Chu 1988
Chu 1988
Chu 1988
Summary
1. Introduction
Thin film cadmium telluride heterojunction solar cells have been under
investigation for many years. A commonly used configuration is: ohmic
contact/p-CdTe/TCS/SnO2 :F/glass, where TCS is a transparent conducting
semiconductor or the window. Its photovoltaic characteristics depend on (i)
the band gap energy, electron affinity, electrical conductivity, lattice param-
eter and thermal expansion coefficient of the window material, (ii) the
resistivity of, and minority carrier diffusion length in, p-CdTe, (iii) the clean-
liness of the p-CdTe-TCS interface, and (iv) the resistance of the contact to
the p-CdTe. Major difficulties associated with the fabrication of efficient
CdTe solar cells include the deposition of low resistivity p-CdTe films and
the formation of stable low-resistance contact to p-CdTe films. In spite of
these difficulties, thin film CdS/CdTe solar cells with a conversion efficiency
of greater than 10% have been prepared by several authors [1 - 3].
In this work, a Zn0.3Cd0.TS film was used as the window material to
improve its optical transparency, the resistivity of p-CdTe films deposited by
the closed-spaced sublimation (CSS) process was controlled by using source
material of appropriate compositions, and a p-HgTe film was used as an
ohmic contact to p-CdTe films. Thin film CdTe solar cells with a conversion
efficiency of nearly 11% were prepared in this manner. The experimental
procedures and results are summarized in this paper.
5
E
u
Q
¢2)
4
¢.)
c~
w
1
I . . . ~ [
2.4 2.6 2.8 3.0 3.2
ENERGYOF PHOTONS, eV
Fig. 1. Optical a b s o r p t i o n o f a Z n x C d l _ x S film d e p o s i t e d o n a C o r n i n g 7 0 5 9 glass sub-
strate.
ited onto the surface of heated substrates by the direct recombination of the
vapor of the elements in a gas flow system, the resistivity of the film may be
controlled, to a limited extent, by intrinsic doping (using a cadmium-
deficient reaction mixture) or extrinsic doping (adding AsH3 or PH3 to the
reaction mixture), and p-CdTe films with a resistivity of about 200 ~ cm can
be deposited [7]. Similar techniques have been applied to the resistivity con-
trol of CSS CdTe films by using source materials of appropriate composition,
since the CSS process provides a direct transport of each c o m p o n e n t of the
source material to the substrate. The CSS technique involves several inter-
related parameters: the temperatures of the source and the substrate, the
separation between the source and the substrate, the nature of the atmo-
sphere and the pressure in the reaction tube, and the composition of the
source material. For example, the pressure in the reaction tube is important
in determining the rate of deposition, since the dissociation pressure of CdTe
depends exponentially on temperature. At a given source temperature, the
sublimation rate increases rapidly as the pressure in the reaction tube is
reduced from the atmospheric pressure. At too low pressures, e.g. 1 Torr,
the mean free path of the gaseous species in the reaction tube increases, and
the condensation process is no longer limited to the space between the
substrate and the source material. Thus, a compromise must be made be-
tween the source temperature, the gas pressure and the deposition rate. The
results of m a n y experiments suggest that a helium pressure of 30 Torr, a
source-to-substrate separation of 0.1 cm, a source temperature of 700 °C and
a substrate temperature of 600 °C are suitable for the CSS process. Thus, the
composition of the source material remains as the only important parameter.
The CdTe source material was prepared by the direct combination of
the elements with different degrees of cadmium deficiency or a stoichio-
metric mixture containing varied concentrations of a dopant, such as arsenic
or antimony. Under the sublimation conditions described above, the elec-
trical resistivity of CSS CdTe films depends strongly on the concentration
of the dopant or cadmium vacancies in the source material. Figure 2 shows
~ lfi4 .
/ i I I I I
i017 1018 I0I? I0~
CONCENTRATIONOF $b IN CdTe SOURCEMAIERIAL, CM"3
~10 3
m
_E
-~I02
I I I I
1018 1019
Concentration of Cd Vacancies in CclTe Source Material, cm"3
Fig. 3. Effect of the c a d m i u m vacancy c o n c e n t r a t i o n in the source material on the resis-
tivity of CSS CdTe films.
electron affinity than p-CdTe (× ~ 5.6 - 5.7 eV). HgTe (X = 5.9 eV) has been
used successfully as low-resistance contacts to p-type single-crystal cadmium
telluride and cadmium mercury telluride, p-type CdTe single crystals of 10 -
15 ~ cm resistivity were first etched in a Br2-CH3OH solution and then
coated with a layer of HgTe 100 pm thick by close spacing isothermal depo-
sition at 550 °C for 20 h. The room temperature specific resistance of this
contact is about 0.1 ~ cm 2 [10].
The use of HgTe as an ohmic contact to p-CdTe films has been investi-
gated in this work. HgTe is thermally unstable; its dissociation pressure in
the temperature range 300 - 500 °C is shown in Fig. 4 (the dissociation pres-
sure of CdTe is included for comparison). The relatively high pressure of
mercury over HgTe, about 1 Torr at 300 °C, suggests that the deposition of
HgTe on CdTe should be carried out at as low a temperature as possible to
minimize the diffusion of mercury along grain boundaries in CdTe and the
interdiffusion between CdTe and HgTe [11]. HgTe is a semimetal; however,
mercury and tellurium vacancies are acceptors and donors respectively. It is
essential to use p-HgTe as the contact to p-CdTe. The deposition of HgTe
films was carried out by the combination of the vapors of the elements in a
gas flow system. This reaction has been used for the epitaxial growth of
HgTe films on CdTe substrates at 470 °C [12]. The apparatus used for the
deposition of HgTe films is shown schematically in Fig. 5. A mercury con-
talner and a tellurium container, both made of fused silica, and the sub-
strate (p-CdTe/SnO2 :F/glass) were placed in a fused silica reaction tube in a
multizone furnace with each zone separately heated and controlled. Since
the vapor pressure of the elements varies exponentially with temperature,
the temperatures of the mercury and Te2 containers must be accurately
controlled. Hydrogen or helium was used to carry mercury and Te2 vapors
TEMPERATURE, °C
500 400 300
100~~' ~ I ' II
10-2
~- 104
10-8 I I "F. N 1 r t x
1.2 1,4 1.6
10001 T, °K-1
F i g . 4. D i s s o c i a t i o n pressure of HgTe in the temperature r a n g e 3 0 0 - 5 0 0 °C.
33
®i
®~ , I
•
TO
L~"AU S T
A, B, C, D MASsFLOWCONTROLLERS
t
(~ SRUT-OFFVALVE
Fig. 5. Schematic diagram of the apparatus for the deposition of HgTe films by direct
combination of the elements.
to the substrate surface. The important process parameters are the substrate
temperature, and the composition and flow rate of the reaction mixture.
Although the substrate temperature is an important factor determining the
deposition rate of HgTe, it is essential to use as low a substrate temperature
as possible. The composition of the reaction mixture, i.e. the [Te2]/[Hg]
ratio, determines the conductivity t y p e of the HgTe films. The [Te2]/[Hg]
ratio required to produce p-type HgTe films depends on the substrate
temperature because of the decreasing sticking coefficient of mercury with
increasing temperature.
A series of experiments have been carried o u t for the deposition of
HgTe films. A substrate temperature of 400 °C was used in the initial experi-
ments; tightly adherent p-type films were obtained at rates of 0.3 - 0.5 gm
min -1 using [Te2]/[Hg] ratios of 1 - 1.2. Films 3 - 5/~m thick had a sheet
resistance of 50 - 80 ~2/[3. However, shunting occurred in most cases, and
the substrate temperature was subsequently reduced. At lower substrate
temperatures, such as 300 °C, the deposition rate was significantly reduced,
and lower [Te2]/[Hg] molar ratios must be used to yield p-type films
because of the higher sticking coefficient of mercury. Further, the sheet
resistance of deposited films also increased. The lowest temperature at
which adherent films can be obtained is 2 7 5 - 300 °C. The composition of
the films was confirmed by X-ray diffraction. At 275 °C, shunting does n o t
occur as frequently as at higher temperatures.
5. Photovoltaic characteristics
1
AREA: c m ~
Voc: 0 . 7 8 V \
Jsc: 23.4rnA/cm2 \
FF: 59.4% \
lO E:E: 10.84'b /
0.2
I
0.4
L
0.6
|1
0.8
VOLTAGE,VOLTS
Fig. 6. Illuminated characteristics of a p-HgTe/p-CdTe/Zno.3Cd0.TS/SnO2:F/glass solar
cell.
Acknowledgment
This work was supported by the Solar Energy Research Institute under
subcontract XL-5-05039-1.
References