University of Southern California Dissertations and Theses

Molecular aspects of photoconversion processes in organic solar cells

(USC Thesis Other)

Molecular aspects of photoconversion processes in organic solar cells

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content
MOLECULAR ASPECTS OF PHOTOCONVERSION PROCESSES IN ORGANIC
SOLAR CELLS

by
Maria Dolores Perez

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

May 2009

Copyright 2009 Maria Dolores Perez
Dedication

To my mother for making me
who I am, Mariana, Pono and
to my love Federico

A mi madre por hacerme lo que
soy, Mariana, Pono y mi amor,
Federico

ii

iii

Acknowledgments
I am forever indebted to Prof. Mark Thompson from whom I learned so much. He
introduced me into the fascinating world of organic photovoltaics and taught me the
wonders of scientific research with unique patience. I am a better scientist thanks to him.
I want to thank my committee members Prof. Hanna Reisler and Prof. Dan Dapkus.
Also Prof. Surya Prakash and Steve Bradforth for their support during the qualifying
process. I will always be greatly thankful to Prof. Curt Wittig whose long talks over
coffee helped me overcome the hardest stone on my Ph.D. road.
The long road to finishing my Ph.D. started long time ago and many people
accompanied me on that process. My first advisor, Prof. Sara Bilmes played a very
important role in introducing me to the world of scientific research. There, my co-worker
Coco Calvo taught me that “you cannot plan an experiment forever; you have to do it one
day, why not today”. My chemistry mentor Silvia Olleros must also take responsibility
for convincing me that “now was the time for graduate studies”. I am forever thankful
for the great friends life has given me, Analia, Mariana and Alcira for their constant
support from the distance, I always knew I could count on their cheerful words when I
needed them. Living in a foreign country was easy thanks to the people I got to work
with and became some of my dearest friends. Dr. Elizabeth Mayo was my mentor and
my friend, always offering the perfect advice and taking care of me. Cody Schlenker is
my friend, confessor and a great coworker. I will forever keep the talks and mate sharing
in my heart. Judy Hom was always there for me and my Ph. D. is possible thanks to her,
iv

a dear friend. I can undoubtedly say that Libby, Cody and Judy are the best American (or
North American or “yanquis”) people I have ever met and I could make it six years in
this country thanks to them. Carsten Borek lightened my days and made cool molecules
for me to discover cool stuff about organic solar cells. Also, I want to thank Peter
Djurovic who introduced me into the photophysics world and was the first one to see a
connection between solar cells and Carsten’s porphyrins. Thanks Peter for your long
talks (or monologues sometimes), for constantly reminding me that I could be better.
Thanks to the wonderful undergraduate students I had the pleasure to work with, Krystal
Sly, Erin Morrison and Grace Cheng are some smart and fun girls. Thanks to Siyi and
Kristin for sharing the work and bearing with my moods. I must also thank all the other
members of the Thompson group that made every day a different one.
My family is the most supportive people I have. They are always there and it gives
me great peace of mind knowing I can always count on all of you. Starting from my aunt
and uncle (Tia Betty and Tio Andres) that wanted to know where I worked so much that
made it all the way here to be with me; to my wonderful cousins Pablo, Mini and Ceci
that are my forever friends. Carlos became part of my family later in my life and I could
not be more thankful for having him around. He is a role model and his sole presence
brings peace to the people around.
Thanks Pa for showing me that you can accomplish anything as long as you set your
mind to it and that it is never late for doing what you like and I will never forget you
shedding a few tears at the airport on a hot January. Mariana, hermana, who despite
hating me being afar, always wanted the best for me and has been a great support during
v

some tough situations. You enlightened me on your knowledge about being a successful
woman in a tough environment. My Ponin kept me grounded. He is my favorite person
in the world, there is nothing else to it, you are my brother, my friend, I always want you
next to me. Thanks Frida for making him happy. Fede, I would not have done it without
you, you are my soul mate and the owner of my heart. You are the entire world I know,
you made this road possible by teaching me how to be happy. You are the main
responsible for my achievement and I owe this title and honor to you. And finally, I must
thank my mother. She represents all I am in life, she taught me that I could do this; she
made me who I am. I learn something from you every day; you taught me that if I knew
what I wanted I could achieve it, that happiness and success are possible by perseverance,
hard work and doing what gives me pleasure. You taught me you can always learn good
things from bad things and that study and dedication will make you a better person.
Thank you.

vi

Table of Contents
Dedication ...................................................................................................................... ii
Acknowledgments........................................................................................................... iii
List of Figures ................................................................................................................. ix
List of Tables ................................................................................................................ xiv
Abstract ................................................................................................................... xvi
Chapter 1. Introduction to organic photovoltaics .............................................................1
1.1 Why solar cell research? ...................................................................................1
1.2 Small molecule organic photovoltaics .............................................................4
1.2.1 A brief review of organic photovoltaic efficiencies ....................................4
1.2.2 Physics of solar cells applied to organic photovoltaics ...............................6
1.2.3 Steps of photoelectric conversion in small molecule organic
photovoltaics (sm-OPVs) ...........................................................................14
1.2.4 Materials for small molecule organic photovoltaic ...................................22
1.2.5 Back to the sun ...........................................................................................25
1.3 Brief overlook of thesis work .........................................................................26
1.4 Chapter 1 References ......................................................................................28
Chapter 2. Organic photovoltaics using metallo-tetraphenyl porphyrins complexes
as donor layers .............................................................................................30
2.1 Introduction .......................................................................................................30
2.2 Experimental......................................................................................................32
2.3 Results and Discussion ......................................................................................34
2.3.1 Optical properties of metal-TPBP films ....................................................34
2.3.2 Electrochemistry and energy levels of PtTPBP and PdTPBP ...................36
2.3.3 Use of PtTPBP and PdTPBP as donor layers in OPVs ..............................39
2.3.4 Optimization of the C
60
layer thickness for PtTPBP cells .........................43
2.3.5 Thickness dependence of metal-TPBP and the exciton diffusion length
limitation ....................................................................................................45
2.3.4 Solar cells fabricated by solution process of PtTPBP as donor layer ........50
2.3.5 PtTPNP optical properties and energy levels.............................................52
2.3.6 PtTPNP as donor layer in solar cells and thickness dependence studies ...55
2.4 Conclusions .......................................................................................................60
2.5 Chapter 2 References .........................................................................................62
Chapter 3. On the molecular nature of the open circuit voltage in organic
photovoltaics – A comprehensive study ......................................................66
vii

3.1 Introduction .......................................................................................................66
3.2 Experimental......................................................................................................68
3.3 The physical origin of the V
oc
............................................................................69
3.4 Results and Discussion ......................................................................................72
3.4.1 Donor analysis ...........................................................................................72
3.4.2 Acceptor analysis .......................................................................................83
3.5 Conclusions .......................................................................................................86
3.6 Chapter 3 References .........................................................................................88
Chapter 4. Charge collection at the organic/metal interface ..........................................92
4.1 Introduction .......................................................................................................92
4.2 Experimental......................................................................................................93
4.3 Results and Discussion ......................................................................................94
4.3.1 Charge collection from the cathode material perspective ..........................94
4.3.2 Charge collection from the organic material or buffer layer perspective 103
4.4 Conclusions .....................................................................................................109
4.5 Chapter 4 References .......................................................................................111
Chapter 5. Exciplex quenching of a luminescent cyclometallated Platinum complex
by extremely poor Lewis bases. ................................................................113
5.1 Introduction .....................................................................................................113
5.2 Experimental....................................................................................................115
5.2.1 Synthesis ..................................................................................................115
5.2.2 Quenching measurements ........................................................................116
5.2.3 Instrumentation ........................................................................................116
5.2.4 X-ray Crystallography .............................................................................117
5.3 Results and Discussion ....................................................................................117
5.3.1 X-Ray diffraction studies .........................................................................117
5.3.2 Electrochemical studies ...........................................................................121
5.3.3 Optical properties .....................................................................................122
5.3.4 Theoretical analysis .................................................................................126
5.3.5 Emission quenching by poor Lewis bases ...............................................127
5.4 Conclusions .....................................................................................................133
5.5 Chapter 5 References .......................................................................................135
Chapter 6. Experimental techniques for the fabrication and testing of small
molecule organic photovoltaics .................................................................138
6.1 Device fabrication ...........................................................................................138
6.1.1 ITO cleaning ............................................................................................138
6.1.2 Vacuum deposition and materials ............................................................139
6.2 Device testing ..................................................................................................142
6.2.1 Dark and simulated white light J-V characteristics ..................................142
6.2.2 Spectral responsivity ................................................................................143
viii

6.3 Spectral Mismatch calculation and standardization of light intensity
procedures ..............................................................................................................145
Bibliography .................................................................................................................149

ix

List of Figures
Figure 1.1: Cost-efficiency analysis for first- (I), second- (II), and third- (III)
generation PV technologies. (Figure taken from Ref 1) ............................ 2
Figure 1.2: Simple solar cell equivalent circuit ............................................................. 6
Figure 1.3: From p-n to D/A junctions .......................................................................... 7
Figure 1.4: J-V characteristics for both dark and light conditions (red). Power as
function of voltage is also depicted (blue). Power = J x V ....................... 8
Figure 1.5: (a) spectral responsivity for a number of inorganic materials
(www.newport.com); (b) absorption spectra of typical films of
photoactive molecules Copper Phthalocyanine (CuPc) in black and
Fullerene (C
60
) in red overlapped to a standard AM1.5G solar
spectrum (blue) ........................................................................................ 14
Figure 1.6: Mechanistic representation of photoconversion in sm-OPV ..................... 15
Figure 1.7: Förster and Dexter transfer mechanisms. .................................................. 18
Figure 1.8: The introduction of a “buffer” transparent layer between the acceptor
layer and the metal cathode prevents damage and defect sites and
enhances charge collection. ..................................................................... 21
Figure 1.9: Typical small molecules for organic photovoltaic applications ................ 24
Figure 1.10: Black body radiation and solar spectrum (www.newport.com). ............. 26
Figure 1.11: Different air masses according to the sunlight angle of incidence (a)
and spectrum variation depending on sky cloud coverage (b)
(www.newport.com). ............................................................................... 26
Figure 2.1: PtTPBP, PdTPBP and PtTPNP molecular structures ................................ 32
Figure 2.2: Absorption spectra of PtTPBP (—) and PdTPBP (-•-) in solution
(black) and as solid film (red) .................................................................. 36
Figure 2.3: Cyclic voltammetry for PtTPBP in DCE. Voltage scale has been
corrected versus the reference potential of the redox couple
Ferrocinium/Ferrocene as internal standard. ............................................ 37
x

Figure 2.4: Cyclic voltammetry for PdTPBP in DCE. Voltage scale has been
corrected versus the reference potential of the redox couple
Ferrocinium/Ferrocene as internal standard. ............................................ 38
Figure 2.5: HOMO and LUMO energy levels for all photoactive compounds in
typical solar cells. Energy values were obtained from literature.
19, 20
.... 39
Figure 2.6: (a) Current density vs. voltage characteristics of
ITO/CuPc(400Å)/C
60
(400Å)/BCP(100Å)/Al(1000Å) (dotted line) and
ITO/PtTPBP(150Å)/C
60
(400Å)/BCP(100Å)/Al(1000Å) (solid line)
under dark (thinner lines) and simulated AM1.5G illumination at 1sun
intensity corrected to accurately match the solar spectrum. (b)
Quantum efficiency (dotted line) for the PtTPBP photovoltaic cell
shown along with the absorption spectra of PtTPBP and C
60
. ................. 41
Figure 2.7: Current density vs. voltage characteristics of ITO/PtTPBP(150Å)/C
60
(
x Å)/BCP(100Å)/Al(1000Å) under simulated 0.5 sun light intensity ..... 45
Figure 2.8. Current density vs. layer thickness of ITO/metal-TPBP(x=100Å -
400Å)/C
60
(400Å)/ BCP(100Å)/Al(1000Å) devices under simulated
AM1.5G illumination at 50mW/cm
2
........................................................ 47
Figure 2.9. Absorption and emission spectra of PtTPBP in degassed toluene
solution (solid line) and solid film (circles) at room temperature. .......... 50
Figure 2.10. Comparison of devices with solution processed (SP) and vapor
deposited (VD) donor layer. Dark and illuminated current density vs.
voltage characteristics corresponding to the following devices:
PtTPBP(SP)(150Å)/C
60
(400Å)/BCP(100Å)/Al (solid line); and
PtTPBP(VD)(150Å)/C
60
(400Å)/BCP(100Å)/Al (filled circles) under
simulated 1 sun intensity, AM1.5G illumination ..................................... 52
Figure 2.11: Absorption spectra of PtTPNP as a solid film (red line) and in
solution dissolved in dichloroethane (black line). ................................... 53
Figure 2.12: (a) Cyclic voltammetry of PtTPNP in DCE with DcFc as the internal
reference and tetrabutyl ammonium hexafluorophosphate as the
conducting electrolyte; (b) Cyclic voltammetry of DcFc with Fc as
the internal reference in the same conditions as before. .......................... 55
Figure 2.13: Comparison of current density vs. voltage characteristics of
ITO/donor (150Å)/C
60
(400Å)/BCP(100Å)/Al(1000Å) under
simulated 1 sun light intensity ................................................................. 57
xi

Figure 2.14: Quantum efficiency (circled line) for the PtTPNP photovoltaic cell
shown along with the absorption spectra of PtTPNP and C
60
films. ....... 57
Figure 2.15: J-V characteristics for devices
ITO/PtTPNP(xÅ)/C
60
(400Å)/BCP(100Å)/Al(1000Å) devices under
simulated 1 sun AM1.5G light intensity. ................................................. 59
Figure 3.1: Molecular structures of donors and the corresponding HOMO (blue)
and LUMO (red) for four of them. Phenyl rings are shown as yellow
space filling surfaces. ............................................................................... 74
Figure 3.2: Dark (closed circles) and Light (open circles) logarithmic J-V
characteristics of tetracene and rubrene based OPVs (see Table 3.1
for device structure). ................................................................................ 75
Figure 3.3: Dark (closed circles) and Light (open circles) logarithmic J-V
characteristics of CuPc, PtTPBP and PtTPNP donor OPVs (see Table
3.1 for device structure). .......................................................................... 76
Figure 3.4: Dark (closed circles) and Light (open circles) logarithmic J-V
characteristics of NPD based OPVs (see Table 3.1 for device
structure). ................................................................................................. 82
Figure 3.5: Molecular structures of acceptors. ............................................................ 83
Figure 3.6: Dark (closed circles) and Light (open circles) logarithmic J-V
characteristics of C
60
, PTCBI and PTCDI based OPVs (see Table 3.3
for device structure). ................................................................................ 84
Figure 4.1: (a) J-V characteristics under approximately 50 mW/cm
2
AM1.5G
simulated illumination and (b) spectral responsivity of CuPc
(200Å)/C
60
(400Å)/BCP (100Å) cells with Al (- ■-) and Ag (- ●-) as
cathode materials. Absorption spectra of CuPc (200Å) and C
60

(400Å) films are overlayed in (b). .......................................................... 95
Figure 4.2: Metal reflectances. (Reproduced from
http://en.wikipedia.org /w iki/Reflectance) ................................................ 96
Figure 4.3: Depiction of experiments performed, total thickness of cathode is
maintained at 1000Å for all devices. ....................................................... 98
xii

Figure 4.4: Device performance with different proportions of Al and Ag for a
CuPc(200 Å)/C
60
(400 Å)/BCP(100 Å)/Al(X Å)/Ag(1000-X Å)
device. Al is always deposited first as shown in Figure 4.3 (a).
Yellow and blue triangles above the plot illustrate the relative
thicknesses of Ag and Al, respectively. ................................................... 99
Figure 4.6: Representation of Al layer on top of the organics with small amount of
Ag addition (a) and inhomogeneous layer of Al next to organics plus
a thick layer of Ag (b) ............................................................................ 102
Figure 5.1: ORTEP diagram for one unique NPt molecule from the asymmetric
unit cell................................................................................................... 120
Figure 5.2: ORTEP diagram showing top view of NPt dimers with short (3.465 Å)
(a), and long (4.476 Å) (b) Pt-Pt separations. ........................................ 120
Figure 5.3: Cyclic voltammogram of NPt recorded in dichloromethane with 0.1 M
TBAPF. .................................................................................................. 121
Figure 5.4: Absorption spectra showing the negative solvatochromic effect for
NPt (a) and ppyPt(acac) (b). .................................................................. 123
Figure 5.5: Absorption and emission (λ
exc
= 450 nm) of NPt in cyclohexane at
room temperature. Inset: Excitation and emission in toluene (77K)
and polystyrene (RT). ............................................................................ 124
Figure 5.6: Excitation and emission spectra at 77K in 3-methylpentane (3MP) and
for the neat solid (a), and at 77 K in 2-methyltetrahydrofuran (b). ....... 125
Figure 5.7: HOMO and LUMO electronic density .................................................... 127
Figure 5.8: Potential surfaces of electron density for the triplet excited state of NPt
(left) and ppyPt(acac) (right). ................................................................ 130
Figure 5.9: Stern–Volmer plots for quenching of NPt by halogenated benzenes in
cyclohexane. ........................................................................................... 132
Figure 5.10: Second-order fit to Stern-Volmer data using toluene as quencher. ....... 133
Figure 6.1: Image of vacuum chamber and schematics of system. ............................ 139
Figure 6.2: Evaporation sources used for the different materials. Source:
www.rdmathis.com ................................................................................ 140
xiii

Figure 6.3: Typical substrate with organic photoactive layers and cathode metal to
make 15 devices with the same structure. .............................................. 141
Figure 6.4: Simulated AM1.5G illumination J-V testing station ............................... 143
Figure 6.5: Picture of tunable wavelength energy source and testing of spectral
responsivity. ........................................................................................... 144
Figure 6.5: Spectral response of reference detectors (a) and geometrical drawings
(b). (Images from www.hamamatsu.com) ............................................. 148

xiv

List of Tables
Table 2.1: Comparison of parameters for
ITO/CuPc(200Å)/C
60
(400Å)/BCP(100Å)/Al(1000Å) and
ITO/metal-TPBP(150Å)/C
60
(400Å)/BCP(100Å)/Al(1000Å) (Metal=Pt
or Pd) under simulated AM1.5G light at 1sun corrected according to
NREL standardized procedures. ................................................................. 42
Table 2.2: Performance comparison of
ITO/PtTPBP(150Å)/C
60
(400Å)/BCP(100Å) /Al(1000Å) and
ITO/PtTPBP(150Å)/C
60
(400Å)/BCP(100Å)/Ag(1000Å) under
simulated AM1.5G illumination at 50mW/cm
2
. ......................................... 42
Table 2.4. Photovoltaic performance parameters for devices
ITO/metal-TPBP(xÅ)/C
60
(400Å)/BCP(100Å)/Al(1000Å) under
simulated AM1.5G illumination at 50mW/cm
2
. ......................................... 48
Table 2.5: Comparison of parameters for ITO/donor/C
60
(400Å)/
BCP(100Å)/Al(1000Å) under simulated AM1.5G light at 1sun
corrected according to NREL standardized procedures. ............................. 56
Table 2.6: Photovoltaic performance parameters for devices
ITO/PtTPNP(xÅ)/C
60
(400Å)/BCP(100Å)/Al(1000Å) under simulated 1
sun AM1.5G illumination. .......................................................................... 60
Table 3.1: Dark and light characteristic parameters for all devices. Structure:
ITO/Donor/C
60
(400 Å)/BCP (100 Å)/Al (1000 Å). ................................... 78
Table 3.2: Comparison of calculated values. J
SO
values are in mA/cm
2
and the
remaining data are in Volts. “Film” refers to the form of the donor thin
film, as determined by X-ray diffraction, Am = amorphous, PC =
polycrystalline. ............................................................................................ 79
Table 3.3: Dark and light characteristic parameters for all devices. Structure:
ITO/CuPc(400 Å)/Acceptor/BCP (100 Å)/Al (1000 Å). ............................ 84
Table 3.4: Comparison of calculated values. J
SO
values are in mA/cm
2
and the
remaining data are in Volts. ........................................................................ 85
Table 4.2: Parameters for the performance of solar cells with different cathodes
materials under AM1.5G simulated illumination with intensity of
approximately of 50 mW/cm
2
for CuPc(200 Å)/C
60
(400 Å)/BCP(100
Å)/Metal(1000 Å) devices. ....................................................................... 102
xv

Table 5.1: NPt extinction coefficients for different solvents ..................................... 122
Table 5.2: Stern-Volmer constants and Gutmann donor numbers (DN) ................... 129

xvi

Abstract
The field of organic photovoltaics research has received a great deal of attention in
the past years. Global climate changes are renewing the need for cleaner alternative
energies that will prevent increased effects in global warming. Solar photoconversion
arises as a natural option for the satisfaction of large energy demands with diminished
impact on the environment. However, the high production costs of the current solar
conversion technologies have prevented the widespread use of the sun as energy source.
Costs reductions will be possible by the use of organic materials as photoactive materials
since manufacturing processes are less demanding that those for the inorganic
counterparts. However, current laboratory efficiencies do not match those required for
commercialization and a huge leap into improved photoconversion must be achieved.
Small molecule organic photovoltaics are a relatively new field and much is still to be
known about the processes involved in solar-to-electric conversion. The introduction of
new materials will provide the insight required for learning more about those processes.
Organic synthesis will play a fundamental role in the evolution of this area.
It is the goal of this thesis to provide new insight into the processes that control the
performance of small molecule organic photovoltaics. By introducing new materials
systematically we can study, for example, the effects of the excited state nature or the
extension of the π system into the exciton diffusion length, or the molecular geometry in
the resultant value of the open circuit voltage. A simple model for the description of the
open circuit voltage will be presented that provides good evidence of the effects of
molecular shape and geometry on the solar cell performance. Studies of the effects of
xvii

charge collection at the organic/cathode interface will also be presented. The
experimental procedures involved in the preparation and testing of the small molecule
organic solar cells are presented and standard procedures for reporting efficiencies are
highlighted. Finally, an additional relevant work on the photophysics of a newly
synthesized cyclometallated platinum complex is presented.
1

Chapter 1. Introduction to organic photovoltaics
1.1 Why solar cell research?
The sun constitutes the major energy source of planet earth, it powers life by means
of photosynthesis and it is essential to support living forms. Solar radiation represented
the original and unique source of energy for the beginning of life on earth until evolution
and civilization growth and new technologies generated new resources to power the new
energy needs. In this constantly changing world, a variety of energy sources are needed
to ensure modern life. In the midst of climate global warming, there is much concern and
effort in developing new energetic technologies that replace contaminating sources
whenever possible. It is estimated that the earth receives more than ~15TW of
electromagnetic energy daily from the sun. The global energy consumption for the recent
years was reported to be in the same value range for one entire year. The sun then arises
as a natural source of clean energy that impacts the earth daily with enormous amount of
power. The sun energy can be harvested to generate both thermal heat and electricity.
Electricity is necessary to power electronics, some types of cars, electrical heating,
nocturnal illumination, etc, and represents a major alternative energy source for active
research and development of new technologies. Some disadvantages of this natural
source of energy include the fact that the sunlight is intermittent such that energy needs to
be stored during the day to be used at night. It also varies with locations, weather
conditions, time of year/day, etc., and does not deliver that much energy to any one place
at any one time and large surface area devices or light collectors are required.

Figure 1.1: Cost-efficiency analysis for first- (I), second- (II), and third- (III) generation PV
technologies. (Figure taken from Ref 1)
Photovoltaic devices harvest solar energy into electrical currents. Three generations
of photovoltaics have been highlighted and they are depicted in Figure 1.1. The first
generation consists mainly of high purity single junction Si devices. These devices are
highly efficient with efficiencies close to their theoretical limit of 33%. A typical Si solar
cell can achieve efficiencies of up to 25%. These cells currently dominate the
photovoltaic market but their elevated costs have prevented this technology to popularize
and substitute the current forms of energy sources. Elevated costs are a consequence of
the processing techniques for obtaining monocrystalline and high purity Si.
Second generation photovoltaics include those devices that employ inorganic
semiconductors with less requirements for processing such that manufacturing costs are
2

3

much reduced. These are normally made in thin polycrystalline films and efficiencies are
much lower than those of the first generation. Second generation photoactive materials
include CdTe, Copper Indium Gallium Selenide (CIGS), amorphous Si and others.
Active research is being developed in this area for the past years and a steady increase on
the efficiencies is being observed. They have been recently introduced into the
commercial market but efficiencies are still lower than the monocrystalline counterparts.
Also, in opposition to Si, the photoactive semiconductors components are limited in
quantities in nature and elevated costs of such components will be detrimental for its
commercialization.
Finally, those devices that involve organic molecules take part on the third
generation. The projection in Figure 1.1 establishes that these solar cells should comprise
much reduced production costs. However, efficiencies are still very low and a large
performance improvement is required in order to become commercially available. Third
generation photovoltaics include dye-sensitized solar cells (DSSC or Grätzel cells),
polymeric and small molecule solar cells. Organic materials represent reduced costs,
both because of the reduced material production cost and because device manufacturing
techniques are much less price demanding. Organic photovoltaics (OPVs) preparation
techniques include solution processing (dip coating, spin coating, ink jet printing) and
vacuum deposition. Organic materials have been successfully proven to be relatively
stable and with low production costs in the Organic Light Emitting Diode (OLED) field.
Many electronics with OLED displays are now commercially available and processing
4

plants have been installed and are fully functional. The same processing plants involved
in OLEDs could be used in the production of OPVs.
The promising reduction of manufacturing costs is tied with the possibility of
achieving elevated efficiencies. The versatility and millions of organic molecular
possibilities by synthetic tuning of the desired properties offer great expectations in
increasing the performances. The field of OPVs is a relatively recent field of study and
much knowledge is still required to improve these efficiencies. Many research groups
around the globe are currently dedicated to the study of the higher efficient DSSC and on
polymeric PVs. Less is known about the mechanisms that dominate small molecule
double heterojunction solar cells. It is the goal of this thesis work to contribute to the
overall understanding of small molecule solar cells by analyzing the different steps of
photovoltaic processes by the introduction of a number of different materials.
1.2 Small molecule organic photovoltaics
1.2.1 A brief review of organic photovoltaic efficiencies
The ultimate goal of photovoltaics research is their use in powering of daily life
needs. In order to understand how far the scientific community has evolved into the
understanding of OPVs, it is necessary to understand what factors limit the present
photoconversion efficiencies. However, many research groups focus mainly on the
technological aspect of the photovoltaic problem and forget that in order to leap into the
required efficiencies for commercialization it is necessary to understand the factors that
control sun light harvesting. A brief summary of the up-to-date progress in organic solar
5

cells is presented here so that we can later move on into the discussions of those scientific
questions that need to be answered in order to improve the performance.
Maximum efficiencies in OPVs lie in the range of ~ 5 % for illuminations of 1 sun.
For both polymeric and small molecule devices, these numbers have been reached in the
years 2004-2005 and no advance on performance has been reported since then.
Two papers report an efficiency of ~ 5 % in the polymeric field, both based on the
same system: P3HT-PCBM blends, however, different preparation conditions were
used.
2, 3
A more recent report by Heeger’s group reported an increase of efficiency up to
5.5 %, record efficiency for polymer solar cells.
4
It is also worth mentioning that many
groups working on this area have also reported fairly high efficiencies, like Padinger et.
al. in 2003
5
with a reported 3.5 % efficiency or like Li et. al. with 4.4%.
6

For small molecules solar cells, the best reported efficiencies have been originated in
the Forrest group at Princeton University. The highest of 5.7 % is based on the concept
of blended heterojunctions plus a tandem architecture design
7
in 2004. When a hybrid
planar-mixed molecular heterojunction is used in the photovoltaic cell, an efficiency of 5
% was obtained.
8
Also, high efficiencies of 3.6% have been reported from the same
group in 2001 on the same system but with planar layers.
9
Other groups using different
donor layers have achieved somewhat high efficiencies but still well below those reported
by Forrest. Some new donor layers employed successfully include pentacene
10
and
tetracene
11
with 2.7 % and 2.3 % reported efficiencies under 1 sun; and oligothiophene
12

with a 3.4 % efficiency. Finally, when both systems are combined in a tandem cell as
that reported by Janssen et. al. an efficiency of 4.6 % is obtained.
13

1.2.2 Physics of solar cells applied to organic photovoltaics
1.2.2.1 Linear J-V response
A typical solar cell can be described by an equivalent circuit that reflects the
performance of the photovoltaic device both in the dark and under illumination
conditions. This model used for inorganic photovoltaics is also applicable for organic
cells and is depicted in Figure 1.2.

Figure 1.2: Simple solar cell equivalent circuit
The device is mainly dominated by a diode component that originates a diode current
J
D
. This means that in the dark, there is a directional flow of current such that there is
only a flow of electrons beyond certain applied voltages. For inorganic cells, this diode
is achieved by the presence of a p-n heterojunction. Semiconductors p-type and n-type
are materials with excess of positive and negative carriers, respectively, either by doping
or intrinsically. For organic semiconductors, a diode can be achieved by an intrinsic
semiconductor-type heterojunction. Organic intrinsic semiconductors are namely donor
materials (p-type) or acceptors (n-type). Donor and Acceptors are molecules that present
energy level offsets such that one layer (donor) donates an electron to the other (acceptor)
6

as represented in Figure 1.3 for the D/A junction analogous to the inorganic p-n junction.
The approach to D/A junctions must be done in terms of individual molecules and their
energy levels. Different degrees of intermolecular overlap appear for a number of
organic molecules that modify the energetics, however, band formation is almost never
an accurate description of the organic solid state, in particular for amorphous films. Even
when band formation has been described for crystalline organic materials, typical solar
cells consist of amorphous films such that a molecular orbital model is necessary for
describing the D/A heterojunction.

Figure 1.3: From p-n to D/A junctions

For simple, well-behaved solar cells, the diode is in most cases coupled to a series
resistant component (R
s
) and to a parallel resistance (R
p
). Current density (J) versus
applied voltage (V) characteristics is described by the Shockley analysis for diodes plus
the additional resistive components as described in Equation 1.1.
7

) ( 1
(
exp V J
R
V
nkT
JR V q
J
R R
R
J
Ph
p
s
S
p s
p
−
⎪
⎭
⎪
⎬
⎫
⎪
⎩
⎪
⎨
⎧
+
⎥
⎦
⎤
⎢
⎣
⎡
− ⎟
⎠
⎞
⎜
⎝
⎛ −
+
=
1.1
where R
s
and R
p
are the series and parallel resistances, respectively, J
S
is the saturation
current, q is the fundamental charge, n is the diode ideality factor and J
ph
is the
photocurrent. The J-V characteristic of organic solar cells for both the dark and
illumination conditions is accurately described by this equation and the corresponding
plot is portrayed in Figure 1.4.

Figure 1.4: J-V characteristics for both dark and light conditions (red). Power as function of
voltage is also depicted (blue). Power = J x V
The J-V response under illumination is characterized by a raise of the current under
reverse bias as a result of absorption of photons, the photocurrent. The direction of the
photogenerated current is opposed to that of the diode current for higher voltages. The
8

0
Dark
Light
V
oc
J
m
Current Density (mA/cm
2
)
Voltage (V)
J
sc
V
m
P
max
= J
m
V
m
Power Density (mW/cm
2
)
9

relevant solar cell parameters that are extracted from this cell and describe the
performance of the device are J
SC
, V
oc
, P, FF and η% for the illumination conditions and
will be described next.
Illumination parameters
• J
SC
is defined as the photocurrent at short circuit conditions and is a measure of the
amount of generated carriers upon absorption of illumination when no voltage is applied.
It is highly dependent on each step of the photovoltaic mechanism as will be explained
later (Figure 1.6), and every effort in optimizing the efficiency of each step will result in
an enhanced short circuit current.
• V
oc
s the open circuit voltage and is the required applied potential that generates a
current with the same opposed value as the photocurrent such that it shuts down the
current flow. Those factors that determine the value of the open circuit voltage are still
not highly understood and its dependence will be evaluated later on this thesis work.
• P is the device power. It is obtained from the product of the current and the voltage
and is a function of the applied potential. P
o
is the incident power that depends on the
lamp irradiation output and is measured independently with a photodetector by
calibration with standard procedures. Standardization procedures for incident power
measurements will be described in the general experimental section in Chapter 6.
• FF or Fill Factor is a measure of the goodness of the cell, i.e. how close to a square
is the cell response. It results from the calculation of the maximum possible power
achieved over the product of the short circuit current and the open circuit voltage
(FF=Pmax/J
SC
V
oc
) and is always < 1. Typical well behaved solar cells result in elevated
10

FF ~ 0.5 - 0.6. Lower values of FF are indicative of high material resistivities and are
usually associated with poor photocurrents and elevated V
oc
, i.e. poor performing cells.
Resistivities arise as a consequence of poor electron or hole conductance determined by
the quality of the packing of the film that ensures good π overlap intermolecularly.
Molecules with extended π systems contribute to charge mobilities positively together
with those the favors film morphology. The deposition rate for the formation of the
amorphous film has been found to have a relevant effect on the quality of the film that
affects the conductivities and therefore enhances or diminishes the FF. Also, the layer
thicknesses affect the FF such that only thin films can be used without an evident
detrimental effect on the resistivity.
• η% is the power conversion efficiency. It is a measure of the quality of the cell and
it provides evidence of how much power will the cell generate per incident photon. It is a
function of the illumination incident power and must be defined for a specific light
Wattage. It is defined as P
max
/P
o
or as (J
SC
x V
oc
x FF)/ P
o
.
Observation of the dark J-V characteristics also offers insight of the quality of the
cell and must be taken into consideration. Fitting of this response by Equation 1.1 with
J
ph
= 0 yields the parameters n, Js, R
s
and R
p
.
Dark parameters
• n is the diode idealiy factor and describes the processes that dominate the dark
current. Typically, for small molecule organic solar cells at room temperature, it is found
that n ~ 2, characteristic of recombination dominated dark currents. Under steady state
conditions, carrier recombination rate (R) equals the generation (G) rate. For dark
11

conditions then, G is the limiting step and can be described by thermal generation of
carriers. Thermal generation of carriers is an activated process and is dependent
exponentially of the activation energy (Ea) for charge generation times a preexponential
factor that describes the intermolecular coupling or delocalization of molecular charges
that favor carrier generation (Jso) (Js = Jso exp(-Ea/nkT)). When n ~ 1, diffusion
dominated dark currents are present and these are often observed for inorganic solar cells.
Diffusion dominated dark currents apply for those materials with very delocalized states
where thermal activation of generation of carriers is not elevated and rather the
movement of minority carriers through the material is the limiting factor for the current.
Dark currents are in this case, directly dependent on the diffusion coefficient of each
carrier (D) and inversely proportional to the carrier lifetime (τ). A higher n ~ 4 applies
for tunneling dominated dark currents and it is sometimes observed for poorly
constructed organic solar cells.
• Js is the dark current under saturation conditions. Organic photovoltaics are well
known for their low dark currents because organic materials are not very effective carrier
conductors with very localized states such that electron and hole generation and flow are
very limited. A thorough analysis of the dark current will be described in Chapter 3 and
the open circuit dependence on the Js will be derived. The possibility of organic
photovoltaics to present such low dark currents is the reason why it is very common to
encounter OPVs with elevated photovoltage as opposed to inorganic cells where the V
oc

is mostly fixed at 0.5 V. A fairly low J
SC
is then compensated by an elevated V
oc

enhancing the efficiency.
12

• R
p
is the parallel resistance that accounts for leakage current through direct contacts
that can arise from defects or other material imperfections. For most cases, leakage
current is minimal by effects of a very elevated parallel resistance (R
p
→ ∞) and the R
p

dependent term can be disregarded in Equation 1.1.
• R
s
is a series resistance that accounts for organic/organic, electrode/organic contact
resistances and bulk material resistivities. As mentioned above, it determines the quality
of the FF and depends on the film morphology and the intermolecular π overlap.
1.2.2.2 Spectral responsivity
Another highly relevant study of the performance of the solar cells is the
measurement of the external quantum efficiency (EQE). EQE refers to the percentage of
incident photons that are converted to electric current (i.e., collected carriers) when the
cell is operated under short circuit conditions. This is a wavelength dependent
measurement and it reflects the ability of the cell to absorb and generate current for
different parts of the solar spectrum. Inorganic materials present band structures such
that the EQE results in a flat response for energies higher than the band gap and cuts off
at energies close to the band gap dropping to zero for lower energies or higher
wavelengths (see Figure 1.5). Organic materials, on the other hand, present an absorption
spectrum characterized by Gaussian-like peaks for each of the corresponding different
transitions. Each absorption peak corresponds to different excited states. The main
disadvantage of OPVs when compared to the inorganic counterparts is the fact that light
absorption may not overlap entirely with the solar spectrum and sometimes sharp peaks,
13

characteristic of molecules, reduce the range of wavelengths that light can be absorbed.
The use of materials that present a strong intermolecular overlap in the solid state that can
form aggregates improves the spectrum overlap because absorption peaks broaden and
more features may even appear covering a larger range of the solar spectrum. This is
typical of molecules like phthalocyanines that present sharp absorption spectra when in
solution but forms aggregates in the condensed state and the absorption spectrum
broadens increasing solar overlap. In this way, the organic materials always present the
opportunity of synthetically tuning the desired molecular properties that improve
spectrum overlap for example. Another advantage is that in principle, organic materials
with absorption into the infrared can be applied that would allow collection of a part of
the spectrum that is otherwise hard to obtain for the inorganic counterparts and accounts
for a relevant amount of solar irradiation.
14

400 500 600 700 800 900
CuPc 200A
C60 400A
Absorbance, a.u.
Wavelength, nm
AM1.5G
Intensity, a.u.
(a)
(b)
Figure 1.5: (a) spectral responsivity for a number of inorganic materials
(www.newport.com); (b) absorption spectra of typical films of photoactive molecules Copper
Phthalocyanine (CuPc) in black and Fullerene (C
60
) in red overlapped to a standard AM1.5G solar
spectrum (blue)
1.2.3 Steps of photoelectric conversion in small molecule organic photovoltaics
(sm-OPVs)
The generation of current upon illumination can be described by the combination of
several steps, as described below. When the OPV is prepared by putting together a
Donor and an Acceptor layer, the mechanistic steps are represented in Figure 1.6, with a
schematic picture of the energy levels (HOMO and LUMO) for each material. Even
when both materials are photoactive, only those processes for light absorption by the
donor are shown for simplicity. The same processes are expected to occur for the
Acceptor side.

Figure 1.6: Mechanistic representation of photoconversion in sm-OPV
Each mechanistic step is associated with a characteristic efficiency for the particular
process. The overal efficiency for the complete process above is determined by the
product of the efficiency of each step.
η
total
= η
A
x η
ED
x η
CS
x η
CT
x η
CC
1.2
Light Absorption
Absorption of photons incident on a film sample is described by the optical density
of the given material in the solid state. Photoexitation of organic molecules
photoexcitation is usually thought as a simple electron promotion from the HOMO level
15

16

into the LUMO. However, for most cases, it is well know that this is an
oversimplification of the process. Upon absorption of light, molecules in the ground state
are promoted into a higher energy excited state. Visible light excitations are typically
characterized by transitions into the first singlet or triplet state (S
o
→S
1
or S
o
→T
1
). These
transitions account for the absorption peaks and differ in intensity depending on the
transition rules. Molecules with extensive π systems are more likely to present intense
absorptivities in the visible region. It was mentioned before that for molecules with
strong interactions in the solid state, the absorption spectra in the solid film differs from
that in solution and gives broader, and flatter peaks. This is a desired feature for OPVs
since the solar spectrum overlap is enhanced. Optical densities in the order of 10
4
-10
5

cm
-1
are typical of these OPV materials.
Exciton diffusion
Upon light absorption, an excited state is generated in the organic film. This excited
state is called an exciton and for organic materials it is highly localized within the
molecule. For organic crystals and amorphous solids, this excited state is called a
Frenkel exciton. In order to generate carriers, the excited state has to move to the
molecules located at the D/A interface where the excess energy creates a hole and an
electron. The excitons split at the D/A interface to generate charged carriers. The
efficiency of this step is determined by the exciton diffusion length (L
d
). L
d
defines how
far do the excited state moves through a film before it recombines and it limits the
thickness of the photoactive layer that offers the best yield of photogenerated carriers. Ld
is defined as:
τ.D L
d
=
1.3
Where τ is the excited state lifetime and D is the diffusion coefficient.
The excited state lifetime varies if we have a singlet or a triple excited state. Singlets
are usually short lived with lifetimes in the order of the μsecs-picosecs and triplets are
longer lived, in the order of the sec-μsecs.
Excitons diffuse to the interface via two different energy transfer mechanism that
also depends on the nature of the excited state, as depicted in Figure 1.7. For singlet
excitons the typical mechanism for energy transfer is called Förster transfer. This is a
long range mechanism with transfer distances of ~100Å. Förster transfer results from
Coulombic interaction, i.e resonant dipole-dipole coupling and is described by the
following equation:
J
R
k
k k
D D
o
D
ET
6
2
2 1
1
κ
=
Rate for Förster transfer 1.4
where k is a constant, κ is an orientation factor, effect from the orientation of the dipoles;
k
D1
o
the radiative rate of the donor molecule; R
D1D2
is the separation between donor
molecule and the acceptor molecule and J the spectral overlap integral between the
absorption spectrum of the acceptor and the fluorescence spectrum of the donor. The rate
transfer is then enhanced when we have large overlap integral, the donor fluoresces
where the acceptor has absorption; large radiative rate constant of D
1
; large extinction
coefficient of D
2
and small spatial separation.
17

Triplet excited states on the other hand move via Dexter mechanism. Energy
transfer occurrs as a result of an electron exchange mechanism. It requires an overlap of
the wavefunctions of the donor energy and the acceptor energy. The transfer rate constant
is given by:
k
ET
∝ [h/(2π)]P
2
J exp [–2R
D1D2
/L] Rate for Dexter transfer 1.5
where L and P are constants not easily related to experimentally determinable quantities,
and J is the spectral overlap integral. For this mechanism, the spin conservation rules are
obeyed. It is a short range mechanism and highly depends on near neighbor overlap.
Förster
transfer
Dexter
transfer

Figure 1.7: Förster and Dexter transfer mechanisms.
Förster transfer is much more efficient than Dexter transfer and therefore singlets
have higher diffusion coefficients (D). The question of which excited state will present a
longer exciton diffusion length remains under study. While singlets are expected to have
larger D, their lifetime is much shorter. And the opposite is true for triplets. Typical
18

19

exciton diffusion lengths for organic photvoltaic materials are in the order of the 50-500
Å. More about this subject will be considered in Chapter 2.
Charge separation
Generation of a hole and an electron ocurrs at the D/A interface. Once the excited
state (exciton) reaches this interface, charge transfer occurs such that an electron is
injected into the LUMO of the acceptor material and a hole is created in the HOMO of
the donor. This process ocurrs thanks to the energy level offset between the donor and
the acceptor. The exciton has an excess of energy such that it is favorable to reduce an
acceptor molecule and oxidize the donor. The same effect is expected whether light is
absorber at the donor level or at the acceptor.
D* + A Æ D
+
+ A
-

D + A* Æ D
+
+ A
-
The efficiency of this process depends on the energy level offsets or how easily
reduced is the acceptor and the donor oxidized such that the photoexcited state has
enough energy to allow the spontaneity of this process.
Charge transport
The introduction of a D/A heterojunction by Tang in 1986
14
contributed to a major
improvement of OPV efficiencies. The heterojunction allows for the creation of opposite
charge carriers in different layers such that holes are allowed to move through the donor
layer and electrons move through the acceptor. The separation of carriers in different
layer prevents high charge recombination and charge transport results more efficient.
Transport of carriers is not a very efficient process in organic materials as compared to
20

inorganic. Conduction is carried out via hopping mechanism and is highly dependent on
film morphology, packing, intermolecular distance and interaction, etc. Molecules with
very delocalized π electron cloud will show improved mobilities similar to those with
good packing. Resistive films prevent carrier transport affecting the FF when J-V
characteristics are measured. For this reason, very thin layers of photoactive material
(100-400 Å) are required for OPVsin order to minimize carrier recombination and
enhance charge transport efficiencies.
Charge collection
Once the photogenerated carriers move through the organic materials towards the
corresponding electrodes, charge is extracted and collected by the external circuit. For
the anode electrode, Indium Tin Oxide (ITO) is usually used as the transparent
conducting electrode, and a low workfunction metal (e.g. Al, Ag) is used as the cathode.
The organic/electrode interface is desirably an ohmic contact such that it does not
introduce an extra resistive component to the device performance. For ohmic contacts,
the barrier to charge collection is very low and the efficiencies for this step should be
close to ~ 100 % if enough care is taken when choosing the right organic and electrode
materials. It has been demonstrated that overall power conversion efficiencies are
improved by the addition of a “buffer” layer between the organic and metal contact.
15

This layer consists of a wide gap organic material to ensure that excitons are not
quenched at the added interface. The origin of the charge collection efficiency
enhancement comes about from the prevention of damage to the acceptor layer. When
the metal is directly evaporated onto the acceptor material, the hot metal diffuses into the
organic layer creating damage sites. The addition of a thin layer of “buffer” material
enhances charge collection without affecting optical properties or any of the other
photvoltaic steps. This effect is demonstrated in Figure 1.8. The choice of the organic
material to be used as buffer layer has a big impact on charge collection and should be
tested experimentally in order to determine how the energy levels align in order to
prevent an extra resistive component that detriments the overall efficiency. Several
different materials are tried as this layer and results are presented in Chapter 4.

Figure 1.8: The introduction of a “buffer” transparent layer between the acceptor layer and
the metal cathode prevents damage and defect sites and enhances charge collection.
A different point of view to the charge collection problem is the choice of electrode
material. The workfunction value is extremelly relevant in order to avoid Schotky
contacts and have an energy barrier to charge collection. The control of the
characteristics of the interface are highly important to avoid elevated barrier heights.
This issue will be discussed in Chapter 4, where the effect of the presence of an
insulating barrier at the organic/metal interface has an evident effect on the measured
photocurrent.
21

22

ITO modifications are also being pursued in order to improve charge collection at the
anode/organic interface. The formation of monolayers of organic molecules on the ITO
surface previous to device fabrication has a positive effect on the quality of the interface
and therefore the charge collection.
1.2.4 Materials for small molecule organic photovoltaic
Typical materials for sm-OPVs include those molecules with relatively small
molecular weight but with extensive amount of π electrons to present strong visible
absorption and relatively efficient charge conduction. Another important requirement is
the thermal stability that allows thermal evaporation in a vacuum chamber. Energy levels
are highly relevant for designing the D/A heterojunction.
A number of materials have been used as donors. The family of phthalocyanines
(Pc) is long known for their photovoltaic properties. Phthalocyanines present intense
absorption in the visible region and good packing properties that ensure high hole
mobilities. Many different metallic Pcs have been successfully tried as donors with high
efficiencies. Lower Pc analog like subphthalocyanine (SubPc) was also employed as
donor. Acenes have also been reported as donor materials with relatively high
performance. Despite their poor absorptivity, molecules like tetracene, rubrene and
pentance have demonstrated high photocurrents and good performances. Dyes from the
family of squaraines are currently under study as donor layers. Excellent performances
are to be reported. The strong absorption into the visible region makes squaraines a great
promise as photoactive materials in OPVs. Porphyrins have long been known as greatly
23

absorbing dyes but due to their low commercial accessibility little has been studied on
their application in OPVs. Chapter 2 is dedicated to the application of a novel porphyrin
molecule to the study of solar cells. General representations of basic structure of
molecules listed here are described in Figure 1.9.
The list of acceptor molecules that have been successfully applied in sm-OPVs is not
so extensive. The number one molecule used as acceptor is the fullerene or C
60
. This
molecule has shown such good performances that it has been hard to replace for another
that perform so well. The characteristic spherical shape provides high surface area that
exposes the vast π cloud of electrons and offers good absorptivities and even more
importantly, good electron mobilities. The other fullerene C
70
has also been successfully
applied to OPVs. Materials with enhanced electronic conductivities are somewhat hard
to find and only a few have been demonstrated to successfully perform as the electron
transport layer. Perylenes have also been used as acceptors because of the large
delocalized π cloud and the planar shape that enhances accessibility to these electrons.
3,4,9,10-perylene-tetracarboxylic-bisbenzimidazole (PTCBI) has been mostly used but
3,4,9,10-perylene-tetracarboxylic-diimide (PTCDI) and perylene-3,4,9,10-
tetracarboxylic-3,4,9,10-dianhydride (PTCDA) has been employed but with lower
performances.
Acenes
Phthaocyanines
NN
B
N
Cl
Porphyrins
Fullerene
Perylenes
N N
O
O
N
N
NH HN
O
O
O
O

Figure 1.9: Typical small molecules for organic photovoltaic applications
24

25

1.2.5 Back to the sun
Light from the sun comprises a characteristic spectrum of electromagnetic radiation
that corresponds to a black body radiation of a temperature T = 5800K. The solar
spectrum has an absorption maxima at λ
max
= 500 nm, corresponding to a photon energy
of 2.48 eV. Upon reaching the earth surface, this radiation is filtered through the several
layers of the atmosphere. Solar radiation is partially absorbed during its passage through
the atmosphere as depicted in Figure 1.10. This absorption is almost entirely caused by
gases of low concentration in the infrared region of the solar spectrum, water vapor, CO
2
,
nitrous oxide, methane, fluorinated hydrocarbons, and dust; and in the UV region by
gases like O
3
and O
2
. Absorption increases with the length of the path through the
atmosphere and therefore with the mass of air through which the radiation passes. From
here, the air mass is defined according to the angle of incidence of the sun to the earth’s
surface normal. AM0 is the spectrum outside the atmosphere, AM1 is that for normal
incidence and AM1.5 corresponds to an angle of incidence of 48° to the surface normal.
The AM1.5 spectrum is regarded as the standard spectrum (see Figure 1.11 (a)). Solar
spectrum also varies with the amount of clouds and the direct or reflected spectrum must
be distinguished. Global AM1.5 is the sum of both direct reflected and diffuse sunlight
(Figure 1.11 (b)).

Figure 1.10: Black body radiation and solar spectrum (www.newport.com).
26

(b)
(a)
Figure 1.11: Different air masses according to the sunlight angle of incidence (a) and
spectrum variation depending on sky cloud coverage (b) (www.newport.com).
1.3 Brief overlook of thesis work
Basic research in the field of organic photovoltaics is fundamental for the
development of devices with increased efficiency in the future. The understanding of
those processes that dominate the photoconversion mechanism will only be achieved by
27

the introduction of new materials and architectures that attend to the different
problematics of the solar cell losses. The key areas of research that need strong focus of
active research and need resolved are: enhancement of spectrum overlap, enlargement of
the exciton diffusion length, reduction of open circuit voltage losses, enhancement of
charge collection at the electrodes and increased carrier mobilities. A number of these
questions are the subject of study of this thesis. First, a novel porphyrin compound is
applied as donor layer that raises the question of the dependence of the exciton diffusion
length on the nature of the excited state. From this work, it was found that the open
circuit voltage results in an unexpected value that does not correlate with the comparison
of the ΔE
DA
value. This outcome led to the study of the origin of the V
oc
by the
exploration of several different devices with a variety of donors. In chapter 3, the
presentation of a simple model that accounts for the losses of the photovoltage relative to
the energetic ΔE
DA
value is evaluated. Chapter 4 accounts for the problems of charge
collection at the organic/metal interface through the use of different metals and
compositions and by modification of the nature of the organic “buffer” layer. For chapter
5, a photochemical study of a newly synthesized cyclometallated Pt complex is presented.
(2-phenyl-5-nitropyridine)Pt(acetylacetonate) (NPt) was originally prepared for its use as
a “buffer” layer in sm-OPVs. Devices were successfully prepared with this compound as
buffer material. However, its photophysical properties turned out to be very exciting
such that it deserved an important amount of work and therefore its own chapter. Finally,
an experimental Chapter is also presented that summarizes the experimental techniques
relevant for solar cell production and testing.
28

1.4 Chapter 1 References
1. Ginley, D., M.A. Green, and R. Collins, Harnessing Materials for Energy, in MRS
Bulletin. 2008. p. 355-372.
2. Ma, W., et al., Thermally Stable, Efficient Polymer Solar Cells with Nanoscale
Control of the Interpenetrating Network Morphology. Advanced Functional Materials,
2005. 15(10): p. 1617-1622.
3. Reyes-Reyes, M., K. Kim, and D.L. Carroll, High-efficiency photovoltaic devices
based on annealed poly(3-hexylthiophene) and 1-(3-methoxycarbonyl)-propyl-1- phenyl-
(6,6)C
61
blends. Applied Physics Letters, 2005. 87(8): p. 083506.
4. Peet, J., et al., Efficiency enhancement in low-bandgap polymer solar cells by
processing with alkane dithiols. Nat Mater, 2007. 6(7): p. 497-500.
5. Padinger, F., R.S. Rittberger, and N.S. Sariciftci, Effects of Postproduction
Treatment on Plastic Solar Cells. Advanced Functional Materials, 2003. 13(1): p. 85-88.
6. Li, G., et al., High-efficiency solution processable polymer photovoltaic cells by self-
organization of polymer blends. Nature Materials, 2005. 4(11): p. 864-868.
7. Xue, J.G., et al., 4.2% efficient organic photovoltaic cells with low series resistances.
Applied Physics Letters, 2004. 84(16): p. 3013-3015.
8. Xue, J.G., et al., A hybrid planar-mixed molecular heterojunction photovoltaic cell.
Advanced Materials, 2005. 17(1): p. 66
9. Peumans, P. and S.R. Forrest, Very-high-efficiency double-heterostructure copper
phthalocyanine/C
60
photovoltaic cells. Applied Physics Letters, 2001. 79(1): p. 126-128.
10. Yoo, S., B. Domercq, and B. Kippelen, Efficient thin-film organic solar cells based
on pentacene/C
60
heterojunctions. Applied Physics Letters, 2004. 85(22): p. 5427-5429.
11. Chu, C.W., et al., Efficient photovoltaic energy conversion in tetracene-C
60
based
heterojunctions. Applied Physics Letters, 2005. 86(24): p. 243506.
12. Schulze, K., et al., Efficient Vacuum-Deposited Organic Solar Cells Based on a New
Low-Bandgap Oligothiophene and Fullerene C
60
. Advanced Materials, 2006. 18(21): p.
2872-2875.
13. Janssen, A.G.F., et al., Highly efficient organic tandem solar cells using an improved
connecting architecture. Applied Physics Letters, 2007. 91(7): p. 073519.
29

14. Tang, C.W., 2-Layer Organic Photovoltaic Cell. Applied Physics Letters, 1986.
48(2): p. 183-185.
15. Peumans, P., A. Yakimov, and S.R. Forrest, Small molecular weight organic thin-
film photodetectors and solar cells. Journal of Applied Physics, 2003. 93(7): p. 3693-
3723.

30

Chapter 2. Organic photovoltaics using metallo-tetraphenyl
porphyrins complexes as donor layers
2.1 Introduction
Given the efficiency requirements for marketing of organic photovoltaics, a long
road towards improvement of the current performance needs to be pursued.
Enhancement of the photoconversion efficiency for organic photovoltaic solar cells will
be possible by a full basic knowledge of the mechanisms that dominate losses and hinder
efficiencies. The introduction and study of new organic molecules as photoactive layers
will provide new knowledge about the fundamental steps involved in solar-to-electric
conversion.
1-7
From the ample understanding of the photoconversion processes in
organic solar cells, improvement of organic photovoltaic efficiencies and stability will
result as a consequence of controlled synthesis of novel organic molecules that tune the
desired molecular features. To increase light conversion efficiency, it is important to
identify new materials with improved absorption overlap with the solar spectrum while
maintaining both high carrier mobility and efficient exciton transport to the
heterojunction. The high versatility and huge possibilities of creating new molecules
constitutes a great promise for the field to advance and become a commercial alternative
to energy sources.
Porphyrins are a very well known family of dyes with high absorptivity on the
visible region that due to the extended π system, it offers potential applications in organic
electronics. However, due to its limited commercial availability, not much is known as
31

their applicability in OPVs. Platinum octaethylporhyrin (PtOEP) has been used as donor
material coupled to C
60
as the acceptor layer with relatively good performance and
enhanced V
oc
, however further annealing after device preparation was required in order to
reduce series resistance and improve the FF and efficiency.
7

Platinum tetraphenylbenzoporphyrin (PtTPBP) was recently found to be an efficient
near infrared electrophosphorescent material for use in organic light emitting devices
(OLEDs).
8
More relevant to OPVs is the extended electron conjugation of this molecule,
relative to other Pt porphyrins, which results in higher optical extinction coefficients and
a marked red shift of the absorption in the visible part of the solar spectrum. An
enhanced absorption of light is required in photovoltaics in order to improve carrier
generation upon exposure to sun light. On the other hand, PtTPBP presents a saddle,
non-planar shape which may affect the film morphology and consequently its electronic
transport properties may be impacted in a different manner than planar molecules. In this
Chapter, the use of PtTPBP and the analogous structure with a Pd core as effective donor
materials in organic photovoltaics is demonstrated (see structures in Figure 2.1). A
thorough study on its thickness dependence performance will be observed and, as a
consequence, the exciton diffusion length characteristic of the porphyrins films will be
discussed. Both Pt and Pd tetraphenylbenzoporphyrins are highly soluble in common
organic solvents like toluene, dichloromethane, dichloroethane, etc, because of its out-of-
plane shape and without the need of solubilizing functional groups that could result in an
effective site for charge trapping. This property allows the use of solution processing
techniques for device fabrication, desirable to reduce production costs. It is the first time
where solar cells with solution processed donors in a double heterojunction fashion has
been successfully applied demonstrating the opportunity of using these porphyrins in
solar cell applications in a more cost effective manner. A porphyrin analog, Pt
tetraphenylnaphthol porphyrin with a more extended conjugation of the π system was
also synthesized and analyzed when used as the donor layer in a solar cell.

Figure 2.1: PtTPBP, PdTPBP and PtTPNP molecular structures
2.2 Experimental
Both Pt and Pd tetraphenylbenzoporphyrin were synthesized according to literature
8

by Dr. Carsten Borek, and purified by vacuum thermal gradient sublimation. The organic
32

33

materials, copper phthalocyanine (CuPc) 99% (Aldrich), fullerene (C
60
) 99.5% (MTR
Limited), and 2,9-dimethyl-4,7-diphenyl-1,10-phenanthroline (BCP) 96% (Aldrich) were
also purified by sublimation with one (C
60
) or two (CuPc and BCP) cycles prior to use.
Metal cathode materials, Al (99.999%) and Ag (99.9999%) (Alfa Aesar) were used as
received. Materials were sequentially grown by vacuum thermal evaporation at the
following rates: metal-TPBP (1 Å/sec) or CuPc (2 Å/sec), C
60
(2 Å/sec), and BCP (2
Å/sec) and metals: 1000 Å thick Al (2 Å/sec) or Ag (4 Å/sec). Lower deposition rates for
the porphyrins were required in order to successfully achieve working devices. If
deposition rates higher than 1 Å/sec were to be used, a flat J-V response, similar to an
open circuit device would be obtained. This is very likely a result of an increased
resistivity for higher deposition rates. Films prepared at evaporation rates > 1 Å/sec
likely yields highly disorganized morphologies decreasing carrier mobilities. On the
other hand, lower deposition rates give molecules enough time to reorganize into a more
stable, more organized, and therefore less resistive films. This is however, not observed
for planar molecules like CuPc. Higher deposition rates can be used without a
detrimental effect on the film morphology since resistivities are reduced given their
molecular shape.
General device preparation and testing are described in Chapter 6. The data given in
the text and tables are those for an average device and the error bars represent the
variation in device performance among all of the devices tested on that substrate
(typically 8-10 devices). Variability in device performance is greater when comparing
devices on different substrates. We have seen the following ranges when we compared
34

devices on five different substrates, with nominally the same structure
(ITO/PtTPBP(150Å)/C
60
(400Å)/BCP(100Å)/Al, prepared by vacuum thermal
evaporation): ΔJ
SC
= 0.5 mA/cm
2
; ΔV
oc
= 0.02 V; ΔFF = 0.05 and Δη = 0.5%.
Absorption spectra were measured in a dichloromethane solution and for 300 Å thick
films on quartz substrates using an Agilent ultraviolet/visible spectrometer. The exciton
diffusion length was measured by spectrally-resolved photoluminescence at the
University of Michigan by Richard Lundt.
9, 10
Experiments were performed for layers of
PtTPBP of thicknesses ~500 nm, with C
60
as the quenching layer and BCP as a buffer
layer
For the solution processed devices, thin films of the donor layer were spin-coated
from 1 mL of a toluene solution of PtTPBP (5 mM), at 1500 rpm for 40 sec to obtain a
150 Å thick layer onto previously cleaned ITO. The thickness was measured by
ellipsometry and UV-vis absorption spectroscopy. The solution-deposited thin films
were heated (90
o
C) under vacuum for 10-20 min to remove any residual solvent. Once
dried, the films were placed in the high vacuum chamber for deposition of C
60
, BCP and
Al layers following the procedures described above.
2.3 Results and Discussion
2.3.1 Optical properties of metal-TPBP films
As was demonstrated in previous reports,
8, 11
the solution absorption spectra present
an intense and sharp Soret band at wavelengths of λ = 430 nm and λ = 444 nm for
PtTPBP and PdTPBP, respectively. The heavy central atoms and the ring structure affect
35

the Q-band, resulting in a narrow and intense peak at λ = 613 nm (FWHM = 18 nm) for
PtTPBP and λ = 629 nm for PdTPBP (FWHM = 22 nm). Film spectra of these
compounds differ by only a minor red shift and slight broadening of the Soret and Q
bands of both complexes compared to their solution spectra, as shown in Figure 2.2. This
is in contrast to reports for metal phthalocyanines
12, 13
where extensive excitonic coupling
between neighboring molecules due to aggregation can significantly affect and modify
the film spectra. For porphyrins, however, the absorption spectral dependence on film
morphology is typically not as significant as in the phthalocyanines.
14-18
Normally, a
small blue shift and broadening of the Soret band is expected in planar porphyrins such as
PtOEP due to dipole-dipole coupling between adjacent molecules
16, 18
. However, the
addition of phenyl rings to the flat molecule is enough to hinder the intermolecular
interaction due to steric effects and diminish the excitonic coupling resulting in a red shift
of the Soret band. This effect was described by Gouterman et. al. in a study of the
photophysics of monomer and dimer porphyrins.
15
Both PtTPBP and PdTPBP present an
out-of-plane saddle-shape arising from the repulsion of the bulky phenyl rings in the
meso positions and the benzannulated pyrrole that reduce intermolecular interactions.
8
A
reduced intermolecular interaction could be related to a decreased excitonic coupling as
reflected by the subtle differences in the absorption spectra of both solution and film
samples.
300 400 500 600 700 800
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Optical density (α x10
-5
), cm
-1
Wavelength, nm
PtTPBP
PdTPBP
Absorption coefficient (ε x 10
-5
), M
-1
cm
-1

Figure 2.2: Absorption spectra of PtTPBP (—) and PdTPBP (-•-) in solution (black) and as
solid film (red)
2.3.2 Electrochemistry and energy levels of PtTPBP and PdTPBP
HOMO and LUMO energies are of relevant importance in the determination of the
role of the photoactive material. A cascade effect that allows excitons to split at the D/A
interface is necessary in order for electrons to be carried through the acceptor material
and holes through the donor. Several techniques are available that measure HOMO and
LUMO energies: UPS and IPES can directly determine HOMO and LUMO values
respectively; UHV-STM has also been demonstrated to yield both values and
electrochemistry can be used to indirectly tie the oxidation and reduction potentials to the
HOMO and LUMO energies according to published correlations.
19, 20
In such fashion,
voltammetric scans were performed for both Pt and Pd compounds.
36

Cyclic voltammetry for the PtTPBP was performed in dry dichloroethane as solvent
and tetrabutyl ammonium hexafluorophosphate as the conducting electrolyte (Figure 2.3).
The redox couple ferrocinium/ferrocene (Fc
+
/Fc) was used as internal standard reference.
The oxidation wave yields two reversible peaks, (E
ox1
1/2
= 0.19 v and E
ox2
1/2
= 0.82 v vs.
Fc
+
/Fc) indicative of a two electron oxidation. Reduction is observed with a
quasireversible peak at E
red
1/2
= -1.76 v vs. Fc
+
/Fc.
-2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5
-1.5x10
-4
-1.0x10
-4
-5.0x10
-5
0.0
5.0x10
-5
Current, A
Voltage, V (vs. Fc
+
/Fc)
Fc
+
/Fc
E
1/2
ox1
= 0.19 v
E
1/2
ox2
= 0.82 v
E
1/2
red
= -1.76 v
PtTPBP

Figure 2.3: Cyclic voltammetry for PtTPBP in DCE. Voltage scale has been corrected
versus the reference potential of the redox couple Ferrocinium/Ferrocene as internal standard.
37

-2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5
-4.0x10
-4
-2.0x10
-4
0.0
2.0x10
-4
E
1/2
red
= -1.85v
Current, A
Voltage, V (vs Fc
+
/Fc)
E
1/2
ox
= 0.16v
Fc
+
/Fc
PdTPBP

Figure 2.4: Cyclic voltammetry for PdTPBP in DCE. Voltage scale has been corrected
versus the reference potential of the redox couple Ferrocinium/Ferrocene as internal standard.
PdTPBP cyclic voltammetry was performed in dry dichloroethane with tetrabutyl
ammonium hexafluorophosphate as the conducting electrolyte and Fc
+
/Fc as the
reference redox couple (Figure 2.4). For the PdTPBP, there is one visible reversible
oxidation potential at E
ox1
1/2
= 0.16 v vs Fc
+
/Fc and a quasireversible reduction peak at
E
red
1/2
= -1.85 v vs Fc
+
/Fc.
HOMO energies are further calculated using the following correlation:
19

HOMO eV 1.4
4.6 2.1
Whereas the LUMO was calculated according to:
20

38

LUMO eV 1.19

4.8 2.2
Energy levels for typical photoactive molecules used in solar cells are depicted in
Figure 2.5. From this, it is possible to predict that both porphyrins will function as donor
materials when C
60
is used as the acceptor material in similar fashion to the well known
donor CuPc. Energetics for molecules other than Pt and Pd tetraphenylbenzoporphyrins
are shown here too from values obtained from UPS and IPES.
19, 20

CuPc
C
60
BCP
PtTPBP
3.3 eV
2.8 eV
3.6 eV
1.6 eV
5.2 eV
4.9 eV
6.2 eV
6.5 eV
PdTPBP
2.7 eV
4.8 eV

Figure 2.5: HOMO and LUMO energy levels for all photoactive compounds in typical solar
cells. Energy values were obtained from literature.
19, 20

2.3.3 Use of PtTPBP and PdTPBP as donor layers in OPVs
Both PtTPBP and PdTPBP were applied as donor layers in solar devices coupled
with C
60
as the acceptor material. Figure 2.6(a) shows the current density vs. voltage (J-
V) characteristics for the cell structure: PtTPBP (150 Å)/C
60
(400 Å)/BCP (100 Å)/Al
(1000 Å), giving J
SC
= 4.48 ± 0.05 mA/cm
2
, V
oc
= 0.69 ± 0.01 V, and FF = 0.63 ± 0.03 at
1 sun, AM1.5G solar spectrally corrected power intensity. This corresponds to a power
39

40

conversion efficiency of η = 1.9 ± 0.1%, which is higher than a control CuPc/C
60
/BCP/Al
cell with η = 1.6 ± 0.1% (see Table 2.1). The difference in efficiency between both
devices is primarily due to the higher V
oc
of the cells that use Pt porphyrin as the donor
layer. The result for the experimental V
oc
is not in agreement with a reduced
HOMO(donor) – LUMO(acceptor) (ΔE
DA
) when compared to a CuPc/C
60
couple
(ΔE
DA(CuPc)
= 1.7 eV; ΔE
DA(PtTPBP)
= 1.4 eV). It has been long proposed and demonstrated
experimentally that the V
oc
depends directly on the ΔE
DA
.
2, 21-27
In this case however, the
relation does not hold, the ΔE
DA
of the CuPc/C
60
is larger than for PtTPBP/C
60
but the V
oc

is smaller. A thorough explanation of the origin of such a disparity will be the topic of
discussion of Chapter 3.

Figure 2.6: (a) Current density vs. voltage characteristics of
ITO/CuPc(400Å)/C
60
(400Å)/BCP(100Å)/Al(1000Å) (dotted line) and
ITO/PtTPBP(150Å)/C
60
(400Å)/BCP(100Å)/Al(1000Å) (solid line) under dark (thinner lines) and
simulated AM1.5G illumination at 1sun intensity corrected to accurately match the solar
spectrum. (b) Quantum efficiency (dotted line) for the PtTPBP photovoltaic cell shown along
with the absorption spectra of PtTPBP and C
60
.
41

A similar photovoltaic response is observed for devices employing PdTPBP, due to
its similar absorption spectrum and energy level positions (see Table 2.1).
Donor
J
SC

(mA/cm
2
,
±0.05)
V
oc

(V, ±0.01)
FF
(±0.03)
ɳ
(%, ±0.1)
CuPc 5.51 0.48 0.60 1.6
PtTPBP 4.48 0.69 0.63 1.9
PdTPBP 4.31 0.65 0.64 1.8
Table 2.1: Comparison of parameters for ITO/CuPc(200Å)/C
60
(400Å)/
BCP(100Å)/Al(1000Å) and ITO/metal-TPBP(150Å)/C
60
(400Å)/BCP(100Å)/Al(1000Å)
(Metal=Pt or Pd) under simulated AM1.5G light at 1sun corrected according to NREL
standardized procedures.
Also, the use of Ag instead of Al as the cathode material results in an increased J
SC

by almost 10%, consistent with reports for CuPc devices with similar cathode
compositions (as described in Table 2.2).
1, 28
The origin of this observed difference will
be addressed in Chapter 4.
Cathode
J
SC

(±0.05mA/cm
2
)
V
oc

(±0.01V)
FF
(±0.03)
ɳ
(±0.1%)
Al 2.33 0.62 0.60 1.7
Ag 2.54 0.62 0.65 2.0
Table 2.2: Performance comparison of ITO/PtTPBP(150Å)/C
60
(400Å)/BCP(100Å)
/Al(1000Å) and ITO/PtTPBP(150Å)/C
60
(400Å)/BCP(100Å)/Ag(1000Å) under simulated
AM1.5G illumination at 50mW/cm
2
.
42

43

Previous reports of devices that employ PtOEP as a donor material suggested that in
the absence of annealing, the series resistance is so important that results in a low FF, and
therefore a low efficiency is observed.
7
In contrast, high fill factors are obtained using
the metal-TPBP donors without the need for post-fabrication thermal treatment.
The J
SC
for metal-TPBP devices is similar to those employing CuPc, which is
remarkable considering the absorption bands for both PtTPBP and PdTPBP films are
narrower than for CuPc, and the Soret band lies at the same wavelength as the principal
C
60
absorption band, resulting in a reduction in overlap with the AM1.5 spectrum. The
external quantum efficiency for the PtTPBP device, along with the absorption spectra, is
shown in Figure 2.4(b). The observation of a high J
SC
suggests that the reduced spectral
coverage by the narrower absorption bands, is compensated by an increased absorbance
at 400 < λ < 700 nm. This is due to the high optical density (α) of metal-TPBP film,
more than double that of the CuPc film for the Q-band, α
CuPc
(λ = 625nm) = 8.06x10
4

cm
-1
; α
PtTPBP
(λ = 625nm) = 1.89x10
5
cm
-1
; and α
PdTPBP
(λ = 635nm) = 1.85x10
5
cm
-1
.
2.3.4 Optimization of the C
60
layer thickness for PtTPBP cells
Various thicknesses of the acceptor C
60
layer were tested in order to optimize the
device performance by maximizing the resulting photocurrent. It is well known that the
maximum optical density depends on the distribution of the various stacked layer
thicknesses such that the optical field is maximized at the D/A interface.
29
Photocurrent
enhancement can be achieved experimentally by thickness variation of both donor and
acceptor layers. The optimization of the donor layer will be described in section 2.3.5.
44

Maximization of the photocurrent is herein obtained experimentally by variation of the
acceptor. Table 2.3 and Figure 2.7 demonstrate that the highest efficiency is achieved for
a C
60
thickness of 400Å. While the V
oc
remains constant for the various devices, the
photocurrent is highly affected. The J
SC
gradually increases as the C
60
thickness grows to
later decrease for thicknesses above 400Å. While the Rsa is not really affected by an
increase of the C
60
thickness, this behavior is a result of excitons that are created further
from the D/A interface and are not able to reach the interface to generate photocarriers.
Only those excitons created at distances within the exciton diffusion length of the
interface will generate charges.

C
60

Thickness
(Å)
J
SC

(±0.05mA/cm
2
)
V
oc

(±0.01v)
FF
(±0.03)
ɳ
(±0.1%)
R
s

(Ω cm
2
)
200 1.61 0.57 0.64 1.16 0.4
300 1.77 0.58 0.63 1.3 0.5
400 2.11 0.58 0.63 1.54 0.7
500 1.87 0.58 0.62 1.32 0.4
Table 2.3: Photovoltaic performance parameters for devices
ITO/PtTPBP(150Å)/C
60
(xÅ)/BCP(100Å)/Al(1000Å) under simulated 0.5 sun light intensity
-0.4 -0.2 0.0 0.2 0.4 0.6
-2.5
-2.0
-1.5
-1.0
-0.5
0.0
0.5
C
60
(200Å)
C
60
(300Å)
C
60
(400Å)
C
60
(500Å)
Current density (mA/cm
2
)
Voltage (V)

Figure 2.7: Current density vs. voltage characteristics of ITO/PtTPBP(150Å)/C
60
( x
Å)/BCP(100Å)/Al(1000Å) under simulated 0.5 sun light intensity
2.3.5 Thickness dependence of metal-TPBP and the exciton diffusion length
limitation
In order to estimate the optimal metal-TPBP layer thickness, OPV cells with the
structure ITO/metal-TPBP (xÅ)/C
60
(400Å)/BCP (100Å)/Al (1000Å) were prepared, with
results shown in Figure 2.8; a similar trend is observed for PdTPBP. The series
resistance (Rs) measured from dark J-V characteristics slightly increases as the PtTPBP
layer thickness is increased (see Table 2.4). However, the FF is not heavily affected with
the thickness indicating that the variations in Rs are not significant enough to modify the
carrier mobility and appreciably affect J
SC
.
From Figure 2.8, it is possible to infer that the maximum photocurrent is obtained at
150 Å for both metal analogs. The photocurrent goes down as the thickness is increased
45

46

beyond 100 Å, even when the absorbance of the metal porphyrin layer is increased, as
shown for the Q-band. This reduced optimum thickness is indicative of a small exciton
diffusion length, shorter than 150 Å. This outcome was unexpected when considering
that the nature of the excited state in solution corresponds to a triplet state. Triplet
excited states are normally long lived because its relaxation into the ground state
corresponds to a spin forbidden transition (T
1
→ S
0
). Since the exciton diffusion length
is proportional to the lifetime (L
d
=√τD), a long exciton lifetime will contribute to an
enhanced L
d
. D is the diffusion coefficient and is described by the efficiency of the
exciton movement through the film. For PtTPBP in solution, τ
triplet
= 53 μs,
8
whereas for
PdTPBP, the triplet lifetime is elongated due to the stronger heavy atom effect (τ
triplet
=
143 μs).
11
However, the nature of the excited state in solution may not be the same as in
the solid. The condensed phase intermolecular interactions may be such that emission is
effectively self-quenched. The phosphorescence for the film samples was in fact very
weak for both metals, with quantum yields ~ < 0.5%, indicating that the triplets are
rapidly quenched in solid state and therefore responsible for a reduced L
d
. A short L
d
and
the observed reduced emission intensity indicates that the triplet nature of the excited
state as described by the fluid solution is not necessarily representative of the excited
state in the solid form
100 200 300 400
-1.4
-1.6
-1.8
-2.0
-2.2
-2.4
J
SC
Abs.
PtTPBP Thickness (Å)
Jsc at 0.5 Sun (mA/cm
2
)
0.2
0.4
0.6
0.8
1.0
PtTPBP Absorbance at 625 nm
100 200 300 400
-1.2
-1.4
-1.6
-1.8
-2.0
-2.2
-2.4
J
SC
Abs.
PdTPBP Thickness (Å)
Jsc at 0.5 Sun (mA/cm
2
)
0.2
0.4
0.6
0.8
PdTPBP Absorbance at 635 nm

Figure 2.8. Current density vs. layer thickness of ITO/metal-TPBP(x=100Å -
400Å)/C
60
(400Å)/ BCP(100Å)/Al(1000Å) devices under simulated AM1.5G illumination at
50mW/cm
2

47

48

In order to support the device evidence of a short exciton diffusion length and to
quantify this value, a direct measurement based on the excited state quenching was
performed as described in section 2.2. A value of L
d
= 57 ± 5 Å for PtTPBP was
obtained, in agreement with the experimental device data of L
d
< 150 Å.
PtTPBP
Thickness (Å)
J
SC

(mA/cm
2
, ±0.05)
V
oc

(V, ±0.01)
FF
(±0.01)
ɳ
(%, ±0.1)
R
s

(Ω cm
2
)
100 2.22 0.59 0.62 1.6 0.3
150 2.26 0.61 0.64 1.8 0.2
200 2.06 0.62 0.64 1.6 0.8
300 1.76 0.63 0.62 1.4 2.6
400 1.56 0.64 0.60 1.2 5.8

PdTPBP
Thickness
(Å)
J
SC

(±0.05mA/cm
2
)
V
oc

(±0.01v)
FF
(±0.03)
ɳ
(±0.1%)
R
s

(Ω cm
2
)
150 2.41 0.61 0.62 1.8 0.3
200 1.86 0.63 0.60 1.4 1.7
300 1.66 0.64 0.60 1.3 10
400 1.36 0.65 0.58 1.0 4.1
Table 2.4. Photovoltaic performance parameters for devices
ITO/metal-TPBP(xÅ)/C
60
(400Å)/BCP(100Å)/Al(1000Å) under simulated AM1.5G illumination
at 50mW/cm
2
.
The origin of a short L
d
for PtTPBP in the condensed phase may be understood in
terms of the nature of the excited state in the solid film. The emission spectrum of a
PtTPBP film shows a broad, featureless band at λ = 978 nm that is significantly red
shifted relative to both the solution emission (see Figure 2.9) and the doped thin film
49

spectra of PtTPBP (solution and doped thin film PL spectra of PtTPBP are nearly
identical).
8
Indeed, the thin film absorption spectrum of PtTPBP shows only a minor
bathochromic shift and broadening relative to the solution spectrum. This red shift and
broadening of the emission spectra suggests that luminescence is due to excimers, or to a
low concentration of dimer/aggregate states in the solid form.
16, 30-34
The excimer energy
in PtTPBP is 0.34 eV, as inferred from the red shift between the solution and thin film
emission spectra. This lower energy can act as an effective exciton trap, which
significantly influences the exciton lifetime, and hence L
d
.
34-37
Equally important may be
trapping due to disorder-related defects arising from the sterically hindered molecular
structure as discussed above. Traps are expected to have reduced lifetime and move
slower through the solid effectively reducing the exciton diffusion length by diminishing
D. As a side experiment, grazing angle x-ray crystallography on the PtTPBP film sample
was performed at “University of Michigan”. No visible crystallinity peak was observed,
suggesting that porphyrin films grow in an amorphous fashion by evaporation, which
supports the observation of poor exciton diffusion through a disorganized film. However,
the film results in a good conductor for holes such that the R
s
is not large, with a high FF,
even for thicknesses of up to 400 Å.
It is highly desirable to enhance device performance by boosting light absorption
(increase of the thickness of photoactive materials). In order to do so it is necessary to
improve the actual values of L
d
for organic materials. An understanding of the nature of
the excited state for the solid samples is the first step towards realizing high L
d
values.
400 600 800 1000 1200
0.0
0.5
1.0
Wavelength (nm)
Normalized absorption, a.u.
ΔE = 0.34eV
0.0
0.5
1.0
Normalized emission intensity, a.u.
Absorption Emission

Figure 2.9. Absorption and emission spectra of PtTPBP in degassed toluene solution (solid
line) and solid film (circles) at room temperature.
2.3.4 Solar cells fabricated by solution process of PtTPBP as donor layer
An appealing attribute of metal-TPBP materials is their high solubility in common
organic solvents. Solution processing is a cost attractive method for production of
photovoltaics and intense research is being developed in the area of polymeric solar cells
for this reason. Solution processed donor layers were fabricated by spin coating various
thicknesses of PtTPBP. These films were later introduced in the high vacuum chamber
for further evaporation of the acceptor (C
60
), buffer (BCP) and metal layers in order to
achieve structures similar to the all-vacuum deposited devices discussed above. Both the
50

51

V
oc
= 0.64 ± 0.01 V and FF = 0.52 ± 0.03 of solution processed devices shown in Figure
2.10 are similar to those prepared by vacuum deposition. However, solution processed
devices give lower J
SC
= 2.47 ± 0.05 mA/cm
2
than the all-vacuum deposited devices
(where J
SC
= 4.48 ± 0.05 mA/cm
2
, see Figure 2.6 and table 2.1). The variability in
performance of solution processed devices is considerably larger than when processed
only in vacuum. For example, devices with a 150 Å thick layer of PtTPBP deposited
from solution have V
oc
ranging from 0.34V to 0.64 V, whereas the analogous devices
with vacuum deposited PtTPBP, the photovoltage varies only between 0.65 V to 0.69 V.
The addition of a 50 Å layer of vapor deposited PtTPBP between the solution processed
PtTPBP film and C
60
leads to devices with a narrow range of V
oc
similar to those
prepared entirely in vacuum. It is likely that improvement of the D/A interface by
addition of a thin layer of evaporated PtTPBP is crucial to recover the same parameters of
the all-vacuum cell.

Figure 2.10. Comparison of devices with solution processed (SP) and vapor deposited (VD)
donor layer. Dark and illuminated current density vs. voltage characteristics corresponding to the
following devices: PtTPBP(SP)(150Å)/C
60
(400Å)/BCP(100Å)/Al (solid line); and
PtTPBP(VD)(150Å)/C
60
(400Å)/BCP(100Å)/Al (filled circles) under simulated 1 sun intensity,
AM1.5G illumination
2.3.5 PtTPNP optical properties and energy levels
As previously reported for the Pd analog in Rogers et. al.,
11
the absorption spectra of
the PtTPNP solution spectra in dichloroethane presents an elevated Soret band at λ = 688
nm (FWHM = 50.7 nm) with an extinction coefficient of ε = 2.28 x 10
5
M
-1
cm
-1
and a
slightly reduced Q-band (FWHM = 22.8 nm) (λ = 435 nm; ε = 1.40 x 10
5
M
-1
cm
-1
). The
naphthol analog is also expected to present a distorted saddle shape as in the case of the
TPBP as a result of the steric bulk from the phenyls in the meso positions that deforms
the molecule in a non-planar fashion in order to accommodate these substituents. The
52

film absorption spectrum is shown in Figure 2.11 and it resembles the absorption
spectrum in solution. Film and solution spectra similarities are very likely a result of the
poor intermolecular interaction of the extended porphyrin in the solid phase as a
consequence of the characteristic out-of-plane geometry (Soret: λ = 443 nm; α = 1.60 x
10
5
cm
-1
; FWHM = 68.6 nm; Q-band: λ = 705 nm; α = 1.9 x 10
5
cm
-1
; FWHM = 48.5
nm)
53

Figure 2.11: Absorption spectra of PtTPNP as a solid film (red line) and in solution
dissolved in dichloroethane (black line).
Electrochemistry was also used to determine the HOMO-LUMO energy levels of
PtTPNP. A cyclic voltammetry scan was performed in dry dichloroethane as solvent and
tetrabutyl ammonium hexafluorophosphate as the conducting electrolite. The first
oxidation wave of PtTPNP was found at almost the same potential as that for the
300 400 500 600 700 800
0.0
0.5
1.0
1.5
2.0
2.5
0.0
0.5
1.0
1.5
2.0
2.5
Wavelength, nm
Extinction coefficient, (ε x 10
-5
) M
-1
cm
-1
Optical density, (α x10
-5
) cm
-1
54

commonly used internal reference Fc
+
/Fc couple, such that decamethylferrocene had to
be used as the internal standard instead. A later scan using decamethylferrocene (DcFc)
and ferrocene was performed in order to determine the potential of the DcFc
+
/DcFc in
reference to Fc
+
/Fc. (see Figures 2.12 (a) and (b)). The anodic scan for the dissolved
PtTPNP yields two reversible oxidation peaks, (E
ox1
1/2
= -0.13 v and E
ox2
1/2
= 0.5 v vs.
Fc
+
/Fc) whereas the reduction wave yields two quasireversible reductions at E
red1
1/2
= -
1.85 v and t E
red2
1/2
= -2.31 v vs Fc
+
/Fc. The extension of the π-system for the naphthol
analog as compared to the benzo platinum porphyrin results in a very reduced oxidation
potential, π electrons in PtTPNP are much more easily removed than for its benzo analog
as expected by the increased number of electrons.
From this experiment we can calculate the corresponding HOMO and LUMO energy
levels from Equations 2.1 and 2.2. (HOMO = 4.42 eV; LUMO = 2.60 eV)
-3 -2 -1 0 1
-3.0x10
-4
-2.0x10
-4
-1.0x10
-4
0.0
1.0x10
-4
2.0x10
-4
3.0x10
-4
E
1/2
red2
= -2.31v

E
1/2
red1
= -1.85v

E
1/2
ox2
= 0.5v

PtTPNP + DcFc
Current, A
Voltage, v (vs. Fc
+
/Fc)
DCE
E
1/2
ox1
= -0.13v

DcFc
(a)

-2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5
-5x10
-4
-4x10
-4
-3x10
-4
-2x10
-4
-1x10
-4
0
1x10
-4
2x10
-4
3x10
-4
4x10
-4
5x10
-4
6x10
-4
7x10
-4
dcfc
dcfc + fc
Current, A
Voltage, v (vs. Fc
+
/Fc)
DCE
E
1/2
DcFc
= -0.57 v (vs Fc/Fc
+
)
(b)

Figure 2.12: (a) Cyclic voltammetry of PtTPNP in DCE with DcFc as the internal reference
and tetrabutyl ammonium hexafluorophosphate as the conducting electrolyte; (b) Cyclic
voltammetry of DcFc with Fc as the internal reference in the same conditions as before.
2.3.6 PtTPNP as donor layer in solar cells and thickness dependence studies
PtTPNP was applied as the donor layer coupled to C
60
as the acceptor in organic
photovoltaics. For a solar cell structure of the type: PtTPNP (150 Å)/C
60
(400 Å)/BCP
55

56

(100 Å)/Al (1000 Å) the illumination performance gives: J
SC
= 4.98 ± 0.05 mA/cm
2
, V
oc

= 0.32 ± 0.01 V, and FF = 0.52 ± 0.03 at 1 sun, AM1.5G solar spectrally corrected power
intensity. This corresponds to a power conversion efficiency of η = 0.83 ± 0.1%, more
than half from that of a cell with PtTPBP as donor (η = 1.9 ± 0.1%) (as described in
section 2.3.3) (see table 2.5 and figure 2.13). The difference in performance arises
principally as a result of the V
oc
value. A lower V
oc
results mainly as a consequence of
the reduced PtTPNP HOMO value that yields a diminished HOMO
D
-LUMO
A
gap
(ΔE
DA
). As mentioned earlier, the effect on the V
oc
will be discussed in Chapter 3 and a
thorough analysis of the origin of this much lower V
oc
will be presented. External
quantum efficiency for the PtTPNP cells is presented in Figure 2.14. The spectral
response follows clearly the absorptivity of the PtTPNP for both Soret and Q bands.
Donor
J
SC

(mA/cm
2
)
V
oc

(V)
FF η%
PtTPBP(150Å) 4.48 0.69 0.63 1.9
PtTPNP(150Å) 4.98 0.32 0.52 0.83
Table 2.5: Comparison of parameters for ITO/donor/C
60
(400Å)/ BCP(100Å)/Al(1000Å)
under simulated AM1.5G light at 1sun corrected according to NREL standardized procedures.

-0.4 -0.2 0.0 0.2 0.4 0.6
-6
-4
-2
0
2
4
6
PtTPBP
PtTPNP
Current density (mA/cm
2
)
Voltage (V)

Figure 2.13: Comparison of current density vs. voltage characteristics of ITO/donor
(150Å)/C
60
(400Å)/BCP(100Å)/Al(1000Å) under simulated 1 sun light intensity
400 500 600 700 800
0
5
10
15
20
25
QE%
Wavelength, nm
0.2
0.4
0.6
0.8
1.0
1.2
PtTPNP 300
Å

C
60
400
Å
Absorbance, a.u.

Figure 2.14: Quantum efficiency (circled line) for the PtTPNP photovoltaic cell shown
along with the absorption spectra of PtTPNP and C
60
films.
57

58

A thickness dependence study was also performed for PtTPNP and data is presented
in figure 2.15 and table 2.6. In a similar fashion to the PtTPBP, the photocurrent raises as
the thickness is increased due to the increase in the amount of absorbed photons. The
photocurrent later declines at thicknesses higher than 150 Å, indicative of an exciton
diffusion length limited photocurrent. For thicknesses larger than the optimum thickness
of 150 Å, the excitons generated near the anode/donor side will inefficiently reach the
D/A interface and recombine before they can separate to generate carriers. This result
suggests that film morphology and the nature of the excited state of the sample in the
solid state are limiting factors in the definition of the exciton diffusion length. Förster
transfer dominates the transport mechanism for the triplet excited states characteristic of
this family of molecules. The very inefficient Förster transfer depends on the
intermolecular overlap between neighboring molecules such that an increase of the
extension of the π system would be expected to yield a more efficient Förster energy
transfer therefore increasing D. Even when an enhanced diffusion coefficient is expected
from a molecule with an extended π-system, little effect is observed in terms of the
resultant L
d
for the film form. This outcome provides even stronger evidence that the
nature of the excited state in the film form is not straightforwardly described by the
solution excited state and that the mechanisms for energy transfer for such a state are not
easily rationalized.
A main difference from the benzo porphyrins performance for diverse layer
thicknesses is the observation of a reduced FF as the layer thickness in increased. A
reduced FF greatly impacts both the J
SC
and the V
oc
such that the overall power
conversion efficiency is diminished. An increase of the value of the series resistance (R
s
)
obtained from the dark J-V characteristics of the PtTPNP layer as the thickness reaches
300 Å is responsible for the reduced FF. As the layer grows in length, carrier mobilities
are greatly affected and the ability of the device to extract the charges before
recombination is reduced giving a low FF. This effect is an indication of a poorer carrier
conductor than that seen for metal-TPBP, high FF of 0.60 were obtained for PtTPBP and
PdTPBP of thicknesses up to 400 Å.
-0.4 -0.2 0.0 0.2 0.4
-6
-5
-4
-3
-2
-1
0
1
2
3
100
Å
150
Å
200
Å
300
Å
Current Density (mA/cm
2
)
Voltage (v)

Figure 2.15: J-V characteristics for devices
ITO/PtTPNP(xÅ)/C
60
(400Å)/BCP(100Å)/Al(1000Å) devices under simulated 1 sun AM1.5G
light intensity.

59

60

PdTPNP
Thickness
(Å)
J
SC

(±0.05mA/cm
2
)
V
oc

(±0.01v)
FF
(±0.03)
ɳ
(±0.1%)
R
s

(Ω cm
2
)
100 4.57 0.26 0.47 0.6 1.8
150 4.98 0.32 0.52 0.8 1.0
200 3.93 0.34 0.47 0.6 1.5
300 2.10 0.19 0.31 0.1 2.7
Table 2.6: Photovoltaic performance parameters for devices
ITO/PtTPNP(xÅ)/C
60
(400Å)/BCP(100Å)/Al(1000Å) under simulated 1 sun AM1.5G
illumination.

2.4 Conclusions
The possibility of fabricating high efficiency, small molecule organic solar cells
using metal-tetraphenyl porphyrins as the donor layer was demonstrated in this Chapter.
These molecules constitute an addition to the currently limited range of donor molecules
that yield highly efficient photovoltaics when paired with the acceptor, C
60
. The
determination of the HOMO and LUMO energies by electrochemistry is highly important
in order to determine the donor ability of the dyes as potential photoactive layers in solar
cells. The V
oc
for metal-TPBP devices results in an unexpected large value that is
responsible for the increased performance compared to a standard CuPc/C
60
cell.
Enlightenment about the nature of the excited state of the amorphous solid form of the
produced films was possible when studying the thickness dependence performance of
solar cells with donor molecules that present diverse excited state nature on the solution.
The exciton diffusion length was not greatly affected by the use of different triplet
61

lifetimes with Pt and Pd TPBP or with diverse π extension with the introduction of the
naphthol porphyrin, demonstrating that the excited state of the films are distinguishable
from that of solution. A limited exciton diffusion length was observed by the making of
solar cell devices with different donor thicknesses and by the observation of the
dependence of J
SC
, and efficiency. A short exciton diffusion length is a result of the
nature of the metal-TPBP solid film excited state that is characterized by a trapped state
with a much reduced exciton diffusion length (L
d
(PtTPBP) = 57 Å), despite its triplet
nature with long lifetimes in solution. Both vacuum and solution processing can be used
for layer deposition with this family of donor materials that provides a non expensive
option for solar cells fabrication. The performance of these cells, smaller than that of the
all-vacuum deposited ones, is still relatively important with no particular losses in FF
demonstrating the possibility of cost effective production techniques.

62

2.5 Chapter 2 References
1. Xue, J.G., S. Uchida, B.P. Rand, and S.R. Forrest, 4.2% efficient organic
photovoltaic cells with low series resistances. Applied Physics Letters, 2004. 84(16): p.
3013-3015.
2. Mutolo, K.L., E.I. Mayo, B.P. Rand, S.R. Forrest, and M.E. Thompson, Enhanced
open-circuit voltage in subphthalocyanine/C-60 organic photovoltaic cells. Journal of the
American Chemical Society, 2006. 128(25): p. 8108-8109.
3. Rand, B.P., J. Xue, F. Yang, and S.R. Forrest, Organic solar cells with sensitivity
extending into the near infrared. Applied Physics Letters, 2005. 87(23): p. 233508.
4. Bailey-Salzman, R.F., B.P. Rand, and S.R. Forrest, Near-infrared sensitive small
molecule organic photovoltaic cells based on chloroaluminum phthalocyanine. Applied
Physics Letters, 2007. 91(1): p. 013508.
5. Yoo, S., B. Domercq, and B. Kippelen, Efficient thin-film organic solar cells based
on pentacene/C
60
heterojunctions. Applied Physics Letters, 2004. 85(22): p. 5427-5429.
6. Chu, C.W., Y. Shao, V. Shrotriya, and Y. Yang, Efficient photovoltaic energy
conversion in tetracene-C
60
based heterojunctions. Applied Physics Letters, 2005.
86(24): p. 243506.
7. Shao, Y. and Y. Yang, Efficient organic heterojunction photovoltaic cells based on
triplet materials. Advanced Materials, 2005. 17(23): p. 2841-2844.
8. Borek, C., K. Hanson, P.I. Djurovich, M.E. Thompson, K. Aznavour, R. Bau, Y.
Sun, S.R. Forrest, J. Brooks, L. Michalski, and J. Brown, Highly Efficient, Near-Infrared
Electrophosphorescence from a Pt-Metalloporphyrin Complex. Angewandte Chemie
International Edition, 2007. 46(7): p. 1109-1112.
9. Lunt, R.R., N.C. Geibink, A.A. Belak, J.B. Benzinger, and S.R. Forrest, Manuscript
in preparation.
10. Vaubel, G. and H. Baessler, Diffusion of Singlet Excitons in Tetracene Crystals.
Molecular Crystals and Liquid Crystals, 1970. 12(1): p. 47-&.
11. Rogers, J.E., K.A. Nguyen, D.C. Hufnagle, D.G. McLean, W.J. Su, K.M. Gossett,
A.R. Burke, S.A. Vinogradov, R. Pachter, and P.A. Fleitz, Observation and
interpretation of annulated porphyrins: Studies on the photophysical properties of meso-
tetraphenylmetalloporphyrins. Journal of Physical Chemistry A, 2003. 107(51): p.
11331-11339.
63

12. Brown, R.J.C., A.R. Kucernak, N.J. Long, and C. Mongay-Batalla, Spectroscopic
and electrochemical studies on platinum and palladium phthalocyanines. New Journal of
Chemistry, 2004. 28(6): p. 676-680.
13. Ferreira, J.A., R. Barral, J.D. Baptista, and M.I.C. Ferreira, Absorption coefficients
and fluorescence quantum yields of porphyrin films determined by optical and
photoacoustic spectroscopies. Journal of Luminescence, 1991. 48-49(1): p. 385-390.
14. Chau, L.K., C.D. England, S.Y. Chen, and N.R. Armstrong, Visible Absorption and
Photocurrent Spectra of Epitaxially Deposited Phthalocyanine Thin-Films -
Interpretation of Exciton Coupling Effects. Journal of Physical Chemistry, 1993. 97(11):
p. 2699-2706.
15. Gouterman, M., D. Holten, and E. Lieberman, Porphyrins XXXV . Exciton coupling
in μ-oxo Scandum dimers. Chemical Physics, 1977. 25(1): p. 139-153.
16. Kalinowski, J., W. Stampor, J. Szmytkowski, M. Cocchi, D. Virgili, V. Fattori, and
P.D. Marco, Photophysics of an electrophosphorescent platinum (II) porphyrin in solid
films. The Journal of Chemical Physics, 2005. 122(15): p. 154710.
17. Stampor, W., Electroabsorption study of vacuum-evaporated films of
Pt(II)octaethylporphyrin. Chemical Physics, 2004. 305(1-3): p. 77-84.
18. Tranthi, T.H., J.F. Lipskier, P. Maillard, M. Momenteau, J.M. Lopezcastillo, and J.P.
Jaygerin, Effect of the Exciton Coupling on the Optical and Photophysical Properties of
Face-to-Face Porphyrin Dimer and Trimer - a Treatment Including the Solvent
Stabilization Effect. Journal of Physical Chemistry, 1992. 96(3): p. 1073-1082.
19. D'Andrade, B.W., S. Datta, S.R. Forrest, P. Djurovich, E. Polikarpov, and M.E.
Thompson, Relationship between the ionization and oxidation potentials of molecular
organic semiconductors. Organic Electronics, 2005. 6(1): p. 11-20.
20. Djurovich, P.I., E.I. Mayo, S.R. Forrest, and M.E. Thompson, Measurement of the
lowest unoccupied molecular orbital energies of molecular organic semiconductors.
Organic Electronics, 2009.
21. Rand, B.P., D.P. Burk, and S. Forrest, R., Offset energies at organic semiconductor
heterojunctions and their influence on the open-circuit voltage of thin-film solar cells.
Physical Review B (Condensed Matter and Materials Physics), 2007. 75(11): p. 115327.
22. Brabec, C.J., A. Cravino, D. Meissner, N.S. Sariciftci, T. Fromherz, M.T. Rispens, L.
Sanchez, and J.C. Hummelen, Origin of the open circuit voltage of plastic solar cells.
Advanced Functional Materials, 2001. 11(5): p. 374-380.
64

23. Gadisa, A., M. Svensson, M.R. Andersson, and O. Inganas, Correlation between
oxidation potential and open-circuit voltage of composite solar cells based on blends of
polythiophenes/ fullerene derivative. Applied Physics Letters, 2004. 84(9): p. 1609-1611.
24. Kooistra, F.B., J. Knol, F. Kastenberg, L.M. Popescu, W.J.H. Verhees, J.M. Kroon,
and J.C. Hummelen, Increasing the Open Circuit Voltage of Bulk-Heterojunction Solar
Cells by Raising the LUMO Level of the Acceptor. Org. Lett., 2007. 9(4): p. 551-554.
25. Scharber, M.C., D. Wuhlbacher, M. Koppe, P. Denk, C. Waldauf, A.J. Heeger, and
C.L. Brabec, Design rules for donors in bulk-heterojunction solar cells - Towards 10 %
energy-conversion efficiency. Advanced Materials, 2006. 18(6): p. 789-+.
26. Vandewal, K., A. Gadisa, W.D. Oosterbaan, S. Bertho, F. Banishoeib, I. Van
Severen, L. Lutsen, T.J. Cleij, D. Vanderzande, and M.J. V., The Relation Between Open-
Circuit Voltage and the Onset of Photocurrent Generation by Charge-Transfer
Absorption in Polymer : Fullerene Bulk Heterojunction Solar Cells. Advanced
Functional Materials, 2008. 18(14): p. 2064-2070.
27. Sarangerel, K., C. Ganzorig, M. Fujihira, M. Sakomura, and K. Ueda, Influence of
the Work Function of Chemically Modified Indium-Tin-Oxide Electrodes on the Open-
circuit Voltage of Heterojunction Photovoltaic Cells. Chemistry Letters, 2008. 37(7): p.
778-779.
28. Peumans, P. and S.R. Forrest, Very-high-efficiency double-heterostructure copper
phthalocyanine/C-60 photovoltaic cells. Applied Physics Letters, 2001. 79(1): p. 126-
128.
29. Peumans, P., A. Yakimov, and S.R. Forrest, Small molecular weight organic thin-
film photodetectors and solar cells. Journal of Applied Physics, 2003. 93(7): p. 3693-
3723.
30. D'Andrade, B. and S.R. Forrest, Formation of triplet excimers and dimers in
amorphous organic thin films and light emitting devices. Chemical Physics, 2003. 286(2-
3): p. 321-335.
31. Dienel, T., H. Proehl, T. Fritz, and K. Leo, Novel near-infrared photoluminescence
from platinum(II)-porphyrin (PtOEP) aggregates. Journal of Luminescence, 2004.
110(4): p. 253-257.
32. Lai, S.W., M.C.W. Chan, T.C. Cheung, S.M. Peng, and C.M. Che, Probing d(8)-d(8)
Interactions in luminescent mono- and binuclear cyclometalated platinum(II) complexes
of 6-phenyl-2,2 '-bipyridines. Inorganic Chemistry, 1999. 38(18): p. 4046-4055.
65

33. Ma, B.W., P.I. Djurovich, S. Garon, B. Alleyne, and M.E. Thompson, Platinum
binuclear complexes as phosphorescent dopants for monochromatic and white organic
light-emitting diodes. Advanced Functional Materials, 2006. 16(18): p. 2438-2446.
34. Ma, B., P.I. Djurovich, and M.E. Thompson, Excimer and electron transfer
quenching studies of a cyclometalated platinum complex. Coordination Chemistry
Reviews, 2005. 249(13-14): p. 1501-1510.
35. Fleeman, W.L. and W.B. Connick, Self-quenching and Cross-quenching Reactions
of Excited Platinum(II) Diimine Complexes. Comments on Inorganic Chemistry, 2002.
23(3): p. 205 - 230.
36. Turro, N.J., Modern Molecular Photochemistry. 1991: University Science Books.
37. Wan, K.T., C.M. Che, and K.C. Cho, Inorganic Excimer - Spectroscopy, Photoredox
Properties and Excimeric Emission of Dicyano(4,4'-Di-Tert-Butyl-2,2'-
Bipyridine)Platinum(Ii). Journal of the Chemical Society-Dalton Transactions, 1991(4):
p. 1077-1080.

66

Chapter 3. On the molecular nature of the open circuit voltage
in organic photovoltaics – A comprehensive study
3.1 Introduction
As mentioned earlier, organic solar cells represent a non-expensive, environmentally
friendly alternative to solar energy conversion. However, in order for OPVs to make a
meaningful contribution to meeting our overall energy needs, the device efficiencies need
to be significantly improved from those current values that lie in the range of 5-6 %.
1, 2

This is compared to inorganic solar cell efficiencies of > 20%. For enhanced
performance and development of higher efficiency OPVs, a basic understanding of the
processes that control and limit the operation of solar cells is crucial and can serve as a
guide for exploration of new materials for high performance devices. A key limitation in
OPVs to date is their low open circuit voltages (V
oc
), which is typically less than half of
the incident photon energy.
From those parameters that define the efficiency of solar cells, the open circuit
voltage is the solar cell characteristic that is less understood. Several reports still argue
about the origin of this value for organic PVs. Little is contended however, on the
reasons that determine the photocurrent which include the light intensity, optical density
of the cell, the degree of overlap between the donor and acceptor layer absorption spectra
and the emission spectrum of the sun, the exciton diffusion lengths and the charge
separation efficiency at the D/A interface. The FF is related to the cell series resistance
(R
s
), with reasonably good OPVs typically giving FF > 0.5. Several models have been
67

proposed that treat the dependence of the photovoltage on a number of D/A material-
independent factors, including electrode choice, light intensity and temperature.
3-5

Cathode materials with different workfunctions result in varied V
oc
originated from
diverse built-in potential generated across the organic layers as a result of the different
energetics between anode and cathode. Light intensity shows a clear logarithmic effect
on the V
oc
. Also, in polymeric photovoltaics, it has been demonstrated that the
morphology and phase segregation of the photoactive layers has an important effect on
the value of the open circuit value by modifications of the interface and carrier
mobilities.
6,7

Additionally, different donor and acceptor materials combinations influence V
oc
.
Specifically, several studies have demonstrated a dependence of V
oc
on the energy
difference between the highest occupied molecular orbital (HOMO) energy of the donor
and the lowest unoccupied molecular orbital (LUMO) energy of the acceptor; the
ΔE
DA
.
5,8-14
However, under normal operating conditions (i.e. room temperature and 1 sun
intensity illumination) the experimental values for V
oc
can differ from the values
predicted by ΔE
DA
for some materials systems. This V
oc
to ΔE
DA
disparity, or the so-
called “voltage losses”, is a result of particular electronic properties of the donor,
acceptor and the D/A interface. A more thorough understanding of molecular materials
properties that influence V
oc
is important to develop new OPV materials that diminish
losses and enhance the V
oc
. In this Chapter, we identify several molecular factors that
govern the open circuit voltage in small molecule based OPVs through the
comprehensive study of several donor materials and their impact on the V
oc
values. The
68

dependence of the open circuit voltage based on the saturation dark current (J
S
) will be
described, together with the portrayal of an accurate model for the dependence of the V
oc
.
Data will then be presented for several different OPVs in which different donor materials
are used with a common acceptor (i.e. C
60
) to illustrate the molecular properties that
influence V
oc
. Several examples are presented that relate molecular structure and
composition with the open circuit voltage. Later the acceptor nature is varied and the
impact of their molecular structure on the V
oc
value is observed.
3.2 Experimental
Pt tetraphenylnaphtholporphyrin (PtTPNP), and both Pt and Pd
tetraphenylbenzoporphyrin (PtTPBP and PdTPBP) were synthesized according to
literature by Dr. Carsten Borek,
15, 16
and purified by vacuum thermal gradient
sublimation. The organic materials, copper phthalocyanine (CuPc, 99% pure), PtOEP,
Tetracene (98%), Rubrene (98%), C
60
(99.5%), PTCBI (90%), PTCDI (90%) and 2,9-
dimethyl-4,7-diphenyl-1,10-phenanthroline (bathocuproine, BCP, 96%) were also
purified by sublimation prior to use. Cathode metal, Al (99.999%), was used as received.
Materials were sequentially grown by vacuum thermal evaporation at the following rates:
metal-TPBP and PtTPNP (1 Å/sec); CuPc, PtOEP, Rubrene, NPD, C
60
, PTCBI, PTCDI,
BCP (2 Å/sec); and Tetracene (15 Å/sec). Dark current characteristics were fit to
Equation 3.1 to obtain parameters J
S
, n and R
s
. J
SO
was calculated using the values for
ΔE
DA
from the literature. Thorough experimental details and testing methods are given in
Chapter 6.
3.3 The physical origin of the V
oc

Organic solar cell current density (J) vs. voltage characteristics can be described
according to the diode equation with the addition of a parallel and series resistance for the
equivalent circuit as referred to in Chapter 1. The equation for J then follows:
17, 18

) ( 1
(
exp V J
R
V
nkT
JR V q
J
R R
R
J
Ph
p
s
S
p s
p
−
⎪
⎭
⎪
⎬
⎫
⎪
⎩
⎪
⎨
⎧
+
⎥
⎦
⎤
⎢
⎣
⎡
− ⎟
⎠
⎞
⎜
⎝
⎛ −
+
= 3.1
where R
s
and R
p
are the series and parallel resistances, respectively, J
S
is the saturation
current, q is the fundamental charge, n is the diode ideality factor and J
ph
is the
photocurrent. At short circuit conditions it is safe to assume that (for very low dark
currents) that the photocurrent equals the short circuit current (J
ph
= J
SC
). For solar cells
with minimal leakage current such that the parallel resistance is extremely high (R
p
→
∝), Equation 3.1 can be simplified to:
17, 19, 20

SC
s
S
J
nkT
JR V q
J J −
⎥
⎦
⎤
⎢
⎣
⎡
− ⎟
⎠
⎞
⎜
⎝
⎛ −
= 1
) (
exp 3.2
The first term of this equation describes the thermal generated current in the dark,
typically dominated by recombination in small molecule OPVs (n ∼ 2), and the second
term refers to the optical carriers generated under illumination (see scheme 3.1). For
open circuit conditions (J = 0, V = V
oc
) the first term must be equal and opposed to the
light generated current such that J = 0. Thus, at a given characteristic J
SC
, a low dark
current (J
S
) will necessarily be reflected in a high V
oc
such that the first term in Equation
3.2 is balanced to equal J
SC
. Solving for V
oc
(J = 0) then yield:
69

⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
+ = 1 ln
S
SC
oc
J
J
q
nkT
V 3.3
In general, it is safe to assume that the photocurrent for reverse bias is always much
higher than the dark current (J
SC
>>J
S
) such that J
SC
/J
S
>>1. Equation 3.3 can then be
simplified to give Equation 3.4, in which V
oc
is directly tied to the ratio of J
SC
to J
S
:
5, 18, 20-
22

3.4
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
≈
S
SC
oc
J
J
q
nkT
V ln
Thus, at a particular J
SC
, a low dark current (given by J
S
) leads to a high V
oc
.

Scheme 3.1: Processes contributing to the current in organic photovoltaic cells comprised of
a donor an acceptor heterojunction.
The dark parameters J
S
and n can be determined by fitting the dark current-voltage
characteristic using Equation 3.1. Insight can be gained by examining the origin of J
S
,
which is the current resulting from carriers generated thermally at the D/A interface or
originating within the film bulk, however it is expected that carrier generation at the
70

interface will dominate the thermal current. The saturation current due to interface
charge generation can be described according to:
5, 17, 18, 20, 23

⎟
⎠
⎞
⎜
⎝
⎛
=
nkT
E
J J
a
SO S
exp 3.4
E
a
is the activation energy for the thermal carrier generation and it is defined by the
HOMO-LUMO energy difference such that E
a
= ΔE
DA
/2. The factor of 2 arises from the
fact that two carriers (an electron and hole) are generated at the D/A heterointerface.
18, 24

The magnitude of J
SO
depends on a number of materials properties which control the
efficiency of carrier generation independent of the D/A thermal barrier, ΔE
DA
. Factors
affecting the J
SO
include the reorganization energy for D→A electron transfer, the bulk
layer conductivities of electron and hole, the area of the D/A interface and the density of
states at the HOMO and LUMO energies of the D and A materials.
Substitution of Equation 3.4 into 3.3 yields:
q
E
J
J
q
nkT
V
DA
SO
SC
oc
2
ln
Δ
+
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
= 3.5
Equation 3.5 suggests a linear dependence of the V
oc
on the interface energy gap, and
a logarithmic in J
SC
/J
SO
. Such a linear correlation of V
oc
and ΔE
DA
has been reported
previously for closely related materials systems.
8, 12, 13
This expression is similar to, but
with important different physical implications from that recently reported by Potscavage,
et. al.
21
Here, ΔE
DA
/2 is used for the thermal activation energy for charge separation at
the D/A interface, such that at low temperatures, the measured V
oc
approaches ΔE
DA
/2. A
71

72

recent experimental study on the temperature dependence of the V
oc
in a series of vapor
deposited OPVs shows this behavior (at 175K V
oc
→ ΔE
DA
/2).
5

In the following section, a study of various material combinations and its effect on
the J
SO
and V
oc
, as expressed in Equation 3.5, will be presented. The film morphology
and intermolecular overlap influence J
SC
by limiting the exciton diffusion length, as well
as impacting the energetics at the D/A interface, thus affecting J
SO
. To achieve the
maximum possible V
oc
for a given D/A pair, J
SO
must be minimized and J
SC
maximized.
From the analysis presented in this Chapter we can further elaborate on the factors that
maximize V
oc
.
3.4 Results and Discussion
3.4.1 Donor analysis
A number of different photovoltaic devices were prepared, and both the light and
dark performance characteristics were studied. The device structure used was:
ITO/donor/C
60
(400Å)/bathocuproine (BCP) (100Å)/Al (1000Å), varying only the donor
material from device to device as depicted in Figure 3.1. In some cases, the donor layer
thickness was optimized to result in the highest J
SC
. Note that in our analysis, J
SO
is
independent of illumination conditions, simplifying comparisons between different
devices.
After device preparation, J-V characteristics in the dark and under illumination are
obtained. The dark response is fit to Equation 3.1 to extract n and J
S
. Device
illumination parameters such as J
SC
, V
oc
, FF and η are obtained under simulated 1 sun,
73

AM1.5 illumination conditions. OPVs with the structure similar to those used here,
utilizing rubrene,
25
tetracene,
26
NPD
7
and CuPc
27
donors have been reported. The data
presented in Table 3.1 were derived from devices prepared for the present study, and are
consistent with the literature reports for V
oc
, J
SC
, FF and η. Dark J-V curves for these
devices were used to determine J
S
and n for the devices studied here. Furthermore,
values of J
SO
were calculated using Equation 3.4, and are listed in Table 3.2. The
interface gap, ΔE
DA
, was calculated using the C
60
LUMO value (3.5eV) measured by
inverse photoelectron spectroscopy (IPES)
28
and the donor HOMOs were obtained both
from ultraviolet photoelectron spectroscopy (UPS) and electrochemistry.
29
The
calculated V
oc
values according to Equation 3.5 are also provided in Table 3.2, along with
the experimental values for comparison. The calculated V
oc
values closely match those
obtained experimentally, revealing the strength of the model presented here.

Figure 3.1: Molecular structures of donors and the corresponding HOMO (blue) and LUMO (red) for four of them. Phenyl rings are
shown as yellow space filling surfaces.
TOP SIDE SIDE

Tetracene CuPc

Rubrene PtTPBP

PtTPNP PtOEP NPD
74

To understand those molecular properties that lead to minimizing J
SO
for a given
ΔE
DA
, we first consider the donors, tetracene and rubrene, with results shown in Figure
3.2 and Table 3.1 that are consistent with those previously reported.
25, 26
Despite the
similarity of the HOMO energies of 5.1 eV for tetracene and 5.3 eV for rubrene,
30
the V
oc

for the two cells are 0.91 V and 0.58 V, respectively. The difference in V
oc
is due to the
large difference in the current dependent terms of Equation 3.5, principally in J
SO
. Here,
J
SO
is two orders of magnitude larger for tetracene than for rubrene (see Table 3.2).
-0.5 0.0 0.5 1.0
1E-5
1E-4
1E-3
0.01
0.1
1
10
100
Tetracene
Rubrene
Current Density ,mA/cm
2
Voltage, V

Figure 3.2: Dark (closed circles) and Light (open circles) logarithmic J-V characteristics of
tetracene and rubrene based OPVs (see Table 3.1 for device structure).
Whereas tetracene is a planar molecule consisting of fused conjugated aromatic
rings, the four nearly orthogonal pendant phenyl rings of rubrene prevent close
association of the tetracene cores of adjacent rubrene molecules or between the tetracene
75

core and the C
60
acceptor (see Figure 3.1). The rubrene poorer intermolecular overlap is
evident from comparisons between the solution and thin film spectra of the two
compounds.
25, 26, 31-34
While rubrene gives similar solution and thin film spectra, strong
intermolecular π-interactions in tetracene thin films give a broadened and red shifted
spectrum relative to solution. Weak intermolecular interactions in rubrene lead to a
decrease in both the rubrene-rubrene and rubrene-C
60
interactions, relative to the
tetracene interactions, and a correspondingly lower J
SO
.
-1.0 -0.5 0.0 0.5 1.0
1E-5
1E-4
1E-3
0.01
0.1
1
10
100
1000
Current Density (mA/cm
2
)
Voltage (V)
CuPc
PtTPBP
PtTPNP

Figure 3.3: Dark (closed circles) and Light (open circles) logarithmic J-V characteristics of
CuPc, PtTPBP and PtTPNP donor OPVs (see Table 3.1 for device structure).
The V
oc
measured for copper phthalocyanine (CuPc) and platinum tetraphenylbenzo-
porphyrin (PtTPBP) containing devices also shows a strong dependence on
intermolecular interactions, and thus J
SO
. The two compounds have comparable
76

π-systems, both with 38 π-electrons. While a larger V
oc
would be expected for the CuPc-
based device due to its higher ΔE
DA
( ΔE
DA
= 1.7eV, vs. ΔE
DA
= 1.4eV for PtTPBP/C
60
),
the experimental CuPc/C
60
V
oc
is actually less than that of PtTPBP by 0.21 V (c.f. Table
3.1). Note that CuPc is a flat molecule with strong intermolecular interactions in the
solid state. In contrast, PtTPBP has a highly distorted saddle-shaped conformation and
four orthogonal phenyl rings, which reduce the availability of the π-system to
intermolecular interactions and electronic delocalization.
15
As observed for tetracene and
rubrene, CuPc shows marked broadening of its low energy absorption in thin films
spectra,
35, 36
while the solution and thin film spectra of PtTPBP are comparable.
37
The
high V
oc
= 0.69 V for the PtTPBP device results from diminished recombination in the
porphyrin-based cells (Figure 3.3 and Table 3.2), and hence a lower J
SO
than for the CuPc
device than for the one with a PtTPBP donor (J
SO
= 1.5x10
4
mA/cm
2
and 12 mA/cm
2
,
respectively).

77

Donor

(Thickness, Å)
J
S

(μA/cm
2
)
V
oc

(V)
J
SC

(mA/cm
2
)
FF η (%) n
SubPc
(130)
3.2x10
-6
0.97 3.36
8
0.57
8
2.1
8
1.8
8

Rubrene
(200)
2.7x10
-3
0.9 2.3 0.50 1.1 2.7
NPD
(100)
5.17x10
-3
0.85 2.35 0.47 0.93 2.5
PtTPBP
(150)
0.020 0.69 4.5 0.63 1.9 2.1
Tetracene
(600)
0.077 0.55 1.8 0.54 0.53 2.2
PtOEP
(150)
0.081 0.52 1.6 0.59 0.49 2.0
PdTPBP*
(150 Å)
0.24 0.61 2.41 0.62 1.8 2.4
CuPc
(400)
1.1 0.48 5.5 0.6 1.6 2.0
PtTPNP
(120)
18 0.31 4.2 0.51 0.66 2.2
*Illumination intensity = 0.5 suns
Table 3.1: Dark and light characteristic parameters for all devices. Structure:
ITO/Donor/C
60
(400 Å)/BCP (100 Å)/Al (1000 Å).
We have also examined PdTPBP as a donor material. The structure of the Pd analog
is close to that of PtTPBP, where the Pd substitution shifts the HOMO energy slightly,
although it does not alter the intermolecular interactions. Hence, the Pd and Pt complexes
give similar J
SO
= 9mA/cm
2
and 12 mA/cm
2
, respectively.

78

*Illumination intensity = 0.5 suns
Donor J
SO

Film
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
so
sc
J
J
q
nkT
ln
q
E
DA
2
Δ
V
oc
calc. V
oc
exp.
SubPc 5.5 Am -0.02 1.0 0.98 0.97
Rubrene 1.1 Am 0.05 0.9 0.95 0.9
NPD 11 Am -0.10 0.95 0.85 0.85
PtTPBP 12 Am -0.05 0.7 0.65 0.69
Tetracene 150 PC -0.24 0.80 0.56 0.55
PtOEP 5.1x10
3
PC -0.40 0.9 0.50 0.52
PdTPBP* 9.1 - -0.082 0.65 0.57 0.61
CuPc 1.5x10
4
PC -0.41 0.85 0.44 0.48
PtTPNP 62 - -0.15 0.45 0.30 0.31
Table 3.2: Comparison of calculated values. J
SO
values are in mA/cm
2
and the remaining
data are in Volts. “Film” refers to the form of the donor thin film, as determined by X-ray
diffraction, Am = amorphous, PC = polycrystalline.
37-44

Electronic interaction or coupling between donor and acceptor molecules can be
modified by varying the level of π-conjugation. To explore these effects, platinum
tetraphenyl naphtholporphyrin (PtTPNP, Figure 3.1) was used as the donor material.
PtTPNP is an analog of PtTPBP whose π-system has been extended by benzanulation of
79

four rings onto the benzo moieties of PtTPBP. The lowest energy conformation of the
naphthol analog has a saddle-shape similar to PtTPBP, resulting from steric repulsions of
the meso and pyrolle substituents. The current-voltage characteristics of the PtTPNP
donor-containing devices are shown in Figure 3.3, along with data for PtTPBP and CuPc-
based devices, with their performance provided in Tables 3.1 and 3.2. While the PtTPNP
and PtTPBP-based cells have similar photocurrent densities, V
oc
is reduced to 0.31 V for
the PtTPNP device. This low V
oc
is due to an increased π-conjugation of the PtTPNP
which both reduces the HOMO energy compared to either PtTPBP or CuPc (PtTPNP
HOMO energy = 4.4 eV) and it also provides enhanced π overlap with near-neighbor
molecules therefore increasing J
SO
. The PtTPNP/C
60
cell has a ΔE
DA
= 0.8 eV, markedly
lower than either PtTPBP or CuPc and the J
SO
is increased by a factor of five for PtTPBP
(J
SO
= 12 mA/cm
2
, compared to and 62 mA/cm
2
for PtTPNP). The benzannulated rings
added to PtTPNP extend beyond the meso-phenyl groups, increasing the π-π overlap, and
thus intermolecular interaction. The resulting increase in J
SO
, along with its decreased
ΔE
DA
, significantly decrease V
oc
for this device, relative to PtTPBP. While π-expansion
leads to a somewhat increased J
SO
relative to PtTPBP, the J
SO
for PtTPNP is more than
200 times lower than that of CuPc suggesting that molecular shape and steric hindrance
has a more relevant effect on the coupling term, J
SO
.
To explore planar donors with π-systems of different sizes, Pt octaethylporphyrin
(PtOEP) and CuPc were compared (Figure 3.1).
45
The CuPc π-system is significantly
larger than that of PtOEP, which have 38 and 22 π-electrons, respectively. The J
SO
of
CuPc is higher than that of PtOEP (J
SO
= 1.5x10
4
mA/cm
2
and 5.1x10
3
mA/cm
2
,
80

respectively), but the difference is not nearly as great as those observed for the
comparison between CuPc and PtTPBP, consistent with significant steric blocking of the
meso phenyls of PtTPBP.
The analysis can be applied to other donor molecules used in literature based
devices.
5, 8
For example, N,N ′-Bis(naphthalen-1-yl)-N,N ′-bis(phenyl)benzidine (NPD) –
based devices (see Figure 3.1 and Figure 3.4 for the J-V performance) give V
oc
= 0.85 V,
consistent with ΔE
DA
= 1.9 eV and reduced recombination losses as reflected by the small
J
SO
. In this case, J
SO
= 11 mA/cm
2
is similar to that of PtTPBP, indicative of reduced π
accessibility and weak interaction with the C
60
acceptor, due to lose intermolecular
packing of the nonplanar NPD molecules. Furthermore, subphthalocyanine (SubPc)
exhibits a larger HOMO energy than CuPc, resulting in ΔE
DA
= 1.9 eV, leading to a high
V
oc
of 0.97 V for a SubPc/C
60
based OPV.
8
SubPc has an out-of-plane cone shape whose
steric effects reduce intermolecular interaction, and thus recombination at the D/A
interface. This gives J
SO
= 5.5 mA/cm
2
, which is sufficiently small and it does not
significantly reduce the value of V
oc
as determined from the interface energy gap. SubPc
is an example of a donor molecule that comprises features that maximizes the V
oc
, strong
absorptivity in the visible region, coupled with high ΔE
DA
and low J
SO
.
81

-0.5 0.0 0.5 1.0
1E-5
1E-4
1E-3
0.01
0.1
1
10
100
Current Density, mA/cm
2
Voltage, V
Dark
Light

Figure 3.4: Dark (closed circles) and Light (open circles) logarithmic J-V characteristics of
NPD based OPVs (see Table 3.1 for device structure).
X-ray diffraction studies of the donor materials investigated here indicate that both
amorphous and polycrystalline thin films are formed, Table 3.2.
37-44
Materials that show
evidence of aggregation in thin film absorption spectra, i.e. tetracene, PtOEP and CuPc,
give polycrystalline thin films. Materials that show little or no evidence of aggregation in
thin film absorption spectra, i.e. SubPc, rubrene, NPD and PtTPBP, form amorphous thin
films. As expected, there is a clear distinction in J
SO
values between the amorphous
materials (J
SO
= 1.1-12 mA/cm
2
) and polycrystalline materials (J
SO
= 150-1.5x10
4

mA/cm
2
) studied here. The π−π stacking forces that encourage crystal growth in thin
films likely lead also to strong donor/C
60
interactions, contributing to the high J
SO
values
observed for the polycrystalline materials.
82

3.4.2 Acceptor analysis
The same analysis can be used to study other D/A couples by changing the nature of
the acceptor layer. The availability of molecules that can be applied as acceptor layer is
much reduced compared to that of the donor and only three molecules are discussed here,
3,4,9,10-perylene-tetracarboxylic-bisbenzimidazole (PTCBI), 3,4,9,10-perylene-
tetracarboxylic-diimide (PTCDI) and C
60
as seen in Figure 3.5. Devices made with CuPc
as the donor and C
60
, PTCBI and PTCDA as the acceptors are presented in Figure 3.6.
Parameters are listed in tables 3.3 and 3.4 in a similar fashion to the previous analysis for
the donors.
83

NH HN
O
O
O
O
N N
O
O
N
N
PTCDI
PTCBI

C
60

Figure 3.5: Molecular structures of acceptors.

-0.5 0.0 0.5
1E-5
1E-4
1E-3
0.01
0.1
1
10
100
1000
10000
Current Density, mA/cm
2
Voltage, V
C
60
PTCBI
PTCDI
dark
light

Figure 3.6: Dark (closed circles) and Light (open circles) logarithmic J-V characteristics of
C
60
, PTCBI and PTCDI based OPVs (see Table 3.3 for device structure).

Acceptor

(Thickness, Å)
J
S

(μA/cm
2
)
V
oc

(V)
J
SC

(mA/cm
2
)
FF η (%) n
C
60

(400)
1.1 0.48 5.5 0.6 1.6 2.0
PTCBI
(400)
0.12 0.44 4.2 0.61 1.1 1.6
PTCDI*
(200)
8.87 0.24 1.8 0.53 0.5 1.7
*Illumination intensity = 0.5 suns
Table 3.3: Dark and light characteristic parameters for all devices. Structure: ITO/CuPc(400
Å)/Acceptor/BCP (100 Å)/Al (1000 Å).
84

The J
SO
values for the flat molecules, PTCBI and PTCDI, come very close to those
values observed for C
60
devices. The PTCBI cell presents an only slightly higher
coupling term, almost double than that of the other two acceptors, probably due to the
increased π system of this molecule compared to the PTCDI. Both PTCBI and PTCDI
are planar molecules with a very large system of π electrons as observed from Figure 3.5.
C
60
presents a spherical 3-D shape (non-planar) with a lower surface contact area but with
little-to-none hindrance of its enormous amount of π electrons such that its J
SO
results
comparable to that of planar molecules. C
60
is known to be a very efficient electron
conductor due to this extensive π cloud which can be related to a good intermolecular
overlap between neighboring molecules. Also, its particular shape allows for efficient
interactions with the donor molecule allowing for good recombination at the interface and
therefore high dark current, J
S
. For the PTCDI case, an elevated dark current, J
S
, that
yields a much reduced V
oc
, is a consequence of both a high J
SO
and a reduced ΔE
DA.

Acceptor J
SO

⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
so
sc
J
J
q
nkT
ln
q
E
DA
2
Δ
V
oc
calc. V
oc
exp.
C
60
1.5x10
4
-0.41 0.85 0.44 0.48
PTCBI 3.8x10
4
-0.38 0.8 0.43 0.44
PTCDI* 1.7x10
4
-0.40 0.63 0.23 0.24

*Illumination intensity = 0.5 suns
Table 3.4: Comparison of calculated values. J
SO
values are in mA/cm
2
and the remaining
data are in Volts.
85

It is proven that the use of this model is also possible for different acceptors and the
calculated V
oc
values come very close to those observed experimentally. An increase of
the V
oc
from the acceptor point of view, could be achieved by hindering the
intermolecular interaction by means of bulking up the molecular shape. However, it is
really hard to maintain good electron conductivities for those systems. Whereas efficient
hole mobility is relatively easily achieved for a number of molecules, good electron
mobilities are a much harder to conceive and very resistive films originate much reduced
FF when applied to organic photovoltaics. Whereas modification of the donor layer is a
much straightforward task, successful acceptors for OPVs have been hardly presented in
literature.
3.5 Conclusions
The correspondence of the calculated values for V
oc
obtained using Equation 3.5 to
experimental values for all D/A materials combinations examined provides strong
evidence for a relationship between molecular structure and film structure, and the
saturation dark current in organic PV cells. Specifically, we show a correspondence
between the strength of intermolecular interaction and the saturation current density, J
SO
.
Weak intermolecular interactions lead to low values of J
SO
, and hence correspondingly
high open circuit voltages. Similar relationships are observed for acceptor materials and
it is demonstrated that C
60
behaves very similarly to planar molecules with extensive π
cloud.
86

The present study focused solely on how changes in the donor and acceptor materials
of a lamellar, vacuum deposited OPV influence J
SO
, and V
oc.
Changes other than the
nature of the photoactive nature of the D or A layers are also expected to affect
recombination rates in the dark. For example, the addition of an electron blocking layer
adjacent to the OPV anode has been shown to reduce significantly the magnitude of the
dark current as was demonstrated by Li, et al.
22
This approach may allow the use of
materials with strong intermolecular overlap leading to comparatively high J
SO
, without a
concomitant reduction in open circuit voltage.
To achieve the highest possible open circuit voltage, we find that materials
combinations with both a high ΔE
DA
and low J
SO
are required. Materials that show weak
intermolecular interactions in thin films often result in poor carrier and exciton transport.
Since weak intermolecular interactions also are found to result in a low J
SO
, materials
optimized for low dark currents will generally also have short exciton diffusion lengths
and poor carrier conductivity, leading to a correspondingly low J
SC
. These contradictory
performance characteristics would appear to provide a fundamental limit to the ability to
achieve an optimally high value of V
oc
as well as J
SC
in a single materials pair. High
power conversion efficiency is possible, however, for materials with significant overlap
between the absorption of the OPV materials and the solar spectrum (high J
SC
), and that
also have steric inaccessibility of the π-system or reduced π-extension (low J
SO
).
Rubrene and PtTPBP donors are examples of such materials.
26, 37
This work, therefore,
can serve as a guide for the molecular design of both D and A materials with improved
spectral overlap, while achieving both a low J
SO
and a high ΔE
DA
.
87

3.6 Chapter 3 References
1. Xue, J.G., S. Uchida, B.P. Rand, and S.R. Forrest, 4.2% efficient organic
photovoltaic cells with low series resistances. Applied Physics Letters, 2004. 84(16): p.
3013-3015.
2. Peet, J., J.Y. Kim, N.E. Coates, W.L. Ma, D. Moses, A.J. Heeger, and G.C. Bazan,
Efficiency enhancement in low-bandgap polymer solar cells by processing with alkane
dithiols. Nat Mater, 2007. 6(7): p. 497-500.
3. Mihailetchi, V.D., P.W.M. Blom, J.C. Hummelen, and M.T. Rispens, Cathode
dependence of the open-circuit voltage of polymer : fullerene bulk heterojunction solar
cells. Journal of Applied Physics, 2003. 94(10): p. 6849-6854.
4. Koster, L.J.A., V.D. Mihailetchi, R. Ramaker, and P.W.M. Blom, Light intensity
dependence of open-circuit voltage of polymer:fullerene solar cells. Applied Physics
Letters, 2005. 86(12): p. 123509.
5. Rand, B.P., D.P. Burk, and S. Forrest, R., Offset energies at organic semiconductor
heterojunctions and their influence on the open-circuit voltage of thin-film solar cells.
Physical Review B (Condensed Matter and Materials Physics), 2007. 75(11): p. 115327.
6. Mandoc, M.M., L.J.A. Koster, and P.W.M. Blom, Optimum charge carrier mobility
in organic solar cells. Applied Physics Letters, 2007. 90: p. 133504.
7. Liu, J., S. Y., and Y. Yang, Solvation-Induced Morphology Effects on the
Performance of Polymer-Based Photovoltaic Devices. Advanced Functional Materials,
2001. 11(6): p. 420-424.
8. Mutolo, K.L., E.I. Mayo, B.P. Rand, S.R. Forrest, and M.E. Thompson, Enhanced
open-circuit voltage in subphthalocyanine/C-60 organic photovoltaic cells. Journal of the
American Chemical Society, 2006. 128(25): p. 8108-8109.
9. Brabec, C.J., A. Cravino, D. Meissner, N.S. Sariciftci, T. Fromherz, M.T. Rispens, L.
Sanchez, and J.C. Hummelen, Origin of the open circuit voltage of plastic solar cells.
Advanced Functional Materials, 2001. 11(5): p. 374-380.
10. Gadisa, A., M. Svensson, M.R. Andersson, and O. Inganas, Correlation between
oxidation potential and open-circuit voltage of composite solar cells based on blends of
polythiophenes/ fullerene derivative. Applied Physics Letters, 2004. 84(9): p. 1609-1611.
11. Kooistra, F.B., J. Knol, F. Kastenberg, L.M. Popescu, W.J.H. Verhees, J.M. Kroon,
and J.C. Hummelen, Increasing the Open Circuit Voltage of Bulk-Heterojunction Solar
Cells by Raising the LUMO Level of the Acceptor. Org. Lett., 2007. 9(4): p. 551-554.
88

12. Scharber, M.C., D. Wuhlbacher, M. Koppe, P. Denk, C. Waldauf, A.J. Heeger, and
C.L. Brabec, Design rules for donors in bulk-heterojunction solar cells - Towards 10 %
energy-conversion efficiency. Advanced Materials, 2006. 18(6): p. 789-+.
13. Vandewal, K., A. Gadisa, W.D. Oosterbaan, S. Bertho, F. Banishoeib, I. Van
Severen, L. Lutsen, T.J. Cleij, D. Vanderzande, and M.J. V., The Relation Between Open-
Circuit Voltage and the Onset of Photocurrent Generation by Charge-Transfer
Absorption in Polymer : Fullerene Bulk Heterojunction Solar Cells. Advanced
Functional Materials, 2008. 18(14): p. 2064-2070.
14. Sarangerel, K., C. Ganzorig, M. Fujihira, M. Sakomura, and K. Ueda, Influence of
the Work Function of Chemically Modified Indium-Tin-Oxide Electrodes on the Open-
circuit Voltage of Heterojunction Photovoltaic Cells. Chemistry Letters, 2008. 37(7): p.
778-779.
15. Borek, C., K. Hanson, P.I. Djurovich, M.E. Thompson, K. Aznavour, R. Bau, Y.
Sun, S.R. Forrest, J. Brooks, L. Michalski, and J. Brown, Highly Efficient, Near-Infrared
Electrophosphorescence from a Pt-Metalloporphyrin Complex. Angewandte Chemie
International Edition, 2007. 46(7): p. 1109-1112.
16. Rogers, J.E., K.A. Nguyen, D.C. Hufnagle, D.G. McLean, W.J. Su, K.M. Gossett,
A.R. Burke, S.A. Vinogradov, R. Pachter, and P.A. Fleitz, Observation and
interpretation of annulated porphyrins: Studies on the photophysical properties of meso-
tetraphenylmetalloporphyrins. Journal of Physical Chemistry A, 2003. 107(51): p.
11331-11339.
17. Bube, H.R. and A.L. Fahrenbruch, Advances in Electronics and Electron Physics.
1981, Academic: New York. p. 163.
18. Fahrenbruch, A.L. and J. Aranovich, Solar Energy Conversion - Solid-State Physics
Aspects. Topics in Applied Physics, ed. B.O. Seraphin. Vol. 31. 1979, Berlin Heidelberg
New York: Springer-Verlag. 257-326.
19. Nelson, J., The Physics of Solar Cells. 2003, London: Imperial College Press.
20. Würfel, P., Physics of Solar Cells: From Principles to New Concepts. 2005,
Weinhem: Wiley-VCH. 186.
21. Potscavage, J.W.J., S. Yoo, and B. Kippelen, Origin of the open-circuit voltage in
multilayer heterojunction organic solar cells. Applied Physics Letters, 2008. 93(19): p.
193308-3.
22. Li, N., B.E. Lassiter, R.R. Lunt, G. Wei, and S.R. Forrest, Open circuit voltage
enhancement due to reduced dark current in small molecule photovoltaic cells. Applied
Physics Letters, 2009. 94(2): p. 023307-3.
89

23. Sze, S.M., Physics of semiconductor devices (2nd edition). 1981, United States:
Wiley-Interscience,New York, NY. Pages: 878.
24. Meier, H., Organic semiconductors : dark and photoconductivity of organic solids
1974: Verlag Chemie.
25. Taima, T., J. Sakai, T. Yamanari, and K. Saito, Realization of large open-circuit
photovoltage in organic thin-film solar cells by controlling measurement environment.
Japanese Journal of Applied Physics Part 2-Letters & Express Letters, 2006. 45(37-41):
p. L995-L997.
26. Chu, C.W., Y. Shao, V. Shrotriya, and Y. Yang, Efficient photovoltaic energy
conversion in tetracene-C
60
based heterojunctions. Applied Physics Letters, 2005.
86(24): p. 243506.
27. Peumans, P., A. Yakimov, and S.R. Forrest, Small molecular weight organic thin-
film photodetectors and solar cells. Journal of Applied Physics, 2003. 93(7): p. 3693-
3723.
28. Schwedhelm, R., L. Kipp, A. Dallmeyer, and M. Skibowski, Experimental band gap
and core-hole electron interaction in epitaxial C60 films. Physical Review B, 1998.
58(19): p. 13176.
29. D'Andrade, B.W., S. Datta, S.R. Forrest, P. Djurovich, E. Polikarpov, and M.E.
Thompson, Relationship between the ionization and oxidation potentials of molecular
organic semiconductors. Organic Electronics, 2005. 6(1): p. 11-20.
30. Sato, N., K. Seki, and H. Inokuchi, Polarization Energies of Organic-Solids
Determined by Ultraviolet Photoelectron-Spectroscopy. Journal of the Chemical Society-
Faraday Transactions II, 1981. 77: p. 1621-1633.
31. Sakurai, T. and S. Hayakawa, Optical-Properties of Tetracene Evaporated-Films.
Japanese Journal of Applied Physics, 1974. 13(11): p. 1733-1740.
32. Pandey, A.K. and J.M. Nunzi, Rubrene/Fullerene Heterostructures with a Half-Gap
Electroluminescence Threshold and Large Photovoltage. Advanced Materials, 2007.
19(21): p. 3613-3617.
33. Badger, G.M. and R.S. Pearce, Absorption spectrum of rubrene in different solvents.
Spectrochimica Acta, 1951. 4(4): p. 280-283.
34. Chan, M.Y., S.L. Lai, M.K. Fung, C.S. Lee, and S.T. Lee, Doping-induced efficiency
enhancement in organic photovoltaic devices. Applied Physics Letters, 2007. 90(2): p.
023504.
90

35. Brown, R.J.C., A.R. Kucernak, N.J. Long, and C. Mongay-Batalla, Spectroscopic
and electrochemical studies on platinum and palladium phthalocyanines. New Journal of
Chemistry, 2004. 28(6): p. 676-680.
36. Ferreira, J.A., R. Barral, J.D. Baptista, and M.I.C. Ferreira, Absorption coefficients
and fluorescence quantum yields of porphyrin films determined by optical and
photoacoustic spectroscopies. Journal of Luminescence, 1991. 48-49(1): p. 385-390.
37. Perez, M.D., C. Borek, P.I. Djurovich, E.I. Mayo, R.R. Lunt, S.R. Forrest, and M.E.
Thompson, Organic photovoltaics using tetraphenylbenzoporphyrin complexes as donor
layers. Advanced Materials, 2009. 21(14-15): p. 1517-1520
38. Gundlach, D.J., J.A. Nichols, L. Zhou, and T.N. Jackson, Thin-film transistors based
on well-ordered thermally evaporated naphthacene films. Applied Physics Letters, 2002.
80(16): p. 2925-2927.
39. Mattheus, C.C., W. Michaelis, C. Kelting, W.S. Durfee, D. Wöhrle, and D.
Schlettwein, Influence of the molecular shape on the film growth of a substituted
phthalocyanine. Synthetic Metals, 2004. 146(3): p. 335-339.
40. Mori, T., S. Oda, N. Ooishi, and Y. Masumoto, Polycrystallization of vaporized
hole-transport materials for organic light-emitting diodes and its suppression using
organic alloy method. Japanese Journal of Applied Physics Part 1-Regular Papers Brief
Communications & Review Papers, 2007. 46(9A): p. 5954-5959.
41. Seo, S., B.-N. Park, and P.G. Evans, Ambipolar rubrene thin film transistors.
Applied Physics Letters, 2006. 88(23): p. 232114-3.
42. Weinberg-Wolf, J.R., L.E. McNeil, S. Liu, and C. Kloc, Evidence of low
intermolecular coupling in rubrene single crystals by Raman scattering. Journal of
Physics: Condensed Matter, 2007. 19(27): p. 276204.
43. Phthalocyanines : properties and applications ed. C.C. Leznoff and A.B.P. Lever.
1989, New York: VCH.
44. Lunt, R.R., N.C. Geibink, A.A. Belak, J.B. Benziger, and S.R. Forrest, Measurement
of the Exciton Diffusion Length of Amorphous and Polycrystalline Organic
Semiconductors by Spectrally Resolved Photoluminescence Quenching. Manuscript in
preparation, 2009.
45. Shao, Y. and Y. Yang, Efficient organic heterojunction photovoltaic cells based on
triplet materials. Advanced Materials, 2005. 17(23): p. 2841-2844.

91

Chapter 4. Charge collection at the organic/metal interface
4.1 Introduction
Organic solar cells operate based primarily on the energy level offset at the pn-like
heterojunction at the donor/acceptor interface. Upon light absorption, excitons reaching
the interface dissociate into an electron (acceptor) and hole (donor) and are further
collected at the cathode and anode respectively. The energetics of both opposing
electrodes must result in a favorable built-in electric field for the charges to migrate and
be collected by the external circuit. Variations in metal work-function have been
therefore reported to play an important role in defining the device efficiency and
performance parameters.
1-5
In addition, charge collection efficiency ( η
cc
) at the electrode
level is also of relevant importance. Charge collection issues in solar cells are normally
disregarded and high efficiency is assumed ( η
cc
~ 100%) presupposing an ideal ohmic
contact between the electrode and the organic material. However, this assumption does
not hold for all cases. In the field of organic light emitting diodes (OLED), for example,
much is known about the nature of the metallic contact, and problems like organic/metal
reactivity or wetting, have been somewhat circumvented to ensure high charge collection
efficiency at the cathode level.
6-8
For OPVs, it is well known that in most cases, a
protective layer of a wide band gap organic material (WGO), the so-called “buffer” layer,
is needed to prevent damage of the photoactive organic films by the cathode and maintain
a high η
cc
, in particular when C
60
is used as the acceptor layer.
9
Bathocuproine (BCP) has
been mostly used as this layer but a number of other materials can provide this function
92

as well, and there is still much to be learned about the effect of this extra non-photoactive
organic layer. While both Ag and Al have been extensively used as metal cathodes in
organic photovoltaics, they have resulted in devices with disparate performance.
10
This
variation is surprising as the two metals were reported to have very similar work
functions.
9, 11-14
A deeper understanding of charge collection at the cathode level is of
major importance in the long term development of small molecule solar cells. This
Chapter will first focus on the origin of the performance differences encountered by
analogous solar cells using Ag and Al as cathode materials. Later, the effects of the
buffer layer materials properties on the charge collection efficiency will be presented.
4.2 Experimental
Organic films with the structure CuPc (200 Å)/C
60
(400 Å)/buffer (x Å) were
sequentially grown by vacuum thermal evaporation at the following rates: CuPc (2
Å/sec), C
60
(2 Å/sec), and buffer (2 Å/sec). Buffer materials, Alq3 and NPD were
commercially available from Sigma-Aldrich and used with prior purification by thermal
sublimation. NPt was synthesized in the lab as described in Chapter 5 and also used after
purification by thermal sublimation. The metal cathodes were evaporated (Ag (4 Å/sec)
or Al (2 Å/sec)) through a shadow mask with 1 mm diameter openings to yield devices of
an area of 0.0075 cm
2
that was measured under an optical microscope. The total cathode
thickness was 1000 Å thick for all combination of metals. Current-voltage (J-V)
characteristics of PV cells were measured under simulated AM1.5G solar illumination
(Oriel Instruments) using a Keithley 2420 3A Source Meter. Incident power was
93

measured using a thermopile detector (Oriel Instruments), therefore power intensities are
only approximated but consistent throughout comparisons. Spectral response
measurements were performed using a tunable wavelength light source from Newport-
Oriel instruments consisting of a 300W Xe lamp coupled to a monochromator.
4.3 Results and Discussion
4.3.1 Charge collection from the cathode material perspective
4.3.1.1 Origin of the differences between Ag and Al cathode
Ag and Al cathodes knowingly result in different performance when used as
cathodes in small molecule solar cell. The origins of such differences are observed from
the performance of standard devices. Figure 4.1 (a) shows the J-V characteristics of a
standard device with the structure CuPc (200Å)/C
60
(400Å)/BCP (100Å) with 1000 Å of
Al or Ag as the cathode. The V
oc
and FF are very similar for these two devices (within
the measurement error) while the J
SC
of the Ag device is significantly larger than for the
Al device. This increase in photocurrent for Ag vs Al devices is consistent with previous
literature reports.
1, 2, 4, 5
A slight decrease in photocurrent when switching from Ag to Al
is expected based on the differences in reflectivity of the two metals. However, optical
field calculations indicate that the photocurrent of the Al device should be at least 94% of
the Ag device, indicating that while reflectivity is a factor, it does not account for the
observed 30% difference in photocurrent.
94

Figure 4.1: (a) J-V characteristics under approximately 50 mW/cm
2
AM1.5G simulated
illumination and (b) spectral responsivity of CuPc (200Å)/C
60
(400Å)/BCP (100Å) cells with Al
(- ■-) and Ag (- ●-) as cathode materials. Absorption spectra of CuPc (200Å) and C
60
(400Å) films
are overlayed in (b).
95

The spectral response of these two devices (Figure 4.1 (b)) can help explain the
differences in photocurrent. The CuPc response (~610 nm and ~700 nm) is higher for the
device with the Ag cathode. The C
60
response, however, is slightly more convoluted.
Above 420 nm, the C
60
response is similarly greater for the Ag device while, below 420
nm there is a decrease in the spectral response for Ag cathode device. As mentioned
previously, there are differences in the reflectivity of these two metals with Ag showing a
significant drop-off in reflectivity at ~ 420 nm (Figure 4.2),
15
indicating that the
differences in the spectral response below this wavelength are due to a decrease in the
number of high energy photons back reflected off the cathode. However, even when
considering only wavelengths where the reflectivity is the same for Ag and Al, it is clear
that there is a broad loss of photoactivity in the Al devices that is indicative of a lower
η
cc
.

Figure 4.2: Metal reflectances. (Reproduced from http://en.wikipedia.org/wiki/Reflectance)
96

There are several loss mechanisms that may attribute to the decreased charge
collection in Al devices relative to Ag. A variety of electronic work functions for these
metals has been reported, and the measured value is highly dependent on the cleanliness
of the samples and the surface quality achieved during deposition.
13, 14, 16-18
However, the
most commonly cited values for the Al and Ag workfunctions are 4.26 eV and 4.28 eV,
respectively.
9, 11-14
Even with variation in work function measurement aside, it is likely
that the metal work functions will shift slightly upon contact with organic materials.
19

This would affect the built-in potential across the device resulting in a different driving
force for exciton dissociation. One must also take into account the fact that these metals
have very distinctive reactivities in the presence of minute amounts of oxygen. The heat
of formation ( ΔH
form
) of Al
2
O
3
is more than an order of magnitude larger than AgO,
13

indicating that there is likely to be a more appreciable amount of oxide in the Al devices.
The presence of an oxide between the metal and the bulk organic film could affect charge
collection in two ways: by shifting the work function, thereby altering the built-in field or
by creating a physical barrier which can affect the degree of ohmic contact.
4.3.1.2 Ag capping effects
In order to probe the origin of the differences in photocurrent, varying cathode
compositions were tested using both Al and Ag in different amounts (Figure 4.3). As
shown in Figure 4.3 (a), Al is deposited first such that it makes direct contact with the
underlying organic layer followed by addition of Ag to act as the top electrode contact.
In this series of experiments, a thin (100 Å) Al under-layer is deposited on the organic,
97

followed by a thick (900 Å) cover-layer of Ag. As the Al thickness is increased, the
capping Ag layer thickness is decreased to always maintain a total cathode thickness of
1000 Å. Alternatively, Ag can be deposited first (Figure 4.3 (b)) with a capping layer of
Al. Once again, as the Ag thickness is increased, the Al capping layer thickness is
adjusted to maintain a 1000 Å total cathode thickness.

Figure 4.3: Depiction of experiments performed, total thickness of cathode is maintained at
1000Å for all devices.
Figure 4.4 shows a series of devices where the thickness of the Al under-layer was
increased from 0 Å (pure Ag cathode) to 1000 Å (pure Al cathode). The device structure
used here was ITO/CuPc (200 Å)/C
60
(400 Å)/BCP (100 Å)/Al (X Å)/Ag (1000-X Å),
Figure 4.3 (a). The efficiency data has been scaled to an efficiency value of 1.0 for a
device with 0 Å of Al, i.e. 1000 Å Ag cathode. Whenever Al is in direct contact with the
organic film, the resulting device efficiency is improved by addition of an over-layer of
98

Ag relative to a device with no Ag cap (i.e. 1000 Å Al cathode). The parameters for the
J-V characteristics of those devices presented in Figure 4.4 are presented in Table 4.1.

Figure 4.4: Device performance with different proportions of Al and Ag for a CuPc(200
Å)/C
60
(400 Å)/BCP(100 Å)/Al(X Å)/Ag(1000-X Å) device. Al is always deposited first as shown
in Figure 4.3 (a). Yellow and blue triangles above the plot illustrate the relative thicknesses of
Ag and Al, respectively.
Al thickness Ag thickness η% J
sc
(mA/cm
2
) V
oc
(V) FF
0 1000 1.60 3.31 0.396 0.61
100 900 1.13 2.62 0.383 0.56
200 800 1.37 2.91 0.405 0.58
300 700 1.35 2.99 0.406 0.56
400 600 1.49 3.18 0.400 0.59
500 500 1.84 3.86 0.401 0.59
600 400 1.45 3.09 0.409 0.57
700 300 1.57 3.25 0.409 0.59
800 200 1.41 3.04 0.399 0.58
900 100 1.28 2.82 0.386 0.59
1000 0 1.18 2.66 0.384 0.58
Table 4.1: Illumination parameters for devices presented in Figure 4.4 under illumination
intensity of 50 mW/cm
2

99

In contrast, when Ag is in physical contract with the organic (Figure 4.3 (b)),
capping with Al shows no effect compared to a pure Ag device, regardless of the
thickness of the underlying Ag layer. Devices with Ag in direct contact with the organic
film give the same efficiencies, within experimental error, independent of the
combination of thicknesses of the two metals, as observed in Figure 4.5. The conversion
efficiencies of devices with cathodes composed of 100-900 Å of Ag match those with
pure Ag cathode. This is most likely because Ag does not undergo spontaneous oxidation
under atmospheric exposure and the organic/metal interface remains undisturbed, even
when reactive Al is placed on top. This evidence agrees with electron collection being
dominated by the nature of the metal/organic interface.

0 200 400 600 800 1000
0.8
0.9
1.0
1.1
1.2
1.3
1.4
1.5
1.6
1.7
1.8
Ag thickness (Å)
η (%)
Al
Ag
1000 800 600 400 200 0
Al thickness (Å)

Figure 4.5: Performance with different proportions of Ag and Al for a
CuPc(200Å)/C
60
(400Å)/BCP(100Å)/Metal(1000Å) device under AM1.5G simulated illumination
with intensity of approximately of 50 mW/cm
2
. Ag is deposited first as shown in Figure 4.3 (b),
such that Ag is always in contact with the organic layer.
100

For all of the devices reported here the V
oc
and FF were similar, with the largest
difference in these parameters being less than 8%. The efficiency differences are due
predominantly to a decreased photocurrent density resulting from lower charge collection
efficiency at the organic/Al interface compared to the organic/Ag interface. Based on the
accepted idea that the work functions of these two metals are very similar, it is clear that
the presence of Al
2
O
3
creates a barrier to charge collection at the metal organic interface
either by physically limiting electron injection to the cathode or through modification of
the work function and subsequently the driving force for charge injection into the
cathode.
For underlying Al thicknesses of 200 Å to 800 Å and subsequent Ag capping
thicknesses of 800 Å to 200 Å (see Figure 4.4, boxed region), the Ag forms a protective
layer that minimizes the degree of aluminum oxide formation at the Al/organic interface.
Since the degree of oxidation is minimal for these devices, the resulting charge collection
efficiencies are similar to that of a pure Ag device due to the similar work functions of
both metals. However, this does not hold for the two outliers in Figure 4.4. When only
100 Å of Ag is used to cap 900 Å of Al in contact with organic, the Ag cannot form a
sufficiently uniform protective barrier over the very thick Al film, resulting in Al oxide
formation (Figure 4.6 (a)). Similarly, when only 100 Å of underlying Al is deposited, the
Al is unlikely to form a continuous layer, making a discussion of what is happening at the
metal organic interface convoluted (Figure 4.6 (b)). However, it is somewhat surprising
that this device, which is quite similar to pure Ag, has such poor charge collection
efficiency.
101

Figure 4.6: Representation of Al layer on top of the organics with small amount of Ag
addition (a) and inhomogeneous layer of Al next to organics plus a thick layer of Ag (b)
The role of Al oxidation in minimizing charge collection is further supported by the
fabrication of a device where the cathode layer was formed by co-depositing both Ag and
Al in same amounts effectively mimicking the “50:50” case but not controlling the
composition of the metal/organic interface (Table 4.2). The photocurrent of the resulting
device was very similar to that of the device with neat Al as cathode. This observation
suggests that Al oxidation is not prevented by just the presence of Ag but through a
capping effect over the underlying Al film.
Cathode ɳ %
Jsc
(mA/cm
2
)
V
oc

(V)
FF
Al 1.18 2.66 0.38 0.58
Ag 1.60 3.31 0.40 0.61
Codeposited
Ag+Al
1.15 2.48 0.38 0.60
Table 4.2: Parameters for the performance of solar cells with different cathodes materials
under AM1.5G simulated illumination with intensity of approximately of 50 mW/cm
2
for
CuPc(200 Å)/C
60
(400 Å)/BCP(100 Å)/Metal(1000 Å) devices.

102

4.3.2 Charge collection from the organic material or buffer layer perspective
It has been proposed that the role of the buffer layer is predominately to block the
excitons generated from the acceptor side by preventing metal damage to the photoactive
layers. Figure 4.7 illustrates the significantly enhanced photocurrent and FF that is
obtained when the organic blocking layer is included on a CuPc/C
60
device. Parameters
for both devices are presented in Table 4.3. The EQE of the BCP device is higher than
the device without BCP, with the largest increase observed in the C
60
response region.
This supports the fact that more C
60
excitons are able to reach the D/A interface and
contribute to the photocurrent rather than being quenched at the acceptor/metal interface.
However, the observed response reduction for both the CuPc and C
60
would indicate that
the metal deposition directly onto the acceptor layer creates damage sites that would
effectively trap charges and reduce the photocurrent. Therefore, the role of the buffer
layer also accounts for preventing the acceptor damage and charge trapping sites besides
preventing exciton quenching.
103

-0.2 0.0 0.2 0.4 0.6
-6
-5
-4
-3
-2
-1
0
1
2
3
4
Current Density (mA/cm
2
)
Voltage (V)

(a)

400 500 600 700 800 900
0
5
10
EQE%
Wavelength, nm
(b)

Figure 4.7: (a) J-V characteristics under approximately 1 sun AM1.5G simulated
illumination and (b) spectral responsivity of CuPc(200Å)/C
60
(400Å)/Al cells (solid line) and
CuPc(200Å)/C
60
(400Å)/BCP(100Å)/Al (- ■-)
104

Device Jsc V
oc
FF ɳ%
CuPc/C
60
/Al 3.35 0.37 0.39 0.49
CuPc/C
60
/BCP/Al 5.51 0.48 0.60 1.59
Table 4.3: Parameters for the performance of solar cells with different cathodes materials
under 1 sun AM1.5G simulated illumination for CuPc(200Å)/C
60
(400Å)/Al(1000Å) and
CuPc(200Å)/C
60
(400Å)/BCP(100Å)/Al(1000Å) devices.
For the case of BCP, energetically, the wide band gap material does not have a
suitable driving force for electron transport through the LUMO. However, it has been
suggested that the metal damages the organic blocking layer upon deposition and creates
defect sites through which the dissociated electrons are collected at the cathode. It is
difficult to completely ascertain the nature and energy of these hopping sites, making a
true discussion of the energetics of OPVs difficult. Based on the understanding of the
buffer layer’s main role of blocking excitons while simultaneously transporting electrons
to the cathode, any organic material that is energetically disfavorable for exciton
dissociation at the acceptor layer and susceptible to metal damage from the cathode
should work similarly to the BCP. A variety of materials with different energy levels
have been shown to be effective buffers in the literature.
20
To further investigate the
mechanism of transport through the buffer layer, a series of wide band gap organics were
employed in CuPc/C
60
/buffer devices and their structures are depicted in Figure 4.8 (a).
All the molecules are significantly harder to reduce than C
60
according to their measured
reduction potentials, and they should effectively block exciton dissociation at the
interface. A depiction of the different energy levels of all materials to be used as buffer
layer is also depicted in Figure 4.8 (b). All of these materials have been shown to
105

transport charge in organic electronics. According to the wide band gaps, these materials
only absorb in the UV with minimal contribution to the photocurrent
106

(a)
NPt
Alq3
BCP
NPD
CuPc C60 BCP NPt Alq3 NPD
-1.17 V
-2.23 V
-2.53 V
0.38 V
1.1 V
1.36 V
0.75 V
-2.8 V
1.26 V

0.43 V
-1.29 V
-1.06 V
(b)

Figure 4.8: (a) Molecular structures of materials used as buffer layer and (b) the
corresponding reduction and oxidation potential as estimation for the LUMO and HOMO levels
repectively.
Figure 4.9 shows the J-V curves for CuPc/C
60
devices employing the different
materials as the buffer layer. Device structures are all
CuPc(400Å)/C
60
(400Å)/buffer(100Å)/Al(1000Å) for all devices with the exception of
NPD, which has reduced transport properties and even 50 Å films show an evident
resistivity that affects greatly the FF. There are significant differences in device

performance with different materials. Table 4.4 reports the device parameters for the
different molecules used as buffers. Clearly, device performance decreases with
diminished oxidation potential. There is a slight increase in V
oc
which correlates with a
drop in the Jsc and FF. The most dramatic change, however, is in the Jsc which varies
significantly across the range of oxidation potentials.
-0.2 0.0 0.2 0.4 0.6
-6
-4
-2
0
2
4
BCP (100A)
NPt (100A)
Alq3 (100A)
NPD (50A)
Current Density (mA/cm
2
)
Voltage (v)

Figure 4.9: J-V characteristics under approximately 1 sun AM1.5G simulated illumination
of devices of the form CuPc(400Å)/C
60
(400Å)/buffer/Al
Buffer J
SC
V
oc
FF η%
BCP 5.51 0.48 0.60 1.59
NPt 4.94 0.48 0.58 1.38
Alq3 4.50 0.48 0.55 1.20
NPD 3.65 0.28 0.21 0.21
Table 4.4: Illumination parameters for the performance of cells described in Figure 4.9 for 1
sun simulated illumination.
107

400 500 600 700 800
0
2
4
6
8
10
BCP (100
Å
)
NPt (100
Å
)
Alq3 (100
Å
)
NPD (50
Å
)
QE%
Wavelength, nm

Figure 4.10: Spectral responsivity of CuPc(400Å)/C
60
(400Å)/buffer/Al cells
All of these observations are consistent with a variation in charge collection
efficiency due to a decrease in charge transport with decreased oxidation potential.
Spectral response measurements for these devices show a decrease in both the CuPc and
the C
60
response, indicating that there is a uniform charge collection problem, rather than
a decrease in exciton dissociation in the C
60
(Figure 4.10). Furthermore, thickness
dependence studies on the buffer layers show similar behavior over 50-150 Å for all
WGOs, except for NPD. It is likely that the metal does diffuse into the organic in a
similar fashion to previous discussions of damage; however it is possible that discrete
damage states for charge hopping are not formed but rather a pseudo continuous pathway
similar to a porous metal structure is present. It is well known that the workfunction of a
metal will shift according to the organic it is in contact with, such that as the organic
108

buffer is varied, the metal energetics would shift according to the electronic properties of
the underlying organic. This observation is consistent with diminished charge collection
efficiency with a decrease in the oxidation potential of the organic material. It is also
likely that the higher HOMO level for as we go from BCP to NPD, would result in
exciton splitting from the C
60
molecules. Such charge separation would result in the
creation of an electron in the acceptor and a hole in the buffer layer. In that case then the
buffer layer is acting as a donor layer creating an overall “sandwich” device were the C
60
is actually acting as a trap for electrons. As more and more C
60
excitons split at both
interfaces we have a build-up of negative charges at the acceptor level, which in turn
leads to a capacitor effect that diminishes charge collection. As we go from BCP to NPD
the extension of C
60
excitons that split at the buffer/acceptor interface increases which
leads to a reduced charge collection and at the extreme, NPD, a complete shut-down of
charge flow for the forward bias.
4.4 Conclusions
The issue of charge collection at the organic/cathode interface has been addressed
and it was demonstrated that energetics are not the only parameters that influences
electron injection into the electrode. The nature and chemistry of the metal is
fundamental to understanding the organic/metal interface. Oxidation of Al creates a
barrier to charge injection and a subsequent loss of photocurrent. This deleterious
chemical reaction can be prevented by capping of the aluminum/organic interface with a
minimal (>100 Å) thickness of Ag. It has been shown that CuPc/C
60
/BCP devices, where
109

the Al/organic interface is sufficiently capped with Ag, have similar efficiencies to
devices with a pure Ag cathode. However, there are reported cases, such as when SubPc
is used as a donor material,
21
higher efficiencies are obtained when the cathode material
is Al rather than Ag. For these cases there may be an advantage to maintaining an
Al/organic interface and minimizing oxidation through Ag capping. It is also important
to note that device performance reported herein was measured under ambient conditions,
making these effects predominately relevant to research laboratory environments. It is
possible that oxidation of aluminum can also be minimized via deposition under UHV
conditions and packaging of the resulting device.
The choice of the buffer material can greatly affect the device performance. The use
of a buffer layer after the acceptor layer is highly important to ensure efficient charge
collection by diminishing the amount of charge trapping at the organic/cathode interface
by bare contact of the acceptor with the deposited metal. However, only materials that
can ensure good electron conducting path for carrier extraction, provided from a high
oxidation potential and a deep HOMO, are efficient buffer layers. Also, the electron
conducting properties have to be favorable for a good ohmic contact at the organic/metal
interface and deep HOMO is necessary to prevent negative charge build-up at the
acceptor level.

110

4.5 Chapter 4 References
1 Xue, J. G., Uchida, S., Rand, B. P. and Forrest, S. R., 4.2% efficient organic
photovoltaic cells with low series resistances. Applied Physics Letters, 2004. 84(16): p.
3013-3015.
2 Peumans, P. and Forrest, S. R., Very-high-efficiency double-heterostructure copper
phthalocyanine/C-60 photovoltaic cells. Applied Physics Letters, 2001. 79(1): p. 126-
128.
3 Chang, E. C., Chao, C. I. and Lee, R. H., Enhancing the efficiency of MEH-PPV and
PCBM based polymer solar cells via optimization of device configuration and processing
conditions. Journal of Applied Polymer Science, 2006. 101(3): p. 1919-1924.
4 Mihailetchi, V. D., Blom, P. W. M., Hummelen, J. C. and Rispens, M. T., Cathode
dependence of the open-circuit voltage of polymer : fullerene bulk heterojunction solar
cells. Journal of Applied Physics, 2003. 94(10): p. 6849-6854.
5 Mihailetchi, V. D., Koster, L. J. A. and Blom, P. W. M., Effect of metal electrodes on
the performance of polymer : fullerene bulk heterojunction solar cells. Applied Physics
Letters, 2004. 85(6): p. 970-972.
6 Lee, C. J., Pode, R. B., Han, J. I. and Moon, D. G., Ca/Ag bilayer cathode for
transparent white organic light-emitting devices. Applied Surface Science, 2007. 253(9):
p. 4249-4253.
7 Jabbour, G. E., Kippelen, B., Armstrong, N. R. and Peyghambarian, N., Aluminum
based cathode structure for enhanced electron injection in electroluminescent organic
devices. Applied Physics Letters, 1998. 73(9): p. 1185-1187.
8 Kim, S. H., Jang, J. and Lee, J. Y., Improvement in Power Efficiency in Organic
Light Emitting Diodes Through Intermediate Mg:Ag Layer in LiF/Mg:Ag/Al Cathodes.
Electrochemical and Solid-State Letters, 2007. 10(10): p. J117-J119.
9 Peumans, P., Yakimov, A. and Forrest, S. R., Small molecular weight organic thin-
film photodetectors and solar cells. Journal of Applied Physics, 2003. 93(7): p. 3693-
3723.
10 Singh, V. P., Singh, R. S., Parthasarathy, B., Aguilera, A., Anthony, J. and Payne,
M., Copper-phthalocyanine-based organic solar cells with high open-circuit voltage.
Applied Physics Letters, 2005. 86(8): p. 082106.
11 Sze, S. M., Physics of semiconductor devices (2nd edition). 1981: p. Pages: 878.
111

12 Veenstra, S. C., Heeres, A., Hadziioannou, G., Sawatzky, G. A. and Jonkman, H. T.,
On interface dipole layers between C60 and Ag or Au. Applied Physics A: Materials
Science & Processing, 2002. 75(6): p. 661-666.
13 Hirose, Y., Kahn, A., Aristov, V., Soukiassian, P., Bulovic, V. and Forrest, S. R.,
Chemistry and electronic properties of metal-organic semiconductor interfaces: Al, Ti,
In, Sn, Ag, and Au on PTCDA. Physical Review B, 1996. 54(19): p. 13748-13758.
14 Stössel, M., Staudigel, J., Steuber, F., Simmerer, J. and Winnacker, A., Impact of the
cathode metal work function on the performance of vacuum-deposited organic light
emitting-devices. Applied Physics A: Materials Science & Processing, 1999. 68(4): p.
387-390.
15 Hass, G., Heaney, J. B., Herzig, H., Osantowski, J. F. and Triolo, J. J., Reflectance
and Durability of Ag Mirrors Coated with Thin-Layers of Al2o3 Plus Reactively
Deposited Silicon-Oxide. Applied Optics, 1975. 14(11): p. 2639-2644.
16 Eastment, R. M. and Mee, C. H. B., Work Function Measurements on (100), (110)
and (111) Surfaces of Aluminum. Journal of Physics F-Metal Physics, 1973. 3(9): p.
1738-1745.
17 Grepstad, J. K., Gartland, P. O. and Slagsvold, B. J., Anisotropic work function of
clean and smooth low-index faces of aluminium. Surface Science, 1976. 57(1): p. 348-
362.
18 Chelvayohan, M. and Mee, C. H. B., Work function measurements on (110), (100)
and (111) surfaces of silver. Journal of Physics C: Solid State Physics, 1982. 15(10): p.
2305-2312.
19 Kampen, T. U., Das, A., Park, S., Hoyer, W. and Zahn, D. R. T., Relation between
morphology and work function of metals deposited on organic substrates. Applied
Surface Science, 2004. 234(1-4): p. 333-340.
20. Song, Q.L., F.Y. Li, H. Yang, H.R. Wu, X.Z. Wang, W. Zhou, J.M. Zhao, X.M.
Ding, C.H. Huang, and X.Y. Hou, Small-molecule organic solar cells with improved
stability. Chemical Physics Letters, 2005. 416(1-3): p. 42-46.
21 Mutolo, K. L., Mayo, E. I., Rand, B. P., Forrest, S. R. and Thompson, M. E.,
Enhanced open-circuit voltage in subphthalocyanine/C
60
organic photovoltaic cells.
Journal of the American Chemical Society, 2006. 128(25): p. 8108-8109.

112

Chapter 5. Exciplex quenching of a luminescent
cyclometallated Platinum complex by extremely poor Lewis
bases.
5.1 Introduction
The photophysical properties of luminescent platinum complexes have received a
great deal of recent attention, particularly for their application in the field of
phosphorescence organic light-emitting diodes (OLEDs).
1, 2
Another area of interest that
exploits the emissive properties of these materials is their use as luminescent probes.
3

Notably, the luminescent efficiency of the complexes in solution often shows a strong
dependence on the identity of the solvent.
4, 5
The variability comes about because the
square planar coordination geometry of the platinum(II) complexes exposes the metal
center to the environment and opens up deactivation pathways for the luminescent
excited state. Common luminescent quenching processes include the formation of
excimers or exciplexes.
6, 7
Exciplex quenching has been reported for several d
8
, d
9
and
d
10
systems,
8-14
and strongly nucleophilic cationic platinum(II) complexes have been
extensively studied by McMillin’s group,
12, 13, 15
Exciplexes form through binding of
donor molecules to the platinum complex in its excited state, resulting in a 5-coordinate
encounter complex that nonradiatively decays. Formation of the 5-coordinate adduct is
governed by interactions between the metal complex in its excited state and the donor
molecule, and depends on the electrophilic strength of the metallic core and the
113

nucleophilicity of the donor. Quenching rates for exciplex formation for different
solvents were shown to somewhat depend on the donor number (DN) of the solvent
molecules, a commonly used nucleophilicity measure,
16
however, it does not distinguish
differences among very weak Lewis bases (DN < 0.1). The Gutmann donor number
(DN) of a quencher is defined as the negative ΔH values for the 1:1 adduct formation
between antimony pentachloride and electron pair solvents (D) in dilute solution in 1,2-
dichloroethane.
16, 17
Exciplex formation with different solvent donors was reported to
occur for relatively strong Lewis bases, with acetonitrile the weakest to show quenching.
In this Chapter a new strong emitter from the family of cyclometallated platinum
complexes, i.e (2-phenyl-5-nitropyridine)Pt(acetylacetonate) (NPt) is introduced. The
original purpose for the production of this complex was to be used as a “buffer” layer in
double heterojunction organic photovoltaics. Several devices were successfully prepared
using NPt as described in Chapter 4, but his photophysical properties were interesting
enough such that they deserved their own study and corresponding Chapter in this thesis.
The NPt complex is found to be a powerful Lewis acid in its excited state and the
first to show a strong emission at room temperature that is effectively quenched upon
exposure to even extremely weak Lewis bases (very small DN), such as aromatic
compounds like toluene. The photophysical properties of this novel complex will be
demonstrated here, together with a thorough study on the emission quenching by several
Lewis bases. Relative nucleophilicities for non-Lewis bases with small and similar donor
numbers will also be demonstrated.
114

5.2 Experimental
5.2.1 Synthesis
Synthesis was performed according to a method previously reported.
18, 19
All
materials were purchased from Sigma-Aldrich and used without further purification. All
synthetic procedures were carried out in inert gas atmosphere. The 2-phenyl-5-
nitropyridine ligand was prepared by Suzuki coupling reaction with commercially
available phenylboronic acid and 2-bromo-5-nitropyridine. The mononuclear
cyclometalated Pt(II) chloro precursor complex
19
was prepared by heating K
2
PtCl
4
salt
with 2-2.5 equiv of cyclometalating ligand in a 3:1 mixture of 2-ethoxyethanol (Aldrich)
and water to 80 °C for 16 h. The chloro complex was isolated in water and subsequently
reacted with 3 equiv of the chelating diketone derivative and 10 equiv of Na
2
CO
3
in 2-
ethoxyethanol at 100 °C for 16 h. The solvent was removed under reduced pressure, and
the compound was purified by flash chromatography using dichloromethane. The product
was further purified by sublimation (yield = 23%). Crystals were grown by slow
diffusion of hexane into a concentrated solution of NPt into dichloroethane.
1
H NMR (250 MHz, CDCl3), ppm: 9.91 (s, 1H), 8.55 (d, 1H, J = 8.94 Hz), 7.67 (m,
2H), 7.49 (d, 1H, J = 7.21 Hz), 7.31 (d, 1H, J = 8.53 Hz), 7.14 (m, 2H), 5.52 (s, 1H), 2.04
(s, 3H), 2.09 (s, 3H). Anal. for C
16
H
14
N
2
O
4
Pt: found C 39.47, H 2.85, N 5.35, calcd C
38.95, H 2.86, N 5.68.

115

5.2.2 Quenching measurements
Luminescence quenching studies were conducted on dilute solutions (1 × 10
–5
M) of
NPt in cyclohexane that were degassed for ten minutes with N
2
. The quencher
concentration was determined volumetrically by known additions of solvent volumes.
Quantum yield and lifetimes were measured after each addition of quencher following
degassing. All solvents were purchased from Sigma-Aldrich and used as received.
5.2.3 Instrumentation
UV-vis spectra were recorded using an Agilent 8453 UV-Vis spectrometer.
Photoluminescent spectra were measured using a Photon Technology International
QuantaMaster C-60SE spectrofluorometer. Emission lifetime measurements were
performed using time-resolved single photon counting on an IBH photon timing
instrument connected to an IBH model TBX-04 PMT detector. Quantum efficiency
measurements were carried out using a calibrated integrating sphere equipped with a
xenon lamp and Hamamatsu Model C10027 photonic multi-channel analyzer. The
quantum efficiency data was processed using the U6039-05 software package provided
by Hamamatsu. The error in the quantum efficiency measurements is + 5%. Cyclic
voltammetry and differential pulse voltammetry were performed using an EG&G
potentiostat/galvanostat model 28. Anhydrous dichloromethane was used as the solvent
under a nitrogen atmosphere, and 0.1 M tetra(n-butyl)ammonium hexafluorophosphate
was used as the supporting electrolyte. A Pt wire acted as the counter electrode, Ag wire
was used as the pseudo reference electrode, and the working electrode was glassy carbon.
116

The redox potentials are based on values measured from differential pulse voltammetry
and reported relative to an internal ferrocenium/ferrocene (Cp
2
Fe
+
/Cp
2
Fe) reference.
Computational calculations were performed by Dr. Azad Hassan.
20

5.2.4 X-ray Crystallography
Diffraction data for NPt was collected at room temperature (T = 23º C) on a Bruker
SMART APEX CCD diffractometer with graphite-monochromated Mo K radiation (λ =
0.71073 Å). The cell parameters for the Pt complex were obtained from the least-squares
refinement of the spots (from 60 collected frames) using the SMART program. A
hemisphere of the crystal data was collected up to a resolution of 0.75 Å, and the
intensity data was processed using the Saint Plus program. All calculations for structure
determination were carried out using the SHELXTL package (version 5.1). Initial atomic
positions were located by Patterson methods using XS, and the structure was refined by
least-squares methods using SHELX with 7063 independent reflections and within the
range of Φ = 1.61–27.51º (completeness 93.8%). Absorption corrections were applied by
using SADABS. Calculated hydrogen positions were input and refined in a riding
manner along with the attached carbons.
5.3 Results and Discussion
5.3.1 X-Ray diffraction studies
Analysis by x-ray diffraction revealed a square planar structure for the complex. All
bond lengths and angles are characteristic of related cyclometalated Pt(β-diketonato)
117

complexes.
21-24
The nitro substituent has a planar nitrogen (average angles; C–N–O =
118°, O–N–O = 125°) and is nearly coplanar with respect to the pyridyl ring; dihedral C–
C–N–O angles approach zero (average = 2.92°) (see ORTEP representations, Figures 5.1
and 5.2). Two unique molecules in the asymmetric unit pack as head-to-tail dimers that
have Pt-Pt separations of 4.476 and 3.465 Å, the latter value close enough to support a
metal-metal interaction.
Summary of structure determination
Empirical formula C16 H14 Cl N2 O4 Pt
Formula weight 528.83
Temperature 110(2) K
Wavelength 0.71073 Å
Crystal system Triclinic
Space group P-1
Unit cell dimensions a = 11.7451(7) Å α= 88.8210(10)°.
b = 12.1959(7) Å β= 86.7640(10)°.
c = 12.6839(7) Å γ= 64.3710(10)°.
Volume 1635.49(16) Å3
Z 4
Density (calculated) 2.148 Mg/m3
Absorption coefficient 8.766 mm-1
F(000) 1004
Crystal size 0.25 x 0.3 x 0.09 mm3
118

Theta range for data collection 1.61 to 27.51°.
Index ranges -15<=h<=14, -15<=k<=8, -16<=l<=15
Reflections collected 10116
Independent reflections 7063 [R(int) = 0.0180]
Completeness to theta = 27.51° 93.8 %
Absorption correction 0.651
Refinement method Full-matrix least-squares on F2
Data / restraints / parameters 7063 / 0 / 428
Goodness-of-fit on F2 1.040
Final R indices [I>2sigma(I)] R1 = 0.0293, wR2 = 0.0767
R indices (all data) R1 = 0.0329, wR2 = 0.0786
Largest diff. peak and hole 4.330 and -1.063 e.Å-3
119

Figure 5.1: ORTEP diagram for one unique NPt molecule from the asymmetric unit cell.

Figure 5.2: ORTEP diagram showing top view of NPt dimers with short (3.465 Å) (a), and
long (4.476 Å) (b) Pt-Pt separations.
120

5.3.2 Electrochemical studies
Electrochemical response of the NPt dissolved in dichloromethane is presented in
Figure 5.3. The introduction of the potent electron-withdrawing nitro group on the 2-
phenylpyridyl (ppy) ligand drastically lowers the reversible reduction potential to -1.43 V
vs. Fc
+
/Fc relative to the unsubstituted analog, (ppyPt(acac) (E
1/2
red
= -2.39 V vs
Fc
+
/Fc).
18
This difference of -1 volt comes about because of the electron withdrawing
effect of the addition of the nitro group. Nitro substitution of the ppy ligand creates a
relative positive charge density and the addition of an electron is favored compared to the
unsubstituted analog. More about this effect will be discussed in section 5.3.4 together
with a description of the calculated energetics obtained by DFT theory. No appreciable
oxidation was found by cyclic voltammetry in dichloromethane.
-2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5
-4
-2
0
2
4
Current (x10
4
), A
Voltage, V (vs. Fc
+
/Fc)

Figure 5.3: Cyclic voltammogram of NPt recorded in dichloromethane with 0.1 M TBAPF.
121

5.3.3 Optical properties
Absorption spectra in cyclohexane shows a characteristic
1
MLCT absorption band at
λ
max
= 495 nm that exhibits a negative solvatochromic effect by shifting to higher
energies in solvents with increasing polarity (λ
max
= 451 nm in acetonitrile), as seen from
Figure 5.4(a). A negative solvatochromism corresponds to a hypsochromic shift, a
change of spectral band position to a shorter wavelength (higher energy), with increasing
solvent polarity. Solvent dependent absorption is typical for this family of complexes
and is also shown for the unsubstituted ppyPtacac in Figure 5.4(b). Extinction
coefficients were measured in several solvents and the wavelength specific data is
presented in Table 5.1.
Solvent
Wavelength
(nm)
Extinction coefficient
(M
-1
cm
-1
)
Acetonitrile
275 2.85 x 10
4

450 4.71 x 10
3

Cyclohexane
343 1.88 x 10
3

492 5.76 x 10
3

Toluene
329 1.60 x 10
4

472 4.86 x 10
3

Table 5.1: NPt extinction coefficients for different solvents
122

350 400 450 500 550
0.0
0.5
1.0
Cyclohexane
Acetonitrile
2-MeTHF
Toluene
Relative Absorbance (a.u.)
Wavelength (nm)
300 350 400 450
0.0
0.5
1.0
1.5
Relative Absorbance (a.u.)
Wavelength (nm)
Acetonitrile
Cyclohexane
(a)
(b)

Figure 5.4: Absorption spectra showing the negative solvatochromic effect for NPt (a) and
ppyPt(acac) (b).
Emission spectra recorded at room temperature in N
2
-deaerated cyclohexane display
an intense, structured phosphorescence (λ
max
= 550 nm, τ = 6 μs, Φ = 0.57 ± 0.05)
(Figure 5.5). When dissolved in solvents other than nonpolar cyclohexane or hexanes,
emission from NPt is completely quenched at room temperature. The emission is
123

recovered if the sample is cooled to 77 K with a characteristic λ
max
= 546 nm and a
lifetime of τ = 10.2 μs. When NPt is dissolved in a rigid matrix, such as polystyrene,
emission is also visible at room temperature (λ
max
= 554 nm, τ = 4.8 μs) (Figure 5.5,
inset). This freezing effect is only shown for toluene in Figure 5.5 (inset) but was also
observed for other solvents like 2-MeTHF as seen in Figure 5.6(b).
400 500 600 700
0
500
1000
1500
Absorption
ε, M
-1
cm
-1
Wavelength, nm
0.0
0.5
1.0
1.5
Intensity, a.u.
Emission
400 500 600
Polystyrene
Toluene
Intensity, a.u.
Wavelength, nm

Figure 5.5: Absorption and emission (λ
exc
= 450 nm) of NPt in cyclohexane at room
temperature. Inset: Excitation and emission in toluene (77K) and polystyrene (RT).
As observed in Figure 5.6, the emission red-shifts and becomes featureless (λ
max
=
620 nm, τ = 13 μs) in 3-methylpentane at 77 K, and is similar to that obtained from the
neat solid at 77 K (λ
max
= 612 nm, τ = 13 μs). 3-methylpentane is used instead of
cyclohexane because of its ability to form glasses at 77 K.
124
(a)

300 400 500 600 700 800
0.0
0.2
0.4
0.6
0.8
1.0

Intensity (a.u.)
Wavelength (nm)
exc in 3MP
em in 3MP
neat, exc
neat, em

300 400 500 600 700
0.0
0.2
0.4
0.6
0.8
1.0
τ = 10 μs
Intensity (a.u.)
Wavelength (nm)
exc @ 77K
em @ 77K
(b)

Figure 5.6: Excitation and emission spectra at 77K in 3-methylpentane (3MP) and for the
neat solid (a), and at 77 K in 2-methyltetrahydrofuran (b).
Emission at room temperature is assigned to
3
MLCT-perturbed, ligand-centered
state, whereas the spectrum at 77 K likely originates from a metal-metal-to-ligand triplet
125

state on Pt dimers, analogous to those seen in the crystal lattice, that are formed on
cooling.
5.3.4 Theoretical analysis
The electrochemical and spectroscopic data are in agreement with results from DFT
calculations. The energies of the HOMO (-5.80 eV) and, especially the LUMO (-2.89
eV), which is localized primarily on the ligand (Figure 5.7), are significantly lower than
that of ppyPt(acac) (-5.65 eV and -1.89 eV, respectively). The strongly reduced
reduction potential is then a consequence of the LUMO being localized on the ppy
ligand. The substitution of the strong electron withdrawing nitro group into the ppy
effectively creates a positive charge density on the LUMO favoring the addition of an
electron by reduction. A TD-DFT calculation indicates that the lowest singlet and triplet
absorption transitions correspond to simple HOMO-LUMO transitions (S
0
–S
1
= 565 nm,
f = 0.0701; S
0
–T
1
= 625 nm). The theoretical singlet-triplet energy matches exactly the
one found experimentally, Δ(S
1
-T
1
) = 0.25 eV.
126

127

HOMO LUMO
Figure 5.7: HOMO and LUMO electronic density
5.3.5 Emission quenching by poor Lewis bases
As mentioned earlier in section 5.3.3, emission is strongly quenched by common
solvents other than hydrocarbons, like cyclohexane or hexane. In order to determine and
quantify the quenching phenomenon, Stern–Volmer analysis was performed for a variety
of compounds using cyclohexane as the inert solvent. Stern-Volmer analysis studies the
variation of quantum yield (Φ) or emission lifetimes with the concentration of the
quencher molecule. In the simplest case, a plot of Φ
o
/Φ vs. concentration of quencher,
[Q], is linear and obeys the equation:
[] Q 1
o
SV
K + =
Φ
Φ
5.1
where Ksv is the Stern–Volmer constant and Φ
o
is the quantum yield in absence of
quencher. In the case of dynamic quenching the constant Ksv is the product of the true
quenching constant (k
q
) and the excited state lifetime (τ
o
) in the absence of quencher. k
q
is

the bimolecular reaction rate constant for the elementary reaction of the excited state with
the particular quencher Q. Equation 5.1 can therefore be replaced by the expression:
[] Q 1
o q
o
τ k + =
Φ
Φ
5.2
The same behavior should be expected for the analysis of emission lifetime
quenching. Both studies were performed in this work but first-order quenching rates (k
q
)
for quantum yield analysis are presented in Table 5.2. Both lifetime and quantum yield
quenching measurements are closely related with similar k
q
values.
Molecules with basic lone pairs, such as acetonitrile, 2-MeTHF, nitrobenzene and
DMSO, show high quenching rates (k
q
= 10
7
to 10
8
M
-1
s
-1
). Weaker Lewis bases give
smaller values of k
q
than stronger bases, but even poor donors are sufficiently
nucleophilic to completely suppress emission when used as a solvent. Most strikingly,
molecules considered to be unreactive and non-coordinating also effectively quench the
emission. For example, dichloromethane, often used as an inert solvent for exciplex
quenching studies, also quenches the NPt phosphorescence, albeit with the lowest value
reported here (k
q
= 2.20 x10
5
M
-1
s
-1
). This deleterious property may account for the low
quantum efficiency observed for other Pt complexes, particularly those with comparable
reduction potentials,
25-28
when measured in this solvent. In contrast, when ppyPtacac is
exposed to these solvents, there is no observable quenching and emission intensity was
not further affected. For toluene, 2-MeTHF and dichloromethane, the lifetimes at room
temperature were τ = 2.7, 2.7 and 3.0 μs, respectively for the ppyPtacac case.
128

Solvent k
q
(M
-1
s
-1
) DN
dichloromethane 2.2 x10
5
N/A
fluorobenzene 4.9 x10
5
3
chlorobenzene 6.3 x10
5
3.3
benzene 7.3 x10
5
0.1
1,3-difluorobenzene 1.2 x10
6
N/A
toluene 1.8 x10
6
0.1
bromobenzene 1.8 x10
6
3
acetonitrile 9.4 x10
6
14.1
1,3-dichlorobenzene 1.2 x10
7
2
naphthalene 3.5 x10
7
N/A
2-methylTHF 7.7 x10
7
18
nitrobenzene 1.6 x10
8
4.4
DMSO 7.0 x10
8
29.8
Table 5.2: Stern-Volmer constants and Gutmann donor numbers (DN).
16,17
The high electrophilicity of the NPt excited state leads to its propensity to form
exciplexes with Lewis bases. Adduct formation is possible due to the open accessibility
and the increased acidity of the metal Pt core imparted by the nitro substituent on the 2-
phenylpyridyl ligand. The enhanced acidity of the NPt can be readily visualized by
electronic potential density plots obtained by DFT calculation of the triplet state for both
substituted and unsubstituted analogs as depicted in Figure 5.8. The red-colored area in
ppyPt(acac) defines a region of high electron density at the metal center, while the
corresponding green area in NPt depicts a diminished electronic density and is the likely
site for interaction with a Lewis base molecule.
129

130

Figure 5.8: Potential surfa ited state of NPt (left) and
ppyPt(acac) (right).
A careful observation of th quenching rates in Table 5.2
reveal that they do no ophiles, it has been
necessarily follow the literature donor
numbers.
10
However, the exten hat dependent on the
relative donor strength, e.g. the k
q
of DMSO > acetonitrile. The strong Lewis acidity of
olecule allows for recognition of slightly different basicities of very
wea
ces of electron density for the triplet exc
e relation of the DN and the
t correlate accurately. Especially for poor nucle
reported before that quenching rates do not
t of exciplex formation is somew
the excited NPt m
k donor molecules. This ability to distinguish small variations in nucleophilicity is
demonstrated by the differing quenching rates observed for aromatic molecules. For
example, benzene and toluene, both very poor nucleophiles (DN = 0.1), have dissimilar
quenching rates, with the more electron-rich toluene being favored by over a factor of
two (k
q
= 7.28 x 10
5
and 1.81 x 10
6
M
-1
s
-1
, respectively). Increasing the size of the
π-system to naphthalene results in a 50-fold increase in the quenching rate compared to
benzene. When the aromatic ring is functionalized with electron-withdrawing groups, as

131

pport the latter process as the principal quenching
mec
in nitrobenzene, substantial quenching of emission is still observed even though the π-
basicity is strongly diminished relative to benzene or toluene. In this case, however,
binding with the excited NPt complex likely occurs through the more basic nitro group
rather than the aromatic π-system.
Halogenated benzenes also quench NPt emission, albeit less effectively than
nitrobenzene. The value of k
q
increases upon multiple halogen substitution or as the
halogen basicity is increased down the period from F to Br (Table 5.2 and Figure 5.9).
There are two possible quenching modes for haloaromatic compounds, through
interaction with the lone pair electrons of the halogen or by coordination to the aromatic
π-system. Several observations su
hanism. The quenching rate is higher for ClC
6
H
5
over CH
2
Cl
2
, even though the latter
has two Cl groups, which suggests that the Cl is a less efficient quencher than the
aromatic π-system. An argument could be made that the aromatic π-electrons increase
the basicity of the halogen, although this effect would be somewhat attenuated by
multiple Cl substitution. However, rather than seeing a diminution of the quenching
efficiency on moving to 1,3-dichlorobenzene, we observe a 20-fold increase in k
q
,
consistent with enhanced basicity of the aromatic π-system. Moreover, the effect of
multiple fluorine substitution is markedly less than that for chlorine, consistent with the
weak π-donating ability of fluorine.

132
Figure 5.9: Stern–Volmer plots for quenching of NPt by halogenated benzenes in
cyclohexane.
Quenching of the phosphorescence by exciplex formation can be described using the
simplistic mechanism shown in Scheme 5.1
Scheme 1: Mechanism for exciplex formation
For very weak Lewis bases such as benzene, toluene and the mono-halogenated
benzenes, Stern–Volmer analysis exhibits non-first order behavior, since the data does
not ts are better fitted by a second order

0.0 0.5 1.0
0
3
6
9
12 1,3-dichlorobenzene
bromobenzene
benzene
chlorobenzene
fluorobenzene
I
o
/I
Quencher Concentration (M)

Q
Pt*
Q Pt*
k
d
q
k
-d
k
Q Pt*
Q Pt

show a strong linear correlation and plo

133
poly

Figure 5.10: Second-order fit to Stern-Volmer data using toluene as quencher.
5.4 Conclusions
In conclusion, a phosphorescent Pt complex has been prepared that undergoes
efficient exciplex quenching with extrem
nomial. Figure 5.10 depicts Stern-Volmer analysis for the quencher toluene fitted by
both a first order and second order polynomial. This behavior has been reported earlier
and was suggested to be a result of a more complicated mechanism for complex
formation in the excited state.
7, 13
For the weakest Lewis-base solvents, the equilibrium
constant for the formation of the 5-membered adduct (K=k
d
/k
-d
, Scheme 5.1) may be
extremely low (poor adduct formation) resulting in a dynamic equilibrium that disfavors
product and reduces the rate of nonradiative decay. Further studies will be necessary in
order to clarify this more complicated quenching mechanism.
0
1
2
3
4
5
0 0.1 0.2 0.3 0.4
I
o
/I
Toluene Concentration, M
ely poor donor bases. The strong Lewis acidity

134
state promotes formation of 5-coordinate adducts that decay
non

of the NPt excited
radiatively. The high sensitivity of NPt has enabled us to quantify a heretofore
undocumented quenching ability of very weak Lewis bases such as dichloromethane.
The differing nucleophilicities of aromatic compounds could be probed by Stern-Volmer
analysis, as well as specific areas of electronic availability within the quenching
molecule. Exciplex quenching was demonstrated to occur by coordination to the basic
π-cloud of aromatic compounds. Modification of the phenyl moiety in NPt should alter
the electrophilicity of the metal center and lead to the creation of a variety of other Pt
derivatives that can serve as finely tuned probes for the Lewis base strength of aromatic
molecules.

5.5 Chapter 5 References
1. Williams, J., Photochemistry and Photophysics of Coordination Compounds:
Platinum, in Photochemistry and Photophysics of Coordination Compounds II. 2007. p.
205-268.
2. Xiang, H.-F., S.-W. Lai, P.T. Lai, and C.-M. Che, Phosphorescent Platinum(II)
Materials for OLED Applications, in Highly Efficient OLEDs with Phosphorescent
Materials, H. Yersin, Editor. 2007, Wiley-VCH: Weinheim. p. 259.
3. Wong, K.M.-C. and V.W.-W. Yam, Luminescence platinum(II) terpyridyl
complexes--From fundamental studies to sensory functions. Coordination Chemistry
Reviews, 2007. 251(17-20): p. 2477-2488.
4. Chan, C.W., L.K. Cheng, and C.M. Che, Luminescent Donor-Acceptor Platinum(Ii)
Complexes. Coordination Chemistry Reviews, 1994. 132: p. 87-97.
5. Lai, S.-W. and C.-M. Che, Luminescent Cyclometalated Diimine Platinum(II)
Complexes: Photophysical Studies and Applications, in Topics in Current Chemistry.
2004. p. 27-63.
6. Ma, B., P.I. Djurovich, and M.E. Thompson, Excimer and electron transfer
quenching studies of a cyclometalated platinum complex. Coordination Chemistry
Reviews, 2005. 249(13-14): p. 1501-1510.
7. Fleeman, W.L. and W.B. Connick, Self-quenching and cross-quenching reactions of
excited platinum(II) diimine complexes. Comments on Inorganic Chemistry, 2002. 23(3):
p. 205-230.
8. Castellano, F.N., I.E. Pomestchenko, E. Shikhova, F. Hua, M.L. Muro, and N.
Rajapakse, Photophysics in bipyridyl and terpyridyl platinum(II) acetylides. Coordination
Chemistry Reviews, 2006. 250(13-14): p. 1819-1828.
9. Horváth, A. and K.L. Stevenson, Transition metal complex exciplexes. Coordination
Chemistry Reviews, 1996. 153: p. 57-82.
10. Stacy, E.M. and D.R. McMillin, Inorganic Exciplexes Revealed by Temperature-
Dependent Quenching Studies. Inorganic Chemistry, 1990. 29(3): p. 393-396.
11. Liu, F., K.L. Cunningham, W. Uphues, G.W. Fink, J. Schmolt, and D.R. McMillin,
Luminescence Quenching of Copper(Ii) Porphyrins with Lewis-Bases. Inorganic
Chemistry, 1995. 34(8): p. 2015-2018.
135

12. Aldridge, T.K., E.M. Stacy, and D.R. McMillin, Studies of the Room-Temperature
Absorption and Emission-Spectra of [Pt(Trpy)X]+ Systems. Inorganic Chemistry, 1994.
33(4): p. 722-727.
13. Crites Tears, D.K. and D.R. McMillin, Exciplex quenching of photoexcited
platinum(II) terpyridines: influence of the orbital parentage. Coordination Chemistry
Reviews, 2001. 211(1): p. 195-205.
14. McMillin, D.R., J.R. Kirchhoff, and K.V. Goodwin, Exciplex quenching of photo-
excitd copper complexes. Coordination Chemistry Reviews, 1985. 64: p. 83-92.
15. McGuire, R., M.H. Wilson, J.J. Nash, P.E. Fanwick, and D.R. McMillin, Lewis acid
and base sensing by platinum(II) polypyridines. Inorganic Chemistry, 2008. 47(8): p.
2946-2948.
16. Gutmann, V., Ion-Pairing and Outer Sphere Effect. Chimia, 1977. 31(1): p. 1-7.
17. Montalti, M., A. Credi, L. Prodi, and M.T. Gandolfi, Handbook of Photochemistry.
Third Edition ed, ed. C. Press. 2006. 650.
18. Brooks, J., Y. Babayan, S. Lamansky, P.I. Djurovich, I. Tsyba, R. Bau, and M.E.
Thompson, Synthesis and characterization of phosphorescent cyclometalated platinum
complexes. Inorganic Chemistry, 2002. 41(12): p. 3055-3066.
19. Cho, J.-Y., K.Y. Suponitsky, J. Li, T.V. Timofeeva, S. Barlow, and S.R. Marder,
Cyclometalated platinum complexes: High-yield synthesis, characterization, and a
crystal structure. Journal of Organometallic Chemistry, 2005. 690(17): p. 4090-4093.
20. Hassan, A., Theoretical, experimental, device fabrication and degradation studies of
materials for optoelectronic devices., in Chemistry. 2007, University of Southern
California: Los Angeles.
21. Ma, B., P.I. Djurovich, M. Yousufuddin, R. Bau, and M.E. Thompson,
Phosphorescent Platinum Dyads with Cyclometalated Ligands: Synthesis,
Characterization, and Photophysical Studies†. The Journal of Physical
Chemistry C, 2008. 112(21): p. 8022-8031.
22. Cho, J.Y., B. Domercq, S. Barlow, K.Y. Suponitsky, J. Li, T.V. Timofeeva, S.C.
Jones, L.E. Hayden, A. Kimyonok, C.R. South, M. Weck, B. Kippelen, and S.R. Marder,
Synthesis and characterization of polymerizable phosphorescent platinum(II) complexes
for solution-processible organic light-emitting diodes. Organometallics, 2007. 26: p.
4816-4829.
23. Yin, B., F. Niemeyer, J.A.G. Williams, J. Jiang, A. Boucekkine, L. Toupet, H.
LeBozec, and V. Guerchais, Synthesis, Structure, and Photophysical Properties of
136

Luminescent Platinum(II) Complexes Containing Cyclometalated 4-Styryl-Functionalized
2-Phenylpyridine Ligands. Inorg. Chem., 2006. 45(21): p. 8584-8596.
24. Ionkin, A.S., W.J. Marshall, and Y. Wang, Syntheses, structural characterization,
and first electroluminescent properties of mono-cyclometalated platinum(II) complexes
with greater than classical pi-pi stacking and Pt-Pt distances. Organometallics, 2005.
24(4): p. 619-627.
25. Cheung, T.C., K.K. Cheung, S.M. Peng, and C.M. Che, Photoluminescent
cyclometallated diplatinum(II,II) complexes: Photophysical properties and crystal
structures of [PtL(PPh(3))]ClO4 and [Pt(2)L2(mu-dppm)][ClO4](2) (HL=6-phenyl-2,2'-
bipyridine, dppm=Ph(2)PCH(2)PPh(2)). Journal of the Chemical Society-Dalton
Transactions, 1996(8): p. 1645-1651.
26. Cummings, S.D. and R. Eisenberg, Tuning the excited-state properties of
platinum(II) diimine dithiolate complexes. Journal of the American Chemical Society,
1996. 118(8): p. 1949-1960.
27. Michalec, J.F., S.A. Bejune, D.G. Cuttell, G.C. Summerton, J.A. Gertenbach, J.S.
Field, R.J. Haines, and D.R. McMillin, Long-lived emissions from 4 '-substituted
Pt(trpy)Cl+ complexes bearing aryl groups. Influence of orbital parentage. Inorganic
Chemistry, 2001. 40(9): p. 2193-2200.
28. Clark, M.L., S. Diring, P. Retailleau, D.R. McMillin, and R. Ziessel, Spectroscopic
properties of orthometalated platinum(II) bipyridine complexes containing various
ethynylaryl groups. Chemistry-a European Journal, 2008. 14(24): p. 7168-7179.

137

Chapter 6. Experimental techniques for the fabrication and
testing of small molecule organic photovoltaics
6.1 Device fabrication
Organic solar cells are fabricated via thermal evaporation of the corresponding
organic materials and metal electrodes. Materials sublimation is performed in a vacuum
chamber with background pressures in the order of 10
-6
torr. The organic layers are
deposited on previously cleaned ITO that is introduced in the vacuum chamber and later
evacuated. Parameters to be controlled are deposition rates and thicknesses. After
deposition of the organic layers, the vacuum is broken and the substrates are momentarily
removed for cathode mask placing. Cathodes are evaporated through a shadow mask that
determines the area of the device. The chamber is further evacuated one more time and
the metal is thermally evaporated and deposited on top of the organic layers. Devices are
now ready to be tested so vacuum is once again broken, the substrates extracted and
subject to testing. A more detailed explanation of all preparation steps are presented
next.
6.1.1 ITO cleaning
The transparent conducting electrode consists of a layer of indium tin oxide, ITO,
deposited on a glass substrate. It needs previous cleaning before deposition of the
organic layers to ensure good organic/anode interface formation quality, to prevent
contact resistances and the presence of dust particles that would create electrical shorts.
138

150nm ITO coated glass substrates were acquired from Thin Film Devices, Inc. with
sheet resistances of R=20±5Ω/ . They were cut to the desired substrate size (1-1.5 cm
2
)
when needed and rubbed with a solution of non-ionic Tergitol detergent to remove any
grease and dust from handling. Substrates are later boiled in tetrachloroethylene, acetone
and methanol in that order for 5 minutes each. Solvents are purchased from Sigma-
Aldrich. Later, ITO substrates are N
2
dried and placed in a UV-Ozone cleaner for 10
minutes. UV-O
3
treatment removes all organic residues by oxidation that may remain on
the surface of the ITO but it also modifies the surface workfunction that will enhance the
electrical contact with the organic layer and allow for good charge collection. The ITOs
are then placed on a substrate holder and introduced in the chamber for deposition.
6.1.2 Vacuum deposition and materials
Organic and metal layers are prepared by thermal deposition on a stainless-steel
vacuum chamber as observed in Figure 6.1.

Chamber
Roughing
Pump
Cryo Pump
Rough Valve
Foreline Valve
Hi Vac
Valve
Vent Valve
(N2 inlet)
Turbo Vent
Valve
(N2 inlet)
Figure 6.1: Image of vacuum chamber and schematics of system.
139

High vacuum is achieved by a cryopump (Cryo-torr® 8.0 from Brooks Automation,
Inc.) attached to the stainless steel body. Organic materials are placed in specially
designed tungsten boats requested from R. D. Mathis, Inc. with dimensions adapted from
the commercially available version shown in Figure 6.2 (a). Aluminum pellets are placed
in tungsten basket boats (Figure 6.2(b)) and Ag in tungsten boats with dimple container
for the material (Figure 6.2 (c)).
(c)
(a)
(b)

Figure 6.2: Evaporation sources used for the different materials. Source:
www.rdmathis.com
Organic materials for evaporation must present a high purity and they must be
thermally stable to ensure proper sublimation inside the chamber. Previous sublimation
of organic materials is always required to ensure material quality. Sublimation
purification by thermal gradient is performed previous to device fabrication in an
independent sublimation tube connected to a turbo pump for achieving low pressures.
The glass tube is inserted in a heating furnace with independent temperature areas that
allow gradient purification. Metal cathodes are used as received.
Control of the evaporation is achieved by a crystal monitor placed inside the
chamber close to the substrate location. The crystal monitor measures the amount of
material that is being depositing on the crystal surface. Material dependent parameters
must be input in the crystal monitor programming mode such that it can then calculate the
140

rate and thickness of material deposited. These parameters include density (d), tooling
factor (TF) and Z-ratio. The density is a material specific parameter and is determined by
preparation of a thick film with an arbitrary density. The thickness of this film is
measured by ellipsometry and then the correct density can be back-calculated using the
read thickness (d
real
x Thickness
real
=d
arb
x Thickness
arb
). The TF is a measure of the
geometry of the chamber and the position of the substrates and is obtained by calibration
of the chamber. This calibration is performed one time and TF values can be applied for
all materials. TF is also dependent on the evaporation boat design. Z-ratio is dependent
on the type of material and is fixed at 1 for organics.
Metal electrodes are deposited onto the organic layers through a shadow mask. This
mask is made from a thin stainless-steel sheet, containing 15 holes of a diameter of 1mm
that determines the active device area. These were later measured with an optical
microscope for accuracy in determination of the real device area for the calculation of the
current density as current/device area. An image of a typical substrate with a set of
devices is shown in Figure 6.3.
141

Exposed
ITO
surface
1cm
Device
Figure 6.3: Typical substrate with organic photoactive layers and cathode metal to make 15
devices with the same structure.

6.2 Device testing
6.2.1 Dark and simulated white light J-V characteristics
Current vs. voltage measurements were performed in an in-house designed testing
station. Voltage dependent current is measured using a Keithley 2420 Sourcemeter.
Electrical contact to devices are possible by the use of microelectrodes that approach the
anode via the exposed ITO and to the cathode by a gold thin wire that is carefully placed
on top of the metal circular area. Illumination is performed by deflecting the light 90°
with a dielectric mirror. Solar simulator consists of a 300W Xe lamp inside a Newport-
Oriel Housing with the corresponding power supply. An AM1.5G filter is used to
simulate the solar spectrum. An image of the testing area used for white illumination
testing is presented in Figure 6.4.
142

Microelectrode probes 300W Xe lamp housing
Substrate holder
90° deflecting mirror

Figure 6.4: Simulated AM1.5G illumination J-V testing station
6.2.2 Spectral responsivity
Spectral response is measured using an in-house designed monochromatic
wavelength dependent optical system. All components were obtained from Newport-
Oriel and assembled in the lab. It consists of a 300W Xe lamp inside housing attached to
the power supply. The outcoming light is directed inside an Oriel Cornerstone
TM
260
monochromator with 2 gratings to accurately cover the entire solar spectrum. The
monochromator output is attached to a chopper to obtain a frequency variable light
output. Later, a filter wheel is placed to ensure elimination of double lines generated
from the monochromator. A fiber optic directs the monochromatic beam into the device
that is placed horizontally, in a similar fashion as the white light testing, and a substrate
143

holder cuts down the illumination beam so that no light is missed away from the device
that will lead to an underestimation of the quantum efficiency percentage. Contacts are
also possible by the use of microelectrode and top approach if a gold wire is possible
thanks to the aid of a microscope positioning. Microelectrode signal is then recorded
using a EG&G 7220 Lock-in amplifier that improves the signal-to-noise ratio, using a
light chopped at a certain frequencies. The lock
-
in signal measures current at short circuit
conditions. Signal acquisition computer program was designed by Dr. Fan Wang during
his doctorate studies at University of Princeton. Figure 6.5 shows the experimental
layout of the monochromatic testing station.

Figure 6.5: Picture of tunable wavelength energy source and testing of spectral responsivity.
144

6.3 Spectral Mismatch calculation and standardization of light intensity
procedures
The performance of PV cells is commonly rated in terms of their efficiency with
respect to standard reporting conditions (SRC) defined by temperature, spectral
irradiance, and total irradiance. The SRC for rating the performance of terrestrial PV
cells are the following: 1000 Wm
–2
irradiance, AM 1.5 (AM: air mass) global reference
spectrum, and 25 °C cell temperature. For a given device to give a unique efficiency, the
incident power P
o
or E
tot
must be with respect to a reference spectral irradiance. The
current reference spectrum adopted by the international terrestrial photovoltaics
community is given in International Electrotechnical Commission (IEC) Standard 60904-
3 and American Society for Testing and Materials (ASTM) Standard G159. A recent
improvement to this spectrum is given in ASTM Standard G173 and is expected to be
adopted by the international photovoltaics community in the next year or two. The
irradiance incident on the PV cell is typically measured with a reference cell. For I–V
measurements with respect to a reference spectrum, there is a spectral error in the
measured short-circuit current (I
SC
) of the PV cell because of the following two reasons:
i) the spectral irradiance of the light source does not match the reference spectrum, which
is computer generated, and ii) the spectral responses of the reference detector and test cell
are different. This error can be derived based upon the assumption that the photocurrent
is the integral of the product of cell responsivity and incident spectral irradiance. This
error can be expressed as spectral mismatch correction factor (M),
145

∫
∫
∫
∫
=
2
1
2
1
2
1
2
1
) (
) (
) (
) (
Ref
Ref
λ
λ
λ
λ
λ
λ
λ
λ
λ λ
λ λ
λ λ
λ λ
d S E
d S E
d S E
d S E
M
R S
T S
T
R
6.1
where E
Ref
(λ) is the reference spectral irradiance, E
S
(λ) is the source spectral
irradiance, S
R
(λ) is the spectral responsivity of the reference cell, and S
T
(λ) is the spectral
responsivity of the test cell, each as a function of wavelength (λ). The limits of
integration λ
1
and λ
2
in the above equation should encompass the range of the reference
cell and the test-device spectral responses, and the simulator and reference spectra should
encompass λ
1
and λ
2
to avoid error. A matched PV reference cell is typically used as the
reference detector and a solar simulator is used as the light source to minimize the
deviation of M from unity.
The total effective irradiance of the light source (E
eff
), which is the total irradiance
seen by the cell, can be determined from the short-circuit current of the reference cell
under the source spectrum (I
R,S
) from the equation:
CN
M I
E
S R,
eff
= 6.2
where CN is the calibration number (in units of AW
–1
m
2
) for the instrument used to
measure the incident irradiance. E
eff
is different from E
tot
since E
tot
usually refers to the
total irradiance integrated over the entire spectrum, and not just the part of the spectrum
the cell responds to. Both E
eff
and E
tot
are derived from integrating E
S
(λ) over an
146

appropriate range of wavelength. The short-circuit current of a test cell (I
T,R
) at the
reference total irradiance (E
Ref
) is given as:
M I
CN E I
I
S R
R T
R T
,
Ref
,
,
= 6.3
where I
T,S
is the short-circuit current of a test cell measured under the source
spectrum. Once M is known, the simulator is adjusted so that E
eff
is equal to E
Ref
, or
M I
I I
I
S R
S T R R
R T
,
, ,
,
= 6.4
where I
R,R
is the calibrated short-circuit current of the reference cell under the
reference spectrum and total irradiance.
The reference cell calibration was performed by NREL for the reference
photodetectors purchased from Hamamatsu. Reference detectors that were used in this
work were chosen because of their similar response to the organic PVs prepared such that
M was as close to 1 as possible. They are the S1787-04 and S1787-12 with responses in
the visible and visible-IR respectively as seen in the Figure 6.6 (a) The specifications and
drawings are shown in Figure 6.6 (b)
147

148

(b)
(a)
Figure 6.5: Spectral response of reference detectors (a) and geometrical drawings (b).
(Images from www.hamamatsu.com)

Bibliography
- Aldridge, T.K., E.M. Stacy, and D.R. McMillin, Studies of the Room-Temperature
Absorption and Emission-Spectra of [Pt(Trpy)X]+ Systems. Inorganic Chemistry, 1994.
33(4): p. 722-727.
- Badger, G.M. and R.S. Pearce, Absorption spectrum of rubrene in different solvents.
Spectrochimica Acta, 1951. 4(4): p. 280-283.
- Bailey-Salzman, R.F., B.P. Rand, and S.R. Forrest, Near-infrared sensitive small
molecule organic photovoltaic cells based on chloroaluminum phthalocyanine. Applied
Physics Letters, 2007. 91(1): p. 013508.
- Borek, C., K. Hanson, P.I. Djurovich, M.E. Thompson, K. Aznavour, R. Bau, Y. Sun,
S.R. Forrest, J. Brooks, L. Michalski, and J. Brown, Highly Efficient, Near-Infrared
Electrophosphorescence from a Pt-Metalloporphyrin Complex. Angewandte Chemie
International Edition, 2007. 46(7): p. 1109-1112.
- Brabec, C.J., A. Cravino, D. Meissner, N.S. Sariciftci, T. Fromherz, M.T. Rispens, L.
Sanchez, and J.C. Hummelen, Origin of the open circuit voltage of plastic solar cells.
Advanced Functional Materials, 2001. 11(5): p. 374-380.
- Brooks, J., Y. Babayan, S. Lamansky, P.I. Djurovich, I. Tsyba, R. Bau, and M.E.
Thompson, Synthesis and characterization of phosphorescent cyclometalated platinum
complexes. Inorganic Chemistry, 2002. 41(12): p. 3055-3066.
- Brown, R.J.C., A.R. Kucernak, N.J. Long, and C. Mongay-Batalla, Spectroscopic and
electrochemical studies on platinum and palladium phthalocyanines. New Journal of
Chemistry, 2004. 28(6): p. 676-680.
- Bube, H.R. and A.L. Fahrenbruch, Advances in Electronics and Electron Physics.
1981, Academic: New York. p. 163.
- Burrows, P.E., Z. Shen, V. Bulovic, D.M. McCarty, S.R. Forrest, J.A. Cronin, and
M.E. Thompson, Relationship between electroluminescence and current transport in
organic heterojunction light-emitting devices. Journal of Applied Physics, 1996. 79(10):
p. 7991-8006.
- Castellano, F.N., I.E. Pomestchenko, E. Shikhova, F. Hua, M.L. Muro, and N.
Rajapakse, Photophysics in bipyridyl and terpyridyl platinum(II) acetylides. Coordination
Chemistry Reviews, 2006. 250(13-14): p. 1819-1828.
149

- Chan, C.W., L.K. Cheng, and C.M. Che, Luminescent Donor-Acceptor Platinum(Ii)
Complexes. Coordination Chemistry Reviews, 1994. 132: p. 87-97.
- Chan, M.Y., S.L. Lai, M.K. Fung, C.S. Lee, and S.T. Lee, Doping-induced efficiency
enhancement in organic photovoltaic devices. Applied Physics Letters, 2007. 90(2): p.
023504.
- Chang, E.C., C.I. Chao, and R.H. Lee, Enhancing the efficiency of MEH-PPV and
PCBM based polymer solar cells via optimization of device configuration and processing
conditions. Journal of Applied Polymer Science, 2006. 101(3): p. 1919-1924.
- Chau, L.K., C.D. England, S.Y. Chen, and N.R. Armstrong, Visible Absorption and
Photocurrent Spectra of Epitaxially Deposited Phthalocyanine Thin-Films -
Interpretation of Exciton Coupling Effects. Journal of Physical Chemistry, 1993. 97(11):
p. 2699-2706.
- Chelvayohan, M. and C.H.B. Mee, Work function measurements on (110), (100) and
(111) surfaces of silver. Journal of Physics C: Solid State Physics, 1982. 15(10): p. 2305-
2312.
- Cheung, T.C., K.K. Cheung, S.M. Peng, and C.M. Che, Photoluminescent
cyclometallated diplatinum(II,II) complexes: Photophysical properties and crystal
structures of [PtL(PPh(3))]ClO4 and [Pt(2)L2(mu-dppm)][ClO4](2) (HL=6-phenyl-2,2'-
bipyridine, dppm=Ph(2)PCH(2)PPh(2)). Journal of the Chemical Society-Dalton
Transactions, 1996(8): p. 1645-1651.
- Cho, J.-Y., K.Y. Suponitsky, J. Li, T.V. Timofeeva, S. Barlow, and S.R. Marder,
Cyclometalated platinum complexes: High-yield synthesis, characterization, and a
crystal structure. Journal of Organometallic Chemistry, 2005. 690(17): p. 4090-4093.
- Cho, J.Y., B. Domercq, S. Barlow, K.Y. Suponitsky, J. Li, T.V. Timofeeva, S.C.
Jones, L.E. Hayden, A. Kimyonok, C.R. South, M. Weck, B. Kippelen, and S.R. Marder,
Synthesis and characterization of polymerizable phosphorescent platinum(II) complexes
for solution-processible organic light-emitting diodes. Organometallics, 2007. 26: p.
4816-4829.
- Chu, C.W., Y. Shao, V. Shrotriya, and Y. Yang, Efficient photovoltaic energy
conversion in tetracene-C
60
based heterojunctions. Applied Physics Letters, 2005.
86(24): p. 243506.
- Clark, M.L., S. Diring, P. Retailleau, D.R. McMillin, and R. Ziessel, Spectroscopic
properties of orthometalated platinum(II) bipyridine complexes containing various
ethynylaryl groups. Chemistry-a European Journal, 2008. 14(24): p. 7168-7179.
150

- Crites Tears, D.K. and D.R. McMillin, Exciplex quenching of photoexcited
platinum(II) terpyridines: influence of the orbital parentage. Coordination Chemistry
Reviews, 2001. 211(1): p. 195-205.
- Cummings, S.D. and R. Eisenberg, Tuning the excited-state properties of platinum(II)
diimine dithiolate complexes. Journal of the American Chemical Society, 1996. 118(8): p.
1949-1960.
- D'Andrade, B. and S.R. Forrest, Formation of triplet excimers and dimers in
amorphous organic thin films and light emitting devices. Chemical Physics, 2003. 286(2-
3): p. 321-335.
- D'Andrade, B.W., S. Datta, S.R. Forrest, P. Djurovich, E. Polikarpov, and M.E.
Thompson, Relationship between the ionization and oxidation potentials of molecular
organic semiconductors. Organic Electronics, 2005. 6(1): p. 11-20.
- Dienel, T., H. Proehl, T. Fritz, and K. Leo, Novel near-infrared photoluminescence
from platinum(II)-porphyrin (PtOEP) aggregates. Journal of Luminescence, 2004.
110(4): p. 253-257.
- Djurovich, P.I., E.I. Mayo, S.R. Forrest, and M.E. Thompson, Measurement of the
lowest unoccupied molecular orbital energies of molecular organic semiconductors.
Organic Electronics, 2009.
- Eastment, R.M. and C.H.B. Mee, Work Function Measurements on (100), (110) and
(111) Surfaces of Aluminum. Journal of Physics F-Metal Physics, 1973. 3(9): p. 1738-
1745.
- Fahrenbruch, A.L. and J. Aranovich, Solar Energy Conversion - Solid-State Physics
Aspects. Topics in Applied Physics, ed. B.O. Seraphin. Vol. 31. 1979, Berlin Heidelberg
New York: Springer-Verlag. 257-326.
- Ferreira, J.A., R. Barral, J.D. Baptista, and M.I.C. Ferreira, Absorption coefficients and
fluorescence quantum yields of porphyrin films determined by optical and photoacoustic
spectroscopies. Journal of Luminescence, 1991. 48-49(1): p. 385-390.
- Fleeman, W.L. and W.B. Connick, Self-quenching and cross-quenching reactions of
excited platinum(II) diimine complexes. Comments on Inorganic Chemistry, 2002. 23(3):
p. 205-230.
- Gadisa, A., M. Svensson, M.R. Andersson, and O. Inganas, Correlation between
oxidation potential and open-circuit voltage of composite solar cells based on blends of
polythiophenes/ fullerene derivative. Applied Physics Letters, 2004. 84(9): p. 1609-1611.
151

- Ginley, D., M.A. Green, and R. Collins, Harnessing Materials for Energy, in MRS
Bulletin. 2008. p. 355-372.
- Gouterman, M., D. Holten, and E. Lieberman, Porphyrins XXXV . Exciton coupling in
[mu]-oxo Scandum dimers. Chemical Physics, 1977. 25(1): p. 139-153.
- Grepstad, J.K., P.O. Gartland, and B.J. Slagsvold, Anisotropic work function of clean
and smooth low-index faces of aluminium. Surface Science, 1976. 57(1): p. 348-362.
- Gundlach, D.J., J.A. Nichols, L. Zhou, and T.N. Jackson, Thin-film transistors based
on well-ordered thermally evaporated naphthacene films. Applied Physics Letters, 2002.
80(16): p. 2925-2927.
- Gutmann, V., Ion-Pairing and Outer Sphere Effect. Chimia, 1977. 31(1): p. 1-7.
- Hass, G., J.B. Heaney, H. Herzig, J.F. Osantowski, and J.J. Triolo, Reflectance and
Durability of Ag Mirrors Coated with Thin-Layers of Al2o3 Plus Reactively Deposited
Silicon-Oxide. Applied Optics, 1975. 14(11): p. 2639-2644.
- Hassan, A., Theoretical, experimental, device fabrication and degradation studies of
materials for optoelectronic devices., in Chemistry. 2007, University of Southern
California: Los Angeles.
- Hirose, Y., A. Kahn, V. Aristov, P. Soukiassian, V. Bulovic, and S.R. Forrest,
Chemistry and electronic properties of metal-organic semiconductor interfaces: Al, Ti,
In, Sn, Ag, and Au on PTCDA. Physical Review B, 1996. 54(19): p. 13748-13758.
- Horváth, A. and K.L. Stevenson, Transition metal complex exciplexes. Coordination
Chemistry Reviews, 1996. 153: p. 57-82.
- Ionkin, A.S., W.J. Marshall, and Y. Wang, Syntheses, structural characterization, and
first electroluminescent properties of mono-cyclometalated platinum(II) complexes with
greater than classical pi-pi stacking and Pt-Pt distances. Organometallics, 2005. 24(4):
p. 619-627.
- Jabbour, G.E., B. Kippelen, N.R. Armstrong, and N. Peyghambarian, Aluminum based
cathode structure for enhanced electron injection in electroluminescent organic devices.
Applied Physics Letters, 1998. 73(9): p. 1185-1187.
- Janssen, A.G.F., T. Riedl, S. Hamwi, H.H. Johannes, and W. Kowalsky, Highly
efficient organic tandem solar cells using an improved connecting architecture. Applied
Physics Letters, 2007. 91(7): p. 073519.
152

- Kalinowski, J., W. Stampor, J. Szmytkowski, M. Cocchi, D. Virgili, V. Fattori, and
P.D. Marco, Photophysics of an electrophosphorescent platinum (II) porphyrin in solid
films. The Journal of Chemical Physics, 2005. 122(15): p. 154710.
- Kampen, T.U., A. Das, S. Park, W. Hoyer, and D.R.T. Zahn, Relation between
morphology and work function of metals deposited on organic substrates. Applied
Surface Science, 2004. 234(1-4): p. 333-340.
- Kim, S.H., J. Jang, and J.Y. Lee, Improvement in Power Efficiency in Organic Light
Emitting Diodes Through Intermediate Mg:Ag Layer in LiF/Mg:Ag/Al Cathodes.
Electrochemical and Solid-State Letters, 2007. 10(10): p. J117-J119.
- Kooistra, F.B., J. Knol, F. Kastenberg, L.M. Popescu, W.J.H. Verhees, J.M. Kroon,
and J.C. Hummelen, Increasing the Open Circuit Voltage of Bulk-Heterojunction Solar
Cells by Raising the LUMO Level of the Acceptor. Org. Lett., 2007. 9(4): p. 551-554.
- Koster, L.J.A., V.D. Mihailetchi, R. Ramaker, and P.W.M. Blom, Light intensity
dependence of open-circuit voltage of polymer:fullerene solar cells. Applied Physics
Letters, 2005. 86(12): p. 123509.
- Lai, S.-W. and C.-M. Che, Luminescent Cyclometalated Diimine Platinum(II)
Complexes: Photophysical Studies and Applications, in Topics in Current Chemistry.
2004. p. 27-63.
- Lai, S.W., M.C.W. Chan, T.C. Cheung, S.M. Peng, and C.M. Che, Probing d(8)-d(8)
Interactions in luminescent mono- and binuclear cyclometalated platinum(II) complexes
of 6-phenyl-2,2 '-bipyridines. Inorganic Chemistry, 1999. 38(18): p. 4046-4055.
- Lee, C.J., R.B. Pode, J.I. Han, and D.G. Moon, Ca/Ag bilayer cathode for transparent
white organic light-emitting devices. Applied Surface Science, 2007. 253(9): p. 4249-
4253.
- Leznoff, C.C. and A.B.P. Lever, Phthalocyanines : properties and applications 1989,
New York: VCH.
- Li, G., V. Shrotriya, J.S. Huang, Y. Yao, T. Moriarty, K. Emery, and Y. Yang, High-
efficiency solution processable polymer photovoltaic cells by self-organization of
polymer blends. Nature Materials, 2005. 4(11): p. 864-868.
- Li, N., B.E. Lassiter, R.R. Lunt, G. Wei, and S.R. Forrest, Open circuit voltage
enhancement due to reduced dark current in small molecule photovoltaic cells. Applied
Physics Letters, 2009. 94(2): p. 023307-3.
153

- Liu, F., K.L. Cunningham, W. Uphues, G.W. Fink, J. Schmolt, and D.R. McMillin,
Luminescence Quenching of Copper(Ii) Porphyrins with Lewis-Bases. Inorganic
Chemistry, 1995. 34(8): p. 2015-2018.
- Liu, J., S. Y., and Y. Yang, Solvation-Induced Morphology Effects on the Performance
of Polymer-Based Photovoltaic Devices. Advanced Functional Materials, 2001. 11(6): p.
420-424.
- Lunt, R.R., N.C. Geibink, A.A. Belak, J.B. Benziger, and S.R. Forrest, Measurement
of the Exciton Diffusion Length of Amorphous and Polycrystalline Organic
Semiconductors by Spectrally Resolved Photoluminescence Quenching. Manuscript in
preparation, 2009.
- Ma, B., P.I. Djurovich, and M.E. Thompson, Excimer and electron transfer quenching
studies of a cyclometalated platinum complex. Coordination Chemistry Reviews, 2005.
249(13-14): p. 1501-1510.
- Ma, B., P.I. Djurovich, M. Yousufuddin, R. Bau, and M.E. Thompson,
Phosphorescent Platinum Dyads with Cyclometalated Ligands: Synthesis,
Characterization, and Photophysical Studies†. The Journal of Physical
Chemistry C, 2008. 112(21): p. 8022-8031.
- Ma, B.W., P.I. Djurovich, S. Garon, B. Alleyne, and M.E. Thompson, Platinum
binuclear complexes as phosphorescent dopants for monochromatic and white organic
light-emitting diodes. Advanced Functional Materials, 2006. 16(18): p. 2438-2446.
- Ma, W., C. Yang, X. Gong, K. Lee, and A.J. Heeger, Thermally Stable, Efficient
Polymer Solar Cells with Nanoscale Control of the Interpenetrating Network
Morphology. Advanced Functional Materials, 2005. 15(10): p. 1617-1622.
- Mandoc, M.M., L.J.A. Koster, and P.W.M. Blom, Optimum charge carrier mobility in
organic solar cells. Applied Physics Letters, 2007. 90: p. 133504.
- Mattheus, C.C., W. Michaelis, C. Kelting, W.S. Durfee, D. Wöhrle, and D.
Schlettwein, Influence of the molecular shape on the film growth of a substituted
phthalocyanine. Synthetic Metals, 2004. 146(3): p. 335-339.
- McGuire, R., M.H. Wilson, J.J. Nash, P.E. Fanwick, and D.R. McMillin, Lewis acid
and base sensing by platinum(II) polypyridines. Inorganic Chemistry, 2008. 47(8): p.
2946-2948.
- McMillin, D.R., J.R. Kirchhoff, and K.V. Goodwin, Exciplex quenching of photo-
excitd copper complexes. Coordination Chemistry Reviews, 1985. 64: p. 83-92.
154

- Meier, H., Organic semiconductors : dark and photoconductivity of organic solids
1974: Verlag Chemie.
- Michalec, J.F., S.A. Bejune, D.G. Cuttell, G.C. Summerton, J.A. Gertenbach, J.S.
Field, R.J. Haines, and D.R. McMillin, Long-lived emissions from 4 '-substituted
Pt(trpy)Cl+ complexes bearing aryl groups. Influence of orbital parentage. Inorganic
Chemistry, 2001. 40(9): p. 2193-2200.
- Mihailetchi, V.D., P.W.M. Blom, J.C. Hummelen, and M.T. Rispens, Cathode
dependence of the open-circuit voltage of polymer : fullerene bulk heterojunction solar
cells. Journal of Applied Physics, 2003. 94(10): p. 6849-6854.
- Mihailetchi, V.D., L.J.A. Koster, and P.W.M. Blom, Effect of metal electrodes on the
performance of polymer : fullerene bulk heterojunction solar cells. Applied Physics
Letters, 2004. 85(6): p. 970-972.
- Montalti, M., A. Credi, L. Prodi, and M.T. Gandolfi, Handbook of Photochemistry.
Third Edition ed, ed. C. Press. 2006. 650.
- Mori, T., S. Oda, N. Ooishi, and Y. Masumoto, Polycrystallization of vaporized hole-
transport materials for organic light-emitting diodes and its suppression using organic
alloy method. Japanese Journal of Applied Physics Part 1-Regular Papers Brief
Communications & Review Papers, 2007. 46(9A): p. 5954-5959.
- Mutolo, K.L., E.I. Mayo, B.P. Rand, S.R. Forrest, and M.E. Thompson, Enhanced
open-circuit voltage in subphthalocyanine/C-60 organic photovoltaic cells. Journal of the
American Chemical Society, 2006. 128(25): p. 8108-8109.
- Nelson, J., The Physics of Solar Cells. 2003, London: Imperial College Press.
- Padinger, F., R.S. Rittberger, and N.S. Sariciftci, Effects of Postproduction Treatment
on Plastic Solar Cells. Advanced Functional Materials, 2003. 13(1): p. 85-88.
- Pandey, A.K. and J.M. Nunzi, Rubrene/Fullerene Heterostructures with a Half-Gap
Electroluminescence Threshold and Large Photovoltage. Advanced Materials, 2007.
19(21): p. 3613-3617.
- Peet, J., J.Y. Kim, N.E. Coates, W.L. Ma, D. Moses, A.J. Heeger, and G.C. Bazan,
Efficiency enhancement in low-bandgap polymer solar cells by processing with alkane
dithiols. Nat Mater, 2007. 6(7): p. 497-500.
- Perez, M.D., C. Borek, P.I. Djurovich, E.I. Mayo, R.R. Lunt, S.R. Forrest, and M.E.
Thompson, Organic photovoltaics using tetraphenylbenzoporphyrin complexes as donor
layers. Advanced Materials, 2009. 21(14-15): p. 1517-1520
155

- Peumans, P. and S.R. Forrest, Very-high-efficiency double-heterostructure copper
phthalocyanine/C
60
photovoltaic cells. Applied Physics Letters, 2001. 79(1): p. 126-128.
- Peumans, P., A. Yakimov, and S.R. Forrest, Small molecular weight organic thin-film
photodetectors and solar cells. Journal of Applied Physics, 2003. 93(7): p. 3693-3723.
- Potscavage, J.W.J., S. Yoo, and B. Kippelen, Origin of the open-circuit voltage in
multilayer heterojunction organic solar cells. Applied Physics Letters, 2008. 93(19): p.
193308-3.
- Rand, B.P., D.P. Burk, and S. Forrest, R., Offset energies at organic semiconductor
heterojunctions and their influence on the open-circuit voltage of thin-film solar cells.
Physical Review B (Condensed Matter and Materials Physics), 2007. 75(11): p. 115327.
- Rand, B.P., J. Xue, F. Yang, and S.R. Forrest, Organic solar cells with sensitivity
extending into the near infrared. Applied Physics Letters, 2005. 87(23): p. 233508.
- Reyes-Reyes, M., K. Kim, and D.L. Carroll, High-efficiency photovoltaic devices
based on annealed poly(3-hexylthiophene) and 1-(3-methoxycarbonyl)-propyl-1- phenyl-
(6,6)C
61
blends. Applied Physics Letters, 2005. 87(8): p. 083506.
- Rogers, J.E., K.A. Nguyen, D.C. Hufnagle, D.G. McLean, W.J. Su, K.M. Gossett,
A.R. Burke, S.A. Vinogradov, R. Pachter, and P.A. Fleitz, Observation and
interpretation of annulated porphyrins: Studies on the photophysical properties of meso-
tetraphenylmetalloporphyrins. Journal of Physical Chemistry A, 2003. 107(51): p.
11331-11339.
- Sakurai, T. and S. Hayakawa, Optical-Properties of Tetracene Evaporated-Films.
Japanese Journal of Applied Physics, 1974. 13(11): p. 1733-1740.
- Sarangerel, K., C. Ganzorig, M. Fujihira, M. Sakomura, and K. Ueda, Influence of the
Work Function of Chemically Modified Indium-Tin-Oxide Electrodes on the Open-circuit
Voltage of Heterojunction Photovoltaic Cells. Chemistry Letters, 2008. 37(7): p. 778-
779.
- Sato, N., K. Seki, and H. Inokuchi, Polarization Energies of Organic-Solids
Determined by Ultraviolet Photoelectron-Spectroscopy. Journal of the Chemical Society-
Faraday Transactions II, 1981. 77: p. 1621-1633.
- Scharber, M.C., D. Wuhlbacher, M. Koppe, P. Denk, C. Waldauf, A.J. Heeger, and
C.L. Brabec, Design rules for donors in bulk-heterojunction solar cells - Towards 10 %
energy-conversion efficiency. Advanced Materials, 2006. 18(6): p. 789-+.
156

- Schulze, K., C. Uhrich, R. Schüppel, K. Leo, M. Pfeiffer, E. Brier, E. Reinold, and P.
Bäuerle, Efficient Vacuum-Deposited Organic Solar Cells Based on a New Low-Bandgap
Oligothiophene and Fullerene C
60
. Advanced Materials, 2006. 18(21): p. 2872-2875.
- Schwedhelm, R., L. Kipp, A. Dallmeyer, and M. Skibowski, Experimental band gap
and core-hole electron interaction in epitaxial C60 films. Physical Review B, 1998.
58(19): p. 13176.
- Seo, S., B.-N. Park, and P.G. Evans, Ambipolar rubrene thin film transistors. Applied
Physics Letters, 2006. 88(23): p. 232114-3.
- Shao, Y. and Y. Yang, Efficient organic heterojunction photovoltaic cells based on
triplet materials. Advanced Materials, 2005. 17(23): p. 2841-2844.
- Shrotriya, V., G. Li, Y. Yao, T. Moriarty, K. Emery, and Y. Yang, Accurate
Measurement and Characterization of Organic Solar Cells. Advanced Functional
Materials, 2006. 16(15): p. 2016-2023.
- Singh, V.P., R.S. Singh, B. Parthasarathy, A. Aguilera, J. Anthony, and M. Payne,
Copper-phthalocyanine-based organic solar cells with high open-circuit voltage. Applied
Physics Letters, 2005. 86(8): p. 082106.
- Song, Q.L., F.Y. Li, H. Yang, H.R. Wu, X.Z. Wang, W. Zhou, J.M. Zhao, X.M. Ding,
C.H. Huang, and X.Y. Hou, Small-molecule organic solar cells with improved stability.
Chemical Physics Letters, 2005. 416(1-3): p. 42-46.
- Stacy, E.M. and D.R. McMillin, Inorganic Exciplexes Revealed by Temperature-
Dependent Quenching Studies. Inorganic Chemistry, 1990. 29(3): p. 393-396.
- Stampor, W., Electroabsorption study of vacuum-evaporated films of
Pt(II)octaethylporphyrin. Chemical Physics, 2004. 305(1-3): p. 77-84.
- Stössel, M., J. Staudigel, F. Steuber, J. Simmerer, and A. Winnacker, Impact of the
cathode metal work function on the performance of vacuum-deposited organic light
emitting-devices. Applied Physics A: Materials Science & Processing, 1999. 68(4): p.
387-390.
- Sze, S.M., Physics of semiconductor devices (2nd edition). 1981, United States: Wiley-
Interscience,New York, NY. Pages: 878.
- Taima, T., J. Sakai, T. Yamanari, and K. Saito, Realization of large open-circuit
photovoltage in organic thin-film solar cells by controlling measurement environment.
Japanese Journal of Applied Physics Part 2-Letters & Express Letters, 2006. 45(37-41):
p. L995-L997.
157

- Tang, C.W., 2-Layer Organic Photovoltaic Cell. Applied Physics Letters, 1986. 48(2):
p. 183-185.
- Tranthi, T.H., J.F. Lipskier, P. Maillard, M. Momenteau, J.M. Lopezcastillo, and J.P.
Jaygerin, Effect of the Exciton Coupling on the Optical and Photophysical Properties of
Face-to-Face Porphyrin Dimer and Trimer - a Treatment Including the Solvent
Stabilization Effect. Journal of Physical Chemistry, 1992. 96(3): p. 1073-1082.
- Turro, N.J., Modern Molecular Photochemistry. 1991: University Science Books.
- Vandewal, K., A. Gadisa, W.D. Oosterbaan, S. Bertho, F. Banishoeib, I. Van Severen,
L. Lutsen, T.J. Cleij, D. Vanderzande, and M.J. V., The Relation Between Open-Circuit
Voltage and the Onset of Photocurrent Generation by Charge-Transfer Absorption in
Polymer : Fullerene Bulk Heterojunction Solar Cells. Advanced Functional Materials,
2008. 18(14): p. 2064-2070.
- Vaubel, G. and H. Baessler, Diffusion of Singlet Excitons in Tetracene Crystals.
Molecular Crystals and Liquid Crystals, 1970. 12(1): p. 47-&.
- Veenstra, S.C., A. Heeres, G. Hadziioannou, G.A. Sawatzky, and H.T. Jonkman, On
interface dipole layers between C
60
and Ag or Au. Applied Physics A: Materials Science
& Processing, 2002. 75(6): p. 661-666.
- Wan, K.T., C.M. Che, and K.C. Cho, Inorganic Excimer - Spectroscopy, Photoredox
Properties and Excimeric Emission of Dicyano(4,4'-Di-Tert-Butyl-2,2'-
Bipyridine)Platinum(Ii). Journal of the Chemical Society-Dalton Transactions, 1991(4):
p. 1077-1080.
- Weinberg-Wolf, J.R., L.E. McNeil, S. Liu, and C. Kloc, Evidence of low
intermolecular coupling in rubrene single crystals by Raman scattering. Journal of
Physics: Condensed Matter, 2007. 19(27): p. 276204.
- Williams, J., Photochemistry and Photophysics of Coordination Compounds:
Platinum, in Photochemistry and Photophysics of Coordination Compounds II. 2007. p.
205-268.
- Wong, K.M.-C. and V.W.-W. Yam, Luminescence platinum(II) terpyridyl complexes--
From fundamental studies to sensory functions. Coordination Chemistry Reviews, 2007.
251(17-20): p. 2477-2488.
- Würfel, P., Physics of Solar Cells: From Principles to New Concepts. 2005, Weinhem:
Wiley-VCH. 186.
158

159

- Xiang, H.-F., S.-W. Lai, P.T. Lai, and C.-M. Che, Phosphorescent Platinum(II)
Materials for OLED Applications, in Highly Efficient OLEDs with Phosphorescent
Materials, H. Yersin, Editor. 2007, Wiley-VCH: Weinheim. p. 259.
- Xue, J.G., B.P. Rand, S. Uchida, and S.R. Forrest, A hybrid planar-mixed molecular
heterojunction photovoltaic cell. Advanced Materials, 2005. 17(1): p. 66-+.
- Xue, J.G., S. Uchida, B.P. Rand, and S.R. Forrest, 4.2% efficient organic photovoltaic
cells with low series resistances. Applied Physics Letters, 2004. 84(16): p. 3013-3015.
- Yin, B., F. Niemeyer, J.A.G. Williams, J. Jiang, A. Boucekkine, L. Toupet, H.
LeBozec, and V. Guerchais, Synthesis, Structure, and Photophysical Properties of
Luminescent Platinum(II) Complexes Containing Cyclometalated 4-Styryl-Functionalized
2-Phenylpyridine Ligands. Inorg. Chem., 2006. 45(21): p. 8584-8596.
- Yoo, S., B. Domercq, and B. Kippelen, Efficient thin-film organic solar cells based on
pentacene/C
60
heterojunctions. Applied Physics Letters, 2004. 85(22): p. 5427-5429.

Abstract (if available)

Abstract The field of organic photovoltaics research has received a great deal of attention in the past years. Global climate changes are renewing the need for cleaner alternative energies that will prevent increased effects in global warming. Solar photoconversion arises as a natural option for the satisfaction of large energy demands with diminished impact on the environment. However, the high production costs of the current solar conversion technologies have prevented the widespread use of the sun as energy source. Costs reductions will be possible by the use of organic materials as photoactive materials since manufacturing processes are less demanding that those for the inorganic counterparts. However, current laboratory efficiencies do not match those required for commercialization and a huge leap into improved photoconversion must be achieved. Small molecule organic photovoltaics are a relatively new field and much is still to be known about the processes involved in solar to electric conversion. The introduction of new materials will provide the insight required for learning more about those processes. Organic synthesis will play a fundamental role in the evolution of this area.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Organic solar cells: molecular electronic processes and device development

PDF

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

PDF

Porphyrin based near infrared‐absorbing materials for organic photovoltaics

PDF

Molecular and polymeric donors for bulk heterojunction organic solar cells

PDF

Properties and applications of dipyrrin-based molecules in organic photovoltaics

PDF

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

PDF

Utilizing n-heterocyclic chromophores for solar energy harvesting.

PDF

Squaraines and their applications to organic photovoltaics

PDF

Energy management in organic photovoltaics

PDF

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

PDF

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

PDF

Organic photovoltaics: from materials development to device application

PDF

Growth, characterization of gallium arsenide based nanowires and application in photovoltaic cells

PDF

First steps of solar energy conversion; primary charge generation processes in condensed phase coupled chromophores

PDF

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

PDF

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

PDF

Synthesis and characterization of novel donor-acceptor systems for solar water splitting

PDF

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

PDF

Molecular design for organic photovoltaics: tuning state energies and charge transfer processes in heteroaromatic chromophores

PDF

Pyrrolic squaraines and energy management in organic photovoltaics

Asset Metadata

Creator Perez, Maria Dolores (author)

Core Title Molecular aspects of photoconversion processes in organic solar cells

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Chemistry

Publication Date 05/07/2009

Defense Date 01/29/2009

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

Tag exciton diffusion length,OAI-PMH Harvest,open circuit voltage,organic semiconductors,organic solar cells,photovoltaics

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Thompson, Mark E. (committee chair), Dapkus, P. Daniel (committee member), Reisler, Hannah (committee member)

Creator Email mdperez@gmail.com,perezd@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m2204

Unique identifier UC152239

Identifier etd-Perez-2797 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-245250 (legacy record id),usctheses-m2204 (legacy record id)

Legacy Identifier etd-Perez-2797.pdf

Dmrecord 245250

Document Type Dissertation

Rights Perez, Maria Dolores

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu