Raising the Efficiency Ceiling in
Multijunction Solar Cells
Richard R. King
Spectrolab, Inc.
A Boeing Company
Stanford Photonics Research Center Symposium
Sep. 14-16, 2009
Stanford, CA
Acknowledgments
• Martha Symko-Davies, Fannie Posey-Eddy, Larry Kazmerski, Manuel Romero,
Carl Osterwald, Keith Emery, John Geisz, Sarah Kurtz – NREL
• Angus Rockett – University of Illinois
• Rosina Bierbaum – University of Michigan, Ann Arbor
• Pierre Verlinden, John Lasich – Solar Systems, Australia
• Kent Barbour, Andreea Boca, Dhananjay Bhusari, Ken Edmondson, Chris Fetzer,
William Hong, Jim Ermer, Russ Jones, Nasser Karam, Geoff Kinsey,
Dimitri Krut, Diane Larrabee, Daniel Law, Phil Liu, Shoghig Mesropian,
Mark Takahashi, and the entire multijunction solar cell team at Spectrolab
This work was supported in part by the U.S. Dept. of Energy through the NREL HighPerformance Photovoltaics (HiPerf PV) program (ZAT-4-33624-12), the DOE
Technology Pathways Partnership (TPP), and by Spectrolab.
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
2
Outline
• Global climate change and the solar resource
contact
AR
n+-GaInAs
n-AlInP window
n-GaInP emitter
• Solar cell theoretical efficiency limits
– Opportunities to change ground rules for higher terrestrial efficiency
– Cell architectures capable of >70% in theory, >50% in practice
• High-efficiency
terrestrial concentrator cells
– Metamorphic (MM) and lattice-matched (LM) 3-junction
solar cells with >40% efficiency
– 4-junction MM and LM concentrator cells
– Inverted metamorphic structure, semiconductor
bonded technology (SBT) for MJ terrestrial concentrator cells
p-GaInP base
p-AlGaInP BSF
p++-TJ
n++-TJ
W
id
n-GaInP window
n-GaInAs emitter
p-GaInAs base
M
E
e-
g
Tu
e
dl
id
ll
el
nn
Ce
ll
p-GaInP BSF
p-GaInAs
step-graded
buffer
p++-TJ
• Metamorphic semiconductor materials
– Control of band gap to tune to solar spectrum
– Dislocations in metamorphic III-Vs imaged by CL and EBIC
Ce
p
To
n++-TJ
Tu
el
nn
nucleation
n+-Ge emitter
t
Bo
Ju
m
to
nc
t io
Ce
n
ll
p-Ge base
and substrate
contact
metal
gridline
semiconductor
bonded
interface
2.0-eV AlGaInP cell 1
1.7-eV AlGaInAs cell 2
1.4-eV GaInAs cell 3
1.1-eV GaInPAs cell 4
0.75-eV GaInAs cell 5
• Concentrator photovoltaic (CPV) systems and economics
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
3
Global Climate Change
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
4
Climate and CO2 Over the
Last 400,000 Years
Vostok Ice Core Data
4
Temperature (°C)
2
0
-2
-4
-6
-8
Years Before Present
0
50
00
0
45
00
00
40
00
00
35
00
00
30
00
00
25
00
00
20
00
00
15
00
00
10
00
00
-10
(J.R. Petit, J. Jouzel, Nature 399:429-436)
• Antarctic ice core data allows for mapping of temperature and CO2 profiles
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
5
Climate and CO2 Over the
Last 400,000 Years
Vostok Ice Core Data
350
Temperature (°C)
2
300
0
-2
250
-4
-6
200
-8
Years Before Present
0
50
00
0
150
45
00
00
40
00
00
35
00
00
30
00
00
25
00
00
20
00
00
15
00
00
10
00
00
-10
CO2 Concentration (ppmv)
4
(J.R. Petit, J. Jouzel, Nature 399:429-436)
• Clear correlation between temperature and CO2 levels
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
6
Climate and CO2 –
Recent History
Vostok Ice Core Data
Temperature (°C)
2
300
0
-2
250
-4
-6
200
-8
0
50
00
0
150
45
00
00
40
00
00
35
00
00
30
00
00
25
00
00
20
00
00
15
00
00
10
00
00
-10
CO2 Concentration (ppmv)
350
315 ppm (NOAA,
1958)
384 ppm (NOAA, 2004)
4
Years Before Present
• CO2 has reached levels never before seen in measured history
• If we do nothing, we allow this rising trend to continue at our own peril
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
7
Temperature Anomaly by
Year
1000 years of Earth temperature history…
and 100 years of projection
IPCC (2001) scenarios
to 2100
Rosina Bierbaum, Univ. of Michigan, IPCC
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
8
The Solar Resource
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
9
The Solar Resource
5
6
Ref.: http://rredc.nrel.gov/solar/old_data/
nsrdb/redbook/atlas/
• Entire US electricity demand can be provided by concentrator PV arrays
using 37%-efficient cells on:
150 km x 150 km area of land
or
or
ten 50 km x 50 km areas
similar division across US
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
10
Concentrator Photovoltaic
(CPV) Electricity Generation
CPV cost superiority
40% cell efficiency
CPV cost superiority
50% cell efficiency
Map source: http://www.nrel.gov/gis/images/map_csp_us_annual_may2004.jpg
Higher multijunction cell efficiency has a huge impact on the economics of
CPV, and on the way we will generate electricity.
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
11
Solar Cell Theoretical
Efficiency
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
12
Energy Transitions in
Semiconductors
insufficient energy
to reach Ec
hν
thermalization
of carriers
Ec
hν
Ev
Ec
e-
Ev
600
1.2
500
1
hν < Eg
400
0.8
300
0.6
200
0.4
100
0.2
0
0
0
0.5
1
1.5
2
2.5
Photon Energy (eV)
3
3.5
4
AM1.5D, ASTM G173-03, 1000 W/m2 1.4
Utilization efficiency of photon energy
to bandgap Eg = 1.424 eV
700
1.2
600
1
500
hν > Eg
400
0.8
300
0.6
200
0.4
100
0.2
Photon Utilization Efficiency
AM1.5D, ASTM G173-03, 1000 W/m2 1.4
Photon Utilization Efficiency
700
Intensity per Unit Photon Energy
(W/m 2 . eV)
Intensity per Unit Photon Energy
(W/m 2 . eV)
h+
0
0
0
0.5
1
1.5
2
2.5
Photon Energy (eV)
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
3
3.5
4
13
LM and MM 3-Junction
Cell Cross-Section
contact
contact
AR
AR
n+-Ga(In)As
n-AlInP window
n-GaInP emitter
T
GaInP top cell
p-GaInP base
p-AlGaInP BSF
Wide-bandgap tunnel junction
p++-TJ
n++-TJ
W
E
eid
n-GaInP window
n-Ga(In)As emitter
Ga(In)As middle cell
p-Ga(In)As base
p-GaInP BSF
Tunnel junction
Buffer region
p++-TJ
M
id
g
e
C
T
el
nn
u
T
e
dl
el
C
p++-TJ
n++-TJ
l
u
lJ
M
id
d
el
nn
u
T
g
le
el
C
l
p-GaInP BSF
p-GaInAs
step-graded
buffer
m
tto
o
B
l
n
tio
nc
n-Ga(In)As buffer
n+-Ge emitter
W
E
eid
n-GaInP window
n-GaInAs emitter
p-GaInAs base
e
nn
Tu
el
C
op
p-GaInP base
p-AlGaInP BSF
n++-TJ
nucleation
Ge bottom cell
op
n+-GaInAs
n-AlInP window
n-GaInP emitter
ll
l
Ce
l
p++-TJ
lJ
n++-TJ
nucleation
n+-Ge emitter
p-Ge base
and substrate
p-Ge base
and substrate
contact
contact
Lattice-Matched (LM)
e
nn
Tu
n
io
ct
n
u
m
tto
o
B
Ce
ll
Lattice-Mismatched
or Metamorphic (MM)
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
14
Photon Utilization Efficiency
3-Junction Solar Cells
500
1
400
0.8
300
0.6
200
0.4
100
0.2
0
Photon utilization efficiency
Intensity per Unit Photon Energy
(W/m 2 . eV)
600
.
AM1.5D, ASTM G173-03, 1000 W/m21.4
Utilization efficiency of photon energy
1-junction cell
3-junction cell
1.2
700
0
0
0.5
1
1.5
2
2.5
Photon Energy (eV)
3
3.5
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
4
15
Energy Transitions in
Semiconductors
Ec
hν
V = voltage of solar cell
Eg
qφn
qV
Ev
qφp
= quasi-Fermi level splitting
=
⏐φp - φn⏐
• Not all of bandgap energy is available to be collected
at terminals, even though electron in conduction band
has energy Eg
• Only qV = q⏐φp - φn⏐ is available at solar cell
terminals
• Due to difference in entropy S of carriers at low
concentration in conduction band, and at high
concentration in contact layers: G = H - TS
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
16
Energy Transitions in
Semiconductors
V = voltage of solar cell
Eg
qφn
hν
qV
qφp
Ev
=
⏐φp - φn⏐
AM1.5D, ASTM G173-03, 1000 W/m2 1.4
Utilization efficiency of photon energy
to bandgap Eg = 1.424 eV
1.2
to Voc at 1000 suns
to Voc at 1 sun
700
Intensity per Unit Photon Energy
2.
(W/m eV)
= quasi-Fermi level splitting
600
500
1
400
0.8
300
0.6
200
0.4
100
0.2
Photon utilization efficiency
Ec
0
0
0
0.5
1
1.5
2
2.5
Photon Energy (eV)
3
3.5
4
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
17
Shockley and Queisser (1961)
Detailed Balance Limit of Solar Cell Efficiency
• 30% efficient
single-gap solar
cell at one sun,
for 1 e-/photon
• 44% ultimate
efficiency for
device with
single cutoff
energy
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
18
Assumptions →
Opportunities
• Assumptions for theoretical efficiency in Shockley and
Quiesser (1961)
• Viewed from a different angle, these assumptions represent
new opportunities, for devices that overcome these barriers
Assumption limiting solar cell efficiency
Device principle overcoming this limitation
Single band gap energy
Multijunction solar cells
Quantum well, quantum dot solar cells
Down conversion
Multiple exciton generation
Avalanche multiplication
Up conversion
One e--h+ pair per photon
Non-use of sub-band-gap photons
Single population of each charge carrier type
One-sun incident intensity
Hot carrier solar cells
Intermediate-band solar cells
Quantum well, quantum dot solar cells
Concentrator solar cells
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
19
Theoretical Multijunction
Cell Efficiency
Detailed balance limit efficiency
Radiative recombination only
Series res. and shadowing, optimized grid spacing
Normalized to experimental efficiency
60%
3J
4J
Efficiency (%)
55%
50%
4J
3J
45%
4J
40%
3J
3J & 4J MM solar cells
35%
1
500
10
100
1000
10000
Incident Intensity (suns) (1 sun = 0.100 W/cm2)
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
20
Maximum Solar Cell
Efficiencies
Measured Theoretical
References
C. H. Henry, “Limiting efficiencies of ideal single and multiple energy gap terrestrial
solar cells,” J. Appl. Phys., 51, 4494 (1980).
W. Shockley and H. J. Queisser, “Detailed Balance Limit of Efficiency of p-n Junction
Solar Cells,” J. Appl. Phys., 32, 510 (1961).
J. H. Werner, S. Kolodinski, and H. J. Queisser, “Novel Optimization Principles and
Efficiency Limits for Semiconductor Solar Cells,” Phys. Rev. Lett., 72, 3851
(1994).
R. R. King et al., "Band-Gap-Engineered Architectures for High-Efficiency
Multijunction Concentrator Solar Cells," 24th European Photovoltaic Solar Energy
Conf., Hamburg, Germany, Sep. 21-25, 2009.
R. R. King et al., "40% efficient metamorphic GaInP / GaInAs / Ge multijunction solar
cells," Appl. Phys. Lett., 90, 183516 (4 May 2007).
M. Green, K. Emery, D. L. King, Y. Hishikawa, W. Warta, "Solar Cell Efficiency Tables
(Version 27)", Progress in Photovoltaics, 14, 45 (2006).
A. Slade, V. Garboushian, "27.6%-Efficient Silicon Concentrator Cell for Mass
Production," Proc. 15th Int'l. Photovoltaic Science and Engineering Conf., Beijing,
China, Oct. 2005.
R. P. Gale et al., "High-Efficiency GaAs/CuInSe2 and AlGaAs/CuInSe2 Thin-Film
Tandem Solar Cells," Proc. 21st IEEE Photovoltaic Specialists Conf., Kissimmee,
Florida, May 1990.
J. Zhao, A. Wang, M. A. Green, F. Ferrazza, "Novel 19.8%-efficient 'honeycomb'
textured multicrystalline and 24.4% monocrystalline silicon solar cells," Appl.
Phys. Lett., 73, 1991 (1998).
95%
93%
Carnot eff. = 1 – T/Tsun
T = 300 K, Tsun ≈ 5800 K
Max. eff. of solar energy conversion
= 1 – TS/E = 1 – (4/3)T/Tsun (Henry)
72%
Ideal 36-gap solar cell at 1000 suns
(Henry)
56%
50%
Ideal 3-gap solar cell at 1000 suns
Ideal 2-gap solar cell at 1000 suns
(Henry)
(Henry)
44%
43%
Ultimate eff. of device with cutoff Eg: (Shockley, Queisser)
1-gap cell at 1 sun with carrier multiplication
(>1 e-h pair per photon) (Werner, Kolodinski, Queisser)
37%
Ideal 1-gap solar cell at 1000 suns
31%
30%
Ideal 1-gap solar cell at 1 sun
(Henry)
Detailed balance limit of 1 gap solar cell at 1 sun
(Shockley, Queisser)
3-gap GaInP/GaInAs/Ge LM cell, 364 suns (Spectrolab) 41.6%
3-gap GaInP/GaInAs/Ge MM cell, 240 suns (Spectrolab) 40.7%
(Henry)
3-gap GaInP/GaAs/GaInAs cell at 1 sun (NREL) 33.8%
1-gap solar cell (silicon, 1.12 eV) at 92 suns (Amonix) 27.6%
1-gap solar cell (GaAs, 1.424 eV) at 1 sun (Kopin) 25.1%
1-gap solar cell (silicon, 1.12 eV) at 1 sun (UNSW) 24.7%
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
21
Metamorphic (MM) 3-Junction Cells
–– Inverted 1.0-eV GaInAs Subcell
Ge or GaAs substrate
Ge or GaAs substrate
Growth
Direction
cap
1.9 eV (Al)GaInP subcell 1
1.4 eV GaInAs subcell 2
graded MM buffer layers
1.0 eV GaInAs subcell 3
Growth on Ge or GaAs substrate,
followed by substrate removal from sunward surface
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
22
0.6
7
eV
0.9
eV 5
100
1.4
1
1.8
9
Solar Spectrum Partition for
3-Junction Cell
100
AM1.5D, low-AOD
AM1.5G, ASTM G173-03
AM0, ASTM E490-00a
80
90
80
70
70
60
60
50
50
40
40
30
30
20
20
10
10
0
300
500
700
900
1100
1300
1500
1700
External Quantum Efficiency (%)
Current Density per Unit
Wavelength (mA/(cm 2μm))
90
0
1900
Wavelength (nm)
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
23
5- and 6-Junction Cells
contact
AR
AR
cap
contact
AR
AR
cap
(Al)GaInP Cell 1
(Al)GaInP Cell 1
2.0 eV
wide-Eg tunnel junction
2.0 eV
wide-Eg tunnel junction
GaInP Cell 2 (low Eg)
1.8 eV
wide-Eg tunnel junction
AlGa(In)As Cell 2
1.7 eV
AlGa(In)As Cell 3
1.6 eV
wide-Eg tunnel junction
wide-Eg tunnel junction
Ga(In)As Cell 3
1.41 eV
Ga(In)As Cell 4
1.41 eV
tunnel junction
tunnel junction
GaInNAs Cell 4
1.1 eV
GaInNAs Cell 5
1.1 eV
tunnel junction
tunnel junction
Ga(In)As buffer
Ga(In)As buffer
nucleation
nucleation
Ge Cell 5
and substrate
0.67 eV
Ge Cell 6
and substrate
0.67 eV
back contact
back contact
• Divides available
current
density above GaAs Eg
among 3-4 subcells
• Allows low-current
GaInNAs cell to
be matched to
other subcells
• Lower series resistance
Ref.: U.S. Pat. No. 6,316,715, Spectrolab, Inc., filed 3/15/00, issued 11/13/01.
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
24
Photon Utilization Efficiency
3-Junction Solar Cells
500
1
400
0.8
300
0.6
200
0.4
100
0.2
0
Photon utilization efficiency
Intensity per Unit Photon Energy
(W/m 2 . eV)
600
.
AM1.5D, ASTM G173-03, 1000 W/m21.4
Utilization efficiency of photon energy
1-junction cell
3-junction cell
1.2
700
0
0
0.5
1
1.5
2
2.5
Photon Energy (eV)
3
3.5
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
4
25
Photon Utilization Efficiency
6-Junction Solar Cells
500
1
400
0.8
300
0.6
200
0.4
100
0.2
0
Photon utilization efficiency
Intensity per Unit Photon Energy
(W/m 2 . eV)
600
.
AM1.5D, ASTM G173-03, 1000 W/m2
1.4
Utilization efficiency of photon energy
1-junction cell
3-junction cell
1.2
6-junction cell
700
0
0
0.5
1
1.5
2
2.5
Photon Energy (eV)
3
3.5
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
4
26
3-Junction Cell Efficiency
Losses from 100%
300
200
100
0
0
0.5
1
80
70
39%
60
subcell 2
14%
26%
40
20
23%
1
0.8
0.6
10
26%
photon
energies
in solar
spectrum
after carrier
thermalization
to band edges
photons
with energy
above lowest
band gap
23%
11%
8%
5%
after carrier
extraction
at quasiFermi levels
-- 1-sun
7%
4%
more
25%
balanced
after carrier
extraction
at quasiFermi levels
-- 500 suns
1.9/1.4/1.0 eV
bandgap
combination
23%
6-junction
terrestrial
concentrator
cell
subcell 3
0.4
thermalization
of carriers
hν
0.2
3.5
20% E
h+
photon
energies
in solar
spectrum
photons
with energy
hν > Eg
above lowest
band gap
600
500
400
300
200
1.2
1
0.8
0.6
0.4
no photogeneration
100
0.2
0
0
0.5
Eg
1
1.5
2
2.5
Photon Energy (eV)
3
3.5
4
Ge or GaAs substrate
after carrier
thermalization
to band edges
700
600
500
400
300
200
8%
qφp
5%
Ge or GaAs substrate
after carrier
extraction
at quasiFermi levels
-- 1-sun
AM1.5D, ASTM G173-03, 1000 W/m2 1.4
Utilization efficiency of photon energy
to bandgap Eg = 1.424 eV
1.2
to Voc at 1000 suns
to Voc at 1 sun
1
0.8
0.6
0.4
100
0.2
11%
cap
AM1.5D, ASTM G173-03, 1000 W/m2
1.4
Utilization efficiency of photon energy
1-junction cell
3-junction cell
1.2
6-junction cell
12%
600
30
500
400
1
0.8
10%
300
0.6
20
200
0.4
8%
100
0
Growth
Direction
after carrier
more
extraction
1.9 eV (Al)GaInP subcellbalanced
1
at quasi1.9/1.4/1.0
eV
1.4 eV GaInAs subcell
2
Fermi levels
bandgap
graded MM buffer layers
-- 500 suns
combination
40
700
0
qφn
qV
Ev
AM1.5D, ASTM G173-03, 1000 W/m2 1.4
Utilization efficiency of photon energy
to bandgap Eg = 1.424 eV
700
0
15%
c
hν
Ev
4
17%
17%
e-
10
25% 0
50
14%
8%
15%
20%
10
20
17%
17%
31%
0
0
10%
subcell
2
20
Ec
60
30
subcell 3
31%
70
12%
23%
31%
50
40
1.2
hν < Eg
3
25%
25%
30
80
60
31%
50
0
90
70
no photogeneration
AM1.5D, ASTM G173-03, 1000 W/m2 1.4
1.5
2
2.5
Photon Energy (eV)
39%
7%1
0.5
0.2
10
1.5
2
2.5
Photon Energy (eV)
4%
6-junction
terrestrial
concentrator
cell
0
3
3.5
4
0
1.0 eV GaInAs subcell 3
0
0
0
0.5
1
1.5
2
2.5
Photon Energy (eV)
3
3.5
4
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
27
.
Remaining Fraction of Available Power (%)
30
80
subcell 1 (top)
100
Photon utilization efficiency
400
40
90
Intensity per Unit Photon Energy
(W/m 2 . eV)
500
50
subcell 1 (top)
Photon utilization efficiency
600
60
90
Intensity per Unit Photon Energy
2.
(W/m eV)
Intensity per Unit Photon Energy
2.
(W/m eV)
700
70
100
Photon Utilization Efficiency
Ev
80
100
Photon Utilization Efficiency
Ec
90
Intensity per Unit Photon Energy
(W/m 2 . eV)
hν
Remaining Fraction of Available Power (%)
100
Metamorphic
Semiconductor
Materials
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
28
Metamorphic (MM)
Semiconductor Materials
• Metamorphic = "changed form"
• Thick, relaxed epitaxial layers grown with different lattice
constant than growth substrate
• Allows access to subcell band gaps desired for more
efficient division of the solar spectrum in multijunction solar
cells
• Also called lattice-mismatched
• Misfit dislocations are allowed to form in metamorphic
buffer, which typically has graded composition and lattice
constant
• Threading dislocations which can propagate up into active
device layers grown on buffer are minimized as much as
possible
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
29
Bandgap vs.
Lattice Constant
Courtesy J. Geisz – NREL
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
30
Bandgap vs.
Lattice Constant
2
Direct Bandgap Eg (eV)
1.8
disordered GaInP
1.6
ordered GaInP
1.4
GaAs
1%-In
1.2
GaInAs
8%-In
12%-In
GaInAs
1
23%-In
GaInAs
0.8
35%-In
Ge
(indirect)
0.6
5.6
5.65
5.7
5.75
5.8
Lattice Constant (angstrom)
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
31
0.96
eV
1.08
1.30
1.26
1.40
1.38
Internal QE of Metamorphic
GaInAs Cells on Ge
Internal Quantum Efficiency (%)
100
90
80
70
60
GaInAs single-junction
solar cells
50
40
1.6% lattice
mismatch
2.4% lattice
mismatch
30
20
10
0
300
400
500
600
700 800 900 1000 1100 1200 1300 1400
Wavelength (nm)
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
32
Cross sectional TEM
Ga0.44In0.56P/ Ga0.92In0.08As/ Ge
Cell
• Low dislocation
density in active cell
layers in top portion of
epilayer stack:
from
~2x
EBIC and CL meas.
105
cm-2
• Dislocations confined
to graded buffer layers in
bottom portion of
epilayer stack
GaInAs cap
GaInP TC
Tunnel junction
GaInAs MC
0.2 μm
GaInAs graded
buffer to 8%-In
Misfit dislocations
Pre-grade buffer
Ge substrate
0.2 μm
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
33
High-Resolution XRD
Reciprocal Space Map (RSM)
• GaInP/ 8%-In GaInAs/ Ge
metamorphic (MM) cell
structure
Ge
Qy (Strain) Å-1
Graded
Buffer
Ga0.92In0.08As MC
GaInP TC
(115)
glancing
exit XRD
• 56%-In GaInP top cell
pseudomorphic with respect
to GaInAs middle cell
Qx (Tilt) Å-1
Line of 0%
relaxation
• Nearly 100% relaxed stepgraded buffer →
removes driving force for
dislocations to propagate
into active cell layers
Line of 100%
relaxation
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
34
Inverted Lattice-Matched (LM)
1.39-eV GaInAs Subcell
hν
metal
contact
1.39-eV GaInAs
inverted LM subcell
base
emitter
buffer layer
nucleation
Ge or GaAs substrate
Growth on Ge or GaAs substrate,
followed by substrate removal from sunward surface
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
35
Metamorphic (MM) 3-Junction Cells
–– Inverted 1.10-eV GaInAs Subcell
hν
metal
contact
1.10-eV GaInAs
inverted MM subcell
base
emitter
transparent MM
graded buffer layers
nucleation and pre-grade buffer
Ge substrate
Growth on Ge or GaAs substrate,
followed by substrate removal from sunward surface
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
36
Metamorphic (MM) 3-Junction Cells
–– Inverted 0.97-eV GaInAs Subcell
hν
metal
contact
0.97-eV GaInAs
inverted MM subcell
base
emitter
transparent MM
graded buffer layers
nucleation and pre-grade buffer
Ge substrate
Growth on Ge or GaAs substrate,
followed by substrate removal from sunward surface
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
37
Metamorphic (MM) 3-Junction Cells
–– Inverted 0.84-eV GaInAs Subcell
hν
metal
contact
0.84-eV GaInAs
inverted MM subcell
base
emitter
transparent MM
graded buffer layers
nucleation and pre-grade buffer
Ge substrate
Growth on Ge or GaAs substrate,
followed by substrate removal from sunward surface
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
38
Dislocations in Inverted
Metamorphic Cells – EBIC
8e-9766-1
50 μm
1.39-eV ILM subcell
GaInAs comp.
Latt. mismatch
Disloc. density
2% In
0.1%
2.5 x 105 cm-2
8e-9756-1
50 μm
0.97-eV IMM subcell
0.84-eV IMM subcell
23% In
1.6%
3.9 x 106 cm-2
33% In
2.3%
5.0 x 106 cm-2
44% In
3.1%
6.3 x 106 cm-2
metal
metal
contact
1.39-eV
GaInAs
base
8e-9783-11
50 μm
1.10-eV IMM subcell
contact
1.39-eV GaInAs
inverted LM subcell
8e-9760-1
50 μm
1.10-eV
GaInAs
1.10-eV GaInAs
inverted MM subcell
base
metal
metal
contact
contact
0.97-eV
GaInAs
buffer layer
transparent MM
graded buffer layers
nucleation
nucleation and pre-grade buffer
Ge or GaAs substrate
Ge substrate
base
base
emitter
emitter
emitter
emitter
0.84-eV
GaInAs
0.840.84-eV GaInAs
inverted MM subcell
0.970.97-eV GaInAs
inverted MM subcell
transparent MM
graded buffer layers
transparent MM
graded buffer layers
nucleation and pre-grade buffer
nucleation and pre-grade buffer
Ge substrate
Ge substrate
EBIC images and dislocation density of inverted metamorphic cell test structures
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
39
Dislocations in Inverted
Metamorphic Cells
Lattice Mismatch Relative to Ge (%)
0.64
1.36
2.07
2.79
3.51
Inverted metamorphic (MM)
GaxIn1-xAs solar cells
1.2
8
7
6
1.0
5
0.8
4
0.6
3
0.4
2
Band gap Eg meas. by ext. QE
Meas. Voc at ~1 sun
Woc = (Eg/q) - Voc = band gap-voltage offset
Dislocation density from EBIC
0.2
1
0.0
Dislocation Density from EBIC (10 6 cm -2)
Band Gap Eg, Open-Circuit Voltage Voc ,
and Band Gap-Voltage Offset (V)
1.4
-0.07
0
0
10
20
30
In Composition for Gax In1-x As (%)
40
50
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
40
Dislocations in Inverted
Metamorphic Cells
Lattice Mismatch Relative to Ge (%)
and Photon Intensity from CL (10 3 cps)
Dislocation Density from EBIC (10 6 cm -2)
9
0.64
1.36
2.07
2.79
3.51
Inverted metamorphic (MM)
GaxIn1-xAs solar cells
8
45
40
7
Dislocation density from EBIC
35
Overall % photon intensity from CL
6
50
% carrier loss at each dislocation from CL
30
5
25
4
20
3
Carrier Loss (%)
-0.07
15
2
10
1
5
0
0
0
10
20
30
In Composition for Gax In1-x As (%)
40
50
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
41
Solar Cell Voltage
Voltage depends on non-equilibrium concentrations of
electrons and holes
pn = n e
2 qV / kT
i
n = NC NV e
2
i
− E g / kT
kT ⎛ pn ⎞
V=
ln⎜⎜ 2 ⎟⎟
q ⎝ ni ⎠
pn = NC NV e
− ( E g − qV ) / kT
= NC NV e − qW / kT
kT ⎛ NC NV
W ≡ (Eg q ) − V =
ln⎜⎜
q ⎝ pn
⎞
⎟⎟
⎠
• Bandgap-voltage offset W ≡ (Eg/q) – V is a useful parameter for
gauging solar cell quality, especially when dealing with semiconductors
of many different bandgaps
• Basically a measure of how close electron and hole quasi-Fermi
levels are to conduction and valence band edges
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
42
Band gap - Voltage Offset (Eg/q) - Voc
for Single-Junction Solar Cells
Voc of solar cells with
wide range of bandgaps and
comparison to radiative limit
d-AlGaInP
d-AlGaInP
d-AlGaInP
d-GaInP
o-GaInP
AlGaInAs
AlGaInAs
1.4 - eV GaInAs
GaAs
1.30-eV GaInAs
1.24-eV GaInAs
0.5
1.10-eV GaInAs
1.0
GaInNAs
1.5
0.97-eV GaInAs
Voc
Eg from EQE
(Eg/q) - Voc
radiative limit
Ge (indirect gap)
Eg/q, Voc, and (Eg/q) - Voc (V)
2.0
0.0
0.6
1
1.4
Bandgap Eg (eV)
1.8
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
2.2
43
Band gap - Voltage Offset (Eg/q) - Voc
for Single-Junction Solar Cells
Voc
Eg from EQE
(Eg/q) - Voc
radiative limit
AM1.5D, low-AOD
800
700
1.5
600
d-AlGaInP
d-AlGaInP
d-GaInP
d-AlGaInP
o-GaInP
AlGaInAs
AlGaInAs
1.4 - eV GaInAs
GaAs
400
1.24-eV GaInAs
1.30-eV GaInAs
0.5
GaInNAs
1.10-eV GaInAs
1.0
0.97-eV GaInAs
500
300
200
100
0.0
Intensity per Unit Photon Energy (W/(m2 .eV))
Voc of solar cells with
wide range of bandgaps and
comparison to radiative limit
Ge (indirect gap)
Eg/q, Voc, and (Eg/q) - Voc (V)
2.0
0
0.6
1
1.4
1.8
2.2
Bandgap Eg (eV)
2.6
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
3
44
High-Efficiency
Multijunction Cells
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
45
LM and MM 3-Junction
Cell Cross-Section
contact
contact
AR
AR
n+-Ga(In)As
n-AlInP window
n-GaInP emitter
GaInP top cell
To
p-GaInP base
p-AlGaInP BSF
Wide-bandgap tunnel junction
p++-TJ
n++-TJ
W
E
eid
n-GaInP window
n-Ga(In)As emitter
Ga(In)As middle cell
p-Ga(In)As base
p-GaInP BSF
Tunnel junction
Buffer region
p++-TJ
M
id
g
dl
l
el
nn
u
T
e
el
C
p++-TJ
n++-TJ
l
tio
nc
u
lJ
W
n-GaInP window
n-GaInAs emitter
M
id
m
tto
o
B
el
nn
u
T
g
dl
l
e
el
C
l
n
n-Ga(In)As buffer
n+-Ge emitter
p
E
eid
p-GaInP BSF
p-GaInAs
step-graded
buffer
Ce
el
C
p-GaInP base
p-GaInAs base
e
nn
u
T
To
p-AlGaInP BSF
n++-TJ
nucleation
Ge bottom cell
p
el
C
n+-GaInAs
n-AlInP window
n-GaInP emitter
ll
p++-TJ
lJ
n++-TJ
nucleation
n+-Ge emitter
p-Ge base
and substrate
p-Ge base
and substrate
contact
contact
Lattice-Matched (LM)
e
nn
Tu
n
tio
c
un
m
tto
o
B
Ce
ll
Lattice-Mismatched
or Metamorphic (MM)
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
46
Metamorphic (MM)
3-Junction Solar Cell
0.3
n+-GaInAs
n-AlInP window
n-GaInP emitter
Ce
p
To
p-GaInP base
p-AlGaInP BSF
p++-TJ
n++-TJ
W
i
-E
de
n-GaInP window
n-GaInAs emitter
p-GaInAs base
M
g
Tu
e
dl
id
ll
el
nn
Ce
ll
p-GaInP BSF
p-GaInAs
step-graded
buffer
n
Tu
p++-TJ
n++-TJ
ne
nucleation
n+-Ge emitter
t
Bo
l
n
t io
nc
u
J
m
to
l
Ce
l
p-Ge base
and substrate
contact
C urrent D ensity / Incident Intensity (A /W )
contact
AR
MJ cell
0.25
subcell 1
subcell 2
0.2
subcell 3
0.15
0.1
0.05
0
0
0.5
1
1.5
2
2.5
3
3.5
Voltage (V)
Lattice-Mismatched
or Metamorphic (MM)
• Metamorphic growth of upper two subcells, GaInAs and GaInP
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
47
External QE of LM and MM
3-Junction Cells
100
AM1.5D, low-AOD
AM1.5G, ASTM G173-03
90
AM0, ASTM E490-00a
90
Current Density per Unit
Wavelength (mA/(cm 2μm))
EQE, lattice-matched
80
EQE, metamorphic
80
70
70
60
60
50
50
40
40
30
30
20
20
10
10
0
300
500
700
900
1100
1300
1500
1700
External Quantum Efficiency (%)
100
0
1900
Wavelength (nm)
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
48
Metamorphic (MM)
3-Junction Solar Cell
Eg1 = Subcell 1 (Top) Bandgap (eV) .
2.1 3-junction Eg1/ Eg2/ 0.67 eV cell efficiency
240 suns (24.0 W/cm2), AM1.5D (ASTM G173-03), 25oC
2 Ideal efficiency -- radiative recombination limit
1.9
MM
40.7%
1.8
LM
40.1%
54%
1.7
52%
1.6
50%
48%
1.5
46%
44%
1.4
42%
40%
1.3
1.0
1.1
1.2
1.3
1.4
38%
1.5
1.6
Eg2 = Subcell 2 Bandgap (eV)
Disordered GaInP top subcell
Ordered GaInP top subcell
• Metamorphic GaInAs and GaInP subcells bring band gap combination
closer to theoretical optimum
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
49
Record 40.7%-Efficient
Concentrator Solar Cell
Spectrolab
Metamorphic
GaInP/ GaInAs/ Ge Cell
Voc
Jsc
FF
Vmp
=
=
=
=
2.911 V
3.832 A/cm2
87.50%
2.589 V
• First solar cell of
any type to reach
over 40% efficiency
Efficiency = 40.7% ± 2.4%
240 suns (24.0 W/cm2) intensity
0.2669 cm2 designated area
25 ± 1°C, AM1.5D, low-AOD spectrum
Ref.: R. R. King et al., "40% efficient
metamorphic GaInP / GaInAs / Ge
multijunction solar cells,"
Appl. Phys. Lett., 90, 183516, 4 May 2007.
Concentrator cell light I-V and efficiency independently verified by J. Kiehl, T. Moriarty, K. Emery – NREL
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
50
New World Record
41.6% Multijunction Solar Cell
• 41.6% efficiency demonstrated for 3J
lattice-matched Spectrolab cell, a new
world record
• Highest efficiency for any type of solar
cell measured to date
• Independently verified by National
Renewable Energy Laboratory (NREL)
• Standard measurement conditions
(25°C, AM1.5D, ASTM G173 spectrum) at
364 suns (36.4 W/cm2)
• Lattice-matched cell structure similar
to C3MJ cell, with reduced grid
shadowing as planned for C4MJ cell
Ref.: R. R. King et al., 24th European Photovoltaic Solar
Energy Conf., Hamburg, Germany, Sep. 21-25, 2009.
• Incorporating high-efficiency 3J
metamorphic cell structure + further
improvements in grid design
→ strong potential to reach 42-43%
champion cell efficiency
Concentrator cell light I-V and efficiency independently verified by C. Osterwald, K. Emery – NREL
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
51
41.6% Solar Cell
Eff., Voc vs. Concentration
Efficiency
Efficiency (%) and Voc x 10 (V)
41.6%
Voc x 10
42
0.98
0.96
Voc fit, 100 to 1000 suns
40
0.94
FF
38
0.92
36
0.90
34
0.88
32
0.86
30
0.84
28
0.82
26
0.80
24
0.78
1000.0
0.1
1.0
10.0
100.0
Fill Factor (unitless)
44
Incident Intensity (suns) (1 sun = 0.100 W/cm2)
• At peak 41.6% efficiency → 364 suns, Voc = 3.192 V, FF = 0.887
• Efficiency still >40% at 820 suns, at 940 suns efficiency is 39.8%
• Diode ideality factor of 1.0 for all 3 junctions fits Voc well from 100 to 1000 suns
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
52
41.6% Solar Cell
LIV Curves vs. Concentration
Current Density / Incident Intensity (A/W)
0.16
41.6%
0.14
Inc. Intensity (suns)
2
1 sun = 0.100 W/cm
0.12
2.6
0.1
6.6
0.08
17.6
59.8
0.06
127.3
364.2
0.04
604.8
0.02
940.9
0
0
0.5
1
1.5
2
2.5
3
3.5
Voltage (V)
• At peak 41.6% efficiency → 364 suns, Voc = 3.192 V, FF = 0.887
• Series resistance causes drop in Vmp above 400 suns, Voc continues to increase
• Efficiency still >40% at 820 suns, at 940 suns efficiency is 39.8%
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
53
Best Research Cell
Efficiencies
Chart courtesy of Larry Kazmerski, NREL
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
54
Inverted Metamorphic (IMM)
3-Junction Cell
Ge or GaAs substrate
Ge or GaAs substrate
Growth
Direction
cap
1.9 eV (Al)GaInP subcell 1
1.4 eV GaInAs subcell 2
graded MM buffer layers
1.0 eV GaInAs subcell 3
Growth on Ge or GaAs substrate,
followed by substrate removal from
sunward surface
Current Density / Incident Intensity (A/W )
0.16
0.14
0.12
0.1
MJ cell
subcell 1
0.08
subcell 2
0.06
subcell 3
0.04
0.02
0
0
0.5
1
1.5
2
2.5
3
3.5
4
Voltage (V)
• Bottom ~1-eV GaInAs subcell is inverted and metamorphic (IMM)
• Upper two GaInAs and GaInP subcells are inverted and lattice matched (ILM)
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
55
Inverted Metamorphic (IMM)
3-Junction Cell
1.6
3-junction 1.9 eV/ Eg2/ Eg3 cell efficiency
2
o
Eg2 = Subcell 2 Bandgap (eV) .
500 suns (50 W/cm ), AM1.5D (ASTM G173-03), 25 C
X
Ideal efficiency -- radiative recombination limit
1.5
1.4
53%
52%
51%
1.3
50%
48%
1.2
46%
1.1
44%
1
0.5
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
Eg3 = Subcell 3 Bandgap (eV)
• Raising band gap of bottom cell from 0.67 for Ge to ~1.0 eV for IMM
GaInAs raises theoretical 3J cell efficiency
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
56
5-Junction Inverted Metamorphic
(IMM) Cells
gro Ge o
wt r G
h s aA
ub s
str
ate
metal
gridline
2.0-eV AlGaInP cell 1
1.7-eV AlGaInAs cell 2
1.4-eV GaInAs cell 3
transparent buffer
1.1-eV GaInAs cell 4
transparent buffer
0.75-eV GaInAs cell 5
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
57
4-Junction
Lattice-Matched Cell
AR
AR
cap
(Al)GaInP Cell 1
1.9 eV
wide-Eg tunnel junction
AlGa(In)As Cell 2
1.6 eV
wide-Eg tunnel junction
Ga(In)As Cell 3
1.4 eV
tunnel junction
Ga(In)As buffer
nucleation
Ge Cell 4
and substrate
0.67 eV
Current Density / Incident Intensity (A/W )
0.25
contact
MJ cell
subcell 1
0.2
subcell 2
subcell 3
0.15
subcell 4
0.1
0.05
0
0
1
back contact
2
3
4
5
Voltage (V)
• Current density in spectrum above Ge cell 4 is divided 3 ways among
GaInAs, AlGa(In)As, GaInP cells
•Lower current and I2R resistive power loss
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
58
4-Junction Upright Metamorphic
(MM) Terrestrial Concentrator Cell
metal
gridline
1.8-eV (Al)GaInP cell 1
1.55-eV AlGaInAs cell 2
1.2-eV GaInAs cell 3
transparent buffer
0.67-eV Ge cell 4
and substrate
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
59
4-Junction Cell
Optimum Band Gap Combinations
1.7
4-junction 1.9 eV/ Eg2/ Eg3/ 0.67 eV cell efficiency
Eg2 = Subcell 2 Bandgap (eV) .
2
1.6
o
500 suns (50 W/cm ), AM1.5D (ASTM G173-03), 25 C
X
Ideal efficiency -- radiative recombination limit
1.5
58%
1.4
56%
1.3
54%
1.2
50%
38%
46%
1.1
34%
42%
1
0.8
0.9
1.0
1.1
1.2
1.3
1.4
1.5
Eg3 = Subcell 3 Bandgap (eV)
• Lowering band gap of subcells 2 and 3, e.g., with MM materials, gives
higher theoretical 4J cell efficiency
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
60
100
1600
90
1400
80
1200
70
AlGaInP subcell 1
1.95 eV
GaInAs subcell 3
1.39 eV
All subcells
60
50
AlGaInAs subcell 2
1.66 eV
Ge subcell 4
0.72 eV
AM1.5D
ASTM G173-03
1000
800
40
600
30
400
20
Intensity Per Unit Wavelength
(W/(m2μ m))
External Quantum Efficiency (%)
Measured 4-Junction Cell
Quantum Efficiency
200
10
0
300
0
500
700
900
1100 1300
Wavelength (nm)
1500
1700
1900
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
61
Light I-V Curves
Record Efficiency Cells
Current Density / Inc. Intensity (A/W) .
0.16
0.14
0.12
0.10
3J Conc.
Cell
0.08
3J Conc.
Cell
Metamorphic
V oc
Jsc /inten.
V mp
FF
conc.
area
0.06
0.04
Lattice-matched
2.911
0.1596
2.589
0.875
240
0.267
LM, 822 suns
3.192 V
0.1467 A/W
2.851 V
0.887
364
suns
0.317 cm2
Eff. 40.7%
41.6%
AM1.5D,
low -AOD spectrum
0.02
3J Conc.
Cell
3.251
0.1467
2.781
0.841
822
0.317
36.9%
AM1.5D,
ASTM G173-03
Independently confirmed meas.
25°C
0.00
0
0.5
1
1.5
4J Cell
4.398 V
0.0980 A/W
3.950 V
0.856
500 suns
0.208 cm2
40.1%
AM1.5D,
ASTM G173-03
4J Conc.
Cell
2
AM1.5D,
ASTM G173-03
Prelim. meas.
25°C
2.5
3
3.5
4
4.5
Voltage (V)
• Light I-V curves for 3-junction upright MM (40.7%), 3J lattice-matched (41.6%),
3J lattice-matched at 822 suns (39.1%), and 4J lattice-matched cell (36.9%)
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
62
Semiconductor-Bonded Technology
(SBT) Terrestrial Concentrator Cell
• Wafer bonding for multijunction solar cells
– Low band gap cells for MJ
cells using high-quality,
lattice-matched materials
– Epitaxial exfoliation and
substrate removal
– Formation of latticeengineered substrate for
later MJ cell growth
– Bonding of high-band-gap
and low-band-gap
cells
after
1.4-eV GaInAs
cell
3
growth
1.7-eV conductance
AlGaInAs cellof
2
– Electrical
semiconductor-bonded
2.0-eV AlGaInP cell 1
interface
– Surface effects
forGe
GaAs or
semiconductor-togrowth substrate
semiconductor bonding
semiconductor
bonded
interface
GaAs or
Ge
metal
gridline
growth
substrate
GaAs
or Ge
growth substrate
2.0-eV AlGaInP cell 1
2.0-eV AlGaInP cell 1
1.7-eV AlGaInAs cell 2
1.7-eV AlGaInAs cell 2
1.4-eV GaInAs cell 3
1.4-eV GaInAs cell 3
1.1-eV GaInPAs cell 4
0.75-eV GaInAs cell 5
InP growth substrate
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
63
6-Junction Solar Cells
0.14
AR
AR
cap
(Al)GaInP Cell 1
2.0 eV
wide-Eg tunnel junction
GaInP Cell 2 (low Eg)
1.78 eV
wide-Eg tunnel junction
AlGa(In)As Cell 3
1.50 eV
wide-Eg tunnel junction
Ga(In)As Cell 4
1.22 eV
tunnel junction
GaInNAs Cell 5
0.98 eV
tunnel junction
Ga(In)As buffer
nucleation
Ge Cell 6
and substrate
0.67 eV
Current Density / Incident Intensity (A/W )
contact
0.12
0.1
0.08
MJ cell
0.06
subcell 1
subcell 2
subcell 3
0.04
subcell 4
subcell 5
0.02
subcell 6
0
0
1
2
3
4
5
6
7
Voltage (V)
back contact
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
64
Photon Utilization Efficiency
6-Junction Solar Cells
500
1
400
0.8
300
0.6
200
0.4
100
0.2
0
Photon utilization efficiency
Intensity per Unit Photon Energy
(W/m 2 . eV)
600
.
AM1.5D, ASTM G173-03, 1000 W/m2
1.4
Utilization efficiency of photon energy
1-junction cell
3-junction cell
1.2
6-junction cell
700
0
0
0.5
1
1.5
2
2.5
Photon Energy (eV)
3
3.5
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
4
65
Concentrator
Photovoltaic (CPV)
Systems
and
Economics
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
66
Concentrator PV Systems
with Multijunction Cells
• 1 football field of ~ 17% solar cells at 1-sun produces ~ 500 kW.
• By using MJ cells (> 35%) at concentration of 500 suns, same power is
produced from smaller semiconductor area (or the football field
produces 500 MW).
Combination of high efficiency & 500X
concentration boosts output per
semiconductor area by a factor of
1000.
MJ cells are replaced by less expensive
optics and common materials.
Leads to reduced cost of energy despite
paying extra for tracking & cooling.
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
67
Solar Systems, Australia
Hermannsburg Power Station
• III-V MJ cells give
56% measured
improvement in
module efficiency
relative to Si
concentrator cells
Equipped
with III-V MJ
cell receivers
Courtesy of Solar Systems Pty. Ltd., Australia
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
68
Balance of System Costs
Optics
Cooling
Tracking
Structure
Operation
and
Maintenance
Courtesy of Solar Systems Pty. Ltd., Australia
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
69
Economics for
Device Physicists
Continuity equation:
∂ρ
∂t
= qG − qR − ∇ ⋅ J
...in $$ rather than charge carriers:
∂$
ρ$
= $$
qG
qR
J$$
gen − $$
exp − ∇ ⋅ F
∂t
change in
value of
=
PV system
(profit or loss)
→
⎛ PV system cost
⎞
⎜
⎟
⎜ per kWh generated in ⎟ =
⎜ 5 year payback period⎟
⎝
⎠
value of kWhr
generated
by PV system
–
operating
expenses
–
for PV system
funds paid out to bank
for interest and principal
on loan to buy PV system
⎛ module ⎞ ⎛ tracking ⎞ ⎛ BOS, area⎞ ⎛ BOS, power ⎞⎛ peak power ⎞ ⎛ cell ⎞
⎜
⎟+⎜
⎟+⎜
⎟+⎜
⎟⎜
⎟+⎜
⎟
⎜ cost / m 2 ⎟ ⎜ cost / m 2 ⎟ ⎜ cos t / m 2 ⎟ ⎜ cos t / W
⎟⎜ output, W/m 2 ⎟ ⎜ cost / m 2 ⎟
⎝
⎠ ⎝
⎠ ⎝
⎠ ⎝
⎠⎝
⎠ ⎝
⎠
5 year
⎛ energy produced, ⎞⎛
⎞
⎜
⎟⎜
⎟
⎜ kWh m 2 ⋅ year ⎟⎜ payback period ⎟
⎝
⎠⎝
⎠
(
⎛ conc. ⎞
⎜
⎟
⎜ ratio ⎟
⎝
⎠
)
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
70
Terrestrial PV System Cost
vs. Cell Cost
Fixed Flat-Plate
PV System Cost / kWh Generated
in 5 Year Period ($/kWh)
3.3
500X Point-Focus
Conc.
Cell cost ranges
0.2
2.2
10%
5-Year Payback Threshold, at $0.15/kWh
0.1
20%
15%
20%
30%
25% cell eff.
40%
50% cell eff.
Decreasing cell cost main priority
for flat-plate
Increasing cell efficiency main
priority for concentrators
⎛ module ⎞ ⎛ tracking ⎞ ⎛ BOS, area⎞ ⎛ BOS, power ⎞⎛ peak power ⎞ ⎛ cell ⎞
⎟
⎟+⎜
⎟⎜
⎟+⎜
⎟+⎜
⎟+⎜
⎛ PV system cost
⎞ ⎜⎜
⎟⎜ output, W/m 2 ⎟ ⎜ cost / m 2 ⎟
⎜
⎟ ⎝ cost / m 2 ⎟⎠ ⎜⎝ cost / m 2 ⎟⎠ ⎜⎝ cos t / m 2 ⎟⎠ ⎜⎝ cos t / W
⎠
⎠ ⎝
⎠⎝
⎜ per kWh generated in ⎟ =
5 year
⎞
⎛ energy produced, ⎞⎛
⎜ 5 year payback period⎟
⎟
⎜
⎟
⎜
⎝
⎠
⎜ kWh m 2 ⋅ year ⎟⎜ payback period ⎟
⎠
⎠⎝
⎝
(
0.0
0.001
0.01
1.1
⎛ conc. ⎞
⎟
⎜
⎜ ratio ⎟
⎠
⎝
)
0.1
1
PV System Cost / Power Output ($/W)
0.3
0.0
10
2
Cell Cost ($/cm )
R. R. King et al., 3rd Int'l. Conf. on Solar Concentrators (ICSC-3), Scottsdale, AZ, May 2005
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
71
Larry Kazmerski, NREL
Larry Kazmerski, NREL
Larry Kazmerski, NREL
Summary
• Urgent global need to address carbon emission, climate change, and
energy security concerns → renewable electric power can help
• Theoretical solar conversion efficiency
– Examining built-in assumptions points out opportunities for higher PV efficiency
– Multijunction architectures, up/down conversion, quantum structures,
intermediate bands, hot-carrier effects, solar concentration → higher η
– Theo. solar cell η > 70%, practical η > 50% achievable
• Metamorphic multijunction cells have begun to realize their promise
– Metamorphic semiconductors offer vastly expanded
of band gaps
– 40.7% metamorphic GaInP/ GaInAs/ Ge 3J cells demonstrated
– First solar cells of any type to reach over 40% efficiency
• New world record efficiency of 41.6% demonstrated
– Highest efficiency yet measured for any type of solar cell
– 41.6% efficiency independently verified at NREL (364 suns, 25°C, AM1.5D)
• Solar cells with efficiencies in this range can transform the way we
generate most of our electricity, and make the PV market explode
R. R. King, Stanford Photonics Research Center Symposium, Stanford, CA, Sep. 14-16, 2009
75