SECOND EDITION

ee a i a ae TEA ae rina eters (ec = = =

AEM ieee ee cee ye ant ar enone tanner aN as Se SS

T. BHIMA SANKAAM
G. S. KUMAR
G. PRASAD
37653012710623
Terry NonFiction
530 BHIMA
Applied physics

This book is one of

155,000 in a special
purchase to upgrade
the CALS collection.
Funds for the project
were approved by
Little Rock
voters on
8/17/04.

CENTRAL ARKANSAS LIBRARY SYSTEM

ADOLPHINE FLETCHER TERRY
BRANCH LIBRARY
LITTLE
ROCK, ARKANSAS
 Digitized by the Internet Archive
in 2023

https://archive.org/details/appliedphysicsOOO0Otbhi
APPLIED PHYSICS
APPLIED PHYSICS

Prof. T. Bhima Sankaram

Prof. G S. Kumar
Dr vGs Prasad
Pepartment-of Physics
OsmaniaUniversity
Hyderabad

Anshan pti nae

_—— BSP BS Publications
Copyright © B.S. Publications

First Published in India by B S Publications, 2005.

This edition published in 2005 by

Anshan Limited
6 Newlands Road
Tunbridge Wells
Kent, TN4 9AT
UK.
Tel/Fax: +44 (0) 1892 557767
E-mail: [email protected]
www.anshan.co.uk

Published by arrangement with

BS Publications, 4-4-309, Giriraj Lane, Sultan Bazar, Hyderabad - 500 095. India

All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or
transmitted in any form or by any means, electronic, mechanical, photocopying, recorded or
otherwise, without the written permission of the publisher.

All care has been taken to bring this publication to incorporate the correct information.
Notwithstanding, errors, omissions and / or discrepancies may not be completely ruled out. The
publication is sold/ exhibited subject to the condition that the author, editor, publisher, typesetter
and printers jointly or severally undertake no liability in respect of any damage, injury or loss
caused to the purchaser and / or any person whatsoever by reason of such sale or exhibition.

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

ISBN 1-904798-10-1

Printed & bound in India by Sanat Printers, Kundli, Haryana - 131028.

 PREFACE

We have a great pleasure in presenting the book on “Applied Physics”, an introductory book on
materials. The book is mainly aimed at covering the syllabi of Physics course for engineering
students, as well as for M.Sc. students of all Universities.

We are passing through a paradox afflicting the undergraduate physics course in

engineering, the paradox being: the subject of physics need not be taught vigorously as physics
is regarded as a tool for engineers. Therefore the students can afford to have a non-mathematical
physics. But outside, in the world of science and technology, there is a great emphasis on the
sound knowledge of physics for utility of technical know-how. A non-mathematical physics is
mainly meant for the students of liberalarts and a few branches of natural sciences. They enjoy
‘Physics’ just like one enjoys the beautiful natural scenery ofa land scape.

A student ofscience of engineering enjoys the learning of contemporary physics through

more rigorous mathematical formalism in electro-magnetics, electrical technology, signal
processing, electro mechanics etc. Therefore, the physics should be taught using mathematical
accessories. The student realises at a later stage that medium of physics is powerful and he has
to use this medium for better understanding of the technical aspects of engineering.

The subject of Applied Physics is fascinating as it unravels the mysteries of many

materials that are used in engineering applications. The science of materials is as vast as the
number ofmaterials that are available of putting a lost of emphasis on the ‘materials’ and their
utility. The book is only a brief and introductory to the vast subject of materials science.

We finally thank all the wonderful and eminent writers of books on Applied Physics
that have given a vast knowledge of the subject. We had the fortune of having been taught by
dedicated teachers. They were sincere with great intellect and creative originality. We dedicate
this book to all our teachers who made the subject of physics fascinating to us.

- Authors

CENTRAL ARKANSAS LIBRARY SYSTEM

ADOLPHINE FLETCHER TERRY BRANCH
LITTLE ROCK. ARKANSAS
 {it oe
e)
AG AOD ies miseries Vantec uns beaitne ? a dood wt ustiaied aden

ghiRoDr gs io orion doled. idnilye sli geneva ib trantia ¢letem reget

chau
ovtimsidl s(ete ntaginse oN vol 26 aided
ithwep nates Peybergolms srt anaaiitle volajy & agent: Gris =
anlayrig en ylewoiegly ‘itmMd onl ton bon evinyilolerooted alt ynitod hes sil
lesifertisitici Hot! #3 «7 SrtA Worigiixa elt ssrcibeenrtj Jesiging wh nore@
yl) we chepentinns & Geel “eolorviua han sunstoe tabce sid mA a
H hetta eaiiaaadtw- dion A whrl-word ingineiaan Vs titite, haoNO Sy aye
ent temetisn Lenni Ws eartonined wat =brn etntneseli ey aterpbane oil at as head
“a vb y. pads basi-t ly manene hmoten \iiused bade ows yao.aki tel pia
|
| shynerisduiehp iceman |n grumasl sift eenics-gniyserige> To sonsnge
Selaesbeape Ay 1a
hint Yiyaliiiay fhoatiels ehitongem -evtasls ni nizilaina inoihensrtarn gn om .
=

emesis grien igual ad biyore eniegie ort} roles wa koinntinad) Gaelsye

andlrah oe ve Tea
2 1 eno tall sail? stat & te baaitany jwehute oT
ead “guntenigns 16 Meck te sinter SiN 11)aaebaisbne solid «Apeas
“epuutivn liste wip even Haas ipiualainbeds alee aa“4
aR rere Lan TR PLE IAge gnessnigiy af bine set
Hap beth, aimiowain oe aca pia ep Meh arid Dane de olla tl ee Tit seyir
Snir teeta SG batsceNOS ; a
ente
ba idiy
d 10 s = AP fo 70 igi, iki : e, se
is Wigs Stood! rat! Wy Sau al tow wy tosigu: ao val el
sant wb ai alan: op aie yey |
4) ol sonttiaveeeme nai magne
wad . a a

Mize YEAHH (
HIMABE Y6fcET 2:02 Te
a Baan
CONTENTS

CHAPTER |

Chemical Bonding and Crystal Structure ..............cc:cccsssesssseeeees 1-52

[1 Introduction. Chemical Bonding mSoldsierty Ui Baal ess l

fee SAREE ALONUIGHEORCOOo carte oferta en oo ee eh ase tacta tue 2
Be daws LOUIE SONS ie meter ten thot nage Soper ey END SURE EPR Ds Toe oe ee 4
Tede eMC OE DCLDY FET TONIC SOLIDS tere erie eter tone tvs tence noone eveecubeatetenes needed: 6
noe Homie Radit and Coordination Number 2 f.0svot vic. scceereeosselsoactiecinsses
eesoshs 8
Aang ib EVES) LOT ceoUU bes22) i bie toa i ol REA Mee a 1]
es PESOLYRE DOI CLV ER POLICIES tact conte rece esas cxiceer occ ne ance urdopssaeec sacar re cca teatime 14
Patan Wines ESORI et ADE CeTOC ESS ts nore, ssc? aatdes Soicn cvear eons twtne asa reece ees Ly
Pe PeE AOPES S194LOUIE HOT SUAIS 4 opirvsns cree) sc cams ernie nce eaeandt Some ener 18
Bi scOV Areal DONG ANG COV AICME CIV SAIS © ciccenctecatoe
gieeusovard abs sssaraasegacssandotes 19
vg leRaisColePeal VSSTe 1 aaa a SI a, oats aot? et ok oem O TY eae Sree te 21
BP eeOC OND cant too tr aor eeo ote eee reel mea aa alc ashen ates wee 22
PM SACS LsANIOR Se pcr devised online cbyont eteaMean alee ata gsc he eas tc tas toencnaeraes aurea! 25
PEN OM AMME STOMA VALLICS osc spc die sea tease opes cach ot Aon ocean mee tee eke Caer 2s)
1.15 Three-Dimensional Lattice or Space Lattice or Bravais Lattices : ............ 30
PG eLCevebal SVStetiis SOU DIG ain.ssecse icger, -ovasgsenutee ces gucal- re sane Wagstoaea une seve eae 32
MM ee SEN EICL UTS ceca ec erces otras cand dont cep eet sam sei haan oa aeene ee 33
( SiSheeL NSnRBMGS (0)0 i Sena ae ate exe eo te a Bak a Ae) TBE Ap At Re ae oe SS 38 -
CMM nnplo WUC SITUCtUTC tt ctumancera 7: het tesco tas eccongerueeterese > aevote grees sossoctse 40
POs ome and Covalent Cubie:Strictures 25 o.iiiien cece cates. sFerverssovnscevesersoraseveduaone 4]
Ue MOY CAA LCUICLUITS eerie ee ree ere eee crete ns snecUeabns secs sancuvansece 42
Ee ais anAS LUC
ENN Nai recety nee reaeece seh nilesave iain cones Abb ameRoseanegi ants AS
ee SOG ina © Mlatice (ROCK Sally SUCUCHITE <........tvsects
ees uasetnssvekweseanecgneiosngesons 46
PaPM Se es0 PAGE. MONO) NE NTE NO Ae IR OS a ne a cer 48
CHAPTER 2A

Crystal Defects icc. acre rect tieccecsee casera 53-88

2:1 Direction in Crystals: cetacean
kien une 53
2,20: Crystal Planes and Miller Indicesiveves: 2a: cc.c5 ee 58
2.3 ‘Inter Planar Distance: = csceaueeek Aue eine ® eee eee teen 64
24--lmpertecttons im Crystals ©! 2 2. ..ucccactrsc te nase eee . 66
24.1 Defects in Crystalsit.e ehmcheett eee ee ee 66
2.4.2 Equilibrium Concentrations of Vacancies in Metals : ..............0000-- 67
2.4.3. Frenkel Detects: £5 mca ied Seca: cee ae 69
gD “NOMIC Crystals? eoigaccsesetosiete acs ROR EE Band tee iW
ZILA “VACANCIES Fres, scacvesceeanteme at ean Aes PS ee me re 71
25.2 Frenkel Defects in:lonte-Crystals = 5.45.23. ae ee 73
2 Ore HOPUNILIES 3 tras. cehdecusen Meta ore dee te Aa, ea ee 74
2.7, Study ef existence of pomts: defects i crystalee.. see. ee ee ie
28s. VAN Cele Ctsss Ae case ee RON cea ae ee 76
2.o/ka... heoretical: Shear Stress.4 ci in.) A ae ee 76
2B.2." dee DistocatiOnce. erat: oc 23s ee Gal ee ee 78
Z.O-3° SCTEWIIISIOCAUION Sa om. p a2 reat yok eae an gene aes oe 80
2.64 Burgers Cireuit-and Burgers Vector: =... 4.4 ee 80
28.5. ENGrgy Ol adislOCAONe 1c 1c) seit Co ee 81
2.8.0. Two dimenstonal Detects: =... .1.,meeeecs nsec) eee 83
CHAPTER 2 B

Principles of X-Ray Diffraction Methods..............:cecccsecssseseeees 89-115 |

29° L-tay Difftacttom s «iiers.,cauic ke cde sead eee ee 89
2.9.1 Brags Law4in.c1.5 Gaeta 3sARN see ee eat eee 89
2.9.2, Orderof Reflection. ...0o. ae eee 92
2.10 Laue Equations for-X-ray Dittraction «.,..0n.0u ee eee 93
2.11 Methods of X-ray Diffraction: a. cece ee ee 96
212 Xatay SOULCES sic. sck see eg ee 98
2.13. Laue Method : ..2.....,.mansup
ene 99
 LEN TONIC TOA CTOU Ha eee Ct RMMM RR Tice eileLacs trices Nearer eles nessa tteeies 103
2.15 Difference in the Diffraction Pattern of fec and bec Crystals + ............... 108
EAD DKRAY DUT ACLONTCLET IS eect ee IE Lah Accel 11
PA ROL
EiOraritun ry Stal MCIN Oat crrsdiccssosvdscnincsssvip
edeeen 112

UNIT - I

CHAPTER 3

Principles of Quantum Mechanics ...............:ccsscccsscesssccesseceeeees 116-148

aN NE OAIIC ERSTE 9 x5 crcasa <a cte cam:
Ad,5foe urea edde cedB arGpRG SIRES 116
Bt2. “de Broglie Hypothesis Of Waweeteantic tens ac ccsssasven.co3erieves st0iscdlatiaaeeiaetaaas 117
Dea. MCMAPACICHISUCS OF CESTORIIE: WAVES Pe eo revipescascosvascesdunevicsenusaroneeaies 119
34. TWaveiGroupsoi Wave Packet Concep tes sci aiitsieb co Uincesssscaceehen
eae bee 120
De moa Mics Oman CICTMET EX DCIIMCEUE oe. . dete cated conc oka tas tose eo 126
S640.) Lhomsonbxperiment ; 55. .abties.ta.ecteotest
nietewter 128
Bay aes CUTOCIIICEE AVE
I LEUUAUION fais 5) chhssecsteu cacao cnc cee saa os eee toons 130
Sco a Line iL
ependent eChrodinger EcMariony 2 ccs. 22 set heck ia 8 saccoternatostee
ene: 131
A AMG = Gepengent OCHIOGINGEL EQUAMOU conc, cnocharaictschoasco eteanesooesencoes 133
Ad |ape Dafafirey
BoTT Ll ada 1tes|) emmanuel 138
SP APplicauionror SCMrOCUIGEL Wave CGUAION “we, cornnerccera
ices terqimtat tere tees 141

CHAPTER 4

BIECEV OND MOOIy OL VI CU AS ooo aco rete erases noadinnseceasuccsesinpenctastanscebee 149-194

ME Alia care sas 70)| te Ameer ee ee MeN ee ree me DPCTN: Se nan PEDO RON MRS ME ere 149
Ba IAeSICAL CS, EMC CIMOMIMIOdED 5 cs:toe.ck feavavataeeyve teubeeeierzanbactharetoppi
wastaeaes.3 150.
mam SeCITE
ACLG SA SULANNT UNIT LMC OES fac cat ada esos sas wailcsiuunbaoraameyiedoe
gasses 158
44 Free-Electron Gas in One-Dimensional BOX: wcccisccncseracarsnevercnesssnserveses 160
As) bree tiectron Gas ina. | htee-Dimenstonal BOX : «6... sssicccsucreossrnevesvoesenes 162
4.6 Enerey Levels and Fermi-Dirac Distribution : ................cc-cenencesnedensenensnoes 164
CMONSEY OSEAN eae cre rce eae cote Nae sterwenee crtecsranseetoreresewnnerdngncenegccutsxos? 166
eSM EC ETICANEOOMOUCTIONT: eee re rer tire aerate acne ee tenotpMicvecsehaneseosvone 166
 49 Temperature Dependance of Electrical RESIStiVity 2 :..ct te-ccecssn--usse=.b05-s0: 170
4.10 Impurity Scattering Phenomena: Electrical Resistivity : 00.0.0... 170
4.11, Scattering due to Phonons-Electrical Resistivity<ea eens eared tate 172
Ail2. Hall Bite cts ncccecosccdsnsheccutch Romeoutteceeceed’ ao emia ates SUPe ONCE edt een ORES acetic: 176
A ee Initroduetton:: Band) Theory OL Solids 2.) sacccccercncss
ae oes a ee Fe,
414 Energy Batted T Ormation *eemereemeertveien ...:0cc.c24-secso0redevesteanrecenebsneserceneeaye 180
415+ The. Kronig - Penny, Mode lpm cx.- cc ocscs.sctescasicee sovaspuonscavanewerresssyee 182
416 WumberofPossiblesWave Functions Per Band: .2.4..2.2.:c2 see 187
Al §©Nelocityof Electrons ‘and Effective Mass. «00 sccsscncncnscnosocs csp 188
418" Metals insulators/and Semiconductors occ eeerecces ee 190

CHAPTER 5

Dielectric Properties Of Solids ...........cccccsscsscsscescessscescsescecees 195-250

Salers LRU OOUCHIO Me r akin Ante aia tua hd ee eee nak tear ete ee eee 193
5.2 Dielectric Medium between the plates of acapacitor : ....ccceccccceccccseeeeesees 196
\J atomic Viewor a Dielectric Polarization” 44. 7een eck ete aS
5.4 Electric Flux Density, Electrical fields and Polarization ina
DiclectriciMediuin eat. mutettun. eo) mtr ee, Meee Pew 199
>.) «Rolarand:non-polandiclectrics «csc: 201
5.6 Polarizability (Electronic, ionic and Orientation) : .....ccccccccccscececceceseececeses. 203
5.7 Local field in a dielectric media classius - Mossotti EQUATION a ceter tier 204
>.SeyBlectronic Polarisabilitye ae eee Oe, be 208
59 lonic Polarization ?..6.0 es ee ed Bixee 210
>.10» Orientation Polarization: 0. ee ee Ze
5.11 Behaviour ofdielectric in SUCH AUN Ske san ctea seen Ne esr hc rs, 216
5.11.1 Frequency dependence ofelectronic POIALIZAD NICS ose. he- eee. a7
>] -2- Joni¢ Polarizability 2.2 ee
222
5.11.3. Dielectric Relaxation and Dielectric Loss ! sceccescseccescesseeseeseece-....
222
 S12) Piezoelectricity and Ferroeler
ial cues cs.coc< issn cnenssnanpweusgincaeaneneserseteonsen 2270
S82: | Unthodcticiy sce eee eI oe sn oeaicnstroasscansaiees
vemsoeaesaerseserenes ps |
Be MP IEZ OCIS CLIIOILY scares eee, Ce sare trees orra ae ey nthe on see Me aceccigeey 228
Be INS
CLIN CMUS eae car cec gays tp cactus nics a nanlyzaroniendeen spicavomhesiamuctencesite 233
Bele eM OCICC LIC WIAECII AIS 1cee. cee cece carota sane p pG Rinses Oper ora ee ee oped 236
BONNE Fe OVEUNNG NLALE EUG 1 cceccercoecsoi a Sains anise Sanele cae seasas oe ges Aan denne R@awetoncind so: 239
Sikessewer 240
Sl ue AAD MAC MMONIG carts race cstetns seetet bari atin Sate Neeede sae eal te iaeataslercereemrsecenonoesr-
5 Ai, MCISCILIG AS INSUlatOrs. atid (CapaCitOES i527 :fetisinatoeesereectoenctcernerneveeonserhense 244

CHAPTER 6

Magnetic Properties .........scscscsscssccsseesesscssssesssseseseeensesseenees 251-303

6.litniveductond cities oe ore eras erates ced iat sa rhancainu atts Zl

6.2 Magnetic properties of atoms : Bohrmagneton | ......... ce eceeeseeeeteteetteees e537

6.3 The external magnetic fields and a magnetic material * ...........:.e ee 260
GIA Dita grit isan 32a o enon orcncbaco cones cusens jon ctnnngncpandeanonvemseationrenses 262
6-5 *Parana Tete SNM 6 eileen aaa er cane ewntennasenevnesce aces -nonaneesesateapasavansiseunoneenest 265
GG PerromagmeriSiis Losses iscecc. ene coecneeaccanessnnnatetansopncenssaanhndowmemanedantantennnseys 269
6.6.1 The characteristics associated with Ferromagnetism : ............+5 269
6.6.2 Molecular field-Weiss domain model : ..........:cccceeeeeeeeeeteeeereens 271
6.6.3. Temperature dependence of spontaneous magnetization :........... O13)
GA WWCIS SLA
PCTS set ee retest ieheaott nn cnaiicacosenbnontouneeepueepreneentoaayane276
6.6.5 Bxchange Emergy ¢ ...2..-.-cscsncsseonsscecevercossanstuovensveonsennecestagenectocarens Pedi
6.6.6 Hysteresis LOOP : .....ccccscsceeseesetereesesesesseseseeesesenenseseneteenenesetseanes 219
6.6.7. Domains and magnetization : .......:ccccceeeesereteeeeeseeeeneeeeeneeenees 282
6.6.8 tereteteeeeeees 284
Magneto crystalline Anisotropy : ........:::csceccceseeereeeeseete
6.6.9 Magnetostriction :.........cccccececseeeesesesanensenenssaenenaeaeeeceseseanees 285
6.6.10 Application of ferromagnetic MAtC CIAL. .a<. coveteaag ieee yews 286

6.7 Antiferromagnetism : ...........csceseneenecenerenrereresestenesenseseeneeneserstssereeesenestes 288

6.8 Ferrimagnetism : Ferrites : .......ccceeceecesseseseeeteneneceeenenenssterererereneseanentaes 290
6.8.1 Structure Of Femmites «....ccsccescressts.cscsstcesteenescaverneeoecesnenmeaeeserense 291
682 Characteristics Of Ferrites :...cvec isi.
cedtebeneen ses
seeesccnesesenen eeneee 295
 CHAPTER 7

Semiconductors and Devices ..........ccccsccsssssesesesesesesessecsoseseceeees 304-347

Tele TMC OTT%sc5, 3, «casa ole ep ee pec op g 304
Tg. Fermi ENergy < .......:..scmeh reas ene, tree eat Sele ae 305
voi Hlecton and Hole Densities: 2.1.7.6 eee 8 ee 306
7 “Tntrinsie Semiconductors =. . ccs. cccscssesssessueecsseerscssractssoscct-sc.
0 307
GAS EXTONSIC SEMUCONGUCLOLS aetr.e a ee ee 311
7.5 Carrier density in extrinsic semiconductors :
Condition of charge mettrality < ......./..:...sseeeenedee
ee.
neel 314
EOF Rec OMBOD i otras Pete ke ol toe atm ee eS 318
die SpUTince Recombination ss iat 46 2h Des leas dy, 319
71.8 Drift and Diffusion in Semiconductors : ........cccssssscv
ecsss+sesecoeessssseccccccszes 320
BedesAbtleL Scattering Ole ete a oct oe og il eee 322
79.1 Scattering due to lattice defects :.......ccccccccsscsc
sseeeeeeeccocs
ccc. 322
7.9.2 Scattering by Ionized DA DURILY: LOLSi: Miciter’ ch tal age
ieae ce323
[eG Cen Set eta nk Delt oe oes se ot Re 324
ee UUSIOUGU Tented este ot Ages Sensei on eee 326
PNET UDC OTN OGC: sie Reta: te gh on elie ater Oe aa 2 327
7.13. Volt-ampere characteristics of pn juncti
on diode: am.. eee . 333
7.14 Direct and Indirect Band-Gap Semiconducto
rs ° voecccccccscscssseeoeeececc 338

CHAPTER 8

Buper Conductivity. ..iiisis

msa
iacan
ee 348-383
Sl introduction... Ata ee
TS 348
8.2 The Properties of Superconductors: nc. eA. 349
8.2.1 Electrical Resistivity. 28 ee MCT.
349
8.2.2 The Magnetic Properties : Meissner Effec
t + o.....ceeccssccccsseecc.. 352
8.2.3. Thermal Properties + 2.0
i CR 358
 SOA ASOLOPe CETC Ute ert ernie eee mrss tater ee sacar enact daneoste 360
Soe OE ONIMON PEE ALON rt can re ona in olen co dancteateas cnsenadeserocens 361
8.3. Type I Superconductors and Type II Superconductors : ..........:eeeeee 364
Bi PG el SUD CICOMGUCLONS cence cc tenet eee wast gsnonneetcostnenusynoncaaieaseene364
ee lyPO Set CONGUCIOLS occ erestenoge-vay-taronne<fectcnsnsonrt
creetearm nce 364
PAM er Ua AETOE epoca
Pas keen hvsenate ene a OOO.
ae EE ENC Ae ae anne ep cae osc a ct dese inn nap tgnactsea fave aeabtionwokios oteasaemse coacomnasaee 368
$3.6. Minick -apegcrns 3/2
Application Of SUPELCONGUCIOTS ©, ..5....c5c00cs1.caces-seonauresrere->npsa
Bor cater egedsanees tetareneres aes: er]
otligh Te Superconductors. 3. 2ec. cic ye.nanneetevsiountobasten

UNIT - VI

CHAPTER 9

MSCS creo acs sc denna Ease evn ncdeebeas naa teo bY ese omald neo e 384-406
stvomeyanceca-Weoanencsanes
ite,1/8)1 pee
OR UD irarel See ee Oe, ten ear See A AR eR ie re es 384
9.2 Principle of Laser Emission : ...........:scsecsescerentereeeseessseesessseasenenseseneanenes 386
OFT ~ Emissiomot radiations cis22 08 oes cae amare nee ee ate oe 386
9.2.2 Population Inversion : ..........csccseeeceeeeseeeeseestseeeeseeeeeeseeeeseneeneeeney 387
ESE INSECT SERIO ICING 2 ose er rctsnecov ses Se ace apd eal aadicse nme enka dy ao umm een 388
9.4 Light Amplification and laser oscillations + ............ceee tee 390
tte teeeteeeeees
OAA Amplification t tivnecsisstiiisicessetcodbuerseenacetentnen- sonscveseceasaiontinevenaes 390
OND -etaser Oscillations y:icaveesc SARE Le ania ieecs ah oeerndeckci ene eeers 392
9.5 Properties of a laser light : 0.2.2... eseeeeseses esse eseseeeeeenenersnenetenenenenenensnesencenes 393
951 Coherence and Monochromaticity : .......2.6--.csssseeccraenarecsnnepernenees 393
5spugs W101 0111 me Neen guins ne ane NURS FEMS re 394,
9.5.3. Speckle mature : ........cccceccceeseresesseseseseeeseseneneneneteeeenenssceenessenenens 394
9.6 Methods for producing Populations Invention : .........-.::s esei 395
tetteeeeees

OF Gas lsasein GN © Liar eri rete seesoaee canta decane beanden conve ctousntenmetesnreensetens 396
9.7.1 Working of He-Ne Laser : ........:.ceseecesceeereeeeseneneesetenetneneteeseneees 396
9.7.2 Applications of He-Ne Lasers : «02.2. .. ens 397
tees reeseseeseeeee
9.73 Other Gas Lasers 5 oct cncce ceo enesscneceecesescencannesterenssasocenensacennanne 398
 9:3 SEmicOnduetor Lasers aaa
en RS her 398
9.8.1 Working of a Semiconducton Easers. sched... vise Shee,
9;8.2 Companison.o£ Lasernand er Deg ee eae ee 401
9.83 Applications of Semiconductors Lasers seq) somes
ee es 403
Dboys tate. aSel SsRUDYLaSCl 5 teint Woe, wed» eR ey trea 403

CHAPTER 10

BIDE OPUS een eee ete er ee ae 407-424

ROSMAN OCUCTION SE 5 ison cnceteecai
in aged eeaih
ne et ioace
rene a e 407
L0-1.1. Structure of the Optiealilatberetie . .25:.0..5-.ceran0ceeeceae
ee 408
10.2 Principle of Optical Fibers eee ca... <ccicsccevocescl
dee eas sa 408
Oza okaldinternal Reflection: .20.....cie.
cee een
ia-..
em 408
NOP ZAP NSEC PUAN COVA IN 6st esses Beasstins ectaose wcsecocceen shee 409
1022-3 NUMELICATIA PerLUTO cali ve eeetaseiaceteesoenaneomeneee
eR 410
LOS envlodes Optical Fibers. take dkcia alee ee 41]
LOA MiSteplndexck beri sencicctec isis ns, SR, NERS eS rs sorb To een te? 413
Dye Oridcdindexhibers samen cn tet eee ee ae 414
ROTO MIOGCOUP seas
cecteaseon
ccscescece, metisa sha Cees eaaueceite: a tee 416
MOP rd
CCD
AEAMOMMORE DES ys erie oo Gad oo) ee a De 416
Moet Cr RACKING. roca. a et. bane Ee Dees as Se ne! 418
L0.On- Losses invOptical Fibers <.: .c sa0-.:0 0 eee aun = Meee, 419
L0:10sApplications of Optical Fibers-t/in:....<.. ee ae © ee eee ale 420
EXD DOME LCOS Forosvsessesasodcodan-tcee
estaststvass:eeoal 425-450
ADIOS. csecc.ssornnscsncssccasesnnzccrse sssesertseeriecite
oe eee 451-453
Index
1
CHEMICAL BONDING
AND CRY S PAL sTRUCTURES
————S S

I Chemical Bonding in Solids :

1.1 Introduction:
state or gaseous state
Solids are formed when the atoms (molecules) in a liquid
is in close proximity of its
are brought closer into a condensed state and each atom
re of atomic arrangement in
neighbours. The condensation into a particular structu
interaction forces between
solids is determined by the nature and magnitude of the
of solidification, the interaction .
atoms, temperature and pressure. In the process
energy of the system. The
between atoms is such that the interaction reduces the free
fact that the electrons occupy
system like a solid, possess lower energy because ofthe
d atom. The state of atoms at
lower energy levels than the electrons of typical isolate
on of stable equilibrium. The
an equilibrium separation thus corresponds to the conditi
be at equilibrium separation to form
interaction forces will be such that the atoms will
a crystal of an ordered internal structure.
 Applied Physics

In general, the chemical bonds that occur in crystals are classified as 1) primary
bonds and 2) secondary bonds.
The primary atomic bonds are formed when the interaction forces are relatively
strong. The primary bonds are classified as 1) Ionic bonds 2) Covalent bonds and
3) Metallic bonds.
The ionic bond is formed between two atoms when one atom transfers an
electron to another acceptor atom to give ions of opposite charge. The bonds are
formed as a result of strong Coulomb forces. The covalent bond is formed between
atoms which share electrons present in the outer most shells. The metallic bond is
formed between two metallic atoms which share electrons in a delocalized manner.
The bond is strong and nondirectional.
The secondary bonds are classified into two groups. 1) Permanent dipole bonds:
These bonds are formed between molecules that possess permanent dipoles.
2) Fluctuating dipole bonds : These bonds take place among atoms as a result of an
asymmetrical distribution of electron densities around their nuclei. The bonding is
very weak and is termed fluctuating since the electrons density continuously changes
with time.
We shall deal with primary bonding between atoms and some crystal structures
as examples in the following sections.

12 Interatomic Forces:

In a gaseous state of matter, the distance of separation between molecules is

large. The interatomic force between the atoms or molecules is very weak.
The
molecules or atoms exist in a continuous motion colliding with each other
and the
walls of the container in which they are present. When they are brought
closer and
closer, they first condense into liquid in which the translational motion
of the
molecules is totally restricted. The energy is then lowered compared
to that of free
atoms. If the liquid is then condensed to forma solid, the atomic motion
is restricted.
The atoms will have only vibrational motion and the rotatory motion and
translational
motion are totally absent. The atoms well be held by the bonding
forces. The energy
of the solid is much lower than that of the liquid state or gaseous
state. The atoms
in a solid are held by an energy called the bond energy.
The liquid, like the melt ofa solid, consists of a short range
order that is not
stable. There is a continuous agitation of molecules or atoms.
The short range order is
broken and restored again. This process takes place continu
ously with a relaxation
time. When a solid is formed out of such a state, there
is a heat of crystallizations that
is evolved. The atoms are brought closer to an equilib
rium distance at a temperature
that is the freezing temperature. In solids, the forces
that operate between the atoms
are the attractive forces and the repulsive forces and these
forces determine the potential
energy associated with the atoms in the solid.
Chemical Bonding and Crystal Structure

Let us examine how does the resultant force between the atoms changes when
the atoms are brought closer to each other. At relatively large distances, the attractive
force between the atoms dominate. It grows as the interatomic distance diminishes.
This variation is shown in Fig. 1.1 as curve 1. When the atoms come sufficiently
l
closer, the repulsive forces come into play Fa oC +) and they grow faster, than
r
attractive forces. When the atoms are brought closer, their charge distributions overlap.
This overlapping causes electrostatic energy to change. The overlap energy is repulsive
at close distances. The repulsion occurs because Pauli exclusion principle prevents
multiple occupancy. When atoms reach a distance of separation r = r, the attractive
forces just balance the repulsive forces and the resultant force becomes zero (curve 3).

The interaction energy U also changes as F = —0U/dr. The interaction energy

at r = r, 1s called binding energy of pair of atoms. The state defined by [U) ie is a

stable equilibrium state. What is said above is valid for all types of primary bonds and
the general character of the forces is the same for all types of solids.

Fig. 1.1 The variation of force between atoms as they are brought closer to each other curve
Qattractive forceQ)repulsive force@)the resultant force between the atoms. The solid
circles and the shaded peaks represent charge distribution overlap.
 Applied Physics

We shall now see the different types of bonding in crystals in the following
sections.

13 Ionic Bond:

Ionic bond is formed when the atoms with small number of electrons on the
outer shells, have a tendency to give up electrons to the atoms with an almost filled
outer shell. Ionic bond is the simplest type of interatomic bond. By loosing electrons,
metallic atoms like alkali-metals, alkaline-earth metals etc., go into the nearest noble
gas configuration, while the atoms like halogens take electrons to gain the inert gas
configuration. Alkali halides are the simplest type of ionic bonded crystals.
The ionic crystals consist ofpositive ions and negative ions. Sodium chloride is
the best example for ionic bonding and it has the highest degree of ionic bonding,
sodium atom has one 3S! electron and upon ionization process gives up electrons to

The formation of Na C/

OS OO
neathceteal
oor ev
aie a bs

ionization energy

Qi — 4G i
electron affinity
+ = + =

binding energy
become sodium ion. The chlorine atom which has half filled p-orbital, takes the electron
from sodium, forming chloride ion. The two ions exist as a single unit. The sodium
atom with 0.19 nm size is reduced in size to sodium ion with size of 0.095 nm.

Let us now see how the interaction forces between the ions operate. Consider
that the two ions, sodium ion and chloride ion are initially far apart. They
approach
each other as they are attracted to each other by Coulomb forces. The Coulomb
force
is given as

f -(Z, q)(-Z, q)
Coul 4n & r

ai Z,Z4q°
A eear
Chemical Bonding and Crystal Structure

where Z,, Z, are the charges of the ions, q is the electron charge, r is the interatomic

distance and ¢, is the permittivity of free space (e,= 8.85x107?C?/N-m? ).

The repulsive force becomes significant as the ions come closer. The repulsive
force is very strong when the ions or atoms are very close to each other and the force
is described by the expression

where n and B are constants. 'n' is called as Born's exponent.

Corresponding to the above forces, the energy changes, associated with the
two forces as the ions come closer, is given by

and Padi sean) & deutie with) Seg) & etary (1.4)

Thus, the total energy change as a pair of unlike ions approach each other is
given as

Tea
142 B
U=- ana Bie ewe er (Te5))

This can be written as

B =A
ie ne aS (1.6)
r r

When fags
a .
ag iad
0U
=
—
0

interaction of
We have inserted a constant A in eq. (1.3) and it takes care of the
the ions.
 Applied Physics

1.4 Lattice Energy in Ionic Solids: -

The above condition that (@U/éar) r= = 0 is the criterion for thermal

equilibrium. The stable configuration is possible only when the energy is minimum
(curve 3 of Fig. 1.2). Let us see the variation of forces with the distance of separation.
This is shown in Fig. 1.1. The net force is zero at the equilbrium position.

When like charges approach each other, they require energy, when unlike charges
approach each other they release energy. When the ions are brought closer and closer,
there exists a repulsion due to electrons of the adjacent atoms. The Pauli exclusion
principle also comes into operation and the electron-repulsion energy becomes
significant as the atoms approach closer and closer. The energy (curve 3 in Fig. 1.2),
possesses a sharp minimum as the inter atomic distance becomes Tp: This is a stable

inter atomic distance. Energy is required to reduce or increase this distance.

Fig. 1.2 Potential energy variation. The solid line (3) gives the
energy ofthe solid, the resultant of
the two energies. The dotted lines (2 and 1)
correspond to the repulsive and attractive energies.

Suppose the lattice is built of N ion pairs, then the energy

of the lattice is
obtained as follows.
Chemical Bonding and Crystal Structure

From eq. (1.6), we have, at r = I, the two forces are equal. Then

B = A :
re - ies tyn-l — B

Kees: (1.8)

Z,Zoq° to"
es =
4n en

Clee Fae ae
U, = - = (i 1) wees 1.9
At &, ag n ay
As the lattice is built up of 'Np' ions pairs, the energy ofthe lattice is

gay ale Beara a

Lat dn
& % -
—AN p ]
ao ies) sone (PaO)

where V,, represents the volume.

The energy trough will be very narrow for strongly bonded atoms. As a result,
the slope, dF/dr, of the force curve at r = rp in Fig. 1.2 becomes very steep. This
oO

produces an interatomic distance which is constant within 0.001 A at a given

temperature. In such a case, we can take the ions as hard balls with definite radii.
We can determine the value of B in the expression in eq. (1.2) for a crystal like
Na Cl. The interionic distance is 2.76 x 107!%m.
The attractive force is

Cae
F att
4n & 1°

leat ren
an (8.55 x 107 C?/N.m?) (2.76 x 107" m)
= —3.02 x 10°N.
Near r = ry, this is equal to the repulsive force.
n+]

Then, tioJeo oe Te oe
n
3.01 x 107? x (2.76 x10-"" J"
= B
n

n= 9 for Na Cl.
 Applied Physics

10
3.0210 > x12.76 <0”
9
= 86x 10 ven Te
Then, the binding energy for Na* CF pair is

=f 8.6 x10 °° Nan.

b 4P & a (2.76 pane) oon m)V

(1.6x -19
10°C}An 86x10 -106

4n(8.85 x 10!?)(2.76 x10") (2,76 x 10°!)

ars GGA ee Nel eis ka 02, Sl Onl
= 74a On ML.
The bond energy due to attraction is — 5.23 eV and the contribution of the
repulsive force to the energy is 0.575 eV.
Let us take the attractive force into consideration. This force is dependent on
r), Z, and Z,. This force varies as 1:4 :9:...:. for monovalent, divalent, and trivalent
ions when they are placed at the same distance of separation. It has been found
experimentally that the deformation ofa crystal requires external forces. The resultant
forces in the solid are due to the forces between the structural elements of lattice. The
attractive forces and the repulsive forces give rise to a potential given by the expression
(1.6). At small distances, the repulsive forces predominate, otherwise, stable equilibrium
would not be possible.

15 lonic Radii and Coordination Number:

In discussing the ionic bond, we have just considered two ions. A crystal has a
certain number ofions per unit volume that are distributed in three dimensions. They
can be considered as having spherical shape with a characteristic value for radius in
crystals. There are a large number of ionic crystals, namely, alkali halides, alkaline
earth halides, oxide, transition metal oxides etc.

Since ions can be treated as hard spheres, carrying a charge of normal parent
atom of the crystal and thus interacting through central electrostatic forces, the crystal
structures of ionic compounds to a large extent, are determined by packing of ions
together as closely as possible under the strong electrostatic forces. The ratio of the
ionic radii in the crystal structure is very important aspect of ionic solids.
Chemical Bonding and Crystal Structure

A crystal of sodium chloride consists of positive ions (cation, Na‘) and negative
ions (anion, Cl). The energy of attraction between the two ions spaced at a distance
Ty apart is given as

Za lot: 7 =i
Uy AN E Mp Sih

where q is the charge of the ion. The reduction of energy on attachment of an electron
to an atom of anion is higher than the increase of energy on detachment of an electron
from an atom of cation. The energy of an aggregate composed of positive and negative
ions is reduced further due to the fact that each ion of one sign has ions of opposite
sign as its nearest neighbours. This does occur as a result of coulomb attraction ofthe
oppositely charged ions. The number of neighbours surrounding an ion and the type
of crystal lattice are determined by the ratio of atomic radii of cation and anion. If this
is R, then
Ritaaiuhts ventiicescé (1.11)
where r, and r, are the radii of the cation and anion respectively. The lattice of the
crystals is stable only when the cores of the ions of one sign are in direct contact with
the cores of opposite sign only.
The coordination number z for the ionic lattices depends on the value of R. The
following table gives the R values and the corresponding Z values.

a Pca 0:77 S-00 oul 20:73 0.22 — 0.41 6.152 0.22

tis Pe eeee
eee
The example for Z = 8 is cesium chloride (CsC/) Fig. L3(C), fot,2 2011S
NaC/ Fig. 1.3(b), and Z = 4, it is CaF, or LiF.

anions
Fig. 1.3(a) Coordination polyhedron for coordination number 6. @ cation O anion. The
1, 2, 3, 4 are at the apices of the polyhedron.
 Applied Physics

The Fig. 1.3(a) gives the

coordination polyhedron of an ionic
bond that has Z = 6. When this
polyhedron is repeated to fill space,
the arrangement of atoms in a crystal
is shown in Fig. 1.3(b). The alkali metal
halides and hydrides, the alkaline earth
oxides and silver halides (except Agl)
belong to this type of ionic bonding.
The above structure called the
rocksalt structure becomes unstable
when the anions come into contact
along the coordination polyhedron
Fig. 1.3(b) The unit cell of acrystal in which Z = 6.
edges. Then , if a further decrease in
The structure is known as rocksalt structure.
r,/r, is caused, the coordination number
falls to 4. In a fourfold structure, tetrahedral structure, the anions are at the apices of a
tetrahedron. Decreasing r,/r, produces an instability on the hard sphere model
of f.c.c when the anions touch along
the edges of the tetrahedron. From
geometrical considerations, the
tetrahedron coordination is possible
only for 0.414 > 1,/1, >°0.225.
Tetrahedral coordination
polyhedra may be arranged or stacked
to fill space in two ways. The are 1)
the structure showing cubic structure
and known as zinc blende 2) wurtzite.
ZnS is the example and it exists in both
the forms with a cordination number
4", Fig. 1.3 (c) CsCl, with coordination number 8.
Crystal Radii:
The ionic radii of the cation and anion are taken such that the sum of the radii
i.e. r, +r, is equal to the equilibrium distance between the corresponding two ions
in
contact in a crystal. The equilibrium distance between the ions is determined by the
nature of electron distributions of ions, but the structure of the crystal and
the ratio of
radii of cation and anion (R).

Sodium chloride is one example in which the r, +r, = 1, and it is taken

asa
standard standard configuration. Other standard crystals are NaF, KC/,
RbBr, and CsI
with r, values 2.13, 3.14, 3.43, and 3.85 A respectively. All these
crystals have rit, ~ 0.75.
The crystal radius of an ion (the radius ofthe ion in the crystal) depends
on the type of
the compound and the forces existing between the ions. The
actual size of the ion

10
Chemical Bonding and Crystal Structure

depends on the most probable distance for the outer valence electrons and is given by
1 : :
ra ——, where Z,,-is the effective charge of the nucleus. The effective nuclear
eff
charge is equal to the actual charge 'Ze' minus the screen effect Se = s of the electrons

in the inner shells. Therefore, R = Fae

Ties where the constant C is determined by the
total quantum number of the outermost electrons in the ions. The effective nuclear
charge for Na‘ is 6.48 e as s for neon configuration is 4.52. The effective nucleon

change for F is 4.48 e. The lattice parameter of naF is 2.13 A. Taking the inverse

ratios of the effective charges we get the radius of Na* as 0.95 A and the Fas 1.36
°

A . The observed values of interionic distance for crystal with Sodium Chloride structure
can be compared with radius sums as given below in Table 1.1
Table 1.1

Crystal

NaF Zoi 3:03: - 23:0)

NaC/ 2.76 3.90 — 347
Nal 3:1} 3.00 3.83

It is found that the observed inte-rionic distance differs from the sum of individual
radii ofthe ions. In ionic crystals like LiC/, LiBr, NaC/, NaBr, CsC/ etc., the anions are
in contact with each other. As R = iat, and if R falls below 0.414, the anion-anion
contact will be better than anion-cation contact. This results in effective change of
radius of the ion in the crystal.

1.6 Madelung Constant:

While discussing about ionic bond, we have considered two ions in the
eq. (1.9). We just have extended it to Ny ion pairs in eq. (1.10). Let us take up now the
same in a general case.
Madelung constant 'A' represents the effect of the interaction of all the ions on
any ion in the ionic crystal. All the ions in the lattice will be contributing to the energy
of cohesion. We shall discuss a simple way of evaluating the Madelung constant for a
crystal like NaC/. In ionic crystal, individual atoms do not exist. Every anion is
surrounded by a certain number of cations and every cation is surrounded by as many
anions. The coordination number is determined by the ratio of cation and anion radii.

1]
 Applied Physics

Consider a crystal as arrays of ions (NaC/) as shown in Fig. 1.4., where a two
dimensional array is shown. Let u,, be the interaction energy between any two ions i
and j. The interactions are pair-wise additive, the total interaction energy between -

©OOCOOOO0O
©O0G0G000
©COGOCOOOO®O
®GO9O000G00
Fig. 1.4 Array of ion pairs in an ionic crystal.
ion taken as a reference ion, and all the other ions in the crystals is given as a summation
over all pairs of ions as given below.

U; = ysl) ee (1.12)

Where Ds _ indicates that the summation is taken to include all ions except
J
j=i(ie., the same ion). This interaction energy does not depend whether the reference
ion 'i' is positive or negative or the position of the ion in the array or crystal. The term
U.. consists of, first, the long-range coulombic interaction. It is due to central force as
has already been stated, the second, force is the short range interaction. We have
defined the coulomb energy between two ions i andj as
ZZ, foe
Fe aes) NIG Se) (i313)

Suppose r;; = p; rq and r, is the shortest distance between unlike ions in the
crystal. Then the total electrostatic energy between the i” ion and other ions in the
crystal is given as (r;; is the distance between reference ion and the j" neighbouring ion)
' be Dh kee
= wads Se rivia
Ure set ame GE Debeed ted fee (1.14)
td Z,ZjA'e° |
iar ree CLENS)

where A'=
ros
' (4) 1
Ate Wore” (0 8)
peg hes Geel Neer! Se of 1.16

The constantA' called Madelung constant. The Madelung constant is the properly
of a particlular crystal structure. It has to computer, + Sign means
different signs of
the charges at distance r, and — means equal signs. A' depends on the
type oflattice and
is independent of the lattice dimensions and the kind of ions. The
coulomb energy is
ZZ.
given as U coul =A; N
PATERT,

12
Chemical Bonding and Crystal Structure

: We consider one row of ions in Fig. 1.4 for NaC/. Consider the ions on either
side of the reference ion ‘i’.
2
Les
ie AT Eq Ip

eee
tet
= Are, fy

ees
The net energy is 7... -
BY 'S ag Ep lp

The next two ions on either side contribute

2e°
At Ey 2%

The next-next-two ions on the either side contribute

=e
At Eq .3%9

This can be done upto the end of the chain on either side which can be taken as
Oe LON} OC

Then

the
Thus, the factor two occurs because there are two ions, one left side and
other on the right side at equal distance.
oe ae
or ON eg ee (1.18)

i.e., RM ce ee i he ie ot (1.19)
Thus, the Madelung constant for one dimensional chain is A'= 2 /n 2.
generic
The cohesive energy decreases monotonously in alkali halides with the
Cl, Br,
formula, M* X~, where M* can be Li’, hla. Kor Rb and x can be/F:.
for a series of
and I- for same M* ion. The Table 1.2 gives the lattice energies
of the lattice and is independ ent of the lattice
compounds. 'A' depends only on the type
energy U of the lattice energy
dimensions and the kind of ions. The attractive potential
: i
is proportional to — .
To
 Applied Physics

The calculated Madelung constants for a few crystals are given below.

NaC/ 1.74756
CsCl 1.76267

MX 5.03878

Rutile 4.816.

For isolated Na*CI molecule, the Madelung constant has a value equal to one.
But in a crystal of NaCl, it is 1.74756 i.e., about 75% more.

The attractive potential V,,,, of the lattice energy is therefore proportional to

Ve The lattice energy is given as

17. Born Repulsive Exponent :

The repulsive energy is given as BBR is evaluated in the eq. (1.6). 'n' is
called
r
the Born exponent. It is evaluated on the basis of compressibility studies
in ionic crystals.
During the compression of a crystal, the ions are pushed to closer
distances and hence
their electrons of outer shells interact vigorously. Therefore, the
study of compressibility
gives the information about repulsive forces. Let us see
how does the study of
compressibility help us in the estimation of Born exponent.

The lattice energy is a function of the volume V the

crystal. The change in
volume can be brought about by the compression of
a solid by external hydrostatic
pressure p. The pressure p should be small. Then, the lattice
energy can be given by
Taylor series expansion at V = Vo
Chemical Bonding and Crystal Structure

At V = Vo, the lattice energy is minimum. (at T = OK and p = 0),

5 oU
oe ol Sy) ee =e ae or (1.24)
ONY Ea
So that, we can write the above expression to a first approximation as
ie
URW)
bp (V)=2 Uso gt(eo (AV) oo an eee b.25
(1.25)
0

Then, the second term is equal to the work done in a compression of the crystal
at T=OK. The W or k in a compression is given as
1 Vv

w=— |pdV=— |[(V-V,)dV

rae !

ZY seeVo 2xVo

if au) a (av)
os ~ al ev , AVY oy Vy (1.26)

(ou) i
lav"), a 7 eer Clea)

Where x, is called the compressibility of the solid. It is defined as x =— 7 [oy,

At T = OK, and the pressure p = 0, the free energy defined as F = U — TS, is

equal to the lattice energy. In terms of volume of the crystal, the expression for lattice
energy is written as

Substituting eq. (1.28) in the above, we get

15
 Applied Physics

ees ae ee (1.29)
x 9 NG

Constant A is given as
Tien
=A!
CaN ys ae Wap.F,
Me 1 “eee (1.30)
1.30

For NaC/ and analogous crystals.

1 x A'z2 2 1 N

ape = SS ee (1.31)
x 9 Ane, 1% Vo

If we denote the number of ion pairs in a volume le as n,,, we get

dpe sol SS VE
Rea*9 Ans Tee
ree (132)
Hence, the measurement of x allows us to determine the repulsive exponent n.
The value ofn is near 10 for many ionic crystals. The following tables gives the values
ofn and the lattice energy of a few alkali halides. (Table 1.2)

Table 1.2 The values of Born's exponent and lattice energies calculated and
measured. The energies mentioned are in KCa//mole.

ee es
S5h\ys'8:05
foaliosfieshie
KCal/mole | 240] 194] 183] 165.7] 210 170 Sy 176 152] 146] 138.6

measured in
KCal/mole | 242 le beesel 214 163 155.4 149] 142.4

The compressibility and repulsion exponent are inversely proportional to one

another. In many cases, the value of n is high confirming the fact that the ions can be
regarded as hard spheres (n —> o ) which are arranged in the lattice in such a way that
the nearest neighbours touch each other. The repulsion force is principally due to Pauli
exclusion principle Knowing the exponent value allows us to calculate the lattice energy,
the main part of which is due to coulomb energy. The values of Madelung constants
for some typical crystals are

WaCr = 1,748) Cs@/ 7630 CaF, — 5.039 ; ZnS — 1.641.

16
Chemical Bonding and Crystal Structure

1.8 The Born - Haber Process :

One can not determine lattice energy directly and in fact, there is no means of
measuring it directly. When energy in the form of heat given to a crystal, it disintegrates
into molecules are neutral atoms and not into mutually independent ions. The lattice
energy is defined as the energy needed to decompose a crystal (NaCJ) into its structural
elements (Na* and Crions). This energy can be determined with the help of the Born-
Haber cycle process. Take NaC/ as our crystal for study. The individual partial process
of the cyclic process are given below. The process is a thermo chemical process. The
process is depicted Fig. 1.5. The process has four steps.

Step I :A disintegration of NaC/ crystal into its structural elements. The structural
elements are the Na* and Cr ions. The process requires an energy equal to
lattice energy U,,,.

Step II : Na” passes into gaseous state. It gives off energy equal to the ionization
energy *[x, CI takes the atomic form C/ by absorption of energy &> which
is the electron affinity of Cl.

Ue ~ =
[NaCl] orig » Nae =

“Ina &

[Na] olig + 2 Cl, Hy, > Na+Cl

a

Fig. 1.5 The Born - Haber Cycle.

Step III : The gaseous Na sublimates under the release of the sublimation heat Hy,
forming a solid that is metallic sodium.

Step IV: Atomic Chlorine passes over to form crystalline NaC/ under the release of
the production heat Qn).

Then the lattice energy is given as

l
Up. = Qnat * Ana + 2 lint Le coh ype i (133)

sal
 Applied Physics
ee aaa

energy
The quantities on the right hand side are known. Therefore, the lattice
given in
can be estimated. The calculated values and the measured values are already
the Table 1.2.

1.9 Properties of Ionic Crystals:

A large class crystals like alkali halides, alkaline earth halides, metal oxides,
sulphides, selenides etc., belong to the group of ionic crystals. Alkali halides are the
most basic crystals for study. We shall briefly give the general properties of crystals.
The ionic bond is a strong bond, The properties of the crystals, depend on the
lattice energy. Some basic properties like heat of sublimation, melting point, boiling
point, solubility etc., show a regular dependence on the inter ionic distance. For example
if we take MF, MCI, MBr, MI, the halogenides, of alkali metals (Li, Na, K, Rb, Cs), the
boiling points and melting points decrease as one goes from fluorides to iodies. In
general, the ionic solids possess high melting points.
The ionic compounds exhibit large band gaps and therefore come under the
class of insulators. The resistivities of alkali halides will be of the order of 10'* Q—cm
or more at room temperature. The oxides exhibit much higher resistivities. For example,
the resistivity of A/, O, is of the order of 10!¢ Q —cm at room temperature.
The resistivity of all ionic solids decreases with the rise of temperature. That is,
they show negative temperature coefficient of resistivity. The resistivity is a strong
function of temperature in ionic crystals.
The conduction ts basically electrolytic in nature. The ions move through the
lattice via the point defects, when electric field is applied to a crystal. The study of
electrica! conductivity helps to determine the characteristics of defects in ionic crystals.
The defects like point defects and line defects are charged defects. The addition of
impurities strongly affects the conductivity of ionic crystals. This is made use of in
making the electrolytic solid state battery cells. Supper ionic conductors are useful
materials for solid state batteries.
Many ionic crystals are transparent to radiation over a wide wavelength range.
Normally the alkalihalides have absorption bands in infrared region. Therefore, they are
useful as infrared filters and windows. When crystals like alkali halides are subjected to
high energy radiation like X - rays or y rays, they get colored i.e., they absorb light in
visible radiation at certain wave lengths. The absorption is due to 'Colour centres'. The
formation of colorcenters is very important in energy storage utility. Many oxides,
sulphides and alkali halides exhibit luminescence properties if they are subjected to
external stimuli like high energy radiation or electric field.

18
Chemical Bonding and Crystal Structure

Ionic crystals exhibit high strength. They are also brittle. The fracture takes
place along the cleavage planes. The ionic solids are hard.
Ionic compounds dissolve in polar solvents.

1.10 Covalent Bond and Covalent Crystals :

It is known that the word ‘covalent! describes the sharing of valence electrons
by adjacent atoms. Hydrogen molecule is a good example of covalent bonding. The
two protons share two electrons of opposite spins. The electrons lie between the
hydrogen atoms the major part of time. To a first approximation, the covalent bond in
a hydrogen molecule can be considered to arise form the attraction of positive ions to
the intervening pair of electrons with opposite spins as shown below.
H, + H® > H :H (Hydrogen molecule)

Sy
ss

geet Pia
i fe i ; ie

electron pair covalent bond and interaction

In the bonding process of forming the hydrogen molecule the potential energy
of the hydrogen atom is lowered as shown in Fig 1.2(b) curve 3.
The molecules of fluorine, chlorine, oxygen, nitrogen etc., (the diatomic molecules)
are formed on the basis of electron-pair covalent bonds. In these cases, two p-electrons
are shared between the atoms. The above examples have single bonds i.e., their
coordination number is one. But many elements form multiple bonds, Carbon is the
best example Carbon can form four covalent bonds even though it should form only
two-covalent bonds. Sulphur, selenium, tellurium etc., form two bonds; Phosphorous,
nitrogen, arsenic, antimony can form three bonds. The covalent bond is pictured as an
overlap of the electron orbitals of the two participating atoms. Pauli's exclusion principle
explains why only two and noger number of electrons can take part in the bond
formation. According to it, only two electrons that have the same quantum numbers
but opposite spins can pair up. If a third electron has to take part in the bonding, it
would have its spin vector in the same sense as either of the earlier electrons. Therefore,
it will be repelled. Hund's rule states that, for equivalent orbitals, electrons tend not to
share the same orbitals and the electrons occupying different but equivalent orbitals
tend to have their spins parallel.
Nitrogen (outer valence electrons 2S? 2p) has three unpaired p-electrons. It
can attain the neon noble gas electron configuration (2S* 2p°) by sharing three
2p-electrons with another nitrogen atom to form the diatomic molecule.
The most important covalent bonded material is carbon. The carbon is the basic
element for many organic materials like polymeric materials. The carbon atom has the

19
 Applied Physics
a ara a nr DB ee

ground state electron configuration 1s? 2s? 2p’. Therefore, it should form two covalent
bonds with its two half filled 2p-orbitals as per Hund's rule.
However, carbon shows a covalency of four and form four covalent bonds of
equal strength. To explain this, let us consider the electron configuration of carbon.
There are 2s states and 2p states. The 2s states and 2p states are closely spaced and
the energy difference between them is small. Therefore electrons can redistribute
themselves as follows (Fig. 1.6)
ieee Hybridization

Is 2s
2p

IGEdela ig dp ksej skPalopeld ca

Four equivalent half filled sp? orbitals sp? may be hybridized orbital arrangement.

Fig. 1.6 Hybridization of carbon.

The carbon promotes one of its 2s electrons to a 2p orbital, the 2s and three 2p
orbitals merge to form sp” hybrid orbitals. These are elliptical orbitals which are disposed
toward the four corners of a tetra hadron. Each orbital contains one electron. This type
of arrangement is known as hybridization. The process of hybridization enables carbon
to form four equivalent covalent bonds instead of only two. This process leads to the
formation of very stable compounds, or elemental compounds like diamond.
This carbon in the form of diamond exhibits the covalent bonding. The four sp?
hybrid orbitals are directed symmetrically toward the corners of a regular tetra hadron.
This is shown in Fig. 1.7. Each carbon is covalently bonded to four adjacent atoms.
Diamond is extremely hard. It has high bond strength and melting point (3550°C). Its
bond energy is about 711 kJ/mol.

Fig. 1.7 In a Diamond structure, each carbon atom is covalently bonded to four adjecent atoms.
The
bond angle is 109.5°C.

20
Chemical Bonding and Crystal Structure

The covalent bonds are highly directional i.e., stereospecific. The coordination
numbers for covalently bonded atoms are mostly controlled by radii ratios.
The covalent bonds follow the following characteristics :
1. Multiple bonds have shorter bond lengths.
2. Double bonds are stronger than single bonds this aspect is very important in
polymers.
Hydrogen peroxide, propylene etc., are other examples for a covalent bonding.
Silicon and Germanium are excellent examples for covalent bonding. Silicon and
Germanium and semiconductors. Silicon and Germanium have the same structure but
have a finite number of electrons which are able to leave the covalent bond locations.

1.11 Metallic Bonding:

This the third type of chemical bonding that is important in materials. It is

somewhat intermediate between covalent and ionic bonding. Metals are the major
class of materials that are used in engineering. Form scientific point of view also, they
place themselves as the most interesting species among different materials.
The atoms are packed relatively close together in a systematic pattern in metallic
solids. Metallic atoms i.e., those which possesses loosely held valence electrons may
be bonded together into a stable solid.
We shall consider the case of a simple metal like sodium. When two sodium
atoms are brought closer, the unpaired 3s electrons of the atoms from an electron-pair
bond with the spins of the 3s electrons opposed. As a third sodium atom approaches
this pair, its 3s electron will be repelled by the 3s electrons of the other two atoms. This
because of Pauliexclusion principle. However, the energy of the unoccupied 3p states
is very nearly the same as the energy of a 3s state. Consequently, the third electron can
go over into this state without violating the Pauli exclusion principle. In a similar fashion,
a large number of sodium atoms can thus surround any sodium atom. The first sodium
atom or the central atom that has only one unpaired electron takes turns forming
electron-pair bonds with each of the surrounding atoms. Therefore, the sodium forms
less than a whole electron-pair bond with each neighour. Solid sodium solidifies into a
body-centered cubic structure in which each sodium atom is surrounded by eight
other atoms. Therefore, each sodium atom can form one-eighth of an electron-pair
bond, or one-quarter of an electron bond, with each of its nearest neighbour.

Similarly, copper atoms in crystalline copper are so close together that their
outer valence electrons are attracted to the nuclei of their numerous neighbours. Each
atom is surrounded by twelve neighbours (closest). The valence electrons are associated
with all nuclei and are spread out among all atoms. Now, electrons exist as ‘electron
gas' in which ion cores are present.

21
 Applied Physics

The metallic bond is essentially a covalent bond without saturation. The metal
atoms are held together by a mutual sharing of valence electrons. The density of
electrons between the atoms is much lower than that is allowed by Pauli exclusion
principle. Therefore, the electrons move freely as gas molecules without expending
energy. The metallic bond is nondirectional. The coordination number is very high.
The metallic lattice is shown in Fig. 1.8.

me Pt rt # fort ? at “en +4 :
x
“ts
oa
“ht.
“5
“A.
n“A. 48
mn
+ a=
positive ion
Theed Dyoad =
aeod oR —
aesPye
4
aes°
%,
om‘4, core
a 5
p. ry
-
Hm are
fate
S :
es
e

a 2.2“
haegs”
REP : “Tang
OUE : :

: 38, oe B88 s.,

+4 *.> ote °. ig =
peg, TS aH 2 electron
on
OA
“a
cloud
iF
ee
pas
CG 2
oe
a z,
° o>Ve
anes
3,
*' * os
é
o ? oie
CG
RT nes Pike s eh ja tt 3 oe a et, e AP Ps is ei
5 3% ina ate on

Fig. 1.8 Metallic lattice

A large number of metals like A/, C/ Mg, Ti, Ni, Zn, etc., have 12 neighbors for
each atom. Other metals like Li, Na, K, V, Mn, W and Fe (at room temperature) have
eight neighbours for each atom. The binding energy for potassium is about 89.6 KJ/
mol, for Nickel it is 423 KJ/mol, Zinc, it is 131 KJ/mol.

Metals possess high thermal and electrical conductivities. This is because of the
free electron movement in the lattice. Metals exhibit ductility. They are brittle below a
certain temperature. Metals exhibit brittle to ductile fracture transition at a certain
temperature. Many metals like tungsten are tough. But some metal like pure silver,
aluminium, gold are soft and malleable. Their melting points vary from 64°C (Potassium)
to 2000°C (Titanium (1 812°C), Molydenum, Tungsten). Metallic alloys are extremely
useful for all engineering applications. Their high conducting values (gold, platinum,
silver, copper etc.,) make them useful as electrodes in scientific and technical devices.

II Crystal Structure :

1.12 Introduction:

In nature matter exists in three forms, Viz, solids, liquids and gasses. When
gasses are cooled they become liquids (condensation) and on further cooling
become
into a solid (solidification). As this transformation is taking place, the interaction between
the atoms/ions increase and start forming different type of chemical bonds
among
them, resulting in a particular arrangement. This general phenomenon takes
place at
different temperatures for different materials. However Helium is an
exception to this,
which on cooling even to 0 K does not become into a solid under normal
pressure and
requires special technique to make it into a solid. When these materials
elements form

yay:
 Chemical Bonding and Crystal Structure

into a solid, due to thermodynamical energy considerations, (free energy to be

minimum), they either solidify into a crystalline form or into an amorphous form.
Most of the materials tend to attain crystalline form, as this state leads to minimum
energy compared to amorphous form. In crystalline form, the atoms/ions are arranged
in a periodically repeating pattern, where as no such regularity of arrangements is
observed in amorphous materials. Glasses, polymers, metallic glasses come under the
category of amorphous materials. In solid form these materials have higher energy
state compared to crystalline state. Crystals in general melt (transform to liquid) at a
fixed temperature, depending upon the pressure, where as glasses or polymers do not
show such tendency and take liquid form gradually over a range of temperatures.
They first become soft and then start becoming into liquid. This does not happen at a
fixed temperature but over a temperature range. They may show some orderly
(periodic) arrangement of atoms/ions but this exists over very small distances, (few
Angstroms) and hence amorphous materials possess short range ordering. Unlike
amorphous materials, atoms/ions in crystals show orderly or periodic arrangement
over large distances.
A particular crystal is formed by cooling the melts of the corresponding
solid. This process is called crystallization. The melts are in the liquid state. This
liquid state lacks the long-range order but exhibits short range order. In crystals,
the radial distribution of atoms is regular. The density of atoms that are present
around a reference atom are fixed as per bonding and energy considerations. This
is repeated with regular periodicity in three dimensions. This radial distribution of
atoms is indicated in Fig. 1.9(a). The distribution is discrete. No atoms will be
present in between.

radial
density
of
atoms
r
Fig. 1.9(a) Radial distribution of atoms in a crystal over a distance form a reference atom.

Amorphous substances are super cooled liquids. A liquid state consists of

molecules that rotate about their axis of rotation. A liquid can also flow showing
viscocity. Consider the atom of a liquid like melt of a substance. Let p(r) be density of
atoms in a small layer (of thickness Ar) in the melt and py be the density averaged over
the entire melt. Then the ratio p(r)/py represents the probability of the occurrence of
an atom at a given distance r from a reference point or an atom. p(r)/py varies radially
with r as one goes from the reference point as shown in Fig. 1.9(b).

23
 Applied Physics

P(r)/Po

i—e To

Fig. 1.9(b) Radial distribution of atoms in a meld (liquid).

The radial density of atoms varies as shown in Fig. 1.1(b). It has maxima and minima.
It extends to a short distance (rj) and fades off. This pattern of radial distribution of
atoms is called short-range order. It is spread over a few interatomic distances and is
lost after that.
The short-range order in liquid or melt is not stable. There will be groups of
molecules or atoms in the melt. The molecules will be grouping together and breaking
off again. This process goes on continuously. There is a time interval between the
breaking off and regrouping again. This time interval is called relaxation time. The
relaxation time is very short (107!° sec) in low molecular weight melts or liquids.
When such a melt is solidified into a three dimensional network, the process of
crystallization is very fast and the specific volume changes abruptly at the crystallization
temperature as shown in Fig. 1.9(c). The process involves the release of the heat of
crystallization.

melt

crystallisation

specific
volume

solid |
crystalline

0 Temperature —»

Fig. 1.9(c) Specific volume is plotted against temperature on a solid solidifies Q) crystal
amorphous solid.

Amorphous substances are formed by sudden cooling of high molecular

inorganic or organic compounds. Near the point of crystallization, the viscocity and
the relaxation time in the melt are very high. Therefore, it becomes impossible to get
ordered structure or crystalline structure as it converts itself into solid state. This
state is called glassy or vitreous or amorphous state.

24
Chemical Bondingaa
a and Crystal Structure
ee ee nee ercere i
Fig. 1.10(a) and (b) exhibit the two dimensional structure of quartz in two
forms, crystalline form and amorphous form. It is silicon dioxide basically. Both
structures are built on silicon oxygen tetrahedrons (Si O,), each having a silicon and
four oxygens at the vertices. Amorphous silicon has a framed structure. Every Si
O, unit is coordinated with four neighbours. A substance in the amorphous state
possesses shear viscocity that is greater than about 1013 poises. A glass is a super
cooled liquid. A glass or amorphous solid is not in absolute thermal equilibrium,
although a glass is stable over years.
e Silicon atom
© Oxygen atom

(b)
Fig. 1.10 Schematic two dimensional analogs. It illustrates two dimensional structure of
(a) crystalline silicon dioxide (b) amorphous or non crystalline silicon dioxide.

Because of different types of periodic arrangement of atoms/ions in crystals,

they exhibit physical properties which are direction dependent. This phenomenon is
called anisotropy. Most of the amorphous materials show isotropic properties, but
deviations from this rule are often observed, because of preparation techniques and
materials used. In the following section, let us try to analyse more closely the atomic/
ionic arrangement in crystals, as they show very interesting properties, symmetry and
are available in abundance in nature. Even materials which generally solidify into
crystalline form, such as metals (Aluminium and other alloys) can be made as glasses
and are known as metallic glasses. Their properties vary drastically compared to
those in the crystalline form (for the same material). These metallic glasses find many
applications in modern world. As this topic is beyond the scope of this book, let us
leave it here. Large number of materials like ceramics and polymers can be made in
crystalline form as well as in amorphous form.
A crystalline solid can either be a single crystal, where solid consists of only
one crystal (Diamond, Ruby, Quartz etc.), or an aggregate of many small crystals,
less than 1mm, separated by well defined boundaries (grain boundaries). In the later
form, a solid is said to be polycrystalline (e.g.: aluminium wire, iron rod) and hence
the surfaces of these can vary in shape.

1.13 Space Lattice:

A crystalline solid, as mentioned above, shows periodicity (regularity) in the
arrangement of atoms/ions. However this arrangement can be of different types
depending upon the bonding between them such as ionic, covalent, metallic etc. Covalent

25
 Applied Physics
ee ee a OO ee

bond requires fixed number of neighbours (bonds) because of its directionality and
hence results in restriction on the arrangement. However in case of ionic and metallic
bonded materials such restrictions are not there because they are non-directional.
However, the number of nearest neighbours are dictated by their relative sizes. We
will study about this in detail in the sections to follow. However, we shall first explore
the possibilities of various arrangements in crystals. Careful crystallographic studies
(study of crystal, through their appearance/morphology) and X-ray crystal structure
studies (study of interior arrangement of atoms) have revealed that there are only few
ways in which the atoms can be arranged in space (three dimensions). The atoms in
groups are arranged ata particular site in space depending upon the bonding and this
gets repeated with a certain periodicity in different ways. This arrangement in space is
governed by the space lattice which the crystal adopts to. Hence we shall first study
about different space lattices.
Space Lattice :
" A space lattice is defined as an infinite array of points in three dimensional
space, in which every point has an identical environment to all others."
For easy understanding let us first consider a two-dimensional array of points
arranged to form squares when consecutive points are joined in both directions
(X and Y) as shown in Fig. 1.11(a) or to form a hexagon as shown in Fig. 1.11(b).

Atom 1 : (0, 0)
a Atom2: (4%)

Unit Cells

Atom1: (4,4)
Atom 2 : (%,%)
Fractional Atomic (x, y) coordinations (As a
fraction of unit cell dimension) i.e., true dimensions
are ax and by
Fig. 1.11 A two dimensional array points (a) square lattice (b) hexagonal network.

26
Chemical Bonding and Crystal Structure

1.14 Two-dimensional Lattice:

By repeated translation of the two vector say a (along X) and b (along Y) in the
plane of the paper, we can generate a square array (if a = b with an angle between
then 90) and other types of arrays depending upon the values of a and b and the
angular relationship between them as shown in Fig. 1.11 (a) and (b). In Fig 1.11, we
can see different types of geometries forming among the array of points, depending
upon how the lines are drawn. Some ofthem include a point in them [(B), (E)], some
are joined with different angular relationships but joined from a point. We can also
observe in (B) in Fig. 1.11 (a), that the square drawn does not contain any point at the
corners but has only one point at the centre. Each of these shapes formed are denoted
by different lattice names. In a square lattice (or any other shaped lattice) at any given
point in the plane of the paper, the arrangement of points is the same irrespective of
the position in the plane. If we consider any point and look due North or due East
form this point, we see another point at a distance of | unit (of a and b). Along North-
East and South-East, we see the nearest point at a distance of ./2 units and along

North , North-East and South, South-East the nearest point is at distance of AG units.
All these form different set of unit cells in a given lattice, when atoms are arranged in
the lattice a shown in Fig. 1.11(a). Unit cell is the smallest unit ofa crystal, like in the
case of a brick wall. The characteristic of the crystal is represented by the unit cell.
Hence we can define unit cell as :
Definition of Unit Cell :" The smallest component of the crystal, which when
stacked together with pure translation repetition reproduces the whole crystal". It is a
small representative of the big crystal.
Definition Primitive unit cell : " Those unit cells which contain only a single
lattice point" as shown in Fig. 1.11(a) (A), (B) or (C).
In the case of (A) and (C), even though they contain four lattice points at the
corners, their contribution to the unit cell is {' from each point as each point is
shared by four unit cells, which are adjacent as shown in Fig 1.11(a), (A). with a
small circle around a corner point.
Similarly in a three-dimensional lattice, joining of such consecutive (very next)
points would lead to a cube. Each corner of a cube is shared by 8 such cubes in a
three dimensional array of points with a = b = c, the basic translatory vectors along X,
Y and Z respectively. Each point at the corner of the cube would contribute th to
the corresponding unit cell.
From Fig. 1.11(b), we can observe that atoms are arranged in a hexagonal
fashion (two-dimensions) forming a periodic arrangement with a fixed distance between
any two atoms. However, we see in the figure that the same arrangement can result in

ve
 Applied Physics

two types of unit cells. First one has four atoms at the corners, placed at (0, 0), (0, b),

(a, 0) and (b, b) and one atom is placed at (= 2.30]. In this unit cell :
3

]
the contributions from four corners are eX i =]

and contribution from one atom inside the cell is = 1 =]

Total a
These two atoms per the primitive cell (unit cell) are called the Basis or Motif.
Hence a structure of a crystal! is built based on the type of lattice and the basis as
shown in Fig. 1.12 (a) and (b). Here there are three primitive lattices (P) and two
non-primitive lattices (NP). Each of these lattices can form different structures,
when different basis such as one atom, two atoms, three atoms etc. as a group with
different orientations among themselves occupy a particular lattice either P or NP.
Different structures can also result when a single basis is associated with different
lattices indicated in the figure.

Lattice Basis/Motif

SE
S,

Crystal Structure

Fig. 1.12(a) Lattice + Basis = Crystal Structure

28
Chemical Bonding and Crystal Structure

The above can also be understood in a very simple way as follows : Let us
assume a class room where chairs are placed with some regularity across the hall and
also along the hall. The gap between any two chairs in the two directions can be
equivalent to the two translatory vectors (a, b) mentioned earlier. Hence, the hall with
chairs in like a lattice with points. These chairs can be arranged in different ways with
different regularity (or periodicity). This is equivalent to different primitive or non-
primitive latices. In the hall, if the students occupy the chairs each one sitting on a
chair, then each student is like a single atom basis. In this case the centre of the atom
coincides with the point. However, this need not be the case always. Let us assume
that there are double the number of students compared to the chairs available. Then
two students would occupy each chair. If the way they sit is similar in all the chairs,
then we are talking about a basis equal to 'two' occupying the same lattice, but leads
to different structure, compared to the earlier. In Fig. 1.12(a), basis is taken as ‘one’
flower, and it is arranged in a given lattice in two ways resulting is two different
structures. In the first, the basis coincides with the lattice point, but in the second, it
is slightly misplaced. "One should remember that lattice is imaginary where as basis is
a reality as it is made up of atoms. These two, together result in a structure either in
two dimensions or in three dimensions.
Lattice + Basis > Structure

Primitive Lattice

=e

Fig. 1.12(a)

Crystal lattices show symmetry i.e. they can be brought into themselves by
lattice translations T where T =n, a+n,b +n, c and other symmetry operation like
the rotation about an axis through a lattice point. There are lattices in which there are
one-, two-, three-, four- and six fold rotation axes that carry the lattice into itself.
These rotations are the integral multiples of 27, 7, 27/3, 21/4, and 27/6 radians. There
2m
axis
is no lattice which will go into itself by rotation of 2n/7, or 5 radius. A five fold

of symmetry can not exist in a lattice because it is not possible to fill out the space. A
lattice point group is the collection of the symmetry operations mentioned above when

29
 Applied Physics

applied about a lattice point leave the lattice in the original condition. We can have
mirror reflections about any plane passing the lattice points. There can be an inversion
operation which is a rotation by x radians followed by reflection in a plane normal to
the rotation axis. The number of lattices that can ultimately be formed will be limited
by these symmetry considerations. In two dimensional space. There are five distinct
lattice types.

1.15 Three-Dimensional Lattice or Space Lattice or Bravais Lattices :

As in the case of two-dimensional lattice, in a real crystal which is a three

dimensional configuration in space also has a structure. This structure is built
ona
three dimensional lattice, with a specific 'basis' or 'motif. A three-dimensional space
lattice is generated by repeated translations of three non-coplanar vectors
a, b, and c.
As in the case of two-dimensions, one can generate many lattices in three
dimensions
with different primitive and non-primitive cells. One would expect
many such lattices
in three-dimensions. However, Bravais, showed that maximum number
of independent
space lattices are 14 only, the figure fourteen is arrived at on
the basis of point
symmetry groups. These 14 space lattices are known as
Bravais lattices, named
after its originator. These 14 lattices are shown in F ig 1.13.
Note that spheres shown
CUBIC
a=b=c
fd 6= y aS 90° Li “J

TETRAGONAL
a=bec Pp
a= p= y= 90°

ORTHORHOMBIC
aebec P
Q= p =y= 90°

HEXAGONAL
TRIGONAL
a=bec
g= B=90°" P
a=b=c¢ P
y= 120°
o=BP=y490°

MONOCLINIC
azbec Pp ATS es of Unit Cell
a= y= 90° / me = Primitive
B # 90° /

TRICLINIC tle
ig +

aebec P 7 Crystal Classes

ae Beye 90° oe —» 14 Bravais Lattices

Fig. 1.13 Crystal Systems and the Corre

sponding Bravais Lattices,

30
Chemical Bonding and Crystal Structure

in the figures are not atoms but represent lattice points. These 14 lattices are further
grouped on the basis of common symmetry among them, based on the dimensions
of the unit vectors along the axis and the angular relationship between them. Each of
this is called a system and they are seven systems in number. These seven systems
along with the correspondings lattices are shown in Fig. 1.13. Notations of lattice
parameters a, b, c, a, B, y are the shown Fig. 1.14. a, b, and ¢ are the unit cell edge
lengths and a, B, y are angles between the axis as shown.
Z

ygine
oe ee,
jet)

Fig. 1.14
Table 1.1 : The 14 Lattice Types in Three Dimensions

Number of Lattices| Lattice symbols Restrictions on

Convetional Cell
Axes and Angels
Triclinic 1 at bs ¢
COA 6 i

Monoclinic Z PoC ae 0+ C
a=y=90° #8
Orthorhombic 4 P.C,Ick ab ec
a =f =y 90°

Tetragonal 2 a= be
a=B=y 90°

Cubic 3 P or sc
I or bec
F or fcc

Trigonal ] iva — Dae

a=B =y< 120°, + 90

le ae

31
 Applied Physics

Table 1.1 gives the details of all the systems and the corresponding lattices 4 types
of unit cells are shown in Figures 1.14, 1.15 and 1.16. They are primitive (P), Body
centered (I), Face-centered (F) and Side-centered or base -centered (C). Except for
'P' all the rest of the unit cells are non-primitive. The means that, they contain lattice
points at the corners, at face centres (F), body-centers (I) and base-centres (C).
Therefore, there are

4 types unit cells (P, I, F and C)

+ 7 systems = 14 lattices.
1.16 Crystal Systems : Cubic
Fourteen Bravais lattices are grouped into 7 crystal systems as mentioned earlier
but are not divided equally among the systems. For instance, a cubic systems has
three lattices namely
(a) Simple cubic (Primitive) points are present only at the corners.

Fig. 1.15 Simple cube. The atoms are at the corners.

(b) Body-centered cubic (BCC) (non-primitive) : Points present
at corners and
also at the centre of the cell.

Body-Centered Cubic

BCC

Fig. 1.15 Body-Centered Cubic structure with projec

tion in 2-dimensions.

32
Chemical Bonding and Crystal Structure

(c) Face-centered cubic (FCC) points present at the corners and also at the
centres of the faces.

(b)
Fig. 1.17 Face centered cubic lattice.
These three lattices have the characteristics of the most symmetric system,
i.e., cubic. Cubic system is characterised by the lattice parameter relationships
a= beac
and a=B=y=90°
That is, any crystal which belongs to cubic system has only one independent
characteristic that is its lattice parameter 'a', since a= b = c and the angular relationships
are already known. Hence for cubic crystals only 'a' has to be obtained. However in
the case of other systems and the respective lattices, other parameters should also be
specified. For instance, for crystals belonging to tetragonal system, we require two
parameters '‘a' and 'c'. For crystals belonging to triclinic system, which is the most
unsymmetric system, all the lattice parameters have to be obtained for complete
information about the crystal.
Most of the metals solidify in cubic system, and have either BCC or FCC
lattice. There are some metals which solidify in hexagonal structure, however we will
only take up cubic system.
If only a single atom becomes the basis, then the resulting structure would look
like the lattice itself and hence such structures can be called by the name of the lattice.
For instance, in the case of metals such as aluminium, copper, lead, platinum, gold
etc., they crystallize into FCC lattice with one atom at each point and hence the
structure of these atoms is called FCC structure.

1.17 FCC Structure:

In this structure, atoms are located at each of the corner and centres of all the
cube faces, and hence called face-centered cubic (FCC) structure. (Fig. 1.16) and
Fig. 1.17 show a hard sphere model for FCC unit cell, where as in Fig. 1.17(b) the
atom centres are represented by small circles as representative of the atoms to give a
clear picture of the structure with respect to atomic positions in the unit cell.

33
 Applied Physics

As in Fig. 1.17(b), the atoms touch one another along a face-diagonal. The
length of the diagonal is 4R, where R is the radius of each atom. The unit cell has a
volume av. ‘a’ is the cell edge length. From the Fig. 1.17(b), we see that the total
length from one corner to its opposite corner is 4R and hence
a’ + a* = (4R)
or qe R

This is the relationship between the edge length 'a' and the radius of the atom 'R’.
The volume (V_.) of the cell in terms of 'R' can be given as

Vo=a= bv/2) (ry = 16/2 R3

Similarly we can also obtain the nearest neighbour distance. "Nearest neighbour
is that which is closest to a given atom and most probably would be in contact with
it." In the case of simple cubic structure, that is the structure formed from primitive
cell of cubic, the atom which is nearest are those which are placed at the corners.
Hence the nearest neighbour distance is 'a' as given in Table 1.2. However, in the case
of FCC, the atoms which are in contact are those which are along the face diagonal of
the cube and not the corner atoms. Therefore the nearest distance between the center
of the atoms (nearest-neighbours) is 2R or when expressed in terms of lattice parameter

'a' (edge length in cubic) is TAs i.e., 0.707 a, as shown in Table 1.2.

Table 1.2 : Characteristics of Cubic Lattices

Body-centered
Volume,
conventional cell a> a3 a3

Nearest-neighbour
distance

Packing fraction*
 Chemical Bonding and Crystal Structure

Effective Number of Atoms in FCC Crystal :

In FCC, these are 8 atoms at the corners and 6 atoms at the face centres of the
cube. As mentioned earlier, all the atoms ina unit cell do not completely belong to that
cell but are shared by the adjacent cells in a solid, as the unit cell is not isolated but is
a part of the solid. Hence the contributions of the atoms to a given cell can be calculated
as follows :
No. of atoms at the corners : 8
No. of atoms at the face-center : 6

Contribution from each corner to the cell consideration is : ye

(as shown in Fig. 1.17(b)

1
Contribution from each face centre to the cell is : my

(as shown in Fig. 1.17(b)

'. Effective total number of atoms belonging to the FCC unit cell is

Hence effective number of atoms or points belonging to FCC structure or

lattice is 4 respectively. Since the volume of the cell is 'a*', hence lattice points or
atoms in a FCC structure per unit volume is (4/a*) as shown in Table 1.2.

Atomic Packing Fraction in FCC Crystal :

In the case of FCC, the coordination number (number of nearest neighbour,

and the atomic packing fraction (APF), i.e., the ratio of the volume actually occupied
by the atoms to the volume ofthe unit cell (i.e., volume available to be occupied by the
atoms). In the case of metals, with single species of atoms, the coordination number
is 12. The student can verify this easily in Fig. 1.19. Let us try to calculate the APF for
ECC.
Fh Volume of the atoms in the unit cell Sey.
eS Total volume of the cell = ve

Vy is the filled volume of the cell by spheres occupying it, as it is assumed that
the atoms are hard spheres with radius R.

we know the value of V< is 16/2 R3.

32
 Applied Physics

[79r’
APE = ere = 0.74
Vo gore
For the FCC structure, the atomic packing fraction is 0.74 or 74% of the space
in the unit cell is occupied. This is the highest packing fraction that can be achieved
where spheres (atoms) of same diameter are packed. For all other structures APF is
less, as can be seen from Table 1.2. Hence 'FCC' structure is also known by the name
"cubic close packed structure" (CCP).
FCC structure as mentioned has the highest packing fraction. This is made
possible due to a particular type of arrangement of similar atoms of single radius. Let
us first see in two-dimensions how best such hard sphere of equal radius can be
placed to give good packing fraction. As shown in Fig. 1.18, there can be six nearest

Fig. 1.18 Single layer 2-Dimensional arrangement of sphere.

neighbours for each space as marked | to 6. Rest of the spheres are placed at longer
distance than '2R', where 'R' is the radius of the sphere. 2R is the minimum distance
between the centres of any two spheres. One can not improve any further packing.
However, if we expand this in the third dimension by stacking each of these layers one
over the other, such that centre of all the atoms in each layer passe through an axis in
the third dimension, then the number of nearest neighbours become eight instead of
six. However this can further be improved by stacking them not exactly one over the
other, placing the second layer of atoms in the gaps left out in the first layer between
the surface of the adjacent spheres as shown in Fig. 1.19. This type of stacking leads

(a)

Fig. 1.19 A portion ofa close-packed plane of atoms, with A, B and

C marked for different
positions that can be occupied is in different layers, layer 'B' above 'A' and
layer 'C' below 'A',

36
Chemical Bonding and Crystal Structure

3 layers

Fig. 1.19 A-Dimensional arrangement of atoms by stacking one over the other leading
to HCP de CCP
to either HCP (Hexagonal close pack) or CCP (cubic close pack) depending upon the
position of the third layer a shown in figure. This type of stacking we commonly
observe in fruit stalls where sweet limes (Mousambi) are kept one over the other. If
you carefully observe the arrangement, you see twelve spheres (Mousambis) touching
a single sphere. Hence the number of nearest neighbours in FCC is twelve and gives
the maximum packing fraction when similar type spheres are stacked. The twelve
nearest neighbours are those which surround the atom under consideration in the
same layer (A) (as shown in Fig.1.19(a) and those which are present in the layers just
above (B) and below (C) that layer. There are six atoms in the same layer (Fig. 1.19(a))

Bi
 Applied Physics

and 3 each in layers just above (B) and below (C)(Fig. 1.19(c)), making a total of 12.
Many of the metals with metallic bond show above arrangement. Alkali halides crystals,
such as NaC/, etc., with ionic bond have FCC lattice with different basis leading to
different structures and the coordination number may differ depending on the ionic
sizes of the cation and anion.
Referring back to Fig. (1.17a), we can describe the FCC structure by writing
the coordinates of the centers of the ions in the given unit cell. If assume one ofthe
bottom corners 'O' as the origin, then the coordinates of the ions (1, 2, 3 and 4) could
be written as

eel ‘lia | 1 1
(0,0, 0), € 5 ),
—,—,0 0,—-,— *] and c
a0, =*),

where the distance between two consecutive ions along any axis is unity. All the rest
of the ions in the cell could be obtained by displacing these four by a factor of 'u' the
lattice constant. For instance, it we substitute 'u' for one of the coordinates in (O50!
0), we get (u, 0, 0), which is the coordinate of the ion '5'. Similarly the coordinates of
ions 6 and 7 are (0, 0, u) and (0, u, 0) respectively. Similarly adding 'u' to each one of
the coordinates of the ions 2, 3 and 4 we can generate all the rest of the points. Hence
FCC structure is-fully described by the coordinates of the four points | to 4. In the
lattice, even if the origin is chosen at any orbitrary given point, we get the same result,
because of the symmetry and periodicity of the lattice.

1.18 BCC Structure:

BCC structure as shown in Fig.1.16, has 9 points, eight at corners and one at
the center. The 2-dimensional projection of the structure is also shown in
F ig 1.16
l lagler I
where, at the centre, a point is marked c
a} This point has coordinates of Z4 >?5]

l
when measured from one of the corners. The marking (5)indicates,
that this ion is

; l
displaced by 5 from the bottom along 'z' direction. The projection
that we see is that
when the cell is observed from the top. Then we see only '4'
corners, which have four
more points below them, and one at the centre. Hence, this
unit cell can be generated

[ alee
with the repetition of coordinates (0, 0, 0) and (Z4
> 3],

38
Chemical Bonding and Crystal Structure

Fig. 1.20 For the body-centered cubic crystal structure, (a) a hard sphere unit cell
representation, (b) a atomic site unit cell.
This structure is commonly observed in metals such as Cr, Fe, W etc., when
the atoms are assumed to be hard sphere, it appears like the one shown in Fig.1.20(a).
The atomic site model is shown in Fig.1.20(b). As can be seen from Fig.1.20(a), the
centre and corner atoms touch one another along the cube diagonals, and the unit cell
length is 'a'. The atomic radius is assumed to be 'R'. From the Fig.1.20(a), we can get
the relation between 'a' and 'R' as

Effective number of points per unit cell :

Number of points or atoms per unit cell in this case is '2' and can be obtained as
follows:
No. of atoms at corners : 8

No. of atoms at the centre : 1

Contribution of corners to the unit cell +g ih Sly

8

Contribution of the atom at the centre of the cell + 1 = 1

Total="2
Therefore number of lattice points per unit volume for this structure in (2/a3).
Number of nearest neighbour (coordination number) in this cell is '8'. For
instance the atom centre of the cell has 8 neighbours, which are at the 8 corners and

39
 Applied Physics

all of them touch the surface of the atom at the centre. Any atom in the lattice can be
chosen to be at the centre and the situation described above would be repeated.
Since the radius of each atom is 'R', the nearest neighbour distance would be
2R, which is the distance between the center of the atom situated at the corner and

av3
centre of the cell. But R is related to 'a' through R = a hence the nearest

neighbour distance in BCC = or

av3 = 0.866 a.

This indicates that the ions at the corners of this cell do not touch each other.

Atomic Packing Factor (APF) :

It is given as

Volume of atoms in unit cell _ As

APF = : =
Volume of unit cell Ve

S x
4 TG 4aR) 64 3
3 > Venn & () bee

81 3
—R
Niprma ties
64 R3
apn8 ae
3v3
The above result indicates that the atoms are less efficiently packed in BCC
compared to FCC, which has APF as 0.74. Reduction of APF indicates that the volume
of the unit cell is un-utilized completely and hence more space is left out un-occupied.

1.19 Simple Cubic Structure :

This structure is generally denoted as SC and is the most simplest among all the
structures. This is formed when single similar atoms occupy the eight lattice points at
the eight corners of the cube as shown in Fig.1.15. This structure is based on the
primitive cell of the cubic system.
In this structure, the distance between any two nearest neighbours is the lattice
constant 'a', which is the edge length of the cube. As in the case of other structures,
we can obtain the nearest neighbour distance, coordination number, number of lattice
points (atoms) per cell and APF. They are as follows,

40
Chemical Bonding and Crystal Structure

(i) Lattice points (atoms) per unit cell

l
8
ee l

(ii) Relation between the lattice parameter 'a' and radius of the atom 'R’ is
a.=-2R,

as the atoms along the edge touch each other Fig.(1.15(b)). Hence the
nearest neighbour distance is 'a' = 2R
(iii) Number of nearest-neighbour : 6

For any given atom at the corner, there are six nearest neighbours, each
one along + X, + Y + and + Z directions at a distance of'a', the edge length.

T
G)MAVEL=A a Se SST = 52)
Ve (QR) 6

Hence we observe that simple cubic structure has the least packing fraction
among the three structure FCC, BCC and SC.
Metals which crystallize under this structure/lattice are very few with one atom
as basis. However, there are many materials which have SC lattice with 2 or more
atoms as basis. CsC/ is one such example. But this structures is known as CsC/
structure and not as SC. We will discuss about it later.

1.20 Lonic and Covalent Cubic Structures :

In the previous sections, we have discussed about the cubic structures such as
FCC, BCC and SC which are normally found in metals, in which metallic bond in
often observed. However there are many other crystals which are non-metallic and
have either ionic or covalent bonds or a combination of them. In the structures so far
discussed, we have assumed that the similar atoms occupy all the lattice points. The
packing fractions have been calculated based on the above assumptions. However,
we see in nature, different minerals and crystals with combination of ions of different
ionic sizes. These combinations lead to different structures under different
crystallographic systems. However, as nature loves symmetry many crystals are found
to have either cubic or tetragonal symmetry. In the following sections let us discuss
about different crystal structures named after natural minerals available in nature,
such as diamond, zinc blends (ZnS), Rocksalt (NaC/) and CsC/. Each of this show
cubic symmetry and follow one of three Bravais lattice viz., FCC, BCC or SC.

4]
 Applied Physics

These structures are basically dependent on two aspects (i) the relative radii of
the ions that combine to form a crystal (ii) their ionic nature (positive or negative),
covalency and electrical neutrality of the crystal. The above two conditions lead to
different types of coordination geometries in ionic solids with coordination numbers
2., 3, 4, 6 and 8. These coordination number are determined by the charge and the
size of the ions that take part in bonding leading to a structure. The above coordinations
are determined by the cation (r,), anion (r,) radius ratio, r-/r, as shown in table 3.

Table 3

Coordination
number

< 0.155

> 0.155 and < 0.225

> 0.225 and < 0.414

> 0.414 and <0.732

SO0.792 e510
Coordination numbers 5 and 7 are not observed due to restriction on crystal
symmetry. These coordination numbers lead to structures such as NaCl, CsC/,CaF,
,
etc. Similarly some crystal structures are formed based on the covalency coordination.
Examples for such structures are Diamond, Zinc blende (ZnS). Here bonding
is either
covalent or a combination of ionic and covalent bonds.

Let us first take up the cubic diamond structure which is based

purely on
covalent bonding between the carbon atoms.

1.21 Diamond Structure:

In crystalline form 'diamond', one of the important precious

stones, is available
in two structures. It crystallizes in cubic system as well
is hexagonal system. We
shall discuss about the diamond crystal which has cubic
symmetry. Its structure is
commonly known as Diamond Cubic (DC). As mentioned
earlier, carbon atoms occupy
the sites in FCC lattice with basis as '2'. That is, there
would be 8 effective atoms in
each of the diamond unit cell. This is observed in figure
1.21, where the covalent

42
Chemical Bonding and Crystal Structure

bonding among the atoms is shown. There 14 carbon atoms situated at the corners
(8) and face centres (6). Apart from those occupying the FCC lattice points, there are

(a)

Fig. 1.21 (a) Unit cell of Diamond cubic structure with the tetrahedron formed by
the carbon atoms. (b) A view of the structure when seeing from the top (along z-
zxis) of the unit cell, with positions of the ions measured from the bottom.

4 carbon atoms situated along the 2 body diagonals at a height of 1/4 of lattice parameter
from the bottom and 3/4 of lattice parameter again from the bottom on the other two
body diagonal which are opposite to each other. Each of the carbon atom has
coordination number (nearest-neighbour) as '4', unlike in the case of FCC structure in
metals where the coordination was '12'. Hence the coordination number varies from
structure to structure even though the lattice is the same. Coordination number in the
case is restricted by the number of bonds that can be formed.

If we calculate the packing efficiency of this structure, we obtain it to be 0.34

which is much less than that any of the structures discussed so far, namely SC, BCC
or FCC.

Let us calculate APF in the diamond structure.

43
 Applied Physics

(a) Effective number of atoms in the Unit cell

!
(i) Corners > 8 x 8 l

(ii) Face centers Sar 5l II es)

(iii) Atoms fully inside the cell -—» 4 x ] 4

Total = 8

4n 3
(b) Volume of each atom = —R™ = ;
3 3

aesituated at a distance a
Since the nearest neighbour to any atom is av3 which
is the diameter of the atom. It we take any atom at the corner, say P, then the atom 'E'
is situated on the body diagonal passing through 'P’, at a distance Ve of the distance

along the body diagonal. Length of body diagonal is aJ3 (a— lattice parameter) and

sean ay 3e
y, of it is a Hence, only those atoms at a distance of a (distance between
the centres of the atoms) touch each other and all others are at distances greater than
this.

ais iam autre Sa

Hence we see that among the cubic structures, packing efficien

cy is least in
this case, because the number of nearest neighbours is dictated
by the number of
bonds and not by their size. This indicates that most of the volume
of the cell remains
unfilled. The tetrahedron formed is shown in Fig. 1.21(a) with PQR
as base and 'H'
as vertex. Same is also marked in the form of a imaginary
cell ABCDEFGH, where the
vertex of the tetrahedron is at 'D' and BFG forms the base.
Other fourth column elements in the periodic table such as Si,
Ge and grey tin
also have diamond cubic structure. As the size of the
ion increases with increase of

44
Chemical Bonding and Crystal Structure

atomic number for Si, Ge and grey tin, the lattice parameter for these crystals also
increases.
Finally we can conclude that FCC lattice with a basis of 2 carbon atoms results
in diamond structure. -
However, instead of two carbon atoms (similar atom) as basis, if we choose
two different types of atoms, one each, then the same FCC lattice results into another
type of structure called zinc blende (ZnS).

1.22 Zince Blende Structure :

The cubic zinc sulphide which has two types of atoms, crystallizes into FCC
lattice with basis as two. These two atoms are 'Zn' and 'S’. Hence the structure would
appear to be different from that of Diamond. The resultant structure, known as zinc
blende structure is shown in Fig. 1.22(a). In this, we observe, that, one type of

(a) (b)
Fig. 1.22 ‘ZnS’

atoms (14) occupy the 14 FCC lattice points and the other type of atoms (4) occupy
positions inside the cell along the body diagonal at the position as that occupied by
carbon is diamond cubic structure. Hence two-dimensional projection Fig.1.22b shows
one type of atoms occupying the corners, edge centres and the centre of the cell,
where as the other type of atoms occupy places along the face diagonals of the
square. It we imagine that sulphur atoms occupy the FCC positions, then Zn atoms
would occupy the sites on the body diagonal. If we imagine exactly in the opposite
as
way, Zn in Fcc and S inside the cube, the resultant structure would be exactly same
the previous one. We can observe that each atom (Zn or S) on the body-diagonal
is
forms four covalent bonds with the other type of atom. That is each Zn atom
to be
bonded to four S atoms, and vice versa. Most often we observe the bonding
show
highly covalent in crystals exhibiting this structure. The other materials which
45
 Applied Physics

this structure are ZnTe, SiC, CdS, CdTe, GaAs, InSb to mention a few. Because of
the covalent nature of the bonds, the materials which crystallize with these (diamond
or ZnS) structures have very high hardness and are used as abrasives. Particular
examples ofthis are diamond and SiC.

The coordination number, the APF and other parameters in this structure would
be same as in the case of diamond structure.

1.23 Sodium Chloride (Rock Salt) Structure :

Sodium chloride (NaCJ), that is mineral Rock salt, crystallizes in fcc lattice
and
it has two atoms as basis, one forming a fcc structure with C/ at (0, 0, 0)
as origin and
another fcc structure forming with origin at (1/2, 0, 0). It has 4 molecule
s of NaC/ in
one unit cell. The coordination number is 6 for each of the ions
(cation and anion).
Cations and anions are topologically identical, that is the structure
remains identical
when a cation or an anion is replaced with an anion or cation
respectively at all
lattice points.

(b) =

Fig. 1.23 (a) actual structure of NaC] (b)

reduced ion-shell model of NaC] (c) two-
dimensional projection of reduced ion-shell model
.
The fcc nature of lattice can be observed
clearly by looking at Fig. 1.23(b),
where two types of ions (big and small) are
shown. If we consider the big ions as Ci,
then they occupy all corners and also the
face centres of the cube as in the case of
structure (lattice). The small ions are fcc
of Na and they occupy the positions
half-way between any two C/ ions. This is exact ly
necessary to keep the electrostatic repulsion
minimum. Each of cation (Na) is surrounded
by 6 anions (C/) and vice versa. Therefor
e

46
Chemical Bonding and Crystal Structure

the coordination number is ‘6’ for each of the ions. As shown in Fig. 1.24, sites
between any of the lattice positions in a fcc can be imagined to be as a space (vacancy)
left out and in this case it is known as octahedral site. If structure is built with 6 ions
surrounding any cation or anion, the structure will have 8 triangles in the form of
octahedron and hence name octohedral site. The ion (cation) is at the centre of ‘6’ an
ions as shown in Fig. 1.24, since the space left out between the six anions has a radius

Fig. 1.24 Critcal radius for various coordination numbers. The most stable structure
is ususlly the one with the maximum coordination allowed by the radius ratio.
of 0.414 times the radius of anion, leading to a coordination number of ‘6’. In
Fig. 1.23(c) a projection of all the ions in the structure as seen from top is shown. The
outer dark ring refers to bigger ion (anion) and inner fainter circle refers to the small
ion (cation). The vertical positions from the base are marked. The mark, 1/2 indicates
the position of the given ion at half the lattice length from the bottom.
Looking at Fig. 1.23(b), we may come to the following conclusion about the
structure :
(a) The structure has two fcc lattices inter-penetrating into each and each lattice displaced
by half the lattice constant along the edge of the cube.
or
(b) The structure has a fcc lattice with basis or motif as 2, i.e., one ‘Na’ and another
‘CI’ ion. Each of these (cation an anion) are displaced from each other by half the
lattice constant along the edge.
or
(c) The structure has fcc lattice with anions occupying the lattice points and the
octohedral sites (the space left out between any two fcc points) which are situated
midway along the each edge and also at the center of cell are occupied by cations.
All the above conclusions lead to the same structure.

47
 Applied Physics

1.24 CsC/ Structure:

Another structure commonly observed among many ionic solids is that of CsC/
structure. This is shown in figure 1.23(c). The structure appears very simple, with C/
(anions) ions occupying the corner of cube (unit cell) and at the centre we observe Cs
ion (cations). The two dimensional projection is also shown. Fig. 1.22 shows a continous
network of Cs and C/ ions in the crystal, with one unit cell marked. Each of the
Cstions is surrounded by 8 Cr ions, leading to a coordination number of '8'. From the
network, we can also construct an unit cell with Cs* ions at the '8' corners of the cell
and one and C/ ion at the centre of it, again making the coordination number as '8'.
Therefore we can look at this structure as two inter penetrating simple cubic structure

3 ;
(lattice) displaced by = distance along the body diagonal of the cube.

“Important :One generally gets confused by looking at CsC/ structure and

assumes it to be of BCC type. This is not correct. In a BCC structure all the lattice
points, i.e., 8 corners and the one at centre are occupied by the SAME atom". _In the
above structure, only corners are occupied by Cr and the centre is occupied by the
Cs* ions or vice versa. Hence CsC/ belongs to a simple cubic lattice with basis as '2'
(two atoms one Ci and one Cs*). The crystals that show CsCl structure are CsI.
CsCl structure is not a close packed structure like fcc structure.

Important Conclusions : A structure is built on a lattice with a basis. Hence the

same lattice can lead to different structures, depending upon the number of atoms
associated with a lattice point and also their mutual placements with respect to each
other.

Sometimes a structure could also be viewed as a combination of two

types
interpenetrating lattices, as in the case of NaCl, Fluorite, CsCl etc.

Examples
(Ie Calculate the densities of (a) MgO and (b) SiC, starting with the
radii of ions or
atoms.

To calculate the density of a solids, knowledge about the structur

e, the volume
of the unit cell, number of formula units in the unit cell, atomic
weights of the
ions or atoms is required.

Weight of the ions within the unit cell

density)
p (density) = Volume of unit cell

48
Chemical Bonding and Crystal Structure

p= Ne (Mc +Ma)/VoNay

where Np = number of formula units within unit cell

Mc = Sum of atomic weights of all cations within the unit cell

»M, = Sum of atomic weights of all anions within the unit cell
Vo = Unit cell volume ; Nx = Avogadro's number

(a) MgO:

MgO crystallizes with NaC/ structure, and hence the ions Mg?* and
O-~ touch each other along the edge of the cube, as they are the nearest
neighbours.

the lattice parameters 'a' = 214g + 21g = 424 pm = 424 x [07 em

radius of Mg*? 1d= 86pm (pico metre or 10-!2m)

radius of O-? = 126 pm
Atomic weight of Mg = 24.31 g/mol

Atomic weight of O = 16 g/mol

There are 4 'Mg' and 4'O' ions in the

lattice as there four formula units, since
it has fcc lattice with effective number 0-2 O-
of lattice points as '4'.
4(24.31+16)
7 ( a 2 4 x1 0 7 ) = 3.51 g/cm}
(6.0221 0
(b) SiC:
In the case of SiC which has ZnS structure the atoms touch each other

along the body diagonal with length a./3 . Therefore the distance between the
centres of these two (Si, C) atoms (which are the nearest neighbours) in equal
to 1/4" length of the body diagonal as explained in ZnS and diamond structures.
Radius of Si atom = 118 pm (covalent radius)
Radius of C atom = 71 pm (covalent radius)

V3
pee or a = 436 pm

49
 Applied Physics

There are four C and four Si atoms in the unit cell, with molecular weights 12
and 20.09 g/mol respectively.

4(12 + 28.09
p (density) = | ( = 3.214 g/cm?

The density of SiC is lower than that of MgO, even though their molecular
weights are almost same. This is a consequence of large lattice parameter resulting
from more open structure of covaiently bonded solids.
Ze On the basis of ionic radii, predict the structure of 'FeO'.
FeO forms basically an ionic structure with ionic bonding. The type of structure
would be of AX type, i.e., one cation and one anion. The cation and anion radius
ratio in this case is

Tre? 7 0.077nm
ey Otomo
This value lies between 0.414 and 0.732 and as given in table '3', and hence the
coordination number is '6'. Hence the predicted crystal structure is that of NaC/, which
has coordination number as '6'.
Note :
Some times the rule of coordination is violated if the bonding is not of pure ionic
but a mixture of ionic and covalent.

Questions
1. Explain the types of interatomic forces.
2. Mention various types of bondings that can exist between atoms.
3. Explain how the force of attraction between two atoms
or ions vary as they are
brought closer.
4. Explain what is ‘ionic bonding’.
5. When two ions are brought closer, what is the nature
of energy components
between them?
6. What is Coulomb energy? How does it vary with the
distance between ions.
7. What do you mean by ‘repulsive force’ between ions?
How does it arise?
8. Derive the expression for the lattice energy of an
ionc crystal like NaC/.
9. What is Madelung constant?
10. What is coordination number? Give examples.
11. How do you calculate Madelung constant?

50
 Chemical Bonding and Crystal Structure

12. How is Madelung constant is evaluated?

. How are compressibility of a solid and Born's exponent are related?
. What do you mean by ‘ionic radii’ and crystal radii?
. Explain Haber cycle.
. Explain covalent bonding? What is Hybridization.
. Explain the bonding in diamond. Give some examples for covalent bonding. In
what way covalent bonding is different form ionic bonding?
. What is a metal? Explain the meaning of electron gas. How is it formed?
. Give the general properties of ionic solids.
. Show the minimum cation/anion radius ratio for a coordination number of 6 is
0.414.
24, Which interstitial site is larger ? the tetrahedral or the octohedral? Calculate the
ratio of the sizes of the tetrahedral and octo hedral sites.
es How many tetrahedral and octohedral sites are there in a unit cell of cubic close
pack structure.
24. Draw the zinc blende structure. Compare this structure with that of diamond
structure.
2: Calculate the nearest neighbour distance between two carbon atoms in a cubic
diamond structure.
26. What is the difference between the space lattice and a structure.
Tt. Describe the important features of the three lattices which belong to the cubic
system.
28. If the atomic radius of aluminium is 143 pm, calculate the volume of its unit cell.
Aluminium has FCC structure.
29) In CsC!/ structure, calculate the distance between the nearest cation and anion.
30. Show that the atomic packing fraction for BCC is 0.68.
Bt: Iron has BCC crystal structure, with atomic radius of 124 pm and atomic weight
of 55.9 g/mol calculate the density.
ope Calculate the coordination number and APF in the case of ZnS structure.
33: What are the important factors that determine the structure of a material or a
crystal?
34. Demonstrate that the minimum cation to anion ratio of 0.732 corresponds to the
coordination number of '8'.

5]
 Applied Physics

3). Compute Atomic packing fraction for CsC/ structure.

36. What are the Bravais lattices of NaC/ and CsC/?
37: How many lattice parameters are required to completely specify a crystal which
belongs to monoclinic system ?
38. What are Bravais Lattice ? Describe them in combination with crystal systems.
ao: Calculate the atomic packing fraction of diamond structure.
40. Calculate the atomic packing fraction in a simple cube.
41. Silicon has the diamond structure. The unit cell has edge a = 0.542 nm. What is
the nearest interatomic separation in it? (refer to page 45)
42. Write the characteristics of seven crystal systems.
43. What are crystal ? Give examples. Explain the formation of an ionic crystal.
44. Sodium chloride crystal has fcc structure and its density is 2.18 gm/cm?. Calculate
the distance between two adjecent atoms. Obtain the expression for cohesive
energy.
45. What are different space lattices in the cubic system.

22
ZA
CRYSAAL DEFECTS

2.1 Directions in Crystals :

It has been said in the earlier chapter that a crystal is a regular array of circuits.
There exists some smallest grouping that repeats itself exactly in three directions. This
means the environment of any atom or ion at one particular atomic site identical to the
environment at a corresponding position or location somewhere else in the crystal. The
crystal possesses properties that are characteristic of chemical bonding, and the nature
of atoms or ions. A crystals contains atomic planes and the number of atoms in a plane
depends on the demands ofthe chemical bonding and the equilibrium conditions in the
crystal. Many crystals exhibit anisotropic properties i.e., the properties are dependent
on the direction in which they are observed or measured and the direction of the
external stimuli to the crystal. The external stimuli can be an electric field, magnetic
field, mechanical stress or heat flux etc., For example, an fcc nickel crystal exhibits
greater magnetic permeability in the diagonal direction of the cubic lattice than that in
other directions like the crystallographic axes. Therefore, it is essential to fix up reference
directions in crystals.

53
 Applied Physics
Da a —

For cubic crystals, the crystallographic direction indices are the vector
components of the direction resolved along each of the coordinate axes and reduced to
the smallest integers. Let us see how the directions in a crystal can be arrived at......

A lattice vector is given as r = ua + vb + we and this vector connects a lattice

point to the origin. The extension of this vector by a repetitive distance will find another
identical point at its end. The atoms in the crystal have fixed points.

Consider a cubic cell that is placed on crystallographic axes as shown in

Fig. 2.1. Let '0' be the origin. The point P is at the position (a, 0, 0). 'a’ represents one
unit distance. Therefore, it is at the position (1, 0, 0). The direction of 0 P is given as

Fig. 2.1 Some directions in a cubic unit cell

[1 0 O], and that ofP 0 is given as (1 0 0]. A number with a bar indicates a negative
direction and index.

Similarly, points Q and R are in the direction [0 1 0] and [0 0 1] respectively.

They are basic crystallographic directions in a cubic unit cell.

Now, consider the direction joining O and S as shown by the arrow. The position

: P 1 Nt ; : : ] 1
coordinates of S are (5. a 0|
. The direction of0S is [2x re x +0 |Led lel).

Suppose the direction [] 1 0] has to be shown. The direction of the arrow on 0S is

reversed as in Fig. 2.1(b) i.e., SO.

Consider a point T shown in both the figures 2.2(a) and (b). The position
coordinates ofT are (4. We Re ) The point T is the body centre of the cell.. The

54
Crystal Defects

direction of T O is [1 1 1]. It is extended to end of the line to meet the corner of the
cubic cell. [1 1 1] is the direction of the cube diagonal. The direction in the opposite
sense gives (1 1 1) direction i.e., TO.

Consider the position of a point shown in Fig. 2.2(b).

(b)
|
PI, 272

The coordinates of L are (4, 1,0 ).The direction of OL is [1 2 0]. To get the

direction, you have to multiply all integers by LCM of the denominator in position
coordinates. The direction [2 1 0] is also shown in the figure.

[3 2 1] direction
(a) (b) (c)
Fig. 2.3 (a) , (b) and (c)

Let us find out what are the points that are connected in a cubic cell to show the
direction [1 1 2], [1 1 2], [321], [3 2 1]. Consider the figures in 2.3(a), (b) and(c)

55
 Applied Physics

The positional coordinates are obtained by dividing all indices by the largest

: aes al bene
number. The positions of the first two points are (4: Mes i), (4.2%. 44)

, ‘ ‘
respectively. Let O be the origin. Then the point (4; Vs 1)isatr. the OPis| 1 F2]

as shown in Fig. 2.3(a). Similarly, the [1 1 a is shown in the Fig. 2.3(b). The family
of <1 1 2> directions pass through the face centers of the cube with the origin at one
of the corners of the opposite face.

[3 2 1]direction
Fig. 2.3(d)

The point (.Ye We is Q in Fig. 2.3(c). The direction is [3 2 1]. To show

[3 2 1] direction, we have to consider the origin or shift the origin to O'. The point

Q now is at é 1, Dy Yh ).Multiply by the common factor in the denominator, we

get (-3, 2 -1) ie., [3 2 1] direction along O' Q shown in Fig. 2.3(d). It should be
remembered that the origin has to be shifted to another corner of the cubic cell to
include negative directions.

The letters u, v, w are used to indicate the general sense ofthe direction indices
along a, b, ¢ (or x, y, z directions) respectively. The directions are written as [u
v w].

All parallel directions or direction vectors have the same direction indices. The
figure below shows the parallel direction of [1 1 0]. Similarly,, [0 1 0] or
[1 0 0],
[0 0 1] have parallel directions. The parallel directions are equivalent.

56
Crystal Defects

Fig. 2.4 Parallel directions in a unit cell of acube

The family of (100) directions are [0 1 0], [001], [100], [0 1 0], [0 0 1],

[1 00]. All of them are groupedas (100). (u v w) represents the family ofdirections.
Similarly (110) directions, (111) directions represent the family of
[1 10] and [1 1 1] directions. They are all present in the cubic cell.

Let us see how a direction like [7 3 1 ] can be shown. Consider the cubic cell
given in Fig. 2.5 below. There are two negative directions. The origin has to be selected
so as to include these negative directions.
: zZ

la ;
> (eal

VA,
Af

Fig. 2.5 The direction OP is gi below the upper plane and A before the backside

(0 1 0) plane.

The position coordinates of P are e Ie y,4 ay.

We have selected a-axis (i.e., x-axis), and c-axis to include the negative directions.
Therefore, the origin is at O'. The direction is shown in the figure. The direction OP is
35.
57
 Applied Physics
a er a rrr

22 Crystal Planes and Miller Indices :

There are several planes possible in a crystal. There are certain rules to be
established for determining the coordinate system for the planes. The position of the
lattice points and the crystal axes are used for the purpose of indexing the plane systems.

A plane in the three dimensions is described from the analytical geometry as

The plane in three directions in space is intercepting the x—, y— and z— coordinate
axes at u, v and w. We may write (2.1) as

hkeeky-i=le fe ee eee. C223

1 ]
Where 2 = Fc k= g and / = <The A, k and / represent the reciprocals of the

corresponding axial intercepts. Therefore, we can conveniently use h, k, and / to

represent the planes or identify the planes. A crystal can be imagined as a composite of
a large number of planes oriented in different directions. The atomic density on these
planes is different. The Fig. 2.6 shows the various planes that can exist in a crystal
lattice.

Fig. 2.6. Some lattice planes are shown. A crystal can be imagined to consist ofa large number of
planes.

58
Crystal Defects

The following procedure for determining the Miller indices for a cubic crystal
plane is followed.
1. First, choose the origin such that the plane does not pass through origin.
2. Find the intercept of the plane in terms of the crystallographic a, b and c or x-, y-
and z-axes for a unit cell. These intercepts may be fractions. Let these intercepts
be u, v and w.
b 4 l
3. Take the reciprocal of the intercepts (= and ~),
4. Clear the fractions with the LCM. of the denominator integers. Then determine
the smallest set of whole numbers which are in the same ratio as the intercepts.
These whole numbers are Miller indices. They are written enclosed in
parenthesis as (h k J).
(a) The figure 2.7(a) shows some of crystallographic planes of cubic crystal
structure. Consider the planes shaded in the figure 2.7 (a).

(a) (b) (c)

(d)
Fig. 2.7. Depicts various planes in a unit cube cell.

a9
 Applied Physics

The intercepts of (1 0 0) plane are 1, «9 ©. By taking reciprocals of the

]
intercepts, we get +> 0, 0,

., the Miller indices are (100).

This plane that cuts the x-axis at one unit distance i.e., the lattice parameter.
The other plane (001) and (010) are also shown in the figure 2.7(a).
Consider the planes given in the figure in Fig. 2.7. (b), (c), (d), (e).
(b) The intercepts of the plane are 1, 1, oo (along x, y, z- axes). Since the reciprocals
of these numbers are 1, 1, 0 which do not involve fractions. The Miller indices
of this plane are (110).
(c) The plane has intercepts 1, 0, 1.
reciprocals 1, 0, 1.
Hence, the plane is (1 0 1)

(d) To find the intercepts, first move the origin to the lower-right back side of the
cube (O')
The intercepts are (-l, 1, «)
reciprocals —1, 1, 0

The plane is (1 1 0).

Find what will be the Miller indices if the origin is shifted to O".

(e) The intercepts are (oo, -1, 1)

reciprocals 0, -1, 1

Miller indices (0 1 1). z

Consider the planes given in Figs. 2.8 in the following.

(a) The intercepts are 1, 1, 1.

reciprocals |ain eat

Miller indices (1 1 1)

the plane is (1 1 1) plane.

Fig. 2.8(a)

60
Crystal Defects

(b) | Consider the plane in Fig.2.8 (b).

The intercepts are dy a l

ie

Reciprocals Bal

Miller indices (2-2: 1)

the plane is Cea) az ee

(c) Consider the plane shown in Fig. 2.8(c). For

this plane, we have to shift the origin to O'.

The intercepts are 1, 1, i

2
Reciprocals ee Wee3
The plane is Cizts).
Fig. 2.8(c)

(d) Let us find out the Miller indices of the plane that is marked in Fig. 2.8 (d).

: iam 8
The intercepts are 799

Reciprocals 2, 2,2

The plane is (2 2 2)
This is parallel to (111) plane. Fig. 2.8(d)
i
(e) In the Fig. 2.8 (e) the plane gives the intercepts at 2/3, 1/3 and | on x-, y— and
Z— axes respectively.

The reciprocals are : “)

3
1 > |
Ww
|N

Multiplying by common LCM. of the

denominator, we get 3, 6, 1. Then the
x
h, k, / values are 3, 6, 1. Hence, the
. Fig. 2.8(e)
plane is (361) plane.

61
 Applied Physics

Suppose we have to show 7, I, | plane one has to take the intercepts | 2

Nl]
S|
Then, we get the (7 | 1) plane.

Fig. 2.8(f)
Exercise :

Take a face centered cubic lattice, mark (i1 i)and (i1 1)

The f.c.c. unit cell is shown in the Fig. 2.9(a). Whenever a negative index has
to be
found, one has to shift the origin appropriately to include the negative directions
on the
axis.

For (i] i)plane, one has to take the origin at 0'. The plane is shown
in the
figure. The p { 1 1 1 } planes contain the < 110 > directions as sides
or edges.

For (1 i)plane, the intercepts are 1, -1, —1 respectively.

The two planes are parallel.

Fig. 2.9 (a)

62
Crystal Defects

Exercise :

Observe the unit cells in the Fig.2.9 (b) and mark the (1 1 1) and (i] 1)planes.
What is the common edge between them ?
The planes are marked in the fig. 2.9 (b).
The common edge is [0 1 | ].
The two planes intersect at [0 1 1 J.
They are the twin planes in the crystal.

As an excercise draw (111) and (111)planes in the cube.

The {111} planes in a face centered cubic crystal are densely packed.

[011]

Fig. 2.9(b)
Sometimes, it is important to determine the atomic densities on various crystal
planes. The planar atomic density is determined by the following relation.
Planar atomic density :

p the area of the plane

Equivalent number of atoms whose centres are interested by
‘ The area

For example,
the copper has an fcc structure.
We have seen in the chapter | that the equivalent number of atoms intersected
by the (110) plane in terms of the surface area inside fcc cell is 2 atoms.

The area of the plane is /2 a’.

63
 Applied Physics

Then, the number of atoms in (110) planes is

Piio -= Seon 2 =1.09x107 atoms/ m2

V2(3.61x10"m)
= 1.1 x 10!'2/mm?.

Exercise :

Calculate the planar atomic density on the (110) plane of the bec molybdenum
that has a lattice constant 3.147A°.
Planar atomic density :

2 atoms

V2 (3.147 x10~ m)?

7 10
~ 1.414 x 9.904

Saat
: 5%
KESPA
14

= 1.42 x10! atoms / m2

23 Inter Planar Distance:

Consider first a two dimensional lattice as shown in Fig. 2.10.

Let the two
planes under consideration be A B and C D. One of them
C D is passing through
origin. Let the intercepts be a/h and a/k along x and y-axes
respectively. Drop a
perpendicular on to the plane A B.

Fig. 2.10 One unit cell ofa square lattice.

64
Crystal Defects

From the Fig. 2.10, ON = d.

OA = a/h ; OB = a/k.
Then, the interplanar spacing = d.
We have, from the figure, triangle OAN and OAB,

ON
OA
_0B
AB

or

or
2k
This can be extended to three dimensions. The two planes in the three dimensions
can be represented interms of Miller indices {h k / }.
aaa
Suppose Wilke are the intercepts on x, y, z - axis of a cubic crystal.

Then, the equation of the plane is

x
eee ay
a ovaryee “ra sh
or hye tky Fiza
The next parallel plane for this plane will have intercepts (see Fig. 2.8)
a a a

in the crystal.
ie £
2
=) (= | (=
a
|on x,
ae
y,
<7
Z axis
==

_ Then, the equation of that plane is

de
eahenlytenol heey bares
2(avh) 2(a/k) 2(a/1)
Lee nx ky = 2a:
This distance between the above parallel planes d,,, is given as
Hig 2-1
hk/ _ 2 si )

a a a

a
 Applied Physics

2.4 Imperfections in Crystals:

2.4.1 Defects in crystal :

We have treated crystals in the earlier chapters as perfect periodic structures. A

‘perfect crystal' is an ideal concept, but real crystals that are grown and used contain
time -averaged deviations from perfect structure. The defects in crystals are such
deviations. The deviations from a perfect three dimensional periodicity of a crystal are
known as lattice defects or imperfections. The lattice,imaperfections are of many types.

A crystal! at absolute zero possesses atom’ the n in their motion. Above

the absolute temperature, the atoms in the cry bout their mean positions.
That means, there is a disturbance or perturbat tperiodic structure which
can vanish again when the temperature is taken B bsolute zero. The vibration
of atoms is a mechanical vibration. The quanta of mechanical vibrational energy is
called ‘phonon’. At temperatures above absolute zero, the vibrations will increase ie.
the number of phonons will also increase. A phonon is a dynamic defect in crystals. Its
existence depends upon the external excitation, if the excitation has ceased, they
disappear. There are many such transient defects such as electronic excitations, exitons,
photons, crowdions, magnons etc. They are created by external excitation. These
imperfections are called the imperfections of first kind. In the present chapter, we are
not interested in such transient defects. We are interested in the imperfections of second
kind which are geometrical defects and arise as a result of thermodynamic reasons and
mechanical reasons. These imperfections are extremely important as they have a definite
role to play in most of the bulk properties of the crystals in some temperature range
or
other. They are structural imperfections. They are classified as 1) Point defects-(atomic
imperfections), vacancies, interstitials, and impurity atoms. They are zero
dimensional
defects. 2) Linear defects such as dislocations, 3) Two-dimensional or planar
defects
such as stacking faults and_ internal grain boundaries. 4) Volume
defects or three
dimensional defects, such as voids or pores, impurity precipitates,
foreign inclusions,
cracks etc.

Point Defects : The point defects are classified as 1) Vacancies 2)

Interstitials
3) Disorder due to impurity atoms.

Vacancies : A vacancy is the simplest defect in a crystal. A vacancy is an atom

site from which the atom is missing. It is known as Schottky defect. Consider a
crystal lattice as shown in the Fig. 2.11. A Schottky defect is created by the removal
of an atom from an atomic site (like position V, or V,) in the crystal and transferring it
to the surface.

66
Crystal Defects

O SC) ah Satan A. i ome 3)pooh O Spee)

i
UGE) WATE ON UATE AIS ON Ore HS’
<— Frenkel

O v,O O PAGi ve F O defect

© OO Onto
0O 0 070" 0
Fig. 2.11 A plane of a metal crystal. V, and V, are the vacancies created. F is the Frenkel Defect.

If a crystal is given an energy in the form of heat, this energy is distributed

among the atoms of the crystals. However, the energy distribution is non uniform. At
any temperature, some atoms which have higher energy than their neighbours will be
able to move considerable distances from one equilibrium position to another and
ultimately go over to the surface. It appears as if the atoms are able to 'vaporise' from
their sites and condense elsewhere. The site left by the atom is a vacancy, called
Schottky defect. The number of vacancies per unit volume of the crystal depends
upon the temperature. The atom that has left the place can go over to an interstitial site
'i' (called interstice) or to a place which can be far away from it. When it goes over to
the next or the nearest interstice, a vacancy-interstitial pair is formed. This vacancy-
interstitial pair is called Frenkel defect (F). The question comes; how does the atom
that is present inside can go to the surface ? At any temperature, the surface of a
crystal is partially evaporated. An atom from the surface evaporates and it leaves the
surface in vapour state. Then an interior atom will move into it leaving behind a vacancy
forming a Schottky defect.

2.4.2 Equilibrium Concentration of Vacancies in Metals :

We shall calculate the density of Schottky or Vacancies as a function of
temperature. We shall do the following assumptions :
1. The crystal consists of N identical atoms. The volume of the crystal is
independent of temperature.
2. There exists an energy of formation for a vacancy (E,). It is independent of
temperature.
3. The formation energy is the energy difference between the energy needed to
remove an atom from the interior of the crystal and take it to infinity and the
energy gained when the atom is placed on the surface of the crystal.
4, The vacancies are indistinguishable particles.
5. The lattice vibrations are not affected by the presence of defects.

67
 Applied Physics

Let 'n' be the number of Schottky defects at any temperature T. The necessary
equilibrium in the crystal at any temperature T is reached thermodynamically when the
free energy F is minimum. The incorporation of a defect in a crystal increases the
entropy of the crystal. It also increases the internal energy of the crystal. There are N
atomic sites in the crystal. The N sites can be filled by N atoms in N! ways. Out of the
N atomic sites, 'n' sites have to be vacancies. The 'n' vacancies can be arranged among
themselves in n! ways. The remaining (N-n) sites are occupied are occupied by atoms.
The atoms can be arranged among them selves in (N-n)! ways. But, they are all
indistinguishable. Effectively, the number of ways in which we can create n vacancies
in the crystal out of the N atoms will be equal to NC,.

e N!
Then Wet ee (N—a)tat

"W' also represents the probability of a distribution in which n vacancies can be crated.
The change in entropy as result of the creation of vacancies is given as
St ikiias Wire oh Sei bigs Serial eS Geet ae (2.6)
where 'S' is the entropy, k is the Boltzmann constant. The above relation is called the
Boltzmann relation between entropy and disorder. The disorder stems from the number
of ways in which 'n' vacancies can be created. (/n refers to log.)

!
S = kin {Gg Sanyal MI (2)

But, according to stirling approximation

Bog Nie Nv NG&-ANT ig! GL FN Ta) eA ere (2.8)

then, eqn(2.7) can be written as

S=k[N/nN-N-(N-n)Mm(N-n)+N-n-ninn+n]

=k(NmN-(N-n)/n(N—-n)-nMgQ] ._.......... (2.9)

In the formation of vacancies, a change in free energy occurs as a result of
change in internal energy and a change in entropy. The internal energy is given as

Ueenk Ss to” 2) ee ee (2.10)

where E, is the energy required for the formation of a vacancy. Then
F =) VE Peers PG Sere ee Gb eee (@.L1)
then, F = nE; —kT[N »N-(N-n) /n(N-n)-n Mo]

68
 Crystal Defects

At equilibrium, the free energy has to be minimum. This is the law of nature.

Then 1
5 F = E,d5n—kT[-(N -n) x (— Sn)
(N =n)

~ In (N~n) (-8n) ~ ~6n— inn (n)]

—OnjE,-klm(N-n)—-Ion| —— = mle)
At equilibrium, SF = 0, then

eee
kT In Nau)
ai a em (2.13)

N >> n, N will be in the order of 1073 and n will be in the order of 107 per unit
volume. Then, we can have

Sad a0
(2 A a an inane (2.14)
'n' is equal to the number of Schottky defects per unit volume. The concentration
of vacancies in metal depends on temperature. As the temperature of the crystal is
raised, the concentration increases rapidly. The formation energy of a vacancy depends
on the nature of the metal. It is a characteristic of a metal. For example, in the case of

copper, the factor = ~ 10-!? at 300 K. But at 700, and 1350 K, it is 10-8 and 10+
respectively.

The formation of vacancies at a temperature is such that there is thermodynamic

equilibrium at that temperature. One can increase the concentration of vacancies in a
metal at room temperature by heating the crystal of the metal to a high temperature and
then suddenly cooling it (this process of sudden cooling is called quenching) to room
temperature. Then the crystal would possess the vacancy density that was there at
high temperature. Because of sudden cooling, the vacancies are frozen in the crystal.

2.43 Frenkel Defects :

We can calculate the equilibrium concentration of Frenkel defects under the

same assumptions as above. Let the energy necessary for the formation of a Frenkel
defect be E, A Frenkel defect consists of a vacancy and an interstitial atom, the
change in entropy consists of two components.
1. n vacancies are created in the crystal out of N atoms. The expression of entropy
is same as in eqn. (2.6).

2. The n atoms from vacancies can be distributed in W, different ways over the N,

69
 Applied Physics

interstitial sites or positions in the crystal. The corresponding contribution to

entropy change is /n W,.
. the total increase in entropy due to the formation of Frenkel defects will be then

As in the earlier case, we can apply stirling approximation and obtain the condition
for minimum free energy.

N! N;!
in———_+/ n >
(N-n)!n! (N;=n)!n!
~NinN+N,
nN; —-(N-—n) /n (N—-7n)
—(N;—n)
In(N; —n) -2n inn

Substituting the above in the relation for free energy, we get

aE
ea = B= Klin ae vise (2.16)
At equilibrium

FAOn )+ a

and N >> n and N,>>n

then ae In ee ea. Sven (2.17)

: Ey J NN;
es KT 2/n :iteas ia haleeind aameaiiaiae (2.18)

or he= 4/NN exp Coe ck) nnn eeerer ete (2.19)

It is possible that some atoms can go to interstial sites (Fig. 2.12). But many
metals are close packed structures. The interstatial sites have small volumes. So that
any atom coming into the interstatial produces lot of strains. The order of energy

70
Crystal Defects

required for the formation of an interstatial is much higher than that required for vacancy
formation. For example, copper has a fractional concentration of vacancies at 1000°C
of about 10~° while the fractional concentration of interstatial atoms in copper at 1000°C
is quite low, about 10-39!

The vacancies migrate in metals. The energy required for this migration is
about | eV for in copper while the interstials require 0.16 eV for migration i.e., the
interstials would be more mobile than vacancies in metals. But, the vacancy
concentration in a particular metal is far higher than that of the interstials (formed due
to dislocated atoms). The diffusion process of atoms (like impurity atoms) in metals is
strongly dependent on vacancy concentration.

In metals, both vacancies and Frenkel defects are formed, even though, the
energy favours the vacancy formation. The interstials and atoms can combine and
form clusters. Similarly, vacancies can coalesce or combine to form voids. A void is
a three dimensional defect. They would have minute spherical structures in crystals
like the bubble structure in a glass paperweight.

2.5 Iloniccrystals:

In ionic crystals also, the point defects are formed. The vacancies (Schottky
defects) and Frenkel defects exist in ionic crystals.

2.5.1 Vacancies :

These are two types of vacancies in ionic crystals. 1) the cation vacancies
(positive ion vacancies) and 2) anion vacancies (negative ion vacancies). A positive ion
vacancy is formed when a positive ion from the interior of the crystal moves out of the
crystal to the surface as shown in Fig. 2.12. When a positive ion vacancy (V,)is
formed inside a crystal, the vacancy possesses an effective negative charge. This
results in an excess negative charge locally inside the crystal. The crystal tries to
maintain charge neutrality. Therefore, a negative - ion moves to the surface to maintain
charge neutrality inside the crystal. The effective charge of a negative - ion vacancy
(V,) is positive. Thus, a negative - ion vacancy positive-ion vacancy pair (P) is created
in the crystal. The Schottky defect in an ionic crystal is an anion vacancy - cation
vacancy pair (V, and V,). We are assuming that they carry equal charges. It should
be remembered that in a crystal like MgC/,, where the magnessium ion is doubly
charged, each vacancy at magnessium sub-lattice is compensated by two anion vacancies
in Chloride sites. In NaC/ crystal, each vacancy of sodium ion tends to result in the
formation of one chloride ion vacancy. A positive ion vacancy and a negative ion
vacancy on adjacent sites (P) can form a neutral point defect.

71
 Applied Physics

Fig. 2.12 Formation of vacancies in a pure ionic crystal like an alkali halide. A layer ofions is
shown. V, is a positive ion vacancy. V, is a negative ion vacancy. A coupled pair of a positive ion
vacancy and anion vacancy can also be formed (P). It is neutral.

Suppose a crystal contains a total of N ions and that n schottky defects are
created by removing n cations and n anions from the interior of the crystal. The
different ways in which each kind of ion can be removed is given as
N!
Ww = Reni 2 salacimigeotenst (2.20)

As the defects occur in pairs, the increase of entropy, is given by

This in turn causes a change in Free energy. Let E, be the formation energy for
a pair of vacancies, then,

[= Te oS

Fn Ey = kT in nom
nol ous
phiaties (2.21)

N!
SIE okt Un (N—n)in!

Sno 2k TiN! In N!
— 7n (N —-n)1 —- nn!]
~n E, —2kT[N InN—N-(N-n
nnn]
)
~nE,—2kT[N nN-N-(N-n)Mm(N—-n)n/n o)....... (2.22)

We
Crystal Defects

At a temperature T, the Free energy is minimum and constant. Taking the first
derivative of the above equation, we get,

(=) =E, — 2kT{[/n(N =n) = inn]

On )+

gt= eS 1= n)
in N=) bblide ad in ehdihc; (223)

or E, = 2kT in ~

N-n
or mere (Eg) 2 eee 2SE are (2.24)

The number of Schottky defects n is much smaller than N i.e., n << N, then .

ne Neo a are (2.25)

The energy needed to remove a pair of ions (of opposite charge) is about 2eV in
NaCl. The number of ions (pairs) is about 102? / cm3. Normally, the Schottky defects
in NaC/ at room temperature are of the order of 10° / cm3. Hence, n << N.

2.5.2 Frenkel Defects in Ionic Crystals :

A Frenkel is formed when the ion does not go to the surface, but, is removed to
an interstitial site. The interstitial site is not normally occupied by ions. Only in the
formation of a Frenkel defect, it does happen. A Frenkel defect is shown in Fig 2.13.
The expression for Frenkel defects in ionic crystals is given by the expression
(2.19) as
ie CDi
NG JUS eX ie 2K) a scre necate (2.26)
where E, is the formation energy for a Frenkel defect in ionic crystal.
In ionic crystals, the concentration of vacancies is much easily formed than the
Frenkel defects. In crystals like alkali halides (NaC/, KC/, KBr, NaBr, LiF, LiCl etc.,)
the vacancies i.e., Schottky defects are predominant defects. Anions are of larger size
than the cations. Therefore cation vacancies are predominant mobile defects in alkali
halides. These vacancies are charged defects in ionic crystals. A positive ion vacancy
has an effective negative charge. A similarly, an anion vacancy has an effective positive
charge. They can join and exist as a neutral pair.

73
 Applied Physics

Fig. 2.13 Frenkel defect (F) in ionic crystal. A Frenkel defect is created when the ion is removed to
an interstitial site.

In Silver halides like AgC/, the dominant defects are Frenkel defects. Even
though all kinds of the above defects are present in ionic solids, it is observed that a
particular type of defects tends to dominate.

2.6 Impurities:
A crystal can never be hundred percent pure. There are always some unavoidable
impurities present. They are regarded as point defects. The impurities can occupy two
different sites in the lattice as illustrated in Fig. 2.14(i). The positions or sites are a)
The substitutional sites, in which the atom of the parent lattice site is replaced by an
impurity atom. They can also replace the existing vacancies, b) The interstial positions
in which the impurity atom is at non-atomic site.
The impurities produce lattice distortion at localized portions as shown in
Fig.2.14(ii) below.

OOOO Neal See.

OOD 0CO0O:
18)Gc GiB OG,
>
ae
(a}

Fig. 2.14 (i) Substitutional impurity atom, (ii) interstitial impurity atom.

Even the vacancies or interstitials produce localized lattice distortions as shown

in the figure 2.14(ii). There are stresses and strains around the defect and these
stresses diminish inversely proportional to the third power of the distance from that
centre. An interstitial atom increases the spacing between the surrounding atoms by
20%. Therefore, the energy of formation for an interstitial is far greater than the energy
required for a point defect like vacancy.
For example, the energy of formation of a vacancy in copper is around
0.96eV, where as 2.5 - 3.5 eV energy is required for the formation ofan interstial
atom. The parent atom can occupy the interstitial positions, but this is not
a
very favourable defect.

74
Crystal Defects

Impurities can be added intentionally to a crystal to alter its properties or to get

required properties for use. The addition of impurities to the crystal and its acceptance
to the lattice is based on Hume-Rothery rules. The addition of impurity to the crystal
results in substitutional solid solutions like copper-nickel, copper-gold alloys. These
alloys are called substitutional solid solutions. Many alloy systems belong to this category.
It is necessary as per Hume-Rothery rules, that the impurity atoms and the parent
atoms should not differ in size by more than 15%. The crystal structures like the
copper and silver alloy, should have the same crystal structure. Preferably the atoms
of the alloy elements should be either electropositive or electronegative. The addition
of impurity can be done to the extent to which the solubility of the impurity in the
crystal is allowed at equilibrium temperature. Interstitial solid solutions are formed
between two elements interestitial positions in the solvent cystal. Some metals can
accommodate carbon atoms interstitially to form carbides (iron carbide). The impurities
alter the properties of the crystals. The alloys of metals possess enhanced strengths,
resistivities, or give different properties as compared to their individual components.
For example, one atomic percent of copper dissolved in silver increases the electrical
resistivity by 7%. Silicon and germanium doped with acceptor or donor impurities
give extrinsic semiconductivity. Similarly mechanical properties of crystals can also
be changed by the incorporation of impurities in lattice. Steel is an alloy in which the
change of impurities (carbon, manganese, chromium etc.,) or the concentration of the
impurities gives various grades of steels.

2.7 Study of existence of points defects in crystals :

The experimental proof for the existence of defects comes from the study of
properties with variations in temperatures of the crystals. The formation of Schottky
defects at constant temperature result in an increase of the crystal volume. As a result
of the increase in volume, the density of the crystal is altered. The difference between
the measured density and the x-ray density is a measure of the Schottky imperfection's
concentration. The x-ray density is given by the atomic mass per unit volume, the
number of the molecules in the cell and the volume of the cell. Frenkel defects do not
change the density of crystals. The expansion of the solid with increasing temperature
is partly due to vacancy creation.
The variation of thermal expansion with temperature or specific heat with
temperature yields information about the point defects.
Atomic imperfections are responsible for the diffusion of the impurity atoms or
parent atoms in the crystals. Point defects possess mobility and they migrate from one
point to another point resulting in the transport of atoms or ions in the crystal. Therefore,
the study of diffusion of impurities gives information about point defects in crystals.

75
 Applied Physics

2.8 Linedefects:
2.8.1 Theoretical Shear Stress :

All materials possess mechanical properties. The crystals undergo elastic

deformation under low stresses and the plastic deformation under high stresses. They
possess mechanical properties like stress-strain relationship, plasticity, yield-stress,
flow-stress, fracture, fatigue etc. All these mechanical properties are closely linked
with a very important type of imperfections called 'dislocations'. Dislocations are line-
defects in crystals and are one dimensional defects. We shall first see how the concept
of dislocation was introduced in crystal physics.
Imagine a perfect crystal that has atoms arranged in regular order schematically
shown in Fig. 2.15. Let 'a' be the distance of separation between atoms. Suppose, a
shear stress is applied along x - direction. Let the stress applied be t.

(A)

()

Fig. 2.15 (A) Displacement of atoms when a shear stress 7 is

appled (B) the stress variation in the
crystal.
The atoms in the upper row are sheared with respect
to the lower adjacent
plane. Let this displacement be'x'. A crystal possesses periodi
city. Due to the symmetry
of the lattice, we have

T= 0. sat CS Kaye) Sm eee sie (227)

where | She Ui be eee eeeenee

The Fig 2.15(B) shows the variation of stréss in the cystal

due to the periodicity
of the crystal.

76
Crystal Defects

At points where t > 0, the lattice offers resistance to the applied stress at
OAx ay 2.
Ta emeal. 6 alo
< aera ace he, (2.28)
Then, the stress in the crystal can be expressed as

Pap lack
"1, sin.
a

Suppose the displacement 'x' is very small which is the case in crystals.

Then, Pea A ROW Ace wale. pals ya eu A (2.29)

a

For small displacements, Hooke's law is valid.

Xx
Stress't =G x =| Ay tee (2.30)

Where 'G' is the shear modulus. In such a case, we get

te ICG 8 ae (2.31)

The above value for tT, is the critical stress that is needed to initiate deformation

: ; : G
of acrystal. More rigorous calculations give the theoretical stress Tt, as 30 for copper,

silver, gold etc., However, the observed values are substantially lower than the calculated
values by several orders. For example, in the case of aluminium single crystal, like the
shear modulus is 2.5 x 10!! dyne/cm? Then, the calculated ty (elastic limit) should be
8.3 x 10? dynes/cm?. But, the observed value is about 4 x 10° dynes/cm?, much lower
that the theoretical value. Similarly, in the case of silver, the theoretical value is about
9 x 109 dynes/cm2, but the observed value is 6 x 10° dynes/cm?. The striking difference
between the calculated and the observed (experimental) values was accounted by the
presence of dislocations and their movements in the crystal. The shear or slip of a
portion of a crystal relative to another portion was introduced as the result of a dislocation
in the crystal. The presence of dislocations, their movements, interactions with one
another and other defects explain the mechanical properties of crystals. Let us now
take up the models of dislocations.

Te
 Applied Physics

There are two types of dislocations. They are 1) edge dislocations 2) screw
dislocations. We will discuss the configurations of them.

2.8.2 Edge Dislocation :

Taylor, Orowan and Polanyi were the first scientists who independently in 1934
suggested that the slip in a crystal propagates by the motion of dislocations. We shall
now describe edge dislocation from the geometrical point of view.
Let us take a simple cubic crystal. The atoms in the crystal can be imagined to
be held by flexible springs. The Fig. 2.16(a) describes the perfect crystal. Make a cut
along CA. Push an extra plane inside as shown in F ig. 2.16(b).

Fig. 2.16 (a) Perfect crystal (b) A linear defect occurs in region just
above ' | '.
Then one would notice that an extra plane is placed in the upper
half of the
crystal. Let this plane be ABCD. As a result the atoms in
the neighbourhood of the
plane ABCD get displaced from their positions as shown in the
Fig.2.16(b). The edge
of the extra plane AB is called 'edge dislocation’ indicated
by 1. Looking into the
paper, the region around the edge of the plane is distorted.
The line at the edge of the
extra plane is an edge dislocation. It is contained in
the plane which is a boundary
between slipped and unslipped (lower half) portions
of the crystal. As the extra plane
is in the upper half of the crystal, the dislocation is called
the positive edge dislocation.
The point at which the dislocation would lie perpendicular
to the plane of the figure is
better shown in Fig.2.17(a) where only the front surface
of the crystal is shown. A
negative edge dislocation is formed if the extra plane
is in the lower half. (shown in Fig.
2.17(b)). Whena positive and negative dislocation
are on the same plane, they annihilate
each other when they meet as a result of stress.
If they are on different planes, they
may create vacancies or interstitials when they
meet. A vacancy creation is shown in
Fig 2.18.

78
Crystal Defects

Fig. 2. 17 (a) Postive edge dislocation ( | )(b) Negative edge dislocation

vacancy

SSchesesacarces
SRSlTEOeso0one
ees 6 SHSSRRRE

(b)
Fig. 2.18 When negative and positive dislocations on different slip planes move under stress and
meet, (a), vacancy is created (b)

Cutting plane

(a)
Fig 2.19 (a) Formation of a screw dislocation. A perfect crystal is cut along AB into the crystal and
sheer stress are applied parallel to the cut plane. (b) A screw dislocation MN is formed, burgers
vector is parallel to dislocation.

79
 Applied Physics

2.8.3 Screw Dislocation :

Consider the perfect crystal once again. Make a cut at AB with the help of a
cutting plane. Now, apply a shear stress parallel to the cutting plane such that the
right(front) portion of the crystal moves down with respect to the left portion of the
crystal. The resulting arrangement is shown in Fig. 2.19(b). A screw dislocation is
formed along MN at the point O. The shear stress applied causes a distortion of the
crystal lattice in the form of a spiral ramp of distorted atoms around the screw
dislocation. The portion or region of distorted crystal is atleast a few atoms in diameter.
The strength of any dislocation is described interms of a parameter called
‘Burger's Vector’. We shall discuss the concept of Burger Vector and Burger Circuit in
the following section.

2.8.4 Burger's Circuit and Burgers Vector:

Let us first take up the edge dislocation. In a perfect crystal, the atoms are in
perfect equilibrium in their positions as determined by the lattice energy. All bond lengths
are in equilibrium state. By creating a dislocation, the bond lengths are disturbed as the
atoms are displaced from their original positions. The bond lengths have to be compressed
first. Therefore the portion of the crystal which contains the extra plane is under
the
state of compression. Just below the edge in the bottom portion of crystal, the atoms
are
displaced apart and therefore, the lower portion is under tension. The bonds are
stretched.
Hence, the potential energy is increased locally. That means there is an extra
energy
locally at the region of the dislocation due to the distortion of the lattice.
The region of
distortion spreads a few lattice distances around the dislocation. The maximum
distortion
is in the immediate region surrounding the edge of the extra plane.
The distortion represents actually the edge dislocation. The plane
in which the
dislocation is present is called slip plane. Over this slip plane,
we have the slipped
portion of the crystal. Below the plane, it is unslipped portion. Therefo
re, the dislocation
forms a boundary between slipped and unslipped portions.
This is true for a screw
dislocation also.
We consider the perfect crystal as given in Fig. 2.16(a). Start
from the point P.
We go down by n steps (n= 4) then take m steps to the left
(m= 5), then n steps up
and finally m steps to the right. We reach the point where
we started. This movement
around a centre is called Burger circuit. We reach the
starting point in a perfect
crystal. The Burger's circuit is closed perfectly in
a perfect crystal.
Now consider the Burger circuit ina crystal (shown in
Fig. 2.16(b)) that contains
edge dislocation, once again Starting from P. When
we reach P’, we need an extra step
b to reach the original starting point P. This extra step
P'P = b is called Burger vector.
The Burger Vector of an edge dislocation is perpen
dicular to the dislocation.

80
Crystal Defects

Burgers vector for screw dislocation :

In a Burger's circuit, the right hand screw convention is followed
(see Fig. 2.19(b)). As one rotates a screw clockwise it goes into the crystal when
placed parallel to the dislocation. Consider the Burger's circuit for screw dislocation.
Start from Q'. Move a few steps to Q. The turn left and take a few steps to R. Turn left
go to S. Then take steps exactly (PS = QR) equal to QR. From P, proceed towards Q.
When you reach P'. The loop is not complete. You have to move a distance P' Q' to
complete the circuit. Then P' Q' = b. The Burger's circuit is Q' QR S P P'Q'. Ina
crystal without dislocation, it is Q R S P Q. The Burgers vector is needed to close
Burgers circuit. Thus in the case of screw dislocation, the Burger's vector is parallel to
the dislocation.
2.8.5 Energy of a dislocation :
It has already been stated that the region around a dislocation has an extra energy.
As the atoms are displaced, the region is strained. Therefore, the dislocation causes a
strain around it and hence possesses a stress field around it. A dislocation thus has an
elastic energy associated with it. The energy of the dislocation i.e. the elastic strain
energy E,,is proportional to the square of the Burger vector in a crystal and the length of
the dislocation. Thus, Burger vector represents the strength of the dislocation. For an
edge dislocation, the elastic strain energy of the dislocation is given as
2
nip at i lg ah Horta (2.32)
2m (1= v) Ip

where, v is called Poisson ratio of the metal or crystal (which is about 0.2 to 0.4 in
metals), / is the length of the dislocation. R represents the radius of the region around
the dislocation to which the distortion is taken into consideration, r, is the radius of the
dislocation core.
The energy of a screw dislocation of length 'L' is given as
Gb? R
Le pepcsel
ne |of
eteApe tthe vie axe (2.33)

Normally 'r,' is comparable with the magnitude of the Burger vector or to the
lattice constant. 'R' can not exceed the dimensions of the crystal.
Thus, the energy of the dislocation per unit length is proportional to the shear
modulus and the square of the Burgers vector. The energies of edge and screw
dislocations are of the same magnitude, but, for the same power, the energy of an edge
dislocation is slightly higher than that of a screw dislocation.
A general dislocation is a mixed dislocation. It is a combination of an edge
dislocation and screw dislocation. We cannot have straight dislocations like pure edge
and pure screw dislocations. The curved dislocation is a mixed dislocation as shown
in Fig. 2.20.

81
 Applied Physics

Schematically, we
can show how a mixed
dislocation is created. Take
acrystal. Make a cut along
ON push the crystal inside
on the upper portion of the
crystal. Then we create an
extra plane above B.
Therefore, at B, we can
have an edge dislocation.
But at A, the dislocation
created is a screw
dislocation. In between A
and B the dislocation is a Fig 2.20 A Mixed Dislocation.
mixed dislocation. It means that the curved dislocation is created starting at A and
ending at B, At any point on the curve in the middle at a point H, the dislocation has a
screw component and an edge component. The curved dislocation is callled the Mixed
dislocation . It is in the horizontal plane whose edge is N N'
At any point H, the Burger vector will have an edge component and a screw
component.

In general, Burgers vector can be resolved into basal vectors of the lattice 1.e.,
b
can be given as b = / i+ mj + nk. There can be an arbitrary angle between
a
dislocation line and the Burger vector. The basal vectors are the crystallo
graphic
directions of the unit cell. If Z, i= OF De sya i , then the dislocation is a
perfect dislocation. If /, m,n are not equal to the integers, the
dislocations are partial
dislocations. The magnitude of Burgers vector is

Drs b bike mlb cobs eee

where a |i b, = mj, b, = nk.
If the magnitude of Burgers vector is equal to the interat
omic spacing, then the
dislocation is called unit dislocation.
For a dislocation having a length of one atomic plane, the
energy comes out to
be a high value. At G = 3.2 x 10!°N/m2and b=3.2 x 10-!0m and the length of the
dislocation equal to b, the energy comes out to be 3 eV.
The length of the dislocation
is normally much greater than b, then, the energy will
be much greater. Dislocations
Can possess energies greater than 10 eV. It is much
higher than the thermal energy (at
300 K, E,,., = 0.025 eV). A large number of dislocations
exist in a crystal and in fact,
a crystal would have a forest of dislocations. Theref
ore, a lot of energy is stored ina
crystal in the form of the elastic energy of disloc
ations.

82
Crystal Defects

When a stress 1 is applied to a crystals, the dislocation is acted upon by a force

tb. It moves under the action of the stress resulting in the deformation of the crystal.
When it moves to the surface of the crystal, it leaves a step on the surface of the crystal.
The dislocation moves on a slip plane. The edge dislocations move on the close
packed planes as the Burgers vector in these planes is smaller. The plane in which the
dislocation glides is called 'glide plane or slip plane' and the direction of slip is known as
slip direction. The slip plane and slip direction become the slip system of the crystal.
For example, in f.c.c.crystal, the densely packed planes are {111} planes. The direction
of slip is <110> and the (111) planes contain these directions. Such slip system is
called the primary slip system. In bcc. crystals, the densely packed plane is {110}
plane, and the slip direction is <111> which corresponds to lower Burgers vector. In
ionic crystals like sodium chloride, the slip plane is (110) plane and slip direction is
< 110 >. At each movement of dislocation which moves like in a relay race, the unlike
atoms or ions move over one another. In each step of movement, the bond is broken
and then reformed. In metals, the bonding is due to electro static attraction between
the valence electrons that are not localized at the sites of the atoms. They are free to
permeate the crystal, and form a sort a of ‘lubricant’ during deformation. Therefore,
metals are highly plastic. The slip planes in bec a — iron are {110}, {112} and {123}.
The dislocation density in metallic crystals would be of the order of 10- — 10!° m~.
This number increases or gets multiplied to very high densities during plastic deformation.
The dislocations, interact with each other and form entangles such that they can not
move, offering resistance to stress. Then, the metals show hardness to stress.

2.8.6 Two dimensional Defects:

The two dimensional defects are the planar defects that seperate two regions of
the crystal or phases in the crystal. These defects are: the external surfaces, grain
boundaries, twin boundaries, stacking faults and phase boundaries. We shall only
discuss about the grain boundaries only.
Any crystal that we use is terminated on free surfaces. The surface of a crystal
itself is a defect. The atoms on the surface are not well bound. The existence of a
surface is based on the surface energy. The bonds of the atoms on the surface are not
complete. A surface can have absorbed gas molecules or water molecules. The
surfaces contain steps which would be formed during the growth of the crystal. They
may contain layers that are not fully grown over the entire surface. Such steps or the
edges of the layers attract the molecules or atoms from the environment.
The most importaint defect in crystalline samples is the grain boundary. When
the melts of materials cool, two types of solids can be formed. If the melt is allowed
to cool very slowly, the atoms are arranged in a regular fashion and we get a single
crystal. if the melt is allowed to cool slowly. We get a poly crystalline sample. Any
metal wire, or rod or fused materials like quartz rods are in the poly crystalline form.

83
 Applied Physics

The solidification of a material is done by cooling its melt. During solidification

(at the freezing point), small crystallites (infinitesimally small) are formed first at some
places in the melt. They are called 'nuclei'. As the solidification proceeds, these nuclei
grow in such as small blocks of the solidified material join them. They are randomly
oriented. The crystallites with particular orientation tend to join those with the some
orientation.
All those crystallites that have the same orientation form a bigger block called
‘grain’. the melt now, as it solidifies would consists of a number of grains attached to
each other. After complete solidification, the resulting materialis a polycrystalline sample.
In poly crystalline samples, the grains combine or coalexe and in between the grains,
there exists a boundary called ‘grain boundary’. This is shown in Fig 2.21. These
grain boundaries which exist
between two grains of different grain
orientations are called 'Large
angle grain boundaries'. They grain boundary
angle is of the order of a few
degrees. They have a width,
and are associated with an
energy and surface tension.
They are defective sites. They Fig. 2.21 Scanning electron micrograph of polycrystal sample.
are the sources and sinks of The grains and large angle grain boundaries are shown.
point defects like vacancies,
interstitials and impurities. Across the grain, the atom mismatch is retained. The
Fig. 2.20 gives the scanning electron micrograph of a poly crystalline sample of a
ceramic. In the photograph, the grains and the grain boundaries are seen. The black
portions represent the voids which are three dimensional defects.
Even in single crystals, the grain boundaries exist because of mechanical
disturbances caused during solidification. The grain boundaries in single crystals are
low angle grain boundaries.
They are called sub-grain
boundaries. Such sub-grain
boundaries are shown in shown
in Fig. 2.22. The sample is a
chemically etched sodium
chloride single crystal. The
lines represent the grain
boundaries. The small pits
randomly distributed are the
isolated dislocation end points Fig. 2.22 The photograph shows the chemically etched
on the surface of the crystal. surface of NaCl crystal. The pits on the photograph indicate
the dislocation end points. The lines indicate the sub grain
boundaries.

84
Crystal Defects

The sub-grain boundaries are of Boundary

two types. They are 1) Tilt boundaries
and 2) Twist boundaries. The grain
boundaries are the arrays or cross grid
of a number of dislocations. An array
of edge dislocations is called Tilt
boundary. The schematic representation
is shown in Fig. 2.23. A cross grid of
screw dislocations is called Twist
boundary.

The grain boundaries in poly

crystalline sample play an important role
in the properties of materials like alloys Fig. 2.23 Schematic representation of tilt boundary
and metals. The affect the creep formed as a vertical array of edge dislocations.
character, toughness, strain resistance.
Yield stress of metals and alloys.

Examples :

a) Calculate the equilibrium number of vacancies per unit volume at a temperature of

1000°C. The energy for the formation of vacancy in copper is 0.90 eV. What is the
vacancy fraction at 500°C. (Boltzmann constant k = 8.62 x 10° eV/K)

The concentration per unit volume of vacancies (n) is given as

n=Ne-E/kT

Now, the value of N, the number of atomic sites in copper is given as

na Na*P

where N, is the Avagadro's number, P is the density and M = the atomic weight.

| 8.96 x 10 °
Then N= 6.023" x 10°" & a

63 .5 m?
= 8.5 «x 1028 atoms/m?

85
 Applied Physics

S
n=(85x10 28
)xex 8.62x10~ |
= 85x10 xexp [-82]

aoe
=e 551072 102"

S510 x2,75x10~
= 2.3x 107° vacancies/m?

b) The vacancy fraction at 500°C

09 13.5
7 =exp wa =e =10 230 =1.61 x 106
N (8.62x10° (773)
Questions :

1) Draw cubic cells and show the following directions..

(ES ORSe > A(aie: ASet (jen [l.0] (BD) aitee0]
©. (3 Gb Spas ee ret ey ie
2) Draw the following directions in a bcc unit cell and give the position coordinates
of the atoms at the ends of the vectors.

(a) [100] (b) [110] (Ce) fleet] (dQ) iT U3]

3) What are the < 1 1 0 > - type directions that lie on the < 1 1 1> plane of a cubic
unit cell?
4) How are the Miller indices for a crystaliographic plane determined?
5) Draw in a unit cube the crystals plane with the following Miler indices:

@)32 1) 232 (©) 201 G1 0s Orionnee te) 121

(NEO oa Gy) alee ba
6) Calculate the planar densities of atoms on( 100) and(1 10) planes of fcc and
bec samples.
7) Calculate the interplanar spacings for the (1 1 0 ) and (22 1 ) sets of planes in
aluminium.
8) Sketch the unit cell of abec structure. Mark (1 1 1 ) plane and ( 2 2 0 ) planes
in it.

86
Crystal Defects

9) What are the Miller indices of a cubic crystal plane that intersects the position
coordinates (0, 0, 1/2), (0, 0, 1) and (1, 1/2, 1)?
10) The latice constant of bec crystal is 3300 Angstroms. Calculate the interplanar
spacing of (1 1 0), (2 2 0), (4 2 2) and (7 1 2) planes.
11) A bec metal has interplanar distance of (3 2 1) planes as 841 x 10-8 cm. What is
the lattice constant.
12) Calculate the planar atomic density on the (1 1 0) plane of a-iron bec lattice in
atoms per square millimeter. The lattice constant of a-iron is 0.287 nm.
13) Copper has a lattice parameter of 3.61 Angstroms. What are the values of d,,,
and d,,o.
14) A direction vector passes through a unit cell of a cube from (3/4, 0, 1/4) to the
position (1/2, 1, 0). what are the indices of the direction.
15) Explain how a vacancy is formed in a metal crystal ? Does the vacancy affect the
bonding of atoms around ?
16) Illustrate the following type of defects in metallic lattice a)Vacancy b) Frenkel
defect c) Interstitial.
17) Derive the expression for the vacancy density in a metal crystal.
18) Explain the formation energy term ? On what factors does it depend ?
19) Derive the expression for the density of Frenkel defects in a metallic crystal ?
Which are the type of defects that predominate in a metallic crystal?
20) The formation energy for a vacancy ina metal crystal is leV. What is the proportion
of vacancies present in the crystal at 300K and 1000K ? Compare them.
21) Pure magnessium has the formation energy for vacancy as 0.9 eV. Calculate the
equilibrium vacancy fraction at 827°C and 527°C.
22) Two metals have the formation energies as 0.7 ev and 0.9 eV. What will the ratio
of their vacancy fractions?
23 —
Write the equation for vacancy concentration in a metal at equilibrium temperature
T. Two solids have atoms that have atomic masses in the ratio 1: 1.3. They have
formation energies in the ratio 1:1.3.Calculate the ratio of the vacancy concentration
in them at temperatureT.
24) The formation energy for a vacancy in pure gold is 0.98 ev. Calculate the
equilibrium concentration of vacancies per cubic meter at 827°C and 27°C. Atomic
weight of gold is 196.97 M/g mol"! the density = 19320 kgm?
25) Explain why vacancies occur in pairs in ionic crystals ? What are the effective
charges of the vacancies in ionic crystals?
26) Derive the expression for the vacancy - concentration in an ionic crystal

87
 Applied Physics

27) Can a cation vacancy interact with an anion vacancy ? If so, what will be the
resultant ?
28) Which defects dominate in alkali halides and in silver halides ?
29) Which is the migration species in ionic crystal when an electric field is applied
across it?
30) Derive the expression for (@) vacancy concentration and Frenkel defect
concentration in an ionic crystal
31) What happens when a metal from high temperature is quenched to room
temperature from the point of the number of vacancies in the crystal ?
32) Write a note on 'impurities' in crystals.
33) Write a note on 'Planar defects’.
34) What are grain boundaries?
35) In the case of expansion of a solid, what will be the fractional contribution of

vacancies at 1000K ina metal whose linear expansion coefficient is 16.2 x Tomei
and the solid expands linearly by 1.9%.
36) Write a short note on 'Point defects?
37) Why do you get dislocations in a crystals?
38) Give the model of 'edge dislocation' in a metallic crystal. Draw the Burger's
circuit.
39) Give the model of 'Screw Dislocation’ in a crystal. Draw the Burger's circuit.
40) Briefly, discuss the movement of an edge dislocation in a crystal.
41) Give the expression for energy of (a) an edge dislocation (b) screw dislocation.
42) What is Burger's vector? In what directions do the Burger's vector lie w.rt (a) an
edge dislocation, (b) screw dislocation.
43) Why do you get the deformation of a crystal when a stress is applied to it?
44) Explain why the metals are highly plastic?

88
Principles of X-Ray Diffraction Methods

2B
PRINCIPLES OF X-RAY
DIFFRACTION METHODS

2.9 X-ray Diffraction :

X-rays are the electromagnetic radiations with a wave length of the order of an
angstrom. Laue considered that the crystal consists of arrays of atoms in three
dimensions. The crystals serve as gratings to the x-rays and produce the diffraction of
x-rays. W. H. Bragg and his son W. L. Bragg while studying the properties of X-rays
studied the X-ray diffraction pattern of a few crystals. This discovery was the first
step towards understanding the crystal structures which subsequently helped in the
fields of metallurgy, chemistry, solid state physics etc.

2.9.1 Bragg Law:

It is the periodicity of the crystal structure that makes X-ray diffraction possible.
Let us consider the diffraction of X-rays by the atomic arrays shown in Fig.2B.1(a). Let

89
 Applied Physics

(a) atomic planes (b) X-ray scattering atom planes

(c) the planes taken as horizontal line

(d) in phase waves (e) out of phase waves

Fig. 2.24
us consider that the atomic array (b) is a stacked parallel planes passing through the
centres of atoms. Each atom is a centre for X-ray scattering. The monochromatic
X-ray beams fall on the parallel planes. They are reflected or scattered by the atoms.
Let the incoming x-rays be MO, MN. Let the scattered rays be OS, NS', These
scattered beams of X-ray form an angle 0 with the planes of atoms. We represent the
rows of atoms as a single horizonal line (shown in Fig. 2.24(c)). The plane of atoms
is like a mirror.
The scattered rays AS, OS' form an angle 9 with the plane (hk /). The total
path lengths of the rays M'AS, MOS' are the same. The rays are said to be in phase
with each other; that means, the waves of the individual rays arriving at SS' again
form
a common wave front.

90
Principles of X-Ray Diffraction Methods

Now, consider the incoming ray M' N and the scattered ray NS". The total path
of the x-ray beam is M'N + NS". This is greater than the path lengths of M'A + AS and
M'O + OS'.
From the Fig. 2B.1(c), the path difference is PN + NQ = 2PN.

But PN = d Sin®é

ie pour ciieremCe.oe= 20 OI = onde (2.34)

If the ray M' N S" arrives at S S'S" in phase with rays M O S' and M'A S or the
two planes { h k/ } scatter x-rays in phase, then total path difference must be equal to
an integral number of wavelengths i.e.

2G St AAT Ay ee i ee(2230)

Where (te Ut feo ae:Bie Oe ec,

This is the condition for diffraction to produce an intense beam.

The meaning of above equation is that the diffraction intensities will build up
only at certain values of 8, corresponding to specific values of and d. 'n' is called the
order of the diffraction. All the rays that are scattered by atoms in a set of parallel
planes will form a common wave front at an angle 8. Here both i and 0 are important.
Under the condition of diffraction given in eqn (2.34), the amplitude of all the waves
reinforce to give a resultant wave having the maximum amplitude. This condition is
shown in Fig.2.24(d). On the other hand, at some other angles, the scattered waves
coming from atoms may not be in phase with each other as shown in Fig.2.24(e). The
amplitude of the resultant is zero.

The relation given in eqn.(2.35) is known as Bragg law. The reflection angle @
is given as

Se te Do ial (2.36)
2d
§ depends ond, , , for a fixed 4 and d, , ; is the interplanar distance.

In principle, when we keep a crystal in the path of monochromatic X-rays, the

diffraction reinforced waves arrive at different angles (6 values). Each diffraction line
then corresponds to a diffraction of a set of wavelets scattered from a set of parallel
planes. The interplanar distances can be determined in the known crystal by determining
sin @ values. This is the most important utility of X-ray diffraction. A

91
 Applied Physics

crystal structure consists of atoms. The atoms in turn consists of electron cloud
around the nuclei. It is the scattering of X-rays by the electrons that fixes the angles of
reflection. The electron cloud as a whole scatters X-rays. X-rays do not penetrate the
electron cloud.

The number of electrons composing the cloud around the nucleus decides the
intensity of scattering. It should be noted by the students that the word 'reflection' ray
is used for 'diffracted ray' in x-ray physics.

2.9.2 Order of Reflection :

We have thus seen that when the lattice planes of indices (hk J) give rise to an
x-ray reflection, the path difference betwen the rays reflected from successive planes
is one wave-length. The Bragg law contains 'n', where 'n' is the order of the reflection.
For example, if the reflection from (100) planes of the lattice occurs, say, at an
angle 0,, then, the path difference between the reflections will be A and the order of
reflection n is equal to one. When, the path difference between the rays reflected from
the planes is 2A instead of 4, the reflection will occur at an angle 0, and the Bragg
equation is satisfied with n = 2. For another angle 93, a third reflection occurs at
n= 3. These reflections are respectively called the first, second and third order reflections
of (100). The order of diffraction or reflection has great significance in optics, but in
x-ray diffraction, it is not necessary to consider the reflections from (200) planes, or
(100) planes seperately.

Let us examine a particular case. Consider (100) planes and (200) planes in a
bee structure. Let the lattice parameter be 'a'. Then, digg = a and dyo9 = a/2.
The
Bragg condition for (200) reflection is 2X = asin @. The Bragg condition for (100)
reflections is given as 2, = a Sin 9. A second order reflections from (100) planes
should satisfy the condition 23, = 2dig9 Sin @. The first order reflection
from (200)
planes should satisfy 7, = 2d,99 Sin 9. The interplanar distance for (200)
planes is
half that of (100) planes. Therefore, the first order of (200) planes and
the second
order of (100) planes will coincide. Similarly, the first order diffraction
lines of (300)
planes will coincide with the third order lines of (100) planes. Therefor
e, it is not
needed to consider the order of diffraction on vigorous terms.

92
Principles of X-Ray Diffraction Methods

2.10 Laue Equations for X-ray Diffraction :

We now find a general approach and obtain-equations which determine the
directions of constructive diffraction (i.e., maxima) for a particular wavelength.

Fig. 2.25 Determination of phase difference between waves scattered by atoms A and B.

The two atoms in the plane scatter X-rays. The direction of the incident X-rays
has changed upon scattering. The effect of the atoms in the plane is to change the
directions of X-rays.
Let sy and s be the unit vectors associated with the two rays, the incident ray
(Sy) and the reflected ray (s) respectively. From the figure, it is seen that the path
difference for incident rays is given as
A, =1cos 0, =F e Sy
where 9, is the angle between the incident ray and the vector r.
Similarly, the path difference for the reflected rays is given as
A, =rcos 0, =Vre Sy
i.e., the total path difference is
Ae 10 (5 —Say
Ses. eee oe he C3y)
Where S=s—Sp
The vector S is normal to the plane that reflects s, into s. (Fig.2.25(b))

Fig. 2.26 Bragg reflection.

95
 Applied Physics

The planes are 'd' apart as in Fig. 2.26. We have to calculate the path difference
between the two reflected rays AP and BP’. From the figure, we have

; d
AB=h; Sin @= +; cos 20 = h

the path difference = A=h-—T/

h = d cosec 80, /=h cos 20 = d cos 20 cosec 0

A = d (cosec 8 — cos 28 cosec 0)

B=2dsin fl "ses = ere) a ee (2.38)

Then, the constructive interference between the rays can be obtained when the
path difference is an integral multiple of A i.e.,

A=nx

DA = 2d SiG Stee Pilea? Vi let ee eee (2.39)

This is Bragg law. Immediately we can see that, the interference of secondary
wavelets (i.e., diffraction) can be obtained only if 4 < 2d, otherwise no diffraction
effect is found. That means, with long wavelengths, this law is not valid. Therefore,
we can not get Bragg reflection with visible light. The largest interplanar distance value
is of the order of 4 x 10-8 cm.
The plane is the reflecting plane for the Bragg condition. Let the angle between
Sy and s is 20. The magnitude of |S| = 2 sin 6 (Fig.2.25(b)) (s and Sp are unit vectors).

If Bees: ee ee ee ee (2.40)
we have constructive interference of the reflected ways.

The phase angle between the rays for constructive interference is

2m
Ore xo fT oat ee ae ee (2.41)

This is true for any two angles.

The crystal has three crystallographic axes. The nearest neighbor-distances along
these three crystallographic axes are a, b, and c. They represent the significant distances

94
Principles of X-Ray Diffraction Methods

(r) in a crystal. Then, the phase difference of the waves that are coming from adjacent
atoms separated by the three translational vectors a, b, and c, is given as

2m
o, = 7
ae" = 227th

21
% = paar ae OR eR Ne (2.42)
2m
, 7
SessiceS t= 2rl

where h, k, / are integers. All the above equations represent the condition for constructive
interference of the X-rays coming from the two atoms.
A dot product can be expressed in terms of direction cosine as
a.S=2asin0@cosa, S|

=2aasinO=Ah
b.S=2bBPsinO0=Ak
tas=—2 ¢7sn 0 = vol

The above equations are called Laue equations.

They determine the direction of the constructive interference of the reflected

waves for a particular wavelength of X-rays.
We can write the above equations as

cela ajc 2 et ao
h k i

rs lel oa so i
ee peianiappeai Set tk shatvagem wii -yims eee (2.44)

2asin9 h
or pan = 5

2Bsin@ k (2.45)
x b

2ysin@/
ry c
where '@' is the angle between the incident or reflected beam direction and the reflecting
plane.

oS
 Applied Physics

Therefore, in the direction of constructive interference, the direction cosines

are proportional to Ye 4. : ow. The above h, k, / are the Miller indices (h k J).
Therefore, the plane under consideration is (h k /) plane. The Bragg equation and Laue
equations are equivalent for X-ray diffraction. We consider a set of planes {h k /}.
They are parallel to each other. The direction coninses of the normal to those planes are

proportional to we ; ve and y . Therefore, any plane parallel to the plane (hk/)

gives the diffraction angle i.e., 6. This is the statement of Bragg's law. If the distance
of separation between the planes is 'd' then

2dsinO=A
which is Bragg's law. Laue reflection plane is the lattice plane with
Miller indices (hk /). Those planes which do not have atoms do not reflect any X-rays:
therefore, no diffraction maxima. The condition for reflection of X-rays is that the
plane should contain atoms.

2.11 Methods of X-ray Diffraction :

When X-rays fall on a crystal, the X-rays are scattered. The scattered waves
interact with each other. From the Bragg equation, we understand that the reflections
can occur in general for a given set of lattice planes only, if there is a variation in X-ray
wavelength (A) or the orientation of the planes (6).
The Bragg law can be given as
D2 de sity.) re ee een) Ss (2.47)
'n' represents the order of reflection. n can be any integer such that sin 0 is not
greater than unity. The magnitude of the distance between two adjacent and parallel
planes of atoms is a function of Miller indices (h, k, J) as well as the lattice parameter
(a). In the case of a cubic crystal,

a
daks = ne ee: aie (2.48)

From equation given in eqn. (2.47), we get

96
Principles of X-Ray Diffraction Methods

for a cubic system where a = b, c = a. The quantity (h?ko ?) is an integer.

For a tetragonal crystal,

A nv? 2 2 r PD
sin?0,,, = rel ik )+rigs BE (2.49(b))

For an orthorhombic crystal,

Ls" alia ‘apie aadsata

sin7O ar a aes ii aoe hea ee (2.49(c))

The purpose of any diffraction method is to find the position of the diffraction
maxima or the points where the Bragg reflections come with respect to the direction of

x-ray a!
source
a |
collimeter yen yy
sample and
mount

photodetector or
photographic film

Fig. 2.27 General X-ray diffraction method.

X-ray beam and then give the Miller indices of the planes responsible for the reflections.
We use these Miller indices to calculate the lattice parameters of the crystal. We shall
confine ourselves to cubic crystal only.
In general, they are two types of X-ray diffraction methods. The first of the
methods employs a single crystal. The second type of method employs polycrystalline
samples.
For a fixed wavelength, the direction of diffraction is not arbitrary. That is, a
single crystal can not be used for diffraction with monochromatic X-rays. It does not
give any results. We have to employ polycrystalline sample or powdered sample in the
case of monochromatic X-rays.
In the case of a single crystal, we have to employ X-rays which have multiple
wavelengths. Such a radiation is called 'white' X-ray radiation.

oF
 Applied Physics

This schematic arrangement for observing X-ray diffraction is shown in

Fig. 2.27. X-ray diffraction methods consist of 1) An X-ray source 2) A sample holder
3) A detector to record the reflections. For detection, a photographic film or a photo
detector is used. The sample holder consists of a holder which is capable of free
rotation and some times a goniometer (for Laue method) to orient the crystal.

2.12 X-ray Sources:

When high energy electrons collide with a metal target, X-rays are produced.
In the production of X-rays, the important things are the target, the source of electrons
and a voltage source to accelerate electrons towards the target. The X-rays that come
from the targets consist of many wavelengths. When the electrons that hit the target
get decelerated rapidly, a continuous spectrum of X-rays is emitted by the target.
The
emission is from the decelerated charge. The spectrum starts at a minimum wavelengt
h.
Heavy metals like tungsten (z = 74), give continuous spectrum.

When the voltage on an X-ray tube is raised above a critical value

that is
characteristic of the target metal, sharp intense lines of X-rays at certain
wavelengths
are produced. The wavelengths are the characteristic of the target metal.
The origin of
the characteristic radiation is as follows. F irst, K electrons are
knocked out of the atom
by highly energetic electrons bombarding the traget, leaving excited
atom. Then, some
electrons in higher shells drop down to lower energy levels to replace
the lost K electrons,
emitting energy of a characteristic wavelength. The transition of
electrons from L shell
to K shell give K, lines.
Molybdenum gives K lines and M lines. K lines are used in
X-ray diffraction
methods. M lines are filtered through filters. The K
lines are composed of Ky ‘ Ky,

and Kg lines. For molybdenum, K. : 0.709A°, K,: 0.71A° and K 54 s0.632A2

a, and q, lines are so close, they are not resolved easily.
In the earlier classes the
students had studied that the characteristic X-ray lines
follow Mosley's law which
states that the wavelength of any particular X-ray line decrea
sed as the atomic number
of the emitter increased. We have to select a proper
target to give proper X-ray
wavelengths for use. A number of targets are availab
le, the most used targets are
molybdenum, copper and tungsten.

An X-ray source called X-ray tube must contain

(a) a source of electrons
(b) a high accelerating voltage, (c) a metal target.
The X-ray tubes (invented first by
Coolidge in an 1913) consist of an evacuated glass
envelope that insulates the anode at
one end from the cathode at the other. The cathod
e is Just a tungsten filament. The

98
 Principels of X-Ray Diffraction Methods

anode gets heated during electron impinging. It has to be cooled continuously through
a water cooled copper block to which the target is attached. The internal structure of
an X-ray tube is shown in Fig. 2.28. Copper and cobalt targets are widely used targets.
The wavelengths emitted by them are given below.

copper vacuum glass

\ tungsten filament

cooling water
circulation

4
beryllium window x-rays metal cup for focusing

Fig.2.28 X-ray tube

Cu Ky, Sof a04 Kg, > 1.392A

Ka, > 1.544A
Co Ky, > 1.788A
— 1.7928A Kg, > 1.621A

2.13 Laue Method:

We will confine ourselves to a method which employs a stationary single crystal.

Laue method was the first diffraction method ever used. In this method, a beam of
white radiation is collinated through a collimeter (c). The 'white’ X-rays fall on the
stationary single crystal (s) shown in Fig. 2.29. The Bragg angle 9 is therefore fixed
for every set of planes in the crystal. Now, each set of planes selects and diffracts that
particular wavelength that satisfies the Bragg's law for particular values of d,,, and 0
involved. Therefore, we get a number of diffacted beams each corresponding to one
set of planes for a particular wavelength. In Laue method A is variable.
The diffracted pattern is caught on a photographic film. Laue pattern can be
obtained either by transmission Laue method (Fig. 2.29(a)) or back-reflection method
(Fig. 2.29(b)).

oo
 Applied Physics

Film Film
Fig. 2.29 Laue method (a) Transmission (b) Back-reflection.

In both the ways, the diffracted beams give a series of spots on the film as
shown in Fig. 2.30.

Fig. 2.30 Laue Pattern of NaC/ crystal

The spots lie on certain curves. These curves are either ellipses or hyperbolas
for transmission laue pattern and hyperbolas for back reflection patterns.

Although, the experimental method of Laue is very simple, the interpretation of

Laue pattern or the diffracted spots is difficult. We can not determine the lattice
parameters of a crystal using this method. Laue method is useful to study the symmetry
of the crystals and in the determination of orientation of crystals. This is possible
because the sample is stationary during the exposure to X-rays. The crystals used
should be thin for transmission laue method. Otherwise, the X-rays are absorbed by
the crystals. Then the density of spots on the photographic film will be less.

This interpretation of Laue photograph requires the knowledge of the concept

of reciprocal lattice; the topic is out side the syllabus.

The Bragg angle for any transmission Laue spot is given from the relation

tan 20 = =
r
Bcd (2.50(a))
100
 Principles of X-Ray Diffraction Methods

where r, is the distance between the centre of the film and the diffraction spot. D is
the distance between the specimen and the film (usually 5 cm). Laue spots, near the
center of the film or close to the directly transmitted X-ray beam, correspond to the
first-order reflections from planes inclined at very small angles to the incident beam.
In this case, only lower wavelengths can satisfy the Bragg condition. One can notice
that the region close to the centre of the pattern is devoid of any diffraction spots and
the size of this region as the intensity of incoming X-ray beam is reduced.
In the case of back reflection pattern, the Bragg angle for any spot on a
back-reflection pattern may be obtained from

tan (180 — 20) = as alley (2.50(b))

where r, is equal to the distance of a spot from the center of the film and D is the
specimen-to-film distance (~3 cm). Normally, the back reflection pattern may have
spots close to the centre of the pattern. Such spots are due to high-order-overlapping
reflections from (hk/) almost perpendicular to the incident beam.
The most essential thing in obtaining the Laue spots is to align the camera (simply
a flat plate) correctly oriented with respect to the X-ray beam. That is the reason we
require collimeter so that the direct X-ray beam is made to pass straight into the crystal
and the centre of the film. Back reflections normally give weak spots. Hence, the
exposure of the film should be longer.
As one moves towards the centre of the Laue pattern, the corresponding values
of 8 decrease. At the same time the indices of planes (hk/) increase. We have
A =2d,,,;sin 8. When we move towards the centre of the Laue photograph, the
corresponding value of d,,, must decrease. Both terms on right hand side of the
Bragg relation decrease as we approach the centre of the film. A limit is set by the
minimum value of the wavelength in the X-ray spectrum. Hence, the spots can not
occur near the centre. If we consider a plane (h, 0 0), then
h
sin
8 = —————————
aieee
a, 6 2a h
en ee * Oa
Whe +k24P Whe ake 42

re 2 ah
2.5]
(7Ek a a
Near the centre of the every photograph, h in every spot is equal to unity.
Therefore, the spots nearest to the centre are the most likely to contain first-order
reflections.

101
 Applied Physics

* Indexing Laue Pattern or Laue Spots :

Laue spots shown in the Fig.2.31. The orientation of the

crystal is such that the
X-rays travel along one of the
crystallographic axes which O <———_p cot 8 ————-> L
we take it to be the c-axis or
z-axis. Then, let O represent
the centre of the film. Let L
be the spot on the film.
N is the point at which
the X-rays fall on the sample. <— pcotd
Then OL = r tan 20. r is the °
distance ON. We have
another point P, called the
gnomonic. pole. Then
construct a circle with radius
p cot (PQ). Let p = 2 cm.
Then, the Laue pattern can Fig. 2.31(a)
be transferred to a graph sheet. The centre of the photograph is determined first.
Then, the distances of all spots from the centre are determined. This distance is r tan
20. We know the value of 'r', calculate '@'. Then calculate p cot @ and for cubic
crystals p = 2cm is taken. We take a crystal that is oriented along the axis [1 0 0].

Fig. 2.31(b) Gromonic projection of Laue pattern of a cubic crystal. r= 5 cm and p=2 cm.

102
 Principles of X-Ray Diffraction Methods

For a value r tan 20 = 1.0, we have p cot 0 as 20.14. For a value of r tan
20 = 2.0, we have p cot @ as 10.39 when r tan 20 = 3.0, p cove 7:22.
r tan 20 values and p cot 0 values are marked on the graph sheet with the scale, one
unit = 2 cm. A gnomogram is built up. The circles on the graph represent the dark
spots in the photograph. (Fig. 2.31(a)).
All the points represent the reciprocal points. The film is perpendicular to
z-axis. The indices of the points are (hk/). We have considered only the points which
fall at the net points.

2.14 Powder Method:

Many materials that are used in industry are polycrystalline materials. They are
used in powder form. Therefore, the X-ray diffraction pattern of such materials is
obtained in the form of powder pattern. The powder patterns are fairly simple and are
widely used for the characterization of crystals or materials by X-rays diffraction.
Laue photographs give spots. The powder pattern is obtained in the form of
lines. The Bragg scattering of X-rays consists of the scattering from all the crystallites
in the specimen corresponding to a particular angle of 20. The diffracted beams must
lie on a cone of apex of which is at the specimen. The specimen is kept on the axis of
the X-ray beam.

sample
Photographic film

Fig. 2.32 A Circular Line is formed on the photographic film.

The axis of the beam coincides with the axis of the cone. The cone has a
semi-vertical angle that is equal to 20. The formation of such a cone is shown
in Fig.2.32.

A number of lines will be produced for each value of the angle @. The flat
cameras are not used in general. The specimen is surrounded by a narrow strip of
film. This strip intercepts a large number of cones of X-rays reflections. As the film
is of the width of about | inch, a part of each cone is recorded. All the reflections for
the angles 0 < @ < 90 are recorded.

103
 Applied Physics

The powder should be extremely fine and the grain size in the powder should be
of the order of 1 —3um. If the grain size is greater than this, the powder pattern will
have '‘spotiness'. The camera is a cylindrical camera as shown in
Fig.2.33 (a) and b. The camera is called Hull/Debye-Scherrer camera.
X-rays are made to pass through a narrow aperture on the left of the camera.
The direct beam after passing through the specimen goes to the other end through a
narrow aperture and falls on a X-ray luminescent screen. A luminous spot appears on
the screen. By adjusting the spot to be exactly at the centre of the screen, the X-rays
can be made to go through the sample exactly. The side view of the camera is shown
in Fig. 2.33(b).

sample mount
[Saeen eee
GEE
Ree ery

X-rays

Fig. 2.33 (a) Acylindrical camera. (b) The top view of the camera. X-rays from left pass
through the
aperture a and through acollimeter C. They fall on the sample S. The direct beam goes to P.

The sample is stacked in a capillary tube of diameter 0.3 — 0.5 mm. It is kept on
the sample holder (M) vertical to the mount.

104
Principles of X-Ray Diffraction Methods

nse The geometry of the powder camera is a section of a hollow cylinder. The film
strip is enclosed on the inner wall of the camera with an yellow paper sheet on it.
Only monochromatic X-rays are used. Usually all the other radiations from
X-ray tube are cut out by keeping a filter before the aperture. The copper target gives
two X-rays wavelengths K, , K, 5. Therefore, the high angle lines (close to 90°)
contain two lines which are due to K, , and K, , X-ray beams. This is shown in
Fig.2.34.
The figure shows the film mounted in the camera and the X-ray powder pattern
obtained.

ete ne aea ace og,eet,

~~“ i ron
f-
ree
6a eR
e.
a isa .
he tine Ts 3 “S04
tae oat gi -
ee Ss age
ot
Oe
34
wa
<= Eas en Oe ered
eran Fi & ee ae
“mnt (Se aN ar

ph ilMig Taig a Db OR AW Aesek oA Oype bao}

Low angle High angle

fines fines

(a)

Fig.2.34 (a) The figure shows the arrangement ofthe film in the camera. The stretched film shows
the diffraction lines. The High angle lines will be two due to K, ,; and K, , wavelengths.
(b) Powder pattern of KC/-KBr mixed crystal taken with Debye-Scherrer camera.

In the case of the cylindrical camera. @ is proportional to S, where the S is the

diameter of the ring that is obtained for an angle 98. Then
aS
4R
Where R represents the radius of the camera.

105
 Applied Physics

After taking the photograph, the film is made flat and the distance S is measured
with the help of a comparator. A comparator is just a magnifying glass with a cross
wire (the cross wire and the film are kept in the same plane) and the glass is fitted to a
metal scale. The magnification of the lens is less than four times the diameter of the
rings. On the left side of the film, (i.e., the right side of the camera), we get low angle
reflections. On the back side i.e., around the point where the X-rays enter the camera,
we get the high reflection rings. They are resolved into K,, and K,, lines.

Suppose the distance between the centres of holes of the film ( they represént
the entry of X-rays and exit of the X-ray beams) is D. Then the diffraction angle 0 is

Q = [90-458 mek (2.52)

The X-rays are scattered by the air molecules in the camera. The direct beam is
not allowed to travel in the space. Narrow conic tubes or collimeters are provided so
that the tips of the collimeters are close to the sample.

Interpretation of the powder pattern :

: ; a
In a cubic system, A = 2d sin 8, dks = vA Sie ee

2
This gives sin? 0 = ioe eee (2.53)

The parameter (1? ey eecsy |3)is an integer. All that is necessary is to calculate
sin? 8 values and N values from them, where N = (h? + k2 + 7),
The characteristic of a cubic crystal is that the values of sin? @ have a common
factor. By dividing sin? @ values with this common factor, we can obtain
the integers
corresponding to each reflection line. They correspond to (1? +k? +1 =) values.
Then ‘a' the lattice parameter can be calculated from the express
ion (2.53).

106
Principles of X-Ray Diffraction Methods

a=
,{common factor
N

The first two lines give the indication that they are from a cubic crystal. The
high angles give the most accurate value of the lattice parameter.

The characteristic of a cubic system is that the values of sin? 6 have a common
factor. This factor can be found by taking the differences of sin? @ values for the
successive lines. The least value of difference is the common factor. Then, we divide

the sin? @ values with this least difference (asin? @}min . We get integer values which

correspond to h? + k? + ? values. The Miller indices can be found. If the Miller

indices come out to be all odd or even, then the cubic system is an f.c.c. lattice.
Ifh +k +/=all even, then the crystal is a b.c.c. crystal. Let us take an example. An
oxide of iron gives the following 9 values. Let us find out the Miller indices. It is done
in the following table.

5)
S.No. (0) Sing Sin?@ = ASin’@ Ge N_ hkl

l 10.67 0.1852 0.03476 : 3.0 sales)

2 17.64 0.3035 0.0929 0.0553 8.0 8 (220)
3 20.80 0.3535 0.12561 0.0328 10.9 11 (311)
4 O72 0.3709 0.1376 0.0115 11.9 12?)
5 25.34 0.4279 0.1831 0.047 16.0 16 (400)
6 31.63 O6246) 0 02753 = 01377 23.9 24 (422)
71 2in33ig14 0.5565 0.3097 0.0344 26.9 27 (333)
The data is taken with x-radiation of wavelength 1.79A

The lattice parameter can be calculated as

/
se 1.79 x _ x 10° =8.3A0

107
 Applied Physics

The above cited example is an f.c.c. cubic crystal because all hkl values are
either odd or even. For a b.c.c. crystal structure, the diffraction occurs only from
those planes whose Miller indices when added together give a total that is even number.
i.e., hh + k + /= even. The principal diffraction planes in b.c.c. structure are {100},
{200}, {211}, {220}, {310}. Those reflections, for which h + k + / = odd, will be
absent i.e., {311}, {100}, {111} will be absent. Higher orders actually are useful for
accurate determination of the cell parameters.
Example :

Determine at which (111), (200), (220), (311) planes of a nickel crystal will
0
diffract X-rays (A = 1.54 A). The crystal is an f.c.c. crystal with a lattice parameter
0
a= 3.52 A. What are the values of inter planar distances ?

Solution :

a
ae ety
ag Cee ITO 0 Sin’@
Vh7+k74/2 A
3.52
(111) ke ie 0.1263 7.256 00159

3.52
(200) 2 re: 0.1100 6.315 0.0121

a 3.52
(220) 2/2 32 0.1361 7.82 0.0185

a 3.52
(311) ait ae 0.0659 3.779 0.0043

2.15 Difference in the Diffraction Pattern of fec and bee C rystals :

In the case of bec crystal,

h+k+/=Even

For fcc crystal, we get reflections from those planes for which the Miller
indices
are all odd or even. The planes that reflect X-rays in the case fcc
crystal are (111),(200),
(220), (311), (222), (400), (33 1), (420), (422), (511), (333),
(440), (531), (442),
(600), (620), (533), (622), (444), (551), (711).

108
 Principles of X-Ray Diffraction Methods

We have already mentioned about the difference in the powdern pattern of bcc
and fcc crystals. Suppose one takes the x-ray diffractogram of a crystal. We have to
know whether it is from a bec crystal or an fcc crystal. First, we have to calculate the
Sin 6 and sjn29 values. Since, the wavelength of x-rays and the lattice parameters

are fixed for the crystal, the ratio of the sjn2@ values for the first two lines is

Sin?@, — hp +ky +17

Since gk lees | gage S bbee Oca

Where 6, and 0, are two different angles associated with the principal diffracting

planes { h,, k,, 4 } and { hy, ky, J, }, respectively.

The first two sets of planes for bcc are (110) and (200).

Sin’@, — 1° +17 +0
a eS
Sin?0, 274+07+07

Thus, if the crystal structure of the unknown cubic metal is bcc, the ratio of the

Sin2@ values which correspond to the first two principal diffracting planes will

DO:

For f c c crystal structure, the first two sets of principal planes are the {111}
and {200} planes. Substitution of (h k /) values for the planes in the relation for

Sin*@» we get

Sib, eek se el
TS
Sin?@, 27+07 +07

Thus, for a crystal of unknown cubic structure, if the ratio of the first two

~ sin2@ values comes out as 0.75, it is fec. The remaining (h k 7) values can be

assigned from the set given.

109
 Applied Physics

The following data is given for NaC/ crystal that is a face centred crystal.
) 20 D Value sin2@ x j000 4H K L Intensity
13.69 of QU38se 30574 56.01 l I 1 86.8
W586 ceue3 lS pee82 74.68 2 0 0 1000.0
22.73s | x45, 47D 19947 149.36 2 2 0 663.7
26.95 > 53:89 1.7011 205.36 3 1 1 23.0
28,25 56.50 1.6287 224.03 2 2 2 215.5
3:13) 566.265) ofA105 298.71 4 0 0 95.1
3655 ~~ 7311 12944 354.72 3 3 l 11.6
37:67 75:33 1.2616 373.39 4 2 0 255.0
42.02 84.04 1.1517 448.07 4 2 2 190.1
45.23 90.47 1.0858 504.07 5 1 l 9.9
45.23 90.47 1.0858 504.07 3 3 3 3.3
50.62 101.24 0.9974 597.42 4 4 0 69.9
53.93 107.87 0.9537 653.43 5 3 1 225
55.07 110.13 0.9403 672.10 4 4 2 135.8
55.07 110.13 0.9403 672.10 6 0 0 33.9
59.79 119.57 0.8921 746.78 6 j3 0 142.5
63.63 127.27 0.8604 802.78 5 3 3 15.9
65.00 130.01 0.8506 821.45 6 2 2 162.4
71.20 142.40 0.8144 896.13 4 4 4 69.7
7736 154.73 0.7900 952.14 5 5 1 36.8
77.36 154.73 0.7900 952.14 7 1 l 36.8
80.16 160.32 0.7824 970.87 6 4 0 394.3
In the observations

Sin’0,
= 0.75. Therefore, the structure belongs to FCC.
Sin’,
The indices can be ascribed from the table given. The Miller
indices should
be either all odd or all even.
Calculating for 422 plane, we get the lattice parameter ‘a' as

EV 9/7/ rin ee = 0.5635 nm

0.4481

110
Principles of X-Ray Diffraction Methods

2.16 X-ray diffractometer :

In the powder method, we have said that the powder pattern can be recorded on
a photographic film. But, these days, the automation has come and the diffracted
beam can be recorded by means of a photodetector. The schematic diffractometer is
given in Fig.2.35. The method is same as Debye-scherrer camera method except a
movable detector replaces the film. It moves on the circumference of a circle with the
sample as its centre.
diffractometer ——*”
circle

Fig. 2.35. A Schematic diagram of sample and photodetector in x-ray diffractometer.

X-ray from S through a collimeter C fall on the sample at the centre P. The
diffractional beam are focussed into the window (F) of the detector (D). The entire
set up is placed ona vibration free unit. A carrieage (E) supporting the detector and the
window rotates about an axis O. The angular motion is accurate to the level of O,
0.19 to 0.019. Now a days, we have automated systems. The data on the intensities,
angular positions of the diffracted beams are collected by a computer through software.
One such data collected is given in the Fig.2.35 for NaCl.
200

Intensity

Diffraction angle 20

Fig. 2.36 X-ray Diffractogram of a FCC crystal

111
 Applied Physics

2.17 Rotating crystal method :

In the rotating crystal method, a single crystal is mounted with one of its axes or
any important crystallographic direction, normal to a monochromatic X-ray beam.
The camera is a cylindrical cylinder as shown in the Fig. 2.37(a). A cylindrical
photographic film is loaded and placed around the inside part of the camera. The
crystal is rotated about a chosen direction, the axis ofthe film coinciding with the axis
of rotation of the crystal. The rotation brings a particular set of lattice planes into the
correct Bragg angle position for the particular monochromatic X-ray beam. Then, the
diffraction of the beam takes place.
The diffracted beams are located on imaginary cones. The axes of the cones
coincide with the rotation axes. Now, a number of spots are formed on the film.
When the photographic film is laid flat lie on ‘layer’ lines Fig. 2.37.

Fig. 2.37 (a) Rotating crystal pattern on a photographic film —(b) Rotating crystal method

The crystal is rotated about only one axis, the Bragg reflections for all angles
between 0° and 90° do not take place for every set of planes. Every set of planes do
not produce diffracted beams.
‘The main use of the rotating crystal method and its variations are found in the
determination of unknown crystal structures. This is normally not used by an average
materials scientist who uses X-ray diffraction as a laboratory tool to find the crystal
system and the lattice parameter. Analysis of the spots in rotating crystal diffraction
pattern is beyond the scope of this book.
Example :

The atomic radius of molybdenum atom is 0.1363 nm. Compute the intermatic
spacing for (111) set of planes in the molybdenum crystal C the crystal structureis
B.C.C. What is the inter planar spacing for (220) set of planes.

112
Principles of X-Ray Diffraction Methods

Solution :
The lattice parameter a is

Ar _ 4x0.1363nM _ 9 3148nm
PRE 1-732

A-~ 01363: nm

bee eae

_ 4x0.1363nm
3

= 0.1816 nm.

This does not give a diffraction line.

4x0.1363nm _ 4x 0.1363nm
eee
d a ee Se er ey ee
za J3xJ44+4+0 J3 x32

=
2 x<t
0.1363aoeb

= 0.1113 nm.

From the above, calculate the diffraction angle for (220) set of planes for first
order
Solution :

a TOROS ——_—
=",
0179
Sin) = —— = —_____
2d59 2x0.1113x10 2x0.113

= 0.7920

9=Sin7! (0.7920) = 52.37°

The angle of diffraction is 26 i.e.,

D6 22 << (52.37°)

= 104.74°

113
 Applied Physics

Example :

An x-ray powder pattern of an element contains the powder lines at the following
20 values : 40.113°, 46.659°, 68.080° and 82.090°. The wavelength of the radiations
used is 0.154 nm. Find out whether the crystal structure is FCC or BCC. Determine
the lattice constant of the element.
Solution :

20° ) Sind Sin70

40.113 20.056 0.3429 0.1176
46.659 23.329 0.3960 0.1568
68.080 34.040 0.5600 0.3136
82.090 41.045 0.6567 0.4313
Consider the first two lines and take the ratio of the sjn29 values
ee?)
Sin’0, _ 0.1176 _ 47.
Sin’@, 0.1568
Therefore, the crystal used in FCC. The first principal line in FCC is (111) ss

0.154nm 2)
ae 7 ne 0.1176 = 0.077 425.5 nm = 0.3885 nm = 3.855 A

Questions :

1) Derive Braggs Law.

2) What is the order of reflection?
3) Calculate the inter planar spacing of atoms for (100), and (110)
planes of BCC
crystal?
4) Describe the Laue method.
5) Write a note on Rotating Crystal method.
6) Why should we use white radiation of X-rays in Laue
method?

114
Principles of X-Ray Diffraction Methods

7) How does Laue diffraction pattern look?

8) Where is Laue method useful?
9) Write a note briefly on 'Rotating crystal method’.
10) Why should you use monochromatic x-ray radiation in the case of powder method?
11) For BCC iron, compute (i) the inter planar spacing (ii) the diffracting angle for
the (220) set of planes. The lattice parameter of iron is 0.2889 nm. Assume that
a monochromatic radiation having a wavelength of 0.1541 nm is used and the
order of reflection is one.
12) Calculate the inter planar spacing for the (110) and (221) sets of planes
in aluminium?
13) —
Rubidium is a BCC crystal. The angle of diffraction for (321) set of planes is at
27.00° for the first order reflection. The wavelength of x-rays used is 0.0711
nm. Calculate the interplanar distance of the planes. What is the atomic radius of
Rubidium atom?
14) How do you distinguish between BCC and FCC structures from powder pattern?
15) What are X-rays ? What should be the order of wavelengths that should be use
for X-ray diffraction of crystals ?
16) What is the characteristic x-ray radiation ?
17) Prove that Laue equation are equivalent to Bragg condition.
18) Obtain the expression for the interplaner distance d,,,, for a cubic lattice
19) Nickel shows a diffraction maximum at 20 = 101.502°. The wavelength of
X-rays is equal to 1.541°A. Calculate the lattice parameter of nickel cell if the
above reflection corresponds to (310) plane (Assume n = 1)
20) A crystal shows the diffraction maxima at the angles 20 =38.184°, 44.392°, 64.576°
and 72.5479. If the wavelength of x-rays used is 1.541A°. Find out whether the
crystal used is b.c.c. or f.c.c. What is the lattice constant of the crystal.
21) Describe Laue method of X-ray diffraction. What does the each intense point in
X-ray diffraction represent in Laue pattern.
22) Describe the powder method of X-ray diffraction.
23) Discuss how powder photograph is taken.
24) Discuss how powder pattern of a cubic crystal can be analysed ?
25) Mention various defects that occur in crystals. Write a brief on each of them.
26) When a monochromatic X-ray beam of X-rays of wavelength 0.1542 nm is
used, the first order reflection from (113) set of planes occurs at 8. What is the
value of 0.

115
 3
PRINCIPLES OF
QUANTUM MECHAN
—————ICS ———

3.1 Introduction:

Interference experiments demonstrate the wave nature of electromagnetic

radiation, which includes visible light, infrared radiation and X-rays.
This type of
experiments where interference is involved, presence of atleast
two waves at the
same position, at the same time is required. However, it is impossib
le to imagine the
existence of two particles occupying the same position at the same
time. On the
other hand, experimental results such as Photoelectric Effect,
Compton Effect, and
X-ray absorption are explained by considering the radiation to
appear as a stream of
particles which are absorbed one at a time. Therefore, from the above, we consider
the light to possess the properties of waves, as well as those of particles
. Experiments,
where wave nature is exhibited, are known as Propagation
Experiments and where
particle nature is used are called Interaction Experiments. This
dual nature of light
was not readily accepted. The main reason is due to, the
contradictory aspects of the
two. A wave is specified by a frequency 'f, wavelength
'A', phase velocity ‘Dy
amplitude 'A' and intensity 'l'. We have a relation among
these as ie fh. A wave is

116
Principles of Quantum Mechanics

necessarily "spread out and occupies a large space". A particle, on the other hand, is
specified by a mass 'm', velocity v, momentum 'p' and energy 'E'. The property of a
particle that conflicts with the wave is, that, it occupies a "definite position" in space.
But in order to explain all the properties of electromagnetic radiation (including visible
light) acceptance of the "dual nature of light is necessary". Hence from the Planck's
radiation and Einstein's Equation, we get a relation between the particle and wave as
h
E = hf and eee foe (3.1)

where E — Photon Energy, h > Planck's Constant, p + photon momentum, ’ >

wavelength and f + frequency of the wave. This dual nature of light concept lead to
another important conclusion, on the basis of symmetry, that the particle also can
behave as a wave. We will learn about this concept and its impact on Physics in
opening the flood gates for a new branch of physics called "Wave or Quantum
Mechanics" and its consequences in following sections of this chapter.

3.2 de Broglie hypothesis of wave-particle :

The particle nature of light i.e., the quantum concept came into existence from
Bohr's theory of atomic structure also. In this, the Quantum or Particle concept of
radiation came from the postulate of Bohr about the absorption or emission of energy
and also about the motion of an electron in a stationary orbit around the nucleus with
constant energy. Even though the postulates were given by Bohr, the implications of
these were not understood completely. The first step in understanding the hidden
meaning of Bohr's Quantum Orbits was made by Louis de Broglie, a French Scientist,
who tried to draw an analogy between the set of discrete energy levels that characterize
the inner state of atoms and the discrete set of mechanical vibrations that are observed
in the case of violin strings, organ pipes, etc., where standing waves form and energy
remains constant. de Broglie thought that the optical properties of atoms are due to
some kind of standing waves enclosed within themselves. On the basis of this, he
came out with a hypothesis that the motion of electrons within an atom are associated
with a peculiar kind of waves, which he called "Pilot Waves". According to this, each
electron circling around an atomic nucleus must be considered as being accompanied
by a standing wave that runs around the electronic orbit. If this is true, the only
possible orbits would be those whose wavelengths are an integral multiple of the
wavelength of the corresponding "de Broglie Wave".

Fig. 3.1 (a) shows the stationary wave pattern found in strings and Fig 3.1(b),
the de Broglie's waves to the first three orbits in Bohr's hydrogen atom model. In
order to have n complete wave lengths(A,) fit into the circumference of the n' orbit,

117
 Applied Physics

Antinode

nodes jE a I eer nodes

Three waves

Fig. 3.1(b)

the following relationship must be true:

Messicc iw extalitecn wll Reunll dart wares te (3.22:)

where r,is the radius of the Bohr's orbit and is given by

n’*h?
de DEE ee ble ee ee ay ere eae 73
4n?kme? a)
where n —> orbit number, h —» Planck's Constant, k —» Boltzmann's Constant, m — mass
of electron and e-» charge of electron.

Substituting equation (3.3) in equation (3.2), we get

nh?
=> —————_-
= — X
h oh
"2 nkme2 m_ 2nke2

The second term in equation (3.4), 5 .he : ;

5 is the inverse of the velocity of the electron
tke
in Bohr's model, i-e.,

2nke?
8) =

nh

118
Principles of Quantum Mechanics

Therefore from equation (3.4), we get

The above relationship states the de Broglie's fundamental hypothesis, that "the
wavelength of the wave associated with a moving particle is equal to Planck's quantum
constant divided by the momentum of the particle".
If the electron behaves like a wave, as per de Broglie's hypothesis in an atomic
orbit, and accompanied by the de Broglie's wavelength, then the same might be true for
the free flight of electrons as observed in free electron beams. And, if the motion of
electrons in the beams is associated with some kind of waves, we should be able to
observe the phenomenon of "Interference" and "Diffraction" of the electron beams in
the same way that we observe these phenomena in beams of light. For a stream of
moving electrons, we assume the formula for the wavelength as that applied to orbital
electrons within the atom:

ities RLS (3.6)

If we calculate the wavelength (A) associated with an electron for a given velocity,
v = 10’m/s, and mass m = 9.11 x lO. h = 6.63 x 10734JS, we get,

X= 7.28 x 107!!m or 0.728 A®

Hence, the wavelength comes out to be shorter than visible light but is comparable
with the wavelength of X-rays (107!°m or 1A°).

33 Characteristics of de Broglie waves :

(i) The de Broglie matter waves could be considered as a wave packet and the
amplitude of such a wave packet is related to probability of detecting the
particle at a particular position at a given time. This amplitude of the wave
packet changes in space and hence it is thought to represent the chance of
finding the particle at a given place. It has a maxima at a certain place at a
given time indicating there by, the probability of detecting the particle at that
place is maximum.
(ii) A wave packet consists of many waves. The particle moves with a velocity
(v), where as a wave associated with the packet travels with a different velocity
known as phase velocity (u). The particle velocity is equal to the group
velocity of all the waves participating within the group (wave packet). We
Cc z
have a relation between v and u as, U=—.,, where v is always less than the
v
light velocity (c). However this expression indicates that u, the phase velocity
is larger than 'c', which is not valid in the case of normal electromagnetic
radiation. Hence the de Broglie waves which form into wave packets are not

119
 Applied Physics

of the kind of normal electromagnetic radiation type. There is no physical

significance attached to this phase velocity and could be thought as the average
velocity of such bunch of waves which make the wave packet, associated
with the particle.

The concept of wave group and group velocity is discussed here under.

3.4 Wave groups or Wave packet concept:

Let us consider superposition of 'n' one-dimensional waves each described by
the equation,
Vig = ASS Unk Re a eh ake ci (3.7)

where k, = (2%,|and a, = (2nf,) represent propagation constant and angular

frequency of the n'" wave. If we use the normal principle of superposition of waves,
the mathematics involved would be very tedious. Hence let us apply the phasor diagram
as shown in the Fig. 3.2(a). Each phasor vector (line) represents each wave and their
relative position in a rotation around the axis represents the sinusoidal motion of the
wave at a constant time or at a constant place in space. Let us try to fix the position in
space and see the effect of different waves (with different ) passing by simultaneously
with time on the overall amplitude. If we imagine that the angular frequencies are
related to each other through
S6@=@) — 0) =O; — W, =.....= ©, —O,_]
Then we have
ee Came
Oise =21tthiceeric gests ier At SORE (3.8)
If all the phases are assumed to be parallel to each other at t = 0, as time passess
on since the w's are different, each would have covered different angles as indicated in
Fig. 3.2(b). The angle between the phasors will be given by

86=(60)t= a =a)t
adt (3.9)
(i) ee ach nip ear dice

120
 Principles of Quantum Mechanics

()

Big. 3.2.4, bc. d)

The magnitude of the vector sum is the amplitude of the resulting wave, while
the projection of the resultant on the vertical axis is the instantaneous displacement.
We assume that 5@ is so small that the amplitudes do not change very much during one
revolution. The amplitude represents the envelope of the oscillations and we will
determine the general shape of it. As the phasors start rotating from time t = 0, at
different instances of time the phasors reach such positions where their sum (vectorial)
becomes zero. This first zero occurs when the fan (four vectors) covers the entire
angle '2n', spreading equally in the four quadrants, and the angle between adjacent

phases is ks At a time t = t,, if the first zero amplitude occurs, then from
n
equation (3.9),
on. 0, =O
= ty
n n=l
2x (n-1)
Le. cpp Vite oa Sane on (Oe Tee (3.10)

For some interval of time after t,, the amplitude would be small, since the
phasors are still spreadout. However, after sometime, all the phasors would catchup
with each other and reach a position that of at t= 0. Fig. 3.2(d) shows the situation
just before this occur. Let us call this time (Fig. 3.2(d) situation) as t, and the angle
between each phase is almost 2x. Now equation (3.9) is written as
 Applied Physics

After time 't,', the process described above gets repeated. The above procedure
produces resultant waves as shown in Fig. 3.3. A similar phenomena also occurs
when plotted as a function of position (x). In this Fig. 3,3, we observe, that there are
regions of very large amplitudes with some time lapse (t,), but the amplitudes of other
waves are not zero as

Displacement ee eeyctoe ee

od time
es v, Group velocity

required by the concept of Wave packet, associated with a particle. As per this, there
should be only one region with maximum amplitude and all the other should have zero
amplitude and amplitude should almost remain zero for rest of the period. We can see,
that, this (single packet) can be achieved, if we make the time 't,', to achieve the
maxima, to be equal to infinity. Equation (3.11) makes it possible if we make n= ©.
Then we achieve a single maxima in infinite time, i.e., only on big packet is associated
with a particle, which is called as 'Wave Packet’. The size of the group in time is 2t,
(since 't,' represents only half the packet) which we call the time duration of the
group. For an infinite number of waves, equation (3.10) becomes.

21
t,(n=00)= Gaga
(n =0o)=-_——_ orrewotl. suohwnine Ie (3.12)
If ©, — ©, = A is the frequence spread, then

A®
where At is the time duration of the wave group. Similarly we get, spatial size of the
group Ax as

an
I=
Ak"

122
Principles of Quantum Mechanics

Let us now find out with what velocity this wave group (Wave Packet) travels.
If the phase velocity of each of the waves participating is u = fA = w/k, in a
non-dispersive medium (where velocity is not a function of wavelength) all the waves
move with the same velocity and group also moves with the same velocity. But in a
dispersive medium, the group velocity differs from that of phase velocity. For simplicity,
let us consider two phase waves (which are part of the group)

Wy Com Re thy Ht) © Se aeteres (3.13)

and ¥,— A Cos(k,x — ®t)

where u, =°! and U7 = a

ky Ky
The resultant wave becomes

1 1
y=y,+y2=2A cos l(k, —k, )x -(o, —o, )t} cos l(k, +k,)x -(0, amot]

Ifk, —k, = Ak and @,

— @, = Ao,

i.e., the differences are very small, then u (k, + k,)*k, and :(o, +@5)® O,
2

Then, the resultant wave is given by

|
cos 5 [dk x—do tly =2A 00s =(dkx—dot)x cos(k,x aa;t) Puc meee (3.14)

In the above equation (3.14), the second cosine term is the original wave function,
term contains a
and rest of the term is the amplitude of that wave. As the amplitude
for this resultant wave,
cosine term, which is a function of x and t, we observe that
function. This
the amplitude doesn't remain constant with time but varies as cosine
original wave has a long waveleng th of m/dk and
"modulation" or the "cover" over the
these two, we can get the velocity of
a low frequency dw/n. From the product of
propagation of the modulation, which is the group velocity v - and

nt \( da do
Ve -(2[*)-¢ ee (3el3)

() varies with
The group velocity depends on, the way in which the frequency
propagation constant 'k’.
small experiment
The concept of group and phase can be understood through a
different from each
conducted with two pocket combs, whose teeth spacing is slightly

123
 Applied Physics

other. Keep one comb over the other, and look at them by placing them before a light
source. You would immediately see well separated dark bands. The separation depends
upon the difference in the teeth spacing between the combs. These bands are due to
‘beats’ like phenomena in sound. If we move any one comb relative to the other, we see
that the bands also start moving. However, we see the dark bands (groups) moving with
a different velocity compared to that of the combs. Direction and velocity of the bands
depends upon which comb is moved. Comb motion is like phase wave and dark band
is
the group. One will realise this easily, when he performs this experiment by himself.
Making use of the mass - energy relation from relativity (E = mC’), we can write,
2 2
v Begnn le eA as oe ee (3.16)

That wave vector k and the angular frequency @ can be written as

2p 27E h
he ees ae [from p = . and E = hf] sae (3.17)

Now the using equation (3.15) and (3.17) the group velocit
y vb, can be
written as

do dE
Dei Gree a coon on hig aa (3.18)
According to relativistic mechanics, the energy of a particl
e is
= 2 - =
E = m,c* + Kinetic energy

Lents pe ya SG a
= (POR Diatys
Vl—v?/c? ek Za

aes eee 2.4 AYA) DD

mge —M,V ¢ Se
1—v*/c? :

where my — rest mass, p —> momentum, E —» energy

and c -» velocity oflight. By
differentiating, we obtain

2EdE = 2pc? dp

cBpiee bei
dp E

124
Principles of Quantum Mechanics

From equations (3.18) and (3.19)

dE pe?
eee eh PT) UP ae, (3.20 a)

We shall now get the relation between group velocity (V,) and the wavevelocity
or phase velocity.

0) da
The wavevelocity i iaing and group velocity we a and
k

poe
shed ta wang pele eA)

ORLA
py
' do do da_ 92) P dr
we have siaientfipeiiecis
wall tir tee!) Thon4Be
2

d ee ee ee Re (3.20 b)
dit ora da g
The above equation gives the relationship between the group velocity and the
phase velocity. For a dispersive medium, V,, = f(X) and V, < V.. This is the case of
deBroglie waves. For a nondispersive medium, v, # f (A) and V, = V,. This is the
case of electro-magnetic waves in vacuum. The particle velocity v = c7/v,.
This equation is same as equation (3.16), which is the expression for the velocity
of the particle. Hence we observe that the group velocity of the wave-packet and
velocity of the particle are same. Therefore, we can conclude that the wave-packet as
a whole is associated with the particle.
Many explanations and possible interpretations were putforth by many, about
the dual nature of matter particles. However all the arguments finally lead to the
following conclusions:
Both wave and the particle are the models we have constructed in attempting
to describe matter. We do not expect the model to give a complete description.
The two models complement each other in that, together they give a description of
matter. Thus, we should say that the electrons are "waves and particles", not
"waves or particles". The same conclusion can be drawn about electromagnetic
radiation. During an experiment, the particular model that is used is determined by
the apparatus used.
Even though de Broglie proposed the hypothesis, the verification of wave nature
of a particle was done through electron diffraction experiments by GP. Thomson
(1928) son of J. J. Thomson and C. J. Davisson and L. H. Germer (1927) separately.
These two are discussed in the following sections.

125
 Applied Physics

3.4 Davisson and Germer Experiment (1927) :

Davisson and Germer were studying the scattering of electrons by a nickel
block containing many tiny crystals. The experiment was to send a beam of electrons
on to the target (Ni block) and study the intensity of the scattered beam of electrons at
different angles, as in the case of any scattering experiments like that of Rutherfod's
alpha-particle scattering experiment. Generally, one would expect the intensity of
scattered beam to decrease with increasing angle away from the incident beam
continuously. During the experiment, the vacuum broke and they had to remove the
set up and start over again after cleaning the Ni surface for the oxide formed on it due
to heating of the crystal during the experiment due to impinging electrons. To their
surprise they found that, the intensity of scattered electrons was not decreasing
continuously but went through maxima and minima, an in the case of diffraction.
Hence, they investigated this carefully and found that electrons were infact getting
diffracted like waves as predicted by de Broglie. They calculated the wavelength of
these electrons and it matched with that obtained from de Broglie formula. Thus, the
proof for matter waves came in an accidental way. The reason for obtaining such
maxima and minima in the second run was found to be due to crystallisation of
polycrystalline nickel almost into a large single crystal due to heat treatment it has
undergone during the first run. Let us now look into their experiment and the results
obtained in detail.
The schematic diagram of the experimental setup used by them is shown in
Figure (3.4). The electron beam produced by the electron gun 'EG' is focussed and
IPSILON EASES YARUSHAMUS ISSESEORCEAER OSRRLELESSEEAASIOYP AMESIRSIANSARI GO
- Vacuum Chamber

Ren
Target (Ni)

for i L
rotation; Focussing :
odie pecs anode :
crystal |oe
diffracted é

beam

|
z

RASA
=

Fig. 3.4 Schematic diagram of Davisson and Germer experimental set up.

accelerated by the anode 'A'. A narrow beam of electrons thus accelerated is directed
on to a target, a large Ni crystal, which could be rotated about an axis parallel to the
axis of the incident beam. The electrons which fall on the target are reflected in
different directions. The electron collector C, which is connected to a galvanometer
(G), is used to collect the electrons coming in different directions from the target the
moving it along a circular scale, so that it is able to receive reflected electrons at all

126
Principles of Quantum Mechanics

angles between 20° to 90°. The accelerating potential used was in the range of 30 to
600 volts. The entire apparatus was enclosed in a highly evacuated chamber.
With the electron beam incident normally, a diffraction pattern as in the case of
For
a grating is produced, where the planes of the crystal act as diffraction elements.
each azimuth (orientation) of the crystal., a beam of electrons is made to fall normally
at
on the crystal. For each of these positions of the crystal the collector 'C' is moved
all angles and the current recorded by the galvanome ter is noted. The current recorded
the
in the form of deflection in galvanometer is plotted against the angle between
incident beam and the beam entering the collector, which is called the colatitude. These
observations were repeated for different accelerating voltages and a number of curves
voltage
are drawn. The graph remained fairly smooth (continuous), till the accelerating
increasing voltage, the
is around 44 volts, where small hump like curve is seen. With
54 V with the correspon ding
hump became more prominent and completely defined at
g and
colatitude of 50°. Again at higher voltage, the hump gradually started diminishin
disappeare d as shown in the Figure (3.5).
for voltages above 68 volts it completely

44 volts 48 volts

54 volts 68 volts

Fig. 3.5
evidence for the existence
The hump seen prominently at 54 volts at 50° is a convincing
the wavelength for a beam of
of the electron waves. According to de Broglie's theory,
y is given by
electrons accelerated with V volts, and velocit
h h 12:27 21)
nn ee a |
m,v ,/2Vem, ay

m,v°
since the kinetic energy E, = Ve= se

For accelerating voltage V = 54 volts, the wavelength

N= —— = 1.66.A

b27
 Applied Physics

Now according to the experiment, the maximum diffracted beam obtained is

with a colatitude of 50°. Applying the known relation of diffraction in a plane grating
na = d Sind
where n, the order is taken as units; 6 = 50° and 'd' the interplanar spacing between the
nickel atomic planes of Miller indices (111) is equal to 2.15 A , we get

dX = 2.15 Sin 50 = 1.65 A

This excellent agreement with the above clearly showed that the electrons behave
like waves. When the direction of t electron
beam and the direction of the detector are fixed
but the accelerating voltage is varied, it was
observed that the current showed maxima at
certain accelerating voltages only, corresponding
to various orders (n) Fig. 3.5. From the curves
the accelerating voltages which produce Current
maximum reflection in the different orders can
be found and hence A. calculated using de Broglie
relation. On the other hand, with the known
values of the glancing angle @ and the grating eee me
spacing 'd', using Bragg's relation, X is found. a/voltage a -

3.5 GP.Thomson Experiment (1928) Fig. 3.5

G. P. Thomson, son of J. J. Thomson extended the work on electron
diffraction
to high speed electrons by accelerating them with 10,000 to 50,000
volts and using a
very thin metallic film as a target. He used a method similar to Debye
Scherrer powder
method of x-ray analysis. The experimental set up is shown in Fig.
3.6.

Photographic
Plate

To vacuum
<—_—
pump

-|_7Fluorescent
screen

Fig. 3.6 G P. Thomson's experimental set-up for observi

ng electron diffraction.

128
Principles of Quantum Mechanics

As shown in figure, a beam of electrons is produced in a discharge tube with

cathode K, by means of an induction coil. In order to obtain a fine pencil of electron
beam, electrons were passed through a fine tube 'A' (Anode). This beam was made to
incident on a thin metallic (Al or Au) foil (M) of thickness 10° cm. A photographic plate
can be introduced at the points PP' to record the picture after the electrons pass through
the film. A fluorescent screen was used for visual examination, before introducing the
photographic plate. The camera part of the discharge tube was maintained at a high
vacuum, even when air was allowed to leak through a needle valve into the discharge
tube part. Now a beam of electrons of known velocity (for a given accelerating voltage)
were made to fall on the photographic plate, after passing through the thin metal foil.
When the plate was developed, a symmetrical pattern consisting of concentric rings
about a central spot was obtained (Fig. 3.7(a)). This pattern is similar to that observed
produced by x-rays in the powdered crystal method (Fig. 3.7(b)). In order

(a) (b)

Fig. 3.7 (a) Electron diffraction pattern of aluminum. (b) X-ray diffraction pattern of aluminum,
using K-radiation from copper.

to make sure that the pattern observed is due to electrons and not due to x-rays, a
magnetic field was applied perpendicular to the path ofthe electrons and a shift in their
path of observed. On removing the metal foil the pattern disappeared showing there by
that diffraction the pattern indeed was produced by the film kept in the path of the
electrons. Clearly this experiment demonstrated that the electrons behave like waves,
since diffraction can be produced only by waves. The de Broglie equation was verified
by G. P. Thomson, by estimating the wavelength of the electrons with the knowledge
of their velocity which depends on accelerating voltage. As in the case of
Debye-Scherrer method, from the knowledge of the geometry of the set-up, such as,
distance between the metallic film and the photographic plate, the radius of each ring,
the wavelength of the electrons was calculated as shown :
OA is the radius (r) of the ring. 'O' is the centre ofthe ring, Q is the point where
the electron strikes a particular plane on the crystal in the metallic film. QO (L) is the
distance between the metallic film and the photographic plate.

129
 Applied Physics

Brage's law is nA = 2d Sind.

From the Fig. (3.8),

os tan 20 ~ 20 (0 being small)

Ne 0
2L

Now Bragg's law is written as

ni=2d.0.=2d.— (for small @)

2L
rd
K=—
nL

From the above expression, the wavelength of electrons was calculated which
matched well with those calculated from de Broglie equation. This provided the ultimate
confirmation for existence of matter waves or wave nature of electrons.
Now a question arises, whether all the particles (objects) in the universe behave
as waves and show diffraction. This can be answered by a small calculation of
de Broglie wavelength for different types of objects namely (i) a dust particle (ii) a
cricket ball (iii) earth etc. In each of these cases, the de Broglie wavelength has
approximately a value of 10°7> to 103° m with varying masses and velocities which
can't exceed the velocity of light. It is a well known fact that diffraction can be
observed only if the grating element, which is 'd', is of the order of wavelength of
the wave which is getting diffracted. We can see from 2d sin@ = nd or dsinO ~ id, sin®

can't exceed unity, hence d~A for 9= 90°. In the universe, we do not come across
grating elements which are of the dimensions 10-5 to 10-3°m. Hence it is not possible
to observe diffraction in the case of particles or objects with larger dimensions
than
those of electrons.
The electron diffraction is used in electron microscopes to analyse
the
microstructure of the materials and also to analyse the crystals structure.

3.6 Schrodinger wave equation :

Schrodinger, in 1926, using the ideas of matter waves, set up a wave equation
to describe a new mechanics called wave Mechanics. Schrodinger used Max
Planck’s
quantum hypothesis of energy quantisation which was proposed
for explanation of
black body radiation* (See Appendix III).

130
Principles of Quantum Mechanics

The Schrodniger equation takes the analogy from partial differential equation of

the classical w ave equation

ti @7 y/@x papi
2y/ w age
O” hereere ‘v’
‘v’ is
is the velocity the
velocity of th

wave, and y is the displacement of the particle in the medium. The solution of the
above given equation is y = A e!(-*) and only the real part of the solution is taken
into consideration. Let us see how the Schrodinger equation can be set up in the
following section.
3.7 Time-dependent Schrodinger Equation:
Generally the simplest plane monochromatic wave can be described by,

are ET ay) adelODNan-eplaciviet Bewrodings:§ (3.22)

where (r, t) is the function possessing the characteristics described for a wave
function, here 'r' is the position vector and 't' is the time.
r=xi+ yj + zk
The above equation, if written in one dimension, say for motion along X-axis, 1S,
Gedy Se Oy (AA SEP _Soease wee BS (3.23)

where K=—, o=a2nt

h
From de Broglie relationship, 4 = , we get, p = Ak and for energy E= fo where

AB
Qn
From the relations given above, equation (3.23) can be written as

(x,t)=A ex (ox-E) er (3.24)

Differentiating the above equation with respect to 't'’, we get

in EY (x,t)
Similarly differentiating (3.24) with respect to 'x’, we get,
oY
== PE (x,t)
ih —=pY(x,t

once again differentiating (3.25), we get,

2
he a? Y Jee
P w(x, t)

2m ax2 2m

13]
 Applied Physics

For a non-relativistic free particle energy

©
B=. (only Kinetic Energy)
2m
A free particle is one which moves only with Kinetic Energy. Therefore, from
equations 3.25 and 3.26 we get,

this equation is called one-dimensional, time-dependent, Schrodinger wave equation

for a free particle. The above equation suggests that Energy E and momentum p can
be represented by the operators
O O
E—> iat and Pp ——>
— jh—
Ox

Hence, equation (3.27) can be written as

2
EW (x, t)= vey ce t)
m
where E & p are operators. The total energy of the particle is
2
E— + V , where V denoes the potential energy.

V can be a function of both x and t. Therefore, now (3.27) gives a more

generalised Schrodinger equation,

The above equation governs the behavior of all particles. Extending the argument
s
to a three-dimensional system, we get,

Oe
ih ata nr(aew
é HORE: 3¥
Ma HORE. + V(r, t) Y (r,t) eee (3.29)

% Bao
Tey a
a
— =| —+
ih
aks
V(r, t)| (r,t

pean
pa ie Im
i Ve iALtv
PeVy adt—e OO! Us (3.30)
H, the Hamitonian is given by,
2 2
H=—+p V=-—_V h 2
Ga ae ’*+V5. |A | SS IO on, ee G31)

132
Principles of Quantum Mechanics

Then, finally, we get the three dimensional time dependent Schrodinger equation as,

Ow hi?
ih—=HY
at =| -—-V’?+V|¥
sar | sent. (3.32)
or in one-dimension.

ov h? a?
Waa =i |-ES + v6) omnes(3.33)
In deriving the above equation we have made use of wave and particle nautre of
matter. As mentioned earlier the wave function ¥ is in general complex. Above
Schrodinger equations are valid for particles moving with velocities much less than the
velocity of light. Hence eq. (3.33) is known as Non-relativistic Schrodinger Equation.
3.8 The Time-independent Schrodinger Equation:
As shown in equation (3.26), any wave function ¥ (x, t) can be written as,
¥(x, t) = wx) o
then, by the method of seperation of variables, we can solve ‘¥(x, t) by solving for
w(x) and @ (t) seperately. Substituting the above equation in time dependent Schrodinger
equation (3.33) we get,
2 ad
_ WO)
vyyg= in VD)
2m 0x? ot
where we have dropped the arguments i.e., the terms in the brackets, and we write y
instead of w(x) and 6 for (t). Since y is a function of x only and independent of 't'
and similarly $ is independent of x and dependent on't' only, the partial derivatives can
be replaced with ordinary derivatives. Thus, we get,

a d? ;
67 +Vwe = ihy do
2m ot
If this equation is divided by ¥ = y9, we get,

We assume that V, the potential, is independent of time and is dependent on x

only. Then the left hand side of the above equation is independent of time and right
hand side is independent of 'x'. Since x and 't' are independent variables, the only way
a function of 'x' can always equal a function of 't' is to have both functions equal to a
constant. The constant, that each side of equation (3.34) must equal, is called the
separation constant E and we obtain two equations:

133
 Applied Physics

and i Ane ES eee (3.36)

That is, we have seperated the partial differntial equation into two ordinary
differential equations which can be written as,
Die?
Sn dy
Sere a EV AA ae ne ee (3.37)
Sew)
called the time-independent Schrodinger equation, and
., do
SS sf EGho hie SA Svs NES WS a ga (3.38)
3.38
Since 'V' is assumed to be independent of 't' (time), hence, equation (3.38) can
be solved by writing it as,
do we
dt A
and the solution is

pede ok eC Oa. | (3.39)

by integrating on both sides with '' and 't'. Equation (3.39) gives the time dependance
of the product solution, yo, and we have

ee-1Et/h ee (3.40)
Because this is similar to a sinusoidal equation, y is called the amplitude and the
differential equation (3.37), is called amplitude equation.
Since must be well behaved, as per definition of wavefunction, the seperation
constant E is restricted to a real number. If E were complex, the exponential would
have a real part and would approach infinity as t approached infinity. In order to know
what 'E' represents, we shall calculate its expectation value from

os a
2B = J ve
Sy oy
—— dx
— 0
co [o@)
a fuel inShyerCE Mey as fu’ w Pe
—00 = 100)

foe}

Since the wavefunction y is normalized, that is fv" Widx — 1 <=" =-F and
similarly <E*>=E?. This yielding a single value for the probability, and hence it is an
=O)

observable. This E ultimately gives the total energy (Kinetic + Potential

).
Schrodinger equation (3.37) describes systems which are said to represen The above
states, there by indicating that the system is stationary in time, t stationary
as potential V is
independent of time. In these states, the position probability density
|Y/ at every point
'x' in space remains independent of time, so also the expectation
values of the all
dynamical variables. The energy of a particle described by a wave
function has a
definite value and in general shows discrete values. Let us see
this in the case of a
particle present in a box in section 3.10.

134
Principles of Quantum Mechanics

Physical significance of Wave Function:

‘V(x, t) is the wave function. V(x) is the potential function assumed to be
independent of time, and m is the mass of the particle. The wave function ‘P(x, t) is
used to describe the behaviour of the system and, mathematically, ‘¥(x, t) can be a
complex quantity. H and E are called Hamiltonian and energy operators respectively.
Max Born postulated in 1926 that function |Y(x, t)/? is the probability of finding
the particle between x and x + dx at a given time, or that |W(x, t)/? is a probability
density function. This is the function that has a physical meaning. The probability
density P = |W? =¥*, where Y* is the complex conjugate of Y, any wave function
can be represented as
P(x, t) = w(x) - O(t)
Where w(x) is independent of time and $(t) is independent of x, the Pues
density can be expressed as
Oe ODS QUGOD? catrieeh nciuipe seers wedtent (3.41)
and is, therefore, independent of the time. Computations often become much simpler
when one considers the time-independent wave function first and deals with the time
component later. Ofcourse, this is only possible when the wave function can be
factored by separation of variables, as shown above in section (3.8).
The time independent Schrodinger wave equation can be written as
H (x) = E y(x)
or SVE),2m(6 Vix) =0
d7 w(x

Since the function |w(x)|* represents the probability density function, then for a
single particle, we must have that
(6)

f(|y(x)| Pax =1 Oona (2.34ia (3.43)

=O).

The above reation is called normalization condition. The wave function that
satesfies the relation (3.43) is called a normalized wave function.
The probability of finding the particle some where is certain. This equation
allows us to normalize the wave function. The wave function Y should be finite, single
valued, continuous and well behaved.
Postulates of Wave Mechanics or Quantum Mechanics:
Wave Mechanics, the mechanics of quanta (discrete packets of energy), can be
developed from a set of postulates and this is known as axiomatic formulation of
quantum mechanics. Following are the fundamental postulates of quantum mechanics:
1) Dynamical states of a particle or a quantum mechanical system can be described by
a wave function which contains all the information about the particle. The wave
function should be single valued, finite and continuous. It should satisfy orthonormality
condition i.e.

135
 Applied Physics

Ree tes. (3.44)

‘Y* = complex conjugate of
oF Oforijs
arene 4
dt = dx dy dz = volume element.
2) Wave TUNCHONS CE patito etict) represent the physical states of a system
and satisfy linear super position theorem.

1.€., w=) CV = CVG Ck... a: (3.45)

‘Y; = discrete states
C; = constants
3) Schrodingers wave equation describes the behaviour of wave function in time and
space (i.e., position).
aioe @
H'Y (x, t) = ees saspens (3.46)
Ot
Time dependent equation is
(ee ao
— am pet POst)+ VON (x,t)= ik S(t) ae a (3.47)

d°y(x) 2m
dx
eS h
| 4)I ES ae ie ieee (3.48)
4) In classical physics, certain definite functions are associated with observable
quantities like the position, momentum, energy etc. In wave mechanics or quantum
mechanics, certain operators are associated with the observable quantities like
momentum, angular momentum, kinetic energy, etc. The operators are given in the
table given below. Every observable (physical parameter) in quantum mechanics
is
represented by operators.
Variable Operator
Momentum p, —1 hO/ 0x
Energy i hd/ot
5) The only possible value, which a measurement of a variable p can yield, is
given by
the equation
PEP Roto E> SSS Wnt ee ae SOR See Cee (3.49)
where P - = operator of variable p

W — = wave function

x — = eigen value

136
Principles of Quantum Mechanics

Therefore, a linear eigen value equaiton is linked with each operator. y is called

; 6 a 9 8) ;
the eigen function of the operator i sacle and A is called the eigen value. The
Tt 0
word, eigen vlaue means " proper or characterstic value". The eigen values are not
restricted to energy alone. They refer to all dynamical variables.

6) When a given system is in a state, the expected mean of the sequence of

measurements of the observable represented by operator A is given by expectation
value of operator A i.e.,

_ PAY
<A> = oseooeeeeee es
[PY Wdr
‘Y* = Complex conjugate of Y
dt = volume element.

Simple Steps to Solve Schrondinger Wave Equation:

The Schrondiger equation is a linear partial differntial equation and admits an

infinite number of solutions. Solving Schrodinger equation for a particular system,
whose potential is known, means to obtain a wave function YY that, it not only obeys
the wave equation but also the boundary conditions. Such boundary conditions fulfill
the requirements of an acceptable wave function. The conditions on wave functions
are given already earlier.
General procedure to solve problems in Wave Mechanics involves the
following steps :
Step II : Set up Schrodingers equation appropriate to the problem.

Step II : Express the operators of Schrodingers equation in appropriate co-ordinate

system (Cartesian, spherical, cylindrical etc.) taking into account the
symmetry considerations of V(x), the potential function.

Step III : Solve Schrodinger wave equation to obtain wave function and eigen
values E.

Step IV : Apply boundary conditions as per the problem requirement.

The procedure for solution of Schrodinger wave equation is illustrated in

section 3.10 for infinite square well potential.

137
 Applied Physics

3.9 Uncertainty Principle:

In classical mechanics based on Newton's Laws, the state of a particle in space
is uniquely defined by position and momentum coordinates. The particles follow
definite trajectories in space. But a quantum particle is different. Each moving particle
is associated with a wave packet that extends in space. If a particle is moving in
x - direction, its position can be any where within the wave packet. Therefore, we
can only define its position to a certain degree of accuracy, Ax, but its momentum
can be defined to a greater accuracy. If the wave extends from minus infinity to plus
infinity, the position is inaccurate to the level of infinity and the momentum of the
particle can be determined precisely.
Suppose, the particle is somewhat confined to within a distance, 'Ax' during
our measurement. Then, the momentum of the particle is
p=ke UNAM 2 - Sle (3.51)
2
(k is the wave vector or propogation constant, k = wv

But Ax Ak ~ 1, i.e., a combination of waves with wave numbers in the range

Ak will yield a wave group spread out in space Ax, where the product is of the order
of unity.
Therefore, a moving particle is associated with a momentum and its momentum
is confined to a width Ap, then,

ies TS GGNCD SAR Rae TM Ta OeOl ae ee MCR (3.52)

The above results gives the minimum uncertainty introduced in the measurement.
The above, expression is normally written as Ax Ap fax it
The above expression is one form of uncertainty relations postulated by
Heisenberg in 1927. The above form of inequality means the following :
A particle can not have simultaneously a definite coordinate and a definite
momentum ( or velocity ). If Ax, Ay, Az and Ap, Ap,» Ap, are
uncertainties in
coordinates x, y, z and projections of momentum P on to the axes respectiv
ely, we
have,
Ax Ap, 2 h; Ay Ap, 2 h; Az Ap, = h.
The above inequalities also lead to the following : The smaller the error
in one
of the quantities, for instance x, the greater is the error in the other,
namely p,. All
dynamical variables like position, momentum, energy, angular
momentum etc. form
pairs coupled by uncertainty relation. The uncertainty relation
of Heisenberg and
de Broglie relation are the two basic postulates which are fundame
ntal in quantum
mechanics. For energy and time, we have the relation as

CNS a ie MR (3.53)
This relation tells us that the uncertainty of energy of the particle
is dependent
on the ume interval used. The particle must be observed for
a time At to be certain that
the particle is at the point of observation.

138
Principles of Quantum Mechanics

Example : An electron remains in the excited state for 10-8 sec, Then, the
energy of the excited state is

h ]
NE) ee a ee = 0.66 x 10°8 eV.
(At) ORS
The above parameter represents the limit of accuracy with which the energy of
an excited state can ever be measured. It gives the natural width of the excited state.
We shall consider two examples to illustrate the uncertainty principle.
Microscope :
Consider that we have to measure the position of a moving electron by means
of a microscope ( y - ray microscope ). aA

When the electron passes through the field

view of the objective, the microscope must be able
to resolve between two positions of the electron.
The resolving power of microscope is (Fig 3.9).

Dane
Ax
2Sin9

Where 'X' is the wavelength of the radiation

used for observation, '@' is the semi-vertical angle
of the cone of light and Ax, the uncertainty in
determining the position of the particle. 'Ax'
represents the minimum distance between the points particle
that can be resolved. If the electron has to be
observed, the light quanta i.e., photon, has to strike Fig. 3.9 Electron position determination
the electron and get reflected into the objective and by means of ay - ray microscope
eye piece. That means, the photons can be
anywhere in the cone of angle 20. If p is the momentum of the electron, the photon
suffers a recoil with momenta +p Sin 9 and —p Sin 8. Therefore, the component of
momentum of the electron, perpendicular to the axis of the microscope after the
process of recoil is indeterminate to the extent of Ap which is given as
A p, =p Sin ®-(—p Sin@)=2p Sin ®

ru h
Hence, Ap, Ax 2 p Sin @ x Sind 2pr 2o(*) 2heh

Ax Ap, 2 h or Ax Ap, mee, Sgt e Eee Gid3)

Electron Diffraction :

The second illustration is the consideration of diffraction through a single slit.

Une
Let a narrow beam of electrons fall on a slit of width Ay as shown in Fig. S02
There is no
electrons have a velocity i.e., the momentum along the x - direction.

139
 Applied Physics

component of momentum along

y - direction. After the slit is crossed,
the electrons will have component
of momentum along y direction. go),

This can be understood from the fact

that we get diffraction patternonthe
screen S. At C we get maximum K
intensity, called the principal
“<<
maximum. At P' we get the first
ji direction
order of the diffraction minimum.
The diffraction condition is
Oe
1 SiG Fig. 3.10

where @ is the angle of diffraction. Ay represents the uncertainty in the position of

electron at the slit as the electron can enter into the slit anywhere within Ay. The
observation of diffraction pattern indicates that the electrons have passed through the
slit. The diffraction pattern consists of two first order minima on the either side of the
central maximum (PP') The momentum component of the electron along the
y-axis
is now p Sin 0.

h h
Then, Ay Sin@=2= — ( Since
1,= —)
p p

an d A be?
~ ye
—x = ae
—: ( Sinceince A Ay = :. )

Ax Apy =)
Thus, h represents an absolute limit to the simultaneous measurement
of
coordinate and momentum. The value o, h, is very small, of order
of 107, Therefore,
the magnitudes of the experimental variables become minute in the
case of atomic-
sized particles, the limitations of accuracy become important.
Applications of Uncertainty Principle :
We can now explain the absence of electrons in the nucleus of
an atom. The
radius of a nucleus is of the order of 107!4 m. If an electron
were present in the
nucleus, the uncertainty in its position is of the order of 10-!4 m. Then
A Py IN By FF;

Tames alae
IB ais fees 194
a VX 102° Ke ins

140
Principles of Quantum Mechanics

For this momentum, the kinetic energy according to relativistic theory is

K.E=pc=1.1 x 10° x 3 x 108= 3.3 x 10-22J
The above result indicates that the kinetic energy must be greater than 20 MeV
in the nucleus. But, it was observed that electrons can not have energies more than
4.5 MeV. The energy is far less than 20 MeV. Therefore, the electrons can not exist
in a nucleus. It can also be proved that the electrons possess de Broglie wave length
equal to the circumference of the electron orbit.
3.10 Application of Schrodinger wave equation :
Particle in a Potential Box (ne dimensional potential well) :
We shall now take up an example of particles constrained y= oo V= 00
to remain localized in some part of space. Let us assume that _
particles are trapped in a potential well with very steep sides as
shown in the Fig.3.11. A particle of mass m moving in the
x - direction width ‘a' potential barriers at x = 0 and x = a. The
barriers are artificially mode steep and high and are infinite in
extent. Therefore, no particle can escape by surmounting the ie aes corr
barriers.
Fig. 3.11 Potential Box
The potential well can mathematically be described as

V (x) = infinity for x <0,

=( forO<x<a
= infinity forx>a

The time independent Schrodinger wave equation is

2 any"
vs) + (=) 2m [E—V@]Y(X)=0 cesses (3.54)
XxX

Where E is the total energy ofthe particle. If E is finite, the wave function must be
zero in the regions outside the box i.e.,
W(x) = 0 forx =Oandx=a
w(x) = finite for0<x<a

Inside the box, we take the potential energy of the particles as zero. Then

d2 2mE
— vx) + =z wx) =0 forO<x<a
dx h

141
 Applied Physics

To find general solution, the above equation can be written as

d*y
—— + |2 =,
dx? “

where k? = ae

This equation has a solution of the form

W(X) = A-COS (KX) eB SIAR) ee ee (55)
Applying the boundary conditions, we get,
w (0) =A cos (k x 0) + B sin (k x 0) =0
.. The relation (3.55) is valid only if we take A = 0. Hence, the solution becomes
vy (x) =B sinkx
At X =a, w (a) =0
B sin k.a=0
ee sinka=0
1.6. ka=nnt

eae
The solution can now be written as

_ (nt
W(x) = B sin Fra ede bees(3.56)

; 2m
Since k= . v¥2mE

h7k2 n2h2
we get energy E.= = yt 1) GRE went Biri.ce (3.57)

We can get the energy spectrum of the particles in the potential well from the
above equation. The ground state energy is obtained for n = 1. It value is h2/8m
ars
The second level (n = 2) has an energy E, = 4 h?/8ma2,

142
Principles of Quantum Mechanics

Similarly E, = 9 h?/8ma*, E, = 16 h? / 8ma and so on. The energy spectrum

is given in the Fig. 3.12.

Wave Functions Probability Functions

ws)

Sisaanear rTM
Eu x=3 ‘

x=2 :
1 2 ae eae ts Le ae eR .i
x=1 |

Fig. 3.12
The total energy of the particle is quantized. That is, the energy of the particle
can only have certain discrete values. The discrete values of energy are called eigen
values. Note that the ground state energy is not zero. The quantization of the particle
energy is contrary to the concepts in classical mechanics where particles can have
continuous energy values.

143
 Applied Physics

The wave function is given by

yw =B sin
mx)
-

The constant B is to be determined. It can be obtained from the fact that the
probability that the particle is located somewhere in the well must be unity. The
probability that a particle is located in length dx is yy* dx. Then

[vw*dx =1
0

The above condition is called the normalization condition. Substituting for w(x),
and since (x) is real (y = w*), we get,

Jlv(x)Pax =|
0

Le. [B?sin (nx /a)dx rol

0

B= V2/a

Hence, the normalized solution of y becomes

NTX
w= J2/a sin =) tae (3.58)

Therefore, the wave function of a bound particle in a potential well is one of

a
set of discrete values, each one corresponding to a different value of the integer
'n’.
This is true for all bound particles. Each allowed value of the wave function
is called
an eigen function. The diagram of Fig.3.12 shows that a particle in the gound
state is
most probably located at the centre of the well. For higher energy state like
E,, the
particle can be located at some position removed from the centre of the
well. There
are positions at which there is a zero probability of finding the particles.

144
Principles of Quantum Mechanics

Excercise 1 :

Determine the wave length associated with an electron having kinetic energy equal to
lKeV.
Energy = 1 x 103 ev= 10? 1.6 x 10719

P= 42x10? x1.6x107!? x9.1x1072” (Cop = Oink)

6.6x10724 A h
per ete ey aT eo ee
210° x16x10729 x9x1072! ( p?

Excercise 2 :

Determine the wavelength associated with a particle like an electron having energy
much greater than its rest mass energy.
The rest mass energy of an electron
me = 9 101 Se(3 710°)"
=82~x 10-4 J
= 0.512 x 10° eV Ct 16 We ey)
Let us take the example of an electron that has an energy 1 MeV.
The total energy = (0.512 x 10®+ 1 x 10%) eV = 1.512 MeV

_ 1512x108
x16 «10-77
m = 2.19 x 1039 Kg
6108
We have for the relativistic particle

ot
2 m2

pe
|(108 = 2.82 x 108 m/s

pub 6.6x10>4 ° h
= 0.00087 A (“A=—=—)
x2.82 x108
~ 2.69x10729 p mv

145
 ey nme Applied
ee eee uO Ae Physics
foe
Excercise 3 :

An electron beam associated with a de Broglie wavelength of 0.2 A falls on a thin

film of
nickel. The inter planar distance for (1 1 1) planes in nickel is 2.04 A . What is

the angle of diffraction for the first order reflection from (1 1 1) planes of a nickel
film?
2d sin@ =A
2 x 2.04 sin 8 = 0.2

ew in 00 ==
sin we
DaAgd = 0.049
aoe

8 = 2.89
Excercise 4 :

A free electron is confined to a potential well of infinite height and

width 1 A . Com-
pute the energy difference between the first two energy states.

h? : 66.105"
= 37.81 eV.
8x9x 1073! x 10729 x 16% 19720

E,=4 x E, = 151.24eV
E, —E, = 113.43
eV.

Questions and Problems

I. Explain the dual nature light, both as a particle and

wave.
2. What are matter waves ? Who proposed this conce
pt ? How did he come to this
conclusion ? What is its importance ?
3. Calculate the de Broglie wavelength of a dust
particle of mass 0.001mg and
moving with velocity 104m/s. Can we observe
diffraction due to dust particle
wave ? Give reasons for your conclusion.

146
Principles of Quantum Mechanics

Calculate the de Broglie wavelength of a cricket ball of mass 500 gm and moving
with velocity 150 km/hr. What should be the size of the grating element to
observe diffraction with a cricket ball ?

Distinguish between group velocity and phase velocity of waves. Show that the
group velocity Vg is equal to the particle velocity.

How did Davisson and Germer detect the presence of matter waves ?

What are the important conclusions of G. P. Thomson experiment ? How did he

arrive at them ?
If an electron is accelerated with a potential difference of 10,000 volts, what
would be its de Broglie wavelength ? Take standard data for the quantities required.
What is the difference between any function y = X(t) and a wave function
Y(x, t). What are the characteristics of a wavefunction ?

Calculate the energies that can be possessed by a particle of mass 9 x 103!kg

which is placed in an infinite potential box of width 10-8 cm.
Write the time-dependent schrodinger wave equation and obtain the time-
independent Schrodinger wave equation from that.
Obtain the energy spectrum ofa free particle moving in a potential box of height
v=

A body has a mass of 0.66 gm and moves with a velocity 2 x 10° cm/s. What
is the de Broglie wavelength associated with it, can you get diffraction with a
beam of such bodies from a nickel foil?

A particle of mass 1 gram is placed in a box of side 10m. A free electron

(m = 9.1 x 10-3! kg) is placed in a box of side 1 A . What are the differences

of energy in the first two levels in both the cases? Which case is possible?

A moving electron has a kinetic energy 64 eV. Compare it with the rest mass
energy of the electron. What is the de Brogile wavelength of the electron waves
(mass of the electron ~ 9 x 1073! kg).
An a - particle is accelerated by a potential difference of 25 KV. What is the de
Broglie wavelength if mass of the a - particle is 4 x 1.67 x 10°?! kg.

[ Hint : use the relation: 7 = h/,/4eVmy ]

147
 Applied Physics

LT: Show that if the uncertainty in the location of a particle is equal to the de
Broglie wavelength, then the magnitude of uncertainty in its velocity is equal to
the magnitude of the velocity of the particle.

What are the conclusions one can draw from the solutions of Schrodinger
wave equation for a particle in a potential box of width a ?

What do you understand by the terms ‘eigen value' and ‘eigen function’ ?

Show that a particle of mass m in a potential well ofinfinite height and width 'a'
possesses discrete energies.

2. Neutrons are in thermal equilibrium at 3000° K. Calculate the energy in eV and

the de Broglie wavelength.

ID Write a short note on 1. de Broglie wavelength 2. Heisenberg's uncertainty

relation 3. electron diffraction

ao. Write short note on ‘uncertainty principle’.

24. Explain why electron can not be present in a nucleus.

DY Give at least one illustration of Heisenberg's uncertainty principle.

26. An electron is confined in a one-dimensional potential well length 0.1 nm. What
is the kinetic energy of the electron in the ground state.

Zit A photon passes through a horizontal slit of width 5 x 10-° m. What is the
uncertainty in vertical momentum ?

28. A slit of width 0.01 mm is placed across the path of a plane light wave. What
are the uncertainties in the transverse position and momentum of a photon that
passes through the slit ?

148
A
ELECTRON THEORY OF METALS

4.1 Introduction:
Metals occupy a rather special position in the study of solids, because of their
variety of striking properties compared to other solids such as quartz, sulfur, or common
salt. They are very good conductors of electricity and heat, are malleable and ductile
and display a striking luster on a freshly cut surface. The challenge to explain the
properties gave the impetus to the modern theory of solids.
Although the majority of solids are non-metallic, metals continued to play
important role in theory of solids right from the beginning of twentieth century, to this
date. Many of the elements in the periodic table belong to the category of metals, as
they favour this state. Even to understand non-metallic behaviour, one must first
understand metallic behaviour for explaining why copper conducts so well, one begins
to learn why common salt does not.
During twentieth century, physicists have tried to explain through simple theories
the metallic behaviour both qualitatively and quantitatively. During this process, success
and failure were tasted time and again. Even the earliest models like free electron
theory (Drude, Sommerfeld) though strikingly wrong in some aspects, remain, when
properly used, of immense use for solid state physicists even today.

149
 Applied Physics

42 Classical Free Electron Model:

As already mentioned, immediately after the discovery of electron by J.J.
Thomson, attempts were made, to explain, the physical properties of metals. Among
these, Drude (1900) explained certain properties of metals based on some assumptions.
This model is now called as “classical free-electron gas model”. According to this
model, 1) the valence electrons of the constituent atoms of the metal are able to move
about freely through the volume of the specimen. 2) The valence electrons are responsible
for the conduction of electricity by the metal, and for this reason these electrons are
termed conduction electrons, as distinguished from the electrons of the filled shells of
the ion cores. 3) The interaction of the conduction electrons with ion cores of the
original atoms is neglected in the free electron approximation. It also neglects the
interaction between the conduction electrons. Therefore, it is assumed that the
conduction electrons move freely like ideal gas molecules, with a small difference that
these electrons move between the ion cores entirely free in the space bounded by the
surface of the specimen. These ion cores and electronic gas (Cloud) are held together
due to the electrostatic interaction between the positively charged ion cores and the
negatively charged electronic cloud. Lorenz added to the assumptions, an assumption
about the distrubution of velocities among the free electron. That assumption is states
that statistical distribution of the electrons is assumed to obey Maxwell - Boltzmann
distribution, where spin of the electron and the effect of Pauli’s exclusion are not
considered. ;
In metals for which the free electron model is assumed to work well, the usefulness
of this model is greatest for experiments that depend essentially on the kinetic properties
of the conduction electrons. This is because of the presence of metallic bonding,
which assumes that in metals, the valence electrons get detached from the atom leaving
the ion cores independent of the sea of electrons formed from the detached valence
electrons. This is in general true only in the case of simplest metals such as Lithium,
Sodium, Potassium, Cesium and Rubidium (Alkali Metals) and in Be, Mg, Ca, Al, Ga,
Pb etc. This is not valid in case of metals where d or fshell electrons overlap. This is
due to the fact that d shell electrons are more localized and less mobile than electrons
ins and p shells. Similarly, the properties of noble metals (Copper, Silver and Gold),
transition metals and the lanthanides are more complicated and can't be easily explained
on the basis of free electron model. As mentioned earlier, conduction electrons in a
simple metal arise from the valence electrons of the constituent atoms. For example,
in the case of sodium atom, the valence electron is in a 3s state; when this forms into
a solid, i.e., a metal, this electron becomes a conduction electron, moving almost
freely
throughout the crystal. If the crystal has N atoms, then N electrons move among the
N ion cores of Na*. Each of these ions would have 10 electrons distributed in 1s, 2s
and 2p states as in the case of free ion. The ions cores occupy only 15% of the total
volume of the crystal, leaving very large space for the free movement of the electrons.
This conclusion is drawn based on the fact that the radius of the Nat ion
is 0.98 A°,
where as the distance between any two nearest ions in the crystal is 3.66
A°. This
clearly indicates that the space left out unoccupied is very large as mentioned
earlier.
150
Electron Theory of Metals

As mentioned earlier, the conduction (Valence) electron in the metal finds itself
in a uniform field of positive ion cores and that of rest of the electrons in a given
volume. As the repulsive force between the electrons is neglected, the theory assumes
that the electrons sees uniform potential even inside the crystal, i.e. the conduction
electrons move freely every where within the specimen without any change in their
energy. As an approximation, the variation in potential energy due to the presence of
cores periodically is neglected and assumed to be constant in the crystal. Therefore,
any variation in the electron energy is assumed to be solely due to the kinetic energy.
However, the conduction electrons are not free to leave the surface of the metal at
room temperature but are confined to the volume of the metal. This is a common
observation made that no electron emission occurs from metals at low temperatures.
Therefore, we may conclude that the potential energy of an electron at rest inside the
metal should be lower than that of an
electron at rest outside the metal. In
such a case, we may assume
arbitrarily, that the potential outside the Surface
metals is very large or infinite. It is
this high potential, which restricts the
electrons from leaving the surface at
lower temperatures. This can be
viewed as though, there is a sharp
potential barrier of height $, at the Energy
surface as shown in Fig. 4.1. This
potential at the surface seen by an
electron is generally known as “work
Distance ————>
function”.

We, thus, can think of a simple

Fig. 4.1 Potential barrier
physical model for metals in which,
the interior of the metal is considered to be a potential well of depth @,, in which all the
electrons are confined to move within that well, without any change in their energy.
This model, even though appears to be very simple but far from truth, surprisingly, has
given considerably satisfactory results with regard to electrical and thermal conductivity
of alkali metals and their ratio, the “Lorenz number” (Weidemann - Franz - Lorenz
law).
Electrical conduction in metals

After the discovery of electron by J.J. Thomson in 1897, Drude constructed his
theory of electrical and thermal conduction by applying the highly successful kinetic
theory of gases to a metal, considered as a gas of electrons as described earlier in the
beginning of this chapter. In its simplest form kinetic theory treats the molecules of a
gas as identical solid spheres, which move in straight lines until they collide with one
another. Except for the forces acting momentarily at the colision, no other forces are
assumed to act between the particles.

151
 Applied Physics

Although there are one kind particles (molecules) in gasses (Kinetic Theory), in
solids, in addition to electrons, there exist other particles (positively charged ions)
which are heavier and hence considered to be immobile. These ion cores consists of
positive charge and bound negative charges other than valence electrons which wonder
in the solid. These valence electrons are called conduction electrons in metals. Drude
applied kinetic theory to this gas of conduction electrons. The basic assumptions are:
(i) Between collisions the interaction of a given electron, both with the others and
with the ions, is neglected. Then in the absence of externally applied electric field each
electron is taken to move uniformly in a straight line. Under an external field the
electron moves following the Newton's Laws of Motion.
(ii) | Collisions in the Drude model
are instantaneous events that abruptly
alter the velocity ofan electron. Drude
attributed these to the collisions
between electrons and ion cores and
not to electron - electron scattering.
A simple model of this is shown in
Fig. 4.2, where an electron bumps at
different ions which are far-off. A
qualitative explanation of conduction
mechanism through this type
scattering was achieved. Fig. 4.2 Scattering ofelectron at ion cores.

(iii) It is also assumed that the electron experiences a collision (i.e., it suffers an
abrupt change in its velocity) with a probability per unit time, 1/t. The time t is known
as the Relaxation Time, the collision time or the mean free time, and it plays a
fundamental role in the theory of metallic conduction. It follows from this assumptio
n
that an electron (chosen randomly) on the average travels for a time t before its
next
collision, and will, on the average, has been travelling for a time t since its last collision.
In this model, t is taken to be independent of position and velocity of an
electron.
(iv) It is also assumed that the electrons achieve thermal equilibrium
with their
surrounding only through collisions with the ion cores. These collision
s are assumed
to maintain local thermo dynamic equilibrium by a simple way. Immedia
tely after each
collision an electron is taken to emerge with a velocity that is not related
to its velocity
just before the collision, but randomly directed and witha speed
appropriate to the
temperature prevailing at the place where the collision occured. Speed
of the emerging
electron would be more if the spot of the collision is at a higher
temperature.

Let us first consider the effect of an electric field E ona free electro
n gas as per
Drude's or classical model. We suppose that there are n electro
ns per unit volume. We
imagine that the electrons are moving in a random way
as described above with a

152
Electron Theory of Metals

velocity distribution (Maxwell - Boltzman speed distribution) appropriate to a condition

of thermal equilibrium at temperature T, undergoing collisions. In the absence of an
applied electric field, the average or drift velocity,

12=
=v,
52s mane (4.1)

will be zero, as the number of electrons across an area moving on either side (left to
right and right to left) in opposite directions are equal (Fig. 4.3).

eas dt —-—_—>

Fig. 4.3 Number of electrons moving to right and to the left are equal across a given area (A) at any
given instant of time, in the absence of external force or field.

Expression for Electrical Conductivity of Metal :

The relaxation time t mentioned earlier is the mean time between the collisions

and is related to the mean free path (7) of the conduction electrons as

an ar are See (4.2)

Therefore, we can define mean free path as the average distance travelled by the
free electrons between any two consequtive collissions. As the electrons are assumed
to behave like gas molecules obeying kinetic theory and Maxwell- Boltzmann distribution,
the mean free path is given by

= oe
Jac jet aor MF es (4.3)
m

: aS
at
Ten,
This average speed is
|
where the average speed of the electron is given by

drift from different velocity, which is present only under the influence of external

153
 Applied Physics

force. If we write the equation of motion as

dVa il =
m dt +—Vq |-
F ha? Valine Die eto (4.4)
T

where F is the average external force acting on the electrons, then we see that in the
absence of external force, the equation of motion becomes,

dVq ]
Va =O) Set ). EL ee (4.5)
dt T

Solving the above leads to a solution

Vd) = Viti? 8 oo ee ay Were (4.6)

Where V4(0) is the initial drift velocity in the non-equilibrium distribution. We
observe from the above that the disturbance dies down (reduces) exponentially. The

|
relaxation time of t is the time taken for the drift velocity to decay to # ofits initial

V
value. In equation (4.4) we see the second term [2 : iP the left side is similar to
)
frictional term or damping force, where m/t is the frictional coefficient. If there is no
friction, the electrons in the system would accelerate continuously and hence
the
current increases to infinity, there by making the resistance of the conductor
zero.
This is in contradiction to the fact that the resistance of the conductor, at a
given
temperature, is constant and obeys Ohm's law.

With a finite relaxation time the equation of motion in a constant electric field
is

nfMa V ,}-e
Aiech (4.7)

when a steady state is reached

dV =
m—4 =+eE— ve = 0;
dt G

154
Electron Theory of Metals

because there would not exist any more, acceleration and the electron moves with a
constant average speed. Hence from the above, we get

mV =
d +e
T

et —
V,=—E
lS I a eR (4.8)

This velocity is the terminal (constant) velocity, as in the case of viscosity,

where a body dropped into a viscous medium, attains a constant velocity after some
time due to viscous force acting against it. V, is the drift velocity, which is zero
under zero field and directly proportional to the field. Due to such continuous net
motion of electrons in one-direction, a current sets in the conductor. If n free
electrons, per unit volume, all move with velocity V4, then the current density they
give rise to will be parallel to V,. Further, in a time dt, the electrons will advance by
a distance V, dt (as shown in Fig. 4.3) in the direction of V4, so that n(V,dt)A
electrons will cross an area A perpendicular to the direction of flow. Since each
electron carries a charge e, the charge crossing A in the time dt will be -ne V,A dt,
and hence the current density is

: I
j=-neV, = celeste te (4.9)

(-. I (current) = AneV, )

From equation (4.8),

oD.
= e| T
eae ees es (4.10)
m m

or j o E, which is Ohm's law.

Equation (4.10) can be written as

j=oE where Sie

and g is known as electrical conductivity, the inverse of it (+ = o|is called resistivity.

oO

15
 Applied Physics

From the expression for conductivity

m
istivi
resistivity P Oe nae Vilee ae Se eee (4.11)

we can estimate the relaxation times for differnt materials. Some typical examples
are given in table 4.1.

Table 4.1 Electrical resistivity (p) in micro ohm-centimeter and relaxation

time (t) in 10-!4s.

Element —(p) (t) (P) (t)

77K 273K ~—s273K

Relaxation times calculated are typically of the order of 10-!4 to 10-158. This in

combination with, Drude's average speed of electrons given by 3kn

, results in
m
mean free path of the order of 1 to 10A°, which is typically the distance
between the
ions. Hence the Drude's original view that collisions are due to electrons
bumping into
the large heavy ions holds good. However we will see later, that in
fact electrons do
not (get scattered) collide with every ion in the path and the mean
free path is much
longer at low temperatures than the result we obtained here. We also
observe that the
resistivity is a function of temperature and increases with increase
of temperature.
Where as the relaxation time decreases with increase of temperature.
This indicates
that the number of collisions per unit time increases with increase
of temperature.
Nothing in free electron theory can account for the temperature dependence
of the DC
conductivity that is observed experimentally. As per equation
(4.11)
156
Electron Theory of Metals

1 net
Qo=—=

ee al

h et
iS
re iO
— =
ae eee ae (4.12)

ut is called as mobility of the electron.

1
o=—=ney

Since, n the number of electrons per unit volume in metals is constant at all
temperatures, only the change in mobility with temperature accounts for the temperature

dependence of 'o' or _ Mobility | has units of m¢v!s"!. Now

Where, 7 —»> mean free path, ¢ —> average speed of electron in the absence of
external field.

ne / m
(ey —

m 3k,Tf

m /3k,T /3mk
ér p=—— JBm ST (4.15)
ne / ne

The above expression shows that the resistivity varies with temperatur as
T” at all temperatures. However, experimental data shows that this is not valid at all
temperatures and the variation again varies from sample to sample depending upon
their purity.The resistivity is independent of the field strength. Hence, Drude's free
electron model failed to explain completely the electrical resistivity behaviour of metals
(conductors). Ofcourse it doesn't tell us anything about the conductivity behaviour of
semiconductors and insulators, where free electrons are almost absent at low
temperatures. However one plus point of the free electron model is, it was able to
predict correctly the ratio between the electrical conductivity and thermal conductivity
(Weideman-Franz ratio) for many metals.

by
 Applied Physics

Wiedeman - Franz Law:

The ratio between thermal conductivity (W) to electrical conductivity (oc) is

W
—=LT
(ey

2 2
or at 2 (48) =2.45x108watt-ohm/K2 si... (4.16)
oT 3 e

Where L —» Wiedeman - Franz - Lorenz number, or generaly known as Lorenz

number. The Lorenz number was thought to be constant for all metals and at all
temperatures. However it was found that the ‘L’ varies with temperature. The
discrepency is again attributed to the assumption made in Drude’s model that all the
free (valence) electrons participate in conduction mechanism and participate in collisions
and hence relaxation time t is calculated taking into account all of them. However we
will see in the next section that the relaxation time is determined by few electrons
which are nearer to the Fermi surface.
However, this theory completely failed to explain the contribution from free
electrons to the heat capacity and the paramagnetic susceptibility. The failure of these
theories viz., Drude model and its improvement by Lorenz, were found to be due to the
assumption that the electrons follow Maxwell - Boltzmann distribution, where the spin
of the electron is neglected. With the advent of Quantum mechanics, Sommerfeld
could resolve some of the difficulties by applying Fermi-Dirac statistics to the distribution
of the electrons, along with Pauli’s exclusion principle, still retaining the concept
of
free electron model.

43 Sommerfeld's Quantum Theory :

Sommerfeld treated electron as quantum particle. They are governe
d by
Heisenberg's uncertainty principle and Pauli's exclusion principle.

The electrons in a metal customarily have both kinetic and potential

energy as a
function of position. For the present we shall consider a free electron
gas in which the
potential energy is arbitrarily set as zero. Then the total energy and
kinetic energy are
synonymous and an electron of velocity v and momentum p =
mv has an energy

Ba
Peal
5 DVee
sofeh.
a rali s

From the deBroglie equivalence of particle aspects and wave

aspects, i.e. the
particle behaves both as a particle and as a wave, this electron
has a wave length and
27 Pp
hence a wave-vector k = ae = (2 h
, where p is the momentum vector and f = oa
T
>

158
Electron Theory of Metals

h being Planck's constant. This electron can be represented by a vector from the origin
to a specific location in y - space, p - space or k -space or by just a point at that
location. One such location in k-space is shown in Fig. 4.4.
We should note that all the points in k-space at the same distance from the origin
will have the same energy, since kinetic energy and hence total energy are given by

The components of
Fig. 4.4 The description ofthe behaviour of an electron by a point in k - space.
.
k describe the periodicity of the electron wave function in various directions

(ii) there
As per classical model (i) any wave vector would be permissible, and
tor. The
would be no restriction on the number of electrons with the same wave-vec
nty principle
first of these classical assumptions was rejected by Heisenberg’s uncertai
of these two, leads
and the second by the Pauli’s exclusion principle. The combination
to the Fermi-Dirac statistics.
e, we require
In formulating rules subject to the Heisenberg uncertainty principl
wave equation subject
that the wave-function for electrons must satisfy the Schrodinger
n gas to be some
to some boundary conditions. Sommerfeld considered the electro
the electron confinement to
percent in a potential box. Let us discuss aspects regarding
the potential box.

159
 Applied Physics

4.4 Free-Electron Gas in One-Dimensional Box :

ioe) co
Let us consider an electron of mass
m confined to a one-dimensional box, of
infinite height, say along x-axis of length “L”
(Fig. 4.5). Let us assume that the potential y
energy everywhere within this box is
constant and is zero. Then the potential
energy for such a box can be written as
V(x) = 0 for0 <x <L x=O x=L x
= © forx<0;x>L _.........(4.18) Fig. 4.5 Variation of potential with distance (x)

i.e., potential energy is zero for all values of x greater than zero and less than ‘L’ and
infinity for x less than or equal to zero and greater than or equal to ‘L’.
The wave-function y,,(x) of the electron in the region 0 < x < L, where V = 0,
is described by the Schrodinger wave equation.

where E, is the total energy of the electron in the state n, and is equal to, in this case,
the kinetic energy only (since potential energy is zero). Then, the general solution for
the above equation can be written as (Chapter 3, Section 3.10). |

Walx) =A sin x] ae (4.20)

which is an equation for standing waves.
Therefore, allowed energy eigen values are,
then, -defined"by™ aree oehee aera aae,
n> =
E, =} ©
UL me
ee ee we es

2 A e
h n 3°
Ke, ee “\ errs (4.21) < ——-------

n
(since kn aL

Thus, the wave function y,, and the

corresponding energy value E, will exist ik] —>
only for integral values of ‘n’. The number Maganitude of wave vector
‘n’ is called the quantum number. Thus, Fig. 4.6 Variation of electron energy with
the concept of continuous energy values quantum number ‘n’ and wave vectork

160
Electron Theory of Metals

allowed in classical theory are prohibited by the boundary conditions and only discrete
energy values are allowed. First few quantum numbers (n) and the corresponding
allowed energy values are shown in Fig. 4.6. The energy E consists of discrete energy
levels with their separation depending on (n/L)* or k? . If ‘L’ is large, the energy levels
are closely spaced. For example
L = 1 metre,

h=66310- JS,

m= 9.11 x 10°!
kg.
Ee on ls ey
for n=3.
However, if the dimensions of ‘L’ are of atomic dimensions (Angstroms i.e.,
10-!m), then the separation between the levels become appreciable.
The normalized wave functions are given as

W (x) = [2sin sk sep (4.22)

’ L I

=|&
See

\
fea}
oa i
Ln n
E
i
&
5
o
(=)
aa]

E 1

0
0 x —> IG.

Fig. 4.7 Energy levels (E) and wave functions y,, for different (n)

161
 Applied Physics

Wave functions for some lower energy state are shown in Fig. 4.7 along with
the corresponding energy levels. If so many energy levels are available with
corresponding wave functions, the electrons have a choice and therefore they distribute
themselves and occupy these levels. Of course this distribution is on the basis of
Fermi-Dirac statistics which is discussed in the following sections.

4.5 Free Electron gas in a three-dimensional box:

Unlike in the previous section, where an electron was assumed to move only in
one dimension (direction), let us now try to be closer to the reality, that, the electron
really moves in three dimensional space and hence let us assume that it is restricted to
remain inside a cube of edge "L". As in the previous section, let us assume that the
potential every where inside the cube is equal to zero and its wave-function y,(r) is
described by the Schrodinger wave equation of the form,

VK) 2 eo 250 Lenina (4.23)

where V? = RIE
seo cs - ie ar , the state
and E, is the total energy of the electron in
OX= BOY = OZ
k. In the present case E, is just kinetic energy. The solution of the
above equation
leads to a standing wave solution, whose normalized wave-function
is
l
8
x)=|—|" 2.sin (nymx).sin fn ay).sin (nynz
ved sn) sn( ae
Where n,, Ny, N; are positive integers. However, as this solutio
n leads to a result
of preferential place for an electron in the box, i.e., where the
probablity of finding an
electron is maximum at antinodes and zero at nodes, which
is against the principle of
quantum mechanics, a plane wave solution is assumed. It can
be written as
Wi (r) = AelKr = Agilkixtkaytky) (4.24)

2m
Where k? = (22 E, and A is an arbitrary constant.
Recalling from the previous

section that a wave function of the above type must

satisfy boundary conditions as
earlier, we write then,

WK RLV AU (X59, 7) ee (4.25(a))

WG Y PLZ r= (ke yoy ee (4.25(b))
LG SAR ASi ean TaChee) (4.25(c))

162
Electron Theory of Metals

Any wave-function which satisfies the above condition would vary periodically
with x, y, z with a period 'L'. The condition (equation 4.25(a)) when applied to
(equation (4.24)) results in

eilk, (X+L}k,ytk3z] “=eilkixtk,y+k,z] See (4.26)

or ekil 4]

or ig Sa Weer bid nena eePopa! Seer (4.27)

where n is any integer. Similarly applying (4.25(b)) and (4.25(c)), we get similar
results. Therefore it is implied that k, or k, or k, are controlled by the number 'n’,
which can be a negative or positive integer or k doesn't vary continuously in a box
can't have continuous varation of energies but has a discrete energy spectrum. In
addition to values of k,, k,, k,, the quantum number m, due to spin is also required to
completely specify the state of an electron. The resulting energy spectrum would be
a discrete one. However the seperation between the energy levels would depend upon
the dimensions of the box. If it is of macroscopic dimensions i.e., in centimetres, the
gap between the energy levels would be very small and could be assumed to be
continuous, even though they are discrete. From the equation (4.26)
E, = (i2/2mk? = (n2/2m)|k? +k24h3] nn (4.28)
one can see that as 'L' becomes smaller and of atomic dimensions, the energy levels
become well seperated. To evaluate the constant 'A' in equation (4.13), we use the
normalisation condition
Vv

Mii iat (4.29)

(Where V is the volume)

2
from which we get A= (=| and hence using eq. (4.24) or (4.29), the wave function

can be written as
1

wcn=(z}e hire (4.30)

Therefore, we understand that an electron placed in a box of finite dimensions
can have discrete energies only.
Let us examine the way, the conduction electrons are distributed among the
different energy levels. The quantum numbers n,, n, and n,, corresponding to kj, k,
and k, and spin quantum number m,, (which has only two states, up or down) restrict

163
 Applied Physics

the number of electrons that can posses the same energy, due to Pauli's exclusion
principle, since the sum of n,, n, and n, lead to quantum number 'n', the number of
electrons which can have the energy corresponding to this 'n' would be only two,
corresponding to the up and down states of spin quantum number. Hence every
energy level corresponding to every 'n' can be occupied by only two electrons.

4.6. Energy Levels andFermi-Dirac Distribution :

We shall discuss how electrons are filled in energy levels. In a three dimensional
system like a conductor, the electrons which are indigoes, can in fact possess any
energy of their choice, except for the restriction put by Pauli’s principle. Hence, if
one assumes that electrons are dropped into a box of discrete energy levels as discussed
earlier, then the very first electron would try to occupy the lowest available energy say
E, and then the next one also would occupy the same energy, as their tendency
always is to occupy the lowest possible energy level. However, when next electron
(third) comes to occupy the same the level ‘E,’, it is prevented by Pauli’s principle
and hence forced to move to the next higher level say E,(E, > E,). Here again two
electrons can occupy this level. But fifth electron has to move to next higher state
E, and so on. Therefore, it is clear that at 0 K (ground state) the electrons would
occupy every energy level in pairs and hence ‘N’ electrons (even) would require
(N/2) levels. The highest energy thus occupied by the last electron in the ground
state, which is a sort of line between the occupied levels and unoccupied levels is
known as Fermi energy level (E;,,). That is “E,” is the highest energy level occupied
by electrons in any given system at 0 K and is the fermi energy for that system as
shown in Fig. 4.8. Here, we note that quantum theory differs from the classical
theory in predicting the
highest energy occupied at 0
K. As per classical theory all
the electrons occupy the
same energy which may be
zero or some fixed value like
E,. But in quantum theory,
different electrons occupy
different energy levels even
Energy
at 0 K, from E, to Es . Fig.
4.8 also shows that i 0 K,
all the energy levels less than
Ep are occupied and all those
Fig. 4.8 Occupation ofenergy levels at 0 K as per
more than are completely
Sommerfeld theory.
empty.

Let us try to understand the effect of temperature on these electrons which

occupy energy levels less than Ep: As the temperature increases, some of the electrons

164
Electron Theory of Metals

gain energy and occupy the higher energy states, leaving the lower energy states
empty. It would be difficult to predict exactly which levels are occupied and which
are empty. Since the number of transitions from lower to higher, are large in number
and random in nature. Hence, to know whether a particular energy state is occupied
or unoccupied, Enrico Fermi suggested a statistical method known as Fermi-Dirac
statistics, which gives information about the probability of occupation of a given
quantum state at energy E by the following expression.
]
Oy
(©) Exp|(E - Er )/kpT]+1
Where E; is the Fermi energy and k, — Boltzmann constant. The behaviour of
this function with the variation of temperature gives us information about the occupation
of energy levels as a function of temperature. The factor k,T is the thermal energy at
temperature T. aoe
Fig. 4.9 shows the 2x fs Ji710K
variation of f(E) with energy T = 0K
‘E’ at different temperatures.
At T= 0 K, we observe the
probability of occupancy of
energy levels below E,, is
maximum that is, unity.
Therefore it is evident that all
the levels below Fermi energy
are fully occupied by electrons
at 0 K. Similarly for all the
levels above E,,, f(E) is zero,
i.e. they are all empty. This
result is obtained from equation Fig. 4.9 Probability of occupation of different energy states
(4.31) by substituting T = 0 with the increase of temperature.
and for E > E,, we get f(E) =
0; for E < E,, we get f(E) = 1. Here ‘one’ in the denominator of equation
(4.31) is neglected compared to the exponential term. E,, can be calculated in terms
of the number of free electrons per unit volume.

h? (3n)3
ee ee imreis Vs yisotne sft tot) wesls ataly (4.32)
RS (=|

The above expression clearly indicates that E,, is essentially dependent on

number of free electrons per unit volume.
For k,T << Es for all temperatures below the melting temperature of metals k,T
is small compared with E, (k,T at room temperature is only 0.025eV). From equation
165
 Applied Physics

(4.31) it follows that for E = E, f(E) = 1/2. This gives another definition or physical,
meaning for E,: at the Fermi level, the probability of occupation is half (1/2). For
energies below E, such that Ep— E >> kgT, the curve f(E) is still practically unity i.e.,
the energy distribution in that region is almost same as that for T = 0. Only in the
vicinity of E, minus a few k,T that f(E) begins to drop below the value at
T=) rep ‘ones
For energies above E,, such that E—E, >> k,T, one may neglect the term unity
in the denominator of (4.31) and one obtains
f(E) ~ exp [{E - E)/kgT] Si dd SS a tineae (4.33)
In this region Fermi distribution becomes closer to that predicted by Boltzmann
distribution.

4.7 Density of States:

Density of states D(E) is defined as the number of energy states or eigen values
available per unit volume which have energies lying between E and E + dE, where dE
is a small energy interval. Similarly N(E), the density of filled elecron states is the
number of electrons per unit energy range about an energy E i.e., between E and E +
dE. It is given by the product of density of states D(E) and Fermi distribution function
CE):

N(E) = D(E) f(E)

48 Electrical Conduction:

As per classical Drude’s Lorenz model, electrons follow Maxwell Boltzmann

distribution of velocities and energies. According to this distribution, many electrons
can simultaneously possess same energy (or velocity) as there are no restrictions on
the numbers. However, we know that electrons possess spin, hence due to Pauli’s
exclusion principle, they cannot behave like gas molecules and hence follow
Fermi-Dirac distribution. According to this distribution, as discussed earlier, not more
than two electrons can occupy the same energy state and hence even at 0K, they
occupy different energy levels which are not continuous. In this arrangement, some
electrons would possess highest among them, which we termed it as Fermi energy.
The velocity of electrons which are moving within the metals also would have similar
distribution, with few of them having higher velocity called v p (Fermi velocity).
Therefore it is clear that the velocity of electrons among the conduction electrons
differs considerably. From these considerations we would expect drastic changes, as
only those electrons near the Fermi surface can participate in collision processes. Let us
try to derive the expression for conductivity with quantum assumptions and Fermi-
Dirac distribution incorporated. A graphical representation of the above statistical
distribution of the energies of the conduction electrons will now be considered, as it
gives better physical picture of the conduction process.

166
Electron Theory of Metals

velocities in 3-d velocity space (a) without field (b) with field along x
Fig. 4.10 Distribution of

Each conduction electron occupies a particular energy state which has an

associated velocity which can be represented as a point in the three-dimensional velocity
space shown in Fig.4.10(a). In the absence of applied electric field, the electron
velocities are random and for every electron travelling in a given direction, there
would be another electron travelling with same speed but in opposite direction. Because
of this, the velocity surface will have a spherical symmetry at any given temperature.
At zero Kelvin, all the levels from the lowest available would be occupied compactly,
upto the highest level at v , (the Fermi velocity), corresponding to Fermi energy (E,).
The outermost boundary containing all possible velocities will be a sphere, with radius

U p, shown dashed in Fig. 4.10(a), since Er = =mop Such a boundary is called a

‘Fermi surface’. Fermi surface need not always be spherical.

At higher temperatures, the boundary of the distribution in velocity space

greater than
becomes more diffuse, since small portion of electrons have velocities
k,T,
Uf. The distortion thus produced is not evident, as it is of the order of few
which is much less than E,; generally.

Now, if we apply an electric field E,, along x-direction, as shown in Fig. 4.10(b),
direction, i.e.,
all electrons in the distribution are subjected to a force eE, in the — x
the imaginary Fermi
opposite to the direction of the field. Only those electrons near
levels immediately
brim (surface) can move, since, only they have unoccupied energy
adjacent to, move into. For rest of the electrons, which are at lower energies, the

167
 Applied Physics

energy supplied by electrical force is too small, to jump out of the Fermi surface. For
these electrons, which are lying below, all the upper levels are occupied and hence
they can’t move into them. Since the electric field is acting on all of them, all the
electrons are subjected to the force and hence equilibrium would be disturbed for a
while, till the velocities of these electrons get adjusted to the new situation. During this
process some electrons return to their equilibrium position in velocity space after a
collision, while others are being accelerated by the field. Hence, on average, the
application of an electric field causes the entire equilibrium velocity distribution to be

shifted slightly by an amount ea in the opposite direction to the field as shown in

m
Fig.4.10(b). We can see from the figure that the velocity distribution is no-longer
symmetrical with respect to ‘0’, the origin and the cancellation of electron velocity
components in the direction of the field does not occur. There is now aa net electron
velocity in a direction opposite to the field direction and is given by the expression.

= = CE ate
D PS DiS eg OVS (Gl Sle MGONET Sik.aes (4.34)
m

where v4 - the drift velocity of electron

<v > - average velocity along x-direction due to field in x-direction

te relaxation time of electron at the Fermi surface

m’ - effective mass of an electron.

Then v ,, the drift velocity, is the difference between the equilibrium and shifted
steady state distributions. This accounts for the resulting current flow on the application
ofa field. The drift velocity v , is of the order of 10m/s for a field of 10V/m. This
is much smaller than the velocity (vu ,) of electron (at the Fermi surface) moving
within the lattice randomly in the absence of the field. Here, we observe that v p is
dependent on temperature and v , on the electric field. Of course, as soon as the field
is removed, the steady distribution relaxes to the symmetrical equilibrium following
the equation (4.6). The relaxation time 1, is not the average for the whole distribution
of electrons but for those which stay near the Fermi surface with energy E, or velocity
Up. Hence the equation for conductivity obtained earlier eq. (4.10) is still valid,
provided the relaxation time is t,.. Similarly m* is the effective of mass of the electron,
about which we will learn more later in band theory.

168
Electron Theory of Metals

Now using these relations, we can write the expression for mean free path and
conductivity as

Since the Fermi energy E;, is relatively insensitive to the temperature,

UR & /2Ep, /m* desir (4.36)

Where Eg, is taken in Joules. If Ep, »7 ev (Fermienergy at 0K). Then

vp *10°m/s. From, condcutivity o = 6x107 S/m, n = 1029/m3 for Cu, we get,

tp = 10's. Hence, the mean free path from eq. (4.35) is of the order of 10°°m,
which is very much longer than the lattice constant, which is of the order of 107!m.
Therefore, collisions are not occuring between the electrons and ion cores ( distance

between the ions is 107!°m) since this would lead to /of the same order as 10°!°m.
Hence one has to investigate for the collision or scattering centres. A detailed study
indicates that scattering or collisions occur between electrons and phonons (A quantised
energy packet of elastic waves generated due to harmonic oscillations of ions. Number
of phonons and their energy depends on the temperature of the system. Ions in the
lattice vibrate to generate phonons with different energy and hence different frequency)
and electrons and defects.

Another discrepancy to be addressed is the variation of resistivity with

temperature. The observed variation is linear with temperature i.e., paT, where as

free electron model predicts p « T!/?. At lower temperature, it is observed that

s is
p a T° and at very low temperatures it becomes constant. None of these variation
show sudden
explained by free-electron model. Some elements such as Pb, Hg, Nb
of the conductor
decrease in resistivity at a particular temperature and the resistance
This phenomenon
becomes zero. This phenomena is known as super conductivity.
show anamoly
also could not be explained on the basis of free electron model. Metals
es.
in resistivity behaviour at low temperatures due to magnetic impuriti

169
 Applied Physics

Therefore free electron model failed to explain all the above phenomena, in
addition to its failure to explain (i) magneto resistance (change of resistance in the
presence of magneticfield), (ii) positive hall coefficient observed in some metals,
(iii) resistivity behaviour of semi conductors and insulators (iv) anamolies in electron
specific heat (iv) paramagnetic susceptibility.

49 ‘Temperature Dependance of Electrical Resistivity :

Electrical resistivity of metals has been observed to increase with increase of
temperature. As mentioned earlier, this is due to the decrease of relaxation time Tes
resulting in the decrease of mean free path of the electron with increase of temperature.
Thus more number of collisions occur at higher temperature. However, the calculation
of Jp has shown that, it is much more than the inter ionic distance. From the theory
of lattice vibrations, and the band theory it can be noted that electrons moving in a
perfectly periodic potential remains unchanged in the absence of external field. Thus,
as a result of the wave nature of the electrons, they can pass through a perfect crystal
without suffering any resistance. This indicates the passing of electron wave without
any interference taking place with lattice waves (phonons). The consequence of this
is that if all the nuclei were at rest (without vibration), the mean free path for electron
scattering would be infinite, leading to infinite conductivity. This is what is expected
at very low temperatures or at OK. But experiment shows that the behaviour of
electrical conductivity or resistivity is different from what is predicted above. The
actual cause of resistivity must therefore be sought in deviations from periodicity of
the potential in which electrons move. The deviations from the periodicity of the
lattice which causes the resistivity in metals could be due to the presence of
(i) Lattice vibrations (non stationary periodic lattice)

(ii) Lattice defects (impurities, vacancies, interstitials etc.,)

(iii) Boundaries.

4.10 Impurity Scattering Phenomena - Electrical Resistivity :

In a pure metal, the electrical conductivity should go to zero as the temperature

approaches absolute zero. However, experiments reveal that the resistivity of
non-super-conducting metals decreases to a small but finite value and remains
practically constant and independent of temperature below 10K down to nearly absolute
zero as shown in Fig. 4.11. This is known as the residual resistance of the metal,
which is attributed to the presence of impurity atoms in small concentrations and also
geometrical imperfections like point defects, grain boundaries, stracking faults etc., in
the metal.

170
Electron Theory of Metals

Po» Po» Po — residual resistivity

(resistivity)
p

Po
Po
Po
0
0 = 20
Temperature

Fig. 4.11 Variation of resistivity of a metal with temperature with different amounts impurities.

The residual resistivity increases with the increase of static defects per unit
volume. Different relaxation times t,, T,,... can be assigned to each type of scattering
solid
taking place at these static defects (impurity, vacancies, dislocations etc.,) in the
the p — 0 as
(metal). If the metal doesn’t contain any defect in the lattice, then
to have
T->0. But this situation can never arise, as according to thermodynamics,
minimum free energy, the system has to have few defects. It will be humanly impossible
residual
to make 100% pure metal by any process. Hence there exists always some
resistivity and corresponding to it the scattering relaxation time (t;) as

me
electron at the
Where 1; is the relaxation time corresponding to the scattering of
simultan eous present ina
impurity centre. If there are many impurity mechanism
ent of each other,
given metal resulting in a residual resistivity, which are independ
al probabilites.
then the total scattering probability becomes the sum of all the imdivudu
relaxation time (if the
Since the scattering probability is inversely proportional to the
s), therefore the
relaxation is large, the probability of occurance scatting decrease
s is given by :
relaxation time due to imperfection and impuritie

to t, given by the
Hence the resistivity at very low temperatures (p,) is related
above equation (4.38).

171
 Applied Physics

We have stated earlier that p; increases with impurity or imperfection

concentration. Let us see how this fact is obtained.
According to kinetic theory model for free electrons,

where 7, is the mean free path electrons in the temperature where scatteing due to
imperfections dominate over other processes of scattering such as electron-phonon
scattering. If A; the scattering cross section of impurity atoms, then we may write
from kinetic theory.

DAG

where n,, is the number of imperfections (impurities) per unit volume and A, the cross
section of the defect. Here cross section is a quantity which gives the probability of
occurrence of scattering and in this case proportional to the actual area of cross of the
defect i.e, area covered by the defect. Area of an impurty atoms or defect can be
assumed to be A, = WAC Substituting equation (4.38) and (4.39) in (4.37) we find
that

m *

P; 5 Dj Aj VR (n = electron concentration)
ne

or nes a ee ee ery A en (4.41)

Hence, we observe that the resistivity of a metal increases with impurity
concentration. This is seen clearly when small amounts of impurities are added to a
metal to make an alloy. It is observed the resistivity of sucyh an alloy is larger than that
of pure (without impurities being added) metal.

4.11 Scattering Due to Phonons - Electrical Resistivity :

In the above section we have seen that at very low temperatures i.e., below the
Debye temperature (0,), scattering of electrons is basically due to imperfections,
where vibrations of ions (atoms) due to temperature (i.e., thermal vibrations of atoms
or phonons) reduce considerably. But if we cross this regime, and increase
the

172
Electron Theory of Metals

temperature, we observe that the resistivity increases as T> initially and after reaching
a certain temperature the resistivity starts increasing linearly that is p a T. Letus look
for the causes for this behaviour of resistivity in metals. As already mentioned earlier,
as the temperature increases the ions which are arranged in a periodic fashion in a
crystal start vibrating like a simple harmonic oscillators. Since, the ions are not
independent of each other and are bound to each other through resulting in electric and
elastic forces and hence, they vibrate with different frequencies. These vibrations are
like stationary waves with fixed energies and are considered to be packets of energy
called phonons. Like electromagnetic spectrum, they also have a phonon spectrum,
which is very much dependent on temperature and internal arrangement of atoms,
defects etc. As vibrations are created due to thermal energy, number of phonons(with
different energies) depends very much on temperature. At lower temperatures their
number reduces cosiderably.
Therefore, as the temperature of the metal increases, more phonons come into
play and the possibility or probability of their colliding with electrons increases. Hence,
we can calculate the variation of resistivity with increase in temperature based on
Debye theory of specific heats. According to this theory each ion is considered to be
an independent oscillator but there exists a highest frequency with which they can
oscillate, which is given by vp, (Debye frequency). Corresponding to this frequency
every solid has a temperature called Debye temperature (0,). According to Debye
T?;
theory, at low temperatures the specific heat of metallic lattice is proportional to
probabilit y for scattering are proportio nal
therefore the density of phonons and the
to T.
In addition to this, the relaxation time which is dependent on the scattering
re
angle, would give rise to resistivity variation with temperature as T’, in this temperatu
region. Consequently, p a T’.T* =T°. This includes argument of relaxation time and
essential
hence is not a very satisfactory explanation. However, it points to the two
the scatterih g
causes for T> law; the decrease in density of phonons and reduction of
angle with decreasing temperature.
increase
2 At much higher temperatures the electrical resistivity is observed to
s putforth
linearly with temperature as mentioned earlier. Here, making use of argument
on for mean free path of electron
earlier for impurity scatting we can write the expressi
due to phonon scattering (9) as

ee ae (4.42)

173
 Applied Physics

where n,,, is the concentration of ions in the lattice, A,., is the cross of the ion. Here
A is not the area of ion, but the cross section of the ion fluctuating due to thermal
energy to the passing electron. If the ion gets displaced from its equilibrium by < x >,
the average displacement, the average cross section is

Adoasaicxeetib
ion — ditw onwihy earl ssonst! bun eae (4.43)

Since the ion is vibrating like a harmonic oscillator, the potential energy,

|
2 Ok
2 oF eee

According to kinetic theory

]
sme? <x? Ss18 Ss = Kei

where, < E > - average energy

a —» Force constant

® > 27Vp

From Debye theory, hvyp =kp@p 2 i ss—(i—‘“—s—s—sS (4.45)

where v, and 0, are Debye Frequency and Debye Temperature.

Therefore, from (4.45)

Oh = 4n*kR0
: 5?
h

Making use of this in (4.44), we get

k - 2
ae Fel h . balbpally ie ic Roeet (4.47)
4n“M )\ kp 9p Mkp 65

174
Electron Theory of Metals

From equation (4.42) and (4.47), we get

Con =i ee

Thus we see that at higher temperature (T >> 0, ) resistivity of a metal increases

linearly with increase of temperature, as observed experimentally.
Hence the total resistivity of metal at any given temperature is a sum of resistivity
due to phonon scattering (Pp, )and impurity scattering (p; ). Thus we can write

Apeptpal(t) —- = | | ae (4.50)
This equation is known as Matheissen rule. From equation (4.50) we can see
by
that the resistivity varies with temperature, and this variation is mostly controlled
that
the phonon scattering part (Dot (T)). At higher temperatures p,,, is so large

p(T) ® Poh (T). However at lower temperatures, influence of Pon decreases and below
nt
10K becomes almost zero. Hence, at those temperatures p(T) = p;, which is independe
of temperature and hence we observe in experiment a constant resistivity, controlled
have
by the quantity of impurity content. Similar to resistivity equation (4.50) we also
a relation between the relaxation times.
1 1 1
Se at ee, TD) RRR (4.51)
O Ce

Thus, from (4.50) and (4.51) we get

es ini m * ]
) Pj + P ph(T)= + ne*t =
p(T)=p;, ) ~s te e

175
 Applied Physics

Deviations from Matheissen rule are observed in many metals in which magnetic
ion (Fe) are doped, which is known as Kondo effect. We observe a small minima at
lower temperatures. Similar type of deviations from Matheissen rule are also observed
in meterials where the band structure is complicated.

412 Hall Effect:

It is a well-known fact, that, when a magnetic field is applied perpendicular to a

conductor (metal) carrying current, a potential difference (voltage) is developed across
the specimen along a direction perpendicular to both the current and magnetic field.
This phenomenon is known as “Hall effect” after its discoverer Hall. The voltage thus
developed is called Hall voltage.
B. :

(c) Steady state condition after the setting up of Hall Voltage (field)

Fig. 4.12 Hall effect

176
Electron Theory of Metals

This phenomenon can be explained as follows:

Consider a slab of material (metal) subjected to an external electric field € long

x-axis and a magnetic field induction B, along z-axis as shown in Fig. 4.12(a). As a
result of applied electric field, a current density J, will flow in the direction € , and
electrons would move opposite to this direction. If we assume that the current is
carried by an electron of change (-e), the electric field & , causes a Lorentz force of
(—E,e) to act on the electron. At the same time, the magnetic induction B,, which is
perpendicular to this motion applies a force perpendicular to both these fields. Therefore,
the electrons which are traveling with velocity v, would be deflected down along
y-axis. When these electrons reach the surface of the conductor, an electrical charge
is built up there, which, in turn produces an additional electric field along ‘y’. Under
equilibrium conditions, this force on the moving carriers just balances that arising from
the magnetic field, and electrons can again move freely along x-direction. All this
happens ina very short interval of time and hence almost impossible to observe. Instead
of electrons, if positive charge carriers are carrying the current, then the electric field
set up along y would be reversed. From this, it is evident that measurement of the ‘Hall
Voltage’ in the y-direction gives information about the sign of the charge carriers.
Further, the information about density of charge carriers can also be obtained.
Let us write down the physical phenomenon described above in mathematical
from as

F = -e[E, sv: 33)| DiI a SY eo Cmerl wee: (4.53)

Since the net force acting on the electron along ‘y’ is zero i.e. Ey = 0, we have

Bsa ey = ¥eBal <0

or Spa Bee > «) ve oma (4.54)

Where v, is the average drift velocity of the electron along x-direction. Then using
free electron theory, current density may be expressed in terms of density of electrons
n (number of electrons per unit volume) as
eter |p: la Zn (4.55)

Substituting (4.55) in (4.54) for v,, we get

lS
maf E (4.56)

a)
 Applied Physics

where RPS soe etect. We oletrve € Srey sae (4.57)

R,, is called the Hall coefficient and is determined by the sign and density of
charge carriers. Observed Hall coefficient for a number of metals are given in
Table 4.2. It is observed that a number of metals have positive Hall coefficient.
Qualitatively, this can be explained on the basis of band theory of metals, since a metal
with a nearly filled band is equivalent to a conductor in which current is carried by
positive holes; this would change the sign of R,,. This concept of holes cannot be
explained on the basis free electron model.
It may be remarked, that the occurrence of positive Hall coefficients and existence
of positive charge carriers (Holes) presented a famous unsolved problem until clarified
by band theory, about which we will study in the next sections.
Further, by combining this result with that of the conductivity measurement,
the mobility may also be obtained:
CONGUCHIVItY C= CU ue ge ee (4.58)
where ‘n’ concentration of the charge carriers, ‘uw’ mobility of the charge carriers and
hence
OR,, = — (nep)(1/ne) = —p ie., p = SGRuy ali-4)2 (4.59)
As we know, all the moving electrons in a conductor do not possess the same
velocity and they follow maxwellian distribution, equation for R,, is modified as

H
8 nle|
From the above, we understand that measurement of R,, is very simple, but at
the same time it gives important information about the type of carrier, positive or
negative, and the concentration density of these charge carriers.
Table 4.2 Hall coefficient of some metals at room temperature in Ry x 10!
(V-Cm/amp-oe)

Observed Calculated from free

electron theory
— 17.0 — 13.1
— 42.0 — 47.0
—7.2 — 10.5
+ 24.4 — 2.5
A383 ~46
+ 6.0 — 6.5

178
Electron Theory of Metals
er ee "et

BAND THEORY OF SOLIDS

413 Introduction
The free electron model of metals, developed in the previous chapters gives us
a good deal of insight into several of the physical properties of metals, yet there are
other properties for which the free electron model is quite uninstructive. This model
leads us to wrong results even in the case of conductivity of metals. As per this model
conductivity 'o' is proportional to electron concentration. If it is so, divalent (Cd, Zn,
etc.) and trivalent (Al, In, etc.) metals should have higher conductivity than monovalent
metals (Cu, Ag, etc.). This is contrary to the experimental observation. Further, the
model cannot help us to understand why some chemical elements crystallize to form
good conductors of electricity and others to form insulators; still others are
semiconductors, with electrical properties varying markedly with temperature. Yet,
the distinction between the resistivity values of conductors, superconductivity apart,
and insulators is striking: the resistivity of a pure metal at low temperature may be of
the order 107! Ohm-—m, and the resistivity of a good insulator may be as high as
107° ohm-m. It may be noted that this observed range of 102° in resistivity, is the
widest range of any common physical property of matter.
We can gain some insight into the nature of the difference between insulators
and conductors only by making such assumptions, which are closer to the reality.
This can be done to certain extent by extending the free electron model to take account
of the interaction of electrons with the ions situated in the periodic lattice of the solid.
We shall encounter some quite remarkable properties possessed by electrons in crystals.
We shall see that they may respond to applied electric field or magnetic field as if the
electrons are endowed with an effective mass m*, which may be larger or smaller than
the free electron mass, or may even be negative. Further, we may encounter situations
where it is convenient to assign a positive charge +|e| to charge carriers in a crystal; we
denote such carriers as holes, in contrast to electrons which behave with their normal
negative charge —le].
As mentioned earlier, one can gain substantial insight into the structure imposed
on the electronic energy levels by a periodic potential, if the potential is very weak.
This approach might once have been regarded as an instructive, but academic exercise.
We now know, however, in many cases this apparently unrealistic assumption gives
results surprisingly close to the mark, particularly in case of metals from I, II, III and
IV group of elements, with atomic structure, consisting of s and p electrons outside of
a closed shell noble gas configuration. Therefore, these elements are referred to as
‘nearly free electron’ metals, because their behaviour can be explained by Sommerfeld
free electron gas model, modified by the presence of a weak periodic potential. In this
chapter, we shall examine some of the broad general features of band structure from
the almost free electron point of view.

ee)
 Applied Physics

It is by no means obvious why the conduction bands of these metals should be

so free-electron like. There are two fundamental reasons why the strong interactions
of the conduction electrons with each other and with the positive ions can have the net .
effect of a very weak potential.
(i) | The electron ion interaction is strongest at small separations, but the conduction
electrons are forbidden (by the Pauli principle) from entering the immediate
neighbourhood of the ions because this region is already occupied by the core
electrons.
(ii) | In the region in which conduction electrons are allowed, their mobility further
diminishes the net potential any single electron experiences, for they can screen
the fields of positively charged ions, diminishing the total effective potential.

4.14 Energy Band Formation :

Formation of Energy Bands :
In order to understand the concept of energy bands formation, Let us first recall
the formation of energy level in an isolated atom. When atoms are at large distance as
in the case of gas molecules, the mutual interaction between the electron orbits of the
neighbouring atoms can be neglected. The atoms have independent energy levels
permitted by quantum rules on the basis of Sommerfeld atomic model. Each energy
level accommodates as many electrons as allowed by Paulis exclusion principle. The
energy levels are degenerate energy levels. However, when the atoms are brought
closer to each other, the electrons respond to the influence of the other nuclei and the
electron orbits of the atoms over lap when the atoms are brought still closer as in the
case ofa solid. The Pauli exclusion principle will then apply as a whole to the system
of atoms that are brought together. Each energy: level associated with the individual
atom now repeats as many times as the number of atoms in the solid. What was an
isolated level in the single atom broadens out into a large number of closely spaced
levels of the atoms. Such set of energy levels that has a finite width is called an energy
band. The electron which originally have the same energies now get their energies
modified so that exclusion principle is not violated. The figure 4.13(a) depicts the
energy band formation in a solid or crystal. At large interatomic distances (represented
by r), different energy levels are shown as discrete energy levels. But, as the distance
between the atoms is reduced, it can be observed that each set of energy level combines
together to form a band of closely spaced, levels under the influence of neighboring
atoms. We can see from the Fig. 4.13(a), that the spreading of levels starts at different
values of r for different energy levels. The spreading of levels starts at larger distances
for outer shells as the electrons in the outer shells interact first with neighbours compared
to the inner shell electrons. At a particular distance 'r,', the inter atomic distance in a
solid, each energy band is separated by an finte energy gap E..

180
Electron Theory of Metals

In a case like Beryllium, in which hybridization of 2s and 2p levels occurs,

overlap of energy levels occurs such that both of them together form a single band as
shown in Fig. 13(b). Thus, the overlapping of energy levels can occur by transfer of
electrons from s-levels to p levels and from p-levels to s-levels, valence band in such
case, will be partially filled.
We can also take up another class of metals in which the valence band is partially
filled. In metal sodium 3s band is half.filled and 2p band is completely filled. This 3s
band is generally denoted as conduction band. In metals such as sodium, the conduction
band is partially filled. In semiconductors like germanium, or silicon, the conduction
band is totally empty. The partial filling of abond arises as a result of hybridization, a
characteristic so common in many metals.
We will summarize now. The band structure of an elemental solid depends upon the
interatomic distance, which in turn depends on chemical bonding. The formation of
‘bands was not a part of free election model in which the distinction between a metal
and an insulator or semiconductor, can not be made. To obtain the real description of
a solid, we have to consider a realistic model like 'band model’ based on band-theory.
The kronig-penney model gives a preliminary band model. The model assumes that the
potential experienced by an electron in a solid is not uniform but periodic in nature. We
will discuss how this simple model leads to the overwhelming successful band model
of a solid.

=
Energy

(a) (b)
Fig. 4.13 Formation of energy bonds in a solid*

181
 Applied Physics
a ——————————————__

415 The Kronig- Penny Model:

One of the most simplest examples of one-dimensional periodic potential was

first studied by Kronig and Penny. It is assumed here that the potential energy of an
electron, when it moves in one-dimensional perfect crystal lattice, is represented in the
form of rectangular wells as shown in Fig. 4.14. The period of the potential is (a + b);
in regions where 0 < x <a, the potential energy is assumed to be zero and in regions
such as —b < x < 0, the potential energy is V,.

Fig. 4.14 Kronig-Penney one-dimensional Periodic Potential Well.

Though this is a hypothetical model, far from truth, it illustrates many important
features of the quantum behaviour of electrons in a periodic lattice. Through this
model, the Schrodinger equation can be solved explicitly in terms of elementary functions.
The Schrodinger equation for the two regions can be written as

dy
ea as (2m/n?) By =i(0 for 0<x<a seevoecs (4.61)

ri d?y
Pe (2m/n?)(E-V,)y=0 for —b<x<0 wn. (4.62)

Let us assume that the energy E of the electrons under consideration is less than V.,.

2mE
Let us put a?= pe and B2=2m(V,—EVAy oaseaees (4.63)

182
Electron Theory of Metals

in the above equations, where « and B are two real quantities. On substitution of /
(4.63) in (4.61) and (4.62), we have

d
ar +o eDe mpoty Ook amb, 5 | ohende. (4.64)
d2

ae Epa Omiigreye Fein"! - 8LS weBe (4.65)

Now from the Bloch theorem, the solution of wave equation for a periodic
potential will be of the form of a plane wave modulated with the periodicity of the
lattice of the form.

Z Ue:k eee ee POY Dent... (4.66)

where U,(x) is the periodic function in x with period (a + b).
Since y = U(x),
dy
al aes 8
' =a:
we get mi ke™ ikx U(x) +e ikx
ee

d’°y
ae a
.
kelXU, (x) af ikeikx
Raed,
ae
se
ae ikelkx
Vacd
dU peak elkx ee

= TA le% oie
Ue
alkx
ea
k“e™U, (x) + 2ike re k +e ikx
a k

Dropping the subscript k in U, and the arguments, and substituting these in

(4.64) and (4.65), we get

d?U dU
ey ik 2 Ge kU 07 OX a Pera (4.67)

d*U dU
+ 2ik—
fe 4+
—— 2ik -(B*+k7)U=0
(B2+k)U=0 -b<x<0
-b<x=0 4.68
a... (4.68)

Solution of the equations (4.67) and (4.68) leads to the condition

g? - iy?
Jo sinh (Bb). sin aa + cosh (Bb). cos aa = COS k(a + b) yee (4.69)
a
In order to express the relation in a more simplified form, Kronig and Penney

183
 Applied Physics

considered the case, for which the potential barrier becomes periodic delta functions,
i.e., V, tends to infinity and b approaches zero, but the product Vb or Bb (since as
Vo 00's B* — o) remains finite. With these limits imposed, eq. (4.69) becomes

(mV ,b/ A? a) sinaa+cosaa=coska __........ (4.70)

Let us define a quantity P such that
2
Pail bane eMimeelit es ee Pee (4.71)
2 b>0 p)
Bo
which is clearly a measure for the ‘area’ V ,b of the potential barrier. In other words,
increasing P means binding a given electron more strongly to a particular potential well.
It also tells us that when P -» 0, the electron behaves like free electron.
Now, (4.70) can be written as

Pe gees COS Gai =Scosika pooriag Sf) 4i (2404. (4.72)

aa
Now, if we assume aa = ka,

then from (4.63) o2 = ag ke

h

k2 72
and hence Es
2m
which was the result obtained in the case of Sommerfeld free electron
model.

gin aa.+ Cos aa

awa

Fig. 4.15 Plot ofthe function P22 %4 3m

+ cos aa for p = re - The allowed values of the energy E
Qa

are given by those ranges ofa= E4

2mE |?
for which the function lies between +] and —1.
h

184
Electron Theory of Metals

If a graph is plotted representing the left hand side of the eqn. (4.72) on the
y-axis as a function of aa, for a value P = 37/2, we get a curve as represented in
Fig. 4.15. Since, a2 = 2mE/h* the abscissa is a measure of energy. Further from
eqn. (4.27) one can see that due to the restriction imposed by the right hand side of the
equation, cos ka, which can take values only between +1 and —1, the left hand side of
the equation is also restricted to vary between those two limits only along the y-axis for
different values of aa. Due to these restrictions, from the Fig. 4.15, the following
important conclusions can be drawn:

Allowed energy band

m

Energy

oP>0 ++ | —+P>0
Fig 4.16 Allowed (shaded) and forbidden (unshaded) energy ranges as a function of P. The extreme
left corresponds to P = 0 (free electrons), the extreme right to P = 00

Conclusions from Kronig - Penny Model :

1) The energy spectrum of electrons consists of an infinite number of allowed energy

bands separated by intervals in which, there are no allowed energy levels. These
are known as forbidden regions. The boundaries of the allowed ranges of
nt
aa correspond to the values of cos ka = + | or ka = nm i.e., k = ar

2) When ata increases, the first term on the left hand side of (4.72) decreases, so the
width of allowed energy bands is increased and forbidden regions become narrower
3) The width of the allowed band decreases, with the increasing value of P i.e., with
increasing ‘binding energy’ of the electrons. When P tends to infinity, an extreme
case, the allowed energy regions become infinitely narrow and the energy spectrum
becomes a line spectrum. In that case, (4.72) has a solution only if sin aa = 0,
1.¢., da = + nt WOIG. Ni lo. 3s nae

From the above and (4.63) we have

185
 Applied Physics

E(t fod Ole Pee ce me (4.73)

In this case energy levels are discrete and electron is completely bound and
behaves like a particle in a constant potential box of atomic dimensions.
Hence we conclude, that in one dimensional problem for the limiting case, the
spectrum of permissible energy values is bound to consist of continuous regions
separated by finite intervals. By varying the quantity P from 0 to 0, we pass from the
case of free, to that of bound electrons and thus study the changes in the allowed and
forbidden ranges of energy and the wavefunction. For a given value of P, the position
and width ofthe allowed and forbidden bands are obtained by erecting a vertical line as
shown in Fig. 4.16, the shaded area correspond to allowed bands.
From (4.72) it is possible to obtain the energy E, as a function of the wave
number k. The variation in E is shown in Fig. 4.17.

15

(8ma’/h?)
E

Fig. 4.17 Plot of energy vs wave number for the Kronig-Penney potential, with P = 37/2

Fig. 4.17 shows discontinuties in E (drawn heavily). These discontinuties occur

nt
atka=+nzork= “a where
n = 1,2,3 .....

These “ka” values define boundaries, the I, Le UU ee etc., called Brillouin Zones.
It must be noted that Fig. 4.17 gives only half of
the complete E (k) curve, for positive

186
Electron Theory of Metals

‘n’ values; thus the first zone extends from —1 to +n. Similarly second zone has two
parts; one extending from 1 to 27 and another part from —7 to —27 (not shown).
Each position of the curve may be called a band. The characteristics of the
curve are :
1) they are horizontal at the top and bottom
2) they are parabolic near the top and the bottom with, curvatures in opposite directions
and
3) d?E/dk? is positive in the lower portion of the band and negative for the upper
region.
A further important conclusion may be drawn from (4.72); within a given energy
band, the energy is a periodic function of k or ka. For example, if one replaces ka by
ka + 2mn, where ‘n’ is an integer, the value of right hand side of (4.72) does not
change. In other words, ka is not uniquenly determined. Therefore, for convenience
a “reduced wave vector” which is limited to the region
Sokaaa” Ga @) (Sew ye & On ote (4.74)

is introduced.

4.16 Number of Possible Wave Functions Per Band :

So far, we have considered the crystal to be of infinite dimensions, as per the

definition of the periodic lattice. But any crystal is of finite dimensions and hence it is
necessary to consider the consequences. of this boundary conditions. For a linear
crystal of length ‘L’ the boundary condition may be written as
TiGe “il 1S ais 0.6) Mme (4.75)

now making use of the Bloch functions

ences UNIX +L) = ee (x).
Because of periodicity
U(x + L) = U(x)
and boundary conditions thus require

a Siete a eee eee (4.76)

the number of possible wavefunctions (or k values) is the range dk is, differentiating
(4.76)

187
 Applied Physics

since k is limited by the relation ka = nz, it follows that the maximum value ofn in 4.76
is given by L/2a = N/2 where N is the number ofunit cells. This leads to the conclusion
that “the total number of possible wavefunctions is any energy band is equal to the
number of unit cells N”.

4.17 Velocity of Electrons and Effective Mass :

According to de Broglie, a particle has dual nature. It can behave as a particle
as well as a wave as seen in chapter 3. The particle velocity v is equal to the group
velocity of a wave packet, which is a representative of the particle.
y ome Hdo
amie are
where @ is the angular frequency. Then the energy of the particle E = fo or
O=E/h.

vifus igTahisahlcnrel
Dae pol kate ofdeste ines anaes (4.78)
eefhkae: an
dE
cra
fe? _ tk
. les mous (4.79)

Fig. 4.18: E vs k curve

Now, from band theory the energy versus k
curve would be as shown in
Fig. 4.18. Because it is a symmetrical curve havin
g points of inflection at M and N, the
value ofdE/dk is different for different portions of the
curve. At points M, O and N,
since curve is flat, dE/dk = 0. That is, at these
points velocity is zero. At the points A
and B, dE/dk is maximum, but, negative at
A and positive at B. That is, at a value

188
Electron Theory of Metals

k = k, the velocity reaches a maximum and decreases on further increase of k, with

increasing energy, reaching zero value at k = n/a or —n/a, as shown in Fig. 4.19. This
behaviour is completely different from that of a free electron.

kK——>

Fig. 4.19 v vs k curve

If we apply an electric field € on such an electron which is moving in one

dimensional periodic potential and which is initially is a state k, then the energy dE
gained by the electron in time dt is

dE ee ot

using 4.78. = e€ [(1/h )(dE/dk)] dt $4, (4.80)

Be dE
Writing ra Rhee Abas dk, we have

tee,
Vie eC /h vitae (4.81)

From Newton’s Second Law, the acceleration ‘a’ is

ae - - <[(v/ayat/4t) = (1/n) CE dk
dk? dt

d°E
substituting for dk/dt, as (ce/n?). rs oie (4.82)

But, we know that for a free electron a = aa

189
 Applied Physics

Therefore, from this expression and (4.82) we can write

That is, mass is not constant

for an electron moving is a periodic
lattice but is a function of k as it
depends on d?E/dk?. Hence the
expression (4.83) for the mass of
an electron is known as effective
mass and is represented as m*. The
effective mass m*-varies with k as
shown in Fig. 4.20. The curve
shows that m* is positive between
-k, and +k, and negative at other
values of k upto + m/a. Because of
this nature of mass of the electron,
Fig. 4.20 Variation of m* with k
the velocity decreases beyond k,, as
the electron gets decelerated.

4.18 Metals, Insulators and Semiconductors:

Metals, insulators and semiconductors can be distinguished on the basis of band

theory. For this, let us consider a particular energy band filled with electrons upto a
certain value k,, as shown in
Fig. 4.21. As for as the influence of
an electric field is concerned,
information about the free electrons
in the band of N electrons would be
of interest. If we can get this
information, conductivity associated
with this band can be estimated. The
effective number of ‘free’ electrons
in the band is Noga 2.1). where
fi,= m/m* and summation extends
over all the occupied states in the
m* band.
Fig. 4.21 Energy Band filled upto k,

190
Electron Theory of Metals

The number of states in the range dk (taking in to a account the spin) for one
dimensional lattice of length L is
L
divtseSidks
n
™

In the filled region of the band (Fig.4.21), within the limits —k, to +k,
k k

ae mabe = 2Lm/nh? cea

= 2Lm/n hh? (=) drdetys' (4.84)

dk k=k,

substituting for f, and m*.

From the expression for N.,- we can draw the following conclusions:
band vanishes,
(i) The effective number of (free) electrons in a completely filled
because dE/dk is zero at the top of the band.
filled upto
(ii) The effective number of (free) electrons reaches a maximum for a band
reaches a maximum .
the inflection point of the E (k) curve, because there dE/dk
a given
From the above discussion, we can say that for any solid for which
band is complete ly empty is an insulator ,
band is completely filled and the next allowed
solid in which a certain band is
as shown in Fig. 4.22(a). On the other hand, a
be called a conducto r
incompletely filled would show metallic character or may
would be valid only at
(Fig. 4.22(c)). The situation shown in case of Fig. 4.22(a)

Empty
Empty ee OD)
conduction band conduction band

(b) (c)
Energy

(a) an insulator (b) a semiconductor and (c) a metal

Fig. 4.22 Electron distribution at T=0Kin

191
 Applied Physics

absolute zero., where the crystal is in its lowest energy state. Ata temperature
different
from zero Kelvin, some electrons from the upper filled band (valence
band) can be
excited into next empty band (conduction band) and the conduction is
possible. Number
of electrons which move into conduction band are determined by
the Fermi-Dirac
Statistics. If the forbidden energy gap is of the order of several
electron volts, the
number of electrons which move into conduction band remain negligibl
e and the crystal
remains to be an insulator for all practical purposes. An example
is diamond, for which
the energy gap is about 7eV. Fora small gap, say leV, the number
of thermally excited
electrons may become appreciable and in this case one
speaks of an intrinsic
semiconductor. Examples are Silicon and Germanium with energy
gaps 1.21 eV and
0.77 eV respectively at absolute zero. It is evident, that, the difference between
semiconductors and insulators is only quantitative. Infact all the intrinsic
semiconductors at T = 0 K are insulators, where as
all the insulators above T>0K
may be considered as semiconductors to some extent.
Because of increase of free
electrons in conduction band in semiconductors, with
the increase in temperature, the
conductivity of semiconductors increases. However,
the conductivity of metals
decreases by a small amount, due to interaction scatterin
g between electrons and lattice,
as the number of free electrons remain constant.
Example :
Calculate the conductivity of Copper at 300K. The
Collision time for electron
scattering is 3 x 107!4s at this temperature.
Solution :

The number of free electrons per m? of Copper

is given by;
Nd
Sere
= £:023%10" x8960 _ eee ie
0.06354
ga BOT _ 85x10" x (1.602) x 1038 x3x 10-4
m O1x108!
= 7.19 x 107Ohm my!
N = Avogadro number = 6.023 x 1023 mol!
A= At. wt = 63.54 amu
d = density = 8960 kg/m3
= charge of electron = 1.602 x 10-!9C
m = rest of mass of electron = 9.1 x 10-3!kg.

192
Electron Theory of Metals

Example :
Calculate the Hall coefficient of sodium on the basis of free electron model,
given that Sodium has a bec (body centered cubic) structure of cell of side 4.28A.
Sine each sodium atom gives out one electron, and there are two sodium atoms
per unit cell of a bec (as affective number of lattice points are two),

number of electrons 2 3

R, n = -—I/ne, e=16 «10 PC

_ (4.28) x10°° am?

lbs. CG
= 45x 10°
= 24.5 x 107-!9m3/c.
Problems

1) Ignoring the random velocities of electrons, calculate the Hall coefficient of copper
which has a FCC structure. If the conductivity of Cu is 108(O0hm m)!, estimate
the electron mobility.
(Hint : FCC structure has 4 effective lattice points)
2) Assuming one free electron per atom in Copper whose conductivity at room
temperature is 108(Ohm m)!, calculate the relation time and mean free path for
electrons at the Fermi level.
3) The Fermi level for potassium is 2.1eV. Calculate the velocity of the electron at the
Fermi level.
4) Assuming the free electron theory, calculate the Fermi energy of bec lithium and
fcc aluminium. The dimensions of the unit cell are 3.50A and 4.04A respectively.

Questions

1) What is free electron gas model of metals. Discuss salient features of Sommerfeld
modified free electron gas model and its success in explaining some physical
properties of metals.
2) (i) What is Fermi energy? Discuss the distribution of electrons in different energy
levels based on Fermi Dirac statistics.
(ii) How does the distribution vary with temperature ?
3) What are the short comings of Sommerfeld modified free electron model?

193
 Applied Physics

4) Explain the phenomenon of electrical conductivity of metals on the basis free

electron model. What is Lorenz number?
5) Discuss the variation of electron energy with quantum number and wave vector.
6) Show that the Kronig-Penney model leads to energy band structure in solids?
7) Discuss the consequences of periodic potential on the variation of velocity and
mass of an electron in a solid under the influence of an electric field?
8) On the basis of band theory, explain how materials are divided into metals
(conductors), semiconductors and insulators. How does the density states curve
change in the above materials?
9) What are the conclusions that are arrived at from Kronig-Penny model?
10) What is meant by 'Effective Mass' ?
11) Show how does velocity of the electron change with K-Vector's magnitude?
12) Give the energy vs k on the basis of Kronig - Penny model.
13) Explain the terms ‘drift velocity' and 'mean free life time’.
14) What is meant by ‘work function’ of a metal?
15) Define the term ‘density of states.
16) Obtain the expression for conductivity of metal in terms of relaxation time, mass
and concentration of electrons.
17) Explain salient features of Sommerfeld Theory.
18) Does Fermi energy of a metal change with temperature?
19) Enumerate the factors that attect the electrical resistivity of metal.
20) What is meant by 'Matheissen Rule’ ?
21) Write down the 'Schrodinger equation' for a free electron ina
potential well. Give
the enrgy spectrum of the electron in the well.
22) What is the relation between Fermi energy of a metal and the
velocity of electrons?
23) How does the drift velocity change with time when the
electric field across a
metallic sample is switched off?
24) Write a short notes on "Hall effect".
25) Define the term 'mobility' of a charge carrier. Does it change
with temperature?
With the electric field ?
26) Write down the expression for Fermi-Dirac distribution
function. Mention how
does it respond to temperature changes.

194
 55
DIFTECTRIC.PROPERTIES-OF- SOLIDS

5.1 Introduction:

Dielectrics are basically electrical insulating materials. This is the definition one
finds for most of the engineering practice. However, the more general definition for a
dielectric is given as a substance whose basic electrical property is the ability to be
polarized and in which an electrostatic field can exist. The electrical insulator is an
engineering material possessing the properties ofa dielectric and preventing the leakage
of electrical charges in electrical engineering devices. Such substances as bakelite,
PVC used in electrical wiring and tubes, many other polymer materials, come under
this category. But in the modern technology, the term ‘dielectric’ has a broader meaning
and they are considered as materials in which electrostatic fields can persist for a long
time. The electrical insulators are passive materials while many dielectrics in use are
active and play a vital role in many electronic applications. Such materials as capacitive
materials in capacitors, ferroelectric, antiferroelectric, piezoelectric, or magnetic materials
come under the umbrella of dielectrics. No electronic circuit can be free from dielectric
devices. Over the years, the importance of dielectrics has over grown in electronic
industry on par with semiconductors from the days of simple dielectrics like

195
 Applied Physics

gutta-percha, natural rubber, paper, or vegetable oils. The development of a dielectric

is linked now with progresses that are taking place in radio electronics, nuclear
engineering, miniaturization of electronic circuits, computer hardware etc.
Dielectrics are electrical insulators. They possess high resistivity values in the
range 10°Q — m — 10!6Q — m. Under high voltage bias, they allow very little current
(in the range 10-°°A — 107!4A). They withstand very high voltages. The conduction
phenomenon is mostly associated with ionic motion through defects or hopping of
charges. They have no free charges. The electrical properties of a dielectric are associated
with its inherent property of possessing electric dipoles. If such a material is placed
between the plates of a parallel plate capacitor (Fig.5.1), its capacitance increases. The
electric field E = V/d.

5.2 Dielectric Medium between the plates ofa capacitor :

Consider that the capacitor (Fig. 5.1) comprises rectangular, parallel plates of
area ofA (meter)? each a distance d apart.

Fig. 5.1
Let the charge across each plate of the capacitor be Q in coulombs due
to the
potential difference between the plates, then the charge is proportional to
the voltage

196
Dielectric Properties of Solids

between the plates of the capacitor i.e. the electric field strength between the plates
V/d, volts/m.

O=CV=2CV
Where C, = ¢, A/d, and €) = 8.854 x 10-!*F/m, the capacitance when free
space is present between the plates, Er is the relative permittivity of the medium. The

electric field E at every point in the medium falls by a factor -of its original field.
Ege,

The new capacitance in the presence of the dielectric medium between the plates is
€,C,. The factor Er is called specific inductive capacitance and is independent of the
shape and size of the capacitor. The purpose of the dielectric medium is to maintain
.two large plates at a very small separation, increase the potential difference such that
the capacitor can withstand without break down and then to increase the capacitance.
€, is always a dimensionless quantity. The parallel plate capacitor as shown in the Fig.
5.1. exhibits the surface charge density on the plates,

ve [E+ ax ya (5.1)
0

R S N

x fedx+ [Edx+ [Ex

0 R S

= a f= ideit ex, |
Eo a 1

= Sla-{i-4]
Es oe

Choe 5 Aoze.
C= V ~ le,d-te, - 1] Sacanese (5.2)

The magnitude of electric field intensity E is same at all points inside the dielectric
which is placed between the two electrodes provided the dielectric under consideration
is homogeneous. In free space, the electric field is defined as the force per unit charge.
q
The electric field E arising from a point charge q is given as E = r Volt/metre;
Aner?
°

where r is the distance between the test charge and q. r is the unit vector in the

197,
 Applied Physics

direction of r. The electric field is the negative gradient of the potential distribution in
the region i.e., E=—dV/dr. It is a measurable quantity in free space. But the measurement
of electric field inside a dielectric is impractical. Therefore a theory or model can be
formulated for the behaviour which relates the externally measurable dielectric constant
to the internal structure of a dielectric. The externally applied electric field is related to
the ability of the dielectrics to become electrically polarized under the action of the
field.

53 Atomic View ofa Dielectric Polarization :

A dielectric is a substance in which the charges or electrons are not free. They
are tightly bound to the nucleus of the atom. An ideal dielectric does not possess free
charges. The molecules or atoms in a dielectric contain positive and negative charge
species. Under no electric field condition, the dielectric is neutral and the positive
charge - negative charge species called dipoles are randomly oriented and they
compensate each other. When an electric field is applied, the molecules or atoms are
acted upon by a force. The positive and negative charges are pushed in directions
Opposite to each other as shown in figure 5.2. These displacements are extremely
small, very small fractions of the molecular diameter. The charge configuration is
altered and there exists a displacement of positive charge in the dielectric relative to the
negative charge. This results in the creation of electric dipoles. This process is called
polarization. The polarization is a molecular phenomena.

atom o—® ©o—®e

e-—® ©-@ O—®
E o—® o-@ o—6
e—-o ©—® o—o
ee 0 e—® ©—-® o—®
the fieldE

Fig. 5.2

A dielectric has a large number of atoms or dipoles in its volume. In the absence
of an external electric field, the volume of the dielectric does not have a net dipole
moment. This is because the algebraic sum of the charges for all the molecules in the
given volume is equal to zero (The dielectric is assumed to consist of very large number
of molecules and the dimensions of the dielectric are large compared to the dimensions
of the molecule). When an electric field is applied, the action of
the field now brings the

198
Dielectric Properties of Solids

charges of the molecules into certain order and molecular dipoles are produced. There
is now a net dipole moment associated with the dielectric. Each molecule contributes
to the dipole moment.
The molecules become polarized in the presence of the electric field. If there are
N molecules and each molecule acts like a dipole with a dipole moment p. The algebraic
sum of the dipole moments is equal to )'p,, i referring a molecule ‘i’.

NAV
> Pi =NAVp
= PAV
i=l
Where P represents the average dipolemoment of each molecule, V is the volume
element.

P= Dehn

P is a vector quantity and is called polarization density and it is a macroscopic

description of molecular phenomena, the tendency of molecules give rise to a dipole
moment. is proportional to E.
Peter sivpalsrdiamewlyrb beisely bate ata (5.3)
This is true for isotropic linear dielectrics. For isotropic dielectrics the directions
of the vectors P and E coincide at each point and the proportionality constant in the
equation (5.3) is a scalar quantity and is called polarizability. For anisotropic dielectrics,
P and E differ in directions and the proportionality constant a is tensorial in nature.

5.4 Electric Flux Density, Electrical field and Polarization in a

Dielectric Medium :
Consider once again the parallel plate capacitor as shown in Fig.5.1. As a result
of the electric field, the plates acquire the charge. At any point between the plates on
the surface (called gaussian surface) say near R, the electric field is simply q/A €,. But
when dielectric is present, the charge on the gaussian surface is reduced by an amount
1
q —q', where q' is the induced surface charge at R. The electric field is vem 2)
r

Then
 Applied Physics

The quantity q/A is equal to D called electric flux density

Dee ee a ee ee ee eee (5.4)
When there is no dielectric,

D = & E
When there is a dielectric,

D= 3 = cE where ¢ = EoE,

cE = EgE sp |b

ie. Shee OU | on a or aS peat Pr > (555)

P
1.€.,
1.€ [=
(ep=s1) otSS an x eC aa? » eee ae 5.6
(5.6)

Where x is the electric susceptibility of the dielectric medium. XEq is called

absolute dielectric susceptibility and has units F/m.
The quantity D is introduced and is conveniently used while referring to different
media. ‘D’ is called electric displacement (electric induction). Suppose there is a charge
placed at the centre of a sphere of radius r. The charge q will emanate q lines of force
and these lines cross the surface of the sphere. The number of lines of force received
by unit area is called electric displacement. D depends only on the distribution and
magnitudes of charges which produce electric field but is independent ofthe nature of
the medium. In the Fig. 5.1, the field lines of the three vectors D, E, P are shown. It
indicates the variations ofthe field in the dielectric medium.
‘D’ has the units of coulomb/m? and D and E have the same direction in isotropic
homogeneous medium. The polarization P is defined in equation (5.3). Polarization is
present in the medium.
Gauss Law :

The permittivity of any material is greater than unity and for vacuum €.= 1 and
P is equal to zero. The dielectric susceptibility is equal to zero in such a case. The most
important relation between D and the charge in a volume is given by Gauss theorem.
Gauss law states that the total normal electric induction from the charge q through a
closed-surface is equal to the charge enclosed. The flux density D at any point on the
surface is q/m r*. Suppose the electric flux is like a flow across the surface dS as
shown in Fig. 5.3. Then total flux passing through dS is dy, = De 7dS="Dico
s
d :
adS = qcosa ye . But dS cosa/r? is the solid angle dQ subtended at q by the elemental
TO

200
Dielectric Properties of Solids

area dS. The electric flux is given as (q/4n) dQ. . The sum of the flux over the whole
surface will be the integral of the solid angle over the whole surface which is 41

F : q
steradians. The total flux is therefore [2 Jan=. Thus for a closed surface

qD, ©d5ity dees AAT-eapidis Gvitcbart wuisbacheone (5.7)

This is called Gauss law. The surface integral of the electric flux lines through a
surface is equal to the charge that the surface encloses. The electric flux is a property
of any vector field. The electric flux is measured by the number of lines crossing or
cutting through a surface. In terms of electric field, the Gauss law is written as

eS. qE*dS=q. If a surface encloses equal and opposite charges, the flux is zero.
A charge outside a surface does not make contribution to the value of q.

Fig. 5.3

D is a parameter which is important for a dielectric, which is made up of several

layers. D is same in all the layers but E is different. The permittivity of a dielectric
determines its ability to form an efficient capacitance. A capacitor stores energy for all
engineering purposes, normally permittivity is considered as dielectric constant especially
in the case of solids. Gauss law is used to evaluate electric field if the charge distribution
is symmetrical and the integral can be evaluated. Coulomb's law can be deduced from
Gauss law. The student should take this as his assignment and do. Gauss law is one
of the fundamental equations of electromagnetism

55 Polar and non polar dielectrics :

The dielectrics can be divided into two classes, I) Polar (dipolar) dielectrics, 2)
Non Polar dielectrics.

201
 Applied Physics

It is already mentioned that the polarization is a molecular phenomena. A molecule

consists of atoms or groups of atoms or ions. Each molecule may possess a net dipole
moment. Each one possesses a charge. The spatial arrangement of charges in the
molecule may be different in different molecules. All the positive charges and negative
charges in the molecule can be replaced by one positive and one negative charge. They
are mutually equal in magnitude and located at the centres of gravity of separate positive
and corresponding negative charges. The centres of gravity may not coincide in some
molecules. Then one would have polar molecules. If they coincide, one would have
non-polar molecules. Even in the absence of an external field, a polar molecule possesses
a net dipole moment p = ql (Fig. 5.4).

—>
@———_———_ ©
+q <—/—> ~~ 9

K @———® |
+g ="¢

(K | Potassium iodide)

Fig. 5.4 Diagram ofa polar molecule, I is the arm ofthe dipole.

The dipole moment would be the order of 10°C-m. All hydrocarbons are non polar or
weakly polar i.e. (P = 0). The other examples for non polar type are monatomic molecules
(He, Ne, Ar, Kr). The molecules consisting of two identical atoms linked by a homo
polar bond like H,, N,, Cl,, etc are also non polar. CO, is non polar. ( Fig 5.5)

202
Dielectric Properties of Solids

H ssh
4

PN
J -AC +4
H =a, H

Qi, HOT Ue

(c)
Fig. 5. 5 (a) CO, (b) HO (C) CHy

Methane CH,, Benzene C,H, are non polar substances. But H,O, CHCl, CH,CL,,
CHCI, etc. are asymmetric and polar. They have a net dipole moment. For example
water molecule has a net dipole moment 6.1.10°°C-m.
Hydrocarbon compositions like polyethylene, palatine, ceresui, poly iso bulytene,
Polypropylene, polystyrene, rubber, electrical insulating petroleum oils etc. are non
polar. But polyvinylchloride is a strong polar substance. Polyvinyl alcohol, cellulose,
phenol formaldehyde, etc., are polar substances. Polytetraflouro ethylene (ftoxlon-4,
Teflon) is non-polar, but polytriflouro ethylene (ftorlon-3) is a polar substance.

5.6 Polarizability (Electronic, Ionic and Orientation

Polarization):
The induced dipole moment in a dielectric as a result of an external electric field
is proportional to the external electric field for linear dielectrics. The reasons for the
dipolemoment in the dielectric will be discussed later. The above statement is expressed
as P = aE. P represents the induced dipole moment. The value of P is measured in
Coul.m. and E in V/m. Therefore the units of a are F.m?. In CGS system o is expressed
in cm}. The above relation is strictly valid only if E is small. In case E is large or very
large, P does not increase at the same rate as earlier and the rate of increase of P will be
field the
smaller and this leads to saturation effects. At particular value of electric
dielectrics saturate at the field of the
polarization reaches a saturation value. Many
order of 107Volts/m.
or
The polarizability a of a dielectric represents the nature of individual atoms
character istic of a dielectric .
molecules. It is an important microscopic electrical
It is its bulk property . The
Polarization takes place in all the molecules of a dielectric.
elastic in
displacement of charges in the presence of the electric field is very small and
return to their
nature. If the field is switched off, the displaced charges may tend to
of charges in the presence of an electric field can be
original states. The displacement
s. There arises a displace ment current
in atoms, ions or dipolar molecules of substance
applied is taken off or when
(capacitive current) in a dielectric when the direct voltage

203
 Applied Physics

the magnitude of the voltage is changed with time. This current can exist only when
the plates of the capacitors containing the dielectric in between them are subjected to
alternating voltages. The displacement current is a quarter of a period ahead of the
voltage. Suppose a dielectric contains n molecules per unit volume. Then the induced
dipole moment per unit volume is n aE. If the induced dipoles are independent, a total
electric dipole moment per unit volume or the polarization P is n aE.
They are several types of polarization. Each type is associated with an intrinsic
physical mechanism. Three basic types of polarization will be mentioned here. They
are 1) Electronic, 2) Ionic and 3) Dipolar polarisation or Orientation polarisation.
Electronic polarisation is the displacement of electrons (or electron cloud) with
respect to the nucleus. Electronic polarization occurs in all atoms or ions and is always
present when a dielectric is under the action ofan electric field. This type ofpolarization
occurs during a very short interval of time (.1O=? sec} which is the time period
of
oscillations of uv rays. Homo polar neutral dielectrics possess this type of polarizati
on
only.
Ionic polarization is the mutual displacement of ions forming ionic molecule
.
This type of polarization occurs during a short time ( 1or?1072 sec} and
this time
is longer than that for electronic polarization. It occurs in some substanc
es where ionic
bonding between the constituent ions exists.
Dipolar polarization is associated with polar molecules. The
tendency of the
dipolar molecules to rotate into the field direction gives rise to this
type of polarization.
It is temperature dependent i.e., it is controlled by the thermal motion
of molecules. It
occurs in gases, liquids and amorphous solids.
Both electronic and ionic polarizations are caused by deforma
tion of an atom or
molecule which is a displacement of charges with respect to each
other in the direction
of electric field. The polarizations are set very fast. The tempera
ture has no significant
effect on their magnitude. Once the electric field is remove
d, the polarizing species
return the electrical energy to the source. Therefore, it
is not associated with any
losses.

5.7 Local field in a dielectric media Claussios - Mossi

tti Equation :
It has been said that if a molecule attains an induced
electric moment P and this
induced electric moment is equal to the product of
the polarizability of the molecule
and the intensity of electric field acting on the molecu
le. A dielectric consists of large
number of molecules. Each molecule is associ
ated with a dipole moment P. Each
molecule is surrounded on all sides by other polari
zed molecules of the dielectric. All
the other molecules now impose an internal electri
c field on the molecule under

204
Dielectric Properties of Solids

consideration in addition to the external electric field. Therefore at a given point in the
dielectric, there exists an effective field which is equal to the sum of the external field
and the internal field due to all other molecules around it. The total intensity of electric
field (E) will be the sum of the external field intensity E and the components of the
internal field caused by the action of other polarized molecules on the molecule under
consideration.

t
+

4c+
tat
te
fat
4eat

Fig. 5.6 A dielectric is placed between two electrodes

Consider a dielectric placed between two electrodes as shown in Fig.5.6. Let a

sphere in a dielectric be described containing a molecule considered at its centre. Let r
be the radius of the sphere and let it be sufficiently large compared to the dimensions of
the molecules, but sufficiently small compared to the dimensions of the dielectric
placed between the electrodes. On the molecule at the centre, there are three fields
acting.
1) The external electric field E,

2) The internal field E,, due to the molecules that are present outside the sphere.
3) The field E, due to the molecules inside the sphere. Then the total internal field
intensity E' is |

E'=E+E,+E, (5.8)

Let it be imagined that the dielectric be removed from the sphere. As the pattern
of the electric field is not disturbed, this is effectively equivalent to placing a charge on
.
the surface of the sphere. Let the charge density on the surface of the sphere be o
Then

o = Pcos@ (5.9)

205
 Applied Physics

Where @ is the angle between radius vector r and the direction of E. Let an
elemental area dS is considered on the surface of the sphere. Then charge dq due to
this surface dS is

dq.#uods.5-PCos0dS. fay moion-ddlt wlieeatine bleed, (5.10)

As a result of this charge, an electric field of intensity dE, is created at the centre
of the sphere. Where

dq P
dE, | = AFAR
= ECos0
ON dS
et (5.11)

This can be resolved into two components (1) dE,Cos@ parallel to the direction
of E and (2) dE,Sin® the component perpendicular to E. The two components are

dE Cos® #— Cos O'dS ff 5 0 (5.1 1a)

€)40r

and dE, Sin@ = Dead Cos@VieSin6 dS be ie

€40r nie a gn (>cKLB}

The perpendicular components of the upper half and lower half of

the sphere
compensate each other. The components parallel to the electric
field will actually
contribute to the total electric field.
The total electric field intensity E' due to the entire charge on the sphere is
determined by integrating dE, over the whole surface of the sphere. The perpendicular
components in the upper and lower halves cancel out each other and only parallel
components contribute to the total field.
Integrating ( 5.11 a), one gets

Tl

E, l = I|dE,Cos@ dS = -P
me Cos*0 dS Fee (S112)

But dS=2nr? Sin@ dO. Therefore equation (5.12)

becomes

E, 1 = ——
Soa leks
|Cos*6 Sin@ dé = ——
P
2, J Bere te TNE ee (5.13)

The direction of E, is same as that of KE.

206
Dielectric Properties of Solids

From the considerations of symmetry, E, is taken as zero. Hence the total internal
field is given as

P
E*= Es SeauRe
—— ThSy wre phish Whtsinnag arth (5.14)

but is=e,(e, —1)E

Inserting the above relation in (5.14), we have

P e, -1 |=E~*—
(c, +2)
E'=E+——=B| 1+ ——
3€, [ 3 ) 3 eee nee (5-15)

Therefore the internal field E' depends on the nature of the dielectric material as
the expression for E contains ¢,. In the case of free space, E’ coincides with E.
Consider a unit volume of a dielectric, then the value of polarization is given as
P=Np=NoEO 222 2 terete (5.16)

Where N represents the number of molecules per unit volume of dielectric.

cole, -)E= Na Ep aed (5.17)

(e, = 1) ns Na
or (e,7 2) 2a esha im |b hooriee 24 oe0 A (5.18)

The above equation is called Clausius-Mosotti equation. It relates the macroscopic

quantity, the dielectric constant and the polarizability, (the microscopic parameter) and N.

Multiplying the equation (5.18} by M/p, where

M = molecular weight and

© = density of the dielectric material, one gets

(c,-1)M_NaM __ aN,
(c,+2) p 3€) p —3€

where N, is the Avogodros number 6.023 x 102°k mole”, is expressed in F.m?, € in

F/m. (8.854 x 10°!? F/m).
¢.-l)M
The quantity eae is called Molar polarisation of a dielectric. Its dimension

is equal to that of the ratio of volume to mass.

207
 Applied Physics

The above expression is effectively valid only for all neutral molecules i.e. neutral
liquids and especially for gases in which molecules are far apart. The expression is not
valid for strongly polarized substances. This aspect will be discussed later. The above
expressions do not indicate how the permittivity depends on the external parameters.
In reality the permittivity depends on the frequency of the voltage applied on the
dielectric, temperature, pressure, humidity etc. When dielectrics are used in devices,
like a capacitor in an electronic circuit, the above external parameters play a vital role in
the device operation.

5.8 Electronic Polarisability :

The process of electronic polarization will now be considered. Consider an

atom with atomic number Z, then the charge on its nucleus is + Ze. When this atom is
subjected to electric field, the nucleus and the electron cloud are pulled apart by an
amount x. Then the force along the direction of the field is

Bi Ze awit filesd) aiitestaib 2tosmudov finy 3 9) (5.20)

e!

Fig. 5.7: An atom without field b. with the field

The nucleus is surrounded by electron cloud of charge -Ze

distributed over a
sphere of radius r. the charge density is given as

Assuming uniform charge distribution of electron cloud,

the total charge in the
electron cloud is

4 = 3
()rx3p= shx3x| —34e |___Zex
3 3 1 a3 ee .22)
Dielectric Properties of Solids

The force (F,) due to the attraction between the nucleus and the electron cloud
now is, i.e., the Coulomb force,

rare Ze ? B55 eee Sakde:

2 Ame )x° 3 Anegr? sone (5:23)

In equilibrium condition, the forces Fy and F, are equal.

ee
(50 9p ll
Aner

Aneor
ii.e., eeZe eee ee (5.24)
24

Electronic dipole moment induced P, is

Pave ke (5.25)
Therefore the electronic polarizability a,, is given as

Ce ee ee (5.26)
Consider the atom of hydrogen. It has a radius r = 0.053 nm. Hence, it has an
electronic polarizability a = 4x (8.85 x 107!) x (0.53 x 10°!) = 1.67.104!Fm2. The
electronic polarizability for halogen atoms are, (1) Fluorine a = 4.45 x 104! Fm?, (2)
Chlorine a = 2.67 x 104! Fm2, (3) Bromine a = 40.05 x 104! Fm?, (4) Iodine
a = 64.5 x 104! Fm?.
The polarization P = no,, E, where n is the number of atoms per unit volume.
The homopolar dielectrics (i.e. without ionic structure) possess only one type of
polarisation i.e. the electronic polarisation.
An optical wave is an electromagnetic wave. When it passes through a dielectric
medium, it undergoes a phase change as well as velocity change. In an electromagnetic
wave, a changing electric field produces a changing magnetic field which in turn
generates an electric field and so on with a resulting propagation of energy in the
forward direction. Maxwell developed Maxwell equations as generalizations of circuit
relations from Gauss law, non-existence of monopoles, Ampere’s law and Faraday’s
law of electric induction. Maxwell’s equations are point relations i.e., they are valid at
every point in the medium. Based on his equations, Maxwell developed the wave
equations for electromagnetic waves in a medium. He showed that the velocity of

propagation of such a wave for unbounded medium is given by V = a1 , where LU is

pe

209
 Applied Physics

the permeability of the medium, and ‘¢’ the permittivity of the medium. Normally for
dielectrics which are nonmagnetic '}!' can be taken as [ 9, the permeability for free

space. The velocity of waves in the free space V, = . As the value of ¢ fora
Ho&o
medium is greater than ¢ 9, the velocity of an electromagnetic wave ( optical wave) in
a medium is less than that in free space. The origin of the refractive index of the
medium is seen from the point of permittivity of the medium. The refractive index is
defined as the ratio of velocity of electromagnetic waves in free space to their velocity
in the medium. Thus the refractive index for a medium is greater than that of the free
space. The refractive index for a medium is equal to the square root of relative permittivity
of the medium. Therefore, Clausius-Mosotti equation can be expressed for an optical
medium as

Ky ol ~ oe,

where 1 is refractive index of the medium. The above relation is called Lorentz-
Lorentz equation. Homo polar dielectrics are dielectrics without ionic structure. They
possess only one type of polarization i.e. electronic polarization. The determination of
Q., is possible from the determination of the refractive index. The dielectric constant
at optical frequencies arises almost totally from the electronic polarizability. Lorentz-
Lorentz equation can also be written as:

n 4 MM NaQey
ae 9) p 3E ee eeeee

The term on the left side is called molar refraction. In the case of non-polar
dielectrics, it is equal to molar polarization. Using the Clausius-Mosotti equation
throughout the visible and ultra violet range of electromagnetic spectra where in the
proper vibrational frequencies of electrons lie, the polarizabilities can be calculated for
an number elements and compounds. The table 5.1. gives the values of the square of
the refractive indices and the relative permittivity for a number of compounds. The
elemental solid dielectrics exhibit the electronic polarizability only.

5.9 Ionic Polarization :

In the case of substances with ionic bonding like NaC/, KC/, CaCO,, Rutile etc.
there appears lot of discrepancy between 1 and €, values. Therefore, it is obvious
that a different polarization mechanism is possible in these compounds apart from

210
Dielectric Properties of Solids

electronic polarization and the polarization is ionic polarization. For these compounds,
normally ¢, > 1%. Table 5.1. gives the static and optical dielectric constants for a few
compounds.
Table 5.1. : Static and Optical dielectric constants for a few compounds
- A comparison

ean
iia a eta
a 1.00027
1.00055
Carbon Tetrachloride
ee a
= |

Seca pe Cone EPPOE teeb re almai ome

In this case of substances with ionic lattices, the electronic polarizability of ions
is evaluated rather than of atoms. The sodium ion has a radius 9.9 x 107!!m, the ratio
of the polarizability of the ion to the cube of the radius does not come out to be 475,
but far less than that. The magnitude of polarizability of an ion when ionic displacement
takes place is about the same order as that of electric polarizability of an atom or ion.
Ionic substances have electronic polarizability but in addition, they have polarization
due to the mechanism of ionic displacement.
The processes of electronic and ionic polarization have common features. Both
are caused by deformation of charge distribution. The deformation is a displacement
of charges with respect to each other in the direction of the electric field. The state of
polarization is set in very fast in both cases. More or less, both are of the same
magnitude. The polarization is practically unaffected by variations in temperature of
the dielectric. Normally the process is reversible and does not suffer from dielectric
losses. Both are present in ionic dielectrics.

21)
 Applied Physics

5.10. Orientation Polarization:

The molecular solids like polymers and other polar dielectrics exhibit a tendency
towards dipole or orientation polarization. A molecular dipole as shown in Fig.5.8 tends
to orient (rotate) into the direction of the external field.

ime YVSth@© (a)

|
eh @ =<? 8 3/6)

oH ) SS
Q@ pe
o2 —
Fig.5.8 Orientation ofdipole molecule Fig. 5.9 Dipole orientation (a) In zero field
(b) In the presence of strong field

Debye, in his theory proposed that the orientation of the dipole is identical with
a sphere rotating in a viscous medium. Each molecule in the polar dielectric is a dipole
having a permanent dipole moment. The polar molecular dipoles are randomly distributed
in space in the absence of an electric field (Fig.5.9). The net dipole moment of the
dielectric is zero (here the contribution from the internal field from other dipoles is
neglected). But when the dielectric is kept under electric field, the field produces a
torque in individual dipoles and there is a tendency for the field to align them with the
field and a net dipole moment per unit volume is originated in the dielectric. If the field
is strong enough, the dipoles may completely be aligned along the field direction and
there will be a saturation polarization in the dielectric. In real dielectrics the polarization
is usually away from saturation values at normal electric fields. This tendency to orient
in an electric field is influenced by thermal motion. Debye proposed that the orientation
of the dipoles is identical with a sphere rotating in a viscous medium. The rotation
of
polar molecules under the action of an electric field is treated as the system of randomly
oriented dipoles going into an orderly arrangement of dipoles in the field. In pure
form,
it can appear only in gases, liquids and amorphous viscous bodies like polymers
and
glassy dielectrics. In many crystalline solids below their melting points, the dipoles
are
said to be frozen. Therefore, orientation polarization is ruled out.

212
Dielectric Properties of Solids

Langevin - Debye Equation:

It is possible to determine the dielectric susceptibility of polar substances
supposing that the molecular dipoles can rotate freely. Let a unit volume of dielectric be
considered and this volume contain n molecules. Each molecule is associated with a
permanent dipole moment p, represented by a vector (Fig.5.10). When a static field E
is applied to the dielectric, there is a tendency to orient the dipoles in the field direction.
The thermal energy works against this and tries to disorganise the orderly state. Consider
a sphere of radius r in the dielectric and a small surface dS on the sphere.

Fig.5.10
Torque on the dipole T = px E.
The potential energy Up of the dipole is given as

U, = [tae z [pEsine d0 = pE,Cos0

) 6

where E, is the applied field.

There are a large number of molecules in the dielectric and these molecules
tend to orient along the field. But the thermal energy tends to disturb it. As a result, the
molecules will be distributed with respect to the field direction in a statistical way.
According to Statistical mechanics, under the state of thermal equilibrium, the probability
of any molecule to be having a potential energy U, due to its position or orientation is
proportional to e~“’*" , the Boltzmann factor; k is the Boltzmann constant. The number
of dipoles per unit volume per unit solid angle making an angle 9 with the direction of
the electric field may be given as

dn(6)=K exp PESEO Joc ie (5.30)

213
 Applied Physics

where K is a constant of proportionality and k is the Boltzmann constant. The conical

solid angle dQ included between @ and @ + d@ is equal to

2nr Sin8 rd@

r
=27uSim Ord ytetu> arrailow eit bee heat (531)

The number of dipoles per unit volume lying in this conical solid angle is given
by the relation
n(0) dé = dn = keP EgCos0/kT Meso de saat (5.32)
The total number of molecules per unit volume over all the values of 0 i.e. from
0 to 7 is

aioe pE,,Cos0
Jan =n= isexp PEaCoS?Joni Ce canny (5.33)

_(pE,
Let
e ar

Then the total polarization (P) acquired is

P= |pCos8 dn = 2n KP exp(a Cos8)Sin@ CosOd0 (5.35)

0 0

This can be written for a molecule or dipole as

2m Kpn Jexp(a Cos 0)Sin@ Cos®@ d@

P= Oe 8 Beh Bioes Sees he eel eee (5.36)
2nK |exp(a Cos@)Sin@ dé
0

Let x = Cos0; i.e. dx = Sin®@ d@, then x changes from —1 to +1, then

+1

np [xexp(ax)dx
Pie +1
a Se ace Ene wie Soc ae ea (e578)
[explax)ax
-I

214
Dielectric Properties of Solids

ete 4 |
i.e., = | zt = np|Cotha 2 Sey (5.38)
e*—e

But np is the maximum possible dipole moment that the substance can have. It
corresponds to a saturation value (P,) and all the molecular dipoles are aligned parallel
to the applied electric field. This represents a state of complete saturation. The above
expression (5.38) is the Langevin-Debye expression for the mean effective polarization
of a polar dielectric in a field E, at temperature T .
Let np = Po, Then

ee ietP
=
= Cota
2pe=o oe (5.39)

]
Let Lowes y. = L(a) where L(a) is called Langevin function. L(a) can be

expressed as

aed: 228:
Maleate oat
If a—o,L(a)>1; a0, L(a)>0

For small values of a, the L(a) vs a plot has a slope of 1/3. (See Fig. 5.11). Here
L(a) curve is practically a straight line. This is true for field strengths of 108 V/m. The
dielectric strengths of many common dielectric are smaller than this value.

1
When L(a) = -,
3 P=n Pp5eee
Pa Ee Ble: (5.40)

The polarization is proportional to the field. The polarization depends inversely

on the temperature. The susceptibility

P np?
x
~ pE 3pkT
where is the P density of the dielectric, n represents the number of atoms per

: i Nap t i : , f
unit volume. Then n reas Where A is the atomic weight and N, is the Avogadro’s

number.

215
 Applied Physics

NEP AE
Liner
If P is of the order of 103° C-m, then P/kT is of the order of 10-%m / Volt.

Then a, , the orientation polarizability is

P. pe
oa; a.Bt
CS —a SKIES ae, oe ee eee (5.43)
5.43

The substance shall have a total polarizability,

p2 p2
AT otal = Ae +O; *3Kr 4 * OKT soraniaials (5.44)

where «, is called deformation polarizability and is equal to a1. + a,. Equation (5.43)
is known Langevin- Debye equation. It relates orientation polarizability to the
temperature. The dielectric constant of polar substances depends on temperature.

3\>o—>

Fig. 5.11

5.11 Behaviour of dielectric in alternating fields:

The permittivity of a dielectric depends on the variable external factors like
frequency of the voltage applied to the dielectric, temperature, humidity, pressure etc..,.
The frequency dependence of the dielectric constant is of a paramount importance in
technology. We will discuss in the following the frequency dependence of different
type of polarization and in the end will summarize the results.

216
Dielectric Properties of Solids

5.11.1 Frequency dependence of electronic polarizability:

An alternating field is one in which the electric field varies sinusoidally with
time. An electromagnetic wave can be made to be incident on a dielectric. As a result
of the time varying electric field, the component of the transverse electromagnetic
wave, the electron in the atom is forced to carry out motion. The electron is displaced
by an amount x relative to the nucleus of the atom and the system is then left to itself.
The restoring force which tends to drive the electron to the nucleus is given by

where x is the displacement and K' is the restoring force constant. In the absence of
the damping force, the equation of motion of the electron is written as

d°x
me =—K'xt+eE expGot) evens (5.46)

where the alternating electric field is expressed as E = E, exp(jot). This is the equation
of motion of a forced harmonic oscillator. Apart from the external field, the electron is
also under the action of a local field. Therefore the above equation can be written as

ds

m2 +K'K=eEc 2 tnnevn (5.47)

d’x K' e P
Ses Bey ape ||] ee
- dt? om ‘i <f 2.)

d°x € P
or Tai O% Tl ‘(es 2 ae? (5.48)

dhip fo _ Net aon

or dt 0 3e,m fa.) ee (5.49)

Where P = Nex. N represents the total number of electrons. The solution of the
equation is of the form

PAM
 Applied Physics

Xi Ky Sim (Ogtsind)) souehmegal pumps 4-.!(5.50)

cE.
Then RCT AT ET SRS LG a oY Gat)
|i re Ne |
3€)m

X, is the maximum displacement. Then the dipole moment p = €X9. The electronic
polarizability is given as

_eX_ Ne?/m os
YS Es ~ (od =Ne? /3e, —0) are (552)

@ is the natural or resonance .angular frequency. For an atom R x 107!%m,

m = 0.9 x 10°3'Kg. N = 10?°m-3, @) = 10!6 radians sec". a, is of the order of
10*°Fm?. The frequencies at which resonance takes place lie in the vacuum ultraviolet
range of electromagnetic spectrum.

i The above equations are based on the assumption that the electron is not acted
upon by a damping force. But in reality, the polarizability does not go to +oo Or—co at
@® = Wo. The electron is constantly accelerated or decelerated, it should radiate energy
“ The equation of motion for an electron should contain a frictional term characterized
by a dissipation constant y and y has the dimensions of frequency. Hence, the equation
of motion for the electron is

d?x ;
ogee +K x =eE,,, SouLeone (5,53)

d? dx
Fn OX = Ecc Sseannce (5.54)

The above equation is a second order differential equation and has a solution for
x which varies sinusoidally in sympathy with a sinusoidally driven field.
The above equation can be written as

d°x dx os x CS
+y—+
fen ?
Taking E = E, exp(—jot), the solution is

218
Dielectric Properties of Solids

f2 ve 2 3 /
Wo - — @~ — jyo
3€)m

Then, the polarization becomes

and the polarizability a, is given as

[22 Ne? /m
a. = > =e SS OS SS eee

: E 5) Ne? De.
Oo — AD =AVoO
3@jm

The relative permittivity ¢, of the medium is given as

els
macy E yk

a ne e° N
r Eom | e2 wees
0 wo — jy
3€ ym

e N
E a =

; Coulee Ne’ — © — jyo

3€ ym

3
4 Ne“

oe 1) (02 - 0? - jyo)- ( pe )e

Pasbe)
 Applied Physics

(Ga-1), > alee N

(#2) 32, m late
aa
7 ee (5.60)

The permittivity €, is a complex quantity ¢, (from equation 5.59)

. ' Ot) e’N

Eye, Oleg | ene ee
9) 2 Ne :
EgM] Wp— O° — — jyo

Simplifying, we get

eraealaae (Ne? /me, \oo30? )

Ea Oa eee
eS:
(0? - PayoBDae TT ae | ee ee (5.61)

a (Ne? /MEo yen

and bs 2 2 ee ae ere TO Oe eer
, +@ Stalen@)

wl
D) e°N
where O, = | {2 | ee (5.62)
0

The equations (5.60), (5.62) show the dispersion equations for solids and liquids.
The equations in the above indicate the frequency dependence of dielectric
constant.
The ratio of ¢’/e" represents the loss factor tan 8.

The polarizability is now a complex quantity. It can be expressed as

OniSio.w ties

Ne? Dap — 0Dash — jy@

Wo he
. mlo2 ~ 0 — jyo) : (02 — @ ~ jyo)

(Ne?smog - 0? _j_LNe?/mhro Ro (5.64)

(02 -w?} + yo (02 - wf +20
Dielectric Properties of Solids

The frequency dependence of the real and imaginary parts of a’, are shown
schematically in Fig. 5.12. The polarizability represents the dielectric constant. The
y- axis can as well be taken as dielectric constant.

Fig. 5.12 Schematic representation of variation of polarizability with frequency

The above curves represent also the variation of permittivity arising from electron
polarizability. The imaginary part vanishes at o = 0. However the real part (a', ) has a
2
value and this is the value of , under the static field (i.e. de) condition.
mas

The real part is positive for all values of @ < @, and negative for all values of
@ > Wy. The real part is zero at @ = Wo. The damping factor is normally very small and
the real part of polarizability does not change with frequency from zero frequency upto
a certain frequency close to @,.
Suppose ) — @) = Ao and Ao << @
@7 — 07 = (@ + @)(@)— ©) * 2 oAw
The real of polarizability is

, Ne? 20, A 2A@/Wo

ee in] Sse gee =Meier acess (5.65)

Then a’, is maximum when 4(Ao) = - 7

Therefore the damping factor is a measure of the separation between maximum
and minimum ofthe polarizability a.

pipe|
 Applied Physics

As ®@ > ,a% > 0 and around @ = @, it is

, _ Ne*y/ma,
OO. ~4(Ao) ee Pe ee eee (5.66)
oy +y°

The af exhibits maximum at @ = @,. This imaginary part ofpolarizability gives

rise to absorption of energy by the dielectric from the field.
The electronic dielectric constant gives the square of the refractive index. The
incident electric field is the radiation wave. The refractive index also will have a real
component and an imaginary component.

At ® ~ @o, the real part of polarizability or permittivity becomes negative. At

these frequencies, the radiation or the field will be reflected. The materials appear
opaque for these fields.

5.11.2 Ionic Polarizability :

For ion pairs of masses M, and M_, the equation of motion for electrically
induced forced oscillation is

bp kis sragin =eE loc: eens (5.67)

Mean ae de
where ©,, represents the natural resonant frequency of ionic pair, the above
equation is in the form as that for electron except for the mass terms. Therefore, ionic
polarizability will also have real and imaginary components and the expressions are
exactly similar to the above expressions for electronic polarizability. The peak position
of the imaginary component occurs in infrared region of the spectrum. The variation
of dielectric constant occurs in infrared region in ionic polarizability. Both the electronic
and ionic polarizability do not change with temperature. Therefore, the dielectric constant
resulting from them does not change.

5.11.3 Dielectric Relaxation and Dielectric Loss :

In the above section, the frequency dependence of the electronic and

ionic
polarization have been discussed. However the frequency dependence of the
orientational
polarization in liquids, polymers and glassy substances is of great importan
ce. Polymers
and glassy substances come under solids only.

Dielectric Loss :

Consider that a dielectric sample is placed between two paralel plates

that work
as electrodes. An oscillating field E is applied across the plates.
The field is described
as b= Ee e!' All dielectrics possess finite conductivity or resistivity values. A real

fae
Dielectric Properties of Solids

dielectric sample now shows the displacement field D with a phase difference with
respect to the E-field.
Then, the displacement field D is

DaseDpeiets?
i.€., ee? 6, tye,

The relative permittivity €, consists of two components: (1) a real component

e'_ and (2) an imaginary component e", and €* is a complex quality represented by
Sate Feel ai
then,

Die eee ie a eee

> €,6, Ee jepe ot
i.e., D,e!* oF = g(e' —je") Ener"

|Diew px soph
beeie= €, — Je,
Ey

eo
Do
E. cos 6

' Dy .
and € >be 8 6
0

and tand = ae
a T

The factor 'tan 5' is called the loss factor or the factor of the dielectric. It gives
the amount of energy that is disipated in the dielectric when an electric field is applied
across it. The dielectric powerloss is given as

WME gE’, V2 tan § watts / cycle, when the AC Voltage V is applied. As a result of
this loss, dielectrics absorb energy and get heated up. The dielectric loss is a funciton
of temperature. It increases with temperature in solids.
In polar dielectrics, there will be a resultant polarization to start with when
dipoles
voltage is applied across two electrodes between which a dielectric is placed the
on, there will be a
tend to orient along the field direction. Suppose the field is switched
on attains an
built up of charge or polarization in the dielectric with time. The polarizati

223
 Applied Physics

equilibrium value after a long time. Similarly when the field is switched off suddenly,
the polarization will not instantaneously become zero. This type of delay is due to the
fact that the dipoles take certain time for their rotation. Therefore, the polarization in
the dielectric is a function of time. In fact, a dipolar substance possesses two terms of
polarization as soon as the electric field is applied, the first term is the initial value of
polarization P), and a second term, the time dependent polarization P(t). The dependence
of polarization on time gives rise to relaxation. This sort of relaxation is responsible for
dielectric loss variation with frequency.
The total polarization of the dielectric is given as
US id ite ltl ce ee (5.68)
After sufficiently long time (t > 0 ), the polarization attains a value P(co ).

Let ot OS eEee wo kee (5.69)

Pi(t) tends to the equilibrium value or steady state value asymptotically. This
tendency towards the equilibrium value after the application of the field or the lack of
instantaneous response of polarization to the external field is known as relaxation.
Fig.5.13 illustrates the decay. of the orientational polarization and growth of the
polarization with time.

Fig. 5.13 The decay (a) and the growth (b) of the polarization with
time.
“ The rate at which P,(t) changes is given as

AP,
dt
_ =(Po+Pi)
Po A _ Pro ———P, (5.70)
dP, loo 1 dt

* Students may skip the mathematical derivations, if it is not

in syllabus.

224
Dielectric Properties of Solids

BE seri

dP, ‘dt
ntegrating
Integrati puSe
a o

In Pichotis
B. ot or (P,,, =P) Es Peraie

P= Pee) ies 6 og aes (5.71)*

The total polarization P is

P(t) = Py + P(t) =fep%p + X89 — PIE (5.72)

When alternating voltage is applied to the two surfaces of the dielectrics, the
polarization can not follow the field variation due to the inertia of polarization. If @ is
the angular frequency of the alternating electric field E, the electric field is given as
E = E, exp(jot). P(t) also oscillates with the field.
Py P,(0)e!!
and Pee XjEpEge*

X1E0Eo
(0) Toray
P(0)=———— ie tadtayhbal fieHes Seaee (5.73)

The orientation polarization P,(t) oscillates with an amplitude

XiE0
P [i0 fs) KACEY oh law AN iedeateoe(5.74)

€
where ea nate=P Ls )
1+ Jot

* X1&0 “alA bit beak

or
€.—-1l=€g%o
r oXo + 14022 Tae safes le's\ste (iid)

" Xie
e, = Xo0Fo i .2
1+@°t

229
 Applied Physics

(aa X1EqOt
and Samat
Dobe

The loss factor is given by the relation ¢”/¢; = tan 5. This represents the
energy dirsipation in the dielectric.
The real and imaginary parts of the dielectric constants are plotted in Fig 5.14.

OR Wc

Fig. 5.14 The frequency dependence of the dielectric constant. Orientation polarization is taken
into consideration.

Frequency Dependence of Dielectric Constant Over a Wide F requency Range:

We have seen how individual polarization mechanisms give rise to frequency

dependence of polarizability. It reflects directly the frequency dependence of dielectric
constant of a solid.

The effects of frequency are especially important because dielectrics are often
used in alternating fields, The polarization depends on the ability of the fundamental
particles to move relative to each other. At high field frequencies, the more massive
molecular particles can not move rapidly enough to follow the direction of the electric
field. Therefore, at high frequencies, dipoles can not rotate rapidly enough to contribute
to polarization. They remain, on the average, at an angle 90° to the field direction.
The
electrons are capable of relaxing rapidly, the electron cloud distortions follow the field
direction more easily at high frequencies. The Fig. 5.15 schematically shows
how the
polarizability might vary with frequency. At zero frequency, the polarization is
static. At
radio wave frequencies, the effective polarization starts to decrease. At
infrared
frequencies the ionic polarization drops off. This then leaves just the electronic
contribution at the optical frequencies.

226
Dielectric Properties of Solids

UHF to
a - space charge Micro Optical
wave

a - dipolar

@ - ionic

a - electronic

10° 102 104 10° 108 10! = go!2—gl# = gle gg

10? 10° 10" 10”

Frequency
Fig. 5.15 The frequency dependence of polarizability.

Temperature Dependence of Dielectric constant. The electronic polarizability

and ionic polarizability do not change with temperature. However polar polarizability
depends on temperature. Normally, the dielectric constant in linear dielectrics decreases
with temperature. The more interesting part comes from the loss variation with
temperature. The dielectric loss in creases with temperature. This is because, the
dielectrics become conducting as temperature is increased. The conductivity of a
dielectric material increases with temperature. This produces an absorption of energy
i.e., the dissipation. Therefore the loss will increase with temperature. The dielectric
relaxation time will decrease with temperature. In non linear dielectrics, the loss may
exhibit a maximum at a certain temperature; A Ferroelectric is one such example. The
defects in the crystals also affect the dielectric constant and dielectric loss variation
with temperature. The defect complexes (like ion-vacancy complexes, or vacancy-
impurity associations) play a vital role in crystals in relaxation phenomena, dielectric
loss and dielectric variation with temperature.

5.12 Piezoelectricity and Ferroelectricity :

5.12.1 Introduction :

The ‘linear’ dielectrics are characterized by the fact that their polarization p and
displacement D are directly proportional to the intensity of electric field E. The relations
are given as P = x) E and D = €,€, E. Thus, in such substances the dielectric constant
¢, and susceptibility x do not depend on the intensity of electric field. If such a dielectric
is used in a capacitor, the capacitance C is independent of the voltage applied to it and

227
 Applied Physics

the charge of the capacitor is directly proportional to the voltage applied to it. Such
capacitors are used as passive elements in electrical circuits. At very high voltages, the
charge of the capacitor may deviate from the linear dependence. However, there are
‘non-linear’ dielectrics which when used in a capacitor give rise to non-linear dependence
of capacitance on the intensity of electric field. The ‘non-linear’ dielectrics are useful
in active devices in various electrical engineering and electronic devices. Ferroelectrics,
piezoelectrics and pyroelectrics fall under this category of non-linear dielectrics. They
possess special characteristics which will be discussed in the following sections.

5.12.2 Piezoelectricity :

In some ionic crystals, the center of positive charge in the unit cell of the crystal
does not coincide with the center of the negative charge. There is a net polarization
associated with the unit cell of a crystal. Pyroelectricity, Piezoelectricity and
Ferroelectricity are the properties for these solids. The first experimental demonstration
of piezoelectricity was done by Jacques and Pierre Curie in 1880 in tourmaline, quartz
and Rochelle salt. There is one important property that a material must have in order to
be piezoelectric or ferroelectric. The unit cell of the crystal exhibiting the piezoelectric
or ferroelectric property should have a structure lacking the center of inversion. Each
unit cell of piezoelectrics develops an electric polarization when subjected to elastic
strain or deformation. In the absence of stress, the free piezoelectric crystal has surface,
charge that attracts free charges from atmosphere making it neutral. The deformation
disrupts the orientation of the dipoles and the acquired charge is not completely cancelled.
This charge can then be measured as a potential difference across the material sample.

Solid substances except wood, polymers or glasses are basically crystalline in

nature. A crystal is defined as a solid which possesses a regular arrangement of atoms
or ions in three-dimensional patterns. This arrangement of atoms in regular manner
satisfying certain chemical bonding conditions and minimum energy considerations is
known as crystalline structure. The internal regularity of arrangement of atoms in
solids automatically leads to symmetry of their unit cells. Then the repetitive arrangement
of unit cells leads to external shapes ofthe crystals. The unit cell is the smallest volume
of the crystal which characterizes the arrangement and positions of atoms. It is a
building block of the crystal. The unit cells of crystals are classified into seven systems,
which are once again divided into fourteen bravais lattices. The unit cell of a crystal is
the basic unit satisfying the chemical bonding and stoichiometry chemically. The unit
cell possesses a number of symmetry elements. For example, a simple cube possesses
mirror planes passing through opposite faces bisecting them. There are axes through

228
Dielectric Properties of Solids

the centres of faces and perpendicular to them. The unit cell can have a center about
which it is symmetrical. But some crystal classes do not have this center of symmetry.
The crystals can be classified into 32 point groups according to their crystallographic
symmetry . These point groups are divided into two large classes, one with a center of
symmetry and the other without. There are 21 point groups, which do not have center
of symmetry . In crystals belonging to 20 of these point groups, positive and negative
charges appear on the surfaces when stresses are applied to the crystal giving rise to
direct piezoelectricity . Later it was predicted by Lippman in 1881 that these solids
become stressed or strained when subjected to an external electric field in certain
directions. Curie verified this converse or reciprocal effect. The strain produced as a
result of the electric field is due to the electromechanical coupling in the crystal. Consider
a row of ions in an ionic crystal as shown in Fig. 5.16. It shows a one dimensional rigid
ion spring model of the crystal lattice. The springs represent equivalently the cohesive
force resulting from the electrostatic Coulomb energy and the quantum mechanical
repulsive energy. In a noncentro symmetric lattice, the springs joining the ions are
different i.e. harder or softer. When an electric field is applied, the cations are drawn in
the direction of the electric field and the anions in the opposite direction leading to the
relative change in the interionic distance. Depending on the direction of the electric
field, the soft spring expands or contracts more than the contraction or expansion of
the hard spring causing a strain 's' (i.e. change in.length of the chain). This strain is
proportional to the electric field applied.

S=dxE

ee
to —_—> a —P
OO HH000 e700 © 0000 +e
|j<—_—_
a ——_ >|

Fig. 5.16 Converse piezoelectric effect

In ordinary solids a stress T causes a proportional strain 's' related by an elastic

modulus; T= Ys. Piezoelectricity is the additional creation of the electric charge by
an applied stress. This is the direct piezoelectric effect. In terms of the dielectric
displacement D, (charge Q per unit area) and stress T, we may write D = Q/A= dT,

229
 Applied Physics

where d is expressed in Coulombs/Newton. ‘d’ is called piezoelectric constant. The

value of 'd', should be high for applications. In the converse effect, an electric field E
produces a proportional strain S, (expansion or contraction) where S = d E (d is
expressed in meters/volt). Joffee defines the piezoelectric constants as partial derivatives
evaluated at constant stress T or a constant electric field E given as

ae =) . =)
Obie a Oly s

In crystals, stress and strain are scalar and tensoric. Therefore the piezoelectric
coefficients are of tensoric form.
Piezoelectricity and Ferroelectricity are exhibited by a few classes of crystals.
Out of the 20 classes of crystals which can show piezoelectricity, ten crystal classes
have a unique polar axis in the unstrained condition. These crystals can develop an
electric charge when uniformly heated. There will be a change in the magnitude of the
dipole moment of the crystal with the change in the temperature. Such crystals are
called pyroelectrics. When heated, they give rise to a pyrocurrent under short circuit
condition. All pyroelectrics are piezoelectric but all piezoelectrics are not pyroelectric.
For example, quartz is a nonpolar piezoelectric crystal. But it is not ferroelectric or
pyroelectric. Such crystals do not contain a net crystal dipole moment. But a crystal
like tourmaline has a net internal dipole moment (polarization). Its polarization is as
high as 80 x 107!°C/m? and gives rise to pyroelectricity .
Piezoelectricity:

Piezoelectricity has already been defined above as the additional creation of an

electric charge by an applied stress. In terms of D the electric displacement, D = d x T
(d is expressed in Coulombs/Newton). An applied field E to a crystal produces a
proportional stress S, the expansion or contraction depends on the polarity. S = d x E.
'd' is a tensor and expressed as d (Cosy Obs), l= SZ, keXY, ZY, ZX, XX, V¥, Zz)
(d is expressed in meters/volt). As already stated, piezoelectricity is a property which is
possible only for ionic solids with crystalline structures lacking a center of inversion.
Consider a molecule in a hypothetical ionic solid which in equilibrium gives three electric
dipoles of equal magnitude at 120° intervals, as shown in Fig. 5.17. This molecule has
a point group 3m. The net dipole moment of the molecule vanishes by symmetry . But
a net dipole moment will appear if the molecule is stretched or compressed along a
direction parallel to, or perpendicular to, one of the three vertices. Similarly, an electric
field applied parallel to one of the three vertices will enforce a distortion of each molecule.
This results in a contraction or elongation.

230
Dielectric Properties of Solids

Stress

fe P=0
Stress | .

@
Stress

Fig. 5.17: A piezoelectric molecule under strain (a) molecule in equilibrium (b) Net polarization
developed in vertical tension or horizontal compression (c) net polarization developed by vertical
compression or horizontal tension of the crystal parallel to the field direction, accompanied by a
length change of opposite sign in the perpendicular direction. But, an electric field applied
perpendicular to the direction of a vertex will not change the crystal dimensions, because the field is
perpendicular to a reflection plane of symmetry.

Most of commercial gadgets based on ferroelectricity and piezoelectricity are in

polycrystalline form. Quartz, Rochelle salt and KDP are the substances which are used
in single crystal form. Some of the piezoelectric are mentioned in Table 5.2.
Table 5.2 : Piezoelectric coefficient of a few compounds

Ba TiO,
Single Crystal 85-120
Ceramic 191
PbTiO3
Single Cryastal Lin
Ceramic 60

LINGO3
PLZT
nna
(plea Ze TiN,
PLZT means La-modified

Pb(Zr, Ti)O, Ceramics

231
 Applied Physics

Energy can be given to a piezoelectric body either mechanically by stressing it or

electrically by charging it. All the energy given to it is not converted into the effect.
Therefore the piezoelectrics are characterized by a strength of piezoelectric effect.
This strength is measured by electromechanical coupling factor K. When an electric
field is applied to the piezoelectric crystal or ceramic, it measures the fraction of the
electrical energy converted to mechanical energy (or vice versa when a crystal is
stressed). The actual relationship is given as

ae electrical energy converted to mechanical energy

Input electrical energy

K2 mechanical energy converted to electric energy

or eee ee
Input mechanical energy

K is also a tensor. The coupling factors are constants for piezoelectric crystals
and ceramics. Normally, the piezoelectric or ferroelectric materials are used in ceramic
form i.e. in polycrystalline form. They are shaped in the form of discs, rectangular
blocks or rings. A polycrystalline disc consists of small crystallites or grains. Each
crystallite has a polarization (P,) oriented in a particular direction. The polarization
vectors are randomly oriented in the sample as shown in the Fig. 5.18. The summation
of the polarization in the entire sample comes out to be a small value or zero compared
to that of a single crystal. Such a sample will not show the proper electromechanical
coupling or ferroelectric parameters.

Unpoled ceramic Poled ceramic

Fig. 5.18
To orient the polarization vectors along the same direction, the disc is placed
between two electrodes and a strong DC electric field (a few KV/m to MV/m) is
applied to the sample. Then the polarization vectors in the grains tend to orient along
the field direction and such a sample is called electrically poled ceramic sample.. These
poled ceramics exhibit the piezoelectric or ferroelectric effects nearly as single crystals

2o2
Dielectric Properties of Solids
a
SR

even though they can not replace a single crystal. Their physical properties related to
piezoelectricity and ferroelectricity depend on the amount of poling. The electrical
poling of a ceramic is analogous to the magnetizing of a permanent magnet.
5.13 Ferroelectricity:
Ferroelectric crystals are those crystals which exhibit spontaneous polarization
i.e. electric polarization without any external electric field. Each unit cell ofa ferroelectric
crystal carries a reversible electric dipole moment, spontaneously oriented parallel to
the dipoles of neighbouring cells. Ferroelectricity is defined as the reversibility, in a
polar crystal, of the direction of the electric dipole by means of an applied electric field.
Since it requires polarity in a
crystal, it can occur only in
pyroelectric crystal classes. But © m™?
all pyroelectrics are not
ferroelectrics. For example |
tourmaline and hexagonal CdS are
Ps
pyroelectrics, but not ferroelectric.
All ferroelectrics are pyroelectric.
The spontaneous polarization is a
function of temperature. The ‘om T ——> Te
spontaneous polarization decreases
with increasing temperature, and
Fig. 5.19 Spontaneous polarization varies with temperature
at a certain temperature the
spontaneous polarization just disappears as shown in Fig. 5.19.
The ferroelectric crystals possess high dielectric constants. They also show
dielectric anomalies i.e., the dielectric constant reaches a maximum value at the transition
temperature called the Curie point (T,). The crystal is ferroelectric below the Curie
point and above the transition temperature the crystal becomes paraelectric in nature.
In paraelectric phase, the dipole moments in the crystal are randomly oriented. Above
the transition temperature T,’ the dielectric constant (€,) changes as per the
Curie-Weiss law expressed as
Cc
GT Pe rier acme ae ae a ee (5,71)
: ages
where C is called the Curie constant. T is the temperature and T, the Curie point. C is
of the order of 103 — 10°K.
It has been pointed out earlier, that Mosotti expression for the internal field ina
dielectric could be applied only to non-dipolar substances. But it is interesting to apply
Mosotti expression for a dipolar substance. The polarizability and polarization are related
as
B= Nig. E loc

233
 Applied Physics

where E,,, = E + P/3€, and the terms involved have the usual meanings.

eae
oo ,_ No ener (5.79)
3€9

From the above expression, it is seen that p tends to infinity when (Na/3€,)
tends to one. As the substance is dipolar, the polarizability is expressed as
O10: heenivenet) sex quimenisttion to:2slc< (5.80)
Suppose the displacement between positive charge and negative charge is x,
then the dipole moment is
Box loc = Rigg ag el Pate ee Cera (5.81)
where F is the restoring force trying to bring the two charges of the dipole together.
The work done in creating N such dipoles is given as

W, =N [rax = [Me[xx

the energy of electric displacement is

2
W, = [Evcdp (= [(E+p/eq)dp Sla JEap (5.83)
Then the net energy associated with a polarized dielectric is

p° Na
ww, = [2 a -1] 4 JE ap ae (5.84)
Hence the net energy is positive even for zero external field. This energy is
stored in the crystal. It is energetically favourable to create highly polarizable dipoles in
an orderly fashion. At sufficiently high temperatures, this orderly arrangement is
destroyed. Thus, there exists a critical temperature above which spontaneous polarization
disappears in the crystal, but still the crystal may possess a high dielectric constant.
For example (Ba, Sr) TiO, ceramic used as capacitor material possesses a dielectric
constant of about 6000 above the transition temperature.
The general requirement for the establishment of a ferroelectric state lies in the
fact that an ordered array of induced ionic dipoles must be able to have a smaller total
energy than the crystal with zero dipole moment. This requires the presence of a very
large number of extremely polarizable ions in the crystal. Above the critical temperature,

234
Dielectric Properties of Solids

called Curie point T., the amplitude of thermal vibrations of atoms is strong enough to
prevent the ordered dipole array to exist. A solid which is ferroelectric below T,, is
paraelectric above T, with susceptibility varying as (T — ayes
We will now see some of the important characteristics of ferroelectrics :

1) As already stated, Ferroelectrics exhibit spontaneous polarization. Every

ferroelectric substance has a characteristic Curie point (T,) which can be
influenced by impurities. The transition from ferroelectric to paraelectric state
may be accompanied by change in crystal structure (example barium titanate) or
change in space group of the crystal structure.
2) They possess very high dielectric constants. The dielectric constant is large
even in paraelectric phase.
3) The spontaneous polarization becomes zero at the transition temperature.
4) All ferroelectrics exhibit polarization reversal. Each unit cell of a ferroelectric
crystal carries a reversible dipole moment, spontaneously oriented parallel to the
dipoles of the neighbouring cells. The commonly accepted criteria for
ferroelectricity is a hysteresis loop on D-E studies (Fig.5.21) The loops can be
square in some cases. Above the transition temperature, the hysteresis loop
disappears and the crystal behaves like an ordinary dielectric.

Ceramic

Single crystal

Fig. 5.20 D—E or P—E graph for a ferroelectric

The P-E curves suggest that there is a net remnant polarization P , The spontaneous
polarization P, is obtained’ by extrapolating the polarization at high fields back to
zero field along a tangent. P,,, itself is a function of the peak field. P, and P, are
nearly equal in single crystals.

5) All ferroelectrics are pyroelectric and piezoelectric but the reverse may not be
true. That is, all pyroelectrics are not ferroelectrics, and all piezoelectrics are not
ferroelectrics. For example tourmaline is pyroelectric and not ferroelectric. Quartz
is peizoelectric but it is not ferroelectric. Ferroelectrics show reversible

Jas)
 Applied Physics

piezoelectric and pyroelectric effects. The observation of hysteresis loop in D-E

studies indicates the presence of domains, called ferroelectric domains. Each
domain is spontaneously polarized.
6) Ferroelectric crystal contains an array of domains with different directions of
polarization. Normally, neighbouring domains in which the direction of Pots
different will surround a domain. When an external field is applied, the domain
wall may move to permit an enlargement of domains in which the direction of p
nearly becomes parallel to the applied field. At the saturation, all domains have
the polarization vectors in the direction ofleast energy i.e., in the direction of the
applied field.
7) The ferroelectric to paraelectric transition is marked by a dielectric peak in the
dielectric constant vs. temperature plots (Fig. 5.21) for a ferroelectric crystal.

Te

T —>
Fig. 5.21 Dielectric constant vs temperature plot for a ferroelectric like BaTiO3.

8) Ferroelectric crystals exhibit birefringence, i.e. double refraction when

a plane
polarized light is passed through them.
9) Both the specific heat and thermal expansion co-efficient exhibit a discontin
uity
at the transition temperature.

5.14 Ferroelectric Materials :

There are a very large number ofcrystals which exhibit ferroele

ctricity.
The ferroelectricity was first discovered in Rochelle Salt
(Na K C,H,0,,
4H,O). It exhibits spontaneous polarization over a range
of temperature
(-18°C to 20°C). The variation of spontaneous polarization is shown
in
Fig.5.22. The figure 5.22 (a) displays the variation of Spontan
eous polarization
in Rochelle salt. The cutting points on temperature axis
give the range of

236
Dielectric Properties of Solids

temperature in which the Rochelle salt is ferroelectric. The deuterated samples

show an increased range of temperature (curve b). The deuterated samples
also show the increased spontaneous polarization.

along
Dielectric
constant
axis
x-

-160 -140-120-100 -60 -40 -20 0 40 60

Temperature in degrees C°
in
polarisation
micorcoulombs/cm?
Spontaneous
Fig. 5.22 The variation of spontaneous polarization and dielectric constant in Rochelle salt.

Ferroelectric crystals are classified into order disorder type and displacive type.
The order-disorder class of crystals includes crystals with hydrogen bonds in which
the motion of protons is related to the ferroelectric properties as in potassium dihydrogen
phosphate (KDP or KH, PO,) and its isomorphous salts. The ferroelectricity is associated
with individual ordering ofions in the crystal lattice. KDP has a transition temperature
of 123K. But the replacement of hydrogen by deuterium increases the transition
temperature drastically. (Table 5.3 ).
Table 5.3. List of a few important ferroelectricals with their transition
temperatures

Substance with chemical formula

Pottassium dihydrogen
Phosphate KH,PO,(KDP)
 Applied Physics

Barium Titanate :

The most important class of ferroelectrics is from displacive type of ferroelectrics.

They possess perovskite crystal structure. Barium titanate is the most important
ferroelectric in this class of crystals. It was discovered around 1943. The crystal
structure of barium titanate is shown in Fig.5.23 (a).c axis of the crystal is along the
arrows shown in Fig. 5.23(b). In ferroelectric phase above room temperature, the
crystal structure has tetragonal symmetry. The prototype crystal is calcium titanate
(perovskite ). The structure is cubic (figure (a)) at high temperatures. The Ba2* ions
are at the cube comers, O?- ions are at the face centers and Ti‘ ion is at the body
center. Below curie temperature i.e. 120°C the structure is deformed. The Ba2+ and

(b) PaSal.s
he = Curie temperature

Fig. 5.23 The Barium titanate structure

Ti** ions are displaced relative to the O2- ions, thereby develop the unit
cell a dipole
moment. The upper and lower oxygen ions (Fig. 5.23) may move downwar
d slightly.
The volume of the cells is 64 x. 10-39 m3, so that dipole moment of the
cell is
2 x 10°?°C.m. The crystal structure below transition temperature
is tetragonal, it is
actually a tetragonal distortion in which the unit cell elongates by
about 10% along one
pseudo-cubic direction (C-axis) and contracts by about 0.5% along
each of the axes
perpendicular to this direction. The sub lattice containing Ba2*+ and
Ti‘* ions shifts
upwards along the C-axis with respect to the Oxygen sub
lattice. The relative
displacement of the two sub lattices is 0.01 nm. This is responsible
for the spontaneous
polarization. The Ti** ion is present at the center of an octahedr
al formed by oxygen
ions. Barium titanate also exhibits two more transitions one at
0°C and another at —
80°C. They are shown in the Fig. 5.24.

238
Dielectric Properties of Solids

20 « 10-7
=a

aa (tLe tee
< Rhombohedra!| Monoclinic Telragonal y

Phebe Dc
(Coulombs/m?)
P
a OS Be es KT. aealed
-100-120-80;-40 0 "40 "80|)
t
-200-169-120 -80 -40 0 40 80 120 '\erew
Temperature(*C)

Fig. 5.24 Spontaneous polarization and dielectric constant vs temperature in Barium Titanate

These are a number of other ferroelectrics similar to BaTiO, with perovskite

structure and these are mentioned with their transition temperatures in Table 5.3. They
have general formula ABO, (A = divalent, B = Tetravalent). The most important
ferroelectrics, other than barium titanate, are lead titanate, lithium niobate, potassium
niobate, etc. But, there are hundreds of ferroelectrics today which have been synthesized,
but it is not possible to give a list of all those ferroelectrics here. The ferroelectrics with
potassium tungsten bronze structure are also important from commercial point of
view. They have the general formula A,B,O,, where A and B are two types of positive
ions. Normally A is a divalent ion and Biis a pentavalent ion (Nb>* or Ta**). PbNb,O,
belongs to this category . These crystals also possess oxygen octahedra.

5.15 Piezoelectric Materials:

All the commercial piezoelectrics are basically ferroelectric except a few like
quartz (SiO). Quartz has piezoelectric coefficient 2.2pC/N, along the piezoelectric
axis, which is low, but it has a good mechanical and thermal strength. It is highly
useful in the generation of ultrasonic waves. When an alternating electric field is applied
to a piezoelectric element or plate, it is compressed and extended thereby vibrating.
These vibrations produce ultrasonic waves. The quartz oscillators are useful in radio
transmitter, quartz clocks and quartz wrist watches.
Piezoelectric effect is known for approximately 1500 materials. Choosing the
suitable material for required application is a difficult problem ofmaterials engineering.
Piezoelectric ceramics are widely used. But ceramics are fragile, difficult to fabricate
in different shapes and possess a high acoustic impedance. Therefore piezoelectric
ceramic polymers or piezoelectric- polymer composites are used. They can be shaped
easily. A piezoelectric ceramic is characterized by an electromechanical coupling factor
(K,,) and piezoelectric coefficient d,,.

2o9
 Applied Physics

Barium titanate was the first ceramic developed and is widely used. It has high
coupling factors (49%), piezoelectric coefficient (19.7 x 10-!*C/m). It is highly stable
chemically. Lead titanate (PbTiO,) has polar axis along 6 directions and it shows a
spontaneous polarization of nearly 80uC/cm?. Pure lead titanate cracks easily under
the action of the field. Stabilization can be obtained by the addition of impurities like
La**. When properly poled, the ceramic gives a d,, of 40 x 10-!°C/N. Lead
zirconate-lead titanate solid solutions are widely used as they show strong piezoelectric
effects, especially those with around 0.48 lead zirconate. The ceramic when poled
properly has d,, more than 230pC/m and K,, about 0.67—0.75. They have low loss
and high coercive field. The solid solution is also modified by the addition of La**, Si?*
or Fe**. They are widely used commercial ceramics.
Piezoelectricity is also found in polymers, for example polyvinylidine fluoride
(PVDP) polymer possesses ferroelectricity but small d constant. The advantages are
lightweight, high elasticity and lead free. They are compatible with human body and
water. They are useful as sensors and actuators. They have low quality factor but a
wide frequency band resonance. In commercial applications, polymer- piezoelectric
composites are used.

5.16 Applications:
All commercial piezoelectric materials used today are ferroelectrics. Then the
applications can also be treated as the applications for ferroelectrics. Piezoelectric
bodies provide a coupling between electrical and mechanical forces and hence serve as
traducers that convert or detect electrical or mechanical signals. Hence they are highly
useful in converting very small mechanical energies into electrical fields: Hence, they
find utility in producing very small mechanical displacements and small amounts of
electric charge per cycle. Several applications are available for piezoelectrics and we
will discuss a few of them.
High frequency electric oscillations can be transformed into mechanical oscillations
in the production of ultrasonic waves. Converse piezoelectric effect is used here.
Quartz is a hexagonal crystal. The symmetry considerations will give three fold polar
axes in this crystal. When the crystal is placed in an oscillating electric field, the direction
ofthe field being along one ofthe polar axes, the crystal will oscillate. The oscillations
are due to changes in thickness ( contraction and expansion). Because of the alternating
Strains, oscillations are produced in the medium to which the crystal makes contact.
n
The frequency f = ie Pe where n= i1,2,3.Ge.es The value of Y/p depends on the

polar axis chosen. The frequency depends on the thickness of the plate. Frequencies
as high as 50 MHz can be achieved by proper selection of thickness. Today we have
piezo electrics which can give much higher frequencies.

240
Dielectric Properties of Solids

1. Phonograph pick-ups:
Prior to 1950, Rochelle salt was used but the ceramics are much superior in
their mechanical and chemical stability . A typical cerafnic phonographs pick up elements
is shown in Fig. 5.25.
Series connected

Force
2 Sule gar
x
xX

Parallel connected

Fig. 5.25 Phonograph pick up. The arrow on the element indicates the polarization direction.

The ceramics plates are cemented on brass vane and are electrically poled in
opposite directions. When the pin attached at the end falls in a groove, there is
compression in one plate and extension in another plate of piezoelectric ceramic.
Therefore, an electric field is generated. Both the ceramic plates give the electric field
in the same direction. PZT modified by impurities is largely used here nowadays.
Another important application is their use as vibrators. When an alternating voltage is
applied across a piezoelectric element, it vibrates and at a particular frequency of the
field, it resonates. The resonant frequency
of the element depends on the thickness.
The mechanical vibrations can be
transferred into solids or liquids as ultrasonic
waves. This frequency range lies from 1000
Hz to 10OMHz. Hard type piezoelectric
ceramics are used. The speakers and
buzzers utilize the piezoelectric vibrations.
One such vibrator is given in Fig. 5.26. Fig. 5.26 Piezoelectric vibrator

1) Elastic membrane
2) Piezoelectric element
3) Electrodes

Piezoelectric ceramics come into their own in generation and reception of sound
waves in water. These ceramics in this sense are called smart materials because they
are actuators as well as detectors. They are useful as ultrasonic cleaning traducers,
(frequency range 20-100KHz), under water detecters of sounds. Under water

241
 Applied Physics

applications comprise of the passive listening (hydrophone) as well as active method

of producing sound pulses and timing their echo from obstacles. This is useful in fish
finding and in sonar. These devices operate at high power levels for maximum range.
They are also useful for submarine detection. Ultrasonic sources and detectors are
extremely useful in factories. Ultrasonic detectors are useful in medical diagnosis.
Piezoelectrics are broadly utilised to induce ultrasonic oscillations. The ultra sound is
used for fault detection i.e. for finding internal cracks and other hidden defects in solid
bodies like bars, rods, plates etc. The simplest application of piezoelectrics one finds at
domestic level is the gas igniter which is shown schematically in Fig. 5.27.

18

out put voltage

=
2
a Electrode
wn
S Voltage
&
°
io

40 60
Time (us)

Fig. 5.27 Piezoelectric igniter. The arrow on piezoelectric element indicates

the polarization
direction.
In the gas igniter, a very high voltage is generated spontaneously. across
a
piezoelectric ceramic element when a strong mechanical stress pulse
is given to the
element. This voltage
gives rise to a spark.
Mechanical
energy is translated
into electric signal. 43
Th Hanieal Original
€ mechanica polling Seismic
Mass
energy is generated direction
by mass and gives a
force F = ma. The
Fig. 5.28 gives a
schematic diagram of
an accelerometer.
Fig. 5.28 A schematic diagram of Accelerometer

242
Dielectric Properties of Solids

The piezoelectric element in tubular shape A is kept around a mount M. Around

the element, a dense metal with a mass m is kept. The whole assembly is encapsulated
in a protective case. As a response to acceleration, the mass imparts a force to the
element which is in turn generates a voltage proportional to the stress. The voltage is
calibrated against acceleration.
The beautiful aspect of piezoelectrics is that they can be used as sources as well
as detectors. The vibrations produce alternating electric field. The alternating electric
fields produce vibrations in them. Here both piezoelectric effect and converse piezoelectric
effect are used. Such materials which have this property of actuators (Producers), as
well as detectors are called smart materials.
Today we require very fine displacement elements in fields of optics, precision
machinery, small motors, and micro electronics. Positioning of various parts of
machinery in micromechanics is needed. Piezoelectrics are developed into micrometer
displacement gadgets, called ceramic actuators which operate on the principle of electric
field induced strain. The strain produced because of the field should be a linear function
of the field (Fig. 5.29). Lead zirconate titanate is primary material for many piezoelectric
applications.

Strain
-——>

(-) 0 Electric field (+) A

m

Fig. 5.29 Field induced strain

Dot matrix printer heads have piezoelectric displacement traducers. The

advantages are high printing speeds, low energy consumption and noiseless printing.
Ink jet printer heads are also based on displacement piezoelectric elements. The
piezoelectrics are extremely useful to-day for the development of microrobots, mini-
monitor positioners, speed arresters etc.

Some of the applications of ferroelectrics are: The ferroelectrics also exhibit

pyroelectric phenomena. pyroelectric effect is the appearance of electric charges on
the surface of dielectrics mainly crystal ones, when they are uniformly heated or
cooled at the rate of 3-5°C/minute. The uniform heating or cooling of the pyroelectric
is essential for a pyroelectric effect to appear. The examples of useful pyroelectrics are

243
 Applied Physics
a

triglicine sulphate (TGS), triglicine fluoro bery! late, triglicine selenate, barium tantalate,
PZT in ceramic form, strontium barium niobate etc.

Pyroelectrics are useful in the detection of infrared radiation. Their sensitivity to

infrared radiation makes them useful in thermal detection, burglars alarm, and infrared
photography. The pyroelectric detectors are useful in power meters for laser radiation.
The ferroelectrics which exhibit a square loop in polarization vs electric field
characteristic are useful in memory technology. The Ferroelectric RAMS (FRAM) are
developed. Barium-strontium titanate, strontium bismuth titanate are useful here. The
high retention and capacity, endurance, access time and stability are the key factors in
utilising square loop ferroelectrics in RAMS.
There are ferroelectrics which exhibit diffuse phase transition i.e., their phase
transition temperature is not sharp. The transition temperature shifts to higher temperature
with the frequency. This behaviour is called relaxor behaviour. Such ferroelectrics in
thin film form are useful in multi layer capacitors. Their size is very small. They are
useful in microelectronic circuits.

5.17 Dielectrics as Insulators and Capacitors :

The application of an electric field gives rise to three phenomenon in dielectrics.
They are (1) electrical conduction, (2) dielectric polarization, and (3) dielectric loss.
When a voltage is applied to a dielectric, the
current through the dielectric increases \
nonlinearly (Fig. 5.30). If the dielectric is
under the action of alternating voltage, the — B
capacitive current increase with the
amplitude of the applied voltage. At a ee
particular voltage the electrical insulation is
damaged and this damage is known as
electrical break down of the dielectric. The Ve V
(Fig. 5.30) schematically represents the
current versus the voltage characteristics of
Fig. 5.30
the dielectric. At point B in the graph. dl /dV -—> a and it corresponds to the moment
of the break down. Dielectric break down is defined as the voltage gradient or electric
field sufficient to cause the short circuit.

At the break down, the strongly conductive break down channel is formed in
the dielectric. V, determines the maximum voltage that can possibly be applied to the
dielectric.
The break down depends on the nature of the dielectric material and the
power ofthe electrical energy. The break down voltage generate a spark or an electric
arc which can crack, or fuse the dielectric. The dielectric is punctured. The dielectric

244
Dielectric Properties of Solids
ee ee Se ee ee ee
now functions improperly. The break down takes place in all types of dielectrics,
solids, liquids or gases at some voltage or other. The voltage at which the break down
occurs depends on the thickness of the dielectric.
The dielectric strengtn or a dielectric is regraded as the break intensity of an
electric field (V3 = Egt), t is the thickness ofthe dielectric and is expressed as MV/m.
The temperaures, humidity, voltage frequency, the time for which the electric field is
given are the factors that decide Vj values. Under the action of an alternating voltage
of 50 Hz, the dielectric strength for mica, is around 100-300 MV/m, for rubber 30-50
MV/m, transformer oil 15-25 MV/m and for air, it is 2-5 MV/m.
In gases, the dielectric break down is followed by discharge of gases and
corona. The gaseous dielectrics exhibit high break down strengths. Elegas (SI),
Freons employed in refrigerators are found to possess high dielectric strengths.
(— 6.5 MV/m).
Petroleum oils, such as transformer oil, cable oil, are good insulating liquids.
The presence of water decreases their dielectric strengths. Transformer oil has a
dielectric strength in the range 17 MV/m -20 MV/m with low viscosity. It can work
up to voltages 30-35 KV. Sovols are very famous synthetic oils that are used in
capacitors. They have a relative permittivity of 4.6 to 5.2 and have the same dielectric
strengths as transformer oils.
Solids form an important class of compounds as dielectrics. The break down
in solids is classified as electric break down, and electro-thermal breakdown. The
forces of an electric field cause the electric break down. It is developed as a result of
the interaction of free charged particles like electrons or ions accelerated by the-
electric field with the atoms or ions of the dielectric. This interaction results in an
inelastic displacement of bound charges on atoms in the dielectric under the action of
the higher external electric field. The electro-thermal break down is basically due to
the dielectric loss. Under the action ofthe electric field, heat is liberated in the dielectric.
These losses increase as the magnitude of the voltage or electric field is increased as
a result of the increase in conductivity ofthe dielectric: Under the action ofalternating
voltage, the higher magnitudes of tan 6 of the dielectric higher and the frequency of
the ac voltage increase the losses. As losses increase, more heat is generated in the
dielectric. The temperature of the dielectric rises. This results in greater losses. The
process continues until the dielectric is heated so much that it gets fused or burnt or
cracked. This phenomenon depends on the nature and strength of chemical bonding
between ions in the dielectric, and the external conditions. A good dielectric should
have low dielectric loss factor, low conductivity, so that the rate of increase of
conductivity and loss factor with growing temperature become insignificant. The
thermal conductivity of the dielectric should be high so that the heat is withdrawn
easily from the dielectric. Such dielectrics work for a long time under a comparatively
high voltage without getting damaged.

245
we cE
Solids are the best variety of dielectrics, useful in electrical insulation and
capacitors. An insulators should have low permittivity values, low loss factor, thermal
stability, good mechanical strength, compatibility with environment high dielectric
strength, and high resistivity values. Most widely used solids are polymeric materials
like polystyrene, polyethylen, polyvinyl chloride (PVC) acrylic plastic, kapron etc.,
Polyethylenes are thermo plastics and are useful in the manufacture of insulating
sheaths and piping. PVC products have high mechanical strengths, low dielectric loss
(10-2) and their dielectric strength is of the order 15-18 MV/m, the relative permittivity
is 3.6. Kapron fibre is highly valued in electrical engineering. It has a dielectric strength
of 18-20 MV/m. Novolack is useful for low voltage applications. Silicone compounds
-are useful dielectrics with dielectric loss in the order of 10-3 and the dielectric strength
of 50-60 MV/m. Electrical ceramics are useful both as insulators as well as capacitor
materials. Porcelain is widely used. It has a loss of 10-3 and the dielectric strength of
30-32 MV/m. Stealite is another ceramic with a dielectric strength of the order of
35-40 MV/m. They are useful as supports for winding coils. A/, O, (alumina) is
another insulating material. It has low permittivity and low loss. The capacitor materials
should have high permittivity values, low loss factors, high resistivity values, low
frequency dependence of loss factor and dielectric constant in the frequency range of
operation. They should have good thermal stability, high dielectric strengths and easy
preparation techniques. Many ceramics like TiO,, SnO, or ZrO,, CaO, MgO or their
mixtures are good capacitor materials. Ceramic capacitor materials based on titanium
dioxide or Titania are called titanates e.g., MgTiO,, CaTiO,, BaTiO, etc. There are
stannates like CaSnO,, and Zirconates like BaZrO, that are useful for high capacity
values and high dielectric strengths. The ferroelectric ceramic like BaTiO, is useful
capacitor material. It has very high dielectric strength.

Exercise 1:

The Diameter of an argon atom is 3.84 x 107!°m. What is polarizability at

S T P. It’s dielectric constant is 1.00044.

x 1076 = 25
The number of molecules per unit volume = A OLD

ae €y{e-1) _ 8.85x10"? x 0.00044

N 2.6x10

= 1.43 x 10-9 farad m2.

246
Dielectric Properties of Solids

Exercise 2 :

The polarizability of Oxygen atom is 1.9 x 1047 Fm2. On being polarized it has
a dipole moment 4.8 x 10-23 C-m. What is the displacement of the centre of negative
charge cloud from the nucleus.
aat.Ox 10 b= m-*
P = 4.8 x 10-23 C-m

oo
Pe
Eo pba ii
Ba lO
t = 2:526x10°™ Volt/m

Lorenz Force = 8 eE

This is equal to the Coulomb force between nuclear and electric charges

8ex8e
8 eE=
4né
x?

: 8e 8x(1.6+10°"}
Ame yXE 4x3.14x885x
107? x2.53x 10°"
, 2BKIOF EK LO?
281.22
0.0455 x 10-2! = 45.5 x 10°34

X= "674-107! -m.
Exercise 3 :

Carbontetrachloride contains 74 electrons in its molecule. Its relative permittivity

is 2.26 when its density is 1.68 x 10° Kg/m?. If
the field acting an the liquid is 5 x 10°
V/m, 1. What is its electronic polarizability and average electron displacement.

6.02x107° x1.68x10° ~=—-10.11

N = ————_- == —_* 10° =6.5 x 1027 molecules/m3
156 156

p 510° 48.85 x10—7 &1,26

ere =8.57x10-2C-m
—33

PN 6.5x 1027

8.57x10°2 Ss 8.57
wv ate ele=——_—___= x1074 =0.724x107°m
ne J4xl6xl0° 118.4

247
 Applied Physics

Exercise 4:

A monatomic gas contains 3 x 10° atoms/m? at a certain temperature at one

atmosphere pressure. The radius of the argon atom is 0.19 nm. What is the relative
permittivity of the gas at the given pressure and temperature. What is the polarizability
of the atom. What is the intensity of polarization.

We know that €) (€,- 1) =N a,

N = 3 x 102° atoms/m? ; € = 8.85 x 10-2 F/m

The dipole moment per unit field P = 4 7, ce

4X 3.14 8:85-<-10-2 x (O19 10 2);

The polarization p = 3 x 1025 x 4 x 3.14 x 8.85 x (0.19)? x 10739
Go a 6 aa COIS? x10
é = 10026

0.0026 x 8.85 x 107!”

Polarizability a, =
3x1075

we eles Sil 26 x 8.85x107'° ‘ = -40

: N 3 10°° = 7.66 x10 farad m2:

Exercise 5 :
A Water molecule has a dipole moment 6.2 x 10-3° C.m. What is the polarization
intensity of awater drop of 0.1 cm radius polarized in the same direction.

4 apg
The Volume of the water drop = 52 ™x10“m

26 =i2
The number of molecules in the drop N= Sel) = uv x 103

Bee eee

4x3.14x10t!”
Pe —aqia ee Oo ee = 26.38x10-3C/m2.

248
Dielectric Properties of Solids

Questions

Define : 1. Electric monopole, 2. Dipole, 3. Quadrupole, 4. Dipolemoment,

5. Polarizability.
Establish a relation between the electric field strength E, polarization P, and electric
displacement D. Write down the units of E, P and D.
Find the magnitude of D in a dielectric material for which y = 2, and P = 3.08 x
10-7 @im*
What are polar molecules and non polar molecules ? Give examples.
Derive the expression for the polarizability of hydrogen atom assuming its negative
charge as distributed in a spherical volume of radius r around the nucleus.
Derive an expression for the total electrical field on a molecule inside an isotropic
dielectric medium, Hence, show that

e-l1 Noa
oe ee 3e5

Derive and discuss Clausius-Mosotti expression. Discuss various sources of

polarization in molecules.
What are dipolar substances ? Explain the temperature variation of their
susceptibility and polarizability. What is saturation polarization ?
~ What is deformation polarizability ? Does it vary with temperature ?
Discuss the frequency variation of electronic polarizability.
Explain how electric polarizability is correlated to refractive index of a material.
What is ionic polarizability ? Is it sensitive to temperature ?
A solid contains 5 x 1028 identical atoms per m3. E~ch atom has polarizability of
2 x 10-40 farads. m2. Assume that the internal field is given by Lorentz relation.
Calculate the ratio of the internal field to the applied field.
Obtain the expression for dipolar polarizability of adipolar system of molecules.
Discuss the temperature variation of dipolar polarizability.
What is “Complex permittivity” ? What is “loss factor” ?
Discuss the frequency dependence of polarizability or dielectric constant of a
dipolar substance.
Explain the phenomenon of “dielectric relaxation”. Hoe does this relaxation depend
on the frequency ofthe applied field across a dielectric ?

249
 Applied Physics

16. Discuss the frequency dependence of complex permittivity ?

Lis Discuss how the total polarizability of a dielectric substance show the marked
difference when studied as a function of the frequency of the applied field.
Discuss the principle of any methods for measuring the dielectric constant of a
solid.
The polarizability of oxygen atom is 1.85 x 1047 F.m2. The atom is being polarized
by an electric field of 2.5 x 10*4 Volt Im. Calculate the distance of the centre of
negative charge cloud to the nucleus. The concentration of the gas is
2.7 x 10°5/m?. Calculate the induced dipolemoment per unit volume.
ZU The Ammonia molecule has a polarizability 2.38 x 10-39 F.m2 and 1.87 x 10739
F.m? attemperatures 280K and 450K respectively. Calculate the value of deformation
polarizability and dipolar polarizability and dipole moment ofthe molecule.
zy Two identical atoms of polarizability 2.5 x 10-4° F.m? are separated by a distance
3 x 107'°m. A homogeneous field of 10° V/m is applied along the line joining the
centres of the atoms. Calculate the magnitude: of internal field.
pad A dielectric which is ionic in nature has a dielectric constant of 5.2 and the
refractive index 2.56. what is the ratio between electronic and ionic polarizabilities
of the dielectric?
23% In a free space D = 2i — 4j + 1.5K C/m2. What is P in a dielectric which has
relative permittivity € = 5.2. The dielectric is placed in the free space.
24. A parallel-plate capacitor, for which C = gA/d, has a static voltage V across the
plates. What is the stored energy in the electric field 2
Peo Find the magnitudes of D, P and €. for a dielectric material in which
E = 0.20MV/m and ¥,, = 4.75.
26. What class of crystals exhibit piezoelectricity, ferroelectricity and pyroelectricity.
past Explain the origin of piezoelectric effect and inverse piezoelectric effect.
28. What is electromechanical coupling coefficient ?
Zz. What is polarization catastropy ? Define a ferroelectric crystal.
30. What are the various important properties of ferroelectrics.
Sale Mention the ferroelectric characteristics of barium titanate.
32; Give examples of crystals which are ferroelectric.
cEx Give examples of piezoelectric materials.
34. What is Electrical poling ? Why is it done.
35: Discuss the applicationsofpiezoelectrics and ferroelectrics.

250
6
MAGNETIC PROPERTIES
OF MAGNETIC MATERIALS

6.1 Introduction:
One of the most important and fascinating branches of Physics is the Physics
of magnetism and electromagnetic phenomena. The electric fields and magnetic fields
can be separate and the two are very dissimilar in many respects. But when they are
time dependent, there is a unified blend of the electric and magnetic fields in the form
of electromagnetic wave. The energy of an electromagnetic wave concentrates
concerned
alternately in the electric and magnetic fields. However, in this chapter, we are
with the origin of magnetic dipolemoment in materials, the material classification and
their utilization. The fundamental entity responsible for electric field or magnetic field
is charge. The static electric field is associated with charge. But we do not have
magnetic charge in nature. A moving charge gives rise to a magnetic field. The magnetic
field has its microscopic sources. The microscopic sources are the electrons, protons
and the neutrons. All these particles have a magnetic moment. But all materials are not
magnetic in the sense that all materials do not show magnetic moment. Materials possess
different types of magnetism. That is inferred from the fact that different materials

Zak
 Applied Physics

behave differently when placed in a magnetic field. In these materials, there is a way of
interactions of the charged particles which will either give or not give magnetic properties
to the material body. Even if there is an interaction among the magnetic dipoles (each
moving charge in an atom is associated with a magnetic dipole moment), the interactions
differ from material to material giving rise to different types of magnetic phenomena.
The magnetic properties of materials are widely used in many applications starting
from permanent magnets used as a toy by a child or a permanent magnets in loud
speakers to computer memory systems. The magnetic materials find wide applications
in all types of motors, dynamos, telephone receivers, ultrasonic generators, transformer
cores, antennas in telecommunication, microwaves, etc. The magnetic materials
dispersed on a substrate or made into thin film coatings on a tape are used for audio,
video, and digital tape recording. Data storage in floppies is nothing but recording on
magnetic tapes. The utility of cylindrical magnetic domains as bubble memories is very
important to achieve high writting density in memory systems. Magnetiostriction
associated with magnetic materials is used in ultrasonic applications or in strain gauges.
The subatomic particles possessing magnetic moments and the subatomic particles
present in the tissue of living beings respond to the external magnetic fields. The
magnetic resonance methods are useful in medical diagnosis. Therefore, the study of
the basics of magnetism, its origin in different materials and the utility of magnetic
materials is essential for electronics engineer. We will discuss the above aspects in the
present chapter .
Before taking up the study of magnetic materials and their classification, we will
study some important fundamental aspects of magnetostatics which are important to
understand the subject on magnetism in materials.
It is known that a moving charge constitutes an electric current and gives rise to
a magnetic field. Consider a long thin conductor like a wire carrying a current I. The
conductor is surrounded by a medium.

Fig. 6.1 The magnetic field induction around a current carrying wire

252
Magnetic Properties

When a compass needle is kept near the wire (Fig 6.1) the needle gets deflected
indicating that the conductor produces a magnetic field around it. When the compass
needle is moved around the wire, the locus of all points at which the needle shows
same deflection will be a circle. We get different concentric circles, around the wire,
which represent the positions of the same deflection. The relation between the direction
of the magnetic field and that of current is given by the right hand rule with the thumb
pointing in the direction of the current, the fingers circling the conductor or wire.
The magnetic field density b at any point on the circle is

_ Hobe y anbe kenttic upset Gad Ae

Baeble ching (6.1)

Consider a current carrying loop as shown in Fig. 6.2. Then, the magnetic field
density B is given as

Fig. 6.2

B= Mo, IR

saek (6.2)
2(R? + z2)p
Z is the distance between P and the centre of the loop. At the centre of the loop,
the magnetic flux density due to the current in the loop is

Bao
OR
HoH,
pet) oe a, GyGkl Sea wn pore (6.3)
6.3

The magnetic flux ym is givenas y,, = BA, where A is the area and the magnetic
field lines are perpendicular to A. The SI unit of the magnetic flux is weber. If B is not
uniform over an area, then we consider an infinitesimally small element of area dS,
then, the flux across this area is B. dS.

Then ja | Be dSéher & Rectorate inBelem ang (6.4)

S
dS is a vector. The magnetic flux over a closed surface is

{ BedS=0
i.e., B is solenoidal. This is the representation of Gauss law in magnetostatics. It
indicates that there is no magnetic monople.
259
 Applied Physics

A current carrying loop acts like a magnetic dipole. We have seen earlier that the
flux density at a distance R from a straight conductor is given by the equation (6.1). B
is integrated around the path of radius R enclosing the wire once. This can be expressed
as

Holt
Bed/=—*+ pl
qdJ/=——.2nR=pl i... 6.6
J i 27R | 2mR & ey

where 4H, =p the permeability of the medium around the wire. If the integration is
taken over a singly closed path H = B/,. H is called the magnetic field intensity and

{iHscr menede Govalnss) abana (6.7)

The above relation is called Ampere’s law. It states that the line integral of H
around a single closed path is equal to the current enclosed. The units of H are
Webers/m7
a a amp.m :
Weber/amp.m

The quantity H is also called magnetomotive force (mmf). It is equal to the

current enclosed. The sign of the integral depends on the direction in which the current
is enclosed. The sign is positive if the path followed by the line integral is parallel to B
and negative, if the path followed in antiparallel. Ampere’s law holds for closed path of
any shape. The current is due to the flow of elections across an area per unit time. We
define a quantity ‘J’ as the current per unit area placed normal to the direction of

current. ‘J’ is expressed as != MI,JedS where dS is a small element of area or

surface. The Ampere’s law can be written as

qBedl=y [J ¥eas leat (6.8)

According to Stoke’s theorem (Appendix-I) of vector analysis for the vector
function B, we can write

{Beal= [JcunB- dS

[[ Curb ds = [[sas

or Curl B = ypJ

254
Magnetic Properties

Forireesspace CanB= oI eg oncnniins (6.10(b))

If there is no current or current sources, J = 0; i.e., Curl B = 0. The equation

(6.10) is called the differential form of Ampere’s law. It is a point to point relation ina
medium. It is valid for steady currents.
The divergence of the magnetic flux density B is always zero. That means B is
solenoidal. This is the differential form of equation (6.5). The lines of B are closed
loops.
Oersted pointed out that the magnetic field is associated with the electric current.
Ina similar way, Faraday and Henry independently suggested that the changing magnetic
field with time should give rise to electric field. Faraday's laws of electromagnetism are
stated as: (1) A magnetic flux, linked with a closed circuit, changes, an induced emf is
set up in the circuit whose magnitude at any instant is proportional to the rate of
change of magnetic flux linked with the circuit. If @ is the induced emf, then

ae aan
ST eae

(2) The direction of induced emf is s:uch that it opposes the change in flux that produces
it. This law is known as Lentz’s law.

A magnet has a magnetic dipole moment. When the north pole of the magnet is
brought towards the coil, the magnetic flux through the coil increases i.e., the flux
the
through the coil is changing with time. It will induce an emf in the coil such that
current has to flow in
induced emf will be anticlockwise as seen from the magnet. If a
pole face the
the coil, it will be circulating in such a way as to make the coils north
and the
approaching magnet. The is, there is a force of repulsion between the loop
magnet.
d
The electric field is irrotational and therefore the electric field can be expresse
field can
as the negative gradient of a scalar function ie., E=—grad >. The electric
is a vector point
be considered at each point in a medium or space. The electric field
medium. An electric
function or a vector field. Similarly B is also a vector field in a
a closed path
field is conservative, hence the workdone in moving a test charge around
effect,
in the field is zero. This is expressed as |Ee d/=0. But in the case of Faraday
is not associated
the changing magnetic field produces electric field. This electric field
The electric
with static charges, but with a changing magnetic field i-e., |Ee d/ #0.

259
 Applied Physics

field produced in Faraday effect is nonconservative. Both conservative and

nonconservative fields are identical, but their sources are different. The emfdeveloped
in the coil depends on the time rate of change of magnetic induction.

Consider a loop C as shown in Fig.6.3

in time varying magnetic induction fields. Me eXEEOC, OO OKO OK
Then the induced emfi.e., an induced electric SOLES. TocE ocprcel
iceman

field E will set the charge carriers in the loop,

there by inducing current.
Ke
eK
oie

The electric field is present at all points x m mx

in the loop C. This electric field will exert a ; 7

force F = q,E ona test charge q, tangentially 3 % yok
at every point ofthe loop. The electric lines Me Ki Ke Ka EE KOK
form concentric circles. Consider a circle x CEM
EK
ee.
KK
EE KP
ES
Re
eo
SE
BE
PR
ee *

with radius r. Then, the work done in

carrying a test charge q around the circle is Fig.6.3 A coil in the changing magnetic
induction x indicates that B is perpendicular to
given as
the plane of paper.

W = dFedi= qq,Eedl = q,E.°2nr

.. The induced emf €=E. 2ar

={Eedl aed ) ee ee
dt (6.13)
That is how Faraday’s law establishes a connection between
E and B. The
“above relation is known as the integral form of Faraday’s
law. It states that the line
integral of Earound a closed path is equal to the rate of change
of magnet flux within
the boundary of the path. The coil C encloses an area A. If
B is the induction field at
any point P (x, y, z), then the magnetic flux B through the area
dS is B,,-dS, where inte
is the normal component of B. The total flux in the coil is

>= [[B, ds

“Gheat =F (JB, eas

nas (6.15)
According to Stoke’s law (Appendix I)

qEed/= {{CurlEeds

256
Magnetic Properties

a {JCurlE*ds=— {|B, edS

BA [Cure + <3 edS=0

dt

dB
or CurlE = Lorry ponead (6517)

B is a function x, y, z, t. The above relation is known as differential form of

Faraday’s law. It is one of the Maxwell’s equation. We will now write the Maxwell’s
equations
(I)V eD=p (2)V eB=0

DIVX ES (4)V eH pSen. emo opens (6.18)

or in the presence of displacement current, the last equation can be written as

dD
VsH = J +—4.
dt

With this background on magnetostatics, we will now study the magnetic nature
of materials.

62 Magnetic properties of atoms : Bohrmagneton :

We have learnt that a current carrying conductor gives rise to magnetic field
around it. A bar magnet gives rise to induction field around it. There is no visible
current circuits of the type we know of a circuit in a bar magnet. But the magnetic
induction field arises only when a current exists. Then, where are those currents ina
magnet ? To answer this question, we have to study the magnetic nature of atoms and
the reasons for this magnetic nature. We shall study this in the following.
An atom of any element consists of election revolving round its nucleus. Each
electron orbitting (Fig. 6.4) around the nucleus produces a loop current I=ev, where
v is the frequency of rotation of the election in the orbit. The electron possesses an
angular momentum L,. The magnetic moment },,, of the loop of current is
Hin — (area of the loop) * (current)
 Applied Physics

The symbol u1,,, should not be confused with the

symbol -H for the permeability of amedium.

The magnetic moment of the electron p,,, due to

its motion around the nucleus is called the orbital
magnetic moment. The direction of 1,,, is perpendicular
to the plane ofthe orbit (in accordance with right screw
rule).

The angular momenlum ofthe electron L,,=mvr,

where m and v are the mass and velocity of the electron. Fig. 6.4 Electron orbital
But according to the quantum theory.

Lys h ay
liRI). ow 8 5 naan ee (6.18)
21

where | is the orbital quantum number. It can assume any value of the integral series of
n, the principal quantum number. L can take values 0, 1, 2, 3,.... An electron with
! = 0 (i.e. S-electron) always has zero angular momentum and hence zero
magneticmoment. But other electrons with / = 1, 2, 3.... (i.e. p, d, f, g..) have nonzero
angular momentum and hence possess magnetic momentum given by

nm=-(sJe = 2m
[ZED
4mm

= aad) ee (6.19)

eh
where Hg = ea oh wen vd weledeng Venodes eeaies # (6.20)

|p is a unit measure of atomic magnetic moment and is called the Bohrmagneton.

One Bohrmagneton is equal to 9.27408 x 10°4Amp m2. The orbital angular
momentum
can orient itself about the chosen direction, such as the direction of
a magnetic field H
only in such a manner that its projection on H direction is an integral
multiple of
h/2 rr 1.e.,
Ligh hay/270.n} peltosio eel te palate te. garn
et ae (6.21)
Here, ‘m/ is called the magnetic quantum number. It can take all integral
values
from —/ to +/.
Mn =U) 0 ie 1, in total, (2/+1) values. From equation (6.21)

We get Hy =—M)Np

258
Magnetic Properties

An electron in p level has three possible components of angular momentum

along the direction of the field, h/2 x , 0,—h/2 x. The magnetic moment component are

p> 9, —Hp-
An electron possesses an intrinsic angular momentum of spin S to which the
spin magneticmoment } , is related. The projection of this moment on the direction of
the H field is equal to the Bohrmagneton. The magnetic moment due to spin in the
direction of the field is

ite ae
wn
wl

The net spin magnetic moment of atoms containing paired electrons or filled
electron shells is zero. Hund’s rule states that the electron spin always add together in
such a way as to give maximum spin angular momentum. Suppose an atom has 10
e
electrons ina d shell, the first five electrons give a moment . The next five electrons
4mm
with opposite spin will nullify the moment. When an atom has odd number of electrons,
one of them will become unpaired, so the atom on the whole will possess a permanent
magnetic moment. When an atom has an even number of electrons, two cases are
possible (1) all electrons are paired and the resultant spin moment is zero. (2) two or
more electrons can be paired parallel and the atom possess a permanent magnetic moment.
Solids whose atoms do not possess a permanent magnetic moment or a net
magnetic moment are diamagnetic.
Solids with atoms in which there is a i t & Mwy J a aN
net magnetic moment can be
paramagnetic, ferromagnetic or t ‘i I I t t t i
antiferromagnetic. If the interaction
between magnetic moments of atoms
does not exist or is negligible, the solid t | t | t | i) |
is said to be paramagnetic (Fig. 6.5(a)).
1 | t | t | ‘ |
If all the neighbouring magnetic
movements tend to align themselves
parallel, the body is ferromagnetic Fig. 6. 5 Schematic representation of magnetic
moments in materials
(Fig 6.5(b)).

If the neighbouring magnetic moments tend to align themselves antiparallel, the

are
body is antiferromagnetic. (Fig. 6.5(c)). If the anti parallel magnetic moments
)).
different in magnitude in a solid then it is said to be ferrimagnetic (Fig.6.5(d
239
 Applied Physics

“We will consider in the following, a moderately detailed discussion on various

types of magnetism.

63 Theexternal magnetic field and a magnetic material - The

three vectors B, H and M and magnetic permeability.
A loop of area A with a current I is associated with a magnetic moment of IA.
The fields at large distances from this loop are identical with those of a bar magnet of
dipolemoment m/, where M is the pole strength and | is the pole separation, provided
the magnetic moment of the bar is equal to that of the loop. Thus ml = JA. A bar
magnet has atoms. These atoms are associated with magnetic moments and they are
aligned in the same direction in the bar. The magnetic bar consists of tiny magnets or
a miniature current loops. The bar as a whole should consist of a magnetic moment
m/. The effect of atomic magnets can be described by a quantity called magnetization
M, defined as the magnetic dipolemoment per unit volume analogous to the polarization

: ; : i _ Mm :
P in a dielectric.Thus M = Woy where m; = mi, V is the volume of the magnet.

Magnetization has the dimension of both magnetic dipolemoment per unit volume and
of magnetic pole strength per unit area. It is expressed in amperes per meter. To define
M at any point in a magnetic bar or rod that has a continuous distribution ofinfinitesimal
magnetic dipoles i.e. a continuous magnetization where as the dipoles are discrete in
size. Assuming continuous magnetization, we can define

Me Lt om,
ANS OUNVede og ee ee Nee ree

or m; = [Mav Am?
Vv
where the integration is carried over the volume of the bar. A magnetized bar or
rod is equivalent to
uniformly wound coil like
solenoid in which a current
exists.
Consider (as shown Cross
in Fig: 6.6) a coil in the sectional N Turns

form of a toroid of area A

and radius R, that has N
turns and initially has air as
core, then the magnetic
flux density B, in the
toroid is Fig. 6.6 Toroid with N turns

260
Magnetic Properties

Suppose the same winding is placed on a ring of a magnetic material and the
core has the same radius R, then the magnetic flux density from the windings is
NI
Bi, = Ho ei
ral |iV aed = ora eae (6.25)
Then the total B is given as
B=B, +B,, =H,(H+M)

or Ba VE ee eg oe sch (6.26)

where B is the magnetic flux density, (T) H, the magnetic field intensity, (Am7!), and
M, the magnetization (Am”!). The above expression is developed for a toroid. But it is
valid in general for the materials kept as cores in any coil. The expression (6.26) can be
written as

B= mC +.*)

M
Then Mer =Ho| 1+ 7c lorilEden alk sediment Aiiil (6.27)

M
then y= Sie es (6.28)

where 1, is the relative permeability of the medium of the core. [1 is assigned value.
The permeability of the medium represents the extent to which field flux lines penetrate
the magnetic subtance. It is a function of temperature. For ferromagnetic substances,
the permeability values are very high and of the order of hundered to a few thousands.
M
Wedefine =X=77 =F: ee eee sree Taree (6.29)

x, is called the magnetic susceptibility and is a dimensionless quantity . In isotropic

media, M and H are is the same direction, so that this quotient is a scalar and } isa
scalar. Both p and y are the characteristics of a magnetic material.
It is already stated that, V e B=0, Le.,

V © B=u, V ¢(H+M)=0

i.e., VeH=-vVemMeAme® — warn (6.30)

261
 Applied Physics

If the divergence of a vector field is not zero, the field should have a source or
a place of origin. Ina polarized dielectric V e P=—p (charge density of polarization
charges) which means that the polarization field originates on the polarization charge at
the dielectric surface. Similarly the H field originates where the magnetization M ends
and H field ends where M field originates. The locations where V e H or V e Mis
not zero are regarded as locations of magnetic poles of a magnetized object.

Take the Curl of the equation (6.26), then

VxB=(VxH+VxM)
yl Porta
VieG My AROS Caer OF @ binprecitese: (6.31)
If there is no magnetization

Vx B=pu,J
and for magnetic materials
VeeBe pe trale)cey acy ae naerm nmin Her uae Gere (6.32)

where J., = V * M(Am-”). By the magnitude and sign of magnetic susceptibility, all
solids can be divided into the three groups, diamagnetic, paramagnetic and ferromagnetic.
If the susceptibility is negative, the material is diamagnetic. If the susceptibility is
positive, it is paramagnetic. The value of susceptibility is low for these materials.

6.4 Diamagnetism:

A diamagnetism substance possesses a net negative magnetism. Even though

the substance consist of atoms that have no net magnetic moment, there is a tendency
of electric charges partially to shield the interior of a body from an applied magnetic
field. The solid responds to external magnetic field. The negative susceptibility arises
from Lentz’s law. The circuit that should be imagined is the circulating electron in an
atom or ion or molecule. A diamagnetic sample moves away from the magnetic field.

A fast spinning top revolves about a vertical axis, along with the rotation about
its own axis. If the axis does not coincide with the vertical, it precesses. (Fig.6.7(a)).
A similar motion occurs for electron in an external magnetic field. An electron revolving
in its orbit precesses a closed loop current possessing a magnetic moment. The magnetic
field strives to turn the electron orbit normal to H, causing a precessional motion of the
orbit (Fig: 6.7(b)) around the field direction.

262
Magnetic Properties ,

"9
Fig. 6.7 (a) Spinning top, (b) An electron and orbital magnetic moment.

Suppose that the atom is placed in a magnetic field. The electron revolving is a -
current loop. When the magnetic field is applied, the induction field changes from zero
to B. Then there is a change of flux through the area of the loop. Hence an emf€ is
produced. & is given as

ai —don d
agra a i Fa OR (6.33)

where H is the external magnetic field intensity, ,, is the magnetic flux. The emf§
gives rise to an electric field E acting longitudinally to the direction of the electron
motion. Then E is given as

ee St te
Bee ee ore: (6.34)

where r is the radius of the orbit of electron. The force on the electron is eE. Then the

e do, ° . |
acceleration of the electron is — J dt ). If we think ofthe finite changes, the

change is the velocity of the electron as a result of the finite change in the flux as
€ = 2

as I>2nRm Joo# is i B)

er
— —|—
(=)|B a (6.35)

The corresponding change in angular velocity is

Av eB 6.36)
Ao = Seo altace ore ,Hi (6.
 Applied Physics
ee ae eS

The change in frequency can be written as

eB
a = ©, =~ bea
The frequency @, is called the Larmor frequency. This is the frequency of
precession. This results in the change in the mag(letic moment. Those electrons whose
orbital moments are parallel to the field are slowed down. The electrons whose orbital
moments are antiparallel are speeded up by an amount eB/2m. This is called Larmor
theorem. The magnetic moment of the orbiting electron is

Na i>)

In the presence of the field, the change in magnetic moment is

Alm Reed
= > ia
tae 6.38
eee (6.38)

The negative sign indicates that Ap is always opposite to the magnetic field
induction B. If they are n electrons per unit volume in the substance, then the magnetic
moment change per unit volume is

gee
ie ea LSE ere or te ne bas
meter (6.39)

The susceptibility x, is

ANS es —nze’r*
eae paral
Nap ez 2
= - teliA =} arene (6.40)

The diamagnetic susceptebility is independent of temperature and field strength.

Monoatomic rare gases like He, Ne, A, Kr etc. have closed shell electronic structure
and are diamagnetic. Similarly, polyatomic gases like H,, N, have filled electron shells.
lonic crystals like NaCl, KC/, MgO, covalent crystals like Si, Ge, are all diamagnetic.
Almost all organic compounds are diamagnetic. Super conductors are perfect
diamagnetic substances. Most of the simple metals like Copper, Silver, Gold, Bismuth,
Sodium etc., are diamagnetic.

264
Magnetic Properties

The table 6.1 gives the susceptibilities of some diamagnetic substance.

0.1 x 10°

2.1 5

a7 x10 12x10
65 Paramagnetism:
Atoms of a sunbstance may contain many electrons. The magnetic moment of
an atom is the vector sum of the magnetic moments of the individual electrons. Similarly
the totoal angular momentum is the vector sum of separate angular momenta. the
method of adition of the orbital and spin angular momenta and obtaining the resultants
J is known as Russel-Sander's coupling of spin and orbital momenta. The resultant
h
orbital angular momentum L takes values VL(t+ Yar . The value of L depends on
the directions in which the various orbital angular momenta points with respect to one
%
another. The maximum value of L is Diy and the minimum is zero. 'Z' is the atomic
1
ys
number of the atom. Therefor L takes value 0, 1, 2, 3.... iy . Similarly the resultant
1
angular momentum S takes values Z/2 (the maximum) and minimum zero when Z is
even, and 1/2 when Z is odd. The rules were given by Hund. Hund’s rule states that for
the ground state of an atom (1) the spins of electrons add to give the maximum possible
S consistent with Pauli’s principle. (2) when the maximum value of S is determined,
the orbital momenta add in like manner to give the maximum possible value of L. (3)
For an incompletely filled sheil J=L-—S fora shell less than half filled and J=L +S for
a Shell more than half filled.
The vectors L and S are combined to give J, the total angular momentum is
h
expressed as VJU+1 On: J changes by | from |L + S| to |L — S| giving (2J + 1)

possible values.

265
 Applied Physics

The resulting magnetic moment now along the direction of an externally applied
magnetic field is expressed as

Hsia git, JIG +1) = gp ee (6.41)

m, takes values from —J to + J. There are (2J + 1) values i.e. orientations for mj. g is
called Lande's splitting factor given in the form

4 -L(L+1)+S(S+1)+3J5(J
+1)
Born ace
=)
The total magnetic moment of an atom specified by the vector J is, there fore,
given as

Ua eeeliag|
Goo pe 1 eae (6.43)
Langevin developed the theory of paramagnetism. This model of paramagnetism
is given in many book. As already said, paramagnetism occurs in those substances
where the individual atoms, ions or molecule possess a permanent magnetic dipole
moment. However, in the modern world, it is appropriate to consider the quantum
theory only. We will see Langevin model in terms quantum parameters.
All atoms possessing odd number of elections exhibit paramagnetism. The
examples are many. A few of them are free sodium ions, gaseous nitric oxide, organic
free radicals. Free atoms or ions with partially filled inner electronic shells, transition
elements ions, rare earth ions, actinide elements, etc. are also examples for
paramagnetism. There are a number of compounds with even number of electrons,
including molecular oxygen, that exhibit paramagnetism. The susceptibility of a
paramagnetic substance increases with decrease of temperature. In a non uniform
field they experience a force towards the stronger regions of the field. The examples
for paramagnetic substances are Alluminium, Tungsten, Oxygen etc.
A paramagnetic material contains very large number of magnetic dipoles. They
are randomly oriented and distributed. The net magnetic moment of the material is zero
without the field. We can treat the material sample as an assembly of magnetic dipoles
or molecular or atom dipoles. This assembly can be called a paramagnetic gas. In the
presence of an external magnetic field, these magnetic gas dipoles tend to orient
themselves along the field direction. But thermal energy gives an agitation to the dipoles
and opposes the tendency. Therefore, the molecular dipoles can be oriented at all
angles with respect to the field direction but majority of them tending to be along the
field. Each dipole has a magnetic moment iL.
. But, according to quantum theory, the dipoles can orient only at certain specific
directions with respect to the magnetic field. For an atom with a particular value of J
and placed in the magnetic field H, the directions are prescribed by the rule that the
possible components of magnetic moment 1,,, have to be Myg Mp, M, can take (21 + 1)
directions. The potential energy of the atom is —M, g Up B. Then the number of atomic
dipoles with a particular value of M, is proportional to Boltzmann factor e~(Ms8#BB/kT)

266
Magnetic Properties

The number of the dipoles which have their M, values from —J to + J is equal to
+J
Sep rE ls (6.44)
My=-J

The total magnetic moment of all those dipoles which have J value from —J to +
J is
+J
> M, exp Mus#eB/KT)
My=-J
If n is the number of atomic dipoles per unit volume, then the average
magnetization per unit volume, M is
J
n> Mjgis exp (*MssuBB/KT)

M= So ee ee (6.45)
5"exp(*Musuat/)
=i

At normal temperatures and low field M, g tg H << kT. Fora field of 1 Tesla,
g = 2, T = 290K, and, exponent value is less than 0.002. Therefore the exponential
functions can be expanded and high powers of exponents can be neglected.
+J

Sia), M, (I+ M gH_B/kT)

=
M=n +J
>"(1+ M gps /kT)
=

netsgupB J(J+1\2I+ »
a 3
(25 +1)
ee,
ng’ HBB HT I)s ce)
eee hae |el eee 6.46
3kT ee eee)
-- The susceptibility x of a paramagnetic substance is given as

_M _ pong?HpJ
+1)
aS 3kT

(B = 1H)
Ake)

267
 Applied Physics

where c is called curie constant and is expressed

Hong” 22Hp J(J +1)ee

= eo eee we ak 6.47
‘ 3kT ( )
If n =n,,, the number of molecule per unit mass, the susccptibility y will be
mass susceptibility x,,. Ifn =N,, the Avogandho's number, then susceptibility will be
molar susceptibility.
The magnetic susceptibility is inversely proportional to the absolute temperature.
For weak fields and high temperatures, the theory predicts a linear dependence of M
on H and inverse dependence of x on T. This is in agreement with the experiments. For
strong fields and low temperatures, the magnetization attains a maximum value, namely,
the s&turation value M, and M, = ngJ Lp
Ma Den) Bae othe See (6.48)
where a = g tp J B/KT and B,(a) is called the function defined as

pine (2) co [22 2J 2)

l cat
anes, coth ay

A paramagnetic salt exhibits saturation magnetization as shown in plot given in

Fig.6.8(a). When 1/x is plotted against temperature T, we get a linear plot and the line
passes through origin at T = 0K as shown in Fig.6.8(b). The curve shown in Fig.
6.8(a) is called magnetization curve.

Saturation
le

Magnetization
——»

Fig. 6.8(a) 1ne magnetic moment ot a paramagnetic 1on as

a tunction ot m,H/T

268
Magnetic Properties

4
x

i") =>

Fig. 6.8(b) The reciprocal of the susceptibility as a function of Temperature TK for a paramagnetic
substance.

Paramagntic substances serve as working substances in pramagnetic amplifier

and oscillators, and in adiabatic demagnetization. The experimentally observed values
of the susceptibility are given in Table 6.2, for a few paramagnetic substances.
Table : 6.2

: Fe,(SO,), | 2.2 x10”:

6.6 Ferromagnetism :

6.6.1 The characterstics associated with Ferromagnetism :

Ferromagnetism is a phenomenon which occurs in some materials which exhibit
spontaneous magnetization. This nature arises in them when the magnetic moments of
adjacent atoms are arranged in a regular order i.e. all parallel to each other. The word,
‘spontaneous’ suggests that the magnetization in the material exists even in the absence
of an external magnetic field. Two important properties are associated with a
ferromagnetic substance. They are:

269
 Applied Physics

1. A ferromagnetic substance possesses spontaneous magnetization only below a certain

temperature T, called Curie temperature. Above the Curie temperature, the substance
suddenly goes into paramagnetic state. The magnetization in the substance is maximum
at OK and in principle, this magnetization is present over a long range of temperature.
Then, the magnetization decreases with the increase in temperature, around the
transition temperature it falloff very rapidly reaching zero value at the transition
temperature T.. If the substance is cooled from above the Curie temperature, the
spontaneous magnetization reappears at T,. This property indicates that there is no
structural or chemical change in the substance at the transition temperature. The
Fig.6.9 depicts the variation of spontaneous magnetization with temperature in a
ferromagnetic substance.

M,(O)

f°)
Peed Gs)naresT, RE
Fig. 6.9 Temperature dependentce of spontaneous magnetization.

At high temperatures, all ferromagnetic substances are paramagnetic. The transition

at T,, is from ferromagnetic state to paramagnetic state .of the substance.
It should
be remembered that all paramagnetics do not become ferromagnetic when
cooled
to a lower temperature. The spontaneous magnetization is a special property
of a
few substances like iron, cobalt and nickel.

2. Ferromagnetic substances, though spontaneously magnetized,

do not show
macroscopically observable magnetization. For example, take a piece
of iron bar .
Iron is basically ferromagnetic in nature, but the iron bar does
not show any
magnetization. But when it is brought into a magnetic field of a
permanent magnet,
it shows appreciable magnetization. Suppose such an iron piece
is taken in the form
of an annular ring and a copper wire of a few turns is wound on
it. Let another wire
be wound on it and be connected toa galvanometer. When a current
is passed in the
first wire ( called primary), a deflection is observed whenever the
current is changed
in the first coil. The current in the first wire produces a magneti
c field of some
intensity and then the material possesses a magnetic induction
field B. The magnetic
induction can be measured and it is a function of the magneti
c field intensity H.
When H is varied from zero value, B increases nonlinearly
and at a certain value of

270
Magnetic Properties

H, it tends to reach a maximum value (Fig.6.10). The magnetization at that value of

the magnetic field reaches a saturation value. The nature is like any paramagnetc
substance, but the magnetization value is enormously large. The saturation is obtained
at low values of the field intensities. When the current in the primary is now decreased,
there by decreasing the magnetic field intensity, the curve does not retrace its earlier
path (Fig.6.10).

(a)
Fig. 6.10 The magnetisation curve. (a) BvH, (b) M Vs H

When the external magnetic field is brought back to zero value, the B as well as
M do not reduce to zero values. Both of them possess remanent values (B, and M_).
This is the important characteristic of a ferromagnetic substance.
The magnetisation curves for iron, cobalt and nickel are different (Fig.6.11).
The way in which they reach saturation is different. The three substances possess
different saturation values. Ferromagnetic substance are characterised by large positive
values of susceptibility. The relative permeability values are very high. The expression
B = uw H is not valid. The permeablity depends on the field strength.

(2)°
>
—

i —
Fig. 6.11 Magnetization curves for iron, cobalt and nickel.

6.6.2 Molecular field-Weiss domain model :

The above characteristics prompted Weiss to propose a hypothesis and model
for a ferromagnetic substance. According to Weiss a ferromagnetic substance consists

274
 Applied Physics

of an internal molecular field which is responsible for spontaneous magnetization. He

also suggests that a ferromagnetic crystal consists of regions, on microscopic level, in
which there is a parallel arrangement of spill moments there by giving a net magnetic
moment equal to the spontaneous moment. However these regions themselves are
oriented with respect to each other randomly, so that the total magnetization of the
ferromagnetic substance like iron piece is negligibly small or zero. These regions of
spontaneous magnetization are known as domains. An unmagnetized sample contains
domains. Each domain is spontaneously magnetized. A poly crystalline sample like an
iron rod contains domaine whose magnetic vectors will be randomly oriented to give
zero magnetic moment (Fig. 6.12 (a).

Fig. 6.12 (a) and (b)

It is assumed that the magnetic moment of atoms that comprise a ferromagnetic

body are of spin (not of orbital) origin. When a ferromagnetic is placed in a magnetic
field, the spin moments are ordered not only by an external magnetic field H, but also
by the cumulative action of all magnetic moments of the atoms or spins. This is the
basis for the hypothesis of molecular field. According Curie-Weiss model, a
ferromagnetic substance consists of magnetic dipoles and these dipoles interact co-
operatively to give an internal molecular field (H,,,). The intensity of the molecular field
is directly proportional to the magnetization.
a. a Te prieets Maes ies... (6.49)
where y is called the molecular field constant.'y M' is a measure ofthe tendency ofthe
environment to align a given dipole parallel to the magnetisation already existing. If H,,
is plotted against M, the resulting plot is a straight line (curve 2 in Fig.6.12(b)). The
curve | represents the magnetization curve, represented by Brillouin function B(a). M
should satisfy both equations (6.48) and (6.49).
The total magnetic moment associated with the ferromagnetic crystal is expressed
as the magnetization M per unit volume and is given as

M = np, g JB(a)

242
Magnetic Properties

2I+1 Beak ) a
where B(a) = Hye coth | >, _| a- jcot{ =)

Initially the point at O is in unstable condition. If a small magnetic field is

momentarily applied, them the magnetization corresponding to P is produced. As a
result of the internal field, the point should be p' instead of p'. This corresponds to a
higher field H,. The field of this magnitude should produce a magnetization of Q,
which should shift to Q'. Thus, the sample will go through O, P, Q, R..... and reaches
the point S which is a stable point for the ferromagnetic substance. Thus the internal
molecular field sends the sample into the saturation.

6.63 Temperature dependence of spontaneous magnetization :

A ferromagnetic substance is looked upon as a paramagnetic sample with a very

large internal molecular field. The value of M at any temperature may be obtained from
the point of intersection of the two plots of equations (6.48) and (6.49). According to
Weiss, the total internal field in a ferromagnetic substance placed in an external magnetic
field of intensity H is
BL Seay) At 93) teed ead) eviiraegset ad (6.50)
The value of a in B(a) is

rm _BJ
fre btomnkeh
If we consider for time being that there is no external field i.e., H = 0, we have
to take only the internal molecular field H,, into consideration. From (6.49) a is given as

gUpyMJ
a=, Sea ee, (6.52)

The saturation magnetization is given as

IRR Oe 1 EE ea i) cc (6.53)

Bes
M, =i]81) Ray TEE, SA ae ear
pe
rt (6.54)

From equation (6.49)

M akT
Pea Re ee ns fo (6.55)
M, yng’ppJ?

295
 Applied Physics

M shouid satisfy both (6.54) as well as (6.55). The values of M/M, are plotted
as function of a as calculated from (6.54) and (6.55) shown in Fig. (6.13).

Eqn. 4.72b

fe) a—>

Fig. 6.13 The graphical solution of equations 6.54 and 6.55.

For temperatures less than T, (i.e. T <T,), always we get a non vanishing value
for M although the external field is zero. Therefore, the above model allows spontaneous
magnetization. The spontaneous magnetization becomes zero as the temperature rises.
AtT=T,, the critical temperature, the plot for relation (6.54) becomes tangential
to the curve of relation (6.55). The spontaneous magnetization vanishes. It means that
there must be a relation between g and T,. The slope of the line is nothing but the
tangent of the curve. Near the origin, a is extremely small i.e., a <<1, then

nc) « CO 2), @3+1) or

2) |(Q3+la 63 *45(25)

1/2),a a
2J}a 63 45(2J)

2 (+1), (+t) Qs? +25+1)

3 3) 30J?
: + 1) ;
The slope of the curve is carl slope of the straight line of equation (6.54) at

kT,
ARES De ie
er
© yng? pBy
274
Magnetic Properties

These two slopes must be equal. Therefore,

(+1). kT,
33 _yng?p3J?
3kT.
or ;a —Reting mediJrl)= Mi santos (6.56)

where he is the total magnetic moment of the atom. It is seen from above T, is
proportional to y. Taking the two relations given in (6.54) and (6.55), we can obtain

Be~ ye ex
cc aimee © 0. Fa (6.57)
Ss

The maximum magnetization occurs at T= 0K. All magnetic moments line up at

that temperature. As temperature is increased, this is disturbed and the magnetization
starts changing. Initially the magnetization decreases by small amounts and then drops

rapidly when (=) reaches more than 0.5 and the magnetization (i.e., spontaneous
T.Cc
magnetization) becomes zero at T = T,. This behavior is shown in Fig. 6.14 for iron,

cobalt and Nickel at = J . J = = value suggests that only spin is involved in the

magnetization process.

0.5 1.0
Ler Le

- ae
Fig. 6.14 M as a function of (=|for Fe, Ni and Co.-All the experimental points lie one the
M Ss Cc

curve for J = A*

275
 Applied Physics

6.6.4 Curie-Weiss Law :

Let us now consider the magnetization above T,. As the material will be in
paramagnetic region, it is possible to notice the magnetization only in the presence of
the external magnetic field of induction B. Then, for low fields and far away from
saturation,
J+1
B(a) = aie

\ Ngiig(J+ 1a
3
here gipi(H+ YM) oat
Se erat ee ee (6.58)
The effective magnetic field intensity on the atom is now H =H + yM.

M ———
bgnp(i+!) | gitpto(H+yM)I
3 kT

3kT

hee
M =
—
NULg Hp
———— Ji + 1) + —
ng HR _—J+1
YM
H 3kT ie 3kT H G+1)

_ nga +1)| — npe2n25(1

+1)
3kT a 3kT

_ mp gpd(I+1)/3k
a . (6.59)
7 _ Mo8 ba +1)
3k
The susceptebility in the paramagnetic region is given as

Seal, sOSe ae (6.60)

where C= nee ta Lok en ee (6.61)

22
fe 9 = yttes
n
HaJ( +1)
JiJ +1
‘cares te
3k

276
Magnetic Properties
(IR Seetiode iclisniSe ac

The relation (6.60) is called Curie-Weiss law. The variation of susceptebility in

the paramagnetic region with temperature. for a ferromagnetic substance is given in
Fig.6.15. The variation is linear, but the line does not pass through origin at OK. There
are deviations between experimental values and the values calculated from (6.60).

aspe
Bei]

Ose a: To 6 ‘’—>-

Fig. 6.15 Behavior oftheferromagnetics above the Curie temperature

The table 6.3 gives the characteristics of the three basic ferromagnetic substances,
namely, iron, cobalt and nickel.
Table 6.3

Magnetic moment at T = 0 M,
per atom per unit volume Amp m'

Cobalt(Co) | 1388 | 1428

133 2A

6.6.5 Exchange Energy:

They are two out comes ofthe above molecular field theory. Firstly, the molecular
field H,, is proportional to the existing magnetization of the sample of the magnetic
material. This implies that the magnetization phenomena involved is a co-operative one.
Secondly, the magnetization is very strong. But it could not explain the physical basis
for the origin of the molecular field. The magnetic properties of ferromagnetism are
due to the magnetism arising out of spin interactions of electrons. The orbital motion
does not contribute to ferromagnetism. One has to explain the nature of forces which
cause a co-operative phenomena i.e. the spin magnetic moments within a domain to

aT,
 Applied Physics

line up parallel to each other and give rise to spontaneous magnetization. The electrons
which are the sources of ferromagnetism are the 3d electrons in the transition elements
Fe, Co and Ni, Only those elements containing incomplete 3d shells and rare earths
with 4f shell give rise to ferromagnetism. The spin moments for iron group elements
are given in Table 6.4
Table 6.4

Element Ti Cr
Resultantspin [1 | 2 | 3 | 5 | 5s [4 [3 [2]
Spin
arrangement | ¥|Wo |WW WWW WY
On the basis of Hund’s rule, we find 4, 3, 2 unpaired 3d electrons in Fe, Co and
Ni respectively. The atoms of chromium and manganese exhibit most uncompensated
spins, therefore the n:lagnetic moments of these atoms must be maximum. We find
that the magnetic moments of the atoms of Fe, Co and Ni respectively should be 4up
3pp and 211, . But experimentally it was found that these atoms have magnetic moments
2.22 i.e. non-integral. In the case of solids, they possess band structure. It has been
suggested that the bands formed with 4s electrons over lap with the narrow bands
formed with 3d electrons in the above metals. The total number of electrons in 4s and
3d states is distributed among 4s and 3d bands. The relative occupation ofthe states in
the band is determined by the Fermi level. The difference in spin quantum moments
between these two occupied states will give rise to the net magnetic moment, for
example, in nickel atom, there are 8 electrons in 3d shell and 2 electrons in 4S shell. But
due to overlapping the electrons are distributed such that 9.40 electrons are in 3d band
and 0.60 in the 4s band. Following Hund’s rule, five electrons set their spins parallel to
the field and the remaining 4.40 set spins antiparallel. Therefore nickel has a net parallel
spin of 5-4.40 = 0.60 i.e. an effective magnetic moment of 0.6 Lp per atom. Similarly,
in iron, 7.8 electrons occupy 4s states. In cobalt, 8.3 electrons out of 9 electrons
occupy 3d band and 0.7 electrons occupy the 4S band. The magnetic moments obtained
this way agree well with the experimental values. It has been observed experimentally
that the addition of copper or zinc to iron decreases the magnetization of iron. The
copper or zinc atoms add electrons to 3d shell.
Heisenberg explained the large value of the molecular field on the basis of exchange
interaction. This interaction is purely of quantum nature and arises as a consequence
of Pauli’s exclusion principle. There exists an exchange interaction between the electrons
within the incomplete bands. This interaction depends on the relative orientation of
spins and not on the-spin magnetic moments. If they are two atoms i, jpossessing
electrpns with the resultant spin Si, Sj, then the energy of the interaction is defined
as
Uey = -21,(S; S)) = - 2, 8; 8; cosh

278
 Magnetic Properties
ee ee ee
where J, is called the exchange integral. 6 is the angle between spins. J, can be positive
or negative, depending on the ratio between the dimensions of the d-shells taking part
in the formation of exchange coupling and also on the distance between the atoms
in
the crystal. For iron group atoms, the exchange integral is positive. For others
it is
negative. This is shown in Fig. 6.16. Mn, Cr show a negative integral.

Fig. 6.16 Exchange integral Vs ratio between lattice parameter (a) and diameter d of the incomplete
inner shell.
The sign of exchange integral indicates which orientation of the spins participating
in the formation of exchange coupling isw advantageous, namely, parallel or antiparallel.
When S,, Si are parallel, the energy is negative i.e., Si bv S; as in the case of Fe, Co,
and Ni. This gives rise to spontaneous magnetization in the crystals as shown in Fig.
6.17. All the magnetic moments are aligned parallel to each other.

Fig. 6.17 Spontaneous magnetization of ferromagnet.

6.6.6 Hysteresis Loop:

The most important aspect of ferromagnetic substances is the hysteresis behavior.

The word ‘hysteresis’ has no connection to the word hysteria, but it refers to the

ZA9
 Applied Physics

physical state of a body. Suppose one applies a slight force to a rubber band and leave,
it will come back to its original state. Take a body like a thin metallic bar and fix at one
end,. Apply a force and leave, it will go back. Now increase the force, the bar will be
bent and it will not go back to its original state when the force is removed. We say that
it is exhibiting hysteresis behavior. Hysteresis happens in many systems. Ferro magnetic
hysteresis is just one among many examples. We shall Study the hysteresis behaviour
of a ferromagnetic substance.
Consider a ferromagnetic ring placed in a coil or torroidal coil as shown in
Fig.6.18(a). The Secondary coil (S) is connected to a fluxmeter (for the measurement
of B). The radius of the ring is kept large compared to its thickness.

(b)
Fig. 6.18 (a) A toroidal coil on aferromagnetic core. The magnetic flux density is measured with flux
meter. (b) Hysteresisloop

Let
The total number of turns in the coil =n
The current in the primary coi aa
The mean radius of the torroid =R
ni
The magnetic field intensity H is given as H = amp.turn.m!,
27R

280
 Magnetic Properties
a

This magnetic field intensity H will induce a flux density B in the material of the
ring and B = pH where 1 is the penneability of the material. The flux meter in the
secondary coil measures the magnetic flux cjlm and the flux density is given as
B = ,,/A where A is the cross sectional area of the ring sample. Initially, the current in
the primary coil is increased from zero in steps so that the magnetic field intensity H
increases in steps. B is measured at every step. This is done till the flux density reaches
a saturation value, the part OS as shown in Fig.6.18(b) of the magnetization curve is
traced. Then the current in the primary coil is decreased in steps to zero. At H = 0, the
flux meter still shows a magnetic flux. Therefore B does not become zero, but reaches
a value Br (part of SB of the curve). B. is called remanent flux density. The direction of
the current is now changed with the help of commutator C. Then the current is increased
in the reverse direction. At a particular value of H, namely H,, the flux meter shows a
zero flux indicating that the flux density has become zero. H, is called coercive field.
As the current is further increased, the flux meter shows a flux and this continues till
the point S' is reached. At this point, the current is decreased till it reaches zero value.
Then the current direction is reversed and the current i.e. H. increased further till the
point S is reached once again. We describe now a loop O S B S'S. This is called the
hysteresis loop. The observation of the hysteresis loop is an indication of the fact that
the magnetisation in a ferromagnetic substance is reversible. A ferromagnetic substance
is characterized by the saturation magnetisation (M,), remanent magnetization (M,),
coercivity (H.). The remanent magnetization is also called retentivity . Ifa ferro magnetic
substance is magnetized once, one has to apply a demagnetizing field H, to demagnatize
it. This is called coercive force. The area of the loop is proportional to the work done
on a unit volume of a ferromagnet in its magnetisation in a reverse sense. In the
process of magnetic reversal, this work done on the material reversal causes
ferromagnets to heat up. Energy lost per cycle due to hysteresis is proportional to the
area of the hysteresis loop.
An empirical relationship between the area of a hysteresis loop and the maximum
value of E attained in the loop (B,,) is given as the work done per unit volume per cycle
as n(B,,,)". 1 has a value of 1.6 for many ferromagnetic substances. 1 is called the
hysteresis coefficient and is a characteristic of the sample.
The energy consumed is proportional to the enclosed area just as occurs in

traversing the hysteresis loop. W = qH e dB. Let us see how this can be arrived at.
The substance possesses n molecular magnets per unit volume. Let the angle between
the applied field and the magnetic vector be 8. On each magnet, a torque t acts and the
torque t = ., mH sin6. If it has to move an additional angle dO, work has to be done.
This work is given as
DWat O20 AO SID GO gers ae ee tennis (6.62)

As 8 changes from 0 to 8 + d@, the intensity of magnetisation changes by an

amount dM.

281
 Applied Physics

dM=-d((Xmcos@)=—Ymsin@ d@ sees (6.63)

Then the dW is given as
dWeSS.reH.dMu tt| ete teed ternary (6.64)
The total work done by the magnetizing field per unit volume of the sample for
completing a cycle in the hysteresis loop is

W= qu,HdM = uqHdM
= uw, x area of the M—H loop
But B=, (H + M)
tes dB = (dH + dM) °
dB
dM = Ls. = FT poner Fe es i re (6.65)

W= fntte( Pan = fH « dB = Lu fH « dH

= fH C7 fs NORA TAE SIE ORSON ES SAM, Ce PLASree (6.66)

6.6.7 Domains and magnetization :

It has been stated earlier that a ferromagnetic crystal contains magnetic domains.
The domains are the spontaneously magnetized regions of the crystal separated by
walls called domain walls. In an ordinary iron crystal, the magnetization vectors of the
domains are arranged in such a way that their vector sum is zero, so that there are
closed magnetic flux paths with in the crystal and no net observable magnetization is
present in sample. The main domain walls are separated by 180° domain walls as
shown in Fig. 6.19, where as closure domains (at the surface) have 90° walls. The
presence of domains and the behaviour of domain walls when subjected to applied field
is a fundamentally important feature in understanding the fundamental properties of
ferromagnetic materials.

Fig. 6.19 Domain structure is formed to reduce the magneto static energy due to flux
leakage.

282
Magnetic Properties
ee

Domains are spontaneously created or nucleated in all ferromagnetic crystals in

order to reduce the magnetostatic energy which is associated with the leakage of
magnetic flux into the surrounding space. When we take a uniformly magnetized block
of a ferromagnet with its associated magnetism, the field lines should be as shown in
Fig. 6.19(a). The leakage of flux can be reduced appreciably if the crystal is divided
into domains as shown in Fig. 6.19(b). The division of domains can not go on without
end. The wall between two domains causes an increase in exchange energy . The
division into domains discontinues as soon as a decrease in the magnetic energy due to
the domain division becomes an offset by an increase in the exchange energy at the
boundary between the
domains. The domain
has a size. The domain
sizein ironis about
0.1 um which is less
than the size of a grain
in polycrystalline iron
sample. The transition
of spin directions from
one domain to another
is not or can not be
abrupt. It is spread Fig. 6.20 Structure of adomain (180°) wall § is the
thickness of the wall
over a certain width.
Therefore, (shown in Fig. 6.20), domain walls have thicknesses and an energy is
associated with them. This is called wall energy and the boundary between one domain
to another is known as Blochwall.

A quantitative analysis of the magnetization curve can be given in terms of

domain movement. Consider a sample with simple domain structure as shown in
Fig.6.21(a). There are three stages of magnetization when the sample is placed in a
magnetic field, the domain whose magnetization vector is closest to the field direction
will grow at the expense of others which are less favorably oriented (Fig. 6.21(b)).
Therefore, the process of magnetization is viewed as the one of domain growth and
displacement of domain walls which are ultimately pushed out of the material at the
magnetization saturation (Fig. 6.21(d)).

283
 Applied Physics

Bomain rotation

Domain
wall motion

(b) (c)
increasing field ——>

Fig. 6.21 Domain growth in a feromagnetic sample

Initially when the applied field is weak, the domain wall motion takes place, but
it is reversible. That is, when the field is removed, the domain boundaries return to the
initial state and the material is demagnatized. At the intermediate field strengths, where
the flux density varies more rapidly, the boundary displacements become irreversible.
Then some domains grow at the expense of others (Fig. 6.21 (b),(c)). The magnetization
remains in the sample even after the removal of the external field. In strong fields, the
domains will also be rotated to line up (Fig.6.21(d)) in the external field direction to
produce saturation magnetisation. The entire crystal may contain a single domain now.
When the external field in reversed, the reorientation does not follow the same path.
The process of demagnetization lags behind the field. This results in the hysteresis
behavior. There is always an energy loss in this process.

6.6.8 Magneto crystalline Anisotropy :

The magnetic substances are crystalline in nature. The observed magnetiz
ation
in a ferromagnetic crystal depends on the magnitude of the field and
also on the
crystallographic direction along which the magnetic field is applied. The
dependence
of magnetic properties on the crystallographic direction is known as
magneto crystalline
an isotropy. In a crystal cell, the number of atoms along different
directions and on
different planes a will be different. Iron has a cubic structure. The importan
t directions
are <111>, <110>, and <100>. Iron goes to saturation easily if the field
is in the direction
284
Magnetic Properties

<100>. The crystals in such directions show saturation at a low field value. This is
shown in Fig. 6.22 for three crystals of iron, and nickel.

Nickel

1
H (104Am* H (104 Amr

Fig. 6.22 Magnetization curves for crystals of (a) iron, (b) nickel, in different directions.

The magnetization along the easy direction produces a hysteresis loop with very
low coercivity and the remanent magnetization, more or less the same as the saturation
magnetization. The other directions for magnetization are called hard axes. The crystal
magnetized in hard direction possesses the highest magnetic energy . The crystal
magnetized in easy direction possesses the lowest energy. The difference between the
energies gives the anisotropy energy. Suppose a,, ,, 0 are the direction cosines for
a direction, the energy of magnetization in a cubic crystal is given as
Beet Ker +K,(a? esnOR se OF a2) + Ko. geosaes

where K,, K,, K, are called magneto anisotropic constants. For iron, the magneto
crystalline anisotropy energy is 1.42 x 10*J/m?. Magneto crystaline anisotropy is of
fundamental importance in determining the magnetic properties of ferromagnetic
materials and hence their stability and suitability for particular applications. For example,
permanent magnet materials require magneto crystalline anisotropy energy to be, uniaxial
and as large as possible so that there is a strong preference for magnetization along a
particular direction, namely, easy direction. It is this property, along with other important
effects, that enables permanent magnets to resist demagnetization in reverse fields. For
high permeability materials, the magneto crystalline anisotropy should be as low as
possible.
6.6.9 Magnetostriction :
The magnetic substances like ferromagnetic or ferrimagnetic materials undergo
a change in dimensions when their state of magnetization is changed by an applied
magnetic field. The magnetostriction arises due to the magnetization change due to

285
 Applied Physics

rotation of the magnetization vector. The magneto strictive effects can be volume or
torsional, or the longitudinal effects. The longitudinal effect is the most important one.
It refers to a fractional change in the length of the specimen in the presence of the
external magnetic field.

The magnetostriction A, is defined as 1, = = , where / is the original length of

the specimen. A/ is the change in length. The effect is strongly dependent on the field
strength. ranges from 10-6 to 10~4 for many substances. Some alloys of ferromagnetic
metals, ferrites have high values of 1. The magnetostrictive effectively results in a
magnetic field controlled variation in Young’s modulus. This variation in Young’s modulus
is known as AE effect. The magnetostriction is anisotropic. When the magneto
striction is that measured when the specimen magnetization is changed from zero to
saturation it is called as the saturation magnetostriction. It is denoted by A,.
Magnetostriction or magnetoelastic effects are fundamental to all ferro and ferrimagnetic
materials. For soft magnetic materials, we require the magnetostriction to be small,
whereas for hard materials, magnetostriction of reasonable value is beneficial.
The magnetostrictive materials can be used as magneto mechanical transducers
for vibrators, ultrasonic machines, sonar, mechanical vibration fitters, sound generators
etc. They require high saturation magnetostriction (i.e. A. > 20 x 10-6), high resistivity
to reduce eddy currents, high magneto mechanical coupling factors, temperature
insensitivity.

6.6.10 Applications of ferromagnetic materials :

Ferromagnetic materials are useful in very large number of applications.
There
are many intrinsic properties for these materials. We have discussed most of
them. Let
us just list out those properties. They are (1) Saturation magnetization M,. (2)
Anisotropy
coefficient K,. K, gives the energy required to rotate the magnetization
vector 90°
away from the easy direction. (3) Anisotropy field, it is the field required
to rotate the
magnetization vector 90° away from easy direction. (4) The thermal
energy destroys
the ferromagnetic alignment in a ferromagnet. The exchange energy
parameter is a
measure of the strength of coupling between aligned magnetic moment
vectors, It is

normally given as * ae kg
where a is the lattice parameter (5) Domain wall energy (6)
Domain wall thickness (7) Permeability. The permeability is defined
from the relation U
= B/H, where B is the flux density produced by the applied field
of intensity H. (The
relative permeability 1 = }1/.,). The word permeability only is
used for all commercial
purposes instead of relative permeability. The flux density
B is entirely due to the
magnetization of the material. Therefore, for high permeab
ilities, we require high
saturation of magnetization. Ideally 11 = oo, for the magneti
c saturation. Soft magnetic
materials have high permeabilities and they possess very
narrow hysteresis loops. 1, is

286
Magnetic Properties
OE SRO SS ee ee

a function of the flux density B. For use in transformer cores, high permeability is
needed and low coercivity is also needed. The permeability calculated from B-H curve
at the point H — O, is called the initial permeability. It is taken as the standard value for
the material. (8) Curie temperature, finally (9) maximum energy product (BH),_.,...
This product is given by the maximum area that can be inscribed in the demagnetization
quadrant of the B-H loop. The available magnetic energy stored by the ferromagnet is
half the (BH). max product.
Ferromagnetic materials are highly useful in a large number of applications,
especially taterials are highly useful, in a large number of transformer cores, and
permanent magnets.

Permanent magnetic materials (Hard Magnetic Materials ) :

The permanent magnets utilize hard magnetic materials. The characteristics of
these hard magnetic substances: They have (1) high saturation magnetization (M,),
(2) high remanence (M,), (3) very high uniaxial magneto crystalline anisotropy energy
K,,.(4) high coercivity. (5) high maximum energy product (BHyy4,x)> (6) High transition
temperature (T,), (7) good thermal stability, (8) high mechanical strength, (9) high
machinibility. Alnico which is an alloy of aluminium, nickel, cobalt and iron.
(Fe-Co-Ni-Al) is the best material for permanent magnet applications. The magnetic
properties depend on the composition. These days rare earth magnets have become
important because large fields can be produced in a small space. The table 6.5 gives the
properties of some permanent magnetic materials.
Table 6.5
B (Tesla)

Nd-Fe-B
Sm(Co-Fe-Cu-Zn), , |
The pernanent magnets are used to produce attractive or repulsive forces without
contact, act through vacuum or non ferromagnetic medium. They do not require any
power. An ideal permanent magnet is one which exhibits rectangular hysteresis loop
with M. = M,. Alnico (Ni 18%, Al 10%, Co 12%, Cu 6%, Fe 54%) is the best permanent
magnetic material.

High Permeability materials :

Very high permeability ferromagnetic materials are needed for transformer cores,
and inductor cores. The permeability values needed are from a few hunderds to a few
thousands. The characteristics of these high permeability materials are (1) soft, (2)

287
 Applied Physics

high saturation magnetization, (3) high remanence, (4) very low coercivity i.e. domain
walls are easily moved or displaced and hysteresis losses are low, (5) very low
magnetocrystaline anisotropyenergy, (6) high Curie temperatures, (7) good temperature
stability, (8) good mechanical strength and non-corrosive nature, (9) high thermal
conductivity. The most useful materials are permendur (alloy of 48% Fe, 50% Co, 2%
V, 4% Si-Fe). The alloy is basically an alloy of silicon, iron and cobalt. Alloy with 96%
Fe- 4% Si, lron-nickel alloys of compositions 68% Ni-Fe, 78% Ni-Fe are useful. They
are called permalloys. Table 6.6 gives characteristics of some high permeability alloys.
Table 6.6 Magnetic properties of soft high permeability materials

ian bea
Si-Fe 96% Fe -4%Si
Psa a] > 7000 1.95
97% Fe -3%Si 9000 | 40000 as

The high permeability materials are useful as relay switches, magnetic shunts,
electromagnets, and indicating meters. The materials with high permeability, very low
or zero remanence, good wear resistance, good corrosion resistance, are useful for the
magnetic recording heads. All the ferromagnetic materials suffer from eddy current
losses. Therefore, they are not useful for high frequency applications.

6.7 Antiferromagnetism:
It has been stated earlier that, if the spin moments of neighbouring atoms
are
oriented in an antiparallel manner, then the antiferromagnetism arises. This
suggests
that the exchange is of negative sign. The magnetic susceptibility of a magnetic
material
is given by Curie-Weiss law at high temperatures. If 0 is negative for any
material, the
material is antiferromagnetic in nature. Landau in 1933 proposed
the presence of
antiferromagnetism in some substances. Some of the substanc
es which exhibit
antiferromagnetism are given in Table 6.7.

288
Magnetic Properties

Table 6.7 Some antiferromagnetic materials or crystals

aa cd ae
16

Some magnetic materials stays in antiferromagnetic state below a certain

temperature, called Neel temperature T,,, named in honor of the French Physicist,
Neel who was awarded noble prize for his work on antiferromagnetism.
The anti ferromagnetic ordering in solids is pictured as composed of two
ferromagnetic sub lattices embedded in each other, interacting via the negative exchange
interaction. The Fig.6.23 depicts the simplest antiferromagnetic structure. What is
shown in Fig. 6.24 is a bec crystal
structure with A atoms occupying
the corner of the elementary cube
and B atoms occupying the centres
of the cube. The structure does not
exist exactly, but it is a representative
model of an antiferromagnetic.
An antiferromagnetic unit cell
consists of we sublattices A and B Fig. 6.23 Antiferromagnetic ordering of magnetic
that have anti parallel magnetic moments in a body centred cubic lattice.
moments. When there is no external
magnetic field on them, the magnetization of the unit cell is zero because the magnetic
moments exactly cancel when summed over the crystal cell. When the sample is
placed in an external magnetic field, small magnetization takes place. Starting from T =
OK hypothetically, the susceptibility increases as the temperature is increased. When
the temperature is equal to T,,, maximum susceptibility is obtained. Above T = OK, the
antiparallel arrangement of spins is gradually upset and the magnetization increases.
Once the magnetization reaches a peak at T,, (Neel point) the spin ordering is lost
completely and the antiferromagnetic material becomes paramagnetic. The

289
 Applied Physics

antiferromagnetism susceptibility can be obtained theoretically from molecular field

theory. But we shall not do it here as it is outside the scope of the book. An
antiferromagnetic substance exhibits a magnetic susceptibility such that the plot of
1/x vs Temperature has an extrapolated intercept —0 on T-axis (shown in Fig. 6.24).

The Fig. 6.24 shows x vs T plots for ferromagnetic, antiferromagnetic and paramagnetic

materials for comparison.

xr
la

-@ On, Or TK) —b

Fig. 6.24 — vs T plot for ferromagnetic, antiferromagnetic and materials

x

Anti ferromagnetics follow Curie-Weiss law in paraelectric state that is about

Ty, - The Curie Law is given as y = C/T +0, where C = np . 9 is the intercept on
3k
the negative side ofT axis as shown in the Fig. 6.24.

68 Ferrimagnetism : Ferrites:
We have seen in the earlier section that if the magnetic moments of sublattic
es in
a crystal cell are equal in magnitude but opposite in direction, they cancel
each other
giving rise to antiferromagnetism in the crystal cell. But there are substanc
es in which
the magnetic moments of the two sublattices are opposite in direction
but not exactly
equal in magnitude (because of the two types of ions in the lattices).
Such crystals
Possess a spontaneous magnetization and most of the properties of ferromag
nets. This
uncompensated antiferromagnetism is known as ferrimagnetism.
The best example
that is known to exhibit ferrimagnetism is the magnetite or lode
stone (Fe,0,). They

290
Magnetic Properties

are several compounds, all of them mixed oxides of metals and transitiol exhibit
ferrimagnetism. The chemical formula of ferrimagnets is MeO, Fe, O,. Where Me is a
divalent ion like Ba, Sr, Zn, Cd, Mg, Ni, CO, or Cu or Zn, Mg or Cd. A large family of
ferrimagnets can be synthesised and used. The ferrimagnets are known as ferrites.
The name of the ferrite is usually ascribed to the name of the divalent ion in the lattice.

6.8.1 Structure of Ferrites :

X-ray diffraction reveals that ferrites have a structure that may be described as
a face centered cubic structure of oxygen ions closely packed together with much
smaller divalent or trivalent ions placed in the interstitial sites. The basic structure is
spinel structure as shown in Fig. 6.25(a), the structure of the mineral spinel (MgAI,0,).
The unit cell contains relatively large oxygen atoms forming a f.c.c. lattice in which
they are 32 octahedral sites (called B sites) surrounded by six oxygen ions, and 64
tetrahedral sites (A-sites) surrounded by 4 oxygen ions. The unit cell contains 32
oxygen ions. Out of the large number of A and B sites,only 24 interstitial sites are
actually occupied by twenty four cations.

.
Fig. 6.25 (a) Spinal structure (b) Inverse spinel structure O 0

Out of these 24 sites, eight are tetrahedral sites, occupied by divalent cations
and sixteen are octahedral sites (0 sites) occupied by trivalent Fe?* ions. The ferrites
having spinel structure of MgA/,O, are known as normal spinels. In some cases, the
Q-sites are occupied by both divant and ferric ions as 8Me2" + 8Fe**, and A sites by
Fe3+ ions. Such structures are known as inverse spinels (Fig. 6.25(b)). The following
table gives the details of the arrangements.

Sate eal A sites (8) | Bsites (16)

Zn Fe,O,,
Mn Fe,0,,
Cd Fe,O,,
2
Inverse 8Fe3* 8Me2t Co Fe,0;,
spinel Fe Fe,O,

291
 Applied Physics

The basic characteristics of the ferrite depend on the distribution of the

Fe3*, and Me?* ions in A and B sites.
The magnetic super exchange interactions between the octahedral B sites and
tetrahedral A sites through the intervening oxygen anions can make the two sub-lattices
to be antiferromagnetically ordered. Depending on the type and the preference of cations,
a large number of magnetic structures are possible. There exist a very large number of
ferrites and mixed ferrites synthesised and used for a number of applications. The
arrangement of cations in different sites is important. Let us take a specific example,
Fe, O,, the magnetite. It can be written as FeO. Fe,O,. The Fe?* ion has five 3d
electrons. The spin number is 5/2 and the magnetic moment should be Sp. There are
two such ions in the formula unit. Therefore, there should be a magnetic moment of
10,13. The Fe** ion has six 3d electrons. The spin number is four and there it should
have a magnetic moment of 4u,. Hence, the total magnetic moment associated with
one formula unit should be 1413. But experimentally found value is only 4.1p,. This
discrepancy is explained on the basis of antiparallel arrangement ofthe iron ions. The
magnetic structure is described as [Fe**] [Fe2*Fe?*] O, The sturcture is an inverse
spinel. The’ spin arrangements are given as
Tertrahedral sites A jepehe thtt et 8Fe3+ § = 5/2
Octahedral sites B Tet |b Vee Ay 8Fe3+ § = 5/2
Vy el S=2
8Fe?*
The structure shows that the formula unit should have 4u, magnetic moment
which agrees well with the experimental value.
The magnetic properties of the ferrimagnetic spinels depend on the interactions
between the ions in the tetrahedral (A) and octa-hedral (B) sites. In some ferrites,
there
is a site preference of particular ions, which is strongly dependent on the temperatu
re
of preparation, the rate of cooling and heat treatment. The cation occupation is given
below for some important ferrites.

CFD. ro faeTT 3
CATT TCH
Titan
F ——— —
MS) |FFL IND Fe*}0,

Mn FeO, 7?
(NS) |Ma*) [Fe*]O,
THTTTt
ttt ttt ttt
ue [Fe™} (CunFe?)
ttt
oo

292
Magnetic Properties
e8
We S ee
of the Ni2* ion is
Let us consider nickel ferrite NiFe,O,. The magnetic moment
unit cell. The magnetic
4u,. So there should be a magnetic moment of 8 x 4p, per
a is the lattice parameter.
polarization J, should be pig. M, = (32p1,)/a> = (),63 Tesla; where
observe d 0.6T. Magnes ium ferrite has an
This value agrees well with the experimentally
uted among A and B sites and
inverse structure. There Fe?* ions should be equally distrib
is zero. Therefore,
Mg2* should be in octahedral sites. The magnetic moment of Mg?" studies
r, neutron diffraction
the unit cell should not have net magnetic moment. Howeve
ions over A and B sites. Due to this
have indicted that there is unequal distribution of Fe3*
Sup = 1.0p, or 8 x lp per unit
difference, there is a magnetic moment p,, = (1.1- 0.9)
t ion may not have preference
cell. Many of the ferrites are mixed spinels i.e. the divalen
prefere nce is based on the internal field
for either octahedral or tetrahedral sites. The
be distrib uted among the A and B
stabilization energy . Therefore, the divalent ion may
cation distrib ution and the magnetic
sites partially. The following table 6.8 gives the accurate
‘moments associated with the ferrite unit cells.
in Ferrites
Table 6.8 Cation distribution of A and B sites

|
value

wads Fell

Mali Fes wasfoss| 10 |10|

1S. |e" [isFe
cigFes0n|

Zinc ferrite unlike all other ferrites

We will just discuss the case of Zinc ferrite.
all the Fe>* ions on the octahedral sites and
has complete normal spinel structure with
ely Zn Fe,O, does not have the magnetic
Zn2* ions on the tetrahedral sites. Unfortunat
ions on the B-sites. Since the magnetic.
moment of 16 5p, per unit due to the Fe3*
, the antiferromagnetic ordering or
moment of Zn?* is zero (it has full 3d!0 shell) coupling on the octahedral
antiferromagnetic
coupling between A and B is replaced by ic
erromagnetic. But zinc ferrite is paramagnet
sites. Therefore, Zn Fe,O, should be antif
and not a ferrite in true sense.

293
 Applied Physics

The ferromagnetism in solids is due to the strong coupling or mutual attraction

of magnetic dipole moments on the atoms in the solids. The origin of coupling behavior
is related to the outer electron interactions of the neighbouring ions. Although the
bonding in oxides is basically ionic, there is certain amount of covalency occurring
between p-orbitals of the filled O2- and the unfilled d-orbitals of the transition metal
cations. This results in a super exchange interaction between outer electrons which
is
responsible for the coupling. This super exchange interaction is characteristic
of the
coupled ions that both have half filled or more than half filled d-orbitals. This
accounts
for both antiferromagnetic or ferrimagnetic ordering in crystals. The Weiss
internal
field theory can be applied to ferrimagnetic materials. There are two sublattices,
therefore
two internal Weiss fields, which are proportional to the sublattice magnetizati
ons.
* Let us assume that the sublattice magnetizations M, and Mg are
due to the
magnetic moments of ions in tetrahedral and octahedral sites respectively.
The internal
fields and the magnetizations are due to exchange interact
ions between A - A ions,
B - B ions and A - B ions. Therefore, the internal fields acting on
A and B are
| gles — XM— ,vMp ; Ba = peicre iIt gm eeraee (6.77)
The internal fields and the magnetizations are vectors. A, Hc,
Vv are all positive
constants incorporating the magnetic constant lo. In the
presence of the fields the
total internal magnetic energy U; is given by the sum of
interactions of the individual
sublattice magnetizations with then effective filed acting
on them, i.e. the sum of the
scalar products. Then, U; = 7l AM* + Rie
]
Mi, - VM,Mg.
The ‘zero’ of energy is taken as that when the two sublatt
ice magnetizations are
zero i.e. M, = Mz = 0, then an even lower energy (i.e.
negative energy) can be obtained if
VM, Mp >>I (imi ae) ee ee ee (6.78)
This expression gives a criterion for ferrimagneti
sm. The two sublattices are
anti ferromagnetically aligned with respect to each
other; then, there is a consequent
lowering of the total magnetic energy of the
crystal cell. The net magnetization
is M, — Mg. Suppose, the Curie constants for
A and B sites are C a and C,. There are
two C's for the two sublattices because there are differ
ent cations on the two sublattices.
Let us consider that only the A - B interaction
is present in the crystal.
Then B, = — vM, and B =— vM,.
_ Then according to mean field approximation,
in the presence of the applied field
or in the presence of a magnetic induction B
due to applied field,
MaT=C,(B -vM,);M3T=Cy(B—VM,) oss (6.79)
Solving for M, and Mz we get and taking T, = v(C,
Cy)!?, we get
) ae H
Se fe) To aieiesisieitie

294
Magnetic Properties

This is a complicated equation for y i.e. the variation of magnetization with

temperature is complicated compared to that in ferromagnets. The variation of magnetic
polarization J. is given in Fig. 6.26(a) for few ferrites.

. |
SSNS T(K) —> Th) ——>-
(a) (b)
Fig. 6.26 (a) Temperature variation of the saturation magnetization for some spinel ferrites.
(b) 1/y vs Temperature for a ferrite.

The temperature variation of 1/x with T is given in Fig. 6.26(b). The plot is not
linear exactly. It has a curvature near the transistion.

6.8.2 Characteristics of Ferrrites :

Ferrites have many properties that are common with ferromagnetic materials.
An important difference is that the spontaneous magnetization is the resultant of two
oppositely magnetized sublattices, each of which is ferromagnetically ordered. They
have become established as one of the most important class of present day materials
and are widely used in linear, digital and microwave branches of electrical, electronic
and communication engineering. They have a very wide range of properties that are
useful in many applications.

1. They exhibit the nonlinear magnetization curve. They exhibit the hysteresis loop.
There is a characteristic retentivity and coercivity for each ferrite. The commercial
ferrites are in ceramic form i.e. polycrystalline hard samples. The coercivity,
retentivity and permeability depend on the grain size of the polycrystalline samples.
Therefore, the preparation of ferrites is a very important aspect in acquiring the
required properties. The crystallites are divided into domains separated by domain
walls. The number and arrangement of domains in a crystallite is such that the
sum of energies, mainly the wall energy and the demagnetizing field energy, is a
minimum. The domain wall movement is responsible for magnetization, as in
the case of ferromagnets. Fig. 6.27 shows various types of hysteresis loops
observed in Ferrites.

295
 Applied Physics

Fig. 6.27 Hysteresis loops of ferrites.

2. Ferrites possess high permeability values normally. But some ferrites can be
made to have low permeability values. The range of penneabilities extends from
15-20 for nickel ferrite to several thousand for some manganese-zinc ferrites.
The permeability is defined as p, = B/p, H. As the field strength or flux density
increases from zero, the permeability rises until it reaches a maximum value at
the knee of the magnetization curve. From there, it falls progressively as the
field is increased. At high fields, B-H relation is nonlinear. The initial permeability

tgs MTS
[dB : eas
}, under the condition Hi H50 p, dH is taken as the characteristic of the

material. 1; is a function of temperature. The saturation magnetization and magneto

crystalline anisotropy co-efficients are both strongly dependent on the
temperature. The anisotropy energy is proportional to the square or higher power
of the saturation magnetization, so the 1, increases with increasing temperature
and then, at the transition temperature, it suddenly falls to a very low value
(paramagnetic state).

3. Another important property of ferrites is their high resistivity . Being oxides,

they are insulators. The resistivity values range from 107 — 10!° ohm-m. Ferrites
have semiconducting nature. But, a ferrite is not a part of any physical electrical
circuit. The resistance of a ferrite decreases very rapidly with the increase of
temperature. Ferrites are widely used in many applications such as transformer
cores or inductor cores. They are subjected to alternating field bias. The
permeability ofa ferrite is frequency dependent and is complex. It is given as
. = p' — jp", where j= \/_]. p' is the real component of the series complex
permeability. 1" is the imaginary component of the series complex permeability.
The finite resistivity gives rise to losses. The magnetic field component in the
time varying field will induce surface currents in the ferrite. These circulating

296
Magnetic Properties

currents are referred to as eddy currents. They cause unnecessary heat losses.
The eddy current loss can be expressed in terms of tan 6 = u"/u'. The magnetic
field intensity H varies as H = Hy exp(jot). The magnetic flux density B has a
phase difference 8 with respect to the field intensity H. Ferrites give rise to very
are
feable eddy currents. In ferromagnetic materials, these eddy current losses
appreciable.
ferrites.
4. Soft Ferrites applications : The practical ferrites are always mixed
spinel, MnFe,O, +
There are three types of mixed ferrites. 1) Mixed normal
Mixed inverse
normal spinel ZnFe,O,. They are called Mn - Zn ferrites. 2)
Ni - Zn
spinel, NiFe,O, and normal spinel ferrite ZnFe,O,. They are called
inverse
ferrites. Several compositions of these ferrites are possible. 3) Mixed
continuo us solid solutions . Ni - Co ferrites
spinel ferrites. Inverse spinels form
at will and the required material with
are examples. The properties can be varied
mixing of the ferrites. In the
needed properties can be obtained by proper
coercivit ies vary from
Mn - Zn ferrites, the saturation magnetic polarization and
Ni - Zn ferrite, the
0.3 to 0.5T and 4 to 100 Am7! respectively. In the case of
The hysteresis
corresponding ranges are 0.1 to 0.4 Tesla and 16-1600 Am!.
on both intrinsic
loop, the permeability and other parameters are dependent
, and composit ion. But, they are
properties of material, their crystal structure
processi ng paramete rs, the grain
also dependent on the extrinsic parameters like
size, density and impurity additions.
magnetostriction
Ferrites possess magnetocrystalline anisotropy. They exhibit
ite Fe,O, has a saturation
under the effect of an external magnetic field. Magnet
, Fe,, O,, has saturation
magnetostriction of +40 x 10-%, cobalt ferrite, Co,
O, has magnetostriction of
magnetostriction as —250 x 10-°, while simple CoFe,
and low losses. The
_110 x 10°°.The soft ferrites possess high permeability
in Table 6.10.
characteristics of Mn - Zn and Ni - Zn ferrites are given
Table 6.10

HL, Saturation Curie Coercivity

Ferrite
magnetization Tesla} temperature T, Am"

500-20,000 0.3-0.5 100-250°C 4-100

10-2000 0.1-0.36 100-500°C | 16-1600

grades, each grade having the -

The mixed ferrites can be made in a variety of
ation. The applications were
properties that suit it to particular class of applic
combination of good magnetic
initiated in the field of carrier telephony where the
als very suitable as cores for
properties and high resistivity made these materi
are used in television deflection
inductors and transformers. Mn - Zn ferrites

297]
 Applied Physics

yokes, cores for television line output/eht transformers (permeability should be

in between 2000-5000) and standard power supplies. They are useful ‘as
induction cores, antennas for medium and longwave broadcasting (permeabi lity
should be 500-1000), pulse transformers (permeability should be 3000-20,000).
Ferrites are widely used in transductors (variable inductors). They are useful in
tuning, automatic frequency control systems, frequency modulation, switching,
filters, etc.
Ni - Zn ferrites are also useful in wide band transformers (frequencies 1-300MHz,
permeabilities 1000), antennas for medium and long wave broad casting, power
transformers cores (frequencies 0.5-5MHz; permeabilities 150-500), inductor
cores, antennas for short wave broadcasting (frequencies 2-30MHz, permeability
70-150).
Some ferrites like Mg,,.Mn,,;Fe, ,O Mg,.,,Mn, Ni, eae,Fe,O t Cu,,,Mn,_.Fe,O,,
Mag
LE pee (NG el ea oe CO. Wink. akeF e,O, give rectangular hysteresis loop. They
are useful in the construction of computer memory system for rapid storage
and retrieval of digital information by switching the direction of magnetization
in
a very small sized toroidal cores. The squareness of the hysteresis loop is defined
by B/B,,,. It is necessary that the ratio should be close to unity . These
ferrites
should have very low coercivity .

- Microwave ferrites : Garnets, such as yttrium iron garnet y,Fe.O,

, and few
other ferrites like lithium ferrites are usefulinmicrowave applicat
ions. Microwave
ferrites are those ferrites which are used at frequencies in the
range 100MHz to
500GHz. The fundamental aspect associated with ferrites is their
nonreciprocity.
The permeability of the ferrites exhibits anisotropy or the magneti
c susceptibility
is a tensor. Ferrites possess an internal field. When an electro
magnetic wave in
the frequency range of microwaves is incident on a ferrite,
it is spontaneously
magnetized. The ferrite is initially kept ina magnetic field.
The aligned magnetic
moment vectors of spins precess about the magnetic field. If
the electromagnetic
wave that is incident is a plane polarized wave, the plane
of polarization of the
wave is rotated by an angle @. When this wave is made
to traverse back in the
ferrite, the wave suffers another rotation by an angle
0. Therefore, the ferrite is
a nonreciprocal device. This property is made use in
a variety of devices like
isolators, circulators, Switches, etc. An isolator is
a device in which the incident
electromagnetic wave can propagate forward, while
there is no reversewave. A
circulator is a device that has n ports which are inputs
and output terminals. A
wave incident at some port like port 1 emerges at port
2. A wave incident at port
2 may not go into port 1, but may emerge from
port 3. Thus the circulator
provides easy isolation between different channels
in a microwave system such
as a radar, or transmission linkage. The most importa
nt ferrites that are useful
are Mg-Mn-Al, Ni-Zn-Al, etc.

298
Magnetic Properties
oe eh Sa

6. Garnets : Garnets have a magnetic structure c|Me?* B|Fes A|Fes Jo,

the net magnetic moment per formula unit is fi. 2 Soi oes SI.
where m, is the magnetic moment of Me** ion (rare earth). The garnets have the
general formula Me,Fe,Fe,O,,. The rare earth ion selected is yttrium. There are
three sites A, B, C. B is octahedral, A is tetrahedral and C site is called skewed
cube polyhedral site. The discussion on the details of the structure is beyond the
scope of this book. A thin film of ferrimagnetic garnet give rise to bubble domains.
The bubble domains are the cylindrical
domains of reverse magnetil2tions
bounded by a vertical domain wall
surrounded by regions of magne tization
and sustained by a bias magnetic field. -
The most important property associated Fig 6.28 Magnetic Bubbles or Bubble
with the bubble is the flux gradient domains in a thin plate ofa granite
across the bubble. If the bubble has to
move, there should be a drive field and this drive field is greater than the coercivity
of the bubble wall. The bubble should have very low wall coercivity. The reverse
domain (Fig.6.28) are mutually repulsive and can be moved through the material
by very small fields (< 100Am-'). The absence of reverse domain may be referred
to as zero.
The presence of reverse domain may be taken as the stored binary digit 1. Thus
a single-crystal platelet of any garnet could be used as the fundamental element
of a computer store. Since the reverse domains can be made very small
(Sp, diameter), one square millimetre of garnet, for example, could store 100000
bits of information.
7. Magnetic Recording : Magnetic recording of audio, visual and digital information
is now one of the most important applications of magnetic materials. The
recorded information stores in the form of remanent magnetization of the material
such that the magnitude of remanence is proportional to the amplitude of the
original electromagnetic input signal. Thus, the material with linear remanent
magnetization is needed. As we have to retrieve the recorded information, the
remanence should be high to give noise less-play back signal. A material with
high coercivity and high magnetization are needed for recording purposes. The
amplitude of ac bias is greater than or equal to the coercivity of the tape material.
Ofcourse, it should be mentioned that the recording process is very complex
and the materials properties play a vital role. The oxides of y-Fe,O,, Fe,O, and
CrO, are the most suitable materiats. Most widely used material is cobalt modified
y-Fe,O, and y-Fe,O,. CrO, is an unusual oxide compound and it is ferromagnetic
rather than ferrimagnetic. Chromium ion is in +4 valence state with electronic

29
 Applied Physics
ce a

configuration of [Ar}” . Hence, there is a magnetic moment of 2p, per ion.

Wherever high recording densities are required, CrO, tapes are preferred. High
quality tapes of audio, video, data tapes and disks are available commercially.
8. Hard Ferrities : Permanent magnetic materials or hard magnetic materials are
characterized by high coercivities and high remanent magnetizations, so that
their hysteresis loop enclose the largest area. The principal function of a permanent
magnet is to supply magnetic flux or magneto motive force. The hexagonal
ferrites are used. The most important ferrites under this class are barium ferrite
(BaFe,,O,,) and Strontium ferrite (SrFe,,O,,) and their combinations. The ferrite
compacts of high density, can have coercivities of the order of 400KAm"! in
combination with a remanence as high as 0.2 Tesla and an energy product of
6.8KJm=. They are the most important permanent magnets in terms of their
large scale production as well as monetary value of scales. Since ferrites are
insulators, they are useful in dynamic applications such as electric motors. They
are used as flat rings for loud speakers, wind screen wiper motors, door catches
(example, refrigerators), small de motors, correction magnets for TV.
Exercise 1 : The magnetic susceptibility of slicon is -004 x 10°. What is the flux
density and the total magnetic moment per unit volume in a magnetic field of intensity
10° amp m".
B=, + M)=p,HC +x)
Me Wi
=—(.4 x 10° x 10° =—0.4 amp m".
B = 4n x 107 x 10°(1 — 0.4 x 10°)
=4n x 10? x 0.99996 = 0.124 wb m”.
Exercise 2 : The atomic weight of gadolinium is 157.25. Its density is 7.8 x 10°Kg/m’.
It is ferromagnetic below 17°C. What is its saturation magnetization.
Avagadros number N, = 6.026 x 107°/Kmol
M, = 2p,N
_ Naxp _ 6.026x10" x7.8x10°
N = 28.89 x 10?’/m?.
A Lf.)

”. We know that m, = 9.27 x 10-4 Amp m?

M =2 x 9.27 x 10 x 29.89 x 1027 = 554.2 x 10? Am”.
Exercise 3 : A paramagnetic system is placed in an external magnetic field of strength
10° amp m'. Calculate the average magnetic moment per dipole at 600K.
The number of dipoles unit volume = n.

300
Magnetic Properties

.. The average magnetic moment = M/n.

M/n = poH pe /kT

ly = 40x 10°’ henry/m; pt, = 9.27 x 10-4 Ampm?

H =10° Amp/m;k =1.38x10°°

JK! ;T =300K .

—M _ 4nx1077 x10° x (9.271077 r

Lin
n 1.38x1077>
x 600

= 1.3x107°Amp m?.

1310"
In terms of Bohrmagnetons pL, = 927x104

= 1.4x10~ Bohrmagnetons/dipole.

Exercise 4 : Iron crystallines in body-centered cubic form. The lattice parameter is

2.9 x 10-°m. Iron has a transition temperature of 765°C. The number of effective
atom in the unit cell is 2. Calculate the saturation magnetization, Weiss field constant
and effective internal field intensity.
M, = 2p,n, vion has moments of two Bohr magnetons per atom.

i ee
a’ (29x10
MeH2 x 927 1077 x 8.2.x 10" = 15.03 = 10° Amr".

Nth
The Curic constant C =
k

28 x4x 22x10 -7 '\9.27x1 -24 = 0.68.

;
_ 82x10 .27x10
(AEEeee

1038
m'
Effective field intensity H, = YM, = 068 * 15.03 x 10°=2.29 x 10° Amp

Internal magnetic flux density B, = m,H; = 2878.8 Tesla.

301
 Applied Physics

Questions
An elctron revolves around a proton in a circular pat of radius 0.35 x 10-!°m.
What is the magnetic field at the proton ?
State the Maxwell's equation for free space.
What is Bohr magneton ? A certain element, that has an atomic weight 24.31,
has a density 1.74 x 10°kg/m’. In the presence of the magnetic field, all the
atomic dipoles align perfectly with the field. What is the magnetic moment ?
Explain the origin of magnetic moment in substances.
What is diamagnetism ? Give some examples. Explain the theory of
diamagnetism.
What is the diamagnetism susceptibility of silicon if the electron radius of an
atom is of the order of 0.1nm. (The experimental value is —0.4 x 10°).
Explain the difference between diamagnetism and paramagnetism.
Explain the Curie law.
Describe how does the magnetic susceptibility vary with temperature for (a)
Paramagnetic salt, (b) Diamagnetic material, (c) Ferromagnetic substance.
Discuss the salient features of Weiss theory of ferromagnetism. Discuss the
expression for Curie constant and Curie temperature.
What is spontaneous magnetization ? Explain the origin.
A ferromagnetic material is placed in a magnetic field of intensity H. What is the
energy stored in the material.
Discuss the spin nature of ferromagnetism. Explain why Fe, Co and Ni are
ferromagnetic in nature ?
What is the hysteresis loop ? What does it represent ? What is the significance
?
Describe how it is recorded ?
Explain the magnetization curve for a ferromagnetic material on the basis
of
domains.
Nickel is subjected to a magnetic field intensity of 5 x 10°A/m. It gets magnetic
ally
saturated. If B = 1.27 Tesla, relative density is 9 and atomic weight
is 58.71.
Compute the magnetic moment per atom.
Silicon is subjected to a field of intensity 1000 A/m. The magnetic suscepti
bility
of silicon is -2.9 x 10-5. What are the magnetization and flux density
values ?
17. Discuss Curie-Weiss law for the susceptibility of a ferromagnet.
18. What are magnetic domains ? What role they play in the hysteresis
behaviour of
a ferromagnet. Give the characteristics of Domains.

302
Magnetic Properties
Se Se a ee

13: A polycrystalline iron rod does not show any magnetization. Explain.
20. What are high permeability substances. What are their characteristics?
PAWe What should be the characteristics of permanent magnetic materials?
22, What is anti ferromagnetism? Discuss its origin in some crystals. Give
one example?
23% Give the concept of ferrimagnetism?
24. What are ferrites? Discuss the salient features of normal spinnel and
Inverse spinnel structures?
23. Describe the magnetic structures of few ferrites.
20% Discuss the application of Soft ferrites?
27. What are Hard ferrites?
28. What should be the characteristics of recording materials?
eB What are magnetic bubbles?
30. How does the permeability of a ferrite change with temperature. How does its
magnetization change with temperature?
Evie What is magnetostriction? What are its applications?
neDie What will be the induced magnetic moment in the hydrogen atom ina fluxdensity
of 1 weber / m??
SS. A magnetic material is subjected to a magnetic field of strength 200Am. Its
relative permeability is 18. Calculate the magnetization and the fluxdensity.
34. The magnetic fluxdensity in a piece of Ferricoxide is 1.4 Tesla. The magnetic
susceptibility at room temperature is 1.6 x 10°. Calculate the magnetization,
magneto motive force, and the relative permeability.
Soe Where are permanent magnets useful?
36. Discuss the general characteristics of ferrites?
cee Discuss the general characteristics of ferromagnetic materials?
38. How does spontaneous magnetization vary with temperature?

303
vi
SEMICONDUCTORS
OOOO

7.1 Introduction:

Electrons ina solid are a collection of large number of interacting particles. It is

impossible to study the dynamical behavior of such as electron gas in a solid by
considering each and every individual electron and then trying to apply the basic
principles of Physics to such a system. In other words, the microscopic approach of
understanding the behavior of electrons in a solid is not practical. However, a
macroscopic approach involving statistical principles will help in predicting the probability
distribution of electrons and currents due to electrons. This approach gives average
values of the properties of all the electrons under study rather than the contributions of
individual electrons.
In order to determine the value of energy of an electron in a solid, it is necessary
to know first of all how the energies are distributed throughout the concentration of
electrons. The types of statistics developed to describe such distributions depend not
only on the type of particles present but also on the possible interactions between them
and their surroundings. Electrons interact with each other quantum mechanically.

304
Semiconductors
a

Electron occupation in different available energy levels is in accordance with Pauli

exclusion principle which allows only one electron to occupy a particular quantum
state. Such an ensemble of electrons when treated statistically yields a distribution
function called Fermi-Dirac distribution function and it is applicable to the electrons in
metals and semiconductors and helps in understanding the electrical phenomena in
these materials.
There are three distribution laws determining the distribution of particles among
available energy states.
1. The Maxwell Boltzmann probability function: It is derived by considering the particles
as distinguishable by numbering each of them (for example, from 1 to N), with no
limit to the number of particles allowed in each energy state. The behaviour of gas
molecules in a container at fairly low pressure is an example of this type of
distribution.
2. The Bose-Einstein probability function: It considers the particle as indistinguishable
and again, there is no limit to the number of particles permitted in each quantum
state. The behavior of photons, or black-body radiation, is an example of this law.
3. The Fermi-Dirac probability function: It assumes that the particles are again
indistinguishable, but now only one particle is permitted in each quantum state.
Electrons in a crystal obey this distribution.

72 FermiEnergy:

Particles like electrons or holes follow Fermi-Dirac statistics and hence they are
called Fermions. The Fermi-Dirac probability distribution function gives the probability
of occupancy of energy states by fermions and is expressed as

[ I
PYfea terres OE Sr gris
e EL. |tex( BoB /KI
iawe &C : F)
E is the energy state of the electron. E, is Fermi energy.
The Fermi Dirac distribution function is valid for only system of particles obeying
Pauli exclusion principle. The P(E) value, since it is probability, can never exceed
unity.
Hole in a semiconductor is essentially the absence of an electron. Hence the
probability distribution of holes can be derived from the probability of unoccupancy:

1
aN) Fmt ee 0) oe 1+exp(E,; — E)/kT

305
 Applied Physics
en ee

Lf, Beaks thenPR. sib idly

Hence, Fermi energy can be defined as the energy state for which probability of
occupancy by an electron or hole is half at T= 0K
Pol or pbs.
= (0 for BE.
Hence at 0 K all the energy states below fermi level are filled and levels above
fermi level are vacant (Fig. 7.1(a)). No electron in the lattice can have energy higher
than fermi energy at 0 K.

Ep

70
P(E) =>;

(a) (b)
Fig. 7.1 Fermi energy in semiconductors

As the temperature is increased, the electrons, which were originally at levels

closer to Fermi level E,, can now occupy states for which E> E, . At low temperatures,
there is a finite but small probability that electrons will occupy available states for
E>E, but the probability rapidly decreases with increase of energy (Fig. 7.1(b)). As
the temperature is increased the tail of the probability distribution curve becomes more
and more pronounced. The probability distribution function is symmetric about the
pointE=E, and P(E)= for E=E, at all temperatures.

The probability of existence of unoccupied levels below fermi level is same as the
probability of occupied levels above fermi level. Thus the probability distribution curve
would be symmetric about fermi level (Fig. 7.1(b)). The fermi energy E, in most of
the solids is about few millivolts and the magnitude of kT is also a fraction of electron
volt at ordinary temperatures. (for example kT = 26 mV at 300 K).

7.2.1 Electron and Hole Densities :

FD statistics discussed in the previous section can be used to find out electron
and hole densities in a solid. Electron/hole density is the number of electrons or holes

306
Semiconductors

per unit volume of the solid. Fermi Dirac distribution function gives the probability of
occupancy of a particular energy state E at equilibrium.

73 Intrinsic Semiconductors:

If a semiconductor is sufficiently pure, then it is known as intrinsic

semiconductor. At low temperatures all the valence electrons are bound to the atoms
through covalent bonds. But at higher temperatures when thermal energy (kT) becomes
comparable with the bond energy, some of the covalent bonds will be broken and
electron hole pairs will be generated. The electrons can occupy conduction band and
holes will be created in the valence band. Each broken covalent bond can generate one
electron and a hole. Thus the electron hole density in an intrinsic semiconductor will
be equal and will be depending on the number of broken covalent bonds. The reason
for rapture of covalent bond can be anything besides thermal excitation. Thus electron
density n = p = hole density = n, = Intrinsic density.
Under equilibrium conditions, the number of electrons per unit volume (n) present
in the conduction band will be equal to the number of holes per unit volume (p) in the
valence band.
The number of electrons dn, per unit volume of a solid in the energy interval dE
between the energies E and E + dE is given as
dis pdb =U ee tees (7.3)

p, is the density of electrons in the energy interval dE and is given by

Or NIE) PAR) en tree (7.4)
N(E) is the number of states available per electron volt per unit volume centered
around E and P,(E) is Fermi Dirac distribution function for electrons.

The density of states N(E) is proportional to the square root of energy and is
given by
N(E) = nE!?
Ant * z :
where 1 = —+(2m; is the proportionality constant

Here m, is the effective mass of the electrons

N (E) = = (om; PE?

4n * 2 is
nse (7.5)
N(E) is the number of electrons per unit volume and unit energy.

307
 Applied Physics
sm a

Therefore,

An 7 ) oe
afoul eS 7.6
ans fee? Phe(E E/T ee
Integration of this equation gives the total number of free electrons per unit
volume of the solid. At T = 0 K, P(E) = 1, hence n is given by

Ep
n= ie$ = 5 lom;PBpa tee: (7.7)

Ep -3% bmi} ?n,

The above equation indicates that Fermi energy in a solid is dependent only on
free electron concentration n, at 0K.

2
32s plage ie 2
ae ea (om2)'n3 =3.64x107Pn3 a (7.8)

Fermi Level in Intrinsic Semiconductor :

To find out the concentration of electrons in conduction band we have to integrate

eq. 7.6 from E, to oo. In such a case we are assuming the origin to be at the edge of
the valance band.

: - (E-B, )2dE
foo)
4n
n, = J—7Gm; P[crexp(i, -EVET]~ (7.9)

E, = energy of conduction band edge ;

E, = Fermi energy.

IfE, —E, >> KT, one in the denominator can be neglected in comparison with
exponential term and Eqn. 7.9 can be rearranged

Let y - “om:) rab. (7.10)

308
Semiconductors

n=yexp|(E, - Ee)er] (©= E.)2 exp(E —E,)/ kr foe

ae
= yexp|(Ec —E,)/kT\kT)2 |x2e*dx
0

(Here x = (E — E, )/kT)

= yet) exp[- (Ec -Ep)/kTa

HeeN expl-(Be—
be KT) eneess (7.11)
; 33
N, = t)2 SE= AT? SS (7.12)

N,, is the effective density of the states in the conduction band.

Equation (7.12) gives the concentration of electrons in the conduction band as
a function of temperature. The proportionality constant N, is called the effective
density of states.
Similarly the hole concentration in the valence band is given by

= he. x Bas exp|-(E,; —E)/kT}E

pi=iN. expe (Ee=EAQiKE| © in waa (gals)

2nm, kT |2
where Ny =2 ata A oy (7.14)

Here Ny, is the density of states in the valance band. m, is the effective mass
of hole.
For an intrinsic semiconductor n = p, hence

N,,exp[- (Ec - Ex)/ kT]=Ny exp[-(E, -Ey)/kT]

309
 Applied Physics
es EE

Na _ expl- (Ey - Ey KT
N, q exp|- (Ec —E,)|/kT

= exp (- E, + Ey + Ec - Ep /kT
= exp. (6, + E72 BE, WkT

Eee Es @ kh WN
E, = —2— ; 5 in
[) pk (7.15)
gs)

Neemme
N, fi eqns. 7.12
(from and
7.12 and 7.14
7.14)

If the effective masses of electrons and holes are equal, then N, = N,, , the
effective density of states in conduction band and valance band is equal in such a case
one gets

Ec +Ey
E, = B)ne Snare

i.e., Fermi level lies exactly in the middle of energy gap (Fig.7.2). Otherwise fermi
level varies with temperature as per the above eqn. (7.15).

Conduction Band

Valence Band

Fig. 7.2 Fermi Energy in Intrinsic Semiconductor

The product of equilibrium concentration of electrons in conduction band and

holes in valance band at a given temperature is constant and is equal to the square of
intrinsic concentration. This is called law of mass action. In an intrinsic semiconductor
electron concentration is equal to hole concentration, but in semiconductors doped
with impurities (extrinsic semiconductors) this is not the case. However, the product
of electron and hole concentration in both types of semiconductors in always constant

310
Semiconductors

and it is always equal to the square of the intrinsic concentration. Product of electron
and hole concentrations np is :
np =n, =N, N, [exp— (E,- E,)/kT] [exp — (Ep — E,)/kT]
TaN NNexpiE
+E, kT]

3 : ‘ 3
2nkT
h? (m:m*) exp(-Eg/kT) ......, GAT)

where E, = E, — E, is the band gap of the semiconductor.

Hence, taking square root of the equation (7.17)

= Al “exp(— Eo/2k]) (ee Sneeeorerl (7.18)

3
2nkm,m, "7
Here A=2 hOpReeS te th (7:19)

The above equation implies that intrinsic concentration of electrons in a

semiconductor is independent of Fermi level. It is also evident that it depends strongly
on temperature as square root of T? and on energy gap, E =

7.4 Extrinsic Semiconductors:

It has been learnt (Chapter 4) that the metals are characterized by overlapping
valance and conduction bands. Insulators have very large forbidden band. Energy gap
of semiconductors is small so that sometimes even thermal excitation can transfer
electrons from valance band to conduction band in these materials. However, it is
extremely difficult to purify semi-conducting materials to high degree of purity and
they always contain certain impurities, however small their concentration may be. The
electrical properties of semi conductors are extremely sensitive to the impurity content
even if their concentration is as small as one impurity atom in 10!! semiconductor
atoms. The pure semiconductors are interesting from theoretical point of view.
A pure semiconductor has high resistivity and is not of much practical significance.
Si, Ge and other materials which have effective valancy four, can act as semiconductors.
A broad classification of semiconductor materials is presented in the table 7.1. Si is the
most popular semiconductor material for various reasons : it is available abundantly as

311
 Applied Physics

silicates which form 25% of earth crust. Energy gap of Ge at room temperature
(300K) is 0.7leV and that of Si is 1.leV. Larger band gap in Si allows it to be used at
higher temperatures. Devices made with Ge cannot be used above 80°C. It is very
easy to oxidize Si to SiO, when compared to Ge. Oxidation of Si is important for
surface passivation in integrated circuit fabrication.

Table 7.1 : Some Common Semiconductor Materials

Type of Semiconductor Example Energy Gap

at 300K (eV
Elemental Si el

I-IV (A! B!Y) Binary alloy CdSe 1.70

Il — V (A™ BY) Binary alloy GaAs 1.42
Teritary (Al Bill CY) HgCdTe 0.12
Narrow band gap PbTe 0.31

In order to control the electrical properties of semiconductors it is necessary to

add impurities. Addition of a small percentage of impurity, in parts per million (PPM)
levels to an extremely pure (99.999% pure) semiconductor is called doping. Addition
of desired impurity to group IV elements from either group V or group III elements of
periodic table produces n — type or donor type and p — type or acceptor type
semiconductors respectively. Extrinsic or doped semiconductors are important in a
large number of device applications. The number of charge carriers, their type and
electrical properties such as conductivity, capacitance etc., are extremely sensitive to
the doping concentration. Doping of impurites creates energy states with in the forbidden
band. The states of n— type impurity will be close to the conduction band and those of
p — type impurity will be close to the valence band (Fig. 7.3).

(a) (b)
Fig. 7.3 Impurity Levels (a) p — type semiconductor (b) n — type semiconductors

Je
Semiconductors

Table 7.2 : Impurity levels in p — type and n — type semiconductor Silicon

Energy of impurity level | Impurity] Energy of impurity level

above valance band eV below conduction band eV
Sb 0.039 B 0.045
Pp 0.045 Al 0.067
AS 0.054 Ga 0.072
In the energy band diagrams, donor levels are just below conduction band
and acceptor levels are just above valance band. Thus donor and acceptor levels in a
doped semiconductor lie in the forbidden region and are discrete and highly localized
levels. At 0 K all the donor levels are filled with electrons and all the acceptor levels are
empty, but at room temperature and above all the donor levels are empty and acceptor
levels are full. (Fig. 7.4)

eo Ec
Ep
Eq Ey e - Electrons Ey
o - Holes

(a) (b)
Fig. 7.4 (a) lonized Acceptor Levels and (b) Ionized Donor Levels

In case of doping with pentavalent impurities like phosphorous, arsenic or

antimony, four of the valance electrons are enough to form covalent bonds with adjacent
silicon atoms. The fifth one occupies a state below the bottom of the conduction band
by an amount equal to ionization energy values of the impurity atom. These energies
for some of the semiconductors is given in the Table 7.2. The impurity is known as
donor and it donates an electron to the lattice of semiconductor. The semiconductor is
then called ann—type semiconductor. Electrons are majority charge carriers in them.
Similarly acceptor impurities like boron, aluminum or gallium can contribute only three
electrons to the lattice for covalent bonding. By accepting an electron from the lattice,
the fourth covalent bond is formed. Hence these materials are called acceptor impurities.
And materials with these impurities are p — type semiconductors. Thus the impurity
ions will be ionized in both types of semiconductors. p — type semiconductors inject
holes into the lattice and holes are majority carriers in these. Energy levels of acceptor
n
impurities lie just above the valence band. In a two dimensional representation the
and p — type semiconductors are represented as shown in the figure 7.5.

313
 Applied Physics

(S)

‘Slits en eae
@ - Si** ion
= Covoent bond

Fig. 7.5(a) Two dimensional representation of semiconductor lattice

al L_|elail QO Pentavalent ton

=H
ialmal
pls @ Electron

mi
OO Sie excess electron

ee es
Fig. 7.5(b) Two dimensional representation of n— type semiconductor lattice

| = ELaI
se & trivalent ion
<3CI
) el[S O hole

=a
Hesee excess hole

no eo
Fig. 7.5 (c) Two dimensional representation ofp— type semiconductor lattice

7.5 Carrier density in extrinsic semiconductors : Condition of

charge neutrality :

The electrical properties like conductivity of a semiconductor can be controlled

by controlling impurity concentration. In an extrinsic semiconductors the charge
carriers available are electrons, holes and ionized impurity atoms. Donor atom will be
positively charged since they give electrons and acceptor atoms will be negatively
charged. Therefore the extrinsic semiconductor will be electrically neutral only if

ns Nop aNe

314
Semiconductors

n, p, Nj,N‘4, are the concentrations of electrons, holes, ionized acceptor and

donor ions respectively. We already know from eqn. 7.11 that
n= No exp [— (Ec— E, kT]
The product of acceptor concentration and probability of finding electron in
acceptor level give the concentration of acceptor ions, if N, is the concentration of
acceptor ions and E, is the energy of acceptor levels, then from equation analogous to
7.1 we can write

N,=
a eer
1+ explE, eee 26
-E,)/kT] sheen Che)

Similarly, for donor doped semiconductor, one gets

Se Nay |
: ; 1+exp|(E, —~E,)/kT|

Na
C22)
Deep eRe vir ed el Se goad
In an n— type semiconductor, the acceptor atoms and thereby the concentration
of holes is extremely small as compared with the concentration of electrons and ionized
donor atoms in the lattice i.e.,n >> p. At 0 K all the donor levels are occupied by the
electrons. When the temperature is increased from 0 K, the electrons form donor
levels jump into the conduction band leaving some of the donor levels empty. The
temperature determines the position of the fermi level (fig. 7.6). In a semiconductor
doped with donor atoms, only N, = 0 and concentration of electrons is much larger
than the concentration of holes, hence from eqn. (7.19(b)), n = Nie

Fig. 7.6 Effect of impurity doping on fermi level.

SIS
 Applied Physics
eer ent eeere
ve

Therefore, from eqns. 7.11 and 7.22, it is evident that

Nc exp [- (Ec — Eg \VkT] = eee) eee

1+ exp|- (E, - E,)/ kT]
ie: Ng = Ncexp [- (Ec — Ep WkT]
[Eorexp "(E, SE yk i (723)

ee= exp[-(E, + E,)/kT] exp(2E; /kT)

Cc

+ exp(—E,/kT)exp(E,/kT) — ....... (7.24)

This is a quadratic equation in E, and its solution is

1
2
it 1+ 44 exp(E¢ -E,)kT}
E,/kT) = -
eas 2exp(- E,/kT)
The negative root is omitted because left hand side of the above equation can
never be negative. There are two cases.
Case I: The doping concentration can be high, or the temperature can be low, i.e.
(N,/N,) exp (Ec — Eq)/kT >> 1. In such a case, the fermi level has to
coincide with the bottom of the conduction band. At these temperatures,
the n — type semiconductor behaves like an insulator. In such a case, the
equation (7.25) reduces to

1
2 re
2 NG exp oe
(;
F ) N 2kT
exp ptt
Cc
A ee

N, |? Eos Ey
posi evened 28) Sraceccage | MN Wyre! 72
| i len

Porte ] N
Bee
F 5 3kT in
ivr ete sol!bepe (7.27)
The equation (7.27) indicates that at 0 K Fermi energy lies in the middle of
energy gap.

316
Semiconductors

E+E
Ep= ee Alias (7.28)

And the egn. (7.27) gives the temperature variation Fermi level position in the
forbidden gap. As the temperature is increased from 0 K or impurity concentration is
changed, the Fermi level starts shifting and the conductivity of the semiconductor
changes accordingly (figure 7.6).

Substituting for E, in equation (7.11)

n= Ncexp(-E, per Be}exp|(E, + Ey)/2kT|

c

2(NgNy)F expl(By=Be)/2kr] 1 (7.29)

Case II: In the second case (N,/N,) exp[(Ec — Ey)/kT] >> 1. The concentration of
the impurity is low such that N, << N, or the temperature is high. Thus
kT >> E,—Ep. Thus, using the binominal expansion of the term in the
radical of equation (7.25) and considering only first two terms, we get

Ey Ec + wy n(
Be) ee (7.30)

Thus, the fermi energy will be slightly below the conduction band and it will
decrease as the temperature is increased.

Substituting for E, from eqn. (7.30), the equilibrium concentration of electrons

in the conduction band over a range of temperature becomes,

“cd Fhe hemega mina eiadimaeanmanmaiminad rt (7.31)

Similar calculations performed for p—type semiconductors give the concentration
of holes in the valance band as
P=Nyexp[(Ey—EpVKT) ene (7:32)

These equations are only approximate. Exact carrier concentrations can be

found out using equation (7.25) for finding Fermi energy and then substituting in
equation (7.21).

317
 Applied Physics

7.6 Recombination :

The broken covalent band due to excitation in semiconductors may result in a

hole in the valence band and an electron in the conduction band. The resultant electrons
and holes contribute for conduction. Electron — hole pairs thus created may be bound
together resulting in excitons. The state of excitation may be brief and they undergo
another transition or may recombine. The recombination of electron hole pairs may be
direct without including any intermediate energy states or through intermediate states
called meta stable states.
In elemental semiconductors like Si or Ge electron hole recombination generally
takes place through one or more intermediate states called recombination centers or
meta stable states. Electrons from the conduction band drop down to these states and
then holes from valence states can recombine with these electrons. The recombination
centers or meta stable states are localized states usually located in forbidden band. The
electrons can also get trapped in these meta stable states. Addition of certain types of
impurities (ex. Gold) to a semiconductor results in recombination states. Hence physical
origin of the recombination centers in a semiconductor is related to impurities,
imperfections and unsaturated bonds near the surface.
The lifetime of an electron in a meta stable state depends on the distance of
this state from conduction band.
In equilibrium, the rate of generation of electrons and holes is equal to rate of
recombination. Recombination processes are :
1. Direct or band to band recombination.
Direct band-to-band recombination is shown figure 7.7 (a) and 7.7(b). In
these, an electron from conduction band recombines directly with a hole of valence
band. There are no intermediate levels involved in these transitions. These are of two
types a) Auger recombination (fig. 7.7 (a) and (b)) radiative recombination (fig.7.7(b)).
2. Indirect recombination, Single-level recombination, Multiple level recombination.
Unlike direct band-to-band recombination, indirect recombination requires a
metastable state, which is in the forbidden region. Impurity atoms doped in the
semiconductor material create these metastable states. These states act as an
intermediate state for the transition ofelectrons or holes. In case if the direct transition
of Electron form conduction band to valence band is not possible, then the transition
can be take place via these metastable states as shown in figure 7.7(c). Depending on
the type of impurity, the metastable states created can be near the valence band or
conduction band.
Indirect recombination processes are further explained in the figure 7.8. If
the metastable state due to an impurity is present near conduction band (figure 7.8a)
then an electron from conduction band can be transferred into this state. Then the
metastable state is acting as an electron trap (figure 7.8b). The electron trapped in

318
Semiconductors

metastable state can combine with a hole of valence band releasing the energy involved
(figure 7.8(c)). Sometimes metastable states can be close to the valence band depending
on the energy of the doped impurity, and then the impurity atom energy levels can act
as hole traps. Figure 7.8(d) shows such hole traps along with the electron traps.

© Ee = easel Ec
Ec
Energy h meta stable
Seca ia Ey Ey state

Dae
a) Auger recombination b) Radiative recombination c) Indirect recombination
Fig. 7.7:

In Auger recombination electrons and holes recombine without involving traps

and released energy is transferred to electron-hole pair. This is an inverse process of
impact ionization in which an energetic electron causes generation of electron -hole
pairs. In radiative recombination electron of conduction band recombines with a hole
of valence band directly and emits radiation. The frequency of emitted radiation is v.

Ec Be
M meta stable

Ey Ey
state

a) Meta stable states (M) b) Trapped electron

F, fo) nas
2 2 2

c) Electron hole recombination d) Electron and hole traps

Fig. 7.8 Indirect recombination of eletron and hole

7.7 Surface Recombination :

Electron and holes in a semiconductor may disappear due to surface

recombination. This type of recombination proceeds at a much faster rate than the
recombination process in the bulk. Surface recombination is due to the absorbed
impurities and various other defects present on the surface. These impurities and defects
create effective recombination centers or traps. Surface recombination becomes
dominant in thin samples where volume is small and surface is large.

a9
 Applied Physics

Since recombination is more at the surface than in the bulk of a semiconductor,

the number of freé charge carriers in the surface will be less than in the bulk. This
difference in concentration causes free carriers to move on to the surface, greater the
concentration difference higher will be velocity of charges moving on to the surface.
In order to supress this recombination semiconductors are specially treated (etching,
washing etc.) to remove the recombination centers at the surface.

78 Driftand Diffusion in Semiconductors:

A solid is said to be in equilibrium if it has a uniform temperature and if the

principle of microscopic reversibility applies. Under such conditions, for every
thermodynamic process, there exists a reverse process which proceeds exactly as the
forward process retracing the path in reverse order in the same way. In case of non-
equilibrium, this is not so. For example, the processes of creation of free charges in a
semiconductor under equilibrium conditions will automatically be followed by
recombination of the charges, resulting in no net increase of charge carrier concentration.
In non-equilibrium conditions, however, there will be a net increase or decrease in
carrier concentration depending on the difference of rates of generation and
recombination of charges.

Continuity Equation :
In equilibrium, all the processes are balanced and there will not be any external
force. The fermi level is constant throughout the volume of the solid. Under equilibrium,
there exists, a steady state, in which the condition of the system does not change.

In non-equilibrium conditions, the Fermi level will be varying through the volume
of the solid. The Fermi level will not be exactly in the middle of the forbidden band.
Hence electrons from the valance band can easily pass to the conduction and vice-
versa. Therefore, the rate of generation of charges will not be equal to the rate of
recombination, and there will be a net change in the concentration of excess electrons
or holes. The condition of the sample will be changing with time. The divergence of
current density gives time rate of change of concentration of electrons or holes.
Number of electrons or holes generated = divergence of current density. When
electron concentration causes the current,

This is the one dimensional equation of continuity which is generally written as

VeJ hipdp
pots —0) TOM Speed eabyersiie, ports este,

320
Semiconductors

J = Current density, p = charge density = n, for electrons; where n is the

concentration of electrons

1 dn
—VJ,
Sa-—=0 a i ie Goss)

Similarly, for holes

1 dp
—VJ,
sid: -—-=0
NapFee. oad el NpAlalae eer (7.36)

Under equilibrium conditions, the rate of generation of charges (electrons or

holes) will be equal to rate of recombination. However, for a non-equilibrium condition
there is a net rate of charge generation. Hence for electrons

dn 1
Sa
- (Ge: he, ce HEA
case metialiith sae ane<i (737)

and for holes

dp
args: =e EE: Poe: (7.38)

ce G, = rates of generation of electrons/holes

Re R, = rates of recombination of electrons/holes

As already mentioned non-equilibrium conditions may be caused by internal or

external interferences. External stimulations include optical excitations with high-energy
photons or impact ionization with large electric fields. Under low injection conditions
(i.e., when injected carrier density is much lower than equilibrium carrier density), the
recombination rate R,, of electrons is given by

n, = equilibrium concentration of electrons.

n =concentration of electrons

Pe electron life time.

dn n-n
; phat ey get
es dt aie 2 e
 Applied Physics

The electron current density J, is sum of diffusion and drift currents due to
electron and is given by:

OPP 1 VER otead)DEAN Ae ae Ee ee (7.41)

Here first term in the above equation gives drift current of electrons with mobility
L,, due to the application of electric field E. Second term gives diffusion current due to
the concentration gradient of electrons.

dn N—No dE a°n
Fe ede. tN rea JP, (7.42)
Similarly, for hole densities

dp P—Po dE sy =enCLp
eae balesFide lla (7.43)

D,, and D,, are diffusion contants of electrons and holes. Continuity equation
enables us to calculate the excess density of electrons or holes in time and space. This
equation is one of the basic equations describing the many semiconductor devices.

7.9 Carrier Scattering :

Electrons accelerating under the influence of an applied electric field, collide

with lattice vibrations, impurity atoms or other electron or holes and get scattered.
Hence, they transfer their energy to the scattering source. The average distance between
two consecutive collisions is called mean free path. A perfect crystalline solid does not
offer any hindrance to the passage of electrons within conduction band because of
periodicity of electrons. Thus the electrons in a perfect periodic lattice would
have
infinite mean free path. An infinite mean free path indicates infinite conductivity since
current density and conductivity are proportional to mean free path.

7.9.1 Scattering due to lattice defects :

The factors responsible for deviation from perfection are of two types.
The
first is caused by physical defects and second is by lattice vibrations. Physical
defects
one presence of vacant lattice sites, atoms in interstitial positions, impurity
atoms or
ions dislocations etc., and variations in the surface morphology. Mean free
path of the
electrons would be infinite if there are no defects and if atoms are at rest
in the
semiconductor.
The vibrational amplitude of the atoms increases with increase of tempera
ture
and therefore, mean free path of electrons decreases. The scattering
of electrons due
to defects may be assumed to be constant. Hence at low temperat
ures, the mean free

$10.92
Semiconductors

path and therefore conductivity will be determined by lattice defects and impurities,
and at high temperatures the lattice vibrations are responsible for scattering of electrons
in materials.
The atoms of a perfect material occupy positions to form a periodic lattice.
These atoms at ordinary temperatures execute oscillations about their mean positions
due to thermal energy. These vibrations will be in various possible modes and each
mode of vibration has certain energy associated with it. The distribution of energy is
as per the principle of equipartition of energy. The number of quanta of energy associated
with each mode of vibration gives number of phonons. This electron - phonon interaction
leads to electron scattering. In the process, the electron gains energy annihilating
phonon.

7.9.2 Scattering by Ionized Impurity Ions:

Ionized impurity atoms in lattice positions can scatter electrons because of the
coulombic interactions. Presence of impurity ions produces an irregularity in the
periodic potential of the material paving a way for the scattering of electrons. At low
temperatures the thermal speed of carriers is small and hence it spends more time near
the impurity ion. As the temperature is increased, the scattering becomes more and
more energetic and mobility of the scattered charge carrier would be directly proportional
to T>’*, where T is the absolute temperature.
An electric field applied to a semiconductor will produce a force on electrons
and holes so that they will experience a net acceleration and net movement, provided
that there are available energy states in the conduction and valence bonds. This net
movement of charge due to an electric field is called the drift. The net drift of charge
gives rise to drift current.
To illustrate the concept of drift, let us consider a schematic model of the random
thermal velocity and position of a hole in a semiconductor.

B B
G D F

Wy €

A’ Er
We C
A
(OE Erde
F
eS l

E
(a) (b)
Fig. 7.9 Motion of Electrons in presence ofElectric Field

The figure 7.1 shows such a semiconductor with zero applied electric field. Let
the motion of hole start from a pointA (figure 7.9(a)). It drifts to point B and undergoes

525
 Applied Physics

a collision. The moving hole can collide with another hole, impurity atom and atoms of
the host lattice. As a result of the collision the direction of the motion of the hole gets
altered and it moves towards C. Like this the hole keeps on colliding and even after a
number of collisions, its average displacement over a time interval is zero. In the
figure 7.9(a) the hole comes to a point (E) after suffering four collisions. These
collisions or scattering events alter the velocity characteristics of the particle. However,
on the average, the displacement of the particle over a time interval remains zero.

Ifa small electric field (E-field) is now applied to this material, as indicated by
figure 7.9(b), there will be a net drift of the hole in the direction of the direction of the
E-field, although the net drift velocity will be only a small perturbation on the random
thermal velocity. There will be a net linear resultant displacement of the hole. A similar
analysis is applied to electrons, but the electron motion will be in the opposite direction
to that of hole.

Both electrons and holes contribute to the drift current, so the total drift
current density is the sum of the individual electron and hole drift current densities,
and is given by

Se eal ep) tot ome Semin eae (7.44)

where pt, and H, are the mobilities of the electrons and holes respectively; n and p
are electron and hole concentrations respectively and e is the magnitude of the
electronic charge. Here Go = e(yn + LH, Dl.” See oer pee (7.44)

Diffusion is the other mechanism that can induce a current in a semiconductor.

It is the process where by the particles flow from a region of high concentration
toward a region of low concentration. When the diffusing particles are electrically
charged (such as electrons and holes in a semiconductor), the net flow of charge
results in diffusion current. We will now discuss about drift currents and diffusion
currents in detail.

7.10 Drift Currents :

Equation of motion of an electron in an electric field is

Be peigy oe De DN,
dt Tr: ences

E = applied electric field

Vp = drift velocity
T, = average time between collisions.

Second term on right hand side in the above equation gives average rate change
of momentum of electrons due to collisions.

324
Semiconductors
————

If the electric field is removed suddenly after attaining steady state, then E = 0.

meet dVp esmVp

dt Lie

Solution of this equation is, Vy =Vp,exp(—t/t.) seas (7.46)

Vp _ is the drift velocity at t= 0

Oo

t, is the time for relaxation of the system to the equilibrium state under steady
state. From the above equation.

Drift velocity is thus proportional to the applied electric field under steady state
condition. The proportionality constant is called mobility. Hence mobility is the drift
velocity per unit applied electric field.

Similarly for the hole diffusion, since the hole moves in opposite direction to the
electrons

From microscopic Ohms law

pgn ead Ol cals (Via

J, = o,E = epy,E

Ae eos a= 0] aii 010 ee | (7.49)

For an intrinsic semiconductor, n = p, we have

6; = e(H, + Hy)n; = e(Hy + Hp) AT?2 et g2kT) |... (7.50)

329
 Applied Physics
in a gl a

7.11 Diffusion Currents:

Diffusion is due to non-uniform distribution (concentration gradient) of charge

carriers. As per the Fick's first law of diffusion, diffusion flux F is proportional to and
directed opposite, to the concentration gradient of charges. Therefore, for electrons
and holes (see figure. 7.10).
d
= e.D,.—n(x
Ree ae)
d PCOwty ovate Ser
Joris ODpere wens (7.51)
Hence Jae nt J,
where D,, and D,, are the diffusion coefficients of electrons and holes, respectively.
xX x + 6x AreaA
<q— OX —p

P + (dp/dx)dx

Fig. 7.10 Diffusion of Charges

The diffusion current in a semiconductor is proportional to the gradient of carrier

(electron or hole) concentration and for electrons it is given, in a one-dimensional case,
by
=e eDe Veen

The total current density in a semiconductor is the sum of the drift and diffusion
current densities :
J = Jet dae oO sgh ae) einen pieced C7352)

The carrier (electron or hole) mobility gives an indication of how well the carrier
moves in a semiconductor due to the force of an electric field.

Einstein's Relation:

The carrier diffusion coefficient gives an indication of how well the carrier
moves in a semiconductor as a result of a density gradient. The mobility and diffusion
coefficient are not independent parameters, and are related via the Einstein relation :
EDHa-KD
Mi ae ee (7233)

326
Semiconductors

Note that this relation between the mobility and diffusion coefficient contains
temperature. It is important to keep in mind that, as the mobilities are strong functions
of temperature (due to lattice scattering and ionized impurity scattering processes), the
diffusion coefficients are also strong functions of temperature. The constant kT/e is
called thermal voltage. These relations are true only for non-degenerate semiconductors.
Electron in the valence band may be excited into the conduction band when, for
example, high-energy photons are incident on a semiconductor. When this happens,
not only is an electron created in the conduction band, but a hole is created in the
valence band; thus an electron-hole pair is generated. The additional electrons and
holes created are called excess electrons and excess holes.

7.12 Junction Diode:

Pure semi-conductors have very few device application possibilities. Extrinsic

or impure semiconductors have relatively many applications possible. As already
discussed in the previous sections, extrinsic semiconductors can be p — type or
n-type semiconductors. Most of the useful and interesting semiconductor device
applications use an intelligent combination of these impure semiconductor materials.
Semiconductor junction devices have a p — type of semiconductor placed beside an
n-type of semiconductor material. Examples of semiconductor junction devices are
diodes, transistors, Silicon controlled rectifier etc. A pn junction may be formed from
a single semiconductor material by doping a part of it with n- type (pentavalent or
donor) impurity and other part with p-type (trivalent or acceptor) impurity.
Fig. 7.11 gives the classification of semiconductor devices.
Semiconductor devices

ane
TTT
Intrinsic
TTD Tay Extrinsic
semiconductors semiconductors

Photo High frequency

cells generators

Diodes Transistors Others

Junction Zener Tunnel Bipolar Field

diodes diode diode Junction effect
Mims
NPN PNP N-channel P-channel

Fig. 7.11 Classification of semiconductor devices

a27
 Applied Physics

The essential feature of all semiconductor junctions is that the impurity charge
7concentration suddenly changes at the junction. The charges are p-type (due to acceptor
impurities) on one side of the junction and they are n-type (due to donor impurities) on
the other side of the junction. This change in concentration takes place in a short
distance, typically of about one micron length. The junctions in which this kind of
sudden change of carrier concentration is present are called abrupt junctions. Other
type of junction where a relatively gradual change of concentration occurs is known as
graded junctions. On p-side of the junctions, holes are majority in number and electrons
are minority and reverse is the case in n—-side of junction diode.

7.12.A PN -Junction Diode:

(a) Depletion Region :

When n and p-type semiconductor materials are brought into contact with
each other, a contact potential V, is created. The potential barrier is created because
the Fermi energies of the separate n- and p-type materials are different and when
they are brought into contact, they must reach an equilibrium. Before the
semiconductor materials are brought into contact, n material has large electron
concentration and p material has large hole concentration. Upon joining the materials,
charge diffusion takes place between the two materials because of the concentration
gradient of the charges. The holes diffuse from p side to n side and electrons diffuse
from n-side to p-side. Diffusion current cannot build up indefinitely, but will reach
equilibrium. The uncompensated ions are left behind by the charge diffusion from n
to p-side or vice-versa. This will result in development ofpositive space charge near
the n-side of the junction and a negative space charge near the p-side of the junction.
Thus an electric field is generated in a direction opposite to the diffusion current.
The electric field builds up till the equilibrium is reached. Fig. 7.12 illustrates the
charge distribution, electric field and energy band bending in an abrupt junction to
equilibrate the Fermi energy. In an abrupt junction, doping of p and n-sides is uniform.
The Fermi energy equilibrates by a transfer of negative charge from n-type to p-type
material. The electric field appears in a region W about the junction, and there is an
equilibrium potential difference V, across W. The region W is called transition region
and contact potential across W is built in a potential barrier and it is necessary to
maintain the equilibrium at the junction. Contact potential cannot be measured with
a meter as no current results from it. Within the transition region, since free charges
are depleted or removed, an electric field is created. This region is also known as
depletion region.

328
Semiconductors

qne

(b)

-—qne

|
Beigel Lal ak ea a
rit G¥o > Epa een
|
a ee ea ee tt appeal poy Erg ne Ee

(C) —

Fig. 7.12 (a) n and p materials are brought into contact, (b) An electric field appears across the
junction, (c) The Fermi energy equilibrates across the junction.

(b) Potential across junciton :

To a good approximation, the charge transfer is achieved by all the donor electrons
within x, of the boundary flowing over and filling all of the acceptor states within Xy of
the boundary. The contact potential is,

Here E,,, and Ep, are F ermi energies of n side and p-side of the semiconductor.
Let n, and ny be acceptor and donor impurity atom on p and n sides.

329
 Applied Physics

The electric field for xy <Ous;

Where P =-qn, is the charge density, and the absolute magnitude of E at x = 0,

E max? iS

max €

For charge conservation,

n.x = nx, and x, = W — x

ap p

=p Ny W ee
eee ee ee Rem (7.58)

Substituting Xp in the equation for V, we get,

Neel oetiaal saan,

Zeon,

(c) Current-Voltage Characteristic :

At the junction, there are four current densities, the current densities due to hole
diffusion, electron diffusion, hole drift and electron drift. Electron and hole diffusion
currents are due to electrons and holes respectively and are in accordance with Fick's
Laws of diffusion. Electron diffusion current would be from n side of the junction to
p-side and hole diffusion currents flow in opposite direction. The electrostatic potential
and width of the potential barrier at the junction is lowered by applied forward bias Ve
(When p-side of the junction connected to the positive electrode of the voltage source
and n-side of the junction is connected to negative electrode of the voltage source).

330
Semiconductors

Then the effective barrier height would be V,—V,. This is due to increase in the p-side
potential due to the applied forward bias. Reverse bias increases the height of the
potential barrier and width ofthe potential barrier at the junction. The electron and hole
diffusion currents at the junction are almost independent of the applied bias voltage.
The total current crossing the junction is sum of the diffusion and drift currents.
Under equilibrium, the net current crossing the junction would be zero as drift and
diffusion components cancel each other. Fig. 7.13 shown current-voltage characteristic
of a pn junction diode.

Reverse saturation
current Is
|

Fig. 7.13. Current-voltage characteristic of a pn junction diode.

The space charge region that is built at the junction is also known as depletion
region, because this region is depleted of mobile charges. It is also called the barrier
layer, as there is a potential barrier. This potential barrier prevents further diffusion of
charges across the junction.

7.12.B Fabrication of pn Junction Diodes :

Diodes can be fabricated formed by causing a transition of the semiconductor
material from p-type to n-type to occur with in a semiconductor material. This can be
done in different ways:
1. Grown junction
2. Alloy junctions
3. Diffused junctions
4. Epitaxial junctions
1. Grown Junctions:
During the crystal growth, the melt of the semiconductor material is doped with
the p-type (or n-type) impurity. At a later time during the growth process (after growing

331
 Applied Physics

sufficient length of p - type material) an N-type (or p-type) impurity is added in sufficient
concentration to the override the effect of p-type (or n-type) so that the subsequently
grown crystal is of n-type (p-type). The resulting crystal is partly p-type and partly n-
type. At the junction, the net impurity concentration is zero. The junction formed in
this process is gradual and the graded pn-junctions are formed in this method. First
junction devices were made using this technique. But this method is not a commercially
used method now because of the difficulties involved in crystal growth.

2. Alloy Junctions :

Alloy junctions are also known as fused junctions. Alloying is widely used for
making semiconductor junction because, this method is simple and economical. A
small n-type semiconductor wafer is taken and surface is cleaned and a smaller dot of
p-type impurity material is pressed on to it. This assembly is heated to a high temperature
using a furnace in a non-oxidizing atmosphere. When the furnace temperature is raised
above the melting point of the eutectic alloy formed by the semiconductor wafer and
doping (dot material) material, the dot material dissolves into the wafer. When the
temperature is decreased, the melt material recrystallizes. During recrystallization process,
an essentially continuous material is formed. This material is rich in p-type impurity.
The structure of the pn junction thus formed depends greatly on time and temperature
of alloying. This method results in an abrupt junction.

3. Diffused Junctions :
Diode manufacture by diffusion technique is the most common fabrication
technique used in the industry. One surface of a slab of material of the type to be used
for the base material (substrate say n- type semiconductor material) is exposed to the
opposite type of gaseous impurity (in this case, p-type impurity) at high temperature.
The gaseous impurity diffuses slowly into the surface of the material. The concentration
of the impurity is the highest at the surface and slowly decreases exponentially inside.
The major part of the substrate remains n-type, but the exposed surface becomes p-
type. A pn junction is thus formed. The net impurity concentration is controlled by the
furnace temperature and impurity gas concentration.

4. Epitaxial junctions :
Epitaxial growth process is another commercially exploited technique for the
large-scale manufacture of the semiconductor junction diodes. In this process a suitable
substrate is selected. The substrate can be p-type or n-type semiconductor material.

332
Semiconductors

On this substrate, say p-type, another n-type semiconductor layer is grown epitaxially.
The suitable chemical reaction is used for the growth of this n-type layer.
Various methods of junction fabrication are shown in figure 7.14. After pn junction
diode is formed by any of the above techniques leads are connected to p and n types of
material to form ohmic contacts.
Junction

p-type
impurity
N

N-type
Single crystals semiconductor

(a) (b)

p-type
N-Type diffused
substrate pee

(c)
Fig. 7.14 Fabrication of p-n junction diode a) Grown Junction b) Alloy Junction c) Diffused
Junction.

713 Volt-ampere characteristics of pn junction diode:

An ideal junction diode is the one, which has no reactive components (capacitance
or inductance associated with it). The working of the pn junction diode can be understood
with the help of energy band picture. It is necessary that in the two semiconductor
materials which are in contact (p-type and n-type), the Fermi level E,, should be at the
same energy level under equilibrium condition. When p-type material is brought into a
contact with the n-type material for the formation of pn-junction diode, immediately
charge transfer takes place in such a way that the Fermi level gets adjusted. The
process of charge transfer continues till the Fermi levels in both the materials get
aligned to the energy level. After this equilibrium condition is reached, no more charge
transfer takes place. In n-type material, the Fermi level is near conduction band and in
p-type semiconductor material, it is near valance band. Energy levels on p-side of the
junction will be at a higher level compared to the energy levels on n-side of the junction
(figure 7.12c). When equilibrium is established the potential barrier that is formed at
the junction prevents further transfer of charges (electron or holes) across the potential
barrier or hill. Hence, under the equilibrium condition, no current would flow across

333
 Applied Physics

the diode. In order for the current to flow through the diode, voltage has to be applied
across the diode. Applying voltage to the diode is called biasing the diode. Diode can be
biased in two ways:
1. Forward bias
2. Reverse bias.

Forward biased diode:

In forward bias, the positive terminal (p-side) of the diode is connected to the
positive of the voltage source and negative terminal (n-side) of the diode is connected
to the negative terminal of the voltage source. The connection is shown in the circuit
given in the figure 7.15(a). For the current to flow through the diode, a voltage should
be applied in such a way that the potential barrier present in the diode at the junction
should be reduced. This is accomplished in the forward bias. The potential barrier
height decreases by applied voltage. If, barrier height is V,, and the applied forward
voltage is V, the effective barrier height after forward bias would be (V, — V). Hence

a) forward bias b) Reverse bias

whe =

anode cathode

c) Circuit symnbol of the diode

Fig. 7.15 Biasing ofa diode

current can flow through the junction because the drift and diffusion currents are no
longer balanced. Thus, the reduction in the barrier height by applied bias voltage causes
holes from p-side of the junction to be injected into n-side of the junction and electron
move form n-side to p-side of the junction. Thus, there is a current flowing through
the external circuit. Four components of current can be identified when a bias is
applied to a diode. These are :
1. Current due to holes of the p-side (I,,)
2. Current due to electrons of the n-side (I,)
3. Current due to electrons of the p-side
4. Current due to holes of the n-side
First two currents are the currents due to the motion of majority carrier (I, I.)

334
Semiconductors

and later two are the minority carriers. Minority carrier current is thermally activated.
On the other hand majority carrier current due to holes and electrons is due to the
diffusion of the charges across the junction due to the reduced barrier height because
of the forward bias. Since hole and electron have opposite charge, they move in opposite
directions, by the application of voltage. Thus the total current flowing through the
diode I is equal to the sum of electron and hole currents and is given by:

[= Iolexp(@=) -1 sestere (7.60)

where I, in the above equation is called saturation current density and it is equal to the
sum of the minority charge currents across the junction, e is the electron charge, V is
the applied voltage, k Boltzmann’s constant and T temperature in degree Kelvin.
The eq. (7.60) is the diode equation. When forward bias is applied to a diode,
the exponential term in the above equation rapidly increases, and becomes much larger
than unity, and then, the above equation can be written as :

t= Iolexp—] ae (7.61)

Here kT is the temperature equivalent of voltage and is equal to 26 mV at 300°C.

This equation gives voltage - current characteristics of a forward biased junction diode.
This indicates that the current across the junction increases exponentially with the
applied voltage. Hence if large voltage (more than a fraction of volt) large current
flows across the junction and junction width and barrier height will reduce. This may
result in thermal breakdown of the diode (Fig. 7.16).

Reverse biased diode :

It can be again seen from the above equation that if a negative voltage is applied
to the diode, the exponential term reduces to a small value and unity present in the
equation becomes larger compared to this. The diode is said to be reverse biased under
this condition Fig 7.15(b). Here the n- side of the junction is connected to the positive
terminal of the voltage source and p-side of the diode is connected to the negative
terminal of the battery. Under limiting condition the maximum value of the current lis
equal to —-I,. This current is called the reverse saturation current and is decided by the
diode parameters only such as the doping concentrations of the diode and is independent
of the applied voltage. This current depends on the minority charge carriers crossing
the junction and does not depend on the barrier height of the junction. Diode is a bipolar
device since both electrons and holes are responsible for currents in the diode.

335
 Applied Physics

The diode equation is plotted in figure 7.16 and this gives volt - ampere
characteristic curve of the junction diode. This is not a straight line. Diode current
does not vary linearly as a straight line with voltage. Hence, diode is a non-linear device
(non-ohmic). Current flowing through a diode when it forward biased (for positive
applied voltages) is large. The point on x-axis where the diode current starts increasing
is called knee voltage or threshold voltage.

I(mA) Forward bais

dI l 1
slope= ——" = ———_ = ——_
dv dynamic lg

knee voltage

v(volts) v(volts)
0.4 Forward
break down bias
voltage 1
=
-

---.
Reverse
bias

I(uA)

Fig. 7.16 Volt-amper characteristic of junction diode.

When a large reverse voltage is applied to the diode, the electric field gradient at
the junction becomes large and the covalent bonds in the material get disrupted and
a
number of charges will be set free. This results in increase in the reverse current of the
diode. This is called breakdown of the junction diode. This excessive reverse current
of the diode damages it. Hence, a diode should not be operated in the break down
region of reverse bias voltage.

A large forward current (milli amperes) and a small reverse (micro amperes)
current allowed by the junction diode is the characteristic of the diode, which
makes it
useful in rectifier applications. This property of the diode is called unilateral
conduction
of the diode. Hence a diode will have small resistance when forward biased
and high

336
Semiconductors

resistance when it is reverse biased. An ideal diode must have zero resistance when
forward biased and infinite resistance when reverse biased.

Static resistance or dc resistance of the diode is give by

chi c lahMeret t = toe

birYat]
at e ver Ske (7.62)
hon Hyer
Dynamic resistance or ac resistance of the diode is give by

kT kT
=a Tew eve On)
dV
(7.63)

The diode resistance for a reverse biased diode extremely large. At high
temperatures, the diode shows a linear variation of resistance with temperature. Hence
diode becomes ohmic at higher temperatures.

Approximate diode models:

As discussed above, a diode is a non-linear and unilateral device. There are two
models of a diode. One is small signal model and the other one is large signal model.
When a diode is being operated at a point close to its knee voltage, small signal model
can be used. If it is being operated at a much higher current compared to the current
corresponding to the knee voltage large signal model should be used. Large signal
model of a diode is described as follows:
1. Since a diode has small resistance when forward biased and large resistance when
reverse biased, a diode acts as a closed switch when forward biased and open
switch when reverse biased.
2. There is a knee voltage or threshold voltage for the diode. This voltage is the minimum
voltage that must be applied to the diode to enable conduction of the current. Hence
a diode can be approximated to a switch which is in series with a battery whose
voltage is equal to the knee voltage.
3. Diode is made out of p and n type semiconductor materials. The bulk of the material
will have certain resistance. This bulk resistance of the diode acts as if it is in series
with the switch, battery that are associated with diode.

In view of the above model, the equivalent circuit of the diode is shown in the
Fig. 7.17. In the small signal model of the diode, in addition to the bulk resistance of the
diode, the dynamic resistance of the junction is also included.

337
 Applied Physics

am Potential source

V, Ph : VB hy,
: Bulk resistor
Switch (r,)
(S)

(a) Reverse bias (b) Forward bias

Fig. 7.17 Equivalent circuit of diode.

7.14 Direct and Indirect Band-Gap Semiconductors:

In energy (E) versus (k) wave vector diagram, if the minimum ofthe conduction
band is directly above the maximum of valance band, then the material is called direct
band gap material (Fig. 7.18). The minimum in the conduction band is important
because electrons excited from the valance band fall to this point much more rapidly
than they fall from conduction band to valance band. Also holes move more quickly to
the maximum of the valance band.
In indirect band gap materials, the conduction band minimum is shifted from
k= 0, and when transition of electron or hole occurs, the wave vector must be conserved
and hence only those photons with certain discrete values of k are absorbed. This
requires simultaneous collision between electrons, holes and the lattice. A three-body
collision is required for transitions in indirect band gap materials and two-body collision
is required for transitions in direct band gap material.

(a) Direct Band Gap

(b) Indirect Band Gap

Fig. 7.18

338
Semiconductors

Band to band transitions have a much higher probability of occurring in a direct

band gap material because only an electron and hole need to collide. .Therefore the
elemental semiconductor materials like Si or Ge are not used for optoelectronic device
fabrication as they are indirect band gap materials. They have the following drawbacks:
1. Fundamental band gap in elemental semiconductors is indirect.
2. They have low optical absorption coefficients with absorption edges in infrared
region.
3. As a consequence of above properties they have low conversion efficiencies.

Example 1 :

What is the minimum energy released in electron-hole recombination in Si at

300°K. Find the wave length of emitted radiation ES me] Dev:

he
creas es minimum energy released.

Substituting E, = 1.12eV, h = 6.6 x 10° J-sec, c is the velocity of

light = 3 x 108m/sec, we get A = 1.1 x10°°m. The minimum energy released
is 1.12 eV.
Example 2 :
Calculate the probability that an energy state 3K Tabove fermi level E,, is occupied
by an electron at 300°K.

f(E)= —E=E) (E-E, =3kT)

+ exp kT

]
=————, = 0.0474
1+ exp(3)

Example 3 :
Calculate the volume density of quantum states in the energy interval 0 to | eV.
dE =1-0=1eV;m,=9.11 x 103! kgs.
ev
An AAn E*dE
3/2 JENpoll? dE
er = 732m)
: 1
steees Om.) [ae eer

= 4.5 x 1027 m3

339
 Applied Physics

Example 4:
For an intrinsic semiconductor of Ge, energy gap is 0.7eV, determine the position
of Fermi level and electron concentration at 300°K if effective mass of electron is six
times that of hole.

Fermi level in intrinsic semiconductor is given by

Here E, =Ec +Ey =0.7eV

Temperature equivalent energy = kT = (1.38 x 10°23) / (1.6 x 107!%) x 300 =

0.026eV
Given ON? =Ne
Substituting the parameters in the expression for E,, we get E, = 0.385eV.
Carrier concentration is given by
3
2nkT 5 * ae —Eg
n=2 m.m, }2 exp} ——
[h? (: ) (=)
= AT*” exp(— Eg/2kT)
where A = 4.83 x 102! K7!
Hence n = 3.5 x 10!9/m3.
Example 5 :
Plot the Fermi -Dirac probability function for — SkT< (E- Ef) < SkT
The Fermi - Dirac distribution is given by

1
a
1+ exp =a

Where E, is the Fermi energy, k is the Boltzmann's constant, and T is the

temperature in Kelvin's.

B=

340
Semiconductors
|B ene eee an ae ee ea

So we need to plot the function

f,(B)= aaa

plotting the graph in the interval B = (—5S, 5)

B
Fig. 7.19 Fermi-Dirac probability function
Note that the probability ofa state being occupied at E = E, (B = 0) is one half.

I l l
Indeed, 4s Fe) 75 en(0)sama
Example 6:
Obtain an expression for electric field as a function of distance in a semiconductor.
Also find an expression for total current in the semiconductor.
The electron current has both diffusion and drif current components, so the
electron current density for this material is given by

ip =e14,.n(xJE+eD,{ n(x)
solving this equation for the electric field E yields:

This electric field distribution can be readily plotted versus x :

The total drift current density is equal to the sum of the individual electron an
hole drift current densities, and is given by
Jap = - (Uy - n+p. P).E
where 1, and 1, are the mobilities of the electrons and holes, respectively, n and p are
electron and hole concentrations, respectively, and e is the magnitude ofthe electronic
change.

341
 Applied Physics

This expression can be written as

Jag = 6 - E where o =e. (yu, . n+p, . P)
where o is the conductivity of the semiconductor materials. Since the mobilities are
functions of temperature and impurity concentrations, conductivity can be a rather
complicated function of temperature and impurity concentration also.
The effective density of states function N,(T) in silicon can be written in the
form :

Ne(1)=2.8:10"( 5) (em )N,(0)

3
Tae w
=1.04.10'°| — |?(m=?)
300

As discussed above, the conductivity o of a semiconductor is given by

o = e.(u,.n+ uP)

for an intrinsic semiconductor, the concentration of electrons in the conduction band

is equal to the concentration of holes in the valence band, i.e.,
Nes bikers
The intrinsic carrier concentration in a semiconductor n, (T) is given by

E
n,(T)=,/N,(T).N, (T).exp| —&
k.T

where N, and N, are the effective density of states functions in the conduction and
valence bands respectively, E, is the band gap energy, and k is Boltzmann's constant.
Substituting this equation for n,(T) into the expression for conductivity, we get

o(T)=en; (TI, (7)+ n, (7)|

The conductivity in intrinsic silicon is rather low. However, conductivity can be
increased by many orders of magnitude by introducing impurities (dopants) into the
lattice of the material. A typical value of conductivity at T= 300K of a silicon crystal
with impurity (donor or acceptor) concentration of 10!8 cm3 is (1Q — cm)!, six
orders of magnitude higher than the intrinsic value.

342
 Semiconductors
rere nc

Example 7 :

Find out the resistivity of a semiconductor for which n, = 2.4 x 10! per m3,
H, = 0.39 m?/V-s and H = 0.19 m?/V-s.
Solution :

Since o= e(u,.n + 1,.P)

1 ]
Resistivity is ae a e(u,n = a)

= —[ ——l —~ = 0.449 ohm — m.

1.6x 107? (0.39 + 0.19) 2.4x 10!

Example 8 :

Calculate the drift velocity of holes and electrons at 300K if the applied electric
field is 100V/m in silicon and germanium. Also find diffusion coefficient at 300K in
these materials.

Solution :

Drift Velocity = H E = 0.0038 x 100 = 0.38 m/sec for electrons

Diffusion coefficient D = (KT/e)y = 0.93 m/sec for electrons.

Similarly we can find out drift velocity and diffusion coefficient for holes.

Example 9:

Calculate the density of donor atoms which have to be added to produce n-type
material of resistivity 0.19 ohm-m, given that mobility of electrons is 0.325 m7/V...

Solution :

We know that ge (en)

n=
A
el

= 1/ (1.6 x 107!9x 0.19 x 0.325) = 1.01 x 10!8 per m3,

343
 Applied Physics

Example 10:

Conductivity of a semiconductor at 300K is 2 mho per m. Find its conductivity

at 600K, if its Fermi energy does not change with temperature.
Solution :

epAT?? oxBets

We know that Oe Beave 2 eee eee CO

O72 3/2 Eas
euwAT;°“
ees exp) ————+
kT,

substituting the values given and simplifying, conductivity at 600K can be evaluated.
Example 11 :

Find the diffusion coefficient of electrons in silicon at 300K if is 0.19m2/V-sec.

Solution :

From Einstein's equation,

D_kT
pL e

D = (1.32 « 10°23 x 300 x 0.19)/ 1.69 x 10°19

= 0.00445m2/sec
Example 12 :

Current flowing through a junction diode is 0.20 microamperes at room

temperature when a large reverse bias is applied. Calculate the current flowing when
0.1 V bias is applied.
Solution :

The current flowing through a diode is given by:

I=1, [exp(—) =i
When large reverse bias is applied the current flowing through the diode is
reverse saturation current I) = 0.2 X 10-7 Amps.
Hence for V = 0.1 V, form the above equation I = 9.1625 microamperes.

344
Semiconductors

Example 13 :
The reverse saturation current of a diode is 1 microampere at 27°C. Determine
its forward resistance when a bias of 0.15 V is applied.
Solution :

The dynamic resistance of a diode is given by

a ORT kT
fe GT IglexpV/AT)| (1+ Ip)
In the problem I, = 1 microampere
T = 27 +273 =300 K; kT = 0.026 V

dV :
= =6 E pas Se 77.65Q
Hence Vac =
dl 1X10 °[exp(0.15/0.026)]

Questions

With the help of energy level diagrams explain insulators, semiconductors and
conductors.
With necessary theory, obtain a relation between energy gap and temperature in
intrinsic semiconductors.
Define Fermi level. Obtain Fermi Dirac distribution function.
Sketch the energy band diagrams of a) an intrinsic b) N-type c) P-type
semiconductors. Indicate Fermi, Donar and acceptor levels wherever present.
What is the significance of volt equivalent of temperature. Obtain an expression
for Fermi energy in terms of this parameter.
Obtain an expression for the hole concentration in a non-degenerate
semiconductor.
Show that the average kinetic energy of a hole in the valance band for a non-
degenerate semiconductor is 3kT/2.
Explain affect of temperature and impurity ions on the Fermi level in a
semiconductor.
Explain what is charge neutrality and compensated doping.
What are the various electron-hole recombination mechanisms?
11) With necessary mathematics, explain the Shockley Read Hall theory of
recombination.

345
 Applied Physics

12) The band gap of a semiconductors is 2eV. Find out wave length of radiation
emitted when electrons of this semiconductor recombine directly.
13) Electrons can be excited across forbidden energy gap in a semiconductor by
photons of sufficient energy. What is the maximum energy of a photon to
create electron-hole pairs in a) germanium, b) silicon, c) diamond.
14) The relative density of a semiconductor is 18.8, its atomic weight is 184 and
there are two free electrons per atom. Find out the density of free electrons and
Fermi energy.
15) In a given n-type semiconductor material, the donor concentration is | atom per
every 2 x 108 semiconductor atoms. Assuming that the effective mass of electron
is equal to its true mass, find the value of temperature at which Fermi level
coincides with the edge of conduction band.
16) Using expression for concentration of holes in a semiconductor i.e.,

p= |lie,~F)! exp E-E,)/kT

dE
Show that dp/dE is maximum at E, — E = kT/2.
17) What is equation of continuity for semiconductors. What physical law leads to
the equation of continuity.
18) What are drift and diffusion currents ? Obtain expression for each of these.
19) What is carrier scattering ? Discuss the reasons for carrier scattering.
20) Define carrier life time and diffusion length.
21) Explain various fabrication methods of junction diode.
22) Describe the variation of electric field, space charge, and potential as a function
of distance from the junction of a diode.
23) What is the significance of barrier potential in a diode?
24) Explain the unilateral conduction of a diode.
25) With necessary circuit diagrams, explain the working of a diode in forward and
reverse bias.
26) Explain the working of a diode with the help of diode equation and hence describe
its volt-ampere characteristics.
21) Sketch the energy band diagram of a diode in forward bias and reverse bias
conditions. Explain the effect of applied bias voltage on the energy band diagram
of the diode.
28) An ideal diode has reverse saturation current of 30 microamperes at
125°C.
Find the dynamic and static resistance of the diode when a bias of 0.2 volts
is
applied in forward and reverse bias.

346
Semiconductors
en

What is Diffusion ? Explain diffusion using Fick's law.

The resistivity of an intrinsic semiconductor at 270K is 3000 ohm — m.
Calculate intrinsic carrier concentration if ae 0.17m2/V-sec and
1 0.0035m2/V-sec.
(Answer : 1 x 10/6 per m>)
31) The conductivity of a semiconductor varies as exp(— 4350/KT), find its band
gap.
(Answer : 0.67 ev)
32) A sample of germanium has an intrinsic resistivity of 0.6 ohm-m. Calculate the
value of current density in an applied field of 1V/m, if there are 10!8 donors and
4x 10!” acceptors in each m3 of the specimen and if mobilities of electron and
hole are L, = 0.0018m?/V -sec and p p = 0.0038 m?/V —sec.
(Answer : 3.65 x 10° A/m?).

347
8
SUPER CONDUCTIVITY

8.1 Introduction:

The discovery of superconductivity was an off shoot of the results of the study
of the properties of gases at low temperatures. During the second half of the nineteenth
century, many scientists were developing methods to condense gases with lower and
lower boiling points. One of the outstanding methods was Joule - Thomson porous
experiment which showed that gases when pumped through a porous plug and made
to expand, they suffer a fall in temperature (cooling). All real gases suffer this coo!ing.
Carbondioxide was first frozen by M.Faraday in 1845. Raoul Pictet and Paul Cailletet
were successful in liquefying air in 1877. James Dewar liquefied hydrogen in 1898 and
he invented Dewar flask, to keep the liquids safe and cool. All the gases have to be
cooled first below their inversion temperatures before subjecting them to further cooling
in porous plug technique. For helium and hydrogen, the inversion temperatures are
much below room temperature (He, 23.6K; H,, 196K). Hydrogen can be precooled by
liquid nitrogen (boiling point 77.3K at one atmospheric pressure) and helium can be
precooled by liquid hydrogen (boiling point 20.4K at one atoms). For a long time the

348
Super Conductivity

liquefaction of helium posed a serious problem till Kammerlingh Onnes in 1908 could
liquefy it applying the method of Joule -Thomson expansion. Liquid helium below
4.22K found a number of applications. The thermal, electrical and magnetic properties
of a number of substances were determined at low temperatures and they exhibited
property - parameters which could be explained only on the basis of quantum physics
thereby paving the way for strong roots of quantum mechanics in physics. One of
them was the discovery of superconductivity in a number of metals. The discovery of
superconductivity was one of the outstanding discoveries made in the twentieth century
and it heralded the dawn of a new era in low temperature physics and engineering.

82 The Properties of Superconductors :

8.2.1 Electrical Resistivity :

Matthiesen predicted that resistivity of pure metal would level off as absolute
temperature approached zero. At higher temperature the resistivity of metal would be
determined by lattice vibrations. The vibrations would be responsible for scattering of
electrons resulting in resistance. Kammerlingh Onnes investigated the electrical resistance
of metals at low temperatures. This was a hot subject in physics in those days. Dewar
predicted that the resistivity of a metal would approach zero at absolute temperature.
The experiments conducted by Onnes metals like gold showed that the resistivity of
pure metals would decrease with the decrease of temperature (below 10K) and at very
low temperatures would tend to a constant value determined by residual impurities or
other defects in the metallic sample. However Kamerlingh Onnes who achieved the

0 150

I || Hg :
>
.
o
© o
= | w
@ |
a Po Super
| conductor

P — >
Alon 444 4.2 4.2 4.4 O Te K
T ——~> Temperature
Temperature

Fig. 8.1 (a) and (b) The variation of resistivity (a) for mercury (b) comparison between pure metal
and a superconductor

349
 Applied Physics

liquefaction of helium and investigated the electrical resistivity of pure mercury below
10K, found that the electrical resistivity of pure mercury dropped abruptly to zero at
around 4.2K. The mercury was incorporated into a ‘U’ shaped glass capillary tube
with electrodes on both ends so that the current could be measured through it while
still in liquid. Finally mercury was cooled to a solid wire. Onnes found a regular decrease
in resistance with decrease of temperature and at liquid helium temperature, the resistance
appeared to have completely vanished. (Fig. 8.1). The result was checked and rechecked,
and the discovery was finally announced.
From his observation, Onnes concluded, “mercury has passed into a new state
which on account of its extraordinary electrical properties may be called the
superconducting state.” The temperature at which the metal had passed into the
superconductivity is known as the superconducting transition temperature (T,).
H.Kamerlingh Onnes received the 1913 Nobel prize in physics for this great discovery.
Onnes in 1914 performed a similar experiment on a coil of lead and established a
‘persistent current’ in it below 7.2K. This discovery of superconductivity opened up
tremendous curiosity about the scope and nature of the phenomenon. It was subsequently
found by several scientists that twenty-one elements and a large number of alloys
become superconducting, the transition temperature being characteristic of the particular
metal or alloy. In the case of pure metals it was found to vary as low as 0.35 K for
hafnium to 9.3K for niobium. Table 8.1 gives the list of superconducting elements
along with their transition temperatures.

Table 8.1 : Superconducting metals and their transition temperatures

De a
FEZ
ee
350
Super Conductivity

Later Onnes also discovered that even a minor magnetic field as a result of the
currents of sufficient strength in the superconductor extinguished the superconductivity.
All elements do not show superconductivity phenomenon. Normally, the superconducting
elements have been found to have the valence electrons between 2 and 8. For example,
alkali metals, noble metals like copper, gold, silver etc., ferromagnetic and
antiferromagnetic metals do not exhibit superconductivity . It was also found that the
purity of the metal determines the sharpness of the transition from normal state to the
superconducting state. For example, in Onnes’ experiment, he observed that the
resistance of mercury decreased from 0.08 W at 4K to about 2.8 x 10-Q over just
0.01 temperature interval. So the transition may not be sharp, but may exhibit a width
near the transition temperature. The Fig. 8.2 depicts the effect of the nature of the
metal on the sharpness of the transition.

Single crystal Pure

o> ja Impure

Te a ==

Fig. 8.2 Superconducting transition depends on the purity and form of superconducting Specimen.

From the observations carried out on many elements and alloys, it became evident
that an important characteristic of superconductivity is the zero resistance (p = 0) or
infinite conductivity. It was also discovered that non-magnetic impurities had little
effect on superconducting properties of metals. However the addition of impurities like
the magnetic ions to a pure superconductor lowers the transition temperature. For
example one atomic percent of gadolinium lowers the transition temperature of
lanthanum from 5.6K to 0.6K. Pure molybdenum (Mo) has a transition temperature of
0.92K., but the addition of a few parts per million of Fe destroys the superconductivity
in it. Many alloys are known to become superconducting at low temperatures. The
most important alloy is Nb;Ge which exhibits a T, of 23K. Table 8.2 gives the list of
some superconducting alloys along with their transition temperatures. Onnes observed
that the increase of superconducting current in the specimen of a superconductor
resulted in the quenching of superconductivity .

350
 Applied Physics

Table 8.2 : Some Superconducting alloys

Np

Pb-Hg
Pb-In 3.473 .
Pbsb,

There were certain properties which were found to have no changes in

superconducting state as compared to the normal state. Some of them are: 1) The
X-ray diffraction pattern of the lattice was same both above and below transition. It
means that there was no crystal lattice change upon transition. 2) No change in reflectivity
of visible or infrared radiation for a metal in normal and superconducting state.
3) There was no change in latent heat or volume upon transition in the absence of
magnetic field.

8.2.2 The Magnetic Properties : Meissner Effect :

The magnetic properties exhibited by superconductors were as dramatic as

their electrical properties. The observation of H. Onnes that the increase of
superconductivity current resulted in destroying the superconductivity lead to the
conclusion that the magnetic fields should have a definite effect on their property .
It
was found by Tuny and Onnes that keeping a super conductor in a magnetic field
of a
definite intensity restored the resistance of the normal conducting state. The magnetic
field intensity at which this restoration takes place is called critical field (H,).
The
critical field needed to quench Superconductivity depends on the temperature.
Its relation
to temperature is expressed empirically by

Fe Hsbc s) 2 | ee lean antec (8.1)

where H, is the critical field at T= 0 K and T, the temperature.

352
Super Conductivity

For many years after the discovery of superconductivity, it was first assumed
that the above magnetic property of a superconductor could be deduced merely from
the infinite conductivity associated. However, in 1933, Meissner and Ochsenfeld proved
this assumption as incorrect. The following will describe it.

Attempts were made to understand the magnetic properties of a superconductor

from a concept which assumed that the superconductor was a perfect conductor. The
characteristic of the superconductor is its zero resistivity and hence it should possess
infinite conductivity . Ohms law which is valid for conductors is expressed as

at en ST a SORE ER RE eee Oe ete ere (8.2)

where E, represents the electric field and J, the current density in the sample. The
current in the sample is proportional to the drift velocity, the charge density, and inversely
to the cross sectional area of the current carrying medium. The current per unit area is
called the current density.
The expression in 8.2 is a point relation i.e., the current density at every point in
the conducting medium is related to the total electric field at that point. J has the units
of amperes per meter? and is a vector quantity .
From the Faraday’s law of induction, Maxwell deduced the differential form of
Faraday’s law as (B =p, H, where 1, is the permeability of free space, 'H' is the
magnetic field intensity).

ur 5 eeeeeeee 43

ul oO at See | )

But, in a perfect conductor o 0, and E -» 0. That means B is constant,

indicating that it is impossible to bring about any change of the lines of induction B in
a perfect conductor. If a perfect conductor is, therefore, subjected to a changing
external magnetic field, a surface current will be induced so as to keep the magnetic
field inside the conductor constant. The external field will be distorted by these currents
so that no lines of force enters the sample. These screening currents do not die away
since there is no resistance. Hence the field inside the conductor remains invariant in
time. The current density also gets frozen in the sample. The infinitely conducting
sample would be left with ‘frozen in’ paramagnetic moments corresponding to the
frozen magnetic flux. A perfect conductor would resist change in flux.

a5)
 Applied Physics

2B
3 _3loulvrwsroqun’ xVexeiabiqcaty Seema need ae (8.5)

If a superconductor were a perfect conductor, it should exhibit the above feature.

The magnetic field should be trapped inside it. If asmall permanent magnet is taken
towards it, it should attract it. But a superconductor does not exhibit these features. It
resets the magnet. Meissner and Ochsenfeld (1933) showed that the assumption of a
mere perfect conductor for a superconductor would lead to a misconcept. Their
experiments on a narrow metallic cylinder showed that the magnetic field strength
outside the metallic sample changed suddenly when the sample was cooled through the
transition temperature. A coil was wound on a thin and long cylinder of metal, which
was to be made superconducting. A magnetic field was switched on parallel to the axis
of the cylinder while the metal was in normal state. A galvanometer connected in series
to the coil showed a ballistic deflection proportional to the magnetic flux present in the
metal. Now the specimen was cooled, a ballistic deflection opposite in sense, but equal
in magnitude to the original one, was produced, at the instant the transition temperature
appropriate to the field was reached. When the field was switched off, there was no
deflection. This clearly demonstrated that the superconductor expelled the magnetic
flux lines from it. The superconducting, sample showed no magnetic flux inside as
shown in Fig. 8.3 .

Fig. 8.3 Meissner Effect in a superconductor

The above observation known as Meissner effect was regarded as an additional

fundamental property apart from zero resistivity. Therefore, a superconductor does

354
Super Conductivity
a

not only behave like a perfect conductor, but behaves like a perfect diamagnetic in
which B = 0. A perfect conductor resists a change in magnetic flux, while a
superconductor expels it.

The resistance of a superconductor depends on the intensity of the external

magnetic field. The current in the superconductor has the same effect as a magnetic
field if it is sufficient to generate a critical field B,. At T = 0, this field must be B, and
zero at T = T,. Expression 8.4 can be written in terms of magnetic induction as

BeBe yo eS wet ee ( 8.6)

The above parabolic relation or equation (8.1) or (8.6) is often called Tuny's law
and it provides a convenient representation of the relation between temperature and the
critical field. The Fig. 8.4 depicts the temperature dependence of the magnetic field
necessary to quench superconductivity. The transition between the normal and
superconducting states is thermodynamically reversible, just as the transition between
liquid — vapour is reversible.

Bo
0.08 + B,(0)

Lead
Tesla

0.02

O ATT Pars pe
TK
Fig. 8.4 Variation of critical induction with temperature for a superconductor. The specimen is
superconducting below the curve and normal above the curve

One can get useful representation by considering a long thin sample in a magnetic
field. The magnetic induction field density B in a magnetic material is given as

B=y,(H+M)=B,t+uM oa. (8.7)

But B = 0 for a superconductor. H is the applied field intensity, B, is the applied
field. The magnetization curve expected for a superconductor as per
Meissner-Oschenfeld’s observation is shown in Fig. 8.5.

fs)
pe
ee Applied Physics

Bey sae

Fig. 8.5 Magnetization vs the applied magnetic field induction for a bulk superconductor.

There is a sudden change of the sign of magnetisation at the critical field. Such
conductors, which exhibit the above behaviour, are known as Type I superconductors.
Above the critical field, the specimen is in the normal state and the magnetization is
extremely small. The negative value of M corresponds to diamagnetism. The original
elemental superconductors like mercury, aluminium, zinc etc., are all classified as type
I superconductors. They possess a critical magnetic field for any given temperature.
They exhibit perfect diamagnetism.

When a superconductor is brought from infinity to a point at a position in the

field of a permanent magnet, work is done. The incremental work done on unit volume
of specimen is given as
GW=—DI edb 2 ae - “TBs (8.9)

this appears as the energy of the magnetic field. dB, is the change in magnetic induction
due to the shifting of the superconductor into the field over small intervals of distance.
M is the magnetization of the sample.

According to the first law of thermodynamics, the change in the internal energy
of the specimen is expressed as

dU= dO. dW) ee Secs (8.10a)

= dO MueidR. Bag e e A (8.10b)
when the B, = B,, the normal state and superconducting state can coexist in equilibrium.
For a superconducting state, the above equation can be written as (from 8.8)

1
dUs= dOrr Saad salem,
Ho

356
 Super Conductivity
i

At absolute zero, the sample is in superconducting state. The sample has to be

moved into the field of the permanent magnet, such that at a particular position, the
field becomes the critical field density (B,), the normal state is achieved. At absolute
zero,

i eS a epee (8.12)
Therefore the increase in the energy density of the superconductor is given by
integration of the equation (8.11) between the limits B, = 0 and B, = B,, and expressed as

2
Usk Bos GB) =UO)+ Peaches (8.13)
219

The sample is stable in either state. When the applied field is equal to the critical
field, AU= U,(B,) — U,(0) (from eqn. 8.12),

i.e., AU =U,(0)-U,O)= Be==

ness (8.14)
29

This is the change in internal energy when the specimen is brought from a
position where the applied field is zero to a position where the applied field density is
B,. AU is the stabilization energy density of the superconducting state at 0K. The
energy of the normal metal is independent of the field (the magnetic susceptibility of
the paramagnetic normal state :s neglected). At the critical field density B,, one gets

OAD SOG ee ee (8.15)

The energies of the superconducting state and the normal state are equal.

The above equation implies that the specimen can be stable in either state. AU is
the stabilization energy at absolute zero. When a metal like Aluminium is considered, it
has a critical induction of 0.0105 Tesla, then its stabilization energy is

gee(L05.¢1 08)
AU S Lee ei = 43.8 Joule —m-?
x 4n x

This is in agreement with the measured values.

357
 Applied Physics
aa

8.2.3 Thermal Properties :

The transition between the normal state and the superconducting state is a
reversible process. This transition is analogous to that of any phase transition like
vapour —> liquid —> solid and vice versa. For any system to be in equilibrium, its free
energy should be a minimum. The free energy is decided by the configuration of the
system. If it is supposed that the normal state has a free energy G,, and the
superconducting state, G,, then the destruction of superconductivity occurs at a critical
field H,, and this transition can be described as
h(E ee a ee ie mn Sey eit: (8.16)
In the absence of the magnetic field, the specific heat of the superconducting
phase is greater than that ofthe normal phase. The specific heat is entirely electronic in
a superconductor and the electronic specific heat increases with temperature
exponentially in the superconducting state. There is a sudden jump at the transition
temperature. (Fig. 8.7, the dotted curve).
Consider a typical isothermal - isobaric process, at a temperature TK and pressure
P, causing a transition of a system from initial state (I) to final state (H). Then, the
work done by it during transformation is AW = pdV, where dV is the change in
volume. AV = V; — V,. The transition is isothermal, pdV = p(V-— V;) < [U; — TS;]
— [U,— TS,], where U; — Ur refers to change in internal energy, and T (S,— S; ) refers
to the quantity of heat absorbed or evolved. (S;— S; ) is called change in entropy.
Hence [pV> — TS-+ Us] < [pV;— TS; + Uj]. We call this quantity U; - TS + pV as G
the Gibb’s free energy.

0 ro
———_}—

Fig. 8.7 There will be discontinuity in the specific heat for gallium.

358
Super Conductivity

This sort of variation at the transition is called a second order phase transition.
A phase transition is called the second order transition when there is no latent heat. The
variation of specific heat is expressed as
=E
Co=AT? Fiyzexp (= Bah (8.17)

In normal state, the specific heat at constant volume C, is expressed as

Cree CE i el ee ee eae (8.18)
A and y, are constants that depend on the nature of the superconductor. The
type of variation that is shown in specific heat with temperature also suggests that
there is an energy gap between the normal state and the superconducting state. This is
taken care of in Bardeen-Cooper-Shrieffer (BCS) theory of superconductivity which
will be discussed later.
The energy gap in a superconductor is of different nature as compared to the
energy gap in intrinsic semiconductors. A metal has conductivity due to electrons
which are treated as Fermi gas. The Fermi gas consists of a number of energy states
and electrons occupy these states. In a superconductor, the electrons interact with
each other lowering their energies. Therefore they go into a condensed state within the
Fermi gas and if the conductor has to become normal conductor, the electron gas has
to be given an energy. Hence it;must have an energy gap. The details of the energy gap
will be discussed in the later section. Here, the gap is tied to the Fermi gas, while the
energy gap in a semiconductor is tied to the crystalline lattice. In a semiconductor the
interaction between atoms of the solid gives rise to the changes in the state of electrons.
The individual energy states of the free atoms combine, overlap and give rise to energy
bands of the crystal. Energy bands are separated from each other by an energy gap.
The argument in the exponential in equation (8.17) is equal to one-half of the gap
energy. Table 8.3 gives the values of energy gap in some superconductors. The energy
gap decreases continuously to zero as the temperature is increased to T, from OK.
Table 8.3

Simei
eecgion gho»peebi!
359
 Applied Physics

772A

Normal metal Super conductor

a) Conduction band in normal state b) Energy gap at the fermi level in the
super conducting state.
Fig. 8.9 Energy gap in a superconductor

8.2.4 Isotope effect:

The effect of isotopic constitution was first studied by Onnes and Tuyn (1922).
They found that ordinary lead (atomic weight 207) and uranium lead (atomic weight
206.06) differ in transition temperature by more than 0.025K. The isotopic constitution
has been found to have an effect on the transition temperatures of elements since then.
It is now known that the transition temperature varies to a good approximation as the
square root of atomic mass. Similarly the critical field also varies with the isotopic
mass in exactly the same way as does T,. The electronic specific heat of an element in
its normal state is independent of its isotopic mass. The experimental results on the
isotopic constitution of an element can be expressed by the relation of the form.
Miles ICONS tne Nn ltd ool ack Fat (8.19)
a is called the isotopic effect coefficient and is given as

_ olny
O(In M)
The value of@ is normally ofthe order of 0.5, but there are deviations for many
elements. There is no theory to explain the variations of « values for different elements.
The above observation indicates that the phenomenon of superconductivity is
determined in some way by the interaction between electrons and atoms in the
superconductor. Electrons are responsible for the conductivity of metals. The electrons
interact with lattice vibrations and their interaction affects the conductivity . In
superconductors also there should be interactions between electrons and lattice
vibrations.
The values of a are given in Table 8.4 for various elements.

360
Super Conductivity

Table 8.4

8.2.5 London penetration depth :

The characteristic property of a superconductor is that B = 0 inside and therefore,

that any current existing must be superficial. The magnetic susceptibility of a
superconductor is -1 (SI units). However, this can not be generalised and the current
can not entirely be superficial. There must be a certain depth, however small, of
penetration of the magnetic field into the superconductor. London (1935, 1939) observed
the penetration effects as important when the samples of superconductors were taken
as thin films. In bulk sample, the depth is very small. Let it be supposed that the field
inside the surface of an infinitely thick sample falls from its value Hp. If the field varies
as exponentially from the surface into the material, then it is expressed as the value at
the surface to a value H (x) inside along the x direction. The fall in H is expressed as

_ H(x)dx
ae tiy

\ Bix) = B, rs

|Sane
Semi infinite
Superconductor

Fig. 8.10 Shows the penetration of an applied magnetic fie/d into semi infinite superconductor.

361
 Applied Physics

i.e, H=-f 0 aeH(xoad 2 2 ioe

Hp

In terms of magnetic induction or magnetic field density, we can write the

above equation as

5 he (8.21)
2d is known as London penetration depth.
The macroscopic description of a superconductor is given as a body with B = 0
and E = 0 inside it. London considered a superconductor as a condensed state consisting
of superelectrons and normal electrons. The total number of electrons in the
superconductors N is equal to the sum of the concentration of normal electrons n, and
the concentration of superelectrons n,. The concentrations are taken as per unit volume.
If the superelectrons move with a velocity v,, then, the equation of motion for
electrons is

m— Fa =eEi eel Mp ha sah eeltga ah (8.22)

LetiJ, represent the super current density, then it is given by

JTS EE SA ra) PAD UA ea (8.23)

dJ dv e’n
dt *=n,.e—*
Ss dt
=| —+
m
|E ee eeeeee (8.24)

Taking curl on both sides of (8.23)

dJ 2 2
Ca -cu a e-(!* Jou|Sea ae (8.25)
dt m m

But Curl E = —6B/6t (equation 8.3) [see Appendex - I]

cal dJ, [| OB
ur aI ae
dt ¥ m ot
or
2
Cail
Curl J, =—- - BIO eeh Cee Gee Seer (8.26)

The electric field can not penetrate, but the magnetic field penetrates the
superconductor. But according to Maxwell’s equation based on Ampere’s law,
Curl B = 1,J, (n, is equal to the density of the superconducting electrons),
it can be
seen that

362
Super Conductivity

CuniGunkB ay Curlty fee ao oer See wee: (8.27)

But Curl Curl B = gradient div B — V’B

2

ise eaMeH =i uel Jip Becen Sue, conlSomes (8.28)

m

: 1
i.e. _vy2B = iB mies (8.29)

h
where r -( ~ ef
<i (epee
one ae ae 8.30
ee (8.30)

i.e. B decays exponentially with distance inside the specimen from the surface.
Equation (8.24) and (8.26) are known as London equations.
From above, it can be seen that J, = 0 wherever B = 0, i.e. well inside the
superconductor. A, the London penetration depth is a function of temperature and the
relation between temperature and A is
jd) Ved We ACU IN of© talc a eee (8.31)
where A(0) is the London penetration depth at T = OK.
London Penetration depths at absolute zero for various metals is given in the
following Table 8.5. The equation (8.31) also implies that the density of superconducting
electrons increases from zero at T, to maximum at 0 K.
Table 8.5

Sn

Thus, in a superconductor, Meissner effect is not complete, London penetration

depth 4 is a fundamental length that characterizes a superconductor. i is of the order
of 10-8m. The penetration depth 4 decreases as temperature is reduced.
The diamagnetic nature of a superconductor can be demonstrated by floating
magnet experiment. This was one of the earliest expriments done. If a small permanent

363
 Applied Physics

magnet is lowered over a superconducting surface, the lines of force of the magnet
can not enter the superconductor and this results in a repulsive force large enough to
over come the weight of the magnet. The Fig. 8.11 shows a floating magnet above a
superconductor. This sort of levitation is important in applications.

1a is bowl of Supercond
uctor
Bis tloating magnet

Fig. 8.11 Amagnet floats over a superconducting bowl

83, Type I Superconductors and Type II Superconductors :

8.3.1 Type I Superconductors :

Type I superconductors are those super conductors in which the magnetic field
flux is expelled totally. When the magnetic field reaches H,, the field lines are totally
expelled. The magnetization M is equal to (—1/j19). The magnetic behaviour of type I
super conductors (Table 8.1) come under this category. The transitions in the type |
super conductors are sharp.

8.3.2 Type If Superconductors :

The magnetization curve for type II superconductors is shown in Fig.8.12.

Normally the type II superconductors are alloys or transition metals with high values
of the electrical resistivity in the normal state. These metals as compared to the good
conducting metals like copper have lower values of electronic mean free path in the
normal state. This is the mean free path, which determines their magnetic properties.
When one considers the example of lead in pure form, it exhibits type I superconducting
behaviour. The critical field is around 0.8 Tesla. However in the case of indium- doped
lead, the flux lines start penetrating around a field less than 0.8 Tesla (Bc,), but the
superconductivity is not destroyed. The specimen shows the superconducting property
upto the critical induction field Be .But above Bc, the flux density is expelled from the
sample. The specimen is said to be in mixed or vortex state between Be, and Bc,. In
the vortex state, the material specimen has both normal and superconducting regions
and can possess zero resistance and partial flux penetration (Type II superconductor

364
Super Conductivity

permits some penetration of magnetic flux above a first critical flux Bc,). It is threaded
through by flux lines at the normal regions. The vortex regions are essentially the
filaments of normal regions that run through the sample when the applied field density
is greater than Bc, as shown in Fig.8.13. As the applied field is increased in strength,
the number of filaments increases until the field density reaches Bc, and the sample
becomes normal.

Fig. 8.12 Magnetization curve for type II superconductor

The vortex state can be viewed as a cylindrical swirl of super currents surrounding
a cylindrical normal metal core which allows some flux to penetrate the interior of the
type II superconductor. Associated with each vortex filament is a magnetic field which
is maximum at the core center and falls off exponentially out side the core with the
characteristic penetration depth 2.. The super currents are the ‘source’ of B for each
vortex state. The radius of normal core is smaller than the penetration depth. The value
of Bc, generally is far greater than Bc,. Many alloys of Nb, Al and Ge etc., are type II
superconductors and the Table 8.6 gives a list of some useful alloys with their Bc,
values.
Table 8.6 : Superconducting alloys along with their Bc, values
 Applied Physics

The type II superconductors can carry large super currents even in fields
approaching 100 KG. This has led to the development of strong field super conducting
magnets. The interface between a normal region and the superconductor region should
have a surface energy . If the surface energy is positive, then the superconductor is
type I superconductor. In type II superconductor, the surface energy is negative as the
magnetic field is increased. The penetration of the applied field into the superconducting
specimen causes the surface energy to become negative. It should be remembered that
(Bc,, Bc,)'’? = B,, the critical field for the pure specimen, a type I superconductor.
They are some superconductors in which Bc, is as large as 35 Tesla at absolute zero.
The type II superconductors have some resistivity in mixed or intermediate state. The
normal material takes the form of laminae or filaments in thin films of superconductors.
Each normal region is surrounded by a superconducting region. In the mixed state, the
external field will penetrate the thin normal regions uniformly. The field will also penetrate
in the superconducting regions to some extent. The mixed state is stable as the
penetration of the applied magnetic field into the superconducting region causes the
surface energy to become negative. If one observes the type II superconducting
specimen between Bc, and Bc,, a pattern as shown in Fig. 8.13 is exhibited. This is
similar to the domain pattern of a ferromagnetic material.

Magnetic flux channels

through normal regions in
the mixed state
IG MUAY Sigtc

Fig. 8.13 The mixed state of'a type I superconductor

8.4 Flux Quantization:

The magnetic flux is ejected out at the transition temperature due to the induction
current in the specimen which produces a flux equal and opposite to the original flux
due to the magnetic field. As there is no resistance, the current due to induction
in the
specimen persists as long as the field is on. As a result of the persistent current,
the
specimen acquires a negative magnetic moment producing a field in a direction
opposite

366
Super Conductivity

to the direction of the applied field. Due to this negative moment, a superconductor is
treated as a diamagnetic substance. When the field is removed, the persistent current
goes to zero.
It appears from above that the persistent current exists as long as there is external
field. This opinion can not fully support the occurrence of superconducting state when
a conductor is cooled in the absence of the field as was experimentally verified. Meissner
observed that persistent currents could be set up in any sample of suitable shape.
Consider that the sample is taken in the form of an anchor ring as shown in the
Fig.8.14.

T<To

Sample first cooled

: and field applied
oe =(=
Sample cooled
in the presence of fet] Cooled

L/
L

Pet flux lines

Fig. 8.14 Aring in magnetic field

Suppose the magnetic field is put on while the ring is in normal state. Then the
magnetic flux penetrates through the ring, When the ring is cooled to the transition
temperature while the field is on, the flux is excluded from the body of the ring as a
result of the induction current on the surface of the ring. However the flux is trapped
inside the ring i.e. in the hole of the ring as the flux lines can not thread through the ring
now. A large current is induced by the collapsing field that maintains the trapped flux.
This trapped current is called persistent current. It will be present even after the field is
switched off as the resistivity of the superconductor is zero.
However, if the ring is cooled first and the field is switched on after the critical
temperature is reached, the magnetic flux is excluded from the ring. There will not be
any flux trapping in the hole of the ring. The current induced in the ring keeps the flux
out of the ring. This current disappears when the field is switched off.

367
 Applied Physics

A superconductor allows the current through. There is a minimum quantity of

current that can exist in a superconductor without destroying the superconductivity.
This minimum current is called critical current I,. Suppose the superconductor is in
the form of a wire of radius r. The current Ic in it will set up a transverse magnetic field
H, and H, = I,/2 mr. For any other current value I, Hig =]/2 xr.
The magnetic flux that passes through the ring is the sum of the flux from the
applied field and the magnetic flux from the superconducting currents which flow in
the ring. The flux lines from the applied field are not quantized. However, the flux from
superconductor adjusts itself such that the total flux is quantized. The magnetic flux is
quantized in integral multiples of ) = h/q, and q is equal to '2e’, the charge of an
electron pair. The value of h/q was found experimentally to be 2.07 x 10-!5 weber. It
confirmed the existence of an electron pair in the superconductor. The unit of flux is
called 'fluxoid'.

835 BCSTheory:

Nearly 50 years after the discovery of superconductivity, Bardeen-Cooper-

Schrieffer in 1957 produced their epoch making theory of superconductivity, which is
now known BCS theory. It explains the microscopic origin of superconductivity . The
theory requires very rigorous mathematical formulation that is outside the scope of this
book, but its basic features can bt’ mentioned here. The basic features on which the
theory stands are
1. The electron-phonon-electron interaction proposed by Frohlich.
2. The formation of Cooper pairs of electrons.

3. The existence of energy gap.

(1) Electron-Phonon-Electron interaction and formation of Cooper pairs
:
The two observations that the critical transition temperature T, varies as M-!2
where M is the atomic weight of the metal that is superconducting and the
sudden
discontinuity in specific heat at transition temperature did make Bardeen
to think that
the superconducting state should involve the dynamics of ionic motions, lattice vibration
s
and associated phonons. Further then T, attains value zero when M approach
es infinity.
When an electron with a wave vector K, encounters a lattice ion (positive
ion cores),
it suffers a Coulomb interaction. As a result of this interaction, the momentu
m of the
electron decreases by an amount proportional to K,—q and the lattice
is set into motion
gaining a momentum from the electron proportional to q. The lattice
vibration is quantised
and characterised by energy packets called phonons (virtual). Another
electron with a
wave vector K, (Fig. 8.15) may enter this volume and this electron
interacts with the
distorted lattice and gains momentum, the gain depending on the
phase of the vibration
at that instant. The electron energy is lowered.

368
Super Conductivity

Fig. 8.15 Electron Phonon-Electron interaction or Frohlich interaction

The lowering of energy indicates that the force between the electrons is attractive
and the interaction is through phonons. Therefore it can be thought that the electron
with wave vector K, emits a virtual phonon q which is absorbed by the second electron
with wave vector K, and the two electrons are scattered with wave vectors K, — q and
K, + q. The process is a virtual one and the energy is not conserved. The process is
very rapid with the emission and absorption of the phonon. The short lived phonon is
called virtual phonon. The two electrons should have equal and opposite spins and their
energies are the same. Then the interaction between them is attractive. As a result of
such attractive interaction, there exists pairs of electrons called Cooper pairs. They are
called the Cooper pairs in honor of L. N. Cooper who was the first to show that
binding of electrons in pairs is energetically advantageous. A cooper pair is denoted by
(Kt, KJ ). Each electron participating in the pair formation is called mate.
Therefore, the electrons in a superconductor are in the form of Cooper pairs. A
Cooper pair is a system of two electrons having equal and opposite momenta and spin
(Kt, KJ). The energy of the pair is lowered when compared to the free separate
electrons. All electrons in the metal can not participate in this process and only a
fraction of the total electrons in the specimen can do so. The energy difference between
the free State of the electron (energy of individual electrons) and the pairs appears as
the energy gap at the Fermi surface. (Fig. 8.16) Therefore, the energy gap is said to be
tied to the Fermi gas. The attraction in between Fermi electrons tends to reduce their
total energy (this total energy must be removed from the electron gas to transfer it to
the superconducting state). The energy spectrum of electrons must then exhibit an
energy gap.
(2) The energy gap:
The existence of energy gap can be proved experimentally. One such experiment
is the study of the spectrum of absorption of electromagnetic waves by a
superconductor in the microwave band. At temperatures close to 0°K, a superconductor
does not absorb energy of incident radiation until the energy of incident radiation exceeds

369
 Applied Physics

the width of the gap (2A). After absorption of energy the electrons become free or
normal.

Energy
E

0.2 0.6 1.0

Electron occupancy
Ve
Tc

Fig. 8.16 Energy gap in a superconductor

Although BCS model is a brilliant model describing the mechanism for

superconductivity, it can not explain which materials are superconductors and which
materials are not. For example gold, copper and silver which are excellent conductors
at room temperatures are not superconductors even at temperatures very close to
0°K.
Mercury, tin, lead etc. which have higher resistivity at room temperature (compared
to
gold, silver) are excellent superconductors at low temperature. We have new materials
which are insulators at room temperature and are now found to be supercondu
cting at
liquid nitrogen temperature (These superconductors are called high T, superconduc
tors).
The energy gap is a function of temperature. At absolute zero, a supercon
ductor
does not absorb any energy when microwave radiation falls on it until
the energy
supplied by the radiation is greater than the width of the energy
gap 2A. When the
absorption takes place, the electrons go to the energy states that lie above
the gap. The
Cooper pairs are broken and the superconductor becomes a normal
conductor. The
energy gap represented the energy needed to break up one of the
cooper pairs. The
width depends on the number of unpaired electrons. The presence
of energy gap also
explains the occurrence of persistent currents. In a Cooper
pair, the momentum of
individual electrons can not be changed. The pair can not
be scattered as a whole
unlike the individual electrons by lattice ions. Each Cooper
pair is associated with a
wave function and this wave function spreads over a large
volume and overlaps with
wave functions of other Cooper pairs. The volume of a given
pair may extend to over
one million other pairs. Therefore, the superconducting
state is a collective state. The
conduction electrons in a superconductor are condensed
into a single entity some thing
like a big macro molecule extending over entire volume
of the superconducting system.
They move as a whole.

370
Super Conductivity

The BCS ground state is separated by a finite energy gap (2A) from its lowest
state. The Cooper pair resists the change of its state as it is a lower energy state and the
Zero resistivity of a superconducting state is based on this fact. As stated earlier, the
amount of energy required to break a pair is 2A (A for each electron). The electrons
become independent of each other at the transition temperature and are said to be
‘normal’. The thermal vibrations of the lattice break the pair. Then the electrons will
appear in the energy states above the gap leaving vacancies below the gap. Thus, a
certain equilibrium concentration of ‘normal’ electrons sets in at every value of
temperature from 0°K onwards. This phenomenon is similar to the existence of holes
and electrons in an intrinsic semiconductor. In a superconductor the current can exist
despite the presence of gap. In a semiconductor, electrons have to be excited into
conduction band. The interaction and pausing of electrons extends over a considerable
distance in the specimen. The maximum distance unto which the states of pairs of
electrons are correlated to produce superconductivity is known as coherence length L.
It can be many thousands of atomic spacing (~10~-m) apart. The properties of a
superconductor depend on the correlating electrons within a volume €}, called
coherence volume. The ratio of London penetration depth to the coherence length is

K = 4/6, . For type I superconductors K < To’ and for type II superconductors

Kee 7 The difference in K values is related to the surface energy . Therefore the

interaction of electrons follows the long-range order. The electron states responsible
for superconductivity exist within k T, (k is Boltzman constant) of the Fermi surface.
Their life time t is obtained from uncertainty principle as t ~ h/2x kT,. The wave
function extends over a distance €, and € = 2kv,/nA where vr is the velocity of
electron at the Fermi surface. This is shown in Fig.(8.16).
BCS theory explains the thermal properties, critical fields and most of the
electromagnetic properties of superconductors in terms of the energy gap. The
penetration depth, coherent length, flux quantization, isotope effect, etc., emerge as
natural consequence of BCS theory . London equations, Meissner effect become a part
of BCS explanation of superconductivity . The rigidity of macromolecule to a change
of its state also explains Meissner effect. When a superconductor is placed in a magnetic
field, the wave function of the macromolecule does not change i.e. the orbital
configuration of the electrons does not change. However, the electrons or the Cooper
pairs, suffer a change in velocity while remaining in the same orbital configuration.
The change of velocity causes an induced current, which expels the magnetic field
from the bulk of the superconductor.

371
 Applied Physics

86 Application ofsuperconductors :

The applications of superconductors are diverse and many. Some of them are
mentioned below.

1) Electromagnets : The type II superconducting wires are wound in the form of

solenoids to generate strong magnetic fields of an induction about 20 tesla. The
size of these magnets will be small compared to the ordinary metal wire wound
electromagnets. The superconducting wire has to be maintained at liquid-He
temperature and the cost of the power will be 1000 times less than the cost of
power required in an ordinary electromagnet of the same capacity. The energy
stored in the field will be substantial about 4 KJ/m3 at 10 Tesla. The
superconducting wires are also used in electric machines and transformers. The
transformers do not require any magnetic cores and the superconducting wires
generate strong fields.

2) Bearings : The Meissner effect is made use of in bearings. The mutual repulsion
between two superconductors due to the repulsion between the magnetic fields
they generate can be embodied in bearings and the bearings would operate without
power loss and friction.

3) High power transmission : The superconducting cables permit high power

transmission without power loss. The superconducting wires used in the above
applications are produced from solid solution alloys such as niobium-zirconium,
niobium-titanium, or niobium-titanium- zirconium. High temperature
superconducting wires may soon replace the wires operating at liquid-He
temperatures.

4) Cryotronics : The new branch of electronics based on superconductors low

temperature electronics or cryogenic electronics or Cryotronics has become a
very important branch of electronics. This includes creating (1) Supersensitive
miniature receivers capable of detecting extremely weak radio signals. (2)
Manufacture of large-scale integrated and very large-scale integrated circuits
(VLSI) for the development of new generation of electronic computers, (3) An
increase in the frequency stability and frequency selectivity of microwave
apparatus, and (4) The extension of radio wave bands into the microwave region
up to the infrared, earlier considered inaccessible for long range reception. There
are many cryogenic devices like memories, logic units, cryotron switches,
oscillators, amplifiers, resonators, modulators etc., A cryotron switch and a
memory cell will be described here. It relies on the effect of inducing a persistent
current in a closed superconducting circuit a destroying the same state by
energetic field.

372
Super Conductivity
ea
er

a) Cryotron Switch : \t consists of (Fig 8.17) a controlled gate | a strip of thin

film tin (T, = 3.7K) and a control strip 2, a thin film of lead (i= 7:26)
deposited on a substrate and separated by a dielectric layer.

(a) (b)

Fig. 8.17 Cryotron switch

Fig. 8.17 a illustrates the schematic diagram of a simple cryotron. It consists

of a controlled (gate) wire A commonly made of tantalum with T, = 4.4 K
and a control coil B from niobium (T, = 9.3K) or lead (T, = 7.2K). At a
working temperature slightly below 4.4K, the gate wire is kept in the
superconducting state and offers no resistance to the flow of current I, through
it. When a current I, passes through control coil, the superconducting state
of the gate breaks down and it is cut into normal state, possessing a finite
resistance R. This is something like a relay.
The cross film cryotron offer high speed operation Fig. 8.17(b). consists of
two crossed strips deposited on the substate and seperated by a thin dielectric
layer. The control strip | is the gate (normally tin) and control strip 2 is made
of lead.
At working temperature T slightly less than 3.7K, the gate is kept in
superconducting state. A current through it passes without resistance. When
current is passed through 2, the superconducting state of the gate 1 is broken
- and it goes into the normal state. Therefore the current in the circuit consisting
the gate (1) is broken. By varying the current in 2 alternately one can get
opening and closing of the gate in the circuit having the gate. The thickness
ofthe strips is about 0.1 microns and by suitable means the switching speeds
can be made 10~’sec. Cryotrons can perform all logic and storage operations
required in computing with essentially zero power dissipation except when
switching. The low power consumption is an attraction to use the
superconducting devices, but the reproducibility, maintenance of the equipment
are the formidable obstacles for actual use.

Sve
 Applied Physics

b) Superconducting Memory Cell: A closed superconducting circuit, in which

persistent currents can be induced, can work as a memory cell. When the
superconductor is in the normal state, there will not be any current in the
circuit. It is called ‘0’ state. In the superconducting state, the circuit will
have a current and it is called ‘1’ state. A superconducting memory cell is in
the binary system depicted in the Fig. 8.18.

at ts @ AB (b)

Fig. 8.18 A superconducting memory cell

A lead substrate S is coated with a thin dielectric layer on it. A thin tin strip T
is deposited on the substrate in the form of loops T and connected into groups
by a bit line L. The tin strip is coated with a dielectric layer on top of it. Over
the tin strip, lead strips A and B are over layered. To write the information, a
current is passed through the bit line.

Simultaneously current is fed into the strips A and B. They produce a magnetic
field H capable of destroying the superconducting state in the tin loop located
below the lead strips. The current is made to flow only in the upper section of
the loop. This state is maintained stable even after cutting off the current in
the lines A and B, although the tin strip becomes fully superconducting. Now
when a current is allowed to go through the bit line, a persistent current will
circulate in the loop, which will then store the delivered information.
(Fig. 8. 18(b))
Reading out the information, requires a total current in A and B that destroys
the superconducting state in the tin loop where this state was broken earlier.
This eliminates the current in the loop and induces the information pulse in
the bit line (Fig. 8.18(c)). The capacity of such a memory cell is up to billion
bits and the speed is in the order of 10-6 — 10-7sec.

374
Super Conductivity
ee
eee

5) Josephson Effect : Josephson junctions are useful in creating fast memory elements
of electronic computers. A brief discussion will be given here. It has been said
that a superconductor possesses cooper pairs. All the cooper pairs are in the
same state and posses one wave function whose amplitude is dependent on the
number of cooper pairs. Based on these, Josephson ( 1962) predicted Josephson
effects. Suppose a piece of a superconductor is connected to a voltage source,
ammeter and voltmeter as shown in Fig. 8.19.

(aj (p) (c)

Fig. 8.19 Super conductor Junction

When the circuit is closed, the ammeter will read a current I which is present in
superconductor. The voltage across it is zero, as there is no resistance. Let this
superconductor be cut, into two halves (Fig. 8.19(b)) and a thin (1 mm) insulator
be inserted into the gap between the superconductors and let them be joined. On
closing the circuit a direct current still flows through the superconductors. That
means, the current also flows through the insulators. This is due to the tunnelling
of charges across the junction. This is called dc Josephson effect. Here the
cooper pairs tunnel through (tunneling can only be explained by quantum theory)
from one conductor to the other through the potential barrier (insulating film) and
return to the first conductor through the external circuit. The currents should not
exceed a critical value (I,). If the currents exceed I,, a potential difference develops
across the junction indicating a resistance in it (Fig.8.19(c)). This principle is
used to develop Josephson tunnel junction cryotrons. Since the transition from
the zero to a finite resistance is not related to the elimination of the superconductivity,
the speed of the transition is much higher than common superconducting
cryotrons. The speeds are of the order of 10-!! sec and the dissipated power is
of the order of only 10-7 Watts. The voltage characteristic of a Josephson’s
junction is given Fig. 8.20.

375
 Applied Physics
INE a ae

current 0)
Tunneling
(T
0)
>

(T
Tunneling
Voltage ———-»

Fig. 8.20 The current voltage characteristic of Josephson junction

Josephsons effects can find application in other fields of applications also.

The tunneling current can be given as I = I, sin g where is the phase difference
in the state of the waves describing the cooper pairs on both sides of the insulator.
When current exceeds I,, a resistance appears and the energies of cooper pairs on
either side of the barrier (insulators) differ by 2qV, but the energy is zero, then the
wave frequency difference Av = 2qV/h and this will appear on both sides of the
junction. The phase difference is @ = 2nt(2qV/h) and the superconducting current

seg _ ( 4nqV rere a, Re

through the junction becomes I = I,sin (a) . This is a sinusoidal current whose

oscillation frequency is equal to 2qV/h. When V = ImV, the frequency is of the order
of 480 GHZ. This is in Microwave region. This type of phenomenon enables one to
construct microwave resonators (with a quality factor of 10!!). The theoretical limit
of the quality factor for conventional resonators lies near 1-8 x 103.
Cryotronics are supposed to become very popular in future as the low temperature
devices operate with low noise levels and raise the performance level of the electronic
devices to a high degree.
The outer space is vacuum, extremely cold and radiation full. Rockets nowadays
are based on cryogenic liquids (liquid oxygen, hydrogen, helium, Nitrogen). Therefore
the satellite borne or rocket-borne apparatus should not only be mini in size but also
should be able to use the low temperatures for their operation. In such cases compact
electronic devices based on Cryotronics are useful.
6) Superconducting magneto meters : They are capable of measuring magnetic
field whose induction B is smaller than 10-!! Tesla. SQUID (Superconducting
Quantum Interference Device) is useful for measurement of very low magnetic

376
Super Conductivity

moments. If a Josephson junction is placed in a magnetic field, the current density

in the junction varies just like the single slit diffraction pattern in optics. If two
junctions are placed side by side, then the super current density exhibits a fringe
pattern like in the double slit arrangement in optics. SQUID magnetometer is
useful for the precision determination of h/e; SQUID are fundamentally
superconducting rings that act as storage houses for magnetic flux. SQUID is
extremely sensitive to very low magnetic fields and it is able to detect the magnetic
field of earth very accurately. SQUID is used to detect the mineral deposits or oil
deposits inside the earth. Superconducting magnetic maps are useful for this
purpose. Squids are useful in medical diagnostics of heart and brain. Nuclear
magnetic resonance imaging is useful in medical diagnostics.

8.7 HighT, Superconductors :

The maximum transition temperature that could be achieved before 1980’s was
23.2K in Nb,Ge. This discovery gave a hope that some metallic alloy systems could be
made to have transition temperatures around liquid hydrogen. But the success could
not come. There was a search for alternate materials which could exhibit higher transition
temperatures. Lithium- Titanium Oxide was found to exhibit a transition temperature
around 13.7K. This gave a hope that ceramic oxides which are regarded as high insulating
materials at normal temperatures can be made superconducting by a proper selection
of the components as mixed oxides. The mixed oxide of Lanthanum-Barium-Copper
oxide exhibited a transition temperature of 30K as discovered by Zurich scientists
Muller and Bednortz in September 1986. The above mixed oxide was found to be in
superconducting state below 30K. It would exist in three structural states. But one of
the structures is perovskite structure and the oxide with perovskite structure only
would exhibit superconductivity . The same mixed oxide when subjected to high
hydrostatic pressure was found to possess a transition temperature of 55K. The
superconductivity of the oxide was linked with variable valancy states of copper in it.
In La,CuO,, the valancy of copper ions is +2, but when a small amount of barium or
strontium (a few atomic percent) was added to La,CuO,, some of the copper ions
would go into +3 valence state. As a result of this the oxygen ion concentration would
deviate from stochiometry and there would be a deficiency of oxygen ions in the oxide
compound. When this deficiency of oxygen was carefully controlled, by keeping the
samples in oxygen atmosphere, it was found that the oxide compound would exhibit
superconductivity in the temperature range 30-40K. American scientists Chu and his
co-workers found in January 1987 that when yttrium (Y) replaced lanthanum in
La-Ba-Cu-O mixed oxide, thereby forming Y-Ba-Cu-O mixed oxide, the resulting mixed
oxide had become superconducting at around 90K i.e., above the boiling point of liquid
nitrogen (77K). This discovery was like a lightning in the basic research on the search
for new superconducting materials. The Y-Ba-Cu-O (Fig. 8.21) also possesses a
perovskite structural unit and the actual chemical formula was Y,Ba,Cu,0, (YBCO).

B77]
 Applied Physics

Y~ 123
Fig. 8.21 Structure of YBCO

The above oxide also contains a deficiency of oxygen with the chemical formula
Y ,Ba,Cu; O,_; where b indicates the deficiency of oxygen and is in the range 0 to 0.1.
The YBCO mixed oxides are known as | 2 3 systems. In YBCO, the oxygen deficiency
has to be regulated by suitable oxygen treatment of the samples. The YBCO oxide can
be prepared in any laboratory very easily. Lanthanum oxide, Barium oxide, Copper
oxide and Yttrium oxide should be taken in appropriate ratios as per the chemical
formula and mixed thoroughly. The mixed powders should be made into pellets by
pressing the powders in a steel die at appropriate pressure. The pellet is kept in a muffle
furnace with a temperature controller. The furnace should be heated to 900°C and
maintained at that temperature for 5 hours and subsequently cooled to room temperature.
The pellets would be in green colour normally. The green pellet should be kept in the
furnace in which oxygen at appropriate partial pressure can be passed over the samples.
The pellets should once again be heated to 900°C in the oxygen atmosphere and retained
there for 2 hours. This treatment allows the deficiency of oxygen to be controlled
systematically. The pellets should be cooled to room temperature and the pellets appear
black. The superconducting nature can be tested on this pellet by simple test. The

378
 Super Conductivity

pellet is kept in a ceramic bowl. A small tiny magnet is kept over the pellet. When liquid
nitrogen is carefully poured in to the bowl, the magnet levitates over the pellet indicating
the superconducting nature of the pellet.
The discovery of high T, superconducting oxides has enthused many materials
scientists, physicists, chemists and technologists throughout the world to try and prepare
more and more oxides for searching superconductivity in them. The discovery also
has given a hope that one day, a superconducting material with a transition temperature
around room temperature could be discovered. The superconductor alloys and pure
materials have transition temperature below 10K. Many practical applications involving
the devices have been found and the devices have been confined to very rich laboratories
and countries. The liquid helium is very costly and maintaining it at very low temperatures
also involves lot of cost. Therefore, the applications of superconductors even though
visualized in many fields could not be brought tc regular usage on large scale. But liquid
nitrogen is easily available, cheaper and does not require much sophistication in
maintenance. The applications of superconductors can be extended even in developing
countries.

The structure of YBCO has oxygen deficient perovskite structure having

orthorhombic symmetry . The oxygen content can be changed from 6 to 7 by
appropriate heat treatment in oxygen atmosphere. The coherence length is of the order
of 10-20 A. It exhibits type II superconductivity . Bc,(0) is of the order of 150-200
tesla. Bc,(0) values are very low and of the order of 0.005 — 0.05 Tesla. Tapes can be
made and the critical current density in tapes is of the order of 10© Amp/cm?. The
penetration depth is 12A. The average specific heat is 200 JKg-! K ~!. In principle if
the tapes or the thin films of YBCO can be made they can carry current densities of the
order of 10!°Amp/cm? without any loss. Being oxides they are not influenced by
atmospheric conditions, and have good flexibility . As there is a great demand for
energy conservation, the need for alternative energy sources in transmission systems
is needed. Probably these oxide superconductors may soon replace copper and
aluminium in high power transmission lines. Oxides may also find their use in other
applications of superconductors. Apart from YBCO compounds, the compounds based
on bismuth like Bi-Sr-Cu-O or Ba-Ca-Cu-O systems are also superconducting in nature.
For example, the compound Bi,Sr,CaCu,Og ,,, (2212) with the transition temperature
85K, Bi,Sr,Ca,Cu,O0I0,9,, (2223) with a transition temperature of 110K belong to
bismuth based compounds. There is a continuous search for new oxide compounds,
which may exhibit transition temperatures much higher than boiling liquid nitrogen
temperature and with reasonably high current density values and high Bc, values. The
following table gives a list of few high temperature superconductors with their transition
temperature.

Sih
 Applied Physics

Table 8.7 List of some high T, superconductors and their transition

temperatures

lee: wK

SmBa,Cu,O,. 88.3

DyBa,Cu,O,, |
‘Oo
© 1h
—

Bi,CaSrCuO 6+x 60420

BESECaci.Or,. ooNn ro)

Bi,Sr,CaCu,0O,,,,

11 Ba€a,Cu.Or. 125.0
THBalCa,Cu,0,4 105.0
HgBa,Ca,Cu,O,,. 130.0

Example |: A long thin superconducting wire of ametal produces a magnetic field 105
x 103 amp/m on its surface due to the current through it at a certain temperature T.
The critical field of the metal is 150 x 103 amp/m at T = OK. The critical temperature
T, of the metal is 9.20K. What is the value ofT ?

H, = H, [I- (1/T,)?)

ae be 43, Ke
Example 2: A lead wire has a critical magnetic field of 6.5 x 103 Am: at OK. The
critical temperature is 7.18K. At what temperature the critical field would
drop to 4.5 x
103 Am!. The diameter of the wire is 2mm. What is the critical current
density at that
temperature ?

380
Super Conductivity

Ho To

[Ho He] _ T*
Ho =
[Ho Help _72
O

— 6.5x10°
=4.5x10°
i.e., T2 6.5x10° 7.18
x 7.18 = 15.86 K2

Le, Loe
The critical current density

Hcx2mr 4.5x10°x2
ee Sealy = = = 9.0 x 106Am? at 3.98 K.
Tr 1x107

Example 3 : The transition temperature of Nb is 9.5K. The critical magnetic field

induction density is 0.2 Tesla at 0 K. What is its stabilization energy ? What is the
magnetic field induction at 3.5K.

By) ;
Ave! e) = (0.2) x7 _ 159x107 Joules.
a, 2x4x22x10°

B=B,[1-(1/T,)’] = ofBaro 0.173 Testa

2 2

Questions:
1) What is Superconductivity?
2) What is Meissner effect?
3) Explain the terms
(a) Critical transition temperature, (b) Critical field, (c) Persistent Current
4) What is the relation between temperature and the critical magnetic field in a
superconductor?
5) Define the term ‘London Penetration depth’.

381
 Applied Physics

6) Does the transition temperature depend on the applied magnetic field in a

superconductor ? If so, how ? A superconductor has T, = 3. 72K in a zero
magnetic field. The critical induction field at 0° K is 3.2 x 10 —2 Tesla. Find the
critical field at 2.5K.
7) What is the important feature of Type I Superconductors. Mention a few type I
superconductors.
8) How does the magnetization M vary with applied magnetic field in type II
superconductors?
9) What is isotope effect in superconductors?
10) What are Cooper pairs? How are they formed in a superconductor?
11) What is the difference between the energy gap in a superconductor and energy
gap in an intrinsic semiconductor?
12) Explain the meaning of coherence length in superconductors?
13) What is magnetic levitation? A superconducting lead wire has a radius of 1 mm.
Is it possible to float it in the magnetic field of earth. The horizontal component of
earth’s magnetic field is 0.5 gauss.
14) What is Josephson effect?
15) What are the properties that do not change in the superconducting state and
normal state?
16) What are Superconductors ? Discuss the earlier experiment on superconductivity
and how zero resistivity was thought to be the characteristic of superconductivity?
17) Can the superconductor be thought of only a zero resistance material ? If not
why ?
18) Explain how Meissner effect proves the superconductor to be a perfect diamagnetic
material? What is the difference in critical fields for lead at 7.1 K and 1K.
19) Obtain the relation between London penetration depth and the applied magnetic
field. How does the London Penetration depth change with temperature ?
20) Distinguish between Type I and Type II Superconductors ?
21) Discuss the effect of magnetic field on critical transition temperature.
What is the
energy stored in a superconductor in the presence of magnetic field
? Determine
the amount of energy in Joules that is required to destroy superconductivit
y in 1
cm? of lead at OK.
22) Arrive at London equations and hence discuss Meissner effect.
Compute the
stabilization energy density of the superconducting state at OK for
aluminum. The:
critical induction field density is 1.1 Tesla.

382
Super Conductivity
oe

23) What is Superconductivity ? Discuss the some properties, which distinguish them
from normal metals? When placed over a superconducting lead plate at what
height will a tiny magnet levitate ? (its magnetic moment is 0.25 Am? and its mass
is 4 gm).
24) Discuss the formation of Cooper pairs and the energy gap in superconductors on
the basis of BCS theory .
25) Discuss the applications of superconductors.
26) What is Josephson effect ? Discuss the applications based on Josephson effect.
27) Explain : (a) Cryotron Switch (b) Memory Cell (c) Energy gap ina superconductor.

28) What are High T, superconductors ? Write a short note on them.

383
Z
LASERS

9.1 Introduction:

In twentieth century, there were a number of outstanding scientific discoveries

and inventions. They include, Planck's law, Einsteins explanation of photo electric
theory, Bohr's model of atom, the birth of quantum mechanics, uncertainty principle,
discovery of high energy particles, positron prediction and its discovery, the inventionsof
bulb, the discoveries for the improvement of communication systems, semiconductors,
transistor, miniarization technologies, optical fibre communication etc. The inventions
that had really a major impact on scientific development were the inventions concerning
the electric bulb, transistor and laser. They had changed the way of life of a common
man. It should be mentioned here that laser is an optical maser. It was in 1954 that a
maser (microwave amplification by stimulated emission of radiation) was made. A
coherent electromagnetic radiation at frequencies of 10° to 10!! Hz was generated in
a maser. Later in 1958, Schwlow and Towner laid down the theoretical work and the
first working laser made of synthetic ruby crystal was made in 1960 at Hughes Research
laboratory. After few months, Bell Telephone laboratory developed He-Ne laser C. K.
N. Patel of Bell laboratories discovered the infrared carbon dioxide Laser. Since then,

384
Lasers

laser science has made enormous advances. We will study in the following, a few
very fundamental aspects of laser physics.

Definition of Laser :

The word 'Laser' is acronym for Light Amplification by Stimulated Emission of

Radiation. A laser source is nothing but a source of light, but a special source of light
that has unique properties. Suppose, we take an ordinary bulb that radiates energy in
all directions. It illuminates the entire room and the objects there in. The radiation
energy density at any point in the room is extremely small. Suppose, one brings all the
beams of light emitted by the bulb into a very narrow pencil of beams, say, of the size
of ball pen reefle, then, there will be a tremendous intensity of light concentrated.
This intensity would be nearly 10° times that of the light at any incident point in the
room. This is the underlying principle with which a laser source works. The laser
makes use of processes that increase or amplify signals after those signals are produced
by other means. The processes are (i) spontaneous emission of light (ii) Stimulated
emission of light (iii) Optical feed back for amplification.

A laser light is 1) coherent 2) Directional and 3) of narrow spectral width. It is

a specialised source of light and is used at places where its specific properties are
required. We have laser sources that emit light in the range 430-750 THZ. This range
spans from far infrared part of the spectrum to the soft x-ray region (A = 3nm). Laser

wavelengths are mentioned in um (microns) in the infrared region, (angstroms) A, in

nm (nanometers) in visible, ultraviolet, extreme ultraviolet, and soft x-ray region of
the spectrum. For the sake of understanding, we give here the wavelength regions.

1 micron (um) = 10,000 angstroms = 1,000 nm. A green light has a wavelength

of 5000 A , or 0.5 um or 500 nm. The infrared radiation spreads from 0.7—1 um
(near infrared), 1 to 10 um (middle infrared), and 10-1000 um (far infrared). The
following are the steps involved in the lasing action.
1. Excitation : The atoms or electrons from lower energy level must be excited or
pumped to a higher energy level.
2. Population Inversion : The life time of atoms or electrons in the higher energy
level should be long enough such that the population inversion can be achieved.
When population inversion is attained, more number of electrons will be present
in the excited level than in the lower level. For this to occur, a continuous pumping
of energy into the system is needed.
3. Light Amplification : It is achieved in a resonant cavity where laser action is
activated.

385
 Applied Physics

92 Principle of Laser Emission :

9.2.1 Emission of radiation:
Let us discuss some fundamental aspects of lasing action. Consider an atom in
the ground state designated as E,. When a radiation is incident on it, the atom absorbs
photons. If the energy of the photon is hv,,, the atoms (electrons) in the ground state
are excited to higher energy state E,, called the excited state (Ey SES,
(Fig.9.1(a))). This state of the atom is not stable. The tendency of the atom is to come
back to the ground state and this can happen by the electron falling back into the
ground state level emitting a photon (Fig.9.1 (b)). A substance consists of many
atoms and each atom has many electrons distributed in the allowed shells or orbits. At
low temperatures, the atoms (electrons) preferably occupy the lowest energy states
or the ground state. The tendency of the excited atoms (electrons) is always to fall
back into the ground states emitting radiation appropriate to the frequency v,,. The
emission of this radiation is a spontaneous process (Fig.9.1(b)). The time that the
atoms (electrons) spends in the excited state is extremely short, normally of the order
of 10-8sec. The spontaneous emission is a natural process. An atom is described
in
terms of quantized energy levels labelled eed SM Oe ch ae and so forth. The population
in any level is the number of atoms per unit volume in that level. Consider
energy
levels as shown in Fig. 9.1. E, — E, = E= hv,,. The population ratio of states
as per
Boltzmann statistics is
-E/kT
N,/N, = EVM EXPCAVyg/KT) saaeaees (9.1)
At any instant of time, there will be the some atoms (electrons) in the energy
state E, say, N, atoms; some atoms in E, state, say N,, atoms;
The population of
atoms in any energy state is proportional to Boltzmann factor exp(—E/k
T). The ground
states are densely populated when compared to the excited state.
This is the normal
condition for thermodynamic equilibrium.

E,

hv.> photon
aa
>

E,

(a) Excitation (b) Spontaneous emission (c) Stimulated emission

Fig. 9.1 Two level laser principle : (a) Excitation : Electro
n goes from E, to E, (b) Spontaneous
emission : Electron falls into E, from E, (c) Stimula
ted emission : Photons are pumped back.
Population inversion take place.
The photons emitted spontaneously propagate rando
mly with different phases.
Therefore the light emitted in spontaneous emissi
on is non-coherent. The stimulated

386
 Lasers
a

emission requires photons to be present already before the process starts. The photon
of energy hv, that is incident on the system can result in excitation and the generated
photon can give rise to a stimulated emission. Let us consider how it is done.

9.2.2 Population Inversion:

Suppose that the emitted radiation is made to interact with the system of atoms
continuously. Then, all the atoms would be present in the radiation field and there are
two processes that can occur now. First, absorption of energy resulting in the transition
of atoms from lower to higher energy levels and second, the emission of radiation by
the transition of atoms from excited state to lower level. The two processes of the
absorption of energy and subsequent emission of radiant energy are controlled by the
radiant energy density and population difference [N, — N,] in the two levels. Higher
the radiation field density, higher is the probability of the occurrence of the above
processes. The radiant energy is pumped into the system of atoms.

When the pumped energy is continuously maintained within the system of atoms,
we disturb the thermodynamic equilibrium. More and more atoms are pumped into the
excited state, and they spend more time in the excited state. Therefore, slowly the
population in the excited level will be built up and as nonequilibrium condition proceeds,
the population in excited state [N,] will be greater than the population in ground state
[N,]. This condition that N, is greater than N, is called population inversion.
Maintaining larger population in higher energy state is called population inversion. The
medium in which this process occurs is known as active medium. The transitions
from E, to E, will prevail over the absorption (transitions E, to E,), when the external
signal of energy hv,, passes through the system continuously. The emission that
takes place is the stimulated emission. This process is shown in the Fig. 9.1): The
difference (8) in the number of transitions from E, to E, and E, to E, is proportional
to the probability of the transition P, the number of quanta U (the radiation density)
and the difference in populations.
i.e., SPUN ee 0 ne ME (9.2)
The transitions from E, to E, can amplify the signal if the radiation output is
reflected back a number of times into the active system. The laser system acts like an
amplifier of electromagnetic radiation. Such an amplifier is called quantum amplifier.
In a laser, a system of atoms electrons and their energy levels is a quantum system.
The system can enter into a mode of self-excitation and operates as a quantum oscillator.
The oscillations are built up under the actions of random and spontaneously emitted
photons. The light can be taken out of the system by following proper reflector
configurations so that the light is first confined to the active medium. Then, the light

387
 Applied Physics

is made to come out as directed beam. If the active medium is a semiconductor, then
the laser system is called a semiconductor laser. Gas laser use gases (He-Ne laser),
solid-state lasers use special solids like ruby as active mediums.
Semiconductor lasers are similar to other type of lasers like gas or solid state
lasers. All the laser systems give highly coherent, monochromatic, and directed beam
of light.

93 Einstein Coefficients :

Einstein in 1917 introduced the concept of stimulated or induced emission of

of radiation by atomic systems. He described the interaction of matter and radiation in
terms of a process in which an excited atom may be induced, by the presence of
radiation to emit a photon, and there by decay to the ground state.
The stimulated emission requires population inversion. The population inversion
is possible if the photon that is emitted by spontaneous emission is made to stay within
the system to induce, once again, the transitions. This will result in stimulated emission.
A stimulated emission is shown in Fig. 9.1(c). Ifa photon can stimulate an electron to
or atom move from a lower energy state E, to a higher energy state E, by means of.
absorption, it can also stimulate an atom to decay from E, to E,. In the case of
absorption the atom goes to excited state, photon disappears. When stimulation
absorption takes place, the system of atoms should radiate additional photons to
conserve energy. This has to occur in order to keep the population of the two energy
levels in equilibrium. The population ratio of the energy states is given by the expression
(9.1). Atoms from a higher level fall into the ground state giving rise to spontaneous
emission. The probability per unit time with which this process takes place represents
the number of transitions.
The number of transitions
from upper level to lower level = A,,N, ou... (9:3)
These transitions depend on the density of the excited atoms in level 2 (E,).A5,
is a. constant.
The pumping of energy does two things. (1) it makes atoms to go from lower
level (E,, the ground state), to excited level (E,).
The number of the stimulated
upward transitions per second = N,B,,U, ........ (9.4)
where U,, is the density of radiation of frequency v. B,, is a constant. The transition
probability is proportional to the radiation energy density (energy per unit area of the
incident radiation), and the population in that level from which the transition
occurs.

388
 Lasers
ee

(2) But pumping also activates the transitions from upper level (E,) to
ground
level (E,). The rate at which these induced or stimulated transitions takes
place is
proportional to the density of radiation and the population in the upper level.

The number of stimulated

downward transitions per second = Ne BU wee... (9.5)
The proportionality constants. A,,, Bj, B,, are called 'Einstein coefficients’.
The net rate of downward transitions must be equal to that of upward transitions,
under equilibrium conditions i.e.,
alte daaeBonalsheeNic palJoreseeee aaddh teratiattored (9.6)
Solving for U,,, we obtain

U, = Vv N,By ze N,B,,
seeeeeee (9.7)
.

But N, = exp| “YT

For simplicity, we replace Vip as V.

Th en, u,= = 22! ee et. en eee (9.8)

:
ese abt Bes hv
—= |exp |—— |-1
B>, kT

According to Einstein, the probability for upward transition and the probability
for downward transition will be the same. Then
eae as) ie ute Mahan Brot ae © ae (9.9)
Then, from eq. (9.8), we get

For atoms in thermal equilibrium with the radiation present, the ratio of stimulated
rate to spontaneous emission rate is given by
Stimulated emission _ B, U
Spontaneousemission A,, ~

389
 Applied Physics

The above ratio is called 'occupation index’.

According to Max Planck's radiation theory of black body radiation we have **
A 8nhv°
ee (9.12)
Bo Cc
Where c is the velocity of light. The above equation refers actually to a black
body radiation. A black body is a pin hole cavity made in a hollow sphere. When the
sphere is heated, the radiation coming from the pin hole would have a spectral
distribution of energy.
According eq. (9.12), the rate of induced emission is extremely small in the
visible region of the spectrum with ordinary optical sources. (T ~ 10°K). Most of the
radiation is emitted through spontaneous transitions. On the other hand, if hv << kT,
the number of stimulated emission far exceeds the number of spontaneous emissions.
In the above discussion, we have considered that the atoms interact with a
photon of single frequency. In practice, atoms can interact with photons of variable
energies. Therefore, the occupation index is a function of frequency.
9.4 Light Amplification and Laser oscillations :
9.4.1 Amplification :
The active medium consists of atoms whose energies are distributed over
different levels like E,, E, E, and so forth. The population inversion is contrary to
normal thermal equilibrium state. The population densities of a two level system are
shown in Fig. 9.2 Fig. 9.2(a) gives the normal or Boltzmann distribution and figure
9.2(b) gives the inverted distribution. The rate of stimulated downward transitions
will exceed that of upward transitions if N, > N;.

5
(a) (b) 3
Fig. 9.2 Population Densities in two level system. (a) Normal or Boltzmann the distribution
(b) Inverted distribution.
In such a case, there would be an exponential increase in the intensity of light
beam. This increase of intensity of light is known as amplification. The Fig. 9.2
depicts the population distribution. The population inversion gives amplification.

** Refer : 'Physics of the atom' by Wehr, Richards, Adhir, Narosa Publishing House.

390
Lasers
i

Let us see qualitatively, in a very simple way, the magnitude of amplification

that takes place in the active medium. Suppose a narrow and parallel beam of light
propagates through the medium in which the population inversion is created. The
energy density U,, of the light is related to the intensity (I,) of the light beam. If the
frequency range of the waves in the beam is equal to Av, then
I,Av

The spectral line has a width Av i.e., the frequencies are in the range v and
v + Av. When we give energy to the medium in the process of excitation or pumping
all the atoms do not acquire the same amount of energy. Only certain number atoms in
any energy level acquire sufficient energy to go from one level to another. Suppose,
ceratin number AN, of the N, atoms in E, are available for absorption. Therefore, the
rate of upward transitions (as seen in the earlier section) is

B, U, AN, = B,,{I14ia eee (9.14)

Similarly, the rate of downward transition is

I
B,, U, AN, =B,, (| AS A tse: (9.15)
¢
During the upward transistion, a part of the incident energy hv is absorbed.
The downward transition adds the same amount of energy to the beam of light. As
more and more transitions take place, the rate of change of spectral energy density in
the frequency interval Av is given as

<(U,Av) = hv(B,, AN, -By AN, )U,

h
> ~~(AN, —N,)Bialy

But {(u,ay) =oat} =46)

d
————
d{
—— hy ee,
Av
u dt ee a
d

d hv
—(I,) = —(N, - AN, JE,)I,
dx c

hv : 2
Let —(AN, - AN, )By = a, where 'O,' is the gain constant at a frequency
C
v. Then

39]
 Applied Physics

yas es er us masd wil ot coview Ca? lo Syne ae (9:16)

we replace AN, and AN, by N, and N, respectively for maximum gain i.e.,

hv
(Oy) max = sain —N, Bi,

Then from equ. (9.12),

: ne | ;
(Oat = ———\(N,-N,)A
Sian Wa d P: sbi ete 2 el

From the above expression (9.17), we can easily understand that the gain constant

'Q,' is positive if N, > N,. That means, the population inversion gives rise to the gain

or amplification of the beam of light.

9.4.22 Laser Oscillations :

The energy due to pumping is confined to the active medium. When sufficient
population inversion is achieved, the radiation emitted is built up in intensity in an
exponential manner as it propagates in the medium. As the gain increases, it will reach
a threshold value when the losses are compensated by gain. This threshold value
depends on the life time of the photon, the frequency and the width of the spectral line

The energy losses in the medium are due to the absorption of energy by small
amounts at the mirrors, medium, or inhomogeneous portions of the medium ete.
When the gain crosses the threshold value, there is a positive feed back, and oscillations
will occur as standing waves are set up in the medium. The entire medium is like a
cavity resonator that emits energy. The selection of the amplification at a particular
frequency takes place. This frequency corresponds to the central point of the spectral
line emitted by spontaneous emission. Out of the laser source, we get a coherent
radiation. In a laser system, the active medium is placed between reflectors like a pair
of mirrors (Fig. 9.3) making the active medium asa sort of cavity resonator. Oscillations
are set in and sustained in the cavity. A laser source is a quantum oscillator. The gain
depends on the population difference.

392
Lasers

f Ctive f yeats =< Output

Mirror ie Mirror
Fig. 9.3 Basic set of mirrors reflectors. In principle, there will be infinite set of reflections
between the mirrors before light beam come out.

95 Properties ofa laser light :

9.5.1 Coherence and Monochromaticity :

A laser beam is highly monochromatic and coherent. In the following, we shall

give the meaning of coherence briefly.
The students have studied about the interference in optics in the earlier classes.
The interference pattern consist of dark and bright fringes. In the formation of fringes,
the two wave trains interfere at the point on the screen satisfying the conditions for
interference. However, normally the dark fringes spread over are not completely dark
and the bright fringe intensity does not correspond to the sum of the squares of the
amplitudes of the two wave trains. This is because of the partial coherence of the
light. The visibility of the fringes and the contrast depends upon the coherence of the
light.
The visibility of fringes is given in terms of the intensity of light at maxima and
minima. The visibility is defined as
tara ite
Veep eaia EME 9 iia i ten inlet. ee (9.18)
ete ef ish

Two waves are said to be coherent if they have the same wavelength, amplitude
and they have a constant phase relationship both with time (temporal coherence) and
in space (spatial coherence).
The visibility is correlated to coherence. If y,, represents the degree of the
coherence then,
V(I, +1,)
Yi w ie

where I, and I, are the intensities of two wave trains that interfere. If y, = 1, then we
have complete coherence ; if y,, = 0, then complete incoherence. If 0 <y,, < 1, then
partial coherence. When y,, = 0, no fringes are formed. The parameter y,, is related
to the nature of the source.

393
 Applied Physics
eees ee ae

Laser source is highly coherent because the photon states or the energy states
between which the transitions take place are correctly, or perfectly selected. That
means, the pumping is suitably selected to induce only those transitions. The very
induction of population inversion is responsible for high coherence of the beam. The
mirrors are arranged such that the amplification is built up and a direct beam comes
out. In ordinary sources, there is no selectivity of photon states.
Different Lasers emit radiation with different degrees of coherence. There are
two parameters associated with coherence. One is, the coherence time t,, the second
is, the coherence length /,. The coherence length /, is related to the monochromaticity
of the light as 7, = ¢/Ay . The coherence time Ty is the time interval between two
successive wave trains emitted by the source. Then /, = cty
The best coherence is found in gas lasers. The coherence time is of the order
of 10-3 sec. That means the wave train length is about 10°m (100 km). The ordinary
source have coherence length of less than a few centimeters. In the case of gas
lasers, the angle of divergence is about | minute.
The solid state lasers emit radiation with a lower degree of coherence. The
coherence time is 10~ sec. The length is about 100 meters. The angle of divergence
is about ten minutes.

9.5.2 Directionality :

A good monochromatic source of Sodium lamp available in the laboratories,

gives light in all directions. Therefore, there is a divergence of light. The light coming
out of a laser source is directional with low divergence. If an ordinary source like an
incandescent source has to give a directional light, the filament should be of the length
less than a micron (10~m). It is difficult to produce such filaments. In lasers that is
the angular spread of the beam. In lasers the divergence will be 6 sec to few minutes
of an arc.

9.5.3 Speckle nature:

Suppose, we focuss a laser beam from a helium-neon laser source or the

pen-torch type lasers available in the stationary shops, on to a wall in a dark room. We
see a bright red colored beam and the red spot on the wall. Let us keep a good lens and
diverge the beam into a bigger spot. Then, the magnified spot appears to 'speckle’,
i.e., dark and bright spots appear and disappear continuously. The speckle pattern of
the light is caused by the high coherence and low divergence. The dark and bright
specks appear because of the interference of coherent beams reflected to our eyes
from different points on the focussed spot.

394
Lasers

9.6 Methods for Producing Population Invention :

There are several ways of producing the population inversions necessary for
optical amplification to take place. Some of the methods are :
1. Optical pumping or photon excitation
2. Electron excitation
3. Inelastic atom-atom collisions.
The first process given in Fig 9.1, is followed in solid state laser like Ruby
Laser. We have also discussed this in 2 level
system. For the second process, we will 1 er cor weer
consider a system consisting of atoms Laser _
possessing an energy level diagram shown transition
in Fig. 9.4.
The excitation of an electron can be
caused by electric discharge (Fig. 9.4) of Electron
a gas. During such an excitation, the laser
medium itself carries the discharge in
discharge tube, the electrons in the
discharge may directly excite the active
medium atoms to produce population Eo
inversion. This method is used in argon
laser which is a gaseous ion laser. Fig. 9.4 Inelastic atom-atom collisions

In the third case also, an electrical discharge is used. A suitable mixture of

gases is selected such that two different type of atoms, (A and B) possess some
excited states A* and B* that coincide or
placed extremely close to each other E,
(shown in Fig. 9.5).,.When-excited-state 7. SN
of atom A, i.e., A*, it occupies a Laser _
metastable level A*, transfer its excitation transition
to the atom B by collision. Therefore, the
population in a level E, will increase and
may become more densely populated than
the lower level of B. Then the excited atom
B* will go to the excited level or decay by
emitting radiation that is stimulated
radiation. This process is applied in He-
tom A
Ne laser. Pe eee é

Types of Lasers : Fig. 9.5 Direct Electron Excitation

1. Gas Lasers
2. Solid State Lasers
3. Semiconductor Lasers
We will discuss about gas laser (He-Ne laser) and a semi conductor laser in the
following.

395
 Applied Physics

9.7 Gas Laser : He- Ne Laser:

9.7.1 Working of He-Ne Laser :
The gas laser have the atomic, ionic or molecular gas as active medium with
desired set of energy levels. The gases are kept at low densities. The gas is always
enclosed in a cylindrical tube arranged between two mirrors.
The first continuous wave (CW) laser that came in 1961 was He-Ne laser. The
gas that used was a mixture of helium and neon.
The Fig. 9.6 shows a typical physical arrangement of gas laser. A mixture of
the gases, Helium and Neon (in the ratio 7 : 1) is enclosed in a narrow and long quartz
discharge tube. The diameter of the tube is of the order of 1 cm and length of about 80
cm. The partial pressure of helium is
about 1 torr and that of neon is about
0.1 torr. On either side of the tube,
reflecting mirrors M,, M, are arranged.
These mirrors should be ideal ly
parallel 1. They are attached to fine
adjusters (a,, a,). The mirrors should Mirror Mirror
be flat to the extent of 4/100 flatness. rf field
The mirrors are used for creating an Fig. 9.6 He-Ne laser arrangement
optical resonator. One of the mirrors is
partially transparent.
The gas is discharged at an
appropriate rf field. Then, the atoms of
5 : ; A H N
helium are excited as the electrons in a \ ‘
the tube collide with them. BA ee ee SE sei
i 1 2S =a
Let us see the energy level L —#8— -.----- >
diagram (Fig. 9.7) of helium and neon.
Lipa ee
The gas discharge is first - —==
established in the tube by applying a |
high voltage (V~10 kV). The Helium =~
atoms are excited. The excited helium |
atoms in turn transfer energy to neon ~ =
atoms.

Helium has energy states 3S

States, 2S states and 1S states. Neon
has energy levels 2S and 35 levels.
When helium atoms are excited by
electron impact in the discharge tube,
the population of the energy states 3S
and 2S states build up. As the energy ° A Ne
Fig. 9.7 Energy level diagram of He and Ne atoms
and laser transitions
396
Lasers

states of neon are close to the metastable levels of helium, there is a high probability of
transfer of energy when the excited helium atoms and electrons collide with neon
atoms. Thus, the collisions will result in energy transfer
He 7S) +Ne -—> He
+ Ne (2S)
He (IS) +Ne — He + Ne (3S)
under suitable discharge condition, the population inversion in Ne(2S) and Ne(3S)
levels can take place. There is a higher probability of levels 3S being populated more.
The stimulated emission comes as a result of the electron transitions in neon
atom which is now in excited state. The transitions and the corresponding wavelengths
of the emitted light are given below.
38, > 2P, > 0.6328 pm
28, —> 2P, —> 1.15 pm
38, > 3P, > 3.39 um
The transitions are shown in Fig. 9.7.
The helium metastable levels in helium-neon laser discharge acquire population
densities of the order of 10!7 per cubic meter. This population density is about three
orders of magnitude higher than that of the order of the other excited energy levels of
the atoms within the discharge. In the process of collision and the energy transfer to
excited states of neon atoms, there is no radiation decay of the helium to ground state.
The population densities of excited levels of helium atom is about 10!7 cubic meter
and suppose there are 1.4 x 10? m™ neon atoms, the excitation rate is of the order of
3.6 x 10° per second. The major spectral line 6328 n is emitted. All the other spectral
lines are eliminated in the process of reflections.

9.7.2 Applications of He-Ne Lasers :

Helium-Neon lasers emit powers in the range 0.5 mW_ to 50 mW. The line

width of the spectral line (6328 A ) is of the order of a few picometers. In low power
tubes the mirrors are fixed to the tube. But in high lower tubes, they are placed away
from the tube. He - Ne laser is a common laser that is found in laboratories. The
applications include interferometry, laser printing, bar code reading, holography etc.
The He-Ne laser provides the stable, single - transverse mode beam necessary for
indentifying optical properties of materials such as surface smoothness. By placing
the mirrors at different angles, the output can taken to be either 6328 A or 1.15 pm
laser light.

The helium-neon lasers are used for surveying, and the target-aiming devices
for guns. These laser are used for larger distance measurements which is laser
modulation telemetry. The laser beam is amplitude modulated and made to fall on a
‘target’ whose position from a point has to be known. The target reflects light and the

397,
 Applied Physics

reflected light is detected. The phase shift between the reflected and the direct beam
2
is measured to calculate the distance C a eth x 2n D|. The lasers are also useful for

surveying and land profiling.

The power of the He-Ne Lasers that are used would be between 2mW to 50
mW. These lasers are highly useful in laboratory experiments in science. They have
an operating period of 50,000 hrs. The laser has a coherence length of 300 meters.
The lasers are used in holography (three dimensional photography).
The 1523 ym laser is used in optical transmission lines, which have minimum
loss in that wavelength region.

9.7.3 Other‘gas lasers :

The argon-ion laser is a noble gas laser that operates in the visible and ultra-
violet spectral regions. They can provide many wavelength lines both in visible and
Uv-regions. Argon and Krypton lasers are used in phototherapy of the eye, for pumping
dye lasers, in laser printers, cellcytometry and stereolithography. Carbondioxide laser
is one of the most powerful and efficient laser available. It operates in the middle
infrared wavelength in the range (9 — 10.6 uum). Both pusled and continuous wave
lasers are available. The carbon dioxide-nitrogen (0.8 : 1) lasers operate in CW mode
to give powers greater than 100KW and pulshed energies of 10KJ. These
lasers are
used in materials processing. This includes cutting, drilling, material removal,
etching,
milling, melting, welding, alloying, annealing, and so forth. If has excellent
medical
applications where it is used for cutting and cauterizing. With these lasers,
a very
intense heating can be produced at a small area or a point.

9.8 Semiconductor Lasers:

A semiconductor is associated with a valence band and a conduction band.
Let
this band gap energy be E, and Eg, and E,, be the quasi fermi states.
In semiconductors
like GaAs, GaP, InS etc., that have direct band gap, the electrons have
high probability
of recombining with holes thereby emitting
radiation-photons. This is shown in Fig.9.8.
The recombination radiation photons
produced may interact with valence
electrons and be absorbed. They can also
interact with electrons in the conduction
band thereby stimulating further photons of
the same energy. The injected carrier
concentration can become large enough and
the stimulated emission can exceed the
absorption. The optical gain can be achieved
in the active region. In diode lasers, it is
necessary to use the external mirrors so that Fig. 9.8 Energy band picture of laser
light emitted by the diode is reflected back transitions

398
Lasers

to provide the positive feed back. The high refractive index of the semiconductor
materials ensures that most of the light is reflected into the materials at the
semiconductor-air interface. We shall study briefly the working of the laser.

9.8.1 Working of a Semiconductor Laser :

The principle of working of a Semiconductor laser is similar to LED working.
Semiconductor lasers are capable of operating as light source for fiber optical
communications. Semiconductor lasers normally operate in 0.8 um, 1.3 um 1.55 pm
region. Table 9.1 summarizes the emission characteristics of certain semiconductors lasers.
Pure semiconductors like GaAs, ZnS, InSb also work as laser sources. But the diodes
are prefered for their stability. No laser action has been observed in silicon and
germanium.
Table 9.1: Emission characteristics of Laser materials
Material Wavelength (um)

PbS - PbSe - PbSn 5.48 - 8.95

The essential conditions for lasing action as discussed earlier are :
1. Carrier injection
2. Stimulated emission
a) Pumping
b) Population inversion
c) Presence of photon in the laser
3. Optical feedback - amplification
Semiconductor lasers work on the same principle as light emitting diode, and
the p-n junction provides the active medium. Thus, for obtaining the laser action, we
have to provide the condition for population inversion and optical feed back. For a
stimulated emission, there must be a region of the device in which there are exited
electrons and holes present together. This is achieved by forward biasing a junction
formed from very heavily doped p — and n — materials. In such a heavily doped
n — material, the fermi level lies in the valence band. The equilibrium forward bias
energy level diagram for a junction formed from such materials is depicted in the
figure 9.9. When forward biased, a large concentration of electrons and holes are
injected across the junction in opposite directions. Sufficient number of electrons are

399
 Applied Physics

pushed into conduction band at the junction on n-side. Therefore a population inversion
is created in a narrow region or active region. The band diagram of a simple PN
junction diode under unbiased and biased conditions is shown in figure 9.9.

p-type p-type n-type

Ee n-type Ss

ES WMITMITIT Dy. ~ brn qv

Ey

Cc

Erp amram

a) Unbiased laser diode b) Forward biased laser diode

Fig. 9.9.
In most semiconductor lasers, the lasing action is made to occur on the p side of
the junction by electron diffusion, as electron-diffusion process has higher efficiency
than hole-injection. Over the diffusion length, the injected carrier density is much
more
than equilibrium charge concentration. This results in population inversion in a narrow
region near the junction. In this region, electrons and holes recombine radiatively
resulting
in the emission of light spontaneously. The photons are made to stay within
the region.
They give rise to stimulated emission. If the emitted radiation is made sufficientl
y intense
and reflected into the diode junction continuously, a gain in the output
beam intensity is
achieved. The laser light comes out as a highly directed and narrow
intense beam of
light. There can be losses due to various reasons but losses are kept at
minimum level so
that lasing action takes place efficiently.

Output +ve

Active region
Roughened surface
Output

ohmic contact
(110) plane
p active region

as AT LILI
100Bae F/T n
(110) plane
ohmic contact

(b)
Fig. 9.10 (a) Semiconductor laser diode (b) a detailed
figure of
(a)

400
Lasers

The semiconductor diode material is a diode laser that acts as an optical cavity.
The active region is parallel to the junction face. The laser beam is taken from the side
of the junction. The optical cavity is formed by the face of the crystal itself, which are
the cleavage planes in GaAs. The arrangement is shown in Fig. 9.10. The (110)
(Fig.9.10) plane of the semiconductor crystal has enough refractive index to confine
the light emitted due to recombination of electrons and holes. Multiple reflections take
place at the boundary and into the active region. This is a very intelligent manipulation
of the optical property of the semiconductor
involved. Due to these multiple reflections,
the light amplification takes place and gain
attained exceeds the losses due to enhanced = (c) Stimulated
stimulated emission. The population £
inversion occurs in active region of ‘
thickness d. This thickness will be &
approximately equal to diffusion length of | a) spontaneous /_(0)Superradiance
the injected charges. The injection current ie
at which lasing occurs is called the threshold Ih
current Cin) (Fig. 9.11). The power radiated Fig. 9.11 Characteristics ofa laser diode
is a function of the forward current I. Below
the threshold current, a laser diode operates as a common LED, emitting spontaneous
radiation in all directions. Most of the energy is lost in reflections in all directions.
Above the threshold current (i.e., 1 > I,,), the ratio between the probabilities of
radiative or nonradiative recombination grows on account of the increased probability
of induced optical transitions. This results in a sharp rise in the power output.
The disadvantage of a semiconductor laser stems from the fact that
semiconductors are extremely sensitive to temperature changes. With a slight increase
in temperature, appreciable increase in forward current takes place and the energy
gap of the material is changed, causing a change in the frequency of light emitted. It
also results in a change in I,,. Due to this, amplification falls off. Therefore heat sinks
are provided for the PN junction.
Confinement of light in Homo junction laser discussed above is extremely poor.
To improve this, Hetro junction Laser are used. The efficiency of the laser is increased
in hetrojunction lasers.
9.8.2 Comparison of Laser and LED:
Laser diodes and LEDs differ in a number of ways, mainly from the point of
view of power, spectral width, speed, linearity, temperature sensitivity, reliability etc.
LEDs are operated with 50 to 300mA and 1.5 to 2V and their power output is 1 to 10m
W. Laser diodes also have similar width input and output powers but they must be
operated above threshold. Spectral width of GaA/As LED is 35 to 5Onm and for laser
diodes it is about 20 times less. Spectral bandwidth of laser diodes would be around 2
to 3nm. Lasers have faster recombination times compared to LEDs. Laser diodes are
more sensitive to temperature variations.

401
 Applied Physics

9.8.3 Applications of semiconductor lasers :

The injection lasers discussed earlier, have a disadvantage in the fact that the
parameters of the lasers are temperature dependent. As temperature increases, the
wavelength of light emitted shifts to longer wavelengths. The temperature also increases
the current in the diode, thereby decreasing the amplification. The heat has to be
removed always from the laser source to get good efficiency.
The semiconductor lasers are of great use in communication electronics and
computer engineering. The driving force in the development of semiconductor lasers
is in the field of communication. In the early 1970s, it was shown that optical fibers of
good quality could transmit certain wavelengths of light at very low loss and were
also relatively inexpensive to fabricate. The laser beam is highly effective for long-
range information transmission. In general, the diode lasers have low power. The
diode lasers have a special place. They are small in size and are useful in hand-held
instruments. They can be integrated with other instruments like in the computer laser
printer. Pulsed lasers as well as continuous wave (CW) lasers are available.
The present age belongs to miniaturization. The effective dimensions of the
semiconductor devices are coming down from three dimensional to two-dimensional
to one-dimensional level to ultimately zero dimensional level. One such device is quantum
dot laser (QD). In these laser systems, the wavelength of the light emitted is
not
determined by the band gap ofthe semi conductor material, but by just a dot of a
semi
conductor that’s created between two semiconductor films. The electrons are confined
to the dots and occupy sharp energy levels. The quantum dots behave like
artificial
atoms and the electron is said to experience a quantum confinement. Ina quantum
dot
laser, gallium arsenide dot is created as an ultra thin plane between two
different semi
conducting thin films, namely the alloy of gallium arsenide and aluminum. As a result
of the quantum confinement, the electrons move freely in the
dot, forming a
two-dimensional electron gas. The size of the device is comparable
with deBroglie
wavelength. The electron-hole pair recombination takes place in the
active region, that
is the galliumarsenide well. These lasers give a coherent radiation
that is highly
monochromatic with low energy loss and insignificant divergence
.
The continuous wave lasers are useful in optical detection
and target ranging.
They are used in communication systems that involve convey
ing information at a
distance. Laser links are useful in the same way as the microw
ave links ofthe coaxial
cables. Nowadays, the coaxial cables are being replaced by optical
fibers.
The optical fibres, free of impurities can transmit contain wavele
ngths of light
at very low loss. The optical fibres can easily be fabrica
ted at low cost as they are
basically silicon dioxide fibres of about 1.3 jum or 1.5 jum
in diameter. The fibre optic
system is slowly replacing the regular micro wave linking.
The semiconductor lasers are useful in high speed comput
er networks, avionic
systems, satellite networks and high definition television.
We can have fibre optic
communication like telephone, but also in video, audio
and computer signals. High
power sources are needed for the above purposes. The
fibre optic system will provide
the largest market for semiconductor lasers.

402
Lasers
ae
an CS

Optically coupled lasers are used for the construction of optical logic elements
that have very high speed (107!° sec). They are going to be the basic elements for the
development of super computers.
The focusing property of laser radiation is also important in low power
applications. A laser beam is used for reading the bar code of consumer items in
supermarkets. The bar code is a series of parallel bars of varying widths giving
information about the consumer product. A semiconductor laser scans these bar codes.
The light reflected from this bar code is detected, and, the product and pricing are
thereby identified.
Semiconductor laser diodes are also used in as 'needle' or 'probe' preparation
and subsequent read out of disks in computer CDROMs (compact disc read only
memory). The information to be stored on the disk is imprinted in the disk in digital
form by forming small pits (using higher powered laser). A lower power laser that
provides a signal for playback of the data can read it. The digitally modulated radiation
that is scattered from the disk is collected by an optical sensor and processed through
the audio amplifier system. Similar system is used in CD (audio) and VCD (video CD)
players.

9.9 Solid State Laser- Ruby Laser:

A Ruby laser system is an example of a three level laser system. Ruby laser is a
proto type of solid state lasers. It was first fabricated by Maimann in 1960. The active
atoms of the laser medium are embedded in a solid. Both glasses and crystals are used
as supporting matrices for solid state lasers. The ruby laser consists of a cylindrical
rod as shown in Fig. 9.12 The crystal is aluminium oxide (Al, O;). Ruby is doped with
chromium (Cr***) ions (0.05 to 0.5%). The aluminium atoms in the crystal are replaced
by chromium ions. The rod is normally 3 to 20 mm in diameter and 30 mm in length.
The ends are optically ground and polished to a high degree of parallelism and flatness.
The mirrors can be coated at the ends of the rod or can be kept externally to make the
rod a resonant cavity.

Light Beam

Coated End machined end faces

Flash Lamp
Laser Rod
(a) Machined
end faces

elliptical pump reflector

pulsed pumping lamp (flash lamp)

(b)
Fig. 9.12 (a) Schematic of aRuby Laser.
(b) asolid state laser with optical pumping lamp, placed inside a reflector parallel to each other.

403
 Applied Physics

The excitation in the laser is caused by optical pumping from an intense source
of light like the xenon lamp or high pressure mercury discharge lamps. A helical flash
lamp can be used as shown in the Fig 9.12 The laser operates at room temperature and
produces high power.
The energy level diagram is given in Fig. 9.13. It should be remembered that the
energy level diagram of solid
consists of energy bands. The
chromium ions absorb the
very fast
radiation emitted by the flash radiationsless
lamp and get excited to the transitions
levels E, and E, from E,. The
energy level E, refers to the
ground state.
The states in the energy
levels E, and E, have a very
small life time (10~? Sec).
There exists a metastable level Energy

M . First, the ions decay to the laser emission

metastable level M by radiation a ean aeak
less transition. The metalstable
level M has a much longer life
time (milliseconds).
Therefore, a population
in version is achieved between E
the metastable levels and the Fig. 9.13 Energy levels of chromium ions in the Ruby crystal.
ground state, i.e., from the
population point of view, we have Ny, > N,. They are some photons which are
spontaneously emitted. They trigger the stimulated emission. As the flash lap is operated
in flashes, the laser gives out pulses of light as a result of the transitions from metastable
level to the ground state.
When the flash lamp operates, the population in the upper level decays very fast
and the lasing action stops till the arrival of the next flash from the lamp. The pulses,
thus emitted, possess very high power. The threshold power per unit volume required
to maintain the population inversion in the three level laser is

P=NoS}¥p _ 1.6x10!9 x6.7x

ee
10 x 6.25 x 10!4 Pe
~ 1100 W/em},3
The average pump frequency used is 6.25 x 10'4 HZ. However, the efficiency
of the pump source is normally 25% and there are absorption losses in the ruby rod as
the light passes through the rod. Even if this loss is about 25%, The power output is
about 18 KW/cm3,

404
 Lasers
a

The Ruby laser and other solid state lasers are useful
in second harmonic
generation. Ruby laser (red light) beam when passed through
a crystal of ammonium
di hydrogen phosphate gets partially converted into the second
harmonic and we get
light in the ultraviolet region from the crystal. The red light
is filtered by means of
optical filters.
Lasers are useful in material processing in engineering applica
tions. The laser
beam that comes out of a laser source is usually a few millimeters
in diameter. If it is
focussed by a lens to a point, the power can be raised to the level
of a few mega watts
per centimeter square. Such focussed beams can be used for materia
l welding. A 0.18
mm thick stainless steel sheet can be welded by a Ruby laser pulsed
beam of 5 Joules
with a pulse length of 5 n sec. This type of precise welding is extreme
ly important in
micro electronics in which thin films are used. Thermo couple
wires can easily be
welded with the help of high power laser beams.
Another important application is hole drilling. A laser pulse with
a width of
10 Sec and an energy of 0.05 Joule can pass through 1 mm thick
strain less steel
sheet leaving behind a hole of radius 0.1 mm. One can drill holes of
the diameter of
10m through very hard substances like diamond. In fact, YAG laser
is very useful in
such applications.
Ruby laser and solid state lasers are useful in welding of a detached retina.
The
precise location of the weld can be made, The time involved in welding
is very short
and the eye does not need any clamping. The pules used have pulse lengths
of the order
of 300 Lt sec at an energy level of one joule.
Ruby laser beam with a power output of 100 mega watts with a total energy
of
200 Joules are useful in the treatment of skin cancer and tumors.
Questions :
1) Expand 'Laser' and Maser'.
2) Mention some important Laser systems
3) What are the steps involved in lasing action.
4) Discuss in detail the absorption, and emission of spontaneous, stimulated
radiation by taking a two level energy level diagram.
Ss) What is population inversion? Explain.
6) Obtain the expression for occupation index in terms of Einsten's
coefficients.
(4) Explain the meaning of light amplification. Is population inversion
necessary for gain in a laser. If so, why?
8) How is light amplification achieved in a laser system.
9) What is coherence? What is the criteria for estimating the coherence of
a light source?
10) Why laser is high coherent source ?

405
 Applied Physics
ee

11) What do you mean by ‘directionality’? Compare gas lasers and a solid
state laser.
Mention the three methods of causing population inversion.
Explain the method of "inelastic atom-atom collision" for achieving
population inversion.
Give the physical arrangement of a gas laser. Why are the mirrors used?
Give the energy level diagram of helium-neon atoms. Explain the emission
of laser light.
Give the applications of He-Ne laser.
Explain why indirect band semiconductor is not suitable for lasing action.
Explain the working of a semiconductor laser.
What are the difference between a laser diode and a LED?
What is the active region in a semi conductor diode laser?
Write a note on the applications of semiconductor lasers.
Explain how laser beam is useful in reading CDs.
What is meant Laser 'speckle'?
Explain the principle of a semiconductor diode laser.
What is 'gain'? How it is obtained in a laser ?
What is meant by 'Laser Oscillations'?
Mention the basic requirements for a 'Laser'.
Discuss about the working of a ruby laser.
Explain the term 'Visibility' of interference fringes.
30) Distinguish between 'Spontaneous Emission’ and ‘Stimulated Emission '.
30) Give the diagram of a He - Ne Laser System. Explain the emission of
radiation from He - Ne medium.
Discuss the working of Ruby Laser. Mention its applications.

406
10
FIBER OPTICS

10.1 Introduction:

Optical fibers are the wave-guides through which electromagnetic waves of optical
frequencies can be guided through them to travel long distances. Optical fibers allow
propagation of light waves in a way, which is very much similar to metallic wave-guides
that allow propagation of microwaves though them. Optical fibers work on the principle
of total internal reflection of light. Light will propagate for long distances in optical fibers
because the absorption coefficient of fibers is exceedingly small. Optical fibers are
made with low refractive index materials like glass. Hence optical fibers work as a
transmission medium for optical signals. Light is an electromagnetic radiation and its
propagation is governed by Maxwell equations. Complete understanding of propagation
of light waves through optical fibers needs good understanding of solutions of these
equations in the context of optical fibers. Maxwell equations, Fresnel laws and Snell’s
law can completely explain the propagation of light through optical fiber wave-guides.
Since most of these are beyond the scope of the present text, a qualitative treatment of
the propagation of light in optical fibers is presented in the next few sections.

407
 Applied Physics

10.1.1 Structure of the Optical Fibers :

Optical fiber has a transparent core with refractive index n, surrounded by a
transparent cladding of slightly lower refractive index n, (Fig. 10.1). The cladding
supports the wave-guide structure and reduces radiation losses into the surroundings.
Light energy travels in both the core and cladding, but it does not penetrate into the
surroundings. The present day optical fibers have losses of about 4.2 decibel/Km. By
refining and purifying glass used for fiber drawing, these losses can be reduced to 1
dB/Km. Optical fibers work as light wave guides. Optical wave-guides can be understood
with the help of laws of electromagnetism and ray optics.
jacket cladding
G 2a
core
Transverse
crossection
Fig. 10.1 Optical fiber
Optical fibers are used in optical communication systems. These systems are
used in digital telecommunications. The light pulses, which are modulated according to
the signal to be transmitted, are guided along an optical fiber. The main advantage of
optical fiber communications is its speed. The optical pulse repetition frequencies will
be in the range of | MBPS (megabytes per second) to 1000 MBPS. Small size, good
overall compatibility, relative simplicity is the essential features of fiber optic
communication systems. The optical fibers are made out of glass, which is essentially
silicon dioxide and is abundantly available and hence is much cheaper than a metal
required for metallic wave- guides (copper or silver). The core and clad are placed in
a jacket for safety.
10.2 Principle of optical fibers :
10.2.1 Total internal reflection :
The propagation of light through a medium is determined by the refractive index
of the medium. Refractive index is the
ratio of velocity of light in air to the
velocity of light in the medium. A light
medium 2
ray travels slowly in denser medium than
in rarer medium. When a ray of light is
incident on the interface of the two
media of different refractive indices, it
bends and refraction occurs (Fig. 10.2).
Let the refractive index n, of the
medium | be greater than the enclive
medium 1
index n, of the medium 2. Let 9, and 8,
be angle of incidence and ee a
Fig. 10.2 Refraction oflight

408
Fiber Optics

refraction respectively. Then by Snell’s law we have

nm sing,
oe Vali Ee) REE ee aan (10.1)

Since n, is greater than n, we always have @, greater than @, in view of the

above equation.
When 6, = 90°, the incident ray emerges parallel to the boundary of the two
media. The angle of incidence 0, = 9, is called the critical angle. At the critical angle,
we have from Snell’s law
n, sin®, =n, sin@,
ie; n, sin@, =n, sin90

: n 2
sin 0, aa oe ee (10.2)
At the angles of incidence 0, > 0,, the light will be reflected back into the
medium | and suffers total internal reflection. Hence, total internal reflection occurs at
the interface of the two media of differing refractive index and angle of incidence is
greater than critical angle. Under these conditions, light can be constrained to move
along the optical fiber. Figure 10.3 illustrates the propagation of light along the fiber via
a series of total internal reflections at the interface of core and cladding of optical fiber.
Core has slightly higher refractive index than cladding material. The wave is transmitted
as guided wave.

era core
axis

__->» light ray

clad

Fig. 10.3 Propagation oflight in optical fiber

10.2.2 Acceptance Angle :

Not all rays falling on the optical fiber will be transmitted through the fiber. Only
those rays with sufficiently shallow grazing angle 0, > 0, at the core-clad interface are
transmitted through the fiber. Also, not all rays entering the fiber will continue to travel
along the fiber. This is shown in the Fig. 10.4. Consider a ray A incident at an angle 0,
to the fiber axis. The ray is refracted at the fiber and refracted ray is incident at an
angle 8, on the core-clad interface. Therefore, it suffers total internal reflection. Whereas
ray B falls on the optical fiber at an angle greater than 0,. This ray does not undergo

409
 Applied Physics

total internal reflection at the core-clad interface and enters the clad material. This ray
will eventually get lost in the clad material itself. Thus, for the light rays to propagate
through optical fiber by total internal reflection, they must be incident on the fiber core
with in acceptance angle defined by conical half angle 8,. This is the maximum angle to
the axis of the optical fiber at which light ray may enter the fiber so that it can be
propagated. Hence, this angle is called acceptance angle.

lost ray
acceptance
cone

Fig. 10.4 Acceptance angle

Acceptance angle is the angle at which light should be incident on the fiber for
the propagation through it.

10.2.3 Numerical Aperture:

Let a light ray be incident on the fiber at an angle 0, to the fiber axis. (8, < 9,)
Fig. 10.5. The ray enters from a medium (air) of the refractive index n, and fiber core

Fig. 10.5 Refractions and reflections in optical fiber.

refractive index be n, which is slightly greater than clad refractive index n,. By Snell’s
law :
Ny SiN, =n, sind,
n, sin, =n, sin(90 + ¢)
1
=n, cosg = ny (~sin? op

es s s
= (n?= n sin2 oP = (n?= n- b Sa stavavavere (10.3)

= Numerical Aperture.
In the limiting case, 8, = 9,, n, sin(9,) is called numerical aperture (NA).

410
Fiber Optics

Hence, we have
NAGS nr sin@a yin 2,7) so ty es (10.4)
In the case of optical fibers used in short communications, the numerical aperture
will be in the range 0.4 to 0.5. In the case of long distance communications, the
numerical aperture will be in the range 0.1 to 0.3.
The following is the summary of the above discussion :
(a) The optical fiber consists of core, clad (made out of glass) and a safety jacket
made out of polymer.
(b) The refractive index of the core glass (n,) should be greater than the refractive
index of clad glass (n,), i.e., Nn; > Np.
(c) The light is propagated through the fiber as a result of multiple total internal
reflections.
(d) Any light wave which is incident at an angle 6 greater than the critical angle undergoes
total internal reflection.
(e) A light that is incident at an angle less than the critical angle, will be lost in the clad.
(f) The angle at which the light ray is incident on the fiber such that it is transmitted
through the fiber is called the acceptance angle. The cone defined by the twice the
acceptance angle is called the 'Acceptance Cone’.
(g) The square root of the difference of the squares of refractive indices of the core
and of refractive indices of the core and clad [ (n? — ie ] power is called the
numerical aperture. It is called the figure of merit of the fiber.

103 Modes in called optical fibers:

An optical wave consists of the electric and magnetic field vectors perpendicular
to each other and perpendicular to the direction of propagation. This type of wave is
known as Transverse Electromagnetic Wave. In the case of bounded medium like an
optical fiber, the optical wave is confined to the fiber that acts like a guide. Asa result,
the electric fields or the magnetic fields are distributed in the guide such that there will
be a component of either the electric field or the magnetic field along the direction of
propagation of energy. The stable field distribution in the y and z- direction with the
periodic dependence in the x-direction is known as 'Mode'. If there is no electric field
component along the direction of propagation of energy (i.e., E, = 0), then the mode is
called TE mode. If there is component of the magnetic field (H-component) along y
and z and H, = 0, then the mode is called TM mode. Both these modes are possible in
an optical guide.

The light ray being guided through an optical fiber is a superposition of bound
waves of different modes. Each of these guided modes consists of variety of
of
electromagnetic field configurations. These field configurations differ in their sense
polarization, phase, amplitude etc. Fora monochrom atic light wave of angular frequency

411
 Applied Physics

® = 2 mn, n being frequency representing a mode traveling in positive x direction is

given by
Y= Yo expr kx) ee (10.5)
where y is the amplitude of the wave at (x, t). y, is the maximum amplitude b is the
x-component of the propagation vector, k = 27/A .This is the solution of wave equation.
For guided waves, k can have only certain discrete values. These values can be
determined by applying boundary conditions at the core-clad interface while solving
the differential equation of wave propagation through optical fiber (Maxwell’s
electromagnetic wave equations). There are three types of modes. They are:
1. Transverse electric (TE) modes (E, = 0)
2. Transverse magnetic (T.M) modes (H, = 0)
3. Hybrid modes
Hybrid modes are combination of transverse AES
otPe
electric and transverse magnetic modes. Different him aie
modes are shown in the Fig. 10.6. seal Calis
———>
Several low order modes are shown in the
Fig. 10.7. The order of the mode is number of zeros (a) TE (b) TM
crossing of the field. The plots show that the electric
and magnetic fields are not completely confined to
the center of the fiber. They vary along the diameter
of the fiber shown. Lower order modes are
concentrated in the center resulting in meridional rays
and in higher order modes the fields are normally
distributed near edges resulting in skew rays. (¢) TE01 (d) TM 01
Increasing the core refractive index, increases the
number of propagating modes. Increasing clad Fig. 10.6 Different modes ofelectric
and magnetic field oscillations in EM
refractive index,decreases the number of propagating
modes. waves.

2 2
Number of propagating modes is proportional to 2na“\N
ye
Xn
a = fiber half width
i = wave length of the light
TE) TE, TE,

clad n,

core ny, —

clad ny

Fig. 10.7 Lower order modes in light through optical fibers.

412
Fiber Optics

10.4 Step Index Fiber:

Fibers in which core of constant refractive index n, is surrounded by cladding
of slightly lower refractive index, with a steep refractive index profile at the interface
of core-clad is known as step index fiber (Fig. 10.8(b)). The variation of the refractive
index of a step index fiber as a function of distance can be mathematically represented
as:
n(r)= n,(r < a inside core) gd oes cn (10.6)
= ny (r > ain clad ) noone (10.7)

Core diameter will be of the order of 2 - 10 mm. Step index fibers can be made
to be multimode fibers or single mode fibers. In multimode fibers, the core diameter
will be large. In single mode fibers, the core has smaller diameter. Step index fibers
support the transmission of transverse electromagnetic radiation. Single mode fibers
transmit single ray along the axis of the fiber (Fig. 10.8(b)). Single mode fibers have

step index fiber

(a = 50pm, b = 100um)
n= 1.00
n, = 1.416 ee light ray
A x 0.002
eo)
Bef, (b) Longitudinal cross section.
Ss.
0 r
(a) Transverse cross section
Fig. 10.8 Single mode step index fibre.

the following advantages:

1. Low inter modal dispersion and lower broadening of light pulses being transmitted.
2. Larger bandwidth can be attained in these. Bandwidth is the difference between the
minimum and maximum frequencies of light that can be sent through the fiber.
Multimode step index fibers allow finite number of guided modes (Fig. 10.9).
Light is an electromagnetic radiation. It

SNe:
SS
has perpendicular electric and magnetic Sie >
field vectors perpendicular to the direction
of propagation. Light can be polarized.
The direction of polarization, alignment
\ clad
of electric and magnetic fields will be
Fig. 10.9 Multimode step index fiber
different in rays of different modes in
multi mode fiber. These modes depend on the boundary conditions used in the solution
of differential equation for the propagation of light wave in optical fiber. Mode volume
of a multimode fiber is the number of modes the fiber can support.

413
 Applied Physics

Normalized frequency of a multimode fiber is the cut-off frequency below which

a particular mode cannot exist. This is related to numerical aperture, radius of the core
and wavelength ofthe light. It is given by:
27
normalised frequenscy = — a(NA) se ee (10.8)
here a= radius of the core
NA = Numerical aperture
dX = Wavelength of the light.
Multi mode fibers have the following advantages:
1. Multimode fibers can use spatially incoherent sources oflight like Light
emitting diodes. Single mode fibers require coherent light sources like laser.
2. These types of fibers have large Numerical! aperture and core diameters.
Therefore it is easier to couple them with other fibers and sources of light.
3. These fibers have lower tolerance requirements on their properties.
10.5 Graded Index fibers :
In graded index fibers, the refractive index of the core varies continuously. It
starts increasing from one end a= 50 pm
of the core diameter and pe ET te
attains a maximum at the ae eae
center (Fig. 10.10). Again
n, = 0,(1 — A)
refractive index value a, = t.1.470
decreases as one moves away
from the center towards other
end of the core diameter. The
refractive index variation of the
graded index fiber can be
represented as:

1
n(r)=nj(l-24(r/a)* P OF Wiesaiy Ja. etna uaa calle iin Hee (10.9)

’ Alepry a 578 or 2 ee ee ae ere (10.10)

D is the fractional difference in the refractive index of the core and
cladding = (n,? - n,”)/2n,?. @ is the profile parameter. If a is infinity, the fiber will be
step index fiber. If a = 1 we get a triangular profile of the refractive index
variation in
the core. Graded index fibers have following characteristics:
|. Refractive index profile is circularly symmetric.
2. Fiber is multimodal with large core diameter..
3. Total internal reflection and refractive index change with in core region
slowly.
4. Refractive index variations are small over distances compara
ble with
wavelength oflight.

414
Fiber Optics

Graded index fiber is a multimode fiber that supports propagation of meridional

rays. Meridional rays pass through the center of the fiber. The gradual decrease in the
refractive index from center of the core towards clad results in many refractions
(Fig. 10.11) and ray gradually gets bent (curved) until total internal reflection takes
place. Multimode graded index fiber exhibits far less inter modal dispersion compared
to the step indexed multimode fiber. The light rays which passes through the edge of
core and does not pass through the fiber are known as skew rays. Skew rays follow a
helical path as shown in the Fig. 10.12.
clad
core
— fiber axis

m7 Ry> 0,7 A,

(a) Light Prpopagation in Graded index fiber. (b) Multiple refractions in graded index fiber
with in core material.
Fig. 10.11
caw
oN

Nee
(a) corss-section view (b) helical path ofa ray

Fig. 10.12 Skew rays.

Light is an electromagnetic radiation, which is transverse in nature. Light wave

is made of electric and magnetic field oscillations that are perpendicular to each other
and also perpendicular to the direction of propagation of the wave. This is the transverse
electromagnetic nature of the light. Meridional rays correspond to transverse electric
and transverse magnetic modes. Skew rays represent yet another mode of the light
waves.
Here is a comparison of different types of fibers:

Single mode Multimode Multimode

Step index fiber step index fiber graded index fibers

Source Laser Laser Laser or

light emitting diode

Cost cheap costly cheapest

Coupling very difficult difficult easy

Application | under water cables Data links Telephone lines

Losses least intermediate maximum

415
 Applied Physics

10.6 Mode coupling:

Optical fiber perturbations such as deviation of fiber axis from straight line, change
in core diameter, irregularities in core and clad interface, refractive index variations may
change the propagation characteristics of the fiber. Individual modes do not normally
propagate through out the length of the fiber without large energy transfers to the adjacent
modes, even when the fiber is exceptionally good without any bends. This mode conversion
or mode mixing is called mode coupling. Mode coupling affects the transmission of light
through fibers in several ways. Mode coupling is illustrated in Fig. 10.13.
defect

(a) mode mixing (b) mode conversion

Fig. 10.13 Mode coupling.

Comparison of Step Index and Graded Index Fibers :

1. Step index fiber has a bandwidth of about 50 MHz. Bandwidth is the difference
between maximum and minimum frequencies of em-wave that can propagate
through fiber. Graded index fibers can have large (theoretically infinite) bandwidth.
2. Graded Index fibers will have lower attenuation than Step Index fibers.
3. Step Index fibers can be single mode fibers where as Graded Index fibers are
always multi-mode fibers.
4. The attenuation in Graded Index fibers is of the order of SdB/Km and the numerical
aperture is 0.16 - 0.2 where as the Step Index fibers of the same physical size
show a loss of order of 12dB/Km and numerical aperture of 0.2 - 0.35.

10.7 Preparation of fibers:

The dielectric material (normally glass) used for the fiber drawing should have
following properties: ‘
1. The dielectric material normally glass used for core must have a higher refractive
index than that of cladding material to ensure guiding the light wave through the
fiber.
2. The dielectric material used must have low loss ( < 10 dB) in the wavelength
region of transmission.
3. Materials and fabrication methods used must be capable of producing graded
index fibers.
4. The dielectric used should be malleable. It should have proper elastic properties
to enable fiber drawing. The core and clad regions should be compatible.

416
Fiber Optics

The glasses used for core, cladding and protective coating, materials are
synthesized and purified. Coating is used
over cladding material to protect the fibers core glass
from atmospheric degradation due to
clad glass
chemical, mechanical attacks. It also
protects the fibers from humidity. furnace
Polymeric materials are used for protective
coating. The melts of these materials of
core and cladding are taken in concentric
conical crucibles hence this method is 0
oo0o0o0o
000 00
0
oo0oo0o0o

called double crucible method. The

arrangement used for fiber drawing is
shown in the Fig. 10.14. The fiber is
drawn form the melt. Refractive index
grading is achieved through the diffusion polymer
of mobile ions across core -clad interface
with in molten glass. The diameter of the QO

fibers is monitored and controlled. After ©} cooling oven

°
drawing the fiber protective coating in
given to the fiber (Fig. 10.14). Liquid cable winding
drum
phase methods have a disadvantage of
obtaining and maintaining extremely pure Fig. 10.14 Optical fiber drawing.
glass which limits their ability to produce
low loss fibers.
Materials used for fiber drawing :
Given below are some of the glass systems used for the optical fiber drawing:
Step index fibers:
Core glass Cladding glass
Na,-Bi,0,-Si0, Na,-Bi,O,-SiO,
Na,-Li,O-CaO-SiO, Na,-Li,O-CaO-SiO,
Na,-CaO-GeO, Na,-CaO-SiO,
P,0;-Ge,03;-GeO, P,0;-Ga,O,-SiO,
Graded index fibers
R,O-GeO, - CaO-SiO, Na,-Bi,0,-SiO,
R,O-Bi,0,-SiO, Na,-Bi,O,-SiO,
Na,-Bi,0,-SiO, Na,-Bi,O,-SiO,
The following are the properties of Multimode Step index fiber:
Core diameter 50 - 100 mm
Cladding diameter 125 - 500 mm
Buffer jacket 250 - 1000 mm
Numerical aperture 0.16 - 0.5
Attenuation 2.6 - 5 dB/Km.
Band width 6 - 50 MHz.

417
 Applied Physics

The following are the properties of Multi mode graded index fiber:
Core diameter 30 - 100 mm
Cladding diameter 100 - 150 mm
Buffer jacket 250 - 1000 mm
Numerical aperture 0.2 -3
Attenuation 2 - 100 dB/Km.
Band width 300 - 3000 MHz.
Properties of Single mode step index fibers are summarized below:
Core diameter 5 - 10 mm
Cladding diameter 125 mm
Buffer jacket 250 - 1000 mm
Numerical aperture 0.08 - 0.15
Attenuation 2-5 dB/Km.
Band width > 500 MHz.

10.8 Fiber Packing:

The glass fibers drawn are packed properly to protect them against the fiber
damage and breakage during installation and maintenance. Bare glass fibers are brittle
and susceptible to damage very easily. It is necessary therefore to cover them properly
into a series of protective layers. There is coating and cabling used. Coating
is to
protect the fibers from chemical and mechanical strains. This enables the cables
to
withstand the mechanical stress and strain and environmental conditions. Around
the
fiber a buffer tube is formed. A rigid member and a strength member are
fabricated
over buffer tube. The strength member is a steel wire, nylon yarn, polymer
film,
terlene yarn, and Kevlar or fiberglass yarn. The buffer tube is normally plastic
and is
made of polyvinyl chloride (PVC), polyethylene, poly urethelene, etc. Strength
members
are added to the cable to minimize the cable elongation and contraction
characteristics.
Optical fibers can be packed in the layered design or bundled design.
These are shown
in the Fig. 10.15.

outer sheath

core

buffer jacket
fiber budle

(a) cross-section (b) layered packing (c) bundled design

Fig. 10.15 Packed optical fiber

418
Fiber Optics

10.9 Losses in Optical Fibers :

A part of the light energy being propagated through optical fiber is lost during its
traversal form the source to destination. These losses are due to various reasons. The
losses are measured through attenuation constant. Its unit is decibels per kilometer.
Attenuation A is defined as

ee ee ae Ste OY os Sy ls (10.11)
Po
Attenuation constant a is attenuation per unit length. Hence

10
le ee ee (10.12)
b Seep)
where | is the length of the fiber.
The following are the loss mechanisms due to which the intensity of light decreases
while propagating through fiber :
1. Absorption losses : Absorption losses could be intrinsic or extrinsic. Depending the
material used intrinsic absorption losses will change. These depend on the absorption
characteristics of the material used for fiber drawing. They depend on the absorption
of the material in infrared and ultra violet region. Extrinsic absorption losses depend
on the purity and perfection of the material
2. Scattering losses : The glass material used for the optical fibers will have many
microscopic inhomogeneities. These are because of the defects and changes in the
structure and result in light scattering.
a) Rayleigh scattering : Is because of small fluctuation in refractive index and
changes in the density. These inhomogeneities are fundamental and intrinsic to
the glass used for fiber preparation and cannot be completely avoided.
b — Mie scattering : This is due to the irregularities in the core-clad interface and
refractive indices, diameter fluctuations, strains, bubbles etc. These defects
normally depend on the material, design and method and conditions of
manufacture of the optical fibers. These defects can be reduced considerably
by proper control of different processes during manufacture of the fibers.
c — Brillouin Scattering : Modulation of light through thermal molecular vibrations
within the fiber material. The incident photon (light packet) produces a phonon
(quantized elastic wave having sound frequency) and a scattered photon. This
type of scattering occurs only above certain minimum threshold frequency.
Hence it can be completely avoided by selecting proper light frequency below
the threshold frequency.
d) Raman scattering : Raman scattering incident photon produces another photon
(optical phonon). This can also be avoided as the Brillouin scattering.
e) Dispersion losses : Losses due to dispersion of the transmitted signal causes
distortion of the signal. Dispersion losses could cause broadening of the pulses
being transmitted resulting in overlap of the adjacent pulses and subsequent
distortion. This limits the rate of bit transfer (bits transmitted per second).

419
 Applied Physics

f) Inter modal dispersion or mode coupling : This is chromatic dispersion and

results in the overlap of signal of different wavelengths. Hence, this will limit
the line width of the light source used.
Attenuation coefficient can be expressed as a function of some of the above
loss mechanism as follows:
Input
a@=AA* +B+C(A) Signal

A is Rayleigh scattering coefficient

B= Wave length independent losses like dispersions
Modulator
C = Wave length dependent loss mechanisms like inter
modal dispersion

10.10 Applications of Optical Fibers :

> Optical fibers can be used in a variety of applications. These Terminal
can be broadly divided in three categories: equipment

i) Communication applications : Light with a large carrier

frequency capability has a potential possibility of packing
extremely high information content into a single channel.
Semiconductor laser and light emitting diodes along with optical
fibers can make this possibility a reality. The main components
of fiber optic communication system are:
1. The optical source : semiconductor lasers, or LED. The
power launched into fiber, rise and fall time are important.
2. A means of modulating the signal with out put of the source
: Intensity modulation is done. Pulse modulation methods
for digital or analog transmission are used. Encoding
schemes are employed. do
3. Transmission medium : optical fibers are used. Multi- is

mode or Single-mode, size, refractive index profile, losses, paaj

4amog
ssaquy
jeo
dispersion, mode coupling, cabling, joints employed are
important. Terminal
4. The detector that converts the received light back into equipment
electrical wave : p-n, p-i-n, or avalanche diodes are used.
5. Amplifiers and signal processor to amplify and process
the signals received : The amplifier design is very important
here.
Detector
As indicated in the third point above, optical fibers are the
transmission media for communication of light signals. Figure
10.16 gives a block diagram of communication using optical fibers.
In the following communications, the dominant criteria Output
for design in the application, using either digital or analog Signal
transmissions, are the required transmission distance and the rate
Fig. 10.16

420
Fiber Optics

of the information transfer. It is necessary to insert repeaters in the long distance

communication. The purpose of the repeater is to convert the optical signal back into
the electrical regime. In the analog transmission, it is transmitted and equalized, then it
is re-transmitted.
Advantages of fiber optic communications:
Communication using an optical carrier wave guided through a glass fiber has a
number of attractive advantages. Some of them are:
a) Enormous Potential Bandwidth : The bandwidth of the optical communication
channel is very large compared to other communication channels. It spreads
form 10!3 Hz to 10! Hz. Because of larger bandwidth, fiber optic systems
have greater information carrying capacity. Bandwidth gives the information
carrying capacity of the communication channel. Enhanced bandwidth utilization
can be obtained in fiber optic systems by transmitting different signals with
different colors (wavelengths) in parallel on the same optical fiber. This is
called wavelength division multiplexing. This kind of multiplexing is not possible
on other systems of communication (e.g. microwaves), where only one signal
can be transmitted at a time.
A fiber has a capacity of 500 channels and its external diameter is not
more than 0.5 mm. This is very small compared to 10 mm for a corresponding
coaxial cable. They offer low attenuation less than SdB/Km of the fiber. The
repeaters can be kept at a very long distance like 45 Kms. The coaxial cables
require repeater every 1.6 Kms. This saves considerable costs.
b) Small size and weight : Fiber optic cables have smaller diameters (about 1-
1.5 mm) Hence they are compact and weigh little in comparison with copper
cables.
Electrical Isolation : Unlike their metallic counterparts (wave guides and co-
axial wave guides) optical fibers are electrically insulated. There is no need of
any other insulation for electrical isolation of fiber optic cables. This makes
optical fibers suitable for use in electrically hazardous environments.
d) Immune to interference and cross talks : Optical fibers are made out of
dielectric materials. Hence, they are free form electrical and electro-magnetic
interference (EMI). Electrical noise present does not affect the signals that are
being carried in the optical fibers. Hence, there will not any cross talk in optical
fibers unlike their metallic counter parts.
Signal security : Unlike copper cables, a transmitted signal in an optical fiber
cannot be obtained without drawing power from the cable. Any such attempt
to draw power from the cable can be detected. This feature is attractive for
military, banking and general secured data transmission applications.
Low transmission losses : Optical fibers have very low transmission losses.
Flexibility : Metallic wave-guides are not flexible. Optical fibers are small in
size and are highly flexible.

421
 Applied Physics

h) Low costs : Optical fibers are made out of glass. Hence, they are cheaper
compared to metallic wave-guides and co-axial cables.
ii) Sensor applications : Many industrial process use optical fiber sensors to monitor
displacement, pressure, temperature, flow rate, liquid level, chemical composition
etc. These parameters are measured by observing phase, state of polarization,
mode coupling and relationship between these properties are used in sensor
applications. Phase modulation and polarization modulation. Single mode fibers are
used as sensors by observing phase and polarization modulation, whereas multimode
fibers use mode coupling and amplitude modulation. Interference is another important
property, which is used in sensing applications.
a) Displacement sensors : Any movable
displacement of a body in the shutter

path of the light between two

fibers can be detected easily.
This application is illustrated in yee
rigeiO.17:
Fig. 10.17 Displacement sensor
b) Temperature and pressure
sensors : Change in the fringe width of interference fringes formed by the
superposition of two single mode light waves traveling in two fibers. One of
the waves contains information about the temperature as it is made to pass
through the region whose temperature/pressure is to be measured. The other
ray is called reference ray. The change the in path length of this wave after
taking into account the extra fiber length and changes in the extraneous refractive
index changes. The interference of these two waves gives fringes. The fringe
width can be used to calculate temperature/pressure. The experimental
arrangement used is shown in Fig. 10.18.

beam splitter Jens

sensing fiber

detector detector

Fig. 10.18 Temperature and pressure sensor

422
Fiber Optics

c) Chemical sensors : Fiber optic chemical sensors are used. Interference is the
technique used in most of these sensing applications. The sensing element is a
modified fiber, which can convert concentration of chemical element being
sensed into phase changes of the light wave. For example in hydrogen sensor,
palladium wire is attached to the sensor. Hydrogen absorption changes the
dimension of palladium wire. This results in the introduction of strain in the
optical fiber. Strain changes the path length of the light propagating through
the fiber.
iii) Light guiding application : Optical fibers are used for light guiding in science,
industry and medicine. Since fibers are flexible, light can travel through these fibers
in any zigzag manner. This is used in medical diagnostics in endoscopies and
angioplasty. Endoscopes are used to see the blocks and tumors in the body and
angioplasty uses optical energy to kill cancer cells or dissolve the blocks of an
artery.
iv) Military and Aero space applications : Fiber optic technology can be used for
military applications for communication, guiding weapons and submarines. As already
mentioned, optical fibers are free from electromagnetic and radio frequency
interferences and the communications through fibers is secured. Fiber optic
communications are suited for military applications. The fibers used must be nuclear
radiation resistant.

Questions

Using rectilinear propagation of light and Snell’s law explain the transmission of
light through optical wave-guides.
Define numerical aperture and acceptance angle. Derive a relation between the
two.
Explain different modes of light waves in optical fibers.
What are skew rays and meridional rays?
With help of refractive index profile distinguish the step index and graded index
fibers.
Distinguish between the single mode and multimode step index fibers.
Discuss various types of losses in the optical fibers.
What are different types of scattering losses in optical fibers?
oo
ND Discuss an arrangement used for fiber drawing and coating using glasses
. Write short notes on:
a) Fiber buffering ) cable strength and structural members
c) Cable sheath

423
 Applied Physics

. Explain various fiber packing methods used

. Discuss various types of materials used in fiber optic cable manufacture.
. Why a fiber is coated with polymer?
. Explain temperature and pressure sensing applications using optical fibers.
. What are chemical sensors? How are they different from temperature sensors?
. Explain the advantages of fiber optic communications
. Explain the medical applications of fiber optic sensors.
. Find the numerical aperture and acceptance angle of an optical fiber with core
refractive index 1.55 and clad refractive index 1.53.
. An optical fiber core refractive index is 1.465 and that of clad is 1.45. Find the
critical angle at the core-clad interface.
20. A Step Index fiber of core diameter 200 um and its numerical aperture is 0.2.
When a light of 850 nm is propagating through fiber. Find the number of
propagating modes.
Zi Discuss the meaning of a 'Mode'.

424
APPENDIX - I
SE
eee eee
Physics deals with vector quantities which are functions of position variable
and time. We talk of two types of fields, (1) Scalar fields and (2) Vector fields. A scalar
field is function of position and time. The temperature of water kept in a vessel on
heater can be considered as an example. The temperature in the vessel may change
initially from point to point and with time. The position of any point is described by (x,
y, Z, t) or T (r,t). T is a quantity which is specified by the magnitude at each point of
space and time. It is a scalar field.
A vector can be differentiated. Suppose A is a vector and is a function of variable
u that is scalar. Then

dA AA . A(u + Au) — A(u

du
ie Limyy 50 AU = Limay50 See)
“ =
This derivative is a vector in the direction of A as u approaches zero.

If A= C+ D, then A = C + D

If A = C x D, then A =16 De D xC

Gradient of a scalar function or field

Let $(r) be a scalar function of position in space. In this space, consider two
levels of , called the surfaces of constant o. Let them be o and o + do

425
 Applied Physics

OP =r; OP'=r+o6r
Then, PP' = 6r
The rate of increase of at P in the direction of PP' is 5d / dr with the limit
ér > 0.
The normal n at P touches the second level at Q. PQ is in the direction of n.
)
The maximum rate of increase of 9 is a) and is along n, i.e. 8 = 0
on

56n=65rCos0

5
5p oo Le
dr Sn ~“Sr On

Pet, 6d 8h
Limit — < —
mm &r-90 ér eh,

0
2 gives the greatest rate of increase of at P. It is called the gradient or grad of

function 9.

Grad $ = = n

The gradient of a scalar function is a vector function.

In general, grad 6 = [Ses

Op db
oe
Ox é Ob
Ans = . au
=

OOS,
= Fon Ns OUP OO PCL
mel Can Com nem,
on eee

0
Where V = pou eave
OX) "OY. Oz

V is a differential operator, del.

The divergence of a vector field :
V is a vector operator and it must operate on an operand. If it operates
on a
scalar function, we get the gradient of a function that is scalar. We
will see what will
happen when it operates on a vector field. Consider a vector A =
A,i+ A,j + A,k

426
Appendix I

The devergence of vector A is written as

VeA=|i—t+j—t+k—|
faeGov «Oo sanon e(A.i i
: CGOx Yay =, Ay +AK)
(A,i+

1) OK go OAyiOag
wadlhda) Oy Oz
Therefore, the divergence of a vector function or fields is a scalar.
In order to understand the concept clearly, we take the example of a flow of
fluid in a medium. Consider a parallel piped section in the mass of the fluid.

— ee

Along y direction, the velocity of the fluid is Vy. With this velocity the fluid
enters the surface A. Then, the rate at which the mass of the fluid that enters the face
A is pV, x = AxAz. The rate at which the fluid mass leaves the surface B is
av,
el baa AxAz

Therefore, the net outward flow in y direction is

es AxAyAz
oy
Similarly for x and z direction, the net outward flows are

fe) os AxAyAz and p oe AxAyAz

oy oy
The total net outward flow from the volume Ax Ay Az of the parallel piped section is

y
p+ a an dB Az
Z
Oy y y OZ

V ¢ (pV) = div (pV)

This is true for heat flow, or for the electronic flux lines or magnetic flux lines.

427
 Applied Physics

If the divergence of a field is positive at a point, it means that there is a

source of the field at that point. If it is negative, there is a sink at that
point. If there are no sources or sinks, the divergence of a vector field F is zero
i.e. div F = 0 such a field is called solenoidal field. The best example is the
magnetic flux density B. The divergence of B is zero i.e. B is solenoidal.
There are three kinds of integrals in vector integration. They are 1) Line integral,
2) Surface integral, 3) Volume integral. The integrand can be vector or scalar field or a
function.
Line integral:

Consider a space. Let a vector function or field F (x, y, z) be defined at any

point in this space. Consider two points A and B in this space. The line integral from A

to B along the path C is denoted by J F.d where dl is an infinitesimal vector

‘
displacement along the curve C.
We divide the path into very short segments. Each segment is represented by a
vector dI and we take the scalar product of the path segment vector dI with the vector
F. Now, we add these products over whole length. If dl is inifiniliseally small, we can
replace the sum by integration.

Line integral J F.dI= i F.dr.

Where dr = dri + drJ + drk, ie. [ Fdr = [(F.dx +F,dy + F,dz)

The line integral may not depend on the path. There can be infinite number of
paths from A to B. Consider two path C, and C,, then for a conservative field

If we travel along path C,, and return along the path C,, then

B (c D A
jF.d= [F.d+[F.d+ [F.d+ [F.d= fra— fra
A B (e D A (

{ is called closed integral.

If F = Vo then
B B B
JF.dI+[Vo.dI =[dd=6,
-0,
A A A

428
Appendix I

This is independent of the path but only depends on the end points. If the closed
integral is equal to zero, then the field is called irrotational field.
Surface integral : i

Ss

Consider F as a vector field

The dot product F. 5S represents the flux through the path of surface area S. If
we add all these products for all patches of surface we get the flux through the entire
surface.

=F .8s= JJ Fads
Ds = I ds, + j ds, + k ds,. The double integral indicates the two dimensional
nature of the surface.

>= [(F,ds, + Fyds, + F,ds,)

Ss

A closed surface encloses a volume e.g. a sphere. Then the surface integral of
the normal component of a vector point function F taken over a closed surface
S enclosing a volume V is equal to the volume integral of the divergence of a vector
F taken through the volume V.

ff ras= JJ] divrav

or

( OF, OF, OF,)

JJ) Gaz, az| dxdydz
JJ4s, + Fydsy ai Gs; ) =

The above theorem is called Gauss's divergence theorem. The proof of this
theorem can be found in any standard text books of vector analysis.
Curl a vector:

The line integral around a closed path is zero. Then the field is called irrotational
field. There are several other vector fields like the magnetic field, whose line integral
around a finite closed path is not zero. These fields are known as rotational fields.

429
 Applied Physics

Consider a small or an incremental plane area As in space. The region As is a

very small region of rotational vector field.

If the vector field is normal to plane, then line integral along its sides is
B Cc D A
fF.dI = [F.dI+ [F.dI+ [F.dI+ [F.dI =0
A B ( P

If the plane is rotating through 90° such that the vector field is parallel to it, then

B C D A B D
fF.dI = [F.di+ [F.dl+ [F.dl+ [F.dl = {F.d1— {F.dI
A B Cc D A C
This gives a finite value. Therefore, the line integral along the edges of the
surface As depends on its orientation with respect to the vector field. At particular
value of the area through which the vector field penetrates, the line integral takes
maximum value. This maximum line integral at any point per unit area in such a way
the vector field forms a closed path or loop, is called Curl of a vector field.

Curl F = Limit,, 9 pF .dl

Curl F = J
Appendix I

The Curl of a vector field is a vector quantity directed along the normal to the
unit area towards which a right handed screw moves when turned in the sense in
which the line integral along the boundary of the test area is maximum.
Referring to the Fig.----- let us calculate the value of the component of F in the
x direction by taking the integral of F around the periphery of the area Ay Az (As) in
four incremental steps. Let F along side 1 equal Fy and along side 4 equal Hz. If the
field is not uniform, the values of F along sides 2 and 3 are given by:

OF,
Baineeyios) Hee!
aia
ay Az

oH
{F.dI = FAy + F,Az + ay AyAz — Fay

OF,
Sid AzAy = F Az = JAyAz

Dividing by the area Ay Az,we obtained the component of Curl F in the

x-direction as

; f F.dI OF, OF,

Curl,F = Lima, az40 iyi = By arean

If the field is in all direction, then Curl F also has components in the y and z
direction.
. Generally

OF, OF, OF, OH OF, OF,

Curl F =i fs rae Tie r=
Oy P1B0Z OZ OX COT OY

=i, + Jy + kJ, = J
Curl F= J

i J k
Vx F=|0/0x d/dy 0d/dzi.
i E ES

If F is replaced by magnetic field intensity then V x H= J

J is the current density. This is differential form of ampere's law.

431
 Applied Physics

Stokes theorem

Consider a infinitesimal area As in the xy plane.

The vector field E have components Ey and Ex as shown.

fE.dl
Curl E = Limit ,, _,o 7

Le an area be x,, y, as shown in Fig. Let this area be divided into infinitesimal
areas as shown. The Curl of E at & point in the small area is equal to the line integral
around it. If the Curl of E is integrated over entire area X,, y,, all contributions to the
line integral cancel except those along the periphery of the area X,, y,- The integrai of
the normal comporient of the curl over the area x, y, must equal the line integral of E
around the periphery of the area. That is

fE.dI = [Ax E.ds

This is called Stoke's theorem. It states that the line integral of a vector function
around a closed contour C is equal to the integral of the normal component of the
having the curl of the that vector function over any surface having the contour C as
its
bounding edge. If E is the electric fields ar every point, then

JJVxE= fed = aes

: ot

The later point of the equation is form Farafays law of electromagnetic

induction.
VxE OB
UNA oS) Onin
ot

432
APPENDIX - II

I. Measurement of Dielectric constant:

A large number of methods exist for the measurement of permittivity. The
natures of the specimen, the range of frequency, the temperature etc., determine the
method that is adopted. We will only describe only two methods suitable for a large
variety of substances.

The measurement of real part of permittivity, &', is done by the measurement

of capacitance of a capacitor with and without the dielectric in between the electrodes.
The imaginary component of permittivity, €", is calculated from the dissipation factor
-D or loss factor tan §.

Capacitance Comparison Bridge :

An ac bridge is used for the measurement of an unknown capacitance by

comparing it with a known capacitance. The schematic diagram of the bridge is
shown in Fig. A-II.1

Fig. A-ll-1

433
 Applied Physics

The ratio arms are both resistive are represented by R, and R,. The source of
emf (E) is connected at the points B and D. The detector is usually a galvanometer or
other sensitive current meter. The current through the detector depends on the potential
difference between A and C. The bridge is said to be balanced when the potential
difference across A and C is zero so that there is no current through the detector. This
condition occurs when the voltage from point A to B equals the voltage from point C
to D. Therefore the bridge is balanced when I, R, = I, Rp.
If the detector does not show any current,

sil
1 3
enee
R, +Z3

d press
an 2 Ro +Zy

Z, and Z, are the impedance in arm 3 and 4 respectively. In the bridge shown above,

Zn Ro
ERT
ac, and Z,=R,

The condition tor balancing the bridge is

R,Z,
= R,Z,

Le, Ry, = a Re- J

OCy OC.

i.e |
Np Meeps J = RR. =R
acy. 2 eC.

The two complex numbers on either side of the above equation should be
equal. That means, the real terms and the imaginary terms are equal.

434
 Appendix II
a eg ee
R, is the variable resistor for adjusting the balance of the bridge. R, is also
kept, as variable resistor in practice and the process of alternative manipulation of
R, and R; is typical nature of this bridge circuit. The frequency of the ac source does
not enter into the final balance equation. Therefore the bridge is said to be independent
of the frequency of the applied voltage. However, the bridge is not suitable for high
frequencies and the loss factor cannot be measured.
The other method is by the employment of Schering Bridge. The Schering
Bridge is usefuf for frequencies 10* — 107 Hz. The basic circuit is shown in
Fig.A-II-2. The arm | contains a parallel combination of a resistor and a capacitor and
the standard arm contains only a capacitor. The standard capacltor is usually a high
quality mica capacitor. The mica capacitor has very low losses and therefore a phase
angle of approximately 90° is incorporated in the arm.

Fig. A-II.2
The balance conditions require that the sum of the phase angles in arms
2 and 3 should be equal to the sum of the phase angles in arms | and 4.
0, + 0, = 0, + 8,
Since the standard capacitor is present in arm 3, the sum ofthe phase angles of arm
2 and arm 3 will be 0° + 90° = 90°. In order to get the 90°-phase angle needed for
balance, the sum ofthe angles of arm | and arm 4 must be equal to 90°. The dielectric
kept in the capacitor in arm 4 does possess a loss factor, the phase angle in arm 4 is
less than 90°. Therefore, it is necessary to give arm | a small capacitive angle by
means of the capacitor C, (small capacitor).

435
 Applied Physics

The balance equations are

or ED
Ry WCy ee wc, Ry

436
 APPENDIX - III
eek 2 el 2 en ee ipee we pedis

Quantum Theory of Radiation

I. Black Body Radiation :

When a body is heated to a high temperature, it emits radiation that is spectrally

continuous. The distribution of energy over different wavelengths is described as the
spectral radiancy or spectral emittance. The spectral radiancy is defined as the radiant
energy flux (power per unit area) emitted by the surface of a glowing body per unit
wavelength interval. For every material, there exists a family of spectral radiancy
curves like the one shown in Fig. A-III.1. The radiation is spread over from short
wavelength to long wavelength (infrared). There exists a peak.

5000

‘E4000
ee
2 3000
= |
=e
53
oO
2000
f=)
oO

1000

1000 2000 3000 4000 5000 6000

wavelength A [nm]

Fig. A-III. 1 Spectral emittance of aglowing body

The flux of the thermal radiation from a glowing body depends on the
characteristics of the surface. The radiation that is coming out of the surface of a
glowing body is generated within the volume of the body by random thermal motion
of atoms of the body. Before it reaches the surface and emerges out, it is absorbed
and re-emitted many times. It attains thermal equilibrium with the atoms. This equilbrium
process shapes the continuous spectrum of radiation completely washing out all of
the original spectral features of the radiation. Only a fraction of radiation comes out
through the surface of the body. The process of absorption and emission is a common
feature of all bodies. A good absorber is a good emitter, and a poor absorber is a poor
emitter.

437
 Applied Physics

A body with a perfectly absorbing surface is called a black body. It is a

hypothetic body that does not reflect light and therefore, appears black, if the
temperature is low enough so as not to be self luminous. All black bodies heated to any
temperature emit thermal radiation.
The ideal characteristics of a black body
are achieved ina cavity radiation. A cavity is a hallow
cube with a tiny hole in one side of the cube as
shown in Fig. A-III.2. The walls of the cavity are
made of suitable materials (tungsten, tantulum,
molybdenum etc). When the cavity is heated, the
hole acts like a black body and the radiation is emitted
through the hole.
The radiation from a black body does not Fig. A-III. 2 A cavity
depend on the type of the material of which it is
made. The radiancy (represented by R,) depends only on the temperature as
Raa oO T*, o is a constant called Stefan - Boltzmann constant. Its measured value is
5.67 x 10-8 watts/m? K4. The above relation is call Stefan - Boltzmann law.
In a black body, the spectral radiancy R, is a universal function of wavelength
A, and temperature T, and of nothing else. A general form of radiation spectrum of a
black body is shown in Fig. A-III.3 for different temperatures.
Wien showed that the product of the wavelength of maximum emission Cn)

2000 K
5000

‘E4000
=
>
2 3000
42}

3 2000
>

oO

S
1000 1 1250K

1000 2000 3000 4000 5000 6000

wavelength A [nm]

Fig. A-III. 3 Spectral emittance of a black body

438
 Appendix
ee III

and the absolute temperature (T) is a constant equal to 0.29 cm K.

This observation
of Wiens is called Wien's displacement law. Wien also showed on empirical
basis that
spectral radiation density varies with wavelength and temperature as
given by the
relation

Where A is the wavelength, T, the absolute temperature, and e, the base

of
natural logarithms. c, and C, are two adjustable constants. Wiens formula was
obeyed
upto long wavelength range. However, it was Lord Rayleigh who made, for
the first
time, an attempt of give a theoretically derived expression for the spectral radiancy
of
a black body. Rayleigh considered that radiation in a cavity consists of a large number
of standing waves (shown in Fig. A-III.4) of electromagnetic waves. Each standing

Number of modes per unit frequency per unit volume

8nv-
3

Less modes More modes

Fig. A-III. 4 Cavity and standing waves

wave is made of vibration of the cavity. Each mode of vibration has an average thermal
energy, kT. (k is the Boltzman constant). According to Rayleigh - Jean, the number of

Z
modes within a frequency interval dv is equal to Bry, dv , c is the velocity oflight.
c

Then, the radiant energy in a frequency interval d v is equal to 8nv2 kT dv . In terms

Cc

of wavelength, this expression is written as

R, dd =8 kT dh
r

439
 Applied Physics

The number of modes depends upon the frequency. Higher the frequency,
higher are the number of modes. The above equation is plotted in the Fig. A III - 5.

* The "Ultraviolet
5000K =, Catastrophe"
2n
:
‘
(=| J
oO ry
i) t

pa ,
oD ‘
i ‘
S ’
o ‘Rayleigh
se %

=
so
. Jeans Law
.

fas}
[a4 Planck
Radiation
Formula
".
"=.
oe SA

1000 2000 3000

Wavelength ofradiation innm
Fig. A-IIl. 5 General form of black body radiation spectrum

The equation fits only at long wavelengths, at short wavelengths or high frequencies,
it leads to infinite values of radiant energy. This deviation is called ultraviolet catastrophe.
The Rayleigh's law was derived on the basis of classical physics. It does not possess
any arbitrary constants. It fits exactly at long-wavelengths. The failure of Rayleigh's
law at short wavelengths presented a crisis on the very concepts on which it was
based. The classical theory was unable to account for the observed radiation spectrum.
This led Max Planck to look into the phenomenon a fresh. Max Planck (1900)
abandoned the wave concept of radiation. He suggested a revolutionary concept,
namely, the quantization of radiant energy. Planck first modified Wien's formula by
putting a mere, — 1, into the denominator of the Wien's formula to get

che
poe
Sox AT

and he adjusted c, and c, to suit Rayleigh's law at long wavelengths. He found that his
formula fitted the experimental observations in every wavelength range.
Planck looked into the theory form a different angle altogether. Rayleigh's
derivation never had a flaw in the derivation and Max Planck thought that the flaw
must ultimately lie in the concepts involved in the classical physics itself. Let us look
‘into Planck's theory without any rigorous mathematical steps. It should by remembered
that Planck's hypothesis of quantization of radiation field gave birth to quantum theory
in physics' which became the spring board for many concepts and theories that followed
in twentieth century in atomic physics, spectroscopy, nuclear physics, solid state
physics etc.

440
 Appendix
(<==
ee
III
RCE ener nee

According Planck, the atoms in the wall of the cavity should be regarded
as
small harmonic oscillators with electric charges. The motions of the oscillators
result
in the emission of electromagnetic radiation into the cavity and absorption
of energy
from the cavity. The cavity is filled with radiation and the average rate of emission
of
radiation energy by the oscillators is equal to the rate of absorption of radiation
energy.
This is the sate of thermal equilbrium. It is now possible to deduce the characteristics
of the radiant energy in the cavity from those of oscillators with which the cavity
is in
equilbrium. Max Planck suggested two postulates. They are : 1. The oscillator
can
emit or absorb energy that are permitted. If, v, is the frequency of the oscillator, then
the energy is B——*0-liv- 2hv. shy nhv i.e., the energy of the oscillator is
quantized, h, is called Plank's constant, and n is called ‘quantum number'. The energy,
hv, is called energy quantum of radiation. This was later called as ‘photon’. The energy
values ofthe oscillator can be written as E,, = nhv ; n, is an integer. The oscillators
do
not, thus, radiate energy continuously but only in quanta. This quanta of energy is
emitted from the oscillator when it goes from higher energy state to lower energy
state. With this quantization condition, Planck calculated the average energy of the
oscillators and from that he derived his radiation law. Although, we will not go into
vigorous derivation, we shall briefly out line the steps involved in the derivation.

There are a large number of oscillators associated with walls of the cavity.
They possess energies given by E, =n. hv. The energy states of the oscillators are
pivenias Es = Ove Zhivs Shiv, a0. nhv..... Each energy vlaue is possessed by large
number of oscillators. Let Ny osicllators be in the energy state Ey, Nj, in the energy
state E, = hy, ..... and so on.

Energy : Eo; E, E,, E yon E,

The number of ocillators : Noein Nos Na race N,

nhv ees
Then, Nasal ex ~) (Boltzmann distribution law)

where N, is the total number of oscillators, and DUNn =N.

n

The energy contributed by the N,, oscillators is obviously

nh
N,nhv=nhvN exp)

44]
 Applied Physics

Then, the average energy contributed by N,, oscillators is given as

— [any)
) nhvN Exe
ok
n=0
ee)
aa NY:
>| Nexp =
kT

Since n is an integer, the above equation can be written as

0 + hv exp a8 +. 2h
v exp fay +3hv exp Eu cheese
= kT kT kT
= i

ex mn) ex ea eX mea rh
ea si Ee eT ans ip RS

Suppose, x site
= ©XPlo.tay,
kT

— (Nee ae
then, © =hvx 5
14+x+xX* 4...

ee ex (1
PAT
The number of modes of vibration, in a unit volume of cavity space, is equal

8TV : ,
10 eae dv. Thus, the radiant energy density in a frequency range dv is given as
Cc

8mv hv
Rydv = 3 my dv
ocala 5 aa)
kT

2 8t v3 h 1 F
= ; “y

kT
 Appendix
a III
ee

or in terms of wavelength A, the radiant energy density in

wavelength range dA is
given as

———__.— dr
a> exp(he/AKT)=1

The above relation is Planck's law for energy density of black body
or cavity
radiation. This law also contains Wien's displacement law as well as
Stefan-Boltzmann
Law. By integrating Plank's law, we obtain the total flux or power
per unit area,
emitted by the black body at all wavelengths

ce oa
8 m he l
. JR
= IR), dA = ad = 4
J je eplheuo G

2nk4
where at eens dc as AW)ak 4
15c*h

The Wiens displacement law is obtained by finding the wavelength for maximum
total energy radiated per unit time per unit area at a fixed temperature, 1.e.,

d | Sachi hh :
dd] exp(ch/AkT)-1

then, we get i rg

ch
At longer wavelength, - el gl be cman cr) ol Nea |

Then, R,da = 8 thc A> exp (— ch/AkT)

da
— Wien's law

443
 Applied Physics

This is shown in Fig. A HI - 5 in which the validity of black body radiation is

given.
Planck's law

5000

4000 Rayleigh-Jeans law

3000
Wien's ey
energy 2000
[J/m*]
density empirical law “**S

1000S RP) ORE ee

Fig. A-III. 6 The validity of the three laws of Blacic body radiation
II. Photoelectric Effect :
When electromagnetic radiation of high energy like ultraviolet from a source
(Fig. A-III.7) strikes a metal, it emits electrons. The emitted electrons are called
photoelectrons. The phenomenon was first experimentally verified by Leonard.
Subsequently J. J. Thomson, R.A. Millikan and several others found that many alkali
metals like lithium, sodium, potassium etc., eject electrons when inadiated with light
-Millikan made a number of experiments regarding the dependence of emission of
electrons on the energy or frequency of the electromagnetic radiation.

© Source

quartz window

Fig. A-III.7 Experimental arrangement for studying photo electric effect

444
Appendix III

The number of electrons emitted and the kinetic energy of the photoelectrons
depend on |. the potential difference between the two electrodes C and A, 2. the
intensity or the flux density of the incident radiation, 3. The frequency of the incident
radiation, 4. the nature of the metal used. The significant features of the effect are
summarized below.
1. The rate of electron emissiion is proportional to the intensity of the incident
light, the energy of the light being kept constant. The relation between photoelecttic
current and the light intensity is linear as shown in Fig. A.III-8.

Photo-electric
current
Light Intensity I

Fig. A-II1.8
2. The photo electric current is a function of frequency of the incident light,
when the intensity of light and the accelerating voltage between the cathode and anode
is kept constant. The aspect is shown in Fig. A-III.9.

Photo-electric
current

Ve Frequency ———>

Fig. A-III.9
3. The number of photo electrons or photo current is a function of the potential
difference V between the cathode and anode. As the voltage V is increased, the
photoelectric current increases and reashes a particular saturation value beyond a
ceratin voltage. If the potential difference is zero, there exists still a photo current.
This indicates that the electromagnetic radiation gives a conducting path to the electrons.

445
 Applied Physics

When the voltage V is made negative, the photo current does not immediately
become zero. There is a particular value of voltage V, at which the photo electric
curent becomes zero. This negative potential of the plate A at which the photoelectric
current becomes zero is called stopping potential (V,).
This aspect is shown in Fig. A-III.10.

Fig. A-III.10 potential difference V vs photoelectric current at different intensity levels

I, > 1, >1,. Fora given metal.

4. The stopping potential V., varies with the frequency of the incident radiation.
The relation is a linear relation (shown in Fig. A-III.11).

Ss
>
s
=o
iS
ow
on
dS Frequency —_»
&
2
N ’

Wo
€

Fig. A-IIL.11

446
Appendix III

5. There exists a limiting frequency v, below which photoelectric current

does not exist. At frequency, vo, the stopping potential V, is zero. The limiting frequency,
Vo, is called the threshold frequency and the corresponding wavelength A, is called
threshold wavelength. It is the maximum wavelength of the incident radiation beyond
which there is no photoelectric emission.
Einsteins photoelectric equation :
The explanation of photoelectric effect was put forward by Einstein in 1905
for which he was given a noble prize. The explanation was the first triumps of quantum
theroy of radiation. Einstein extended Planck's hypothesis that quantam nature or
quantization was an intrinisic property of electromagnetic field. Einstein proposed that
the absorption of light of frequency v by electrons takes place in discrete quanta of
energy, hv. These quanta of energy preserve their identity throughout their life.
When a light quanta called photon impinges an atom or absorbed by the
metal, it gives all its energy to the electron and the electron is emitted or ejected
violently and such an event follows probability function. Ifthe light intensity is small,
the chance or probability of photons hitting the electrons is less. Therefore, the
magnitude of photo electric current is less. The photon has an energy |hv |, where v
is the frequency. For light frequency zero and the threshold value, vo, there are no
photoelectrons produced. The energy, hvo, is called the threshold energy or the work
function, W, of the metal. This is the characteristic of the metal. There is an upper
limit to the energy of the emitted photo-electrons produced. No photoelectron can
have energy in excess of the energy of the incoming photon less the threshold energy,
or the work function. Since, the stopping potential is a measure of the maximum
kinetic energy of the ejected electrons, the graph between the stopping potential and
the frequency should be a straight line and this was what was observed.

Einstein's interpretation can be put down as an equation, called photoelectric

equation. The energy of the incident photon is hv. This is equal to the sum of the
work function, hvo, of the emitter/collector metal and the maximum kinetic energy
that an ejected photo electron can acquire.
|
hv =-hye 5) me Vics ( V = veolcity )

I 2 =
or fi 04 aR hv — hvo

or eVo= hv— Wo

This is a linear equation. The slope ofthe line, V, plotted as a function of v is

equal to h/e. vg is the v-axis-intercept. The above equation was first verified by
Millikan experimentally.

447
 Applied Physics

III Compton Effect

Prof. A.H. Compton observed that when monochromatic X-rays fall on a metal
foil, they are scattered. The scattered X-rays in any given direction have wavelength
higher than that of indicent X-rays. The scattered beam is thus modified. Apart from
the scattered radiation, the electrons are also recoiled. There is also an unmodified X-
ray beam which emerges out of the metal foil in the same direction as the incident
radiation.
Compton assumed that X-rays of wavelength, A, consist of a stream of bundles
of energy, called Quanta of Energy,

oes

The quanta of energy is known as Photon. When one of the photons passing
through the target (metal foil) hits an electron, the electrn would recoil. Its kinetic
energy has to come from the incident photon. Therefore, the photon, would be left
with lesser energy E'. After the collision, the electron with a momentum, p,, the the
photon with a momentom, p, will be emerge out of the metal foil.
The Fig. IlI.12 shows the process.

Fig. A-III. 12 Compton scattering

Before collision, the energy of the photon = E = hv; the moment

um of the photon,
E hv
a a eae The electron has only the rest-mass energy m,c?, where
m, is the rest
mass of the electron. Its momentum is equal to zero.

448
Appendix III

After the collision, the X-ray photon, with an energy E'= hv' and momentum p'
goes in the direction @ to the incident radiation. The electron recoils at an angle with
respect to the direction of the incident angle. The momentum of the electron is p, and
the energy is K,. From the conservation of momentum
i)

é gee Pp, cos >

hv' . ’
Ot 0 ee sin 8 — p, sin >

or p.c sing = hv' sin >

p. ¢ cos @ = hv —hv' cos6
Squaring and adding, we get from the above expressions,
p2c?=h?[
v?2+ v2-2vv'cos 6]
h2
or po =i (VW twee v V'cos 9)
e

hy
=> [w-vyP+2vv'-2vv'cos 0]
c

ae, pene
a Si il ow arene
2 vv

v-viP
But (v — v')? << vv’. Therefore, we can neglect the term rae We can

rewrite the above expression as

Oh 2 vy aetetvee I
Oe = 5 fti—cos 0] og 0
c
Conservation of energy demands that the kinetic energy of the electron equals
the energy lost by the photon,

I 2
i.€., 5 Me Ve =h(v- Vv’)

or Poem Rea I

449
 ee Applied Physics

equating, eqs I and II, we see that

Kinet ay"
2m,,h(v —v')= % (1 — cos0)

2
or Le (I -cos6)
Mec sere Il

This result, first obtained by Compton, is valid even if the recoil speed
ofthe
; Mo
electron is relativistic. In such a case, the mass-of the electron
m = [ae :
lve
where m,, is the rest-mass of the electron. However the express
ion given by III is
exact, even when relativistic expresion for mass is taken into
consideration.

The constant, he is called the compton wavelength. Its value is

ef

= 2.43 x 10-'? m = 0.0243 A . In spite of its name, this is not the

mec wavelength
of anything. The compton shift, 0' — X', is a function of
0. For 6 = 180° (i.e., a head-

on collision), the compton shift is maximum; the value

being 0.0485 A . The compton
effect provided the final confirmation ofthe validity of
Planck's hypthesis.

450
 Units for Magnetic Properties

Quantity Symbol | Gaussian & Cgs | Conversion | SI & rationalized

emu Factor, C mks
Magnetic flux density Fa ew Gauss (G) ieee 3Tesla (T), Wb/m?
magnetic induction
Magnetic flux ,y Maxwell (Mx), Weber (Wb), volt
G-cm’ second (V.s)
Magnetic potential Gilbert (Gb) 10/47 Ampere (A)
difference, mag-
netomotive force
Magnetic field H Oersted (Oe), 107/4r

:
strength, magnetizing Gb/cm

Magnetic polari-
zation, intensity of
magnetization ee eee ee
(Mass) magnetization | o,M emu/g l
4n x 10”
Magnetic moment emu, egG 10° A-m’, joule per
tesla (J/T)

v.16 Dimensionless, | 47 Dimensionless

emu/cm? (4m)? x 10°77 |henry per meter
(H/m), Wb/(A-m)
An m'/kg
(4x)? x10" |H-m'/kg
4nx10* |m’/mol
(4n)°x 10° H.m?/mol

H/m, Wb(A-m)
Dimensionless

Demagnetization Dimensionless
factor —

451
 Units for Electrical Properties

Quantity Symbol SI & rationalized mks

Change of Electron voor mane Coulombs, C
Boltzman constant Joule Kelvin', J K"'
Electric field intensity E -!

Dielectric polarization i ee Cm:

Spontaneous polarization
Electric Susceptibility
Pyroelectric coefficient
Dielectric loss angle
Dielectric loss tangent
Relative permittivity / Dielectric constant
Permittivity
Permittivity of vaccum
Capacitance a Farads C V"
Electrical conductivity

[ese
7 are a
Seebeck coefficient om Wake:
rane coefficient T K A’! dt
Hall constant R,, a x |
Hall mobility My, Visi =
! Refractive index n Numeral
| Electromechanical coupling factor K, Numeral |
Piezoelectric coefficient d CN’ =
Bohrmagneton Te dine
Nuclear Magneton Ly sara
Activation Energy U eV, J mot!
Piezoresistive constant 4 m* N’

452
 Values of Physical constants

Quantity Symbol Nominal Value More accurate

. value

Electron charge pone 1.60 x 10°C -1.6022 x 10°”

Rest Mass of the electron a 9.11 x 10°' Kg 9.10956 x 10°!
Electron change to mass ratio

a
(electron specific change) 1.76x 10"C kg" 1.758803 x 10"

Proton rest mass 1.67 x 10°’ Kg

3.00x 10'ms’ ‘|2.9979x10°
Avagacho’s number NA 6.0221 x 10”

5.292x10"m | 5.2917x10"
Permittivity of vacuum 1.260 nHm'

€) 8.854185

Planck’s constant ae h 6.63 x 10°7JS 5.62607 x 10"

Boltzmann constant k 1.38 x 107 JK"! 1.38062 x 10”

Bohr magneton B
Pant: 9.2740 x 10-24 JT"
Nuclear Magneton 6.05 x10" JT 5.0508 x 10°?’

Gyromagnetic ratio of proton 26) 100Se Tt: 2.6752 x 10°

P .

_|Gravitational constant G 6.673 x 10°" 6.67259 x 10"!

n’kg's?

Radian a3 57.29580

Impedance of free space 376.7 Q 376.730 (uC)

Atomic mass unit 1.66 x 10°’ kg 1.6605 x 10?”

(931.49 Mev)

AU 1.5x if Kim 1.496

Pi 3.14 3.1415927

453
 index

A D
Acceptance 409, 410, 411
Depletion Region 328, 331
Ampere’s law 254
Diamagnetism 262
Antiferromagnetism 288, 289, 290
Diamond 25, 41, 42, 43, 45, 46, 49
Apmere’s law 254
Dielectric 204, 205, 207

B Dielectric constant 198, 210, 211

Diode 327, 328, 331, 332, 333, 334, 335,
Band theory 170
336, 337, 344
BCS theory 368, 371
Dipolar polarization 204
Black body 390, 437, 438, 439, 444
Directional 385, 394
Black body radiation 390, 437
Domains 252,272
Bohr magneton 301
Bravais Lattices 30, 32
E
C Effective 168
Electric field 152, 154, 167, 168
Carrier 312, 313, 314, 317, 320, 3215323,
326, 328, 335, 342 Electronic polarizability 209, 210, 211

Claussios 204 Electronic polarization 204, 208, 210, 211

Coercive force 281
Coherence 398 F

Computer 252
Fermi level 166
Conductivity 151, 155, 156, 157, 166, 168,
169, 170 Fermi-Dirac 158, 159, 164, 165, 166
Cooper pair 368, 369, 370,375, 376 Fiber Packing 418
Covalent bond 25, 41,42, 45 Flash lamp 404
Curie temperature 270, 277
Fluorite 48
Curie-Weiss law 288, 290
Free electron gas 152, 158
Curl 255, 262
Free Electron Model 150, 157, 170
 O
sarnet 298, 299 Optical fibers 407, 408, 412, 418, 419,
420, 421, 423

dex fiber 413, 414, 415, 416, 417, 418

Paramagnetism 265, 266
\frared 384, 385
Perfect conductor 353, 354, 355
nsulators 157, 170
Permanent magnets 252
ntrinsic Semiconductor 307, 308, 310, 340
Permeability 197, 208, 210, 254, 258, 260,
nverse spinel 291, 292, 297
261
onic bonding 50
Photoelectric Effect 444

J Planck’s law 384, 443

Pumping 385, 388, 389, 391, 392, 394,
Josephson Effect 375 395, 398

L R
_aser 384, 385, 386, 387, 388, 390, 392 Ruby Laser 395, 403, 405

attice 25, 26, 27, 28, 29 30, 31, 32, 33,

34, 40, 41, 42, 43, 45, 46, 47 S
_ight Amplification 385, 390, 401 268, 273, 395
Saturation magnetization

Semiconductors 305, 306, 310, 312, 313,

314,318, 327,339

Magnetic 288, 289, 290 Simple cubic 32, 34, 41, 48

Magnetic recording 288, 299 Solid state laser 388, 395, 403, 405

Magnetization 260, 272, 273, ZI4A, 215, Space Lattice 25, 30

Dione 1, 216,219,201 SQUID 376, 377
Maxwell equations 407 Step index 413, 414, 415, 416, 417
Metals 25, 33, 39, 43 Superconductivity 348, 349, 350; 354352,
Minority carriers 335 355, 359, 368, 371

Mode coupling 416, 420, 422 Symmetry 25, 29, 30, 38, 41, 42

455
 7:
V
Tetragonal 33 Valence band 309, 312, 318, 342
Triclinic 33
Type ll Superconductors 364, 365, 366 WwW
U Wave packet 119, 120, 123, 138

Uncertainty principle 384 Z

Unit cell 33, 35, 38, 39, 40, 43, 46, 48, 49
Zinc blende 42
Zinc blende 45

456
 - SECOND EDITION

j
y \

ice
pa a Le ee Saia1
pee) Se ig : eee %
SS ee 2 Br cs 4 i sins = Se i ss Teer = es
oe & a EFT ene REI MSE Te rs To (pai ieee aeLae cy

—— Se i a ai a i a i a a ae 2
a i ?
aus ei Sd a aga

* x :

iol —eEaeSSSsSsSsS
=a
i