Junction Structure and Dark Current

1
Junction Structure and Dark Current
The junction potential at the p-n junction has an
important effect on the semiconductor energy
levels. Consider first the separated p-type and
n-type phases first. They have the same
conduction and valence band edge energies,
separated by the same and gap, but different
Fermi levels and therefore different work
functions, fp and fn.
Fermi level -
When the two doped materials are pushed together
and equilibrated, the Fermi level must be the
same throughout the device causing the band edge
energies, the semiconductor energy levels) to
bend across the junction in response to the local
electric field. The equilibrium band-bending
energy is qVbo is related to the difference in
the work functions of the separate, uncharged
materials qVbo fp - fn
2
P-N Junction

• A p-n junction is formed within a single crystal
• Part of it is n-type and part is p-type
• To understand the p-n junction, we can consider
  separate p and n regions
• Due to diffusion (or difference in Fermi levels),
  electrons move to the p region and holes to the n
  region
• Positive charge (ionized donor atoms) in the n
  region and negative charge in the p region
• An internal electric field is created pointing
  from n to p regions

n-type
p-type
After contact
3
As the Fermi level in a doped semiconductor
normally is in the gap, but near the
majority-carrier band edge, qVbo is normally
slightly smaller than the bandgap energy. There
is a junction potential that exists at the
interface across the n-p Si solar cell in the
dark. At equilibrium, no net current flows in
the cell, but small movements back and forth of
electrons and holes in the valence band occur.
These are called the generation, ih,gen and
ie,gen thermal generation currents from the
minority carriers (e- on the p side and holes on
the n-side. Also the recombination currents,
ioh,rec and ioe, rec, present from the majority
carriers) are present, both generated in the
device at a miniscule rates under thermal
conditions.
4
(No Transcript)
5
Those minority carriers that reach the junction
without first recombining are swept across it in
opposite directions by the strong electric field.
On the other hand, the majority carriers must
flow up the band bending barrier (energetically
unfavorable), but entropically favorable because
the carriers move from a region of high to low
concentration.
At equilibrium, the generation and recombination
currents in each band exactly balance each other,
and the sum of the hole and electron thermal
generation currents is called the saturation
current density, io of the junction. io
ih,gen ie,gen ioh,rec ioe,rec
6
When a forward bias voltage, Vj gt0, is applied
across the junction of the dark cell, the barrier
height is reduced to qVb q(Vbo-Vj). The
generation currents are not affected, but the
recombination currents are. The net current
across the junction, which is the difference
between the recombination current and the
generation current is call the dark current or
junction current ij. ij(Vj) ih,rec(Vj)
ie,rec(Vj) ih,gen ie,gen ih,rec(Vj)
ie,rec(Vj) ioh,rec ioe,rec
When a reverse bias, Vjlt0, is applied, the
barrier height is increased to qVb q(VboVj),
as shown. the generation currents are still
unaffected, but the recombination currents are
now suppressed. Only a very small
bias-independent saturation current passes ij
(Vjlt0) -io
7
The dependence of the recombination currents
ih,rec(Vj) and ie,rec(Vj) on Vj, is determined by
the recombination mechanism of the carriers
injected into the junction. In most cells, the
dark current-voltage characteristic conforms to
the empirical diode eqn. ij(Vj)
ioexp(qVj/(bkT)) 1 b is
the diode ideality factor. For an ideal
junction, in which no injected carriers recombine
in the junction, b1. For an ideal junction,
in which some carriers do recombine in the
junction, 1ltblt2. For thin-film cells a better
eqn. to use is ij(Vj) io1exp(qVj/(kT)) 1
io2exp(qVj/(2kT)) 1 The 1st term
corresponds to carriers that move across the
junction without recombining, and the second to
the carriers that recombine in mid-gap.
8
Independent of the exact form of the diode eqn.,
all PV cells behave as rectifiers in the dark,
showing highly nonlinear current-voltage
character. Junctions must show rectifying
properties in the dark, if they are to show
photovoltaic properties in the light. When a PV
cell is illuminated, a photocurrent and
photovoltage are generated. Fig 1.9
Absorption of photons of energy greater than the
bandgap energy of the semiconductor creates
excess minority carriers throughout the
illuminated region of the cell. The light at the
interior of the cell falls off exponentially with
distance into the cell. Once the minority
carriers are generated by the photons, (same song
and dance), they diffuse from the quasi neutral
region (QNR) toward the junction where they are
swept across the junction by the junction
potential. The sum of the two currents,
electrons and holes, is the photocurrent, iph
ih,ph ie,ph
9
The photocurrent is directly proportional to the
absorbed photon flux, but independent of the
junction potential Under open circuit, no
current is drawn from the cell and the
photocurrent must be balanced by the
recombination current, so the junction
self-biases in the forward direction by the
open-circuit voltage Voc. At this voltage, the
recombination current exactly opposes the
photocurrent such that iph ij(Voc)
0 Fig 1.9 Under short circuit conditions, the
illuminated cell will deliver maximum output
current, but at zero output voltage. If the
internal resistance effects are negligible, the
junction bias, Vj is also zero, so the band
bending is the same as in the dark junction at
equilibrium. Here the short circuit current is
given by isc iph - io
10
Under normal circuit conditions, the band bending
and junction current are intermediate between the
open-circuit and short circuit ideals, the cell
delivers current i at output voltage V Vj and
the current is given by i iph ij(Vj) If
the photocurrent is bias-independent, the
current-voltage character of the dark and
illuminated cell will look similar, but the
latter will be shifted down with respect to the
former by the constant amount iph. This is
called Superposition. Fig 1.10 p 21 Clean
Electricity The superposition is an ideal case.
It is not to be expected where the photocurrent
is bias dependent, which can happen for a number
of reasons. In the amorphous silicon cell, the
field in the junction region is weak and the
extent of the recombination in it is bias
dependent. Also cells operating a high
injection mode, where the concentration of
photogenerated minority carriers becomes
comparable with that of the majority carriers,
dont show superposition since the majority
carrier concentrations and fluxes are not the
same in the light and dark. Also cells that have
a significant internal series resistance or shunt
conductance also depart from superposition.
11
Current Voltage Characteristics (i-V) The i-V
characteristic of the illuminated cell can be
found by substituting the diode eqn. into the
normal circuit condition eqn. i iph ijVj)
iph ioexp(qV/(bkT))-1. The output power
is a product of i and V, iV, which is the area of
the rectangle of sides i and V inscribed in the
i-V curve. The power should be zero for both
the open circuit and short circuit conditions.
The fill factor, defined previously is a measure
of the squareness of the i-V curve. nfill imp
Vmp / (isc Voc) In efficient cells, the fill
factor is around 0.7 to 0.8. In poor cells it can
be 0.5 or lower.
12
The Voc can be found from the above eqn. by
setting i0 and VVoc. The Voc is given
by Voc bkT/q ln (1 iph/io) bkT/q
ln(iph/io) For good performance iph and Voc must
be as large as possible. What can be
expected? The max value of iph would be obtained
if all the photogenerated electron-hole pairs
were collected as photocurrent, and iph can
achieve 80-90 of this limit if light absorption
and minority carrier collection are both highly
efficient. The limiting value of Voc is the
built in voltage, Vbo. This corresponds to a
complete flattening of the bands across the
junction. This could only happen under extremely
intense illumination. At AM1 conditions, the
values of Voc are usually not much greater than
0.7 Vbo. Thus Vbo should be as large as
possible, essentially limited by the band gap of
the material, or more succinctly, the work
function difference between the two sides of the
junction should be as large as possible.
13
Note that Voc increases as the saturation
current, io, decreases. The io has no minimum
value. In thin cells with well passivated
surfaces, io can be driven down toward zero, and
Voc toward its upper limit of Vbo. In thicker
cells where volume recombination occurs, the
lower limit on io is determined by the rate of
radiative recombination of minority carriers, but
usually non-radiative recombination also occurs
and this raises io by several orders of
magnitude. The band bending diagrams shown so
far are for the simplest and most common type of
photovoltaic junction. This is a p-n
homojunction. There are others such as a p-i-n
homojunction, p-n heterojunction formed between
semiconductors with different band gaps, a metal
semiconductor junction formed between a metal and
an n-type semiconductor etc. Fig 1.11
14
Polycrystalline Silicon Polycrystalline and
amorphous silicon cells have advantages over
single-crystal Si cells. There are cost
reductions Inexpensive refining processes Less
Si for each device Less costly and less energy-
and time-consuming wafer or layer deposition
steps Problems making these types of cells
include Lower efficiencies Device performance
degradations Lack of in-depth knowledge of
materials and devices What is polycrystalline
Si? Material made up of many grains of
single-crystal silicon Larger more perfect
grains give better performance, the
electrical Behavior more nearly resembles cell
made from single-crystal Si.
15
(No Transcript)
16
When light-generated charge carriers encounter
grain boundaries, problems result. With larger
grains means fewer grain boundaries so better
cell performance. Charge carries tend to flow
from top to bottom, a cell whose grains are wide
vertical columns is a better cell. (Fig p 69
SERI) Polycrystalline solar cells have been
fabricated with efficiencies over 10 and
have significant long-term reliability, with
decrease raw material
17
Czochralski wafers are often 300mm thick and an
equal or greater amount Of Si than is contained
in the active cell is lost when Si ingots are
sawn into Wafers. In
Polycrystalline fabrication, ribbon growth
techniques of flat rectangular Polycrystalline,
150mm cell wafers result in a factor of 2x300/150
(a factor of 4) in savings of material.
18
Grain Boundaries Grain boundaries hinder the
flow of charge carriers either by adversely
affecting Their movement or by capturing them.
If light-generated charge carriers encounter a
grain boundary before separation by the junction
potential, then recombination of electrons and
holes can result. If they are held back from
free movement on their way to the electrical
contacts then it is a resistance
issue. Impurities may build up at grain
boundaries as well sometimes short circuiting the
cell. Grain boundaries can also provide a
pathway for light generated carriers that have
been separated by the junction to move back
across it, decreasing the voltage. Grain
boundaries can contain numerous recombination
centers, which are dangling bonds from either
broken or distorted quality of the Si lattice or
impurities. Dangling bonds have the same charge
as the majority carrier in n-type Si they are
negatively charged and in p-type, positively.
But an electron in a dangling bond is less
stable than one on normal Si site. So, such
captured electrons naturally fall into a
more- stable, lower energy site recombining with
holes.
19
Fabrication Must be produced in relatively
uncontaminated way with wide (cm-sized) columnar
grains spanning the cell from back to front.
Formed in directional solidification process. If
not cooled in this way, from one direction but
from all sides, then randomly oriented and
oddly-shaped boundaries would result. Grains can
have atom-to-atom structural defects called
dislocations which behave as grain boundaries on
a smaller scale. Another problem is SiC
inclusions that result from heating up carbon
impurities during fabrication. Casting A vat of
molten Si is allowed to solidify, and impurity
segregation is good because contaminants tend to
remain n the liquid during solidification and are
not incorporated into the crystal. Grains can be
large. Some dislocations and SiC inclusions
result. Also, cast Si must be sawed, resulting
in a 50 material loss.
20
Ribbon Growth A pair of elongated silicon
crystallites, dendrites are extended into the Si
melt. Then as the dendrites are raised a web
forms between them, making an almost perfect
nearly single crystal Si ribbon. In a seeded
version a seed crystal is oriented to cause
vertical grain boundaries of highly controlled
on nondestructive character to propagate
down ward from the seed into the web. This can
result in large grains about 1cm wide and cells
with up to 13 efficiency. Silicon on Ceramic
SOC Dip coating involves slowly withdrawing a
piece of carbon-coated ceramic from a pool of
molten silicon resulting in the deposition of a
polycrystalline layer of Si. The ceramic form
must be designed into the cell as the Si cannot
be removed from it. Silicon on Inexpensive
Substrates Silicon vapor deposition on cheap,
abundant, easily handled materials such as steel
or glass. However impurities from the substrate
(or from process equipment likely enter the Si.
The best efficiencies are about 3.6.
21
Amorphous Silicon The structure in amorphous Si
is very disordered. Until recently (1974)
amorphous Si was thought to be totally
inappropriate. However, by controlling the
conditions under which it is made and by
modifying its composition, it can be used in
solar cells In fact even though it is not well
understood, it is a leading possibility for
future solar cell production. In amorphous
Si the tetrahedral units do not line up with each
other. They are randomly rotated with respect to
each other. The tetrahedral relationship of
neighboring atoms is almost preserved, but the
long range order is not. The small random
rotation destroys this making the amorphous Si
dense with chains or clusters of linked
tetrahedral groups ending in broken uncompleted
dangling bonds. Amorphous Si absorbs light more
strongly for a given thickness than crystalline
Si.
22
A main barrier to amorphous Si being an excellent
photovoltaic material is the presence of the
dangling bonds. They can result from disorder of
the Si atoms or from non-silicon impurity atoms
which impede the motion of the charge carriers.
However, if the amorphous Si is deposited so
that it contains hydrogen in 5-10
concentration), the dangling bonds are removed
and the efficiency is raised. When an electron
is promoted to a conduction band in crystalline
Si, it can travel great distances, but in
amorphous Si, there are localized states that
result at the lowest energies in the conduction
band that restrict the motion of the electrons
that occupy them. Light is so restricted that in
amorphous Si, the carriers do not move far from
the point of generation. This means that it is
likely that the light generated electrons and
holes recombine rather than remaining mobile
and contributing to conductivity.
23
Amorphous Si Solar Cell Since the electrons and
holes cannot move far, the conventional
pn-junction solar cell design does not work.
Doping amorphous Si p-type or n-type causes
further structural distortion and reduces the
already poor minority charge carrier mobility.
Although a junction filed would exist,
photogenerated carriers would simply recombine
before they can be accelerated and separated by
the field Its bandgap energy is 1.65eV, which is
greater than the bandgap of crystalline Si. A
cells output voltage is directly related to the
size of its bandgap, so cells made of amorphous
silicon have higher output voltages. This
compensates for the fact that lower energy
photons are not absorbed by amorphous Si.
Amorphous silicon absorbs light about 40 times
more strongly than crystalline Si. So to absorb
the same amount of light, one-fortieth of the
thickness is required. This strong absorption is
the key to amorphous Si. Since carriers in
amorphous Si have low mobility and recombine
rapidly, the only way they can be collected
during illumination is for the time needed for
their collection to be short. This requires a
very thin cell.
24
Workable hydrogenated amorphous silicon cells are
designed to have an ultrathin 0.008 micrometers
highly doped p-doped top top layer, a thicker
(0.5-1 micrometer) undoped (intrinsic) middle
layer, and a very thin (0.02 micrometer) bottom
n-doped layer. The top layer is so thin and
relatively transparent that most incident
sunlight passes right through it and then
generates electron-hole pairs in the undoped
amorphous silocon. The top p-type and bottom
n-type induce an electric field across the entire
intrinsic region similarly to the induced
junction potential in a regular crystalline pn
device. A conventional crystalline silicon pn
device has a thickness of about 100 micrometers.
However, the extent of the junction field is
about 1 micron thick, which is about the same as
the electric field which is induced across the
entire intrinsic region in the amorphous Si solar
cells. Figure 7-2 This design compensates for
the very low charge mobility in the tip and
bottom doped layers as the light induced charge
carriers are not induced within them. The p-i-n
design also makes it possible for the
light-generated charge carriers to be separated
in the field of the sandwiched intrinsic layer.
25
Even after dangling bonds and impurities have
been reduced by hydrogenation, intrinsic
amorphous silicon still has a very low intrinsic
charge mobility. Photogenerated charge carriers
lose their momentum so quickly and stay so near
their origination point that they would likely
return to their bound states. But the electric
field induced across the intrinsic region gives
the electrons and holes just enough extra motion
to keep them going and out of trapping defects,
keeping them as free charge carriers. The
electric field acts in opposite directions on the
free electrons and holes, separating them,
sending the electrons toward the n-doped bottom
layer and the holes toward the p-doped top layer.
Even though the free carrier mobility is still
low, the intrinsic layer is so thin and the
field with it is so strong that charges can be
successfully separated and a photocurrent
produced. Efficiencies of 6, voltages of 0.8 V
(higher than crystalline silicons) and currents
greater than 10 mA/cm2. have been attained with
the simplest kinds of p-i-n devices.
Efficiencies for more complex structures have
attained 10 and higher ones are expected.
26
Fabrication Techniques Making solar cells from
amorphous silicon needs to be closely controlled,
but it is a thin film material, and little of it
is needed, so even complex processing can produce
a cheap solar cell. In most processes, a
suitable gaseous form of silicon-hydrogen alloy
is decomposed and deposited as a thin film of
amorphous Si onto a clean substrate. H2 is
included in the deposition gas so the film is
deposited as a SiH alloy material. This would
have fewer structural imperfections. The
addition of a p-dopant like boron or an n-dopant
like phosphorus can easily be done in the same
process with the addition of diborane or
phosphine gas. The substrate on which the Si is
deposited must be kept between 200 and 350C. If
the temp is hotter then the film becomes
microcrystalline with many tiny, randomly
oriented crystal grains. At temperatures cooler
than 200, the hydrogen and silicon do not form a
simple structure free of electronic defects.
Also at temps above 350, hydrogen is driven from
the material, rejuvenating the dangling bonds.
27
There are many methods for depositing thin films
of amorphous Si. Three are a) Glow
discharge b) Sputtering c) Chemical Vapor
Deposition Glow Discharge This technique has
been used to make the most efficient amorphous
silicon cells. There are two major types of
glow discharge direct current and alternating
current (rf discharge). Direct current glow
discharge silicon cells are less efficient than
those made by AC, radiofrequency glow discharge
methods. In rf discharge a stream of silane and
hydrogen gas is passed between a pair of
electrodes with power alternating at 13.56 MHz.
This rf induces an oscillation of energetic
electrons between the electrodes, and the
electrons collide with the silane, breaking it
apart into molecular fragments which deposit on a
substrate placed on top of one of the electrodes
at a rate of about 10-50nm/min. this produces a
thin film of hydrogenated amorphous Si. Doping
can be done by adding diborane or phosphine gas.
This process does not use much energy and is
potentially able to produce large-area
cells Figure 7-3
28
3 SiCl4 Si 2 H2 ? 4 HSiCl3 4 HSiCl3 ? 3 SiCl4
SiH4 SiH4 ? Si 2 H2
29
Sputtering This is a process where a focused
stream of energetic particles such as argon ions
hits a silicon target, driving off silicon atoms.
The Si atoms are ejected from the target in a
beam that reaches an appropriate substrate.
Hydrogen or dopant gases can be passed over the
substrate to react with the gaseous silicon,
forming hydrogenated/doped amorphous Si.
Hydrogenation can also be done by adding H2 to
the argon sputtering gas striking the target.
Fig. 7.4 Chemical Vapor Deposition (CVD) In
CVD gases such as disilane thermally decompose to
form a solid on an appropriate substrate. Both
polycrystalline and amorphous Si can be deposited
this way. The gas reactions are potentially
highly controllable. As yet, polySi or amorphous
silicon cells made via CVD have lower efficiency
than those made by rf discharge. Fig. 7.5
30
n-i-p cell It is also possible to reverse the
order of the layers in an amorphous Si cell..
The n-doped layer can be placed on top, and the
p-type layer on the bottom. Much work has been
done on this. Undoped amorphous Si is slightly
n-ype in its electrical behavior. This is
disadvantageous in terms of the n-i-p design, as
there is a less abrupt interface at the front of
the cell. This lowers the junctions voltage
because more holes leak backward into the n
layer, where they recombine with electrons. To
compensate, the intrinsic layer is often lightly
p-doped to make the i layer more nearly intrinsic
or even slightly p-type. An advantage of the
n-i-p design is that an n-doped amorphous silicon
layer absorbs less light than a p-doped layer.
This is because n-type amorphous Si has a
slightly wider bandgap and is thus transparent to
a greater proportion of the spectrum. As a
result, more light reaches the intrinsic region
in the n-i-p design.
31
Stacked Cell There are two types of amorphous Si
and Si alloy stacked cels a) multilayered cells
made of the same material b) multilayered cells
made with different materials This is one of the
most important recent advances in amorphous Si
research. In a simple stacked cells made of the
same material, the cells act like batteries
joined in series the individual voltages add.
So if each has a 0.5V output, then the combined
output would be 1.0V, and without much change in
the total power. So the current must be about
half that of single cell since the voltage is
twice. For effective stacked cell design, light
must be absorbed and collected equally in both
the top and bottom cells for equal current to be
generated by each. The top cell must be thin
enough to let half the light pass right through
it, while the bottom must be thick enough to
absorb the passed light.
32
The advantages of stacking two similar
amorphous Si cells on top of each other are a)
The cell voltages add, and higher voltage is
often of practical use in actual devices. b)
Carrier collection within each layer is
facilitated because the layers are thinner than
ina normal cell (to absorb only half the light)
so there are fewer electron-hole pairs lost by
recombination. The stacked cell design works
only for thin film solar cells. Since the
depositions needed to make thin films dont use
much energy. Two ways to exploit the stacked
cell geometry are being pursued a) Developing
multilayered amorphous Si cells. b) Developing
multilayered combinations of amorphous Si with
other thin films. Up to five layers of similar
amorphous Si cells have been joined into stacked
cells. Layer thickness is controlled such that
light is absorbed and collected in each cell
equally. Lower current results, but the power
loss is less and also thin conductive oxides,
rather than metal grids, can be used.
33
The most sophisticated and promising stacked cell
design is a layered combination of several
different thin film materials. If the top cell
has a band gap fo 1.65 eV and the lower cell has
an absorbance of 1.4 eV then light with energy
above 1.65 eV is absorbed in the top cell, light
between 1.4 and 1.65eV would pass right through
the top cell and would be absorbed in the lower
bandgap cell. By using a part of the suns
spectrum that would otherwise be lost.kllklkl
34
Cadmium Telluride Solar Cells Until recently,
silicon was used for almost all photovoltaic
applications, with good success but at relatively
high cost. Only two additional semiconductors
have shown real promise for replacing silicon as
the primary material for photovoltaic power
generation CdTe and Cu(InGa)S2. Other
materials including Se, Cu2S, Cu2O, InP, CdSe,
and Zn3P2 have been investigated, but due to
disappointing results or high cost they have
fallen by the wayside. Also GaAs is being
developed for special applications where very
high efficiency is required. CdTe is thought to
be an optimum material for use in solar cells.
It is produced as a thin film, as it has a high
absorption coefficient. Its action is similar to
our discussion of crystalline silicon in that
there is a p-n junction formed with a junction
potential at the interface. The absorbed photons
create electron-hole pairs and the potential
energy of the excited minority carriers is
converted into electrical energy as it is swept
through the junction potential. As in Si, the
separation of the carrier from its opposite leads
to a photoinduced voltage, which can drive an
electron through an external circuit.
35
This type of solar cell then requires high
minority carrier lifetimes and mobilities which
can be only obtained by good crystalline
properties, chemical purity, suitable doping and
low resistance. In this type of thin-film solar
cell, the diode is created by two materials know
as the window layer and the absorber and it is
usually a heterojunction p-n solar cell. CdTe
is the absorber. CdTe can be used in p-n
homojunction cells, but there has been very
limited success here due to the strong light
absorption, CdTe is a direct bandgap
semiconductor, coupled with a high surface
recombination rate limiting the minority carrier
lifetime. Also, it is difficult to manufacture
CdTe p-n junctions in thin film form as the
interdiffusion severely distorts the
junction. Heterojunction p-n cells are the most
promising. The first was n-CdTe/p-CuTe2 with
about 7 efficiency, but diffusion of Cu stopped
further development. Next was n-CdS/p-CdTe
heterojunction. This is effective as Cd Te has a
bandgap energy of 1.45eV and an absorption
coefficient of gt105 cm-1 for visible light, so
that the absorber layer needs to be only a few mm
thick to absorbgt90 of the photons at
energiesgt1.45eV. Current densitite of 27mA/cm2
and open circuit voltages of 880mV with AM1.5
efficiencies of 18.5 can be expected for cells
made from CdTe.
36
How can the material be both n-type and p-type.
In the high temperature phase of CdTe, a slight
nonstoichiometry is present in the form of a
slight Cd deficiency. This leads to p-type
material. No excessive care is necessary for
preparing p-type CdTe films as long as the
substrate temperature is sufficiently high.
Since the bond energies are so high, 5.75eV,
solar photons do not normally lead to
dissociation and destabilization. There is a
significant mismatch of the lattice parameters
between CdS and CdTe. This leads to some
problems with forming the junction between the
two materials. Post deposition treatments can
partially alleviate this problem The
n-CdS/p-CdTe heterojunction solar cell must be
illuminated through the CdS window, so that the
light is absorbed in the CdTe close to the
junctions. The preferred fabrication procedure
is to deposit CdS onto a transparent tin(IV)
oxide coated glass substrate. Next the CdTe is
deposited onto the CdS, and finally a low
resistance contact is made to the CdTe followed
by a back electrode, which does not have to
transmit light. Fig 6.3
37
The CdS will absorb part of the light to be
converted in the CdTe so it should be a thin as
feasible. CdS grows natively n-type without
additional foreign doping. Data indicate that
when the CdS layer is decreased, only half of the
short-circuit current due to light below 515nm
can be realized before the fill factor starts to
fall off due to weak areas. The optimum CdS
thickness is in the 50-80nm range, but production
tends to keep it a bit thicker. Physical vapor
deposition (sublimation/condensation) and
chemical spraying are the main ways to make CdS
films. An alternative is chemical bath
deposition from a metastable aqueous solution of
Cd(C2H3O2)2, thiourea, ammonium hydroxide, and
ammonium acetate at temps. of about 70oC can
deposit CdS on TCO glass. The CdTe deposition
process must utilize the advantages of the
material the native p-type, good crystallinity,
and high minority carrier mobility. About ten
procedures have been developed to do this.
38
. Among them are Sublimation-condensation Chem
ical Spraying Galvanic Deposition Screen
Printing Chemical Vapor Deposition Atomic Layer
Epitaxy Sputtering