Two-source coevaporation technique for synthesis of indium phosphide films with controlled composition

Article abstract of DOI:10.1016/j.jallcom.2012.03.119

Synthesis of indium phosphide films with controlled composition by coevaporation technique has been demonstrated. Films with three different In:P ratios were deposited to represent phosphorous rich, stoichiometric and phosphorous poor InP films. Microstructural and compositional studies indicated films to be polycrystalline in nature, with grain size and shape varied with In:P ratio in the films. X-ray diffraction pattern indicated reflections from (1 1 1), (2 2 0) and (3 1 1) planes of InP only. The surface roughness of the films estimated from AFM studies was found to be 30 nm. PL spectrum measured at 10 K was dominated by a strong peak located ～1.35 eV. Characteristic Raman peaks for InP at ～306 cm^-1 (TO) and ～341 cm^-1 (LO) were observed.

Full text of DOI:10.1016/j.jallcom.2012.03.119

Journal of Alloys and Compounds 531 (2012) 34–40
Contents lists available at SciVerse ScienceDirect
Journal of Alloys and Compounds
journal homepage: www.elsevier.com/locate/jallcom
Two-source coevaporation technique for synthesis of indium phosphide films
with controlled composition
R.N. Gayen, S. Hussain1, D. Ghosh, R. Bhar, A.K. Pal∗
Department of Instrumentation Science, USIC Building, Jadavpur University, Calcutta-700 032, India
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 1 March 2012
Received in revised form 26 March 2012
Accepted 30 March 2012
Available online 10 April 2012
Synthesis of indium phosphide films with controlled composition by coevaporation technique has been
demonstrated. Films with three different In:P ratios were deposited to represent phosphorous rich, stoi-
chiometric and phosphorous poor InP films. Microstructural and compositional studies indicated films to
be polycrystalline in nature, with grain size and shape varied with In:P ratio in the films. X-ray diffraction
pattern indicated reflections from (1 1 1), (2 2 0) and (3 1 1) planes of InP only. The surface roughness of the
films estimated from AFM studies was found to be 30 nm. PL spectrum measured at 10 K was dominated
by a strong peak located ∼1.35 eV. Characteristic Raman peaks for InP at ∼306 cm−1 (TO) and ∼341 cm−1
(LO) were observed.
PACS:
68.55.−a
78.66.Fd
81.15.Fg
78.20.−e
© 2012 Elsevier B.V. All rights reserved.
Keywords:
InP films
III–V semiconductor
Optical properties
1. Introduction
for synthesis of InP in polycrystalline form is the need of the day
to expand the use of this material in different device technologies.
High radiation resistance and potentiality for achieving high
efficiency makes indium phosphide a very promising photovoltaic
(PV) material for space solar cell applications [1–5]. The electro-
optical properties of InP along with its carrier recombination
mechanism indicate InP and related compound to be the best can-
didate for optoelectronic and microwave devices. But large scale
deployment of this material in PV industry is being deterred by the
dearth of process technology for obtaining the material in thin film
form by low-cost and scalable method. Among the different tech-
niques, reported so far for the deposition of InP films, liquid phase
epitaxy [6], sputtering [7], electrodeposition [8], spray pyrolysis
[9], flash evaporation [10] and organic chemical vapor deposition
[11] are worth mentioning. InP films were also deposited by three-
temperature method by Rusu [12] while InP deposited/grown in
different nanocrystalline forms (nanoparticles, nano wires, etc.)
have been reported by others [13–18]. But none of the above meth-
ods could effectively ensure large area deposition of InP films and
the need of a reliable, low-cost and scalable process technology
Recently, Gayen et al. [19] demonstrated that high quality InP could
be effectively deposited in polycrystalline form by co-evaporating
In and P. This study mainly focused the dependence of growth of InP
films on substrate temperature during deposition and the results
were concentrated on the studies of InP film having stoichiometric
composition. But for proper utilization of the InP layer in photo-
voltaic solar cell structure, one may need to vary the composition
of the adjacent layers constituting the depletion region. The quality
of the diode structure essentially depends on the carrier concen-
tration and thickness of the above adjacent layers. Thus, there is
a need for a complete and comprehensive study on InP polycrys-
talline films with different In:P ratios so that these films could be
fruitfully exploited in PV technology.
In the present paper, studies on InP films deposited on glass
substrates with controlled composition have been reported. The
structural properties such as crystallinity, strain, and degree of
disorder have been studied by a combination of X-ray diffraction
(XRD) and Raman spectroscopy. Raman spectroscopy gives use-
ful information about the solid on a scale of the order of a few
lattice constants, and can therefore be used to study the micro-
scopic nature of structural disorder. Here, we obtain a relationship
between the lattice disorder investigated by Raman spectroscopy
and the crystallite size obtained from XRD. Raman spectra have
∗
Corresponding author.
E-mail address: msakp2002@yahoo.co.in (A.K. Pal).
1
1UGC-DAE CSR, Kalpakkam Node, Kokilamedu-603104, India.
0925-8388/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jallcom.2012.03.119

R.N. Gayen et al. / Journal of Alloys and Compounds 531 (2012) 34–40
35
been analyzed within the purview of the spatial correlation model
to estimate the phonon correlation length which is an estimate of
the lattice disorder.
of InP reduced significantly. The XRD traces for phosphorous rich
films (Fig. 2(c)) resembled that for stoichiometric films with dif-
ferent intensity ratios of the peaks. The lattice constant (0.58 nm),
calculated from the XRD pattern, was in good agreement with that
reported earlier for InP bulk (0.5869 nm) with zinc blende structure
[21].
Line profile analysis (LPA) was adapted to characterize the above
films to assess the size–strain relationship and correlate them with
physical properties of the resulting materials. We have utilized the
Scherrer relation [22] to extract the size of the coherent diffract-
ing domains from diffraction peak. The “apparent” domain size, a
volume-weighted quantity, is given by:
2. Experimental
Co-evaporation technique adopted here is capable of tuning the composition
of InP films deposited on glass substrates. The films were deposited at a substrate
temperature of ∼423 K which was found to be the optimal deposition tempera-
ture (discussed later). The purity of the elemental components (In and red-P) was
∼99.99% as obtained from Sigma Aldrich. The system pressure during deposition
was better that 10−6 Torr. Two independent alumina crucibles, heated indirectly
by heavy duty tungsten basket heaters, were used to evaporate the materials at
the desired rate of evaporation. The rate of evaporation was monitored by a quartz
crystal oscillator and was calibrated against the current flowing through the indi-
vidual basket heaters used for evaporating In and P. Thus, the composition of the
films deposited by this procedure could be tuned easily. The complete deposition
technique has been reported in our earlier publication [19].
Compositional information was obtained by EDAX (Oxford Instruments’ INCA
Energy 250 Microanalysis System), which indicated that the films were nearly sto-
ichiometric. The surface morphology of the films was studied by Field Emission
Scanning Electron Microscope (FESEM) (Carl Zeiss SUPRA® 55 with GEMINI® Tech-
nology having a resolution ∼0.8 nm @ 15 kV). AFM pictures were recorded by using
a Nanosurf Easy Scan 2 in contact mode while X-Ray Diffraction (XRD) studies were
carried out by Rigaku MiniFlex using CuK␣ line (0.154 nm). Photoluminescence (PL)
measurements were recorded at 10 K by using a 300 W xenon arc lamp as the emis-
sion source. A Hamamatsu photomultiplier along with a 1/4 m monochromator was
used as the detecting system.
Kꢁ
t =
(1)
FWHM cos ꢀ
where K is a constant close to unity, FWHM corresponds to the
full width at half maximum of the peak and ꢀ the Bragg angle of
the [h k l] reflection. Another volume-weighted quantity defined by
Wilson [23] as an “integral breadth apparent size” is given by:
Kꢁ
D
v
=
(2)
ˇ cos ꢀ
where, the integral breadth ˇ = A/I0, A being the peak area and I0 the
height of the observed line profile. In both the relations, the peak
broadening was attributed to the effect of the diffracting coherent
domain size. When the broadening is solely due to the strain effect,
the following Stokes and Wilson relation [24] may be applied:
3. Results and discussion
3.1. Microstructural studies
ꢂ = ˇ cot ꢀ
(3)
(4)
Indium phosphide is known to have a congruent temperature
of ∼630 K beyond which phosphorous tends to ooze out of InP.
We have observed this experimentally and found that InP films
deposited at substrate temperature ∼630 K produced phosphorous
deficient films while films deposited at significantly lower temper-
atures resulted in phosphorous rich composition with amorphous
characteristics. We have analyzed the composition of these InP
films by EDAX and confirmed the optimal substrate temperature
for deposition to be ∼423 K. The results discussed below are based
on films deposited at ∼423 K.
ꢂ
ˇ
e =
=
4
4 tan ꢀ
where ꢂ is the “apparent” strain and e, the maximum strain. In all
practical cases, line broadening occurs due to simultaneous size and
lattice distortion effects. To such cases, Williamson–Hall plot [25]
may conveniently be utilized to separate these two effects:
Kꢁ
ˇ =
+ 4e tan ꢀ
cos ꢀ
(5)
D
v
In this technique, plot of ˇ for all the reflections of the sample
are expressed in terms of reciprocal unit (ˇ* = ˇ cos ꢀ/ꢁ) and as a
function of d* (d* = 2 sin ꢀ/ꢁ).
Fig. 1(a)–(c) shows the SEM picture of three representative InP
films deposited with three different In:P ratios such that three sets
of In rich, stoichiometric and phosphorous rich InP films could be
obtained. The films were deposited under identical experimental
conditions discussed in the previous section with the only variant
being indium flux arriving at the substrates. It may be noted here
that all the films appear to be compact in nature and the size and
shape depended critically on the composition of the films. Stoi-
chiometric films had oblong grains (Fig. 1(b)) while phosphorous
rich films consisted of nearly spherical grains (Fig. 1(c)) with nar-
rower size distribution. In contrast to the above, indium rich films
consisted of varied shape and sizes of grains.
Fig. 2(a)–(c) shows the X-ray diffraction patterns for the above
three InP films. The film with stoichiometric composition (Fig. 2(b))
indicated a strong peak at 2ꢀ = 26.5◦ for reflections from (1 1 1)
planes of InP alongwith reflections from peaks for (2 2 0) and (3 1 1)
planes located at 43.8◦ and 51.7◦ [20] of InP, respectively. Contrary
to the above, the indium rich films showed (Fig. 2(a)) a stronger
peak for reflections from (2 0 2) plane of InP located at 2ꢀ = 33◦
while the intensities of peaks for reflections from (2 2 0) and (3 1 1)
ꢀ
ꢁ
ˇ cos ꢀ
ꢁ
K
2 sin ꢀ
ꢁ
=
+ 2e
(6)
D
v
It is apparent that the plots of ˇ* versus d* would be a straight
line and the intercept would give the particle size (K/D ) while the
v
slope would give the strain (Fig. 2(d)). The particle size and the
strain values obtained from the above calculations are shown in
Table 1. It is observed that the stoichiometric films had largest
crystallite size (∼57 nm) while the phosphorous rich films had
the lowest grain size of ∼40 nm. It could be seen that the crys-
tallite size determined from the above calculations is lower than
that obtained from the SEM studies. This is basically due to con-
glomeration of such crystallites forming polycrystallites which are
apparently visible in the SEM micrograph. The strain value obtained
from the above calculations indicated that residual strain in the
film was quite low for indium rich and stoichiometric films. The
films became more strained with increased phosphorous content
Table 1
Values of thickness, band gap, crystallite size, strain and other parameters of the InP films deposited here.
Sample name
Thickness (nm)
Crystallite size(nm)
Strain (×10−4
)
Band Gap (eV)
ε␣
ωp(×1013) (sec−1
)
In-rich
Stoichiometric
P-rich
1550
1220
1160
53.0
57.1
40.7
7.35
8.70
13.85
1.34
1.42
1.47
11.2
10.1
9.9
9.297
8.984
8.975

36
R.N. Gayen et al. / Journal of Alloys and Compounds 531 (2012) 34–40
Fig. 1. SEM micrographs of three representative InP films: (a) In-rich, (b) stoichiometric and (c) P-rich.
140
120
100
80
100
50
0
(a)
(b)
60
40
20
0
20
30
40
50
60
20
30
40
50
60
2θ (degree)
2θ (degree)
0.018
0.016
0.014
0.012
0.010
0.008
0.006
60
40
20
0
(a)
(d)
(c)
−Δ−In rich
−Ο−P-rich
stoichiometric
3.0 3.5 4.0 4.5 5.0 5.5 6.0
20
30
40
50
60
β
2θ (degree)
Fig. 2. XRD patterns of the three representative InP films whose SEM pictures are shown in Fig. 1: (a) In-rich, (b) stoichiometric, (c) P-rich and (d) plots of ˇ* versus d* for
the InP films.

R.N. Gayen et al. / Journal of Alloys and Compounds 531 (2012) 34–40
37
4
3
2
1
0
gap ∼1.47 eV. The band gap values have been calculated considering
the overall uncertainty of 0.3% for transmittance measurements.
The band gap values obtained for the stoichiometric films agreed
well with that for bulk InP [21].
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
(a)
(b)
m=0.496
10.8
10.5
Simultaneous determination of the thickness (d) of the thin
film and its complex refractive index (ꢂ = n + ik, where n = refractive
index and k = extinction coefficient) is quite difficult experimen-
tally in most practical cases and especially, when the transmittance
spectra of a film do not indicate interference fringes needed for
the determination of the thickness. This would render the deter-
mination of the optical constants difficult and uncertain. We have
computed the values of film thickness (d), refractive index (n) and
extinction coefficient (k) of the films deposited here by using the
modified KK theory approach given by Xue et al. [27] by over-
coming the difficulties of simultaneous computation of the optical
parameters of transparent thin film along with its thickness [28].
The values for the real part of refractive index (n) and the extinc-
tion coefficient (k) were calculated and hence the real and complex
part of the dielectric constant (ε1 and ε2 respectively) were esti-
mated by using the relations:
-2.0 -1.5
ln (hν-Eg)
0.0
0.5
1.0
hν (eV)
1.5
2.0
0.4 0.6 0.8 1.0 1.2 1.4 1.6
hν (eV)
4
3
2
1
0
(c)
ε1 = n2(ꢁ) − k2(ꢁ)
ε2 = 2n(ꢁ)k(ꢁ)
(9)
(10)
It was observed (Fig. 4(a)) that the indium rich films are more
dense than both the stoichiometric and phosphorous rich films.
The refractive indices (n) were found to vary (Fig. 4(a)) from 3.52
and 3.16 within the measured wavelength range of 1000–2000 nm
for the films deposited for this study. These values tallied quite
well with the reported value of 3.54 for the bulk InP [21]. Variation
of extinction coefficient (k) with wavelength is shown in Fig. 4(b).
Extinction coefficient of indium rich films was larger than those for
stoichiometricand phosphorousrich films and all of them increased
near the band gap energy.
Variations of ε1 and ε2 with ꢁ for the InP films are shown in
Fig. 4(c) and (d), respectively. Values of both the ε1 and ε2 obtained
here are greater for the phosphorous rich films. The variation of ε1
with incident photon energy depends on the plasma frequency ωp
and can be expressed as [28,29]:
0.0 0.3 0.6 0.9 1.2 1.5 1.8
hν (eV)
Fig. 3. Plots of (˛hꢃ)2 versus hꢃ for representative films: (a) In-rich, (b) stoichio-
metric, (c) P-rich. Inset of (b) shows the plot of ln(˛hꢃ) versus ln(hꢃ − Eg).
in the films. Thus peeling-off of the films with subsequent anneal-
ing required for formation of heterojunction with stoichiometric
composition would be minimal.
3.2. Optical measurements
ε
ω2
ω2
∞
Transmittance and reflectance spectra of all the InP films were
recorded in the wavelength range of 400–2300 nm (not shown
here) .The absorption coefficient (˛) was calculated from them and
determined by using the method described in references [25,26].
The absorption coefficient (˛) may be written as a function of the
incident photon energy (hꢃ) as:
p
ε1 = ε
−
(11)
∞
where, ε is the limiting value of the high frequency dielectric
∞
constant. Fig. 4(e) shows the variation of ε1 with inverse of the
square of the frequency (1/ω2) of incident radiation for all the InP
films studied here. The value of ωp and ε were determined from
∞
the slope and the intercept of the linear portion of the ε1 versus
A
m
˛ =
{hv − Eg}
(7)
1/ω2 plots (Fig. 4(e)) and are shown in Table 1. The value of ε
hv
∞
determined as above for the InP films varied between 9.9 and 11.2.
where, A is a constant which is different for different transitions
indicated by different values of m and Eg is the corresponding band
gap. Eq. (7) may be rewritten as:
The value of plasma frequency, ωp, obtained as above was higher
(∼9.3 × 1013
s
−1) for indium rich films than that for both the stoi-
chiometric and phosphorous rich films (∼8.9 × 1013
s
−1).
d(ln ˛hꢃ)
d(hꢃ)
m
In a semiconductor, the carrier concentration (N) varies accord-
ing to the square of plasma frequency (ωp) and we have [28,30,31]:
=
(8)
hꢃ − Eg
4ꢄNe2
Eq. (8) indicates that a plot of d[ln(˛hꢃ)]/d[hꢃ] versus hꢃ would
show a divergence at hꢃ = Eg, from which a rough estimate of Eg
could be obtained. Thus, by using Eq. (8), the value of m can eas-
ily be computed from the slope of the plot (inset of Fig. 3(b)) of
ln(˛hꢃ) versus ln(hꢃ − Eg) plots. The value of m, obtained as above,
for InP films deposited here varied between 0.5 and 0.6, which indi-
cated allowed direct transition to be present in the InP films. The
band gap was determined by extrapolating the linear portion of
the plot of (˛hꢃ)2 versus hꢃ (Fig. 3(a)–(c)). It may be observed that
the band gap for the indium rich films (Fig. 3(a)) was significantly
lower (∼1.34 eV) than that for stoichiometric films (Fig. 3(b) and
Table 1) while the phosphorous rich films indicated a higher band
ωp2
=
(12)
me∗ε
∞
Thus, if N is known, the effective mass of charge carriers, me*, could
be determined from (p and vice versa. Now, we know that the band
gap (Eg) may be related to the effective mass as [32]:
1
m∗
p2
= 1 +
(13)
2m Eg
∗ , m is the free electron mass and p = ꢀG, G being
m
m
where m∗
=
e
the smallest reciprocal lattice vector. Also, p is related to ‘a’ as
p ∼ h/a, ‘a’ being the lattice constant. Thus, using the experimental

38
R.N. Gayen et al. / Journal of Alloys and Compounds 531 (2012) 34–40
3.52
3.44
3.36
3.28
3.20
3.12
0.18
P-rich
(a)
(b)
P-rich
In-rich
0.15
0.12
0.09
0.06
0.03
In-rich
Stoichiometric
Stoichiometric
800 1000 1200 1400 1600 1800 2000
1000 1200 1400 1600 1800 2000
λ (nm)
λ (nm)
1.2
12.5
(c)
P-rich
(d)
P-rich
In-rich
In-rich
12.0
11.5
11.0
10.5
10.0
0.9
0.6
0.3
0.0
Stoichiometric
Stoichiometric
1000 1200 1400 1600 1800 2000
800 1000 1200 1400 1600 1800 2000
λ (nm)
λ (nm)
(e)
11.0
P-rich
In-rich
10.5
10.0
9.5
Stoichiometric
6
7
8
9
10
11
12
1/ω2 (x10-31. sec-2)
Fig. 4. (a) Plots of n versus ꢁ, (b) plots of k versus ꢁ, (c) plots of ε1 versus ꢁ, (d) plots of ε2 versus ꢁ and (e) ε1 versus 1/ω2 for the representative films.
values of Eg and the bulk value of the lattice constant (a), the value
of me* was estimated to be ∼0.183 me. Using the above values of ωp
and me*, the value of carrier concentration was estimated from Eq.
as shallow donors. With the reduction of indium in the InP films
(for stoichiometric and phosphorous rich films), indium vacancies
(VIn) and/or phosphorous interstitials acting as shallow acceptors
would participate in the transition resulting in the reduction of the
intensity of the transition at ∼1.35 eV. It may be noted here that
Kim et al. [33] attributed this peak to DA pair transition involving
a relatively deep donor level (∼33 meV) appearing due to native
defects or related complexes. Temkin et al. [34] also indicated the
presence of such donor like phosphorous vacancies, acceptor like
indium vacancy and P interstitials in explaining the PL studies of
their system.
(12). Carrier concentration thus evaluated was 19.51 × 1024
m
−3 for
a representative InP film.
3.3. Photoluminescence studies
PL spectra recorded at 10 K for three representative InP films (In-
rich, stoichiometric and P-rich) with an excitation energy of 2.0 eV
are shown in Fig. 5(a)–(c). All the PL spectra were dominated by
a very strong peak located ∼1.35 eV followed by a low intensity
peak located ∼1.52 eV. The intensity of the peak ∼1.35 eV was sig-
nificantly larger for indium rich films ∼1.35 eV compared to those
for stoichiometric and phosphorous rich films. The observed peak
located at ∼1.52 eV could be related to band gap transitions for all
the films. It may be noted here that the details of the PL peak located
around the band gap energy was over shadowed by the presence
of the very strong peak located at ∼1.35 eV. The observed PL peak
located at ∼1.35 eV is due to DA transitions due to native defects.
If we take into consideration the strong intensity of this peak in
indium rich films, then this transition may be associated with phos-
phorous vacancies (VP) and associated indium interstitials acting
3.4. Raman studies
Raman spectra for three representative indium rich, stoichio-
metric and phosphorous rich InP films are shown in Fig. 6(a)–(c),
respectively. One may observe that all the Raman spectra are dom-
inated by the presence of two characteristic Raman peaks for InP
located at ∼306 cm−1 and 341 cm−1. The peak at ∼306 cm−1 could
be identified as arising due to TO phonon scattering, which is sym-
metry forbidden for the ([0 0 1]||[0 0 1]) configuration of InP only.
The other peak at ∼342 cm−1 could be attributed to LO phonon
scattering. According to the selection rules for the backscattering

R.N. Gayen et al. / Journal of Alloys and Compounds 531 (2012) 34–40
39
70
0.08
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0.00
-0.01
916.5 nm
(a)
(a)
60
50
40
30
20
10
0
814 nm
200
300
400
500
600
-1
Raman shift (cm )
750
800
850
900
950
1000
100
λ (nm)
(b)
80
60
40
20
0
918.1 nm
0.025
0.020
0.015
0.010
0.005
0.000
(b)
200
400
600
Raman shift (cm-1)
816.03 nm
100
80
(c)
750
800
850
900
950
1000
λ (nm)
918.8 nm
0.025
0.020
0.015
0.010
0.005
0.000
(c)
60
40
200
400
600
Raman shift (cm-1)
Fig. 6. Raman spectra of three representative InP films: (a) In-rich, (b) stoichiometric
and (c) P-rich.
817.29 nm
4. Conclusions
750 800 850 900
950 1000
One can easily tune in the composition of the films deposited by
this coevaporation technique described here. Films obtained were
of high quality polycrystalline nature suitable for device applica-
tions. The grain size in the films was found to vary between 40 nm
and 57 nm. The residual strain of the stoichiometric films was sig-
nificantly lower than that in indium rich films. Band gap of the
stoichiometric films was ∼1.42 eV. PL spectra for all the films were
dominated by a strong peak at ∼1.35 eV followed by a low intensity
peak ∼1.52 eV which was ascribed to be bound excitonic in nature.
XRD patterns indicated peaks at 2ꢀ ∼ 26.50◦, 43.8◦ and 51.7◦ for
reflections from (1 1 1), (2 2 0) and (3 1 1) planes, respectively for
zinc blende structure. Raman peaks for InP located at ∼306 cm−1
and 341 cm−1 could be identified as arising due to TO phonon and
LO phonon scattering, respectively, were also observed.
λ (nm)
Fig. 5. PL spectra of three representative InP films recorded at 10 K: (a) In-rich, (b)
stoichiometric and (c) P-rich.
geometry of crystals with zincblende structure, the incident light
is expected to be scattered by LO phonons if the crystal surface
is (1 0 0) and/or (1 1 1) oriented. The phonon energies are in good
agreement with the phonon modes found in the literature for InP
single crystals (303.7 0.3 and 345.0 0.3 cm−1 for TO and LO
peaks, respectively [21]). It may be noted here that these values
tallied extremely well with the values obtained by Mooradian and
Wright [35] for single crystalline InP. Raman lines at 304 cm−1 and
345 cm−1 obtained by Perna et al. [36] for polycrystalline InP films
are also close to the values obtained by us. It may be noted that
both the TO and LO phonon peaks showed shift in frequency as
the composition of the film changed. The TO peak shifted to higher
wave number while the LO peak shifted to lower wave number
as the film composition changed from indium rich to indium poor
condition.
Acknowledgment
The authors wish to thank the Department of Science and
Technology, Government of India for the financial assistance for
executing this programme. One of us (RNG) wishes to thank Coal
S&T, Government of India, for granting him fellowship. The authors

40
R.N. Gayen et al. / Journal of Alloys and Compounds 531 (2012) 34–40
would also like to thank Dr. G. Amarendra, Scientist-in-charge,
UGC-DAE CSR, Kalpakkam Node, Kalpakkam for his kind support.
[17] S. Adam, D.V. Talapin, H. Borchert, et al., Journal of Chemical Physics 123 (2005)
084706.
[18] J. Guo, H. Huang, X. Ren, X. Yan, et al., Journal of Vacuum Science and Technology
B 29 (2011) 030603.
References
[19] R.N. Gayen, S. Hussain, R. Bhar, A.K. Pal, Vacuum 1 (2012).
[20] Joint Committee on Powder Diffraction-International Centre for Diffraction
Data (JCPDS-ICDD, 1997) Card no. 32-0452.
[21] Bornstein L, Numerical Data and Functional Relationships in Science and Tech-
nology New Series Group III 17a (Springer Verlag, Berlin, 1982), p. 281.
[22] P. Scherrer, Nachr Ges Wiss Göttingen 26 (1918) 98.
[23] Wilson A.C.J., Mathematical Theory of X-ray Powder Diffractometry (Philips,
Eindhoven, 1963).
[24] A.R. Stokes, A.C.J. Wilson, Proceedings of the Physical Society London 56 (1944)
174.
[25] G.K. Williamson, W.H. Hall, Acta Metallurgica 1 (1953) 22.
[26] J.S. Pan, A.T.S. Wee, C.H.A. Huan, H.S. Tan, K.L. Tan, Journal of Physics D: Applied
Physics 29 (1996) 2997.
[27] S.W. Xue, X.T. Zu, W.G. Zheng, H.X. Deng, X. Xiang, Physica B 381 (2006) 209.
[28] S.R. Bhattacharyya, R.N. Gayen, R. Paul, A.K. Pal, Thin Solid Films 517 (2009)
5530.
[29] J.I. Pankove, Optical Processes in Semiconductors, Prentice-Hall Inc., NJ, USA,
1971, p. 89.
[30] J.C. Manifacier, M.D. Murcia, J.P. Fillard, E. Vicario, Thin Solid Films 41 (1977)
127.
[31] D.K. Ferry, Semiconductors, McMillan, New York, 1995, Ch. 5.
[32] D. Bhattacharya, S. Chaudhuri, A.K. Pal, Vacuum 46 (1995) 1.
[33] T.S. Kim, S.D. Lester, B.G. Streetman, Journal of Applied Physics 62 (1987)
1363.
[1] I. Weinberg, Solar Cells 29 (1990) 225.
[2] H.–J Schimper, Z. Kollonitsch, K. Möller, U. Seidel, et al., Journal of Crystal
Growth 287 (2006) 642.
[3] M. Yamagucuchi, Solar Energy Materials and Solar Cells 68 (2001) 31.
[4] M.W. Wanlass, T.A. Gessert, K.A. Emery, T.J. Coutts, Proceedings 20th IEEE Pho-
tovoltaic Specialists’ Conference, Las Vegas, 1988, pp. 491–496.
[5] X. Li, M.W. Wanlass, T. Gessert, K.A. Emery, T.J. Coutts, Applied Physics Letters
54 (1989) 2674.
[6] K.Y. Choi, C.C. Shen, B.I. Miller, Proceedings 19th IEEE Photovoltaic Specialists’
Conference, New York, 1987, pp. 255–260.
[7] P. Mitra, P. Gupta, S. Chaudhuri, A.K. Pal, Materials Chemistry and Physics 38
(1994) 81.
[8] S.N. Sahu, A. Bourdilon, Physica Status Solidi (a) 111 (1989) K179.
[9] M. Öztas, M. Bedir, R. Kayah, F. Aksoy, Journal of Materials Science - Materials
in Electronics 17 (2006) 841.
[10] R.N. Gayen, S.N. Das, S. Dalui, R. Paul, R. Bhar, A.K. Pal, Thin Solid Films 518
(2010) 3595.
[11] N. Shahid, S. Naureen, M.Y. Li, M. Swillo, S. Anand, Journal of Vacuum Science
and Technology B 29 (2011) 031202.
[12] G.I. Rusu, Physica Status Solidi (a) 4 (1971) 755.
[13] M. Mattila, T. Hakkarainen, H. Jiang, E.I. Kauppinen, H. Lipsanen, Nanotechnol-
ogy 18 (2007) 155301.
[14] X. Duan, Y. Huang, Y. Cui, J. Wang, C.M. Lieber, Nature 409 (2001) 66.
[15] J. Jasinski, V.J. Leppert, Si-Ty. Lam, G.A. Gibson, et al., Solid State Communica-
tions 114 (2007) 624.
[16] K.W. Jun, P.K. Khanna, K.B. Hong, et al., Materials Chemistry and Physics 96
(2006) 494.
[34] H. Temkin, B.V. Dutt, W.A. Bonner, Applied Physics Letters 38 (1981) 431.
[35] A. Mooradian, G.B. Wright, Solid State Communications 4 (1966) 431.
[36] G. Perna, V. Capozzi, V. Augelli, et al., Semiconductor Science and Technology
16 (2001) 377.