ED-27
, 692 (1980).
73. Bergmann R
et al
.,
Semicond. Sci. Technol.
12
, 224 (1997).
74. Schropp R, Zeman M,
Amorphous and Microcrystalline Silicon Solar Cells: Modeling, Mate-
rials and Device Technology
, Kluwer Academic Publications, Boston/Dordrecht (1988).
75. Baba T
et al
.,
MRS Proc.
Fall
1994, 895 (1995).
76. Haque M, Naseem H, Brown W,
J. Appl. Phys.
75
, 3928 (1994).
77. Tu K,
Appl. Phys. Lett.
27
, 221 (1975).
78. Ashtikar M, Sharma G,
J. Appl. Phys.
78
, 913 (1995).
79. Masaki Y, Ogata T, Ogawa H, Jones D,
J. Appl. Phys.
76
, 5225 (1994).
80. Sok-Woon Lee
et al
.,
IEEE Electron Device Lett.
17
(8), 407 (1996).
81. Sopori B
et al
.,
Mat. Res. Soc. Symp. Proc.
470
, 419 (1997).
82. Cullity B,
Elements of X-Ray Diffraction
, Addison-Wesley, Reading, MA (1956).
83. Chen W
et al
.,
MRS Meeting
, (April 2001).
84. Ishihara T
et al
.,
Mat. Res. Soc. Symp. Proc.
485
, 3 (1997).
85. Cao M
et al
.,
IEEE Trans. Electron Devices
ED-43
, 561 (1996).
86. Efremov M
et al
.,
Proc. SPIE
2801
, 263 (1996).
87. Sopori B,
J. Electron. Mater.
31
, 972 (2002).
88. Narayanan S, Wenham S, Green M,
IEEE Trans.
ED-37
, 382 (1990).
89. Joshi S, Gosele U, Tan T,
J. Appl. Phys.
77
, 3858 (1995).
90. Gilles D, Weber E, Hahn S,
Phys. Rev. Lett.
64
, 196 (1990).
91. Sana P, Rohatgi A, Kalejs J, Bell R,
Appl. Phys. Lett.
64
, 97 (1994).
92. Sopori B, Jastrzebski L, Tan T, Narayanan S,
Proc. 12
th
Photovoltaic Science and Engineer-
ing Conf.
, 1003 (1994).
93. Istratov A, Hieslmair H, Weber E,
Appl. Phys.
A69
, 13 (1999).
94. Perichaud I, Martinuizi S,
Proc. 22
nd
IEEE Photovoltaic Specialists conf.
, 877 (1991).
95. Sopori B, Chen W, Tan T, Plekhanov P,
AIP Conf. /Proc.
462
, 341 (1998).
96. Sopori B, Zhang Y, N Ravindra,
J. Electron. Mater.
30
, 1616 (2001).
97. Sopori B
et al
.,
Sol. Energy Mater. Sol. Cells
41/42
, 159 (1996).
REFERENCES
357
98. Symko M, Sopori B, Reedy R, Jones K, in Davies G, Nazare M, Eds,
Proc. ICDS-19
, 527,
Tran. Tech. Publ. Aveiro, Portugal (1997).
99. Sopori B,
Appl. Phys. Lett.
52
, 1718 (1988).
100. Sopori B,
J. Appl. Phys.
64
, 5264 (1988).
101. Estreicher S, Hastings J, Fedders P,
Phys. Rev. B
57
, 12663 (1998).
102. Zhang Y,
Modeling Hydrogen Diffusion for Solar Cell Passivation and Process Optimization
,
Ph.D. thesis, New Jersey Institute of Technology, NJ (2002).
This Page Intentionally Left Blank
9
High-Efficiency III-V Multijunction
Solar Cells
J. M. Olson, D. J. Friedman and Sarah Kurtz
National Renewable Energy Laboratory, Golden, CO
9.1 INTRODUCTION
The large-scale use of photovoltaics is slowly becoming a reality. Small scale
(
∼
10–20 kW) power systems using Si solar cells now compete with fossil-fueled electric
generators for remote applications, where “remote” in the United States means less than
one kilometer from the electrical grid. The total worldwide solar cell production in the
year 2000 was 0.3 GW, mostly in the form of flat-plate Si solar cells. Compared to the PV
production capacity 20 years ago, this represents remarkable progress. Silicon solar cells
have reached efficiencies exceeding 20%, and the cost has been reduced to under $10/W.
However, in the context of world energy consumption, 0.3 GW is a miniscule number.
The problem is related to the diffuse nature of solar radiation. For example, to generate
1 GW of electrical power using Si solar cells requires an aperture area on the order of
10
7
m
2
. The main problem is not the land area, but the daunting task of producing 10
7
m
2
of what has been termed
solar-grade silicon
, which in reality is virtually indistinguishable
from semiconductor-grade silicon. One solution to this problem is to use “concentrator
technology.” Here, lenses or mirrors focus the sunlight (usually the direct portion) on a
smaller solar cell. The concentration ratio can be as large as 200X to 300X for Si and
1000X to 2000X for a GaAs solar cell. At these concentration ratios, the cost of the cell
becomes less important than its efficiency. For example, a GaInP/GaAs/Ge tandem cell
with an efficiency of 34% at 1000X and a cost of $10/cm
2
may be more cost-effective
than a Si concentrator cell with an efficiency of 28% at 200X and a cost of $0.50/cm
2
.
The trade-offs are complex and currently not well quantified, but it seems clear that
concentrator photovoltaics must be a dominant player if photovoltaics have to supply a
significant fraction of the world’s energy needs.
Handbook of Photovoltaic Science and Engineering
. Edited by A. Luque and S. Hegedus
2003 John Wiley & Sons, Ltd
ISBN: 0-471-49196-9
360
HIGH-EFFICIENCY III-V MULTIJUNCTION SOLAR CELLS
n
/
p
GaInP top cell
E
g
=
1.85 eV
Front grids
Au back contact
Antireflection
coat
p
/
n
tunnel junction
n
/
p
GaAs bottom cell
E
g
=
1.42 eV
GaAs or Ge substrate
h
n
Figure 9.1
Schematic of GaInP/GaAs multijunction solar cell. When grown on a Ge substrate,
there is an option for introducing a third junction in the Ge substrate, thus boosting the voltage and
efficiency of the overall device. Dimensions are not to scale
It was in this context that researchers at the National Renewable Energy Laboratory
(NREL) conceived and began work on the GaInP/GaAs tandem solar cell more than
a decade ago [1]. A schematic of the cell is shown in Figure 9.1. The cell consists
of a Ga
x
In
1
−
x
P top cell (with a band gap of 1.8–1.9 eV) grown monolithically on a
lattice-matched interconnecting tunnel junction and a GaAs bottom cell. As shown in
Figure 9.2, for x
≈
0
.
5, Ga
x
In
1
−
x
P has the same lattice constant as GaAs with a band
gap energy between 1.8 and 1.9 eV. Prior to this, several groups were working on tandem
device designs that theoretically should achieve efficiencies approaching 36 to 40% [2].
These included mechanical stacks of a high band gap top cell on a Si bottom cell and
monolithic combinations of AlGaAs, GaAs, and GaInAs or GaAsP on Si. However, the
mechanical stacks were viewed as too costly and cumbersome (perhaps prematurely).
The defects generated by the lattice mismatch between top and bottom cells in some
of the monolithic structures (i.e. GaAs and GaInAs or GaAsP and Si) were a problem
that could not be solved easily. The AlGaAs/GaAs tandem cell is lattice matched with
a theoretical efficiency of 36% [2]. However, the sensitivity of AlGaAs to trace levels
of oxygen in all growth systems and source materials made it difficult to produce high
yield and, thus, limited its use in a production environment. The novel NREL idea was
to trade manufacturability (i.e. lattice-matched top and bottom cells and oxygen-tolerant
device materials) for a slightly lower theoretical efficiency of 34%.
By most standards, progress was rapid (see Figure 9.3). Despite initial problems
with the growth of GaInP due to metalorganic chemical vapor deposition (MOCVD) and
complications associated with an anomalous red shift of the band gap energy, by 1988
reasonably good GaInP top cells could be fabricated [3–5]. In 1990, efficiencies greater
INTRODUCTION
361
2.4
2.0
1.6
1.2
0.8
0.4
0.0
Band gap
[eV]
6.2
6.0
5.8
5.6
5.4
Lattice constant
[Å]
AlAs
AlSb
GaP
GaAs
GaSb
InP
InAs
Ge
Si
X
Γ
L
Figure 9.2
Estimated band gap as a function of lattice constant for Si, Ge, III-V binaries and
their alloys
40
30
20
10
0
2000
1995
1990
1985
Year
Ef
ficienc
y
[%]
Two-junction
practical limit
NREL Invention
of GaInP/GaAs
solar cell
Patent
issued
Tandem-
powered
satellite
flown
Commercial
production of
tandem
Production
levels reach
300 kW y
−
1
Three-junction concentrator
Figure 9.3
These GaInP/GaAs cell efficiencies were measured at one sun with the AM1.5
global spectrum. The triangles were measured under concentrated sunlight for three-junction
GaInP/GaAs/Ge cells
than 27% one-sun air mass 1.5 global (AM1.5G) were achieved by changing the top-cell
thickness to achieve current matching [6, 7]. This tuning of the top-cell thickness can also
be used to achieve current matching under different solar spectra, for example, AM0 and
AM1.5direct (AM1.5D). Using this feature of the GaInP/GaAs tandem solar cell, NREL,
over the next three years, set records at AM1.5G with an efficiency
η
=
29
.
5% [8], at
160-suns AM1.5D with
η
=
30
.
2%, [9] and at one-sun AM0 with
η
=
25
.
7% [10]. Soon,
numerous laboratories around the world were studying this device, and the 29.5% record
was eventually eclipsed by researchers at the Japan Energy Corporation with an efficiency
Do'stlaringiz bilan baham: |