SERIES  ARTICLE

Snippets of Physics
11. Isochronous Potentials

T Padmanabhan

O sc illa to r y m o tio n o f a p a rtic le in a o n e d im e n -

sio n a l p o te n tia l b e lo n g s to a c la ss o f e x a c tly so lv -
a b le p ro b le m s in c la ssic a l m e c h a n ic s. In th is in -
sta llm e n t, w e e x a m in e so m e le sse r k n o w n a sp e c ts
o f th e o sc illa tio n s in so m e p o te n tia ls .
T Padmanabhan works at
IUCAA, Pune and is T h e m o tio n o f a p a rticle o f m ass m in o n e d im en sio n u n -
interested in all areas d er th e a ctio n o f a p o ten tia l V (x ) is th e sim p lest p ro b lem
of theoretical physics, w h ich o n e stu d ies in cla ssica l m ech a n ics. In fa ct, a fo r-
especially those which
have something to do with
m a list w ill co n sid er th is a s a so lv ed p ro b lem , in th e sen se
gravity. th a t th e d i® eren tia l eq u a tio n g ov ern in g th e m o tio n ca n
b e red u ced to a q u a d ra tu re; ie., th e tra jecto ry o f th e
p a rticle ca n b e ex p ressed a s a n in d e¯ n ite in teg ra l. In
sp ite o f th is a p p a ren t triv ia lity o f th e p ro b lem th ere a re
so m e in terestin g su rp rises o n e en co u n ters in th eir stu d y.
U sin g th e co n sta n cy o f th e to ta l en erg y, E = (1 = 2 )m x_2 +
V (x ), o n e ca n w rite d ow n th e eq u a tio n d eterm in g th e
tra jecto ry o f th e p a rticle x (t) in th e fo rm o f th e in teg ra l
r Zx
m dx
t(x ) = p : (1 )
2 E ¡ V (x )

T h is d eterm in es th e in v erse fu n ctio n t(x ) fo r a g iv en

V (x ) a n d th e p ro b lem is co m p letely so lv ed . In th is in -
sta llm en t, w e a re in terested in th e ca se o f b o u n d ed o s-
cilla tio n s o f a p a rticle in a p o ten tia l w ell V (x ) w h ich
h a s th e g en era l sh a p e sh ow n in F igu re 1 . T h e p o ten tia l
h a s a sin g le m in im u m a n d in crea ses w ith o u t b o u n d a s
jx j ! 1 . F o r a g iv en va lu e o f en erg y E , th e p a rticle
w ill o scilla te b etw een th e tw o tu rn in g p o in ts x 1 (E ) a n d
Keywords x 2 (E ) w h ich a re th e ro o ts o f th e eq u a tio n V (x ) = E .
Classical mechanics, oscillation. T h e p erio d o f o scilla tio n ca n b e im m ed ia tely w ritten

998 RESONANCE  November 2008

 SERIES  ARTICLE

Figure 1. A one-dimen-
sional potential with a
single minimum which
supports oscillations.

d ow n u sin g eq u a tio n (1 ). W e g et
r Z x 2 (E )
m dx
T (E ) = p : (2 )
2 x 1 (E ) E ¡ V (x )

F o r a g en era l p o ten tia l V (x ), th e resu lt o f in teg ra tio n

o n th e rig h t h a n d sid e w ill d ep en d o n th e va lu e o f th e
en erg y E . In o th er w o rd s, th e p erio d o f o scilla tio n w ill
d ep en d o n th e en erg y o f th e p a rticle; eq u iva len tly, if o n e
im a g in es relea sin g th e p a rticle fro m th e lo ca tio n x =
x 1 , say, o n e m ig h t say th a t th e p erio d d ep en d s o n th e
a m p litu d e o f o scilla tio n .
F o r a sim p le cla ss o f p o ten tia ls, it is q u ite ea sy to d e-
term in e th e sca lin g o f th e p erio d T w ith th e en erg y E .
C o n sid er, fo r ex a m p le, a cla ss o f p o ten tia ls o f th e fo rm
V (x ) = k x 2 n w h ere n is a n in teg er. T h ese p o ten tia ls a re
sy m m etric in th e x -a x is a n d h av e a m in im u m a t x = 0
w ith th e m in im u m va lu e b ein g V m in = 0 . In th is ca se, b y
in tro d u cin g a va ria b le q su ch th a t q = (k = E )1 = 2 n x , th e
en erg y d ep en d en ce o f th e in teg ra l in (2 ) ca n b e ea sily
id en ti¯ ed to g iv e
Z1
1 dq 1 1¡ n We find that, for all
T (E ) / p E 1=2n
p / E 2 ( n ) : (3 )
E 0 1¡ q 2 n values of n other
than n = 1, the period
W e ¯ n d th a t, fo r a ll va lu es o f n o th er th a n n = 1 , th e T has a non-trivial
p erio d T h a s a n o n -triv ia l d ep en d en ce o n th e en erg y. dependence on the
W h en n = 1 , w h ich co rresp o n d s to th e h a rm o n ic o s- energy.
cilla to r p o ten tia l, V (x ) = k x 2 , w e ¯ n d th a t th e p erio d

RESONANCE  November 2008 999

 SERIES  ARTICLE

is in d ep en d en t o f th e en erg y. T h is, o f co u rse, is th e w ell-

k n ow n resu lt th a t th e p erio d o f a h a rm o n ic o scilla to r
d o es n o t d ep en d o n th e a m p litu d e o f th e o scilla to r. T h e
a b ov e a n a ly sis a lso sh ow s th a t a m o n g st a ll th e sym m et-
ric p o ten tia ls o f th e fo rm V (x ) / x 2 n , o n ly th e h a rm o n ic
o scilla to r h a s th is p ro p erty.
L et u s n ow co n sid er th e in v erse p ro b lem . S u p p o se w e
a re g iv en th e fu n ctio n T (E ). Is it th en p o ssib le fo r u s
to d eterm in e th e p o ten tia l V (x )? F o r ex a m p le, if w e a re
to ld th a t th e p erio d is in d ep en d en t o f th e a m p litu d e,
w h a t ca n o n e say a b o u t th e form o f th e p o ten tia l V (x )?
S h o u ld it n ecessa rily b e a h a rm o n ic o scilla to r p o ten tia l
o r ca n it b e m o re g en era l?
B efo re la u n ch in g in to a m a th em a tica l a n a ly sis, let m e
d escrib e a sim p le ex a m p le w h ich d eserv es to b e b etter
k n ow n th a n it is. C o n sid er a p o ten tia l o f th e fo rm
b
V (x ) = a x 2 + (4 )
x2
in th e reg io n x > 0 . In th is reg io n , th is p o ten tia l h a s a
d istin ct m in im u m a t x m in = (b=pa )1 = 4 w ith th e m in im u m
va lu e o f th e p o ten tia l b ein g 2 a b. (B ein g sy m m etric
in x , th e p o ten tia l h a s tw o m in im a in th e fu ll ra n g e
¡ 1 < x < 1 , b u t w e sh a ll co n ¯ n e o u r a tten tio n to
It turns out that the th e ra n g e x > 0 . B y sh iftin g th e o rig in su ita b ly w e ca n
period of oscillation m a k e th e p o ten tia l in th is ra n g e to lo o k lik e th e o n e in
in this potential is F igu re 1 ). F o r a n y ¯ n ite en ergy, a p a rticle w ill ex ecu te
independent of the p erio d ic o scilla tio n s in th is p o ten tia l. It tu rn s o u t th a t
amplitude just as th e p erio d o f o scilla tio n in th is p o ten tia l is in d ep en d en t
in the case of a o f th e a m p litu d e ju st a s in th e ca se o f a h a rm o n ic o s-
harmonic oscillator
cilla to r p o ten tia l! S o clea rly h a rm o n ic o scilla to r is n o t
potential! So
u n iq u e in h av in g th is p ro p erty.
clearly harmonic T h ere a re sev era l w ay s to p rove th is resu lt, th e h a rd est
oscillator is not ro u te b ein g to eva lu a te th e in teg ra l in (2 ) w ith V (x )
unique in having g iv en b y (4 ); th e cu test p ro ced u re is p ro b a b ly th e fo l-
this property. low in g . C o n sid er a p a rticle m ov in g n o t in o n e d im en sio n

1000 RESONANCE  November 2008

 SERIES  ARTICLE

b u t in tw o (say in th e x y p la n e) u n d er th e a ctio n o f th e
tw o -d im en sio n a l h a rm o n ic o scilla to r p o ten tia l
1
V (x ;y ) = m ! 2 (x 2 + y 2 ): (5 )
2
C lea rly, su ch a p a rticle w ill o scilla te w ith a p erio d w h ich
is in d ep en d en t o f its en erg y. N ow co n sid er th e sa m e
p ro b lem in p o la r co o rd in a tes in stea d o f C a rtesia n co o r-
d in a tes. T h e co n serva tio n o f en erg y n ow b eco m es
1 1
E = m (x_2 + y_2 ) + m ! 2 (x 2 + y 2 )
2 2
1 2 _2 1
= m (r_ + r µ ) + m ! 2 r 2 :
2
(6 )
2 2
U sin g th e fa ct th a t fo r su ch a m o tio n { u n d er th e cen tra l
fo rce V (r ) / r 2 { th e a n g u la r m o m en tu m J = m r 2 µ_ is
co n serv ed , th e en erg y ca n b e ex p ressed in th e fo rm
1 1 1 J2 1 B
E = m r_2 + m ! 2 r 2 + 2
= m r_2 + A r 2 + 2
2 2 2m r 2 r
(7 )
w ith A = (1 = 2 )m ! 2 ;B = J 2 = 2 m . W e n ow see th a t,
m a th em a tica lly, th is is id en tical to th e p ro b lem o f a p a r-
ticle m ov in g in o n e d im en sio n u n d er th e a ctio n o f a p o -
ten tia l o f th e fo rm in (4 ). B u t w e k n ow b y co n stru ctio n
th a t th e p erio d o f o scilla tio n d o es n o t d ep en d o n th e
co n serv ed en erg y E in th e case o f (7 ). It fo llow s th a t
th e p o ten tia l in (4 ) m u st h av e th is p ro p erty. T h e a ctu a l Potentials like that
freq u en cy o f o scilla tio n is ! 0 = (8 a = m )1 = 2 w h ich is m o st of the harmonic
ea sily fo u n d b y u sin g th e fa ct th a t th e freq u en cy m u st oscillator or the
b e th e sa m e a s th a t fo r v ery sm a ll o scilla tio n s n ea r th e one in equation (4)
m in im u m . O n e m ay th in k th at sin ce ! 0 is in d ep en d en t are called
o f b, it m u st b e (2 a = m )1 = 2 fo r b = 0 . T h is is, h ow ev er, ‘isochronous
n o t tru e b eca u se h ow ev er sm a ll b m ay b e, th e p o ten - potentials’, the
tia l d o es rise to in ¯ n ity a t x = 0 th ereb y d o u b lin g th e term referring to
freq u en cy. the property that
the period is
P o ten tia ls lik e th a t o f th e h a rm o n ic o scilla to r o r th e o n e
independent of the
in (4 ) a re ca lled `iso ch ro n o u s p o ten tia ls', th e term refer-
rin g to th e p ro p erty th a t th e p erio d is in d ep en d en t o f amplitude.

RESONANCE  November 2008 1001

 SERIES  ARTICLE

In fact, for every th e a m p litu d e. It is n o t d i± cu lt to see th a t th ere a re

function T(E), one a ctu a lly a n in ¯ n ite n u m b er o f su ch p o ten tia ls. In fa ct,
can construct an fo r ev ery fu n ctio n T (E ), o n e ca n co n stru ct a n in ¯ n ite
infinite number of n u m b er o f p o ten tia ls V (x ) su ch th a t eq u a tio n (4 ) h o ld s.
potentials V(x) T h ere is a n elem en ta ry w ay b y w h ich o n e ca n co n stru ct
such that th em w h ich w e w ill n ow d escrib e [1 ].
equation (4) holds. N o te th a t th e p erio d T (E ) is d eterm in ed b y th e in te-
g ra l in (2 ) w h ich is essen tia lly th e a rea u n d er th e cu rv e
(E ¡ V (x ))¡ 1 = 2 . S u p p o se w e a re g iv en a p o ten tia l V 1 (x )
fo r w h ich th e en erg y d ep en d en ce o f th e p erio d is g iv en
b y a fu n ctio n T (E ). L et u s n ow co n stru ct a n o th er p o -
ten tia l V 2 (x ) b y `sh ea rin g ' th e o rig in a l p o ten tia l V 1 (x )
p a ra llel to th e x -a x is. T h is is d o n e b y sh iftin g th e p o -
ten tia l cu rv e h o rizo n ta lly b y a n a m o u n t ¢ (V ) a t ev ery
va lu e o f V u sin g so m e a rb itra ry fu n ctio n ¢ (V ). T h e
o n ly restrictio n o n th e fu n ctio n ¢ (V ) is th a t th e re-
su ltin g p o ten tia l sh o u ld b e sin g le va lu ed ev ery w h ere. A
m o m en t o f th o u g h t sh ow s th a t su ch a sh ift leav es th e
a rea u n d er th e cu rv e in va ria n t a n d h en ce T (E ) d o es n o t
ch a n g e. In o th er w o rd s, g iv en a n y p o ten tia l V (x ), th ere
a re a n in ¯ n ite n u m b er o f o th er p o ten tia ls fo r w h ich w e
w ill g et th e sa m e p erio d { en erg y d ep en d en ce T (E ); ea ch
o f th ese p o ten tia ls is d eterm in ed b y th e fo rm o f th e fu n c-
tio n ¢ (V ).
In th e ca se o f th e h a rm o n ic o scilla to r p o ten tia l, th e d is-
ta n ce h (V p) b etw een th e tw o tu rn in g p o in ts (`w id th ')
va ries a s V w h en th e p o ten tia l is m ea su red fro m its
m in im a . S in ce (4 ) h a s th e iso ch ro n o u s p ro p erty, w e
w o u ld su sp ect th a t it is o b ta in ed fro m th e h a rm o n ic o s-
cilla to r p o ten tia l b y a sh ea rin g m o tio n k eep in g th e w id th
h (V ) va ry in g a s (V ¡ V m in )1 = 2 . T h is is in d eed tru e a n d
ca n b e d em o n stra ted a s fo llow s. F ro m (4 ), w e ca n d eter-
m in e th e in v erse, d o u b le-va lu ed fu n ctio n x (V ) th ro u g h
th e eq u a tio n
a x 4 + b ¡ V x 2 = 0: (8 )
If th e ro o ts o f th is eq u a tio n a re x 21 a n d x 22 , w e im m ed i-

1002 RESONANCE  November 2008

 SERIES  ARTICLE

a tely h av e x 21 + x 22 = V = a a n d x 21 x 22 = b= a . E lem en ta ry For those of you

a lg eb ra n ow g iv es who do not like such
r
V b a geometric
h (V )2 = (x 1 ¡ x 2 )2 = ¡ 2 : (9 ) argument, here is a
a a
more algebraic
O r, eq u iva len tly,
derivation of the
1 same result.
h (V ) = p (V ¡ V m in )1 = 2 : (1 0 )
a
T h is sh ow s th a t th e p o ten tia l in (4 ) is in d eed o b ta in ed
b y a sh ea rin g o f th e h a rm o n ic o scilla to r p o ten tia l.
F o r th o se o f y o u w h o d o n o t lik e su ch a g eo m etric a r-
g u m en t, h ere is a m o re a lg eb ra ic d eriv a tio n o f th e sa m e
resu lt [2 ]. L et u s su p p o se th a t w e a re g iv en th e fu n c-
tio n T (E ) a n d a re a sk ed to d eterm in e th e p o ten tia l V (x )
w h ich is a ssu m ed to h av e a sin g le m in im u m a n d a sh a p e
ro u g h ly lik e th e o n e in F igu re 1 . W e ca n a lw ay s a rra n g e
th e co o rd in a tes su ch th a t th e m in im u m o f th e p o ten tia l
lies a t th e o rig in o f th e co o rd in a te sy stem . T h e sh a p e o f
th e cu rv e in th e reg io n s x > 0 a n d x < 0 w ill, o f co u rse,
b e d i® eren t. In o rd er to m a in ta in sin g le va lu ed n ess o f
th e in v erse fu n ctio n x (V ), w e w ill d en o te th e fu n ctio n a s
x 1 (V ) in th e reg io n x < 0 a n d x 2 (V ) in th e reg io n x > 0 .
O n ce th is is d o n e, w e ca n rep la ce d x in th e in teg ra l in
(2 ) b y (d x = d V )d V . T h is a llow s u s to w rite
p ZE · ¸
dx2 dx1 dV
T (E ) = 2m ¡ p : (1 1 )
0 dV dV E ¡ V
T h is is a n in teg ra l eq u a tio n w h ich , fo rtu n a tely, ca n b e
in v erted b y a sta n d a rd trick . W e d iv id e b o th sid es o f
th e eq u a tio n b y (z ¡ E )1 = 2 , w h ere z is a p a ra m eter a n d
in teg ra te w ith resp ect to E fro m 0 to z . T h is g iv es
Zz
T (E ) d E
p =
0 z¡ E

p Zz ZE · ¸
dx2 dx1 dV dE
2m ¡ p : (1 2 )
0 0 dV dV [(z ¡ E )(E ¡ V )]

RESONANCE  November 2008 1003

 SERIES  ARTICLE

The family of In th e rig h t h a n d sid e w e ca n ch a n g e th e o rd er o f in te-

curves which has g ra tio n a n d u se th e fa ct th a t th e in teg ra l ov er E ju st
the same width will g iv es ¼ . T h e rest is triv ia l a n d w e o b ta in
give rise to the
Zz p
T (E ) d E
same T(E) and p = ¼ 2 m [x 2 (z ) ¡ x 1 (z )]: (1 3 )
0 z¡ E
vice versa.
T h is is a n im p licit eq u a tio n va lid fo r a n y z . C a llin g th e
va ria b le z a s V , g iv es th e fu n ctio n a l fo rm o f x 2 (V ) ¡
x 1 (V ). W e g et th e ¯ n a l resu lt
ZV
1 T (E ) d E
x 2 (V ) ¡ x 1 (V ) = p p : (1 4 )
¼ 2m 0 V ¡ E
T h is resu lt sh ow s ex p licitly th a t th e fu n ctio n T (E ) ca n
o n ly d eterm in e fo r u s th e `w id th ' o f th e cu rv e x 2 (V ) ¡
x 1 (V ). T h e fa m ily o f cu rv es w h ich h a s th e sa m e w id th
w ill g iv e rise to th e sa m e T (E ) a n d v ice v ersa . T h e
sh ea rin g m o tio n b y w h ich w e tra n sfo rm o n e p o ten tia l to
a n o th er p reserv es th is w id th a n d h en ce th e fu n ctio n a l
fo rm o f T (E ).
O n e ca n a lso o b ta in so m e in terestin g rela tio n s in q u a n -
tu m m ech a n ics fo r th e co rresp o n d in g sy stem s. In q u a n -
tu m th eo ry, th e p o ten tia ls lik e th e o n e in F igu re 1 w ill
h av e a set o f d iscrete en erg y lev els E n . F o rm a lly in -
v ertin g th e fu n ctio n E (n ) { w h ich is o rig in a lly d e¯ n ed
o n ly fo r in teg ra l va lu es o f n { o n e ca n o b ta in th e in v erse
fu n ctio n n (E ) fo r th is sy stem . T h is fu n ctio n essen tia lly
p lay s th e ro le a n a lo g o u s to T (E ) in th e ca se o f q u a n tu m
th eo ry. W e ca n n ow a sk w h eth er o n e ca n d eterm in e th e
p o ten tia l V (x ) g iv en th e en erg y lev els E n o r, eq u iva -
len tly, th e fu n ctio n n (E ). It tu rn s o u t th a t o n e ca n d o
th is fa irly ea sily in th e sem i-cla ssica l lim it co rresp o n d -
in g to la rg e n . T o see th is, reca ll th a t th e en erg y E n o f
th e n -th lev el o f a q u a n tu m m ech a n ica l sy stem is g iv en
b y th e B o h r q u a n tiza tio n co n d itio n
Z r Z x2 p
1 x2 2m
n (E ) ' pdx = E ¡ V d x : (1 5 )
~ x1 ~2 x 1

1004 RESONANCE  November 2008

 SERIES  ARTICLE

T rea tin g x a s a fu n ctio n o f V , w e ca n tra n sfo rm th is

rela tio n to g iv e
r Z x2
2m p
n (E ) ' E ¡ V dx
~2 x 1
r ZE p
2m dx
= E ¡ V dV
~ 2 dV
r ZE
m
= (E ¡ V )¡ 1 = 2 x (V ) d V . (1 6 )
2 ~2
H ere w e h av e d o n e a n in teg ra tio n b y p a rts a n d h av e
trea ted th e in teg ra l a s a fu n ctio n o f th e u p p er lim it.
T h is in teg ra l eq u a tio n ca n a g a in b e so lv ed b y ex a ctly
th e sa m e trick w h ich w e u sed in th e ca se o f (1 1 ). T h is
w ill lea d to th e resu lt
r Z
2 ~2 n (V ) dn
x (V ) = p : (1 7 )
m V ¡ E (n )

w h ich d eterm in es th e fo rm o f th e p o ten tia l V (x ) { in

term s o f th e in v erse fu n ctio n x (V ) { su ch th a t in th e
sem i-cla ssica l lim it it w ill h av e th e en erg y lev els g iv en
b y th e fu n ctio n E (n ).
E v en th o u g h w e w o rk ed it o u t fo r a o n e-d im en sio n a l
m o tio n w ith a C a rtesia n x -a x is, it is o b v io u s th a t th e
sa m e fo rm u la sh o u ld b e a p p lica b le fo r en erg y lev els in a
sp h erica lly sy m m etric p o ten tia l V (r) p rov id ed w e o n ly
co n sid er th e zero a n g u la r m o m en tu m q u a n tu m sta tes. The formula given in
A s a cu rio sity, co n sid er th e p o ten tia l w h ich w ill rep ro - equation (17) should
d u ce th e en erg y lev els (w h ich w e k n ow is th e o n e a risin g be applicable for
in th e ca se o f th e C o u lo m b p ro b lem ) g iv en b y energy levels in a
spherically symmetric
4 2
m e Z potential V(r)
E n = ¡ : (1 8 )
2 ~2 n 2 provided we only
consider the zero
T h is g iv es n (V ) = (¡ 2 ~2 V = m e 4 Z 2 )¡1 = 2 so th a t a n ele-
angular momentum
m en ta ry in teg ra tio n u sin g (1 7 ) w ith a su ita b le ch o ice fo r
quantum states.

RESONANCE  November 2008 1005

 SERIES  ARTICLE

th e co n sta n t o f in teg ra tio n g ives

r Z · µ 4 2 ¶¸¡ 1 = 2
2 ~2 n (V ) m e Z Z e2
r= V + d n = ¡
m 2 ~2 n 2 V
(1 9 )

th ereb y lea d in g to V (r ) = ¡ Z e 2 = r w h ich , o f co u rse, w e

k n ow is ex a ct. (T h is is o n e o f th e m a n y cu rio sities in
th e C o u lo m b p ro b lem w h ich w e w ill tu rn to in a fu tu re
in sta llm en t.)
W e w ill n ow d escrib e a n o th er in terestin g fea tu re in q u a n -
tu m th eo ry rela ted to iso ch ro n o u s p o ten tia ls. It is w ell
k n ow n th a t w h en w e m ov e fro m cla ssica l to q u a n tu m
m ech a n ics, th e h a rm o n ic o scilla to r p o ten tia l lea d s to
eq u id ista n t en erg y lev els. C u rio u sly en o u g h , a ll th e iso ch -
ro n o u s p o ten tia ls h av e th is p ro p erty in the sem i-classical
lim it. T h is is m o st ea sily seen b y d i® eren tia tin g (1 5 )
w ith resp ect to E a n d u sin g (2 ) so a s to o b ta in
r Z x2
dn m dx T (E )
= p = : (2 0 )
dE 2~ x 1
2
E ¡ V ~

In o th er w o rd s, th e q u a n tu m n u m b ers a re g iv en b y th e
eq u iva len t fo rm u la
Z
1
n (E ) ' T (E )d E (2 1 )
~

In the case of the

w h ich n icely co m p lem en ts th e ¯ rst eq u a tio n in (1 5 ). If
potential in
th e p o ten tia l is iso ch ro n o u s, th en T (E ) = T 0 is a co n -
sta n t in d ep en d en t o f E a n d th e in teg ra l im m ed ia tely
equation (4),
g iv es th e lin ea r rela tio n b etw een E a n d n o f th e fo rm
something more
E = ® n + ¯ , w h ere ® = (~= T 0 ). C lea rly, th ese en erg y
surprising happens:
lev els a re eq u a lly sp a ced ju st as in th e ca se o f h a rm o n ic
The exact solution
o scilla to rs.
to the Schrödinger
equation itself has In th e ca se o f th e p o ten tia l in (4 ), so m eth in g m o re su r-
equally spaced p risin g h a p p en s: T h e ex a ct so lu tio n to th e S ch rÄo d in g er
energy levels! eq u a tio n itself h a s eq u a lly sp a ced en erg y lev els! I w ill

1006 RESONANCE  November 2008

 SERIES  ARTICLE

in d ica te b rie° y h ow th is a n a ly sis p ro ceed s leav in g th e

d eta ils fo r y o u to w o rk o u t. (Y o u ca n n o t rea ch th is
co n clu sio n b y th e tw o -d im en sio n a l trick u sed ea rlier in
cla ssica l p h y sics.) T o b eg in w ith , w e ca n red e¯ n e th e
p o ten tia l to th e fo rm
· ¸2
B
V (x ) = A x ¡ ; A 2 ´ a; B 2 ´ b (2 2 )
x

b y a d d in g a co n sta n t so th a t th e m in im u m va lu e o f th e
p o ten tia l is zero a t x = (B =A )1 = 2 . T h e freq u en cy o f
o scilla tio n s in th is p o ten tia l is ! 0 = (8 a = m )1 = 2 . T o
stu d y th e S ch rÄo d in g er eq u a tio n fo r th e p o ten tia l in (2 2 ),
it is co n v en ien t to in tro d u ce th e u su a l d im en sio n less
va ria b les » = (m ! 0 = ~)1 = 2 x ; ² = 2 E = (~! 0 ) a n d ¯ =
B (2 m )1 = 2 = ~, in term s o f w h ich th e S ch rÄo d in g er eq u a tio n
ta k es th e fo rm :
" µ ¶2 #
1 ¯
à 00+ ² ¡ »¡ à = 0: (2 3 )
2 »

A s » ! 1 , th e ¯ = » term b eco m es n eg lig ib le a n d { a s

in th e ca se o f th e sta n d a rd h a rm o n ic o scilla to r { th e
w av efu n ctio n s w ill d ie a s ex p [¡ (1 = 4 )» 2 ]. N ea r th e o ri-
g in , th e S ch rÄo d in g er eq u a tio n ca n b e a p p rox im a ted a s
» 2 à 00 ¼ ¯ 2 à w h ich h a s so lu tio n s o f th e fo rm à / » s w ith
s b ein g th e p o sitiv e ro o t o f s(s ¡ 1 ) = ¯ 2 . W e n ow fo l-
low th e sta n d a rd p ro ced u re a n d w rite th e w av efu n ctio n
in th e fo rm à = Á (» )[» s ex p (¡ (1 = 4 )» 2 )] a n d lo o k fo r a
p ow er-law ex p a n sio n fo r Á o f th e fo rm
X1
Á (» ) = cn » n : (2 4 )
n= 0

S u b stitu tin g th is fo rm in to th e S ch rÄo d in g er eq u a tio n w ill

lea d , a fter so m e a lg eb ra , to th e recu rren ce rela tio n

cn + 2 n + s ¡ ² ¡ ¯ + (1 = 2 )
= : (2 5 )
cn (n + 2 )(n + 2 s + 1 )

RESONANCE  November 2008 1007

 SERIES  ARTICLE

Do all isochronous A sy m p to tica lly, th is w ill lea d to th e b eh av io u r c n + 2 = c n '

potentials lead to
(1 = n ) so th a t Á (» ) ' ex p [(1 = 2 )» 2 ] m a k in g à d iv erg e
evenly spaced
u n less th e series term in a tes. S o , ² m u st b e so ch o sen
th a t th e n u m era to r o f (2 5 ) va n ish es fo r so m e va lu e o f n .
energy levels as
C lea rly, o n ly ev en p ow ers o f » a p p ea r in Á (» ) a llow in g
exact solutions to
u s to w rite n = 2 k , w h ere k is a n in teg er. P u ttin g ev ery -
Schrödinger
th in g b a ck , th e en erg y o f th e k -th lev el ca n b e w ritten
equation rather
in th e fo rm
than only in the " µ ¶1 = 2 #
asymptotic limit? 1 1
E k = (k + C )~! 0 ; C = 1 ¡ ¯ + ¯2 +
2 4
(2 6 )

sh ow in g th a t th e en erg y lev els a re eq u a lly sp a ced w ith

th e w id th ~! 0 b u t w ith C rep la cin g (1 = 2 ) in th e ca se o f
th e h a rm o n ic o scilla to r. Y o u ca n co n v in ce y o u rself th a t
a ll th e lim itin g b eh av io u r is co rrectly rep ro d u ced .
D o a ll iso ch ro n o u s p o ten tia ls lea d to ev en ly sp a ced en -
erg y lev els a s ex a ct so lu tio n s to S ch rÄo d in g er eq u a tio n
ra th er th a n o n ly in th e a sy m p to tic lim it? T h e a n sw er
is \ n o " . T h e sim p le co u n ter-ex a m p le is p rov id ed b y tw o
p a ra b o lic w ells co n n ected to g eth er sm o o th ly a t th e m in -
im a w ith V (x ) = (1 = 2 )m ! R2 x 2 fo r x ¸ 0 a n d V (x ) =
(1 = 2 )m ! L2 x 2 fo r x · 0 . It is o b v io u s th a t th is p o ten -
tia l is iso ch ro n o u s cla ssica lly. S o lv in g th e S ch rÄo d in g er
eq u a tio n req u ires a b it o f e® o rt b eca u se y o u n eed to en -
su re co n tin u ity o f à a n d à 0 a t th e o rig in . T h is lea d s
to a set o f en erg y lev els w h ich n eed to b e so lv ed fo r
n u m erica lly. O n e ¯ n d s th a t th e en erg y lev els a re n o t
Address for Correspondence
eq u a lly sp a ced b u t th e d ep a rtu re fro m ev en sp a cin g is
T Padmanabhan su rp risin g ly sm a ll. T o th e ex ten t I k n ow , th ere is n o sim -
IUCAA, Post Bag 4 p le ch a ra cteriza tio n o f p o ten tia ls w h ich lea d to ev en ly
Pune University Campus sp a ced en erg y lev els in q u a n tu m th eo ry.
Ganeshkhind
Pune 411 007, India. Suggested Reading
Email:
[email protected] [1] A B Pippard, The Physics of Vibration, Omnibus Edition, Cambridge
[email protected] University Press, p.15, 1989.
[2] L D Landau and E M Lifshitz, Mechanics, Pergammon Press, 1960.

1008 RESONANCE  November 2008