EXTENSIBLE MODELLING FRAMEWORK FOR
NANOSTRUCTURED III-V SOLAR CELLS
Markus F Führer
1
, Jessica GJ Adams
1
, Keith WJ Barnham
1
, Ben C Browne
1
, Ngai LA Chan
1
, Daniel J Farrell
1
,
Louise Hirst
1
, Kan-Hua Lee
1
, Ned J Ekins-Daukes
1
, Akio Ogura
2
, Katsuhisa Yoshida
2
, Yoshitaka Okada
2
1
Imperial College London, London, United Kingdom
2
Research Center for Advanced Science and Technology, The University of Tokyo, Tokyo, Japan
ABSTRACT
The use of nanostructures has been shown to provide
practical performance enhancements to high-efficiency III-
V based solar cells by permitting sub-bandgap tuneable
absorption. Nanostructures present a fertile ground for
new solar cell technologies, and an improved
understanding of fundamental processes may even lead to
functional intermediate band and hot-carrier devices. As
the fundamental processes occurring in nanostructured
solar cells are complex and not easily observable, the
study of such devices often requires the analysis of data
derived from experimental characterisation techniques
using computer models. Models exist for many individual
aspects of these nanostructured solar cells, but as yet no
comprehensive modelling solution exists. We report on our
progress to produce an extendable abstract modelling
framework written in the high-level programming language
Python. The framework is intended for deployment both as
back-end to a variety of interfaces for specialised
modelling purposes, and as a library of methods and
classes for use at source-code level, allowing adaptation
to a wide variety of research problems. Significant code
abstraction, such as sequestering complex materials
parameterisation behind a simple material object allows
simple scripts to do complex work. Modules underway
cover several device simulation tiers, including
fundamental processes such as quantum well and dot
absorption and recombination, as well as device level
simulations such as spatial bias mapping using equivalent
circuits and multijunction IV characteristics. These
simulations correlate with and derive experimental data
from characterisation techniques including spatially and
temporally resolved electro- and photoluminescence
spectroscopy, fourier-transform infrared spectroscopy, and
others.
NANOSTRUCTURED III-V SOLAR CELLS
The incorporation of nanostructures such as quantum
wells or quantum dots into the intrinsic region of p-i-n solar
cells has been studied extensively. Recent results show
that directional emission from strained quantum wells
provides a small but fundamental efficiency enhancement
to such devices [1]. An improved understanding of the
fundamental processes in these structures may even lead
to functional high efficiency cell concepts, such as hot-
carrier devices, wherein above-bandgap carriers are
collected prior to thermalisation, and intermediate-band
devices, where an additional band in the band-gap permits
sub-bandgap absorption as a two step process.
Preliminary evidence of non-equilibrium carrier
populations suggesting delayed cooling is actively being
investigated [2,3].
A more immediate benefit to nanostructures comes in the
form of subcell bandgap tuning in multijunction solar cells:
the inclusion of lower bandgap regions in current-limiting
junctions trades extra current for a slight reduction in
junction voltage, allowing the other junctions to operate
closer to their maximum power point [4], as well as
permitting multijunction cells to be tuned to particular
insolation spectra. The recent world-record efficiency
achieved by a single-junction quantum well solar cell
shows that even single-junction nanostructured devices
show promise. [5]
Nanostructure Example: Quantum Well Solar Cells
The production of quantum well solar cells involves
growing several interleaved layers of different band-gap
lattice-matched materials. There are no binary or ternary
alloys with a suitable band-gap that are lattice-matched to
GaAs. One candidate, InGaAs, satisfies the band-gap
constraints, but its larger lattice constant limits the number
of quantum wells; only a few can be grown under
compressive strain before relaxation occurs and
dislocations form. To counter this problem, it is possible to
space the quantum wells not with GaAs, but rather with
another material with a smaller lattice constant, such as
Figure 1: Quantum Well Solar Cell band structure
978-1-4244-9965-6/11/$26.00 ©2011 IEEE 002615