University of Southern California Dissertations and Theses

Nanocrystal surface engineering as a route towards improved photovoltaics

(USC Thesis Other)

Nanocrystal surface engineering as a route towards improved photovoltaics

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content NANOCRYSTAL SURFACE ENGINEERING AS A ROUTE TOWARDS
IMPROVED PHOTOVOLTAICS

by
Matthew J. Greaney
__________________________________________________________________
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)

August 2015

Copyright 2015 Matthew J. Greaney

ii

Acknowledgements
I will begin by acknowledging my advisor, Professor Richard L. Brutchey for his
guidance, support, and friendship, all of which have helped get me to where I am today.
His high standards and expectations along with our common scientific interests brought
me to USC and kept me motivated and engaged throughout the entirety of my graduate
work. Richard’s genuine concern for the well-being and the success of his students is an
admirable trait for which I will personally be forever thankful to have experienced.
In addition to Richard, I must extend my gratitude to my other committee
members, Professors Stephen Bradforth, Aiichiro Nakano, Mark Thompson, and Brent
Melot. Their support and thought-provoking questions certainly helped me along the
way with this academic endeavor, and for that I will always be appreciative. My
colleagues in the Brutchey Group, both past and present, are also due thanks for their
tutelage, camaraderie, and time spent drinking fine beers and libations together. Dr. Dan
Ruddy and Dr. Jane Frommer are two people who have long been inspirations to me, and
they are both due acknowledgement for their contributions to my academic career.
I would also like to acknowledge my mother, father, brother for their love and
support throughout my academic career. My mother, Judi Greaney, is an amazing
woman who has shown me what it means to overcome adversity and always follow
through on a commitment. Finally, I extend my deepest appreciation and love to my
girlfriend Anna, for standing next to me throughout this long journey and helping me to
keep things in perspective along the way.

iii

Table of Contents

Acknowledgements ii

List of Tables vi

List of Figures

List of Schemes
vii

xiv

Abstract xv

Chapter 1. Ligand Engineering in Hybrid Polymer:Nanocrystal Solar Cells
1.1. Abstract
1.2. Introduction
1.3. Semiconductor Nanocrystals as Electron Acceptors
1.4. Surface Trap States and Nanocrystal Coupling
1.5. Surface Ligands
1.6. Ligand Engineering for Device Optimization
1.7. State-of-the-Art Hybrid Bulk Heterojunction Solar Cells
1.8. Future Directions
1.9. References

1
1
1
2
4
7
10
18
20
22

Chapter 2. Improving Open Circuit Potential in Hybrid P3HT:CdSe Bulk
Heterojunction Solar Cells via Colloidal tert-Butylthiol Ligand
Exchange
2.1. Abstract
2.2. Introduction
2.3. Results and Discussion
2.3.1. Colloidal Ligand Exchange
2.3.2. Electrochemical Characterization
2.3.3. Hybrid Solar Cells
2.4. Experimental
2.4.1. General Considerations
2.4.2. Synthesis of Native-Ligand CdSe Nanocrystals
2.4.3. Nanocrystal Ligand Exchange
2.4.4. Characterization
2.4.5. Device Fabrication and Testing
2.5. Conclusions
2.6. References

29
29
30
33
33
38
42
53
53
54
55
56
58
59
60

iv

Chapter 3. Novel Semi-Random and Alternating Copolymer Hybrid Solar
Cells Using CdSe Multipods as Versatile Acceptors
3.1. Abstract
3.2. Introduction
3.2. Results and Discussion
3.2.1. Nanocrystal Synthesis and Characterization
3.2.3. Device Fabrication and Optimization
3.4. Experimental
3.4.1. CdSe Multipod Synthesis and Ligand Exchange
3.4.2. Materials Characterization
3.4.3. Donor Polymers
3.4.4. Device Fabrication and Characterization
3.5. Conclusions
3.6. References

65
65
65
67
67
69
75
75
75
77
78
79
80

Chapter 4. Direct Spectroscopic Evidence of Ultrafast Electron Transfer
from a Low Band Gap Polymer to CdSe Quantum Dots in
Hybrid Photovoltaic Thin Films
4.1. Abstract
4.2. Introduction
4.3. Results and Discussion
4.3.1. Steady-State Characterization of the Hybrid Blends
4.3.2. Transient Absorption Spectra of Hybrid Components
4.3.3. Transient Absorption Spectra of Hybrid Films
4.3.4. Accessible QD Electron Energy Levels by Injection
from the PCPDTBT S
1
State
4.3.5. Evaluation of the Electronic Coupling Between
PCPDTBT and CdSe(tBT) QDs
4.3.6. Prompt Yield for Electron Transfer and Model for the
Electron Population Time Evolution
4.3.7. Comparison of Yields Obtained from the Transient
Absorption Measurements to Solar Cell Performance
4.4. Experimental
4.4.1. Sample Preparation
4.4.2. Steady-State Spectroscopies
4.4.3. Femtosecond Transient Absorption Spectroscopy
4.4.4. Preparation of Hybrid Thin Films
4.4.5. Steady-State Optical Characterization of Hybrid Thin
Films Containing Various QD Quantities
4.4.6. Photovoltaic Device Testing
4.4.7. Ultrafast Transient Absorption Spectroscopy
4.5. Conclusions

83
83
84
86
86
91
97

99

101

103

111
112
112
113
113
114

115
116
116
125

v

4.6. References 127

Chapter 5. Ligand- and Size-Dependent Ultrafast Electron Transfer at
Hybrid Polymer:Nanocrystal Interfaces
5.1. Introduction
5.2. Results and Discussion
5.3. Experimental
5.3.1. CdSe Synthesis and Ligand Exchange
5.3.2. Materials Characterization
5.3.3. Transient Absorption
5.4. Conclusions
5.5. References

Chapter 6. Controlling the Trap State Landscape of Colloidal CdSe
Nanocrystals with Cadmium Halide Ligands
6.1. Abstract
6.2. Introduction
6.3. Results and Discussion
6.3.1. Nanocrystal Synthesis and Characterization
6.3.2. Ligand Exchange and Characterization
6.3.3. Thermal Treatment
6.3.4. Photocurrent Measurements
6.3.5. Surface Photovoltage Spectroscopy: Probing Sub-Gap
Energy States
6.3.6. Time-Resolved Photoluminescence and Transient
Absorption: Excited State Dynamics
6.4. Experimental
6.4.1. General Considerations
6.4.2. Synthesis and Ligand Exchange of CdSe(native ligand)
Nanocrystals
6.4.3. Surface Photovoltage Spectroscopy
6.4.4. Femtosecond Transient Absorption Spectroscopy
6.5. Conclusions
6.6. References

Bibliography

134
134
136
150
150
150
152
155
156

159
159
160
164
164
165
175
181

184

189
201
201

201
204
205
206
207

214

vi

List of Tables

Table 1.1: A summary of high efficiency hybrid polymer:nanocrystal BHJ
solar cells.

19

Table 2.1: Optical band gaps and electrochemically determined
HOMO/LUMO levels for P3HT and CdSe nanocrystals.

39

Table 2.2: Short-circuit current (J
SC
), open circuit voltage (V
OC
), fill factor
(FF), and power conversion efficiency (η
P
) for P3HT:CdSe
devices under simulated AM 1.5G light illumination.

42

Table 2.3: Fitting parameters for the PL lifetime measurements of hybrid
films.

49

Table 2.4: Fitting parameters for the PL lifetime measurements of neat
films.

Table 3.1: Photovoltaic device parameters for the polymer:CdSe(tBT)
multipod BHJ hybrid solar cells.

Table 4.1: Decay dynamics of the excited species identified in
PCPDTBT:CdSe(tBT) hybrid films.

Table 5.1: Comparison between the various quantities observed in the text:
external quantum efficiencies of various devices measured at 800
nm, fraction of short ligands in the ligand corona, LUMO level
of the QD acceptor, and amplitude of the spectral signature for
charge transfer.

Table 6.1: Nanocrystal composition as determined by TEM-EDX and ICP-
OES.

53

73

94

141

173

vii

List of Figures

Figure 1.1: Calculated power conversion efficiencies of a hybrid
polymer:CdSe nanocrystal solar cell as a function of donor
polymer band gap and CdSe nanocrystal acceptor diameter.

3

Figure 1.2: Transient short-circuit photocurrent of a,b) P3HT:PCBM BHJ
device and c,d) P3HT:CdSe BHJ device in response to a 200 µs,
525 nm pulse at varying intensities.

5

Figure 1.3: Schematic of stoichiometry dependent surface coordination
chemistry of CdSe nanocrystals.

8

Figure 1.4: Qualitative molecular orbital diagrams illustrating the bonding
interactions between undercoordinated Cd
2+
surface cations on
CdSe nanocrystals and s-donating ligands.

9

Figure 1.5: TGA and FT-IR of CdSe Nanocrystals before and after ligand
exchange.

12

Figure 1.6: Spectroelectrochemistry and cyclic voltammagrams for
CdSe(tBT) and CdSe(Py).

14

Figure 1.7: Structures of the polymers used in the hybrid polymer:CdSe
multipod BHJ solar cells and the corresponding I–V curves
under dark AM1.5G 1 sun illumination.

15

Figure 1.8: Transient absorption spectra of a neat CdSe quantum dot film, a
neat PCPDTBT film, and a hybrid PCPDTBT:CdSe BHJ film
upon excitation at 800 nm.

17

Figure 2.1: UV-vis absorption and PL emission spectra for a dispersion of
CdSe(NL) in toluene.

34

Figure 2.2: TEM image and corresponding particle size histogram for a
typical batch of CdSe(NL) nanocrystals.

35

Figure 2.3: TGA traces of CdSe(NL), CdSe(Py), and CdSe(tBT)
nanocrystals.

36

viii

Figure 2.4: FT-IR spectra of CdSe(NL), CdSe(Py), and CdSe(tBT)
nanocrystals.

37

Figure 2.5: Cyclic voltammagrams of CdSe(NL), CdSe(Py), and CdSe(tBT)
nanocrystals on ITO.

39

Figure 2.6: Differential pulse voltammagrams of CdSe(NL), CdSe(Py),
and CdSe(tBT) nanocrystals on ITO.

40

Figure 2.7: Spectroelectrochemistry: visible absorption spectra of
CdSe(NL) (A), CdSe(Py) (B) and CdSe(tBT) (C) spun-cast
onto ITO collected with decreasing applied bias.

41

Figure 2.8: I-V curves for devices with ITO/PEDOT:PSS/P3HT:CdSe/Al
structure, made with CdSe(NL), CdSe(Py), or CdSe(tBT)
nanocrystal acceptors.

42

Figure 2.9: Dark I-V characteristics for P3HT:CdSe(NL), P3HT:CdSe(Py),
and P3HT:CdSe(tBT) hybrid solar cells.

44

Figure 2.10: CV of P3HT spun-cast onto ITO.

44
Figure 2.11: EQE and absorption spectra of neat and hybrid P3HT:CdSe
solar cells and thin films.

46

Figure 2.12: Steady-state PL spectra (λ
ex
= 550 nm) for films of neat P3HT,
P3HT:CdSe(NL), P3HT:CdSe(Py), and P3HT:CdSe(tBT).

47

Figure 2.13: PL decay measurements with the corresponding fitted curves
(λ
ex
= 550 nm; λ
em
= 650 nm) for neat P3HT and P3HT blended
with CdSe(NL), CdSe(Py), and CdSe(tBT).

48

Figure 2.14: AFM images of hybrid BHJ films. 51

Figure 2.15: TEM images of hybrid BHJ films. 51

Figure 2.16: PL lifetime decay traces for films of neat CdSe films. 52

Figure 2.17: Normalized PL lifetime traces of the neat P3HT film and
hybrid blends.

52

ix

Figure 3.1: (a) TGA thermograms, (b) EDX spectra, (c) TEM micrograph,
and (d) FT-IR spectra of CdSe multipods.

Figure 3.2: Structure of the polymers used in the hybrid solar cells and the
corresponding hybrid polymer:CdSe(tBT) multipod thin film
absorption spectra.

Figure 3.3: Absorption spectrum for CdSe(tBT) multipod thin film.

Figure 3.4: PL spectra for neat polymers (solid lines) versus hybrid blends
with CdSe(tBT) multipods.

Figure 3.5: Device characteristics of the champion polymer:CdSe(tBT)
multipod BHJ hybrid solar cells.

Figure 4.1: Normalized absolute absorbance spectra of a CdSe(tBT) QDs
solution, a neat PCPDTBT film, and a PCPDTBT solution.

Figure 4.2: Absolute absorbance spectra of a neat PCPDTBT film and of
hybrid films containing various weight fractions of QDs.

Figure 4.3: I–V curve for a PCPDTBT:CdSe(tBT) QD hybrid solar cell in
a polymer/QD 1:8 wt/wt ratio.

Figure 4.4: Photoluminescence efficiency as a function of the QD content
in the hybrid, for an excitation wavelength of 780 nm.

Figure 4.5: Time-resolved transient absorption signal for a neat PCPDTBT
and a PCPDTBT:CdSe(tBT) QD hybrid film.

Figure 4.6: Transient absorption spectra of a neat CdSe(tBT) film, of a
neat PCPDTBT film (15−20 nm), and of a hybrid
PCPDTBT:CdSe(tBT) film (1:8 wt/wt, 45−55 nm) at different
time delays.

Figure 4.7: Comparison of the decay dynamics of the excited species in
the neat PCPDTBT film and in the hybrid film.

Figure 4.8: Transient absorption spectra of a neat PCPDTBT film (15-20
nm) and of a hybrid PCPDTBT:CdSe(tBT) film (1:8 wt/wt
polymer/QDs, 45-55 nm) at different time delays.

68

70

71

71

72

87

88

89

90

91

93

94

95

x

Figure 4.9: Transient absorption decays of the signals measured in a neat
PCPDTBT film.

Figure 4.10: Transient absorption spectra of a PCPDTBT in ODCB
solution.

Figure 4.11: Density of electrons located on the QD conduction band as a
function of time.

Figure 4.12: Time evolution of the QD electron population in the hybrid
film compared to the time evolution of the 610 nm ground
state bleach in the neat QD films.

Figure 4.13: Normalized electron population decay as a function of time for
different pump fluences.

Figure 4.14: Different solutions obtained with different values of the
Onsager radius r
c
, initial charge separation distribution and
diffusion coefficient D.

Figure 4.15: Comparison between the ground state bleach observed in
transient absorption 75 fs after the pump pulse in the neat
polymer film and in the hybrid film.

Figure 5.1: External quantum efficiencies of hybrid PCPDTBT:CdSe
devices.

Figure 5.2: Representative I-V curves of hybrid solar cells whose EQEs
are shown in Figure 5.1.

Figure 5.3: Thermogravimetric analysis of the 3.9 nm CdSe QDs with
various ligands and schematic representations of the QD
surface coverage.

Figure 5.4: Effect of QD surface ligand on ultrafast electron transfer.

Figure 5.5: Thermogravimetric analysis of the CdSe(6 nm) QDs before
and after each type of ligand exchange.

Figure 5.6. Absolute absorbance of the neat PCPDTBT and
PCDPTBT:CdSe (6 nm) films used in transient absorption.

98

100

105

106

108

110

115

137

138

139

144

146

146

xi

Figure 5.7: Transient absorption spectra for various pump-probe delays
(λ
exc
= 800 nm) and for a series of samples made with 6 nm
CdSe QDs with different ligands.

Figure 5.8: The transient absorption spectra from Figure 5.4 for tBT, Py,
and BA based hybrids were deconvoluted into a QD
component (a) and a polymer component (b).

Figure 6.1: Optical absorption spectra of aliquots (in 1-octadecene) taken
over the course of the reaction representing CdSe nanocrystal
growth.

Figure 6.2: XRD pattern for a CdSe(native ligand) nanocrystal film on
silicon.

Figure 6.3: TGA thermograms of (a) CdSe(butylamine), (b) CdSe(CdI
2
),
(c) CdSe(CdBr2), (d) CdSe(CdCl
2
), and (e) CdSe(native
ligand).

Figure 6.4:
1
H NMR spectra for (a) CdSe(native ligand), (b)
CdSe(butylamine), (c) CdSe(CdCl
2
), (d) CdSe(CdBr
2
), and (e)
CdSe(CdI
2
) after digestion in aqua regia and extraction with
d
6
-benzene.

Figure 6.5: Quantitative FT-IR spectra of unannealed (dashed lines) and
250 °C annealed (solid lines) nanocrystal samples.

Figure 6.6: Absorption spectra of as-cast CdSe nanocrystal thin films.

Figure 6.7: High-resolution XPS spectra of CdSe(CdX
2
) nanocrystal
samples.

Figure 6.8: STEM-EDX maps of CdSe(CdCl
2
) nanocrystal ensemble.

Figure 6.9: STEM-EDX maps of CdSe(CdBr
2
) nanocrystal ensemble.

Figure 6.10: STEM-EDX elemental map of an ensemble of CdSe(CdI
2
)
nanocrystals.

Figure 6.11: Absorption spectra of CdSe nanocrystal thin films after
annealing at 250 °C under nitrogen for 10 min.

147

149

164

165

167

168

169

170

171

172

172

173

176

xii

Figure 6.12: TEM images of as-cast (top) and 250 °C annealed CdSe
nanocrystals (bottom).

Figure 6.13: XRD patterns of CdSe nanocrystal thin films.

Figure 6.14: SEM top-down images of CdSe(CdCl
2
) nanocrystal films
before and after 250 °C annealing.

Figure 6.15: XPS spectrum for annealed CdSe(butylamine) showing a
reduction in the N 1s peak near 399 eV after thermal
annealing.

Figure 6.16: Photoelectrochemical responses of CdSe(native ligands) and
CdSe(CdX
2
) photoelectrodes under chopped 472 nm
illumination.

Figure 6.17: Measurement configuration for SPV and sample spectra.

Figure 6.18: SPV (solid green) and diffuse reflectance absorption (dashed)
spectra of a CdSe(native ligands) nanocrystal film.

Figure 6.19: SPV (solid colored lines) and diffuse reflectance absorption
(dashed) spectra of unannealed nanocrystal films.

Figure 6.20: SPV (solid colored lines) and absorption (dashed) spectra of
(a) CdSe(CdCl
2
) (b) CdSe(CdBr
2
), and (c) CdSe(butylamine)
nanocrystal films after annealing at 250 °C.

Figure 6.21: Steady-state and time-resolved photoluminescence spectra for
CdSe nanocrystal samples.

Figure 6.22: TA difference spectra for nanocrystal films at 77 K for
different probe times after excitation at 400 nm.

Figure 6.23: Dynamics of the bleach of the 2S
3/2
-1S
e
transition (at 540 nm)
for the nanocrystal samples at 77 K and 298 K.

Figure 6.24: Experimental 2S
3/2
-1S
e
bleach decay d
obs
for the CdSe(native
ligand) and CdSe(CdX
2
) nanocrystal films.

177

178

178

180

182

184

186

186

188

190

193

194

198

xiii

Figure 6.25: TA dynamics of the bleach of the 2S
3/2
-1S
e
transition (at 540
nm) for the annealed CdSe(CdCl
2
) nanocrystal film at 77 K
and 298 K.

Figure 6.26: Proposed energy landscape in CdX
2
induced hole trap states
for CdSe(CdX
2
) nanocrystals and electron trap landscape
dependence on thermal treatment.
199

200

xiv

List of Schemes

Scheme 2.1: Ligand exchange of as-prepared CdSe(NL) nanocrystals with
pyridine (Py) and tert-butylthiol (tBT).

Scheme 5.1: Schematic representation of the driving force for charge
transfer at the hybrid interface.

Scheme 6.1: Representative ligand exchange reaction that depicts L-, X-,
and Z-type binding modes to the CdSe nanocrystal surface.

33

141

174

xv

Abstract
Understanding how nanocrystal surface chemistry influences the transport of
photogenerated charges is of paramount importance for the development of successful
nanocrystal-based photovoltaics. As synthesized, semiconductor nanocrystals typically
possess electrically insulating long-chain organic molecules that offer excellent colloidal
stability, yet present an obstacle for efficient charge transport. With this in mind, we
have developed novel organic and inorganic surface passivation strategies that enable
quantitative displacement of the native synthesis ligands by replacement (i.e., ligand
exchange) with small organic molecules and inorganic compounds, while still
maintaining colloidal stability. This approach facilitates the production of optoelectronic
devices from nanocrystal “inks” that exhibit lower resistance and better performance than
non ligand-exchanged nanocrystals.

We have utilized nanocrystals after ligand exchange with small organic molecules to
fabricate hybrid polymer:nanocrystal solar cells, and we have observed a strong
dependence of device performance on the nature of the nanocrystal surface ligand.
Hybrid solar cells fabricated from poly(3-hexylthiophene) and CdSe nanocrystals ligand-
exchanged with tert-butylthiol (tBT) show significantly improved device performance as
a result of enhanced short current density (J
SC
) and increased open-circuit potential (V
OC
)
relative to nonligand-exchanged or pyridine exchanged nanocrystals. Through a
combination of (spectro)electrochemistry and ultrafast photoluminescence lifetime
measurements, we have shown that the enhancement in V
OC
can be attributed to ligand-
induced modulation of the conduction band edge (i.e., LUMO energy) of the CdSe
nanocrystals, and the increase in J
SC
can be explained by a reduction in the number of
nanocrystal surface traps. We have shown the dependence on nanocrystal LUMO energy
on surface ligand to be robust for a series of different organic ligands. Employing the
tBT ligand-exchange in combination with a spectrally complementary polymer enables
the fabrication of high-efficiency hybrid solar cells that demonstrate state-of-the-art
power conversion efficiencies in excess of 4%.

To gain better insight into the influence of nanocrystal surface chemistry at the hybrid
nanocrystal:polymer interface, we have used ultrafast transient absorption spectroscopy
to probe the electron transfer dynamics between the low band gap polymer, poly[2,6-
(4,4-bis(2-ethylhexyl)-4H-cyclopenta[2,1-b:3,4-b’]-dithiophene)-alt-4,7-(2,1,3-
benzothiadiazole)] (i.e., PCPDTBT), and CdSe nanocrystals. We find that ultrafast
electron transfer from PCPDTBT to CdSe(tBT) occurs in under 40 fs. This electron
transfer rate and yield is heavily dependent on nanocrystal surface chemistry, with
pyridine and butylamine ligand-exchanged CdSe exhibiting slower and lower yielding
electron transfer compared to tBT treatment.

xvi

In addition to organic surface ligands, we have explored the use of inorganic cadmium
halide complexes as ligands for CdSe nanocrystals. We developed a simple ligand
exchange procedure for replacing native ligands from the surface of CdSe nanocrystals
with cadmium halide (CdX
2
; X = Cl, Br, or I). Photoelectrochemical measurements on
CdX
2
-treated CdSe nanocrystal films reveal a strong dependence of the observed
photocurrent on the cadmium halide ligand used for ligand exchange, with CdCl
2

treatment leading to the greatest observed photocurrents, and CdI
2
the lowest
photocurrents. The dependence of photocurrent on surface ligand is thought to result
from ligand-induced changes to the nanocrystal electronic structure. We arrive at this
conclusion using a combination of ultrafast photoluminescence lifetime, transient
absorption, and surface photovoltage spectroscopies. These measurements help to
establish a trend in ligand-related sub-band gap trap states that are generated upon
cadmium halide treatment. CdCl
2
-treatment results in the most energetically shallow
traps, whereas CdI
2
-treatment leads to nanocrystal films that possess a large density of
deep trap states. These results help rationalize the observed photocurrents, and they
suggest that the trap state landscape in nanocrystal films can be controlled through
careful control over nanocrystal surface chemistry.

1
Chapter 1. Ligand Engineering in Hybrid Polymer:Nanocrystal Solar Cells*
*Published in Materials Today 2015, 18, 31–38.

1.1. Abstract
Blends of semiconducting polymers and inorganic semiconductor nanocrystals are
receiving renewed interest as a type of inexpensive, solution-processed third generation
solar cell. In these hybrid bulk heterojunctions (BHJs), the interface between the
disparate organic and inorganic phases is a dominating factor in the overall performance
of the resulting devices. Paramount to this interface is the ligand landscape on the
nanocrystal surface, which as a result of the inherently large surface area to volume ratio
of the nanocrystals, has a significant spatial and electronic influence on the boundary
between the donor polymer and acceptor nanocrystal. We have investigated the
importance of this three-part polymer/ligand/nanocrystal interface by studying the ligand
effects in hybrid BHJ solar cells. In this article, we highlight the major research advances
and the state-of-the-art in hybrid BHJ solar cells with respect to ligand engineering, as
well as outline future research avenues deemed necessary for continued technological
advancement.

1.2. Introduction
Organic photovoltaics (OPVs) are an important component in the growing portfolio of
third generation solar cell technologies because of their potential to be manufactured at a
disruptively low cost through solution processing (e.g., rapid roll-to-roll printing), in

2
addition to having the added benefit of utilizing a very thin absorber layer which
translates to very low material cost. One of the most commonly employed OPV device
geometries is a bulk heterojunction (BHJ), which consists of an active layer of a
conjugated polymer donor blended with a fullerene acceptor. Such polymer:fullerene
BHJs have achieved power conversion efficiencies (PCEs) of in excess of 9%;
1,2

however, optimization of the donor component may be nearing its limit to further
increase PCE,
3
thereby highlighting the importance of exploring new acceptor types.

1.3. Semiconductor Nanocrystals as Acceptors
In 2002, Alivisatos and coworkers reported the first example of a hybrid BHJ solar cell
by blending a poly(3-hexylthiophene) (P3HT) donor polymer with CdSe quantum dots,
which acted as an organic acceptor phase in place of the fullerenes.
4
In such a device
architecture, semiconductor nanocrystals (such as CdSe quantum dots) possess several
attributes that should make them attractive substitutes for the more well-established
fullerene acceptors; namely, (i) tunable band gaps and frontier energy levels through both
compositional control and quantum confinement effects, (ii) strong, broad absorption at
energies higher than the band edge, (iii) high dielectric constants to help overcome the
strong exciton binding energy of organic materials and (iv) high electron mobilities.
5,6

Furthermore, the ability to tune the dimensionality of semiconductor nanocrystals can be
used to improve device performance through morphological control, with anisotropic
nanorod and branched multipod-based acceptors facilitating directional electron transport
along their principal rod axis and thereby reducing the number of electron hopping events

3
between nanocrystals required for charge collection.
7,8
While champion PCEs for hybrid
BHJ solar cells have recently reached 5–6%,
9
their device performance still lags behind
that of their all-organic counterparts utilizing fullerene acceptors. Moreover, champion
PCEs of these hybrid BHJ solar cells do not yet meet the performance levels expected
from predictions based on the band gaps and frontier energy levels of the polymer and
nanocrystal components using the treatment of Scharber et al.
10
Using empirically
determined band gaps and exciton binding energies for donor polymers and CdSe
acceptor nanocrystals,
11
along with very conservative estimates for external quantum
efficiency (EQE; i.e., 40% EQE for the polymer component and 30% EQE for CdSe
nanocrystal component) and fill factor (FF = 0.50) based on current state-of-the-art
device performance, we believe that hybrid polymer:CdSe BHJ devices should be able to
achieve PCEs in excess of 10% (Fig. 1.1).

Figure 1.1. Calculated power conversion efficiencies of a hybrid polymer:CdSe nanocrystal
solar cell as a function of donor polymer band gap and CdSe nanocrystal acceptor diameter. CdSe
nanocrystal energy levels and exciton binding energies taken from ref. 11.

4
This is not a theoretical efficiency ceiling for these devices, but rather the efficiency that
hybrid BHJ solar cells should be capable of achieving using currently available materials.
Therefore, research into understanding the origin of this performance gap in hybrid BHJ
solar cells is very much needed; that is, a better understanding of the factors that govern
charge generation and separation at the interface between the two disparate phases in
hybrid solar cells. A dominant issue in hybrid BHJs is thought to be the electronic
coupling between the nanocrystals, their ligands and the polymer chains; this is poorly
understood because of the complicated interplay between the chemical and electronic
structures at this interface, yet is frequently considered to be responsible for the
efficiency gap.
12,13
Understanding how such interfacial factors relate to device efficiency
remains a central challenge in hybrid solar cell research.
14–17
Herein, we focus on the
ligands present on the surface of the semiconductor nanocrystal acceptors and their effect
on hybrid BHJ solar cell performance. While much of the hybrid BHJ device
optimization has heretofore focused on the donor polymer and/or core of the
semiconductor nanocrystal acceptor (i.e., its composition and morphology), much less
effort has gone into these ligands, which as we will discuss, crucially affect both the
short-circuit current (J
SC
) and the open circuit potential (V
OC
) that can be achieved.

1.4. Surface Traps States and Nanocrystal Coupling
It is generally thought that nanocrystal surface traps strongly contribute to the limited
efficiency of exciton separation and charge transport in photovoltaic devices.
18
These
surface traps are inherent to the semiconductor nanocrystals and are a consequence of

5
their composition, stoichiometry and surface chemistry.
19,20
The effect of surface traps
on the performance of hybrid BHJ solar cells has been investigated using several steady-
state and transient techniques. McNeil and coworkers employed transient photovoltage
and photocurrent measurements to compare the dynamics of charge carriers between
P3HT:phenyl-C61-butyric acid methyl ester (PCBM) BHJs and hybrid P3HT:CdSe
BHJs.
18
Figure 1.2 compares transient photocurrent data for P3HT:PCBM and
P3HT:CdSe blends under different light intensities.

Figure 1.2. Transient short-circuit photocurrent of a,b) P3HT:PCBM BHJ device and c,d)
P3HT:CdSe BHJ device in response to a 200 µs, 525 nm pulse at varying intensities. a,c) show
the raw data while b,d) are given as normalized to the photocurrent value at 200 µs. The insets in
b) show the slight dependence of turn-on/turn-off dynamics of P3HT:PCBM as a function of light
intensity.
18

6
Distinct differences are observed for the two systems, with P3HT:PCBM showing fast
turn-on and turn-off behavior, indicative of a relatively trap-free landscape. Conversely,
the hybrid P3HT:CdSe BHJ displays both fast and slow photocurrent components. Rapid
charge trapping is observed, which they surmise to be caused by nanocrystal surface trap
states or defects; in turn, this leads to poor charge extraction of the long-lived carriers.
More recently, Greenham et al. reaffirmed that nanocrystal surface trap-mediated
recombination is a major contributor to loss of photogenerated charge carriers in hybrid
BHJ solar cells.
21
Using a combination of transient absorption spectroscopy and transient
photocurrent measurements, they probed the carrier dynamics in P3HT:CdSe BHJ solar
cells with a series of three nanocrystal acceptor diameters (i.e., 3.3, 4.3 and 5.3 nm). They
concluded that the devices utilizing the smaller nanocrystal acceptors have more
dominant trap-mediated recombination (with a higher interfacial surface area and higher
degree of trap states) than those with larger nanocrystal acceptors due to the increasing
density of surface states with decreasing nanocrystal diameter.
In the case of hybrid systems with PbS nanocrystals, it has been demonstrated by several
groups that charge transfer events can occur between the inorganic nanocrystal phase and
an organic phase, be it either P3HT or PCBM.
22,23
In the case of ternary blends of P3HT,
PCBM and PbS nanocrystals, it has been demonstrated that interfacial charge transfer
(ICT) processes are more efficient at the P3HT/PCBM and PbS/PCBM interfaces than
the P3HT/PbS interface.
23
Using a combination of ultrafast transient absorption and
time-domain terahertz spectroscopies, Schins and Siebbeles inferred that charge
localization on PbS nanocrystals leads to a strongly bound coulomb pair with the
opposite charge residing on the organic phase (i.e., P3HT or PCBM).
13
This strong

7
coulombic attraction is thought to reduce charge separation and mobility, and ultimately
lead to loss via recombination. It follows that the relatively weak electronic coupling
between nanocrystals in the BHJ prevents charge delocalization and separation of the
opposite charges. This effect is exacerbated by long chain, insulating ligands, which
were oleate ligands in the case of the PbS nanocrystals used in this study.

1.5. Surface Ligands
It follows that to manage surface traps and nanocrystal coupling in hybrid BHJ solar
cells, the ligands on the surface of semiconductor nanocrystal acceptors must be
understood and rationally controlled. Colloidal semiconductor nanocrystals are most
commonly synthesized in the presence of insulating, long-chain organic ligands. These
aliphatic organic ligands act as stabilizing agents during nanocrystal synthesis and
provide postsynthetic colloidal dispersibility in nonpolar solvents through van der Waals
stabilization. They are usually amphiphilic molecules that possess a polar headgroup that
binds to the nanocrystal surface and a nonpolar aliphatic tail. Typically utilized
headgroups include amines, carboxylic acids, thiols, phosphine oxides and phosphonic
acids. A neutral L-type ligand (e.g., amine, thiol, or phosphine oxide) coordinates to the
nanocrystal surface by dative bonding through donation of lone pairs on the ligand
headgroup. An X-type ligand (e.g., carboxylate, phosphonate, or thiolate) interacts with
the surface through an anionic moiety on the ligand headgroup (Fig. 1.3).
24,25

8

Figure 1.3. Schematic of stoichiometry dependent surface coordination chemistry of CdSe
nanocrystals.
22

Because many semiconductor nanocrystals that are used in hybrid BHJ solar cells are
nonstoichiometric at their surface, this distinction between L- and X-type ligands is an
important one. It is known that CdSe nanocrystals can possess Cd/Se ratios as high as
6:1,
24,26,27
and PbSe nanocrystals can similarly be nonstoichiometric and lead rich.
28,29

For example, Weiss and coworkers showed that the Cd/Se ratio in CdSe nanocrystals was
dependent upon the purity of the high-boiling point solvent tri-n-octylphosphine oxide
(TOPO) used in the synthesis as well as the final nanocrystal size, with the smallest
nanocrystals exhibiting the highest Cd/Se values.
30
This excess of metal cations at the
nanocrystal surface requires charge balance with the appropriate number of anionic
ligands. This means that the surface of a nonstoichiometric semiconductor nanocrystal
has to be coordinated to some amount of strongly binding X-type ligands for charge
balance, with L-type ligands or X-type ligands counterbalanced with extra-particle
cations filling out the coordination sphere of the nanocrystal.

9
As shown schematically in Fig. 1.4, if the excess metal cations on the nanocrystal surface
are undercoordinated, then the resulting surface sites will result in electronic states within
the band gap that can act as trap states (i.e., via ‘dangling bonds’).
31

Figure 1.4. Qualitative molecular orbital diagrams illustrating the bonding interactions between
undercoordinated Cd
2+
surface cations on CdSe nanocrystals and σ-donating ligands. a)
Undercoordinated Cd
2+
surface sites create a "non-bonding" state within the band gap that serves
as a trap. This creates a state frequently referred to as "dangling bonds". b) A strong σ-donating
ligand passivates the surface traps by bonding with the surface Cd
2+
and creating new bonding
and antibonding states that lie outside of the band gap. c) A weak σ -donating ligand does not
eliminate the trap because it forms an antibonding orbital with the surface Cd
2+
that is lower in
energy than the LUMO of the nanocrystal.
31

Similarly, if the organic ligands that are coordinated to the excess surface metal cations
result in bonding/antibonding states that lay within the band gap, then those electronic
states can also act as trap states. Conversely, if coordination of organic ligands to the
excess surface metal cations removes trap states via elimination of electronic states
within the band gap, then those ligands are said to pacify the surface states of the
semiconductor nanocrystal. Therefore, there are several important factors that must be
taken into account when considering the ligands on the semiconductor nanocrystal
surface: (i) whether the ligand is L- or X-type, (ii) whether the ligands are introducing or
eliminating midgap trap states and (iii) how this surface landscape changes when the

10
insulating native ligands are exchanged for smaller organic ligands.
1.6. Ligand Engineering for Device Applications
The application of semiconductor nanocrystals as acceptors in hybrid BHJ solar cells
mandates the removal or replacement of the electrically insulating native ligands, which
impair donor/ acceptor and acceptor/acceptor coupling and efficient charge transport.
32–34

This may be achieved by performing a post-synthetic ligand exchange to replace the
typically X-type insulating ligands with smaller organic ligands. This ligand exchange
can be carried out either in the colloidal phase or in the solid phase after the BHJ layer
deposition. Colloidal ligand exchange is carried out by addition of the desired
replacement ligand to a dispersion of native ligand-coated nanocrystals, followed by
addition of a suitable antisolvent to assist in the removal of displaced, long-chain native
ligands. This process may be repeated several times to increase the efficacy of ligand
exchange. Alternatively, solid phase ligand exchange can be carried out by infiltration of
the replacement ligand into the solid BHJ film. In general, a colloidal ligand exchange is
conceptually preferable because it eliminates kinetic problems associated with diffusion
of replacement ligands into and native ligands out of the solid polymer:nanocrystal BHJ
films; however, both approaches have yielded positive results in terms of device
performance (vide infra). In the context of ligand engineering for hybrid BHJ solar cells,
a thorough understanding of the ligand exchange chemistry is important for several
reasons. Firstly, the degree of ligand exchange is important for understanding the amount
of insulating material left behind on the nanocrystal acceptor. Nonquantitative ligand
exchange reactions will result in some (potentially large) fraction of long-chain native

11
ligands remaining on the nanocrystal surface, which may negatively affect charge transfer
and carrier transport. Secondly, the creation or removal of surface traps, and the density
of those trap states, as a function of ligand exchange must be considered. This has the
potential to affect both the J
SC
and the V
OC
that can be achieved in the resulting BHJ
devices.
Traditionally, the principal rationale behind this concept of ligand engineering in hybrid
BHJ solar cells had been to remove the native insulating ligands from the nanocrystal
surface to better facilitate charge transfer between the donor and acceptor and charge
collection from the nanocrystal phase.
8,32,35,36
Among the most commonly used ligand
exchange procedures for nanocrystal acceptors in hybrid BHJ solar cells is to remove the
insulating native ligands with pyridine;
34
however, ligand exchange is generally
incomplete with such a weak and neutral L-type ligand because complete exchange of
anionic X-type native ligands (charged balanced by an excess of metal cations near the
particle surface) is unfavorable without surface restructuring through loss of CdX
2
.
24
This
therefore typically leaves behind some complement of deleterious insulating ligands even
after ligand exchange. It was therefore of interest to explore new ligand systems that are
able to quantitatively displace the insulating native ligands from the nanocrystal surface.
It should be noted, however, that even with an incomplete ligand exchange with pyridine,
hybrid polymer:CdSe BHJ solar cells have reached PCEs in excess of 3%.
8,35
It is
possible that even though the ligand exchange is incomplete and insulating ligands
remain on the nanocrystal surface, the weakly binding pyridine creates ‘hot spots’ where
charge transfer becomes spatially possible between the donor polymer and the acceptor.

12
Our strategy was to explore small (low organic content), strongly binding ligands that can
exact quantitative removal of the insulating ligands and yield good interparticle coupling
once deposited as a thin film. We reported CdSe ligand exchange with tert-butylthiol,
which in the presence of base is one of only a few small organic ligand systems that have
been shown to quantitatively exchange native ligands from the nanocrystal surface, as
evidenced by thermogravimetric analysis (TGA) and FT-IR spectroscopy (Fig. 1.5a,b),
while still maintaining colloidal stability and solution processability.
37

Figure 1.5. a) Thermogravimetric analysis (TGA) of CdSe nanocrystals with their original
ligand shell (NL), and after ligand exchange with pyridine (Py) and tert-butylthiol (tBT). The
thiol-exchanged nanocrystal possess a much lower organic content and the absence of the high
temperature mass loss event at ~375 ˚C suggests that tert-butylthiol quantitatively exchanges the
native ligands whereas pyridine does not. b) ν(C–H) stretching region in the FT-IR spectrum of
CdSe nanocrystals with NL, Py, and tBT ligand shells. The spectrum for the thiol-exchanged
nanocrystals confirms a much lower organic content relative to those nanocrystals with the native
ligands or that have been pyridine exchanged. c) PL lifetime decay traces for films of neat CdSe
nanocrystals (λ
ex
= 400 nm; λ
em
= 650 nm). Lifetimes of 1.4, 3.3, and 4.4 ns were measured for
the nanocrystals with native ligands and those that had been pyridine and thiol-exchanged,
respectively.
37,38

As a result of improved interparticle coupling, electrochemical photocurrent
measurements showed a >70-fold increase in photocurrent for tert-butylthiol-exchanged
CdSe nanocrystal films as compared to pyridine-exchanged CdSe nanocrystal films,
while field effect transistor (FET) measurements corroborated a substantial increase in

13
carrier mobility with the tert-butylthiol-exchanged nanocrystals. In addition to having a
much lower degree of organic content (from quantitative ligand exchange),
photoluminescence lifetime experiments suggest that the tert-butylthiol-exchanged
nanocrystals also contain a lower density of surface traps when compared to the pyridine-
exchanged CdSe nanocrystals (Fig. 1.5c).
38
All of these data point to tert-butylthiol
being a far superior ligand for nanocrystal acceptors with regards to charge transfer and
collection, which was subsequently confirmed through the observation of higher EQE in
P3HT:CdSe BHJ solar cells utilizing tert-butylthiol ligands as compared to pyridine.
38

In addition to the acceptor surface ligands contributing to the extractable photocurrent in
devices, we introduced another effect within this context, which is the ability for ligands
to contribute to the HOMO/LUMO energies of the semiconductor nanocrystal and
consequently affect the energy offset (ΔE
DA
) between the HOMO of the donor polymer
and the LUMO of the nanocrystal acceptor. As the LUMO energy of the semiconductor
nanocrystal is increased relative to the HOMO of the conjugated polymer via ligand
exchange, the V
OC
of the device can be maximized to increase the device performance of
the hybrid BHJ solar cells. Rationally introducing small, strongly binding, electron-
donating ligands such as tert-butylthiol enables improvement in the open circuit potential
of hybrid BHJ solar cells by raising the LUMO energy level of the nanocrystal acceptor
phase and thereby increasing the energy offset from the polymer HOMO.
39
Hybrid BHJ
solar cells fabricated from blends of tert-butylthiol-exchanged CdSe nanocrystals and
P3HT achieved PCEs of 2.0%. Compared to devices made from pyridine-exchanged
(V
OC
= 0.57 V) and nonligand exchanged CdSe (V
OC
= 0.70 V), the thiol-exchanged CdSe
nanocrystals were found to consistently exhibit the highest open circuit potentials with

14
V
OC
= 0.80 V.
38
The high V
OC
associated with devices using the tert-butylthiol-
exchanged acceptors is attributed to an elevated nanocrystal LUMO energy relative to the
HOMO energy of P3HT, which produces the greatest DEDA of the three ligand types
investigated. Indeed, electrochemical determination of LUMO levels using cyclic
voltammetry, differential pulse voltammetry and spectroelectrochemistry suggests that
the thiol-exchanged CdSe nanocrystals possess the highest lying LUMO of the three
ligand sets, which translates to the highest V
OC
(Fig. 1.6).

Figure 1.6. Visible absorption spectra for a) pyridine and b) tert-butylthiol exchanged CdSe
nanocrystals spun-cast onto ITO and collected with decreasing applied bias. All potentials are
listed relative to NHE. Initial exciton bleaching is first observed at 0.54 and 0.74 V for the
pyridine and thiol-exchanged nanocrystals, respectively. c) Cyclic voltammograms for CdSe
nanocrystals exchanged with pyridine and tert-butylthiol demonstrating a higher lying LUMO for
the thiol-exchanged nanocrystals.
38

While the ability to tune the LUMO level of CdSe nanocrystals by various ligand
treatments has been previously reported,
39–43
we demonstrated for the first time that a
direct correlation exists between the ligand induced LUMO energy shift and the observed
V
OC
in operable solar cells. Therefore, simultaneous increases in both J
SC
and V
OC
in
hybrid BHJ solar cells may be realized through judicious ligand choice on the nanocrystal
surface.

15
As an extension of this concept, we demonstrated that anisotropic tert-butylthiol-
exchanged CdSe multipods could be used as the acceptor phase in BHJ hybrid solar cells
with a series of four different donor polymers, including a novel semi-random poly(3-
hexylthiophene-thiophene-diketopyrrolopyrrole) (P3HTT-DPP) copolymer, an
alternating copolymer with a low-lying HOMO level poly[N-9’-heptadecanyl-2,7-
carbazole-alt-5,5-(4’,7’-di-2-thienyl-2’,1’,3’-benzothiadiazole)] (PCDTBT), in addition
to the more widely used P3HT and poly[2,6-(4,4-bis(2-ethylhexyl)-4H-cyclopenta[2,1-
b:3,4-b’]-dithiophene)-alt-4,7-(2,1,3-benzothiadiazole)] (PCPDTBT) donors (Fig. 7).
44

Figure 1.7. Structures of the polymers used in the hybrid polymer:CdSe multipod BHJ solar
cells and the corresponding I–V curves under dark (dashed lines) and AM1.5G 1 sun illumination
(solid lines).
44

The CdSe multipod acceptors gave high performing devices for each of these very
different donor polymers using the same device structure and very similar CdSe/donor
ratios, which demonstrates the versatility of this acceptor system. PCDTBT, PCPDTBT
and P3HTT-DPP all give devices with high current densities (9.5–11.6 mA cm
−2
), while

16
PCDTBT:CdSe gives the highest open circuit potential relative to the other hybrid solar
cells (V
OC
= 0.82 V). Champion PCEs of 3.1 and 3.2% were obtained for P3HTT-DPP
and PCDTBT:CdSe hybrid BHJ solar cells, respectively, while the highest PCE was
achieved with the low band gap PCPDTBT donor polymer (η
champ
= 4.1%). These devices
outperform a reference P3HT:CdSe device (J
SC
= 8.3 mA cm
−2
, V
OC
= 0.71 V, η
champ
=
2.9%) as a result of higher photocurrent generation from the lower band gap copolymers.
Reporting high efficiencies across a range of four very different polymers using identical
processing and device architectures is unprecedented for hybrid BHJ solar cells, and is
promising with respect to using nanocrystals as generally applicable alternatives to
fullerene acceptors. This versatility is made possible by the processability and low
organic content of the tert-butylthiol-exchanged CdSe multipod acceptors.
Despite the improving device performance in hybrid BHJ solar cells, the physical
mechanisms that occur at the organic/inorganic interfaces are still poorly understood–in
particular the formation and evolution of charge transfer states and of mobile charge
carriers. While many studies of conjugated polymers and fullerene BHJs have resulted in
the detection of ICT states,
45
spectral signatures for ICT states in hybrid BHJs continue to
be more elusive.
46–48
Consequently, direct measurement of the charge transfer processes
in hybrid blends is of particular interest. We used ultrafast transient absorption
spectroscopy to probe the carrier dynamics in a PCPDTBT:CdSe hybrid BHJ utilizing
tert-butylthiol exchanged CdSe quantum dots.
49
This donor/acceptor combination gives
champion device efficiencies of 4.1% for CdSe tetra- pod acceptors and 2.1% for 4.5 nm
CdSe quantum dots.
44,49
By selectively pumping the low band gap PCPDTBT polymer,
we were able to determine that electron transfer from the polymer to the nanocrystal

17
happens on an ultrafast timescale (<65 fs) through observation of an unambiguous
spectral signature for the reduced quantum dot acceptor (Fig. 1.8a).

Figure 1.8. a) Transient absorption spectra of a neat CdSe quantum dot film (yellow), a neat
PCPDTBT film (blue), and a hybrid PCPDTBT:CdSe BHJ film (green) upon excitation at 800
nm. The negative spectral signature at 610 nm reveals that the quantum dot 1S
e
energy level is
populated by charge transfer. b) Density of electrons located on the nanocrystal conduction band
as a function of time. This curve is obtained from the deconvolution of the 610 nm TA bleach.
46

From this timescale, the electronic coupling between the polymer chains and the quantum
dots is estimated to be J ≥ 17 meV. The amplitude of the unambiguous spectral bleach
signature on the reduced quantum dot acceptors allows for the first direct calculation of
the absolute electron transfer yield in a hybrid solar cell of η
et
= 82 ± 5%. It is important
to note that the timescale for electron transfer is comparable to polymer:fullerene BHJs.
50

Therefore, the relatively high electron transfer yield does not appear to be the major
limitation in EQE. We demonstrated that a limitation of the hybrid BHJ is rapid and
measurable geminate recombination due to the small separation of the initial charge pair
(Fig. 1.8b), with a charge separation efficiency of only η
sep
= 29.5 ± 4.5%. This fast
recombination is consistent with the internal quantum efficiency of the corresponding
PCPDTBT:CdSe BHJ solar cell. We have therefore identified and quantified a major
loss mechanism in this type of solar cell. A potential means to increase the initial

18
electron–hole separation could be to use anisotropic nanostructures. This has already
proven effective in several hybrid polymer:CdSe BHJ solar cell devices with CdSe
nanorods and/or tetrapods; however, the higher photocurrents in these devices are usually
rationalized by there being fewer hopping events between nanocrystals for charge
collection. Our results suggest that in addition to this effect, these anisotropic structures
may be aiding in dissociation of the charge transfer state, which is consistent with the
work of Dayal et al.
51

1.7. State-of-the-Art Hybrid BHJ Solar Cells
The performance of hybrid solar cells has made steady progress over the past four years,
with recent reports of conversion efficiencies exceeding 5%. A summary of champion
efficiencies for several polymer:nanocrystal BHJ solar cells is presented in Table 1.1. The
highest performances are obtained with cadmium or lead chalcogenide based acceptors
that have been optimized through ligand engineering; however, even for the champion
systems, there is still significant room for further improvement. The best performing
hybrid BHJ solar cell is based on ternary PbS
x
Se
1−x
nanocrystal acceptors and has a PCE
of 5.5%.
9
This recent breakthrough was the result of coupling the nanocrystal acceptors
using a bifunctional benzenedithiol ligand, in addition to a favorable BHJ morphology in
which optimized phase segregation was achieved by using a 15:1 nanocrystal/polymer
weight loading ratio and an empirically determined thermal treatment profile. These
processing conditions resulted in nanocrystal enrichment near the cathode and polymer
enrichment near the anode. The result is promising as the device performance is not too

19
far behind the best all quantum dot solar cells with a demonstrated efficiency of 7–8%.
52

It should be noted that most of the hybrid BHJ solar cells listed in Table 1.1 utilized a
thiol ligand exchange (i.e., tert-butylthiol, 1,3-benzenedithiol or 1,2-ethanedithiol).
Although it is well known that thiols act as midgap hole traps for CdSe nanocrystals,
53

relatively high efficiencies can still be achieved with that acceptor platform. Improved
surface passivation through further ligand engineering may lead to increase in J
SC
and FF
by reducing these losses associated with surface traps and poor interfacial coupling.
Table 1.1.
A summary of high efficiency hybrid polymer:nanocrystal BHJ solar cells.

Ligand Donor Acceptor V
OC

(V)
J
SC

(mA cm
–2
)
FF PCE
(%)
Ref
EDT PCPDTBT CdSe 0.74 12.8 0.50 4.7 51
tBT PCPDTBT CdSe 0.71 11.6 0.49 4.1 41
HA PCPDTBT CdSe 0.63 8.7 0.56 3.1 52
tBT PCDTBT CdSe 0.82 10.0 0.39 3.2 41
tBT P3HTT-DPP CdSe 0.70 9.5 0.46 3.1 41
1,3-BDT PSBTBT-NH
2
CdTe 0.79 7.2 0.56 3.2 53
EDT PDTPBT PbS 0.57 13.6 0.51 3.8 54
EDT P3HT CdS 1.1 10.9 0.35 4.1 55
1,3-BDT PDTPBT PbS
x
Se
1–x
0.57 14.7 0.66 5.5 8
YD2 P3HT TiO
2
0.52 12.1 0.50 3.1 56
1,4-BDT PDPPTPT PbS 0.47 12.5 0.49 2.9 57
YD2 = zinc(II) 5,15-bis(3,5-di-tert-butylphenyl)-10-(bis(4-hexylphenyl)amino)-20-
(triisoproylsilylethynyl)porphyrin
BDT = benzenedithiol
EDT = 1,2-ethanedithiol
HA = hexanoic acid
P3HTT-DPP = poly(3-hexylthiophene-thiophene-diketopyrrolopyrrole)
PCDTBT = poly[N-9′-heptadecanyl-2,7-carbazole-alt-5,5-(4′,7′-di-2-thienyl-2′,1′,3′-
benzothiadiazole)]
PCPDTBT = poly[2,6-(4,4-bis(2-ethylhexyl)-4H-cyclopenta[2,1-b:3,4-b′]-dithiophene)-alt-4,7-
(2,1,3-benzothiadiazole)]
PDTPBT = poly(2,6-(N-(1-octylnonyl)dithieno[3,2-b:20,30-d]pyrrole)-alt-4,7-(2,1,3-
benzothiadiazole))
PSBTBT-NH
2
= poly[(4,4′-bis(2-ethylhexyl)-dithieno[3,2-b:2′,3′-d]silole)-2,6-diyl-alt-(2,1,3-
benzothiadiazole)-4,7-diyl]
PDPPTPT = poly[{2,5-bis(2-hexyldecyl)-2,3,5,6-tetrahydro-3,6-dioxopyrrolo[3,4-c]pyrrole-1,4-
diyl}-alt-{[2,2′-(1,4-phenylene)bisthiophene]-5,5′-diyl}]

20
1.8. Future Directions
The recent improvements in hybrid polymer:nanocrystal BHJ solar cells have come from
a combination of improved nanocrystal surface engineering in addition to a better
understanding of the fundamental processes that limit EQE within hybrid films.
Currently semiconductor nanocrystals are the most successful non-fullerene acceptors for
solution-processed BHJ solar cells.
61
Despite these recent gains, a more complete
fundamental under- standing into the various photophysical processes that govern charge
generation, transport and extraction in hybrid solar cells is needed, in addition to a
rational ability to explore how the BHJ morphology affects these parameters.
Understanding how these factors govern device efficiency still remains a crucial and out-
standing goal in hybrid solar cell research. Ultimately, this line of inquiry will be needed
to make this photovoltaic technology competitive with all-organic and all-nanocrystal
based solar cells. Continued focus on nanocrystal surface chemistry, specifically ligands
that are able to quantitatively replace native ligands without introducing trap states, and
ligands that can enhance nanocrystal–nanocrystal and nanocrystal–polymer interfacial
coupling, will be needed for further improvements.
Ideally, advances made through ligand engineering can be transferred to new types of
nanocrystal acceptors. Complementary spectral coverage between the donor and
acceptor has the potential to significantly broaden the photoresponse in hybrid solar cells,
leading to enhancements in J
SC
and PCE. Low band gap lead chalcogenide nanocrystals
have already been employed for this purpose and, not surprisingly, are the acceptor phase
in the current record hybrid BHJ device. However, much like the cadmium

21
chalcogenides, concerns over heavy metal toxicity and long-term stability may limit their
end potential. Intriguing, yet less investigated, electron acceptors are CuInS
2
(CIS) and
Cu
2
ZnSn(S,Se)
4
(CZTS/Se) nanocrystals. These are attractive for several reasons
including low toxicity, high earth abundance of the constituent elements and optimal
band gaps for solar photon harvesting. Initial reports of hybrid solar cells using CIS
nanocrystals have yielded very moderate results with champion PCEs around 3%.
62

Better understanding of surface chemistries as well as advances in ligand engineering in
these systems is required if they are to become relevant as acceptors in hybrid solar cells.
An additional, previously overlooked complexity of nanocrystals is their tendency to
exhibit a dynamic surface stoichiometry.
25,63
This has recently begun to receive attention,
and has been recognized as a crucial parameter to control for optoelectronic device
optimization. While typically metal rich immediately after synthesis, ligand exchange
procedures as well as solution equilibria may drastically alter the metal/chalcogenide
ratio.
63
Recent work has elucidated the dependence of transport and photoluminescence
properties on the surface stoichiometry, with metal rich surfaces generally exhibiting the
highest photoluminescence quantum yields and carrier mobilities.
63–65
Future work on
nanocrystal ligand engineering should focus on understanding the dependence of surface
stoichiometry on polymer/nanocrystal and nanocrystal/nanocrystal electronic coupling.
The ability to rationally control both the ligand environment as well as the exact
metal/chalcogenide ratios may prove vital for increasing the PCEs of hybrid solar cells
beyond the 10% threshold.

22
1.9. References
(1) Zhang, W.; Wu, Y.; Bao, Q.; Gao, F.; Fang, J. Morphological Control for Highly
Efficient Inverted Polymer Solar Cells Via the Backbone Design of Cathode
Interlayer Materials. Adv. Energy Mater. 2014, 4, 1400359.
(2) Service, R. F. Outlook Brightens for Plastic Solar Cells. Science 2011, 332, 293.
(3) Carsten, B.; Szarko, J. M.; Lu, L.; Son, H. J.; He, F.; Botro, Y. Y.; Chen, L. X.;
Yu, L. Mediating Solar Cell Performance by Controlling the Internal Dipole
Change in Organic Photovoltaic Polymers. Macromolecules 2012, 45, 6390-6395.
(4) Huynh, W. U., Dittmer, J. J., and Alivisatos, A. P. Hybrid Nanorod-Polymer Solar
Cells. Science 2002, 295, 2425-2457.
(5) Milliron D. J., Gur, I., and Alivisatos, A. P. Hybrid Organic-Nanocrystal Solar
Cells. MRS Bull. 2005, 30, 41-44.
(6) McGehee, M. D. Nanostructured Organic-Inorganic Hybrid Solar Cells. MRS
Bull. 2009, 34, 95-100.
(7) Gur, I.; Fromer, N. A.; Chen, C. P.; Kanaras, A. G.; Alivisatos, A. P. Hybrid Solar
Cells with Prescribed Nanoscale Morphologies Based on Hyperbranched
Semiconductor Nanocrystals. Nano Lett. 2007, 7, 409-414.
(8) Jeltsch, K. F.; Schadel, M.; Bonekamp, J. B.; Niyamakom, P.; Rauscher, F.;
Lademann, H. W. A.; Dumsch, I.; Allard, S.; Scherf, U.; Meerholz, K. Efficiency
Enhanced Hybrid Solar Cells Using a Blend of Quantum Dots and Nanorods. Adv.
Funct. Mater. 2012, 22, 397-404.
(9) Liu, Z.; Sun, Y.; Yuan, J.; Wei, H.; Huang, X.; Han, L.; Wang, W.; Wang, H.;
Ma, W. High-Efficiency Hybrid Solar Cells Based on Polymer/PbS
x
Se
1-x

Nanocrystals Benefiting from Vertical Phase Segregation. Adv. Mater. 2013, 25,
5772-5778.
(10) Scharber, M. C.; Muhlbacher, D.; Koppe, M.; Denk, P.; Waldauf, C.; Heeger, A.
J.; Brabec, C. J. Design Rules for Donors in Bulk-Heterojunction Solar Cells-
Towards 10% Energy-Conversion Efficiency. Adv. Mater. 2006, 18, 789-794.
(11) Meulenberg, R. W.; Lee, J. R. I.; Wolcott, A.; Zhang, J. Z.; Terminello, L. J.; van
Buuren, T. Determination of the Exciton Binding Energy in CdSe Quantum Dots.
ACS Nano 2009, 3, 325-330.

23
(12) Greenham, N. C.; Peng, X.; Alivisatos, A. P. Charge Separation and Transport in
Conjugated-Polymer/Semiconductor-Nanocrystal Composites Studied by
Photoluminescence Quenching and Photoconductivity. Phys. Rev. B. 1996, 54,
17628-17637.
(13) ten Cate, S.; Schins, J. M.; Siebbeles, L. D. A. Origin of Low Sensitizing
Efficiency of Quantum Dots in Organic Solar Cells. ACS Nano 2012, 6, 8983-
8988.
(14) Zacher, B.; Gantz, J. L.; Richards, R. E.; Armstrong, N. R. Organic Solar Cells-At
the Interface. J. Phys. Chem. Lett. 2013, 4, 1949-1952.
(15) Goh, C.; Scully, S. R.; McGehee, M. D. Effects of Molecular Interface
Modification in Hybrid Organic-Inorganic Photovoltaic Cells. J. Appl. Phys.
2007, 101, 114503.
(16) Lin, Y. Y.; Chu, T. H.; Li, S. S; Chuang, C. H.; Chang, C. H.; Su, W. F.; Chang,
C. P.; Chu, M. W.; Chen, C. W. Interfacial Nanostructuring on the Performance
of Polymer/TiO
2
Nanorod Bulk Heterojunction Solar Cells. J. Am. Chem. Soc.
2009, 131, 3644-3649.
(17) Liu, J.; Kadnikova, E. N.; Liu, Y.; McGehee, M. D.; Frechet, J. M. J.
Polythiophene Containing Thermally Removable Solubilizing Groups Enhances
the Interface and the Performance of Polymer-Titania Hybrid Solar Cells. J. Am.
Chem. Soc. 2004, 126, 9486-9487.
(18) Li, Z.; Gao, F.; Greenham, N. C.; McNeill, C. R. Comparison of the Operation of
Polymer/Fullerene, Polymer/Polymer, and Polymer/Nanocrystal Solar Cells: A
Transient Photocurrent and Photovoltage Study. Adv. Funct. Mater. 2011, 21,
1419-1431.
(19) Ginger, D. S.; Greenham, N. C. Charge Injection and Transport in Films of CdSe
Nanocrystals. J. Appl. Phys., 2000, 87, 1361-1368.
(20) Lifshitz, E.; Dag, I.; Litvitn, I. D.; Hodes, G. Optically Detected Magnetic
Resonance Study of Electron/Hole Traps on CdSe Quantum Dots. J. Phys. Chem.
B, 1998, 102, 9245-9250.
(21) Gao, F.; Li, Z.; Wang, J.; Rao, A.; Howard, I. A.; Abrusci, A.; Massip, S.;
McNeill, C. R.; Greenham, N. C. Trap-Induced Losses in Hybrid Photovoltaics.
ACS Nano, 2014, 8, 3213-3221.
(22) Jarzab, D.; Szendrei, K.; Yarema, M.; Pichler, S.; Heiss, W.; Loi, M. A. Charge-
Separation Dynamics in Inorganic-Organic Ternary Blends for Efficient Infrared
Photodiodes. Adv. Funct. Mater. 2011, 21, 1988-1992.

24
(23) Itskos, G.; Othonos, A.; Rauch, T.; Tedde, S. F.; Hayden, O.; Kovalenko, M. V.;
Heiss, W.; Choulis, S. A. Optical Properties of Organic Semiconductor Blends
with Near-Infrared Quantum-Dot Sensitizers for Light Harvesting Applications.
Adv. Energy Mater. 2011, 1, 802-812.
(24) Owen, J. S.; Park, J.; Trudeau, P. E.; Alivisatos, A. P. Reaction Chemistry and
Ligand Exchange at Cadmium-Selenide Nanocrystal Surfaces. J. Am. Chem. Soc.
2008, 130, 12279-12281.
(25) Anderson, N. C.; Owen, J. S. Soluble, Chloride-Terminated CdSe Nanocrystals:
Ligand Exchange Monitored by
1
H and
31
P NMR Spectroscopy. Chem. Mater.
2013, 25, 69-76.
(26) Rosenthal, S. J.; McBride, J.; Pennycook, S. J.; Feldman, L. C. Synthesis, Surface
Studies, Composition and Structural Characterization of CdSe, Core/Shell, and
Biologically Active Nanocrystals. Surf. Sci. Rep. 2007, 62, 111-157.
(27) Morris-Cohen, A. J.; Donakowski, M. D.; Knowles, K. E.; Weiss, E. A. The
Effect of a Common Purification Procedure on the Chemical Composition of the
Surfaces of CdSe Quantum Dots Synthesized with Trioctylphosphine Oxide. J.
Phys. Chem. C 2010, 114, 897-906.
(28) Petkov, V.; Moreels, I.; Hens, Z.; Ren, Y. PbSe Quantum Dots: Finite, Off-
Stoichiometric, and Structurally Distorted. Phys. Rev. B 2010, 81, 241304.
(29) Moreels, I.; Fritzinger, B.; Martins, J. C.; Hens, Z. Surface Chemistry of Colloidal
PbSe Nanocrystals. J. Am. Chem. Soc. 2008, 130, 15081-15086.
(30) Morris-Cohen, A. J.; Frederick, M. T.; Lilly, G. D.; McArthur, E. A.; Weiss, E. A.
Organic Surfactant-Controlled Composition of the Surfaces of CdSe Quantum
Dots. J. Phys. Chem. Lett. 2010, 1, 1078-1081.
(31) Knowles, K. E.; Tice, D. B.; McArthur, E. A.; Solomon, G. C.; Weiss, E. A.
Chemical Control of the Photoluminescence of CdSe Quantum Dot-Organic
Complexes with a Series of Para-Substituted Aniline Ligands. J. Am. Chem. Soc.
2010, 132, 1041-1050.
(32) Radychev, N.; Lokteva, I.; Witt, F.; Kolny-Olesiak, J.; Borchert, H.; Parisi, J.
Physical Origin of the Impact of Different Nanocrystal Surface Modifications on
the Performance of CdSe/P3HT Hybrid Solar Cells. J. Phys. Chem. C 2011, 115,
14111-14122.
(33) Lek, J. Y.; Xi, L.; Kardynai, B. E.; Wong, L. H.; Lam, Y. M. Understanding the
Effect of Surface Chemistry on Charge Generation and Transport in Poly (3-
hexylthiophene)/CdSe Hybrid Solar Cells. ACS Appl. Mater. Interfaces 2011, 3,
287-292.

25
(34) Lokteva, I.; Radychev, N.; Witt, F.; Borchert, H.; Parisi, J.; Kolny-Olesiak, J.
Surface Treatment of CdSe Nanoparticles for Application in Hybrid Solar Cells:
The Effect of Multiple Ligand Exchange with Pyridine. J. Phys. Chem. C 2010,
114, 12784-12791.
(35) Dayal, S.; Kopidakis, N.; Olsen, D. C.; Ginley, D. S.; Rumbles, G. Photovoltaic
Devices with a Low Band Gap Polymer and CdSe Nanostructures Exceeding 3%
Efficiency. Nano Lett. 2010, 10, 239-242.
(36) Zhou, Y.; Eck, M.; Kruger, M. Bulk-Heterojunction Hybrid Solar Cells Based on
Colloidal Nanocrystals and Conjugated Polymers. Energy Environ. Sci. 2010, 3,
1851-1864.
(37) Webber, D. H.; Brutchey, R. L. Ligand Exchange on Colloidal CdSe Nanocrystals
Using Thermally Labile tert-Butylthiol for Improved Photocurrent in Nanocrystal
Films. J. Am. Chem. Soc. 2012, 134, 1085-1092.
(38) Greaney, M. J.; Das, S.; Webber, D. H.; Bradforth, S. E.; Brutchey, R. L.
Improving Open Circuit Potential in Hybrid P3HT:CdSe Bulk Heterojunction
Solar Cells via Colloidal tert-Butylthiol Ligand Exchange. ACS Nano 2012, 6,
4222-4230.
(39) Dennler, G.; Scharber, M. C.; Brabec, C. J. Polymer-Fullerene Bulk-
Heterojunction Solar Cells. Adv. Mater. 2009, 21, 1323-1338.
(40) Albero, J.; Martinez-Ferrero, E.; Iacopio, D.; Vidal-Ferran, A.; Palomares, E.
Interfacial Charge Transfer Dynamics in CdSe/Dipole Molecules Coated
Quantum Dot Polymer Blends. Phys. Chem. Chem. Phys. 2010, 12, 13047-13051.
(41) Munro, A. M.; Zacher, B.; Graham, A.; Armstrong, N. R. Photoemission
Spectroscopy of Tethered CdSe Nanocrystals: Shifts in Ionization Potential and
Local Vacuum Level As a Function of Nanocrystal Capping Ligand. ACS Appl.
Mater. Interfaces 2010, 2, 863-869.
(42) Soreni-Harari, M.; Yaacobi-Gross, N.; Steiner, D.; Aharoni, A.; Banin, U.; Milo,
O.; Tessler, N. Tuning Energetic Levels in Nanocrystal Quantum Dots Through
Surface Manipulations. Nano Lett. 2008, 8, 678-684.
(43) Jasieniak, J.; Califano, M.; Watkins, S. E. Size-Dependent Valence and
Conduction Band-Edge Energies of Semiconductor Nanocrystals. ACS Nano
2011, 5, 5888-5902.
(44) Greaney, M. J.; Araujo, J.; Burkhart, B.; Thompson, B. C.; Brutchey, R. L. Novel
Semi-Random and Alternating Copolymer Hybrid Solar Cells Utilizing CdSe
Multipods as Versatile Acceptors. Chem. Commun. 2013, 49, 8602-8604.

26
(45) Clarke, T. M.; Durrant, J. R. Charge Photogeneration in Organic Solar Cells.
Chem. Rev. 2010, 110, 6736-6767.
(46) Witt, F.; Krusynska, M.; Borchert, H.; Parisi, J. Charge Transfer Complexes in
Organic-Inorganic Hybrid Blends for Photovoltaic Applications Investigated by
Light-Induced Electron Spin Resonance Spectroscopy. J. Phys. Chem. Lett. 2010,
1, 2999-3003.
(47) Grancini, G.; Biasiucci, M.; Mastria, R.; Scotognella, F.; Tassone, F.; Polli, D.;
Gigli, G.; Lanzani, G. Dynamic Microscopy Study of Ultrafast Charge Transfer in
a Hybrid P3HT/Hyperbranched CdSe Nanoparticle Blend for Photovoltaics. J.
Phys. Chem. Lett. 2012, 3, 517-523.
(48) Bansal, N.; Reynolds, L. X.; MacLachlan, A.; Lutz, T.; Ashraf, R. S.; Zhang, W.;
Nielsen, C. B.; McCulloch, I.; Rebois, D. G.; Kirchartz, T.; Hill, M. S.; Molloy,
K. C.; Nelson, J.; Haque, S. A. Influence of Crystallinity and Energetics on
Charge Separation in Polymer-Inorganic Nanocomposite Films for Solar Cells.
Sci. Rep. 2013, 3, 1531.
(49) Couderc, E.; Greaney, M. J.; Brutchey, R. L.; Bradforth, S. E.; Direct
Spectroscopic Evidence of Ultrafast Electron Transfer from a Low Band Gap
Polymer to CdSe Quantum Dots in Hybrid Photovoltaic Thin Films. J. Am.
Chem. Soc. 2013, 135, 18418-18426.
(50) Park, S. H.; Roy, A.; Beaupre, S.; Cho, S.; Coates, N.; Moon, J. S.; Moses, D.;
Leclerc, M.; Lee, K.; Heeger, A. J. Bulk Heterojunction Solar Cells with Internal
Quantum Efficiency Approaching 100%. Nat. Photon. 2009, 5, 297-302.
(51) Dayal, S.; Reese, M. O.; Ferguson, A. J.; Ginley, D. S.; Rumbles, G.; Kopidakis,
N. The Effect of Nanoparticle Shape on the Photocarrier Dynamics and
Photovoltaic Device Performance of Poly(3-hexylthiophene):CdSe Nanoparticle
Bulk Heterojunction Solar Cells. Adv. Funct. Mater. 2010, 20, 2629-2635.
(52) Ip, A. H.; Thon, S. M.; Hoogland, S.; Voznyy, O.; Zhitomirsky, D.; Debnath, R.;
Levina, L.; Rollny, L. R.; Carey, G. H.; Fischer, A.; Kemp, K. W.; Kramer, I. J.;
Ning, Z.; Labelle, A. J.; Chou, K. W.; Amassian, A.; Sargent, E. H. Hybrid
Passivated Colloidal Quantum Dot Solids. Nat. Nanotech. 2012, 7, 577-582.
(53) Munro, A. M.; Plante, I. J. L.; Ng, M. S.; Ginger, D. S. Quantitative Study of the
Effects of Surface Ligand Concentration on CdSe Nanocrystal
Photoluminescence. J. Phys. Chem. C 2007, 111, 6220-6227.
(54) Zhou, R.; Stadler, R.; Xie, D.; Cao, W.; Zheng, Y.; Yang, Y.; Plaisant, M.;
Holloway, P. H.; Schanze, K. S.; Reynolds, J. R.; Xue, J. Enhancing the
Efficiency of Solution-Processed Polymer:Colloidal Nanocrystal Hybrid
Photovoltaic Cells Using Ethanedithiol Treatment. ACS Nano 2013, 7, 4846-
4854.

27
(55) Zhou, Y.; Eck, M.; Men, C.; Rauscher, F.; Niyamakom, P.; Yilmaz, S.; Dumsch,
I.; Allard, S.; Scherf, U.; Kruger, M. Efficient Polymer Nanocrystal Hybrid Solar
Cells by Improved Nanocrystal Composition. Sol. Energy Mater. Sol. Cells 2011,
95, 3227-3232.
(56) Chen, H. C.; Lai, C. W.; Wu, I. C.; Pan, H. R.; Chen, I. W. P.; Peng, Y. K.; Liu,
C. L.; Chen, C. H.; Chou, P. T. Enhanced Performance and Air Stability of 3.2%
Hybrid Solar Cells: How the Functional Polymer and CdTe Nanostructure Boost
the Solar Cell Efficiency. Adv. Mater. 2011, 23, 5451-5455.
(57) Seo, J.; Cho, M. J.; Lee, D.; Cartwright, A. N.; Prasad, P. N. Efficient
Heterojunction Photovoltaic Cell Utilizing Nanocomposites of Lead Sulfide
Nanocrystals and a Low-Bandgap Polymer. Adv. Mater. 2011, 23, 3984-3988.
(58) Ren, S.; Chang, L. Y.; Lim, S. K.; Zhao, J.; Smith, M.; Zhao, N.; Bulovic, V.;
Bawendi, M.; Gradecak, S. Inorganic-Organic Hybrid Solar Cells: Bridging
Quantum Dots to Conjugated Polymer Nanowires. Nano Lett. 2011, 11, 3998-
4002.
(59) Moon, S. J.; Baranoff, E.; Zakeeruddin, S. M.; Yeh, C. Y.; Diau, E. W. G.;
Gratzel, M.; Sivula, K. Enhanced Light Harvesting in Mesoporous TiO
2
/P3HT
Hybrid Solar Cells Using a Porphyrin Dye. Chem. Commun. 2011, 47, 8244-
8246.
(60) Piliego, C.; Manca, M.; Kroon, R.; Yarema, M.; Szendrei, K.; Andersson, M. R.;
Heiss, W.; Loi, M. A. Charge Separation Dynamics in a Narrow Band Gap
Polymer-PbS Nanocrystal Blend for Efficient Hybrid Solar Cells. J. Mater. Chem.
2012, 22, 24411-24416.
(61) Eftaiha, A. F.; Sun, J. P.; Hill, I. G.; Welch, G. C. Recent Advances of Non-
Fullerene, Small Molecular Acceptors for Solution Processed Bulk
Heterojunction Solar Cells. J. Mater. Chem. A 2014, 2, 1201-1213.
(62) Rath, T.; Edler, M.; Haas, W.; Fischereder, A.; Moscher, S.; Scheck, A.; Trattnig,
R.; Sezen, M.; Mauthner, G.; Pein, A.; Meischler, D.; Bartl, K.; Saf, R.; Bansal,
N.; Haque, S. A.; Hofer, F.; List, E. J. W.; Trimmel, G. A Direct Route Towards
Polymer/Copper Indium Sulfide Nanocomposite Solar Cells. Adv. Energy Mater.
2011, 1, 1046-1050.
(63) Anderson, N. A.; Hendricks, M. P.; Choi, J. J.; Owen, J. S. Ligand Exchange and
the Stoichiometry of Metal Chalcogenide Nanocrystals: Spectroscopic
Observation of Facile Metal-Carboxylate Displacement and Binding. J. Am.
Chem. Soc. 2013, 135, 18536-18548.
(64) Kim, D. K.; Fafarman, A. T.; Diroll, B. T.; Chan, S. H.; Gordon, T. R.; Murray,
C. B.; Kagan, C. R. Solution-Based Stoichiometric Control Over Charge
Transport in Nanocrystalline CdSe Devices. ACS Nano 2013, 7, 8760-8770.

28
(65) Dirin, D. N.; Dreyfuss, S.; Bodnarchuk, M. I.; Nedelcu, G.; Papagiogis, P.; Itskos,
G.; Kovalenko, M. V. Lead Halide Perovskites and Other Metal Halide
Complexes as Inorganic Capping Ligands for Colloidal Nanocrystals. J. Am.
Chem. Soc. 2014, 136, 6550-6553.

29
Chapter 2. Improving Open Circuit Potential in Hybrid P3HT:CdSe Bulk
Heterojunction Solar Cells via Colloidal tert-Butylthiol Ligand Exchange*
*Published in ACS Nano 2012, 6, 4222-4230.

2.1. Abstract
Organic ligands have the potential to contribute to the reduction potential, or LUMO
energy, of semiconductor nanocrystals. Rationally introducing small, strongly binding,
electron-donating ligands should enable improvement in the open circuit potential of
hybrid organic/inorganic solar cells by raising the LUMO energy level of the nanocrystal
acceptor phase and thereby increasing the energy offset from the polymer HOMO.
Hybrid organic/inorganic solar cells fabricated from blends of tert-butylthiol-treated
CdSe nanocrystals and poly(3-hexylthiophene) (P3HT) achieved power conversion
efficiencies of 1.9%. Compared to devices made from pyridine-treated and non-ligand
exchanged CdSe, the thiol-treated CdSe nanocrystals are found to consistently exhibit the
highest open circuit potentials with V
OC
= 0.80 V. Electrochemical determination of
LUMO levels using cyclic voltammetry and spectroelectrochemistry suggest that the
thiol-treated CdSe nanocrystals possess the highest lying LUMO of the three, which
translates to the highest open circuit potential. Steady-state and time-resolved
photoluminescence quenching experiments on P3HT:CdSe films provide insight into how
the thiol-treated CdSe nanocrystals also achieve greater current densities in devices
relative to pyridine-treated nanocrystals, which are thought to contain a higher density of
surface traps.

30
2.2. Introduction
Organic photovoltaics (OPVs) are attractive candidates for third generation solar cell
technologies because they can be inexpensively solution processed (e.g., allowing roll-to-
roll printing), in addition to having the added benefits of being thin and flexible if the
appropriate substrates are used.
1-4
The most commonly employed bulk heterojunction
(BHJ) geometry consists of an active layer wherein a conjugated polymer donor is
blended with a fullerene acceptor.
5,6
Such types of OPVs have achieved power
conversion efficiencies (PCEs) of η
P
= 8.3%.
7
Hybrid organic/inorganic solar cells
utilize inorganic semiconductor nanocrystals as the acceptor component. Semiconductor
nanocrystals possess several attributes that should make them attractive substitutes for
fullerene acceptors; namely, (i) tunable band gaps and energy levels through
compositional control and quantum confinement effects, (ii) strong, broad absorption at
energies higher than the band edge, (iii) high dielectric constants to help overcome the
strong exciton binding energy of conjugated polymers, and (iv) high electron mobilities
relative to organic materials.
8-12
In addition, the interfacial band structure of
semiconductor nanocrystals can potentially be synthetically tuned.
13,14
Although the
PCEs of hybrid organic/inorganic solar cells currently lag behind their all-organic
counterparts, theoretical efficiencies of η
P
>12% should be attainable for conjugated
polymer/CdSe solar cells that utilize donor polymers with band gaps between 1.5-1.6 eV
and HOMO energies near -4.8 eV relative to vacuum.
12,15

Since the seminal work of Huynh, Dittmer, and Alivisatos in 2002 that described poly(3-
hexylthiophene) (P3HT):CdSe nanocrystal-based BHJ devices,
16
the P3HT:CdSe system

31
has been used as a model for hybrid organic/inorganic solar cells. The current record
efficiency for a P3HT:CdSe BHJ device stands at η
P
= 2.6%, which used 65-nm long
CdSe nanorods as the acceptor phase.
17
The anisotropic nature of the CdSe nanorods
resulted in increased current densities by facilitating electron transport along the principal
rod axis, thereby reducing the number of electron hopping events between individual
nanocrystals required for charge collection. This design criterion has also been
demonstrated to work with other anisotropic CdSe morphologies, such as tetrapods and
hyperbranched structures.
18-20
A second means through which the efficiency of hybrid
organic/inorganic solar cells can be improved is to utilize lower band gap polymers to
maximize absorption at wavelengths longer than 640 nm (i.e., the absorption edge of
P3HT). Along these lines, Dayal et al. first utilized a conjugated low band gap polymer
(PCPDTBT, E
g
≈ 1.4 eV) in conjunction with anisotropic CdSe tetrapods to achieve
champion efficiencies of η
P
= 3.2%.
21
More recently, Jeltsch et al. have combined
PCPDTBT with a mixture of 0-D and 1-D CdSe nanocrystals to achieve efficiencies of η
P

= 3.6%.
22
Here, it is thought that parallel oriented CdSe nanorods are percolated through
the active layer with CdSe quantum dot interconnects to more efficiently collect charge.
One design criterion that can be used to improve hybrid organic/inorganic BHJ cells that
is unique to these systems is the ability to molecularly tune the donor/acceptor interface
with organic ligands on the surface of the inorganic nanocrystals using simple
coordination chemistry.
22-26
Up to this point, the principal rationale behind this concept
has been to remove the large and insulating native (or legacy) ligands from the
nanocrystal surface to better facilitate charge transfer between the donor and acceptor and
charge collection from the nanocrystal phase. This has been achieved to varying degrees

32
of success by exchanging the native ligands with small molecules, such as, pyridine,
butylamine, and hexanoic acid.
21,22,27-30
In all cases, the short circuit current (J
SC
) of the
resulting devices increased considerably relative to devices prepared with the native
ligand remaining on the nanocrystal surface, achieving short circuit current densities up
to 6 mA cm
–2
for ligand exchanged P3HT:CdSe quantum dot based BHJ devices.
30

Inherent to this ligand exchange process is the creation or removal of surface trap states
on the nanocrystal, which will also contribute to the efficiency of charge collection.
Herein we consider another effect, which is the potential for ligands to contribute to the
HOMO/LUMO energies of the semiconductor nanocrystal. This could have important
consequences for device performance since V
OC
is directly proportional to the energy
offset (ΔE
DA
) between the HOMO of the donor polymer and the LUMO of the
nanocrystal acceptor.
1,5,31
If the LUMO energy of the semiconductor nanocrystal can be
increased relative to the HOMO of the conjugated polymer via facile ligand exchange,
then this would be yet another avenue to increase the device performance of
organic/inorganic hybrid solar cells.
We report an increase in device performance for P3HT:CdSe BHJ solar cells through a
colloidal ligand exchange of CdSe nanocrystals with tert-butylthiol (t-BuSH, tBT).
32

While the ligand exchange with tBT results in increased J
SC
and external quantum
efficiency (EQE) as a result of the more effective removal of insulating native ligands
when compared to pyridine (Py) exchange (Scheme 2.1), we also observed an increase in
V
OC
as a result of the tBT raising (making less negative) the level of the CdSe nanocrystal
LUMO relative to the HOMO of P3HT.

33

Scheme 2.1. Ligand exchange of as-prepared CdSe(NL) nanocrystals with pyridine (Py) and
tert-butylthiol (tBT). More complete ligand exchange is achieved with a thiol as compared to
pyridine.

A combination of thermogravimetric analysis, FT-IR spectroscopy,
(spectro)electrochemistry, steady-state and time-resolved photoluminescence
spectroscopy, and device measurements are used to elucidate the role that the tBT ligand
plays on the interfacial effects and device performance of model P3HT:CdSe BHJ
devices.

2.3. Results and Discussion
2.3.1. Colloidal Ligand Exchange
The CdSe nanocrystals were prepared according literature protocol,
33
in the presence of
stearic acid and as-received tri-n-octylphosphine oxide (TOPO), which is known to
contain phosphonic acid impurities. The nanocrystals were purified by washing four
times using toluene as the dispersant and ethanol as the flocculant. The as-prepared CdSe

34
nanocrystals possess negatively charged and strong binding native ligands (NLs; i.e.,
stearate, alkylphosphonates) that originate from the synthetic preparation. The resulting
CdSe(NL) nanocrystals were 4.5 nm in diameter, as evidenced by a clear first exciton
peak of 600 nm in the UV-vis absorption spectrum (see Figure 2.1) of a nanocrystal
dispersion in toluene.
33
A photoluminescence spectrum (see Figure 2.1) was obtained
using an excitation wavelength of 550 nm; it displayed a clear near-band edge emission at
λ
max
= 635 nm.

Figure 2.1. UV-vis absorption and PL emission spectra for a dispersion of CdSe(NL) in toluene.
A 550 nm excitation wavelength was used for the PL acquisition.
Transmission electron microscopy (TEM) analysis confirmed that the CdSe(NL)
nanocrystals were quasispherical, with a mean diameter of 4.2 ± 0.5 nm (see Figure 2.2),
in agreement with the UV-vis absorption spectra.

35

Figure 2.2. TEM image and corresponding particle size histogram for a typical batch of
CdSe(NL) nanocrystals. Based on the statistical analysis of 93 nanocrystals, the average diameter
was found to be 4.2 ± 0.5 nm, in agreement with the size determined from UV-vis analysis.

Facile ligand exchange of the CdSe(NL) nanocrystals was carried out by rinsing the
nanocrystals with tBT in tetramethyluruea (TMU) at room temperature, flocculating with
methanol/pentane, and redispersing the final suspension in TMU with a small amount of
tBT to act as an antioxidant.
32
The resulting CdSe(tBT) nanocrystals form suspensions in
TMU at concentrations up to 100 mg mL
–1
that are stable for several months when stored
at 4 ˚C in the dark. During the tBT ligand exchange process, care was taken to avoid
exposure to light to prevent formation of disulfides from photo-oxidation of tBT. For
comparison, the CdSe(NL) nanocrystals were also exchanged with the prototypical Py
ligand by stirring the as-prepared nanocrystals in neat pyridine at 80 ˚C, flocculating with
pentane, and redispersing the final suspension of CdSe(Py) in pyridine at concentrations
up to 80 mg mL
–1
.

36
To characterize the efficacy of ligand exchange, the CdSe(Py) and CdSe(tBT)
nanocrystals were analyzed using thermogravimetric analysis (TGA) and FT-IR
spectroscopy and compared to the as-prepared CdSe(NL) sample. Figure 2.3 shows an
overlay of the results obtained from typical TGA traces (ambient to 600 ˚C, 10 ˚C min
–1
,
under flowing nitrogen) for CdSe(NL), CdSe(Py), and CdSe(tBT) samples.

Figure 2.3. TGA traces of CdSe(NL) (1), CdSe(Py) (2), and CdSe(tBT) (3) obtained under
flowing nitrogen with a ramp rate of 10 ˚C min
–1
. The %mass losses for CdSe(NL), CdSe(Py)
and CdSe(tBT) were 12.0, 9.6, and 4.4%, respectively.

It is evident that ligand exchange with both Py and tBT reduces the overall organic
content on the CdSe nanocrystal surface. Taking the mass at 550 ˚C as approximating
pure CdSe, it was found that CdSe(NL) nanocrystals typically contained 12.0% organic
content by weight after purification by four toluene/ethanol washes. After ligand
exchange with Py, the organic content is reduced to 9.1%, while exchange with tBT
reduces the organic content to 4.4%, indicating a more efficient ligand exchange.
Significantly, the mass loss event at ~400 ˚C, which is attributed to loss of strongly

37
bound stearate and alkylphosphonates, is greatly reduced upon ligand exchange with
tBT.
32
This suggests that the thiol is able to bind to the nanocrystal surface in the
deprotonated thiolate form, thereby displacing the negatively charged stearate and
alkylphosphonate ligands. In contrast, Py is unable to efficiently displace these
negatively charged ligands. FT-IR spectroscopy provides further validation of the
increased effectiveness of tBT over Py ligand exchange (Figure 2.4).

Figure 2.4. FT-IR spectra of CdSe(NL) (1), CdSe(Py) (2) and CdSe(tBT) (3). The spectra were
normalized to the 2089 cm
-1
ν(CN) stretching peak of a measured Fe
4
[Fe(CN)
6
]
3
internal standard
(not shown). The spectra are offset for clarity.

A clear reduction in the ν(C-H) stretching intensity at 2850 and 2920 cm
-1
is observed
upon introduction of both ligands relative to the CdSe(NL) nanocrystals; however, ligand
exchange with tBT results in a greater decrease in ν(C-H) stretching intensity as
compared to Py treatment.

38
2.3.2. Electrochemical Characterization
Cyclic voltammetry (CV) was used to estimate the relative LUMO energies of the CdSe
nanocrystals. Films of the CdSe(NL), CdSe(Py), and CdSe(tBT) nanocrystals were
prepared by spin-casting nanocrystal suspensions onto ITO, drying the nanocrystal films
under flowing nitrogen, and measuring them under identical nonaqueous conditions. An
approximation of the effective energy of the LUMO levels can be obtained from the onset
potential of CdSe nanocrystal reduction waves.
34-36
Correcting the measured onset values
with respect to vacuum after calibration against the ferrocene/ferrocinium (Fc/Fc
+
) redox
couple, LUMO energies (relative to vacuum) are obtained according to equation 1:
36, 37

E
LUMO
= –(E
red
+ 4.5) eV (1),
where E
red
is the onset value of the reduction wave obtained from CV versus the normal
hydrogen electrode (NHE). Although there is no apparent difference in the optical band
gap after Py or tBT treatment, the reduction onset potential, and thus the position of the
LUMO relative to vacuum, is shifted following ligand treatment (Figure 2.5).

39

Figure 2.5. CV traces for CdSe(NL) (1), CdSe(Py) (2) and CdSe(tBT) (3) spun-cast onto ITO.
Data were collected with a scan rate of 20 mV s
–1
in 0.10 M TBAP with a Pt wire counter
electrode and Ag wire pseudo-reference electrode calibrated against the Fc/Fc
+
redox couple. All
potentials are given relative to NHE. Reduction onset values of -0.57 V, -0.50 V, and -0.66 V
were obtained for CdSe(NL), CdSe(Py), and CdSe(tBT), respectively.

Based on the reduction peak onsets by CV, the energy of the LUMO levels are
approximated as -3.93, -4.00 and -3.84 eV for CdSe(NL), CdSe(Py) and CdSe(tBT),
respectively. The complete set of electrochemically and optically determined parameters
for the CdSe nanocrystals and P3HT are given in Table 2.1.
Table 2.1. Optical band gaps and electrochemically determined HOMO/LUMO levels for
P3HT and CdSe nanocrystals. CV was used for the electrochemically determined values.
E
g, optical
(eV) E
red
(V) E
ox
(V)

E
LUMO
(eV) E
HOMO
(eV)

P3HT 1.9 n/a 0.70 n/a -5.25
CdSe(NL) 2.0 -0.57 n/a -3.93 -5.93
*

CdSe(Py) 2.0 -0.50 n/a -4.00 -6.00
*

CdSe(tBT) 2.0 -0.66 n/a -3.84 -5.84
*

*
E
HOMO
values were estimated by adding the optical band gap to the electrochemically determined
E
LUMO
.

In addition, differential pulse voltammetry (DPV) was employed as a means of
corroborating the trend in reduction potentials for CdSe following ligand exchange.

40
Based on the reduction peak onsets measured by DPV, the energy of the LUMO levels
are approximated as -3.86, -3.92 and -3.73 eV for CdSe(NL), CdSe(Py) and CdSe(tBT),
respectively (see Figure 2.6).

Figure 2.6. DPV of CdSe(NL), CdSe(Py), and CdSe(tBT) spun-cast on ITO collected at 10 mV
s
–1
using a Pt wire counter electrode and Ag-wire reference electrode in 0.10 M TBAP.

These LUMO energies are all slightly less negative compared to those approximated by
CV, but the absolute differences are similar and the general trend is the same. The
HOMO levels are more difficult to approximate by CV and DPV as a result of the
appearance of peak shoulders under oxidizing conditions, which may be the result of trap
states.
38

To confirm the relative positioning of CdSe nanocrystal LUMO levels for the three
samples, spectroelectrochemistry (SEC) was performed under reducing conditions. The
lowest energy exciton peak, with an absorption λ
max
= 600 nm, was monitored under
increasing negative bias in a nonaqueous electrochemical cell. The UV-vis spectra of
CdSe(NL), CdSe(Py), and CdSe(tBT) nanocrystal films collected at decreasing bias are
compared in Figure 2.7.

41

Figure 2.7. Visible absorption spectra CdSe(NL) (A), CdSe(Py) (B) and CdSe(tBT) (C) spun-
cast onto ITO collected with decreasing applied bias. Data were acquired under nitrogen in 0.10
M TBAP using a Pt wire counter electrode and Ag wire pseudo-reference electrode calibrated
against the Fc/Fc
+
redox couple. All potentials are listed relative to NHE. Visible absorption
spectra were obtained after a 2 min equilibration period following each potential step. Initial
exciton bleaching is first observed at -0.54 V, -0.59 V, and -0.74 V for CdSe(NL), CdSe(Py), and
CdSe(tBT), respectively.

As the electrochemical reduction of the CdSe nanocrystals occurs, electrons are injected
into the LUMO, which causes a bleach of the first exciton peak.
39,40
At potentials down
to -500 mV (vs. NHE), all three spectra are virtually unperturbed. At -590 mV, the
CdSe(NL) nanocrystal film begins to exhibit a bleach of the first exciton peak intensity.
Similar bleaching is observed for CdSe(Py) below -540 mV and for CdSe(tBT) below
−740 mV. Using the potentials where bleaching is first observed, the LUMO energies
can be approximated as -3.91, -3.96, and -3.76 eV for CdSe(NL), CdSe(Py), and
CdSe(tBT), respectively. Although bleaching of the first exciton peak is not observed at
potentials that exactly correspond to the reduction wave onsets from CV analysis, the
deviation is <80 mV for all three samples and the general energy trend is in agreement
with CV and DPV results. It is of interest to note that bleaching of the first exciton peak
is reversible as long as the CdSe nanocrystals are not subjected to extremely reducing or
oxidizing potentials, in which case the films undergo fatal degradation and delamination.
GREANEY ET AL.VOL.6 ’ NO. 5 ’ 4222–4230 ’ 2012
www.acsnano.org
4226
shown an inverse relationship between V
OC
and the
reverse saturation current (J
O
).
41
To confirm that the
V
OC
differencesinourhybridsolarcellswereinfactdue
to changes inΔE
DA
and not to J
O
, the dark I!V curves
for P3HT:CdSe(NL), P3HT:CdSe(Py), and P3HT:CdSe-
(tBT) were compared and no significant differences in
J
O
were observed (see Supporting Information, Figure
S4). Since P3HT is used as the donor material for all of
the devices (with an electrochemically measured
HOMO energy of!5.20 eV; see Supporting Informa-
tion, Figure S5), the relative differences in V
OC
can be
attributedtothenanocrystalLUMOenergydifferences
induced by ligand exchange.
Several groups have observed a dependence of
nanocrystal frontier energy levels on surface ligands,
withshiftsupto0.25!0.57eVobserveduponachange
in ligand.
24,26,39,42,43
While the dipole moment of or-
ganic ligands has been shown to influence the energy
levelsof2-Dsemiconductorsurfacesattheinterface,
44
no
such correlation between the frontier energy level shift
andthemagnitudeorsignofliganddipolehasyetbeen
establishedfor0-Dsemiconductornanocrystals.
24,42
Even
though the liganddipolemay not have acleareffecton
thefrontierenergylevels,ithasbeensuggestedthatthe
anchoringgroupoftheligandmayplayamoreimportant
roleinfrontierenergylevelshifts.
42
Interestingly,Jasieniak
et al. reported a lower lying HOMO for CdSe(Py) than
CdSe(alkane thiol) using photoelectron spectroscopy in
air,
43
whichagreeswiththetrendmeasuredhereforthe
LUMOenergiesofCdSe(Py)andCdSe(tBT).Itisimportant
to consider that because the ligand exchange reactions
are not always quantitative (vide supra), the resulting
nanocrystal may contain a mixed ligand shell with an
electronic structure influenced by many factors contri-
buting to the frontier energy levels.
The J
SC
values measured from I!V curves were
confirmed by integrated EQE data. Figure 6A shows
an overlay of the EQE data for devices with the three
different CdSe acceptors. The corresponding absorp-
tionspectraforneatP3HT,CdSe(tBT),andhybridP3HT:
CdSe(tBT) films are given in Figure 6B. The EQE and
absorption spectra indicate that the P3HT and CdSe
nanocrystals both contribute to the generated photo-
current. In agreement with I!V data, devices with the
CdSe(tBT) acceptors gave the highest photocurrents
withpeakEQEsgreaterthan35%at430nm.Whilethe
absorption features of P3HT overlap with the charac-
teristicfirst excitonpeak oftheCdSe nanocrystals, the
increasingphotocurrent below500nm isindicativeof
contribution by the nanocrystals.
Thedifferencesin V
OC
canberationalizedbasedon
the (spectro)electrochemical data, but the source of
the differences in J
SC
for the hybrid devices with the
three CdSe(NL), CdSe(Py), and CdSe(tBT) acceptors is
less clear. Lower overall organic content in the nano-
crystal ligand sphere is expected to facilitate greater
photocurrent,consistentwiththetrenddemonstrated
byacombinationofTGAanddevicedata.However,itis
known that the density of traps and/or defect sites on
the surface of semiconductor nanocrystals can also
affectphotoinducedchargecarrierlifetimeandrecom-
bination probability.
8,9,45
While extending the charge
carrier lifetimes can be beneficial for harvesting, it
can significantly impede charge transfer and hamper
Figure 4. Visible absorption spectra for CdSe(NL) (A), CdSe(Py) (B), and CdSe(tBT) (C) spun-cast onto ITO collected with
decreasing applied bias. Data were acquired under nitrogen in 0.10 M TBAP using a Pt wire counter-electrode and Ag wire
pseudo-reference-electrode calibrated against the Fc/Fc
þ
redox couple. All potentials are listed relative to NHE. Visible
absorptionspectrawereobtainedaftera2minequilibrationperiodfollowingeachpotentialstep.Initialexcitonbleachingis
first observed at!0.54,!0.59, and!0.74 V for CdSe(NL), CdSe(Py), and CdSe(tBT), respectively.
TABLE2. Short-CircuitCurrent(J
SC
),OpenCircuitVoltage
(V
OC
), Fill Factor (FF), and Power Conversion Efficiency
(η
P
) for P3HT:CdSe Devices under Simulated AM 1.5G
Light Illumination. The HOMO/LUMO Energy Offset
Determined by CV (ΔE
DA,CV
) for P3HT and CdSe with
eachLigandIs Givenfor Comparison
acceptor ligand J
SC
(mA cm
!2
) V
OC
(V) FF η
P
(%) ΔE
DA,CV
(eV)
NL 1.95 0.70 0.36 0.5 1.27
Py 3.69 0.57 0.47 1.0 1.20
tBT 5.62 0.80 0.43 1.9 1.36
Figure5. I!VcurvesfordeviceswithITO/PEDOT:PSS/P3HT:
CdSe/Al structure, made with CdSe(NL) (1), CdSe(Py) (2), or
CdSe(tBT) (3) nanocrystal acceptors. Device areas were
0.8 mm
2
.
ARTICLE

42
2.3.3. Hybrid Solar Cells
To investigate the effect of ligand treatment on photovoltaic performance, BHJ solar cells
with a simple ITO/PEDOT:PSS/P3HT:CdSe/Al device architecture were studied. All
devices were fabricated and tested in the ambient atmosphere without any thermal
annealing. The device characteristics are summarized in Table 2.2, and I-V curves are
given in Figure 2.8.

Table 2.2. Short-circuit current (J
SC
), open circuit voltage (V
OC
), fill factor (FF), and power
conversion efficiency (η
P
) for P3HT:CdSe devices under simulated AM 1.5G light
illumination. The HOMO/LUMO energy offset determined by CV (ΔE
DA,CV
) for P3HT and
CdSe with each ligand is given for comparison.
Acceptor Ligand J
SC
(mA cm
–2
) V
OC
(V) FF η
P
(%) ΔE
DA,CV
(eV)
NL 1.95 0.70 0.36 0.5 1.27
Py 3.69 0.57 0.47 1.0 1.20
tBT 5.62 0.80 0.43 1.9 1.36

Figure 2.8. I-V curves for devices with ITO/PEDOT:PSS/P3HT:CdSe/Al structure, made with
CdSe(NL) (1), CdSe(Py) (2), or CdSe(tBT) (3) nanocrystal acceptors. Device areas were 0.8
mm
2
.

43
In each case, an 8:1 weight ratio of CdSe nanocrystals to polymer was used, with active
layer thicknesses between 65-75 nm, as determined by X-ray reflectivity measurements.
As expected, the CdSe(NL) acceptor produces the lowest efficiency solar cells, primarily
due to low current densities and poor fill factors (FF). This can be rationalized by
considering the insulating nature of the ligand shell from the high organic content, as
determined by TGA. Considering the ambient processing conditions and lack of thermal
treatment, devices based on CdSe(Py) acceptors give reasonable efficiencies of η
P
=
1.0%. An increase in J
SC
(3.69 mA cm
–2
) and FF (0.47) relative to devices using
CdSe(NL) acceptors can be explained by the reduced barrier to interfacial charge transfer
facilitated by the smaller Py ligand shell and overall reduced organic content.
27
The V
OC

for devices using CdSe(Py) acceptors (0.57 V) is significantly less than those with
CdSe(NL) acceptors (0.70 V). Comparatively, the best P3HT:CdSe solar cells result
from devices using CdSe(tBT) acceptors and demonstrate PCEs of 1.9%. These devices
have the highest J
SC
(5.6 mA cm
–2
) and V
OC
(0.80 V), although there tends to be a slight
decrease in FF between CdSe(Py) and CdSe(tBT). Based on the electrochemically
determined LUMO energies for the three types of CdSe nanocrystal acceptors, the
observed V
OC
values directly correlate with the energy level offset (i.e., ΔE
DA
) between
the HOMO energy of P3HT and the LUMO energy of CdSe. For OPVs, it is often
observed that the theoretical maximum V
OC
is dependent on the energy level difference
between the HOMO of the donor material and the LUMO of the acceptor material.
1,5,31

In addition to the dependence of V
OC
on ΔE
DA
, Potscavage et al. have shown an inverse
relationship between V
OC
and the reverse saturation current (J
O
).
41
To confirm that the
V
OC
differences in our hybrid solar cells were in fact due to changes in ΔE
DA
and not to

44
J
O
, the dark I-V curves for P3HT:CdSe(NL), P3HT:CdSe(Py), and P3HT:CdSe(tBT)
were compared and no significant differences in J
O
were observed (see Figure 2.9).

Figure 2.9. Dark I-V characteristics for P3HT:CdSe(NL), P3HT:CdSe(Py), and
P3HT:CdSe(tBT) hybrid solar cells.

Since P3HT is used as the donor material for all of the devices, (with an
electrochemically measured HOMO energy of -5.20 eV; see Figure 2.10), the relative
differences in V
OC
can be attributed to the nanocrystal LUMO energy differences induced
by ligand exchange.

Figure 2.10. CV of P3HT spun-cast onto ITO obtained at a 20 mV s
–1
scan rate using a Pt
counter electrode and Ag wire reference electrode in acetonitrile with 0.1 M TBAP as the
supporting electrolyte.

45
Several groups have observed a dependence of nanocrystal frontier energy levels on
surface ligands, with shifts up to 0.25–0.57 eV observed upon change in ligand.
24,26,39,42,43

While the dipole moment of organic ligands has been shown to influence the energy
levels of 2-D semiconductor surfaces at the interface,
44
no such correlation between the
frontier energy level shift and the magnitude or sign of ligand dipole has yet been
established for 0-D semiconductor nanocrystals.
24,42
Even though the ligand dipole may
not have a clear effect on the frontier energy levels, it has been suggested that the
anchoring group of the ligand may play a more important role in frontier energy level
shifts.
42
Interestingly, Jasieniak et al. reported a lower lying HOMO for CdSe(Py) than
CdSe(alkane thiol) using photoelectron spectroscopy in air,
43
which agrees with the trend
measured here for the LUMO energies of CdSe(Py) and CdSe(tBT). It is important to
consider that because the ligand exchange reactions are not always quantitative (vide
supra), the resulting nanocrystal may contain a mixed ligand shell with an electronic
structure influenced by many factors contributing to the frontier energy levels.
The J
SC
values measured from I-V curves were confirmed by integrated EQE data.
Figure 2.11a shows an overlay of the EQE data for devices with the three different CdSe
acceptors. The corresponding absorption spectra for neat P3HT, CdSe(tBT), and hybrid
P3HT:CdSe(tBT) films are given in Figure 2.11b.

46

Figure 2.11. (A) EQE spectra for hybrid solar cells fabricated with P3HT:CdSe(NL) (1),
P3HT:CdSe(Py) (2), and P3HT:CdSe(tBT) (3). (B) UV-vis absorption spectra of 1:8 (wt/wt)
ratio P3HT:CdSe(tBT) (1), neat CdSe(tBT) (2), and neat P3HT (3) films on glass. The neat
P3HT and CdSe(tBT) were spun-cast from o-dichlorobenzene solutions of 2.5 mg mL
–1
P3HT
and 20 mg mL
–1
CdSe(tBT), respectively.

The EQE and absorption spectra indicate that the P3HT and CdSe nanocrystals both
contribute to the generated photocurrent. In agreement with I-V data, devices with the
CdSe(tBT) acceptors gave the highest photocurrents with peak EQEs greater than 35% at
430 nm. While the absorption features of P3HT overlap with the characteristic first
exciton peak of the CdSe nanocrystals, the increasing photocurrent below 500 nm is
indicative of contribution by the nanocrystals.
The differences in V
OC
can be rationalized based on the (spectro)electrochemical data, but
the source of the differences in J
SC
for the hybrid devices with the three CdSe(NL),
CdSe(Py), and CdSe(tBT) acceptors is less clear. Lower overall organic content in the
nanocrystal ligand sphere is expected to facilitate greater photocurrent, consistent with
the trend demonstrated by a combination of TGA and device data. However, it is known
that the density of traps and/or defect sites on the surface of semiconductor nanocrystals

47
can also affect photoinduced charge carrier lifetime and recombination probability.
8,9,45

While extending the charge carrier lifetimes can be beneficial for harvesting, it can
significantly impede charge transfer and hamper device performance if the extension is
due to carrier trapping.
46,47
To probe the dynamics of charge transfer or energy transfer
between the P3HT donor and the CdSe nanocrystal acceptor, a series of steady-state and
time-resolved photoluminescence (PL) quenching studies were performed on the three
hybrid blends. Samples were prepared with the same 8:1 wt/wt ratio of CdSe to P3HT
and under identical processing conditions to the working devices. From the steady-state
PL spectra for neat P3HT and hybrid films with the three CdSe nanocrystal acceptors
(Figure 2.12) it is clear that CdSe(Py) and CdSe(tBT) nanocrystal acceptors result in
enhanced quenching of the P3HT fluorescence when compared to the CdSe(NL)
acceptor, with approximately 85% and 90% quenching efficiency of the 635 nm peak
P3HT emission (λ
ex
= 550 nm) for blends of P3HT:CdSe(Py) and P3HT:CdSe(tBT),
respectively, compared to just 65% quenching efficiency for films of P3HT:CdSe(NL).

Figure 2.12. Steady-state PL spectra (λ
ex
= 550 nm) for films of neat P3HT (1), P3HT:CdSe(NL)
(2), P3HT:CdSe(Py) (3), and P3HT:CdSe(tBT) (4). All hybrid films were prepared using an 8:1
(wt/wt) ratio of CdSe to P3HT. Spectra were normalized to reflect the relative intensities
compared to neat P3HT emission at 635 nm.

48
The steady-state PL quenching suggests that charge transfer or energy transfer between
the P3HT and nanocrystal phases is most efficient for the CdSe(Py) and CdSe(tBT)
acceptors; however, interpretation of the steady-state data can be complicated by
variations in sample thickness and orientation within the instrument. These
complications can be avoided through the use of time-resolved photoluminescence
measurements. The PL lifetime decay plots of hybrid films with the three CdSe(NL),
CdSe(Py), and CdSe(tBT) nanocrystal acceptors are compared in Figure 2.13 (λ
em
= 650
nm; λ
ex
= 550 nm).

Figure 2.13. PL decay measurements with the corresponding fitted curves (λ
ex
= 550 nm; λ
em
=
650 nm2) for neat P3HT (A) and P3HT blended with CdSe(NL) (D), CdSe(Py) (C), and
CdSe(tBT) (B). For the hybrid films, data were fit with a sum of two exponential functions
whereas for neat P3HT one exponential function was used. P3HT lifetimes in the hybrid films
were calculated using a weighted average to be 160 ps for P3HT:CdSe(NL), 110 ps for
P3HT:CdSe(Py), and 90 ps for P3HT:CdSe(tBT). The lifetime of the neat P3HT film is 280 ps.
The raw data are shown in black, and the fit to the data are shown in red. The instrument
response of the TCSPC apparatus is shown in green. The fitting parameters for the hybrid films
are given in Table 2.3.

49
Table 2.3. Fitting parameters for the PL lifetime measurements of hybrid films.
Hybrid Film t
1
(ps) a
1
t
2
(ps) a
2
P3HT:CdSe(NL) 70 0.924 371 0.076
P3HT:CdSe(Py) 57 0.797 182 0.203
P3HT:CdSe(tBT) 30 0.839 156 0.162

Based on the neat absorption spectra for P3HT and CdSe(tBT), approximately 70% of
light absorption at 550 nm stems from P3HT with the remaining absorption due to
CdSe(tBT). Although both P3HT and CdSe nanocrystals absorb at 550 nm, the PL
lifetime measurements on hybrid films largely monitor the decay of emission from the
donor polymer since the steady-state emission line shape resembles that of P3HT (see
Figure 2.12).
As expected from the steady-state PL results, significant reduction of the P3HT PL
lifetime is observed for hybrid blends with all three CdSe acceptors. The lifetime of a
neat P3HT film was determined to be t = 280 ps. For the hybrid films, the data cannot be
fit with a single exponential, but it is most useful to characterize an average or weighted
lifetime.
1
For P3HT:CdSe(NL) the average PL lifetime was t = 160 ps, whereas for
P3HT:CdSe(Py) and P3HT:CdSe(tBT) the average lifetimes were t = 110 and 90 ps,
respectively. This suggests that compared to P3HT:CdSe(NL), the quenching of P3HT

1
Weighted average lifetime was calculated using τ = , where a
i
and t
i
correspond
to the amplitudes and time constants of individual exponents, respectively (Berberan-
Santos, M. N.; Bodunov, E. N.; Valeur, B. Mathematical Functions for the Analysis of
Luminescence Decays with Underlying Distributions: 2. Becquerel (Compressed
Hyperbola) and Related Decay Functions. Chem. Phys. 2005, 317, 57).

50
PL is more efficient when the nanocrystals are treated with Py or tBT ligands, again
indicating more efficient charge transfer or energy transfer for nanocrystals containing a
diminished ligand shell. Based on the measured average luminescence lifetimes, an
approximate P3HT PL quenching rate can be computed ignoring inhomogeneities in the
quenching; for CdSe(NL), CdSe(Py) and CdSe(tBT) this quenching measure is 2.6, 5.3
and 7.2 ns
–1
, respectively. For CdSe(tBT) the quenching rate is almost three times
greater than with the CdSe(NL) acceptor, and nearly two times greater than for the
CdSe(Py) acceptor.
From the time-resolved and steady-state PL spectra obtained for hybrid films, CdSe(tBT)
seems to enable more efficient charge transfer or energy transfer than CdSe(Py) from
P3HT based on the shorter luminescence lifetime. Furthermore, devices fabricated from
CdSe(tBT) acceptors consistently exhibit higher J
SC
and EQE than those made from
CdSe(Py) acceptors. We suspected that differences in hybrid film morphologies could in
part be responsible for the superior behavior of the P3HT:CdSe(tBT) films. However,
investigation by AFM and TEM revealed little indication of morphological differences
between P3HT:CdSe(Py) and P3HT:CdSe(tBT) films, suggesting another cause for the
increased current densities (see Figures 2.14, 2.15).

51

Figure 2.14. AFM images of hybrid BHJ films. (a) AFM topological image of P3HT:CdSe(NL).
Image was obtained using 5 × 5 µm window with a 53 nm maximum vertical height and an rms
roughness of 12 nm. (b) AFM topological image of P3HT:CdSe(Py). Image was obtained using
5 × 5 µm window with a 39 nm maximum vertical height and an rms roughness of 5 nm. (c)
AFM topological image of P3HT:CdSe(tBT). Image was obtained using 5 × 5 µm window with
a 43 nm maximum vertical height and an rms roughness of 10 nm.

Figure 2.15. TEM images of hybrid BHJ films. Panel (a) shows P3HT:CdSe(NL); panel (b)
shows P3HT:CdSe(Py), and panel (c) shows P3HT:CdSe(tBT).

The PL lifetime decay curves of neat CdSe(NL), CdSe(Py), and CdSe(tBT) nanocrystal
films (λ
em
= 650 nm;

λ
ex
= 400 nm) are given in Figure 2.16, and the normalized traces
are shown in Figure 2.17. The weighted average radiative lifetimes are determined as t =
1.4, 3.3, and 4.4 ns for CdSe(NL), CdSe(Py) and CdSe(tBT), respectively.

52

Figure 2.16. PL lifetime decay traces for films of neat CdSe films (λ
ex
= 400 nm; λ
em
= 650 nm).
Data were fit with a sum of three exponential functions, and lifetimes were determined using a
weighted average. Lifetimes of 1.4, 3.3, and 4.4 ns were measured for CdSe(NL), CdSe(Py) and
CdSe(tBT), respectively. The fitting parameters are given in Table 2.4.

Figure 2.17. Normalized PL lifetime traces of the neat P3HT film and hybrid blends (λ
ex

= 550 nm; λ
em
= 650 nm) at two different excitation fluences, 3 µJ cm
–2
(black) and 6 µJ
cm
–2
(red).

53
Table 2.4. Fitting parameters for the PL lifetime measurements of neat CdSe films.
Acceptor t
1
(ns) a
1
t
2
(ns) a
2
t
3
(ns) a
3

CdSe(NL) 0.20 0.630 0.89 0.220 2.24 0.15
CdSe(Py) 0.31 0.517 1.24 0.314 4.82 0.169
CdSe(tBT) 0.45 0.493 1.29 0.351 6.60 0.156

Assuming trap states hamper the radiative recombination of holes and electrons, it seems
plausible that semiconductor nanocrystals with the greatest density of traps will exhibit
the shortest PL lifetimes. Thus, due to the more efficient charge transfer or energy
transfer between P3HT and CdSe(tBT), as compared to CdSe(Py), as well as a greater
density of trap states in the CdSe(Py) nanocrystals, the extraction of charges from
P3HT:CdSe(Py) based devices is more difficult (as reflected by a lower J
SC
and
integrated EQE) than for P3HT:CdSe(tBT) devices.
2.4. Experimental
2.4.1. General Considerations
CdCO
3
(99.998% metals basis, “Puratronic” grade, Alfa Aesar), selenium (200 mesh
powder, 99.999% metals basis, Alfa Aesar,), tri-n-octylphosphine oxide (TOPO, 98%,
Alfa Aesar), tri-n-octylphosphine (TOP, ≥97%, Strem), stearic acid (95%, Sigma-
Aldrich), 2-methyl-2-propanethiol (tert-butylthiol, tBT, 99%, Sigma-Aldrich), and
pyridine (≥99.0%, “GR ACS” grade, EMD) were all used as received. P3HT (93% RR,
52 kDa MW, 2.2 PDI) was purchased from Reike Metals. Tetramethylurea (TMU, 99%,
Alfa Aesar) was distilled at atmospheric pressure under nitrogen before use, discarding
~5-10% of the residue in the distillation flask.

54

2.4.2. Synthesis of Native-Ligand CdSe Nanocrystals
The synthesis is based on literature methods.
33
In a typical synthesis, CdCO
3
(1.17 g,
6.79 mmol), stearic acid (10.0 g), and tri-n-octylphosphine oxide (TOPO, 10.0 g) were
stirred at 100 ˚C under flowing nitrogen (1.5 h), then held at 360 ˚C (1 h) during which
time the solution became transparent. With rapid stirring, TOP/TOPSe (0.77 g, 9.7 mmol
of selenium previously dissolved in 10 mL, 22 mmol of TOP) was quickly injected (~4
s). Exactly 2 min after the start of injection, the flask was removed from the heating bath
and the reaction was quenched via air-cooling.
The reaction product was split between three 45-mL centrifuge tubes and the reaction
flask was rinsed with 6 mL of toluene added equally to each centrifuge tube. To this was
added EtOH (20 mL/tube), and the mixture was centrifuged down (6000 rpm, 2 min), and
then the supernatant was discarded. To wash the nanocrystals, the solid was redispersed
in toluene (10 mL/tube), after which EtOH (20 mL/tube) was added as a flocculant, the
mixture was centrifuged (6000 rpm, 1 min), and then the supernatant was discarded. This
washing procedure was repeated twice. The final dispersion was made in 6 mL of
toluene, which was passed through a 0.45 µm PTFE syringe filter to give the final
product. The product was stored in the dark at 10 ˚C.

55
2.4.3. Nanocrystal Ligand Exchange
Pyridine-Exchange of CdSe Nanocrystals, CdSe(Py). A CdSe(NL) dispersion in toluene
(100 mg CdSe; 1 mL) was added to 10 mL of pyridine in a 50-mL round-bottomed flask
fitted with a small water-cooled condenser. The system was vigorously purged with
nitrogen for ~10 min, and then introduced into an 80 ˚C oil bath with slow stirring and
heated for several hours. After cooling, the dispersion was divided between two 45-mL
centrifuge tubes. Pentane (40-42 mL per tube) was added, the mixture was centrifuged
down (6000 rpm, 30 s), and the supernatant was discarded and either pyridine or CdSe-
in-pyridine from previous tubes was immediately added in order to disperse the solid.
The total volume of pyridine used for redispersion was 1-2 mL. The dispersion was
centrifuged (4000 rpm, 2 min) and carefully decanted from the small precipitate of
agglomerates in order to give the final product. The product was stored in the dark at 10
˚C.
tert-Butylthiol-Exchange of CdSe Nanocrystals, CdSe(tBT). CdSe(NL) (100 mg in 1 mL
of toluene) was placed in a 45-mL centrifuge tube, an equal volume of TMU and tBT (2
mL) were added, and the mixture was allowed to sit in the dark for 30 min. For the first
washing, 8 mL of methanol was added to flocculate before centrifugation (6000 rpm, 1
min). The colorless supernatant was discarded, and the nanocrystals were redispersed in
TMU (2 mL). Following redispersion, tBT (1 mL) was added and the mixture was
allowed to sit in the dark for 30 min after which time pentane (15 mL) was added to
induce flocculation prior to centrifugation (6000 rpm, 1 min). This tBT washing was
repeated three more times. After five tBT washings, the CdSe(tBT) nanocrystals were

56
dispersed in 1 mL of TMU, and the mixture was stirred very rapidly before bubbling
nitrogen through the liquid for ~1 min to evaporate residual pentane. At this point, tBT
(0.1 mL, as an antioxidant) was added and the mixture was centrifuged (6000 rpm, 1 min)
before decanting off the supernatant from the very tiny mass of precipitated
agglomerates. The dispersion was passed through a 0.45 µm PTFE syringe filter and
then stored in the dark without signs of agglomeration.

2.4.4. Characterization
UV-vis spectra were acquired on a Shimadzu UV-1800 spectrophotometer, using a quartz
cuvette for liquid samples or a borosilicate glass microscope slide substrate for films.
TGA measurements were made on a TA Instruments TGA Q50 instrument, using sample
sizes of 5-15 mg in an alumina crucible under a flowing nitrogen atmosphere. TGA
samples were prepared by drying the colloid under flowing nitrogen at 80-100 ˚C for up
to 90 min, then lightly crushing with a spatula. CV, DPV, and SEC experiments were
conducted using a BASi Epsilon-EC potentiostat with a C3 cell stand. All measurements
were done using dry, degassed acetonitrile under a nitrogen atmosphere. The supporting
electrolyte, tetra-n-butylammonium hexafluorophosphate (TBAP, 98%, Sigma-Aldrich),
was recrystallized from ethanol and stored in a dessicator after vacuum drying.
Acetonitrile (HPLC grade, EMD) was purified using a VAC solvent purifier. ITO
cuvette slides (Delta Technologies Ltd., Loveland, CO, 7 × 50 × 0.7 mm, 5-15 Ω) were
cleaned prior to film casting by sequential sonication in detergent, water,
tetrachloroethylene, isopropanol, and finally subjected to a 10 min UV-ozone treatment.

57
For CV, DPV, and SEC experiments, a Pt wire counter electrode and a Ag wire pseudo-
reference electrode were used. The Ag reference electrode was calibrated against the
Fc/Fc
+
redox couple, and all potentials were reported relative to NHE. Film thicknesses
were determined by X-ray reflectometry using samples spun-cast on glass slides under
identical conditions to those used for device processing. Reflectometry was performed
on a Rigaku Ultima IV diffractometer in parallel beam geometry (2-mm beam width)
using Cu Kα radiation (λ = 1.54 Å), a reduced beam intensity (40 kV, 3 mA), no
attenuator, 0-2˚ scan range, continuous scan mode, 0.02˚ step size, 0.1˚ min
–1
scan rate.
Oscillation patterns were fit with Rigaku GRX software, taking care to accurately fit the
oscillation period to ensure reliable film thicknesses.
TEM images of hybrid films were obtained on a JEOL JEM-2100 microscope at an
operating voltage of 200 kV, equipped with a Gatan Orius CCD camera. Samples were
prepared with the same 8:1 (wt/wt) ratio of CdSe to P3HT and under identical processing
conditions to the working devices. The P3HT:CdSe layer was floated off of the
PEDOT:PSS layer in water and collected on a grid for imaging. No significant
morphological differences were observed upon comparison of the three different CdSe
samples. For each sample, it appeared that the 8:1 (wt/wt) CdSe:P3HT loading produced
a uniform distribution of nanocrystals throughout the polymer matrix.
Plain-view AFM images of hybrid films on glass were obtained on a Digital Instruments
Dimension 3100 Atomic Force Microscope using intermittent contact mode equipped
with a Vista Probes Al coated Si tip with a force constant of 10 N m
-1
and a nominal tip
radius of <10 nm.

58
Time correlated single photon counting measurements (TCSPC, 22 ps time resolution)
were performed using a R3809U-50 Hamamatsu PMT with a B&H SPC-630 module.
The grating placed in the monochromator was blazed at 600 nm with 1200 g mm
–1
.
Samples were spun cast onto glass under conditions that produced optical densities
between 0.1–0.2 a.u. at the excitation wavelength. Photoluminescence lifetimes were
measured by detecting the fluorescence decay of the films at 650 nm, after excitation at
400 nm (neat CdSe films) or 550 nm (P3HT and hybrid films). Data that appear in
Figure 2.17 were measured using an excitation fluence of 3 µJ cm
–2
. These data were
found to scale linearly with PL decay traces measured at two times this excitation fluence
(6 µJ cm
–2
, see Figure 2.17), suggesting that nonlinear effects due to exciton annihilation
do not significantly contribute to the plotted data.

2.4.5. Device Fabrication and Testing
Devices were fabricated and tested in air. Aluminum shot (Al; Alfa Aesar, 99.999%) was
purchased and used as received. Patterned ITO-coated glass substrates (10 Ω cm
–2
, Thin
Film Devices, Inc.) were sequentially cleaned by sonication in detergent, water,
tetrachloroethylene, acetone, and isopropanol followed by 10 min of UV-ozone
treatment. A layer of PEDOT:PSS (Baytron P VP AI 4083, filtered through a 0.45 µm
PTFE syringe filter) was spun-cast onto the cleaned ITO and heated at 120 ˚C for 30 min
under vacuum (0.01 mmHg). Solutions of 10 mg mL
–1
P3HT were prepared in o-
dichlorobenzene, and were mixed with o-dichlorobenzene dispersions of CdSe to final
concentrations of 22.5 mg mL
–1
(2.5 mg mL
–1
P3HT, 20 mg mL
–1
CdSe), and were

59
stirred for 12-18 h in the dark prior to use. CdSe(Py) dispersions required 5% pyridine as
a co-solvent to prevent nanocrystal agglomeration. Likewise, CdSe(tBT) required 5%
TMU as a co-solvent to prevent agglomeration. Active layers passed through 0.45 µm
PTFE syringe filters were spun-cast in air onto annealed PEDOT:PSS layers (900 rpm, 45
s) forming films with thicknesses of 65-75 nm. Following spin-casting, active layer films
were dried under a nitrogen atmosphere for 20 min before loading into a high vacuum
(~2 µTorr) thermal deposition chamber (Kurt J. Lesker Co.) for Al deposition at a rate of
2 Å s
-1
. Device active areas were 0.8 mm
2
.
Current-density dependence on applied test voltage measurements were performed under
ambient conditions using a Keithley 2420 SourceMeter (sensitivity = 100 pA) in the dark
and under ASTM G173-03 spectral mismatch corrected 1000 W m
–2
white light
illumination from an AM 1.5G filtered 300 W Xenon arc lamp (Newport Oriel).
Chopped and filtered monochromatic light (250 Hz, 10 nm FWHM) from a Cornerstone
260 1/4 M double grating monochromator (Newport 74125) was used in conjunction with
an EG&G 7220 lock-in amplifier to perform all spectral responsivity measurements.

2.5. Conclusions
We fabricated hybrid solar cells that utilize CdSe nanocrystals ligand exchanged with the
small, strongly binding, and electron donating tBT ligand. P3HT:CdSe(tBT) devices
demonstrate efficiencies of 1.9% and a maximum V
OC
of 0.80 V. The high open circuit
potential associated with devices using the CdSe(tBT) acceptors is attributed to an

60
elevated nanocrystal LUMO energy relative to the HOMO energy of P3HT, which
produces the greatest ΔE
DA
of the three ligand types investigated. While the ability to
tune the LUMO level of CdSe nanocrystals by various ligand treatments has been
previously reported, we have, for the first time, demonstrated a direct correlation between
the ligand- induced LUMO energy shift and the observed open circuit potential in
operable solar cells. Excellent solution stability and processability, ambient atmosphere
fabrication, and high overall performance render the tBT ligand system attractive for
future hybrid solar cell applications.

2.6. References
(1) Lunt, R. R.; Osedach, T. P.; Brown, P. R.; Rowehl, J. A.; Bulovic, V. Practical
Roadmap and Limits to Nanostructured Photovoltaics. Adv. Mater. 2011, 23,
5712-5727.
(2) Shaheen, S. E.; Ginley, D. S.; Jabbour, G. E. Organic-Based Photovoltaics:
Toward Low-Cost Power Generation. MRS Bull. 2005, 30, 10-19.
(3) Brabec, C. J.; Dyakonov, V.; Scherf, U. Organic Photovoltaics: Materials, Device
Physics, and Manufacturing Technologies. Weinheim: Wiley-VCH, 2008.
(4) Kippelen, B.; Bredas, J. Organic Photovoltaics. Energy Environ. Sci. 2009, 2,
251-261.
(5) Dennler, G.; Scharber, M. C.; Brabec, C. J. Polymer-Fullerene Bulk-
Heterojunction Solar Cells. Adv. Mater. 2009, 21, 1323-1338.
(6) Chen, D.; Nakahara, A.; Wei, D.; Nordlund, D.; Russell, T. P. P3HT/PCBM Bulk
Heterojunction Organic Photovoltaics: Correlating Efficiency and Morphology.
Nano Lett. 2011, 11, 561-567.
(7) Green, M. A.; Emery, K.; Hishikawa, Y.; Warta, W. Solar Cell Efficiency Tables
(Version 37). Prog. Photovolt. Res. Appl. 2011, 19, 84-92.

61
(8) Moule, A. J.; Chang, L.; Thambidurai, C.; Vidu, R. Stroeve, P. Hybrid Solar
Cells: Basic Principles and the Role of Ligands. J. Mater. Chem. 2012, 22, 2351-
2368.
(9) Cozzoli, P. D.; Pellegrino, T.; Manna, L. Synthesis, Properties and Perspectives of
Hybrid Nanocrystal Structures. Chem. Soc. Rev. 2006, 35, 1195-1208.
(10) McGehee, M. D. Nanostructured Organic-Inorganic Hybrid Solar Cells. MRS.
Bull. 2009, 34, 95-100.
(11) Habas, S. E.; Platt, H. A. S.; Van Hest, M. F. A. M.; Ginley, D. S. Low-Cost
Inorganic Solar Cells: From Ink to Printed Device. Chem. Rev. 2010, 6571-6594.
(12) Xu, T.; Qiao, Q. Conjugated Polymer-Inorganic Semiconductor Hybrid Solar
Cells. Energy Environ. Sci. 2011, 4, 2700-2720.
(13) Sharma, S. N.; Sharms, H.; Singh, G.; Shivaprasad. Studies of Interaction of
Amines with TOPO/TOP Capped CdSe Quantum Dots: Role of Crystallite Size
and Oxidation Potential. Mater. Chem. Phys. 2008, 110, 471-480.
(14) Querner, C.; Reiss, P.; Sadki, S.; Zagorska, M.; Pron, A. Size and Ligand Effects
on the Electrochemical Responses of CdSe Nanocrystals. Phys. Chem. Chem.
Phys. 2005, 7, 3204-3209.
(15) Scharber, M. C.; Muhlbacher, D.; Koppe, M.; Denk, P.; Waldauf, C.; Heeger, A.
J.; Brabec, C. J. Design Rules for Donors in Bulk-Heterojunction Solar Cells –
Towards 10% Energy-Conversion Efficiency. Adv. Mater. 2006, 18, 789-794.
(16) Huynh, W. U.; Dittmer, J. J.; Alivisatos, A. P. Hybrid Nanorod-Polymer Solar
Cells. Science 2002, 295, 2425-2427.
(17) Sun, B.; Greenham, N. C. Improved Efficiency of Photovoltaics Based on CdSe
Nanorods and poly(3-Hexylthiophene) Nanofibers. Phys. Chem. Chem. Phys.
2006, 8, 3557-3560.
(18) Gur, I.; Fromer, N. A.; Chen, C. P.; Kanaras, A. G.; Alivisatos, A. P. Hybrid Solar
Cells With Prescribed Nanoscale Morphologies Based on Hyperbranched
Semiconductor Nanocrystals. Nano Lett. 2007, 7, 409-414.
(19) Talapin, D. V.; Lee, J. S.; Kovalenko, M. V.; Shevchenko, E. V. Prospects of
Colloidal Nanocrystals for Electronic and Optoelectronic Applications. Chem.
Rev. 2010, 110, 389-458.
(20) Dayal, S.; Reese, M. O.; Ferguson, A. J.; Ginley, D. S.; Rumbles, G.; Kopidakis,
N. The Effect of Nanoparticle Shape on the Photocarrier Dynamics and
Photovoltaic Device Performance of poly(3-Hexylthiophene):CdSe Nanoparticle
Bulk Heterojunction Solar Cells. Adv. Funct. Mater. 2010, 16, 2629-2635.

62
(21) Dayal, S.; Kopidakis, N.; Olson, D. C.; Ginley, D. S.; Rumbles, G. Photovoltaic
Devices with a Low Band Gap Polymer and CdSe Nanostructures Exceeding 3%
Efficiency. Nano Lett. 2010, 10, 239-242.
(22) Jeltsch, K. F.; Schadel, M.; Bonekamp, J. B.; Niyamakom, P.; Rauscher, F.;
Lademann, H. W. A.; Dumsch, I.; Allard, S.; Scherf, U.; Meerholz, K. Efficiency
Enhanced Hybrid Solar Cells Using a Blend of Quantum Dots and Nanorods. Adv.
Funct. Mater. 2012, 22, 397-404.
(23) Lefrancois, A.; Couderc, E.; Vincent, J. F.; Sadki, S.; Pron, A. Reiss, P. Effect of
the Treatment with Diamines and Dithiols on the Spectroscopic, Electrochemical
and Electrical Properties of CdSe Nanocrystals’ Thin Films. J. Mater. Chem.
2011, 21, 11524-11531.
(24) Albero, J.; Martinez-Ferrero, E.; Iacopino, D.; Vidal-Ferran, A.; Palomares, E.
Interfacial Charge Transfer Dynamics in CdSe/Dipole Molecules Coated
Quantum Dot Polymer Blends. Phys. Chem. Chem. Phys. 2010, 12, 13047-13051.
(25) Yang, S.; Prendergast, D.; Neaton, J. B. Tuning Semiconductor Band Edges
Energies for Solar Photocatalysis via Surface Ligand Passivation. Nano Lett.
2012, 12, 383-388.
(26) Munro, A. M.; Zacher, B.; Graham, A.; Armstrong, N. R. Photoemission
Spectroscopy of Tethered CdSe Nanocrystals: Shifts in Ionization Potential and
Local Vacuum Level As a Function of Nanocrystal Capping Ligand. ACS Appl.
Mater. Interfaces 2010, 2, 863-869.
(27) Zhou, Y.; Riehle, F. S.; Yuan, Y.; Schleiermacher, H. F.; Niggemann, M.; Urban,
G. A.; Kruger, M. Improved Efficiency of Hybrid Solar Cells Based on Non
Ligand-Exchanged CdSe Quantum Dots and poly(3-Hexylthiophene). Appl. Phys.
Lett. 2010, 96, 013304.
(28) Zhou, Y.; Eck, M.; Kruger, M. Bulk-Heterojunction Hybrid Solar Cells Based On
Colloidal Nanocrystals and Conjugated Polymers. Energy Environ. Sci. 2010, 3,
1851-1864.
(29) Radychev, N.; Lokteva, I.; Witt, F.; Kolny-Olesiak, J.; Borchert, H.; Parisi, J.
Physical Origins of the Impact of Different Nanocrystal Surface Modifications on
the Performance of CdSe/P3HT Hybrid Solar Cells. J. Phys. Chem. C 2011, 115,
14111-14122.
(30) Zhou, Y.; Eck, M.; Veit, C.; Zimmermann, B.; Rauscher, F.; Niyamalom, P.;
Yilmaz, S.; Dumsch, I.; Allard, S.; Scherf, U.; Kruger, M. Efficiency
Enhancement for Bulk-Heterojunction Hybrid Solar Cells Based on Acid Treated
CdSe Quantum Dots and Low Bandgap Polymer PCPDTBT. Sol. Energy Mater.
Sol. Cells 2011, 95, 1232-1237.

63
(31) Perez, M. D.; Borek, C.; Forrest, S. R.; Thompson, M. E. Molecular and
Morphological Influences on the Open Circuit Voltages of Organic Photovoltaic
Devices. J. Am. Chem. Soc. 2009, 131, 9281-9286.
(32) Webber, D. H.; Brutchey, R. L. Ligand Exchange on Colloidal CdSe Nanocrystals
Using Thermally Labile tert-Butylthiol for Improved Photocurrent in Nanocrystal
Films. J. Am. Chem. Soc. 2012, 134, 1085-1092.
(33) Yu, W. W.; Qu, L.; Guo, W.; Peng, X. Experimental Determination of the
Extinction Coefficient of CdTe, CdSe and CdS Nanocrystals. Chem. Mater. 2003,
15, 2854-2860.
(34) Li, Y.; Zhong, H.; Li, R.; Zhou, Y.; Yang, C.; Li, Y. High-Yield Fabrication and
Electrochemical Characterization of Tetrapodal CdSe, CdTe, and CdSe
x
Te
1-x

Nanocrystals. Adv. Funct. Mater. 2006, 16, 1705-1716.
(35) Baumgardner, W. J.; Choi, J. J.; Lim, Y. F.; Hanrath, T. SnSe Nanocrystals:
Synthesis, Structure, Optical Properties, and Surface Chemistry. J. Am. Chem.
Soc. 2010, 132, 9519-9521.
(36) Cardona, C. M.; Li, W.; Kaifer, A. E.; Stockdale, D.; Bazan, G. C.
Electrochemical Considerations for Determining Absolute Frontier Orbital
Energy Levels of Conjugated Polymers for Solar Cell Applications. Adv. Mater.
2011, 23, 2367-2371.
(37) Bard, Allen J., and Larry R. Faulkner. Electrochemical Methods: Fundamentals
and Applications. New York: John Wiley, 2001.
(38) Kucar, E.; Bucking, W.; Giernoth, R.; Nann, T. Determination of Defect States in
Semiconductor Nanocrystals by Cyclic Voltammetry. J. Phys. Chem. B 2005,
109, 20355-20360.
(39) Wang, C.; Shim, M.; Guyot-Sionnest, P. Electrochromic Nanocrystal Quantum
Dots. Science 2001, 291, 2390-2392.
(40) Araci, Z. O.; Shallcross, C. R.; Armstrong, N. R.; Saavedra, S. S. Potential-
Modulated Attenuated Total Reflectance Characterization of Charge Injection
Processes in Monolayer-Tethered CdSe Nanocrystals. J. Phys. Chem. Lett. 2010,
1, 1900-1905.
(41) Potscavage, W. J.; Yoo, S.; Kippelen, B. Origin of the Open-Circuit Voltage in
Multilayer Heterojunction Organic Solar Cells. Appl. Phys. Lett. 2008, 93,
193308.
(42) Soreni-Harari, M.; Yaacobi-Gross, N.; Steiner, D.; Aharoni, A.; Banin, U.; Millo,
O.; Tessler, N. Tuning Energetic Levels in Nanocrystal Quantum Dots through
Surface Manipulations. Nano Lett. 2008, 8, 678-684.

64
(43) Jasieniak, J; Califano, M.; Watkins, S. E. Size-Dependent Valence and
Conduction Band-Edge Energies of Semiconductor Nanocrystals. ACS Nano
2011, 5, 5888-5902.
(44) Vilan, A.; Shanzer, A.; Cahen, D. Molecular Control over Au/GaAs Diodes.
Nature 2000, 404, 166-168.
(45) Wuister, S. F.; Donega, C. M.; Meijerink, A. Influence of Thiol Capping on the
Exciton Luminescence and Decay Kinetics of CdTe and CdSe Quantum Dots. J.
Phys. Chem. B 2004, 108, 17393-17397.
(46) Li, Z.; Gao, F.; Greenham, N. C.; McNeill, C. Comparison of the Operation of
Polymer/Fullerene, Polymer/Polymer, and Polymer/Nanocrystal Solar Cells: A
Transient Photocurrent and Photovoltage Study. Adv. Funct. Mater. 2011, 21,
1419-1431.
(47) Noone, K. M.; Subramaniyan, S.; Zhang, Q.; Cao, G.; Jenejhe, S. A.; Ginger, D.
S. Photoinduced Charge Transfer and Polaron Dynamics in Polymer and Hybrid
Photovoltaic Thin Films: Organic vs. Inorganic Acceptors. J. Phys. Chem. C
2011, 115, 24403-24410.

65
Chapter 3. Novel Semi-Random and Alternating Co-Polymer Hybrid Solar
Cells Utilizing CdSe Multipods as Versatile Acceptors*
*Published in Chem. Commun. 2013, 49, 8602–8604.

3.1. Abstract
Hybrid solar cells with tert-butylthiol exchanged CdSe multipod acceptors and novel
semi-random poly(3-hexylthiophene-thiophene-diketopyrrolopyrrole) (P3HTT-DPP) and
alternating poly[N-9’-heptadecanyl-2,7-carbazole-alt-5,5-(4’,7’-di-2-thienyl-2’,1’,3’-
benzothiadiazole)] (PCDTBT) co-polymer donors were studied, giving high performance
devices with power conversion efficiencies >3%.

3.2. Introduction
The use of semiconductor nanocrystals as electron acceptors in bulk heterojunction (BHJ)
hybrid solar cells has garnered significant interest because nanocrystals offer spectral
tunability through quantum confinement effects, high dielectric constants, and high
electron mobilities when compared to the more commonly used fullerene acceptors.
1-5

Furthermore, the ability to tune the morphology of semiconductor nanocrystals has been
used to significantly improve device performance, with anisotropic nanorod and branched
multipod-based acceptors giving devices that exhibit increased short circuit current
densities (J
SC
) relative to those devices made from spherical quantum dots.
1,6,7
In
addition to spectral and morphological tunability, recent work has demonstrated the
ability to simultaneously increase both J
SC
and the open circuit potential (V
OC
) in hybrid
solar cells through judicious choice of ligands on the nanocrystal surface.
8

66
While numerous reports have investigated the effects of nanocrystal size, morphology,
composition, and surface ligands on hybrid solar cell performance, the majority of these
have been carried out using a singular donor material per study from a small sub-set of
polymers.
2-4
By far, the most commonly used donor polymer for hybrid solar cells has
been poly(3-hexylthiophene) (P3HT), which gives power conversion efficiencies up to
3.0% with CdSe acceptors;
9
however, the band gap (E
g
= 2.00 eV) of this polymer
provides poor spectral overlap with the solar spectrum. Consequently, more recent focus
has been placed on poly[2,6-(4,4-bis(2-ethylhexyl)-4H-cyclopenta[2,1-b:3,4-b’]-
dithiophene)-alt-4,7-(2,1,3-benzothiadiazole)] (PCPDTBT), which is a low band gap (E
g

= 1.45 eV) co-polymer comprised of alternating electron rich and electron poor subunits.
As a result of improved photon harvesting from the PCPDTBT, device efficiencies from
3.2-4.6% have been reported for hybrid solar cells that use CdSe acceptors.
10-12
Although
there have been several reports that use PCPDTBT as the donor co-polymer in hybrid
solar cells, there have been relatively few reports that use other donor co-polymer types
in hybrid systems, which is a key requirement for extension to high-performance systems
based on tandem and ternary blend hybrid solar cells.
Herein, we extend the versatility of CdSe nanocrystal acceptors by applying them to two
novel donor co-polymers. We utilize a semi-random co-polymer based on electron rich
3-hexylthiophene (3HT) and thiophene (T) units and an electron poor
diketopyrrolopyrrole (DPP) unit for the first time in a hybrid solar cell. Poly(3-
hexylthiophene-thiophene-diketopyrrolopyrrole) (P3HTT-DPP) containing 10% DPP,
80% 3HT, and 10% thiophene not only possesses a low band gap (E
g
= 1.51 eV), but also
has a very broad and intense absorption profile across the whole visible spectrum into the

67
near-infrared.
13-15
We also utilize the alternating co-polymer poly[N-9’-heptadecanyl-
2,7-carbazole-alt-5,5-(4’,7’-di-2-thienyl-2’,1’,3’-benzothiadiazole)] (PCDTBT). In
addition to possessing a low band gap (E
g
= 1.88 eV), the carbazole unit of PCDTBT
gives it a deep HOMO level, which in turn leads to higher V
OC
values as compared to
P3HT- and PCPDTBT-based devices.
16,17
High efficiency polymer:CdSe BHJ hybrid
solar cells were fabricated with P3HTT-DPP and PCDTBT using CdSe multipods as the
versatile acceptor phase, and these results were compared to P3HT and PCPDTBT
reference devices.

3.3. Results and Discussion
3.3.1. Nanocrystal Synthesis and Characterization
The CdSe multipods were synthesized following a modified literature procedure.
11
The
native ligand (NL) shell on the resulting CdSe multipods was subsequently ligand
exchanged by treatment with tert-butylthiol (tBT) utilizing pyridine as the base.
18

Thermogravimetric analysis (TGA) and a transmission electron micrograph for the
CdSe(tBT) multipods are given in Fig. 3.1.

68

Figure 3.1. (a) TGA thermograms, (b) EDX spectra, (c) TEM micrograph, and (d) FT-IR spectra
of CdSe multipods. TGA analysis shows a clear mass loss difference in the multipods before and
after ligand exchange with tBT. The efficacy of ligand exchange is corroborated by EDX and
FT-IR spectroscopies.

A majority of the multipods have an average diameter of 5 nm and arm lengths ranging
between 25-35 nm by TEM analysis. TGA clearly displays the efficacy of tBT treatment
in reducing the total organic mass content of the nanocrystals as compared to CdSe(NL).
The CdSe(NL) multipods display a relatively large mass loss (~13.0%) at ca. 400 °C,
which can be primarily attributed to loss of octylphosphonate; whereas the CdSe(tBT)
multipods have a substantially reduced overall mass loss (~4.8%) that corresponds to the
decomposition of tBT at ca. 200 °C.
18
Energy dispersive X-ray spectroscopy (EDX) and
FT-IR spectroscopy were used to further corroborate the efficacy of tBT exchange on the

69
CdSe multipods. EDX spectra clearly show the presence of a phosphorus peak in
CdSe(NL) from the octylphosphonate ligands, which is replaced by the appearance of a
peak corresponding to sulfur in CdSe(tBT) (Fig. 3.1b). Furthermore, quantitative FT-IR
spectra utilizing an internal standard show a significant reduction in the ν(C–H)
stretching intensities at 2800-3050 cm
–1
in CdSe(tBT) relative to CdSe(NL), again
corroborating exchange of tBT for the native octylphosphonate ligands (Fig. 3.1d).

3.3.2. Device Fabrication and Characterization
Active layers for hybrid solar cells with the general device structure ITO/PEDOT:PSS
(40 nm)/polymer:CdSe BHJ/ZnO (20-25 nm)/Al (100 nm) were fabricated by spin-
casting blends of P3HTT-DPP, PCDTBT, P3HT, or PCPDTBT with CdSe(tBT)
multipods dispersed in 1,2-dichlorobenzene. The exact wt/wt ratios used for the
polymer:CdSe(tBT) multipod blends studied here were determined empirically during
device optimizations for each polymer; however the optimized wt/wt ratios for these four
different polymers ranged only between 6.2-4.0:1 (CdSe/polymer). The thicknesses of
the active layers were measured by spectroscopic ellipsometry to be between 65-75 nm
for P3HTT-DPP:CdSe, 75-85 nm for PCDTBT:CdSe, 80-90 nm for P3HT:CdSe, and 90-
100 nm for PCPDTBT:CdSe. This range of thicknesses between polymers is a result of
the different concentrations used for each optimized polymer:CdSe(tBT) multipod blend
(ESI†), in which the final film thickness was proportional to the total concentration of the
deposited solution. Zinc oxide (ZnO) nanocrystals were used as an electron transport
layer between the active layer and the Al cathode to improve fill factor (FF) and J
SC
.
19

70
The devices were annealed at 160 °C for 10 min prior to thermal deposition of the Al
cathode.
The chemical structures of the four polymers used and the normalized absorption spectra
of the polymer:CdSe(tBT) multipod hybrid films are given in Fig. 3.2.

Figure 3.2. (a) Structure of the polymers used in the hybrid solar cells and (b) the corresponding
hybrid polymer:CdSe(tBT) multipod thin film absorption spectra.

The CdSe multipods possess an optical band gap of E
g
=1.90 eV with an absorption cut-
off at 665 nm (see Fig. 3.3). PCDTBT and P3HTT-DPP complement the absorption
profile of the CdSe(tBT) multipods by extending absorption into the red and near-infrared

71
region of the solar spectrum, respectively. Photoluminescence (PL) quenching
measurements were performed for each of the hybrid films, using the optimized wt/wt
ratios for each device (see Fig. 3.4).

Figure 3.3. Absorption spectrum for CdSe(tBT) multipod thin film.

Figure 3.4. PL spectra for neat polymers (solid lines) versus hybrid blends with CdSe(tBT)
multipods (dashed lines) made under identical conditions as the optimized devices. The data show
significant PL quenching for each hybrid film compared to the neat polymers, suggesting the
occurrence of energy or charge transfer. The figure legend gives the donor polymer in each
hybrid blend.
These results suggest charge transfer or energy transfer between the two phases in the
form of significantly reduced PL intensity in the hybrid films as compared to the neat
polymer films, with 61% and 47% PL quenching observed at the peak emission

72
wavelength for hybrid films with PCDTBT and P3HTT-DPP donor co-polymers,
respectively. This compares with 86% and 73% PL quenching observed at the peak
emission wavelength for hybrid films with P3HT and PCPDTBT donor polymers,
respectively.
An overlay of the current-voltage (I-V) curves and external quantum efficiency (EQE)
spectra for the four optimized polymer:CdSe(tBT) multipod hybrid solar cells is given in
Fig. 3.5, and a summary of champion and average device parameters is given in Table
3.1.

Figure 3.5. Device characteristics of the champion polymer:CdSe(tBT) multipod BHJ hybrid
solar cells with the polymer donor phases indicated in the figure captions. (a) I–V curves under
dark (dashed lines) and AM1.5G 1 sun illumination (solid lines). (b) EQE spectra of the four
corresponding champion hybrid devices.

73
Table 3.1. Photovoltaic device parameters for the polymer:CdSe(tBT) multipod BHJ
hybrid solar cells
a
Polymer J
SC

(mA cm
–2
)
V
OC

(V)
FF η
champ

(%)
η
avg
(σ)
(%)
P3HTT-DPP 9.5 0.70 0.46 3.06 2.87 (0.17)
PCDTBT 10.0 0.82 0.39 3.16 2.85 (0.18)
P3HT 8.3 0.71 0.49 2.88 2.70 (0.15)
PCPDTBT 11.6 0.71 0.49 4.05 3.75 (0.22)
a
The device parameters were measured under AM1.5G illumination at 1 sun. Average power
conversion efficiencies (η
avg
) and the associated standard deviation (σ) were determined over a
total of twelve pixels.

The integrated EQE values reproduced the J
SC
values obtained from I-V measurements
for each system to within 8-13%. In each polymer:CdSe(tBT) multipod blend, the EQE
spectra demonstrate cooperative spectral response from both the polymer and nanocrystal
phases. The contribution from the CdSe(tBT) multipod phase is apparent in the
appearance of excitonic peak-like features in the EQE spectra at ca. 505 and 620 nm, and
an increasing photoresponse at high energies presumably resulting from contributions
from the nanocrystals. Furthermore, the EQE onset matches the absorption onset for
each of the hybrid films (Fig. 3.5b).
Similar short circuit current densities were obtained for the PCDTBT:CdSe(tBT) and
P3HTT-DPP:CdSe(tBT) hybrid solar cells (J
SC
= 10.0 and 9.5 mA cm
–2
, respectively),
even though the band gap of P3HTT-DPP is 370 meV lower than that of PCDTBT. The
high current obtained for the PCDTBT:CdSe(tBT) device results from a surprisingly
strong photoresponse, with an EQE >35% from ca. 400-610 nm that surpasses 60% from
470-520 nm. As a result of its lower band gap, the P3HTT-DPP:CdSe(tBT) device gives

74
an EQE response that extends past 800 nm, with EQE between 30-50% at 400-700 nm.
Among the four systems investigated, the PCDTBT:CdSe(tBT) hybrid solar cell gives the
highest observed V
OC
of 0.82 V. This compares to a V
OC
of 0.70 V for P3HTT-
DPP:CdSe(tBT), which has a similar open circuit potential to hybrid solar cells made
from P3HT and PCPDTBT (both 0.71 V). As previously discussed, the higher V
OC
for
PCDTBT:CdSe(tBT) results from the deeper HOMO level of PCDTBT (~400 meV)
relative to the other donor polymers.
16,17
This leads to a power conversion efficiency of
3.16% in the resulting PCDTBT:CdSe(tBT) hybrid solar cell, which represents a
significant improvement in performance over the one previous example of a
PCDTBT:CdSe device (J
SC
= 6.2 mA cm
–2
, V
OC
= 0.67 V, FF = 0.46, η
champ
= 1.91%).
20

For the P3HTT-DPP:CdSe(tBT) hybrid solar cell, a similar power conversion efficiency
of 3.06% was achieved. These two devices slightly outperform a reference
P3HT:CdSe(tBT) device (J
SC
= 8.3 mA cm
–2
, η
champ
= 2.88%) as a result of higher
photocurrent generation with the lower band gap co-polymers. Conversely, the reference
PCPDTBT:CdSe(tBT) hybrid solar cell gives the highest power conversion efficiency out
of the series, with η
champ
= 4.05%. This result is consistent with PCPDTBT:CdSe(tBT)
giving the highest photocurrent (J
SC
= 11.6 mA cm
–2
) and a strong EQE response out to
~850 nm.

75
3.4. Experimental
3.4.1. CdSe Multipod Synthesis and Ligand Exchange
CdO (0.512 g, 3.99 mmol), octylphosphonic acid (1.55 g, 8.00 mmol) and
trioctylphosphine oxide (3.85 g, 9.97 mmol) were heated to 300 °C under a nitrogen
atmosphere. Once the solution became transparent, it was cooled to 25 °C and allowed to
age for 48 h. After aging, the solution was brought back to 300 °C and stirred for 1 h. A
solution of selenium (0.632 g, 8.00 mmol) in tributylphosphine (2.05 g, 10.0 mmol) was
cooled to 0 °C and then rapidly injected into the cadmium precursor solution. The
temperature was lowered to 260 °C immediately after selenium injection, and stabilized
at 250 °C. After 60 min, the reaction was quenched in a 25 °C oil bath, and 10 mL of
toluene was injected to prevent solidification. The CdSe multipod product was purified
by flocculation with isopropanol and redispersion in toluene (3×) to afford multipods
coated with the insulating native ligands, CdSe(NL). CdSe(NL) multipods (200 mg)
dispersed in 3 mL of toluene were added to 6 mL of pyridine and 2 mL of tBT. The
suspension was kept in the dark at room temperature for 1 h, and then flocculated with
pentane and isolated via centrifugation. This procedure was repeated five to six times to
ensure complete ligand exchange.

3.4.2. Materials Characterization
TEM images of hybrid films were obtained on a JEOL JEM-2100 microscope at an
operating voltage of 200 kV, equipped with a Gatan Orius CCD camera. Samples were

76
prepared with the same (wt/wt) ratio of polymer to CdSe(tBT) and under identical
processing conditions to the working devices. The polymer:CdSe(tBT) layer was floated
off of the PEDOT:PSS layer in water and collected on a copper grid for imaging. No
significant morphological differences were observed upon comparison of the four
different polymer:CdSe(tBT) samples. For each sample, it appeared that a uniform
distribution of nanocrystals throughout the polymer matrix was achieved.
FT-IR spectra were collected under nitrogen on a Jasco FT-IR 4100 from 4000-400 cm
–1

by averaging over 64 scans with a resolution of 2 cm
–1
. Semi-quantitative FT-IR was
performed on CdSe(NL) and CdSe(tBT) to help gauge the efficiency of tBT ligand
exchange. Nanocrystal samples were dried for 24 h under flowing nitrogen, and then
quantitatively mixed into a KBr matrix containing an internal standard. Specifically, 7
mg of nanocrystals were mixed into KBr (200 mg) containing a Fe
3
[Fe(CN)
6
]
4
(0.17 mg)
internal standard, and ground with a mortar and pestle for at least 10 min. A small
amount (20 mg) of each thoroughly mixed sample was pressed into a pellet and
immediately analyzed. A similarly prepared KBr blank was employed for the
background run. The spectra were normalized to the absorbance at 2090 cm
–1

corresponding the main C–H stretch of the internal standard.
UV-vis spectra were acquired on a Shimadzu UV-1800 spectrophotometer, using
borosilicate glass microscope slides as film substrates. TGA measurements were made
on a TA Instruments TGA Q50 instrument, using sample sizes of 5-15 mg in an alumina
crucible under a flowing nitrogen atmosphere heated at 10 °C min
–1
. TGA samples were
prepared by drying the colloid under flowing nitrogen at 80-100 °C for up to 90 min, and

77
then lightly crushing them with a spatula. EDX spectra were acquired using an EDAX
Apollo silicon-drift detector (model JSM 6490) mounted on a JEOL JSM-6610 SEM with
an accelerating voltage of 15 kV. Amp time was 51.2 µs, with count rate of ~3400-3700
cps giving a dead time of ~30-34%. Counts were acquired for 50 s live time. For each
sample, data were obtained for five random areas of several tens to hundred of square
microns each, and the mean value was reported.
Film thicknesses were determined using a J.A. Woollam variable angle spectroscopic
ellipsometer equipped with a 150 W Xe-arc lamp. Data were collected from 1000-1500
nm at 64°, 69°, and 74° angles incident to the sample and were fit with a Cauchy model,
in which the index of refraction can be represented by a slowly varying function of
wavelength with the form n(λ) = B + C/λ
2
, where λ is given in µm. The thickness and B
and C parameters were simultaneously fit using the Cauchy model, which is applicable to
the optical data collected between 1000-1500 nm since the films do not have any
measurable absorption in this spectral region. Reported thicknesses were derived from at
least three measurements per sample substrate. The sample thicknesses were found to
vary between 10-20% across a single film, yet the values for B and C remained within the
measured error. All samples were prepared analogously to optimized devices, but on Si
wafers.

78
3.4.3. Donor Polymers
PCPDTBT (28.5 kDa, 1.9 PDI) and PCDTBT (3.0 PDI, 56 kDa) were purchased from 1-
Material, Inc. (Quebec, Canada). P3HT (93% R.R., 52 kDa, 2.2 PDI) was purchased from
Reike Metals. P3HTT-DPP was prepared as previously reported (Macromolecules 2011,
44, 5079-5084).

3.4.4. Device Fabrication and Characterization
Devices were fabricated and tested in air. Aluminum shot (Al; Alfa Aesar, 99.999%) was
purchased and used as received. Patterned ITO-coated glass substrates (10 Ω cm
–2
, Thin
Film Devices, Inc.) were sequentially cleaned by sonication in tetrachloroethylene,
acetone, and isopropanol followed by 30 min of UV-ozone treatment. A 40 nm layer of
PEDOT:PSS (Baytron P VP AI 4083, filtered through a 0.45 µm PTFE syringe filter) was
spun-cast onto the clean ITO and heated at 120 °C for 30 min under vacuum. Donor
polymer solutions of 15 mg mL
–1
were prepared in 1,2-dichlorobenzene by dissolving
under mild heating (40-50 °C) and filtering through a 0.45 µm PTFE syringe filter. The
CdSe(tBT) multipods were probe sonicated in 1,2-dichlorobenzene for 30-45 min and
filtered through a 1 µm PTFE syringe filter. The polymer stock solutions were then
mixed with the pre-filtered 1,2-dichlorobenzene dispersions of CdSe to final
concentrations of 20:4, 21:3.4, 16:4, and 24:4 mg mL
–1
with equivalent vol/vol ratios of
1:1, 1.3:1, 0.8:1, and 1.2:1 (assuming CdSe and polymer densities of 5.8 and 1.2 g cm
–2
,
respectively) for CdSe:P3HT, CdSe:P3HTT-DPP, CdSe:PCDTBT, and CdSe:PCPDTBT,

79
respectively. Active layers were spun-cast in air onto annealed PEDOT:PSS layers (700
rpm, 60 s) and allowed to dry for 25-40 min in a dark nitrogen filled cabinet, forming
films with thicknesses between 65-100 nm. Zinc oxide nanocrystals dispersed in ethanol
(20 mg mL
–1
) were spun-cast on top of the active layers (4000 rpm, 60 s) to produce a
20-25 nm layer, and then the device was immediately annealed at 160 °C for 10 min
under flowing nitrogen. Finally, the devices were loaded into a high vacuum (~2 µTorr)
thermal deposition chamber (Angstrom Engineering) for deposition of 100-nm thick Al
cathodes through a shadow mask at a rate of 2 Å s
–1
. Device active areas were 4.3 mm
2
.
Current-density dependence on applied test voltage measurements were performed under
ambient conditions using a Keithley 2400 SourceMeter (sensitivity = 100 pA) in the dark
and under ASTM G173-03 spectral mismatch corrected 1000 W m–2 white light
illumination from an AM 1.5G filtered 450 W Xenon arc lamp (Newport Oriel). Chopped
and filtered monochromatic light (250 Hz, 10 nm FWHM) from a Cornerstone 260 1/4 M
double grating monochromator (Newport 74125) was used in conjunction with an EG&G
7220 lock-in amplifier to perform all spectral responsivity measurements. Reported
device metrics were averaged over 12 individual pixels from at least three separate
substrates.

3.5. Conclusions
In summary, we used tBT-exchanged CdSe multipods as the acceptor phase in BHJ
hybrid solar cells with four different donor polymers, including a novel semi-random co-

80
polymer and a novel alternating co-polymer. The CdSe(tBT) multipod acceptors gave
high performing devices for each of these very different polymers using the same device
structure and very similar CdSe/polymer ratios, which demonstrates the versatility of this
acceptor system. The use of tBT as a surface ligand for the CdSe multipods allows for
quantitative removal of the native ligands, and assists in charge collection in the hybrid
solar cells. Both PCDTBT and P3HTT-DPP give devices with high current densities,
while PCDTBT:CdSe(tBT) gives the highest open circuit potential relative to the other
hybrid solar cells. Champion power conversion efficiencies of 3.06 and 3.16% were
obtained for P3HTT-DPP:CdSe(tBT) and PCDTBT:CdSe(tBT) hybrid solar cells,
respectively.

3.6. References
(1) Huynh, W. U., Dittmer, J. J., and Alivisatos, A. P. Hybrid Nanorod-Polymer Solar
Cells. Science 2002, 295, 2425-2457.
(2) Günes, S.; Sariciftci, N. S. Hybrid Solar Cells. Inorg. Chim. Acta 2008, 361, 581-
588.
(3) Moulé, A. J.; Chang, L.; Thambidurai, C.; Vidu, R.; Stroeve, P. Hybrid Solar
Cells: Basic Principles and the Role of Ligands. J. Mater. Chem. 2012, 22, 2351-
2368.
(4) Xu, T.; Qiao, Q. Conjugated Polymer-Inorganic Semiconductor Hybrid Solar
Cells. Energy Environ. Sci. 2011, 4, 2700-2720.
(5) Ren, S.; Chang, L. Y.; Lim, S. K.; Zhao, J.; Smith, M.; Zhao, N.; Bulović, V.;
Bawendi, M.; Gradečak, S. Inorganic–Organic Hybrid Solar Cell: Bridging
Quantum Dots to Conjugated Polymer Nanowires. Nano Lett. 2011, 11, 3998-
4002.

81
(6) Dayal, S.; Reese, M. O.; Ferguson, A. J.; Ginley, D. S.; Rumbles, G.; Kopidakis,
N. The Effect of Nanoparticle Shape on the Photocarrier Dynamics and
Photovoltaic Device Performance of Poly(3-hexylthiophene):CdSe Nanoparticle
Bulk Heterojunction Solar Cells. Adv. Funct. Mater. 2010, 20, 2629-2635.
(7) Gur, I.; Fromer, N. A.; Chen, C. P.; Kanaras, A. G.; Alivisatos, A. P. Hybrid Solar
Cells with Prescribed Morphologies Based on Hyperbranched Semiconductor
Nanocrystals. Nano Lett. 2007, 7, 409-414.
(8) Greaney, M. J.; Das, S.; Webber, D. H.; Bradforth, S. E.; Brutchey, R. L.
Improving Open Circuit Potential in Hybrid P3HT:CdSe Bulk Heterojunction
Solar Cells via Colloidal tert-Butylthiol Ligand Exchange. ACS Nano 2012, 6,
4222-4230.
(9) Fu, W.; Shi, Y.; Qiu, W.; Wang, L.; Nan, Y.; Shi, M.; Li, H.; Chen, H. High
Efficiency Hybrid Solar Cells Using Post-Deposition Ligand Exchange by
Monothiols. Phys. Chem. Chem. Phys. 2012, 14, 12094-12098.
(10) Jeltsch, K. F.; Schadel, M.; Bonekamp, J. B.; Niyamakom, P.; Rauscher, F.;
Lademann, H. W. A.; Dumsch, I.; Allard, S.; Scherf, U.; Meerholz, K. Enhanced
Efficiency Hybrid Solar Cells Using a Blend of Quantum Dots and Nanorods.
Adv. Funct. Mater. 2012, 22, 397-404.
(11) Dayal, S.; Kopidakis, N.; Olson, D. C.; Ginley, D. S.; Rumbles, G. Photovoltaic
Devices with a Low Band Gap Polymer and CdSe Nanostructures Exceeding 3%
Efficiency. Nano Lett. 2010, 10, 239-242.
(12) Zhou, R.; Stalder, D.; Xie, D.; Cao, W.; Zheng, Y.; Yang, Y.; Plaisant, M.;
Halloway, P. H.; Shanze, K. S.; Reynolds, J. R.; Xue. Enhancing the Efficiency of
Solution-Processed Polymer:Colloidal Nanocrystal Hybrid Photovoltaic Cells
Using Ethanedithiol Treatment. ACS Nano 2013, 7, 4846-4854.
(13) Khylabich, P. P.; Burkhart, B.; Thompson, B. C. Compositional Dependence of
the Open-Circuit Voltage in Ternary Blend Bulk Heterojunction Solar Cells
Based on Two Donor Polymers. J. Am. Chem. Soc. 2012, 134, 9074-9077.
(14) Khylabich, P. P.; Burkhart, B.; Ng, C. F.; Thompson, B. C. Efficient Solar Cells
From Semi-Random P3HT Analogues Incorporating Diketopyrrolopyrrole.
Macromolecules 2011, 44, 5079-5084.
(15) Burkhart, B.; Khlyabich, P. P.; Thompson, B. C. Influence of the Acceptor
Composition on Physical Properties and Solar Cell Performance in Semi-Random
Two-Acceptor Copolymers. ACS Macro Lett. 2012, 1, 660-666.
(16) Park, S. H.; Roy, A.; Beaupre, S.; Cho, S.; Coates, N.; Moon, J. S.; Moses, D.;
Leclerc, M.; Lee, K.; Heeger, A. J. Bulk Heterojunction Solar Cells with Internal
Quantum Efficiency Approaching 100%. Nature Photon. 2009, 3, 297-302.

82
(17) Blouin, N.; Michaud, A.; Gendron, D.; Wakim, S.; Blair, E.; Neagu-Plesu, R.;
Belletete, M.; Durocher, G.; Tao, Y.; Leclerc, M. Toward a Rational Design of
Poly(2, 7-Carbazole) Derivatives for Solar Cells. J. Am. Chem. Soc. 2008, 130,
732-742.
(18) Webber, D. H.; Brutchey, R. L. Ligand Exchange on Colloidal CdSe Nanocrystals
Using Thermally Labile tert-Butylthiol for Improved Photocurrent in Nanocrystal
Films. J. Am. Chem. Soc. 2012, 134, 1085-1092.
(19) Qian, L.; Yang, J.; Zhou, R.; Tang, A.; Zheng, Y.; Tseng, T. K.; Bera, D.; Xue, J.;
Holloway, P. H. Hybrid Polymer-CdSe Solar Cells with a ZnO Nanoparticle
Buffer Layer for Improved Efficiency and Lifetime. J. Mater. Chem. 2011, 21,
3814-3817.
(20) Lek, J. Y.; Lam, Y. M.; Niziol, J.; Marzec, M. Understanding Polycarbazole-
Based Polymer:CdSe Hybrid Solar Cells. Nanotechnology 2012, 23, 315401.

83
Chapter 4. Direct Spectroscopic Evidence of Ultrafast Electron Transfer
from a Low Band Gap Polymer to CdSe Quantum Dots in Hybrid
Photovoltaic Thin Films *
*Published in J. Am. Chem. Soc. 2013, 135, 18418–18426.

4.1. Abstract
Ultrafast transient absorption spectroscopy is used to study charge transfer dynamics in
hybrid films composed of the low band gap polymer PCPDTBT and CdSe quantum dots
capped with tert-butylthiol ligands. By selectively exciting the polymer, a spectral
signature for electrons on the quantum dots appears on ultrafast time scales (≲ 65 fs),
which indicates ultrafast electron transfer. From this time scale, the coupling between the
polymer chains and the quantum dots is estimated to be J ≳ 17 meV. The reduced
quantum dot acceptors exhibit an unambiguous spectral bleach signature, whose
amplitude allows for the first direct calculation of the absolute electron transfer yield in a
hybrid solar cell (82 ± 5%). We also show that a limitation of the hybrid system is rapid
and measurable geminate recombination due to the small separation of the initial charge
pair. The fast recombination is consistent with the internal quantum efficiency of the
corresponding solar cell. We therefore have identified and quantified a main loss
mechanism in this type of third generation solar cell.

84
4.2. Introduction
Bulk heterojunction (BHJ) hybrid solar cells comprised of semiconductor nanocrystals
blended with conjugated polymers are demonstrating improved photovoltaic device
performance (up to η = 4−5%) over the past 12 years.
1−3
Semiconductor nanocrystals
offer spectral tunability through compositional and quantum confinement effects, high
dielectric constants, and high electron mobilities when compared to commonly used
fullerene acceptors; however, hybrid device efficiencies currently lag behind those of all-
organic heterojunctions. Design rationale specific to hybrid systems are being developed
to further enhance the efficiencies of these solar cells,
2,4
but the physical mechanisms that
occur at the hybrid interfaces are still poorly understood − in particular the formation and
evolution of charge transfer states and of mobile charge carriers. While many studies of
conjugated polymers and fullerene derivative blends have resulted in the detection of
interfacial charge-transfer (ICT) states
5
(e.g., through emission below the band gap of the
donor6 or through new photoinduced absorption (PIA) bands in transient absorption (TA)
measurements
7
), possible spectral signatures for such ICT states in hybrid blends have
been more elusive.
8−10
A dominant issue in hybrids is the electronic coupling between the
nanocrystals, their ligands, and the polymer chains, which is poorly understood yet
frequently considered to be responsible for the lower efficiencies of hybrid devices
11,12

when compared to all-organic solar cells. Consequently, direct measurement of the
charge transfer processes in hybrid blends is of particular interest. It has heretofore been
difficult, however, to disentangle the spectral signatures of all the transient species at
play; as a result, measuring how fast these transient species form and decay and/or
determining the absolute yield of charge carrier formation has remained a challenge. In

85
most BHJs, the charge transfer yields are evaluated from the contribution of the excited
species residing on the polymer because the spectral signature of the acceptor anion is
only poorly resolved.
13,14
Despite the spectroscopic signatures of transient species on the
acceptor phase in hybrid BHJs,
15−17
the question of charge transfer yield is still largely
unanswered.
12,14,18
Furthermore, ultrafast spectroscopy studies have focused on hybrid
heterojunctions giving relatively poor photovoltaic efficiencies as a result of long,
insulating ligands on the nanocrystal acceptor,
12,19
or reduced absorption of one of the
hybrid components (i.e., either wide band gap nanocrystal acceptors
15,16,20
or donor
polymer
17
).
Herein, we characterize the electron transfer dynamics and electron transfer yield in a
hybrid BHJ blend that (i) demonstrates high power conversion efficiency,
21
as a result of
ligand engineering,
2,4
and (ii) allows for clear elucidation of the electron transfer rate and
yield because the donor and acceptor phases can be spectrally resolved. The BHJ blends
of poly[2,6-(4,4-bis(2-ethylhexyl)-4H-cyclopenta[2,1-b;3,4-b′]-dithiophene)-alt-4,7-
(2,1,3-benzothiadiazole)] (PCPDTBT, E
g, opt
≈ 1.5 eV) with tert-butylthiol (tBT)-capped
CdSe quantum dots, or CdSe(tBT) QDs, give a power conversion efficiency of 2.1%. To
study the excited state dynamics, we used spectrally resolved TA spectroscopy with a
time resolution of 65 fs. In such an experiment, the sample is excited by an ultrafast laser
pulse (pump), and its absorption spectrum is measured at a time delay Δt after the pump
by a broadband laser pulse (probe). Spectra presented in this paper are given as
differences between the steady-state absorption (pump off) and the excited state
absorption (pump on) of the sample. The TA spectra therefore explicitly show the
depletion of ground state- based transitions and the appearance of excited state

86
transitions. We designed the hybrid system and TA experiment to minimize the number
of physical processes occurring at the donor/acceptor interfaces, therefore allowing for
straightforward extraction of physical parameters. In particular, when the PCPDTBT is
selectively photoexcited, due to the relative alignment of the energy levels of the donor
and acceptor, polymer excitons may only dissociate through electron transfer into the QD
conduction band. Consequently, this work reveals an unambiguous spectral signature for
the QD anion, providing evidence of electron transfer from a donor polymer to a
semiconductor nanocrystal acceptor, which occurs within 65 fs for this system. The
spectral evidence of excited state species on the acceptor phase is unique to hybrids and
allows the electron transfer yield (82 ± 5%) to be determined and the evolution of the
electron population on the QDs to be tracked. Finally, rapid geminate recombination is
identified and quantified as the main loss mechanism in the corresponding solar cell.

4.3. Results and Discussion
4.3.1. Steady-State Characterization of the Hybrid Blends
We investigated BHJs of 4.5 nm CdSe(tBT) QDs with the low band gap conjugated
polymer PCPDTBT.
24
The extended absorption profile of PCPDTBT (i.e., absorption
cutoff at ∼850 nm) coupled with the effects of the tBT ligands on the electronic coupling
between QDs
25
result in an efficient BHJ solar cell. Figure 4.1 shows the absorption
spectra of films of PCPDTBT, CdSe(tBT) QDs, and the resulting hybrid BHJ.

87

Figure 4.1. (a) Normalized absolute absorbance spectra of a CdSe(tBT) QDs solution (solid line),
a neat PCPDTBT film (dashed line), and a PCPDTBT solution (dotted line). The inset shows the
structure of the donor polymer, PCPDTBT. (b) Absolute absorbance of a PCPDTBT:CdSe(tBT)
hybrid film with 1:8 wt/wt polymer/QD spin-cast from 1,2-dichlorobenzene onto quartz, and
external quantum efficiency of the corresponding ITO/PEDOT:PSS/PCPDTBT:CdSe-
(tBT)/ZnO/Al device. The inset shows the energy diagram of the BHJ relative to vacuum, with
energy levels of PCPDTBT24 and of 4.5 nm CdSe(tBT) QDs,
4
all determined from
electrochemistry measurements.

The 4.5 nm CdSe(tBT) QDs (E
g
≈ 2.1 eV) exhibit excitonic transitions at 600 nm
(overlapping 1S
e
−1S
3/2
and 1S
e
−2S
3/2
transitions) and 500 nm (1P
e
−1P
3/2
transition).
26,27

A change in the polymer morphology in the presence of the QDs is evidenced by a blue
shift of the PCPDTBT peak absorbance relative to that of the neat polymer film (see
Figure 4.2), resulting from the QDs inhibiting packing of polymer chains and/or from the
cosolvent required during processing of the hybrid BHJ films.
28

88

Figure 4.2. a) Absolute absorbance spectra of a neat PCPDTBT film and of hybrid films
containing various weight fractions of QDs. b) The same, normalized to 1.0 at their reddest λ
max
,
and compared to the normalized absolute absorbance spectra of neat PCPDTBT solution and of
neat CdSe(tBT) solution.

The external quantum efficiency (EQE) provides evidence for dissociation of excitons
generated on both the donor and acceptor phases (Figure 4.1b). This is evidenced in the
EQE spectrum by cooperative spectral response from both the nanocrystals (e.g.,
excitonic features at 500 and 600 nm) and the polymer at energies lower than the band
gap of the QDs (e.g., 750 nm). The corresponding PCPDTBT:CdSe(tBT) QD BHJ hybrid
solar cell gives a short circuit current of J
SC
= 7.25 mA cm
−2
, an open circuit potential of
V
OC
= 0.76 V, and a fill factor (FF) of 0.38, leading to a power conversion efficiency of
η
P
= 2.1% for a ITO/PEDOT:PSS/PCPDTBT:CdSe(tBT)/ ZnO/Al device structure (see
Figure 4.3).
! S3!
!
Figure S1. a) Absolute absorbance spectra of a neat PCPDTBT film and of hybrid films
containing various weight fractions of QDs. b) The same, normalized to 1.0 at their reddest λ
max
,
and compared to the normalized absolute absorbance spectra of neat PCPDTBT solution and of
neat CdSe(tBT) solution.
!
Photovoltaic device testing
The dark and one-sun illumination I-V characteristics of a photovoltaic device, whose
active layer is identical to that study by TA, are given in Figure S2. Current density dependence
on applied test voltage measurements were performed under ambient conditions using a Keithley
2400 SourceMeter in the dark and under ASTM G173-03 spectral mismatch corrected 1000 W
m
–2
white light illumination from an AM 1.5G filtered 300 W xenon arc lamp (Newport Oriel).
The resulting device parameters were J
SC
= 7.25 mA cm
–2
, V
OC
= 0.76 V, and fill factor of 0.38.
1
0
Absolute absorbance (normalized)
1000 800 600 400
Wavelength (nm)
Nanocrystals
PCPDTBT solution
PCPDTBT film
Hybrids (polymer:QD weight; nanocrystal fractions)
1:0.66 ; 40 wt. % ; 10 vol. %
1:1.5 ; 60 wt. % ; 20 vol. %
1:4 ; 80 wt. % ; 40 vol. %
1:8 ; 90 wt. % ; 60 vol. %
b)
0.15
0.10
0.05
0.00
Absolute absorbance (OD)
1000 800 600 400
Wavelength (nm)
Neat PCPDTBT
Hybrids
40 wt. % ; 10 vol. %
60 wt. % ; 20 vol. %
80 wt. % ; 40 vol. %
90 wt. % ; 60 vol. %
a)

89

Figure 4.3. I–V curve for a PCPDTBT:CdSe(tBT) QD hybrid solar cell in a polymer/QD 1:8
wt/wt ratio.

The active layer of this device is the same as that studied by TA spectroscopy (vide
infra). Efficiencies of 4.1% have been achieved for this system by substituting the CdSe
QDs with anisotropic CdSe multipods.
2

In order to understand how the two components of the hybrid heterojunction give rise to
photocurrent upon photoexcitation, we performed both steady-state and time-resolved
spectroscopic measurements. Photoluminescence tracks the excited state population and
dynamics. The presence of CdSe(tBT) QDs in a PCPDTBT matrix results in the
reduction of the polymer photoluminescence and is an indication of the interaction
between the two components. Photoluminescence quenching experiments (Figure 4.4)
illustrate that the addition of CdSe(tBT) QDs reduces the photo- luminescence efficiency
of PCPDTBT by as much as 75 ± 5% for the maximal QD content studied here (1:10
wt/wt polymer/QD).
! S4!
!
Figure S2. I–V curve for a PCPDTBT:CdSe(tBT) QD hybrid solar cell in a polymer/QD 1:8
wt/wt ratio.

Ultrafast transient absorption spectroscopy
1. Experimental setup
White light supercontinuum probe pulses were polarized perpendicular to the pump pulse
and generated in different ways depending on the probed wavelength range. The 380-620 nm
and 580-890 nm broadband continua were generated using a rotating, 2 mm thick CaF
2
window
seeded with 800 nm and 1315 nm pulses respectively and detected with a Si Hamamatsu S3901-
256Q photodiode array. The 850-1150 nm and 1000-1480 nm broadband continua were
generated in a sapphire window, seeded with 800 nm and 1315 nm pulses respectively and
detected with an InGaAs Hamamatsu G9213-256S photodiode array. The 1315 nm WL seed was
generated using ca. 10% of the amplifier’s output to seed a type II Spectra Physics OPA-800C.
To minimize probe dispersion, a pair of off-axis aluminum parabolic mirrors was used to
collimate the probe and focus it into the sample, while a CaF
2
lens focused the pump. Spectra
were measured for a range of pump fluences from 30 to 2400 µJ cm
–2
, with a typical fluence of
180 µJ cm
–2
for the 800 nm pump.
-10
-5
0
5
Current density (mA/cm
2
)
-0.5 0.0 0.5
Potential (V)
dark
light
J
sc
= 7.25 mA/cm
2
V
oc
= 0.76 V
fill factor 0.38
efficiency 2.1 %

90

Figure 4.4. Photoluminescence efficiency as a function of the QD content in the hybrid, for an
excitation wavelength of 780 nm. The quantum dot content was evaluated from the ratio of the
absorbance at the wavelength at which the polymer absorbs least (at 480 nm the absorbance is
mainly due to QDs) and of the absorbance at the wavelength at which the polymer absorbs most
(714−774 nm depending on the film). The photoluminescence efficiency was determined by
integrating the absorbance-corrected PL signals of each film from 850 to 1100 nm. The relative
absorbance of the sample presented in the transient absorption section is shown by a black arrow.

To understand the type of interaction that results in photoluminescence quenching, it is
important to note that the respective band gaps of the components prevent uphill energy
transfer from the polymer to the QDs (inset of Figure 4.1b). Furthermore, while the
modifications of the polymer packing detected by absorbance measurements can affect
the photoluminescence efficiency by altering radiative and nonradiative decay pathways,
we would expect that the more disordered, glassy phase of PCPDTBT observed in the
hybrid films would be more emissive than the more ordered phase in neat PCPDTBT
films, as observed in P3HT.
29
Consequently, the photoluminescence quenching
experiments suggest that charge transfer does indeed occur at the interface between the
PCPDTBT and the CdSe(tBT) QDs, as confirmed by the experimental evidence
presented below.

91
4.3.2. Transient Absorption Spectra of Hybrid Components
TA measurements were performed to directly quantify the charge-transfer processes at
play in the hybrid blend. The low band gap polymer was selectively excited by pumping
the films at 800 nm, which is in the low-energy tail of the S
1
←S
0
transition of
PCPDTBT.
30
First, we established that all measurements were performed in the linear
regime, as determined by probing the initial singlet population in neat PCPDTBT films
for a range of pump fluence (Figure 4.5a).

Figure 4.5. (a) Transient absorption signal amplitude of a neat PCPDTBT film at 1400 nm, taken
at 1 ps delay after the pump pulse. (b) Time-resolved transient absorption signal for a
PCPDTBT:CdSe(tBT) QD hybrid film probed at 1400 nm (singlet exciton, see text) after
excitation by a 800 nm pump pulse for two different pump fluences.

92
The inflection of the curve at around 250 µJ cm
−2
(corresponding to 1.4 × 10
14
absorbed
800 nm photons per cm
2
) shows the threshold above which exciton−exciton and
exciton−charge annihilation effects become non-negligible. The threshold fluence 250 µJ
cm
−2
corresponds to a number density ρ
c
of absorbed photons of 9 × 10
−2
nm
−3
in a 15 nm
film. The pump fluence used for the production run TA experiments was set at 180 µJ
cm
−2
for the 800 nm and for the 550 nm pumps. We further checked that this pump
fluence indeed corresponds to the annihilation-free regime in hybrid films by recording
the time-resolved exciton decay for different fluences. Figure 4.5b shows that the exciton
decays are comparable for fluences of 180 µJ cm
−2
and 52 µJ cm
−2
, indicating the absence
of nonlinear processes at these pump fluences.
Now spectral features for neat CdSe(tBT) QD, neat PCPDTBT polymer, and hybrid films
are compared in Figure 4.6 for different time delays between the pump and the probe.
The dynamics of the different excited state populations are extracted and summarized in
Figure 4.7.

93

Figure 4.6. (a) Transient absorption spectra of a neat CdSe(tBT) film, of a neat PCPDTBT film
(15−20 nm), and of a hybrid PCPDTBT:CdSe(tBT) film (1:8 wt/wt, 45−55 nm) at different time
delays between the pump and the probe pulses. I
0
is the incident pump photon density (I
0
∼ 7.3 ×
10
14
cm
−2
for the 800 nm pump; 5 × 10
14
cm
−2
for the 550 nm pump), while F
a
is the fraction of
800 nm photons absorbed by the films (F
a
∼ 0.14 for both the polymer and the QD films, F
a
∼
0.09 for the hybrid). We identify the spectral features discussed in the text by color-coding. (b)
Open symbols show the QD bleach formation (600−620 nm, violet shading) within the time
resolution of the setup. The small signal in the trace from the neat polymer film, full symbols, is
due to the overlapping polymer ground state bleach.

94

Figure 4.7. Comparison of the decay dynamics of the excited species in the neat PCPDTBT film
(squares, blue symbols) and in the hybrid film (triangles, brown symbols), with short time
dynamics given as insets. Time traces were obtained by averaging data over the following
wavelength ranges: green 855−885 nm, orange 900−1050 nm, red 1250−1300 nm, brown
1400−1480 nm. The full lines are multiexponential fits to the data to serve as guides to the eyes;
their coefficients can be found in Table 4.1.

Table 4.1. Decay dynamics of the excited species identified in the main text: multi-
exponential fit parameters for the data presented in Figure 4.7.

The data point traces for the hybrid samples in the orange, red, and brown wavelength ranges
were smoothed by 5 points boxcar averaging.

! S10!
Table S1. Decay dynamics of the excited species identified in the main text: multi-exponential
fit parameters for the data presented in Figure 5.
Wavelength range PCPDTBT film PCPDTBT:CdSe QD film
1:8 wt:wt ratio
855-885 nm
(attributed to PCPDTBT
polarons II)
t
1
= 0.1 ps (6%)
t
2
= 2 ps (60%)
t
3
= 50 ps (18%)
t
4
> 500 ps (16%)
t
1
= 0.2 ps (5%)
t
2
= 3 ps (64%)
t
3
= 136 ps (24%)
t
4
> 500 ps (7%)
900-1050 nm
(attributed to electrons on
QD)
t
1
= 1 ps (68%)
t
2
= 20 ps (32%)
t
3
> 500 ps (0%)
t
1
= 0.2 ps (15%)
t
2
= 7ps (62%)
t
3
> 500 ps (22%)
1250-1300nm
(attributed to PCPDTBT
polarons I)
t
1
= 1 ps (15%)
t
2
= 20 ps (79%)
t
3
> 100 ps (6%)
t
1
= 0.4 ps (19%)
t
2
= 18 ps (61%)
t
3
> 100 ps (20%)
1400-1480 nm
(S
n
 S
1
, attributed to
PCPDTBT singlets)
t
1
= 1 ps (66%)
t
2
= 17 ps (34%)
t
3
> 100 ps (0%)
t
1
= 0.7 ps (25%)
t
2
= 45 ps (73%)
t
3
> 100 ps (2%)
!

6. The 500 nm region contains overlapping features including a PIA signal from PCPDTBT
singlet excitons
The PCPDBT ground state bleach overlaps with a photo-induced absorption (PIA) band.
The PIA band is detected as a positive contribution around 500 nm and through the deformation
of the ground state bleach (the high energy polymer absorption band, located around 410 nm in
the steady state absorption, is almost absent in the transient absorption spectra). Because this
PIA feature is present in spectra of PCPDTBT solutions in ODCB (Figure S6), we attribute the
band to high-energy singlet transitions (S
n’
 S
1
).
2

95
The TA data have been normalized by the excitation densities in each sample to eliminate
the slight differences in absorbance of the neat component films and of the hybrid films
due to thickness and morphology differences; the raw data (signals in mOD) is shown in
Figure 4.8 for completeness. Here, the spectra of the neat CdSe(tBT) QD film and the
spectra of the neat PCPDTBT film serve as control experiments to disentangle transient
species present in the hybrid data.

Figure 4.8. Transient absorption spectra of a neat PCPDTBT film (15-20 nm) and of a hybrid
PCPDTBT:CdSe(tBT) film (1:8 wt/wt polymer/QDs, 45-55 nm) at different time delays between
the 800 nm pump and the probe pulses. Both samples were measured with the same pump fluence
and under identical conditions, but show slight differences in their absorbance. Figure 4.6 shows
the same data where the absorbance difference has been normalized out.

First, Figure 4.6a shows the absence of signal from the neat CdSe(tBT) QD film when it
is excited at 800 nm, ruling out two-photon absorption and/or trap-state absorption as
! S6!
!
Figure S3. Transient absorption spectra of a neat PCPDTBT film (15-20 nm) and of a hybrid
PCPDTBT:CdSe(tBT) film (1:8 wt/wt polymer/QDs, 45-55 nm) at different time delays between
the 800 nm pump and the probe pulses. Both samples were measured with the same pump
fluence and under identical conditions, but show slight differences in their absorbance. Figure 4
of the main paper shows the same data with the absorbance difference has been normalized out.
It is interesting to note that the ground state bleach amplitude is larger in the case of the
hybrid than in the case of the polymer in Figure 4 of the main text. This difference does not arise
from the absorbance correction, as shown in Figure S3. Both samples were measured under
identical conditions. This effect likely takes its origin in the photo-induced absorption bands that
overlap spectrally with the ground state bleach, modifying the shape and amplitude of the ground
state bleach in different ways in the two samples. Figure S4b shows that the bleaches in the two
samples exhibit slightly different shapes (for example for wavelengths above 750 nm and below
650 nm in Figure S4c and Figure S4d, as discussed in the main text and in paragraph 6 of the
Supporting Information), suggesting that the overlapping photo-induced absorption bands
∆A (mOD) ∆A (mOD)
Wavelength (nm)
PCPDTBT
Hybrid
4
2
0
-2
1400 1200 1000 800 600 400
1 ps
3 ps
10 ps
60-100 ps
4
2
0
-2
1 ps
3 ps
10 ps
60-100 ps

96
possible QD excitation pathways at this excitation wavelength. This stands in contrast to
transient spectra obtained when a 550 nm pump is used that directly excites the
CdSe(tBT) QD film. The latter experiment reveals bleaches of the overlapping 1S
e
−1S
3/2

and 1S
e
−2S
3/2
transitions of the QDs around 610 nm and of the 1P
e
−1P
3/2
transition
around 500 nm.
26,27

Second, PCPDTBT singlet excitons can be monitored directly in the near-infrared (NIR)
region through the induced absorption band around 1400−1480 nm (brown shading in
Figure 4.6a), which is attributed to transitions from the lowest lying singlet excited state
to higher lying singlet states (S
n
← S
1
).
31
The 1250 nm PIA signal of neat PCPDTBT (red
shading in Figure 4.6a) has been previously associated with dissociated/delocalized
polarons
31,32
(labeled “polarons I”). This signal overlaps with the broad singlet exciton
PIA around 1400 nm, but the polarons I band can be detected at long times (≳100 ps)
when the singlet band disappears.
31
The positive feature in the 855−885 nm range (green
shading) has also been observed in absorption spectra of chemically oxidized
PCPDTBT
33,34
and is therefore also attributed to polarons. In our TA measurements, this
feature shows a significant long-lived component (lifetime >500 ps, Figure 4.7) that
distinguishes it from the polaron I band. It is therefore likely related to a different polaron
species, labeled “polarons II”. Polarons II may correspond to trapped/localized
polarons,
35,36
possibly caused by morphological defects.

97
4.3.3. Transient Absorption Spectra of Hybrid Films
For hybrid films, the TA measurements provide clear evidence that electron transfer
occurs from PCPDTBT to the CdSe(tBT) QDs upon excitation at 800 nm. The hybrid
film exhibits an additional negative peak around 610 nm when compared to the neat
polymer film (violet shading in Figure 4.6a). This peak corresponds to a bleach of the
CdSe(tBT) QD 1S
e
−1S
3/2
and 1S
e
−2S
3/2
excitonic transitions (Figure 4.1a and Figure
4.6a). As noted before, and as shown on the inset of Figure 4.1b, this signal can only arise
from electron transfer from the excited polymer to the QDs; both energy transfer from the
exciton state of PCPDTBT to the exciton state of CdSe(tBT) QDs and hole transfer from
PCPDTBT to the QDs are uphill processes. Moreover, the 610 nm bleach arises very
rapidly within the time resolution of our ultrafast measurement, namely 65 fs (Figure
4.6b).
In the 900−1050 nm (orange shaded) region, the hybrid films exhibit a long-lived
component that is fully absent in the neat polymer. This long-lived PIA is therefore
related to the presence of QDs in the blend. It could arise from electronic transitions in
the reduced CdSe(tBT) QDs
37,38
after charge transfer or from transitions from states akin
to CT states that involve positive polarons on the polymer chains and electrons on the
QDs. Although there is a small signal in this 900−1050 nm region in the neat polymer
films, its decay dynamics are similar to that of the singlet and polarons bands (see Figure
4.9).

98

Figure 4.9. Transient absorption decays of the signals measured in a neat PCPDTBT film
(spectra in Figure 4.6).

Therefore, this neat polymer film signal can be distinguished from the large and long-
lived signal observed in the hybrid films in the orange shaded region. In addition, both
the prompt polymer bleach (observed at 600−800 nm) and the prompt polaron II signal
(green shading) are enhanced in the hybrid, suggesting a higher yield of charge carriers
by a factor 2−3 at 1 ps, comparable to the yield enhancement in PDTPQx-HD:PbS hybrid
hetero junctions determined by steady-state PIA.
39
Finally, the PCPDTBT singlet exciton
S
n
← S
1
transition(s) amplitude is partially quenched by ∼33% at initial times (75 fs) in
! S9!

Figure S5. Transient absorption decays of the signals measured in a neat PCPDTBT film
(spectra in Figure 4 of the main text).
!
5. Multi-exponential fits of the decays dynamics of the various excited species
The results of analysis of the decay of the excited species populations are summarized in
the following table for the neat PCPDTBT and hybrid films, obtained using multi-exponential fits
.
!
1.0
0.5
0.0
A (norm)

Neat PCPDTBT, normalized traces
900 - 1050 nm
1250 - 1300 nm (polarons I)
1.0
0.5
0.0
A (norm)
0.1 1 10 100
Time (ps)

Neat PCPDTBT, normalized traces
900 - 1050 nm
1400 - 1480 nm (singlets)
1.0
0.5
0.0
A (norm)

Neat PCPDTBT, normalized traces
855 - 885 nm (polarons II)
900 - 1050 nm
!
A
i
e
"(t"t
0
)/#
i
i
$

99
the hybrid, as compared to the neat polymer film (Figure 4.7). This can result from (i) a
reduction of the initial S
1
exciton population caused by electron transfer to the QDs,
and/or (ii) a modified oscillator strength of the S
n
← S
1
transitions caused by packing
modifications of the polymer, and/or (iii) overlapping bands from various excited species.

4.3.4 Accessible QD Electron Energy Levels by Injection from the PCPDTBT S
1

State
The TA experiments presented in this paper lead to a clearer understanding of the
electronic couplings at play in the PCPDTBT:CdSe(tBT) QD hybrid system. First, the
spectral information given by the TA data set helps discriminate between filled electron
levels in the CdSe(tBT) QDs. The bleached 1S
e
−1S
3/2
and 1S
e
−2S
3/2
transitions indicate
that the 1S
e
level of the QD is partly filled upon electron transfer from the polymer donor.
On the other hand, due to the spectral overlap in the 500 nm region with a singlet state
absorption of the polymer (blue shading in Figure 4.6 and 4.10), it is difficult to confirm
filling of the 1P
e
electron level from TA experiments; however, the energy level
alignment (Figure 4.1b) gives information on the filling of the 1P
e
level.

100

Figure 4.10. Transient absorption spectra of a PCPDTBT in ODCB solution at 1.6 × 10
-3
mg
mL
-1

In CdSe QDs, most of the energy difference between the first (1S
e
−1S
h,1/2
) and second
(1P
e
−1P
h,1/2
) excitonic transitions derives from the energy level difference in the
conduction band due to the near degeneracy of the valence band.
40
This excitonic
transition energy difference is on the order of 400 meV for the 4.5 nm CdSe QDs used
here, which implies that the 1P
e
level of the 4.5 nm CdSe(tBT) QD probably lies
shallower in energy than the ionization potential of the PCPDTBT excited state,
IP*
PCPDTBT
(vide infra). Therefore, it is likely that electron transfer from PCPDTBT
excited at 800 nm to the 1P
e
level of CdSe QDs is energetically unfavorable.

! S11!

Figure S6. Transient absorption spectra of a PCPDTBT in ODCB solution at 1.6 × 10
‑
3
mg mL
‑
1

excited with an 800 nm pump. The pump fluence has been adjusted to get a number of absorbed
photons similar to that absorbed in the films, i.e. 1.05 × 10
14
cm
–2
. The data points are given as
circular symbols, the full line is smoothed data.

7. Deconvoluting the spectral contributions of the polymer bleach and the QD bleach
We use Bayes’ theorem to deconvolute the polymer and QD bleaches with a linear
combination model.
3
This approach gives access to statistically based error bars for the
deconvolution process.
We used a doorway function for the prior probability density function on model
parameters a and b (0,0)<(a,b)<(10,5): our only a priori is that the contributions of the neat QD
and neat polymer bleaches cannot be negative and cannot diverge. Because the ground state
bleach of the polymer is enhanced in the hybrid relative to that in the neat film, b can be greater
than one. The likelihood is given by
!
exp "
1
2
d
obs
"d
pred
( )
2
e
2
#
$
%
%
&
'
(
(
, where e is the experimental
uncertainty and d
obs
and d
pred
are defined as follows:
-1.0
-0.8
-0.6
-0.4
-0.2
0.0
0.2
A (mOD)
600 500 400
Wavelength (nm)
PCPDTBT in ODCB solution
1 ps after the 800 nm pump

101
4.3.5 Evaluation of the Electronic Coupling Between PCPDTBT and CdSe(tBT)
QDs
The ultrafast electron transfer time t
et
≲ 65 fs (k
et
≳ 1.5 × 10
13
s
−1
) shows that the
PCPDTBT S
1
excited state and the QDs 1S
e
electron state are well coupled. Under the
assumption of a three-level scheme
41
(i.e., ground state, polymer excited state, and
electron-transferred state), the electronic coupling J between the S
1
excitonic state of
PCPDTBT and that of the CdSe(tBT) QDs can be evaluated from Marcus theory: k
et
=
(t
et
)
−1
= 2πJ
2
/ħ(4πλk
BT
)
1/2
exp(−(ΔG + λ)
2
/4λk
BT
), where J is the electronic coupling
matrix element, ħ is the reduced Planck constant, λ is the reorganization energy, k
B
is the
Boltzmann constant, T is the temperature, and ΔG is the driving force for the electron
transfer reaction. The driving force ΔG cannot directly be inferred from the electron
affinities measured by electrochemistry and pictured in Figure 4.1b because this would
neglect the exciton binding energy. We therefore use
42
ΔG ∼ EA
QD
− IP*
PCPDTBT
, where
EA
QD
is the adiabatic electron affinity of the QD and IP*
PCPDTBT
is the adiabatic
ionization potential of the excited PCPDTBT donor. Approximating IP*
PCPDTBT
as
IP
PCPDTBT
− E
exciton
, and with the electrochemically determined energy levels for
IP
PCPDTBT
and EA
QD
, we find ΔG ∼ −100 meV. The reorganization energies for charge
transfer from a conjugated polymer to a QD are poorly documented at this stage, but the
minimal value of the electronic coupling J can be obtained using the optimal
reorganization energy value (i.e., suchthat λ + ΔG ∼ 0meV), which yields J ≳ 17meV.
This value is comparable to the optimal coupling matrix element obtained for electron
transfer from PbS QDs to PCBM molecules derived with the same hypothesis.
12
The

102
measured electron transfer time is also consistent with the measured electron injection
times in dye-sensitized solar cells
43
and with calculated electron injection times of 90 fs
from dodecathiophene chains physisorbed to Si nanocrystals.
44
To date, no signature of
an optically bright CT state has been unambiguously detected in this hybrid system,
supporting the assumption of a three-level scheme.
Although electron transfer from the S
1
state of PCPDTBT and the 1S
e
state of CdSe QDs
is “ultrafast” in our hybrid system (i.e., below the 65 fs resolution of our setup), it does
not directly imply that it always outcompetes all other processes, leading to 100% of the
excitons dissociated at donor/acceptor interfaces, as is commonly assumed.
45,46
Indeed, in
highly disordered materials such as hybrid and organic BHJs, the interface is a
heterogeneous ensemble that exists as a distribution of donor−acceptor conformations
and couplings. Moreover, the distribution of couplings is likely widened in hybrid
materials compared to PCBM-based BHJs as a result of the intrinsically heterogeneous
nature of QD surfaces (i.e., variety of crystalline facets, ligand configurations, and
polydispersity of QD diameters). In these experiments, we therefore likely observe a
subset of donor/acceptor interface conformations, namely those exhibiting the strongest
couplings. It is possible that additional, slower electron transfer processes take place and
that these slower charge separation events are hidden by the fast decay of the rapidly
charge-transferred pairs. Disorder, via the most unfavorable conformations, or
equivalently the slowest electron transfer processes, can reduce the absolute yield of
charge transfer below 100%.

103
4.3.6 Prompt Yield for Electron Transfer and Model for the Electron Population
Time Evolution
The QD spectral bleach provides a way to quantify the absolute yield of charge transfer at
the donor/acceptor interface. This task is generally difficult in all-organic BHJs because
the numerous PIA signatures of excited states in the NIR overlap strongly, making
quantitative analysis difficult. Here, the 610 nm signal arises from CdSe(tBT) QD
reduction only, so that its amplitude reflects the number of electrons located on the QD at
each delay time of the TA experiment. In order to eliminate the polymer bleach
contribution in this region, the hybrid TA spectra can be decomposed at each time by a
linear combination of the CdSe(tBT) QD and of the neat PCPDTBT TA spectra. For each
time step, the observed data d
obs
(corresponding to the hybrid spectrum) can be compared
to the data d
pred
predicted by a linear combination model with parameters (a,b):
d
obs
= ΔA
hybrid

d
pred
= a × ΔA
polymer
+ b × ΔA
QD
(t = 75 fs) (1)
where ΔA
i
is the TA signal of sample i (Figure 4.6a). The comparison between d
obs
and
d
pred
is made for a large number of (a,b) parameter couples. The most likely model
parameter couple (a
mean
, b
mean
) is extracted using Bayes’ theorem and a Metropolis
algorithm.
47
It is possible to fit the data, within experimental uncertainties, with various
(a,b) couples; therefore, the standard deviations of the parameters (a
s.d
, b
s.d.
) reflect the
dispersion of (a,b) couples that fit the experimental data. Simply put, this procedure gives
access to rigorous, statistically relevant error bars (see the Experimental, section 4.4).

104
In this decomposition, the parameter b represents the contribution of the QD electrons to
the hybrid spectrum. The parameter b can be converted to the density of electrons, N
e
,
residing on the QDs:
N
e
= b × I
0, QDs
× F
a, neat QDs
× F
1Se
(2)
where I
0,QDs
, F
a,neatQD
, and F
1Se
are the experimental parameters used for the control
experiment on the neat QD film: I
0,QDs
∼ 5 × 10
14
cm
−2
is the density of 550 nm photons
incident on the neat QD film, F
a
,
neat QD
∼ 0.14 is the fraction of 550 nm photons absorbed
by the neat QD film, and F
1Se
∼ 0.68 is the fraction of absorbed photons involving the 1S
e

state of the QD conduction band (see the Experimental section, 4.4). In other words, the
610 nm bleach magnitude in the neat QD film, obtained after the generation of I
0,QDs
×
F
a,neat QD
× F
1Se
∼ 5 × 10
13
cm
−2
electrons on the 1S
e
energy level is used to deduce the
density of electrons on the QDs in the hybrid film from the 610 nm bleach magnitude.
The population density of electrons on the conduction band of the QDs as a function of
time is represented in Figure 4.11.

105

Figure 4.11. Density of electrons located on the QD conduction band as a function of time. This
curve is obtained from the deconvolution of the 610 nm TA bleach, with corresponding statistical
error bars (brown bars). A solution to the electron−hole pair diffusion problem, under mutual
Coulomb attraction, is shown with the black line and described in the text.

Overall, our deconvolution model assumes that (i) each absorbed photon yields one
electron−hole pair in a neat QD film or one polymer exciton in a hybrid film, (ii) an extra
electron on a QD results in the same magnitude of bleach (in mOD) as an optically
excited electron−hole pair,
48
and (iii) the transient spectra measured in the hybrid films
can be fitted by a linear combination of the TA spectra measured in the neat QD film and
in the neat polymer film in the bleach region (i.e., the amplitude of the PIA features in the
spectra of the hybrid are proportional to the PIA features in the neat component films).
Having gained access to the QD electron density as a function of time, we now (i) deduce
the prompt yield of electron transfer and (ii) analyze the temporal evolution of the QD
electron density with a physically relevant model that describes diffusion of
electron−hole pairs, taking account of their mutual Coulomb attraction. Thereby, we
avoid the more commonly used (but less physical) fit to multiexponential decays.

106
i. Prompt Yield of Electron Transfer. The yield of electron transfer is obtained by
normalizing the initial electron density (i.e., the electron density extracted at 75 fs) by the
exciton density initially present on the PCPDTBT chains: η
CT
∼ N
e
/ (I
0
F
a
), where I
0
∼ 7.3
× 10
14
cm
−2
is the density of 800 nm photons incident on the hybrid film, F
a
∼ 0.09 is the
fraction of photons absorbed by the hybrid film. We obtain η
CT
∼ 82 ± 5%. This number
is in general agreement with the charge transfer yield obtained from photoluminescence
data.
ii. Temporal Evolution of the QD Electron Density. The QD electron density exhibits a
rapid temporal decay that we attribute to recombination with the PCPDTBT polarons.
Indeed, the decay observed in the PCPDTBT:CdSe(tBT) QDs hybrids is greatly
accelerated compared to the decay observed in neat films of CdSe QDs (the decays are
compared in Figure 4.12).

Figure 4.12. Time evolution of the QD electron population in the hybrid film compared to the
time evolution of the 610 nm ground state bleach in the neat QD films, capped with either native
ligands (black circles) and tBT (red symbols), after 550 nm excitation.

! S14!
state and therefore ground state bleach recovery. The ground state bleach recovery pictured in
red in Figure S7 is therefore representative of the dynamics of the excited, localized hole states.
c) In the PCPDTBT:CdSe(tBT) QD hybrid, the bleach at 600 nm probes only the electron
dynamics in the context of the QDs, as argued in the text. The decay of the electron population
in the hybrid is greatly accelerated compared to the ground state bleach in neat QD films, Figure
S7. This means that the accelerated decay observed in the hybrid film is not related to localized
excited states that already exist in the neat QD films. Assuming that the blending with the
polymer does not induce electron traps, we therefore attribute the rapid decay of the electron
density in the hybrids to recombination with the PCPDTBT polarons.

Figure S7. Time evolution of the QD electron population in the hybrid film compared to the
time evolution of the 610 nm ground state bleach in the neat QD films, capped with either native
ligands (black circles) and tBT (red symbols), after 550 nm excitation.

10. The rapid decay of the electron population is due to geminate recombination
In Figure S8, we show that the electron decay time does not vary with the pump fluence,
confirming monomolecular recombination pathway. As for Figure 6, electron density decays
were obtained from the deconvolution of the 610 nm TA bleach in two contributions, that of the
60x10
12
50
40
30
20
10
0
Density of electrons
on the QDs (cm
-2
)
100 80 60 40 20 0
Time in ps
neat CdSe(NL) film, λ
exc
= 550 nm, scaled
neat CdSe(tBT), λ
exc
= 550 nm, scaled
electron density in PCPDTBT:CdSe(tBT) hybrids

107
This observation rules out the hypothesis that the rapid dynamics observed in hybrids are
related to localized excited states (traps) that exist at the surface of the QDs, which would
result in longer ground state bleach recovery. We further assume that the blending with
the polymer does not introduce localized electron states at the QD surface.
A simple physical model for the recombination of CdSe(tBT) QD electrons and
PCPDTBT polarons is that of charges diffusing under a spherically symmetric potential
U:
49

!"
!"
= ∇ ∙ ∇+
!
!
!
!
∇ + −
!
−
!
(3)
where ρ is the density of charges. = k
B
T(r
c
/|r| ⃗) + E ⃗·r ⃗ is the potential, where the first
term represents the Coulomb attraction between charges via the Onsager radius r
c
= e
2
/
4πε
0
ε
r
k
B
T and where the second term is the potential from an external electric field E ⃗. D
is the diffusion coefficient and is homogeneous in our case, with D = (k
B
T)/q(µ
e
+ µ
n
). In
this model, recombination between QD electrons and PCPDTBT polarons is geminate. In
effect, the decay of the electron population in the hybrid film does not depend on the
fluence (see Figure 4.13), providing evidence for geminate recombination. If bimolecular
recombination dominated, the electron population decay would be slower for lower pump
fluences.

108

Figure 4.13. Normalized electron population decay as a function of time for different pump
fluences. The top panel is equivalent to the data in Figure 4.11; the black line is the result of our
diffusion model.

! S16!
!
1
0
Electron population (normalized)
0.1 1 10 100
Time (ps)
32 µJ/cm
2
63 µJ/cm
2
150 µJ/cm
2
180 µJ/cm
2
1
0
1
0
1
0

109
The solution of this equation in the long-time limit is the Braun−Onsager model,
50,51

commonly used to describe the dynamics of charge carrier population at longer time
scales.
45,52
Here, the external electric field E ⃗ is null (to match the conditions of the TA
experiment), and the force felt by the charges is solely from their Coulomb attraction. We
use a differential equation solver
53
to obtain the time-dependent solution of eq. 3 (i.e., the
electron density as a function of time). The specific solution presented in black in Figure
6 has been obtained for a diffusion coefficient D = 5 × 10
−5
cm
2
s
−1
, calculated from
measured mobility values for electrons and holes in hybrid blends made of P3HT and
CdSe QDs.
28
We find that diffusive models that match our data require that most of the
charges are initially located within their Coulomb attraction radius (see Figure 4.14).

110

Figure 4.14. a) Different solutions obtained with different values of the Onsager radius r
c
, initial
charge separation distribution and diffusion coefficient D; b) for the parameters used for the black
line solution (Figure 4.11, and 4.14a), the spatial profiles of the Coulomb potential and of the pair
survival probability density for various time after the generation event at time 0.

This means that most electrons and polarons are spatially located very close to one
another and charge separation suffers from their mutual Coulomb attraction. Other charge
pairs progressively diffuse away from each other and escape their Coulomb attraction. In
particular, the TA measurements show that the yield of electrons decreases from 82 ± 5%
at initial times to 24 ± 3% after 1 ps, 10 ± 2% after 10 ps, and 4.5 ± 1% after 500 ps. It is
important to point out here that these yield values do not depend on the assumptions
made in the diffusion limited recombination model. As noted above, the yield of
! S19!
!
Figure S9. a) Different solutions obtained with different values of the Onsager radius r
c
, initial
charge separation distribution and diffusion coefficient D; b) for the parameters used for the
black line solution (main text, Figure 6, and Figure S9a), the spatial profiles of the Coulomb
potential and of the pair survival probability density for various time after the generation event at
time 0.

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
1
Kovalenko, S. A.; Dobryakov, A. L.; Ruthmann, J.; Ernsting, N. P. Phys. Rev. A, 1999, 59,
2369-2384.
2
Grancini, G.; Martino, N.; Antognazza, M. R.; Celebrano, M.; Egelhaaf, H-J.; Lanzani, G. J.
Phys. Chem. C, 2012, 116, 9838-9844.
1
0
Normalized electron population
on the QDs
0.01 1 100
Time in ps
10
-6
10
-4
10
-2
Pair survival probability density
4 6 8
10
2 4 6 8
100
2 4 6 8
1000
Distance (A)
-1.6
-1.2
-0.8
-0.4
0.0
Coulomb potential (k
B
T)
Black line model
Pair survival probability density
t = 0 ns
t = 10
-5
ns
t=10
-4
ns
t=10
-3
ns
t=10
-2
ns
t=10
-1
ns

Coulomb potential
Onsager radius, r
c
= 5 A
o
L, initial distribution Gaussian characteristic length in A 1 3.4 20
r
c
, Onsager radius in A (effective dielectric constant ε
r
) 5 (112) 15 (37) 100 (5.6)
o
o
D, diffusion coefficient in cm
2
/s 5x10
-5
10
-3
5x10
-2
a)
b)
o

111
electrons at 75 fs is referred to as the charge transfer yield or exciton dissociation
efficiency and is η
CT
∼ 82 ± 5%. The fraction of these dissociated excitons that remains
after 1 ps (i.e., after the fast recombination) is called charge separation efficiency: η
sep
∼
(24 ± 3%)/(82 ± 5%) = 29.5 ± 4.5%.

4.3.7. Comparison of Yields Obtained from the Transient Absorption
Measurements to Solar Cell Performance
These yields for electron density formation and decay obtained by TA experiments can
be compared to the internal quantum efficiency at 800 nm (where IQE is equal to the
EQE divided by the fraction of incident photons absorbed in the solar cell). While TA
reports on the exciton dissociation efficiency and recombination, the IQE also probes the
charge transport and collection efficiencies and therefore should logically be smaller than
the TA transfer yield. Let us consider a solar cell with an active layer identical to the
hybrid film studied by TA spectroscopy (ITO/PEDOT:PSS/PCPDTBT:CdSe(tBT)/
ZnO/Al). It possesses an EQE at 800 nm of 3.7%. For a 2-pass absorption process, the
fraction of absorbed photons in a similar active layer deposited on quartz is η
abs
∼ 17.5%
(equal to 2 × (1 − %T), where %T is the transmission of the film at 800 nm), resulting in
an IQE value at 800 nm of approximately 20%. This suggests that around 80% of the
charge carriers remaining at 1 ps (i.e., after the fast recombination time range) are
extracted and yield electrical current in a solar cell device under the influence of the
electric field induced by the difference in the electrode work functions. Altogether, we

112
have quantified the efficiencies of the following steps in our materials for 800 nm
illumination. The external quantum efficiency can be written as EQE = η
abs
× η
CT
× η
sep
×
η
coll
with η
abs
∼ 17.5% the light absorption efficiency (2-pass absorption), η
CT
∼ 82 ± 5%
the exciton dissociation efficiency (from TA), and η
sep
∼ 29.5 ± 4.5% the charge
separation efficiency (from TA), so that η
coll
∼ 80 ± 20% the charge collection efficiency.
The absorption and charge separation efficiencies are here the limiting process in our
hybrid solar cell. It is important to mention that the fraction of absorbed photons in a
solar cell configuration comprising electrodes and transport layers (rather than on quartz)
will be different from the one cited above.
54,55
In effect, it has been shown that the
multiple reflections at the various interfaces of the solar cell stack can result in a
redistribution of the optical electric field in the active layer, thereby modifying its
absorption efficiency. We are currently investigating the magnitude of this effect in our
solar cells, and our preliminary results suggest than the fraction of absorbed photons by
the active layer in a solar cell configuration is 24% at 800 nm. This results in a charge
collection efficiency η
coll
∼ 66 ± 14%.

4.4. Experimental
4.4.1. Sample Preparation. CdSe quantum dot synthesis and tBT ligand exchange were
performed as described in ref. 4. PCPDTBT was used as received from 1-Material, with a
molecular weight and a polydispersity index of 28.5 kDa and 1.9, respectively.
Photovoltaic devices were fabricated and tested in air as described in ref. 2 and in section

113
4.4.4. Device active areas were 4.4 mm
2
, and active layers were ∼45−55 nm thick. For
spectroscopic studies, hybrid active layers were deposited through a similar spin-casting
procedure onto quartz slides. Their thicknesses were adjusted to obtain similar optical
densities at 800 nm (i.e., neat PCPDTBT, 15−20 nm; hybrids, 45−55 nm).

4.4.2. Steady-State Spectroscopies
Thin film absolute absorbances were measured using an integrating sphere. Chopped and
filtered monochromatic light (250 Hz, 10 nm fwhm) from a Cornerstone 260 1/4 m
double grating monochromator (Newport 74125) was used in conjunction with an EG&G
7220 lock-in amplifier to perform spectral responsivity (or EQE) measurements. Thin
film photoluminescence spectra were collected on a Horiba Jobin Yvon Nanolog spectro-
fluorometer equipped with a 450W xenon short-arc excitation source and an IGA-020
InGaAs photodiode detector using 0.5 s integrating time and 3 scan averages, and then
they were corrected for the InGaAs detector response function. The photoluminescence
intensity is normalized by the absorbance of the film at 780 nm to account for slight
differences in film thicknesses.

4.4.3. Femtosecond Transient Absorption (TA) Spectroscopy
TA measurements were carried out using the output of a Coherent Legend Ti:sapphire
amplifier (1 kHz, 3.5 mJ, 35 fs), as described in detail elsewhere.
22,23
The amplifier

114
output was used as the 800 nm pump pulse. White light supercontinuum probe pulses
were polarized perpendicular to the pump pulse and generated using a rotating CaF
2

window or a sapphire window depending on the probed wavelength range (see section
4.4.7). The probe pulse was dispersed by an Oriel MS1271 spectrograph onto a 256 pixel
Si or InGaAs photodiode array depending on the wavelength range probed. To obtain
satisfactory signal-to-noise on the optically thin samples, transient spectra were measured
for typical pump fluences of 180 µJ cm
−2
. Samples were translated perpendicular to the
path of the pump and probe to prevent photodamage.

4.4.4. Preparation of Hybrid Thin Films
Patterned ITO-coated glass substrates (10 Ω cm
–2
, Thin Film Devices, Inc.) were
sequentially cleaned by sonication in tetrachloroethylene, acetone, and isopropyl alcohol
followed by 30 min of UV-ozone treatment. A layer of PEDOT:PSS (Clevios PH 500,
filtered through a 0.45 µm PTFE syringe filter) was spun-cast onto the cleaned ITO and
heated at 120 °C for 30 min under vacuum (0.01 mmHg). Solutions of 15 mg mL
–1

PCPDTBT were prepared in dichlorobenzene, and were mixed with dichlorobenzene
dispersions of CdSe to final concentrations of 22.5 mg mL
–1
(2.5 mg mL
–1
PCPDTBT, 20
mg mL
–1
CdSe), and were stirred for 2-4 h in the dark prior to use. CdSe(tBT) dispersions
required 5% tetramethylurea as a cosolvent to prevent agglomeration. Active layers
passed through 0.45 µm PTFE syringe filters were spun-cast in air onto annealed
PEDOT:PSS layers (900 rpm, 45 s) forming films with thicknesses of ~45-55 nm.
Following spin-casting, active layer films were dried under a nitrogen atmosphere for 20

115
min. ZnO nanoparticles were deposited by spin-casting to form a 20-30 nm film on top of
the bulk heterojunction layer. Following the ZnO deposition, films were dried under a
nitrogen atmosphere for 20 min. Dry devices were annealed at 160 °C for 10 min under
nitrogen atmosphere. Finally, samples were loaded into a high vacuum (∼2 µTorr)
thermal deposition chamber (Angstrom Engineering) for Al deposition (Al; Alfa Aesar,
99.999%) at a rate of 2 Å s
–1
. For spectroscopic studies, hybrid blends were deposited
through a similar spin-casting procedure onto clean quartz slides (no PEDOT:PSS, ZnO,
nor upper Al electrode), and encapsulated by a second quartz side glued with epoxy in a
nitrogen-filled glove bag.

4.4.5. Steady-State Optical Characterization of Hybrid Thin Films Containing
Various QD Quantities
The as-prepared hybrid thin films are studied by optical steady-state methods. The hybrid
film morphology was characterized by looking at the absorbance of thin films with
various QD fractions (Figure 4.2). The data shows a blue shift of the PCPDTBT peak
absorbance with increasing CdSe(tBT) QD fraction, while maintaining a contribution to
the absorbance above 800 nm. This suggests that in the hybrid films, part of the polymer
chains are packed similarly to the neat polymer film, whereas another fraction of the
chains is in a glassy (disordered) phase.
The legend indicates the weight and volume fraction of the CdSe(tBT) QDs in the hybrid
films. The weight fractions are those used when blending the components in solution, for

116
which the CdSe(tBT) QD weight concentrations were corrected for the ligand
contributions (typically 5% for 4.5 nm tBT-exchanged CdSe QDs). The volume fractions
are calculated from the weight fraction for densities of 5.8 g cm
–3
for CdSe and 1 g cm
–3

for PCPDTBT.

4.4.6. Photovoltaic Device Testing
The dark and one-sun illumination I-V characteristics of a photovoltaic device, whose
active layer is identical to that study by TA, are given in Figure 4.3. Current density
dependence on applied test voltage measurements were performed under ambient
conditions using a Keithley 2400 SourceMeter in the dark and under ASTM G173-03
spectral mismatch corrected 1000 W m
–2
white light illumination from an AM 1.5G
filtered 300 W xenon arc lamp (Newport Oriel). The resulting device parameters were J
SC

= 7.25 mA cm
–2
, V
OC
= 0.76 V, and fill factor of 0.38.

4.4.7. Ultrafast Transient Absorption Spectroscopy
White light supercontinuum probe pulses were polarized perpendicular to the pump pulse
and generated in different ways depending on the probed wavelength range. The 380-620
nm and 580-890 nm broadband continua were generated using a rotating, 2 mm thick
CaF
2
window seeded with 800 nm and 1315 nm pulses respectively and detected with a
Si Hamamatsu S3901- 256Q photodiode array. The 850-1150 nm and 1000-1480 nm

117
broadband continua were generated in a sapphire window, seeded with 800 nm and 1315
nm pulses respectively and detected with an InGaAs Hamamatsu G9213-256S
photodiode array. The 1315 nm WL seed was generated using ca. 10% of the amplifier’s
output to seed a type II Spectra Physics OPA-800C. To minimize probe dispersion, a pair
of off-axis aluminum parabolic mirrors was used to collimate the probe and focus it into
the sample, while a CaF
2
lens focused the pump. Spectra were measured for a range of
pump fluences from 30 to 2400 µJ cm
–2
, with a typical fluence of 180 µJ cm
–2
for the 800
nm pump.
Spectra were acquired by averaging each time slice over 250 laser shots, and delay traces
were scanned 5 to 10 times depending on the noise. Once acquired, data were corrected
for accidental phase flips and averaged over the different time delay scans. Data obtained
in different wavelength ranges were stitched together to obtain the spectra shown in the
main text. In order to account for the relative absorbances of the different samples, the
spectra presented in the main text are given in ΔA/(I
0
× F
a
), where ΔA is the transient
absorption signal, I
0
is the incident pump photon density and F
a
is the fraction of photons
absorbed by a sample.
In the main text, we show the data normalized by the number of absorbed photons in
order to account for slight variations in film thickness and morphologies. The transient
absorption spectra for a neat PCPDTBT film and a PCPDTBT:CdSe(tBT) 1:8 wt/wt
polymer/QDs film, presented in Figure 4.6 of the main text is shown here with the signal
in mOD.

118
It is interesting to note that the ground state bleach amplitude is larger in the case of the
hybrid than in the case of the polymer in Figure 4.6 of the main text. This difference does
not arise from the absorbance correction, as shown in Figure 4.8. Both samples were
measured under identical conditions. This effect likely takes its origin in the photo-
induced absorption bands that overlap spectrally with the ground state bleach, modifying
the shape and amplitude of the ground state bleach in different ways in the two samples.
Figure 4.15b shows that the bleaches in the two samples exhibit slightly different shapes
(for example for wavelengths above 750 nm and below 650 nm in Figure 4.15c and
Figure 4.15d, as previously discussed), suggesting that the overlapping photo-induced
absorption bands presumably have different amplitudes in the neat PCPDTBT film and in
the hybrid film.

Figure 4.15. Comparison between the ground state bleach observed in transient absorption 75 fs
after the pump pulse in the neat polymer film and in the hybrid film: the raw data in mOD is
shown in (a); the normalized data is shown in (b). Comparison between the ground state bleach
and the inverted steady-state absorption spectrum: in the neat PCPDTBT film (c) and in the
hybrid film (d).
! S7!
presumably have different amplitudes in the neat PCPDTBT film and in the hybrid film. In
addition, Figure S4c and Figure S4d show that the spectral mismatches between the inverted
steady-state absorption and the ground state bleach in the TA experiment are different for the
neat PCPDTBT and the hybrid samples, which demonstrates different overlapping photo-induced
absorption bands.

Figure S4. Comparison between the ground state bleach observed in transient absorption
75 fs after the pump pulse in the neat polymer film and in the hybrid film: the raw data in
mOD is shown in (a); the normalized data is shown in (b). Comparison between the ground
state bleach and the inverted steady-state absorption spectrum: in the neat PCPDTBT film
(c) and in the hybrid film (d).

Absolute absorbance (inverted)
-1.0
-0.5
0.0
0.5
TA signal at 75 fs (normalized)
PCPDTBT
800 700 600 500
Wavelength (nm)
-1.0
-0.5
0.0
0.5
Hybrid
800 700 600 500
-1.0
-0.5
0.0
0.5
-4
-2
0
2
TA signal at 75 fs (normalized)
TA signal at 75 fs
Wavelength (nm)
a)
b)
c)
d)
-80x10
-3
-40
0
40
-80x10
-3
-40
0
40
PCPDTBT
Hybrid

119
In addition, Figure 4.15c and Figure 4.15d show that the spectral mismatches between the
inverted steady-state absorption and the ground state bleach in the TA experiment are
different for the neat PCPDTBT and the hybrid samples, which demonstrates different
overlapping photo-induced absorption bands.
The time resolution of the system, i.e., the pump pulse duration τ
pump
, was found to be 65
fs based on cross-correlation between the pump and probe in two 1 mm quartz substrates
(similar to encapsulated samples). The cross-correlation response was fitted to the
following equation
58
:

is the spectral component of the probe, t(ω
probe
) is the frequency-dependent pump-probe
time delay, τ
pump
and τ
probe
are the pump and probe pulse durations, β is the chirp rate and
t0(ω
probe
) is the time-zero function, which is fitted at the peak of the cross-phase
modulation signals.
We observe a strong and long-lived PIA signal in the hybrid film, in the 900-1050 nm
range (orange shading in Figure 4.6 and Figure 4.7). A smaller PIA signal is initially
present in the neat PCPDTBT films, and based on the very similar decays observed in
Figure 4.9, we attribute the PIA in neat polymers in this wavelength region to tails of the
broad PIA absorptions of the excitons and polarons.
The results of analysis of the decay of the excited species populations are summarized in
the following table for the neat PCPDTBT and hybrid films, obtained using multi-
! S8!
3. Determination of the time resolution of the transient absorption setup
The time resolution of the system, i.e. the pump pulse duration τ
pump
, was found to be 65 fs
based on cross-correlation between the pump and probe in two 1 mm quartz substrates (similar to
encapsulated samples). The cross-correlation response was fitted to the following equation:
1

!
"A(#
probe
,t)$exp %
t
2
(#
probe
)
&
pump
2
'
(
)
)
*
+
,
,
-sin
1
2.&
pump
2
%
t
2
(#
probe
)
.&
pump
2
%
t
0
(#
probe
)t(#
probe
)
.&
pump
2
&
probe
2
'
(
)
)
*
+
,
,

where ΔA(ω
probe
, t) is the experimentally measured pump-induced change in optical density, ω
probe

is the spectral component of the probe, t(ω
probe
) is the frequency-dependent pump-probe time
delay, τ
pump
and τ
probe
are the pump and probe pulse durations, β is the chirp rate and t
0
(ω
probe
) is the
time-zero function, which is fitted at the peak of the cross-phase modulation signals.

4. The 900-1050 nm signal in neat PCPDTBT films likely comes from the tails of the broad
exciton and polaron PIA bands
We observe a strong and long-lived PIA signal in the hybrid film, in the 900-1050 nm
range (orange shading in Figure 4 and Figure 5 of the main text). A smaller PIA signal is
initially present in the neat PCPDTBT films, and based on the very similar decays observed in
Figure S5, we attribute the PIA in neat polymers in this wavelength region to tails of the broad
PIA absorptions of the excitons and polarons.

120
exponential fits;

The PCPDBT ground state bleach overlaps with a photo-induced absorption (PIA) band.
The PIA band is detected as a positive contribution around 500 nm and through the
deformation of the ground state bleach (the high energy polymer absorption band, located
around 410 nm in the steady state absorption, is almost absent in the transient absorption
spectra). Because this PIA feature is present in spectra of PCPDTBT solutions in ODCB
(Figure 4.10), we attribute the band to high-energy singlet transitions (S
n
ç S
1
).
59

We use Bayes’ theorem to deconvolute the polymer and QD bleaches with a linear
combination model.
60
This approach gives access to statistically based error bars for the
deconvolution process.
We used a doorway function for the prior probability density function on model
parameters a and b (0,0)<(a,b)<(10,5): our only a priori is that the contributions of the
neat QD and neat polymer bleaches cannot be negative and cannot diverge. Because the
ground state bleach of the polymer is enhanced in the hybrid relative to that in the neat
film, b can be greater than one. The likelihood is given by,

where e is the experimental uncertainty and d
obs
and d
pred
are defined as follows:
! S9!

Figure S5. Transient absorption decays of the signals measured in a neat PCPDTBT film
(spectra in Figure 4 of the main text).
!
5. Multi-exponential fits of the decays dynamics of the various excited species
The results of analysis of the decay of the excited species populations are summarized in
the following table for the neat PCPDTBT and hybrid films, obtained using multi-exponential fits
.
!
1.0
0.5
0.0
A (norm)

Neat PCPDTBT, normalized traces
900 - 1050 nm
1250 - 1300 nm (polarons I)
1.0
0.5
0.0
A (norm)
0.1 1 10 100
Time (ps)

Neat PCPDTBT, normalized traces
900 - 1050 nm
1400 - 1480 nm (singlets)
1.0
0.5
0.0
A (norm)

Neat PCPDTBT, normalized traces
855 - 885 nm (polarons II)
900 - 1050 nm
!
A
i
e
"(t"t
0
)/#
i
i
$
! S11!

Figure S6. Transient absorption spectra of a PCPDTBT in ODCB solution at 1.6 × 10
‑
3
mg mL
‑
1

excited with an 800 nm pump. The pump fluence has been adjusted to get a number of absorbed
photons similar to that absorbed in the films, i.e. 1.05 × 10
14
cm
–2
. The data points are given as
circular symbols, the full line is smoothed data.

7. Deconvoluting the spectral contributions of the polymer bleach and the QD bleach
We use Bayes’ theorem to deconvolute the polymer and QD bleaches with a linear
combination model.
3
This approach gives access to statistically based error bars for the
deconvolution process.
We used a doorway function for the prior probability density function on model
parameters a and b (0,0)<(a,b)<(10,5): our only a priori is that the contributions of the neat QD
and neat polymer bleaches cannot be negative and cannot diverge. Because the ground state
bleach of the polymer is enhanced in the hybrid relative to that in the neat film, b can be greater
than one. The likelihood is given by
!
exp "
1
2
d
obs
"d
pred
( )
2
e
2
#
$
%
%
&
'
(
(
, where e is the experimental
uncertainty and d
obs
and d
pred
are defined as follows:
-1.0
-0.8
-0.6
-0.4
-0.2
0.0
0.2
A (mOD)
600 500 400
Wavelength (nm)
PCPDTBT in ODCB solution
1 ps after the 800 nm pump

121

The posterior probability density function is given by the product of the prior probability
density function and the likelihood. A Metropolis-Hastings algorithm (Matlab Statistical
Toolbox) is used to scan the parameter space (20,000 samples, 1,000 burnt samples) and
obtain the most likely model parameters (a
mean
, b
mean
), i.e., the ones that maximize the
posterior probability density function, as well as their standard deviations (a
std
, b
std
).
The parameter b of the above deconvolution gives the QD bleach contribution to the
hybrid spectra. In order to backtrack this number b to the actual density of electrons on
QDs, we need a “calibration” to relate the bleach amplitude to the density of electrons.
For this calibration, we use the spectrum obtained from neat QD films excited directly at
550 nm, shown in Figure 4.6 of the main text.
We consider the amplitude of the 1S
e
-1S
3/2
/1S
e
-2S
3/2
QD bleach when the neat QD film is
directly excited at 550 nm. This spectrum was obtained for a density of absorbed photons
I
0
,
QDs
× F
a,neat QD
= 5 × 10
14
cm
–2
× 0.14 ~ 7 × 10
13
cm
–2
. We assume that each absorbed
photons results in an excited electron, part of an electron-hole pair and occupying either
the 1S
e
level or the 1P
e
level of the conduction band of the QD. The spectrum in Figure
4.6 includes the bleaches of transitions involving these two electron states (i.e., 1S
e
at 610
nm and 1P
e
at 510 nm); and we estimate the relative populations of the 1S
e
and 1P
e
levels
from the ratio of the integrated TA bleaches at 610 nm and 510 nm – the 1P
e
level
! S12!
!
d
obs
=
"OD
hybrid
I
0
# F
a,hybrid
d
pred
= a#
"OD
polymer
I
0
# F
a,polymer
+b#
"OD
QD
(t = 75fs)
I
0
# F
a,QD

The posterior probability density function is given by the product of the prior probability density
function and the likelihood. A Metropolis-Hastings algorithm (Matlab Statistical Toolbox) is
used to scan the parameter space (20,000 samples, 1,000 burnt samples) and obtain the most
likely model parameters (a
mean
, b
mean
), i.e. the ones that maximize the posterior probability density
function, as well as their standard deviations (a
std
, b
std
).

8. Electron transfer yield calculation
The parameter b of the above deconvolution gives the QD bleach contribution to the
hybrid spectra. In order to backtrack this number b to the actual density of electrons on QDs, we
need a “calibration” to relate the bleach amplitude to the density of electrons. For this
calibration, we use the spectrum obtained from neat QD films excited directly at 550 nm, shown
in Figure 4 of the main text.
We consider the amplitude of the 1S
e
-1S
3/2
/1S
e
-2S
3/2
QD bleach when the neat QD film is
directly excited at 550 nm. This spectrum was obtained for a density of absorbed photons I
0,QDs
×
F
a,neat QD
= 5 × 10
14
cm
–2
× 0.14 ~ 7 × 10
13
cm
–2
. We assume that each absorbed photons results in
an excited electron, part of an electron-hole pair and occupying either the 1S
e
level or the 1P
e

level of the conduction band of the QD. The spectrum in Figure 4 includes the bleaches of
transitions involving these two electron states (i.e., 1S
e
at 610 nm and 1P
e
at 510 nm); and we
estimate the relative populations of the 1S
e
and 1P
e
levels from the ratio of the integrated TA
bleaches at 610 nm and 510 nm – the 1P
e
level contains F
1Pe
∼ 0.32 of the electron population and

122
contains F
1
P
e
∼ 0.32 of the electron population and the 1S
e
level contains F
1
S
e
∼ 0.68 of
the electron population. Consequently, the bleach amplitude at 610 nm in the spectrum
for neat QDs corresponds to a density of electrons on the 1S
e
energy level I
0
,
QDs
× F
a,neat

QD
× F
1
S
e
= ~ 5 × 10
13
cm
–2
.
Two control experiments were carried out to support the claim that the rapid decay of the
electron density in the hybrids is due to recombination with PCPDTBT polarons and not
to trapping into localized states. The ground state bleach recovery of the 610 nm bleach
was measured in neat QD films, with either native ligands or tBT ligands. Trapping of
electrons into localized states would result in longer bleach recovery than in the absence
of trapping. Here, the bleach recovery is much faster in the hybrid (ps timescale) than in
both the neat QD film (ns time scale), see Figure 4.10, suggesting an efficient decay
pathway to the ground state. In particular:
a) The dynamics of the ground state bleach of neat CdSe QD films is usually dominated
by the electron relaxation due to the near degeneracy of the valence band.
61
The ground
state bleach recovery pictured in black in Figure 4.12 is therefore representative of the
excited electron dynamics. The fact that the decay is much slower than in the hybrid
shows that electrons specifically are not trapped.
b) Thiols induce hole traps at the surface of CdSe(tBT) QD. The ground state bleach
recovery is greatly slowed down compared to that observed in films of CdSe capped with
native ligands, as observed in the literature
61
and in Figure 4.12. This slower decay is
attributed to hole transfer to localized, thiol-induced excited trap states that prevent full
relaxation to the ground state and therefore ground state bleach recovery. The ground

123
state bleach recovery pictured in red in Figure 4.12 is therefore representative of the
dynamics of the excited, localized hole states.
c) In the PCPDTBT:CdSe(tBT) QD hybrid, the bleach at 600 nm probes only the electron
dynamics in the context of the QDs, as argued in the text. The decay of the electron
population in the hybrid is greatly accelerated compared to the ground state bleach in neat
QD films, Figure 4.12. This means that the accelerated decay observed in the hybrid film
is not related to localized excited states that already exist in the neat QD films. Assuming
that the blending with the polymer does not induce electron traps, we therefore attribute
the rapid decay of the electron density in the hybrids to recombination with the
PCPDTBT polarons.
In Figure 4.13, we show that the electron decay time does not vary with the pump
fluence, confirming monomolecular recombination pathway. As for Figure 4.11, electron
density decays were obtained from the deconvolution of the 610 nm TA bleach in two
contributions, that of the ground state bleach of the polymer and that of the bleach of the
QDs. A solution to the electron-hole pair diffusion problem, under mutual Coulomb
attraction, is shown with the black line for all fluences.
We solve the diffusion equation describing the diffusive recombination of charge pairs
under the influence of their Coulombic mutual attraction (in our TA experiments, there is
no externally applied bias) with a numerical solver.
62
In polar coordinates, the diffusion
equation writes:

124

where ρ is the density of charges, r
c
is the Onsager radius;
,
and D is the diffusion coefficient,

Perfectly absorbing boundaries in r = 3 Å and r = 1000 Å limit the diffusion space. r = 3
Å was chosen because it correspond the tert-butylthiol ligand length: the recombination
rate is infinite (perfectly absorbing boundary) when the charges are separated by the
ligand only (i.e., by 3 Å). This means that we assume that recombination is diffusion
limited. The initial charge pair density probability is a Gaussian distribution centered at r
= 0 and we varied the Gaussian characteristic length L = σ/ (2)
0.5
(see Figure 4.14). Only
the charges that are generated at r > 3 Å are taken into account in the diffusion model
(lower boundary r = 3 Å). This is a strong assumption: we hypothesize that the diffusion
model accounts for all the measured charges.
The solution presented in Figure 6 of the main text (black line, also in Figure 4.14) was
obtained with the diffusion coefficient D calculated from measured mobility values
63
(µ
e

~ µ
h
~10
-3
cm
2
V
–1
s
–1
) D = 5 × 10
-5
cm
2
s
–1
and the Onsager radius r
c
= 5 Å
! S17!
Figure S8. Normalized electron population decay as a function of time for different pump
fluences. The top panel is equivalent to the data in Figure 6 of the main text; the black line is the
result of our diffusion model.

11. Diffusive recombination of charge pairs under the influence of an electric field
We solve the diffusion equation describing the diffusive recombination of charge pairs
under the influence of their Coulombic mutual attraction (in our TA experiments, there is no
externally applied bias) with a numerical solver.
5
In polar coordinates, the diffusion equation
writes:
!
where ρ is the density of charges, r
c
is the Onsager radius
!
r
c
=
e
2
4"#
0
#
r
k
B
T
and D is the diffusion
coefficient
!
D =
k
B
T
q
µ
e
+ µ
h
( ). Perfectly absorbing boundaries in r = 3 Å and r = 1000 Å limit
the diffusion space. r = 3 Å was chosen because it correspond the tert-butylthiol ligand length:
the recombination rate is infinite (perfectly absorbing boundary) when the charges are separated
by the ligand only, i.e. by 3 Å. This means that we assume that recombination is diffusion
limited. The initial charge pair density probability is a Gaussian distribution centered at r = 0 Å
and we varied the Gaussian characteristic length
!
L =
"
2
(see Figure S9a). Only the charges
that are generated at r > 3 Å are taken into account in the diffusion model (lower boundary r = 3
Å). This is a strong assumption: we hypothesize that the diffusion model accounts for all the
measured charges.
!
"#
"t
= D
"
2
#
"r
2
+
2
r
$
r
c
r
%
&
'
(
)
*
"#
"r
+
,
-
.
/
0
+
1 r
0
$r ( )1 t
0
$t ( )
42r
2
! S17!
Figure S8. Normalized electron population decay as a function of time for different pump
fluences. The top panel is equivalent to the data in Figure 6 of the main text; the black line is the
result of our diffusion model.

11. Diffusive recombination of charge pairs under the influence of an electric field
We solve the diffusion equation describing the diffusive recombination of charge pairs
under the influence of their Coulombic mutual attraction (in our TA experiments, there is no
externally applied bias) with a numerical solver.
5
In polar coordinates, the diffusion equation
writes:
!
where ρ is the density of charges, r
c
is the Onsager radius
!
r
c
=
e
2
4"#
0
#
r
k
B
T
and D is the diffusion
coefficient
!
D =
k
B
T
q
µ
e
+ µ
h
( ). Perfectly absorbing boundaries in r = 3 Å and r = 1000 Å limit
the diffusion space. r = 3 Å was chosen because it correspond the tert-butylthiol ligand length:
the recombination rate is infinite (perfectly absorbing boundary) when the charges are separated
by the ligand only, i.e. by 3 Å. This means that we assume that recombination is diffusion
limited. The initial charge pair density probability is a Gaussian distribution centered at r = 0 Å
and we varied the Gaussian characteristic length
!
L =
"
2
(see Figure S9a). Only the charges
that are generated at r > 3 Å are taken into account in the diffusion model (lower boundary r = 3
Å). This is a strong assumption: we hypothesize that the diffusion model accounts for all the
measured charges.
!
"#
"t
= D
"
2
#
"r
2
+
2
r
$
r
c
r
%
&
'
(
)
*
"#
"r
+
,
-
.
/
0
+
1 r
0
$r ( )1 t
0
$t ( )
42r
2
! S17!
Figure S8. Normalized electron population decay as a function of time for different pump
fluences. The top panel is equivalent to the data in Figure 6 of the main text; the black line is the
result of our diffusion model.

11. Diffusive recombination of charge pairs under the influence of an electric field
We solve the diffusion equation describing the diffusive recombination of charge pairs
under the influence of their Coulombic mutual attraction (in our TA experiments, there is no
externally applied bias) with a numerical solver.
5
In polar coordinates, the diffusion equation
writes:
!
where ρ is the density of charges, r
c
is the Onsager radius
!
r
c
=
e
2
4"#
0
#
r
k
B
T
and D is the diffusion
coefficient
!
D =
k
B
T
q
µ
e
+ µ
h
( ). Perfectly absorbing boundaries in r = 3 Å and r = 1000 Å limit
the diffusion space. r = 3 Å was chosen because it correspond the tert-butylthiol ligand length:
the recombination rate is infinite (perfectly absorbing boundary) when the charges are separated
by the ligand only, i.e. by 3 Å. This means that we assume that recombination is diffusion
limited. The initial charge pair density probability is a Gaussian distribution centered at r = 0 Å
and we varied the Gaussian characteristic length
!
L =
"
2
(see Figure S9a). Only the charges
that are generated at r > 3 Å are taken into account in the diffusion model (lower boundary r = 3
Å). This is a strong assumption: we hypothesize that the diffusion model accounts for all the
measured charges.
!
"#
"t
= D
"
2
#
"r
2
+
2
r
$
r
c
r
%
&
'
(
)
*
"#
"r
+
,
-
.
/
0
+
1 r
0
$r
( )
1 t
0
$t
( )
42r
2

125
(corresponding to an effective medium dielectric constant,

The pair survival probability density at various times after the generation event and the
Coulomb potential are represented in Figure 4.13
Reasonable agreement with the data can also be obtained using higher Onsager radii (i.e.,
more reasonable, lower dielectric constants) together with higher initial charge
separations as long as the charges are generated within r
c
. On the other hand, these
alternative models require high diffusion constants, up to D = 5 × 10
-2
cm
2
s
–1
(i.e., three
orders of magnitude larger than the diffusion coefficient calculated from typical
measured mobility values). Two alternative models are shown in Figure 4.14.
Concluding, we show that our data can be understood in a physically meaningful
framework, namely that of diffusion limited recombination. While these models can
explain the data, it is perfectly possible that a finite recombination rate at the interface
can account for the fast population decay (<1 ps), while the diffusion of charge carriers
would account for the later and slower decay (> 1 ps).

4.5. Conclusions
We have characterized an efficient hybrid BHJ composed of PCPDTBT and 4.5 nm CdSe
QDs capped with tert-butylthiol by ultrafast TA spectroscopy. After selective excitation
! S18!
The solution presented in Figure 6 of the main text (black line, also in Figure S9a) was
obtained with the diffusion coefficient D calculated from measured mobility values
6
(µ
e
~ µ
h

~10
-3
cm
2
V
–1
s
–1
) D = 5 × 10
-5
cm
2
s
–1
and the Onsager radius r
c
= 5 Å (corresponding to an
effective medium dielectric constant
€
ε
r
=
e
2
4πε
0
k
B
Tr
c
≈112). The pair survival probability density
at various times after the generation event and the Coulomb potential are represented in Figure
S8b.
Reasonable agreement with the data can also be obtained using higher Onsager radii (i.e.
more reasonable, lower dielectric constants) together with higher initial charge separations as
long as the charges are generated within r
c
. On the other hand, these alternative models require
high diffusion constants, up to D = 5 × 10
-2
cm
2
s
–1
, i.e. three orders of magnitude larger than the
diffusion coefficient calculated from typical measured mobility values. Two alternative models
are shown in Figure S9a.
Concluding, we show that our data can be understood in a physically meaningful
framework, namely that of diffusion limited recombination. While these models can explain the
data, it is perfectly possible that a finite recombination rate at the interface can account for the
fast population decay (<1 ps), while the diffusion of charge carriers would account for the later
and slower decay (> 1 ps).

126
of the low band gap polymer, a clear spectroscopic signature for ultrafast electron transfer
to the CdSe(tBT) QDs is observed. The amplitude of this spectral band gives a photon-to-
electron transfer yield of 82 ± 5% at initial times (75 fs), while its spectral position
identifies the QD energy level filled (1S
e
) by charge transfer. From the decay of the
electron population, we argue that charge pairs are generated in close proximity to each
other, which results in strong, unfavorable geminate recombination. This suggests that
this is the major loss mechanism for the hybrid blend tested here. Finally, the electron
coupling between the PCPDTBT polymer chains and the CdSe(tBT) QDs is found to be J
≳ 17 meV from the measured lower bound for the forward electron transfer rate, k ≳ 1.5
× 10
13
s
−1
.
Recently, it was shown for BHJ containing donor−acceptor polymers blended with
PCBM that higher energy excitation led to higher CT rates.
30,56,57
In the case of hybrids,
because the electronic coupling probed for low energy excitons (as here with the 800 nm
pump) is already very good, we think that higher energy excitons would open new
transfer pathways to other CdSe excited electronic levels rather than increase the rate of
electron transfer from the PCPDTBT S
1
exciton to the CdSe(tBT) QDs 1S
e
level. This
variety of transfer pathways could result in a higher global yield of charge transfer. This
scenario will be tested in a subsequent study by using various excitation wavelengths in
the transient absorption experiment and by comparing the EQE of the solar cell to the
absorbance of the active layer in a solar cell configuration. Further studies will also focus
on elucidating the impact of the polymer structure, nanocrystal morphology, and ligand
chemistry on the electron transfer yield under solar cell operating conditions.

127
4.6. References

(1) Zhou, R.; Stadler, R.; Xie, D.; Cao, W.; Zheng, Y.; Yang, Y.; Plaisant, M.;
Holloway, P. H.; Schanze, K. S.; Reynolds, J. R.; Xue, J. Enhancing the
Efficiency of Solution-Processed Polymer:Colloidal Nanocrystal Hybrid
Photovoltaic Cells Using Ethanedithiol Treatment. ACS Nano 2013, 7, 4846-
4854.
(2) Greaney, M. J.; Araujo, J.; Burkhart, B.; Thompson, B. C.; Brutchey, R. L. Novel
Semi-Random and Alternating Copolymer Hybrid Solar Cells Utilizing CdSe
Multipods as Versatile Acceptors. Chem. Commun. 2013, 49, 8602-8604.
(3) Reiss, P.; Couderc, E.; De Girolamo, J.; Pron, A. Conjugated
Polymers/Semiconductor Nanocrystals Hybrid Materials-Preparation, Electrical
Transport Properties and Applications. Nanoscale 2011, 3, 446-489.
(4) Greaney, M. J.; Das, S.; Webber, D. H.; Bradforth, S. E.; Brutchey, R. L.
Improving Open Circuit Potential in Hybrid P3HT:CdSe Bulk Heterojunction
Solar Cells via Colloidal tert-Butylthiol Ligand Exchange. ACS Nano 2012, 6,
4222-4230.
(5) Clarke, T. M.; Durrant, J. R. Charge Photogeneration in Organic Solar Cells.
Chem. Rev. 2010, 110, 6736-6767.
(6) Piliego, C.; Loi, M. A. J. Charge Transfer State in Highly Efficient Polymer-
Fullerene Bulk Heterojunction Solar Cells. J. Mater. Chem. 2012, 22, 4141-4150.
(7) Deibel, C.; Strobel, T.; Dyankonov, V. Role of the Charge Transfer State in
Organic Donor-Acceptor Solar Cells. Adv. Mater. 2010, 22, 4097-4111.
(8) Witt, F.; Kruszynska, M.; Borchert, H.; Parisi, J. Charge Transfer Complexes in
Organic-Inorganic Hybrid Blends for Photovoltaic Applications Investigated by
Light-Induced Electron Spin Resonance Spectroscopy. J. Phys. Chem. Lett. 2010,
1, 2999-3003.
(9) Grancini, G.; Biasiucci, M.; Mastria, R.; Scotognella, F.; Tassone, F.; Polli, D.;
Gigli, G.; Lanzani, G. Dynamic Microscopy Study of Ultrafast Charge Transfer in
a Hybrid P3HT/Hyperbranched CdSe Nanoparticle Blend for Photovoltaics. J.
Phys. Chem. Lett. 2012, 3, 517-523.
(10) Bensal, N.; Reynolds, L. X.; MacLachlan, A.; Lutz, T.; Ashraf, R. S.; Zhang, W.;
Nielsen, C. B.; McCulloch, I.; Rebois, D. G.; Kirchhartz, T.; Hill, M. S.; Molloy,
K. C.; Nelson, J.; Haque, S. A. Influence of Crystallinity and Energetics on

128
Charge Separation in Polymer-Inorganic Nanocomposite Films for Solar Cells.
Sci. Rep. 2013, 3 1531.
(11) Greenham, N. C.; Peng, X.; Alivisatos, A. P. Charge Separation and Transport in
Conjugated-Polymer/Semiconductor-Nanocrystal Composites Studied by
Photoluminescence Quanching and Photoconductivity. Phys. Rev. B. 1996, 54,
17628-17637.
(12) ten Cate, S.; Schins, J. M.; Siebbeles, L. D. A. Origin of Low Sensitizing
Efficiency of Quantum Dots in Organic Solar Cells. ACS Nano 2012, 6, 8983-
8988.
(13) Piris, J.; Dykstra, T. E.; Bakulin, A. A.; van Loosdrecht, P. H. M.; Knulst, W.;
Tuan Trinh, M.; Schins, J. M.; Siebbeles, L. D. A. Photogeneration and Ultrafast
Dynamics of Excitons and Charges in P3HT/PCBM Blends. J. Phys. Chem. C
2009, 113, 14500-14506.
(14) Herrmann, D.; Niesar, S.; Scharsich, C.; Köhler, A.; Stutzmann, M.; Riedle, E.
Role of Structural Order and Excess Energy on Ultrafast Free Charge Generation
in Hybrid Polythiophene/Si Photovoltaics Probed in Real Time by Near-Infrared
Broadband Transient Absorption. J. Am. Chem. Soc. 2011, 113, 18220-18233.
(15) Beek, W. J. E.; Wienk, M. M.; Janssen, R. A. J. Efficient Hybrid Solar Cells from
Zinc Oxide Nanoparticles and a Conjugated Polymer. Adv. Mater. 2004, 16,
1009-1013.
(16) Beek, W. J. E.; Wienk, M. M.; Kemerink, M.; Yang, X.; Janssen, R. A. J. Hybrid
Zinc Oxide Conjugated Polymer Bulk Heterojunction Solar Cells. J. Phys. Chem.
B 2005, 109, 9505-9516.
(17) Tagliazucchi, M.; Tice, D. B.; Sweeney, C. M.; Morris-Cohen, A. J.; Weiss, E. A.
Ligand-Controlled Rates of Photoinduced Electron Transfer in Hybrid CdSe
Nanocrystal/Poly(viologen) Films. ACS Nano 2011, 5, 9907-9917.
(18) Heinemann, M. D.; von Maydell, K.; Zutz, F.; Kolny-Olesiak, J.; Borchert, H.;
Riedel, I.; Parisi, J. Photo-Induced Charge Transfer and Relaxation of Persistent
Charge Carriers in Polymer/Nanocrystal Composites for Applications in Hybrid
Solar Cells. Adv. Funct. Mater. 2009, 19, 3788-3795.
(19) Itskos, G.; Othonos, A.; Rauch, T.; Tedde, S. F.; Hayden, O.; Kovalenko, M. V.;
Heiss, W.; Choulis, S. A. Optical Properties of Organic Semiconductor Blends
with Near-Infrared Quantum-Dot Sensitizers for Light Harvesting Applications.
Adv. Energy Mater. 2011, 1, 802-812.

129
(20) Boucle, ́ J.; Chyla, S.; Shaffer, M. S. P.; Durrant, J. R.; Bradley, D. D. C.; Nelson,
J. Hybrid Solar Cells from a Blend of Poly(3-hexylthiophene) and Ligand-Capped
TiO
2
Nanorods. Adv. Funct. Mater. 2008, 18, 622-633.
(21) Dayal, S.; Kopidakis, N.; Olson, D. C.; Ginley, D. S.; Rumbles, G. Dayal, S.;
Kopidakis, N.; Olsen, D. C.; Ginley, D. S.; Rumbles, G. Photovoltaic Devices
with a Low Band Gap Polymer and CdSe Nanostructures Exceeding 3%
Efficiency. Nano Lett. 2010, 10, 239-242.
(22) Rivera, C. A.; Winter, N.; Harper, R. V.; Benjamin, I.; Bradforth, S. E. The
Dynamical Role of Solvent on the ICN Photodissociation Reaction: Connecting
Experimental Observables Directly with Molecular Dynamics Simulations. Phys.
Chem. Chem. Phys. 2011, 13, 8269-8283.
(23) Roberts, S. T.; Schlenker, C. W.; Barlier, V.; McAnally, R. E.; Zhang, Y.;
Mastron, J. N.; Thompson, M. E.; Bradforth, S. E. Observation of Triplet Exciton
Formation in a Platinum-Sensitized Organic Photovoltaic Device. J. Phys. Chem.
Lett. 2011, 2, 48-54.
(24) Mühlbacher, D.; Scharber, M.; Morana, M.; Zhu, Z.; Waller, D.; Gaudiana, R.;
Brabec, C. High Photovoltaic Performance of a Low-Bandgap Polymer. Adv.
Mater. 2006, 18, 2884-2889.
(25) Webber, D. H.; Brutchey, R. L. Ligand Exchange on Colloidal CdSe Nanocrystals
Using Thermally Labile tert-Butylthiol for Improved Photocurrent in Nanocrystal
Films. J. Am. Chem. Soc. 2012, 134, 1085-1092.
(26) Norris, D. J.; Bawendi, M. G. Measurment ans Assignment of the Size-Dependent
Optical Spectrum in CdSe Quantum Dots. Phys. Rev. B 1996, 53, 16338-16346.
(27) Klimov, V. I.; McBranch, D. W.; Leatherdale, C. A.; Bawendi, M. G. Electron
and Hole Relaxation Pathways in Semiconductor Quantum Dots. Phys. Rev. B
1999, 60, 13740-13749.
(28) Fiore, A.; Chandezon, F.; Djurado, D.; Reiss, R.; Faure-Vincent, J. Charge
Transport in Poly(3-hexylthiophene):CdSe Nanocrystals Hybrid Films
Investigated with Time-of-Flight Measurements. Appl. Phys. Lett. 2012, 101,
133301.
(29) Tsoi, W. C.; Spencer, S. J.; Yang, L.; Ballantyne, A. M.; Nicholson, P. G.;
Turnbull, A.; Shard, A. G.; Murphy, C. E.; Bradley, D. D. C.; Nelson, J.; Kim, J.-
S. Effect of Crystallization on the Electronic Energy Levels and Thin Film
Morphology of P3HT:PCBM Blends. Macromolecules 2011, 44, 2944-2952.

130
(30) Grancini, G.; Maiuri, M.; Fazzi, D.; Petrozza, A.; Egelhaaf, H.-J.; Brida, D.;
Cerullo, G.; Lanzani, G. Hot Exciton Dissociation in Polymer Solar Cells. Nat.
Mater. 2013, 12, 29-33.
(31) Hwang, I.-W.; Soci, C.; Moses, D.; Zhu, Z.; Waller, D.; Gaudiana, R.; Brabec, C.
J.; Heeger, A. J. Ultrafast Electron Transfer and Decay Dynamics in a Small Band
Gap Heterojunction Material. Adv. Mater. 2007, 19, 2307-2312.
(32) Yamamoto, S.; Ohkita, H.; Benten, H.; Ito, S. Role of Interfacial Charge Transfer
State in Charge Generation and Recombination in Low-Bandgap Polymer Solar
Cell. Phys. Chem. C 2012, 116, 14804-14810.
(33) Tautz, R.; Da Como, E.; Limmer, T.; Feldmann, J.; Egelhaaf, H.- J.; von Hauff,
E.; Lemaur, V.; Beljonne, D.; Yilmaz, S.; Dumsch, I.; Allard, S.; Scherf, U.,
Structural Correlations in the Generation of Polaron Pairs in Low-Bandgap
Polymers for Photovoltaics. Nat. Commun. 2012, 3, 970.
(34) Wiebeler, C.; Tautz, R.; Feldmann, J.; von Hauff, E.; Da Como, E.; Schumacher,
Spectral Signatures of Polarons in Conjugated Co-polymers. J. Phys. Chem. B
2013, 117, 4454-4460.
(35) Kaake, L. G.; Jasieniak, J. J.; Bakus, B. C.; Welch, G. C.; Moses, D.; Bazan, G.
C.; Heeger, A. J. Photoinduced Charge Generation in a Molecular Bulk
Heterojunction Material. J. Am. Chem. Soc. 2012, 134, 19828-19838.
(36) Etzold, F.; Howard, I. A.; Forler, N.; Cho, D. M.; Meister, M.; Mangold, H.; Shu,
J.; Hansen, M. R.; Muellen, K.; Laquai, F. The Effect of Solvent Additives on
Morphology and Excited-State Dynamics in PCPDTBT:PCBM Photovoltaic
Blends. J. Am. Chem. Soc. 2012, 134, 10569-10583.
(37) McArthur, E. A.; Morris-Cohen, A. J.; Knowles, K. E.; Weiss, E. A. Charge
Carrier Resolved Relaxation of the Excitonic State in CdSe Quantum Dots Probed
with Near-Infrared Transient Absorption Spectroscopy J. Phys. Chem. B 2010,
114, 14514−14520.
(38) Knowles, K. E.; McArthur, E. A.; Weiss, E. A. A Multi-Timescale Map of
Radiative and Nonradiative Decaty Pathways for Excitons in CdSe Quantum
Dots. ACS Nano 2011, 5, 2026-2035.
(39) Colbert, A. E.; Janke, E. M.; Hsieh, S. T.; Subramaniyan, S.; Schlenker, C. W.;
Jenekhe, S. A.; Ginger, D. S. Hole Transfer from Low Band Gap Quantum Dots
to Conjugated Polymers in Organic/Inorganic Hybrid Photovoltaics. J. Phys.
Chem. Lett. 2013, 4, 280−284.

131
(40) Moulé, A. J.; Chang, L.; Thambidurai, C.; Vidu, R.; Stroeve, P. Hybrid Solar
Cells: Basic Principles and the Role of Ligands. J. Mater. Chem. 2012, 22, 2351-
2368.
(41) Xu, T.; Qiao, Q. Conjugated Polymer-Inorganic Semiconductor Hybrid Solar
Cells. Energy Environ. Sci. 2011, 4, 2700-2720.
(42) Ren, S.; Chang, L. Y.; Lim, S. K.; Zhao, J.; Smith, M.; Zhao, N.; Bulović, V.;
Bawendi, M.; Gradečak, S. Inorganic–Organic Hybrid Solar Cell: Bridging
Quantum Dots to Conjugated Polymer Nanowires. Nano Lett. 2011, 11, 3998-
4002.
(43) Dayal, S.; Reese, M. O.; Ferguson, A. J.; Ginley, D. S.; Rumbles, G.; Kopidakis,
N. The Effect of Nanoparticle Shape on the Photocarrier Dynamics and
Photovoltaic Device Performance of Poly(3-hexylthiophene):CdSe Nanoparticle
Bulk Heterojunction Solar Cells. Adv. Funct. Mater. 2010, 20, 2629-2635.
(44) Gur, I.; Fromer, N. A.; Chen, C. P.; Kanaras, A. G.; Alivisatos, A. P. Hybrid Solar
Cells with Prescribed Morphologies Based on Hyperbranched Semiconductor
Nanocrystals. Nano Lett. 2007, 7, 409-414.
(45) Greaney, M. J.; Das, S.; Webber, D. H.; Bradforth, S. E.; Brutchey, R. L.
Improving Open Circuit Potential in Hybrid P3HT:CdSe Bulk Heterojunction
Solar Cells via Colloidal tert-Butylthiol Ligand Exchange. ACS Nano 2012, 6,
4222-4230.
(46) Fu, W.; Shi, Y.; Qiu, W.; Wang, L.; Nan, Y.; Shi, M.; Li, H.; Chen, H. High
Efficiency Hybrid Solar Cells Using Post-Deposition Ligand Exchange by
Monothiols. Phys. Chem. Chem. Phys. 2012, 14, 12094-12098.
(47) Jeltsch, K. F.; Schadel, M.; Bonekamp, J. B.; Niyamakom, P.; Rauscher, F.;
Lademann, H. W. A.; Dumsch, I.; Allard, S.; Scherf, U.; Meerholz, K. Enhanced
Efficiency Hybrid Solar Cells Using a Blend of Quantum Dots and Nanorods.
Adv. Funct. Mater. 2012, 22, 397-404.
(48) Dayal, S.; Kopidakis, N.; Olson, D. C.; Ginley, D. S.; Rumbles, G. Photovoltaic
Devices with a Low Band Gap Polymer and CdSe Nanostructures Exceeding 3%
Efficiency. Nano Lett. 2010, 10, 239-242.
(49) Zhou, R.; Stalder, D.; Xie, D.; Cao, W.; Zheng, Y.; Yang, Y.; Plaisant, M.;
Halloway, P. H.; Shanze, K. S.; Reynolds, J. R.; Xue. Enhancing the Efficiency of
Solution-Processed Polymer:Colloidal Nanocrystal Hybrid Photovoltaic Cells
Using Ethanedithiol Treatment. ACS Nano 2013, 7, 4846-4854.
(50) Khylabich, P. P.; Burkhart, B.; Thompson, B. C. Compositional Dependence of
the Open-Circuit Voltage in Ternary Blend Bulk Heterojunction Solar Cells
Based on Two Donor Polymers. J. Am. Chem. Soc. 2012, 134, 9074-9077.

132
(51) Khylabich, P. P.; Burkhart, B.; Ng, C. F.; Thompson, B. C. Efficient Solar Cells
From Semi-Random P3HT Analogues Incorporating Diketopyrrolopyrrole.
Macromolecules 2011, 44, 5079-5084.
(52) Burkhart, B.; Khlyabich, P. P.; Thompson, B. C. Influence of the Acceptor
Composition on Physical Properties and Solar Cell Performance in Semi-Random
Two-Acceptor Copolymers. ACS Macro Lett. 2012, 1, 660-666.
(53) Park, S. H.; Roy, A.; Beaupre, S.; Cho, S.; Coates, N.; Moon, J. S.; Moses, D.;
Leclerc, M.; Lee, K.; Heeger, A. J. Bulk Heterojunction Solar Cells with Internal
Quantum Efficiency Approaching 100%. Nature Photon. 2009, 3, 297-302.
(54) Blouin, N.; Michaud, A.; Gendron, D.; Wakim, S.; Blair, E.; Neagu-Plesu, R.;
Belletete, M.; Durocher, G.; Tao, Y.; Leclerc, M. Toward a Rational Design of
Poly(2, 7-Carbazole) Derivatives for Solar Cells. J. Am. Chem. Soc., 2008, 130,
732-742.
(55) Webber, D. H.; Brutchey, R. L. Ligand Exchange on Colloidal CdSe Nanocrystals
Using Thermally Labile tert-Butylthiol for Improved Photocurrent in Nanocrystal
Films. J. Am. Chem. Soc. 2012, 134, 1085-1092.
(56) Qian, L.; Yang, J.; Zhou, R.; Tang, A.; Zheng, Y.; Tseng, T. K.; Bera, D.; Xue, J.;
Holloway, P. H. Hybrid Polymer-CdSe Solar Cells with a ZnO Nanoparticle
Buffer Layer for Improved Efficiency and Lifetime. J. Mater. Chem., 2011, 21,
3814-3817.
(57) Lek, J. Y.; Lam, Y. M.; Niziol, J.; Marzec, M. Understanding Polycarbazole-
Based Polymer:CdSe Hybrid Solar Cells. Nanotechnology, 2012, 23, 315401.
(58) Kovalenko, S. A.; Dobryakov, A. L.; Ruthmann, J.; Ernsting, N. P. Femtosecond
Spectroscopy of Condensed Phases with Chirped Supercontinum Probing. Phys.
Rev. A, 1999, 59, 2369-2384.
(59) Grancini, G.; Martino, N.; Antognazza, M. R.; Celebrano, M.; Egelhaaf, H-J.;
Lanzani, G. Influence of Blend Composition on Ultrafast Charge Generation and
Recombination Dynamics in Low Band Gap Polymer-BasedOrganic
Photovoltaics. J. Phys. Chem. C, 2012, 116, 9838-9844.
(60) Press, W. H.; Teukolsky, S. A.; Vetterling, W. T.; Flannery, B. P. Numerical
Recipes in C: the Art of Scientific Computing; Cambridge University Press, 1988.
(61) Burda, C.; Link, S.; Mohamed, M.; El-Sayed, M. The Relaxation Pathways of
CdSe Nanoparticles Monitored with Femtosecond Time-Resolution from the
Visible to the IR: Assignment of the Transient Features by Carrier Quenching. J.
Phys. Chem. B, 2001, 105, 12286-12292.

133
(62) Krissinel, E. B.; Agmon, N. Proton=-Transfer Reaction Dynamics Within the
Human Serum Albumin Protein. J. Comput. Chem., 1996, 17, 1085–1096.
(63) Couderc, E.; Bruyant, N.; Fiore, A.; Chandezon, F.; Djurado, D.; Reiss, R.; Faure-
Vincent, J. Charge transport in Poly(3-hexylthiophene):CdSe Nanocrystals
Hybrid Thin Films Investigated With Time-of-Flight Measurements Appl. Phys.
Lett., 2012, 101, 133301.

134
Chapter 5. Ligand- and Size-Dependent Ultrafast Electron Transfer at
Hybrid Polymer:Nanocrystal Interfaces

5.1. Introduction
Organic and hybrid bulk heterojunction (BHJ) solar cells are developed to harvest solar
energy through the use of low cost, thin film materials chosen for their high
absorptivities. Within the junction, inorganic acceptors are used for the potential to
further benefit from their tunable absorption edge and energy levels, their high dielectric
constants, and their high intrinsic charge carrier mobilities. Champion hybrid BHJs
comprised of a conjugated polymer blended with semiconductor quantum dots (QDs)
have been reported to achieve 5.5% power conversion efficiency.
1-3
High efficiencies
have primarily been achieved by engineering the surface chemistry of the QDs, which is
thought to control charge formation and charge transport processes in these blends.
4

Indeed, charge transfer (CT) at the donor-acceptor interface governs the efficiency of
charge formation in BHJ solar cells. However, a clear picture of how various material
parameters influence the CT efficiency has not emerged so far due to various reasons:
first because CT can occur over a wide range of time scales (fs-ns),
5-7
and secondly
because it is difficult to systematically compare works of different groups due to the wide
variety of donors and acceptors that have been investigated.
Here, we set out to understand the impact of some materials parameters on the CT
efficiencies with femtosecond time resolution. To do so, we focus on hybrid BHJ solar
cells comprised of semiconductor quantum dots (QDs) blended with a low band gap

135
conjugated polymer: by modifying nanocrystal parameters such as their size and surface
chemistry, we can study a range of interface parameters, like the distance between the
donor and the acceptor components and the energy level alignment at their interface.
For instance, QD surface chemistry can be controlled to change the thickness of the
ligand cap, which has been shown to widely modify the electron transport properties of
nanocrystal films by increasing electronic coupling thus leading to more facile electron
hopping.
8,9
In the context of CT at a donor/acceptor (D/A) interface, a thicker insulating
layer of ligands can also decrease the electronic coupling between the initial and final
states of the CT process and therefore drive down the initial CT yield. On the other hand,
we expect the increased separation between the electron and hole on both sides of the
D/A interface to reduce attractive Coulombic interactions and hence geminate
recombination. The influence of the delocalization of interfacial CT states on free carrier
photogeneration is currently a hot topic of debate
10,11
and we propose here to use the
ligand layer thickness to tune the initial electron-hole separation after charge transfer.
Furthermore, nanocrystals offer tunable energy levels, both via size and surface
chemistry adjustments. The energy offset between the excited state of the donor and the
charge-transferred state is believed to dictate the CT efficiency: Marcus theory predicts
that the efficiency of CT increases with the energy offset up to an optimal value and then
decreases (normal and inverted regions, respectively). In organic BHJs, reports have
been made that the nanosecond CT yield indeed depends on the energy offset: CT occurs
either in the normal or in the inverted region of Marcus theory, depending on the energy
level of the fullerene acceptor.
12
In organic-inorganic BHJ, the influence of the energy

136
level offset on the CT efficiency is not mapped out precisely yet, as the most recent
studies
13,14
(i) do not distinguish between electron and hole transfer processes and (ii)
report on nanosecond polaron populations, thereby not resolving the initial CT
efficiencies from the population decay due to early recombination.
Here, we set out to resolve these issues by selectively exciting one component of the
blend to observe only electron transfer and by probing femtosecond time scales to
unravel the effect of recombination of charge carriers on the observed CT efficiencies.
We probe the effects of the nanocrystal surface chemistry and size on the initial CT yield
and early recombination. We do this while keeping in mind that disorder leads to a
variety of interfacial configurations. In particular, the inhomogeneties of the ligand cap,
the molecular orientations of polymer chains relative to QD facets, and the degree of
phase separation likely lead to a distribution of electronic couplings at the interface and
consequently to a distribution of CT times.

5.2. Results and Discussion
The photovoltaic performances of hybrid devices, in which the active layers are made of
CdSe QDs blended with the low band gap polymer poly[2,6-(4,4-bis(2-ethylhexyl)-4H-
cyclopenta[2,1-b;3,4-b’]-dithiophene)-alt-4,7-(2,1,3-benzothiadiazole)] (PCPDTBT),
depend strongly on the QD ligands and on the QD size. This is visible from Figure 5.1a,
where the external quantum efficiencies (EQEs) of various devices are shown (also see
Table 5.1). The devices made with CdSe QDs capped with native ligands (NL) exhibit

137
extremely low power conversion efficiencies (PCE) between 0.001- 0.1%, with the
smallest QDs giving the lowest PCEs. The devices made with pyridine (Py) and n-
butylamine (BA) treated QDs show low PCE (0.47 ± 0.11% and 0.25 ± 0.04%
respectively), but recognizable features from the QD and the polymer. Using tert-
butylthiol (tBT) as a ligand greatly enhances the PCE (1.57 ± 0.32%) and the EQE at all
wavelengths. Finally, using bigger QD resulted in more efficient devices (PCE=0.57 ±
0.04 % for 3.4 nm QDs, 1.93 ± 0.31% for 3.9 nm QDs, and 2.35 ± 0.12 % for 5.3 nm
QDs).

Figure 5.1. (a) External quantum efficiencies of hybrid PCPDTBT:CdSe devices. A series of QD
ligands were used (NL, BA, Py and tBT) and three sizes are shown for the CdSe(tBT) hybrids
(3.4 nm, 3.9 nm, and 5.3 nm). The device architecture is
ITO/PEDOT:PSS/PCPDTBT:CdSe/ZnO/Al in all cases. The thick and thin arrows indicate
regions of polymer and QD absorption, respectively. (b) Absolute absorbance of the neat
PCPDTBT film and of the hybrid PCPDTBT:CdSe QD BHJ films whose TA spectra are shown
in Figure 5.4.

polymer
contribution
QDs
contribution
PCPDTBT
native
ligands
butylamine
pyridine
tert-butylthiol
Hybrids 3.9 nm:
a
b
0.2
0.1
0.0
Absorbance (OD)
900 800 700 600 500 400
Wavelength (nm)
30
20
10
0
External Quantum Efficiency (%)
native ligands
butylamine
pyridine
tert-butylthiol
3.4 nm
3.9 nm
5.3 nm

138
The size trend stems from two different effects: the increase observed from 3.4 nm
CdSe(tBT) QDs to 3.9 nm CdSe(tBT) QDs is due to a large overall increase in EQE
across the entire wavelength range whereas the increase from 3.9 nm CdSe(tBT) QDs to
5.3 nm CdSe(tBT) QDs is mainly due to the red-shifted absorption edge of the QDs,
contrasting with the lower absorbance shown in Figure 5.1b. Representative I-V curves
are shown in Figure 5.2.

Figure 5.2. Representative I-V curves of hybrid solar cells whose EQEs are shown in Figure 5.1.
The device architecture is ITO/PEDOT:PSS/PCPDTBT:CdSe QDs/ZnO/Al. The devices based
on tBT-exchanged QD show higher J
sc
and V
oc
.

Characterization of the surface chemistry of the QDs helps rationalize the overall
variations of the photovoltaic performances. In particular, thermogravimetric analysis
(TGA) provides a simple method for quantifying the efficacy of ligand exchange, and
therefore the ligand coverage. Figure 5.3 shows the mass loss of CdSe(ligand, 3.9 nm)
QDs as a function of temperature. The TGA trace for the CdSe(NL) QDs shows a mass
loss event between 260 °C and 500 °C corresponding to decomposition of the strongly
bound, long chain carboxylate ligands, while the ligand exchanged CdSe QDs exhibit
-6
-4
-2
0
2
4
6
Current (A)
0.8 0.6 0.4 0.2 0.0 -0.2
Voltage (V)
PCPDTBT:CdSe(3.9 nm, ligands)
pyridine
tert-butylthiol

139
mass loss events beginning at lower temperatures. In the case of CdSe(tBT) QDs, 86%
of the total mass loss is already complete by 260 °C, indicating that 94% of the molecules
at the surface of the CdSe(tBT) QDs are tBT ligands (calculated from the ratio of molar
masses between tBT and tetradecanoate, see the Experimental Section for details). On
the other hand, both CdSe(BA) and CdSe(Py) QDs give a non-negligible mass loss event
in the 260-500 °C range, suggesting that Py and BA only partially exchange with NL.
The inability of neutral Py and BA to quantitatively displace NL is expected since NLs
may bind a neutral carboxylic acids or as anionic carboxylates. The requirement to
maintain an overall charge balance prevents such nitrogen-based ligands from achieving
complete ligand exchange with carboxylic acid-based NLs. This highlights the dual
composition of the ligand corona in these samples: the BA- and Py-exchanged CdSe QDs
possess both long and short ligands on their surfaces whereas CdSe(NL) and tBT-
exchanged CdSe QDs possess only long or only short ligands, respectively, as
schematized in Figure 5.3. The presence of long ligands in the NL, Py, and BA devices
correlates with the lower PV efficiencies, which may suggest an impediment to charge
transfer across the QD:polymer interface due to electrically insulating organic ligands.

Figure 5.3. Thermogravimetric analysis of the 3.9 nm CdSe QDs with various ligands and
schematic representations of the QD surface coverage, as discussed in the text.
1.0
0.9
0.8
0.7
0.6
Mass (normalized)
500 400 300 200
Temperature (°C)
native ligands
butylamine
pyridine
tert-butylthiol
QD
QD QD
QD
QD
QD

140
The energy offset between the excited state of the donor and the reduced state of the
acceptor influences the exciton separation via CT in organic devices, and hence can also
contribute to the variations in PV efficiency. In particular, changes in size and surface
chemistry shift the QD LUMO levels and therefore tune the energy offset for electron
transfer. For example, the deeper LUMO of bigger QDs would favor electron transfer
when it occurs in the normal Marcus region. According to the parameterization by
Meulenberg et al.
15
, the LUMO level of CdSe QDs shifts by tens of meV between 3.4 nm
and 5.3 nm. Additionally, modifications in the CdSe QD surface chemistry also leads to
shifts in the LUMO levels of the QDs as measured using cyclic and differential pulse
voltammetries (see Table 5.1). The trends in LUMO energy observed for the 3.9 nm
CdSe(ligand) QDs are fully consistent with our previously published results for NL, tBT,
and Py ligands on 4.5 nm CdSe QDs, in which CdSe(tBT) exhibits the highest energy
(i.e., most shallow) LUMO and CdSe(Py) exhibits the lowest energy (i.e., deepest)
LUMO level.
16
We also find that BA ligands lower the LUMO level by an additional 50
meV compared to the Py ligands. The maximum difference between the LUMO levels is
observed between the CdSe(tBT) and CdSe(BA) QDs and is on the order of 200 meV.
In Scheme 5.1, we organize the energetic driving force for charge transfer at the
polymer:QD interface, estimated from the LUMO levels of the neat QDs. By fabricating
hybrid solar cells with QDs of different sizes and different QD surface chemistry, we
attempt to elucidate the relation between the PV performance and the energy level
alignment. Table 5.1 shows that the EQE at 800 nm for each hybrid system does not
correlate linearly with the LUMO level of the QDs as the NL sample exhibits an
intermediate LUMO level but minimal EQE values. This suggests that the effect of

141
energy offset on the EQE of hybrid devices cannot be completely explained in the
framework of Marcus theory. Since the EQE depends on the efficiency of many
successive steps – the absorption of photons of a given energy, their dissociation by
charge transfer at interfaces, and the equilibrium between charge extraction and charge
recombination – this finding motivates our further investigation of the individual steps of
light-to-current conversion.

Scheme 5.1. Schematic representation of the driving force for charge transfer at the hybrid
interface, assuming that the energies of the charge transferred states for different QD sizes and
ligands vary in a similar trend to the LUMO levels of the neat QDs (given in Table 5.1).

energy (eV)
PCPDTBT
excited state
ground state
3.9 nm native ligands
3.9 nm butylamine
3.9 nm pyridine
3.9 nm, tert-butylthiol
5.3 nm, tert-butylthiol
3.4 nm, tert-butylthiol
hybrid CT state

142
Table 5.1. Comparison between the various quantities observed in the text: external
quantum efficiencies of various devices measured at 800 nm, fraction of short ligands in the
ligand corona, LUMO level of the QD acceptor († measured by cyclic voltammetry, see
details in the Experimental, section 5.3.2; * deduced from the shifts from Ref. [14]) and
amplitude of the spectral signature for charge transfer.
Ligand and size of
the QD in the
PCPDTBT:CdSe
hybrid
EQE of the
solar cell
device at
800 nm
Number
fraction of
short ligands
in the ligand
corona
LUMO level (eV) Amplitude of the first
exciton signature,
relative to that of the
PCPDTBT: CdSe(tBT,
3.9 nm)
NL, 3.9 nm < 0.1% 0% -3.90 ± 0.05 eV † n/a
Py, 3.9 nm 0.5% 37% -4.00 ± 0.05 eV † 0.53 ± 0.11
BA, 3.9 nm 0.5% 52% -4.05 ± 0.05 eV † 0.63 ± 0.11
tBT, 3.9 nm 7.5% 94% -3.85 ± 0.05 eV † 1.00 ± 0.07
tBT, 3.4 nm 1% 86% -3.79 ± 0.05 eV * 1.77 ± 0.24
tBT, 5.3 nm 10.3% 68% -3.96 ± 0.05 eV * XX
PCPDTBT 2
tBT, 4.5 nm
3.7% 96% -3.8 eV DA/(I_0 x F_a) ~4.3-
4.6 x 10
-14

PCPDTBT 2
tBT, 6 nm
No data 76% -4.01 ± 0.05 eV * DA/(I_0 x F_a) ~5.9-
6.3 x 10
-14

In order to understand exactly why the surface chemistry and QD size have such strong
impacts on PV efficiencies, we focused on the earliest steps of energy conversion and
probed the charge transfer processes at very early times with ultrafast transient absorption
measurements. Transient absorption is a powerful spectroscopic technique that lets us
identify the transient species present in the hybrid films (i.e.; excitons and polarons) and
follow the dynamics of these excited species.

143
Here, a 45 fs, 800 nm laser pulse was used to excite the low energy edge of the
PCPDTBT absorbance. In the neat polymer film, this excitation pulse primarily leads to
the formation of PCPDTBT excitons, which results in the bleaching of the ground state
absorption and the appearance of photo-induced absorption bands specific to the transient
species.
5,17
In the hybrid films, the 800 nm excitation pulse selectively excites the
PCPDTBT component of the BHJ. The consequent generation of polymer excitons can
only result in electronic relaxation to the ground state or in electron transfer to the CdSe
QD LUMO, as seen from the Type-II energy level alignment shown in Scheme 5.1.
Detection of electron transfer in this hybrid system is accomplished by monitoring the
550-800 nm region in transient absorption experiments. Indeed, as a result of electron
transfer, the polymer bleach is complemented by additional features composed of a
(negative) bleach of the 1S
e
-1S
h,3/2
CdSe QD transition (at 560-590 nm for the 3.9 nm
QDs used in the samples of Figure 5.4) and a (positive) photo-induced absorption (PIA)
feature arising from Coulomb interactions in CdSe QD (at 590-620 nm in Figure 5.4).
19

144

Figure 5.4. Effect of QD surface ligand on ultrafast electron transfer: transient absorption spectra
for various pump-probe delays (λ
exc
= 800 nm) and for a series of samples made with 3.9 nm
CdSe QDs with different ligands. The ligand that corresponds to each set of TA spectra is listed
in the lower right corner. The excitation fluence was 90 µJ/cm
2
.

−6
−4
−2
0
ΔA (mOD)

0 ps
0.3 ps
1 ps
3 ps
10 ps
100 ps
−4
−3
−2
−1
0

ΔA (mOD)
0 ps
0.3 ps
1 ps
3 ps
10 ps
100 ps
−2
−1
0

ΔA (mOD)
−3
0 ps
0.3 ps
1 ps
3 ps
10 ps
100 ps
−4
−3
−2
−1

ΔA (mOD)
0
0 ps
0.3 ps
1 ps
3 ps
10 ps
100 ps
−2
−1
0

550 600 650 700 750 800
Wavelength (nm)
ΔA (mOD)
0 ps
0.3 ps
1 ps
3 ps
10 ps
100 ps
−3
neat
polymer
NL
Pyr
BA
tBT

145
First, we address the effect of QD surface chemistry on the photophysical dynamics of
hybrid BHJs made with 3.9 nm CdSe QDs. In the case of the hybrid
PCPDTBT:CdSe(NL) films, no spectral signature of electron transfer is visible at any
pump-probe delay (see Figure 5.4). This suggests that the negligible EQE response
(<0.1%, Figure 5.1a) at 800 nm of the PCPDTBT:CdSe(NL) solar cells arises from the
very low yield of electron transfer from the polymer chains to the 3.9 nm CdSe(NL) QDs
after excitation at 800 nm rather than from ulterior loss mechanisms. This comes as no
surprise, since electrically insulating NLs are known to impede charge transfer and
transport at QD interfaces. On the other hand, in the case of hybrid films utilizing the
ligand exchanged CdSe(tBT), CdSe(Py), and CdSe(BA) QD acceptors, the spectral
signature of electron transfer is clearly visible as a negative bleach at 580 nm that
corresponds to QD reduction (Figure 5.4). It is interesting to note that for these three
ligand-exchanged CdSe QD acceptors, the first excitonic transition bleach appears within
the time resolution of our transient absorption apparatus (i.e., 45 fs, see the Experimental
section 5.3.3). In other words, electron transfer occurs on an extremely rapid timescale
for all of the ligand exchanged QDs. Such a rapid electron transfer event could be
followed by slower transfer events, but these are not detected as a resolved enhancement
of the QD bleach signal; if slower transfer events occur, their signature is compensated in
the TA signal by the population decay.
The amplitude of the negative 580 nm spectral features varies with the ligands on the
CdSe QDs – it is most intense in the case of tBT ligands while BA and Py ligands much
smaller amplitudes but are comparable with each other. This ligand trend is robust and
was also observed for hybrid films made with 6 nm CdSe QDs and PCPDTBT of a

146
different molecular weight, as shown in Figures 5.5-5.7).

Figure 5.5. Thermogravimetric analysis of the CdSe(6 nm) QDs before and after each type of
ligand exchange.

Figure 5.6. Absolute absorbance of the neat PCPDTBT and PCDPTBT:CdSe (6 nm) films used
in transient absorption (Figure 5.7). (a) Raw data; (b) data normalized to the red maxima to
facilitate the comparison of spectral shapes.

1.00
0.95
0.90
0.85
Mass (normalized)
500 400 300 200
Temperature (C)
6nm CdSe QDs
native ligands
butylamine
pyridine
tert-butylthiol

147

Figure 5.7. Transient absorption spectra for various pump-probe delays (λ
exc
= 800 nm) and for a
series of samples made with 6 nm CdSe QDs with different ligands. The excitation fluence was
60 µJ/cm
2
.

By spectrally deconvoluting the QD bleach from the polymer bleach (explained in
−2
−1
0

∆A (mOD)
−30
−20
−10
0

∆A (mOD)
−6
−4
−2
0

∆A (mOD)
−3
−2
−1
0

∆A (mOD)
550 600 650 700 750 800
Wavelength (nm)
−1
0

∆A (mOD)

neat
polymer
NL
Pyr
BA
tBT
0 ps
0.3 ps
1 ps
3 ps
10 ps
100 ps
0 ps
0.3 ps
1 ps
3 ps
10 ps
100 ps
0 ps
0.3 ps
1 ps
3 ps
10 ps
100 ps
0 ps
0.3 ps
1 ps
3 ps
10 ps
100 ps
0 ps
0.3 ps
1 ps
3 ps
10 ps
100 ps

148
Experimental, section 5.3.3, Figure 5.8), we can estimate the relative efficiencies of
electron transfer between PCPDTBT and CdSe (see Table 5.1). We thereby
quantitatively confirm that electron transfer is less efficient for PCPDTBT:CdSe(Py) and
PCPDTBT:CdSe(BA) hybrids than for PCPDTBT:CdSe(tBT) hybrids, with only 63 ±
11% and 51 ± 16 % of electron transfers relative to tBT, respectively. The photovoltaic
efficiencies follow the same ligand trend, with the EQE at 800 nm being the largest for
CdSe(tBT) based samples (7.5%), followed by CdSe(Py) and CdSe(BA) based samples
(both 0.5%), and finally CdSe(NL) (<0.1%). This important correlation identifies the
very initial step of CT as one of the sources of limitation in working hybrid devices.
We discussed above that PV efficiencies do not correlate with the LUMO of the QDs.
Based on TA measurements, we conclude that the band alignment is not the dominating
factor in determining the electron transfer efficiency at short times in our hybrid
PCPDTBT:CdSe QD BHJs. We believe that the dominating factor here is the
wavefunction overlap between the initial and final states of electron transfer: it is
negligible when the polymer chain and the QD are separated by the native ligand length
(~1-2 nm) but it becomes significant when the counterparts are separated by shorter
ligands (~0.3 nm for tBT). The mixed ligand shells of Py and BA treated QDs results in
a distribution of large and small electronic couplings between the donor and acceptor
depending on whether the polymer chain and the QD acceptor are separated by a short or
long ligand. Heterogeneities in the active layer therefore explain that charge transfer can
be ultrafast without being 100% efficient.

149
We have also studied recombination of these charge pairs by deconvoluting transient
absorption signals from the QD and from the polymer in the 550-800 nm region (see
Experimental section 5.3.3). The normalized decay dynamics are similar within the
signal-to-noise ratio for all tBT, Py, and BA based hybrids as shown in Figure 5.8.

Figure 5.8. The transient absorption spectra from Figure 5.4 for tBT, Py, and BA based hybrids
were deconvoluted into a QD component (a) and a polymer component (b). The circle symbols
represent the most likely values of the deconvolution and the error bars represent the statistical
uncertainties in the deconvolution obtained by Bayesian fits, as detailed in the Experimental
section. The polymer component reflects the populations of both polymer excitons and polarons
and therefore exhibits a different dynamics than the QDs.

This is consistent with the idea that electron transfer occurs at specific sites, where the
polymer chains and QDs are separated by short ligands, and therefore results in charge
pairs of small extent.
0.1
1
0.1 1 10 100
tert-butylthiol
pyridine
butylamine
0.1
1
a
b
Time (ps)
Deconvoluted TA signals (normalized)

150
5.3. Experimental
5.3.1. CdSe QD Synthesis and Ligand Exchange
Quantum dots were synthesized in a one-pot procedure using SeO
2
and
Cd(tetradecanoate)
2
precursors.
20
The native ligands in this synthesis are tetradecanoate
and oleate. The native ligands consist of long insulating alkyl chains; this requires
exchanging these native ligands for shorter ones to fabricate operable solar cells.
Therefore, ligand exchanges were performed to displace the insulating ligands (native
ligands, NL) with smaller organic ligands, namely tert-butylthiol (tBT),
18
pyridine (Py),
and n-butylamine (BA). The resulting ligand exchanged QDs are denoted CdSe(ligand)
in the text, where ligand = tBT, Py, or BA. The choice of this set of ligands was
motivated by previous literature that has demonstrated performance gains in hybrid
polymer:nanocrystal BHJ solar cells when these ligands are utilized.
1,2,18
The as-
synthesized CdSe(NL) QDs show a well defined first exciton peak at 580 nm,
corresponding to a diameter of 3.9 nm,
21
also confirmed by TEM.
5.3.2. Materials Characterization
Absolute absorbances of thin films were measured using an integrating sphere. Chopped
and filtered monochromatic light (250 Hz, 10 nm fwhm) from a Cornerstone 260 1/4 m
double grating monochromator (Newport 74125) was used in conjunction with an EG&G
7220 lock-in amplifier to perform spectral responsivity (or EQE) measurements.
TGA data were acquired for all QDs employed in this study using a TA Instruments TGA
Q50 instrument under flowing nitrogen. All samples heated at a ramp rate of 10 °C min
−1

151
up to 600 °C. Nanocrystal samples were prepared by drying out a minimum of 5 mg of
CdSe under flowing nitrogen for 1-24 h. Once dried, nanocrystal samples were lightly
crushed with a spatula before being loaded into an alumina crucible for TA analysis.
Electrochemical measurements were performed on a BASi potentiostat using a glassy
carbon working electrode, a platinum counter electrode, and a silver psuedo-reference
electrode. Cyclic voltammetry (CV) and differential pulse voltammetry (DPV) were
conducted using this three-electrode setup in a 0.1 M tetrabutylammonium
hexafluorphosphate/acetonitrile electrolyte solution with approximately 1 mM ferrocene
as an internal reference. The working electrode was prepared by drying out a thin film of
nanocrystals on the glassy carbon surface under nitrogen for several minutes prior to
submersion in the electrolyte solution. Multiple scans were run on every sample, and the
second CV and DPV scans were used for quantitative energy level calculations.
Photovoltaic devices were fabricated on patterned ITO (Thin Film Devices, 10 Ω cm
–2
).
Aluminum shot (Al; Alfa Aesar, 99.999%) was purchased and used as received. Prior to
film deposition, ITO was sequentially cleaned by sonication in tetrachloroethylene,
acetone, and isopropanol followed by 30 min of UV-ozone treatment. A layer of
PEDOT:PSS (Clevios PH 500, H.C. Starck, Inc., filtered through a 0.45 µm PTFE
syringe filter) was spun-cast onto the cleaned ITO and heated at 120 °C for 30 min under
vacuum. Polymer solutions of 20 mg mL
–1
were prepared in 1,2-dichlorobenzene by
dissolving under mild heating (40-50 °C) and filtering through a 0.45 µm PTFE syringe
filter. These polymer stock solutions were then mixed with pre-filtered 1,2-
dichlorobenzene dispersions of CdSe. Active layers were spun-cast in air onto annealed

152
PEDOT:PSS layers (700 rpm, 60 s) and allowed to dry for 25-40 min in a dark nitrogen
filled cabinet, forming films with thicknesses between 45-60 nm. Zinc oxide
nanocrystals dispersed in ethanol (20 mg mL
–1
) were spun-cast on top of the active layers
(400 rpm, 60 s), and then immediately annealed at 160 °C for 10 min under flowing
nitrogen. Finally, the devices were loaded into a high vacuum (~2 µTorr) thermal
deposition chamber (Angstrom Engineering) for Al deposition through a shadow mask at
a rate of 2 Å s
–1
. Device active areas were 4.3 mm
2
.
Hybrid PCPDTBT:CdSe BHJ films were obtained by blending PCPDTBT (purchased
from ONE-Material; the polydispersity index (PDI) is 2.4 and the molecular weight is M
w

= 38,000 Da) and the CdSe(ligand) QDs in 1,2-dichlorobenzene for several hours prior to
spin coating. The absorbance of the hybrid films shown in Figure 5.1b combines both
polymer and CdSe QDs features. Namely, the broad absorption bands from roughly 600–
850 nm can be solely attributed to PCPDTBT while the relatively sharp absorption bands
at 572 nm and 536 nm can be assigned to the first two exciton peaks of the CdSe QDs.
Absorption at wavelengths less than 500 nm results from a combination of features from
PCPDTBT and the CdSe QDs.

5.3.3. Transient Absorption
TA measurements were carried out using the output of a Coherent Legend Ti:sapphire
amplifier (1 kHz, 3.5 mJ, 35 fs), as described in detail elsewhere.
5,22,23
The amplifier
output was used as the 800 nm pump pulse. White light supercontinuum probe pulses

153
were polarized perpendicular to the pump pulse and generated using a rotating CaF
2

window. The probe pulse was dispersed by an Oriel MS1271 spectrograph onto a 256
pixel Si photodiode array. To obtain satisfactory signal-to-noise on the optically thin
samples, transient spectra were measured for typical pump fluences of 90 mJ cm
-2
.
Samples were translated perpendicular to the path of the pump and probe to prevent
photodamage.
The time resolution of the system (i.e. the pump pulse duration τ
pump
) was found to be 45
fs based on cross-correlation between the pump and probe in two 1 mm quartz substrates
(similar to encapsulated samples). The cross-correlation response was fitted to the
following equation:
22

Eq. 1
where ΔA(ω
probe
, t) is the experimentally measured pump-induced change in optical
density, ω
probe
is the spectral component of the probe, t(ω
probe
) is the frequency-dependent
pump-probe time delay, τ
pump
and τ
probe
are the pump and probe pulse durations, β is the
chirp rate and t
0
(ω
probe
) is the time-zero function, which is fitted at the peak of the cross-
phase modulation signals.
As detailed in our previous work
6
, the QD spectral bleach provides a way to quantify the
yield of electron transfer at the donor/acceptor interface. In order to eliminate the
polymer bleach contribution in this region, the hybrid TA spectra can be decomposed at
€
ΔA(ω
probe
,t)≈exp −
t
2
(ω
probe
)
τ
pump
2
'
(
)
)
*
+
,
,
×sin
1
2βτ
pump
2
−
t
2
(ω
probe
)
βτ
pump
2
−
t
0
(ω
probe
)t(ω
probe
)
βτ
pump
2
τ
probe
2
'
(
)
)
*
+
,
,

154
each time by a linear combination of the CdSe(tBT) QD and of the neat PCPDTBT TA
spectra. For each time step, the observed data d
obs
(corresponding to the hybrid
spectrum) can be compared to the data d
pred
predicted by a linear combination model with
parameters (a,b):

where DA
i
is the TA signal of sample i. The comparison between d
obs
and d
pred
is made
for a large number of (a,b) parameter couples. The most likely model parameter couple
(a
mean
, b
mean
) is extracted using Bayes’ theorem and a Metropolis algorithm.
22
It is
possible to fit the data, within experimental uncertainties, with various (a,b) couples;
therefore, the standard deviations of the parameters (a
s.d
, b
s.d.
) reflect the dispersion of
(a,b) couples that fit the experimental data. This procedure gives access to rigorous,
statistically relevant error bars. In this decomposition, the parameter b represents the
contribution of the QD electrons to the hybrid spectrum.
Here, we used a doorway function for the prior probability density function on model
parameters a and b (0,0)<(a,b)<(2,2): our only a priori is that the contributions of the
neat QD and neat polymer bleaches cannot be negative and cannot diverge. The
likelihood is given by , where e is the experimental uncertainty.

The posterior probability density function is given by the product of the prior probability
€
d
obs
=ΔA
hybrid
d
pred
= a×ΔA
polymer
+b×ΔA
QD
(t = 75fs)
€
exp −
1
2
d
obs
−d
pred ( )
2
e
2
#
$
%
%
&
'
(
(

155
density function and the likelihood. A Metropolis-Hastings algorithm (Matlab Statistical
Toolbox) is used to scan the parameter space (20,000 samples, 1,000 burnt samples) and
obtain the most likely model parameters (a
mean
, b
mean
), i.e. the ones that maximize the
posterior probability density function, as well as their standard deviations (a
std
, b
std
).

5.4. Conclusions
In conclusion, the finding that the ligand coverage determines the charge transfer
efficiency underlines not only the need for surface chemistry modification from the
native ligands, but also the requirement for full displacement of the native ligands in
order to maximize the number of ultrafast transfer sites. After electron transfer, the
electron and hole are initially separated by the ligand length, which determines the
intensity of their Coulomb attraction. Without a driving force for further separation of
the charge pair, geminate recombination rapidly reduces the charge pair population.
5

Using spectral deconvolution, we have investigated recombination dynamics of these
charge pairs, and we found that the decay dynamics are quite similar for all short ligands
studied herein (i.e., tBT, Py, BA). Finally, the effect of the ligands is greater on the EQE
than on the CT efficiency. This is most likely due to improved charge transport for tBT-
based hybrid solar cells, as these shorter ligands have been shown to improve the charge
mobility in QD networks.
19

156
5.5. References
(1) Greaney, M. J.; Araujo, J.; Burkhart, B.; Thompson, B. C. ; Brutchey, R. L. Novel
Semi-Random and Alternating Copolymer Hybrid Solar Cells Utilizing CdSe
Multipods as Versatile Acceptors. Chem. Commun. 2013, 49, 8602–8604.
(2) Zhou, R.; Stadler, R.; Xie, D.; Cao, W.; Zheng, Y.; Yang, Y.; Plaisant, M.;
Holloway, P. H.; Schanze, K. S.; Reynolds, J. R.; Xue, J. Enhancing the
Efficiency of Solution-Processed Polymer:Colloidal Nanocrystal Hybrid
Photovoltaic Cells Using Ethanedithiol Treatment. ACS Nano 2013, 7, 4846–
4854.
(3) Liu, Z.; Sun, Y.; Yuan, J.; Wei, H.; Huang, X.; Han, L.; Wang, W.; Wang, H. ;
Ma, W. High-Efficiency Hybrid Solar Cells Based on Polymer/PbS
x
Se
1-x

Nanocrystals Benefiting from Vertical Phase Segregation. Adv. Mater. 2013, 25,
5772–5778.
(4) Greaney, M. J. ; Brutchey, R. L. Ligand Engineering in Hybrid
Polymer:Nanocrystal Solar Cells. Materials Today 2015, 18, 31-38.
(5) Kraabel, B.; Lee, C.H.; McBranch, D.; Moses, D.; Sariciftci, N.S. ; Heeger A.J.
Ultrafast Photoinduced Electron Transfer in Conducting Polymer-
Buckminsterfullerene Composites. Chem. Phys. Lett. 1993, 213, 389–394.
(6) Couderc, E.; Greaney, M. J.; Brutchey, R. L.; Bradforth, S. E. Direct
Spectroscopic Evidence of Ultrafast Electron Transfer from a Low Band Gap
Polymer to CdSe Quantum Dots in Hybrid Photovoltaic Thin Films. J. Am.
Chem. Soc. 2013, 135, 18418−18426.
(7) Morgenstern, F. S. F.; Rao, A.; Böhm, M. L.; Kist, R. J. P.; Vaynzof, Y. ;
Greenham N. C. Ultrafast Charge- and Energy-Transfer Dynamics in Conjugated
Polymer: Cadmium Selenide Nanocrystal Blends. ACS Nano, 2014, 8, 1647–
1654.
(8) Barkhouse, D. A. R.; Pattantyus-Abraham, A. G.; Levina, L. ; Sargent, E. H.
Thiols Passivate Recombination Centers in Colloidal Quantum Dots Leading to
Enhanced Photovoltaic Device Efficiency. ACS Nano, 2008, 2, 2356–2362.
(9) Liu, Y.; Gibbs, M.; Puthussery, J.; Gaik, S.; Ihly, R.; Hillhouse, H. W.; Law, M.
Dependence of Carrier Mobility on Nanocrystal Size and Ligand Length in PbSe
Nanocrystal Solids. Nano Lett. 2010, 10, 1960–1969.
(10) Bakulin, A. A.; Rao, A.; Pavelyev, V. G.; van Loosdrecht, P. H. M.;
Pshenichnikov, M. S.; Niedzialek, D.; Cornil, J.; Beljonne, D.; Friend, R. H. The
Role of Driving Energy and Delocalized States for Charge Separation in Organic
Semiconductors. Science 2012, 335, 1340–1344.

157
(11) Vandewal, K.; Albrecht, S.; Hoke, E. T.; Graham, K. R.; Widmer, J.; Douglas, J.
D.; Schubert, M.; Mateker, W. R.; Bloking, J. T.; Burkhard, G. F.; Sellinger, A.;
Fréchet, J. M. J.; Amassian, A.; Riede, M. K.; McGehee, M. D.; Neher, D.;
Salleo, A. Efficient Charge Generation by Relaxed Charge-transfer States at
Organic Interfaces. Nature Mater. 2014, 13, 63–68.
(12) Coffey, D. C.; Larson, B. W.; Hains, A. W.; Whitaker, J. B.; Kopidakis, N.;
Boltalina, O. V.; Strauss, S. H.; Rumbles, G. An Optimal Driving Force for
Converting Excitons Into Free Carriers in Excitonic Solar Cells. J. Phys. Chem.
C 2012, 116, 8916–8923.
(13) Nagaoka, H.; Colbert, A. E.; Strein, E.; Janke, E. M.; Salvador, M.; Schlenker, C.
W. and Ginger, D. S. Size-Dependent Charge Transfer Yields in Conjugated
Polymer/Quantum Dot Blends. J. Phys. Chem. C 2014, 118, 5710–5715.
(14) Gao, F.; Li, Z.; Wang, J.; Rao, A.; Howard, I. A.; Abrusci, A.; Massip, S.;
McNeill, C. R.; Greenham, N. C. Trap-Induced Losses in Hybrid Photovoltaics,
ACS Nano 2014, 8, 3213–3221.
(15) Meulenberg, R. W.; Lee, J. R.I.; Wolcott, A.; Zhang, J. Z.; Terminello, L. J.; van
Buuren, T. Determination of the Exciton Binding Energy in CdSe Quantum Dots,
ACS Nano 2009, 3, 325–330.
(16) Greaney, M. J.; Das, S.; Webber, D. H.; Bradforth, S. E. and Brutchey, R. L.
Improving Open Circuit Potential in Hybrid P3HT:CdSe Bulk Heterojunction
Solar Cells via Colloidal tert-Butylthiol Ligand Exchange. ACS Nano 2012, 6,
4222−4230.
(17) Hwang, I. W.; Soci, C.; Moses, D.; Zhu, Z.; Waller, D.; Gaudiana, R.; Brabec, C.
J.; Heeger, A. J. Ultrafast Electron Transfer and Decay Dynamics in a Small
Band Gap Bulk Heterojunction Material. Adv. Mater. 2007, 19, 2307−2312.
(18) Klimov, V. In Handbook on Nanostructured Materials and Nanotechnology;
Nalwa, H., Ed.; Academic Press: San Diego, CA, 1999; Volume 4, p 451.
(19) Webber, D. H.; Brutchey, R. L. Ligand Exchange on Colloidal CdSe Nanocrystals
Using Thermally Labile tert-Butylthiol for Improved Photocurrent in Nanocrystal
Films. J. Am. Chem. Soc. 2012, 134, 1085-1092.
(20) Chen, O.; Chen, X.; Yang, Y.; Lynch, J.; Wu, H.; Zhuang, J.; Cao, Y. C.
Synthesis of metal-selenide nanocrystals using selenium dioxide as the selenium
precursor, Angew. Chem. Int. Ed. 2008, 47, 8638−8441.
(21) Jasieniak, J.; Smith, L.; van Embden, J.; Mulvaney, P. and Califano, M. Re-
examination of the Size-Dependent Absorption Properties of CdSe Quantum
Dots. J. Phys. Chem. C 2009, 113,19468–19474.

158
(22) Rivera, C. A.; Winter, N.; Harper, R. V.; Benjamin, I.; Bradforth, S. E. The
Dynamical Role of Solvent on the ICN Photodissociation Reaction: Connecting
Experimental Observables Directly with Molecular Dynamics Simulations, Phys.
Chem. Chem. Phys. 2011, 13, 8269–8283.
(23) Roberts, S. T.; Schlenker, C. W.; Barlier, V.; McAnally, R. E.; Zhang, Y.;
Mastron, J. N.; Thompson, M. E.; Bradforth, S. E. Observation of Triplet Exciton
Formation in a Platinum-Sensitized Organic Photovoltaic Device, J. Phys.
Chem. Lett. 2011, 2, 48–54.

159
Chapter 6. Controlling the Trap State Landscape of Colloidal CdSe
Nanocrystals with Cadmium Halide Ligands*
*Published in Chem. Mater. 2015, 27, 744–756.

6.1. Abstract
We developed a simple and robust colloidal route for the installation of CdX
2
(X = Cl,
Br, I) ligands on the surface of CdSe nanocrystals, which effectively displace the native
ligands and form stable suspensions. After colloidal ligand exchange, these nanocrystals
can be easily solution cast into nanocrystal films. Photoelectrochemical measurements on
solution-cast nanocrystal films reveal a striking influence of surface cadmium halide
on photocurrent response, with mildly annealed, CdCl
2
-treated CdSe
nanocrystals showing the greatest enhancement in photocurrent to above band
gap illumination. The strong dependence of photoresponse on surface halide is thought to
result from ligand-induced changes to the electronic structure of the nanocrystal samples.
We arrive at this conclusion using a combination of ultrafast transient absorption, time-
resolved photoluminescence, and surface photovoltage spectroscopies, which are being
applied together for the first time to investigate nanocrystal trap states. From these
measurements, we establish a trend for ligand-related sub-band gap states that accounts
for electron and hole trapping at the nanocrystal surface. The nature of the electron and
hole traps in the nanocrystal films are dependent on the thermal history of the sample as
well as the specific halide surface treatment employed. After subjecting the nanocrystal
films to mild thermal annealing, we find evidence that suggests a drastic reduction in
electron trap states. Additionally, depending on the surface halide treatment employed,

160
the energy of the hole trap states varies, with CdCl
2
treatment resulting in energetically
shallow hole trap states, and CdBr
2
and CdI
2
treatments leading to much deeper hole
traps. Thus, judicious choice of cadmium halide surface treatment can be used to
manipulate the trap state landscape of these ligand exchanged CdSe nanocrystals.

6.2. Introduction
Developing schemes to manipulate the electronic and optoelectronic properties of
semiconductor nanocrystals will be a critical enabler for the widespread deployment of
nanocrystal-based technologies. An inherently high surface- area-to-volume ratio requires
that a special emphasis be placed on understanding how to tune nanocrystal surfaces in
order to rationally control their resulting properties. High-performance nanocrystal-based
optoelectronic devices require the facile transport of charges through percolation
networks, yet the nanocrystal surface species often present barriers for such transport.
1−3

The native ligand species present on the vast majority of colloidal semiconductor
nanocrystals are usually good surface passivants while also enabling excellent
dispersibility in nonpolar solvents; however, these native ligands are often insulating,
long-chain aliphatic compounds (e.g., C
14
− C
18
fatty acids such as myristic or oleic acid)
and therefore can have a deleterious effect on electrical transport properties in
nanocrystalline solids.
4
To mitigate this effect, widespread efforts have been made to
exchange native ligands with new ligands that simultaneously enhance nanocrystal
coupling, passivate surface states, and (ideally) maintain the colloidal stability of
nanocrystal dispersions for efficient solution processing.

161
The development of rational nanocrystal surface chemistries requires a basic
understanding of ligand chemistry and surface stoichiometry. Generally, II-VI and IV-VI
nanocrystals such as CdE and PbE (E = S, Se, Te) contain metal-rich surfaces, and thus
necessitate an excess of anionically bound ligands to maintain overall charge balance.
5-9

This fact has guided the design of ligand exchange strategies that tend to focus on the
utilization of small species having the ability to displace native ligands as neutral or
anionic moieties. In an effort to organize the various binding motifs available at
nanocrystal surfaces, Owen et al. have adopted Green’s Covalent Bond Classification
scheme
10
for categorizing ligands to balance charge and fill out the nanocrystal
coordination sphere.
11
For convenience, we have also chosen to utilize this scheme in
which neutral and anionic Lewis basic ligands are referred to as L- or X-type,
respectively, while neutral or cationic Lewis acidic ligands are classified as Z-type. The
possibilities for ligation at nanocrystal surfaces are dependent upon surface
stoichiometry, but in most cases, saturated surface passivation requires a combination of
L-, X-, and/or Z-type ligands.
Relevant ligand types that have received increasing attention include extremely small
species such as molecular chalcogenidometalates (e.g., Cu
7
S
4
–
and Sn
2
S
6
4–
),
12-15

thermally labile organic ligands (e.g., tert-butylthiol),
16
and small, simple inorganic
anions (e.g., SCN
–
, BF
4
–
, S
2–
, Cl
–
).
17-25
While these examples are promising for the
future development of solution-processed nanocrystalline electronics, potential
drawbacks include the use of pyrophoric phosphines
11,22
demanding the use of an inert
atmosphere, the use of toxic and hazardous hydrazine as a solvent/reagent,
12-15
as well as
limited colloidal stability of the ligand-exchanged species
17
that necessitates immediate

162
device processing after ligand exchange. Nonetheless, the aforementioned ligands have
been very successfully leveraged for the purpose of producing nanocrystal solids with
tailored electronic properties such as enhanced carrier mobility or improved
photoresponse.
24-27

Aside from the predictable improvement in transport properties stemming from decreased
interparticle separation associated with sterically reduced ligand shells, much less is
understood about the influence of such surface ligands on the electronic structure of the
nanocrystals.
28,29
It is quite possible that the electronic consequences of surface
engineering with extremely small ligands will turn out to be equally if not more important
for optoelectronic applications that the associated improvements in spatial coupling. For
example, halide ion treatments have been shown to influence the electronic structure of
IV-VI nanocrystal-based devices by reducing the density and depth of mid-gap trap states
in PbS colloidal nanocrystal devices.
30-32
However, similar types of ligand treatments
have been shown to reduce the photoluminescence quantum yield of CdSe nanocrystals,
33

much like some organic and inorganic ligands such as phenylchalcogenolates
34
on CdSe
nanocrystals and molecular chalcogenidometalate ligands on CdSe-CdS core-shell
nanocrystals.
29
The disparity in ligand effect for different types of nanocrystals
highlights one rationale for pursuing systematic studies into the relationship between
surface ligands and nanocrystal electronic structure. Chemically and spatially similar
classes of ligands, such as the halides, are ideal systems to investigate for such a study,
since the ligand-induced spatial and electronic effects on optoelectronic performance can
be reasonably separated from one another. A recent example from Talapin and
coworkers describes using methylammonium halides or sodium halides in a two-phase

163
ligand exchange as a means of replacing native ligands for halide ions, which leads to
nanocrystal films with enhanced FET mobilities.
35
The improvement in nanocrystal
transport properties was ascribed to the reduced interparticle separation, although
changes in nanocrystal electronic structure stemming from the new halide ligands were
not investigated. It is possible that both ligand-induced changes in nanocrystal sterics
and electronics were responsible for the observed mobility changes.
In light of recent work that highlights the potential benefits of halide surface ligands (e.g.,
high quantum yields and large electron mobilities
11,22,24,33,35
), we sought to develop a
robust colloidal ligand exchange route providing passivation of both cadmium and
selenium surface sites, while facilitating the installation of surface halides. Additionally,
we were interested in conducting a systematic investigation into the influence of different
surface halides on the nanocrystal electronic structure by tracking changes in carrier
dynamics and trap state energetics associated with related, but chemically distinct,
inorganic ligands. In an effort to address our interests, we developed an extremely simple
and resilient route for the colloidal ligand exchange of native ligands (i.e., tetradecanoate,
tetradecanoic acid, oleate, and oleic acid) with CdX
2
(X = Cl, Br, I) ligands. Optimized
ligand exchange enables the nearly quantitative replacement of native ligands with CdBr
2

and CdI
2
and semi-quantitative native ligand replacement with CdCl
2
. We utilize this
simple platform to probe the influence of colloidal cadmium halide treatments on the
optoelectronics and photophysics of CdSe nanocrystals, and we relate our findings to the
associated trap state landscape, which is strongly dictated by the nature of the nanocrystal
surface ligand. This is the first study where time-resolved transient absorption (TA)
spectroscopy has been applied together with photoelectrochemical (PEC) measurements

164
and surface photovoltage (SPV) spectroscopy to elucidate the trap state landscape of
semiconductor nanocrystals.

6.3. Results and Discussion
6.3.1. Nanocrystal Synthesis and Characterization
The CdSe nanocrystals were prepared using a modified version of a previously published
protocol.
36
Aliquots were taken during the course of the reaction and analyzed by
absorption spectroscopy to monitor nanocrystal growth by tracking the evolution of the
first exciton peak (see Figure 6.1).

Figure 6.1. Optical absorption spectra of aliquots (in 1-octadecene) taken over the course of the
reaction representing CdSe nanocrystal growth, as evidenced by a continually red-shifting first
exciton peak. The emission spectrum is shown for the final aliquot taken.

The purified CdSe(native ligands) nanocrystals used in this study have a zinc blende
structure (see Figure 6.2) with a first exciton peak at 596 nm, and exhibit sharp near-band

165
edge emission with a maximum intensity at 608 nm. Using the E
1s
absorption maximum
and the empirical CdSe sizing equation established by Jasieniak et al., the CdSe(native
ligands) nanocrystals should have a diameter of 4.48 nm.
37
Indeed, size analysis using
transmission electron microscopy (TEM) gave an average nanocrystal diameter of 4.54 ±
0.28 nm, demonstrating a particle size distribution of less than 7%.

Figure 6.2. XRD pattern for a CdSe(native ligand) nanocrystal film on silicon.

6.3.2. Ligand Exchange and Characterization
Portions of CdSe(native ligands) were ligand exchanged with CdX
2
(X = Cl, Br, I) and
butylamine, as described in detail in the Experimental Section. In general, CdSe(native
ligands) nanocrystals were treated with solutions of CdX
2
in methanol, which caused
flocculation during the initial treatment. After separation of the supernatant, the
cadmium halide-treated nanocrystals were easily dispersed in several polar chlorinated
solvents (i.e., 1,2-dichlorobenzene, chlorobenzene, and chloroform) after the first CdX
2

166
treatment, but chlorobenzene was chosen as a preferred solvent for optimal film-casting
conditions. Several CdX
2
treatments were performed for each cadmium halide type, with
the exact number of iterations required for optimal ligand exchange being different for
each halide (see Experimental, Section 6.4.2). It was found that CdI
2
and CdBr
2
were
most effective for native ligand removal, followed CdCl
2
. Both CdI
2
and CdBr
2
were
able to quantitatively displace the native ligands from the CdSe(native ligand)
nanocrystals; however, in the case of CdCl
2
, quantitative removal of native ligands was
never achieved.
To increase the efficacy of CdX
2
ligand exchange, we tried enhancing the colloidal
stability of the partially treated nanocrystals by addition of butylamine, which is known
to act as an L-type ligand on the surface of metal chalcogenide nanocrystals.
22,38,39

Partial surface ligation with butylamine leads to greatly enhanced colloidal stability as
evidenced by very slow or complete prevention of nanocrystal flocculation upon addition
of methanolic CdX
2
. Preventing nanocrystal agglomeration is especially critical for
efficient ligand exchange since agglomeration upon addition of methanolic solutions of
CdX
2
limits exposure of the nanocrystal surface to the incoming ligands. Control
experiments in which butylamine alone is used for ligand exchange with CdSe(native
ligand) produce partially exchanged nanocrystals that still have a significant amount of
X-type native ligand as suggested by thermogravimetric analysis (TGA),
1
H NMR, and
FT-IR of the resulting CdSe(butylamine) nanocrystals (see Figures 6.3-6.5).
The explanation for the inability of butylamine to quantitatively displace native ligands is
straightforward when taking charge balance into consideration. Butylamine can only

167
bind to surfaces as a neutral ligand so it is limited to displacing neutrally bound L-type
carboxylic acids, and is unable to displace anionically bound X-type ligands in the
absence of surface reconstruction through the etching of Cd(carboxylate)
2
from the
surface. The partial removal of native ligands upon butylamine exchange produces
nanocrystals with a mixed ligand shell that exhibit excellent colloidal stability in medium
polarity solvents, such as chlorobenzene and methanol. By using butylamine after
several CdX
2
treatments, the colloidal stability of the nanocrystals after addition of
methanolic solutions of CdX
2
is greatly enhanced, with stable suspensions observed for
over one year when stored in the dark.

Figure 6.3. TGA thermograms of (a) CdSe(butylamine), (b) CdSe(CdI
2
), (c) CdSe(CdBr
2
), (d)
CdSe(CdCl
2
), and (e) CdSe(native ligand).

168

Figure 6.4. Normalized
1
H NMR spectra for (a) CdSe(native ligand), (b) CdSe(butylamine), (c)
CdSe(CdCl
2
), (d) CdSe(CdBr
2
), and (e) CdSe(CdI
2
) after digestion in aqua regia and extraction
with d
6
-benzene. The spectra demonstrate the removal of NL aliphatic protons after ligand
exchange as evidenced by the reduced aliphatic proton signal from 0.8-2.2 ppm. All spectra are
normalized to the residual benzene solvent peak at 7.16 ppm (not shown for clarity). The
remaining native ligands in the CdSe(butylamine) and CdSe(CdCl
2
) nanocrystal samples are
visible as small signals in the aliphatic proton range. The aliphatic proton signals for the
CdSe(CdBr
2
) and CdSe(CdI
2
) samples are almost completely gone, which supports our
assessment of nearly quantitative CdX
2
ligand exchange. Proton signal from butylamine is not
observable since amine protonation during the acid digestions prevents extraction into the organic
d
6
-benzne layer. The asterisked triplet around 0.7 ppm is from residual pentane, which was used
as an anti-solvent for CdSe(CdBr
2
) and CdSe(CdI
2
) nanocrystals.

169

Figure 6.5. Quantitative FT-IR spectra of unannealed (dashed lines) and 250 °C annealed (solid
lines) nanocrystal samples. From bottom to top: CdSe(native ligand) (green), CdSe(butylamine)
(pink), CdSe(CdCl
2
) (red), CdSe(CdBr
2
) (orange), and CdSe(CdI
2
) (blue). An internal standard,
Fe
4
[Fe(CN)
6
]
3
with ν(C≡N) at 2090 cm
–1
, was mixed into a KBr matrix and used for pellet
preparation so that spectra could be quantitatively compared. All thermal annealing steps were
performed prior to mixing with the KBr/internal standard matrix. From integration of the ν(C-H)
signals between 2750-3050 cm
–1
, it was found that thermal annealing at 250 °C reduces the ν(C-
H) intensity by 44.6, 74.3, and 51.6% for CdSe(CdCl
2
), CdSe(CdBr
2
), and CdSe(CdI
2
),
respectively.

The thin film absorption spectra of ligand-exchanged nanocrystals are given in Figure
6.6. The absorption spectra of the as-cast nanocrystal films are very similar to one
another after ligand exchange, with a slight blue shift of the first exciton peak relative to
the CdSe(native ligand) sample, as well as decreased absorption intensity of the second
exciton peak. The former may be explained by a surface polarity and/or polarizability
changes (e.g., native ligand relative to polar cadmium halide), whereas the cause for the
decreased absorption strength of second exciton feature is less clear. This phenomenon
has been observed in prior studies of phosphine- and halide-passivated CdSe
nanocrystals, and the authors suggested that the spectral differences were possibly related
to nanocrystal surface reconstruction.
11,22

170

Figure 6.6. Absorption spectra of as-cast CdSe nanocrystal thin films. The spectra were
collected inside of an integrating sphere to account for scattering and/or reflections.

TGA of the CdSe(native ligand) nanocrystals showed the native ligand content to
comprise ca. 33% of the overall mass content of the sample with a high temperature mass
loss event onset at 350 °C (Figure 6.3) that we attribute to decomposition of native
ligands. After treatment with CdX
2
, the mass loss around 350 °C is eliminated for
CdSe(CdBr
2
) and CdSe(CdI
2
) nanocrystal samples, and it is significantly reduced for
CdSe(CdCl
2
) samples.
1
H NMR spectroscopy and XPS experiments were conducted on
all samples to further probe the ligand exchange chemistry. Nanocrystal samples for
NMR analyses were digested in aqua regia, extracted with benzene, and dried with
MgSO
4
before obtaining
1
H NMR spectra (see Figure 6.4).
1
H NMR spectra for all
samples were normalized relative to residual benzene at 7.16 ppm for the purpose of

171
semi-quantitative comparison. All ligand exchanged nanocrystal samples showed
significant reductions in the aliphatic proton signals associated with tetradecanoate and
oleate native ligands. Consistent with TGA interpretation, the CdSe(CdBr
2
) and
CdSe(CdI
2
) nanocrystal samples appear to undergo the most efficient ligand exchange
based upon the greatest reduction in aliphatic proton signal from 0.8-2.2 ppm. The
CdSe(butylamine) and CdSe(CdCl
2
) samples exhibited a reduced proton signal relative to
the as-prepared CdSe nanocrystals, albeit not as extensive as CdSe(CdBr
2
) and
CdSe(CdI
2
) nanocrystal samples. The presence of halide was positively confirmed using
XPS measurements, which showed halide signals with the expected binding energies and
peak separations for each of the CdSe(CdX
2
) nanocrystal samples (Figure 6.7). The
binding energies were in agreement with literature values with core energy levels of
618.9,
40
69.0, and 198.7 eV
41
measured for I 3d
5/2
, Br 3d
5/2
, and Cl 2p
3/2
, respectively, for
the corresponding CdSe(CdX
2
) nanocrystals.

Figure 6.7. High-resolution XPS spectra of CdSe(CdX
2
) nanocrystal samples. (a) Cl 2p
spectrum for CdSe(CdCl
2
), (b) Br 3d spectrum for CdSe(CdBr
2
), and (c) I 3d spectrum for
CdSe(CdI
2
) nanocrystals.

172
While the combination of TGA, NMR, and XPS experiments provide evidence for the
efficient removal of native ligand and installation of halide ligands, conclusive proof
regarding the homogenous distribution of halide species was obtained by elemental
mapping. Using spatially resolved STEM-EDX as a probe of the elemental composition
in nanocrystal samples, we observed uniform distribution of halide in ensembles of
CdSe(CdX
2
) nanocrystals (see Figures 6.8 and 6.9), with a representative set of elemental
maps given for a grouping of CdSe(CdI
2
) nanocrystals in Figure 6.10. These results
suggest that the halide is indeed bound to the nanocrystal surface.

Figure 6.8. STEM-EDX maps of CdSe(CdCl
2
) nanocrystal ensemble demonstrating a uniform
distribution of chloride throughout the nanocrystals. (a) STEM field of view; (b) Cl (K-line); (c)
Se (K-line); (d) Cd (L-line). The scale bars are 50 nm.

Figure 6.9. STEM-EDX maps of CdSe(CdBr
2
) nanocrystal ensemble demonstrating a uniform
distribution of bromide throughout the nanocrystals. (a) STEM field of view; (b) Se (K-line); (c)
Br (K-line); (d) Cd (L-line). The scale bars are 35 nm.

173

Figure 6.10. STEM-EDX elemental map of an ensemble of CdSe(CdI
2
) nanocrystals showing
uniform distribution of iodide throughout sample. (a) STEM images of a cluster of CdSe(CdI
2
)
nanocrystals; (b) Cd (L-line) net intensity elemental map; (c) I (L-line) net intensity elemental
map; (d) Se (L-line) net intensity elemental map. All scale bars are 20 nm.

As would be expected for surface ligation by Cd
2+
species, the Cd:Se ratio increased after
all CdX
2
treatments; multiple TEM-EDX measurements taken for each sample supported
an increasing Cd:Se ratio, which was also confirmed by ICP-OES analysis (Table 6.1).

Table 6.1. Nanocrystal composition as determined by TEM-EDX and ICP-OES.

Nanocrystal Sample at% X
(by EDX)
at% Cd
(by EDX)
at% Se
(by EDX)
Cd:Se
EDX ICP
CdSe(native ligands) n/a 51.7 48.3 1.07 ± 0.11 1.18 ± 0.02
CdSe(butylamine) n/a 51.1 48.8 1.04 ± 0.07 1.06 ± 0.04
CdSe(CdCl
2
) 3.3 52.5 45.2 1.16 ± 0.15 1.24 ± 0.03
CdSe(CdBr
2
) 6.9 50.7 42.4 1.20 ± 0.11 1.36 ± 0.06
CdSe(CdI
2
) 10.6 49.2 40 1.23 ± 0.18 1.51 ± 0.05

The increased Cd:Se ratio suggests binding of Cd
2+
species to the nanocrystal surface at
under-coordinated Se-surface sites through a Z-type Lewis donor (Se) – Lewis acceptor

174
(Cd) interaction. This type of surface ligation has been previously reported by Owen et
al. for Cd(O
2
CR)
2
species on the surface of CdE and PbE (E = S, Se) nanocrystals.
11
As
illustrated in Scheme 6.1, we believe that multiple surface binding motifs are occurring in
which the aforementioned Z-type ligation leads to passivation of Se surface sites with
Cd
2+
species that must be charge balanced, likely by two additional halides.

Scheme 6.1. Representative ligand exchange reaction that depicts L-, X-, and Z-type binding
modes to the CdSe nanocrystal surface. For clarity, only one example of each type of binding
mode is shown, and the nanocrystals and ligands are not drawn to scale.

The X-type ligation of halide ions onto Cd surface sites results in the efficient
displacement of native ligand species (vide infra). Finally, L-type binding of butylamine
to charge-balanced Cd surface sites fills out the coordination sphere and provides
colloidal stability in medium polarity solvents. We would like to note that our depiction
of the monodentate X-type interaction between carboxylate native ligands and CdSe is

175
partially based upon previously published work by Hens and coworkers in which they
used multi-dimensional
1
H NMR to confirm the binding motif and stoichiometry for oleic
acid ligands to the surface of CdSe nanocrystals.
42

6.3.3. Thermal Treatment
Nanocrystal solids are often subjected to thermal annealing after being cast into films for
the purpose of driving off weakly bound surface ligands and altering electronic properties
of the nanocrystal film such as conductivity and photoresponse. This type of treatment
can be mild enough to prevent particle sintering and gross grain growth, yet still enable
some degree of surface reconstruction.
22,43
As a side note, when choosing a thermal
profile for nanocrystal films on transparent conducting oxide (TCO) substrates, it is
important to consider the possibility for migration of the TCO components into the
nanocrystal film, which can significantly change the optoelectronic performance and
complicate interpretation of results (e.g., ITO is known to undergo indium diffusion
above 250 °C)
44

The effect of thermal annealing on the optical properties of CdSe(CdX
2
) nanocrystals was
investigated using absorption and FT-IR spectroscopies. Nanocrystal thin films were
made by spin-coating the nanocrystals from chlorobenzene dispersions onto glass
substrates, except for CdSe(native ligand) which was not dispersible in chlorobenzene
and was cast from toluene. FT-IR demonstrates significant reduction in ν(C-H)
stretching intensities for CdSe(CdX
2
) samples, consistent with removal of the oleate and
myristate native ligands (see Figure 6.4). All CdSe(CdX
2
) nanocrystal samples yielded

176
optically transparent and homogeneous films with excellent film quality. As previously
stated, the as-cast nanocrystal films exhibit very similar absorption features, with the
major difference appearing as a 2-3 nm (7-9 meV) blue shift in the first exciton peak for
all cadmium halide-treated nanocrystal samples relative to the CdSe(native ligand)
nanocrystals. More interesting are the changes observed after mild heat treatment at 250
°C under nitrogen for 10 min. The CdSe(CdX
2
) nanocrystal films all exhibit an apparent
reduction in quantum confinement as evidenced by drastic red shifts and broadening of
the first exciton peaks, as shown in Figure 6.11.

Figure 6.11. Absorption spectra of CdSe nanocrystal thin films after annealing at 250 °C under
nitrogen for 10 min. The spectra were collected inside an integrating sphere to account for
scattering and/or reflections, which suggests that the observed red shifts are due to intrinsic
changes in the properties of the nanocrystal films.

177
Compared to red shifts of 8 and 10 meV observed for the heat-treated CdSe(native
ligand) or CdSe(butylamine) nanocrystal thin films, respectively, heat-treated
CdSe(CdX
2
) nanocrystal thin films show significantly larger red shifts in the first exciton
feature of 35, 42, and 55 meV for CdSe(CdI
2
), CdSe(CdBr
2
) and CdSe(CdCl
2
),
respectively.
Reduction in the quantum confinement observed in nanocrystal absorption spectra can be
caused by several factors, including increased electronic coupling between nanocrystal
and/or particle sintering and grain growth. We initially hypothesized that the large red
shift was the combined result of particle sintering and enhanced nanocrystal coupling.
We cannot rule out the possibility of some degree of nanocrystal necking or fusing, but
we have been unable to observe this experimentally. TEM, XRD, and SEM analyses
before and after annealing do not reveal evidence of grain growth (Figures 6.12-6.14).

Figure 6.12. TEM images of as-cast (top) and 250 °C annealed CdSe nanocrystals (bottom). (a)
CdSe(native ligands), (b) CdSe(CdCl
2
), (c) CdSe(CdBr
2
), (d) CdSe(CdI
2
), and (e)
CdSe(butylamine) nanocrystals. The annealed samples do not show obvious signs of nanocrystal
growth. Nanocrystal sizes and distributions were calculated by counting at least 50 nanocrystals.

178

Figure 6.13. XRD patterns of CdSe nanocrystal thin films comparing the (111) peak before
(color) and after (black) 250 °C annealing. From top left to bottom right: (a) CdSe(native ligand),
(b) CdSe(CdCl
2
), (c) CdSe(CdBr
2
), and (d) CdSe(CdI
2
).

Figure 6.14. SEM top-down images of CdSe(CdCl
2
) nanocrystal films before (left side) and after
(right side) 250 °C annealing. The images are shown at magnifications of 2500x in (a) and (d),
25000x in (b) and (e), and 250000x in (c) and (f). A contaminant particle is shown in the lower
magnification panels to demonstrate proper focus. The images show no obvious grain growth
upon heat treatment, although the appearance of a crack is seen after thermal annealing.

179

Thin film XRD patterns do not demonstrate line narrowing that would be expected for
particle growth and/or sintering, although minor changes to the grain structure like mild
nanocrystal necking or fusing would be difficult to detect using wide angle powder XRD
techniques (see Figure 6.13). Furthermore, no obvious change in average particle size or
size distribution was observed by TEM, with average nanocrystal diameters of 4.54 ±
0.28, 4.73 ± 0.32, 4.80 ± 0.43, and 4.81 ± 0.48 nm measured prior to thermal treatment
for CdSe(native ligand), CdSe(CdCl
2
), CdSe(CdBr
2
), and CdSe(CdI
2
), respectively.
After annealing, particle sizes of 4.56 ± 0.46, 4.57 ± 0.33, 4.71 ± 0.38, and 4.52 ± 0.48
nm were measured for CdSe(native ligand), CdSe(CdCl
2
), CdSe(CdBr
2
), and CdSe(CdI
2
),
respectively (Figure 6.12). It should be noted that the average interparticle separation
observed by TEM analysis is significantly smaller (0.1-0.3 nm) for CdSe(CdX
2
) samples
than for CdSe(native ligand) or CdSe(butylamine) nanocrystals (2-3 nm). Such short
interparticle distances would naturally be expected for the short atomic ligand shells in
CdSe(CdX
2
) films, whereas CdSe(native ligand) and CdSe(butylamine) would be
expected to have interparticle separations on the order of several nm due to the sterically
bulky native ligands.
The large red shifts observed for annealed CdSe(CdX
2
) films may be due to enhanced
electronic coupling between nanocrystals, and while gross grain growth is not occurring
we do not rule out the possibility that the nanocrystals are fusing with one another.
24

However, the temperature profile employed for annealing nanocrystal films (i.e., 250 °C
for 10 min) is significantly lower than what is typically used to induce large grain growth
in sintered nanocrystal solids.
44,45
Considering the relatively weak binding affinity of L-

180
type ligands, the 250 °C annealing step should drive off most of the surface-bound
butylamine, further reducing the interparticle separation in CdSe(CdX
2
) samples. Indeed,
thermal loss of butylamine is observed by TGA as a 6-7% mass loss between 150-280 °C
for the CdSe(butylamine) nanocrystals (Figure 6.3). After annealing at 250 °C,
significant loss of butylamine is confirmed by XPS and FT-IR (see Figure 6.15 and 6.5).

Figure 6.15. XPS spectrum for annealed CdSe(butylamine) showing a reduction in the N 1s peak
near 399 eV after thermal annealing.

XPS analysis of annealed CdSe(butylamine) nanocrystal powders demonstrated a
reduction of the N 1s photoelectron line near 399 eV. Based on integration of the FT-IR
ν(C-H) stretching intensity between 2750-3050 cm
−1
for CdSe(butylamine) nanocrystals,
thermal annealing results in an ca. 25% reduction of the total integrated ν(C-H) stretching
intensity. Based on TGA results, we attribute this to a loss of surface butylamine ligands.
It should be noted that the ν(C-H) stretching signals from butylamine and native ligand
molecules overlap and are difficult to distinguish from one another; however, from the
TGA of CdSe(native ligand) nanocrystals, we are confident that native ligands are not

181
lost at 250 °C, and the reduction in ν(C-H) stretching intensity can therefore be primarily
attributed to the loss of butylamine ligands.

6.3.4. Photocurrent Measurements
To probe the influence of various CdX
2
and heat treatments on optoelectronic properties,
we performed PEC measurements on as-cast and 250 °C annealed nanocrystal films.
PEC measurements offer a convenient method for evaluating the effect of surface
treatments on the photoresponse of semiconductor nanocrystals without having to deal
with the multiple contacts and interfaces present in completed solid-state devices.
16,46-51

The nanocrystal films were prepared as photoelectrodes by spin-casting onto ITO
substrates under conditions that gave the same optical densities at 472 nm, and they were
annealed in an identical manner to the thin films used for optical characterizations (vide
infra). Based on ellipsometry measurements, these films were between 45-55 nm in
thickness. Photocurrent measurements were performed according to previously
published procedures for the photoelectrochemical characterization of CdSe
nanocrystals.
16,51
Photoelectrochemical measurements were carried out using an aqueous
50 mM sulfide/polysulfide redox couple using a three-electrode configuration, with Pt-
wire counter and Ag-wire pseudo-reference electrodes and two 472 nm LEDs for
illumination. While it is possible that the sulfide electrolyte may influence the surface
chemistry of the nanocrystals, we used identical experimental conditions for all of the
samples under investigation, and therefore any interaction between the aqueous species
and nanocrystal electrodes should be systematic throughout our comparison. A typical

182
experiment involved measuring the photocurrent response of the nanocrystal
photoelectrodes across a 1 V window while sweeping at a potential rate of 3 mV s
–1
with
5 s dark and illuminated intervals.
The photocurrent responses for the as-cast samples exhibit cathodic behavior, as
evidenced by an enhanced photocurrent magnitude with increasing negative applied
biases (Figure 6.16a).

Figure 6.16. Photoelectrochemical responses of CdSe(native ligands) and CdSe(CdX
2
)
photoelectrodes under chopped 472 nm illumination. (a) As-cast nanocrystal films exhibiting
cathodic photoresponse. (b) Annealed nanocrystal films exhibiting anodic response with
enhanced current densities relative to unannealed nanocrystal films. PEC responses for
CdSe(native ligands) are given in green, CdSe(CdCl
2
) in red, CdSe(CdBr
2
) in orange, and
CdSe(CdI
2
) in blue.

Nanocrystal thin films of CdSe(CdCl
2
) exhibit the largest photocurrents (141 µA cm
−2

from -0.05 to -0.4 V vs the Ag-wire pseudo-reference electrode) of all the samples,
whereas the CdSe(native ligand) films produce the lowest photocurrents (< 10 µA cm
−2
).
The butylamine surface ligands may possibly influence the nature of the photocurrent
response as we observed significant changes to the PEC data upon thermal removal of
butylamine ligands at 250 °C. More rationale for this will be provided below.

183
After annealing for 10 min at 250 °C, the nanocrystal thin films exhibit stark differences
in photoresponse when compared to the as-cast unannealed films (Figure 6.16b). As
observed before heat treatment, the specific CdX
2
surface treatment has a large effect on
the magnitude of the photoresponse, with CdSe(CdCl
2
) nanocrystal films again exhibiting
the largest photocurrents up to nearly 500 µA cm
–2
at +0.2 V vs the Ag-wire pseudo-
reference. However, the photoresponse changes from cathodic (as-cast) to anodic after
annealing, which may be related to changes in surface passivation and/or surface
reconstruction during heat treatment. As previously explained, a substantial portion of
surface-bound butylamine is removed after annealing (vide supra). Indeed, upon thermal
treatment, all of the CdSe(CdX
2
) samples show a substantial decrease in the total FT-IR
ν(C-H) stretching intensity as evidenced by a decrease in the integrated area from 2750-
3050 cm
−1
. Compared to unannealed CdSe(CdX
2
) samples, we calculate reductions in
the ν(C-H) stretching signal of approximately 45, 75, and 50% for CdSe(CdCl
2
),
CdSe(CdBr
2
), and CdSe(CdI
2
), respectively. We attribute the primary cause for this
reduction in FT-IR signal to removal of surface-bound butylamine. It is possible that the
changes from cathodic to anodic photoresponse that we observe in PEC measurements
after annealing are related to this.
The large electrochemical photocurrent measured for the annealed CdSe(CdCl
2
)
nanocrystal films may be the combined result of strong interparticle coupling from the
removal of native ligand content as well as beneficial alteration of the trap state landscape
from annealing induced butylamine removal and/or surface reconstruction. The former is
observed in TEM analysis, in which CdSe(native ligand) and CdSe(butylamine) samples

184
show interparticle separation greater than 1-2 nm (and thus smaller interparticle
coupling), whereas all CdX
2
treated samples exhibit extremely small particle-particle
distances (vide supra). Although each of the CdX
2
ligand exchanges results in close
interparticle contacts, the very different photocurrent responses among CdX
2
-treated
nanocrystals suggest varying trap state landscapes. We sought to understand the
differences in trap states resulting from different CdX
2
treatments by using a combination
of steady-state and time-resolved spectroscopic techniques that are described in the
following sections.

6.3.5. Surface Photovoltage Spectroscopy: Probing Sub-gap Energy States
Surface photovoltage (SPV) spectroscopy is a powerful contactless technique that can
provide information on the band gap, sub-band gap states, and majority carriers in
semiconductor nanocrystal powders and films.
52-54
In SPV, the surface potential of a
nanocrystal film on a conductive substrate is measured in a contact-free configuration
with a Kelvin probe (see Figure 6.17) as a function of photon energy.

Figure 6.17. Measurement configuration for SPV and sample spectra.

185
A positive contact potential difference (ΔCPD) signal indicates hole transport towards the
substrate, whereas a negative signal implies electron transport. Signals below the band
gap energy are caused by excitation of defect states. If the charge carriers in these defect
states are non-mobile, they produce ΔCPD signals through polarization by the built-in-
field at the sample-substrate or sample-ambient interface. SPV spectroscopy relies upon
the measurement of voltage rather than current, which enables the detection of very low
concentrations of states (10
–8
cm
–3
).
55-57
If the charge carriers in these defect states are
nonmobile, they produce ΔCPD signals through polarization by the built-in field at the
sample− substrate or sample−ambient interface.
58,59
Here, we employ SPV to gain a
better understanding of the influence that CdX
2
ligand exchanges have on the energies of
sub-band gap states, as well as whether the resulting traps are electron or hole specific.
The onset of the photovoltage signal for the as-cast CdSe(native ligand) nanocrystal films
is observed at 1.95 eV, near the band gap of the material. The negative sign of the
voltage suggests that the signal is due to injection of electrons into the ITO substrate
(Figure 6.18). No significant sub-gap features are present, which implies an
electronically well-passivated surface. After CdX
2
treatments with butylamine, the SPV
spectra of the CdSe(CdX
2
) nanocrystal films are drastically altered, exhibiting clear
differences in sign, magnitude, and onset as presented in Figure 6.19. A large positive
photovoltage signal (+350 mV at 2.5 eV) is observed for CdSe(CdCl
2
) nanocrystal films,
whereas CdSe(CdBr
2
) exhibits a much weaker positive signal (+18 mV at 2.5 eV).
Compared to CdSe(native ligand), both samples show lower energy onsets near 1.5 eV
for CdSe(CdCl
2
) and 1.7 eV for CdSe(CdBr
2
) nanocrystal films, suggesting the presence
of surface traps. Furthermore, the reversal of sign from negative to positive indicates a

186

Figure 6.18. SPV (solid green) and diffuse reflectance absorption (dashed) spectra of a
CdSe(native ligands) nanocrystal film with the arrow indicating the photovoltage signal onset.

Figure 6.19. SPV (solid colored lines) and diffuse reflectance absorption (dashed) spectra of
unannealed (a) CdSe(CdCl
2
), (b) CdSe(CdBr
2
), (c) CdSe(CdI
2
), and (d) CdSe(butylamine)
nanocrystal films.

187
switch in carrier injection type from electrons in CdSe(native ligand) to holes in the
CdSe(CdCl
2
) and CdSe(CdBr
2
) nanocrystal films. In the case of CdSe(CdI
2
), a negligible
SPV signal is observed Considering that the CdSe(CdI
2
) nanocrystal films exhibit the
smallest photocurrents of the three CdSe(CdX
2
) samples, this SPV observation may be a
result of enhanced recombination between deeply trapped holes and mobile electrons, or
it is also probable that it is a result of the superposition of positive and negative SPV
signals resulting from the presence of strong trap sites for both electrons and holes,
respectively. The exact origin of the change in the dominant type of trap upon going
from CdSe(native ligand) to CdSe(CdCl
2
) or CdSe(CdBr
2
) may be rooted in several
factors. One possibility is the introduction of additional Cd
2+
surface ions with empty
states just below the CdSe conduction band. Upon trapping of photoexcited electrons,
these ions can undergo reduction to Cd(0). Similar trap states in CdSe nanocrystal films
introduced by quantum confinement-related shifts in the Cd
2+
surface states were
previously shown to cause a transition from cathodic to anodic photoelectrochemical
response when going from large to small CdSe quantum dots.
60
Another possible
explanation may be the introduction of new surface trap states associated with surface
reconstruction during the butylamine-assisted CdX
2
exchange (vide supra). As a control,
SPV spectra of CdSe(butylamine) nanocrystal films were also collected (Figure 6.19).
The sub-band gap SPV signal onset and the photovoltage sign in the CdSe(butylamine)
nanocrystal films are similar to those observed in CdSe(CdCl
2
) and CdSe(CdBr
2
) films.
This suggests that butylamine may be somehow involved in creating surface traps,
possibly through assisting in reconstruction of surface Cd
2+
sites or by facilitating the
creation of new surface states.

188
After annealing at 250 °C, we find the SPV spectra to change considerably (Figure 6.20),
which can be reconciled with the photocurrent changes observed after annealing at the
same temperature.

Figure 6.20. SPV (solid colored lines) and absorption (dashed) spectra of (a) CdSe(CdCl
2
) (b)
CdSe(CdBr
2
), and (c) CdSe(butylamine) nanocrystal films after annealing at 250 °C. An SPV
spectrum for a CdSe(CdI
2
) nanocrystal film is not shown because no significant photovoltage
signal was observed.

The most significant difference in the SPV spectra between the as-cast and annealed
samples is the disappearance of the sub-band gap signals near 1.5 eV for CdSe(CdCl
2
)

189
and CdSe(butylamine) and 1.7 eV for CdSe(CdBr
2
), suggesting a thermally induced
reduction of surface traps. One plausible explanation is surface reconstruction that is
facilitated by removal of butylamine during annealing. In addition to the alignment of
SPV signal with the absorption onset, the SPV signals for the CdSe(CdCl
2
) and
CdSe(CdBr
2
) nanocrystal films decrease significantly. Considering the change from
cathodic to anodic photoresponse in annealed CdSe(CdCl
2
) and CdSe(CdBr
2
) nanocrystal
films, the photovoltage signal reduction suggests a decrease in the density of electron
surface traps upon heating and an enhanced role of surface hole traps. A similar
reduction in SPV signal for CdSe(butylamine) nanocrystal films after annealing was also
observed (Figure 6.20), which supports the idea of thermally induced electron trap state
removal or surface reconstruction.

6.3.6. Time-Resolved Photoluminescence and Transient Absorption: Excited State
Dynamics
We used ultrafast, time-resolved optical spectroscopy as a probe of the electronic
structure of the CdSe nanocrystals after ligand exchange with CdX
2
. By systematically
altering the halide in the CdX
2
series, and by using low temperature measurements, we
were able to quantify hole trapping rates and determine the relative depths of the mid-gap
states introduced via ligand exchange.
The steady-state near band edge photoluminescence is clearly modified by the CdX
2

ligand exchange. Both in solution and in thin films, the near band edge
photoluminescence is quenched relative to the CdSe(native ligand) nanocrystals for all of

190
the CdX
2
treatments (Figure 6.21). The solution quenching can be attributed to the
trapping of photogenerated holes onto mid-gap states introduced by the CdX
2
ligands,
61

whereby the electron and hole wavefunction overlap is reduced and the charge transfer
state is not emissive.

Figure 6.21. Steady-state and time-resolved photoluminescence spectra for CdSe nanocrystal
samples. (a) Steady-state photoluminescence of CdSe nanocrystal suspensions; (b) steady-state
photoluminescence of CdSe nanocrystal films; and (c) TCSPC plots for (1) CdSe(native ligands),
(2) CdSe(CdCl
2
), (3) CdSe(CdBr
2
), and (4) CdSe(CdI
2
) nanocrystal films at 77 K. The
instrument response function is shown as trace (5). The excitation wavelength was 400 nm.

191
The photoluminescence quenching is stronger in the case of the CdI
2
and CdBr
2
ligands
than it is for the CdCl
2
ligands. While it is difficult to rule out effects of varying ligand
coverage, this trend matches the trend of the redox potential of the halides in aqueous
solution found in the literature.
62
With a lower redox potential (-0.535 V vs NHE), CdI
2

offers a larger driving force for transfer of photogenerated holes to the surface halide
level than CdCl
2
(-1.358 V vs NHE) or CdBr
2
(-1.087 V vs NHE). This trend also
matches the first-principles study of Cl
–
and I
–
binding onto Cd
33
Se
33
clusters, which
showed that the halide ions introduce new, halide-participating states close to the valence
band. These halide-related states are located within the band gap except when the Cl
–
ion
is coordinated to 3 Cd atoms at the same time (on the (000 ) facet), in which case the
halide-based level is pushed within the valence band, losing its hole-trapping
characteristics.
61
The halide-exchanged nanocrystal film photoluminescence is further
quenched compared to solutions. This observation suggests that an additional quenching
mechanism may be at play, consistent with the work of Johnson et al., where reduced
interparticle distances induce exciton dissociation and local charge transport to
nonradiative, non-geminate recombination sites.
63
Here, the reduction in interparticle
distance is expected to be similar across the halide series (vide supra), so that the
observed variation in photoluminescence quenching for the different nanocrystal films
(Figure 6.21b) is likely a results of the energy and/or density of dark states. Finally,
while the observed photoluminescence trend within the halide series was robust and
reproducible, the effect of butylamine on nanocrystal photoluminescence was strongly
dependent on the sample history and nanocrystal concentrations.
64

€
1

192
TCSPC studies were used to further quantify the hole transfer dynamics of the CdX
2

ligand exchanged CdSe nanocrystals. At low temperature, phonon-assisted relaxation
pathways are reduced so that the photoluminescence is stronger and ligand effects are
clearer. Figure 6.21c shows that at 77 K the near band-edge photoluminescence of CdX
2
-
treated nanocrystals is rapidly quenched relative to the CdSe(native ligand) nanocrystals.
The characteristic times for hole transfer to the ligand states are evaluated by fitting each
decay with 3 components; all samples exhibit similar slow decay times around 200 ps and
1.8 ns, but the fastest component is ligand dependent. This fast decay can be fitted with a
characteristic time that decreases for heavier halides: 68 ± 4 ps for CdCl
2
, 44 ± 1 ps for
CdBr
2
, and < 30 ps (instrument-limited) for CdI
2
. This trend is consistent with the idea
that the driving force for hole transfer is larger for heavier halide atoms.
Temperature-dependent ultrafast TA measurements were used as a complement to SPV
spectroscopy and TCSPC photoluminescence spectroscopy to probe the electronic
structure of the CdSe nanocrystals after CdX
2
ligand exchange. The TA measurements
allow us (i) to probe the populations of a number of different electronic levels and (ii) to
access sub-ps timescales beyond the resolution of the time-resolved photoluminescence
experiments. Monitoring the growth and recovery of ground state bleach and
photoinduced absorption features in the TA difference spectra let us develop a description
of the trend in trap state energies that results from the different surface treatments.
The TA experiments revealed several spectral features that are common to all of the CdSe
nanocrystal samples. First, negative signals near 580, 540, 490, and 475 nm indicate the
bleaching of 1S
3/2
-1S
e
, 2S
3/2
-1S
e
, 1P
3/2
-1P
e
, and 2S
1/2
-1S
e
excitonic transitions,

193
corresponding to population of the initial and/or the final energy level(s) of the transition
(see Figure 6.22). Second, the positive features indicate photo-induced absorption (PIA)
from a populated excited state to higher energy state. In particular, the positive 600 nm
feature is associated with the formation of a second exciton when a first excitation
(exciton or charge) is already present on the nanocrystal.
65

Figure 6.22. TA difference spectra for nanocrystal films at 77 K for different probe times after
excitation at 400 nm. Spectra are color coded: CdSe(native ligand) (green), CdSe(butylamine)
(pink), CdSe(CdCl
2
) (red), CdSe(CdBr
2
) (orange), and CdSe(CdI
2
) (blue). Left side depicts raw
difference spectra, and the right side shows normalized spectra.

194
Careful analysis of the PIA and bleach recovery signals reveals ligand-dependent
dynamics. For example, Figure 6.23 shows the initial 1000 ps evolution of the 2S
3/2
-1S
e

transition at 77 and 298 K.

Figure 6.23. Dynamics of the bleach of the 2S
3/2
-1S
e
transition (at 540 nm) for the nanocrystal
samples at 77 K (black traces) and 298 K (colored traces) for (a) CdSe(native ligands), (b)
CdSe(butylamine), (c) CdSe(CdCl
2
), (d) CdSe(CdBr
2
), and (e) CdSe(CdI
2
) nanocrystals. A pump
wavelength λ
exc
= 400 nm and a pump fluence of 9 µJ cm
–2
were used for all measurements.

195
As a result of the valence band degeneracy, this negative signal indicates that the 1S
e

level in the conduction band is populated.
66
Interestingly, the bleach rises over very
different time scales depending on the ligands; the rise is finished after 1 ps for
CdSe(native ligands) but takes longer for all CdSe(CdX
2
) nanocrystal samples. Further,
the bleach rise time increases upon going from CdCl
2
to CdI
2
and is complete after 2 ps
for CdSe(CdCl
2
), 10 ps for CdSe(CdBr
2
), and 30 ps for CdSe(CdI
2
) nanocrystals. This
rise time is related to the intraband relaxation of electrons in the nanocrystal conduction
band; the 400 nm excitation produces high-energy electrons that then thermalize to the
band edge, populating the 1S
e
level and inducing the bleach of 1S
e
-based transitions. The
thermalization of electrons in CdSe(native ligands) occurs by Auger scattering (i.e., by
energy transfer from the hot electron to the hole) followed by rapid hole thermalization
through the dense valence band.
67,68
When photogenerated holes are trapped onto ligand-
induced trap states in the CdX
2
-treated samples, Auger scattering is quenched and the
thermalization of electrons is slower. The slower thermalization of hot electrons upon
CdI
2
treatment indicates that CdSe(CdI
2
) nanocrystals possess a larger density of and/or
deeper photogenerated hole traps than do CdSe(CdBr
2
) and CdSe(CdCl
2
) nanocrystals, in
agreement with steady-state and time-resolved photoluminescence measurements.
Control measurements on CdSe(butylamine) films (Figure 6.23b) show no thermalization
slowdown compared to native ligands, and thereby reinforce the idea of halide-specific
hole trapping rather than electron trapping. Thermalization of hot electrons is a current
area of interest because this process directly competes with hot carrier extraction and
carrier multiplication processes, which have the potential to increase the thermodynamic
limit in nanocrystal solar cells via hot carrier extraction. Thus, the ability to understand

196
and control the factors slowing down hot electron thermalization (e.g., through ligand
engineering of trap states) is key for optimizing hot carrier extraction and multiple
exciton generation.

The ligands also affect the long time dynamics of the 1S
e
population. In particular, both
CdSe(native ligands) and CdSe(butylamine) nanocrystals exhibit a slow decay of the 1S
e

population with a decay time constant of several ns independent of temperature between
77-298 K (Figure 6.23). In contrast, the 1S
e
population is temperature dependent for the
CdX
2
-treated samples, which exhibit similarly slow decay dynamics at 77 K, but at
higher temperature their decay accelerates. This faster decay component suggests that
the 1S
e
level can be depopulated by a thermally activated mechanism, with the fastest rate
observed for CdSe(CdCl
2
) and the slowest rate observed for CdSe(CdI
2
) nanocrystals.
While our other steady-state and time-dependent spectroscopic observations of ligand
effects (i.e., SPV, photoluminescnece quenching, TCSPC and TA rise times) could not
distinguish between the effects of hole trap densities and hole trap energies, the thermally
activated decays specifically point to differences in trap energies (not ruling out
differences in trap densities). Furthermore, the control experiment on CdSe(butylamine)
nanocrystals shows temperature-independent bleach recovery behavior similar to
CdSe(native ligand) nanocrystals, which further suggests that the thermally activated
ground state bleach recovery is related to halide-induced hole trapping. We therefore
hypothesize that the relaxation mechanism of the 1S
e
electron population can occur via
recombination with thermally detrapped holes; detrapping from the CdSe(CdCl
2
) levels
that are close to the valence edge is more facile as compared to detrapping from the
CdSe(CdBr
2
) and CdSe(CdI
2
) levels that are located deeper into the band gap. To

197
estimate the depth of the trap levels relative to the valence band edge, we fitted the long
time bleach recoveries measured at 77 K and at room temperature for both CdSe(native
ligands) and CdSe(CdX
2
) to a very simple model. We assumed that trapping is complete
at 100 ps and considered two pathways for bleach recovery: (i) relaxation from the
valence band (ligand independent) and (ii) relaxation from CdX
2
-induced trap states. The
extracted parameters for the CdSe(CdX
2
) samples are the fraction of trapped excitations
at 100 ps and the depth of the trap relative to the valence band. Details of the model and
of the fit can be found in the Supporting Information (see Figure 6.24).

198

Figure 6.24. Experimental 2S
3/2
-1S
e
bleach decay d
obs
for the CdSe(native ligand) and
CdSe(CdX
2
) nanocrystal films (circles) and a number of predictions d
pred
given our simplified
model (black lines).
As suggested by the time-resolved photoluminescence results and by the bleach rise
observed by TA, the traps are deeper for the CdI
2
ligands (ca. 210 meV) than for CdBr
2

(ca. 200 meV) than for CdCl
2
ligands (ca. 140 meV). Correspondingly, the fraction of
trapped charges at 100 ps is larger for the heavier halides; that is, around 68% for
−1
0
−1
0
−1
0
600 1000
−1
0
200
CdSe(NL)
CdSe(CdCl
2
)
CdSe(CdBr
2
)
CdSe(CdI
2
)
77 K 298 K
600 1000 200
Time (ps)
ΔA (norm.) ΔA (norm.) ΔA (norm.) ΔA (norm.)

199
CdSe(CdI
2
), 59% for CdSe(CdBr
2
), and 37% for CdSe(CdCl
2
) nanocrystal films. The
model points to a combined effect of larger trap densities and larger trap energies for
heavier halides. Of course, the exact trap energy and density values extracted from this
simplistic model should be taken with caution, particularly since an average, well-defined
trap level was implemented, rather than a more realistic trap energy distribution.
The transient absorption measurement of annealed CdSe(CdCl
2
) films confirms the red
shifts and broadening of electronic transitions observed in the steady-state measurements
(see Figure 6.25). The bleach recovery dynamics after annealing is not as thermally
activated as it is before annealing. This could be due to a broadening of the trap level
energy distribution or to a shift of the average trap energy.

Figure 6.25. TA dynamics of the bleach of the 2S
3/2
-1S
e
transition (at 540 nm) for the annealed
CdSe(CdCl
2
) nanocrystal film at 77 K (black traces) and 298 K (colored traces).

In summary, based on the results from TA, PEC, and TCSPC photoluminescence
experiments, we propose a simple picture that illustrates the differences in trap state
energies in CdSe(CdX
2
) nanocrystals (Figure 6.26). In this depiction, CdX
2
ligands
induce surface hole traps that vary in energy based on the halide. The shallowest trap
levels are thought to exist in CdSe(CdCl
2
) whereas the deepest hole traps are found in

200

Figure 6.26. Proposed energy landscape in CdX
2
induced hole trap states for CdSe(CdX
2
)
nanocrystals and electron trap landscape dependence on thermal treatment. From temperature-
dependent ultrafast TA traces, the thermal activation energies for hole detrapping are estimated to
be ca. 140 meV for CdSe(CdCl
2
), 200 meV for CdSe(CdBr
2
), and 210 meV for CdSe(CdI
2
). SPV
analysis suggests thermally induced rearrangement of electron traps concurrent with removal of
the majority of butylamine ligands, as depicted in the diagram after 250 °C treatment. An
estimation of the electron trap state energies relative to the CdSe LUMO is taken from the
energies of the pre-band gap features observed in SPV measurements. For visual clarity, the
energy axis on the figure is not drawn to scale.

CdSe(CdI
2
) nanocrystals. The butylamine ligands cooperatively utilized with CdX
2
for
ligand exchange induce a surface state that results in the creation of electron traps
observed in SPV measurements and in the cathodic behavior in the as-cast CdSe(CdX
2
)
nanocrystal films. Thermal treatment reduces the effect of electron trap states on the
charge transport properties, likely via thermal removal of butylamine and/or associated
surface reconstructions, and results in strong anodic photoconductivity in CdSe(CdCl
2
)
based photoelectrochemical devices.

201
6.4. Experimental
6.4.1. General Considerations
The following chemicals were used as received without further purification.
Tetradecanoic acid (>99.0%, Sigma-Aldrich), sodium hydroxide (pellets, 97%, Sigma-
Aldrich), Cd(NO
3
)
2
•4 H
2
O (98.0 %, Sigma-Aldrich), Cd(CH
3
CO
2
)
2
hydrate (>99.99%,
Sigma-Aldrich), selenium dioxide (≥99.9%, Sigma-Aldrich), oleic acid (technical grade,
90%, Sigma-Aldrich), 1-octadecene (technical grade, 90%, Sigma-Aldrich), methanol
(production grade, VWR), CdCl
2
•H
2
O (99.99%, Alfa Aesar), CdBr
2
(98%, Alfa Aesar),
CdI
2
(99.5 %, Alfa Aesar), butylamine (Acros), Na
2
S hydrate (>60%, Lancaster), sulfur
(precipitated, 99.5%, Alfa Aesar).

6.4.2. Synthesis and Ligand Exchange of CdSe(native ligand) Nanocrystals
The cadmium myristate precursor was prepared according to a previously published
procedure.
36
In a 1-L flask equipped with a 2" magnetic stir bar and fitted with two
rubber septa and a 14" condenser, Cd(C
14
H
27
O
2
)
2
(10.0 g, 17.6 mmol) was added
followed by SeO
2
(2.67 g, 22.0 mmol) and Cd(C
2
H
3
O
2
)
2
(0.45 g, 1.8 mmol). The mixture
was suspended in 1-octadecene (ODE, 250 mL), and the flask was fitted a nitrogen-
connected condenser and two rubber septa, one of which was used to fix a stainless steel
thermocouple in the reaction suspension in order to monitor the internal temperature over
the entire reaction course. An initial period of 1-2 h under vacuum at 130 °C was found
to be necessary to obtain monodisperse nanocrystal size distributions. After the initial

202
heating step, the reaction was vented to nitrogen and ramped to 240 °C (6-7 °C min
–1
).
Aliquots were taken every 30 s, and nanocrystal nucleation was first observed around 200
°C, as evidenced by the appearance of excitonic features in the absorption spectra. The
reaction was quenched by removal from heat, cooling in a room temperature oil bath, and
injection of a 1:1 vol/vol oleic acid:ODE (60 mL) mixture. After cooling to room
temperature, the crude reaction mixture was purified by flocculation with acetone and
redispersion in toluene. This flocculation/redispersion cycle was repeated 4 additional
times before filtering the final toluene dispersion of CdSe(NL) nanocrystals through a
0.45 µm syringe filter. The final product was stable in the dark for several months at
concentrations up to 150 mg mL
–1
CdSe(NL).
All CdX
2
ligand exchanges were done using 50 mM CdX
2
methanolic solutions. This
was found to be the solubility limit for CdCl
2
in methanol. It should be noted that the
number of CdX
2
treatments can be significantly reduced (i.e., 2-5 total treatments for
quantitative native ligand exchange) in the case of CdI
2
and CdBr
2
if more concentrated
solutions are employed.
In general, a 50 mM solution of CdX
2
(X = Cl, Br, I) in methanol (1 mL) was added to a
toluene suspension of CdSe(native ligands) nanocrystals (3 mL, 100 mg mL
–1
) and
shaken for 1-2 min. Methanol (5 mL) was added to the nanocrystal suspension, and the
nanocrystals were isolated by centrifugation (6000 rpm, 1 min) and the clear supernatant
was discarded. Chlorobenzene (3 mL) was used to redisperse the nanocrystals. A
significantly enhanced dispersability in chlorobenzene (along with CHCl
3
and 1,2-
dichlorobenzene) after the initial CdX
2
treatment suggested an immediate change in

203
nanocrystal surface chemistry. Additional CdX
2
treatments on the chlorobenzene-
dispersed nanocrystals were performed in the same manner as the first.
After 4-5 treatments the 50 mM CdI
2
treated nanocrystals, henceforth referred to as
CdSe(CdI
2
), did not easily disperse in chlorobenzene. Addition of 25 µL of butylamine
to the mixture of partially dispersed CdSe(CdI
2
) in chlorobenzene afforded complete
dispersability. The resulting CdSe(CdI
2
) nanocrystals were then flocculated with pentane
(10 mL) and isolated via centrifugation (6000 rpm, 1 min). An additional chlorobenzene
dispersion/pentane flocculation was performed to remove residual CdI
2
and butylamine.
The final chlorobenzene dispersion of CdSe(CdI
2
) nanocrystals was passed through a 0.2
µm syringe filter and stored in the dark at ambient temperature.
Similar to the CdI
2
treatment, the CdBr
2
-treated nanocrystals, henceforth referred to as
CdSe(CdBr
2
), became difficult to disperse after 7-8 CdBr
2
treatments. At this point,
butylamine (25 µL) was added to the CdSe(CdBr
2
)/chlorobenzene mixture, and the
nanocrystals quickly dispersed forming a completely homogeneous colloidal dispersion.
Pentane (10 mL) was used to flocculate the nanocrystals, which were isolated by
centrifugation (6000 rpm, 1 min), and redispersed in chlorobenzene (3 mL). An
additional pentane/chlorobenzene purification step was performed, and the final
dispersion of CdSe(CdBr
2
) nanocrystals in chlorobenzene was filtered through a 0.2 µm
syringe filter and stored in the dark.
CdCl
2
-treated nanocrystals (also referred to from here on as CdSe(CdCl
2
)) were prepared
in a similar way to CdSe(CdBr
2
), except 7-8 additional CdCl
2
treatments were performed
after the first addition of butylamine. After a maximum of 14-16 CdCl
2
treatments,

204
butylamine (25 µL) was again added to the mixture of CdSe(CdCl
2
) and chlorobenzene,
and the nanocrystals became completely dispersable in chlorobenzene. Addition of
pentane (15 mL) resulted in nearly complete flocculation of CdSe(CdCl
2
) with only a
slight brown color remaining in the supernatant. The nanocrystals were redispersed in
chlorobenzene (3 mL), and two additional flocculation/redispersion cycles were
performed to remove residual butylamine and CdCl
2
. The final chlorobenzene dispersion
of CdSe(CdCl
2
) nanocrystals was passed through a 0.2 µm syringe filter and stored in the
dark.

6.4.3. Surface Photovoltage Spectroscopy
Contact potential difference (CPD) measurements were recorded on nanocrystal films on
ITO substrates (Thin Film Devices, Inc., ρ = 20-30 Ω sq
−1
) in vacuum (10
−7
bar) with a
vibrating gold mesh reference electrode (Kelvin Probe, 3 mm in diameter, Delta PHI
Besocke) and a Kelvin control 07 (Delta PHI Besocke) with a sensitivity of 1 mV. The
sample-probe distance was kept consistent (ca. 1 mm). A 175 W Xe arc lamp (PE175-
BF) was used as the light source, in conjunction with a monochromator (Cornerstone
130) to provide monochromatic illumination with an average bandwidth of 30 nm. The
light power at the sample was < 1.0 mW cm
–2
. Photovoltage spectra were corrected for
drift effects due to pressure variations by subtracting a linear background. Nanocrystal
thin films were deposited by spin-casting (1000 rpm, 60 s) onto pre-cleaned ITO
substrates under a nitrogen atmosphere. After spin-casting, films were placed into a
nitrogen filled cabinet for 20-30 min to complete the drying process before being used for

205
experiments. CdSe(native ligands) nanocrystals were cast from toluene suspensions,
whereas all other nanocrystal films were cast from chlorobenzene suspensions. All
nanocrystal films were between 45-55 nm in thickness, as determined by ellipsometry.

6.4.4. Femtosecond Transient Absorption Spectroscopy
Nanocrystal films that were prepared identically to samples used for SPV measurements
were placed in the sample chamber of a ST-100 Janis cryostat. TA measurements were
carried out using pulses derived from a Coherent Legend Ti:sapphire amplifier (1 kHz,
3.5 mJ, 35 fs), as described in detail elsewhere.
69,70
The amplifier output was frequency-
doubled to serve as the 400 nm pump pulse. White light supercontinuum probe pulses
were polarized perpendicular to the pump pulse and generated using a moving CaF
2

window. The probe pulse was dispersed by an Oriel MS1271 spectrograph onto a 256
pixel Si photodiode array. To obtain satisfactory signal-to-noise on the optically thin
samples, transient spectra were measured for typical pump fluences of 9 µJ cm
–2
.
Samples were translated perpendicular to the path of the pump and probe to prevent
photodamage.

206
6.5. Conclusions
In summary, we have developed a simple and air-tolerant colloidal procedure for the
exchange of native ligands on CdSe nanocrystals with CdX
2
. Butylamine is employed as
a thermally labile L-type coligand to impart additional colloidal stability to the
CdSe(CdX
2
) nanocrystals. In PEC experiments, we observe a photoresponse change in
which the CdSe(CdX
2
) nanocrystal films initially exhibit a cathodic photoresponse but
switch to an anodic photoresponse upon heating. We propose that this change may be
related to the removal of surface electron traps, possibly facilitated by surface
reconstruction during heating and the thermally induced expulsion of butylamine. In
support of this interpretation, SPV measurements suggest the removal of sub-band gap
trap states after annealing the CdSe(CdX
2
) nanocrystal films. This implies that thermally
induced surface changes, such as removal of butylamine and/or reconstruction, contribute
to better overall surface trap passivation. We further demonstrate that the excited state
dynamics are strongly influenced by the nanocrystal ligand treatment, with clear
differences observed for different CdX
2
ligand exchanges. Ultrafast TA spectroscopy
reveals the fastest bleach growth and recovery for CdSe(CdCl
2
) nanocrystals, suggesting
the lowest degree of hole trapping among the CdX
2
-treated nanocrystals. These results
suggest that periodic trends may exist within chemically similar classes of surface ligands
that may be useful for controlling the depth, density, and specificity of nanocrystal
surface traps. We believe these findings may be useful for the development of all-
inorganic nanocrystal-based optoelectronics, in particular for solar energy conversion and
devices designed to harvest hot carriers and/or carrier multiplication.

207
6.6. References
(1) Kim, J. Y.; Voznyy, O.; Zhitomirsky, D.; Sargent, E. H. 25th Anniversary Article:
Colloidal Quantum Dot Materials and Devices: a Quarter-Century of Advances.
Adv. Mater. 2013, 25, 4986–5010.
(2) Talapin, D. V.; Lee, J. S.; Kovalenko, M. V.; Shevchenko, E. V. Prospects of
Colloidal Nanocrystals for Electronic and Optoelectronic Applications. Chem.
Rev. 2010, 110, 389–458.
(3) Kim, J. Y.; Kotov, N. A. Charge Transport Dilemma of Solution-Processed
Nanomaterials. Chem. Mater. 2013, 26, 134–152.
(4) Zabet-Khosousi, A.; Dhirani, A. A. Charge Transport in Nanoparticle Assemblies.
Chem. Rev. 2008, 108, 4072–4124.
(5) Fritzinger, B.; Capek, R. K.; Lambert, K.; Martins, J. C.; Hens, Z. Utilizing Self-
Exchange to Address the Binding of Carboxylic Acid Ligands to CdSe Quantum
Dots. J. Am. Chem. Soc. 2010, 132, 10195–10201.
(6) Gomes, R.; Hassinen, A.; Szczygiel, A.; Zhao, Q.; Vantomme, A.; Martins, J. C.;
Hens, Z. Binding of Phosphonic Acids to CdSe Quantum Dots: a Solution NMR
Study. J. Phys. Chem. Lett. 2011, 2, 145–152.
(7) Luther, J. M.; Pietryga, J. M. Stoichiometry Control in Quantum Dots: a Viable
Analog to Impurity Doping of Bulk Materials. ACS Nano 2013, 7, 1845–1849.
(8) Oh, S. J.; Berry, N. E.; Choi, J. H.; Gaulding, E. A.; Paik, T.; Hong, S. H.;
Murray, C. B.; Kagan, C. R. Stoichiometric Control of Lead Chalcogenide
Nanocrystal Solids to Enhance Their Electronic and Optoelectronic Device
Performance. ACS Nano 2013, 7, 2413–2421.
(9) Yuan, M.; Zhitomirsky, D.; Adinolfi, V.; Voznyy, O.; Kemp, K. W.; Ning, Z.;
Lan, X.; Xu, J.; Kim, J. Y.; Dong, H.; Sargent, E. H. Doping Control via
Molecularly Engineered Surface Ligand Coordination. Doping Control via
Molecularly Engineered Surface Ligand Coordination. Adv. Mater. 2013, 25,
5586-5592.
(10) Green, M. L. H. A New Approach to the Formal Classification of Covalent
Compounds of the Elements. J. Organomet. Chem. 1995, 500, 127–148.
(11) Anderson, N. C.; Hendricks, M. P.; Choi, J. J.; Owen, J. Ligand Exchange and the
Stoichiometry of Metal Chalcogenide Nanocrystals: Spectroscopic Observation of
Facile Metal-Carboxylate Displacement and Binding. J. Am. Chem. Soc. 2013,
135, 18536–18548.

208
(12) Kovalenko, M. V.; Scheele, M.; Talapin, D. V. Colloidal Nanocrystals with
Molecular Metal Chalcogenide Surface Ligands. Science 2009, 324, 1417–1420.
(13) Kovalenko, M. V.; Bodnarchuk, M. I.; Zaumseil, J.; Lee, J. S.; Talapin, D. V.
Expanding the Chemical Versatility of Colloidal Nanocrystals Capped with
Molecular Metal Chalcogenide Ligands. J. Am. Chem. Soc. 2010, 132, 10085–
10092.
(14) Liu, W.; Lee, J. S.; Talapin, D. V. III–V Nanocrystals Capped with Molecular
Metal Chalcogenide Ligands: High Electron Mobility and Ambipolar
Photoresponse. J. Am. Chem. Soc. 2013, 135, 1349–1357.
(15) Buckley, J. J.; Greaney, M. J.; Brutchey, R. L. Ligand Exchange of Colloidal
CdSe Nanocrystals with Stibnates Derived from Sb
2
S
3
Dissolved in a Thiol-
Amine Mixture. Chem. Mater. 2014, 26, 6311-6317.
(16) Webber, D. H.; Brutchey, R. L. Ligand Exchange on Colloidal CdSe Nanocrystals
Using Thermally Labile Tert-Butylthiol for Improved Photocurrent in Nanocrystal
Films. J. Am. Chem. Soc. 2012, 134, 1085–1092.
(17) Fafarman, A. T.; Koh, W. K.; Diroll, B. T.; Kim, D. K.; Ko, D. K.; Oh, S. J.; Ye,
X.; Doan-Nguyen, V.; Crump, M. R.; Reifsnyder, D. C.; Murray, C. B.; Kagan, C.
R. 3Thiocyanate-Capped Nanocrystal Colloids: Vibrational Reporter of Surface
Chemistry and Solution-Based Route to Enhanced Coupling in Nanocrystal
Solids. J. Am. Chem. Soc. 2011, 133, 15753–15761.
(18) Dong, A.; Ye, X.; Chen, J.; Kang, Y.; Gordon, T.; Kikkawa, J. M.; Murray, C. B.
A Generalized Ligand-Exchange Strategy Enabling Sequential Surface
Functionalization of Colloidal Nanocrystals. J. Am. Chem. Soc. 2011, 133, 998–
1006.
(19) Rosen, E. L.; Buonsanti, R.; Llordes, A.; Sawvel, A. M.; Milliron, D. J.; Helms,
B. A. Exceptionally Mild Reactive Stripping of Native Ligands From Nanocrystal
Surfaces by Using Meerwein’s Salt. Angew. Chem. Int. Ed. 2011, 51, 684–689.
(20) Nag, A.; Kovalenko, M. V.; Lee, J. S.; Liu, W.; Spokoyny, B.; Talapin, D. V.
Metal-Free Inorganic Ligands for Colloidal Nanocrystals: S
2–
, HS
–
, Se
2–
, HSe
–
,
Te
2–
, HTe
–
, TeS3
2–
, OH
–
, and NH
2–
As Surface Ligands. J. Am. Chem. Soc. 2011,
133, 10612–10620.
(21) Liu, Y.; Tolentino, J.; Gibbs, M.; Ihly, R.; Perkins, C. L.; Liu, Y.; Crawford, N.;
Hemminger, J. C.; Law, M. PbSe Quantum Dot Field-Effect Transistors with Air-
Stable Electron Mobilities Above 7 cm
2
V
–1
s
–1
. Nano Lett. 2013, 13, 1578–1587.
(22) Anderson, N. C.; Owen, J. S. Soluble, Chloride-Terminated CdSe Nanocrystals:
Ligand Exchange Monitored by
1
H and
31
P NMR Spectroscopy. Chem. Mater.
2013, 25, 69–76.

209
(23) Zanella, M.; Maserati, L.; Pernia Leal, M.; Prato, M.; Lavieville, R.; Povia, M.;
Krahne, R.; Manna, L. Atomic Ligand Passivation of Colloidal Nanocrystal Films
via Their Reaction with Propyltrichlorosilane. Chem. Mater. 2013, 25, 1423–
1429.
(24) Norman, Z. M.; Anderson, N. C.; Owen, J. S. Electrical Transport and Grain
Growth in Solution-Cast, Chloride-Terminated Cadmium Selenide Nanocrystal
Thin Films. ACS Nano 2014, 8, 7513–7521.
(25) Niu, G.; Wang, L.; Gao, R.; Li, W.; Guo, X.; Dong, H.; Qiu, Y. Inorganic
Halogen Ligands in Quantum Dots: I
-
, Br
-
, Cl
-
and Film Fabrication Through
Electrophoretic Deposition. Phys. Chem. Chem. Phys. 2013, 15, 19595-19600.
(26) Lee, J. S.; Kovalenko, M. V.; Huang, J.; Chung, D. S.; Talapin, D. V. Band-Like
Transport, High Electron Mobility and High Photoconductivity in All-Inorganic
Nanocrystal Arrays. Nat. Nanotech. 2011, 6, 348–352.
(27) Chung, D. S.; Lee, J. S.; Huang, J.; Nag, A.; Ithurria, S.; Talapin, D. V. Low
Voltage, Hysteresis Free, and High Mobility Transistors From All-Inorganic
Colloidal Nanocrystals. Nano Lett. 2012, 12, 1813–1820.
(28) Tsokkou, D.; Papagiorgis, P.; Protesescu, L.; Kovalenko, M. V.; Choulis, S. A.;
Christofides, C.; Itskos, G.; Othonos, A. Photophysics of PbS Quantum Dot Films
Capped with Arsenic Sulfide Ligands. Adv. Energy Mater. 2014, 4, 1301547.
(29) Cordones, A. A.; Scheele, M.; Alivisatos, A. P.; Leone, S. R. Probing the
Interaction of Single Nanocrystals with Inorganic Capping Ligands: Time-
Resolved Fluorescence From CdSe–CdS Quantum Dots Capped with
Chalcogenidometalates. J. Am. Chem. Soc. 2012, 134, 18366–18373.
(64) Ip, A. H.; Thon, S. M.; Hoogland, S.; Voznyy, O.; Zhitomirsky, D.; Debnath, R.;
Levina, L.; Rollny, L. R.; Carey, G. H.; Fischer, A.; Kemp, K. W.; Kramer, I. J.;
Ning, Z.; Labelle, A. J.; Chou, K. W.; Amassian, A.; Sargent, E. H. Hybrid
Passivated Colloidal Quantum Dot Solids. Nat. Nanotech. 2012, 7, 577-582.
(30) Ning, Z.; Ren, Y.; Hoogland, S.; Voznyy, O.; Levina, L.; Stadler, P.; Lan, X.;
Zhitomirsky, D.; Sargent, E. H. All-Inorganic Colloidal Quantum Dot
Photovoltaics Employing Solution-Phase Halide Passivation. Adv. Mater. 2012,
24, 6295–6299.
(31) Zhitomirsky, D.; Voznyy, O.; Levina, L.; Hoogland, S.; Kemp, K. W.; Ip, A. H.;
Thon, S. M.; Sargent, E. H. Engineering Colloidal Quantum Dot Solids Within
and Beyond the Mobility-Invariant Regime. Nat. Commun. 2014, 5, 1–7.

210
(32) Dirin, D. N.; Dreyfuss, S.; Bodnarchuk, M. I.; Nedelcu, G.; Papagiorgis, P.;
Itskos, G.; Kovalenko, M. V. Lead Halide Perovskites and Other Metal Halide
Complexes as Inorganic Capping Ligands for Colloidal Nanocrystals. J. Am.
Chem. Soc. 2014, 136, 6550–6553.
(33) Buckley, J. J.; Couderc, E.; Greaney, M. J.; Munteanu, J.; Riche, C. T.; Bradforth,
S. E.; Brutchey, R. L. Chalcogenol Ligand Toolbox for CdSe Nanocrystals and
Their Influence on Exciton Relaxation Pathways. ACS Nano 2014, 8, 2512–2521.
(34) Zhang, H.; Jang, J.; Liu, W.; Talapin, D. V. Colloidal Nanocrystals with
Inorganic Halide, Pseudohalide, and Halometallate Ligands. ACS Nano 2014, 8,
7359-7369.
(35) Chen, O.; Chen, X.; Yang, Y.; Lynch, J.; Wu, H.; Zhuang, J.; Cao, Y. C.
Synthesis of Metal-Selenide Nanocrystals Using Selenium Dioxide as the
Selenium Precursor. Angew. Chem. Int. Ed. 2008, 47, 8638–8641.
(36) Jasieniak, J.; Smith, L.; Embden, J. V.; Mulvaney, P.; Califano, M. Re-
Examination of the Size-Dependent Absorption Properties of CdSe Quantum
Dots. J. Phys. Chem. C 2009, 113, 19468–19474.
(37) Yang, J.; Xue, C.; Yu, S. H.; Zeng, J. H.; Qian, Y. T. General Synthesis of
Semiconductor Chalcogenide Nanorods by Using the Monodentate Ligand n-
Butylamine as a Shape Controller. Angew. Chem. Int. Ed. 2002, 114, 4891–4894.
(38) Ji, X.; Copenhaver, D.; Sichmeller, C.; Peng, X. Ligand Bonding and Dynamics
on Colloidal Nanocrystals at Room Temperature: The Case of Alkylamines on
CdSe Nanocrystals. J. Am. Chem. Soc. 2008, 130, 5726–5735.
(39) Niu, G.; Li, N.; Wang, L.; Li, W.; Qiu, Y. Combined Post-Modification of Iodide
Ligands and Wide Band Gap ZnS in Quantum Dot Sensitized Solar Cells. Phys.
Chem. Chem. Phys. 2014, 1–6.
(40) Meyns, M.; Iacono, F.; Palencia, C.; Geweke, J.; Coderch, M. D.; Fittschen, U. E.
A.; Gallego, J. M.; Otero, R.; Juárez, B. H.; Klinke, C. Shape Evolution of CdSe
Nanoparticles Controlled by Halogen Compounds. Chem. Mater. 2014, 26, 1813–
1821.
(41) Fritzinger, B.; Capek, R. C.; Lambert, K.; Martins, J. C.; Hens, Z. Utilizing Self-
Exchange to Address the Binding of Carboxylic Acid Ligands to CdSe Quantum
Dots. JACS, 2010, 132, 10195-10201.
(42) Dzhagan, V. M.; Lokteva, I.; Himcinschi, C.; Kolny-Olesiak, J.; Valakh, M. Y.;
Schulze, S.; Zahn, D. R. T. The Influence of Pyridine Ligand Onto the Structure
and Phonon Spectra of CdSe Nanocrystals. Appl. Phys. 2011, 109, 084334.

211
(43) Kim, D. K.; Fafarman, A. T.; Diroll, B. T.; Chan, S. H.; Gordon, T. R.; Murray,
C. B.; Kagan, C. R. Solution-Based Stoichiometric Control Over Charge
Transport in Nanocrystalline CdSe Devices. ACS Nano 2013, 7, 8760-8770.
(44) MacDonald, B. I.; Martucci, A.; Rubanov, S.; Watkins, S. E.; Mulvaney, P.;
Jasieniak, J. J. Layer-by-Layer Assembly of Sintered CdSe
x
Te
1-x
Nanocrystal
Solar Cells. ACS Nano, 2012, 6, 5885-6004.
(45) Riha, S. C.; Fredrick, S. J.; Sambur, J. B.; Liu, Y.; Prieto, A. L.; Parkinson, B. A.
Photoelectrochemical Characterization of Nanocrystalline Thin-Film
Cu
2
ZnSnS
4
Photocathodes. ACS Appl. Mater. Interfaces 2011, 3, 58–66.
(46) Chen, Y.; Chuang, C. H.; Lin, K. C.; Shen, S.; McCleese, C.; Guo, L.; Burda, C.
Composition-Dependent Photoelectrochemical Properties of Nonstoichiometric
Cu
2
ZnSnS
4
Nanoparticles. J. Phys. Chem. C 2014, 118, 11954-11963.
(47) Kramer, T. J.; Babu, S. S.; Saeki, A.; Seki, S.; Aimi, J.; Nakanishi, T. CdSe
Nanocrystal/C
60
-Liquid Composite Material with Enhanced Photoelectrochemical
Performance. J. Mater. Chem. 2012, 22, 22370-22373.
(48) Van Embden, J.; Latham, K.; Duffy, N. W.; Tachibana, Y. Near-Infrared
Absorbing Cu
12
Sb
4
S
13
and Cu
3
SbS
4
Nanocrystals: Synthesis, Characterization,
and Photoelectrochemistry. JACS 2013, 135, 11562-11571.
(49) Norako, M. E.; Greaney, M. J.; Brutchey, R. L. Synthesis and Characterization of
Wurtzite-Phase Copper Tin Selenide Nanocrystals. JACS 2012, 134, 23-26.
(50) Chakrapani, V.; Baker, D.; Kamat, P. V. Understanding the Role of the Sulfide
Redox Couple (S
2–
/S
n
2–
) In Quantum Dot-Sensitized Solar Cells. J. Am. Chem.
Soc. 2011, 133, 9607–9615.
(51) Osterloh, F. E. Inorganic Nanostructures for Photoelectrochemical and
Photocatalytic Water Splitting. Chem. Soc. Rev. 2013, 42, 2294.
(52) Muthuswamy, E.; Zhao, J.; Tabatabaei, K.; Amador, M. M.; Holmes, M. A.;
Osterloh, F. E.; Kauzlarich, S. M. Thiol-Capped Germanium Nanocrystals:
Preparation and Evidence for Quantum Size Effects. Chem. Mater. 2014, 26,
2138–2146.
(53) Zhao, J.; Osterloh, F. E. Photochemical Charge Separation in Nanocrystal
Photocatalyst Films: Insights From Surface Photovoltage Spectroscopy. J. Phys.
Chem. Lett. 2014, 5, 782–786.
(54) Lagowski, J. Semiconductor Surface Spectroscopies: The Early Years. Surf. Sci.
1994, 299, 92-101.

212
(55) Kronik, L.; Shapira, Y. First-Principles Study of the Halide-Passivation Effects on
the Electronic Structures of CdSe Quantum Dots. Surf. Sci. Rep. 1999, 37, 1-206.
(56) Kronik, L.; Shapira, Y. Surface
Photovoltage
Spectroscopy
of
Semiconductor

Structures:
At
the
Crossroads
of
Physics,
Chemistry
and
Electrical

Engineering.

Surf. Interface Anal. 2001, 31, 954-965.
(57) Osterloh, F. E.; Holmes, M. A.; Zhao, J.; Chang, L.; Kawula, S.; Roehling, J. D.;
Moulé, A. J. P3HT:PCBM Bulk-Heterojunctions: Observing Interfacial and
Charge Transfer States with Surface Photovoltage Spectroscopy. J. Phys. Chem.
C 2014, 118, 14723-14731.
(58) Osterloh, F. E.; Holmes, M. A.; Chang, L.; Moule, A. J.; Zhao, J. Photochemical
Charge Separation in Poly(3-hexylthiophene) (P3HT) Films Observed with
Surface Photovoltage Spectroscopy. J. Phys. Chem. C 2013, 117, 26905–26913.
(59) Zhao, J.; Holmes, M. A.; Osterloh, F. E. Quantum Confinement Controls
Photocatalysis: A Free Energy Analysis for Photocatalytic Proton Reduction at
CdSe Nanocrystals. ACS Nano 2013, 7, 4316–4325.
(60) Wang, Y.; Zhang, Y.; Zhang, W. First-Principles Study of the Halide-Passivation
Effects on the Electronic Structures of CdSe Quantum Dots. RSC Adv. 2014, 4,
19302.
(61) Bard, A. J. Electrochemical Methods: Fundamentals and Applications; 2nd ed.;
Wiley: New York, 2001.
(62) Gao, J.; Johnson, J. C. Charge Trapping in Bright and Dark States of Coupled PbS
Quantum Dot Films. ACS Nano, 2012, 6, 3292–3303.
(63) Hartmann, L.; Kumar, A.; Welker, M.; Fiore, A.; Julien-Robant, C.; Gromova,
M.; Bardet, M.; Reiss, P.; Baxter, P. N. W.; Chandezon, F.; Pansu, R. B.
Quenching Dynamics in CdSe Nanoparticles: Surface-Induced Defects Upon
Dilution. ACS Nano, 2012, 6, 9033-9041.
(64) Norris, D. J.; Bawendi, M. G. Measurement and Assignment of the Size-
Dependent Optical Spectrum in CdSe Quantum Dots. Phys. Rev. B 1996, 53,
16338–16346.
(65) Klimov, V. I. Linear and Nonlinear Spectroscopy of Semiconductor Nanocrystals.
In Handbook on Nanostructered Materials and Nanotechnology; Nalwa, H., Ed.;
Elsevier: San Diego, 1999; Vol. 4, pp. 451–527.
(66) Klimov, V. I.; Mikhailovsky, A. A.; McBranch, D. W.; Leatherdale, C. A.;
Bawendi, M. G. Mechanisms for Intraband Energy Relaxation in Semiconductor
Quantum Dots: the Role of Electron-Hole Interactions. Phys. Rev. B 2000, 61,
R133349–R13352.

213
(67) Guyot-Sionnest, P.; Wehrenberg, B.; Yu, D. Intraband Relaxation in CdSe
Nanocrystals and the Strong Influence of the Surface Ligands. J. Chem. Phys.
2005, 123, 074709.
(68) Rivera, C. A.; Winter, N.; Harper, R. V.; Benjamin, I.; Bradforth, S. E. The
Dynamical Role of Solvent on the ICN Photodissociation Reaction: Connecting
Experimental Observables Directly with Molecular Dynamics Simulations. Phys.
Chem. Chem. Phys. 2011, 13, 8269-8283.
(69) Roberts, S. T.; Schlenker, C. W.; Barlier, V.; McAnally, R. E.; Zhang, Y.;
Mastron, J. N.; Thompson, M. E.; Bradforth, S. E. Observation of Triplet Exciton
Formation in a Platinum-Sensitized Organic Photovoltaic Device. J. Phys. Chem.
Lett. 2011, 2, 48–54.

214
Bibliography

Albero, J.; Martinez-Ferrero, E.; Iacopio, D.; Vidal-Ferran, A.; Palomares, E. Interfacial
Charge Transfer Dynamics in CdSe/Dipole Molecules Coated Quantum Dot Polymer
Blends. Phys. Chem. Chem. Phys. 2010, 12, 13047-13051.
Anderson, N. C.; Owen, J. S. Soluble, Chloride-Terminated CdSe Nanocrystals: Ligand
Exchange Monitored by
1
H and
31
P NMR Spectroscopy. Chem. Mater. 2013, 25, 69–76.
Anderson, N. C.; Hendricks, M. P.; Choi, J. J.; Owen, J. Ligand Exchange and the
Stoichiometry of Metal Chalcogenide Nanocrystals: Spectroscopic Observation of Facile
Metal-Carboxylate Displacement and Binding. J. Am. Chem. Soc. 2013, 135, 18536–
18548.

Araci, Z. O.; Shallcross, C. R.; Armstrong, N. R.; Saavedra, S. S. Potential-Modulated
Attenuated Total Reflectance Characterization of Charge Injection Processes in
Monolayer-Tethered CdSe Nanocrystals. J. Phys. Chem. Lett. 2010, 1, 1900-1905.

Bakulin, A. A.; Rao, A.; Pavelyev, V. G.; van Loosdrecht, P. H. M.; Pshenichnikov, M.
S.; Niedzialek, D.; Cornil, J.; Beljonne, D.; Friend, R. H. The Role of Driving Energy and
Delocalized States for Charge Separation in Organic Semiconductors. Science 2012, 335,
1340–1344.

Bard, Allen J., and Larry R. Faulkner. Electrochemical Methods: Fundamentals and
Applications. New York: John Wiley, 2001.

Barkhouse, D. A. R.; Pattantyus-Abraham, A. G.; Levina, L. ; Sargent, E. H. Thiols
Passivate Recombination Centers in Colloidal Quantum Dots Leading to Enhanced
Photovoltaic Device Efficiency. ACS Nano, 2008, 2, 2356–2362.

Baumgardner, W. J.; Choi, J. J.; Lim, Y. F.; Hanrath, T. SnSe Nanocrystals: Synthesis,
Structure, Optical Properties, and Surface Chemistry. J. Am. Chem. Soc. 2010, 132, 9519-
9521.

Beek, W. J. E.; Wienk, M. M.; Janssen, R. A. J. Efficient Hybrid Solar Cells from Zinc
Oxide Nanoparticles and a Conjugated Polymer. Adv. Mater. 2004, 16, 1009-1013.

Bensal, N.; Reynolds, L. X.; MacLachlan, A.; Lutz, T.; Ashraf, R. S.; Zhang, W.;
Nielsen, C. B.; McCulloch, I.; Rebois, D. G.; Kirchhartz, T.; Hill, M. S.; Molloy, K. C.;
Nelson, J.; Haque, S. A. Influence of Crystallinity and Energetics on Charge Separation
in Polymer-Inorganic Nanocomposite Films for Solar Cells. Sci. Rep. 2013, 3 1531.

Blouin, N.; Michaud, A.; Gendron, D.; Wakim, S.; Blair, E.; Neagu-Plesu, R.; Belletete,
M.; Durocher, G.; Tao, Y.; Leclerc, M. Toward a Rational Design of Poly(2, 7-
Carbazole) Derivatives for Solar Cells. J. Am. Chem. Soc., 2008, 130, 732-742.

215
Boucle, ́ J.; Chyla, S.; Shaffer, M. S. P.; Durrant, J. R.; Bradley, D. D. C.; Nelson, J.
Hybrid Solar Cells from a Blend of Poly(3-hexylthiophene) and Ligand-Capped TiO
2

Nanorods. Adv. Funct. Mater. 2008, 18, 622-633.

Brabec, C. J.; Dyakonov, V.; Scherf, U. Organic Photovoltaics: Materials, Device
Physics, and Manufacturing Technologies. Weinheim: Wiley-VCH, 2008.
Buckley, J. J.; Couderc, E.; Greaney, M. J.; Munteanu, J.; Riche, C. T.; Bradforth, S. E.;
Brutchey, R. L. Chalcogenol Ligand Toolbox for CdSe Nanocrystals and Their Influence
on Exciton Relaxation Pathways. ACS Nano 2014, 8, 2512–2521.
Buckley, J. J.; Greaney, M. J.; Brutchey, R. L. Ligand Exchange of Colloidal CdSe
Nanocrystals with Stibnates Derived from Sb
2
S
3
Dissolved in a Thiol-Amine Mixture.
Chem. Mater. 2014, 26, 6311-6317.
Burda, C.; Link, S.; Mohamed, M.; El-Sayed, M. The Relaxation Pathways of CdSe
Nanoparticles Monitored with Femtosecond Time-Resolution from the Visible to the IR:
Assignment of the Transient Features by Carrier Quenching. J. Phys. Chem. B, 2001,
105, 12286-12292.
Burkhart, B.; Khlyabich, P. P.; Thompson, B. C. Influence of the Acceptor Composition
on Physical Properties and Solar Cell Performance in Semi-Random Two-Acceptor
Copolymers. ACS Macro Lett. 2012, 1, 660-666.
Cardona, C. M.; Li, W.; Kaifer, A. E.; Stockdale, D.; Bazan, G. C. Electrochemical
Considerations for Determining Absolute Frontier Orbital Energy Levels of Conjugated
Polymers for Solar Cell Applications. Adv. Mater. 2011, 23, 2367-2371.
Carsten, B.; Szarko, J. M.; Lu, L.; Son, H. J.; He, F.; Botro, Y. Y.; Chen, L. X.; Yu, L.
Mediating Solar Cell Performance by Controlling the Internal Dipole Change in Organic
Photovoltaic Polymers. Macromolecules 2012, 45, 6390-6395.
Chakrapani, V.; Baker, D.; Kamat, P. V. Understanding the Role of the Sulfide Redox
Couple (S
2–
/S
n
2–
) In Quantum Dot-Sensitized Solar Cells. J. Am. Chem. Soc. 2011, 133,
9607–9615.
Chen, D.; Nakahara, A.; Wei, D.; Nordlund, D.; Russell, T. P. P3HT/PCBM Bulk
Heterojunction Organic Photovoltaics: Correlating Efficiency and Morphology. Nano
Lett. 2011, 11, 561-567.
Chen, H. C.; Lai, C. W.; Wu, I. C.; Pan, H. R.; Chen, I. W. P.; Peng, Y. K.; Liu, C. L.;
Chen, C. H.; Chou, P. T. Enhanced Performance and Air Stability of 3.2% Hybrid Solar
Cells: How the Functional Polymer and CdTe Nanostructure Boost the Solar Cell
Efficiency. Adv. Mater. 2011, 23, 5451-5455.

216
Chen, O.; Chen, X.; Yang, Y.; Lynch, J.; Wu, H.; Zhuang, J.; Cao, Y. C. Synthesis of
metal-selenide nanocrystals using selenium dioxide as the selenium precursor, Angew.
Chem. Int. Ed. 2008, 47, 8638−8441.
Chen, Y.; Chuang, C. H.; Lin, K. C.; Shen, S.; McCleese, C.; Guo, L.; Burda, C.
Composition-Dependent Photoelectrochemical Properties of Nonstoichiometric
Cu
2
ZnSnS
4
Nanoparticles. J. Phys. Chem. C 2014, 118, 11954-11963.
Chung, D. S.; Lee, J. S.; Huang, J.; Nag, A.; Ithurria, S.; Talapin, D. V. Low Voltage,
Hysteresis Free, and High Mobility Transistors From All-Inorganic Colloidal
Nanocrystals. Nano Lett. 2012, 12, 1813–1820.
Clarke, T. M.; Durrant, J. R. Charge Photogeneration in Organic Solar Cells. Chem. Rev.
2010, 110, 6736-6767.
Coffey, D. C.; Larson, B. W.; Hains, A. W.; Whitaker, J. B.; Kopidakis, N.; Boltalina, O.
V.; Strauss, S. H.; Rumbles, G. An Optimal Driving Force for Converting Excitons
Into Free Carriers in Excitonic Solar Cells. J. Phys. Chem. C 2012, 116, 8916–8923.
Colbert, A. E.; Janke, E. M.; Hsieh, S. T.; Subramaniyan, S.; Schlenker, C. W.; Jenekhe,
S. A.; Ginger, D. S. Hole Transfer from Low Band Gap Quantum Dots to Conjugated
Polymers in Organic/Inorganic Hybrid Photovoltaics. J. Phys. Chem. Lett. 2013, 4,
280−284.
Cordones, A. A.; Scheele, M.; Alivisatos, A. P.; Leone, S. R. Probing the Interaction of
Single Nanocrystals with Inorganic Capping Ligands: Time-Resolved Fluorescence From
CdSe–CdS Quantum Dots Capped with Chalcogenidometalates. J. Am. Chem. Soc. 2012,
134, 18366–18373.
Couderc, E.; Bruyant, N.; Fiore, A.; Chandezon, F.; Djurado, D.; Reiss, R.; Faure-
Vincent, J. Charge Transport in Poly(3-hexylthiophene):CdSe Nanocrystals Hybrid Thin
Films Investigated With Time-of-Flight Measurements Appl. Phys. Lett., 2012, 101,
133301.
Couderc, E.; Greaney, M. J.; Brutchey, R. L.; Bradforth, S. E.; Direct Spectroscopic
Evidence of Ultrafast Electron Transfer from a Low Band Gap Polymer to CdSe
Quantum Dots in Hybrid Photovoltaic Thin Films. J. Am. Chem. Soc. 2013, 135, 18418-
18426.
Cozzoli, P. D.; Pellegrino, T.; Manna, L. Synthesis, Properties and Perspectives of
Hybrid Nanocrystal Structures. Chem. Soc. Rev. 2006, 35, 1195-1208.
Dayal, S.; Kopidakis, N.; Olson, D. C.; Ginley, D. S.; Rumbles, G. Dayal, S.; Kopidakis,
N.; Olsen, D. C.; Ginley, D. S.; Rumbles, G. Photovoltaic Devices with a Low Band Gap
Polymer and CdSe Nanostructures Exceeding 3% Efficiency. Nano Lett. 2010, 10, 239-
242.

217
Dayal, S.; Reese, M. O.; Ferguson, A. J.; Ginley, D. S.; Rumbles, G.; Kopidakis, N. The
Effect of Nanoparticle Shape on the Photocarrier Dynamics and Photovoltaic Device
Performance of Poly(3-hexylthiophene):CdSe Nanoparticle Bulk Heterojunction Solar
Cells. Adv. Funct. Mater. 2010, 20, 2629-2635.
Deibel, C.; Strobel, T.; Dyankonov, V. Role of the Charge Transfer State in Organic
Donor-Acceptor Solar Cells. Adv. Mater. 2010, 22, 4097-4111.
Dennler, G.; Scharber, M. C.; Brabec, C. J. Polymer-Fullerene Bulk-Heterojunction Solar
Cells. Adv. Mater. 2009, 21, 1323-1338.
Dirin, D. N.; Dreyfuss, S.; Bodnarchuk, M. I.; Nedelcu, G.; Papagiogis, P.; Itskos, G.;
Kovalenko, M. V. Lead Halide Perovskites and Other Metal Halide Complexes as
Inorganic Capping Ligands for Colloidal Nanocrystals. J. Am. Chem. Soc. 2014, 136,
6550-6553.
Dong, A.; Ye, X.; Chen, J.; Kang, Y.; Gordon, T.; Kikkawa, J. M.; Murray, C. B. A
Generalized Ligand-Exchange Strategy Enabling Sequential Surface Functionalization of
Colloidal Nanocrystals. J. Am. Chem. Soc. 2011, 133, 998–1006.
Dzhagan, V. M.; Lokteva, I.; Himcinschi, C.; Kolny-Olesiak, J.; Valakh, M. Y.; Schulze,
S.; Zahn, D. R. T. The Influence of Pyridine Ligand Onto the Structure and Phonon
Spectra of CdSe Nanocrystals. Appl. Phys. 2011, 109, 084334.
Eftaiha, A. F.; Sun, J. P.; Hill, I. G.; Welch, G. C. Recent Advances of Non-Fullerene,
Small Molecular Acceptors for Solution Processed Bulk Heterojunction Solar Cells. J.
Mater. Chem. A 2014, 2, 1201-1213.
Etzold, F.; Howard, I. A.; Forler, N.; Cho, D. M.; Meister, M.; Mangold, H.; Shu, J.;
Hansen, M. R.; Muellen, K.; Laquai, F. The Effect of Solvent Additives on Morphology
and Excited-State Dynamics in PCPDTBT:PCBM Photovoltaic Blends. J. Am. Chem.
Soc. 2012, 134, 10569-10583.
Fafarman, A. T.; Koh, W. K.; Diroll, B. T.; Kim, D. K.; Ko, D. K.; Oh, S. J.; Ye, X.;
Doan-Nguyen, V.; Crump, M. R.; Reifsnyder, D. C.; Murray, C. B.; Kagan, C. R.
3Thiocyanate-Capped Nanocrystal Colloids: Vibrational Reporter of Surface Chemistry
and Solution-Based Route to Enhanced Coupling in Nanocrystal Solids. J. Am. Chem.
Soc. 2011, 133, 15753–15761.
Fiore, A.; Chandezon, F.; Djurado, D.; Reiss, R.; Faure-Vincent, J. Charge Transport in
Poly(3-hexylthiophene):CdSe Nanocrystals Hybrid Films Investigated with Time-of-
Flight Measurements. Appl. Phys. Lett. 2012, 101, 133301.
Fritzinger, B.; Capek, R. K.; Lambert, K.; Martins, J. C.; Hens, Z. Utilizing Self-
Exchange to Address the Binding of Carboxylic Acid Ligands to CdSe Quantum Dots. J.
Am. Chem. Soc. 2010, 132, 10195–10201.

218
Fu, W.; Shi, Y.; Qiu, W.; Wang, L.; Nan, Y.; Shi, M.; Li, H.; Chen, H. High Efficiency
Hybrid Solar Cells Using Post-Deposition Ligand Exchange by Monothiols. Phys. Chem.
Chem. Phys. 2012, 14, 12094-12098.
Gao, F.; Li, Z.; Wang, J.; Rao, A.; Howard, I. A.; Abrusci, A.; Massip, S.; McNeill, C.
R.; Greenham, N. C. Trap-Induced Losses in Hybrid Photovoltaics, ACS Nano 2014, 8,
3213–3221.
Gao, J.; Johnson, J. C. Charge Trapping in Bright and Dark States of Coupled PbS
Quantum Dot Films. ACS Nano, 2012, 6, 3292–3303.
Ginger, D. S.; Greenham, N. C. Charge Injection and Transport in Films of CdSe
Nanocrystals. J. Appl. Phys., 2000, 87, 1361-1368.
Goh, C.; Scully, S. R.; McGehee, M. D. Effects of Molecular Interface Modification in
Hybrid Organic-Inorganic Photovoltaic Cells. J. Appl. Phys. 2007, 101, 114503.
Gomes, R.; Hassinen, A.; Szczygiel, A.; Zhao, Q.; Vantomme, A.; Martins, J. C.; Hens,
Z. Binding of Phosphonic Acids to CdSe Quantum Dots: a Solution NMR Study. J. Phys.
Chem. Lett. 2011, 2, 145–152.
Grancini, G.; Biasiucci, M.; Mastria, R.; Scotognella, F.; Tassone, F.; Polli, D.; Gigli, G.;
Lanzani, G. Dynamic Microscopy Study of Ultrafast Charge Transfer in a Hybrid
P3HT/Hyperbranched CdSe Nanoparticle Blend for Photovoltaics. J. Phys. Chem. Lett.
2012, 3, 517-523.
Grancini, G.; Maiuri, M.; Fazzi, D.; Petrozza, A.; Egelhaaf, H.-J.; Brida, D.; Cerullo, G.;
Lanzani, G. Hot Exciton Dissociation in Polymer Solar Cells. Nat. Mater. 2013, 12, 29-
33.
Grancini, G.; Martino, N.; Antognazza, M. R.; Celebrano, M.; Egelhaaf, H-J.; Lanzani,
G. Influence of Blend Composition on Ultrafast Charge Generation and Recombination
Dynamics in Low Band Gap Polymer-Based Organic Photovoltaics. J. Phys. Chem. C,
2012, 116, 9838-9844.
Greaney, M. J.; Araujo, J.; Burkhart, B.; Thompson, B. C.; Brutchey, R. L. Novel Semi-
Random and Alternating Copolymer Hybrid Solar Cells Utilizing CdSe Multipods as
Versatile Acceptors. Chem. Commun. 2013, 49, 8602-8604.
Greaney, M. J. ; Brutchey, R. L. Ligand Engineering in Hybrid Polymer:Nanocrystal
Solar Cells. Materials Today 2015, 18, 31-38.
Greaney, M. J.; Das, S.; Webber, D. H.; Bradforth, S. E.; Brutchey, R. L. Improving
Open Circuit Potential in Hybrid P3HT:CdSe Bulk Heterojunction Solar Cells via
Colloidal tert-Butylthiol Ligand Exchange. ACS Nano 2012, 6, 4222-4230.

219
Green, M. A.; Emery, K.; Hishikawa, Y.; Warta, W. Solar Cell Efficiency Tables
(Version 37). Prog. Photovolt. Res. Appl. 2011, 19, 84-92.
Green, M. L. H. A New Approach to the Formal Classification of Covalent Compounds
of the Elements. J. Organomet. Chem. 1995, 500, 127–148.
Greenham, N. C.; Peng, X.; Alivisatos, A. P. Charge Separation and Transport in
Conjugated-Polymer/Semiconductor-Nanocrystal Composites Studied by
Photoluminescence Quenching and Photoconductivity. Phys. Rev. B. 1996, 54, 17628-
17637.
Günes, S.; Sariciftci, N. S. Hybrid Solar Cells. Inorg. Chim. Acta 2008, 361, 581-588.
Gur, I.; Fromer, N. A.; Chen, C. P.; Kanaras, A. G.; Alivisatos, A. P. Hybrid Solar Cells
with Prescribed Nanoscale Morphologies Based on Hyperbranched Semiconductor
Nanocrystals. Nano Lett. 2007, 7, 409-414.
Guyot-Sionnest, P.; Wehrenberg, B.; Yu, D. Intraband Relaxation in CdSe Nanocrystals
and the Strong Influence of the Surface Ligands. J. Chem. Phys. 2005, 123, 074709.
Habas, S. E.; Platt, H. A. S.; Van Hest, M. F. A. M.; Ginley, D. S. Low-Cost Inorganic
Solar Cells: From Ink to Printed Device. Chem. Rev. 2010, 6571-6594.
Hartmann, L.; Kumar, A.; Welker, M.; Fiore, A.; Julien-Robant, C.; Gromova, M.;
Bardet, M.; Reiss, P.; Baxter, P. N. W.; Chandezon, F.; Pansu, R. B. Quenching
Dynamics in CdSe Nanoparticles: Surface-Induced Defects Upon Dilution. ACS Nano,
2012, 6, 9033-9041.
Heinemann, M. D.; von Maydell, K.; Zutz, F.; Kolny-Olesiak, J.; Borchert, H.; Riedel, I.;
Parisi, J. Photo-Induced Charge Transfer and Relaxation of Persistent Charge Carriers in
Polymer/Nanocrystal Composites for Applications in Hybrid Solar Cells. Adv. Funct.
Mater. 2009, 19, 3788-3795.
Herrmann, D.; Niesar, S.; Scharsich, C.; Köhler, A.; Stutzmann, M.; Riedle, E. Role of
Structural Order and Excess Energy on Ultrafast Free Charge Generation in Hybrid
Polythiophene/Si Photovoltaics Probed in Real Time by Near-Infrared Broadband
Transient Absorption. J. Am. Chem. Soc. 2011, 113, 18220-18233.
Huynh, W. U., Dittmer, J. J., and Alivisatos, A. P. Hybrid Nanorod-Polymer Solar Cells.
Science 2002, 295, 2425-2457.
Hwang, I.-W.; Soci, C.; Moses, D.; Zhu, Z.; Waller, D.; Gaudiana, R.; Brabec, C. J.;
Heeger, A. J. Ultrafast Electron Transfer and Decay Dynamics in a Small Band Gap Bulk
Heterojunction Material. Adv. Mater. 2007, 19, 2307-2312.
Ip, A. H.; Thon, S. M.; Hoogland, S.; Voznyy, O.; Zhitomirsky, D.; Debnath, R.; Levina,
L.; Rollny, L. R.; Carey, G. H.; Fischer, A.; Kemp, K. W.; Kramer, I. J.; Ning, Z.;

220
Labelle, A. J.; Chou, K. W.; Amassian, A.; Sargent, E. H. Hybrid Passivated Colloidal
Quantum Dot Solids. Nat. Nanotech. 2012, 7, 577-582.
Itskos, G.; Othonos, A.; Rauch, T.; Tedde, S. F.; Hayden, O.; Kovalenko, M. V.; Heiss,
W.; Choulis, S. A. Optical Properties of Organic Semiconductor Blends with Near-
Infrared Quantum-Dot Sensitizers for Light Harvesting Applications. Adv. Energy Mater.
2011, 1, 802-812.
Jarzab, D.; Szendrei, K.; Yarema, M.; Pichler, S.; Heiss, W.; Loi, M. A. Charge-
Separation Dynamics in Inorganic-Organic Ternary Blends for Efficient Infrared
Photodiodes. Adv. Funct. Mater. 2011, 21, 1988-1992.
Jasieniak, J; Califano, M.; Watkins, S. E. Size-Dependent Valence and Conduction Band-
Edge Energies of Semiconductor Nanocrystals. ACS Nano 2011, 5, 5888-5902.
Jasieniak, J.; Smith, L.; Embden, J. V.; Mulvaney, P.; Califano, M. Re-Examination of
the Size-Dependent Absorption Properties of CdSe Quantum Dots. J. Phys. Chem. C
2009, 113, 19468–19474.
Jeltsch, K. F.; Schadel, M.; Bonekamp, J. B.; Niyamakom, P.; Rauscher, F.; Lademann,
H. W. A.; Dumsch, I.; Allard, S.; Scherf, U.; Meerholz, K. Enhanced Efficiency Hybrid
Solar Cells Using a Blend of Quantum Dots and Nanorods. Adv. Funct. Mater. 2012, 22,
397-404.
Ji, X.; Copenhaver, D.; Sichmeller, C.; Peng, X. Ligand Bonding and Dynamics on
Colloidal Nanocrystals at Room Temperature: The Case of Alkylamines on CdSe
Nanocrystals. J. Am. Chem. Soc. 2008, 130, 5726–5735.
Kaake, L. G.; Jasieniak, J. J.; Bakus, B. C.; Welch, G. C.; Moses, D.; Bazan, G. C.;
Heeger, A. J. J. Am. Chem. Soc. 2012, 134, 19828-19838.
Khylabich, P. P.; Burkhart, B.; Ng, C. F.; Thompson, B. C. Efficient Solar Cells From
Semi-Random P3HT Analogues Incorporating Diketopyrrolopyrrole. Macromolecules
2011, 44, 5079-5084.
Khylabich, P. P.; Burkhart, B.; Thompson, B. C. Compositional Dependence of the
Open-Circuit Voltage in Ternary Blend Bulk Heterojunction Solar Cells Based on Two
Donor Polymers. J. Am. Chem. Soc. 2012, 134, 9074-9077.
Kim, D. K.; Fafarman, A. T.; Diroll, B. T.; Chan, S. H.; Gordon, T. R.; Murray, C. B.;
Kagan, C. R. Solution-Based Stoichiometric Control Over Charge Transport in
Nanocrystalline CdSe Devices. ACS Nano 2013, 7, 8760-8770.
Kim, J. Y.; Kotov, N. A. Charge Transport Dilemma of Solution-Processed
Nanomaterials. Chem. Mater. 2013, 26, 134–152.

221
Kim, J. Y.; Voznyy, O.; Zhitomirsky, D.; Sargent, E. H. 25th Anniversary Article:
Colloidal Quantum Dot Materials and Devices: a Quarter-Century of Advances. Adv.
Mater. 2013, 25, 4986–5010.
Kippelen, B.; Bredas, J. Organic Photovoltaics. Energy Environ. Sci. 2009, 2, 251-261.
Klimov, V. I. Linear and Nonlinear Spectroscopy of Semiconductor Nanocrystals. In
Handbook on Nanostructered Materials and Nanotechnology; Nalwa, H., Ed.; Elsevier:
San Diego, 1999; Vol. 4, pp. 451–527.
Klimov, V. I.; McBranch, D. W.; Leatherdale, C. A.; Bawendi, M. G. Electron and Hole
Relaxation Pathways in Semiconductor Quantum Dots. Phys. Rev. B 1999, 60, 13740-
13749.
Klimov, V. I.; Mikhailovsky, A. A.; McBranch, D. W.; Leatherdale, C. A.; Bawendi, M.
G. Mechanisms for Intraband Energy Relaxation in Semiconductor Quantum Dots: the
Role of Electron-Hole Interactions. Phys. Rev. B 2000, 61, R133349–R13352.
Knowles, K. E.; Tice, D. B.; McArthur, E. A.; Solomon, G. C.; Weiss, E. A. Chemical
Control of the Photoluminescence of CdSe Quantum Dot-Organic Complexes with a
Series of Para-Substituted Aniline Ligands. J. Am. Chem. Soc. 2010, 132, 1041-1050.
Kovalenko, M. V.; Bodnarchuk, M. I.; Zaumseil, J.; Lee, J. S.; Talapin, D. V. Expanding
the Chemical Versatility of Colloidal Nanocrystals Capped with Molecular Metal
Chalcogenide Ligands. J. Am. Chem. Soc. 2010, 132, 10085–10092.
Kovalenko, M. V.; Scheele, M.; Talapin, D. V. Colloidal Nanocrystals with Molecular
Metal Chalcogenide Surface Ligands. Science 2009, 324, 1417–1420.
Kovalenko, S. A.; Dobryakov, A. L.; Ruthmann, J.; Ernsting, N. P. Femtosecond
Spectroscopy of Condensed Phases with Chirped Supercontinum Probing. Phys. Rev. A,
1999, 59, 2369-2384.
Kraabel, B.; Lee, C.H.; McBranch, D.; Moses, D.; Sariciftci, N.S. ; Heeger A.J. Ultrafast
Photoinduced Electron Transfer in Conducting Polymer-Buckminsterfullerene
Composites. Chem. Phys. Lett. 1993, 213, 389–394.
Kramer, T. J.; Babu, S. S.; Saeki, A.; Seki, S.; Aimi, J.; Nakanishi, T. CdSe
Nanocrystal/C
60
-Liquid Composite Material with Enhanced Photoelectrochemical
Performance. J. Mater. Chem. 2012, 22, 22370-22373.
Krissinel, E. B.; Agmon, N. Proton=-Transfer Reaction Dynamics Within the Human
Serum Albumin Protein. J. Comput. Chem., 1996, 17, 1085–1096.
Kronik, L.; Shapira, Y. First-Principles Study of the Halide-Passivation Effects on the
Electronic Structures of CdSe Quantum Dots. Surf. Sci. Rep. 1999, 37, 1-206.

222
Kronik, L.; Shapira, Y. Surface
Photovoltage
Spectroscopy
of
Semiconductor

Structures:
At
the
Crossroads
of
Physics,
Chemistry
and
Electrical
Engineering.

Surf.
Interface Anal. 2001, 31, 954-965.
Kucar, E.; Bucking, W.; Giernoth, R.; Nann, T. Determination of Defect States in
Semiconductor Nanocrystals by Cyclic Voltammetry. J. Phys. Chem. B 2005, 109,
20355-20360.
Lagowski, J. Semiconductor Surface Spectroscopies: The Early Years. Surf. Sci. 1994,
299, 92-101.
Lee, J. S.; Kovalenko, M. V.; Huang, J.; Chung, D. S.; Talapin, D. V. Band-Like
Transport, High Electron Mobility and High Photoconductivity in All-Inorganic
Nanocrystal Arrays. Nat. Nanotech. 2011, 6, 348–352.
Lefrancois, A.; Couderc, E.; Vincent, J. F.; Sadki, S.; Pron, A. Reiss, P. Effect of the
Treatment with Diamines and Dithiols on the Spectroscopic, Electrochemical and
Electrical Properties of CdSe Nanocrystals’ Thin Films. J. Mater. Chem. 2011, 21,
11524-11531.
Lek, J. Y.; Lam, Y. M.; Niziol, J.; Marzec, M. Understanding Polycarbazole-Based
Polymer:CdSe Hybrid Solar Cells. Nanotechnology, 2012, 23, 315401.
Li, Y.; Zhong, H.; Li, R.; Zhou, Y.; Yang, C.; Li, Y. High-Yield Fabrication and
Electrochemical Characterization of Tetrapodal CdSe, CdTe, and CdSe
x
Te
1-x

Nanocrystals. Adv. Funct. Mater. 2006, 16, 1705-1716.
Li, Z.; Gao, F.; Greenham, N. C.; McNeill, C. Comparison of the Operation of
Polymer/Fullerene, Polymer/Polymer, and Polymer/Nanocrystal Solar Cells: A Transient
Photocurrent and Photovoltage Study. Adv. Funct. Mater. 2011, 21, 1419-1431.
Lifshitz, E.; Dag, I.; Litvitn, I. D.; Hodes, G. Optically Detected Magnetic Resonance
Study of Electron/Hole Traps on CdSe Quantum Dots. J. Phys. Chem. B, 1998, 102,
9245-9250.
Lin, Y. Y.; Chu, T. H.; Li, S. S; Chuang, C. H.; Chang, C. H.; Su, W. F.; Chang, C. P.;
Chu, M. W.; Chen, C. W. Interfacial Nanostructuring on the Performance of
Polymer/TiO
2
Nanorod Bulk Heterojunction Solar Cells. J. Am. Chem. Soc. 2009, 131,
3644-3649.
Liu, J.; Kadnikova, E. N.; Liu, Y.; McGehee, M. D.; Frechet, J. M. J. Polythiophene
Containing Thermally Removable Solubilizing Groups Enhances the Interface and the
Performance of Polymer-Titania Hybrid Solar Cells. J. Am. Chem. Soc. 2004, 126, 9486-
9487.

223
Liu, W.; Lee, J. S.; Talapin, D. V. III–V Nanocrystals Capped with Molecular Metal
Chalcogenide Ligands: High Electron Mobility and Ambipolar Photoresponse. J. Am.
Chem. Soc. 2013, 135, 1349–1357.
Liu, Y.; Gibbs, M.; Puthussery, J.; Gaik, S.; Ihly, R.; Hillhouse, H. W.; Law, M.
Dependence of Carrier Mobility on Nanocrystal Size and Ligand Length in PbSe
Nanocrystal Solids. Nano Lett. 2010, 10, 1960–1969.
Liu, Y.; Tolentino, J.; Gibbs, M.; Ihly, R.; Perkins, C. L.; Liu, Y.; Crawford, N.;
Hemminger, J. C.; Law, M. PbSe Quantum Dot Field-Effect Transistors with Air-Stable
Electron Mobilities Above 7 cm
2
V
–1
s
–1
. Nano Lett. 2013, 13, 1578–1587.
Liu, Z.; Sun, Y.; Yuan, J.; Wei, H.; Huang, X.; Han, L.; Wang, W.; Wang, H. ; Ma, W.
High-Efficiency Hybrid Solar Cells Based on Polymer/PbS
x
Se
1-x
Nanocrystals Benefiting
from Vertical Phase Segregation. Adv. Mater. 2013, 25, 5772–5778.
Lokteva, I.; Radychev, N.; Witt, F.; Borchert, H.; Parisi, J.; Kolny-Olesiak, J. Surface
Treatment of CdSe Nanoparticles for Application in Hybrid Solar Cells: The Effect of
Multiple Ligand Exchange with Pyridine. J. Phys. Chem. C 2010, 114, 12784-12791.
Lunt,
R.
R.;
Osedach,
T.
P.;
Brown,
P.
R.;
Rowehl,
J.
A.;
Bulovic,
V.
Practical
Roadmap

and
Limits
to
Nanostructured
Photovoltaics.
Adv.
Mater.
2011,
23,
5712-‐5727.

Luther, J. M.; Pietryga, J. M. Stoichiometry Control in Quantum Dots: a Viable Analog
to Impurity Doping of Bulk Materials. ACS Nano 2013, 7, 1845–1849.
MacDonald, B. I.; Martucci, A.; Rubanov, S.; Watkins, S. E.; Mulvaney, P.; Jasieniak, J.
J. Layer-by-Layer Assembly of Sintered CdSe
x
Te
1-x
Nanocrystal Solar Cells. ACS Nano,
2012, 6, 5885-6004.
McArthur, E. A.; Morris-Cohen, A. J.; Knowles, K. E.; Weiss, E. A. Charge Carrier
Resolved Relaxation of the Excitonic State in CdSe Quantum Dots Probed with Near-
Infrared Transient Absorption Spectroscopy J. Phys. Chem. B 2010, 114, 14514−14520.
McGehee, M. D. Nanostructured Organic-Inorganic Hybrid Solar Cells. MRS. Bull. 2009,
34, 95-100.
Meulenberg, R. W.; Lee, J. R.I.; Wolcott, A.; Zhang, J. Z.; Terminello, L. J.; van Buuren,
T. Determination of the Exciton Binding Energy in CdSe Quantum Dots, ACS Nano
2009, 3, 325–330.
Meyns, M.; Iacono, F.; Palencia, C.; Geweke, J.; Coderch, M. D.; Fittschen, U. E. A.;
Gallego, J. M.; Otero, R.; Juárez, B. H.; Klinke, C. Shape Evolution of CdSe
Nanoparticles Controlled by Halogen Compounds. Chem. Mater. 2014, 26, 1813–1821.
Milliron D. J., Gur, I., and Alivisatos, A. P. Hybrid Organic-Nanocrystal Solar Cells.
MRS Bull. 2005, 30, 41-44.

224
Moon, S. J.; Baranoff, E.; Zakeeruddin, S. M.; Yeh, C. Y.; Diau, E. W. G.; Gratzel, M.;
Sivula, K. Enhanced Light Harvesting in Mesoporous TiO
2
/P3HT Hybrid Solar Cells
Using a Porphyrin Dye. Chem. Commun. 2011, 47, 8244-8246.
Moreels, I.; Fritzinger, B.; Martins, J. C.; Hens, Z. Surface Chemistry of Colloidal PbSe
Nanocrystals. J. Am. Chem. Soc. 2008, 130, 15081-15086.
Morgenstern, F. S. F.; Rao, A.; Böhm, M. L.; Kist, R. J. P.; Vaynzof, Y. ; Greenham N.
C. Ultrafast Charge- and Energy-Transfer Dynamics in Conjugated Polymer: Cadmium
Selenide Nanocrystal Blends. ACS Nano, 2014, 8, 1647–1654.
Morris-Cohen, A. J.; Donakowski, M. D.; Knowles, K. E.; Weiss, E. A. The Effect of a
Common Purification Procedure on the Chemical Composition of the Surfaces of CdSe
Quantum Dots Synthesized with Trioctylphosphine Oxide. J. Phys. Chem. C 2010, 114,
897-906.
Morris-Cohen, A. J.; Frederick, M. T.; Lilly, G. D.; McArthur, E. A.; Weiss, E. A.
Organic Surfactant-Controlled Composition of the Surfaces of CdSe Quantum Dots. J.
Phys. Chem. Lett. 2010, 1, 1078-1081.
Moulé, A. J.; Chang, L.; Thambidurai, C.; Vidu, R.; Stroeve, P. Hybrid Solar Cells: Basic
Principles and the Role of Ligands. J. Mater. Chem. 2012, 22, 2351-2368.
Mühlbacher, D.; Scharber, M.; Morana, M.; Zhu, Z.; Waller, D.; Gaudiana, R.; Brabec,
C. High Photovoltaic Performance of a Low-Bandgap Polymer. Adv. Mater. 2006, 18,
2884-2889.
Munro, A. M.; Plante, I. J. L.; Ng, M. S.; Ginger, D. S. Quantitative Study of the Effects
of Surface Ligand Concentration on CdSe Nanocrystal Photoluminescence. J. Phys.
Chem. C 2007, 111, 6220-6227.
Munro, A. M.; Zacher, B.; Graham, A.; Armstrong, N. R. Photoemission Spectroscopy of
Tethered CdSe Nanocrystals: Shifts in Ionization Potential and Local Vacuum Level As a
Function of Nanocrystal Capping Ligand. ACS Appl. Mater. Interfaces 2010, 2, 863-869.
Muthuswamy, E.; Zhao, J.; Tabatabaei, K.; Amador, M. M.; Holmes, M. A.; Osterloh, F.
E.; Kauzlarich, S. M. Thiol-Capped Germanium Nanocrystals: Preparation and Evidence
for Quantum Size Effects. Chem. Mater. 2014, 26, 2138–2146.
Nag, A.; Kovalenko, M. V.; Lee, J. S.; Liu, W.; Spokoyny, B.; Talapin, D. V. Metal-Free
Inorganic Ligands for Colloidal Nanocrystals: S
2–
, HS
–
, Se
2–
, HSe
–
, Te
2–
, HTe
–
, TeS3
2–
,
OH
–
, and NH
2–
As Surface Ligands. J. Am. Chem. Soc. 2011, 133, 10612–10620.
Nagaoka, H.; Colbert, A. E.; Strein, E.; Janke, E. M.; Salvador, M.; Schlenker, C. W.;
Ginger, D. S. Size-Dependent Charge Transfer Yields in Conjugated Polymer/Quantum
Dot Blends. J. Phys. Chem. C 2014, 118, 5710–5715.

225
Ning, Z.; Ren, Y.; Hoogland, S.; Voznyy, O.; Levina, L.; Stadler, P.; Lan, X.;
Zhitomirsky, D.; Sargent, E. H. All-Inorganic Colloidal Quantum Dot Photovoltaics
Employing Solution-Phase Halide Passivation. Adv. Mater. 2012, 24, 6295–6299.
Niu, G.; Li, N.; Wang, L.; Li, W.; Qiu, Y. Combined Post-Modification of Iodide
Ligands and Wide Band Gap ZnS in Quantum Dot Sensitized Solar Cells. Phys. Chem.
Chem. Phys. 2014, 1–6.
Niu, G.; Wang, L.; Gao, R.; Li, W.; Guo, X.; Dong, H.; Qiu, Y. Inorganic Halogen
Ligands in Quantum Dots: I
-
, Br
-
, Cl
-
and Film Fabrication Through Electrophoretic
Deposition. Phys. Chem. Chem. Phys. 2013, 15, 19595-19600.
Noone, K. M.; Subramaniyan, S.; Zhang, Q.; Cao, G.; Jenejhe, S. A.; Ginger, D. S.
Photoinduced Charge Transfer and Polaron Dynamics in Polymer and Hybrid
Photovoltaic Thin Films: Organic vs. Inorganic Acceptors. J. Phys. Chem. C 2011, 115,
24403-24410.
Norako, M. E.; Greaney, M. J.; Brutchey, R. L. Synthesis and Characterization of
Wurtzite-Phase Copper Tin Selenide Nanocrystals. JACS 2012, 134, 23-26.
Norman, Z. M.; Anderson, N. C.; Owen, J. S. Electrical Transport and Grain Growth in
Solution-Cast, Chloride-Terminated Cadmium Selenide Nanocrystal Thin Films. ACS
Nano 2014, 8, 7513–7521.
Norris, D. J.; Bawendi, M. G. Measurement and Assignment of the Size-Dependent
Optical Spectrum in CdSe Quantum Dots. Phys. Rev. B 1996, 53, 16338-16346.
Oh, S. J.; Berry, N. E.; Choi, J. H.; Gaulding, E. A.; Paik, T.; Hong, S. H.; Murray, C. B.;
Kagan, C. R. Stoichiometric Control of Lead Chalcogenide Nanocrystal Solids to
Enhance Their Electronic and Optoelectronic Device Performance. ACS Nano 2013, 7,
2413–2421.
Osterloh, F. E. Inorganic Nanostructures for Photoelectrochemical and Photocatalytic
Water Splitting. Chem. Soc. Rev. 2013, 42, 2294.
Osterloh, F. E.; Holmes, M. A.; Chang, L.; Moule, A. J.; Zhao, J. Photochemical Charge
Separation in Poly(3-hexylthiophene) (P3HT) Films Observed with Surface Photovoltage
Spectroscopy. J. Phys. Chem. C 2013, 117, 26905–26913.
Osterloh, F. E.; Holmes, M. A.; Zhao, J.; Chang, L.; Kawula, S.; Roehling, J. D.; Moulé,
A. J. P3HT:PCBM Bulk-Heterojunctions: Observing Interfacial and Charge Transfer
States with Surface Photovoltage Spectroscopy. J. Phys. Chem. C 2014, 118, 14723-
14731.
Owen, J. S.; Park, J.; Trudeau, P. E.; Alivisatos, A. P. Reaction Chemistry and Ligand
Exchange at Cadmium-Selenide Nanocrystal Surfaces. J. Am. Chem. Soc. 2008, 130,
12279-12281.

226
Park, S. H.; Roy, A.; Beaupre, S.; Cho, S.; Coates, N.; Moon, J. S.; Moses, D.; Leclerc,
M.; Lee, K.; Heeger, A. J. Bulk Heterojunction Solar Cells with Internal Quantum
Efficiency Approaching 100%. Nat. Photon. 2009, 5, 297-302.
Perez, M. D.; Borek, C.; Forrest, S. R.; Thompson, M. E. Molecular and Morphological
Influences on the Open Circuit Voltages of Organic Photovoltaic Devices. J. Am. Chem.
Soc. 2009, 131, 9281-9286.
Petkov, V.; Moreels, I.; Hens, Z.; Ren, Y. PbSe Quantum Dots: Finite, Off-
Stoichiometric, and Structurally Distorted. Phys. Rev. B 2010, 81, 241304.
Piliego, C.; Manca, M.; Kroon, R.; Yarema, M.; Szendrei, K.; Andersson, M. R.; Heiss,
W.; Loi, M. A. Charge Separation Dynamics in a Narrow Band Gap Polymer-PbS
Nanocrystal Blend for Efficient Hybrid Solar Cells. J. Mater. Chem. 2012, 22, 24411-
24416.
Piris, J.; Dykstra, T. E.; Bakulin, A. A.; van Loosdrecht, P. H. M.; Knulst, W.; Tuan
Trinh, M.; Schins, J. M.; Siebbeles, L. D. A. Photogeneration and Ultrafast Dynamics of
Excitons and Charges in P3HT/PCBM Blends. J. Phys. Chem. C 2009, 113, 14500-
14506.
Potscavage, W. J.; Yoo, S.; Kippelen, B. Origin of the Open-Circuit Voltage in
Multilayer Heterojunction Organic Solar Cells. Appl. Phys. Lett. 2008, 93, 193308.
Press, W. H.; Teukolsky, S. A.; Vetterling, W. T.; Flannery, B. P. Numerical Recipes in
C: the Art of Scientific Computing; Cambridge University Press, 1988.
Qian, L.; Yang, J.; Zhou, R.; Tang, A.; Zheng, Y.; Tseng, T. K.; Bera, D.; Xue, J.;
Holloway, P. H. Hybrid Polymer-CdSe Solar Cells with a ZnO Nanoparticle Buffer
Layer for Improved Efficiency and Lifetime. J. Mater. Chem. 2011, 21, 3814-3817.
Querner, C.; Reiss, P.; Sadki, S.; Zagorska, M.; Pron, A. Size and Ligand Effects on the
Electrochemical Responses of CdSe Nanocrystals. Phys. Chem. Chem. Phys. 2005, 7,
3204-3209.
Radychev, N.; Lokteva, I.; Witt, F.; Kolny-Olesiak, J.; Borchert, H.; Parisi, J. Physical
Origin of the Impact of Different Nanocrystal Surface Modifications on the Performance
of CdSe/P3HT Hybrid Solar Cells. J. Phys. Chem. C 2011, 115, 14111-14122.
Rath, T.; Edler, M.; Haas, W.; Fischereder, A.; Moscher, S.; Scheck, A.; Trattnig, R.;
Sezen, M.; Mauthner, G.; Pein, A.; Meischler, D.; Bartl, K.; Saf, R.; Bansal, N.; Haque,
S. A.; Hofer, F.; List, E. J. W.; Trimmel, G. A Direct Route Towards Polymer/Copper
Indium Sulfide Nanocomposite Solar Cells. Adv. Energy Mater. 2011, 1, 1046-1050.
Reiss, P.; Couderc, E.; De Girolamo, J.; Pron, A. Conjugated Polymers/Semiconductor
Nanocrystals Hybrid Materials-Preparation, Electrical Transport Properties and
Applications. Nanoscale 2011, 3, 446-489.

227
Ren, S.; Chang, L. Y.; Lim, S. K.; Zhao, J.; Smith, M.; Zhao, N.; Bulovic, V.; Bawendi,
M.; Gradecak, S. Inorganic-Organic Hybrid Solar Cells: Bridging Quantum Dots to
Conjugated Polymer Nanowires. Nano Lett. 2011, 11, 3998-4002.
Riha, S. C.; Fredrick, S. J.; Sambur, J. B.; Liu, Y.; Prieto, A. L.; Parkinson, B. A.
Photoelectrochemical Characterization of Nanocrystalline Thin-Film
Cu
2
ZnSnS
4
Photocathodes. ACS Appl. Mater. Interfaces 2011, 3, 58–66.
Rivera, C. A.; Winter, N.; Harper, R. V.; Benjamin, I.; Bradforth, S. E. The Dynamical
Role of Solvent on the ICN Photodissociation Reaction: Connecting Experimental
Observables Directly with Molecular Dynamics Simulations. Phys. Chem. Chem. Phys.
2011, 13, 8269-8283.
Roberts, S. T.; Schlenker, C. W.; Barlier, V.; McAnally, R. E.; Zhang, Y.; Mastron, J. N.;
Thompson, M. E.; Bradforth, S. E. Observation of Triplet Exciton Formation in a
Platinum-Sensitized Organic Photovoltaic Device. J. Phys. Chem. Lett. 2011, 2, 48-54.
Rosen, E. L.; Buonsanti, R.; Llordes, A.; Sawvel, A. M.; Milliron, D. J.; Helms, B. A.
Exceptionally Mild Reactive Stripping of Native Ligands From Nanocrystal Surfaces by
Using Meerwein’s Salt. Angew. Chem. Int. Ed. 2011, 51, 684–689.
Rosenthal, S. J.; McBride, J.; Pennycook, S. J.; Feldman, L. C. Synthesis, Surface
Studies, Composition and Structural Characterization of CdSe, Core/Shell, and
Biologically Active Nanocrystals. Surf. Sci. Rep. 2007, 62, 111-157.
Scharber, M. C.; Muhlbacher, D.; Koppe, M.; Denk, P.; Waldauf, C.; Heeger, A. J.;
Brabec, C. J. Design Rules for Donors in Bulk-Heterojunction Solar Cells – Towards
10% Energy-Conversion Efficiency. Adv. Mater. 2006, 18, 789-794.
Seo, J.; Cho, M. J.; Lee, D.; Cartwright, A. N.; Prasad, P. N. Efficient Heterojunction
Photovoltaic Cell Utilizing Nanocomposites of Lead Sulfide Nanocrystals and a Low-
Bandgap Polymer. Adv. Mater. 2011, 23, 3984-3988.
Service, R. F. Outlook Brightens for Plastic Solar Cells. Science 2011, 332, 293.
Shaheen, S. E.; Ginley, D. S.; Jabbour, G. E. Organic-Based Photovoltaics: Toward Low-
Cost Power Generation. MRS Bull. 2005, 30, 10-19.
Sharma, S. N.; Sharms, H.; Singh, G.; Shivaprasad. Studies of Interaction of Amines with
TOPO/TOP Capped CdSe Quantum Dots: Role of Crystallite Size and Oxidation
Potential. Mater. Chem. Phys. 2008, 110, 471-480.
Soreni-Harari, M.; Yaacobi-Gross, N.; Steiner, D.; Aharoni, A.; Banin, U.; Millo, O.;
Tessler, N. Tuning Energetic Levels in Nanocrystal Quantum Dots through Surface
Manipulations. Nano Lett. 2008, 8, 678-684.

228
Sun, B.; Greenham, N. C. Improved Efficiency of Photovoltaics Based on CdSe
Nanorods and poly(3-Hexylthiophene) Nanofibers. Phys. Chem. Chem. Phys. 2006, 8,
3557-3560.
Tagliazucchi, M.; Tice, D. B.; Sweeney, C. M.; Morris-Cohen, A. J.; Weiss, E. A.
Ligand-Controlled Rates of Photoinduced Electron Transfer in Hybrid CdSe
Nanocrystal/Poly(viologen) Films. ACS Nano 2011, 5, 9907-9917.
Talapin, D. V.; Lee, J. S.; Kovalenko, M. V.; Shevchenko, E. V. Prospects of Colloidal
Nanocrystals for Electronic and Optoelectronic Applications. Chem. Rev. 2010, 110, 389-
458.
Tautz, R.; Da Como, E.; Limmer, T.; Feldmann, J.; Egelhaaf, H.- J.; von Hauff, E.;
Lemaur, V.; Beljonne, D.; Yilmaz, S.; Dumsch, I.; Allard, S.; Scherf, U., Nat. Commun.
2012, 3, 970.
ten Cate, S.; Schins, J. M.; Siebbeles, L. D. A. Origin of Low Sensitizing Efficiency of
Quantum Dots in Organic Solar Cells. ACS Nano 2012, 6, 8983-8988.
Tsoi, W. C.; Spencer, S. J.; Yang, L.; Ballantyne, A. M.; Nicholson, P. G.; Turnbull, A.;
Shard, A. G.; Murphy, C. E.; Bradley, D. D. C.; Nelson, J.; Kim, J.-S. Effect of
Crystallization on the Electronic Energy Levels and Thin Film Morphology of
P3HT:PCBM Blends. Macromolecules 2011, 44, 2944-2952.
Tsokkou, D.; Papagiorgis, P.; Protesescu, L.; Kovalenko, M. V.; Choulis, S. A.;
Christofides, C.; Itskos, G.; Othonos, A. Photophysics of PbS Quantum Dot Films
Capped with Arsenic Sulfide Ligands. Adv. Energy Mater. 2014, 4, 1301547.
Van Embden, J.; Latham, K.; Duffy, N. W.; Tachibana, Y. Near-Infrared Absorbing
Cu
12
Sb
4
S
13
and Cu
3
SbS
4
Nanocrystals: Synthesis, Characterization, and
Photoelectrochemistry. JACS 2013, 135, 11562-11571.
Vandewal, K.; Albrecht, S.; Hoke, E. T.; Graham, K. R.; Widmer, J.; Douglas, J. D.;
Schubert, M.; Mateker, W. R.; Bloking, J. T.; Burkhard, G. F.; Sellinger, A.; Fréchet, J.
M. J.; Amassian, A.; Riede, M. K.; McGehee, M. D.; Neher, D.; Salleo, A. Efficient
Charge Generation by Relaxed Charge-transfer States at Organic Interfaces. Nature
Mater. 2014, 13, 63–68.
Vilan, A.; Shanzer, A.; Cahen, D. Molecular Control over Au/GaAs Diodes. Nature
2000, 404, 166-168.
Wang, C.; Shim, M.; Guyot-Sionnest, P. Electrochromic Nanocrystal Quantum Dots.
Science 2001, 291, 2390-2392.
Wang, Y.; Zhang, Y.; Zhang, W. First-Principles Study of the Halide-Passivation Effects
on the Electronic Structures of CdSe Quantum Dots. RSC Adv. 2014, 4, 19302.

229
Webber, D. H.; Brutchey, R. L. Ligand Exchange on Colloidal CdSe Nanocrystals Using
Thermally Labile Tert-Butylthiol for Improved Photocurrent in Nanocrystal Films. J. Am.
Chem. Soc. 2012, 134, 1085–1092.
Wiebeler, C.; Tautz, R.; Feldmann, J.; von Hauff, E.; Da Como, E.; Schumacher, S. J.
Phys. Chem. B 2013, 117, 4454-4460.
Witt, F.; Kruszynska, M.; Borchert, H.; Parisi, J. Charge Transfer Complexes in Organic-
Inorganic Hybrid Blends for Photovoltaic Applications Investigated by Light-Induced
Electron Spin Resonance Spectroscopy. J. Phys. Chem. Lett. 2010, 1, 2999-3003.
Wuister, S. F.; Donega, C. M.; Meijerink, A. Influence of Thiol Capping on the Exciton
Luminescence and Decay Kinetics of CdTe and CdSe Quantum Dots. J. Phys. Chem. B
2004, 108, 17393-17397.
Xu, T.; Qiao, Q. Conjugated Polymer-Inorganic Semiconductor Hybrid Solar Cells.
Energy Environ. Sci. 2011, 4, 2700-2720.
Yamamoto, S.; Ohkita, H.; Benten, H.; Ito, S. Phys. Chem. C 2012, 116, 14804-14810.
Yang, J.; Xue, C.; Yu, S. H.; Zeng, J. H.; Qian, Y. T. General Synthesis of
Semiconductor Chalcogenide Nanorods by Using the Monodentate Ligand n-Butylamine
as a Shape Controller. Angew. Chem. Int. Ed. 2002, 114, 4891–4894.
Yang, S.; Prendergast, D.; Neaton, J. B. Tuning Semiconductor Band Edges Energies for
Solar Photocatalysis via Surface Ligand Passivation. Nano Lett. 2012, 12, 383-388.
Yu, W. W.; Qu, L.; Guo, W.; Peng, X. Experimental Determination of the Extinction
Coefficient of CdTe, CdSe and CdS Nanocrystals. Chem. Mater. 2003, 15, 2854-2860.
Yuan, M.; Zhitomirsky, D.; Adinolfi, V.; Voznyy, O.; Kemp, K. W.; Ning, Z.; Lan, X.;
Xu, J.; Kim, J. Y.; Dong, H.; Sargent, E. H. Doping Control via Molecularly Engineered
Surface Ligand Coordination. Doping Control via Molecularly Engineered Surface
Ligand Coordination. Adv. Mater. 2013, 25, 5586-5592.
Zabet-Khosousi, A.; Dhirani, A. A. Charge Transport in Nanoparticle Assemblies. Chem.
Rev. 2008, 108, 4072–4124.
Zacher, B.; Gantz, J. L.; Richards, R. E.; Armstrong, N. R. Organic Solar Cells-At the
Interface. J. Phys. Chem. Lett. 2013, 4, 1949-1952.
Zanella, M.; Maserati, L.; Pernia Leal, M.; Prato, M.; Lavieville, R.; Povia, M.; Krahne,
R.; Manna, L. Atomic Ligand Passivation of Colloidal Nanocrystal Films via Their
Reaction with Propyltrichlorosilane. Chem. Mater. 2013, 25, 1423–1429.
Zhang, H.; Jang, J.; Liu, W.; Talapin, D. V. Colloidal Nanocrystals with Inorganic
Halide, Pseudohalide, and Halometallate Ligands. ACS Nano 2014, 8, 7359-7369.

230
Zhang, W.; Wu, Y.; Bao, Q.; Gao, F.; Fang, J. Morphological Control for Highly
Efficient Inverted Polymer Solar Cells Via the Backbone Design of Cathode Interlayer
Materials. Adv. Energy Mater. 2014, 4, 1400359.
Zhao, J.; Holmes, M. A.; Osterloh, F. E. Quantum Confinement Controls Photocatalysis:
A Free Energy Analysis for Photocatalytic Protoneduction at CdSe Nanocrystals. ACS
Nano 2013, 7, 4316–4325.
Zhao, J.; Osterloh, F. E. Photochemical Charge Separation in Nanocrystal Photocatalyst
Films: Insights From Surface Photovoltage Spectroscopy. J. Phys. Chem. Lett. 2014, 5,
782–786.
Zhitomirsky, D.; Voznyy, O.; Levina, L.; Hoogland, S.; Kemp, K. W.; Ip, A. H.; Thon, S.
M.; Sargent, E. H. Engineering Colloidal Quantum Dot Solids Within and Beyond the
Mobility-Invariant Regime. Nat. Commun. 2014, 5, 1–7.
Zhou, Y.; Eck, M.; Kruger, M. Bulk-Heterojunction Hybrid Solar Cells Based on
Colloidal Nanocrystals and Conjugated Polymers. Energy Environ. Sci. 2010, 3, 1851-
1864.
Zhou, Y.; Eck, M.; Veit, C.; Zimmermann, B.; Rauscher, F.; Niyamalom, P.; Yilmaz, S.;
Dumsch, I.; Allard, S.; Scherf, U.; Kruger, M. Efficiency Enhancement for Bulk-
Heterojunction Hybrid Solar Cells Based on Acid Treated CdSe Quantum Dots and Low
Bandgap Polymer PCPDTBT. Sol. Energy Mater. Sol. Cells 2011, 95, 1232-1237.
Zhou, Y.; Eck, M.; Men, C.; Rauscher, F.; Niyamakom, P.; Yilmaz, S.; Dumsch, I.;
Allard, S.; Scherf, U.; Kruger, M. Efficient Polymer Nanocrystal Hybrid Solar Cells by
Improved Nanocrystal Composition. Sol. Energy Mater. Sol. Cells 2011, 95, 3227-3232.
Zhou, Y.; Riehle, F. S.; Yuan, Y.; Schleiermacher, H. F.; Niggemann, M.; Urban, G. A.;
Kruger, M. Improved Efficiency of Hybrid Solar Cells Based on Non Ligand-Exchanged
CdSe Quantum Dots and poly(3-Hexylthiophene). Appl. Phys. Lett. 2010, 96, 013304.
Zhou, R.; Stadler, R.; Xie, D.; Cao, W.; Zheng, Y.; Yang, Y.; Plaisant, M.; Holloway, P.
H.; Schanze, K. S.; Reynolds, J. R.; Xue, J. Enhancing the Efficiency of Solution-
Processed Polymer:Colloidal Nanocrystal Hybrid Photovoltaic Cells Using Ethanedithiol
Treatment. ACS Nano 2013, 7, 4846-4854.

Abstract (if available)

Abstract Understanding how nanocrystal surface chemistry influences the transport of photogenerated charges is of paramount importance for the development of successful nanocrystal-based photovoltaics. As synthesized, semiconductor nanocrystals typically possess electrically insulating long-chain organic molecules that offer excellent colloidal stability, yet present an obstacle for efficient charge transport. With this in mind, we have developed novel organic and inorganic surface passivation strategies that enable quantitative displacement of the native synthesis ligands by replacement (i.e., ligand exchange) with small organic molecules and inorganic compounds, while still maintaining colloidal stability. This approach facilitates the production of optoelectronic devices from nanocrystal “inks” that exhibit lower resistance and better performance than non ligand-exchanged nanocrystals. ❧ We have utilized nanocrystals after ligand exchange with small organic molecules to fabricate hybrid polymer:nanocrystal solar cells, and we have observed a strong dependence of device performance on the nature of the nanocrystal surface ligand. Hybrid solar cells fabricated from poly(3-hexylthiophene) and CdSe nanocrystals ligand-exchanged with tert-butylthiol (tBT) show significantly improved device performance as a result of enhanced short current density (JSC) and increased open-circuit potential (VOC) relative to nonligand-exchanged or pyridine exchanged nanocrystals. Through a combination of (spectro)electrochemistry and ultrafast photoluminescence lifetime measurements, we have shown that the enhancement in VOC can be attributed to ligand-induced modulation of the conduction band edge (i.e., LUMO energy) of the CdSe nanocrystals, and the increase in JSC can be explained by a reduction in the number of nanocrystal surface traps. We have shown the dependence on nanocrystal LUMO energy on surface ligand to be robust for a series of different organic ligands. Employing the tBT ligand-exchange in combination with a spectrally complementary polymer enables the fabrication of high-efficiency hybrid solar cells that demonstrate state-of-the-art power conversion efficiencies in excess of 4%. ❧ To gain better insight into the influence of nanocrystal surface chemistry at the hybrid nanocrystal:polymer interface, we have used ultrafast transient absorption spectroscopy to probe the electron transfer dynamics between the low band gap polymer, poly[2,6-(4,4-bis(2-ethylhexyl)-4H-cyclopenta[2,1-b:3,4-b’]-dithiophene)-alt-4,7-(2,1,3-benzothiadiazole)] (i.e., PCPDTBT), and CdSe nanocrystals. We find that ultrafast electron transfer from PCPDTBT to CdSe(tBT) occurs in under 40 fs. This electron transfer rate and yield is heavily dependent on nanocrystal surface chemistry, with pyridine and butylamine ligand-exchanged CdSe exhibiting slower and lower yielding electron transfer compared to tBT treatment. ❧ In addition to organic surface ligands, we have explored the use of inorganic cadmium halide complexes as ligands for CdSe nanocrystals. We developed a simple ligand exchange procedure for replacing native ligands from the surface of CdSe nanocrystals with cadmium halide (CdX₂

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Expanding the library of surface ligands for semiconductor nanocrystal synthesis and photovoltaic applications

PDF

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

PDF

Semiconductor inks for solution processed electronic thin‐films

PDF

Quantifying the surface chemistry of semiconductor nanocrystals spanning covalent to ionic materials

PDF

Non-radiative energy transfer for photovoltaic solar energy conversion: lead sulfide quantum dots on silicon nanopillars

PDF

Non-linear surface spectroscopy of photoswitches annd photovoltaics

PDF

The chemistry of polymorphism in semiconductor nanocrystals

PDF

Exciton dynamics in photovoltaic materials

PDF

Excited state dynamics of organic photovoltaics probed using ultrafast spectroscopy

PDF

Understanding organic and inorganic semiconductor surfaces by coherent nonlinear spectroscopy

PDF

Synthesis and surface modification of perovskite-based nanocrystals for use in high energy density nanocomposites

PDF

Solution‐phase synthesis and deposition of earth‐abundant metal chalcogenide semiconductors

PDF

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

PDF

Interfacial vibrational and electronic excited states of photovoltaic materials probed using ultrafast nonlinear spectroscopy

PDF

Charge separation in transition metal and quantum dot systems

PDF

Energy management in organic photovoltaics

PDF

Synthesis and characterization of metal chalcogenide semiconductor nanocrystals using dialkyl dichalcogenide precursors

PDF

Molecular and morphological effects on the operational parameters of organic solar cells

PDF

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

PDF

Singlet fission in disordered acene films: from photophysics to organic photovoltaics

Asset Metadata

Creator Greaney, Matthew J. (author)

Core Title Nanocrystal surface engineering as a route towards improved photovoltaics

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Chemistry

Publication Date 07/09/2015

Defense Date 07/09/2015

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

Tag CdSe,ligand engineering,nanocrystals,OAI-PMH Harvest,photovoltaics,surface chemistry

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Brutchey, Richard L. (committee chair), Bradforth, Stephen E. (committee member), Nakano, Aiichiro (committee member)

Creator Email greaney@usc.edu,greaney19@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-590793

Unique identifier UC11301937

Identifier etd-GreaneyMat-3580.pdf (filename),usctheses-c3-590793 (legacy record id)

Legacy Identifier etd-GreaneyMat-3580.pdf

Dmrecord 590793

Document Type Dissertation

Format application/pdf (imt)

Rights Greaney, Matthew J.

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