Thin Solid Films 476 (2005) 231 – 236
www.elsevier.com/locate/tsf
Studies on micro-structural and electrical properties of spray-deposited
fluorine-doped tin oxide thin films from low-cost precursor
E. Elangovan*, K. Ramamurthi
Department of Physics, Bharathidasan University, Tiruchirappalli-620 024, India
Received 15 August 2003; received in revised form 10 March 2004; accepted 11 September 2004
Available online 27 October 2004
Abstract
Thin films of fluorine-doped tin oxide (SnO2:F) on glass were prepared by spray pyrolysis technique using stannous chloride (SnCl2) and
ammonium fluoride (NH4F) as precursors. The as-prepared films were characterized for their structural and electrical properties and are
discussed in detail in this article. The surface morphology studies revealed that the films are grainy and the roughness of undoped films has
been reduced on fluorine doping. X-ray diffraction (XRD) studies revealed that the films are polycrystalline. It further revealed that the
undoped films grow along the preferred orientation of (211), whereas all the doped films grow along (200). The minimum sheet resistance
1.75 V/5 achieved in the present study for the films doped with 15 wt.% F is the lowest among the reported values for these materials
prepared using SnCl2 precursor. The electrical transport phenomenon has been analyzed in order to find out the possible scattering
mechanism that limiting the mobility of charge carriers.
D 2004 Elsevier B.V. All rights reserved.
PACS: 78.20.e; 78.68.+m; 78.40.q; 68.55.a
Keywords: Spray pyrolysis; Tin oxide; Fluorine doping; Electrical properties; Doped oxides
1. Introduction
Recently, studies on thin films of transparent conducting
oxide (TCO) semiconductors have attracted the attention of
many researchers because of their wide range of applications in science and technology [1–3]. These materials are
very efficient in reflecting broadband infrared heat radiation
in a manner similar to highly conducting metal-like
materials and in transmitting the light in the visible region
as if they are insulators. Such spectrally selective coatings
have wide applications in solar thermal energy and solar
photovoltaic conversions, solar heating, window insulation,
and thermal insulation in lamps [3]. Among the available
TCOs, highly transparent and conducting fluorine-doped tin
* Corresponding author. Present address: Materials Research Centre,
Indian Institute of Science, Bangalore-560 012, India. Tel.: +91 80 2293
2782; mobile: +91 98861 67687; fax: +91 80 2360 0683.
E-mail addresses: [email protected], [email protected]
(E. Elangovan).
0040-6090/$ - see front matter D 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.tsf.2004.09.022
oxide (SnO2:F) films are promising candidates for solar
thermal energy conversion [4]. The incorporation of fluorine
in SnO2 films results in enhancing the electrical properties
without declining the optical transmittance [5].
Therefore, production of SnO2:F layers with high
conductivity becomes essential. Though thin films of
SnO2:F were convincingly prepared using various thin film
deposition techniques such as chemical vapour deposition
[6], reactive sputtering [7], evaporation [8], and spray
pyrolysis [9], the techniques such as reactive sputtering,
and evaporation need vacuum environment and thus become
quite expensive when large-scale production is in demand.
The films produced by chemical vapour deposition may
have good quality but the requirement of high purity and
huge amount of precursors for large area coatings makes
these techniques rather expensive. Spray pyrolysis has a
simple and inexpensive experimental arrangement and has
the advantages like ease of adding doping material,
reproducibility, high growth rate, and mass production
capability for uniform large area coatings, which are
232
E. Elangovan, K. Ramamurthi / Thin Solid Films 476 (2005) 231–236
desirable for industrial and solar cell applications. In this
context, the present study is aimed at preparing SnO2:F
films using inexpensive spray pyrolysis technique. Quite a
lot of different materials are reported in the existing
literature as suitable spray precursors for both tin and
fluorine. For example, SnCl4 [10] or SnCl2 [1] is used as the
source of tin, whereas NH4F [11], HF [12] or Freon [13] is
used as the source of fluorine. Instead, the use of tin (II)
fluoride as the source of both tin and fluorine is also
reported [14]. A thorough understanding of literature survey
showed that the reports on SnO2:F thin films using SnCl2
precursor [15] is very scarce. The SnCl2 and NH4F
precursors are very cost-effective when compared to other
precursors reported for spray pyrolysis of these films.
Taking all these acute points into consideration, it is targeted
to prepare low cost–high conducting SnO2:F films by spray
pyrolysis technique using SnCl2 and NH4F (0–30 wt.% in
spray solution) precursors. Further, the structural and
electrical properties of these films were carried out to
understand the effect of fluorine doping.
2. Experimental details
Thin films of SnO2:F on glass were prepared using a
homemade spray pyrolysis experimental setup [16]. Stannous chloride (SnCl2) was used as the source for tin. The
fluorine doping was achieved using ammonium fluoride
(NH4F). The fluorine doping concentration was varied
from 0 to 30 wt.% in the spray solution. Microscopic glass
slides (75251.4 mm3) were used as substrates. The
substrates were cleaned using distilled water and various
organic solvents such as methanol, ethanol, and trichloroethylene. The substrate temperature was fixed at 400 8C.
Films of ~1.2 Am thick were grown. The spray solution
was prepared in the following manner. SnCl2 (11 g) was
dissolved in 5 ml of concentrated hydrochloric acid by
heating at 90 8C for 10 min. The resultant transparent
solution was then diluted with methanol formed the
starting solution. For fluorine doping, NH4F dissolved in
double-distilled water was added to the starting solution.
The amount of NH4F was varied to achieve different [F]/
[Sn] weight ratio in the spray solution. The deposition time
was 40 min for all the depositions. The carrier gas flow
rate was maintained at 6 l/min at a pressure of 6.50104 N
m2. The normalized distance between the spray nozzle
and the substrate is 35 cm. The spray time was maintained
at b1 s and the spray interval at ~3 min. For each
concentration, several sets of films were prepared and
found to be reproducible. Phase formation of the films was
studied by X-ray diffraction (XRD) system using Cu-Ka
radiation. The surface morphology of the films, crystallites
size and distribution were examined by scanning electron
microscopy (SEM). Roughness of the films was measured
by atomic force microscopy (AFM). The electrical properties such as sheet resistance, carrier concentration and
mobility of charge carriers were measured using Hall
measurements setup in van der Pauw configuration.
3. Results and discussion
3.1. Structural properties
The XRD patterns obtained for SnO2 thin films deposited
at 400 8C are shown in Fig. 1. The XRD studies clearly
reveal that all the as-deposited films are polycrystalline in
nature that was confirmed after matching the XRD profile
with JCPDS card no. 41-1445 [17]. It is perceptible from the
figure that the undoped films grow along the preferred
orientation of (211) whereas all the doped films grow along
(200). This has been reflected in our SEM studies, where the
doped and undoped films have different grain shapes.
Presence of other (hkl) reflections such as (110), (101),
(310), (301), and (321) have also been detected with lower
intensities. Further, the intensity of the (200) plane increased
with increasing fluorine doping concentration initially but
then decreased for further increase in F doping. The high
intensity of (200) reflection is observed for the films doped
with 15 wt.% of NH4F. This increase in intensity leads to
better crystallinity of the films.
As-grown films were uniform and free from pinholes.
These films were well adherent to substrate as examined by
the adhesive tape peel test. The thickness of the films was
Fig. 1. XRD pattern of SnO2:F films: (a) 0 wt.%, (b) 5 wt.%, (c) 10 wt.%,
(d) 15 wt.%, (e) 20 wt.%, (f) 25 wt.%, and (g) 30 wt.% of fluorine in the
spraying solution.
E. Elangovan, K. Ramamurthi / Thin Solid Films 476 (2005) 231–236
estimated from the cross-sectional SEM micrographs and
found to be ~1.2 Am. The SEM micrographs obtained on the
surface of SnO2:F films are shown in Fig. 2 for different
fluorine doping concentrations. The micrographs indicate
that the crystallites are well formed and densely packed.
Undoped films have well-faceted grains with large crystallite size (~500 nm). Whereas the doped films comprise of
largely distributed needle shaped crystallites (~150 nm)
together with the well-faceted crystallites. The distribution
of needle shaped crystallites in the doped films is random
and such grains are seen more in number at higher fluorine
doping concentration (N20 wt.%). These observations are
well corroborated with the different orientation of the grains
as observed from XRD studies. For undoped films, the
faceted grains (observed by SEM) give rise to high intense
(211) reflection. Whereas for the fluorine-doped films, the
needle shaped grains (observed by SEM) strongly orient
along (200) reflection [13].
To examine the roughness and microstructure of the films
in more detail, AFM was employed. The 2D topographic
views of AFM images obtained for films without and with
15 wt.% of fluorine in spray solution are shown in Fig. 3a
and b, respectively. The root mean square (RMS) roughness
extracted from the AFM data was found to be 86.53 nm for
the undoped film and 24.76 nm for fluorine-doped films.
This clearly indicates that the surface of the undoped films is
rougher than that of film doped with fluorine. It is
perceptible from the figures that the smoothness of the
undoped film has been increased on fluorine doping. This
can be attributed to the fact that the undoped films have
larger crystallites than doped films as seen in the SEM
study. Thus the AFM study corroborates SEM findings.
It has been reported earlier that the films prepared from
SnCl2 contain a disordered growth [18]. But the films
Fig. 2. SEM micrographs of SnO2 films obtained for different fluorine
concentrations in the spraying solution; (a) 0 wt.%, (b) 5 wt.%, (c) 10 wt.%,
(d) 15 wt.%, (e) 20 wt.%, and (f) 25 wt.%, respectively.
233
Fig. 3. 2D topographic view of AFM images of SnO2 films: (a) without and
(b) 15 wt.% fluorine in the spraying solution.
prepared in the present study showed no disorder in growth
and showed the preferential growth along (200). The
preferred growth of (200) for doped films remained
predominant irrespective of the fluorine doping level. The
most conspicuous feature from the XRD analysis is that the
films in the present study also oriented along (200) plane.
Gorodillo et al. [19] reported that the SnO2 films prepared
using SnCl2 showed a tendency to grow preferentially along
the (101), (211) and (301) directions. They have analyzed
the precursor chemistry and growth rate for the deposition
of SnO2 films from SnCl4 and SnCl2. The exact reason for
the orientation along (200) plane may be sought from the
differences in the preparation of starting solutions.
Smith et al. [18] have analyzed the relation between
solution chemistry and morphology of SnO2 films for with
and without the addition of HCl in SnCl2d 2H2O solution. It
was reported that if HCl was added with the starting solution
the films were highly oriented along (200) plane. This can
be ascribed to the formation of different intermediate
molecules in the starting solution. Though SnCl2d 2H2O
can partly ionize into Sn2+ and Cl, it could also form tinbased polymer molecules [20]. Addition of HCl that resulted
in transparent solution, during the experiment, may be due
to the breakdown of these polymer molecules. The effect of
association between SnCl2d 2H2O and Cl or H+ has to be
considered for further explanation. SnCl2d 2H2O is known
to react with HCl to give HSnCl3. This neutral HSnCl3 is
unstable and highly reactive [21]. At the pyrolysis temperature, HSnCl3 is thermally decomposed to form the hydrated
234
E. Elangovan, K. Ramamurthi / Thin Solid Films 476 (2005) 231–236
SnO2 molecule. The hydrated SnO2 molecule is also formed
when SnCl4 is dissolved in methanol. This leads to an
exothermic reaction in which part of the SnCl4 in methanol
ionizes into Sn4+ and Cl and the rest form complexes like
SnCl4d 2CH3OH and SnCl4d 5CH3OH [22] by reacting with
alcohol. These two complexes are neutral molecules and
thus can react in the pyrolysis region to form hydrated SnO2
on the substrate. It is now clear that irrespective of precursor
and additives, the similar hydrated SnO2 molecule is formed
in both the process. The starting solution in the present
study was prepared by dissolving SnCl2d 2H2O in HCl at 90
8C. Further, the SEM micrograph analysis shows the film
morphology is similar to that of obtained with SnCl4 [18].
Hence, it is not a revelation that the films are highly oriented
along (200) plane.
3.2. Electrical properties
Sheet resistance (R sh) is a useful parameter in comparing
thin films, particularly those of the same material deposited
under different conditions. The variation of sheet resistance
and electrical resistivity of SnO2:F films with different
fluorine doping is plotted in Fig. 4. It is found that the sheet
resistance of the pure tin oxide thin films (38.22 V/5)
decreases with increasing fluorine concentration initially
and reaches a minimum value (1.75 V/5 at 15 wt.% F)
afterwards increases on higher doping levels. The minimum
value of sheet resistance achieved in the present study is
lower than those reported earlier for these films prepared
from SnCl2 precursor. The reported values of sheet
resistance of SnO2:F films prepared by spray pyrolysis
from SnCl2 precursor are given in Table 1.
When fluorine is incorporated in tin oxide films, each F
anion substitutes an O2 anion in the lattice and the
substituted O2 anion introduces more free electrons [23].
This results in an increase in free electrons and decreases the
value of R sh. This can be attributed as the reason for
decreasing R sh with increasing fluorine doping. The increase
in the value of R sh beyond a certain doping concentration of
Fig. 4. Variation of sheet resistance and resistivity of SnO2:F films with
different fluorine concentrations in the spraying solution.
Table 1
Sheet resistance of SnO2:F films prepared from SnCl2 precursor
Film
NH4F
R sh (V/5)
Reference
SnO2:F
15 wt.%
20 wt.%
5 wt.%
10 wt.%
4.5 wt.%
0.7 mol%
1.75
3.18
4.61
5.65
16.40
30.00
Present work
Present work
Present work
[1]
[20]
[15]
fluorine probably represents a solubility limit of fluorine in
the tin oxide lattice. The excess F atoms do not occupy the
proper lattice positions to contribute to the free carrier
concentration, while at the same time increase the disorder
of the structure leading to an increase in sheet resistance.
This was observed in the films beyond 15 wt.% of fluorine
doping.
The Hall coefficient measurements were carried out on
the SnO2:F films as a function of fluorine doping. Hall
mobility measurements indicated that the films are n-type.
The variation of Hall mobility and carrier concentration of
SnO2:F thin films as a function of fluorine doping
concentrations is plotted in Fig. 5. The figure clearly reveals
that the Hall mobility of charge carriers is decreasing with
increasing fluorine concentration till 15 wt.% of fluorine
doping but then increases for higher doping concentrations.
On the other hand, the carrier concentration is increasing
with increase in fluorine doping initially and reaches a peak
value but then decreases with further increase in the fluorine
doping. The electrical measurements of the present work
suggest that the 15 wt.% of fluorine doping is the optimum
doping level for achieving better electrical properties.
The initial increase in carrier concentration (n) suggests
that the fluorine dopant substitutes oxygen [23]. This
substitution is decided by ionic size and charge of the
dopant. In the case of SnO2:F films, fluorine appears to be
the most favoured substituent because of the following
reasons: (i) its ionic size (F: 0.133 nm) very closely
matches with that of oxygen (O2: 0.132 nm), (ii) the
energy of the SnUF bond (~26.75 D8/kJ mol1) is
Fig. 5. Variation of Hall mobility and carrier concentration of SnO2:F thin
films with different fluorine concentrations in the spraying solution.
E. Elangovan, K. Ramamurthi / Thin Solid Films 476 (2005) 231–236
comparable to that of the SnUO bond (~31.05 D8/kJ mol1)
and (iii) since the charge on the fluorine ion is only half of
the charge of the oxygen ion, Coulomb forces that bind the
lattice together are reduced. Thus, geometrically, the lattice
is nearly unable to distinguish between fluorine and oxygen
ions. But this increase does not precede much further. The
reduction in carrier concentration for higher doping levels
suggests a probable interstitial incorporation of the dopant,
taking place in the SnO2 lattice. The corresponding peak
value of n has been obtained for the films prepared with 15
wt.% of NH4F.
The resistivity measurement in the temperature range of
30 to 200 8C revealed that the films are degenerate. The
variation of the resistivity as a function of temperature of the
films is comparatively shown in Fig. 6. It is perceptible from
the figure that both the undoped and doped films show no
variation in film resistivity and it is almost constant through
the entire temperature range applied in the present study.
This is the evident for the degeneracy of semiconducting
materials. The film degeneracy was further confirmed by
evaluating Fermi energy using the relation
EF ¼
h2
8m4
3n
p
2=3
ð1Þ
For the value of effective mass, the mean value that is
evaluated from plasma frequency for fluorine-doped SnO2
films, 0.19m e (m e=rest mass of electron), has been used
[16]. For pure SnO2 films, the reported value of 0.12m e
has been used [24]. The calculated E F values (around 1.5
eV) are very high compared to the energy corresponding
to the room temperature that is the evidence for
degenerate nature of materials. The electrical data
obtained for SnO2:F thin films are given in Table 2 as
a function of fluorine doping. Further, the E F values are
proportional to n 2/3 that is the characteristic for degeneracy of materials. Hence, it is established that the films are
degenerate semiconductors.
235
Table 2
Electrical properties of SnO2:F films as a function of fluorine concentrations in the spraying solution
[F]/[Sn]
(wt.%)
R sh
(V/5)
q
(104V cm)
l
(cm2/V s)
n
(1020 cm3)
EF
(eV)
0
5
10
15
20
25
30
38.22
4.61
3.47
1.75
3.18
4.32
5.68
46.25
5.58
4.20
2.12
3.85
5.23
6.87
109.5
26.8
24.3
22.1
24.7
26.0
29.0
0.12
4.18
6.12
13.30
6.58
4.59
3.13
0.10
1.09
1.41
2.36
1.48
1.16
0.90
3.3. Electrical transport phenomenon
In order to explore the transport phenomena in doped
semiconductors it is very indispensable to converse the kind
of scattering mechanism that determines the actual value of
the mobility of the carriers in these materials [25]. The
influence of grain boundary should also be considered for
polycrystalline thin film semiconductors. In the layers
produced by pyrolysis an ideal lattice cannot be expected,
even if no donor atoms are present. Hence, the scattering of
electrons by the thermal vibrations of the lattice atoms can
be omitted in the present case [26]. For tin oxide thin films,
it is commonly reported that either grain boundary scattering
[26–29] or ionized impurity scattering [1,23,30–33] as the
dominant scattering mechanism. Sometimes, it is also
reported as optical lattice scattering to be the dominant
mechanism [28,33,34]. The possibility of grain boundary
and ionized impurity scattering mechanisms as to be
dominant has been verified using the calculations available
in literature [1,25,26,29]. The parameters that were useful in
analyzing the possibility of different scattering mechanisms
are given in Table 3. It is obvious from the table that both
the grain boundary and impurity ion scatterings are not the
dominant mechanisms in limiting the mobility. This conclusion has been arrived since neither the condition for grain
boundary scattering nor the impurity ion scattering is
satisfied.
The present situation can be explained using the
following arguments. Scattering by grain boundaries can
be neglected to the first approximation since the grain size
is much greater than the mean free path (l). Another
Table 3
Mean free path, screening radius and mobility data of SnO2:F films as a
function of fluorine concentrations in the spraying solution
Fig. 6. Variation of resistivity as a function of temperature of SnO2:F thin
films for different fluorine concentrations in the spraying solution.
NH4F
(wt.%)
Crystallite
size L (2)
l (2)
0
5
10
15
20
25
30
366.69
256.30
412.20
412.45
322.14
378.35
241.32
51.6
40.9
42.2
49.7
43.8
41.0
40.3
R S (2)
6.29
2.77
2.60
2.28
2.57
2.73
2.91
Mobility (cm2/V sec)
lH
l cal
109.56
26.82
24.33
22.14
24.68
26.04
29.06
63.56
10.09
8.84
6.95
8.63
9.89
11.42
236
E. Elangovan, K. Ramamurthi / Thin Solid Films 476 (2005) 231–236
approach to the discussion of the grain boundary influence
is based on treating polycrystalline material as a two-phase
system [35]. This model states that the optical mobility is
not impacted by grain boundaries as long as the grain size
is much greater than the mean free path. The collision
frequency is determined by scattering processes occurring
in the bulk of a grain, and is a reciprocal of the bulk
relaxation/collision time, s. Thus l opt is a better measure
of real mobility in the grain bulk. If the l H and l opt values
are close to each other, that means the contribution of
grain boundary to the film resistance is negligible and if
they are not close to each other the contribution of grain
boundary is significant. Hence, the efforts are on to find
out the optical mobility and thus the effect of grain
boundary on mobility.
References
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
4. Conclusions
[13]
Polycrystalline thin films of SnO2 with different
fluorine doping concentrations were prepared by spray
pyrolysis technique. The undoped films showed the
preferential growth along (211) which has been found
shifted to (200) on fluorine doping. The AFM studies
showed that the fluorine-doped films has smoother surface.
The minimum sheet resistance achieved in the present
study is found to be the lowest among the reported values
for these films prepared from SnCl2 precursor. The detailed
analysis of electrical transport phenomenon revealed that
the generally reported grain boundary and impurity ion
scatterings are not the dominant scattering mechanisms for
these films. Further efforts are under progress to find out
the possible scattering mechanism that limiting the
mobility of the charge carriers of the films reported in
the present study.
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
Acknowledgements
[26]
The authors gratefully acknowledge S. Uthanna, Department of Physics, Sri Venkateswara University, for his help
in recording the optical spectra and Hall measurements. One
of the authors (E.E.) sincerely thanks S.A. Shivashankar,
Materials Research Centre, Indian Institute of Science for
extending the facilities of his laboratory. B. Thangaraju
(Wipro Technologies, Bangalore) is acknowledged for his
valuable suggestions in preparing this manuscript. The
authors are thankful to the refereeTs valuable suggestions
that were indeed useful in enhancing the quality of this
manuscript, TSF scientific staff for their inspiring comments/suggestions, and as well to the Editorial office for
their kind co-operation.
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
B. Thangaraju, Thin Solid Films 402 (2002) 71.
P.S. Patil, Mater. Chem. Phys. 59 (1999) 185.
A.L. Dawar, J.C. Joshi, J. Mater. Sci. 19 (1984) 1.
B. Gottlieb, R. Koropecki, R. Arce, R. Crisalle, J. Ferron, Thin Solid
Films 199 (1991) 13.
R. Asomoza, A. Maldonado, J. Rickards, E.P. Zironi, M.H. Farias, L.
Cota-Araiza, G. Soto, Thin Solid Films 203 (1991) 195.
H.L. Ma, D.H. Zhang, S.Z. Win, S.Y. Li, Y.P. Chen, Sol. Energy
Mater. Sol. Cells 40 (1996) 371.
G. Beensh-Marchwicka, L.K. Stephiewska, A. Misiuk, Thin Solid
Films 113 (1984) 215.
H. Demiryont, N. Tezey, Thin Solid Films 101 (1983) 345.
S.P.S. Arya, Cryst. Res. Technol. 21 (1996) K38.
G. Mavrodiev, M. Gajdardziska, N. Novkovski, Thin Solid Films 113
(1984) 93.
P. Cowache, D. Lincot, J. Vedel, J. Electrochem. Soc. 136 (1989)
1646.
A.E. Rakshani, Y. Makdisi, H.A. Ramazaniyan, J. Appl. Phys. 83
(1998) 1049.
D. Belanger, J.P. Dodelet, B.A. Lombos, J.I. Dickson, J. Electrochem.
Soc. 132 (1985) 1398.
G.C. Morris, A.E. McElnea, Appl. Surf. Sci. 92 (1996) 167.
D.R. Acosta, E.P. Zironi, E. Montoya, W. Estrada, Thin Solid Films
288 (1996) 1.
E. Elangovan, K. Ramamurthi, J. Optoelectron. Adv. Mater. 5 (2003)
45.
Joint Committee on Powder Diffraction Standards (JCPDS), International Centre for Diffraction Data, 1997, card no. 41-1445.
A. Smith, J.M. Laurent, D.S. Smith, J.P. Bonnet, R.R. Clemente, Thin
Solid Films 266 (1995) 20.
G. Gorodillo, L.C. Moreno, W. de la Cruz, P. Teheran, Thin Solid
Films 252 (1994) 61.
J.C. Guintini, W. Granier, J.V. Zanchetta, A. Taha, J. Mater. Sci. Lett.
9 (1990) 1983.
P. Pascal, Nouveau Traité de Chimie Minérale, vol. 8, Masson, Paris,
France, 1963, p. 313 and 369.
L. Gmelin, Hanbuch der Anorganische Chemie—Tin Part C5,
Springer-Verlag, New York, 1977, p. 68.
C. Agashe, S.S. Major, J. Mater. Sci. 31 (1996) 2965.
S. Shanthi, C. Subramanian, P. Ramasamy, Cryst. Res. Technol. 34
(1999) 1037.
V.I. Fistul, Heavily Doped Semiconductors, Plenum Press, New York,
1969, p. 86.
E. Shanthi, V. Dutta, A. Banerjee, K.L. Chopra, J. Appl. Phys. 51
(1980) 6243.
C. Erginsoy, Phys. Rev. 79 (1950) 1013.
E.H. Putley, The Hall Effect and Semiconductor Physics, Dover, New
York, 1968, p. 146.
E. Shanthi, A. Banerjee, V. Dutta, K.L. Chopra, J. Appl. Phys. 53
(1982) 1615.
G. Frank, E. Kaur, H. Kostlin, Sol. Energy Mater. 8 (1983) 387.
Z.M. Jarzebski, J.P. Marton, J. Electrochem. Soc. 123 (1976) 199C;
299C; 333C.
T. Arai, J. Phys. Soc. Jpn. 15 (1960) 916.
H. Haitjema, J.J. Ph. Elich, C.J. Hoogendoorn, Sol. Energy Mater. 18
(1989) 283.
P. Grosse, F.J. Schmitte, G. Frank, H. Kostlin, Thin Solid Films 90
(1982) 309.
A.S. Gilmore, A. Al-Kaoud, V. Kaydanov, T.R. Ohno, Mater. Res.
Soc. Symp. Proc. 666 (2001) F 3.10.1.