University of Southern California Dissertations and Theses

Efficient ternary blend bulk heterojunction solar cells with tunable open-circuit voltage

(USC Thesis Other)

Efficient ternary blend bulk heterojunction solar cells with tunable open-circuit voltage

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content
EFFICIENT TERNARY BLEND BULK HETEROJUNCTION SOLAR CELLS WITH
TUNABLE OPEN-CIRCUIT VOLTAGE

by

Petr P. Khlyabich

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(CHEMISTRY)

December 2013

Copyright 2013 Petr P. Khlyabich
ii
Epigraph

The Science of Today Is the Technology of Tomorrow
- Edward Teller

iii
Dedication

To My Parents For Always Supporting Me

iv
Acknowledgements

First of all, I would like to thank my parents, Nellya and Petr Khlyabich, for all
their support, love and attention not only during my graduate school, but always. Thank
you for instilling in me a passion for science, arts and sports. I always cherish our talks
and your suggestions, and not once you guided me through the hardest times of my life. I
always know that I have two people who always ready to back me up no matter what and
I am proud to be your son. You have made me the man I am.
I am grateful to my supervisor Prof. Barry C. Thompson for his guidance and
support through all these years. Thank you for giving me an opportunity to work in the
exciting field of organic photovoltaics and letting a physicist be a member of an organic
chemistry team. You were an excellent mentor to me and taught me to look at problems
from a different angle. Your goal-oriented and hard-working attitude towards work will
continue to be an example for me in my future life. But most of all I will miss our
scientific discussions.
Of course all the success and fun through my graduate school would never be
possible without my awesome officemate Beate Burkhart. You were the important part of
our team and your outstanding experience in organic chemistry and throughout education
let us achieve the heights we are at right now. I am very thankful for your friendship and
will always value our discussions, laughter and smiles.
Thank you to my committee members Prof. Mark E. Thompson and Prof.
Aiichiro Nakano for going through all this process with me. I am very thankful to my
v
screening committee members Prof. Steve E. Bradforth and Prof. Alex V. Benderskii.
Additional thanks to Prof. Mark E. Thompson for letting me work in his lab for the first
two years and Prof. Steve E. Bradforth for introducing me to the field of spectroscopy
and valuable discussions.
I am grateful to all the collaborators through all these years: Dr. Robert A. Street,
Prof. Steve E. Bradforth, Prof. Alex V. Benderskii and Prof. Joe C. Campbell.
I was lucky to work with great fellow graduate students and post-docs: Beate
Burkhart, Alejandra Aviles, Andrey Rudenko, Tuba Cakir Canak, Kejia Li, Saptaparna
Das, Purnim Dhar and Dr. Sean T. Roberts. Special thanks to Dr. Cody W. Schlenker for
teaching me how to make solar cells and spending a lot of time answering my questions.
I appreciate all the work done by the machine shop: Donald Wiggins, Ramon and
Mike as well as Ross Lewis. You have made my life easier and experiments faster.
I never forget my first steps in real science, which would never be possible
without my undergraduate and graduate advisor, Natalia Yevlampieva, at Saint-
Petersburg State University. These were an amazing 4 years that accelerated my entire
scientific life and let me see myself as a fellow scientist.
I am very grateful to Maria Frushicheva for her love, care and support. Thank you
for making me smile every day and realizing my dream to live by the ocean.
I am thankful to my sister and my brother-in-law for all their help and time they
have spent supporting me. Thank you for my English and time in Europe.
Finally, special thanks to my dog and cat. You were my extraordinary friends. I
miss you.
vi
Table of Contents
Epigraph ii
Dedication iii
Acknowledgments iv
List of Tables ix
List of Figures xii
Abstract xxiii
Chapter 1 Polymer-Based Solar Cells: State-of-the-Art Principles for the
Design of Active Layer Components 1
1.1 Introduction 1
1.2 Electronic Processes and Interactions in the Donor-Acceptor Composite 12
1.2.1 The Charge Photogeneration Pathway 12
1.2.2 Influence of Electronic Structure on the V
oc
24
1.2.3 Defining the Electronics of the Optimal Donor-Acceptor Pair 29
1.3 Physical Interactions in the Donor-Acceptor Composite: Defining the
Optimal Morphology 36
1.4 Architectures and Stability in Bulk Heterojunction Solar Cells 44
1.4.1 Device Architecture 44
1.4.2 Long-Term Stability 47
1.5 Outlook 52
1.6 Chapter Bibliography 53
Chapter 2 Tuning the Open-Circuit Voltage in Ternary Blend Organic Solar
Cells as a Path to Higher Efficiency 65
2.1 Introduction 65
2.2 Ternary Blend Bulk Heterojunction Solar Cells 71
2.2.1 Basis Concept – Increase of Light Absorption 73
2.2.2 Tuning the V
oc
– Early Evidence 81
2.2.3 Tuning the V
oc
85
2.2.4 Models 98
2.3 Future Challenges and Directions 108
2.4 Chapter Bibliography 112
Chapter 3 Efficient Ternary Blend Bulk Heterojunction Solar Cells with
Tunable Open-Circuit Voltage 126
3.1 Introduction 126
3.2 Results and Discussion 127
vii
3.3 Conclusion 136
3.4 Chapter Bibliography 137
Chapter 4 Semi-Random and Random Copolymers 141
4.1 Efficient Solar Cells from Semi-Random P3HT Analogues
Incorporating Diketopyrrolopyrrole 141
4.1.1 Introduction 141
4.1.2 Results and Discussion 142
4.1.3 Conclusion 155
4.2 Influence of the Ethylhexyl Side-Chain Content on the Open-Circuit
Voltage in rr-Poly(3-hexylthiophene-co-3-(2-ethylhexyl)thiophene)
Copolymers 156
4.2.1 Introduction 156
4.2.2 Results and Discussion 157
4.2.3 Conclusion 176
4.3 Chapter Bibliography 178
Chapter 5 Compositional Dependence of the Open-Circuit Voltage in Ternary
Blend Bulk Heterojunction Solar Cells Based on Two Donor
Polymers 185
5.1 Introduction 185
5.2 Results and Discussion 186
5.3 Conclusion 196
5.4 Chapter Bibliography 197
Chapter 6 Origin of the Tunable Open-Circuit Voltage in Ternary Blend Bulk
Heterojunction Organic Solar Cells 200
6.1 Introduction 200
6.2 Results and Discussion 201
6.3 Conclusion 213
6.4 Chapter Bibliography 214
Perspectives 217

Bibliography 220

Appendices
Appendix A Ternary Blend Bulk Heterojunction Solar Cells in the Case of
Complementary Acceptors 249
A.1 Materials and Methods 249
A.2 Synthetic Procedures 251
A.3 UV-vis and GIXRD 252
A.4 Device Fabrication and Characterization 253
A.5 J-V and EQE Curves 255
viii
A.6 TEM Images 257
A.7 Appendix Bibliography 257
Appendix B Semi-Random and Random Copolymers 258
B.1 Semi-Random Copolymers 258
B.1.1 Materials and Methods 258
B.1.2 UV-vis and GIXRD 260
B.1.3 CV Traces 263
B.1.4 Device Fabrication and Characterization 263
B.1.5 J-V Curves 267
B.1.6 TEM Images 268
B.1.7 Mobility Measurements 268
B.2 Random Copolymers 269
B.2.1 Materials and Methods 269
B.2.2 UV-vis 272
B.2.3 CV Traces 273
B.2.4 DSC Curves 275
B.2.5 Device Fabrication and Characterization 275
B.2.6 TEM Images 278
B.3 Appendix Bibliography 278
Appendix C Ternary Blend Bulk Heterojunction Solar Cells in the Case of
Complementary Donors 280
C.1 Materials and Methods 280
C.2 Synthetic Procedures 281
C.3 UV-vis and GIXRD 282
C.4 Device Fabrication and Characterization 284
C.5 J-V Curves 287
C.6 TEM Images 288
C.7 Mobility Measurements 289
C.8 Raw J-V Data 290
C.9 Appendix Bibliography 292

ix
List of Tables

Table 1.1 Overview of some of the highest performing BHJ solar cells.
The optical band gap (E
g
) and the electrochemical values for
HOMO and LUMO levels are given where reported (HOMO
and LUMO values have been adjusted for consistency so that
Fc/Fc
+
is taken as a value of 5.1 eV below the vacuum for all
cases) 8

Table 1.2 The integrated photon flux and maximum current density
available for a photovoltaic device that harvests light from
280 nm up to the wavelength quoted, assuming that every
photon is converted to an electron that is collected in the
external circuit 9

Table 1.3 Measured and calculated performance parameters for all
devices

based on the polymers illustrated in Figure 1.6 28

Table 1.4 The properties of polymer-C
60
-PCBM blend combinations 34

Table 2.1 Photovoltaic properties of ternary blend BHJ solar cells with
unpinned open-circuit voltage 84

Table 2.2 Photovoltaic properties of P3HT:PC
61
BM:ICBA ternary
blend BHJ solar cells at different fullerene ratios 87

Table 2.3 Photovoltaic properties of P3HTT-DPP-10%:P3HT
75
-co-
EHT
25
:PC
61
BM ternary blend BHJ solar cells at optimized
ratios 90

Table 3.1 Photovoltaic properties of P3HT:PC
61
BM:ICBA ternary
blend BHJ solar cells at different fullerene ratios 131

Table 4.1 Photovoltaic properties of P3HT, P3HTT-DPP-5%, P3HTT-
DPP-10% and P3HTT-DPP-15% with PC
61
BM as an
acceptor 151

Table 4.2 Molecular and Electronic Properties of Polymers 165

Table 4.3 Photovoltaic properties of P3HT, P3HT
90
-co-EHT
10
, P3HT
75
-
co-EHT
25
, P3HT
50
-co-EHT
50
and P3EHT with PC
61
BM as
the acceptor 171

x
Table 5.1 Photovoltaic properties of P3HTT-DPP-10%:P3HT
75
-co-
EHT
25
:PC
61
BM ternary blend BHJ solar cells at optimized
ratios 190

Table B.1 Optical properties of P3HT, P3HTT-DPP-5%, P3HTT-DPP-
10% and P3HTT-DPP-15% in o-dichlorobenzene (o-DCB)
and thin films, spin-coated from o-DCB 260

Table B.2 Raw short-circuit current densities (J
sc
), spectral-mismatch
factor (M), spectral mismatch-corrected short-circuit current
densities (J
sc,corr
) and integrated short-circuit current densities
(J
sc,EQE
) 266

Table C.1 Optical properties of neat P3HTT-DPP-10%, P3HT
75
-co-
EHT
25
and P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM
blends at different ratios in thin films spin-coated from o-
DCB 283

Table C.2 2θ, interchain distances (100) and GIXRD intensities of neat
P3HTT-DPP-10%, P3HT
75
-co-EHT
25
and P3HTT-DPP-
10%:P3HT
75
-co-EHT
25
:PC
61
BM blends at different ratios in
thin films spin-coated from o-DCB 284

Table C.3 Hole mobilities of neat P3HTT-DPP-10%, P3HT
75
-co-EHT
25

and P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM blends at
different ratios in thin films spin-coated from o-DCB 290

Table C.4 Raw short-circuit current densities (J
sc
), spectral-mismatch
factor (M), spectral mismatch-corrected short-circuit current
densities (J
sc,corr
) and integrated short-circuit current densities
(J
sc,EQE
) for Table 5.1 290

Table C.5 Raw short-circuit current densities (J
sc
) for Table 5.1 without
spectral-mismatch factor correction, open-circuit voltage
(V
oc
), fill factor (FF) and efficiency (η) of P3HTT-DPP-
10%:P3HT
75
-co-EHT
25
:PC
61
BM ternary blend BHJ solar
cells under optimized conditions 291

Table C.6 Raw short-circuit current densities (J
sc
), open-circuit voltage
(V
oc
), fill factor (FF) and efficiency (η) of P3HTT-DPP-
10%:P3HT
75
-co-EHT
25
:PC
61
BM ternary blend BHJ solar
cells at constant overall polymer:fullerene ratio of 1:1.1 in
ternary blend composition regime 291

xi
Table C.7 Raw short-circuit current densities (J
sc
), open-circuit voltage
(V
oc
), fill factor (FF) and efficiency (η) of P3HTT-DPP-
10%:P3HT
75
-co-EHT
25
:PC
61
BM ternary blend BHJ solar
cells at constant overall polymer:fullerene ratio of 1:1.0 in
ternary blend composition regime 292

xii
List of Figures

Figure 1.1 Schematic illustration of a donor-acceptor bilayer solar cell
(upper left) as well as a bulk-heterojunction (BHJ) solar cell
(upper right). The magnified areas represent the bilayer
structure and the bicontinuous morphology respectively. The
typical current-voltage curves for dark and light current in
solar cells are shown at the bottom left and illustrate the
important parameters such as J
sc
(short-circuit current
density), V
oc
(open-circuit voltage), FF (fill factor) and J
m
as
well as V
m
which are current and voltage at the maximum
power point. The power conversion efficiency (η in %) can
be defined as the ratio of power out (P
out
) to power in (P
in
) or
in terms of the characteristic parameters of a solar cell, as
shown in the figure. The FF is the ratio of maximum power
divided by J
sc
× V
oc
. 4

Figure 1.2 Representative chemical and electronic structures of some of
the best performing polymers to date as well as the relevant
acceptors. HOMO and LUMO values are from
electrochemical measurements except for the LUMO level of
PBDTT-NAP (no electrochemical LUMO reported) and the
HOMO level of Indene-C
60
bisadduct (no electrochemical
HOMO reported). In these instances, optical data is used to
illustrate the band structure. Efficiencies for the BHJ solar
cells as well as the respective acceptor are given. Note that
for P3HT efficiencies are reported for BHJ cells using either
PC
60
BM or Indene-C
60
bisadduct as the acceptor. HOMO and
LUMO values have been adjusted for consistency so that
Fc/Fc
+
is taken as 5.1 eV below the vacuum in all cases. 7

Figure 1.3 Simplified schematic of the charge photogeneration process.
(i) light absorption and bound exciton formation; (ii) exciton
diffusion; (iii) charge transfer (CT) state population; (iv)
exciton dissociation to free charge carriers (charge separated
(CS) state population); (v) charge transport; (vi) charge
collection. Note that the energy levels labeled LUMO
-
and
HOMO
+
represent the energetic difference between the
ground state HOMO and LUMO and the HOMO and LUMO
corresponding to the excited state or charged species. 14

xiii
Figure 1.4 Energetic process that lead favorably to charge
photogeneration. (i) and (ii) light absorption and bound
exciton formation; (iii) CT state population; (iv) direct
formation of CS states; (v) formation of free charge carriers
(CS states); (vi) thermalization of free charge carriers. 15

Figure 1.5 Energetic processes that do not lead to charge
photogeneration. (i) luminescence; (ii) phosphorescence; (iii)
intersystem crossing on the donor; (iv) thermalization of
“hot” excitons; (v) intersystem crossing on the CT state; (vi)
geminate recombination on the donor’s triplet state; (vii)
geminate recombination to the donor’s singlet state; (viii)
geminate recombination to the ground state; (ix)
thermalization of the CT state; (x) recombination from the
CS* state to the CT state; (xi) bimolecular recombination. 20

Figure 1.6 The chemical structures of the six polymers based on the
PNDT-DTBT backbone. NDT = naphto[2,1-b:3,4-
b’]dithiophene; DTBT = 4,7-di(thiophen-2-
yl)benzothiadiazole. 28

Figure 1.7 Energy levels for the excited states of the described donor
polymer with a balance between the maximum V
oc
and the
minimal corresponding optical band gap relative to C
60
-
PCBM as the acceptor. 32

Figure 1.8 HOMO-LUMO diagram of the described donor polymer
relative to C
60
-PCBM as the acceptor. 33

Figure 1.9 Levels of BHJ solar cell structural organization. (i). Overall
organic solar cell; (ii). Layers of the BHJ device; (iii). BHJ
structure of the active layer; (iv). Donor-acceptor patterns in
the active layer; (v). Unknown structure at the donor-
acceptor interface. 43

Figure 1.10 Device architectures of a standard

cell (left)

with a
P3HT:PCBM active layer, an inverted

cell (center)

with
electrodes in reverse order and additional transition metal
oxide layer, and a tandem

cell (right) with P3HT:C
70
BM and
PCPDTBT:PCBM active layers. 45

Figure 1.11 Structures of polymer and fullerene components designed to
improve the stability of the morphology in the BHJ active
layer. 50
xiv
Figure 2.1 Simplified energy band diagram of organic donor:acceptor
solar cell. 69

Figure 2.2 Global total photon flux from the sun and integrated short-
circuit current density (J
sc
). 70

Figure 2.3 Device architecture and morphology of ternary blend bulk
heterojunction solar cells in case of two donors (top) and two
acceptors (bottom). 73

Figure 2.4 Absorption profile of an acceptor (A), high bad gap donor
(D1), low band gap donor (D2) as well as binary (A+D1),
(A+D2) and ternary (A+D1+D2) blends. 74

Figure 2.5 Solar spectrum covering in binary and ternary blends. 75

Figure 2.6 Chemical structures and external quantum efficiency (EQE)
values of binary and ternary blend BHJ solar cells with
PCBM acceptors. 77

Figure 2.7 Open-circuit voltage (V
oc
) of corresponding binary blend
BHJ solar cells. 79

Figure 2.8 Proposed origin of the V
oc
pinning in ternary blend BHJ solar
cells upon light absorption by D2. 80

Figure 2.9 Proposed origin of the V
oc
pinning in ternary blend BHJ solar
cells upon light absorption by D1. 80

Figure 2.10 Chemical structures of polymer and fullerene pairs used in
ternary blend BHJ solar cells with the non-pinned V
oc
. 82

Figure 2.11 Structures and corresponding electro-optical properties of
P3HT, ICBA and PC
61
BM. 86

Figure 2.12 The open-circuit voltage (V
oc
) in P3HT:PC
61
BM:ICBA
ternary blend BHJ solar cells. 87

Figure 2.13 Structures and corresponding electro-optical properties of
PC
61
BM, P3HT
75
-co-EHT
25
and P3HTT-DPP-10%. 89

xv
Figure 2.14 Open-circuit voltage (V
oc
) (black squares – left axis) and
short-circuit current density (J
sc
) (red circles – right axis) of
the individually optimized ternary blend BHJ solar cells from
Table 2.3 with different fraction of the polymer P3HT
75
-co-
EHT
25
component in the blends. 90

Figure 2.15 Open-circuit voltage (V
oc
) of the individually optimized
ternary blend BHJ solar cells (open squares), with fixed
overall polymer:PC
61
BM ratio at 1:1.1 (blue stars) and with
fixed overall polymer:PC
61
BM ratio at 1:1.0 (green
triangles). 91

Figure 2.16 Tunable open-circuit voltage (V
oc
) in AnE-PVba:AnE-
PVab:PC
61
BM ternary blend BHJ solar cells. 94

Figure 2.17 Tunable open-circuit voltage (V
oc
) in
P3HT:OXCMA:OXCBA ternary blend BHJ solar cells. 95

Figure 2.18 Tunable open-circuit voltage (V
oc
) in P3HT:CdSe
NR
:CdSe
NC

ternary blend BHJ solar cells. 95

Figure 2.19 Photocurrent spectral response (PSR) and energy of CT state
compared to the values of the V
oc
data for the
P3HT:PC
61
BM:ICBA. 101

Figure 2.20 Photocurrent spectral response (PSR) and energy of CT state
compared to the values of the V
oc
data for the P3HTT-DPP-
10%:P3HT
75
-co-EHT
25
:PC
61
BM ternary blend solar cells. 101

Figure 2.21 Expanded plot of the peaks near 1.7 eV in the PSR spectra of
P3HT:PC
61
BM:ICBA. The peak centered above 1.7 eV
corresponds to PC
61
BM absorption and the peak centered
below 1.7 eV corresponds to ICBA. 103

Figure 2.22 High energy PSR data of P3HTT-DPP-10%:P3HT
75
-co-
EHT
25
:PC
61
BM show the exciton peaks from the donor
mixture. 103

Figure 2.23 Exciton and charge transfer (CT) exciton in organic BHJ
solar cells. 106

xvi
Figure 2.24 Composition dependent HOMO energy level (HOMO
D1-D2
)
of an organic alloy blend based on two donors (top) and
composition dependent LUMO energy level (LUMO
A1-A2
) of
an organic alloy blend based on two acceptors (bottom).
Arrows indicate composition dependent CT state. LUMO
D1-
D2
and HOMO
D1-D2
drawn unchanged for simplification. 107

Figure 3.1 Structures and corresponding HOMO and LUMO energy
levels of P3HT, ICBA and PC
61
BM. 128

Figure 3.2 Open-circuit voltage (V
oc
) of the ternary blend BHJ solar
cells as a function of the amount of ICBA in the blends. 132

Figure 3.3 UV-vis absorption spectra of thin films spin-coated from
chlorobenzene (CB) and annealed at 150 °C under N
2
for 20
min with P3HT:PC
61
BM:ICBA ratios: (i) is 1:1:0 (red line),
(ii) is 1:0.8:0.2 (blue line), (iii) is 1:0.5:0.5 (green line), (iv)
is 1:0.2:0.8 (black line) and (v) is 1:0:1 (purple line). 134

Figure 4.1 Synthesis and structures of P3HTT-DPP-5%, P3HTT-DPP-
10% and P3HTT-DPP-15%. 145

Figure 4.2 UV-vis absorption spectra of polymers in solution (o-
dichlorobenzene or o-DCB) where (i) is P3HT, (ii) is
P3HTT-DPP-5%, (iii) is P3HTT-DPP-10% and (iv) is
P3HTT-DPP-15%. 149

Figure 4.3 UV-vis absorption spectra of polymers in thin film (spin-
coated from o-DCB) where (i) is P3HT (annealed at 150

°C
for 30 min for the thin films), (ii) is P3HTT-DPP-5% (thin
film as-cast), (iii) is P3HTT-DPP-10% (thin film as-cast) and
(iv) is P3HTT-DPP-15% (thin film as-cast). 149

Figure 4.4 External quantum efficiency of the BHJ solar cells based on
P3HT (black squares), P3HTT-DPP-5% (red circles),
P3HTT-DPP-10% (green triangles) and P3HTT-DPP-15%
(blue stars) with PC
61
BM as the acceptor, under optimized
conditions for device fabrication. 154

Figure 4.5 Synthesis of monomer 2-bromo-5-trimethyltin-3-(2-
ethylhexyl)thiophene (3) and Stille polymerization for
poly(3-hexylthiophene-co-3-(2-ethylhexyl)thiophene). 165

xvii
Figure 4.6 UV-vis absorption of all five polymers in solution (CB)
where (i) is P3HT (purple line), (ii) is P3HT
90
-co-EHT
10

(green line), (iii) is P3HT
75
-co-EHT
25
(blue line), (iv) is
P3HT
50
-co-EHT
50
(red line) and (v) is P3EHT (orange line). 167

Figure 4.7 UV-vis absorption of all five polymers in thin films (spin
coated from CB and annealed for 30 min under N
2
at 150 °C
for (i), (ii) and (iii), 100 °C for (iv) and 40 °C for (v)) where
(i) is P3HT (purple line), (ii) is P3HT
90
-co-EHT
10
(green
line), (iii) is P3HT
75
-co-EHT
25
(blue line), (iv) is P3HT
50
-co-
EHT
50
(red line) and (v) is P3EHT (orange line). 167

Figure 4.8 Grazing-incidence X-ray diffraction of thin films spin-coated
from CB and annealed for 30 min under N
2
(annealing
temperature was 150 °C for (i)-(iii), 100 °C for (iv) and 40
°C for (v)) are shown where (i) is P3HT (purple line), (ii) is
P3HT
90
-co-EHT
10
(green line), (iii) is P3HT
75
-co-EHT
25

(blue line), (iv) is P3HT
50
-co-EHT
50
(red line) and (v) is
P3EHT (orange line). The inset shows the region around 2θ =
4º - 8º in greater detail. 169

Figure 4.9 Melting points of polymers as measured by DSC. 170

Figure 4.10 J-V curves of the BHJ solar cells based on (i) P3HT (purple
line), (ii) P3HT
90
-co-EHT
10
(green line), (iii) P3HT
75
-co-
EHT
25
(blue line), (iv) P3HT
50
-co-EHT
50
(red line) and (v)
P3EHT (orange line) with PC
61
BM as the acceptor under AM
1.5G illumination (100 mW/cm
2
) at the optimal conditions
for solar cell performance. 173

Figure 4.11 HOMO levels of the polymers in the solid state (filled
squares) and V
oc
(circles) of the optimized solar cells as a
function of amount of 2-ethylhexyl side-chains in the
polymer backbone. 174

Figure 4.12 TEM images of polymer:PC
61
BM films (optimized
conditions for best solar cell performance were used to make
the films) where (a) is P3HT, (b) is P3HT
90
-co-EHT
10
, (c) is
P3HT
75
-co-EHT
25
, (d) is P3HT
50
-co-EHT
50
and (e) is
P3EHT. 175

Figure 5.1 Structures, corresponding HOMO energy levels and
absorption profiles of PC
61
BM (black line), P3HT
75
-co-
EHT
25
(red line) and P3HTT-DPP-10% (green line). 188
xviii
Figure 5.2 Open-circuit voltage (V
oc
) (black squares – left axis) and
short-circuit current density (J
sc
) (red circles – right axis) of
the individually optimized ternary blend BHJ solar cells from
Table 5.1 with different fraction of the polymer P3HT
75
-co-
EHT
25
component in the blends. 192

Figure 5.3 Open-circuit voltage (V
oc
) of the individually optimized
ternary blend BHJ solar cells (open squares), with fixed
overall polymer:PC
61
BM ratio at 1:1.1 (blue stars) and with
fixed overall polymer:PC
61
BM ratio at 1:1.0 (green
triangles). 192

Figure 5.4 External quantum efficiency of ternary blend BHJ solar cells
where (i) is 1:0:1.3 (red line), (ii) is 0.9:0.1:1.1 (green line),
(iii) is 0.8:0.2:1.0 (blue line), (iv) is 0.7:0.3:1.0 (cyan line),
(v) is 0.6:0.4:1.0 (magenta line), (vi) is 0.5:0.5:0.9 (wine-red
line), (vii) is 0.4:0.6:0.9 (olive line), (viii) is 0.3:0.7:0.8 (dark
yellow line), (ix) is 0.2:0.8:0.8 (purple line), (x) is 0.1:0.9:0.9
(yellow line) and (xi) is 0:1:0.8 (black line). 194

Figure 5.5 Grazing-incidence X-ray diffraction of thin films where (i) is
1:0:1.3 (red line), (ii) is 0.9:0.1:1.1 (green line), (iii) is
0.8:0.2:1.0 (blue line), (iv) is 0.7:0.3:1.0 (cyan line), (vi) is
0.5:0.5:0.9 (wine-red line), (viii) is 0.3:0.7:0.8 (dark yellow
line), (x) is 0.1:0.9:0.9 (yellow line) and (xi) is 0:1:0.8 (black
line). 195

Figure 6.1 Photocurrent spectral response (PSR) data for the
P3HT:PCBM:ICBA (D:A1
X
A2
(1-X)
) ternary blend solar cells
plotted as a function of ICBA fraction in PCBM:ICBA pair.
The inset indicates the CT transition or interface band gap
that is being measured and the pair of dashed lines indicates
the range over which the interface band gap energy is
extracted. 203

Figure 6.2 Expanded plot of the peaks near 1.7 eV in the PSR spectra of
Figure 6.1, with the background subtracted. The peak
centered above 1.7 eV corresponds to PCBM absorption and
the peak centered below 1.7 eV corresponds to ICBA. 204

xix
Figure 6.3 Plot of the estimated interface band gap energy defined by
PSR at photocurrent (PC) values of 0.1 and 0.01 from the
data of Figure 6.1 (closed symbols) compared to the values
of the V
oc
(open triangles) for P3HT:PCBM:ICBA
(D:A1
X
A2
(1-X)
) ternary blend solar cells. The solid lines are
the model of Equation 1 with the same bowing parameter b. 205

Figure 6.4 Photocurrent spectral response (PSR) data for the P3HTT-
DPP-10%:P3HT
75
-co-EHT
25
:PCBM (D1
X
D2
(1-X)
:A) ternary
blend solar cells plotted as a function of P3HT
75
-co-EHT
25

fraction in the polymer donor pair. The pair of dashed lines
indicates the range over which the interface band gap energy
is extracted. The inset shows the photocurrent signal at 1.6
eV as a function of P3HT
75
-co-EHT
25
fraction in the polymer
donor pair. 207

Figure 6.5 High energy PSR data from Figure 6.4 plotted on a linear
scale to show the exciton peaks from the donor mixture.
Solid lines are fits to the E1, E2 and E3 peaks. The P3HT
75
-
coEHT
25
fraction is indicated for each data set. 208

Figure 6.6 Plot of the estimated interface band gap energy defined by
PSR at photocurrent (PC) values of 0.1 and 0.01 from the
data of Figure 6.4 (closed symbols) compared to the values
of the V
oc
(open triangles), as a function of P3HT
75
-co-EHT
25

fraction for the mixed donor system. Solid lines are linear fit
to the data with the same slope (zero bowing factor). 209

Figure A.1 UV-vis absorption spectra of thin films spin-coated from
chlorobenzene (CB) and annealed at 150 °C under N
2
for 20
min, where (i) is P3HT (black line), (ii) is PC
61
BM (red line)
and (iii) is ICBA (blue line). 252

Figure A.2 Grazing-incidence X-ray diffraction of thin films of
P3HT:PC
61
BM:ICBA spin-coated from chlorobenzene (CB)
and annealed at 150 °C under N
2
for 20 min, where (i) is
1:1:0 (red line), (ii) is 1:0.2:0.8 (blue line), (iii) is 1:0.5:0.5
(green line), (iv) is 1:0.8:0.2 (black line) and (v) is 1:0:1
(purple line). 252

xx
Figure A.3 J-V curves of the ternary blend BHJ solar cells based on
P3HT:PC
61
BM:ICBA at different ratios: 1:1:0 (red line),
1:0.9:0.1 (navy line), 1:0.8:0.2 (blue line), 1:0.7:0.3 (olive
line), 1:0.6:0.4 (cyan line), 1:0.5:0.5 (green line), 1:0.4:0.6
(pink line), 1:0.3:0.7 (wine red line), 1:0.2:0.8 (black line),
1:0.1:0.9 (dark yellow line) and 1:0:1 (purple line) under AM
1.5G illumination (100 mW/cm
2
) at thicknesses 95 – 105 nm
presented in Table 3.1. 255

Figure A.4 J-V curves of the optimized ternary blend BHJ solar cells
based on P3HT:PC
61
BM:ICBA at different ratios: 1:1:0 (red
line), 1:0.5:0.5 (green line), 1:0:1 (purple line) under AM
1.5G illumination (100 mW/cm
2
) at thicknesses 104 nm, 137
nm and 174 nm, respectively. 256

Figure A.5 External quantum efficiency of the ternary blend BHJ solar
cells based on P3HT:PC
61
BM:ICBA at different ratios: 1:1:0
(red circles), 1:0.5:0.5 (green circles), 1:0:1 (purple circles)
at thicknesses 95 – 105 nm presented in Table 3.1. 256

Figure A.6 TEM images of P3HT:PC
61
BM:ICBA at (a) 1:1:0, (b)
1:0.5:0.5 and (c) 1:0:1 at thicknesses 95 – 105 nm for BHJ
solar cells presented in Table 3.1 (scale bar is 50 nm). 257

Figure B.1 Grazing-incidence X-ray diffraction of thin films of (i) P3HT
(spin-coated from cholorobenzene and annealed at 150

°C for
30 min under N
2
) (black line), (ii) P3HTT-DPP-5% (spin-
coated from o-dichlorobenzene (o-DCB) and annealed at 150

°C for 30 min under N
2
) (red line), (iii) P3HTT-DPP-10%
(spin-coated from o-DCB and tested as-cast) (green line), (iv)
P3HTT-DPP-10% (spin-coated from o-DCB and annealed at
150

°C for 30 min under N
2
) (purple line) and (v) P3HTT-
DPP-15% (spin-coated from o-DCB and annealed at 150

°C
for 30 min under N
2
) (blue line). 261

Figure B.2 CV traces for the oxidation of P3HTT-DPP-5%, P3HTT-
DPP-10% and P3HTT-DPP-15%. 262

Figure B.3 J-V curves of the BHJ solar cells based on P3HT (black line),
P3HTT-DPP-5% (red line), P3HTT-DPP-10% (green line)
and P3HTT-DPP-15% (blue line) with PC
61
BM as the
acceptor under AM 1.5G illumination (100 mW/cm
2
) at the
optimal conditions for solar cell performance. 267

xxi
Figure B.4 TEM images of (a) P3HT:PC
61
BM, (b) P3HTT-DPP-
5%:PC
61
BM, (c) P3HTT-DPP-10%:PC
61
BM and (d) P3HTT-
DPP-15%:PC
61
BM prepared under optimal solar cells
conditions. 268

Figure B.5 UV-vis absorption of all five polymers in as cast thin films
(spin coated from CB) where P3HT is purple line, P3HT
90
-
co-EHT
10
is green line, P3HT
75
-co-EHT
25
is blue line,
P3HT
50
-co-EHT
50
is red line and P3EHT is orange line. 272

Figure B.6 CV traces for the oxidation of thin films (as cast) where a) is
P3HT, b) is P3HT
90
-co-EHT
10
, c) is P3HT
75
-co-EHT
25
, d) is
P3HT
50
-co-EHT
50
and e) is P3EHT. Ferrocene was used as a
reference and values were converted to the vacuum scale
using the approximation that the ferrocene redox couple is
5.1 eV relative to vacuum. 273

Figure B.7 CV traces for the oxidation of polymers in solution (CHCl
3

with tetrabutylammonium tetrafluoroborate as supporting
electrolyte) where a) is P3HT, b) is P3HT
90
-co-EHT
10
, c) is
P3HT
75
-co-EHT
25
, d) is P3HT
50
-co-EHT
50
and e) is P3EHT.
Ferrocene was used as a reference and values were converted
to the vacuum scale using the approximation that the
ferrocene redox couple is 5.1 eV relative to vacuum. 274

Figure B.8 DSC curves where a) is P3HT, b) is P3HT
90
-co-EHT
10
, c) is
P3HT
75
-co-EHT
25
and d) is P3HT
50
-co-EHT
50
. 275

Figure B.9 TEM images of polymer:PCBM films (optimized conditions
for best solar cell performance) where a) is P3HT, b) is
P3HT
90
-co-EHT
10
, c) is P3HT
75
-co-EHT
25
, d) is P3HT
50
-co-
EHT
50
and e) is P3EHT. 278

Figure C.1 UV-vis absorption spectra of thin films spin-coated from o-
dichlorobenzene (o-DCB) and placed to N
2
cabinet for 30
min with P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM
ratios, where (i) is 1:0:1.3 (red line), (ii) is 0.9:0.1:1.1 (green
line), (iii) is 0.8:0.2:1.0 (blue line), (iv) is 0.7:0.3:1.0 (cyan
line), (v) is 0.5:0.5:0.9 (wine-red line), (vi) is 0.3:0.7:0.8
(dark yellow line), (vii) is 0.1:0.9:0.9 (yellow line) and (viii)
is 0:1:0.8 (black line). 282

xxii
Figure C.2 Grazing-incidence X-ray diffraction of thin films of P3HTT-
DPP-10% and P3HT
75
-co-EHT
25
spin-coated from o-
dichlorobenzene (o-DCB) and placed to N
2
cabinet for 20
min, where (i) is P3HT
75
-co-EHT
25
(red line) and (ii) is
P3HTT-DPP-10% (green line). 283

Figure C.3 J-V curves of the ternary blend BHJ solar cells based on
P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM at different
ratios: (i) is 1:0:1.3 (red line), (ii) is 0.9:0.1:1.1 (green line),
(iii) is 0.8:0.2:1.0 (blue line), (iv) is 0.7:0.3:1.0 (cyan line),
(v) is 0.6:0.4:1.0 (magenta line), (vi) is 0.5:0.5:0.9 (wine-red
line), (vii) is 0.4:0.6:0.9 (olive line), (viii) is 0.3:0.7:0.8 (dark
yellow line), (ix) is 0.2:0.8:0.8 (purple line), (x) is 0.1:0.9:0.9
(yellow line) and (xi) is 0:1:0.8 (black line) under AM 1.5G
illumination (100 mW/cm
2
) presented in Table 5.1. 287

Figure C.4 TEM images of P3HTT-DPP-10%:P3HT
75
-co-
EHT
25
:PC
61
BM at (a) 1:0:1.3, (b) 0.9:0.1:1.1, (c) 0.8:0.2:1.0,
(d) 0.5:0.5:0.9, (e) 0.2:0.8:0.8 and (f) 0:1:0.8 for BHJ solar
cells presented in Table 5.1 (scale bar is 50 nm). 288

xxiii
Abstract

The growing energy demands facilitate the search of cheap and renewable energy.
The vision of organic photovoltaics is that of a low cost solar energy conversion platform
that provides lightweight, flexible solar cells that are easily incorporated into existing
infrastructure with minimal impact on land usage. Organic solar cells have been a subject
of growing research interest over the past quarter century, and are now developed to the
point where they are on the verge of introduction into the market. Polymer solar cells
provide all important characteristics necessary to fill this niche and power conversion
efficiency exceeding 10% was recently achieved. However, such a high efficiencies are
obtained using multi-junction solar cells which require additional steps in the
manufacturing process and thus contradicts the attractive simplicity of the single step
solution processing of the active layer in bulk heterojunction solar cells. A basic
introduction to polymer:fullerene solar cells is presented in Chapter 1.
The current dissertation is focused on the recently emerged approach of ternary
blend bulk heterojunction solar cells, where three components are mixed in the active
layer, providing an alternative route to high efficiencies while preserving the simplicity
of the solar cell fabrication. A general overview of this class of solar cells is presented in
Chapter 2.
The key component to the efficiency increase in ternary blend bulk heterojunction
solar cells is the discovered open-circuit voltage (V
oc
) tunability in the three component
systems, which is introduced in Chapter 3. Together with the short-circuit current density
xxiv
(J
sc
) increase obtained using polymers with complementary absorption profiles (discussed
in Chapter 4), the efficiencies of the ternary blend bulk heterojunction solar cells are
increased beyond that of the corresponding binary blend solar cells, as discussed in
Chapter 5. By extension, this result suggests that ternary blends provide a potentially
effective route toward maximizing the attainable J
sc
×V
oc
product (which is directly
proportional to the solar cell efficiency) in bulk heterojunction solar cells and that with
judicious selection of donor and acceptor components, solar cells with efficiencies
exceeding the theoretical limits for binary blend solar cells could be possible without
sacrificing the simplicity of a single active layer processing step.
Finally, the formation of an organic alloy in ternary blend bulk heterojunction
solar cells, introduced in Chapter 6, unravels and explains the origin of the V
oc
tunability
and simultaneous J
sc
increase in three component systems.

1
Chapter 1

Polymer-Based Solar Cells: State-of-the-Art Principles for the Design of Active
Layer Components

1.1 Introduction

With the world energy demand increasing, current trends in energy supply are
economically and environmentally unsustainable. Consequently, there is a growing
realization of the necessity for clean and renewable energy sources, such as hydropower,
geothermal, wind, and solar. As the planet’s most abundant potential source of energy,
the sun can provide enough energy to the surface of the earth in a single hour to meet the
energy demand of the world population for an entire year. However, solar energy is a
largely untapped energy source, accounting for less than 1% of the global energy
supply.
1,2

The photovoltaic market is currently dominated by silicon-based solar cells.
Single crystal silicon cells have efficiencies exceeding 20%, although application of
single or multi crystalline silicon solar cells is primarily limited due to high fabrication
costs. Further restrictions come from the size of solar panels as current silicon
photovoltaic technology is bulky and heavy. As a growing potential alternative, thin film
solar cell technology has rapidly progressed in recent years.
3,4
Materials like amorphous
silicon, CdTe and CIS/CIGS (Copper Indium Gallium Selenide) as well as organic
2
materials, offer the advantage of higher absorption coefficients, use of less material and
an inexpensive fabrication process. Inorganic thin film photovoltaics have laboratory
efficiencies of up to 19.9% (CIGS) and organic devices are rapidly catching up with
efficiencies now being over 8%.
5
With growing efficiencies and considerably lower
potential costs of production, thin film solar technologies are projected to accelerate the
usage of solar power around the world.
The vision of organic photovoltaics is that of a low cost solar energy conversion
platform that provides lightweight, flexible solar cells that are easily incorporated into
existing infrastructure with minimal impact on land usage. Organic solar cells have been
a subject of growing research interest over the past quarter century, but are now
developed to the point where they are on the verge of introduction into the market.
Compared to inorganic semiconductors, organic materials generally have the advantage
of very high absorption coefficients, which likely exceed 10
5
cm
-1
in thin films (~ 1000
times higher than silicon).
6,7
These high absorption coefficients allow the use of very thin
films (generally between 50 and 200 nm), which will absorb a large fraction of the
incident sunlight.
8
In addition, organic materials also have much lower densities than
inorganic semiconductors (poly(3-hexylthiophene) or P3HT = 1.1 g/cm
3
, silicon = 2.33
g/cm
3
).
9,10
The combination of high absorption coefficient and low density leads to
extremely small masses of active material used in organic solar cells, which promises
lightweight and potentially inexpensive devices.
The fundamental operating principle of an organic solar cell is based on the
cooperative interaction of molecular or polymeric electron donors and acceptors.
3
Typically, photoexcitation of the donor generates excitons (bound electron-hole pairs), as
opposed to free charges in the inorganic solar cells, due to the low dielectric constant of
organics.
11
These excitons will only find sufficient energetic driving force for
dissociation into free charges at the interface with an electron acceptor of suitably high
electron affinity. Excitons must therefore diffuse through the donor in order to reach an
acceptor site where charges can be generated and then finally be transported through the
donor phase (holes) and the acceptor phase (electrons). It is this necessity of having two
distinct and interacting species that is the defining characteristic of the organic solar cell.
Despite this common attribute, many different types of organic solar cells exist, which
can be grouped in two general categories distinguished by the architecture of the active
layer, with either a donor-acceptor bilayer or a bicontinuous donor-acceptor composite,
known as a bulk-heterojunction (BHJ). Figure 1.1 illustrates these two general organic
solar cell platforms. Dye sensitized solar cells or Grätzel cells,
12,13
do not fit into either
category as they operate on a fundamentally different principle based on a
photoelectrochemical process and will not be considered here.
14
Organic solar cells are
characterized by the output parameters of power conversion efficiency (η or PCE), open-
circuit voltage (V
oc
), short-circuit current density (J
sc
), and fill factor (FF), as are
illustrated in Figure 1.1.

4

Figure 1.1. Schematic illustration of a donor-acceptor bilayer solar cell (upper left) as
well as a bulk-heterojunction (BHJ) solar cell (upper right). The magnified areas
represent the bilayer structure and the bicontinuous morphology respectively. The typical
current-voltage curves for dark and light current in solar cells are shown at the bottom
left and illustrate the important parameters such as J
sc
(short-circuit current density), V
oc

(open-circuit voltage), FF (fill factor) and J
m
as well as V
m
which are current and voltage
at the maximum power point. The power conversion efficiency (η in %) can be defined as
the ratio of power out (P
out
) to power in (P
in
) or in terms of the characteristic parameters
of a solar cell, as shown in the figure. The FF is the ratio of maximum power divided by
J
sc
× V
oc
.

The first organic bilayer solar cell, consisting of vacuum deposited layers of
copper phthalocyanine and a perylene tetracarboxylic derivative, was published by Tang
in 1986.
15
The maximum PCE that has since been reached with molecular bilayer
(vacuum deposited) devices is 4.2% using copper phthalocyanine as the donor and C
60
as
the acceptor.
16
Adding a mixed donor acceptor interlayer to this system improved the
5
PCE to 5%.
17
The highest PCE for a polymer-polymer (solution processed) bilayer solar
cell is 2% and was achieved using poly[3-(4-n-octyl)-phenylthiophene] as the donor and
Cyano-PPV (PPV = poly-para-phenylenevinylene) as the acceptor.
18
In general, bilayer
devices have performances that are limited by the short exciton diffusion length in
organics. If the exciton is not created within a very short distance (generally < 10 nm) of
the donor-acceptor interface it will simply decay back to the ground state and not
contribute to the photocurrent.
11
Despite the high absorption coefficients of most
organics, more than a few nanometers of material are necessary to absorb a sufficient
fraction of incident light for practical function, and this inherently limits the efficiency of
bilayer solar cells.
The problem of short exciton diffusion length can largely be solved by using the
donor-acceptor BHJ approach introduced in the mid 90’s.
19,20
Conceptually, a BHJ is a
bicontinuous donor-acceptor composite with domain sizes on the order of the exciton
diffusion length, such that all excitons can reach an interface and dissociate into free
charge carriers, rendering the entire volume of the active layer effective for light
absorption. BHJ devices also have the advantage that the active layer is solution
processed in one single step, which is especially effective when at least one component is
a high molecular weight polymer. Solution processing uses minimal energy and can be
done at ambient temperature and pressure. At the laboratory level, spin coating is the
primary processing method used to make thin films, however, solution processing can be
easily adapted for large area devices
21,22
and is suitable for low cost mass production.
Some of the relevant techniques that are used are doctor blading,
23
screen printing,
21

6
inkjet printing,
24
pad printing,
25
knife-over-edge and slot-die coating,
26
spray coating,
27

roller painting,
28
and brush painting.
29

Numerous different types of donors and acceptors have been combined in BHJ
solar cells. The highest PCE for a BHJ solar cell based on two small molecules (using a
diketopyrrolopyrrole-oligothiophene donor and a fullerene acceptor, C
70
-PCBM) is
4.4%.
7
Polymer-polymer BHJ cells have shown relatively low power conversion
efficiencies with 1.8% being the maximum efficiency reported.
30–32
Combinations of
organic and inorganic components have given efficiencies of 2.6%
33
using P3HT and
CdSe tetrapods whereas a low band gap polymer with CdSe nanocrystals gave a PCE of
3.13%.
34

By far the most successful and most studied combination in BHJ solar cells is that
of conjugated polymer donors and fullerene acceptors. Fullerenes are an excellent choice
for an electron acceptor due to their high electron affinity and superior ability amongst
organics to function as electron transport materials.
35
This combination provides the
highest efficiencies of any organic solar cell to date, with efficiencies in excess of 8%
now reported.
5
In contrast, combinations of polymers with other small molecule
acceptors such as vinazenes, result in devices with much lower efficiencies.
36
Figure 1.2
illustrates the chemical and electronic structures of a representative selection of some of
the polymers and fullerenes that have led to some of the highest efficiency BHJ solar
cells reported in the literature to date, and the characteristics are summarized in Table
1.1.
7

Figure 1.2. Representative chemical and electronic structures of some of the best
performing polymers to date as well as the relevant acceptors. HOMO and LUMO values
are from electrochemical measurements except for the LUMO level of PBDTT-NAP (no
electrochemical LUMO reported) and the HOMO level of Indene-C
60
bisadduct (no
electrochemical HOMO reported). In these instances, optical data is used to illustrate the
band structure. Efficiencies for the BHJ solar cells as well as the respective acceptor are
given. Note that for P3HT efficiencies are reported for BHJ cells using either PC
60
BM or
Indene-C
60
bisadduct as the acceptor. HOMO and LUMO values have been adjusted for
consistency so that Fc/Fc
+
is taken as 5.1 eV below the vacuum in all cases.

Examination of the chemical and electronic structures in Figure 1.2 leads to
insight into the design principles that have been used until now to improve published
champion level efficiencies from 2.5% in 2001
37
to nearly 8% in 2010.
38
These design
principles are directly related to the output parameters in solar cells, where the PCE can
be expressed as η(%) = (J
sc
× V
oc
× FF) / (Input Power), when standard AM1.5G
conditions with light intensity of 100 mW/cm
2
are used.
8
Table 1.1. Overview of some of the highest performing BHJ solar cells. The optical band
gap (E
g
) and the electrochemical values for HOMO and LUMO levels are given where
reported (HOMO and LUMO values have been adjusted for consistency so that Fc/Fc
+
is
taken as a value of 5.1 eV below the vacuum for all cases)
Polymer
E
g

(eV)
HOMO
(eV)
LUMO
(eV)
Acceptor
(polymer:fullerene)
V
oc

(V)
J
sc
(mA/cm
2
)

FF
(%)
η
(%)
P3HT
39
1.9 5.2 3.2 PC
60
BM, 1:0.8 0.63 9.50 68.0 4.9
P3HT
40
1.9 5.2 3.2
Indene-C
60

bisadduct, 1:1
0.84 10.61 72.7 6.5
PCPDTBT
41
1.5 5.1 3.3 PC
70
BM, 1:2 0.62 16.20 55.0 5.5
TQ1
42
1.7 5.7 3.3 PC
70
BM, 1:3 0.89 10.50 64.0 6.0
PCDTBT
43
1.8 5.5 3.6 PC
70
BM, 1:4 0.88 10.60 66.0 6.1
PBDTTT-NAP
44
1.7 5.7

- PC
60
BM, 1:1.5 0.85 11.50 68.0 6.8
PBDTTT-CF
38
1.6 5.5

3.8

PC
70
BM, 1:1.5 0.76 15.20 66.9 7.7

An important design principle that has been used is based on the J
sc
, which is
related to the breadth of spectral absorption, where the broader the wavelength range of
light that can be absorbed, the greater the number of photons that can be harvested and
potentially lead to charge carriers.
45
Table 1.2 illustrates a correlation between the long
wavelength onset of absorption and the maximum possible photocurrent under AM1.5G
conditions. As such, the focus was placed on generating lower band gap polymers
capable of absorbing longer wavelength light, since P3HT, which is capable of generating
~5% efficient solar cells
39,46
with C
60
-PCBM, has an optical band gap of ~1.9 eV, which
was deemed to be too large. All of the polymers illustrated in Figure 1.2 give higher
efficiencies than P3HT using a PCBM acceptor and were designed to give longer
wavelength absorption and ultimately a broader absorption when considering the
combined absorption envelope with the acceptor. Synthetically, an alternating pattern of
electron-rich and electron-poor units was used in these polymer backbones to achieve the
longer wavelength absorption, based on the well-known donor-acceptor route to low band
gap polymers.
47

9
Table 1.2. The integrated photon flux and maximum current density available for a
photovoltaic device that harvests light from 280 nm up to the wavelength quoted,
assuming that every photon is converted to an electron that is collected in the external
circuit
45

Wavelength
Max. % harvested
(280 nm → )
Current density
(mA/cm
2
)
500 8.0 5.1
600 17.3 11.1
650 22.4 14.3
700 27.6 17.6
750 35.6 20.8
800 37.3 23.8
900 46.7 29.8
1000 53.0 33.9
1250 68.7 43.9
1500 75.0 47.9

Another design principle that has been used is based on the empirical correlation
that the magnitude of the V
oc
is directly related to the magnitude of the energetic
difference between the HOMO of the donor and the LUMO of the acceptor. For example,
the combination of P3HT and C
60
-PCBM, yields an offset of ~1 eV and the resulting V
oc
is ~0.6 eV. Based on this concept, a lower lying HOMO is thought to be desirable for any
polymer to be used in combination with PCBM and it can be seen that all but one
polymer in Figure 1.2 has a lower lying HOMO than P3HT, which is also a consequence
of the electron-poor units in the polymer backbone.
An alternative approach to increase the V
oc
, also based on the same concept of
maximizing the offset between the donor HOMO and the acceptor LUMO, is to use
fullerene acceptors that have higher lying LUMO energies than PCBM. Bis-Indene-C
60
serves as the most successful example of a fullerene acceptor with a higher lying
LUMO,
40
although other examples are known.
48,49
The effect is easily seen with P3HT
10
and bis-Indene-C
60
where the V
oc
increases from 0.6 V to 0.84 V when the acceptor is
changed (Table 1.1), resulting in an overall increase in PCE from 4.9% to 6.5%.
40

A further design principle that was followed for all the polymers in Figure 1.2 is
based on the empirical observation that a sufficient energetic offset between the LUMO
of the donor and the LUMO of the acceptor is required to generate a driving force for
electron transfer from donor to acceptor. It is thought that at least 0.3 eV downhill driving
force is required and that any energy offset significantly in excess of this amount
constitutes wasted energy.
35
Note that all of the polymers in Figure 1.2 exhibit LUMO
energies closer to that of the fullerene acceptors than for the case of P3HT.
It is clear based on the data shown in Table 1.1 that all of these highly optimized
donor-acceptor pairs lead to extremely high values of FF, approaching or exceeding 0.7.
This is a consequence of extensive optimization of the device fabrication and processing
methods, which has led to optimum values for active layer composition (donor-acceptor
ratio), active layer thickness, electrode choice, solvent choice for spin-coating, and
treatment of the active layer by judicious thermal and/or solvent annealing techniques.
50

The consequence of these optimizations has been an improvement in morphology
(bicontinuity and domain size), charge transport, and carrier collection, all combining to
give high FF and PCE values.
The rapid increase in efficiency from 2.5% to more than 8% in the span of less
than a decade has moved polymer-fullerene solar cells to the verge of commercial
viability. However, it is generally agreed that further increases in efficiency will be
required before these organic solar cells can become competitive with their thin film
11
inorganic counterparts. Several research groups have tried to predict the maximum
attainable efficiency that can be achieved with polymer-fullerene BHJ solar cells,
although different methods have been used, most estimates are between 10 and 11%.
51–55

A specific route toward such efficiencies is not well-defined, although it does appear that
development of new polymer and fullerene derivatives will be required. Such
development must be based on a new set of more detailed design principles that can only
be established through the rigorous elucidation of the fundamental physical principles
that govern the photovoltaic process. The preceding discussion of the current state-of-the-
art in polymer-fullerene BHJ solar cells highlights a critical point about this field in that
most of the progress toward high efficiency has come largely as a result of a trial and
error approach based loosely on a number of empirical correlations.
In BHJ solar cells, the two most fundamental aspects that lead to high efficiency
are the electronic and physical interactions between the donor and acceptor components
in the composite film. Electronic interactions are defined as those interactions that are
related to the electronic structures of the donor and acceptor components and the
relationship between these electronic structures, whereas physical interactions are defined
as the interactions that determine the miscibility of the components and the physical
structure of the composite at both the bulk and molecular levels. These interactions
govern the V
oc
, the J
sc
, and (to a large extent) the FF of the solar cell. These two
interactions are also intimately intertwined. In order to arrive at a more rigorous set of
design principles for more optimal polymer and fullerene components for BHJ solar cells,
the fundamental questions that must be answered are: 1. What is the complete electronic
12
pathway from absorbed photon to collected charges at the electrodes? 2. What is the
optimal relationship between the electronic structures of a donor and an acceptor? 3.
What is the optimal bulk morphology and how does this relate to the chemical structure
of the donor and acceptor component? 4. What is the optimal structure of the donor-
acceptor interface at the molecular level? If these questions can be answered, design
principles for donors and acceptors that will give optimally performing BHJ solar cells
should follow and ultimately lead to competitive efficiencies in excess of 10%. This
chapter is designed to unite the results of selected recent works that significantly
contribute to answering these questions with an eye toward the design of optimal donor-
acceptor pairs, and includes consideration of the overall device architecture and the
necessity for long-term stability.

1.2 Electronic Processes and Interactions in the Donor-Acceptor Composite

1.2.1 The Charge Photogeneration Pathway

A critical point in designing new components for BHJ polymer-fullerene solar
cells is the establishment of rational design principles that are rooted in a deep
understanding of the fundamental processes that govern the charge photogeneration
pathway, which has undergone significant clarification in recent years. Figure 1.3
illustrates a diagram of the principle steps in the photocurrent generation process that lead
to charge generation and collection, assuming the donor is the only light absorbing
13
component, although the process is facilitated by acceptor absorption as well. In the first
step, light is absorbed by the polymer donor, resulting in the creation of the strongly
bound exciton, which can relax to the S
1
state, although higher energy excitons (S
n
) exist
in the system.
56
Exciton formation is possible only if the energy of the absorbed photon is
larger than the bandgap of the donor (the same mechanism can also be applied if the
photon is absorbed by the acceptor).
57,58
Because the dielectric constant of organic
molecules is small (ε
r
≈ 3) in comparison to inorganic materials (ε
r
≈ 10),
11
strong
binding between the electron and the hole results in the formation of Frenkel-type
excitons, which are characterized by large binding energies (E
B
exc
) on the order of 0.3 –
0.5 eV (this is much greater than the thermal energy kT ≈ 25 meV).
11,59–61
The exciton
created is a singlet, since the spin-orbit coupling to the triplet state (T
1
) is typically
negligible. However, in the presence of heavy atoms, intersystem crossing from singlet to
triplet state may be enhanced in some cases.
62–64
The exciton lifetime can vary from a
range of 100 ps to 1 ns for singlets,
49,63,65–67
up to microseconds for triplets
62,68,69
in neat
polymers, and dramatically decreases when an acceptor is present in the system.
49,70

14

Figure 1.3. Simplified schematic of the charge photogeneration process. (i) light
absorption and bound exciton formation; (ii) exciton diffusion; (iii) charge transfer (CT)
state population; (iv) exciton dissociation to free charge carriers (charge separated (CS)
state population); (v) charge transport; (vi) charge collection. Note that the energy levels
labeled LUMO
-
and HOMO
+
represent the energetic difference between the ground state
HOMO and LUMO and the HOMO and LUMO corresponding to the excited state or
charged species.

If the exciton is not generated directly at the donor-acceptor interface, it will
diffuse within the donor phase via a random walk process,
71–73
which constitutes the
second principle step in the charge photogeneration pathway. Singlet exciton diffusion is
generally described as a hopping process via a Förster mechanism, in which no direct
contact between the hopping sites is required. On the other hand, triplets diffuse
according to a Dexter-type mechanism, which requires spatial overlap of wavefunctions
via direct molecular contact.
74
Because of the difference in hopping processes and
exciton lifetimes, singlet exciton diffusion length is in general about 5-10 nm,
70,75–78

while triplets can diffuse tens of nanometers.
79–82
Absorption of light with the fullerene,
creates excitons with diffusion length in the range 5-40 nm, depending on the exciton
15
type,
83
however triplet states dominate due to the near-unity (95%) intersystem crossing
efficiency.
82

Figure 1.4. Energetic process that lead favorably to charge photogeneration. (i) and (ii)
light absorption and bound exciton formation; (iii) CT state population; (iv) direct
formation of CS states; (v) formation of free charge carriers (CS states); (vi)
thermalization of free charge carriers.

If the exciton reaches the donor-acceptor interface, an initial electron transfer
event can occur, generating an interfacial bound electron-hole pair referred to as a charge
transfer state (abbreviated CT state), which is illustrated in step (iii) in Figure 1.3. In the
CT state, there is a Coulombic binding between a hole that is primarily in the HOMO of
the donor and an electron that is primarily in the LUMO of the acceptor. This CT state
has been referred to by many names including, geminate pair, bound polaron pair, charge
16
transfer exciton, and exciplex. It is important to consider which factors favor the
formation of the CT state, what are the possible fates of this CT state once formed, and
which factors favor conversion to fully separated charges (referred to as charge separated
states or CS states or polaron pairs) that can be collected as photocurrent at the
electrodes.
Figure 1.4 illustrates a detailed energy diagram centered around the CT state,
focusing on the processes that lead to efficient charge separation. The CT state will be
populated at the interface of the donor and the acceptor if there is a suitable energetic
driving force for the initial electron transfer event. Forward electron transfer from the S
1
state of the donor (with energy = E
S1
) to the interfacial CT state (with energy = E
CT
)
occurs when the free energy change in the process ( ∆G
CT
= E
S1
- E
CT
) is
thermodynamically favorable, but as little as 0.1 eV of driving force is sufficient to favor
the formation of the CT state.
84
The value of E
CT
for a given donor-acceptor pair is
estimated as: E
CT
= (IP
D
– EA
A
) + E
B
CT
, where IP
D
refers to the ionization potential of the
donor (or HOMO energy of the donor), EA
A
refers to the electron affinity of the acceptor
(or LUMO energy of the acceptor), and E
B
CT
refers to the binding energy of the CT state,
which is typically somewhat smaller than the binding energy of the exciton (E
B
exc
) due to
the larger intermolecular spatial charge separation of the CT state and will be on the order
of 0.1 – 0.5 eV.
84
Conceptually, the binding energy of the CT state can be thought of as
the energy difference between the thermally relaxed CT state (CT
0
) and the two fully
dissociated and thermally relaxed free charge carriers (CS states).
17
Recently, theoretical calculations as well as experimental data have shown that
population of the CT state is more efficient beginning with higher lying singlet states (S
n
)
due to the better overlap with the CT states, and thus facilitating their population.
59,85–87

In general, CT* states (referred to as hot CT states) will be initially populated due to the
“horizontal” electron transfer from the S
n
states or the S
1
state, as dictated under Marcus
electron transfer theory.
84
The CT* states have smaller binding energies than the CT
0
,
due to a greater electron-hole separation in these vibrationally hot states, most likely on
the order of only ~ 0.1 eV.
84,86,88–90

The remainder of the favorable pathway to charge separation occurs once the CT*
states are populated and is illustrated in Figure 1.4. In order for a CT* state to lead to
free charge carriers (CS states), there must be an energetic driving force to overcome the
Coulombic attraction of the electron-hole pair. A horizontal transfer from a CT* state to a
hot CS state (CS*) state can occur. The initially formed hot CS states will then relax
down to the ground CS state. This process will be favorable if the free energy difference
between CT
0
and CS exceeds the magnitude of the binding energy of the CT
0
state. The
energy of the CS

state is equal to the energetic difference of IP
D
and EA
A
. As such, the
overall energetic driving force for the charge separation process ( ∆G
CS
) is equal to the
energetic difference of the first singlet excited state (E
S1
) and the free, thermally relaxed
charge carriers (IP
D
– EA
A
), or ∆G
CS
= E
S1
- (IP
D
– EA
A
), as illustrated in Figure 1.4.
In the overall charge separation process the key step is the dissociation of the CT
states into free charge carriers. Recently the importance of the CT* states has
emerged.
87,88,91
The model describing photocurrent generation discussed above, implies
18
formation of a CT* state rather than CT
0
after the initial photoinduced electron transfer.
The excess of the thermal energy increases the electron-hole separation distance of the
CT state, and thereby decreases the binding energy.
88
In the solar cell, dissociation of the
CT* state must compete with thermalization to CT
0
. If a CT* state thermally relaxes to
CT
0
before charge separation, it is likely to undergo geminate recombination (as will be
described below) rather than charge dissociation. This can be explained by the lower-
lying energy level of the CT
0
state with respect to CS*, as seen in Figure 1.4. In this case
CT
0
dissociation is thermodynamically unfavorable and population of the CT
0
state is
viewed as evidence of a failed charge dissociation pathway. If the CT* does not undergo
thermalization to CT
0
, the smaller binding energy improves the dissociation yield to the
CS state.
84

Finally, the necessary energetic driving force for efficient charge carrier
dissociation can be established. As it was mentioned above, without an excess of the
thermal energy, the CT
0
state has little chance to dissociate and yield a CS state. This
means that there must be some excess in energy available beyond that which would be
minimally estimated from the relation ∆G
CS
= E
S1
- (IP
D
– EA
A
) in order to serve as a
driving force for the efficient dissociation of the CT state and indeed the entire charge
generation process. Ohkita et al.
92
were able to empirically correlate the free energy
change for charge separation and the charge generation yield. It is found that the
minimum free energy difference required for efficient charge dissociation (∆G
CS
eff
) is on
the order of 0.5 - 0.7 eV, which is significantly larger than the energy needed for exciton
19
separation at the donor/acceptor interface (~0.1 eV) and confirms that an excess of
thermal/vibrational energy of the CT* states increases the dissociation efficiency.

Based on this, a strong correlation between the yield of free charge carriers (CS
states) and the free energy of charge separation (∆G
CS
) was achieved. For example, it was
shown, for example that an increase of ∆G
CS
by 0.3 eV from 0.6 to 0.9 eV increased the
yield of free charge carriers by more than two orders of magnitude.
92
After this
publication, the correlation was tested on different polymer systems
91,93–96
and using
different acceptors.
87,97,98
The only systems that were found to deviate from the trend
described above are the ones processed with solvent additives.
84
The mismatch is
prescribed to the increased crystallinity, which caused a reduction of the ionization
potential of the donor and reduced Coulombic attraction in the CT state, hence,
improving charge photogeneration. The description of charge photogeneration described
above was also general for polymer/polymer blends, implying a correct description of the
processes at the donor/acceptor interface.
66,99–104

20

Figure 1.5. Energetic processes that do not lead to charge photogeneration. (i)
luminescence; (ii) phosphorescence; (iii) intersystem crossing on the donor; (iv)
thermalization of “hot” excitons; (v) intersystem crossing on the CT state; (vi) geminate
recombination on the donor’s triplet state; (vii) geminate recombination to the donor’s
singlet state; (viii) geminate recombination to the ground state; (ix) thermalization of the
CT state; (x) recombination from the CS* state to the CT state; (xi) bimolecular
recombination.

As such, a clear pathway to efficient charge photogeneration is seen in Figure 1.3
and 1.4. However, this pathway is based on a delicate balancing of energetic relationships
at each point in the process, and can break down at any point along the way. For example,
if an exciton is generated in the donor phase at a distance greater than the exciton
diffusion length from the interface, it is most likely that the exciton will simply relax
back to the ground state (S
0
). Figure 1.5 summarizes this and other unfavorable pathways
that can arise in the overall charge photogeneration process. Specifically, if the exciton
21
reaches the donor-acceptor interface and there is a suitable energetic driving force for the
initial electron transfer event to form the CT state, a number of circumstances can prevent
the CT state from effectively dissociating into free charge carriers that can be collected as
photocurrent. Once populated, a CT state can face a number of possible fates. First, the
CT* state can relax to the CT
0
state, which lacks the energy necessary to lead to a CS
state. The CT
0
state can then undergo geminate recombination to reform the S
0
state or
the CT
0
state can undergo spin mixing to form a triplet state that can undergo geminate
recombination to the T
1
state of the donor, which would ultimately relax to the ground
state, although the latter is not shown.
84
Either pathway would ultimately lead to a failed
photogeneration step and decrease the overall yield of generated charges.
Other factors that inhibit effective charge generation are related to the relationship
between the electronic structure of the donor and acceptor components and the relative
energy of the CT state. It is also important to consider that the energy of electronic states
are affected by the relative crystallinity of conjugated polymers. Ohkita et al. studied the
effect of polymer crystallinity on charge carrier yield.
92
In the case of amorphous
polymers, the triplet energy (T
1
a
) is usually about 1-1.5 eV lower than S
1
. If the T
1
a
is the
lowest lying energy level with respect to CT
0
, the formation of charge carriers will be a
competition between charge dissociation and geminate recombination to the triplet on the
donor component. When a more crystalline polymer was used in the blend, the triplet
energy (T
1
c
) was no longer the lowest lying energy level (∆E
ST
= 0.45 eV) with respect to
the CT
0
state. Thus the formation of the triplet on the donor molecule is
thermodynamically unfavorable and the yield of photogenerated charge carriers was
22
increased going from amorphous to crystalline polymer. This correlation between the
triplet energy and the crystallinity may not be a general one, but it does illustrate the
importance of all the energy states as well as the physical state of the polymer in the
overall effectiveness of the charge generation process. Similarly, the relationship among
the singlet and triplet energies of the acceptor and the CT
0
state are also important. For
reference, the singlet and triplet energies of C
60
-PCBM are 1.7 and 1.5 eV,
respectively.
105

Along these same lines, the degree of phase separation and the domain size in
donor-acceptor composites can also strongly affect the yield of photogenerated charges.
Both bimolecular recombination and geminate recombination are sensitive to
morphology. Bimolecular recombination refers to the recombination of completely
separated free charge carriers (CS states). Bimolecular recombination of free charges in
organic low mobility materials is described via Langevin theory.
106–108
In the case of
bimolecular recombination, free charges must diffuse to a separation distance shorter than
Coulomb radius (r
c
≈ 20 nm)
109
before recombination can happen. Therefore, the
bimolecular recombination is a slower process with respect to geminate recombination.
Geminate recombination involves recombination of bound charge carriers and follows
first-order kinetics, while bimolecular recombination of two completely separated free
charge carriers follows second-order kinetics.
84,110
As already described, the ideal
morphology in BHJ devices should be based on a bicontinuous network of donor and
acceptor phase with the scale of phase separation on the order of exciton diffusion length.
Having too large a degree of phase separation will lead to an increase of exciton
23
recombination back to the ground state, because in this case the majority of the excitons
will not be able to reach the interface. At the same time it is observed that domain sizes
that are too small will increase bimolecular and geminate recombination. This can be
explained with the increase of the density of CT states and CS states in both the donor
and acceptor phases.
93,111–115
Due to the fact that charge transport is a random hopping
process,
71
the probability of recombination increases for the small phase separation.
The last step in the photocurrent generation after free CS state formation is charge
transport and collection at the electrodes (Figure 1.3 (v) and (vi)). If the CS states in
Figure 1.3 are efficiently populated, free charge carriers must then migrate to the
electrodes (step (v) in Figure 1.3), with holes collected at the ITO anode and electrons at
the metal cathode (step (vi) in Figure 1.3). The efficiency of these two steps strongly
depends on the morphology of the blend. In the case of small phase separation, free
charge diffusion is hindered due to a lack of well percolating paths to the electrodes.
95,116

The role of charge carrier mobility in each phase on the overall charge photogeneration
process has been reviewed elsewhere.
111
However, the importance of balanced mobilities
must be considered.
117
An increase in the ordering within each phase of the blend also
improves charge transport (reducing bimolecular recombination and facilitating electron
and hole mobility) and enhances charge collection efficiency.
111

The above discussion has focused on the fundamental charge photogeneration
pathway and the factors that serve to increase or decrease the overall yield of free charge
carriers. It is clear that a detailed understanding of this process is finally emerging and is
critical to the design of the next generation of donors and acceptors for highly efficient
24
BHJ solar cells. The energetic processes and limitations discussed above also have great
consequence on the attainable V
oc
in BHJ solar cells.

1.2.2 Influence of Electronic Structure on the V
oc

The origin of the V
oc
in BHJ solar cells has long been a topic of debate. As
described in the introduction, the correlation between the energetic difference of the
donor HOMO and acceptor LUMO has long been a guiding principle for material’s
design. In most cases, the V
oc
can be reliably estimated as the difference between HOMO
level of the donor and LUMO level of the acceptor minus an empirical value of ~0.3
V,
51,118–120
however for some polymer-fullerene pairs this correlation fails.
121
The
preceding discussion of the centrality of the CT state in the photogeneration process also
makes it clear that the simple HOMO-LUMO picture of solar cell operation overlooks the
actual energetic pathway that is undertaken in the charge photogeneration process. Thus,
several new models pertaining to the origins of the V
oc
have appeared in recent years.
Several groups have developed correlations between the energy of the CT state
and the observed V
oc
. In a model developed empirically by Janssen et al.
54
using a series
of donor polymers, acceptor polymers, and C
60
-PCBM, it was found that the V
oc
could be
related to the E
CT
via the relationship eV
oc
= E
CT
– 0.5 eV (where e is the elementary
charge). In a similar approach, Vandewal et al.
122
empirically correlated the V
oc
with the
onset of photocurrent generation via direct CT absorption through the relation eV
oc
=
E
CT(onset)
– 0.43 eV, using a family of well-known polymers in combination with C
60
-
25
PCBM. In both cases, the empirical correlation provides a significantly higher level of
sophistication than a simple HOMO-LUMO model and these models were tested with
numerous other polymer/fullerene systems and excellent agreement with the correlations
was obtained in majority of cases.
93,96,123–126
The fact that the V
oc
is always less than the
E
CT
by a value of nearly 0.5 eV is attributed to recombination of the CT state, which
amounts to losses in the number of free charge carriers that can be formed and directly
relates to a lowering of the V
oc
.
90
This recombination can actually be directly correlated to
the dark current (J
S
). As shown below, a reduction in the J
S
can lead to a direct increase
in the magnitude of the V
oc
and can be controlled via a tuning of the coupling between the
donor and the acceptor.
P
S
() R
= { [exp( ) 1] } ( )
R
S
S Ph
PP
q V JR V
J J JV
R nkT R
−
− + −
+
(1)
()
= [exp( ) 1] ( )
S
S Ph
q V JR
JJ J V
nkT
−
− − (2)
ln( )
SC
OC
S
J nkT
V
q J
≈ (3)
exp( )
2
DA
S SO
E
JJ
nkT
−∆
= (4)
ln( )
2
SC DA
OC
SO
J E nkT
V
qJ q
∆
= + (5)
In an organic solar cell the generalized Shockley equation (equation 1) is used to
describe the relationship between current density (J) and voltage (V).
127,128
Here, R
P
is the
parallel resistance, q is the fundamental charge, n is the diode ideality factor, and J
Ph
(V)
is the voltage-dependent photocurrent density. The term J
S
is defined as the saturation
26
current density (dark current). For solar cells characterized by vanishingly small leakage
currents (when R
p
>> R
S
), the Shockley equation can be written in the form of equation
2, where the first term describes thermally generated current (primarily from the donor-
acceptor interface or more exactly via thermal excitation to the CT state of interacting
ground state donor and acceptor components) and the second term describes current
based on photogenerated carriers. Under open circuit conditions, equation 2 can be re-
written in the form of equation 3 (in the limit where R
S
is negligible and J
ph
(V) >> J
S
).
The J
S
term is defined as shown in equation 4 and substitution of equation 4 into
equation 3 yields the final expression in equation 5 relating the V
oc
and the pre-
exponential dark current term J
S0
(which is influenced by the reorganization energy
associated with the electron transfer, the intermolecular overlap at the donor-acceptor
interface, the electrical conductivities of the donor and acceptor phases, the area of the
donor-acceptor interface, and the density of states at the HOMO and LUMO energies of
the donor and acceptor materials). From this relationship, it can be concluded that the
smaller the value of J
S0
, the larger the V
oc
. A clear correlation between the V
oc
and the J
S0

is observed for both bilayer organic solar cells based on vapor deposited layers of small
molecules
127
and for conjugated polymers in fullerene-based BHJ solar cells.
128
Figure
1.6 illustrates the chemical structures of a homologous series of donor-acceptor based
conjugated polymers that were studied in BHJ solar cells with C
60
-PCBM
128
and Table
1.3 summarizes the solar cell results for 1:1 blends of polymer and fullerene. A clear
correlation can be seen in that the smaller the value of the J
S0
, the larger the V
oc
. Further,
27
there is excellent agreement between the calculated V
oc
values (using the theoretical
treatment described above) and the experimentally observed V
oc
values.
It is interesting to note the very subtle differences in the polymer structures in
Figure 1.6 led to V
oc
values ranging from 0.41 – 0.81 V. Only the nature of the alkyl side
chains were varied, leading to essentially no change in the band gap or HOMO energy of
the polymers and dispelling the notion that the V
oc
is directly correlated to the offset of
the acceptor LUMO and the donor HOMO. Note that the polymer with the highest
content of large, bulky, branched side chains (C10,6-C6,2) displayed the highest V
oc
and
as the branched chains were systematically converted to short unbranched chains, to the
extreme of the polymer C8-C8, the V
oc
systematically decreased to a minimum. This is
consistent with the notion that J
S0
is indicative of the strength of the intermolecular
interactions at the donor-acceptor interface.
127
The stronger the intermolecular interaction
of the donor and the acceptor, the larger the J
S0
and consequently the smaller the V
oc
. This
clearly correlates with the observation that the large bulkier side chains on the polymers
in Figure 1.6 inhibit close molecular interaction between the donor and acceptor, hence
minimizing the π molecular orbital overlap between them and resulting in a lower J
S0
and
a larger V
oc
.

28

Figure 1.6. The chemical structures of the six polymers based on the PNDT-DTBT
backbone. NDT = naphto[2,1-b:3,4-b’]dithiophene; DTBT = 4,7-di(thiophen-2-
yl)benzothiadiazole.
128

Table 1.3. Measured and calculated performance parameters for all devices
a
based on the
polymers illustrated in Figure 1.6
Polymer J
so
(mA/cm
2
) HOMO (eV) V
oc
(V), calcd V
oc
(V), exptl
C10,6-C8 148 5.32 0.60 0.59
C10,6-C6,2 3.38 5.33 0.83 0.81
C8-C8 399 5.13 0.39 0.41
C8-C12 254 5.27 0.53 0.52
C8-C6,2 68.8 5.30 0.60 0.59
C6,2-C6,2 22.6 5.34 0.70 0.69
a
Devices were obtained with polymer and [6,6]-phenyl C
60
-butyric acid methyl ester (PCBM) blend with
1:1 ratio.

At a deeper level of analysis, it is clear that the influence of the intermolecular
interaction, which correlates so well with the J
S0
is also strongly correlated to the energy
of the CT state. When the donor and the acceptor have a very close intermolecular
interaction, the CT state is characterized by a relatively short charge separation distance
between electron and hole. If the intermolecular interaction is considerably weaker, the
separation distance between electron and hole in the CT state will be significantly larger
and the energy (E
CT
) will consequently be lower. Vandewal et al.
129
more rigorously
29
developed this relationship between E
CT
, J
S
, and the V
oc
. In organic solar cells the J
S
is
dominated by recombination. A reduction in dark current is tantamount to a reduction in
the non-radiative recombination of the CT state and favors a large V
oc
. This issue of
recombination is still the subject of investigation to clarify the full physical significance
and lead to concrete design principles.
90

1.2.3 Defining the Electronics of the Optimal Donor-Acceptor Pair

As a consequence of the recent progress toward understanding of the charge
photogeneration pathway and the origin of the V
oc
, a revised picture of the optimal
electronic relationship between donor and acceptor can be established. In the following
discussion, the optimal electronic structure of a polymeric donor for use in combination
with C
60
-PCBM, will be proposed as an example of the logical process of defining an
optimal donor for a specific acceptor. This example is based on current state-of-the-art
understanding and based entirely on electronic considerations.
It is clear from the previous discussion of the charge photogeneration process that
a number of interacting energetic parameters must be considered with any donor-acceptor
pair. Ultimately it is desired that the product of J
sc
× V
oc
will reach a maximum. The J
sc
is
limited by the absorption envelope of the donor-acceptor pair, so it is desired that the
wavelength range of light absorption is as large as possible and consequently to minimize
the E
g
of the donor. It is also desired to maximize the V
oc
and the charge photogeneration
efficiency, which are both closely related to the energy of the CT state (E
CT
), and thus to
30
the HOMO and LUMO energies of the donor and the acceptor, respectively.
Additionally, the energy of the lowest lying triplet states of both the donor (T
1
D
) and the
acceptor (T
1
A
) are important as they must not lie at a significantly lower energy than the
CT state, or else risk the activation of a loss mechanism in the charge photogeneration
process.
Ultimately, there are two levels of energetic variables to consider in determining
the maximum attainable product of J
sc
× V
oc
. For simplicity, only forward electron
transfer from photoexcited donor to acceptor are considered. At the first level are the
electronic variables related to the individual donor and acceptor components, which
include the band gap (E
g
), or lowest singlet excited state (S
1
) of both components, the
HOMO and LUMO energies, or ionization potentials (IP) and electron affinities (EA), the
lowest triplet state energies (T
1
), and the exciton binding energy (E
B
exc
). At the second
level, are the energetic variables based on the interaction of the donor and acceptor,
which include the energy of the CT state (E
CT
), the binding energy of the CT state (E
B
CT
),
the minimal energetic driving force available for forward electron transfer from the donor
exciton to the CT state (∆G
CT
= E
S1
– E
CT
), and the energetic driving force for the overall
charge photogeneration process (∆G
CS
= E
S1
– (IP
D
– EA
A
)).
For the purposes of the following discussion, several assumptions are made. First,
the HOMO energy is considered to be the IP measured by electrochemistry and the
LUMO energy is considered to be the EA measured by electrochemistry. Also, based on
empirical observations the optical band gap of a conjugated polymer is observed to be
smaller than the electrochemical band gap (IP-EA) by a discrete value, that is taken to be
31
typically ~0.4 eV although this value depends on the specific chemical structure of the
donor. Often this difference is cited as corresponding to the E
B
exc
,
84
and while this value
is of a magnitude suitable to make this correlation, we will not attempt to assign the
precise physical significance of this empirical difference between the optical and
electrochemical band gaps.
As a starting point, the S
1
and T
1
of C
60
-PCBM are 1.7 and 1.5 eV respectively
(Figure 1.7). As a consequence of these values, the E
CT
of any donor-acceptor pair
involving C
60
-PCBM cannot exceed 1.7 eV, as a downhill driving force of > 0.2 eV
between the CT state and T
1
of the acceptor will favor a loss pathway via population of
T
1
in PCBM.
54
It necessarily follows that the upper limit for the V
oc
is 1.2 V, based on the
empirical relation (eV
oc
= E
CT
– 0.5 eV) that was previously discussed. Furthermore, the
smallest band gap of a donor polymer that can be used to give this maximal E
CT
and V
oc
is
1.8 eV, as determined by ∆G
CT
= E
S1
– E
CT
, which describes the minimum energy
required for population of the CT state from the donor exciton, which is thought to be
only about 0.1 eV. The ionization potential of the donor in this case is predicted based on
the relation E
CT
= (IP
D
– EA
A
) + E
B
CT
, considering that E
CT
is 1.7 eV, EA
A
is 4.2 eV, and
E
B
CT
is a typical value of 0.4 eV. As such, the HOMO energy of the donor (IP
D
) should
be 5.5 eV. Also, considering the empirical difference between the optical band gap and
the electrochemical band gap of ~0.4 eV, with a HOMO of 5.5 eV and an optical band
gap of 1.8 eV, the LUMO (as measured electrochemically) should be at ~3.3 eV. Figure
1.8 shows a sketch of the HOMO-LUMO diagram (electrochemical) for this donor-
PCBM pair. A donor of the specific electronic structure shown in Figure 1.7 and 1.8
32
would possess the minimum possible band gap for a donor capable of generating the
maximum possible V
oc
with C
60
-PCBM.

Figure 1.7. Energy levels for the excited states of the described donor polymer with a
balance between the maximum V
oc
and the minimal corresponding optical band gap
relative to C
60
-PCBM as the acceptor.

A further important point of importance is related to the charge photogeneration
efficiency. If a maximum J
sc
is to be realized, the photocurrent generation process must
be as effective as possible and thus a suitable energetic driving force is required via ∆G
CS

= E
S1
– (IP
D
– EA
A
), where ∆G
CS
eff

should be minimally 0.5 eV. This condition is
satisfied if IP
D
= 5.5 eV and E
S1
= 1.8 eV, for the case of the acceptor C
60
-PCBM. As
such, the donor electronic structure illustrated in Figure 1.7 and 1.8 should be capable of
efficiently generating a maximum allowable photocurrent at the maximum possible V
oc
for a PCBM-based donor-acceptor pair. The characteristics of such a solar cell are
summarized in entry 9 of Table 1.4. Here it is assumed that the V
oc
(V
oc
(max)) is based
on the relation eV
oc
= E
CT
– 0.5 eV and that a FF of 0.75 will be possible along with a J
sc

33
corresponding to an average EQE (external quantum efficiency) of 0.75 across the entire
range of absorption from 280 nm to the band edge of the donor (see Table 1.2).

Figure 1.8. HOMO-LUMO diagram of the described donor polymer relative to C
60
-
PCBM as the acceptor.

Donors with band gap energies of less than 1.8 eV can also be used very
effectively in C
60
-PCBM BHJs, but at the price of sacrificing the V
oc
in exchange for
enhanced J
sc
. When attempting to use polymers with band gaps narrower than 1.8 eV, a
shift in the HOMO and LUMO energies of the donor must be done with great care. In
order to ensure efficient charge photogeneration, a ∆G
CS
of minimally 0.5 eV must be
retained, requiring that the LUMO of the donor should not dip below the value of 3.3 eV
derived above. Consequently, the HOMO energy of low band gap polymers (E
g
< 1.5 eV)
should always be above 5.5 eV. Table 1.4 illustrates the properties of donor-acceptor
34
combinations where the LUMO of the donor is held constant at 3.3 eV, and band gap is
systematically decreased from 1.8 to 1.0 eV. By holding the LUMO of the donor constant
at 3.3 eV and thus keeping the ∆G
CS
at a minimum of 0.5 eV, effective charge
photogeneration and maximized V
oc
are ensured for each band gap. As such, the
maximum possible efficiency for each given donor band gap is illustrated in Table 1.4
for the case of a C
60
-PCBM donor.

Table 1.4. The properties of polymer-C
60
-PCBM blend combinations
a
Entry E
g
(D) (eV) HOMO(D) (eV) V
oc
(emp) (eV) V
oc
(max) (eV) J
sc
(mA/cm
2
) η (%)
1 1.0 4.7 0.1 0.4 32.6 2.5 - 9.8
2 1.1 4.8 0.2 0.4 29.2 4.4 – 11.0
3 1.2 4.9 0.3 0.6 26.4 5.9 - 11.9
4 1.3 5.0 0.4 0.7 24.0 7.2 - 12.6
5 1.4 5.1 0.5 0.8 21.7 8.2 – 13.0
6 1.5 5.2 0.6 0.9 19.1 8.6 - 12.9
7 1.6 5.3 0.7 1.0 16.7 8.8 - 12.6
8 1.7 5.4 0.8 1.1 14.6 8.8 – 12.0
9 1.8 5.5 0.9 1.2 12.7 8.6 - 11.4
a
These properties are calculated considering the donor LUMO is constant at 3.3 eV, FF = 75%, and the
average EQE = 75% across full range of light absorption. V
oc
(max) and V
oc
(emp) are described in the text
and η(%) is given as a range based on these two different V
oc
values.

It should be noted that the V
oc
(max)

values shown in Table 1.4 represent a value
that is related to the HOMO energy of the donor and LUMO energy of the acceptor via
the relation eV
oc
= E
CT
– 0.5 eV, which effectively translates to eV
oc
= E
HOMO(D)
– E
LUMO(A)

-0.1 eV. This can be taken as an upper limit estimate of the V
oc
, where empirically the V
oc
is more closely approximated by eV
oc
= E
HOMO(D)
– E
LUMO(A)
-0.4 eV in typical cases (e.g.
P3HT:PCBM). Lower limit estimates for the V
oc
are thus also given in Table 1.4 as V
oc

(emp), along with the efficiencies that would result. This uncertainty in the V
oc
can also
be related to the influence of J
S0
(molecular interaction at the donor acceptor:interface) on
35
the V
oc
, where two polymer with the same HOMO-LUMO energy can give significantly
different V
oc
values even with the same acceptor. It appears from this tabulated data that
the optimal donor for C
60
-PCBM would have an optical band gap in the range of 1.4 – 1.7
eV, a LUMO energy (EA) of 3.3 eV, and a HOMO energy (IP) in the range of 5.1 – 5.4
eV. Lowering the HOMO below 5.5 eV will increase the E
CT
to the point where
population of the PCBM T
1
can become a problematic loss mechanism. Lowering the
LUMO energy below 3.3 eV could result in decreased charge photogeneration efficiency
via a decreased energetic driving force for the charge separation process, while
significantly increasing the LUMO above 3.3 eV will result in a non-optimal balance
between J
sc
and V
oc
. Importantly, this suggests that the commonly used approximation
that a LUMO-LUMO offset of 0.3 eV is required for charge photogeneration is not
accurate and that for the most efficient charge photogeneration a LUMO-LUMO offset
approaching 0.9 eV appears to be more optimal. Notice that polymers shown in Figure
1.2 have HOMO energies in the range of 5.1 – 5.7 eV, optical band gaps in the range of
1.5 – 1.9 eV, and LUMO energies of 3.2 – 3.8 eV (not taking into account PBDTTT-
NAP as only the optical LUMO is known). Further optimization of the electronic
structures of high performing polymers such as those shown in Figure 1.2 could perhaps
move the observed efficiencies from the 5.0 – 7.7% range toward 10% or even higher.
Clearly though, there are a number of other factors that play a role in determining
the V
oc
and the J
sc
that are attainable with a given donor-acceptor pair, which include the
dark current (and the pre-exponential factor, J
S0
), the dielectric constant, the charge
carrier mobilities of the donor and acceptor, the absorption coefficient, the morphology as
36
it relates to the length scale of phase segregation and the molecular structure of the
donor-acceptor interface, and the device superstructure as it relates to such factors as light
absorption/reflection and electrode contact resistance. For BHJ solar cells the
morphology of the donor-acceptor blend is known to play a central role in determining
the device characteristics and thus will be considered below, in light of the fundamental
questions posed in this review concerning the optimal bulk morphology of the polymer-
fullerene BHJ and the role of the molecular structure of the donor-acceptor interface.

1.3 Physical Interactions in the Donor-Acceptor Composite: Defining the Optimal
Morphology

In the previous sections the operation and optimization of polymer-fullerene BHJ
solar cells was discussed strictly from the point of view of the electronic interactions
between the donor and acceptor. As mentioned in the introduction, the operation of a BHJ
solar cell is based on an intimately intertwined set of electronic and physical interactions
between the donor and acceptor. While these interactions cannot be viewed as truly
separate, in this section the state-of-the-art understanding of how physical interactions
influence device performance will be discussed. Importantly, physical interactions
between donor and acceptor at the level of the bulk, phase-separated structure will be
considered in addition to the role of molecular interactions at the donor-acceptor
interface.
37
During the past 15 years, a picture of the “optimal” morphology of a BHJ solar
cell has emerged, based on the necessity to develop a maximum interfacial area, a defined
minimum domain size, and continuous pathways for charge transport. This picture is
largely driven by the short exciton diffusion length of organic polymers, which is
typically thought to be less than 10 nm. In essence, an ideal BHJ is a bicontinuous
composite with domain sizes on the order of ~20 nm, such that excitons generated in
either phase will have a significant probability of diffusing to the donor-acceptor
interface where the CT state can be populated. This domain size restriction of 20 nm
imposes a delicate balance between exciton harvesting and bimolecular recombination.
As discussed in the previous section, the Coulomb radius of free charge carriers in
organics is on the order of ~20 nm and thus if two oppositely charged species approach
within less than 20 nm of one another, there is a distinct probability that their attraction
will lead to bimolecular charge recombination. Therefore, an average domain size that
significantly exceeds 20 nm will lead to a loss mechanism due to exciton relaxation to the
ground state, while a domain size any smaller than 20 nm can significantly increase the
rate of bimolecular recombination. There are a number of other considerations related to
bimolecular recombination that will be dependent on the specific properties of the donor
and acceptor and the random walk of the exciton to the donor/acceptor interface.
95,112,130

An optimal bicontinuous composite will also possess sufficient interfacial contact of the
donor with the anode (ITO) and acceptor with the cathode (typically aluminum) for
efficient hole and electron collection, respectively.
38
It is quite simple to define an optimal morphology based on the pertinent
electronic considerations of exciton harvesting, charge generation, charge transport, and
charge collection; however, it is significantly more complicated to actually realize such a
structure in a real polymer-fullerene composite. A great deal of effort has been devoted to
the optimization of morphology over the past nine years since the first understanding of
the relationship between processing conditions and bulk morphology was developed.
37

Since then, numerous processing techniques have been explored and these have been
thoroughly reviewed.
50,131–133
Primary among the techniques utilized have been thermal
annealing,
39
solvent annealing,
46
and the use of solvent additives.
41,134
All of these
techniques operate on a similar concept. After a film is formed via spin-coating, a
controlled reorganization of the polymer and fullerene components into separate domains
containing some level of intramolecular order suitable for efficient and balanced charge
transport is pursued through the application of heat, slow evaporation of solvent, or
selective solubilization of one component in a solvent mixture with different evaporation
rates. When the desired level of morphology evolution is achieved, the process is halted
in order to “trap” the desired morphology. This morphology does not represent an
equilibrium structure, but rather a kinetically trapped state that can be subject to further
evolution given the right circumstances.
Only a few polymer-fullerene combinations have undergone extensive
optimization of processing conditions, with the combination of P3HT and C
60
-PCBM
representing the most well studied pair.
135
With this combination the influence of
annealing temperature and time on the nanostructure of the composite is well known.
39
Further, solvent annealing (the so-called “slow-growth” method
136,137
has also been
optimized to give efficient solar cells with well-characterized morphologies. While
diverse characterization techniques have allowed for a deep understanding of the effect of
these processing methods on the bulk structure of the P3HT-PCBM pair, as well as other
well-known pairs,
41,134
what is missing are the quantitative design principles that can
relate molecular structure to a specific tendency for a donor-acceptor pair to self-organize
into an optimal morphology under a specific set of processing conditions.
With P3HT, the influence of molecular weight
138
and regioregularity
139
have been
examined for their influence on the development of a suitable morphology under a
variety of processing conditions. Recently, significant efforts have begun to quantify the
relationship between intrinsic properties of the donor and the acceptor as a means of
rigorously defining the physical basis for the optimal donor-acceptor ratio in the
composite film and the processing conditions. Several groups have attempted to draw
conclusions based on the construction of a phase diagram for P3HT-PCBM blends.
140,141
This approach promises to lead to predictive power of the optimal composition and
processing conditions for a given donor-acceptor pair given a complete mapping of the
thermal properties of the blend across the full range of donor-acceptor compositions. In
another approach, the strength of the repulsive interactions between polymer and
fullerene components has been quantified through the calculation of the Flory-Huggins
interaction parameter for the two components cast from a specific solvent, by
measurement of the surface energies via a contact angle technique.
142
These types of
quantitative approaches have not lead to a general set of design principles for a donor-
40
acceptor pair with designed optimal morphology under a pre-selected processing method,
however, such efforts are indicative of the level of detailed study that will be required to
develop the deep understanding necessary for the realization of quantitative design
principles that will shift morphology optimization from a trial-and-error or combinatorial
process
143
to one of execution of rational design.
It is clear that the bulk morphology is critical for solar cell operation, but a
growing body of theoretical and experimental data strongly suggests that the molecular
level structure at the donor-acceptor interface is perhaps even more profoundly
responsible for the operational characteristics of organic solar cells. This point was
highlighted in the previous section when the correlation of the pre-exponential dark-
current term (J
S0
) with the V
oc
was discussed. In both polymer-fullerene BHJ solar cells
and small molecule bilayer devices this effect has been observed and correlated to the
degree of interaction of the donor and acceptor components at the interface. The more
significant the interaction between the donor and the acceptor, the greater the extent of
orbital overlap between the donor and the acceptor. As discussed, this not only influences
J
S0
and V
oc
,
127,128
but also influences other fundamental parameters such as E
CT
and E
B
CT
,
which are profoundly involved in the charge photogeneration process. A detailed picture
of what constitutes the optimal molecular level packing at the donor-acceptor interface is
currently not known. The design principle suggested by the works discussed, is that the
V
oc
can be maximized by inhibiting tight molecular packing and significant orbital
overlap at the donor-acceptor interface through the incorporation of bulky substituents.
However, it should be noted that this also runs counter to the most efficient molecular
41
arrangement for photoinduced charge transfer, so some optimal compromise must be
found.
At an even deeper level of analysis, the molecular orientation of the donor and
acceptor with respect to one another at the interface can also have strong effects on the
charge photogeneration process. Theoretical studies
144,145
have indicated that the relative
geometric orientation of the donor and acceptor at the interface will not only affect the
degree of electronic coupling between the two, but will also influence the local interface
dipole, which plays a critical role in determining the precise energy of molecular
electronic states. As such, the specific geometry at interface will have a strong influence
on the rates of charge transfer and recombination as well the achievable photovoltage as
it is strongly related to the energetics of the interface. Much of the work done in this area
of molecular geometry and orientation at the interface has been theoretical and has
mostly focused on small molecule donor-acceptor pairs. When considering the significant
complication of this picture describing the molecular interface when moving to BHJs of
high molecular weight polymer donors and fullerene acceptors, it is a truly daunting task
to fully understand the structure of this molecular interface and the role that this structure
(and associated structural disorder) will play in determining the characteristics of a BHJ
solar cell. Recent theoretical studies of the P3HT/C
60
interface picture do not describe in
details the influence of interfacial geometry on the device characteristics.
146
However, it
will ultimately be important to develop this level of understanding and translate it into
defined design principles for a next generation of highly optimized and tailored
polymeric donor and fullerene acceptor combinations.
42
One characterization method that allows for the direct observation of the
molecular structure at the interface involved a combination of 1-D and 2-D solid-state
NMR methods, which were used to examine the close contacts between PCBM and
P3HT at the interface.
147
In this direct study, it was observed that when films were cast at
room temperature, the hexyl groups of the P3HT chains are associated with the fullerene
cage of PCBM. However upon thermal annealing these associations are no longer
observed and the fullerene cage is in closer spatial proximity to the thiophene rings. The
work clarified the structural interaction PCBM and P3HT in this paper is helpful in
understanding the effect of thermal annealing on morphology at the interface and set an
important precedent in using a well-known experimental method to probe the molecular
structure at the interface. Certainly more studies employing this and additional
characterization methods will be required in order to ultimately understand the interface
at the molecular level and to apply this information to generate rational design principles.
Figure 1.9 illustrates a pictorial summary of the different levels of structure that
influence device performance. From a bottom-up perspective, the nature of the influence
of the molecular interface is poorly understood, while the role of the bulk morphology is
relatively better understood. Ultimately the superstructure of the device is significant to
device performance as well, and will be discussed in the final section along with the
importance of device stability.

43

Figure 1.9. Levels of BHJ solar cell structural organization. (i). Overall organic solar
cell; (ii). Layers of the BHJ device; (iii). BHJ structure of the active layer; (iv). Donor-
acceptor patterns in the active layer; (v). Unknown structure at the donor-acceptor
interface.

44
1.4 Architectures and Stability in Bulk Heterojunction Solar Cells

The focus of this chapter is to present the state-of-the-art understanding of
polymer-fullerene BHJ solar cells with an emphasis on the electronic and physical
interactions between the donor and the acceptor components as the central element of
device performance. In a broader sense, the device superstructure, or overall architecture,
also plays a significant role in the device performance and thus is addressed. Of equal
importance is the stability of these BHJ solar cells, since no matter how high the
efficiency or how inexpensive the processing method, if the solar cells are not stable
enough to provide a useful lifetime, practical application will remain elusive.

1.4.1 Device Architecture

The standard device architecture for a polymer-fullerene BHJ solar cell is based
on a multilayer structure (Figure 1.10) on a glass substrate of ITO/PEDOT:PSS/active
layer (polymer-fullerene blend)/Al and has been extensively described and
optimized.
35,82,148
In the case of the standard cell, PEDOT-PSS serves as a hole-transport
layer adjacent to the hole collecting ITO electrode and the lower work-function
aluminum serves as the electron collecting electrode. As it stands, the standard device
structure has been able to provide efficiencies close to 8%.
38
Modifications to this
standard device architecture are often been employed, where the most well-known is the
insertion of a titanium oxide layer between the active layer and the aluminum
45
cathode.
43,149
The incorporation of this so-called “optical-spacer” is used to improve
device efficiency by increasing the fraction of photons absorbed within the active layer.

Figure 1.10. Device architectures of a standard

cell (left)
35
with a P3HT:PCBM active
layer, an inverted

cell (center)
132
with electrodes in reverse order and additional transition
metal oxide layer, and a tandem

cell (right)
150
with P3HT:C
70
BM and PCPDTBT:PCBM
active layers.

A so-called inverted device configuration, also shown in Figure 1.10, employs a
reversal of charge collection as a remedy for perceived shortcomings of the standard
device such as the acidic nature of PEDOT:PSS, the oxidation of standard devices and
the vertical phase separation of the active layer, which favors the polymer to be the
dominant component at the upper interface of the active layer near the electron-collecting
electrode.
132
By coating the substrate surface with a lower work function layer, like ZnO,
TiO
2
, V
2
O
5
, CsCO
3
or MoO
3
, the ITO is utilized as the cathode, and a high work function
metal, Au or Ag, as the anode.
132,151
Thus far, an inverted device structure using a
46
carboxylic acid-based C
60
-SAM (self-assembled monolayer) formed on the surface of
ZnO and P3HT/PCBM blend has attained a PCE of 4.4%.
152

A tandem cell architecture has also been employed using both the standard and
inverted architectures.
150,153,154
A multi-layer or stacked tandem cell is shown in Figure
1.10. Tandem cells are designed to cover a broad absorption range by stacking two or
more distinct active layers with donors having complementary band gaps. Tandem cells
based on polymer-fullerene BHJ cells were reported to approach PCE’s of 6.5%.
148,150

Folded reflective tandem devices having a V-shaped geometry (as opposed to the stacked
geometry that is shown in Figure 1.10) have the potential advantage of reflecting
unabsorbed light from one cell to the other, thus creating a larger area for light absorption
and a light-trapping effect.
155

In the three architectures described above, the device processing involves at least
one vacuum deposition step for the top metal electrode. As such, the devices are not fully
solution processed, and while this is acceptable for laboratory scale optimization, it is
clear that the only truly low-cost processing of polymer-based solar cells will be based on
a process free of vacuum deposition steps.
131
This implies a continuous roll-to-roll
procedure in which all the layers of the solar cell are processed from solution. Most
notably, Krebs et al.
156
developed an entirely solution processed method for roll-to-roll
printing on flexible substrates. In this case, using an inverted device geometry, the bottom
electrode (cathode) is comprised of a printed silver nanoparticle dispersion on top of
which a solution processed ZnO-nanoparticle electron transport layer is deposited. To
complete the device, the active layer of P3HT-PCBM was coated from solution, followed
47
by a hole transporting PEDOT-PSS layer and finally a screen-printed silver paste ink was
applied in a grid pattern as the top electrode (anode). This example not only offers the
advantage of a completely solution processed fabrication method, but also offers an ITO-
free solar cell, which is highly relevant as the ITO electrodes are thought to be the most
expensive component in polymer-based solar cells.
22
A few other examples of fully
solution-processed polymer solar cells are also known and represent an important shift
toward more practical device processing techniques and architectures.
22,50,157,158

1.4.2 Long-Term Stability

For practical application, polymer-based solar cells must exhibit sufficient long-
term stability. Several studies have focused on the long-term stability under simulated or
actual ambient conditions. A well-known study by Hauch et al. focused on the long-term
outdoor operation of P3HT-PCBM BHJ solar cells.
159
In this case the devices were
encapsulated with a transparent barrier film and it was found that the solar cells were
stable over a period of 14 months, and in fact, a slight increase in efficiency was
observed. Several other accelerated studies on outdoor stability have been published,
160–
164
although a relatively small amount of information has been collected on long-term
stability.
Before too much effort is spent on studying the long-term stability of polymer
solar cells, it is first important to understand the factors that could lead to device
degradation. This will allow the development of design principles for the next generation
48
of device architectures as well the next generation of polymer and fullerene components.
The factors that have been observed to cause degradation of device performance can be
lumped into three broad categories. The first category consists of environmental factors
that can be excluded via appropriate device encapsulation strategies. These include:
oxygen and/or water,
165,166
ultraviolet illumination,
167,168
temperature, humidity, and
atmosphere.
161

The second category consists of interactions between the component layers in the
solar cell. As mentioned above, a major reason for using the inverted solar cell
architecture is based on the acidity of the PEDOT-PSS layer.
132
Another major source of
degradation is linked to the metal/organic interface. Studies by Reese et al. were
performed to elucidate the nature of a possible organic-inorganic interfacial instability in
BHJ solar cells.
169
Results with P3HT-PCBM solar cells under a variety of conditions
reveal that the interaction between the metal and organic layer can be a significant source
of degradation through both photoactivated and non-photoactivated degradation
pathways. However degradation appears to be dependent on the choice of the metal
electrode and can be mitigated through device encapsulation. It was also observed that
the morphology at the metal organic interface can play a role in device degradation
characteristics as well.
170

A third category of degradation effects is related to the active layer itself,
specifically the morphological stability of the BHJ. As described in previous section, the
morphology of BHJ solar cells is primarily achieved via a kinetic trapping of a desirable
morphology and thus does not represent a thermodynamically stable structure. As a
49
consequence, the morphology of BHJ solar cells is often unstable to temperature
fluctuations. This was demonstrated with the P3HT-PCBM solar cell
139,171
where
operation of the device at elevated temperatures resulted in a significant decrease in
performance that was directly related to a coarsening of the phase separation between
polymer and fullerene. Based on this level of instability a number of approaches have
been designed to improve the morphological stability at the molecular level of the active
layer components. Several promising approaches are highlighted below, although it
should be noted that none have resulted in equivalent or increased efficiencies relative to
the simple polymer-fullerene BHJ solar cell.
In a simple approach for the P3HT-PCBM system, it was found that a reduction in
the regioregularity of the P3HT
139,171
resulted in a decrease in the tendency of the P3HT
to undergo large-scale crystallization, which is a primary driving force for polymer-
fullerene phase separation. Several other methods of specifically designed chemical
structures have been employed in order to stabilize the BHJ. Figure 1.11 illustrates
several chemical structures that are representative of the approaches that will be
discussed below.

50

Figure 1.11. Structures of polymer and fullerene components designed to improve the
stability of the morphology in the BHJ active layer.

One approach to stabilizing the morphology of the BHJ is chemical crosslinking
to lock-in the desired morphology.
172
Several research groups
173–175
studied this process
by using PCBM derivatives that have crosslinkable groups. As an example, structure 1 in
Figure 1.11 is a phenyl C
60
-butyric acid glycidol ester (PCBG). The blend formed with 1
and P3HT was studied by means of AFM, TEM, and PL to observe an improved
nanomorphology.
173
Zhu et al. similarly reported PCBG with success, and also
synthesized cross-linkable polythiophenes with furoyl and epoxy side chains, however,
due to solubility problems no films were made nor devices fabricated.
174
Likewise, efforts
have been made with a crosslinkable P3HT analogue P3HNT (structure 2 in Figure
1.11), with hexenyl chains.
176
Devices with this crosslinkable polymer and C
60
-PCBM
51
achieved performance comparable to that of the benchmark P3HT/PCBM and with
extended thermal treatment. In this case, it was found that PCBM aggregates were
eliminated due to the cross-linking abilities of P3HNT.
176

Another approach to stabilizing the morphology of polymer-fullerene BHJ solar
cells is through the use of thermocleavable (or thermally removable) solubilizing groups.
These groups allow one to chemically alter the structure of the active layer after
processing from solution. The solubilizing side chains in a conjugated polymer are
removed at high temperatures (> 200 °C), therefore increasing the density of the
chromophores, inhibiting the diffusion of small molecules (e.g fullerenes) and ultimately
achieving more rigid systems that are proven to be more stable in devices.
172,177
Examples
of polymers of this class that have been used in fullerene-based BHJ solar cells are 3 and
4 in Figure 1.11.
178,179
In both cases, the thermocleavable group is a tertiary ester, which
thermally decomposes to release a volatile alkene and yield a carboxylic acid
functionality on the polymer backbone.
A final strategy that has been explored for the stabilization of the BHJ
morphology has been via the use of block copolymers. Specifically, diblock copolymers
have been used in two different approaches. In the first approach the diblock copolymers
were used as phase compatibilizers.
180–182
In a representative example, a block copolymer
based on a P3HT block and a fullerene-functionalized block was used as an additive to
the P3HT-PCBM BHJ. It was found that the morphology and performance of the BHJ
with the additive was significantly more stable to long-term heating than the simple
P3HT-PCBM BHJ.
182
The second approach to using block copolymers is the generation
52
of the active layer based entirely on the controlled phase segregation of a donor-acceptor
block copolymer. This approach was designed to take advantage of the fact that diblock
copolymers are known to phase segregate into bicontinuous structures with a nanometer
length scale of phase separation in many cases.
183–185
Further, the well-defined
bicontinuous morphologies that are observed occur at thermodynamic minima and
therefore promise a thermally stable morphology.

The structures 5 and 6 in Figure 1.11
are examples of donor-acceptor block copolymers that were developed toward this
end.
186,187

It is clear that while significant work has been done to understand and improve
the stability of BHJ solar cells, much work is still needed. This will certainly involve
optimization of device structure at all levels illustrated in Figure 1.9, as well as the
thorough optimization of the chemical structure of the polymer and fullerene components
in the active layer.

1.5 Outlook

This chapter has served to summarize the state-of-the-art in the current
understanding of the factors that influence the operation of polymer-fullerene BHJ solar
cells. The primary focus has been on the electronic and physical interactions between the
donor and the acceptor. Developing a deep understanding of theses interactions will serve
as the cornerstone in the development of design principles for the next generation of
polymer and fullerene components, which promise to deliver 10% efficient solar cells.
53
Great strides have been made in the last few years toward this level of fundamental
understanding. Coupling this with the significant advances in device efficiency that have
been achieved during this same time period and the developing understanding of device
stability, polymer-fullerene BHJ solar cells are now more promising than ever as
candidates for practical application.

1.6 Chapter Bibliography

(1) Shaheen, S. E.; Ginley, D. S.; Jabbour, G. E. MRS Bull. 2005, 30, 10.

(2) Ginley, D.; Green, M. A.; Collins, R. MRS Bull. 2008, 33, 355.

(3) Aberle, A. G. Thin Solid Films 2009, 517, 4706.

(4) Green, M. A. J. Mater. Sci.: Mater. Electron. 2007, 18, 15.

(5) Solarmer Energy, Inc., http://www.solarmer.com/rd.php (Accessed October 12,
2010).

(6) Günes, S.; Neugebauer, H.; Sariciftci, N. S. Chem. Rev. 2007, 107, 1324.

(7) Walker, B.; Tamayo, A. B.; Dang, X.-D.; Zalar, P.; Seo, J. H.; Garcia, A.;
Tantiwiwat, M.; Nguyen, T.-Q. Adv. Funct. Mater. 2009, 19, 3063.

(8) Coakley, K. M.; McGehee, M. D. Chem. Mater. 2004, 16, 4533.

(9) Yang, X.; Loos, J.; Veenstra, S. C.; Verhees, W. J. H.; Wienk, M. M.; Kroon, J.
M.; Michels, M. A. J.; Janssen, R. A. J. Nano Lett. 2005, 5, 579.

(10) Van Bavel, S.; Sourty, E.; De With, G.; Frolic, K.; Loos, J. Macromolecules 2009,
42, 7396.

(11) Blom, P. W. M.; Mihailetchi, V. D.; Koster, L. J. A.; Markov, D. E. Adv. Mater.
2007, 19, 1551.

(12) O’Regan, B.; Gra ̈tzel, M. Nature 1991, 353, 737.
54
(13) Gra ̈tzel, M. Nature 2001, 414, 338.

(14) Chen, C.-Y.; Wang, M.; Li, J.-Y.; Pootrakulchote, N.; Alibabaei, L.; Ngoc-le, C.;
Decoppet, J.-D.; Tsai, J.-H.; Gra ̈tzel, C.; Wu, C.-G.; Zakeeruddin, S. M.; Gra ̈tzel,
M. ACS Nano 2009, 3, 3103.

(15) Tang, C. W. Appl. Phys. Lett. 1986, 48, 183.

(16) Xue, J.; Uchida, S.; Rand, B. P.; Forrest, S. R. Appl. Phys. Lett. 2004, 84, 3013.

(17) Xue, J.; Rand, B. P.; Uchida, S.; Forrest, S. R. J. Appl. Phys. 2005, 98, 124903.

(18) Holcombe, T. W.; Woo, C. H.; Kavulak, D. F. J.; Thompson, B. C.; Fréchet, J. M.
J. J. Am. Chem. Soc. 2009, 131, 14160.

(19) Yu, G.; Heeger, A. J. J. Appl. Phys. 1995, 78, 4510.

(20) Halls, J. J. M.; Walsh, C. A.; Greenham, N. C.; Marseglia, E. A.; Friend, R. H.;
Moratti, S. C.; Holmes, A. B. Nature 1995, 376, 498.

(21) Krebs, F. C.; Jørgensen, M.; Norrman, K.; Hagemann, O.; Alstrup, J.; Nielsen, T.
D.; Fyenbo, J.; Larsen, K.; Kristensen, J. Sol. Energy Mater. Sol. Cells 2009, 93,
422.

(22) Krebs, F. C.; Tromholt, T.; Jørgensen, M. Nanoscale 2010, 2, 873.

(23) Schilinsky, P.; Waldauf, C.; Brabec, C. J. Adv. Funct. Mater. 2006, 16, 1669.

(24) Hoth, C. N.; Schilinsky, P.; Choulis, S. A.; Brabec, C. J. Nano Lett. 2008, 8, 2806.

(25) Krebs, F. C. Sol. Energy Mater. Sol. Cells 2009, 93, 484.

(26) Krebs, F. C. Sol. Energy Mater. Sol. Cells 2009, 93, 465.

(27) Hoth, C. N.; Steim, R.; Schilinsky, P.; Choulis, S. A.; Tedde, S. F.; Hayden, O.;
Brabec, C. J. Org. Electron. 2009, 10, 587.

(28) Jung, J. W.; Jo, W. H. Adv. Funct. Mater. 2010, 20, 2355.

(29) Kim, S.-S.; Na, S.-I.; Kang, S.-J.; Kim, D.-Y. Sol. Energy Mater. Sol. Cells 2010,
94, 171.

(30) McNeill, C. R.; Greenham, N. C. Adv. Mater. 2009, 21, 3840.
55
(31) McNeill, C. R.; Abrusci, A.; Zaumseil, J.; Wilson, R.; McKiernan, M. J.;
Burroughes, J. H.; Halls, J. J. M.; Greenham, N. C.; Friend, R. H. Appl. Phys. Lett.
2007, 90, 193506.

(32) Kietzke, T.; Hörhold, H.-H.; Neher, D. Chem. Mater. 2005, 17, 6532.

(33) Sun, B.; Greenham, N. C. Phys. Chem. Chem. Phys. 2006, 8, 3557.

(34) Dayal, S.; Kopidakis, N.; Olson, D. C.; Ginley, D. S.; Rumbles, G. Nano Lett.
2010, 10, 239.

(35) Thompson, B. C.; Fréchet, J. M. J. Angew. Chem., Int. Ed. 2008, 47, 58.

(36) Woo, C. H.; Holcombe, T. W.; Unruh, D. A.; Sellinger, A.; Fréchet, J. M. J. Chem.
Mater. 2010, 22, 1673.

(37) Shaheen, S. E.; Brabec, C. J.; Sariciftci, N. S.; Padinger, F.; Fromherz, T.;
Hummelen, J. C. Appl. Phys. Lett. 2001, 78, 841.

(38) Chen, H.-Y.; Hou, J.; Zhang, S.; Liang, Y.; Yang, G.; Yang, Y.; Yu, L.; Wu, Y.;
Li, G. Nat. Photonics 2009, 3, 649.

(39) Ma, W.; Yang, C.; Gong, X.; Lee, K.; Heeger, A. J. Adv. Funct. Mater. 2005, 15,
1617.

(40) Zhao, G.; He, Y.; Li, Y. Adv. Mater. 2010, 22, 4355.

(41) Peet, J.; Kim, J. Y.; Coates, N. E.; Ma, W. L.; Moses, D.; Heeger, A. J.; Bazan, G.
C. Nat. Mater. 2007, 6, 497.

(42) Wang, E.; Hou, L.; Wang, Z.; Hellstroem, S.; Zhang, F.; Inganaes, O.; Andersson,
M. R. Adv. Mater. 2010, 22, 5240.

(43) Park, S. H.; Roy, A.; Beaupré, S.; Cho, S.; Coates, N.; Moon, J. S.; Moses, D.;
Leclerc, M.; Lee, K.; Heeger, A. J. Nat. Photonics 2009, 3, 297.

(44) Piliego, C.; Holcombe, T. W.; Douglas, J. D.; Woo, C. H.; Beaujuge, P. M.;
Fréchet, J. M. J. J. Am. Chem. Soc. 2010, 132, 7595.

(45) Bundgaard, E.; Krebs, F. Sol. Energy Mater. Sol. Cells 2007, 91, 954.

(46) Li, G.; Shrotriya, V.; Huang, J.; Yao, Y.; Moriarty, T.; Emery, K.; Yang, Y. Nat.
Mater. 2005, 4, 864.
56
(47) Kroon, R.; Lenes, M.; Hummelen, J.; Blom, P.; De Boer, B. Polym. Rev. 2008, 48,
531.

(48) Lenes, M.; Wetzelaer, G.-J. A. H.; Kooistra, F. B.; Veenstra, S. C.; Hummelen, J.
C.; Blom, P. W. M. Adv. Mater. 2008, 20, 2116.

(49) Jarzab, D.; Cordella, F.; Lenes, M.; Kooistra, F. B.; Blom, P. W. M.; Hummelen, J.
C.; Loi, M. A. J. Phys. Chem. B 2009, 113, 16513.

(50) Peet, J.; Senatore, M. L.; Heeger, A. J.; Bazan, G. C. Adv. Mater. 2009, 21, 1521.

(51) Scharber, M. C.; Mühlbacher, D.; Koppe, M.; Denk, P.; Waldauf, C.; Heeger, A.
J.; Brabec, C. J. Adv. Mater. 2006, 18, 789.

(52) Kirchartz, T.; Taretto, K.; Rau, U. J. Phys. Chem. C 2009, 113, 17958.

(53) Servaites, J. D.; Ratner, M. A.; Marks, T. J. Appl. Phys. Lett. 2009, 95, 163302.

(54) Veldman, D.; Meskers, S. C. J.; Janssen, R. A. J. Adv. Funct. Mater. 2009, 19,
1939.

(55) Koster, L. J. A.; Mihailetchi, V. D.; Blom, P. W. M. Appl. Phys. Lett. 2006, 88,
093511.

(56) Scholes, G. D.; Rumbles, G. Nat. Mater. 2006, 5, 683.

(57) Bakulin, A. A.; Hummelen, J. C.; Pshenichnikov, M. S.; Van Loosdrecht, P. H. M.
Adv. Funct. Mater. 2010, 20, 1653.

(58) Cook, S.; Katoh, R.; Furube, A. J. Phys. Chem. C 2009, 113, 2547.

(59) Scholes, G. D. ACS Nano 2008, 2, 523.

(60) Krause, S.; Casu, M. B.; Schöll, A.; Umbach, E. New J. Phys. 2008, 10, 085001.

(61) Knupfer, M. Appl. Phys. A: Mater. Sci. Process. 2003, 77, 623.

(62) Zhao, X.; Piliego, C.; Kim, B.; Poulsen, D. A.; Ma, B.; Unruh, D. A.; Fréchet, J.
M. J. Chem. Mater. 2010, 22, 2325.

(63) Park, M.-J.; Lee, J.; Kwak, J.; Jung, I. H.; Park, J.-H.; Kong, H.; Lee, C.; Hwang,
D.-H.; Shim, H.-K. Macromolecules 2009, 42, 5551.

57
(64) Wong, W.-Y.; Wang, X.-Z.; He, Z.; Djurišić, A. B.; Yip, C.-T.; Cheung, K.-Y.;
Wang, H.; Mak, C. S. K.; Chan, W.-K. Nat. Mater. 2007, 6, 521.

(65) Cook, S.; Liyuan, H.; Furube, A.; Katoh, R. J. Phys. Chem. C 2010, 114, 10962.

(66) Benson-Smith, J. J.; Wilson, J.; Dyer-Smith, C.; Mouri, K.; Yamaguchi, S.;
Murata, H.; Nelson, J. J. Phys. Chem. B 2009, 113, 7794.

(67) Meng, K.; Ding, Q.; Wang, S.; He, Y.; Li, Y.; Gong, Q. J. Phys. Chem. B 2010,
114, 2602.

(68) Suzuki, S.; Sugimura, R.; Kozaki, M.; Keyaki, K.; Nozaki, K.; Ikeda, N.;
Akiyama, K.; Okada, K. J. Am. Chem. Soc. 2009, 131, 10374.

(69) Ogawa, K.; Guo, F.; Schanze, K. S. J. Photochem. Photobiol., A 2009, 207, 79.

(70) Piris, J.; Dykstra, T. E.; Bakulin, A. A.; Loosdrecht, P. H. M. van; Knulst, W.;
Trinh, M. T.; Schins, J. M.; Siebbeles, L. D. A. J. Phys. Chem. C 2009, 113,
14500.

(71) Emelianova, E. V.; Van der Auweraer, M.; Ba ̈ssler, H. J. Chem. Phys. 2008, 128,
224709.

(72) Athanasopoulos, S.; Emelianova, E. V.; Walker, A. B.; Beljonne, D. Phys. Rev. B
2009, 80, 195209.

(73) Pivrikas, A.; Sariciftci, N. S.; Juška, G.; Österbacka, R. Prog. Photovolt: Res.
Appl. 2007, 15, 677.

(74) Brédas, J.-L.; Norton, J. E.; Cornil, J.; Coropceanu, V. Acc. Chem. Res. 2009, 42,
1691.

(75) Markov, D. E.; Amsterdam, E.; Blom, P. W. M.; Sieval, A. B.; Hummelen, J. C. J.
Phys. Chem. A 2005, 109, 5266.

(76) Lewis, A.; Ruseckas, A.; Gaudin, O.; Webster, G.; Burn, P.; Samuel, I. Org.
Electron. 2006, 7, 452.

(77) Scully, S. R.; McGehee, M. D. J. Appl. Phys. 2006, 100, 034907.

(78) Westenhoff, S.; Howard, I.; Friend, R. Phys. Rev. Lett. 2008, 101, 016102.

(79) Gommans, H.; Schols, S.; Kadashchuk, A.; Heremans, P.; Meskers, S. C. J. J.
Phys. Chem. C 2009, 113, 2974.
58
(80) Lebental, M.; Choukri, H.; Chénais, S.; Forget, S.; Siove, A.; Geffroy, B.; Tutiš, E.
Phys. Rev. B 2009, 79, 165318.

(81) Lunt, R. R.; Giebink, N. C.; Belak, A. A.; Benziger, J. B.; Forrest, S. R. J. Appl.
Phys. 2009, 105, 053711.

(82) Heremans, P.; Cheyns, D.; Rand, B. P. Acc. Chem. Res. 2009, 42, 1740.

(83) Cook, S.; Furube, A.; Katoh, R.; Han, L. Chem. Phys. Lett. 2009, 478, 33.

(84) Clarke, T. M.; Durrant, J. R. Chem. Rev. 2010, 110, 6736.

(85) Guo, J.; Ohkita, H.; Benten, H.; Ito, S. J. Am. Chem. Soc. 2009, 131, 16869.

(86) Zhu, X.-Y.; Yang, Q.; Muntwiler, M. Acc. Chem. Res. 2009, 42, 1779.

(87) Shoaee, S.; Clarke, T. M.; Huang, C.; Barlow, S.; Marder, S. R.; Heeney, M.;
McCulloch, I.; Durrant, J. R. J. Am. Chem. Soc. 2010, 132, 12919.

(88) Pensack, R. D.; Asbury, J. B. J. Phys. Chem. Lett. 2010, 1, 2255.

(89) Hallermann, M.; Haneder, S.; Da Como, E. Appl. Phys. Lett. 2008, 93, 053307.

(90) Deibel, C.; Strobel, T.; Dyakonov, V. Adv. Mater. 2010, 22, 4097.

(91) Clarke, T. M.; Ballantyne, A. M.; Tierney, S.; Heeney, M.; Duffy, W.; McCulloch,
I.; Nelson, J.; Durrant, J. R. J. Phys. Chem. C 2010, 114, 8068.

(92) Ohkita, H.; Cook, S.; Astuti, Y.; Duffy, W.; Tierney, S.; Zhang, W.; Heeney, M.;
McCulloch, I.; Nelson, J.; Bradley, D. D. C.; Durrant, J. R. J. Am. Chem. Soc.
2008, 130, 3030.

(93) Pal, S. K.; Kesti, T.; Maiti, M.; Zhang, F.; Ingana ̈s, O.; Hellström, S.; Andersson,
M. R.; Oswald, F.; Langa, F.; O ̈sterman, T.; Pascher, T.; Yartsev, A.; Sundström,
V. J. Am. Chem. Soc. 2010, 132, 12440.

(94) Benson-Smith, J. J.; Ohkita, H.; Cook, S.; Durrant, J. R.; Bradley, D. D. C.;
Nelson, J. Dalton Trans. 2009, 10000.

(95) Clarke, T. M.; Jamieson, F. C.; Durrant, J. R. J. Phys. Chem. C 2009, 113, 20934.

(96) Zhou, Y.; Tvingstedt, K.; Zhang, F.; Du, C.; Ni, W.-X.; Andersson, M. R.;
Inganäs, O. Adv. Funct. Mater. 2009, 19, 3293.
59
(97) Shoaee, S.; An, Z.; Zhang, X.; Barlow, S.; Marder, S. R.; Duffy, W.; Heeney, M.;
McCulloch, I.; Durrant, J. R. Chem. Commun. 2009, 5445.

(98) Howard, I. A.; Laquai, F.; Keivanidis, P. E.; Friend, R. H.; Greenham, N. C. J.
Phys. Chem. C 2009, 113, 21225.

(99) Ford, T. A.; Ohkita, H.; Cook, S.; Durrant, J. R.; Greenham, N. C. Chem. Phys.
Lett. 2008, 454, 237.

(100) Huang, Y.; Westenhoff, S.; Avilov, I.; Sreearunothai, P.; Hodgkiss, J. M.;
Deleener, C.; Friend, R. H.; Beljonne, D. Nat. Mater. 2008, 7, 483.

(101) McNeill, C. R.; Westenhoff, S.; Groves, C.; Friend, R. H.; Greenham, N. C. J.
Phys. Chem. C 2007, 111, 19153.

(102) Dyer-Smith, C.; Benson-Smith, J. J.; Bradley, D. D. C.; Murata, H.; Mitchell, W.
J.; Shaheen, S. E.; Haque, S. A.; Nelson, J. J. Phys. Chem. C 2009, 113, 14533.

(103) Westenhoff, S.; Howard, I. A.; Hodgkiss, J. M.; Kirov, K. R.; Bronstein, H. A.;
Williams, C. K.; Greenham, N. C.; Friend, R. H. J. Am. Chem. Soc. 2008, 130,
13653.

(104) Panda, P.; Veldman, D.; Sweelssen, J.; Bastiaansen, J. J. A. M.; Langeveld-Voss,
B. M. W.; Meskers, S. C. J. J. Phys. Chem. B 2007, 111, 5076.

(105) Cook, S.; Ohkita, H.; Kim, Y.; Benson-Smith, J. J.; Bradley, D. D. C.; Durrant, J.
R. Chem. Phys. Lett. 2007, 445, 276.

(106) Koster, L. J. A.; Mihailetchi, V. D.; Blom, P. W. M. Appl. Phys. Lett. 2006, 88,
052104.

(107) Jus ̌ka, G.; Genevic ̌ius, K.; Nekras ̌as, N.; Sliauz ̌ys, G.; O ̈sterbacka, R. Appl. Phys.
Lett. 2009, 95, 013303.

(108) Juska, G.; Arlauskas, K.; Stuchlik, J.; Osterbacka, R. J. Non-Cryst. Solids 2006,
352, 1167.

(109) Pivrikas, A.; Juška, G.; Mozer, A.; Scharber, M.; Arlauskas, K.; Sariciftci, N.;
Stubb, H.; Österbacka, R. Phys. Rev. Lett. 2005, 94, 176806.

(110) Deibel, C.; Dyakonov, V. Rep. Prog. Phys. 2010, 73, 096401.

(111) Guo, J.; Ohkita, H.; Benten, H.; Ito, S. J. Am. Chem. Soc. 2010, 132, 6154.

60
(112) Shoaee, S.; Eng, M. P.; Espíldora, E.; Delgado, J. L.; Campo, B.; Martín, N.;
Vanderzande, D.; Durrant, J. R. Energy Environ. Sci. 2010, 3, 971.

(113) Clarke, T. M.; Ballantyne, A. M.; Nelson, J.; Bradley, D. D. C.; Durrant, J. R. Adv.
Funct. Mater. 2008, 18, 4029.

(114) Hallermann, M.; Kriegel, I.; Da Como, E.; Berger, J. M.; Von Hauff, E.;
Feldmann, J. Adv. Funct. Mater. 2009, 19, 3662.

(115) Ballantyne, A. M.; Ferenczi, T. A. M.; Campoy-Quiles, M.; Clarke, T. M.;
Maurano, A.; Wong, K. H.; Zhang, W.; Stingelin-Stutzmann, N.; Kim, J.-S.;
Bradley, D. D. C.; Durrant, J. R.; McCulloch, I.; Heeney, M.; Nelson, J.; Tierney,
S.; Duffy, W.; Mueller, C.; Smith, P. Macromolecules 2010, 43, 1169.

(116) Street, R.; Cowan, S.; Heeger, A. Phys. Rev. B 2010, 82, 121301.

(117) Mihailetchi, V.; Wildeman, J.; Blom, P. Phys. Rev. Lett. 2005, 94, 126602.

(118) Maurano, A.; Hamilton, R.; Shuttle, C. G.; Ballantyne, A. M.; Nelson, J.;
O’Regan, B.; Zhang, W.; McCulloch, I.; Azimi, H.; Morana, M.; Brabec, C. J.;
Durrant, J. R. Adv. Mater. 2010, 22, 4987.

(119) Cravino, A. Appl. Phys. Lett. 2007, 91, 243502.

(120) Gadisa, A.; Svensson, M.; Andersson, M. R.; Ingana ̈s, O. Appl. Phys. Lett. 2004,
84, 1609.

(121) Yamanari, T.; Taima, T.; Sakai, J.; Saito, K. Sol. Energy Mater. Sol. Cells 2009,
93, 759.

(122) Vandewal, K.; Gadisa, A.; Oosterbaan, W. D.; Bertho, S.; Banishoeib, F.; Van
Severen, I.; Lutsen, L.; Cleij, T. J.; Vanderzande, D.; Manca, J. V. Adv. Funct.
Mater. 2008, 18, 2064.

(123) Nuzzo, D. D.; Aguirre, A.; Shahid, M.; Gevaerts, V. S.; Meskers, S. C. J.; Janssen,
R. A. J. Adv. Mater. 2010, 22, 4321.

(124) Vandewal, K.; Tvingstedt, K.; Gadisa, A.; Inganäs, O.; Manca, J. V. Phys. Rev. B
2010, 81, 125204.

(125) Tvingstedt, K.; Vandewal, K.; Gadisa, A.; Zhang, F.; Manca, J.; Ingana ̈s, O. J. Am.
Chem. Soc. 2009, 131, 11819.

61
(126) Hallermann, M.; Da Como, E.; Feldmann, J.; Izquierdo, M.; Filippone, S.; Martín,
N.; Ju ̈chter, S.; Von Hauff, E. Appl. Phys. Lett. 2010, 97, 023301.

(127) Perez, M. D.; Borek, C.; Forrest, S. R.; Thompson, M. E. J. Am. Chem. Soc. 2009,
131, 9281.

(128) Yang, L.; Zhou, H.; You, W. J. Phys. Chem. C 2010, 114, 16793.

(129) Vandewal, K.; Tvingstedt, K.; Gadisa, A.; Inganäs, O.; Manca, J. V. Nat. Chem.
2009, 8, 904.

(130) Hamilton, R.; Shuttle, C. G.; O’Regan, B.; Hammant, T. C.; Nelson, J.; Durrant, J.
R. J. Phys. Chem. Lett. 2010, 1, 1432.

(131) Krebs, F. C. Sol. Energy Mater. Sol. Cells 2009, 93, 394.

(132) Chen, L.-M.; Hong, Z.; Li, G.; Yang, Y. Adv. Mater. 2009, 21, 1434.

(133) Moulé, A. J.; Meerholz, K. Adv. Funct. Mater. 2009, 19, 3028.

(134) Lee, J. K.; Ma, W. L.; Brabec, C. J.; Yuen, J.; Moon, J. S.; Kim, J. Y.; Lee, K.;
Bazan, G. C.; Heeger, A. J. J. Am. Chem. Soc. 2008, 130, 3619.

(135) Li, G.; Shrotriya, V.; Yao, Y.; Huang, J.; Yang, Y. J. Mater. Chem. 2007, 17,
3126.

(136) Shrotriya, V.; Yao, Y.; Li, G.; Yang, Y. Appl. Phys. Lett. 2006, 89, 063505.

(137) Chu, C.-W.; Yang, H.; Hou, W.-J.; Huang, J.; Li, G.; Yang, Y. Appl. Phys. Lett.
2008, 92, 103306.

(138) Schilinsky, P.; Asawapirom, U.; Scherf, U.; Biele, M.; Brabec, C. J. Chem. Mater.
2005, 17, 2175.

(139) Woo, C. H.; Thompson, B. C.; Kim, B. J.; Toney, M. F.; Fréchet, J. M. J. J. Am.
Chem. Soc. 2008, 130, 16324.

(140) Zhao, J.; Swinnen, A.; Van Assche, G.; Manca, J.; Vanderzande, D.; Mele, B. V. J.
Phys. Chem. B 2009, 113, 1587.

(141) Müller, C.; Ferenczi, T. A. M.; Campoy-Quiles, M.; Frost, J. M.; Bradley, D. D.
C.; Smith, P.; Stingelin-Stutzmann, N.; Nelson, J. Adv. Mater. 2008, 20, 3510.

62
(142) Nilsson, S.; Bernasik, A.; Budkowski, A.; Moons, E. Macromolecules 2007, 40,
8291.

(143) Alstrup, J.; Jørgensen, M.; Medford, A. J.; Krebs, F. C. ACS Appl. Mater.
Interfaces 2010, 2, 2819.

(144) Yi, Y.; Coropceanu, V.; Brédas, J.-L. J. Am. Chem. Soc. 2009, 131, 15777.

(145) Lennartz, C.; Idé, J.; Méreau, R.; Aurel, P.; Ducasse, L.; Castet, F.; Linares, M.;
Beljonne, D.; Cornil, J.; Lancaster, K.; Brédas, J.-L.; Verlaak, S.; Mityashin, A.;
Heremans, P.; Fuchs, A. J. Phys. Chem. C 2010, 114, 3215.

(146) Kanai, Y.; Grossman, J. C. Nano Lett. 2007, 7, 1967.

(147) Yang, C.; Hu, J. G.; Heeger, A. J. J. Am. Chem. Soc. 2006, 128, 12007.

(148) Brabec, C. J.; Gowrisanker, S.; Halls, J. J. M.; Laird, D.; Jia, S.; Williams, S. P.
Adv. Mater. 2010, 22, 3839.

(149) Kim, J. Y.; Kim, S. H.; Lee, H.-H.; Lee, K.; Ma, W.; Gong, X.; Heeger, A. J. Adv.
Mater. 2006, 18, 572.

(150) Kim, J. Y.; Lee, K.; Coates, N. E.; Moses, D.; Nguyen, T.-Q.; Dante, M.; Heeger,
A. J. Science 2007, 317, 222.

(151) Li, C.-Y.; Wen, T.-C.; Lee, T.-H.; Guo, T.-F.; Huang, J.-C.-A.; Lin, Y.-C.; Hsu,
Y.-J. J. Mater. Chem. 2009, 19, 1643.

(152) Hau, S. K.; Cheng, Y.-J.; Yip, H.-L.; Zhang, Y.; Ma, H.; Jen, A. K.-Y. ACS Appl.
Mater. Interfaces 2010, 2, 1892.

(153) Tvingstedt, K.; Andersson, V.; Zhang, F.; Ingana ̈s, O. Appl. Phys. Lett. 2007, 91,
123514.

(154) Hadipour, A.; De Boer, B.; Wildeman, J.; Kooistra, F. B.; Hummelen, J. C.;
Turbiez, M. G. R.; Wienk, M. M.; Janssen, R. A. J.; Blom, P. W. M. Adv. Funct.
Mater. 2006, 16, 1897.

(155) Inganäs, O.; Zhang, F.; Tvingstedt, K.; Andersson, L. M.; Hellström, S.;
Andersson, M. R. Adv. Mater. 2010, 22, E100.

(156) Krebs, F. C. Org. Electron. 2009, 10, 761.

(157) Krebs, F. C.; Norrman, K. ACS Appl. Mater. Interfaces 2010, 2, 877.
63
(158) Zhou, Y.; Li, F.; Barrau, S.; Tian, W.; Inganäs, O.; Zhang, F. Sol. Energy Mater.
Sol. Cells 2009, 93, 497.

(159) Hauch, J. A.; Schilinsky, P.; Choulis, S. A.; Childers, R.; Biele, M.; Brabec, C. J.
Sol. Energy Mater. Sol. Cells 2008, 92, 727.

(160) Medford, A. J.; Lilliedal, M. R.; Jørgensen, M.; Aarø, D.; Pakalski, H.; Fyenbo, J.;
Krebs, F. C. Optics Express 2010, 18, A272.

(161) Gevorgyan, S. A.; Jørgensen, M.; Krebs, F. C. Sol. Energy Mater. Sol. Cells 2008,
92, 736.

(162) Krebs, F. C.; Gevorgyan, S. A.; Alstrup, J. J. Mater. Chem. 2009, 19, 5442.

(163) Girtan, M.; Rusu, M. Sol. Energy Mater. Sol. Cells 2010, 94, 446.

(164) Krebs, F. C. Sol. Energy Mater. Sol. Cells 2008, 92, 715.

(165) Jørgensen, M.; Norrman, K.; Krebs, F. C. Sol. Energy Mater. Sol. Cells 2008, 92,
686.

(166) Norrman, K.; Gevorgyan, S. A.; Krebs, F. C. ACS Appl. Mater. Interfaces 2009, 1,
102.

(167) Tromholt, T.; Katz, E. A.; Hirsch, B.; Vossier, A.; Krebs, F. C. Appl. Phys. Lett.
2010, 96, 073501.

(168) Reese, M. O.; Nardes, A. M.; Rupert, B. L.; Larsen, R. E.; Olson, D. C.; Lloyd, M.
T.; Shaheen, S. E.; Ginley, D. S.; Rumbles, G.; Kopidakis, N. Adv. Funct. Mater.
2010, 20, 3476.

(169) Reese, M. O.; Morfa, A. J.; White, M. S.; Kopidakis, N.; Shaheen, S. E.; Rumbles,
G.; Ginley, D. S. Sol. Energy Mater. Sol. Cells 2008, 92, 746.

(170) Paci, B.; Generosi, A.; Rossi Albertini, V.; Perfetti, P.; De Bettignies, R.; Sentein,
C. Sol. Energy Mater. Sol. Cells 2008, 92, 799.

(171) Sivula, K.; Luscombe, C. K.; Thompson, B. C.; Fréchet, J. M. J. J. Am. Chem. Soc.
2006, 128, 13988.

(172) Helgesen, M.; Søndergaard, R.; Krebs, F. C. J. Mater. Chem. 2010, 20, 36.

64
(173) Drees, M.; Hoppe, H.; Winder, C.; Neugebauer, H.; Sariciftci, N. S.; Schwinger,
W.; Schäffler, F.; Topf, C.; Scharber, M. C.; Zhu, Z.; Gaudiana, R. J. Mater.
Chem. 2005, 15, 5158.

(174) Zhu, Z.; Hadjikyriacou, S.; Waller, D.; Gaudiana, R. J. Macromol. Sci., Part A:
Pure Appl. Chem. 2004, 41, 1467.

(175) Kim, B. J.; Miyamoto, Y.; Ma, B.; Fréchet, J. M. J. Adv. Funct. Mater. 2009, 19,
2273.

(176) Miyanishi, S.; Tajima, K.; Hashimoto, K. Macromolecules 2009, 42, 1610.

(177) Krebs, F. C.; Spanggaard, H. Chem. Mater. 2005, 17, 5235.

(178) Helgesen, M.; Gevorgyan, S. A.; Krebs, F. C.; Janssen, R. A. J. Chem. Mater.
2009, 21, 4669.

(179) Helgesen, M.; Krebs, F. C. Macromolecules 2010, 43, 1253.

(180) Ball, Z. T.; Sivula, K.; Fréchet, J. M. J. Macromolecules 2006, 39, 70.

(181) Rajaram, S.; Armstrong, P. B.; Kim, B. J.; Fréchet, J. M. J. Chem. Mater. 2009, 21,
1775.

(182) Sivula, K.; Ball, Z. T.; Watanabe, N.; Fréchet, J. M. J. Adv. Mater. 2006, 18, 206.

(183) King, S.; Sommer, M.; Huettner, S.; Thelakkat, M.; Haque, S. A. J. Mater. Chem.
2009, 19, 5436.

(184) Zhang, Q.; Cirpan, A.; Russell, T. P.; Emrick, T. Macromolecules 2009, 42, 1079.

(185) Segalman, R. A.; McCulloch, B.; Kirmayer, S.; Urban, J. J. Macromolecules 2009,
42, 9205.

(186) Sommer, M.; Lindner, S. M.; Thelakkat, M. Adv. Funct. Mater. 2007, 17, 1493.

(187) Sommer, M.; Lang, A. S.; Thelakkat, M. Angew. Chem., Int. Ed. 2008, 47, 7901.

65
Chapter 2

Tuning the Open-Circuit Voltage in Ternary Blend Organic Solar Cells as a Path to
Higher Efficiency

2.1 Introduction

Obtaining cheap, renewable and ecologically friendly source of energy is a long
and important quest for a modern society.
1,2
Renewable energy sources, such as wind,
geothermal, hydropower and biomass cannot single-handedly satisfy the modern society
growing demands.
3–6
On the other hand, every hour on the Earth’s surface falls enough
energy
7,8
to meet world energy needs for an entire year.
9
As such, solar energy becomes
the only solution as an easily attainable renewable energy source.
5,6
Unfortunately, solar
energy is responsible only for around 1% of the total energy supply worldwide right
now
4,5
but has all necessary components to compete with the non-renewable resources of
energy (petroleum, gas, coal, nuclear), which currently dominate the worldwide energy
market.
5

In order to harvest and convert solar energy to electricity and heat, solar cells are
used. All solar cells can be roughly divided into few groups, depending on the materials
and mechanisms responsible for device operation: inorganic solar cells, dye-sensitized
(Grätzel) solar cells (DSSC), hybrid solar cells, and organic solar cells. Current
photovoltaic (PV) market is controlled by inorganic solar cells with crystalline silicon
66
being the most commonly used solar irradiance absorbing material.
10–12
As a result of
almost 60 years of intense research, crystalline silicon solar cells have reached
efficiencies of 25%.
13
New inorganic thin film technologies
10–12
using cadmium telluride
(CdTe), copper indium gallium selenide (CIGS) and gallium arsenide (GaAs) are current
best competitors to crystalline silicon solar cells with efficiencies exceeding 25% for the
best PV devices.
13
But despite all efforts, multi- and mono-crystalline silicon represents
more than 85% of the PV market.
10,11,14
The main limitation of inorganic solar cells and
silicon in particular is the production cost. This is determined by the necessity of
applying high temperatures, sometimes exceeding 1000 °C, in order to enhance grain
grow and recrystalization of the active layer.
7
Furthermore, crystalline silicon is an
indirect bandgap material with a low optical absorption coefficient (α ~ 100 cm
-1
)
throughout the absorption breadth (bandgap (E
g
) = 1.1 eV)
10
thus requiring several
hundred micron thick films to absorb light efficiently which in turn increases the cost of
the silicon solar cell.
5,7,11,12,14,15
The low absorption coefficients do not appear to be a
problem in the majority of the inorganic thin film technologies with α ~ 10
5
cm
-1
at the
peak maximum,
10,11
but low abundance of the inorganic materials and device
manufacturing process which still requires high temperatures
10,11
significantly increases
the cost of the inorganic thin film solar cells.
10–12,14,16

Dye-sensitized solar cells
17–20
have recently showed impressive 12.3% power
conversion efficiency (PCE).
21
But despite all achievements, DSSC have a list of
disadvantages which currently prevent them from the widespread commercial
application. First of all, the use of temperature sensitive liquid electrolyte narrows the
67
temperature range for solar cell application.
18
Also, as in case of inorganic solar cells,
high cost of the active layer materials, predominantly ruthenium and platinum dyes,
significantly increases the cost of the DSSC.
17,18
Solid-state DSSC can help overcome
some of these limitations,
17
but the use of solid-state DSSC eliminates an attractive
feature of thin film solar cells to be flexible and current solid-state DSSC show low
efficiencies
22,23
and suffer from rapid degradation.
17
Finally, the practical efficiency limit
for DSSC is predicted to be 13.4%
17,24
which combined with the high cost of device
fabrication makes this type of solar cells economically less attractive.
Hybrid organic-inorganic nanoparticle solar cells
25–27
are intended to combine the
attractive features of both inorganic and organic materials, such as strong absorption of
organic compound,
28,29
inorganic component shape and dimension tunability with an aim
to tune short-circuit current density (J
sc
) and open-circuit voltage (V
oc
)
26
and the ability to
solution process the active layer. However, hybrid solar cells experience serious
problems which predominantly come from the inorganic counterpart and include
unfavorable morphology with large phase separation in bulk heterojunction (BHJ) solar
cells, necessity for an active layer to consist more than 80% of nanoparticles for efficient
charge transport, and presence of insulating ligands necessary to provide solubility of
nanoparticles.
26,30
As a result, efficiencies of the hybrid solar cells are still in the 3 – 5%
range.
27
Considering high cost and low abundance of the inorganic material,
26
hybrid
solar cells require significant efficiency increase to be considered as a competitor to
inorganic solar cells.
Among all of the PV types of solar cells, organic vapour deposited and especially
68
solution processable BHJ solar cells
31–39
provide the most attractive route to efficient,
cheap and ecologically friendly way for solar energy conversion. Organic materials have
important properties necessary for mass-production of cheap thin film solar cells such as
high absorption coefficients which lead to the thin films of only 200 – 300 nm necessary
to absorb almost 100% of incident light,
28,40,41
flexibility in tuning of the chemical
composition to effectively alter J
sc
and V
oc
,
42,43
presence of solubilizing groups for
solution processability.
42–44
As a result, in the last 10 years organic solar cells show the
steepest among all PV technologies and constant increase in the PCE,
6,32
now
approaching
45–47
and exceeding 10%.
6,13,48,49
This is attributed to the better understanding
of structure-property relation between donor material structure and device
performance
36,42–44,50,51
which lead to large number polymers/small molecules
synthesized and applied in the efficient organic solar cells.
42–44,52–55
Even though there are
still some not well understood mechanisms in photocurrent generation,
56–62
stability
63–66

and scale-up,
67,68
intense research of the device physics
28,56–60,69,70
and corresponding BHJ
active layer morphology
71–74
sheds light on the current obstacles of the organic solar cells
and the ways to overcome them. From the economical point of view, organic solar cells
have advantages over other PV technologies. The ability to obtain thin films from
solution allows fabrication of cheap, flexible, lightweight and large area solar cells at
high manufacturing speed using roll-to-roll coating and printing methods without the
necessity of applying high temperatures.
68,75,76
Furthermore, organic solar cell printing is
easily integrated in the existing PV manufacturing process. Therefore, the combination of
cheap organic solar cell fabrication, constantly increasing efficiency and promising
69
stability
77
make organic photovoltaics (OPV) the strongest potential competitor to
inorganic solar cells with the expected cost of kW*h similar and even less than the one of
the inorganic PV.
31,60,78–81

Figure 2.1. Simplified energy band diagram of organic donor:acceptor solar cell.

The main reason for small commercial interest to organic solar cells is current low
power conversion efficiencies with respect to inorganic solar cells.
13
The predicted
practical efficiency limit for single layer binary blend BHJ organic solar cell is found to
be in the 10 – 12% range,
31,56–59,82–88
even though higher efficiencies are potentially
achievable.
89–92
The efficiency (η) and the J
sc
× V
oc
product in particular (η = (J
sc
× V
oc
×
FF)/P
in
, where FF is fill factor and P
in
is input power in the form of solar radiation) are
limited by the interplay between J
sc
and V
oc
in the organic solar cells, which are opposing
quantities, as shown in Figure 2.1. The decrease of the donor material band gap leads to
the increase of the absorption breadth and consequently J
sc
(Figure 2.2) but is
70
accompanied by unavoidable decrease of the V
oc
because of the correlation between the
V
oc
and the difference between the donor HOMO energy level (HOMO
D
) and the LUMO
level of the acceptor (LUMO
A
).
86,93
Furthermore, the donor LUMO energy level
(LUMO
D
) cannot be lowered to the LUMO
A
energy level since driving force is necessary
to separate excitons at the donor/acceptor interface.
28,56,94

Figure 2.2. Global total photon flux from the sun and integrated short-circuit current
density (J
sc
).

The efficiency can be pushed to 14 – 15%
8,82–85,95,96
by using tandem organic solar
cells with two absorbing layers connected in series or parallel.
6,8,96–100
But tandem
organic solar cells loose one of the main attractive features of OPV which is the
simplicity of the device fabrication with an active layer being processed in a single step.
Together with the necessity of intermediate recombination layer,
101
the final cost of the
tandem organic solar cells should increase with respect to single layer binary blend solar
cells.
71
A few other approaches offer the possibility to increase the efficiency beyond the
one of the binary blend organic solar cells. Among them are singlet fission
102–104
and
increase of the dielectric constant of the active layer.
89,92,105
Despite the intriguing
possibilities with the efficiency increase, no solar cells showing actual results are yet
exist.
In current review, we introduce the novel approach of the efficiency increase
beyond the one of the binary blend solar cells while preserving the simplicity of an active
layer fabrication in a single processing step – ternary blend BHJ solar cells. The key
element in the power conversion efficiency increase is found to be not only the
photoresponse broadening and hence J
sc
increase, but simultaneous enhancement of both
parameters J
sc
and V
oc
, due to the V
oc
tunability in the three component systems. The
concept of ternary blend BHJ solar cells will be introduced, followed by the development
timeline of the ternary blend BHJ approach and discussion of the recent achievements of
the ternary blend BHJ solar cell approach. The models used for the explanation of the J
sc

and the V
oc
behaviour will be discussed in great details.

2.2 Ternary Blend Bulk Heterojunction Solar Cells

Ternary blend BHJ organic solar cells recently attracted significant attention as
one of the potential candidates to increase the efficiency beyond the one of the binary
blend BHJ solar cells while maintaining the simplicity of the solar cell design and
processability.
106–109
The device architecture of the ternary blend BHJ solar cell is the
72
same as in case of binary blend solar cell
56,110,111
with the only difference in the active
layer, as can be seen in Figure 2.3. Ternary blend BHJ solar cell active layer consists of
two donors and one acceptor
107–109,112,113
or one donor and two acceptors.
106,112,114–117

Even though, the morphology of the ternary blends is a big question so far,
118
high FF
106–
109,119,120
and balanced charge carrier mobilities
107,116,121,122
obtained in the most efficient
ternary blend BHJ solar cells indicate the presence of the interpenetrating network of
donor and acceptor components with a scale of phase separation on the order of exciton
diffusion length, schematically represented in Figure 2.3. Using three component
systems, efficiencies higher than that of corresponding binary blends were recorded for
numerous ternary blend combinations,
107–109,119,120
thus proving the potential for overall
efficiency increase beyond 10 – 12% limit set for binary blend BHJ solar cells. The
power conversion efficiency increase is possible due to simultaneous increase of the J
sc

and, more importantly intermediate V
oc
. Three major models describing the efficiency
increase, as well as J
sc
and V
oc
behaviour in the ternary blend BHJ solar cells are used:
organic alloy model,
123
parallel-like model
108
and the sensitization/cascade model.
124
The
concept, the device physics and the efficiency increase models of ternary blend BHJ solar
cells are presented below.

73

Figure 2.3. Device architecture and morphology of ternary blend bulk heterojunction
solar cells in case of two donors (top) and two acceptors (bottom).

2.2.1 Basis Concept – Increase of Light Absorption

The initial interest in the ternary blend BHJ solar cells was the increase of the
light harvesting possibilities by the active layer. Since the J
sc
of the solar cell is
proportional to the product of the absorption breadth and the absorption intensity as
shown in Figure 2.2, the use of the additional absorbing material with the absorption
envelope complimentary to the first absorber can be used to better cover the solar
spectrum. The concept is shown in Figure 2.4 and 2.5. In the binary blend with high band
gap donor D1 and fullerene acceptor A, absorption and, as a consequence, J
sc
are limited
by the small absorption breadth of the binary blend (A+D1). This translates to the poor
covering of the solar spectrum, as shown in Figure 2.5. Introduction of the low band gap
donor D2, as a third absorber in the blend, which on its own has broad absorption in the
74
binary blend (A+D2) but lacks of absorption in the visible, allows increasing the
absorption breadth of the ternary system (A+D1+D2) while maintaining strong
absorption throughout the visible and near-IR (NIR) parts of the solar spectrum and
enhances the number of photons that can be absorbed by the three component system,
leading to J
sc
increase.

Figure 2.4. Absorption profile of an acceptor (A), high bad gap donor (D1), low band
gap donor (D2) as well as binary (A+D1), (A+D2) and ternary (A+D1+D2) blends.

75

Figure 2.5. Solar spectrum covering in binary and ternary blends.

The first attempt to sensitize the polymer:fullerene blend was approached by
Brabec et al.
125
, where two polymers or one polymer and absorbing dye were mixed with
phenyl-C
61
-butyric acid methyl ester (PC
61
BM). Even though authors haven’t seen the
increase of the device parameters of ternary blend solar cells, they were able to detect
spectral response from all three components in the solar cells. As such, it was proven that
charge and/or energy transfer are possible in the three component systems. Later
photoluminescence and transient absorption studies of the photocurrent generation in
different ternary blends showed the dynamics and pathways of charge/energy transfer in
numerous ternary blend BHJ solar cells.
109,113,124,126–134
Since then, a large number of
ternary blend systems, both polymer:polymer
107–109,119,120,124,135,136
and polymer:small
molecule/dye based,
137–143
targeting increase of the solar spectrum coverage and J
sc
were
published. In some cases, J
sc
increase by more than 30% was recorded with respect to
corresponding binary blend BHJ solar cells upon adding low band gap polymer/small
molecule.
107–109,119,137,139,140,144
Figure 2.6 shows the chemical structures and external
76
quantum efficiency (EQE) values of some binary and ternary blend BHJ solar cells. It is
important to mention, that the EQE of the ternary blend strongly depends on the relative
ratio between two donors. As can be seen in Figure 2.6, addition of small amount of the
low band gap polymer/dye (10%) leads to a noticeable increase of the photocurrent in the
NIR. Further increase of the low band gap donor amount in the ternary blend system
leads to the strengthening of the photoresponse in the NIR. But the increase of the EQE
in the NIR sometimes is accompanied with the decrease of the photoresponse in the
visible (Figure 2.6b and d), especially if film thicknesses are kept constant, since the
amount of high band gap donor is decreased in the overall donor:acceptor ternary blend.
Judicious optimization of film thicknesses at each polymer:polymer/dye ratio is necessary
in order to obtain uniform and strong photoresponse throughout the absorption spectra of
the three component system.
107–109

77

Figure 2.6. Chemical structures and external quantum efficiency (EQE) values of binary
and ternary blend BHJ solar cells with PCBM acceptors. Reproduced with permission
from [
119
], [
108
], [
138
].

Despite the proven ability to broaden the absorption and photoresponse in ternary
blend BHJ solar cells, many ternary blend systems do not show the J
sc
and efficiency
increase and more important suffer from significant decrease in FF as the third
component is introduced in the system.
119,124,145,146
Even though the origin of the FF
decrease is not well understood, two major explanations of the behaviour can be
proposed. First possible explanation relies on the enhancement of the hole and electron
mobility differences in the ternary blend with the increase of the third component content,
thus leading to the space-charge build-up.
146,147
Second assumption is based on the well-
known bad miscibility between polymers,
148
which can lead to the formation of an
unfavourable morphology
71,72
with potentially large phase separation upon introduction
of the third component. Furthermore, J
sc
and efficiency increase in ternary blend BHJ
78
solar cells is usually achieved at small doping ratios of the third component in the blend,
since at such low doping concentrations FF is not affected or even improves in the
ternary blend BHJ solar cell due to small changes in the morphology and change
transport of the system.
143,149–157
The increase of the third component percentage in the
ternary blend is commonly described with the significant J
sc
, FF and overall efficiency
decrease.
109,114,138,141,158–171

Finally, the main limitation of the ternary blend BHJ solar cells was low V
oc
. In
addition, Brabec et al.
124
predicted the inability to tune V
oc
in ternary blend BHJ solar
cells and proposed that the V
oc
to be pinned to the smallest V
oc
of the corresponding
binary blend solar cells, as depicted in Figure 2.7 – 2.9. Considering two donor
materials, high band gap D1 and small band gap D2, with HOMO level of D2 (HOMO
D2
)
be higher than that of D1 (HOMO
D1
), corresponding binary blends will have two
different V
oc
, with V
oc
D1-A
larger than V
oc
D2-A
(Figure 2.7). In the ternary blend D1:D2:A,
absorption of the photon by the low band gap D2 leads to the formation of exciton with
the electron in LUMO
D2
and hole in HOMO
D2
, as depicted in Figure 2.8(i) and 2.8(ii).
As a result, charge transport of electron from D2 to A (Figure 2.8(iii)) leads to the V
oc
of
the ternary blend (V
oc
D1-D2-A
) be equal to V
oc
D2-A
. In case of photon absorption by the high
band gap D1, shown in Figure 2.9, generated exciton can split at two different interfaces
(D1:A and D1:D2) and thus electron has two pathways to produce photocurrent: either
direct charge transfer to acceptor (A) or through the charge transfer to D2 with the
subsequent charge transfer to A, as presented in Figure 2.9(iii) and 2.9(iv). At the same
time, generated hole on D1 has only one pathway in photocurrent generation, which
79
consists of charge transfer to D2 (Figure 2.9(v)). In case of the Förster resonance energy
transfer from D1 to D2,
172
shown in Figure 2.9(vi), the electron is then transferred to A
via the charge transfer (Figure 2.9(iv)), while the hole remains on the HOMO
D2
energy
level. As a result, it was proposed that in ternary blend BHJ solar cells all holes,
independent of the origin of generation and the photocurrent generation mechanism, have
to migrate to the higher-lying HOMO level of two donors (in our case HOMO
D2
) before
being collected at the electrode. Thus, based on the correlation between the V
oc
and
HOMO
D
-LUMO
A
, it was proposed that the V
oc
of the ternary blend BHJ solar cell (V
oc
D1-
D2-A
) has to be pinned to the smallest V
oc
of the two corresponding binary blend BHJ solar
cells (V
oc
D2-A
). The proposed pinned V
oc
was found in numerous
polymer:polymer/dye:fullerene ternary blend BHJ solar cells.
109,113,119,138,142,143,163

Figure 2.7 Open-circuit voltage (V
oc
) of corresponding binary blend BHJ solar cells.

80

Figure 2.8. Proposed origin of the V
oc
pinning in ternary blend BHJ solar cells upon light
absorption by D2.

Figure 2.9. Proposed origin of the V
oc
pinning in ternary blend BHJ solar cells upon light
absorption by D1.

Overall, until 2011 the ternary blend BHJ solar cells were seen as a way to
increase the absorption breadth of the blend and, as a result, enhance the photoresponse
and J
sc
. But high sensitivity of the FF to the ternary blend composition, leading to narrow
operating polymer:polymer/dye ratio, as well as pinning of the V
oc
to the smallest one of
the corresponding binary blends minimized the interest in the ternary blend BHJ solar
cells considering them as a hopeless way to increase the efficiency.

81
2.2.2 Tuning the V
oc
– Early Evidence

Despite the proposed pinning of the V
oc
in ternary blend BHJ solar cells, in a
limited number of cases, the V
oc
was found to be tunable in three component systems, but
at the expense of significant decrease of the FF. Chu et al.
158
synthesized conjugated
small molecule 7,7’-{5,5’-[10,12-bis(4-tert-butylphenyl)dibenzo[f,h]thieno[3,4-
b]quinoxaline-2,7-diyl]bis(thiophene-5,2-diyl)}bis(9,9-d-ihexyl-N,N-diphenyl-9H-
fluoren-2-amine) (TQTFA), shown in Figure 2.10a for the use in the ternary blend BHJ
solar cells with a structure of poly(3-hexylthiophene) (P3HT):TQTFA:PC
71
BM with a
goal to increase the absorption in the ultraviolet (UV) part of the solar spectrum.
Furthermore, HOMO level of the small molecule was found to be intermediate between
P3HT and PC
71
BM, thus delivering the V
oc
of 0.90 V in TQTFA:PC
71
BM binary blend
solar cells. Upon blending three components at different polymer:small molecule ratios,
increase of the J
sc
and decrease of the FF was observed at small doping ratios of the
TQTFA, followed by drastic decrease of J
sc
and FF at higher TQTFA concentrations, as
tabulated in Table 1. Despite the expected J
sc
and FF behaviour, the V
oc
of the ternary
system showed the continuous increase as the amount of the TQTFA with deeper-lying
HOMO was increased in the three component blend. Thus, it was demonstrated that the
V
oc
is not necessary pinned to the smallest one in ternary blend BHJ solar cells. As a
result of simultaneous increase of J
sc
and V
oc
, overall efficiency of the ternary blend was
found higher than the one of the corresponding binary blends, but was limited from
82
further enhancement by the drastic FF decrease as the amount of the TQTFA in the
system was increased.

Figure 2.10. Chemical structures of polymer and fullerene pairs used in ternary blend
BHJ solar cells with the non-pinned V
oc
.

Similar J
sc
, FF and V
oc
behaviour was found when in P3HT:PC
61
BM blend was
added low band gap diketopyrrolopyrrole-based small molecule (SMD1) with deeper-
lying HOMO level.
141
J
sc
was increased from 7.7 mA/cm
2
to 8.6 mA/cm
2
by adding just
20% of SMD1 but was accompanied with the decrease of FF. At the same time, the V
oc

was increased from 0.60 V to 0.63 V by adding 20% of SMD1 and was further enhanced
at higher SMD1 loadings.
Upon mixing two polymers as donors, different V
oc
trends were observed. Egbe et
83
al.
122
found that the V
oc
of the ternary blend BHJ solar cell based on two
phenylenevinylene (PPV) polymers is higher than either of the corresponding binary
blends. Moreover, the measured mobility of the MEH-PThE
1
-PPV
2
:DO-PThE
1
-PPV
2

blend was found to be at least one order of magnitude higher than that of neat polymers,
thus confirming efficient charge transport in some polymer:polymer blends. Same V
oc

behaviour was obtained in different polymer:polymer:PC
71
BM ternary blend.
173

Diindeno[1,2-b:2’,1’-d]thiophene (DIDT) based donor-acceptor alternating polymer
PDIDTDTBT and random copolymer PTDIDTTBT, presented in Figure 2.10b, have
identical HOMO energy level and, as a result, the V
oc
of 0.70 V and 0.68 V in binary
blend BHJ solar cells with PC
71
BM as an acceptor, as seen in Table 1. Surprisingly, the
V
oc
of the ternary blend BHJ solar cell of 0.85 V was found to be higher than that of
corresponding binary blends.
Intermediate V
oc
were also recorded for polymer:polymer:PC
61
BM/PC
71
BM
ternary blend solar cells. 4H-cyclopenta[2,1-b;3,4-b’]dithiophene (CPDT) based random
copolymer with benzothiadiazole (BTD) acceptor P1 or its structural analogue with
dithieno[3,2-b;2’,3’-d]silole (DTS) as a donor monomer P5 (Figure 2.10c and 2.10d)
were mixed with perfectly alternating poly[2,6-(4,4-bis(2-ethylhexyl)-4H-cyclopenta[2,1-
b;3,4-b’]dithiophene)-alt-4,7-(2,1,3-benzothiadiazole)] (PCPDTBT) to form a ternary
blend.
174
The V
oc
in ternary blend BHJ solar cells was found to be intermediate, as shown
in Table 1, with P1:PCPDTBT:PC
71
BM having V
oc
closer to the largest one of the
corresponding binary blends, while P5:PCPDTBT:PC
71
BM closer to the smallest one.
Similar V
oc
trend was found in other ternary blends as well.
175,176

84
Table 2.1. Photovoltaic properties of ternary blend BHJ solar cells with unpinned open-
circuit voltage
Ternary Blend J
sc
(mA/cm
2
) V
oc
(V) FF η (%)
P3HT:PC
71
BM (1:1)
P3HT:TQTFA:PC
71
BM (1:0.2:1)
P3HT:TQTFA:PC
71
BM (1:0.25:1)
P3HT:TQTFA:PC
71
BM (1:0.3:1)
P3HT:TQTFA:PC
71
BM (1:0.35:1)
P3HT:TQTFA:PC
71
BM (0.75:0.5:1)
P3HT:TQTFA:PC
71
BM (0.5:0.75:1)
TQTFA:PC
71
BM (1:1)
158

9.7
10.0
10.6
9.1
8.5
6.2
4.7
4.6
0.60
0.65
0.69
0.71
0.73
0.77
0.80
0.90
67
63
61
52
45
45
38
32
3.9
4.1
4.5
3.3
2.8
2.1
1.5
1.3
PTDIDTTBT:PC
71
BM (1:4)
PTDIDTTBT:PDIDTDTBT:PC
71
BM (0.5:0.5:4)
PDIDTDTBT:PC
71
BM (1:2)
173

6.2
6.4
5.3
0.68
0.85
0.70
47
43
44
2.0
2.4
1.7
P1:PC
71
BM (1:2)
P1:PCPDTBT:PC
71
BM (0.5:0.5:2)
PCPDTBT:PC
71
BM (1:2)
P5:PCPDTBT:PC
71
BM (0.5:0.5:2)
P5:PC
71
BM (1:2)
174

9.8
11.1
6.3
9.5
9.6
0.58
0.64
0.65
0.54
0.51
36
36
34
37
45
2.0
2.5
1.4
1.9
2.2
P3HT:PCBS (1:1)
P3HT:PCBS:PC
61
BM (1:0.5:0.5)
P3HT:PCBS:PC
61
BM (1:0.2:0.8)
P3HT:PC
61
BM (1:1)
159

6.75
9.65
9.82
10.26
0.52
0.56
0.60
0.60
56
56
65
65
1.9
3.0
3.8
4.1

In case of two acceptors and one polymer donor unpinning of the V
oc
was also
observed. Hsu et al.
159
used P3HT as a donor polymer, while PC
61
BM and phenyl-C
61
-
butyric acid styryl ester (PCBS) were used as acceptors, as seen in Figure 2.10e. Overall
P3HT:fullerene ratio was kept at 1:1 while ratio between PCBS and PC
61
BM was varied
as shown in Table 1. As can be seen, the V
oc
increases as the amount of PC
61
BM in the
ternary blend increases. The increase of the V
oc
is accompanied with the enhancement of
J
sc
, FF and efficiency. Similar unpinned V
oc
behaviour was observed for other
polymer:fullerene:fullerene blends.
117,160,177

The examples described above establish the fact that the V
oc
is not necessary
pinned in all ternary blend BHJ solar cells. The V
oc
of the ternary blend can be tuned
between the V
oc
of the corresponding binary blends and in some cases even exceed both
85
of them. The main drawback of the majority of ternary blend solar cells with the
unpinned V
oc
discussed so far is FF decrease with respect to corresponding binary blend
solar cells. As a result, as in case of ternary blends with the pinned V
oc
, efficiency
increase was observed only for small doping ratios of the third component in the blend.
Moreover, the V
oc
behaviour was unpredictable and not understood, thus reducing the
potential of the ternary blend BHJ solar cells.

2.2.3 Tuning the V
oc

Intrigued with the reports of unpinned V
oc
, Thompson et al.
106
designed ternary
blend system with a goal to better understand the V
oc
behaviour in the ternary blend BHJ
solar cells and demonstrate constant high FF at different ternary blend compositions. The
model system was based on P3HT donor, while PC
61
BM and indene-C
60
bisadduct
(ICBA) were chosen as acceptors. The choice of this acceptor pair was stipulated by a
few reasons, among which are similarities in the chemical structures, high efficiencies in
binary blend solar cells processed under similar device fabrication conditions and
different LUMO energy levels of 4.2 eV and 4.0 eV for PC
61
BM and ICBA, which
translated into the V
oc
of 0.6 V and 0.84 V for P3HT:PC
61
BM and P3HT:ICBA binary
blend BHJ solar cells, respectively, as shown in Figure 2.11.

86

Figure 2.11. Structures and corresponding electro-optical properties of P3HT, ICBA and
PC
61
BM.

Binary and ternary blend BHJ solar cells in conventional device configuration
were fabricated to show optimal device performance keeping active layer thickness
constant at around 100 nm. Overall polymer:fullerene ratio was kept at 1:1 while the
PC
61
BM:ICBA ratio was varied from 1:0 to 0:1 at 10% increment step. Table 2 lists the
average values of J
sc
, V
oc
, FF and efficiency of binary and ternary blend BHJ solar cells.
The most notable observation in the Table 2 is the V
oc
behaviour as the amount of ICBA
in the ternary blend is increased. The V
oc
not only increases from 0.605V to 0.844 V, but
also does that in the continuous fashion, as shown in Figure 2.12. Furthermore, no
decrease of the FF is observed at all three component compositions, which was attributed
to similar nanometer scale phase separation and lack of charge trapping in all ternary
87
blends independent of composition.
71,72,178–180
These observations are the first
demonstration of the predictable and tunable V
oc
between limiting V
oc
values of
corresponding binary blend BHJ solar cells accompanied with high FF at all
compositions in ternary blend BHJ solar cells.

Table 2.2. Photovoltaic properties of P3HT:PC
61
BM:ICBA ternary blend BHJ solar cells
at different fullerene ratios
106

P3HT:PC
61
BM:ICBA J
sc
(mA/cm
2
) V
oc
(V) FF η (%)
1:1:0
a
9.90 0.605 0.60 3.57
1:0.9:0.1
b
9.22 0.618 0.59 3.29
1:0.8:0.2
b
9.11 0.631 0.57 3.28
1:0.7:0.3
c
8.58 0.649 0.58 3.22
1:0.6:0.4
d
8.31 0.669 0.58 3.11
1:0.5:0.5
e
8.27 0.688 0.57 3.18
1:0.4:0.6
c
8.18 0.709 0.57 3.22
1:0.3:0.7
f
8.14 0.741 0.57 3.34
1:0.2:0.8
b
8.19 0.769 0.59 3.69
1:0.1:0.9
b
8.18 0.804 0.60 3.91
1:0:1
f
8.23 0.844 0.58 3.98
All devices were spin-coated from chlorobenzene (CB) and after aluminum deposition annealed at 150 °C
under N
2
for
a
60 min,
b
20 min,
c
40 min,
d
30 min,
e
50 min and
f
10 min.

Figure 2.12. The open-circuit voltage (V
oc
) in P3HT:PC
61
BM:ICBA ternary blend BHJ
solar cells. Reproduced with permission from [
106
].

88
The demonstrated V
oc
tunability in ternary blend BHJ solar cells opened the
practical road towards efficiency increase beyond the one of the binary blend BHJ solar
cells while retaining the simplicity of a single active layer processing step. Since the J
sc

increase can be achieved through broadening of the absorption profile of the three
component active layer, intermediate V
oc
and high FF provide necessary push of the J
sc
×
V
oc
× FF product and hence the efficiency increase.
Despite the first demonstration of the controlled and tunable V
oc
and high FF,
P3HT:PC
61
BM:ICBA ternary blend BHJ solar cells lacked to provide the increase of the
absorption breadth and consequently J
sc
increase due to the overlap of the absorption
profiles of PC
61
BM and ICBA and high band gap of P3HT, as seen in Figure 2.11,
although all important properties of P3HT necessary for device performance were
preserved. As a result, the next goal was to show simultaneous increase of the absorption
breadth, high FF and tunable V
oc
.
Thompson et al.
107
studied the ternary blend BHJ solar cell system based on two
donor polymers and PC
61
BM acceptor with a goal to show J
sc
× V
oc
× FF increase. The
polymers had to satisfy certain requirements, including high efficiencies in binary blend
BHJ solar cells, complimentary absorption profiles for J
sc
increase and finally different
HOMO energy levels to show intermediate and tunable V
oc
in ternary blend BHJ solar
cells. Low band gap semi-random poly(3-hexylthiophene-thiophene-
diketopyrrolopyrrole) (P3HTT-DPP-10%)
181
and high band gap random poly(3-
hexylthiophene-co-3-(2-ethylhexyl)thiophene) (P3HT
75
-co-EHT
25
)
182
were synthesized
and chosen as suitable candidates with the electro-optical properties shown in Figure
89
2.13.

Figure 2.13. Structures and corresponding electro-optical properties of PC
61
BM, P3HT
75
-
co-EHT
25
and P3HTT-DPP-10%.

Binary and ternary blend BHJ solar cells in conventional device configuration
were fabricated and individually optimized to show optimal device performance. The
optimization included finding overall polymer:fullerene ratio at each polymer:polymer
ratio, as well as, modulation of active layer thickness at each ratio. Table 3 lists the
average values of J
sc
, V
oc
, FF and efficiency of binary and ternary blend BHJ solar cells.
As a result of the individual optimization of each ternary blend BHJ solar cell, the
efficiency of the ternary blend solar cell increased beyond the one of the corresponding
binary blends. The increase of the ternary blend BHJ solar cell efficiency was stipulated
90
by simultaneous increase of J
sc
and intermediate value of the V
oc
, as well as high FF.
Overall efficiency increase of the ternary blend BHJ solar cells beyond the one of the
corresponding binary blend solar cells supports the vision of ternary blend solar cells as
an effective way to overcome efficiency limit set by binary blend solar cells.

Table 2.3. Photovoltaic properties of P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM
ternary blend BHJ solar cells at optimized ratios
107

P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM J
sc
(mA/cm
2
) V
oc
(V) FF η (%)
1:0:1.3
a
14.38 0.574 0.62 5.07
0.9:0.1:1.1
b
15.05 0.603 0.61 5.51
0.8:0.2:1.0
b
14.60 0.608 0.61 5.37
0.7:0.3:1.0
c
11.54 0.614 0.59 4.15
0.6:0.4:1.0
c
11.19 0.619 0.59 4.12
0.5:0.5:0.9
a
10.89 0.622 0.59 4.00
0.4:0.6:0.9
a
10.19 0.626 0.59 3.74
0.3:0.7:0.8
a
9.77 0.633 0.59 3.64
0.2:0.8:0.8
a
8.57 0.639 0.60 3.27
0.1:0.9:0.9
a
8.25 0.646 0.59 3.10
0:1:0.8
d
7.96 0.675 0.59 3.16
All devices were spin-coated from o-dichlorobenzene (o-DCB) and placed to the N
2
cabinet before
aluminum deposition for
a
30 min,
b
60 min,
c
45 min and
d
20 min.

Figure 2.14. Open-circuit voltage (V
oc
) (black squares – left axis) and short-circuit
current density (J
sc
) (red circles – right axis) of the individually optimized ternary blend
BHJ solar cells from Table 2.3 with different fraction of the polymer P3HT
75
-co-EHT
25

component in the blends. Reproduced with permission from [
107
].

91

Figure 2.15. Open-circuit voltage (V
oc
) of the individually optimized ternary blend BHJ
solar cells (open squares), with fixed overall polymer:PC
61
BM ratio at 1:1.1 (blue stars)
and with fixed overall polymer:PC
61
BM ratio at 1:1.0 (green triangles). Reproduced with
permission from [
107
].

The V
oc
was found to be tunable in P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM
ternary blend BHJ solar cells but the behaviour was different than in case of
P3HT:PC
61
BM:ICBA,
106
as illustrated in Figure 2.14. The transition from binary to
ternary blends was accompanied with the rapid change of the V
oc
by 29 mV. The V
oc
of
individually optimized solar cells changed linearly in the ternary blend regime (Figure
2.14), steadily increasing as the amount of P3HT
75
-co-EHT
25
with lower-lying HOMO
energy level increased in the blend. The introduction of the low bad gap polymer into the
blend increased the absorption breadth of the ternary system, as shown in Figure 2.13.
This translated to the photoresponse of the ternary blend BHJ solar cells in the NIR, with
the EQE values and hence J
sc
increase as the amount of the P3HTT-DPP-10% in the
ternary blend is increased as can be seen in Figure 2.6a and 2.14. Different behaviour of
the V
oc
was recorded when the overall polymer:fullerene ratio was not individually
optimized at each polymer:polymer ratio, as shown in Figure 2.15. Decrease from
92
linearity and reduction of the V
oc
values were observed when overall polymer:fullerene
ratio was fixed at 1:1.1 and 1:1. Furthermore, in cases with non-optimized ternary blend
BHJ solar cells, J
sc
, FF and efficiency were significantly reduced, thus highlighting the
importance of optimization of ternary blend BHJ solar cells.
The groundbreaking works by Thompson et al.
106,107
proved and demonstrated the
V
oc
tunability in ternary blend BHJ without negative impact on the J
sc
and FF which
translated to the overall efficiency increase beyond the corresponding binary blend solar
cells.
Since these original publications, the field of the ternary blend BHJ solar cells
attracted significant scientific attention. You et al.
108
used the ternary blend of two
polymers and PC
61
BM, as represented in Figure 2.6c, and showed increase of the
efficiency of ternary blend BHJ solar cells with respect to corresponding binary blend
BHJ solar cells, going from 6.3% to 7.0%, respectively. The efficiency increase became
possible due to enhancement of the breadth and strength of the photoresponse by the
three component blends (Figure 2.6c), as well as intermediate V
oc
and high FF at all
ternary blend compositions. Brabec et al.
109,119,120
applied NIR sensitization upon
blending high band gap P3HT and low band gap Si-PCPDTBT, shown in Figure 2.6b,
with different fullerene acceptors (PC
61
BM or ICBA). In all ternary blend systems,
increase of the J
sc
was observed upon addition of Si-PCPDTBT due to enhancement of
photoresponse in the NIR (Figure 2.6b). The V
oc
was found to be pinned to the smallest
one of the corresponding binary blend solar cells with PC
61
BM as an acceptor,
109,119

while in case of P3HT:Si-PCPDTBT:ICBA ternary blend solar cells the V
oc
was found to
93
be intermediate.
120
The main problem which authors encountered was the low operating
window for efficient ternary blend BHJ solar cells. Introduction of more than 20 – 30%
of Si-PCPDTBT in the ternary blend lead to steady decrease of FF. As a result of
thickness optimization, ternary blend BHJ solar cells showed best device performance at
20 – 30% of Si-PCPDTBT in the blend, which was higher than that of corresponding
binary blend solar cells.
109,120
A few other publications also studied P3HT NIR
photoresponse sensitization by adding PCPDTBT as a low bang gap polymer and
PC
61
BM as an acceptor with a goal to increase J
sc
and efficincy.
136,183
Even though the
pinned V
oc
behaviour in P3HT:PCPDTBT:PC
61
BM ternary blend solar cells was found to
be similar to the previous results,
124
careful optimization of the ternary blend active layer
using solvent additive 1,8-diiodooctane (DIO) helped maintaining high and constant FF
around 0.50 at PCPDTBT loading up to 30% which together with the expected J
sc

increase translated into 17% increase in the PCE over P3HT:PC
61
BM binary blend solar
cells. Several other polymer:polymer:fullerene ternary blend BHJ solar cells also showed
efficiency increase based on the J
sc
enhancement and/or intermediate V
oc
.
121,144,145,184

Different approach to increase the V
oc
and efficiency was used by Hoppe et al.
145

Upon mixing amorphous and semi-crystalline donor polymers the efficiencies beyond the
one of the corresponding binary blends were achieved. Anthracene-containing poly(p-
phenylene-ethynylene)-alt-poly(p-phenylene-vinylene) (PPE-PPV) copolymers (AnE-
PVs) were used: amorphous AnE-PVba and semi-crystalline AnE-PVab. Upon adding
from 10 to 70% of AnE-PVba, with the smaller V
oc
in binary blend solar cell, as the third
component in the ternary blend with PC
61
BM acceptor, the V
oc
increased beyond the one
94
of both of corresponding binary blend solar cells, as can be seen in Figure 2.16. At the
same time, high FF were recorded when the amount of amorphous AnE-PVba in the
ternary blend was below 50%. Even though both polymers had similar band gaps and no
increase in the J
sc
of the three component blends was recorded, high FF and enhanced V
oc

translated into overall PCE increase in ternary blend BHJ solar cells with respect to
corresponding binary blend solar cells. Similar strategy was adopted by Murata et al.
171

where regiorandom P3HT (RRa-P3HT) with lower lying HOMO energy level was added
as the third component into the regioregular P3HT (RR-P3HT):PC
61
BM blend. Addition
of up to 20% of RRa-P3HT lead to marginal changes in J
sc
, while the V
oc
, FF and
correspondingly efficiency were improved.

Figure 2.16. Tunable open-circuit voltage (V
oc
) in AnE-PVba:AnE-PVab:PC
61
BM
ternary blend BHJ solar cells. Reproduced with permission from [
145
].

95

Figure 2.17. Tunable open-circuit voltage (V
oc
) in P3HT:OXCMA:OXCBA ternary
blend BHJ solar cells. Reproduced with permission from [
146
].

Figure 2.18. Tunable open-circuit voltage (V
oc
) in P3HT:CdSe
NR
:CdSe
NC
ternary blend
BHJ solar cells. Reproduced with permission from [
185
].

Addition of small molecule or dye also lead to the increase of the ternary blend
BHJ solar cell efficiency based of the J
sc
× V
oc
enhancement. Park et al.
137
used
anthracene-based star-shaped conjugated small molecules and either P3HT or poly((5,5-
E-alpha-((2-thienyl)methylene)-2-thiopheneacetonitrile)-alt-2,6-[(1,5-
96
didecyloxy)naphthalene])) (PBTADN) as donors in ternary blend BHJ solar cells with
PC
71
BM acceptor. It was observed that J
sc
was increased due to the additional absorption
by the small molecule, while the V
oc
was found to be intermediate. As a consequence, the
efficiency of the ternary blend system was increased by almost 87% relative to
PBTADN:PC
71
BM binary blend solar cells.
In many cases the incorporation of the small molecule/dye lead to minor changes
in the V
oc
, while J
sc
was significantly increased. Forrest et al.
138
and Taylor et al.
134
used
squaraine (SQ) dyes, which were only different in the terminal groups, to enhance the
light absorption in the NIR in P3HT:SQ:PC
61
BM ternary blend solar cells. As shown in
Figure 2.6d, incorporation of just 10% of 2,4-bis[4-(N,N-diphenylamino)-2,6-
dihydroxyphenyl] squaraine (DPSQ) translates into significant broadening of the
photoresponse of the three component blend and J
sc
increase by almost 20%.
Furthermore, FF remained unaffected at the studied dye doping ratios and thus the PCE
increase was observed in ternary blend BHJ solar cells. Similar results were obtained in
other polymer:small molecule/dye:fullerene solar cells.
139,163,186–190

The controlled and tunable non-linear V
oc
behaviour similar to
P3HT:PC
61
BM:ICBA
106
ternary blend BHJ solar cells was obtained upon mixing of
P3HT with two fullerene-based acceptors o-xylenyl C
60
mono-adduct (OXCMA) with
LUMO energy level of 3.83 eV and o-xylenyl C
60
bis-adduct (OXCBA) with LUMO
energy level of 3.66 eV.
146
The V
oc
was tuned from 0.63 V to 0.84 V in the continuous
fashion as the amount of OXCBA with higher-lying LUMO energy level in the ternary
blend was increased, as can be seen in Figure 2.17. The obtained J
sc
was high (around 10
97
mA/cm
2
) at all P3HT:OXCMA:OXCBA ratios due to similar absorption by OXCMA and
OXCBA. Furthermore, FF remained above 0.51 throughout the studied range of ternary
blend BHJ solar cells. As in case of P3HT:PC
61
BM:ICBA,
106
despite the intermediate V
oc

and high FF at all P3HT:OXCMA:OXCBA ratios, the efficiency of ternary blend BHJ
solar cells was intermediate between the corresponding binary blends due to the lack of
the photoresponse broadening and enhancement. Efficiency increase and in some cases
similar tunable V
oc
behaviours were observed for other polymer:fullerene:fullerene
systems.
114,116,191–193
Interesting to mention, that substitution of one or both fullerene
derivatives in polymer:fullerene:fullerene system with inorganic nanoparticles facilitates
J
sc
increase and efficiency of the ternary blend systems,
185,194
while the V
oc
dependence
was found to be linear in the ternary blend regime, as shown in Figure 2.18.
185

The pioneering work by Thompson et al.
106,107
showed the predictable V
oc

tunability in ternary blend BHJ solar cells without negative impact on J
sc
and FF and
demonstrated the possibility of the efficiency increase beyond the one of binary blend
solar cells. Furthermore, photoresponse broadening and increase, charge carrier mobility
enhancement and resistance optimization, as well as preserving the nanometer scale
morphology in ternary blend BHJ solar cells, showed by others
108,109,116,119,120,122,134,138

further supports the vision of ternary blend solar cells as an effective way to overcome
efficiency limit set by binary blend solar cells.

98
2.2.4 Models

The ability to tune the V
oc
was shown in a couple of organic ternary blend systems
so far.
106,107,146,185
But the origin of the non-pinned V
oc
as well as the difference in the V
oc

behaviour in the ternary blend regime (linear
107
versus nonlinear
106
) remained unclear.
Three major models were introduced to explain the J
sc
and V
oc
behaviour in ternary blend
BHJ solar cells: sensitization/cascade model,
113,124
parallel-like model
108,195
and organic
alloy model.
123
All of these models explain the J
sc
increase, however some of them fail to
explain tunable V
oc
behaviour in ternary blend BHJ solar cells but give explanation to the
V
oc
pinning in other three component systems.
The sensitization/cascade model was proposed by Brabec et al.
113,124
Considering
D1:D2:A ternary blend solar cell, this model assumes that the HOMO and LUMO energy
levels of the three components are intermediate (HOMO
D1
< HOMO
D2
< HOMO
A
and
LUMO
D1
> LUMO
D2
> LUMO
A
), thus allowing cascade exciton dissociation and charge
transfer. As a result, charges can be transported through the individual domains but also
hop from one donor to another if the energy difference is sufficient, thus allowing to use
the three diode model with two diodes in parallel
191
for the charge transport modelling.
Consequently, it was concluded that for the sensitization/cascade model in ternary blend
BHJ solar cells all holes have to migrate to the highest-lying HOMO level of two donors
(HOMO
D2
) before being collected at the electrode, independent of the origin of
generation and the photocurrent transport mechanism. Thus, the model explains the
photocurrent and hence J
sc
increase, however the V
oc
of the ternary blend solar cells
99
would have to remain pinned to the one of the smallest one of the corresponding binary
blend solar cells for the sensitization/cascade model.
Another explanation of the V
oc
tunability proposed by You et al.
108
is based on the
idea of parallel connection of the binary blend “sub-cells” in ternary blend BHJ solar
cells – parallel-like model. This model implies that excitons generated in each donor
domain, considering ternary blend system D1:D2:A, migrate to their respective
donor:acceptor interface and then dissociate into free charge carriers. After this,
generated holes travel through their corresponding donor domains to the anode, while
electrons are transported through the fullerene domains to the cathode. It is proposed that
the holes generated in one donor domain cannot migrate into the other donor domain,
even if the energy between the HOMO levels is sufficient, thus creating the analogue of
the parallel connection of the two individual binary blend solar cells.
195
In order to
explore this idea, P3HT:PC
61
BM:ICBA ternary blend BHJ solar cells were studied.
191,196

The equivalent circuit based on three diode model with two diodes, which represented
P3HT:PC
61
BM and P3HT:ICBA, in parallel was proposed. Unfortunately, the authors did
not present the results of the fitting but only used the simplified expression for the V
oc

using Shockley equation.
56,70

The idea of the V
oc
tunability assuming parallel-like model in ternary blend BHJ
solar cells fails to explain intermediate V
oc
. If the parallel connection of the “sub-cells”
would take place, the overall V
oc
of the ternary blend would obey the rules similar to the
one as in case of tandem solar cells with parallel connection of the sub-cells, where the
V
oc
is limited by the minimum V
oc
of the individual sub-cells.
8,98,100
Recent study of the
100
ternary blend BHJ solar cells based on the P3HT:PC
6
PT:PC
61
BM, where PC
6
PT is
poly(2,3-dihexyl thieno(3,4-b)pyrazine), demonstrated the V
oc
pinning to the smallest V
oc

among corresponding binary blend solar cells upon using the equivalent circuit model
with parallel connection of binary blend “sub-cells”.
197

In case of an organic alloy, upon deeper analysis of the composition dependent
V
oc
behaviour in ternary blend BHJ solar cells an analogy can be drawn to electronic
properties in the inorganic alloys where the continuous change of the HOMO and LUMO
energy levels with composition is common.
198–200
Despite the similarity in the V
oc

behaviour in ternary blend systems and inorganic alloys, the significant deference in the
electro-optical properties is observed. Inorganic alloys do not preserve the individual
properties of the components of the blend forming a new set of electro-optical properties
at each composition, in contrast to organic ternary blends, where each component retains
the set of individual physical properties, for example absorption, as can be seen in Figure
2.6 and 2.13. With a purpose to unravel the nature of the V
oc
behaviour Street et al.
123

studied two ternary blend systems discussed above: P3HT:PC
61
BM:ICBA and P3HTT-
DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM.

101

Figure 2.19. Photocurrent spectral response (PSR) and energy of CT state compared to
the values of the V
oc
data for the P3HT:PC
61
BM:ICBA.

Figure 2.20. Photocurrent spectral response (PSR) and energy of CT state compared to
the values of the V
oc
data for the P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM ternary
blend solar cells.

The V
oc
of the organic BHJ solar cells depends on the difference between HOMO
energy level of the donor and LUMO energy level of the acceptor (Figure 2.1). Recently,
102
it was determined that the energy of charge transfer (CT) state, which can be imagined as
the hole at the donor’s HOMO and electron at the acceptor’s LUMO, gives a better
description of the V
oc
in binary blend BHJ solar cells.
56,61,62
In order to probe the V
oc
in
ternary blend BHJ solar cells photocurrent spectral response (PSR) measurements, which
measure the optical absorption of the heterojunction interface (CT state energy),
201
were
performed. As can be seen in Figure 2.19 and 2.20, the energy of CT state increases
continuously in both cases, as the amount of higher-lying LUMO ICBA in
P3HT:PC
61
BM:ICBA and deeper-lying HOMO P3HT
75
-co-EHT
25
in P3HTT-DPP-
10%:P3HT
75
-co-EHT
25
:PC
61
BM is increased. Furthermore, the changes in the CT state
energy track in the same way as the measured V
oc
in the ternary blend BHJ solar cells,
confirming that the V
oc
accurately measures the change of the CT state energy and that
the CT state energy changes continuously with the composition of two component
acceptor or donor. The qV
oc
in ternary blend BHJ solar cells was found to be smaller than
the energy of CT state by 0.55 eV, which is the same difference found in binary blend
BHJ solar cells.
56,61,62,87,201–204
Similar dependences of the energy of CT state and the V
oc
,
as well as the difference between the CT state energy and the qV
oc
of 0.55 eV, as in case
of binary blend solar cells, allowed authors to conclude the formation of the organic
alloys (from the electronic and not morphological point of view) in P3HT:PC
61
BM:ICBA
and P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM ternary blends.

103

Figure 2.21. Expanded plot of the peaks near 1.7 eV in the PSR spectra of
P3HT:PC
61
BM:ICBA. The peak centered above 1.7 eV corresponds to PC
61
BM
absorption and the peak centered below 1.7 eV corresponds to ICBA.

Figure 2.22. High energy PSR data of P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM
show the exciton peaks from the donor mixture.

Despite the formation of the organic alloy, it was determined that in both ternary
blends individual absorptions of all components are preserved, as shown in Figure 2.21
and 2.22. For the case of P3HT:PC
61
BM:ICBA (Figure 2.21), the absorption of the
104
acceptor blend is not a single energy transition but rather a weighted sum of two
absorptions of each of the acceptors. The same behaviour was observed for P3HTT-DPP-
10%:P3HT
75
-co-EHT
25
:PC
61
BM ternary blends (Figure 2.22), where the addition of low
band gap P3HTT-DPP-10% lead to the increase of the photocurrent in the low energy
region, while the position of peaks denoted E1, E2 and E3 did not change. Overall, in
both P3HT:PC
61
BM:ICBA and P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM systems,
the optical transitions retain the properties of the individual materials of the mixture,
while the energy of CT state exhibits the properties of an average composition of the two
component donor or acceptor.
The difference between the V
oc
and energy of CT state behaviour according to an
organic alloy model and retaining the optical properties of the individual materials of the
mixture were explained based on the nature of excitons and CT excitons, as shown in
Figure 2.23 and 2.24. Optical transitions are excitonic in character. In the organic
materials, upon light absorption the created Frenkel-type exciton
205,206
is described with
high Coulomb interaction between the hole and the electron.
56,61
As a result of strong
hole and electron wave function overlap, exciton is localized on a single molecule or
segment of the molecule,
207–209
thus remaining independent from the other components in
the blend and preserving the optical properties of the individual components, rather than
an average of the blend (Figure 2.23). At the same time, CT excitons, created after
photoinduced charge transfer,
56–58,61
are described with the hole on the donor and the
electron on the acceptor (Figure 2.23). Since donor and acceptor are spatially separated,
the Coulomb interaction is reduced with respect to an exciton, thus increasing the
105
delocalization of a CT exciton. The delocalized hole and electron extend to more than
one molecule,
207,208,210,211
consequently representing the average composition of donor
and acceptor at the interface, rather than individual components of the blend. On the
simplified energy diagram, the V
oc
composition dependence and tunability of the ternary
blend BHJ solar cells can be presented as a weighted average HOMO energy level
(HOMO
D1-D2
) of two donors D1 and D2 or a weighted average LUMO energy level
(LUMO
A1-A2
) of two acceptors A1 and A2, shown in Figure 2.24. Independent of the
origin of the exciton generation, HOMO energy level of a donor alloy system will be
limited by the HOMO energy levels of the corresponding alloy components, as well as
LUMO energy level of an acceptor alloy system will be limited by the LUMO energy
levels of the corresponding alloy components. In other words, each component of the
ternary blend generate an exciton but after photoinduced charge transfer, an extensive
electronic mixing of an organic alloy controls the position of the HOMO (or LUMO)
level, CT state energy and the V
oc
of the ternary blend BHJ solar cell.

106

Figure 2.23. Exciton and charge transfer (CT) exciton in organic BHJ solar cells.

The V
oc
tunability is unlikely to be described using parallel-like model or
sensitization/cascade model, but these models can explain the V
oc
pinning observed in
some ternary blend BHJ solar cells. So far, an organic alloy model is the only explanation
of the V
oc
tunability in ternary blend BHJ solar cells supported by the experimental
evidence.
Furthermore, in case of the intermediate V
oc
observed in P3HT:PC
61
BM:ICBA
191

and DTffBT:DTPyT:PC
61
BM
108
ternary blend BHJ solar cells, the chemical structures of
PC
61
BM and ICBA, as well as DTffBT and DTPyT (Figure 2.6c) are similar which allow
to assume good electronic interaction between the components. As was discussed
above,
123
P3HT:PC
61
BM:ICBA and P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM form
an organic alloy and it is plausible that DTffBT and DTPyT also form an alloy and
provide extensive electronic interaction necessary for the V
oc
tunability.
107
At the same time, the increase of the V
oc
in ternary blend BHJ solar cells beyond
the V
oc
of corresponding binary blends recorded in some three component solar cells can
be explained based on the decrease of the crystallinity of the host
donor
122,139,140,144,145,171,176,212–214
or acceptor
116,192,215
and thus the decrease of the HOMO
energy level in case of two donors and one acceptor or increase of the LUMO energy
level in case of one donor and two acceptors can be observed.

Figure 2.24. Composition dependent HOMO energy level (HOMO
D1-D2
) of an organic
alloy blend based on two donors (top) and composition dependent LUMO energy level
(LUMO
A1-A2
) of an organic alloy blend based on two acceptors (bottom). Arrows indicate
composition dependent CT state. LUMO
D1-D2
and HOMO
D1-D2
drawn unchanged for
simplification.
108
As was described above, an organic alloy model is the only model that can
explain simultaneous J
sc
increase and intermediate V
oc
in ternary blend BHJ solar cells
using the difference between the exciton and CT exciton nature. In case of J
sc
, the low
band gap donor provides high energy optical transitions to more fully cover the solar
energy spectrum, while the addition of the high energy material pushes the one electron
LUMO states of the donor mixture higher in energy by the alloying effect. The band
offset at the interface
56,87,216,217
is therefore increased and the exciton is more efficiently
split at the interface, even though the exciton energies are unchanged. The changes in the
V
oc
show that the HOMO and LUMO levels change continuously with composition in the
respective two component donor or acceptor pair, consistent with the formation of an
organic alloy. Other models, like parallel-like model or sensitization/cascade model, can
explain the J
sc
increase but lack the explanation of the V
oc
tunability in ternary blend BHJ
solar cells. At the same time, parallel-like model or sensitization/cascade model explain
well the examples of the V
oc
pinning in some ternary blend BHJ solar cells.

2.3 Future Challenges and Directions

Recent achievements in the field of the organic ternary blend BHJ solar cells
stimulate the extensive research and support the vision of ternary BHJ blend solar cells as
a promising and effective way to overcome efficiency limit set by binary blend solar cells
without complicating the device fabrication procedure and preserving the attractive
features of OPV. The main achievement in understanding of ternary blend BHJ solar cells
109
was the discovery of the formation an organic alloy in three component blends
123
which
is responsible for simultaneous J
sc
increase, high FF and tunable V
oc
between the V
oc
of
the corresponding binary blend BHJ solar cells.
106,107
The formation of an organic alloy
allowed the efficiency increase of the ternary blend BHJ solar cells beyond the one of the
corresponding binary blend solar cells.
In order to further pursue the efficiency increase in ternary blend BHJ solar cells a
set of questions and challenges has to be clarified first. The main challenge is the
understanding of the morphology in the three component solar cells. The first attempts to
probe the ternary blend morphology were recently undertaken
118,134,186,218–220
but were
unable to give a conclusive answer about the nano-morphology. Techniques such as,
transmission electron microscopy (TEM) and atomic force microscopy (AFM), can only
provide the morphology of binary polymer:fullerene blends and, even though the
formation of bicontinious blends in the most efficient ternary blend BHJ solar cells is
shown,
107
cannot depict the morphology between two polymers in
polymer:polymer:fullerene ternary blends due to similar electron densities of the majority
of polymers. Near-edge X-ray absorption experiments (NEXAFS) which analyze the
element specific X-ray absorption spectra of the film and TEM energy dispersive X-ray
elemental mapping can possibly answer the questions about the morphology of the
ternary blends, but no studies has yet to be performed. By extension, the morphology
studies should shed light on the difference between the V
oc
tunability in organic alloys
and the V
oc
pinning when organic alloy is not formed, i.e. organic alloy model vs. parallel
connection of the binary blend “sub-cells” model and energy level cascade model. In
110
addition, time-dependent morphology studies should evaluate the effect of ternary mixing
on the stability/degradation of the active layer morphology in three component blends
and should be correlated with the stability studies of the ternary blend BHJ solar cells.
Several recent studies related to the hydrophobicity/hydrophilicity of the
components in the ternary blend were conducted with the goal to study their effects on
J
sc
, V
oc
, FF, efficiency, as well as, series and parallel resistance.
142,221–223
It was concluded
that the surface energy of the third component plays an important role on the solar cell
performance and that with judicious choice of the third component J
sc
can be improved
without negative impact on FF and also decreasing series resistance, thus leading to the
efficiency increase. Further studies are necessary to identify the effect of the
hydrophobicity/hydrophilicity of the three components in the blend on the V
oc
, since only
small doping ratios were used in current studies. In general, the miscibility of all three
components in the blend should influence the morphology and thus the device
parameters: J
sc
, FF and V
oc
. General guidelines about the polymer miscibility and its
effect on the solar cell parameters are necessary for the future polymer design.
The next challenge to tackle is to determine the origin of the different V
oc

behaviour in ternary blend regime in case of polymer:polymer:fullerene (linear)
107
and
polymer:fullerene:fullerene (non-linear).
106,146,191
Since it was determined that in both
cases the formation of an organic alloy is responsible for the V
oc
tunability,
123

understanding the differences in the V
oc
behaviour should facilitate better control and
predictability of the tunable V
oc
in ternary blend BHJ solar cells and provide the next
steps for the efficiency increase.
111
From the electronic point of view, charge transport in ternary blend BHJ solar
cells still remains not well understood. Furthermore, charge transport dynamics
measurements were only performed on the ternary blends with the pinned V
oc
109,113,124

and no studies were yet performed on the ternary blends that form an organic alloy. The
transient absorption spectroscopy (TAS) studies should also explain the high contribution
of the small band gap polymers to the photocurrent at all polymer:polymer ratios in
ternary blend BHJ solar cells even when its concentration is below the usual percolation
threshold: either the exciton can diffuse easily through the dilute component, or the
transition to the interface is by a direct energy transfer process that does not involve
diffusion.
Finally, in order to simplify the donor polymers structure design, the origin of
frequently observed charge trapping
154,224–226
in ternary blend BHJ solar cells has to be
studied. It is of great importance to understand the relationship between the positions of
the HOMO and LUMO energy levels in three component systems and determine general
rules for the HOMO-HOMO (or LUMO-LUMO) offset in polymer:polymer
(fullerene:fullerene) based ternary blends necessary to avoid charge trapping and
facilitate the formation of an organic alloy instead.
Overcoming the challenges which are currently facing ternary blend BHJ solar
cells will further push the efficiencies of organic solar cells while retaining the simplicity
of the solar cell fabrication. As was discussed in current review, a suitably designed
ternary blend system has the potential for much higher solar cell efficiency than binary
blend systems.
112
2.4 Chapter Bibliography

(1) Jacobson, M. Z.; Delucchi, M. A. Energy Policy 2011, 39, 1154.

(2) Delucchi, M. A.; Jacobson, M. Z. Energy Policy 2011, 39, 1170.

(3) Annual Energy Outlook 2013 - Energy Information Administration
http://www.eia.gov/forecasts/aeo/ (accessed May 28, 2013).

(4) Annual Energy Review - Energy Information Administration
http://www.eia.gov/totalenergy/data/annual/ (accessed May 28, 2013).

(5) Armaroli, N.; Balzani, V. Energy Environ. Sci. 2011, 4, 3193.

(6) Ameri, T.; Li, N.; Brabec, C. J. Energy Environ. Sci. 2013, DOI:
10.1039/c3ee40388b.

(7) Tao, M. Electrochem. Soc. Interface 2008, 17, 30.

(8) Ameri, T.; Dennler, G.; Lungenschmied, C.; Brabec, C. J. Energy Environ. Sci.
2009, 2, 347.

(9) Devabhaktuni, V.; Alam, M.; Shekara Sreenadh Reddy Depuru, S.; Green, R. C.;
Nims, D.; Near, C. Renew. Sust. Energ. Rev. 2013, 19, 555.

(10) Avrutin, V.; Izyumskaya, N.; Morkoç, H. Superlattices Microstruct. 2011, 49, 337.

(11) Razykov, T. M.; Ferekides, C. S.; Morel, D.; Stefanakos, E.; Ullal, H. S.;
Upadhyaya, H. M. Sol. Energy 2011, 85, 1580.

(12) Hegedus, S. WIREs Energy Environ. 2013, 2, 218.

(13) Green, M. A.; Emery, K.; Hishikawa, Y.; Warta, W.; Dunlop, E. D. Prog.
Photovolt: Res. Appl. 2013, 21, 1.

(14) Jäger-Waldau, A. Int. J. Photoenergy 2012, 2012, 1.

(15) Jäger-Waldau, A. Green 2011, 1, 277.

(16) Schmidtke, J. Opt. Express 2010, 18, A477.

(17) Hardin, B. E.; Snaith, H. J.; McGehee, M. D. Nat. Photonics 2012, 6, 162.
113
(18) Hagfeldt, A.; Boschloo, G.; Sun, L.; Kloo, L.; Pettersson, H. Chem. Rev. 2010,
110, 6595.

(19) Jung, H. S.; Lee, J.-K. J. Phys. Chem. Lett. 2013, 4, 1682.

(20) Zhang, S.; Yang, X.; Numata, Y.; Han, L. Energy Environ. Sci. 2013, 6, 1443.

(21) Yella, A.; Lee, H.-W.; Tsao, H. N.; Yi, C.; Chandiran, A. K.; Nazeeruddin, M. K.;
Diau, E. W.-G.; Yeh, C.-Y.; Zakeeruddin, S. M.; Gratzel, M. Science 2011, 334,
629.

(22) Burschka, J.; Dualeh, A.; Kessler, F.; Baranoff, E.; Cevey-Ha, N.-L.; Yi, C.;
Nazeeruddin, M. K.; Grätzel, M. J. Am. Chem. Soc. 2011, 133, 18042.

(23) Chung, I.; Lee, B.; He, J.; Chang, R. P. H.; Kanatzidis, M. G. Nature 2012, 485,
486.

(24) Snaith, H. J. Adv. Funct. Mater. 2010, 20, 13.

(25) Xu, T.; Qiao, Q. Energy Environ. Sci. 2011, 4, 2700.

(26) Wright, M.; Uddin, A. Sol. Energy Mater. Sol. Cells 2012, 107, 87.

(27) He, M.; Qiu, F.; Lin, Z. J. Phys. Chem. Lett. 2013, 4, 1788.

(28) Thompson, B. C.; Fréchet, J. M. J. Angew. Chem. Int. Ed. 2008, 47, 58.

(29) Beaujuge, P. M.; Amb, C. M.; Reynolds, J. R. Acc. Chem. Res. 2010, 43, 1396.

(30) Moulé, A. J.; Chang, L.; Thambidurai, C.; Vidu, R.; Stroeve, P. J. Mater. Chem.
2012, 22, 2351.

(31) Dennler, G.; Scharber, M. C.; Brabec, C. J. Adv. Mater. 2009, 21, 1323.

(32) Brabec, C. J.; Gowrisanker, S.; Halls, J. J. M.; Laird, D.; Jia, S.; Williams, S. P.
Adv. Mater. 2010, 22, 3839.

(33) Su, Y.-W.; Lan, S.-C.; Wei, K.-H. Mater. Today 2012, 15, 554.

(34) Li, G.; Zhu, R.; Yang, Y. Nat. Photonics 2012, 6, 153.

(35) Mayukh, M.; Jung, I. H.; He, F.; Yu, L. J. Polym. Sci., Part B: Polym. Phys. 2012,
50, 1057.

114
(36) He, F.; Yu, L. J. Phys. Chem. Lett. 2011, 2, 3102.

(37) Son, H. J.; Carsten, B.; Jung, I. H.; Yu, L. Energy Environ. Sci. 2012, 5, 8158.

(38) Beaujuge, P. M.; Fréchet, J. M. J. J. Am. Chem. Soc. 2011, 133, 20009.

(39) Schlenker, C. W.; Thompson, M. E. Top. Curr. Chem. 2011, 312, 175.

(40) Günes, S.; Neugebauer, H.; Sariciftci, N. S. Chem. Rev. 2007, 107, 1324.

(41) Coakley, K. M.; McGehee, M. D. Chem. Mater. 2004, 16, 4533.

(42) Zhou, H.; Yang, L.; You, W. Macromolecules 2012, 45, 607.

(43) Chochos, C. L.; Choulis, S. A. Prog. Polym. Sci. 2011, 36, 1326.

(44) Bian, L.; Zhu, E.; Tang, J.; Tang, W.; Zhang, F. Prog. Polym. Sci. 2012, 37, 1292.

(45) Li, W.; Furlan, A.; Hendriks, K. H.; Wienk, M. M.; Janssen, R. A. J. J. Am. Chem.
Soc. 2013, 135, 5529.

(46) He, Z.; Zhong, C.; Su, S.; Xu, M.; Wu, H.; Cao, Y. Nat. Photonics 2012, 6, 593.

(47) Dou, L.; Chang, W.-H.; Gao, J.; Chen, C.-C.; You, J.; Yang, Y. Adv. Mater. 2013,
25, 825.

(48) You, J.; Dou, L.; Yoshimura, K.; Kato, T.; Ohya, K.; Moriarty, T.; Emery, K.;
Chen, C.-C.; Gao, J.; Li, G.; Yang, Y. Nat. Commun. 2013, 4, 1446.

(49) You, J.; Chen, C.-C.; Hong, Z.; Yoshimura, K.; Ohya, K.; Xu, R.; Ye, S.; Gao, J.;
Li, G.; Yang, Y. Adv. Mater. 2013, DOI: 10.1002/adma.201300964.

(50) Son, H. J.; He, F.; Carsten, B.; Yu, L. J. Mater. Chem. 2011, 21, 18934.

(51) Uy, R. L.; Price, S. C.; You, W. Macromol. Rapid Commun. 2012, 33, 1162.

(52) Boudreault, P.-L. T.; Najari, A.; Leclerc, M. Chem. Mater. 2011, 23, 456.

(53) Kularatne, R. S.; Magurudeniya, H. D.; Sista, P.; Biewer, M. C.; Stefan, M. C. J.
Polym. Sci., Part A: Polym. Chem. 2013, 51, 743.

(54) Lin, Y.; Li, Y.; Zhan, X. Chem. Soc. Rev. 2012, 41, 4245.

(55) Mishra, A.; Bäuerle, P. Angew. Chem. Int. Ed. 2012, 51, 2020.
115
(56) Thompson, B. C.; Khlyabich, P. P.; Burkhart, B.; Aviles, A. E.; Rudenko, A.;
Shultz, G. V.; Ng, C. F.; Mangubat, L. B. Green 2011, 1, 29.

(57) Kippelen, B.; Brédas, J.-L. Energy Environ. Sci. 2009, 2, 251.

(58) Brédas, J.-L.; Norton, J. E.; Cornil, J.; Coropceanu, V. Acc. Chem. Res. 2009, 42,
1691.

(59) Deibel, C.; Dyakonov, V.; Brabec, C. J. IEEE J. Sel. Top. Quantum Electron.
2010, 16, 1517.

(60) Deibel, C.; Dyakonov, V. Rep. Prog. Phys. 2010, 73, 096401.

(61) Clarke, T. M.; Durrant, J. R. Chem. Rev. 2010, 110, 6736.

(62) Deibel, C.; Strobel, T.; Dyakonov, V. Adv. Mater. 2010, 22, 4097.

(63) Grossiord, N.; Kroon, J. M.; Andriessen, R.; Blom, P. W. M. Org. Electron. 2012,
13, 432.

(64) Lee, J. U.; Jung, J. W.; Jo, J. W.; Jo, W. H. J. Mater. Chem. 2012, 22, 24265.

(65) Jørgensen, M.; Norrman, K.; Gevorgyan, S. A.; Tromholt, T.; Andreasen, B.;
Krebs, F. C. Adv. Mater. 2012, 24, 580.

(66) Jørgensen, M.; Norrman, K.; Krebs, F. C. Sol. Energy Mater. Sol. Cells 2008, 92,
686.

(67) Krebs, F. C.; Tromholt, T.; Jørgensen, M. Nanoscale 2010, 2, 873.

(68) Søndergaard, R. R.; Hösel, M.; Krebs, F. C. J. Polym. Sci., Part B: Polym. Phys.
2013, 51, 16.

(69) Blom, P. W. M.; Mihailetchi, V. D.; Koster, L. J. A.; Markov, D. E. Adv. Mater.
2007, 19, 1551.

(70) Schlenker, C. W.; Thompson, M. E. Chem. Commun. 2011, 47, 3702.

(71) Van Bavel, S.; Veenstra, S.; Loos, J. Macromol. Rapid Commun. 2010, 31, 1835.

(72) Dang, M. T.; Hirsch, L.; Wantz, G.; Wuest, J. D. Chem. Rev. 2013, 113, 3734.

(73) Brabec, C. J.; Heeney, M.; McCulloch, I.; Nelson, J. Chem. Soc. Rev. 2011, 40,
1185.
116
(74) Brady, M. A.; Su, G. M.; Chabinyc, M. L. Soft Matter 2011, 7, 11065.

(75) Krebs, F. C. Sol. Energy Mater. Sol. Cells 2009, 93, 394.

(76) Søndergaard, R.; Hösel, M.; Angmo, D.; Larsen-Olsen, T. T.; Krebs, F. C. Mater.
Today 2012, 15, 36.

(77) Peters, C. H.; Sachs-Quintana, I. T.; Kastrop, J. P.; Beaupré, S.; Leclerc, M.;
McGehee, M. D. Adv. Energy Mater. 2011, 1, 491.

(78) Yue, D.; Khatav, P.; You, F.; Darling, S. B. Energy Environ. Sci. 2012, 5, 9163.

(79) Lizin, S.; Van Passel, S.; De Schepper, E.; Vranken, L. Sol. Energy Mater. Sol.
Cells 2012, 103, 1.

(80) Lunt, R. R.; Osedach, T. P.; Brown, P. R.; Rowehl, J. A.; Bulović, V. Adv. Mater.
2011, 23, 5712.

(81) Espinosa, N.; Hösel, M.; Angmo, D.; Krebs, F. C. Energy Environ. Sci. 2012, 5,
5117.

(82) Scharber, M. C.; Sariciftci, N. S. Prog. Polym. Sci. 2013, DOI:
10.1016/j.progpolymsci.2013.05.001.

(83) Janssen, R. A. J.; Nelson, J. Adv. Mater. 2013, 25, 1847.

(84) Kotlarski, J. D.; Blom, P. W. M. Appl. Phys. Lett. 2011, 98, 053301.

(85) Kotlarski, J. D.; Blom, P. W. M. Appl. Phys. Lett. 2012, 100, 013306.

(86) Scharber, M. C.; Mühlbacher, D.; Koppe, M.; Denk, P.; Waldauf, C.; Heeger, A.
J.; Brabec, C. J. Adv. Mater. 2006, 18, 789.

(87) Veldman, D.; Meskers, S. C. J.; Janssen, R. A. J. Adv. Funct. Mater. 2009, 19,
1939.

(88) Nayak, P. K.; Garcia-Belmonte, G.; Kahn, A.; Bisquert, J.; Cahen, D. Energy
Environ. Sci. 2012, 5, 6022.

(89) Koster, L. J. A.; Shaheen, S. E.; Hummelen, J. C. Adv. Energy Mater. 2012, 2,
1246.

(90) Gruber, M.; Wagner, J.; Klein, K.; Hörmann, U.; Opitz, A.; Stutzmann, M.;
Brütting, W. Adv. Energy Mater. 2012, 2, 1100.
117
(91) Giebink, N. C.; Wiederrecht, G. P.; Wasielewski, M. R.; Forrest, S. R. Phys. Rev.
B 2011, 83, 195326.

(92) Camaioni, N.; Po, R. J. Phys. Chem. Lett. 2013, 4, 1821.

(93) Qi, B.; Wang, J. J. Mater. Chem. 2012, 22, 24315.

(94) Zhang, W.; Wang, Y.-W.; Hu, R.; Fu, L.-M.; Ai, X.-C.; Zhang, J.-P.; Hou, J.-H. J.
Phys. Chem. C 2013, 117, 735.

(95) Dennler, G.; Scharber, M. C.; Ameri, T.; Denk, P.; Forberich, K.; Waldauf, C.;
Brabec, C. J. Adv. Mater. 2008, 20, 579.

(96) Siddiki, M. K.; Li, J.; Galipeau, D.; Qiao, Q. Energy Environ. Sci. 2010, 3, 867.

(97) You, J.; Dou, L.; Hong, Z.; Li, G.; Yang, Y. Prog. Polym. Sci. 2013, DOI:
10.1016/j.progpolymsci.2013.04.005.

(98) Hadipour, A.; de Boer, B.; Blom, P. W. M. Org. Electron. 2008, 9, 617.

(99) Hadipour, A.; de Boer, B.; Blom, P. W. M. Adv. Funct. Mater. 2008, 18, 169.

(100) Sista, S.; Hong, Z.; Chen, L.-M.; Yang, Y. Energy Environ. Sci. 2011, 4, 1606.

(101) Yuan, Y.; Huang, J.; Li, G. Green 2011, 1, 65.

(102) Smith, M. B.; Michl, J. Chem. Rev. 2010, 110, 6891.

(103) Hanna, M. C.; Nozik, A. J. J. Appl. Phys. 2006, 100, 074510.

(104) Tayebjee, M. J. Y.; Gray-Weale, A. A.; Schmidt, T. W. J. Phys. Chem. Lett. 2012,
3, 2749.

(105) Lenes, M.; Kooistra, F. B.; Hummelen, J. C.; Van Severen, I.; Lutsen, L.;
Vanderzande, D.; Cleij, T. J.; Blom, P. W. M. J. Appl. Phys. 2008, 104, 114517.

(106) Khlyabich, P. P.; Burkhart, B.; Thompson, B. C. J. Am. Chem. Soc. 2011, 133,
14534.

(107) Khlyabich, P. P.; Burkhart, B.; Thompson, B. C. J. Am. Chem. Soc. 2012, 134,
9074.

(108) Yang, L.; Zhou, H.; Price, S. C.; You, W. J. Am. Chem. Soc. 2012, 134, 5432.

118
(109) Koppe, M.; Egelhaaf, H.-J.; Clodic, E.; Morana, M.; Lüer, L.; Troeger, A.; Sgobba,
V.; Guldi, D. M.; Ameri, T.; Brabec, C. J. Adv. Energy Mater. 2013, DOI:
10.1002/aenm.201201076.

(110) Brabec, C. J. Organic photovoltaics: concepts and realization; Springer series in
materials science; Springer: New York, 2003.

(111) Kumar, P.; Chand, S. Prog. Photovolt: Res. Appl. 2012, 20, 377.

(112) Chen, Y.-C.; Hsu, C.-Y.; Lin, R. Y.-Y.; Ho, K.-C.; Lin, J. T. ChemSusChem 2013,
6, 20.

(113) Ameri, T.; Khoram, P.; Min, J.; Brabec, C. J. Adv. Mater. 2013, DOI:
10.1002/adma.201300623.

(114) Chen, L.; Yao, K.; Chen, Y. J. Mater. Chem. 2012, 22, 18768.

(115) Zhang, Z.-G.; Li, H.; Qi, Z.; Jin, Z.; Liu, G.; Hou, J.; Li, Y.; Wang, J. Appl. Phys.
Lett. 2013, 102, 143902.

(116) Liu, H.-W.; Chang, D.-Y.; Chiu, W.-Y.; Rwei, S.-P.; Wang, L. J. Mater. Chem.
2012, 22, 15586.

(117) Tai, Q.; Li, J.; Liu, Z.; Sun, Z.; Zhao, X.; Yan, F. J. Mater. Chem. 2011, 21, 6848.

(118) Li, N.; Machui, F.; Waller, D.; Koppe, M.; Brabec, C. J. Sol. Energy Mater. Sol.
Cells 2011, 95, 3465.

(119) Ameri, T.; Min, J.; Li, N.; Machui, F.; Baran, D.; Forster, M.; Schottler, K. J.;
Dolfen, D.; Scherf, U.; Brabec, C. J. Adv. Energy Mater. 2012, 2, 1198.

(120) Ameri, T.; Heumüller, T.; Min, J.; Li, N.; Matt, G.; Scherf, U.; Brabec, C. J.
Energy Environ. Sci. 2013, 6, 1796.

(121) Chen, M.-C.; Liaw, D.-J.; Chen, W.-H.; Huang, Y.-C.; Sharma, J.; Tai, Y. Appl.
Phys. Lett. 2011, 99, 223305.

(122) Adam, G.; Pivrikas, A.; Ramil, A. M.; Tadesse, S.; Yohannes, T.; Sariciftci, N. S.;
Egbe, D. A. M. J. Mater. Chem. 2011, 21, 2594.

(123) Street, R. A.; Davies, D.; Khlyabich, P. P.; Burkhart, B.; Thompson, B. C. J. Am.
Chem. Soc. 2013, 135, 986.

119
(124) Koppe, M.; Egelhaaf, H.-J.; Dennler, G.; Scharber, M. C.; Brabec, C. J.;
Schilinsky, P.; Hoth, C. N. Adv. Funct. Mater. 2010, 20, 338.

(125) Winder, C.; Matt, G.; Hummelen, J. C.; Janssen, R. A. J.; Sariciftci, N. S.; Brabec,
C. J. Thin Solid Films 2002, 403-404, 373.

(126) Kaake, L.; Dang, X.-D.; Leong, W. L.; Zhang, Y.; Heeger, A.; Nguyen, T.-Q. Adv.
Mater. 2013, 25, 1706.

(127) Ito, S.; Ohkita, H.; Benten, H.; Honda, S. AMBIO 2012, 41, 132.

(128) Jarzab, D.; Szendrei, K.; Yarema, M.; Pichler, S.; Heiss, W.; Loi, M. A. Adv.
Funct. Mater. 2011, 21, 1988.

(129) Hesse, H. C.; Weickert, J.; Hundschell, C.; Feng, X.; Müllen, K.; Nickel, B.;
Mozer, A. J.; Schmidt-Mende, L. Adv. Energy Mater. 2011, 1, 861.

(130) Honda, S.; Yokoya, S.; Ohkita, H.; Benten, H.; Ito, S. J. Phys. Chem. C 2011, 115,
11306.

(131) Bakulin, A. A.; Martyanov, D.; Paraschuk, D. Y.; Loosdrecht, P. H. M. van;
Pshenichnikov, M. S. Chem. Phys. Lett. 2009, 482, 99.

(132) Bakulin, A. A.; Zapunidy, S. A.; Pshenichnikov, M. S.; van Loosdrecht, P. H. M.;
Paraschuk, D. Y. Phys. Chem. Chem. Phys. 2009, 11, 7324.

(133) Johansson, E. M. J.; Yartsev, A.; Rensmo, H.; Sundström, V. J. Phys. Chem. C
2009, 113, 3014.

(134) Huang, J.-S.; Goh, T.; Li, X.; Sfeir, M. Y.; Bielinski, E. A.; Tomasulo, S.; Lee, M.
L.; Hazari, N.; Taylor, A. D. Nat. Photonics 2013, DOI: 10.1038/nphoton.2013.82.

(135) Thompson, B. C.; Kim, Y.-G.; Reynolds, J. R. Macromolecules 2005, 38, 5359.

(136) Hu, Z.; Tang, S.; Ahlvers, A.; Khondaker, S. I.; Gesquiere, A. J. Appl. Phys. Lett.
2012, 101, 053308.

(137) Cha, H.; Chung, D. S.; Bae, S. Y.; Lee, M.-J.; An, T. K.; Hwang, J.; Kim, K. H.;
Kim, Y.-H.; Choi, D. H.; Park, C. E. Adv. Funct. Mater. 2013, 23, 1556.

(138) Cho, Y. J.; Lee, J. Y.; Chin, B. D.; Forrest, S. R. Org. Electron. 2013, 14, 1081.

(139) Sharma, G. D.; Singh, S. P.; Roy, M. S.; Mikroyannidis, J. A. Org. Electron. 2012,
13, 1756.
120
(140) Suresh, P.; Balraju, P.; Sharma, G. D.; Mikroyannidis, J. A.; Stylianakis, M. M.
ACS Appl. Mater. Interfaces 2009, 1, 1370.

(141) Peet, J.; Tamayo, A. B.; Dang, X.-D.; Seo, J. H.; Nguyen, T.-Q. Appl. Phys. Lett.
2008, 93, 163306.

(142) Honda, S.; Ohkita, H.; Benten, H.; Ito, S. Adv. Energy Mater. 2011, 1, 588.

(143) Honda, S.; Nogami, T.; Ohkita, H.; Benten, H.; Ito, S. ACS Appl. Mater. Interfaces
2009, 1, 804.

(144) Lu, K.; Fang, J.; Zhu, X.; Yan, H.; Li, D.; Di, C.; Yang, Y.; Wei, Z. New J. Chem.
2013, 37, 1728.

(145) Kästner, C.; Rathgeber, S.; Egbe, D. A. M.; Hoppe, H. J. Mater. Chem. A 2013, 1,
3961.

(146) Kang, H.; Kim, K.-H.; Kang, T. E.; Cho, C.-H.; Park, S.; Yoon, S. C.; Kim, B. J.
ACS Appl. Mater. Interfaces 2013, 5, 4401.

(147) Mihailetchi, V.; Wildeman, J.; Blom, P. Phys. Rev. Lett. 2005, 94, 126602.

(148) Robeson, L. M. Polymer blends: a comprehensive review; Carl Hanser Verlag:
Munich, 2007.

(149) Wu, Y.-C.; Chao, Y.-H.; Wang, C.-L.; Wu, C.-T.; Hsu, C.-S.; Zeng, Y.-L.; Lin, C.-
Y. J. Polym. Sci., Part A: Polym. Chem. 2012, 50, 5032.

(150) Singh, T. B.; Chen, X.; Wong, W. W. H.; Ehlig, T.; Kemppinen, P.; Chen, M.;
Watkins, S. E.; Winzenberg, K. N.; Holdcroft, S.; Jones, D. J.; Holmes, A. B. Appl.
Phys. A: Mater. Sci. Process. 2012, 108, 515.

(151) Gao, D.; Hollinger, J.; Seferos, D. S. ACS Nano 2012, 6, 7114.

(152) Kozycz, L. M.; Gao, D.; Hollinger, J.; Seferos, D. S. Macromolecules 2012, 45,
5823.

(153) Park, S. J.; Cho, J. M.; Byun, W.-B.; Lee, J.-C.; Shin, W. S.; Kang, I.-N.; Moon,
S.-J.; Lee, S. K. J. Polym. Sci., Part A: Polym. Chem. 2011, 49, 4416.

(154) Mandoc, M. M.; Kooistra, F. B.; Hummelen, J. C.; de Boer, B.; Blom, P. W. M.
Appl. Phys. Lett. 2007, 91, 263505.

(155) Lim, E.; Lee, S.; Lee, K. K. Chem. Commun. 2011, 47, 914.
121
(156) Sharma, S. S.; Sharma, G. D.; Mikroyannidis, J. A. Sol. Energy Mater. Sol. Cells
2011, 95, 1219.

(157) Honda, S.; Ohkita, H.; Benten, H.; Ito, S. Chem. Commun. 2010, 46, 6596.

(158) Huang, J.-H.; Velusamy, M.; Ho, K.-C.; Lin, J.-T.; Chu, C.-W. J. Mater. Chem.
2010, 20, 2820.

(159) Cheng, Y.-J.; Hsieh, C.-H.; Li, P.-J.; Hsu, C.-S. Adv. Funct. Mater. 2011, 21, 1723.

(160) Koppolu, V. R.; Gupta, M. C.; Bagienski, W.; Shen, Y.; Shu, C.; Gibson, H. W.;
Dorn, H. C. In Conf. Rec. IEEE Photovoltaic Spec. Conf., 34th; IEEE:
Philadelphia, PA, USA, 2009; p. 001252.

(161) Chen, M. C.; Liaw, D. J.; Huang, Y. C.; Wu, H. Y.; Tai, Y. Sol. Energy Mater. Sol.
Cells 2011, 95, 2621.

(162) Chen, M.-C.; Kar, S.; Liaw, D.-J.; Chen, W.-H.; Huang, Y.-C.; Tai, Y. Org.
Electron. 2012, 13, 2702.

(163) Lee, J.; Yun, M. H.; Kim, J.; Kim, J. Y.; Yang, C. Macromol. Rapid Commun.
2012, 33, 140.

(164) Choi, J. W.; Kulshreshtha, C.; Kennedy, G. P.; Kwon, J. H.; Jung, S.-H.; Chae, M.
Sol. Energy Mater. Sol. Cells 2011, 95, 2069.

(165) Shin, M.; Kim, H.; Nam, S.; Park, J.; Kim, Y. Energy Environ. Sci. 2010, 3, 1538.

(166) Hori, T.; Masuda, T.; Fukuoka, N.; Hayashi, T.; Miyake, Y.; Kamikado, T.;
Yoshida, H.; Fujii, A.; Shimizu, Y.; Ozaki, M. Org. Electron. 2012, 13, 335.

(167) Lou, Y.; Wang, Z.; Naka, S.; Okada, H. Appl. Phys. Lett. 2011, 99, 033305.

(168) Veysel Tunc, A.; De Sio, A.; Riedel, D.; Deschler, F.; Da Como, E.; Parisi, J.; von
Hauff, E. Org. Electron. 2012, 13, 290.

(169) Yilmaz Canli, N.; Günes, S.; Pivrikas, A.; Fuchsbauer, A.; Sinwel, D.; Sariciftci,
N. S.; Yasa, Ö.; Bilgin-Eran, B. Sol. Energy Mater. Sol. Cells 2010, 94, 1089.

(170) Jeong, S.; Woo, S.-H.; Lyu, H.-K.; Han, Y. S. Sol. Energy Mater. Sol. Cells 2011,
95, 1908.

(171) Vohra, V.; Higashimine, K.; Murakami, T.; Murata, H. Appl. Phys. Lett. 2012,
101, 173301.
122
(172) Feron, K.; Belcher, W.; Fell, C.; Dastoor, P. Int. J. Mol. Sci. 2012, 13, 17019.

(173) Chen, C.-H.; Cheng, Y.-J.; Dubosc, M.; Hsieh, C.-H.; Chu, C.-C.; Hsu, C.-S.
Chem. Asian J. 2010, 5, 2483.

(174) Chen, C.-H.; Hsieh, C.-H.; Dubosc, M.; Cheng, Y.-J.; Hsu, C.-S. Macromolecules
2010, 43, 697.

(175) Mikroyannidis, J. A.; Tsagkournos, D. V.; Balraju, P.; Sharma, G. D. J. Power
Sources 2011, 196, 2364.

(176) Kim, Y.; Shin, M.; Kim, H.; Ha, Y.; Ha, C.-S. J. Phys. D: Appl. Phys. 2008, 41,
225101.

(177) Jung, J. W.; Jo, J. W.; Jo, W. H. Adv. Mater. 2011, 23, 1782.

(178) Tress, W.; Merten, A.; Furno, M.; Hein, M.; Leo, K.; Riede, M. Adv. Energy
Mater. 2013, 3, 631.

(179) Kotlarski, J. D.; Moet, D. J. D.; Blom, P. W. M. J. Polym. Sci., Part B: Polym.
Phys. 2011, 49, 708.

(180) Qi, B.; Wang, J. Phys. Chem. Chem. Phys. 2013, 15, 8972.

(181) Khlyabich, P. P.; Burkhart, B.; Ng, C. F.; Thompson, B. C. Macromolecules 2011,
44, 5079.

(182) Burkhart, B.; Khlyabich, P. P.; Thompson, B. C. Macromolecules 2012, 45, 3740.

(183) Chang, S.-Y.; Liao, H.-C.; Shao, Y.-T.; Sung, Y.-M.; Hsu, S.-H.; Ho, C.-C.; Su,
W.-F.; Chen, Y.-F. J. Mater. Chem. A 2013, 1, 2447.

(184) Yao, K.; Chen, L.; Chen, X.; Chen, Y. Chem. Mater. 2013, 25, 897.

(185) Jeltsch, K. F.; Schädel, M.; Bonekamp, J.-B.; Niyamakom, P.; Rauscher, F.;
Lademann, H. W. A.; Dumsch, I.; Allard, S.; Scherf, U.; Meerholz, K. Adv. Funct.
Mater. 2012, 22, 397.

(186) Graham, K. R.; Stalder, R.; Wieruszewski, P. M.; Patel, D. G.; Salazar, D. H.;
Reynolds, J. R. ACS Appl. Mater. Interfaces 2013, 5, 63.

(187) Andrew, T. L.; Bulović, V. ACS Nano 2012, 6, 4671.

123
(188) Kubo, Y.; Watanabe, K.; Nishiyabu, R.; Hata, R.; Murakami, A.; Shoda, T.; Ota,
H. Org. Lett. 2011, 13, 4574.

(189) Chang, S.-W.; Waters, H.; Kettle, J.; Horie, M. Org. Electron. 2012, 13, 2967.

(190) Min, J.; Ameri, T.; Gresser, R.; Lorenz-Rothe, M.; Baran, D.; Troeger, A.; Sgobba,
V.; Leo, K.; Riede, M.; Guldi, D. M.; Brabec, C. J. ACS Appl. Mater. Interfaces
2013, DOI: 10.1021/am400923b.

(191) Li, H.; Zhang, Z.-G.; Li, Y.; Wang, J. Appl. Phys. Lett. 2012, 101, 163302.

(192) Andersson, L. M.; Hsu, Y.-T.; Vandewal, K.; Sieval, A. B.; Andersson, M. R.;
Inganäs, O. Org. Electron. 2012, 13, 2856.

(193) Susarova, D. K.; Troshin, P. A.; Moskvin, Y. L.; Babenko, S. D.; Razumov, V. F.
Thin Solid Films 2011, 519, 4132.

(194) Lee, J. M.; Kwon, B.-H.; Park, H. I.; Kim, H.; Kim, M. G.; Park, J. S.; Kim, E. S.;
Yoo, S.; Jeon, D. Y.; Kim, S. O. Adv. Mater. 2013, 25, 2011.

(195) Yang, L.; Yan, L.; You, W. J. Phys. Chem. Lett. 2013, 4, 1802.

(196) Guerrero, A.; Ripolles-Sanchis, T.; Boix, P. P.; Garcia-Belmonte, G. Org.
Electron. 2012, 13, 2326.

(197) Duck, B. C.; Vaughan, B.; Cooling, N.; Zhou, X.; Holdsworth, J. L.; Wen, L. L.;
Rasmussen, S. C.; Dastoor, P. C.; Belcher, W. J. Sol. Energy Mater. Sol. Cells
2013, 114, 65.

(198) Rockett, A. The materials science of semiconductors; Springer: New York, 2008.

(199) Wu, J.-C.; Zheng, J.; Zacherl, C. L.; Wu, P.; Liu, Z.-K.; Xu, R. J. Phys. Chem. C
2011, 115, 19741.

(200) Moses, P. G.; Van de Walle, C. G. Appl. Phys. Lett. 2010, 96, 021908.

(201) Street, R. A.; Song, K. W.; Northrup, J. E.; Cowan, S. Phys. Rev. B 2011, 83,
165207.

(202) Vandewal, K.; Tvingstedt, K.; Gadisa, A.; Inganäs, O.; Manca, J. V. Nat. Mater.
2009, 8, 904.

(203) Street, R. A.; Schoendorf, M.; Roy, A.; Lee, J. H. Phys. Rev. B 2010, 81, 205307.

124
(204) Vandewal, K.; Tvingstedt, K.; Gadisa, A.; Inganäs, O.; Manca, J. V. Phys. Rev. B
2010, 81, 125204.

(205) Brazovskii, S.; Kirova, N. Chem. Soc. Rev. 2010, 39, 2453.

(206) Scholes, G. D. ACS Nano 2008, 2, 523.

(207) Kaake, L. G.; Jasieniak, J. J.; Bakus, R. C.; Welch, G. C.; Moses, D.; Bazan, G. C.;
Heeger, A. J. J. Am. Chem. Soc. 2012, 134, 19828.

(208) Bakulin, A. A.; Rao, A.; Pavelyev, V. G.; van Loosdrecht, P. H. M.;
Pshenichnikov, M. S.; Niedzialek, D.; Cornil, J.; Beljonne, D.; Friend, R. H.
Science 2012, 335, 1340.

(209) Wang, C.; Zhuang, L.-Q.; Chen, R.-A.; Li, S.; George, T. F. J. Phys. Chem. B
2013, 117, 3258.

(210) Grancini, G.; Maiuri, M.; Fazzi, D.; Petrozza, A.; Egelhaaf, H.-J.; Brida, D.;
Cerullo, G.; Lanzani, G. Nat. Mater. 2012, 12, 29.

(211) Jailaubekov, A. E.; Willard, A. P.; Tritsch, J. R.; Chan, W.-L.; Sai, N.; Gearba, R.;
Kaake, L. G.; Williams, K. J.; Leung, K.; Rossky, P. J.; Zhu, X.-Y. Nat. Mater.
2012, 12, 66.

(212) Imamura, S. J. Photon. Energy 2011, 1, 011119.

(213) Ma, W.; Kim, J. Y.; Lee, K.; Heeger, A. J. Macromol. Rapid Commun. 2007, 28,
1776.

(214) Campoy-Quiles, M.; Kanai, Y.; El-Basaty, A.; Sakai, H.; Murata, H. Org.
Electron. 2009, 10, 1120.

(215) Richards, J. J.; Rice, A. H.; Nelson, R. D.; Kim, F. S.; Jenekhe, S. A.; Luscombe,
C. K.; Pozzo, D. C. Adv. Funct. Mater. 2013, 23, 514.

(216) Di Nuzzo, D.; Wetzelaer, G.-J. A. H.; Bouwer, R. K. M.; Gevaerts, V. S.; Meskers,
S. C. J.; Hummelen, J. C.; Blom, P. W. M.; Janssen, R. A. J. Adv. Energy Mater.
2013, 3, 85.

(217) Faist, M. A.; Kirchartz, T.; Gong, W.; Ashraf, R. S.; McCulloch, I.; de Mello, J.
C.; Ekins-Daukes, N. J.; Bradley, D. D. C.; Nelson, J. J. Am. Chem. Soc. 2012,
134, 685.

125
(218) Machui, F.; Rathgeber, S.; Li, N.; Ameri, T.; Brabec, C. J. J. Mater. Chem. 2012,
22, 15570.

(219) Chen, J.; Yu, X.; Hong, K.; Messman, J. M.; Pickel, D. L.; Xiao, K.; Dadmun, M.
D.; Mays, J. W.; Rondinone, A. J.; Sumpter, B. G.; Kilbey II, S. M. J. Mater.
Chem. 2012, 22, 13013.

(220) Machui, F.; Abbott, S.; Waller, D.; Koppe, M.; Brabec, C. J. Macromol. Chem.
Phys. 2011, 212, 2159.

(221) Lobez, J. M.; Andrew, T. L.; Bulović, V.; Swager, T. M. ACS Nano 2012, 6, 3044.

(222) Liu, J.; Han, Y. Polym. Bull. 2012, 68, 2145.

(223) Kim, C. S.; Tinker, L. L.; DiSalle, B. F.; Gomez, E. D.; Lee, S.; Bernhard, S.; Loo,
Y.-L. Adv. Mater. 2009, 21, 3110.

(224) Cowan, S. R.; Leong, W. L.; Banerji, N.; Dennler, G.; Heeger, A. J. Adv. Funct.
Mater. 2011, 21, 3083.

(225) Leong, W. L.; Cowan, S. R.; Heeger, A. J. Adv. Energy Mater. 2011, 1, 517.

(226) Street, R. A.; Krakaris, A.; Cowan, S. R. Adv. Funct. Mater. 2012, 22, 4608.

126
Chapter 3

Efficient Ternary Blend Bulk Heterojunction Solar Cells with Tunable Open-
Circuit Voltage

3.1 Introduction

In order to explore the potential of ternary blend bulk heterojunction (BHJ)
photovoltaics as a general platform for increasing the attainable performance of organic
solar cells, a model system based on poly(3-hexylthiophene) (P3HT) as the donor and
two soluble fullerene acceptors, phenyl-C
61
-butyric acid methyl ester (PC
61
BM) and
indene-C
60
bisadduct (ICBA), was examined. In all solar cells, the overall ratio of
polymer to fullerene was maintained at 1:1, while varying the composition of the
fullerene component (PC
61
BM:ICBA ratio). Photovoltaic devices showed high short-
circuit current densities (J
sc
) and fill factors (FF) (above 0.57) at all fullerene ratios, while
the open-circuit voltage (V
oc
) was found to vary from 0.61 to 0.84 V as the fraction of
ICBA was increased. These results indicate that the V
oc
in ternary blend BHJ solar cells is
not limited to the smallest V
oc
of the corresponding binary blend solar cells, but can be
varied between the extreme V
oc
values without significant effect on the J
sc
or the FF. By
extension, this result suggests that ternary blends provide a potentially effective route
toward maximizing the attainable J
sc
× V
oc
product (which is directly proportional to the
solar cell efficiency) in BHJ solar cells and that with judicious selection of donor and
127
acceptor components, solar cells with efficiencies exceeding the theoretical limits for
binary blend solar cells could be possible without sacrificing the simplicity of a single
active layer processing step.

3.2 Results and Discussion

Bulk heterojunction (BHJ) solar cells based on a binary blend of a polymeric
donor and a fullerene acceptor have seen rapid improvements in efficiency in recent
years, from 2.5%
1
to ~8%.
2–6
However, the ultimate efficiency of such solar cells appears
to be limited to ~10 – 12%.
7–9
Ternary blend solar cells based on two donor components
and one acceptor component (or one donor and two acceptors) have received far less
attention, but have been recognized as a potential route to increase the absorption breadth
of a solar cell and consequently the short-circuit current density (J
sc
).
10–16
Despite this
potential advantage, it has been proposed that the open-circuit voltage (V
oc
) of ternary
blend solar cells would necessarily be pinned to the smaller V
oc
of the corresponding
binary blends of the constituent components, thus limiting the perceived impact of this
device platform.
14
Here, using a model three-component system (Figure 3.1), we
demonstrate for the first time that the V
oc
of ternary blend BHJ solar cells is composition
dependent and can be tuned across the full range defined by the corresponding limiting
binary blends without negatively impacting the fill factor (FF) or the J
sc
of the solar cells.
The consequence is the possibility that with judicious choice of components, the
attainable product of J
sc
× V
oc
(and by extension the efficiency, η = (J
sc
× V
oc
× FF) / P
in
,
128
where P
in
is the intensity of the incident light) in a single-layer ternary blend solar cell
could be higher than is achievable with a standard binary blend solar cell.

Figure 3.1. Structures and corresponding HOMO and LUMO energy levels of P3HT,
ICBA and PC
61
BM.

In a binary blend BHJ polymer-fullerene solar cell, the J
sc
is proportional to the
product of absorption breadth and absorption intensity of the active layer, which is
generally limited by the band gap (E
g
) of the donor polymer.
17
In principle, the smaller
the E
g
, the broader the wavelength range of light absorption and consequently the higher
the J
sc
. Conversely, the V
oc
cannot exceed the energetic difference between the donor
HOMO (HOMO
D
) and the acceptor LUMO (LUMO
A
).
7,18
Taking into account that an
energetic driving force for charge transfer must exist (approximated by the LUMO
D
–
LUMO
A
offset
19
), a high J
sc
is favored by a shallow HOMO
D
and a high V
oc
is favored by
a deep HOMO
D
. Ultimate optimization is found through adjusting the frontier orbital
energies of the donor and acceptor components to balance the opposing quantities of J
sc

129
and V
oc
and to target a maximum J
sc
× V
oc
product rather than targeting a maximum
attainable value for either J
sc
or V
oc
. As a consequence, the ultimate efficiency of binary
polymer-fullerene BHJ solar cells is limited to 10 – 12%.
As a route toward higher efficiency, tandem solar cells are an increasingly
explored alternative to simple BHJ solar cells where two (or more) sub-cells absorbing
light in different regions of the solar spectrum are connected either in series or
parallel.
20,21
The ultimate achievable efficiency of a tandem cell with two absorbing
layers is predicted to be around 14 – 15%.
8,22
In the case of serial connection of sub-cells,
the V
oc
of the tandem device can approach the sum of the V
oc
values of the individual sub-
cells
23
while the J
sc
can at best approach the highest J
sc
of the corresponding sub-
cells.
20,23,24
On the other hand, with parallel connection of the sub-cells the J
sc
approaches
the sum of individual sub-cells,
20
but the V
oc
is limited to the minimum V
oc
of the
individual sub-cells.
20
As such, it is clear that tandem cells also do not allow independent
and concurrent optimization of both J
sc
and V
oc
beyond that of the individual sub-cells.
Another drawback of tandem solar cells is the increase of complexity in the cell design
and fabrication,
20,23
which is in the contrast to the attractive simplicity of the single step
solution processing of the active layer in BHJ solar cells.
7

Ternary blend BHJ solar cells offer a distinct platform and an alternative approach
to increase the attainable J
sc
× V
oc
product, while retaining the simplicity of a single active
layer processing step. There is a growing body of literature describing a number of
variants of the ternary blend system based on either two polymer donors and a fullerene
acceptor,
10,12–16
one polymer donor and two acceptors,
25–29
a polymer donor, fullerene
130
acceptor and a small molecule/dye,
11,30–38
or a polymer donor with a nanoparticle and
fullerene acceptor.
39–43
In many cases an increase in breadth of the spectral response of
the ternary blends has been observed in reference to the corresponding limiting binary
blends, often leading to a larger J
sc
.
10–16
Conversely, the V
oc
is proposed to be pinned to
the smallest V
oc
of corresponding binary blends.
14,16,44
To this end, it is thought that a
limiting HOMO
D
– LUMO
A
interaction controls the V
oc
due to the fact that dominant
hole transport and collection occurs through the donor component with the highest-lying
HOMO (and analogously electron transport and collection through the lowest-lying
LUMO), independent of the origin of photocurrent generation.
14
However, in a limited
number of cases the V
oc
seems to be tunable in the three-component system, but at the
expense of marked and steady decrease in the FF as the amount of the third component
increases.
25,30,31
Nonetheless, these isolated observations of composition tunable V
oc

suggest that both J
sc
and V
oc
are composition dependent in ternary blend BHJ solar cells
and that neither value is necessarily limited to the lesser quantity of the corresponding
binary blend solar cells. In support of this, here we communicate for the first time an
example of a ternary blend BHJ solar cell in which the V
oc
is tunable across the full
composition range of the components, yielding high FF at all compositions.
The model ternary system of poly(3-hexylthiophene) (P3HT) as the donor and
phenyl-C
61
-butyric acid methyl ester (PC
61
BM) and indene-C
60
bisadduct (ICBA) as
acceptors used here (Figure 3.1) is chosen for several reasons. First is the similarity in
the chemical structures of the acceptors, which is envisioned to give good miscibility
between the acceptor components. Second is the excellent miscibility known in the
131
limiting binary polymer-fullerene blends, resulting in similar polymer-fullerene ratios
(close to 1:1) and processing conditions (post aluminum annealing at 150 ºC) necessary
for the optimal solar cell performance.
45,46
Moreover, high efficiencies with FF over 0.6
have been observed in both binary blend BHJ solar cells. Finally, the two limiting
polymer-fullerene binary blends give significantly different values of V
oc
at 0.6 V for
P3HT:PC
61
BM
45
and 0.84 V for P3HT:ICBA
46
due to the different positions of the
acceptor LUMOs,
7
as shown in Figure 3.1.

Table 3.1. Photovoltaic properties of P3HT:PC
61
BM:ICBA ternary blend BHJ solar cells
at different fullerene ratios
P3HT:PC
61
BM:ICBA J
sc
(mA/cm
2
) V
oc
(V)* FF η (%)
1:1:0
a
9.90 0.605 0.60 3.57
1:0.9:0.1
b
9.22 0.618 0.59 3.29
1:0.8:0.2
b
9.11 0.631 0.57 3.28
1:0.7:0.3
c
8.58 0.649 0.58 3.22
1:0.6:0.4
d
8.31 0.669 0.58 3.11
1:0.5:0.5
e
8.27 0.688 0.57 3.18
1:0.4:0.6
c
8.18 0.709 0.57 3.22
1:0.3:0.7
f
8.14 0.741 0.57 3.34
1:0.2:0.8
b
8.19 0.769 0.59 3.69
1:0.1:0.9
b
8.18 0.804 0.60 3.91
1:0:1
f
8.23 0.844 0.58 3.98
All devices were spin-coated from chlorobenzene (CB) and after aluminum deposition annealed at 150 °C
under N
2
for
a
60 min,
b
20 min,
c
40 min,
d
30 min,
e
50 min and
f
10 min. *Standard deviations of less than
0.005 were observed in all cases averaged over eight pixels.

Photovoltaic devices containing ternary blends in a conventional device
configuration ITO/PEDOT:PSS/P3HT:PC
61
BM:ICBA/Al were fabricated in air. In order
to compare the device parameters, all active layer thicknesses of the optimized devices
were kept between 95 – 105 nm. Additionally, the concentration of all polymer-fullerene
solutions was kept constant (10 mg/ml in P3HT) as was the annealing temperature (150
132
°C). Furthermore, in all cases the overall weight ratio between P3HT and the fullerene
component was maintained at 1:1. Table 3.1 lists the average values of J
sc
, V
oc
, FF and η
obtained under simulated AM 1.5G illumination (100 mW/cm
2
) as the ratio of PC
61
BM to
ICBA was varied.

Figure 3.2. Open-circuit voltage (V
oc
) of the ternary blend BHJ solar cells as a function
of the amount of ICBA in the blends.

Several significant observations can be made from the data in Table 3.1.
Importantly, as is also illustrated in Figure 3.2, as the amount of the ICBA component in
the ternary blend is increased, the V
oc
of the three-component solar cells shows a
continuous increase from the 0.605 V to 0.844 V. This establishes that in ternary blend
BHJ solar cells V
oc
is not necessarily pinned to the smallest V
oc
of the corresponding
binary blends. It is also seen in Table 3.1 that high FF values above 0.57 are observed for
all photovoltaic devices independent of the fullerene composition. This can be attributed
to a balanced and trap free charge transport through the bulk,
47
and favorable
133
morphology.
48
Transmission electron microscopy (TEM) (see Appendix A) shows very
similar, bicontinious blends, with nanometer-scale phase separation, independent of the
fullerene ratios. As such, charge separation and transport do not appear to be hindered in
the ternary blend solar cells.
In contrast to the V
oc
trend observed in Table 3.1, J
sc
is found to decrease with the
increase of the ICBA content. This observation was explained using external quantum
efficiency (EQE) measurements (see Appendix A) and comparing with the absorption
coefficients of the various blend films. The highest photocurrent response was observed
for P3HT:PC
61
BM blend, while introduction of the ICBA in the blends leads to the
gradual reduction in the photocurrent intensity with the minimum value reached for
P3HT:ICBA solar cells. In order to further investigate the origin of the decrease in EQE
at high ICBA contents, optical properties of annealed P3HT:PC
61
BM:ICBA blends at
various ratios of fullerenes in thin films spin-coated from chlorobenzene (CB) were
studied using UV-vis spectroscopy as shown in Figure 3.3. The introduction of the ICBA
results in the observed decrease in the absorption coefficients of the thin films as the
amount of ICBA in the blends is increased. The decrease in the intensities was explained
based on the absorption coefficients and profiles of PC
61
BM and ICBA (see Appendix
A). For ICBA the absorption strength in the visible is significantly less in comparison to
PC
61
BM. Considering that all the devices from Table 3.1 are of the same active layer
thickness, the introduction of ICBA into the films decreases the number of photons
absorbed, thus leading to a decrease in the EQE and by extension the J
sc
.

134

Figure 3.3. UV-vis absorption spectra of thin films spin-coated from chlorobenzene (CB)
and annealed at 150 °C under N
2
for 20 min with P3HT:PC
61
BM:ICBA ratios: (i) is 1:1:0
(red line), (ii) is 1:0.8:0.2 (blue line), (iii) is 1:0.5:0.5 (green line), (iv) is 1:0.2:0.8 (black
line) and (v) is 1:0:1 (purple line).

To verify the possibility of achieving higher J
sc
in the case of high contents of
ICBA, P3HT:PC
61
BM:ICBA ternary blend solar cells at 1:0.5:0.5 and 1:0:1 ratios were
optimized (see Appendix A) to film thicknesses of 137 nm and 174 nm, respectively.
These optimized photovoltaic devices at 1:0.5:0.5 and 1:0:1 showed improved J
sc
, FF and
η of 9.82 mA/cm
2
, 0.59, 3.92% and 9.23 mA/cm
2
, 0.59, 4.55%, respectively, with
essentially no change in the V
oc
at 0.682 V and 0.839 V, relative to the devices reported
in Table 3.1. Thus, high J
sc
and FF are possible in the ternary blend solar cells
independent of the ratios of three components in the blends and without effect on the
compositional dependence of the V
oc
.
Another important feature in thin film absorption spectra (Figure 3.3) is the
presence of the strong vibronic feature around 600 nm for all P3HT:PC
61
BM:ICBA
blends. These shoulders are common for P3HT thin films
49
and generally are ascribed to
135
the interchain vibrational absorption induced by a high degree of ordering and strong
interchain interaction.
50
To study the effect of ternary blends on the degree of P3HT
crystallinity, grazing-incidence X-ray diffraction (GIXRD) was used (see Appendix A).
In all films the peaks corresponding to the interchain distance (100) for P3HT were
observed in the range of 16.4 – 16.7 Å and with similar intensities. As such, the ability to
obtain semi-crystallinity in P3HT is not hindered in the ternary blends.
Thinking beyond this model ternary system, which has a fixed spectral range for
both limiting binary blends, it is clear that the composition tunable V
oc
in ternary blends
offers the potential for higher efficiencies than are attainable in binary blends if a
simultaneous tuning of the J
sc
is also targeted. Consider a hypothetical case of two donor
polymers (D1 and D2) and a fullerene acceptor (A), where E
gD1
> E
gD2
and HOMO
D1
is
lower than HOMO
D2
. Consider that the binary blend of D1:A displays a V
oc
of 0.8 V, a J
sc

of 8.0 mA/cm
2
, and a FF of 0.6 for an efficiency of 3.8% and the binary blend of D2:A
shows a V
oc
of 0.5 V, a J
sc
of 14.0 mA/cm
2
, and a FF of 0.6 for an efficiency of 4.2%.
Combining D1, D2, and A in a ternary blend could lead to higher J
sc
for the ternary blend
than in either limiting binary blend due to a more uniform spectral coverage of absorption
(which has been demonstrated previously). Even a modest increase in J
sc
to 15.0 mA/cm
2

could couple with an intermediate V
oc
of 0.65 V to give an efficiency of 5.9% at FF = 0.6.
The ternary blend could thus give a higher efficiency than either limiting binary blend
due to the higher attainable J
sc
× V
oc
product. Based on the tunability of the V
oc

established here, it is proposed that given the judicious choice of components, ternary
blends with efficiencies exceeding the 10 – 12% maximum predicted for binary blends
136
could be achieved. Future work is focused on broadening the scope of investigated
ternary systems beyond the illustrative model system investigated in this work.

3.3 Conclusion

In summary, we have fabricated ternary blend BHJ solar cells containing P3HT as
the donor and two soluble fullerenes, PC
61
BM and ICBA, as acceptors. Devices were
tested at different acceptor ratios and showed uniformly high J
sc
and FF. Importantly, the
V
oc
of the three-component solar cells was tuned between the limiting V
oc
values of the
corresponding binary blend solar cells. Taken together with previous literature, it is now
established that both V
oc
and the spectral response (J
sc
) of ternary blend solar cells are
dependent on the composition. This suggests that J
sc
and V
oc
are not necessarily
constrained by the same factors and with the same limitations in ternary blends as they
are in binary blends. These results indicate that ternary blend BHJ solar cells are
promising candidates for the next generation of solution processable solar cells with the
potential to overcome predicted ultimate efficiencies for binary blend photovoltaic
devices with potentially higher attainable products of J
sc
× V
oc
through judicious
component selection and compositional control and without sacrificing the attractive
simplicity of processing in single active layer BHJ solar cells.

137
3.4 Chapter Bibliography

(1) Shaheen, S. E.; Brabec, C. J.; Sariciftci, N. S.; Padinger, F.; Fromherz, T.;
Hummelen, J. C. Appl. Phys. Lett. 2001, 78, 841.

(2) Chen, H.-Y.; Hou, J.; Zhang, S.; Liang, Y.; Yang, G.; Yang, Y.; Yu, L.; Wu, Y.;
Li, G. Nat. Photonics 2009, 3, 649.

(3) Liang, Y.; Xu, Z.; Xia, J.; Tsai, S.-T.; Wu, Y.; Li, G.; Ray, C.; Yu, L. Adv. Mater.
2010, 22, E135.

(4) Chu, T.-Y.; Lu, J.; Beaupré, S.; Zhang, Y.; Pouliot, J.-R.; Wakim, S.; Zhou, J.;
Leclerc, M.; Li, Z.; Ding, J.; Tao, Y. J. Am. Chem. Soc. 2011, 133, 4250.

(5) Amb, C. M.; Chen, S.; Graham, K. R.; Subbiah, J.; Small, C. E.; So, F.; Reynolds,
J. R. J. Am. Chem. Soc. 2011, 133, 10062.

(6) Green, M. A.; Emery, K.; Hishikawa, Y.; Warta, W. Prog. Photovolt: Res. Appl.
2011, 19, 84.

(7) Thompson, B. C.; Khlyabich, P. P.; Burkhart, B.; Aviles, A. E.; Rudenko, A.;
Shultz, G. V.; Ng, C. F.; Mangubat, L. B. Green 2011, 1, 29.

(8) Kotlarski, J. D.; Blom, P. W. M. Appl. Phys. Lett. 2011, 98, 053301.

(9) Scharber, M. C.; Mühlbacher, D.; Koppe, M.; Denk, P.; Waldauf, C.; Heeger, A.
J.; Brabec, C. J. Adv. Mater. 2006, 18, 789.

(10) Thompson, B. C.; Kim, Y.-G.; Reynolds, J. R. Macromolecules 2005, 38, 5359.

(11) Suresh, P.; Balraju, P.; Sharma, G. D.; Mikroyannidis, J. A.; Stylianakis, M. M.
ACS Appl. Mater. Interfaces 2009, 1, 1370.

(12) Chen, C.-H.; Hsieh, C.-H.; Dubosc, M.; Cheng, Y.-J.; Hsu, C.-S. Macromolecules
2010, 43, 697.

(13) Chen, C.-H.; Cheng, Y.-J.; Dubosc, M.; Hsieh, C.-H.; Chu, C.-C.; Hsu, C.-S.
Chem. Asian J. 2010, 5, 2483.

(14) Koppe, M.; Egelhaaf, H.-J.; Dennler, G.; Scharber, M. C.; Brabec, C. J.;
Schilinsky, P.; Hoth, C. N. Adv. Funct. Mater. 2010, 20, 338.

138
(15) Adam, G.; Pivrikas, A.; Ramil, A. M.; Tadesse, S.; Yohannes, T.; Sariciftci, N. S.;
Egbe, D. A. M. J. Mater. Chem. 2011, 21, 2594.

(16) Mikroyannidis, J. A.; Tsagkournos, D. V.; Balraju, P.; Sharma, G. D. J. Power
Sources 2011, 196, 2364.

(17) Bundgaard, E.; Krebs, F. Sol. Energy Mater. Sol. Cells 2007, 91, 954.

(18) Schlenker, C. W.; Thompson, M. E. Chem. Commun. 2011, 47, 3702.

(19) Clarke, T. M.; Durrant, J. R. Chem. Rev. 2010, 110, 6736.

(20) Siddiki, M. K.; Li, J.; Galipeau, D.; Qiao, Q. Energy Environ. Sci. 2010, 3, 867.

(21) Sista, S.; Hong, Z.; Chen, L.-M.; Yang, Y. Energy Environ. Sci. 2011, 4, 1606.

(22) Dennler, G.; Scharber, M. C.; Ameri, T.; Denk, P.; Forberich, K.; Waldauf, C.;
Brabec, C. J. Adv. Mater. 2008, 20, 579.

(23) Ameri, T.; Dennler, G.; Lungenschmied, C.; Brabec, C. J. Energy Environ. Sci.
2009, 2, 347.

(24) Hadipour, A.; De Boer, B.; Blom, P. W. M. Adv. Funct. Mater. 2008, 18, 169.

(25) Cheng, Y.-J.; Hsieh, C.-H.; Li, P.-J.; Hsu, C.-S. Adv. Funct. Mater. 2011, 21, 1723.

(26) Koppolu, V. R.; Gupta, M. C.; Bagienski, W.; Shen, Y.; Shu, C.; Gibson, H. W.;
Dorn, H. C. In Conf. Rec. IEEE Photovoltaic Spec. Conf., 34
th
2009, 001252.

(27) Wei, Q.; Nishizawa, T.; Tajima, K.; Hashimoto, K. Adv. Mater. 2008, 20, 2211.

(28) Lin Leong, W.; Cowan, S. R.; Heeger, A. J. Adv. Energy Mater. 2011, 1, 517.

(29) Mandoc, M. M.; Kooistra, F. B.; Hummelen, J. C.; De Boer, B.; Blom, P. W. M.
Appl. Phys. Lett. 2007, 91, 263505.

(30) Huang, J.-H.; Velusamy, M.; Ho, K.-C.; Lin, J.-T.; Chu, C.-W. J. Mater. Chem.
2010, 20, 2820.

(31) Peet, J.; Tamayo, A. B.; Dang, X.-D.; Seo, J. H.; Nguyen, T.-Q. Appl. Phys. Lett.
2008, 93, 163306.

(32) Sharma, S. S.; Sharma, G. D.; Mikroyannidis, J. A. Sol. Energy Mater. Sol. Cells
2011, 95, 1219.
139
(33) Cooling, N.; Burke, K. B.; Zhou, X.; Lind, S. J.; Gordon, K. C.; Jones, T. W.;
Dastoor, P. C.; Belcher, W. J. Sol. Energy Mater. Sol. Cells 2011, 95, 1767.

(34) Honda, S.; Ohkita, H.; Benten, H.; Ito, S. Chem. Commun. 2010, 46, 6596.

(35) Honda, S.; Ohkita, H.; Benten, H.; Ito, S. Adv. Energy Mater. 2011, 1, 588.

(36) Honda, S.; Yokoya, S.; Ohkita, H.; Benten, H.; Ito, S. J. Phys. Chem. C 2011, 115,
11306.

(37) Johansson, E. M. J.; Yartsev, A.; Rensmo, H.; Sundström, V. J. Phys. Chem. C
2009, 113, 3014.

(38) Bakulin, A. A.; Martyanov, D.; Paraschuk, D. Y.; Van Loosdrecht, P. H. M.;
Pshenichnikov, M. S. Chem. Phys. Lett. 2009, 482, 99.

(39) Khan, M. T.; Bhargav, R.; Kaur, A.; Dhawan, S. K.; Chand, S. Thin Solid Films
2010, 519, 1007.

(40) De Freitas, J. N.; Grova, I. R.; Akcelrud, L. C.; Arici, E.; Sariciftci, N. S.;
Nogueira, A. F. J. Mater. Chem. 2010, 20, 4845.

(41) Naidu, B. V. K.; Park, J. S.; Kim, S. C.; Park, S.-M.; Lee, E.-J.; Yoon, K.-J.; Joon
Lee, S.; Wook Lee, J.; Gal, Y.-S.; Jin, S.-H. Sol. Energy Mater. Sol. Cells 2008,
92, 397.

(42) Chang, M.-Y.; Chen, Y.-F.; Tsai, Y.-S.; Chi, K.-M. J. Electrochem. Soc. 2009,
156, B234.

(43) Wang, D. H.; Kim, D. Y.; Choi, K. W.; Seo, J. H.; Im, S. H.; Park, J. H.; Park, O.
O.; Heeger, A. J. Angew. Chem. 2011, 123, 5633.

(44) Kim, Y.; Shin, M.; Kim, H.; Ha, Y.; Ha, C.-S. J. Phys. D: Appl. Phys. 2008, 41,
225101.

(45) Ma, W.; Yang, C.; Gong, X.; Lee, K.; Heeger, A. J. Adv. Funct. Mater. 2005, 15,
1617.

(46) Zhao, G.; He, Y.; Li, Y. Adv. Mater. 2010, 22, 4355.

(47) Kotlarski, J. D.; Moet, D. J. D.; Blom, P. W. M. J. Polym. Sci. B Polym. Phys.
2011, 49, 708.

(48) Van Bavel, S.; Veenstra, S.; Loos, J. Macromol. Rapid Commun. 2010, 31, 1835.
140
(49) Kim, Y.; Cook, S.; Tuladhar, S. M.; Choulis, S. A.; Nelson, J.; Durrant, J. R.;
Bradley, D. D. C.; Giles, M.; McCulloch, I.; Ha, C.-S.; Ree, M. Nat. Mater. 2006,
5, 197.

(50) Gurau, M. C.; Delongchamp, D. M.; Vogel, B. M.; Lin, E. K.; Fischer, D. A.;
Sambasivan, S.; Richter, L. J. Langmuir 2007, 23, 834.

141
Chapter 4

Semi-Random and Random Copolymers

4.1 Efficient Solar Cells from Semi-Random P3HT Analogues Incorporating
Diketopyrrolopyrrole

4.1.1 Introduction

Three novel semi-random poly(3-hexylthiophene) (P3HT) based donor-acceptor
copolymers containing 5 – 15 % of the acceptor diketopyrrolopyrrole (DPP) were
synthesized by Stille copolymerization and their optical, electrochemical, charge
transport and photovoltaic properties were investigated. Poly(3-hexylthiophene-
thiophene-diketopyrrolopyrrole) polymers with various percentages of DPP, represented
as P3HTT-DPP-5%, P3HTT-DPP-10% and P3HTT-DPP-15% had high molecular
weights (M
n
17,500 – 24,500 g/mol) and broad, intense absorption spectra, covering the
spectral range from 350 nm up to 850 nm with absorption maxima at 685 and 703 nm for
P3HTT-DPP-10% and P3HTT-DPP-15%. The low content of acceptor units allowed
preservation of many important properties of P3HT, among which are semi-crystallinity
(as verified by GIXRD) and high hole mobilities (µ
h
= 1 – 2.3*10
-4
cm
2
/(V*s)).
Photovoltaic devices fabricated from P3HTT-DPP polymers blended with PC
61
BM
showed higher short-circuit current densities (J
sc
) of 9.57 – 13.87 mA/cm
2
compared to
142
P3HT (9.49 mA/cm
2
), leading to average power conversion efficiencies of 3.6 – 4.9%,
which exceed the average value of 3.4% for P3HT under AM 1.5G illumination (100
mW/cm
2
). External quantum efficiency (EQE) measurements revealed a strong
photoresponse from the semi-random polymers up to 850 nm, with EQE values above
40% across the visible and into the near-infrared for P3HTT-DPP-10% and P3HTT-DPP-
15%. These results indicate that semi-random P3HT analogues provide a simple and
effective route toward polymers with a broad photocurrent response in bulk
heterojunction solar cells.

4.1.2 Results and Discussion

The ongoing development of an understanding of the operating principles of
polymer-based bulk heterojunction (BHJ) photovoltaic devices
1–4
has given rise to
increases in reported efficiencies, which now exceed 7%.
5–9
This rapid improvement in
efficiency, along with the possibility for the economical fabrication of flexible,
lightweight, and large area solar cells,
10
make organic photovoltaics a potential
competitor to the crystalline silicon solar cells that currently dominate the market.
11

However, in order for polymer solar cells to become a viable technology, it is widely
agreed that further increases in the efficiency (η) are needed (>10%).
12
The primary
strategy toward this end has been the design of novel donor polymers, within the context
of the highly successful and ubiquitous fullerene-based acceptors.
13,14
In this design
process, there are numerous parameters of the polymeric donor that must be optimized.
143
The first is the ability of the polymer to absorb light broadly across the solar spectrum
with a high absorption coefficient (~10
5
cm
-1
), as the short-circuit current density (J
sc
) of
the solar cell is proportional to the product of the spectral absorption breadth and
absorption intensity of the active layer.
15
Second, the polymer must also have energy
level offsets (with respect to the fullerene acceptor) positioned to provide sufficient
driving force for charge generation and maximization of the open-circuit voltage (V
oc
).
1

The polymer must also have a sufficiently high hole mobility to ensure effective charge
extraction,
2
which is closely tied to the solid-state structure of the polymer and strongly
affects the fill factor (FF).
16–18
Finally, any donor polymer must be capable of effective
mixing with the fullerene acceptor in order to generate a bicontinuous morphology.
19,20

For the first design parameter, broadening of the spectral absorption is usually
targeted via the so-called donor-acceptor approach,
15,21
where electron-rich and electron-
deficient units are polymerized in an alternating fashion along the polymer backbone,
leading to a narrowing of the polymer bandgap (E
g
). The desired consequence is the
absorption of a larger fraction of solar photons and an increase in the J
sc
and overall
efficiency.
22
The main weakness of the alternating donor-acceptor strategy is the
frequently observed shifting of the polymer absorption profile (relative to the pure donor
homopolymer) to the long wavelength region as opposed to a true broadening across both
the visible and near-infrared regions.
5,9,23–25
Decreasing the polymer absorption in the
visible region of the solar spectrum can hinder the desired increase in the J
sc
and
ultimately the efficiency.
144
Recently, we reported a new class of donor-acceptor copolymers based on
regioregular poly(3-hexylthiophene) (rr-P3HT or simply P3HT) called semi-random
polymers, where only a small amount of acceptor units (10 – 17.5%) were introduced in a
randomized fashion into the polymer backbone.
26
In this approach, monomer
incorporation is controlled via a restricted linkage pattern of the donor and acceptor units,
preventing the formation of acceptor-acceptor linkages as well as sterically unfavorable
linkages, and preserving continuous segments of head-to-tail (regioregular) coupled 3-
hexylthiophene. The randomized sequence distribution and nature of the donor-acceptor
chromophores (based on the numerous possible chromophores defined by the effective
conjugation length) generated broadly absorbing, multichromophoric polymers with
intense absorption across the visible and up to 1000 nm. Importantly, the small
percentage of the acceptor moieties preserved many of the important properties of P3HT
that are conducive to high solar cell performance, among which are: semi-crystallinity,
high hole mobility and good miscibility with the fullerene acceptor PC
61
BM. Polymer
miscibility with the fullerene is important for minimizing the amount of fullerene
required to generate a percolated network and maintaining the ability of the polymer to
absorb a meaningful fraction of long wavelength light within the constraints of an active
layer that cannot typically exceed a thickness of ~100 – 200 nm.
18,27–30
Solar cells, based
on the previously synthesized semi-random polymers, showed moderate performance, but
nonetheless indicated that semi-random polymers are interesting for further research as
potentially attractive donor polymers for BHJ solar cells.
145
Here we report the first examples of semi-random P3HT analogues that give high
efficiency in bulk heterojunction solar cells. In this study we have chosen
diketopyrrolopyrrole (DPP)
31–35
as the acceptor unit and the effect of acceptor content on
polymer properties and photovoltaic response is described. Three novel donor-acceptor
polymers with different percentages of the DPP unit (5 – 15 %) were synthesized. These
polymers were fully characterized and used in BHJ solar cells with PC
61
BM as the
acceptor, giving efficiencies reaching nearly 5%.

Figure 4.1. Synthesis and structures of P3HTT-DPP-5%, P3HTT-DPP-10% and P3HTT-
DPP-15%.

The synthesis of all three P3HTT-DPP polymers (P3HTT-DPP stands for poly(3-
hexylthiophene – thiophene – diketopyrrolopyrrole)) were carried out in analogy to the
previously reported semi-random P3HT analogues,
26
as illustrated in Figure 4.1.
Copolymerization of 2-bromo-5-trimethyltin-3-hexylthiophene (1) with 2,5-
bis(trimethyltin)thiophene (2) and varying amounts of dibromo-bisthiophene-
diketopyrrolopyrrole (3) in DMF at 95 °C with Pd(PPh
3
)
4
as the catalyst, gave polymers
146
with M
n
17,500 – 24,500 g/mol (see Appendix B). Polymer structures were confirmed by
1
H-NMR and it was found that polymer composition matched the monomer feed ratios,
as we have previously observed with analogous semi-random polymers.
26
The resulting
polymers are represented by the acronyms P3HTT-DPP-X%, where the percentage of
DPP monomer is indicated, giving P3HTT-DPP-5%, P3HTT-DPP-10%, and P3HTT-
DPP-15%.
The choice of the bisthiophene-diketopyrrolopyrrole (DPP) unit (monomer 3) as
the acceptor was influenced by the growing number of recent reports of high efficiency in
BHJ solar cells with polymers
32–34
and small molecules
31
containing this unit, albeit with
variable N-alkyl substitution. Introduction of the DPP unit into polymer backbones with
different donor units
36–39
led to low optical band gaps (1.2 – 1.6 eV), photocurrent
response up to 1100 nm,
40,41
J
sc
of more than 10 mA/cm
2
and solar cell efficiencies of 4 –
5.5%.
32–34,38,42,43
The presence of one thiophene on each side of the DPP acceptor
minimizes steric hindrance and induces planarity, which enhances chain packing and
intermolecular π-π interaction.
38,42,43
As a result, high hole mobilities are observed for
DPP-containing polymers.
34
The DPP acceptor also influences the energy levels of the
polymer, lowering the position of the HOMO and LUMO levels.
33,38
As a result, V
oc
values in the range 0.8 – 0.9 V have been obtained.
32,38,39,44
These physical and electronic
properties of DPP-based polymers, together with their observed thermal stability,
35,45

make DPP a very attractive unit for incorporation into semi-random copolymers.
The optical properties of the semi-random P3HTT-DPP polymers in o-
dichlorobenzene (o-DCB) solutions and thin films were studied using UV-vis
147
spectroscopy as shown in Figure 4.2 and 4.3. As a reference, data is shown for P3HT
synthesized using analogous Stille polymerization conditions.
26
The introduction of DPP
into the P3HT backbone is observed to significantly decrease the optical bandgap and
lead to the formation of a distinct dual band absorption in solutions and thin films. This
type of absorption profile is often ascribed to π-π* (short wavelength band) and ICT
(intramolecular charge transfer) transitions (long wavelength band).
46,47
In the case of
semi-random polymers, the dual band absorption could more specifically be assigned to
π-π* transitions of segments in the randomized polymer that are thiophene-rich (short
wavelength band) and ICT transitions in segments that are rich in donor-acceptor
linkages (long wavelength band). This is especially evident in solution (Figure 4.2)
where pristine P3HT has a single absorption band with a peak at 463 nm and this
absorption band is retained in all three P3HTT-DPP copolymers. With increasing content
of DPP acceptor, the intensity and breadth of the long wavelength absorption band
(donor-acceptor or ICT band) increases at the expense of the short wavelength band
(thiophene π-π*). More specifically, an increase of the DPP content leads to a red-shift
and intensity increase of the ICT band, accompanied by a blue-shift and intensity
decrease in the π-π* band, which is even more pronounced in thin films (Figure 4.3).
Absorption coefficients in thin films of the ICT band of P3HTT-DPP-10% and
especially P3HTT-DPP-15% are approaching 10
5
cm
-1
and are comparable to the peak
value of P3HT. Furthermore, the ICT peak positions of the P3HTT-DPP polymers are
located close to the maximum of the photon flux from the sun, which is at 700 nm (1.8
eV).
15
It is notable that small red-shifts of the ICT peak positions for P3HTT-DPP
148
polymers (10 – 15 nm), when going from solution to film, imply that already in solution
the polymers adopt a planar conformation and upon film formation only a small
reorganization and increase in packing between the polymer chains occurs (especially in
the case of P3HTT-DPP-15%). This is in contrast to P3HT, which displays a 96 nm red-
shift induced by significant ordering in the solid state. As a further point, thermal
annealing is observed to enhance the thin film absorption of P3HT and affect the
absorption profile.
48
The P3HTT-DPP thin film absorption spectra were collected with
thin films spin-coated from o-DCB solution. It was observed that the films exhibited a
rapid color change from dark green to grey-purple immediately after spin-coating was
completed. Unlike P3HT, neither thermal annealing nor slow solvent evaporation had any
effect on the absorption profiles. Overall, increasing the content of DPP in the polymer
backbone gives rise to a relatively uniform absorption profile at low DPP content (5%),
analogous to the multichromophoric description that we proposed for previously reported
semi-random donor acceptor copolymers.
26
However, as the DPP content increases (10 –
15%), the polymer absorption profile begins to converge toward that observed for
perfectly alternating thiophene-DPP polymers, which contain significantly higher
contents (50%) of DPP acceptor.
32–34,42,43

149

Figure 4.2. UV-vis absorption spectra of polymers in solution (o-dichlorobenzene or o-
DCB) where (i) is P3HT, (ii) is P3HTT-DPP-5%, (iii) is P3HTT-DPP-10% and (iv) is
P3HTT-DPP-15%.

Figure 4.3. UV-vis absorption spectra of polymers in thin film (spin-coated from o-DCB)
where (i) is P3HT (annealed at 150

°C for 30 min for the thin films), (ii) is P3HTT-DPP-
5% (thin film as-cast), (iii) is P3HTT-DPP-10% (thin film as-cast) and (iv) is P3HTT-
DPP-15% (thin film as-cast).

Another interesting feature in the thin film absorption spectra is the presence of
the vibronic features in the ICT band. The same vibrational shoulders were observed in
150
the case of other DPP-based polymers and small molecules,
31,39
and were ascribed to the
high degree of ordering and strong intermolecular (π-π) interactions.
33,49
To verify the
formation of semi-crystalline polymer thin films, grazing-incidence X-ray diffraction
(GIXRD) was used (see Appendix B). P3HTT-DPP polymers were spin-coated from o-
DCB solutions under identical conditions used for the preparation of films for absorption
spectra. P3HTT-DPP-10% was found to exhibit evidence of crystallinity in the as-cast
films with an interchain distance (100) of 14.7 Å (for comparison, the P3HT interchain
distance is 16.6 Å). In contrast, P3HTT-DPP-5% and P3HTT-DPP-15% show features
indicative of crystalline order only upon annealing at 150

°C showing 16.0 and 16.2 Å
interchain distances, respectively. Under the same thermal annealing condition, P3HTT-
DPP-10% shows a more intense peak than that observed in the as-cast film and a larger
interchain distance of 15.3 Å. The origin of the difference in initial ordering behavior of
the as-cast films is not clear, although it could be partially attributed to the lower
molecular weight and larger PDI of P3HTT-DPP-5% and P3HTT-DPP-15% (M
n
=
19,000 g/mol, PDI = 2.8 and M
n
= 17,570 g/mol, PDI = 2.9, respectively) compared to
P3HTT-DPP-10% (M
n
= 24,570 g/mol, PDI = 2.3). Slow solvent evaporation was
observed to have no effect for any of the P3HTT-DPP polymers. However, all of the
P3HTT-DPP polymers showed evidence of a semi-crystalline structure.
The space-charge limited current (SCLC) method was employed to determine the
hole mobilities of the P3HTT-DPP polymers. High hole mobilities in the range of 1 –
2.3*10
-4
cm
2
/(V*s) (Table 4.1) were obtained, which are close to that of P3HT and are
attributed to the semi-crystalline nature of the polymers revealed with the GIXRD. In
151
contrast to GIXRD measurements, the maximum mobilities were obtained when a slow
solvent evaporation technique was employed, where thin films were placed in a N
2

cabinet for 20 minutes before aluminum deposition.
The HOMO and LUMO energy levels for P3HT and the P3HTT-DPP polymers
were measured by cyclic voltammetry (CV) with ferrocene as a reference, and converted
to the vacuum scale using the approximation that the ferrocene redox couple is 5.1 eV
relative to vacuum (see Appendix B).
13,50,51
All P3HTT-DPP polymers, independent on
the DPP content, showed a HOMO level of 5.2 eV, which is equivalent to that of P3HT.
The measured position of the HOMO levels also indicates that the polymers should be
resistive to air oxidation and thus facilitate device operational lifetime.
52,53

Table 4.1. Photovoltaic properties of P3HT, P3HTT-DPP-5%, P3HTT-DPP-10% and
P3HTT-DPP-15% with PC
61
BM as an acceptor
Polymer:PC
61
BM
(ratio)
Thickness
(nm)
c
SCLC hole
mobility
(cm
2
/(V*s))
d

J
sc

(mA/cm
2
)
e
V
oc

(V)
FF

η
(%)
P3HT (1:1)
a
95 2.3 x 10
-4
9.49 0.61 0.61 3.42
P3HTT-DPP-5% (1:1)
b
74 1.1 x 10
-4
9.57 0.66 0.58 3.60
P3HTT-DPP-10% (1:1.3)
b
71 2.3 x 10
-4
13.87 0.57 0.63 4.94
P3HTT-DPP-15% (1:2.6)
b
75 1.3 x 10
-4
13.44 0.50 0.60 4.10
a
Spin-coated from chlorobenzene (CB) and annealed at 150

°C for 30 min under N
2
after aluminum
deposition.
b
Spin-coated from o-DCB and tested after spending 20 min in a N
2
cabinet before aluminum
deposition.
c
Measured by X-ray reflectivity.
d
Measured for neat polymer films.
e
Mismatch corrected
54
(see
Appendix B).

The observed characteristics (high molecular weight, low bandgap, high
absorption coefficient, high hole mobility) of the P3HTT-DPP polymers make them
excellent candidates for photovoltaic devices. BHJ solar cells in a conventional device
configuration of ITO/PEDOT:PSS/polymer:PC
61
BM/Al were fabricated in air (see
152
Appendix B). The optimized polymer:PC
61
BM weight ratios for P3HTT-DPP-5%,
P3HTT-DPP-10%, P3HTT-DPP-15% were found to be 1:1, 1:1.3 and 1:2.6, respectively.
Optimal processing conditions include slow solvent evaporation (solvent annealing) from
the polymer:PC
61
BM composites for 20 minutes in a N
2
cabinet after spin-coating and
prior to aluminum deposition, analogous to the conditions observed to give the highest
mobilities for the polymers in the SCLC measurements. Interestingly the same solvent
annealing process was not observed to give any changes in the GIXRD data or in the
absorption spectra of the polymers. Shorter or longer solvent annealing times for the solar
cells led to a decrease in J
sc
and thermal annealing across a range of temperatures was
also observed to decrease the performance of the solar cells. Table 4.1 lists the average
values of η, V
oc
, FF, and mismatch corrected
54
J
sc
obtained under simulated AM 1.5G
illumination (100 mW/cm
2
) (J-V curves are provided in the Appendix B). For reference,
the average efficiency measured here for P3HT:PC
61
BM solar cells is 3.42%. While this
value is lower than reported champion level values of 4.5 – 5.0%, the difference is
thought to primarily reflect the consequences of device fabrication and testing in air as
opposed to an inert environment. High values of FF for all devices can be attributed to
the high hole mobilities of the polymers, presumably leading to balanced charge transport
in the devices and a reduction of recombination.
17,18,27
It is observed that the V
oc
of the
solar cells varies from 0.66 V with P3HTT-DPP-5% to 0.57 V with P3HTT-DPP-10%, to
0.50 V with the P3HTT-DPP-15%. The 160 mV range in the V
oc
cannot be explained by
differences between the HOMO of the donor and LUMO of the acceptor,
1
because the
positions of the HOMO levels of the three P3HTT-DPP polymers are the same. We
153
speculate that one possible explanation for the changes in the V
oc
could be related to the
increase of the degree of aggregation, when going from P3HTT-DPP-5% to P3HTT-
DPP-10% to P3HTT-DPP-15%, supported by the observed decrease in the solubility of
the polymers in o-DCB with increasing DPP content.
55
As such, the recombination rate
could increase with increasing DPP content,
55–57
thus slowing down the kinetics of
molecular electron transfer
58
and leading to the observed V
oc
reduction. However, it
should be noted that V
oc
values observed for perfectly alternating DPP copolymers, which
utilize thiophenes as the donor units were 0.63 – 0.68 V in PC
61
BM blends.
42,43
As such,
the origins of this trend in the V
oc
are under further investigation.
A more easily explained trend is that of the J
sc
, where the decrease of the
polymer bandgap, relative to P3HT (J
sc
= 9.49 mA/cm
2
, mismatch correction M = 1.05),
results in significant increases in the J
sc
, giving 9.57 mA/cm
2
for P3HTT-DPP-5% (M =
0.87), 13.87 mA/cm
2
for P3HTT-DPP-10% (M = 0.76) and 13.44 mA/cm
2
for P3HTT-
DPP-15% (M = 0.71). The highest observed J
sc
value for P3HTT-DPP-10% can
explained by the more balanced, intense absorption across the visible and near-infrared
regions with respect to the other two DPP containing polymers and the smaller quantity
of PC
61
BM needed to optimize device performance in comparison to P3HTT-DPP-15%.
The photocurrent response for all the optimized BHJ solar cells is shown in
Figure 4.4. All devices showed strong photocurrent response in the range 350 – 850 nm,
with EQE values of 41% and 46% at 750 nm for P3HTT-DPP-10% and P3HTT-DPP-
15%, respectively. Photocurrent peaks around 400 nm are assigned to PC
61
BM light
absorption, while photocurrent responses in the longer wavelength regions are attributed
154
primarily to the polymers. The integrated photocurrents from the EQE measurement
match within 5% to that of the mismatch corrected photocurrents measured under
simulated AM 1.5G illumination (see Appendix B for mismatch corrected (J
sc,corr
) and
integrated (J
sc,EQE
) photocurrents). A uniformly strong photocurrent response from the
P3HTT-DPP-10% in the 350 – 850 nm range is explained by the favorable ratio of
polymer to PC
61
BM of 1:1.3. In the case of perfectly alternating DPP-polymers, the best
results were achieved at much higher polymer:fullerene ratios (1:2 and higher),
32–34,38,42,43

leading to a decrease in the uniformity of the photocurrent generation, especially in the
near-infrared region, and hence smaller obtained J
sc
.
38,42,43

Figure 4.4. External quantum efficiency of the BHJ solar cells based on P3HT (black
squares), P3HTT-DPP-5% (red circles), P3HTT-DPP-10% (green triangles) and P3HTT-
DPP-15% (blue stars) with PC
61
BM as the acceptor, under optimized conditions for
device fabrication.

As further characterization of the BHJ solar cells, transmission electron
microscopy (TEM) images (see Appendix B) show the presence of uniform, bicontinuous
155
thin films, with small length-scales of phase separation in PC
61
BM blends for all the
P3HTT-DPP polymers. The observed morphologies are indistinguishable from optimized
blends of P3HT and PC
61
BM and result in an apparent large interfacial area for efficient
charge separation, which helps to explain the high attainable FFs and J
sc
values for P3HT
and P3HTT-DPP polymers. In general, the semi-random approach allows favorable
morphology formation without application of any solvent additives
32,33
or thermal
annealing
28
at a close to 1:1 polymer:fullerene ratio, when the acceptor content in the
polymer backbone is low.

4.1.3 Conclusion

In summary, we have synthesized a family of novel semi-random P3HTT-DPP
copolymers containing different contents (5 – 15 %) of the DPP acceptor unit. These
polymers combine broad absorption profiles, high absorption coefficients, high hole
mobilities and semi-crystalline structures similar to P3HT. In BHJ solar cells with
PC
61
BM, the polymers show effective film formation with optimized polymer:fullerene
ratios that vary based on the content of DPP in the polymer backbone and efficiencies of
nearly 5.0% are observed for P3HTT-DPP-10% at a 1:1.3 polymer:fullerene ratio. A
broad photocurrent response, representative of the polymer absorption profile confirms
that semi-random donor-acceptor copolymers are an effective platform for improving
light harvesting in BHJ solar cells that further benefits from a simple and highly modular
156
synthetic protocol. Ongoing studies are focused on elucidating the effect of acceptor-
monomer content and identity on photovoltaic device parameters.

4.2 Influence of the Ethylhexyl Side-Chain Content on the Open-Circuit Voltage in
rr-Poly(3-hexylthiophene-co-3-(2-ethylhexyl)thiophene) Copolymers

4.2.1 Introduction

Although recently considerable attention has been paid to the impact of polymer
alkyl side-chains on conjugated-polymer:fullerene solar cell performance, and especially
the V
oc
and J
sc
, a clear and comprehensive picture of the effect of side-chain positioning,
length and branching has yet to evolve. In order to address some of these questions we
designed a simple, modular system of random copolymers based on rr-P3HT. The
influence of increasing amounts of branched 2-ethylhexyl side-chains (10, 25 and 50%)
in rr-poly(3-hexylthiophene-co-3-(2-ethylhexyl)thiophene) copolymers on properties
such as UV-vis absorption, polymer crystallinity, HOMO energy levels,
polymer:PC
61
BM solar cell performance and especially the V
oc
was studied and compared
to the corresponding homopolymers P3HT and poly(3-(2-ethylhexyl)thiophene)
(P3EHT). Polymers with 50% or less 2-ethylhexyl side chains (P3HT
90
-co-EHT
10
,
P3HT
75
-co-EHT
25
, P3HT
50
-co-EHT
50
) have the same band gap, similar absorption
properties and also retain the semi-crystalline nature of P3HT whereas P3EHT has higher
band gap and lower absorption coefficient. Polymer HOMO levels were determined by
157
electrochemistry in solution and thin film and are virtually identical for all polymers in
solution whereas in the solid state an increase in the amount of 2-ethylhexyl side-chains
leads to marked and correlated decrease in the HOMO levels. This decrease is directly
reflected in the V
oc
measured in polymer:PC
61
BM solar cells which increases with
increasing 2-ethylhexyl side-chain content indicating a relatively straightforward
HOMO
DONOR
-LUMO
ACCEPTOR
dependence of the V
oc
for this family of polymers.
P3HT
75
-co-EHT
25
benefits from an increased V
oc
(0.69 V), a J
sc
(9.85 mA/cm
2
) on the
same order of P3HT and a high FF and ultimately achieves an efficiency of 3.85%
exceeding that measured for P3HT (V
oc
= 0.60 V, J
sc
= 9.67 mA/cm
2
, 3.48%). The
observed efficiency increase suggests that the random incorporation of branched alkyl
side-chains could also be successfully used in other polymers to maximize the V
oc
while
maintaining the band gap and improve the overall polymer:fullerene solar cell
performance.

4.2.2 Results and Discussion

Polymer solar cells have attracted considerable attention over the last decade due
to their potential advantages over inorganic counterparts. The vision of low-cost energy
production using lightweight and flexible solar cells, which are easily incorporated in
existing infrastructure, has spurred extensive research efforts.
1,59,60
Even though
efficiencies have reached over 8%, further studies in order to increase them to 10 – 12%
are necessary to make conjugated polymer solar cells likely to become commercially
158
successful.
5,6,9,61–63
Among the great number of conjugated polymers that have been
investigated, regioregular-poly(3-hexylthiophene) (rr-P3HT) is the most studied
conjugated polymer. P3HT offers a unique property set and has a semi-crystalline
structure and high hole mobility as well as a favorable mixing ratio with fullerene
acceptors such as PCBM (~ 1:1 as opposed to ratios of 1:3 or even 1:4 for many Donor-
Acceptor (D/A) copolymers), which allows for maximum amounts of polymer, and
consequently strong light absorption, in active layers with a constrained thickness.
1
The
combination of these positive factors leads to solar cell efficiencies of up to 5% for
P3HT:PCBM.
64
However, due to unfavorable positioning of the frontier energy levels
(HOMO and LUMO) and a wide band gap (E
g
), the short-circuit current density (J
sc
) as
well as the open-circuit voltage (V
oc
) in fullerene blends are restricted to relatively small
values thus limiting the achievable power conversion efficiency of P3HT-based solar
cells.
15

The most commonly used approaches to higher efficiency solar cells are a
decrease in the polymer E
g
and thus increase in J
sc
by absorbing more photons and/or the
lowering of the polymer HOMO to achieve a higher V
oc
(which can be roughly
approximated by the HOMO
DONOR
-LUMO
ACCEPTOR
offset).
2
For both strategies the D/A
approach is often used where copolymerization of electron-rich monomers with electron-
poor monomers leads to lower E
g
polymers and at the same time gives control over the
position of the HOMO (and LUMO) level thus potentially also allowing for an increase
of the V
oc
.
21
As a consequence, P3HT and poly(3-alkylthiophenes) (P3ATs) in general,
have lately mainly been used to study fundamental behavior such as polymer and
159
fullerene crystallization or interdiffusion with fullerenes and are less associated with
strategies to increase efficiency.
65–68

Until recently in the design of new conjugated polymers for solar cells it was
commonly accepted that the conjugated backbone of a polymer determines the electronic
properties whereas the attached (non-conjugated) alkyl side-chains were considered
merely a necessity for the solubility of polymers in organic solvents to allow for solution
processing.
69
Nevertheless, considerable influence of side-chains on the morphology
(crystalline vs. amorphous, lamellar stacking distance, mixing with fullerenes) as well as
thermal properties (glass transition temperature and melting point) of the polymer has
been widely known.
70,71
A common example for the pronounced influence of alkyl side-
chains is rr-P3HT (head-to-tail ordering of 3-hexylthiophene), which shows high hole
mobility, semi-crystallinity and high solar cell efficiencies, whereas regiorandom-P3HT
is amorphous with considerably lower efficiencies.
72,73
P3HT also exemplifies the fact
that the alkyl chain length is important as out of all investigated P3ATs the hexyl side-
chain seems to be the optimal compromise between solubility (processability), good
morphology (e.g. phase separation between polymer and fullerene and crystallinity) as
well as chromophore density.
74
It has also been found with P3ATs that the optimal
polymer:fullerene ratio is influenced by the side-chain density on the conjugated
backbone, where a lower density allows not only for interdigitation of side-chains but
also leads to intercalation of fullerene and thus higher required ratios of acceptor in the
solar cells.
75
As such the identity, content and distribution of alkyl side-chains play a
strong role in the solar cell performance of P3ATs and are slowly drawing more
160
attention, although no comprehensive or conclusive picture has yet emerged of the
specific influences of all those structural variations.
Recently a more profound influence of alkyl side-chains on polymer properties
has been reported for both D/A-copolymers and P3ATs. Several studies found a
correlation between the alkyl side-chain length and the V
oc
of the polymer:fullerene solar
cells where typically a longer side-chain leads to a lower HOMO level of the polymer
(while E
g
remains constant) and larger V
oc
while keeping the conjugated backbone
constant.
76,77
It needs to be mentioned though, that these changes in V
oc
and HOMO level
tend to be relatively small (often within 100 mV) while processing conditions such as
annealing can also have a large influence on V
oc
. In addition, other reports show less clear
trends so that it is difficult to draw any general conclusions.
74,78,79
With P3ATs, Hou et al.
reported that by decreasing the alkyl side-chain density on a thiophene backbone using a
long branched alkyl chain on every third thiophene ring they were able to increase the V
oc

considerably compared to P3HT (as a consequence of a lower HOMO level of the
polymer) without increasing the band gap. However, no improvement of efficiency was
achieved compared to P3HT solar cells.
80
Ko et al. used a similar approach with P3ATs
to lower the HOMO energy and increase the V
oc
by only putting alkyl chains on the 3 and
4 position of every third thiophene ring. In this case obtained efficiencies of up to 4.2%
with PC
71
BM were found to exceed those observed with P3HT.
81
In both of these cases
with P3ATs the decrease in density of electron donating alkyl side-chains was cited as the
reason for a lower HOMO and consequently a larger V
oc
. Interestingly in the study by Ko
et al. the V
oc
was also observed to increase with side-chain length (going from hexyl side-
161
chain to dodecyl) whereas the HOMO levels remained almost constant (but lower than
P3HT). Note that this is not consistent with the trend reported for P3ATs where longer
side-chains lead to a lower HOMO level.
76,77
This result suggests that side-chain length
has an effect on V
oc
that is more complex than simply via an influence on the HOMO
level.
In D/A copolymers it is even more difficult to identify the influence of alkyl side-
chains as not only the type (length, branched or linear) but also the position (e.g. on the
donor or acceptor moiety) of the solubilizing groups can be changed. In addition studies
often do not contain enough data points or too many parameters are changed
simultaneously to draw any general conclusions on the impact of side-chains on the
V
oc
.
82–85
However, there are a several reports in the literature which show a pronounced
influence of side-chains on the V
oc
with D/A polymers in polymer:fullerene solar cells. Li
et al. published a family of cyclopentadithiophene (CPDT)-thienopyrroledione
copolymers with varying alkyl chains where they found that longer/branched alkyl chains
on CPDT lead to lower HOMO energies and increased V
oc
.
86
Yang et al. reported a study
on naphtodithiophene (NDT)-dithiophenebenzothiadiazole (DTBT) copolymers where
they changed the alkyl side-chains on the NDT and DTBT units, varying the length
(octyl, dodecyl) as well as the branching (2-hexyldecyl, 2-ethylhexyl) of the side-chains
but leaving placement constant.
87
In this case the authors found a pronounced influence
of the type of side-chains on the V
oc
as well as J
sc
of the polymer:PCBM devices, where
long and branched side-chains lead to large V
oc
but small J
sc
values. A correlation
between the preexponential term J
so
(from the generalized Shockley equation for solar
162
cells) and the V
oc
of the polymer:PCBM devices was found (similar observations on small
molecule solar cells had already been made by Perez et al.
88
) and it was stated that long
branched alkyl chains weaken intermolecular interaction between polymer and fullerene
and thus lead to small J
so
values and consequently large V
oc
values. It has to be mentioned
though that HOMO levels of polymers in this study were not constant with different side-
chain substitution and that for most cases the HOMO level of these polymers follows the
same trend as the V
oc
of the polymer:PCBM devices, with the highest HOMO level
belonging to the polymer with short and linear side-chains (which also has the smallest
V
oc
value) whereas longer as well as branched side chains lower the HOMO level and
give higher V
oc
. It was concluded that short, branched alkyl side chains (such as 2-
ethylhexyl) are optimal for polymer:fullerene device performance as they allow for both a
large V
oc
and large J
sc
and thus the highest efficiency.
From this short literature overview it can easily be seen that no comprehensive
picture of the effect of side-chain positioning, length and branching on polymer:fullerene
solar cell performance has evolved. In addition there seems to be no general agreement
on the reasons for the observed effects of the alkyl side-chains on the V
oc
(influence of
interfacial interactions (J
so
) vs. lowering of polymer HOMO level). In order to address
some of these questions, and specifically the influence of branched alkyl side-chains, we
designed a simple, modular system of random copolymers based on rr-P3HT. This
system is intended to help gain insight into the effect the introduction of varying amounts
of short and branched alkyl side-chains (2-ethylhexyl) have on the properties and solar
cell performance (and especially the V
oc
) of the resulting polymers. This system takes
163
into account the beneficial impact of 2-ethylhexyl chains on device properties that has
been shown in D/A-copolymers as well as the fact that a certain degree of disorder in the
polymer backbone is beneficial for solar cell performance.
69,87,89,90
The influence of
increasing amounts of branched 3-(2-ethylhexyl)thiophene (10, 25 and 50%) in rr-poly(3-
hexylthiophene-co-3-(2-ethylhexyl)thiophene) copolymers on important properties such
as UV-vis absorption, polymer crystallinity, HOMO energy levels and polymer:PC
61
BM
solar cell performance (especially the V
oc
) is studied and compared to the results of the
corresponding homopolymers P3HT and poly(3-(2-ethylhexyl)thiophene) (P3EHT). It
should be noted that P3EHT has received relatively little attention and no solar cell data
has been reported to date. Somanathan et al. characterized P3EHT, which was
synthesized by chemical oxidative polymerization,
91
whereas Hashimoto et al. studied
copolymers between 3-hexylthiophene and 3-(2-ethylhexyl)thiophene (block as well as
random copolymers) and were able to find a distinct influence of the ratio of comonomers
on the polymer film morphology and the field-effect hole mobilities for block copolymers
as well as random copolymers.
92–94
Segalman et al. demonstrated that P3EHT has
comparable optical and charge transport properties as P3HT but did not report solar cell
data.
95
Here we show that even small amounts of 3-(2-ethylhexyl)thiophene added as a
comonomers to 3-hexylthiophene lead to an increased V
oc
of the polymer:PC
61
BM solar
cells (because of lowered HOMO levels in the solid state) while J
sc
remains high due to
high absorption coefficients ultimately leading to an efficiency increase compared to
P3HT.
164
Polymers were synthesized using an established Stille coupling
26,96
procedure with
2-bromo-5-trimethyltin-3-hexylthiophene and 2-bromo-5-trimethyltin-3-(2-
ethylhexyl)thiophene as co-monomers to insure polymerization with a head-to-tail
regioregularity as is shown in Figure 4.5. Synthesis of 2-bromo-5-trimethyltin-3-
hexylthiophene has been reported previously
26
and 2-bromo-5-trimethyltin-3-(2-
ethylhexyl)thiophene 3 was synthesized starting from 3-bromothiophene using Kumada
coupling
93
to give 1 followed by bromination in the two-position using N-
bromosuccinimide to give 2 and finally stannylation using LDA and trimethyltinchloride
to give the monomer 3 (Figure 4.5). The ratio of monomers was varied starting from
polymerizations with only 2-bromo-5-trimethyltin-3-hexylthiophene to give pure P3HT
and then introducing 10, 25 and 50% of 2-bromo-5-trimethyltin-3-(2-
ethylhexyl)thiophene. Polymers are named P3HT
(100-x)%
-co-EHT
x%
in order to indicate
the respective amounts of 3-hexylthiophene and 3-(2-ethylhexyl)thiophene in the
copolymers. In all cases very soluble polymers with high molecular weights (determined
via GPC using polystyrene standards) and good yields were obtained after Soxhlet
purification (Table 4.2). The polymer structures were analyzed using
1
H NMR and match
closely with previously reported spectra of copolymers of 3-hexylthiophene and 3-(2-
ethylhexyl)thiophene.
92
By integrating the aromatic peaks at 6.96 ppm (aromatic CH
from 3-hexylthiophene) as well as 6.94 ppm (aromatic CH from 3-(2-
ethylhexyl)thiophene) and comparing with the integration of the benzylic protons of both
monomers at ~2.8 ppm it was confirmed that the monomer feed ratio is equivalent to the
incorporation of monomers in the backbone.
165

Figure 4.5. Synthesis of monomer 2-bromo-5-trimethyltin-3-(2-ethylhexyl)thiophene (3)
and Stille polymerization for poly(3-hexylthiophene-co-3-(2-ethylhexyl)thiophene).

Table 4.2. Molecular and Electronic Properties of Polymers
Polymer
M
n
g/mol
(PDI)
a
HOMO (eV)
(solution)
b
HOMO (eV)
(thin film)
c
E
g
(optical)
(eV)
d
P3HT 24,240 (2.7) 5.25 5.17 1.9
P3HT
90
-co-EHT
10
21,330 (2.5) 5.25 5.30 1.9
P3HT
75
-co-EHT
25
26,120 (2.5) 5.30 5.43 1.9
P3HT
50
-co-EHT
50
40,130 (2.0) 5.32 5.48 1.9
P3EHT 22,180 (2.9) 5.28 5.57 2.0
a
Determined by GPC with polystyrene as standard and o-DCB as eluent.
b
Cyclic voltammetry (vs Fc/Fc
+
)
in chloroform containing 0.1 M TBABF
4
.
c
Cyclic voltammetry (vs Fc/Fc
+
) in acetonitrile containing 0.1 M
TBAPF
6
.
d
Optical band gaps from onset of absorption in UV-vis spectra in annealed films.

UV-vis absorption spectra were recorded for all five polymers in solution
(chlorobenzene (CB)), as-cast films (see Appendix B) and annealed thin films (annealing
temperatures were chosen to be below the melting point of the polymer). The results are
summarized in Figure 4.6 and 4.7. As can be seen from the UV-vis absorption in
solution P3HT, P3HT
90
-co-EHT
10
and P3HT
75
-co-EHT
25
all have the same peak
absorption wavelength (456 nm) whereas the peak positions of P3HT
50
-co-EHT
50
and
P3EHT are slightly blue shifted (454 and 445 nm respectively) (Figure 4.6). The
166
absorptivity decreases with increasing ethylhexyl side-chain content, which can be
explained as a dilution effect of the chromophore due to the increase in larger side-chains
with increasing ethylhexyl content. In thin films the absorption maxima red shift for all
five polymers relative to the solution measurement (100 nm for P3HT, 68 nm for
P3HT
90
-co-EHT
10
, 65 nm for P3HT
75
-co-EHT
25
, 64 nm for P3HT
50
-co-EHT
50
and 40 nm
for P3EHT) which is an indication of planarization of the polymer backbones in the thin
film (Figure 4.7). All absorption profiles show vibronic features after annealing in the
longer wavelength region, also indicating ordering in the solid state and suggesting a
semi-crystalline structure (with the exception of P3HT
50
-co-EHT
50
all polymers show
vibronic features already in the unannelaed films, see Appendix B). The onset of
absorption in the thin films is the same for all polymers except for P3EHT which has a
slightly higher E
g
of 2.0 eV probably due to more significant steric interactions, which
inhibit backbone planarity. All other polymers have an E
g
of 1.9 eV which is identical
with P3HT. Interestingly in thin films the absorption coefficients of all polymers (except
P3EHT) are virtually the same showing almost no dilution effect of the chromophore due
to the side-chains. It can be concluded that the introduction of 50% or less of 2-
ethylhexyl side-chains does not considerably alter the absorption properties of the thin
films compared to P3HT.

167

Figure 4.6. UV-vis absorption of all five polymers in solution (CB) where (i) is P3HT
(purple line), (ii) is P3HT
90
-co-EHT
10
(green line), (iii) is P3HT
75
-co-EHT
25
(blue line),
(iv) is P3HT
50
-co-EHT
50
(red line) and (v) is P3EHT (orange line).

Figure 4.7. UV-vis absorption of all five polymers in thin films (spin coated from CB
and annealed for 30 min under N
2
at 150 °C for (i), (ii) and (iii), 100 °C for (iv) and 40
°C for (v)) where (i) is P3HT (purple line), (ii) is P3HT
90
-co-EHT
10
(green line), (iii) is
P3HT
75
-co-EHT
25
(blue line), (iv) is P3HT
50
-co-EHT
50
(red line) and (v) is P3EHT
(orange line).

168
Thin polymer films prepared under identical annealing conditions as for UV-vis
were analyzed by grazing-incidence X-ray diffraction (GIXRD) to confirm their semi-
crystalline nature. As can be seen in Figure 4.8 all polymers exhibit peaks in the 2θ range
of 5º-7º which are referenced as (100) reflections in relation to P3HT. The interplane
packing distance (100) for P3HT is calculated as 16.7 Å, for P3HT
90
-co-EHT
10
as 17.0 Å,
for P3HT
75
-co-EHT
25
as 16.9 Å, for P3HT
50
-co-EHT
50
as 16.0 Å and for P3EHT as 14.9
Å. P3EHT has a considerably smaller interplane distance than P3HT, which is consistent
with the literature
92,95
and is likely due to the significant steric interactions introduced by
the 2-ethylhexyl side-chains which inhibit backbone planarity and thus decrease the
density of side-chains in the (100) direction and lead to a shorter interplane packing
distance. That effect can also already be seen in P3HT
50
-co-EHT
50
where the interplane
distance is decreased to 16.0 Å. The slightly larger interplane distances for P3HT
90
-co-
EHT
10
, P3HT
75
-co-EHT
25
compared to P3HT can be explained by the introduction of the
bulkier 2-ethylhexyl side-chains which require more space, but at low percentages do not
introduce significant steric interactions so that the polymer backbone remains planar and
the density of side-chains in the (100) direction is not reduced. As a consequence the
stacking distance in the (100) direction increases as determined by the larger 2-ethylhexyl
side-chains. The degree of crystallinity, as inferred from the peak intensity, of the
polymer thin films decreases with introduction of 2-ethylhexyl side-chains as can be seen
in Figure 4.8 and is lowest for P3EHT. This can be explained by the unfavorable steric
interactions introduced by the 2-ethylhexyl side-chains, which are expected to inhibit
crystallization.
169

Figure 4.8. Grazing-incidence X-ray diffraction of thin films spin-coated from CB and
annealed for 30 min under N
2
(annealing temperature was 150 °C for (i)-(iii), 100 °C for
(iv) and 40 °C for (v)) are shown where (i) is P3HT (purple line), (ii) is P3HT
90
-co-
EHT
10
(green line), (iii) is P3HT
75
-co-EHT
25
(blue line), (iv) is P3HT
50
-co-EHT
50
(red
line) and (v) is P3EHT (orange line). The inset shows the region around 2θ = 4 º - 8º in
greater detail.

In order to expand on the results obtained by GIXRD and analyze the melting and
crystallization behavior of the polymers, differential scanning calorimetry (DSC) traces
were recorded. Melting points (see Appendix B) decrease with increasing percentage of
2-ethylhexyl side-chains as expected from the literature and match well for those
polymers which have been reported previously (Figure 4.9).
92
We were unable to
observe a melting transition for P3EHT, although it has been previously reported in the
literature (two peaks at ~ 70 and ~ 90 °C attributed to two coexisting crystal structures),
which is probably due to its relatively slow crystallization kinetics.
95,97
The presented
GIXRD and DSC data confirms that even though the polymers contain branched 2-
ethylhexyl side-chains and have a random distribution of 3-hexylthiophene and 3-(2-
170
ethylhexyl)thiophene co-monomers, the semi-crystalline nature of rr-P3HT is retained
and even P3EHT shows a certain degree of lamellar packing (as evidenced by GIXRD).

Figure 4.9. Melting points of polymers as measured by DSC.

In order to gain insight on the effect of the 2-ethylhexyl side-chain content on the
polymer HOMO levels, cylic voltammetry (CV) vs. ferrocene (see Appendix B) was
employed in both solution and thin films. The values are summarized in Table 4.2.
HOMO levels for polymers in solution are virtually identical (within the range of 5.25-
5.32 eV), whereas in the solid state an increase in the amount of 2-ethylhexyl side-chains
leads to a marked and correlated decrease in the HOMO level. In thin films the
introduction of only 10% 2-ethylhexyl chains led to a 0.13 eV decrease in the HOMO
level compared to P3HT, while P3EHT has a HOMO level of 5.57 eV (compared to 5.17
for P3HT). Considering that the optical band gap does not increase for P3HT
90
-co-EHT
10
,
P3HT
75
-co-EHT
25
and P3HT
50
-co-EHT
50
compared to P3HT, the LUMO level is most
171
likely shifting down simultaneously with the HOMO level. The fact that HOMO levels of
all five polymers are essentially identical in solution implies that the observed decrease in
HOMO level energy in thin films is a solid state organization effect and dependent on the
bulk properties of the polymers. This decrease in HOMO levels, especially for P3HT
90
-
co-EHT
10
, P3HT
75
-co-EHT
25
and P3HT
50
-co-EHT
50
comes with almost no decrease in
the absorption coefficient and with E
g
equivalent to P3HT. These results are consistent
with previous literature reports that show lowered HOMO levels (and constant E
g
) when
longer side-chains were introduced into polymers and are not only expected to allow for
larger V
oc
in solar cells but also make the polymers more stable against oxidation in
air.
52,76,77

Table 4.3. Photovoltaic properties of P3HT, P3HT
90
-co-EHT
10
, P3HT
75
-co-EHT
25
,
P3HT
50
-co-EHT
50
and P3EHT with PC
61
BM as the acceptor
Polymer:PC
61
BM
(ratio)
SCLC hole mobility
(cm
2
/(V*s))
f

J
sc
(mA/cm
2
)

V
oc
(V) FF η (%)
P3HT (1:1)
a
2.3*10
-4
9.67 0.60 0.60 3.48
P3HT
90
-co-EHT
10
(1:0.8)
b
1.77*10
-4
9.26 0.63 0.51 2.80
P3HT
75
-co-EHT
25
(1:0.8)
c
1.39*10
-4
9.85 0.69 0.57 3.85
P3HT
50
-co-EHT
50
(1:3.5)
d
1.07*10
-4
2.52 0.85 0.35 0.74
P3EHT (1:3.0)
e
2.87*10
-5
2.54 0.90 0.36 0.83
a
Spin-coated from chlorobenzene (CB) and annealed at 145

°C for 60 min under N
2
after aluminum
deposition.
b
Spin-coated from CB and annealed at 110

°C for 60 min under N
2
after aluminum deposition.
c
Spin-coated from CB and annealed at 110

°C for 30 min under N
2
after aluminum deposition.
d
Spin-coated
from CB and annealed at 110

°C for 10 min under N
2
after aluminum deposition.
e
Spin-coated from CB and
tested as-cast.
f
Measured for neat polymer films.

The space-charge limited current (SCLC) method was employed to determine the
hole mobilities of all five polymers and the values are summarized in Table 4.3. With
increasing 2-ethylhexyl content the mobility decreases, which correlates well with the
172
decreased degree of crystallinity of the polymer films. However, with the exception of
P3EHT, the measured mobilities are still extremely close to the hole mobility of P3HT.
After characterization of the neat polymers in solution and thin films, all polymers
were used in photovoltaic devices with PC
61
BM as the acceptor. BHJ solar cells in a
conventional device configuration of ITO/PEDOT:PSS/polymer:PC
61
BM/Al were
fabricated in air. P3HT
90
-co-EHT
10
and P3HT
75
-co-EHT
25
have a optimized
polymer:PC
61
BM ratios of 1:0.8, which is close to that found here for P3HT:PC
61
BM at
1:1. As mentioned earlier, this ratio is very favorable as it allows for a maximum amount
of polymer (and consequently strong light absorption) in an active layer with constrained
thickness and it also implies that the behavior of P3HT
90
-co-EHT
10
and P3HT
75
-co-
EHT
25
in blends is similar to P3HT showing good miscibility with the fullerene. Upon
further increase of the 2-ethylhexyl side-chain content (P3HT
50
-co-EHT
50
and P3EHT),
significant changes in the optimal polymer:fullerene ratios were observed and an excess
of PC
61
BM is necessary for peak device operation (1:3.5 and 1:3 for P3HT
50
-co-EHT
50

and P3EHT respectively), indicating a fundamental change in the behavior of these
polymers compared to P3HT. Average values of η (efficiency), V
oc
, FF (fill factor), and
J
sc
obtained under simulated AM 1.5G illumination (100 mW/cm
2
) for all polymers are
listed in Table 4.3 and J-V curves are shown in Figure 4.10.

173

Figure 4.10. J-V curves of the BHJ solar cells based on (i) P3HT (purple line), (ii)
P3HT
90
-co-EHT
10
(green line), (iii) P3HT
75
-co-EHT
25
(blue line), (iv) P3HT
50
-co-EHT
50

(red line) and (v) P3EHT (orange line) with PC
61
BM as the acceptor under AM 1.5G
illumination (100 mW/cm
2
) at the optimal conditions for solar cell performance.

The V
oc
of the optimized solar cells increases upon introduction of the 2-
ethylhexyl side-chains into the polymer backbone as shown in Figure 4.11 following the
trend suggested by the decreasing HOMO levels as measured by CV of the thin films.
This seems to indicate that the increase in V
oc
for this family of polymers is primarily
determined by the HOMO
DONOR
-LUMO
ACCEPTOR
offset and less by other effects
suggested in the literature (such as the polymer:fullerene interface interactions
represented by J
so
).
2,87
Even 10% of 2-ethylhexyl side-chains are enough to enhance V
oc

from 0.60 V for P3HT to 0.63 V and this value is even further increased to 0.69, 0.85 and
0.90 V in case of P3HT
75
-co-EHT
25
, P3HT
50
-co-EHT
50
and P3EHT.

174

Figure 4.11. HOMO levels of the polymers in the solid state (filled squares) and V
oc

(circles) of the optimized solar cells as a function of amount of 2-ethylhexyl side-chains
in the polymer backbone.

At the same time the J
sc
values for P3HT, P3HT
90
-co-EHT
10
and P3HT
75
-co-
EHT
25
are high and close to each other in the range of 9.26 – 9.85 mA/cm
2
while a
noticeable drop in J
sc
was recorded for P3HT
50
-co-EHT
50
and P3EHT. A high FF (around
60%) for P3HT

and P3HT
75
-co-EHT
25
suggests a favorable morphology of the
polymer:PC
61
BM blends as well as balanced charge transport in the devices. This
combination of increased V
oc
and high J
sc
as well as a large FF ultimately allows P3HT
75
-
co-EHT
25
:PC
61
BM to exceed the efficiency of P3HT:PC
61
BM (3.85 vs 3.48%). It is
unclear why the fill factor of P3HT
90
-co-EHT
10
, after extensive solar cell optimization,

is
only 51% even though the absorption coefficient, degree of crystallinity, hole mobility
and mixing ratio are comparable to P3HT

and P3HT
75
-co-EHT
25
. Ultimately the lower FF
causes the efficiency of the optimized P3HT
90
-co-EHT
10
device to drop to 2.80% even
though J
sc
is high and the V
oc
is increased compared to P3HT. For P3HT
50
-co-EHT
50
and
P3EHT the combination of a decrease in the absorption coefficient, the degree of
175
crystallinity, and the hole mobility, as well as the unfavourable mixing ratios with PCBM
(less polymer in the active layer) explains the low J
sc
, FF and ultimately the low
efficiencies. Nevertheless the observed trend in V
oc
with increasing 2-ethylhexyl side-
chain content is still maintained.

Figure 4.12. TEM images of polymer:PC
61
BM films (optimized conditions for best solar
cell performance were used to make the films) where (a) is P3HT, (b) is P3HT
90
-co-
EHT
10
, (c) is P3HT
75
-co-EHT
25
, (d) is P3HT
50
-co-EHT
50
and (e) is P3EHT.

For further characterization the BHJ morphology of the solar cells was analyzed
by transmission electron microscopy (TEM) and the recorded images are shown in
176
Figure 4.12. P3HT and P3HT
90
-co-EHT
10
(Figure 4.12a and b) appear to have almost
the same morphology with an overall homogeneous distribution of polymer and PC
61
BM
whereas P3HT
75
-co-EHT
25
and P3HT
50
-co-EHT
50
(Figure 4.12c and d) show a slightly
finer phase separation. Apparently the introduction of small amounts of 2-ethylhexyl
side-chains allows for the good properties of P3HT:PC
61
BM blends, such as the
formation of bicontinuous pathways for charge transport and domain sizes favourable for
exciton splitting, to be retained. In contrast, P3EHT (Figure 4.12e) shows large, (roughly
on the scale of 100 nm), round, dark regions throughout the entire film, which appear to
be PC
61
BM aggregates. These large fullerene aggregates likely prevent efficient exciton
dissociation and hinder charge transport (see Appendix B for larger area films).

4.2.3 Conclusion

The work presented here has been motivated by the desire to demonstrate and
understand the influence of conjugated polymer alkyl side-chains on the performance of
polymer:fullerene BHJ solar cells. Specifically the relationship between alkyl side-chain
identity (length, branching) and distribution (patterns of alkyl-substituted and non-
substituted rings) and the V
oc
is of greatest interest for not only the design of new
polymers, but also for furthering understanding of the mechanism of operation in
polymer:fullerene solar cells. In order to address this incredibly complex issue, we have
developed a model system of random alkylthiophene copolymers in order to logically test
one variable, in particular the influence of branched alkyl chain content (specifically 2-
177
ethylhexyl) on P3ATs. With this model system we have been able to demonstrate a
correlation between increasing branched alkyl chain content, a physical property of the
polymer (decreasing HOMO energy measured in the solid state), and the increasing V
oc

of polymer:fullerene BHJ solar cells. The observed decrease of HOMO levels in the solid
state, and simultaneous increase of the V
oc
upon increasing branched side-chain content is
consistent with some previously reported work on P3ATs and, according to our results,
implies a solid state organization effect of the polymer, as the introduction of even small
amounts of 2-ethylhexyl side-chains leads to a marked decrease of the HOMO levels of
polymers in the solid state, whereas they are virtually identical in solution.
76,77
While the
physical origin of the decreasing HOMO energy in the solid state is not clear, an
important observation is that even with significant branched side-chain content (up to
50%), the decreasing HOMO energy is not accompanied by any change in the polymer
optical band gap. Further, it is clear that, for the polymers studied here, the V
oc
is strongly
correlated to the HOMO
DONOR
-LUMO
ACCEPTOR
offset as the decrease in HOMO levels is
reflected in the measured increase in V
oc
values of the polymer:PC
61
BM solar cells. More
detailed explanations, such as interfacial interactions between the polymer and fullerene,
which other literature reports show to be relevant for increased V
oc
are not necessary to
explain the observed results.
87,88

Further work is clearly needed on the path to developing a comprehensive picture
of alkyl side-chain effects on polymer:fullerene solar cells and specifically on the V
oc
. In
the present study we have not addressed the influence of alkyl chain length and
distribution nor have we addressed the potential differences between polymers with
178
homogenous backbone content (e.g. P3ATs) or copolymers, such as D/A polymers.
However, it is clear that logical model studies of isolated variables, analogous to that
presented here, can provide an important path toward demonstrating the influence of
alkyl chain substitution in conjugated polymers and opening the door to a unified
description and ultimately a detailed fundamental understanding. As a further interesting
consequence of the model study presented here, we have found that the good properties
of P3HT (high absorption coefficient, semi-crystallinity and high hole mobility) and the
optical band gap are essentially retained in P3HT
90
-co-EHT
10
, P3HT
75
-co-EHT
25
and
P3HT
50
-co-EHT
50
. The relatively large increases in the V
oc
of the polymer:fullerene solar
cells with introduction of only small amounts of 2-ethylhexyl chains, with almost no
change in all other properties, suggests that this is also a viable way of tuning the V
oc
in
other polymers as a method for seeking higher efficiencies.

4.3 Chapter Bibliography

(1) Thompson, B. C.; Khlyabich, P. P.; Burkhart, B.; Aviles, A. E.; Rudenko, A.;
Shultz, G. V.; Ng, C. F.; Mangubat, L. B. Green 2011, 1, 29.

(2) Clarke, T. M.; Durrant, J. R. Chem. Rev. 2010, 110, 6736.

(3) Deibel, C.; Dyakonov, V. Rep. Prog. Phys. 2010, 73, 096401.

(4) Kippelen, B.; Brédas, J.-L. Energy Environ. Sci. 2009, 2, 251.

(5) Chen, H.-Y.; Hou, J.; Zhang, S.; Liang, Y.; Yang, G.; Yang, Y.; Yu, L.; Wu, Y.;
Li, G. Nat. Photonics 2009, 3, 649.

(6) Liang, Y.; Xu, Z.; Xia, J.; Tsai, S.-T.; Wu, Y.; Li, G.; Ray, C.; Yu, L. Adv. Mater.
2010, 22, E135.
179
(7) Zhou, H.; Yang, L.; Stuart, A. C.; Price, S. C.; Liu, S.; You, W. Angew. Chem. Int.
Ed. 2011, 50, 2995.

(8) Price, S. C.; Stuart, A. C.; Yang, L.; Zhou, H.; You, W. J. Am. Chem. Soc. 2011,
133, 4625.

(9) Chu, T.-Y.; Lu, J.; Beaupré, S.; Zhang, Y.; Pouliot, J.-R.; Wakim, S.; Zhou, J.;
Leclerc, M.; Li, Z.; Ding, J.; Tao, Y. J. Am. Chem. Soc. 2011, 133, 4250.

(10) Krebs, F. C. Sol. Energy Mater. Sol. Cells 2009, 93, 394.

(11) Avrutin, V.; Izyumskaya, N.; Morkoç, H. Superlattices Microstruct. 2011, 49, 337.

(12) Brabec, C. J.; Gowrisanker, S.; Halls, J. J. M.; Laird, D.; Jia, S.; Williams, S. P.
Adv. Mater. 2010, 22, 3839.

(13) Thompson, B. C.; Fréchet, J. M. J. Angew. Chem. Int. Ed. 2008, 47, 58.

(14) Dennler, G.; Scharber, M. C.; Brabec, C. J. Adv. Mater. 2009, 21, 1323.

(15) Bundgaard, E.; Krebs, F. Sol. Energy Mater. Sol. Cells 2007, 91, 954.

(16) Mihailetchi, V.; Wildeman, J.; Blom, P. Phys. Rev. Lett. 2005, 94, 126602.

(17) Koster, L. J. A.; Mihailetchi, V. D.; Blom, P. W. M. Appl. Phys. Lett. 2006, 88,
052104.

(18) Kotlarski, J. D.; Moet, D. J. D.; Blom, P. W. M. J. Polym. Sci. B Polym. Phys.
2011, 49, 708.

(19) Van Bavel, S.; Veenstra, S.; Loos, J. Macromol. Rapid Commun. 2010, 31, 1835.

(20) Moulé, A. J.; Meerholz, K. Adv. Funct. Mater. 2009, 19, 3028.

(21) Kroon, R.; Lenes, M.; Hummelen, J.; Blom, P.; De Boer, B. Polymer Revs. 2008,
48, 531.

(22) Peet, J.; Kim, J. Y.; Coates, N. E.; Ma, W. L.; Moses, D.; Heeger, A. J.; Bazan, G.
C. Nat. Mater. 2007, 6, 497.

(23) Liang, Y.; Feng, D.; Wu, Y.; Tsai, S.-T.; Li, G.; Ray, C.; Yu, L. J. Am. Chem. Soc.
2009, 131, 7792.

180
(24) Hou, J.; Chen, H.-Y.; Zhang, S.; Chen, R. I.; Yang, Y.; Wu, Y.; Li, G. J. Am.
Chem. Soc. 2009, 131, 15586.

(25) Allard, N.; Aich, R. B.; Gendron, D.; Boudreault, P.-L. T.; Tessier, C.; Alem, S.;
Tse, S.-C.; Tao, Y.; Leclerc, M. Macromolecules 2010, 43, 2328.

(26) Burkhart, B.; Khlyabich, P. P.; Cakir Canak, T.; LaJoie, T. W.; Thompson, B. C.
Macromolecules 2011, 44, 1242.

(27) Kim, M.-S.; Kim, B.-G.; Kim, J. ACS Appl. Mater. Interfaces 2009, 1, 1264.

(28) Van Bavel, S.; Sourty, E.; De With, G.; Frolic, K.; Loos, J. Macromolecules 2009,
42, 7396.

(29) Lenes, M.; Koster, L. J. A.; Mihailetchi, V. D.; Blom, P. W. M. Appl. Phys. Lett.
2006, 88, 243502.

(30) Min Nam, Y.; Huh, J.; Ho Jo, W. Sol. Energy Mater. Sol. Cells 2010, 94, 1118.

(31) Walker, B.; Tamayo, A. B.; Dang, X.-D.; Zalar, P.; Seo, J. H.; Garcia, A.;
Tantiwiwat, M.; Nguyen, T.-Q. Adv. Funct. Mater. 2009, 19, 3063.

(32) Bijleveld, J. C.; Gevaerts, V. S.; Di Nuzzo, D.; Turbiez, M.; Mathijssen, S. G. J.;
De Leeuw, D. M.; Wienk, M. M.; Janssen, R. A. J. Adv. Mater. 2010, 22, E242.

(33) Woo, C. H.; Beaujuge, P. M.; Holcombe, T. W.; Lee, O. P.; Fréchet, J. M. J. J. Am.
Chem. Soc. 2010, 132, 15547.

(34) Bronstein, H.; Chen, Z.; Ashraf, R. S.; Zhang, W.; Du, J.; Durrant, J. R.; Shakya
Tuladhar, P.; Song, K.; Watkins, S. E.; Geerts, Y.; Wienk, M. M.; Janssen, R. A.
J.; Anthopoulos, T.; Sirringhaus, H.; Heeney, M.; McCulloch, I. J. Am. Chem. Soc.
2011, 133, 3272.

(35) Tieke, B.; Rabindranath, A. R.; Zhang, K.; Zhu, Y. Beilstein J. Org. Chem. 2010,
6, 830.

(36) Zou, Y.; Gendron, D.; Neagu-Plesu, R.; Leclerc, M. Macromolecules 2009, 42,
6361.

(37) Bijleveld, J. C.; Karsten, B. P.; Mathijssen, S. G. J.; Wienk, M. M.; De Leeuw, D.
M.; Janssen, R. A. J. J. Mater. Chem. 2011, 21, 1600.

(38) Huo, L.; Hou, J.; Chen, H.-Y.; Zhang, S.; Jiang, Y.; Chen, T. L.; Yang, Y.
Macromolecules 2009, 42, 6564.
181
(39) Zoombelt, A. P.; Mathijssen, S. G. J.; Turbiez, M. G. R.; Wienk, M. M.; Janssen,
R. A. J. J. Mater. Chem. 2010, 20, 2240.

(40) Zhou, E.; Yamakawa, S.; Tajima, K.; Yang, C.; Hashimoto, K. Chem. Mater.
2009, 21, 4055.

(41) Zhou, E.; Wei, Q.; Yamakawa, S.; Zhang, Y.; Tajima, K.; Yang, C.; Hashimoto, K.
Macromolecules 2010, 43, 821.

(42) Wienk, M. M.; Turbiez, M.; Gilot, J.; Janssen, R. A. J. Adv. Mater. 2008, 20, 2556.

(43) Bijleveld, J. C.; Zoombelt, A. P.; Mathijssen, S. G. J.; Wienk, M. M.; Turbiez, M.;
De Leeuw, D. M.; Janssen, R. A. J. J. Am. Chem. Soc. 2009, 131, 16616.

(44) Badrou Aïch, R.; Zou, Y.; Leclerc, M.; Tao, Y. Org. Electron. 2010, 11, 1053.

(45) Tamayo, A. B.; Walker, B.; Nguyen, T.-Q. J. Phys. Chem. C 2008, 112, 11545.

(46) Zhu, Y.; Champion, R. D.; Jenekhe, S. A. Macromolecules 2006, 39, 8712.

(47) Beaujuge, P. M.; Amb, C. M.; Reynolds, J. R. Acc. Chem. Res. 2010, 43, 1396.

(48) Kim, Y.; Cook, S.; Tuladhar, S. M.; Choulis, S. A.; Nelson, J.; Durrant, J. R.;
Bradley, D. D. C.; Giles, M.; McCulloch, I.; Ha, C.-S.; Ree, M. Nat. Mater. 2006,
5, 197.

(49) Jo, J.; Gendron, D.; Najari, A.; Moon, J. S.; Cho, S.; Leclerc, M.; Heeger, A. J.
Appl. Phys. Lett. 2010, 97, 203303.

(50) Thompson, B. C.; Kim, Y.-G.; McCarley, T. D.; Reynolds, J. R. J. Am. Chem. Soc.
2006, 128, 12714.

(51) Cardona, C. M.; Li, W.; Kaifer, A. E.; Stockdale, D.; Bazan, G. C. Adv. Mater.
2011, 23, 2367.

(52) Thompson, B. C.; Kim, Y.-G.; Reynolds, J. R. Macromolecules 2005, 38, 5359.

(53) Jørgensen, M.; Norrman, K.; Krebs, F. C. Sol. Energy Mater. Sol. Cells 2008, 92,
686.

(54) Shrotriya, V.; Li, G.; Yao, Y.; Moriarty, T.; Emery, K.; Yang, Y. Adv. Funct.
Mater. 2006, 16, 2016.

182
(55) Kim, B.-G.; Jeong, E. J.; Park, H. J.; Bilby, D.; Guo, L. J.; Kim, J. ACS Appl.
Mater. Interfaces 2011, 3, 674.

(56) Maurano, A.; Hamilton, R.; Shuttle, C. G.; Ballantyne, A. M.; Nelson, J.;
O’Regan, B.; Zhang, W.; McCulloch, I.; Azimi, H.; Morana, M.; Brabec, C. J.;
Durrant, J. R. Adv. Mater. 2010, 22, 4987.

(57) Deibel, C.; Strobel, T.; Dyakonov, V. Adv. Mater. 2010, 22, 4097.

(58) Schlenker, C. W.; Thompson, M. E. Chem. Commun. 2011, 47, 3702.

(59) Gendron, D.; Leclerc, M. Energy Environ. Sci. 2011, 4, 1225.

(60) Chochos, C. L.; Choulis, S. A. Prog. Polym. Sci. 2011, 36, 1326.

(61) Amb, C. M.; Chen, S.; Graham, K. R.; Subbiah, J.; Small, C. E.; So, F.; Reynolds,
J. R. J. Am. Chem. Soc. 2011, 133, 10062.

(62) Service, R. F. Science 2011, 332, 293.

(63) He, Z.; Zhong, C.; Huang, X.; Wong, W.-Y.; Wu, H.; Chen, L.; Su, S.; Cao, Y.
Adv. Mater. 2011, 23, 4636.

(64) Dang, M. T.; Hirsch, L.; Wantz, G. Adv. Mater. 2011, 23, 3597.

(65) Agostinelli, T.; Lilliu, S.; Labram, J. G.; Campoy-Quiles, M.; Hampton, M.; Pires,
E.; Rawle, J.; Bikondoa, O.; Bradley, D. D. C.; Anthopoulos, T. D.; Nelson, J.;
Macdonald, J. E. Adv. Funct. Mater. 2011, 21, 1701.

(66) Treat, N. D.; Shuttle, C. G.; Toney, M. F.; Hawker, C. J.; Chabinyc, M. L. J.
Mater. Chem. 2011, 21, 15224.

(67) Treat, N. D.; Brady, M. A.; Smith, G.; Toney, M. F.; Kramer, E. J.; Hawker, C. J.;
Chabinyc, M. L. Adv. Energy Mater. 2011, 1, 82.

(68) Snyder, C. R.; Henry, J. S.; DeLongchamp, D. M. Macromolecules 2011, 44, 7088.

(69) Thompson, B. C.; Kim, B. J.; Kavulak, D. F.; Sivula, K.; Mauldin, C.; Fréchet, J.
M. J. Macromolecules 2007, 40, 7425.

(70) Szarko, J. M.; Guo, J.; Liang, Y.; Lee, B.; Rolczynski, B. S.; Strzalka, J.; Xu, T.;
Loser, S.; Marks, T. J.; Yu, L.; Chen, L. X. Adv. Mater. 2010, 22, 5468.

(71) Wu, P.-T.; Ren, G.; Jenekhe, S. A. Macromolecules 2010, 43, 3306.
183
(72) Osaka, I.; McCullough, R. D. Acc. Chem. Res. 2008, 41, 1202.

(73) Mauer, R.; Kastler, M.; Laquai, F. Adv. Funct. Mater. 2010, 20, 2085.

(74) Nguyen, L. H.; Hoppe, H.; Erb, T.; Günes, S.; Gobsch, G.; Sariciftci, N. S. Adv.
Funct. Mater. 2007, 17, 1071.

(75) Kline, R. J.; DeLongchamp, D. M.; Fischer, D. A.; Lin, E. K.; Richter, L. J.;
Chabinyc, M. L.; Toney, M. F.; Heeney, M.; McCulloch, I. Macromolecules 2007,
40, 7960.

(76) Al-Ibrahim, M.; Roth, H.-K.; Schroedner, M.; Konkin, A.; Zhokhavets, U.;
Gobsch, G.; Scharff, P.; Sensfuss, S. Org. Electron. 2005, 6, 65.

(77) Gadisa, A.; Oosterbaan, W. D.; Vandewal, K.; Bolsée, J.-C.; Bertho, S.; D’Haen,
J.; Lutsen, L.; Vanderzande, D.; Manca, J. V. Adv. Funct. Mater. 2009, 19, 3300.

(78) Friedel, B.; McNeill, C. R.; Greenham, N. C. Chem. Mater. 2010, 22, 3389.

(79) Stevens, D. M.; Speros, J. C.; Hillmyer, M. A.; Frisbie, C. D. J. Phys. Chem. C
2011, 115, 20806.

(80) Hou, J.; Chen, T. L.; Zhang, S.; Huo, L.; Sista, S.; Yang, Y. Macromolecules 2009,
42, 9217.

(81) Ko, S.; Verploegen, E.; Hong, S.; Mondal, R.; Hoke, E. T.; Toney, M. F.;
McGehee, M. D.; Bao, Z. J. Am. Chem. Soc. 2011, 133, 16722.

(82) Zoombelt, A. P.; Leenen, M. A. M.; Fonrodona, M.; Nicolas, Y.; Wienk, M. M.;
Janssen, R. A. J. Polymer 2009, 50, 4564.

(83) Wang, E.; Hou, L.; Wang, Z.; Ma, Z.; Hellström, S.; Zhuang, W.; Zhang, F.;
Ingana ̈s, O.; Andersson, M. R. Macromolecules 2011, 44, 2067.

(84) Biniek, L.; Fall, S.; Chochos, C. L.; Anokhin, D. V.; Ivanov, D. A.; Leclerc, N.;
Léve ̂que, P.; Heiser, T. Macromolecules 2010, 43, 9779.

(85) Li, Z.; Zhang, Y.; Tsang, S.-W.; Du, X.; Zhou, J.; Tao, Y.; Ding, J. J. Phys. Chem.
C 2011, 115, 18002.

(86) Li, W.; Qin, R.; Zhou, Y.; Andersson, M.; Li, F.; Zhang, C.; Li, B.; Liu, Z.; Bo, Z.;
Zhang, F. Polymer 2010, 51, 3031.

(87) Yang, L.; Zhou, H.; You, W. J. Phys. Chem. C 2010, 114, 16793.
184
(88) Perez, M. D.; Borek, C.; Forrest, S. R.; Thompson, M. E. J. Am. Chem. Soc. 2009,
131, 9281.

(89) Woo, C. H.; Thompson, B. C.; Kim, B. J.; Toney, M. F.; Fréchet, J. M. J. J. Am.
Chem. Soc. 2008, 130, 16324.

(90) Sivula, K.; Luscombe, C. K.; Thompson, B. C.; Fréchet, J. M. J. J. Am. Chem. Soc.
2006, 128, 13988.

(91) Somanathan, N.; Radhakrishnan, S.; Thelakkat, M.; Schmidt, H. W. Macromol.
Mater. Eng. 2002, 287, 236.

(92) Zhang, Y.; Tajima, K.; Hashimoto, K. Macromolecules 2009, 42, 7008.

(93) Zhang, Y.; Tajima, K.; Hirota, K.; Hashimoto, K. J. Am. Chem. Soc. 2008, 130,
7812.

(94) Zhang, Y.; Tajima, K.; Hashimoto, K. Synth. Met. 2011, 161, 225.

(95) Ho, V.; Boudouris, B. W.; Segalman, R. A. Macromolecules 2010, 43, 7895.

(96) Khlyabich, P. P.; Burkhart, B.; Ng, C. F.; Thompson, B. C. Macromolecules 2011,
44, 5079.

(97) Boudouris, B. W.; Ho, V.; Jimison, L. H.; Toney, M. F.; Salleo, A.; Segalman, R.
A. Macromolecules 2011, 44, 6653.

185
Chapter 5

Compositional Dependence of the Open-Circuit Voltage in Ternary Blend Bulk
Heterojunction Solar Cells Based on Two Donor Polymers

5.1 Introduction

Ternary blend bulk heterojunction (BHJ) solar cells, containing two P3HT
analogues as donor polymers, high bandgap poly(3-hexylthiophene-co-3-(2-
ethylhexyl)thiophene) (P3HT
75
-co-EHT
25
) and low bandgap poly(3-hexylthiophene-
thiophene-diketopyrrolopyrrole) (P3HTT-DPP-10%), with phenyl-C
61
-butyric acid
methyl ester (PC
61
BM) as an acceptor, were studied. Upon variation of the ratio of the
three components, the open-circuit voltage (V
oc
) was found to increase as the amount of
the P3HT
75
-co-EHT
25
was increased. The V
oc
dependence on polymer composition for
the ternary blend regime was linear when the overall polymer:fullerene ratio was
optimized for each polymer:polymer ratio. Additionally, short-circuit current densities
(J
sc
) for the ternary blends were improved in comparison to the binary blends due to
complimentary polymer absorption, verified using external quantum efficiency (EQE)
measurements. High fill factors (FF) in all cases (above 0.59) were achieved and
attributed to high charge-carrier mobilities in the ternary blends. As a result of the
intermediate V
oc
, increased J
sc
and high FF, ternary blend BHJ solar cells showed power
conversion efficiencies up to 5.51%, exceeding the corresponding binary blends (3.16
186
and 5.07%). Importantly, this work demonstrates that upon optimization of the overall
polymer:fullerene ratio at each polymer:polymer ratio, high FF, regular variations in the
V
oc
, and enhanced J
sc
are possible throughout the ternary blend composition regime. This
result supports growing evidence that ternary blends are a general and effective strategy
for efficient organic photovoltaics manufactured in a single active layer processing step.

5.2 Results and Discussion

Extensive research on bulk heterojunction (BHJ) solar cells in the last decade
resulted in deep understanding of the operating principles of binary blend solar cells
composed of a polymeric donor and a fullerene acceptor.
1–6
Numerous polymers were
synthesized
7,8
and efficiencies exceeding 8% have been achieved.
9–11
As an approach to
further increase power conversion efficiencies, tandem solar cells, where two (or more)
sub-cells absorbing light in different regions of the solar spectrum are connected either in
series or parallel, were studied.
12–14
Despite the increase of either the open-circuit voltage
(V
oc
)
14
or short-circuit current density (J
sc
),
13
only slightly higher efficiencies have been
obtained with tandem cells.
15
In addition, tandem cells require more complex design and
fabrication and cannot be manufactured in a single active layer processing step,
13,14
in
contrast to the desire for inexpensive solar cells.
4
Importantly, the efficiencies (η)
achieved with the described strategies are approaching their theoretical limits of ~10 –
12% for binary solar cells
1,16–19
and around 14 – 15% for tandem solar cells with two
absorbing layers.
13,14,17

187
Recently, we demonstrated for the first time that the V
oc
of ternary blend BHJ
solar cells based on one donor and two acceptors is composition dependent and can be
tuned across the full range defined by the corresponding limiting binary blends without
negatively impacting the fill factor (FF) or the J
sc
.
20
This finding opened the door towards
BHJ solar cells with the potential for exceeding the efficiency limit, even for tandem
cells, but fabricated in a single active layer processing step. Following our initial
discovery of the potential of ternary blend BHJ solar cells and our prediction of increased
efficiency with judicious choice of donor polymers and a fullerene acceptor,
20
You et al.
recently demonstrated that blends of two donor polymers and a fullerene acceptor, at
constant overall polymer:fullerene ratio, could exceed the efficiency of the limiting
binary blends via increase of J
sc
and modulation of V
oc
.
21
However, a vast number of
fundamental questions remain to be answered, as there is no clear set of structure-
function relationships for component selection and the mechanism of operation is not
established.
Here, our focus is to investigate the dependence of the V
oc
on composition in a
ternary blend system based on two donor polymers and a fullerene acceptor where the
limiting binary blends do not give optimal performance at the same polymer:fullerene
ratio. The focus is thus also to demonstrate evidence for the generality of this ternary
blend approach. The model ternary blend system (Figure 5.1) of two donor polymers,
high bandgap poly(3-hexylthiophene-co-3-(2-ethylhexyl)thiophene) (P3HT
75
-co-
EHT
25
)
22
and low bandgap poly(3-hexylthiophene-thiophene-diketopyrrolopyrrole)
(P3HTT-DPP-10%),
23
and phenyl-C
61
-butyric acid methyl ester (PC
61
BM) as an acceptor
188
is used. In this work, the overall polymer:fullerene ratio of the ternary blend BHJ solar
cells is individually optimized at each polymer:polymer ratio. In this case, we observe
that even the smallest amount of the second polymer in the ternary blend has a large
effect on the V
oc
, which evolves linearly with composition across the ternary blend
regime. In cases where the overall polymer:fullerene ratio is not individually optimized at
each polymer:polymer ratio, a significant decrease of the V
oc
and deviation from linearity
in the ternary blend composition regime was observed, which is accompanied by a
decrease of J
sc
, FF and efficiency.

Figure 5.1. Structures, corresponding HOMO energy levels and absorption profiles of
PC
61
BM (black line), P3HT
75
-co-EHT
25
(red line) and P3HTT-DPP-10% (green line).

The choice of donor polymers was dictated by the desire to simultaneously
increase J
sc
, through complementary absorption,
20
and have an intermediate V
oc
in ternary
189
blends. As can be seen in Figure 5.1, P3HTT-DPP-10%, P3HT
75
-co-EHT
25
and PC
61
BM
provide broad and uniformly strong absorption from 300 to 830 nm, which should
facilitate J
sc
increase with respect to either binary blend solar cell. At the same time,
different HOMO energy levels of 5.4 eV for P3HT
75
-co-EHT
25
22
and 5.2 eV for P3HTT-
DPP-10%
23
enables V
oc
tunability in the ternary blend solar cells at different polymer
ratios. Furthermore, high efficiencies were obtained in both binary blend solar cells.
22,23

Importantly, both the semi-random
23,24
P3HTT-DPP-10% and the random P3HT-co-
EHT
22
are P3HT analogues containing 80% and 75% of 3-hexylthiophene repeat units,
respectively. Copolymerization allowed us to tune the bandgap of P3HTT-DPP-10% and
HOMO energy of P3HT
75
-co-EHT
25
relative to P3HT in order to give a complementary
set of properties ideal for analyzing ternary blend behavior in the context of a
“P3HT:PCBM” model system, which allows the use of similar processing conditions at
each composition.
Photovoltaic devices containing ternary blends in a conventional device
configuration ITO/PEDOT:PSS/P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM/Al were
fabricated in air. Ternary blend BHJ solar cells were optimized at each polymer:polymer
ratio to obtain the highest efficiencies. Optimal processing conditions include slow
solvent evaporation (solvent annealing) from the P3HTT-DPP-10%:P3HT
75
-co-
EHT
25
:PC
61
BM blends after spin-coating and prior to aluminum deposition. Since ternary
blend solar cells contain more variables that can be tuned the optimization of the three
component system included increase in the number of cells needed to achieve optimal
device performance while keeping the same optimization steps as for the binary blend
190
solar cells.
25
Table 5.1 lists the optimized average values of J
sc
, V
oc
, FF and η obtained
under simulated AM 1.5G illumination (100 mW/cm
2
) as the ratio of P3HTT-DPP-10%
to P3HT
75
-co-EHT
25
was varied. As can be seen in Table 5.1, the overall
polymer:fullerene ratio of the individually optimized ternary blend solar cells tracks
regularly with the P3HTT-DPP-10%:P3HT
75
-co-EHT
25
ratio. Efficiencies of the ternary
blends are noticeably increased by 0.44% and 0.30% going from 5.07% with P3HTT-
DPP-10%:PC
61
BM to 5.51% and 5.37% with 0.9:0.1:1.1 and 0.8:0.2:1.0 ratios,
respectively. Furthermore, at all other three component combinations, except 0.1:0.9:0.9
(where the J
sc
increase is smaller than the V
oc
decrease, relative to 0:1:0.8), power
conversion efficiencies are higher than for the binary P3HT
75
-co-EHT
25
:PC
61
BM solar
cell. This observation supports ternary blend BHJ solar cells as an effective way to
overcome the efficiencies of the corresponding binary blend solar cells.

Table 5.1. Photovoltaic properties of P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM
ternary blend BHJ solar cells at optimized ratios
P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM J
sc
(mA/cm
2
) V
oc
(V) FF η (%)
1:0:1.3
a
14.38 0.574 0.62 5.07
0.9:0.1:1.1
b
15.05 0.603 0.61 5.51
0.8:0.2:1.0
b
14.60 0.608 0.61 5.37
0.7:0.3:1.0
c
11.54 0.614 0.59 4.15
0.6:0.4:1.0
c
11.19 0.619 0.59 4.12
0.5:0.5:0.9
a
10.89 0.622 0.59 4.00
0.4:0.6:0.9
a
10.19 0.626 0.59 3.74
0.3:0.7:0.8
a
9.77 0.633 0.59 3.64
0.2:0.8:0.8
a
8.57 0.639 0.60 3.27
0.1:0.9:0.9
a
8.25 0.646 0.59 3.10
0:1:0.8
d
7.96 0.675 0.59 3.16
All devices were spin-coated from o-dichlorobenzene (o-DCB) and placed to the N
2
cabinet before
aluminum deposition for
a
30 min,
b
60 min,
c
45 min and
d
20 min.

191
Another interesting observation from Table 5.1, also illustrated in Figure 5.2, is
that the V
oc
in the individually optimized ternary blends is composition dependent, as was
shown for the case of one donor and two acceptors in earlier papers.
20,26
A noticeable
difference from the previous results is the specific evolution of the V
oc
. Upon introduction
of the second polymer component into either limiting polymer:fullerene binary blend, the
V
oc
rapidly changes by 29 mV in both cases. After this, for the ternary blend composition
regime, the V
oc
monotonically increases from 0.603 V to 0.646 V as the amount of
P3HT
75
-co-EHT
25
(with lower-lying HOMO) is increased. On the other hand, as can be
seen in Figure 5.3, in the case where the overall polymer:fullerene ratio is not
individually optimized at each polymer:polymer ratio, but rather held constant, a
significant decrease of the V
oc
and deviation from linearity in the ternary blend
composition regime is observed. In the case of fixed overall polymer:PC
61
BM ratio at
1:1.1 in the ternary blends, linear behavior of the V
oc
on the composition is observed,
expect for the case of 10% of P3HT
75
-co-EHT
25
, which is an optimal overall
polymer:fullerene ratio. Across the remaining composition range, the V
oc
values are
significantly lower than for individually optimized ternary blends. Non-linear behavior is
observed for the V
oc
compositional dependence in ternary blends for the fixed overall
polymer:PC
61
BM ratio of 1:1.0. Here, linear behavior at P3HT
75
-co-EHT
25
loadings
below 50% is followed by a saturation regime with almost constant V
oc
up to 80%, before
an increase at higher P3HT
75
-co-EHT
25
loadings. Importantly, all cases where the overall
polymer:fullerene ratio is not individually optimized at each polymer:polymer ratio are
accompanied by a decrease of J
sc
, FF and efficiency (see Appendix C). This implies that
192
individual optimization of the overall composition at each polymer:polymer ratio is
necessary for achieving maximum efficiency in ternary blend solar cells and upon the
introduction of a donor polymer with lower-lying HOMO, a significant increase in the
V
oc
can be achieved even at small amounts.

Figure 5.2. Open-circuit voltage (V
oc
) (black squares – left axis) and short-circuit current
density (J
sc
) (red circles – right axis) of the individually optimized ternary blend BHJ
solar cells from Table 5.1 with different fraction of the polymer P3HT
75
-co-EHT
25

component in the blends.

Figure 5.3. Open-circuit voltage (V
oc
) of the individually optimized ternary blend BHJ
solar cells (open squares), with fixed overall polymer:PC
61
BM ratio at 1:1.1 (blue stars)
and with fixed overall polymer:PC
61
BM ratio at 1:1.0 (green triangles).

193
In contrast to the V
oc
trend observed in Table 5.1, J
sc
is found first to increase
with increasing P3HT
75
-co-EHT
25
content and then decrease (Figure 5.2), but still
remain higher than the P3HT
75
-co-EHT
25
:PC
61
BM solar cells in all cases. This is
explained based on the complementary absorption of the P3HTT-DPP-10%:P3HT
75
-co-
EHT
25
:PC
61
BM blend as shown in Figure 5.1. Addition of 10 and 20% of high bandgap
P3HT
75
-co-EHT
25
in the ternary blends leads to an increase of the absorption coefficient
in the 400 – 600 nm region (see Appendix C), while the intensity of the long wavelength
absorption, attributed to the P3HTT-DPP-10%, remains almost the same. Moreover, the
position of the peak in the high energy region red-shifts as the amount of the P3HT
75
-co-
EHT
25
is increased, while the position of the long wavelength peak remains the same. As
a result, the number of absorbed photons is increased and J
sc
is increased. Upon addition
of more than 30% of P3HT
75
-co-EHT
25
, the absorption coefficients in the long
wavelength region start to decrease dramatically, while significant absorption increase in
the visible is observed. Therefore, fewer low energy photons are absorbed and a decrease
in J
sc
is observed.
To further investigate the origin of the J
sc
changes in the ternary blend solar cells,
external quantum efficiency (EQE) measurements were performed. As can be seen in
Figure 5.4, P3HTT-DPP-10%:PC
61
BM and P3HT
75
-co-EHT
25
:PC
61
BM binary blend
solar cells have strong photoresponse in the 400 – 800 nm and 400 – 650 nm wavelength
regions, respectively. Addition of 10 and 20% of the P3HT
75
-co-EHT
25
to P3HTT-DPP-
10%:PC
61
BM increases the EQE values in the visible region of the solar spectrum while
keeping the long wavelength photoresponse from the P3HTT-DPP-10% unchanged.
194
Therefore, an enhanced J
sc
was recorded for 0.9:0.1:1.1 and 0.8:0.2:1.0 ratios. All device
thicknesses were optimized in order to show best overall efficiencies and FF. This
required film thicknesses of 75 nm for binary blend solar cells to 85 – 90 for ternary
blend solar cells. Consistent with the absorption profiles, further increase of the P3HT
75
-
co-EHT
25
content in the ternary blends leads to the decrease of the photocurrent
generated by the P3HTT-DPP-10% and thus a decrease in J
sc
.

Figure 5.4. External quantum efficiency of ternary blend BHJ solar cells where (i) is
1:0:1.3 (red line), (ii) is 0.9:0.1:1.1 (green line), (iii) is 0.8:0.2:1.0 (blue line), (iv) is
0.7:0.3:1.0 (cyan line), (v) is 0.6:0.4:1.0 (magenta line), (vi) is 0.5:0.5:0.9 (wine-red line),
(vii) is 0.4:0.6:0.9 (olive line), (viii) is 0.3:0.7:0.8 (dark yellow line), (ix) is 0.2:0.8:0.8
(purple line), (x) is 0.1:0.9:0.9 (yellow line) and (xi) is 0:1:0.8 (black line).

To gain deeper insight, the morphology was also investigated. Both polymers are
semi-crystalline with interchain distances (100) of 15.07 Å and 16.72 Å for P3HTT-DPP-
10% and P3HT
75
-co-EHT
25
, respectively (see Appendix C). In binary blends, the
polymers retain their semi-crystalline nature as can be seen from the presence of the
vibronic shoulders in the UV-vis,
27
and in the grazing-incidence X-ray diffraction
195
(GIXRD) measurements illustrated in Figure 5.5. In the ternary blends, both polymers
remain semi-crystalline and two peaks are present for the 0.8:0.2:1.0 ratio in the GIXRD
profile (Figure 5.5 – (iii)). Furthermore, vibronic shoulders for both polymers are present
in the UV-vis absorption for the P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM ternary
blends. The interchain distances for the P3HTT-DPP-10% and P3HT
75
-co-EHT
25

domains change in the opposite directions as the amount of the P3HT
75
-co-EHT
25
in the
ternary blends increases. The interchain distance for the P3HTT-DPP-10% increased
from 15.27 Å to 15.41 Å going from 1:0:1.3 to 0.8:0.2:1.0, while P3HT
75
-co-EHT
25

packs tighter with interchain distances decreasing from 16.69 Å to 16.02 Å for 0:1:0.8
and 0.8:0.2:1.0. The ability of both polymers to remain semi-crystalline and pack closer
in the ternary blends should be a contributing factor for the high J
sc
observed in the
ternary blend solar cells.

Figure 5.5. Grazing-incidence X-ray diffraction of thin films where (i) is 1:0:1.3 (red
line), (ii) is 0.9:0.1:1.1 (green line), (iii) is 0.8:0.2:1.0 (blue line), (iv) is 0.7:0.3:1.0 (cyan
line), (vi) is 0.5:0.5:0.9 (wine-red line), (viii) is 0.3:0.7:0.8 (dark yellow line), (x) is
0.1:0.9:0.9 (yellow line) and (xi) is 0:1:0.8 (black line).

196
Finally, high FF values above 0.59 were observed for all optimized ternary blend
ratios (Table 5.1). This can be attributed to a balanced and trap free charge transport
through the bulk
28
and favorable morphology.
29,30
Hole mobilities for the binary and
ternary blends are in the range from 1.11*10
-3
to 3.23*10
-3
cm
2
/(V*s), increasing with
greater amount of P3HTT-DPP-10% (see Appendix C). These values are close to
literature values for solvent annealed P3HT:PC
61
BM binary blends.
31
High values can be
also attributed to the ability of both polymers to retain their semi-crystalline nature in the
ternary blends. Furthermore, high FF and hole mobilities indicate inefficient trapping of
charges and as a result no increase in recombination in the system containing polymer
with shallower-lying HOMO level. Transmission electron microscopy (TEM) images
(see Appendix C) show similar bicontinious blends with nanometer length-scale phase
separation at different P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM ratios. As a result,
introduction of the third component in the blend does not change the overall
polymer:fullerene morphology, suggesting no negative impact on charge transport
through the ternary blends. This however emphasizes the importance of investigating the
morphology of ternary blends.
32

5.3 Conclusion

In summary, we have fabricated ternary blend BHJ solar cells containing two
donor polymers, P3HT
75
-co-EHT
25
and P3HTT-DPP-10%, and PC
61
BM as an acceptor.
The overall polymer:fullerene ratio of the ternary blend BHJ solar cells was individually
197
optimized at each polymer:polymer ratio. This individual optimization is necessary in
order to obtain a simultaneous increase in J
sc
, intermediate V
oc
, high FF and high
efficiency. Furthermore, we observe that introduction of even the smallest amount of the
second polymer into either limiting polymer:fullerene binary blend has a large effect on
the V
oc
, and for the present system, the V
oc
evolves linearly with composition across the
ternary blend regime, for the optimized case. In the case where the overall
polymer:fullerene ratio is not individually optimized at each polymer:polymer ratio, a
significant decrease in V
oc
and deviation from linearity in the ternary blend composition
regime is observed, which is accompanied with the decrease of J
sc
, FF and efficiency.
Overall, the results of this work support the growing evidence that ternary blend BHJ
solar cells are an effective strategy toward more efficient organic photovoltaics
manufactured in a single active layer processing step with the possibility to exceed power
conversion efficiency limits for binary blend solar cells. The results presented here also
demonstrate that judicious choice of paired components with smaller bandgaps and
lower-lying HOMO energy levels must be accompanied by careful optimization of film
composition and processing if the potentially paradigm-shifting nature of this platform is
to be realized.

5.4 Chapter Bibliography

(1) Thompson, B. C.; Khlyabich, P. P.; Burkhart, B.; Aviles, A. E.; Rudenko, A.;
Shultz, G. V.; Ng, C. F.; Mangubat, L. B. Green 2011, 1, 29.

(2) Schlenker, C. W.; Thompson, M. E. Chem. Commun. 2011, 47, 3702.
198
(3) Clarke, T. M.; Durrant, J. R. Chem. Rev. 2010, 110, 6736.

(4) Deibel, C.; Dyakonov, V. Rep. Prog. Phys. 2010, 73, 096401.

(5) Brédas, J.-L.; Norton, J. E.; Cornil, J.; Coropceanu, V. Acc. Chem. Res. 2009, 42,
1691.

(6) Beaujuge, P. M.; Fréchet, J. M. J. J. Am. Chem. Soc. 2011, 133, 20009.

(7) Chochos, C. L.; Choulis, S. A. Prog. Polym. Sci. 2011, 36, 1326.

(8) Boudreault, P.-L. T.; Najari, A.; Leclerc, M. Chem. Mater. 2011, 23, 456.

(9) Small, C. E.; Chen, S.; Subbiah, J.; Amb, C. M.; Tsang, S.-W.; Lai, T.-H.;
Reynolds, J. R.; So, F. Nat. Photonics 2011, 6, 115.

(10) He, Z.; Zhong, C.; Huang, X.; Wong, W.-Y.; Wu, H.; Chen, L.; Su, S.; Cao, Y.
Adv. Mater. 2011, 23, 4636.

(11) Green, M. A.; Emery, K.; Hishikawa, Y.; Warta, W.; Dunlop, E. D. Prog.
Photovolt: Res. Appl. 2011, 19, 565.

(12) Sista, S.; Hong, Z.; Chen, L.-M.; Yang, Y. Energy Environ. Sci. 2011, 4, 1606.

(13) Siddiki, M. K.; Li, J.; Galipeau, D.; Qiao, Q. Energy Environ. Sci. 2010, 3, 867.

(14) Ameri, T.; Dennler, G.; Lungenschmied, C.; Brabec, C. J. Energy Environ. Sci.
2009, 2, 347.

(15) Dou, L.; You, J.; Yang, J.; Chen, C.-C.; He, Y.; Murase, S.; Moriarty, T.; Emery,
K.; Li, G.; Yang, Y. Nat. Photonics 2012, 6, 180.

(16) Nayak, P. K.; Garcia-Belmonte, G.; Kahn, A.; Bisquert, J.; Cahen, D. Energy
Environ. Sci. 2012, 5, 6022.

(17) Kotlarski, J. D.; Blom, P. W. M. Appl. Phys. Lett. 2011, 98, 053301.

(18) Veldman, D.; Meskers, S. C. J.; Janssen, R. A. J. Adv. Funct. Mater. 2009, 19,
1939.

(19) Kippelen, B.; Brédas, J.-L. Energy Environ. Sci. 2009, 2, 251.

(20) Khlyabich, P. P.; Burkhart, B.; Thompson, B. C. J. Am. Chem. Soc. 2011, 133,
14534.
199
(21) Yang, L.; Zhou, H.; Price, S. C.; You, W. J. Am. Chem. Soc. 2012, 134, 5432.

(22) Burkhart, B.; Khlyabich, P. P.; Thompson, B. C. Macromolecules 2012, 45, 3740.

(23) Khlyabich, P. P.; Burkhart, B.; Ng, C. F.; Thompson, B. C. Macromolecules 2011,
44, 5079.

(24) Burkhart, B.; Khlyabich, P. P.; Cakir Canak, T.; LaJoie, T. W.; Thompson, B. C.
Macromolecules 2011, 44, 1242.

(25) Alstrup, J.; Jørgensen, M.; Medford, A. J.; Krebs, F. C. ACS Appl. Mater.
Interfaces 2010, 2, 2819.

(26) Jeltsch, K. F.; Schädel, M.; Bonekamp, J.-B.; Niyamakom, P.; Rauscher, F.;
Lademann, H. W. A.; Dumsch, I.; Allard, S.; Scherf, U.; Meerholz, K. Adv. Funct.
Mater. 2012, 22, 397.

(27) Gurau, M. C.; Delongchamp, D. M.; Vogel, B. M.; Lin, E. K.; Fischer, D. A.;
Sambasivan, S.; Richter, L. J. Langmuir 2007, 23, 834.

(28) Kotlarski, J. D.; Moet, D. J. D.; Blom, P. W. M. J. Polym. Sci., Part B: Polym.
Phys. 2011, 49, 708.

(29) Brabec, C. J.; Heeney, M.; McCulloch, I.; Nelson, J. Chem. Soc. Rev. 2011, 40,
1185.

(30) Van Bavel, S.; Veenstra, S.; Loos, J. Macromol. Rapid Commun. 2010, 31, 1835.

(31) Shrotriya, V.; Yao, Y.; Li, G.; Yang, Y. Appl. Phys. Lett. 2006, 89, 063505.

(32) Li, N.; Machui, F.; Waller, D.; Koppe, M.; Brabec, C. J. Sol. Energy Mater. Sol.
Cells 2011, 95, 3465.

200
Chapter 6

Origin of the Tunable Open-Circuit Voltage in Ternary Blend Bulk Heterojunction
Organic Solar Cells

6.1 Introduction

Ternary blend bulk heterojunction organic solar cells comprising either a
polythiophene donor and two fullerene acceptors or two polythiophene donors and a
fullerene acceptor are shown to have unique electronic properties. Measurements of the
photocurrent spectral response and the open-circuit voltage show that the HOMO and
LUMO levels change continuously with composition in the respective two-component
acceptor or donor pair, consistent with the formation of an organic alloy. However,
optical absorption of the exciton states retains the individual molecular properties of the
two materials across the blend composition. This difference is attributed to the highly
localized molecular nature of the exciton, and the more delocalized intermolecular nature
of electrons and holes that reflect the average composition of the alloy. As established
here, the combination of molecular excitations that can harvest a wide range of photon
energies, and electronic alloy states that can adjust the open-circuit voltage provide the
underlying basis of ternary blends as a platform for highly efficient next generation
organic solar cells.

201
6.2 Results and Discussion

Ternary blend organic solar cells are bulk heterojunction (BHJ) devices in which
typically either the donor or acceptor component consists of two complementary
materials.
1–17
The blends are of the type D:A1
X
A2
(1-X)
or D1
X
D2
(1-X)
:A where x is varied
over the composition range 0 ≤x≤1 and D and A represent donor and acceptor materials.
BHJ solar cells in which the acceptor is a blend of fullerenes with a poly(3-
hexylthiophene) (P3HT) donor,
1
and in which the donor is a blend of polythiophenes with
a phenyl-C
61
-butyric acid methyl ester (PCBM) acceptor have been reported.
2
An
unprecedented feature in each of these cases is that the open-circuit voltage (V
oc
) is
observed to change continuously with composition x, without detriment to the fill factor
(FF) of the solar cell.
1,2
The inference is that the acceptor LUMO in the case of
complementary acceptors (D:A1
X
A2
(1-X)
)
1
and the donor HOMO in the case of
complementary donors (D1
X
D2
(1-X)
:A),
2
change energy continuously with composition in
the mixture. While this is the normal behavior for an inorganic binary alloy,
18,19
it is less
obviously expected in a molecular mixture where the separate energy levels of the two
species might be preserved in the blend. Indeed, polymer blends generally have an optical
absorption that is the weighted sum of the two materials rather than characteristic of a
uniform material of average composition.
2,3,7,12,13,15

In a BHJ solar cell, V
oc
is given by the energy of the interface band gap minus the
quasi-Fermi energies, where the interface band gap is the energy difference between the
donor HOMO and the acceptor LUMO.
20–22
Hence, while the obvious origin of the
202
continuous change in the V
oc
in ternary blends is the corresponding change in the LUMO
or HOMO level, it is possible that the effect reflects a change in the quasi-Fermi energies.
This paper reports measurements to identify the electronic states in ternary blend systems
and confirms the continuous change in the energy of the HOMO or LUMO of the
complementary two-component material, indicative of alloy formation. However, the
experiments also show that the optically excited exciton states within the two-component
pairs do not reflect the average composition, but instead retain individual molecular
characteristics. An explanation for these observations is proposed, which advances a
working mechanism of operation for ternary blend BHJ organic solar cells. The reported
results support the potential of ternary blends to enable significant enhancements in the
efficiency of BHJ solar cells.
For this study the ternary blend systems previously reported to show a
compositional dependence of the V
oc
were further investigated.
1,2
The system of type
D:A1
X
A2
(1-X)
consists of P3HT as the donor (D), PCBM as A1, and indene-C
60
bisadduct
(ICBA) as A2.
1
The system of type D1
X
D2
(1-X)
:A consists of the low band gap polymer
poly(3-hexylthiophene-thiophene-diketopyrrolopyrrole) (P3HTT-DPP-10%)
23
as D1,
poly(3-hexylthiophene-co-3-(2-ethylhexyl)thiophene) (P3HT
75
-co-EHT
25
),
24
a P3HT
analogue with a deep HOMO as D2, and PCBM as the acceptor (A).
2
Polymers and
fullerenes were prepared as previously described.
1,2
Devices used for measurements were
also prepared without deviation from previous reports in a standard device architecture of
ITO/PEDOT:PSS/Active layer blend/Al. In each case, the composition of the
complementary two-component pair was varied across the full range (0≤x≤1) between
203
the corresponding limiting binary blends in steps of 10% (mass basis). In the D:A1
X
A2
(1-
X)
system the overall ratio of donor to acceptor was constant at an optimal 1:1 for all
compositions, while in the D1
X
D2
(1-X)
:A system the overall donor to acceptor ratio was
individually optimized at each polymer:polymer ratio.

Figure 6.1. Photocurrent spectral response (PSR) data for the P3HT:PCBM:ICBA
(D:A1
X
A2
(1-X)
) ternary blend solar cells plotted as a function of ICBA fraction in
PCBM:ICBA pair. The inset indicates the CT transition or interface band gap that is
being measured and the pair of dashed lines indicates the range over which the interface
band gap energy is extracted.

Measurements of the photocurrent spectral response (PSR) were made to explore
the electronic states of these ternary blend solar cells. The measurements are described in
detail elsewhere, and are made at open-circuit in the solar cells.
25
PSR measures the
optical absorption of those transitions that result in the generation of mobile carriers in
the solar cell and therefore provides information about the electronic states. Previous
studies have shown that the PSR spectrum at energy below the optical band gap of the
separate materials arises from direct excitation at the interface from the donor HOMO to
204
the acceptor LUMO (see inset in Figure 6.1) and measures the optical absorption of the
heterojunction interface.
25,26
The excitation is usually referred to as the charge transfer
(CT) state.
27–29

Figure 6.2. Expanded plot of the peaks near 1.7 eV in the PSR spectra of Figure 6.1, with
the background subtracted. The peak centered above 1.7 eV corresponds to PCBM
absorption and the peak centered below 1.7 eV corresponds to ICBA.

Figure 6.1 shows the PSR spectrum of the mixed acceptor system (D:A1
X
A2
(1-X)
)
and Figure 6.2 shows details of the spectrum at 1.6-1.8 eV with the background
subtracted. The two fullerenes have a similar optical band gap and a reported 0.2 eV
difference in their LUMO levels.
1,27
The spectrum above 1.6 eV reflects bulk absorption
in the donor and acceptor to create excitons that diffuse to the interface and separate.
From the data below 1.6 eV in Figure 6.1, it is evident that the interface CT absorption
shifts in energy continuously with composition. The actual interface band gap is not
clearly delineated in the PSR spectrum, but previous studies have shown that it
corresponds to a photocurrent of about 10
-4
of the peak value, and where the shape of the
PSR spectrum changes from an exponential at lower energy to a broader band at higher
energy.
25
Figure 6.3 compares the interface absorption measured at two different values
205
(0.1 and 0.01) of the photocurrent (dashed lines in Figure 6.1) with the change in the V
oc
.
The compositional dependence is the same within experimental uncertainty. Figure 6.3
shows that the V
oc
is about 0.55 V smaller that the interface band gap which is consistent
with the expected quasi-Fermi energies.
25,29–32

Figure 6.3. Plot of the estimated interface band gap energy defined by PSR at
photocurrent (PC) values of 0.1 and 0.01 from the data of Figure 6.1 (closed symbols)
compared to the values of the V
oc
(open triangles) for P3HT:PCBM:ICBA (D:A1
X
A2
(1-X)
)
ternary blend solar cells. The solid lines are the model of Equation 1 with the same
bowing parameter b.

The data confirm that the V
oc
accurately measures the change in the interface band
gap (E
GI
) and that the band gap changes continuously with the composition of two-
component acceptor. The composition dependence of the band gap for an alloy is often
described by a quadratic dependence on composition, using an extension of Vegard’s
law,
18,33

) 1 ( ) 1 ( ) (
2 1
x bx xE E x x E
G G GI
− − + − = (1)
206
where b is known as the bowing factor and E
G1
and E
G2
are the band gaps of the two
materials. Figure 6.3 shows that this model fits the data well. The same bowing factor
(b=0.18) and value of E
G2
-E
G1
(= 0.23 eV) are used to fit the three sets of data, and the fit
value of E
G2
-E
G1
is close to the measured difference in LUMO level of the two fullerenes.
Figure 6.2 shows the changes in the fullerene absorption peak near 1.7 eV as a
function of composition in the ternary blends. The data is obtained by subtracting a
smooth background from the PSR spectrum between 1.55 and 1.85 eV, to give the
absorption. The data are scaled vertically to give approximately the same integrated
absorption strength for each composition. The absorption peak positions differ by about
0.045 eV with the ICBA system having the lower energy peak. It is evident from the data
and confirmed by modeling that the intermediate compositions have an absorption that is
the weighted sum of the two peaks of the pure materials, rather than a single peak at an
average energy. Hence the alloy model of Equation 1 does not apply to the optical
excitation of the bulk acceptor material.
Figures 6.4, 6.5 and 6.6 show analogous data for the solar cells that have mixed
polymer donors (D1
X
D2
(1-X)
:A). The two polymers have only a slightly different HOMO
level (5.2 eV for P3HTT-DPP-10% and 5.4 eV for P3HT
75
-co-EHT
25
) but widely
differing optical band gaps (1.5 and 1.9 eV, respectively). In this system the V
oc
is also
found to change continuously with donor composition, but the change is smaller than for
the mixed acceptor case. Figure 6.6 shows that the composition dependence of the
interface band gap energy corresponding to two values of the photocurrent in Figure 6.4
is consistent with the changes in the V
oc
. The lines are a linear fit, showing zero bowing
207
factor. The composition end points deviate from the fit, particularly at x=1, and this point
is discussed further below.

Figure 6.4. Photocurrent spectral response (PSR) data for the P3HTT-DPP-10%:P3HT
75
-
co-EHT
25
:PCBM (D1
X
D2
(1-X)
:A) ternary blend solar cells plotted as a function of
P3HT
75
-co-EHT
25
fraction in the polymer donor pair. The pair of dashed lines indicates
the range over which the interface band gap energy is extracted. The inset shows the
photocurrent signal at 1.6 eV as a function of P3HT
75
-co-EHT
25
fraction in the polymer
donor pair.

Figure 6.5 shows the spectral response data plotted on a linear vertical scale in
order to focus on the high energy optical excitations of the polymers. The spectra clearly
display the optical excitations of the two polymers since their onset differs by almost 0.5
eV. The low energy peaks marked E1 and E2 correspond to P3HTT-DPP-10%, while the
higher energy peaks are a combination of P3HTT-DPP-10% and P3HT
75
-co-EHT
25
(as
well as PCBM). Inspection shows that these marked peaks change intensity with
composition but do not change in energy. This observation is confirmed by fitting the
spectra to a sum of Gaussian bands, which is shown by the solid lines in Figure 6.5,
208
where each fit uses the same peak energies. Evidently therefore the composition
dependence of the high-energy absorption spectrum again has a different behavior from
the continuously changing interface band gap. In both the mixed-polymer and mixed-
fullerene systems, the high-energy transitions retain the properties of the individual
materials of the mixture, while the interface band gap exhibits the properties of an
average composition of the two-component donor or acceptor.

Figure 6.5. High energy PSR data from Figure 6.4 plotted on a linear scale to show the
exciton peaks from the donor mixture. Solid lines are fits to the E1, E2 and E3 peaks. The
P3HT
75
-coEHT
25
fraction is indicated for each data set.

209

Figure 6.6. Plot of the estimated interface band gap energy defined by PSR at
photocurrent (PC) values of 0.1 and 0.01 from the data of Figure 6.4 (closed symbols)
compared to the values of the V
oc
(open triangles), as a function of P3HT
75
-co-EHT
25

fraction for the mixed donor system. Solid lines are linear fit to the data with the same
slope (zero bowing factor).

We propose that the explanation for these contrasting properties arises from the
nature of the electronic excitations. The high-energy transitions are excitonic in character
comprising a set of peaks that are narrow enough to include distinguishable phonon
sidebands – for example, E2 is a sideband of E1. The complete overlap of the electron
and hole wavefunctions and the strong Coulomb interaction of the exciton leads to a
highly localized state which is therefore confined within a single molecule (or localized
segment of a polymer chain)
34
and is not subject to a variation based on blending of
components, as in an alloy. Hence the exciton states are characteristic of the different
individual components rather than the average material composition. On the other hand,
the interface band gap is an excitation comprising a hole in the donor material and an
electron in the acceptor. Due to the phase separated bulk heterojunction structure, the
electron and hole are physically apart from each other and therefore have a much smaller
210
Coulomb interaction and consequently are more delocalized.
34,35
The delocalized
wavefunction extends over more than a single molecule and therefore is determined by
the average composition of the donor and acceptor phases at the interface rather than by a
single molecule of either type. Therefore the CT state energy (and thus the V
oc
) is
observed to vary as the composition of either the donor or acceptor is changed. The
variation in CT energy is regular and described well by an alloy model, as developed
here.
As such, the data presented here is consistent with a model in which the two-
component material in a ternary blend forms an alloy with a HOMO and LUMO energy
based on the average composition. Alloy formation implies that the components are
subject to intimate and uniform electronic interaction and that delocalization of states
should be extensive. Consistent with this model, the polymer donors in the D1
X
D2
(1-X)
:A
system are of similar structure and are both semicrystalline as determined by x-ray
analysis
2,23,24
and hence the hole wavefunction is expected to be substantially delocalized.
The crystallinity of the fullerene acceptors in D:A1
X
A2
(1-X)
system is undetermined in
this specific case (although PCBM is well-known to form crystallites in polymer
blends),
36
but delocalizion of the electron wavefunction is evidently sufficient to lead to
an electron energy determined by the average composition. The delocalization and weak
electron-hole interaction of the interface CT excitation is consistent with the absence of
obvious excitonic peaks in the low energy region of the spectral response, and the
measured interface band gap is consistent with the values of the V
oc
, which relates to one-
electron energies.
211
The observations made here supporting the formation of an organic alloy in
ternary blends are unprecedented in BHJ solar cells and differ from previous results with
an analogous system discussed below. Recent studies have measured the PSR spectra of a
BHJ solar cell with a fullerene mixture of PC
84
BM and PC
70
BM, which have LUMO
levels differing by about 0.2 eV.
37
At concentrations of 1-10%, the PC
84
BM acts as a
localized trap having a separate absorption band below the interface transitions associated
with PC
70
BM, rather than as an alloy material with an average composition. A transition
from a localized state characteristic of a dopant to alloy behavior is expected as the
concentration increases, following the inorganic alloy analogy. The electron
wavefunction is determined by the average material only when the concentration of the
dilute phase of the mixture is sufficiently high. It appears that >10% concentration is
sufficient to show alloy properties, since the dopant state behavior is not evident in
Figure 6.1 for the material with 10% PCBM. The large drop in energy between a
P3HT
75
-co-EHT
25
fraction of 1 and 0.9 in Figure 6.6 and the slightly different shape of
the x=0.9 data in Figure 6.1 might be due to the transition from dopant to alloy. The
corresponding transition from pure P3HTT-DPP-10%:PCBM to 10% of P3HT
75
-co-
EHT
25
is not expected to show the same doping effect because P3HT
75
-co-EHT
25
has the
larger band gap and therefore leads to states within the band rather than within the band
gap.
While the alloy nature of ternary blends is clearly manifest in the variation of the
V
oc
with composition, the molecular nature of the excitonic properties is equally apparent.
The inset to Figure 6.4 shows that the photocurrent at 1.6 eV, associated with the low
212
band gap polymer (P3HTT-DPP-10%) is proportional to the polymer concentration even
at low concentrations down to 10%. The implication is that an exciton created in this
dilute component reaches the interface and splits into an electron-hole pair, with the same
efficiency as it does at high concentrations. The P3HTT-DPP-10% exciton energy is
much less than for the P3HT
75
-co-EHT
25
material and so the transport of the exciton
cannot be through the latter material. Either the exciton can diffuse easily through the
dilute component even when its concentration is below the usual percolation threshold, or
the transition to the interface is by a direct energy transfer process that does not involve
diffusion. Further studies are needed to understand this process. However, it is clear that
photon can be harvested effectively by both components in the alloy.
The property of having molecular excitonic optical excitations together with
material-averaged one-electron states in these ternary blends, offers interesting pathways
to improve the efficiency of organic BHJ solar cells. Specifically for multi-donor systems
(D1
X
D2
(1-X)
:A), suitably paired donors can harvest light over a much broader range than
either individual component as a method of increasing the short-circuit current density
(J
sc
) while a V
oc
larger than that achievable for the donor with higher HOMO can be
achieved providing a route to a higher J
sc
×V
oc
product (and efficiency) in ternary blends
than possible in binary blends.
1,2
Further, a substantial contribution to the efficiency loss
of an organic solar cell arises from the band offset needed to separate the electron-hole
pair.
38
It is generally found that if the exciton energy is too close to the interface band
gap, then efficient exciton separation does not occur even though it is energetically
favorable.
31,39,40
The problem is often attributed to a small band offset between the donor
213
and acceptor LUMO.
27
However, if the same material is mixed with a higher energy
donor, two benefits occur. First, the second donor provides high energy optical
transitions to more fully cover the solar energy spectrum, as discussed previously.
2,5,13–15

The second benefit is that the addition of the high energy material pushes the one-
electron LUMO states of the donor mixture higher in energy by the alloying effect. The
band offset at the interface is therefore increased and the exciton is more efficiently split
at the interface, even though the exciton energies are unchanged. A suitably designed
ternary system therefore has the potential for much higher solar cell efficiency than
binary systems.

6.3 Conclusion

The results presented here support the previously observed tunability of the V
oc
in
ternary blend BHJ solar cells and provide a description of how this phenomenon occurs
by invoking an alloy model for the two-part component, either the acceptor in
D:A1
X
A2
(1-X)
or the donor in D1
X
D2
(1-X)
:A. In this model, optical excitation within the
alloy is molecular in nature due to the highly localized nature of the exciton. As such two
donors within distinct band gap can be used in complementary fashion to harvest light
over a broad range and target high J
sc
. Most importantly however, the donor-acceptor
interface and corresponding interface band gap, which determines the V
oc
, displays a
material averaged electronic structure due to the more delocalized nature of the one-
electron states and indeed the CT state energy in general, resulting in a V
oc
that reflects
214
the average composition of the interface. In the case of a D1
X
D2
(1-X)
:A system, this
allows a higher V
oc
to be observed in the ternary blend than could be possible in the
binary blend of the lower band gap (higher HOMO) polymer and the acceptor, as we
have previously reported,
2
leading to higher efficiencies. In a broader sense, an
implication of the alloy model is that the two complementary components should be
mixed for effective electronic interaction. Further studies are needed to develop an
understanding of the role of morphology in these ternary blends.

6.4 Chapter Bibliography

(1) Khlyabich, P. P.; Burkhart, B.; Thompson, B. C. J. Am. Chem. Soc. 2011, 133,
14534.

(2) Khlyabich, P. P.; Burkhart, B.; Thompson, B. C. J. Am. Chem. Soc. 2012, 134,
9074.

(3) Honda, S.; Nogami, T.; Ohkita, H.; Benten, H.; Ito, S. ACS Appl. Mater. Interfaces
2009, 1, 804.

(4) Huang, J.-H.; Velusamy, M.; Ho, K.-C.; Lin, J.-T.; Chu, C.-W. J. Mater. Chem.
2010, 20, 2820.

(5) Chen, C.-H.; Hsieh, C.-H.; Dubosc, M.; Cheng, Y.-J.; Hsu, C.-S. Macromolecules
2010, 43, 697.

(6) Cheng, Y.-J.; Hsieh, C.-H.; Li, P.-J.; Hsu, C.-S. Adv. Funct. Mater. 2011, 21, 1723.

(7) Honda, S.; Ohkita, H.; Benten, H.; Ito, S. Adv. Energy Mater. 2011, 1, 588.

(8) Chen, L.; Yao, K.; Chen, Y. J. Mater. Chem. 2012, 22, 18768.

(9) Li, H.; Zhang, Z.-G.; Li, Y.; Wang, J. Appl. Phys. Lett. 2012, 101, 163302.

215
(10) Liu, H.-W.; Chang, D.-Y.; Chiu, W.-Y.; Rwei, S.-P.; Wang, L. J. Mater. Chem.
2012, 22, 15586.

(11) Jeltsch, K. F.; Schädel, M.; Bonekamp, J.-B.; Niyamakom, P.; Rauscher, F.;
Lademann, H. W. A.; Dumsch, I.; Allard, S.; Scherf, U.; Meerholz, K. Adv. Funct.
Mater. 2012, 22, 397.

(12) Lee, J.; Yun, M. H.; Kim, J.; Kim, J. Y.; Yang, C. Macromol. Rapid Commun.
2012, 33, 140.

(13) Yang, L.; Zhou, H.; Price, S. C.; You, W. J. Am. Chem. Soc. 2012, 134, 5432.

(14) Sharma, G. D.; Singh, S. P.; Roy, M. S.; Mikroyannidis, J. A. Org. Electron. 2012,
13, 1756.

(15) Ameri, T.; Min, J.; Li, N.; Machui, F.; Baran, D.; Forster, M.; Schottler, K. J.;
Dolfen, D.; Scherf, U.; Brabec, C. J. Adv. Energy Mater. 2012, 2, 1198.

(16) Cha, H.; Chung, D. S.; Bae, S. Y.; Lee, M.-J.; An, T. K.; Hwang, J.; Kim, K. H.;
Kim, Y.-H.; Choi, D. H.; Park, C. E. Adv. Funct. Mater. 2013, DOI:
10.1002/adfm.201201913.

(17) Andersson, L. M.; Hsu, Y.-T.; Vandewal, K.; Sieval, A. B.; Andersson, M. R.;
Inganäs, O. Org. Electron. 2012, 13, 2856.

(18) Rockett, A. Materials science of semiconductors; Springer: New York, 2008.

(19) Moses, P. G.; Van de Walle, C. G. Appl. Phys. Lett. 2010, 96, 021908.

(20) Garcia-Belmonte, G.; Bisquert, J. Appl. Phys. Lett. 2010, 96, 113301.

(21) Schlenker, C. W.; Thompson, M. E. Chem. Commun. 2011, 47, 3702.

(22) Qi, B.; Wang, J. J. Mater. Chem. 2012, 22, 24315.

(23) Khlyabich, P. P.; Burkhart, B.; Ng, C. F.; Thompson, B. C. Macromolecules 2011,
44, 5079.

(24) Burkhart, B.; Khlyabich, P. P.; Thompson, B. C. Macromolecules 2012, 45, 3740.

(25) Street, R.; Song, K.; Northrup, J.; Cowan, S. Phys. Rev. B 2011, 83, 165207.

216
(26) Goris, L.; Poruba, A.; Hod’ákova, L.; Vane ̌c ̌ek, M.; Haenen, K.; Nesládek, M.;
Wagner, P.; Vanderzande, D.; De Schepper, L.; Manca, J. V. Appl. Phys. Lett.
2006, 88, 052113.

(27) Thompson, B. C.; Khlyabich, P. P.; Burkhart, B.; Aviles, A. E.; Rudenko, A.;
Shultz, G. V.; Ng, C. F.; Mangubat, L. B. Green 2011, 1, 29.

(28) Clarke, T. M.; Durrant, J. R. Chem. Rev. 2010, 110, 6736.

(29) Deibel, C.; Strobel, T.; Dyakonov, V. Adv. Mater. 2010, 22, 4097.

(30) Vandewal, K.; Tvingstedt, K.; Gadisa, A.; Inganäs, O.; Manca, J. V. Nat. Mater.
2009, 8, 904.

(31) Veldman, D.; Meskers, S. C. J.; Janssen, R. A. J. Adv. Funct. Mater. 2009, 19,
1939.

(32) Street, R. A.; Schoendorf, M.; Roy, A.; Lee, J. H. Phys. Rev. B 2010, 81, 205307.

(33) Wu, J.-C.; Zheng, J.; Zacherl, C. L.; Wu, P.; Liu, Z.-K.; Xu, R. J. Phys. Chem. C
2011, 115, 19741.

(34) Kaake, L. G.; Jasieniak, J. J.; Bakus, R. C.; Welch, G. C.; Moses, D.; Bazan, G. C.;
Heeger, A. J. J. Am. Chem. Soc. 2012, 134, 19828.

(35) Bakulin, A. A.; Rao, A.; Pavelyev, V. G.; Van Loosdrecht, P. H. M.;
Pshenichnikov, M. S.; Niedzialek, D.; Cornil, J.; Beljonne, D.; Friend, R. H.
Science 2012, 335, 1340.

(36) Kim, Y.; Nelson, J.; Zhang, T.; Cook, S.; Durrant, J. R.; Kim, H.; Park, J.; Shin,
M.; Nam, S.; Heeney, M.; McCulloch, I.; Ha, C.-S.; Bradley, D. D. C. ACS Nano
2009, 3, 2557.

(37) Street, R. A.; Krakaris, A.; Cowan, S. R. Adv. Funct. Mater. 2012, 22, 4608.

(38) Kirchartz, T.; Taretto, K.; Rau, U. J. Phys. Chem. C 2009, 113, 17958.

(39) Di Nuzzo, D.; Wetzelaer, G.-J. A. H.; Bouwer, R. K. M.; Gevaerts, V. S.; Meskers,
S. C. J.; Hummelen, J. C.; Blom, P. W. M.; Janssen, R. A. J. Adv. Energy Mater.
2013, 3, 85.

(40) Faist, M. A.; Kirchartz, T.; Gong, W.; Ashraf, R. S.; McCulloch, I.; De Mello, J.
C.; Ekins-Daukes, N. J.; Bradley, D. D. C.; Nelson, J. J. Am. Chem. Soc. 2012,
134, 685.
217
Perspectives

The field of organic electronics attracts significant attention and has expanded
with an accelerating pace over the last 15 years. Organic light-emitting diodes, organic
field effect transistors, organic sensors and organic solar cells become a regular part of
our everyday lives. However, despite all the success in the field, organic electronics is far
behind the inorganic “big” brother. Of course this can be explained due to a short period
of time for organic electronics being on the market relative to inorganic electronics, as
well as better understanding of the working mechanisms and industrial manufacturing of
the latter, however the potential of organic electronics is tremendous. Organic materials
offer a set of the properties such as flexibility, lightweight, semi transparency and, of
critical importance, inexpensive and fast manufacturing of the final product, which is
unachievable with the existing traditional inorganic materials. Furthermore, organic
materials allow fine tuning of the chemical composition in order to target a desired set of
properties. As a result, the projected and existing cost of many organic-based electronic
devices should be significantly lower than that of the inorganic analogues.
In terms of organic photovoltaics, the year 2011 brought the first organic solar
cell with the efficiency exceeding 10%. Since then, power conversion efficiencies
reached 12% for vapor-deposited multi-junction organic solar cells. At the same time,
solution-processed OPVs were lagging behind the vapor-deposited devices in terms of
efficiency for abrief time, but recently single junction solution-processed devices reached
efficiencies above 9%, while 10.6% was recorded for the solution-processed tandem solar
218
cells in 2012. In the case of all-polymer solar cells, Polyera recently reached 6.4% in a
binary blend solar cell configuration, thus proving the potential of substituting the
fullerene acceptor with a polymer acceptor, but further research is needed to achieve even
higher efficiencies. However, still all the efficiencies reported to date are obtained using
small area often less than 1 cm
2
, multi-step synthesized air sensitive polymers and
complicated device architectures (tandem solar cells, and use of hole and/or electron
blocking layers in case of binary blend solar cells), which contradicts the idea of organic
electronics as a cheap, fast and simple means for device manufacturing. Thus, the next
important goal for the entire field of organic solar cells is manufacturing of large area
photovoltaic devices with the efficiencies at least 10% and long operational stability, in
order to be considered a potential competitor to the inorganic solar cells. Furthermore,
organic solar cells should utilize their unique set of properties and be used in the areas
where traditional inorganic materials cannot be applied, such as UV transparent solar
windows, solar clothes and bags, etc. Moreover, the published efficiencies approach the
practical efficiency limits for binary and tandem organic solar cells (12% and 15%,
respectively) and novel concepts are needed to push the efficiencies forward.
Ternary blend bulk heterojunction solar cells have a potential to become the
leading concept to increase the efficiency of the solar cells beyond the practical limits of
binary blend solar cells while keeping the simplicity of the photovoltaic device
fabrication, as discussed in Chapter 2. The increase is possible due to the simultaneous
increase of the short-circuit current density and, what is the most important, open-circuit
voltage as presented in Chapter 3 and 5. Furthermore, the individual optical properties of
219
the components are retained in the blend, thus allowing broad coverage of the solar
spectrum, while the energy of the charge-transfer state and hence the open-circuit voltage
is found to be composition dependent are vary between the open-circuit voltage of the
individual corresponding binary blend solar cells, as shown in Chapter 6. Though many
questions have to be answered about the working mechanisms, optimal donor and
acceptor materials and morphology, the ternary blend approach has all the necessary
features to be the key player in the OPV field in years to come.

220
Bibliography

Annual Energy Outlook 2013 - Energy Information Administration
http://www.eia.gov/forecasts/aeo/ (accessed May 28, 2013).

Annual Energy Review - Energy Information Administration
http://www.eia.gov/totalenergy/data/annual/ (accessed May 28, 2013).

Solarmer Energy, Inc., http://www.solarmer.com/rd.php (Accessed October 12, 2010)
(accessed Oct 12, 2010).

Aberle, A. G. Thin Solid Films 2009, 517, 4706.

Adam, G.; Pivrikas, A.; Ramil, A. M.; Tadesse, S.; Yohannes, T.; Sariciftci, N. S.; Egbe,
D. A. M. J. Mater. Chem. 2011, 21, 2594.

Agostinelli, T.; Lilliu, S.; Labram, J. G.; Campoy-Quiles, M.; Hampton, M.; Pires, E.;
Rawle, J.; Bikondoa, O.; Bradley, D. D. C.; Anthopoulos, T. D.; Nelson, J.;
Macdonald, J. E. Adv. Funct. Mater. 2011, 21, 1701.

Al-Ibrahim, M.; Roth, H.-K.; Schroedner, M.; Konkin, A.; Zhokhavets, U.; Gobsch, G.;
Scharff, P.; Sensfuss, S. Org. Electron. 2005, 6, 65.

Allard, N.; Aich, R. B.; Gendron, D.; Boudreault, P.-L. T.; Tessier, C.; Alem, S.; Tse, S.-
C.; Tao, Y.; Leclerc, M. Macromolecules 2010, 43, 2328.

Alstrup, J.; Jørgensen, M.; Medford, A. J.; Krebs, F. C. ACS Appl. Mater. Interfaces
2010, 2, 2819.

Amb, C. M.; Chen, S.; Graham, K. R.; Subbiah, J.; Small, C. E.; So, F.; Reynolds, J. R. J.
Am. Chem. Soc. 2011, 133, 10062.

Ameri, T.; Dennler, G.; Lungenschmied, C.; Brabec, C. J. Energy Environ. Sci. 2009, 2,
347.

Ameri, T.; Khoram, P.; Min, J.; Brabec, C. J. Adv. Mater. 2013, DOI:
10.1002/adma.201300623.

Ameri, T.; Li, N.; Brabec, C. J. Energy Environ. Sci. 2013, DOI: 10.1039/c3ee40388b.

Ameri, T.; Min, J.; Li, N.; Machui, F.; Baran, D.; Forster, M.; Schottler, K. J.; Dolfen,
D.; Scherf, U.; Brabec, C. J. Adv. Energy Mater. 2012, 2, 1198.
221
Andersson, L. M.; Hsu, Y.-T.; Vandewal, K.; Sieval, A. B.; Andersson, M. R.; Inganäs,
O. Org. Electron. 2012, 13, 2856.

Andrew, T. L.; Bulović, V. ACS Nano 2012, 6, 4671.

Armaroli, N.; Balzani, V. Energy Environ. Sci. 2011, 4, 3193.

Athanasopoulos, S.; Emelianova, E. V.; Walker, A. B.; Beljonne, D. Phys. Rev. B 2009,
80, 195209.

Avrutin, V.; Izyumskaya, N.; Morkoç, H. Superlattices Microstruct. 2011, 49, 337.

Badrou Aïch, R.; Zou, Y.; Leclerc, M.; Tao, Y. Org. Electron. 2010, 11, 1053.

Bakulin, A. A.; Rao, A.; Pavelyev, V. G.; van Loosdrecht, P. H. M.; Pshenichnikov, M.
S.; Niedzialek, D.; Cornil, J.; Beljonne, D.; Friend, R. H. Science 2012, 335, 1340.

Bakulin, A. A.; Hummelen, J. C.; Pshenichnikov, M. S.; van Loosdrecht, P. H. M. Adv.
Funct. Mater. 2010, 20, 1653.

Bakulin, A. A.; Martyanov, D.; Paraschuk, D. Y.; Loosdrecht, P. H. M. van;
Pshenichnikov, M. S. Chem. Phys. Lett. 2009, 482, 99.

Bakulin, A. A.; Zapunidy, S. A.; Pshenichnikov, M. S.; van Loosdrecht, P. H. M.;
Paraschuk, D. Y. Phys. Chem. Chem. Phys. 2009, 11, 7324.

Ball, Z. T.; Sivula, K.; Fréchet, J. M. J. Macromolecules 2006, 39, 70.

Ballantyne, A. M.; Ferenczi, T. A. M.; Campoy-Quiles, M.; Clarke, T. M.; Maurano, A.;
Wong, K. H.; Zhang, W.; Stingelin-Stutzmann, N.; Kim, J.-S.; Bradley, D. D. C.;
Durrant, J. R.; McCulloch, I.; Heeney, M.; Nelson, J.; Tierney, S.; Duffy, W.;
Mueller, C.; Smith, P. Macromolecules 2010, 43, 1169.

Beaujuge, P. M.; Amb, C. M.; Reynolds, J. R. Acc. Chem. Res. 2010, 43, 1396.

Beaujuge, P. M.; Fréchet, J. M. J. J. Am. Chem. Soc. 2011, 133, 20009.

Benson-Smith, J. J.; Ohkita, H.; Cook, S.; Durrant, J. R.; Bradley, D. D. C.; Nelson, J.
Dalton Trans. 2009, 10000.

Benson-Smith, J. J.; Wilson, J.; Dyer-Smith, C.; Mouri, K.; Yamaguchi, S.; Murata, H.;
Nelson, J. J. Phys. Chem. B 2009, 113, 7794.

Bian, L.; Zhu, E.; Tang, J.; Tang, W.; Zhang, F. Prog. Polym. Sci. 2012, 37, 1292.
222
Bijleveld, J. C.; Gevaerts, V. S.; Di Nuzzo, D.; Turbiez, M.; Mathijssen, S. G. J.; de
Leeuw, D. M.; Wienk, M. M.; Janssen, R. A. J. Adv. Mater. 2010, 22, E242.

Bijleveld, J. C.; Karsten, B. P.; Mathijssen, S. G. J.; Wienk, M. M.; de Leeuw, D. M.;
Janssen, R. A. J. J. Mater. Chem. 2011, 21, 1600.

Bijleveld, J. C.; Zoombelt, A. P.; Mathijssen, S. G. J.; Wienk, M. M.; Turbiez, M.; de
Leeuw, D. M.; Janssen, R. A. J. J. Am. Chem. Soc. 2009, 131, 16616.

Biniek, L.; Fall, S.; Chochos, C. L.; Anokhin, D. V.; Ivanov, D. A.; Leclerc, N.; Lévêque,
P.; Heiser, T. Macromolecules 2010, 43, 9779.

Blom, P. W. M.; Mihailetchi, V. D.; Koster, L. J. A.; Markov, D. E. Adv. Mater. 2007,
19, 1551.

Boudouris, B. W.; Ho, V.; Jimison, L. H.; Toney, M. F.; Salleo, A.; Segalman, R. A.
Macromolecules 2011, 44, 6653.

Boudreault, P.-L. T.; Najari, A.; Leclerc, M. Chem. Mater. 2011, 23, 456.

Brabec, C. J. Organic photovoltaics: concepts and realization; Springer series in
materials science; Springer: New York, 2003.

Brabec, C. J.; Gowrisanker, S.; Halls, J. J. M.; Laird, D.; Jia, S.; Williams, S. P. Adv.
Mater. 2010, 22, 3839.

Brabec, C. J.; Heeney, M.; McCulloch, I.; Nelson, J. Chem. Soc. Rev. 2011, 40, 1185.

Brady, M. A.; Su, G. M.; Chabinyc, M. L. Soft Matter 2011, 7, 11065.

Brazovskii, S.; Kirova, N. Chem. Soc. Rev. 2010, 39, 2453.

Brédas, J.-L.; Norton, J. E.; Cornil, J.; Coropceanu, V. Acc. Chem. Res. 2009, 42, 1691.

Bronstein, H.; Chen, Z.; Ashraf, R. S.; Zhang, W.; Du, J.; Durrant, J. R.; Shakya
Tuladhar, P.; Song, K.; Watkins, S. E.; Geerts, Y.; Wienk, M. M.; Janssen, R. A.
J.; Anthopoulos, T.; Sirringhaus, H.; Heeney, M.; McCulloch, I. J. Am. Chem. Soc.
2011, 133, 3272.

Bundgaard, E.; Krebs, F. Sol. Energy Mater. Sol. Cells 2007, 91, 954.

Burkhart, B.; Khlyabich, P. P.; Cakir Canak, T.; LaJoie, T. W.; Thompson, B. C.
Macromolecules 2011, 44, 1242.

223
Burkhart, B.; Khlyabich, P. P.; Thompson, B. C. Macromolecules 2012, 45, 3740.

Burschka, J.; Dualeh, A.; Kessler, F.; Baranoff, E.; Cevey-Ha, N.-L.; Yi, C.;
Nazeeruddin, M. K.; Grätzel, M. J. Am. Chem. Soc. 2011, 133, 18042.

Camaioni, N.; Po, R. J. Phys. Chem. Lett. 2013, 4, 1821.

Campoy-Quiles, M.; Kanai, Y.; El-Basaty, A.; Sakai, H.; Murata, H. Org. Electron.
2009, 10, 1120.

Cardona, C. M.; Li, W.; Kaifer, A. E.; Stockdale, D.; Bazan, G. C. Adv. Mater. 2011, 23,
2367.

Cha, H.; Chung, D. S.; Bae, S. Y.; Lee, M.-J.; An, T. K.; Hwang, J.; Kim, K. H.; Kim,
Y.-H.; Choi, D. H.; Park, C. E. Adv. Funct. Mater. 2013, DOI:
10.1002/adfm.201201913.

Chang, M.-Y.; Chen, Y.-F.; Tsai, Y.-S.; Chi, K.-M. J. Electrochem. Soc. 2009, 156,
B234.

Chang, S.-Y.; Liao, H.-C.; Shao, Y.-T.; Sung, Y.-M.; Hsu, S.-H.; Ho, C.-C.; Su, W.-F.;
Chen, Y.-F. J. Mater. Chem. A 2013, 1, 2447.

Chang, S.-W.; Waters, H.; Kettle, J.; Horie, M. Org. Electron. 2012, 13, 2967.

Chen, C.-Y.; Wang, M.; Li, J.-Y.; Pootrakulchote, N.; Alibabaei, L.; Ngoc-le, C.;
Decoppet, J.-D.; Tsai, J.-H.; Grätzel, C.; Wu, C.-G.; Zakeeruddin, S. M.; Grätzel,
M. ACS Nano 2009, 3, 3103.

Chen, C.-H.; Cheng, Y.-J.; Dubosc, M.; Hsieh, C.-H.; Chu, C.-C.; Hsu, C.-S. Chem.
Asian J. 2010, 5, 2483.

Chen, C.-H.; Hsieh, C.-H.; Dubosc, M.; Cheng, Y.-J.; Hsu, C.-S. Macromolecules 2010,
43, 697.

Chen, H.-Y.; Hou, J.; Zhang, S.; Liang, Y.; Yang, G.; Yang, Y.; Yu, L.; Wu, Y.; Li, G.
Nat. Photonics 2009, 3, 649.

Chen, J.; Yu, X.; Hong, K.; Messman, J. M.; Pickel, D. L.; Xiao, K.; Dadmun, M. D.;
Mays, J. W.; Rondinone, A. J.; Sumpter, B. G.; Kilbey II, S. M. J. Mater. Chem.
2012, 22, 13013.

Chen, L.-M.; Hong, Z.; Li, G.; Yang, Y. Adv. Mater. 2009, 21, 1434.

224
Chen, M. C.; Liaw, D. J.; Huang, Y. C.; Wu, H. Y.; Tai, Y. Sol. Energy Mater. Sol. Cells
2011, 95, 2621.

Chen, M.-C.; Kar, S.; Liaw, D.-J.; Chen, W.-H.; Huang, Y.-C.; Tai, Y. Org. Electron.
2012, 13, 2702.

Chen, M.-C.; Liaw, D.-J.; Chen, W.-H.; Huang, Y.-C.; Sharma, J.; Tai, Y. Appl. Phys.
Lett. 2011, 99, 223305.

Chen, Y.-C.; Hsu, C.-Y.; Lin, R. Y.-Y.; Ho, K.-C.; Lin, J. T. ChemSusChem 2013, 6, 20.

Cheng, Y.-J.; Hsieh, C.-H.; Li, P.-J.; Hsu, C.-S. Adv. Funct. Mater. 2011, 21, 1723.

Cho, Y. J.; Lee, J. Y.; Chin, B. D.; Forrest, S. R. Org. Electron. 2013, 14, 1081.

Chochos, C. L.; Choulis, S. A. Prog. Polym. Sci. 2011, 36, 1326.

Choi, J. W.; Kulshreshtha, C.; Kennedy, G. P.; Kwon, J. H.; Jung, S.-H.; Chae, M. Sol.
Energy Mater. Sol. Cells 2011, 95, 2069.

Chu, C.-W.; Yang, H.; Hou, W.-J.; Huang, J.; Li, G.; Yang, Y. Appl. Phys. Lett. 2008,
92, 103306.

Chu, T.-Y.; Lu, J.; Beaupré, S.; Zhang, Y.; Pouliot, J.-R.; Wakim, S.; Zhou, J.; Leclerc,
M.; Li, Z.; Ding, J.; Tao, Y. J. Am. Chem. Soc. 2011, 133, 4250.

Chung, I.; Lee, B.; He, J.; Chang, R. P. H.; Kanatzidis, M. G. Nature 2012, 485, 486.

Clarke, T. M.; Ballantyne, A. M.; Nelson, J.; Bradley, D. D. C.; Durrant, J. R. Adv.
Funct. Mater. 2008, 18, 4029.

Clarke, T. M.; Ballantyne, A. M.; Tierney, S.; Heeney, M.; Duffy, W.; McCulloch, I.;
Nelson, J.; Durrant, J. R. J. Phys. Chem. C 2010, 114, 8068.

Clarke, T. M.; Durrant, J. R. Chem. Rev. 2010, 110, 6736.

Clarke, T. M.; Jamieson, F. C.; Durrant, J. R. J. Phys. Chem. C 2009, 113, 20934.

Coakley, K. M.; McGehee, M. D. Chem. Mater. 2004, 16, 4533.

Cook, S.; Furube, A.; Katoh, R.; Han, L. Chem. Phys. Lett. 2009, 478, 33.

Cook, S.; Katoh, R.; Furube, A. J. Phys. Chem. C 2009, 113, 2547.

225
Cook, S.; Ohkita, H.; Kim, Y.; Benson-Smith, J. J.; Bradley, D. D. C.; Durrant, J. R.
Chem. Phys. Lett. 2007, 445, 276.

Cooling, N.; Burke, K. B.; Zhou, X.; Lind, S. J.; Gordon, K. C.; Jones, T. W.; Dastoor, P.
C.; Belcher, W. J. Sol. Energy Mater. Sol. Cells 2011, 95, 1767.

Cowan, S. R.; Leong, W. L.; Banerji, N.; Dennler, G.; Heeger, A. J. Adv. Funct. Mater.
2011, 21, 3083.

Cravino, A. Appl. Phys. Lett. 2007, 91, 243502.

Dang, M. T.; Hirsch, L.; Wantz, G. Adv. Mater. 2011, 23, 3597.

Dang, M. T.; Hirsch, L.; Wantz, G.; Wuest, J. D. Chem. Rev. 2013, 113, 3734.

Dayal, S.; Kopidakis, N.; Olson, D. C.; Ginley, D. S.; Rumbles, G. Nano Lett. 2010, 10,
239.

De Freitas, J. N.; Grova, I. R.; Akcelrud, L. C.; Arici, E.; Sariciftci, N. S.; Nogueira, A.
F. J. Mater. Chem. 2010, 20, 4845.

Deibel, C.; Dyakonov, V. Rep. Prog. Phys. 2010, 73, 096401.

Deibel, C.; Dyakonov, V.; Brabec, C. J. IEEE J. Sel. Top. Quantum Electron. 2010, 16,
1517.

Deibel, C.; Strobel, T.; Dyakonov, V. Adv. Mater. 2010, 22, 4097.

Delucchi, M. A.; Jacobson, M. Z. Energy Policy 2011, 39, 1170.

Dennler, G.; Scharber, M. C.; Ameri, T.; Denk, P.; Forberich, K.; Waldauf, C.; Brabec,
C. J. Adv. Mater. 2008, 20, 579.

Dennler, G.; Scharber, M. C.; Brabec, C. J. Adv. Mater. 2009, 21, 1323.

Devabhaktuni, V.; Alam, M.; Shekara Sreenadh Reddy Depuru, S.; Green, R. C.; Nims,
D.; Near, C. Renew. Sust. Energ. Rev. 2013, 19, 555.

Di Nuzzo, D.; Wetzelaer, G.-J. A. H.; Bouwer, R. K. M.; Gevaerts, V. S.; Meskers, S. C.
J.; Hummelen, J. C.; Blom, P. W. M.; Janssen, R. A. J. Adv. Energy Mater. 2013,
3, 85.

Dou, L.; Chang, W.-H.; Gao, J.; Chen, C.-C.; You, J.; Yang, Y. Adv. Mater. 2013, 25,
825.
226
Dou, L.; You, J.; Yang, J.; Chen, C.-C.; He, Y.; Murase, S.; Moriarty, T.; Emery, K.; Li,
G.; Yang, Y. Nat. Photonics 2012, 6, 180.

Drees, M.; Hoppe, H.; Winder, C.; Neugebauer, H.; Sariciftci, N. S.; Schwinger, W.;
Schäffler, F.; Topf, C.; Scharber, M. C.; Zhu, Z.; Gaudiana, R. J. Mater. Chem.
2005, 15, 5158.

Duck, B. C.; Vaughan, B.; Cooling, N.; Zhou, X.; Holdsworth, J. L.; Wen, L. L.;
Rasmussen, S. C.; Dastoor, P. C.; Belcher, W. J. Sol. Energy Mater. Sol. Cells
2013, 114, 65.

Dyer-Smith, C.; Benson-Smith, J. J.; Bradley, D. D. C.; Murata, H.; Mitchell, W. J.;
Shaheen, S. E.; Haque, S. A.; Nelson, J. J. Phys. Chem. C 2009, 113, 14533.

Emelianova, E. V.; van der Auweraer, M.; Bässler, H. J. Chem. Phys. 2008, 128, 224709.

Espinosa, N.; Hösel, M.; Angmo, D.; Krebs, F. C. Energy Environ. Sci. 2012, 5, 5117.

Faist, M. A.; Kirchartz, T.; Gong, W.; Ashraf, R. S.; McCulloch, I.; de Mello, J. C.;
Ekins-Daukes, N. J.; Bradley, D. D. C.; Nelson, J. J. Am. Chem. Soc. 2012, 134,
685.

Feron, K.; Belcher, W.; Fell, C.; Dastoor, P. Int. J. Mol. Sci. 2012, 13, 17019.

Ford, T. A.; Ohkita, H.; Cook, S.; Durrant, J. R.; Greenham, N. C. Chem. Phys. Lett.
2008, 454, 237.

Friedel, B.; McNeill, C. R.; Greenham, N. C. Chem. Mater. 2010, 22, 3389.

Gadisa, A.; Oosterbaan, W. D.; Vandewal, K.; Bolsée, J.-C.; Bertho, S.; D’Haen, J.;
Lutsen, L.; Vanderzande, D.; Manca, J. V. Adv. Funct. Mater. 2009, 19, 3300.

Gadisa, A.; Svensson, M.; Andersson, M. R.; Inganäs, O. Appl. Phys. Lett. 2004, 84,
1609.

Gao, D.; Hollinger, J.; Seferos, D. S. ACS Nano 2012, 6, 7114.

Garcia-Belmonte, G.; Bisquert, J. Appl. Phys. Lett. 2010, 96, 113301.

Gendron, D.; Leclerc, M. Energy Environ. Sci. 2011, 4, 1225.

Gevorgyan, S. A.; Jørgensen, M.; Krebs, F. C. Sol. Energy Mater. Sol. Cells 2008, 92,
736.
227
Giebink, N. C.; Wiederrecht, G. P.; Wasielewski, M. R.; Forrest, S. R. Phys. Rev. B 2011,
83, 195326.

Ginley, D.; Green, M. A.; Collins, R. MRS Bull. 2008, 33, 355.

Girtan, M.; Rusu, M. Sol. Energy Mater. Sol. Cells 2010, 94, 446.

Gommans, H.; Schols, S.; Kadashchuk, A.; Heremans, P.; Meskers, S. C. J. J. Phys.
Chem. C 2009, 113, 2974.

Goris, L.; Poruba, A.; Hod’ákova, L.; Vaněček, M.; Haenen, K.; Nesládek, M.; Wagner,
P.; Vanderzande, D.; De Schepper, L.; Manca, J. V. Appl. Phys. Lett. 2006, 88,
052113.

Graham, K. R.; Stalder, R.; Wieruszewski, P. M.; Patel, D. G.; Salazar, D. H.; Reynolds,
J. R. ACS Appl. Mater. Interfaces 2013, 5, 63.

Grancini, G.; Maiuri, M.; Fazzi, D.; Petrozza, A.; Egelhaaf, H.-J.; Brida, D.; Cerullo, G.;
Lanzani, G. Nat. Mater. 2012, 12, 29.

Grätzel, M. Nature 2001, 414, 338.

Green, M. A. J. Mater. Sci.: Mater. Electron. 2007, 18, 15.

Green, M. A.; Emery, K.; Hishikawa, Y.; Warta, W. Prog. Photovolt: Res. Appl. 2011,
19, 84.

Green, M. A.; Emery, K.; Hishikawa, Y.; Warta, W.; Dunlop, E. D. Prog. Photovolt: Res.
Appl. 2013, 21, 1.

Grossiord, N.; Kroon, J. M.; Andriessen, R.; Blom, P. W. M. Org. Electron. 2012, 13,
432.

Gruber, M.; Wagner, J.; Klein, K.; Hörmann, U.; Opitz, A.; Stutzmann, M.; Brütting, W.
Adv. Energy Mater. 2012, 2, 1100.

Guerrero, A.; Ripolles-Sanchis, T.; Boix, P. P.; Garcia-Belmonte, G. Org. Electron.
2012, 13, 2326.

Günes, S.; Neugebauer, H.; Sariciftci, N. S. Chem. Rev. 2007, 107, 1324.

Guo, J.; Ohkita, H.; Benten, H.; Ito, S. J. Am. Chem. Soc. 2009, 131, 16869.

Guo, J.; Ohkita, H.; Benten, H.; Ito, S. J. Am. Chem. Soc. 2010, 132, 6154.
228
Gurau, M. C.; Delongchamp, D. M.; Vogel, B. M.; Lin, E. K.; Fischer, D. A.;
Sambasivan, S.; Richter, L. J. Langmuir 2007, 23, 834.

Hadipour, A.; de Boer, B.; Blom, P. W. M. Org. Electron. 2008, 9, 617.

Hadipour, A.; de Boer, B.; Blom, P. W. M. Adv. Funct. Mater. 2008, 18, 169.

Hadipour, A.; de Boer, B.; Wildeman, J.; Kooistra, F. B.; Hummelen, J. C.; Turbiez, M.
G. R.; Wienk, M. M.; Janssen, R. A. J.; Blom, P. W. M. Adv. Funct. Mater. 2006,
16, 1897.

Hagfeldt, A.; Boschloo, G.; Sun, L.; Kloo, L.; Pettersson, H. Chem. Rev. 2010, 110,
6595.

Hallermann, M.; Da Como, E.; Feldmann, J.; Izquierdo, M.; Filippone, S.; Martín, N.;
Jüchter, S.; von Hauff, E. Appl. Phys. Lett. 2010, 97, 023301.

Hallermann, M.; Kriegel, I.; Da Como, E.; Berger, J. M.; von Hauff, E.; Feldmann, J.
Adv. Funct. Mater. 2009, 19, 3662.

Halls, J. J. M.; Walsh, C. A.; Greenham, N. C.; Marseglia, E. A.; Friend, R. H.; Moratti,
S. C.; Holmes, A. B. Nature 1995, 376, 498.

Hamilton, R.; Shuttle, C. G.; O’Regan, B.; Hammant, T. C.; Nelson, J.; Durrant, J. R. J.
Phys. Chem. Lett. 2010, 1, 1432.

Hanna, M. C.; Nozik, A. J. J. Appl. Phys. 2006, 100, 074510.

Hardin, B. E.; Snaith, H. J.; McGehee, M. D. Nat. Photonics 2012, 6, 162.

Hau, S. K.; Cheng, Y.-J.; Yip, H.-L.; Zhang, Y.; Ma, H.; Jen, A. K.-Y. ACS Appl. Mater.
Interfaces 2010, 2, 1892.

Hauch, J. A.; Schilinsky, P.; Choulis, S. A.; Childers, R.; Biele, M.; Brabec, C. J. Sol.
Energy Mater. Sol. Cells 2008, 92, 727.

He, F.; Yu, L. J. Phys. Chem. Lett. 2011, 2, 3102.

He, M.; Qiu, F.; Lin, Z. J. Phys. Chem. Lett. 2013, 4, 1788.

He, Y.; Chen, H.-Y.; Hou, J.; Li, Y. J. Am. Chem. Soc. 2010, 132, 1377.

He, Z.; Zhong, C.; Huang, X.; Wong, W.-Y.; Wu, H.; Chen, L.; Su, S.; Cao, Y. Adv.
Mater. 2011, 23, 4636.
229
He, Z.; Zhong, C.; Su, S.; Xu, M.; Wu, H.; Cao, Y. Nat. Photonics 2012, 6, 593.

Hegedus, S. WIREs Energy Environ. 2013, 2, 218.

Helgesen, M.; Gevorgyan, S. A.; Krebs, F. C.; Janssen, R. A. J. Chem. Mater. 2009, 21,
4669.

Helgesen, M.; Krebs, F. C. Macromolecules 2010, 43, 1253.

Helgesen, M.; Søndergaard, R.; Krebs, F. C. J. Mater. Chem. 2010, 20, 36.

Heremans, P.; Cheyns, D.; Rand, B. P. Acc. Chem. Res. 2009, 42, 1740.

Hesse, H. C.; Weickert, J.; Hundschell, C.; Feng, X.; Müllen, K.; Nickel, B.; Mozer, A.
J.; Schmidt-Mende, L. Adv. Energy Mater. 2011, 1, 861.

Ho, V.; Boudouris, B. W.; Segalman, R. A. Macromolecules 2010, 43, 7895.

Holcombe, T. W.; Woo, C. H.; Kavulak, D. F. J.; Thompson, B. C.; Fréchet, J. M. J. J.
Am. Chem. Soc. 2009, 131, 14160.

Honda, S.; Nogami, T.; Ohkita, H.; Benten, H.; Ito, S. ACS Appl. Mater. Interfaces 2009,
1, 804.

Honda, S.; Ohkita, H.; Benten, H.; Ito, S. Chem. Commun. 2010, 46, 6596.

Honda, S.; Ohkita, H.; Benten, H.; Ito, S. Adv. Energy Mater. 2011, 1, 588.

Honda, S.; Yokoya, S.; Ohkita, H.; Benten, H.; Ito, S. J. Phys. Chem. C 2011, 115,
11306.

Hori, T.; Masuda, T.; Fukuoka, N.; Hayashi, T.; Miyake, Y.; Kamikado, T.; Yoshida, H.;
Fujii, A.; Shimizu, Y.; Ozaki, M. Org. Electron. 2012, 13, 335.

Hoth, C. N.; Schilinsky, P.; Choulis, S. A.; Brabec, C. J. Nano Lett. 2008, 8, 2806.

Hoth, C. N.; Steim, R.; Schilinsky, P.; Choulis, S. A.; Tedde, S. F.; Hayden, O.; Brabec,
C. J. Org. Electron. 2009, 10, 587.

Hou, J.; Chen, H.-Y.; Zhang, S.; Chen, R. I.; Yang, Y.; Wu, Y.; Li, G. J. Am. Chem. Soc.
2009, 131, 15586.

Hou, J.; Chen, T. L.; Zhang, S.; Huo, L.; Sista, S.; Yang, Y. Macromolecules 2009, 42,
9217.
230
Howard, I. A.; Laquai, F.; Keivanidis, P. E.; Friend, R. H.; Greenham, N. C. J. Phys.
Chem. C 2009, 113, 21225.

Hu, Z.; Tang, S.; Ahlvers, A.; Khondaker, S. I.; Gesquiere, A. J. Appl. Phys. Lett. 2012,
101, 053308.

Huang, J.-H.; Velusamy, M.; Ho, K.-C.; Lin, J.-T.; Chu, C.-W. J. Mater. Chem. 2010, 20,
2820.

Huang, J.-S.; Goh, T.; Li, X.; Sfeir, M. Y.; Bielinski, E. A.; Tomasulo, S.; Lee, M. L.;
Hazari, N.; Taylor, A. D. Nat. Photonics 2013, DOI: 10.1038/nphoton.2013.82.

Huang, Y.; Westenhoff, S.; Avilov, I.; Sreearunothai, P.; Hodgkiss, J. M.; Deleener, C.;
Friend, R. H.; Beljonne, D. Nat. Mater. 2008, 7, 483.

Huo, L.; Hou, J.; Chen, H.-Y.; Zhang, S.; Jiang, Y.; Chen, T. L.; Yang, Y.
Macromolecules 2009, 42, 6564.

Imamura, S. J. Photon. Energy 2011, 1, 011119.

Inganäs, O.; Zhang, F.; Tvingstedt, K.; Andersson, L. M.; Hellström, S.; Andersson, M.
R. Adv. Mater. 2010, 22, E100.

Ito, S.; Ohkita, H.; Benten, H.; Honda, S. AMBIO 2012, 41, 132.

Jacobson, M. Z.; Delucchi, M. A. Energy Policy 2011, 39, 1154.

Jäger-Waldau, A. Green 2011, 1, 277.

Jäger-Waldau, A. Int. J. Photoenergy 2012, 2012, 1.

Jailaubekov, A. E.; Willard, A. P.; Tritsch, J. R.; Chan, W.-L.; Sai, N.; Gearba, R.;
Kaake, L. G.; Williams, K. J.; Leung, K.; Rossky, P. J.; Zhu, X.-Y. Nat. Mater.
2012, 12, 66.

Janssen, R. A. J.; Nelson, J. Adv. Mater. 2013, 25, 1847.

Jarzab, D.; Cordella, F.; Lenes, M.; Kooistra, F. B.; Blom, P. W. M.; Hummelen, J. C.;
Loi, M. A. J. Phys. Chem. B 2009, 113, 16513.

Jarzab, D.; Szendrei, K.; Yarema, M.; Pichler, S.; Heiss, W.; Loi, M. A. Adv. Funct.
Mater. 2011, 21, 1988.

231
Jeltsch, K. F.; Schädel, M.; Bonekamp, J.-B.; Niyamakom, P.; Rauscher, F.; Lademann,
H. W. A.; Dumsch, I.; Allard, S.; Scherf, U.; Meerholz, K. Adv. Funct. Mater.
2012, 22, 397.

Jeong, S.; Woo, S.-H.; Lyu, H.-K.; Han, Y. S. Sol. Energy Mater. Sol. Cells 2011, 95,
1908.

Jo, J.; Gendron, D.; Najari, A.; Moon, J. S.; Cho, S.; Leclerc, M.; Heeger, A. J. Appl.
Phys. Lett. 2010, 97, 203303.

Johansson, E. M. J.; Yartsev, A.; Rensmo, H.; Sundström, V. J. Phys. Chem. C 2009,
113, 3014.

Jørgensen, M.; Norrman, K.; Gevorgyan, S. A.; Tromholt, T.; Andreasen, B.; Krebs, F.
C. Adv. Mater. 2012, 24, 580.

Jørgensen, M.; Norrman, K.; Krebs, F. C. Sol. Energy Mater. Sol. Cells 2008, 92, 686.

Jung, H. S.; Lee, J.-K. J. Phys. Chem. Lett. 2013, 4, 1682.

Jung, J. W.; Jo, J. W.; Jo, W. H. Adv. Mater. 2011, 23, 1782.

Jung, J. W.; Jo, W. H. Adv. Funct. Mater. 2010, 20, 2355.

Juska, G.; Arlauskas, K.; Stuchlik, J.; Osterbacka, R. J. Non-Cryst. Solids 2006, 352,
1167.

Juška, G.; Genevičius, K.; Nekrašas, N.; Sliaužys, G.; Österbacka, R. Appl. Phys. Lett.
2009, 95, 013303.

Kaake, L.; Dang, X.-D.; Leong, W. L.; Zhang, Y.; Heeger, A.; Nguyen, T.-Q. Adv.
Mater. 2013, 25, 1706.

Kaake, L. G.; Jasieniak, J. J.; Bakus, R. C.; Welch, G. C.; Moses, D.; Bazan, G. C.;
Heeger, A. J. J. Am. Chem. Soc. 2012, 134, 19828.

Kanai, Y.; Grossman, J. C. Nano Lett. 2007, 7, 1967.

Kang, H.; Kim, K.-H.; Kang, T. E.; Cho, C.-H.; Park, S.; Yoon, S. C.; Kim, B. J. ACS
Appl. Mater. Interfaces 2013, 5, 4401.

Kästner, C.; Rathgeber, S.; Egbe, D. A. M.; Hoppe, H. J. Mater. Chem. A 2013, 1, 3961.

232
Khan, M. T.; Bhargav, R.; Kaur, A.; Dhawan, S. K.; Chand, S. Thin Solid Films 2010,
519, 1007.

Khlyabich, P. P.; Burkhart, B.; Ng, C. F.; Thompson, B. C. Macromolecules 2011, 44,
5079.

Khlyabich, P. P.; Burkhart, B.; Thompson, B. C. J. Am. Chem. Soc. 2011, 133, 14534.

Khlyabich, P. P.; Burkhart, B.; Thompson, B. C. J. Am. Chem. Soc. 2012, 134, 9074.

Kietzke, T.; Hörhold, H.-H.; Neher, D. Chem. Mater. 2005, 17, 6532.

Kim, B.-G.; Jeong, E. J.; Park, H. J.; Bilby, D.; Guo, L. J.; Kim, J. ACS Appl. Mater.
Interfaces 2011, 3, 674.

Kim, B. J.; Miyamoto, Y.; Ma, B.; Fréchet, J. M. J. Adv. Funct. Mater. 2009, 19, 2273.

Kim, C. S.; Tinker, L. L.; DiSalle, B. F.; Gomez, E. D.; Lee, S.; Bernhard, S.; Loo, Y.-L.
Adv. Mater. 2009, 21, 3110.

Kim, J. Y.; Lee, K.; Coates, N. E.; Moses, D.; Nguyen, T.-Q.; Dante, M.; Heeger, A. J.
Science 2007, 317, 222.

Kim, J. Y.; Kim, S. H.; Lee, H.-H.; Lee, K.; Ma, W.; Gong, X.; Heeger, A. J. Adv. Mater.
2006, 18, 572.

Kim, M.-S.; Kim, B.-G.; Kim, J. ACS Appl. Mater. Interfaces 2009, 1, 1264.

Kim, S.-S.; Na, S.-I.; Kang, S.-J.; Kim, D.-Y. Sol. Energy Mater. Sol. Cells 2010, 94,
171.

Kim, Y.; Cook, S.; Tuladhar, S. M.; Choulis, S. A.; Nelson, J.; Durrant, J. R.; Bradley, D.
D. C.; Giles, M.; McCulloch, I.; Ha, C.-S.; Ree, M. Nat. Mater. 2006, 5, 197.

Kim, Y.; Nelson, J.; Zhang, T.; Cook, S.; Durrant, J. R.; Kim, H.; Park, J.; Shin, M.;
Nam, S.; Heeney, M.; McCulloch, I.; Ha, C.-S.; Bradley, D. D. C. ACS Nano 2009,
3, 2557.

Kim, Y.; Shin, M.; Kim, H.; Ha, Y.; Ha, C.-S. J. Phys. D: Appl. Phys. 2008, 41, 225101.

King, S.; Sommer, M.; Huettner, S.; Thelakkat, M.; Haque, S. A. J. Mater. Chem. 2009,
19, 5436.

Kippelen, B.; Brédas, J.-L. Energy Environ. Sci. 2009, 2, 251.
233
Kirchartz, T.; Taretto, K.; Rau, U. J. Phys. Chem. C 2009, 113, 17958.

Kline, R. J.; DeLongchamp, D. M.; Fischer, D. A.; Lin, E. K.; Richter, L. J.; Chabinyc,
M. L.; Toney, M. F.; Heeney, M.; McCulloch, I. Macromolecules 2007, 40, 7960.

Knupfer, M. Appl. Phys. A: Mater. Sci. Process. 2003, 77, 623.

Ko, S.; Verploegen, E.; Hong, S.; Mondal, R.; Hoke, E. T.; Toney, M. F.; McGehee, M.
D.; Bao, Z. J. Am. Chem. Soc. 2011, 133, 16722.

Koppe, M.; Egelhaaf, H.-J.; Clodic, E.; Morana, M.; Lüer, L.; Troeger, A.; Sgobba, V.;
Guldi, D. M.; Ameri, T.; Brabec, C. J. Adv. Energy Mater. 2013, DOI:
10.1002/aenm.201201076.

Koppe, M.; Egelhaaf, H.-J.; Dennler, G.; Scharber, M. C.; Brabec, C. J.; Schilinsky, P.;
Hoth, C. N. Adv. Funct. Mater. 2010, 20, 338.

Koppolu, V. R.; Gupta, M. C.; Bagienski, W.; Shen, Y.; Shu, C.; Gibson, H. W.; Dorn,
H. C. In Conf. Rec. IEEE Photovoltaic Spec. Conf., 34th; Philadelphia, PA, USA,
2009; p. 001252.

Koster, L. J. A.; Mihailetchi, V. D.; Blom, P. W. M. Appl. Phys. Lett. 2006, 88, 052104.

Koster, L. J. A.; Shaheen, S. E.; Hummelen, J. C. Adv. Energy Mater. 2012, 2, 1246.

Kotlarski, J. D.; Blom, P. W. M. Appl. Phys. Lett. 2011, 98, 053301.

Kotlarski, J. D.; Blom, P. W. M. Appl. Phys. Lett. 2012, 100, 013306.

Kotlarski, J. D.; Moet, D. J. D.; Blom, P. W. M. J. Polym. Sci. B Polym. Phys. 2011, 49,
708.

Kozycz, L. M.; Gao, D.; Hollinger, J.; Seferos, D. S. Macromolecules 2012, 45, 5823.

Krause, S.; Casu, M. B.; Schöll, A.; Umbach, E. New J. Phys. 2008, 10, 085001.

Krebs, F. C. Sol. Energy Mater. Sol. Cells 2008, 92, 715.

Krebs, F. C. Org. Electron. 2009, 10, 761.

Krebs, F. C. Sol. Energy Mater. Sol. Cells 2009, 93, 394.

Krebs, F. C.; Gevorgyan, S. A.; Alstrup, J. J. Mater. Chem. 2009, 19, 5442.

234
Krebs, F. C.; Jørgensen, M.; Norrman, K.; Hagemann, O.; Alstrup, J.; Nielsen, T. D.;
Fyenbo, J.; Larsen, K.; Kristensen, J. Sol. Energy Mater. Sol. Cells 2009, 93, 422.

Krebs, F. C.; Norrman, K. ACS Appl. Mater. Interfaces 2010, 2, 877.

Krebs, F. C.; Spanggaard, H. Chem. Mater. 2005, 17, 5235.

Krebs, F. C.; Tromholt, T.; Jørgensen, M. Nanoscale 2010, 2, 873.

Kroon, R.; Lenes, M.; Hummelen, J.; Blom, P.; de Boer, B. Polym. Rev. 2008, 48, 531.

Kubo, Y.; Watanabe, K.; Nishiyabu, R.; Hata, R.; Murakami, A.; Shoda, T.; Ota, H. Org.
Lett. 2011, 13, 4574.

Kularatne, R. S.; Magurudeniya, H. D.; Sista, P.; Biewer, M. C.; Stefan, M. C. J. Polym.
Sci., Part A: Polym. Chem. 2013, 51, 743.

Kumar, P.; Chand, S. Prog. Photovolt: Res. Appl. 2012, 20, 377.

Laird, D. W.; Richter, H.; Vejins, V.; Scott, L. T.; Lada, T. A. Synthesis and use of
fullerenes in polymer solar cells. WO 2009086210, July 2009.

Lebental, M.; Choukri, H.; Chénais, S.; Forget, S.; Siove, A.; Geffroy, B.; Tutiš, E. Phys.
Rev. B 2009, 79, 165318.

Lee, J. K.; Ma, W. L.; Brabec, C. J.; Yuen, J.; Moon, J. S.; Kim, J. Y.; Lee, K.; Bazan, G.
C.; Heeger, A. J. J. Am. Chem. Soc. 2008, 130, 3619.

Lee, J. U.; Jung, J. W.; Jo, J. W.; Jo, W. H. J. Mater. Chem. 2012, 22, 24265.

Lee, J. M.; Kwon, B.-H.; Park, H. I.; Kim, H.; Kim, M. G.; Park, J. S.; Kim, E. S.; Yoo,
S.; Jeon, D. Y.; Kim, S. O. Adv. Mater. 2013, 25, 2011.

Lee, J.; Yun, M. H.; Kim, J.; Kim, J. Y.; Yang, C. Macromol. Rapid Commun. 2012, 33,
140.

Lenes, M.; Kooistra, F. B.; Hummelen, J. C.; Van Severen, I.; Lutsen, L.; Vanderzande,
D.; Cleij, T. J.; Blom, P. W. M. J. Appl. Phys. 2008, 104, 114517.

Lenes, M.; Koster, L. J. A.; Mihailetchi, V. D.; Blom, P. W. M. Appl. Phys. Lett. 2006,
88, 243502.

Lenes, M.; Wetzelaer, G.-J. A. H.; Kooistra, F. B.; Veenstra, S. C.; Hummelen, J. C.;
Blom, P. W. M. Adv. Mater. 2008, 20, 2116.
235
Lennartz, C.; Idé, J.; Méreau, R.; Aurel, P.; Ducasse, L.; Castet, F.; Linares, M.;
Beljonne, D.; Cornil, J.; Lancaster, K.; Brédas, J.-L.; Verlaak, S.; Mityashin, A.;
Heremans, P.; Fuchs, A. J. Phys. Chem. C 2010, 114, 3215.

Leong, W. L.; Cowan, S. R.; Heeger, A. J. Adv. Energy Mater. 2011, 1, 517.

Lewis, A.; Ruseckas, A.; Gaudin, O.; Webster, G.; Burn, P.; Samuel, I. Org. Electron.
2006, 7, 452.

Li, C.-Y.; Wen, T.-C.; Lee, T.-H.; Guo, T.-F.; Huang, J.-C.-A.; Lin, Y.-C.; Hsu, Y.-J. J.
Mater. Chem. 2009, 19, 1643.

Li, G.; Shrotriya, V.; Huang, J.; Yao, Y.; Moriarty, T.; Emery, K.; Yang, Y. Nat. Mater.
2005, 4, 864.

Li, G.; Shrotriya, V.; Yao, Y.; Huang, J.; Yang, Y. J. Mater. Chem. 2007, 17, 3126.

Li, G.; Zhu, R.; Yang, Y. Nat. Photonics 2012, 6, 153.

Li, H.; Zhang, Z.-G.; Li, Y.; Wang, J. Appl. Phys. Lett. 2012, 101, 163302.

Li, N.; Machui, F.; Waller, D.; Koppe, M.; Brabec, C. J. Sol. Energy Mater. Sol. Cells
2011, 95, 3465.

Li, W.; Furlan, A.; Hendriks, K. H.; Wienk, M. M.; Janssen, R. A. J. J. Am. Chem. Soc.
2013, 135, 5529.

Li, W.; Qin, R.; Zhou, Y.; Andersson, M.; Li, F.; Zhang, C.; Li, B.; Liu, Z.; Bo, Z.;
Zhang, F. Polymer 2010, 51, 3031.

Li, Z.; Zhang, Y.; Tsang, S.-W.; Du, X.; Zhou, J.; Tao, Y.; Ding, J. J. Phys. Chem. C
2011, 115, 18002.

Liang, Y.; Feng, D.; Wu, Y.; Tsai, S.-T.; Li, G.; Ray, C.; Yu, L. J. Am. Chem. Soc. 2009,
131, 7792.

Liang, Y.; Xu, Z.; Xia, J.; Tsai, S.-T.; Wu, Y.; Li, G.; Ray, C.; Yu, L. Adv. Mater. 2010,
22, E135.

Lim, E.; Lee, S.; Lee, K. K. Chem. Commun. 2011, 47, 914.

Lin, Y.; Li, Y.; Zhan, X. Chem. Soc. Rev. 2012, 41, 4245.

Lin Leong, W.; Cowan, S. R.; Heeger, A. J. Adv. Energy Mater. 2011, 1, 517.
236
Liu, H.-W.; Chang, D.-Y.; Chiu, W.-Y.; Rwei, S.-P.; Wang, L. J. Mater. Chem. 2012, 22,
15586.

Liu, J.; Han, Y. Polym. Bull. 2012, 68, 2145.

Lizin, S.; Van Passel, S.; De Schepper, E.; Vranken, L. Sol. Energy Mater. Sol. Cells
2012, 103, 1.

Lobez, J. M.; Andrew, T. L.; Bulović, V.; Swager, T. M. ACS Nano 2012, 6, 3044.

Lou, Y.; Wang, Z.; Naka, S.; Okada, H. Appl. Phys. Lett. 2011, 99, 033305.

Lu, K.; Fang, J.; Zhu, X.; Yan, H.; Li, D.; Di, C.; Yang, Y.; Wei, Z. New J. Chem. 2013,
37, 1728.

Lunt, R. R.; Giebink, N. C.; Belak, A. A.; Benziger, J. B.; Forrest, S. R. J. Appl. Phys.
2009, 105, 053711.

Lunt, R. R.; Osedach, T. P.; Brown, P. R.; Rowehl, J. A.; Bulović, V. Adv. Mater. 2011,
23, 5712.

Ma, W.; Yang, C.; Gong, X.; Lee, K.; Heeger, A. J. Adv. Funct. Mater. 2005, 15, 1617.

Ma, W.; Kim, J. Y.; Lee, K.; Heeger, A. J. Macromol. Rapid Commun. 2007, 28, 1776.

Machui, F.; Abbott, S.; Waller, D.; Koppe, M.; Brabec, C. J. Macromol. Chem. Phys.
2011, 212, 2159.

Machui, F.; Rathgeber, S.; Li, N.; Ameri, T.; Brabec, C. J. J. Mater. Chem. 2012, 22,
15570.

Mandoc, M. M.; Kooistra, F. B.; Hummelen, J. C.; de Boer, B.; Blom, P. W. M. Appl.
Phys. Lett. 2007, 91, 263505.

Markov, D. E.; Amsterdam, E.; Blom, P. W. M.; Sieval, A. B.; Hummelen, J. C. J. Phys.
Chem. A 2005, 109, 5266.

Mauer, R.; Kastler, M.; Laquai, F. Adv. Funct. Mater. 2010, 20, 2085.

Maurano, A.; Hamilton, R.; Shuttle, C. G.; Ballantyne, A. M.; Nelson, J.; O’Regan, B.;
Zhang, W.; McCulloch, I.; Azimi, H.; Morana, M.; Brabec, C. J.; Durrant, J. R.
Adv. Mater. 2010, 22, 4987.

237
Mayukh, M.; Jung, I. H.; He, F.; Yu, L. J. Polym. Sci., Part B: Polym. Phys. 2012, 50,
1057.

McNeill, C. R.; Westenhoff, S.; Groves, C.; Friend, R. H.; Greenham, N. C. J. Phys.
Chem. C 2007, 111, 19153.

McNeill, C. R.; Abrusci, A.; Zaumseil, J.; Wilson, R.; McKiernan, M. J.; Burroughes, J.
H.; Halls, J. J. M.; Greenham, N. C.; Friend, R. H. Appl. Phys. Lett. 2007, 90,
193506.

McNeill, C. R.; Greenham, N. C. Adv. Mater. 2009, 21, 3840.

Medford, A. J.; Lilliedal, M. R.; Jørgensen, M.; Aarø, D.; Pakalski, H.; Fyenbo, J.;
Krebs, F. C. Optics Express 2010, 18, A272.

Meng, K.; Ding, Q.; Wang, S.; He, Y.; Li, Y.; Gong, Q. J. Phys. Chem. B 2010, 114,
2602.

Mihailetchi, V.; Wildeman, J.; Blom, P. Phys. Rev. Lett. 2005, 94, 126602.

Mikroyannidis, J. A.; Tsagkournos, D. V.; Balraju, P.; Sharma, G. D. J. Power Sources
2011, 196, 2364.

Min, J.; Ameri, T.; Gresser, R.; Lorenz-Rothe, M.; Baran, D.; Troeger, A.; Sgobba, V.;
Leo, K.; Riede, M.; Guldi, D. M.; Brabec, C. J. ACS Appl. Mater. Interfaces 2013,
DOI: 10.1021/am400923b.

Min Nam, Y.; Huh, J.; Ho Jo, W. Sol. Energy Mater. Sol. Cells 2010, 94, 1118.

Mishra, A.; Bäuerle, P. Angew. Chem. Int. Ed. 2012, 51, 2020.

Miyanishi, S.; Tajima, K.; Hashimoto, K. Macromolecules 2009, 42, 1610.

Moses, P. G.; Van de Walle, C. G. Appl. Phys. Lett. 2010, 96, 021908.

Moulé, A. J.; Chang, L.; Thambidurai, C.; Vidu, R.; Stroeve, P. J. Mater. Chem. 2012,
22, 2351.

Moulé, A. J.; Meerholz, K. Adv. Funct. Mater. 2009, 19, 3028.

Müller, C.; Ferenczi, T. A. M.; Campoy-Quiles, M.; Frost, J. M.; Bradley, D. D. C.;
Smith, P.; Stingelin-Stutzmann, N.; Nelson, J. Adv. Mater. 2008, 20, 3510.

238
Naidu, B. V. K.; Park, J. S.; Kim, S. C.; Park, S.-M.; Lee, E.-J.; Yoon, K.-J.; Joon Lee,
S.; Wook Lee, J.; Gal, Y.-S.; Jin, S.-H. Sol. Energy Mater. Sol. Cells 2008, 92,
397.

Nayak, P. K.; Garcia-Belmonte, G.; Kahn, A.; Bisquert, J.; Cahen, D. Energy Environ.
Sci. 2012, 5, 6022.

Nguyen, L. H.; Hoppe, H.; Erb, T.; Günes, S.; Gobsch, G.; Sariciftci, N. S. Adv. Funct.
Mater. 2007, 17, 1071.

Nilsson, S.; Bernasik, A.; Budkowski, A.; Moons, E. Macromolecules 2007, 40, 8291.

Norrman, K.; Gevorgyan, S. A.; Krebs, F. C. ACS Appl. Mater. Interfaces 2009, 1, 102.

Nuzzo, D. D.; Aguirre, A.; Shahid, M.; Gevaerts, V. S.; Meskers, S. C. J.; Janssen, R. A.
J. Adv. Mater. 2010, 22, 4321.

Ogawa, K.; Guo, F.; Schanze, K. S. J. Photochem. Photobiol., A 2009, 207, 79.

Ohkita, H.; Cook, S.; Astuti, Y.; Duffy, W.; Tierney, S.; Zhang, W.; Heeney, M.;
McCulloch, I.; Nelson, J.; Bradley, D. D. C.; Durrant, J. R. J. Am. Chem. Soc.
2008, 130, 3030.

O’Regan, B.; Grätzel, M. Nature 1991, 353, 737.

Osaka, I.; McCullough, R. D. Acc. Chem. Res. 2008, 41, 1202.

Paci, B.; Generosi, A.; Rossi Albertini, V.; Perfetti, P.; de Bettignies, R.; Sentein, C. Sol.
Energy Mater. Sol. Cells 2008, 92, 799.

Pal, S. K.; Kesti, T.; Maiti, M.; Zhang, F.; Inganäs, O.; Hellström, S.; Andersson, M. R.;
Oswald, F.; Langa, F.; Österman, T.; Pascher, T.; Yartsev, A.; Sundström, V. J.
Am. Chem. Soc. 2010, 132, 12440.

Panda, P.; Veldman, D.; Sweelssen, J.; Bastiaansen, J. J. A. M.; Langeveld-Voss, B. M.
W.; Meskers, S. C. J. J. Phys. Chem. B 2007, 111, 5076.

Park, M.-J.; Lee, J.; Kwak, J.; Jung, I. H.; Park, J.-H.; Kong, H.; Lee, C.; Hwang, D.-H.;
Shim, H.-K. Macromolecules 2009, 42, 5551.

Park, S. J.; Cho, J. M.; Byun, W.-B.; Lee, J.-C.; Shin, W. S.; Kang, I.-N.; Moon, S.-J.;
Lee, S. K. J. Polym. Sci., Part A: Polym. Chem. 2011, 49, 4416.

239
Park, S. H.; Roy, A.; Beaupré, S.; Cho, S.; Coates, N.; Moon, J. S.; Moses, D.; Leclerc,
M.; Lee, K.; Heeger, A. J. Nat. Photonics 2009, 3, 297.

Peet, J.; Kim, J. Y.; Coates, N. E.; Ma, W. L.; Moses, D.; Heeger, A. J.; Bazan, G. C.
Nat. Mater. 2007, 6, 497.

Peet, J.; Tamayo, A. B.; Dang, X.-D.; Seo, J. H.; Nguyen, T.-Q. Appl. Phys. Lett. 2008,
93, 163306.

Peet, J.; Senatore, M. L.; Heeger, A. J.; Bazan, G. C. Adv. Mater. 2009, 21, 1521.

Pensack, R. D.; Asbury, J. B. J. Phys. Chem. Lett. 2010, 1, 2255.

Perez, M. D.; Borek, C.; Forrest, S. R.; Thompson, M. E. J. Am. Chem. Soc. 2009, 131,
9281.

Peters, C. H.; Sachs-Quintana, I. T.; Kastrop, J. P.; Beaupré, S.; Leclerc, M.; McGehee,
M. D. Adv. Energy Mater. 2011, 1, 491.

Piliego, C.; Holcombe, T. W.; Douglas, J. D.; Woo, C. H.; Beaujuge, P. M.; Fréchet, J.
M. J. J. Am. Chem. Soc. 2010, 132, 7595.

Piris, J.; Dykstra, T. E.; Bakulin, A. A.; Loosdrecht, P. H. M. van; Knulst, W.; Trinh, M.
T.; Schins, J. M.; Siebbeles, L. D. A. J. Phys. Chem. C 2009, 113, 14500.

Pivrikas, A.; Juška, G.; Mozer, A.; Scharber, M.; Arlauskas, K.; Sariciftci, N.; Stubb, H.;
Österbacka, R. Phys. Rev. Lett. 2005, 94, 176806.

Pivrikas, A.; Sariciftci, N. S.; Juška, G.; Österbacka, R. Prog. Photovolt: Res. Appl. 2007,
15, 677.

Price, S. C.; Stuart, A. C.; Yang, L.; Zhou, H.; You, W. J. Am. Chem. Soc. 2011, 133,
4625.

Qi, B.; Wang, J. J. Mater. Chem. 2012, 22, 24315.

Qi, B.; Wang, J. Phys. Chem. Chem. Phys. 2013, 15, 8972.

Rajaram, S.; Armstrong, P. B.; Kim, B. J.; Fréchet, J. M. J. Chem. Mater. 2009, 21, 1775.

Razykov, T. M.; Ferekides, C. S.; Morel, D.; Stefanakos, E.; Ullal, H. S.; Upadhyaya, H.
M. Sol. Energy 2011, 85, 1580.

240
Reese, M. O.; Morfa, A. J.; White, M. S.; Kopidakis, N.; Shaheen, S. E.; Rumbles, G.;
Ginley, D. S. Sol. Energy Mater. Sol. Cells 2008, 92, 746.

Reese, M. O.; Nardes, A. M.; Rupert, B. L.; Larsen, R. E.; Olson, D. C.; Lloyd, M. T.;
Shaheen, S. E.; Ginley, D. S.; Rumbles, G.; Kopidakis, N. Adv. Funct. Mater.
2010, 20, 3476.

Richards, J. J.; Rice, A. H.; Nelson, R. D.; Kim, F. S.; Jenekhe, S. A.; Luscombe, C. K.;
Pozzo, D. C. Adv. Funct. Mater. 2013, 23, 514.

Robeson, L. M. Polymer blends: a comprehensive review; Carl Hanser Verlag: Munich,
2007.

Rockett, A. Materials science of semiconductors; Springer: New York, 2008.

Scharber, M. C.; Mühlbacher, D.; Koppe, M.; Denk, P.; Waldauf, C.; Heeger, A. J.;
Brabec, C. J. Adv. Mater. 2006, 18, 789.

Scharber, M. C.; Sariciftci, N. S. Prog. Polym. Sci. 2013, DOI:
10.1016/j.progpolymsci.2013.05.001.

Schilinsky, P.; Waldauf, C.; Brabec, C. J. Adv. Funct. Mater. 2006, 16, 1669.

Schilinsky, P.; Asawapirom, U.; Scherf, U.; Biele, M.; Brabec, C. J. Chem. Mater. 2005,
17, 2175.

Schlenker, C. W.; Thompson, M. E. Top. Curr. Chem. 2011, 312, 175.

Schlenker, C. W.; Thompson, M. E. Chem. Commun. 2011, 47, 3702.

Schmidtke, J. Opt. Express 2010, 18, A477.

Scholes, G. D. ACS Nano 2008, 2, 523.

Scholes, G. D.; Rumbles, G. Nat. Mater. 2006, 5, 683.

Scully, S. R.; McGehee, M. D. J. Appl. Phys. 2006, 100, 034907.

Segalman, R. A.; McCulloch, B.; Kirmayer, S.; Urban, J. J. Macromolecules 2009, 42,
9205.

Servaites, J. D.; Ratner, M. A.; Marks, T. J. Appl. Phys. Lett. 2009, 95, 163302.

Service, R. F. Science 2011, 332, 293.
241
Shaheen, S. E.; Brabec, C. J.; Sariciftci, N. S.; Padinger, F.; Fromherz, T.; Hummelen, J.
C. Appl. Phys. Lett. 2001, 78, 841.

Shaheen, S. E.; Ginley, D. S.; Jabbour, G. E. MRS Bull. 2005, 30, 10.

Sharma, G. D.; Singh, S. P.; Roy, M. S.; Mikroyannidis, J. A. Org. Electron. 2012, 13,
1756.

Sharma, S. S.; Sharma, G. D.; Mikroyannidis, J. A. Sol. Energy Mater. Sol. Cells 2011,
95, 1219.

Shin, M.; Kim, H.; Nam, S.; Park, J.; Kim, Y. Energy Environ. Sci. 2010, 3, 1538.

Shoaee, S.; An, Z.; Zhang, X.; Barlow, S.; Marder, S. R.; Duffy, W.; Heeney, M.;
McCulloch, I.; Durrant, J. R. Chem. Commun. 2009, 5445.

Shoaee, S.; Clarke, T. M.; Huang, C.; Barlow, S.; Marder, S. R.; Heeney, M.;
McCulloch, I.; Durrant, J. R. J. Am. Chem. Soc. 2010, 132, 12919.

Shoaee, S.; Eng, M. P.; Espíldora, E.; Delgado, J. L.; Campo, B.; Martín, N.;
Vanderzande, D.; Durrant, J. R. Energy Environ. Sci. 2010, 3, 971.

Shrotriya, V.; Li, G.; Yao, Y.; Moriarty, T.; Emery, K.; Yang, Y. Adv. Funct. Mater.
2006, 16, 2016.

Shrotriya, V.; Yao, Y.; Li, G.; Yang, Y. Appl. Phys. Lett. 2006, 89, 063505.

Siddiki, M. K.; Li, J.; Galipeau, D.; Qiao, Q. Energy Environ. Sci. 2010, 3, 867.

Singh, T. B.; Chen, X.; Wong, W. W. H.; Ehlig, T.; Kemppinen, P.; Chen, M.; Watkins,
S. E.; Winzenberg, K. N.; Holdcroft, S.; Jones, D. J.; Holmes, A. B. Appl. Phys. A:
Mater. Sci. Process. 2012, 108, 515.

Sista, S.; Hong, Z.; Chen, L.-M.; Yang, Y. Energy Environ. Sci. 2011, 4, 1606.

Sivula, K.; Ball, Z. T.; Watanabe, N.; Fréchet, J. M. J. Adv. Mater. 2006, 18, 206.

Sivula, K.; Luscombe, C. K.; Thompson, B. C.; Fréchet, J. M. J. J. Am. Chem. Soc. 2006,
128, 13988.

Small, C. E.; Chen, S.; Subbiah, J.; Amb, C. M.; Tsang, S.-W.; Lai, T.-H.; Reynolds, J.
R.; So, F. Nat. Photonics 2011, 6, 115.

Smith, M. B.; Michl, J. Chem. Rev. 2010, 110, 6891.
242
Snaith, H. J. Adv. Funct. Mater. 2010, 20, 13.

Snaith, H. J. Energy Environ. Sci. 2012, 5, 6513.

Snyder, C. R.; Henry, J. S.; DeLongchamp, D. M. Macromolecules 2011, 44, 7088.

Somanathan, N.; Radhakrishnan, S.; Thelakkat, M.; Schmidt, H. W. Macromol. Mater.
Eng. 2002, 287, 236.

Sommer, M.; Lindner, S. M.; Thelakkat, M. Adv. Funct. Mater. 2007, 17, 1493.

Sommer, M.; Lang, A. S.; Thelakkat, M. Angew. Chem., Int. Ed. 2008, 47, 7901.

Son, H. J.; Carsten, B.; Jung, I. H.; Yu, L. Energy Environ. Sci. 2012, 5, 8158.

Son, H. J.; He, F.; Carsten, B.; Yu, L. J. Mater. Chem. 2011, 21, 18934.

Søndergaard, R.; Hösel, M.; Angmo, D.; Larsen-Olsen, T. T.; Krebs, F. C. Mater. Today
2012, 15, 36.

Søndergaard, R. R.; Hösel, M.; Krebs, F. C. J. Polym. Sci., Part B: Polym. Phys. 2013,
51, 16.

Stevens, D. M.; Speros, J. C.; Hillmyer, M. A.; Frisbie, C. D. J. Phys. Chem. C 2011,
115, 20806.

Street, R.; Cowan, S.; Heeger, A. Phys. Rev. B 2010, 82, 121301.

Street, R.; Song, K.; Northrup, J.; Cowan, S. Phys. Rev. B 2011, 83, 165207.

Street, R. A.; Schoendorf, M.; Roy, A.; Lee, J. H. Phys. Rev. B 2010, 81, 205307.

Street, R. A.; Davies, D.; Khlyabich, P. P.; Burkhart, B.; Thompson, B. C. J. Am. Chem.
Soc. 2013, 135, 986.

Street, R. A.; Krakaris, A.; Cowan, S. R. Adv. Funct. Mater. 2012, 22, 4608.

Su, Y.-W.; Lan, S.-C.; Wei, K.-H. Mater. Today 2012, 15, 554.

Sun, B.; Greenham, N. C. Phys. Chem. Chem. Phys. 2006, 8, 3557.

Suresh, P.; Balraju, P.; Sharma, G. D.; Mikroyannidis, J. A.; Stylianakis, M. M. ACS
Appl. Mater. Interfaces 2009, 1, 1370.

243
Susarova, D. K.; Troshin, P. A.; Moskvin, Y. L.; Babenko, S. D.; Razumov, V. F. Thin
Solid Films 2011, 519, 4132.

Suzuki, S.; Sugimura, R.; Kozaki, M.; Keyaki, K.; Nozaki, K.; Ikeda, N.; Akiyama, K.;
Okada, K. J. Am. Chem. Soc. 2009, 131, 10374.

Szarko, J. M.; Guo, J.; Liang, Y.; Lee, B.; Rolczynski, B. S.; Strzalka, J.; Xu, T.; Loser,
S.; Marks, T. J.; Yu, L.; Chen, L. X. Adv. Mater. 2010, 22, 5468.

Tai, Q.; Li, J.; Liu, Z.; Sun, Z.; Zhao, X.; Yan, F. J. Mater. Chem. 2011, 21, 6848.

Tamayo, A. B.; Walker, B.; Nguyen, T.-Q. J. Phys. Chem. C 2008, 112, 11545.

Tang, C. W. Appl. Phys. Lett. 1986, 48, 183.

Tao, M. Electrochem. Soc. Interface 2008, 17, 30.

Tayebjee, M. J. Y.; Gray-Weale, A. A.; Schmidt, T. W. J. Phys. Chem. Lett. 2012, 3,
2749.

Thompson, B. C.; Khlyabich, P. P.; Burkhart, B.; Aviles, A. E.; Rudenko, A.; Shultz, G.
V.; Ng, C. F.; Mangubat, L. B. Green 2011, 1, 29.

Thompson, B. C.; Kim, B. J.; Kavulak, D. F.; Sivula, K.; Mauldin, C.; Fréchet, J. M. J.
Macromolecules 2007, 40, 7425.

Thompson, B. C.; Kim, Y.-G.; McCarley, T. D.; Reynolds, J. R. J. Am. Chem. Soc. 2006,
128, 12714.

Thompson, B. C.; Kim, Y.-G.; Reynolds, J. R. Macromolecules 2005, 38, 5359.

Thompson, B. C.; Fréchet, J. M. J. Angew. Chem., Int. Ed. 2008, 47, 58.

Tieke, B.; Rabindranath, A. R.; Zhang, K.; Zhu, Y. Beilstein J. Org. Chem. 2010, 6, 830.

Treat, N. D.; Brady, M. A.; Smith, G.; Toney, M. F.; Kramer, E. J.; Hawker, C. J.;
Chabinyc, M. L. Adv. Energy Mater. 2011, 1, 82.

Treat, N. D.; Shuttle, C. G.; Toney, M. F.; Hawker, C. J.; Chabinyc, M. L. J. Mater.
Chem. 2011, 21, 15224.

Tress, W.; Merten, A.; Furno, M.; Hein, M.; Leo, K.; Riede, M. Adv. Energy Mater.
2013, 3, 631.

244
Tromholt, T.; Katz, E. A.; Hirsch, B.; Vossier, A.; Krebs, F. C. Appl. Phys. Lett. 2010,
96, 073501.

Tvingstedt, K.; Andersson, V.; Zhang, F.; Inganäs, O. Appl. Phys. Lett. 2007, 91, 123514.

Tvingstedt, K.; Vandewal, K.; Gadisa, A.; Zhang, F.; Manca, J.; Inganäs, O. J. Am.
Chem. Soc. 2009, 131, 11819.

Uy, R. L.; Price, S. C.; You, W. Macromol. Rapid Commun. 2012, 33, 1162.

Van Bavel, S.; Sourty, E.; de With, G.; Frolic, K.; Loos, J. Macromolecules 2009, 42,
7396.

Van Bavel, S.; Veenstra, S.; Loos, J. Macromol. Rapid Commun. 2010, 31, 1835.

Vandewal, K.; Gadisa, A.; Oosterbaan, W. D.; Bertho, S.; Banishoeib, F.; Van Severen,
I.; Lutsen, L.; Cleij, T. J.; Vanderzande, D.; Manca, J. V. Adv. Funct. Mater. 2008,
18, 2064.

Vandewal, K.; Tvingstedt, K.; Gadisa, A.; Inganäs, O.; Manca, J. V. Nat. Mater. 2009, 8,
904.

Vandewal, K.; Tvingstedt, K.; Gadisa, A.; Inganäs, O.; Manca, J. V. Phys. Rev. B 2010,
81, 125204.

Veldman, D.; Meskers, S. C. J.; Janssen, R. A. J. Adv. Funct. Mater. 2009, 19, 1939.

Veysel Tunc, A.; De Sio, A.; Riedel, D.; Deschler, F.; Da Como, E.; Parisi, J.; von Hauff,
E. Org. Electron. 2012, 13, 290.

Vohra, V.; Higashimine, K.; Murakami, T.; Murata, H. Appl. Phys. Lett. 2012, 101,
173301.

Walker, B.; Tamayo, A. B.; Dang, X.-D.; Zalar, P.; Seo, J. H.; Garcia, A.; Tantiwiwat,
M.; Nguyen, T.-Q. Adv. Funct. Mater. 2009, 19, 3063.

Wang, C.; Zhuang, L.-Q.; Chen, R.-A.; Li, S.; George, T. F. J. Phys. Chem. B 2013, 117,
3258.

Wang, D. H.; Kim, D. Y.; Choi, K. W.; Seo, J. H.; Im, S. H.; Park, J. H.; Park, O. O.;
Heeger, A. J. Angew. Chem. 2011, 123, 5633.

Wang, E.; Hou, L.; Wang, Z.; Hellstroem, S.; Zhang, F.; Inganaes, O.; Andersson, M. R.
Adv. Mater. 2010, 22, 5240.
245
Wang, E.; Hou, L.; Wang, Z.; Ma, Z.; Hellström, S.; Zhuang, W.; Zhang, F.; Inganäs, O.;
Andersson, M. R. Macromolecules 2011, 44, 2067.

Wei, Q.; Nishizawa, T.; Tajima, K.; Hashimoto, K. Adv. Mater. 2008, 20, 2211.

Westenhoff, S.; Howard, I.; Friend, R. Phys. Rev. Lett. 2008, 101, 016102.

Westenhoff, S.; Howard, I. A.; Hodgkiss, J. M.; Kirov, K. R.; Bronstein, H. A.; Williams,
C. K.; Greenham, N. C.; Friend, R. H. J. Am. Chem. Soc. 2008, 130, 13653.

Wienk, M. M.; Turbiez, M.; Gilot, J.; Janssen, R. A. J. Adv. Mater. 2008, 20, 2556.

Winder, C.; Matt, G.; Hummelen, J. C.; Janssen, R. A. J.; Sariciftci, N. S.; Brabec, C. J.
Thin Solid Films 2002, 403-404, 373.

Wong, W.-Y.; Wang, X.-Z.; He, Z.; Djurišić, A. B.; Yip, C.-T.; Cheung, K.-Y.; Wang,
H.; Mak, C. S. K.; Chan, W.-K. Nat. Mater. 2007, 6, 521.

Woo, C. H.; Beaujuge, P. M.; Holcombe, T. W.; Lee, O. P.; Fréchet, J. M. J. J. Am.
Chem. Soc. 2010, 132, 15547.

Woo, C. H.; Holcombe, T. W.; Unruh, D. A.; Sellinger, A.; Fréchet, J. M. J. Chem.
Mater. 2010, 22, 1673.

Woo, C. H.; Thompson, B. C.; Kim, B. J.; Toney, M. F.; Fréchet, J. M. J. J. Am. Chem.
Soc. 2008, 130, 16324.

Wright, M.; Uddin, A. Sol. Energy Mater. Sol. Cells 2012, 107, 87.

Wu, J.-C.; Zheng, J.; Zacherl, C. L.; Wu, P.; Liu, Z.-K.; Xu, R. J. Phys. Chem. C 2011,
115, 19741.

Wu, P.-T.; Ren, G.; Jenekhe, S. A. Macromolecules 2010, 43, 3306.

Wu, Y.-C.; Chao, Y.-H.; Wang, C.-L.; Wu, C.-T.; Hsu, C.-S.; Zeng, Y.-L.; Lin, C.-Y. J.
Polym. Sci., Part A: Polym. Chem. 2012, 50, 5032.

Xu, T.; Qiao, Q. Energy Environ. Sci. 2011, 4, 2700.

Xue, J.; Rand, B. P.; Uchida, S.; Forrest, S. R. J. Appl. Phys. 2005, 98, 124903.

Xue, J.; Uchida, S.; Rand, B. P.; Forrest, S. R. Appl. Phys. Lett. 2004, 84, 3013.

Yamanari, T.; Taima, T.; Sakai, J.; Saito, K. Sol. Energy Mater. Sol. Cells 2009, 93, 759.
246
Yang, C.; Hu, J. G.; Heeger, A. J. J. Am. Chem. Soc. 2006, 128, 12007.

Yang, L.; Yan, L.; You, W. J. Phys. Chem. Lett. 2013, 4, 1802.

Yang, L.; Zhou, H.; Price, S. C.; You, W. J. Am. Chem. Soc. 2012, 134, 5432.

Yang, L.; Zhou, H.; You, W. J. Phys. Chem. C 2010, 114, 16793.

Yang, X.; Loos, J.; Veenstra, S. C.; Verhees, W. J. H.; Wienk, M. M.; Kroon, J. M.;
Michels, M. A. J.; Janssen, R. A. J. Nano Lett. 2005, 5, 579.

Yao, K.; Chen, L.; Chen, X.; Chen, Y. Chem. Mater. 2013, 25, 897.

Yella, A.; Lee, H.-W.; Tsao, H. N.; Yi, C.; Chandiran, A. K.; Nazeeruddin, M. K.; Diau,
E. W.-G.; Yeh, C.-Y.; Zakeeruddin, S. M.; Gratzel, M. Science 2011, 334, 629.

Yi, Y.; Coropceanu, V.; Brédas, J.-L. J. Am. Chem. Soc. 2009, 131, 15777.

Yilmaz Canli, N.; Günes, S.; Pivrikas, A.; Fuchsbauer, A.; Sinwel, D.; Sariciftci, N. S.;
Yasa, Ö.; Bilgin-Eran, B. Sol. Energy Mater. Sol. Cells 2010, 94, 1089.

You, J.; Chen, C.-C.; Hong, Z.; Yoshimura, K.; Ohya, K.; Xu, R.; Ye, S.; Gao, J.; Li, G.;
Yang, Y. Adv. Mater. 2013, DOI: 10.1002/adma.201300964.

You, J.; Dou, L.; Hong, Z.; Li, G.; Yang, Y. Prog. Polym. Sci. 2013, DOI:
10.1016/j.progpolymsci.2013.04.005.

You, J.; Dou, L.; Yoshimura, K.; Kato, T.; Ohya, K.; Moriarty, T.; Emery, K.; Chen, C.-
C.; Gao, J.; Li, G.; Yang, Y. Nat. Commun. 2013, 4, 1446.

Yu, G.; Heeger, A. J. J. Appl. Phys. 1995, 78, 4510.

Yuan, Y.; Huang, J.; Li, G. Green 2011, 1, 65.

Yue, D.; Khatav, P.; You, F.; Darling, S. B. Energy Environ. Sci. 2012, 5, 9163.

Zhang, Q.; Cirpan, A.; Russell, T. P.; Emrick, T. Macromolecules 2009, 42, 1079.

Zhang, S.; Yang, X.; Numata, Y.; Han, L. Energy Environ. Sci. 2013, 6, 1443.

Zhang, W.; Wang, Y.-W.; Hu, R.; Fu, L.-M.; Ai, X.-C.; Zhang, J.-P.; Hou, J.-H. J. Phys.
Chem. C 2013, 117, 735.

Zhang, Y.; Tajima, K.; Hashimoto, K. Macromolecules 2009, 42, 7008.
247
Zhang, Y.; Tajima, K.; Hashimoto, K. Synth. Met. 2011, 161, 225.

Zhang, Y.; Tajima, K.; Hirota, K.; Hashimoto, K. J. Am. Chem. Soc. 2008, 130, 7812.

Zhang, Z.-G.; Li, H.; Qi, Z.; Jin, Z.; Liu, G.; Hou, J.; Li, Y.; Wang, J. Appl. Phys. Lett.
2013, 102, 143902.

Zhao, G.; He, Y.; Li, Y. Adv. Mater. 2010, 22, 4355.

Zhao, J.; Swinnen, A.; Van Assche, G.; Manca, J.; Vanderzande, D.; Mele, B. V. J. Phys.
Chem. B 2009, 113, 1587.

Zhao, X.; Piliego, C.; Kim, B.; Poulsen, D. A.; Ma, B.; Unruh, D. A.; Fréchet, J. M. J.
Chem. Mater. 2010, 22, 2325.

Zhou, E.; Wei, Q.; Yamakawa, S.; Zhang, Y.; Tajima, K.; Yang, C.; Hashimoto, K.
Macromolecules 2010, 43, 821.

Zhou, E.; Yamakawa, S.; Tajima, K.; Yang, C.; Hashimoto, K. Chem. Mater. 2009, 21,
4055.

Zhou, H.; Yang, L.; Stuart, A. C.; Price, S. C.; Liu, S.; You, W. Angew. Chem. Int. Ed.
2011, 50, 2995.

Zhou, H.; Yang, L.; You, W. Macromolecules 2012, 45, 607.

Zhou, Y.; Tvingstedt, K.; Zhang, F.; Du, C.; Ni, W.-X.; Andersson, M. R.; Inganäs, O.
Adv. Funct. Mater. 2009, 19, 3293.

Zhou, Y.; Li, F.; Barrau, S.; Tian, W.; Inganäs, O.; Zhang, F. Sol. Energy Mater. Sol.
Cells 2009, 93, 497.

Zhu, X.-Y.; Yang, Q.; Muntwiler, M. Acc. Chem. Res. 2009, 42, 1779.

Zhu, Y.; Champion, R. D.; Jenekhe, S. A. Macromolecules 2006, 39, 8712.

Zhu, Z.; Hadjikyriacou, S.; Waller, D.; Gaudiana, R. J. Macromol. Sci., Part A: Pure
Appl. Chem. 2004, 41, 1467.

Zoombelt, A. P.; Leenen, M. A. M.; Fonrodona, M.; Nicolas, Y.; Wienk, M. M.; Janssen,
R. A. J. Polymer 2009, 50, 4564.

Zoombelt, A. P.; Mathijssen, S. G. J.; Turbiez, M. G. R.; Wienk, M. M.; Janssen, R. A. J.
J. Mater. Chem. 2010, 20, 2240.
248
Zou, Y.; Gendron, D.; Neagu-Plesu, R.; Leclerc, M. Macromolecules 2009, 42, 6361.

249
Appendix A

Ternary Blend Bulk Heterojunction Solar Cells in the Case of Complementary
Acceptors

A.1 Materials and Methods

All reagents from commercial sources were used without further purification,
unless otherwise noted. All reactions were performed under dry N
2
, unless otherwise
noted. All dry reactions were performed with glassware that was oven dried and then
flamed under high vacuum and backfilled with N
2
. Flash chromatography was performed
using a Teledyne CombiFlash R
f
instrument in combination with RediSep R
f
normal
phase disposable columns. Solvents were purchased from VWR and used without further
purification except for THF which was dried over sodium/benzophenone before being
distilled.
All compounds were characterized by
1
H NMR (400 MHz) and
13
C NMR (100
MHz) on a Mercury 400. Polymer
1
H NMRs (500 MHz) were obtained on a Varian
VNMRS-500. For polymer molecular weight determination, polymer samples were
dissolved in HPLC grade o-dichlorobenzene at a concentration of 1 mg/ml, briefly heated
and then allowed to turn to room temperature prior to filtering through a 0.2 μm PTFE
filter. SEC was performed using HPLC grade o-dichlorobenzene at a flow rate of 1
ml/min on one 300 × 7.8 mm TSK-Gel GMH
H R
-H column (Tosoh Corporation) at 70 °C
250
using a Viscotek GPC Max VE 2001 separation module and a Viscotek TDA 305 RI
detector. The instrument was calibrated vs. polystyrene standards (1,050 – 3,800,000
g/mol) and data was analyzed using OmniSec 4.6.0 software. MALDI data was obtained
using an Applied Biosystems Voyager-DE STR mass spectrometer and 2,5-
dihydroxybenzoic acid as matrix.
For thin film measurements solutions were spin-coated onto pre-cleaned glass
slides from chlorobenzene solutions at 7 mg/ml for P3HT, 20 mg/ml for PC
61
BM and
ICBA and 10 mg/ml in P3HT for P3HT:PC
61
BM:ICBA blends. UV-vis absorption
spectra were obtained on a Perkin-Elmer Lambda 950 spectrophotometer. The thickness
of the thin films and GIXRD measurements were obtained using Rigaku Diffractometer
Ultima IV using Cu Kα radiation source (λ = 1.54 Å) in the reflectivity and grazing
incidence X-ray diffraction mode, respectively.
Transmission electron microscopy (TEM) was performed on the JEOL JEM-2100
microscope equipped with the Gatan Orius CCD camera. The accelerating voltage was
200 kV. Films for the TEM measurements were prepared from the chlorobenzene
solutions of P3HT:PC
61
BM:ICBA blends at the 1:1:0, 1:0.5:0.5 and 1:0:1 ratios and
optimized annealing conditions. Films for TEM were prepared by first spin-casting on
KBr plates, which were then placed in de-ionized water and upon salt dissolution the
floated P3HT:PC
61
BM:ICBA films were picked up with the 600 hex mesh copper grid
(Electron Microscopy Sciences).

251
A.2 Synthetic Procedures

Synthetic procedures for the synthesis of poly(3-hexylthiophene) (P3HT) were
used without modifications as reported in the literature.
1

Scheme A.1.

Indene-C
60
bisadduct (ICBA)

(1):
Modified from the literature.
2
200 mg (0.27 mmol) C
60
and 0.82 ml (7.02 mmol)
indene were combined with 20 ml o-DCB and brought to reflux for 22 hours. The
reaction mixture was cooled down and poured in methanol. The precipitate was filtered
and washed with methanol before being purified with flash chromatography (Hexanes:
Toluene, 7:1). 22 mg mono-adduct, 137 mg indene-C
60
bisadduct (ICBA) (53% yield)
and 124 mg tris-adduct was obtained. MALDI: 951.41 (calculated for C
78
H
16
: 952).

252
A.3 UV-vis and GIXRD

Figure A.1. UV-vis absorption spectra of thin films spin-coated from chlorobenzene
(CB) and annealed at 150 °C under N
2
for 20 min, where (i) is P3HT (black line), (ii) is
PC
61
BM (red line) and (iii) is ICBA (blue line).

Figure A.2. Grazing-incidence X-ray diffraction of thin films of P3HT:PC
61
BM:ICBA
spin-coated from chlorobenzene (CB) and annealed at 150 °C under N
2
for 20 min, where
(i) is 1:1:0 (red line), (ii) is 1:0.2:0.8 (blue line), (iii) is 1:0.5:0.5 (green line), (iv) is
1:0.8:0.2 (black line) and (v) is 1:0:1 (purple line).

253
A.4 Device Fabrication and Characterization

All steps of device fabrication and testing were performed in air. ITO-coated glass
substrates (10 Ω/□, Thin Film Devices Inc.) were sequentially cleaned by sonication in
detergent, de-ionized water, tetrachloroethylene, acetone, and isopropyl alcohol, and
dried in a nitrogen stream. A thin layer of PEDOT:PSS (Baytron
®
P VP AI 4083, filtered
with a 0.45 μm PVDF syringe filter – Pall Life Sciences) was first spin-coated on the pre-
cleaned ITO-coated glass substrates and baked at 130 ºC for 60 minutes under vacuum.
Separate solutions of 20 mg/ml P3HT, PC
61
BM and ICBA each were prepared in
chlorobenzene solvent. The solutions were stirred for 24 hrs before they were mixed at
the desired ratios and stirred for 24 hrs to form a homogeneous mixture. Subsequently,
the P3HT:PC
61
BM:ICBA active layer was spin-coated (with a 0.45 µm PTFE syringe
filter - Whatman) on top of the PEDOT:PSS layer. Upon spin-coating (900 rpm for 60
sec) of P3HT:PC
61
BM:ICBA, at all fullerene ratios (10 mg/ml in P3HT) films formed 95
– 105 nm thick layers and were directly placed in the vacuum chamber for aluminum
deposition. At the final stage, the substrates were pumped down to high vacuum (< 7×10
-
7
Torr) and aluminum (100 nm) was thermally evaporated at 3 – 4 Å/sec using a Denton
Benchtop Turbo IV Coating System onto the active layer through shadow masks to
define the active area of the devices as 4.9 mm
2
. Thermal annealing of
P3HT:PC
61
BM:ICBA blends was carried out by directly placing the completed devices in
the nitrogen oven at 150 °C for: 10 min for 1:0.3:0.7 and 1:0:1; 20 min for 1:0.9:0.1,
1:0.8:0.2, 1:0.2:0.8 and 1:0.1:0.9; 30 min for 1:0.6:0.4; 40 min 1:0.7:0.3 and 1:0.4:0.6; 50
254
min 1:0.5:0.5; and 60 min 1:1:0 ternary blend ratios. In case of optimized
P3HT:PC
61
BM:ICBA blends at 1:0.5:0.5 and 1:0:1 ratios thicker films of 137 nm (spin-
coated from 10 mg/ml in P3HT) (700 rpm for 60 sec) and 174 nm (spin-coated from 13
mg/ml in P3HT) were required (700 rpm for 60 sec). After aluminum deposition films
were annealed in the nitrogen oven at 150 °C for 10 min in both cases. After annealing,
the devices were cooled down to room temperature before measurements were carried
out.
The current-voltage (I-V) characteristics of the photovoltaic devices were
measured under ambient conditions using a Keithley 2400 source-measurement unit. An
Oriel
®
Sol3A class AAA solar simulator with Xenon lamp (450 Watt) and an AM 1.5G
filter was used as the solar simulator. An Oriel PV reference cell system 91150V was
used as the reference cell. To calibrate the light intensity of the solar simulator (to 100
mW/cm
2
), the power of the Xenon lamp was adjusted to make the short-circuit current
density (J
sc
) of the reference cell under simulated sun light as high as it was under the
calibration condition.
External quantum efficiency measurements were performed using a 300 W Xenon
arc lamp (Newport Oriel), chopped and filtered monochromatic light (250 Hz, 10 nm
FWHM) from a Conerstone 260 1/4 M double grating monochromator (Newport 74125)
together with an EG&G 7220 lock-in amplifier. A silicon photodiode (Hamamatsu
S1787-04, 8RA filter) calibrated at the National Renewable Energy Laboratory (NREL)
was utilized as the reference cell.

255
A.5 J-V and EQE Curves

Figure A.3. J-V curves of the ternary blend BHJ solar cells based on
P3HT:PC
61
BM:ICBA at different ratios: 1:1:0 (red line), 1:0.9:0.1 (navy line), 1:0.8:0.2
(blue line), 1:0.7:0.3 (olive line), 1:0.6:0.4 (cyan line), 1:0.5:0.5 (green line), 1:0.4:0.6
(pink line), 1:0.3:0.7 (wine red line), 1:0.2:0.8 (black line), 1:0.1:0.9 (dark yellow line)
and 1:0:1 (purple line) under AM 1.5G illumination (100 mW/cm
2
) at thicknesses 95 –
105 nm presented in Table 3.1.

256

Figure A.4. J-V curves of the optimized ternary blend BHJ solar cells based on
P3HT:PC
61
BM:ICBA at different ratios: 1:1:0 (red line), 1:0.5:0.5 (green line), 1:0:1
(purple line) under AM 1.5G illumination (100 mW/cm
2
) at thicknesses 104 nm, 137 nm
and 174 nm, respectively.

Figure A.5. External quantum efficiency of the ternary blend BHJ solar cells based on
P3HT:PC
61
BM:ICBA at different ratios: 1:1:0 (red circles), 1:0.5:0.5 (green circles),
1:0:1 (purple circles) at thicknesses 95 – 105 nm presented in Table 3.1.

257
A.6 TEM Images

Figure A.6. TEM images of P3HT:PC
61
BM:ICBA at (a) 1:1:0, (b) 1:0.5:0.5 and (c) 1:0:1
at thicknesses 95 – 105 nm for BHJ solar cells presented in Table 3.1 (scale bar is 50
nm).

A.7 Appendix Bibliography

(1) Burkhart, B.; Khlyabich, P. P.; Cakir Canak, T.; LaJoie, T. W.; Thompson, B. C.
Macromolecules 2011, 44, 1242.

(2) Laird, D. W.; Richter, H.; Vejins, V.; Scott, L. T.; Lada, T. A. Organic
photovoltaic devices comprising fullerenes and derivatives thereof and improved
methods of making fullerene derivatives. World Intellectual Property Organization
WO 2009/086210, July 9, 2009.

258
Appendix B

Semi-Random and Random Copolymers

B.1 Semi-Random Copolymers

B.1.1 Materials and Methods

All reagents from commercial sources were used without further purification,
unless otherwise noted. All reactions were performed under dry N
2
, unless otherwise
noted. All dry reactions were performed with glassware that was oven dried and then
flamed under high vacuum and backfilled with N
2
. Flash chromatography was performed
using a Teledyne CombiFlash R
f
instrument in combination with RediSep R
f
normal
phase disposable columns. Solvents were purchased from VWR and used without further
purification except for THF which was dried over sodium/benzophenone before being
distilled.
All compounds were characterized by
1
H NMR (400 MHz) and
13
C NMR (100
MHz) on a Mercury 400. Polymer
1
H NMRs (500 MHz) were obtained on a Varian
VNMRS-500. For polymer molecular weight determination, polymer samples were
dissolved in HPLC grade o-dichlorobenzene at a concentration of 1 mg/ml, briefly heated
and then allowed to turn to room temperature prior to filtering through a 0.2 μm PTFE
filter. SEC was performed using HPLC grade o-dichlorobenzene at a flow rate of 1
259
ml/min on one 300 x 7.8 mm TSK-Gel GMH
H R
-H column (Tosoh Corporation) at 70 ºC
using a Viscotek GPC Max VE 2001 separation module and a Viscotek TDA 305 RI
detector. The instrument was calibrated vs. polystyrene standards (1,050 – 3,800,000
g/mol) and data was analyzed using OmniSec 4.6.0 software.
Cyclic voltammetry was collected using an EG&G instruments Model 263A
potentiostat under the control of PowerSuite Software. A standard three electrode cell
based on a Pt wire working electrode, a silver wire pseudo reference electrode (calibrated
vs. Fc/Fc
+
which is taken as 5.1 eV vs. vacuum)
1
and a Pt wire counter electrode was
purged with nitrogen and maintained under nitrogen atmosphere during all
measurements. Acetonitrile was distilled over CaH
2
prior to use and tetrabutyl
ammonium hexafluorophosphate (0.1 M) was used as the supporting electrolyte. Polymer
films were made by repeatedly dipping the Pt wire in a 1% (w/w) polymer solution in
chloroform or o-dichlorobenzene and dried under nitrogen prior to measurement.
For thin film measurements polymers were spin coated onto pre-cleaned glass
slides from o-dichlorobenzene solutions (7 mg/ml). UV-vis absorption spectra were
obtained on a Perkin-Elmer Lambda 950 spectrophotometer. The thickness of the thin
films and GIXRD measurements were obtained using Rigaku Diffractometer Ultima IV
using Cu Kα radiation source (λ = 1.54 Å) in the reflectivity and grazing incidence X-
Ray diffraction mode, respectively.
Transmission electron microscopy (TEM) was performed on the JEOL JEM-2100
microscope equipped with the Gatan Orius CCD camera. The accelerating voltage was
200 kV. Films for the TEM measurements were prepared from the chlorobenzene or o-
260
dichlorobenzene solutions of polymer:PC
61
BM blend at the optimized ratios found for the
most efficient solar cells. Films for TEM were prepared by first spin-casting on KBr
plates, which were then placed in de-ionized water and upon salt dissolution the floated
polymer:PC
61
BM films were picked up with the 600 hex mesh copper grid (Electron
Microscopy Sciences).

B.1.2 UV-vis and GIXRD

Table B.1. Optical properties of P3HT, P3HTT-DPP-5%, P3HTT-DPP-10% and P3HTT-
DPP-15% in o-dichlorobenzene (o-DCB) and thin films, spin-coated from o-DCB

λ
max,abs
(nm)
solution
Absorptivity
(L/cm*g)
λ
max,abs

(nm)
film
Absorption coefficient
(cm
-1
)
E
g

(nm/eV)
c

P3HT 463 40.5 559 122068
a
648/1.91
P3HTT-DPP-5% 449 25.3 532 56503
b
814/1.52
670 17.8 682 43528
b

743 36547
b

P3HTT-DPP-10% 438 25.6 502 42107
b
821/1.51
671 33.7 685 71460
b

746 61924
b

P3HTT-DPP-15% 452 14.6 472 36400
b
848/1.46
696 40.5 703 97210
b

767 93677
b

a
Annealed at 150

ºC for 30 min under N
2
.
b
As-cast.
c
Calculated from the absorption band edge in thin films,
E
g
= 1240/λ
edge
.

261

Figure B.1. Grazing-incidence X-ray diffraction of thin films of (i) P3HT (spin-coated
from cholorobenzene and annealed at 150

°C for 30 min under N
2
) (black line), (ii)
P3HTT-DPP-5% (spin-coated from o-dichlorobenzene (o-DCB) and annealed at 150

°C
for 30 min under N
2
) (red line), (iii) P3HTT-DPP-10% (spin-coated from o-DCB and
tested as-cast) (green line), (iv) P3HTT-DPP-10% (spin-coated from o-DCB and
annealed at 150

°C for 30 min under N
2
) (purple line) and (v) P3HTT-DPP-15% (spin-
coated from o-DCB and annealed at 150

°C for 30 min under N
2
) (blue line).

262
B.1.3 CV Traces

Figure B.2. CV traces for the oxidation of P3HTT-DPP-5%, P3HTT-DPP-10% and
P3HTT-DPP-15%.

263
B.1.4 Device Fabrication and Characterization

All steps of device fabrication and testing were performed in air. ITO-coated glass
substrates (10 Ω/□, Thin Film Devices Inc.) were sequentially cleaned by sonication in
detergent, de-ionized water, tetrachloroethylene, acetone, and isopropyl alcohol, and
dried in a nitrogen stream. A thin layer of PEDOT:PSS (Baytron
®
P VP AI 4083, filtered
with a 0.45 μm PVDF syringe filter – Pall Life Sciences) was first spin-coated on the pre-
cleaned ITO-coated glass substrates and baked at 130 °C for 60 minutes under vacuum.
Separate solutions of polymers and PC
61
BM were prepared in cholorobenzene or o-
dichlorobenzene solvent. The solutions were stirred for 24 hrs before they were mixed at
the desired ratios and stirred for 24 hrs to form a homogeneous mixture. Subsequently,
the polymer:PC
61
BM active layer was spin-coated (with a 0.45 µm PTFE syringe filter -
Whatman ) on top of the PEDOT:PSS layer. The P3HT:PC
61
BM film was spin-coated
from chlorobenzene solution (10 mg/ml in P3HT), forming 95 nm thick layer and directly
placed in the vacuum chamber for aluminum deposition. The P3HTT-DPP-5%:PC
61
BM
and P3HTT-DPP-10%:PC
61
BM layers were spin cast from a solution in o-
dichlorobenzene containing 10 mg/ml P3HTT-DPP-5% or P3HTT-DPP-10% and 10
mg/ml PC
61
BM (P3HTT-DPP-5%) or 13.5 mg/ml PC
61
BM (P3HTT-DPP-10%). Films
were placed in a nitrogen cabinet for 20 min, before being transferred to the vacuum
chamber. Due to limited solubility, the solution of P3HTT-DPP-15%:PC
61
BM was
prepared by dissolving the polymer (8 mg/ml) and PC
61
BM (20.8 mg/ml) in a o-
dichlorobenzene solvent. As in case of P3HTT-DPP-5% and P3HTT-DPP-10% devices
264
were placed for 20 min in a nitrogen cabinet before loading in the vacuum chamber. At
the final stage, the substrates were pumped down to high vacuum (< 7×10
-7
Torr) and
aluminum (100 nm) was thermally evaporated at 3 – 4 Å/sec using a Denton Benchtop
Turbo IV Coating System onto the active layer through shadow masks to define the
active area of the devices as 4.9 mm
2
. Thermal annealing of P3HT:PC
61
BM blends was
carried out by directly placing the completed devices in the nitrogen oven for 60 min at
145 °C. After annealing, the devices were cooled down to room temperature before
measurements were carried out. All P3HTT-DPP:PC
61
BM BHJ solar cell was tested
without thermal treatment.
The current-voltage (I-V) characteristics of the photovoltaic devices were
measured under ambient conditions using a Keithley 2400 source-measurement unit. An
Oriel
®
Sol3A class AAA solar simulator with xenon lamp (450 Watt) and an AM 1.5G
filter was used as the solar simulator. An Oriel PV reference cell system 91150V was
used as the reference cell. To calibrate the light intensity of the solar simulator (to 100
mW/cm
2
), the power of the xenon lamp was adjusted to make the short-circuit current
(J
sc
) of the reference cell under simulated sun light as high as it was under the calibration
condition.
External quantum efficiency measurements were performed using a 300 W Xenon
arc lamp (Newport Oriel), chopped and filtered monochromatic light (250 Hz, 10 nm
FWHM) from a Conerstone 260 1/4 M double grating monochromator (Newport 74125)
together with an EG&G 7220 lock-in amplifier. A silicon photodiode (Hamamatsu
265
S1787-04, 8RA filter) calibrated at the National Renewable Energy Laboratory (NREL)
was utilized as the reference cell.
The spectral mismatch correction (spectral-mismatch factor (M)) was performed
according to Shrotriya et al.,
2
where mismatch factor is defined as:
22
11
22
11
() () () ()
() () () ()
Ref R S T
Ref T S R
E S dE S d
M
E S dE S d
λλ
λλ
λ λ
λ λ
λ λ λ λ λ λ
λ λ λ λ λ λ
=
∫∫
∫∫
(1),
where E
Ref
(λ) is the reference spectral irradiance; E
S
(λ) is the source spectral irradiance;
S
R
(λ) is the spectral responsivity; and S
T
(λ) is the spectral responsivity of the test cell,
each as a function of wavelength (λ). Spectral responsivities S(λ) for the tested devices
were calculated based on the external quantum efficiency (EQE) values, according to
equation 2:
() ()
q
S EQE
hc
λ
λ λ = (2),
where the constant term q/hc equals 8.0655 x 10
5
for wavelength in units of meters and
S(λ) in units of AW
-1
. Based on the spectral responsivities S(λ) obtained using equation
2, integrated short-circuit current densities (J
sc,EQE
) can be obtained:
2
1
,
() ()
sc EQE Ref T
J E Sd
λ
λ
λ λ λ =
∫
(3).
In order to mismatch-correct the efficiencies of the BHJ solar cells, short-circuit
current densities (J
sc
) were divided by the M, as defined in equation 4. The raw data (J
sc
),
266
spectral-mismatch factor (M) and the spectral mismatch-corrected short-circuit current
densities (J
sc,corr
) are summarized in Table B.2.
,
sc
sc corr
J
J
M
= (4).

Table B.2. Raw short-circuit current densities (J
sc
), spectral-mismatch factor (M),
spectral mismatch-corrected short-circuit current densities (J
sc,corr
) and integrated short-
circuit current densities (J
sc,EQE
)
Polymer:PC
61
BM
(ratio)
J
sc
(mA/cm
2
)
M
J
sc,corr
(mA/cm
2
)
J
sc,EQE
(mA/cm
2
)
(J
sc,corr
/ J
sc,EQE
) x 100
(%)
P3HT (1:1)
a
9.90 1.05 9.49 9.02 4.9
P3HTT-DPP-5% (1:1)
b
8.33 0.87 9.57 9.07 5.2
P3HTT-DPP-10% (1:1.3)
b
10.54 0.76 13.87 13.24 4.5
P3HTT-DPP-5% (1:2.6)
b
9.54 0.71 13.44 12.78 4.9
a
Spin-coated from chlorobenzene (CB) and annealed at 150

ºC for 30 min under N
2
after aluminum
deposition.
b
Spin-coated from o-DCB and tested after 20 minutes of slow solvent evaporation in a N
2

cabinet.

267
B.1.5 J-V Curves

Figure B.3. J-V curves of the BHJ solar cells based on P3HT (black line), P3HTT-DPP-
5% (red line), P3HTT-DPP-10% (green line) and P3HTT-DPP-15% (blue line) with
PC
61
BM as the acceptor under AM 1.5G illumination (100 mW/cm
2
) at the optimal
conditions for solar cell performance.

268
B.1.6 TEM Images

Figure B.4. TEM images of (a) P3HT:PC
61
BM, (b) P3HTT-DPP-5%:PC
61
BM, (c)
P3HTT-DPP-10%:PC
61
BM and (d) P3HTT-DPP-15%:PC
61
BM prepared under optimal
solar cells conditions.

B.1.7 Mobility Measurements

Polymer mobility was measured using a hole-only device configuration of
ITO/PEDOT:PSS/Polymer/Al in the space charge limited current regime. The devices
preparations were the same as described above for solar cells. The dark current was
measured under ambient conditions. At sufficient potential the mobilities of charges in
the device can be determined by fitting the dark current to the model of SCL current and
described by equation 5:
269
2
0 3
9
8
SCLC R
V
J
L
ε εµ = (5),
where J
SCLC
is the current density, ε
0
is the permittivity of space, ε
R
is the dielectric
constant of the polymer (assumed to be 3), μ is the zero-field mobility of the majority
charge carriers, V is the effective voltage across the device (V = V
applied
– V
bi
– V
r
), and L
is the polymer layer thickness. The series and contact resistance of the device (14 – 20 Ω)
was measured using a blank (ITO/PEDOT/Al) configuration and the voltage drop due to
this resistance (V
r
) was subtracted from the applied voltage. The built-in voltage (V
bi
),
which is based on the relative work function difference of the two electrodes, was also
subtracted from the applied voltage. The built-in voltage can be determined from the
transition between the ohmic region and the SCL region and is found to be about 1 V.
Polymer film thicknesses were measured using GIXRD in the reflectivity mode.

B.2 Random Copolymers

B.2.1 Materials and Methods

The synthetic procedure for 2-bromo-5-trimethyltin-3-hexylthiophene was
previously published.
3
3-(2-Ethylhexyl)thiophene (1) was synthesized according to the
literature.
4
All reagents from commercial sources were used without further purification,
unless otherwise noted. N-bromosuccinimide was recrystallized from water prior to use.
All reactions were performed under dry N
2
, unless otherwise noted. All reactions were
performed with glassware that was oven dried and then flamed under high vacuum and
270
backfilled with N
2
. Flash chromatography was performed using a Teledyne CombiFlash
R
f
instrument in combination with RediSep R
f
normal phase disposable columns.
Solvents were purchased from VWR and used without further purification except for
THF which was dried over sodium/benzophenone before being distilled.
All compounds were characterized by
1
H NMR (400 MHz) and
13
C NMR (100
MHz) on a Mercury 400. Polymer
1
H NMRs (500 MHz) were obtained on a Varian
VNMRS-500. For polymer molecular weight determination, polymer samples were
dissolved in HPLC grade o-dichlorobenzene at a concentration of 1 mg/ml, briefly heated
and then allowed to turn to room temperature prior to filtering through a 0.2 μm PTFE
filter. SEC was performed using HPLC grade o-dichlorobenzene at a flow rate of 1
ml/min on one 300 × 7.8 mm TSK-Gel GMH
H R
-H column (Tosoh Corporation) at 70
°
C
using a Viscotek GPC Max VE 2001 separation module and a Viscotek TDA 305 RI
detector. The instrument was calibrated vs. polystyrene standards (1,050 – 3,800,000
g/mol) and data was analyzed using OmniSec 4.6.0 software.
Cyclic voltammetry was collected using an Princeton Applied Research
VersaStat3 potentiostat under the control of VersaStudio Software. A standard three
electrode cell based on a Pt wire working electrode, a silver wire pseudo reference
electrode (calibrated vs. Fc/Fc
+
which is taken as 5.1 eV vs. vacuum)
1,5
and a Pt wire
counter electrode was purged with nitrogen and maintained under nitrogen atmosphere
during all measurements. For cyclic voltammetry of thin films acetonitrile was distilled
over CaH
2
prior to use and tetrabutyl ammonium hexafluorophosphate (0.1 M) was used
as the supporting electrolyte. Polymer films were made by repeatedly dipping the Pt wire
271
in a 1% (w/w) chloroform or o-dichlorobenzene solution and dried under nitrogen prior
to measurement. For cyclic voltammetry of solutions chloroform was distilled over CaH
2

prior to use and tetrabutyl ammonium tetrafluoroborate (0.1 M) was used as the
supporting electrolyte.
UV-vis absorption spectra were obtained on a Perkin-Elmer Lambda 950
spectrophotometer. Thickness of the samples and GIXRD measurements were obtained
using Rigaku Diffractometer Ultima IV using a Cu Kα radiation source (λ = 1.54 Å) in
the reflectivity and grazing incidence X-ray diffraction mode, respectively. For thin film
measurements polymers were spin coated onto pre-cleaned glass slides from
chlorobenzene solutions (7 mg/ml).
DSC traces were obtained using a Perkin Elmer DSC 8000 with a scan rate of 10
ºC/min. Sample size was ~ 3 mg and polymers were used as obtained after purification.
Transmission electron microscopy (TEM) was performed on the JEOL JEM-2100
microscope equipped with the Gatan Orius CCD camera. The accelerating voltage was
200 kV. Films for the TEM measurements were prepared from the chlorobenzene
solutions of polymer:PC
61
BM blend at the optimized ratios and annealing temperatures
found for the most efficient solar cells. Films for TEM were prepared by first spin-casting
on KBr plates, which were then placed in de-ionized water and upon salt dissolution the
floated polymer:PC
61
BM films were picked up with the 600 hex mesh copper grid
(Electron Microscopy Sciences).

272
B.2.2 UV-vis

Figure B.5. UV-vis absorption of all five polymers in as cast thin films (spin coated from
CB) where P3HT is purple line, P3HT
90
-co-EHT
10
is green line, P3HT
75
-co-EHT
25
is
blue line, P3HT
50
-co-EHT
50
is red line and P3EHT is orange line.

273
B.2.3 CV Traces

Figure B.6. CV traces for the oxidation of thin films (as cast) where a) is P3HT, b) is
P3HT
90
-co-EHT
10
, c) is P3HT
75
-co-EHT
25
, d) is P3HT
50
-co-EHT
50
and e) is P3EHT.
Ferrocene was used as a reference and values were converted to the vacuum scale using
the approximation that the ferrocene redox couple is 5.1 eV relative to vacuum.
274

Figure B.7. CV traces for the oxidation of polymers in solution (CHCl
3
with
tetrabutylammonium tetrafluoroborate as supporting electrolyte) where a) is P3HT, b) is
P3HT
90
-co-EHT
10
, c) is P3HT
75
-co-EHT
25
, d) is P3HT
50
-co-EHT
50
and e) is P3EHT.
Ferrocene was used as a reference and values were converted to the vacuum scale using
the approximation that the ferrocene redox couple is 5.1 eV relative to vacuum.

275
B.2.4 DSC Curves

Figure B.8. DSC curves where a) is P3HT, b) is P3HT
90
-co-EHT
10
, c) is P3HT
75
-co-
EHT
25
and d) is P3HT
50
-co-EHT
50
.

B.2.5 Device Fabrication and Characterization

All steps of device fabrication and testing were performed in air. ITO-coated glass
substrates (10 Ω/□, Thin Film Devices Inc.) were sequentially cleaned by sonication in
276
detergent, de-ionized water, tetrachloroethylene, acetone, and isopropyl alcohol, and
dried in a nitrogen stream. A thin layer of PEDOT:PSS (Baytron
®
P VP AI 4083, filtered
with a 0.45 μm PVDF syringe filter – Pall Life Sciences) was first spin-coated on the pre-
cleaned ITO-coated glass substrates and baked at 130 °C for 60 minutes under vacuum.
Separate solutions of polymers and PC
61
BM were prepared in cholorobenzene solvent.
The solutions were stirred for 24 hrs before they were mixed at the desired ratios and
stirred for further 24 hrs to form a homogeneous mixture. Subsequently, the
polymer:PC
61
BM active layer was spin-coated (with a 0.45 μm PTFE syringe filter - Pall
Life Sciences) on top of the PEDOT:PSS layer. All polymer:PC
61
BM films were spin-
coated from chlorobenzene solution (10 mg/ml in polymer) and directly placed in the
vacuum chamber for aluminum deposition. At the final stage, the substrates were pumped
down to high vacuum (< 2×10
-6
Torr) and aluminum (100 nm) was thermally evaporated
at 3 – 4 Å/sec using a Denton Benchtop Turbo IV Coating System onto the active layer
through shadow masks to define the active area of the devices as 4.4 mm
2
. Thermal
annealing of P3HT:PC
61
BM blends was carried out by directly placing the completed
devices in the nitrogen oven for 60 min at 145 °C, P3HT
90
-co-EHT
10
:PC
61
BM blends for
60 min at 110 °C, P3HT
75
-co-EHT
25
:PC
61
BM blends for 30 min at 110 °C, P3HT
50
-co-
EHT
50
:PC
61
BM blends for 10 min at 110 °C. After annealing, the devices were cooled
down to room temperature before measurements were carried out. P3EHT:PC
61
BM BHJ
solar cells were tested without thermal treatment.
The current-voltage (I-V) characteristics of the photovoltaic devices were
measured under ambient conditions using a Keithley 2400 source-measurement unit. An
277
Oriel
®
Sol3A class AAA solar simulator with Xenon lamp (450 Watt) and an AM 1.5G
filter was used as the solar simulator. An Oriel PV reference cell system 91150V was
used as the reference cell. To calibrate the light intensity of the solar simulator (to 100
mW/cm
2
), the power of the Xenon lamp was adjusted to make the short-circuit current
density (J
sc
) of the reference cell under simulated sun light as high as it was under the
calibration condition.
Polymer mobility was measured using a hole-only device configuration of
ITO/PEDOT:PSS/Polymer/Al in the space charge limited current regime as described
previously.
3,6

278
B.2.6 TEM Images

Figure B.9. TEM images of polymer:PCBM films (optimized conditions for best solar
cell performance) where a) is P3HT, b) is P3HT
90
-co-EHT
10
, c) is P3HT
75
-co-EHT
25
, d)
is P3HT
50
-co-EHT
50
and e) is P3EHT.

B.3 Appendix Bibliography

(1) Thompson, B. C.; Kim, Y.-G.; McCarley, T. D.; Reynolds, J. R. J. Am. Chem. Soc.
2006, 128, 12714.

(2) Shrotriya, V.; Li, G.; Yao, Y.; Moriarty, T.; Emery, K.; Yang, Y. Adv. Funct.
Mater. 2006, 16, 2016.
279
(3) Burkhart, B.; Khlyabich, P. P.; Cakir Canak, T.; LaJoie, T. W.; Thompson, B. C.
Macromolecules 2011, 44, 1242.

(4) Zhang, Y.; Tajima, K.; Hirota, K.; Hashimoto, K. J. Am. Chem. Soc. 2008, 130,
7812.

(5) Cardona, C. M.; Li, W.; Kaifer, A. E.; Stockdale, D.; Bazan, G. C. Adv. Mater.
2011, 23, 2367.

(6) Khlyabich, P. P.; Burkhart, B.; Ng, C. F.; Thompson, B. C. Macromolecules 2011,
44, 5079.

280
Appendix C

Ternary Blend Bulk Heterojunction Solar Cells in the Case of Complementary
Donors

C.1 Materials and Methods

All reagents from commercial sources were used without further purification,
unless otherwise noted. Solvents were purchased from VWR and used without further
purification except for THF which was dried over sodium/benzophenone before being
distilled.
For thin film measurements solutions were spin-coated onto pre-cleaned glass
slides from o-dichlorobenzene solutions at 7 mg/ml for P3HTT-DPP-10% and P3HT
75
-
co-EHT
25
, and at optimal ratios found for P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM
solar cells. UV-vis absorption spectra were obtained on a Perkin-Elmer Lambda 950
spectrophotometer. The thickness of the thin films and GIXRD measurements were
obtained using Rigaku Diffractometer Ultima IV using Cu Kα radiation source (λ = 1.54
Å) in the reflectivity and grazing-incidence X-ray diffraction mode, respectively.
Transmission electron microscopy (TEM) was performed on the JEOL JEM-2100
microscope equipped with the Gatan Orius CCD camera. The accelerating voltage was
200 kV. Films for the TEM measurements were prepared from the o-dichlorobenzene
solutions of P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM blends at the 1:0:1.3,
281
0.9:0.1:1.1, 0.8:0.2:1.0, 0.5:0.5:0.9, 0.2:0.8:0.8 and 0:1:0.8 ratios and optimized
processing conditions. Films for TEM were prepared by first spin-casting on KBr plates,
which were then placed in de-ionized water and upon salt dissolution the floated P3HTT-
DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM films were picked up with the 600 hex mesh
copper grid (Electron Microscopy Sciences).

C.2 Synthetic Procedures

Synthetic procedures for the synthesis of poly(3-hexylthiophene-co-3-(2-
ethylhexyl)thiophene) (P3HT
75
-co-EHT
25
) and poly(3-hexylthiophene-thiophene-
diketopyrrolopyrrole) (P3HTT-DPP-10%) were used without modifications as reported in
the literature.
1,2

282
C.3 UV-vis and GIXRD

Figure C.1. UV-vis absorption spectra of thin films spin-coated from o-dichlorobenzene
(o-DCB) and placed to N
2
cabinet for 30 min with P3HTT-DPP-10%:P3HT
75
-co-
EHT
25
:PC
61
BM ratios, where (i) is 1:0:1.3 (red line), (ii) is 0.9:0.1:1.1 (green line), (iii) is
0.8:0.2:1.0 (blue line), (iv) is 0.7:0.3:1.0 (cyan line), (v) is 0.5:0.5:0.9 (wine-red line), (vi)
is 0.3:0.7:0.8 (dark yellow line), (vii) is 0.1:0.9:0.9 (yellow line) and (viii) is 0:1:0.8
(black line).

283
Table C.1. Optical properties of neat P3HTT-DPP-10%, P3HT
75
-co-EHT
25
and P3HTT-
DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM blends at different ratios in thin films spin-coated
from o-DCB
P3HTT-DPP-10%:
P3HT
75
-co-EHT
25
:
PC
61
BM
λ
max,abs
(nm)
film
Absorption coefficient (cm
-1
)
E
g
(nm/eV)
a

P3HTT-DPP-10% 501 52205 835/1.49
687 77265
1:0:1.3 463 30936 8635/1.49
685 38109
0.9:0.1:1.1 473 31478 835/1.49
685 37770
0.8:0.2:1.0 482 31887 834/1.49
687 36096
0.7:0.3:1.0 488 32155 834/1.49
686 30630
0.5:0.5:0.9 512 34856 833/1.49
688 23925
0.3:0.7:0.8 512 41577 832/1.49
686 18009
0.1:0.9:0.9 516 48718 828/1.50
0:1:0.8 519 42577 655/1.89
P3HT
75
-co-EHT
25
555 85789 656/1.89
a
Calculated from the absorption band edge in thin films, E
g
= 1240/λ
edge
.

Figure C.2. Grazing-incidence X-ray diffraction of thin films of P3HTT-DPP-10% and
P3HT
75
-co-EHT
25
spin-coated from o-dichlorobenzene (o-DCB) and placed to N
2
cabinet
for 20 min, where (i) is P3HT
75
-co-EHT
25
(red line) and (ii) is P3HTT-DPP-10% (green
line).

284
Table C.2. 2θ, interchain distances (100) and GIXRD intensities of neat P3HTT-DPP-
10%, P3HT
75
-co-EHT
25
and P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM blends at
different ratios in thin films spin-coated from o-DCB
P3HTT-DPP-10%:
P3HT
75
-co-EHT
25
:
PC
61
BM
2θ (deg) d (Å) Intensity (a.u.) 2θ (deg) d (Å) Intensity (a.u.)
P3HTT-DPP-10% - - - 5.858 15.07 26.9
1:0:1.3 - - - 5.783 15.27 37.5
0.9:0.1:1.1 - - - 5.748 15.36 38.3
0.8:0.2:1.0 5.513 16.02 38.1 5.732 15.41 39.2
0.7:0.3:1.0 5.500 16.06 43.3 - - -
0.5:0.5:0.9 5.427 16.27 69.0 - - -
0.3:0.7:0.8 5.393 16.37 92.2 - - -
0.1:0.9:0.9 5.298 16.67 97.3 - - -
0:1:0.8 5.290 16.69 116.8 - - -
P3HT
75
-co-EHT
25
5.281 16.72 80.5 - - -

C.4 Device Fabrication and Characterization

All steps of device fabrication and testing were performed in air. ITO-coated glass
substrates (10 Ω/□, Thin Film Devices Inc.) were sequentially cleaned by sonication in
detergent, de-ionized water, tetrachloroethylene, acetone, and isopropyl alcohol, and
dried in a nitrogen stream. A thin layer of PEDOT:PSS (Baytron
®
P VP AI 4083, filtered
with a 0.45 μm PVDF syringe filter – Pall Life Sciences) was first spin-coated on the pre-
cleaned ITO-coated glass substrates and baked at 130 °C for 60 minutes under vacuum.
Separate solutions of P3HTT-DPP-10%, P3HT
75
-co-EHT
25
and PC
61
BM were prepared
in o-dichlorobenzene solvent. The solutions were stirred for 24 hrs before they were
mixed at the desired ratios and stirred for 24 hrs to form a homogeneous mixture.
Subsequently, the P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM active layer was spin-
coated (with a 0.45 µm PTFE syringe filter - Pall Life Sciences) on top of the
PEDOT:PSS layer. Concentrations of the binary blends were 10 mg/ml in polymer to
285
give ~ 75 nm thick films. Concentrations were of 11 mg/ml in total P3HTT-DPP-
10%:P3HT
75
-co-EHT
25
for ternary blends to give ~ 85 – 90 nm thick films. Upon spin-
coating of P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM, films were first placed to the
N
2
cabinet for: 1:0:1.3, 0.5:0.5:0.9, 0.4:0.6:0.9, 0.3:0.7:0.8, 0.2:0.8:0.8 and 0.1:0.9:0.9 for
30 min; 0.9:0.1:1.1 and 0.8:0.2:1.0 for 60 min; 0.7:0.3:1.0 and 0.6:0.4:1.0 for 45 min;
0:1:0.8 for 20 min and then placed in the vacuum chamber for aluminum deposition. At
the final stage, the substrates were pumped down to high vacuum (< 9×10
-7
Torr) and
aluminum (100 nm) was thermally evaporated at 3 – 4 Å/sec using a Denton Benchtop
Turbo IV Coating System onto the active layer through shadow masks to define the
active area of the devices as 4.4 mm
2
.
The current-voltage (I-V) characteristics of the photovoltaic devices were
measured under ambient conditions using a Keithley 2400 source-measurement unit. An
Oriel
®
Sol3A class AAA solar simulator with Xenon lamp (450 Watt) and an AM 1.5G
filter was used as the solar simulator. An Oriel PV reference cell system 91150V was
used as the reference cell. To calibrate the light intensity of the solar simulator (to 100
mW/cm
2
), the power of the Xenon lamp was adjusted to make the short-circuit current
density (J
sc
) of the reference cell under simulated sun light as high as it was under the
calibration condition.
External quantum efficiency measurements were performed using a 300 W Xenon
arc lamp (Newport Oriel), chopped and filtered monochromatic light (250 Hz, 10 nm
FWHM) from a Conerstone 260 1/4 M double grating monochromator (Newport 74125)
286
together with a light-bias lock-in amplifier. A silicon photodiode calibrated at Newport
was utilized as the reference cell.
The spectral mismatch correction (spectral-mismatch factor (M))
3
was performed
according to Shrotriya et al.,
4
where mismatch factor is defined as:
22
11
22
11
() () () ()
() () () ()
Ref R S T
Ref T S R
E S dE S d
M
E S dE S d
λλ
λλ
λ λ
λ λ
λ λ λ λ λ λ
λ λ λ λ λ λ
=
∫∫
∫∫
(1),
where E
Ref
(λ) is the reference spectral irradiance; E
S
(λ) is the source spectral irradiance;
S
R
(λ) is the spectral responsivity; and S
T
(λ) is the spectral responsivity of the test cell,
each as a function of wavelength (λ). Spectral responsivities S(λ) for the tested devices
were calculated based on the external quantum efficiency (EQE) values, according to
equation 2:
() ()
q
S EQE
hc
λ
λ λ = (2),
where the constant term q/hc equals 8.0655 x 10
5
for wavelength in units of meters and
S(λ) in units of AW
-1
. Based on the spectral responsivities S(λ) obtained using equation
2, integrated short-circuit current densities (J
sc,EQE
) can be obtained:
2
1
,
() ()
sc EQE Ref T
J E Sd
λ
λ
λ λ λ =
∫
(3).
In order to mismatch-correct the efficiencies of the BHJ solar cells, short-circuit
current densities (J
sc
) were divided by the M, as defined in equation 4. The raw data (J
sc
),
287
spectral-mismatch factor (M) and the spectral mismatch-corrected short-circuit current
densities (J
sc,corr
) are summarized in Table C.4.
,
sc
sc corr
J
J
M
= (4).

C.5 J-V Curves

Figure C.3. J-V curves of the ternary blend BHJ solar cells based on P3HTT-DPP-
10%:P3HT
75
-co-EHT
25
:PC
61
BM at different ratios: (i) is 1:0:1.3 (red line), (ii) is
0.9:0.1:1.1 (green line), (iii) is 0.8:0.2:1.0 (blue line), (iv) is 0.7:0.3:1.0 (cyan line), (v) is
0.6:0.4:1.0 (magenta line), (vi) is 0.5:0.5:0.9 (wine-red line), (vii) is 0.4:0.6:0.9 (olive
line), (viii) is 0.3:0.7:0.8 (dark yellow line), (ix) is 0.2:0.8:0.8 (purple line), (x) is
0.1:0.9:0.9 (yellow line) and (xi) is 0:1:0.8 (black line) under AM 1.5G illumination (100
mW/cm
2
) presented in Table 5.1.

288
C.6 TEM Images

Figure C.4. TEM images of P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM at (a) 1:0:1.3,
(b) 0.9:0.1:1.1, (c) 0.8:0.2:1.0, (d) 0.5:0.5:0.9, (e) 0.2:0.8:0.8 and (f) 0:1:0.8 for BHJ solar
cells presented in Table 5.1 (scale bar is 50 nm).

289
C.7 Mobility Measurements

Polymer mobility was measured using a hole-only device configuration of
ITO/PEDOT:PSS/ P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM /Al in the space charge
limited current regime. The devices preparations were the same as described above for
solar cells. The dark current was measured under ambient conditions. At sufficient
potential the mobilities of charges in the device can be determined by fitting the dark
current to the model of SCL current and described by equation 5:
2
0 3
9
8
SCLC R
V
J
L
ε εµ = (5),
where J
SCLC
is the current density, ε
0
is the permittivity of space, ε
R
is the dielectric
constant of the polymer (assumed to be 3), μ is the zero-field mobility of the majority
charge carriers, V is the effective voltage across the device (V = V
applied
– V
bi
– V
r
), and L
is the polymer layer thickness. The series and contact resistance of the device (16 – 20 Ω)
was measured using a blank (ITO/PEDOT/Al) configuration and the voltage drop due to
this resistance (V
r
) was subtracted from the applied voltage. The built-in voltage (V
bi
),
which is based on the relative work function difference of the two electrodes, was also
subtracted from the applied voltage. The built-in voltage can be determined from the
transition between the ohmic region and the SCL region and is found to be about 0.6 V.
Polymer film thicknesses were measured using GIXRD in the reflectivity mode.

290
Table C.3. Hole mobilities of neat P3HTT-DPP-10%, P3HT
75
-co-EHT
25
and P3HTT-
DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM blends at different ratios in thin films spin-coated
from o-DCB
P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM
Mobility (cm
2
/(V*s))
P3HTT-DPP-10%
2.30*10
-4

1:0:1.3
3.23*10
-3

0.9:0.1:1.1
2.38*10
-3

0.8:0.2:1.0
2.37*10
-3

0.7:0.3:1.0
2.19*10
-3

0.5:0.5:0.9
2.13*10
-3

0.3:0.7:0.8
2.01*10
-3

0.1:0.9:0.9
1.92*10
-3

0:1:0.8
1.11*10
-3

P3HT
75
-co-EHT
25

1.39*10
-4

C.8 Raw J-V Data

Table C.4. Raw short-circuit current densities (J
sc
), spectral-mismatch factor (M),
spectral mismatch-corrected short-circuit current densities (J
sc,corr
) and integrated short-
circuit current densities (J
sc,EQE
) for Table 5.1
P3HTT-DPP-10%:
P3HT
75
-co-EHT
25
:
PC
61
BM
J
sc

(mA/cm
2
)
M
J
sc,corr
(mA/cm
2
)
J
sc,EQE
(mA/cm
2
)
(J
sc,corr
/ J
sc,EQE
) ×
100 (%)
1:0:1.3
a
11.36 0.79 14.38 13.81 3.97
0.9:0.1:1.1
b
11.93 0.80 15.05 14.55 3.30
0.8:0.2:1.0
b
11.55 0.80 14.60 14.06 3.72
0.7:0.3:1.0
c
9.50 0.82 11.54 11.03 4.41
0.6:0.4:1.0
c
9.63 0.86 11.19 10.74 4.00
0.5:0.5:0.9
a
9.50 0.87 10.89 10.45 4.00
0.4:0.6:0.9
a
9.28 0.91 10.19 9.76 4.24
0.3:0.7:0.8
a
9.19 0.94 9.77 9.56 2.16
0.2:0.8:0.8
a
8.32 0.97 8.57 8.21 4.25
0.1:0.9:0.9
a
8.25 1.00 8.25 7.91 4.12
0:1:0.8
d
8.27 1.05 7.96 7.61 4.36
All devices were spin-coated from o-dichlorobenzene (o-DCB) and placed to the N
2
cabinet before
aluminum deposition for
a
30 min,
b
60 min,
c
45 min and
d
20 min.

291
Table C.5. Raw short-circuit current densities (J
sc
) for Table 5.1 without spectral-
mismatch factor correction, open-circuit voltage (V
oc
), fill factor (FF) and efficiency (η)
of P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM ternary blend BHJ solar cells under
optimized conditions
P3HTT-DPP-10%:
P3HT
75
-co-EHT
25
:
PC
61
BM
J
sc
(mA/cm
2
) V
oc
(V) FF η (%)
1:0:1.3
a
11.36 0.574 0.62 4.01
0.9:0.1:1.1
b
11.93 0.603 0.61 4.37
0.8:0.2:1.0
b
11.55 0.608 0.61 4.24
0.7:0.3:1.0
c
9.50 0.614 0.59 3.41
0.6:0.4:1.0
c
9.63 0.619 0.59 3.54
0.5:0.5:0.9
a
9.50 0.622 0.59 3.47
0.4:0.6:0.9
a
9.28 0.626 0.59 3.40
0.3:0.7:0.8
a
9.19 0.633 0.59 3.42
0.2:0.8:0.8
a
8.32 0.639 0.60 3.18
0.1:0.9:0.9
a
8.25 0.646 0.59 3.10
0:1:0.8
d
8.27 0.675 0.59 3.29
All devices were spin-coated from o-dichlorobenzene (o-DCB) and placed to the N
2
cabinet before
aluminum deposition for
a
30 min,
b
60 min,
c
45 min and
d
20 min.

Table C.6. Raw short-circuit current densities (J
sc
), open-circuit voltage (V
oc
), fill factor
(FF) and efficiency (η) of P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM ternary blend
BHJ solar cells at constant overall polymer:fullerene ratio of 1:1.1 in ternary blend
composition regime
P3HTT-DPP-10%:
P3HT
75
-co-EHT
25
:
PC
61
BM
J
sc
(mA/cm
2
) V
oc
(V) FF η (%)
1:0:1.3
a
11.36 0.574 0.62 4.01
0.9:0.1:1.1
b
11.93 0.603 0.61 4.37
0.8:0.2:1.1
b
10.06 0.591 0.57 3.39
0.7:0.3:1.1
c
9.45 0.597 0.57 3.21
0.6:0.4:1.1
c
9.08 0.602 0.56 3.06
0.5:0.5:1.1
a
8.71 0.607 0.52 2.75
0.4:0.6:1.1
a
7.76 0.617 0.52 2.49
0.3:0.7:1.1
a
7.50 0.619 0.49 2.27
0.2:0.8:1.1
a
8.35 0.625 0.47 2.45
0.1:0.9:1.1
a
6.02 0.634 0.51 1.95
0:1:0.8
d
8.27 0.675 0.59 3.29
All devices were spin-coated from o-dichlorobenzene (o-DCB) and placed to the N
2
cabinet before
aluminum deposition for
a
30 min,
b
60 min,
c
45 min and
d
20 min.

292
Table C.7. Raw short-circuit current densities (J
sc
), open-circuit voltage (V
oc
), fill factor
(FF) and efficiency (η) of P3HTT-DPP-10%:P3HT
75
-co-EHT
25
:PC
61
BM ternary blend
BHJ solar cells at constant overall polymer:fullerene ratio of 1:1.0 in ternary blend
composition regime
P3HTT-DPP-10%:
P3HT
75
-co-EHT
25
:
PC
61
BM
J
sc
(mA/cm
2
) V
oc
(V) FF η (%)
1:0:1.3
a
11.36 0.574 0.62 4.01
0.9:0.1:1.0
b
9.89 0.601 0.62 3.66
0.8:0.2:1.0
b
11.55 0.608 0.61 4.24
0.7:0.3:1.0
c
9.50 0.614 0.59 3.41
0.6:0.4:1.0
c
9.63 0.619 0.59 3.54
0.5:0.5:1.0
a
9.06 0.622 0.54 3.04
0.4:0.6:1.0
a
8.79 0.626 0.54 2.94
0.3:0.7:1.0
a
8.34 0.626 0.50 2.61
0.2:0.8:1.0
a
8.58 0.629 0.51 2.72
0.1:0.9:1.0
a
6.98 0.635 0.53 2.75
0:1:0.8
d
8.27 0.675 0.59 3.29
All devices were spin-coated from o-dichlorobenzene (o-DCB) and placed to the N
2
cabinet before
aluminum deposition for
a
30 min,
b
60 min,
c
45 min and
d
20 min.

C.9 Appendix Bibliography

(1) Khlyabich, P. P.; Burkhart, B.; Thompson, B. C. J. Am. Chem. Soc. 2011, 133,
14534.

(2) Burkhart, B.; Khlyabich, P. P.; Thompson, B. C. Macromolecules 2012, 45, 3740.

(3) Snaith, H. J. Energy Environ. Sci. 2012, 5, 6513.

(4) Shrotriya, V.; Li, G.; Yao, Y.; Moriarty, T.; Emery, K.; Yang, Y. Adv. Funct.
Mater. 2006, 16, 2016.

Abstract (if available)

Abstract The growing energy demands facilitate the search of cheap and renewable energy. The vision of organic photovoltaics is that of a low cost solar energy conversion platform that provides lightweight, flexible solar cells that are easily incorporated into existing infrastructure with minimal impact on land usage. Organic solar cells have been a subject of growing research interest over the past quarter century, and are now developed to the point where they are on the verge of introduction into the market. Polymer solar cells provide all important characteristics necessary to fill this niche and power conversion efficiency exceeding 10% was recently achieved. However, such a high efficiencies are obtained using multi-junction solar cells which require additional steps in the manufacturing process and thus contradicts the attractive simplicity of the single step solution processing of the active layer in bulk heterojunction solar cells. A basic introduction to polymer:fullerene solar cells is presented in Chapter 1. ❧ The current dissertation is focused on the recently emerged approach of ternary blend bulk heterojunction solar cells, where three components are mixed in the active layer, providing an alternative route to high efficiencies while preserving the simplicity of the solar cell fabrication. A general overview of this class of solar cells is presented in Chapter 2. ❧ The key component to the efficiency increase in ternary blend bulk heterojunction solar cells is the discovered open-circuit voltage (Voc) tunability in the three component systems, which is introduced in Chapter 3. Together with the short-circuit current density (Jsc) increase obtained using polymers with complementary absorption profiles (discussed in Chapter 4), the efficiencies of the ternary blend bulk heterojunction solar cells are increased beyond that of the corresponding binary blend solar cells, as discussed in Chapter 5. By extension, this result suggests that ternary blends provide a potentially effective route toward maximizing the attainable Jsc × Voc product (which is directly proportional to the solar cell efficiency) in bulk heterojunction solar cells and that with judicious selection of donor and acceptor components, solar cells with efficiencies exceeding the theoretical limits for binary blend solar cells could be possible without sacrificing the simplicity of a single active layer processing step. ❧ Finally, the formation of an organic alloy in ternary blend bulk heterojunction solar cells, introduced in Chapter 6, unravels and explains the origin of the Voc tunability and simultaneous Jsc increase in three component systems.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Molecular and polymeric donors for bulk heterojunction organic solar cells

PDF

Improving the efficiency and stability of organic solar cells through ternary strategies

PDF

Synthesis and characterization of 3-hexylesterthiophene based random and semi-random polymers and their use in ternary blend solar cells

PDF

Development of a family of semi-random multichromophoric polymers for application in organic solar cells

PDF

Influence of polymer structure on electronic properties and performance as donor materials in bulk heterojunction solar cells

PDF

Structure-function studies of random and semi-random poly(3-hexyl thiophene) copolymers

PDF

From perfectly alternating to random multichromophoric conjugated polymers for ternary blend bulk heterojunction solar cells

PDF

Direct C−H arylation for the synthesis of conjugated polymers

PDF

Improving the sustainability of conjugated polymer synthesis via direct arylation polymerization

PDF

Excited state dynamics of organic photovoltaics probed using ultrafast spectroscopy

PDF

Molecular and morphological effects on the operational parameters of organic solar cells

PDF

Direct arylation polymerization (DARP) for the synthesis of conjugated polymers

PDF

Organic solar cells: molecular electronic processes and device development

PDF

Organic photovoltaics: from materials development to device application

PDF

Molecular aspects of photoconversion processes in organic solar cells

PDF

Non-traditional architectures for semiconducting polymers

PDF

Growth, characterization of gallium arsenide based nanowires and application in photovoltaic cells

PDF

Development of polystyrene sulfonic acid-polyvinylidene fluoride (PSSA-PVDF) blends for direct methanol fuel cells (DMFCS)

PDF

Properties and applications of dipyrrin-based molecules in organic photovoltaics

PDF

The influence of CdSe surface ligands on hybrid inorganic/organic solar cell performance

Asset Metadata

Creator Khlyabich, Petr P. (author)

Core Title Efficient ternary blend bulk heterojunction solar cells with tunable open-circuit voltage

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Chemistry

Publication Date 10/21/2013

Defense Date 10/10/2013

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag conjugated polymer,fullerene,OAI-PMH Harvest,organic alloy,random copolymer,semi-random copolymer,solar cell,ternary blend

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Thompson, Barry C. (committee chair), Nakano, Aiichiro (committee member), Thompson, Mark E. (committee member)

Creator Email khlyabic@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-341334

Unique identifier UC11296339

Identifier etd-KhlyabichP-2110.pdf (filename),usctheses-c3-341334 (legacy record id)

Legacy Identifier etd-KhlyabichP-2110.pdf

Dmrecord 341334

Document Type Dissertation

Format application/pdf (imt)

Rights Khlyabich, Petr P.

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA