Solar Energy Materials & Solar Cells 81 (2004) 363–369

XRD study of the grain growth in CdTe films

annealedat different temperatures

$

Joel Pantoja Enr

!ıquez, Xavier Mathew*

Centro de Investigaci

!on en Energ!ıa-UNAM, 62580 Temixco, Morelos, Mexico

Abstract

The CdTe thin films electrodeposited on stainless steel substrates were annealed in air at

various temperatures and time durations in order to investigate the influence of post-
deposition heat treatments on the grain growth of the films. The recrystallization process at
lower annealing temperature is different from that of the high-temperature annealing. The
annealing at lower temperature promotes better grain growth by maintaining the preference
for the (1 1 1) plane. In general the grain size increases due to annealing and the
recrystallization happens in three phases. The grain growth exponent is a function of
temperature andtime. In the beginning of the annealing, irrespective of the annealing
temperatures the grain growth obeys the ideal parabolic law and for longer annealing times it
deviates from the ideal case.
r

2003 Elsevier B.V. All rights reserved.

Keywords: CdTe; Re-crystallization; Post-deposition treatments; Grain growth

1. Introduction

In CdTe-based solar cells one of the key steps in optimizing the photovoltaic

parameters is the post-deposition heat treatments of the CdTe film. There are reports
about the effect of the post-deposition heat treatments on the structural changes of
CdTe thin films and its influence on the performance of CdTe–CdS solar cells

[1–5]

.

It is known that the annealing of CdTe film in air or in CdCl

2

environment has a

significant effect on morphology of the films, promoting grain growth or sometimes

ARTICLE IN PRESS

$

This work is part of the Ph.D. Thesis of J.P. Enriquez at the Universidad Nacional Aut

!onoma de

Mexico (UNAM).

*Corresponding author. Tel.: [phone]; fax: [phone].

E-mail address:

[e-mail] (X. Mathew).

0927-0248/$ - see front matter r 2003 Elsevier B.V. All rights reserved.
doi:10.1016/j.solmat.2003.11.012

disintegration of grains and reducing internal stress

[1,3]

. It was foundthat in the

case of CdTe films developed on CdS layers, the grain size increases by 6–14 times
due to annealing. Further it was observed that the re-crystallization influenced the
preferredorientation of the crystallites.

A detailed study of the effect of post-deposition heat treatment on the

microstructure of the electrodeposited CdTe was reported by Bin Qi et al.

[5]

. It

was observedthat the re-crystallization process has two steps, in the first stage the
crystallites loose the preferred(1 1 1) orientation followedby a secondre-crystal-
lization process, which once again arranges the crystallites along the (1 1 1) plane due
to a longer annealing time. In this article we are presenting a study of the influence of
the post-deposition annealing on the grain size of the electrodeposited CdTe films on
stainless steel (SS) substrates. XRD measurements were performedto investigate the
effects of annealing on the re-crystallization andthe grain growth.

2. Experimental

The potentiostatic electro deposition of the CdTe films from acidic baths is

described elsewhere

[6–9]

. The films were deposited at a potential of –580 mV

with respect to an Ag/AgCl reference electrode. The counter electrode was a
pure platinum wire and0.05-mm thick SS foil was usedas the working electrode.
The thickness of all the CdTe films used in this study was about 1.4 mm. The films
were annealedin the temperature range from 300

C to 450

C at different time

intervals ranging from 0 to 60 min. The samples were annealedin air without any
chemical treatments. The XRD data over a 2y range of 20–90

were collectedusing a

Rigaku X-ray diffractometer with CuK

a

radiation of wavelength 1.54056 (

A. The

grain size was calculatedusing the full-wid

th at half-maximum (FWHM) of the

(1 1 1) peak.

3. Results and discussion

The X-ray diffraction patterns of the as deposited and annealed CdTe films are

shown in

Figs. 1 and2

. The films were annealedin air at 350

C and450

C for

different time intervals between 5 and60 min. The spectra were obtainedby scanning
y=2y in the range of 20–90

. The CdTe samples have a cubic zinc-blend structure and

the as-deposited CdTe films exhibit a strong preferred orientation along the (1 1 1)
planes parallel to the substrate. For the samples annealedat 450

C the formation of

oxides of cadmium and tellurium such as CdO, TeO

x

andCdTe

y

O

x

is evident as we

can see from the spectra (

Fig. 2

). In the case of samples annealedat 350

C and

400

C the oxide peaks were not observed.

Fig. 3

shows the variation of integral intensity of (3 1 1) peak with annealing time

for the samples annealedat 350

C, 400

C and450

C. The large angle (2y)

diffraction gives more details of the structural changes and hence we selected the
(3 1 1) peak at 2y ¼ 62

for the analysis. For the samples annealedat 400

C and

ARTICLE IN PRESS

J.P. Enr

!ıquez, X. Mathew / Solar Energy Materials & Solar Cells 81 (2004) 363–369

364

450

C the intensity increases first, reach a maximum value, andthen decreases with

annealing time. This behavior suggests that the orientation of lattice planes tendto
random initially and then again to (1 1 1) direction. The integral intensity of the
sample annealedat 350

C does not achieve a maximum indicating that the

recrystallization process at 350

C is different from that at 400

C and450

C. The

annealing at lower temperature may be promoting better grain growth by
maintaining the preference for the (1 1 1) plane as we can see in the discussions

ARTICLE IN PRESS

20

40

60

80

3 5 0

°

C

SS

SS

6 0 m in

3 0 m in

1 5 m in

1 0 m in

5 m in

V irg in

CdT

e

(440)

C

d

T

e

(5

11

)/(3

33)

Cd

Te

(4

2

2

)

Cd

Te

(3

3

1

)

CdT

e

(311)

Cd

Te

(2

2

0

)

CdT

e(111)

Intensity (a.u)

2

θ

(deg.)

Fig. 1. XRD patterns of the virgin andthe annealedCdTe thin films. The films were annealedin air at
350

C for 5, 10, 15, 30, and60 min.

20

40

60

80

450

°

C

5min

CdT

e(

422)

SS

SS

CdO

CdO

CdTe

2

O

5

CdTe(

331)

CdTe(400)

CdTe(440)

CdTe(311)

CdTe(220)

CdTe(111)

60min

30min

15min

10min

Intensity (a.u)

2

θ

(deg.)

Fig. 2. XRD patterns of the CdTe thin films annealed in air at 450

C for 5, 10, 15, 30, and60 min.

J.P. Enr

!ıquez, X. Mathew / Solar Energy Materials & Solar Cells 81 (2004) 363–369

365

below. Even though there are changes in the intensity of the various peaks due to
annealing, the samples continue to show a preference for the (1 1 1) direction in each
stage of the annealing process. Many authors have reportedchanges in XRD
patterns due to annealing in different environments

[1–5]

.

The average diameter of the CdTe grains was calculated using the Scherrer

relation

[10,11]

:

D ¼

Kl

b cos y

;

ð1Þ

where y is the Bragg angle, l ¼ 1:54056 (

A is the CuK

a

radiation, D is the

average diameter of the grains, and K is the shape factor which is approximately
unity and

b

2

¼ ðFWHMÞ

2

 b

2

:

ð2Þ

The integral breadth b was obtainedfrom a powder sample of polycrystalline silicon.

Fig. 4

shows the variation of grain size with the annealing time andtemperature. As

one can see from

Fig. 4

there are three phases for the re-crystallization, in the first

phase the grain growth occurs irrespective of the annealing temperature andit takes
place between 5 and15 min of the annealing time. In this interval the grain size
reaches a maximum. The secondphase is between 15 and30 min of annealing
duration where the grain size slowly decreases. Beyond 30 min the grain size remains
more or less constant with a tendency to slightly increase in size. This behavior is

ARTICLE IN PRESS

0

10

20

30

40

50

60

7

8

9

10

11

12

13

14

15

16

Virgin samples

I

311

(%)

Annealing time (min.)

350

°

C

400

°

C

450

°

C

Fig. 3. The variation of the integral intensity of the peak (3 1 1) with annealing time. The films were
annealedin air at 350

C, 400

C and450

C for different annealing times. The markers are experimental

data and the line is a guide to the eye.

J.P. Enr

!ıquez, X. Mathew / Solar Energy Materials & Solar Cells 81 (2004) 363–369

366

different from the grain growth observed in CdTe electrodeposited on CdS/SnO

2

/

glass structures reportedin literature in which the grains show a tendency to grow
throughout the annealing duration

[5]

. One can see from

Fig. 4

that the grain size as

well as the growth rate is better for lower annealing temperature. This is supported
by the observation in

Fig. 3

where the integral intensity of the (3 1 1) peak of the

sample annealedat 350

C do not attains a maximum indicating that the

recrystallization was always along the (1 1 1) plane. This can be due to the fact
that at lower temperatures, the formation of oxide layers is negligible and also
the stress induced due to the thermal mismatch between CdTe and SS substrate is
also less.

In order to compare the time dependent grain growth to the ideal case described

by the parabolic grain growth law

[12]

:

ðD

2

 D

2
0

Þ

1=2

¼ At

n

;

ð3Þ

where D

0

and D are the difference of the average grain sizes before and after the

annealing, t is the annealing time, A is a constant and n is the grain growth exponent
(the ideal value of n above the half-melting temperature is 0.5). A plot of the n
against annealing time is shown in

Fig. 5

. As we can see in

Fig. 5

the n value is a

function of annealing time, in general decrease with time. In an ideal grain growth
obeying the parabolic law, the value of n is 0.5. It can be seen from

Fig. 5

that in the

beginning of the annealing, irrespective of the annealing temperature the grain
growth obeys the ideal parabolic law and for longer annealing time it deviates from
the ideal case.

ARTICLE IN PRESS

0

10

20

30

40

50

60

0.10

0.12

0.14

0.16

0.18

0.20

0.22

0.24

0.26

0.28

Annealing time (minutes)

D (

µ

m)

350

°

C

400

°

C

450

°

C

Fig. 4. The variation of average grain size with annealing time. The data for the films annealed at 350

C,

400

C and450

C are presented. The markers are experimental data and the line is a guide to the eye.

J.P. Enr

!ıquez, X. Mathew / Solar Energy Materials & Solar Cells 81 (2004) 363–369

367

4. Conclusion

The grain growth mechanisms in a CdTe film electrodeposited on stain less steel

substrates during post-deposition annealing have been investigated using XRD
analysis. The as-deposited films have a cubic zinc-blend structure with a strong
preferredorientation along (1 1 1) direction. The recrystallization process at lower
annealing temperature is different from the high-temperature annealing. The
annealing at lower temperature promotes better grain growth by maintaining the
preference for the (1 1 1) plane. The grain growth has three phases, in the first phase
the grain size reaches a maximum, in the secondphase the grain size slowly decreases
andin the thirdphase the size remains more or less constant. In general, the grain
size increasedafter annealing. The grain growth exponent is a function of time, and
at the beginning of the annealing the grain growth obeys the ideal parabolic law.

Acknowledgements

This work was supportedby CONACYT andPAPIIT-UNAM through the projects

38542-U, G38618-U andIN115102-3, respectively. One of the authors (J.P. Enr

!ıquez)

acknowledges the grant received from DGIA-UNAM for the Ph.D. program.

References

[1] A. Romeo, D.L. Batzner, H. Zogg, A.N. Tiwari, Thin SolidFilms 361 (2000) 420.

ARTICLE IN PRESS

0

10

20

30

40

50

60

0.1

0.2

0.3

0.4

0.5

0.6

350

°

C

400

°

C

450

°

C

Grain growth exponent n

Time (min.)

Fig. 5. The graph shows the dependence of the grain growth exponent (n) on the annealing time of the
samples. The markers are experimental data and the line is a guide to the eye.

J.P. Enr

!ıquez, X. Mathew / Solar Energy Materials & Solar Cells 81 (2004) 363–369

368

[2] K.D. Rogers, J.D. Painter, D.W. Lane, M. Healy, J. Electron. Mater. 28 (1999) 112.
[3] L.R. Cruz, L.L. Kazmerski, H.R. Moutinho, F. Hasoon, R.G. Dhere, R. de Avillez, Thin Solid Films

350 (1999) 44.

[4] H.R. Moutinho, M.M. Al-Jassim, F.A. Abufoltuh, D.H. Levi, P.C. Dippo, R.G. Dhere, L.L.

Kazmerski, 26th IEEE Photovoltaic Specialists Conference, Anaheim, CA, 1997.

[5] B. Qi, D. Kim, D.L. Williamson, J.U. Trefny, J. Electrochem. Soc. 143 (1996) 517.
[6] G.P. Hernandez, X. Mathew, J.P. Enriquez, N.R. Mathews, P.J. Sebastian, Sol. Energy Mater. Sol.

Cells 70 (2001) 269.

[7] X. Mathew, P.J. Sebastian, A. Sanchez, J. Campos, Sol. Energy Mater. Sol. Cells 59 (1999) 99.
[8] X. Mathew, J. Mater. Sci. Lett. 21 (2002) 529.
[9] X. Mathew, J. Phys. D 33 (2000) 1565.

[10] E. Lifshin, X-Ray Characterization of Materials, Wiley-VCH, New York, 1999, p. 37.
[11] H.P. Klug, L.E. Alexander, X-Ray Diffraction Procedures for Polycrystalline and Amorphous

Material, Wiley, New York, 1954, p. 491.

[12] D.A. Porter, K.E. Easterling, Phase Transformations in Metals andAlloys, VNR, New York, 1981,

p. 110.

ARTICLE IN PRESS

J.P. Enr

!ıquez, X. Mathew / Solar Energy Materials & Solar Cells 81 (2004) 363–369

369