Optimisation of Coated Nanowire Solar

Cells by Simulation

A thesis submitted to Imperial College London

for the degree of Doctor of Philosophy in Physics
2015

Thomas Edwards

Centre for Doctoral Training in

Theory and Simulation of Materials,
Department of Physics

March 21, 2016

Contents

1 Introduction 19
1.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
1.1.1 A Brief History of Solar Cells . . . . . . . . . . . . . . . . . . . . . . . 19
1.1.2 Development of Solar Cells . . . . . . . . . . . . . . . . . . . . . . . . . 21
1.1.2.1 First-generation: Crystalline Silicon . . . . . . . . . . . . . . . 21
1.1.2.2 Second-generation: Thin-film, Amorphous Silicon and Poly-
crystalline Silicon . . . . . . . . . . . . . . . . . . . . . . . . . 21
1.1.2.3 Third generation: Organic Solar Cells . . . . . . . . . . . . . . 22
1.1.3 Improving the Efficiency of Emerging PV . . . . . . . . . . . . . . . . . 23
1.1.3.1 Bilayer Organic Solar Cells . . . . . . . . . . . . . . . . . . . . 23
1.1.3.2 Bulk Heterojunction Devices . . . . . . . . . . . . . . . . . . . 23
1.1.3.3 Hybrid Organic/Inorganic Solar Cells . . . . . . . . . . . . . . 23
1.1.3.4 Dye-Sensitised Solar Cells . . . . . . . . . . . . . . . . . . . . . 24
1.1.3.5 Solid-state Sensitised and Extremely Thin Absorber Solar Cells 24
1.1.3.6 Perovskite Solar Cells . . . . . . . . . . . . . . . . . . . . . . . 26
1.1.4 Nanowire Solar Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
1.2 Introduction to Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
1.2.1 Optical Modelling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
1.2.2 Charge Transport Modelling . . . . . . . . . . . . . . . . . . . . . . . . 27
1.2.3 Coupling of Optical and Charge Transport Methods . . . . . . . . . . . 29
1.3 Thesis Outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
1.3.1 Devices being Studied . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
1.3.1.1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
1.3.2 Summary of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

2 Literature Review 33
2.1 Review of ETA cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
2.1.1 Porous TiO2 cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
2.1.1.1 Light Scattering in Porous TiO2 . . . . . . . . . . . . . . . . . 33
2.1.1.2 Inorganic Absorbers . . . . . . . . . . . . . . . . . . . . . . . . 34
2.1.1.3 3D cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
2.1.1.4 Thin Film Absorbers . . . . . . . . . . . . . . . . . . . . . . . 35
2.1.1.5 Nanoparticle/Quantum-dot Absorbers . . . . . . . . . . . . . . 36
2.1.2 ZnO Nanorod Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.1.2.1 Thin Film Absorbers . . . . . . . . . . . . . . . . . . . . . . . 37
2.1.2.2 Nanoparticle/Quantum-dot Absorbers . . . . . . . . . . . . . . 38
2.1.3 TiO2 Nanowire Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
2.1.4 Perovskite Sensitised ETA cells . . . . . . . . . . . . . . . . . . . . . . . 40
2.1.5 Optimisation of ETA cells . . . . . . . . . . . . . . . . . . . . . . . . . . 41
2.1.5.1 Modelling of ETA cells . . . . . . . . . . . . . . . . . . . . . . 44
2.1.6 Summary of ETA cells Performance . . . . . . . . . . . . . . . . . . . . 45
2.2 Light-Trapping in Solar Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
2.3 Electromagnetic Modelling of Nanostructured Solar Cells . . . . . . . . . . . . 48

1
 2.3.1 FDTD Simulations of Nanowires . . . . . . . . . . . . . . . . . . . . . . 48
2.3.1.1 TiO2 Nanotubes . . . . . . . . . . . . . . . . . . . . . . . . . . 48
2.3.1.2 Semiconductor Nanowires . . . . . . . . . . . . . . . . . . . . . 49
2.3.2 FEM and Multiphysics Simulations of Solar Cells . . . . . . . . . . . . . 52
2.3.2.1 Multiphysics simulations . . . . . . . . . . . . . . . . . . . . . 52
2.4 Charge Transport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
2.4.1 Charge Transport in Nanowire Geometries . . . . . . . . . . . . . . . . . 53
2.4.2 Numerical Methods for Charge Transport . . . . . . . . . . . . . . . . . 54
2.4.3 Charge Transport in Polymers, Hybrids and Metal-Oxides . . . . . . . . 56
2.4.3.1 Charge Transport in TiO2 . . . . . . . . . . . . . . . . . . . . 56
2.4.3.2 Charge Transport and Optical Properties of Polymers . . . . . 57

3 Optical Methods 59
3.1 Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
3.1.1 Differential Equation Methods . . . . . . . . . . . . . . . . . . . . . . . 59
3.1.1.1 Separation of Variables Method . . . . . . . . . . . . . . . . . 59
3.1.2 Mie Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
3.1.2.1 Homogeneous Cylinder . . . . . . . . . . . . . . . . . . . . . . 61
3.1.2.2 Finite Length Cylinders . . . . . . . . . . . . . . . . . . . . . . 63
3.1.2.3 Multiple Finite Cylinders . . . . . . . . . . . . . . . . . . . . . 64
3.1.2.4 Multiple Cylinders in an Absorbing Medium . . . . . . . . . . 64
3.1.3 Transfer-Matrix Method . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
3.2 Numerical Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
3.2.1 The Finite-Difference Time-Domain Method . . . . . . . . . . . . . . . . 65
3.2.1.1 Yee Implementation . . . . . . . . . . . . . . . . . . . . . . . . 66
3.2.2 Lumerical FDTD Solutions . . . . . . . . . . . . . . . . . . . . . . . . . 66
3.2.2.1 Simulation Setup . . . . . . . . . . . . . . . . . . . . . . . . . . 68
3.2.2.2 Illumination and Fourier Transform . . . . . . . . . . . . . . . 68
3.2.2.3 Absorption Calculation . . . . . . . . . . . . . . . . . . . . . . 70
3.2.2.4 Meshing and Dielectric Function . . . . . . . . . . . . . . . . . 70
3.2.2.5 PML Boundary Conditions . . . . . . . . . . . . . . . . . . . . 74
3.2.2.6 Oblique Incidence . . . . . . . . . . . . . . . . . . . . . . . . . 77
3.3 Modelling Core-Shell Nanowires . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
3.3.1 Simulation Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
3.3.2 Material Optical Properties . . . . . . . . . . . . . . . . . . . . . . . . . 83
3.4 Modelling nanoparticle Nanowires . . . . . . . . . . . . . . . . . . . . . . . . . 84
3.4.1 Simulation Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
3.4.2 Material Optical Properties . . . . . . . . . . . . . . . . . . . . . . . . . 84
3.5 Computational Resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
3.5.1 Mie Theory Calculations . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
3.5.2 FDTD Simulations and Analysis . . . . . . . . . . . . . . . . . . . . . . 86

4 Charge Transport Methods 87

4.1 Theory of Charge Transport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
4.1.1 Basic Photovoltaic Principles . . . . . . . . . . . . . . . . . . . . . . . . 87
4.1.2 The Quantum Mechanics of Electrons and Holes in Semiconductors . . 89
4.1.2.1 Band Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
4.1.2.2 Electron and Hole Statistics . . . . . . . . . . . . . . . . . . . 90
4.1.2.3 Boltzmann Approximation for Charge Carrier Densities . . . . 92
4.1.2.4 Doping and Bias . . . . . . . . . . . . . . . . . . . . . . . . . . 92
4.1.3 Charge Carrier Dynamics in Crystalline Semiconductors . . . . . . . . . 94
4.1.3.1 Current Density . . . . . . . . . . . . . . . . . . . . . . . . . . 94
4.1.3.2 Drift-Diffusion Model . . . . . . . . . . . . . . . . . . . . . . . 95
4.1.3.3 Recombination of Charge Carriers . . . . . . . . . . . . . . . . 96
4.1.4 The p-n Junction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
4.1.4.1 Theoretical Analysis of the 1D p-n Junction . . . . . . . . . . 98
4.1.4.2 The p-i-n junction . . . . . . . . . . . . . . . . . . . . . . . . . 101
4.2 Numerical Methods - Microvolt . . . . . . . . . . . . . . . . . . . . . . . . . . . 101

2
 4.2.1 The Model used by Microvolt . . . . . . . . . . . . . . . . . . . . . . . . 102
4.2.1.1 Boundary Conditions . . . . . . . . . . . . . . . . . . . . . . . 104
4.2.2 Finite-Difference Discretisation . . . . . . . . . . . . . . . . . . . . . . . 104
4.2.3 Solution of Discretised Equations . . . . . . . . . . . . . . . . . . . . . . 105
4.3 Implementation of Microvolt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
4.3.1 Coupling FDTD to Microvolt . . . . . . . . . . . . . . . . . . . . . . . . 107
4.3.2 Microvolt Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
4.3.2.1 Reproducing Deinega’s Results . . . . . . . . . . . . . . . . . . 107
4.3.2.2 Microvolt Heterojunction Test . . . . . . . . . . . . . . . . . . 112
4.3.2.3 Debugging the Heterojunction . . . . . . . . . . . . . . . . . . 112
4.3.2.4 Solution to Numerical Instabilities . . . . . . . . . . . . . . . . 114
4.3.3 Microvolt Setup for TiO2 /CdSe Core-shell Nanowires . . . . . . . . . . 116

5 Results 117
5.1 Optical Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
5.1.1 Coated Cylinders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
5.1.1.1 Varying Core Radius . . . . . . . . . . . . . . . . . . . . . . . 118
5.1.1.2 Varying Coating Thickness . . . . . . . . . . . . . . . . . . . . 118
5.1.2 Planar Slabs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
5.1.2.1 Pure CdSe Slabs . . . . . . . . . . . . . . . . . . . . . . . . . . 120
5.1.2.2 TiO2 /CdSe Slab . . . . . . . . . . . . . . . . . . . . . . . . . . 120
5.1.2.3 Nanowires vs Planar Solar Cells . . . . . . . . . . . . . . . . . 120
5.1.3 Rough Coating vs Smooth Coating . . . . . . . . . . . . . . . . . . . . . 122
5.2 Core-shell Nanowires and the effect of Geometry on Absorption . . . . . . . . . 124
5.2.1 Optical Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
5.2.1.1 Light Absorption vs Geometry . . . . . . . . . . . . . . . . . . 125
5.2.1.2 Analysis of Spectra . . . . . . . . . . . . . . . . . . . . . . . . 127
5.2.1.3 Field Profiles: Frequency Domain . . . . . . . . . . . . . . . . 129
5.2.1.4 Field Profiles: Time Domain . . . . . . . . . . . . . . . . . . . 130
5.2.2 Coupled Optical / Charge Transport Results . . . . . . . . . . . . . . . 130
5.2.2.1 Generation Rate . . . . . . . . . . . . . . . . . . . . . . . . . . 131
5.2.2.2 Conversion to 1D . . . . . . . . . . . . . . . . . . . . . . . . . 131
5.2.2.3 Energy Levels and Charge Densities . . . . . . . . . . . . . . . 132
5.2.2.4 J-V Curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
5.2.2.5 Maximum Power Yield . . . . . . . . . . . . . . . . . . . . . . 135
5.3 Nanoparticle Nanowires . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
5.3.1 Absorption Spectra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
5.3.2 Absorption vs Height . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
5.3.3 Random vs Ordered Nanoparticles . . . . . . . . . . . . . . . . . . . . . 139
5.3.4 Total Absorption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
5.3.5 Absorption per Unit Volume,  . . . . . . . . . . . . . . . . . . . . . . . 141
5.3.6 Conformal Coating vs Nanoparticles . . . . . . . . . . . . . . . . . . . . 142
5.4 Discussion of Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
5.4.1 Optical Properties of TiO2 /CdSe Nanowires . . . . . . . . . . . . . . . . 144
5.4.2 Charge Transport in TiO2 /CdSe Nanowires . . . . . . . . . . . . . . . . 145
5.4.3 Optical Properties of Nanoparticle Nanowires . . . . . . . . . . . . . . . 146
5.5 Overall Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146

6 Conclusions 147
6.1 Summary of Work Done . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
6.2 Complications and Future Work . . . . . . . . . . . . . . . . . . . . . . . . . . 148
6.2.1 Optical Properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
6.2.1.1 Electronic Structure Calculations . . . . . . . . . . . . . . . . 148
6.2.2 Charge Transport . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
6.2.2.1 TiO2 Charge Transport . . . . . . . . . . . . . . . . . . . . . . 149
6.2.2.2 Interfacial Charge Transport . . . . . . . . . . . . . . . . . . . 149
6.2.2.3 Organic Charge Transport . . . . . . . . . . . . . . . . . . . . 149
6.2.2.4 Numerical Problems . . . . . . . . . . . . . . . . . . . . . . . . 149

3
 6.2.3 Other Parameters to Test . . . . . . . . . . . . . . . . . . . . . . . . . . 150
6.3 Software Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150

Bibliography 151

A Derivation of Absorption per Unit Volume in Nanoparticles 169

B Permission Documents 171

4
List of Figures

1.1 Best research solar cell efficiencies, from 1976 onwards. This plot is
courtesy of the National Renewable Energy Laboratory, Golden, CO. . 21
1.2 Different types of silicon used in solar cells. Monocrystalline silicon (c-
Si), poly-crystalline or multi-crystalline silicon (p-Si or multi-Si) and
amorphous silicon (a-Si). Figure from Wikimedia commons [31]. . . . 22
1.3 Dye-sensitised solar cell [65]. Photons are absorbed in the dye, creat-
ing an exciton. The exciton dissociates, injecting an electron into the
conduction band of the TiO2 , while the electrolyte regenerates the dye.
Original figure by Gratzel [65], permissions granted. . . . . . . . . . . 25
1.4 The current-voltage curve (black) and power-density (grey) of an ideal
solar cell [137]. The maximum current density (short-circuit) is Jsc . The
maximum voltage (open-circuit) is Voc . The maximum power density,
P = Jm Vm is given by the area of the small rectangle. The larger
rectangle gives the power density P = Jsc Voc if the fill-factor F F is
one. Original figure by Nelson [137], permissions granted. . . . . . . . 28
1.5 (a) Schematic of the nanowire array solar cell. The nanowires consist of
CdSe coated TiO2 . These are embedded in a hole conducting polymer.
The diagram shows a small section. The solar cell is modelled with
periodic boundary conditions resulting in an infinite array of nanowires.
In each nanowire, the TiO2 core connects to a TiO2 substrate which is
in contact with the anode. The polymer fills in all the space between
the gold cathode, TiO2 substrate and nanowires. (b) Side view . . . . 31
1.6 Diagram of the nanoparticle sensitised solar cells (B.ElZein, E.Traversa
2014). The n-type layer consists of ZnO nanowires whilst the p-type
light-absorbing layer consists of PbS nanoparticles. . . . . . . . . . . . 31

2.1 Efficiency of a solar cell as a function of bandgap [137], based on the

Shockley-Quiesser limit for a single p-n junction [160]. Original figure
by Nelson [137], permissions granted. . . . . . . . . . . . . . . . . . . . 34
2.2 SEM image of a 3D solar cell [11, 19]. CdTe is electrochemically de-
posited onto TiO2 , acting as both the absorber and the hole-collector.
The thick black line highlights the interface between the CdTe and
TiO2 . The highly convoluted interface between the absorber layer and
electron-conductor (in this case TiO2 ) leads to increased light scattering
and hence optical path length enhancement. Original figure by Belaidi
et al [11], permissions granted. . . . . . . . . . . . . . . . . . . . . . . 35

5
2.3 External quantum efficiency of CdSe coated ZnO nanorod solar cells,
under different durations of annealing [135]. The efficiency increases
when annealed up to 24 hours. Further annealing decreases the effi-
ciency. Original figure by Nadarajah et al [135], permissions granted. . 38
2.4 TiO2 /CdS/CdSe nanowires with 4 CdS layers and n CdSe layers where
n = 0 to 4 [111]. Original figure by Li et al [111], permissions granted. 39
2.5 (a) Reflectance of TiO2 nanowires calculated from EMA method and
experiment, and reflectance of bulk TiO2 [154]. There is good agreement
with experiment, and the nanowires have less reflectance than bulk
TiO2 . (b) SEM image of TiO2 nanowires [154]. Original figures by Sani
et al [154], permissions granted. . . . . . . . . . . . . . . . . . . . . . . 40
2.6 Perovskite ETA cell, synthesised by Lee et al [106]. The mesoporous
TiO2 electron conducting layer is sensitised with a perovskite absorber.
Spiro-OMeTAD is used as a hole-conducting polymer. Original figure
by Lee et al [106], permissions granted. . . . . . . . . . . . . . . . . . . 41
2.7 Band diagrams comparing perovskite ETA cell with TiO2 or Al2 O3
[106]. In the Al2 O3 case, the conduction band is higher than the per-
ovskite, so electrons cannot be transported through. Instead, they are
transported through the perovskite to the electrode. When TiO2 is
used, this transports the electrons and the perovskite functions solely
as an absorber. Original figure by Lee et al [106], permissions granted. 42
2.8 Perovskite sensitised TiO2 nanowire arrays by Qiu et al [149]. Fig-
ures (a) and (b) show SEM images of the bare TiO2 nanowires. Figure
(c) and (d) show SEM images of the TiO2 nanowires coated with per-
ovskites. Figure (e) shows a cross-sectional SEM image of the full solar
cell including the spin-coated spiro-OMeTAD. Figure (f) shows a dia-
gram of the full solar cell. Original figure by Qiu et al [149], permissions
granted. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
2.9 Schematic of ETA solar cell as modelled by Taretto et al [169]. (a)
shows the real structure with its convoluted folds. The band diagram
for the segment A to A’ is shown in Figure 2.10. (b) shows the comb-
like structure used in this model where there are N p-i-n type elements
connected in parallel, a dashed line enclosing each element. A p-i-n
junction consists of a p-type layer, intrinsic layer (no doping) and n-
type layer connected in series (see section 4.1.4.2). Original figure by
Taretto et al [169], permissions granted. . . . . . . . . . . . . . . . . . 44
2.10 Simplified p-i-n type band diagram for the segment A to A’ in Figure
2.9a [169]. The bands of the n-type and p-type materials are staggered
in such a way that the bandgap of the absorber is equal to the change in
workfunctions. Original figure by Taretto et al [169], permissions granted. 45
2.11 Plasmonic light trapping schemes [3]. (a) Light scattering off metallic
nanoparticles, (b) excitation of localised surface plasmons, (c) grating
coupling to surface-plasmon-polaritons. Original figure by Atwater et
al [3], permissions granted. . . . . . . . . . . . . . . . . . . . . . . . . . 47
2.12 Optimised light trapping structure resulting from genetic algorithms
[183]. (a) Unit cell of the optimised light trapping pattern. (b) Multiple
tiles of this pattern. (c) Absorption spectra of the optimised pattern
and a random pattern for comparison. Original figure by Wang et al
[183], permissions granted under the Creative Commons licence. . . . . 48

6
2.13 (a) 2D FDTD model of TiO2 nanotubes [133]. (b) Measured results vs
FDTD results of the transmittance of TiO2 nanotubes [133]. Original
figures by Mor et al [133], permissions granted. . . . . . . . . . . . . . 49
2.14 Plots of specular reflectance of the base-tapered InP nanowires [39].
Measured results (a) and FDTD simulations (b). Original figures by
Diedenhofen et al [39]. Reprinted (adapted) with permission from Died-
enhofen et al [39]. Copyright (2011) American Chemical Society. . . . 50
2.15 Absorption as a function of wavelength and nanowire diameter for 2.5µm
long GaAs nanowire arrays embedded in a dielectric of n = 1.67 [100].
Results obtained using FDTD simulations. Original figure by Krogstrup
et al [100], permissions granted. . . . . . . . . . . . . . . . . . . . . . . 52
2.16 Absorption profiles (in a unit cell) of the plasmonic solar cell under
illumination at different wavelengths [113]. Original figure by Li et al
[113], permissions granted. . . . . . . . . . . . . . . . . . . . . . . . . . 53
2.17 Efficiency vs cell thickness and electron diffusion length for planar Si
solar cells (a) and nanorod solar cells (b). Original figure by Kayes et
al [85] , permissions granted. . . . . . . . . . . . . . . . . . . . . . . . 55
2.18 Meshes used in the 2D(r, ) cylindrical discretisation (top) and 3D(x, y, z)
cartesian discretisation (bottom) for the nanowire drift-diffusion model
by Deinega and John [35]. Original figure by Deinega et al [35] , per-
missions granted. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
2.19 Diagram representing the transfer of a polaron from initial state A to
final state B [38]. The corresponding equilibrium geometries are qA
and qB , and the transition state is at qC (see Figure 2.20). G⇤ is the
diabatic activation energy, is the reorganisation energy (the energy of
the final state B at the geometry qA ) and VAB the electronic coupling
matrix element. Original figure by Deskins and Dupuis [38], permissions
granted. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
2.20 Diagram representing the transfer of an electron (via polaron) [38]. The
electron is localised on the left Ti ion (qA geometry), shared between the
two ions (qC geometry) then localised on the right ion (qB geometry).
Original figure by Deskins and Dupuis [38], permissions granted. . . . 58

3.1 Infinitely long cylinder with a radius of a, used in the Mie theory anal-
ysis. k represents the wave-vector of the incoming transverse-electric
light, E and H the electric and magnetic fields. . . . . . . . . . . . . . 62
3.2 A diagram of the “Yee Cell” [30] demonstrating the interweaving of the
electric and magnetic fields. The mesh points are labeled (i,j,k) for the
x, y and z directions respectively. The curl of the magnetic field through
the faces gives the time derivatives of the electric fields at the edges and
vice versa. Original figure from Wikimedia commons [30]. . . . . . . . 67
3.3 Leapfrog algorithm [165]. At each half time step, either the electric fields
or magnetic fields are calculated, using values of the fields at previous
time steps and at half step positions. Original figure by Taflove [165],
permissions granted. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67

7
3.4 (a) 2D Simulation model of the TiO2 /CdSe nanowire. The grey circle in
the centre represents the TiO2 core. The blue ring represents the CdSe
coating. The inner yellow box represents the ‘total field monitor’. The
outer yellow box is the ‘scattered field monitor’. The pink arrow rep-
resents the x-direction of the ‘total-field scattered-field’ (TFSF) plane
wave source, the blue arrow the oscillation of the electric field in the
y-axis. The grey box attached to the source represents the cutoff for
the incident field, leaving only the scattered field. Absorbing boundary
conditions are enforced outside the simulation region. (b) 3D simulation
model of the nanowire with top-down illumination. Periodic boundary
conditions are enforced in the x and y directions. Absorbing boundary
conditions are enforced in the ±z directions. . . . . . . . . . . . . . . . 69
3.5 Electric field intensity of the place-wave light source used in FDTD
simulations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
3.6 Snapshots in time (femtoseconds) of the 2D FDTD simulations of light
scattering off the TiO2 /CdSe nanowire. The illumination is from the
side (perpendicular to nanowire axis). Blue represents zero intensity,
red is maximum intensity. The two concentric circles in the centre
represent the coated nanowire. . . . . . . . . . . . . . . . . . . . . . . 72
3.7 Conformal Meshing techniques for FDTD modelling [120], developed by
Yu and Mittra [197], on which Lumerical’s conformal meshing techno-
logy is based. (a) This model provides greater accuracy for resolving
PEC (perfect electrical conductor) / dielectric interfaces. The contour
integral for Maxwell’s integral equations is reduced from the contour C
to the contour C1 such that the PEC with the vanishing electric field
is not included. (b) This model takes into account interfaces between
two dielectrics. The Yee Cell’s effective permittivity in the x and y
directions are weighted by the fraction of the mesh that is inside each
material in those directions. Original figures by Lumerical Solutions
Inc. [120], permissions granted. . . . . . . . . . . . . . . . . . . . . . . 73
3.8 Lumerical simulation setup for a coated nanowire with core radius a =
100nm and coating thickness h = 20nm, perspectives from the top (a)
and from the side (b). The blue and green translucent regions represent
symmetric and anti-symmetric boundaries respectively (reducing the
computational requirement by a factor of 4). Three different meshes are
used. For all cases, the minimum mesh step was set to 3nm. On the left,
the maximum mesh steps in the x and y directions were xmax = 20nm
and ymax = 20nm, and for the z direction the number of mesh cells
per wavelength was set to nz = 9. In the middle, xmax = 15nm,
ymax = 15nm and nz = 15. On the right, xmax = 10nm, ymax =
10nm and nz = 18. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
3.9 Absorption spectra obtained using three different meshes for a periodic
array of nanowires with a = 100nm and h = 20nm. The solid line shows
the results for a mesh with xmax = 20nm, ymax = 20nm and nz = 9
(coarsest mesh). Dashed line: xmax = 15nm, ymax = 15nm and
nz = 15. Dotted line: xmax = 10nm, ymax = 10nm and nz = 18
(finest mesh). The results have converged for the finest mesh. . . . . . 76

8
3.10 The PML method [14]. Perfectly matched layers are applied on the
boundaries of the simulation region to absorb electromagnetic waves.
The conductivities ( x , x⇤ , y , y⇤ ) of each PML are specially chosen to
prevent reflections. Original figure by Berenger [14], permissions granted. 78
3.11 Schematic of injection angle in broadband simulations performed with
Lumerical FDTD solutions [120]. For a single oblique broadband simu-
lation, the in-plane component of the wavevector is kept constant over
the range of frequencies. This results in an angle of injection which var-
ies as a function of frequency. Original figure by Lumerical Solutions
Inc. [120], permissions granted. . . . . . . . . . . . . . . . . . . . . . . 79
3.12 Solar intensity in Wm 2 nm 1 as a function of wavelength. . . . . . . 80
3.13 (a) 2D nanowire simulation setup. The absorption monitor (red box)
measures the amount of light that is removed by the nanowire by cal-
culating the total transmission of light out of the box. The Total-Field
Scattered-Field box (yellow) confines the incident light field, such that
any light outside the box is only scattered. The Perfectly Matched Lay-
ers (PML) remove light from the simulation region while minimising
reflection. (b) 3D nanowire simulation setup. Plane wave light (S)
propagates from above. Periodic boundary conditions are enforced in
the x and y directions to simulate an infinite array of nanowires. PML
boundary conditions are enforced in the z directions. The amount of
light reflected (R) or transmitted (T) was monitored for each wavelength. 81
3.14 Different nanowire arrangements as viewed from above. In (a) each unit
cell has one nanowire whilst in (b) each unit cell has two nanowires.
Periodic boundary conditions are enforced on the outermost dotted
lines. In (c), (d) and (e) multiple nanowires are placed within a large
super-cell which has periodic boundary conditions enforced. The inner-
dotted lines in (d) and (e) refer to the unit cells (analogous to (a) and
(b)) where the position of each nanowire is confined. For (d), the max-
imum displacement in the x or y directions is ±(p/2 a) where p is the
unit cell lattice constant and a is the core radius. For (e), the maximum
x and y displacements from ideal position are ± 23 (p/2 a) . . . . . . 82
3.15 The complex refractive index of CdSe. Red refers to the real part, blue
to the imaginary part. The circles refer to the experimental data from
Palik [144]. The lines are Lumerical’s multi-coefficient models [120]
fitted to the experimental data. . . . . . . . . . . . . . . . . . . . . . 83
3.16 (a) Three different types of nanowire coating: Ordered nanoparticles,
randomised nanoparticles and conformal coating. (b) Four additional
parameters were varied: nanowire radius, nanoparticle size, number of
nanoparticles and lattice constant (or spacing). . . . . . . . . . . . . 85
3.17 The complex refractive index of PbS. Purple refers to the real part, teal
to the imaginary part. The cross refer to the experimental data from
Palik [144]. The lines are Lumerical’s multi-coefficient models [120]
fitted to the experimental data. . . . . . . . . . . . . . . . . . . . . . 85

4.1 The quantum efficiency of GaAs as a function of wavelength, in com-

parison with the photon flux density of solar irradiation [137]. The
quantum efficiency is high at wavelengths where there is high solar in-
tensity, making GaAs suitable for photovoltaic devices. Original figure
by Nelson [137], permissions granted. . . . . . . . . . . . . . . . . . . . 88

9
4.2 Band structure diagram for a direct bandgap semiconductor. The en-
ergy bands are represented at the Brillouin zone in a one-dimensional
cross-section of k-space. The conduction band Ec (k), and valence band
Ev (k) are separated by the bandgap Eg . . . . . . . . . . . . . . . . . . 91
4.3 Energy level diagrams for semiconductors with p-type doping (left),
n-type doping (right), and no doping (middle). The Fermi level is rep-
resented by the dashed lines while the acceptor and donor levels are
represented by dotted lines. . . . . . . . . . . . . . . . . . . . . . . . 93
4.4 The energy levels of electrons in semiconductors with p-type dopants
(left) and n-type dopants (right). Evac is the vacuum energy, at which
an electron becomes unbound from the material. Ec and Ev mark
the lowest and highest levels of the conduction band and valence band
respectively. EF is the Fermi level, which resides in the bandgap. The
workfunction, denoted by , is the energy required to excite an electron
from the Fermi level to Evac (i.e, the lowest energy required to release
an electron from the semiconductor). . . . . . . . . . . . . . . . . . . 97
4.5 (a) The p-n junction. (b) Energy level diagram of the p-n junction
at equilibrium. One can picture the energy levels in Figure 4.4 with
different Fermi levels. When joined together, near the junction (fine
dotted line) the energy bands ‘bend’ by an amount qVbi to accommodate
the equidistribution of the Fermi level. Electrons are represented as
filled circles, holes as empty circles. . . . . . . . . . . . . . . . . . . . . 99
4.6 Depletion zone in the p-n junction, marked by the faded area within the
dashed lines with a width of wn + wp . In this region, under diffusion
the electrons have left the n-type donor atoms leaving them as posit-
ively charge ions. Likewise, holes have left the acceptors leaving them
negatively charged. The resulting electric field (which resides only in
the depletion zone) acts as a barrier to further diffusion. . . . . . . . 100
4.7 Energy level diagram of p-i-n junction. The region in the middle has no
p
doping, but has an intrinsic charge density ni = np. Band-bending
occurs over a longer distance than in the p-n junction. . . . . . . . . . 102
4.8 Band diagram representing how semiconductor devices are modelled.
Original figure taken from Microvolt website [131], permissions granted. 103
4.9 Gummel method as implemented in Microvolt . . . . . . . . . . . . . 107
4.10 Diagram of silicon nanowire charge transport model [35]. Red lines
represent the metal contacts, blue lines represent the boundary with
the insulating medium. The green line separates the p-doped region
from the n-doped region. Original figure by Deinega and John [35],
permissions granted. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
4.11 Optical setup of Deinega’s nanowire in Lumerical. The grey cylinder is
the nanowire itself, the blue region is the silica medium. The orange
box represents the simulation region and the pink arrow represents the
solar irradiance. Periodic boundary conditions are enforced. The silica
medium extends to infinite depth, such that no reflection occurs between
the boundary of the silica and vacuum. . . . . . . . . . . . . . . . . . . 109

10
4.12 Optical absorption profile averaged over azimuthal angle. Results from
Lumerical are on the left, results from Deinega et al are on the right
[35]. For the Lumerical results, the highest generation rate (dark red)
is 1.4 ⇥ 1022 cm 3 s 1 . For Deinega’s results (which were obtained using
EMTL [44]) the colour scale goes up to 1.0 ⇥ 1022 cm 3 s 1 but the
generation rate probably goes higher than this due to the fact that the
redness is over-saturated. Figure on the right by Deinega and John [35],
permissions granted. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
4.13 Solar spectrum with interpolation on top. . . . . . . . . . . . . . . . . 111
4.14 Current voltage curve for silicon nanowire solar cell. Short circuit cur-
rent is 15mA/cm2 , open-circuit voltage is about 0.37V. . . . . . . . . . 111
4.15 Absorption spectrum for the nanowire array. The dashed black line is
that calculated by Deinega using EMTL. The coloured lines are results
I obtained using Lumerical for increasing mesh resolutions. . . . . . . 111
4.16 (a) Refractive index data for Lumerical vs EMTL (Deinega). (b) A
closeup of the imaginary component of the Lumerical refractive index fit.112
4.17 Diagrams of devices simulated by Metzger [129]. Light is incident from
the top. The ZnO and CdS are n-doped. (a) Planar, 1D. (b) Interdig-
itated, 2D. The CIGS is p-doped in (b), and in (a) it has alternating
regions of n-type and p-type doping separated by width W . Original
Figures by Metzger [129], permissions granted. . . . . . . . . . . . . . 113
4.18 Short-circuit current density (left) and open-circuit voltage (right) for
the Metzger cell [129]. The results are plotted against W (µm), the
width of the interdigitated regions. The straight black lines show results
for cells without interdigitations. Original figures by Metzger [129],
permissions granted. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
4.19 Printout of the Jacobian matrix for the continuity equations 4.21 and
4.22, for a mesh of 101 points (resulting in a 202⇥202 matrix). The bot-
tom half represents the electron continuity derivatives for all positions,
the top half represents the hole continuity derivatives for all positions.
Note the large gap in the centre full of zeroes, making the Jacobian
ill-posed for the solution of the continuity equations. . . . . . . . . . . 115

5.1 Mie absorption efficiency Qabs for an infinite coated cylinder with core
radius a = 200nm and coating thickness 20nm. Red crosses are results
from the Lumerical simulation, the black line represents results from
Mie theory calculations. . . . . . . . . . . . . . . . . . . . . . . . . . . 118
5.2 Absorption efficiency Qabs against wavelength for a TiO2 /CdSe cylinder
with a coating thickness of 20nm. Mie theory results are coloured lines,
FDTD results are coloured circles. Purple is a coated cylinder with core
radius a = 50nm, teal 80nm and blue 100nm. The dashed black line
represent the imaginary part of the refractive index of CdSe. . . . . . 119
5.3 Absorption efficiency Qabs against wavelength for a TiO2 /CdSe cylinder
with a core radius of 80nm. Mie theory results are coloured lines, FDTD
results are coloured circles. Orange represents a coating thickness of
h = 10nm, teal 20nm and yellow 30nm. . . . . . . . . . . . . . . . . . 120
5.4 Absorbance against wavelength for planar slabs of pure CdSe. Theor-
etical results are solid lines, FDTD results are crosses. . . . . . . . . . 121

11
5.5 Absorption against wavelength for a TiO2 slab with a 20nm thick CdSe
coating. Lines are theoretical results, crosses are FDTD results. Purple
represents a TiO2 thickness of 100nm, teal 160nm and blue 200nm. . 121
5.6 Absorption against wavelength for a 160nm thick TiO2 slab coated with
various thicknesses of CdSe. Lines are theoretical results, crosses are
FDTD results. Orange represents a coating of 10nm, purple 20nm, teal
30nm and blue 40nm. . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
5.7 Absorption for a planar CdSe slab (purple 60nm and teal 300nm) as
well as a TiO2 /CdSe nanowire cell with core radius 100nm and coating
thickness 20nm (blue dashed line). . . . . . . . . . . . . . . . . . . . . 123
5.8 2D cross-sectional view of the TiO2 nanowire (light blue) with a rough
CdSe coating (dark blue). This coating was procedurally generated us-
ing a random number generator and gaussian filter. The default coating
thickness was set at 20nm and the maximum amplitude of the roughness
was 5nm. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
5.9 Absorption spectra for nanowires with a smooth coating (solid purple
line) and rough coating (dashed teal line). . . . . . . . . . . . . . . . 124
5.10 The size of the unit cell (dashed line) in the periodic FDTD simulations
scales with the size of the core. The radius of the core is a, the width
of the unit cell is 4a. Increasing the coating thickness h by a factor of ⇠
does not change the unit cell. Increasing the core radius a by a factor
of ⇠ will increase the width of the unit cell by the same factor. The
core filling ratio fC is such that the unit cell width is twice the core
diameter, i.e. fC = ⇡a2 /(4a)2 = ⇡/16. . . . . . . . . . . . . . . . . . . 125
5.11 Colormap of photocurrent density for a periodic square lattice of nanowires
with different core radii and shell thicknesses. Colours represent the
magnitude of the photocurrent (red is high, blue is low). The geometry
of one nanowire is shown pictorially for each data point as a ring whose
size scales with core radius and width scales with shell thickness. The
unit cell is twice the diameter of the nanowire core, leading to a filling
ratio of fC = ⇡a2 /(4a)2 = ⇡/16 = 0.196 . . . . . . . . . . . . . . . . . 126
5.12 Enhancement of photocurrent density, Jsquare J
, in different nanowire ar-
rangements when compared to the ordered square lattice results from
Figure 5.11. Red represents greater photocurrent, while blue represents
less photocurrent. Each ring is split into four quadrants representing
the enhancement of each type of nanowire array: top-left (semi-random
square), bottom-left (semi-random hexagonal), bottom-right (random)
and top-right (ordered hexagonal). . . . . . . . . . . . . . . . . . . . . 127
5.13 Absorption spectra for different nanowire arrangements with a shell
thickness of 30 nm. In (a), the total absorption percentage is calculated
for the ordered and random arrays of nanowires. In (b), the Mie absorp-
tion efficiency Qabs is calculated for the isolated nanowires. The black
lines in (a) and (b) show the absorption for the case where the core
radius a = 200 nm. One can see the major peaks lining up between the
isolated nanowires, ordered arrays and random arrays. The coloured
lines show the absorption for other core thicknesses. . . . . . . . . . . 128

12
5.14 Heat maps of absorption per unit volume in one nanowire of a periodic
array with core radius 200nm and shell thickness 30 nm. Top shows
absorption averaged over the axial dimension, bottom shows the ab-
sorption averaged over the z dimension. Absorption per unit volume
for three different wavelengths is shown: 400nm, 490nm and 550nm.
These wavelengths were chosen because the largest absorption peak in
Figure 5.13a is at 490nm, while 400nm and 550nm lie on either side of
the peak. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
5.15 The electric field intensity, |E|2 , of the light propagating through the
nanowire and subsequently reflecting back up. This is a cross-sectional
view. Four snapshots in time are taken. The colour scales range from
red for high intensity to blue for low intensity. . . . . . . . . . . . . . . 130
5.16 2D generation rate profiles of TiO2 /CdSe Core-shell nanowires. The
generation rate is measured in free electron-hole pairs per second per
cm3 . On the left the nanowire has core radius of 100nm and coating
thickness of 20nm. On the right the core radius is 200nm and coating
thickness is 30nm. The heat map ranges in colour from black (zero
generation rate) to yellow (maximum generation rate). . . . . . . . . 131
5.17 Plot of energy levels, charge densities and currents across the 1D Silicon
diode. The p-n junction is located in the centre at x = 0.5µm. . . . . 133
5.18 Plot of energy levels, charge densities and currents across the TiO2 /CdSe/polymer
device, in 1D. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
5.19 Current density J plotted against Voltage V for different nanowire para-
meters. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
5.20 Maximum power yield for core radii of 100, 150 and 200nm and coat-
ing thicknesses of 10, 20 and 30nm. The thinnest nanowires with the
thickest coatings have the largest power output. . . . . . . . . . . . . 136
5.21 Absorption spectra for nanoparticle coated nanowires. The thick lines
represent the nanowires with 500 nanoparticles. The dashed lines rep-
resent 100 nanoparticles. Different colours represent different nano-
particle diameters. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
5.22 (a) Absorption at = 400nm against the cube of nanoparticle radius for
nanowires coated with 1000 nanoparticles. The lattice spacing is 192nm.
Results for three different nanowire radii are shown. The radius of the
nanoparticle is r. (b) Logarithmic plots of 1 A vs size of nanoparticle,
showing that 1 A decays exponentially with r3 . . . . . . . . . . . . 137
5.23 (a) Plot of absorption at = 400nm vs number of nanoparticles. (b)
JMAPD vs number of nanoparticles . . . . . . . . . . . . . . . . . . . . 138
5.24 Average absorption (m 3 s 1 ) against height of the nanowire. The ab-
sorption is fairly even over the length of the coating. The orange line
shows a moving average. The red dashed line is the line of best fit. . 138
5.25 Absorption spectra for nanowires coated with randomised nanoparticles
(crosses) and ordered nanoparticles (lines). The nanowire spacing is
120nm and the nanoparticle diameters are 9nm. Results for three
nanowire thicknesses are shown with core radii of 25nm (purple), 40nm
(blue) and 50nm (orange). . . . . . . . . . . . . . . . . . . . . . . . . 139

13
5.26 Plot of maximally achievable photo current density, JMAPD , for different
nanowire radii, nanoparticle diameters and numbers of nanoparticles
with a lattice constant of 192nm. The nanoparticles are arranged on
the nanowires randomly. Ordered nanoparticles give similar results.
Each pictorial element represents a single value of nanowire radius and
nanoparticle diameter, but is split into 3 sections for three numbers of
nanoparticles: 100, 500 and 1000 nanoparticles. . . . . . . . . . . . . 140
5.27 Same as Figure 5.26 but for two other lattice constants, (a) 120nm and
(b) 240nm. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
5.28 JMAPD per unit volume for different nanowire parameters. . . . . . . . 141
5.29 Plot of absorption per unit volume, , against wavelength for a variety
of different numbers of nanoparticles and radii of nanowires.  is calcu-
lated by gathering results for nanoparticles ranging from 4nm to 9nm
in diameter. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
5.30 Plot of  against number of nanoparticles for lattice constant p = 192nm
and wavelength = 400nm. The blue curve is for nanowires with a core
radius of a = 25nm, and the green curve is for a = 50nm. . . . . . . . 143
5.31 Plot of  against radius of nanowire for = 400nm, p = 192nm and
N = 1000. The maximum absorption per unit volume is at a = 60nm. 143
5.32 Absorption spectra of conformal coatings (solid lines) and nanoparticles
(dashed lines). N = 500, a = 25nm, p = 120nm . . . . . . . . . . . . . 144
5.33 Absorption per unit volume of conformal coatings (solid lines) and nan-
oparticles (dashed lines). . . . . . . . . . . . . . . . . . . . . . . . . . . 145

A.1 Diagram of nanowire coated with nanoparticles. Light is incident from

above and propagates through the nanowire . . . . . . . . . . . . . . . 169

14
List of Tables

4.1 Table of properties for Metzger’s thin film solar cell [129] . . . . . . . 113
4.2 Microvolt J-V results of the Metzger Heterojunction. . . . . . . . . . . 114
4.3 Charge transport properties of TiO2 , CdSe and the polymer MEH-PPV.
The effective masses aren’t defined in the polymer, but these can be
calculated from the effective densities of states using equation 4.6. . . 116

5.1 Ratios of core (total) diameter, 2a (2b), to peak wavelength (of the
peaks in Figure 5.13b). 1 is the wavelength for the right peak, and 2
is the wavelength for the left peak. . . . . . . . . . . . . . . . . . . . . 128

15
Abstract

Radial junction nanowire solar cells are expected to give a higher performance
than planar junction solar cells. This is due to the enhanced optical path length
for absorption, and the shorter distance for excitons to travel to interfaces where
they can dissociate. By means of simulation, we have optimised, with respect
to geometrical parameters, solar cells consisting of light absorber coated metal-
oxide nanowires. We have primarily studied nanowires with a TiO2 core and
CdSe shell, as well as ZnO cores with PbS nanoparticles. Optical scattering
and absorption is determined using Finite-Difference Time-Domain (FDTD)
simulations. Charge transport is studied using finite-difference drift-diffusion
modelling. The generation rate calculated from FDTD is inputted into the
charge transport model. The aim is to determine which combination of nanowire
parameters (such as coating thickness, height, radius and spacing) lead to the
the highest power efficiencies. We found that core and shell thickness has more
effect on absorption than nanowire arrangement.

16
Declarations

Declaration of Originality
I declare that the work submitted in this thesis is my own and that all other
work is sufficiently referenced.

Declaration of Copyright
The copyright of this thesis rests with the author and is made available un-
der a Creative Commons Attribution Non-Commercial No Derivatives license.
Researchers are free to copy, distribute or transmit the thesis on the condition
that they attribute it, that they do not use it for commercial purposes and that
they do not alter, transform or build upon it. For any reuse or redistribution,
researchers must make clear to others the license terms of this work.

17
Acknowledgements

I would like to express my utmost gratitude to my PhD supervisor Dr. Andrew

Horsfield, for spending so much time helping me through derivations and anal-
yses, discussing results, reading and checking my thesis and all other guidance.
In addition, I’d like to thank the following people and institutions: The
Theory and Simulation of Materials Centre for Doctoral Training (TSM CDT)
for accepting my PhD application, nurturing my academic skills and providing
many extra-curricular oppurtunities; The Engineering and Physical Sciences
Research Council (EPSRC) for funding my PhD stipend; Dr. Danieli Dini for
being a great personal tutor and taking me and the rest of my cohort out to
regular lunches; Miranda Smith and Lilian Wanjohi for solving many adminis-
trative problems for me and generally being very helpful; Dr. James Spencer for
solving general computing and HPC issues; Fabian Renn and Aeneas Weiner
from the physics department for helping me with the Finite-Difference Time-
Domain method, Mie Theory and other theoretical optics discussions; Prof.
Stefan Maier for lending me access to his computing cluster and granting me
a license to use the Lumerical FDTD Solutions software; Prof. Jason Riley for
his experimental insight and inspiration for studying TiO2 /CdSe nanowires;
Lumerical Solutions, Inc. for the usage of their FDTD software and the Lumer-
ical support team for their patience in helping me solve technical issues; Math-
works for the usage of MATLAB for various calculations and data analyses;
Dr. Alexei Deinega and Prof. Sajeev John for developing the Microvolt soft-
ware used for charge transport simulations; Dr. Alexei Deinega again for in-
vesting lots of time to help me learn how to use Microvolt and debug errors.
I’d also like to thank Dr. Roy Schestowitz for usage of his thesis template:
http://www.schestowitz.com/IMG/research/thesis_examples/.
Finally, I’d like to express thanks to my friends, family and girlfriend for
supporting me throughout the PhD.

18
Chapter 1

Introduction

1.1 Motivation
One of the largest problems that society faces is global warming. The scientific
consensus that human action is a major contributor is overwhelming [32], with a
large component being due to greenhouse gases released from burning fossil fuels.
In 2009, fossil fuels contributed 81% to global energy production, nuclear power
2.8%, and renewable energy 16% [151]. To mitigate global warming while serving
ever increasing energy demands, the proportion of energy produced from fossil
fuels must reduce and renewable energy production must expand. Currently,
fossil fuels are much cheaper to use than renewable energy. This problem is
both technological and economical in scope.
Photovoltaics, the conversion of solar radiation to electrical power, is one
promising type of renewable energy. The main problem facing the full-scale
implementation of photovoltaics (and that of any other renewable energy tech-
nology) is that of economics. For photovoltaics to compete with fossil fuels, the
price per watt of solar cells will have to decrease by a factor between 2 and 5
[3]. It is expected that photovoltaic energy production will grow from 5 GW in
2008 to 50 GW in 2020 [3]. Current research focuses on increasing the efficiency
of photovoltaic devices (i.e, solar cells) or reducing the cost of manufacture.
Together these two branches of research will reduce the cost per watt of solar
power, working towards a future where solar cells are competitive with fossil
fuels.

1.1.1 A Brief History of Solar Cells

The photovoltaic effect, which is responsible for the operation of solar cells, was
first demonstrated experimentally by Edmond Becquerel in 1839 [10]. Silver
chloride was placed in an acidic solution and connected to platinum electrodes,
which produced a voltage while illuminated. This work was followed by Wil-
loughby Smith in 1873, who discovered that the photoconductivity of selenium
varies significantly with the amount of light incident on it [1].

19
 Charles Fritts produced the first solid-state solar cell in 1883. This device
consisted of selenium coated with a thin layer of gold [54]. A photovoltaic effi-
ciency of 1% was achieved. This device was based on the photoelectric effect,
discovered by Heinrich Hertz in 1887 [74]. He observed that metals exposed to
ultraviolet light would emit electrons. Only light above a threshold frequency
would dislodge electrons from the metal, while light below that frequency would
have no effect. In 1905, Albert Einstein was able to explain this effect by pos-
tulating that light was discretised into packets of energy called quanta [42]. His
work eventually led to the foundations of quantum mechanics. In this frame-
work, light consists of electromagnetic wave-packets called photons, where each
photon carries an amount of energy E = hf , where h is Planck’s constant and f
is the frequency of the photon. The theoretical underpinning of quantum mech-
anics and photons would be crucial later on for understanding the behaviour of
semiconductors and hence solar cells.
In 1939, Russell Ohl discovered the p-n junction, the interface between two
materials of different electrical conductivity in a semiconductor [152]. His work
led to the development of semiconductor diodes (and eventually transistors),
which allow current to flow more readily in one direction than the other. In
1946, Ohl patented the first silicon solar cell, which implements the p-n junction.
The first practical solar cell was publicly demonstrated in 1954 at Bell Labor-
atories [146]. Solar cells were first used in space on the Vanguard I satellite in
1958. Space satellites demanded a high power-to-weight ratio, due to the ex-
tremely high costs per kilogram of launching objects into space. This meant
that research was more aligned towards developing the highest quality solar
cells, without much effort going into lower-efficiency and lower-cost solutions,
so solar cells remained prohibitively expensive for anything other than the space
industry.
The development of integrated circuits in the 1960s led to larger single crys-
tals (boules) of silicon being produced at lower costs. This lowered the costs of
solar cells to $100 per watt by 1971 [145]. The oil crisis of 1973 and 1979 led to a
greater interest in alternative energy sources [121, 61]; investment in renewable
energy increased and photovoltaics began to be used in off-grid buildings and
homes.
For the next few decades, solar panels have steadily decreased in cost per
watt and increased in power conversion efficiency. Figure 1.1 shows the record
of the highest efficiency research solar cells over the past few decades. The
development of solar cells has branched out into different technologies, ranging
from the crystalline semiconductor solar cells (such as Si, GaAs, and multi-
junction variants) to second generation thin-film solar cells and emerging third
generation solar cells (dye-sensitised, perovskite, organic and quantum-dot solar
cells). The emerging PV have a relatively low efficiency. However, their cost is
low and they have a promising potential for higher efficiency, putting them on
par with conventional solar cells.

20
Figure 1.1: Best research solar cell efficiencies, from 1976 onwards. This plot is
courtesy of the National Renewable Energy Laboratory, Golden, CO.

1.1.2 Development of Solar Cells

1.1.2.1 First-generation: Crystalline Silicon
Most commercial solar cells are made from crystalline silicon (c-Si). The crys-
tallinity ensures there are no defects to trap the charge carriers. Such solar
cells can reach efficiencies as high as 25.6% for single-junction [126] and 40% for
multi-junction cells [67]. The problem with this type of solar cell is that it is
expensive, due to the high-temperature processing required [71] and difficulty
with synthesising pure crystalline silicon. Also, a large thickness is needed due
to the low absorption coefficient of silicon.

1.1.2.2 Second-generation: Thin-film, Amorphous Silicon and Poly-

crystalline Silicon
Thin-film solar cells have a thickness ranging from the order of 10nm to 10µm,
much thinner than first-generation solar cells (c-Si, 200µm). They tend to be
made from either cadmium telluride, copper indium gallium diselenide (CIGS)
or amorphous silicon. While being cheaper but less efficient than crystalline
silicon, thin film solar cells made from CdTe or CIGS have now reached 20%
power conversion efficiency, about the same as polycrystalline silicon solar cells
(p-Si). CdTe and CIGS have higher absorption coefficients than silicon, and
therefore require less thickness to absorb as much light.
CdTe cells form the majority of the thin-film market. The problem with
CdTe cells is that tellurium is rare, and cadmium is a toxic element. CIGS cells

21
Figure 1.2: Different types of silicon used in solar cells. Monocrystalline sil-
icon (c-Si), poly-crystalline or multi-crystalline silicon (p-Si or multi-Si) and
amorphous silicon (a-Si). Figure from Wikimedia commons [31].

solve both of these problems, but still have low manufacturing yields due to the
complexity of the materials [164].
Amorphous silicon (a-Si) solar cells are the third most common type of
thin-film solar cells. Silicon is still an attractive material to use because of
its abundance and non-toxicity. Amorphous silicon requires a low processing
temperature compared to c-Si. a-Si is good for diffuse or weak sunlight, making
it ideal for locations with cloudy weather. However, due to the Staebler-Wronski
effect [163] where the conductivity changes under prolonged solar illumination,
the efficiency of a-Si cells decreases by 10-30% over the first 6 months of usage.
p-Si (or multi-Si) solar cells are made from a polycrystalline form of silicon,
produced using the Siemens process [194], rather than a single crystal (see Figure
1.2). This is easier to produce than single crystal boules of silicon. However,
the grain-boundaries induce recombination. Currently, p-Si solar cells comprise
the highest share of the PV market at 54.9%.

1.1.2.3 Third generation: Organic Solar Cells

An alternative to using silicon or other crystalline semiconductors is to make
solar cells out of semiconducting organic materials. Organic photovoltaics (OPV)
use conjugated polymers or small organic molecules. With molecular engineer-
ing, one can manipulate the bandgap (or HOMO-LUMO gap) of polymers by
changing their lengths and functional groups. Organic solar cells are much
cheaper to produce than crystalline PV, require less material and are more flex-
ible. They can be produced using roll-to-roll printing technology [101, 138] and
ink-jet technology. However, they are a lot less efficient and degrade easily [79].
They are also a formidable challenge to develop due to the use of novel materials,
manufacturing methods and device architectures as compared to traditional in-
organic solar cells [71]. The highest efficiency achieved by an organic solar cell
so far is 10% [66]. However, the biggest problem with OPV is that they are
prone to environmental degradation.
A typical single layer device consists of an organic material sandwiched
between two electrodes with different work functions. This sets up an elec-
tric field which pulls apart excitons generated under illumination. Electrons

22
are excited to the lowest unoccupied molecular orbital (LUMO), leaving behind
holes in the highest occupied molecular orbital (HOMO). Single layer organic
devices tend to have extremely low efficiencies and a better approach is to use
more than one layer [71].

1.1.3 Improving the Efficiency of Emerging PV

1.1.3.1 Bilayer Organic Solar Cells
A bilayer (or planar heterojunction) device has two organic layers with different
electron affinities. One layer is an electron donor and the other is an electron
acceptor. A strong effective field is set up at the interface between the mater-
ials, which is better at dissociating excitons than the electric field in a single
layer device. This is because the HOMO-LUMO gaps (analogous to bangaps
in inorganic semiconductors) of each material are staggered so as to encourage
charge separation. However, the diffusion length of excitons in organic materials
is usually only about 10nm, meaning that a lot of the excitons generated away
from the interface become trapped or recombine before they can undergo charge
separation and contribute to the current.

1.1.3.2 Bulk Heterojunction Devices

To increase the efficiency of bilayer organic solar cells, more advanced designs
were conceived, aiming to improve either the charge separation, diffusion length
or light absorption. In bulk heterojunction devices, the donor and acceptor ma-
terials are blended together such that they interpenetrate [138]. This leads to a
large interfacial area and small domain size, ensuring that a high percentage of
excitons reach the donor acceptor interface and undergo charge separation. The
highest efficiency recorded with this kind of device is 6.77% [26]. The drawback
is that this leads to discontinuous regions of donors and acceptors, preventing
charge carriers from reaching their electrodes since there is no clear path. This
can be mitigated by engineering a continuous nanostructured interface. Yang et
al [193] solved this problem in bulk heterojunctions with a method for growing
crystalline organic films in a controlled way, using organic vapour-phase depos-
ition. This resulted in a substantial increase in efficiency compared to previous
bulk heterojunction devices.

1.1.3.3 Hybrid Organic/Inorganic Solar Cells

A further development is that of the hybrid organic/inorganic solar cells, typic-
ally consisting of inorganic nanocrystals embedded in organic conducting poly-
mers [203]. Hybrid cells combine the low costs and flexibility of organic PV with
the environmental stability of inorganic materials [150]. In addition, inorganic
materials such as metal-oxides can be synthesised into vertically aligned nano-
structures [62, 83] (e.g, nanowires), combining the efficient charge separation
of a bulk heterojunction with clear electron transport paths. Nanowires also

23
provide enhanced optical properties due to their light trapping effects [56] and
waveguide effects [39]. This is elaborated on later in section 1.1.4.
Although inorganic electron acceptors should have multiple advantages (en-
hanced absorption and charge transport), the efficiencies of current devices have
not surpassed polymer-fullerene OPV [190]. This is mainly due to the problems
of nanoparticle surface chemistry (ligands forming an insulating layer) and the
morphology of the light absorbing layer [190].

1.1.3.4 Dye-Sensitised Solar Cells

It was known that the photosensitivity of platinum electrodes in an electrolyte
could be extended to longer wavelengths using a dye [187]. Dye-sensitised photo-
electrochemical solar cells were first developed in the late 60s [60]. O’Regan and
Gratzel took these concepts and extended them to nanocrystalline structures
coated with dye molecules, forming the modern DSSC [65, 141]. Encapsulated
in these devices is a mesoporous inorganic matrix (such as TiO2 nanocrystals)
coated with organic dye molecules and a liquid electrolyte (Figure 1.3). The
dye absorbs light, injects electrons into the TiO2 and regenerates electrons via a
redox reaction with the electrolyte. The highest efficiency recorded with DSSCs
so far is 11.5% [26].
The metal-oxide part of the cell while good at transporting electrons has a
wide bandgap and hence does not absorb visible light. Taking inspiration from
photosynthesis, the functions of light absorption and charge transport can be
separated, leaving the dye to absorb the light and the metal-oxide to transport
the electrons. The advantages are that such cells are not as sensitive to defects as
conventional semiconductors (such as Si) and they are much cheaper to produce
[187]. The energy produced from the absorption of photons can also be directly
used for photoelectrochemical processes such as splitting water into H and O2 .
In addition, DSSCs are flexible and the transparency of the electrolytic solution
allows novel applications such as photovoltaic windows [201].
The problem with these cells is that they easily degrade due to the liquid
electrolyte leaking, freezing or expanding. This problem can be solved by repla-
cing the electrolyte with a solid p-type semiconductor, but brings about further
challenges as it is difficult to establish intimate contact between the organic dye
and both solid semiconductors [65].

1.1.3.5 Solid-state Sensitised and Extremely Thin Absorber Solar

Cells
Many of the short-comings of DSSC solar cells can be overcome by replacing
the liquid electrolyte with a solid-state conductor, resulting in a solid-state dye-
sensitised solar cell (ssDSSC). Efficiencies of over 20% have been predicted for
such cells [40]. However, the highest efficiency of a ssDSSC (with organic dyes)
to this date remains at 7% [22].
One of the problems with the DSSC is that the best performance can only
be achieved by depositing a single mono-layer of dye over the surface of the

24
Figure 1.3: Dye-sensitised solar cell [65]. Photons are absorbed in the dye,
creating an exciton. The exciton dissociates, injecting an electron into the
conduction band of the TiO2 , while the electrolyte regenerates the dye. Original
figure by Gratzel [65], permissions granted.

mesoporous metal-oxide [40, 90]. One cannot increase the light absorption by
depositing more layers of dye molecules. A solution to this problem is by using
materials with higher absorption coefficients that continue to perform well when
more than a monolayer is deposited.
Semiconductor-sensitised solar cells (SSSCs) have been developed which
solve this problem. There are two types: extremely thin absorber (ETA) solar
cells [82] and quantum-dot sensitised solar cells (QDSSC). Cells sensitised with
an inorganic semiconductor (such as perovskite) instead of an organic dye have
reached efficiencies as high as 15% [40, 23]. SSSCs have the added benefit of
being more thermodynamically and optically stable than DSSCs [40]. Using a
solid-state hole conductor (such as P3HT) rather than a liquid electrolyte can
increase the stability as well as increase charge separation efficiency and reduce
interfacial charge recombination [92].
In the ETA cell, light absorption occurs in a thin solid-state semiconductor
coating (as opposed to a light-absorbing dye) on the order of 10nm thick. Being
so thin means that the material can be of very low quality (i.e. short diffusion
length) and shortens the distance the exciton has to travel, therefore reducing
the chance of recombination. By making the p-n interface structured [45], such
as in nanowire solar cells [110], one can use the coating as an intermediate
material where light absorption occurs separately from the charge transport.
The longer optical path length is effectively decoupled from the shorter charge-
carrier length.
SSSCs perform best when the coating has a near complete uniform distribu-

25
tion over the surface. This can be achieved either by chemical bath deposition
(CBD) or successive ion-layer adsorption and reaction (SILAR) [9]. However,
the thickness of the coating is crucial. Too thin and not enough light is absorbed.
Too thick and the recombination rate increases.

1.1.3.6 Perovskite Solar Cells

Perovskite materials have become very popular in the solar cell community
recently [17, 29]. Efficiencies of solar cells using perovskite absorbers have risen
from 3.8% to 18% in the space of 6 years. Perovskite-based solar cells are covered
in more detail in section 2.1.4, but are not part of the main topic of this thesis.

1.1.4 Nanowire Solar Cells

Using nanoscale columnar structures in the solar cell, (i.e. nanowires), one can
gain many benefits over solar cells with a planar interface [56]. Nanowires
are commonly synthesised using two methods: chemical vapour deposition and
patterned chemical etching [56]. Junctions in the nanowires are formed either
in a single material via dopant diffusion or by layering a secondary material
(coated nanowires). There are many potential benefits of using nanowires. They
have anti-reflection properties, since the change in effective refractive index (i.e.
volume averaged refractive index) between the nanowire layer and the continu-
ous layer is less abrupt than it would be in a planar interface. Nanowires also
have light-trapping properties, both due to inter-nanowire photonics and the
geometry of the nanowires themselves [39, 94]. For example, InP nanowire solar
cells have surpassed the ray-optics limit for light absorption and delivered 83%
of the photocurrent of planar devices even though nanowires covered only 12%
of the surface [182].
However, when compared to mono-crystalline solar cells, rather than increas-
ing the efficiency beyond current limits they reduce the quality and quantity of
materials required to reach those limits [56]. There is often a trade-off between
the improved absorption and the increased surface recombination [55]. On the
other hand, they have been experimentally demonstrated (with further support
from simulations) to significantly enhance the efficiency of hybrid solar cells
[176].
If the nanowires and the hole conducting medium are not sensitive to light,
they can be sensitised by coating with a thin light absorbing layer in the form
of nanoparticles [92, 184] or a conformal shell [45]. For example, nanowires
made from ZnO (which only absorb at ultra-violet frequencies) can be sensitised
to light by coating with CdSe or PbS (which have bandgaps that are small
enough to allow exciton formation by visible light). Furthermore, by distributing
the thin coating over the surface of the nanowires the optical path length is
enhanced, without compromising the charge transport. Having a thick planar
layer of CdSe would result in the same optical path length, but would lead to
lower efficiency due to a large charge-separation distance, and higher costs due
to the large amount of material required.

26
1.2 Introduction to Method
1.2.1 Optical Modelling
To study the light absorption in nanowire solar cells, one might at first consider
modelling the nanowires as an effective medium and using a simple Lambert-
Beer model. The intensity of light through a distance l in the effective medium
would simply be proportional to e ↵l , where ↵ is the absorption coefficient in
the material. However, it has been shown that an optimised nanowire solar cell
absorbs 500% more photons per unit volume than implied in the Lambert-Beer
model [188, 182]. Thus, a wave-optics approach is ultimately required to model
light absorption.
To calculate how much light is absorbed in coated nanowire solar cells,
a study was carried out on how light behaves in the separate components
(nanowire core, coating, and substrate). One can analytically calculate the
scattering and absorption cross-sections of side-illuminated homogeneous cylin-
ders using Mie Theory [178]. This can been extended to core-shell cylinders [89],
oblique angles [196], multiple cylinders [52] and absorbing mediums [107, 108].
Finite length cylinders can be treated to a certain extent [185]. However, for
more complicated structures and for illumination parallel to the nanowire axis it
is difficult to find analytical solutions to the electromagnetic fields and therefore
absorption spectra. One has to calculate the absorption and scattering of light
numerically.
For this thesis, simulations were performed of light scattering off the solar
cell components. To do this, Maxwell’s equations of electromagnetism were nu-
merically solved using the Finite-Different Time-Domain (FDTD) method. 2D
FDTD simulations were verified by comparing the scattering and absorption
spectra against analytical calculations for infinitely long cylinders, using Mie
Theory [178]. This allowed one to test that the simulations were setup correctly
before running the more complicated 3D simulations. Through these simula-
tions, one can vary the geometrical parameters of the solar cell to maximise
the amount of light absorbed. However, light absorption is only part of the full
picture. One must also take into account the charge transport, using the absorp-
tion calculated in the optical simulation as an electron-hole generation rate in
the charge transport model. By coupling both the optical and charge-transport
models, one can calculate the photocurrent of the device and then optimise the
efficiency of the solar cell.

1.2.2 Charge Transport Modelling

The aim of the charge transport model is to calculate J-V (current-voltage)
curves of the solar cell. A typical J-V curve for a solar cell is shown in Figure
1.4. From this, one can obtain the overall power efficiency,
J m Vm Jsc Voc ⇥ F F
⌘= =
Ps Ps

27
Figure 1.4: The current-voltage curve (black) and power-density (grey) of an
ideal solar cell [137]. The maximum current density (short-circuit) is Jsc . The
maximum voltage (open-circuit) is Voc . The maximum power density, P =
Jm Vm is given by the area of the small rectangle. The larger rectangle gives the
power density P = Jsc Voc if the fill-factor F F is one. Original figure by Nelson
[137], permissions granted.

28
where Jsc is the short-circuit current density, Voc is the open-circuit voltage, Ps
is the incident power intensity, Jm and Vm are the current density and voltage at
the maximum power yield and F F = Jm Vm /(Jsc Voc ) is the fill-factor, describing
the “squareness” of the J-V curve. To obtain the J-V curves, one has to solve
the drift-diffusion equations for the electrons/holes and the Poisson equation
for the electrostatic field, taking into account the charge transport properties
of each material and the interfaces between them. The generation rate of the
electrons and holes is determined by the light absorption calculated in the optical
simulations.
The drift-diffusion equations describe the charge carrier densities and cur-
rents as a function of position (and time, when not looking at steady state
solutions). Making a few approximations allow these equations to be solved
analytically for a 1D p-n junction [137]. Kayes et al have extended this model
to radial p-n junctions [85].
To model the charge transport of a coated nanowire array embedded in a
hole-conducting medium, more complicated geometries and higher dimensions
are required due to the interfaces and wave-optics. Such models can only be
solved by numerical means. The drift-diffusion equations can be discretised in
the finite-difference scheme and solved using the Newton method [158].
Full coupled optical-electrical device physics models have been utilised by
Li et al [112, 113] and Deinega et al [35]. Li et al used COMSOL to couple
optical absorption profiles to carrier transport calculations order to compute
the photocurrent of plasmonic solar cells. They wrote a custom plugin for the
carrier transport and used the in-built RF modules to do the optical modelling.
COMSOL uses a finite-element method (FEM), where the system is discretised
into triangular elements and solved in the frequency domain.
Deinega et al have coupled FDTD optical simulations to a finite-difference
drift-diffusion solver. The latter uses a C++ code called Microvolt. Since this
is free, open source, and remarkably good at simulating cylindrical geometry,
this is what I have used for my research.

1.2.3 Coupling of Optical and Charge Transport Methods

The charge transport method uses a generation rate of elecron-hole pairs to
calculate the charge carrier dynamics. This generation rate could be calculated
using the Lambert-Beer model for attenuation of light. For more detail, tak-
ing into account sub-wavelength features, one should use FDTD simulation to
calculate the generation rate (as mentioned earlier). In either case, the optical
illumination and charge transport are treated separately (the coupling is only in
one direction). One could use the electric fields calculated from the charge trans-
port model and input those into the FDTD simulation (bi-directional coupling).
Such an approach has been taken for modelling the high frequency conductivity
of doped Silicon [189]. However, the fields generated by charge carriers are so
small compared to the solar illumination, that for photovoltaic applications it
is not necessary to couple the charge transport fields into FDTD.

29
1.3 Thesis Outline
There are many types of solar cell and the literature is extensive. There are few
studies of ETA nanowire solar cells, and only a couple of theory and simulation
investigations of such: one study is that of Mora-Sero et al [134]. Another
study which looks at core-shell nanowires is by Tsai et al [176]. Currently there
are no studies of how to optimise coated nanowire solar cells with respect to
geometrical parameters such as length, radius, coating thickness and spacing.
Thus, this subject area warrants research.

1.3.1 Devices being Studied

This thesis focuses on two different solar cells. The first consists of n-type TiO2
nanowires coated with a conformal layer of CdSe (light sensitiser) and embed-
ded in a hole-conducting polymer (see Figure 1.5). Most studies of TiO2 /CdSe
nanowires have focused on those with CdSe nanoparticles [123, 184, 204]. It
should be possible to create a compact conformal layer of CdSe, using the SILAR
method [166] or by annealing nanoparticles. So far studies of this kind of device
have focused on the optics. I have carried out full optical simulations of the
device using Lumerical FDTD Solutions [120]. This software uses the Finite-
Difference Time-Domain method (FDTD). The amount of light absorbed in
Watts/m3 can be obtained. Results were obtained showing that nanowires out-
perform planar solar cells optically (see Figure 5.7) and that thin nanowires
with thick coatings are optimal. Results are obtained from coupling FDTD
generation profiles to finite-difference charge transport calculations.
Another device (Figure 1.6) I have looked at consists of n-type ZnO nanowires
coated with PbS nanoparticles. Like the former device, FDTD simulations have
been carried out to calculate the absorption spectra for different sweeps of para-
meters: length, radius, spacing, and number and size of nanoparticles. The
absorption per unit volume per nanoparticle is derived and conformal coatings
are compared to nanoparticle coatings.

1.3.1.1

1.3.2 Summary of Contents

Chapter 2 of this thesis reviews the literature on coated nanostructured solar
cells. This begins with ETA solar cells, is followed by studies on light trapping
and the electromagnetic modelling of solar cells, and then covers the charge
transport studies.
Chapter 3 covers the optical methods used for analysing the light scattering
and absorption of solar cells. It begins with the theory (primarily Mie Theory)
and numerical methods (FDTD). This is followed by a detailed description of
how the core-shell and nanoparticle nanowires were modelled in FDTD, ending
in a section devoted to the computational resources used.
Chapter 4 covers the theory and modelling of charge transport in solar cells.
It starts by constructing a theory of charge transport in crystalline semiconduct-

30
 (a) (b)

Figure 1.5: (a) Schematic of the nanowire array solar cell. The nanowires consist
of CdSe coated TiO2 . These are embedded in a hole conducting polymer. The
diagram shows a small section. The solar cell is modelled with periodic boundary
conditions resulting in an infinite array of nanowires. In each nanowire, the
TiO2 core connects to a TiO2 substrate which is in contact with the anode.
The polymer fills in all the space between the gold cathode, TiO2 substrate and
nanowires. (b) Side view

PbS QD TiO2
50nm diameter = 4 to 9nm 70nm

1.2μm

Figure 1.6: Diagram of the nanoparticle sensitised solar cells (B.ElZein,

E.Traversa 2014). The n-type layer consists of ZnO nanowires whilst the p-
type light-absorbing layer consists of PbS nanoparticles.

31
ors, an analysis of the p-n junction, and the derivation of the drift-diffusion equa-
tions used for calculating photocurrent. This is followed by a detailed section
explaining the numerical method used for solving the drift-diffusion equations
how it was used to calculate the photocurrent in TiO2 /CdSe nanowires.
Chapter 5 details the results obtained from all the optical and charge trans-
port modelling. It begins with the tests to verify that FDTD gives the same
results as analytical Mie Theory. This is followed by the absorption spectra,
absorption field profiles and maximal photocurrent densities of the TiO2 /CdSe
core-shell nanowires. The next section details the charge transport calculations
for the core-shell nanowires and the maximum power yields. After this is the
absorption spectra and absorption per unit volume of various ZnO/PbS QD
nanowires. The chapter ends with a discussion of the results and what they
mean.
Finally, chapter 6 contains the conclusions and final remarks of the thesis.

32
Chapter 2

Literature Review

2.1 Review of ETA cells

Briscoe and Dunn have written a comprehensive review on extremely thin ab-
sorber (ETA) solar cells [19]. The concept of ETA cells originated from dye-
sensitised solar cells (DSSC) [64]. The high surface area of DSSCs allows a thin
dye to be used for absorbing a significant amount of light. DSSCs are however
plagued with poor stability due to reactions between the electrolyte and electron
transporter [175] and degradation of the dye or other components [191]. These
problems could be solved if one were to replace the liquid electrolyte with a solid-
state hole-conductor and the dye with an inorganic semiconductor. The first
solid-state dye-sensitised solar cell (ssDSSC) was created in 1995 [174], where
the electrolyte was replaced with CuI. This was followed by another ssDSSC in
1998 using CuSCN as the electrolyte replacement. [143]. Further improvements
in solar cell design have followed, leading to the full solid-state ETA solar cell,
as detailed in the following sections.

2.1.1 Porous TiO2 cells

TiO2 is often used in DSSCs and ETA cells for the electron conductor, being
n-type due to oxygen vacancies. It has a high bandgap and therefore does not
absorb light in the visible spectrum, so it is usually sensitised with dye molecules
or inorganic semiconductors which can absorb light. The TiO2 in solar cells,
consisting of nanoparticles sintered together, is porous and hence provides a lot
of surface area for light sensitisation.

2.1.1.1 Light Scattering in Porous TiO2

Ferber et al [48] have conducted a study of light scattering and absorption in
TiO2 DSSCs. Since the porous TiO2 consists of 10-30nm sized coated particles
sintered together, they use Mie theory applied to coated spheres. The nanopor-
ous structure of the TiO2 increases the effective surface area 1000-fold compared

33
Figure 2.1: Efficiency of a solar cell as a function of bandgap [137], based on
the Shockley-Quiesser limit for a single p-n junction [160]. Original figure by
Nelson [137], permissions granted.

to a planar interface. If one increases the size of the TiO2 particles, one can
increase the light scattering and hence absorption. However, the surface area
is reduced. Their calculations show that using a mixture of different sizes op-
timises the absorption, since there is both a high amount of Mie scattering and
large surface area.

2.1.1.2 Inorganic Absorbers

In 1990, porous TiO2 was sensitised to light by coating with quantum sized CdS
particles (4-20nm) rather than an organic dye [181]. However, the bandgap of
CdS is quite high (2.4eV), leading it to be transparent to most of the solar spec-
trum. Figure 2.1 shows how important the bandgap is, bounding the theoretical
efficiency at 33% due to the Shockley-Quiesser limit [160]. The maximum effi-
ciency ⌘ is a balance between a small bandgap where the photocurrent is high
(more light absorbed) but photovoltage low (since V < Eg /e where e is the
electronic charge) and large bandgap where the photocurrent is low (less light is
absorbed) but photovoltage is high. In addition to having a prohibitively large
bandgap, the CdS reacted with the electrolyte, reducing the IPCE by 50% over
a period of a few weeks [181].
In 1993, CdSe was used instead of CdS [118], which has a lower bandgap of
1.7eV. This sensitised the TiO2 from previously 300nm to up to 700nm. Like
with the CdS device mentioned earlier, the oxidation of the electrolyte at the
CdSe interface caused the photocurrent to decay after just 5 seconds of illu-
mination, due to the increase in recombination and decrease in hole extraction
[118].

34
Figure 2.2: SEM image of a 3D solar cell [11, 19]. CdTe is electrochemically
deposited onto TiO2 , acting as both the absorber and the hole-collector. The
thick black line highlights the interface between the CdTe and TiO2 . The highly
convoluted interface between the absorber layer and electron-conductor (in this
case TiO2 ) leads to increased light scattering and hence optical path length
enhancement. Original figure by Belaidi et al [11], permissions granted.

2.1.1.3 3D cells
Cells have been created with an inorganic absorber but with no dedicated hole-
conductor layer. For this type of cell the absorber is more than 100nm thick,
acts as a hole-conductor and makes direct contact with the electrode. These
cells are designated as 3D cells [19]. They have a convoluted n-type layer (typ-
ically porous TiO2 ) on to which a solid-state semiconductor absorber layer is
deposited. Figure 2.2 shows the cross-section of a 3D cell [11]. These do not
strictly count as ETA cells, since the absorber layer is too thick and doubles as
a hole-conductor [19]. Examples of absorber layers used in 3D cells are CdTe
[45] and CuInS2 [136]. The porous TiO2 leads to a higher efficiency than planar
TiO2 due to the light scattering, enhancing the optical path length by a factor
of 5 [19]. The highest efficiency achieved with this type of cell to date is a
TiO2 /CuInS2 3D cell with an efficiency of 7% [63].

2.1.1.4 Thin Film Absorbers

Extremely thin absorber (ETA) solar cells are defined as those with an inorganic
semiconductor absorber layer on the order of 10nm thick, and separate electron
conductor and hole-conductor layers. The first complete solid-state ETA cell was
created in 1998 [173]. This consisted of a 6µm thick porous TiO2 film coated
with a 23nm Se thin film formed by electrodeposition and annealing. This was
then filled in with 10µm of CuSCN from an acetonitride solution. This cell

35
has a low efficiency of 0.13% due to the high surface recombination and short-
circuiting between the Se absorber and electrode due to voids in the TiO2 [173].
However, this cell remained stable after illumination, unlike dye-sensitised or
electrolytic cells.
Other ETA films [19] deposited on porous TiO2 include CdTe [132, 46], ZnTe
[46] and CuSCN [99, 82] by electrodeposition and CdS by SILAR [132]. Using
SILAR led to inhomogeneities and incomplete filling. ILGAR [132] was later
used to solve these problems for depositing CdS, Cu2S, In2 S3 and CuInS2 . In
this method, the absorber metal cation precursor is adsorbed onto the TiO2 and
then sulpherised using a hydrogen sulphide gas. This resulted in a conformal
and homogeneous coating with a controllable thickness and better infiltration.
A cell made using TiO2 with ILGAR deposited CuIn2 S2 and electrochemically
deposited CuSCN [82] resulted in a good rectification and no short-circuiting.
Organic polymers have also been used as the hole-conductor in ETA cells,
specifically polystyrene sulphonic acid doped poly(3,4-ethylenedioxythiophene),
abbreviated as PEDOT:PSS. Bayon et al [9] have spin-coated PEDOT:PSS from
an aqueous solution onto porous TiO2 , where PbS was used as an absorber
deposited by CBD or SILAR. Performance was limited in this type of cell due
to the fact that PEDOT:PSS is not completely transparent.

2.1.1.5 Nanoparticle/Quantum-dot Absorbers

The absorption coefficient and bandgap increase as particles get smaller. Quantum-
dots allow narrower bandgap materials to be used due to quantum confinement
effects which increase the bandgap [19]. Varying the size of the quantum-dots
allows the bandgap to be tuned to optimise performance. The first solid-
state cell based on nanoparticle absorbers was created by Plass et al in 2002
[147]. They used the organic hole conductor spiro-OMeTAD [4], which is short
for 2,2’,7,7’-tetrakis(N,N-di-p-methoxyphenyl-amine)9,9’-spirobifluorene. PbS
quantum-dots were grown on the surface of the TiO2 , using a lead nitrate solu-
tion followed by a Na2 S solution. This process was repeated to get different
sized quantum-dots. In 2009, Lee et al [103] also created a PbS quantum-dot
sensitised solar cell but used methanol instead of water as a solvent. This res-
ulted in a coverage of 30-40%. Attempts to increase the efficiency by increasing
the coverage were not successful due to a reduction in quantum confinement
and hence reduction of the efficiency of electron transfer.
Microstructuring has also been attempted in solar cells to increase light
scattering and hence absorbance. In 2007, Biancardo et al [15] created mi-
crostructured TiO2 /PbS/PEDOT:PPS ETA cells using a grooved silicon press.
Despite the increased absorbance, the microstructuring did not produce bet-
ter performance than a planar cell. In fact, the efficiency was decreased, due
to short-circuiting. Care must be taken when implementing micro or nano-
structuring in a solar cell to increase absorbance.

36
2.1.2 ZnO Nanorod Cells
In addition to porous TiO2 , ZnO nanorods have been synthesised for usage
in solar cells. The advantage of this morphology is the direct path to con-
tact, increase in optical path length and the ease at which the nanorods can
be coated compared to porous TiO2 [19]. Garnett et al [56] have reviewed
nanowire/nanorod type solar cells where the benefits of this geometry are fur-
ther explained. In addition, ZnO has a higher electron mobility than TiO2 (200
cm2 V 1 s 1 compared to 10 cm2 V 1 s 1 for TiO2 ) [19].

2.1.2.1 Thin Film Absorbers

In 2000, ZnO nanorods had a-Si deposited on them using chemical vapour de-
position (CVD), resulting in a uniform and conformal coating [98]. ZnO nanor-
ods with a CuSCN hole-conductor have been sensitised with CdTe [170], and
annealed CdSe resulting in an efficiency of 2.3% [172].
In2 S3 has also been used as an absorber, deposited by the ILGAR method
[13]. ILGAR allowed the coating to be controllable and conformal, from 10nm to
75nm. Increasing the coating thickness decreased Jsc (due to the diffusion length
being ⇠10nm) but increased Voc , due to reduction in tunnelling recombination
through the absorber. The optimum thickness of In2 S3 was found to be 25nm.
The effect of changing the length of the nanorods (from 0 to 3.3µm) was
also studied [91]. Increasing the length led to an increase in Jsc but decrease in
Voc and fill factor, due to the increase in surface recombination. The optimum
length was found to be 800nm. The performance of the cell could be improved
by annealing at 200 C for 2 minutes [12]. The result of the annealing was
that the Cu would diffuse to the interface and react with the In2 S3 absorber,
creating extra sub-bandgap states [12, 19]. This would shift the In2 S3 bandgap
and hence reduce the recombination, increasing Voc and shifting absorption to
longer wavelengths thus increasing Jsc . The resulting efficiency was 3.4%, the
highest yet for ETA solar cells [19].
ZnO nanorods with a polymer hole-conductor have also been produced. The
nanorods were coated with CdSe QDs and then annealed to form a continuous
film of CdSe [135]. This annealing led to an increase in efficiency and shifted
the absorption to longer wavelengths, due to the loss of quantum confinement.
Figure 2.3 shows the increase in efficiency during the annealing process. An
efficiency of 0.9% was achieved for cells using the MEH-PPV polymer, and
1.5% for P3HT. In 2007, Greene et al [68] created ZnO nanorod cells with
a P3HT hole-conductor. They found that coating ZnO nanorods with TiO2
increased the efficiency by more than 5 times. However the efficiency was so
low anyway, increasing from 0.05% to 0.29%. Two reasons were given for the
poor performance: the ZnO nanorods do not form a good charge separation
interface with P3HT and the nanorod array was too densely packed (especially
after coating with TiO2 ).

37
Figure 2.3: External quantum efficiency of CdSe coated ZnO nanorod solar
cells, under different durations of annealing [135]. The efficiency increases when
annealed up to 24 hours. Further annealing decreases the efficiency. Original
figure by Nadarajah et al [135], permissions granted.

2.1.2.2 Nanoparticle/Quantum-dot Absorbers

ZnO nanorods have had quantum-dots deposited on them, ranging from CdTe
[25], CdS [202], CuInS2 [102] and CdSe [109]. These cells used a liquid electrolyte
rather than a solid-state hole-conductor, so in that sense they are not proper
ETA cells as they are not fully solid-state [19]. In the case of the CuInS2 and
CdSe, QDs were grown before deposition. Capping molecules were attached to
the QDs to prevent agglomeration and the nanorods were soaked in solution for
several days. This resulted in less than a monolayer on the surface, and the
lower surface area of nanorods compared to porous TiO2 meant that these cells
were limited to less than 1% efficiency because they could not absorb much light
[19].

2.1.3 TiO2 Nanowire Arrays

In addition to ZnO nanorods, solar cells consisting of TiO2 nanorods (or nanowires
as they are often called) have been synthesised, using chemical vapour deposition
(CVD) [104] or the hydrothermal method [117].
In 2012 Zou et al fabricated TiO2 nanorod arrays coated with CdSe nan-
oparticles [204] and filled in with CuSCN hole-conductor. They use thermal
annealing to remove the organic molecules on the CdSe (to improve electron
transport). The annealing process makes the CdSe nanoparticles lose their

38
Figure 2.4: TiO2 /CdS/CdSe nanowires with 4 CdS layers and n CdSe layers
where n = 0 to 4 [111]. Original figure by Li et al [111], permissions granted.

quantum confinement effect, due to the grain growth [204]. An efficiency of

0.21% was achieved.
In 2012, Wang et al [184] studied the effect of CdSe quantum-dots on TiO2
nanowires with a P3HT polymer hole conductor. The quantum-dots were shown
to enhance absorption and charge separation, as well as reduce recombination
due the more gradual stepwise change in band edges in P3HT/CdSe/TiO2 .
In 2010 Li et al [111] coated TiO2 nanowires with both CdS and CdSe
quantum-dots at the same time, to optimise both the absorption and charge
transport. They used the SILAR method and found that cells were optimal
with 4 layers of CdS and 3 layers of CdSe, achieving an efficiency of 1.14%. CdS
has a larger bandgap than CdSe. It was not made clear in this study whether
co-sensitisation was better than just having 7 layers of CdSe. Figure 2.4 shows
the absorbance of the co-sensitised nanowires under various numbers of SILAR
cycles.
Sani et al studied the optical properties of TiO2 nanowires [154], comparing
theoretical calculations with experiment. In their study, they use the effective
medium approximation (EMA) method to calculate the optical properties such
as dielectric function, absorption coefficient and reflectance. Figure 2.5a shows
that the reflectance for nanowires is less than bulk TiO2 . This is due to scatter-
ing from the TiO2 nanowires and increase in voids roughness [154]. Their EMA
calculations also agree well with experiment. It has to be noted however that
the nanowires in their study lie horizontal and randomly arranged as opposed
to vertical and ordered (see Figure 2.5b).

39
 (a) (b)

Figure 2.5: (a) Reflectance of TiO2 nanowires calculated from EMA method and
experiment, and reflectance of bulk TiO2 [154]. There is good agreement with
experiment, and the nanowires have less reflectance than bulk TiO2 . (b) SEM
image of TiO2 nanowires [154]. Original figures by Sani et al [154], permissions
granted.

2.1.4 Perovskite Sensitised ETA cells

Hybrid organic-inorganic materials with the perovskite structure have recently
shown considerable promise and become popular in the solar cell community
[17, 29]. This is because of the direct bandgap, large absorption coefficient
and high carrier mobility of perovskites [47]. Efficiencies of solar cells using
perovskite as the absorber have risen rapidly from 3.8% in 2009 [97] to more
than 18% in 2015 [77]. Most of these solar cells have a thin-film of perovskite
on the order of hundreds of nanometers thick. There are however a few solar
cells with extremely thin coatings of perovskite (⇠10nm).
Lee et al [106] have created ETA cells using a mesoporous TiO2 or alumina
scaffold and a perovskite absorber, shown in Figure 2.6. They compare using
Al2 O3 to TiO2 . Interestingly, the efficiency was increased from 8% to 10.9%
when replacing the TiO2 with Al2 O3 . In this case, the Al2 O3 is optically and
electrically inert and the perovskite acts as both the absorber and electron
conductor. See Figure 2.7 for the band diagrams of the TiO2 and Al2 O3 versions
of the cell. Further enhancements in performance can be expected by using
new perovskites with extended absorption edges [106]. Etgar et al [47] have
also synthesised perovskite sensitised TiO2 solar cells. Rather than having a
separate hole-conducting material, they use the perovskite as both an absorber
and hole conductor. An efficiency of 5.5% was achieved.
Lead iodide perovskite sensitised solar cells have been synthesised by Kim et
al [93]. An efficiency greater than 9% was achieved. In this case, nanoparticles
were embedded into porous TiO2 which was then filled in with a hole-conducting

40
Figure 2.6: Perovskite ETA cell, synthesised by Lee et al [106]. The mesoporous
TiO2 electron conducting layer is sensitised with a perovskite absorber. Spiro-
OMeTAD is used as a hole-conducting polymer. Original figure by Lee et al
[106], permissions granted.

polymer, spiro-OMeTAD. Since the perovskite absorber does not consist of a

continuous layer, this cell does not strictly fall under the ETA category.
TiO2 nanowire arrays (embedded in spiro-OMeTAD) have been sensitised
with perovskite nanoparticles [149] by Qiu et al, shown in Figure 2.8. These
nanoparticles range in size from 10 to 20nm. An efficiency of 4.8% was achieved.
This solar cell is interesting as it is one of the very few all-solid-state cells
consisting of TiO2 nanowire arrays. Since publication of this study in 2013,
only two other perovskite sensitised TiO2 nanowire solar cells have been found
in the literature [78, 200]. Again, this cell is not a proper ETA cell due to the
absorber not being a continuous thin film.

2.1.5 Optimisation of ETA cells

The optimisation of absorber thickness in ETA cells is very complex. For ex-
ample, for ZnO nanorods with a In2 S3 absorber [13], increasing the absorber
thickness leads to lower Jsc , due to the high recombination losses. On the
other hand, Voc increases due the decreased tunnelling recombination. Studies
of light scattering on ZnO nanorods have been done [171] showing that up to
88% absorption is achievable for a CdSe shell thickness of 20nm. The optimum
absorber thickness has been calculated theoretically for ETA cells [169]. This is
50nm for CdTe and 23nm for CuInS2 . However, the model used in this study
is full of many assumptions (as discussed in the next section) and does not take
wave-optics into account, which is important for nanowires.
In 2013, Levy-Clement et al [110] optimised ETA ZnO-nanorod/CdSe solar
cells. They tried increasing the length of the nanorods, aiming to increase
performance by increasing the scattering and decreasing the required CdSe ab-
sorber thickness. However, by increasing the nanorod length, the performance
decreased due to the extra surface recombination. So they tried varying other
parameters. The diameter of the nanorods was increased from 100nm to 330nm.

41
Figure 2.7: Band diagrams comparing perovskite ETA cell with TiO2 or Al2 O3
[106]. In the Al2 O3 case, the conduction band is higher than the perovskite, so
electrons cannot be transported through. Instead, they are transported through
the perovskite to the electrode. When TiO2 is used, this transports the electrons
and the perovskite functions solely as an absorber. Original figure by Lee et al
[106], permissions granted.

42
Figure 2.8: Perovskite sensitised TiO2 nanowire arrays by Qiu et al [149]. Fig-
ures (a) and (b) show SEM images of the bare TiO2 nanowires. Figure (c) and
(d) show SEM images of the TiO2 nanowires coated with perovskites. Figure (e)
shows a cross-sectional SEM image of the full solar cell including the spin-coated
spiro-OMeTAD. Figure (f) shows a diagram of the full solar cell. Original figure
by Qiu et al [149], permissions granted.

43
Figure 2.9: Schematic of ETA solar cell as modelled by Taretto et al [169].
(a) shows the real structure with its convoluted folds. The band diagram for
the segment A to A’ is shown in Figure 2.10. (b) shows the comb-like structure
used in this model where there are N p-i-n type elements connected in parallel, a
dashed line enclosing each element. A p-i-n junction consists of a p-type layer,
intrinsic layer (no doping) and n-type layer connected in series (see section
4.1.4.2). Original figure by Taretto et al [169], permissions granted.

This increased the range of wavelengths at which scattering occurred, closer to

the bandgap of CdSe, thereby increasing the absorption. The optimum power
efficiency of 3.2% was achieved with a CdSe thickness of 30nm, 230nm diameter
nanorods and 1.3 µm length.

2.1.5.1 Modelling of ETA cells

In 2004 Taretto et al [169] constructed a theoretical model of ETA cells and
used it to calculate the optimum parameters. In this study they determined
that the minimum thicknesses should be 15nm for CdTe and 20nm for CuInS2 ,
the optimum thicknesses 50nm and 23nm respectively and that a maximum
efficiency of 15% was possible with these absorbers. They modelled the ETA
cell as a simplified comb-like structure (see Figure 2.9), where the convolutions
in the electron-conductor/absorber/hole-conductor interface are approximated
as multiple p-i-n type elements connected in parallel. Each of these elements
has a simplified band energy diagram applied to it, shown in Figure 2.10.
Immediately there is an issue with applying this model to the TiO2 /CdSe
solar cell studied in this thesis. If the difference in workfunction of the n and p
layers is much less than the bandgap of the absorber, then modelling at different
length scales is necessary [169]. The band diagram of TiO2 /CdSe/polymer looks
nothing like the simplified model.
Assuming one uses the simplified band diagram, the j-V curve of each element
can be calculated using the p-i-n type expression for j(V ) derived by Burgelman
and Grasso [21, 169]. This model also makes some other assumptions. One is
that the generation rate G in the absorber is spatially constant. The problem
with this assumption is that while it is acceptable for simple geometries, for

44
Figure 2.10: Simplified p-i-n type band diagram for the segment A to A’ in
Figure 2.9a [169]. The bands of the n-type and p-type materials are staggered
in such a way that the bandgap of the absorber is equal to the change in work-
functions. Original figure by Taretto et al [169], permissions granted.

nanowire cells the position dependence of G is important. Another assumption

made is that µ and ⌧ are the same for electrons and holes. For equal ⌧ to hold,
the difference between the conduction band or valence band and the defect level
must be the same, i.e. En = Ep = Eg /2 [169]. This is a direct consequence
of the simplified band model and does not hold for the TiO2 /CdSe/polymer cell
studied in this thesis. This model also makes the assumption that resistivities in
the n and p layers are so low that the voltage drop is negligible, but for TiO2 and
the polymer this is not true. The final assumption is that the n and p layers are
assumed not to be photoactive. This is acceptable for TiO2 (since its bandgap
is in the UV region) and the polymer (which is chosen to be transparent for this
thesis).
While there are many issues with this model, it importantly shows that
moderate efficiencies of around 15% are possible with ETA cells despite the
usage of cheap and low quality materials.

2.1.6 Summary of ETA cells Performance

ETA cells were produced with efficiencies ranging from less than 1% up to 7.6%
(10.9% if the absorber doubles as an electron-conductor). Efficiencies of up to
7% were achieved for TiO2 3D cells [63], 3.37% for TiO2 ETA cells [76], 3.4% for
ZnO-nanorod ETA cells [13] and 7.6% for perovskite ETA cells [106] (10.9% if
the perovskite acts as both an absorber and electron-conductor). Non-perovskite
ETA cells have a much lower efficiency than 3D cells due to the insufficient ab-
sorption of light and the fact that the hole-collector/absorber interface increases
recombination [19]. So far, only the perovskite-based ETA cells have exceeded
the 7.12% efficiency of the original DSSC [19, 141]. Based on modelling studies

45
[169] and the recent achievements of perovskite based ETA cells [17], efficiencies
approaching those of conventional silicon solar cells (20-30%) are likely possible.
The next sections detail optical modelling studies on nanostructured solar cells,
in particular nanowires, determining whether performance enhancements can
be gained by controlling the nanostructure.

2.2 Light-Trapping in Solar Cells

Nanostructures in solar cells can trap light by coupling it into localised and
guided modes [49]. Using these nanostructures enhances the absorption and
reduces the amount or quality of absorber material required, making thin films
comparable to crystalline solar cells. There are a variety of different ways to
nanostructure the metals, dielectrics or semiconductors in the solar cell. These
include photonic crystals, plasmonic nanoparticles, nanowires and gratings [49].
Electromagnetic modelling methods are useful because they allow high through-
put calculations to determine the best shapes, sizes and arrangements of nano-
structures for light-trapping. Such methods include finite-element (FEM), rigor-
ous coupled wave analysis (RCWA), transfer-matrix methods (TMM), discrete
dipole approximations and semi-empirical methods [49].
EM modelling can be coupled with electrical models to ultimately determ-
ine the photocurrent. For example, Li et al [112] have used multiphysics FEM
methods to model the optical absorption and electrical charge transport of plas-
monic solar cells. Deinega et al [35] have coupled finite difference time-domain
(FDTD) optical simulations with finite-difference charge transport to model
nanowire solar cells.
Vivian et al [49] have used EM modelling to study nanostructured solar cells.
They investigate localised modes, which are coupled into surface plasmons in
plasmonic nanoparticles, and waveguide modes which propagate throughout
the cell and are photonic in nature. They determined that with an optimised
nanostructure design, waveguide modes enhance absorption in the red end of
the spectrum and localised modes enhance absorption in the blue end, when
compared to flat and Asahi [50] textures. However, the absorption remains
below the Yablonovitch 4n2 limit [192] for random textures, where n is the
refractive index.
Battaglia et al [7] have investigated whether periodic nanostructures can
outperform random nanostructures, focusing on a-Si cells. Previously, the best
a-Si thin film solar cells had randomised pyramid structures on the transparent
ZnO electrodes. Yu et al [198] show that theoretically a maximum enhancement
of about 14.5n2 is achievable for triangular gratings. Battaglia et al show that
enhancement above the 4n2 Yablonovitch limit is possible for a broad spectral
range. However, these limits are derived assuming that the absorber thickness is
much greater than the wavelength, which is not necessarily the case. Guided by
these theoretical results, they develop periodic nanostructured solar cells with an
efficiency of 10.9%, matching those with random nanostructures. If it weren’t
for the electrical losses in Voc and FF, their periodic cells would outperform

46
Figure 2.11: Plasmonic light trapping schemes [3]. (a) Light scattering off
metallic nanoparticles, (b) excitation of localised surface plasmons, (c) grating
coupling to surface-plasmon-polaritons. Original figure by Atwater et al [3],
permissions granted.

random cells.
Plasmonics can improve light trapping in one of three ways [3]: light scatter-
ing from metallic nanoparticles embedded at the surface, excitation of localised
surface plasmons in embedded nanoparticles and excitation of surface-plasmon-
polaritons (SPPs) at corrugated metal/semiconductor interfaces. Figure 2.11
shows diagrams of these processes. Plasmonic nanoparticles have been exper-
imentally confirmed to improve the light-trapping to almost the same amount
as random textures [167].
Another way to improve light-trapping is by using nanowires. By carefully
engineering the nanowires diameter and distribution of p-i-n doping, Wallentin
et al [182] have achieved 13.8% efficiency using InP nanowires covering only 12%
of the surface. This is due to the resonant light trapping in the nanowires. The
amount of light converted to photocurrent was shown to be six times greater
than what would have been calculated with a ray-optics model. They used
3D electromagnetic simulations to model the optical absorption, taking into
account wave-optics and allowing them to understand what parameters lead to
the highest absorption.
Wang et al [183] used genetic algorithms to optimise periodic light-trapping
structures. Rather than manually varying pre-defined topologies (such as nanowires,
pyramids, cones, etc) they generated 2D scattering patterns (represented as ar-
rays of bits), calculated their absorption coefficient with the RCWA method
then used a combination of selection, crossover and mutation of these patterns
over several iterations to achieve the best structure. The resulting structure
(see Figure 2.12) achieves a 3-fold increase over the Yablonovitch limit. This
study shows how crucial electromagnetic modelling methods are for improving
the light-trapping properties of solar cells.

47
Figure 2.12: Optimised light trapping structure resulting from genetic al-
gorithms [183]. (a) Unit cell of the optimised light trapping pattern. (b) Mul-
tiple tiles of this pattern. (c) Absorption spectra of the optimised pattern and a
random pattern for comparison. Original figure by Wang et al [183], permissions
granted under the Creative Commons licence.

2.3 Electromagnetic Modelling of Nanostructured

Solar Cells
A wide variety of electromagnetic methods have been developed and used for
modelling solar cells. These include the transfer-matrix method (TMM) and
scattering matrix methods, used for analysing optical absorption in periodic
structures such as nanowire arrays [75, 2]. Then there are the 3D electro-
magnetic methods which solve Maxwell’s equations numerically, such as rigor-
ous coupled wave analysis (RCWA), finite-element method (FEM) and finite-
difference time-domain method (FDTD). Zanuccoli et al to used RCWA to ana-
lyse absorption in heterojunction silicon nanowires [199]. The finite-element
method (FEM) was used by Li et al [112, 113] for modelling plasmonic solar
cells. The FDTD method is most common for modelling nanowires, due to
the ease of use, computational efficiency and wide selection of software tools
available.

2.3.1 FDTD Simulations of Nanowires

2.3.1.1 TiO2 Nanotubes
Mor et al [133] have written a review on solar cells (and other solar energy
devices) based on ordered, vertical TiO2 nanotube arrays. They investigate

48
 (a) (b)

Figure 2.13: (a) 2D FDTD model of TiO2 nanotubes [133]. (b) Measured results
vs FDTD results of the transmittance of TiO2 nanotubes [133]. Original figures
by Mor et al [133], permissions granted.

cylindrical vs tapered nanotubes, pore size, length and wall thickness. In this
case the solar cells are dye-sensitised and contain liquid electrolyte. Of particular
relevance to this thesis are the FDTD simulations (Figure 2.13a). Ong et al [140]
have also carried out FDTD simulations of nanotube arrays. They show that the
absorption is increased with nanotube length, smaller pore size and increased
surface roughness. As shown in Figure 2.13b, the measured results appear to
agree well with the simulations, despite the simulations being 2D. It should be
noted, however, that the geometry of the nanotube is more correctly represented
in 2D than a solid cylindrical nanowire (which could support three-dimensional
waveguide modes).
Foster and John [53] have taken this a step further and numerically demon-
strated how photonic arrays of TiO2 nanotubes (with a sinusoidal diameter)
enhance the absorption by 1/3 compared to conventional TiO2 porous DSSCs.
The spacing and diameter of the tubes are chosen to promote dielectric modes
that concentrate light into the interior of the tubes. The sinusoidal modula-
tion in the diameter enhance the anti-reflection properties. The method used
for simulating these nanotubes was a frequency-domain Finite-Element Method
(FEM) rather than FDTD.

2.3.1.2 Semiconductor Nanowires

There is a large amount of literature on FDTD simulations of semiconductor
nanowires. Kim et al have used FDTD to study Fabry-Perot resonances in
core-shell p-i-n nanowires as a function of morphology [94]. However, they
study side illumination on a single nanowire. While this is useful for verifying

49
Figure 2.14: Plots of specular reflectance of the base-tapered InP nanowires
[39]. Measured results (a) and FDTD simulations (b). Original figures by
Diedenhofen et al [39]. Reprinted (adapted) with permission from Diedenhofen
et al [39]. Copyright (2011) American Chemical Society.

2D simulations (by comparing with analytical Mie Theory), it does not tackle
the problem of an array of nanowires illuminated from above.
Fischer et al [51] have written a paper on the modelling of light propagation
in a hexagonal array of dielectric cylinders. They used Lumerical FDTD Solu-
tions [120] to calculate the optical modes of an infinite hexagonal array of finite
length cylinders, applying periodic boundary conditions in the x-y directions and
terminating the ±z directions with perfectly matched layers (PMLs are absorb-
ing boundaries that remove outgoing waves from the simulation with minimal
reflection). Although their simulation setup is similar to the one used for this
thesis, they have used constant refractive indices for the cylinders (n = 1.408)
and the medium (n = 1.34). The nanowire solar cells studied in this thesis
on the other hand are highly dispersive (i.e. frequency dependent permittivity),
light absorbing (complex permittivity) and consist of a core-shell structure.
Diedenhofen et al [39] have investigated the geometrical dependence of ab-
sorption in InP nanowires. Cylinders offer the highest absorption to volume
ratio and cones the highest absolute absorption, while base-tapered nanowires
offer a good compromise. By using FDTD one can visualise the Fabry-Perot res-
onances occurring in different positions of the nanowire according to wavelength.
This paper shows how it is possible with the FDTD method to design the geo-
metry of nanowires to maximise absorption at specific positions. There are a
couple of issues however: there is residual absorption at the red end of the spec-
trum, despite experimental data showing that the absorption coefficient is zero
here. This is due to an artefact of FDTD (the refractive index data in FDTD
must be fitted to Drude and Lorentz functions, then Fourier transformed to the
time-domain [165]). Secondly, the refractive index data used was that of zinc
blende InP, not Wurtzite InP (since there is no optical data for Wurtzite InP).
Despite this, the FDTD results agree well with experiment, as shown in the
specular reflectance plots in Figure 2.14.

50
 Deinega, Demésy et al [35, 36, 37] have performed FDTD simulations of
modulated silicon nanowires. They demonstrate an absorption of 75% of in-
cident sunlight (wavelengths 400nm to 1200nm), due to the antireflection and
light-trapping provided by the modulation. Furthermore, the efficiency remains
nearly unchanged for sunlight incidence angles from normal to 60 degrees. For
the frequency dependent refractive index of silicon, they use a model where the
dielectric polarisation depends on the time derivative of the electric field as well
as the electric field itself [34]. This results in a more accurately modelled refract-
ive index for silicon in the FDTD method, compared to traditional fits based
on Debye, Drude and Lorentz approximations. Though the Lumerical software
has its own multi-coefficient procedure for fitting the dielectric function, it is
not perfect (see section 3.2.2.4). If there were better ways to fit the dielectric
functions for the TiO2 /CdSe cell materials then ideally they would be used.
Guo et al [70] have analysed optical absorption in GaAs nanowire arrays.
They determine optimised geometric parameters (diameter, length, diameter to
lattice constant ratio) and investigated oblique incidence. However, they have
not been compared to experimental results.
There are many papers on FDTD simulations of core-shell nanowires [69,
168, 122, 176, 87, 86]. Most studies do not calculate the absorption in the shell
itself, with a couple of exceptions. The paper by Kelzenberg et al [86] does
calculate the absorption in the shell (in addition to the core) and couples it to a
charge-transport model. However, the core and shell in this case are just n and p
doped silicon respectively, unlike the nanowire cells studied in this thesis where
the core and shell are composed of different materials with different properties.
Krogstrup et al [100] have performed simulations of a radial p-i-n GaAs
nanowire. The two main resonant branches in the absorption rate (Figure 2.15)
are compared to the Mie resonances of nanowires lying horizontally on a sub-
strate [24], while the periodic modulation with wavelength is a result of Fabry-
Pérot interference in the polymer layer [100]. They produce a single nanowire
cell with an efficiency of 40%, showing how the light-concentrating properties of
nanowires lead to efficiencies beyond the Shockley-Quiesser limit (33%). Again,
the core-shell structure in this study consists of a single material homojunction
(GaAs) rather than a heterojunction of different materials.
Tsai et al [176] carried out a study which included FDTD simulations of a
core-shell nanowire where the core is silicon and the shell is P3HT (polymer),
but all they do is compare three different geometries (planar, P3HT infiltrated
silicon nanowire arrays, and P3HT coated silicon nanowire arrays). They do not
vary parameters such as the length of the nanowires, thickness of the coating or
the periodicity.
To date no studies have been published on FDTD simulations of TiO2
nanowires (but there are simulations on TiO2 nanotubes [133, 140]). No pa-
per has been published on the in-depth study of light absorption in the coating
of a core-shell nanowire either. This is motivates the research in the thesis.

51
Figure 2.15: Absorption as a function of wavelength and nanowire diameter for
2.5µm long GaAs nanowire arrays embedded in a dielectric of n = 1.67 [100].
Results obtained using FDTD simulations. Original figure by Krogstrup et al
[100], permissions granted.

2.3.2 FEM and Multiphysics Simulations of Solar Cells

The finite-element method [148], as opposed to the finite-difference time-domain
method, is another numerical technique for simulating electromagnetic systems.
The simulation region is discretised into triangular elements and Maxwell’s equa-
tions are solved in the frequency domain instead of the time-domain. The ad-
vantage of this method is that it is easier to resolve curved interfaces, and the
frequency dependent dielectric function can directly be used. However, this
method is a lot more computationally demanding. Li et al [112, 113] have
coupled optical and carrier transport calculations of solar cells in 1D, 2D and
3D using COMSOL Multiphysics. With the 3D model they were able to account
for the presence of plasmonic nanoparticles, which requires detailed 3D spatial
information as they concentrate the light (Figure 2.16). Since large sweeps of
simulations needed to be carried out for the research in this thesis, the FDTD
method was chosen over the more computationally demanding FEM method.

2.3.2.1 Multiphysics simulations

Deceglie et al [33] have coupled optical and electrical modelling to optimise
both the light absorption and charge carrier collection. They designed nano-
structured a-Si solar cells that maximised absorption in the intrinsic region and
while minimising the absorption in the doped regions. The carrier generation
rate profile was calculated using FDTD simulations. This was then input into
a finite-element device physics model for the charge transport, to calculate the

52
Figure 2.16: Absorption profiles (in a unit cell) of the plasmonic solar cell under
illumination at different wavelengths [113]. Original figure by Li et al [113],
permissions granted.

photocurrent. They showed how coupled optical-electrical modelling can help

design more efficient solar cells.
Willis et al [189] have coupled the FDTD method with molecular dynamics
(MD) and the ensemble Monte Carlo method (EMC). In this case, rather than
simulating the optical and electrical aspects separately, they are coupled in every
time step. The charge and current densities are calculated using EMC, input
into the MD and FDTD models respectively, then the resulting electromagnetic
fields are fed back into the EMC model. This method is useful for simulating
high frequency charge transport in highly conductive materials. In the case
of solar cells, this is not required because the fields generated by the photo-
excited charges are negligible compared to the solar illumination. There also
exist multiscale methods which combine FDTD with quantum mechanics in the
form of time-dependent density functional theory [27, 128]. This is useful for
determining the absorption cross-section of a dye molecule, for example, but too
fine-grained for full solar cell simulations.

2.4 Charge Transport

Jenny Nelson’s book, “The Physics of Solar Cells”, is a useful text covering the
theory of charge transport in crystalline semiconductors [137]. This book de-
scribes how to derive an analytical expression for the current of a one-dimensional
p-n junction, the most basic model in semiconductor device physics. To model
heterojunctions and nanowires, a more complicated approach is required. Stud-
ies have been carried out on the charge transport properties of nanowire solar
cells, organic solar cells and metal-oxides.

2.4.1 Charge Transport in Nanowire Geometries

Kayes et al [85] have developed an analytical 1D model for the radial p-n junction
in nanowire solar cells. They carried out calculations of the short circuit current,

53
open circuit voltage and power efficiency of nanowire solar cells vs planar solar
cells for many different cell thicknesses and electron diffusion lengths (Figure
2.17). Included in their appendix are the equations for heterojunction planar
solar cells, extending those of Nelson’s. The equations in this paper were use-
ful for testing and learning about the drift-diffusion model (via Mathematica).
Unfortunately this paper does not treat radial p-i-n junctions, which will be
needed to model the nanowire geometry of the TiO2 /CdSe solar cell. Kayes
et al’s paper deals with inorganic semiconductors and as such does not treat
excitons, since they are weakly bound and dissociate almost immediately.
Whether excitons would be important to model depends on the exciton bind-
ing energy. In CdSe, excitons immediately dissociate due to having a low exciton
binding energy of 15meV [130], making it unnecessary to model exciton trans-
port. Using this paper as a starting point, one may be able to develop an
analytical model for radial p-i-n junctions. Failing that, numerical simulations
of the charge-carrier transport are required.

2.4.2 Numerical Methods for Charge Transport

To calculate the charge transport properties of a solar cell numerically, the drift-
diffusion equations are solved [158] (see section 4.2.2). Deinega and John [35]
have implemented a finite-difference discretisation of the drift-diffusion equa-
tions for nanowire solar cells. They compare two schemes: 2D(r, ) cylindrical
discretisation and 3D(x, y, z) cartesian discretisation. In the 2D cylindrical
scheme, the variables for potential (r, ), electron density n(r, ) and hole
density p(r, ) are expanded into a sum of trigonometric Fourier series (partial
waves). The charge carrier generation rate (obtained from FDTD simulations)
is averaged over the azimuthal angle (which is shown to to be a good approx-
imation despite the slight anisotropy). The drift-diffusion equations are then
solved in cylindrical coordinates. They used a higher mesh resolution near the
p-n junction due to the fact that , n and p vary rapidly there. The results
were compared against the 3D cartesian scheme, showing that the Jsc and Voc
results in the 2D cylindrical scheme converged more rapidly as a function of
mesh resolution.
The method used for modelling charge transport for the TiO2 /CdSe solar cell
in this thesis follows that of Deinega and John. Ideally the 2D cylindrical scheme
would be used, as it is fast, accurate and seems well suited to the problem.
However, due to the large bandgaps and rapid changes in electronic properties
over a short distance, the 2D cylindrical scheme is too numerically unstable, as
explained in chapter 4. Only 1D charge transport simulations were successfully
carried out in the end.
As mentioned earlier (section 2.3.2), Li et al [113] developed a multidi-
mensional model for optical and carrier transport simulations of solar cells
using COMSOL. However, their model uses finite-elements instead of finite-
differences. Although their model is useful for simulating arbitrary 3D solar
cells, it is very computationally intensive. Furthermore, their model requires
the usage of closed-source software, making it difficult to customise the charge

54
 114302-8 Kayes, Atwater, and Lewis J. Appl. Ph

FIG. 9. !Color online" Short-circuit current density

and minority-electron diffusion length Ln for !a" a
Figure 2.17:FIG.Efficiency vs cellEfficiency
8. !Color online" thickness and
vs cell electron
thickness diffusion
L and length forjunction
minority-electron planargallium arsenide cell and !b" a radial p-n ju
Si solar cells (a) and
diffusion nanorod
length Ln for !a"solar cells (b).
a conventional Original
planar figure
p-n junction siliconbycellKayes
and et al [85]cell. In both cases the short-circuit current
arsenide
!b" decreasing the trap density in the depletion region. I
, permissions granted.
a radial p-n junction nanorod silicon cell. In both cases the top surface
nanorod case, the cell radius R is set equal to Ln, a c
shown in the plot has a depletion-region trap density fixed at 1014 cm−3, so
that # , # = 1 "s, while the bottom surface has a depletion-region trap to be near optimal.
n0 p0
density equal to the trap density in the quasineutral region, at each value of
Ln. In the radial p-n junction nanorod case, the cell radius R is set equal to
Ln, a condition that was found to be near optimal. For a silicon solar cell with Ln = 100
! 1018 cm−3" in the quasineutral regions
the open-circuit voltage dropped from 0.59 to 0.24 V over in the depletion region, the maximal effic
the same range, or from 0.59 to 0.49 V if Nr in the depletion p-n junction nanorod geometry was 11
region was held fixed. 1.5% in the planar geometry. This max
curred for a radial p-n junction nanorod c
55
500 "m thick, whereas the efficiency sat
C. Cell efficiency, silicon
cell thicker than 450 nm. If Nr was se
Taken together, the factors discussed above imply that throughout the cell, the maximal efficien
the efficiency of a radial p-n junction nanorod solar cell can -n junction nanorod geometry was 1%, co
remain high despite a high quasineutral-region trap density, in the planar geometry. In this case, the
provided that the depletion-region trap density remains rela- occurred for a radial p-n junction nanorod
tively low !Fig. 8". In the planar geometry, a high 30 "m thick, whereas the efficiency sat
quasineutral-region trap density leads to a very low short- cell thicker than 1 "m.
Figure 2.18: Meshes used in the 2D(r, ) cylindrical discretisation (top) and
3D(x, y, z) cartesian discretisation (bottom) for the nanowire drift-diffusion
model by Deinega and John [35]. Original figure by Deinega et al [35] , permis-
sions granted.

transport method to handle numerically unstable problems.

2.4.3 Charge Transport in Polymers, Hybrids and Metal-

Oxides
Kannan et al [83] were the first to carry out a systematic theoretical ana-
lysis of hybrid nanowire solar cells. They examine two configurations: polymer
nanowires in an inorganic matrix (which is analytically solvable) and inorganic
nanowires in a polymer matrix (which must be solved numerically). In either
case, excitons are generated in the polymer, diffuse to the inorganic/polymer in-
terface and dissociate. The electron hops onto the delocalised conduction band
of the inorganic phase and the hole is left behind in the polymer. This model
does not reflect the charge transport of the TiO2 /CdSe solar cell in this thesis.
This is because in the this solar cell, the exciton is generated in the CdSe and
almost immediately dissociates (due to the the low exciton binding energy of
15meV [130]) leaving an electron to hop onto the conduction band of the TiO2
and the hole to hop onto the HOMO of the polymer.

2.4.3.1 Charge Transport in TiO2

Deskins and Dupuis [38] have carried out a DFT study on the electron transport
in bulk TiO2 . In this paper it is assumed that conducting electrons in TiO2 are

56
Figure 2.19: Diagram representing the transfer of a polaron from initial state
A to final state B [38]. The corresponding equilibrium geometries are qA
and qB , and the transition state is at qC (see Figure 2.20). G⇤ is the diabatic
activation energy, is the reorganisation energy (the energy of the final state B
at the geometry qA ) and VAB the electronic coupling matrix element. Original
figure by Deskins and Dupuis [38], permissions granted.

described by a small polaron hopping model [43], and that experimental and
theoretical work add to the validity of this model. They use Marcus Theory [125]
to calculate the mobility of the electrons, using DFT to calculate the parameters
in Marcus Theory (see Figures 2.19 and 2.20).

2.4.3.2 Charge Transport and Optical Properties of Polymers

The electronic and optical properties of polymeric semiconductors arise mainly
from delocalised ⇡ orbitals [71]. The length that these delocalised bonds extend
is the conjugation length. Real polymer chains have defects which break up the
conjugation length, resulting in a distribution of delocalised segments and hence
band gaps. The literature seems to show that polymers have a rich range of
electronic and optical properties depending on the molecular structure, disorder,
doping and preparation methods [71]. Since in the ETA cell the absorption will
mainly occur in the thin CdSe coating, one may not need to worry too much
about optical absorption and excitonic transport in the polymer. Hole transport
is more important.

57
Figure 2.20: Diagram representing the transfer of an electron (via polaron) [38].
The electron is localised on the left Ti ion (qA geometry), shared between the
two ions (qC geometry) then localised on the right ion (qB geometry). Original
figure by Deskins and Dupuis [38], permissions granted.

58
Chapter 3

Optical Methods

3.1 Theory
3.1.1 Differential Equation Methods
A review on the subject of numerical methods (some analytical) in electromag-
netic scattering theory has been written by Kahnert [81]. These are broken
down into two categories: differential equation and integral equation methods.
All of the differential equation methods are based on either solving Maxwell’s
curl equations (3.1) subject to an initial condition or by solving the vector
Helmholtz equations (3.2) subject to boundary conditions,

✏(r) @E(r, t)
r ⇥ H(r, t) = , (3.1)
c0 @t
µ(r) @H(r, t)
r ⇥ E(r, t) =
c0 @t

[r2 + k 2 (r)]H(r) = 0, (3.2)

2 2
[r + k (r)]E(r) = 0
where H and E are the magnetic and electric fields, k is the wavenumber inside
the medium, ✏ is the dielectric permitivitty and µ is the magnetic permeability.
Equation 3.2 follows by taking the curl of equation 3.1 and assuming harmonic
fields (i.e, E(r, t) = E(r)e i!t where ! = k/c is the angular frequency).
Such methods include the separation of variables method (SVM), the finite-
difference time-domain (FDTD) method and the finite-element method (FEM).
Other methods are mentioned in Kahnert’s review [81].

3.1.1.1 Separation of Variables Method

The SVM consists of solving the scalar Helmholtz equation [r2r + k 2 ] (r) = 0
in spherical coordinates for each component (r) of the electric and magnetic

59
fields [81]. One assumes that (r) is separable such that
(r, ✓, ) = R(r)⇥(✓) ( ) (3.3)
where ✓ is the polar angle and is the azimuthal angle. Substituting this into
the scalar Helmholtz equation, one obtains the set of solutions
j
n,m (r, ✓, ) = znj (r)Pnm (✓)eim

where znj (r) are spherical Bessel functions (j = 1, 2 refer to Bessel functions
of the first and second kind, j = 3, 4 refer to Hankel functions of the first and
second kind) and Pnm (✓) are Legendre functions (m = n, . . . , n. n2 N). From
the solutions one can construct spherical vector wave functions M and N [81]
where
M = r ⇥ (r )
1
N = r⇥M
k
These vector wave functions satisfy the Helmholtz equation (3.2). The incident,
scattered and internal fields are then expanded in terms of these wave function
functions such that,
1 X
X n
⇥ ⇤
E inc (k0 r) = aM 1 N 1
n,m Mn,m (k0 r) + an,m Nn,m (k0 r) ,
n=0 m= n
X1 X n
⇥ ⇤
E sca (k0 r) = pM 3 N 3
n,m Mn,m (k0 r) + pn,m Nn,m (k0 r) ,
n=0 m= n
X1 X n
⇥ ⇤
E int (k0 r) = cM 1 N 1
n,m Mn,m (k0 r) + cn,m Nn,m (k0 r) ,
n=0 m= n

where k0 is the wavenumber outside the scattering medium [81] and an,m , pn,m
and cn,m are expansion coefficients. The infinite expansions are truncated to a
finite number of terms depending on the computational resources and accuracy
required. The tangential components of E and H must be continuous across
the surface of the scatterer. By applying this boundary condition, one obtains
a set of linear equations relating the expansion coefficients pn,m and cn,m to
the known coefficients of the incident field an,m . After solving these equations,
one can determine the scattered and internal fields E sca (k0 r) and E int (k0 r),
allowing other quantities to be calculated such as the scattering and absorption
cross-sections. In the case of spherical particles the system of equations can be
solved analytically, leading to the widely used Mie Theory solutions [88, 178].

3.1.2 Mie Theory

To understand how light interacts with a sub-wavelength sized particle, one must
calculate the following quantities: scattering cross-section sca , absorption cross-
section abs , and the extinction cross-section ext . The scattering cross-section

60
(i.e. the effective area of collision) is a measure of how much light is refracted
or reflected by a particle, which usually increases with the particle size. The
absorption cross-section abs quantifies how much light is absorbed by a particle.
The extinction cross-section ext is a sum of the scattering and absorption cross-
sections. The scattering and extinction efficiencies, Qsca and Qext , where the
efficiency Q is divided by cross-sectional area, can be calculated using Mie
Scattering Theory.
Mie Scattering Theory is a rigorous treatment of how light scatters from
arbitrarily sized spherical/cylindrical particles. The reason Mie theory is used
is because the nanowires studied in this thesis (being ⇠100nm wide) are too
large to treat with Rayleigh theory and too small to treat using geometrical
optics [127]. The Mie solution for an infinitely long homogeneous cylinder is
derived in books by van de Hulst [178] and Bohren [16]. For a cylinder, the
formulae derived are:
1
2 X
Qext = Re(an )
x n= 1
1
2 X
Qsca = |an |2
x n= 1

where an is the scattering coefficient, x = ka is the size parameter, k is wavenum-

ber and a is the radius of the cylinder. The absorption efficiency can then be
found using Qabs = Qext Qsca . The derivation of the coefficient an is detailed
in the next section. For this thesis, FDTD simulations have been benchmarked
against the Mie solutions by carrying out two-dimensional simulations of light
incident on a cylindrical particle (see section 5.1.1).

3.1.2.1 Homogeneous Cylinder

To derive the scattering coefficient an , one starts by assuming light is perpendic-
ularly incident on an infinitely long homogenous cylinder in a transverse electric
polarisation (Figure 3.1). Then Maxwell’s equations are solved in cylindrical co-
ordinates, applying the appropriate boundary conditions.
It is assumed that the electromagnetic fields are plane waves such that
E(r, t) = E0 ei(!t kx) . By substituting this and a similar expression for H
into Maxwell’s curl equations (3.1) one obtains

r⇥E = ikH
r⇥H = ikm2 E

where m = ck/! is the refractive index of the cylinder. The cartesian compo-
nents of E and H are expanded in terms of cylindrical waves Zn (kr)ein✓ inside
(r < a) and outside (r > a) the cylinder [178] such that

61
Figure 3.1: Infinitely long cylinder with a radius of a, used in the Mie theory
analysis. k represents the wave-vector of the incoming transverse-electric light,
E and H the electric and magnetic fields.

1
X
= ei!t ( i)n Zn (kr)ein✓
n= 1
(
Jn (kr) r<a
Zn (kr) = (1)
Hn (kr) ⌘ Jn (kr) + iYn (kr) r>a
where represents the cartesian field components, Jn (kr) is the Bessel function
(1)
of the 1st kind, Yn (kr) is the Bessel function of the 2nd kind, and Hn (kr) is the
Hankel function of the 1 kind (note that in Van de Hulst’s treatment of Mie
st
(2)
theory [178], Hn (kr) is used because they define the complex refractive index
as m = n i instead of m = n + i). The reasons these functions are chosen
(1)
are that Jn (kr) is finite at r = 0 and Hn (kr) is finite as r ! 1. Solving for
incident and scattered fields together [178],

(P
1
1 Fn [Jn (kr) an Hn (kr)] r<a
= Pn=
1
n= 1 Fn cn Jn (mkr) r>a
i(n✓+!t)
Fn = e ( 1)n

The coefficient an is the term one aims to derive, since this determines the
scattering and absorption cross-sections. By applying the boundary conditions
that the normal components of the electric and magnetic fields are continuous
across the cylinder boundary,

n ⇥ (Hr>a Hr<a ) = 0
n ⇥ (Er>a Er<a ) = 0

62
one can solve for an ,

Jn0 (mx)Jn (x) mJ n (mx)Jn0 (x)

an =
Jn0 (mx)Hn (x) mJ n (mx)Hn0 (x)

where x = ka.
The Mie solution for a coated cylinder (two coaxial cylinders with different
refractive indices) with perpendicularly incident light was first derived by Kerker
[89]. Using the same procedure as for the single cylinder case, one can solve the
scattering coefficient for a coated cylinder,

m0 Jn (m0 x1 ) m1 Hn (m1 x1 ) m1 Jn (m1 x1 ) 0

Jn0 (m0 x1 ) Hn0 (m1 x1 ) Jn0 (m1 x1 ) 0
0 m1 Hn (m1 x2 ) m1 Jn (m1 x2 ) m2 Jn (m2 x2 )
0 Hn0 (m1 x2 ) Jn0 (m1 x2 ) Jn0 (m2 x2 )
an =
m0 Hn (m0 x1 ) m1 Hn (m1 x1 ) m1 Jn (m1 x1 ) 0
Hn0 (m0 x1 ) Hn0 (m1 x1 ) Jn0 (m1 x1 ) 0
0 m1 Hn (m1 x2 ) m1 Jn (m1 x2 ) m2 Jn (m2 x2 )
0 Hn0 (m1 x2 ) Jn0 (m1 x2 ) Jn0 (m2 x2 )

where x1 = kb, x2 = ka, a and m2 are the radius and refractive index of the
inner cylinder, b and m1 are the radius and refractive index of the outer cylinder
respectively, and m0 the refractive index of the medium. The Mie solution for
a coated cylinder was extended to the case of oblique incidence by Shah [159]
and Yousif [196]. These solutions could be used to verify FDTD simulation
of obliquely incident light. Such simulations are however complicated due to
the dispersion of injection angles in broadband simulations (details in section
3.2.2.6).
The theoretical results described above are valid for an infinitely long nanowire
illuminated from the side (perpendicular to nanowire axis). However, modelling
nanowire array solar cells requires understanding the scattering properties of an
array of finite length nanowires, immersed in an absorbing medium illuminated
from above (parallel to the nanowire axis). To begin to tackle this problem one
must extend Mie Theory with multiple scattering and finite length effects. As
detailed below, this is quite difficult.

3.1.2.2 Finite Length Cylinders

For the case of finite length cylinders, Wang and Hulst [185] derived a Mie the-
ory approximation using Huygen’s principle. Note that this is an approximation
and works best for long cylinders illuminated close to the normal (not illumin-
ated from above). Seker and Schneider [157] calculated the far-field scattering
amplitude for a finite-length cylinder, by assuming the internal fields are the
same as those computed from the Mie solution of an infinite cylinder. Their
formulation must be applied with caution for near end-on incidence, and only
works for lengths much larger than the radius and greater than or comparable
to the wavelength.

63
 Rother et al [153] have derived formulae for cylinders with non-circular cross-
sections. Such work is useful for calculating the scattering behaviour of ice crys-
tals, by modelling each crystal as a hexagonal column. However, the nanowires
in the solar cell under study are roughly circular in shape. Furthermore, Rother
et al only consider the far-field scattering, and do not derive the internal field
which is required for calculating the absorption profile in the nanowire coating.
Their paper has, however, offered a useful review on the subject of scattering of
finite cylinders.
Lin and Sarabandi [115] wrote a paper describing a model for the electro-
magnetic scattering of a tree trunk on a tilted plane. Though radar applications
were primarily in mind, one could use this to model the scattering of light of a
nanowire on a substrate. Again, infinitely long cylinders are used to approximate
the internal fields of finite cylinders, and only the far-field scattering properties
can be calculated reliably, but the model does take into account the effect of
the bark layer of the tree (which is analogous to the coating on the nanowire).

3.1.2.3 Multiple Finite Cylinders

Many papers on the subject of scattering of multiple finite cylinders exist but
most use numerical methods [59, 180]. A couple of papers on the radar scat-
tering of multiple tree trunks over a rough interface [80, 84] would at first sight
look useful to model nanowires over a substrate. Such papers model the internal
fields of tree trunks as infinite cylinders and use Huygen’s principle and statist-
ical methods (for the multiple scattering) to extrapolate the far-field scattering
properties.

3.1.2.4 Multiple Cylinders in an Absorbing Medium

The only studies which take into account multiple scattering of oriented coated
cylinders in an absorbing medium, and calculate the internal fields, are those
by Siu-Chun Lee. One of the papers is about a single coated infinite cylinder
in an absorbing medium [107]. This can be used to verify FDTD simulations
of a single coated infinite nanowire embedded in a polymer. Two of the papers
are on multiple scattering of multi-layered cylinders in an absorbing medium
[107, 108].
Although this helps when testing whether FDTD simulations are set up
correctly, it will not help in the case when the light is incident from above the
nanowires. In that case the area the light is initially incident on is flat, so
there is no exact analytical method to compute the scattering. Fully numerical
methods (i.e, FDTD) will need to be depended on.

3.1.3 Transfer-Matrix Method

In order to compare nanowire arrays to planar heterojunctions, the reflectance
R, transmittance T and absorbance A = 1 (R + T ) were calculated for planar
slabs of TiO2 , CdSe, and CdSe coated TiO2 using the Transfer-Matrix Method.

64
By considering what happens to a normally incident wave Ei exp[i(k0 z !t)]
where k0 = m0 !/c and m0 is the refractive index of the non-absorbing medium,
one can calculate the reflection and transmission coefficients [16],
r̃[1 exp(i2km1 h)]
r̃slab =
1 r̃2 exp(i2km1 h)
4⌘ exp( i2km0 h)
t̃slab = ,
(⌘ + 1)2 [exp( ikm1 h) r̃2 exp(ikm1 h)]
where ⌘ = m m0 is the relative refractive index of the slab compared to the
1

medium, h is the slab thickness and r̃ = 11+⌘⌘ . The reflectance and and trans-
mittance are defined as,

R = |r̃slab |2 , T = |t̃slab |2

One can follow a similar procedure for a coated slab. If layer j has a thickness hj ,
refractive index mj and wavenumber kj = mj k, the transmission and reflection
coefficients are,
(1 ⌘1 )(1 + ⌘12 )(1 + ⌘2 ) exp( ik2 h2 ik1 h1 )
+ (1 ⌘1 )(1 ⌘12 )(1 ⌘2 ) exp(+ik2 h2 ik1 h1 )
(1 + ⌘1 )(1 ⌘12 )(1 + ⌘2 ) exp( ik2 h2 + ik1 h1 )
(1 + ⌘1 )(1 + ⌘12 )(1 ⌘2 ) exp(+ik2 h2 + ik1 h1 )
r̃slab = ,
(1 + ⌘1 )(1 + ⌘12 )(1 + ⌘2 ) exp( ik2 h2 ik1 h1 )
+ (1 + ⌘1 )(1 ⌘12 )(1 ⌘2 ) exp(+ik2 h2 ik1 h1 )
(1 ⌘1 )(1 ⌘12 )(1 + ⌘2 ) exp( ik2 h2 + ik1 h1 )
(1 ⌘1 )(1 + ⌘12 )(1 ⌘2 ) exp(+ik2 h2 + ik1 h1 )

8⌘12 ⌘2 exp[ ik0 (h1 + h2 )]

t̃slab = ,
(1 + ⌘1 )(1 + ⌘12 )(1 + ⌘2 ) exp( ik2 h2 ik1 h1 )
+ (1 + ⌘1 )(1 ⌘12 )(1 ⌘2 ) exp(+ik2 h2 ik1 h1 )
(1 ⌘1 )(1 ⌘12 )(1 + ⌘2 ) exp( ik2 h2 + ik1 h1 )
(1 ⌘1 )(1 + ⌘12 )(1 ⌘2 ) exp(+ik2 h2 + ik1 h1 )
where the relative refractive indices are ⌘1 = m0 ,
m1
⌘2 = m2
m0 and ⌘12 = m2 .
m1

3.2 Numerical Methods

3.2.1 The Finite-Difference Time-Domain Method
To simulate light scattering off a cylinder, Maxwell’s equations were solved
numerically using the Finite-Difference Time-Domain (FDTD) method [165].
Maxwell’s curl equations for a lossy material are
@H 1 1
= r⇥E (Msource + ⇤ H)
@t µ µ
@E 1 1
= r⇥H (Jsource + E)
@t " "

65
where Jsource is an external current density and Msource is an external magnetic
current density [165]. There are therefore six coupled partial differential equa-
tions describing the electromagnetics, one for each component of E and H. For
example, the equation for Ex is

@Ex 1 @Hz @Hy
= (Jsource,z Ex ) (3.4)
@t ✏ @y @z
and similar equations describe the other components.

3.2.1.1 Yee Implementation

Under the Yee implementation of FDTD the Maxwell equations are discretised
in space and time using central differences. This is achieved by partitioning
the simulation region into a spatial mesh of Yee Cells, as shown in Figure 3.2.
The field components in each cell are staggered such that every E component is
encircled by four H components and vice versa. The interweaving of the electric
and magnetic field components and the central difference approximation allows
the finite-difference equations to be solved using a leapfrog algorithm (see Figure
3.3).
Applying central differences to equation 3.4 and re-arranging, one gets
0 1 0 1
i,j+1/2,k+1/2 t t
n+1/2
1 2"i,j+1/2,k+1/2 "i,j+1/2,k+1/2
Ex |i,j+1/2,k+1/2 = @ A Ex |n 1/2 @ A
i,j+1/2,k+1/2 t i,j+1/2,k+1/2 + i,j+1/2,k+1/2 t
1 + 2"i,j+1/2,k+1/2 1 + 2" i,j+1/2,k+1/2
n n n n
Hz |i,j+1,k+1/2 Hz |i,j,k+1/2 Hy |i,j+1/2,k+1 Hy |i,j+1/2,k
·
y z
⌘
n
Jsource,x |i,j+1/2,k+1/2

and similar expressions are derived for the other field components [165].
If there is a boundary between two or more materials in the same cell, the
dielectric function ✏ is sampled according to the meshing algorithm (see section
3.2.2.4).

3.2.2 Lumerical FDTD Solutions

Lumerical FDTD Solutions [120] is a commercial TCAD software package for
designing optoelectronic devices and simulating them using the FDTD method.
The two main features that set Lumerical apart from other FDTD software
are its conformal meshing technology and multi-coefficient material modelling
(MCM). The MCM is described in section 3.2.2.4.
While the standard Yee-Cell implementation of FDTD requires discretising
the system into a cartesian mesh, it is unable to take into account variations of
structure within each cell. This results in “staircasing”, where the permittivity
takes a single value in each cell depending on which material takes up most of
the cell. A more accurate representation of the structure can be acquired by

66
Figure 3.2: A diagram of the “Yee Cell” [30] demonstrating the interweaving
of the electric and magnetic fields. The mesh points are labeled (i,j,k) for the
x, y and z directions respectively. The curl of the magnetic field through the
faces gives the time derivatives of the electric fields at the edges and vice versa.
Original figure from Wikimedia commons [30].

Figure 3.3: Leapfrog algorithm [165]. At each half time step, either the electric
fields or magnetic fields are calculated, using values of the fields at previous time
steps and at half step positions. Original figure by Taflove [165], permissions
granted.

67
accounting for these subcell features, for example by using the Maxwell-Garnett
rule [57],
ˆ x+ x ˆ y+ y ˆ z+ z
✏(i, j, k) 1 1 ✏(x0 , y 0 , z 0 ) 1
= dx0 dy 0 dz 0
✏(i, j, k) + 2 x y z x y z ✏(x0 , y 0 , z 0 ) + 2

Lumerical has a built-in conformal meshing technique (CMT) which takes

into account dispersive materials, and is based on work by Yu and Mittra [197].
Their CMT is able to account for subcell features much more accurately with
a significantly coarser mesh than with a staircasing meshing algorithm. Since
computation time in FDTD varies as 1/dx4 (in 3D), it enables much faster and
more accurate simulations. More details of the meshing algorithm are in section
3.2.2.4.

3.2.2.1 Simulation Setup

To validate the FDTD method implemented in Lumerical, tests were performed
in two dimensions. One can compare the simulation of light illuminating the side
of a coated nanowire with the Mie solution for a coated cylinder [89]. Figure 3.4a
shows the simulation model of the TiO2 /CdSe nanowire for this test. Since the
nanowire was modelled as infinitely long in the z-direction with perpendicular
illumination, only a 2D simulation was required. The light source was polarised
in the transverse electric (TE) configuration such that E oscillates in plane
(y-axis) and H oscillates out of the plane, along the nanowire axis (z-axis).
The total field monitor (inner yellow box) was used to calculate Qabs , the Mie
absorption efficiency, whereas the scattered-field monitor (outer yellow box)
was used to calculate Qsca . The Lumerical total-field scattered-field (TFSF)
box is used. Outside the TFSF region the incident field is subtracted out,
allowing monitors to measure the scattered field. Scripts to calculate the cross-
sections this way were downloaded from the Lumerical website [120] and edited.
Lumerical and Bash scripts were written to carry out sweeps of simulations
with different parameters. While the simulation runs, a plane wave source is
injected, then after the fields have decayed to 10 5 of their original magnitude
the simulation terminates.
To model the illumination of an array of finite-length cylindrical nanowires,
the FDTD simulations require three dimensions. These type of problems have
no analytical solution. Figure 3.4b shows the simulation setup for a periodic
array of nanowires.

3.2.2.2 Illumination and Fourier Transform

Although FDTD simulations were carried out in the time-domain, the results
were Fourier-transformed to obtain quantities for each frequency. A broadband
source of light predominantly in the 400nm to 800nm region (near ultra-violet
to infra-red) was used while 100 frequency points in this range were monitored
(see Figure 3.5). The simulation time must be long enough for the fields to
decay, otherwise the frequency (wavelength) domain calculations will be wrong.

68
 (a) (b)

Figure 3.4: (a) 2D Simulation model of the TiO2 /CdSe nanowire. The grey
circle in the centre represents the TiO2 core. The blue ring represents the CdSe
coating. The inner yellow box represents the ‘total field monitor’. The outer
yellow box is the ‘scattered field monitor’. The pink arrow represents the x-
direction of the ‘total-field scattered-field’ (TFSF) plane wave source, the blue
arrow the oscillation of the electric field in the y-axis. The grey box attached to
the source represents the cutoff for the incident field, leaving only the scattered
field. Absorbing boundary conditions are enforced outside the simulation region.
(b) 3D simulation model of the nanowire with top-down illumination. Periodic
boundary conditions are enforced in the x and y directions. Absorbing boundary
conditions are enforced in the ±z directions.

69
Cutting off a simulation too early while fields are still oscillating with significant
magnitude will cause ‘ripples’ to appear in the spectra. To save computational
time while getting accurate results, an ‘auto-shutoff’ feature was enabled so
that when the fields decayed to 10 5 of their original magnitude, the simulation
would terminate.
By using a special type of plane wave source, the total-field scattered-field
(TFSF) source, the incident field can be subtracted outside the grey box in
Figure 3.4a, leaving only the scattered field. This is to ensure the outer yellow
box only measures the scattered field. Figure 3.6 shows a few snapshots in time
of the FDTD simulation. In Figure 3.6a, light has just reached the nanowire.
Figure 3.6b shows the fields penetrating the nanowire. In Figure 3.6c, one can
see how the outline of the box used in the TFSF source where the incident field
gets subtracted off. Figure 3.6d shows the fields oscillating inside and radiating
from the nanowire after the illumination has passed.

3.2.2.3 Absorption Calculation

Calculating the absorption in a material using a total field monitor does not work
in an absorbing medium, or if there is more than one material. An alternative
method to calculate the absorption is needed. The power absorbed per unit
volume in a material is given by,
1
Pabs = Re(r · S)
2
where S = E ⇥ H is the Poynting vector. Divergence calculations are susceptible
to numerical errors. However, this equation can be reformulated into one that
is more numerically stable [120],
1
Pabs = !|E|2 Im(") (3.5)
2
where ! is the angular frequency and " is the dielectric permittivity. The spatial
profile of power absorbed per unit volume can be obtained using these methods.

3.2.2.4 Meshing and Dielectric Function

A fine mesh is required for sudden changes in refractive index or for highly
curved interfaces. Lumerical has an automatic meshing feature which makes
the mesh finer as one approaches boundaries between two different dielectric
media. There are many different ways to sample the dielectric function in a
cell when there is a boundary between two materials. The simplest way is the
‘stair-casing method’, where a cell will sample the dielectric function of the ma-
terial that takes up the most volume in a cell. Another method is volumetric
averaging, where the dielectric function in a cell is the volume average of the
dielectric functions of the two materials in that cell. Lumerical uses confor-
mal meshing technology. Though the exact meshing algorithm Lumerical uses
is closed-source, in general conformal meshing algorithms account for sub-cell

70
 1

0.9

0.8

0.7

0.6
|E|2

0.5

0.4

0.3

0.2

0.1

0
0 2 4 6 8 10 12 14
Time (fs)

(a) Linear plot, time domain (truncated to 15 femtoseconds)

10000

100

0.01
|E|2

0.0001

1e-06

1e-08

1e-10

1e-12
100 1000
Wavelength (nm)

(b) Log-log plot, wavelength domain (truncated to the dominant 100nm to

1000nm region)

Figure 3.5: Electric field intensity of the place-wave light source used in FDTD
simulations.

71
 (a) 5.00fs (b) 6.66fs

(c) 9.17fs (d) 12.08fs

Figure 3.6: Snapshots in time (femtoseconds) of the 2D FDTD simulations of

light scattering off the TiO2 /CdSe nanowire. The illumination is from the side
(perpendicular to nanowire axis). Blue represents zero intensity, red is maximum
intensity. The two concentric circles in the centre represent the coated nanowire.

72
 (a) (b)

Figure 3.7: Conformal Meshing techniques for FDTD modelling [120], developed
by Yu and Mittra [197], on which Lumerical’s conformal meshing technology is
based. (a) This model provides greater accuracy for resolving PEC (perfect
electrical conductor) / dielectric interfaces. The contour integral for Maxwell’s
integral equations is reduced from the contour C to the contour C1 such that the
PEC with the vanishing electric field is not included. (b) This model takes into
account interfaces between two dielectrics. The Yee Cell’s effective permittivity
in the x and y directions are weighted by the fraction of the mesh that is inside
each material in those directions. Original figures by Lumerical Solutions Inc.
[120], permissions granted.

variations by solving Maxwell’s integral equations near structure boundaries

[120]. It is based on the models developed by Yu and Mittra [197], described in
Figure 3.7 but updated to take into account arbitrary dispersive dielectrics.
To use the FDTD method, the electromagnetic fields are Fourier transformed
from the frequency domain to the time domain, solved in the time-domain,
then inverse-Fourier transformed back to get solutions in terms of frequency.
The same process is carried out for the complex refractive index (or dielectric
function) of each material. To ensure this process is not too computationally
intensive, the refractive index must be expressible in terms of functions that
are easily Fourier transformed. Traditionally, this is achieved by fitting the
experimental data of the refractive indices to idealised Drude, Lorentz and Debye
models. In Lumerical the refractive index is fitted to experimental data using
multi-coefficient models (MCM’s). These have a more extensive set of basis
functions to take into account the highly dispersive nature of such materials.
Figure 3.15 shows the MCM refractive index of CdSe fitted to experimental
data from Palik [144]. The polynomial expressions for the real and imaginary

73
parts of the refractive index are manually parametrised in Lumerical by varying
the number of coefficients, fit tolerance and the imaginary weight (how much
consideration to give to the fitting of the imaginary part) until satisfactory fits
are obtained. Care is taken not to use too many coefficients (for computational
speed). The degree to which the MCM fits the real or imaginary parts could
be controlled, though fitting well to the real part could worsen the fit to the
imaginary part, and vice versa. For calculating the absorption, it is important
to attach more weight to the imaginary part, since this is directly proportional
to how much light is absorbed.

Mesh Resolution To work out what mesh resolution was required, 3D FDTD
simulations were performed on a periodic array of nanowires for three different
meshes. Each mesh used is a customised non-uniform mesh which is graded
so that the resolution is proportional to the refractive index. Figure 3.8 shows
the Lumerical setup, and Figure 3.9 shows the absorption spectra for the three
meshes. The results converge at the finest mesh. This is defined by setting
the minimum mesh step to 3nm, maximum mesh step to 10nm in the x and y
directions, and 18 mesh cells per wavelength in the z direction.

3.2.2.5 PML Boundary Conditions

To simulate the scattering of light with a particle in an unbounded region in
FDTD, special boundary conditions are applied to the edges of the simulation
region [165]. Ideally, these boundaries would absorb electromagnetic waves with
zero reflection for all frequencies and angles of incidence. A matched layer im-
plementation is often used [14], where the simulation region is surrounded with
an absorbing layers which match the impedance of the medium. However, in
the past these implementations of matching layers were fraught with problems:
electromagnetic waves only had zero reflection when specific conditions were
met (i.e, planar waves normally incident on the boundary) [14]. Berenger de-
veloped a more robust matched layer boundary condition called the Perfectly
Matched Layer (PML) [14]. In this case, theoretically the reflection is zero
for any frequency or angle of incidence. To derive the PML medium, one first
writes the Maxwell equations for a medium with an electric conductivity of
and magnetic conductivity of ⇤ ,
@Ex @Hz
✏0 + Ex =
@t @y
@Ey @Hz
✏0 + Ey =
@t @x
@Hz @E x @Ey
µ0 + ⇤ Hz =
@t @y @x
For no reflection to occur (with normal incidence), the impedance of the PML
medium and vacuum must be matched, such that

74
 (a)

(b)

Figure 3.8: Lumerical simulation setup for a coated nanowire with core radius
a = 100nm and coating thickness h = 20nm, perspectives from the top (a)
and from the side (b). The blue and green translucent regions represent sym-
metric and anti-symmetric boundaries respectively (reducing the computational
requirement by a factor of 4). Three different meshes are used. For all cases,
the minimum mesh step was set to 3nm. On the left, the maximum mesh steps
in the x and y directions were xmax = 20nm and ymax = 20nm, and for
the z direction the number of mesh cells per wavelength was set to nz = 9.
In the middle, xmax = 15nm, ymax = 15nm and nz = 15. On the right,
xmax = 10nm, ymax = 10nm and nz = 18.

75
 �
� � ����� �� � ����� �� � ����� �� � �
� � ����� �� � ����� �� � ����� �� � ��
� � ����� �� � ����� �� � ����� �� � ��
���
����������

���

���

���

�
��� ��� ��� ��� ��� ��� ��� ��� ���
���������� ����

Figure 3.9: Absorption spectra obtained using three different meshes for a peri-
odic array of nanowires with a = 100nm and h = 20nm. The solid line shows
the results for a mesh with xmax = 20nm, ymax = 20nm and nz = 9 (coarsest
mesh). Dashed line: xmax = 15nm, ymax = 15nm and nz = 15. Dotted line:
xmax = 10nm, ymax = 10nm and nz = 18 (finest mesh). The results have
converged for the finest mesh.

76
 ⇤
=
✏0 µ0
The electromagnetic field components are then split into orthogonal subcom-
ponents. This splitting of the components is the basis for Berenger’s original
PML method, also known as split-PML. For example, in the transverse electric
(TE) case where the electric field is confined to the xy-plane and the magnetic
field lies out of the plane, Hz is split into two components Hzx + Hzy , such that

@Ex @(Hzx + Hzy )

✏0 + y Ex =
@t @y
@Ey @(Hzx + Hzy )
✏0 + x Ey =
@t @x (3.6)
@Hzx @Ey
µ0 + x⇤ Hzx =
@t @x
@Hzy @Ex
µ0 + y⇤ Hzx =
@t @y
To prevent reflections at interfaces between media, the x and y components of
the electric and magnetic conductivities, ( x , x⇤ , y , y⇤ ), are specially chosen:
y and y are matched for interfaces normal to x, x and x are matched for
⇤ ⇤

interfaces normal to y [14]. Figure 3.10 shows how to setup the PML boundaries
in an FDTD simulation. Though in theory PML boundaries will prevent reflec-
tions from any frequency or angle of incidence, in practise there will always be
small reflections due to the numerical discretisation of the PML equations (Eq
3.6).
The PML boundary condition that Lumerical uses is based on the PML
method implementation developed by Gedney and Zhao [58]. Their method,
which is based on a complex frequency shifted (CFS) PML, has improved anti-
reflection properties compared to the original split-PML method developed by
Berenger. In Lumerical these boundaries must be placed far away enough from
the object so that they do not interact with the evanescent fields. Lumerical
tech support suggests that placing them at least half a wavelength away will
suffice. Care must be taken for oblique incidence on the PML boundaries, as
this is where reflection is the highest.

3.2.2.6 Oblique Incidence

FDTD simulations for oblique incidence still need to be verified against Mie
Theory. The difficulty is that in broadband FDTD simulations, the angle of
incidence changes as a function of frequency in order to keep the in-plane wave
vector constant due to periodic boundary conditions (see Figure 3.11). To get
results for one angle of incidence, one would have to carry out a sweep of narrow
band simulations for each frequency. Alternatively, one could carry out a sweep
of broadband simulations at various angles and interpolate the data. To verify
the FDTD results against Mie Theory, one would compare with Shah [159] for a

77
Figure 3.10: The PML method [14]. Perfectly matched layers are applied on
the boundaries of the simulation region to absorb electromagnetic waves. The
conductivities ( x , x⇤ , y , y⇤ ) of each PML are specially chosen to prevent re-
flections. Original figure by Berenger [14], permissions granted.

78
Figure 3.11: Schematic of injection angle in broadband simulations performed
with Lumerical FDTD solutions [120]. For a single oblique broadband simula-
tion, the in-plane component of the wavevector is kept constant over the range
of frequencies. This results in an angle of injection which varies as a func-
tion of frequency. Original figure by Lumerical Solutions Inc. [120], permissions
granted.

coated infinite cylinder under oblique incidence, and Wang and Hulst [185] for
a finite cylinder.

3.3 Modelling Core-Shell Nanowires

The first device studied in this thesis is covered in section 1.3.1. This is a
TiO2 nanowire array, where each nanowire is coated in a conformal CdSe shell.
The nanowires are 1µm long, with radii ranging from 100nm to 200nm. CdSe
shell thicknesses range from 10nm to 30nm. The FDTD method implemented
in Lumerical FDTD Solutions was used to simulate light propagating through
arrays of these nanowires and measure the absorption and scattering properties.
A planewave light source of 400-800nm was simulated, since this covers most
of the solar spectrum (Figure 3.12) and was easy to fit the MCM dielectric
functions to (since TiO2 is transparent at wavelengths above 400nm).

3.3.1 Simulation Setup

Different nanowire arrangements have been compared in this study, as shown
in Figure 3.14. The default arrangement is square periodic, where there is

79
 ���

���

����� ���������� � ��� ����

���

���

���

���

���

���

�
� ��� ���� ���� ���� ����
����������� ��

2
Figure 3.12: Solar intensity in Wm nm 1
as a function of wavelength.

one nanowire per unit cell, centred so as to produce a periodic square lattice.
Hexagonal periodic arrays are also considered. The random arrangements in
Figure 3.14c are produced by placing the nanowires at random positions within a
larger super-cell, but retaining the same filling ratio as the square and hexagonal
periodic arrangements. A random number generator is used to produce the (x, y)
coordinates of each nanowire. In this case the results are averaged over 3 batches
of 25 nanowires with different random number seeds. The semi-random square
and hexagonal arrangements shown in Figures 3.14d and 3.14e are produced in a
similar way, but each nanowire is displaced by a limited random amount from an
ideal site so as to retain some of their periodicity while replicating experimental
conditions more realistically. The maximum deviations in the x, y directions
are defined as ±(p/2 a) for square and ± 23 (p/2 a) for hexagonal lattices,
where a is the core radius and p is the unit cell lattice constant. The supercell
has 16 (13) nanowires for the semi-random square (hexagonal) lattice. For each
arrangement of nanowires, results were collected for nine different combinations
of core radii and shell thicknesses: thin cores (100 nm), medium cores (150
nm) and thick cores (200 nm) were combined with thin shells (10 nm), medium
shells (20 nm) and thick shells (30 nm). Each nanowire had a length of 1 µm.
The absorption, A( ), was calculated by measuring the transmission T ( ) and
reflection R( )

A( ) = 1 (R( ) + T ( ))
This formula is much more computationally efficient to use for calculating ab-
sorption than integrating equation 3.5, but does not record any spatial inform-
ation of the absorption profile.
Simulations were also carried out for isolated nanowires. This is to provide
reference data that allows one to identify those effects that are due to the ar-
rangement of the wires, and those due to the geometry of individual wires. The
simulation region is similar to that shown in Figure 3.13b but with perfectly

80
 Reflection
PML monitor (R)

Source
TFSF bounding box

~100nm
Absorption monitor

Coating
~1000nm

Core

Light
source

Transmission
monitor (T)

~1000nm

(a) (b)

Figure 3.13: (a) 2D nanowire simulation setup. The absorption monitor (red
box) measures the amount of light that is removed by the nanowire by calcu-
lating the total transmission of light out of the box. The Total-Field Scattered-
Field box (yellow) confines the incident light field, such that any light outside
the box is only scattered. The Perfectly Matched Layers (PML) remove light
from the simulation region while minimising reflection. (b) 3D nanowire simu-
lation setup. Plane wave light (S) propagates from above. Periodic boundary
conditions are enforced in the x and y directions to simulate an infinite ar-
ray of nanowires. PML boundary conditions are enforced in the z directions.
The amount of light reflected (R) or transmitted (T) was monitored for each
wavelength.

matching layers (PMLs) on all boundaries rather than periodic boundary con-
ditions. Also, instead of calculating the fraction of light absorbed, the Mie
absorption efficiency, Qabs , is calculated,
abs
Qabs =
G
where abs is the absorption cross-section and G is the geometrical cross-section.
The absorption cross section is measured by integrating the total power loss
over the box containing the nanowire and dividing by the source intensity. The
geometrical cross-section is given by cross-sectional area of the cylinder, G = ⇡b2
where b is the total radius (core and shell included).

81
 a) Ordered b) Ordered
Square Hexagonal

c) Random

d) Semi-random Square e) Semi-random Hexagonal

Figure 3.14: Different nanowire arrangements as viewed from above. In (a)

each unit cell has one nanowire whilst in (b) each unit cell has two nanowires.
Periodic boundary conditions are enforced on the outermost dotted lines. In
(c), (d) and (e) multiple nanowires are placed within a large super-cell which
has periodic boundary conditions enforced. The inner-dotted lines in (d) and
(e) refer to the unit cells (analogous to (a) and (b)) where the position of each
nanowire is confined. For (d), the maximum displacement in the x or y directions
is ±(p/2 a) where p is the unit cell lattice constant and a is the core radius.
For (e), the maximum x and y displacements from ideal position are ± 23 (p/2 a)

82
 2.9 0.7

2.85 0.6
Refractive index, Re(n)

0.5

Extinction, Im(n)
2.8
0.4
2.75
0.3
2.7
0.2
Re(n) - experimental
2.65 Re(n) - fitted 0.1
Im(n) - experimental
Im(n) - fitted
2.6 0
400 450 500 550 600 650 700 750 800
Wavelength, nm

Figure 3.15: The complex refractive index of CdSe. Red refers to the real part,
blue to the imaginary part. The circles refer to the experimental data from
Palik [144]. The lines are Lumerical’s multi-coefficient models [120] fitted to the
experimental data.

3.3.2 Material Optical Properties

The complex refractive index for CdSe (Figure 3.15) was fitted to experimental
data from Palik [144] using Lumerical’s multi-coefficient models (MCM’s). As
TiO2 only absorbs light in the ultraviolet region (wavelength less than 400nm)
it was modelled as a non-absorbing dielectric with a real refractive index of
n = 3. Attempting to fit the dielectric function of TiO2 from ultraviolet to infra-
red would otherwise be quite difficult and would result in non-zero absorption
across the visible range. For this study one assumes that the polymer is a
non-absorbing dielectric with n = 1. The fitting for CdSe involved a necessary
compromise. The degree to which the MCM fit the real or imaginary parts
could be controlled, though fitting well to the real part could worsen the fit
to the imaginary part, and vice versa. In this case it was important to attach
more weight to the imaginary part, since this is directly proportional to how
much light is absorbed. After generating the MCM refractive indices, they were
exported and used in the Mie theory calculations in addition to the FDTD
simulations.

83
3.4 Modelling nanoparticle Nanowires
Like with the core-shell nanowires in the previous section, I have carried out
Lumerical FDTD simulations of nanoparticle coated nanowires. As covered
in Chapter 2, many solar cells have used semiconductor nanoparticles as light
absorbers, since this is how semiconductor nanocrystals form (before anneal-
ing them into a conformal layer). The nanowires are made from ZnO and are
arranged in a square lattice. Each nanowire is 1.2µm long and have radii ran-
ging from 25nm to 50nm. The nanowires are coated with PbS nanoparticles,
with diameters ranging from 4nm to 9nm and number densities from 100 to
1000 nanoparticles per nanowire. A planewave light source of 400-800nm was
used. Absorption spectra for various nanowires and nanoparticle parameters
were measured.

3.4.1 Simulation Setup

A schematic of the solar cells is shown in Figure 1.6. Three different coating
schemes were tested, as shown in Figure 3.16a. In the first scheme, nanoparticles
were arranged in a square periodic lattice conformal to the cylindrical surface of
the nanowire. In the second scheme, the nanoparticle were arranged randomly
on the surface of the nanowire. In the third scheme, a conformal shell was used
rather than a distribution of nanoparticles.
In addition, different nanowire and nanoparticle parameters were varied, as
shown in Figure 3.16b. Simulations were carried out with nanowires at radii of
25, 40 and 50nm, nanoparticles with diameters of 4 to 9nm, lattice constants of
120nm, 192nm and 240nm and coating densities from 100 to 1000 nanoparticles
per nanowire.

3.4.2 Material Optical Properties

The complex refractive index of PbS was fitted to experimental data from Palik
[144] using the Lumerical MCM, as shown in Figure 3.17. ZnO was modelled
as a non-absorbing dielectric with a refractive index of n = 2. This is due
to the fact that the bandgap is in the ultraviolet regime, outside the range of
wavelengths used in the light source.

3.5 Computational Resources

3.5.1 Mie Theory Calculations
Initially, Mathematica was used to calculate the Mie Theory cross-sections. It
was relatively fast at calculating an for a pure cylinder. However, for a coated
cylinder it took ⇠30s to calculate an for a particular core radius and coating
thickness. So the Mathematica code was ported to MATLAB, where each iter-
ation took ⇠5s.

84
 (a) (b)

Figure 3.16: (a) Three different types of nanowire coating: Ordered nano-
particles, randomised nanoparticles and conformal coating. (b) Four additional
parameters were varied: nanowire radius, nanoparticle size, number of nano-
particles and lattice constant (or spacing).

���

�
���������� �����

���

���

��� ����� � �����

���� �� � �����
� ����� � �����
���� �� � �����
���
��� ��� ��� ��� ��� ��� ��� ��� ���
����������� ��

Figure 3.17: The complex refractive index of PbS. Purple refers to the real
part, teal to the imaginary part. The cross refer to the experimental data from
Palik [144]. The lines are Lumerical’s multi-coefficient models [120] fitted to the
experimental data.

85
3.5.2 FDTD Simulations and Analysis
To run FDTD simulations, I would connect to a remote machine and launch an
instance of Lumerical FDTD Solutions. At first, simulations were run within
the graphical user interface (GUI) of the software. Lumerical has a scripting
language which allows you to setup, run and analyse simulations. For running
sweeps of simulations, scripts were written in Bash and used for interfacing
with Lumerical scripts (though one could also run parameter sweeps within
the Lumerical GUI). To allow more complex sweeps of simulations and data
analyses, Python was eventually used for interfacing with Lumerical. This was
especially useful as managing and automating large sweeps of simulations and
data analysis on remote machine becomes much easier when using Python rather
Bash.
A Python application has been written to automate and queue the FDTD
simulations of nanowire solar cells. It launches jobs which sweep over multiple
parameters of the solar cell (such as nanowire length, spacing, coating thickness
etc.). For each parameter, it will invoke Lumerical FDTD Solutions to create the
structures, and carry out an FDTD simulations and analyse the data in order to
determine the absorption spectra and profiles. The optical absorption profiles
are exported from Lumerical and analysed using MATLAB. Although Lumerical
has the capabilities of analysing 3D optical absorption data, for high resolution
simulations this becomes too computationally intensive (the Lumerical software
runs out of memory and crashes). Using MATLAB allowed multiple cores to
be used in parallel. This helped when integrating the absorption over the solar
spectrum to get the generation rate for example. Once the generation rate (as
a function of position) is calculated from MATLAB, this is used as an input
to the finite-difference charge transport code, Microvolt [35], which is described
in the next chapter. Bash scripts are used for the remainder of the simulation
pipeline (converting the generate rate from MATLAB to Microvolt format and
running Microvolt simulations).

86
Chapter 4

Charge Transport Methods

4.1 Theory of Charge Transport

Jenny Nelson has written a comprehensive book, “The Physics of Solar Cells”
[137], which describes the basic principles of photovoltaics, electrons and holes in
semiconductors, and generation/recombination processes. It goes into detailed
analysis of the p-n junction, the most widely used interface in photovoltaics.
There are areas the book does not cover that are relevant to the modelling
of nanowire ETA solar cells such as: radial p-n junctions [85], numerics (how
to solve the drift-diffusion equations) [158] and organic charge transport (only
briefly mentioned) [6]. This is covered in other literature.

4.1.1 Basic Photovoltaic Principles

When a photon is absorbed in a material, it excites an electron to a higher
energy state, leaving behind a positively charged hole. Normally the electron
would quickly relax back to its ground state, recombining with the hole. In solar
cells, an in-built asymmetry in the electronic properties of the material allows
the excited electron to be pulled away from the hole before it can recombine.
The potential difference generated by the surplus energy of the excited electron
is used to drive it through a load in an external circuit.
Solar cells are often compared by measuring the short-circuit current density
Jsc , open-circuit voltage Voc , and fill-factor F F ; these quantities are defined
below. The effectiveness of a solar cell is determined by the quantum efficiency,
QE(E), which is the probability that an incident photon with energy E will
drive one electron to the external circuit. Jsc is related to QE by the following
expression [137], ˆ
Jsc = q bs (E)QE(E)dE

where bs (E) is the incident spectral photon flux density (number of photons
incident on unit area per unit time with energy E to E + dE). The quantum

87
Figure 4.1: The quantum efficiency of GaAs as a function of wavelength, in
comparison with the photon flux density of solar irradiation [137]. The quantum
efficiency is high at wavelengths where there is high solar intensity, making GaAs
suitable for photovoltaic devices. Original figure by Nelson [137], permissions
granted.

efficiency of a solar cell is dependent on many factors, such as the absorption

coefficient, recombination rate and charge separation efficiency of the underlying
materials. For optimum performance one desires a high quantum efficiency at
wavelengths where the solar spectrum is most intense. Figure 4.1 shows the
quantum efficiency of GaAs, a commonly used photovoltaic semiconductor, as
compared the solar spectrum.
The open-circuit voltage, Voc is the potential difference across the contacts
of the solar cell when they are isolated, which is the maximum voltage. This is
related to the short circuit current by
✓ ◆
kB T Jsc
Voc = ln +1
q J0
where kB is Boltzmann’s constant, T is the temperature, q is the electron charge
and J0 is a constant. This expression is derived in Chapter 1 of Nelson’s book
[137].
The operation of a solar cell is characterised by measuring or computing its
J-V curve, defined over a range of voltages from 0 to Voc , as shown in Figure
1.4. The power density is given by the product of current density and voltage,
P = JV . As mentioned in section 1.2.2, the overall efficiency, ⌘, of a solar cell is
determined by the power density delivered at the maximum power point divided
by the incident solar power density Ps ,
J m Vm Jsc Voc F F
⌘= =
Ps Ps
where F F is the fill-factor, describing the ‘squareness’ of the J-V curve. The

88
fill-factor is reduced by leakage currents around the edges of the solar cell, the
internal resistance and the series resistance between the contacts and solar cell.
The solar cell parameters Jsc , Voc , F F and ⌘ can all be determined by meas-
urement. In order to calculate these properties theoretically, one must formulate
a theory of charge transport in semiconductors. The following sections describe
the quantum mechanics of electrons and holes in semiconductors, leading to
the drift-diffusion equations and culminating in the numerical methods used to
solve them and hence calculate the solar cell parameters.

4.1.2 The Quantum Mechanics of Electrons and Holes in

Semiconductors
Atoms without a full outer electron shell will form a molecule when brought
together. This is due to electrostatic forces between the electrons and nuclei of
each atom. The shell of electrons in an atom occupy atomic orbitals, solutions to
the Schrödinger equation which describe their probabilistic position and energy.
Each atom’s atomic orbitals will overlap when brought together, with their
associated energy levels each splitting into multiple energy levels. As the number
of atoms increases to a macroscopic number, the number of energy levels will
effectively form a continuum of energies, known as a band. Solid-state crystalline
materials consist of 3D lattices of such atoms bound together via ionic, covalent
and metallic bonds.
For a temperature of absolute zero, the only kinetic energy atoms have is
due to zero point motion, and the electrons occupy their lowest energy levels.
Due to the Pauli Exclusion principle, two electrons cannot occupy the same
single particle quantum state, which means that each spatial orbital can only
be occupied by at most two electrons, one spin-up and one spin-down. Therefore,
electrons in a solid will fill up energy levels by increasing energy until they reach
the highest occupied level. In an insulator or semiconductor, the electrons
completely fill the valence band, contributing to cohesion but not electrical
conductivity. If some electrons are excited to a high enough energy they can
cross the band gap which has no states, and enter the conduction band, leaving
holes behind in the valence band. They are now free to move, and contribute
to electrical conductivity. For metals the valence and conduction bands overlap
(there is no bandgap), meaning that electrons are free to conduct even with very
low excitation energies. For insulators, the bandgap is large enough that almost
no electrons at room temperature have enough energy to cross the bandgap from
the valence band to the conduction band. For semiconductors the bandgap is
small enough that some electrons will have enough energy to cross the bandgap
into the conduction band and become free charge carriers. Since holes can be
filled in by a neighbouring electron, they move in opposite direction to the
electrons. As such they are treated as charge carriers with opposite charge to
the electrons, and their own effective mass and mobility.

89
4.1.2.1 Band Structure
Atoms in a crystalline solid form identical repeating units, with the exception of
defects. Therefore, to model the electronic properties, one solves the Schrödinger
equation for electrons in an infinite periodic lattice. The solutions take the form
of Bloch wavefunctions [137]

(k, r) = un,k (r)eik·r

where un,k (r) is a function with the same periodicity as the lattice, k is the
wavevector and n denotes the crystal energy band. The energies En (k) (eigen-
values to (k, r)) are a rich set of solutions which can be expressed as a band-
structure diagram: see Figure 4.2. For semiconductors with direct bandgaps, the
minimum of the conduction bands E(k0c ) and maximum of the valence bands
E(k0v ) have the same k values, k0c = k0v . Close to these points the energy is
quadratic in k and can be expressed using the parabolic band approximation.
In 1D,
~2 |k k0c |2
E(k) = E0c +
2m⇤c
for electrons and
~2 |k k0v |2
E(k) = E0v
2m⇤v
for holes, where E0c = Ec (k0c ), E0v = Ev (k0v ), and m⇤c and m⇤v are the effective
masses of electrons and holes [137], defined by

1 1 @ 2 Ec (k) 1 1 @ 2 Ev (k)
= , = .
m⇤c ~2 @k 2 m⇤v ~2 @k 2

In higher dimensions the effective masses are second-rank tensors rather than
scalars.

4.1.2.2 Electron and Hole Statistics

In semiconductor physics, every position r is treated as though it were part of
a perfectly periodic system, so that band structures apply. The position depen-
dent densities of electrons and holes in semiconductors are therefore calculated
by integrating over the density of states g(k), and the distribution function
f (k, r), such that
ˆ
n(r) = gc (k)fc (k, r)dk
cond band
ˆ .
p(r) = gv (k)(1 fv (k, r))dk
val band

The density of states gives the number of single particle states per unit energy,
per unit volume. For a given band it corresponds the number of k states per

90
 E

Ec(k)

Eg

Ev(k)

Figure 4.2: Band structure diagram for a direct bandgap semiconductor. The
energy bands are represented at the Brillouin zone in a one-dimensional cross-
section of k-space. The conduction band Ec (k), and valence band Ev (k) are
separated by the bandgap Eg .

unit volume with energy in the neighbourhood of a given value. For a quadratic
conduction band, the DOS in terms of energy is [137]
✓ ◆3/2
1 2m⇤c
gc (E) = (E Ec0 )3/2 (4.1)
2⇡ 2 ~2

and for a quadratic valence band it is

✓ ◆3/2
1 2m⇤v
gv (E) = (Ev0 E)3/2 . (4.2)
2⇡ 2 ~2

The distribution function f (k,r) is the probability that an electron at point r

occupies state k. At a temperature of absolute zero all the available states with
energy below a threshold are occupied. The distribution function then takes the
form of a step-function, (
1 E  EF
f (E) =
0 E > EF
where EF , the Fermi energy, is the highest occupied energy state at T = 0. For
non-zero temperatures in equilibrium (when EF no longer depends on position),
the distribution function takes the Fermi-Dirac form [137],
1
f (k, r) = f0 (E(k), EF , T ) =
e(E(k) EF )/kB T +1
which approaches the step function as T ! 0. The electron and hole densities
are then given by ˆ 1
n= gc (E)f0 (E, EF , T )dE (4.3)
Ec

91
and ˆ Ev
p= gv (E)(1 f0 (E, EF , T ))dE (4.4)
1

4.1.2.3 Boltzmann Approximation for Charge Carrier Densities

By assuming the Fermi energy lies far away from the band edges (e.g, in the
middle of the bandgap) one can use the Maxwell Boltzmann approximation for
the distribution function, f0 [137]. In the conduction band for E EF ,

f0 (E, EF , T ) ⇡ e(EF E)/kB T

and in the valence band for E ⌧ EF ,

1 f0 (E, EF , T ) ⇡ e(E EF )/kB T

.

One can then substitute these expressions and those of equations 4.1 and 4.2
into the integrals for n and p (equations 4.3 and 4.4) to get

n = Nc e(EF Ec )/kB T
, p = Nv e(Ev EF )/kB T
(4.5)

where ✓ ◆3/2 ✓ ◆3/2

m⇤c kB T m⇤v kB T
Nc = 2 , Nv = 2 (4.6)
2⇡~2 2⇡~2
are constants, defined as the effective density of states for the conduction and
valence bands respectively [137]. Equation 4.5 tells us that the electron (hole)
density grows (decays) exponentially with the Fermi level. One can then define
a constant property, the intrinsic carrier density ni as

n2i = np = Nc Nv e Eg /kB T
,

which is the density of electrons (holes) excited to the conduction band (va-
lence band) of a pure semiconductor in thermal equilibrium [137]. This decays
exponentially with the size of the bandgap.

4.1.2.4 Doping and Bias

By introducing impurities with a valence different from that of the atoms in
the host crystal, the electrical properties of the semiconductor can be changed.
To demonstrate this concept with an example [137], take silicon. Each atom
is covalently bonded to 4 other atoms, by sharing its 4 outer electrons with its
neighbours. They in turn share their electrons back, so that each silicon atom
has a full electron shell with 8 covalent bonds. If one introduces a phosphorous
atom, which has 5 outer electrons, then one of the electrons is left when all
covalent bonds are formed. The extra electron exists in a loosely bound state
centred on the phosphorous atom. The energy required to excite this electron to
the conduction band is much lower than for the electrons contributing to bond

92
 p-type intrinsic n-type
Evac

EC
EF
Energy donor level
EF
acceptor level

EF
EV

Figure 4.3: Energy level diagrams for semiconductors with p-type doping (left),
n-type doping (right), and no doping (middle). The Fermi level is represented by
the dashed lines while the acceptor and donor levels are represented by dotted
lines.

formation, so at room temperature is it likely to be a free charge carrier and

contribute to the electrical current. The phosphorous atom is known as a donor,
since it donates extra electrons to the conduction band. Introducing electron
donors into the semiconductor is known as n-type doping, since it creates n-type
charge carriers. The Fermi level is raised, since extra states have been created
closer to the conduction band (see Figure 4.3).
One can also create p-type doping, by introducing electron acceptors in the
semiconductor. For example, introducing boron, which has a valence of 3, into
a silicon semiconductor [137]. The 3 outer electrons form covalent bonds with
neighbouring silicon atoms. There is a remaining covalent bond left to be filled
by an electron. With enough energy, the boron (electron acceptor) removes an
electron from the valence band creating a hole. The energy required is much less
than the bandgap, with the energy level of the grabbed electron residing closer
to the valence band maximum than the middle of the bandgap. The Fermi level
therefore resides somewhere between Ev and this energy level (see Figure 4.3).

Quasi Fermi levels Under a small bias (either by illumination or applied

voltage), the Fermi level EF is split into two quasi Fermi levels for electrons
and holes, EF,n and EF,p . This is because the populations of electrons and
holes have been perturbed, and particles are able to reach a state of quasi
thermal equilibrium with particles of their own type, but not with particles of
the opposite type [137]. If the bias is not too great then the electrons and holes
have equilibrium-like distributions with their associated quasi-Fermi levels. The
electron and hole densities are re-expressed by substituting EF,n and EF,p for
EF in equation 4.5,
(Ec EF,n )/kB T (EF,p Ev )/kB T
n = Nc e , p = Nv e . (4.7)
Increasing the bias and therefore the splitting of the Fermi levels causes an
exponential increase in the population density of electrons and holes, since
np = n2i eEF,n EF,p

93
4.1.3 Charge Carrier Dynamics in Crystalline Semicon-
ductors
4.1.3.1 Current Density
The current density can be found by weighting the carrier density by the group
velocity, v = ~1 rk E(k) [137], which for parabolic bands is

q~
ˆ
Jn (r) = kgc (k)fc (k, r)d3 k
m⇤c cond band
. (4.8)
q~
ˆ
3
Jp (r) = ⇤ kgv (k)(1 fv (k, r))d k
mv val band

The Fermi-Dirac distribution f0 is symmetric in E (or k), so when integrated

with k this contributes nothing to the current. Under bias, the distribution
functions have a small anti-symmetric term fA (k, r) term added to them so
that the current is non-zero [137],

fc (k, r) = f0 (E, EF,n , T ) + fA,c (k, r)

.
fv (k, r) = f0 (E, EF,p , T ) + fA,v (k, r)

These non-equilibrium contributions can be found by solving the Boltzmann

equation. In the relaxation time approximation this is

@fc ~k @fc F @fc (fc f0 )

+ · + · =
@t m @r ~ @k ⌧
where ⌧ is the relaxation time constant. In the absence of external fields (F = 0),
this can be solved for the steady state to give [137]
✓ ◆
⌧v
fc = f0 1 · rr EF,n
kB T

and similarly for the valence distribution function fv . By substituting fc and

fv into equation 4.8, one obtains the current densities as proportional to the
quasi-Fermi level gradients [137],

Jn (r) = µn nrr EF,n

(4.9)
Jp (r) = µp prr EF,p

where
✓
◆2
q ~k
ˆ
µn n = ⌧ gc (k)f0 (E(k))d3 k
kB T cond band m⇤c
✓ ◆2
q ~k
ˆ
µp p = ⌧ gv (k)f0 (E(k))d3 k
kB T val band m⇤v

94
4.1.3.2 Drift-Diffusion Model
By using the Boltzmann approximation (equation 4.5) the conduction band,
valence band and Fermi energy can be expressed as [137]

Ec = Evac
Ev = Evac Eg
EF = Evac kB T ln Nc = Evac Eg + kB T ln Nv

where = Evac Ec is the electron affinity (the minimum energy required to

remove an electron from the solid).
Using these definitions for the energy levels, one can re-express the currents
as
Jn (r) = qDn rn + µn n(qF r kB T r ln Nc )
Jp (r) = qDp rn + µn p(qF r rEg + kB T r ln Nv )
where F = (rEvac )/q is the electric field. For a homogeneous material the
gradients in , Eg , and dopant densities disappear. This allows us to express
the currents in terms of the drift-diffusion model [137],

Jn = qDn rn + qµn En
Jp = qDp rp + qµp Ep (4.10)

where J and E are the 3D vectorial representations of the current and elec-
tric field, and D is the diffusion constant given by the Einstein relation µ =
qD/(kB T ). The essence of the drift-diffusion model is that electrons and holes
diffuse down concentration gradients (first term of equation 4.10) and drift along
electric fields (second term of equation 4.10).
Conservation of charge requires that

@n 1
= r · J n + Gn R n
@t q
@p 1
= r · J p + Gp R p (4.11)
@t q
where G is the generation rate of free charge carriers (due to photo-absorption)
and R is the recombination rate (the rate at which free electrons recombine with
holes and become bound again) [137]. The recombination rate is discussed fur-
ther in section 4.1.3.3. These equations are solved together with the electrostatic
Poisson equation
q
r · (r ) = (NA ND + n p) (4.12)
"
where is the electrostatic potential ( = qV ), " is the dielectric function
and Nd/a is the donor/acceptor density. In charge transport calculations, one
is normally interested in the steady state where
@n @p
= =0
@t @t

95
Solar cell parameters (such as Jsc , Voc and ⌘) are calculating by solving this
equation for a range of voltages.

4.1.3.3 Recombination of Charge Carriers

There are three different types of recombination: Radiative, Auger and Shockley-
Read-Hall.
Radiative recombination is unavoidable, in the sense that it is a fundamental
material property. It is due to spontaneous relaxation of electrons from a higher
energy state (conduction band) to a lower energy state (valence band). This is
based on first order perturbation theory; Fermi’s Golden Rule, which states the
probability of transition per unit time from a state hi| with energy Ei to a state
0
|f i with energy Ef under the perturbation Hamiltonian H is
2⇡ 0 2
i!f = hi| H |f i (Ef Ei ± E).
~
In Nelson’s book [137], from this Golden rule, the radiative recombination rate
is derived as
Rrad = Brad (np n2i )
where 1
1 2⇡
ˆ
Brad = n2s ↵(E)e E/kB T
E 2 dE
n2i h3 c2 0

and ↵ is the absorption coefficient.

Auger recombination is also unavoidable and is caused by the collision of
identical charge carriers. For example, as two electrons in the conduction band
collide, one is excited to a higher energy whilst the other undergoes a transition
to the valence band, recombining with a hole. The excited electron subsequently
relaxes back to the conduction band edge, losing that energy as heat. The Auger
recombination rate is proportional to the densities of the three carriers involved
in this process (two electrons and one hole, or two holes and one electron),
resulting in the following expression

RAug = Cp (n2 p n20 p0 ) + Cn (np2 n0 p20 )

where Cp is the Auger coefficient for two-electron collisions and Cn is the coef-
ficient for two-hole collisions [137].
The dominant recombination type is Shockley-Read-Hall recombination, which
is avoidable because it depends on material quality. This type of recombination
involves defects or trap states in the band gap, which capture free electrons
in the conduction band. Electrons that reside in a trap state can be freed via
thermal activation, allowing the capture of holes from the valence band. For a
semiconductor containing a density of Nt trap states at an energy level of Et ,
the recombination rate is
np n2i
RSRH =
⌧n (p + pt ) ⌧p (n + nt )

96
 p-type n-type
Evac

Φn
EC

Energy Φp

EF

EV

Figure 4.4: The energy levels of electrons in semiconductors with p-type dopants p-type n-type
(left) and n-type dopants
Evac (right). Evac is the vacuum energy, at which an
electron becomes unbound from the material. Ec and Ev mark the lowest and
highest levels of the Econduction
C band and valence band respectively. EF is the
Fermi level, Energy
which resides in the bandgap.
qVbi The workfunction, denoted by , is
the energy required to excite an electron from the Fermi level to Evac (i.e, the
lowest energy requiredE to release an electron from the semiconductor).
F

EV
where nt is the electron density when EF,n = Et , pt is the hole density when
EF,p = Et , and ⌧n and ⌧p are the lifetimes for the electron and hole respectively
[137]. These are defined as
1 1
⌧n = , ⌧p =
⌫n n Nt ⌫p p Nt

where ⌫n (⌫p ) is the mean thermal velocity of the electron (hole) and n ( p) is
the cross-section of capture for electrons (holes) by the trap state.

4.1.4 The p-n Junction

The p-n junction is the standard model which describes the basic fundamentals
of how most solar cells work. Two layers of a material are doped differently, such
that one layer has an abundance of electron donors (n-type) and the other layer
has an abundance of electron acceptors (p-type) (see Figure 4.4). The doping
shifts the position of the Fermi level such that in the p-type layer, the Fermi
level is closer to the valence band (larger workfunction p ) and in the n-type
layer it is closer to the conduction band (smaller workfunction n ).
When the two layers of the material are brought together (as in Figure 4.5a),
the interface is defined as the p-n junction. Due to the mismatch in Fermi levels,
electrons diffuse from the n-type layer to the p-type layer (and holes in the
other direction), leaving behind positively charged donors (negatively charged
acceptors). A finite region around the p-n junction becomes depleted of charge
carriers and an electric field builds up. Photo-generated electrons are driven by
the electric field towards the n-type side and holes towards the p-type side. At
equilibrium, the diffusion is balanced by the drift of carriers due to the built-up
electric field, so no net current occurs. In this situation the Fermi levels of the
n and p type layers are equalised. The difference in workfunctions of the layers

97
is taken up by the change in conduction and valence bands, which is manifested
as a built-in bias or potential difference across the layers,
1
Vbi = ( n p)
q
Figure 4.5b illustrates the band-bending that occurs in the p-n junction. By
applying an external bias V to the p-side, the potential across the junction
becomes
Vj = Vbi V
A positive bias will reduce the size of the step between the energy bands and
hence reduce the driving force of the electrons, whereas a negative bias will
increase it.

4.1.4.1 Theoretical Analysis of the 1D p-n Junction

Nelson has carried out a theoretical analysis of the 1D p-n junction [137]. This
offers a starting point to formulate a charge transport model for the TiO2 /CdSe
solar cell. By making a few assumptions about the formation of the p-n junction
[137] the charge carrier dynamics can be solved analytically.
The first approximation is that the charged region around the p-n junction
is depleted of free carriers (depletion zone) and that the built-in potential drops
exclusively in this region (see Figure 4.6). The boundaries of this region are
at a fixed distance from the p-n junction, wp into the p region and wn into
the n region. This is also known as the depletion approximation. Outside the
depletion zone the electric field is zero, the n and p regions are neutral, so the
currents (consisting of minority carriers) are driven only by diffusion. In this
case the electron (hole) density in the neutral n-type (p-type) region takes its
equilibrium value, n = Nd (p = Na ). Inside the depletion zone, the applied bias
Vj drives majority carrier currents. The difference in the quasi Fermi levels,
qVj , provides a boundary condition on the minority carrier density at the edges
of the depletion zone, allowing the minority carrier currents to be solved in the
neutral regions [137].
The second approximation is that in the neutral regions, the recombination
rate is linear in minority carrier density. Also known as the superposition ap-
proximation, this ensures that there is no coupling of the effects of illumination
and bias [137]. Together these approximations decouple the neutral n and p
regions and simplify the current equations, allowing solutions to be obtained
analytically.

The current solutions The net current, J(x), in the solar cell is spatially
constant when in steady state. Therefore, one can evaluate J(x) at any point.
By evaluating at the p-type boundary of the depletion zone [137], wp , the net

98
p-type n-type

Φn

p-type n-type
Evac

Φn
EC

Energy Φp p-type n-type

EF

qVbi
EV

(a)
p-type n
Evac

EC
qVbi
Energy

EF

EV

(b)

Figure 4.5: (a) The p-n junction. (b) Energy level diagram of the p-n junction
at equilibrium. One can picture the energy levels in Figure 4.4 with different
Fermi levels. When joined together, near the junction (fine dotted line) the
energy bands ‘bend’ by an amount qVbi to accommodate the equidistribution
of the Fermi level. Electrons are represented as filled circles, holes as empty
circles.

99
 n dopant atom
n dopant ion
depletion zone
p-type n-type p dopant atom
p dopant ion
n carrier (electron)
p carrier (hole)

x
wp 0 wn

Figure 4.6: Depletion zone in the p-n junction, marked by the faded area within
the dashed lines with a width of wn + wp . In this region, under diffusion the
electrons have left the n-type donor atoms leaving them as positively charge
ions. Likewise, holes have left the acceptors leaving them negatively charged.
The resulting electric field (which resides only in the depletion zone) acts as a
barrier to further diffusion.

100
current is given by

J= Jn ( w p ) Jp ( w p )
= Jn ( wp ) Jp ( wn ) Jscr
ˆ ˆ ˆ
= jn (E, wp )dE jp (E, wn )dE jscr (E)dE

where Jn (x) is the electron current at position x, Jp (x) is the hole current and
Jscr is the net adjustment to electron and hole currents across the space-charge
region (depletion zone). The current is expressed as an integral over the solar
spectrum, where j(E, x) is the current under monochromatic light of energy E.
The third term, Jscr , is also known as the recombination-generation current
[137]. It can be split into terms for recombination and generation,
ˆ wn ˆ ˆ wn
Jscr (V ) = Jrec + Jgen = q U dx q g(E, x)dxdE.
wp wp

The full solutions to the currents and charge densities for the 1D p-n junction
are derived in chapter 6 of Nelson’s book [137].

4.1.4.2 The p-i-n junction

The p-i-n junction is a slightly more complicated version of the p-n junction,
in that there is a layer of undoped semiconductor sandwiched in between the
p and n layers. The processes are similar as there is a built-in bias over the
p-i-n junction, but it extends over a longer distance, as shown in Figure 4.7.
Compared to the simpler p-n junction, the advantage is that carriers generated
in the i-region survive longer than in the doped p or n layers. However, the
disadvantage is that the i-region has greater resistance and increased chance of
recombination [137].

4.2 Numerical Methods - Microvolt

Microvolt is a free C++ object-oriented code for semiconductor device modelling
[131]. The Poisson and current continuity equations for electrons and holes
(Eq 4.13) are discretised in the finite-difference scheme and solved iteratively
using the Newton-Raphson method. In this method, the differential equations
describing the semiconductor device physics (4.13) are converted into a linear
set of algebraic equations. These equations are solved on a cartesian mesh
where the values of n, p and are sampled. Selberherr’s book “Analysis and
Simulation of Semiconductor Devices” describes in detail the finite-difference
scheme which Microvolt uses for simulating semiconductor devices [158]. The
Microvolt code allows for many different meshes of differing resolutions, with
cartesian or cylindrical coordinates. I use it for the simulations reported here.

101
 p-type intrinsic n-type

Evac

EC

Energy

EF

EV

Figure 4.7: Energy level diagram of p-i-n junction. The region in the middle
p
has no doping, but has an intrinsic charge density ni = np. Band-bending
occurs over a longer distance than in the p-n junction.

4.2.1 The Model used by Microvolt

The band diagram in Figure 4.8 illustrates some of energy level concepts in
Microvolt. The formulation of the drift-diffusion model is slightly different than
from Nelson’s. The vacuum level, Evac , is fixed at E = 0 and other energy levels
are defined relative to it. Also, the distance between the bottom of the conduc-
tion band and vacuum level is split into two contributions, electron affinity
and electrostatic potential , so that an electron in the conduction band needs
an energy equal to + to leave the semiconductor. The electron and hole
concentrations are now calculated as
✓ ◆
EF,n (Ei q )
n = ni exp
kB T
✓ ◆
(Ei q ) EF,p
p = ni exp
kB T

where Ei q is the intrinsic Fermi level (Fermi level of an intrinsic semicon-

ductor), defined at = 0 as
✓ ◆
Eg 1 Nv
Ei = + kB T ln .
2 2 Nc

This is different to the Nelson formulation as , the electrostatic potential, has

been exposed in the expressions for n and p, where as before it was implicit in
the variation of the non-fixed vacuum energy Evac .
In Microvolt the Poisson and continuity equations (4.12 and 4.11) are solved
in the steady state such that rate of change of n and p are zero. Rearranging,
one gets

102
Figure 4.8: Band diagram representing how semiconductor devices are modelled.
Original figure taken from Microvolt website [131], permissions granted.

r · ("r ) = q(p n + ND NA )
r · Jn = q(G R) (4.13)
r · Jp = q(G R)
The generation rate G = G(x, y, z) can be calculated by an external optical
simulation (for example, using Lumerical FDTD Solutions [120]).
The current densities can be expressed in terms of the electron/hole quasi-
fermi levels as in equation 4.9,

Jn = µn nrEF,n
.
Jp = µp prEF,p
Following the definitions of EF,n and EF,p in equation 4.7, one can also express
the current densities in terms of electron/hole concentrations,

Jn = µn nr(q Ei log ni ) + qDn rn

.
Jp = µp pr(q Ei log ni ) qDp rp
As a reminder, the intrinsic charge density is given by
Eg
n2i = Nc Nv e kT

At equilibrium, the density of electrons and holes are defined as

n0 p0 = n2i , n 0 + N A = p0 + N D .
Recombination is a sum of three terms: Radiation recombination Rrad , Auger
recombination Raug and Shockley-Read-Hall recombination RSRH
Rrad = B(np n 0 p0 )
Raug = Cn n(np n0 p0 ) + Cp p(np n 0 p0 )
(np n0 p0 )
RSRH =
⌧p (n + n1 ) + ⌧n (p + p1 )

103
where ⌧ = ⌫th1Nt , and n1 and p1 are the equilibrium electron and hole con-
centrations at the surface, defined in the following section. The mechanisms of
recombination are covered in section 4.1.3.3.

4.2.1.1 Boundary Conditions

The boundary conditions are explained in detail in the Microvolt website [131].
At a semiconductor-metal interface, the boundary conditions for the current are

Jn · n = qsn (n n1 ), Jp · p = qsp (p p1 )
where sn and sp are the electron/hole recombination velocities. n1 and p1 are
are given by

✓ ◆ ✓ ◆
B m
n1 = n0 exp = NC exp (4.14)
kT kT
✓ ◆ ✓ ◆
B m Eg
p1 = p0 exp = NV exp (4.15)
kT kT

where is the electron affinity, m is the metal workfunction, B = m s is

the Schottky potential barrier and s is the semiconductor workfunction. B is
positive for n-doped semiconductors and negative for p-doped semiconductors.
For Ohmic contacts, B = 0 and therefore the surface equilibrium concentra-
tions are equal to those of the bulk,

n1 = n0
(4.16)
p1 = p 0
In the case of semiconductor-insulator interfaces, the boundary conditions for
the current densities are
Jn · n = qRsurf
(4.17)
Jp · p = qRsurf
where the surface recombination is given in the Shockley-Read-Hall form,
np n0 p0
Rsurf = 1 (4.18)
sp (n + n1 ) + s1n (p + p1 )

4.2.2 Finite-Difference Discretisation

The finite-difference discretisation of the Poisson equation (Eq 4.13) in 2D is (if
" is constant),
i+1,j i,j i,j i 1,j i,j+1 i,j i,j i,j 1
!
xi xi yi yi
" xi + xi 1
1
+ yi + y i 1
1
(4.19)
2 2
ni,j + pi,j + Ci,j = 0 (4.20)

104
where Ci,j is the total fixed charge density due to dopants ND NA at the
location (xi , yj ). For the current continuity equations,

Jn|i+1/2,j Jn|i 1/2,j Jn|i,j+1/2 Jn|i,j 1/2

xi + xi 1
+ y i + yi 1
= q(Gi,j Ri,j ) (4.21)
2 2
Jp|i+1/2,j Jp|i 1/2,j Jp|i,j+1/2 Jp|i,j 1/2
xi + xi 1
+ y i + yi 1
= q(Gi,j Ri,j ) (4.22)
2 2

where we define the subscripted values as

ui,j = u (xi , yj )
✓ ◆
xi + xi+1
ui+1/2,j = u , yj
2
✓ ◆
yj + yj+1
ui,j+1/2 = u xi ,
2

The current densities are expressed using the Scharfetter-Gummel approx-

imation [155]. For example, one of the components of the current density is
given by
0 0 0 0
B( i+1,j i,j )ni+1,j B( i,j i+1,j )ni,j
Jn|i+1/2,j = qDn|i+1/2,j (4.23)
xi,j
0
where = /kB T is the scaled potential and B(x) = exx 1 is the Bernoulli
function. This approximation is numerically effective when dealing with large
variations in the charge concentration.

4.2.3 Solution of Discretised Equations

The discretisation of the Poisson and continuity equations results in a system of
nonlinear algebraic equations. These equations must be solved iteratively. First
we rewrite them in the following form [158]
0 1
f (w)
F(w) = @fn (w)A = 0 (4.24)
fp (w)

where 0 1

w = @nA (4.25)
p
is the vector of solutions , n and p defined at all mesh points and F is a vector
function of rank three, consisting of the vector solution functions f , fn and fp .
These correspond to the discretised Poisson and continuity equations (rearrange
equations 4.20, 4.21 and 4.22 so that zero is on the right-hand side). In the case
of Microvolt, one uses the quasi-Fermi level representation where En and Ep are

105
used instead of n and p for the solving stage of the algorithm. One can find the
solutions to equation 4.24 using the Newton method. This is a method to find
the root of an equation, for example in 1D it’s a method to solve f (w) = 0. In
this method you iteratively update the solution
f (wk )
wk+1 = wk (4.26)
f 0 (wk )

for the k th step. For solving the semiconductor equations in 3D, this is extended
to
J(wk ) · wk = F (wk ) (4.27)
where wk = wk+1 wk and J(w) is the Jacobian matrix consisting of all the
partial derivatives of F(w),
0 @f @f @f
1
@ @n @p
B n @fn C
J(w) = @ @f
@
@fn
@n @p A (4.28)
@fp @fp @fp
@ @n @p

The iterations are repeated until |F(wk )| is smaller than some predefined value
(i.e, the solution has converged). To calculate w, the Jacobian must be in-
verted. Since the Jacobian is sparse (only the terms along and adjacent to the
diagonal are non-zero) one can use an optimised linear algebra procedure in-
vert it. One such method is LU decomposition [179], where equation 4.27 is
reformulated as
0 @f 10 1 0 1 0 @f @f
10 1
@ 0 0 f 0 @n @p
B @fn @fn C B @fn C @ n A
@@ @n 0 A @ nA = @ fn A @0 0 @p
A
@fp @fp @fp p k+1 fp k+1 p k
@ @n @p
0 0 0
(4.29)
After this reformulation the equations can be solved by a series of substitution
steps.
There are two different ways to implement this method described above. One
can use the classical Newton method, where a full Newton iteration consists of
inverting a 3N ⇥ 3N matrix, since there are N mesh points and 3 equations for
each (Poisson, electron and hole continuity equations). Alternatively, one can
use the Gummel Method, where the Poisson equation and continuity equations
are solved in separate iterations. This algorithm is detailed in Figure 4.9 and
is the one used by default in Microvolt. This method converges faster, but
should not be used in exceptionally high doping situations because the coupling
between the continuity equations and Poisson equation is stronger.

4.3 Implementation of Microvolt

The Microvolt application is a heavily optimised object-oriented C++ code,
making use of advanced C++ features and numerical libraries. The source code

106
 (1) Make a few Newton iterations for the continuity equations (2N ⇥ 2N
matrix)
(2) Substitute En ,Ep into Poisson equation, f
(3) Make a few Newton iterations for the Poisson equation (N ⇥N matrix)
(4) Substitute into continuity equations, fn and fp . Return to (1)

Figure 4.9: Gummel method as implemented in Microvolt

is split into different parts describing the user interface functions, the physical
system, the gridding, finite-difference scheme and numerical solvers. LAPACK
and PARDISO numerical libraries are used for the matrix inversions and multi-
plications. Significant computational power and usage of the Intel MKL libraries
are required to run calculations in a reasonably short time. For this reason, Mi-
crovolt is compiled and run on a Linux workstation on the Imperial College
CMTH network.

4.3.1 Coupling FDTD to Microvolt

To accurately calculate the photocurrent in a 3D solar cell, one must feed into
a Microvolt simulation a profile of the generation rate as calculated by FDTD.
As described in section 3.5.2, a script was needed to take the optical absorption
profile from Lumerical and integrate over the solar spectrum. This results in
the generation rate in m 3 s 1 ,
max
ˆ
G(r) = ↵( , r) ( )d (4.30)
min
hc

where ↵ is the optical absorption per unit volume multiplied by the area of the
unit cell (resulting in units of m 1 ), is the wavelength, is the solar spectrum
intensity (in units of Wm 2 nm 1 ), h and c are Planck’s constant and the speed
of light respectively. ↵ is obtained from the FDTD simulation, whereas G is
calculated by numerical integration using MATLAB.

4.3.2 Microvolt Tests

4.3.2.1 Reproducing Deinega’s Results
In order to test the scheme to couple the optical and charge transport simu-
lations, I reproduced simulations by Deinega et al [35] for radial p-n junction
silicon nanowires. In particular this means getting a short-circuit current dens-
ity of Jsc = 9.48 mA cm 2 and an open-circuit voltage of Voc = 0.36 V. Figure
4.10 shows a diagram of Deinega’s nanowire model. In this case the parameters
are L = 2µm, R = 200nm, d = 100nm. Deinega et al used the same silicon
parameters as Kayes et al [85]. Their input to the charge-transport model,

107
the FDTD absorption profile, was obtained using an FDTD simulator from the
Electromagnetic Template Library, EMTL [44]. For this, the nanowires were
arranged in a square periodic lattice with period a = 800nm and embedded in
silica (refractive index n = 1.5) to the top surface. For reproducing the FDTD
results I used Lumerical. A diagram of the Lumerical simulation setup is shown
in Figure 4.11. After carrying out Lumerical simulations and extracting the

Figure 4.10: Diagram of silicon nanowire charge transport model [35]. Red
lines represent the metal contacts, blue lines represent the boundary with the
insulating medium. The green line separates the p-doped region from the n-
doped region. Original figure by Deinega and John [35], permissions granted.

optical absorption profile, I averaged over the azimuthal angle to produce a plot
of the absorption that depends on radius and height (r, z), as shown in Figure
4.12.
To produce the generation rate, I used MATLAB to integrate the absorption
with the solar intensity, as described in section 4.3.1. A numerical integration
was performed, but first the solar spectrum had to be interpolated to make the
integration computationally tractable. This is shown in Figure 4.13. I then
imported the generation rate into Microvolt, ran a simulation over a sweep of
voltages, resulting in the current-voltage curve in Figure 4.14.
The short circuit current I calculated was 15mA, but Deinega obtained
10mA. The disparity was due to the differing amount of absorption when com-
paring the Lumerical FDTD data with Deinega’s FDTD data (see Figure 4.15).
At first it seemed to be a problem of not having a fine enough mesh, but the
absorption spectrum converged on the wrong values as the resolution increased.
This prompted me to look at the dielectric function fit I used, in particular
the imaginary part (since the absorption is proportional to Im(")). Figure 4.16
shows the fits used by Deinega (EMTL) and Lumerical. Lumerical has an in-
built multi-coefficient model for fitting the dielectric function to the sampled
experimental data. Although the Lumerical fit looks similar to Deinega’s, the
imaginary parts differ significantly at the larger wavelengths, as shown in Figure

108
Figure 4.11: Optical setup of Deinega’s nanowire in Lumerical. The grey cylin-
der is the nanowire itself, the blue region is the silica medium. The orange box
represents the simulation region and the pink arrow represents the solar irradi-
ance. Periodic boundary conditions are enforced. The silica medium extends to
infinite depth, such that no reflection occurs between the boundary of the silica
and vacuum.

109
Figure 4.12: Optical absorption profile averaged over azimuthal angle. Results
from Lumerical are on the left, results from Deinega et al are on the right
[35]. For the Lumerical results, the highest generation rate (dark red) is 1.4 ⇥
1022 cm 3 s 1 . For Deinega’s results (which were obtained using EMTL [44])
the colour scale goes up to 1.0 ⇥ 1022 cm 3 s 1 but the generation rate probably
goes higher than this due to the fact that the redness is over-saturated. Figure
on the right by Deinega and John [35], permissions granted.

110
 Figure 4.13: Solar spectrum with interpolation on top.

��

��
������� �������� �� ����

��

��

���
� ���� ��� ���� ��� ���� ��� ���� ���
�������� �

Figure 4.14: Current voltage curve for silicon nanowire solar cell. Short circuit
current is 15mA/cm2 , open-circuit voltage is about 0.37V.

���
��������� ���� �
��� ��������� ���� �
��������� ���� �
��� ����
���
����������

���
���
���
���
�
����
����
��� ��� ��� ��� ��� ��� ��� ����
����������� ��

Figure 4.15: Absorption spectrum for the nanowire array. The dashed black
line is that calculated by Deinega using EMTL. The coloured lines are results I
obtained using Lumerical for increasing mesh resolutions.

111
4.16b. This explains the disparity of absorption, and prompted me to model the
TiO2 and ZnO as non-absorbing dielectrics in the cells studied in this thesis.

�
��������� �����
������� �����
� ��������� �����
������� �����
�
����������

�
��� ��� ��� ��� ��� ��� ��� ����
����������� ��

(a) (b)

Figure 4.16: (a) Refractive index data for Lumerical vs EMTL (Deinega). (b)
A closeup of the imaginary component of the Lumerical refractive index fit.

It turns out that Deinega uses a Modified Lorentz fit for silicon, for which
the source code for is not available: it is contained in a precompiled binary in
EMTL. However, this fit is discussed in his paper [34]. In this case, the dielectric
function depends both on the electric field and its first derivative with respect
to time. They achieve an extremely close fit to the experimental data.

4.3.2.2 Microvolt Heterojunction Test

All the work mentioned above is done in preparation for eventually being able
to simulate the TiO2 /CdSe solar cell. However, the work done by Deinega et al
does not cover heterojunctions.
In order to test the capability of Microvolt to simulate heterojunction de-
vices, I looked for a paper that had modelled a heterojunction device electrically.
Metzger [129] had written a paper on interdigitated thin-film solar cells consist-
ing of ZnO, CdS and CIGS (copper indium gallium arsenide). In this paper,
2D charge transport simulations performed for devices where the n-type and p-
type regions of CIGS are distributed in a comb like structure (see Figure 4.17a),
topped with n-type CdS and ZnO. A simple test case for a 1D planar structure
is also studied in the paper (Figure 4.17b), so I started with that. The electrical
properties are listed in Table 4.1.
Figure 4.18 shows the J-V results from the Metzger heterojunction cell.
I tested the non-interdigitating cells (see Figure 4.17b). Metzger gets Jsc =
2
23 mA/cm and Voc = 0.64 V for an equilibrium hole density of p0 = 3 ⇥
2
1017 cm 3 , and Jsc = 27 mA/cm and Voc = 0.57 V for p0 = 1 ⇥ 1016 cm 3 .
These are the results which I aimed to reproduce using Microvolt.

4.3.2.3 Debugging the Heterojunction

Initial attempts to simulate this simple planar heterojunction were fraught with
problems. As I ran the Microvolt calculation, numerical instabilities occurred

112
 (a) (b)

Figure 4.17: Diagrams of devices simulated by Metzger [129]. Light is incident

from the top. The ZnO and CdS are n-doped. (a) Planar, 1D. (b) Interdigitated,
2D. The CIGS is p-doped in (b), and in (a) it has alternating regions of n-type
and p-type doping separated by width W . Original Figures by Metzger [129],
permissions granted.

Properties ZnO CdS CIGS

Eg (eV) 3.3 2.4 2.3 [28]
(eV) 4.6 4.6 4.6
µe (cm2 V 1 s 1 ) 100 100 100
µh (cm2 V 1 s 1 ) 25 25 12.5
Doping (cm 3 ) n: 1018 n: 6 ⇥ 1017 Varied
"/"0 9 10 13.6
me /m0 0.2 1.2 0.09
mh /m0 1.2 0.8 0.72

Table 4.1: Table of properties for Metzger’s thin film solar cell [129]

Figure 4.18: Short-circuit current density (left) and open-circuit voltage (right)
for the Metzger cell [129]. The results are plotted against W (µm), the width of
the interdigitated regions. The straight black lines show results for cells without
interdigitations. Original figures by Metzger [129], permissions granted.

113
 p0 , hole density in base (cm 3
) Jsc (mA/cm2 ) Voc (V)
3 ⇥ 1017 23 0.64
1 ⇥ 1016 27 0.57

Table 4.2: Microvolt J-V results of the Metzger Heterojunction.

leading to pathological solutions, or NaNs (Not a Number). In order to determine

what was causing these issues, I tried varying different material parameters and
used the gnu debugger, gdb, to get to the bottom of the issue.
The bandgap of ZnO is 3.3eV, which is quite wide compared to silicon with its
bandgap of 1.12eV. This can produce numerical problems because the intrinsic
concentration ni decays exponentially as the bandgap increases (since ni ⇠
exp( Eg /2kB T )), leading to a value of 10 8 cm 3 (compare this with Silicon
which has ni = 1010 cm 3 ). The reason why this would be an issue is because of
the extremely large variation of electron and hole concentrations over a distance
of a few nanometers.
I discovered that the bug was related to solving the Jacobian linear system
of equations. In this case, one is trying to solve the matrix equation

J(wk ) · wk = F (wk ) (4.31)

Rather than trying to inverting the Jacobian matrix J directly, one can use a
much more efficient algorithm which indirectly inverts it via LU decomposition
(see section 4.29). The problem with the Jacobian was that it was populated
with full rows and columns of zeroes (see Figure 4.19), making it impossible to
invert.

4.3.2.4 Solution to Numerical Instabilities

The pathological rows and columns of zeroes are due to how the derivatives in
the Jacobian are computed. The method of forward differences is used, where
one approximates the derivatives as f / x where x is set to a small finite
value such as 10 5 and the resulting change in f is f ⇡ f (x + x) f (x).
This could be a problem if the change in f is several orders of magnitude smaller
than f itself. To avoid this problem, I set the generation rate G to zero while
calculating f , since this quantity is constant and dominates the expression for
f . Mathematically this is the same result,

f 0 = [f (x + x) G] [f (x) G] = f (x + x) f (x) = f

but avoids losing numerical precision.

The final results obtained from the Microvolt simulation of the Metzger
heterojunction are listed in Table 4.2. The current density and voltage are a bit
higher than the results from Metzger [129], but the numerical method is stable
and the behaviour is consistent: lower p0 leads to higher Jsc and lower Voc .

114
Figure 4.19: Printout of the Jacobian matrix for the continuity equations 4.21
and 4.22, for a mesh of 101 points (resulting in a 202⇥202 matrix). The bottom
half represents the electron continuity derivatives for all positions, the top half
represents the hole continuity derivatives for all positions. Note the large gap
in the centre full of zeroes, making the Jacobian ill-posed for the solution of the
continuity equations.

115
4.3.3 Microvolt Setup for TiO2 /CdSe Core-shell Nanowires
Microvolt simulations were run for the TiO2 /CdSe core-shell nanowires. Peri-
odic boundary conditions were enforced, and the nanowires were embedded in
MEH-PPV polymer (since this is a typical hole-conducting polymer used in
solar cells and its charge transport properties are well known). Due to numer-
ical instabilities, 3D simulations or 2D radial simulations were impossible to
run. Therefore, the simulations were run in 1D after collapsing the 3D genera-
tion rate profile into 1D (see section 5.2.2.2). The material properties of TiO2 ,
CdSe and MEH-PPV were obtained from the literature and are listed in Table
4.3. The generation rate used in the Microvolt simulation was calculated from
Lumerical FDTD Solutions [120].

Properties TiO2 CdSe Polymer (MEH-PPV)

Eg (eV) 3.2 [177] 1.74 [96] 2.3 [28]
(eV) 4.1 [116] 4.53 [20] 3.2 [114]
µe (cm2 V 1 s 1 ) 0.0524 [38] 700 [124] 0.42
µh (cm2 V 1 s 1 ) 0.0524 40 [124] 0.42 [6]
Doping (cm 3 ) Nd : 1017 cm 3 [142] 0 0
"/"0 86 [72] 10 [72] 3 [186]
me /m0 10 [73] 0.13 [8] -
mh /m0 1 0.44 [8] -
Nc (cm 3 ) - - 5.3 ⇥ 1025 [161]
Nv (cm 3 ) - - 4.2 ⇥ 1025 [161]

Table 4.3: Charge transport properties of TiO2 , CdSe and the polymer MEH-
PPV. The effective masses aren’t defined in the polymer, but these can be
calculated from the effective densities of states using equation 4.6.

116
Chapter 5

Results

5.1 Optical Tests

The coated nanowires and planar substrate form the main components of the
TiO2 /CdSe solar cells under study in this thesis. To study the absorption
spectra of arrays of nanowires illuminated from above, 3D numerical methods
are required (such as FDTD). In order to be confident of the FDTD method,
the 2D results were compared to analytical Mie Theory. Coated nanowires
were modelled in 2D as infinitely long concentric cylinders, for which there
exist Mie Theory solutions. For the FDTD simulations, isolated nanowires were
illuminated from the side, as in Figure 3.13a. The results show how the core
radius and coating thickness effect the absorption spectra, providing insight into
the results of the 3D simulation.
Apart from the nanowires, the main component in the TiO2 /CdSe cell is
the CdSe-coated TiO2 substrate beneath the nanowires. FDTD simulations of
planar slabs in 1D were carried out, to see what effect the TiO2 has on the
absorption in the CdSe, and to see what difference the cylindrical geometry of
the nanowires has on the absorption spectra. These planar simulations were
verified against the Transfer-Matrix method.

5.1.1 Coated Cylinders

Figure 5.1 shows the Mie absorption efficiency calculated by FDTD simulation
and Mie theory for a CdSe coated TiO2 cylinder. This is defined as Qabs =
abs /2b, where abs is the absorption cross-section and b is the total radius
of the coated cylinder. As can be seen, the simulation results match up well
with Mie theory, validating the FDTD method (at least in 2D). The peaks in
absorption are faithfully reproduced.

117
 � �
��������� ����������
��� ������
��� ���

��� ���
����

��� ���

��� ���

� �
��� ��� ��� ��� ��� ��� ��� ��� ���
���������� ����

Figure 5.1: Mie absorption efficiency Qabs for an infinite coated cylinder with
core radius a = 200nm and coating thickness 20nm. Red crosses are results
from the Lumerical simulation, the black line represents results from Mie theory
calculations.

5.1.1.1 Varying Core Radius

In Figure 5.2 the absorption efficiency has been plotted for TiO2 cylinders with
varying radii (50nm, 80nm, 100nm) surrounded with a 20nm CdSe coating.
As the core radius increases, the number of resonances increase, due to a lar-
ger number of modes fitting inside the cylinder. The peaks in absorption are
weighted by the imaginary part of the refractive index of CdSe. The sudden
drop in absorption at about 720nm is due to the imaginary part of the refractive
index of CdSe dropping to zero.

5.1.1.2 Varying Coating Thickness

Figure 5.3 shows the absorption efficiency for varying thicknesses of the CdSe
coating (10nm, 20nm and 30nm). As the coating thickness increases the peaks
broaden and get red-shifted. The red-shift is due to the larger total radius,
whilst the broadening is due to the dampening of the resonance from having
two materials of different refractive index. Overall absorption increases with
coating thickness, with the exception of the near infra-red wavelengths.

5.1.2 Planar Slabs

Results were gathered for planar slabs of pure CdSe and TiO2 /CdSe, to see
what effect the TiO2 medium has on the absorption in the CdSe. Theoretical

118
 � �
� � ����� �� � ����
� � ����� �� � ����
� � ������ �� � ����

���������� ���������� �� ����� �

��� ����� � ���� ���

��� ���
����

��� ���

��� ���

� �
��� ��� ��� ��� ��� ��� ��� ��� ���
���������� ����

Figure 5.2: Absorption efficiency Qabs against wavelength for a TiO2 /CdSe
cylinder with a coating thickness of 20nm. Mie theory results are coloured
lines, FDTD results are coloured circles. Purple is a coated cylinder with core
radius a = 50nm, teal 80nm and blue 100nm. The dashed black line represent
the imaginary part of the refractive index of CdSe.

119
 � �
� � ����� �� � ����
� � ����� �� � ����
� � ����� �� � ����

���������� ���������� �� ����� �

��� ����� � ���� ���

��� ���
����

��� ���

��� ���

� �
��� ��� ��� ��� ��� ��� ��� ��� ���
���������� ����

Figure 5.3: Absorption efficiency Qabs against wavelength for a TiO2 /CdSe
cylinder with a core radius of 80nm. Mie theory results are coloured lines,
FDTD results are coloured circles. Orange represents a coating thickness of
h = 10nm, teal 20nm and yellow 30nm.

results are calculated using the Transfer Matrix method. Simulation results are
gathered from Lumerical FDTD Solutions.

5.1.2.1 Pure CdSe Slabs

The absorbance of slabs of pure CdSe are plotted against wavelength for various
thicknesses in Figure 5.4. The absorbance saturates for most wavelengths after
a threshold thickness of 20nm, but carries on increasing slightly in the violet to
ultra-violet range.

5.1.2.2 TiO2 /CdSe Slab

The absorption for CdSe coated TiO2 slabs is plotted against wavelength in Fig-
ure 5.5 for constant CdSe thickness and Figure 5.6 for constant TiO2 thickness.
As the thickness of the TiO2 increases from 100nm to 200nm, more absorption
peaks occur due to a larger number of resonances fitting inside the combined
slab. In contrast to CdSe on its own (Figure 5.4), increasing the thickness of the
CdSe beyond 20nm in the TiO2 /CdSe slab increases the absorption in CdSe.

5.1.2.3 Nanowires vs Planar Solar Cells

Results were obtained that justify the usage of nanowires instead of planar slabs
in the solar cell. By stripping the CdSe coating off the nanowires and making a

120
 ���
� � ����
���� � � ����
� � ����
��� � � ����

����

���
�

����

���

����

�����
��� ��� ��� ��� ��� ��� ��� ��� ���
���������� ����

Figure 5.4: Absorbance against wavelength for planar slabs of pure CdSe. The-
oretical results are solid lines, FDTD results are crosses.

���
�� � ����� �� � �����
���� �� � ����� �� � �����
�� � ����� �� � �����
���
����
���
����
�

���
����
���
����
�
�����
��� ��� ��� ��� ��� ��� ��� ��� ���
���������� ����

Figure 5.5: Absorption against wavelength for a TiO2 slab with a 20nm thick
CdSe coating. Lines are theoretical results, crosses are FDTD results. Purple
represents a TiO2 thickness of 100nm, teal 160nm and blue 200nm.

121
 ���
�� � ����� �� � �����
�� � ����� �� � �����
��� �� � ����� �� � �����
�� � ����� �� � �����
���

���
�

���

���

����
��� ��� ��� ��� ��� ��� ��� ��� ���
���������� ����

Figure 5.6: Absorption against wavelength for a 160nm thick TiO2 slab coated
with various thicknesses of CdSe. Lines are theoretical results, crosses are FDTD
results. Orange represents a coating of 10nm, purple 20nm, teal 30nm and blue
40nm.

planar slab with the same volume of CdSe, one will get less absorption of light.
For an array of 1000nm long nanowires with core radius a = 100nm and coating
thickness h = 20nm, a planar slab of CdSe with the same volume would be
60nm thick. Figure 5.7 shows the absorption for a 60nm slab is significantly less
than in the nanowire case, and that a 300nm thick planar slab is required for
comparable absorption. This shows that the optical path length is significantly
enhanced when the CdSe is distributed in a coated-nanowire geometry.

5.1.3 Rough Coating vs Smooth Coating

3D FDTD simulations were carried out of light incident on nanowires from above
to check whether an ideal smooth coating can accurately model a nanowire with
a more realistic rough coating (see Figure 5.8). The rough coating was generated
by creating a matrix of random values in k space, applying a gaussian filter,
Fourier transforming back to real space then wrapping around a cylindrical
surface. The mean coating thickness was 20nm and the roughness amplitude
was set at 5nm, which reflects the typical size of CdSe nanoparticles [105].
As Figure 5.9 shows, the results for rough and smooth coatings are sim-
ilar, showing that coated nanowires can be represented faithfully with smooth
coatings.

122
 ��� ���
���� ���� ����
����� ���� ����
��������� ���������� �������� �������
�� ��
���������� �

�� ��

�� ��

�� ��

� �
��� ��� ��� ��� ��� ��� ��� ��� ���
���������� ����

Figure 5.7: Absorption for a planar CdSe slab (purple 60nm and teal 300nm) as
well as a TiO2 /CdSe nanowire cell with core radius 100nm and coating thickness
20nm (blue dashed line).

Figure 5.8: 2D cross-sectional view of the TiO2 nanowire (light blue) with a
rough CdSe coating (dark blue). This coating was procedurally generated using
a random number generator and gaussian filter. The default coating thickness
was set at 20nm and the maximum amplitude of the roughness was 5nm.

123
 ��������� ��������� ���� ������ ������ ������� ��������� ����
� ������ �������
����� �������
���
����������

���

���

���

�
��� ��� ��� ��� ��� ��� ��� ��� ���
����������� ��

Figure 5.9: Absorption spectra for nanowires with a smooth coating (solid purple
line) and rough coating (dashed teal line).

5.2 Core-shell Nanowires and the effect of Geo-

metry on Absorption
The main device studied in this thesis is the TiO2 /CdSe nanowire array, as
described in section 3.3. Each nanowire consists of a TiO2 electron-conducting
core surrounded by a CdSe extremely thin absorber. The array of nanowires is
embedded in a hole-conducting transparent polymer and is sandwiched between
gold and TCO (transparent conducting oxide) electrodes. A diagram of this
device is shown in Figure 1.5. The following sections detail the results obtained
from optical and charge transport simulations.

5.2.1 Optical Results

FDTD simulations were performed on the TiO2 /CdSe nanowire array, using
Lumerical FDTD solutions. Results were obtained for the total absorption as
a function of geometry (section 5.2.1.1), which would allow one to optimise the
solar cell. Absorption spectra for different nanowire parameters were calculated
(section 5.2.1.2), to see what effect the geometry has on absorption. Electromag-
netic field profiles were plotted in the time and frequency domains to understand
how light behaves in the solar cell (sections 5.2.1.3 and 5.2.1.4).

124
 (a)
(c)
h a

2a

ξa
h
2ξa
(b)

ξh a

2a

Figure 5.10: The size of the unit cell (dashed line) in the periodic FDTD simu-
lations scales with the size of the core. The radius of the core is a, the width of
the unit cell is 4a. Increasing the coating thickness h by a factor of ⇠ does not
change the unit cell. Increasing the core radius a by a factor of ⇠ will increase
the width of the unit cell by the same factor. The core filling ratio fC is such
that the unit cell width is twice the core diameter, i.e. fC = ⇡a2 /(4a)2 = ⇡/16.

5.2.1.1 Light Absorption vs Geometry

By performing multiple FDTD simulations with a range of geometrical para-
meters, one can optimise the light absorption of the solar cell with respect
to nanowire arrangement, core radius and shell thickness. Figure 5.11 shows
the maximally achievable photocurrent density for a periodic square lattice of
nanowires with different core and shell thicknesses. This is given by
e
ˆ
JMAPD = ( )↵( ) d (5.1)
hc
where e is the electron charge, h is Planck’s constant, c is the speed of light, is
wavelength, ( ) is the ground-level solar intensity in Wm 2 nm 1 , and ↵( ) is
the absorption. This calculation assumes perfect quantum efficiency, i.e. a one-
to-one ratio of photons absorbed to charge carriers collected by the electrodes.
The filling ratio, fC , defined as the ratio of the nanowire core cross-sectional
area to the base area of the unit cell, is kept constant at a value of ⇡/16 = 0.196
(see Figure 5.10). This means that the size of the unit cell scales with the size
of the core but does not change with the thickness of the shell.
As might be expected, the nanowires with the thinnest cores and thickest
shells absorb the most light, since the density of absorbing material is greatest.
However, the medium core (a = 150 nm) yields the same absorption as the thin
core (a = 100 nm) despite having a lower density of absorber material. This

125
Figure 5.11: Colormap of photocurrent density for a periodic square lattice of
nanowires with different core radii and shell thicknesses. Colours represent the
magnitude of the photocurrent (red is high, blue is low). The geometry of one
nanowire is shown pictorially for each data point as a ring whose size scales
with core radius and width scales with shell thickness. The unit cell is twice the
diameter of the nanowire core, leading to a filling ratio of fC = ⇡a2 /(4a)2 =
⇡/16 = 0.196

is due to the medium core supporting more waveguide modes than the thinner
core. Hence, one can use less CdSe for medium core nanowires than thin cores
and achieve the same amount of absorption.
As shown in Figure 3.14, results were gathered for different lattice arrange-
ments of nanowires. Figure 5.12 shows the change in photocurrent density,
relative to the ordered square lattice case, for these arrangements of nanowires
while keeping the filling ratio (fC = 0.196) fixed. For the ordered hexagonal
arrangement, the results show enhancement for the thick cores (between 5% and
12%) but otherwise the differences are negligible (about ±1%). For the random
nanowire simulations, a periodic supercell of 25 nanowires was generated, and
results averaged over 3 simulations with different random number seeds. The
results show that random nanowires are only advantageous for thin cores, where
the enhancement goes up to 10%. Otherwise it is disadvantageous (up to 18%
lower).
Similarly to the fully random case, the semi-random square lattice enhances
the absorption for thin cores by up to 8% and is worse for medium cores by
up to 5%. The thick cores show a very small improvement of up to 3%. The
semi-random hexagonal lattice shows similar results except that for thin cores
the enhancement is up to 3% and for the thick cores it is up to 7%.

126
 Semi-random Ordered
Square Hexagonal

Semi-random Random
Hexagonal

Figure 5.12: Enhancement of photocurrent density, Jsquare

J
, in different nanowire
arrangements when compared to the ordered square lattice results from Figure
5.11. Red represents greater photocurrent, while blue represents less photocur-
rent. Each ring is split into four quadrants representing the enhancement of
each type of nanowire array: top-left (semi-random square), bottom-left (semi-
random hexagonal), bottom-right (random) and top-right (ordered hexagonal).

5.2.1.2 Analysis of Spectra

In order to understand why certain nanowire characteristics lead to better
photovoltaic performance, the absorption as a function of wavelength was ana-
lysed. Figure 5.13 shows the absorption spectra for selected nanowire geomet-
ries: ordered square lattice arrays, random arrays and single isolated nanowires.
Resonances in the absorption spectra match up between ordered, random and
single nanowires at wavelengths 490nm, 635nm and 695nm for the large cores
(a = 200 nm). These peaks are therefore due to the geometry of individual
nanowires rather than their arrangement. For the ordered nanowire arrays, sec-
ondary peaks occur due to periodic effects which are not present for random
nanowires, particularly at wavelengths 550nm, 560nm, 590nm, and 675nm.
By comparing results for isolated nanowires of different thicknesses (Figure
5.13b), it is evident that the wavelengths of absorption resonances are determ-
ined strongly by the core radius. There are two prominent resonances that occur
for the isolated nanowires, on the blue and red sides of the spectrum (e.g. for
a = 200 nm these occur at 490 nm and 740 nm). Table 5.1 shows that there
is an almost constant ratio between the wavelengths of these absorption peaks
and the core radius or total radius (core plus shell). These are due to specific
wavelengths exciting waveguide modes in the nanowire, as shown in sections
5.2.1.3 and 5.2.1.4.

127
 a) 100
Arrays of nanowires Ordered, a = 200nm
80 Random, a = 200nm
Absorption %
60

40

20

0
b) 5
Isolated nanowires a = 200nm
4 190nm
180nm
3 170nm
Qabs

160nm
2

0
400 450 500 550 600 650 700 750 800
Wavelength, nm

Figure 5.13: Absorption spectra for different nanowire arrangements with a

shell thickness of 30 nm. In (a), the total absorption percentage is calculated
for the ordered and random arrays of nanowires. In (b), the Mie absorption
efficiency Qabs is calculated for the isolated nanowires. The black lines in (a)
and (b) show the absorption for the case where the core radius a = 200 nm.
One can see the major peaks lining up between the isolated nanowires, ordered
arrays and random arrays. The coloured lines show the absorption for other
core thicknesses.

Core radius, a (nm) 160 170 180 190 200

2a/ 1 0.500 0.507 0.522 0.528 0.541
2a/ 2 0.780 0.773 0.800 0.800 0.816
Total radius, b (nm) 190 200 210 220 230
2b/ 1 0.589 0.597 0.609 0.611 0.622
2b/ 2 0.927 0.909 0.933 0.926 0.939

Table 5.1: Ratios of core (total) diameter, 2a (2b), to peak wavelength (of the
peaks in Figure 5.13b). 1 is the wavelength for the right peak, and 2 is the
wavelength for the left peak.

128
Figure 5.14: Heat maps of absorption per unit volume in one nanowire of a
periodic array with core radius 200nm and shell thickness 30 nm. Top shows
absorption averaged over the axial dimension, bottom shows the absorption
averaged over the z dimension. Absorption per unit volume for three different
wavelengths is shown: 400nm, 490nm and 550nm. These wavelengths were
chosen because the largest absorption peak in Figure 5.13a is at 490nm, while
400nm and 550nm lie on either side of the peak.

5.2.1.3 Field Profiles: Frequency Domain

Figure 5.14 shows the absorption per unit volume for thick nanowires (a =
200nm) in an ordered periodic array at three different wavelengths: 400 nm, 490
nm and 550 nm. Hotspots appear at 490 nm, as this is where the peak in ab-
sorption occurs (see Figure 5.13a), whereas 400 nm and 550 nm are wavelengths
either side of the peak. This suggests that the 490 nm waves can maintain a
large amplitude at the surface, which suggests some compatibility between the
nanowire radius and the wavelength of the light. The exact nature of this com-
patibility is not entirely clear, but some insight can be gained by considering the
time domain results in section 5.2.1.4, and observing how the scattered waves
evolve.

129
Figure 5.15: The electric field intensity, |E|2 , of the light propagating through
the nanowire and subsequently reflecting back up. This is a cross-sectional
view. Four snapshots in time are taken. The colour scales range from red for
high intensity to blue for low intensity.

5.2.1.4 Field Profiles: Time Domain

Figure 5.15 shows the intensity of the electric field at different time snapshots.
This demonstrates both the scattering and wave-guide effects of the nanowire on
the electromagnetic field. The incident field is an x polarised pulse of planewave
light with most power in the 400nm to 800nm band. As the light enters the
nanowire, the change of refractive index causes it to refract from the corners,
where it acquires a z-component and subsequently undergoes multiple internal
reflections. It is important to note that the top interface of the nanowire has a
strong effect on the scattering and absorption of light. This means that analytic
theories developed for infinitely long dielectric cylinders (i.e. fibre optic cables)
cannot be used straightforwardly to calculate the amount of light absorbed in
the nanowire. However, from Figure 5.15 we observe that certain wavelengths
are strongly guided inside the nanowire, thus defining wavelengths at which the
intensity is greatest in the shell and therefore achieving the highest absorption.

5.2.2 Coupled Optical / Charge Transport Results

FDTD simulations were carried out of CdSe-coated TiO2 nanowires on TiO2
substrates. The absorption profiles obtained from these simulations were integ-
rated over the solar spectrum using MATLAB to get the generation rate. The
resulting generation rate profiles were coupled with charge transport simulations
using the Microvolt C++ code. Results were obtained for photocurrent density
of different nanowire geometries. Due to numerical instabilities in 2D and 3D,
Microvolt simulations were carried out in 1D.

130
 �������� ������� �������� �������
��� �������� ��� ��������

��� ���

������ ������
��� ���

� ������ � ������
� ���������

� ���������
��� ���

������ ������
��� ���

��� ������ ��� ������

��� ���

������ ������
� �

���� � ���� �
���� ���� ���� � ��� ��� ��� ���� ���� ���� ���� ���� � ��� ��� ��� ��� ���
� ��������� � ���������

Figure 5.16: 2D generation rate profiles of TiO2 /CdSe Core-shell nanowires.

The generation rate is measured in free electron-hole pairs per second per cm3 .
On the left the nanowire has core radius of 100nm and coating thickness of 20nm.
On the right the core radius is 200nm and coating thickness is 30nm. The heat
map ranges in colour from black (zero generation rate) to yellow (maximum
generation rate).

5.2.2.1 Generation Rate

Figure 5.16 presents the generation rate profiles obtained from Lumerical and
MATLAB calculations in 3D. The nanowires have radii of 100nm and 200nm,
a height of 1µm and coating thicknesses of 20nm and 30nm respectively. The
TiO2 substrate is 0.2µm thick and the distance from the top of the nanowires
to the electrode is also 0.2µm. The highest generation rate is 1.2 ⇥ 1022 s 1 cm 3
and this occurs near the top of the nanowire. The lowest generation rate occurs
at the bottom of the nanowire at about 4 ⇥ 1021 s 1 cm 3 .

5.2.2.2 Conversion to 1D
Due to numerical instabilities, only 1D simulations could be performed. The
nanowires in the model have to be collapsed into planar slabs. To do this,

131
first the spatial generation rates from the 3D simulations are approximated by
uniform generation rates over the shells,
Vreg
Gshell = Gavg
Vshell
where Vreg is the volume of the simulation region, Vshell is the volume of the
nanowire shell and Gavg is the average generation rate over the whole simulation
region. The simulation region is a cuboid with the volume Vreg = p2 (L + h)
where p is the lattice constant, L is the length of the nanowire and h is the shell
thickness. The shell covers the substrate (not included in the simulation region)
and the core cylinder. Its volume is Vshell = ⇡L(b2 a2 ) + p2 h where a is the
core radius and b = a + h is the total radius. To get the 1D generation rate,
Gshell is then multiplied by a factor nopt corresponding to the extra volume a
nanowire shell takes up compared to a planar slab in the same unit cell,

G1D = nopt Gshell

Vshell
= Gshell
Vslab
⇡L(b2 a2 ) + p2 (b a)
= Gshell
p2 (b a)
✓ ◆
⇡L(b + a)
= + 1 Gshell
p2
The TiO2 and polymer thicknesses were both set to 0.7µm. This is half the
length of the nanowire plus the initial substrate thickness (0.5µm + 0.2µm =
0.7µm).

5.2.2.3 Energy Levels and Charge Densities

Figure 5.17 shows the energy levels for a 1D silicon p-n diode for reference. Note
the large changes in carrier density across the p-n junction (almost 10 orders
of magnitude). However, the changes in carrier density are relatively smooth
and the numerical method for solving this system is stable. The conduction and
valence bands change smoothly.
Figure 5.18 shows the energy levels for the TiO2 /CdSe/polymer device in
1D. The changes in carrier density are much higher, at about 30 orders of
magnitude (note that double precision floating point numbers can only have up
to 15 significant digits). In addition, the bandgaps are large and the band edges
have large step changes over a short distance. Without implementing the fixes
in section 4.3.2.4, the numerical method would be unstable for this problem,
even in 1D.

5.2.2.4 J-V Curves

Figure 5.19 shows the J-V curves obtained for the TiO2 /CdSe/polymer device
for different radii a and coating thicknesses h. For nanowires with core radii

132
Figure 5.17: Plot of energy levels, charge densities and currents across the 1D
Silicon diode. The p-n junction is located in the centre at x = 0.5µm.

133
Figure 5.18: Plot of energy levels, charge densities and currents across the
TiO2 /CdSe/polymer device, in 1D.

134
 ��

��

������� ������� ��� �����

��

��

�
�������� ������
� �������� ������
�������� ������
�������� ������
� �������� ������
�������� ������
�
� ��� ��� ��� ��� ��� ���
������� ���

Figure 5.19: Current density J plotted against Voltage V for different nanowire
parameters.

of 100nm, increasing the coating thickness from 10nm to 30nm increased Jsc
from 10.1 mA cm 2 to 15.1 mA cm 2 , decreased Voc from 0.58V to 0.56V and
decreased the fill-factor from 74% to 61%. For a core radius of 200nm, Jsc
increased from 6.3 mA cm 2 to 10 mA cm 2 , Voc decreased from 0.58V to
0.56V and the fill-factor decreased from 77% to 64%.

5.2.2.5 Maximum Power Yield

Figure 5.20 shows the power yield obtained for different coating thicknesses and
core radii. The highest power yield is for the thinnest core and thickest coating,
a = 100nm and h = 30nm. This nanowire geometry yields a maximum power
density of 52 W m 2 and an efficiency of 5.3%. This is theoretically enough to
power 5 energy efficient light bulbs with one square metre of solar cells. The
lowest power achieved (28 W m 2 ) is for the nanowires with the largest core
and thinnest coating, a = 200nm and h = 10nm.

5.3 Nanoparticle Nanowires

Many nanowire solar cells are sensitised using nanoparticles to improve their op-
tical absorption. Semiconductor absorbers often grow in nano-crystalline forms,
and it is not until they are annealed that the coalesce into a conformal shell.
Nanoparticles have a higher absorption per unit volume than conformal coat-
ings. It is for these reasons that nanoparticle coatings are studied in this thesis.

5.3.1 Absorption Spectra

Figure 5.21 shows the absorption spectra for the nanoparticle coated nanowires
calculated by FDTD simulations. The absorption decays with wavelength, mir-

135
Figure 5.20: Maximum power yield for core radii of 100, 150 and 200nm and
coating thicknesses of 10, 20 and 30nm. The thinnest nanowires with the thickest
coatings have the largest power output.

roring the imaginary part of the dielectric function of PbS. This shows that the
nanoparticles do not appear to have any wavelength dependent effect on the
absorption apart from that of the dielectric function. This is due to the fact
that the nanoparticles are much smaller than the wavelength of light.
There is a clear increase in absorption with the size of the nanoparticles, as
shown in Figure 5.22a. At first the absorption, A, increases linearly for small
sizes of nanoparticles, but then increases sub-linearly for larger sizes, suggesting
an exponential decay of 1 A with volume of the nanoparticles. Indeed, as
Figure 5.22b shows, this is the case.
As the number of nanoparticles, N , increases, the absorption increases at a
slower and slower rate (see Figure 5.23a). Figure 5.23b shows the product of
the absorption and solar intensity integrated over all wavelengths, which results
in the maximally achievable photocurrent density, JMAPD .

5.3.2 Absorption vs Height

Figure 5.24 shows the absorption profile along the length of the nanowire. The
absorption is slightly higher near the top of the nanowire, suggesting a shadow-
ing effect (nanoparticles at the top prevent light being absorbed lower down).

136
 ���
��� �� ��������� ��� ���
��� �� ��������� ��� ���
��� ��� �� ��������� ��� ���
��� �� ��������� ��� ���
��� �� ��������� ��� ���
��� ��� �� ��������� ��� ���
��� �� ��������� ��� ���
����������

��� �� ��������� ��� ���

��� �� ��������� ��� ���
��� ��� �� ��������� ��� ���

���

���

�
��� ��� ��� ��� ��� ���
����������� ��

Figure 5.21: Absorption spectra for nanoparticle coated nanowires. The thick
lines represent the nanowires with 500 nanoparticles. The dashed lines represent
100 nanoparticles. Different colours represent different nanoparticle diameters.

(a) (b)

Figure 5.22: (a) Absorption at = 400nm against the cube of nanoparticle

radius for nanowires coated with 1000 nanoparticles. The lattice spacing is
192nm. Results for three different nanowire radii are shown. The radius of the
nanoparticle is r. (b) Logarithmic plots of 1 A vs size of nanoparticle, showing
that 1 A decays exponentially with r3 .

137
 (a) (b)

Figure 5.23: (a) Plot of absorption at = 400nm vs number of nanoparticles.

(b) JMAPD vs number of nanoparticles

#10 -7Generation Rate of Kaust Nanowires (a=25nm, th=9nm N=1000) vs height

6

2
z (m)

-2

-4

-6
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Average Generation Rate (/m3/s) #10 27

Figure 5.24: Average absorption (m 3 s 1 ) against height of the nanowire. The

absorption is fairly even over the length of the coating. The orange line shows
a moving average. The red dashed line is the line of best fit.

138
 ���
������ ���� ������
��� ������� ���� ������
������ ���� ������
������� ���� ������
��� ������ ���� ������
������� ���� ������
���������� ���

���

���

���

���

�
��� ��� ��� ��� ��� ��� ��� ��� ���
����������� ��

Figure 5.25: Absorption spectra for nanowires coated with randomised nano-
particles (crosses) and ordered nanoparticles (lines). The nanowire spacing is
120nm and the nanoparticle diameters are 9nm. Results for three nanowire
thicknesses are shown with core radii of 25nm (purple), 40nm (blue) and 50nm
(orange).

5.3.3 Random vs Ordered Nanoparticles

Figure 5.25 compares the absorption spectra of nanowires with randomised nan-
oparticles and ordered nanoparticles. Results are shown for various radii with
lattice constant p = 120nm and nanoparticle diameter 9nm. There is little dif-
ference in total absorption whether the nanoparticles are ordered or randomised,
and the benefits vary with nanoparticle size: random arrangement is beneficial
for nanoparticles of size 40 nm, but ordered is beneficial otherwise.

5.3.4 Total Absorption

Figure 5.26 shows the maximally achievable photo current density, JMAPD , for
different nanowire radii, nanoparticle diameter and numbers of nanoparticles,
with lattice constant p = 192nm. This is calculated by using the same expression
as equation 5.1. JMAPD was calculated using FDTD and integrated with the
solar spectrum, as before.
Figure 5.27 shows the same JMAPD results but for two other lattice spacings:
p = 120nm and p = 240nm. The behaviour is the same, but the overall photo-
current density is scaled up by the increase in nanowire (and coating) density.
So comparing the results of Figure 5.27a with 5.27b, the density is higher by a
factor of 4 (since the radius is halved) so the photocurrent is approximately 4
times greater. Therefore, there is not much dependence on the wavelength or
lattice spacing.
Figure 5.28 shows the JMAPD per unit volume of coating material. This
shows that the nanowires with the lowest numbers of nanoparticles have the
highest absorption per unit volume. In addition, the nanowires with larger radii

139
Figure 5.26: Plot of maximally achievable photo current density, JMAPD , for dif-
ferent nanowire radii, nanoparticle diameters and numbers of nanoparticles with
a lattice constant of 192nm. The nanoparticles are arranged on the nanowires
randomly. Ordered nanoparticles give similar results. Each pictorial element
represents a single value of nanowire radius and nanoparticle diameter, but is
split into 3 sections for three numbers of nanoparticles: 100, 500 and 1000
nanoparticles.

(a) (b)

Figure 5.27: Same as Figure 5.26 but for two other lattice constants, (a) 120nm
and (b) 240nm.

140
 Figure 5.28: JMAPD per unit volume for different nanowire parameters.

have higher absorption density. This is probably due to a waveguide effect, as

these nanowires that take up more of the unit cell will guide more light than
those that take up less of the unit cell.

5.3.5 Absorption per Unit Volume, 

We can calculate the absorption per unit volume of one nanoparticle, . The
relation between  and the total absorption, A, is derived in Appendix A as
ND VD
A=1 e
where ND is the number of nanoparticles per nanowire and VD = 43 ⇡r3 is the
volume of a nanoparticle. We find  by plotting log(1 A) against r3 and finding
the gradient of the line of best fit, , such that log(1 A) = r3 where
3
= .
4⇡ND
Figure 5.29 shows the dependence of  against wavelength for nanowires
with different sets of parameters. There are ripples in the  curves for N = 100,
where the nanoparticles are more sparse. This could be due to wavelength
interference effects. As the number of nanoparticles increases, the wavelength

141
Figure 5.29: Plot of absorption per unit volume, , against wavelength for
a variety of different numbers of nanoparticles and radii of nanowires.  is
calculated by gathering results for nanoparticles ranging from 4nm to 9nm in
diameter.

interference cancels out and  follows the same behaviour as the base material
extinction coefficient (shown in Figure 3.17). As can be seen, each nanoparticle’s
ability to absorb light decreases as more nanoparticles are added, as photons
absorbed in one nanoparticle are not absorbed in another nanoparticle. This can
be seen in Figure 5.30 where  curves have been plotted against the number of
nanoparticles. One can see that the effectiveness of each nanoparticle decreases
as they are packed closer together ( decreases with N ).
Non-trivially,  is higher for the nanowires with larger radii, even though the
overall density of nanoparticles has not changed. This is due to the fact that
light captured in the nanowire is directed to the edge (due to internal reflection)
where it gets absorbed by the coating. If the radius is small compared to the
unit cell, then less light will enter the nanowire and therefore less will be directed
to the coating. Figure 5.31 shows that the optimum nanowire radius is 85nm
for a lattice spacing of 192nm.

5.3.6 Conformal Coating vs Nanoparticles

The solid lines in Figure 5.32 show the absorption spectra for conformal coatings
of different thicknesses. The absorption is higher than when using nanoparticles
and remains above 50% for most of the solar spectrum. However, as seen in
Figure 5.33 the absorption per unit volume is much lower. This may be due to
the enhanced scattering and absorption cross-sections of the nanoparticles. This

142
Figure 5.30: Plot of  against number of nanoparticles for lattice constant
p = 192nm and wavelength = 400nm. The blue curve is for nanowires with a
core radius of a = 25nm, and the green curve is for a = 50nm.

Figure 5.31: Plot of  against radius of nanowire for = 400nm, p = 192nm

and N = 1000. The maximum absorption per unit volume is at a = 60nm.

143
 � ��� ���
��� ��� ���
��� ���
��� ��� ���
��� ���
��� ��� �����
��� �����
����������

��� ��� �����

��� �����
��� ��� �����
���
���
���
���
�
��� ��� ��� ��� ��� ���
����������� ��

Figure 5.32: Absorption spectra of conformal coatings (solid lines) and nano-
particles (dashed lines). N = 500, a = 25nm, p = 120nm

shows that if economising on material use, it is better to sensitise the nanowires

with separated nanoparticles rather than a conformal layer.

5.4 Discussion of Results

5.4.1 Optical Properties of TiO2 /CdSe Nanowires
The results in section 5.2.1.1 show that the best solution to optimise the per-
formance of core-shell nanowires is to have a thin core, thick shell and a random
arrangement of nanowires.
Square or hexagonal lattice arrangements make a negligible difference in
photocurrent except for large radius nanowires (where the hexagonal lattice is up
to 16% better). Semi-random arrangements show little performance difference
(within ±8%). Random arrangements show the greatest change in performance
(+12% for thin cores and -18% for medium cores).
When comparing different core and shell thicknesses, the differences in pho-
tocurrent are much greater. For the square periodic arrangement, having a thick
core and a thin shell (10 nm) leads to a photocurrent of 60 mA/cm2 , where-as
thin and medium cores with a thick shell (30 nm) leads to a photocurrent of
150 mA/cm2 , a 150% enhancement.
For thick cores, the geometry (radius and shell thickness) correlates well
with the absorption spectra. Prominent peaks in absorption line up at specific
frequencies due to the excitation of waveguide modes. The peaks are inde-
pendent of nanowire arrangement (ordered, random, isolated), although they
are somewhat smoothed out for random nanowires. Secondary peaks appear in

144
 �������� ��� ���
������ ��� ���
��� ���
�������� ��� ���
��� ���
���������� ��� ���
������ ��� �����
��� �����
�������� ��� �����
������ ��� �����
��� �����
��������
������
��������
������
������
�
��� ��� ��� ��� ��� ���
����������� ��

Figure 5.33: Absorption per unit volume of conformal coatings (solid lines) and
nanoparticles (dashed lines).

the ordered nanowire absorption spectra due to periodic effects, but these are
too insignificant to contribute much to the total absorption. The field profiles
in section 5.2.1.3 show that waveguide modes are excited in the nanowires at
wavelengths corresponding to peaks of absorption.
The results in section 5.2.1.4 show that the planar interface at the top of
the nanowire is an important factor in the optical scattering. It is therefore
difficult to calculate the light absorption analytically. A simple theory of optical
waveguides [162] cannot be used in this case, since this assumes infinitely long
cylinders and therefore cannot treat the boundary conditions at the end of the
cylinder.
In conclusion, while differences in arrangement can increase or decrease op-
tical absorption, the geometry of individual nanowires (core radius and shell
thickness) is far more important. In addition, results show that one doesn’t
need to put effort into ensuring the nanowires are ordered, since this does not
enhance their ability to absorb light.

5.4.2 Charge Transport in TiO2 /CdSe Nanowires

In section 5.2.2 are the results of the coupled optical and charge transport simu-
lations. Optical absorption profiles obtained from Lumerical FDTD simulations
were integrated with the solar spectrum and then used as the input generation
rate in Microvolt. Due to numerical instabilities caused by the large variations
in charge density and material properties, simulations could only be run in 1D.
Therefore, the 3D generation profiles were collapsed into 1D. Results were ob-
tained for coating thicknesses of 10nm, 20nm and 30nm, as well as core radii of
100nm, 150nm and 200nm. The highest power yield was for the thinnest core

145
(100nm radius) and thickest coating (30nm) with a power density of 52 W m 2
and efficiency 5.3%. This is because of the higher packing density of nanowires
and the fact that the carrier diffusion length in CdSe is greater than its thick-
ness. Thicker coatings lead to a lower Voc and fill-factor due to the internal
resistance, but this is more than offset by the high Jsc .

5.4.3 Optical Properties of Nanoparticle Nanowires

In addition to the TiO2 /CdSe nanowires, optical absorption results were gathered
for ZnO nanowires coated with PbS nanoparticles. It was shown that the ab-
sorption spectra curves match the shape of the imaginary part of the dielectric
function of PbS. In other words, the nanoparticles (which had diameters from
4nm to 9nm) were too small to have any wavelength dependent effect on the
absorption.
The nanoparticles were found to have a high absorption per unit volume
when compared to a conformal coating. It was also determined that the ab-
sorption depends on the radius of the nanowire cores, in that it is higher for
tightly packed nanowires, where the core diameter is comparable to the unit cell.
This is due to the waveguide effect the nanowires have, as light that enters the
centre of the nanowire gets refracted towards the edges where it is subsequently
absorbed by the nanoparticles.

5.5 Overall Conclusions

Overall, it was determined that the positional arrangement of nanowires does
not matter as much as the individual nanowire geometries. Random or ordered
placements make little difference in the total absorption. Results from the simu-
lations show that power yield is maximised when densely packing thin nanowires
with thick coatings. For nanowires coated with nanoparticles, the arrangement
of the nanoparticles does not matter as much as the size and density of the nan-
oparticles, or the size of the nanowires. Nanoparticles have a high absorption
per unit volume but conformal shell coatings are preferred over nanoparticles
due to the much higher overall absorption.
There are limitations in the modelling due to the wide bandgaps and com-
plicated device physics. More work is need to study the charge transport in
metal-oxides and interfaces, and to determine the ultimate combination of ma-
terials and parameters for coated nanowire solar cells. This would be helped
by writing an open-source, multiphysics simulator that other researchers and
software engineers could contribute to.

146
Chapter 6

Conclusions

6.1 Summary of Work Done

The purpose of this thesis was to optimise nanostructured solar cells. We sur-
veyed the literature on coated nanowire solar cells, and determined what theor-
etical work needed doing. Then we ran simulations to understand and ultimately
optimise the photocurrent of such solar cells.
We used the Finite-Difference Time-Domain (FDTD) method for determ-
ining the optical properties of the nanowires, implemented using Lumerical
FDTD Solutions [120]. This method was tested and verified against analyt-
ical Mie Theory [178]. Charge transport in the nanowires was simulated using
a finite-difference discretisation of the drift-diffusion equations [158]. This was
implemented using the Microvolt C++ code [131] developed by Deinega et al
[35]. After fixing bugs in the code which caused numerical instabilities for wide
bandgap semiconductors, Microvolt was tested against results for heterojunction
solar cells in the literature [129]. Optical absorption profiles generated from the
FDTD method were input into the charge transport simulations for determining
the overall photocurrent.
The optical properties were determined for nanowires with a TiO2 core and
CdSe shell, as well as ZnO nanowires coated with PbS nanoparticles. Using the
FDTD method we determined the absorption profiles as a function of position,
absorption spectra as a function of wavelength and total absorption as a function
of geometrical parameters. For the TiO2 /CdSe nanowires, the core thickness
(radius) and coating thickness were varied. Different arrangements were tested
ranging from isolated nanowires to periodic or randomised arrays of nanowires.
For the ZnO/PbS nanowires, the core thickness (radius), number density of
nanoparticles and nanoparticle radius were all varied. Different types of coatings
were studied for the ZnO/PbS nanowires, including conformal coating (as in the
TiO2 /CdSe case), randomised position of nanoparticles and ordered positions
of nanoparticles. A model was constructed for relating the absorption per unit
volume in the ZnO/PbS nanowires to the number and radius of nanoparticles.

147
 Charge transport was studied in the TiO2 /CdSe nanowires. The generation
rate calculated from FDTD was input into the charge transport model. The aim
was to determine which combination of nanowire parameters (coating thickness
and core radius) lead to the the highest power efficiencies.
The absorption of light inside the CdSe layer on cylindrical and planar TiO2
substrates has been analysed. This has been achieved using Mie scattering the-
ory, equations of reflectance and transmittance in slabs, and FDTD simulations.
It is evident that the absorption in the CdSe layer is largely effected by the TiO2
and the geometry.

6.2 Complications and Future Work

6.2.1 Optical Properties
Due to the difficulties of obtaining dielectric fits for organic polymers, the poly-
mer in the solar cell was treated as a non-absorbing dielectric. Also, the TiO2
and ZnO were treated as non-absorbing dielectrics. In reality, they absorb light
in the UV range. However, there were difficulties of obtaining an accurate
FDTD fit of the dielectric function over the combined UV and visible range.
Simulations were restricted to the visible 400-800nm range. This problem could
be be overcome by performing two different optical simulations: one for visible
light where > 400nm (metal-oxides are transparent dielectrics) and one for
UV light where < 400nm (metal-oxides have complex dielectric functions and
non-zero absorption). Then these two simulations could be combined to obtain
the absorption over almost all of the solar spectrum.

6.2.1.1 Electronic Structure Calculations

We are assuming the bulk dielectric functions for the materials in the solar
cell. One may have to take into account the finite-width of materials to calcu-
late a more accurate dielectric function. For the extremely thin CdSe layer in
the TiO2 /CdSe solar cell, there exist no results in the literature for the dielec-
tric function of CdSe in this environment. To calculate the dielectric function,
ab-initio electronic structure calculations of the solar cell components (bulk ma-
terial and interfaces) could be done using Density-Functional Theory.
DFT is an efficient method (in principal) for solving the many-body Schrödinger
equation. Solving it directly cannot be done in polynomial time, as the number
of interactions between each electron grows factorially with the number of elec-
trons. DFT reformulates the Schrödinger equation such that each electron feels
an effective potential which is dependent on the electron density only, not the
position of each electron separately. In other words, the many-body problem is
mapped onto a mean-field problem. Time-dependent DFT extends the theory
to time-dependent potentials. This is needed for example to calculate the polar-
isability and hence the dielectric function. Using the CRYSTAL computational
tool [41], one could calculate the dielectric function as a function of position,
taking into account the proximity of TiO2 /CdSe and CdSe/polymer interfaces.

148
 In addition, the electronic structure calculations could provide the charge
carrier mobilities (for bulk and interface), surface recombination and charge
injection rates to input into the charge transport model.

6.2.2 Charge Transport

6.2.2.1 TiO2 Charge Transport
The TiO2 was treated as a crystalline semiconductor. In reality, the charge
transport is more complicated and occurs through polaronic hopping [73, 195].
Using Marcus Theory [125], one could construct a more accurate model of charge
transport through TiO2 . Polaron hopping rates in the bulk could in theory be
calculated by using DFT calculations as an input to Marcus Theory.

6.2.2.2 Interfacial Charge Transport

Charge transport occurs across the interfaces from the TiO2 to the CdSe and
from the CdSe to the polymer. The mechanisms for charge transport across
these interfaces are not well understood and could be investigated using DFT.

6.2.2.3 Organic Charge Transport

Microvolt was designed for crystalline semiconductors by default. However,
the Russian computational materials science lab, Kintech [95], was working on
a version of Microvolt that would support more advanced models for organic
semiconductors. So far they have implemented a gaussian density of states [5],
different organic mobility models where the mobility depends on the charge
density and field [18, 139], and boundary conditions which account for image
charge [156]. Ultimately, if we want to achieve more accurate results when
modelling the TiO2 /CdSe/polymer device, then we will need to implement these
organic charge-transport models.
While the organic version of Microvolt can treat organic/metal interfaces,
it does not tackle organic/inorganic interfaces. A review on organic/inorganic
interfacial charge-transport has been written by Lonergan [119]. One could use
this to help write some additional code in Microvolt to model such interfaces.

6.2.2.4 Numerical Problems

Due to complications in the charge transport modelling, only 1D simulations
were performed. Given more time, it should have been possible to determine
what were causing the numerical instabilities and represent the effect that the
geometry of the nanowire has on charge transport. For example, increasing
the nanowire length may increase the photocurrent due to increased optical
scattering, or decrease the photocurrent due to extra surface recombination.
This was investigated experimentally on ZnO/CdSe nanorods by Levy-Clement
et al [110].

149
6.2.3 Other Parameters to Test
There are other geometrical parameters remain to be tested for coated nanowire
solar cells: different lengths of nanowires, different core filling ratios, and com-
bining and randomising individual nanowire geometries in an array (such as
coating thickness and core radius, rather than just the position). So far I have
only looked at TiO2 /CdSe and ZnO/PbS nanowires. It would be useful to test
other combinations of materials and see whether one can optimise the pairing
of two different materials for the core and coating. This would be a non-trivial
problem as some material combinations are great optically, but not electric-
ally, or vice versa. One would have to perform coupled optical and electrical
simulations to determine the effectiveness of a specific pair of materials.

6.3 Software Design

For coupling optical and charge transport simulations and ultimately determ-
ining the power efficiency of solar cells, a combination of different pieces of
software were used. Lumerical for optical simulations, MATLAB for calculating
generation rate, Microvolt for charge transport simulations and a collection of
bash and python scripts for managing simulations and analysing data.
Lumerical is proprietary and requires a purchased license to use. Due to the
limited number of licences available for running simulations, one would often
have to wait for others to finish their simulations before you could run any. The
source code is closed and therefore it is difficult to figure out exactly how the
FDTD simulations are run.
Microvolt, while being open-source, is a heavily optimised C++ code. This
makes it very fast since C++ is a compiled language and is usually by default
faster than using an interpreted language (such as Python). However, code
written to make use of the fast C++ features is difficult to reverse engineer and
debug.
Ideally, one would use a single piece of software for carrying out the optical
and charge-transport simulations. Python is a good candidate language for
writing a new solar cell simulator. The syntax is easy to read, easy to debug
and its very well suited for object oriented programming. While Python is
an interpreted language as opposed to a compiled language, it has numerical
algorithm libraries in NumPy which call optimised routines written in C and
FORTRAN. To allow other people to use, scrutinise and help develop the code,
it should be open-source and hosted on a public repository. Hosting on github.
com would allow people to easily contribute to the code while implementing the
git version control system.

150
Bibliography

[1] Effect of Light on Selenium During the Passage of An Electric Current.

Nature, 7:303, February 1873.

[2] N Anttu and H Q Xu. Coupling of Light into Nanowire Arrays and
Subsequent Absorption. Journal of Nanoscience and Nanotechnology,
10(11):7183–7187, November 2010.
[3] Harry A. Atwater and Albert Polman. Plasmonics for improved photovol-
taic devices. Nat Mater, 9(3):205–213, 03 2010.
[4] U. Bach, D. Lupo, P. Comte, J. E. Moser, F. Weissortel, J. Salbeck,
H. Spreitzer, and M. Gratzel. Solid-state dye-sensitized mesoporous TiO2
solar cells with high photon-to-electron conversion efficiencies. Nature,
395(6702):583–585, 10 1998.

[5] H. Bässler. Charge Transport in Disordered Organic Photoconductors a

Monte Carlo Simulation Study. Physica Status Solidi B Basic Research,
175:15–56, January 1993.
[6] Heinz Bässler and Anna Köhler. Charge Transport in Organic Semicon-
ductors. pages 1–65. Springer Berlin Heidelberg, Berlin, Heidelberg, Oc-
tober 2011.
[7] Corsin Battaglia, Ching-Mei Hsu, Karin Söderström, Jordi Escarré, Franz-
Josef Haug, Mathieu Charrière, Mathieu Boccard, Matthieu Despeisse,
Duncan TL Alexander, and Marco Cantoni. Light trapping in solar cells:
can periodic beat random? ACS Nano, 6(3):2790–2797, 2012.

[8] M G Bawendi, A R Kortan, M L Steigerwald, and L E Brus. X-ray

structural characterization of larger CdSe semiconductor clusters. The
Journal of Chemical Physics, 91(11):7282, 1989.
[9] Rocío Bayón, Robinson Musembi, Abdelhak Belaidi, Marcus Bär, Tatjana
Guminskaya, Christian Herbert Fischer, Martha Christina Lux-Steiner,
and Thomas Dittrich. Highly structured TiO2 /In(OH)x Sy /PbS/PEDOT:
PSS to be used in photovoltaic applications. Comptes Rendus Chimie,
9(5):730–734, 2006.

151
[10] A. E. Becquerel. Mémoire sur les effets électriques produits sous l’influence
des rayons solaires. Comptes Rendus des Séances Hebdomadaires, 9:561–
567, 1839.
[11] A Belaidi, R Bayón, L Dloczik, K Ernst, M Ch Lux-Steiner, and R Könen-
kamp. Comparison of different thin film absorbers used in eta-solar cells.
Thin Solid Films, 431-432:488–491, May 2003.
[12] A. Belaidi, Th. Dittrich, D. Kieven, J. Tornow, K. Schwarzburg, M. Kunst,
N. Allsop, M.-Ch. Lux-Steiner, and S. Gavrilov. ZnO-nanorod arrays for
solar cells with extremely thin sulfidic absorber. Solar Energy Materials
and Solar Cells, 93(6–7):1033 – 1036, 2009. 17th International Photovol-
taic Science and Engineering Conference.
[13] Abdelhak Belaidi, Thomas Dittrich, David Kieven, Julian Tornow, Klaus
Schwarzburg, and Martha Lux-Steiner. Influence of the local absorber
layer thickness on the performance of ZnO nanorod solar cells. Physica
status solidi (RRL) - Rapid Research Letters, 2(4):172–174, August 2008.

[14] Jean-Pierre Berenger. A perfectly matched layer for the absorption of

electromagnetic waves. Journal of Computational Physics, 114(2):185 –
200, 1994.
[15] Matteo Biancardo and Frederik C Krebs. Microstructured extremely thin
absorber solar cells. Solar Energy Materials and Solar Cells, 91(18):1755–
1762, November 2007.
[16] C.F. Bohren and D.R. Huffman. Absorption and Scattering of Light by
Small Particles. Wiley Science Series. Wiley, 2008.
[17] Pablo P. Boix, Kazuteru Nonomura, Nripan Mathews, and Sub-
odh G. Mhaisalkar. Current progress and future perspectives for or-
ganic/inorganic perovskite solar cells. Materials Today, 17(1):16 – 23,
2014.
[18] M Bouhassoune, S L M van Mensfoort, P A Bobbert, and R Coehoorn.
Organic Electronics. Organic Electronics, 10(3):437–445, May 2009.

[19] J Briscoe and S Dunn. Extremely thin absorber solar cells based
on nanostructured semiconductors. Materials Science and Technology,
27(12):1741–1756, 2011.
[20] Fredrik Buch, Alan L Fahrenbruch, and Richard H Bube. Photovol-
taic properties of five II-VI heterojunctions. Journal of Applied Physics,
48(4):1596, 1977.
[21] Marc Burgelman and Catelijne Grasso. Network of flatband solar cells
as a model for solid-state nanostructured solar cells. Journal of Applied
Physics, 95(4):2020–2024, 2004.

152
[22] Julian Burschka, Amalie Dualeh, Florian Kessler, Etienne Baranoff, Ngoc-
Lê Cevey-Ha, Chenyi Yi, Mohammad K. Nazeeruddin, and Michael
Grätzel. Tris(2-(1H-pyrazol-1-yl)pyridine)cobalt(III) as p-Type Dopant
for Organic Semiconductors and Its Application in Highly Efficient Solid-
State Dye-Sensitized Solar Cells. Journal of the American Chemical So-
ciety, 133(45):18042–18045, 11 2011.
[23] Julian Burschka, Norman Pellet, Soo-Jin Moon, Robin Humphry-Baker,
Peng Gao, Mohammad K. Nazeeruddin, and Michael Graetzel. Sequential
deposition as a route to high-performance perovskite-sensitized solar cells.
Nature, 499(7458):316+, Jul 18 2013.
[24] Linyou Cao, Justin S White, Joon-Shik Park, Jon A Schuller, Bruce M
Clemens, and Mark L Brongersma. Engineering light absorption in semi-
conductor nanowire devices. Nature Materials, 8(8):643–647, July 2009.
[25] Xuebo Cao, Peng Chen, and Yang Guo. Decoration of Textured ZnO
Nanowires Array with CdTe Quantum Dots: Enhanced Light-Trapping
Effect and Photogenerated Charge Separation. The Journal of Physical
Chemistry C, 112(51):20560–20566, 2008.
[26] Chia-Yuan Chen, Mingkui Wang, Jheng-Ying Li, Nuttapol Pootrak-
ulchote, Leila Alibabaei, Cevey-ha Ngoc-le, Jean-David Decoppet, Jia-
Hung Tsai, Carole Grätzel, Chun-Guey Wu, Shaik M Zakeeruddin, and
Michael Gratzel. Highly Efficient Light-Harvesting Ruthenium Sensitizer
for Thin-Film Dye-Sensitized Solar Cells. ACS Nano, 3(10):3103–3109,
October 2009.
[27] Hanning Chen, Jeffrey M McMahon, Mark A Ratner, and George C
Schatz. Classical Electrodynamics Coupled to Quantum Mechanics for
Calculation of Molecular Optical Properties: a RT-TDDFT/FDTD Ap-
proach. The Journal of Physical Chemistry C, 114(34):14384–14392,
September 2010.
[28] W.C.H. Choy. Organic Solar Cells: Materials and Device Physics. Green
Energy and Technology. Springer London, 2012.
[29] Silvia Collavini, Sebastian F. Völker, and Juan Luis Delgado. Understand-
ing the outstanding power conversion efficiency of perovskite-based solar
cells. Angewandte Chemie International Edition, 54(34):9757–9759, 2015.
[30] Wikimedia Commons. Yee Cell implementation of FDTD method, 2013.
[31] Wikimedia Commons. Schematic of atomic arrangements of allotropic
forms of silicon, 2015.
[32] John Cook, Dana Nuccitelli, Sarah A Green, Mark Richardson, Bärbel
Winkler, Rob Painting, Robert Way, Peter Jacobs, and Andrew Skuce.
Quantifying the consensus on anthropogenic global warming in the sci-
entific literature. Environmental Research Letters, 8(2):024024, 2013.

153
[33] Michael G Deceglie, Vivian E Ferry, A Paul Alivisatos, and Harry A
Atwater. Design of Nanostructured Solar Cells Using Coupled Optical
and Electrical Modeling. Nano Letters, 12(6):2894–2900, June 2012.
[34] Alexei Deinega and Sajeev John. Effective optical response of silicon
to sunlight in the finite-difference time-domain method. Optics Letters,
37(1):112–114, 2012.
[35] Alexei Deinega and Sajeev John. Finite difference discretization of semi-
conductor drift-diffusion equations for nanowire solar cells. Computer
Physics Communications, 183(10):2128–2135, October 2012.

[36] Alexei Deinega and Sajeev John. Solar power conversion efficiency in
modulated silicon nanowire photonic crystals. Journal of Applied Physics,
112(7):074327, 2012.
[37] Guillaume Demésy and Sajeev John. Solar energy trapping with mod-
ulated silicon nanowire photonic crystals. Journal of Applied Physics,
112(7):074326, 2012.
[38] N Aaron Deskins and Michel Dupuis. Electron transport via polaron hop-
ping in bulk TiO2 : A density functional theory characterization. Physical
Review B, 75(19):195212, 2007.
[39] Silke L Diedenhofen, Olaf T A Janssen, Grzegorz Grzela, Erik P A M
Bakkers, and Jaime Gómez Rivas. Strong Geometrical Dependence of the
Absorption of Light in Arrays of Semiconductor Nanowires. ACS Nano,
5(3):2316–2323, March 2011.
[40] Pablo Docampo, Stefan Guldin, Tomas Leijtens, Nakita K. Noel, Ullrich
Steiner, and Henry J. Snaith. Lessons learned: From dye-sensitized solar
cells to all-solid-state hybrid devices. Advanced Materials, 26(24):4013–
4030, 2014.
[41] Roberto Dovesi, Roberto Orlando, Alessandro Erba, Claudio M Zicovich-
Wilson, Bartolomeo Civalleri, Silvia Casassa, Lorenzo Maschio, Matteo
Ferrabone, Marco De La Pierre, Philippe D’Arco, et al. CRYSTAL14: A
program for the ab initio investigation of crystalline solids. International
Journal of Quantum Chemistry, 114(19):1287–1317, 2014.
[42] A. Einstein. Über einen die Erzeugung und Verwandlung des Lichtes
betreffenden heuristischen Gesichtspunkt. Annalen der Physik, 322:132–
148, 1905.

[43] David Emin and T Holstein. Studies of small-polaron motion IV. Adia-
batic theory of the Hall effect. Annals of Physics, 53(3):439 – 520, 1969.
[44] Electromagnetic Template Library (EMTL), 2014.

154
[45] K Ernst, A Belaidi, and R Könenkamp. Solar cell with extremely thin
absorber on highly structured substrate. Semiconductor science and tech-
nology, 18:475, 2003.
[46] K Ernst, I Sieber, M Neumann-Spallart, M.-Ch Lux-Steiner, and R Kön-
enkamp. Characterization of II–VI compounds on porous substrates. Thin
Solid Films, 361–362:213 – 217, 2000.
[47] Lioz Etgar, Peng Gao, Zhaosheng Xue, Qin Peng, Aravind Kumar
Chandiran, Bin Liu, Md. K. Nazeeruddin, and Michael Grätzel. Meso-
scopic CH3 NH3 PbI3 /TiO2 Heterojunction Solar Cells. Journal of the
American Chemical Society, 134(42):17396–17399, 2012. PMID: 23043296.
[48] Jörg Ferber and Joachim Luther. Computer simulations of light scattering
and absorption in dye-sensitized solar cells. Solar Energy Materials and
Solar Cells, 54(1):265–275, 1998.
[49] Vivian E. Ferry, Albert Polman, and Harry A. Atwater. Modeling light
trapping in nanostructured solar cells. ACS Nano, 5(12):10055–10064,
2011. PMID: 22082201.
[50] Vivian E. Ferry, Marc A. Verschuuren, Hongbo B. T. Li, Ewold Verhagen,
Robert J. Walters, Ruud E. I. Schropp, Harry A. Atwater, and Albert
Polman. Light trapping in ultrathin plasmonic solar cells. Opt. Express,
18(S2):A237–A245, Jun 2010.
[51] Leigh Fischer, Andrei Zvyagin, Taras Plakhotnik, and Misha Vorobyev.
Numerical modeling of light propagation in a hexagonal array of dielectric
cylinders. JOSA A, 27(4):865–872, 2010.
[52] C Flesia and P Schwendimann. Analytical multiple scattering extension
of the mie theory. Applied Physics B: Lasers and Optics, 56(3):157–163,
1993.
[53] Stephen Foster and Sajeev John. Light-trapping in dye-sensitized solar
cells. Energy & Environmental Science, 6(10):2972, 2013.
[54] C.E. Fritts. On a new form of selenium photocell. American J. of Science,
26:465, 1883.
[55] Erik Garnett and Peidong Yang. Light trapping in silicon nanowire solar
cells. Nano Letters, 10(3):1082–1087, 2010. PMID: 20108969.
[56] Erik C Garnett, Mark L Brongersma, Yi Cui, and Michael D McGehee.
Nanowire Solar Cells. Annual Review of Materials Research, 41(1):269–
295, August 2011.
[57] JC Maxwell Garnett. Colours in metal glasses, in metallic films, and
in metallic solutions. II. Philosophical Transactions of the Royal Society
of London. Series A, Containing Papers of a Mathematical or Physical
Character, pages 237–288, 1906.

155
[58] S.D. Gedney and Bo Zhao. An auxiliary differential equation formulation
for the complex-frequency shifted pml. Antennas and Propagation, IEEE
Transactions on, 58(3):838–847, March 2010.
[59] L G Georgiadis and E E Kriezis. Scattering from two parallel cylinders of
finite length. Archiv für Elektrotechnik, 63(2):69–75, 1981.

[60] H. Gerischer, M.E. Michel-Beyerle, F. Rebentrost, and H. Tributsch. Sens-

itization of charge injection into semiconductors with large band gap. Elec-
trochimica Acta, 13(6):1509 – 1515, 1968.
[61] José Goldemberg, Thomas B Johansson, Amulya KN Reddy, Robert H
Williams, et al. Energy for a sustainable world. Wiley New York, 1988.
[62] I. Gonzalez-Valls and M. Lira-Cantu. Vertically-aligned nanostructures of
zno for excitonic solar cells: A review. Energy and Environmental Science,
2(1):19–34, 2009. cited By 386.
[63] Albert Goossens and Joris Hofhuis. Spray-deposited CuInS2 solar cells.
Nanotechnology, 19(42):424018, 2008.
[64] Michael Gratzel. Photoelectrochemical cells. Nature, 414(6861):338–344,
2001.
[65] Michael Gratzel. Dye-sensitized solar cells. Journal of Photochemistry
and Photobiology C: Photochemistry Reviews, 4(2):145–153, 2003.
[66] Martin A. Green, Keith Emery, Yoshihiro Hishikawa, Wilhelm Warta,
and Ewan D. Dunlop. Solar cell efficiency tables (version 39). Progress in
Photovoltaics: Research and Applications, 20(1):12–20, 2012.
[67] Martin A. Green, Mark J. Keevers, Ian Thomas, John B. Lasich, Keith
Emery, and Richard R. King. 40 Progress in Photovoltaics: Research and
Applications, 23(6):685–691, 2015.
[68] L E Greene, M Law, B D Yuhas, and P Yang. ZnO-TiO2 Core-Shell
Nanorod/P3HT Solar Cells. The Journal of Physical Chemistry C,
111(50):18451–18456, December 2007.

[69] Anjia Gu, Yijie Huo, Shu Hu, T Sarmiento, Evan Pickett, Dong Liang,
Shuang Li, A Lin, S Thombare, and Zongfu Yu. Design and growth of
III–V nanowire solar cell arrays on low cost substrates. pages 2034–2037,
2010.
[70] Haomin Guo, Long Wen, Xinhua Li, Zhifei Zhao, and Yuqi Wang. Ana-
lysis of optical absorption in GaAs nanowire arrays. Nanoscale Research
Letters, 6(1):617, December 2011.
[71] Jonathan JM Halls and Richard H Friend. Organic photovoltaic devices.

156
[72] W.M. Haynes. CRC Handbook of Chemistry and Physics, 94th Edition.
CRC Handbook of Chemistry and Physics. Taylor & Francis Limited,
2013.
[73] Euan Hendry, F Wang, J Shan, T F Heinz, and Mischa Bonn. Electron
transport in TiO2 probed by THz time-domain spectroscopy. Physical
Review B, 69(8):081101 (R)–081101 (R), 2004.
[74] H. Hertz. Ueber einen Einfluss des ultravioletten Lichtes auf die electrische
Entladung. Annalen der Physik, 267:983–1000, 1887.
[75] Lu Hu and Gang Chen. Analysis of Optical Absorption in Silicon Nanowire
Arrays for Photovoltaic Applications. Nano Letters, 7(11):3249–3252,
November 2007.
[76] Yafit Itzhaik, Olivia Niitsoo, Miles Page, and Gary Hodes. Sb2 S3 -
Sensitized Nanoporous TiO2 Solar Cells. The Journal of Physical Chem-
istry C, 113(11):4254–4256, 2009.

[77] Nam Joong Jeon, Jun Hong Noh, Woon Seok Yang, Young Chan Kim,
Seungchan Ryu, Jangwon Seo, and Sang Il Seok. Compositional engin-
eering of perovskite materials for high-performance solar cells. Nature,
517(7535):476–480, 01 2015.
[78] Qinglong Jiang, Xia Sheng, Yingxuan Li, Xinjian Feng, and Tao Xu.
Rutile TiO2 nanowire-based perovskite solar cells. Chem. Commun.,
50:14720–14723, 2014.
[79] Mikkel Jørgensen, Kion Norrman, and Frederik C. Krebs. Stabil-
ity/degradation of polymer solar cells. Solar Energy Materials and Solar
Cells, 92(7):686–714, jul 2008.

[80] M L Juan. An electromagnetic scattering model for multiple tree trunks

above a tilted rough ground plane. Geoscience and Remote Sensing, IEEE
Transactions on, 37(2), 1999.
[81] F Michael Kahnert. Numerical methods in electromagnetic scattering
theory. Journal of Quantitative Spectroscopy and Radiative Transfer, 79-
80:775–824, June 2003.
[82] I. Kaiser, K Ernst, C.H. Fischer, R Könenkamp, C. Rost, I. Sieber, and
M C Lux-Steiner. The ETA-solar cell with CuInS2 : A photovoltaic cell
concept using an extremely thin absorber (ETA). Solar Energy Materials
and Solar Cells, 67(1):89–96, 2001.

[83] Balaji Kannan, Kenneth Castelino, and Arun Majumdar. Design of Nano-
structured Heterojunction Polymer Photovoltaic Devices. Nano Letters,
3(12):1729–1733, December 2003.

157
[84] MA Karam and AK Fung. Electromagnetic scattering from a layer of finite
length, randomly oriented, dielectric, circular cylinders over a rough inter-
face with application to vegetation. Remote Sensing, (January 2013):37–
41, 1988.
[85] Brendan M Kayes, Harry A Atwater, and Nathan S Lewis. Comparison
of the device physics principles of planar and radial p-n junction nanorod
solar cells. Journal of Applied Physics, 97(11):114302, 2005.
[86] M D Kelzenberg, M C Putnam, D B Turner-Evans, N S Lewis, and H A
Atwater. Predicted efficiency of Si wire array solar cells. In Photovoltaic
Specialists Conference (PVSC), 2009 34th IEEE, pages 001948–001953,
2009.
[87] Thomas J Kempa, James F Cahoon, Sun-Kyung Kim, Robert W Day,
David C Bell, Hong-Gyu Park, and Charles M Lieber. Coaxial multishell
nanowires with high-quality electronic interfaces and tunable optical cav-
ities for ultrathin photovoltaics. Proceedings of the National Academy of
Sciences, 109(5):1407–1412, 2012.
[88] M. Kerker. The scattering of light, and other electromagnetic radiation.
Physical chemistry. Academic Press, 1969.
[89] M. Kerker and E. Matijević. Scattering of electromagnetic waves from
concentric infinite cylinders. JOSA, 51(5):506–508, 1961.

[90] AC Khazraji, S Hotchandani, S Das, and PV Kamat. Controlling dye

(Merocyanine-540) aggregation on nanostructured TiO2 films. An organ-
ized assembly approach for enhancing the efficiency of photosensitization.
Journal of Physical Chemistry B, 103(22):4693–4700, Jun 3 1999.

[91] D. Kieven, T. Dittrich, A. Belaidi, J. Tornow, K. Schwarzburg, N. Allsop,

and M. Lux-Steiner. Effect of internal surface area on the performance
of ZnO/In2 S3 /CuSCN solar cells with extremely thin absorber. Applied
Physics Letters, 92(15):153107, April 2008.
[92] Heejin Kim, Hyuncheol Jeong, Tae Kyu An, Chan Eon Park, and Kijung
Yong. Hybrid-Type Quantum-Dot Cosensitized ZnO Nanowire Solar Cell
with Enhanced Visible-Light Harvesting. ACS Applied Materials & Inter-
faces, 5(2):268–275, January 2013.
[93] Hui-Seon Kim, Chang-Ryul Lee, Jeong-Hyeok Im, Ki-Beom Lee, Thomas
Moehl, Arianna Marchioro, Soo-Jin Moon, Robin Humphry-Baker, Jun-
Ho Yum, Jacques E Moser, Michael Gratzel, and Nam-Gyu Park. Lead
Iodide Perovskite Sensitized All-Solid-State Submicron Thin Film Meso-
scopic Solar Cell with Efficiency Exceeding 9%. Scientific Reports, 2,
August 2012.

158
 [94] Sun-Kyung Kim, Robert W Day, James F Cahoon, Thomas J Kempa,
Kyung-Deok Song, Hong-Gyu Park, and Charles M Lieber. Tuning Light
Absorption in Core/Shell Silicon Nanowire Photovoltaic Devices through
Morphological Design. Nano Letters, 12(9):4971–4976, September 2012.
[95] Kintech Laboratory - R&D Service and Software for Predictive Physico-
chemical Modeling, 2014.
[96] C. Kittel. Introduction to solid state physics. Wiley, 1986.
[97] A. Kojima, K. Teshima, Y. Shirai, and T. Miyasaka. Organometal halide
perovskites as visible-light sensitizers for photovoltaic cells. Journal of the
American Chemical Society, 131(17):6050–6051, 2009. cited By 861.
[98] R Könenkamp, K Boedecker, M C Lux-Steiner, M Poschenrieder, F Zenia,
C Levy-Clement, and S Wagner. Thin film semiconductor deposition on
free-standing ZnO columns. Applied Physics Letters, 77(16):2575, 2000.
[99] R. Könenkamp, K. Ernst, Ch.-H. Fischer, M.C. Lux-Steiner, and C. Rost.
Semiconductor growth and junction formation within nano-porous oxides.
Physica status solidi (a), 182(1):151–155, 2000.
[100] Peter Krogstrup, Henrik Ingerslev Jørgensen, Martin Heiss, Olivier De-
michel, Jeppe V Holm, Martin Aagesen, Jesper Nygard, and Anna
Fontcuberta i Morral. Single-nanowire solar cells beyond the Shockley–
Queisser limit. Nature Photonics, 7(4):306–310, March 2013.
[101] Pankaj Kumar and Suresh Chand. Recent progress and future aspects of
organic solar cells. Progress in Photovoltaics: Research and Applications,
20(4):377–415, 2012.
[102] Kuan-Ting Kuo, Dean-Mo Liu, San-Yuan Chen, and Chin-Ching Lin.
Core-shell CuInS2 /ZnS quantum dots assembled on short ZnO nanowires
with enhanced photo-conversion efficiency. J. Mater. Chem., 19:6780–
6788, 2009.
[103] HyoJoong Lee, Henry C. Leventis, Soo-Jin Moon, Peter Chen, Seigo Ito,
Saif A. Haque, Tomas Torres, Frank Nüesch, Thomas Geiger, Shaik M.
Zakeeruddin, Michael Grätzel, and Md. Khaja Nazeeruddin. Pbs and cds
quantum dot-sensitized solid-state solar cells: “old concepts, new results”.
Advanced Functional Materials, 19(17):2735–2742, 2009.
[104] Jung-Chul Lee, Tae Geun Kim, Wonjoo Lee, Sung-Hwan Han, and Yun-
Mo Sung. Growth of CdS Nanorod-Coated TiO2 Nanowires on Con-
ductive Glass for Photovoltaic Applications. Crystal Growth & Design,
9(10):4519–4523, 2009.
[105] Jung-Chul Lee, Yun-Mo Sung, Tae Geun Kim, and Heon-Jin Choi. TiO2 –
CdSe nanowire arrays showing visible-range light absorption. Applied
Physics Letters, 91(11), 2007.

159
[106] Michael M. Lee, Joël Teuscher, Tsutomu Miyasaka, Takurou N. Murakami,
and Henry J. Snaith. Efficient hybrid solar cells based on meso-
superstructured organometal halide perovskites. Science, 338(6107):643–
647, 2012.
[107] S.C. Lee. Light scattering by a coated infinite cylinder in an absorbing
medium. JOSA A, 28(6):1067–1075, 2011.
[108] Siu-Chun Lee. Near field scattering by multiple infinite cylinders in an
absorbing medium at oblique incidence. Journal of Quantitative Spectro-
scopy and Radiative Transfer, 2012.

[109] Kurtis S. Leschkies, Ramachandran Divakar, Joysurya Basu, Emil Enache-

Pommer, Janice E. Boercker, C. Barry Carter, Uwe R. Kortshagen,
David J. Norris, and Eray S. Aydil. Photosensitization of ZnO Nanowires
with CdSe Quantum Dots for Photovoltaic Devices. Nano Letters,
7(6):1793–1798, 2007. PMID: 17503867.

[110] C Lévy-Clément, R Tena-Zaera, M A Ryan, A Katty, and G Hodes. CdSe-

Sensitized p-CuSCN/Nanowire n-ZnO Heterojunctions. Advanced Mater-
ials, 17(12):1512–1515, June 2005.
[111] Ming Li, Yong Liu, Hai Wang, Hui Shen, Wenxia Zhao, Hong Huang,
and Chaolun Liang. CdS/CdSe cosensitized oriented single-crystalline
TiO2 nanowire array for solar cell application. Journal of Applied Physics,
108(9):094304, 2010.
[112] Xiaofeng Li, Nicholas P Hylton, Vincenzo Giannini, Kan-Hua Lee, Ned J
Ekins-Daukes, and Stefan A Maier. Bridging electromagnetic and carrier
transport calculations for three-dimensional modelling of plasmonic solar
cells. Opt. Express, 19(S4):A888–A896, 2011.

[113] Xiaofeng Li, Nicholas P Hylton, Vincenzo Giannini, Kan-Hua Lee, Ned J
Ekins-Daukes, and Stefan A Maier. Multi-dimensional modeling of solar
cells with electromagnetic and carrier transport calculations. Progress in
Photovoltaics: Research and Applications, 21(1):109–120, February 2012.

[114] Z. Li, Z.R. Li, and H. Meng. Organic Light-Emitting Materials and
Devices. Optical science and engineering. CRC Press, 2006.
[115] Yi-Cheng Lin and Kamal Sarabandi. Electromagnetic scattering model
for a tree trunk above a tilted ground plane. Geoscience and Remote
Sensing, IEEE Transactions on, 33(4):1063–1070, 1995.

[116] Amy L Linsebigler, Guangquan Lu, and John T Yates. Photocatalysis on

TiO2 Surfaces: Principles, Mechanisms, and Selected Results. Chemical
reviews, 95(3):735–758, May 1995.

160
[117] Bin Liu and Eray S Aydil. Growth of Oriented Single-Crystalline Rutile
TiO 2Nanorods on Transparent Conducting Substrates for Dye-Sensitized
Solar Cells. Journal of the American Chemical Society, 131(11):3985–3990,
March 2009.
[118] Di Liu and Prashant V. Kamat. Photoelectrochemical behavior of thin
cadmium selenide and coupled titania/cadmium selenide semiconductor
films. The Journal of Physical Chemistry, 97(41):10769–10773, 1993.
[119] Mark Lonergan. Charge transport at conjugated polymer–inorganic semi-
conductor and conjugated polymer–metal interfaces. Annual Review of
Physical Chemistry, 55(1):257–298, June 2004.

[120] Lumerical FDTD Solutions, 2014.

[121] Henrik Lund. Renewable energy strategies for sustainable development.
Energy, 32(6):912–919, 2007.
[122] Christin Lundgren, Rene Lopez, Joan Redwing, and Kathleen Melde.
FDTD modeling of solar energy absorption in silicon branched nanowires.
Optics Express, 21(S3):A392, 2013.
[123] Jingshan Luo, Lin Ma, Tingchao He, Chin Fan Ng, Shijie Wang, Handong
Sun, and Hong Jin Fan. TiO2 /(CdS, CdSe, CdSeS) Nanorod Heterostruc-
tures and Photoelectrochemical Properties. Journal of Physical Chemistry
C, 116(22):11956–11963, June 2012.
[124] O Madelung. Semiconductors Data Handbook. Springer, Berlin, 2004.
[125] RA Marcus. Chemical and electrochemical electron-transfer theory. An-
nual Review of Physical Chemistry, 15(1):155–196, 1964.

[126] K. Masuko, M. Shigematsu, T. Hashiguchi, D. Fujishima, M. Kai,

N. Yoshimura, T. Yamaguchi, Y. Ichihashi, T. Mishima, N. Matsubara,
T. Yamanishi, T. Takahama, M. Taguchi, E. Maruyama, and S. Okamoto.
Achievement of more than 25% conversion efficiency with crystalline sil-
icon heterojunction solar cell. Photovoltaics, IEEE Journal of, 4(6):1433–
1435, Nov 2014.

[127] L E McNeil and R H French. Multiple scattering from rutile TiO2

particles. Acta materialia, 48(18-19):4571–4576, 2000.
[128] Lingyi Meng, ChiYung Yam, SiuKong Koo, Quan Chen, Ngai
Wong, and GuanHua Chen. Dynamic Multiscale Quantum Mechan-
ics/Electromagnetics Simulation Method. Journal of Chemical Theory
and Computation, 8(4):1190–1199, April 2012.
[129] Wyatt K Metzger. The potential and device physics of interdigitated thin-
film solar cells. Journal of Applied Physics, 103(9):094515–094515, 2008.

161
[130] Robert W. Meulenberg, Jonathan R.I. Lee, Abraham Wolcott, Jin Z.
Zhang, Louis J. Terminello, and Tony van Buuren. Determination of the
exciton binding energy in cdse quantum dots. ACS Nano, 3(2):325–330,
2009.
[131] Microvolt, a free C++ program for modelling semiconductor devices, 2014.
[132] J Möller, Ch. H Fischer, H. J Muffler, R Könenkamp, I Kaiser, C Kelch,
and M. C Lux-Steiner. A novel deposition technique for compound semi-
conductors on highly porous substrates: Ilgar. Thin Solid Films, 361–
362:113–117, 2 2000.
[133] Gopal K Mor, Oomman K Varghese, Maggie Paulose, Karthik Shankar,
and Craig A Grimes. A review on highly ordered, vertically oriented
TiO2 nanotube arrays: Fabrication, material properties, and solar energy
applications. Solar Energy Materials and Solar Cells, 90(14):2011–2075,
September 2006.
[134] Iván Mora-Seró, Sixto Giménez, Francisco Fabregat-Santiago, Eneko Aza-
ceta, Ramón Tena-Zaera, and Juan Bisquert. Modeling and characteriza-
tion of extremely thin absorber (eta) solar cells based on ZnO nanowires.
Physical Chemistry Chemical Physics, 13(15):7162, 2011.
[135] A. Nadarajah, Robert C. Word, Kaitlyn VanSant, and R. Könenkamp.
Nanowire–quantum-dot–polymer solar cell. Physica status solidi B,
245(9):1834–1837, 2008.
[136] M. Nanu, J. Schoonman, and A. Goossens. Inorganic nanocomposites of
n- and p-type semiconductors: A new type of three-dimensional solar cell.
Advanced Materials, 16(5):453–456, 2004.
[137] J. Nelson. The Physics of Solar Cells. Series on Properties of Semicon-
ductor Materials. Imperial College Press, 2003.
[138] Jenny Nelson. Organic photovoltaic films. Current Opinion in Solid State
and Materials Science, 6(1):87–95, 2002.
[139] S V Novikov, D H Dunlap, V M Kenkre, P E Parris, and A V Vannikov.
Essential role of correlations in governing charge transport in disordered
organic materials. Physical Review Letters, 81(20):4472, 1998.
[140] Keat G Ong, Oomman K Varghese, Gopal K Mor, Karthik Shankar, and
Craig A Grimes. Application of finite-difference time domain to dye-
sensitized solar cells: The effect of nanotube-array negative electrode di-
mensions on light absorption. Solar Energy Materials and Solar Cells,
91(4):250–257, February 2007.
[141] Brian O’Regan and Michael Gratzel. A low-cost, high-efficiency solar cell
based on dye-sensitized colloidal tio2 films. Nature, 353(6346):737–740, 10
1991.

162
[142] Brian O’Regan, Jacques Moser, Marc Anderson, and Michael Graetzel.
Vectorial electron injection into transparent semiconductor membranes
and electric field effects on the dynamics of light-induced charge separa-
tion. Journal of Physical Chemistry, 94(24):8720–8726, 1990.
[143] Brian O’Regan and Daniel T. Schwartz. Large Enhancement in Photocur-
rent Efficiency Caused by UV Illumination of the Dye-Sensitized Hetero-
junction TiO2 /RuLL‘NCS/CuSCN: Initiation and Potential Mechanisms.
Chemistry of Materials, 10(6):1501–1509, 1998.
[144] E.D. Palik. Handbook of Optical Constants of Solids II. Academic Press
handbook series. Academic Press, 1991.

[145] J. Perlin. From Space to Earth: The Story of Solar Electricity. Harvard
University Press, 1999.
[146] J. Perlin. Silicon Solar Cell Turns 50. Aug 2004.
[147] Robert Plass, Serge Pelet, Jessica Krueger, Michael Grätzel, and Udo
Bach. Quantum dot sensitization of organic-inorganic hybrid solar cells.
The Journal of Physical Chemistry B, 106(31):7578–7580, 2002.
[148] A.C. Polycarpou. Introduction to the Finite Element Method in Electro-
magnetics. Synthesis lectures on computational electromagnetics. Morgan
& Claypool Publishers, 2006.

[149] Jianhang Qiu, Yongcai Qiu, Keyou Yan, Min Zhong, Cheng Mu, He Yan,
and Shihe Yang. All-solid-state hybrid solar cells based on a new organo-
metal halide perovskite sensitizer and one-dimensional TiO2 nanowire ar-
rays. Nanoscale, 5:3245–3248, 2013.
[150] S. Ren, L.-Y. Chang, S.-K. Lim, J. Zhao, M. Smith, N. Zhao, V. Bu-
lović, M. Bawendi, and S. Gradečak. Inorganic-organic hybrid solar cell:
Bridging quantum dots to conjugated polymer nanowires. Nano Letters,
11(9):3998–4002, 2011. cited By 206.
[151] REN21. Renewables 2011 global status report. July 2011.

[152] Michael Riordan and Lillian Hoddeson. The origins of the pn junction.
IEEE Spectr., 34(6):46–51, June 1997.
[153] T Rother, S Havemann, and K Schmidt. Scattering of plane waves on fi-
nite cylinders with non-circular cross-sections. Journal of Electromagnetic
Waves and Applications, 13(8):1037–1038, 1999.

[154] S Ramezani Sani, A Morteza Ali, and R Jafari. A comparison between

optical properties of TiO2 nanowires obtained by EMA method and ex-
periment. Physica B: Physics of Condensed Matter, 406(18):3383–3388,
September 2011.

163
[155] D L Scharfetter and H K Gummel. Large-signal analysis of a silicon read
diode oscillator. Electron Devices, IEEE Transactions on, 16(1):64–77,
1969.
[156] J C Scott and G G Malliaras. Charge injection and recombination at the
metal-organic interface. Chemical Physics Letters, 299(2):115–119, 1999.

[157] S S Seker and A Schneider. Electromagnetic scattering from a dielectric

cylinder of finite length. IEEE Transactions on Antennas and Propaga-
tion, 36(2):303–307, 1988.
[158] S. Selberherr. Analysis and Simulation of Semiconductor Devices.
Springer-Verlag, 1984.
[159] G A Shah. Scattering of plane electromagnetic waves by infinite concent-
riccircular cylinders at oblique incidence. Monthly Notices of the Royal
Astronomical Society, 148:93, 1970.
[160] William Shockley and Hans J Queisser. Detailed balance limit of efficiency
of p-n junction solar cells. Journal of Applied Physics, 32(3):510–519,
1961.
[161] Douglas W. Sievers, Vishal Shrotriya, and Yang Yang. Modeling optical
effects and thickness dependent current in polymer bulk-heterojunction
solar cells. Journal of Applied Physics, 100(11), 2006.

[162] E. Snitzer. Cylindrical dielectric waveguide modes⇤. J. Opt. Soc. Am.,

51(5):491–498, May 1961.
[163] D. L. Staebler and C. R. Wronski. Optically induced conductivity changes
in discharge-produced hydrogenated amorphous silicon. Journal of Ap-
plied Physics, 51:3262–3268, June 1980.

[164] Billy J. Stanbery. Copper indium selenides and related materials for photo-
voltaic devices. Critical Reviews in Solid State and Materials Sciences,
27(2):73–117, 2002.
[165] A. Taflove and S.C. Hagness. Computational Electrodynamics: The Finite-
difference Time-domain Method. Artech House antennas and propagation
library. Artech House, 2005.
[166] Youngjo Tak, Suk Joon Hong, Jae Sung Lee, and Kijung Yong. Fab-
rication of ZnO/CdS core/shell nanowire arrays for efficient solar energy
conversion. Journal of Materials Chemistry, 19(33):5945, 2009.

[167] Hairen Tan, Rudi Santbergen, Arno H M Smets, and Miro Zeman. Plas-
monic Light Trapping in Thin-film Silicon Solar Cells with Improved Self-
Assembled Silver Nanoparticles. Nano Letters, 12(8):4070–4076, August
2012.

164
[168] Jinyao Tang, Ziyang Huo, Sarah Brittman, Hanwei Gao, and Peidong
Yang. Solution-processed core-shell nanowires for efficient photovoltaic
cells. Nature Nanotechnology, 6(9):568–572, August 2011.
[169] K Taretto and U Rau. Modeling extremely thin absorber solar cells for
optimized design. Progress in Photovoltaics: Research and Applications,
12(8):573–591, 2004.
[170] R. Tena-Zaera, A. Katty, S. Bastide, C. Lévy-Clément, B. O’Regan, and
V. Muñoz-Sanjosé. ZnO/CdTe/CuSCN, a promising heterostructure to
act as inorganic ETA-solar cell. Thin Solid Films, 483(1–2):372 – 377,
2005.

[171] Ramón Tena-Zaera, Jamil Elias, and Claude Lévy-Clément. ZnO nanowire
arrays: Optical scattering and sensitization to solar light. Applied Physics
Letters, 93(23):233119, 2008.
[172] Ramon Tena-Zaera, Margaret A. Ryan, Abou Katty, Gary Hodes,
Stéphane Bastide, and Claude Lévy-Clément. Fabrication and character-
ization of ZnO nanowires/CdSe/CuSCN eta-solar cell. Comptes Rendus
Chimie, 9(5–6):717 – 729, 2006.
[173] K Tennakone, G R R A Kumara, I R M Kottegoda, V P S Perera, and
G M L P Aponsu. Nanoporous n-TiO2 /selenium/p-CuCNS photovoltaic
cell. Journal of Physics D: Applied Physics, 31(18):2326, 1998.

[174] K Tennakone, G R R A Kumara, A R Kumarasinghe, K G U Wijayantha,

and P M Sirimanne. A dye-sensitized nano-porous solid-state photovoltaic
cell. Semiconductor Science and Technology, 10(12):1689, 1995.
[175] K Tennakone, G.R.R.A Kumara, I.R.M Kottegoda, V.P.S Perera, and
P.S.R.S Weerasundara. Sensitization of nano-porous films of TiO2 with
santalin (red sandalwood pigment) and construction of dye-sensitized
solid-state photovoltaic cells. Journal of Photochemistry and Photobio-
logy A: Chemistry, 117(2):137 – 142, 1998.
[176] Shin-Hung Tsai, Hung-Chih Chang, Hsin-Hua Wang, Szu-Ying Chen,
Chin-An Lin, Show-An Chen, Yu-Lun Chueh, and Jr-Hau He. Significant
Efficiency Enhancement of Hybrid Solar Cells Using Core–Shell Nanowire
Geometry for Energy Harvesting. ACS Nano, 5(12):9501–9510, December
2011.
[177] Sergio Valencia, Juan Miguel Marín, and Gloria Restrepo. Study of the
Bandgap of Synthesized Titanium Dioxide Nanoparticules Using the Sol-
Gel Method and a Hydrothermal Treatment. Open Materials Science
Journal, 2010.
[178] H.C. van de HULST. Light Scattering by Small Particles. Structure of
Matter Series. John Wiley & Sons, 1957.

165
[179] D. Vasileska and S.M. Goodnick. Computational Electronics. Synthesis
lectures on computational electromagnetics. Morgan & Claypool Publish-
ers, 2006.
[180] V G Vereshchagin, R A Dynich, and A N Ponyavina. Light extinction
by two finite parallel dielectric cylinders. Journal of applied spectroscopy,
66(1):62–67, 1999.
[181] Ralf Vogel, Klaus Pohl, and Horst Weller. Sensitization of highly porous,
polycrystalline TiO2 electrodes by quantum sized CdS. Chemical Physics
Letters, 174(3):241 – 246, 1990.
[182] J Wallentin, N Anttu, D Asoli, M Huffman, I Aberg, M H Magnusson,
G Siefer, P Fuss-Kailuweit, F Dimroth, B Witzigmann, H Q Xu, L Samuel-
son, K Deppert, and M T Borgstrom. InP Nanowire Array Solar Cells
Achieving 13.8Exceeding the Ray Optics Limit. 339(6123):1057–1060,
February 2013.
[183] Chen Wang, Shuangcheng Yu, Wei Chen, and Cheng Sun. Highly Efficient
Light-Trapping Structure Design Inspired By Natural Evolution. Scientific
Reports, 3, January 2013.
[184] Jingyang Wang, Tianjin Zhang, Duofa Wang, Ruikun Pan, Qingqing
Wang, and Hanming Xia. Influence of CdSe quantum dot interlayer on the
performance of polymer/TiO2 nanorod arrays hybrid solar cell. Chemical
Physics Letters, 541(C):105–109, July 2012.
[185] Ru T Wang and H C Van de Hulst. Application of the exact solution for
scattering by an infinite cylinder to the estimation of scattering by a finite
cylinder. Applied Optics, 34(15):2811–2821, 1995.
[186] Andrew Watt, Troy Eichmann, Halina Rubinsztein-Dunlop, and Paul
Meredith. Carrier transport in pbs nanocrystal conducting polymer com-
posites. Applied Physics Letters, 87(25), 2005.
[187] Di Wei. Dye sensitized solar cells. International Journal of Molecular
Sciences, 11(3):1103–1113, 2010.
[188] Long Wen, Zhifei Zhao, Xinhua Li, Yanfen Shen, Haoming Guo, and
Yuqi Wang. Theoretical analysis and modeling of light trapping in high
efficicency GaAs nanowire array solar cells. Applied Physics Letters,
99(14):143116, 2011.
[189] K J Willis, S C Hagness, and I Knezevic. Multiphysics simulation of high-
frequency carrier dynamics in conductive materials. Journal of Applied
Physics, 110(6):063714, 2011.
[190] Matthew Wright and Ashraf Uddin. Organic—inorganic hybrid solar cells:
A comparative review. Solar Energy Materials and Solar Cells, 107:87 –
111, 2012.

166
[191] Guogang Xue, Yong Guo, Tao Yu, Jie Guan, Xirui Yu, Jiyuan Zhang,
Jianguo Liu, and Zhigang Zou. Degradation mechanisms investigation for
long-term thermal stability of dye-sensitized solar cells. Int. J. Electro-
chem. Sci, 7:1496–1511, 2012.
[192] Eli Yablonovitch. Statistical ray optics. J. Opt. Soc. Am., 72(7):899–907,
Jul 1982.
[193] Fan Yang, Max Shtein, and Stephen R. Forrest. Controlled growth of a
molecular bulk heterojunction photovoltaic cell. Nat Mater, 4(1):37–41,
01 2005.

[194] Kouji Yasuda, Kunio Saegusa, and ToruH. Okabe. Production of solar-
grade silicon by halidothermic reduction of silicon tetrachloride. 42(1):37–
49, 2011.
[195] Abdullah Yildiz, Felicia Iacomi, and Diana Mardare. Polaron transport
in TiO2 thin films. Journal of Applied Physics, 108(8):083701, 2010.

[196] H A Yousif, R E Mattis, and K Kozminski. Light scattering at oblique in-

cidence on two coaxial cylinders. Applied Optics, 33(18):4013–4024, 1994.
[197] Wenhua Yu and R. Mittra. A conformal FDTD software package modeling
antennas and microstrip circuit components. Antennas and Propagation
Magazine, IEEE, 42(5):28–39, Oct 2000.

[198] Zongfu Yu, Aaswath Raman, and Shanhui Fan. Fundamental limit of
nanophotonic light trapping in solar cells. Proceedings of the National
Academy of Sciences, 107(41):17491–17496, 2010.
[199] Mauro Zanuccoli, Igor Semenihin, Jérôme Michallon, Enrico Sangiorgi,
and Claudio Fiegna. Advanced electro-optical simulation of nanowire-
based solar cells. Journal of Computational Electronics, 12(4):572–584,
November 2013.
[200] Jie Zhang, Philippe Barboux, and Thierry Pauporté. Electrochemical
Design of Nanostructured ZnO Charge Carrier Layers for Efficient Solid-
State Perovskite-Sensitized Solar Cells. Advanced Energy Materials, 4(18),
2014.
[201] Kun Zhang, Chuanjiang Qin, Xudong Yang, Ashraful Islam, Shufang
Zhang, Han Chen, and Liyuan Han. High-performance, transparent, dye-
sensitized solar cells for see-through photovoltaic windows. Advanced En-
ergy Materials, 4(11), 2014. 1301966.

[202] Yu Zhang, Tengfeng Xie, Tengfei Jiang, Xiao Wei, Shan Pang, Xi Wang,
and Dejun Wang. Surface photovoltage characterization of a ZnO
nanowire array/CdS quantum dot heterogeneous film and its application
for photovoltaic devices. Nanotechnology, 20(15):155707, 2009.

167
[203] Y. Zhou, M. Eck, and M. Krüger. Bulk-heterojunction hybrid solar cells
based on colloidal nanocrystals and conjugated polymers. Energy and
Environmental Science, 3(12):1851–1864, 2010. cited By 150.
[204] Yu Zou, Dongsheng Li, Xia Sheng, Lei Wang, and Deren Yang. Fabrication
of TiO2 nanorod array/semiconductor nanocrystal hybrid structure for
photovoltaic applications. Solar Energy, 86(5):1359–1365, May 2012.

168
Appendix A

Derivation of Absorption per

Unit Volume in Nanoparticles

One can imagine the nanowire sectioned into n layers such as in Figure A.1. Each
layer has an absorption coefficient ↵ (absorption per layer) and the number of
nanoparticles per layer is nL .

Figure A.1: Diagram of nanowire coated with nanoparticles. Light is incident

from above and propagates through the nanowire

The intensity of light going into layer i is related to the intensity of light

169
coming out of layer i by the following equation
(i) (i) (i)
Iin Iout = ↵Iin (A.1)
(i) (i)
Iout = (1 ↵)Iin (A.2)
(i) (i 1) (i) (i 1)
By definition, Iin = Iout . Therefore, Iout = (1 ↵)Iout . Hence, the total
absorption of the nanowire is given by
(n) (0)
(Iout Iin )
A= (0)
(A.3)
Iin
= ((1 ↵)n 1) = 1 (1 ↵)n (A.4)

The coefficient ↵ is absorption per layer. We can express the equation above in
terms of absorption per nanoparticle ↵1 and nanoparticles per layer nL

A =1 (1 ↵ 1 nL ) n (A.5)
n
=1 (1 V1 nL ) (A.6)

where V1 = 4⇡r3 /3 is the volume of one nanoparticle, and  is the absorption per
unit volume of one nanoparticle. It is this coefficient  which we aim to investi-
gate in these simulations. By using a binomial expansion, we can approximate
equation A.6 like so
Xn ✓ ◆
n
A =1 ( ↵)p (A.7)
p=0
p
n
X n! ( ↵)p
=1 (A.8)
p=0
(n p)! p!
Xn
n! 1 ( ↵n)p
=1 (A.9)
p=0
(n p)! np p!
Xn
( ↵1 N )p
⇡1 (A.10)
p=0
p!
↵1 N
⇡1 e (A.11)
V1 N
=1 e (A.12)

where N is the total number of nanoparticles on the nanowire. From this we

get the dependence of the absorption on the radius of the nanoparticle r,
r3
1 A=e (A.13)

where is related to  by
3
= (A.14)
4⇡N

170
Appendix B

Permission Documents

Permissions have been granted to republish all third party copyrighted figures
in this thesis. The following pages contain the permissions documents obtained
for these figures.

171
 ELSEVIER LICENSE
TERMS AND CONDITIONS
Jan 07, 2016

This is a License Agreement between Thomas Edwards ("You") and Elsevier ("Elsevier")
provided by Copyright Clearance Center ("CCC"). The license consists of your order details,
the terms and conditions provided by Elsevier, and the payment terms and conditions.

All payments must be made in full to CCC. For payment instructions, please see
information listed at the bottom of this form.

Supplier Elsevier Limited

The Boulevard,Langford Lane
Kidlington,Oxford,OX5 1GB,UK
Registered Company 1982084
Number

Customer name Thomas Edwards

Customer address Imperial College London

London, SW7 2AZ

License number 3773320358294

License date Dec 20, 2015

Licensed content publisher Elsevier

Licensed content publication Physica B: Condensed Matter

Licensed content title A comparison between optical properties of TiO2 nanowires obtained
by EMA method and experiment
Licensed content author S. Ramezani Sani,A. Morteza Ali,R. Jafari

Licensed content date 15 September 2011

Licensed content volume 406

number

Licensed content issue 18

number
Number of pages 6

Start Page 3383

End Page 3388

Type of Use reuse in a thesis/dissertation

Intended publisher of new other

work

Portion figures/tables/illustrations
Number of 2
figures/tables/illustrations

Format both print and electronic

Are you the author of this No

Elsevier article?

Will you be translating? No

Original figure numbers Figures 2 and 6

Title of your Optimisation of Coated Nanowire Solar Cells by Simulation

thesis/dissertation

Expected completion date Nov 2015

Estimated size (number of 170
pages)
Elsevier VAT number GB 494 6272 12

Permissions price 0.00 GBP

VAT/Local Sales Tax 0.00 GBP / 0.00 GBP

Total 0.00 GBP

Terms and Conditions

INTRODUCTION
1. The publisher for this copyrighted material is Elsevier. By clicking "accept" in
connection with completing this licensing transaction, you agree that the following terms
and conditions apply to this transaction (along with the Billing and Payment terms and
conditions established by Copyright Clearance Center, Inc. ("CCC"), at the time that you
opened your Rightslink account and that are available at any time at
http://myaccount.copyright.com).
GENERAL TERMS
2. Elsevier hereby grants you permission to reproduce the aforementioned material subject to
the terms and conditions indicated.
3. Acknowledgement: If any part of the material to be used (for example, figures) has
appeared in our publication with credit or acknowledgement to another source, permission
must also be sought from that source. If such permission is not obtained then that material
may not be included in your publication/copies. Suitable acknowledgement to the source
must be made, either as a footnote or in a reference list at the end of your publication, as
follows:
"Reprinted from Publication title, Vol /edition number, Author(s), Title of article / title of
chapter, Pages No., Copyright (Year), with permission from Elsevier [OR APPLICABLE
SOCIETY COPYRIGHT OWNER]." Also Lancet special credit - "Reprinted from The
Lancet, Vol. number, Author(s), Title of article, Pages No., Copyright (Year), with
permission from Elsevier."
4. Reproduction of this material is confined to the purpose and/or media for which
permission is hereby given.
5. Altering/Modifying Material: Not Permitted. However figures and illustrations may be
altered/adapted minimally to serve your work. Any other abbreviations, additions, deletions
and/or any other alterations shall be made only with prior written authorization of Elsevier
Ltd. (Please contact Elsevier at permissions@elsevier.com)
6. If the permission fee for the requested use of our material is waived in this instance, please
be advised that your future requests for Elsevier materials may attract a fee.
7. Reservation of Rights: Publisher reserves all rights not specifically granted in the
combination of (i) the license details provided by you and accepted in the course of this
licensing transaction, (ii) these terms and conditions and (iii) CCC's Billing and Payment
terms and conditions.
8. License Contingent Upon Payment: While you may exercise the rights licensed
immediately upon issuance of the license at the end of the licensing process for the
transaction, provided that you have disclosed complete and accurate details of your proposed
use, no license is finally effective unless and until full payment is received from you (either
by publisher or by CCC) as provided in CCC's Billing and Payment terms and conditions. If
full payment is not received on a timely basis, then any license preliminarily granted shall be
deemed automatically revoked and shall be void as if never granted. Further, in the event
that you breach any of these terms and conditions or any of CCC's Billing and Payment
terms and conditions, the license is automatically revoked and shall be void as if never
granted. Use of materials as described in a revoked license, as well as any use of the
materials beyond the scope of an unrevoked license, may constitute copyright infringement
and publisher reserves the right to take any and all action to protect its copyright in the
materials.
9. Warranties: Publisher makes no representations or warranties with respect to the licensed
material.
10. Indemnity: You hereby indemnify and agree to hold harmless publisher and CCC, and
their respective officers, directors, employees and agents, from and against any and all
claims arising out of your use of the licensed material other than as specifically authorized
pursuant to this license.
11. No Transfer of License: This license is personal to you and may not be sublicensed,
assigned, or transferred by you to any other person without publisher's written permission.
12. No Amendment Except in Writing: This license may not be amended except in a writing
signed by both parties (or, in the case of publisher, by CCC on publisher's behalf).
13. Objection to Contrary Terms: Publisher hereby objects to any terms contained in any
purchase order, acknowledgment, check endorsement or other writing prepared by you,
which terms are inconsistent with these terms and conditions or CCC's Billing and Payment
terms and conditions. These terms and conditions, together with CCC's Billing and Payment
terms and conditions (which are incorporated herein), comprise the entire agreement
between you and publisher (and CCC) concerning this licensing transaction. In the event of
any conflict between your obligations established by these terms and conditions and those
established by CCC's Billing and Payment terms and conditions, these terms and conditions
shall control.
14. Revocation: Elsevier or Copyright Clearance Center may deny the permissions described
in this License at their sole discretion, for any reason or no reason, with a full refund payable
to you. Notice of such denial will be made using the contact information provided by you.
Failure to receive such notice will not alter or invalidate the denial. In no event will Elsevier
or Copyright Clearance Center be responsible or liable for any costs, expenses or damage
incurred by you as a result of a denial of your permission request, other than a refund of the
amount(s) paid by you to Elsevier and/or Copyright Clearance Center for denied
permissions.
LIMITED LICENSE
The following terms and conditions apply only to specific license types:
15. Translation: This permission is granted for non-exclusive world English rights only
unless your license was granted for translation rights. If you licensed translation rights you
may only translate this content into the languages you requested. A professional translator
must perform all translations and reproduce the content word for word preserving the
integrity of the article.
16. Posting licensed content on any Website: The following terms and conditions apply as
follows: Licensing material from an Elsevier journal: All content posted to the web site must
maintain the copyright information line on the bottom of each image; A hyper-text must be
included to the Homepage of the journal from which you are licensing at
http://www.sciencedirect.com/science/journal/xxxxx or the Elsevier homepage for books at
http://www.elsevier.com; Central Storage: This license does not include permission for a
scanned version of the material to be stored in a central repository such as that provided by
Heron/XanEdu.
Licensing material from an Elsevier book: A hyper-text link must be included to the Elsevier
homepage at http://www.elsevier.com . All content posted to the web site must maintain the
copyright information line on the bottom of each image.

Posting licensed content on Electronic reserve: In addition to the above the following
clauses are applicable: The web site must be password-protected and made available only to
bona fide students registered on a relevant course. This permission is granted for 1 year only.
You may obtain a new license for future website posting.
17. For journal authors: the following clauses are applicable in addition to the above:
Preprints:
A preprint is an author's own write-up of research results and analysis, it has not been peer-
reviewed, nor has it had any other value added to it by a publisher (such as formatting,
copyright, technical enhancement etc.).
Authors can share their preprints anywhere at any time. Preprints should not be added to or
enhanced in any way in order to appear more like, or to substitute for, the final versions of
articles however authors can update their preprints on arXiv or RePEc with their Accepted
scan the printed edition to create an electronic version. Posting to a repository: Authors are
permitted to post a summary of their chapter only in their institution's repository.
19. Thesis/Dissertation: If your license is for use in a thesis/dissertation your thesis may be
submitted to your institution in either print or electronic form. Should your thesis be
published commercially, please reapply for permission. These requirements include
permission for the Library and Archives of Canada to supply single copies, on demand, of
the complete thesis and include permission for Proquest/UMI to supply single copies, on
demand, of the complete thesis. Should your thesis be published commercially, please
reapply for permission. Theses and dissertations which contain embedded PJAs as part of the
formal submission can be posted publicly by the awarding institution with DOI links back to
the formal publications on ScienceDirect.

Elsevier Open Access Terms and Conditions
You can publish open access with Elsevier in hundreds of open access journals or in nearly
2000 established subscription journals that support open access publishing. Permitted third
party re-use of these open access articles is defined by the author's choice of Creative
Commons user license. See our open access license policy for more information.
Terms & Conditions applicable to all Open Access articles published with Elsevier:
Any reuse of the article must not represent the author as endorsing the adaptation of the
article nor should the article be modified in such a way as to damage the author's honour or
reputation. If any changes have been made, such changes must be clearly indicated.
The author(s) must be appropriately credited and we ask that you include the end user
license and a DOI link to the formal publication on ScienceDirect.
If any part of the material to be used (for example, figures) has appeared in our publication
with credit or acknowledgement to another source it is the responsibility of the user to ensure
their reuse complies with the terms and conditions determined by the rights holder.
Additional Terms & Conditions applicable to each Creative Commons user license:
CC BY: The CC-BY license allows users to copy, to create extracts, abstracts and new
works from the Article, to alter and revise the Article and to make commercial use of the
Article (including reuse and/or resale of the Article by commercial entities), provided the
user gives appropriate credit (with a link to the formal publication through the relevant
DOI), provides a link to the license, indicates if changes were made and the licensor is not
represented as endorsing the use made of the work. The full details of the license are
available at http://creativecommons.org/licenses/by/4.0.
CC BY NC SA: The CC BY-NC-SA license allows users to copy, to create extracts,
abstracts and new works from the Article, to alter and revise the Article, provided this is not
done for commercial purposes, and that the user gives appropriate credit (with a link to the
formal publication through the relevant DOI), provides a link to the license, indicates if
changes were made and the licensor is not represented as endorsing the use made of the
work. Further, any new works must be made available on the same conditions. The full
details of the license are available at http://creativecommons.org/licenses/by-nc-sa/4.0.
CC BY NC ND: The CC BY-NC-ND license allows users to copy and distribute the Article,
provided this is not done for commercial purposes and further does not permit distribution of
the Article if it is changed or edited in any way, and provided the user gives appropriate
credit (with a link to the formal publication through the relevant DOI), provides a link to the
license, and that the licensor is not represented as endorsing the use made of the work. The
full details of the license are available at http://creativecommons.org/licenses/by-nc-nd/4.0.
Any commercial reuse of Open Access articles published with a CC BY NC SA or CC BY
NC ND license requires permission from Elsevier and will be subject to a fee.
Commercial reuse includes:

- Associating advertising with the full text of the Article

- Charging fees for document delivery or access
- Article aggregation
- Systematic distribution via e-mail lists or share buttons
 ELSEVIER LICENSE
TERMS AND CONDITIONS
Jan 07, 2016

This is a License Agreement between Thomas Edwards ("You") and Elsevier ("Elsevier")
provided by Copyright Clearance Center ("CCC"). The license consists of your order details,
the terms and conditions provided by Elsevier, and the payment terms and conditions.

All payments must be made in full to CCC. For payment instructions, please see
information listed at the bottom of this form.

Supplier Elsevier Limited

The Boulevard,Langford Lane
Kidlington,Oxford,OX5 1GB,UK
Registered Company 1982084
Number

Customer name Thomas Edwards

Customer address Imperial College London

London, SW7 2AZ

License number 3773530858960

License date Dec 21, 2015

Licensed content publisher Elsevier

Licensed content publication Journal of Computational Physics

Licensed content title A perfectly matched layer for the absorption of electromagnetic
waves
Licensed content author Jean-Pierre Berenger

Licensed content date October 1994

Licensed content volume 114

number

Licensed content issue 2

number
Number of pages 16

Start Page 185

End Page 200

Type of Use reuse in a thesis/dissertation

Intended publisher of new other

work

Portion figures/tables/illustrations
Number of 1
figures/tables/illustrations

Format both print and electronic

Are you the author of this No

Elsevier article?

Will you be translating? No

Original figure numbers Figure 3

Title of your Optimisation of Coated Nanowire Solar Cells by Simulation

thesis/dissertation

Expected completion date Nov 2015

Estimated size (number of 170
pages)
Elsevier VAT number GB 494 6272 12

Permissions price 0.00 GBP

VAT/Local Sales Tax 0.00 GBP / 0.00 GBP

Total 0.00 GBP

Terms and Conditions

INTRODUCTION
1. The publisher for this copyrighted material is Elsevier. By clicking "accept" in
connection with completing this licensing transaction, you agree that the following terms
and conditions apply to this transaction (along with the Billing and Payment terms and
conditions established by Copyright Clearance Center, Inc. ("CCC"), at the time that you
opened your Rightslink account and that are available at any time at
http://myaccount.copyright.com).
GENERAL TERMS
2. Elsevier hereby grants you permission to reproduce the aforementioned material subject to
the terms and conditions indicated.
3. Acknowledgement: If any part of the material to be used (for example, figures) has
appeared in our publication with credit or acknowledgement to another source, permission
must also be sought from that source. If such permission is not obtained then that material
may not be included in your publication/copies. Suitable acknowledgement to the source
must be made, either as a footnote or in a reference list at the end of your publication, as
follows:
"Reprinted from Publication title, Vol /edition number, Author(s), Title of article / title of
chapter, Pages No., Copyright (Year), with permission from Elsevier [OR APPLICABLE
SOCIETY COPYRIGHT OWNER]." Also Lancet special credit - "Reprinted from The
Lancet, Vol. number, Author(s), Title of article, Pages No., Copyright (Year), with
permission from Elsevier."
4. Reproduction of this material is confined to the purpose and/or media for which
permission is hereby given.
5. Altering/Modifying Material: Not Permitted. However figures and illustrations may be
altered/adapted minimally to serve your work. Any other abbreviations, additions, deletions
and/or any other alterations shall be made only with prior written authorization of Elsevier
Ltd. (Please contact Elsevier at permissions@elsevier.com)
6. If the permission fee for the requested use of our material is waived in this instance, please
be advised that your future requests for Elsevier materials may attract a fee.
7. Reservation of Rights: Publisher reserves all rights not specifically granted in the
combination of (i) the license details provided by you and accepted in the course of this
licensing transaction, (ii) these terms and conditions and (iii) CCC's Billing and Payment
terms and conditions.
8. License Contingent Upon Payment: While you may exercise the rights licensed
immediately upon issuance of the license at the end of the licensing process for the
transaction, provided that you have disclosed complete and accurate details of your proposed
use, no license is finally effective unless and until full payment is received from you (either
by publisher or by CCC) as provided in CCC's Billing and Payment terms and conditions. If
full payment is not received on a timely basis, then any license preliminarily granted shall be
deemed automatically revoked and shall be void as if never granted. Further, in the event
that you breach any of these terms and conditions or any of CCC's Billing and Payment
terms and conditions, the license is automatically revoked and shall be void as if never
granted. Use of materials as described in a revoked license, as well as any use of the
materials beyond the scope of an unrevoked license, may constitute copyright infringement
and publisher reserves the right to take any and all action to protect its copyright in the
materials.
9. Warranties: Publisher makes no representations or warranties with respect to the licensed
material.
10. Indemnity: You hereby indemnify and agree to hold harmless publisher and CCC, and
their respective officers, directors, employees and agents, from and against any and all
claims arising out of your use of the licensed material other than as specifically authorized
pursuant to this license.
11. No Transfer of License: This license is personal to you and may not be sublicensed,
assigned, or transferred by you to any other person without publisher's written permission.
12. No Amendment Except in Writing: This license may not be amended except in a writing
signed by both parties (or, in the case of publisher, by CCC on publisher's behalf).
13. Objection to Contrary Terms: Publisher hereby objects to any terms contained in any
purchase order, acknowledgment, check endorsement or other writing prepared by you,
which terms are inconsistent with these terms and conditions or CCC's Billing and Payment
terms and conditions. These terms and conditions, together with CCC's Billing and Payment
terms and conditions (which are incorporated herein), comprise the entire agreement
between you and publisher (and CCC) concerning this licensing transaction. In the event of
any conflict between your obligations established by these terms and conditions and those
established by CCC's Billing and Payment terms and conditions, these terms and conditions
shall control.
14. Revocation: Elsevier or Copyright Clearance Center may deny the permissions described
in this License at their sole discretion, for any reason or no reason, with a full refund payable
to you. Notice of such denial will be made using the contact information provided by you.
Failure to receive such notice will not alter or invalidate the denial. In no event will Elsevier
or Copyright Clearance Center be responsible or liable for any costs, expenses or damage
incurred by you as a result of a denial of your permission request, other than a refund of the
amount(s) paid by you to Elsevier and/or Copyright Clearance Center for denied
permissions.
LIMITED LICENSE
The following terms and conditions apply only to specific license types:
15. Translation: This permission is granted for non-exclusive world English rights only
unless your license was granted for translation rights. If you licensed translation rights you
may only translate this content into the languages you requested. A professional translator
must perform all translations and reproduce the content word for word preserving the
integrity of the article.
16. Posting licensed content on any Website: The following terms and conditions apply as
follows: Licensing material from an Elsevier journal: All content posted to the web site must
maintain the copyright information line on the bottom of each image; A hyper-text must be
included to the Homepage of the journal from which you are licensing at
http://www.sciencedirect.com/science/journal/xxxxx or the Elsevier homepage for books at
http://www.elsevier.com; Central Storage: This license does not include permission for a
scanned version of the material to be stored in a central repository such as that provided by
Heron/XanEdu.
Licensing material from an Elsevier book: A hyper-text link must be included to the Elsevier
homepage at http://www.elsevier.com . All content posted to the web site must maintain the
copyright information line on the bottom of each image.

Posting licensed content on Electronic reserve: In addition to the above the following
clauses are applicable: The web site must be password-protected and made available only to
bona fide students registered on a relevant course. This permission is granted for 1 year only.
You may obtain a new license for future website posting.
17. For journal authors: the following clauses are applicable in addition to the above:
Preprints:
A preprint is an author's own write-up of research results and analysis, it has not been peer-
reviewed, nor has it had any other value added to it by a publisher (such as formatting,
copyright, technical enhancement etc.).
Authors can share their preprints anywhere at any time. Preprints should not be added to or
enhanced in any way in order to appear more like, or to substitute for, the final versions of
articles however authors can update their preprints on arXiv or RePEc with their Accepted
scan the printed edition to create an electronic version. Posting to a repository: Authors are
permitted to post a summary of their chapter only in their institution's repository.
19. Thesis/Dissertation: If your license is for use in a thesis/dissertation your thesis may be
submitted to your institution in either print or electronic form. Should your thesis be
published commercially, please reapply for permission. These requirements include
permission for the Library and Archives of Canada to supply single copies, on demand, of
the complete thesis and include permission for Proquest/UMI to supply single copies, on
demand, of the complete thesis. Should your thesis be published commercially, please
reapply for permission. Theses and dissertations which contain embedded PJAs as part of the
formal submission can be posted publicly by the awarding institution with DOI links back to
the formal publications on ScienceDirect.

Elsevier Open Access Terms and Conditions
You can publish open access with Elsevier in hundreds of open access journals or in nearly
2000 established subscription journals that support open access publishing. Permitted third
party re-use of these open access articles is defined by the author's choice of Creative
Commons user license. See our open access license policy for more information.
Terms & Conditions applicable to all Open Access articles published with Elsevier:
Any reuse of the article must not represent the author as endorsing the adaptation of the
article nor should the article be modified in such a way as to damage the author's honour or
reputation. If any changes have been made, such changes must be clearly indicated.
The author(s) must be appropriately credited and we ask that you include the end user
license and a DOI link to the formal publication on ScienceDirect.
If any part of the material to be used (for example, figures) has appeared in our publication
with credit or acknowledgement to another source it is the responsibility of the user to ensure
their reuse complies with the terms and conditions determined by the rights holder.
Additional Terms & Conditions applicable to each Creative Commons user license:
CC BY: The CC-BY license allows users to copy, to create extracts, abstracts and new
works from the Article, to alter and revise the Article and to make commercial use of the
Article (including reuse and/or resale of the Article by commercial entities), provided the
user gives appropriate credit (with a link to the formal publication through the relevant
DOI), provides a link to the license, indicates if changes were made and the licensor is not
represented as endorsing the use made of the work. The full details of the license are
available at http://creativecommons.org/licenses/by/4.0.
CC BY NC SA: The CC BY-NC-SA license allows users to copy, to create extracts,
abstracts and new works from the Article, to alter and revise the Article, provided this is not
done for commercial purposes, and that the user gives appropriate credit (with a link to the
formal publication through the relevant DOI), provides a link to the license, indicates if
changes were made and the licensor is not represented as endorsing the use made of the
work. Further, any new works must be made available on the same conditions. The full
details of the license are available at http://creativecommons.org/licenses/by-nc-sa/4.0.
CC BY NC ND: The CC BY-NC-ND license allows users to copy and distribute the Article,
provided this is not done for commercial purposes and further does not permit distribution of
the Article if it is changed or edited in any way, and provided the user gives appropriate
credit (with a link to the formal publication through the relevant DOI), provides a link to the
license, and that the licensor is not represented as endorsing the use made of the work. The
full details of the license are available at http://creativecommons.org/licenses/by-nc-nd/4.0.
Any commercial reuse of Open Access articles published with a CC BY NC SA or CC BY
NC ND license requires permission from Elsevier and will be subject to a fee.
Commercial reuse includes:

- Associating advertising with the full text of the Article

- Charging fees for document delivery or access
- Article aggregation
- Systematic distribution via e-mail lists or share buttons
3/20/2016 RightsLink Printable License

ELSEVIER LICENSE
TERMS AND CONDITIONS
Jan 08, 2016

This is a License Agreement between Thomas Edwards ("You") and Elsevier ("Elsevier")
provided by Copyright Clearance Center ("CCC"). The license consists of your order details,
the terms and conditions provided by Elsevier, and the payment terms and conditions.

All payments must be made in full to CCC. For payment instructions, please see
information listed at the bottom of this form.

Supplier Elsevier Limited
The Boulevard,Langford Lane
Kidlington,Oxford,OX5 1GB,UK
Registered Company 1982084
Number

Customer name Thomas Edwards

Customer address Imperial College London

  London, SW7 2AZ
License number 3783881008503

License date Jan 07, 2016

Licensed content publisher Elsevier
Licensed content publication Solar Energy Materials and Solar Cells

Licensed content title A review on highly ordered, vertically oriented TiO2 nanotube
arrays: Fabrication, material properties, and solar energy
applications

Licensed content author Gopal K. Mor,Oomman K. Varghese,Maggie Paulose,Karthik
Shankar,Craig A. Grimes
Licensed content date 6 September 2006

Licensed content volume 90
number
Licensed content issue 14
number

Number of pages 65
Start Page 2011

End Page 2075
Type of Use reuse in a thesis/dissertation
Portion figures/tables/illustrations

Number of 2
figures/tables/illustrations

Format both print and electronic
Are you the author of this No
Elsevier article?

Will you be translating? No
Original figure numbers Figures 19a and 20
Title of your Optimisation of Coated Nanowire Solar Cells by Simulation

file:///Users/tre11/Dropbox%20(Personal)/Tom_papers/thesis/permissions/obtained_permissions/%20%20A%20review%20on%20highly%20ordered,%20vertica… 1/6
3/20/2016 RightsLink Printable License
thesis/dissertation
Expected completion date Nov 2015

Estimated size (number of 170
pages)
Elsevier VAT number GB 494 6272 12

Permissions price 0.00 USD
VAT/Local Sales Tax 0.00 USD / 0.00 GBP
Total 0.00 USD

Terms and Conditions

INTRODUCTION
1. The publisher for this copyrighted material is Elsevier.  By clicking "accept" in connection
with completing this licensing transaction, you agree that the following terms and conditions
apply to this transaction (along with the Billing and Payment terms and conditions
established by Copyright Clearance Center, Inc. ("CCC"), at the time that you opened your
Rightslink account and that are available at any time at http://myaccount.copyright.com).
GENERAL TERMS
2. Elsevier hereby grants you permission to reproduce the aforementioned material subject to
the terms and conditions indicated.
3. Acknowledgement: If any part of the material to be used (for example, figures) has
appeared in our publication with credit or acknowledgement to another source, permission
must also be sought from that source.  If such permission is not obtained then that material
may not be included in your publication/copies. Suitable acknowledgement to the source
must be made, either as a footnote or in a reference list at the end of your publication, as
follows:
"Reprinted from Publication title, Vol /edition number, Author(s), Title of article / title of
chapter, Pages No., Copyright (Year), with permission from Elsevier [OR APPLICABLE
SOCIETY COPYRIGHT OWNER]." Also Lancet special credit - "Reprinted from The
Lancet, Vol. number, Author(s), Title of article, Pages No., Copyright (Year), with
permission from Elsevier."
4. Reproduction of this material is confined to the purpose and/or media for which
permission is hereby given.
5. Altering/Modifying Material: Not Permitted. However figures and illustrations may be
altered/adapted minimally to serve your work. Any other abbreviations, additions, deletions
and/or any other alterations shall be made only with prior written authorization of Elsevier
Ltd. (Please contact Elsevier at permissions@elsevier.com)
6. If the permission fee for the requested use of our material is waived in this instance,
please be advised that your future requests for Elsevier materials may attract a fee.
7. Reservation of Rights: Publisher reserves all rights not specifically granted in the
combination of (i) the license details provided by you and accepted in the course of this
licensing transaction, (ii) these terms and conditions and (iii) CCC's Billing and Payment
terms and conditions.
8. License Contingent Upon Payment: While you may exercise the rights licensed
immediately upon issuance of the license at the end of the licensing process for the
transaction, provided that you have disclosed complete and accurate details of your proposed
use, no license is finally effective unless and until full payment is received from you (either
by publisher or by CCC) as provided in CCC's Billing and Payment terms and conditions.  If
full payment is not received on a timely basis, then any license preliminarily granted shall be
deemed automatically revoked and shall be void as if never granted.  Further, in the event
that you breach any of these terms and conditions or any of CCC's Billing and Payment
terms and conditions, the license is automatically revoked and shall be void as if never
granted.  Use of materials as described in a revoked license, as well as any use of the

file:///Users/tre11/Dropbox%20(Personal)/Tom_papers/thesis/permissions/obtained_permissions/%20%20A%20review%20on%20highly%20ordered,%20vertica… 2/6
3/20/2016 RightsLink Printable License

materials beyond the scope of an unrevoked license, may constitute copyright infringement
and publisher reserves the right to take any and all action to protect its copyright in the
materials.
9. Warranties: Publisher makes no representations or warranties with respect to the licensed
material.
10. Indemnity: You hereby indemnify and agree to hold harmless publisher and CCC, and
their respective officers, directors, employees and agents, from and against any and all
claims arising out of your use of the licensed material other than as specifically authorized
pursuant to this license.
11. No Transfer of License: This license is personal to you and may not be sublicensed,
assigned, or transferred by you to any other person without publisher's written permission.
12. No Amendment Except in Writing: This license may not be amended except in a writing
signed by both parties (or, in the case of publisher, by CCC on publisher's behalf).
13. Objection to Contrary Terms: Publisher hereby objects to any terms contained in any
purchase order, acknowledgment, check endorsement or other writing prepared by you,
which terms are inconsistent with these terms and conditions or CCC's Billing and Payment
terms and conditions.  These terms and conditions, together with CCC's Billing and Payment
terms and conditions (which are incorporated herein), comprise the entire agreement
between you and publisher (and CCC) concerning this licensing transaction.  In the event of
any conflict between your obligations established by these terms and conditions and those
established by CCC's Billing and Payment terms and conditions, these terms and conditions
shall control.
14. Revocation: Elsevier or Copyright Clearance Center may deny the permissions described
in this License at their sole discretion, for any reason or no reason, with a full refund payable
to you.  Notice of such denial will be made using the contact information provided by you. 
Failure to receive such notice will not alter or invalidate the denial.  In no event will Elsevier
or Copyright Clearance Center be responsible or liable for any costs, expenses or damage
incurred by you as a result of a denial of your permission request, other than a refund of the
amount(s) paid by you to Elsevier and/or Copyright Clearance Center for denied
permissions.
LIMITED LICENSE
The following terms and conditions apply only to specific license types:
15. Translation: This permission is granted for non-exclusive world English rights only
unless your license was granted for translation rights. If you licensed translation rights you
may only translate this content into the languages you requested. A professional translator
must perform all translations and reproduce the content word for word preserving the
integrity of the article.
16. Posting licensed content on any Website: The following terms and conditions apply as
follows: Licensing material from an Elsevier journal: All content posted to the web site must
maintain the copyright information line on the bottom of each image; A hyper-text must be
included to the Homepage of the journal from which you are licensing at
http://www.sciencedirect.com/science/journal/xxxxx or the Elsevier homepage for books at
http://www.elsevier.com; Central Storage: This license does not include permission for a
scanned version of the material to be stored in a central repository such as that provided by
Heron/XanEdu.
Licensing material from an Elsevier book: A hyper-text link must be included to the Elsevier
homepage at http://www.elsevier.com . All content posted to the web site must maintain the
copyright information line on the bottom of each image.

Posting licensed content on Electronic reserve: In addition to the above the following
clauses are applicable: The web site must be password-protected and made available only to
bona fide students registered on a relevant course. This permission is granted for 1 year only.
You may obtain a new license for future website posting.
17. For journal authors: the following clauses are applicable in addition to the above:
file:///Users/tre11/Dropbox%20(Personal)/Tom_papers/thesis/permissions/obtained_permissions/%20%20A%20review%20on%20highly%20ordered,%20vertica… 3/6
3/20/2016 RightsLink Printable License

for classroom teaching and internal training at the institution (including use in course packs
and courseware programs), and inclusion of the article for grant funding purposes.
Gold Open Access Articles: May be shared according to the author-selected end-user
license and should contain a CrossMark logo, the end user license, and a DOI link to the
formal publication on ScienceDirect.
Please refer to Elsevier's posting policy for further information.
18. For book authors the following clauses are applicable in addition to the above:  
Authors are permitted to place a brief summary of their work online only. You are not
allowed to download and post the published electronic version of your chapter, nor may you
scan the printed edition to create an electronic version. Posting to a repository: Authors are
permitted to post a summary of their chapter only in their institution's repository.
19. Thesis/Dissertation: If your license is for use in a thesis/dissertation your thesis may be
submitted to your institution in either print or electronic form. Should your thesis be
published commercially, please reapply for permission. These requirements include
permission for the Library and Archives of Canada to supply single copies, on demand, of
the complete thesis and include permission for Proquest/UMI to supply single copies, on
demand, of the complete thesis. Should your thesis be published commercially, please
reapply for permission. Theses and dissertations which contain embedded PJAs as part of
the formal submission can be posted publicly by the awarding institution with DOI links
back to the formal publications on ScienceDirect.
 
Elsevier Open Access Terms and Conditions
You can publish open access with Elsevier in hundreds of open access journals or in nearly
2000 established subscription journals that support open access publishing. Permitted third
party re-use of these open access articles is defined by the author's choice of Creative
Commons user license. See our open access license policy for more information.
Terms & Conditions applicable to all Open Access articles published with Elsevier:
Any reuse of the article must not represent the author as endorsing the adaptation of the
article nor should the article be modified in such a way as to damage the author's honour or
reputation. If any changes have been made, such changes must be clearly indicated.
The author(s) must be appropriately credited and we ask that you include the end user
license and a DOI link to the formal publication on ScienceDirect.
If any part of the material to be used (for example, figures) has appeared in our publication
with credit or acknowledgement to another source it is the responsibility of the user to
ensure their reuse complies with the terms and conditions determined by the rights holder.
Additional Terms & Conditions applicable to each Creative Commons user license:
CC BY: The CC-BY license allows users to copy, to create extracts, abstracts and new
works from the Article, to alter and revise the Article and to make commercial use of the
Article (including reuse and/or resale of the Article by commercial entities), provided the
user gives appropriate credit (with a link to the formal publication through the relevant
DOI), provides a link to the license, indicates if changes were made and the licensor is not
represented as endorsing the use made of the work. The full details of the license are
available at http://creativecommons.org/licenses/by/4.0.
CC BY NC SA: The CC BY-NC-SA license allows users to copy, to create extracts,
abstracts and new works from the Article, to alter and revise the Article, provided this is not
done for commercial purposes, and that the user gives appropriate credit (with a link to the
formal publication through the relevant DOI), provides a link to the license, indicates if
changes were made and the licensor is not represented as endorsing the use made of the
work. Further, any new works must be made available on the same conditions. The full
details of the license are available at http://creativecommons.org/licenses/by-nc-sa/4.0.
CC BY NC ND: The CC BY-NC-ND license allows users to copy and distribute the Article,
provided this is not done for commercial purposes and further does not permit distribution of
the Article if it is changed or edited in any way, and provided the user gives appropriate
credit (with a link to the formal publication through the relevant DOI), provides a link to the
file:///Users/tre11/Dropbox%20(Personal)/Tom_papers/thesis/permissions/obtained_permissions/%20%20A%20review%20on%20highly%20ordered,%20vertica… 5/6
 ROYAL SOCIETY OF CHEMISTRY LICENSE
TERMS AND CONDITIONS
Jan 07, 2016

This Agreement between Thomas Edwards ("You") and Royal Society of Chemistry ("Royal
Society of Chemistry") consists of your license details and the terms and conditions provided
by Royal Society of Chemistry and Copyright Clearance Center.

License Number 3773320720072

License date Dec 20, 2015

Licensed Content Publisher Royal Society of Chemistry

Licensed Content Publication Nanoscale

Licensed Content Title All-solid-state hybrid solar cells based on a new organometal halide
perovskite sensitizer and one-dimensional TiO2 nanowire arrays

Licensed Content Author Jianhang Qiu,Yongcai Qiu,Keyou Yan,Min Zhong,Cheng Mu,He

Yan,Shihe Yang

Licensed Content Date Mar 4, 2013

Licensed Content Volume 5
Number

Licensed Content Issue 8

Number

Type of Use Thesis/Dissertation

Requestor type academic/educational

Portion figures/tables/images

Number of 1
figures/tables/images

Format print and electronic

Distribution quantity 10

Will you be translating? no

Order reference number None

Title of the Optimisation of Coated Nanowire Solar Cells by Simulation
thesis/dissertation

Expected completion date Nov 2015

Estimated size 170

Requestor Location Thomas Edwards

Imperial College London
Exhibition Road
South Kensington
London, United Kingdom SW7 2AZ
Attn: Thomas Edwards

Billing Type Invoice

Billing Address Thomas Edwards

Imperial College London
Exhibition Road
South Kensington
London, United Kingdom SW7 2AZ
Attn: Thomas Edwards

Total 0.00 USD

Terms and Conditions

This License Agreement is between {Requestor Name} (“You”) and The Royal Society of
Chemistry (“RSC”) provided by the Copyright Clearance Center (“CCC”). The license consists of
your order details, the terms and conditions provided by the Royal Society of Chemistry, and the
payment terms and conditions.
RSC / TERMS AND CONDITIONS
INTRODUCTION
The publisher for this copyrighted material is The Royal Society of Chemistry. By clicking
“accept” in connection with completing this licensing transaction, you agree that the
following terms and conditions apply to this transaction (along with the Billing and Payment
terms and conditions established by CCC, at the time that you opened your RightsLink
account and that are available at any time at .
LICENSE GRANTED
The RSC hereby grants you a non-exclusive license to use the aforementioned material
anywhere in the world subject to the terms and conditions indicated herein. Reproduction of
the material is confined to the purpose and/or media for which permission is hereby given.
RESERVATION OF RIGHTS
The RSC reserves all rights not specifically granted in the combination of (i) the license
details provided by your and accepted in the course of this licensing transaction; (ii) these
terms and conditions; and (iii) CCC’s Billing and Payment terms and conditions.
REVOCATION
The RSC reserves the right to revoke this license for any reason, including, but not limited
to, advertising and promotional uses of RSC content, third party usage, and incorrect source
figure attribution.
THIRD-PARTY MATERIAL DISCLAIMER
If part of the material to be used (for example, a figure) has appeared in the RSC publication
with credit to another source, permission must also be sought from that source. If the other
source is another RSC publication these details should be included in your RightsLink
request. If the other source is a third party, permission must be obtained from the third party.
The RSC disclaims any responsibility for the reproduction you make of items owned by a
third party.
PAYMENT OF FEE
If the permission fee for the requested material is waived in this instance, please be advised
that any future requests for the reproduction of RSC materials may attract a fee.
ACKNOWLEDGEMENT
The reproduction of the licensed material must be accompanied by the following
acknowledgement:
Reproduced (“Adapted” or “in part”) from {Reference Citation} (or Ref XX) with
permission of The Royal Society of Chemistry.
If the licensed material is being reproduced from New Journal of Chemistry (NJC),
Photochemical & Photobiological Sciences (PPS) or Physical Chemistry Chemical Physics
(PCCP) you must include one of the following acknowledgements:
For figures originally published in NJC:
Reproduced (“Adapted” or “in part”) from {Reference Citation} (or Ref XX) with
permission of The Royal Society of Chemistry (RSC) on behalf of the European Society for
Photobiology, the European Photochemistry Association and the RSC.
For figures originally published in PPS:
Reproduced (“Adapted” or “in part”) from {Reference Citation} (or Ref XX) with
permission of The Royal Society of Chemistry (RSC) on behalf of the Centre National de la
Recherche Scientifique (CNRS) and the RSC.
For figures originally published in PCCP:
Reproduced (“Adapted” or “in part”) from {Reference Citation} (or Ref XX) with
permission of the PCCP Owner Societies.
HYPERTEXT LINKS
With any material which is being reproduced in electronic form, you must include a
hypertext link to the original RSC article on the RSC’s website. The recommended form for
the hyperlink is http://dx.doi.org/10.1039/DOI suffix, for example in the link
http://dx.doi.org/10.1039/b110420a the DOI suffix is ‘b110420a’. To find the relevant DOI
suffix for the RSC article in question, go to the Journals section of the website and locate the
article in the list of papers for the volume and issue of your specific journal. You will find
the DOI suffix quoted there.
LICENSE CONTINGENT ON PAYMENT
While you may exercise the rights licensed immediately upon issuance of the license at the
end of the licensing process for the transaction, provided that you have disclosed complete
and accurate details of your proposed use, no license is finally effective unless and until full
payment is received from you (by CCC) as provided in CCC's Billing and Payment terms
and conditions. If full payment is not received on a timely basis, then any license
preliminarily granted shall be deemed automatically revoked and shall be void as if never
granted. Further, in the event that you breach any of these terms and conditions or any of
CCC's Billing and Payment terms and conditions, the license is automatically revoked and
shall be void as if never granted. Use of materials as described in a revoked license, as well
as any use of the materials beyond the scope of an unrevoked license, may constitute
copyright infringement and the RSC reserves the right to take any and all action to protect its
copyright in the materials.
WARRANTIES
The RSC makes no representations or warranties with respect to the licensed material.
INDEMNITY
You hereby indemnify and agree to hold harmless the RSC and the CCC, and their
respective officers, directors, trustees, employees and agents, from and against any and all
claims arising out of your use of the licensed material other than as specifically authorized
pursuant to this licence.
NO TRANSFER OF LICENSE
This license is personal to you or your publisher and may not be sublicensed, assigned, or
transferred by you to any other person without the RSC's written permission.
NO AMENDMENT EXCEPT IN WRITING
This license may not be amended except in a writing signed by both parties (or, in the case
of “Other Conditions, v1.2”, by CCC on the RSC's behalf).
OBJECTION TO CONTRARY TERMS
You hereby acknowledge and agree that these terms and conditions, together with CCC's
Billing and Payment terms and conditions (which are incorporated herein), comprise the
entire agreement between you and the RSC (and CCC) concerning this licensing transaction,
to the exclusion of all other terms and conditions, written or verbal, express or implied
(including any terms contained in any purchase order, acknowledgment, check endorsement
or other writing prepared by you). In the event of any conflict between your obligations
established by these terms and conditions and those established by CCC's Billing and
Payment terms and conditions, these terms and conditions shall control.
JURISDICTION
This license transaction shall be governed by and construed in accordance with the laws of
the District of Columbia. You hereby agree to submit to the jurisdiction of the courts located
in the District of Columbia for purposes of resolving any disputes that may arise in
connection with this licensing transaction.
LIMITED LICENSE
The following terms and conditions apply to specific license types:
Translation
This permission is granted for non-exclusive world English rights only unless your license
was granted for translation rights. If you licensed translation rights you may only translate
this content into the languages you requested. A professional translator must perform all
translations and reproduce the content word for word preserving the integrity of the article.
Intranet
If the licensed material is being posted on an Intranet, the Intranet is to be password-
protected and made available only to bona fide students or employees only. All content
posted to the Intranet must maintain the copyright information line on the bottom of each
image. You must also fully reference the material and include a hypertext link as specified
above.
Copies of Whole Articles
All copies of whole articles must maintain, if available, the copyright information line on the
bottom of each page.
Other Conditions
v1.2
Gratis licenses (referencing $0 in the Total field) are free. Please retain this printable license
for your reference. No payment is required.
If you would like to pay for this license now, please remit this license along with
yourpayment made payable to "COPYRIGHT CLEARANCE CENTER" otherwise you will
be invoiced within 48 hours of the license date. Payment should be in the form of a check or
money order referencing your account number and this invoice number {Invoice Number}.
Once you receive your invoice for this order, you may pay your invoice by credit card.
Please follow instructions provided at that time.
Make Payment To:
Copyright Clearance Center
Dept 001
P.O. Box 843006
Boston, MA 02284-3006
For suggestions or comments regarding this order, contact Rightslink Customer Support:
customercare@copyright.com or +1-855-239-3415 (toll free in the US) or +1-978-646-2777.
Questions? customercare@copyright.com or +1-855-239-3415 (toll free in the US) or
+1-978-646-2777.
 AIP PUBLISHING LLC LICENSE
TERMS AND CONDITIONS
Jan 07, 2016
All payments must be made in full to CCC. For payment instructions, please see
information listed at the bottom of this form.

License Number 3773311236709

Order Date Dec 20, 2015

Publisher AIP Publishing LLC

Publication Journal of Applied Physics

Article Title CdS/CdSe cosensitized oriented single-crystalline TiO2 nanowire
array for solar cell application
Author Ming Li,Yong Liu,Hai Wang, et al.

Online Publication Date Nov 2, 2010

Volume number 108

Issue number 9
Type of Use Thesis/Dissertation

Requestor type Student

Format Print and electronic

Portion Figure/Table

Number of figures/tables 1

Title of your thesis / Optimisation of Coated Nanowire Solar Cells by Simulation

dissertation
Expected completion date Nov 2015

Estimated size (number of 170

pages)

Total 0.00 GBP

Terms and Conditions

AIP Publishing LLC -- Terms and Conditions: Permissions Uses

AIP Publishing LLC ("AIPP"") hereby grants to you the non-exclusive right and license to use
and/or distribute the Material according to the use specified in your order, on a one-time basis,
for the specified term, with a maximum distribution equal to the number that you have ordered.
Any links or other content accompanying the Material are not the subject of this license.

1. You agree to include the following copyright and permission notice with the reproduction of
the Material:"Reprinted with permission from [FULL CITATION]. Copyright [PUBLICATION
YEAR], AIP Publishing LLC." For an article, the copyright and permission notice must be
printed on the first page of the article or book chapter. For photographs, covers, or tables,
the copyright and permission notice may appear with the Material, in a footnote, or in the
reference list.
2. If you have licensed reuse of a figure, photograph, cover, or table, it is your responsibility
to ensure that the material is original to AIPP and does not contain the copyright of another
entity, and that the copyright notice of the figure, photograph, cover, or table does not
indicate that it was reprinted by AIPP, with permission, from another source. Under no
circumstances does AIPP, purport or intend to grant permission to reuse material to which
it does not hold copyright.
3. You may not alter or modify the Material in any manner. You may translate the Material
into another language only if you have licensed translation rights. You may not use the
Material for promotional purposes. AIPP reserves all rights not specifically granted herein.
4. The foregoing license shall not take effect unless and until AIPP or its agent, Copyright
Clearance Center, receives the Payment in accordance with Copyright Clearance Center
Billing and Payment Terms and Conditions, which are incorporated herein by reference.
5. AIPP or the Copyright Clearance Center may, within two business days of granting this
license, revoke the license for any reason whatsoever, with a full refund payable to you.
Should you violate the terms of this license at any time, AIPP, AIP Publishing LLC, or
 Copyright Clearance Center may revoke the license with no refund to you. Notice of such
revocation will be made using the contact information provided by you. Failure to receive
such notice will not nullify the revocation.
6. AIPP makes no representations or warranties with respect to the Material. You agree to
indemnify and hold harmless AIPP, AIP Publishing LLC, and their officers, directors,
employees or agents from and against any and all claims arising out of your use of the
Material other than as specifically authorized herein.
7. The permission granted herein is personal to you and is not transferable or assignable
without the prior written permission of AIPP. This license may not be amended except in a
writing signed by the party to be charged.
8. If purchase orders, acknowledgments or check endorsements are issued on any forms
containing terms and conditions which are inconsistent with these provisions, such
inconsistent terms and conditions shall be of no force and effect. This document, including
the CCC Billing and Payment Terms and Conditions, shall be the entire agreement between
the parties relating to the subject matter hereof.

This Agreement shall be governed by and construed in accordance with the laws of the State
of New York. Both parties hereby submit to the jurisdiction of the courts of New York
County for purposes of resolving any disputes that may arise hereunder.
Questions? customercare@copyright.com or +1-855-239-3415 (toll free in the US) or
+1-978-646-2777.
 ELSEVIER LICENSE
TERMS AND CONDITIONS
Jan 07, 2016

This is a License Agreement between Thomas Edwards ("You") and Elsevier ("Elsevier")
provided by Copyright Clearance Center ("CCC"). The license consists of your order details,
the terms and conditions provided by Elsevier, and the payment terms and conditions.

All payments must be made in full to CCC. For payment instructions, please see
information listed at the bottom of this form.

Supplier Elsevier Limited

The Boulevard,Langford Lane
Kidlington,Oxford,OX5 1GB,UK
Registered Company 1982084
Number

Customer name Thomas Edwards

Customer address Imperial College London

London, SW7 2AZ

License number 3772510912712

License date Dec 19, 2015

Licensed content publisher Elsevier

Licensed content publication Thin Solid Films

Licensed content title Comparison of different thin film absorbers used in eta-solar cells
Licensed content author A. Belaidi,R. Bayón,L. Dloczik,K. Ernst,M.Ch. Lux-Steiner,R.
Könenkamp

Licensed content date 1 May 2003

Licensed content volume 431

number

Licensed content issue n/a

number
Number of pages 4

Start Page 488

End Page 491

Type of Use reuse in a thesis/dissertation

Intended publisher of new other

work

Portion figures/tables/illustrations
Number of 1
figures/tables/illustrations

Format both print and electronic

Are you the author of this No

Elsevier article?

Will you be translating? No

Original figure numbers Figure 2

Title of your Optimisation of Coated Nanowire Solar Cells by Simulation

thesis/dissertation

Expected completion date Nov 2015

Estimated size (number of 170
pages)
Elsevier VAT number GB 494 6272 12

Permissions price 0.00 GBP

VAT/Local Sales Tax 0.00 GBP / 0.00 GBP

Total 0.00 GBP

Terms and Conditions

INTRODUCTION
1. The publisher for this copyrighted material is Elsevier. By clicking "accept" in
connection with completing this licensing transaction, you agree that the following terms
and conditions apply to this transaction (along with the Billing and Payment terms and
conditions established by Copyright Clearance Center, Inc. ("CCC"), at the time that you
opened your Rightslink account and that are available at any time at
http://myaccount.copyright.com).
GENERAL TERMS
2. Elsevier hereby grants you permission to reproduce the aforementioned material subject to
the terms and conditions indicated.
3. Acknowledgement: If any part of the material to be used (for example, figures) has
appeared in our publication with credit or acknowledgement to another source, permission
must also be sought from that source. If such permission is not obtained then that material
may not be included in your publication/copies. Suitable acknowledgement to the source
must be made, either as a footnote or in a reference list at the end of your publication, as
follows:
"Reprinted from Publication title, Vol /edition number, Author(s), Title of article / title of
chapter, Pages No., Copyright (Year), with permission from Elsevier [OR APPLICABLE
SOCIETY COPYRIGHT OWNER]." Also Lancet special credit - "Reprinted from The
Lancet, Vol. number, Author(s), Title of article, Pages No., Copyright (Year), with
permission from Elsevier."
4. Reproduction of this material is confined to the purpose and/or media for which
permission is hereby given.
5. Altering/Modifying Material: Not Permitted. However figures and illustrations may be
altered/adapted minimally to serve your work. Any other abbreviations, additions, deletions
and/or any other alterations shall be made only with prior written authorization of Elsevier
Ltd. (Please contact Elsevier at permissions@elsevier.com)
6. If the permission fee for the requested use of our material is waived in this instance, please
be advised that your future requests for Elsevier materials may attract a fee.
7. Reservation of Rights: Publisher reserves all rights not specifically granted in the
combination of (i) the license details provided by you and accepted in the course of this
licensing transaction, (ii) these terms and conditions and (iii) CCC's Billing and Payment
terms and conditions.
8. License Contingent Upon Payment: While you may exercise the rights licensed
immediately upon issuance of the license at the end of the licensing process for the
transaction, provided that you have disclosed complete and accurate details of your proposed
use, no license is finally effective unless and until full payment is received from you (either
by publisher or by CCC) as provided in CCC's Billing and Payment terms and conditions. If
full payment is not received on a timely basis, then any license preliminarily granted shall be
deemed automatically revoked and shall be void as if never granted. Further, in the event
that you breach any of these terms and conditions or any of CCC's Billing and Payment
terms and conditions, the license is automatically revoked and shall be void as if never
granted. Use of materials as described in a revoked license, as well as any use of the
materials beyond the scope of an unrevoked license, may constitute copyright infringement
and publisher reserves the right to take any and all action to protect its copyright in the
materials.
9. Warranties: Publisher makes no representations or warranties with respect to the licensed
material.
10. Indemnity: You hereby indemnify and agree to hold harmless publisher and CCC, and
their respective officers, directors, employees and agents, from and against any and all
claims arising out of your use of the licensed material other than as specifically authorized
pursuant to this license.
11. No Transfer of License: This license is personal to you and may not be sublicensed,
assigned, or transferred by you to any other person without publisher's written permission.
12. No Amendment Except in Writing: This license may not be amended except in a writing
signed by both parties (or, in the case of publisher, by CCC on publisher's behalf).
13. Objection to Contrary Terms: Publisher hereby objects to any terms contained in any
purchase order, acknowledgment, check endorsement or other writing prepared by you,
which terms are inconsistent with these terms and conditions or CCC's Billing and Payment
terms and conditions. These terms and conditions, together with CCC's Billing and Payment
terms and conditions (which are incorporated herein), comprise the entire agreement
between you and publisher (and CCC) concerning this licensing transaction. In the event of
any conflict between your obligations established by these terms and conditions and those
established by CCC's Billing and Payment terms and conditions, these terms and conditions
shall control.
14. Revocation: Elsevier or Copyright Clearance Center may deny the permissions described
in this License at their sole discretion, for any reason or no reason, with a full refund payable
to you. Notice of such denial will be made using the contact information provided by you.
Failure to receive such notice will not alter or invalidate the denial. In no event will Elsevier
or Copyright Clearance Center be responsible or liable for any costs, expenses or damage
incurred by you as a result of a denial of your permission request, other than a refund of the
amount(s) paid by you to Elsevier and/or Copyright Clearance Center for denied
permissions.
LIMITED LICENSE
The following terms and conditions apply only to specific license types:
15. Translation: This permission is granted for non-exclusive world English rights only
unless your license was granted for translation rights. If you licensed translation rights you
may only translate this content into the languages you requested. A professional translator
must perform all translations and reproduce the content word for word preserving the
integrity of the article.
16. Posting licensed content on any Website: The following terms and conditions apply as
follows: Licensing material from an Elsevier journal: All content posted to the web site must
maintain the copyright information line on the bottom of each image; A hyper-text must be
included to the Homepage of the journal from which you are licensing at
http://www.sciencedirect.com/science/journal/xxxxx or the Elsevier homepage for books at
http://www.elsevier.com; Central Storage: This license does not include permission for a
scanned version of the material to be stored in a central repository such as that provided by
Heron/XanEdu.
Licensing material from an Elsevier book: A hyper-text link must be included to the Elsevier
homepage at http://www.elsevier.com . All content posted to the web site must maintain the
copyright information line on the bottom of each image.

Posting licensed content on Electronic reserve: In addition to the above the following
clauses are applicable: The web site must be password-protected and made available only to
bona fide students registered on a relevant course. This permission is granted for 1 year only.
You may obtain a new license for future website posting.
17. For journal authors: the following clauses are applicable in addition to the above:
Preprints:
A preprint is an author's own write-up of research results and analysis, it has not been peer-
reviewed, nor has it had any other value added to it by a publisher (such as formatting,
copyright, technical enhancement etc.).
Authors can share their preprints anywhere at any time. Preprints should not be added to or
enhanced in any way in order to appear more like, or to substitute for, the final versions of
articles however authors can update their preprints on arXiv or RePEc with their Accepted
scan the printed edition to create an electronic version. Posting to a repository: Authors are
permitted to post a summary of their chapter only in their institution's repository.
19. Thesis/Dissertation: If your license is for use in a thesis/dissertation your thesis may be
submitted to your institution in either print or electronic form. Should your thesis be
published commercially, please reapply for permission. These requirements include
permission for the Library and Archives of Canada to supply single copies, on demand, of
the complete thesis and include permission for Proquest/UMI to supply single copies, on
demand, of the complete thesis. Should your thesis be published commercially, please
reapply for permission. Theses and dissertations which contain embedded PJAs as part of the
formal submission can be posted publicly by the awarding institution with DOI links back to
the formal publications on ScienceDirect.

Elsevier Open Access Terms and Conditions
You can publish open access with Elsevier in hundreds of open access journals or in nearly
2000 established subscription journals that support open access publishing. Permitted third
party re-use of these open access articles is defined by the author's choice of Creative
Commons user license. See our open access license policy for more information.
Terms & Conditions applicable to all Open Access articles published with Elsevier:
Any reuse of the article must not represent the author as endorsing the adaptation of the
article nor should the article be modified in such a way as to damage the author's honour or
reputation. If any changes have been made, such changes must be clearly indicated.
The author(s) must be appropriately credited and we ask that you include the end user
license and a DOI link to the formal publication on ScienceDirect.
If any part of the material to be used (for example, figures) has appeared in our publication
with credit or acknowledgement to another source it is the responsibility of the user to ensure
their reuse complies with the terms and conditions determined by the rights holder.
Additional Terms & Conditions applicable to each Creative Commons user license:
CC BY: The CC-BY license allows users to copy, to create extracts, abstracts and new
works from the Article, to alter and revise the Article and to make commercial use of the
Article (including reuse and/or resale of the Article by commercial entities), provided the
user gives appropriate credit (with a link to the formal publication through the relevant
DOI), provides a link to the license, indicates if changes were made and the licensor is not
represented as endorsing the use made of the work. The full details of the license are
available at http://creativecommons.org/licenses/by/4.0.
CC BY NC SA: The CC BY-NC-SA license allows users to copy, to create extracts,
abstracts and new works from the Article, to alter and revise the Article, provided this is not
done for commercial purposes, and that the user gives appropriate credit (with a link to the
formal publication through the relevant DOI), provides a link to the license, indicates if
changes were made and the licensor is not represented as endorsing the use made of the
work. Further, any new works must be made available on the same conditions. The full
details of the license are available at http://creativecommons.org/licenses/by-nc-sa/4.0.
CC BY NC ND: The CC BY-NC-ND license allows users to copy and distribute the Article,
provided this is not done for commercial purposes and further does not permit distribution of
the Article if it is changed or edited in any way, and provided the user gives appropriate
credit (with a link to the formal publication through the relevant DOI), provides a link to the
license, and that the licensor is not represented as endorsing the use made of the work. The
full details of the license are available at http://creativecommons.org/licenses/by-nc-nd/4.0.
Any commercial reuse of Open Access articles published with a CC BY NC SA or CC BY
NC ND license requires permission from Elsevier and will be subject to a fee.
Commercial reuse includes:

- Associating advertising with the full text of the Article

- Charging fees for document delivery or access
- Article aggregation
- Systematic distribution via e-mail lists or share buttons
 AIP PUBLISHING LLC LICENSE
TERMS AND CONDITIONS
Jan 07, 2016
All payments must be made in full to CCC. For payment instructions, please see
information listed at the bottom of this form.

License Number 3773521187219

Order Date Dec 21, 2015

Publisher AIP Publishing LLC

Publication Journal of Applied Physics

Article Title Comparison of the device physics principles of planar and radial p-n
junction nanorod solar cells
Author Brendan M. Kayes,Harry A. Atwater,Nathan S. Lewis

Online Publication Date May 23, 2005

Volume number 97

Issue number 11
Type of Use Thesis/Dissertation

Requestor type Student

Format Print and electronic

Portion Figure/Table

Number of figures/tables 2

Title of your thesis / Optimisation of Coated Nanowire Solar Cells by Simulation

dissertation
Expected completion date Nov 2015

Estimated size (number of 170

pages)

Total 0.00 GBP

Terms and Conditions

AIP Publishing LLC -- Terms and Conditions: Permissions Uses

AIP Publishing LLC ("AIPP"") hereby grants to you the non-exclusive right and license to use
and/or distribute the Material according to the use specified in your order, on a one-time basis,
for the specified term, with a maximum distribution equal to the number that you have ordered.
Any links or other content accompanying the Material are not the subject of this license.

1. You agree to include the following copyright and permission notice with the reproduction of
the Material:"Reprinted with permission from [FULL CITATION]. Copyright [PUBLICATION
YEAR], AIP Publishing LLC." For an article, the copyright and permission notice must be
printed on the first page of the article or book chapter. For photographs, covers, or tables,
the copyright and permission notice may appear with the Material, in a footnote, or in the
reference list.
2. If you have licensed reuse of a figure, photograph, cover, or table, it is your responsibility
to ensure that the material is original to AIPP and does not contain the copyright of another
entity, and that the copyright notice of the figure, photograph, cover, or table does not
indicate that it was reprinted by AIPP, with permission, from another source. Under no
circumstances does AIPP, purport or intend to grant permission to reuse material to which
it does not hold copyright.
3. You may not alter or modify the Material in any manner. You may translate the Material
into another language only if you have licensed translation rights. You may not use the
Material for promotional purposes. AIPP reserves all rights not specifically granted herein.
4. The foregoing license shall not take effect unless and until AIPP or its agent, Copyright
Clearance Center, receives the Payment in accordance with Copyright Clearance Center
Billing and Payment Terms and Conditions, which are incorporated herein by reference.
5. AIPP or the Copyright Clearance Center may, within two business days of granting this
license, revoke the license for any reason whatsoever, with a full refund payable to you.
Should you violate the terms of this license at any time, AIPP, AIP Publishing LLC, or
 Copyright Clearance Center may revoke the license with no refund to you. Notice of such
revocation will be made using the contact information provided by you. Failure to receive
such notice will not nullify the revocation.
6. AIPP makes no representations or warranties with respect to the Material. You agree to
indemnify and hold harmless AIPP, AIP Publishing LLC, and their officers, directors,
employees or agents from and against any and all claims arising out of your use of the
Material other than as specifically authorized herein.
7. The permission granted herein is personal to you and is not transferable or assignable
without the prior written permission of AIPP. This license may not be amended except in a
writing signed by the party to be charged.
8. If purchase orders, acknowledgments or check endorsements are issued on any forms
containing terms and conditions which are inconsistent with these provisions, such
inconsistent terms and conditions shall be of no force and effect. This document, including
the CCC Billing and Payment Terms and Conditions, shall be the entire agreement between
the parties relating to the subject matter hereof.

This Agreement shall be governed by and construed in accordance with the laws of the State
of New York. Both parties hereby submit to the jurisdiction of the courts of New York
County for purposes of resolving any disputes that may arise hereunder.
Questions? customercare@copyright.com or +1-855-239-3415 (toll free in the US) or
+1-978-646-2777.
Confirmation Number: 11524856 Print this page
Order Date: 01/08/2016

Customer Information

Customer: Thomas Edwards

Account Number: 3000972200
Organization: Thomas Edwards
Email: thomas.edwards11@imperial.ac.uk
Phone: +44 123456789
Payment Method: Invoice

This is not an invoice

Order Details

Computational electrodynamics : the finite-difference time-domain method Billing Status:

N/A
Order detail ID: 69491682
Permission Status: Granted
ISBN: 978-0-89006-792-5
Permission type: Republish or display content
Publication Type: Book
Type of use: Thesis/Dissertation
Publisher: ARTECH HOUSE,
Order License Id: 3784161013655
Author/Editor: TAFLOVE, ALLEN
Requestor type Academic institution
Format Print, Electronic
Portion chart/graph/table/figure
Number of
charts/graphs/tables/fi 1
gures
Title or numeric
reference of the Figure 3.2
portion(s)
Title of the article or
chapter the portion is Chapter 3
from
Editor of portion(s) N/A
Author of portion(s) N/A
Volume of serial or
N/A
monograph
Page range of portion 61
Publication date of
2005
portion
Rights for Main product
Duration of use Life of current edition
Creation of copies for
no
the disabled
With minor editing
no
privileges
U.K. and Commonwealth
For distribution to
(excluding Canada)
In the following Original language of
language(s) publication
With incidental
no
promotional use
Lifetime unit quantity of
Up to 499
new product
Made available in the
Academic
following markets
The requesting Thomas Edwards, Imperial
person/organization College London
Order reference number
Author/Editor Thomas Edwards
 The standard identifier N/A
Optimisation of Coated
Title Nanowire Solar Cells by
Simulation
Publisher Imperial College London
Expected publication
Jan 2016
date
Estimated size (pages) 170

$
Note: This item was invoiced separately through our RightsLink service. More info 0.00

Total order items: 1 Order Total: $0.00

About Us | Privacy Policy | Terms & Conditions | Pay an Invoice

Copyright 2016 Copyright Clearance Center

 Title: Strong Geometrical Dependence Logged in as:
of the Absorption of Light in Thomas Edwards
Arrays of Semiconductor Account #:
Nanowires 3000972200

Author: Silke L. Diedenhofen, Olaf T. A.

Janssen, Grzegorz Grzela, et al
Publication: ACS Nano
Publisher: American Chemical Society
Date: Mar 1, 2011
Copyright © 2011, American Chemical Society

PERMISSION/LICENSE IS GRANTED FOR YOUR ORDER AT NO CHARGE

This type of permission/license, instead of the standard Terms & Conditions, is sent to you because
no fee is being charged for your order. Please note the following:

Permission is granted for your request in both print and electronic formats, and
translations.
If figures and/or tables were requested, they may be adapted or used in part.
Please print this page for your records and send a copy of it to your publisher/graduate
school.
Appropriate credit for the requested material should be given as follows: "Reprinted
(adapted) with permission from (COMPLETE REFERENCE CITATION). Copyright
(YEAR) American Chemical Society." Insert appropriate information in place of the
capitalized words.
One-time permission is granted only for the use specified in your request. No additional
uses are granted (such as derivative works or other editions). For any other uses, please
submit a new request.

If credit is given to another source for the material you requested, permission must be obtained
from that source.

Copyright © 2016 Copyright Clearance Center, Inc. All Rights Reserved. Privacy statement. Terms and Conditions .
Comments? We would like to hear from you. E-mail us at customercare@copyright.com
 ELSEVIER LICENSE
TERMS AND CONDITIONS
Jan 07, 2016

This is a License Agreement between Thomas Edwards ("You") and Elsevier ("Elsevier")
provided by Copyright Clearance Center ("CCC"). The license consists of your order details,
the terms and conditions provided by Elsevier, and the payment terms and conditions.

All payments must be made in full to CCC. For payment instructions, please see
information listed at the bottom of this form.

Supplier Elsevier Limited

The Boulevard,Langford Lane
Kidlington,Oxford,OX5 1GB,UK
Registered Company 1982084
Number

Customer name Thomas Edwards

Customer address Imperial College London

London, SW7 2AZ

License number 3742571138598

License date Nov 05, 2015

Licensed content publisher Elsevier

Licensed content publication Journal of Photochemistry and Photobiology C: Photochemistry

Reviews

Licensed content title Dye-sensitized solar cells

Licensed content author Michael Grätzel

Licensed content date 31 October 2003

Licensed content volume 4

number

Licensed content issue 2

number
Number of pages 9

Start Page 145

End Page 153

Type of Use reuse in a thesis/dissertation

Portion figures/tables/illustrations

Number of 1
figures/tables/illustrations
Format both print and electronic

Are you the author of this No

Elsevier article?

Will you be translating? No

Original figure numbers Figure 1

Title of your Optimisation of Coated Nanowire Solar Cells by Simulation

thesis/dissertation

Expected completion date Nov 2015

Estimated size (number of 170

pages)
Elsevier VAT number GB 494 6272 12

Permissions price 0.00 GBP

VAT/Local Sales Tax 0.00 GBP / 0.00 GBP

Total 0.00 GBP

Terms and Conditions

INTRODUCTION
1. The publisher for this copyrighted material is Elsevier. By clicking "accept" in
connection with completing this licensing transaction, you agree that the following terms
and conditions apply to this transaction (along with the Billing and Payment terms and
conditions established by Copyright Clearance Center, Inc. ("CCC"), at the time that you
opened your Rightslink account and that are available at any time at
http://myaccount.copyright.com).
GENERAL TERMS
2. Elsevier hereby grants you permission to reproduce the aforementioned material subject to
the terms and conditions indicated.
3. Acknowledgement: If any part of the material to be used (for example, figures) has
appeared in our publication with credit or acknowledgement to another source, permission
must also be sought from that source. If such permission is not obtained then that material
may not be included in your publication/copies. Suitable acknowledgement to the source
must be made, either as a footnote or in a reference list at the end of your publication, as
follows:
"Reprinted from Publication title, Vol /edition number, Author(s), Title of article / title of
chapter, Pages No., Copyright (Year), with permission from Elsevier [OR APPLICABLE
SOCIETY COPYRIGHT OWNER]." Also Lancet special credit - "Reprinted from The
Lancet, Vol. number, Author(s), Title of article, Pages No., Copyright (Year), with
permission from Elsevier."
4. Reproduction of this material is confined to the purpose and/or media for which
permission is hereby given.
5. Altering/Modifying Material: Not Permitted. However figures and illustrations may be
altered/adapted minimally to serve your work. Any other abbreviations, additions, deletions
and/or any other alterations shall be made only with prior written authorization of Elsevier
Ltd. (Please contact Elsevier at permissions@elsevier.com)
6. If the permission fee for the requested use of our material is waived in this instance, please
be advised that your future requests for Elsevier materials may attract a fee.
7. Reservation of Rights: Publisher reserves all rights not specifically granted in the
combination of (i) the license details provided by you and accepted in the course of this
licensing transaction, (ii) these terms and conditions and (iii) CCC's Billing and Payment
terms and conditions.
8. License Contingent Upon Payment: While you may exercise the rights licensed
immediately upon issuance of the license at the end of the licensing process for the
transaction, provided that you have disclosed complete and accurate details of your proposed
use, no license is finally effective unless and until full payment is received from you (either
by publisher or by CCC) as provided in CCC's Billing and Payment terms and conditions. If
full payment is not received on a timely basis, then any license preliminarily granted shall be
deemed automatically revoked and shall be void as if never granted. Further, in the event
that you breach any of these terms and conditions or any of CCC's Billing and Payment
terms and conditions, the license is automatically revoked and shall be void as if never
granted. Use of materials as described in a revoked license, as well as any use of the
materials beyond the scope of an unrevoked license, may constitute copyright infringement
and publisher reserves the right to take any and all action to protect its copyright in the
materials.
9. Warranties: Publisher makes no representations or warranties with respect to the licensed
material.
10. Indemnity: You hereby indemnify and agree to hold harmless publisher and CCC, and
their respective officers, directors, employees and agents, from and against any and all
claims arising out of your use of the licensed material other than as specifically authorized
pursuant to this license.
11. No Transfer of License: This license is personal to you and may not be sublicensed,
assigned, or transferred by you to any other person without publisher's written permission.
12. No Amendment Except in Writing: This license may not be amended except in a writing
signed by both parties (or, in the case of publisher, by CCC on publisher's behalf).
13. Objection to Contrary Terms: Publisher hereby objects to any terms contained in any
purchase order, acknowledgment, check endorsement or other writing prepared by you,
which terms are inconsistent with these terms and conditions or CCC's Billing and Payment
terms and conditions. These terms and conditions, together with CCC's Billing and Payment
terms and conditions (which are incorporated herein), comprise the entire agreement
between you and publisher (and CCC) concerning this licensing transaction. In the event of
any conflict between your obligations established by these terms and conditions and those
established by CCC's Billing and Payment terms and conditions, these terms and conditions
shall control.
14. Revocation: Elsevier or Copyright Clearance Center may deny the permissions described
in this License at their sole discretion, for any reason or no reason, with a full refund payable
to you. Notice of such denial will be made using the contact information provided by you.
Failure to receive such notice will not alter or invalidate the denial. In no event will Elsevier
or Copyright Clearance Center be responsible or liable for any costs, expenses or damage
incurred by you as a result of a denial of your permission request, other than a refund of the
amount(s) paid by you to Elsevier and/or Copyright Clearance Center for denied
permissions.
LIMITED LICENSE
The following terms and conditions apply only to specific license types:
15. Translation: This permission is granted for non-exclusive world English rights only
unless your license was granted for translation rights. If you licensed translation rights you
may only translate this content into the languages you requested. A professional translator
must perform all translations and reproduce the content word for word preserving the
integrity of the article.
16. Posting licensed content on any Website: The following terms and conditions apply as
follows: Licensing material from an Elsevier journal: All content posted to the web site must
maintain the copyright information line on the bottom of each image; A hyper-text must be
included to the Homepage of the journal from which you are licensing at
http://www.sciencedirect.com/science/journal/xxxxx or the Elsevier homepage for books at
http://www.elsevier.com; Central Storage: This license does not include permission for a
scanned version of the material to be stored in a central repository such as that provided by
Heron/XanEdu.
Licensing material from an Elsevier book: A hyper-text link must be included to the Elsevier
homepage at http://www.elsevier.com . All content posted to the web site must maintain the
copyright information line on the bottom of each image.

Posting licensed content on Electronic reserve: In addition to the above the following
clauses are applicable: The web site must be password-protected and made available only to
bona fide students registered on a relevant course. This permission is granted for 1 year only.
You may obtain a new license for future website posting.
17. For journal authors: the following clauses are applicable in addition to the above:
Preprints:
A preprint is an author's own write-up of research results and analysis, it has not been peer-
reviewed, nor has it had any other value added to it by a publisher (such as formatting,
copyright, technical enhancement etc.).
Authors can share their preprints anywhere at any time. Preprints should not be added to or
enhanced in any way in order to appear more like, or to substitute for, the final versions of
articles however authors can update their preprints on arXiv or RePEc with their Accepted
Author Manuscript (see below).
If accepted for publication, we encourage authors to link from the preprint to their formal
19. Thesis/Dissertation: If your license is for use in a thesis/dissertation your thesis may be
submitted to your institution in either print or electronic form. Should your thesis be
published commercially, please reapply for permission. These requirements include
permission for the Library and Archives of Canada to supply single copies, on demand, of
the complete thesis and include permission for Proquest/UMI to supply single copies, on
demand, of the complete thesis. Should your thesis be published commercially, please
reapply for permission. Theses and dissertations which contain embedded PJAs as part of the
formal submission can be posted publicly by the awarding institution with DOI links back to
the formal publications on ScienceDirect.

Elsevier Open Access Terms and Conditions
You can publish open access with Elsevier in hundreds of open access journals or in nearly
2000 established subscription journals that support open access publishing. Permitted third
party re-use of these open access articles is defined by the author's choice of Creative
Commons user license. See our open access license policy for more information.
Terms & Conditions applicable to all Open Access articles published with Elsevier:
Any reuse of the article must not represent the author as endorsing the adaptation of the
article nor should the article be modified in such a way as to damage the author's honour or
reputation. If any changes have been made, such changes must be clearly indicated.
The author(s) must be appropriately credited and we ask that you include the end user
license and a DOI link to the formal publication on ScienceDirect.
If any part of the material to be used (for example, figures) has appeared in our publication
with credit or acknowledgement to another source it is the responsibility of the user to ensure
their reuse complies with the terms and conditions determined by the rights holder.
Additional Terms & Conditions applicable to each Creative Commons user license:
CC BY: The CC-BY license allows users to copy, to create extracts, abstracts and new
works from the Article, to alter and revise the Article and to make commercial use of the
Article (including reuse and/or resale of the Article by commercial entities), provided the
user gives appropriate credit (with a link to the formal publication through the relevant
DOI), provides a link to the license, indicates if changes were made and the licensor is not
represented as endorsing the use made of the work. The full details of the license are
available at http://creativecommons.org/licenses/by/4.0.
CC BY NC SA: The CC BY-NC-SA license allows users to copy, to create extracts,
abstracts and new works from the Article, to alter and revise the Article, provided this is not
done for commercial purposes, and that the user gives appropriate credit (with a link to the
formal publication through the relevant DOI), provides a link to the license, indicates if
changes were made and the licensor is not represented as endorsing the use made of the
work. Further, any new works must be made available on the same conditions. The full
details of the license are available at http://creativecommons.org/licenses/by-nc-sa/4.0.
CC BY NC ND: The CC BY-NC-ND license allows users to copy and distribute the Article,
provided this is not done for commercial purposes and further does not permit distribution of
the Article if it is changed or edited in any way, and provided the user gives appropriate
credit (with a link to the formal publication through the relevant DOI), provides a link to the
license, and that the licensor is not represented as endorsing the use made of the work. The
full details of the license are available at http://creativecommons.org/licenses/by-nc-nd/4.0.
Any commercial reuse of Open Access articles published with a CC BY NC SA or CC BY
NC ND license requires permission from Elsevier and will be subject to a fee.
Commercial reuse includes:

- Associating advertising with the full text of the Article

- Charging fees for document delivery or access
- Article aggregation
- Systematic distribution via e-mail lists or share buttons

Posting or linking by commercial companies for use by customers of those companies.

 THE AMERICAN ASSOCIATION FOR THE ADVANCEMENT OF SCIENCE LICENSE
TERMS AND CONDITIONS
Jan 07, 2016

This is a License Agreement between Thomas Edwards ("You") and The American
Association for the Advancement of Science ("The American Association for the
Advancement of Science") provided by Copyright Clearance Center ("CCC"). The license
consists of your order details, the terms and conditions provided by The American
Association for the Advancement of Science, and the payment terms and conditions.

All payments must be made in full to CCC. For payment instructions, please see
information listed at the bottom of this form.

License Number 3773320482674

License date Dec 20, 2015
Licensed content publisher The American Association for the Advancement of Science

Licensed content publication Science

Licensed content title Efficient Hybrid Solar Cells Based on Meso-Superstructured

Organometal Halide Perovskites
Licensed content author Michael M. Lee, Joël Teuscher, Tsutomu Miyasaka, Takurou N.
Murakami, Henry J. Snaith

Licensed content date Nov 2, 2012

Volume number 338

Issue number 6107

Type of Use Thesis / Dissertation

Requestor type Scientist/individual at a research institution

Format Print and electronic

Portion Figure

Number of figures/tables 2

Order reference number None

Title of your thesis / Optimisation of Coated Nanowire Solar Cells by Simulation
dissertation

Expected completion date Nov 2015

Estimated size(pages) 170

Total 0.00 GBP

Terms and Conditions

American Association for the Advancement of Science TERMS AND CONDITIONS

Regarding your request, we are pleased to grant you non-exclusive, non-transferable
permission, to republish the AAAS material identified above in your work identified above,
subject to the terms and conditions herein. We must be contacted for permission for any uses
other than those specifically identified in your request above.
The following credit line must be printed along with the AAAS material: "From [Full
Reference Citation]. Reprinted with permission from AAAS."
All required credit lines and notices must be visible any time a user accesses any part of the
AAAS material and must appear on any printed copies and authorized user might make.
This permission does not apply to figures / photos / artwork or any other content or materials
included in your work that are credited to non-AAAS sources. If the requested material is
sourced to or references non-AAAS sources, you must obtain authorization from that source
as well before using that material. You agree to hold harmless and indemnify AAAS against
any claims arising from your use of any content in your work that is credited to non-AAAS
sources.
If the AAAS material covered by this permission was published in Science during the years
1974 - 1994, you must also obtain permission from the author, who may grant or withhold
permission, and who may or may not charge a fee if permission is granted. See original
article for author's address. This condition does not apply to news articles.
The AAAS material may not be modified or altered except that figures and tables may be
modified with permission from the author. Author permission for any such changes must be
secured prior to your use.
Whenever possible, we ask that electronic uses of the AAAS material permitted herein
include a hyperlink to the original work on AAAS's website (hyperlink may be embedded in
the reference citation).
AAAS material reproduced in your work identified herein must not account for more than
30% of the total contents of that work.
AAAS must publish the full paper prior to use of any text.
AAAS material must not imply any endorsement by the American Association for the
Advancement of Science.
This permission is not valid for the use of the AAAS and/or Science logos.
AAAS makes no representations or warranties as to the accuracy of any information
contained in the AAAS material covered by this permission, including any warranties of
merchantability or fitness for a particular purpose.
If permission fees for this use are waived, please note that AAAS reserves the right to
charge for reproduction of this material in the future.
Permission is not valid unless payment is received within sixty (60) days of the issuance of
this permission. If payment is not received within this time period then all rights granted
herein shall be revoked and this permission will be considered null and void.
In the event of breach of any of the terms and conditions herein or any of CCC's Billing and
Payment terms and conditions, all rights granted herein shall be revoked and this permission
will be considered null and void.
AAAS reserves the right to terminate this permission and all rights granted herein at its
discretion, for any purpose, at any time. In the event that AAAS elects to terminate this
permission, you will have no further right to publish, publicly perform, publicly display,
distribute or otherwise use any matter in which the AAAS content had been included, and all
fees paid hereunder shall be fully refunded to you. Notification of termination will be sent to
the contact information as supplied by you during the request process and termination shall
be immediate upon sending the notice. Neither AAAS nor CCC shall be liable for any costs,
expenses, or damages you may incur as a result of the termination of this permission, beyond
the refund noted above.
This Permission may not be amended except by written document signed by both parties.
The terms above are applicable to all permissions granted for the use of AAAS material.
Below you will find additional conditions that apply to your particular type of use.
FOR A THESIS OR DISSERTATION
If you are using figure(s)/table(s), permission is granted for use in print and electronic
versions of your dissertation or thesis. A full text article may be used in print versions only of
a dissertation or thesis.
Permission covers the distribution of your dissertation or thesis on demand by ProQuest /
UMI, provided the AAAS material covered by this permission remains in situ.
If you are an Original Author on the AAAS article being reproduced, please refer to your
License to Publish for rules on reproducing your paper in a dissertation or thesis.
FOR JOURNALS:
Permission covers both print and electronic versions of your journal article, however the
AAAS material may not be used in any manner other than within the context of your article.
FOR BOOKS/TEXTBOOKS:
If this license is to reuse figures/tables, then permission is granted for non-exclusive world
rights in all languages in both print and electronic formats (electronic formats are defined
below).
 AMERICAN PHYSICAL SOCIETY LICENSE
TERMS AND CONDITIONS
Jan 07, 2016

This Agreement between Thomas Edwards ("You") and American Physical Society
("American Physical Society") consists of your license details and the terms and conditions
provided by American Physical Society and Copyright Clearance Center.

License Number 3773530007139

License date Dec 21, 2015

Licensed Content Publisher American Physical Society

Licensed Content Publication Physical Review B

Licensed Content Title Electron transport via polaron hopping in bulk $\mathrm{Ti}
{\mathrm{O}}_{2}$: A density functional theory characterization

Licensed Content Author N. Aaron Deskins and Michel Dupuis

Licensed Content Date May 22, 2007

Licensed Content Volume 75

Number

Type of Use Thesis/Dissertation

Requestor type Student

Format Print, Electronic

Portion chart/graph/table/figure

Number of 2
charts/graphs/tables/figures

Portion description Figures 1 and 2

Rights for Main product

Duration of use Life of Current Edition

Creation of copies for the no

disabled
With minor editing privileges no

For distribution to Worldwide

In the following language(s) Original language of publication

With incidental promotional no

use

The lifetime unit quantity of 0 to 499

new product

The requesting Imperial College London

person/organization is:

Order reference number None

Title of your thesis / Optimisation of Coated Nanowire Solar Cells by Simulation

dissertation

Expected completion date Nov 2015

Expected size (number of 170

pages)
Requestor Location Thomas Edwards
Imperial College London
Exhibition Road
South Kensington
 London, United Kingdom SW7 2AZ
Attn: Thomas Edwards
Billing Type Invoice

Billing Address Thomas Edwards

Imperial College London
Exhibition Road
South Kensington
London, United Kingdom SW7 2AZ
Attn: Thomas Edwards

Total 0.00 GBP

Terms and Conditions

Terms and Conditions

The American Physical Society (APS) is pleased to grant the Requestor of this license a non-
exclusive, non-transferable permission, limited to [print and/or electronic format,
depending on what they chose], provided all criteria outlined below are followed.
1. For electronic format permissions, Requestor agrees to provide a hyperlink from the
reprinted APS material using the source material’s DOI on the web page where the work
appears. The hyperlink should use the standard DOI resolution URL,
http://dx.doi.org/{DOI}. The hyperlink may be embedded in the copyright credit line.
2. For print format permissions, Requestor agrees to print the required copyright credit line
on the first page where the material appears: "Reprinted (abstract/excerpt/figure) with
permission from [(FULL REFERENCE CITATION) as follows: Author's Names, APS
Journal Title, Volume Number, Page Number and Year of Publication.] Copyright (YEAR)
by the American Physical Society."
3. Permission granted in this license is for a one-time use and does not include permission for
any future editions, updates, databases, formats or other matters. Permission must be sought
for any additional use.
4. Use of the material does not and must not imply any endorsement by APS.
5. Under no circumstance does APS purport or intend to grant permission to reuse materials
to which it does not hold copyright. It is the requestors sole responsibility to ensure the
licensed material is original to APS and does not contain the copyright of another entity, and
that the copyright notice of the figure, photograph, cover or table does not indicate that it
was reprinted by APS, with permission from another source.
6. The permission granted herein is personal to the Requestor for the use specified and is not
transferable or assignable without express written permission of APS. This license may not
be amended except in writing by APS.
7. You may not alter, edit or modify the material in any manner.
8. You may translate the materials only when translation rights have been granted.
9. You may not use the material for promotional, sales, advertising or marketing purposes.
10. The foregoing license shall not take effect unless and until APS or its agent, Copyright
Clearance Center (CCC), receives payment in full in accordance with CCC Billing and
Payment Terms and Conditions, which are incorporated herein by reference.
11. Should the terms of this license be violated at any time, APS or CCC may revoke the
license with no refund to you and seek relief to the fullest extent of the laws of the USA.
Official written notice will be made using the contact information provided with the
permission request. Failure to receive such notice will not nullify revocation of the
permission.
12. APS reserves all rights not specifically granted herein.
13. This document, including the CCC Billing and Payment Terms and Conditions, shall be
the entire agreement between the parties relating to the subject matter hereof.
Other Terms and Conditions
Questions? customercare@copyright.com or +1-855-239-3415 (toll free in the US) or
+1-978-646-2777.
 ELSEVIER LICENSE
TERMS AND CONDITIONS
Jan 07, 2016

This is a License Agreement between Thomas Edwards ("You") and Elsevier ("Elsevier")
provided by Copyright Clearance Center ("CCC"). The license consists of your order details,
the terms and conditions provided by Elsevier, and the payment terms and conditions.

All payments must be made in full to CCC. For payment instructions, please see
information listed at the bottom of this form.

Supplier Elsevier Limited

The Boulevard,Langford Lane
Kidlington,Oxford,OX5 1GB,UK
Registered Company 1982084
Number

Customer name Thomas Edwards

Customer address Imperial College London

London, SW7 2AZ

License number 3774170155875

License date Dec 22, 2015

Licensed content publisher Elsevier

Licensed content publication Computer Physics Communications

Licensed content title Finite difference discretization of semiconductor drift-diffusion

equations for nanowire solar cells
Licensed content author Alexei Deinega,Sajeev John

Licensed content date October 2012

Licensed content volume 183

number

Licensed content issue 10

number
Number of pages 8

Start Page 2128

End Page 2135

Type of Use reuse in a thesis/dissertation

Intended publisher of new other

work

Portion figures/tables/illustrations
Number of 3
figures/tables/illustrations

Format both print and electronic

Are you the author of this No

Elsevier article?

Will you be translating? No

Original figure numbers Figures 2, 5 and 8

Title of your Optimisation of Coated Nanowire Solar Cells by Simulation

thesis/dissertation

Expected completion date Nov 2015

Estimated size (number of 170
pages)
Elsevier VAT number GB 494 6272 12

Permissions price 0.00 GBP

VAT/Local Sales Tax 0.00 GBP / 0.00 GBP

Total 0.00 GBP

Terms and Conditions

INTRODUCTION
1. The publisher for this copyrighted material is Elsevier. By clicking "accept" in
connection with completing this licensing transaction, you agree that the following terms
and conditions apply to this transaction (along with the Billing and Payment terms and
conditions established by Copyright Clearance Center, Inc. ("CCC"), at the time that you
opened your Rightslink account and that are available at any time at
http://myaccount.copyright.com).
GENERAL TERMS
2. Elsevier hereby grants you permission to reproduce the aforementioned material subject to
the terms and conditions indicated.
3. Acknowledgement: If any part of the material to be used (for example, figures) has
appeared in our publication with credit or acknowledgement to another source, permission
must also be sought from that source. If such permission is not obtained then that material
may not be included in your publication/copies. Suitable acknowledgement to the source
must be made, either as a footnote or in a reference list at the end of your publication, as
follows:
"Reprinted from Publication title, Vol /edition number, Author(s), Title of article / title of
chapter, Pages No., Copyright (Year), with permission from Elsevier [OR APPLICABLE
SOCIETY COPYRIGHT OWNER]." Also Lancet special credit - "Reprinted from The
Lancet, Vol. number, Author(s), Title of article, Pages No., Copyright (Year), with
permission from Elsevier."
4. Reproduction of this material is confined to the purpose and/or media for which
permission is hereby given.
5. Altering/Modifying Material: Not Permitted. However figures and illustrations may be
altered/adapted minimally to serve your work. Any other abbreviations, additions, deletions
and/or any other alterations shall be made only with prior written authorization of Elsevier
Ltd. (Please contact Elsevier at permissions@elsevier.com)
6. If the permission fee for the requested use of our material is waived in this instance, please
be advised that your future requests for Elsevier materials may attract a fee.
7. Reservation of Rights: Publisher reserves all rights not specifically granted in the
combination of (i) the license details provided by you and accepted in the course of this
licensing transaction, (ii) these terms and conditions and (iii) CCC's Billing and Payment
terms and conditions.
8. License Contingent Upon Payment: While you may exercise the rights licensed
immediately upon issuance of the license at the end of the licensing process for the
transaction, provided that you have disclosed complete and accurate details of your proposed
use, no license is finally effective unless and until full payment is received from you (either
by publisher or by CCC) as provided in CCC's Billing and Payment terms and conditions. If
full payment is not received on a timely basis, then any license preliminarily granted shall be
deemed automatically revoked and shall be void as if never granted. Further, in the event
that you breach any of these terms and conditions or any of CCC's Billing and Payment
terms and conditions, the license is automatically revoked and shall be void as if never
granted. Use of materials as described in a revoked license, as well as any use of the
materials beyond the scope of an unrevoked license, may constitute copyright infringement
and publisher reserves the right to take any and all action to protect its copyright in the
materials.
9. Warranties: Publisher makes no representations or warranties with respect to the licensed
material.
10. Indemnity: You hereby indemnify and agree to hold harmless publisher and CCC, and
their respective officers, directors, employees and agents, from and against any and all
claims arising out of your use of the licensed material other than as specifically authorized
pursuant to this license.
11. No Transfer of License: This license is personal to you and may not be sublicensed,
assigned, or transferred by you to any other person without publisher's written permission.
12. No Amendment Except in Writing: This license may not be amended except in a writing
signed by both parties (or, in the case of publisher, by CCC on publisher's behalf).
13. Objection to Contrary Terms: Publisher hereby objects to any terms contained in any
purchase order, acknowledgment, check endorsement or other writing prepared by you,
which terms are inconsistent with these terms and conditions or CCC's Billing and Payment
terms and conditions. These terms and conditions, together with CCC's Billing and Payment
terms and conditions (which are incorporated herein), comprise the entire agreement
between you and publisher (and CCC) concerning this licensing transaction. In the event of
any conflict between your obligations established by these terms and conditions and those
established by CCC's Billing and Payment terms and conditions, these terms and conditions
shall control.
14. Revocation: Elsevier or Copyright Clearance Center may deny the permissions described
in this License at their sole discretion, for any reason or no reason, with a full refund payable
to you. Notice of such denial will be made using the contact information provided by you.
Failure to receive such notice will not alter or invalidate the denial. In no event will Elsevier
or Copyright Clearance Center be responsible or liable for any costs, expenses or damage
incurred by you as a result of a denial of your permission request, other than a refund of the
amount(s) paid by you to Elsevier and/or Copyright Clearance Center for denied
permissions.
LIMITED LICENSE
The following terms and conditions apply only to specific license types:
15. Translation: This permission is granted for non-exclusive world English rights only
unless your license was granted for translation rights. If you licensed translation rights you
may only translate this content into the languages you requested. A professional translator
must perform all translations and reproduce the content word for word preserving the
integrity of the article.
16. Posting licensed content on any Website: The following terms and conditions apply as
follows: Licensing material from an Elsevier journal: All content posted to the web site must
maintain the copyright information line on the bottom of each image; A hyper-text must be
included to the Homepage of the journal from which you are licensing at
http://www.sciencedirect.com/science/journal/xxxxx or the Elsevier homepage for books at
http://www.elsevier.com; Central Storage: This license does not include permission for a
scanned version of the material to be stored in a central repository such as that provided by
Heron/XanEdu.
Licensing material from an Elsevier book: A hyper-text link must be included to the Elsevier
homepage at http://www.elsevier.com . All content posted to the web site must maintain the
copyright information line on the bottom of each image.

Posting licensed content on Electronic reserve: In addition to the above the following
clauses are applicable: The web site must be password-protected and made available only to
bona fide students registered on a relevant course. This permission is granted for 1 year only.
You may obtain a new license for future website posting.
17. For journal authors: the following clauses are applicable in addition to the above:
Preprints:
A preprint is an author's own write-up of research results and analysis, it has not been peer-
reviewed, nor has it had any other value added to it by a publisher (such as formatting,
copyright, technical enhancement etc.).
Authors can share their preprints anywhere at any time. Preprints should not be added to or
enhanced in any way in order to appear more like, or to substitute for, the final versions of
articles however authors can update their preprints on arXiv or RePEc with their Accepted
scan the printed edition to create an electronic version. Posting to a repository: Authors are
permitted to post a summary of their chapter only in their institution's repository.
19. Thesis/Dissertation: If your license is for use in a thesis/dissertation your thesis may be
submitted to your institution in either print or electronic form. Should your thesis be
published commercially, please reapply for permission. These requirements include
permission for the Library and Archives of Canada to supply single copies, on demand, of
the complete thesis and include permission for Proquest/UMI to supply single copies, on
demand, of the complete thesis. Should your thesis be published commercially, please
reapply for permission. Theses and dissertations which contain embedded PJAs as part of the
formal submission can be posted publicly by the awarding institution with DOI links back to
the formal publications on ScienceDirect.

Elsevier Open Access Terms and Conditions
You can publish open access with Elsevier in hundreds of open access journals or in nearly
2000 established subscription journals that support open access publishing. Permitted third
party re-use of these open access articles is defined by the author's choice of Creative
Commons user license. See our open access license policy for more information.
Terms & Conditions applicable to all Open Access articles published with Elsevier:
Any reuse of the article must not represent the author as endorsing the adaptation of the
article nor should the article be modified in such a way as to damage the author's honour or
reputation. If any changes have been made, such changes must be clearly indicated.
The author(s) must be appropriately credited and we ask that you include the end user
license and a DOI link to the formal publication on ScienceDirect.
If any part of the material to be used (for example, figures) has appeared in our publication
with credit or acknowledgement to another source it is the responsibility of the user to ensure
their reuse complies with the terms and conditions determined by the rights holder.
Additional Terms & Conditions applicable to each Creative Commons user license:
CC BY: The CC-BY license allows users to copy, to create extracts, abstracts and new
works from the Article, to alter and revise the Article and to make commercial use of the
Article (including reuse and/or resale of the Article by commercial entities), provided the
user gives appropriate credit (with a link to the formal publication through the relevant
DOI), provides a link to the license, indicates if changes were made and the licensor is not
represented as endorsing the use made of the work. The full details of the license are
available at http://creativecommons.org/licenses/by/4.0.
CC BY NC SA: The CC BY-NC-SA license allows users to copy, to create extracts,
abstracts and new works from the Article, to alter and revise the Article, provided this is not
done for commercial purposes, and that the user gives appropriate credit (with a link to the
formal publication through the relevant DOI), provides a link to the license, indicates if
changes were made and the licensor is not represented as endorsing the use made of the
work. Further, any new works must be made available on the same conditions. The full
details of the license are available at http://creativecommons.org/licenses/by-nc-sa/4.0.
CC BY NC ND: The CC BY-NC-ND license allows users to copy and distribute the Article,
provided this is not done for commercial purposes and further does not permit distribution of
the Article if it is changed or edited in any way, and provided the user gives appropriate
credit (with a link to the formal publication through the relevant DOI), provides a link to the
license, and that the licensor is not represented as endorsing the use made of the work. The
full details of the license are available at http://creativecommons.org/licenses/by-nc-nd/4.0.
Any commercial reuse of Open Access articles published with a CC BY NC SA or CC BY
NC ND license requires permission from Elsevier and will be subject to a fee.
Commercial reuse includes:

- Associating advertising with the full text of the Article

- Charging fees for document delivery or access
- Article aggregation
- Systematic distribution via e-mail lists or share buttons
 From: Miroslav Benes mbenes.support@lumerical.com
Subject: Re: Permission to use Figures from Lumerical knowledge base in.. [Case ID#: 54801]
Date: 21 December 2015 at 22:01
To: Edwards Thomas thomas.edwards11@imperial.ac.uk

Dear Thomas,

Thank you for contacting Lumerical. Please consider this letter to be our written permission to reprint the following resources in your
thesis at Imperial College:

https://kb.lumerical.com/en/usr_broadband_injection_angle_k_zoom59.png
https://kb.lumerical.com/en/ref_yu_mittra1.png

https://kb.lumerical.com/en/ref_yu_mittra2.png

Signed on behalf of Lumerical Inc,

________________________
Miroslav Benes | Applications Manager
Lumerical Solutions, Inc.
300 – 535 Thurlow Street, Vancouver, Canada
Tel: 1.604.733.9006 Ext. 509
www.lumerical.com

mbenes.support@lumerical.com

Knowledge Exchange | Knowledge Base | Webinars | Videos | LinkedIn | Twitter | YouTube

This email is CONFIDENTIAL. If you are not the intended recipient, please inform me by telephone or return email and delete any copy of

this email. Thank you.

-----Original Message-----
From: Edwards, Thomas <thomas.edwards11@imperial.ac.uk>
Sent: Monday December 21, 2015 01:26 AM
To: support@lumerical.com <support@lumerical.com>
Subject: Permission to use Figures from Lumerical knowledge base in thesis

Dear Lumerical Support,

I am completing my PhD thesis at Imperial College London entitled ‘Optimisation of Coated Nanowire Solar Cells by Simulation’.

I seek your permission to reprint, in my thesis some figures from the Lumerical support knowledge base. The figures to be reproduced
are:

https://kb.lumerical.com/en/usr_broadband_injection_angle_k_zoom59.png

https://kb.lumerical.com/en/ref_yu_mittra1.png

https://kb.lumerical.com/en/ref_yu_mittra2.png

I would like to include the figures in my thesis which will be added to Spiral, Imperial's​ ​institutional​​
repository http://spiral.imperial.ac.uk/ and made available to the public under a Creative Commons Attribution-NonCommercial-NoDerivs
licence.

If you are happy to grant me all the permissions requested, please return a signed copy of this letter. If you wish to grant only some of the
permissions requested, please list these and then sign.

Yours sincerely,

Thomas Edwards
 JOHN WILEY AND SONS LICENSE
TERMS AND CONDITIONS
Jan 07, 2016

This Agreement between Thomas Edwards ("You") and John Wiley and Sons ("John Wiley
and Sons") consists of your license details and the terms and conditions provided by John
Wiley and Sons and Copyright Clearance Center.

License Number 3773321035637

License date Dec 20, 2015

Licensed Content Publisher John Wiley and Sons

Licensed Content Publication Progress in Photovoltaics

Licensed Content Title Modeling extremely thin absorber solar cells for optimized design
Licensed Content Author K. Taretto,U. Rau

Licensed Content Date Jul 29, 2004

Pages 19

Type of use Dissertation/Thesis

Requestor type University/Academic

Format Print and electronic

Portion Figure/table

Number of figures/tables 2
Original Wiley figure/table Figures 1 and 2
number(s)

Will you be translating? No

Title of your thesis / Optimisation of Coated Nanowire Solar Cells by Simulation

dissertation

Expected completion date Nov 2015

Expected size (number of 170
pages)

Requestor Location Thomas Edwards

Imperial College London
Exhibition Road
South Kensington
London, United Kingdom SW7 2AZ
Attn: Thomas Edwards

Billing Type Invoice

Billing Address Thomas Edwards

Imperial College London
Exhibition Road
South Kensington
London, United Kingdom SW7 2AZ
Attn: Thomas Edwards
Total 0.00 GBP

Terms and Conditions

TERMS AND CONDITIONS

This copyrighted material is owned by or exclusively licensed to John Wiley & Sons, Inc. or
one of its group companies (each a"Wiley Company") or handled on behalf of a society with
which a Wiley Company has exclusive publishing rights in relation to a particular work
(collectively "WILEY"). By clicking "accept" in connection with completing this licensing
transaction, you agree that the following terms and conditions apply to this transaction
(along with the billing and payment terms and conditions established by the Copyright
Clearance Center Inc., ("CCC's Billing and Payment terms and conditions"), at the time that
you opened your RightsLink account (these are available at any time at
http://myaccount.copyright.com).

Terms and Conditions

The materials you have requested permission to reproduce or reuse (the "Wiley
Materials") are protected by copyright.

You are hereby granted a personal, non-exclusive, non-sub licensable (on a stand-
alone basis), non-transferable, worldwide, limited license to reproduce the Wiley
Materials for the purpose specified in the licensing process. This license, and any
CONTENT (PDF or image file) purchased as part of your order, is for a one-time
use only and limited to any maximum distribution number specified in the license. The
first instance of republication or reuse granted by this license must be completed
within two years of the date of the grant of this license (although copies prepared
before the end date may be distributed thereafter). The Wiley Materials shall not be
used in any other manner or for any other purpose, beyond what is granted in the
license. Permission is granted subject to an appropriate acknowledgement given to the
author, title of the material/book/journal and the publisher. You shall also duplicate
the copyright notice that appears in the Wiley publication in your use of the Wiley
Material. Permission is also granted on the understanding that nowhere in the text is a
previously published source acknowledged for all or part of this Wiley Material. Any
third party content is expressly excluded from this permission.

With respect to the Wiley Materials, all rights are reserved. Except as expressly
granted by the terms of the license, no part of the Wiley Materials may be copied,
modified, adapted (except for minor reformatting required by the new Publication),
translated, reproduced, transferred or distributed, in any form or by any means, and no
derivative works may be made based on the Wiley Materials without the prior
permission of the respective copyright owner.For STM Signatory Publishers
clearing permission under the terms of the STM Permissions Guidelines only, the
terms of the license are extended to include subsequent editions and for editions
in other languages, provided such editions are for the work as a whole in situ and
does not involve the separate exploitation of the permitted figures or extracts,
You may not alter, remove or suppress in any manner any copyright, trademark or
other notices displayed by the Wiley Materials. You may not license, rent, sell, loan,
lease, pledge, offer as security, transfer or assign the Wiley Materials on a stand-alone
basis, or any of the rights granted to you hereunder to any other person.

The Wiley Materials and all of the intellectual property rights therein shall at all times
remain the exclusive property of John Wiley & Sons Inc, the Wiley Companies, or
their respective licensors, and your interest therein is only that of having possession of
and the right to reproduce the Wiley Materials pursuant to Section 2 herein during the
continuance of this Agreement. You agree that you own no right, title or interest in or
to the Wiley Materials or any of the intellectual property rights therein. You shall have
no rights hereunder other than the license as provided for above in Section 2. No right,
license or interest to any trademark, trade name, service mark or other branding
("Marks") of WILEY or its licensors is granted hereunder, and you agree that you
shall not assert any such right, license or interest with respect thereto

NEITHER WILEY NOR ITS LICENSORS MAKES ANY WARRANTY OR

REPRESENTATION OF ANY KIND TO YOU OR ANY THIRD PARTY,
EXPRESS, IMPLIED OR STATUTORY, WITH RESPECT TO THE MATERIALS
OR THE ACCURACY OF ANY INFORMATION CONTAINED IN THE
MATERIALS, INCLUDING, WITHOUT LIMITATION, ANY IMPLIED
WARRANTY OF MERCHANTABILITY, ACCURACY, SATISFACTORY
QUALITY, FITNESS FOR A PARTICULAR PURPOSE, USABILITY,
INTEGRATION OR NON-INFRINGEMENT AND ALL SUCH WARRANTIES
ARE HEREBY EXCLUDED BY WILEY AND ITS LICENSORS AND WAIVED
BY YOU.

WILEY shall have the right to terminate this Agreement immediately upon breach of
this Agreement by you.

You shall indemnify, defend and hold harmless WILEY, its Licensors and their
respective directors, officers, agents and employees, from and against any actual or
threatened claims, demands, causes of action or proceedings arising from any breach
of this Agreement by you.

IN NO EVENT SHALL WILEY OR ITS LICENSORS BE LIABLE TO YOU OR

ANY OTHER PARTY OR ANY OTHER PERSON OR ENTITY FOR ANY
SPECIAL, CONSEQUENTIAL, INCIDENTAL, INDIRECT, EXEMPLARY OR
PUNITIVE DAMAGES, HOWEVER CAUSED, ARISING OUT OF OR IN
CONNECTION WITH THE DOWNLOADING, PROVISIONING, VIEWING OR
USE OF THE MATERIALS REGARDLESS OF THE FORM OF ACTION,
WHETHER FOR BREACH OF CONTRACT, BREACH OF WARRANTY, TORT,
NEGLIGENCE, INFRINGEMENT OR OTHERWISE (INCLUDING, WITHOUT
LIMITATION, DAMAGES BASED ON LOSS OF PROFITS, DATA, FILES, USE,
BUSINESS OPPORTUNITY OR CLAIMS OF THIRD PARTIES), AND
WHETHER OR NOT THE PARTY HAS BEEN ADVISED OF THE POSSIBILITY
OF SUCH DAMAGES. THIS LIMITATION SHALL APPLY
NOTWITHSTANDING ANY FAILURE OF ESSENTIAL PURPOSE OF ANY
LIMITED REMEDY PROVIDED HEREIN.

Should any provision of this Agreement be held by a court of competent jurisdiction to

be illegal, invalid, or unenforceable, that provision shall be deemed amended to
achieve as nearly as possible the same economic effect as the original provision, and
the legality, validity and enforceability of the remaining provisions of this Agreement
shall not be affected or impaired thereby.

The failure of either party to enforce any term or condition of this Agreement shall not
constitute a waiver of either party's right to enforce each and every term and condition
of this Agreement. No breach under this agreement shall be deemed waived or
excused by either party unless such waiver or consent is in writing signed by the party
granting such waiver or consent. The waiver by or consent of a party to a breach of
any provision of this Agreement shall not operate or be construed as a waiver of or
consent to any other or subsequent breach by such other party.

This Agreement may not be assigned (including by operation of law or otherwise) by

you without WILEY's prior written consent.

Any fee required for this permission shall be non-refundable after thirty (30) days
from receipt by the CCC.

These terms and conditions together with CCC's Billing and Payment terms and
conditions (which are incorporated herein) form the entire agreement between you and
WILEY concerning this licensing transaction and (in the absence of fraud) supersedes
all prior agreements and representations of the parties, oral or written. This Agreement
may not be amended except in writing signed by both parties. This Agreement shall be
 binding upon and inure to the benefit of the parties' successors, legal representatives,
and authorized assigns.

In the event of any conflict between your obligations established by these terms and
conditions and those established by CCC's Billing and Payment terms and conditions,
these terms and conditions shall prevail.

WILEY expressly reserves all rights not specifically granted in the combination of (i)
the license details provided by you and accepted in the course of this licensing
transaction, (ii) these terms and conditions and (iii) CCC's Billing and Payment terms
and conditions.

This Agreement will be void if the Type of Use, Format, Circulation, or Requestor
Type was misrepresented during the licensing process.

This Agreement shall be governed by and construed in accordance with the laws of the
State of New York, USA, without regards to such state's conflict of law rules. Any
legal action, suit or proceeding arising out of or relating to these Terms and Conditions
or the breach thereof shall be instituted in a court of competent jurisdiction in New
York County in the State of New York in the United States of America and each party
hereby consents and submits to the personal jurisdiction of such court, waives any
objection to venue in such court and consents to service of process by registered or
certified mail, return receipt requested, at the last known address of such party.

WILEY OPEN ACCESS TERMS AND CONDITIONS

Wiley Publishes Open Access Articles in fully Open Access Journals and in Subscription
journals offering Online Open. Although most of the fully Open Access journals publish
open access articles under the terms of the Creative Commons Attribution (CC BY) License
only, the subscription journals and a few of the Open Access Journals offer a choice of
Creative Commons Licenses. The license type is clearly identified on the article.
The Creative Commons Attribution License
The Creative Commons Attribution License (CC-BY) allows users to copy, distribute and
transmit an article, adapt the article and make commercial use of the article. The CC-BY
license permits commercial and non-
Creative Commons Attribution Non-Commercial License
The Creative Commons Attribution Non-Commercial (CC-BY-NC)License permits use,
distribution and reproduction in any medium, provided the original work is properly cited
and is not used for commercial purposes.(see below)

Creative Commons Attribution-Non-Commercial-NoDerivs License

The Creative Commons Attribution Non-Commercial-NoDerivs License (CC-BY-NC-ND)
permits use, distribution and reproduction in any medium, provided the original work is
properly cited, is not used for commercial purposes and no modifications or adaptations are
made. (see below)
Use by commercial "for-profit" organizations
Use of Wiley Open Access articles for commercial, promotional, or marketing purposes
requires further explicit permission from Wiley and will be subject to a fee.
Further details can be found on Wiley Online Library
http://olabout.wiley.com/WileyCDA/Section/id-410895.html

Other Terms and Conditions:

 JOHN WILEY AND SONS LICENSE
TERMS AND CONDITIONS
Jan 07, 2016

This Agreement between Thomas Edwards ("You") and John Wiley and Sons ("John Wiley
and Sons") consists of your license details and the terms and conditions provided by John
Wiley and Sons and Copyright Clearance Center.

License Number 3773330416077

License date Dec 20, 2015

Licensed Content Publisher John Wiley and Sons

Licensed Content Publication Progress in Photovoltaics

Licensed Content Title Multi-dimensional modeling of solar cells with electromagnetic and
carrier transport calculations

Licensed Content Author Xiaofeng Li,Nicholas P. Hylton,Vincenzo Giannini,Kan-Hua Lee,Ned

J. Ekins-Daukes,Stefan A. Maier

Licensed Content Date Feb 2, 2012

Pages 12

Type of use Dissertation/Thesis

Requestor type University/Academic

Format Print and electronic

Portion Figure/table

Number of figures/tables 1
Original Wiley figure/table Figure 9
number(s)
Will you be translating? No

Title of your thesis / Optimisation of Coated Nanowire Solar Cells by Simulation

dissertation

Expected completion date Nov 2015

Expected size (number of
pages)
Requestor Location
170
Thomas Edwards
Imperial College London
Exhibition Road
South Kensington
London, United Kingdom SW7 2AZ
Attn: Thomas Edwards

Billing Type Invoice

Billing Address Thomas Edwards

Imperial College London
Exhibition Road
South Kensington
London, United Kingdom SW7 2AZ
Attn: Thomas Edwards
Total 0.00 GBP

Terms and Conditions

TERMS AND CONDITIONS

This copyrighted material is owned by or exclusively licensed to John Wiley & Sons, Inc. or
one of its group companies (each a"Wiley Company") or handled on behalf of a society with
which a Wiley Company has exclusive publishing rights in relation to a particular work
(collectively "WILEY"). By clicking "accept" in connection with completing this licensing
transaction, you agree that the following terms and conditions apply to this transaction
(along with the billing and payment terms and conditions established by the Copyright
Clearance Center Inc., ("CCC's Billing and Payment terms and conditions"), at the time that
you opened your RightsLink account (these are available at any time at
http://myaccount.copyright.com).

Terms and Conditions

The materials you have requested permission to reproduce or reuse (the "Wiley
Materials") are protected by copyright.

You are hereby granted a personal, non-exclusive, non-sub licensable (on a stand-
alone basis), non-transferable, worldwide, limited license to reproduce the Wiley
Materials for the purpose specified in the licensing process. This license, and any
CONTENT (PDF or image file) purchased as part of your order, is for a one-time
use only and limited to any maximum distribution number specified in the license. The
first instance of republication or reuse granted by this license must be completed
within two years of the date of the grant of this license (although copies prepared
before the end date may be distributed thereafter). The Wiley Materials shall not be
used in any other manner or for any other purpose, beyond what is granted in the
license. Permission is granted subject to an appropriate acknowledgement given to the
author, title of the material/book/journal and the publisher. You shall also duplicate
the copyright notice that appears in the Wiley publication in your use of the Wiley
Material. Permission is also granted on the understanding that nowhere in the text is a
previously published source acknowledged for all or part of this Wiley Material. Any
third party content is expressly excluded from this permission.

With respect to the Wiley Materials, all rights are reserved. Except as expressly
granted by the terms of the license, no part of the Wiley Materials may be copied,
modified, adapted (except for minor reformatting required by the new Publication),
translated, reproduced, transferred or distributed, in any form or by any means, and no
derivative works may be made based on the Wiley Materials without the prior
permission of the respective copyright owner.For STM Signatory Publishers
clearing permission under the terms of the STM Permissions Guidelines only, the
terms of the license are extended to include subsequent editions and for editions
in other languages, provided such editions are for the work as a whole in situ and
does not involve the separate exploitation of the permitted figures or extracts,
You may not alter, remove or suppress in any manner any copyright, trademark or
other notices displayed by the Wiley Materials. You may not license, rent, sell, loan,
lease, pledge, offer as security, transfer or assign the Wiley Materials on a stand-alone
basis, or any of the rights granted to you hereunder to any other person.

The Wiley Materials and all of the intellectual property rights therein shall at all times
remain the exclusive property of John Wiley & Sons Inc, the Wiley Companies, or
their respective licensors, and your interest therein is only that of having possession of
and the right to reproduce the Wiley Materials pursuant to Section 2 herein during the
continuance of this Agreement. You agree that you own no right, title or interest in or
to the Wiley Materials or any of the intellectual property rights therein. You shall have
no rights hereunder other than the license as provided for above in Section 2. No right,
license or interest to any trademark, trade name, service mark or other branding
("Marks") of WILEY or its licensors is granted hereunder, and you agree that you
shall not assert any such right, license or interest with respect thereto

NEITHER WILEY NOR ITS LICENSORS MAKES ANY WARRANTY OR

REPRESENTATION OF ANY KIND TO YOU OR ANY THIRD PARTY,
EXPRESS, IMPLIED OR STATUTORY, WITH RESPECT TO THE MATERIALS
OR THE ACCURACY OF ANY INFORMATION CONTAINED IN THE
MATERIALS, INCLUDING, WITHOUT LIMITATION, ANY IMPLIED
WARRANTY OF MERCHANTABILITY, ACCURACY, SATISFACTORY
QUALITY, FITNESS FOR A PARTICULAR PURPOSE, USABILITY,
INTEGRATION OR NON-INFRINGEMENT AND ALL SUCH WARRANTIES
ARE HEREBY EXCLUDED BY WILEY AND ITS LICENSORS AND WAIVED
BY YOU.

WILEY shall have the right to terminate this Agreement immediately upon breach of
this Agreement by you.

You shall indemnify, defend and hold harmless WILEY, its Licensors and their
respective directors, officers, agents and employees, from and against any actual or
threatened claims, demands, causes of action or proceedings arising from any breach
of this Agreement by you.

IN NO EVENT SHALL WILEY OR ITS LICENSORS BE LIABLE TO YOU OR

ANY OTHER PARTY OR ANY OTHER PERSON OR ENTITY FOR ANY
SPECIAL, CONSEQUENTIAL, INCIDENTAL, INDIRECT, EXEMPLARY OR
PUNITIVE DAMAGES, HOWEVER CAUSED, ARISING OUT OF OR IN
CONNECTION WITH THE DOWNLOADING, PROVISIONING, VIEWING OR
USE OF THE MATERIALS REGARDLESS OF THE FORM OF ACTION,
WHETHER FOR BREACH OF CONTRACT, BREACH OF WARRANTY, TORT,
NEGLIGENCE, INFRINGEMENT OR OTHERWISE (INCLUDING, WITHOUT
LIMITATION, DAMAGES BASED ON LOSS OF PROFITS, DATA, FILES, USE,
BUSINESS OPPORTUNITY OR CLAIMS OF THIRD PARTIES), AND
WHETHER OR NOT THE PARTY HAS BEEN ADVISED OF THE POSSIBILITY
OF SUCH DAMAGES. THIS LIMITATION SHALL APPLY
NOTWITHSTANDING ANY FAILURE OF ESSENTIAL PURPOSE OF ANY
LIMITED REMEDY PROVIDED HEREIN.

Should any provision of this Agreement be held by a court of competent jurisdiction to

be illegal, invalid, or unenforceable, that provision shall be deemed amended to
achieve as nearly as possible the same economic effect as the original provision, and
the legality, validity and enforceability of the remaining provisions of this Agreement
shall not be affected or impaired thereby.

The failure of either party to enforce any term or condition of this Agreement shall not
constitute a waiver of either party's right to enforce each and every term and condition
of this Agreement. No breach under this agreement shall be deemed waived or
excused by either party unless such waiver or consent is in writing signed by the party
granting such waiver or consent. The waiver by or consent of a party to a breach of
any provision of this Agreement shall not operate or be construed as a waiver of or
consent to any other or subsequent breach by such other party.

This Agreement may not be assigned (including by operation of law or otherwise) by

you without WILEY's prior written consent.

Any fee required for this permission shall be non-refundable after thirty (30) days
from receipt by the CCC.

These terms and conditions together with CCC's Billing and Payment terms and
conditions (which are incorporated herein) form the entire agreement between you and
WILEY concerning this licensing transaction and (in the absence of fraud) supersedes
all prior agreements and representations of the parties, oral or written. This Agreement
 may not be amended except in writing signed by both parties. This Agreement shall be
binding upon and inure to the benefit of the parties' successors, legal representatives,
and authorized assigns.

In the event of any conflict between your obligations established by these terms and
conditions and those established by CCC's Billing and Payment terms and conditions,
these terms and conditions shall prevail.

WILEY expressly reserves all rights not specifically granted in the combination of (i)
the license details provided by you and accepted in the course of this licensing
transaction, (ii) these terms and conditions and (iii) CCC's Billing and Payment terms
and conditions.

This Agreement will be void if the Type of Use, Format, Circulation, or Requestor
Type was misrepresented during the licensing process.

This Agreement shall be governed by and construed in accordance with the laws of the
State of New York, USA, without regards to such state's conflict of law rules. Any
legal action, suit or proceeding arising out of or relating to these Terms and Conditions
or the breach thereof shall be instituted in a court of competent jurisdiction in New
York County in the State of New York in the United States of America and each party
hereby consents and submits to the personal jurisdiction of such court, waives any
objection to venue in such court and consents to service of process by registered or
certified mail, return receipt requested, at the last known address of such party.

WILEY OPEN ACCESS TERMS AND CONDITIONS

Wiley Publishes Open Access Articles in fully Open Access Journals and in Subscription
journals offering Online Open. Although most of the fully Open Access journals publish
open access articles under the terms of the Creative Commons Attribution (CC BY) License
only, the subscription journals and a few of the Open Access Journals offer a choice of
Creative Commons Licenses. The license type is clearly identified on the article.
The Creative Commons Attribution License
The Creative Commons Attribution License (CC-BY) allows users to copy, distribute and
transmit an article, adapt the article and make commercial use of the article. The CC-BY
license permits commercial and non-
Creative Commons Attribution Non-Commercial License
The Creative Commons Attribution Non-Commercial (CC-BY-NC)License permits use,
distribution and reproduction in any medium, provided the original work is properly cited
and is not used for commercial purposes.(see below)

Creative Commons Attribution-Non-Commercial-NoDerivs License

The Creative Commons Attribution Non-Commercial-NoDerivs License (CC-BY-NC-ND)
permits use, distribution and reproduction in any medium, provided the original work is
properly cited, is not used for commercial purposes and no modifications or adaptations are
made. (see below)
Use by commercial "for-profit" organizations
Use of Wiley Open Access articles for commercial, promotional, or marketing purposes
requires further explicit permission from Wiley and will be subject to a fee.
Further details can be found on Wiley Online Library
http://olabout.wiley.com/WileyCDA/Section/id-410895.html

Other Terms and Conditions:

 JOHN WILEY AND SONS LICENSE
TERMS AND CONDITIONS
Jan 07, 2016

This Agreement between Thomas Edwards ("You") and John Wiley and Sons ("John Wiley
and Sons") consists of your license details and the terms and conditions provided by John
Wiley and Sons and Copyright Clearance Center.

License Number 3773311018021

License date Dec 20, 2015

Licensed Content Publisher John Wiley and Sons

Licensed Content Publication physica status solidi (b) basic solid state physics
Licensed Content Title Nanowire–quantum-dot–polymer solar cell
Licensed Content Author A. Nadarajah,Robert C. Word,Kaitlyn VanSant,R. Könenkamp

Licensed Content Date Aug 7, 2008

Pages 4

Type of use Dissertation/Thesis

Requestor type University/Academic

Format Print and electronic

Portion Figure/table

Number of figures/tables 1
Original Wiley figure/table Figure 3
number(s)

Will you be translating? No

Title of your thesis / Optimisation of Coated Nanowire Solar Cells by Simulation

dissertation

Expected completion date Nov 2015

Expected size (number of 170
pages)

Requestor Location Thomas Edwards

Imperial College London
Exhibition Road
South Kensington
London, United Kingdom SW7 2AZ
Attn: Thomas Edwards

Billing Type Invoice

Billing Address Thomas Edwards

Imperial College London
Exhibition Road
South Kensington
London, United Kingdom SW7 2AZ
Attn: Thomas Edwards
Total 0.00 GBP

Terms and Conditions

TERMS AND CONDITIONS

This copyrighted material is owned by or exclusively licensed to John Wiley & Sons, Inc. or
one of its group companies (each a"Wiley Company") or handled on behalf of a society with
which a Wiley Company has exclusive publishing rights in relation to a particular work
(collectively "WILEY"). By clicking "accept" in connection with completing this licensing
transaction, you agree that the following terms and conditions apply to this transaction
(along with the billing and payment terms and conditions established by the Copyright
Clearance Center Inc., ("CCC's Billing and Payment terms and conditions"), at the time that
you opened your RightsLink account (these are available at any time at
http://myaccount.copyright.com).

Terms and Conditions

The materials you have requested permission to reproduce or reuse (the "Wiley
Materials") are protected by copyright.

You are hereby granted a personal, non-exclusive, non-sub licensable (on a stand-
alone basis), non-transferable, worldwide, limited license to reproduce the Wiley
Materials for the purpose specified in the licensing process. This license, and any
CONTENT (PDF or image file) purchased as part of your order, is for a one-time
use only and limited to any maximum distribution number specified in the license. The
first instance of republication or reuse granted by this license must be completed
within two years of the date of the grant of this license (although copies prepared
before the end date may be distributed thereafter). The Wiley Materials shall not be
used in any other manner or for any other purpose, beyond what is granted in the
license. Permission is granted subject to an appropriate acknowledgement given to the
author, title of the material/book/journal and the publisher. You shall also duplicate
the copyright notice that appears in the Wiley publication in your use of the Wiley
Material. Permission is also granted on the understanding that nowhere in the text is a
previously published source acknowledged for all or part of this Wiley Material. Any
third party content is expressly excluded from this permission.

With respect to the Wiley Materials, all rights are reserved. Except as expressly
granted by the terms of the license, no part of the Wiley Materials may be copied,
modified, adapted (except for minor reformatting required by the new Publication),
translated, reproduced, transferred or distributed, in any form or by any means, and no
derivative works may be made based on the Wiley Materials without the prior
permission of the respective copyright owner.For STM Signatory Publishers
clearing permission under the terms of the STM Permissions Guidelines only, the
terms of the license are extended to include subsequent editions and for editions
in other languages, provided such editions are for the work as a whole in situ and
does not involve the separate exploitation of the permitted figures or extracts,
You may not alter, remove or suppress in any manner any copyright, trademark or
other notices displayed by the Wiley Materials. You may not license, rent, sell, loan,
lease, pledge, offer as security, transfer or assign the Wiley Materials on a stand-alone
basis, or any of the rights granted to you hereunder to any other person.

The Wiley Materials and all of the intellectual property rights therein shall at all times
remain the exclusive property of John Wiley & Sons Inc, the Wiley Companies, or
their respective licensors, and your interest therein is only that of having possession of
and the right to reproduce the Wiley Materials pursuant to Section 2 herein during the
continuance of this Agreement. You agree that you own no right, title or interest in or
to the Wiley Materials or any of the intellectual property rights therein. You shall have
no rights hereunder other than the license as provided for above in Section 2. No right,
license or interest to any trademark, trade name, service mark or other branding
("Marks") of WILEY or its licensors is granted hereunder, and you agree that you
shall not assert any such right, license or interest with respect thereto

NEITHER WILEY NOR ITS LICENSORS MAKES ANY WARRANTY OR

REPRESENTATION OF ANY KIND TO YOU OR ANY THIRD PARTY,
EXPRESS, IMPLIED OR STATUTORY, WITH RESPECT TO THE MATERIALS
OR THE ACCURACY OF ANY INFORMATION CONTAINED IN THE
MATERIALS, INCLUDING, WITHOUT LIMITATION, ANY IMPLIED
WARRANTY OF MERCHANTABILITY, ACCURACY, SATISFACTORY
QUALITY, FITNESS FOR A PARTICULAR PURPOSE, USABILITY,
INTEGRATION OR NON-INFRINGEMENT AND ALL SUCH WARRANTIES
ARE HEREBY EXCLUDED BY WILEY AND ITS LICENSORS AND WAIVED
BY YOU.

WILEY shall have the right to terminate this Agreement immediately upon breach of
this Agreement by you.

You shall indemnify, defend and hold harmless WILEY, its Licensors and their
respective directors, officers, agents and employees, from and against any actual or
threatened claims, demands, causes of action or proceedings arising from any breach
of this Agreement by you.

IN NO EVENT SHALL WILEY OR ITS LICENSORS BE LIABLE TO YOU OR

ANY OTHER PARTY OR ANY OTHER PERSON OR ENTITY FOR ANY
SPECIAL, CONSEQUENTIAL, INCIDENTAL, INDIRECT, EXEMPLARY OR
PUNITIVE DAMAGES, HOWEVER CAUSED, ARISING OUT OF OR IN
CONNECTION WITH THE DOWNLOADING, PROVISIONING, VIEWING OR
USE OF THE MATERIALS REGARDLESS OF THE FORM OF ACTION,
WHETHER FOR BREACH OF CONTRACT, BREACH OF WARRANTY, TORT,
NEGLIGENCE, INFRINGEMENT OR OTHERWISE (INCLUDING, WITHOUT
LIMITATION, DAMAGES BASED ON LOSS OF PROFITS, DATA, FILES, USE,
BUSINESS OPPORTUNITY OR CLAIMS OF THIRD PARTIES), AND
WHETHER OR NOT THE PARTY HAS BEEN ADVISED OF THE POSSIBILITY
OF SUCH DAMAGES. THIS LIMITATION SHALL APPLY
NOTWITHSTANDING ANY FAILURE OF ESSENTIAL PURPOSE OF ANY
LIMITED REMEDY PROVIDED HEREIN.

Should any provision of this Agreement be held by a court of competent jurisdiction to

be illegal, invalid, or unenforceable, that provision shall be deemed amended to
achieve as nearly as possible the same economic effect as the original provision, and
the legality, validity and enforceability of the remaining provisions of this Agreement
shall not be affected or impaired thereby.

The failure of either party to enforce any term or condition of this Agreement shall not
constitute a waiver of either party's right to enforce each and every term and condition
of this Agreement. No breach under this agreement shall be deemed waived or
excused by either party unless such waiver or consent is in writing signed by the party
granting such waiver or consent. The waiver by or consent of a party to a breach of
any provision of this Agreement shall not operate or be construed as a waiver of or
consent to any other or subsequent breach by such other party.

This Agreement may not be assigned (including by operation of law or otherwise) by

you without WILEY's prior written consent.

Any fee required for this permission shall be non-refundable after thirty (30) days
from receipt by the CCC.

These terms and conditions together with CCC's Billing and Payment terms and
conditions (which are incorporated herein) form the entire agreement between you and
WILEY concerning this licensing transaction and (in the absence of fraud) supersedes
all prior agreements and representations of the parties, oral or written. This Agreement
may not be amended except in writing signed by both parties. This Agreement shall be
 binding upon and inure to the benefit of the parties' successors, legal representatives,
and authorized assigns.

In the event of any conflict between your obligations established by these terms and
conditions and those established by CCC's Billing and Payment terms and conditions,
these terms and conditions shall prevail.

WILEY expressly reserves all rights not specifically granted in the combination of (i)
the license details provided by you and accepted in the course of this licensing
transaction, (ii) these terms and conditions and (iii) CCC's Billing and Payment terms
and conditions.

This Agreement will be void if the Type of Use, Format, Circulation, or Requestor
Type was misrepresented during the licensing process.

This Agreement shall be governed by and construed in accordance with the laws of the
State of New York, USA, without regards to such state's conflict of law rules. Any
legal action, suit or proceeding arising out of or relating to these Terms and Conditions
or the breach thereof shall be instituted in a court of competent jurisdiction in New
York County in the State of New York in the United States of America and each party
hereby consents and submits to the personal jurisdiction of such court, waives any
objection to venue in such court and consents to service of process by registered or
certified mail, return receipt requested, at the last known address of such party.

WILEY OPEN ACCESS TERMS AND CONDITIONS

Wiley Publishes Open Access Articles in fully Open Access Journals and in Subscription
journals offering Online Open. Although most of the fully Open Access journals publish
open access articles under the terms of the Creative Commons Attribution (CC BY) License
only, the subscription journals and a few of the Open Access Journals offer a choice of
Creative Commons Licenses. The license type is clearly identified on the article.
The Creative Commons Attribution License
The Creative Commons Attribution License (CC-BY) allows users to copy, distribute and
transmit an article, adapt the article and make commercial use of the article. The CC-BY
license permits commercial and non-
Creative Commons Attribution Non-Commercial License
The Creative Commons Attribution Non-Commercial (CC-BY-NC)License permits use,
distribution and reproduction in any medium, provided the original work is properly cited
and is not used for commercial purposes.(see below)

Creative Commons Attribution-Non-Commercial-NoDerivs License

The Creative Commons Attribution Non-Commercial-NoDerivs License (CC-BY-NC-ND)
permits use, distribution and reproduction in any medium, provided the original work is
properly cited, is not used for commercial purposes and no modifications or adaptations are
made. (see below)
Use by commercial "for-profit" organizations
Use of Wiley Open Access articles for commercial, promotional, or marketing purposes
requires further explicit permission from Wiley and will be subject to a fee.
Further details can be found on Wiley Online Library
http://olabout.wiley.com/WileyCDA/Section/id-410895.html

Other Terms and Conditions:

E-theses letter template: request to reproduce an extract from
a third party's published work

5th November 2015

Dear Alexei Deinega,

I am completing my PhD thesis at Imperial College London entitled ‘Optimisation of

Coated Nanowire Solar Cells by Simulation’.

I seek your permission to reprint, a figure from the Microvolt website. The figure to be
reproduced is:

https://sites.physics.utoronto.ca/sajeevjohn/software/microvolt/images/bands.png/@
@images/3a29e1dc-e079-4680-bef4-6903c40a1011.png

I would like to include the figure in my thesis which will be added to Spiral,
Imperial's institutional repository http://spiral.imperial.ac.uk/ and made available to
the public under a Creative Commons Attribution-NonCommercial-NoDerivs licence.

If you are happy to grant me all the permissions requested, please return a signed
copy of this letter. If you wish to grant only some of the permissions requested,
please list these and then sign.

Yours sincerely,

Thomas Edwards

Permission granted for the use requested above:

I confirm that I am the copyright holder of the extract above and hereby give
permission to include it in your thesis which will be made available, via the internet,
for non-commercial purposes under the terms of the user licence.

Signed:

Name: Alexei Deinega

Organisation: University of Toronto

Job title: postdoc

 NATURE PUBLISHING GROUP LICENSE
TERMS AND CONDITIONS
Jan 07, 2016

This is a License Agreement between Thomas Edwards ("You") and Nature Publishing
Group ("Nature Publishing Group") provided by Copyright Clearance Center ("CCC"). The
license consists of your order details, the terms and conditions provided by Nature
Publishing Group, and the payment terms and conditions.

All payments must be made in full to CCC. For payment instructions, please see
information listed at the bottom of this form.

License Number 3773321235344

License date Dec 20, 2015

Licensed content publisher Nature Publishing Group

Licensed content publication Nature Materials

Licensed content title Plasmonics for improved photovoltaic devices

Licensed content author Harry A. Atwater, Albert Polman

Licensed content date Feb 19, 2010

Volume number 9

Issue number 3

Type of Use reuse in a dissertation / thesis

Requestor type academic/educational

Format print and electronic

Portion figures/tables/illustrations

Number of 3
figures/tables/illustrations
High-res required no

Figures Figures 2a, 2b and 2c

Author of this NPG article no

Your reference number None

Title of your thesis / Optimisation of Coated Nanowire Solar Cells by Simulation

dissertation
Expected completion date Nov 2015

Estimated size (number of 170

pages)

Total 0.00 GBP

Terms and Conditions

Terms and Conditions for Permissions

Nature Publishing Group hereby grants you a non-exclusive license to reproduce this
material for this purpose, and for no other use,subject to the conditions below:
1. NPG warrants that it has, to the best of its knowledge, the rights to license reuse of this
material. However, you should ensure that the material you are requesting is original to
Nature Publishing Group and does not carry the copyright of another entity (as credited in
the published version). If the credit line on any part of the material you have requested
indicates that it was reprinted or adapted by NPG with permission from another source,
then you should also seek permission from that source to reuse the material.

2. Permission granted free of charge for material in print is also usually granted for any
 electronic version of that work, provided that the material is incidental to the work as a
whole and that the electronic version is essentially equivalent to, or substitutes for, the
print version.Where print permission has been granted for a fee, separate permission must
be obtained for any additional, electronic re-use (unless, as in the case of a full paper, this
has already been accounted for during your initial request in the calculation of a print
run).NB: In all cases, web-based use of full-text articles must be authorized separately
through the 'Use on a Web Site' option when requesting permission.

3. Permission granted for a first edition does not apply to second and subsequent editions and
for editions in other languages (except for signatories to the STM Permissions Guidelines,
or where the first edition permission was granted for free).

4. Nature Publishing Group's permission must be acknowledged next to the figure, table or
abstract in print. In electronic form, this acknowledgement must be visible at the same
time as the figure/table/abstract, and must be hyperlinked to the journal's homepage.

5. The credit line should read:

Reprinted by permission from Macmillan Publishers Ltd: [JOURNAL NAME] (reference
citation), copyright (year of publication)
For AOP papers, the credit line should read:
Reprinted by permission from Macmillan Publishers Ltd: [JOURNAL NAME], advance online
publication, day month year (doi: 10.1038/sj.[JOURNAL ACRONYM].XXXXX)

Note: For republication from the British Journal of Cancer, the following credit
lines apply.
Reprinted by permission from Macmillan Publishers Ltd on behalf of Cancer Research UK:
[JOURNAL NAME] (reference citation), copyright (year of publication)For AOP papers, the
credit line should read:
Reprinted by permission from Macmillan Publishers Ltd on behalf of Cancer Research UK:
[JOURNAL NAME], advance online publication, day month year (doi: 10.1038/sj.[JOURNAL
ACRONYM].XXXXX)

6. Adaptations of single figures do not require NPG approval. However, the adaptation should
be credited as follows:

Adapted by permission from Macmillan Publishers Ltd: [JOURNAL NAME] (reference

citation), copyright (year of publication)

Note: For adaptation from the British Journal of Cancer, the following credit line
applies.
Adapted by permission from Macmillan Publishers Ltd on behalf of Cancer Research UK:
[JOURNAL NAME] (reference citation), copyright (year of publication)

7. Translations of 401 words up to a whole article require NPG approval. Please visit
http://www.macmillanmedicalcommunications.com for more information.Translations of up
to a 400 words do not require NPG approval. The translation should be credited as follows:

Translated by permission from Macmillan Publishers Ltd: [JOURNAL NAME] (reference

citation), copyright (year of publication).

Note: For translation from the British Journal of Cancer, the following credit line
applies.
Translated by permission from Macmillan Publishers Ltd on behalf of Cancer Research UK:
[JOURNAL NAME] (reference citation), copyright (year of publication)

We are certain that all parties will benefit from this agreement and wish you the best in the
use of this material. Thank you.
Special Terms:
v1.1
Questions? customercare@copyright.com or +1-855-239-3415 (toll free in the US) or
+1-978-646-2777.
 NATURE PUBLISHING GROUP LICENSE
TERMS AND CONDITIONS
Jan 07, 2016

This is a License Agreement between Thomas Edwards ("You") and Nature Publishing
Group ("Nature Publishing Group") provided by Copyright Clearance Center ("CCC"). The
license consists of your order details, the terms and conditions provided by Nature
Publishing Group, and the payment terms and conditions.

All payments must be made in full to CCC. For payment instructions, please see
information listed at the bottom of this form.

License Number 3773330260721

License date Dec 20, 2015

Licensed content publisher Nature Publishing Group

Licensed content publication Nature Photonics

Licensed content title Single-nanowire solar cells beyond the Shockley-Queisser limit
Licensed content author Peter Krogstrup, Henrik Ingerslev Jørgensen, Martin Heiss, Olivier
Demichel, Jeppe V. Holm, Martin Aagesen

Licensed content date Mar 24, 2013

Volume number 7

Issue number 4

Type of Use reuse in a dissertation / thesis

Requestor type academic/educational

Format print and electronic

Portion figures/tables/illustrations

Number of 1
figures/tables/illustrations
High-res required no

Figures Figure 2a

Author of this NPG article no

Your reference number None

Title of your thesis / Optimisation of Coated Nanowire Solar Cells by Simulation

dissertation

Expected completion date Nov 2015

Estimated size (number of 170

pages)

Total 0.00 GBP

Terms and Conditions

Terms and Conditions for Permissions

Nature Publishing Group hereby grants you a non-exclusive license to reproduce this
material for this purpose, and for no other use,subject to the conditions below:
1. NPG warrants that it has, to the best of its knowledge, the rights to license reuse of this
material. However, you should ensure that the material you are requesting is original to
Nature Publishing Group and does not carry the copyright of another entity (as credited in
the published version). If the credit line on any part of the material you have requested
indicates that it was reprinted or adapted by NPG with permission from another source,
then you should also seek permission from that source to reuse the material.

 2. Permission granted free of charge for material in print is also usually granted for any
electronic version of that work, provided that the material is incidental to the work as a
whole and that the electronic version is essentially equivalent to, or substitutes for, the
print version.Where print permission has been granted for a fee, separate permission must
be obtained for any additional, electronic re-use (unless, as in the case of a full paper, this
has already been accounted for during your initial request in the calculation of a print
run).NB: In all cases, web-based use of full-text articles must be authorized separately
through the 'Use on a Web Site' option when requesting permission.

3. Permission granted for a first edition does not apply to second and subsequent editions and
for editions in other languages (except for signatories to the STM Permissions Guidelines,
or where the first edition permission was granted for free).

4. Nature Publishing Group's permission must be acknowledged next to the figure, table or
abstract in print. In electronic form, this acknowledgement must be visible at the same
time as the figure/table/abstract, and must be hyperlinked to the journal's homepage.

5. The credit line should read:

Reprinted by permission from Macmillan Publishers Ltd: [JOURNAL NAME] (reference
citation), copyright (year of publication)
For AOP papers, the credit line should read:
Reprinted by permission from Macmillan Publishers Ltd: [JOURNAL NAME], advance online
publication, day month year (doi: 10.1038/sj.[JOURNAL ACRONYM].XXXXX)

Note: For republication from the British Journal of Cancer, the following credit
lines apply.
Reprinted by permission from Macmillan Publishers Ltd on behalf of Cancer Research UK:
[JOURNAL NAME] (reference citation), copyright (year of publication)For AOP papers, the
credit line should read:
Reprinted by permission from Macmillan Publishers Ltd on behalf of Cancer Research UK:
[JOURNAL NAME], advance online publication, day month year (doi: 10.1038/sj.[JOURNAL
ACRONYM].XXXXX)

6. Adaptations of single figures do not require NPG approval. However, the adaptation should
be credited as follows:

Adapted by permission from Macmillan Publishers Ltd: [JOURNAL NAME] (reference

citation), copyright (year of publication)

Note: For adaptation from the British Journal of Cancer, the following credit line
applies.
Adapted by permission from Macmillan Publishers Ltd on behalf of Cancer Research UK:
[JOURNAL NAME] (reference citation), copyright (year of publication)

7. Translations of 401 words up to a whole article require NPG approval. Please visit
http://www.macmillanmedicalcommunications.com for more information.Translations of up
to a 400 words do not require NPG approval. The translation should be credited as follows:

Translated by permission from Macmillan Publishers Ltd: [JOURNAL NAME] (reference

citation), copyright (year of publication).

Note: For translation from the British Journal of Cancer, the following credit line
applies.
Translated by permission from Macmillan Publishers Ltd on behalf of Cancer Research UK:
[JOURNAL NAME] (reference citation), copyright (year of publication)

We are certain that all parties will benefit from this agreement and wish you the best in the
use of this material. Thank you.
Special Terms:
v1.1
Questions? customercare@copyright.com or +1-855-239-3415 (toll free in the US) or
+1-978-646-2777.
 AIP PUBLISHING LLC LICENSE
TERMS AND CONDITIONS
Jan 07, 2016
All payments must be made in full to CCC. For payment instructions, please see
information listed at the bottom of this form.

License Number 3774161401234

Order Date Dec 22, 2015

Publisher AIP Publishing LLC

Publication Journal of Applied Physics

Article Title The potential and device physics of interdigitated thin-film solar cells
Author Wyatt K. Metzger

Online Publication Date May 9, 2008

Volume number 103

Issue number 9
Type of Use Thesis/Dissertation
Requestor type Student

Format Print and electronic

Portion Figure/Table

Number of figures/tables 2

Title of your thesis / Optimisation of Coated Nanowire Solar Cells by Simulation

dissertation
Expected completion date Nov 2015

Estimated size (number of 170

pages)

Total 0.00 GBP

Terms and Conditions

AIP Publishing LLC -- Terms and Conditions: Permissions Uses

AIP Publishing LLC ("AIPP"") hereby grants to you the non-exclusive right and license to use
and/or distribute the Material according to the use specified in your order, on a one-time basis,
for the specified term, with a maximum distribution equal to the number that you have ordered.
Any links or other content accompanying the Material are not the subject of this license.

1. You agree to include the following copyright and permission notice with the reproduction of
the Material:"Reprinted with permission from [FULL CITATION]. Copyright [PUBLICATION
YEAR], AIP Publishing LLC." For an article, the copyright and permission notice must be
printed on the first page of the article or book chapter. For photographs, covers, or tables,
the copyright and permission notice may appear with the Material, in a footnote, or in the
reference list.
2. If you have licensed reuse of a figure, photograph, cover, or table, it is your responsibility
to ensure that the material is original to AIPP and does not contain the copyright of another
entity, and that the copyright notice of the figure, photograph, cover, or table does not
indicate that it was reprinted by AIPP, with permission, from another source. Under no
circumstances does AIPP, purport or intend to grant permission to reuse material to which
it does not hold copyright.
3. You may not alter or modify the Material in any manner. You may translate the Material
into another language only if you have licensed translation rights. You may not use the
Material for promotional purposes. AIPP reserves all rights not specifically granted herein.
4. The foregoing license shall not take effect unless and until AIPP or its agent, Copyright
Clearance Center, receives the Payment in accordance with Copyright Clearance Center
Billing and Payment Terms and Conditions, which are incorporated herein by reference.
5. AIPP or the Copyright Clearance Center may, within two business days of granting this
license, revoke the license for any reason whatsoever, with a full refund payable to you.
Should you violate the terms of this license at any time, AIPP, AIP Publishing LLC, or
Copyright Clearance Center may revoke the license with no refund to you. Notice of such
 revocation will be made using the contact information provided by you. Failure to receive
such notice will not nullify the revocation.
6. AIPP makes no representations or warranties with respect to the Material. You agree to
indemnify and hold harmless AIPP, AIP Publishing LLC, and their officers, directors,
employees or agents from and against any and all claims arising out of your use of the
Material other than as specifically authorized herein.
7. The permission granted herein is personal to you and is not transferable or assignable
without the prior written permission of AIPP. This license may not be amended except in a
writing signed by the party to be charged.
8. If purchase orders, acknowledgments or check endorsements are issued on any forms
containing terms and conditions which are inconsistent with these provisions, such
inconsistent terms and conditions shall be of no force and effect. This document, including
the CCC Billing and Payment Terms and Conditions, shall be the entire agreement between
the parties relating to the subject matter hereof.

This Agreement shall be governed by and construed in accordance with the laws of the State
of New York. Both parties hereby submit to the jurisdiction of the courts of New York
County for purposes of resolving any disputes that may arise hereunder.
Questions? customercare@copyright.com or +1-855-239-3415 (toll free in the US) or
+1-978-646-2777.