Temperature And PV Panels
User Manual: Home
Open the PDF directly: View PDF 
.
Page Count: 23
Photovoltaic
Cell
Efficiency
at
Elevated
Temperatures
by
Katherine Leung Ray
Submitted
to
the Department
of
Mechanical
Engineering
in
Partial
Fulfillment
of
the
Requirements
for the
Degree
of
Bachelor
of
Science
at
the
Massachusetts
Institute
of
Technology
June
2010
C
2010
Massachusetts Institute
of
Technology
All
rights
reserved
A
ARCHIVES
MASSACHUSETTS
INSTiTUTE
OF
TECHNOLOGY
JUN
3 0
2010
LIBRARIES
Signature
of
Author:
DepartmentW
Mechanical
Engineering
May
10,
2010
Certified
by:
Gang
Chen
Carl
Richard Soderberg
Professor
of
Power
Engineering
Thesis Supervisor
Accepted
by:
_
John
H.
Lienhard
V
lonins
Professor
of
Mechanical
Engineering
Chairman, Undergraduate
Thesis
Committee
Photovoltaic
Cell Efficiency
at
Elevated
Temperatures
by
Katherine
Leung Ray
Submitted
to
the Department
of
Mechanical
Engineering
on May
10,
2010
in
Partial
Fulfillment
of
the
Requirements
for the Degree
of
Bachelor
of
Science
in
Mechanical
Engineering
Abstract
In
order
to
determine
what
type
of
photovoltaic solar
cell
could
best
be
used
in
a
thermoelectric
photovoltaic hybrid
power
generator,
we
tested
the change
in efficiency
due
to
higher
temperatures
of
three
types
of
solar
cells:
a
polymer
cell, an
amorphous silicon
cell
and
a
CIS
cell.
Using
an
AM1.5
G
solar
simulator
at
973
W/m
2
we
took
the
I-V
curve
of
each
of
the
three
cells
at
increasing
temperatures.
We
used
the
I-V
curve
to
find the
maximum power
and
determine the
efficiency
of
each
cell
with respect
to temperature.
We
found
that
the
CIS
cell
had
an
efficiency
of
10%
and
the
performance decreased
with respect
to
temperature
in
a
non-linear
manner.
The
efficiency
at
83*C
was
a
peak
and
the
same
efficiency
as
at
40"C.
We
found
that
the
amorphous
silicon
cell
tested had
an
efficiency
of
4%
at
45
0
C
that
decreased
with
respect
to
temperature
in
a
linear
manner
such
that
an
800
C
increase in
temperature
resulted
in
an
efficiency
of
3%.
We
further found
that
the
polymer
cell
efficiency decreased
from
1.1%
to
1%
with
a
60*C
increase
in temperature,
but that
the
polymer
cell
is
destroyed
at
temperatures higher
than
1
00*C.
We
determined
that
CIS
or
amorphous
silicon
could
be
suitable
materials
for
the
photovoltaic portion
of
the
hybrid system.
Thesis
Supervisor: Gang
Chen
Title:
Carl
Richard Soderberg
Professor
of
Power
Engineering
Table
of
Contents
Abstract...........................................................................................................................................2
List
of
Figures.................................................................................................................................
4
Acknowledgem
ents.........................................................................................................................
5
Introduction.....................................................................................................................................6
Background
.....................................................................................................................................
8
Experim
ental Set-Up
.....................................................................................................................
12
Results
...........................................................................................................................................
14
Discussion.....................................................................................................................................20
Conclusion....................................................................................................................................21
References
.............................................................................................
23
List
of
Figures
Figure
1
The solar
spectrum.
1120
nm
is
marked.
All
longer
wavelengths
of
light
cannot
be
used
by
standard
silicon
solar
cells.
The
shorter
wavelengths
create
electron-hole
pairs
of
high
energy
that
relax to
the
bandgap
energy. The difference
is
lost
as
heat..................7
Figure
2
Best recorded
efficiency
under
"standard"
conditions,
1000
W/m
2,
25*C.
Note
that
simply
putting
the
cell
in
the sun
generally
results
in
temperatures
of
about
40*C
(from
K
azm
erski
2006,
106).........................................................................................9
Figure
3
A
thermoelectric
device
situated
between
a
hot
side
and a
cold
side
produces
a
current.
The
temperatures
may
vary.
A
larger
AT
is
ideal. The
heat
collector may or
may
not
be
a
photovoltaic
cell....................................................................................11
Figure
4
Diagram
of
the
set-up.........................................................................13
Figure
5
The I-V
curve
for
the
polymer
cell
at
47,
60,
82, 102,
and
109*C.
Note that
the
cell
begins
to
break down
at
109*C..........................................................................14
Figure
6
Graph
of
the
maximum power
vs.
temperature for
the
polymer
cell....................15
Figure
7
This
is
a
graph
of
the
I-V
curve
of
the
polymer
cell
at
65*C
after
it
operated
at
high
temperature.
The
curve
changed
based
on
whether the
current was
increased or decreased.
The
maximum
power
was
also
10
times
less...........................................................15
Figure
8
I-V
curve
at
44'C
for
the three days
of
testing.
Day
1
the
cell
went
to
higher
currents
for
the
same
voltage............................................................................16
Figure
9
Maximum
power
vs.
temperature
for
the
amorphous silicon
cell.
The
cell
behaved
differently successive
testings............................................................................17
Figure
10
I-V
curves for the
CIS
cell..................................................................17
Figure
11
Maximum
power
for the
CIS
cell...........................................................18
Figure
12
Efficiency
comparisons
of
the
three
cells....................................................19
Figure
13
I-V
curves
at
82
and
47"C......................................................................19
Figure
14
V0
c
for the
three
cells. The
polymer
cell
has
a
slope
of
-0.0061
V/*C,
the
amorphous
silicon
cell
has
a
slope
of
-0.059
V/*C,
and
the
CIS
cell
has
a
slope
of
-0
.0
13
V
/*C
...................................................................................................
2
0
Acknowledgements
I'd
like
to
thank
Daniel
Kraemer,
Professor
Chen
and
MIT
NanoEngineering
Group
for
providing
the
problem
statement for
this thesis
and
the
materials,
space
and knowledge needed to
complete
it.
Special
thanks
to
Ken McEnaney
for
showing
me
how
to
spark
weld
the
thermocouples.
I'd
also
like
to
thank
my
parents
Lydia
Young and
Alan
Ray for
their
encouragement,
and
Hannah
Ray
and
Ian
Leroux for the
much
needed moral support back in
March.
Introduction
In
the
past
decade
concern about
the effect
of
excessive amounts
of
greenhouse
gases
in
the
atmosphere has
been
rising.
Carbon
dioxide
and
methane
are
two
main
greenhouse
gases,
and
carbon dioxide
is
produced
whenever
gasoline,
coal,
oil
or
methane
is
burned
to
produce
electricity
or
power
transportation.
Thus,
work
has
been
done
to
reduce
the
amount
of
carbon
dioxide
going
into the
atmosphere.
Some
of
the
options
for
reducing
the
output
of
carbon
dioxide
include
wind
power,
geothermal
power,
fuel
cells,
carbon sequestration
and
solar
power.
Several
methods
of
utilizing
solar
power
have
been
found.
One is
solar
panels,
wherein light
strikes a surface,
excites
an
electron,
and
electricity
is
produced.
Another
is
thermal
power,
where
the
light
from the sun
is
concentrated
in
one
spot
and
heats water which
is
then
used
in
a
steam
turbine.
Another
possibility
involves
the Seebeck effect,
which results
in electricity
generation
due
to a large
difference
in
temperature over
a
short
distance.
Solar
power
comes
to
Earth
in
the form
of
light,
or
radiation.
Given
what
we
know
about
the
Sun,
we can
calculate
an
amount
of
power hitting the
Earth
per
square
meter.
The
power
that
comes to
the
Earth from
the
Sun,
if
it can
be
completely
gathered,
provides
more
than
enough
power
for
everything
humans
do
now
and
reasonable growth expectations. Currently there
are
methods
of
harvesting
solar
power
into
electricity
that
have
an
efficiency
of
25%
at
best, but
a
more
efficient power collection mechanism
could
do
quite
a
bit.
Let's
first
talk
about photovoltaic
cells.
Most
cells
are
made
of
silicon,
mainly
because
it
works, it
is
plentiful,
and the
computer industry
also
uses
it,
which
means
that
methods
of
producing
and
working with silicon
wafers
are
well
understood.
Crystalline
silicon
has
a
bandgap
of
1.1
eV.
This
means
that
if
a
photon
of
light
comes
in
with
an
energy
of
less
than
1.1
eV,
it
fails to
create an
electron
and
hole
pair
across
the
bandgap,
and
if
it
has an energy
of
more
than
1.1
eV
it
does
excite
an
electron
and
hole
pair, but
any
extra
energy
is
lost
because
the
electron
and
hole
settle at
1.1
eV
quickly.
A
photon's
energy
relates
to
its
wavelength
by
hc
E
=
-I
where
E
is
energy,
h
is
Planck's
constant,
c
is
the speed
of
light
and
2
is
the
wavelength.
Thus
photovoltaic
cells
do
not
use
any
of
the
energy
from
electromagnetic
radiation
with
longer
wavelengths
than
1120
nm
(which corresponds
to
1.1
eV).
See
Fig.
1.
1120
nm
is
infrared
light.
Lower wavelengths
of
light
tend
to
manifest
as
heat
when
it
hits
an
object.
Consider
that
infrared goggles
will
show warm objects,
and
microwaves
can
heat
food. Thus,
combining
a
photovoltaic
cell
and
a
mechanism
to
collect
the
thermal
energy
could
greatly
improve
the
energy
collection
efficiency
of
the
machine.
ASTM
6173-03
ROWrnce
Spectra
2.00
1.75
E
1.50
E
t
1.25
1.00
0.75
0.s0
0.25
0.00
'L
250
-Grt+circumsdar
Wm-2'fwlI
Energy
not
used
by
PV
500
750
1000
1250
1500 1750
2000
2250
2500
2750
3000
3250
3500
3750
4000
Wavelength
nm
Figure
1:
The solar
spectrum.
1120
nm
is
marked.
All
longer
wavelengths
of
light
cannot
be
used
by
standard silicon
solar
cells. The
shorter wavelengths
create
electron-hole pairs
of
high
energy
that relax
to the
bandgap energy.
The
difference
is
lost
as
heat.
The
Nanoengineering
Group
at
MIT
has
been working
on
a
design
of
a
machine
that
will
combine
photovoltaic
and thermal solar
power.
The
thermal
power
in
this
case
will
be collected
using
a
thermoelectric
device
that
creates
a
current
from
a
temperature
gradient.
There needs
to
be
a
temperature difference
of
at
least
500
C
for
the
thermoelectric
device
to
work
efficiently,
which
means
one
side
will
be
25"C
and the
other
75"C,
for
example.
Higher
temperature
differences
tend
to
be
better.
A
difference
of
25
0
C
to
150*C
would
be
excellent.
However,
this
means
that
the
photovoltaic
cell
might
be very,
very hot.
There
is
not
a
lot
of
research into solar
cell
performance
at
high
temperatures.
This
paper
investigates
how
different
types
of
solar
cells
behave
at
higher
temperatures.
.... ..... .. .. .. .
.....
.
....
.
......
...
........
Background
Available
Solar
Energy
The
Sun
radiates
energy
based on
its
temperature
(about
5700
K).
The
amount
of
power
that
hits
the
Earth's
upper
atmosphere
is
1366
W/m
2
(NREL).
By
the time
it penetrates
the
atmosphere,
the
power
is
down
to
approximately
1000
W/m
2.
The
Earth
has
1.49
x
108
km
2 of
land
(Pidwirny
2006),
which
means
that
collecting
solar
power
from
all
the
available land mass
gives
about
1.4
x
105
TW
of
power, although
since
weather
plays
a
large
part,
3.6
x
104
TW
may
be
more
realistic
(Buonassisi
2009,
3).
World
power
consumption
in
2004 was
14
TW
(IEA
2006,
6)
which
is
much
smaller
than
36,000
TW.
If
there
were
a
100%
efficient
solar
power
collection system,
1/1 0 0 0 th
of
the
Earth's
dry
surface
would
provide
more
than
enough power.
As
it
is,
with
photovoltaic
cells
averaging
15%
efficiency,
the
U.S.
could
be
powered
with
130,000
km
2
or
about
50%
of
Nevada
(Buonassisi
2009,
82).
Types
of
Photovoltaic Solar
Cells
Photovoltaic
solar
cells
work
by
absorbing
light,
creating
electron-hole pairs,
separating
charges and
running
them
through
an
external load.
The
main types
of
solar
cells
in
use
today
are
crystalline
silicon, both
single
and
multi-crystalline,
and
what
are
known
as
thin
film
solar
cells,
which
include
amorphous
silicon,
cadmium
telluride,
copper
indium
gallium
diselenide
(CIGS),
and
copper
indium
diselenide
(CIS).
There
has
also
been work
on
organic
photovoltaic
cells
and
dye-sensitized
cells.
The
best record efficiencies
of
the
various types
of
cells
is
shown
in
Fig.
2.
0
1975
1980
1995
1990
1995
2000
2006
Figure
2:
Best
recorded
efficiency
under "standard" conditions,
1000
W/m
2,
25"C.
Note
that
simply
putting
the
cell
in
the
sun
generally
results
in
temperatures
of
about
40*C
(from
Kazmerski
2006,
106)
Measuring
Efficiency
of
Photovoltaic
Cells
Efficiency
is
the
ratio
of
the
amount
of
energy
or
power
output
by
a
device
to
the
amount
of
energy or
power
received.
For
photovoltaic
cells,
this
is
the
amount
of
electric
power
out
(voltage
multiplied
by
current)
divided
by the
amount
of
power
of
the
incoming
sunlight
(the
standard
1000
W/m
2).
The
most
straight
forward
way
to
measure
the
efficiency
of
a
solar
cell
is
to
put
it
in
1000
W/m
2
light,
measure
the
I-V
(current-voltage)
curve
and
find
the
maximum
power
point. That
power
divided by
1000
W/m
2
multiplied
by
the
area
of
the
cell
gives
the
efficiency
of
the
cell.
Since
sunlight that reaches
the
ground
is
different
in
Alaska than
in
Singapore
(there
is
more
atmosphere
to
travel
through),
different
in
summer
than
winter,
and
different
on
cloudy
days than
cloudless
days,
photovoltaic
research has
come
up
with
a
standard
against which
to
test
all
cells.
The
standard
we
used
is
called
AM1
.5
G
because
it
is
sunlight
after going
through
about
1.5
atmosphere thicknesses with
global,
G,
collection
of
light.
Another
standard
is
the
AM
1.5
D,
(the
D
is
for
direct),
that
is
used when
the cell
tracks
the sun.
The
standard
is
maintained by
having
a
reference
cell
measured
at
a
central
location
and
used
to
. ............................................................. ..........................
. .. ............................................................... . .. .. .....
... ... .. .......
I-~
I'll, " I'll"
compare
against
the
new
cell
being
tested
(Green
1986,
99-100).
In
this
case, we
used
the
straight forward
method
of
measuring
efficiency
by
measuring
the
I-V
curve.
When
describing
I-V
curves, three
numbers
are
commonly
used:
the
open-circuit
voltage
(Voc),
the
short-circuit
current
(Is,),
and
the
fill
factor
(FF).
The
open-circuit
voltage
is
where
the
curve
crosses
the
voltage
axis
and
the
short-circuit
current
is
where
the curve
crosses the
current
axis.
The
fill
factor
is
defined
as
FF-
mp
Vmp
FF
=S
O
where
Ip
and
V.p
are
the current
and
voltage
at
maximum
power
and
FF,
Voc,
and
ISc
are
as
given
above.
The
closer
the
fill
factor
is
to
1
the
better
the solar
cell
is.
Commercial
solar
modules
sold
by
the company
Evergreen
Solar have
a
fill
factor
of
0.75
(Evergreen
Solar).
Thermal
Solar
A
typical
large
scale
power plant
runs
on
a
steam
cycle.
The fuel
is
used
to
boil
the
water
and
heat
the steam
which then
runs
through
a
turbine
to
generate
the electricity.
Newer
power
plants
might
use
what
is
known
as
a
combined
cycle,
which
uses
high
temperature
gas
through
one
turbine
and the
exhaust
gas
is
used
to
heat
up
the
steam
which
then
runs
through
another
turbine.
Some
large
scale
power plants use
solar
power,
not
from
photovoltaics,
which
is
expensive,
but
using mirrors
to
focus the
sunlight
onto
a
fluid in order
to
heat
the
fluid.
The
fluid
is
then
used
to
provide
heat
for
the standard steam
cycle.
There
is
another way to use
heat
to get
electricity, without stopping
at
mechanical
energy
first.
It is
known
as a
thermoelectric
effect.
If
there
is
a
temperature gradient
between
two
different
conductors
or
semiconductors connected
in
a
loop
then
a
voltage
is
created
by
the
different
response
to
temperature
of
the
two
materials.
This
is
known
as
the Seebeck effect.
Using
this
effect,
one
can
create
a
device
that,
given
a
current
will
create a
hot
side
and
a cold
side,
or,
given
a
temperature
difference,
will create
electricity.
The
greater
the
temperature
difference, the
more
electricity
obtainable.
Thermoelectric
Deice
Heat
Sik
Heat
Collector
Photovoltaic
Cell
Low
T
High
T
Current
Figure
3:
A
thermoelectric
device situated
between
a
hot
side
and
a
cold
side
produces
a
current.
The
temperatures
may
vary.
A
larger
AT
is
ideal.
The
heat
collector
may
or
may
not
be
a
photovoltaic
cell.
Thermoelectric
devices
are
evaluated
using
something
called
the
figure
of
merit,
or
ZT.
ZT
is
defined
as
aS
2T
ZT
k
where
a
is
the
electrical
conductivity,
S
is
the Seebeck
coefficient,
T
is
the
average temperature,
and
k
is
the
thermal
conductivity.
In
order
for
a
thermoelectric
device to
be
comparable
to
the
standard steam cycles
used
in
power
plants,
ZT
needs to
be
3
to
4.
Most
thermoelectric
devices
from
1950
to
2000
only
achieved
a
ZT
of
1
or less.
At
the turn
of
the
century,
due
to
engineering
at
the
nanoscale
level,
some devices
achieved
a
ZT
of
2.5,
and
could conceivably
do
better
(Majumdar
2004,
777).
This
is
the
reason
that
only
now
thermoelectric
devices
are
being
considered for
power
generation.
One
possible
set-up
for
thermoelectric power
is
to have
a
photovoltaic
cell
collect
heat
for
the
hot
side
of
the
thermoelectric
device, and
have
the ambient temperature
be
the
cold
side,
11
as
shown
in
Fig.
3.
In
this
hybrid system,
the
sunlight
is
transformed
into
electricity
by
both the
photovoltaic
cell
and
the
thermoelectric
device.
The
photovoltaic
portion
takes
care
of
the
higher wavelength
radiation
(visible
light)
and
the
thermoelectric
portion
takes
care
of
the
lower
wavelength
radiation
(infrared,
microwave, "heat")
as
well
as
excess
energy above
the
bandgap.
The
photovoltaic
portion
would
operate
at
a
much higher temperature
than
has
been
done before,
and
thus,
we
are
interested
in
how
much
the
efficiency degrades
as
a
function
of
temperature.
If
the
performance degrades
too
much,
it
may
not
be
worthwhile
to
pursue
the
hybrid concept.
Previous Research
There have
been
a
number
of
studies
into the
way
that
silicon
solar
cells
react
at
different
temperatures.
There
is
one
study
on the
way
CIS
cells
perform
at
different
temperatures.
Meneses-Rodriguez
et
al.
(2005)
compares
one
CIS
cell to several
types
of
silicon
cells
over the
range
of
25*C
to
800
C.
One
of
the
silicon
cells
is
an
amorphous silicon
cell
tested
from
25*C
to
800
C.
Carlson
et
al.
(2000) examined amorphous
silicon
cells
from
20
0
C
to
1
00*C
and Carlson
(1977)
has
information
about
amorphous silicon
performance from
100
K
to
300
K,
-173*C
to
23*C
that
includes
open circuit voltage
and
short circuit current,
but not
fill
factor,
maximum
power
or
efficiency.
Experimental
Set-Up
The
objective
of
this
project
is
to
determine
whether
any
of
several
types
of
photovoltaic
cells
are
candidates
for
the
hybrid
thermoelectric
photovoltaic
system.
Our
method
of
doing
so
is
to
measure
the
efficiency
of
the
various
cells
at
elevated
temperatures
by
measuring their
I-V
curves.
For
these
experiments,
we
used
a
xenon
lamp
with
an
AM
1.5
G
filter for our solar
simulator.
The
light
was
not
collimated,
so
there was
a
spread
of
intensities.
The
sample
was
positioned
3
feet from
the
exit
of
the
lamp,
and
the
power
of
the
lamp
was
set
at
1600
W. This
resulted
in a
incident
power
of
973±57
W/m
2.
The
error
is
due
to
the
non-uniformity
of
the
light
over
an
area.
We
tested
three
cells. The
first was
a
polymer
cell
from
Konarka.
The
second
was
an
amorphous
silicon
cell
from Fuji Electronics.
The
third
was a
CIS
cell.
The
polymer
cell
and
the
amorphous
silicon
cell
were
attached
to
hot
plates with
a
piece
of
copper
in
between
to
spread the
heat
evenly
across
the
entire plate. Omegabond
300
High Temperature Cement
12
worked well
to
attach
the piece
of
copper
to
the
hot
plate
-
there was
decent thermal contact
-
however
it
fell
apart
a
few
days
after
testing, which
is
either good because
it
means
the
hotplate
was
undamaged
or
bad
because
we
wanted
something firmer.
The
cement
failed
to
hold
the
copper
to
the
plastic
coating
of
the
amorphous silicon
cell.
We
instead
used
gasket
maker
and
thermal paste.
Since
the
hot
plate
only
heats
and
has
no
cooling functions,
the
lowest
temperature
available for these two
plates
was
about
45*C.
The
CIS cell
was
attached
to
a
Peltier
device,
which
can
either
heat or cool
depending
on
the
voltage given
to
it.
It
was
attached
with
high temperature
gasket
maker and
put Dow Coming
340
silicone
heat
sink
compound
in
between
the
cell
and
the
device
to
ensure good
thermal contact.
A
K-type
thermocouple
kept
track
of
the
cell
temperature.
One
thermocouple was
between
the
heating element
and
the
cell;
another
was
taped
to
the
front
of
the
cell.
We
used
a
Keithley
2430
1
kW Pulse
mode
SourceMeter
to
run
a
particular
current
through
the
circuit
containing
the solar
cell
and
an
Agilent
34401A
6
2
Digit
Multimeter
to
measure
the
voltage
across the
cell
while
that
particular current
was
active.
This allowed us
to trace
an
I-V
curve
for
each
cell at
each
temperature. Figure
4
shows
a
diagram
of
the
set-up.
Solar
Heating
LiCel
Element
Light
Source
Voltage
Measurement
Current
Source
T r
Temperature
Measurement
Figure
4:
Diagram
of
the
set-up.
Results
A
proper
characterization
of
the
cells
would
include
the
open-circuit
voltage,
short-circuit
current
and
fill
factor.
Unfortunately the short-circuit
current
was
not
easily
measured
and
cannot
be
presented
here.
We
only
obtained
the
open-circuit
voltage
and
the
maximum
power.
Polymer
Cell
The
polymer
cell
performance did
not
change
much with
temperature,
but
suffered
a
total
breakdown
at
temperatures
above
102*C.
After operating
at
this
very
high temperature,
the
maximum
power
out
of
the
cell
decreased
by
an
order
of
magnitude.
I
gathered
four
sets
of
data
before
the
cell
degraded;
however there
is
another
sample
that
may
be
used
for
further testing
in
the
future.
Figure
5
is
the
I-V
curves for the
cell
from
47*C
to
102*C
with
the measurements
from
when
the
cell
broke
down
at
109"C.
The
short-circuit
current
is
between
80
and
90
mA,
which
means
the
fill
factor
is
between
0.53
and
0.57,
but
cannot
be
determined
with
greater
accuracy.
90
01
80
X
-X6
w
XX
XO
750
X-( --
X--
~
__w
30
X
XX]
X
AK*
X t~
XO1
20
*47~~~$0
0C
0
0
20
0 1
2
3
4
5 6
Voltage
(V)
Figure
5:
The
I-V
curve
for
the
polymer
cell at 47,
60,
82, 102,
and
109*C.
Note
that the
cell
begins
to
break down
at
109*C.
Figure
6
shows
the
maximum
power
at
the
four
temperatures determined
by
fitting
a
polynomial
to
the
I-V
curve
and
finding
the
maximum
power.
0.012
0.01
0.008
0.006
0.004
0.002
0
0
20
40
60
80
100
120
Temperature
("C)
Figure
6:
Graph
of
the
maximum
power
vs.
temperature for the polymer
cell.
Figure
7
is
the
I-V
curve
of
the
degraded
polymer
cell. The
short-circuit
current
is
10
times
less
than
before,
and
the
behavior
differs
based
on
whether
one
moves
from
2
mA
to
3
mA
or
3
mA
to
2
mA.
It is
interesting
that
there
is
a
hysteresis
effect,
but
as
this
is
the degraded
behavior
it
does
not pertain
to
the
investigation
at
hand.
7
6
5 0D
4
3 0
S2
1 - D
0
0
1 2 3
4
5 6
Voltage
(V)
Figure
7:
This
is a
graph
of
the
I-V
curve
of
the
polymer
cell at
65*C
after
it
operated
at
high
temperature.
The
curve
changed
based
on
whether
the current
was increased
or
decreased.
The
maximum
power
was
also
10
times
less.
Amorphous
Silicon
Cell
There
are
19
I-V
curves
for
the
amorphous silicon
cell
taken
on
three
separate
days
ranging
from
45*C
to
122*C.
On the
first
day
of
testing,
the
cell
had
a
higher
short-circuit
current
and
a
correspondingly higher
maximum
power
than
on
the
next
two days
of
testing,
as
can be
seen
in
Fig.
8.
The
short
circuit
current
is
on
the order
of
120
mA,
which
results
in
a
fill
factor
between
0.56
and
0.64.
140
4*
120
100
*Day
1
K
Day
2
ADay
3
40
20
0
010
Voltage
(V)
20
Figure
8:
I-V
curve
at
44*C
for the
three
for
the
same
voltage.
days
of
testing.
Day
1
the
cell
went
to
higher
currents
The
maximum
power
of
the
cell
is
linear
with
respect
to
temperature, ignoring
the
higher
power
on
the
first
day
of
testing
(see
Fig.
9).
Amorphous
silicon solar
cells
are
well
known
for
being
more
efficient
the
first time
they
are
exposed
to
light than
the
second time,
so
the
data
from
the
second
and
third
day
of
testing
are
more
to
be
trusted.
0.05
0.045
0.04
0.035
0.03
0.025
0.02
0.015
0.01
0.005
0
4LA&
*day
1
day
2
A
day
3
0
20 40
60
80
100 120
140
Temperature
Figure
9:
Maximum power
vs.
temperature for
the
amorphous silicon
cell.
The
cell
behaved
differently
between successive
testings.
CIS
Cell
The
CIS
cell
had
24
I-V
curves
taken
on three separate
days
ranging
from
7*C
to
104*C.
The
short
circuit
current is somewhere
around
68-70
mA,
which
gives
a
fill
factor
of
0.29
to
0.34
at
lower
temperatures
(7-60*C)
and
0.37
to
0.41
at
higher temperatures
(80-104*C).
80
1
2
3 4
Voltage
(V)
Figure
10: I-V
curves for
the
CIS
cell.
The
maximum
power
of
the
CIS
cell
with respect
to
temperature
does
not fit
a
straight
line
(see
Fig.
11).
The
power
on the
first
day was
worse
than
the
power
on the two
later
days.
There
was
a
slight bump
at
83*C.
This
may
have
to
do
with
the
improvement
in
fill
factor
even
though
the
open-circuit
voltage
is
decreasing.
Further
study
may
be
warranted.
0.12
0.1
-
0.08
0.06
*Day
1
0.04
ODay
2
ADay3
0.02
0-
0
20
40
60
80
100
120
Temperature
(*C)
Figure
11:
Maximum power
for
the
CIS
cell.
All
Three
Cells
Each solar
cell
was a
different
size,
so
when
we
want
to
compare the
cells,
it
is
necessary
to
use efficiency.
Figure
12
shows
the
efficiency
based
on
an
input
power
of
973
W/m
2.
The
CIS
cell
is
the
most
efficient
of
the
three.
The
amorphous silicon
cell
is
next
best
and
the
polymer
cell
has
the
lowest efficiency.
The
polymer
cell
performance
has
the
least
amount
of
temperature dependence.
The
amorphous
silicon
cell
output
vs.
temperature behaves
in
a
predictable
and
linear manner.
The
CIS
cell
behaves
in a
non-linear
manner.
0.14
0.12
0.1
0.08
0.06
0.04
0.02
A
A
A
A
-A
AA
A-AA
*
Polymer
*
a-Si
A
CIS
0
20
40
60
80
100 120
140
Temperature
(*C)
Figure
12:
Efficiency
comparisons
of
the three
cells.
Figure
13
shows the
I-V
curves
at
45*C
and
820
C
for
the
three
cells. The
CIS
cell's
I-V
curve
changes
shape,
the
polymer
cell's
I-V
curve shifts
to
the
left,
and
the
amorphous
silicon
cell's'
I-V
curve
changes
the
most.
120
10
100
Polymer,
82*C
m
a-Si, 82*C
CIS,
83
0C
X
Polymer,
47*C
X
a-Si,
45*C
+CIS,
50
0C
20
Voltage
(V)
Figure
13:
I-V
curves
at
82
and
47
0
C.
60
XX
X:x
x
X
X
-)K
X
x
X
X
x
X
x
Figure
14
shows the
open
circuit
voltage
vs.
temperature.
20
18
16
14
~12
10
rolymer
8
Oa-Si
6
4
2
0
0
20 40
60
80
100
120
140
Temperature
(*C)
Figure
14:
V.c
for
the
three
cells. The
polymer
cell
has
a
slope
of
-0.0061
V/*C,
the amorphous
silicon cell
has
a
slope
of
-0.059
V/*C,
and
the
CIS
cell has
a
slope
of
-0.013
V/*C.
Discussion
We
can
compare
the
cell
performance tested
in
this
paper
to
cells from
literature.
Meneses-Rodriguez
et
al.
(2005)
used
a
measurement
called
KT
to
report
the
behavior
of
the
tested
cells.
KT
is
defined
as
K
=
dPmax
* 1000
dT
Pmax
or
the change
in
maximum power based
on
temperature
normalized
by
the
maximum
power,
reported
by
the
authors
in
parts
per
thousand
and
Kelvin. Meneses-Rodriguez reported
a
KT
of
-2.5
for
their
CIS
cell
testing
from
25*C
-80*C,
and
further stated
that
other
literature
reported
a
KT
of
-5.9
to
-2.4.
There
is
more
than
one
dPmax/dT
for
the
CIS
cell
in
this
report.
Using the
slope
between
13*C
and
61
C
and
the
minimum
and
maximum power
over
that
range
as
shown
in Fig.
11,
we
have
a
KT
of
-2.7
to
-3.3,
or
-3.0
using
average
power.
That
is
comparable
to
the
KT
found
by
Meneses-Rodriguez.
Referring
to
the
same Figure,
using
the slope
between
61"C
and
104*C
and
the
minimum
and
maximum power
for
that
range,
we
have
KT
is -0.82
to
-0.88,
which
is
much
flatter.
Meneses-Rodriguez
et
al.
(2005) also
reported
a
KT
of
amorphous silicon testing
from
25*C
to
80*C
of
-2.1
and
literature
values
of
-2.2.
This
amorphous silicon
cell
had
a
KT
of
-3.1
using
the
projected maximum power
at
25*C,
-3.4
from
the
maximum power
at
44*C
and
-4.5
from
the
maximum
power
at
122*C
as
shown
in
Fig.
9.
These
numbers
are
high. Carlson
et
al.
2000
found
that
the
efficiency
of
a
single
junction
amorphous silicon
cell
at
1
00"C
was
87%
of
what
it
was
at
25"C,
and
that
the
efficiency
at
25*C
was the
same
within
1%
as
the
efficiency
at
45*C.
They
also
used
tandem
cells
for
which
the
efficiency
at
100*C
was
75%
of
the
efficiency
at
25*C
so
long
as
the
cell had sat
in light for
642
hours
(~4
weeks)
first.
At
103*C
this project's
amorphous silicon
cell
was
at
83%
of
the
efficiency
at
44*C
and
76%
of
the
projected
efficiency
at
25
0
C.
Carlson
(1977)
measures from
-173'C
to
25*C.
His data
show
that
current
falls
abruptly
at
temperatures
lower
than
150K
(-123
0
C).
The
slope
of
open circuit voltage
per
degree
Kelvin
is
-2.3x10
3
V/K in
Carlson's
report,
compared to
-5.9x10
2
V/K
for
the
cell
in
this
study
at
higher
temperatures.
The
polymer
cell
performance
at
high
temperatures
has
no
comparison
in
literature.
Its
efficiency was
measured
at
1%
(see
Fig.
12).
The
best
recorded
polymer
cell
in 2006 had
an
efficiency
of
4%
(see
Fig.
2).
Liang
et
al.
(2009)
reported
a
new
polymer
cell
with
an
efficiency
of
7.4%. The
commercially
available
polymer
cells
are
a
non-ideal
choice
due
to
their
low
efficiency
and
maximum temperature
of
100*C.
The
polymer
cell
in
this project
has
a
KT
of
-1.5,
which
is
less
temperature
dependence
than
either
the
CIS
cell
or
the
amorphous silicon
cell.
The
amorphous
silicon
had
a
measured efficiency
of
2.9
to
3.8%.
The
best
recorded
amorphous
silicon
cell
has
an
efficiency
of
12%
(see
Fig.
2).
Carlson
et
al.
(2000)
tested
amorphous silicon
cells
that
had
4%
and
8%
efficiencies.
The
CIS
cell
had
a
measured
efficiency
of
10.0%
to
10.7%.
The
best recorded
CIS
cell
efficiency
was
16%
(Fig.
2).
This
means the
CIS
cell
had relatively
high
efficiency
for its type while the
amorphous
silicon
cell
used
in
this project
had
worse
performance
than
amorphous silicon
cells
potentially
could
have.
Before
picking
a
candidate
for
the
hybrid
system
it
may
be
advisable
to
run
another
set
of
tests
on the
CIS
cell
(or
a
second
CIS
cell)
in order
to
pin
down
the
temperature
dependant
behavior,
and
a
set
of
tests
on
a
higher
efficiency amorphous
silicon
cell.
Conclusion
In
order
to
build
a good
hybrid thermoelectric photovoltaic
device,
the
photovoltaic part
needs
to
have
a
good
efficiency
at
high
temperatures.
The
CIS
cell
had
a
higher
efficiency
and
operated
at
temperatures
up
to
102*C
without
degrading.
The
amorphous silicon
cell
had
a
lower
efficiency,
but
other amorphous
silicon
cells
that
have
been
tested
have
had efficiencies
21
comparable
to
the
CIS
cell's
efficiency.
The
polymer
cell
cannot
be
operated
over
100*C
without degradation,
and
this
should
be
factored into
any
design
that
requires
the
polymer
cell
to
operate
at
high temperatures.
The
polymer
cell's
efficiency
was
very
low.
It
does
not
seem
a
suitable candidate for the
photovoltaic
part
of
the
hybrid
system.
Both
CIS
and
amorphous
silicon
are
suitable
materials for
the
photovoltaic
part
of
the
solar
cell.
Further testing
could
be
done
to
ensure
that
the
maximum
operable
temperature
of
both
types
of
cells
is
not
under
the
desired operating temperature
of
the
hybrid
system. More
testing
could
be
done on
a
higher efficiency
amorphous
silicon
cell,
and
more
testing
could
be
done
on the
CIS
cells
to
pin
down the
CIS
cell's
non-linear
temperature
dependant
behavior.
References
Buonassisi,
T.
"The
Solar
Resource."
Lecture
Notes,
2.627:
Fundamentals
of
Photovoltaics,
Fall
2009.
Carlson,
D.E.
"The
Effects
of
Impurities
and
Temperature
on
Amorphous
Silicon
Solar
Cells."
IEEE
(1977)
214-217.
Carlson,
D.E., G.
Lin,
G.
Ganguly
"Temperature Dependence
of
Amorphous
Silicon
Solar
Cell
PV Parameters."
IEEE
(2000)
707-712.
Evergreen
Solar,
Inc.
ES-A
Series
Photovoltaic
Solar
Panels.
Datasheet.
http://www.evergreensolar.com/app/en/products/item/739/.
Accessed
May
7,
2010.
Green, M.
Solar
Cells:
Operating
Principles,
Technology
and
System
Applications,
Sydney:
University
of
New
South Wales,
1986.
International Energy
Agency
(IEA).
"1973
and
2004
Fuel
Shares
of
Total
Primary
Energy
Supply."
Key
World
Energy
Statistics,
2006.
Available
from
http://www.iea.org/Textbase/publications/freenewDesc.asp?PUBSID=1
199/.
Kazmerski,
L.L.
"Solar
photovoltaics
R&D
at
the
tipping
point:
A
2005
technology overview"
Journal
of
Electron
Spectroscopy
and
Related
Phenomena
150
(2006):
105-135.
Liang,
Y.,
Z.
Xu,
J.
Xia,
S.T.
Tsai,
Y.
Wu,
G.
Li,
C.
Ray,
L.
Yu.
"For
the Bright
Future
-
Bulk
Heterojunction Polymer
Solar Cells
with
Power Conversion Efficiency
of
7.4%"
Advanced
Materials
22
(2010):
1-4.
Majumdar,
A.
"Thermoelectricity
in
Semiconductor Nanostructures."
Science
303,
(2004):
777-
778.
Meneses-Rodriguez,
D.
et
al.
"Photovoltaic
solar
cells
performance
at
elevated
temperatures"
Solar
Energy
78
(2005)
243-250.
National
Renewable Energy Laboratory (NREL).
"Solar
Spectra:
Standard
Air
Mass
Zero."
http://rredc.nrel.gov/solar/spectra/am0/ASTM2000.html/.
(accessed
April
26,
2010)
Pidwirny,
M.
"Introduction
to
the
Oceans."
in
Fundamentals
of
Physical
Geography,
2nd
Edition.
2006.
http://www.physicalgeography.net/fundamentals/8o.html/.
