Pseudo moiré dislocations appearing in xray diffraction topography

J. Yoshimura

Citation: J. Appl. Phys. 80, 2138 (1996); doi: 10.1063/1.363828

View online: http://dx.doi.org/10.1063/1.363828
View Table of Contents: http://jap.aip.org/resource/1/JAPIAU/v80/i4
Published by the American Institute of Physics.

Related Articles
Critical shell thickness for InAs-AlxIn1−xAs(P) core-shell nanowires
J. Appl. Phys. 112, 124305 (2012)
A visualization of threading dislocations formation and dynamics in mosaic growth of GaN-based light emitting
diode epitaxial layers on (0001) sapphire
Appl. Phys. Lett. 101, 231911 (2012)
Role of the Ge surface during the end of range dissolution
Appl. Phys. Lett. 101, 162103 (2012)
The origin of 0.78eV line of the dislocation related luminescence in silicon
J. Appl. Phys. 112, 063528 (2012)
Growth regimes during homoepitaxial growth of GaN by ammonia molecular beam epitaxy
J. Appl. Phys. 112, 054903 (2012)

Additional information on J. Appl. Phys.

Journal Homepage: http://jap.aip.org/
Journal Information: http://jap.aip.org/about/about_the_journal
Top downloads: http://jap.aip.org/features/most_downloaded
Information for Authors: http://jap.aip.org/authors

Downloaded 28 Dec 2012 to 139.184.30.132. Redistribution subject to AIP license or copyright; see http://jap.aip.org/about/rights_and_permissions
Pseudo moiré dislocations appearing in x-ray diffraction topography
J. Yoshimuraa)
Institute of Inorganic Synthesis, Faculty of Engineering, Yamanashi University, 4-3-11 Takeda, Kofu 400,
Japan
~Received 26 January 1996; accepted for publication 14 May 1996!
In recent x-ray moiré topographic experiments a dislocationlike discontinuity of moiré fringes has
been found despite the fact that specimens ~Si bicrystal! were dislocation free. This discontinuity,
although similar in appearance to the fringe discontinuity known as moiré dislocation, should be
essentially distinguished from it. Preliminary considerations suggest that the outbreak of such
pseudo-moiré dislocations is related not only to the wave-field phase due to the moiré effect, but to
the phase of extinction fringes. Pseudo-moiré dislocations commonly occur in plane-wave x-ray
moiré topography of slightly strained specimens. © 1996 American Institute of Physics.
@S0021-8979~96!04616-6#

I. INTRODUCTION mass difference between the two. A minute pulling force by

a load L 0.0–0.3 g was applied further to the rear component
It is well known that a crystal-lattice moiré pattern can
crystal so as to rotate it about the the goniometer axis ~y
appear arising from the interference of electron or x-ray
axis!. Conventional-type high-resolution x-ray films ~Fuji no.
waves diffracted successively by two crystal plates.1–11
50! were used. Exposure time was 25–35 s. A rocking curve
When either of the two crystals contains dislocations, a dis-
measured from a 10.0 ~x direction!30.3 mm2 ~y direction!
locationlike discontinuity called moiré dislocation appears
area on the specimen exhibited a half-width @full width at
correspondingly in the moiré pattern.3,4,10,11 In recent plane-
half-maximum ~FWHM!# of 1.61 arcsec against the calcu-
wave x-ray topographic experiments using dislocation-free
lated width for the unstrained state of 1.44 arcsec. This
Si crystal devices,12–17 we, however, have found that similar
shows that the specimen was free of heavy strain but a weak
dislocationlike discontinuities appear in the moiré patterns
strain that gives an angular broadening ;0.2 arcsec was
despite the complete absence of dislocations. This new-kind
present ~see Ref. 17 for more details of the experiment!.
fringe discontinuity is of interest from the viewpoint of
x-ray-diffraction physics, as well as significant for the
characterization of crystals. This article describes the sum-
mary of experimental facts with preliminary considerations
on the origin of the discontinuity.

II. EXPERIMENT
The setup of the main experiment for the present topic is
shown in Fig. 1. The experiment was conducted using syn-
chrotron radiation at Station BL-15C,18 at the Photon Fac-
tory, KEK, Japan. X rays from the synchrotron source were
monochromatized and collimated by a Si ~111! premono-
chromator and an asymmetrically cut Si~220! collimator
~asymmetry factor b51/40!. The wavelength of the incident
x rays onto the specimen was centered at l050.72 Å with the
spread Dl/l0>1023 and the angular spread was 0.34 arcsec.
The specimen was a monolithic Si bicrystal. It was set in the
parallel setting with the upstream Si~220! collimator so as to
give rise to the plane-wave diffraction images of it. The an-
gular spread above is much narrower than the Darwin width FIG. 1. ~a! General view of the experimental setup. S: specimen; P: thin Pt
2.15 arcsec for Si~220!. The specimen bicrystal had a narrow wires stretched in a rigid frame; F: x-ray films; L~1! and L~2!: load for
causing rotation about the y axis @~1,2! give the sign of the load#. x axis i
gap between the two component crystals, and a cut perpen-
@11̄0#, y axis i @112̄#, z axis i @111#. The (x,z) plane lies in the vertical plane
dicular to the crystal surfaces at its bottom. The thicknesses with the x axis inclined by 2.4° from the vertical axis. The Pt wires were to
of the front and rear components and the gap were 1.615, make a reference for position and/or orientation on the topographs by their
1.510, and 0.225 mm, respectively. Observed moiré patterns shadow images. Exposure to many films was for another purpose of the
experiment. ~b! Schematic illustration of the ray geometry in the specimen
were mainly caused by a relative rotation between the two bicrystal. The O and G waves emerging from surface a 8 are superposed on
component crystals about an axis ~z axis! normal to the crys- surface b with a mutual displacement of 86 mm. ~c! Detail drawing of the
tal surface, which occurred unforcedly because of a slight specimen. Dimensions are given in mm. Hatching shows portions where
adhesive ~Araldite! was painted to fix the pulling wires, and to attach a
balancer for adjusting the mass difference between the two component crys-
a!
Electronic mail: [email protected] tals.

2138 J. Appl. Phys. 80 (4), 15 August 1996 0021-8979/96/80(4)/2138/4/$10.00 © 1996 American Institute of Physics

Downloaded 28 Dec 2012 to 139.184.30.132. Redistribution subject to AIP license or copyright; see http://jap.aip.org/about/rights_and_permissions
FIG. 2. ~a! Moiré topographs taken when the pulling load is L50.0 g; O image. ~b! The same as ~a!, but G image. ~c! L520.2 g; O image. In ~a!–~c! the
diffraction vector g2 ¯2 0 is from left- to right-handside. The coordinates (X,Y ) of the O and G images are so graduated as to refer to the same position on each
specimen. The coordinates of the G images show the real length on the films, with the Y axis parallel to @112̄#. The crossing black lines are the shadows of
thin Pt wires @see Fig. 1~a!#. ~d! Taken by a similar but separate setup ~see Ref. 19! for plane-wave topography, using a conventional x-ray source and another
bicrystal specimen ~the thicknesses of the front and rear crystals and the gap were 1.56,1.41, and 0.26 mm, respectively!. Si 220 reflection, Mo K a radiation.
O image. The moiré pattern was caused by lattice expansion induced by heating the top of the rear crystal. In all topographs ~a!–~d! white contrast indicates
stronger intensity.

III. RESULTS ing 0.4–0.5 mm corresponding to the relative lattice-plane

rotation 4–531027 rad. Dislocationlike discontinuities are
Figure 2 presents x-ray topographs illustrating discon-
observed at sites noted by arrows. Figure 2~d! illustrates the
tinuous moiré fringes. The topographs in Figs. 2~a!–2~c! in-
clude a pair of transmitted- and diffracted-beam images ~O case of a nearly parallel moiré pattern caused mainly by a
and G images, respectively! taken simultaneously from the lattice-spacing difference between two compared crystals.
same specimen, and another O image taken when a minute The fringe spacing is 0.6 ~near the top! to 0.8 mm ~near the
force L was applied to that specimen. These moiré patterns bottom!, which corresponds to the lattice-spacing difference
are close to a so-called rotation moiré, with the fringe spac- Dd/d52–331027. On the way from the top to bottom,

J. Appl. Phys., Vol. 80, No. 4, 15 August 1996 J. Yoshimura 2139

Downloaded 28 Dec 2012 to 139.184.30.132. Redistribution subject to AIP license or copyright; see http://jap.aip.org/about/rights_and_permissions
 ~3! Like real dislocations and true moiré dislocations,
PMDs are generated with either positive or negative senses
~Fig. 3!. As well as single PMDs, seemingly paired PMDs
with the opposite senses are observed @e.g., O4 and O5 in
Fig. 3~a!; O6 and O7 in Fig. 3~b!#.
~4! As shown by the examples in Figs. 2~a!–2~d!, PMDs
are generally encountered in any type of moiré pattern when
topographs of a nonuniformly strained specimen are taken
under nearly plane-wave conditions. Although not evidenced
here, PMDs occur also in moiré patterns produced with an
x-ray interferometer.
~5! Besides the abnormality of PMD, the outbreak of an
extra subfringe or the split of one regular fringe occurs in
several sites. A clear example can be seen in a circled area in
Fig. 2~b! ~see also Fig. 4 below!. Although less clear, similar
subfringes appear also at other sites and in other topographs.
Seemingly, a subfringe is produced so as to join two main
fringes present on both of its sides. This is called bridging.
The bridging could be regarded as a kind of paired PMDs
whose members are located most closely each other.

IV. DISCUSSION
The generation of PMDs is not explained satisfactorily at
this stage of study. The difference in the PMD maps @Figs.
3~a! and 3~b!# between the O and G images demonstrates
that what matters with PMDs is the total phase difference of
interference, and the formation of PMDs is a problem of
phase distribution. The wave-field intensities for the present
moiré topographs of the O and G images can be given by
I o ~ r! 5I oo 8 1I go 8 12 AI oo 8 I go 8 cos C o ~ r! , ~1a!

I g ~ r! 5I og 8 1I gg 8 12 AI og 8 I gg 8 cos C g ~ r! , ~1b!
with
FIG. 3. Plots of fringe discontinuities observed in topographs. ~a! Case of C o ~ r! 5 f M1 f PO1 f gap , ~2a!
Figs. 2~a! and 2~b!; ~b! case of Fig. 2~c!. Letters O and G refer to the O and
G images. C g ~ r! 5 f M1 f PG1 f gap . ~2b!
~for details see Ref. 17!. Here I oo 8 , I go 8 and I og 8 , I gg 8 denote
fringes between the two marking lines increase by two, with the intensities of interfering partial wave fields in the O and
two outbreaks of a discontinuous new fringe. G beams @see Fig. 1~b!#. fM in Eqs. ~2a! and ~2b! is given by
The following features are remarkable on these discon- fM52p@~gB 2gA !i•rb #, representing the phase difference due
tinuities: to the moiré effect; gA and gB are the reciprocal lattice vec-
~1! Comparison of Figs. 2~a! and 2~b! @see also Fig. 3~a!# tors of the two component crystals, and rb is a position vec-
shows that discontinuities appear at mutually displaced sites tor lying on the entrance surface of the rear crystal; the sym-
and in different numbers between the paired O and G im- bol i denotes the parallel component to the crystal surface.
ages. In addition, the discontinuity sites in the O and G fPO and fPG are the phases made by the extinction action of
images seem to have a weak partial correlation, not distrib- the crystal, and are connected complicatedly with the diffrac-
uted independently ~e.g., O2 and G2, and O4 and G4 in Fig. tion intensities I oo 8 , I go 8 , etc. fgap is given by
3~b!#. fgap52Kt gap Du sin u B / g g , representing an additional phase
~2! Comparison of Figs. 2~a! and 2~c! @see also Figs. 3~a! difference made by an interspacing gap in the bicrystal; here
and 3~b!# shows that the distribution of discontinuities K means the wave number in vacuum, t gap the gap width, Du
changes completely with the application of a minute force. If the deviation angle from the exact Bragg position, uB the
the observed discontinuities were true moiré dislocations, Bragg angle, and gg the direction cosine between the surface
they should appear at the same fixed sites irrespective of the normal and the G beam direction.
difference between the O and G images and of the applied When the specimen has a nonuniform strain aside from
force. The observations ~1! and ~2! make it clear that the the lattice mismatch giving rise to fM , intensities I oo 8 , I go8,
discontinuities are not caused by dislocations in the real lat- etc., and phases fPO , fPG , and fgap all get variable with
tice. We call them pseudo-moiré dislocations, ~PMDs!. position, and the fringe pattern made by fM is modulated by

2140 J. Appl. Phys., Vol. 80, No. 4, 15 August 1996 J. Yoshimura

Downloaded 28 Dec 2012 to 139.184.30.132. Redistribution subject to AIP license or copyright; see http://jap.aip.org/about/rights_and_permissions
 comes from that between fPO and fPG , since fM and fgap
are common in the O and G images according to Eq. ~2!.
The change of the PMD distribution with a minute force
@from Fig. 2~a! to Fig. 2~c!# also seems to be connected with
fPO , which in general changes sensitively with strain. Thus,
what seems certain at present is that the extinction phases
fPO and fPG take part importantly in the generation of
PMDs. Of interest is the question of how sharp fringe dis-
continuities can be created while dislocations or other local-
ized strains responsible for the localized phase jump do not
exist in the real crystal. Satisfactory explanation must wait
for the detailed analysis of the diffraction optics concerned.

ACKNOWLEDGMENTS
The synchrotron experiment was conducted under the
approval of the Photon Factory Advisory Committee ~Nos.
88-086 and 90-113!. The work was supported by a Grant-in-
Aid for Scientific Research from the Ministry of Education,
FIG. 4. Plots of moiré fringes in Fig. 2~a! ~dotted line: O image! and Fig.
2~b! ~solid line: G image!. Fringe lines of the two images are plotted in the Science, Sports, and Culture ~Contract No. 02452283!.
same coordinate plane as they were on the exit surface of the specimen. The
positions of fringe lines were determined as the midpoint on the fringe- 1
T. Mitsuishi, H. Nagasaki, and R. Uyeda, Proc. Jpn. Acad. 27, 86 ~1951!.
profile width at 0.7 times the peak height measured from a background 2
Y. Seki, J. Phys. Soc. Jpn. 8, 149 ~1953!.
curve ~see Ref. 17!. In blank areas such as the ones around PMDs, plotting 3
H. Hashimoto and R. Uyeda, Acta Crystallogr. 10, 143 ~1957!.
fringe positions is omitted because of ambiguous fringe patterns. The blank 4
D. W. Pashley, J. W. Menter, and G. A. Bassett, Nature 179, 752 ~1957!.
D is an area where fringe positions are indefinite because of a point like 5
U. Bonse and M. Hart, Appl. Phys. Lett. 6, 155 ~1965!.
defect image. Thin solid lines joining thick lines show bridging ~only those 6
J. Chikawa, Appl. Phys. Lett. 7, 193 ~1965!.
in the G image are indicated for simplicity!. 7
A. R. Lang and V. F. Miuscov, Appl. Phys. Lett. 7, 214 ~1965!.
8
U. Bonse and M. Hart, Z. Phys. 190, 455 ~1966!.
9
D. Simon and A. Authier, Acta Crystallogr. A 24, 527 ~1968!.
fPO , fPG , and fgap . Fringe lines in the moiré topographs in 10
A. R. Lang, Nature 220, 652 ~1968!.
Figs. 2~a! and 2~b! are replotted in Fig. 4 to make the moiré
11
M. Hart, Philos. Mag. 26, 821 ~1972!.
12
J. Yoshimura, Acta Crystallogr. A 43, C221 ~1987!.
patterns clearer as such a phase pattern. The plotted lines 13
J. Yoshimura, J. Phys. Soc. Jpn. 58, 1283 ~1989!.
may be understood to approximate the patterns of 14
J. Yoshimura and T. Ishikawa, Acta Crystallogr. A 46, C424 ~1990!.
Co ~r!52 p n and Cg ~r!52 p n, where n is an integer. As J. Yoshimura, Acta Crystallogr. A 47, 139 ~1991!.
15
16
J. Yoshimura, J. Crystallogr. Soc. Jpn. 34, 19 ~1992! ~in Japanese!.
seen from Fig. 4, PMDs occur at sites where the local fringe 17
J. Yoshimura, Acta Crystallogr. A 52, 312 ~1996!.
patterns of Co ~r!52 p n and Cg ~r!52 p n greatly disagree 18
T. Ishikawa, Rev. Sci. Instrum. 60, 2490 ~1989!.
with one another. The difference between Co ~r! and Cg ~r! 19
J. Yoshimura, Phys. Status Solidi A 125, 429 ~1991!.

J. Appl. Phys., Vol. 80, No. 4, 15 August 1996 J. Yoshimura 2141

Downloaded 28 Dec 2012 to 139.184.30.132. Redistribution subject to AIP license or copyright; see http://jap.aip.org/about/rights_and_permissions