University of Southern California Dissertations and Theses

Influence of polymer structure on electronic properties and performance as donor materials in bulk heterojunction solar cells

(USC Thesis Other)

Influence of polymer structure on electronic properties and performance as donor materials in bulk heterojunction solar cells

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Copyright 2015 Bing Xu

INFLUENCE OF POLYMER STRUCTURE ON ELECTRONIC PROPERTIES AND
PERFORMANCE AS DONOR MATERIALS IN BULK HETEROJUNCTION
SOLAR CELLS

by

Bing Xu

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)

May 2015
ii ii
ACKNOWLEDGEMENTS
As I finally approach the end of this long journey, I would like to express my deepest
appreciation to the many people who have been extremely supportive and helpful in the past
years. The accomplishment of this thesis would not be possible if there were no such
guidance, blessing and encouragement from them, where I have always been so fortunate to
find the warmth and inspiration I need.
I owe the most special thanks to my advisor Prof. Barry C. Thompson. He has led me into
the jungle of science and guided me through every single challenge with great patience, so
that I am able to reach as far as I can. From developing plans to solving problems, to
analyzing results and to writing up publications, Barry has taught me all the tiniest details of
the research process. More importantly, it is Barry who set the great example himself and
showed me how to face the obstacles in research and how to keep being confident and
positive even when walking through frustrations. His encouragement and his belief in my
ability have also become a lighthouse of my life, giving me strength to fight for the best in
every aspect.
I would like to thank all the members of our research group for being supportive colleagues
and good friends. I am grateful to my senior group members Dr. Beate Burkhart, Dr. Petr
Khlyabich, Dr. Alejandra Beier and Dr. Andrey Rudenko who taught me many things. I had
iii iii
a privilege to work with Sangtaik Noh on several projects and I want to thank him for being a
great research collaborator. I also want to thank other group members Alia Latif, Jenna
Howard, Seyma Ekiz, Nemal Gobalasingham and Betsy Melenbrink for sharing a lot of
pleasant time both in lab and out of lab.
I am grateful to my previous advisor Thieo E. Hogen-Esch, not only for the patient
guidance he provided as I first entered graduate school in 2009, but also for his
understanding and support when I decided in 2011 to work on polymer photovoltaics with
Barry. I also want to thank Prof. G. K. Surya Prakash, Prof. Chao Zhang, Prof. Matthew R.
Pratt, and Prof. Malancha Gupta, for agreeing to serve on my qualifying examination.
I would like to thank those professors, postdocs and my fellow graduate students at USC
who have done a lot of favors to my own research projects. I thank Prof. Mark E. Thompson
for his permission to use their atomic force microscope and ellipsometer, and Dr. Rui Zhang
and John Chen for their help with the instruments. I thank Prof. Malancha Gupta, Scott
Seidel and Benny Chen for allowing me to use their goniometer. I thank Dr. Ralf Haiges for
his advice on working with explosive chemicals. I also thank both Prof. Stephen B. Cronin
and Prof. Chongwu Zhou for their permissions to access the atomic force microscopes in
their groups, as well as the time and effort of Dr. Jesse Theiss and Dr. Jia Liu in helping me
with the instruments.
iv iv
I appreciate both the warm company and the professional discussion with all my close
friends in the chemistry department. Special thanks go to Dr. Wenbo Hou, Dr. Bo Yang, Dr.
Yifei Liu, Dr. Rui Zhang, Zhe Zhang, Yan Song, Xu Liu and many others. I will always
cherish the great moments I have spent with all of you throughout these years. I am also
thankful to all my other classmates and friends, for sharing part of our lifetime together. I
also would like to thank my teachers Prof. Linqi Shi in Nankai University and Jianchi Zhang
in my high school who made a lasting impression on me and guided into the world of
chemistry.
At last but not least, I hold a special place in my heart to my family who love me
unconditionally. Mom and dad, thank you for everything you have done. I love you.

v v
TABLE OF CONTENTS
ACKNOWLEDGEMENTS ··········································································· ii
LIST OF TABLES ···················································································· viii
LIST OF FIGURES ···················································································· xi
ABSTRACT ·························································································· xxii
CHAPTER 1. Random Conjugated Copolymers for Organic Photovoltaics ···················· 1
1.1 Introduction ···················································································· 1
1.2 Regioregular poly(3-substituted thiophene)s ·············································· 5
1.2.1 Alkyl side chain effects ······························································· 8
1.2.2 Non-alkyl thiophene copolymers ··················································· 13
1.3 A-A type bridging unit plus two or more B-B type comonomers ······················ 18
1.3.1 Special cases with M1 and M2 sharing the same core structure ················ 20
1.3.2 Thiophene or ethylene as bridging units in Stille polymerizations ············· 27
1.3.3 Other random D-A copolymers by Stille or Suzuki polymerizations ·········· 47
1.3.4 Random copolymers prepared by other polymerization
methods or post-functionalization reactions ······································ 69
1.3.5 Importance of comonomer linkage patterns ······································· 71
1.4 Semi-random copolymers ··································································· 74
1.5 Other copolymerization strategies ························································ 82
1.6 Summary and conclusion ···································································· 88
1.7 References ····················································································· 90
CHAPTER 2. Fine Tuning of Polymer Properties by Incorporating
Strongly Electron-Donating 3-Hexyloxythiophene Units into
Random and Semi-random Copolymers ····················································· 102
2.1 Introduction ············································································ 102
2.2 Monomer and polymer syntheses ··················································· 106
2.3 Electronic properties ·································································· 110
2.4 Solar cell characterization ··························································· 117
2.5 Conclusion ············································································· 122
2.6 References ············································································· 123

vi vi
CHAPTER 3. Random Terpolymers Based on 3-Hexyloxythiophene,
2,1,3-Benzothiadiazole and 2,7-Carbazole ················································· 127
3.1 Introduction ············································································ 127
3.2 Monomer and polymer syntheses ··················································· 131
3.3 Optoelectronic properties of polymers ············································· 135
3.4 Summary ··············································································· 138
3.5 References ············································································· 140
CHAPTER 4. Morphology and Device Performance of Polymer:Small
Molecule:Fullerene Ternary Blend Solar Cells in Comparison to
Binary Blend Cells Based on Polymers with Covalently Attached
Small Molecules ··········································································· 144
4.1 Introduction ············································································ 144
4.2 Material syntheses ···································································· 147
4.3 Device performance of ternary blends ············································· 151
4.4 Morphology of ternary blends ····················································· 153
4.5 Characterization and device performance of P1-P3 ······························ 157
4.6 Conclusion ············································································· 160
4.7 References ············································································· 162
CHAPTER 5. Stereoregular Non-conjugated Polymers with
Diketopyrrolopyrrole Electroactive Pendant Groups ······································ 165
5.1 Introduction ············································································ 165
5.2 System Design ········································································· 170
5.3 Materials Syntheses ··································································· 172
5.3.1 Stereorandom and stereoregular free radical polymerizations
of PNIPAM ·········································································· 172
5.3.2 Synthesis of alkyne functionalized polystyrene and
polyacrylamides (PS-alkyne, PA-atactic and PA-isotactic) ·················· 175
5.3.3 “Click” reactions to generate electroactive pendant polymers ················ 178
5.4 Characterizations of Polymer Samples and DPP-N
3
····························· 178
5.5 Conclusions and Summary ·························································· 184
5.6 References ············································································· 186
BIBLIOGRAPHY ···················································································· 189

vii vii
APPENDIX 1. Fine Tuning of Polymer Properties by Incorporating
Strongly Electron-Donating 3-Hexyloxythiophene Units into
Random and Semi-random Copolymers ····················································· 204
A1.1 Materials and Methods ··························································· 204
A1.2 Synthetic Procedures ······························································ 206
A1.3 Device fabrication and characterization ········································· 210
A1.4 Structure Verifications of Small Molecules and Polymers ···················· 214
A1.5 Characterization of Polymer Samples ············································ 230
A1.6 References ··········································································· 235
APPENDIX 2. Random Terpolymers Based on 3-Hexyloxythiophene,
2,1,3-Benzothiadiazole and 2,7-Carbazole ················································· 237
A2.1 Materials and Methods ··························································· 237
A2.2 Synthetic Procedures and Structural Characterizations ························ 238
A2.3 Structure Verifications of Small Molecules and Polymers ···················· 245
A2.4 Characterization of Polymer Samples ············································ 256
A2.5. References ··········································································· 257
APPENDIX 3. Morphology and Device Performance of Polymer:Small
Molecule:Fullerene Ternary Blend Solar Cells in Comparison to
Binary Blend Cells Based on Polymers with Covalently Attached
Small Molecules ················································································ 258
A3.1 Materials and Methods ··························································· 258
A3.2 Synthetic Procedures and Structural Characterizations ························ 259
A3.3 Device Fabrication and Characterization ········································ 279
A3.4 Characterization of Polymer Samples and Blends ······························ 280
A3.5. References ··········································································· 282
APPENDIX 4. Stereoregular Non-conjugated polymers with
Diketopyrrolopyrrole Electroactive pendant groups ······································· 283
A4.1 Materials and Methods ··························································· 283
A4.2 Synthetic Procedures ······························································ 284
A4.3 Structure Verifications of Small Molecules and Polymers ···················· 295
A4.4 Device Fabrication and Mobility Measurements ······························· 310
A4.5 Characterization of Polymer Samples and DPP-N
3
···························· 311
viii viii
LIST OF TABLES
Table 1.1. Typical conditions of common polymerization methods for
preparing regioregular poly(3-substituted thiophene)s. ······································· 6
Table 1.2. Summary of polymer properties and device performance parameters
of polythiophene-based copolymers with side chain engineering. ························· 12
Table 1.3. Summary of polymer properties and solar cell performance
parameters of polythiophene-based copolymers with electronically
distinct comonmer units. ········································································· 17
Table 1.4. Summary of polymer properties and device performance
parameters of polyfluorene-based copolymers following the strategy
described in Section 1.3.1. ······································································· 24
Table 1.5. Summary of polymer properties and device performance parameters
of other copolymers following the strategy described in Section 1.3.1. ··················· 27
Table 1.6. Polymer properties and device performance parameters of
copolymers shown in Figure 1.9 and 1.10. ···················································· 33
Table 1.7. Polymer properties and device performance parameters of
copolymers shown in Figure 1.11. ······························································ 36
Table 1.8. Polymer properties and device performance parameters of
copolymers shown in Figure 1.12. ······························································ 39
Table 1.9. Polymer properties and device performance parameters of
copolymers shown in Figure 1.13. ······························································ 42
Table 1.10. Polymer properties and device performance parameters of
copolymers shown in Figure 1.14 and Figure 1.15. ·········································· 45
Table 1.11. Polymer properties and device performance parameters of
random copolymers shown in Figure 1.17. ···················································· 50

ix ix
Table 1.12. Polymer properties and device performance parameters of
random copolymers shown in Figure 1.18. ···················································· 55
Table 1.13. Polymer properties and device performance parameters of
random copolymers shown in Figure 1.19 and 1.20. ········································· 59
Table 1.14. Polymer properties and device performance parameters of
random copolymers depicted in Figure 1.21. ················································· 64
Table 1.15. Polymer properties and device performance parameters of
random copolymers depicted in Figure 1.22. ················································· 67
Table 1.16. Polymer properties and device performance parameters of
1A/2D copolymers depicted in Figure 1.23. ·················································· 68
Table 1.17. Polymer properties and device performance parameters of
copolymers depicted in Figure 1.24 and 1.25. ················································ 70
Table 1.18. Polymer properties and device performance parameters of
copolymers depicted in Figure 1.26 and 1.27. ················································ 73
Table 1.19. Polymer properties and device performance parameters of
semi-random copolymers depicted in Figure 1.29. ··········································· 81
Table 1.20. Polymer properties and device performance parameters of
semi-random copolymers depicted in Figure 1.31-1.33. ····································· 86
Table 2.1. Molecular Weight and Electronic Properties of All Polymers. ··················· 109
Table 2.2. Average Photovoltaic Parameters and SCLC Hole Mobilities
of Random P3HT-co-3HOT and Semi-random P3HTT-HOT-DPP
Copolymers. ····················································································· 117
Table 3.1. Molecular weights and compositions of polymers ································· 134
Table 3.2. Optoelectronic properties of all polymers ··········································· 138
Table 4.1. PhDPP compositions and molecular weights of P1-P3. ··························· 150
Table 4.2. Device performance of P3HT:PhDPP:PC
61
BM ternary blend
solar cells before and after annealing. ························································ 152
x x
Table 4.3. Device performance of solar cells based on P1-P3 blended
with PC
61
BM. ···················································································· 158
Table 4.4. HOMO levels of PhDPP-attached P3HT polymers P1-P3 in
both films and solutions, together with those of P3HT and PhDPP
small molecule. ·················································································· 160
Table 5.1. Polymerizations of NIPAM at 60
o
C for 24h. ······································· 173
Table 5.2. Molecular weights and dyad tacticities of polymer samples. ····················· 178
Table 5.3. HOMO energy levels, optical bandgaps and hole mobilities of
all samples. ······················································································· 183
Table A1.1. Summary of UV-Vis Absorption Data of All Polymers ························· 231
Table A1.2. Summary of XRD Data of All Polymers ·········································· 234
Table A1.3. Summary of 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 BHJ solar cells based on
semi-random copolymers. ····································································· 235
Table A1.4. SCLC hole mobilities of semi-random copolymers measured
in blends with PC
61
BM at their optimized conditions. ····································· 235
Table A2.1. Summary of UV-Vis Absorption Data of All Polymers. ························ 256
Table A3.1. GIXRD data summary of P3HT:PhDPP:PC
61
BM ternary blend
films before and after annealing. ······························································ 282

xi xi
LIST OF FIGURES
Figure 1.1. (a) Typical device architecture and active layer structure of
bulk heterojunction polymer solar cells. (b) Molecular structure of
previously reported PTB7 as an example of D-A alternating copolymers. ··············· 3
Figure 1.2. Schematic syntheses and structures of polythiophene-based
copolymers with side chain engineering. ······················································· 8
Figure 1.3. Schematic syntheses and structures of polythiophene-based
copolymers with side chain engineering (continued). ········································ 11
Figure 1.4. Schematic syntheses and structures of polythiophene-based
copolymers with electronically distinct comonomer units. ·································· 14
Figure 1.5. General scheme of the synthetic route to prepare random
conjugated copolymers using one A-A type bridging unit and two
B-B type comonomer units, where A and B in the scheme
represents complementary functionalities. ···················································· 18
Figure 1.6. Schematic syntheses and structures of polyfluorene-based
copolymers following the strategy described in Section 1.3.1. ····························· 21
Figure 1.7. Syntheses and structures of other copolymers following the
strategy described in Section 1.3.1. ····························································· 25
Figure 1.8. General schematic synthesis of copolymers with thiophene or
ethylene bridging units via Stille polymerization. ············································ 28
Figure 1.9. Schematic syntheses and comonomer structures for copolymers
with thiophene bridging units. Composition being m:n = 1:1 unless specified. ·········· 29
Figure 1.10. Comonomer structures for copolymers with thiophene bridging
units (continued) synthesized via the schematic Stille polymerization
method depicted in Figure 1.9. Composition being m:n = 1:1 unless specified. ········· 31
Figure 1.11. Comonomer structures for copolymers with thiophene bridging
units (continued) synthesized via the schematic Stille polymerization
method depicted in Figure 1.9. ·································································· 34
xii xii
Figure 1.12. Schematic syntheses and comonomer structures for copolymers
with ethylene bridging units. Composition being m:n = 1:1 for all copolymers. ········· 38
Figure 1.13. Schematic syntheses and comonomer structures for copolymers
with thiophene or ethylene bridging units and two other acceptor units.
Composition being m:n = 1:1 unless specified. ··············································· 40
Figure 1.14. Schematic syntheses and comonomer structures for copolymers
containing thiophene bridging units as well as 3-hexylthiophene. ························· 43
Figure 1.15. Syntheses and structures of copolymers with thiophene bridging
units and three other comonomer units. ························································ 46
Figure 1.16. General scheme of the two sub-types of copolymers
containing an electron-donating unit as M1. ·················································· 47
Figure 1.17. Schematic syntheses via Stille polymerization and comonomer
structures of random copolymers that fall into the description in Figure 1.16a. ········· 48
Figure 1.18. Schematic syntheses via Suzuki polymerization and comonomer
structures of random copolymers that fall into the description in
Figure 1.16a. Composition being m:n = 1:1 unless specified. ······························ 53
Figure 1.19. Schematic syntheses via Stille polymerization and comonomer
structures of random copolymers that fall into the description in
Figure 1.16b. Composition being m:n = 1:1 unless specified. ······························ 56
Figure 1.20. Comonomer structures for random copolymers synthesized
by Stille polymerization following the synthetic scheme depicted in
Figure 1.19 (continued). ·········································································· 60
Figure 1.21. Comonomer structures for random copolymers synthesized
by Stille polymerization following the synthetic scheme depicted in
Figure 1.19 (continued). ·········································································· 63
Figure 1.22. Schematic syntheses via Suzuki polymerization and comonomer
structures of random copolymers that fall into the description in Figure 1.16b. ········· 66
Figure 1.23. Schematic syntheses comonomer structures of 1A/2D copolymers. ············ 67

xiii xiii
Figure 1.24. Syntheses and structures of copolymers prepared by
Suzuki-Heck polymerizations. ·································································· 69
Figure 1.25. Syntheses and structures of random copolymers prepared
by post-functionalizations. ······································································· 71
Figure 1.26. Structures of random copolymers with isomeric comonomer linkages. ········ 72
Figure 1.27. Structures of random copolymers with isomeric comonomer
linkages (continued). ············································································· 73
Figure 1.28. General scheme of synthetic route to prepared semi-random
copolymers by Stille polymerizations. ························································· 74
Figure 1.29. Structures of semi-random copolymers that follow the synthetic
strategy shown in Figure 1.28. ·································································· 77
Figure 1.30. General scheme of synthetic route to prepare random copolymers
through other polymerization strategies. ······················································· 82
Figure 1.31. Schematic syntheses and structures of random copolymers prepared
by Gilch polymerizations. ······································································ 84
Figure 1.32. Schematic synthesis and structure of random copolymers prepared
by Yamamoto polymerization, in comparison to the alternating copolymer
prepared by Suzuki polymerization. ·························································· 87
Figure 1.31. Schematic syntheses and structures of random copolymers prepared
by ADMET polymerization. ····································································· 88
Figure 2.1. Syntheses and Structures of (a) P3HTT-DPP-DTP (reported
previously
12
) and (b) P3HTT-HOT-DPP (reported here). ······························ 105
Figure 2.2. Synthetic Scheme of (a) Monomer 2-Bromo-5-trimethyltin-3-
hexyloxythiophene (4), (b) Stille Polymerization for Semi-random
P3HTT-HOT-DPP Copolymers and (c) Stille Polymerization for Random
P3HT-co-3HOT Copolymers. ································································ 107

xiv xiv
Figure 2.3. UV−vis absorption of (a) random P3HT-co-3HOT copolymers
and (b) semi-random P3HTT-HOT-DPP copolymers in thin films spin-coated
from o-DCB and annealed for 30 min under N
2
at 150 °C (P3HOT thin film is
as-cast): (i) P3HT, (ii) P3HT
90
-co-3HOT
10
, (iii) P3HT
75
-co-3HOT
25
,
(iv) P3HT
50
-co-3HOT
50
, (v) P3HOT, (vi) P3HTT-DPP,
(vii) P3HTT-HOT-DPP(75:5), (viii) P3HTT-HOT-DPP(70:10),
(ix) P3HTT-HOT-DPP(65:15), (x) P3HTT-HOT-DPP(60:20)
and (xi) P3HTT-HOT-DPP(40:40). ························································· 112
Figure 2.4. Grazing-Incidence X-ray Diffraction of thin films of (a) random
P3HT-co-3HOT copolymers and (b) semi-random P3HTT-HOT-DPP
copolymers (spin-coated from o-DCB, as cast for P3HOT and annealed
for 30 min under N
2
at 150 °C for other polymers): (i) P3HT,
(ii) P3HT
90
-co-3HOT
10
, (iii) P3HT
75
-co-3HOT
25
, (iv) P3HT
50
-co-3HOT
50
,
(v) P3HOT, (vi) P3HTT-DPP, (vii) P3HTT-HOT-DPP(75:5),
(viii) P3HTT-HOT-DPP(70:10), (ix) P3HTT-HOT-DPP(65:15),
(x) P3HTT-HOT-DPP(60:20) and (xi) P3HTT-HOT-DPP(40:40). ················· 115
Figure 2.5. HOMO levels in the solid state (filled squares) and V
oc
(open circles)
of the optimized solar cells of (a) random P3HT-co-3HOT copolymers and
(b) semi-random P3HTT-HOT-DPP copolymers as a function of 3HOT
content in the polymer backbone. ····························································· 119
Figure 2.6. EQE of the BHJ solar cells based on P3HTT-DPP (black squares),
P3HTT-HOT-DPP(75:5) (red circles), P3HTT-HOT-DPP(70:10) (green
triangles), P3HTT-HOT-DPP(65:15) (blue triangles), P3HTT-HOT-DPP(60:20)
(purple stars), and P3HTT-HOT-DPP(40:40) (pink triangles) with PC
61
BM
as the acceptor, under optimized condition for device fabrication. ······················· 121
Figure 3.1. Structure of polymers (a) PCDTBT, (b) PCDTBT-OR reported in
Ref 17 and 18, and (c) PCDTBT
m
DHOTBT
n
and PCDHOTBT reported here. ····· 128
Figure 3.2. Synthetic scheme to prepare brominated comonomer DHOTBT (5). ·········· 131
Figure 3.3. Isomeric structure to the brominated DHOTBT comonomer (Compound 5). · 132
Figure 3.4. Synthetic scheme of Suzuki polymerization to prepare random
terpolymers and the alternating copolymers PCDTBT and PCDHOTBT. ············ 133

xv xv
Figure 3.5. UV−vis absorption of copolymers in (a) thin films spin-coated from
CHCl
3
and annealed for 30 min under N
2
at 150 °C and (b) o-DCB solutions:
(black) PCDTBTT, (red) PCDTBT
3
DHOTBT
1
, (green) PCDTBT
1
DHOTBT
1
,
(dark blue) PCDTBT
1
DHOTBT
3
, (light blue) PCDHOTBT. ·························· 137
Figure 4.1. Structures of materials used in this work. ·········································· 146
Figure 4.2. Structures, energy levels and UV-Vis absorption of materials used in
the ternary blend solar cells in this work. ···················································· 148
Figure 4.3. (a) “Click” reaction to covalently attach PhDPP onto the side chains of
P3HT; (b) synthetic routes towards azide group functionalized P3HT. Reaction
condition: (i) CuBr/PMDETA, THF, 35 °C, 3d. ············································ 149
Figure 4.4. GIXRD of P3HT:PhDPP:PC
61
BM ternary blend as-cast films spin-coated
from CHCl
3
solution: (i) P3HT80:PhDPP20:PC
61
BM, (ii) P3HT60:PhDPP40:
PC
61
BM, (iii) P3HT40:PhDPP60:PC
61
BM, (iv) P3HT20: PhDPP80:PC
61
BM. ········ 154
Figure 4.5. GIXRD of P3HT:PhDPP:PC
61
BM ternary blend films spin-coated
from CHCl
3
solution followed by annealing with o-DCB vapor for 2 min:
(ii) P3HT80:PhDPP20: PC
61
BM, (iii) P3HT60:PhDPP40:PC
61
BM,
(iv) P3HT40:PhDPP60:PC
61
BM, (v) P3HT20:PhDPP80:PC
61
BM, along with
(i) P3HT and (vi) PhDPP films thermal annealed at 150 °C. ····························· 154
Figure 4.6. AFM topology images of (a and e) P3HT80:PhDPP20:PC
61
BM,
(b and f) P3HT60:PhDPP40:PC
61
BM, (c and g) P3HT40:PhDPP60:PC
61
BM
and (d and h) P3HT20:PhDPP80: PC
61
BM films spin-coated from CHCl
3

solution, before (a, b, c and d) and after (e, f, g and h) annealing with o-DCB
vapor for 2 min. ················································································· 156
Figure 4.7. GIXRD of films of PhDPP-attached P3HT polymers, as well as
P3HT, spin-coated from CHCl3 solution and annealed at 150 °C for 30 min:
(black) P3HT, (red) P1, (green) P2 and (blue) P3. ······································· 158
Figure 4.8. Solar cell V
oc
values of as a function of PhDPP composition of
(open circle) P3HT:PhDPP:PC
61
BM ternary blends before annealing, (filled
square) P3HT:PhDPP:PC
61
BM ternary blends after annealing and (filled
triangle) PhDPP-attached P3HT polymers P1-P3. ········································· 159

xvi xvi
Figure 5.1. (a) Cartoon representation of the process of phase separation in A-B
diblock copolymers. (b) Cartoons illustrating the different morphologies that
are commonly observed in A-B diblock copolymers. When the fraction of A (fA)
is greater than the fraction of B (fB), spherical (S), cylindirical (C), or gyroid (G)
phases of B can exist within a continuous matrix of A. The converse is true when
fA < fB. If fA = fB, a lamellar phase (L) can exist. Note that C, G, and L phases
are bicontinuous. ················································································ 166
Figure 5.2. Chemical structure of an atactic block copolymer reported by
Thelakkat.
14
······················································································ 167
Figure 5.3. (a) Illustration of the axis of p-overlap generated via face-to-face packing
in conjugated molecules that leads to effective charge transport along the p-axis.
(b) Illustration of the effective p-overlap generated in isotactic poly(N-
vinylcarbazole) in the extended all-trans form. (c) Illustration of the poor
p-overlap generated in atactic poly(N-vinylcarbazole) in the extended all-trans
form. (1) Poly(N-vinylcarbazole). (2) Poly(N-ethylcarbazole acrylate). ················ 169
Figure 5.4. Polymer and small molecule samples investigated in this work. ················ 170
Figure 5.5. Cartoon illustrating the mechanism of stereoregularity control in
Lewis Acid-mediated radical polymerizations that lead to isotactic polymers. ······· 172
Figure 5.6.
1
H NMR spectra of (a) atactic PNIPAM from Entry 1 and (b) isotactic
PNIPAM from Entry 5 in DMSO-d
6
at 125 °C. ············································ 174
Figure 5.7. Synthetic routes to prepare atactic alkyne functionalized polystyrene
(PS-alkyne). ····················································································· 176
Figure 5.8. Synthetic routes to prepare alkyne-functionalized polyacrylamides with
both stereorandom and stereoregular backbones (PA-TMS-atactic
and PA-TMS-isotactic respectively). ························································ 176
Figure 5.9. 1H NMR spectra of (a) PA-TMS-atactic and (b) PA-TMS-isotactic
in DMSO-d
6
at 125 °C. ········································································· 177
Figure 5.10. UV-Vis Absorption of DPP-N
3
(black), PS-DPP (red), PA-DPP-atactic
(green) and PA-DPP-isotactic (blue) in (a) o-DCB solution and (b) thin
films spin-coated from chloroform followed by thermal annealing at 110 °C
for 20 min. ······················································································· 180
xvii xvii
Figure 5.11. GIXRD of thin film of DPP-N
3
(black), PS-DPP (red), PA-DPP-atactic
(green) and PA-DPP-isotactic (blue) spin-coated from CHCl
3
and annealed at
110 °C for 20 min. ············································································· 181
Figure A1.1.
1
H NMR of 3-methoxythiophene (1). ············································· 214
Figure A1.2.
1
H NMR of 3-hexyloxythiophene (2). ············································ 215
Figure A1.3.
1
H NMR of 2-bromo-3-hexyloxythiophene (3). ································· 216
Figure A1.4.
1
H NMR of 2-bromo-5-trimethyltin-3-hexyloxythiophene (4). ··············· 217
Figure A1.5.
13
C NMR of 2-bromo-5-trimethyltin-3-hexyloxythiophene (4). ·············· 218
Figure A1.6.
1
H NMR of P3HT. ·································································· 219
Figure A1.7.
1
H NMR of P3HT
90
-co-3HOT
10
. ················································· 220
Figure A1.8.
1
H NMR of P3HT
75
-co-3HOT
25
. ················································· 221
Figure A1.9.
1
H NMR of P3HT
50
-co-3HOT
50
. ················································· 222
Figure A1.10.
1
H NMR of homopolymer P3HOT. ············································· 223
Figure A1.11.
1
H NMR of P3HTT-DPP. ························································ 224
Figure A1.12.
1
H NMR of P3HTT-HOT-DPP(75:5). ········································· 225
Figure A1.13.
1
H NMR of P3HTT-HOT-DPP(70:10). ······································· 226
Figure A1.14.
1
H NMR of P3HTT-HOT-DPP(65:15). ······································· 227
Figure A1.15.
1
H NMR of P3HTT-HOT-DPP(60:20). ······································· 228
Figure A1.16.
1
H NMR of P3HTT-HOT-DPP(40:40). ······································· 229

xviii xviii
Figure A1.17. UV−vis absorption of (a) random P3HT-co-P3HOT copolymers in
CB solutions and (b) semi-random P3HTT-HOT-DPP copolymers in o-DCB
solutions: (i) P3HT, (ii) P3HT
90
-co-HOT
10
, (iii) P3HT
75
-co-HOT
25
,
(iv) P3HT
50
-co-HOT
50
, (vi) P3HTT-DPP, (v) P3HOT, (vii) P3HTT-HOT-
DPP(75:5), (viii) P3HTT-HOT-DPP(70:10), (ix) P3HTT-HOT-DPP(65:15),
(x) P3HTT-HOT-DPP(60:20) and (xi) P3HTT-HOT-DPP(40:40). ·················· 230
Figure A1.18. CV traces of random P3HT-co-3HOT copolymers as well as
homopolymers P3HT and P3HOT. ·························································· 232
Figure A1.19. CV traces of semi-random P3HTT-HOT-DPP copolymers. ················ 233
Figure A1.20. Representative J-V curves of semi-random copolymers:PC
61
BM BHJ
solar cells under AM 1.5G illumination (100 mW/cm
2
) at optimized conditions
for solar cell performance. ····································································· 234
Figure A2.1. Synthetic scheme to prepare the isomeric comonomer Compound 6. ········ 242
Figure A2.2.
1
H NMR of Compound 2 in CDCl
3
. ·············································· 245
Figure A2.3.
1
H NMR of Compound 4 in CDCl
3
. ·············································· 246
Figure A2.4.
1
H NMR of Compound 5 in CDCl
3
. ·············································· 247
Figure A2.5.
1
H NMR of polymer PCDTBT
3
DHOTBT
1
in CDCl
3
. ························ 248
Figure A2.6.
1
H NMR of polymer PCDTBT
1
DHOTBT
1
in CDCl
3
. ························ 249
Figure A2.7.
1
H NMR of polymer PCDTBT
1
DHOTBT
3
in CDCl
3
. ························ 250
Figure A2.8.
1
H NMR of polymer PCDHOTBT in CDCl
3
. ·································· 251
Figure A2.9.
1
H NMR of Compound 8 in CDCl
3
. ·············································· 252
Figure A2.10.
1
H NMR of Compound 9 in CDCl
3
. ············································· 253
Figure A2.11.
1
H NMR of Compound 10 in CDCl
3
. ··········································· 254
Figure A2.12.
1
H NMR of Compound 11 in CDCl
3
. ··········································· 255
Figure A2.13.
1
H NMR of Compound 6 in CDCl
3
. ············································· 256
xix xix
Figure A2.14. GIXRD of films of PCDTBT derivative polymers spin-coated from
CHCl
3
solution and annealed at 150 °C for 30 min: (black) PCDTBT,
(red) PCDTBT
3
DHOTBT
1
, (green) PCDTBT
1
DHOTBT
1
,
(blue) PCDTBT
3
DHOTBT
1
and (light blue) PDHOTBT. ····························· 257
Figure A3.1. Syntheses of monomers 2,5-dibromo-3- hexylthiophene (1)
and 2,5-Dibromo-3-{2-bromoethyl} thiophene (3). ········································ 260
Figure A3.2. Syntheses of alkyne-functionalized DPP molecule (11). ······················· 262
Figure A3.3.
1
H NMR of Compound 11 in CDCl
3
. ············································· 266
Figure A3.4.
1
H NMR of polymer P3BrT-1 in CDCl
3
. ········································ 268
Figure A3.5.
1
H NMR of polymer P3BrT-2 in CDCl
3
. ········································ 269
Figure A3.6.
1
H NMR of polymer P3BrT-3 in CDCl
3
. ········································ 270
Figure A3.7.
1
H NMR of polymer P3N
3
T-1 in CDCl
3
. ········································ 271
Figure A3.8.
1
H NMR of polymer P3N
3
T-2 in CDCl
3
. ········································ 272
Figure A3.9.
1
H NMR of polymer P3N
3
T-3 in CDCl
3
. ········································ 273
Figure A3.10.
1
H NMR of polymer P1 in CDCl
3
. ·············································· 275
Figure A3.11.
1
H NMR of polymer P2 in CDCl
3
. ·············································· 276
Figure A3.12.
1
H NMR of polymer P3 in CDCl
3
. ·············································· 277
Figure A3.13.
1
H NMR of P3HT in CDCl
3
. ····················································· 278
Figure A3.14. Synthesis of 2,5-di(2-ethylhexyl)-3,6-bis(5-phenylthiophene-2-yl)
pyrrolo[3,4-c]-pyrrole-1,4-dione (13). ······················································· 278
Figure A3.15.
1
H NMR of Compound 13 in CDCl
3
. ··········································· 279
Figure A3.16. UV-Vis Absorption of P3HT:PhDPP blend films as well as pristine
P3HT and PhDPP films spin-coated from CHCl
3
and annealed at 150 °C for
30 min. (i) P3HT, (ii) P3HT80:PhDPP20, (iii) P3HT60:PhDPP40,
(iv) P3HT40:PhDPP60, (v) P3HT20:PhDPP80 and (vi) PhDPP. ························ 280
xx xx
Figure A3.17. UV-Vis Absorption of P1-P3 and P3HT solutions in o-DCB.
(i) P3HT, (ii) P1, (iii) P2 and (iv) P3. ························································ 281
Figure A3.18. UV-Vis Absorption of P1-P3 and P3HT films spin-coated from
CHCl
3
and annealed at 150 °C for 30 min. (i) P3HT, (ii) P1, (iii) P2 and (iv) P3. ···· 282
Figure A4.1. Synthesis of N-(3-(trimethylsilyl)prop-2-yn-1-yl)acrylamide (1). ············ 284
Figure A4.2. Syntheses of DPP-N
3
. ······························································· 286
Figure A4.3.
1
H NMR spectrum of Compound 1a in CDCl
3
. ································· 295
Figure A4.4.
1
H NMR spectrum of Compound 1 in CDCl
3
. ··································· 296
Figure A4.5.
1
H NMR spectrum of Compound 2 in CDCl
3
. ··································· 297
Figure A4.6.
1
H NMR spectrum of Compound 3 in CDCl
3
. ··································· 298
Figure A4.7.
1
H NMR spectrum of Compound 4 in CDCl
3
. ··································· 299
Figure A4.8.
1
H NMR spectrum of Compound 5 in CDCl
3
. ··································· 300
Figure A4.9.
1
H NMR spectrum of Compound 6 in CDCl
3
. ··································· 301
Figure A4.10.
1
H NMR spectrum of Compound 7 in CDCl
3
. ································· 302
Figure A4.11.
1
H NMR spectrum of DPP-N
3
in CDCl
3
. ······································· 303
Figure A4.12.
13
H NMR spectrum of DPP-N
3
in CDCl
3
. ······································ 304
Figure A4.13.
1
H NMR spectrum of PS-alkyne in DMSO-d
6
. ······························· 305
Figure A4.14.
1
H NMR spectrum of PA-atactic in DMSO-d
6
. ······························· 306
Figure A4.15.
1
H NMR spectrum of PA-isotactic in DMSO-d
6
. ····························· 307
Figure A4.16.
1
H NMR spectrum of PS-DPP in CHCl
3
. ······································ 308
Figure A4.17.
1
H NMR spectrum of PA-DPP-atactic in CDCl
3
. ···························· 309
Figure A4.18.
1
H NMR spectrum of PA-DPP-isotactic in CDCl
3
. ·························· 310
xxi xxi
Figure A4.19. AFM topology images of thin films of (a) PS-DPP, (b) PA-DPP-
atactic, (c) PA-DPP-isotactic and (d) DPP-N
3
spin-coated from 10 mg/mL
CDCl
3
solutions followed by thermal annealing at 110 °C for 20 min. ················· 311
Figure A4.20. Second heating in differential scanning calorimetry of
DPP-N
3
(black), PS-DPP (red), PA-DPP-atactic (green) and PA-DPP-
isotactic (blue). Dash line indicates the temperature to anneal thin films
for UV-Vis absorption and hole mobility measurements. ································· 312

xxii xxii
ABSTRACT
Essential to the achievement of higher efficiencies in polymer:fullerene bulk
heterojunction solar cells is the development of a polymer donor material that simultaneously
meets the requirements of a broad absorption profile, suitable energy levels and a favorable
morphology with fullerene derivatives. In an effort to pursue the ideal polymer,
copolymerizing electron-rich and electron-poor monomers in a perfectly alternating manner
(alternating donor-acceptor (D-A) copolymers) has produced state-of-the-art polymer
samples through the “monomer approach”. As an alternative strategy, polymer properties can
be modulated by the strategic combination of monomers through the “polymer approach”.
Following this concept, in particular, a number of random copolymers have been reported in
literature, revealing their distinct advantages in the ease of synthesis, as well as in the
convenience and broad scope of property tuning ability.
This thesis has been inspired by the previous works on electroactive polymers for organic
photovoltaics that are synthesized through the “polymer approach”, especially those with
random conjugated backbones. Focus in this thesis is on the synthesis of electroactive
polymers following the “polymer approach” and their structure-property relationships.
Chapter 1 reviews the reported examples of random conjugated polymers in the context of
organic solar cell applications, with an emphasis on their synthetic strategies,
structure-property relationships and photovoltaic device performance. Several specific
strategies to produce random copolymers are recognized and discussed according to
xxiii xxiii
comonomer structures and reaction features. In particular, three popular strategies reviewed
in Chapter 1, namely the random regioregular polythiophenes, random D-A copolymers and
semi-random D-A copolymers, will be reflected in the polymer structures reported in Chapter
2, 3 and 4. In addition, a brief summary is included in Chapter 1 on the advantages of random
conjugated copolymers, especially in comparison to alternating D-A copolymers.
Chapter 2 describes the investigation of two sets of copolymers containing varying
percentages of strongly electron-donating 3-hexyloxythiophene units (3HOT). The polymers
synthesized here include both random regioregular polythiophenes and semi-random
diketopyrrolopyrrole (DPP) -containing copolymers. The influence of 3HOT content on
UV−Vis absorption, HOMO energy levels, polymer crystallinity and polymer:fullerene solar
cell performance, especially the open-circuit voltage (V
oc
), is discussed in detail. Importantly,
this study demonstrates that significant changes in polymer electronic properties can be
induced with only small percentage of 3HOT comonomers in random and semi-random
conjugated polymers.
Chapter 3 continues the investigation on the effect of 3HOT units on polymer properties,
but with another parent polymer based on alternating 2,7-carbazole and
dithienyl-2,1,3-benzothiadiazole. A 2,1,3-benzothiadiazole-based comonomer containing
3HOT units was used to replace varied amounts of dithienyl-2,1,3-benzothiadiazole and
incorporated into polymer backbones following the random D-A copolymer strategy. The
relationships between polymer composition and properties are described.
xxiv xxiv
Chapter 4 describes the preparation of a series of random DPP-functionalized regioregular
poly(3-hexylthiophene)s (rr-P3HTs) and their application as donor materials in
fullerene-based solar cells. Functionalized rr-P3HTs with varied amounts of azide side
groups were synthesized, and post-polymerization “click” reactions were used to attach DPP
derivatives onto the rr-P3HT backbones. Structure-property-performance relationships of
these polymers are analyzed, particularly in terms of UV-Vis absorption, HOMO energy
levels, polymer crystallinity and solar cell V
oc
. Due to the separated conjugation between
P3HT backbones and DPP side chains, these polymers are considered as two donor
components that are physically constrained in one phase. Comparison is made between the
binary blend solar cells based on these DPP-functionalized rr-P3HTs and the ternary blend
devices based on separate P3HT and DPP derivative donors. Dramatic differences in their
device performance shed light on the working principles of ternary blend solar cells.
Chapter 5 describes the use of post-polymerization “click” reactions in preparing
non-conjugated polymers with electroactive pendent groups. The synthetic strategy toward
functionalization of stereoregular polymers with electroactive pendents is explored. This
work demonstrates another platform of the “polymer approach” in achieving distinct
structural features of electroactive polymers.
!
1
CHAPTER 1. Random Conjugated Copolymers for Organic
Photovoltaics
1.1 Introduction
An enormous amount of research has been done to improve the power conversion
efficiency (PCE) of organic photovoltaics (OPVs) over the past few years, inspired by the
vision of a low-cost, light-weight and flexible platform that is alternative to conventional
inorganic cells to harness solar energy.
1–8
Among OPVs, bulk heterojunction (BHJ) polymer
solar cells (PSCs), which contain an interpenetrated network of conjugated polymer donor
material (p-type material for hole transport) and fullerene derivative acceptor material (n-type
material for electron transport) in the active layer between electrodes, have been studied
mostly extensively (Figure 1.1a).
1–6
Such device architectures have led to PCE of 8–9% in
single junction cells and over 10% in tandem cells, leading the state-of-the-art efficiencies of
PSCs.
9–17

A variety of approaches can be noticed as efforts to improve the efficiency of PSCs, such
as those focusing on the improvement of device architectures, the understanding of operating
mechanisms and/or the discovery of novel materials. Crucial to the breakthrough for rapid
PCE enhancement is the development of new conjugated polymers, which are used as donor
materials in the device active layer.
18–21
Considering the efficiency of organic solar cells
!
2
defined as η = (J
sc
V
oc
FF)/P
in
, where P
in

is the input power in the form of solar radiation and
the output parameters are J
sc
as the short-circuit current density, V
oc
as the open-circuit
voltage, and FF as the fill factor, realization of a high PCE is directly dependent on the
conjugated polymer properties that affect parameters J
sc
, V
oc
and FF.
1,6
Generally speaking,
J
sc
is associated with the ability of the polymer in harvesting photons, and to some extent can
be improved by lowering the polymer bandgap and broadening the absorption band. V
oc
is
correlated to the magnitude of the energetic difference between the highest occupied
molecular orbital (HOMO) of the donor and the lowest unoccupied molecular orbital (LUMO)
of the acceptor, and can be increased by lowering the HOMO of the donor conjugated
polymer.

Finally, a high FF is mainly attributed to improved active layer morphologies and
balanced charge transport.

Therefore, an ideal conjugated polymer should simultaneously
meet the requirements of a suitable absorption profile matching the photon flux of the solar
spectrum, an appropriate HOMO and LUMO energy levels, a high hole mobility for efficient
charge transport and an optimized morphology realized in their blend with fullurene
derivatives.
!
3

Figure 1.1. (a) Typical device architecture and active layer structure of bulk
heterojunction polymer solar cells. (b) Molecular structure of previously
reported PTB7 as an example of D-A alternating copolymers.
22

The development of alternating push–pull type copolymers consisting of electron-rich (D)
and electron-deficient (A) units in the polymer backbone (D–A alternating copolymers) is
one of the milestones in the exploration of conjugated polymers.
20,21,23,24
Figure 1.1b shows
the structure of previously reported PTB7 as an example of D-A alternating copolymer.
22

Following this approach, low polymer band gaps can be achieved due to electron-donating
and electron-withdrawing interaction between D and A units, resulting in an effective
absorption over the visible light range typically from 500 to 800 nm.
20,24
Top efficiencies of
PSCs mentioned in the first paragraph have been realized with some of these D-A alternating
copolymers.
However, two major drawbacks should be noted for D–A alternating copolymers, which
limits their potential to achieve even higher efficiencies.
25,26
On one hand, instead of
broadening spectral coverage, these polymers tend to produce a primary absorption band
!
4
being red shifted to longer wavelength region, while losing the coverage of higher energy
visible range from 350 to 500 nm. Therefore, the upper limit of J
sc
achievable with these
polymers is suppressed for any specific pair of polymer-fullerene blends. On the other hand,
the state-of-the-art D-A alternating copolymers commonly involve lengthy and challenging
monomer syntheses. Due to limited structural variations available in alternating polymer
backbones, the properties of polymers have to be controlled by fine-tuned manipulation of
the monomer structures, which has been referred to as the “monomer approach” in
conjugated polymer design principles. The difficulty in developing new monomer units then
causes the major bottleneck against the library expansion of well-performing polymers. In
addition, although not emphasized very often, the highly regular alternating structure of
polymer backbones may lead to strong tendency of molecular packing and hence low
solubility of the polymer samples, causing difficulties in the solution processing of active
layers.
In this regard, the family of copolymers that consists of randomized linkages among two or
more comonomer building blocks holds the potential to overcome the drawbacks of D-A
alternating copolymers mentioned above.
25,26
The so-called random or semi-random
conjugated copolymers offer a chance to tune polymer properties through strategic
combination of monomers, in a manner referred to as the “polymer approach”.
25
Through
judicious control over the comonomer combinations and molar ratios, the previously reported
D and A moieties can be fully exploited and thus delicate manipulation of polymer properties
!
5
can be achieved, especially in terms of absorption behavior, energy levels and polymer
solubility. In this chapter, the syntheses and OPV applications of random or semi-random
conjugated copolymers are briefly reviewed. The reported examples are discussed and
classified according to their synthetic strategies and/or general features of the resulting
polymers.
1.2 Regioregular poly(3-substituted thiophene)s
Regioregular poly(3-substituted thiophene)s are among the most popular types of OPV
donor materials in recent years.
27
The typical regioregular poly(3-hexylthiophene) (rr-P3HT)
consistently shows 3-5% PCE, thanks to its good film forming ability, high hole mobility and
the ability to form favorable morphology when blended with the acceptor material
PC
61
BM.
28,29
However, achieving even higher efficiencies with rr-P3HT is challenging
because the narrow absorption breadth (from 400 to 650 nm)
30
and the relatively high
HOMO level (5.2 eV)
31
inherently limit the device J
sc
and V
oc
respectively. Therefore,
well-tailored structural modifications have been made to poly(3-substituted thiophene)s by
the means of copolymerizations. Modern synthetic methods, including McCullough, GRIM,
Rieke, Stille and DArP polymerizations, which lead to regioregular poly(3-substituted
thiophene)s (>90% RR),
27,32
have also been used to produce random poly(3-substituted
thiophene)-based copolymers and/or poly(3-substituted selenophene)s that replaces the sulfur
atom in polythiophenes with selenium. (Table 1.1)
!
6
Table 1.1. Typical conditions of common polymerization methods
for preparing regioregular poly(3-substituted thiophene)s.

When these methods are employed to prepare random copolymers of regioregular
polythiophene derivatives, the synthetic strategy and hence the resulting copolymers share
the following common features:
1. The polymerizations start with at least two thiophene or selenophene derivatives,
which are functionalized unsymmetrically and in an analogous way at 2- and 5- positions.
Typical methods involve bromination at 2-position and activation with organo metal at
5-position to allow cross-coupling between monomers, except that direct arylation
polycondensation (DArP) requires no activation of the 5-position proton.
!
7
2. The monomer functionalities allow for possibly random distribution of comonomer
units along the polymer backbone, without inherent restriction, although the monomer
sequences and/or incorporated composition of monomers may be dictated by the reactivity
difference of comonomers. It is not uncommon that the polymer composition differs
significantly from the feed ratio;
3. The use of regio-specific thiophene or selenophene monomers ensures a high degree
of head-to-tail (H-T) regioregularity of the comonomer linkages;
4. The functionalized comonomers can be subject to post-polymerization reactions that
lead to further modifications of the polymer structures, and thus introduce desired features
into the polymer materials.
Among the existing examples of random copolythiophenes/copolyselenophenes, two types
of modification strategies can be recognized. On one hand, some have focused on the effect
of side chain patterns or functionalities on polymer chain packing, polymer:fullerene blend
morphology and/or hole mobilities, and hence on the solar cell performance, without
rendering significant changes in the energy levels of the polymers. On the other hand, some
other research efforts mainly aimed at tuning the polymer electronic properties, such as
bandgaps and/or energy levels, which had substantial impact on their performance in OPV.
The two classes of polymers will be discussed in Section 1.2.1 and 1.2.2, repectively.

!
8
1.2.1 Alkyl side chain effects
As mentioned above, engineering of side chain patterns or functionalities of
polythiophenes can be used to modulate polymer chain packing, and thus to influence the
polymer:fullerene blend morphology and/or hole mobility as well as the solar cell
performance. Previously reported research works following this strategy are summarized in
Table 1.2, with the polymer structures shown in Figure 1.2 (P1-P6) and Figure 1.3 (P7-P11).

Figure 1.2. Schematic syntheses and structures of
polythiophene-based copolymers with side chain engineering.
!
9
As an effort to explore the effect of polythiophene side chain length, random
poly(3-butylthiophene-co-3-octylthiophene) of several compositions (P1) were synthesized
and characterized.
33
It was found that not only the melting (T
m
) and recrystallization (T
c
)
temperatures, but also the interlayer stacking distance of the lamellar crystalline structure of
the copolymers, had a linear dependence on the composition of the random copolythiophenes.
Fullerene-based BHJ solar cells fabricated from the random copolythiophenes were
significantly enhanced compared to the homopolymers or physical blends, mainly due to the
enhanced charge transport and a more favorable nanoscale morphology of the BHJ thin films
of the random copolythiophenes. The influence of alkyl chain length was also investigated by
the Lipomi group with a focus on the polymer mechanical properties (P2).
34
However, the
random copolymer of 3-hexylthiophene and 3-octylthiophene did not beat the homopolymer
of 3-heptylthiophene in either mechanical compliance or PCE of PC
61
BM-containing solar
cells.
In 2007, the Fréchet group copolymerized 3-hexylthiophene with a regio-symmetric
thiophene monomer with n-hexyl chains at both the 3 and 4 positions (P3).
35
The resulting
polythiophenes had reduced regioregularity from >96% to 91%, which was under rational
control. Investigation revealed the influence of P3HT regioregularity on its performance in
solar cells. Particularly, the 91% regioregular copolymer-based BHJ solar cell did not suffer
in average PCE with prolonged thermal annealing as was seen with rr-P3HT, mainly due to
the formation of a thermally stable interpenetrating network of the copolymer:PCBM blends
!
10
without large-scale phase segregation. Later, they examined the effect of substituent
sequence distribution on polymer crystallinity and solar cell performance in PCBM based
BHJ devices. They compared the random copolymer poly(3-dodecylthiophene-co-thiophene)
(P4), which are composed of equal parts of 3-dodecylthiophene and unsubstituted thiophene,
with its alternating analog (P5) with identical molecular weight, composition, and electronic
structure.
36
In addition to the drop in HOMO level from -5.35 eV for P5 to -5.45 eV for P4, a
more than 3-fold increase in solar cell efficiency was observed for P4 relative to P5. The
significant enhancement in OPV performance was attributed to the ability to form
bicontinuous structure in the random copolymer, resulted from the decreased tendency for
crystallization.
The influence of branched side chains was investigated by comparing
rr-poly(3-hexylthiophene-co-3-(2-ethylhexyl)thiophene) copolymers (P6) with varied content
of branched 2-ethylhexyl side chains (10, 25, and 50%) to the corresponding homopolymers
P3HT and poly(3-(2-ethylhexyl)thiophene) (P3EHT).
37
Importantly, Polymer HOMO levels
in the solid state decreased remarkably with an increase in the amount of 2-ethylhexyl side
chains, and the decrease was directly reflected in the V
oc
measured in polymer:PC
61
BM solar
cells which increased with increasing 2-ethylhexyl side-chain content. Particularly P6-b

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
, PCE = 3.48%).
!
11

Figure 1.3. Schematic syntheses and structures of
polythiophene-based copolymers with side chain engineering
(continued).
A variety of functional groups were introduced to the side chains of polythiophenes by
Campo and coworkers (P7-P10).
38,39
They prepared a set of rr-P3HT-based random
copolymers containing varying ratios of ester functionalized alkyl side chains using the Rieke
method (P7 and P8),
38
and also converted the ester functions to alcohol and
photo-crosslinkable cinnamoyl groups by post-polymerization reactions (P9 and P10).
39

Their characterization and device performance results proved it possible to introduce a
certain amount of functionalized side chains into the polymer molecular structure and
maintain solar cell performance comparable to P3HT. Additionally, the effect of side chain
!
12
functionalities has also been explore by the Yassar group (P11).
40
They successfully
copolymerized 3-hexylthiophene with a new 3-functionalized-thiophene (propyl
5-(2-(thiophen-3-yl)ethoxy) pentanoate), bearing an ester function (P11), which produced a
moderately performing copolymer donor.
Table 1.2. Summary of polymer properties and device performance parameters of
polythiophene-based copolymers with side chain engineering.
Polymer
Composition
(m:n)
M
n
(kg/mol)/PDI
J
sc

(mA/cm
2
)
V
oc

(V)
FF
PCE
(%)
a

P1-a 20:80 77.8/1.45 7.27 0.56 0.59 2.40
P1-b 34:66 12/1.09 7.01 0.55 0.54 2.10
P1-c 50:50 15/1.36 7.92 0.56 0.57 3.00
P1-d 61:39 9.7/1.18 7.00 0.62 0.65 2.80
P1-e 72:28 8.3/1.13 6.22 0.64 0.62 2.50
P2 50:50 17/1.55 5.8 0.549 0.47 1.50
P3 96:4 22/1.5 10.3 0.606 0.600 3.74
P4 50:50 19.4/1.99 5.77 0.68 0.47 1.84
P5 N/A 19.8/2.11 2.78 0.59 0.33 0.54
P6-a 90:10 21.3/2.5 9.26 0.63 0.51 2.80
P6-b 75:25 26.2/2.5 9.85 0.69 0.57 3.85
P6-c 50:50 40.1/2.0 2.52 0.85 0.35 0.74
P7-a 90:10 24.7/1.9 7.43 0.54 0.62 2.49
P7-b 70:30 21.6/2.1 6.40 0.56 0.55 1.98
P7-c 50:50 17.3/1.5 6.25 0.56 0.57 1.97
P7-d 95:5 17.3/1.53 10.2 0.58 0.62 3.69
P8-a 90:10 30/1.9 7.85 0.50 0.62 2.43
P8-b 70:30 39.3/2.3 6.60 0.52 0.62 2.13
P8-c 50:50 44.5/2.7 5.11 0.50 0.63 1.61
P8-d 95:5 17.9/1.62 7.97 0.54 0.62 2.67
P9-a 90:10 --
b
7.17 0.5 0.57 2.04
P9-b 70:30 --
b
5.01 0.47 0.44 1.14
P9-c 50:50 --
b
3.43 0.44 0.3 0.47
P9-d 95:5 19.2/1.76 8.72 0.56 0.63 3.07
P10 95:5 18.7/1.69 9.33 0.56 0.64 3.34
P11 64:36 16.6/1.12 4.31 0.66 0.44 1.20
a
Device performance was measured using PC
61
BM as acceptor material.
b
No data was
provided in literature.
!
13
1.2.2 Non-alkyl thiophene copolymers
The replacement of alkyl-substituted thiophenes with other functional thiophene or
selenophene derivatives may lead to more significant tuning of the electronic properties of
polythiophenes. (P12-P18 as shown in Figure 1.4 and Table 1.3) For example, Kozycz and
others reported a conjugated statistical copolymer,
poly(3-hexylthiophene)-stat-(3-thiohexylthiophene) (P3HT-s-P3THT), with
3-hexylthiophene:3-thiohexylthiophene monomer ratios ranging from 50:50 to 99:1 (P12).
41

When testing the copolymers in bulk heterojunction devices with a fullerene-derivative
(PC
71
BM) acceptor, the copolymers have an 11−18% increase in the open-circuit voltage (V
oc
)
relative to the P3HT:PC
71
BM device due to the deeper HOMO level of the
3-thiohexylthiophene unit. This increase was independent of copolymer composition over the
50:50 to 85:15 range and was still observed when statistical copolymers contained as low as
1% of 3-thiohexylthiophene unit in the polymer chain.
!
14

Figure 1.4. Schematic syntheses and structures of polythiophene-based copolymers with
electronically distinct comonomer units.
!
15
Highly electron-donating 3-alkyloxythiophenes have been incorporated into regioregular
poly(3-substituted thiophene) copolymers as well. The copolymer of 3-octylthiophene and
3-decyloxythiophene POT-co-DOT (P13, Figure 1.4a) was synthesized by GRIM in 2006.
42

The copolymer had higher oxidation potential as well as better film-forming ability of the
PCBM blends compared to the homopolymer poly(3-decyloxythiophene) P3DOT, and thus
lead to a substantial improvement in solar cell performance from a PCE of 0.0007% based on
P3DOT to 0.054%. However, only one monomer composition was tested, and it differed
significantly from the feed ratio 1:1. Similar copolymers containing 3-hexylthiophene and
3-hexyloxythiophene (3HOT) at varied ratios were synthesized by the Thompson group (P14,
Figure 1.4a).
43
Stille polymerizations resulted in copolymers with the same compositions as
feed ratios. Bandgaps and HOMO levels, as well as semi-crystallinity were observed to
change regularly with increasing amount of 3HOT, due to its strong electron-donating nature.
The results proved that the properties of random copolymers can be readily tuned by varying
the comonomer compositions.
Low band-gap polythiophene derivatives, with bulky conjugated side-chains composed of
the triphenylamine, thiophene, and vinylene groups (TPATh), were synthesized by Hsing-Ju
Wang and coworkers.
44
The regioregular copolymer of 3-hexylthiophene and the side-chain
conjugated comonomer was synthesized by GRIM polymerization (P15). Its regio-random
analogue (P15-r) was synthesized via a different strategy that will be discussed in detail in
Section 1.3.2. Low HOMO levels of -5.38 eV was obtained for P15, and -5.35 eV for P15-r,
!
16
due to the effect of bulky conjugated moiety curtailed on effective conjugation length in the
main chain. Additionally, side-chain conjugated structures can be used to introduce
intramolecular donor–acceptor interactions. For example, conjugated
octylphenanthrenyl-imidazole moieties were covalently attached to the 3-position of
thiophenes, and the intramolecular donor–acceptor thiophene derivative was copolymerized
with 3-hexylthiophene by Chang and coworkers (P16).
45,46
The copolymer exhibited lowered
bandgaps and enhanced electron transfer compared to the parent polymer, which lead to
much higher short-circuit current densities relative to those of pure rr-P3HT. The
short-circuit current density of a device prepared from a copolymer containing 90 mol%
octylphenanthrenyl-imidazole moieties was 13.7 mA/cm
2
which was a 65% increase
compared to the 8.3 mA/cm
2
observed for a device based on rr-P3HT. The maximum power
conversion efficiency of this particular copolymer reached 3.45%.
Poly(3-heptylselenophene)-stat-poly(3-hexylthiophene) (P17) was synthesized and
characterized in terms of its crystallinity and performance in an OPV cell.
47
Despite the
random distribution of units along the polymer main chain, the material is semi-crystalline,
as demonstrated by differential scanning calorimetry and wide-angle X-ray diffraction. The
optoelectronic properties of P17, such as solution and thin-film absorption and EQE response,
behave as the average of the two constituent homopolymers. OPV performance is enhanced
by a broadened absorption profile and a favorable morphology and reaches a peak
performance of 3.27% PCE.
!
17
Table 1.3. Summary of polymer properties and solar cell performance parameters of
polythiophene-based copolymers with electronically distinct comonmer units.
Polymer
Composition
(m:n)
M
n
(kg/mol)
/PDI
E
g

(eV)
HOMO
(V)
J
sc

(mA/cm
2
)
V
oc

(V)
FF
PCE

(%)
P12-a
a
99:1 17/1.8 1.90 -5.09 9.49 0.62 0.6315 3.72
P12-b
a
97:3 19/1.7 1.90 -5.57 8.04 0.62 0.6334 3.16
P12-c
a
85:15 10/1.5 1.90 -5.43 5.88 0.66 0.5680 2.21
P12-d
a
80:20 13/1.7 1.90 -5.46 5.45 0.66 0.5153 1.86
P12-e
a
67:33 9.0/1.8 1.90 -5.60 2.97 0.66 0.4348 0.85
P12-f
a
50:50 9.0/1.5 1.85 -5.45 3.50 0.66 0.4197 0.97
P13
b
1:2.5 14.5/1.75 1.64 -4.55 0.60 0.22 0.412 0.05
P14-a
b
90:10 19.5/2.11 1.85 -5.13 5.28 0.43 0.41 1.04
P14-b
b
75:25 19.4/2.56 1.77 -5.07 3.62 0.37 0.45 0.60
P14-c
b
50:50 11.9/3.00 1.70 -4.96 1.34 0.21 0.41 0.11
P15
b
2.04:1 7.8/1.7 1.90 -5.38 2.21 0.66 0.37 0.49
P15-r
b
1.05:1 5.1/2.3 1.82 -5.35 6.03 0.69 0.35 1.45
P16-a 80:20 11.0/1.56 1.88 -4.70 --
d
--
d
--
d
--
d
P16-b
b
60:40 10.5/1.44 1.86 -4.70 10.1 0.64 0.374 2.42
P16-c
b
40:60 9.8/1.46 1.85 -4.65 11.7 0.66 0.342 2.63
P16-d
b
20:80 9.0/1.54 1.83 -4.65 13.0 0.67 0.331 2.85
P16-e
b
10:90 8.9/1.56 1.8 -4.65 13.7 0.68 0.372 3.45
P17
c
48:52 24.5/1.09 1.65 --
e
8.08 0.70 0.599 3.27
P18-a
b
95:5 11/1.9 1.90 -5.30 7.02 0.72 0.58 2.96
P18-b
b
90:10 12/2.4 1.88 -5.31 8.16 0.75 0.55 3.33
P18-c
b
85:15 10/2.1 1.90 -5.34 7.56 0.81 0.55 3.28
P18-d 80:20 14/2.1 1.87 -5.31 --
d
--
d
--
d
--
d

a
Device performance was measured using PC
71
BM as acceptor material.
b
Device
performance was measured using PC
61
BM as acceptor material.
c
Device performance was
measured using ICBA as acceptor material.
d
No data provided in literature due to poor
solubility of polymers. e. No data provided in literature.
Random P3HT-based copolymers containing 5–20% of 3-cyanothiophene （P18）were
synthesized to investigate the influence of this electron-deficient thiophene derivative
unit.
48,49
All copolymers had decreased HOMO levels compared to P3HT by over 0.1 eV.
Enhanced V
oc
that correlated well with the content of 3-cyanothiophene were observed with a
PC
61
BM acceptor in organic solar cells. With 20% of the comonomer baring cyano
!
18
substituents, the V
oc
reached 0.81 V, more than 0.2 V higher than that of P3HT. Due to the
benefits of increased V
oc
, the PCE of P18-b and P18-c based solar cells exceeded that of
P3HT. Later, the same copolymers were prepared by DArP as well.
49
Comparable results
were obtained from characterizations and BHJ solar cells with PC
61
BM as acceptor.
1.3 A-A type bridging unit plus two or more B-B type comonomers
Among the strategies to produce random copolymers, it is most popular to use the
combination of three (or more) comonomers with A-A/B-B type of functionalities, as
depicted in Scheme Figure 1.5. When three comonomers are used, this strategy can be
considered as adding a third (or more) component into the perfectly alternating copolymer
backbone, and this concept has been emphasized in the recent review paper from the Kim
group.
26
In the discussion here, this category of random copolymers will be described from a
different point of view.

Figure 1.5. General scheme of the synthetic route to prepare random conjugated copolymers
using one A-A type bridging unit and two B-B type comonomer units, where A and B in the
scheme represents complementary functionalities.
Generally speaking, this synthetic strategy and hence the resulting copolymers share the
following common features:
!
19
1. The polymerizations start with three or more comonomers. Among them,
Comonomer 1 (M1) has functionality A on both ends, and the others (M2, M3 and so on)
have complementary functionality B on both ends of each;
2. The monomer functionalities allow for cross-coupling between M1 and any other
comonomer, but no cross-coupling can occur without M1;
3. M1 will constitute 50% of the building blocks in the resulting polymer backbones,
perfectly alternating with the other comonomers (M2, M3 and so on), while the distribution
of the other comonomers follow a randomized pattern, which is influenced to some extent by
the relative rate of the cross-coupling reactions;
4. In the ideal case where the conversion of all functionalities approaches 1, the
composition of polymer backbones will be the same as feed ratio. However, due to the
limited reactivity of some cross-coupling reactions, the real composition may deviate from
feed ratio, but M1 always makes up 50% of the polymer backbones;
5. The polymerization reaction is not regio-selective. That being said, if any of the
comonomer is unsymmetrical, it is not possible to achieve regioregular copolymers using this
strategy.
As the comonomers with functionality B can be of unlimited number, it is actually not a
comprehensive discussion of Kim and coworkers to cover terpolymers only. In fact, the
copolymers made with this strategy can be better considered as an alternating copolymer
!
20
with one alternating unit being M1 and the other unit being random among M2, M3 and/or
others. In terms of specific polymerization methods, Stille and Suzuki polymerizations are
most widely employed in this strategy, utilizing the combination of stannyl groups and
bromines, and boronic esters and bromines, respectively. In addition, DArP is also available
for this strategy, using bromines and unactivated hydrogens as the complementary
“functionality”.
1.3.1 Special cases with M1 and M2 sharing the same core structure
In a special case, depicted in Figure 1.6 and 1.7, M1 and M2 can share the same core
structure (M*, highlighted with yellow backgrounds in Figure 1.6 and 1.7), but have
complementary functionalities just as described in the general picture. As such, after
elimination of functional groups during cross-couplings, the building blocks coming from
M1 and M2 will be the same (M*) and undistinguishable. Therefore, the resulting
copolymers will appear to have M* segments of different length (repeating units being 1 or
more) interrupted by discrete comonomer M3 (or others). So far, this special case has mostly
been seen for polyfluorene-based copolymers, shown in Figure 1.6 (P19-P27).
!
21

Figure 1.6. Schematic syntheses and structures of polyfluorene-based copolymers
following the strategy described in Section 1.3.1.
A series of copolymers between 9,9-dioctylfluorene and
4,7-di-2-thienyl-2,1,3-benzothia-diazole were reported by the Cao group with the content of
4,7-di-2-thienyl-2,1,3-benzothia-diazole (DBT) unit ranging from 5% to 35% (P19).
50,51

Compared to the homopolymer of fluorenes PFO, the copolymers show a new peak in
UV-Vis absorption at longer wavelength, even further red-shifted compared to the peak from
alternating copolymer PFO-DBT,
52
and this peak increase linearly with increasing content of
DBT. At the same time, the incorporation of DBT raised the V
oc
significantly. Therefore, the
!
22
copolymer containing 35% DBT benefits from both increased J
sc
and V
oc
compared to PFO
and reaches a PCE of 2%, which is almost two orders of magnitude higher than PFO and
comparable to the alternating copolymer PFO-DBT. The same group later reported similar
copolymers P20-P22 but replacing selected sulfur atoms in P19 by seleniums,
53,54
as well as
another copolymer P23 derived from 9,9-dioctylfluorene (DOF) and
2,3-dimethyl-5,7-dithien-2-yl-thieno[3,4-b]pyrazine (DTTP).
55
The actual ratios of
substituted fluorene to the other comonomer in the copolymers were estimated by elemental
analysis and proved to be in good agreement with the feed ratios. These copolymers exhibit
interesting photoluminescence and electroluminescence properties, but only perform
moderately in OPV devices.
Schulz and coworkers tried to incorporate Iridium complexes into the random copolymer
backbone in order to take advantage of charge generation from triplets in photovoltaic
devices.
56
Fluorene-based copolymer containing 2-phenylpyridine-coordinated iridium
complex P25 was synthesized, as well as the one containing only 2-phenylpyridine P24 as a
control experiment. To be noted, X-ray fluorescence (XRF) spectroscopy determined the
actual amount of iridium complex incorporated was 13 mol % among all building blocks of
polymer P25 backbone, instead of the 25% feed ratio. No experimental data was reported
about actual composition of P24. Characteristic absorption and phosphorescence of the
iridium complex can be observed for P25, and this phosphorescence is completely quenched
upon blending PCBM into the film. Due to the contribution from the iridium complex triplet,
!
23
the OPV devices based on P25 clearly exhibits an improved device performance relative to
P24, with a J
sc

of 0.44 mA/cm
2
and PCE of 0.07% for P25 in comparison to the J
sc
of 0.01
mA/cm
2
and PCE of 0.002% for P24.
Park and coworkers copolymerized fluorenes with varied percentage of
4,7-bis(3′,3′-dihepyl-3,4-propylenedioxythienyl)-2,1,3-benzothiadiazole (PTBT), resulting in
the copolymer series P26.
57
Similar to P19-P23, this class of copolymers also shows two
peaks in UV-Vis absorption whose intensities are influenced by the copolymer composition.
At the ratio 65:35 of fluorene to PTBT, the copolymer leads to the best performing OPV
device within this series, with a J
sc
of 1.55 mA/cm
2
, a V
oc
of 0.75 V, a FF of 0.31, and a PCE
of 0.36%.
In 2008, a set of fluorene-based copolymers incorporating different
phenothiazine-arylcyanovinyl units (P27-1 to P27-6) were reported.
58
Some of these
copolymers contain unsymmetrical building blocks and therefore are regio-random. This set
of copolymers offered a screening of several similar comonomers and proved that the
bandgaps and energy levels could be tuned by introducing small variation in comonomer
strucutures. However, no solar cell device performance data were reported with these
copolymers.

!
24
Table 1.4. Summary of polymer properties and device performance parameters of
polyfluorene-based copolymers following the strategy described in Section 1.3.1.
Polymer
Compo-
sition
(m:n)
M
n
(kg/mol)
/PDI
E
g

(eV)
HOMO
(V)
J
sc

(mA/cm
2
)
V
oc

(V)
FF
PCE
(%)
a

P19-a 45:5 35/3.1

2.08 -5.65 0.12 0.65 0.339 0.035
P19-b 40:10 33/2.7

2.07 -5.61 0.18 0.70 0.343 0.056
P19-c 35:15 34/2.6

2.03 -5.55 0.29 0.80 0.218 0.064
P19-d 25:25 29/2.4

2.02 -5.50 1.20 0.80 0.292 0.35
P19-e 15:35 11/1.7

2.01 -5.47 4.59 0.95 0.35 1.95
P20-a 49:1 21/1.60 2.92 -5.70 --
b
--
b
--
b
--
b

P20-b 45:5 28/1.43 1.88 -5.70 --
b
--
b
--
b
--
b

P20-c 40:10 25/2.01 1.88 -5.68 --
b
--
b
--
b
--
b

P20-d 35:15 24/1.59 1.87 -5.60 --
b
--
b
--
b
--
b

P20-e 20:30 12/1.78 1.85 -5.51 2.53 1.00 0.374 1.00
P21-a 49:1 23/1.86 2.92 -5.73 --
b
--
b
--
b
--
b

P21-b 45:5 20/1.83 1.78 -5.71 --
b
--
b
--
b
--
b

P21-c 40:10 28/1.95 1.77 -5.71 --
b
--
b
--
b
--
b

P21-d 35:15 13/1.72 1.78 -5.53 0.51 0.80 0.233 0.10
P22 15:35 --
b
1.77 --
b
2.53 0.85 0.327 0.91
P23-a 49:1 24.6/3.1 2.00 -5.71 0.14 0.55 0.3201 0.03
P23-b 45:5 27.5/2.4 1.88 -5.70 0.45 0.55 0.3479 0.09
P23-c 35:15 19.7/2.2 1.80 -5.68 0.65 0.55 0.2762 0.10
P23-d 20:30 9.5/1.9 1.77 -5.22 4.10 0.70 0.2891 0.83
P24 --
b
16.3/2.0 --
b
-5.73 0.01 0.6 0.24 0.002
P25 37:13 8.7/1.4 --
b
-5.68 0.44 0.63 0.19 0.07
P26-a 40:10 22.1/1.9 1.88 -5.39 0.18 0.55 0.28 0.028
P26-b 34:16 6.6/2.3 1.86 -5.43 0.32 0.38 0.28 0.034
P26-c 15:35 6.7/1.5 1.73 -5.16 1.55 0.75 0.31 0.36
P27-1 25:25 13.4/2.00 2.06 -5.38 --
b
--
b
--
b
--
b

P27-2 25:25 12.9/1.45 1.77 -5.29 --
b
--
b
--
b
--
b

P27-3 25:25 20.7/2.40 2.10 -5.40 --
b
--
b
--
b
--
b

P27-4 25:25 22.3/2.65 1.79 -5.31 --
b
--
b
--
b
--
b

P27-5 25:25 28.0/3.68 1.73 -5.32 --
b
--
b
--
b
--
b

P27-6 25:25 18.5/1.63 1.55 -5.16 --
b
--
b
--
b
--
b

a
Device performance was measured using PC
61
BM as acceptor material.
b
No data provided
in literature.
!
25
In addition to the polyfluorene-based copolymers, other examples that fit in the description
of this special category are shown in Figure 1.7 (P28-P29), and the properties and device
performance data are summarized in Table 1.5.

Figure 1.7. Syntheses and structures of other copolymers following the
strategy described in Section 1.3.1.
D-A copolymers composed of varying !amounts of benzodithiophene (BDT) donor with the
thienoisoindoledione !(TID) acceptor (P28) were synthesized by Stille polymerizations to
investigate the influence of backbone composition on the polymer physical ! properties.
59

!
26
Alternating copolymers featuring the same donor and acceptor ratio in backbones were
synthesized as well, but through a much more complex and challenging route. It was found
that the absorption spectra of the alternating copolymers are significantly more red-shifted
compared to random copolymers with analogous donor:acceptor ratios. But the random
copolymers outperformed well-defined alternating copolymers in OPV devices.
Bui and coworkers recently reported a series of donor-rich copolymers with various
benzodithiophene (BDT, donor unit)/fluorinated-thienothiophene (TT-F, acceptor unit) ratios
(P29).
60
The perfectly alternating copolymer with these two building blocks was also
prepared for comparison. The optical and electrochemical properties of these polymers
were tuned with the amount of donor units in their polymer backbones. Although regular
small-area solar cells based on these random copolymers did not exceed that of the
alternating copolymer, which had a PCE of 7.63%, it was found that the device performance
of P29-c was less dependent on the thickness of the active layer than those of other polymers.
A large area (6270 mm
2
) P29-c-based device exhibit a best PCE of 6.54%, which is
significantly better than that of the equivalent alternating-copolymer-based large area device
(4.03%).

!
27
Table 1.5. Summary of polymer properties and device performance parameters of other
copolymers following the strategy described in Section 1.3.1.
Polymer
Composition
(m:n)
M
n
(kg/mol)
/PDI
E
g

(eV)
HOMO
(V)
J
sc

(mA/
cm
2
)
V
oc

(V)
FF
PCE
(%)
P28-a

1:2 21/2.9 1.50 -5.4 --
d
--
d
--
d
--
d

P28-b
a
1:1 20/4.3 1.54 -5.4 6.8 0.78 0.52 2.7
P28-c
a
2:1 19/3.5 1.64 -5.4 6.2 0.84 0.38 1.9
P28-d
a
3:1 17/4.2 1.70 -5.4 6.0 0.86 0.37 1.9
P28-e

6:1 17/3.9 1.80 -5.4 --
d
--
d
--
d
--
d

P29-a
b
1.08:1 39.6/2.70 1.65 -5.64 11.69 0.86 0.7 7.04
P29-b
b
2.42:1 27.2/2.09 1.68 -5.66 11.22 0.88 0.71 7.02
P29-c
b
4.15:1 14.5/5.55 1.74 -5.67 10.95 0.89 0.71 6.92
P29-c
b,c
4.15:1 14.5/5.55 1.74 -5.67 9.40 1.14 0.61 6.54
P29-d
b
5.95:1 9.2/3.98 1.78 -5.67 10.12 0.89 0.71 6.45
a
Device performance was measured using PC
61
BM as acceptor material.
b
Device
performance was measured using PC
71
BM as acceptor material.
c
Performance data were
measured on large area device (6270 mm
2
). d. No data was provided in literature.
1.3.2 Thiophene or ethylene as bridging units in Stille polymerizations
Following the strategy described at the beginning of Section 1.3, a massive amount of
research has been reported, exploring different combinations of comonomers and varied
ratios between M2 and M3. Among them, a number of works have involved the use of
bis-stannylated thiophene or bis-stannylated ethylene as comonomer M1, which was
copolymerized with dibrominated comonomers M2 and M3 (or others), as depicted in Figure
1.8.
!
28

Figure 1.8. General schematic synthesis of copolymers with thiophene or ethylene
bridging units via Stille polymerization.
The bis-stannylated thiophenes or ethylenes are not only easily accessible, but also capable
of maintaining planarity of polymer backbones, and thus leading to low bandgap polymers.
Comonomers M2 and M3 are typically a pair of push-pull donor and acceptor, as shown in
Figure 1.9-1.11 and summarized in Table 1.6 and 1.7 (P30-P43). As such, three distinct
advantages can be achieved. First, the use of thiophene or ethylene linkers result in
randomized distributions of M2 and M3, as well as the opportunity to vary the ratio between
the two, which differs the resulting polymers from perfectly alternating copolymers. Second,
as thiophene or ethylene units are weakly electron-donating, a π−π* peak in UV-Vis
absorption can be expected from their linkage with the donor unit M2, and an ICT peak with
the acceptor unit M3, which contributes to a wide coverage of the solar spectrum. Third, this
synthetic route also avoids the challenges of converting aromatic units M2 or M3 into
bis-stannylated or bis-boronic ester functionalized comonomer structures as what is needed
in preparing alternating D-A copolymers via typical Stille or Suzuki polymerizations.
!
29

Figure 1.9. Schematic syntheses and comonomer structures for copolymers
with thiophene bridging units. Composition being m:n = 1:1 unless
specified.
The Ting group used a fused thiophene/phenylene/thiophene (TPT) derivative as the donor
comonomer and coplymerized it with six different acceptor units respectively in the presence
of bis-stannylated thiophenes as the linker bridge to achieve low bandgap conjugated
polymers P30-1 to P30-6.
61
By incorporating various electron-withdrawing comonomers, the
obtained TPT-based donor/acceptor copolymers exbihited optical bandgaps ranging from 1.0
!
30
to 1.8 eV, which followed a decreasing trend in the values of the bandgaps in in agreement
with order of the acceptor strength. When the polymers were blended with PC
71
BM in solar
cell devices, copolymer P30-2 reached the highest power conversion efficiency of 4.3%.
Actually, this polymer had been reported in an earlier paper,
62
together with the comparison
between copolymer P30-biTh, where thiophene linkers were replaced by bithiophenes.
Photophysical studies reveal bandgaps of 1.76 eV for P30-biTh, which is slightly larger than
the 1.70 eV for P30-2, indicating the influence on the electronic properties of copolymers
from aromatic linker structures.
Similarly, as shown in Figure 1.10, with the bridging thiophenes, the
4H-cyclopentadithiophene (CPDT) unit was used as donor in the copolymerization with
benzothiadiazole (BT), quinoxaline (QU), dithienoquinoxaline, or thienopyrazine (TP)
acceptor units respectively to affored copolymers P31-1 to P31-4.
63
Optical band gaps can be
well adjusted between 1.2 and 1.8 eV according to the relative quinoid form population of
the acceptor units in a decreasing order for P31-4, P31-1 and P31-2. Compared with the
perfectly alternating version, random copolymer P31-1 showed a more blue-shifted, broader
absorption spectrum and a higher PCE of 2.0%. P32-1 and P32-2 was also prepared as
structural analogs to P31-1 and P31-2 except for using the dithienosilole (DTS) unit as the
donor. It was found that the two bridged bithiophene units, CPDT and DTS, have similar
steric and electronic effects on the P31-1 and P32-1 as well as P31-2 and P32-2, respectively,
leading to comparable intrinsic properties and device performances.
!
31

Figure 1.10. Comonomer structures for copolymers with thiophene bridging
units (continued) synthesized via the schematic Stille polymerization method
depicted in Figure 1.9. Composition being m:n = 1:1 unless specified.
Li and coworkers synthesized a polymer with the electron acceptor diketopyrrolopyrrole
(DPP) unit linked with the 3,5-dialkyldithienothiophene (DTT) unit by using thiophene as a
bridge in 2011 (P33).
64
This donor–acceptor randomly incorporated copolymer performed
well OPV devices, with a PCE as high as 5.02% when PC
71
BM was used as acceptor. Two
!
32
years later, the same group expanded the scope of their polymer structures to a series
of DPP/DTT based donor–acceptor random copolymers (P34-1 to P34-4).
65
In this work, the
solution processability of the copolymers was modified through the attachment of different
alkyl chains to the lactam N-atoms of the DPP moiety. More importantly, the optical and
electronic properties were tuned by changing the donor–acceptor ratio in the backbone.
Under optimized donor-acceptor ratio of 2:1 for P33, although the hole mobility was at least
one order of magnitude lower than that of alternating copolymer P34-3-alt, random
copolymer P33 provided a higher PCE of 5.02% when blended with PC
71
BM in OPV
devices compared to the 3.44% achieved with P34-3-alt.
The copolymerization of a diketone derivative as acceptor unit in the currently discussed
strategy was reported by Zhao and coworkers, as depicted in Scheme (P35).
66
Additionally, a
maleimide derivative was explored by Chan and others as acceptor unit in the combination
with different fused thiophene donor units in random D-A copolymers in 2012, using
thiophene as the bridging unit (P35-1 and P35-2).
67
The results from Chan’s paper
demonstrated the enhancement in coplanarity and effective conjugated length when fused
thiophene rings were incorporated. However, the copolymers only performed moderately in
OPV devices.

!
33
Table 1.6. Polymer properties and device performance parameters of copolymers shown in
Figure 1.9 and 1.10.
Polymer
Compo-
sition
(m:n)
M
n
(kg/mol)
/PDI
E
g

(eV)
HOMO
(V)
J
sc

(mA/
cm
2
)
V
oc

(V)
FF
PCE

(%)
P30-1
a
1:1 41.1/1.52 1.80 -5.30 9.0 0.81 0.57 4.20
P30-2
a
1:1 22.2/1.74 1.70 -5.30
c
10.1 0.80 0.53 4.30
P30-biTh
a
1:1 16.9/1.55 1.76 -5.46 8.7 0.84 0.53 3.90
P30-3
a
1:1 28.5/1.82 1.40 -5.25 10.3 0.75 0.54 4.20
P30-4
a
1:1 20.1/1.61 1.20 -5.03 1.94 0.51 0.41 0.40
P30-5
a
1:1 18.2/2.02 1.20 -5.12 3.6 0.54 0.43 0.84
P30-6
a
1:1 16.1/1.45 1.00 -5.16 2.57 0.54 0.36 0.50
P31-1
a
1:1 26/2.50 1.59 -5.17 9.8 0.58 0.36 2.00
P31-2
a
1:1 27/2.15 1.77 -5.05 1.4 0.62 0.33 0.30
P31-3
a
1:1 19/2.00 1.82 -5.13 2.1 0.63 0.41 0.70
P31-4
a
1:1 36/2.17 1.2 -5.02 3.5 0.44 0.38 0.60
P32-1
a
1:1 22/1.73 1.57 -5.15 9.6 0.51 0.45 2.20
P32-2
a
1:1 44/2.18 1.79 -5.07 1.1 0.64 0.4 0.30
P33
a
1:1 52.4/3.4 1.37 -5.14 12.76 0.584 0.6727 5.02
P34-1-a
b
1:1 15.7/3.4 1.37 -5.25 8.71 0.480 0.4933 2.06
P34-1-b
b
2:1 21.7/3.2 1.37 -5.24 10.93 0.536 0.6100 3.58
P34-1-c
b
3:1 25.4/2.1 1.37 -5.19 10.17 0.536 0.5621 3.06
P34-2
a
2:1 17.7/3.6 1.37 -5.15 13.75 0.548 0.5033 3.79
P34-3
a
2:1 15.4/3.0 1.37 -5.15 12.76 0.584 0.6727 5.02
P34-3-alt
a
N/A --
d
--
d
--
d
9.91 0.53 0.6565 3.44
P34-4
a
2:1 29.1/2.3 1.37 -5.15 3.24 0.593 0.6760 1.30
P35-a
b
5:1 11.6/1.55 1.97 -5.05 0.43 0.53 0.24 0.11
P35-b
b
5:2 8.5/1.4 1.97 -5.06 1.02 0.62 0.26 0.33
P35-c
b
1:1 --
e
--
d
-5.05 0.72 0.54 0.33 0.26
P36-1
a
1:1.14 11/4.30 2.09 -5.67 7.9 0.70 0.20 1.08
P36-2
a
1:1.27 8.0/1.87 2.08 -5.73 7.4 0.74 0.22 1.20
a
Device performance was measured using PC
71
BM as acceptor material.
b
Device
performance was measured using PC
61
BM as acceptor material.
c
HOMO level was reported
as -5.43 eV in Ref 62.
d
No data provided in literature.
e
No data provided due to poor
solubility of polymer.

!
34

Figure 1.11. Comonomer structures for copolymers with thiophene bridging
units (continued) synthesized via the schematic Stille polymerization method
depicted in Figure 1.9.
A novel ladder-type interfused TPTPT unit was explored in P37-2 and P37-3, where each
thiophene ring is covalently fastened with the adjacent benzene ring by a carbon bridge,
!
35
embedding four cyclopentadiene rings in the nonacyclic structure.
68
The pentacyclic DIDT
unit was copolymerized to prepared P37-1 as a comparison in an earlier report.
69
The optical
band gaps and hole mobilities were modulated by changing either the comonomer structures
or the donor-acceptor ratios, which was reflected in device performance. Blends of
P37-3-b/PC
71
BM worked as the most efficient active layer among these copolymers, due to
its broad absorption ability and high hole mobility, reaching a J
sc

of 10.78 mA/cm
2
and an
impressive PCE of 4.3%.
Donor unit 3-(hexyloxy)thiophene was copolymerized with 2,2-dihexyl-2H-benzimidazole
(HBI) in the presence of thiophene bridges (P38).
70
The spectra of the solid films show
absorption bands with maximum peaks at about 408–526 nm and the absorption onsets at
550–692 nm, corresponding to band gaps of 1.79–2.25 eV. The onset wavelengths of the
absorption spectra in thin films exhibit a gradual blue-shift with increasing amount of
dihexyl-2H-benzimidazol unit, from 692 nm with P38-a to 550 nm with P38-e. However,
solar cell devices based on these copolymers did not perform well. Similar
dialkyl-2H-benzimidazole acceptor unit was used in combination with carbazole and fluorene,
respectively to prepared P39 and P40, with thiophene bridges by the same group in following
works.
71,72
Their analogs with bithiophene bridges instead of thiophenes were synthesized as
well (P39-biTh and P40-biTh). Both sets of polymers showed a red-shift in UV-Vis
absorption peaks when thiophenes were replaced by bithiophenes.

!
36
Table 1.7. Polymer properties and device performance parameters of copolymers shown in
Figure 1.11.
Polymer
Compo-
sition
(m:n)
M
n
(kg/mol)
/PDI
E
g

(eV)
HOMO
(V)
J
sc

(mA/
cm
2
)
V
oc

(V)
FF
PCE

(%)
P37-1
a
1:1 11.0/1.36 1.73 -5.35 6.2 0.68 0.47 2.00
P37-2-a
a
1:1 38.6/2.46 1.78 -5.16

7.8 0.72 0.49 2.80
P37-2-b
a
1:2 16.5/2.29 1.78 -5.18 8.34 0.74 0.50 3.10
P37-3-a
a
1:1 30.1/2.16 1.50 -5.16 9.64 0.70 0.54 3.60
P37-3-b
a
1:2 12.9/2.19 1.44 -5.21 10.78 0.68 0.58 4.30
P37-3-c
a
1:3 --
c
1.44 -5.24 9.27 0.68 0.65 4.10
P38-a
b
34:66 7.0/2.02 2.25 -5.34 0.72 0.34 0.32 0.08
P38-b
b
48:52 7.6/1.79 2.08 -5.26 0.58 0.29 0.34 0.06
P38-c
b
59:41 8.1/2.02 1.94 -5.23 1.05 0.36 0.38 0.14
P38-d
b
77:23 5.0/2.01 1.85 -5.15 1.20 0.36 0.37 0.16
P38-e
b
93:7 3.0/1.64 1.79 -4.93 1.23 0.16 0.42 0.08
P39-a
b
7:3 10.7/4.1 1.53 -5.60 1.50 0.40 0.33 0.20
P39-b
b
1:1 11.4/2.3 1.49 -5.43 1.26 0.47 0.40 0.23
P39-c
b
3:7 10.0/2.4 1.48 -5.41 0.98 0.30 0.30 0.09
P39-biTh-a
b
7:3 5.8/3.0 1.56 -5.44 4.31 0.60 0.35 0.91
P39-biTh-b
b
1:1 4.8/1.2 1.53 -5.39 2.76 0.40 0.33 0.37
P39-biTh-c
b
3:7 4.6/1.2 1.52 -5.40 1.92 0.54 0.36 0.37
P40-a
b
80:20 15/3.6 1.57 -5.66 0.69 0.66 0.29 0.13
P40-b
b
77:23 16.8/2.6 1.54 -5.64 0.73 0.63 0.31 0.14
P40-c
b
57:43 9.2/2.0 1.53 -5.47 0.65 0.52 0.28 0.09
P40-biTh-a
b
80:20 83/2.1 1.60 -5.56 2.61 0.69 0.35 0.63
P40-biTh-b
b
71:29 102/3.7 1.58 -5.39 3.10 0.68 0.35 0.74
P40-biTh-c
b
58:52 6.9/2.9 1.57 -5.24 2.60 0.61 0.31 0.49
P41-a
b
1:1.55 7.4/1.65 1.70 -4.95 4.55 0.46 0.39 0.82
P41-b
b
1:1.33 8.5/1.72

1.64 -5.00 6.83 0.54 0.44 1.63
P42-a
b
3:1 21.3/2.11 1.76 -5.36 3.559 0.831 0.372 1.10
P42-b
b
2:1 19,6/1.77 1.75 -5.36 6.46 0.791 0.376 1.90
P42-c
b
1:1 18.5/1.92 1.71 -5.30 2.625 0.797 0.338 0.70
P42-d
b
1:2 15.3/2.06 1.71 -5.26 1.68 0.626 0.264 0.30
P42-e
b
1:3 14.9/2.39 1.66 -5.28 1.068 0.413 0.213 0.10
P42-alt
b
N/A 18.8/1.95 1.78 -5.38 7.93 0.800 0.365 2.30
P43-a
b
1.82:1 14.0/3.12 1.57 -5.38 4.07 0.823 0.353 1.20
P43-b
b
0.88:1 17.6/2.57 1.46 -5.34 3.02 0.748 0.289 0.70
P43-c
b
0.65:1 21.5/2.36 1.39 -5.31 6.46 0.763 0.578 2.80
P43-TT
b
1:3 17.8/2.78 1.42 -5.29 7.78 0.748 0.405 2.40
a
Device performance was measured using PC
71
BM as acceptor material.
b
Device
performance was measured using PC
61
BM as acceptor material.
c
No data was provided in
literature due to poor solubility of the polymer.
!
37
In addition, an indolocarbazole derivative was explored as donor unit in such type of
random donor-acceptor copolymers (P41). Fine modulation of the photophysical and
electrochemical properties of these copolymers could be easily achieved adjusting the ratios
of comonomers.
73
Additionally, the 6H-phenanthrocarbazole (PC) unit has been used as
donor unit in such type of random donor-acceptor copolymers. This unit was copolymerized
by Deng and coworkers with benzothiadiazole (BT) at a number of varied ratios with the
thiophene bridging units (P42).
74
Compared with the perfectly alternating copolymer P42-alt
which had the same backbone composition as P42-c, random copolymers P42-c, P42-d and
P42-e obtained more pronounced intramolecular charge transfer band and extended
absorption. The D/A ratio also influenced power conversion efficiency of these
copolymer-based devices significantly, through variation in the spectrum absorption and
active layer morphology. Later, this group also reported the copolymerization of this PC unit
with DPP following the same strategy (P43). In addition, P43-TT was synthesized by
replacing the thiophene bridges by thieno[3,2-b]thiophene (TT) units, and this polymer
produced a slightly improved PCE of 2.2% when blended with PC
61
BM compared that of
P43-c. Nevertheless, with the addition of 1-chloronaphthalene (CN) in the solar cell active
layer, PCE based on P43-c:PC
61
BM was improved to 2.8%, higher than the CN-treated
P43-TT:PC
61
BM device (2.4%).
75

Similar to the bridging effects of thiophene units, ethylene was used as the linker bridge in
Stille polymerizations by Ko and coworkers to prepare random donor-acceptor copolymers
!
38
(P44 and P45), as shown in Figure 1.12.
76
Absorption of the random fluorene-based
copolymer P44-1 was red-shifted by 36 nm and enhanced compared to its alternating analog.
When the adjacent alkylated thiophene groups around the acceptor core (BT) were removed
in copolymers (P45-1 and P45-2, compared to P44-1 and P44-2 respectively), the strength of
the optical absorption was improved at longer wavelengths. by removing due to the better
HOMO-LUMO overlap. Nevertheless, P44-1 had a deeper HOMO energy level than the
more strongly absorbing P45-1 and P45-2, which resulted in a much higher value of V
oc
and
a better PCE of 1.42%.

Figure 1.12. Schematic syntheses and comonomer structures for
copolymers with ethylene bridging units. Composition being m:n = 1:1 for
all copolymers.

!
39
Table 1.8. Polymer properties and device performance parameters of copolymers shown in
Figure 1.12.
Polymer
Compo-
sition
(m:n)
M
n
(kg/mol)
/PDI
E
g

(eV)
HOMO
(V)
J
sc

(mA/ cm
2
)
V
oc

(V)
FF
PCE

(%)
a

P44-1

1:1 25/1.60 2.01 -5.06 5.82 0.765 0.32 1.42
P44-2 1:1 13/1.68 1.74 -4.83

2.03 0.495 0.32 0.32
P45-1 1:1 8.1/1.65 1.88 -5.15 1.49 0.675 0.36 0.36
P45-2 1:1 7.6/1.97 1.57 -4.91 4.42 0.525 0.41 0.95
a
Device performance was measured using PC
71
BM as acceptor material.
Other than the above-mentioned random D-A copolymers, two strong electron-acceptor
units have been used as M2 and M3 in a few cases. The general polymerization schemes
involving either thiophene or ethylene linker bridges are shown in Figure 1.13a and 1.13b,
respectively. The previously reported structures are depicted under each polymerization
scheme in Figure 1.13 (P46-P49). Their device performance parameters and summarized in
Table 1.9. Such combination of comonomers takes advantage of the weak electron-donating
feature of thiophenes or ethylenes, achieving two different ICT absorption peaks from their
linkage with the two acceptor units M2 and M3 respectively.
!
40

Figure 1.13. Schematic syntheses and comonomer structures for
copolymers with thiophene or ethylene bridging units and two other
acceptor units. Composition being m:n = 1:1 unless specified.
The weakly electron-withdrawing thiophene-3,4-dicarboxylate unit was introduced into the
polymer backbone (P46), in order to increase the V
oc
of polymer solar cells based on
DPP-containing polymers.
77
Compared to the alternating analog without thiophene linkers,
!
41
the HOMO energy level of random copolymer P46 was 0.18 V higher, which resulted in a
lower V
oc
in solar cell devices blended with PC
61
BM. Nevertheless, the deviced based on P46
achieved a much better PCE of 3.53% compared to 0.92% for alternating copolymer,
benefiting from both an enhanced J
sc
due to a lowered bandgap and a remarkably improved
FF. Jung and coworkers copolymerized bis-stannylated thiophenes with two accepting units,
isoindigo and DPP, at raried ratios (P47).
78
The resulting polymers exhibited broadened
absorption due to the so-called panchromatic effect. The optical and electrochemical
properties could be optimized by varying the ratio of the two acceptor units. P47-b-based
solar cells exhibited a promising PCE of 6.04% with a V
oc

of 0.77 V, a J
sc

of 13.52 mA/cm
2
and a FF of 0.58, which was an improvement compared to the two homopolymers The
combination of acceptor pair (TTD and TBT) also provided fine-tuning of polymer optical
and electronic properties according to the molar ratio of acceptor units (P48).
79
The
copolymers with a specific composition (TTD to TBT ratios between 70:30 and 50:50)
exhibited promising potentials as OPV donor materials, showing relatively strong absorption
covering the entire visible to NIR region as well as suitable energy levels, although no OPV
performance data was provided.
Similarly, ethylene units were used as the linker bridges in Stille polymerization with a
mixture of two acceptors DPP and BT (P49).
80
The variation in ratio of the two acceptor
units not only affected the polymer optoelectronic properties, but also influenced the
!
42
molecular packing in both pristine polymer films and polymer:PCBM blend films. Solar cells
based on P49-a reached a PCE of 2.4%.
Table 1.9. Polymer properties and device performance parameters of copolymers shown in
Figure 1.13.
Polymer
Compo-
sition
(m:n)
M
n
(kg/mol)
/PDI
E
g

(eV)
HOMO
(V)
J
sc

(mA/ cm
2
)
V
oc

(V)
FF
PCE

(%)
P46
a
1:1 5.11/2.8 1.31 -5.14 8.53 0.66 0.63 3.53
P47-a
b
7:3 31.4/1.13 1.46 -5.20

7.84 0.78 0.57 3.49
P47-b
b
1:1 25.3/1.17 1.4 -5.32 13.52 0.77 0.58 6.04
P47-a
b
3:7 24.7/1.24 1.38 -5.33 12.35 0.69 0.57 4.86
P48-a

92:8 6.4/2.5 1.00 -5.34 --
c
--
c
--
c
--
c

P48-b

79:21 3.2/3.7 1.03 -5.26 --
c
--
c
--
c
--
c

P48-c

56:44 3.7/3.8 1.08 -5.23 --
c
--
c
--
c
--
c

P48-d

32:68 5.0/3.4 1.12 -5.15 --
c
--
c
--
c
--
c

P48-e

13:87 9.3/2.6 1.14 -4.93 --
c
--
c
--
c
--
c

P49-a
a
3:7 18.0/1.68 1.36 -5.07 7.35 0.60 0.58 2.40
P49-b
a
1:1 18.8/1.64 1.31 -5.13 3.54 0.59 0.59 1.30
P49-c
a
7:3 22.1/1.34 1.26 -5.16 1.52 0.59 0.56 0.50
a
Device performance was measured using PC
61
BM as acceptor material.
b
Device
performance was measured using PC
71
BM as acceptor material.
c
No data provided in
literature.
As mentioned in the beginning of Section 1.3, this general type of synthetic strategy does
not involve regio-selective reactions. In other words, if any of the comonomers have
unsymmetrical structures, the resulting copolymers will be regio-random. For example, in the
sub-category incorporating thiophene linkers through Stille polymerizations, a number of
research has involved 3-hexylthiophene as M2, in combination with another
dibromo-functionalized M3, as shown in Figure 1.14 (P50-P55). The resulting regio-random
copolymers are discussed in the following paragraphs and summarized in Table 1.10.
!
43

Figure 1.14. Schematic syntheses and comonomer structures for copolymers
containing thiophene bridging units as well as 3-hexylthiophene.
The Y. Li group used a two-dimensional (2-D) conjugated thiophene derivative as M3,
and prepared polythiophenes with different ratios between hexyl and bi(thienylenevinylene)
side chains (P50-1).
81
These copolymers showed a broader absorption band than P3HT,
covering the region 350 to 650 nm, while their HOMO levels were determined to be 0.2 eV
!
44
lower than that of P3HT, which led to a higher V
oc
by 0.1 V in solar cells. The maximum
power conversion efficiency (PCE) of the PSCs based on P50-1-b reached 3.18%, which is
38% increased in comparison with that of the devices based on P3HT (2.41%). The same
group also reported the copolymers with ethylene units instead of thiophenes being the linker
bridges (P50-2).
82
Although these copolymers did not perform well in solar cells, an even
broader coverage of the solar spectrum was achieved in the absorpotion of these copolymers,
reaching 780 nm at the onset. In the same year, they also reported the use of another 2-D
conjugated thiophene derivative with terthiophene-vinylene side chains in preparing random
copolymer series P51.
83
As the content of the conjugated side chains increased, the
absorption peak corresponding to the branched thiophene segments enhanced and
blue-shifted. The hole mobility raised from 1.16 × 10
-4
cm
2
/V to 6.35 × 10
-4
cm
2
/V when the
conjugated side chain content increased from 10% to 20%, but it decreased as the content
further increased over 20%. This trend in hole mobility also translated in the PCE of
polymer:PC
61
BM-based solar cells. Similarly, Wang and coworkers explored thiophene
derivatives with bulky conjugated side-chains composed of the triphenylamine, thiophene,
and vinylene groups (TPATh), and copolymerized it following the general scheme shown in
Figure 1.14 (P15-r, refer to Section 1.2.2).
44
Compared to the regioregular copolymer of
3-hexylthiophene (3HT) and this 2-D conjugated comonomer (refer to P15 in Section 1.2.2),
the regio-random copolymer exhibited a better conjugation along the polymer backbone and
higher absorption intensity in the vision region. Solar cells fabricated with the blend of
P15-r:PC
71
BM showed a PCE of 1.75%.
!
45
Table 1.10. Polymer properties and device performance parameters of copolymers shown in
Figure 1.14 and Figure 1.15.
Polymer
Composition
(m:n) or
(m:n:k)
M
n
(kg/mol)
/PDI
E
g

(eV)
HOMO
(V)
J
sc

(mA/
cm
2
)
V
oc

(V)
FF
PCE

(%)
P50-1-a
a
1:0.99 48/1.34 1.82 -4.94 8.74 0.69 0.43 2.57
P50-1-b
a
1:0.59 50/1.62 1.82 -4.93

10.3 0.72 0.43 3.18
P50-2-a 1:1.7 16/2.41 1.63 -4.77 --
b
--
b
--
b
--
b

P50-2-b 1:0.63 26/1.33 1.63 -4.77 --
b
--
b
--
b
--
b

P50-2-c
a
1:0.35 12/3.01 1.63 -4.77 2.27 0.48 0.30 0.32
P51-a
a
9:1 34/1.48 1.88 -5.07 6.58 0.58 0.39 1.47
P51-b
a
8:2 36/1.56 1.88 -5.07 6.85 0.73 0.38 1.91
P51-c
a
7:3 40/1.74 1.91 -5.09 5.69 0.74 0.39 1.65
P51-d
a
6:4 43/1.51 1.91 -5.09 6.42 0.63 0.36 1.47
P15-r
c
1.05:1 5.0/2.34 1.82 -5.35 6.83 0.71 0.36 1.75
P52-a
a
1:1 95/2.90 1.27 -5.05 6.61 0.507 0.37 1.29
P52-b
a
1.9:1 46/2.37 1.31 -5.10 6.64 0.543 0.40 1.41
P52-c
a
3.7:1 132/2.47 1.45 -5.17 8.66 0.595 0.38 1.93
P52-d
a
6.3:1 47/2.50 1.78 -5.25 6.84 0.608 0.41 1.79
P53-1

1:0.96 12/2.21 2.04 -5.58 --
b
--
b
--
b
--
b

P53-2

1:0.92 74/1.84 1.82 -5.35 --
b
--
b
--
b
--
b

P53-3

1:0.59 38/2.60 1.55 -5.39 --
b
--
b
--
b
--
b

P54
a
1:1 28.8/1.67 1.58 -5.09 2.60 0.72 0.30 0.57
P55-1-a
a
82:18 M
w
=5.6 2.21 -5.10 0.98 0.75 0.29 0.20
P55-1-b
a
54:46 M
w
=4.2 1.99 -5.01 3.80 0.59 0.26 0.60
P55-2
a
71:29 M
w
=5.3 1.76 -4.99 5.03 0.60 0.29 0.90
P56-a
c
1:0.55 38.6/1.97 1.74 -5.21 10.40 0.740 0.740 3.71
P56-b
c
1:0.34:0.14 27.3/2.49 1.56 -5.16 9.67 0.724 0.473 3.31
P56-c
c
1:0.26:0.34 26.0/2.50 1.48 -5.12 10.10 0.715 0.431 3.12
a
Device performance was measured using PC
61
BM as acceptor material.
b
No data provided
in literature.
c
Device performance was measured using PC
71
BM as acceptor material.
Liang and coworkers prepared a series of copolymers based on thieno[3,4-b]thiophene (TT)
and thiophene units (P52).
84
When the ratio of TT unit to 3HT in the copolymer increased,
the energy gap of copolymers narrowed, resulting from the decrease in LUMO and increase
in HOMO energy levels. At optimized composition of 3.7:1 between TT and 3HT, the
copolymer led a PCE of almost 2% in solar cells based on blends with PC
61
BM. In addition,
!
46
the copolymers containing different benzothiadiazole derivatives (P53-1, P53-2 and P53-3),
quinoxaline derivative (P54) and 2-pyran-4-ylidene-malononitrile derivatives (P55-1 and
P55-2) provided another data point in the structural-property relationship of such random
regio-irregular copolymers.
85–87

In addition, thiophene has been used to copolymerize three other comonomer units in a
work reported by Tsai and coworkers in 2010 (P56), as shown in Figure 1.15.
88
Using
thiophene-phenylene-thiophene derivative (TPT) as an electron-donor, in combination with
the acceptor dithienylquinoxalines (DTQ) or a mixture of DTQ and DPP, the authors
achieved three low bandgap polymers. Increase of the DPP unit significantly enhanced the
near infrared absorption, lowered the band gap, and improved the carrier mobility. This work
demonstrated the vast scope of possible comonomer combinations and linkages in the
random copolymer approach.

Figure 1.15. Syntheses and structures of copolymers with thiophene bridging units and three
other comonomer units.
!
47
1.3.3 Other random D-A copolymers by Stille or Suzuki polymerizations
Thiophene and ethylene are only slightly electron-donating, therefore they do not interact
effectively with the electron acceptor unit to generate a strong donor-acceptor pair. Therefore,
a number of research works have utilized stronger electron-donating units as the alternating
M1 building block. Among these, two sub-types of comonomer arrangements can be
recognized, one with the M2 and M3 comonomers being a donor-acceptor pair, the other
with M2 and M3 being two different acceptor units, as illustrated in Figure 1.16.

Figure 1.16. General scheme of the two sub-types of copolymers
containing an electron-donating unit as M1.
The combination of strongly electron-donating M1 and a donor-acceptor pair as M2 and
M3 (Figure 1.16a) will be discussed first. Following this synthetic route, the choice of
suitable comonomer combinations leads to strong and broadened absorption profile,
benefiting from both the π−π* and ICT peaks from the D-D and D-A interactions
respectively. Stille polymerization is the most popular method to generate random
copolymers following the above-mentioned route. Examples are summarized in Figure 1.17
(P57-P64) and Table 1.11.
!
48
!
Figure 1.17. Schematic syntheses via Stille polymerization and comonomer
structures of random copolymers that fall into the description in Figure 1.16a.
!
49
A series of conjugated polymers were synthesized containing alternating electron-donating
cyclopentadithiophenes (CPDT), as well as randomly distributed acceptor BT and second
donor bithiophene (biTh) (P57).
89
UV-Vis absorptions of these copolymers were proved to be
a sum of the two alternating copolymers based on CPDT/BT and CPDT/biTh respectively,
covering a broad range from 400 to 800 nm. The relative absorption intensity between the
ICT and π−π* peaks could be modulated by varying the ratio of BT and biTh units. PCE up
to 3% was achieved from solar cells based on the blends of each of these copolymers with
PC
61
BM.
Yuan and coworkers used Stille polymerization to prepare two random copolymers that
contain electron-rich benzodithiophene (BDT) and CPDT units, respectively, as the
alternating building block (P58-1 and P58-2), together with a donor-acceptor comonomer
pair that random distributed in each polymer backbone.
90
When compared to the alternating
copolymer analogs without the second donor M3, P58-1 and P58-2 showed a slightly larger
bandgap, but significantly broadened absorption profile covering the shorter wavelength
region due to the π−π* contribution. Additionally, the morphologies of the blends of P58-1
and P58-2 with PC
71
BM were observed to be more homogeneous than that of the alternating
copolymer analogs, which led to superior device performance. Solar cells based on
P58-2:PC
71
BM blend achieved a J
sc
of 10.85 mA/cm
2
and a PCE of 3.15%.
Bithiophene was used by Shi and coworkers to copolymerize the content-tunable pair of
benzothiadiazole and carbazole moieties (P59).
91
Broad absorption covering the visible
!
50
region from 300 to 800 nm was achieved for these copolymers, while strong interchain
stacking and semicrystallinity was observed at the same time. PCE of 1.1-1.2% could be
reached by the solar cell devices based on a blend of such copolymer with PC
61
BM.
Table 1.11. Polymer properties and device performance parameters of random copolymers
shown in Figure 1.17.
Polymer
Compo-
sition
(m:n)
M
n
(kg/mol)
/PDI
E
g

(eV)
HOMO
(V)
J
sc

(mA/
cm
2
)
V
oc

(V)
FF
PCE

(%)
P57-a
a,b
2:1 28/1.4 1.70 -5.25 -- -- -- ~3
P57-b
a,b
1:1 30/1.6 1.83 -5.38

-- -- -- ~3
P57-c
a,b
1:2 23/1.4 1.75 -5.30 -- -- -- ~3
P58-1
c
1:1 22.1/2.54 1.60 -5.18 6.41 0.71 0.517 2.35
P58-2
c
1:1 5.0/1.79 1.49 -5.13 10.85 0.63 0.461 3.15
P59-a
a
82:19 10.4/1.27 1.59 -5.12 4.04 0.67 0.44 1.18
P59-b
a
72:28 9.7/1.13 1.62 -5.12 3.23 0.64 0.47 1.12
P59-c
a
51:49 8.6/1.80 1.64 -5.10 0.68 4.43 0.36 1.09
P60-a
a
1:2 18.5/2.92 2.14 -5.30 2.74 0.838 0.458 1.05
P60-b
a
1:1 23.4/2.13 2.01 -5.22 4.06 0.751 0.444 1.35
P60-c
a
2:1 26.3/2.47 1.87 -5.12 4.58 0.671 0.373 1.14
P61
a
1:1 15.4/1.68 1.93 -5.53 12.63 0.79 0.49 4.85
P62-a
a
1:3 26/2.23 1.51 -5.58 2.24 0.72 0.41 0.67
P62-b
a
1:2 22.1/2.85 1.44 -5.55 11.79 0.78 0.50 4.64
P62-c
a
1:1 22.4/1.92 1.46 -5.66 15.13 0.76 0.49 5.62
P62-d
a
3:1 37.2/2.24 1.26 -5.23 10.08 0.74 0.51 3.75
P63-a
a
3:1 14.3/1.92 1.43 -5.29 6.16 0.72 0.49 2.16
P63-b
a
1:1 17.1/1.85 1.47 -5.36 12.21 0.74 0.46 4.17
P63-c
a
1:3 16.6/2.01 1.52 -5.47 14.13 0.78 0.48 5.29
P64-a
c
3:1 190/2.7 1.37 -5.2 11.10 0.68 0.61 4.58
P64-b
c
1:1 165/2.6 1.37 -5.1 12.07 0.66 0.53 4.28
a
Device performance was measured using PC
61
BM as acceptor material.
b
No detailed
performance data was provided.
c
Device performance was measured using PC
71
BM as
acceptor material.
Three random donor–acceptor copolymers containing variable contents of the electron
acceptor DBT and the electron donor fluorene derivatives with terphenyl mesogen side
!
51
chains were synthesized through the bridging of 3HT units (P60).
92
The results showed that
increasing the amount of mesogen-attached fluorene units not only provided control over the
bandgaps but also influenced the liquid crystallinity of the copolymers. A good correlation
was observed between the fluorene/DBT ratio and HOMO levels of the polymers, and hence
the V
oc
values in solar cells based on the polymer:PC
61
BM blends. Therefore, copolymer
P62-a had the highest V
oc
of 0.86 V among the three due to its deepest HOMO energy level,
while the superior absorption behavior of P60-c led to the largest J
sc
of 4.58 mA/cm
2
.
The random terpolymer containing alternating bridge unit alkylthienyl substituted
benzodithiophene (BDTT) and the randomly incorporated donor-acceptor pair of a 2-D
conjugated thiophene derivative and DPP was synthesized by Stille polymerization by the
Tan group (P61).
93
By incorporating the dialkylthienylbenzothiadiazolethienyl vinylene
(TDTBT) side chain on thiophene units, the absorption spectrum of P61 was significantly
broadened, while the HOMO energy level was lowered due to suitable backbone twists in the
main chain. Therefore, P61 exhibited an improved PCE of 4.85% in OPV devices when
blended with PC
61
BM. The good performance of P61 inspired the following studies of the
Tan group to expand their copolymer library. Isoindigo (TID) based side chains were used to
replace the original TDTBT side chains on thiophenes (P62).
94
A series of copolymers with
varied ratio between DPP and TID-functionalized thiophene units were prepared and tested.
It was found that the increase in TID-substituted thiophene content induced an increased
absorption between 450 and 600 nm and a lower HOMO energy level, while the absorption
!
52
between 600 and 900 nm became stronger with higher DPP content. At optimized
comonomer composition, P62-c-based solar cells reached a high J
sc
of 15 mA/cm
2
and a PCE
of 5.62%. The same group also explored the alkyloxy-substituted BDT instead of BDTT in
similar random copolymers (P63).
95
A broad absorption range from 300 to 900 nm and deep
HOMO energy levels were achieved. P63-c benefited from superior optical and electronic
properties due to a suitable ratio between DPP and substituted thiophenes and produced a
PCE as high as 5.29% in solar cell devices based on its blend with PC
61
BM.
Nielsen and coworkers employed benzotrithiophene (BTT) unit as the bridging linker to
copolymerize DPP and a differently substituted BTT (P64).
96
The increase in the overall
BTT-content resulted in the appearance of a new absorption band around 500 nm, which was
attributed to the π−π* transition of oligomeric BTT chromophores within the polymer chain.
A raised LUMO level of P64-b was observed due to a larger portion of electron-donating
units in the copolymer backbone, which contributed to an improved charge transfer from
donor (polymer) to acceptor (PC
71
BM) materials in the active layer of solar cell devices.
Additionally, the blends of copolymers P64 with PC
71
BM were found to be homogeneous
according to AFM images. As a consequence of their broad absorptions, suitable energy
levels and favorable blend morphologies, both copolymers performed well in solar cell
devices with PC
71
BM as the acceptor material, with PCE above 4%.
In addition to the copolymers synthesized by Stille polymerizations, Suzuki polymerization
has also been employed to prepare random copolymers that fall into the general category
!
53
depicted in Figure 1.16a, such as the examples shown in Figure 1.18 (P65-P68) and Table
1.12.

Figure 1.18. Schematic syntheses via Suzuki polymerization and comonomer
structures of random copolymers that fall into the description in Figure 1.16a.
Composition being m:n = 1:1 unless specified.
Fluorene or dithienyl-fluorene derivatives were used as the bridging unit to copolymerize
dithienyl-benzothiadiazole (DTBT) and perylene at varied ratios (P65-1 and P65-2).
97
The
introduction of perylene units not only modulated the optical and electronic properties of the
!
54
resulting copolymers, but also led to improved charge-transport ability due to the
enhancement in π−π interaction between polymer chains. Solar cells based on the copolymer
P65-1-b showed a maximum PCE of 3.16%. Fluorene was also copolymerized as the
alternating linker with dibromo-functiionalized dithienylpyrazine and bis(3-hexylthiophene)
units (P66).
98
The random terpolymer was characterized and tested in solar cell devices based
on its blend with PC
61
BM. Combined optical and e with a variation of the charge carrier
mobilities supports our suspicion of a predominant limitation of the charge carrier mobility.
A large discrepancy was observed between the measured current density of 3 mA/cm
2
of
P66:PC
61
BM devices and the current density potentially available calculated by optical
simulations (15 mA/cm
2
), which was explained with electrical simulation results in the paper.
A dithienylpyrrole unit was used as the alternating building block in the random
copolymerization of BT with either indenofluorene or fluorene derivatives, respectively
(P67-1 and P67-2).
99
Compared to indenofluorene-containing copolymer P67-1, the
fluorene-containing P67-2 had a lower bandgap and absorbs up to 975 nm. However, P67-1
performed slightly better in solar cell devices when blended with PC
71
BM in comparison to
P67-2. The same dithienylpyrrole structure was incorporated into a set of copolymers P68 as
one of the dibrominated comonomers, together with BT as the other diborminated
comonomer and 3-octylthiophene (3OT) as the alternating bridge.
100
The three copolymers
with different BT/3OT ratios were found to exhibit quite different absorption profiles, but the
same bandgap of 1.80 eV. The polymer with equal amount of BT and 3OT units had the most
effective light harvesting ability among the three polymer samples, with a broad and flat
!
55
absorption across the region from 440 to 560 nm which led to the highest PCE of the device
with P68-a:PC
71
BM as the active layer compared to that of the other copolymers.
Table 1.12. Polymer properties and device performance parameters of random copolymers
shown in Figure 1.18.
Polymer
Composi-
tion
(m:n)
M
n
(kg/mol)
/PDI
E
g

(eV)
HOMO
(V)
J
sc

(mA/
cm
2
)
V
oc

(V)
FF
PCE

(%)
P65-1-a
a
97.2:2.8 14/3.5 1.9 -5.56 6.13 0.72 0.38 1.69
P65-1-b
a
88.6:11.4 12/2.7 1.92 -5.55

7.19 0.74 0.4 2.16
P65-1-c
a
82.7:17.3 11/3.0 1.93 -5.51 6.23 0.7 0.38 1.67
P65-2-a
a
93.9:6.1 12/3.3 1.82 -5.41 5.54 0.67 0.36 1.35
P65-2-b
a
86.7:13.3 11/2.7 1.81 -5.41 5.95 0.64 0.39 1.51
P66
b
1:1 8.6/1.5 --
c
--
c
3.16 0.647 0.42 0.79
P67-1
a
1:1 7.6/1.83 1.69 -5.05 7.28 0.43 0.378 1.18
P67-2
a
1:1 4.7/1.76 1.27 -5.00 6.74 0.36 0.371 0.90
P68-a
a
1:1 6.2/1.98 1.80 -5.12 9.27 0.57 0.384 2.03
P68-b
a
2:1 5.5/1.69 1.80 -5.10 7.00 0.57 0.36 1.44
P68-c
a
1:2 6.7/1.86 1.80 -5.03 5.08 0.56 0.36 1.03
a
Device performance was measured using PC
71
BM as acceptor material.
b
Device
performance was measured using PC
61
BM as acceptor material.
c
No data was provided in the
literature.
The examples discussed above fall into the general scheme depicted in Figure 1.16a. As an
alternative strategy, two acceptor units can be used as M2 and M3 to generate two sets of
donor-acceptor pairs along the polymer backbone and hence two different ICT bands in
UV-Vis absorption profile, as shown in Figure 1.16b. By changing the ratio of the two
acceptor units, one can achieve modulation of not only the relative intensity of the two ICT
absorption peaks, but also the HOMO energy level of the copolymers. Examples that have
been reported are summarized in Figure 1.19-1.22 (P69-P83) and Table 1.13-1.15.
!
56

Figure 1.19. Schematic syntheses via Stille polymerization and comonomer
structures of random copolymers that fall into the description in Figure
1.16b. Composition being m:n = 1:1 unless specified.
!
57
DPP was randomly copolymerized with two different acceptor units, BT and
thienopyrrolodione (TPD) respectively, with the bridging of benzotrithiophene units (P69-1
and P69-2).
96
The variation of the third component identity led to different HOMO levels of
the copolymers. The lowering HOMO level of P69-2 afforded an improved V
oc in solar cell
devices compared to P69-1.
However, device performance was found better based on P69-1
due to its superior J
sc
and FF values, reaching a PCE of 5.14%, almost twice of that based on
the alternating copolymer between BTT and DPP.
Three random copolymers based on both TPD and isoindigo units have been synthesized
with different alternating linker (P70-1, P70-2 and P70-3).
101
All three copolymers displayed
broad absorption profiles due to a mixture of two donor-acceptor ICT bands. The difference
in HOMO levels of these copolymers also proved the importance of optimizing the linker
unit structure. Among them, P70-1 afforded the best device performance with a PCE of
1.75%.
As benzodithiophene (BDT) derivatives have proved to be a promising building block in
alternating D-A copolymers,
102,103
including PTB7,
104,105
this type of donor units has also
found it use in a massive amount of random D-A copolymers. In most cases, BDT
derivatives are incorporated as the bridging comonomer M1 through Stille polymerizations,
as shown in Figure 1.19 and 1.20.
!
58
Alkylthienyl-substituted BDT (BDTT) units was employed to copolymerize
2,5-difluorobenzene (DFB) and DPP moieties (P71).
106
The random copolymer exhibited a
broad absorption band that extend from 300 to 850 nm, as well as a low-lying HOMO energy
level, which contributed to enhancement of both J
sc
and V
oc
in P71:PC
61
BM-based solar cells,
which reached an overall PCE of 2.50%. Even better performing copolymers have been
achieved based on similar alkylthienyl-attached BDT derivatives. For example, Jiang and
coworkers reported a series of copolymers containing electron-deficient benzooxadiazole
(BO) and DPP units (P72).
107
Due to the complementary absorption behavior of the two
donor-acceptor pairs (BDTT/BO and BDTT/DPP), a broad visible light absorption from 400
to 900 nm, as well as a low HOMO level deeper than -5.22 eV, was obtained for each of
these copolymers, while polymer bandgap and the maximum absorption wavelength could be
modulated by varying the ratio of BO and DPP units. At optimized conditions, P72-b-based
devices exhibited excellent photovoltaic performance, with a J
sc
of 17 mA/cm
2
, a V
oc

of 0.73
V, and a promising PCE of 6.8%. The series of copolymers based on DPP and TPD moieties
were synthesized by Kang and coworkers (P73).
108
Thanks to the different
electron-withdrawing strengths of DPP and TPD, an increase in TPD induced increased
absorption between 400 and 650 nm and a lower HOMO energy level, while higher DPP
contents resulted in stronger absorption between 600 and 900 nm. In addition, the polymer
packing structure and the electron distribution in the polymers could also be modulated with
the acceptor unit ratio variation. The best PCE of 6.33% was obtained from P73-b with
PC
71
BM, with an impressive J
sc
of over 16 mA/cm
2
.
!
59
Table 1.13. Polymer properties and device performance parameters of random copolymers
shown in Figure 1.19 and 1.20.
Polymer
Composi-
tion
(m:n)
M
n
(kg/mol)
/PDI
E
g

(eV)
HOMO
(V)
J
sc

(mA/
cm
2
)
V
oc

(V)
FF
PCE

(%)
P69-1
a
3:1 130/2.1 1.36 -5.2 10.95 0.68 0.69 5.14
P69-2
a
3:1 90/1.4 1.34 -5.3

8.87 0.72 0.66 4.23
P70-1
b
1:1 32/2.11 1.61 -5.57 7.35 0.78 0.30 1.75
P70-2
b
1:1 13/1.63 1.60 -5.35 0.83 0.67 0.39 0.22
P70-3
b
1:1 7.9/1.25 1.61 -5.22 2.52 0.74 0.35 0.67
P71
b
1:1 35.7/3.3 1.53 -5.42 5.97 0.75 0.56 2.50
P72-a
a
3:7 20.3/3.7 1.31 -5.22 12.5 0.70 0.62 5.50
P72-b
a
1:1 35.1/2.5 1.34 -5.32 17.0 0.72 0.55 6.80
P72-c
a
7:3 37.6/2.7 1.46 -5.36 12.0 0.75 0.59 5.30
P73-a
a
9:1 41/3.04 1.38 -5.33 10.95 0.74 0.61 4.91
P73-b
a,c
3:1 87/2.44 1.40 -5.35 16.32 0.76 0.51 6.33
P73-c
a
1:1 100/2.26 1.42 -5.42 12.95 0.76 0.53 5.26
P73-d
a
1:3 33/2.67 1.49 -5.47 7.61 0.79 0.50 3.02
P73-e
a
1:9 38/2.12 1.55 -5.48 6.09 0.82 0.42 2.09
P74-1
d
1:4 --
e
1.81
f
-5.07 17.25 0.683 0.672 7.93
P74-2-a
a
1:3 51/2.4 1.59 -5.26 16.19 0.73 0.54 6.38
P74-2-b
a
1:1 52/2.5 1.60 -5.26 16.63 0.75 0.57 7.01
P74-2-c
a
3:1 40/2.9 1.60 -5.28 16.32 0.76 0.59 7.28
P74-3-a
a
5.3:94.7 78.6/3.37 1.65 -4.94 16.2 0.62 0.462 4.60
P74-3-b
a
19.2:90.8 77.8/3.92 1.68 -4.93 16.2 0.62 0.54 5.50
P74-3-c
a
13.6:86.4 105/3.36 1.63 -5.01 16.0 0.59 0.411 3.90
P74-3-d
a
79.9:20.1 28.3/5.01 1.68 -4.99 16.1 0.69 0.51 5.70
P75-a
a
9:1 41.7/1.97 1.57 -5.42 1.67 0.83 0.59 0.82
P75-b
a
8:2 35.1/1.83 1.49 -5.37 2.35 0.81 0.63 1.20
P75-c
a
6:4 33.5/1.77 1.34 -5.25 5.02 0.67 0.53 1.78
P76-1
a
1:1 31/1.6 1.77 -5.21 6.40 0.74 0.49 2.35
P76-1
a
1:1 61.3/3.21 1.75 -5.17 10.3 0.81 0.60 5.01
P76-2
a
1:1 52/2.5 1.65 -5.20 7.70 0.76 0.51 3.00
a
Device performance was measured using PC
71
BM as acceptor material.
b
Device
performance was measured using PC
61
BM as acceptor material.
c
The device active layer was
treated with methanol.
d
Careful device engineering was carried out (Ref 109).
e
No
information was provided in literature.
f
Bandgap obtained from electrochemical
characterization of HOMO and LUMO energy levels with CV measurements.
!
60

Figure 1.20. Comonomer structures for random copolymers
synthesized by Stille polymerization following the synthetic
scheme depicted in Figure 1.19 (continued).
Additionally, based on the promising performance of PTB7, several groups have tried to
explore random copolymers with varied ratio between fluorinated and non-fluorinated TT,
while keeping BDT derivatives as the alternating bridge. In 2012, P74-1 containing
dialkyl-BDT was investigated in device engineering.
109
The inverted OPV device with
plasmonic enhancement from Au nanoparticles and reduced electron-hole recombination
from deposited ZnO thin layers reached a PCE of 7.9%. In 2014, BDTT units were used
!
61
instead of dialkyl-BDT to produce a family of copolymers with different amount of fluorine
decoration (P74-2).
110
The increase in fluorine TT content resulted in slight deepening of the
HOMO energy levels without significantly influencing the bandgap and the light harvesting
properties of the polymers, which led to an improved V
oc

in OPV devices while retaining
high J
sc
. Increasing fluorine content also contributed to a smaller scale phase separation in
the active layer with PC
71
BM, affording more favorable morphology and correlating with
improved PCE of the corresponding devices. Alkyloxy-substituted BDT units were employed
by He and coworkers in the syntheses of copolymer series P74-3.
111
Unfortunately, a strong
molecular weight effect was observed that largely masked any underlying fluorine-dependent
effect. Relatively high device efficiencies were achieved with low degrees of fluorination and
sufficient molecular weights, with an efficiency of 4.6% achieved for a degree of fluorination
of only 5.3%.
In addition to the use of alkyloxy-substituted BDT moities in P74-3, such structures were
also employed in other copolymers. For example, a set of copolymers containing varied
amount of randomly copolymerized TPD and DPP was reported by Zhang and coworkers
(P75). The copolymers exhibited broad and strong absorption covering the range from 350
nm to 922 nm, with detailed absorption behavior being tunable with TPD/DPP ratios.
However, compared to the copolymers P73 that are based on BDTT instead of alkyloxy-BDT
units, none of the P75 series performed well in solar cell devices.
112
Moreover, two
copolymers based on BT or a BT derivative in together with pyridylthiadiazole (BTz) were
!
62
prepared with alkyloxy-BDT as the alternating linker.
113
Compared to the alternating
copolymer of BDT and BTz, the random copolymers showed broadened the UV-Vis
absorption range, but adverse effects on the bulk molecular order and semiconducting
properties. Only moderate PCE was achieved in OPV devices based on either P76-1 or P76-2.
The same polymer P76-1 was also investigated by Kotowski and coworkers, with
comparison to the perfectly alternating copolymer at the same backbone composition.
114
Due
to the better solubility of the random copolymer relative the alternating one, the authors were
able to achieve P76-1 at very high molecular weight. A PCE of 5% was achieved in their
report.
Some other research works following the 1D/2A strategy have focused on the effect of side
chain substitutions on molecular packing and blend morphology with PCBM, as an
alternative pathway to improve OPV performance, such as the copolymers shown in Figure
1.21 and summarized in Table 1.14. Triethylene glycol (TEG) side chain was randomly
introduced onto DPP moieties of the low-bandgap polymers based on BDTT and DPP as a
stack-inducing agent (P77).
115
Due to the enhancement in the aggregation propensity of the
polymers induced by TEG modification, a more favorable degree of phase separation was
observed in the TEG-containing polymer:PC
71
BM blends, which led to over 10%
improvement in PCE (from 6.2% to 7.0%) with 10% TEG modification. Additionally,
poly(isoindigo-dithiophene)-based copolymers were subject to side chain engineering in that
varying amounts of low molecular weight polystyrene (PS) side chains (M
n

= 1300 g/mol)
!
63
were introduced though random copolymerizations (P78).
116,117
The better solubility of
polymers containing the PS side chains resulted in better film processability, without
affecting the electronic and optical properties, when the molar percentage of the
PS-containing repeating units were ≤10%. A much higher PCE of 7% was obtained in
comparison to the 6.3% without PS-side chains. Gedefaw and coworkers synthesized a
copolymer with hydroxyl functional group to improve the blend morphology with PCBM
(P79).
118
PCE of 2-3% could be achieved based on this copolymer.

Figure 1.21. Comonomer structures for random copolymers
synthesized by Stille polymerization following the synthetic scheme
depicted in Figure 1.19 (continued).

!
64
Table 1.14. Polymer properties and device performance parameters of random copolymers
depicted in Figure 1.21.
Polymer
Compo-
sition
(m:n)
M
n
(kg/mol)
/PDI
E
g

(eV)
HOMO
(V)
J
sc

(mA/ cm
2
)
V
oc

(V)
FF
PCE

(%)
P77-a
a
19:1 29.0/2.1 1.49 -5.31 13.8 0.73 0.63 6.30
P77-b
a
9:1 29.2/2.1 1.47 -5.31

14.3 0.72 0.68 7.00
P77-c
a
3:1 28.9/2.1 1.46 -5.28 10.3 0.7 0.65 4.70
P77-d
a
1:1 20.2/2.1 1.45 -5.25 8.1 0.68 0.58 3.20
P78-a
b
1:19 185/6.1 --
c
--
c
11.1 0.92 0.56 5.70
P78-b
b
1:9 341/5.3 --
c
--
c
11.6 0.96 0.63 7.00
P78-c
b
1:2 106/2.5 --
c
--
c
1.50 0.92 0.51 0.74
P79
b
9:1 17/2.18 1.97 -5.80 5.01 0.99 0.57 2.84
a
Device performance was measured using PC
71
BM as acceptor material.
b
Device
performance was measured using PC
61
BM as acceptor material.
c
No data was provided in
literature.
In addition to the 1D/2A copolymers synthesized by Stille polymerizations, Suzuki
polymerization method has also been utilized to generate random copolymers based one
electron-donating bridge unit and two other acceptor units, as shown in Figure 1.22. For
example, the copolymer based on carbazole in combination with a mixture of
dithienylbenzothiadiazole (TBT) and dithienylquinoxaline (TQ) was synthesized at high
molecular weight (P80).
119
The random copolymer exhibited comparable optical and
electronic properties as its alternating parent polymer PCDTBT, as well as a similar device
performance parameters. Song and coworkers copolymerized fluorene units with varied ratio
of TBT and thiadiazoloquinoxaline moieties (P81).
120
An increase in thiadiazoloquinoxaline
content resulted in a stronger absorption at the longer wavelength region between 600 and
900 nm, but a decrease in the absorption intensity between 480 and 600 nm, which
corresponded to the TBT contribution. HOMO energy levels were also tuned with the ratio
!
65
between TBT and thiadiazoloquinoxaline. Different N-heterocyclic moieties were screened
in conjugation with BT with the bridging linkage of fluorenes (P82-1, P82-2 and P82-3).
121

The resulting copolymers all exhibited strong absorption in the range from 300 nm to 700 nm,
as well as good solubility and thermal stability, but did not afford satisfying device
performance. In addition, the copolymer containing a carbazole donor and varying
compositions of perylene diimide (PDI) and naphthalene diimide (NDI) acceptors were
demonstrated to be potential acceptor materials in the all-polymer active layer of OPV
devices (P83).
122

!
66

Figure 1.22. Schematic syntheses via Suzuki polymerization and
comonomer structures of random copolymers that fall into the description in
Figure 1.16b.

!
67
Table 1.15. Polymer properties and device performance parameters of random copolymers
depicted in Figure 1.22.
Polymer
Composition
(m:n)
M
n
(kg/mol)
/PDI
E
g

(eV)
HOMO
(V)
J
sc
(mA/
cm
2
)
V
oc

(V)
FF
PCE

(%)
P80
a
1:1 216 /1.47 1.89 -5.45 9.00 0.87 0.44 3.50
P81-a
b
3:1 26/1.78 1.45 -5.47

5.25 0.69 0.46 1.67
P81-b
b
1:1 10.7/1.67 1.44 -5.37 5.25 0.70 0.48 1.80
P81-c
b
1:3 8.5/1.94 1.40 -5.30 4.25 0.65 0.48 1.32
P82-1
a
1:1 13.5/1.04 1.86

-5.70 2.20 0.59 0.26 0.34
P82-2
a
51:49 2.12/2.11 1.88 -5.75 2.49 0.59 0.32 0.47
P82-3
a
46:54 87.7/1.15 1.90 -5.45 3.42 0.37 0.28 0.35
P83-a
c
3:1 6.9/2.5 1.80 1.12 1.12 0.78 0.29 0.25
P83-b
c
1:1 6.4/2.0 1.80 1.11 1.11 0.82 0.26 0.24
P83-c
c
1:3 6.5/2.1 1.82 0.88 0.88 0.82 0.28 0.20
a
Device performance was measured using PC
71
BM as acceptor material.
b
Device
performance was measured using PC
61
BM as acceptor material.
c
Device performance was
measured based on the blends with PBDTTT-C.
In contrast to 1D/2A copolymers discussed above, a small number of 1A/2D copolymers
have also been investigated, as shown in Figure 1.23 (P84-P85) and Table 1.16.

Figure 1.23. Schematic syntheses comonomer structures of 1A/2D copolymers.
!
68
The alternating copolymer of CPDT and BT was modified by either changing some side
chains from branched 2-ethylhexyl to linear hexyl gorups or replacing CPDT with fluorene
units partially (P84-1 and P84-2).
123
It was found that incorporation of linear side chains
resulted in polymers with lower solubility but higher hole mobility, while introduction of
fluorenes afforded better absorption coverage of the mid-visible region of the solar spectrum.
But no device performance was reported. Anthradithiophene was introduced and randomly
copolymerized with BDT units with the bridging of either thienothiophene (TT) or
fluorinated TT, in order to elucidate the effect of the extended conjugation length on polymer
properties and OPV performance (P85 and P85-F).
124
The band gap and hole mobility of
donor polymers were systematically changed with the mole ratio of the monomers. However,
higher anthradithiophene proportion also led to a decrease in V
oc
and poor morphology with
PCBM, which compromised the enhancement in J
sc
and limited the gain in overall device
performance.
Table 1.16. Polymer properties and device performance parameters of 1A/2D copolymers
depicted in Figure 1.23.
Polymer
Composition
(m:n)
M
n
(kg/mol)
/PDI
E
g

(eV)
HOMO
(V)
J
sc

(mA/
cm
2
)
V
oc

(V)
FF
PCE

(%)
P84-1

1:1 --
a
1.55 --
a
--
a
--
a
--
a
--
a

P84-2 1:1 >20/--
a
1.77 -5.37

--
a
--
a
--
a
--
a

P85-a
b
1:3 20.2/1.3 1.57 -4.96 11.6 0.568 0.609 4.10
P85-b
b
1:1 9.5/3.7 1.54 -5.04 7.3 0.537 0.499 1.96
P85-c
b
3:1 2.7/2.2 1.52

-5.08 6.0 0.494 0.592 1.76
P85-F
b
3:1 2.12/2.11 1.57 -5.22 13.4 0.659 0.635 5.61
a
No data was provided in literature.
b
Device performance was measured with PC
71
BM as
acceptor material.
!
69
1.3.4 Random copolymers prepared by other polymerization methods or
post-functionalization reactions
The random copolymers discussed above were prepared with the most popular
cross-coupling polycondensation reactions, namely Stille and Suzuki polymerizations. In fact,
other polymerization methods can also be used to synthesize random copolymers that feature
an alternating bridging unit, such as the Suzuki-Heck polymerization shown in Figure 1.24
(P86).
125
The variation in the relative amount of comonomer units enabled control over the
bandgap and energy levels of the copolymers, which again demonstrated the possibility and
advantage to easily access a series of polymers by simply tuning the comonomer
combinations in random copolymers.

Figure 1.24. Syntheses and structures of copolymers prepared by Suzuki-Heck
polymerizations.

!
70
Table 1.17. Polymer properties and device performance parameters of copolymers depicted
in Figure 1.24 and 1.25.
Polymer
Compo-
sition
(m:n:o)
M
n
(kg/mol)
/PDI
E
g

(eV)
HOMO
(V)
J
sc

(mA/
cm
2
)
V
oc

(V)
FF
PCE

(%)
P86-a
a
0:1:1 7.1/1.4

1.98 -5.56

1.03 0.48 0.40 0.20
P86-b
a
0:1:3 6.9/1.6

2.12 -5.60

0.28 0.54 0.31 0.05
P86-c
a
1:1:0 4.1/1.7 1.77 -5.16 1.14 0.67 0.36 0.27
P86-d
a
1:1:2 10.0/1.3 2.02 -5.47 1.00 0.65 0.37 0.24
P86-e
a
2:1:1 9.2/1.3 1.96 -5.33 1.61 0.76 0.31 0.38
P87-1
a
1:1 12.9/1.81 1.57 -5.25 6.59 0.90 0.42 2.83
P87-2
a
1:1 6.0/1.28 1.57 -5.27 5.06 0.90 0.34 1.77
a
Device performance was measured using PC
71
BM as acceptor material.
In addition, Duan and coworkers prepared random copolymers by attaching two different
acceptor groups malononitrile and 2-(1,2-dihydro-1-oxoinden-3-ylidene)malononitrile
(DCNIO) onto alternating D-A copolymer backbones via successive Knoevenagel
condensations, as shown in Figure 1.25.
126
The absorption spectra of P87-1 and P87-2 were
observed to greatly extend due to a combination of acceptors with different
electron-withdrawing ability. P87-1 produced a PCE of 2.83% when blended with PC
71
BM
as the solar cell active layer.
!
71

Figure 1.25. Syntheses and structures of random copolymers prepared by
post-functionalizations.
1.3.5 Importance of comonomer linkage patterns
In addition to the examples discussed above within Section 1.3, some research works have
investigated the difference between isomeric polymer structures and revealed the significant
impact of comonomer linkage patterns on the polymer optical/electronic properties and hence
their performance in solar cell devices.
For example, two low bandgap copolymers composed of fluorene, CPDT and
dithienyl-benzothiadiazole (DBT) were synthesized and compared (P88-1 and P88-2), as
shown in Figure 1.26.
127
Due to the difference in electron-donating intensity between
fluorene and CPDT, the copolymer P88-1 featured a stronger donor-acceptor interaction of
!
72
directly bonded CPDT and DBT, which led to a lower bandgap and stronger absorption in the
longer wavelength region in comparison to P88-2, as well as a raised HOMO level by about
0.1 eV. As a result, devices based on P88-1 and P88-2 exhibited either a higher J
sc
value or a
superior V
oc
value, respectively, while their overall efficiencies ended up to be comparable.
Additionally, polymers P89-1 and P89-2, also shown in Figure 1.26, were reported as a pair
with different comonomer sequences.
128
In particular, the presence of oligofluorene block in
P89-2 had a positive impact on device efficiencies mainly due to the enhancement in FF
values.

Figure 1.26. Structures of random copolymers with isomeric comonomer
linkages.
In another example, the two isomeric random poly(arylene−vinylene)s shown in Figure
1.27 (P90) were synthesized by Suzuki-Heck polymerizations, with ethylene being the
bridging unit.
129
The two polymers exhibited remarkably different UV−Vis absorption
!
73
spectra, and P90-2 covered a broader wavelength range of the visible region relative to P90-1,
which also led to a higher PCE from solar cells based on P90-2 compared to P90-1.

Figure 1.27. Structures of random copolymers with isomeric comonomer linkages
(continued).

Table 1.18. Polymer properties and device performance parameters of copolymers depicted
in Figure 1.26 and 1.27.
Polymer
Compo-
sition
(m:n)
M
n
(kg/mol)
/PDI
E
g

(eV)
HOMO
(V)
J
sc

(mA/
cm
2
)
V
oc

(V)
FF
PCE

(%)
P88-1
a
1:1 9.8/1.81 1.62 -5.13 4.92 0.66 0.35 1.13
P88-2
a
1:1 12.9/2.12 1.86 -5.21 2.12 0.81 0.39 0.66
P89-1
b
66:34 9.0/2.5 1.49 -5.2 6.38 0.61 0.499 1.94
P89-2
b
94:6 17.8/1.93 1.46 -5.1 4.94 0.665 0.368 1.21
P90-1
a
1:1 6.7/1.4 1.93 -5.70 1.20 0.62 0.30 0.20
P90-2
a
1:1 8.6/2.0 1.95 -5.76 2.91 0.67 0.30 0.60
a
Device performance was measured using PC
61
BM as acceptor material.
b
Device
performance was measured using PC
71
BM as acceptor material.

!
74
1.4 Semi-random copolymers
Considering the two strategies of preparing copolymers that have been discussed in Section
1.2 and 1.3, it is obviously beneficial if the advantages from either of them could be
combined in a single synthetic platform. On one hand, the widely investigated rr-P3HT is
well-known for its semi-crystallinity, high hole mobility and favorable morphology with
PCBM. P3HT-based regioregular copolymers (See Section 1.2) can retain these promising
properties while offering tuning ability of the polymer energy levels. However, these
polymers typically suffer from large bandgaps and consequently narrow absorption breadth.
On the other hand, the random D/A approach (See Section 1.3) is successful in broadening
the polymer absorption due to the multichromophoric nature of polymer backbones. However,
the hole mobilities of these polymers are generally low due to the loss of ordering structure in
thin films introduced by the random polymerizations.
Therefore, another category of copolymers, namely “semi-random” copolymers, have been
designed and investigated. This type of copolymers is typically synthesized by Stille
polymerizations, as illustrated in Figure 1.28. Detailed features of this synthetic strategy are
explained as follows.

Figure 1.28. General scheme of synthetic route to prepared semi-random copolymers by
Stille polymerizations.
!
75
1. The polymerization starts with regio-specifically functionalized 3-hexylthiophenes
as a major fraction of the comonomer mixture, together with a small number of
electron-deficient comonomer units (M3) that are functionalized with bromines and another
comonomer moiety with complementary functionalities to balance out stoichiometry.
Typically, Stille polymerization method is employed, due to its excellent functional group
tolerance, with the complementary functionality being stannyl groups.
2. Due to the careful choice of functional groups and comonomer compositions, the
linkage pattern of monomers are restricted in that the polymer backbone is heavily dominated
by head-to-tail coupled 3-hexylthiophene, with only a small percentage of acceptor
monomers that do not bond to themselves. As such, a large degree of structural order is
retained, in contrast to the cases of completely random copolymers.
In principle, the semi-random copolymers can be considered as rr-P3HT-based copolymers
that contain a small number of randomized D/A linkages. Such structural features are capable
of retaining some favorable properties from rr-P3HT, while introducing multiple
chromophores at the same time. This hypothesis has been proved in several research works,
which will be discussed below.
Most of the existing examples of this so-called semi-random copolymer approach are
reported by the Thompson group, as shown in Figure 1.29 (P91-P99). Unsubstituted
thiophene comonomer is used not only to balance the functionality stoichiometry in
!
76
polymerizations, but also to further alleviate any unfavorable steric interactions. As
mentioned, the restrictions in monomer connectivity will help retain a considerable degree of
structural order, and therefore overcome some limitations of random copolymers, such as an
amorphous morphology and generally low hole mobilities.
!
77

Figure 1.29. Structures of semi-random copolymers that follow the synthetic
strategy shown in Figure 1.28.
The first generation of semi-random copolymers reported by the Thompson group include
P91 (10% benzothiadiazole (BTD) content), P92 (10% thienopyrazine (TP) content) and P93
!
78
(8.75% TP and 8.75% BTD content).
130
All three semi-random polymers exhibited a lower
band gap than P3HT and much broadened absorption profile, with P93 able to absorb almost
twice as many photons as P3HT. The copolymers were found semicrystalline except P92.
Specifically, the GIXRD peak of P93 corresponding to the (100) plane had a shift to larger
angles compared to P3HT, indicating that ordering in these semi-random polymers is not
simply due to packing of continuous 3-hexylthiophene sequences, but also dependent on the
nature of other comonomers. Because of the effective ordering, hole mobilities were
determined to be similar to that of rr-P3HT.
The Moon group has also investigated the semi-random copolymer series containing BTD
(P94).
131
Their copolymers incorporated BTD at varied percentage among all comonomers,
ranging from less than 6% to 20%. Semi-crystallinity and broad absorption range were
observed similar to the findings of the Thompson group. The best performance was achieved
with solar cell devices based on the copolymer containing BTD units of slightly over 10%.
In the following generation of semi-random copolymers, DPP unit was used as the acceptor
unit at different quantities ranging from 5 to 10% (P95).
132
The introduction of DPP into the
P3HT backbone was observed to significantly decrease the optical band gap and lead to the
formation of a dual band absorption, due to the π−π* transitions corresponding to
thiophene-rich segments and the ICT transitions correlated with segments rich in D/A
linkages. Semi-crystallinity was observed for all three DPP-containing polymers, which
resulted in high SCLC hole mobilities. Benefiting from the broad and intense absorption and
!
79
high mobility, an impressively high J
sc
, and consequently satisfying efficiencies of close to
6%,

was achieved from solar cells based on P95-b:PC
61
BM blends.
Another acceptor unit TPD has also been used in semi-random copolymers (P96).
133
The
absorption profiles of the TPD-containing polymers were slightly broadened compared to
P3HT but much narrower than for DPP-containing polymers. Although V
oc
of these
copolymers were higher (0.72 V and 0.68 V) than for DPP-based polymers, the solar cell
efficiencies were moderate due to a decreased low J
sc
. Additionally, two-acceptor
copolymers were made with a combination of TPD and DPP comonomers at varied ratios
(P97). The total acceptor monomer content was kept as 15% to ensure good solubility of the
resulting copolymers. The absorption peaks were observed to rise and fall according to the
change in acceptor ratio and the overall absorption profiles mimic the weighted sum of the
corresponding absorption profiles of P3HTT-DPP and P3HTT-TPD. Solar cell efficiencies
close to 5% were obtained due to a combination of very high J
sc

and FF.
In addition, the influence of the simultaneous incorporation of electron rich and electron
poor monomers investigated. The electron-donor dithienopyrrole (DTP) was introduced as a
dibrominated monomer, in combination with different acceptors (BTD, TP or DPP) in
semi-random copolymers (P98-1 to P98-4).
134
The DTP-containing copolymers were found
to be amorphous and have low hole mobilities except for DPP-DTP. Unfortunately, due to
the choices of functional groups, the electron-donating DTP cannot directly bond to the
acceptor units.
!
80
In addition, the influence of pairing two distinct acceptors in the semi-random polymer
backbone was explored (P99-1 to P99-4).
135
Interestingly, it was found that the shape of the
absorption profile two-acceptor copolymers appeared as a linear combination of those of the
respective one-acceptor copolymers, which made it possible to predict the absorption
behavior and bandgap of two-acceptor copolymers. Efficiencies of up to 3% were achieved
with this set of semi-random copolymers.

!
81
Table 1.19. Polymer properties and device performance parameters of semi-random
copolymers depicted in Figure 1.29.
Polymer
Composition
(m:n:o) or
(m:n:o:p)
M
n
(kg/mol)
/PDI
E
g

(eV)
HOMO
(V)
J
sc

(mA/
cm
2
)
V
oc

(V)
FF
PCE

(%)
P91
a
8:1:1 15.3/2.45 1.62 -5.41 2.87 0.79 0.33 0.75
P92
a
8:1:1 16.7/2.35 1.36 -5.23

3.22 0.44 0.50 0.71
P93
a

13:3.5:1.75:1.
75
16.3/2.05 1.27 -5.11 3.04 0.39 0.37 0.43
P94-a
b
3:1:1 28.5/1.82 1.40 -5.55 5.5 0.64 0.320 1.10
P94-b
b
6:1:1 20.1/1.61 1.20 -5.33 5.5 0.82 0.346 1.60
P94-c
b
9:1:1 18.2/2.02 1.20 -5.42 2.4 0.84 0.277 0.55
P94-d
b
12:1:1 16.1/1.45 1.00 -5.46 2.4 0.70 0.352 0.59
P94-e
b
15:1:1 26/2.50 1.59 -5.47 2.9 0.80 0.338 0.78
P95-a
a
9:0.5:0.5 19.0/2.8 1.52 -5.35 9.57 0.66 0.58 3.60
P95-b
a
8:1:1 24.6/2.3 1.51 -5.43 13.87 0.57 0.63 4.94
P95-c
a
7:1.5:1.5 17.6/2.9 1.46 -5.32 13.44 0.50 0.60 4.10
P96-a
a
8:1:1 22/1.73 1.82 -5.40 5.38 0.72 0.58 2.22
P96-b
a
7:1.5:1.5 44/2.18 1.80 -5.40 5.33 0.68 0.56 2.02
P97-a
a
14:3:1.5:1.5 52.4/3.4 1.48 -5.35 15.26 0.51 0.64 4.93
P97-b
a
14:3:2:1 15.7/3.4 1.50 -5.30 11.67 0.55 0.62 3.94
P97-c
a
14:3:1:2 21.7/3.2 1.47 -5.20 16.37 0.50 0.61 3.19
P98-1
a
13:3.5:3.5 20.2/3.5 1.96 -5.22 1.81 0.63 0.33 0.36
P98-2
a

13:3.5:1.75:
1.75
20.0/2.8 1.61 -5.15 1.48 0.47 0.31 0.21
P98-3
a

13:3.5:1.75:
1.75
21.2/3.0 1.38 -5.05 1.81 0.35 0.28 0.18
P98-4
a

13:3.5:1.75:
1.75
19.6/3.6 1.47 -5.18 10.77 0.53 0.50 2.83
P99-1
a
2:1 15.4/3.4 1.47 -5.22 10.91 0.50 0.55 2.97
P99-2
a
5:1 22.8/2.4 1.32 -5.14 7.94 0.36 0.48 1.37
P99-3
a
5:2 16.8/3.0 1.64 -5.40 6.89 0.58 0.62 2.48
P99-4
a
1:1 12.6/3.3

1.32 -5.29 4.56 0.36 0.48 0.78
a
Device performance was measured using PC
61
BM as acceptor material.
b
Device
performance was measured using PC
71
BM as acceptor material.
Importantly, within this class of semi-random copolymers, acceptor monomers are not only
used in a very small quantity, but are typically synthesized in only a few steps. This is in
strong contrast to the ever-increasing complexity of monomers used in perfectly alternating
!
82
D/A polymers, and makes the semi-random approach a low-cost platform to achieve
high-performance polymer materials for OPV applications.
1.5 Other copolymerization strategies
The synthetic strategy that will be discussed in this section includes several reaction
methods that may seem to differ from each other remarkably. However, some essential
elements are shared among these synthetic methods, which is depicted in Figure 1.30 and
described as follows.

Figure 1.30. General scheme of synthetic route to prepare random
copolymers through other polymerization strategies.
1. The polymerization starts with two (or more) comonomers with the same functional
groups, forming linkages through homocouplings or other reactions that possess
homocoupling features;
2. Distribution of the two (or more) comonomer building blocks is statistically random,
with no inherent restriction from monomer functionalities. However, the monomer sequences,
as well as incorporated composition of monomers, may be dictated by the reactivity
difference of comonomers and thus deviate from truly random patterns or monomer feed
ratios;
!
83
3. The reaction between monomers is not regioselective. Therefore, when the
comonomers are not symmetrical in the core structure, the resulting polymers will have
regio-random linkages in their backbone;
4. Some polymerization methods may result in alternating linker bridges between
arylene rings, such as Gilch polymerization and Acrylic Diene Metathesis (ADMET)
polymerization that will be discussed below. As such, the covalent bonding formed after
polymerization, as shown in Figure 1.30, is depicted as wavy lines, indicating the formation
of ethylene linkages between the core structures of comonomers in some cases.
Within this category, poly(arylene vinylene)s made by Gilch polymerization method are
among the most common structures prepared and investigated. As typical examples of
poly(arylene vinylene)s, poly[2-methoxy-5-(2′-ethylhexyloxy)-1,4-phenylenevinylene]
(MEH-PPV) or Poly[2-methoxy-5-(3′,7′-dimethyloctyloxy)-1,4-phenylenevinylene]
(MDMO-PPV) has reached PCE of over 2.5%.
136
Research efforts have been made to
introduce specific advantages to MEH-PPV by preparing random copolymers of
2-methoxy-5-(2′-ethylhexyloxy)-1,4-phenylenevinylene (MEH-PV) and other functionalized
arylene vinylene units. Gilch polymerization starts with two bis(halomethyl)arene compunds,
forming vinylene linkages that alternate with either of the two arylene building blocks.
Examples of random copolymers made by Gilch polymerizations are shown in Figure 1.31
(P100-P103).
!
84

Figure 1.31. Schematic syntheses and structures of random copolymers
prepared by Gilch polymerizations.
In particular, the Yongfang Li group reported the synthesis of random copolymers of
MEH-PV and 2,5-thienylene-vinylene (ThV) at two different compositions (P100).
137
The
copolymers exhibited combined advantages of lower bandgap from ThV contributions and
good film-forming ability from MEH-PV contributions. As a result, the OPV based on the
copolymer containing 18% ThV showed both enhanced J
sc
and FF relative to that of
MEH-PPV, and a 47% increase in the final PCE, reaching 1.13%. The authors also
synthesized an alternating copolymer of MEH-PV and ThV by Hornor–Emmons reaction,
which led to a PCE of only 0.63% (P101). But they failed to provide random copolymers that
!
85
match the 1:1 ratio of two repeating units for direct comparison. In 2008, the same group
incorporated a dendronized phenylene derivative into MEH–PPV to afford random
copolymer BE-co-MEH-PPV (P102).
138
The power conversion efficiency of the device based
on the blend of BE-co-MEH–PPV and PC
61
BM reached 1.41%, which was an improvement
from both the dendronized homopolymer BE–PPV and MEH–PPV under the same
experimental conditions.
In 2009, a poly(p-phenylenevinylene) PPV based copolymer with electron deficient
oxadiazole segments as the side chain was reported (P103).
139
The photovoltaic devices
based on 3C-OXD-PPV and PCBM blend film exhibit a V
oc
of 0.81V, J
sc
of 4.93 mA/cm
2
,
FF of 0.40 and PCE up to 1.60% which indicates that 3C-OXD-PPV is a promising polymer
for photovoltaic application.

!
86
Table 1.20. Polymer properties and device performance parameters of semi-random
copolymers depicted in Figure 1.31-1.33.
Polymer
Compo-
sition
(m:n)
M
n

(kg/mol)
/PDI
E
g,opt

(eV)
HOMO
(eV)
J
sc

(mA
/cm
2
)
V
oc

(V)
FF

PCE
(%)
P100-a
a
89:11 106/3.13 2.18 --
d
2.23 0.66 0.418 0.88
P100-b
a
81:19 139/2.33 2.17 --
d
2.68 0.66 0.426 1.13
P101
a
N/A 15.6/2.23 2.07 --
d
1.79 0.69 0.412 0.63
P102
b
1:1 110/1.64 2.16 −5.02 3.37 0.81 0.42 1.41
P103
b
71:29 125/4.28 2.10 −5.05 4.93 0.81 0.40 1.60
P104
c
1:1 34.6/3.27 1.29 -5.71 2.31 0.74 0.61 1.0
P104-alt
c
N/A 50/2.74 1.42 -5.78 9.95 0.78 0.68 5.3
P105-a
c
77:23 38.8/2.5 1.77 -5.81 4.91 0.920 0.37 1.67
P105-b
c
61:39 31.2/2.3 1.65 -5.62 7.42 0.852 0.416 2.63
P105-c
c
52:48 21.9/2.3 1.58 -5.60 7.01 0.780 0.377 2.06
P105-d
c
42:58 13.0/1.8 1.57 -5.51 4.90 0.682 0.323 1.08
P105-e
c
23:77 8.4/1.7 1.51 -5.42 6.46 0.53 0.499 1.71
a
Device performance was measured using C
60
as acceptor material.
b
Device performance
was measured using PC
61
BM as acceptor material.
c
Device performance was measured
using PC
71
BM as acceptor material.
d
No data provided in literature.
In addition, Yamamoto homocoupling can be used to randomly copolymerize two
brominated arylene comonomers. Hendriks and coworkers prepared a random copolymer
of dithienyl-DPP and dithienyl-TPD in 2014 by Yamamoto polymerization (P104), and
compared the random copolymer with regular alternating copolymer at the same
composition (P104-alt), as shown in Figure 1.31. DPP2T-rich sections in the random
copolymer led to higher HOMO level and a smaller optical bandgap compared to the
regular copolymer. However, the randomization of DPP and TPD units along the
backbone has a negative effect on the photovoltaic performance, mainly due to the low
lying LUMO energy level hampering exciton dissociation in solar cells.
140

!
87

Figure 1.32. Schematic synthesis and structure of random copolymers
prepared by Yamamoto polymerization, in comparison to the alternating
copolymer prepared by Suzuki polymerization.
In addition, acyclic diene metathesis (ADMET) was employed to prepare a series of
copolymers containing randomly distributed fluorene and TBT units as well as
alternating vinylenes, as shown in Figure 1.32 (P105).
141
The polymerizations were
carried out with both divinyl comonomers under the catalysis of Ru complexes, while the
resulting copolymers feature similar backbone sequences with those prepared with the
strategy described in Section 1.3.2. Particularly, within the polymer series reported by
Paulsen and coworkers, it was found that the HOMO level was tunable over a 400 meV
range through variation of the donor–acceptor unit ratio, which resulted in a similarly
tunable V
oc
value. Peak photovoltaic performance was found in solar cells based on donor
rich statistical D–A copolymers, demonstrating the significance of random copolymers
with non-stoichiometric D–A compositions.
!
88

Figure 1.31. Schematic syntheses and structures of random copolymers
prepared by ADMET polymerization.

1.6 Summary and conclusion
In this chapter, the use of random and semi-random conjugated copolymers as donor
materials in OPV devices has been summarized in four subsections according to the
specific synthetic strategy of copolymers. From the examples mentioned above, two
major aspects of property tuning can be realized using the random/semi-random
copolymer approaches. On one hand, the polymer electronic properties, including the
bandgaps, absorption range and HOMO/LUMO energy levels, can be finely tailored
when comonomers with different electron-donating or electron-withdrawing ability are
combined. On the other hand, optimization of polymer solubility or molecular packing
behavior can be realized by varying the solubilizing groups bonded to polymer backbones.
Additionally, although not a focus in the discussion above, the combination of random
copolymer preparation and post-polymerization modification reactions can further
produce functionalized conjugated polymers at desired composition, which can be
applied to a vast scope of investigations.
In comparison to homopolymers and/or the state-of-the-art alternating D-A copolymers,
random or semi-random copolymers exhibit several distinct advantages.
!
89
(1) The ratios between comonomer components can be easily varied in random
copolymers over a wide range. As such, the properties of the polymers, including
absorption behavior, HOMO/LUMO energy levels, charge mobility and solubility, can be
finely tuned according to the nature of each comonomer unit. This offers a fast and easy
pathway to screen over a series of polymers that may be potential well-performing
candidates, without investing time and effort in developing complex monomer structures.
(2) In most of the cases, random or semi-random copolymers consist of several
distinct comonomer units bonded in randomized linkages. A number of different
monomer sequences being present in polymer backbones results in a combination of
multiple chromophores and hence a truly broadened absorption band covering a wide
range of the visible light spectrum. This effect is remarkable in the D-A random
copolymers discussed in Section 1.3 and the semi-random copolymers discussed in
Section 1.4, and it has led to impressive Jsc values around 16-17 mA/cm
2
in several
examples.
(3) Due to the randomized backbone structures, these copolymers usually do not
suffer from poor solubility as observed for some highly ordered alternating D-A
copolymers.
(4) Random copolymers can result in a more favorable and thermodynamically
stable bicontinuous morphology when blended with fullerene derivatives, due to the
decreased tendency for crystallization.
These advantages make the random/semi-random copolymer approach a promising
platform to develop new well-performing conjugated polymers in OPV applications.
!
90
Even without the extensive focus being made as for alternating D-A copolymers, some
high efficiencies exceeding 7% have already been achieved with several
random/semi-random copolymers.
Inspired by the promising potential of these copolymers and the relatively limited
investigations on them, this thesis has a focus on further developing the
structure-property-performance relationship of random/semi-random copolymers
(Chapter 2 and Chapter 3). The use of post-polymerization “click” reactions was also
employed in connection with random copolymer synthesis, in order to generate a
functionalized rr-P3HT with the aim of providing comparisons to a ternary blend system
(Chapter 4). Another piece of work that involved the same “click” chemistry in
preparation of non-conjugated polymers with electroactive pendants is included in
Chapter 5 as well. As a whole, the research presented in the following chapters should
enrich the family of random/semi-random conjugated copolymers for OPV applications,
providing fundamental understanding in the structure-property-performance relationships
of such polymers and expanding the scope of their usage into investigation of ternary
blend solar cells.
1.7 References
(1) Thompson, B. C.; Fréchet, J. M. J. Angew. Chem. Int. Ed. 2008, 47, 58–77.
(2) Brabec, C. J.; Gowrisanker, S.; Halls, J. J. M.; Laird, D.; Jia, S.; Williams, S. P.
Adv. Mater. 2010, 22, 3839–3856.
(3) Nelson, J. Mater. Today 2011, 14, 462–470.
(4) Kumar, P.; Chand, S. Prog. Photovolt. Res. Appl. 2012, 20, 377–415.
!
91
(5) Li, G.; Zhu, R.; Yang, Y. Nat. Photonics 2012, 6, 153–161.
(6) Singh, R. P.; Kushwaha, O. S. Macromol. Symp. 2013, 327, 128–149.
(7) Su, Y.-W.; Lan, S.-C.; Wei, K.-H. Mater. Today 2012, 15, 554–562.
(8) Cao, W.; Xue, J. Energy Environ. Sci. 2014, 7, 2123–2144.
(9) Cabanetos, C.; El Labban, A.; Bartelt, J. A.; Douglas, J. D.; Mateker, W. R.;
Fréchet, J. M. J.; McGehee, M. D.; Beaujuge, P. M. J. Am. Chem. Soc. 2013, 135, 4656–
4659.
(10) Chen, H.-C.; Chen, Y.-H.; Liu, C.-C.; Chien, Y.-C.; Chou, S.-W.; Chou, P.-T.
Chem. Mater. 2012, 24, 4766–4772.
(11) Deng, Y.; Liu, J.; Wang, J.; Liu, L.; Li, W.; Tian, H.; Zhang, X.; Xie, Z.; Geng,
Y.; Wang, F. Adv. Mater. 2013, 26, 471–476.
(12) Osaka, I.; Kakara, T.; Takemura, N.; Koganezawa, T.; Takimiya, K. J. Am.
Chem. Soc. 2013, 135, 8834–8837.
(13) Yang, T.; Wang, M.; Duan, C.; Hu, X.; Huang, L.; Peng, J.; Huang, F.; Gong, X.
Energy Environ. Sci. 2012, 5, 8208–8214.
(14) Zhang, M.; Gu, Y.; Guo, X.; Liu, F.; Zhang, S.; Huo, L.; Russell, T. P.; Hou, J.
Adv. Mater. 2013, 25, 4944–4949.
(15) He, Z.; Zhong, C.; Su, S.; Xu, M.; Wu, H.; Cao, Y. Nat. Photonics 2012, 6, 591–
595.
(16) You, J.; Chen, C.-C.; Hong, Z.; Yoshimura, K.; Ohya, K.; Xu, R.; Ye, S.; Gao, J.;
Li, G.; Yang, Y. Adv. Mater. 2013, 25, 3973–3978.
(17) 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.
!
92
(18) Helgesen, M.; Søndergaard, R.; Krebs, F. C. J. Mater. Chem. 2009, 20, 36–60.
(19) Zhan, X.; Zhu, D. Polym. Chem. 2010, 1, 409–419.
(20) Chochos, C. L.; Choulis, S. A. Prog. Polym. Sci. 2011, 36, 1326–1414.
(21) Son, H. J.; He, F.; Carsten, B.; Yu, L. J. Mater. Chem. 2011, 21, 18934–18945.
(22) Lu, L.; Yu, L. Adv. Mater. 2014, 26, 4413–4430.
(23) Duan, C.; Huang, F.; Cao, Y. J. Mater. Chem. 2012, 22, 10416–10434.
(24) Bian, L.; Zhu, E.; Tang, J.; Tang, W.; Zhang, F. Prog. Polym. Sci. 2012, 37,
1292–1331.
(25) Khlyabich, P. P.; Burkhart, B.; Rudenko, A. E.; Thompson, B. C. Polymer 2013,
54, 5267–5298.
(26) Kang, T. E.; Kim, K.-H.; Kim, B. J. J. Mater. Chem. A 2014, 2, 15252–15267.
(27) Osaka, I.; McCullough, R. D. Acc. Chem. Res. 2008, 41, 1202–1214.
(28) Ma, W.; Yang, C.; Gong, X.; Lee, K.; Heeger, A. J. Adv. Funct. Mater. 2005, 15,
1617–1622.
(29) Dang, M. T.; Hirsch, L.; Wantz, G. Adv. Mater. 2011, 23, 3597–3602.
(30) Bundgaard, E.; Krebs, F. C. Sol. Energy Mater. Sol. Cells 2007, 91, 954–985.
(31) 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, 29–54.
(32) Rudenko, A. E.; Thompson, B. C. J. Polym. Sci. Part Polym. Chem. 2015, 53,
135–147.
(33) Wu, P.-T.; Ren, G.; Jenekhe, S. A. Macromolecules 2010, 43, 3306–3313.
(34) Savagatrup, S.; Printz, A. D.; Rodriquez, D.; Lipomi, D. J. Macromolecules
2014, 47, 1981–1992.
!
93
(35) Sivula, K.; Luscombe, C. K.; Thompson, B. C.; Fréchet, J. M. J. J. Am. Chem.
Soc. 2006, 128, 13988–13989.
(36) Thompson, B. C.; Kim, B. J.; Kavulak, D. F.; Sivula, K.; Mauldin, C.; Fréchet, J.
M. J. Macromolecules 2007, 40, 7425–7428.
(37) Burkhart, B.; Khlyabich, P. P.; Thompson, B. C. Macromolecules 2012, 45,
3740–3748.
(38) Campo, B. J.; Bevk, D.; Kesters, J.; Gilot, J.; Bolink, H. J.; Zhao, J.; Bolsée,
J.-C.; Oosterbaan, W. D.; Bertho, S.; D’Haen, J.; Manca, J.; Lutsen, L.; Van Assche, G.;
Maes, W.; Janssen, R. A. J.; Vanderzande, D. Org. Electron. 2013, 14, 523–534.
(39) Campo, B. J.; Oosterbaan, W. D.; Gilot, J.; Cleij, T. J.; Lutsen, L.; Janssen, R. A.
J.; Vanderzande, D. In SPIE 7416, Organic Photovoltaics X; Kafafi, Z. H.; Lane, P. A.,
Eds.; 2009; Vol. 7416, p. 74161G – 74161G – 11.
(40) Suspène, C.; Miozzo, L.; Choi, J.; Gironda, R.; Geffroy, B.; Tondelier, D.;
Bonnassieux, Y.; Horowitz, G.; Yassar, A. J. Mater. Chem. 2012, 22, 4511–4518.
(41) Kozycz, L. M.; Gao, D.; Seferos, D. S. Macromolecules 2013, 46, 613–621.
(42) Shi, C.; Yao, Y.; Yang; Pei, Q. J. Am. Chem. Soc. 2006, 128, 8980–8986.
(43) Xu, B.; Noh, S.; Thompson, B. C. Macromolecules 2014, 47, 5029–5039.
(44) Wang, H.-J.; Chan, L.-H.; Chen, C.-P.; Lin, S.-L.; Lee, R.-H.; Jeng, R.-J.
Polymer 2011, 52, 326–338.
(45) Chang, Y.-T.; Hsu, S.-L.; Su, M.-H.; Wei, K.-H. Adv. Funct. Mater. 2007, 17,
3326–3331.
(46) Chang, Y.-T.; Hsu, S.-L.; Chen, G.-Y.; Su, M.-H.; Singh, T. A.; Diau, E. W.-G.;
Wei, K.-H. Adv. Funct. Mater. 2008, 18, 2356–2365.
!
94
(47) Hollinger, J.; Sun, J.; Gao, D.; Karl, D.; Seferos, D. S. Macromol. Rapid
Commun. 2013, 34, 437–441.
(48) Khlyabich, P. P.; Rudenko, A. E.; Thompson, B. C. J. Polym. Sci. Part Polym.
Chem. 2014, 52, 1055–1058.
(49) Rudenko, A. E.; Khlyabich, P. P.; Thompson, B. C. ACS Macro Lett. 2014, 3,
387–392.
(50) Hou, Q.; Xu, Y.; Yang, W.; Yuan, M.; Peng, J.; Cao, Y. J. Mater. Chem. 2002,
12, 2887–2892.
(51) Zhou, Q.; Hou, Q.; Zheng, L.; Deng, X.; Yu, G.; Cao, Y. Appl. Phys. Lett. 2004,
84, 1653–1655.
(52) Svensson, M.; Zhang, F.; Veenstra, S. c.; Verhees, W. j. h.; Hummelen, J. c.;
Kroon, J. m.; Inganäs, O.; Andersson, M. r. Adv. Mater. 2003, 15, 988–991.
(53) Yang, R.; Tian, R.; Yan, J.; Zhang, Y.; Yang, J.; Hou, Q.; Yang, W.; Zhang, C.;
Cao, Y. Macromolecules 2005, 38, 244–253.
(54) Luo, J.; Hou, Q.; Chen, J.; Cao, Y. Synth. Met. 2006, 156, 470–475.
(55) Xia, Y.; Luo, J.; Deng, X.; Li, X.; Li, D.; Zhu, X.; Yang, W.; Cao, Y. Macromol.
Chem. Phys. 2006, 207, 511–520.
(56) Schulz, G. L.; Holdcroft, S. Chem. Mater. 2008, 20, 5351–5355.
(57) Park, J. S.; Ryu, T. I.; Song, M.; Yoon, K.-J.; Lee, M. J.; Shin, I. A.; Lee, G.-D.;
Lee, J. W.; Gal, Y.-S.; Jin, S.-H. J. Polym. Sci. Part Polym. Chem. 2008, 46, 6175–6184.
(58) Li, K.-C.; Hsu, Y.-C.; Lin, J.-T.; Yang, C.-C.; Wei, K.-H.; Lin, H.-C. J. Polym.
Sci. Part Polym. Chem. 2008, 46, 4285–4304.
(59) Braunecker, W. A.; Oosterhout, S. D.; Owczarczyk, Z. R.; Kopidakis, N.;
!
95
Ratcliff, E. L.; Ginley, D. S.; Olson, D. C. ACS Macro Lett. 2014, 3, 622–627.
(60) Bui, T. T. T.; Jahandar, M.; Song, C. E.; Hoang, Q. V.; Lee, J.-C.; Lee, S. K.;
Kang, I.-N.; Moon, S.-J.; Shin, W. S. Sol. Energy Mater. Sol. Cells 2015, 134, 148–156.
(61) Yu, C.-Y.; Chen, C.-P.; Chan, S.-H.; Hwang, G.-W.; Ting, C. Chem. Mater.
2009, 21, 3262–3269.
(62) Chen, C.-P.; Chan, S.-H.; Chao, T.-C.; Ting, C.; Ko, B.-T. J. Am. Chem. Soc.
2008, 130, 12828–12833.
(63) Chen, C.-H.; Hsieh, C.-H.; Dubosc, M.; Cheng, Y.-J.; Hsu, C.-S.
Macromolecules 2010, 43, 697–708.
(64) Li, J.; Ong, K.-H.; Lim, S.-L.; Ng, G.-M.; Tan, H.-S.; Chen, Z.-K. Chem.
Commun. 2011, 47, 9480–9482.
(65) Li, J.; Ong, K.-H.; Sonar, P.; Lim, S.-L.; Ng, G.-M.; Wong, H.-K.; Tan, H.-S.;
Chen, Z.-K. Polym. Chem. 2013, 4, 804.
(66) Zhao, C.; Chen, X.; Zhang, Y.; Ng, M. J. Polym. Sci. Part Polym. Chem. 2008,
46, 2680–2688.
(67) Chan, L.-H.; Juang, S.-Y.; Chen, M.-C.; Lin, Y.-J. Polymer 2012, 53, 2334–
2346.
(68) Chen, C.-H.; Cheng, Y.-J.; Chang, C.-Y.; Hsu, C.-S. Macromolecules 2011, 44,
8415–8424.
(69) Chen, C.-H.; Cheng, Y.-J.; Dubosc, M.; Hsieh, C.-H.; Chu, C.-C.; Hsu, C.-S.
Chem. – Asian J. 2010, 5, 2483–2492.
(70) Song, S.; Park, S. H.; Jin, Y.; Goo, Y.; Kim, I.; Lee, H.; Lee, K.; Suh, H. Synth.
Met. 2010, 160, 2618–2622.
!
96
(71) Song, S.; Kim, G.; Kang, I.; Jin, Y.; Kim, I.; Kim, J. Y.; Suh, H. J. Polym. Sci.
Part Polym. Chem. 2011, 49, 3751–3758.
(72) Song, S.; Kim, J.; Shim, J.; Kim, J.; Hoon Lee, B.; Jin, Y.; Kim, I.; Lee, K.; Suh,
H. Sol. Energy Mater. Sol. Cells 2012, 98, 323–330.
(73) Chan, L.-H.; Lin, L.-C.; Yao, C.-H.; Liu, Y.-R.; Jiang, Z.-J.; Cho, T.-Y. Thin
Solid Films 2013, 544, 386–391.
(74) Deng, Z.; Chen, L.; Chen, Y. J. Polym. Sci. Part Polym. Chem. 2013, 51, 4885–
4893.
(75) Deng, Z.; Chen, L.; Wu, F.; Chen, Y. J. Phys. Chem. C 2014, 118, 6038–6045.
(76) Ko, S.; Mondal, R.; Risko, C.; Lee, J. K.; Hong, S.; McGehee, M. D.; Brédas,
J.-L.; Bao, Z. Macromolecules 2010, 43, 6685–6698.
(77) Hu, X.; Zuo, L.; Fu, W.; Larsen-Olsen, T. T.; Helgesen, M.; Bundgaard, E.;
Hagemann, O.; Shi, M.; Krebs, F. C.; Chen, H. J. Mater. Chem. 2012, 22, 15710–15716.
(78) Jung, J. W.; Liu, F.; Russell, T. P.; Jo, W. H. Energy Environ. Sci. 2013, 6,
3301–3307.
(79) Zhou, J.; Xie, S.; Amond, E. F.; Becker, M. L. Macromolecules 2013, 46, 3391–
3394.
(80) Li, H.; Liu, F.; Wang, X.; Gu, C.; Wang, P.; Fu, H. Macromolecules 2013, 46,
9211–9219.
(81) Hou, J.; Tan, Z.; Yan, Y.; He, Y.; Yang, C.; Li, Y. J. Am. Chem. Soc. 2006, 128,
4911–4916.
(82) Hou, J.; Tan, Z.; He, Y.; Yang, C.; Li, Y. Macromolecules 2006, 39, 4657–4662.
(83) Zhou, E.; Tan, Z.; Huo, L.; He, Y.; Yang, C.; Li, Y. J. Phys. Chem. B 2006, 110,
!
97
26062–26067.
(84) Liang, Y.; Xiao, S.; Feng, D.; Yu, L. J. Phys. Chem. C 2008, 112, 7866–7871.
(85) He, Y.; Wang, X.; Zhang, J.; Li, Y. Macromol. Rapid Commun. 2009, 30, 45–51.
(86) Huo, L.; Tan, Z.; Wang, X.; Zhou, Y.; Han, M.; Li, Y. J. Polym. Sci. Part Polym.
Chem. 2008, 46, 4038–4049.
(87) Peng, Q.; Park, K.; Lin, T.; Durstock, M.; Dai, L. J. Phys. Chem. B 2008, 112,
2801–2808.
(88) Tsai, J.-H.; Chueh, C.-C.; Chen, W.-C.; Yu, C.-Y.; Hwang, G.-W.; Ting, C.;
Chen, E.-C.; Meng, H.-F. J. Polym. Sci. Part Polym. Chem. 2010, 48, 2351–2360.
(89) Zhu, Z.; Waller, D.; Gaudiana, R.; Morana, M.; Mühlbacher, D.; Scharber, M.;
Brabec, C. Macromolecules 2007, 40, 1981–1986.
(90) Yuan, M.-C.; Chiu, M.-Y.; Chiang, C.-M.; Wei, K.-H. Macromolecules 2010, 43,
6270–6277.
(91) Shi, F.; Fang, G.; Liang, F.; Wang, L.; Mu, Z.; Zhang, X.; Xie, Z.; Su, Z. Eur.
Polym. J. 2010, 46, 1770–1777.
(92) Yao, K.; Chen, L.; Chen, Y.; Li, F.; Ren, X.; Wang, H.; Li, Y. Polym. Chem.
2012, 3, 710–717.
(93) Zhang, M.; Cao, Z.; Wu, F.; Chen, H.; Zhang, M.; Qiao, H.; Tan, S. Eur. Polym.
J. 2014, 57, 83–90.
(94) Zhang, M.; Wu, F.; Cao, Z.; Shen, T.; Chen, H.; Li, X.; Tan, S. Polym. Chem.
2014, 5, 4054–4060.
(95) Huang, Y.; Zhang, M.; Chen, H.; Wu, F.; Cao, Z.; Zhang, L.; Tan, S. J. Mater.
Chem. A 2014, 2, 5218–5223.
!
98
(96) Nielsen, C. B.; Ashraf, R. S.; Schroeder, B. C.; D’Angelo, P.; Watkins, S. E.;
Song, K.; Anthopoulos, T. D.; McCulloch, I. Chem. Commun. 2012, 48, 5832–5834.
(97) Kim, J.-H.; Kim, H. U.; Mi, D.; Jin, S.-H.; Shin, W. S.; Yoon, S. C.; Kang, I.-N.;
Hwang, D.-H. Macromolecules 2012, 45, 2367–2376.
(98) Janietz, S.; Krueger, H.; Schleiermacher, H.-F.; Würfel, U.; Niggemann, M.
Macromol. Chem. Phys. 2009, 210, 1493–1503.
(99) Tamilavan, V.; Song, M.; Ban, T.-W.; Jin, S.-H.; Hyun, M. H. Polym. Bull. 2012,
69, 439–454.
(100) Tamilavan, V.; Song, M.; Jin, S.-H.; Hyun, M. H. Macromol. Res. 2013, 21,
406–413.
(101) Deng, P.; Xiong, J.; Li, S.; Wu, Y.; Yang, J.; Zhang, Q. Chin. J. Chem. 2014, 32,
521–526.
(102) Cabanetos, C.; El Labban, A.; Bartelt, J. A.; Douglas, J. D.; Mateker, W. R.;
Fréchet, J. M. J.; McGehee, M. D.; Beaujuge, P. M. J. Am. Chem. Soc. 2013, 135, 4656–
4659.
(103) Huo, L.; Hou, J. Polym. Chem. 2011, 2, 2453–2461.
(104) He, Z.; Zhong, C.; Huang, X.; Wong, W.-Y.; Wu, H.; Chen, L.; Su, S.; Cao, Y.
Adv. Mater. 2011, 23, 4636–4643.
(105) Liang, Y.; Xu, Z.; Xia, J.; Tsai, S.-T.; Wu, Y.; Li, G.; Ray, C.; Yu, L. Adv. Mater.
2010, 22, E135–E138.
(106) Wang, Z.; Li, Z.; Zhang, H.; Liu, W.; Xu, X.; Si, C.; Cui, Y.; Sun, Q.; Wang, H.;
Shi, F.; Hao, Y. Macromol. Chem. Phys. 2014, 215, 2119–2124.
(107) Jiang, J.-M.; Chen, H.-C.; Lin, H.-K.; Yu, C.-M.; Lan, S.-C.; Liu, C.-M.; Wei,
!
99
K.-H. Polym. Chem. 2013, 4, 5321–5328.
(108) Kang, T. E.; Cho, H.-H.; Kim, H. jun; Lee, W.; Kang, H.; Kim, B. J.
Macromolecules 2013, 46, 6806–6813.
(109) Lim, D. C.; Kim, K.-D.; Park, S.-Y.; Hong, E. M.; Seo, H. O.; Lim, J. H.; Lee, K.
H.; Jeong, Y.; Song, C.; Lee, E.; Kim, Y. D.; Cho, S. Energy Environ. Sci. 2012, 5, 9803–
9807.
(110) Liu, P.; Zhang, K.; Liu, F.; Jin, Y.; Liu, S.; Russell, T. P.; Yip, H.-L.; Huang, F.;
Cao, Y. Chem. Mater. 2014.
(111) He, X.; Mukherjee, S.; Watkins, S.; Chen, M.; Qin, T.; Thomsen, L.; Ade, H.;
McNeill, C. R. J. Phys. Chem. C 2014.
(112) Zhang, G.; Fu, Y.; Qiu, L.; Xie, Z. Polymer 2012, 53, 4407–4412.
(113) Wong, W. W. H.; Subbiah, J.; Puniredd, S. R.; Pisula, W.; Jones, D. J.; Holmes,
A. B. Polym. Chem. 2014, 5, 1258–1263.
(114) Kotowski, D.; Luzzati, S.; Bianchi, G.; Calabrese, A.; Pellegrino, A.; Po, R.;
Schimperna, G.; Tacca, A. J. Mater. Chem. A 2013, 1, 10736–10744.
(115) Chang, W.-H.; Gao, J.; Dou, L.; Chen, C.-C.; Liu, Y.; Yang, Y. Adv. Energy
Mater. 2014, 4, 1300864.
(116) Fang, L.; Zhou, Y.; Yao, Y.-X.; Diao, Y.; Lee, W.-Y.; Appleton, A. L.; Allen, R.;
Reinspach, J.; Mannsfeld, S. C. B.; Bao, Z. Chem. Mater. 2013, 25, 4874–4880.
(117) Zhou, Y.; Kurosawa, T.; Ma, W.; Guo, Y.; Fang, L.; Vandewal, K.; Diao, Y.;
Wang, C.; Yan, Q.; Reinspach, J.; Mei, J.; Appleton, A. L.; Koleilat, G. I.; Gao, Y.;
Mannsfeld, S. C. B.; Salleo, A.; Ade, H.; Zhao, D.; Bao, Z. Adv. Mater. 2014, 26, 3767–
3772.
!
100
(118) Gedefaw, D.; Zhou, Y.; Hellström, S.; Lindgren, L.; Andersson, L. M.; Zhang,
F.; Mammo, W.; Inganäs, O.; Andersson, M. R. J. Mater. Chem. 2009, 19, 5359–5363.
(119) Kim, D. H.; Song, H. J.; Heo, S. W.; Song, K. W.; Moon, D. K. Sol. Energy
Mater. Sol. Cells 2014, 120, Part A, 94–101.
(120) Song, J.; Zhang, C.; Li, C.; Li, W.; Qin, R.; Li, B.; Liu, Z.; Bo, Z. J. Polym. Sci.
Part Polym. Chem. 2010, 48, 2571–2578.
(121) Scaria, R.; Dhawan, S. K.; Chand, S. Synth. Met. 2014, 191, 168–176.
(122) Kozycz, L. M.; Gao, D.; Tilley, A. J.; Seferos, D. S. J. Polym. Sci. Part Polym.
Chem. 2014, 52, 3337-3345.
(123) Zhu, Z.; Waller, D.; Gaudiana, R. J. Macromol. Sci. Part A 2007, 44, 1249–
1253.
(124) Lim, Y.; Ihn, S.-G.; Bulliard, X.; Yun, S.; Chung, Y.; Kim, Y.; Chang, H.; Choi,
Y. S. Polymer 2012, 53, 5275–5284.
(125) Colella, S.; Melcarne, G.; Mazzeo, M.; Gigli, G.; Grisorio, R.; Suranna, G. P.;
Mastrorilli, P. Polymer 2011, 52, 2740–2746.
(126) Duan, C.; Hu, X.; Chen, K.-S.; Yip, H.-L.; Li, W.; Huang, F.; Jen, A. K.-Y.; Cao,
Y. Sol. Energy Mater. Sol. Cells 2012, 97, 50–58.
(127) Jung, I. H.; Kim, S. H.; Jeong, E.; Yang, R.; Lee, K.; Woo, H. Y.; Shim, H.-K. J.
Polym. Sci. Part Polym. Chem. 2011, 49, 1248–1255.
(128) Lange, A.; Krueger, H.; Ecker, B.; Tunc, A. V.; von Hauff, E.; Morana, M. J.
Polym. Sci. Part Polym. Chem. 2012, 50, 1622–1635.
(129) Grisorio, R.; Allegretta, G.; Romanazzi, G.; Suranna, G. P.; Mastrorilli, P.;
Mazzeo, M.; Cezza, M.; Carallo, S.; Gigli, G. Macromolecules 2012, 45, 6396–6404.
!
101
(130) Burkhart, B.; Khlyabich, P. P.; Cakir Canak, T.; LaJoie, T. W.; Thompson, B. C.
Macromolecules 2011, 44, 1242–1246.
(131) Lee, J.-Y.; Choi, M.-H.; Song, H.-J.; Moon, D.-K. J. Polym. Sci. Part Polym.
Chem. 2010, 48, 4875–4883.
(132) Khlyabich, P. P.; Burkhart, B.; Ng, C. F.; Thompson, B. C. Macromolecules
2011, 44, 5079–5084.
(133) Burkhart, B.; Khlyabich, P. P.; Thompson, B. C. ACS Macro Lett. 2012, 1, 660–
666.
(134) Burkhart, B.; Khlyabich, P. P.; Thompson, B. C. J. Photonics Energy 2012, 2,
021002–1.
(135) Burkhart, B.; Khlyabich, P. P.; Thompson, B. C. Macromol. Chem. Phys. 2013,
214, 681–690.
(136) Shaheen, S. E.; Brabec, C. J.; Sariciftci, N. S.; Padinger, F.; Fromherz, T.;
Hummelen, J. C. Appl. Phys. Lett. 2001, 78, 841–843.
(137) Hou, J.; Yang, C.; Qiao, J.; Li, Y. Synth. Met. 2005, 150, 297–304.
(138) Tan, Z.; Tang, R.; Zhou, E.; He, Y.; Yang, C.; Xi, F.; Li, Y. J. Appl. Polym. Sci.
2008, 107, 514–521.
(139) Wen, S.; Pei, J.; Zhou, Y.; Xue, L.; Xu, B.; Li, Y.; Tian, W. J. Polym. Sci. Part
Polym. Chem. 2009, 47, 1003–1012.
(140) Hendriks, K. H.; Heintges, G. H. L.; Wienk, M. M.; Janssen, R. A. J. J. Mater.
Chem. A 2014, 2, 17899–17905.
(141) Paulsen, B. D.; Speros, J. C.; Claflin, M. S.; Hillmyer, M. A.; Frisbie, C. D.
Polym. Chem. 2014, 5, 6287–6294.

102
CHAPTER 2. Fine Tuning of Polymer Properties by
Incorporating Strongly Electron-Donating 3-Hexyloxythiophene
Units into Random and Semi-random Copolymers
2.1 Introduction
As described in Section 1.1, essential to an efficient solar cell is the ideal polymer donor
material, which is generally described as one that has a low band gap (E
g
) and high
absorption coefficient, an effective matching of energy levels with fullerenes, and a suitable
level of ordering as well as moderate miscibility with fullerenes.
1–4
In an effort to pursue the
ideal polymer, numerous alternating D-A copolymers with novel and finely-tailored
monomer structures are synthesized and investigated to tune the properties of such polymers
through “monomer approach”.
3,4
Despite the remarkable improvement in efficiency, these
polymers typically require complex, demanding syntheses of monomers, which is a deviation
from the ultimate vision of a simple, low-cost platform desired for OPV.
In contrast, polymer properties can be tuned by the strategic combination of simple
monomers through the so-called “polymer approach”.
4
As summarized in Chapter 1, a
considerable number of random and semi-random copolymers have proved to be quite
successful, giving significantly broadened absorption band and suitable energy levels.
Thanks to the convenience in tuning the properties of random/semi-random copolymers,
some of them have reached an efficiency around 7% in BHJ solar cells with fullerene

103
derivatives.
5–7
Specifically, our group has introduced a class of semi-random copolymers
based on regioregular poly(3-hexylthiophene) (rr-P3HT), characterized by a small amount of
acceptor monomer, typically 5-15%, being copolymerized in a randomized fashion with 2-
bromo-5-trimethyltin-3-hexylthiophene and 2,5-bis(trimethyltin)thiophene.
8–12
These
polymers possess restricted linkage patterns (due to rational choice of functional groups)
where discrete acceptor units are randomly distributed along the polymer backbone, among
head-to-tail linked 3-hexythiophene (3HT) segments. As such, the semi-random copolymers
present a simple perturbation of the rr-P3HT structure but maintain critical elements of the
polymeric structure, and thus are capable of combining the advantages of rr-P3HT (high hole
mobility, semi-crystallinity, and effective miscibility with PCBM) with truly broadened
absorption profiles, instead of red-shifted absorption bands typical of alternating donor-
acceptor copolymers.
The potential of this new class of semi-random polymers has been demonstrated in the
studies of poly(3-hexylthiophene-thiophene-diketopyrrolopyrrole) P3HTT-DPP and some
so-called two-acceptor polymers containing two distinct acceptor units, where efficiencies as
high as 5-6% were achieved.
10,11
Moreover, investigations into semi-random copolymers have
shed light on our fundamental understanding of the structure-property relationships of
polymers, in terms of band gaps, HOMO and LUMO energy levels, semi-crystallinity as well
as thermal properties. So far, most studies on semi-random copolymers have primarily
focused on strong electron acceptor units. However, the simultaneous incorporation of both
electron donors and acceptors into the rr-P3HT backbone is also of interest. With both
electron donors and acceptors being embedded in polymer backbones, more pronounced D-A
interactions can be envisioned, tailoring band gaps and HOMO/LUMO energy levels. It is of

104
fundamental significance to explore how such structural features and how specific monomer
choices affect the function of semi-random copolymers in terms of band gaps, energy levels,
hole mobility and/or semi-crystallinity. It is also interesting to test the range of control over
HOMO levels of semi-random copolymers afforded by incorporating small amounts of
strongly electron-rich units. In addition to relevance in solar cells, the possibly high HOMO
levels of the resulting polymers may make them potential candidates as electrochromics or
conductive components in coatings and bulk plastics as a consequence of low oxidation
potentials and highly stable conducting states.
13

Only one previous example of semi-random copolymers containing both strong donors and
strong acceptors has been reported. In this case, the strongly electron-donating
dithienopyrrole (DTP) was copolymerized with various electron acceptors including
diketopyrrolopyrrole (DPP).
12
(Figure 2.1a) The synthetic strategy involved functionalization
of both the DTP comonomer and the electron-accepting comonomers with bromine. As such,
DTP and the electron acceptors could not bond to each other, excluding direct D-A
interactions. As a result, the semi-random copolymers containing DTP exhibited slightly
reduced band gaps and a small decrease in V
oc
corresponding to raised HOMO levels, relative
to those containing only acceptor units. Although these results demonstrated the capability of
tuning polymer properties by introducing both electron donors and acceptors to semi-random
copolymers, the structural restrictions led to limited scope of structure-function relationships.
Further exploration will clearly require direct bonding of the strong donor with acceptors so
as to facilitate maximal D-A interactions.

105

Figure 2.1. Syntheses and Structures of (a) P3HTT-DPP-DTP (reported previously
12
) and (b)
P3HTT-HOT-DPP (reported here).
Here we report the syntheses and characterization of a series of rr-P3HT based random and
semi-random copolymers, where a portion of the 3HT component is replaced by strongly
electron-donating 3-hexyloxythiophene (3HOT). Previously, 3-alkyloxythiophenes (3AOT)
have been introduced into polymer backbones by several research groups. Specifically,
successful synthesis of rr-P3AOT was reported by McCullough and Koeckelberghs
independently in 2005 using GRIM polymerization and the so-called “reversed McCullough”
method.
14–16
Mechanistic study of the regioregular polymerization of 3AOT was carried out
later.
17
When 3AOT was introduced into polymers, lower band gaps and significantly raised
HOMO levels were consistently seen relative to the 3-alkylthiophene (3AT) based
analogs.
15,16,18–21
Specifically, the regioregular homopolymer of 3-decyloxythiophene was
reported to have a HOMO level 0.28V higher than rr-P3HT and a much lower bandgap of
1.60 eV as opposed to the 1.92 eV of rr-P3HT.
22
However, the effect of 3AOT in OPV
applications has not been widely explored.
22–25
Further, it still remains a synthetic challenge
to regio-specifically functionalize 3AOT, which restricts the way that 3AOT can be

106
copolymerized with other monomers.
25
As a consequence, the existing 3AOT-containing
copolymers investigated in OPV are either of the alternating D-A type,
23,25
or with regio-
random linkages of 3AOT,
24
except for a regioregular random copolymer poly(3-
octylthiophene-2,5-diylco-3-decyloxythiophene-2,5-diyl) (POT-co-DOT) reported by Shi
and coworkers in 2006.
22
As such, we have great interest in introducing 3AOT into more
complex copolymer backbones in a regioregular manner, particularly into the class of semi-
random copolymers, where small amounts of 3AOT promise rational control over polymer
electronic properties. Compared to the DTP-containing semi-random copolymers previously
reported by our group, this work not only introduces the potential for bonding between the
electron donor 3HOT and the electron acceptor DPP, but also minimizes the structural
disorder due to remarkable resemblance between 3HT and 3HOT as well as good control
over regioregularity. (Figure 2.1b) This work should provide valuable insight into the
structure-property-performance relationships of random and semi-random copolymers.
2.2 Monomer and polymer syntheses
Regio-specific functionalization of 3-hexyloxythiophene (3HOT) is critical to develop a
universal platform toward regioregular copolymerizations. At the same time, our established
Stille polymerization is ideal for producing semi-random polymers with head-to-tail 3HT
segments and randomly distributed acceptor units. Therefore, 2-bromo-5-trimethyltin-3-
hexyloxythiophene (compound 4, Figure 2.2a) is desired as a comonomer, analogous to the
2-bromo-5-trimethyltin-3-hexylthiophene monomer for rr-P3HT, to retain favorable features
of semi-random configuration and allow bonding between 3HOT and the chosen acceptor
DPP.

107
The compound 3-hexyloxythiophene (2) was readily brominated at the 2-position using
NBS (Figure 2.2a) as described before.
20
However, standard conditions to stannylate the 5-
position of 2-bromo-3-hexylthiophene using lithium diisopropylamide (LDA) failed to
translate into the alkyloxy analog, consistent with what had been observed by others.
25
A
mixture of 2- and 5-stannylated 3HOT was obtained, resulting from the lithium-halogen
exchange competing with the desired deprotonation at 5-position. In order to solve this
problem, we employed a commercially available solution of 2,2,6,6-
tetramethylpiperidinylmagnesium chloride lithium chloride complex (TMPMgCl·LiCl)
instead of LDA. This so-called Knochel-Hauser Base was reported in 2006 and proved to
have high functional group tolerance and regioselectivity in metalation of arenes and
heteroarenes.
28–31
Employment of this base successfully produced the desired monomer 4
(Figure 2.2a) with high purity, which could be further purified via vacuum distillation and
isolated in 94% yield.

Figure 2.2. Synthetic Scheme of (a) Monomer 2-Bromo-5-trimethyltin-3-hexyloxythiophene
(4), (b) Stille Polymerization for Semi-random P3HTT-HOT-DPP Copolymers and (c) Stille
Polymerization for Random P3HT-co-3HOT Copolymers.

108
Monomer 4 was then copolymerized via the established Stille polymerization method with
other comonomers as presented in Figure 2.2b to afford a series of novel semi-random
poly(3-hexylthiophene-thiophene-3-hexyloxythiophene-diketopyrrolopyrrole) copolymers
referred to as P3HTT-HOT-DPPs. The compositions of non-alkylated thiophene and DPP
units were held constant at 10% each, consistent with the best performing DPP-containing
semi-random copolymer reported before (previously defined as P3HTT-DPP-10% and
referred to simply as P3HTT-DPP here).
11
The ratio of 3HT and 3HOT, however, was varied
in a wide range and indicated in the polymer names within parentheses. The parent
copolymer P3HTT-DPP containing no 3HOT was also synthesized for comparison. In
addition, a set of random copolymers comprising only 3HT and 3HOT, named as P3HT
x
-co-
3HOT
y
, were synthesized as well in order to further elucidate the intrinsic properties of
3HOT-containing polymers. (Figure 2.2c) These random copolymers are analogous to the
random poly(3-hexylthiophene)-co-(3-cyanothiophene)
31,32
and poly(3-hexylthiophene)-co-
(3-(2-ethylhexyl)thiophene)
33
copolymers we have previously reported, along with the
poly(3-hexylthiophene)-stat-(3-thiohexylthiophene) copolymers reported by Seferos, et al.
34
,
all of which have demonstrated how small changes in composition could have large
influences on copolymer energy levels and device parameters. Homopolymers P3HT and
P3HOT were also synthesized with the same method.

109
Table 2.1. Molecular Weight and Electronic Properties of All Polymers.
Polymer M
n
(kg/mol) /PDI
a
E
g
(nm/eV)
b
HOMO (eV)
c
P3HT 16.1/2.46 650/1.91 -5.18
P3HT
90
-co-3HOT
10
19.5/2.11 670/1.85 -5.13
P3HT
75
-co-3HOT
25
19.4/2.56 700/1.77 -5.07
P3HT
50
-co-3HOT
50
11.9/3.00 730/1.70 -4.96
P3HOT 2.0/1.41 780/1.59 -4.73
P3HTT-DPP 12.7/2.56 830/1.49 -5.20
P3HTT-HOT-DPP(75:5) 14.7/3.06 850/1.46 -5.14
P3HTT-HOT-DPP(70:10) 18.5/2.66 860/1.44 -5.11
P3HTT-HOT-DPP(65:15) 14.7/2.90 870/1.43 -5.07
P3HTT-HOT-DPP(60:20) 15.8/2.59 880/1.41 -4.98
P3HTT-HOT-DPP(40:40) 5.3/3.82 920/1.35 -4.93
a
Determined by SEC with polystyrene as standard and o-DCB as eluent.
b
Optical band
gaps from the onset of absorption in UV−Vis spectra of annealed films.
c
Determined by
cyclic voltammetry (vs Fc/Fc
+
) in acetonitrile containing 0.1 M TBAPF
6
.
Molecular weights of the copolymers were analyzed by SEC in o-dichlorobenzene (o-DCB)
against polystyrene standards. Most copolymers have molecular weights above 12,000 g/mol
as shown in Table 2.1. For P3HTT-HOT-DPP(40:40), the molecular weight is dramatically
lower, which may result from the observed poor solubility of this polymers and early
precipitation from the reaction mixture. For the case of P3HOT, there is no clear explanation
for the low molecular weight and repeated polymerization attempts did not result in higher
molecular weights. We have not tried other polymerization methods, as we preferred
consistency to prepare all polymers under the same conditions to avoid introducing other
potential variations in structure.
1
H NMR spectra of the random and semi-random
copolymers verify the comonomer compositions are consistent with the monomer feed ratio.
(Figure A1.6-A1.16) This is in stark contrast to the previously reported random copolymer of
3-octylthiophene and 3-decyloxythiophene referred to as POT-co-DOT prepared by the
GRIM polymerization method.
22
It was stated that the molar ratio of 3-decyloxythiophene to
3-octylthiophene units in POT-co-DOT was 2.5:1 when they were fed in ratio of 1:1 for
GRIM copolymerization. The observed discrepancy likely indicates a dramatic difference in

110
reactivity of the intermediate Grignard species when alkyl side chains were replaced by
alkyloxy groups. In this context, the Stille method used here shows a significant advantage in
producing random copolymers with controlled monomer composition. In addition to
verifying the composition of the copolymers,
1
H NMR also indicates that P3HT-co-3HOT
copolymers are not block copolymers or mixtures of homopolymers, but do indeed contain
heterocoupled linkages as evidenced by a peak at ~7.07 ppm that is unique relative to either
homopolymer.
35,36
Additionally, high regioregularity is retained for all copolymers as
evidenced by the integrated peak areas corresponding to H-T linkages of 3HT at 2.81 ppm
and H-H linkages at 2.58 ppm. The small peak at 6.08 ppm in the
1
H NMR of P3HOT
corresponds to the peak previously assigned to protons at the 2-position of 3-
alkyloxythiophene rings at polymer chain ends.
37
Integration of this peak suggests a relatively
low molecular weight of the polymer, which is consistent with the result from SEC
measurement.
2.3 Electronic properties
UV-Vis absorption profiles of P3HT-co-3HOT and P3HTT-HOT-DPP copolymers were
obtained in both solutions and annealed thin films. The thin film data are presented in Figure
2.3. (Summarized data is available in the Appendix 1, Table A1.1.) As shown in Figure 2.3a,
the random copolymer P3HT
90
-co-3HOT
10
has the same peak absorption coefficient and the
same absorption maxima as the parent P3HT, whereas a small red-shift of 20 nm can be seen
in the absorption onset and thereby a slight broadening of the band. At the same time, the
vibronic shoulder associated with semi-crystallinity and ordering in the solid state is less
pronounced. When the amount of 3HOT further increases in these random copolymers, the

111
red-shift of the absorption onset and absorption band broadening becomes more prominent,
while the vibronic shoulder completely disappears. A continuous decrease in optical band
gap is observed for random P3HT-co-3HOT copolymers with increasing 3HOT content,
without decreasing the absorption coefficient. With 50% of 3HOT incorporated, the
copolymer P3HT
50
-co-3HOT
50
shows a band gap of 1.7 eV, which is 0.2 eV smaller than
that of P3HT. The homopolymer P3HOT has the most red-shifted absorption peak, which
appears in similar region as observed before for regioregular poly(3-octyloxythiophene) and
poly(3-decyloxythiophene),
22,38
and is red-shifted compared to regio-irregular poly(3-
hexyloxythiophene) synthesized by oxidative polymerization with FeCl
3
.
18
However, the
absorption coefficient of P3HOT is about 40% lower than other polymers. This could
possibly be due to its low molecular weight, as molecular weight is known to influence peak
absorption coefficient in several cases
39,40
, or an intrinsic decrease in oscillator strength for
the P3HOT polymer relative to P3HT. In addition, no vibronic shoulder can be observed for
P3HOT, as opposed to what Koeckelberghs et al. reported for highly regioregular (RR
~100%) poly(3,7-dimethyloctyloxythiophene) with M
n
of 13.5 kg/mol.
15
This discrepancy
may result from a slightly lower degree of regioregularity (RR ~90%) or lower molecular
weight of P3HOT synthesized here compared to the polymer from Koeckelberghs. Similar
trends in absorption profile with increasing 3HOT content can also be seen in solution.
(Supporting Information, Figure A1.17a) Absorption maxima and onsets in thin films are all
red-shifted relative to the solution data, which indicates planarization of the polymer
backbones in thin films.

112

Figure 2.3. UV−vis absorption of (a) random P3HT-
co-3HOT copolymers and (b) semi-random P3HTT-
HOT-DPP copolymers in thin films spin-coated from
o-DCB and annealed for 30 min under N
2
at 150 °C
(P3HOT thin film is as-cast): (i) P3HT, (ii) P3HT
90
-
co-3HOT
10
, (iii) P3HT
75
-co-3HOT
25
, (iv) P3HT
50
-co-
3HOT
50
, (v) P3HOT, (vi) P3HTT-DPP, (vii)
P3HTT-HOT-DPP(75:5), (viii) P3HTT-HOT-
DPP(70:10), (ix) P3HTT-HOT-DPP(65:15), (x)
P3HTT-HOT-DPP(60:20) and (xi) P3HTT-HOT-
DPP(40:40).
When 3HOT units are incorporated into semi-random P3HTT-HOT-DPP copolymers, the
characteristic dual band absorption of P3HTT-DPP in both solution and thin film is retained.
This pattern was attributed to π-π* transitions of thiophene-rich segments (short wavelength

113
band) and intramolecular charge transfer (ICT) transitions of segments containing D-A
linkages (long wavelength band), as can be seen in Figure 2.3b.
8
Similar to random
copolymers, an increasing content of 3HOT in semi-random copolymers leads to a more red-
shifted absorption onset and broadened absorption profile. The effect of 3HOT in lowering
optical band gaps in D-A copolymers is consistent with previous literature,
20,21,23–25
and the
trend can be clearly seen as summarized in Table 2.1. Moving from solutions (Appendix 1,
Figure A1.17b) to thin films, the semi-random copolymers also experience a red-shift of
absorption bands, indicating coplanarization of polymer backbones. A closer look reveals a
much smaller red-shift of the absorption onset for P3HTT-HOT-DPP(40:40) (50 nm)
compared to P3HTT-DPP and other semi-random copolymers (around 90 nm) when
comparing the solution and solid film spectra of each polymer. This suggests a certain degree
of polymer aggregation in the solution of P3HTT-HOT-DPP(40:40), which is consistent
with our hypothesis that poor solubility of this polymer limited its molecular weight.
In order to learn about the influence of 3HOT unit on the semi-crystallinity of copolymers,
grazing-incidence X-ray diffraction (GIXRD) was used. As can be seen in Figure 2.4, all
polymers show peaks within the range of 2θ = 5°-6°, which correlate with the (100) planes of
semi-crystalline polythiophenes. (Summary of XRD data is available in Appendix 1, Table
A1.2.) In the set of random copolymers, lower intensity of the diffraction peak is observed
with an increasing amount of 3HOT, which suggests a decreasing degree of crystallinity.
(Figure 2.4a) This observation is consistent with the disappearing vibronic shoulder in thin
film absorption profiles. Compared to P3HT, which has a lamellar distance of 16.71 Å,
P3HT
90
-co-3HOT
10
and P3HT
75
-co-3HOT
25
show slightly larger lamellar spacing in the
(100) direction, which are calculated to be 16.94 Å and 17.23 Å respectively. This is likely

114
caused by the introduction of slightly longer hexyloxy side chains. At this low content of
3HOT (no more than 25%), the molecular packing in copolymer films is expected to
resemble that of rr-P3HT, but with more interplane space needed for the slightly longer
hexyloxy chains, resulting in an increase in lamellar distances. In contrast, the stacking
distance of P3HT
50
-co-3HOT
50
is determined to be 15.74 Å, much lower than the other
polymers. This suggests significant perturbation to the packing mode of rr-P3HT induced by
the hexyloxy side chains. The disappearance of diffraction peak from the homopolymer
P3HOT likely further suggests the disruption to semi-crystallinity induced by a large amount
of hexyloxy groups. To be noted, the regio-irregular P3HOT prepared by Hu and Xu via
oxidative polymerization with FeCl
3
showed an X-ray diffraction peak at 2θ = 4.5°,
corresponding to a lamellar distance of 19.8 Å.
18
The origin of this discrepancy is unclear.

115

Figure 2.4. Grazing-Incidence X-ray Diffraction of
thin films of (a) random P3HT-co-3HOT copolymers
and (b) semi-random P3HTT-HOT-DPP copolymers
(spin-coated from o-DCB, as cast for P3HOT and
annealed for 30 min under N
2
at 150 °C for other
polymers): (i) P3HT, (ii) P3HT
90
-co-3HOT
10
, (iii)
P3HT
75
-co-3HOT
25
, (iv) P3HT
50
-co-3HOT
50
, (v)
P3HOT, (vi) P3HTT-DPP, (vii) P3HTT-HOT-
DPP(75:5), (viii) P3HTT-HOT-DPP(70:10), (ix)
P3HTT-HOT-DPP(65:15), (x) P3HTT-HOT-
DPP(60:20) and (xi) P3HTT-HOT-DPP(40:40).
As for the set of semi-random P3HTT-HOT-DPP copolymers, the peak position and
intensity are not significantly changed compared to P3HTT-DPP. (Figure 2.4b) The lamellar
spacing gradually becomes larger as the 3HOT content increases, from 15.76 Å for P3HTT-

116
DPP to 16.01 Å for P3HTT-HOT-DPP(40:40). Nevertheless, this shift is quite subtle
compared to random P3HT-co-3HOT copolymers. This probably indicates that the solid-
state packing of DPP-containing semi-random copolymers is dominated by the DPP unit, and
not much affected by changes in the segments that are rich in 3-substituted thiophenes.
In addition to the optical properties described above, the electronic properties of the
polymers, including HOMO levels and hole mobility, are also investigated. HOMO levels of
all polymers were measured by cyclic voltammetry (CV) with ferrocene as reference and the
results are summarized in Table 2.1. Incorporation of 3HOT into random P3HT-co-3HOT
copolymers

affords significantly higher HOMO levels compared to P3HT and the rise in
HOMO level closely correlates with the 3HOT content in copolymers. The homopolymer
P3HOT exhibits a HOMO level of -4.73 eV, which is 0.45 eV above that of P3HT. The
effect of 3AOT in raising polymer HOMO levels relative to 3-alkylthiophene-based
polymers has been reported with different polymer structures,
20–25
and is not difficult to
understand considering the highly electron-rich nature of 3HOT. Interestingly, our previous
work revealed that the strong acceptor DPP would override the effect of electron donor DTP
in P3HTT-DPP-DTP, preventing its HOMO levels from rising although a strongly electron-
rich unit DTP was incorporated.
12
Opposed to this previous observation, the HOMO levels of
3HOT-containing semi-random P3HTT-HOT-DPP copolymers are affected substantially by
the content of 3HOT and could be raised by almost 0.3 eV with no more than 40% 3HOT
loading. This dramatic difference in polymer electronic property may arise from the fact that
DTP could not bond directly to DPP in P3HTT-DPP-DTP due to the choice of functional
groups in the monomers, while linkage between 3HOT and DPP is possible in the semi-
random P3HTT-HOT-DPP copolymers. The results here demonstrate that introducing

117
3HOT into polymer backbones with control over monomer composition can be a good way
to perform rational tuning of polymer energy levels.
2.4 Solar cell characterization
The photovoltaic properties of the random P3HT-co-3HOT

copolymers and the semi-
random P3HTT-HOT-DPP copolymers in BHJ OPV devices were studied in a conventional
device configuration of ITO/PEDOT:PSS/polymer:PC
61
BM/Al. All of the devices were
individually optimized and the average parameters of the devices, namely J
sc
, V
oc
, FF, and
PCE, are summarized in Table 2.2. The results from P3HT:PC
61
BM and P3HTT-
DPP:PC
61
BM are also shown as a reference.
Table 2.2. Average Photovoltaic Parameters and SCLC Hole Mobilities of Random P3HT-
co-3HOT and Semi-random P3HTT-HOT-DPP Copolymers.
a
Polymer:PC
61
BM
J
sc
(mA/cm
2
)
b

V
oc
(V)
FF
PCE
(%)
b

SCLC Hole
Mobility (cm
2
/Vs)
e
P3HT

c
9.85 0.62 0.55 3.36 6.5×10
-4

P3HT
90
-co-3HOT
10
c
5.28 0.43 0.41 1.04 3.0×10
-4

P3HT
75
-co-3HOT
25
d
3.62 0.37 0.45 0.60 6.8×10
-5

P3HT
50
-co-3HOT
50
d
1.34 0.21 0.41 0.11 1.0×10
-5

P3HOT
d
0.66 0.024 0.28 0.0045 1.8×10
-4

P3HTT-DPP
d
12.78 0.61 0.59 4.58 6.8×10
-4

P3HTT-HOT-DPP(75:5)
d
12.64 0.53 0.57 3.79 2.8×10
-4

P3HTT-HOT-DPP(70:10)
d
11.02 0.45 0.53 2.64 2.0×10
-4

P3HTT-HOT-DPP(65:15)
d
10.92 0.40 0.48 2.11 7.8×10
-5

P3HTT-HOT-DPP(60:20)
d
8.65 0.37 0.47 1.51 2.3×10
-5

P3HTT-HOT-DPP(40:40)
d
4.89 0.23 0.36 0.40 4.4×10
-5

a
Average of 4 to 7 pixels in the best working device.
b
Mismatch corrected.
41

c
The optimized
ratio of polymer:PC
61
BM is 1:0.8.
d
The optimized ratio of polymer:PC
61
BM is 1:1.
e
Measured from pristine polymer films in their optimized annealing condition.
A clear trend is that the V
oc
of the random and semi-random copolymers are observed to
regularly decrease as the content of 3HOT unit increases. With only 10% of 3HOT
incorporated in P3HT
90
-co-3HOT
10
, the V
oc
drops by almost 0.2 V compared to P3HT, from

118
0.62 V to 0.43 V. The copolymers P3HT
75
-co-3HOT
25
and P3HT
50
-co-3HOT
50
exhibit
further decreased V
oc
of 0.37 V and 0.21 V respectively. The homopolymer P3HOT shows
an extremely low V
oc
of 0.024 V, which is consistent with the value of 0.02 V reported
previously for poly(3-decyloxythiophene).
22
As for the semi-random P3HTT-HOT-DPP
copolymers, the V
oc
also decreases continuously with an increasing content of 3HOT. The
value reaches 0.23 V when the copolymer contains 40% of 3HOT, as opposed to the 0.61 V
of the parent P3HTT-DPP.
The trend in V
oc
upon 3HOT incorporation mentioned above is clearly depicted and
compared with the change in HOMO levels in Figure 2.5, where the V
oc
and HOMO levels
are plotted as a function of 3HOT content. For both sets of copolymers, the regular decrease
in V
oc
is well correlated with the rising HOMO levels as the content of 3HOT goes up, further
indicating that the composition of 3HOT comonomer in the polymer chain is reflected in the
electronic properties of the polymers. However, the change in V
oc
is slightly more
pronounced than that in HOMO energy. To be specific, V
oc
drops by 0.59 V from P3HT to
P3HOT, whereas the HOMO level rises by only 0.45 eV, from -5.18 eV for P3HT to -4.73
eV for P3HOT. Similarly, the V
oc
of P3HTT-HOT-DPP(40:40) is 0.38 V lower than
P3HTT-DPP, while the rise in HOMO level is only 0.27 eV. This behavior resembles our
previous observation with rr-poly(3-hexylthiophene-co-3-(2-ethylhexyl)thiophene).
33
In this
earlier work, we reported the trend of increasing V
oc
as branched 2-ethylhexyl side chains
replaced some hexyl chains in rr-poly(3-hexylthiophene), correlating with the decreasing
HOMO levels of the copolymers. Interestingly, the increase in V
oc
was slightly less than the
decrease in HOMO levels in the previous case. Nonetheless, in the previous case and the
present case, regular variation in the HOMO energy with comonomer content is reflected by

119
a changing HOMO
DONOR
−LUMO
ACCEPTOR
offset and ultimately correlated to the changing
V
oc
.
42,43

Figure 2.5. HOMO levels in the solid state (filled
squares) and V
oc
(open circles) of the optimized solar
cells of (a) random P3HT-co-3HOT copolymers
and (b) semi-random P3HTT-HOT-DPP
copolymers as a function of 3HOT content in the
polymer backbone.

Importantly, the J
sc
of semi-random P3HTT-HOT-DPP copolymers with low 3HOT
monomer content (up to 15%) is above 10 mA/cm
2
, comparable to that of the P3HTT-
DPP:PC
61
BM reference device and superior to that of the previously studied semi-random
copolymer P3HTT-DPP-DTP.
12
At the same time, the FF of these devices are around or

120
above 0.5, leading to efficient solar cells with PCE of above 2%. However, when the
composition of 3HOT unit further increases to above 20%, the photovoltaic parameters start
to deteriorate for semi-random copolymers. As for the random P3HT-co-3HOT copolymers,
both J
sc
and FF drop drastically as soon as 3HOT is introduced and further deterioration in J
sc

is observed as the amount of 3HOT increases. Such poor performance of the random
copolymers is likely caused by the unfavorable morphology of polymer:fullurene blends, as
inefficient polymer chain packing is suggested by their lower diffraction peak intensity in
GIXRD characterization. The very low FF for P3HOT could be a consequence of low
molecular weight, which is known to negatively influence device performance.
39,40,44
The very
low FF observed for P3HTT-HOT-DPP(40:40) can potentially be ascribed to the low
molecular weight analogous to P3HOT. While no evidence exists to support that these
polymers suffer from oxidative instability on the time-frame of device fabrication and testing,
the influence of possible polymer oxidation in air with increasing 3HOT content (and
correspondingly raised HOMO energies) cannot be totally ruled out as a contributing factor
to decreasing device performance.
External quantum efficiency measurements were conducted in order to study the
photocurrent response of the solar cells based on semi-random copolymers. As illustrated in
Figure 2.6, at low 3HOT composition (up to 10%), the semi-random copolymers have broad
and strong photoresponse that is comparable to that of the P3HTT-DPP:PC
61
BM device. As
the content of 3HOT increases, the photoresponse at longer wavelength around 850 nm
becomes evident, but its contribution does not compensate the significant decrease in the
external quantum efficiency within the range of 450 nm to 825 nm. As a consequence,

121
significantly lower J
sc
is observed from the solar cells based on copolymers that contain high
content of 3HOT.

Figure 2.6. EQE of the BHJ solar cells based on
P3HTT-DPP (black squares), P3HTT-HOT-
DPP(75:5) (red circles), P3HTT-HOT-DPP(70:10)
(green triangles), P3HTT-HOT-DPP(65:15) (blue
triangles), P3HTT-HOT-DPP(60:20) (purple stars),
and P3HTT-HOT-DPP(40:40) (pink triangles) with
PC
61
BM as the acceptor, under optimized condition for
device fabrication.
Hole mobilities of the polymers were determined using the space-charge limited current
(SCLC) method. The results shown in Table 2.2 indicate, in general, that the hole mobility
decreases as the composition of the 3HOT monomer increases, which also helps to explain
the deterioration of the photovoltaic properties. However, with a small amount of 3HOT
incorporated (up to 10%), the mobility of copolymers is still very close to that of P3HT or
P3HTT-DPP reference.

122
2.5 Conclusion
The strongly electron-donating 3HOT monomer was functionalized in a regio-specific
manner for Stille polymerizations. Starting from rr-P3HT and previously reported semi-
random P3HTT-DPP, 3HOT was copolymerized at various contents in the polymer
backbones. The rational choice of functional groups not only ensures the retention of
regioregular linkage patterns, but also allows possible bonding between the electron donor
3HOT and DPP acceptor in the semi-random P3HTT-HOT-DPP copolymers so that
maximal effect from 3HOT even at low contents can be exploited due to the direct D-A
interactions in polymer backbone.
It is demonstrated that the introduction of 3HOT leads to smaller band gaps, broader
absorption spectra and higher HOMO energy levels of the resulting copolymers. By varying
the content of 3HOT units, we have obtained rational control over these electronic properties,
and to a substantial extent on HOMO levels. Importantly, the copolymers are strongly
absorbing and semi-crystalline materials, which is relevant for achieving efficient charge
generation and favorable morphology in OPV devices. Polymer:PC
61
BM solar cell devices
based on 3HOT-containing copolymers provide V
oc
values that are tunable over hundreds of
mV’s with varied 3HOT content, which closely correlate with the measured HOMO energy
levels. With low content of 3HOT (no more than 15%), the semi-random P3HTT-HOT-DPP
copolymers can work as efficient donor materials in polymer:PC
61
BM solar cells, affording
high J
sc
of above 10 mA/cm
2
and FF of around or above 0.5.
This work reveals valuable information on the structure-function relationship in electron-
rich random and semi-random copolymers. The results suggest that the incorporation of

123
small amounts of strongly electron-donating 3HOT units has remarkable impact on the
electronic properties of conjugated polymers, which provides a practical way to rationally
tune polymer properties. Furthermore, this work is another demonstration of the “polymer
approach” for designing conjugated polymers that display highly-tailored properties
achievable from the strategic combination of relatively simple monomers.
2.6 References
(1) Li, G.; Zhu, R.; Yang, Y. Nat. Photonics 2012, 6, 153–161.
(2) Thompson, B. C.; Fréchet, J. M. J. Angew. Chem. Int. Ed. 2008, 47, 58–77.
(3) Li, Y. Acc. Chem. Res. 2012, 45, 723–733.
(4) Khlyabich, P. P.; Burkhart, B.; Rudenko, A. E.; Thompson, B. C. Polymer 2013, 54,
5267–5298.
(5) Lim, D. C.; Kim, K.-D.; Park, S.-Y.; Hong, E. M.; Seo, H. O.; Lim, J. H.; Lee, K. H.;
Jeong, Y.; Song, C.; Lee, E.; Kim, Y. D.; Cho, S. Energy Environ. Sci. 2012, 5, 9803–9807.
(6) Liu, P.; Zhang, K.; Liu, F.; Jin, Y.; Liu, S.; Russell, T. P.; Yip, H.-L.; Huang, F.; Cao,
Y. Chem. Mater. 2014.
(7) Chang, W.-H.; Gao, J.; Dou, L.; Chen, C.-C.; Liu, Y.; Yang, Y. Adv. Energy Mater.
2014, 4, 1300864.
(8) Burkhart, B.; Khlyabich, P. P.; Cakir Canak, T.; LaJoie, T. W.; Thompson, B. C.
Macromolecules 2011, 44, 1242–1246.
(9) Burkhart, B.; Khlyabich, P. P.; Thompson, B. C. Macromol. Chem. Phys. 2013, 214,
681–690.
(10) Burkhart, B.; Khlyabich, P. P.; Thompson, B. C. ACS Macro Lett. 2012, 1, 660–666.

124
(11) Khlyabich, P. P.; Burkhart, B.; Ng, C. F.; Thompson, B. C. Macromolecules 2011, 44,
5079–5084.
(12) Burkhart, B.; Khlyabich, P. P.; Thompson, B. C. J. Photonics Energy 2012, 2,
021002–1.
(13) Skotheim, T. A.; Reynolds, J. Conjugated Polymers: Theory, Synthesis, Properties,
and Characterization; CRC Press, 2006.
(14) Sheina, E. E.; Khersonsky, S. M.; Jones, E. G.; McCullough, R. D. Chem. Mater.
2005, 17, 3317–3319.
(15) Koeckelberghs, G.; Vangheluwe, M.; Samyn, C.; Persoons, A.; Verbiest, T.
Macromolecules 2005, 38, 5554–5559.
(16) Koeckelberghs, G.; Vangheluwe, M.; Doorsselaere, K. V.; Robijns, E.; Persoons, A.;
Verbiest, T. Macromol. Rapid Commun. 2006, 27, 1920–1925.
(17) Van den Bergh, K.; De Winter, J.; Gerbaux, P.; Verbiest, T.; Koeckelberghs, G.
Macromol. Chem. Phys. 2011, 212, 328–335.
(18) Hu, X.; Xu, L. Polymer 2000, 41, 9147–9154.
(19) He, J.; Su, Z.; Yan, B.; Xiang, L.; Wang, Y. J. Macromol. Sci. Part A 2007, 44, 989–
993.
(20) Guo, X.; Watson, M. D. Org. Lett. 2008, 10, 5333–5336.
(21) Guo, X.; Ortiz, R. P.; Zheng, Y.; Kim, M.-G.; Zhang, S.; Hu, Y.; Lu, G.; Facchetti, A.;
Marks, T. J. J. Am. Chem. Soc. 2011, 133, 13685–13697.
(22) Shi, C.; Yao, Y.; Yang; Pei, Q. J. Am. Chem. Soc. 2006, 128, 8980–8986.
(23) Cho, S.; Seo, J. H.; Kim, S. H.; Song, S.; Jin, Y.; Lee, K.; Suh, H.; Heeger, A. J. Appl.
Phys. Lett. 2008, 93, 263301.

125
(24) Song, S.; Park, S. H.; Jin, Y.; Goo, Y.; Kim, I.; Lee, H.; Lee, K.; Suh, H. Synth. Met.
2010, 160, 2618–2622.
(25) Hendriks, K. H.; Li, W.; Wienk, M. M.; Janssen, R. A. J. Adv. Energy Mater. 2013, 3,
674–679.
(26) Keegstra, M. A.; Peters, T. H. A.; Brandsma, L. Tetrahedron 1992, 48, 3633–3652.
(27) Shrotriya, V.; Li, G.; Yao, Y.; Moriarty, T.; Emery, K.; Yang, Y. Adv. Funct. Mater.
2006, 16, 2016–2023.
(28) Krasovskiy, A.; Krasovskaya, V.; Knochel, P. Angew. Chem. Int. Ed. 2006, 45, 2958–
2961.
(29) Lin, W.; Baron, O.; Knochel, P. Org. Lett. 2006, 8, 5673–5676.
(30) Mosrin, M.; Knochel, P. Org. Lett. 2008, 10, 2497–2500.
(31) Khlyabich, P. P.; Rudenko, A. E.; Thompson, B. C. J. Polym. Sci. Part Polym. Chem.
2014, 52, 1055–1058.
(32) Rudenko, A. E.; Khlyabich, P. P.; Thompson, B. C. ACS Macro Lett. 2014, 3, 387–
392.
(33) Burkhart, B.; Khlyabich, P. P.; Thompson, B. C. Macromolecules 2012, 45, 3740–
3748.
(34) Kozycz, L. M.; Gao, D.; Seferos, D. S. Macromolecules 2013, 46, 613–621.
(35) Palermo, E. F.; McNeil, A. J. Macromolecules 2012, 45, 5948–5955.
(36) Hollinger, J.; Jahnke, A. A.; Coombs, N.; Seferos, D. S. J. Am. Chem. Soc. 2010, 132,
8546–8547.
(37) Daoust, G.; Leclerc, M. Macromolecules 1991, 24, 455–459.

126
(38) Aoyama, Y.; Yamanari, T.; Koumura, N.; Tachikawa, H.; Nagai, M.; Yoshida, Y.
Polym. Degrad. Stab. 2013, 98, 899–903.
(39) Liu, C.; Wang, K.; Hu, X.; Yang, Y.; Hsu, C.-H.; Zhang, W.; Xiao, S.; Gong, X.; Cao,
Y. ACS Appl. Mater. Interfaces 2013, 5, 12163–12167.
(40) Intemann, J. J.; Yao, K.; Yip, H.-L.; Xu, Y.-X.; Li, Y.-X.; Liang, P.-W.; Ding, F.-Z.;
Li, X.; Jen, A. K.-Y. Chem. Mater. 2013, 25, 3188–3195.
(41) Rudenko, A. E.; Noh, S.; Thompson, B. C. Nanotechnology 2013, 24, 484002.
(42) Yang, L.; Zhou, H.; You, W. J. Phys. Chem. C 2010, 114, 16793–16800.
(43) Veldman, D.; İpek, Ö.; Meskers, S. C. J.; Sweelssen, J.; Koetse, M. M.; Veenstra, S.
C.; Kroon, J. M.; Bavel, S. S. van; Loos, J.; Janssen, R. A. J. J. Am. Chem. Soc. 2008, 130,
7721–7735.
(44) Schilinsky, P.; Asawapirom, U.; Scherf, U.; Biele, M.; Brabec, C. J. Chem. Mater.
2005, 17, 2175–2180.

!
127
CHAPTER 3. Random Terpolymers Based on
3-Hexyloxythiophene, 2,1,3-Benzothiadiazole and 2,7-Carbazole
3.1 Introduction
As discussed in Chapter 1 and 2, an ideal polymer donor material should have a low band
gap (E
g
) and high absorption coefficient to reach a high short-circuit current density (J
sc
), an
effective matching of energy levels with fullerene acceptors to ensure a high open-circuit
voltage (V
oc
), and a suitable level of ordering as well as moderate miscibility with fullerenes
to achieve a high fill factor (FF).
1–4
Numerous alternating donor-acceptor (D-A) copolymers
have been synthesized and investigated as an approach to develop desired polymer materials
for polymer:fullerene bulk heterojunction (BHJ) solar cells.
5–8
In fact, state-of-the-art
efficiencies of organic photovoltaic (OPV) have been achieved by polymers that fall into the
alternating D-A copolymers, reaching over 9% in single junction cells,
9,10
and over 10% in
tandem cells.
11–13

Among alternating D-A copolymers, poly[N-9
’’
-hepta-decanyl-2,7-carbazole-
alt-5,5-(4
’
,7
’
-di-2-thienyl-2
’
,1
’
,3
’
-benzothiadiazole) (PCDTBT), as shown in Figure 3.1a,
has been widely studied and proved to be a promising polymer with a V
oc
value of 0.86-0.92
V, and PCE of 4-7% when blended with PC
61
BM or PC
71
BM as the solar cell active
layer.
14,15
In addition, the internal quantum efficiency of the solar cells based on its blends
!
128
with PC
71
BM was found to approach 100%.
16
However, this alternating D-A copolymer has
a band gap around 1.90 eV and a relatively narrow absorption band, which leaves out a large
portion of visible light photons not being harvested and results in J
sc
no more than 10
mA/cm
2
with fullerene derivatives. Therefore, it is of interest for reaching higher J
sc
and PCE
values to adjust the band gap and energy levels of PCDTBT by introducing structural
variations. In this regard, the random copolymer strategies described in Chapter 1 can be
useful in examining the possibility of property tuning of this copolymer.

Figure 3.1. Structure of polymers (a) PCDTBT, (b)
PCDTBT-OR reported in Ref 17 and 18, and (c)
PCDTBT
m
DHOTBT
n
and PCDHOTBT reported here.
In addition, the use of ternary blends with two donor materials with complementary
absorption bands can be an effective way to enhance the photon harvesting capability and to
improve the solar cell J
sc
.
19–22
Several research groups have reported their success in
!
129
achieving higher J
sc
values via this ternary blend approach.
23–29
Importantly, it has been
demonstrated in several cases that the V
oc
of ternary blend solar cells can be
composition-dependent and tunable between the two individual binary blend cells.
30–33
This
dual gain in J
sc
and V
oc
offers a promising pathway of reaching higher efficiencies. However,
in a paper published by our group in 2014, it was shown that the combination of PCDTBT
and P3HTT-DPP-10% as two donors blended with PC
61
BM led to pinned V
oc
around that of
the P3HTT-DPP-10%:PC
61
BM-based binary blend cells, which was around 0.59 V, much
lower than the 0.92 V of the PCDTBT:PC
61
BM-based ones.
34
This behavior has been mostly
attributed to the incompatibility of the two polymers. However, other variables may not be
deconvoluted completely, such as the large HOMO level offset of the two polymers being
over 300 mV. Targeted modifications over the polymer PCDTBT will help to further
understand the cause of pinned V
oc
in the previously reported system. In particular, the
possible influence from the energy level offset can be investigated by raising the HOMO
level of PCDTBT under rational control without changing other properties significantly.
According to the discussions above, introducing structural variations to the widely studied
PCDTBT is of interest in both practical and fundamental point of view. Previously, the Bo
group synthesized the derivative of PCDTBT with octyloxy groups on benzothiadiazole, as
shown in Figure 3.1b.
17
It was found to have a lower band gap and a planar backbone, which
facilitates photon capture and charge transport. The polymer with similar structures but
hexyloxy side chains on 2,1,3-benzothiadiazole units (Figure 3.1b) was synthesized by Meng
!
130
and coworkers and found to reach an efficiency around 7% in polymer:fullerene BHJ solar
cells after careful annealing process.
18

At the same time, the incorporation of the comonomer unit of 3-hexyloxythiophene (3HOT)
into random and semi-random copolymer backbones was investigated as described in
Chapter 2. All the resulting copolymers exhibited lower band gaps and
composition-dependent energy levels. Other reported copolymers containing
3-alkyloxythiophene (3AOT) units, such as the alternating copolymer prepared by Cho and
coworkers that is based on 2,7-dihydroindenoindene and 2,1,3-benzothiadiazole,
35
as well as
the one based on diketopyrrolopyrrole and phenylene,
36
were found to have significant
impact from 3AOT on the polymer energy levels as well.

Because of the simplicity of this
3HOT unit and its strong influence on the properties of conjugated copolymers, we find it
interesting to include it into PCDTBT derivative polymers and investigate its effect in a
systematic way. Here, the synthesis and characterization of a set of random terpolymers that
are PCDTBT derivatives but contain 4,7-bis(3-hexyloxythiophen-2-yl)-2,1,3-
benzothiadiazole units (DHOTBT) instead of 4,7-dithien-2-yl-2,1,3-benzothiadiazoles
(DTBT), as shown in Figure 3.1c. The investigation into the optical and electrochemical
properties of these polymers not only reveals the influence of this comonomer unit on
polymer properties, but also probes into the potential of property tuning based on PCDTBT.

!
131
3.2 Monomer and polymer syntheses
The DHOTBT unit has been reported previously and copolymerized with
2,7-dihydroindenoindene.
35
However, no detailed synthetic procedures could be found to the
best of our knowledge. In the work reported here, starting with
2-bromo-3-hexyloxythiophene (Compound 1, Figure 3.2) and
4,7-dibromo-2,1,3-benzothiadiazole (Compound 3), the DHOTBT unit (Compound 4) was
easily achieved by two reaction steps, followed by bromination under mild conditions to
afford the comonomer for Suzuki polymerizations (Compound 5). Due to the strong
electron-donating and electron-withdrawing interactions between 3HOT and
2,1,3-benzothiadiazole, this comonomer appears as dark red crystals in solid form and bright
red in organic solutions, distinct from the orange/red color of dithienyl-benzothiadiazole
comonomer.

Figure 3.2. Synthetic scheme to prepare brominated comonomer DHOTBT (5).
Compound 5 did not exhibit evidence of decomposition after months of storage at around 4
°C after purging with N
2
and occasional operations at room temperature in air. The
preparation of its isomeric structure, 4,7-bis(5-bromo-4-hexyloxythiophen-2-yl)-2,1,3-
!
132
benzothiadiazole (Compound 6) with hexyloxy side chains located on the 4-position of
thiophene rings as shown in Figure 3.3, was attempted as well. Although this structure has
been reported previously,
37
there was no detailed description on the synthesis or purification.
In our attempt to achieve this comonomer, Compound 6 suffered from decomposition in air
and over silica gel column, which rendered it extremely challenging to purify.

Figure 3.3. Isomeric structure to the
brominated DHOTBT comonomer
(Compound 5).
In fact, over 90% pure product of Compound 6 could be achieved after a simple work-up
(Appendix 2.2). However, the remaining impurity, which has a characteristic peak in
1
H
NMR spectrum at 7.84 ppm, could not be removed by either additional column
chromatography or recrystallization. In addition, during the process of handling the product,
decomposition was evidenced by the visual observation of chemical color changes on thin
layer chromatography (TLC) plate from red to blue and then to brownish black over the
period of ten minutes or less. Due to the unsatisfactory impurity, Compound 6 has not been
used in polymerizations. Further purifications under more careful control of conditions and
environment, such as using different chromatography columns and/or inert atmosphere, may
be necessary to achieve pure Compound 6 for polymer preparation.
!
133
Random terpolymers that are based on 2,7-carbazole and varied amounts of DHOTBT and
DTBT were synthesized by Suzuki polymerizations according to the scheme shown in Figure
3.4. These terpolymers expand the library of random conjugated copolymers described in
Section 1.3.3. The alternating copolymers PCDTBT and PCDHOTBT were synthesized and
compared as well.

Figure 3.4. Synthetic scheme of Suzuki polymerization to prepare random
terpolymers and the alternating copolymers PCDTBT and PCDHOTBT.
Molecular weights of copolymers were analyzed by SEC in o-dichlorobenzene (o-DCB)
against polystyrene standards and summarized in Table 3.1. All polymers have high
molecular weights around or above 11 kg/mol. Due to the solubilizing effect of additional
hexyloxy side chains in DHOTBT, the molecular weights of the copolymers increase when
more of this unit is present in the polymer backbone, except that polymer
PCDTBT
3
DHOTBT
1
falls out of the trend but has comparable molecular weights as
PCDTBT.

!
134
Table 3.1. Molecular weights and compositions of polymers.
Polymer
M
n
(kg/mol)
/PDI
a
Feed ratio
(m:n)
Composition
(m:n)
b

PCDTBT 11.8/2.43 1:0 1:0
PCDTBT
3
DHOTBT
1
11.0/2.40 3:1 2.6:1
PCDTBT
1
DHOTBT
1
12.8/3.70 1:1 1:1.0
PCDTBT
1
DHOTBT
3
17.7/2.33 1:3 1:3.3
PCDHOTBT 18.8/2.92 0:1 0:1
a
Determined by SEC with polystyrene as standard and o-DCB as eluent.
b

Determined by peak integrations of
1
H NMR spectra.
The compositions of copolymers are determined from integration of characteristic peaks in
the polymer
1
H NMR spectra, and are also listed in Table 3.1 with the comparison to feed
ratios. Particularly, the methyne proton next to the nitrogen atom of 2,7-carbazole shows up
as a singlet peak at 4.72 ppm, while the methylene protons next to the oxygens in DHOTBT
appear as a peak at 4.31 ppm. Considering the number of protons present in each repeating
unit, the molar ratio between 2,7-carbazole and DHOTBT can be obtained. With the
assumption that 2,7-carbazole should alternate with either of the two benzothiadiazole-based
comonomer according to the polymerization strategy, the 2,7-carbazole:DHOTBT ratio can
be easily transformed into the ratio between DTBT and DHOTBT in the terpolymer
backbones. Compared with the comonomer feed ratios, the compositions calculated using the
above-mentioned method reveal a slightly stronger tendency of DHOTBT incorporation into
polymer backbones than that of the DTBT units.

!
135
3.3 Optoelectronic properties of polymers
UV-Vis absorption profiles of all copolymers were obtained in both solutions and annealed
thin films, and are presented in Figure 3.5. (Summarized data is available in the Appendix 2,
Table A2.1.) As shown in Figure 3.5a, all copolymers exhibit similar absorption patterns in
annealed thin films. Two distinct absorption bands, with the maxima wavelength at around
400 nm and 600 nm, are observed for each polymer sample, which can be attributed to the
π−π* transition and intramolecular charge transfer (ICT) transition respectively. There is a
slight red shift of both absorption peaks with increasing amount of DHOTBT units in the
polymer backbone. However, the band gaps of polymers containing DHOTBT do not differ
significantly from the parent polymer PCDTBT. In fact, some previously reported
terpolymers containing one alternating electron-donor unit and two randomly distributed
electron-acceptor units have shown a composition-dependent combination of two distinct
ICT absorption bands due to the presence of two D-A pairs.
38–41
The observation of a single
ICT band for random terpolymers reported here indicates that the DTBT and DHOTBT units
do not behave differently to a considerable extent to generate distinct chromophores with
carbazoles. In addition, the absence of vibronic shoulders for each copolymer sample
reported here suggests a lack of structural ordering in the polymer thin films.
The polymer absorptions in o-DCB solutions, shown in Figure 3.5b, appear to be similar to
the profiles in thin films. Going from solutions to thin films, the absorption peaks shift to
longer wavelength due to the planaring of polymer backbones in solid states. To be noted, the
!
136
red shift in absorption onset from solutions to thin films is more significant when smaller
amount of DHOTBT units are present in copolymers. In particular, the absorption onset is
670 nm for PCDTBT in thin film, which differs from the 630 nm onset in solution by 40 nm,
while the onset difference for PCDHOTBT is only 20 nm from 660 nm in solution to 680
nm in thin film. This observation suggests that copolymers with a higher content of
DHOTBT units do not experience significant change in the planarity of polymer backbones
when they pack into solid films from solubilized chains. This is probably because the
hexyloxy side chains pointing towards the 2,1,3-benzothiadiazole core creates steric effect
and results in a dihedral angle between the 2,1,3-benzothiadiazole and thiophene rings, as
what has been observed for 4,7-bis(3-hexylthiophen-2-yl)-2,1,3-benzothiadiazole.
37

!
137

Figure 3.5. UV−vis absorption of copolymers in (a)
thin films spin-coated from CHCl
3
and annealed for
30 min under N
2
at 150 °C and (b) o-DCB solutions:
(black) PCDTBTT, (red) PCDTBT
3
DHOTBT
1
,
(green) PCDTBT
1
DHOTBT
1,
(dark blue)
PCDTBT
1
DHOTBT
3
, (light blue) PCDHOTBT.
In addition, the HOMO energy levels of polymers are measured using cyclic voltammetry
(CV) in the solid state. The obtained data of HOMO levels, as well as the UV-Vis absorption
onsets and band gaps, are summarized in Table 3.2. Although the polymer band gaps are not
influenced much by composition within the investigated sample set, their HOMO energies
vary over a large range from -5.53 eV to -5.30 eV. This observation further demonstrates the
!
138
strong influence of 3HOT units on the energy levels of polymers due to their
electron-donating nature.
Table 3.2. Optoelectronic properties of all polymers
Polymer
λ
onset

(film, nm)

a
λ
onset

(solution, nm)

b

E
g
(eV)
c
HOMO (eV)
d
PCDTBT 670 630 1.85 -5.53
PCDTBT
3
DHOTBT
1
674 640 1.84 -5.50
PCDTBT
1
DHOTBT
1
678 650 1.83 -5.43
PCDTBT
1
DHOTBT
3
678 655 1.83 -5.35
PCDHOTBT 680 660 1.82 -5.30
a
Wavelength of absorption onset of polymer thin films spin-coated from CHCl
3
and
annealed at 150 °C for 30 min.
b
Wavelength of absorption onset of polymer solutions in
o-DCB.
c
Optical band gaps from the onset of absorption in UV−Vis spectra of annealed
films.
d
Determined by cyclic voltammetry (vs Fc/Fc
+
) in acetonitrile containing 0.1 M
TBAPF
6
.
Additionally, grazing incidence X-ray diffraction (GIXRD) was used to probe the packing
of polymer chains within annealed thin films. The absence of diffraction peaks suggests the
amorphous nature of the solid state of these polymers, which is consistent with the
observation of no vibronic shoulder in thin film UV-Vis absorption profiles. These
above-mentioned properties of the polymer set indicate that the incorporation of DHOTBT
units into the parent polymer PCDTBT leads to modulated polymer energy levels while
retaining other features of PCDTBT such as band gaps and the amorphous nature.
3.4 Summary
A set of random terpolymers was synthesized by Suzuki polymerizations to replace varied
amounts of DTBT units in polymer PCDTBT with 3HOT-capped 2,1,3-benzothiadiazoles
!
139
(DHOTBT), with the aim to modulate the polymer properties. The alternating copolymer
PCDHOTBT was synthesized at the same time as a comparison. Integration of
1
H NMR
peaks verifies the terpolymer compositions and indicates that DHOTBT units are slightly
more favorable to react than DTBT.
The optoelectronic properties of all copolymers were investigated by UV-Vis absorption,
CV and GIXRD measurements. It is found that the HOMO levels of the polymers follow a
trend to increase when more DHOTBT units are present in the polymer backbones, ranging
from -5.53 eV of PCDTBT to -5.30 eV of PCDHOTBT. However, the optical band gaps do
not experience a significant change although the absorption peak maxima shift slightly to
longer wavelength with higher content of DHOTBT. At the same time, all polymers appear
to be amorphous in annealed thin films. These observations demonstrate the ability to tune
polymer energy levels with varied amount of 3HOT units, consistent with the findings
discussed in Chapter 2. Importantly, the HOMO level of PCDTBT can be raised under
control based on the terpolymer compositions, while the other properties of PCDTBT are
generally retained.
Considering their properties, these polymer samples can be promising candidates for
polymer:fullerene solar cells. Device performance measurements are currently underway in
our research group. This polymer set also opens the gate to further explore the working
principles of ternary blend solar cells, because HOMO energy offset between the two
polymer donors can be regulated in comparison with the system reported earlier with
!
140
PCDTBT and P3HTT-DPP-10%,
34
without giving significant changes on other parameters.
It will be interesting to test this polymer set with the combination of P3HTT-DPP-10% in
fullerene-based ternary blend solar cells, in order to investigate the requirements of a certain
HOMO level offset between the two polymer donors for highly efficient ternary blend cells.
3.5 References
(1) Li, G.; Zhu, R.; Yang, Y. Nat. Photonics 2012, 6, 153–161.
(2) Thompson, B. C.; Fréchet, J. M. J. Angew. Chem. Int. Ed. 2008, 47, 58–77.
(3) Li, Y. Acc. Chem. Res. 2012, 45, 723–733.
(4) Khlyabich, P. P.; Burkhart, B.; Rudenko, A. E.; Thompson, B. C. Polymer 2013, 54,
5267–5298.
(5) Helgesen, M.; Søndergaard, R.; Krebs, F. C. J. Mater. Chem. 2009, 20, 36–60.
(6) Li, Y.; Zou, Y. Adv. Mater. 2008, 20, 2952–2958.
(7) Zhan, X.; Zhu, D. Polym. Chem. 2010, 1, 409–419.
(8) Duan, C.; Huang, F.; Cao, Y. J. Mater. Chem. 2012, 22, 10416–10434.
(9) He, Z.; Zhong, C.; Su, S.; Xu, M.; Wu, H.; Cao, Y. Nat. Photonics 2012, 6, 591–595.
(10) Liao, S.-H.; Jhuo, H.-J.; Cheng, Y.-S.; Chen, S.-A. Adv. Mater. 2013, 25, 4766–
4771.
(11) You, J.; Chen, C.-C.; Hong, Z.; Yoshimura, K.; Ohya, K.; Xu, R.; Ye, S.; Gao, J.; Li,
G.; Yang, Y. Adv. Mater. 2013, 25, 3973–3978.
(12) You, J.; Dou, L.; Yoshimura, K.; Kato, T.; Ohya, K.; Moriarty, T.; Emery, K.; Chen,
!
141
C.-C.; Gao, J.; Li, G.; Yang, Y. Nat. Commun. 2013, 4, 1446.
(13) Yusoff, A. R. bin M.; Kim, D.; Kim, H. P.; Shneider, F. K.; Silva, W. J. da; Jang, J.
Energy Environ. Sci. 2014, 8, 303–316.
(14) Wakim, S.; Beaupré, S.; Blouin, N.; Aich, B.-R.; Rodman, S.; Gaudiana, R.; Tao, Y.;
Leclerc, M. J. Mater. Chem. 2009, 19, 5351–5358.
(15) Sun, Y.; Seo, J. H.; Takacs, C. J.; Seifter, J.; Heeger, A. J. Adv. Mater. 2011, 23,
1679–1683.
(16) 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, 297–302.
(17) Qin, R.; Li, W.; Li, C.; Du, C.; Veit, C.; Schleiermacher, H.-F.; Andersson, M.; Bo,
Z.; Liu, Z.; Inganäs, O.; Wuerfel, U.; Zhang, F. J. Am. Chem. Soc. 2009, 131, 14612–14613.
(18) Meng, B.; Fang, G.; Fu, Y.; Xie, Z.; Wang, L. Nanotechnology 2013, 24, 484004.
(19) Ameri, T.; Khoram, P.; Min, J.; Brabec, C. J. Adv. Mater. 2013, 25, 4245–4266.
(20) Yang, L.; Yan, L.; You, W. J. Phys. Chem. Lett. 2013, 4, 1802–1810.
(21) Goubard, F.; Wantz, G. Polym. Int. 2013, n/a – n/a.
(22) Chen, Y.-C.; Hsu, C.-Y.; Lin, R. Y.-Y.; Ho, K.-C.; Lin, J. T. ChemSusChem 2013, 6,
20–35.
(23) 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, 1556–1565.
(24) Sharma, S. S.; Sharma, G. D.; Mikroyannidis, J. A. Sol. Energy Mater. Sol. Cells
2011, 95, 1219–1223.
!
142
(25) Honda, S.; Nogami, T.; Ohkita, H.; Benten, H.; Ito, S. ACS Appl. Mater. Interfaces
2009, 1, 804–810.
(26) Koppe, M.; Egelhaaf, H.-J.; Dennler, G.; Scharber, M. C.; Brabec, C. J.; Schilinsky,
P.; Hoth, C. N. Adv. Funct. Mater. 2010, 20, 338–346.
(27) 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, 1198–1202.
(28) Lim, B.; Bloking, J. T.; Ponec, A.; McGehee, M. D.; Sellinger, A. ACS Appl. Mater.
Interfaces 2014, 6, 6905–6913.
(29) Lee, J.; Yun, M. H.; Kim, J.; Kim, J. Y.; Yang, C. Macromol. Rapid Commun. 2012,
33, 140–145.
(30) Huang, J.-H.; Velusamy, M.; Ho, K.-C.; Lin, J.-T.; Chu, C.-W. J. Mater. Chem.
2010, 20, 2820–2825.
(31) Khlyabich, P. P.; Burkhart, B.; Thompson, B. C. J. Am. Chem. Soc. 2011, 133,
14534–14537.
(32) Khlyabich, P. P.; Burkhart, B.; Thompson, B. C. J. Am. Chem. Soc. 2012, 134,
9074–9077.
(33) Yang, L.; Zhou, H.; Price, S. C.; You, W. J. Am. Chem. Soc. 2012, 134, 5432–5435.
(34) Khlyabich, P. P.; Rudenko, A. E.; Street, R. A.; Thompson, B. C. ACS Appl. Mater.
Interfaces 2014, 6, 9913–9919.
(35) Cho, S.; Seo, J. H.; Kim, S. H.; Song, S.; Jin, Y.; Lee, K.; Suh, H.; Heeger, A. J.
Appl. Phys. Lett. 2008, 93, 263301.
!
143
(36) Hendriks, K. H.; Li, W.; Wienk, M. M.; Janssen, R. A. J. Adv. Energy Mater. 2013,
3, 674–679.
(37) Lévêque, P.; Biniek, L.; Fall, S.; Chocos, C. L.; Leclerc, N.; Heiser, T. Energy
Procedia 2012, 31, 38–45.
(38) Jiang, J.-M.; Chen, H.-C.; Lin, H.-K.; Yu, C.-M.; Lan, S.-C.; Liu, C.-M.; Wei, K.-H.
Polym. Chem. 2013, 4, 5321–5328.
(39) Zhang, G.; Fu, Y.; Qiu, L.; Xie, Z. Polymer 2012, 53, 4407–4412.
(40) Kang, T. E.; Cho, H.-H.; Kim, H. jun; Lee, W.; Kang, H.; Kim, B. J.
Macromolecules 2013, 46, 6806–6813.
(41) Zhou, J.; Xie, S.; Amond, E. F.; Becker, M. L. Macromolecules 2013, 46, 3391–
3394.

144
CHAPTER 4. Morphology and Device Performance of
Polymer:Small Molecule:Fullerene Ternary Blend Solar Cells in
Comparison to Binary Blend Cells Based on Polymers with
Covalently Attached Small Molecules
4.1 Introduction
Ternary blend bulk heterojunction (BHJ) solar cells have received increasing attention in
recent years as a simple and cost-effective way to improve the performance of polymer-based
organic photovoltaics (OPV).
1,2
A ternary blend device is a single active layer device whose
active layer is composed of three distinct electron donor and acceptor species. In many cases,
the combination of materials with complementary absorption leads to more effective
harvesting of photons, which results in an increase in the short-circuit current density (J
sc
)
compared to either binary blend solar cells of the constituent components. This enhancement
in J
sc
has contributed to higher efficiencies from ternary blend solar cells than that of binary
blends in a number of three-component systems.
3–6
At the same time, besides the benefits in
J
sc
, it has been demonstrated recently that the open-circuit voltage (V
oc
) of ternary blend solar
cells can be intermediate and composition-dependent between that of the two binary cells,
which endows the ternary blend strategy with more promising potential toward higher
efficiencies.
7–9

145
The V
oc
tuning ability of ternary blends correlated with composition has been observed in
the work of our group for both the one donor and two acceptors (D:A1:A2) system, as well as
two donors and one acceptor (D1:D2:A) system.
7,8
Similar behavior was also reported by
other groups.
10
The origin of this behavior has been ascribed to the formation of an organic
alloy of the synergistic components (D1:D2 or A1:A2) with HOMO and LUMO energies
based on average composition.
11
This organic alloy model was supported by the empirical
fact found by the authors that, in ternary blends with a tunable V
oc
, a continuous change was
observed in the energy of the charge-transfer (CT) state with composition, suggesting an
averaging of frontier orbital (HOMO and LUMO) energies.
However, the composition-dependent V
oc
was only observed in limited examples of ternary
blends so far. In many cases, V
oc
was pinned to the lower value of binary blend systems.
3,5,6

Our group also recently reported a ternary blend system containing poly[N-9′-heptadecanyl-
2,7-carbazole-alt-5,5-(4′,7′-di-2-thienyl-2′,1′,3′-benzothiadiazole)] (PCDTBT) and poly(3-
hexylthiophene-thiophene-diketopyrrolopyrrole) (P3HTT-DPP-10%) donors with phenyl-
C61-butyric acid methyl ester (PC
61
BM) acceptor that showed a V
oc
pinned to the lower value
of the P3HTT-DPP-10%:PC
61
BM binary blend even up to 95% PCDTBT in the polymer
fraction.
12
This work suggests that miscibility of the synergistic components is a necessary
condition for alloy formation and an enabling parameter for composition-dependent V
oc
in
ternary blends. However, it still remains unclear if miscibility is sufficient for alloy formation
and the design and evaluation of new ternary systems is necessary to further understand the
structure-performance relationships of ternary blends and to deconvolute additional variables
that may influence alloy formation such as HOMO energy difference and relative
crystallinity of the components.

146
So far the ternary blends containing one polymer donor, one small molecule donor and a
fullerene acceptor have not been discussed in the context of the alloy model. Although a
large amount of work has been done on such systems as a means to broaden the absorption
spectrum and to enhance J
sc
, the small molecule (or organic dye) usually only constitutes a
very small percentage of the ternary blends and is typically considered as a sensitizer to
introduce near IR absorption.
13–20
The trend in V
oc
of small molecule containing ternary cells
is often not obvious due to the lack of data over the full polymer:small molecule ratio range
and for the small molecule binary cells themselves. Only in very few examples of
polymer:small molecule:fullerene ternary blends, a composition-dependent V
oc
was observed,
and still not over the full range of polymer:small molecule ratio.
21
It is interesting and
important to investigate further into polymer:small molecule:fullerene ternary blends in the
context of V
oc
tuning ability, as an approach to understand how variables can influence
organic alloy formation.

Figure 4.1. Structures of materials used in this work.

147
Therefore, we report here the combination of a polymer poly(3-hexylthiophene) (P3HT), a
small molecule bisphenyl-diketopyrrolopyrrole (PhDPP), and PCBM as active layer in solar
cell devices (Figure 4.1), and the effect of ternary blend composition and processing
conditions on the device performance, especially in the context of V
oc
tuning ability. In
comparison to these ternary blends, where each component is possible to either self-
crystallize and segregate or co-crystallize and intermix, depending on the processing
conditions and thermodynamic minima of the blends, we also report the design and synthesis
a functionalized P3HT with the diketopyrrolopyrrole (DPP) -based small molecule covalently
attached, shown as P1-P3 in Figure 4.1, where self-crystallization and segregation of either
P3HT or the small molecule is forbidden and intimate mixing of the two components is
enforced physically. The comparison between the two systems will provide information on
how molecular structures and morphology of the small molecule containing ternary blend
active layer influences the alloy formation and device V
oc
.
4.2 Material syntheses
As stated above, we intended to investigate the possibility of alloy formation in
polymer:small molecule:fullerene ternary blend solar cells as a way to further elucidate the
working principle of ternary blends. The particular materials chosen in this study are rr-P3HT
as the polymer donor, a DPP-based small molecule PhDPP donor and PC
61
BM acceptor, as
depicted in Figure 4.2 along with the energy levels and absorption profiles of each material.
These two donors were selected because the binary blend solar cells based on either of them,
respectively, with PC
61
BM can not only work effectively, but also exhibit V
oc
values that
differ by 300 mV. This V
oc
value offset offers an opportunity to clearly track the trend in the

148
ternary blend V
oc
, and thus the tendency of alloy formation correlated with blend
compositions or processing conditions can be easily indicated.

Figure 4.2. Structures, energy levels and UV-Vis
absorption of materials used in the ternary blend solar
cells in this work.
As a comparison to this ternary blend system, we also designed and synthesized the
functionalized P3HT analogs to afford covalent attachment between P3HT and PhDPP

149
donors, as presented in Figure 4.3a. The covalent bonding is realized through “click”
reactions between azide groups on functionalized P3HT side chains and the alkyne groups at
the end of PhDPP molecules. A short aliphatic spacer is kept between the two donor
conjugated systems so that their electronic properties are not significantly affected. Azide-
functionalized P3HT syntheses start with the Grignard Metathesis (GRIM) copolymerization
between dibrominated 3-hexylthiophene (3HT) and 3-(2-bromoethyl)thiophene (3BrET)
monomers, followed by S
N
2 substitution of the bromines by azide groups, as shown in Figure
4.3b. Full conversions from bromine to azide groups and from azide to PhDPP attached side
chains are evidenced by the
1
H NMR peak of the methylene protons completely shifted, as
can be seen in Supporting Information, Figure A3.4-A3.12. After careful purification by
Soxhlet extractions, pure polymers P1-P3 with three different PhDPP compositions are
achieved (Table 4.1).

Figure 4.3. (a) “Click” reaction to covalently attach PhDPP onto the side chains of P3HT; (b)
synthetic routes towards azide group functionalized P3HT. Reaction condition: (i)
CuBr/PMDETA, THF, 35 °C, 3d.

The molar ratios between 3HT comonomers and PhDPP-attached comonomers in each of
the polymer backbone are calculated from
1
H NMR integrations based on methylene protons

150
that are characteristic of 3HT and PhDPP fragments respectively, and the values are
summarized in Table 4.1. The calculated numbers deviate slightly from the initial feed ratio
in GRIM copolymerizations, which may be due to reactivity difference between the two
comonomers. The weight ratios between corresponding P3HT and PhDPP in P1-P3 are also
determined and shown in Table 4.1. The weight percentage of small molecule PhDPP in the
two donor blends range from about 13% to 35%. Polymers with even higher composition of
PhDPP are not obtained, mainly due to the poor solubility of P3N
3
Ts. Polymers P1-P3
achieved and studied here all exhibit high molecular weights and good solubility in common
organic solvents, such as CHCl
3
and o-DCB.
Table 4.1. PhDPP compositions and molecular weights of P1-P3.
Sample
Feed ratio
(m:n)
Composition
(m:n)
a
Weight
ratio
b
M
n

(kg/mol)
c

PDI
c
P1 95:5 96.2:3.8 86.6:13.4 47.1 2.02
P2 90:10 91.8:8.2 75.0:25.0 42.9 2.07
P3 85:15 86.7:13.3 64.9:35.1 21.7 2.31
a
Determined by peak integrations of
1
H NMR spectra.
b
Calculated weight ratio between
corresponding P3HT and PhDPP fragments in each polymer.
c
Determined by SEC with
polystyrene as standard and o-DCB as eluent.

To be noted, efforts to attach small molecules on the side chains of functionalized P3HT
have been reported before, as a way to broaden the absorption of polymers, with
phthalocyanines and porphyrins respectively.
22,23
However, this work reported here is the first
attempt to investigate the effect of such structures on V
oc
as a physically constrained
counterpart to ternary blend systems.

151
4.3 Device performance of ternary blends
Table 4.2 summarizes the performance of P3HT:PhDPP:PC
61
BM ternary blends in both as-
cast and annealed devices, over the full range of P3HT:PhDPP ratios, along with the two
binary blend end points. Interestingly, in as-cast devices, V
oc
exhibits a trend to regularly
increase with more PhDPP in the blend, from 0.61 V for P3HT:PC
61
BM binary blend cell to
0.85 V for that of PhDPP:PC
61
BM. Similar behavior of increasing V
oc
correlated with more
small molecule constitution was reported by Huang and coworkers with a system based on
P3HT, PC
71
BM and TQTFA small molecule donor.
21
However, their data did not include
tuning of the P3HT:TQTFA ratios over the full range. To the best of our knowledge, the data
we report here is the first example of full range V
oc
tuning in polymer:small
molecule:fullerene ternary blend solar cells. This trend in V
oc
resembles that observed in our
previous work with either two polymer donors and one fullerene acceptor, or one polymer
donor and two fullerene acceptors, and it suggests the formation of an organic alloy of the
two donor components.
7,8
However, these devices were tested before annealing, and J
sc
and
FF are low probably due to the lack of continuous charge transport pathways for either holes
or electrons, which is not surprising for unannealed devices.

152
Table 4.2. Device performance of P3HT:PhDPP:PC
61
BM ternary blend solar cells
before and after annealing.
P3HT:PhDPP:PCBM
(weight ratio)
As-Cast
a
J
sc
(mA/cm
2
) V
oc
(V) FF PCE (%)

1:0:1 2.02 0.612 0.326 0.401
0.8:0.2:1 1.22 0.634 0.295 0.227
0.6:0.4:1 1.55 0.692 0.288 0.309
0.4:0.6:1 1.70 0.761 0.288 0.37
0.2:0.8:1 1.69 0.829 0.268 0.375
0:1:1.5 2.55 0.851 0.287 0.619
P3HT:PhDPP:PCBM
(weight ratio)
Annealed
b
J
sc
(mA/cm
2
) V
oc
(V) FF PCE (%)

1:0:1 6.79 0.504 0.446 1.52
0.8:0.2:1 6.19 0.520 0.361 1.16
0.6:0.4:1 4.85 0.521 0.38 0.956
0.4:0.6:1 2.47 0.512 0.319 0.402
0.2:0.8:1 1.63 0.489 0.36 0.288
0:1:1.5 1.54 0.843 0.411 0.53
a
The ternary blend active layer was spin-coated from CHCl
3
solutions without
any annealing process.
b
The ternary blend active layer was spin-coated from
CHCl
3
solutions followed by o-DCB solvent vapor annealing for 2 min.
J
sc
and FF can be both considerably improved after annealing process, as shown in Table
4.2. However, in the meantime, V
oc
values drop to around 0.5 V for all blend ratios as well as
P3HT binary cell. The drop in V
oc
after annealing is as much as 0.35 V when the two donor
mixture contains 80% PhDPP and 20% P3HT. This data suggests that the annealing process
disrupts the organic alloy formation. Actually it has been reported by Belcher and coworkers
that when porphyrin small molecules were included into blends of P3HT and PC
61
BM to
afford ternary blends, the annealing process could lead to large-scale aggregation of
porphyrins.
24
Considering the general high crystallizing tendency of conjugated small

153
molecules, we hypothesize that the self-crystallization and aggregation of PhDPP small
molecules destroys the intimate mixing (alloy) between P3HT and PhDPP donor, and thus
prevents the averaging of electronic energies of these two donors.
4.4 Morphology of ternary blends
In order to support our hypothesis about PhDPP self-crystallization as a factor inhibiting
alloy formation, we investigated the morphology of ternary blends before and after annealing
with grazing-incidence X-ray diffraction (GIXRD) and Atomic Force Microscopy (AFM)
measurements. Figure 4.4 shows the GIXRD diffraction patterns of ternary blend films
before annealing. The peak correlating with the (100) planes of crystalline domains is very
weak in all blend films with different component ratios, as evidenced by the low signal-to-
noise ratio. These low-intensity diffraction peaks appear in the region of 5-6°, which is
between the characteristic peaks of pristine P3HT and PhDPP films at 5.2° and 6.3°
respectively (shown in Figure 4.5). This may suggest a certain degree of co-crystallization of
P3HT and PhDPP, possibly as well as PC
61
BM, in as-cast films, which facilitates the alloy
formation.

154

Figure 4.4. GIXRD of P3HT:PhDPP:PC
61
BM ternary
blend as-cast films spin-coated from CHCl
3
solution: (i)
P3HT80:PhDPP20:PC
61
BM, (ii) P3HT60:PhDPP40:
PC
61
BM, (iii) P3HT40:PhDPP60:PC
61
BM, (iv) P3HT20:
PhDPP80:PC
61
BM.

Figure 4.5. GIXRD of P3HT:PhDPP:PC
61
BM ternary blend
films spin-coated from CHCl
3
solution followed by
annealing with o-DCB vapor for 2 min: (ii) P3HT80:
PhDPP20:PC
61
BM, (iii) P3HT60:PhDPP40:PC
61
BM, (iv)
P3HT40:PhDPP60:PC
61
BM, (v) P3HT20:PhDPP80:
PC
61
BM, along with (i) P3HT and (vi) PhDPP films thermal
annealed at 150 °C.

155
The diffraction patterns of ternary blend films after annealing are shown in Figure 4.5,
together with those of the pristine P3HT and PhDPP films. It is clear that a sharp and intense
peak at 6.3° now appears in all ternary blend films, which is in the same position as the peak
of pristine PhDPP films. This indicates the self-crystallization and aggregation of the small
molecule PhDPP taking place after the annealing process, similar to what Belcher and
coworkers reported for porphyrins.
24
Additionally, in ternary blends that contain 80% and 60%
P3HT, another peak between 5-6° can be seen. The position of this peak moves away from
that of the pristine P3HT when PhDPP constitutes more in blends, and its intensity decreases
at the same time, relative to the peak at 6.3°. In the blend containing 40% P3HT, this peak
diminishes into a slight shoulder to the left of the 6.3° peak, while in the blend with 20%
P3HT, it disappears completely. This observation may be a consequence of co-crystallization
of P3HT and PhDPP. However, the device performance, especially V
oc
, seems to be dominant
by the strong PhDPP self-crystallization.

156

Figure 4.6. AFM topology images of (a and e) P3HT80:
PhDPP20:PC
61
BM, (b and f) P3HT60:PhDPP40:PC
61
BM, (c and g)
P3HT40:PhDPP60:PC
61
BM and (d and h) P3HT20:PhDPP80: PC
61
BM
films spin-coated from CHCl
3
solution, before (a, b, c and d) and after (e,
f, g and h) annealing with o-DCB vapor for 2 min.
Figure 4.6 shows the AFM topology images of ternary blend films before and after
annealing. The films appear to be very smooth without any obvious features before annealing
for all blend ratios. The surface root mean square (RMS) roughness is determined to be 0.483

157
nm, 0.815 nm, 0.397 nm and 0.299 nm respectively with PhDPP content ranging from 20%
to 80%. After annealing, large surface features as big as 500 nm in diameter can be seen for
all blend ratios. The RMS roughness values also increase drastically to 4.17 nm, 8.14 nm,
5.24 nm and 6.50 nm, which are about one magnitude higher correspondingly. This further
demonstrates the fact that the degree of crystallization in these ternary blends is significantly
enhanced by the annealing process, which probably lead to the failure of V
oc
tuning in
annealed solar cells.
4.5 Characterization and device performance of P1-P3
The polymers P1-P3 with PhDPP covalently attached to the side chains of P3HT can be
considered as a mixture of P3HT and PhDPP donors where the two components are
physically constrained. Self-crystallization and aggregation of either component will be
suppressed in this way, enforcing the intimate mixing of the two. The blends of P1-P3 with
PC
61
BM are expected to form different morphology than P3HT:PhDPP:PC
61
BM ternary
blends and thus perform differently in solar cell devices. In order to explore this hypothesis,
we investigated the lamellar ordering of P1-P3 with GIXRD. Essentially, none of the three
polymers show obvious diffraction peaks, which demonstrates that crystallization is
sufficiently suppressed in these samples as expected.

158

Figure 4.7. GIXRD of films of PhDPP-attached P3HT
polymers, as well as P3HT, spin-coated from CHCl
3

solution and annealed at 150 °C for 30 min: (black)
P3HT, (red) P1, (green) P2 and (blue) P3.

Table 4.3. Device performance of solar cells based on P1-P3 blended with
PC
61
BM.
Polymer:PC
61
BM J
sc
(mA/cm
2
) V
oc
(V) FF
PCE
(%)

P3HT
a
9.49 0.540 0.616 3.14
P1
a
8.12 0.653 0.589 3.11
P2
a
6.29 0.677 0.513 2.17
P3
a
3.13 0.787 0.312 0.765
a
The optimized ratio of polymer:PC
61
BM is 1:1.
Device performance of these polymer samples after annealing is summarized in Table 4.3.
In stark contrast to the annealed ternary blend solar cells, V
oc
increases regularly with a
higher composition of PhDPP in these functionalized polymers. With about 35% by weight
of PhDPP incorporated in such polymers, the V
oc
value reaches a value almost 250 mV higher
than P3HT:PC
61
BM solar cells. At the same time, J
sc
and FF remain comparable to P3HT for
P1-based solar cells, where the composition of PhDPP is 13% by weight, affording a PCE

159
almost the same as P3HT. When the content of PhDPP increases to 25% in P2, the solar cell
can still work efficiently with J
sc
over 6 mA/cm
2
and FF above 0.5, leading to a PCE over 2%.
This demonstrates the ability of such polymers to perform as efficient donor materials with
PC
61
BM acceptors, regardless of the bulky side chain attachment and their amorphous nature.

Figure 4.8. Solar cell V
oc
values of as a function of PhDPP
composition of (open circle) P3HT:PhDPP:PC
61
BM ternary
blends before annealing, (filled square) P3HT:PhDPP:
PC
61
BM ternary blends after annealing and (filled triangle)
PhDPP-attached P3HT polymers P1-P3.
The trends in V
oc
as a function of PhDPP composition based on annealed P1-P3, as well as
ternary blends both before and after annealing, are plotted in Figure 4.8. The composition-
dependence of V
oc
can be clearly seen in as-cast ternary blend solar cells as well as those
based on polymers P1-P3. In addition, the polymers P1-P3 give V
oc
values even higher than
the as-cast ternary blend devices under the same corresponding composition. To explore
more deeply into this phenomenon, HOMO levels of P1-P3 in both films and solutions are
measured by cyclic voltammetry (CV).

160
Table 4.4. HOMO levels of PhDPP-attached P3HT polymers P1-
P3 in both films and solutions, together with those of P3HT and
PhDPP small molecule.
Sample Film HOMO Solution HOMO
P3HT -5.22 eV -5.33 eV
P1 -5.27 eV -5.32 eV
P2 -5.32 eV -5.32 eV
P3 -5.39 eV -5.33 eV
PhDPP -5.5 eV --

As seen in Table 4.4, the HOMO levels of P1-P3 films decrease regularly with an
increasing amount of PhDPP, from the -5.22 eV of P3HT to -5.39 eV of P3 when 35% of
PhDPP is incorporated. However, the HOMO levels in solutions remains constant for all
polymer samples and the same as that of P3HT at -5.3 eV. Therefore, we propose that the
extremely high V
oc
observed with these polymer samples may partly result from the steric
effect of these bulky side groups influencing the conformation and packing of the polymer
backbone, similar to what was previously observed by our group for poly(3-hexylthiophene-
co-3-(2-ethylhexyl)thiophene) copolymers (P3HT-co-EHTs).
25
However, the increase in V
oc

observed for these DPP-attached P3HT polymers are more significant than that observed for
P3HT-co-EHTs. With only about 13 mol% of the comonomers functionalized with the bulky
DPP side chains in P3, the copolymers already exhibit a V
oc
that is almost 250 mV higher
than P3HT itself. This difference in solar cell V
oc
between P3HT and P3 is also stronger than
that in HOMO energy levels, which is less than 0.2 eV. This behavior probably suggests that
there are other factors playing a role in determining the device V
oc
of copolymers P1-P3,
which is likely to be the cooperating effect of both P3HT and PhDPP donors that form
organic alloys when constrained in one phase.

161
4.6 Conclusion
In the work described in this chapter, the ternary blends P3HT:PhDPP:PCBM as active
layer in solar cell devices were compared to the cases where the two donor components
P3HT and PhDPP were covalently attached. In particular, the DPP derivatives were
connected to functionalized P3HT through “click” reactions at various amounts to produce
polymers P1-P3. In comparison to the ternary blends, where each component is free to phase
separate, the movement of P3HT and/or the small molecule segments in P1-P3 is constrained.
In fact, GIXRD results and AFM topology images suggest formation of PhDPP-dominant
crystallites after annealing of P3HT:PhDPP:PCBM ternary blend thin films. Such strong self-
crystallization of one donor component leads to unfavorable phase separation in the active
layer of ternary blend solar cells. As a result, the annealed devices display a low V
oc
value
around 0.5 V that is about pinned to the lower value of the corresponding binary blend cells
(the P3HT:PCBM-based one), which is in stark contrast to the tunable V
oc
with blend
compositions observed for unannealed devices. Such findings demonstrate the necessity of
controlling the ternary blend morphology in order to achieve formation of organic alloys and
hence composition-dependent V
oc
values.
On the contrary, the self-crystallization of DPP derivatives is sufficiently suppressed in
polymer samples P1-P3, suggesting that intimate mixing of the two components is physically
enforced. Solar cells using any of these polymers as the donor material also exhibit V
oc
tuning
ability even after annealing process. At a small amount of DPP derivative being attached to
P3HT (around 3.5 mol% relative to 3HT repeating units), polymer P1 leads to a V
oc
of 0.653
V when blended with PCBM as the solar cell active layer, which is over 100 mV higher than

162
that based on P3HT:PCBM. Due to the enhancement in V
oc
, the PCE of P1-based devices
exceeds 3%, which is about the same as that of P3HT-based cells, even though J
sc
and FF are
slightly lower for P1-based ones. At the same time, measurements of the HOMO energy
levels of P1-P3 suggest possible impact from molecular packing on the device V
oc
values due
to the bulky nature of the DPP-based side groups. However, the remarkable degree of
increase in V
oc
probably results from a combined effect from both molecular packing effect
and organic alloy formation. As such, further investigation is necessary to explain the origin
of V
oc
tuning behavior of solar cells based on these polymer samples.
This work provides a detailed examination on solar cell performance of polymer:small
molecule:PCBM ternary blends over the entire range of blend ratios, especially in terms of
V
oc
tuning ability. Due to the strong tendency of conjugated small molecule to self-crystallize,
formation of organic alloy is greatly affected by the processing condition of ternary blends,
which influences the device V
oc
thereby. It is also shown that covalent attachment of the two
donor components offers a way to suppress the self-crystallization and unfavorable phase
separation in ternary blends, which help to meet the prerequisites for tunable device V
oc
based
on blend compositions.
4.7 References
(1) Ameri, T.; Khoram, P.; Min, J.; Brabec, C. J. Adv. Mater. 2013, 25, 4245–4266.
(2) Khlyabich, P. P.; Burkhart, B.; Rudenko, A. E.; Thompson, B. C. Polymer 2013, 54,
5267–5298.
(3) Koppe, M.; Egelhaaf, H.-J.; Dennler, G.; Scharber, M. C.; Brabec, C. J.; Schilinsky,
P.; Hoth, C. N. Adv. Funct. Mater. 2010, 20, 338–346.

163
(4) Mikroyannidis, J. A.; Tsagkournos, D. V.; Balraju, P.; Sharma, G. D. J. Power
Sources 2011, 196, 2364–2372.
(5) Hu, Z.; Tang, S.; Ahlvers, A.; Khondaker, S. I.; Gesquiere, A. J. Appl. Phys. Lett.
2012, 101, 053308.
(6) 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, 1198–1202.
(7) Khlyabich, P. P.; Burkhart, B.; Thompson, B. C. J. Am. Chem. Soc. 2011, 133,
14534–14537.
(8) Khlyabich, P. P.; Burkhart, B.; Thompson, B. C. J. Am. Chem. Soc. 2012, 134, 9074–
9077.
(9) Khlyabich, P. P.; Burkhart, B.; Rudenko, A. E.; Thompson, B. C. Polymer 2013, 54,
5267–5298.
(10) Yang, L.; Zhou, H.; Price, S. C.; You, W. J. Am. Chem. Soc. 2012, 134, 5432–5435.
(11) Street, R. A.; Davies, D.; Khlyabich, P. P.; Burkhart, B.; Thompson, B. C. J. Am.
Chem. Soc. 2013, 135, 986–989.
(12) Khlyabich, P. P.; Rudenko, A. E.; Street, R. A.; Thompson, B. C. ACS Appl. Mater.
Interfaces 2014, 6, 9913–9919.
(13) Sharma, S. S.; Sharma, G. D.; Mikroyannidis, J. A. Sol. Energy Mater. Sol. Cells
2011, 95, 1219–1223.
(14) Sharma, G. D.; Singh, S. P.; Roy, M. S.; Mikroyannidis, J. A. Org. Electron. 2012, 13,
1756–1762.
(15) Lee, J.; Yun, M. H.; Kim, J.; Kim, J. Y.; Yang, C. Macromol. Rapid Commun. 2012,
33, 140–145.

164
(16) Ye, L.; Xia, H.; Xiao, Y.; Xu, J.; Miao, Q. RSC Adv. 2013, 4, 1087–1092.
(17) 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, 1556–1565.
(18) Honda, S.; Ohkita, H.; Benten, H.; Ito, S. Adv. Energy Mater. 2011, 1, 588–598.
(19) Honda, S.; Nogami, T.; Ohkita, H.; Benten, H.; Ito, S. ACS Appl. Mater. Interfaces
2009, 1, 804–810.
(20) Lim, B.; Bloking, J. T.; Ponec, A.; McGehee, M. D.; Sellinger, A. ACS Appl. Mater.
Interfaces 2014, 6, 6905–6913.
(21) Huang, J.-H.; Velusamy, M.; Ho, K.-C.; Lin, J.-T.; Chu, C.-W. J. Mater. Chem. 2010,
20, 2820–2825.
(22) Campo, B. J.; Duchateau, J.; Ganivet, C. R.; Ballesteros, B.; Gilot, J.; Wienk, M. M.;
Oosterbaan, W. D.; Lutsen, L.; Cleij, T. J.; Torre, G. de la; Janssen, R. A. J.; Vanderzande,
D.; Torres, T. Dalton Trans. 2011, 40, 3979–3988.
(23) Angiolini, L.; Cocchi, V.; Lanzi, M.; Salatelli, E.; Tonelli, D.; Vlamidis, Y. Mater.
Chem. Phys. 2014, 146, 464–471.
(24) Belcher, W. J.; Wagner, K. I.; Dastoor, P. C. Sol. Energy Mater. Sol. Cells 2007, 91,
447–452.
(25) Burkhart, B.; Khlyabich, P. P.; Thompson, B. C. Macromolecules 2012, 45, 3740–
3748.

165
CHAPTER 5. Stereoregular Non-conjugated Polymers with
Diketopyrrolopyrrole Electroactive Pendant Groups
5.1 Introduction
Conjugated polymer-fullerene (donor-acceptor) composites are generally considered as
the most promising active layer materials in organic photovoltaics. The state-of-the-art
power conversion efficiencies based on such materials have reached over 9% in single-
junction OPV devices,
1
and over 10% in tandem cells.
2,3
However, blends of conjugated
polymers and fullerenes are not ideal in certain points of view.
One of the requirements to meet such high efficiencies is the favorable morphology
within the active layer of polymer-fullerene blends, which should be a bicontinuous
network with nanometer length-scale phase separation to ensure efficient exciton and
charge transport. In typical cases, within the bulk-heterojunction layer of polymer-
fullerene blends, the suitable bicontinuous morphology is kinetically trapped after
thermal or solvent annealing.
4
Achieving optimal morphology via kinetic trapping is
challenging and is a limiting factor for expanding the scope of donor-acceptor pairs. In
addition to the morphological drawbacks, conjugated polymers generally do not possess
ideal mechanical properties that are needed in flexible devices, casting doubt over the
vision of flexible photovoltaics.
5,6
Moreover, the stability of these extended conjugated
systems toward light and ambient conditions (oxidation and moisture) is a concern.
7

166
In order to realize robust favorable morphology, diblock copolymers with suitable
electronic properties of the two blocks (donor and acceptor blocks) can be used, as the
only one material to generate active layer of photovoltaics. Block copolymers are well
known to undergo phase separation on the nanometer length-scale,
8
giving bicontinuous
morphologies with well-defined interfaces, shown in Figure 5.1, as is desired for the
bulk-heterojunction active layers. Importantly, the morphologies shown in Figure 5.1
represent thermodynamic minima and are not kinetically trapped structures.

Figure 5.1. (a) Cartoon representation of the process of phase
separation in A-B diblock copolymers. (b) Cartoons illustrating
the different morphologies that are commonly observed in A-B
diblock copolymers. When the fraction of A (f
A
) is greater than
the fraction of B (f
B
), spherical (S), cylindirical (C), or gyroid (G)
phases of B can exist within a continuous matrix of A. The
converse is true when f
A
< f
B
. If f
A
= f
B
, a lamellar phase (L) can
exist. Note that C, G, and L phases are bicontinuous.

A number of examples have been reported regarding to the performance of donor-
acceptor diblock copolymers as OPV active layers.
9,10
In the majority of cases, two
conjugated blocks are used (rod-rod block copolymers), or one conjugated block is end-

167
functionalized followed by the growth of a second block via radical polymerization (rod-
coil block copolymers). Such rod-containing block copolymers often do not display the
well-defined morphologies shown in Figure 5.1 as there is an additional driving force to
the order-disorder transition (microphase separation) from the crystallization of the rod
blocks.
10,11
Notably, the Thelakkat group exploited diblock copolymers in photovoltaics
that were synthesized via radical polymerizations with triphenylamine pendants as donors
and perylene pendants as acceptors, as represented by the structure shown in Figure
5.2.
12–14
The non-conjugated backbones used in their study typically display improved
mechanical and ambient stability than conjugated polymers. More importantly, it is
observed that these polymers can form phase-separated structures with naonometer
features, as should be optimal for photovoltaics. However, photovoltaic efficiency is
observed to be low (0.32%) and can be attributed to poor charge transport in the polymer
active layer, caused by suspected disorder within one or both of the phases.

Figure 5.2. Chemical structure of an
atactic block copolymer reported by
Thelakkat.
14

168
For effective charge transport in organic semiconductors, there must be effective face-
to-face π−π overlap between adjacent molecules.
15
In fact, the radical polymerizations
employed in the synthesis of Thelakkat’s work tend to give inherently stereorandom
(atactic) polymers. It has been observed that charge mobilities are higher when perylene
pendants are attached to polyacrylate backbones with long alkyl-spacers.
16,17
The long
flexbile spacers create sufficient distance between the perylenes and the atactic backbone
so that the liquid crystalline perylenes can self-organize into ordered motifs and provide
efficient charge transport pathways. The same principle does not apply to noncrystalline
pendants such as triphenylamines,
18,19
nor to other potentially attractive pendants that do
not exhibit the tendency to self-organize in crystalline structures. Therefore, if non-
conjugated polymers with electroactive pendants are to be improved towards OPV
applications, some judicious synthetic control is necessary to ensure ordering of the
pendants.
In fact, it has been theoretically proposed a high degree of intermolecular p-overlap for
effective charge transport can be ensured through control over backbone stereoregularity
of pendant polymers.
20–22
Poly(vinyl carbazole) (Figure 5.3(1)) has been used as the
prototypical system for exploring these concepts, where extended isotactic sequences
(Figure 5.3b) are thought to dramatically increase π-overlap, as opposed to atactic
sequences (Figure 5.3c). Additionally, it has been shown empirically that increasing the
isotacticty of carbazole moieties functionalized polyacrylates (Figure 5.3(2)) results in a
strong increase in the hole mobility of the polymer.
23
These results suggest that
controlling the backbone stereoregularity of electroactive pendant polymers could serve
as a general route towards polymers with high mobilities, irrespective of the nature of the

169
pendant group and without the requirement of using long alkyl tethers, which introduce
insulating character.

Figure 5.3. (a) Illustration of the axis of π-overlap
generated via face-to-face packing in conjugated
molecules that leads to effective charge transport along
the π-axis. (b) Illustration of the effective π-overlap
generated in isotactic poly(N-vinylcarbazole) in the
extended all-trans form. (c) Illustration of the poor π-
overlap generated in atactic poly(N-vinylcarbazole) in
the extended all-trans form. (1) Poly(N-vinylcarbazole).
(2) Poly(N-ethylcarbazole acrylate).
However, the hypothesis that stereoregular pendant polymers will display improved
charge transport relative to their stereorandom analogues has not been carefully tested
with common pendant structures and modern experimental techniques. Such investigation
will be of fundamental significance in expanding our understandings on the structure-
property relationships of organic semiconductors, as well as in exploring new material

170
platforms for organic photovoltaics. Toward this direction, the synthesis of both atactic
and moderately isotactic polyacrylamides with diketopyrrolopyrrole (DPP) derivatives
attached to the backbones via “click” reaction, as well as the brief investigation in their
electronic properties, is reported here. This work is also another application of “click”
chemistry in the synthesis of semiconducting polymers, in addition to the work shown in
Chapter 4.
5.2 System Design
Structures shown in Figure 5.4a are designed as a proof-of-concept to evaluate the
synthetic strategies toward stereoregular pendant polymers. Three important aspects
should be recognized in the design principles.

Figure 5.4. Polymer and small molecule samples investigated in this work.

171
First, post-polymerization alkyne-azide “click” reaction is used to attach the
electroactive pendants onto the non-conjugated polymer backbones. Alkyne-azide “click”
reaction is famous for its tolerance of a wide range of functional groups and high yields
under a wide range of conditions.
24
This strategy will avoid the challenging
polymerization of bulky and electroactive monomers, especially in the case of
stereoregular polymer synthesis. Second, DPP-based structures are used as pendants,
mainly due to its promising potential in organic electronic applications.
25,26
Third,
polyacrylamides are employed as the non-conjugated backbones and are to be
synthesized into both atactic and isotactic chains that are available for pendant attachment.
On one hand, radical polymerizations do not suffer from the lack of functional group
tolerance as anionic or Zeigler-Natta polymerizations do. On the other hand,
polyacrylamides are readily available to functionalize for “click” chemistry.
In addition, an atactic polystyrene-based polymer (PS-alkyne), as shown in Figure 5.4b
is used as the backbone precursor as well, being converted to DDP-functionalized
pendant polymers. This material not only serves as a model system to prove the working
efficiency of “click” reactions, but also provides a comparison among different polymer
backbone structures and their corresponding influence on pendant polymer electronic
properties. The azide functionalized DPP small molecule, shown in Figure 5.4c is also
subject to characterizations, as a control to investigate the difference in molecular
packings and electronic properties while the conjugated molecules are with or without
covalent linkage confinements.

172
5.3 Materials Syntheses
5.3.1 Stereorandom and stereoregular free radical polymerizations of PNIPAM
Before conducting polymerizations of functionalized acrylamides, N-
isopropylacrylamide (NIPAM) is used as a model monomer for radical polymerizations
to optimize reaction conditions that will lead to either atactic and isotactic
polyacrylamides. There has been reported methods to generate isotactic polyacrylamides
by simply adding a bulky Lewis Acid, such as Yttrium (III) Triflate (Y(OTf)
3
), into the
radical polymerization mixtures.
27
The bulky Lewis Acid is thought to enforce a multisite
coordination (Figure 5.5) with the polar carbonyl functionality of the monomer, leading
to the preferential meso (isotactic) placement of each monomer added to the polymer
chain.
28

Figure 5.5. Cartoon illustrating the mechanism of
stereoregularity control in Lewis Acid-mediated
radical polymerizations that lead to isotactic
polymers.
28

In this work, a number of reaction conditions with varied monomer concentrations,
types of Lewis Acids and Lewis Acid loadings were screened to test the reproducibility
of the reported methods. Table 5.1 lists some of our experimental results.

173
Due to the poor solubility of PNIPAM in the solvent system o-DCB employed in our
SEC instrument, molecular weight information on the PNIPAM samples was not
collected. The dyad tacticity of PNIPAM backbone is determined from the integration
ratio of
1
H NMR peaks between 1.25-1.75 ppm that correspond to meso and racemo
dyads respectively. More specifically, the backbone methylene protons in racemo dyads
show up as a singlet around 1.5 ppm, while those of meso dyads split into two peaks
sitting on both sides of the racemo peak, as represented in the
1
H NMR spectra taken
from Entry 1 and 5, shown in Figure 5.6.
29,30

Table 5.1. Polymerizations of NIPAM at 60
o
C for 24h.
Entry # Solvent Lewis Acid [Lewis Acid] Yield m/r
c
1
a
MeOH None N/A 55% 47/53
2
a
MeOH Y(OTf)
3
0.05 M, 0.1 eq 67% 82/18
3
a
MeOH Y(OTf)
3
0.10 M, 0.2 eq 40% 83/17
4
a
MeOH Y(OTf)
3
0.25 M, 0.5 eq 53% 85/15
5
b
MeOH Y(OTf)
3
0.2 M, 0.1 eq 52% 83/17
6
a
toluene Y(OTf)
3
0.05 M, 0.1 eq 48% 53/47
7
a
toluene Y(OTf)
3
0.10 M, 0.2 eq 30% 65/35
8
a
toluene Y(OTf)
3
0.25 M, 0.5 eq 13% 65/35
9
a
MeOH Sc(OTf)
3
0.05 M, 0.1 eq 49% 62/38
10
a
MeOH Sc(OTf)
3
0.10 M, 0.2 eq 22% 65/35
11
a
MeOH Sc(OTf)
3
0.25 M, 0.5 eq 35% 66/34
12
a
MeOH Zn(OTf)
2
0.05 M, 0.1 eq 43% 47/53
13
a
MeOH Zn(OTf)
2
0.10 M, 0.2 eq 15% 47/53
14
a
MeOH Zn(OTf)
2
0.25 M, 0.5 eq 9% 51/49
a
[NIPAM]=0.5 M, [AIBN]= 0.01M.!
b
[NIPAM]= 2 M, [AIBN]= 0.04 M.
c
Dyad
tacticities m/r calculated from
1
H NMR spectra measured in DMSO-d
6
at 125
o
C.

174

Figure 5.6.
1
H NMR spectra of (a) atactic PNIPAM from
Entry 1 and (b) isotactic PNIPAM from Entry 5 in
DMSO-d
6
at 125 °C.
The calculated results suggest that the free radical polymerization without Lewis Acid
actually produces PNIPAM that is slightly richer in racemo configurations (m/r = 47/53),
propably due to the inherent steric effect during monomer addition, even though the
polymer is referred to as “atactic” for simplicity in this Chapter. Results from the other
entries reveal a strong influence from the identity of Lewis Acid on the tacticity of
resulting PNIPAM, but not much impact from the Lewis Acid equivalence or monomer

175
concentration. In particular, when the polymerizations are carried out in MeOH at 60
o
C,
the use of Y(OTf)
3
results in a dyad tacticity above m/r=80/20 with all monomer
concentrations or Y(OTf)
3
equivalence being tested. However, reactions with Sc(OTf)
3

only lead to m/r around 65/35, while the effect of Zn(OTf)
2
is negligible. In addition, the
effect of Y(OTf)
3
is found to be much more significant in the polar solvent MeOH than in
non-polor toluene. These observations lead to a conclusion that is consistent with
previous literatures, that is, the m/r dyad tacticities resulted with Y(OTf)
3
in MeOH
(Entry 2-5) are above 80/20 and among the highest numbers compared to other Lewis
Acid species or solvents, and the results are not significantly influenced by the monomer
concentrations or Y(OTf)
3
loadings.
31,32

5.3.2 Synthesis of alkyne functionalized polystyrene and polyacrylamides (PS-alkyne,
PA-atactic and PA-isotactic)
The synthetic strategies to prepare alkyne-functionalized polymers are described in the
following paragraphs. Molecular weights of the polymers achieved here will be
summarized in Table 5.2 within Section 5.3.3. Alkyne functionalized polystyrene was
achieved from the commercially available poly(4-hydroxystyrene) with one step of S
N
2
reaction, as illustrated by Figure 5.7. The polymer precursor is atactic, and therefore it
leads to atactic polymers in following reactions. Full conversion from hydroxy group to
propargyl ether is confirmed by
1
H NMR. (See Appendix 4, Figure A4.13)

176

Figure 5.7. Synthetic routes to prepare atactic alkyne
functionalized polystyrene (PS-alkyne).
Alkyne-functionalized polyacrylamides with both stereorandom and stereoregular
backbones (PA-TMS-atactic and PA-TMS-isotactic respectively) were synthesized in
two steps as shown in Figure 5.8. The same conditions that produced atactic PNIPAM
(Entry 1, Table 5.1) and isotactic PNIPAM (Entry 2, Table 5.1) were adapted to
polymerize the trimethylsilyl-protected N-propargylacrylamide (Compound 1, Figure 5.8).

Figure 5.8. Synthetic routes to prepare alkyne-functionalized polyacrylamides
with both stereorandom and stereoregular backbones (PA-TMS-atactic and PA-
TMS-isotactic respectively).
As shown in Figure 5.9, the backbone methine proton exhibits similar splitting pattern
in
1
H NMR as that of PNIPAM, and thus the dyad tacticities can be calculated in the
same way. It is determined that PA-TMS-atactic has a dyad tacticity of m/r=50/50, while
PA-TMS-isotactic prepared in the presence of Y(OTf)
3
contains 78% of meso dyads.

177
The isotacticity of PA-TMS-isotactic is actually less than the isotactic PNIPAM
prepared in same conditions. This is probably because the bulkier trimethylsilyl group
affects the Lewis Acid coordination efficiency to some extent.

Figure 5.9.
1
H NMR spectra of (a) PA-TMS-atactic and
(b) PA-TMS-isotactic in DMSO-d
6
at 125 °C.
The second step is the deprotection of terminal alkynes in poly(N-
propargylacrylamide)s catalyzed by a mild base Na
2
CO
3
. According to the integration of
the remaining methyl proton peak in
1
H NMR spectra corresponding to trimethylsilyl

178
groups, the deprotection efficiency is around 99% for PA-atactic, and approaching 100%
for PA-isotactic. (Appendix 4, Figure A4.14 and A4.15)
5.3.3 “Click” reactions to generate electroactive pendant polymers
Azide functionalized DPP derivative was used to react with the alkyne functionality on
polystyrene or polyacrylamides (Figure 5.4). After careful Soxhlet extractions with
MeOH, hexanes and chloroform, pure polymers with almost 100% of DPP pendant
attachment were achieved (Figure A4.16-A4.18). The polymer molecular weight and
dyad tacticity information is summarized in Table 5.2. The molecular weights of DPP-
attached polymers are under-estimated by SEC relative to those of the precursors,
probably due to the conformation change of the polymer chains that leads to a decrease in
the size of molecules.
Table 5.2. Molecular weights and dyad tacticities of polymer samples.
Polymer M
n
, kDa
a
PDI
a
Dyad tacticity, m/r
b
PA-TMS-atactic 10.6 1.75 50/50
PA-DPP-atactic 10.0 2.15 --
c
PA-TMS-isotactic 13.1 1.81 78/22
PA-DPP-isotactic 16.6 2.12 --
d
PS-alkyne 23.5 2.13 N/A
PS-DPP 43.9 2.84 N/A
a
Measured by SEC with polystyrene as standard and THF as eluent.
b
Dyad tacticities m/r
calculated from
1
H NMR spectra measured in DMSO-d
6
at 125
o
C.
c
Dyad tacticities m/r
presumably being the same as that of PA-TMS-atactic.
d
Dyad tacticities m/r presumably
being the same as that of PA-TMS-isotactic.
5.4 Characterizations of Polymer Samples and DPP-N
3

The optical properties of DPP-pendant polymers, together with the small molecule
DPP-N
3
were obtained using UV-Vis spectroscopy in both solutions and thin films as
shown in Figure 5.10.

179
In solutions (Figure 5.10a), the polymers exhibit absorption maxima at shorter
wavelength compared to DPP-N
3
(573 nm for polymers vs. 590 nm for DPP-N
3
), as well
as lower absorptivities. The absorption coefficient is smallest in the case of PS-DPP,
which is easy to understand considering the fact that polystyrene backbones that do not
absorb visible light constitute a considerable fraction of the polymer total weight. As the
backbone molecular weights of polyacrylamides are relatively less, their DPP segments
are not diluted as much as PS-DPP. Therefore, PA-DPP-atactic and PA-DPP-isotactic
do not suffer from decrease in absorptivity significantly.
As for the UV-Vis absorptions in thin films (Figure 5.10b), all the polymer samples
exhibit the same peak positions and intensities that are remarkably different from the
DPP-N
3
. There is a strong vibronic shoulder between 600 and 700 nm in the absorption
profile of DPP-N
3
, which is associated with semicrystallinity and ordering in the solid
state and indicates crystallization of the DPP-derivative small molecule. In contrast, none
of the polymer samples have vibronic shoulders in their thin film absorption profile.
However, the peaks of polymer samples are red-shifted compared to the major peak of
DPP-N
3
, from 542 nm to 577 nm. This difference between polymer samples and the
small molecule DPP-N
3
suggest that polymer backbones can introduce significant
influence in the molecular packing of pendant segments. At the same time, the absorption
coefficients of polymer samples are about half of DPP-N
3
, mostly due to the diluting
effect from polymer backbones.

180

Figure 5.10. UV-Vis Absorption of DPP-N
3
(black), PS-
DPP (red), PA-DPP-atactic (green) and PA-DPP-
isotactic (blue) in (a) o-DCB solution and (b) thin films
spin-coated from chloroform followed by thermal
annealing at 110 °C for 20 min.
With the aim at investigating non-conjugated polymer backbone impacts on the
packing of pendant groups, GIXRD was used to probe the molecular ordering in the
thermally annealed thin films of the DPP-pendant polymer samples, as well as that of
DPP-N
3
. As shown in Figure 5.11, the small molecule DPP-N
3
exhibits a sharp and
intense diffraction peak that corresponds to the (100) plane parallel to the substrate
surface. The peak appears at 5.78 degrees, which correlates to a lamellar spacing of 15.29

181
Å. On the contrary, none of the other DPP-pendant polymer samples shows any evidence
of structural ordering in the direction perpendicular to the substrate, indicating a likely
amorphous nature of these polymer-based thin films. This observation demonstrates again
the significant impact on the packing behavior of conjugated side groups influenced by
the non-conjugated polymer backbones.

Figure 5.11. GIXRD of thin film of DPP-N
3
(black), PS-DPP
(red), PA-DPP-atactic (green) and PA-DPP-isotactic (blue)
spin-coated from CHCl
3
and annealed at 110 °C for 20 min.
Consistent with the GIXRD data, AFM topology images shown in Figure A4.19
confirm the formation of large crystalline structures at micron-scale in the annealed film
of DPP-N
3
, but lacking of bulk structural features in those of the pendant polymer thin
films as evidenced by extremely flat film surfaces. In addition, the thermal behaviors of
DPP-N
3
and pendant polymers are investigated by differential scanning calorimetry, and
the second heating profiles (shown in Figure A4.20). The small molecule DPP-N
3
shows
an exothermic peak around 125 °C that corresponds with the melting process of

182
crystalline domains. This behavior is in good agreement with the vibronic shoulder
observed in its UV-Vis absorption profile and the high roughness of film surface
determined by AFM. In contrast, there is no melting peak observed with the polymer
samples, indicating their amorphous nature in solid state.
The bandgaps of all samples calculated from UV-Vis absorption onsets are summarized
in Table 5.3, together with the HOMO energy levels measured by cyclic voltammetry
(CV) and the hole mobilities determined from the space charge limiting current (SCLC)
method. There is a small difference in HOMO energy levels among the three pendant
polymers, suggesting that polymer backbone structures may have impact on the
electronic properties of the active pendant species via their influence on molecular
packings.
Essential to the goal and expectations of this work is the hole mobility of each sample.
The hypothesis is that stereoregular polymer backbones can enhance the molecular
packing of electroactive pendant groups, and thus provide more efficient pathways for
charge transport and lead to higher charge mobilities than that of stereorandom pendant
polymers. As shown in Table 5.3, the hole mobility of DPP-N
3
after annealing
experiences a huge increase of more than one order of magnitude compared to as-cast
films, which is consistent with the fact that DPP-N
3
forms crystalline domains after
annealing. On the contrary, the three polymer samples do not show such a significant
improvement after annealing. In fact, the hole mobility of PS-DPP is determined to be
quite low either with or without annealing, suggesting that the polystyrene backbone
interrupts efficient packing of DPP segments and impedes charge transport.

183
Table 5.3. HOMO energy levels, optical bandgaps and hole mobilities of all samples.
Sample
HOMO,
film
(eV)
E
g, opt
,
film
(nm/eV)
E
g, opt
,
solution
(nm/eV)
Hole Mobility (cm
2
/Vs)
As cast
b
Annealed
d
DPP-N
3
--
a
700, 1.77 650, 1.91
8.15*10
-6
2.2*10
-4

PS-DPP -5.40 685, 1.81 665, 1.86
1.67*10
-6
1.03*10
-6

PA-DPP-atactic -5.50 685, 1.81 670, 1.85
2.8*10
-6
6.8*10
-6

PA-DPP-isotactic -5.46 685, 1.81 670, 1.85 --
c
8.0*10
-6

a
Film of DPP-N
3
dissolves in acetonitrile and does not produce stable signal in cyclic
voltammetry measurement.
b
Measured from pristine polymer films spin-coated from
CHCl
3
.
c
No valid mobility data can be extrapolated from SCLC measurements.
d
Measured from pristine polymer films spin-coated from CHCl
3
followed by thermal
annealing at 110 °C for 20 min.
Polyacrylamide-based polymers show higher mobilities relative to PS-DPP, but the
values still are much lower than annealed DPP-N
3
due to lack of crystalline domains in
the polymer samples. PA-DPP-isotactic shows a slightly higher hole mobility than PA-
DPP-atactic, which falls in the same direction as expected from hypothesis. However,
the difference is not strong between atactic and isotactic polyacrylamides. The major
reason is believed to be the insufficient degree of isotacticity. Due to synthetic challenges,
the PA-DPP-isotactic used in this study has a dyad tacticity of no more than m/r=78/22.
The percentage of racemo dyad is not small enough, and theoretically the segments of
continuing meso dyad are only four-repeating-unit long in average. Such configurations
probably do not lead to effective impact on the packing of pendant groups. Therefore, it is
necessary to explore new approaches toward even higher stereoregularity in order to
investigate our original hypothesis. Nevertheless, the synthetic strategies shown in this
work, particularly the use of post-polymerization “click” chemistry to attach pendant
groups onto non-conjugated polymer backbones, is a good demonstration of a universal
functionalization strategy in similar pendant polymer synthesis.

184
5.5 Conclusions and Summary
The research reported in this chapter is a fundamental exploration in the field of
electroactive polymers with novel structural features. In particular, a DPP-based small
molecule DPP-N
3
was attached to three different non-conjugated polymers as pendant
groups via “click reaction”. The three DPP-containing pendant polymers differ from
each other in their backbone structures, including the stereoregularity. Careful evaluation
of the
1
H NMR spectra provides information on the dyad tacticity of the pendant polymer
samples. Atactic configurations are found for PS-DPP that was derived from a
commercially available poly(4-hydroxystyrene), as well as PA-DPP-atactic that was
synthesized from free radical polymerization and functionalized later. At the same time,
PA-DPP-isotactic was observed to be dominated by meso dyads in the backbone,
resulting in a relatively isotactic configuration. The polymer samples and the small
molecule DPP-N
3
were characterized and compared in order to reveal the backbone
impact on their pendant group packing behaviors and hence the polymer properties.
Importantly, the molecular packing of DPP-N
3
is found to differ drastically from all
other polymer samples, as evidenced by its distinct UV-Vis absorption profile, X-ray
diffraction pattern and surface features. In contrast to the annealed DPP-N
3
film being
highly crystalline, the polymer films appear to lack ordering and obvious features at
surface. These results indicate the remarkable influence on the packing of pendant groups
caused by the covalent constraints of polymer backbones. In addition, the slight
difference in HOMO energy levels of the polymer samples suggests that such influence is
reflected in the electronic properties of these samples as well. Therefore, it is important to

185
have precise control over backbone structures in pendant polymers containing
electroactive pendant groups in order to achieve suitable properties for electronic
applications.
However, the three pendant polymer samples do not exhibit significant difference
among each other, in terms of UV-Vis absorption, GIXRD results and surface features.
The SCLC hole mobility values increase slightly in the order of PS-DPP, PA-DPP-
atactic and PA-DPP-isotactic, but appear in the same order of magnitude and much
lower than the crystalline DPP-N
3
. This does not agree well with our initial hypothesis
that stereoregular polymer backbones can ensure efficient π−π packing of electroactive
pendants and thus improve the charge mobility of the pendant polymers relative to
stereorandom polymers. Nevertheless, this discrepancy does not disapprove our
hypothesis because the sample PA-DPP-isotactic only has moderate stereoregularity.
Instead, the results indicate that even higher degree of stereoregularity, being either
isotactic or syndiotactic, is probably necessary to illustrate the ultimate impact of this
parameter. A screening of various linker groups and lengths between backbone and
pendants may also be needed to achieve maximum effects. To this end, it is essential to
explore effective synthetic methods toward highly stereoregular polymers with well-
controlled functionalities. Besides the measurements done in this work focusing on the
bulk structure and electronic properties of these non-conjugated polymers, test in their
mechanical properties and stability will also be of great value, and as it will probably
reveal remarkable improvement than conjugated polymers.
This research work is a starting point for exploring a novel type of semi-conducting
polymers, namely stereoregular non-conjugated polymers with electroactive pendants. In

186
the big picture of polymer-based photovoltaics, such investigation is preliminary to
discover efficient coil-coil block copolymers as a single-component active layer that can
be more stable, more mechanically favorable and superior in morphology controls
compared to the currently popular conjugated polymer-fullerene blends. Following the
possible future work directions mentioned in the previous paragraph, the investigation in
the current topic may eventually lead to a promising new path towards efficient polymer-
based photovoltaics.
5.6 References
(1) He, Z.; Zhong, C.; Su, S.; Xu, M.; Wu, H.; Cao, Y. Nat. Photonics 2012, 6, 591–
595.
(2) You, J.; Chen, C.-C.; Hong, Z.; Yoshimura, K.; Ohya, K.; Xu, R.; Ye, S.; Gao, J.;
Li, G.; Yang, Y. Adv. Mater. 2013, 25, 3973–3978.
(3) 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.
(4) Li, G.; Shrotriya, V.; Huang, J.; Yao, Y.; Moriarty, T.; Emery, K.; Yang, Y. Nat.
Mater. 2005, 4, 864–868.
(5) Tahk, D.; Lee, H. H.; Khang, D.-Y. Macromolecules 2009, 42, 7079–7083.
(6) Savagatrup, S.; Makaram, A. S.; Burke, D. J.; Lipomi, D. J. Adv. Funct. Mater.
2014, 24, 1169–1181.
(7) Jørgensen, M.; Norrman, K.; Krebs, F. C. Sol. Energy Mater. Sol. Cells 2008, 92,
686–714.
(8) Bates, F. S.; Fredrickson, G. H. Phys. Today 2008, 52, 32–38.

187
(9) Nakabayashi, K.; Mori, H. Materials 2014, 7, 3274–3290.
(10) Topham, P. D.; Parnell, A. J.; Hiorns, R. C. J. Polym. Sci. Part B Polym. Phys.
2011, 49, 1131–1156.
(11) Lee, M.; Cho, B.-K.; Zin, W.-C. Chem. Rev. 2001, 101, 3869–3892.
(12) Lindner, S. M.; Kaufmann, N.; Thelakkat, M. Org. Electron. 2007, 8, 69–75.
(13) Lindner, S. M.; Hüttner, S.; Chiche, A.; Thelakkat, M.; Krausch, G. Angew. Chem.
Int. Ed. 2006, 45, 3364–3368.
(14) Sommer, M.; Lindner, S. M.; Thelakkat, M. Adv. Funct. Mater. 2007, 17, 1493–
1500.
(15) Bässler, H.; Köhler, A. In Unimolecular and Supramolecular Electronics I;
Metzger, R. M., Ed.; Topics in Current Chemistry; Springer Berlin Heidelberg, 2012; pp.
1–65.
(16) Hüttner, S.; Sommer, M.; Thelakkat, M. Appl. Phys. Lett. 2008, 92, 093302.
(17) Lang, A. S.; Neubig, A.; Sommer, M.; Thelakkat, M. Macromolecules 2010, 43,
7001–7010.
(18) Shirota, Y.; Kageyama, H. Chem. Rev. 2007, 107, 953–1010.
(19) Maria, S.; Susha, A. S.; Sommer, M.; Talapin, D. V.; Rogach, A. L.; Thelakkat,
M. Macromolecules 2008, 41, 6081–6088.
(20) Ishizawa, H.; Nakano, T.; Yade, T.; Tsuji, M.; Nakagawa, O.; Yamaguchi, T. J.
Polym. Sci. Part Polym. Chem. 2004, 42, 4656–4665.
(21) Gallego, J.; Pérez-Foullerat, D.; Mendicuti, F.; Mattice, W. L. J. Polym. Sci. Part
B Polym. Phys. 2001, 39, 1272–1281.
(22) Karali, A.; Dais, P.; Mikros, E.; Heatley, F. Macromolecules 2001, 34, 5547–5554.

188
(23) Uryu, T.; Ohkawa, H.; Oshima, R. Macromolecules 1987, 20, 712–716.
(24) Romulus, J.; Henssler, J. T.; Weck, M. Macromolecules 2014, 47, 5437–5449.
(25) Qu, S.; Tian, H. Chem. Commun. 2012, 48, 3039–3051.
(26) Li, Y.; Sonar, P.; Murphy, L.; Hong, W. Energy Environ. Sci. 2013, 6, 1684–1710.
(27) Isobe, Y.; Fujioka, D.; Habaue, S.; Okamoto, Y. J. Am. Chem. Soc. 2001, 123,
7180–7181.
(28) Kamigaito, M.; Satoh, K. Macromolecules 2008, 41, 269–276.
(29) Ray, B.; Isobe, Y.; Matsumoto, K.; Habaue, S.; Okamoto, Y.; Kamigaito, M.;
Sawamoto, M. Macromolecules 2004, 37, 1702–1710.
(30) Idota, N.; Nagase, K.; Tanaka, K.; Okano, T.; Annaka, M. Langmuir 2010, 26,
17781–17784.
(31) Okamoto, Y.; Habaue, S.; Isobe, Y.; Suito, Y. Macromol. Symp. 2003, 195, 75–80.
(32) Habaue, S.; Isobe, Y.; Okamoto, Y. Tetrahedron 2002, 58, 8205–8209.

!
189
BIBLIOGRAPHY
Ameri, T.; Khoram, P.; Min, J.; Brabec, C. J. Adv. Mater. 2013, 25, 4245–4266.
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, 1198–1202.
Angiolini, L.; Cocchi, V.; Lanzi, M.; Salatelli, E.; Tonelli, D.; Vlamidis, Y. Mater. Chem.
Phys. 2014, 146, 464–471.
Aoyama, Y.; Yamanari, T.; Koumura, N.; Tachikawa, H.; Nagai, M.; Yoshida, Y. Polym.
Degrad. Stab. 2013, 98, 899–903.
Bässler, H.; Köhler, A. In Unimolecular and Supramolecular Electronics I; Metzger, R. M.,
Ed.; Topics in Current Chemistry; Springer Berlin Heidelberg, 2012; pp. 1–65.
Bates, F. S.; Fredrickson, G. H. Phys. Today 2008, 52, 32–38.
Belcher, W. J.; Wagner, K. I.; Dastoor, P. C. Sol. Energy Mater. Sol. Cells 2007, 91, 447–
452.
Bian, L.; Zhu, E.; Tang, J.; Tang, W.; Zhang, F. Prog. Polym. Sci. 2012, 37, 1292–1331.
Brabec, C. J.; Gowrisanker, S.; Halls, J. J. M.; Laird, D.; Jia, S.; Williams, S. P. Adv. Mater.
2010, 22, 3839–3856.
Braunecker, W. A.; Oosterhout, S. D.; Owczarczyk, Z. R.; Kopidakis, N.; Ratcliff, E. L.;
Ginley, D. S.; Olson, D. C. ACS Macro Lett. 2014, 3, 622–627.
Bui, T. T. T.; Jahandar, M.; Song, C. E.; Hoang, Q. V.; Lee, J.-C.; Lee, S. K.; Kang, I.-N.;
Moon, S.-J.; Shin, W. S. Sol. Energy Mater. Sol. Cells 2015, 134, 148–156.
Bundgaard, E.; Krebs, F. C. Sol. Energy Mater. Sol. Cells 2007, 91, 954–985.
Burkhart, B.; Khlyabich, P. P.; Cakir Canak, T.; LaJoie, T. W.; Thompson, B. C.
Macromolecules 2011, 44, 1242–1246.
Burkhart, B.; Khlyabich, P. P.; Thompson, B. C. ACS Macro Lett. 2012, 1, 660–666.
!
190
Burkhart, B.; Khlyabich, P. P.; Thompson, B. C. J. Photonics Energy 2012, 2, 021002–1.
Burkhart, B.; Khlyabich, P. P.; Thompson, B. C. Macromol. Chem. Phys. 2013, 214, 681–
690.
Burkhart, B.; Khlyabich, P. P.; Thompson, B. C. Macromolecules 2012, 45, 3740–3748.
Cabanetos, C.; El Labban, A.; Bartelt, J. A.; Douglas, J. D.; Mateker, W. R.; Fréchet, J. M. J.;
McGehee, M. D.; Beaujuge, P. M. J. Am. Chem. Soc. 2013, 135, 4656–4659.
Campo, B. J.; Bevk, D.; Kesters, J.; Gilot, J.; Bolink, H. J.; Zhao, J.; Bolsée, J.-C.;
Oosterbaan, W. D.; Bertho, S.; D’Haen, J.; Manca, J.; Lutsen, L.; Van Assche, G.;
Maes, W.; Janssen, R. A. J.; Vanderzande, D. Org. Electron. 2013, 14, 523–534.
Campo, B. J.; Oosterbaan, W. D.; Gilot, J.; Cleij, T. J.; Lutsen, L.; Janssen, R. A. J.;
Vanderzande, D. In SPIE 7416, Organic Photovoltaics X; Kafafi, Z. H.; Lane, P. A.,
Eds.; 2009; Vol. 7416, p. 74161G – 74161G – 11.
Cao, W.; Xue, J. Energy Environ. Sci. 2014, 7, 2123–2144.
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, 1556–1565.
Chan, L.-H.; Juang, S.-Y.; Chen, M.-C.; Lin, Y.-J. Polymer 2012, 53, 2334–2346.
Chan, L.-H.; Lin, L.-C.; Yao, C.-H.; Liu, Y.-R.; Jiang, Z.-J.; Cho, T.-Y. Thin Solid Films
2013, 544, 386–391.
Chang, W.-H.; Gao, J.; Dou, L.; Chen, C.-C.; Liu, Y.; Yang, Y. Adv. Energy Mater. 2014, 4,
1300864.
Chang, Y.-T.; Hsu, S.-L.; Chen, G.-Y.; Su, M.-H.; Singh, T. A.; Diau, E. W.-G.; Wei, K.-H.
Adv. Funct. Mater. 2008, 18, 2356–2365.
Chang, Y.-T.; Hsu, S.-L.; Su, M.-H.; Wei, K.-H. Adv. Funct. Mater. 2007, 17, 3326–3331.
Chen, C.-H.; Cheng, Y.-J.; Chang, C.-Y.; Hsu, C.-S. Macromolecules 2011, 44, 8415–8424.
Chen, C.-H.; Cheng, Y.-J.; Dubosc, M.; Hsieh, C.-H.; Chu, C.-C.; Hsu, C.-S. Chem. – Asian J.
2010, 5, 2483–2492.
!
191
Chen, C.-H.; Hsieh, C.-H.; Dubosc, M.; Cheng, Y.-J.; Hsu, C.-S. Macromolecules 2010, 43,
697–708.
Chen, C.-P.; Chan, S.-H.; Chao, T.-C.; Ting, C.; Ko, B.-T. J. Am. Chem. Soc. 2008, 130,
12828–12833.
Chen, H.-C.; Chen, Y.-H.; Liu, C.-C.; Chien, Y.-C.; Chou, S.-W.; Chou, P.-T. Chem. Mater.
2012, 24, 4766–4772.
Chen, Y.-C.; Hsu, C.-Y.; Lin, R. Y.-Y.; Ho, K.-C.; Lin, J. T. ChemSusChem 2013, 6, 20–35.
Cho, S.; Seo, J. H.; Kim, S. H.; Song, S.; Jin, Y.; Lee, K.; Suh, H.; Heeger, A. J. Appl. Phys.
Lett. 2008, 93, 263301.
Chochos, C. L.; Choulis, S. A. Prog. Polym. Sci. 2011, 36, 1326–1414.
Colella, S.; Melcarne, G.; Mazzeo, M.; Gigli, G.; Grisorio, R.; Suranna, G. P.; Mastrorilli, P.
Polymer 2011, 52, 2740–2746.
Dang, M. T.; Hirsch, L.; Wantz, G. Adv. Mater. 2011, 23, 3597–3602.
Daoust, G.; Leclerc, M. Macromolecules 1991, 24, 455–459.
Deng, P.; Xiong, J.; Li, S.; Wu, Y.; Yang, J.; Zhang, Q. Chin. J. Chem. 2014, 32, 521–526.
Deng, Y.; Liu, J.; Wang, J.; Liu, L.; Li, W.; Tian, H.; Zhang, X.; Xie, Z.; Geng, Y.; Wang, F.
Adv. Mater. 2013, 26, 471–476.
Deng, Z.; Chen, L.; Chen, Y. J. Polym. Sci. Part Polym. Chem. 2013, 51, 4885–4893.
Deng, Z.; Chen, L.; Wu, F.; Chen, Y. J. Phys. Chem. C 2014, 118, 6038–6045.
Duan, C.; Hu, X.; Chen, K.-S.; Yip, H.-L.; Li, W.; Huang, F.; Jen, A. K.-Y.; Cao, Y. Sol.
Energy Mater. Sol. Cells 2012, 97, 50–58.
Duan, C.; Huang, F.; Cao, Y. J. Mater. Chem. 2012, 22, 10416–10434.
Fang, L.; Zhou, Y.; Yao, Y.-X.; Diao, Y.; Lee, W.-Y.; Appleton, A. L.; Allen, R.; Reinspach,
J.; Mannsfeld, S. C. B.; Bao, Z. Chem. Mater. 2013, 25, 4874–4880.
Gallego, J.; Pérez-Foullerat, D.; Mendicuti, F.; Mattice, W. L. J. Polym. Sci. Part B Polym.
!
192
Phys. 2001, 39, 1272–1281.
Gedefaw, D.; Zhou, Y.; Hellström, S.; Lindgren, L.; Andersson, L. M.; Zhang, F.; Mammo,
W.; Inganäs, O.; Andersson, M. R. J. Mater. Chem. 2009, 19, 5359–5363.
Goubard, F.; Wantz, G. Polym. Int. 2013, 63, 1362-1367.
Grisorio, R.; Allegretta, G.; Romanazzi, G.; Suranna, G. P.; Mastrorilli, P.; Mazzeo, M.;
Cezza, M.; Carallo, S.; Gigli, G. Macromolecules 2012, 45, 6396–6404.
Guo, X.; Ortiz, R. P.; Zheng, Y.; Kim, M.-G.; Zhang, S.; Hu, Y.; Lu, G.; Facchetti, A.;
Marks, T. J. J. Am. Chem. Soc. 2011, 133, 13685–13697.
Guo, X.; Watson, M. D. Org. Lett. 2008, 10, 5333–5336.
Habaue, S.; Isobe, Y.; Okamoto, Y. Tetrahedron 2002, 58, 8205–8209.
He, J.; Su, Z.; Yan, B.; Xiang, L.; Wang, Y. J. Macromol. Sci. Part A 2007, 44, 989–993.
He, X.; Mukherjee, S.; Watkins, S.; Chen, M.; Qin, T.; Thomsen, L.; Ade, H.; McNeill, C. R.
J. Phys. Chem. C 2014.
He, Y.; Wang, X.; Zhang, J.; Li, Y. Macromol. Rapid Commun. 2009, 30, 45–51.
He, Z.; Zhong, C.; Huang, X.; Wong, W.-Y.; Wu, H.; Chen, L.; Su, S.; Cao, Y. Adv. Mater.
2011, 23, 4636–4643.
He, Z.; Zhong, C.; Su, S.; Xu, M.; Wu, H.; Cao, Y. Nat. Photonics 2012, 6, 591–595.
Helgesen, M.; Søndergaard, R.; Krebs, F. C. J. Mater. Chem. 2009, 20, 36–60.
Hendriks, K. H.; Heintges, G. H. L.; Wienk, M. M.; Janssen, R. A. J. J. Mater. Chem. A 2014,
2, 17899–17905.
Hendriks, K. H.; Li, W.; Wienk, M. M.; Janssen, R. A. J. Adv. Energy Mater. 2013, 3, 674–
679.
Hollinger, J.; Jahnke, A. A.; Coombs, N.; Seferos, D. S. J. Am. Chem. Soc. 2010, 132, 8546–
8547.
Hollinger, J.; Sun, J.; Gao, D.; Karl, D.; Seferos, D. S. Macromol. Rapid Commun. 2013, 34,
!
193
437–441.
Honda, S.; Nogami, T.; Ohkita, H.; Benten, H.; Ito, S. ACS Appl. Mater. Interfaces 2009, 1,
804–810.
Honda, S.; Ohkita, H.; Benten, H.; Ito, S. Adv. Energy Mater. 2011, 1, 588–598.
Hou, J.; Tan, Z.; He, Y.; Yang, C.; Li, Y. Macromolecules 2006, 39, 4657–4662.
Hou, J.; Tan, Z.; Yan, Y.; He, Y.; Yang, C.; Li, Y. J. Am. Chem. Soc. 2006, 128, 4911–4916.
Hou, J.; Yang, C.; Qiao, J.; Li, Y. Synth. Met. 2005, 150, 297–304.
Hou, Q.; Xu, Y.; Yang, W.; Yuan, M.; Peng, J.; Cao, Y. J. Mater. Chem. 2002, 12, 2887–
2892.
Hu, X.; Xu, L. Polymer 2000, 41, 9147–9154.
Hu, X.; Zuo, L.; Fu, W.; Larsen-Olsen, T. T.; Helgesen, M.; Bundgaard, E.; Hagemann, O.;
Shi, M.; Krebs, F. C.; Chen, H. J. Mater. Chem. 2012, 22, 15710–15716.
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,
2820–2825.
Huang, Y.; Zhang, M.; Chen, H.; Wu, F.; Cao, Z.; Zhang, L.; Tan, S. J. Mater. Chem. A 2014,
2, 5218–5223.
Huo, L.; Hou, J. Polym. Chem. 2011, 2, 2453–2461.
Huo, L.; Tan, Z.; Wang, X.; Zhou, Y.; Han, M.; Li, Y. J. Polym. Sci. Part Polym. Chem. 2008,
46, 4038–4049.
Hüttner, S.; Sommer, M.; Thelakkat, M. Appl. Phys. Lett. 2008, 92, 093302.
Idota, N.; Nagase, K.; Tanaka, K.; Okano, T.; Annaka, M. Langmuir 2010, 26, 17781–17784.
Intemann, J. J.; Yao, K.; Yip, H.-L.; Xu, Y.-X.; Li, Y.-X.; Liang, P.-W.; Ding, F.-Z.; Li, X.;
Jen, A. K.-Y. Chem. Mater. 2013, 25, 3188–3195.
!
194
Ishizawa, H.; Nakano, T.; Yade, T.; Tsuji, M.; Nakagawa, O.; Yamaguchi, T. J. Polym. Sci.
Part Polym. Chem. 2004, 42, 4656–4665.
Isobe, Y.; Fujioka, D.; Habaue, S.; Okamoto, Y. J. Am. Chem. Soc. 2001, 123, 7180–7181.
Janietz, S.; Krueger, H.; Schleiermacher, H.-F.; Würfel, U.; Niggemann, M. Macromol.
Chem. Phys. 2009, 210, 1493–1503.
Jiang, J.-M.; Chen, H.-C.; Lin, H.-K.; Yu, C.-M.; Lan, S.-C.; Liu, C.-M.; Wei, K.-H. Polym.
Chem. 2013, 4, 5321–5328.
Jørgensen, M.; Norrman, K.; Krebs, F. C. Sol. Energy Mater. Sol. Cells 2008, 92, 686–714.
Jung, I. H.; Kim, S. H.; Jeong, E.; Yang, R.; Lee, K.; Woo, H. Y.; Shim, H.-K. J. Polym. Sci.
Part Polym. Chem. 2011, 49, 1248–1255.
Jung, J. W.; Liu, F.; Russell, T. P.; Jo, W. H. Energy Environ. Sci. 2013, 6, 3301–3307.
Kamigaito, M.; Satoh, K. Macromolecules 2008, 41, 269–276.
Kang, T. E.; Cho, H.-H.; Kim, H. jun; Lee, W.; Kang, H.; Kim, B. J. Macromolecules 2013,
46, 6806–6813.
Kang, T. E.; Kim, K.-H.; Kim, B. J. J. Mater. Chem. A 2014, 2, 15252–15267.
Karali, A.; Dais, P.; Mikros, E.; Heatley, F. Macromolecules 2001, 34, 5547–5554.
Keegstra, M. A.; Peters, T. H. A.; Brandsma, L. Tetrahedron 1992, 48, 3633–3652.
Khlyabich, P. P.; Burkhart, B.; Ng, C. F.; Thompson, B. C. Macromolecules 2011, 44, 5079–
5084.
Khlyabich, P. P.; Burkhart, B.; Rudenko, A. E.; Thompson, B. C. Polymer 2013, 54, 5267–
5298.
Khlyabich, P. P.; Burkhart, B.; Thompson, B. C. J. Am. Chem. Soc. 2011, 133, 14534–14537.
Khlyabich, P. P.; Burkhart, B.; Thompson, B. C. J. Am. Chem. Soc. 2012, 134, 9074–9077.
Khlyabich, P. P.; Rudenko, A. E.; Street, R. A.; Thompson, B. C. ACS Appl. Mater.
Interfaces 2014, 6, 9913–9919.
!
195
Khlyabich, P. P.; Rudenko, A. E.; Thompson, B. C. J. Polym. Sci. Part Polym. Chem. 2014,
52, 1055–1058.
Kim, D. H.; Song, H. J.; Heo, S. W.; Song, K. W.; Moon, D. K. Sol. Energy Mater. Sol. Cells
2014, 120, Part A, 94–101.
Kim, J.-H.; Kim, H. U.; Mi, D.; Jin, S.-H.; Shin, W. S.; Yoon, S. C.; Kang, I.-N.; Hwang,
D.-H. Macromolecules 2012, 45, 2367–2376.
Ko, S.; Mondal, R.; Risko, C.; Lee, J. K.; Hong, S.; McGehee, M. D.; Brédas, J.-L.; Bao, Z.
Macromolecules 2010, 43, 6685–6698.
Koeckelberghs, G.; Vangheluwe, M.; Doorsselaere, K. V.; Robijns, E.; Persoons, A.;
Verbiest, T. Macromol. Rapid Commun. 2006, 27, 1920–1925.
Koeckelberghs, G.; Vangheluwe, M.; Samyn, C.; Persoons, A.; Verbiest, T. Macromolecules
2005, 38, 5554–5559.
Koppe, M.; Egelhaaf, H.-J.; Dennler, G.; Scharber, M. C.; Brabec, C. J.; Schilinsky, P.; Hoth,
C. N. Adv. Funct. Mater. 2010, 20, 338–346.
Kotowski, D.; Luzzati, S.; Bianchi, G.; Calabrese, A.; Pellegrino, A.; Po, R.; Schimperna, G.;
Tacca, A. J. Mater. Chem. A 2013, 1, 10736–10744.
Kozycz, L. M.; Gao, D.; Seferos, D. S. Macromolecules 2013, 46, 613–621.
Kozycz, L. M.; Gao, D.; Tilley, A. J.; Seferos, D. S. J. Polym. Sci. Part Polym. Chem. 2014,
52, 3337-3345.
Krasovskiy, A.; Krasovskaya, V.; Knochel, P. Angew. Chem. Int. Ed. 2006, 45, 2958–2961.
Kumar, P.; Chand, S. Prog. Photovolt. Res. Appl. 2012, 20, 377–415.
Lang, A. S.; Neubig, A.; Sommer, M.; Thelakkat, M. Macromolecules 2010, 43, 7001–7010.
Lange, A.; Krueger, H.; Ecker, B.; Tunc, A. V.; von Hauff, E.; Morana, M. J. Polym. Sci.
Part Polym. Chem. 2012, 50, 1622–1635.
Lee, J.; Yun, M. H.; Kim, J.; Kim, J. Y.; Yang, C. Macromol. Rapid Commun. 2012, 33,
140–145.
!
196
Lee, J.-Y.; Choi, M.-H.; Song, H.-J.; Moon, D.-K. J. Polym. Sci. Part Polym. Chem. 2010, 48,
4875–4883.
Lee, M.; Cho, B.-K.; Zin, W.-C. Chem. Rev. 2001, 101, 3869–3892.
Lévêque, P.; Biniek, L.; Fall, S.; Chocos, C. L.; Leclerc, N.; Heiser, T. Energy Procedia
2012, 31, 38–45.
Li, G.; Shrotriya, V.; Huang, J.; Yao, Y.; Moriarty, T.; Emery, K.; Yang, Y. Nat. Mater. 2005,
4, 864–868.
Li, G.; Zhu, R.; Yang, Y. Nat. Photonics 2012, 6, 153–161.
Li, H.; Liu, F.; Wang, X.; Gu, C.; Wang, P.; Fu, H. Macromolecules 2013, 46, 9211–9219.
Li, J.; Ong, K.-H.; Lim, S.-L.; Ng, G.-M.; Tan, H.-S.; Chen, Z.-K. Chem. Commun. 2011, 47,
9480–9482.
Li, J.; Ong, K.-H.; Sonar, P.; Lim, S.-L.; Ng, G.-M.; Wong, H.-K.; Tan, H.-S.; Chen, Z.-K.
Polym. Chem. 2013, 4, 804.
Li, K.-C.; Hsu, Y.-C.; Lin, J.-T.; Yang, C.-C.; Wei, K.-H.; Lin, H.-C. J. Polym. Sci. Part
Polym. Chem. 2008, 46, 4285–4304.
Li, Y. Acc. Chem. Res. 2012, 45, 723–733.
Li, Y.; Sonar, P.; Murphy, L.; Hong, W. Energy Environ. Sci. 2013, 6, 1684–1710.
Li, Y.; Zou, Y. Adv. Mater. 2008, 20, 2952–2958.
Liang, Y.; Xiao, S.; Feng, D.; Yu, L. J. Phys. Chem. C 2008, 112, 7866–7871.
Liang, Y.; Xu, Z.; Xia, J.; Tsai, S.-T.; Wu, Y.; Li, G.; Ray, C.; Yu, L. Adv. Mater. 2010, 22,
E135–E138.
Liao, S.-H.; Jhuo, H.-J.; Cheng, Y.-S.; Chen, S.-A. Adv. Mater. 2013, 25, 4766–4771.
Lim, B.; Bloking, J. T.; Ponec, A.; McGehee, M. D.; Sellinger, A. ACS Appl. Mater.
Interfaces 2014, 6, 6905–6913.
Lim, D. C.; Kim, K.-D.; Park, S.-Y.; Hong, E. M.; Seo, H. O.; Lim, J. H.; Lee, K. H.; Jeong,
!
197
Y.; Song, C.; Lee, E.; Kim, Y. D.; Cho, S. Energy Environ. Sci. 2012, 5, 9803–
9807.
Lim, Y.; Ihn, S.-G.; Bulliard, X.; Yun, S.; Chung, Y.; Kim, Y.; Chang, H.; Choi, Y. S.
Polymer 2012, 53, 5275–5284.
Lin, W.; Baron, O.; Knochel, P. Org. Lett. 2006, 8, 5673–5676.
Lindner, S. M.; Hüttner, S.; Chiche, A.; Thelakkat, M.; Krausch, G. Angew. Chem. Int. Ed.
2006, 45, 3364–3368.
Lindner, S. M.; Kaufmann, N.; Thelakkat, M. Org. Electron. 2007, 8, 69–75.
Liu, C.; Wang, K.; Hu, X.; Yang, Y.; Hsu, C.-H.; Zhang, W.; Xiao, S.; Gong, X.; Cao, Y.
ACS Appl. Mater. Interfaces 2013, 5, 12163–12167.
Liu, P.; Zhang, K.; Liu, F.; Jin, Y.; Liu, S.; Russell, T. P.; Yip, H.-L.; Huang, F.; Cao, Y.
Chem. Mater. 2014.
Lu, L.; Yu, L. Adv. Mater. 2014, 26, 4413–4430.
Luo, J.; Hou, Q.; Chen, J.; Cao, Y. Synth. Met. 2006, 156, 470–475.
Ma, W.; Yang, C.; Gong, X.; Lee, K.; Heeger, A. J. Adv. Funct. Mater. 2005, 15, 1617–1622.
Maria, S.; Susha, A. S.; Sommer, M.; Talapin, D. V.; Rogach, A. L.; Thelakkat, M.
Macromolecules 2008, 41, 6081–6088.
Meng, B.; Fang, G.; Fu, Y.; Xie, Z.; Wang, L. Nanotechnology 2013, 24, 484004.
Mikroyannidis, J. A.; Tsagkournos, D. V.; Balraju, P.; Sharma, G. D. J. Power Sources 2011,
196, 2364–2372.
Mosrin, M.; Knochel, P. Org. Lett. 2008, 10, 2497–2500.
Nakabayashi, K.; Mori, H. Materials 2014, 7, 3274–3290.
Nelson, J. Mater. Today 2011, 14, 462–470.
Nielsen, C. B.; Ashraf, R. S.; Schroeder, B. C.; D’Angelo, P.; Watkins, S. E.; Song, K.;
Anthopoulos, T. D.; McCulloch, I. Chem. Commun. 2012, 48, 5832–5834.
!
198
Okamoto, Y.; Habaue, S.; Isobe, Y.; Suito, Y. Macromol. Symp. 2003, 195, 75–80.
Osaka, I.; Kakara, T.; Takemura, N.; Koganezawa, T.; Takimiya, K. J. Am. Chem. Soc. 2013,
135, 8834–8837.
Osaka, I.; McCullough, R. D. Acc. Chem. Res. 2008, 41, 1202–1214.
Palermo, E. F.; McNeil, A. J. Macromolecules 2012, 45, 5948–5955.
Park, J. S.; Ryu, T. I.; Song, M.; Yoon, K.-J.; Lee, M. J.; Shin, I. A.; Lee, G.-D.; Lee, J. W.;
Gal, Y.-S.; Jin, S.-H. J. Polym. Sci. Part Polym. Chem. 2008, 46, 6175–6184.
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, 297–302.
Paulsen, B. D.; Speros, J. C.; Claflin, M. S.; Hillmyer, M. A.; Frisbie, C. D. Polym. Chem.
2014, 5, 6287–6294.
Peng, Q.; Park, K.; Lin, T.; Durstock, M.; Dai, L. J. Phys. Chem. B 2008, 112, 2801–2808.
Qin, R.; Li, W.; Li, C.; Du, C.; Veit, C.; Schleiermacher, H.-F.; Andersson, M.; Bo, Z.; Liu,
Z.; Inganäs, O.; Wuerfel, U.; Zhang, F. J. Am. Chem. Soc. 2009, 131, 14612–
14613.
Qu, S.; Tian, H. Chem. Commun. 2012, 48, 3039–3051.
Ray, B.; Isobe, Y.; Matsumoto, K.; Habaue, S.; Okamoto, Y.; Kamigaito, M.; Sawamoto, M.
Macromolecules 2004, 37, 1702–1710.
Romulus, J.; Henssler, J. T.; Weck, M. Macromolecules 2014, 47, 5437–5449.
Rudenko, A. E.; Khlyabich, P. P.; Thompson, B. C. ACS Macro Lett. 2014, 3, 387–392.
Rudenko, A. E.; Noh, S.; Thompson, B. C. Nanotechnology 2013, 24, 484002.
Rudenko, A. E.; Thompson, B. C. J. Polym. Sci. Part Polym. Chem. 2015, 53, 135–147.
Savagatrup, S.; Makaram, A. S.; Burke, D. J.; Lipomi, D. J. Adv. Funct. Mater. 2014, 24,
1169–1181.
Savagatrup, S.; Printz, A. D.; Rodriquez, D.; Lipomi, D. J. Macromolecules 2014, 47, 1981–
!
199
1992.
Scaria, R.; Dhawan, S. K.; Chand, S. Synth. Met. 2014, 191, 168–176.
Schilinsky, P.; Asawapirom, U.; Scherf, U.; Biele, M.; Brabec, C. J. Chem. Mater. 2005, 17,
2175–2180.
Schulz, G. L.; Holdcroft, S. Chem. Mater. 2008, 20, 5351–5355.
Sharma, G. D.; Singh, S. P.; Roy, M. S.; Mikroyannidis, J. A. Org. Electron. 2012, 13, 1756–
1762.
Sharma, S. S.; Sharma, G. D.; Mikroyannidis, J. A. Sol. Energy Mater. Sol. Cells 2011, 95,
1219–1223.
Sheina, E. E.; Khersonsky, S. M.; Jones, E. G.; McCullough, R. D. Chem. Mater. 2005, 17,
3317–3319.
Shi, C.; Yao, Y.; Yang; Pei, Q. J. Am. Chem. Soc. 2006, 128, 8980–8986.
Shi, F.; Fang, G.; Liang, F.; Wang, L.; Mu, Z.; Zhang, X.; Xie, Z.; Su, Z. Eur. Polym. J. 2010,
46, 1770–1777.
Shirota, Y.; Kageyama, H. Chem. Rev. 2007, 107, 953–1010.
Shrotriya, V.; Li, G.; Yao, Y.; Moriarty, T.; Emery, K.; Yang, Y. Adv. Funct. Mater. 2006,
16, 2016–2023.
Singh, R. P.; Kushwaha, O. S. Macromol. Symp. 2013, 327, 128–149.
Sivula, K.; Luscombe, C. K.; Thompson, B. C.; Fréchet, J. M. J. J. Am. Chem. Soc. 2006, 128,
13988–13989.
Skotheim, T. A.; Reynolds, J. Conjugated Polymers: Theory, Synthesis, Properties, and
Characterization; CRC Press, 2006.
Sommer, M.; Lindner, S. M.; Thelakkat, M. Adv. Funct. Mater. 2007, 17, 1493–1500.
Son, H. J.; He, F.; Carsten, B.; Yu, L. J. Mater. Chem. 2011, 21, 18934–18945.
Song, J.; Zhang, C.; Li, C.; Li, W.; Qin, R.; Li, B.; Liu, Z.; Bo, Z. J. Polym. Sci. Part Polym.
!
200
Chem. 2010, 48, 2571–2578.
Song, S.; Kim, G.; Kang, I.; Jin, Y.; Kim, I.; Kim, J. Y.; Suh, H. J. Polym. Sci. Part Polym.
Chem. 2011, 49, 3751–3758.
Song, S.; Kim, J.; Shim, J.; Kim, J.; Hoon Lee, B.; Jin, Y.; Kim, I.; Lee, K.; Suh, H. Sol.
Energy Mater. Sol. Cells 2012, 98, 323–330.
Song, S.; Park, S. H.; Jin, Y.; Goo, Y.; Kim, I.; Lee, H.; Lee, K.; Suh, H. Synth. Met. 2010,
160, 2618–2622.
Street, R. A.; Davies, D.; Khlyabich, P. P.; Burkhart, B.; Thompson, B. C. J. Am. Chem. Soc.
2013, 135, 986–989.
Su, Y.-W.; Lan, S.-C.; Wei, K.-H. Mater. Today 2012, 15, 554–562.
Sun, Y.; Seo, J. H.; Takacs, C. J.; Seifter, J.; Heeger, A. J. Adv. Mater. 2011, 23, 1679–1683.
Suspène, C.; Miozzo, L.; Choi, J.; Gironda, R.; Geffroy, B.; Tondelier, D.; Bonnassieux, Y.;
Horowitz, G.; Yassar, A. J. Mater. Chem. 2012, 22, 4511–4518.
Svensson, M.; Zhang, F.; Veenstra, S. c.; Verhees, W. j. h.; Hummelen, J. c.; Kroon, J. m.;
Inganäs, O.; Andersson, M. r. Adv. Mater. 2003, 15, 988–991.
Tahk, D.; Lee, H. H.; Khang, D.-Y. Macromolecules 2009, 42, 7079–7083.
Tamilavan, V.; Song, M.; Ban, T.-W.; Jin, S.-H.; Hyun, M. H. Polym. Bull. 2012, 69, 439–
454.
Tamilavan, V.; Song, M.; Jin, S.-H.; Hyun, M. H. Macromol. Res. 2013, 21, 406–413.
Tan, Z.; Tang, R.; Zhou, E.; He, Y.; Yang, C.; Xi, F.; Li, Y. J. Appl. Polym. Sci. 2008, 107,
514–521.
Thompson, B. C.; Fréchet, J. M. J. Angew. Chem. Int. Ed. 2008, 47, 58–77.
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, 29–54.
Thompson, B. C.; Kim, B. J.; Kavulak, D. F.; Sivula, K.; Mauldin, C.; Fréchet, J. M. J.
Macromolecules 2007, 40, 7425–7428.
!
201
Topham, P. D.; Parnell, A. J.; Hiorns, R. C. J. Polym. Sci. Part B Polym. Phys. 2011, 49,
1131–1156.
Tsai, J.-H.; Chueh, C.-C.; Chen, W.-C.; Yu, C.-Y.; Hwang, G.-W.; Ting, C.; Chen, E.-C.;
Meng, H.-F. J. Polym. Sci. Part Polym. Chem. 2010, 48, 2351–2360.
Uryu, T.; Ohkawa, H.; Oshima, R. Macromolecules 1987, 20, 712–716.
Van den Bergh, K.; De Winter, J.; Gerbaux, P.; Verbiest, T.; Koeckelberghs, G. Macromol.
Chem. Phys. 2011, 212, 328–335.
Veldman, D.; İpek, Ö.; Meskers, S. C. J.; Sweelssen, J.; Koetse, M. M.; Veenstra, S. C.;
Kroon, J. M.; Bavel, S. S. van; Loos, J.; Janssen, R. A. J. J. Am. Chem. Soc. 2008,
130, 7721–7735.
Wakim, S.; Beaupré, S.; Blouin, N.; Aich, B.-R.; Rodman, S.; Gaudiana, R.; Tao, Y.; Leclerc,
M. J. Mater. Chem. 2009, 19, 5351–5358.
Wang, H.-J.; Chan, L.-H.; Chen, C.-P.; Lin, S.-L.; Lee, R.-H.; Jeng, R.-J. Polymer 2011, 52,
326–338.
Wang, Z.; Li, Z.; Zhang, H.; Liu, W.; Xu, X.; Si, C.; Cui, Y.; Sun, Q.; Wang, H.; Shi, F.; Hao,
Y. Macromol. Chem. Phys. 2014, 215, 2119–2124.
Wen, S.; Pei, J.; Zhou, Y.; Xue, L.; Xu, B.; Li, Y.; Tian, W. J. Polym. Sci. Part Polym. Chem.
2009, 47, 1003–1012.
Wong, W. W. H.; Subbiah, J.; Puniredd, S. R.; Pisula, W.; Jones, D. J.; Holmes, A. B. Polym.
Chem. 2014, 5, 1258–1263.
Wu, P.-T.; Ren, G.; Jenekhe, S. A. Macromolecules 2010, 43, 3306–3313.
Xia, Y.; Luo, J.; Deng, X.; Li, X.; Li, D.; Zhu, X.; Yang, W.; Cao, Y. Macromol. Chem. Phys.
2006, 207, 511–520.
Xu, B.; Noh, S.; Thompson, B. C. Macromolecules 2014, 47, 5029–5039.
Yang, L.; Yan, L.; You, W. J. Phys. Chem. Lett. 2013, 4, 1802–1810.
Yang, L.; Zhou, H.; Price, S. C.; You, W. J. Am. Chem. Soc. 2012, 134, 5432–5435.
!
202
Yang, L.; Zhou, H.; You, W. J. Phys. Chem. C 2010, 114, 16793–16800.
Yang, R.; Tian, R.; Yan, J.; Zhang, Y.; Yang, J.; Hou, Q.; Yang, W.; Zhang, C.; Cao, Y.
Macromolecules 2005, 38, 244–253.
Yang, T.; Wang, M.; Duan, C.; Hu, X.; Huang, L.; Peng, J.; Huang, F.; Gong, X. Energy
Environ. Sci. 2012, 5, 8208–8214.
Yao, K.; Chen, L.; Chen, Y.; Li, F.; Ren, X.; Wang, H.; Li, Y. Polym. Chem. 2012, 3, 710–
717.
Ye, L.; Xia, H.; Xiao, Y.; Xu, J.; Miao, Q. RSC Adv. 2013, 4, 1087–1092.
You, J.; Chen, C.-C.; Hong, Z.; Yoshimura, K.; Ohya, K.; Xu, R.; Ye, S.; Gao, J.; Li, G.;
Yang, Y. Adv. Mater. 2013, 25, 3973–3978.
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, C.-Y.; Chen, C.-P.; Chan, S.-H.; Hwang, G.-W.; Ting, C. Chem. Mater. 2009, 21, 3262–
3269.
Yuan, M.-C.; Chiu, M.-Y.; Chiang, C.-M.; Wei, K.-H. Macromolecules 2010, 43, 6270–
6277.
Yusoff, A. R. bin M.; Kim, D.; Kim, H. P.; Shneider, F. K.; Silva, W. J. da; Jang, J. Energy
Environ. Sci. 2014, 8, 303–316.
Zhan, X.; Zhu, D. Polym. Chem. 2010, 1, 409–419.
Zhang, G.; Fu, Y.; Qiu, L.; Xie, Z. Polymer 2012, 53, 4407–4412.
Zhang, M.; Cao, Z.; Wu, F.; Chen, H.; Zhang, M.; Qiao, H.; Tan, S. Eur. Polym. J. 2014, 57,
83–90.
Zhang, M.; Gu, Y.; Guo, X.; Liu, F.; Zhang, S.; Huo, L.; Russell, T. P.; Hou, J. Adv. Mater.
2013, 25, 4944–4949.
Zhang, M.; Wu, F.; Cao, Z.; Shen, T.; Chen, H.; Li, X.; Tan, S. Polym. Chem. 2014, 5, 4054–
4060.
!
203
Zhao, C.; Chen, X.; Zhang, Y.; Ng, M. J. Polym. Sci. Part Polym. Chem. 2008, 46, 2680–
2688.
Zhou, E.; Tan, Z.; Huo, L.; He, Y.; Yang, C.; Li, Y. J. Phys. Chem. B 2006, 110, 26062–
26067.
Zhou, J.; Xie, S.; Amond, E. F.; Becker, M. L. Macromolecules 2013, 46, 3391–3394.
Zhou, Q.; Hou, Q.; Zheng, L.; Deng, X.; Yu, G.; Cao, Y. Appl. Phys. Lett. 2004, 84, 1653–
1655.
Zhou, Y.; Kurosawa, T.; Ma, W.; Guo, Y.; Fang, L.; Vandewal, K.; Diao, Y.; Wang, C.; Yan,
Q.; Reinspach, J.; Mei, J.; Appleton, A. L.; Koleilat, G. I.; Gao, Y.; Mannsfeld, S.
C. B.; Salleo, A.; Ade, H.; Zhao, D.; Bao, Z. Adv. Mater. 2014, 26, 3767–3772.
Zhu, Z.; Waller, D.; Gaudiana, R. J. Macromol. Sci. Part A 2007, 44, 1249–1253.
Zhu, Z.; Waller, D.; Gaudiana, R.; Morana, M.; Mühlbacher, D.; Scharber, M.; Brabec, C.
Macromolecules 2007, 40, 1981–1986.
204
APPENDIX 1. Fine Tuning of Polymer Properties by
Incorporating Strongly Electron-Donating 3-Hexyloxythiophene
Units into Random and Semi-random Copolymers
A1.1 Materials and Methods
The monomers 2-bromo-5-trimethyltin-3-hexylthiophene, 2,5-bis(trimethyltin)thiophene
and 2,5-di(2-ethylhexyl)-3,6-bis(5-bromothiophene-2-yl)pyrrolo[3,4-c]-pyrrole-1,4-dione
were synthesized following published procedures.
1,2
All reagents from commercial sources
were used as received unless otherwise noted. N-Bromosuccinimide (NBS) was
recrystallized from hot water. CuBr was purified by washing three times with acetic acid,
twice with ethanol and twice with diethyl ether, and then stored in a desiccator. Solvents
were purchased from VWR and used without further purification except for THF, which was
dried over sodium/benzophenone before being distilled. All reactions were performed under
dry N
2
in glassware that was pre-dried in oven, unless otherwise noted. Flash
chromatography was performed on a Teledyne CombiFlash R
f
instrument with RediSep R
f

normal phase disposable columns.
1
H NMR spectra of small molecules were recorded on a Varian Mercury 400 NMR
Spectrometer.
13
C NMR spectrum of 2-bromo-5-trimethyltin-3-hexyloxythiophene and
1
H
NMR spectra of polymers were obtained on a Varian VNMRS-500 instrument at 50 °C. All
NMR spectra were recorded in CDCl
3
.
205
Number average molecular weight (M
n
) and polydispersity (PDI) were determined by size
exclusion chromatography (SEC) using a Viscotek GPC Max VE 2001 separation module
and a Viscotek TDA 305 RI detector, with 70 °C HPLC grade o-DCB as eluent at a flow rate
of 1 mL/min on one 300 × 7.8 mm TSK-Gel GMH
HR
-H column (Tosoh Corp). The
instrument was calibrated vs polystyrene standards (1050−3,800,000 g/mol), and data were
analyzed using OmniSec 4.6.0 software. To prepare polymer samples for SEC measurements,
a polymer was dissolved in HPLC grade o-dichlorobenzene (o-DCB) at a concentration of
1.0 mg/mL and then stirred overnight at 60 °C prior to filtering through a 0.2 mm PTFE filter.
Cyclic voltammetry was performed on a 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), and a Pt wire counter electrode was purged with N
2

and maintained under N
2
atmosphere during all measurements. Polymer films were made by
drop-casting an o-DCB solution of polymer (10 mg/mL) and tetrabutylammonium
hexafluorophosphate (TBAPF
6
) (30 mg/mL) onto the Pt wire and drying under N
2
prior to
measurement. Acetonitrile was distilled over CaH
2
prior to use as the solvent, and TBAPF
6

(0.1 M) was used as the supporting electrolyte.
UV−Vis absorption spectra were obtained on a Perkin-Elmer Lambda 950
spectrophotometer. For thin film measurements, polymers were spin-coated onto pre-cleaned
glass slides from 10 mg/mL o-DCB solutions. Thickness of the samples and GIXRD
measurements were obtained using Rigaku diffractometer Ultima IV using a Cu Kα radiation
206
source (λ = 1.54 Å) in the reflectivity and grazing incidence X-ray diffraction mode,
respectively.
A1.2 Synthetic Procedures
Detailed synthetic procedures to prepare compounds shown in Scheme 2.2 are described as
follows.
3-methoxythiophene (1). Modified from literature.
3
A solution of 30 wt% sodium
methoxide in methanol (5.5 mL, 29.7 mmol, 1.6 eq) was added to a 3-neck round bottom
flask under N
2
. Then 3-bromothiophene (3.00 g, 18.4 mmol) and CuBr (0.265 g, 1.84 mmol,
0.1 eq) were added successively with stirring. The mixture was allowed to stir at 65 °C
overnight. After cooling, the reaction mixture was poured into water and extracted three
times with diethyl ether. The ether solution was dried with MgSO
4
and then concentrated
under reduced pressure. Purification by column chromatography with a gradient of hexanes
and DCM (10-50% DCM) afforded the product as a colorless liquid (1.18g, 10.4 mmol, 56%).
1
H NMR (400 MHz, CDCl
3
): δ 7.18 (dd, 1H), 6.76 (dd, 1H), 6.25 (dd, 1.01H), 3.82 (s,
3.76H).
3-hexyloxythiophene (2). Modified from literature.
4
A 3-neck flask was equipped with a
condenser, which connected to a dry N
2
line through a drying tube filled with molecular
sieves. To this flask, 10 mL toluene was added, followed by the sequential addition of 3-
methoxythiophene (1.18 g, 10.4 mmol), 1-hexanol (2.6 mL, 20.8 mmol, 2.0 eq) and p-
toluenesulfonic acid monohydrate (0.196 g, 1.03 mmol, 0.1 eq). The mixture was heated to
120 °C under N
2
for 20h. After cooling, the mixture was poured into 150 mL of water and
207
extracted three times with DCM. The organic layer was washed with water twice and then
dried over MgSO
4
. After solvent evaporation under reduced pressure, the residue was
purified by column chromatography with 4:1 hexanes/DCM to afford the product as a
colorless liquid (1.41 g, 7.66 mmol, 74%).
1
H NMR (400 MHz, CDCl
3
): δ 7.17 (dd, 1H),
6.76 (dd, 0.91H), 6.23 (dd, 0.95H), 3.94 (t, 2.47H), 1.77 (m, 2.38H), 1.55 (m, 2.35H), 1.34
(m, 4.58H), 0.91 (t, 3.29H).
2-bromo-3-hexyloxythiophene (3). Modified from literature.
4
NBS (1.38 g, 7.74 mmol,
1.01 eq) was added in one portion to 3-hexyloxythiophene (1.41 g, 7.66 mmol) in 10 mL
DMF at 0 °C. The mixture was warmed to room temperature and stirred overnight. It was
then poured into water followed by extraction with diethyl ether. The organic layer was
washed with water twice and then dried over MgSO
4
before being concentrated under
reduced pressure. Purification by column chromatography with 19:1 hexanes/ethyl acetate
afforded the product as a yellowish liquid (1.71 g, 6.51 mmol, 85 %).
1
H NMR (400 MHz,
CDCl
3
): δ 7.19 (d, 1H), 6.75 (d, 1.1H), 4.03 (t, 2.41H), 1.76 (m, 2.52H), 1.46 (m, 2.51H),
1.35 (m, 4.79H), 0.91 (t, 3.42H).
2-bromo-5-trimethyltin-3-hexyloxythiophene (4). To a flame-dried 3-neck flask, a
solution of 1.0 M 2,2,6,6-tetramethylpiperidinylmagnesium chloride lithium chloride
complex (TMPMgCl·LiCl) in toluene (3.5 mL, 3.5 mmol, 1.15 eq) was added under N
2

atmosphere. After cooled to -78 °C, 2-bromo-3-hexyloxythiophene (0.784 g, 2.98 mmol) was
added dropwise with stirring. The mixture was kept at -78 °C for 3 h before a 1.0 M solution
of trimethyltin chloride in hexane (4.5 mL, 4.5 mmol, 1.50 eq) was added slowly. The
mixture was allowed to warm and kept at room temperature overnight. The reaction was
208
quenched by adding 20 mL of water to the flask, followed by pouring into another 150 mL of
water. After extraction with diethyl ether, the organic layer was washed with water three
times and then dried over MgSO
4
. Purified product was achieved by vacuum distillation as a
yellowish oil (0.837 g, 1.92 mmol, 94%).
1
H NMR (400 MHz, CDCl
3
): δ 6.78 (s, 1H), 4.05 (t,
2.18H), 1.76 (m, 2.08H), 1.47 (m, 2.07H), 1.35 (m, 4.33H), 0.91 (t, 3.07H), 0.36 (s, 8.87H).
13
C NMR (125 MHz, CDCl
3
): δ 156.01, 137.03, 124.68, 96.12, 72.18, 31.53, 29.54, 25.51,
22.58, 14.02, 8.24.
General procedures for Stille Polymerization. All monomers were dissolved in dry
DMF to afford a 0.04 M solution. The solution was then degassed by purging with N
2
for 15
min before 4 mol% of Pd(PPh
3
)
4
(relative to the total moles of all comonomers) was added in
one portion. The solution was degassed for another 10 minutes and then allowed to stir at
95 °C for 48 h. Then the reaction mixture was cooled and precipitated into methanol.
Purification was achieved through Soxhlet extractions with a sequence of solvents, which are
described individually below. The last fraction was concentrated under reduced pressure,
reprecipitated in MeOH, vacuum filtered and then dried overnight under high vacuum.
P3HT. Soxhlet extracted with methanol, hexanes and chloroform. Yield 67%. M
n
= 16.1
kg/mol, PDI = 2.46.
1
H NMR (500 MHz, CDCl
3
): δ 6.98 (m, 1H), 2.81 (s, 1.86H), 2.58 (s,
0.11H), 1.72 (s, 2.01H), 1.46 (m, 3.40H), 1.37 (m, 4.27H), 0.92 (m, 3.05H).
P3HT
90
-co-3HOT
10
. Soxhlet extracted with methanol, hexanes and chloroform. Yield 65%.
M
n
= 19.5 kg/mol, PDI = 2.11.
1
H NMR (500 MHz, CDCl
3
): δ 6.98 (m, 1H), 4.17 (s, 0.21H),
2.81 (s, 1.72H), 2.58 (s, 0.14H), 1.89 (s, 0.18H), 1.70, 1.60 (m, 2.13H), 1.45, 1.36 (m, 7.13H),
0.92 (m, 2.81H).
209
P3HT
75
-co-3HOT
25
. Soxhlet extracted with methanol, hexanes and chloroform. Yield 50%.
M
n
= 19.4 kg/mol, PDI = 2.56.
1
H NMR (500 MHz, CDCl
3
): δ 6.98 (m, 1H), 4.18 (s, 0.49H),
2.80 (s, 1.41H), 2.58 (s, 0.1H), 1.89 (s, 0.52H), 1.70, 1.60 (m, 2.06H), 1.47, 1.36 (m, 6.45H),
0.92 (m, 2.87H).
P3HT
50
-co-3HOT
50
. Soxhlet extracted with methanol, hexanes and chloroform. Yield 50%.
M
n
= 11.9 kg/mol, PDI = 3.00.
1
H NMR (500 MHz, CDCl
3
): δ 6.98 (m, 1H), 4.17 (s, 1.04H),
2.81 (s, 1.02H), 2.58 (s, 0.03H), 1.89 (s, 1.01H), 1.72, 1.60 (m, 2.21H), 1.42, 1.36 (m, 6.29H),
0.92 (m, 3.02H).
P3HOT. Soxhlet extracted with methanol, hexanes and chloroform. Yield 30%. M
n
= 2.0
kg/mol, PDI = 1.41.
1
H NMR (500 MHz, CDCl
3
): δ 6.91 (s, br, 1H), 6.08 (s, 0.06H), 4.18 (s,
2.02H), 3.96 (s, 0.23H), 1.90 (m, 2.61H), 1.60 (m, 2.61H), 1.41 (m, 5.36H), 0.95 (s, 3.92H).
P3HTT-DPP. Soxhlet extracted with methanol, DCM and chloroform. Yield 74%. M
n
=
12.7 kg/mol, PDI = 2.56.
1
H NMR (500 MHz, CDCl
3
): δ 8.90 (s, 2H), 7.16-6.98 (m, 11.31H),
4.07 (s, 4.12H), 2.82 (s, 15.12H), 2.58 (s, 1.50H), 1.96 (s, 2.09H), 1.72 (m, 17.36H), 1.47,
1.36 (m, 87.05H), 0.92 (m, 39.78H).
P3HTT-HOT-DPP(75:5). Soxhlet extracted with methanol, DCM and chloroform. Yield
75%. M
n
= 14.7 kg/mol, PDI = 3.06.
1
H NMR (500 MHz, CDCl
3
): δ 8.90 (s, 2H), 7.16-6.99
(m, 9.95H), 4.18, 4.07 (m, 4.90H), 2.81 (s, 13.89H), 2.58 (s, 1.04H), 1.94 (m, 3.21H), 1.72
(m, 17.13H), 1.40, 1.36 (m, 76.75H), 0.92 (m, 38.21H).
P3HTT-HOT-DPP(70:10). Soxhlet extracted with methanol, DCM and chloroform. Yield
63%. M
n
= 18.5 kg/mol, PDI = 2.66.
1
H NMR (500 MHz, CDCl
3
): δ 8.90 (s, 2H), 7.16-6.99
210
(m, 10.64H), 4.18, 4.07 (m, 6.04H), 2.81 (s, 13.1H), 2.58 (s, 0.87H), 1.94 (m, 3.75H), 1.72
(m, 16.53H), 1.40, 1.36 (m, 80.86H), 0.92 (m, 39.27H).
P3HTT-HOT-DPP(65:15). Soxhlet extracted with methanol, DCM and chloroform. Yield
70%. M
n
= 14.7 kg/mol, PDI = 2.90.
1
H NMR (500 MHz, CDCl
3
): δ 8.90 (s, 2H), 7.16-6.99
(m, 10.75H), 4.18, 4.07 (m, 7.08H), 2.81 (s, 12.08H), 2.58 (s, 0.89H), 1.93 (m, 5.15H), 1.72
(m, 18.18H), 1.44, 1.37 (m, 79.70H), 0.93 (m, 42.53H).
P3HTT-HOT-DPP(60:20). Soxhlet extracted with methanol, DCM and chloroform. Yield
53%. M
n
= 15.8 kg/mol, PDI = 2.59.
1
H NMR (500 MHz, CDCl
3
): δ 8.90 (s, 2H), 7.16-6.98
(m, 9.10H), 4.17, 4.06 (m, 7.82H), 2.80 (s, 11.14H), 2.58 (s, 1.05H), 1.90 (m, 5.03H), 1.71
(m, 15.81H), 1.44, 1.36 (m, 90.59H), 0.92 (m, 43.23H).
P3HTT-HOT-DPP(40:40). Soxhlet extracted with methanol, DCM and chloroform. Yield
74%. M
n
= 5.3 kg/mol, PDI = 3.82.
1
H NMR (500 MHz, CDCl
3
): δ 8.90 (s, 2H), 6.97 (m, br,
11.45H), 4.16 (m, br, 12.59H), 2.80 (s, 7.88H), 2.58 (s, 0.62H), 1.94 (m, 12.79H), 1.72, 1.60
(m, 24.51H), 1.42, 1.37 (m, 98.38H), 0.93 (m, 52.95H).
A1.3 Device fabrication and characterization
All steps of device fabrication and characterization were performed in air. ITO-coated
glass substrates (10 Ω/sq, Thin Film Devices Inc.) were sequentially cleaned by sonication in
detergent, deionized water, tetrachloroethylene, acetone, and isopropyl alcohol, and dried in a
N
2
stream. About 40 nm of PEDOT:PSS (Baytron P VP AI 4083, filtered with a 0.45 μm
PVDF syringe filter from Pall Life Sciences) was first spin-coated on the pre-cleaned ITO-
coated glass substrates and annealed at 120 °C for 60 min under vacuum. Separate solutions
211
of the polymers and PC
61
BM were prepared in o-DCB for all polymers (except
chlorobenzene (CB) for P3HT, P3HT
75
-co-3HOT
25
and P3HT
50
-co-3HOT
50
). The solutions
were stirred for 6 h before they were mixed at the desired ratios and stirred overnight to form
a homogeneous solution. Subsequently, the polymer:PC
61
BM active layer was spin-coated
(filtered with a 0.45 μm PVDF syringe filter from Pall Life Sciences) on top of the
PEDOT:PSS layer. The concentrations of the blends were 10 mg/mL in polymer (with the
exception of 11 mg/mL for P3HTT-DPP:PC
61
BM and P3HTT-HOT-DPP(75:5):PC
61
BM,
and 7.5 mg/mL for P3HT
75
-co-3HOT
25
:PC
61
BM). For optimized conditions, devices of all
the P3HTT-HOT-DPP polymers, P3HT, and P3HT
90
-co-3HOT
10
were stored under N
2
for
40 min after spin-coating and then placed in the vacuum chamber for aluminum deposition.
The substrates were pumped down to high vacuum (< 9×10
-7
Torr) and aluminum (100 nm)
was thermally evaporated at 3–6 Å/s using a Denton Benchtop Turbo IV Coating System
onto the active layer through shadow masks to define the active area of the devices as 5.18
mm
2
. The P3HT device was annealed after aluminum deposition at 150 °C for 30 min.
P3HT
75
-co-3HOT
25
and P3HT
50
-co-3HOT
50
devices were annealed at 120 °C for 40 min and
20 min respectively in a N
2
atmosphere. 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 S11 solar simulator with a Xenon lamp (450
W) and an AM 1.5G filter was used as the solar simulator. An Oriel PV reference cell system
91150 V was used as the reference cell to calibrate the light intensity of the solar simulator
(to 100 mW/cm
2
).
External quantum efficiency (EQE) measurements and spectral mismatch correction.
EQE measurements were performed using a 300 W Xenon arc lamp (Newport Oriel),
212
chopped and filtered monochromatic light (250 Hz, 10 nm FWHM) from a Conerstone 260
1/4 M double grating monochromator (Newport 74125) 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)) was performed according
to Shrotriya et al.,
5
where mismatch factor is defined as:
(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 the equation
2:
(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 3,
integrated short-circuit current densities (J
sc,EQE
) can be obtained:
(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.
213
(4)
Mobility Measurements. Mobility of the polymers was measured using a hole-only device
configuration of ITO/PEDOT:PSS/Polymer/Al in the space charge limited current regime.
The device preparation was 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 space-charge-limited
(SCL) current and described by equation 5:
(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 thickness of
the polymer film. The series and contact resistance of the device (20 – 30 Ω) was measured
using a blank (ITO/PEDOT:PSS/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 XRD in the reflectivity mode.

214
A1.4 Structure Verifications of Small Molecules and Polymers

Figure A1.1.
1
H NMR of 3-methoxythiophene (1).
215

Figure A1.2.
1
H NMR of 3-hexyloxythiophene (2).
216

Figure A1.3.
1
H NMR of 2-bromo-3-hexyloxythiophene (3).
217

Figure A1.4.
1
H NMR of 2-bromo-5-trimethyltin-3-hexyloxythiophene (4).
218

Figure A1.5.
13
C NMR of 2-bromo-5-trimethyltin-3-hexyloxythiophene (4).
219

Figure A1.6.
1
H NMR of P3HT
.

220

Figure A1.7.
1
H NMR of P3HT
90
-co-3HOT
10
.
221

Figure A1.8.
1
H NMR of P3HT
75
-co-3HOT
25
.
222

Figure A1.9.
1
H NMR of P3HT
50
-co-3HOT
50
.
MeOH
223

Figure A1.10.
1
H NMR of homopolymer P3HOT.
Acetone
224

Figure A1.11.
1
H NMR of P3HTT-DPP.
225

Figure A1.12.
1
H NMR of P3HTT-HOT-DPP(75:5).
226

Figure A1.13.
1
H NMR of P3HTT-HOT-DPP(70:10).
227

Figure A1.14.
1
H NMR of P3HTT-HOT-DPP(65:15).
228

Figure A1.15.
1
H NMR of P3HTT-HOT-DPP(60:20).
229

Figure A1.16.
1
H NMR of P3HTT-HOT-DPP(40:40).
230
A1.5 Characterization of Polymer Samples

Figure A1.17. UV−vis absorption of (a) random P3HT-co-
P3HOT copolymers in CB solutions and (b) semi-random
P3HTT-HOT-DPP copolymers in o-DCB solutions: (i) P3HT,
(ii) P3HT
90
-co-HOT
10
, (iii) P3HT
75
-co-HOT
25
, (iv) P3HT
50
-co-
HOT
50
, (vi) P3HTT-DPP, (v) P3HOT, (vii) P3HTT-HOT-
DPP(75:5), (viii) P3HTT-HOT-DPP(70:10), (ix) P3HTT-
HOT-DPP(65:15), (x) P3HTT-HOT-DPP(60:20) and (xi)
P3HTT-HOT-DPP(40:40).

231
Table A1.1. Summary of UV-Vis Absorption Data of All Polymers
Sample
λ
max,abs

(nm)
solution
Absorptivity
(L/cm*g)
λ
max,abs
(nm)
film
Absorption
Coefficient
*10
-5
(cm
-1
)
Absorption
Onset (nm)
solution/film
P3HT 457 55.0 550 0.750 550/650
P3HT
90
-co-3HOT
10
469 52.8 537 0.745 570/670
P3HT
75
-co-3HOT
25
488 53.9 557 0.778 610/700
P3HT
50
-co-3HOT
50
527 53.3 590 0.794 660/730
P3HOT 593 46.2 632 0.488 720/780
P3HTT-DPP 438/673 31.1/44.4 512/683 0.384/0.600 740/830
P3HTT-HOT-
DPP(75:5)
444/676 27.4/42.5 512/687 0.407/0.655 760/850
P3HTT-HOT-
DPP(70:10)
452/679 28.3/42.1 512/692 0.436/0.645 770/860
P3HTT-HOT-
DPP(65:15)
456/682 28.0/43.5 522/696 0.402/0.646 785/870
P3HTT-HOT-
DPP(60:20)
465/685 26.7/41.5 538/703 0.421/0.654 790/880
P3HTT-HOT-
DPP(40:40)
489/704 24.2/41.7 --/729 --/0.654 870/920

232

Figure A1.18. CV traces of random P3HT-co-3HOT copolymers as well as homopolymers
P3HT and P3HOT.
233

Figure A1.19. CV traces of semi-random P3HTT-HOT-DPP copolymers.

234
Table A1.2. Summary of XRD Data of All Polymers
Sample 2θ (°) d (Å) FWHM (°)
P3HT 5.285 16.71 0.516
P3HT
90
-co-3HOT
10
5.211 16.94 0.483
P3HT
75
-co-3HOT
25
5.125 17.23 0.677
P3HT
50
-co-3HOT
50
5.609 15.74 0.672
P3HOT -- -- --
P3HTT-DPP 5.602 15.76 0.728
P3HTT-HOT-DPP(75:5) 5.601 15.77 0.732
P3HTT-HOT-DPP(70:10) 5.599 15.77 0.817
P3HTT-HOT-DPP(65:15) 5.526 15.98 0.759
P3HTT-HOT-DPP(60:20) 5.517 16.00 0.776
P3HTT-HOT-DPP(40:40) 5.517 16.01 0.694

Figure A1.20. Representative J-V curves of semi-random
copolymers:PC
61
BM BHJ solar cells under AM 1.5G
illumination (100 mW/cm
2
) at optimized conditions for solar
cell performance.

235
Table A1.3. Summary of 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 BHJ solar cells based on semi-random copolymers.
Polymer:PC
61
BM (1:1)
J
sc
(mA/cm
2
)
M
J
sc,corr

(mA/cm
2
)
J
sc,EQE
(mA/cm
2
)
J
sc
% error
a

P3HTT-DPP 11.10 0.87 12.78 13.07 2.2
P3HOT-HOT-DPP(75:5) 10.84 0.86 12.64 12.25 3.2
P3HOT-HOT-DPP(70:10) 9.69 0.88 11.02 11.13 1.0
P3HOT-HOT-DPP(65:15) 9.40 0.86 10.92 10.49 4.1
P3HOT-HOT-DPP(60:20) 7.54 0.87 8.65 8.99 3.8
P3HOT-HOT-DPP(40:40) 3.84 0.79 4.89 4.98 1.8
a
J
sc
% error was calculated using the equation A1:
(A1)

Table A1.4. SCLC hole mobilities of semi-random copolymers
measured in blends with PC
61
BM at their optimized conditions.
Polymer SCLC hole mobilities (cm
2
/Vs)

P3HTT-DPP 2.5×10
-2

P3HOT-HOT-DPP(75:5) 1.6×10
-2

P3HOT-HOT-DPP(70:10) 8.9×10
-3

P3HOT-HOT-DPP(65:15) 6.0×10
-3

P3HOT-HOT-DPP(60:20) 8.0×10
-3

P3HOT-HOT-DPP(40:40) 4.3×10
-3

A1.6 References
(1) Burkhart, B.; Khlyabich, P. P.; Cakir Canak, T.; LaJoie, T. W.; Thompson, B. C.
Macromolecules 2011, 44, 1242–1246.
(2) Khlyabich, P. P.; Burkhart, B.; Ng, C. F.; Thompson, B. C. Macromolecules 2011, 44,
5079–5084.
(3) Keegstra, M. A.; Peters, T. H. A.; Brandsma, L. Tetrahedron 1992, 48, 3633–3652.
(4) Guo, X.; Watson, M. D. Org. Lett. 2008, 10, 5333–5336.
236
(5) Shrotriya, V.; Li, G.; Yao, Y.; Moriarty, T.; Emery, K.; Yang, Y. Adv. Funct. Mater.
2006, 16, 2016–2023.

237
APPENDIX 2. Random Terpolymers Based on
3-Hexyloxythiophene, 2,1,3-Benzothiadiazole and 2,7-Carbazole
A2.1 Materials and Methods
For general materials and methods, please refer to Appendix 1. All reagents from
commercial sources were used as received unless otherwise noted. Trimethylsilyl chloride
was distilled over CaH
2
before use. Solvents were purchased from VWR and used without
further purification except for THF, which was dried over sodium/benzophenone before
being distilled. All reactions were performed under dry N
2
in glassware that was pre-dried in
oven, unless otherwise noted.
1
H NMR spectra of small molecules were recorded on a Varian Mercury 400 NMR
Spectrometer.
1
H NMR spectra of polymers were obtained on a Varian 600 NMR
Spectrometer at 50 °C. All NMR spectra were recorded in CDCl
3
.
Cyclic voltammetry (CV) was performed on a 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), and a Pt wire counter electrode was purged with N
2

and maintained under N
2
atmosphere during all measurements. Polymer films were made by
dipping the Pt wire tip into a CHCl
3
solution of polymer (10 mg/mL) and
tetrabutylammonium hexafluorophosphate (TBAPF
6
) (30 mg/mL) and drying under N
2
prior

238
to measurement. Acetonitrile was distilled over CaH
2
prior to use as the solvent, and TBAPF
6

(0.1 M) was used as the supporting electrolyte.
UV−Vis absorption spectra were obtained on a Perkin-Elmer Lambda 950
spectrophotometer. For thin film measurements, polymers or blends were spin-coated onto
pre-cleaned glass slides from 7 mg/mL CHCl
3
solutions. 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.
A2.2 Synthetic Procedures and Structural Characterizations
Compound 1 2-bromo-3-hexyloxythiophene shown in Figure 3.2 was synthesized
according to the procedure described in Appendix 1. Compound 3
4,7-dibromo-2,1,3-benzothiadiazole shown in Figure 3.2 were synthesized following
published procedure.
1
Compound 7 9-(heptadecan-9-yl)-2,7-bis(4,4,5,5-tetramethyl-1,3-
dioxolan-2-yl)-9H-carbazole and Compound 8 4,7-bis(5-bromothiophen-2-yl)-2,1,3-
benzothiadiazole shown in Figure 3.4 as well as polymer PCDTBT shown in Figure 3.1,
were obtained from Seyma Ekiz. Synthetic procedures for the other small molecules shown
in Figure 3.2 and polymers shown in Figure 3.4 are described as follows.
2-trimethylstannyl-3-hexyloxythiophene (2). A solution of 0.872 g
2-bromo-3-hexyloxythiophene (1) (3.32 mmol) in 8 mL THF was cooled to -78 °C under N
2

atmosphere, before the addition of 2.5 mL 1.6 M n-BuLi solution in hexanes (4 mmol, 1.20
eq) with stirring. The mixture was kept at -78 °C for 3 h before a 1.0 M solution of

239
trimethyltin chloride in hexane (5.0 mL, 5.0 mmol, 1.51 eq) was added slowly. The mixture
was allowed to warm and kept at room temperature overnight. The reaction was quenched by
adding 10 mL of water to the flask, followed by pouring into another 150 mL of water. After
extraction with diethyl ether, the organic layer was washed with water three times and then
dried over MgSO
4
. The product with approximately 3% of the isomeric impurity
(5-trimethyltin-3-hexyloxythiophene) was achieved by vacuum distillation as a colorless oil
(0.80 g, 2.31 mmol, 70% yield).
1
H NMR (400 MHz, CDCl
3
): δ 7.46 (d, 1H), 6.94 (d, 1.06H),
3.94 (t, 2.41H), 1.73 (m, 2.47H), 1.44 (m, 2.59H), 1.32 (m, 4.62H), 0.90 (m, 3.58H), 0.33 (s,
10.63H).
4,7-bis(3-(hexyloxy)thiophen-2-yl)-2,1,3-benzothiadiazole (4). Starting materials 0.112
g 4,7-dibromo-2,1,3-benzothiadiazole (3) (0.38 mmol) and 0.300 g Compound 2 (0.86 mmol,
2.28 eq) were dissolved in 15 mL dry DMF under N
2
atmosphere. After degasing by purging
N
2
into the reaction mixture for 15 min, 13 mg Pd(PPh
3
)
4
(0.011 mmol, 3 mol%) was added
quickly. The mixture was degased for another 10 min, and then immersed into pre-heated oil
bath at 90 °C and kept stirring overnight. After that, the mixture was poured into 100 mL of
water and extracted with DCM until the aqueous layer became colorless. The DCM solutions
were combined and concentrated under reduced pressure. The dark red residue was subject to
silica gel column chromatography with gradient hexanes/DCM (v/v 7:3 to 1:1) to afford pure
product from the first separated band as a dark red solid. (0.132 g, 0.26 mmol product, 68%
yield).
1
H NMR (400 MHz, CDCl
3
): δ 8.46 (s, 1H), 7.39 (d, 0.89H), 7.00 (d, 1.01H), 4.17 (t,
2.13H), 1.87-1.83 (m, 2.08H), 1.53 (m, 2.54H), 1.35 (m, 3.99H), 0.90 (m, 2.80H).

240
4,7-bis(5-bromo-3-(hexyloxy)thiophen-2-yl)-2,1,3-benzothiadiazole (5). NBS (94 mg,
0.53 mmol, 2.0 eq) was added in one portion to Compound 4 (0.132 g, 0.26 mmol) in 5 mL
DMF at 0 °C. The mixture was stirred at 0 °C for another hour and then warmed to room
temperature and stirred overnight. After that, it was then poured into water followed by
extraction with DCM until the aqueous layer became colorless. The DCM solution was
concentrated under reduced pressure, and the dark red residue was subject to silica gel
column chromatography with gradient hexanes/DCM (v/v 7:3 to 1:1) to afford pure product
from the first separated band as a dark red solid. (0.115 g, 0.17 mmol product, 67% yield).
1
H
NMR (400 MHz, CDCl
3
): δ 8.43 (s, 1H), 6.98 (s, 1.03H), 4.14 (t, 2.35H), 1.84 (m, 2.25H),
1.50 (m, 2.77H), 1.36 (m, 3.89H), 0.91 (m, 3.07H).
General procedures for Suzuki Polymerization. In a flame-dried three-neck flask
equipped with a condenser and under N
2
atmosphere, desired amounts of Compound 5, 7 and
8, as well as K
2
CO
3
(17 eq relative to Compound 7), were dissolved in a mixture of THF and
water (v/v 4:1) to afford a 0.03 M solution. To this mixture, 2 mol% of Pd(PPh
4
)
3
(relative to
Compound 7) was added quickly. The reaction flask was then immersed into an oil bath that
was preheated to 85 °C. Polymerization was allowed to occur over 3 d under reflux,
gradually producing dark purple precipitates in the reaction mixture. After that, the mixture
was cooled and precipitated in methanol. The crude polymer was purified by Sohxlet
extraction successively with methanol, hexanes and CHCl
3
. The CHCl
3
fraction was
concentrated under reduced pressure, reprecipitated in methanol, vacuum filtered and then
dried overnight under high vacuum.

241
PCDTBT
3
DHOTBT
1
. Yield 50%. M
n
= 11.0 kg/mol, PDI = 2.40.
1
H NMR (600 MHz,
CDCl
3
): δ 8.57 (s, 1H), 8.21-8.10 (m, 5.56H), 7.96-7.33 (m, 14.43H), 4.70 (s, 1.80H), 4.32 (s,
2.00H), 2.43 (m, 3.78H), 2.08-1.95 (m, 6.05H), 1.61 (s, 2.53H), 1.47-1.19 (m, 62.58H), 0.96
(s, 4.19H), 0.81 (s, 12.74H).
PCDTBT
1
DHOTBT
1
. Yield 66%. M
n
= 12.8 kg/mol, PDI = 3.70.
1
H NMR (600 MHz,
CDCl
3
): δ 8.57 (s, 1H), 8.21-8.10 (m, 2.88H), 7.96-7.33 (m, 7.10H), 4.72 (s, 0.96H), 4.31 (s,
1.94H), 2.43 (s, 1.86H), 2.07-1.94 (m, 3.81H), 1.61 (s, 1.90H), 1.47-1.19 (m, 30.51H), 0.96
(s, 3.27H), 0.80 (s, 6.19H).
PCDTBT
1
DHOTBT
3
. Yield 73%. M
n
= 17.7 kg/mol, PDI = 2.33.
1
H NMR (600 MHz,
CDCl
3
): δ 8.57 (s, 1H), 8.22-8.10 (m, 1.70H), 7.96-7.33 (m, 4.49H), 4.72 (s, 0.66H), 4.32 (s,
2.02H), 2.43 (s, 1.32H), 2.07 (s, 1.30H), 1.94 (s, 2.04H), 1.59 (s, 2.04H), 1.47-1.19 (m,
22.49H), 0.96 (s, 3.13H), 0.80 (m, 4.54H).
PCDHOTBT. Yield 64%. M
n
= 18.8 kg/mol, PDI = 2.92.
1
H NMR (600 MHz, CDCl
3
): δ
8.57 (s, 1H), 8.10-7.33 (m, 4.25H), 4.72 (s, 0.50H), 4.31 (s, 2.02H), 2.43 (s, 0.96H),
2.07-1.94 (m, 2.96H), 1.59 (s, 1.97H), 1.47-1.19 (m, 18.17H), 0.96 (s, 3.29H), 0.80 (s,
3.25H).
The isomeric comonomer Compound 6 was synthesized according to the procedures shown
in Figure A2.1, which is described as follows.

242

Figure A2.1. Synthetic scheme to prepare the isomeric comonomer Compound 6.
2-trimethylsilyl-3-hexyloxythiophene (8). A solution of 1.402 g
2-bromo-3-hexyloxythiophene (1) (5.33 mmol) in 12 mL THF was cooled to -78 °C under N
2

atmosphere, before the addition of 4.0 mL 1.6 M n-butyl lithium solution in hexanes (6.40
mmol, 1.20 eq) with stirring. The mixture was kept at -78 °C for 3 h before 0.870 g distilled
trimethylsilyl chloride (8.0 mmol, 1.50 eq) was added slowly. The mixture was allowed to
warm and kept at room temperature overnight. The reaction was poured into water and then
extracted with diethyl ether. The organic layer was washed with water three times and then
dried over MgSO
4
. Over 95% pure product was achieved by two successive vacuum
distillations as a colorless oil (0.73 g, 2.85 mmol, 54% yield).
1
H NMR (400 MHz, CDCl
3
): δ
7.40 (d, 1H), 6.90 (d, 1.08H), 3.96 (t, 2.45H), 1.75-1.71 (m, 2.35H), 1.46 (m, 2.33H), 1.34 (m,
6.28H), 0.90 (m, 4.67H), 0.29 (s, 9.63H).
2-trimethylsilyl-5-trimethylstannyl-3-hexyloxythiophene (9). Two three-neck flasks
were flame-dried right before use. In flask 1, 0.73 g Compound 8 (2.85 mmol, 1 eq) was
dissolved in 8 mL dry THF. In flask 2, 0.407 g diisopropylamine (4.00 mmol, 1.4 eq) was
dissolved in 5 mL dry THF, and the solution was cooled to -78 °C. At this temperature, 1.9
mL 1.6 M n-butyl lithium solution in hexanes (3.00 mmol, 1.05 eq) was added dropwise. The

243
mixture was kept stirring at -78 °C for another 1 h, after which it was transferred into
previously cooled flask 1 (-78 °C) via cannula. The whole mixture was stirrd for another 3 h
at -78 °C, followed by the dropwise addition of 4.3 mL 1.0 M trimethyltin chloride in
hexanes (4.3 mmol, 1.5 eq). The reaction was allowed to occur at room temperature
overnight. After that, it was then quenched by addition of water and product was extracted
with diethyl ether twice. The ether solution was washed with brine and water and dried with
MgSO
4
, before ether was evaporated under reduced pressure. Pure product was obtained by
vacuum filtration as a colorless oil (0.85 g, 2.03 mmol product, 71%).
1
H NMR (400 MHz,
CDCl
3
): δ 6.96 (s, 1H), 3.99 (t, 2.13H), 1.72 (m, 2.13H), 1.46 (m, 2.19H), 1.33 (m, 4.18H),
0.91 (m, 3.09H), 0.35 (s, 8.10H), 0.28 (s, 9.90H).
4,7-bis(4-(hexyloxy)-5-(trimethylsilyl)thiophen-2-yl)-2,1,3-benzothiadiazole(10). Starti
ng materials 0.280 g 4,7-dibromo-2,1,3-benzothiadiazole (3) (0.95 mmol) and 0.850 g
Compound 2 (2.02 mmol, 2.14 eq) were dissolved in 15 mL dry DMF under N
2
atmosphere.
After degasing by purging N
2
into the reaction mixture for 15 min, 22 mg Pd(PPh
3
)
4
(0.019
mmol, 3 mol%) was added quickly. The mixture was degased for another 10 min, and then
immersed into pre-heated oil bath at 90 °C and kept stirring overnight. After that, the mixture
was poured into 100 mL of water and extracted with DCM until the aqueous layer became
colorless. The DCM solutions were combined and concentrated under reduced pressure before it was poured into cold methanol to form red precipitates. The red solid was collected
by vacuum filtration and then dried under high vacuum.
1
H NMR spectrum confirmed the
solid being a mixture of 90% product, 10% 4,7-bis(4-(hexyloxy)thiophen-2-yl)-2,1,3-
benzothiadiazole (Compound 11) and a small percentage of impurity. The mixture was used
in the following step without further purification. (0.504 g product, 0.78 mmol, 82% yield).

244
1
H NMR (400 MHz, CDCl
3
): δ 7.97 (s, 1H), 7.81 (s, 1.25H), 6.37 (s, 0.09H), 4.10 (t, 2.04H),
4.02 (t, 0.22H), 1.80 (m, 2.34H), 1.49 (m, 2.38H), 1.36 (m, 4.63H), 0.91 (m, 4.51H), 0.35 (s,
8.74H).
4,7-bis(4-(hexyloxy)thiophen-2-yl)-2,1,3-benzothiadiazole (11). A solution of 0.480 g
Compound 10 (a mixture from the above-mentioned synthesis, 0.504 mmol Compound 10
mixed with 0.050 mmol Compound 11) in 5 mL THF was cooled to 0 °C. To this mixture,
3.0 mL tetrabutylammonium fluoride (TBAF) solution (1.0 M in THF, 3.0 mmol, 6 eq) was
added dropwise. The reaction mixture was then kept stirring at room temperature overnight.
After that, the mixture was poured into H
2
O and extracted with DCM until the aqueous layer
became colorless. The DCM solution was concentrated under reduced pressure before being
poured into cold methanol. The orange precipitates formed in methanol was collected by
vacuum filtration and then dried under high vacuum. The resulting orange crystals were
determined to be about 95% pure product. (0.280 g, 0.56 mmol, 100%).
1
H NMR (400 MHz,
CDCl
3
): δ 7.81 (s, 2.35H), 7.73 (s, 0.09H), 6.37 (s, 1H), 4.18 (t, 0.10H), 4.00 (t, 2.42H), 1.81
(m, 2.53H), 1.49 (m, 8.63H), 1.35 (m, 4.89H), 0.90 (m, 3.49H).
4,7-bis(5-bromo-4-(hexyloxy)thiophen-2-yl)-2,1,3-benzothiadiazole (6). NBS (44.5 mg,
1.12 mmol, 2.0 eq) was added in one portion to Compound 10 (0.28 g, 0.56 mmol) in 10 mL
DMF at 0 °C. The mixture was stirred at 0 °C for another hour and then warmed to room
temperature and stirred overnight. After that, it was then poured into water followed by
extraction with DCM until the aqueous layer became colorless. The DCM solution was
concentrated under reduced pressure, and then poured into cold methanol. Recrystallization
from a solvent mixture of DCM and ethanol afforded over 90% pure product as a dark red

245
solid. (0.313 g, 0.47 mmol, 85% yield).
1
H NMR (400 MHz, CDCl
3
): δ 7.84 (s, 0.08H), 7.80
(s, 1H), 7.74 (s, 1.06H), 4.16 (t, 2.18H), 1.85 (m, 2.18H), 1.54 (m, 8.38H), 1.36 (m, 4.11H),
0.92 (m, 2.89H).
A2.3%Structure%Verifications%of%Small%Molecules%and%Polymers%

Figure A2.2.
1
H NMR of Compound 2 in CDCl
3
.

246

Figure A2.3.
1
H NMR of Compound 4 in CDCl
3
.

247

Figure A2.4.
1
H NMR of Compound 5 in CDCl
3
.

248

Figure A2.5.
1
H NMR of polymer PCDTBT
3
DHOTBT
1
in CDCl
3
.

249

Figure A2.6.
1
H NMR of polymer PCDTBT
1
DHOTBT
1
in CDCl
3
.

250

Figure A2.7.
1
H NMR of polymer PCDTBT
1
DHOTBT
3
in CDCl
3
.

251
!
Figure A2.8.
1
H NMR of polymer PCDHOTBT in CDCl
3
.

252

Figure A2.9.
1
H NMR of Compound 8 in CDCl
3
.

253

Figure A2.10.
1
H NMR of Compound 9 in CDCl
3
.

254
!
Figure A2.11.
1
H NMR of Compound 10 in CDCl
3
.

255
!
Figure A2.12.
1
H NMR of Compound 11 in CDCl
3
.

256
!
Figure A2.13.
1
H NMR of Compound 6 in CDCl
3
.
!
!
A2.4 Characterization of Polymer Samples
Table A2.1. Summary of UV-Vis Absorption Data of All Polymers.
Sample
λ
max,abs

(nm)
solution
Absorptivity
(L/cm*g)
λ
max,abs
(nm)
film
Absorption
Coefficient
*10
-5
(cm
-1
)
PCDTBT 552 54.0 566 0.705
PCDTBT
3
DHOTBT
1
558 51.8 568 0.688
PCDTBT
1
DHOTBT
1
564 50.6 572 0.671
PCDTBT
1
DHOTBT
3
574 52.1 576 0.665
PCDHOTBT 580 52.0 578 0.635

257

Figure A2.14. GIXRD of films of PCDTBT derivative
polymers spin-coated from CHCl
3
solution and
annealed at 150 °C for 30 min: (black) PCDTBT, (red)
PCDTBT
3
DHOTBT
1
, (green) PCDTBT
1
DHOTBT
1
,
(blue) PCDTBT
3
DHOTBT
1
and (light blue)
PDHOTBT.

A2.5.%References%
(1) Burkhart, B.; Khlyabich, P. P.; Cakir Canak, T.; LaJoie, T. W.; Thompson, B. C.
Macromolecules 2011, 44, 1242–1246.
%

!

258
APPENDIX 3. Morphology and Device Performance of
Polymer:Small Molecule:Fullerene Ternary Blend Solar Cells in
Comparison to Binary Blend Cells Based on Polymers with
Covalently Attached Small Molecules
A3.1 Materials and Methods
For general materials and methods, please refer to Appendix 1. All reagents from
commercial sources were used as received unless otherwise noted. CuBr was purified by
washing three times with acetic acid, twice with ethanol and twice with diethyl ether, and
then stored in a desiccator. Isopropylmagnesium chloride (iPrMgCl) solution in THF was
titrated before use.
1
Solvents were purchased from VWR and used without further
purification except for THF, which was dried over sodium/benzophenone before being
distilled. All reactions were performed under dry N
2
in glassware that was pre-dried in oven,
unless otherwise noted.
1
H NMR spectra of small molecules were recorded on a Varian Mercury 400 NMR
Spectrometer.
1
H NMR spectra of polymers were obtained on a Varian 600 NMR
Spectrometer at 50 °C. All NMR spectra were recorded in CDCl
3
.
Cyclic voltammetry (CV) was performed on a 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
+

259
which is taken as -5.1 eV vs vacuum), and a Pt wire counter electrode was purged with N
2

and maintained under N
2
atmosphere during all measurements. For solid state CV, polymer
films were made by drop-casting an o-DCB solution of polymer (10 mg/mL) and
tetrabutylammonium hexafluorophosphate (TBAPF
6
) (30 mg/mL) onto the Pt wire and
drying under N
2
prior to measurement. Acetonitrile was distilled over CaH
2
prior to use as
the solvent, and TBAPF
6
(0.1 M) was used as the supporting electrolyte. For solution CV,
polymer was dissolved in chloroform (CHCl
3
) to form a 0.01 wt% solution. CHCl
3
was
distilled over CaH
2
prior to use and tetrabutylammonium tetrafluoroborate (TBABF
4
) (0.1 M)
was used as the supporting electrolyte.
UV−Vis absorption spectra were obtained on a Perkin-Elmer Lambda 950
spectrophotometer. For thin film measurements, polymers or blends were spin-coated onto
pre-cleaned glass slides from 10 mg/mL CHCl
3
solutions. 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. AFM topology images were recorded on an Agilent 5420 instrument under
tapping mode. In order to prepare thin films of the ternary blends for AFM measurements,
the blends were spin-coated onto pre-cleaned glass slides from CHCl
3
solutions with the
concentrations of the overall donors being 7.5 mg/mL. Annealing is done by putting the films
in a petri dish full of o-DCB vapor for about 2 min.
A3.2 Synthetic Procedures and Structural Characterizations
Compound 4 2,5-di(2-ethylhexyl)-3,6-bis(thiophene-2-yl)pyrrolo[3,4-c]-pyrrole-1,4-dione
and 12 2,5-di(2-ethylhexyl)-3,6-bis(5-bromothiophene-2-yl)pyrrolo[3,4-c]-pyrrole-1,4-dione

260
shown in Figure A3.3 and A3.14 were synthesized following published procedures.
2,3

Synthetic procedures for other small molecules shown in Figure A3.1, A3.2 and A3.14 and
polymers shown in Figure 4.3 are described as follows.

Figure A3.1. Syntheses of monomers 2,5-dibromo-3-
hexylthiophene (1) and 2,5-Dibromo-3-{2-bromoethyl}
thiophene (3).
2,5-dibromo-3-hexylthiophene (1). NBS (1.62 g, 9.08 mmol, 1.01 eq) was added in one
portion to the solution of 3-hexylthiophene (1.51 g, 8.99 mmol) in a mixture of 10 mL CHCl
3

and 10 mL acetic acid. The mixture was heated up to 70 °C and stirred for 6 h. It was then
cooled to room temperature and poured into water, followed by extraction with DCM. The
organic layer was washed with water twice and then dried over MgSO
4
before being
concentrated under reduced pressure. Purification of product was done by silica gel column
chromatography with hexanes, followed by a vacuum distillation (300 mtorr, 65 °C). The
product was obtained as a colorless liquid (2.49 g, 7.64 mmol, 85 %).
1
H NMR (400 MHz,
CDCl
3
): δ 6.78 (s, 1H), 2.50 (t, 2.38H), 1.52 (m, 3.33H), 1.30 (m, 6.85H), 0.89 (m, 3.25H).
2,5-Dibromo-3-{2-hydroxyethyl}thiophene (2). Adopted from literature.
4
To a solution
of 3-thiopheneethanol (0.565 g, 4.42 mmol) in 9 mL DCM, 1.422g TBAB (4.42 mmol, 1.0
eq) and 0.705g MeOH (2.21 mmol, 0.5 eq) were added. After stirring at room temperature
for 10 min, about 0.5 mL Br
2
(1.55 g, 9.72 mmol, 2.2 eq) was added slowly. The mixture was

261
stirred for 5 h at room temperature. The reaction was then quenched by addition of 2 M
NaOH (10 mL) and diluted with 150 mL DCM. The organic phase was washed successively
with saturated Na
2
S
2
O
3
, brine and water. After drying with MgSO
4
, solvent was removed
from the organic layer, and crude product 2,5-dibromo-3-(2-hydroxy)ethylthiophene (2), was
purified by silica gel column chromatography with 1:1 hexanes/DCM. After removal of
solvents under reduced pressure, the product was obtained as an yellow viscous oil (1.04 g,
3.62 mmol, 82%).
1
H NMR (CDCl
3
): δ (ppm) 6.87 (s, 1H), 3.81 (t, 2H), 2.80 (t, 2H), 1.62 (s,
1H).
2,5-Dibromo-3-{2-bromoethyl}thiophene (3). To a solution of 0.51 g
2,5-Dibromo-3-{2-hydroxyethyl}thiophene (2) (1.78 mmol) in 10 mL DCM, 0.621g CBr
4

(1.87 mmol, 1.05 eq) was added under N
2
atmosphere. The mixture was cooled to 0 °C and
stirred for 15 min. 0.565 g PPh
3
(2.16 mmol, 1.2 eq) was then added in portions. The mixture
was kept stirring at 0 °C for another 1 h before it was warmed up to room temperature to stir
overnight. The reaction was quenched by pouring into 200 mL diethyl ether under stirring.
After vacuum filtration to remove the white precipitates, the ether filtrate was concentrated
under reduced pressure. The residue was purified by silica gel column chromatography with
9:1 hexane/DCM to afford the product as a colorless oil. (0.584 g, 1.67 mmol, 94%).
1
H
NMR (400 MHz, CDCl
3
): δ 6.86 (s, 1H), 3.49 (t, 2.50H), 3.10 (t, 2.51H).

262

Figure A3.2. Syntheses of alkyne-functionalized DPP molecule (11).
3-(5-bromothiophen-2-yl)-2,5-bis(2-ethylhexyl)-6-(thiophen-2-yl)-2,5-dihydropyrrolo[
3,4-c]pyrrole-1,4-dione (5). A solution of 2,5-bis(2-ethylhexyl)-3,6-di(thiophen-2-yl)-2,5-
dihydropyrrolo[3,4-c]pyrrole-1,4-dione (4) (1.20 g, 2.29 mmol) in 80 mL of CHCl3 was
cooled to 0 °C under N
2
atmosphere. To this solution, 0.408 g NBS (2.29 mmol, 1.0 eq)
dissolved in 40 mL CHCl
3
was blocked from light and added dropwise over 2 h at 0 °C.
After the addition was finished, the reaction mixture was warmed up to room temperature
and kept stirring in dark overnight. The mixture was then concentrated under reduced
pressure before being subject to silica gel column chromatography with gradient
hexanes/DCM (7:3 to 0:1). The product was eluted out as the second band (after
dibrominated side product as the first band and before starting material as the third). After
drying, 0.773 g product was obtained as a dark red solid (1.28 mmol, 56%).
1
H NMR (400
MHz, CDCl
3
): δ 8.90 (dd, 1H), 8.63 (d, 0.99H), 7.65 (dd, 0.98H), 7.28 (dd, 0.85H), 7.23 (d,
0.93H), 4.03-3.93 (m, 4.41H), 1.85 (m, 2.21), 1.35-1.25 (m, 18.23H), 0.87 (m, 13.25H).
2,5-bis(2-ethylhexyl)-3-(5-phenylthiophen-2-yl)-6-(thiophen-2-yl)-2,5-dihydropyrrolo[
3,4-c]pyrrole-1,4-dione (6). To a solvent mixture of 30 mL toluene, 8 mL ethanol and 2 mL
H
2
O, 0.975 g 3-(5-bromothiophen-2-yl)-2,5-bis(2-ethylhexyl)-6-(thiophen-2-yl)-2,5-

263
dihydropyrrolo[3,4-c] pyrrole-1,4-dione (5) (1.62 mmol), 0.246 g phenylboronic acid (2.02
mmol, 1.25 eq) and 0.279 g K
2
CO
3
(2.02 mmol, 1.25 eq) were added and dissolved. The
mixture was degassed by purging N
2
for 20 min before the addition of 75 mg Pd(PPh
3
)
4

(0.065 mmol, 4 mol%). After degassing for another 5 min after the addition, the mixture was
heated up and kept stirring at 85 °C for 20 h, during which process the solution turned from
dark red to purple. The reaction was then cooled down to room temperature, poured into H
2
O
and extracted with DCM afterwards until the aqueous layer became colorless. The DCM
solution was concentrated under reduced pressure before being subject to silica gel column
chromatography with gradient hexanes/DCM (2:3 to 1:4). The pure product was obtained as
a dark purple solid (0.960 g, 1.60 mmol, 99%).
1
H NMR (400 MHz, CDCl
3
): δ 8.96 (d, 1H),
8.88 (d, 0.94H), 7.68 (d, 2.16H), 7.62 (d, 0.96H), 7.47 (d, 1.24H), 7.44 (m, 2.14H), 7.37 (m,
1.11H), 7.28 (m, 0.84H), 4.07 (m, 4.30H), 1.94-1.86 (m, 2.03H), 1.39-1.26 (m, 18.19H),
0.91-0.85 (m, 13.24H).
3-(5-bromothiophen-2-yl)-2,5-bis(2-ethylhexyl)-6-(5-phenylthiophen-2-yl)-2,5-dihydro
pyrrolo[3,4-c]pyrrole-1,4-dione (7). To a solution of 3-(5-bromothiophen-2-yl)-2,5-
bis(2-ethylhexyl)-6-(thiophen-2-yl)-2,5-dihydropyrrolo[3,4-c]pyrrole-1,4-dione (6) (0.954 g,
15.9 mmol) in 50 mL CHCl
3
, 0.311 g NBS (17.5 mmol, 1.1 eq) was added in one portion.
The reaction mixture was blocked from light and stirred at room temperature overnight, after
which it was concentrated and then subject to silica gel column chromatography with
gradient hexanes/DCM (1:1 to 1:4). The product was obtained as a dark purple solid (0.880 g,
13.0 mmol, 82%)
1
H NMR (400 MHz, CDCl
3
): δ 8.96 (d, 1H), 8.62 (d, 1.02H), 7.69 (d,
2.16H), 7.48-7.42 (m, 3.28H), 7.37 (m, 1.07H), 7.23 (m, 1.19H), 4.07 (m, 4.56H), 1.94-1.86
(m, 2.20H), 1.37-1.26 (m, 18.56H), 0.91-0.87 (m, 13.70H).

264
((4-bromophenyl)ethynyl)triisopropylsilane (8). In a flame-dried three-neck round
bottom flask, 1.00 g 1-bromo-4-iodobenzene (3.53 mmol), 0.771 g
(Triisopropylsilyl)acetylene (4.24 mmol, 1.2 eq) and 0.185 g PPh
3
(0.707 mmol, 0.20 eq)
were dissolved in a solvent mixture of 5.5 mL dry THF and 3 mL triethylamine that is
previous distilled over CaH
2
. The mixture was degassed by purging N
2
for 20 min before the
sequential addition of 67 mg CuI (0.353 mmol, 10 mol%) and 50 mg Pd(PPh
3
)Cl
2
(0.071
mmol, 2 mol%). After degassing for another 5 min, the mixture was heated up and kept
stirring at 70 °C overnight. The reaction was then cooled down to room temperature, poured
into H
2
O and extracted with DCM for three times. The DCM solution was concentrated
under reduced pressure before being subject to silica gel column chromatography with
hexanes. The product was eluted out together with the two starting materials. Pure product
was obtained as a colorless oil residue after removal of starting materials under high vacuum
at 70 °C over 1.5 h (1.14 g, 3.38 mmol, 96%).
1
H NMR (400 MHz, CDCl
3
): δ 7.44 (d, 2H),
7.34 (d, 2H), 1.13 (m, 23.35H).
Triisopropyl((4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)phenyl)ethynyl)silane (9).
In a flame-dried three-neck round bottom flask, 1.14 g
((4-bromophenyl)ethynyl)triisopropylsilane (8) (3.38 mmol) was dissolved in 15 mL dry
THF. The mixture was cooled down to -78 °C and then 2.4 mL n-BuLi (1.6 M in hexanes,
3.90 mmol, 1.15 eq) was added dropwise. The mixture was kept stirring at -78 °C for 3 h,
after which 1.05 mL Isopropoxyboronic acid pinacol ester (0.943 g, 5.07 mmol, 1.5 eq) was
added at -78 °C in one portion. The reaction was then allowed to proceed at room
temperature for 16 h, followed by quenching with H
2
O. The aqueous mixture was extracted
with diethyl ether for three times, and the combined ether solution was then washed with H
2
O

265
twice. After removal of diethyl ether, purification of product was done by recrystallization
from a mixture of hexanes and ethanol. Pure product was obtained as white crystals (0.596 g,
1.55 mmol, 46%).
1
H NMR (400 MHz, CDCl
3
): δ 7.44 (d, 2H), 7.45 (d, 2.03H), 1.34 (s,
12.67H), 1.13 (m, 22.12H).
2,5-bis(2-ethylhexyl)-3-(5-phenylthiophen-2-yl)-6-(5-(4-((triisopropylsilyl)ethynyl)phe
nyl)thiophen-2-yl)-2,5-dihydropyrrolo[3,4-c]pyrrole-1,4-dione (10). To a solvent mixture
of 45 mL toluene and 3 mL H
2
O, 0.850 g Compound 7 (1.25 mmol), 0.529 g Compound 9
(1.38 mmol, 1.1 eq), 0.215 g K
2
CO
3
(1.56 mmol, 1.25 eq) and 3 drops of Aliquat 336 were
added successively with stirring. The mixture was degassed by purging N
2
for 20 min before
the addition of 58 mg Pd(PPh
3
)
4
(0.050 mmol, 4 mol%). After degassing for another 5 min,
the mixture was heated up and kept stirring at 85 °C for 20 h, during which process the
solution turned from dark red to purple. The reaction was then cooled down to room
temperature, poured into H
2
O and extracted with DCM afterwards until the aqueous layer
became colorless. The DCM solution was concentrated under reduced pressure before being
subject to silica gel column chromatography with gradient hexanes/DCM (1:1 to 3:7). The
pure product was obtained as a dark purple solid (0.960 g, 1.12 mmol, 90%).
1
H NMR (400
MHz, CDCl
3
): δ 8.97-8.93 (m, 2H), 7.69 (d, 2.17H), 7.60 (d, 2.17H), 7.53 (m, 2.30H), 7.48
(d, 2.30H), 7.44 (m, 2.28H), 7.37 (m, 0.95H), 4.10 (m, 4.48H), 1.93 (m, 2.27H), 1.39-1.29 (m,
18.72H), 1.15 (m, 24.42H), 0.92-0.87 (m, 13.86H).
2,5-bis(2-ethylhexyl)-3-(5-(4-ethynylphenyl)thiophen-2-yl)-6-(5-phenylthiophen-2-yl)-2
,5-dihydropyrrolo[3,4-c]pyrrole-1,4-dione (11). A solution of 0.525 g Compound 10 (0.613
mmol) in 40 mL THF was cooled to 0 °C. To this mixture, 3.1 mL tetrabutylammonium

266
fluoride (TBAF) solution (1.0 M in THF, 3.1 mmol, 5 eq) was added dropwise. The reaction
mixture was then kept stirring at room temperature overnight. After that, the mixture was
poured into H
2
O and extracted with DCM until the aqueous layer became colorless. The
DCM solution was concentrated under reduced pressure before being subject to silica gel
column chromatography with 7:3 hexanes/DCM. The product was obtained as a dark purple
solid (0.290 g, 0.413 mmol, 67%).
1
H NMR (400 MHz, CDCl
3
): δ 8.99 (d, 1H), 8.94 (d,
1.03H), 7.68 (d, 2.04H), 7.63 (d, 2.05H), 7.55 (d, 2.05H), 7.48 (m, 2.06H), 7.44 (m, 2.08H),
7.37 (m, 1.03H), 4.08 (m, 4.06H), 3.19 (s, 0.84H), 1.94 (m, 2.06H), 1.38-1.29 (m, 16.97H),
0.92-0.87 (m, 12.54H).

Figure A3.3.
1
H NMR of Compound 11 in CDCl
3
.

267
General procedures for GRIM Polymerization. In a flame-dried three-neck flask
equipped with a condenser and under N
2
atmosphere, all monomers were dissolved in dry
THF to afford a 0.067 M solution. To this mixture, 1 eq of isopropylmagnesium chloride
solution was added dropwise. Then the mixture was gently refluxed with stirring for 2 h.
After that it was cooled down to room temperature before 0.8 mol% of Ni(dppp)Cl
2
catalyst
was added. The solution was heated up to gentle reflux again for another 6 h before the
formed polymer was precipitated in cold methanol. The crude polymer was purified by
Sohxlet extraction successively with methanol, hexanes and CHCl
3
. The CHCl
3
fraction was
concentrated under reduced pressure, reprecipitated in methanol, vacuum filtered and then
dried overnight under high vacuum.
P3HT. Yield 70%. M
n
= 42.0 kg/mol, PDI = 1.67.
1
H NMR (600 MHz, CDCl
3
): δ 6.98 (m,
1H), 2.81 (t, 1.86H), 2.58 (t, 0.09H), 1.72 (m, 1.96H), 1.46 (m, 2.67H), 1.37 (m, 4.17H), 0.92
(m, 3.03H).
P3BrT-1. Yield 68%. M
n
= 48.5 kg/mol, PDI = 1.66.
1
H NMR (600 MHz, CDCl
3
): δ 6.98
(m, 1H), 3.64 (t, 0.07H), 3.40 (t, 0.07H), 2.80 (t, 1.84H), 2.58 (t, 0.07), 1.72 (m, 1.93H), 1.44
(m, 3.14H), 1.35 (m, 4.11H), 0.91 (m, 3.03H).
P3BrT-2. Yield 67%. M
n
= 43.1 kg/mol, PDI = 1.69.
1
H NMR (600 MHz, CDCl
3
): δ 6.98
(m, 1H), 3.64 (t, 0.17H), 3.40 (t, 0.15H), 2.81 (t, 1.73H), 2.58 (m, 0.09H), 1.72 (m, 1.92H),
1.47-1.36 (m, 6.76H), 0.91 (m, 2.82H).

268
P3BrT-3. Yield 65%. M
n
= 23.2 kg/mol, PDI = 1.81.
1
H NMR (600 MHz, CDCl
3
): δ 6.98
(m, 1H), 3.64 (t, 0.27H), 3.40 (t, 0.26H), 2.80 (t, 1.65H), 2.58 (m, 0.07H), 1.72 (m, 1.73H),
1.44-1.35 (m, 6.51H), 0.91 (m, 2.79H).

Figure A3.4.
1
H NMR of polymer P3BrT-1 in CDCl
3
.

269

Figure A3.5.
1
H NMR of polymer P3BrT-2 in CDCl
3
.

270

Figure A3.6.
1
H NMR of polymer P3BrT-3 in CDCl
3
.
General procedures for azide functionalization. To a flame-dried three-neck flask under
N
2
atmosphere, P3BrT polymers were fully dissolved in o-DCB under 50 °C with stirring to
afford an 8 mg/mL solution. DMSO (1/8 of the DCB volume) was added to the mixture via
syringe very slowly so as to prevent polymer precipitation. After the addition of DMSO was
finished, NaN
3
(10 eq relative to bromine functional groups) was added quickly and the
reaction flask was then wrapped with aluminum foil to block light. The reaction mixture was
kept stirring in dark for 3 days before the polymer was precipitated in cold methanol. The
precipitated product was collected by vacuum filtration, dried under high vacuum and used in
following reactions without further purification.

271
P3N
3
T-1. Yield 82%.
1
H NMR (600 MHz, CDCl
3
): δ 6.98 (m, 1H), 3.58 (t, 0.07H), 3.13 (t,
0.09H), 2.80-2.58 (m, 1.89H), 1.70 (m, 1.92H), 1.47-1.35 (m, 6.78H), 0.91 (m, 2.94H).
P3N
3
T-2. Yield 80%.
1
H NMR (600 MHz, CDCl
3
): δ 6.99 (m, 1H), 3.59 (t, 0.16H), 3.13 (t,
0.17H), 2.82-2.58 (m, 1.87H), 1.72 (m, 1.89H), 1.47-1.37 (m, 6.69H), 0.92 (m, 2.85H).
P3N
3
T-3. Yield 80%.
1
H NMR (600 MHz, CDCl
3
): δ 6.99 (m, 1H), 3.59 (t, 0.27H), 3.13 (t,
0.27H), 2.81-2.58 (m, 1.74H), 1.72 (m, 1.75H), 1.47-1.35 (m, 5.90H), 0.91 (m, 2.62H).

Figure A3.7.
1
H NMR of polymer P3N
3
T-1 in CDCl
3
.

272

Figure A3.8.
1
H NMR of polymer P3N
3
T-2 in CDCl
3
.

273

Figure A3.9.
1
H NMR of polymer P3N
3
T-3 in CDCl
3
.
General procedures for “click” reaction. To a flame-dried three-neck flask under N
2
atmosphere, azide-functionalized polymer and alkyne-functionalized DPP (1.05 eq relative to
azide functional groups) were dissolved in DCB at 35 °C with stirring to afford an 8 mg/mL
solution. PMDETA (5 eq relative to azide functional groups) was added via syringe. The
mixture was then degased by purging N
2
for 15 min, before CuBr (5 eq relative to azide
functional groups) was added quickly. Degasing was continued for 5 min and the reaction
mixture was then kept stirring in dark at 35 °C for 3 days. The resulting polymer was
precipitated in a mixture of acetone and methanol (1:1 v/v) and then purified by Soxhlet
extraction with acetone, ethyl acetate and CHCl
3
successively. The CHCl
3
fraction was

274
reprecipitated in cold methanol, followed by vacuum filtration to collect the final polymer
product before it is finally dried under high vacuum.
P1. Yield 88%. M
n
= 47.1 kg/mol, PDI = 2.02.
1
H NMR (600 MHz, CDCl
3
): δ 8.93 (m,
0.07H), 7.86 (m, 0.06H), 7.65 (m, 0.20H), 7.44 (m, 0.21), 6.98 (m, 1H), 4.73 (m, 0.07H),
4.06 (m, 0.15H), 3.52 (m, 0.06H), 2.80-2.58 (m, 1.86H), 1.96 (m, 0.08H), 1.69 (m, 1.88H),
1.46-1.27 (m, 6.89H), 0.94-0.87 (m, 3.36H).
P2. Yield 85%. M
n
= 42.9 kg/mol, PDI = 2.07.
1
H NMR (600 MHz, CDCl
3
): δ 8.92 (m,
0.16H), 7.86 (m, 0.17H), 7.65 (m, 0.43H), 7.43 (m, 0.43H), 6.98 (m, 1H), 4.73 (m, 0.17H),
4.06 (m, 0.32H), 3.49 (m, 0.17H), 2.80-2.56 (m, 1.79H), 1.94 (m, 0.17H), 1.71 (m, 1.87H),
1.47-1.27 (m, 7.84H), 0.94-0.87 (m, 3.87H).
P3. Yield 83%. M
n
= 21.7 kg/mol, PDI = 2.31.
1
H NMR (600 MHz, CDCl
3
): δ 8.91 (m,
0.26H), 7.86 (m, 0.26H), 7.64 (m, 0.66H), 7.39 (m, 0.68H), 6.98 (m, 1H), 4.71 (m, 0.24H),
4.05 (m, 0.53H), 3.50 (m, 0.25H), 2.80-2.56 (m, 1.72H), 1.93 (m, 0.30H), 1.70 (m, 1.80H),
1.44-1.28 (m, 8.77H), 0.92-0.86 (m, 4.27H).

275

Figure A3.10.
1
H NMR of polymer P1 in CDCl
3
.

276

Figure A3.11.
1
H NMR of polymer P2 in CDCl
3
.

277

Figure A3.12.
1
H NMR of polymer P3 in CDCl
3
.

278
!
Figure A3.13.
1
H NMR of P3HT in CDCl
3
.
!
!
!
Figure A3.14. Synthesis of 2,5-di(2-ethylhexyl)-3,6-
bis(5-phenylthiophene-2-yl)pyrrolo[3,4-c]-pyrrole-1,4-dione (13).
2,5-di(2-ethylhexyl)-3,6-bis(5-phenylthiophene-2-yl)pyrrolo[3,4-c]-pyrrole-1,4-dione
(13). Same procedure as Compound 6 except that the amount of phenylboronic acid is 2.5 eq
relative to compound 12. Product is obtained as a dark purple solid (85%).
1
H NMR (400
MHz, CDCl
3
): δ 8.97 (d, 2H), 7.69 (d, 4.26H), 7.47 (d, 2.15H), 7.44 (m, 4.48H), 7.37 (m,
2.30H), 4.09 (m, 4.43H), 1.95 (m, 2.23H), 1.41-1.28 (m, 19.39H), 0.91-0.86 (m, 14.15H).

279
!
Figure A3.15.
1
H NMR of Compound 13 in CDCl
3
.
A3.3 Device Fabrication and Characterization
For general device fabrication and characterization methods, please refer to Appendix 1. In
order to process P3HT:PhDPP:PCBM ternary blends into device active layers, separate
solutions of the P3HT, PhDPP, and PC
61
BM were prepared in CHCl
3
. The solutions were
stirred for 6 h before they were mixed at the desired ratios and stirred overnight to form a
homogeneous solution. Subsequently, the P3HT:PhDPP:PC
61
BM active layer was
spin-coated (filtered with a 0.45 µm PVDF syringe filter from Pall Life Sciences) on top of
the PEDOT:PSS layer. The concentrations of the overall donors were 7.5 mg/mL. For
annealed devices, films were put in a petri dish full of o-DCB vapor for about 2 min before

280
placed in the vacuum chamber for aluminum deposition. For the preparation of solar cell
devices based on polymers P1-P3, the polymer:PCBM active layer was spin-coated from
o-DCB solution with the polymer concentration being 10 mg/mL, followed by exposure to
N
2
atmosphere for 20 min.
A3.4 Characterization of Polymer Samples and Blends
!
Figure A3.16. UV-Vis Absorption of P3HT:PhDPP blend
films as well as pristine P3HT and PhDPP films
spin-coated from CHCl
3
and annealed at 150 °C for 30
min. (i) P3HT, (ii) P3HT80:PhDPP20, (iii)
P3HT60:PhDPP40, (iv) P3HT40:PhDPP60, (v)
P3HT20:PhDPP80 and (vi) PhDPP.
!
!

281

Figure A3.17. UV-Vis Absorption of P1-P3 and
P3HT solutions in o-DCB. (i) P3HT, (ii) P1, (iii)
P2 and (iv) P3.
!
Figure A3.18. UV-Vis Absorption of P1-P3 and P3HT
films spin-coated from CHCl
3
and annealed at 150 °C
for 30 min. (i) P3HT, (ii) P1, (iii) P2 and (iv) P3.
! !

282
Table A3.1. GIXRD data summary of P3HT:PhDPP:PC
61
BM ternary blend films before and
after annealing.
Sample 2θ (°) d (Å) FWHM
As Cast
a
P3HT80:PhDPP20:PC
61
BM 5.25 16.82 0.800
P3HT60:PhDPP40:PC
61
BM 5.60 15.77 1.105
P3HT40:PhDPP60:PC
61
BM N/A N/A N/A
P3HT20:PhDPP80:PC
61
BM N/A N/A N/A
Annealed
b
P3HT 5.20 16.98 0.834
P3HT80:PhDPP20:PC
61
BM 5.50/6.30 16.06/14.02 0.722/0.503
P3HT60:PhDPP40:PC
61
BM 5.55/6.30 15.90/14.02 0.738/0.638
P3HT40:PhDPP60:PC
61
BM 6.30 14.02 0.468
P3HT20:PhDPP80:PC
61
BM 6.35 13.91 0.523
PhDPP 6.30 14.02 0.687
a
Thin films of blends were spin-coated from CHCl
3
solutions.
b
Thin films of blends were
spin-coated from CHCl
3
solutions, followed by o-DCB solvent vapor annealing for 2 min.

A3.5.%References%
(1) Org. Synth. 1968, 48, 47.
(2) Burkhart, B.; Khlyabich, P. P.; Cakir Canak, T.; LaJoie, T. W.; Thompson, B. C.
Macromolecules 2011, 44, 1242–1246.
(3) Khlyabich, P. P.; Burkhart, B.; Ng, C. F.; Thompson, B. C. Macromolecules 2011, 44,
5079–5084.
(4) Strover, L.; Roux, C.; Malmström, J.; Pei, Y.; Williams, D. E.; Travas-Sejdic, J. Synth.
Met. 2012, 162, 381–390.
!

!

283
APPENDIX 4. Stereoregular Non-conjugated polymers with
Diketopyrrolopyrrole Electroactive pendant groups
A4.1 Materials and Methods
For general materials and methods, please refer to Appendix 1. For method of Cyclic
voltammetry (CV), please refer to Appendix 3. All reagents from commercial sources were
used as received unless otherwise noted. N-Bromosuccinimide (NBS) was recrystallized
from hot water. CuBr was purified by washing three times with acetic acid, twice with
ethanol and twice with diethyl ether, and then stored in a desiccator. N-isopropylacrylamide
(NIPAM) was recrystallized from hexanes. 2,2-Azobisisobutyronitrile (AIBN) was
recrystallized twice from methanol. Solvents were purchased from VWR and used without
further purification unless otherwise noted. Dry THF was stirred over sodium/benzophenone
before being distilled. Other dry solvents were obtained by stirring over CaH
2
overnight
before being distilled. All reactions were performed under dry N
2
in glassware that was
pre-dried in oven, unless otherwise noted.
1
H NMR spectra of small molecules were recorded on a Varian Mercury 400 NMR
Spectrometer in CDCl
3
.
1
H NMR spectra of polymer samples were obtained on a Varian
VNMRS-600 instrument, at 50 °C in CDCl
3
for DPP-pendant polymers and at 125 °C in
DMSO-d
6
for other polymers. DSC profiles were recorded on Perkin-Elmer DSC 8000 from

284
the second heating cycle with a scan rate of 10 °C/min. Sample size was around 5 mg, and
polymers were used as obtained after purification.
A4.2 Synthetic Procedures
The synthetic procedure for 3-(5-bromothiophen-2-yl)-2,5-bis(2-
ethylhexyl)-6-(thiophen-2-yl)-2,5-dihydropyrrolo[3,4-c]pyrrole-1,4-dione (6) is described in
Appendix 3. The syntheses of other compounds are described as follows.

Figure A4.1. Synthesis of N-(3-(trimethylsilyl)prop-2-yn-1-yl)acrylamide (1).
N-propargylacrylamide (1a). In a three-neck flask under N
2
atmosphere, the solution of
3.84 mL propargyl amine (3.3 g, 0.06 mol) and 9.74 mL dry triethylamine (7.07 g, 0.07 mol,
1.17 eq) were dissolved in 30 mL dry DCM and was cooled to 0 °C. A solution of 5.69 mL
acryloyl chloride (6.34 g, 0.07 mol, 1.17 eq) in 15 mL dry DCM was added dropwise to the
mixture via syringe, which gradually resulted in a lot of crystals suspended in a bright yellow
solution. After stirring at 0 °C for another 30 min, the reaction mixture was then warmed up
to room temperature and stirred overnight. After that, the mixture was diluted with DCM and
washed with brine and water. The DCM solution was then dried with MgSO
4
, condensed
under reduced pressure and then subject to silica gel column chromatography with
hexanes/ethyl acetate (1:2 v/v) to afford 4.30 g product as white crystals (0.046 mol, 76%).

285
1
H NMR (400 MHz, CDCl
3
): δ 6.34 (dd, 1H), 6.11 (dd, 0.97H), 5.81-5.70 (s+dd, 1.89H),
4.14 (dd, 1.99H), 2.25 (t, 0.86H).
N-(3-(trimethylsilyl)prop-2-yn-1-yl)acrylamide (1). Into a dry three-neck flask under N
2

atmosphere, 0.716 g AgCl (5 mmol, 0.25 eq) was added and suspended in 80 mL dry DCM.
With stirring, 2.18 g N-propargyl acrylamide (0.02 mol, 1 eq) and 7.5 mL
1,8-Diazabicycloundec-7-ene (DBU) (7.6 g, 0.05 mol, 2.5 eq) were added. The mixture was
stirred at room temperature for 15 min, during which process white cloudiness gradually
formed. Distilled trimethylsilyl chloride (13.0 mL, 11.3 g, 0.1 mol, 5 eq) was then added
dropwise. The mixture cleared up after the addition of trimethylsilyl chloride and was then
warmed up to 45 °C to stir overnight. After that, it was cooled to room temperature, diluted
in DCM and washed with saturated NaHCO
3
solutions. The DCM solution containing the
product was dried with MgSO
4
and condensed under reduced pressure. Purification by silica
gel column chromatography with hexanes/ethyl acetate (2:1 v/v) afforded 1.45 g pure
products as white crystals (7.84 mmol, 39%).
1
H NMR (400 MHz, CDCl
3
): δ 6.32 (dd, 1H),
6.12 (dd, 0.96H), 5.90 (s+dd, 2.01H), 4.16 (d, 2.11H), 0.16 (s, 8.93H).

286

Figure A4.2. Syntheses of DPP-N
3
.
2-(2-bromoethoxy)tetrahydro-2H-pyran (2). A solution of 2.4 mL 3,4-dihydro-2H-pyran
(2.21 g, 26.3 mmol, 1.25 eq) in DCM was cooled to 0 °C under N
2
atmosphere, and then 40
mg p-toluenesulfonic acid monohydrate (p-TsOH) (0.21 mmol, 0.01 eq) was added. With
stirring, 1.5 mL 2-bromoethanol (2.64 g, 21.1mmol, 1 eq) was added dropwise. The mixture
was kept at 0 °C for another 1 h after addition, and was then warmed up to room temperature
to stir overnight. After that, about 8 g NaHCO
3
was added to the resulting greenish black
solution to quench the reaction, which turned the color of mixture into brown. The excess salt
was filtered out after stirring for 50 min, followed by evaporation of DCM. Pure product was
collected by vacuum distillation as a colorless liquid (3.30 g, 15.7 mmol, 75%).
1
H NMR
(400 MHz, CDCl
3
): δ 4.67 (t, 1H), 4.02 (m, 1.07H), 3.89 (m, 1.17H), 3.80-3.74 (m, 1.07H),

287
3.52 (m, 3.23H), 1.85-1.83 (m, 1.20H), 1.74 (m, 1.19H), 1.64-1.59 (m, 5.12H), 1.54 (m,
0.79H).
2-(2-([2,2'-bithiophen]-5-yl)ethoxy)tetrahydro-2H-pyran (3). Two three-neck flasks
were flame-dried right before use. In flask 1, 2.09 g Compound 2 (10 mmol, 1 eq) was
dissolved in 25 mL dry THF. In flask 2, 2.0 g bithophene (12 mmol, 1.2 eq) was dissolved in
25 mL dry THF, and the solution was cooled to -40 °C. At this temperature, 7.5 mL 1.6 M
n-butyl lithium solution in hexanes (12 mmol, 1.2 eq) was added dropwise. The mixture was
kept stirring at -40 °C for another 1.5 h, after which the previously cooled flask 1 content
(-40 °C) was transferred into flask 2 via cannula. The whole mixture was allowed to warm up
to room temperature and kept stirring overnight. The reaction was then quenched by addition
of water and product was extracted with diethyl ether twice. The ether solution was washed
with brine and water and dried with MgSO
4
, before ether was evaporated under reduced
pressure. Purification by silica gel column chromatography with petroleum ether/ethyl
acetate (19:1 v/v) afforded 1.34 g product as a green oil with small percentage of unidentified
impurity (4.56 mmol, 46%).
1
H NMR (400 MHz, CDCl
3
): δ 7.17 (dd, 1H), 7.11 (dd, 1.05H),
6.99 (m, 2.22H), 6.76 (d, 1.18H), 4.65 (t, 1.24H), 3.99-3.97 (m, 1.33H), 3.83 (m, 1.35H),
3.65-3.63 (m, 1.35H), 3.51-3.49 (m, 1.37H), 3.09 (t, 2.66H), 1.87-1.85 (m, 1.38H), 1.73 (m,
1.41H), 1.66-1.57 (m, 2.15H), 1.53 (m, 2.89H).
2-([2,2'-bithiophen]-5-yl)ethanol (4). To the solution of 1.33 g Compound 3 (25.2 mmol)
in THF, a mixture of acetic acid and water (4:1 v/v) was added. The mixture was warmed up

288
to 45 °C and kept stirring overnight. It was then poured into water and extracted with DCM.
After washing the DCM solution seven times with water and drying with MgSO
4
, the solvent
was evaporated under reduced pressure. The green oil residue was used in next step without
further purifications (0.93 g, 44.3 mmol, 98%).
1
H NMR (400 MHz, CDCl
3
): δ 7.19 (dd, 1H),
7.12 (dd, 1.03H), 7.02-6.99 (m, 2.21H), 6.78 (d, 1.08H), 3.88 (t, 2.38H), 3.05 (m, 2.34H).
5-(2-bromoethyl)-2,2'-bithiophene (5). To a solution of 0.93 g Compound 4 (4.43 mmol)
in 10 mL DCM under N
2
atmosphere, 1.54g CBr
4
(4.65 mmol, 1.05 eq) was added. The
mixture was cooled to 0 °C and stirred for 15 min, before the addition of 1.39 g PPh
3
(5.31
mmol, 1.2 eq) in portions. The mixture was kept stirring at 0 °C for another 1 h and then was
warmed up to room temperature to stir overnight. The reaction was quenched by pouring into
200 mL diethyl ether with stirring. After vacuum filtration to remove the white precipitates,
the ether filtrate was concentrated under reduced pressure. The residue was purified by silica
gel column chromatography with 8:2 hexane/DCM (v/v) to afford 0.91 g product as a light
yellow oil. (3.33 mmol, 77%).
1
H NMR (400 MHz, CDCl
3
): δ 7.20 (d, 1H), 7.13 (d, 1.02H),
7.02-6.99 (m, 2.11H), 6.79 (d, 1.06H), 3.58 (t, 2.44H), 3.34 (t, 2.46H).
5-(2-azidoethyl)-2,2'-bithiophene (6). To the solution of 0.91 g Compound 5 (3.33 mmol)
in 10 mL DMF, 1.3 g sodium azide (20 mmol, 6 eq) was added in one portion. The mixture
was wrapped to block light and heated to 60 °C to stir overnight. After that, the mixture was
poured into water and then extracted twice with ethyl acetate. The organic solution was dried
with MgSO
4
and concentrated under reduced pressure. Pure product was achieved after silica

289
gel column chromatography with gradient hexanes/DCM (1:0 to 3:2 v/v) as a colorless oil
(0.722 g, 3.07 mmol, 92%).
1
H NMR (400 MHz, CDCl
3
): δ 7.20 (dd, 1H), 7.13 (dd, 1.03H),
7.02-6.99 (m, 2.11H), 6.79 (dt, 1.07H), 3.56 (t, 2.40H), 3.07 (td, 2.42H).
(5'-(2-azidoethyl)-[2,2'-bithiophen]-5-yl)trimethylstannane (7). Two three-neck flasks
were flame-dried right before use. In flask 1, 0.50 g Compound 6 (2.13 mmol, 1 eq) was
dissolved in 6 mL dry THF. In flask 2, 0.35 mL diisopropylamine (2.80 mmol, 1.3 eq) was
dissolved in 2.5 mL dry THF, and the solution was cooled to -78 °C. At this temperature, 1.5
mL 1.5 M n-butyl lithium solution in hexanes (2.25 mmol, 1.05 eq) was added dropwise. The
mixture was kept stirring at -78 °C for another 1 h, after which it was transferred into
previously cooled flask 1 (-78 °C) via cannula. The whole mixture was stirrd for another 1.5
h at -78 °C, followed by the dropwise addition of 2.8 mL 1.0 M trimethyltin chloride in
hexanes. The reaction was allowed to occur at room temperature overnight. After that, it was
then quenched by addition of water and product was extracted with diethyl ether twice. The
ether solution was washed with brine and water and dried with MgSO
4
, before ether was
evaporated under reduced pressure. The brown oil residue was determined by
1
H NMR to be
a mixture of about 75 mol% product and 25 mol% starting material, and it was used in next
step without further purifications (0.73 g, 1.53 mmol product, 55%).
1
H NMR (400 MHz,
CDCl
3
): δ 7.23 (d, 1H), 7.20 (dd, 0.35H), 7.13 (dd, 0.42H), 7.08 (d, 1.02H), 7.02-7.00 (m,
1.75H), 6.79-6.77 (m, 1.39H), 3.57-3.54 (m, 2.91H), 3.49 (m, 0.49H), 3.08-3.05 (m, 3.03H),
1.21 (m, 0.61H), 0.39 (s, 7.98H).

290
DPP-N
3
. Starting materials 0.70 g 3-(5-bromothiophen-2-yl)-2,5-
bis(2-ethylhexyl)-6-(thiophen-2-yl)-2,5-dihydropyrrolo[3,4-c]pyrrole-1,4-dione (8) (1.16
mmol) and 0.602 mg impure Compound 7 (1.51 mmol Compound 7, 1.3 eq) were dissolved
in 100 mL dry toluene under N
2
atmosphere. After degasing by purging N
2
into the reaction
mixture for 15 min, 67 mg Pd(PPh
3
)
4
(0.058 mmol, 0.05 eq) was added quickly. The mixture
was degased for another 10 min, and then immersed into pre-heated oil bath at 85 °C. The
reaction was allowed to occur at 85 °C over 3 d. After that, toluene was evaporated and pure
product was achieved by silica gel column chromatography with 1:9 hexanes/chloroform (v/v)
as a dark purple solid (0.49 g, 0.65 mmol, 56%).
1
H NMR (400 MHz, CDCl
3
): δ 8.94 (d, 1H),
8.89 (d, 1H), 7.62 (d, 0.97H), 7.30 (d, 1.03H), 7.28 (d, 0.88H), 7.22 (d, 1H), 7.07 (m, 1.97H),
6.82 (d, 0.98H), 4.04 (m, 4.41H), 3.58 (t, 2.13H), 3.08 (t, 2.13H), 1.92-1.87 (m, 2.25H),
1.38-1.26 (m, 18.78H), 0.91-0.86 (m, 13.65H).
13
C NMR (100 MHz, CDCl
3
): δ 161.77,
161.59, 142.47, 140.38, 139.89, 139.84, 138.27, 136.83, 135.58, 135.14, 134.53, 130.40,
129.90, 128.41, 127.99, 126.75, 125.84, 124.59, 124.34, 124.10, 108.15, 52.21, 45.92, 39.26,
39.08, 30.37, 30.23, 29.95, 28.56, 28.36, 23.55, 23.06, 14.02, 10.50.
General procedures for free radical polymerizations of acrylamides. To an oven-dried
three-neck flask under N
2
atmosphere, monomers of desired amount (N-isopropylacrylamide
(NIPAM) or Compound 1 N-propargylacrylamide) were dissolved in anhydrous methanol
with stirring. In the case of isotactic polymerizations, Lewis Acid Y(OTf)
3
was dissolved at
the same time. After purging N
2
into the solution for 20 min, 2,2-Azobisisobutyronitrile

291
(AIBN) was added quickly. The reaction flask was wrapped to block light and immersed into
pre-heated oil bath at 60 °C. The polymerization was allowed to occur in 20 h, followed by
precipitation of polymers in hot water. The polymer precipitation was then collected for
tacticity measurements in the case of PNIPAM. As for alkyne functionalized
polyacrylamides PA-TMS-atactic and PA-TMS-isotactic, the water suspension was filtered
through thimble and washed with a lot of water. After drying in desiccator, the polymer
solids were subject to Soxhlet extractions with hexanes and chloroform. Polymer samples
PA-TMS-atactic and PA-TMS-isotactic were obtained by evaporating all solvents from the
chloroform solution and drying under high vacuum overnight.
PNIPAM-atactic. Yield 55%.
1
H NMR (600 MHz, DMSO-d
6
): δ 6.64 (m, 1H), 3.88 (s,
1.01H), 2.05 (m, 1.02H), 1.72 (s, 0.46H), 1.51 (m, 1.09H), 1.33 (m, 0.49H), 1.12 (s, 6.15H).
PNIPAM-isotactic. Yield 67%.
1
H NMR (600 MHz, DMSO-d
6
): δ 6.71-6.65 (m, 1H),
3.87 (s, 0.98H), 2.09-1.99 (m, 1.03H), 1.71 (s, 0.78H), 1.52 (m, 0.32H), 1.37-1.30 (m, 0.77H),
1.10 (s, 6.02H).
PA-TMS-atactic. Yield 37%. M
n
= 10.6 kg/mol, PDI = 1.75.
1
H NMR (600 MHz,
DMSO-d
6
): δ 7.25 (m, 0.92H), 3.98 (s, 2H), 2.20-2.08 (m, 0.93H), 1.72 (s, 0.50H), 1.58-1.51
(m, 0.99H), 1.42-1.29 (m, 0.48H), 0.18 (s, 9.56H).

292
PA-TMS-isotactic. Yield 35%. M
n
= 13.1 kg/mol, PDI = 1.81.
1
H NMR (600 MHz,
DMSO-d
6
): δ 7.31-7.22 (m, 1H), 3.98 (s, 2.13H), 2.15-2.03 (m, 1.07H), 1.73 (s, 0.82H),
1.56-1.52 (m, 0.47H), 1.42-1.29 (m, 0.82H), 0.18 (s, 9.69H).
General procedures for deprotection of alkyne functionalized polyacrylamides
PA-TMS-atactic and PA-TMS-isotactic. Polymer sample PA-atactic or PA-isotactic was
dissolved in dry methanol in a three-neck flask at 40 °C under N
2
to afford a 60 mg/mL
solution. After purging N
2
into the solution for 15 min, 6 eq of Na
2
CO
3
was added in one
portion. The reaction mixture was then kept stirring at 40 °C overnight. After that, the
mixture was passed through a short silica plug with methanol eluent in order to remove
excess Na
2
CO
3
. The polymer solution in methanol was then condensed and precipitated in
hexanes. Product polymers were collected as yellow precipitates after filtration and drying
under high vacuum overnight.
PA-atactic. Yield 100%.
1
H NMR (600 MHz, DMSO-d
6
): δ 7.94 (m, 1H), 3.86 (s, 2.09H),
3.05 (s, 0.94H), 1.97 (m, 1.17H), 1.48-1.24 (m, 2.30H), 0.07 (s, 0.12H).
PA-isotactic. Yield 100%.
1
H NMR (600 MHz, DMSO-d
6
): δ 8.09-7.95 (m, 1H), 3.82 (s,
2.25H), 3.01 (m, 0.91H), 2.01 (m, 1.05H), 1.23 (m, 2.43H).
Preparation of PS-alkyne. In a dry three-neck flask equipped with a condenser, 200 mg
poly(4-hydroxystyrene) (1.67 mmol in repeating unit), 345 mg K
2
CO
3
(2.50 mmol, 1.5 eq)
and 17.9 mg 18-crown-6 (0.068 mmol, 0.04 eq) were dissolved in 3 mL THF under N
2

293
atmosphere. The mixture was then refluxed with stirring for 1 h, before 193 µL propargyl
bromide (2.17 mmol, 1.3 eq) was added quickly. The reaction was allowed to occur under
reflux overnight. After that, the mixture was cooled down to room temperature and
precipitated in methanol. The solids were collected in thimble and purified by Soxhlet
extractions with methanol, hexanes and chloroform. The chloroform solution was then
condensed and reprecipitated in methanol. Solids collected from vacuum filtration were dried
under high vacuum overnight to afford 200 mg products as yellowish solids (1.25 mmol,
75%). M
n
= 23.5 kg/mol, PDI = 2.13.
1
H NMR (600 MHz, DMSO-d
6
): δ 6.73-6.58 (m, 4H),
4.74 (s, 2.01H), 3.51 (s, 0.92H), 1.72-1.41 (m, 2.92H).
General procedures for “click” reactions. To a dry three-neck flask under N
2
atmosphere,
alkyne-functionalized polymer and DPP-N3 (1.1 eq relative to alkyne functional groups)
were dissolved desired solvents at a certain temperature with stirring (THF at room
temperature for PS-alkyne, and a mixture of THF/DMF (1:1 v/v) at 60 °C for
polyacrylamides). PMDETA (5 eq relative to alkyne functional groups) was added via
syringe. The mixture was then degased by purging N
2
for 15 min, before CuBr (2.5 eq
relative to alkyne functional groups) was added quickly. Degasing was continued for 5 min
and the reaction mixture was then kept stirring in dark at the previously stated temperatures
for 2 days. The resulting polymer was precipitated in methanol and then purified by Soxhlet
extraction with methanol, hexanes and CHCl
3
successively. The CHCl
3
fraction was

294
reprecipitated in cold methanol, followed by vacuum filtration to collect the final polymer
product before it is finally dried under high vacuum.
PS-DPP. Yield 75%. M
n
= 43.9 kg/mol, PDI = 2.84.
1
H NMR (600 MHz, CDCl
3
): δ 8.84
(m, 2H), 7.91-7.80 (m, 1.11H), 7.50 (s, 1.07H), 7.14-7.02 (m, 3.06H), 6.87 (s, 2.15H), 6.59
(m, 4.49H), 5.05 (s, 1.70H), 4.55 (s, 1.90H), 3.91 (s, 4.05H), 3.31 (s, 1.86H), 1.82 (s, 2.46H),
1.31-1.21 (m, 18.67H), 0.82 (s, 13.31H).
PA-DPP-atactic. Yield 72%. M
n
= 10.0 kg/mol, PDI = 2.15.
1
H NMR (600 MHz, CDCl
3
):
δ 8.80 (s, 2H), 7.48 (s, 2.05H), 7.14-6.58 (m, 5.70H), 4.51 (s, 3.35H), 3.91 (s, 4.15H), 3.28 (s,
2.26H), 1.82 (s, 2.75H), 1.30-1.21 (m, 17.29H), 0.82 (s, 12.76H).
PA-DPP-isotactic. Yield 74%. M
n
= 16.6 kg/mol, PDI = 2.12.
1
H NMR (600 MHz, CDCl
3
):
δ 8.79 (s, 2H), 7.47 (s, 2.56H), 7.14-6.58 (m, 5.49H), 4.51 (s, 2.98H), 3.89 (s, 3.80H), 3.28 (s,
1.82H), 1.81 (s, 2.33H), 1.27-1.20 (m, 17.43H), 0.83 (s, 13.11H).

295
A4.3 Structure Verifications of Small Molecules and Polymers

Figure A4.3.
1
H NMR spectrum of Compound 1a in CDCl
3
.

296

Figure A4.4.
1
H NMR spectrum of Compound 1 in CDCl
3
.

297

Figure A4.5.
1
H NMR spectrum of Compound 2 in CDCl
3
.

298

Figure A4.6.
1
H NMR spectrum of Compound 3 in CDCl
3
.

299

Figure A4.7.
1
H NMR spectrum of Compound 4 in CDCl
3
.

300

Figure A4.8.
1
H NMR spectrum of Compound 5 in CDCl
3
.

301

Figure A4.9.
1
H NMR spectrum of Compound 6 in CDCl
3
.

302

Figure A4.10.
1
H NMR spectrum of Compound 7 in CDCl
3
.

303

Figure A4.11.
1
H NMR spectrum of DPP-N
3
in CDCl
3
.

304

Figure A4.12.
13
H NMR spectrum of DPP-N
3
in CDCl
3
.

305

Figure A4.13.
1
H NMR spectrum of PS-alkyne in DMSO-d
6
.

306

Figure A4.14.
1
H NMR spectrum of PA-atactic in DMSO-d
6
.

307

Figure A4.15.
1
H NMR spectrum of PA-isotactic in DMSO-d
6
.

308

Figure A4.16.
1
H NMR spectrum of PS-DPP in CHCl
3
.

309

Figure A4.17.
1
H NMR spectrum of PA-DPP-atactic in CDCl
3
.

310

Figure A4.18.
1
H NMR spectrum of PA-DPP-isotactic in CDCl
3
.
A4.4 Device Fabrication and Mobility Measurements
For general device fabrication and mobility measurements, please refer to Appendix 1. For
SCLC hole mobility measurements, the pristine polymer was spin-coated from 10 mg/mL
chlorobenzene (CB) solutions. Thermal annealing at 110 °C for 20 min under N
2
atmosphere
was done for annealed devices.

311
A4.5 Characterization of Polymer Samples and DPP-N
3

Figure A4.19. AFM topology images of thin films of (a) PS-DPP, (b)
PA-DPP-atactic, (c) PA-DPP-isotactic and (d) DPP-N
3
spin-coated from 10
mg/mL CDCl
3
solutions followed by thermal annealing at 110 °C for 20 min.

312

Figure A4.20. Second heating in differential scanning
calorimetry of DPP-N
3
(black), PS-DPP (red),
PA-DPP-atactic (green) and PA-DPP-isotactic (blue). Dash
line indicates the temperature to anneal thin films for UV-Vis
absorption and hole mobility measurements.

Exotherm

Abstract (if available)

Abstract Essential to the achievement of higher efficiencies in polymer:fullerene bulk heterojunction solar cells is the development of a polymer donor material that simultaneously meets the requirements of a broad absorption profile, suitable energy levels and a favorable morphology with fullerene derivatives. In an effort to pursue the ideal polymer, copolymerizing electron‐rich and electron‐poor monomers in a perfectly alternating manner (alternating donor‐acceptor (D-A) copolymers) has produced state‐of‐the‐art polymer samples through the “monomer approach”. As an alternative strategy, polymer properties can be modulated by the strategic combination of monomers through the “polymer approach”. Following this concept, in particular, a number of random copolymers have been reported in literature, revealing their distinct advantages in the ease of synthesis, as well as in the convenience and broad scope of property tuning ability. ❧ This thesis has been inspired by the previous works on electroactive polymers for organic photovoltaics that are synthesized through the “polymer approach”, especially those with random conjugated backbones. Focus in this thesis is on the synthesis of electroactive polymers following the “polymer approach” and their structure‐property relationships. Chapter 1 reviews the reported examples of random conjugated polymers in the context of organic solar cell applications, with an emphasis on their synthetic strategies, structure‐property relationships and photovoltaic device performance. Several specific strategies to produce random copolymers are recognized and discussed according to comonomer structures and reaction features. In particular, three popular strategies reviewed in Chapter 1, namely the random regioregular polythiophenes, random D‐A copolymers and semi‐random D‐A copolymers, will be reflected in the polymer structures reported in Chapter 2, 3 and 4. In addition, a brief summary is included in Chapter 1 on the advantages of random conjugated copolymers, especially in comparison to alternating D‐A copolymers. ❧ Chapter 2 describes the investigation of two sets of copolymers containing varying percentages of strongly electron‐donating 3-hexyloxythiophene units (3HOT). The polymers synthesized here include both random regioregular polythiophenes and semi‐random diketopyrrolopyrrole (DPP) ‐containing copolymers. The influence of 3HOT content on UV−Vis absorption, HOMO energy levels, polymer crystallinity and polymer:fullerene solar cell performance, especially the open‐circuit voltage (Voc), is discussed in detail. Importantly, this study demonstrates that significant changes in polymer electronic properties can be induced with only small percentage of 3HOT comonomers in random and semi‐random conjugated polymers. ❧ Chapter 3 continues the investigation on the effect of 3HOT units on polymer properties, but with another parent polymer based on alternating 2,7-carbazole and dithienyl-2,1,3-benzothiadiazole. A 2,1,3-benzothiadiazole‐based comonomer containing 3HOT units was used to replace varied amounts of dithienyl-2,1,3-benzothiadiazole and incorporated into polymer backbones following the random D‐A copolymer strategy. The relationships between polymer composition and properties are described. ❧ Chapter 4 describes the preparation of a series of random DPP‐functionalized regioregular poly(3-hexylthiophene)s (rr-P3HTs) and their application as donor materials in fullerene‐based solar cells. Functionalized rr-P3HTs with varied amounts of azide side groups were synthesized, and post‐polymerization “click” reactions were used to attach DPP derivatives onto the rr-P3HT backbones. Structure‐property‐performance relationships of these polymers are analyzed, particularly in terms of UV‐Vis absorption, HOMO energy levels, polymer crystallinity and solar cell Voc. Due to the separated conjugation between P3HT backbones and DPP side chains, these polymers are considered as two donor components that are physically constrained in one phase. Comparison is made between the binary blend solar cells based on these DPP‐functionalized rr-P3HTs and the ternary blend devices based on separate P3HT and DPP derivative donors. Dramatic differences in their device performance shed light on the working principles of ternary blend solar cells. ❧ Chapter 5 describes the use of post‐polymerization “click” reactions in preparing non‐conjugated polymers with electroactive pendent groups. The synthetic strategy toward functionalization of stereoregular polymers with electroactive pendents is explored. This work demonstrates another platform of the “polymer approach” in achieving distinct structural features of electroactive polymers.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Development of a family of semi-random multichromophoric polymers for application in organic solar cells

PDF

Molecular and polymeric donors for bulk heterojunction organic solar cells

PDF

Improving the sustainability of conjugated polymer synthesis via direct arylation polymerization

PDF

Structure-function studies of random and semi-random poly(3-hexyl thiophene) copolymers

PDF

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

PDF

Improving the efficiency and stability of organic solar cells through ternary strategies

PDF

Direct arylation polymerization (DARP) for the synthesis of conjugated polymers

PDF

Synthesis and characterization of 3-hexylesterthiophene based random and semi-random polymers and their use in ternary blend solar cells

PDF

Direct C−H arylation for the synthesis of conjugated polymers

PDF

From perfectly alternating to random multichromophoric conjugated polymers for ternary blend bulk heterojunction solar cells

PDF

Non-traditional architectures for semiconducting polymers

PDF

Non-conjugated pendant electroactive polymers as potential materials for opto-electronic applications

PDF

The influence of CdSe surface ligands on hybrid inorganic/organic solar cell performance

PDF

Improving the efficiency and sustainability of direct arylation polymerization for conjugated polymer synthesis

PDF

Blends of polystyrene sulfonic acid copolymers and polyvinylidene fluoride as polyelectrolyte membranes

Asset Metadata

Creator Xu, Bing (author)

Core Title Influence of polymer structure on electronic properties and performance as donor materials in bulk heterojunction solar cells

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Chemistry

Publication Date 03/12/2015

Defense Date 03/05/2015

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag electro‐active polymers,OAI-PMH Harvest,polymer bulk heterojunction solar cells,random conjugated polymers,structure‐property relationship

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Thompson, Barry C. (committee chair), Gupta, Malancha (committee member), Prakash, G. K. Surya (committee member)

Creator Email bingx@usc.edu,bxbingxu@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-541051

Unique identifier UC11298498

Identifier etd-XuBing-3231.pdf (filename),usctheses-c3-541051 (legacy record id)

Legacy Identifier etd-XuBing-3231.pdf

Dmrecord 541051

Document Type Dissertation

Format application/pdf (imt)

Rights Xu, Bing

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