University of Southern California Dissertations and Theses

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

(USC Thesis Other)

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

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Synthesis and Characterization of
Novel Donor-Acceptor Systems for
Solar Water Splitting

by

Rebecca Jean Wilson
A dissertation submitted to the
Faculty of the USC Graduate School
in partial fulfillment of the degree of
Doctor of Philosophy

in
Chemical Engineering
August 2019

Advisor: Mark E. Thompson

University of Southern California
Viterbi School of Engineering

Acknowledgements

First and foremost, I want to thank, from the bottom of my heart, Professor Mark
Thompson. The support and encouragement he gave me to pursue my passions led me
to many experiences I could never have imagined. He let me grow on my own path,
which I am eternally grateful for. I feel incredibly fortunate that I had him as a mentor
throughout this adventure.

The journey to becoming a doctor of philosophy is littered with unforseen hurdles, and
my parents, Bob and Jean Wilson, were always there to catch me if I tripped over one.
Their unconditional love assured me that I would always land on my feet, since they
would always be there for me. My sisters, Ashley, Kristina, Kerry and Faith, always gave
me support when I needed it, and their love and admiration encouraged me to keep
going.

The opportunity to live and do research in Strasbourg, France was a dream come true,
and would not have been possible without Prof. Luisa De Cola. I cannot thank her
enough. Dr. Eko Prasetyanto was an excellent mentor during my time there.

Prof. Peter Brüggeller welcomed me to his lab, and working in the heart of the Alps was
a wonderful experience. Christof Strabler and Johann Pann’s help with these
experiments were instrumental to my research.

I am grateful to my co-workers who taught and collaborated with me. Specifically, Dr.

Jessica Golden for making a BODIPY molecule for me and Savannah Kapper for being
my comrade in the bis(oxazoline) project. Judy Hom and Peter Djurovich are an
integral part of the MET lab, and without them nothing is possible.

Niki Bayat is a continuous source of inspiration for me to work hard toward my goals. I
feel so lucky I was able to go through graduate school with her by my side. I could not
have survived without Serena Cararra. She is the one person I know I can call up with
any problem, from anywhere in the world, and she can fix it for me. My roommate,
Rahne Avant, kept my life clean when I made it a mess. I love these ladies a lot and can’t
thank them enough for their love and support.

My friends from graduate school made this time of my life completely unforgettable.
Camping, game nights, pool parties and beaching with the best – Sean, Nick, Lena,
Carrie, Mark, Karol, Ola, Shruti, Harsh, Devang and Rahne – I’ll always love these fools.
Life in the MET group would have been boring without my coworkers. I’ve learned so
much from all of them: Niki, Rasha, Patrick, Roby, Matt, John Chen, Abegail, Narcisse,
Savannah, Muazzam, Andrew, John F., and many more. My time in Strasbourg was
made memorable from my friends and colleauges: Serena, Ingrid, Ondrej, John, Pilar,
Ricardo, Maurizio, Remi, Belinda, Elena, Mike, Christof and Stephan. Thank you for
making life fun!

Starting a company from scratch and competing for funding was an incredible learning
experience. I thank Andrew, Paulo and Niki for all of their hard work and motivation in
this process. Co-chairing the 2018 GRS on Electronic Processes in Organic Materials

was truly an honor, and would not have been possible without my fellow co-chair, David
Kiefer. The support of Prof. Antoine Kahn and Prof. Anna Kohler was instrumental in a
successful conference.

I could not have possibly achieved what I have without the guidance and help of my
undergraduate professor, Dr. Ihab Farag. His mentorship led me into graduate school,
and I am forever grateful to him. My forever friends from undergrad and high school
will always be my support – Sam, Gina and Katherine!

And last but not least, I want to thank myself, for making it to the end.

Table of Contents

1. Solar Energy ......................................................................................................................... 1
2. Electron Transport Layer in Inverted OPVs ...................................................................... 27
3. Microparticulate Light Harvesting Units ........................................................................... 62
4. Chirality-Driven Assembly of Push-Pull Molecules ......................................................... 125

1

1 Solar Energy
Why we need it and how to harvest it

2

Table of Contents

Climate Change ...................................................................................................................3
1.1.1 Historical Perspective on Climate Change ............................................................................................... 3
1.1.2 Impact of Climate Change ........................................................................................................................ 7
Photovoltaics (PVs) ........................................................................................................... 10
1.2.1 History of PVs ......................................................................................................................................... 10
1.2.2 First- and second- generation PVs ......................................................................................................... 10
1.2.3 Organic Photovoltaics (OPVs) ................................................................................................................ 13
Hydrogen Economy ........................................................................................................... 14
1.3.1 History of the Fuel Cell ........................................................................................................................... 14
1.3.2 Photoelectrochemistry .......................................................................................................................... 15
1.3.3 Photocatalytic Water Splitting ............................................................................................................... 16
1.3.1 Natural Photosynthesis .......................................................................................................................... 18
1.3.2 Artificial Photosynthesis ........................................................................................................................ 19
Current Study .................................................................................................................... 22
Bibliography ...................................................................................................................... 23

3

Climate Change
1.1.1 Historical Perspective on Climate Change
In the 18
th
century, the discovery of coal led to the invention of the steam engine by
Thomas Newcomen in 1712. This prompted the Industrial Revolution, which was a
significant step forward for humanity, shifting from human- to machine power. The
steam engine changed the face of manufacturing, as it powered machinery in mines,
factories and mills. It powered locomotives, revolutionizing travel. Society was
transformed and innovation continued into the 20
th
century with the internal combustion
engine. Today, we are now more reliant than ever on fossil fuels. Fossil fuels are complex
hydrocarbons, which when burned in the presences of O2, result in heat for energy, as well
as a plethora of unconsumed by-products (most notably CO2 and H2O).
Over a century ago, in 1896, Svante Arrhenius published the first article to
postulate on the detrimental effects of burning fossil fuels. He hypothesized that the
increase in carbon dioxide in the atmosphere, due to burning fossil fuels, would
simultaneously increase the temperature of the Earth, known as the “greenhouse effect.”
1

The Earth is warmed by solar heat, and this heat is reflected back toward space. However,
greenhouse gases, or GHG’s (carbon dioxide, methane, water vapor, nitrous oxide and
fluorinated gases), absorb this heat and re-emit it in the lower atmosphere, resulting in
the lower atmosphere warming. Few took his work seriously, one being G. S. Callendar,
who in the 1930’s used temperature measurements and the radiation absorption
coefficients of carbon dioxide and water to discuss how atmospheric carbon dioxide
increases the global temperature.
2
He explained that should the warming be caused by
an increase in solar heat, the entire atmosphere would be warmed, not exclusively the
lower atmosphere.
3

4

It wasn’t until 1960 that scientists started to seriously consider the impact of
burning fossil fuels. That year, C.D. Keeling released a study of measured carbon dioxide
levels in the atmosphere at various locations, most notably Mauna Loa, a volcano in
Hawaii. Due to it’s high elevation away from industrial activities, the levels of CO2 had
low variability. The sequential increase in CO2 over 3 years was a clear indicator that
human-related activities were GHG’s.
4
This study finally convinced scientists to further
study the effects of GHG’s on the atmosphere, and came at a time when environmental
conservation was gaining momentum. The 1960’s in the U.S. brought in a movement of
environmental conservation, with the passing of the Clean Air Act, the Clean Water Act,
the Air Quality Act and many others. Keeling continued his measurements, his record
from Mauna Loa is kept still today, and climate scientists have developed models to
predict the increase in global temperature. It’s a challenge though, as the Earth has
innumerable variables that are constantly changing. Over the next few decades,
questions, research and skeptisicm abounded, in attempts to deconvolute the intricate
climate and human’s impact on it. By 1988, the majority of scientists agreed that a
doubling in carbon dioxide levels in the atmosphere would result in a 1°C increase in
global temperature. To combat this, the International Panel on Climate Change (IPCC)
was formed by the United Nations.
5
The IPCC is an international aggregation of
scientists, representing their governments, to produce and assess evidence on climate
change globally. The reports include the risk of human-induced climate change, its
potential impacts and possible preventitive measures.
6

Today, climate change, or the increased warming of the Earths atmosphere due to
human activity, is unequivocal. According to the 2014 IPCC report, “each of the last three
decades has been successively warmer at the Earth’s surface than any preceding decade
5

since 1850. The period from 1983 to 2012 was likely the warmest 30-year period of the
last 1400 years in the Northern Hemisphere. The globally averaged combined land and
ocean surface temperature data as calculated by a linear trend show a warming of 0.85
[0.65 to 1.06] °C over the period 1880 to 2012, when multiple independently produced
datasets exist.”
7
These rising temperatures have resulted in glacial melts and a
subsequent rise in ocean levels (Figure 1-1, b). The trend with which the temperature and
sea levels are rising follows the release of GHG’s due to human-related activities (Figure
1-1).

6

Figure 1-1: (a) Annually and globally averaged combined land and ocean surface temperature
anomalies relative to the average over the period 1986 to 2005. Colors indicate different data
sets. (b) Annually and globally averaged sea-level change relative to the average over the period
1986 to 2005 in the longest-running dataset. Colors indicate different data sets. (c) Atmospheric
concentrations of the greenhouse gases carbon dioxide (CO 2, green), methane (CH 4, orange),
and nitrous oxide (N 2O, red) determined from ice core data (dots) and from direct atmospheric
measurements (lines). Indicators: (d) Global anthropogenic CO 2 emissions from forestry and
other land use as well as from burning of fossil fuel, cement production, and flaring. Cumulative
emissions of CO 2 from these sources and their uncertainties are shown as bars and whiskers,
respectively, on the right-hand side.
7

1.1.2 Impact of Climate Change
In October 2018, the IPCC released a special report evaluating the changes an
increase in the global average temperature of 1.5°C (from pre-industrial levels) would be
in the world.
8
It was compiled by 91 scientist from 40 different countries, and describes
a vastly different world. Currently, the global temperature has risen 1°C from pre-
industrial levels, and significant changes would be required to achieve solely a 1.5°C
increase in the next 30 years. This would require reducing GHG emissions to pre-2010
levels by 45% by 2030, and 100% by 2050. Were a 2°C increase to occur, GHG emissions
would have to be at pre-2010 levels by 2075. Both of these scenarios require an upheaval
of the way in which society operates. In the 1.5°C scenario, coal would have to be reduced
from current use of 40% to 1-7%, and technology to remove emissions from the
atmosphere have to be widely employed. One example of the devastation to the world due
to an increase in 1.5-2°C in temperature is exposing 30-80 million people to flooding from
sea level rise in 2100. Table 1-1 describes further devastation due to increase in global
temperatures.
Table 1-1: IPCC reported changes in the environment due to a temperature rise
8

Increase in
Temperature
Plants and
Animal Range
Arctic
Summer Ice
Coral Reefs

Water
Scarcity
1.5°C
6% insects
8% plants
4% vertebrates
Ice free once a
century

Frequent mass
mortalities

350 million
people

2°C
18% insects
16% plants
8% vertebrates
Ice free once a
decade

Mostly
disappear

411 million
people

8

One of the greatest contributers to
global warming is anthropogenic CO2, due
to the burning of fossil fuels (Figure 1-1, c,
d). Though release of anthropogenic CO2
started during the Industrial Revolution,
half of the total anthropogenic CO2
emissions were released between 1970-
2010, with 78% of them due to fossil fuel
combustion and industrial processes.
7

This increase in greenhouse gas emissions is due to the significant global population
increase, from 2 billion in 1927, to 7.6 billion in 2018. GHG emissions today come from
electricity, industry, transportation, buildings, agriculture and other energy needs (Figure
1-2).
Agriculture is one source of GHG’s not directly caused by fossil fuels. Figure 1-2 has
agriculture contributing 24% of greenhouse gases, but other estimates are as high as 51%.
9

Agriculture’s impact should not be ignored, especially as it is expected to grow 80% by
2050, whereas energy-related CO2 is estimated to increase 15% by 2040.
10,11
Agriculture
also releases mostly methane, which is 30 times as potent as a GHG as CO2. This estimate
of CO2 increase is based on the movement towards renewable energy, and the U.S. will
have to make significant changes to achieve it. The U.S. in 2017 consumed 9.8x10
16
Btu’s
of energy (80% being from burning fossil fuels), which is 17% of the 5.84x10
17
Btu’s
worldwide (Figure 1-3, a).
12
As the demand for energy in the U.S. is huge, a shift towards
renewable energies is the only way to cut back on energy-related emission. An ideal
situation would be to cut CO2 emissions from energy to half of 2011 levels. To do this, one

Figure 1-2: Global greenhouse gas
emissions by economic sector.
11

9

model predicts that 66% of global energy needs to come from renewable energy sources
(and more than 27% from solar energy).
13

Renewable energy sources are naturally replenished on a human timescale, such as
solar, wind, tides, waves and geothermal, making them an ideal energy source.
Unfortunately, in 2017 only about 11% of the total energy consumption in the U.S. was
renewable energy (Figure 1-3, a). Of this, 45% was from biomass, which still releases
GHGs, and is not as readily available as other energy sources, such as wind and solar
(Figure 1-3, b). The potential energy that is available from solar is greater than all other
energy sources today, combined (Figure 1-3, b). Despite it’s abundance, only 0.66% of
U.S.’s energy is from solar power. Inhibiting widespread implementation of solar power
is the technology used to harness its energy, or photovoltaic.

(a) (b)

Figure 1-3: (a) 2017 U.S. energy consumption by energy source
14
(b) Comparison of global
energy sources, with renewable energy being the amount available annually, the finite being the
total reserves.
15

10

Photovoltaics (PVs)
1.2.1 History of PVs
Alexandre-Edmond Becquerel first observed the photoelectric effect in 1839, when
by shining light on a solution containing silver chloride attached to two platinum
electrodes, he generated a voltage and current.
16
Albert Einstein published the atomic
theory behind the photoelectric effect in 1905, affording him the Nobel Prize in Physics in
1921.
17
In 1950, William Shockley laid the groundwork for a p-n junction, of which Daryl
Chapin, Clavin Fuller and Gerald Pearson at Bell Labs used to develop the first silicon
solar cell in 1954.
18
This foundation paved the way for commercialization of
photovoltaics.
Photovoltaics (PV) are the prevailing method of harnessing solar energy,
constituting 98% of solar energy conversion globally. The three principle obstacles facing
widespread implementation of photovoltaics are cost, scaling and intermittancy. The cost
of photovoltaics is still greater than fossil fuels, and the photovoltaics industry would need
to scale dramatically to meet global energy needs. Scaling photovoltaic technology
requires large swaths of land. The availability of solar energy at any location is
intermittant, and unpredictable; this requires the simultaneous scaling of energy-storage
technologies.
13

Despite these obstacles, photovoltaics remain one of the most promising
renewable energy solutions, due to the ubiquitous nature of sunlight and continuously
improving PV technology.
1.2.2 First- and second- generation PVs
First generation PV’s account for 90% of global PV technology. They are wafer-
based and composed of crystalline silicon (c-Si) (Figure 1-4).
19
Silicon is an an indirect
11

bandgap semiconductor and needs to be on the order of 100 μm thick to absorb sufficient
photons to produce photocurrent. Single crystal Si solar cells have reached solar
conversion efficiencies of 26% with lifetimes of 25-30 years, making them
commercializable (Figure 1-5).
19
The high cost of c-Si PVs comes from the energy required
to manufacture large amounts of pure material with long range order.

Figure 1-4: Typical solar PV device structures, divided into wafer-based and thin-film
technologies. Primary absorber layers are labeled in white, and thicknesses are shown to scale.
c-Si encompasses sc-Si and mc-Si technologies. GaAs cells use thin absorbing films but require
wafers as templates for crystal growth.
19

Second generation photovoltaics were developed to reduce the cost of PV
materials. They are thin film semiconductors with a direct bandgap, such as cadmium
telluride (CdTe) and copper-indium-gallium-selium (CIGS) PVs (Figure 1-4). These
materials have high absorption coefficients, allowing them to absorb light 10-100 times
more efficiently than silicon, which requires less material. CIGS technology has reached
efficiencies of up to 23%, making it competitive with c-Si (Figure 1-5). However, concerns
over the earth-abundance and the toxicity of these materials has led to third generation
PVs: organic photovoltaics.
12

Figure 1-5: 2018 Record research photovoltaic efficiencies.
13

1.2.3 Organic Photovoltaics (OPVs)
OPVs are thin film technologies that function differently than their inorganic
counterpart (Figure 1-4). Organic materials have dielectric constants that are a factor of
four less than inorganic materials, so, instead of directly forming free charges (a hole and
an electron) when a photon is absorbed, an excited state, or exciton, is produced. To split
an exciton into free charges, a chemical potential gradient is required; this is applied
through the contact of two materials with different energetics.
20
The photoelectric
processes undergone in OPVs is discussed in more detail in Chapter 2.
Anthracene was the first organic compound in which the photoelectric effect was
observed, by Pochettino in 1906.
21
The first organic photovoltaic was a bilayer device,
with copper phthalcyanine (CuPc) as a donor and a perylene tetracarboxylic derivative
acceptor, developed by C. Tang in 1986.
22
In 1993, the first organic solar cell with
fullerene (C60) as an acceptor was made.
23
The excellent charge conduction and high
electron affinity of fullerene made it the premier acceptor used in OPVs. Subsequent
research focused on optimizing the donor used in conjunction with C60, as well as device
physics and modelling. Current perspectives on the state of the art in organic
photovoltaics is given in Chapter 2.
PV technology is constantly being improved upon, but it does not change the
intermittancy with which sunlight hits the Earth. Batteries allow energy harvested by
PV’s to be utilized at night and on cloudy days for PV’s integrated in homes.
Unfortunately, they don’t provide long term energy storage for transportation
applications or large industrial processes. This is enabled by converting solar energy
directly to chemical energy, such as in hydrogen bonds (H2).
14

Hydrogen Economy
As climate change has been attributed to the release of GHGs from burning fossil
fuels, one promising solution would be to use hydrogen instead; this is known as the
“hydrogen economy”. Hydrogen is converted in a fuel cell into electric energy by the
reaction in eqn(1-1):
(1-1) 2 𝐻 2
+ 𝑂 2
↔ 2 𝐻 2
𝑂 − ∆ 𝐺
(1-2) 2 𝐻 2
↔ 4 𝐻 +
+ 4 𝑒 −

(1-3) 𝑂 2
+ 4 𝐻 +
+ 4 𝑒 −
↔ 2 𝐻 2
𝑂

Where the Gibbs free energy, ΔG, is 236 kJ mol
-1
.
24
The half reactions, eqns (1-2) and
(1-3) occur at the negative and positive electrodes of a fuel cell, respectively.
1.3.1 History of the Fuel Cell
Christian Friedrich Schoenbein discovered the principles of the fuel cell in 1839. It
was demonstrated in 1845 by Sir William Grove, who was deemed the “Father of the Fuel
Cell.” In 1939, Francis Bacon made the first stationary hydrogen-air fuel cell, known as
the Bacon Cell. Fuel cell technology did not gain popularity until the 1990’s.
25,26
Fuel cell
technology is available today, with the first fuel cell car commercially available in 2013,
the Hyundai Tucson FCEV. There are 39 fuel cell stations in the US, with 35 of them in
California.
27

It is clear from eqn (1-1) that the only byproduct in a hydrogen fuel cell is water.
Elemental hydrogen is not common, and so must be derived from another primary energy
source. Currently, fossil fuels are used to produce hydrogen (Figure 1-6). An ideal
“hydrogen economy” would be independent of fossil fuels, and utilize renewable energy
to electrolyze water for hydrogen. The one direct pathway from primary renewable energy
to hydrogen energy is through photoelectrochemical energy.
15

Figure 1-6: A sustainable hydrogen economy.
28

1.3.2 Photoelectrochemistry
The field of photoelectrochemistry actually originated through the development of
photography. Some of the first photographic images were printed on paper coated with
silver halide, made by Fox Talbot in 1839. In 1883, Hermann Vogel was able to extend
the photosensitivity to longer wavelengths by sensitizing the silver halide emulsions with
a dye.
29
J. Moser carried this concept over to photoelectrochemical (PEC) cells in 1887
by using the dye erythosine on silver halide electrodes.
30
Photoassisted water splitting
was first demonstrated in 1921 by Buar and Rebmann, using AgCl/TlCl in water and UV
light.
31
Subsequent studies expanded the field of photoelectrochemistry, with a surge of
16

publications during the oil crisis of 1973, when the Organization of Arab Petroleum
Exporting countries proclaimed an oil embargo.
32
The 21
st
century saw an outpouring of
research as alternative fuels are found to be critical to our world’s future.
A photoelectrochemical (PEC) cell consists of a photoactive semiconductor
electrode (either n- or p- type) and a metal counter-electrode, immersed in a redox
electrolyte. Two types of PEC cells were first developed: a regenerative PEC cell, in which
it converts light to electric power with no net chemical change (ΔG=0) and a
photoelectrosynthetic cell, that utilizes the phototelectric power to aid a chemical reaction
( ∆ 𝐺 ≶ 0). A phototelectrolytic cell has ΔG>0, for example, 𝐻 2
𝑂 → 𝐻 2
+
1
2
𝑂 2
.
1.3.3 Photocatalytic Water Splitting
Photolysis of water occurs through two half reactions, reduction at the photocathode,
eqn (1-4), and oxidation at the photoanode, eqn (1-5). The energy required to drive the
reaction forward depends on the energy of the reaction pathway involved. The pH of the
solution facilitates different reaction pathways for the reduction of H2 and oxidation of O2
half reactions (eqns ( 1 - 4) and ( 1 - 5), respectively). The free energy change for water
splitting is ΔE=-1.23 V.
Alkaline Conditions Acidic Conditions
(1-4) 2 𝐻 2
𝑂 + 2 𝑒 −
→ 𝐻 2
+ 2 𝑂𝐻
−
2 𝐻 +
+ 2 𝑒 −
→ 𝐻 2

(1-5) 2 𝑂𝐻
−
→
1
2
𝑂 2
+ 𝐻 2
𝑂 + 2 𝑒 −
𝐻 2
𝑂 →
1
2
𝑂 2
+ 2 𝐻 +
+ 2 𝑒 −

Photoelectrochemical (PEC) water splitting uses nanomaterials to harvest solar
energy and catalyze the water to hydrogen gas reaction. The seminal paper on
electrochemical photolysis of water was published in 1972 by Fujishima and Honda using
a titanium dioxide photoanode with a platinum cathode.
33
TiO2 is the preferrred material
for PEC reactors, though various other inorganic materials, such as metal oxides,
17

oxynitrides, n- and p- GaP are also researched, to improve low solar-to-hydrogen (STH)
conversion efficiencies (eqn (1-6).
34

(1-6) 𝑆𝑇 𝐻 =
𝐽 𝑆𝐶
( mA cm
− 2
) × 1 . 23 V
𝑃 𝑖𝑛
( mW cm
− 2
)

Many different device designs and material sets have been investigated, with the
goal of a single catalyst that can undergo both reduction and oxidation. However,
overpotential losses at the cathode and anode, as well as low conductivity in solution,
requires the applied voltage for water splitting to be 1.5-1.8V. This high voltage is difficult
for a single material to overcome independent of an external bias.
35
Thus,
semiconductors for either hydrogen evolution, oxygen evolution or overall water splitting
have been investigated (Figure 1-7). However, many are unstable in the harsh acidic or
alkaline conditions, or they are limited by their STH conversion efficiency.

Figure 1-7: Band edges of semiconductors for solar water splitting

Limits for commercialization have been set at >10% efficiency with a 10 year life
stability, and hydrogen fuel being cost competitive with fossil fuels.
36
The ability to utilize
solar energy and convert it into chemical energy is naturally done in plants through
photosynthesis Biomimicry has paved the way for scientists to develop artificial
photoysnthesis.

18

1.3.1 Natural Photosynthesis
Photoelectrochemical water splitting is often referred to as artificial
photosynthesis, as it mimics the processes undergone by photosynthetic organisms. In
photosynthesis, energy from sunlight is harvested by complexes, which subsequently
transfer this excited-state energy within the chromophores to a reaction site. At the
reaction site, the excited state separate into charges. These charges are utilized to catalyze
biochemical reactions, to generate stored chemical potential.
37
Lessons that chemists
have learned from photosynthesis are applied to PEC water splitting, such as (1)
development of light harvesting complexes and (2) charge separation by an electron
donor-acceptor moiety. Reaction centers are surrounded by ~200 complexes that
concentrate solar energy and transfer it to the reaction center. This light harvesting
antenna allows plants to thrive under low-light conditions. Not only are many complexes
utilized, but the complexes’ size, shape and molar absorptivity also influences their light
harvesting ability. Arrangement of chromophores in the antennae and reaction center is
critical to allow for energy transfer and inhibit quenching by charge transfer.
Natural photosynthesis in plants is an exemplar for designing reaction schemes,

(a) (b)
Figure 1-8: (a) Z-scheme in natural photosynthesis and (b) energy diagram of
photocatalytic water splitting by two-step photoexcitation.
38

19

specifically the Z-scheme (see Figure 1-8, a). Photosystem II (PSII) contains light
harvesting chlorophyll that absorbs photons and transport energy to reaction centers,
which generate electron-hole pairs. The pairs charge separate and split water into
oxygen, protons and electrons. The protons accumulate to build electric potential for
ADP to ATP conversion while the electrons are transported to photosystem I (PSI), to be
excited again. The excited electrons in PSI are used to reduce NADP
+
to NADPH.
NADPH and ATP are used to convert CO2 to sugars (Figure 1-8, a).
38
Artificial
photosynthesis mimics this process with synthesized complexes and carefully chosen
dyes and semiconductors (Figure 1-8, b).
1.3.2 Artificial Photosynthesis
The two different types of PEC cells for water splitting are based on light harvesting
using dyes (Figure 1-9, a, b) and those that use PV’s immersed in an aqueous electrolyte
(Figure 1-9, c,d).
39
Type I and II PEC devices are reactors that utilize particle suspensions,
typically with nanomaterials that are loaded with a dye to aid in reduction and/or
oxidation (see Figure 1-9, a and b).

20

Production Cost = $1.60 / kg H2

Production Cost = $3.20 / kg H2

Production Cost = $10.40 / kg
H2

Production Cost = $4.10 / kg H2

Figure 1-9: (a) Type 1, (b) Type 2, (c) Type 3 and (d) Type 4 PEC reactors, and the associated
STH efficiency required for commercialization as well as final production cost per kg H 2.
39

In Type I, the particles in suspension contain both photoanode and photocathode, so
hydrogen and oxygen evolution reactions occur in the same vessel. By not having a
physical separator, the hydrogen and oxygen can recombine, making it an explosive
mixture. Type II reactors separate the two half reactions, and contains an additional
redox shuttle. The simple materials and design reduces overall costs, so only 5-10% STH
conversion efficiency is required to be competitive with gasoline, making them
economically feasible.
40
Variations on these PEC cells include: photosensitizers (PS)
bound to the surface of nanoparticle semiconductors, bound molecular PS and
photocatalysts, or molecular PS and photocatalysts in solution. Ruthenium (II)
polypyridyl, [Ru(bpy)3]
2+
, is the most studied photosensitizer since it was discovered in
21

the 1970’s that it underwent electron transfer to a sacrificial acceptor.
41

PECs built with the light absorbers being solar cells that are immersed, or within
the vicinity of, an aqueous electolyte are denoted Type III and IV, and require an STH of
10% and 15% (respectively) to be cost competitive with gasoline (Figure 1-9, c and d).
42

Development of Type III and IV devices has corresponded with the growth of PV
technology. In the early 2000s, high efficiencies were achieved based on single junction
III-V materials, and Si/III-V two-junction architectures, which reached STH conversion
efficiencies of 12% and 18%, respectively. Since 2010, the main research tasks in PEC
development have been to: (1) replace noble metal electrocatalysts with earth-abundant
materials utilizing metal-oxides as stable light absorbers and; (2) to create a fully
integrated device (close contact between the light absorbers and catalysts).
34
In 2013,
the first organic PEC (OPEC) was developed by Bourgeteau et al. using an OPV motif:
PEDOT:PSS/P3HT:PCBM/TiO2:MoSx. The thick HER photocatalyst, TiO2:MoSx helped
protect the organics from the electrolyte solution (Figure 1-10).
43
The OPV generates a
charge that the catalyst utilizes for hydrogen
evolution, and in turn the catalyst protects the
OPV from exposure to the electrolyte, which
would decompose it. In this device, light
unharvested that passes through the OPV is
then transmitted and lost in the electrolyte
solution, whereas as an OPV it would be
reflected back to be harvested. Optical losses
in PECs account for significant STH conversion
loss, which were recently modeled by Cendula et al, who found that more than half the

Figure 1-10: Architecture of the
MoS 3:TiO 2/P3HT:PCBM H 2 evolving
photocathode inserted in the cell.
43

22

visible light in a device is lost through transmission.
44

Though PEC water splitting technologies have not advanced as quickly as OPVs,
research into these devices is growing as STH conversion efficiencies approach
commercializable values.
Current Study
The work herein addresses multiple approaches to tackling the problems with
developing alternative energy technologies through energy and electron management. In
Chapter 2, optimizing the OPV bilayer device structure for improved efficiency and
stability is investigated. This is done by incorporating an electron transport layer of zinc
oxide nanoparticles in an inverted device structure. This was found to have superior
device performance than the conventional structure counterpart.
Chapter 3 discusses a novel supramolecular light harvesting unit (LHU) used in
the hydrogen evolution reaction (HER) to achieve solar water splitting. The LHU was
designed as a microparticulate OPV, to be used in conjunction with an oxygen or hydrogen
evolution catalyst. The LHU has benefits from both particulate (Type 1 and 2) and
substrate (Type 3 and 4) devices; the materials are inexpensive, so costs are kept low,
while generating a current that is directly used for water splitting. This will ideally
improve energy and electron transfer to a HER catalyst.
Chapter 4 takes a molecular approach towards artificial photosynthesis, by using
chirality to drive assembly of donor-acceptor molecules. Intramolecular charge transfer
states were found to be an inherent property of the dyads synthesized, allowing for
intramolecular electron transfer. These are ideal candidates as light harvesting
complexes in artificial photosynthesis.
45

23

Bibliography
1. Arrhenius, S. On the Influence of Carbonic Acid in the Air upon the Temperature
of the Ground. Philos. Magaine J. Sci. 41, 237–276 (1896).
2. Callendar, G. S. The Artificial Production of Carbon Dioxide and its Influence on
Temperature. Q. J. R. Meterological Soc. 64, 223–224 (1938).
3. Shaftel, H. Global Climate Change. NASA (2018). Available at:
https://climate.nasa.gov/causes/.
4. Keeling C.D. The concentration and isotopic aboundances of carbon dioxide in the
atmosphere. Tellus 12, 200–203 (1960).
5. Weart, S. The discovery of global warming. (Harvard University Press, 2008).
6. IPCC. Principles Governing IPCC Work. (1998).
7. Intergovernmental Panel on Climate Change. Climate Change 2014 Synthesis
Report Fifth Assessment Report. (2018). Available at: http://ar5-
syr.ipcc.ch/topic_summary.php.
8. Intergovernmental Panel on Climate Change. Special Report: Global Warming of
1.5C. WMO, UNEP (2018). Available at: https://www.ipcc.ch/sr15/.
9. Goodland, R. & Anhang, J. Livestock and Climate Change: What if the key actors
in climate change are...cows, pigs and chickens? World Watch 22, (2009).
10. Tilman, D. & Clark, M. Global diets link environmental sustainability and human
health. Nature 515, 518–522 (2014).
11. US Environmental Protection Agency. Global Greenhouse Gas Emissions Data.
(2017). Available at: https://www.epa.gov/ghgemissions/global-greenhouse-gas-
emissions-data.
12. U.S. Energy Information Administration. Frequently Asked Questions. (2018).
24

Available at: https://www.eia.gov/tools/faqs/faq.php?id=87&t=1.
13. MIT Energy Initiative. The Future of Solar Energy. (2015).
14. US Energy Information Administration. Monthly Energy Review. (2018).
15. Perez, R. & Perez, M. A Fundamental Look at the Energy Reserves for the Planet.
Wind Energy 9–11 (2016).
16. Becquerel, E., Becquerel, H. & Becquerel, C. Mémoire sur les effets électriques
produits sous l’influence des rayons solaires. Comptes Rendus 9, 561–567 (1839).
17. Einstein, A. Concerning an Heuristic Point of View Toward the Emission and
Transformation of Light. Ann. Phys. 17, 132 (1905).
18. Mertens, K. & Hanser, K. F. Photovoltaics : Fundamentals, Technology and
Practice. (John Wiley & Sons, Inc., 2013).
19. Jean, J., Brown, P. R., Jaffe, R. L., Buonassisi, T. & Bulović, V. Pathways for solar
photovoltaics. Energy Environ. Sci. 8, 1200–1219 (2015).
20. Schlenker, C. W. & Thompson, M. E. The molecular nature of photovoltage losses
in organic solar cells. Chem. Commun. (Camb). 47, 3702–3716 (2011).
21. Pochettino, A. & Sella, A. Photoelectric behavior of anthracene. Acad. Lincei
Rendus 15, 355–363 (1906).
22. Tang, C. W. Two-layer organic photovoltaic cell. Appl. Phys. Lett. 48, (1986).
23. Sariciftci, N. S. et al. Semiconducting polymer-buckminsterfullerene
heterojunctions: Diodes, photodiodes, and photovoltaic cells. Appl. Phys. Lett.
62, (1993).
24. Sørensen, B. & Spazzafumo, G. Hydrogen. in Hydrogen and Fuel Cells (Third
Edition) (eds. Sørensen, B. & Spazzafumo, G.) 107–220 (Academic Press, 2018).
doi:https://doi.org/10.1016/B978-0-08-100708-2.00003-5
25

25. National Hydrogen Association; US Department Energy. The History of
Hydrogen. (2010).
26. Hultman, M. & Nordlund, C. Energizing technology: Expectations of fuel cells and
the hydrogen economy, 1990-2005. Hist. Technol. 29, 33–53 (2013).
27. U.S. Department of Energy. Alternative Fueling Station Locator. (2019). Available
at: https://afdc.energy.gov/stations/#/find/nearest?fuel=HY.
28. Rand, D. A. J. & Dell, R. M. Chapter 1 Why Hydrogen Energy? in Hydrogen
Energy: Challenges and Prospects 1–33 (The Royal Society of Chemistry, 2008).
doi:10.1039/9781847558022-00001
29. Graetzel, M. Photoelectrochemical Cells. Nature 414, 338–344 (2001).
30. Moser, J. Notiz ü ber die Verstä rkung photoelectrischer Strö me durch optische
Sensibilisierung. Monatshefte für Chemie 8, 373 (1887).
31. Baur, E. & Rebmann, A. Über Versuche zur Photolyse des Wassers. Helv. Chim.
Acta 4, 256–262 (1921).
32. E. Funk, J. Thermochemical hydrogen production: past and present. Int. J.
Hydrogen Energy 26, 185–190 (2001).
33. Fujishima, A. & Honda, K. Electrochemical Photolysis of Water at a
Semiconductor Electrode. Nature 238, 37–38 (1972).
34. Ager, J. W., Shaner, M. R., Walczak, K. A., Sharp, I. D. & Ardo, S. Experimental
Demonstrations of Spontaneous, Solar-Driven Photoelectrochemical Water
Splitting. Energy Environ. Sci. 8, 2811–2824 (2015).
35. Steier, L. & Holliday, S. A bright outlook on organic photoelectrochemical cells for
water splitting. J. Mater. Chem. A 6, 21809–21826 (2018).
36. Bard, A. J. & Fox, M. A. Artificial Photosynthesis: Solar Splitting of Water to
26

Hydrogen and Oxygen. Acc. Chem. Res. 28, 141–145 (1995).
37. Scholes, G. D., Fleming, G. R., Olaya-Castro, A. & Van Grondelle, R. Lessons from
nature about solar light harvesting. Nat. Chem. 3, 763–774 (2011).
38. Wang, Y. et al. Mimicking Natural Photosynthesis: Solar to Renewable H2Fuel
Synthesis by Z-Scheme Water Splitting Systems. Chem. Rev. 118, 5201–5241
(2018).
39. Pinaud, B. A. et al. Technical and economic feasibility of centralized facilities for
solar hydrogen production via photocatalysis and photoelectrochemistry. Energy
Environ. Sci. 6, 1983–2002 (2013).
40. James, B. D., Baum, G. N., Perez, J. & Baum, K. N. Technoeconomic Analysis of
Photoelectrochemical (PEC) Hydrogen Production. (2009).
41. Kärkäs, M. D., Verho, O., Johnston, E. V. & Åkermark, B. Artificial
photosynthesis: Molecular systems for catalytic water oxidation. Chem. Rev. 114,
11863–12001 (2014).
42. Fabian, D. M. et al. Particle suspension reactors and materials for solar-driven
water splitting. Energy Environ. Sci. 8, 2825–2850 (2015).
43. Bourgeteau, T. et al. A H2-evolving photocathode based on direct sensitization of
MoS3 with an organic photovoltaic cell. Energy Environ. Sci. 6, 2706–2713
(2013).
44. Cendula, P., Steier, L., Losio, P. A., Grätzel, M. & Schumacher, J. O. Analysis of
Optical Losses in a Photoelectrochemical Cell: A Tool for Precise Absorptance
Estimation. Adv. Funct. Mater. 28, 1702768
45. Li, X. et al. Engineering heterogeneous semiconductors for solar water splitting. J.
Mater. Chem. A 3, 2485–2534 (2015).
27

2 Electron Transport Layer in
Inverted OPVs

28

Table of Contents

2.1 Introduction .................................................................................................................... 29
2.1.1 Photocurrent Generation in OPVs ....................................................................................................... 29
2.1.1.1 Photon Absorption ..................................................................................................................... 30
2.1.1.2 Exciton Diffusion ........................................................................................................................ 31
2.1.1.3 Charge Transfer .......................................................................................................................... 33
2.1.1.4 Charge Separation ...................................................................................................................... 33
2.1.1.5 Charge Transport ........................................................................................................................ 34
2.2 Current Perspectives in OPVs ........................................................................................... 34
2.2.1 OPV Characterization ........................................................................................................................... 35
2.2.2 State-of-the-Art OPVs .......................................................................................................................... 37
2.2.2.1 Polymer Solar Cells (PSC) ............................................................................................................ 37
2.2.2.2 Small Molecule Solar Cells (SM-SC) ............................................................................................ 38
2.2.2.3 Perovskite Solar Cells ................................................................................................................. 39
2.2.2.4 Inverted OPVs ............................................................................................................................. 39
2.2.3 ZnO as an Electron Transport Layer ..................................................................................................... 40
2.3 Results ............................................................................................................................. 42
2.3.1 Optimization of Electron and Hole Transport in iOPVs ........................................................................ 43
2.3.1.1 ZnO Deposition Characterization ............................................................................................... 44
2.3.1.2 ZnO np post-treatment .............................................................................................................. 44
2.3.1.3 Effect of MoO3 thickness on device performance ...................................................................... 46
2.3.2 Optimizing Electron Transfer in ZnO .................................................................................................... 47
2.3.3 Donor/Acceptors in an Inverted Bilayer Device................................................................................... 48
2.3.4 Role of the Acceptor in an i-OPV ......................................................................................................... 50
2.3.4.1 Fullerene Devices ....................................................................................................................... 50
2.3.4.2 PTCBI Devices ............................................................................................................................. 51
2.3.5 Optimized i-OPV................................................................................................................................... 51
2.4 Conclusions ..................................................................................................................... 53
2.4.1 Future Work ......................................................................................................................................... 53
2.5 Experimental Methods ..................................................................................................... 55
2.5.1 ZnO np Synthesis ................................................................................................................................. 55
2.6 Bibliography .................................................................................................................... 56

29

2.1 Introduction
OPVs are third generation solar cells; they are an emerging thin-film technology, and
have much potential after the success of their counterpart, organic light emitting diodes
(OLEDs).
1
They can be composed of small molecules, polymers or perovskites, and formed
in either a bulk- or bilayer- heterojunction. OPVs can be produced through cost effective
roll-to-roll processing, allowing for inexpensive, large scale manufacturing on flexible
substrates. This enables them to be semi-transparent and portable. Despite all these
advantages, OPVs still suffer from inefficient power conversion efficiencies and poor long
term stability under ambient conditions that would enable them to be a commerically
scalable technology.
2

2.1.1 Photocurrent Generation in OPVs
OPVs generate photocurrent differently
than their inorganic counterpart; optical
excitation generates excitons (a bound e
-
/h
+
pair) in organic materials (versus free
charges in inorganic semiconductors). The
organic active layer is sandwiched between
the two electrodes, and is comprised of a
donor material (D) and an acceptor counterpart (A), whose energy level offset enables
photocurrent generation (see Figure 2-1 (a)). These are either oriented in a bilayer
configuration (one distinct D/A interface), or a bulk heterojunction (mixed D/A) (see
Figure 2-1 (b) and (c)).
2

Figure 2-1: (a) Energy level
diagram of the layers in an OPV,
(b) a bilayer configuration and (c) a
bulk heterojunction
2

30

Photocurrent generation occurs through multiple steps: photon absorption, exciton
diffusion, charge transfer, charge separation, charge transport and finally, current
generation. A schematic of the processes described above is given in Figure 2-2.

Figure 2-2: Photophysical processes occurring in an OPV to generate current.

2.1.1.1 Photon Absorption
Photons are absorbed in the active layer of the OPV by both the D and A. Organic
molecules in OPVs have very high extinction coefficients compared to their inorganic
counterparts, allowing them to absorb a large amount of solar photon flux.
1
Organic D/A
heterojunctions are able to capture more photons than their c-Si and GaAs counterparts
(Figure 2-3). In addition, with the development of ternary and tandem devices, photons
in the red and near-infrared (NIR) are able to be harvested and utilized.
31

To absorb as much of the solar photon flux possible, it is prudent when designing
D/A heterojunctions that the donor and acceptor absorb at different wavelengths. In
Figure 2-3 (b), the absorption spectra of C60 (or fullerene, the most universally used
acceptor) and CuPc (donor), are in the blue and red region, respectively to optimize
photon absorption.
3
C60 has limited absorption in the blue. Thus, extensive research on
donors with extended absorption has been done to optimize overall photon absorption in
the device. This collectively absorbs the most photons over the entire OPV.
2.1.1.2 Exciton Diffusion
Immediately following photon absorption, an exciton is formed due to the low
dielectric constant of organic materials. These excitons diffuse to the D/A interface by two
mechanisms: dipole-dipole interactions (also known as Förster resonance energy
transfer, FRET), or electron exchange interactions (also known as Dexter energy transfer,
DET) (see Figure 2-4).
4
FRET is the process by which energy that is absorbed by a donor

(a) (b)
Figure 2-3: (a) top panel: Solar photon flux as a function of wavelength,
middle panel: the absorption coefficient (α) for the active region of an
archetypal organic donor/acceptor pair, CuPc/C 60 (solid red trace), gallium
arsenide (dashed black trace) and crystalline silicon (dotted blue trace).
Bottom panel: percentage of solar photons captured by 1000 Å active layers
with α identical to those in the middle panel, assuming a two-pass optical
path, (b) Spectra of solar irradiance and absorption of CuPC and C60
(chemical structures shown on the right)

32

(to make an excited state, D*) is transferred to an acceptor molecule by the overlap of
dipolar electric fields of D* with A. This is facilitated when the electronic transition dipole
moment (ETDM) of the donor chromophore is aligned with that of the acceptor. The
ETDM is the oscillatory strength of an atom’s electrons transition between two orbits
when interacting with an electric field. FRET can occur on length scales longer than the
sum of a molecules van der Waals radii, and is the primary mechanism for singlet exciton
diffusion in OPVs.
1
Electron exchange, or Dexter transfer, occurs through the direct
overlap of the orbitals of the excited state molecule and ground state molecule. This limits
it to the van der Waals radii of the two molecules. DET is the primary mechanism of
triplet exciton diffusion in OPV’s, since their lifetimes are on the order of μs-ms. The
lifetime of the excited state impacts the exciton diffusion length, lD, which is the length
the exciton diffuses befor either decaying to the ground state, or forming a charge transfer
state. Singlet excitons in organic
materials typically diffuse on the
order of 10 nm before either
transferring charge to the acceptor
(favored) or radiatively
recombining (an OPV loss
mechanism).
5

Figure 2-4: Energy diagram for FRET (top)
and DET (middle and bottom) processes

33

2.1.1.3 Charge Transfer
An exciton formed in the donor that
diffuses to the D/A interface will dissociate
rapidly through electron transfer to the
acceptor, creating a charge transfer state
between the D/A pair. Thermodynamically,
this occurs given that the energy of the exciton
in the donor is greater than that of the D/A
charge transfer state (D
+
A
-
). The energy of
the CT state generated must be overcome to
dissociate into free charges (Figure 2-5).
2.1.1.4 Charge Separation
After CT state formation, it can either generate free charges (resulting in
photocurrent) or recombine (another OPV loss mechanism). The probability that a CT
state will dissociate at a D/A interface is given by the Onsager-Braun model.
6–8
CT state
recombination is either radiative or non-radiative. Radiative recombination is the result
of back electron transfer from the acceptor to the donor, thus regenerating the donor
exciton, which can decay to the ground state. Non-radiative recombination undergoes
geminate recombination (electron-hole pair recombines into original CT state, which
relaxes to the ground state) or non-geminate recombination (two independently
generated CT states of opposite polarity that come in contact to recombine).
9
These are
all loss pathways for OPV photocurrent generation. Free charges that form then transport
to the corresponding electrodes.

Figure 2-5: (a) Molecular orbital
(MO) description for the charge
transfer state formation in a donor-
acceptor system (b) Description for
CT state exiplex emission energy
h CT
Em
for the MO’s in (a).
1

34

2.1.1.5 Charge Transport
Once charges are generated, they then transport through the active layer to the
electrodes via localized hopping. Charge transfer occurs through hopping mechanisms,
as the charge is localized on individual molecules since intermolecular interactions in
organic semiconductors are weak (~1 cm
2
V
-1
s
-1
versus 1000 cm
2
V
-1
s
-1
in inorganic
photovoltaic materials).
10
Charge transport rates are expressed by Marcus Theory,
developed by Rudolph Marcus in the 1950’s, and for which he won the 1992 Nobel Prize
in Chemistry.
11
Ideally, charge transfport is balanced in the donor and acceptor material.
Low charge mobility can impact device performance, resulting in bimolecular charge
recombination (reducing photocurrent), reducing the fill factor through high resistance,
and low local charge mobility can impact how efficiently charges separate. Fullerene’s 3-
D shape enables excellent charge extraction, however, it has limited photon absorption.
For decades, non fullerene acceptors (NFA) could not match the charge extraction
generated in a fullerene device. Only recently have devices with NFA been competitive
with fullerene devices, truly enabling band gap tuning between the D/A.
2.2 Current Perspectives in OPVs
Given that all the above processes occur, and charges make it to the interface of the
organic material and electrode, the charges are extracted by the electrodes and generate
current. It is also possible that trap sites at this interface causes recombination. A
transparent conducting electrode (TCO) extracts charges, and allows light to penetrate
through it to the active layer. Currently, indium-doped tin oxide (ITO) is the preferred
transparent electrode that the active material is deposited on, while aluminum, silver or
gold are most commonly used as the top electrode. In conventional devices, holes are
35

extracted by ITO while electrons flow to the top electrode. ITO is one of the limiting
factors to commercialization, due to the high cost of indium.

2.2.1 OPV Characterization
To quantify the efficiency of charge extraction in an OPV, current-voltage curves are
measured. This is done by irradiating the OPV at room temperature with 1000 W m
-2
,
using an AM 1.5 global reference spectrum.
Applied voltage is swept from V<0 (reverse
bias) to V>0 (forward bias) and the current
density is measured (see Figure 2-6). In the
dark, the OPV behaves like a diode, whereas
under illumination a photocurrent is
generated, allowing current to flow through
the OPV. The short circuit current (JSC) is the
current at which the voltage is 0, whereas the
open circuit voltage (VOC) is the voltage at
which the current is 0.
The fill factor (FF) is the ratio of area of the filled gray rectangle (P max) to unfilled gray
rectangle (J sc × V oc) (Figure 2-6), and is given by equation ( 1).
( 1) 𝐹𝐹 =
𝑃 𝑚 𝑎𝑥 𝐽 𝑆𝐶
𝑉 𝑂𝐶

Figure 2-6: Simulated solar cell
electrical behavior in the dark
(dotted traces) and under
illumination (solid black and red
traces) given as a semi-log plot
(top) and a linear representation
(bottom).
2

36

Various properties of an OPV can impact
each of these characteristics. The JSC is limited
by photon harvesting, specifically the spectral
overlap of the donor and acceptor with the solar
spectrum, as well as the diffusion length and
dissociation of the excitons generated. The VOC
is related to the charge transfer state (CTS)
energy, ECT, which is the difference between the
CTS ground and excited state. The energy loss
in an OPV, ΔE, is given in equation ( 2).
12

( 2) ∆ 𝐸 = 𝐸 𝑜𝑝 𝑡 − 𝑒 𝑉 𝑂𝐶

Thus, the smaller the gap between Eopt and ECTS,
the less VOC loss. However, the energy offset must be large enough for excitons to
dissociate. This trade-off between VOC and JSC in device performance is a struggle when
finding matching donor/acceptor materials (see Figure 2-7). The FF is limited by low
carrier conductivity, and interfacial resistance.
Another important measurement for OPVs is the external quantum efficiency
(EQE). This is a measurement of the electron-hole pairs generated per incident photon,
measured at the short circuit condition (zero applied bias). This is differet from the
internal quantum efficiency, which is a measurement of the photons that are absorbed by
the solar cell to the number of charge carriers collected.

Figure 2-7: Schematic
representation of two hypothetical
donor materials embodying the
efficiency limiting trade-off between
suppressed voltage losses on the left
and robust spectral coverage on the
right. Note that E i* is identical for
both donors, making the enthalpic
driving force for charge transfer
equivalent in both cases.
2

37

2.2.2 State-of-the-Art OPVs
As previously mentioned, OPVs suffer from low efficiencies, and poor device
stability. Chemists, physicists and engineers work on improving OPV device efficiency
and stability through optimizing donors and acceptors energy levels and structure
through molecular engineering, improving device morphology and/or modelling the
various photoelectrochemical processes to design better molecules.
13–15
OPVs are
primarily classified by the material with which they are made: small molecules, polymers
or perovskites. OPV’s can be further categorized by the device architecture; bilayer,
wherein the donor and acceptor are stacked on top of each other with limited mixing
between layers, or bulk heterojunction, where the donor and acceptor are mixed together.
Devices can be of inverted orientation, a ternary device and/or multiple subcells
comprising one tandem device.
2.2.2.1 Polymer Solar Cells (PSC)
In 2000, Alan Heeger, Alan MacDiarmid and Hideki Shirakawa received the Nobel
Prize in Chemistry for the development of conductive polymers. As polymers are easily
synthesized, tuned, and solution-processable, they are ideal OPV candidates. The most
studied PSC is made with poly(3-hexathiophene) (P3HT) as a donor, and solution
processable fullerene, phenyl-C61-butyric acid methyl ester (PC61BM) as the acceptor ( and
PEDOT:PSS as a hole transport layer (HTL)). P3HT:PCBM make devices with unusually
high efficiencies, making them the standard in PSC’s.
16,17
Currently, research into donor
and acceptor hybrid polymer devices is popular. This would replace the small molecule
fullerene acceptors, which reduces the cost of production, allows readily tunable
absorption/energy levels between the D/A, as well as good has morphological stability
and improved mechanical durability. Other popoular research areas currently for
38

polymer devices is to build ternary devices (three components in the active layer),
incorporating additives to improve BHJ morphology and utilize benzodithiophene (BDT)
as a donor building block. BDT has demonstrated high stability and planar molecular
structure (for strong intermolecular pi-pi stacking).
18,19
All-polymer solar cells achieved
maximum efficiencies of 9% recently.
20,21

2.2.2.2 Small Molecule Solar Cells (SM-SC)
As in PSCs, current research topics focus on small molecule non-fullerene acceptors
(NFA). Acceptors with an A-D-A motif (also known as push-pull) have shown to have
good charge extraction and can be solution processed. One of the most popular push-pull
acceptors is ITIC, which was first synthesized in 2015 by Lin et. al, and resulted in a PCE
of 6.8%.
22
ITIC was novel in that the side chains allowed it to twist and not crystallize,
but there was still significant π-π interactions. Iterations on this design have further
optimized small molecule NFA. Perylene diimide (PDI) and naphthalene diimide (NDI)
are some of the most popular NFA components for their π-π interactions. Dimers are
made that twist to reduce acceptor-phase crystallinity, which enables better mixing with
donors in bulk heterojunctions.
19,23
Fluorinated pentacene, arylene diimides, BDT and
others are also popular NFA molecules and building blocks.
The majority of NFA development focuses on solution processable small molecules,
which enables them to be utilized with polymer donors. The most efficient solar cells to
date are a combination of polymer and small molecule, and NFAs devices have surpassed
solar efficiencies of fullerene devices. The most efficient single junction SC to date has a
polymer donor and non-fullerene acceptor has achieved 14.2% efficiency.
24
Tandem
devices recently reached a stunning 17.3% efficiency, pushing OSC commercialization to
within reach.
25

39

2.2.2.3 Perovskite Solar Cells
In 1991, Michael Grätzel developed a new technology, based on a thin TiO2 layer
coated with dye molecules in an electrolyte solution: the dye-sensitized solar cell
(DSSC).
26
This opened up another subset of organic photovoltaics, with excellent
efficiencies over 10%, but it was difficult to transfer this technology to a solid state system.
The evolution of the DSSC eventually led to perovskite solar cells. eventually leading to
the discovery of perovskite solar cells in 2009 by Miyasaka et. al., achieving 3.8%
efficiency, which was promptly optimized to 10% within 3 years by Snaith.
27,28
In 2012,
one of the first perovskite solar cells was reported to have a 10.9% efficiency, an
unprecedentedly high efficiency for an OPV at the time.
29
The methylammonium lead
iodide chloride form of perovskite is the standar. Perovskites have the benefits of being
solution processed, excellent optical, mechanical and electronic properties, as well as
achieving efficiencies over 20%.
30
However, perovskites suffer from instability, inhibiting
commercialization.
Instability in OPVs arises from oxygen and water diffusing into the active layer of the
device and creating trap sites for recombination pathways. Diffusion of gases can be
slowed down by encapsulating devices, but encapsulation also has its limits. For OPVs to
compete with existing technologies, they have to last for years. Thus, stabilizing OPVs by
optimizing the device structure is ideal.
2.2.2.4 Inverted OPVs
Inverted devices improve the stability of OPVs, making them the ideal device
architecture. The cathode in a conventional device is made of a low work function metal,
which allows diffusion of oxygen into the active layer, and is inherently reactive with
oxygen and water. Oxygen diffusion into the active layer creates trap sites that reduce the
40

overall conductivity of the organic material. Device stability is reduced dramatically if it
is not encapsulated.
31
By inverting the device architecture, a high work function layer
(such as a transition metal oxide
interfacial layer) can modify the
aluminum cathode and stabilize the
active layer.
32
Figure 2-8, a shows the
device architecture of a conventional
OPV (c-OPV) versus an inverted OPV (i-
OPV) with power conversion efficiency
(PCE) of normal and inverted devices
over time.
2.2.3 ZnO as an Electron Transport Layer
It is important to have a transparent bottom electrode in organic photovoltaics, to
allow sunlight to penetrate through the active layer. ITO is a highly conductive,
transparent material, that is the standard bottom electrode in organic electronics. In a
conventional device, it acts as the anode, whereas in an inverted device, ITO has to be
modified to harvest electrons. An electron transport layer is deposited on top of it to
facilitate charge extraction. A transition metal oxide, such as ZnO or TiO2, has ideal
energy level alignment with the active layer (Figure 2-9, a). Zinc oxide (ZnO) is ideal for
its optical and electrical properties; it is transparent and highly conducting.
31
ZnO
nanoparticles (np) also exhibit quantum confinement effects; as their size decreases their
band gap widens, enabling tuning from 3.2-3.6 eV (Figure 2-9, b).
34

Figure 2-8: PCE of an inverted and normal
device as a function of device operation
time (Inset: Device structure of a normal
device (left) versus an inverted device
(right))
33

41

Figure 2-9: (a) Work function of common OPV materials and selected metal oxides.
35
(b)
photoluminescence spectra of nanocrystalline ZnO with different crystal sizes36

ZnO np can be deposited by a variety of methods, such as atomic layer deposition
(ALD), the sol-gel method and the precipitation method. ALD is form of chemical vapor
deposition (CVD), but divides the chemical reaction into two half-reactions to control the
layer. This creates a stable, conformal coating, but requires high temperature.
37
The sol-
gel method creates ZnO networks through the formation of a colloid suspension (sol),
which then gelate to form a continuous liquid phase (gel). This creates an amorphous
layer with size controllable nanocrystals or nanorods, but has to be heated and coooled.
38

The precipitation method forms ZnO np through a base-initiated
hydrolysis/condensation reaction. This allows for size controllable nanoparticles,
synthesied at room temperature.
39
When considering processing techniques of a ZnO np
layer in a device, surface roughness of the film is one of the properties that can vary the
most. This is important, as a rough film can cause shorts if the active layer does not fully
cover it.
42

Comparing a layer of ZnO processed by ALD versus solution processed
(synthesized via the precipitation method) films, the solution processed films are
noticeably smoother (see Figure 2-10).

Figure 2-10: AFM images (10μm x 10μm) (a) ITO/ZnO/C60 (40 nm)
37
(b) ITO/CuPc; (c)
ITO/ALD ZnO and (d) ITO/spincast ZnO

Thus, it is expected that spin coating solution processed ZnO as an electron
transport layer in iOPVs will improve device efficiency.
2.3 Results
The current study investigates a solution-processed ZnO electron transport layer in
inverted, small molecule bilayer OPVs. The conditions for the ZnO np layer were
optimized with the standard device, CuPc as a donor and C60 as an acceptor. MoO3 was
used as a hole transport layer (Figure 2-9, a). Doping ZnO with aluminum to improve
electron transport and optimization of photon harvesting by utilizing different
donors/acceptors were also investigated.

43

2.3.1 Optimization of Electron and Hole Transport in iOPVs
As solution processed zinc oxide nanoparticles in bilayer OPVs had not previously
been investigated, the first step was to optimize the zinc oxide layer. This work was
completed in three steps:
(1) Synthesis of the ZnO nanoparticles
(2) ZnO deposition characterization
(3) Optimizing the inverted OPV performance.
The precipitation method was used to synthesize the ZnO np – these were then
solution processed on ITO. Solution processing was used because it allows deposition at
low temperatures, with control of the uniformity and thickness of the film. Solution
processing is done in with zinc oxide nanoparticles, which exhibit quantum confinement
effects. This results in size-dependent optical effects, resulting in a wide, tunable band
gap (3.2-3.6 eV).
34

Devices were characterized by measuring the
open circuit voltage (Voc), short circuit
current density (Jsc), fill factor (a ratio
between the maximum output power and the
theoretical output power, FF), external
quantum efficiency (EQE) and the power
conversion efficiency (PCE). The size of the
ZnO nanparticles was measured through
UV-Vis (see Figure 2-11). As the peak absorbance is around 330 nm, these zinc oxide
nanoparticles are less than 5 nm, as the correlation between size and absorption is
demonstrated in Figure 2-9, b.
250 275 300 325 350 375 400
0.0
0.1
0.2
0.3
0.4
0.5

Absorbance (a.u.)
Wavelength (nm)

Figure 2-11: UV-vis of zinc oxide
nanoparticles in ethanol.

44

2.3.1.1 ZnO Deposition Characterization
ZnO is a semiconductor with a high electron mobility and high degree of transparency
in the visible wavelength, it as an optimal cathode material (Figure 2-11). The goal was to
create a uniform ZnO nanoparticle layer of about 20 nm using solution processing, to
compare with devices previously made with a 20 nm layer of ZnO deposited by ALD.
40

The ZnO nanoparticles were spin cast with varying concentrations and rotation speeds
onto silicon substrates and the thickness was measured by ellipsometry. The spin cast
conditions varied were rotation speed, and solution concentration (Figure 2-12).
15 20 25 30
35
40
45
50
55
60
65
70
Thickness (nm)
Rotation (RPM, x100)
0 5 10 15 20 25
20
25
30
35
40

Thickness (nm)
Concentration (mg ZnO/mL)

Figure 2-12: (a) Thickness (nm) vs. rotation speed of spin coater (RPM) at
concentration of 20 mg/mL; (b) Thickness (nm) versus concentration (mg/mL) at 3000
RPM for 1 min.

There is little variation in thickness with concentrations from 5-15 mg/mL, whereas at
higher concentrations, the thickness increases. The thickness also increased at lower
rotation speeds. The optimal ZnO layer thickness (about 20 nm) was made by spin casting
at 3000 RPM for one minute using between 5-15 mg/mL ZnO solution.
2.3.1.2 ZnO np post-treatment
After the ZnO deposition conditions had been determined, OPV’s were assembled.
First, the post-processing treatment of the ZnO np was investigated. Previous methods
45

of processing interlayers include thermal annealing (TA) and uv-ozone (UVO) treatment
to remove the organic ligands that control the growth kinetics of the nanoparticles. Both
methods were investigated.
41,42

OPV devices were made with various post-treatment conditions, including: no
treatment, 10 min UVO treatment, 10 min TA at 150°C, and 10 min UVOC followed by 10
min thermal annealing at 150°C. The device performance for various post-treatment of
ZnO np is shown in Table 2-1 and Figure 2-13.
Table 2-1: Device performance for post-treatment of ZnO nanoparticles

i-OPV: ITO/ZnO (20nm)/C60 (40 nm)/CuPC (40 nm)/MoO3 (20nm)/Al
c-OPV: ITO/CuPC (40nm)/C60 (40nm)/BCP (10 nm)/Al

No treatment and UVOC+TA had the worst
performance, as seen by the low JSC, VOC, FF and
 compared to those with UVO or thermal
annealing treatment. These two devices are not
rectifying under forward bias, indicating a large
series resistance within the device (Figure 2-13).
UVO treatment of ZnO np makes the best
devices, with the highest JSC, VOC, FF and ,
(Table 2-1). Thus, UVO treatment was utilized
in following experiments. The low FF indicates
JSC (mA/cm
2
) VOC (V) FF (%)
i-OPV: No Treatment 3.26 0.38 0.23 0.29
i-OPV: UVOC 3.79 0.34 0.38 0.50
i-OPV: TA 3.72 0.34 0.31 0.39
iOPV: UVOC + TA 0.22 0.34 0.09 0.01
c-OPV 5.38 0.47 0.59 1.5

-1.00 -0.75 -0.50 -0.25 0.00 0.25 0.50 0.75 1.00
-8
-6
-4
-2
0
2
4
6

Current (mA/cm
2
)
Voltage (V)

Figure 2-13: Current versus voltage
for devices with: no treatment
(blue), UVO treatment (green),
thermal annealing (orange), both
thermal annealing and UVO (red)
and a standard (black).

46

a problem with charge extraction from either the electron transport layer (ZnO) or hole
transport layer (MoO3). The ZnO has been investigated for different thicknesses,
uniformity and treatment, so the thickness of the MoO3 was varied.
2.3.1.3 Effect of MoO3 thickness on device performance
To improve the fill factor of the devices, the thickness of the MoO3 extraction layer
was varied from 10-40 nm. The goal was to determine the best charge extraction efficiency
without compromising the optical properties.
Table 2-2: Device performance for inverted devices with varying MoO3 interlayers
MoO3 Thickness (nm) JSC (mA/cm
2
) VOC (V) FF (%)
i-OPV: 10 3.2 0.40 0.21 0.3
i-OPV: 20 4.3 0.47 0.32 0.6
i-OPV: 30 4.6 0.40 0.48 0.9
i-OPV: 40 4.2 0.37 0.50 0.8
c-OPV 5.4 047 0.59 1.5
i-OPV: ITO/ZnO (20nm)/C60 (40 nm)/CuPC (40 nm)/MoO3/Al
c-OPV: ITO/CuPC (40nm)/C60 (40nm)/BCP (10 nm)/Al

The results indicate that the devices with a
MoO3 layer of 20 nm or less have not yet
exhibited adopted the bulk properties (i.e.
work function) of the MoO3 (Table 2-2 and
Figure 2-14). A MoO3 layer of 40 nm was too
thick, so conductivity within the device was
inhibited. The only device that rectified under
forward bias was that with a 30 nm layer of
MoO3. Thinner layers of MoO3 had resistance
from poor contact with the electrode.

-1.00 -0.75 -0.50 -0.25 0.00 0.25 0.50 0.75 1.00
-8
-6
-4
-2
0
2
4
6

Current (mA/cm
2
)
Voltage (V)

Figure 2-14: Current versus voltage for
devices with 10 nm MoO3 (blue), 20
nm MoO3 (green), 30 nm MoO3
(orange), 40 nm MoO3 (red) and a
standard (black).

47

The work done so far has indicated that solution processed deposition with
5-15 mg/ml ZnO solution at around 3000 RPM for 1 minute gives an optimal cathode
interface layer (ZnO layer ranging from 20-25 nm). i-OPVs with a solution processed ZnO
layer demonstrate better performance than the ALD counterpart, having an overall power
conversion efficiency (η) of 0.9% versus 0.6% (Table 2-3). This along with the other
benefits (ease of synthesis, room temperature processing), is why solution processed zinc
oxide nanoparticles were studied in devices.
Table 2-3: Device Performance for Inverted and Normal Devices

JSC (mA/cm
2
) VOC (V) FF η (%)
c-OPV 5.38 0.47 0.59 1.5
i-OPV - ALD ZnO
37
3.66 0.33 0.52 0.6
i-OPV - Spincast ZnO 4.58 0.4 0.48 0.9
c-OPV: Glass/ITO/CuPC (40nm)/C60 (40nm)/BCP(10nm) /Al
i-OPV: Glass /ITO/ZnO (20nm)/C60 (40nm)/CuPC(40nm)/MoO3(30nm)/Al

ZnO np have proven to be a conductive electron transport layer, but the device
performance of iOPVs still do not exceed that of the conventional device. Doping ZnO
with aluminum will further modify the work function of the electron transport layer,
improving device efficiency.
2.3.2 Optimizing Electron Transfer in ZnO
To improve electron transport in the zinc oxide nanoparticles, aluminum was doped
into them. Aluminum-doped zinc oxide (AZO) is highly conductive and transparent,
making it an ideal electrode material. It has been used in spin-coated polymer OPVs and
proven to work as well as ITO, and was made by the procedure outlined by Murdoch et.
al.
46

48

Standard devices were made with an inverted structure with a ZnO and AZO layer.
These were compared with conventional devices (Table 2-4 and Figure 2-19).
Table 2-4: Device performance of inverted OPVs with an ETL of ZnO or AZO.

JSC (mA/cm
2
) VOC (V) FF PCE (%)
i – ZnO/C60/CuPc
2.3 0.4 0.3 0.3
i – AZO/C60/CuPc
2.3 0.4 0.4 0.4
c – CuPc/C60 5.3 0.5 0.6 1.6
i: ITO/ETL (20nm)/C60 (40nm)/CuPc (40nm)/ MoO3 (30nm)/Al
c: ITO/ CuPc (40nm)/ C60 (40nm)/BCP
(10nm)/Al

A distinct S-shape is seen in the J-V curve of
the inverted devices, indicating an imbalance
in charge extraction.
47
The poor device
performance is caused by the low JSC, which
is half that seen in the conventional device.
Trinh et al. also observed a reduced JSC from
conventional to inverted device (using the
same stack structure, except with ALD ZnO), which was attributed to imbalance in the
optical electric field distribution. This was investigated in devices with both AZO and ZnO
ETL layers.
37

2.3.3 Donor/Acceptors in an Inverted Bilayer Device
Inversion of the donor/acceptor layer in a bilayer device changes the optical field
distribution within it. Trinh et al. used a transfer-matrix formalism to model how the
optical electric field varies from a conventional to an inverted OPV (see Figure 2-16).
37

The optical electric field distribution similar, with long wavelength (600-800 nm) close
to the substrate and short wavelengths further away (400-600 nm) (Figure 2-16, a, b).
-1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0
-8
-6
-4
-2
0
2
4

Current (mA/cm2)
Voltage (V)

Figure 2-15: J-V curve of inverted
devices with ZnO (red) and AZO
(blue) as an ETL.
49

However, the absorbed power in a conventional device is 30% higher than an inverted
device (Figure 2-16, c, d). In a conventional device, red-light absorbing CuPc is closer to
the substrate, with blue- absorbing C60 further away, optimizing light absorption. In an
inverted device, the materials are switched, so photons are not harvested as well in a
CuPc/C60 devices, and the overall power output is reduced.

(e) (f)

Figure 2-16: Optical modeling of c-OPV and i-OPV: optical E-field in (a) c-OPV (D1), (b)
i-OPV (D2); absorbed power (Qj) of (c) D1 (integrated Qj = 0.19); (d) D2 (integrated
Qj = 0.28). The y-axis presents distances from the glass surface. OPV structure of (e)
conventional OPV and (f) inverted OPV.

Inverted devices with ZnO and AZO spincast ETL were optimized by investigating donors
that absorb at short wavelengths, and non-fullerene acceptors that absorb at long
wavelengths.
50

2.3.4 Role of the Acceptor in an i-OPV
To investigate how well an acceptor absorbs in an inverted device when it is either
blue- or red- absorbing, C60 and perylene- 3,4,9,10-bis-benzimidazole (PTCBI)
(respectivly) were used with donor molecule, NPD. NPD is a wide-band gap, transparent
material and thus exciton formation is localized in the acceptor.
48
C60 absorbs from ~300-
550 nm, whereas PTCBI absorbs from ~400-800 nm.
49

2.3.4.1 Fullerene Devices
Conventional devices with C60 as the acceptor have overall better device performance
than the inverted structure (see Table 2-5 and ).
Table 2-5: Device performance of OPVs with C60 and NPD.

JSC (mA/cm
2
) VOC (V) FF PCE (%)
i- ZnO/C60/NPD 2.0 0.8 0.5 0.8
i- AZO/C60/NPD 2.6 0.8 0.5 1.1
c- NPD/C60 3.3 0.9 0.5 1.5
i: ITO/ETL (20nm)/C60 (40nm)/NPD (40nm)/ MoO3 (30nm)/Al
c: ITO/ NPD (40nm)/ C60 (40nm)/BCP (10nm)/Al

These were the best devices made
thus far, due to the high VOC. The FF was
also higher than the devices thus studied,
demonstrating improved charge
extraction. Utilizing AZO as an ETL
versus ZnO resulted in a higher JSC, giving
a better device efficiency.

-1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0
-4
-2
0
2
4

Current (mA/cm2)
Voltage (V)

Figure 2-17: Inverted devices with ZnO
(red) and AZO (blue) as an ETL and a
conventional device (black) with C60
and NPD as an acceptor and donor,
respectively.
51

2.3.4.2 PTCBI Devices
PTCBI was used as an acceptor with NPD as a donor. PTCBI is a flat, small molecule
acceptor. It does not conduct electrons as well as C60 so it is used as an exciton blocking
layer as.
50

Table 2-6: Device performance of OPVs with PTCBI and NPD.

JSC (mA/cm
2
) VOC (V) FF PCE (%)
i- ZnO/PTCBI/NPD 1.5 0.6 0.5 0.4
i- AZO/PTCBI/NPD 1.6 0.6 0.4 0.4
c- NPD/PTCBI 0.7 0.8 0.4 0.2
i: ITO/ETL (20nm)/PTCBI (15 nm)/NPD (11 nm)/ MoO3 (30nm)/Al
c: ITO/ NPD (11nm)/ PTCBI (15 nm)/BCP (10nm)/Al

The inverted devices had PCE’s twice that
of the conventional device (Table 2-6 and
Figure 2-22). This is due to a higher JSC in the
inverted devices, which can be attributed to
optimizing the absorbed power in the inverted
devices with PTCBI. The VOC is lower in the
inverted devices, which could be caused by
increased charge recombination at the rough
ETL/A interface, as previously observed in
similar devices.
37

2.3.5 Optimized i-OPV
The inverted devices had a higher efficiency than conventional devices by using PTCBI,
whereas the opposite holds true when C60 is an acceptor. PTCBI was investigated in
conjunction with CuPc to observe how it will work with a visible-light absorbing donor.
-0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0
-2
0
2
4

Current (mA/cm2)
Voltage (V)

Figure 2-18: Inverted devices with
ZnO (red) and AZO (blue) as an
ETL and a conventional device
(black) with PTCBI and NPD as an
acceptor and donor, respectively.

52

Inverted devices were again made with both ZnO and AZO and compared to a
conventional device (Table 2-7 and Figure 2-23).

Table 2-7: Device performance of OPVs with PTCBI and CuPC.

JSC (mA/cm
2
) VOC (V) FF PCE (%)
i- ZnO/PTCBI/CuPc 3.5 0.5 0.5 0.8
i- AZO/PTCBI/CuPc 3.6 0.5 0.5 0.8
c- CuPc/PTCBI 0.9 0.5 0.3 0.1
i: ITO/ETL (20nm)/PTCBI (40nm)/CuPc (40nm)/ MoO3 (30nm)/Al
c: ITO/ CuPc (40nm)/ PTCBI(40nm)/BCP (10nm)/Al

The inverted devices with PTCBI/CuPC had
significantly better performance than that
of the conventional device (Table 2-7 and
Figure 2-23). This is due to a greatly
improved Jsc. This confirms the theory that
the optical electric field is inverted in i-
OPVs versus c-OPVs, necessitating the use
of blue-absorbing donors and red-
absorbing acceptors. While this device has
both red-absorbing materials, excitons are formed in both layers. A slight S-shape in the
JV curve of the inverted devices above 0.6 V indicates poor charge extraction (see Figure
2-23). This was common in inverted devices with CuPc (Figure 2-19), and not observed
in the NPD devices ( and Figure 2-22). Thus, it is most likely due to poor hole extraction
between the CuPC and MoO3. This can be improved upon with a different donor
molecule.

-0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0
0
10

Current (mA/cm
2
)
Voltage (V)

Figure 2-19: Inverted devices with ZnO
(red) and AZO (blue) as an ETL and a
conventional device (black) with PTCBI
and CuPC as an acceptor and donor,
respectively.

53

2.4 Conclusions
Inverted OPVs have demonstrated improved stability, though are limited by electron
extraction.
35
ZnO nanoparticles have been used as an electron tranport layer (ETL)
between ITO and the active layer to improve electron extraction. The ZnO layer was
optimized for electron extraction through post-treatment and doping aluminum in the
nanoparticles. Utilizing AZO nanoparticles did not significantly improve the devices over
ZnO. As XPS was not performed on these layers, it is possible that the aluminum was not
doped into the nanoparticles.
The theory proposed by Trinh et al. that inverted bilayer devices have a reduced JSC
due to the optical electric field not optimized for standard OPV materials (C60 and CuPC)
was investigated and proven experimentally. c-OPV devices with C60 (a blue absorber)
had a better JSC that the corresponding i-OPV, whereas PTCBI (red absorber) had an
improved JSC in an i-OPV versus a c-OPV. These results were consistent across devices
with different donors (NPD and CuPC).
2.4.1 Future Work
Next steps in this project include fabricating an i-OPV with AZO and PTCBI (A) with
a red-absorbing donor, such as tetracene or -sexithiophene.
13
In 2015, a revolutionary
non-fullerene acceptor was developed, ITIC, resulting in an OPV with a PCE of 6.8%,
previously unheard of for nonfullerene OPVs.
22
Since then, a plethora of non-fullerene
acceptors have been developed, many of them being red absorbers (Figure 2-20).
51

54

Figure 2-20: Absorption spectra of ITIC-like non-fullerene acceptors

These have been used in inverted structures, and the highest efficiency single
junction OPV to date was an inverted device with a ternary structure, obtaining a PCE of
η=14%.
52
Also, the highest efficiency OPV to date, a tandem device, with each subcell
having an inverted structure using ZnO as an ETL.
25
The strategy employed herein of
optimizing the optical electric field in an inverted device by employing red-absorbing
acceptors and blue-absorbing donors is potentially impacting these devices. However,
the record breaking devices mentioned are bulk heterojunctions, so an analysis of the D/A
distribution within the active layer would need to be further analyzed.
Inverted device structures employing ZnO as an ETL have become ubiquitous within
the field of OPVs. Further investigation into optimizing device performance by
maximizing photon absorbance in donor/acceptor regions is imperative to the growth of
this field.
55

2.5 Experimental Methods
2.5.1 ZnO np Synthesis
ZnO nanoparticles were synthesized by reacting a 0.1 M zinc acetate in dimethyl
sulfoxide solution (30 mL) with about 2.5 mL of TMAOH in 8.5 mL ethanol added drop
wise over 5 minutes, and then mixing for one hour.
39
The resulting ZnO particles were
then dissolved in ethanol before being crashed out of solution by adding hexane and
centrifuging for 3 minutes at 6000 RPM. The ZnO particles were then redissolved and
crashed once more before they were suspended in ethanol to obtain a solution of ~8-
15 mg/mL ZnO.

56

2.6 Bibliography
1. Schlenker, C. W. & Thompson, M. E. Current Challenges in Organic Photovoltaic
Solar Energy Conversion. Top Curr Chem 312, 175–212 (2012).
2. Schlenker, C. W. & Thompson, M. E. The molecular nature of photovoltage losses
in organic solar cells. Chem. Commun. (Camb). 47, 3702–3716 (2011).
3. Zhao, X. et al. Structure optimization of organic planar heterojunction solar cells.
J. Phys. D. Appl. Phys. 46, (2013).
4. Turro, N., Ramamurthy, V. & Scaiano, J. C. Modern Molecular Photochemistry of
Organic Molecules. (University Science Books, 2010).
5. Siegmund, B. et al. Exciton Diffusion Length and Charge Extraction Yield in
Organic Bilayer Solar Cells. Adv. Mater. 1604424 (2017).
doi:10.1002/adma.201604424
6. Onsager, L. Deviations from Ohm’s law in weak electrolytes. J. Chem. Phys. 2,
599–615 (1934).
7. Onsager, L. Initial recombination of ions. Phys. Rev. 54, 554–557 (1938).
8. Braun, C. L. Electric field assisted dissociation of charge transfer states as a
mechanism of photocarrier production. J. Chem. Phys. 80, 4157–4161 (1984).
9. Deibel, C., Strobe, T. & Dyakonov, V. Role of the charge transfer state in organic
donor-acceptor solar cells. Adv. Mater. 22, 4097–4111 (2010).
10. Nelson, J., Kwiatkowski, J. O. E. J., Kirkpatrick, J. & Frost, J. M. Modeling Charge
Transport in Organic Photovoltaic Materials Introduction. Acc. Chem. Res. 42,
(2009).
11. Marcus, R. A. Electron Transfer Reactions in Chemistry. Theory and Experiment.
Rev. Mod. Phys. 65, 599–610 (1993).
57

12. Faist, M. A. et al. Competition between the charge transfer state and the singlet
states of donor or acceptor limiting the efficiency in polymer: Fullerene solar cells.
J. Am. Chem. Soc. (2012). doi:10.1021/ja210029w
13. Lin, Y., Li, Y. & Zhan, X. Small molecule semiconductors for high-efficiency
organic photovoltaics. Chem. Soc. Rev. 41, 4245–4272 (2012).
14. Roncali, J., Leriche, P. & Blanchard, P. Molecular materials for organic
photovoltaics: Small is beautiful. Adv. Mater. 26, 3821–3838 (2014).
15. Lee, H., Park, C., Sin, D. H., Park, J. H. & Cho, K. Recent Advances in Morphology
Optimization for Organic Photovoltaics. Adv. Mater. 1800453, 1800453 (2018).
16. Dennler, G., Scharber, M. C. & Brabec, C. J. Polymer-Fullerene Bulk-
Heterojunction Solar Cells. Adv. Mater. 21, 1323–1338
17. Dang, M. T., Hirsch, L. & Wantz, G. P3HT:PCBM, best seller in polymer
photovoltaic research. Adv. Mater. 23, 3597–3602 (2011).
18. Holliday, S., Li, Y. & Luscombe, C. K. Recent advances in high performance
donor-acceptor polymers for organic photovoltaics. Prog. Polym. Sci. 70, 34–51
(2017).
19. Vogelbaum, H. S. & Sauvé, G. Recently developed high-efficiency organic
photoactive materials for printable photovoltaic cells: a mini review. Synth. Met.
223, 107–121 (2017).
20. Li, Z. et al. 9.0% Power Conversion Efficiency From Ternary All-Polymer Solar
Cells. Energy Environ. Sci. (2017). doi:10.1039/C7EE01858D
21. Kolhe, N. B. et al. All-Polymer Solar Cells with 9.4% Efficiency from Naphthalene
Diimide-Biselenophene Copolymer Acceptor. Chem. Mater. 30, 6540–6548
(2018).
58

22. Lin, Y. et al. An Electron Acceptor Challenging Fullerenes for Efficient Polymer
Solar Cells. Adv. Mater. 27, 1170–1174
23. Zhan, X. et al. Rylene and related diimides for organic electronics. Adv. Mater.
23, 268–284 (2011).
24. Li, S. et al. A Wide Band Gap Polymer with a Deep Highest Occupied Molecular
Orbital Level Enables 14.2% Efficiency in Polymer Solar Cells. J. Am. Chem. Soc.
140, 7159–7167 (2018).
25. Meng, L. et al. Organic and solution-processed tandem solar cells with 17.3%
efficiency. Science (80-. ). 361, 1094–1098 (2018).
26. O’Regan, B. & Gratzel, M. A Low-Cost, High-Efficiency Solar-Cell Based on Dye-
Sensitized Colloidal TiO2 Films. Nature 353, 737–740 (1991).
27. Kojima, A., Teshima, K., Shirai, Y. & Miyasaka, T. Organometal Halide Perovskites
as Visible-Light Sensitizers for Photovoltaic Cells. J. Am. Chem. Soc. 131, 6050–
6051 (2009).
28. Snaith, H. J. Perovskites: The emergence of a new era for low-cost, high-efficiency
solar cells. J. Phys. Chem. Lett. 4, 3623–3630 (2013).
29. Lee, M. M., Teuscher, J., Miyasaka, T., Murakami, T. N. & Snaith, H. J. Efficient
Hybrid Solar Cells Based on Meso-Superstructured Organometal Halide
Perovskites. Science (80-. ). 338, 643–647 (2012).
30. Boix, P. P., Nonomura, K., Mathews, N. & Mhaisalkar, S. G. Current progress and
future perspectives for organic/inorganic perovskite solar cells. Mater. Today 17,
16–23 (2014).
31. Liu, J. P., Wang, S. S., Bian, Z. Q., Shan, M. N. & Huang, C. H. Inverted
photovoltaic device based on ZnO and organic small molecule heterojunction.
59

Chem. Phys. Lett. 470, 103–106 (2009).
32. Zhang, F. et al. Recent development of the inverted configuration organic solar
cells. Sol. Energy Mater. Sol. Cells 95, 1785–1799 (2011).
33. Tan, M. J., Zhong, S., Li, J., Chen, Z. & Chen, W. Air-Stable Efficient Inverted
Polymer Solar Cells Using Solution-Processed Nanocrystalline ZnO Interfacial
Layer. ACS Appl. Mater. Interfaces 5, 4696–4701 (2013).
34. Fan, X., Zhang, M., Wang, X., Yang, F. & Meng, X. Recent progress in organic-
inorganic hybrid solar cells. J. Mater. Chem. A 1, 8694–8709 (2013).
35. Steim, R., Kogler, F. R. & Brabec, C. J. Interface materials for organic solar cells.
J. Mater. Chem. 20, 2499–2512 (2010).
36. Kestell, A. E. & DeLorey, G. T. Nanoparticles : properties, classification,
characterization, and fabrication. (Nova Science Publishers, 2010).
37. Trinh, C. et al. Power losses in bilayer inverted small molecule organic solar cells.
Appl. Phys. Lett. 101, 4 (2012).
38. Cohn, A. W., Janßen, N., Mayer, J. M. & Gamelin, D. R. Photocharging ZnO
Nanocrystals: Picosecond Hole Capture, Electron Accumulation, and Auger
Recombination. J. Phys. Chem. C 116, 20633–20642 (2012).
39. Meulenkamp, E. A. Synthesis and Growth of ZnO Nanoparticles. J. Phys. Chem. B
102, 5566–5572 (1998).
40. Trinh, C. et al. Power losses in bilayer inverted small molecule organic solar cells.
Applied Physics Letters (2012). doi:10.1063/1.4769440
41. Kyaw, A. K. K. et al. An inverted organic solar cell employing a sol-gel derived
ZnO electron selective layer and thermal evaporated MoO3hole selective layer.
Appl. Phys. Lett. 93, 1–4 (2008).
60

42. Weber, D., Botnaraş, S., Pham, D. V., Steiger, J. & De Cola, L. Functionalized ZnO
nanoparticles for thin-film transistors: Support of ligand removal by non-thermal
methods. J. Mater. Chem. C 1, 3098–3101 (2013).
43. Aldrich, S. ITO. (2018). Available at:
https://www.sigmaaldrich.com/catalog/substance/ito123455092611911?lang=en
®ion=US.
44. Aldrich, S. Fluorine doped tin oxide coated glass slide. (2018). Available at:
https://www.sigmaaldrich.com/catalog/product/aldrich/735140?lang=en®ion
=US.
45. Kwak, D.-J., Moon, B.-H., Lee, D.-K., Park, C.-S. & Sung, Y.-M. Comparison of
transparent conductive indium tin oxide, titanium-doped indium oxide, and
fluorine-doped tin oxide films for dye-sensitized solar cell application. J. Electr.
Eng. Technol. 6, 684–687 (2011).
46. Murdoch, G. B. et al. Aluminum doped zinc oxide for organic photovoltaics. Appl.
Phys. Lett. 94, 17–20 (2009).
47. Tress, W. et al. Imbalanced mobilities causing S-shaped IV curves in planar
heterojunction organic solar cells. Appl. Phys. Lett. 98, (2011).
48. Bartynski, A. N. et al. A fullerene-based organic exciton blocking layer with high
electron conductivity. Nano Lett. 13, 3315–3320 (2013).
49. Luhman, W. A. & Holmes, R. J. Investigation of energy transfer in organic
photovoltaic cells and impact on exciton diffusion length measurements. Adv.
Funct. Mater. 21, 764–771 (2011).
50. Mishra, A. & Bäuerle, P. Small molecule organic semiconductors on the move:
Promises for future solar energy technology. Angew. Chemie - Int. Ed. 51, 2020–
61

2067 (2012).
51. Hou, J., Inganäs, O., Friend, R. H. & Gao, F. Organic solar cells based on non-
fullerene acceptors. Nat. Mater. 17, 119–128 (2018).
52. Xiao, Z., Jia, X. & Ding, L. Ternary organic solar cells offer 14% power conversion
efficiency. Sci. Bull. 62, 1562–1564 (2017).

62

3 Microparticulate Light
Harvesting Units

63

Table of Contents

3.1 Project Description........................................................................................................... 64
3.2 Background...................................................................................................................... 65
3.2.1 Zeolite-L ................................................................................................................................................................65
3.2.2 Pore Filling ............................................................................................................................................................66
3.2.2.1 Low vs. High Loading of Dye Molecules ...............................................................................................70
3.2.3 Pore Capping ........................................................................................................................................................71
3.2.4 Energy and Electron Transfer in Zeolite-L.......................................................................................................72
3.3 Results............................................................................................................................. 76
3.3.1 Energy Transfer in Zeolite-L...............................................................................................................................76
3.3.1.1 Tetracene Encapsulation in Zeolite .......................................................................................................78
3.3.1.2 Photophysical Characterization of Z-T Samples ..................................................................................78
3.3.1.3 Pore Capping with BODIPY-dipyridinium .............................................................................................81
3.3.2 Electron Transfer in LHUs ..................................................................................................................................85
3.3.2.1 Anthracene Loading in Zeolite-L ............................................................................................................85
3.3.2.2 Z-A-BD Photophysics ................................................................................................................................88
3.3.3 Water Splitting in LHUs ......................................................................................................................................91
3.3.3.1 Hydrogen Evolution Reaction (HER)......................................................................................................91
3.3.3.2 Molecular HER Catalysts..........................................................................................................................91
3.3.3.3 Catalytic Palladium Nanoparticles .........................................................................................................93
3.3.4 Optimizing Electron Transfer in LHUs ..............................................................................................................94
3.3.4.1 Diluent Stopcock Molecules ...................................................................................................................94
3.3.4.2 Diluting BODIPY-dipyridinium on zeolite..............................................................................................96
3.3.4.3 Energy and Electron Transfer in LHUs ..................................................................................................98
3.3.5 Optimizing LHU Electron Transfer ................................................................................................................. 101
3.4 Conclusions and Future Outlooks ................................................................................... 104
3.4.1 Future Recommendations .............................................................................................................................. 104
3.4.2 Challenges with LHUs ...................................................................................................................................... 105
3.5 Experimental Methods ................................................................................................... 106
3.5.1 Synthesis of Zeolite-L....................................................................................................................................... 106
3.5.2 Pore Loading ..................................................................................................................................................... 108
3.5.3 Pore Stoppering................................................................................................................................................ 111
3.5.4 Synthesis of BODIPY-dipyridinium ................................................................................................................ 111
3.5.4.1 Synthesis of Naphthalene-Pyridinium ............................................................................................... 116
3.5.4.2 Diether-pyridinum synthesis ............................................................................................................... 116
3.5.4.3 HER experiments ................................................................................................................................... 117
3.6 Bibliography .................................................................................................................. 119

64

3.1 Project Description
Solar water splitting has attracted considerable attention as a potential renewable
energy source with no carbon emissions. Overall water splitting requires both hydrogen
evolution reaction (HER) and oxygen evolution reaction (OER) photocatalysts to be
present; as this system is complex, individual compenents are investigated and optimized
separately. Research areas in solar water splitting include: HER catalysts, OER catalysts
and photosensitizers.
A novel approach to a photosensitizer design for photocatalytic HER was
developed. It is a supramolecular assembly using a thin, disc-shaped zeolite-L to organize
donor and acceptor small molecules in a bilayer. This is a light-harvesting unit (LHU) that
facilitates charge separation at the surface of the zeolite-L (see Figure 3-1). This assembly
is designed after bilayer organic photovoltaics. It is based on the pioneering work by Gion
Calzaferri on energy transfer in zeolite-L, and that of Thomas Mallouk looking at electron
transfer in zeolite-L.
1 –3

Figure 3-1: (a) Steps to make a LHU loaded with a donor material and capped with an acceptor
material (b) complete photoelectrochemical system, modelled with a Pt catalyst, bound to the
surface.

The system created herein is optimized for exciton diffusion and electron transfer
without the use of heavy metals. Energy transfer through the loaded dye allows efficient
exciton diffusion after photon absorption. The exciton will diffuse through the 1-D pores

65

to the surface of the zeolite-L, where the excited state will donate an electron to the
acceptor stopcock to form a charge transfer state. This will dissociate into the respective
electron and holes; the holes will migrate through the donor, and the electron on the
stopcock will be utilized by an HER catalyst. The entire zeolite-donor-acceptor system is
referred to as a light-harvesting unit (LHU) (Figure 3-1, a).
The advantage of using zeolite-L as a scaffolding to organize the dye molecules is to
provide stability in the solid state. The PEC system can complete either HER or OER,
depending on the energetics of the dyes in place. The LHU can be either functionalized
on the surface with an electrode catalyst (eg. Pt), or placed in solution with an electrode
catalyst. For HER, the electrode catalyst will collect electrons and subsequently catalyze
a water-splitting reaction to generate hydrogen in the presence of a sacrificial donor
(Figure 3-1, b). These micron- scale PEC systems will be suspended in solution; this will
improve efficiency over immobilized PEC systems (Type III and IV) since zeolite
efficiently scatters light as it has a similar refractive index to glass. All incident photons
can be absorbed through the use of various dyes that absorb throughout the entire solar
spectrum.
3.2 Background
3.2.1 Zeolite-L
Zeolite frameworks have structured pores of varying size, shape, dimensionality, and
charge.
4
Zeolites are thermally stable, and can capture small molecules in their pores.
Zeolite-type framework structures are normally based on fully cross-linked tetrahedra
containing Si, Al, P and occasionally other atoms (bridges formed mostly by O atoms).
The specific zeolite used herein is zeolite-L (or Linde-type L, LTL).
Zeolite-L are crystalline aluminosilicates with defined channels and hexagonal
66

symmetry (Figure 3-2). SiO4
−
and AlPO5
−
tetrahedra are the primary building units.
Zeolite-L is composed of cancrinite cages linked along the c-axis (see Figure 3-2
a, b).
5
Crystals can be synthesized with varying sizes and aspect ratios, though the
dimensions of the cages are fixed with a 7.1Å pore opening, 12.6Å in the center and
7.5Å width (see Figure 3-2, c). Zeolite-L is synthesized by hydrothermal sythesis (See
section 3.5.1). Zeolite-L has an anionic framework, so the stoichiometry of zeolite-L with
monovalent cations, M
+
, is (M)9d[Al9−dSi27+dO72]*nH2O (where d depends on the
synthesis procedure and n corresponds to the water hydrogen bonded inside). The charge
compensating cations are located at four different sites (A, B, C, D) in fully hydrated
(n=21) zeolite-L (Figure 3-2, d).
6
Only cations at site D are exchangeable at room
temperature in aqueous environments. The cation exchanged in the zeolite-L framework
can impact the rate of intersystem crossing of the dye loaded due to the heavy atom effect
(decay from the excited singlet state to the excited triplet state).

Figure 3-2: Zeolite-L structure and dimensions (a) top view (b) cancrinite cages stacked along
the c-axis (c) dimensions of the pores; (d) Cation sites in a zeolite- L cage

3.2.2 Pore Filling
The defined cages and channels unique to zeolite-L allow chromophores to organize
in the pores. There are different methods for loading dyes. The method used to load dyes
67

depends on the molecule being loaded and the desired host- guest morphology. How the
molecule will orient in the channel is related to the size, shape, charge and concentration
of the dye molecule, as well as the co-solvent in the channel (Figure 3-3).
7

Figure 3-3: (A) Four representative orientations of molecules with different packing. (B)
Orientation of molecules which align their ETDM parallel and (B’) perpendicular to the channel
axis and which have no interaction because their shape keeps them at sufficiently large distance.
(C) Orientation of large molecules which align their ETDM parallel and (C’) perpendicular to the
channel axis and which are so close that Davydov coupling is important.
8

The shape and size of the dye entering the pore is crucial to this study because this
will impact the orientation and chromophore- chromophore interactions. This has been
well studied by investigating various methods for dye loading. The quantity of dye loaded
is determined by thermogravimetric analysis (TGA) or HF analysis and is most often
reported by the p value (given in equation (3-1)).
68

(3-1) 𝑝 =
𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑜𝑐𝑐𝑢𝑝𝑖𝑒𝑑 𝑠𝑖𝑡𝑒𝑠 𝑡𝑜𝑡𝑎𝑙 𝑎𝑚𝑜𝑢𝑛𝑡 𝑜𝑓 𝑠𝑖𝑡𝑒𝑠

Loading is also reported as moles of dye per gram of zeolite (mol g
−1
), which is related to
p through the concentration of dye in the zeolite (c(p)) (eq. (3-2).
(3-2) 𝑐 ( 𝑝 ) = 0.752
𝑝 (𝑚𝑜𝑙 )
𝑛 𝑠 (𝐿 )

in which ns is the number of unit cells in zeolite-L that are occupied by dye molecules.
9

The different methods for loading zeolite-L are:
8

1) Dye loading by sorption
a) Gas Phase
b) Solution
2) Dye loading by ion exchange
3) Dye synthesis in nanopores (ie. Ship in the bottle)
4) Crystallization inclusion
The most popular methods for loading dyes in zeolite-L are sorption loading in the gas
phase and ion exchange. These techniques are the easiest to execute and allow a large
selection of molecules.
Dye Loading by Sorption
This method has versatile loading rates, from low to high loading, but the molecules
allowed are limited to those that can fit inside the mouth of the pore (<7Å). Diffusion of
the sorbed molecule depends on the size of the molecule as well as the strength of the
host-guest interaction.
Gas Phase Loading
Gas phase loading allows incorporation without significantly altering the lattice
structure, although this can change when dehydrating the zeolite.
10
A few methods for
69

gas phase adsorption have been documented, both including and excluding dehydration
before sublimation of the material. Neutral dyes are loaded by gas phase adsorption. A
perylene dye, DXP (a donor molecule) is a popular molecule for gas phase loading and
host-guest interaction studies because it is small enough to enter the pore, but large
enough to only orient parallel to the pore channel.
9

The single ampoule method for dehydrated loading was the one used in this study.
This allowed for high loading of acene molecules, in a relatively short period of time,
without the co-absorption of water molecules.
Solution Phase Loading
Solution phase loading incorporates neutral species by mixing zeolite and dye in
solution. This technique is not as popular because it has relatively low loading; however,
it is an easy method to incorporate neutral molecules that cannot be sublimed, or to
readily prepare materials under N2.
11

Dye Loading by Ion Exchange
Dye loading by ion exchange is done by exchanging the cations present in zeolite-L
with the cationic dye, in solution. The cations exchanged must fit the diameter of the
internal pores. High loading results in the formation of aggregates on the outer surface.
12

Low loading leads to uniform distribution of molecules in the pores. Ion exchange
typically has less loading than gas phase adsorption because diffusion is slower in
solution, which is based on the exchange time. The majority of ion exchange studies have
been done utilizing cationic molecules pyronine (Py
+
) and oxonine (Ox
+
), because they
are small enough to enter the pore, can easily ion exchange within the pore wall, and have
well-established photophysical data.
8 ,9,13

70

Dye Synthesis in Nanopores (Ship-in-the-Bottle)
Dye encapsulation by ship-in-the-bottle allows for inclusion of dyes that are larger
than the pore opening of the zeolite. The first precursor is inserted into the zeolite, which
is then activated before the second precursor is added and reacts with the first to form the
final product. This method has almost exclusively been done in faujasite zeolite (aka
zeolite- Y). This technique allows for complexes to be prepared in situ, and thus they
cannot diffuse out of the solid matrix.
14

Crystallization Inclusion
Crystallization inclusion incorporates dye molecules into zeolite pores during the
hydrothermal synthesis of the zeolite. This requires the dye molecule to be soluble in the
gel mixture as well as remain relatively stable under the harsh reaction conditions. One
benefit of crystallization inclusion is that the diameter of the dye molecule can be larger
than that of the zeolite pore, though in that case, they will be defects in the crystalline
lattice. This method is most often utilized in NaY, AlPO5
−4
, SAPO
−5
and FAPO
−5
crystals,
and is rare in zeolite-L.
15

These methods for incorporating materials in zeolite-L nanoparticles vary in their
ease and applicability depending on the material being inserted. While gas phase
adsorption and ion exchange are most popular for zeolite-L, ship-in-the-bottle is a
recently emerging technique for zeolite-L. Crystallization inclusion
is more popular for the neutral form of zeolite-L, AlPO5
−4
, and has not really
been applied for dye molecules inclusion in zeolite-L.
3.2.2.1 Low vs. High Loading of Dye Molecules
The extent of molecular loading in zeolite-L can strongly impact how the molecules
interact with each other, as well as with the host. As most of the studies done to date have
71

focused on FRET from dyes in the zeolite-L to a stopcock molecule at the mouth of the
channel, experiments have used low loading to prevent electronic coupling between
molecules, and self-absorption. Molecules that have been studied have had their
electronic transition dipole moment (ETDM) aligned parallel to the channel to facilitate
dipole-dipole interactions to study FRET in zeolite. Studies have shown that at higher
loading, self-absorption causes red shifting of the fluorescence of dye molecules loaded in
zeolite-L; thus, most reports have been limited to low loading. This is confirmed by Gigli,
who stated “the structure, properties, and behavior of highly packed dye-ZL materials,
i.e., characterized by a high degree of dye loading, have never been explored to date,
neither by experiment nor by modeling.”
7

3.2.3 Pore Capping
Pore capping using “stopcock” molecules is a technique to plug the channels of
zeolite-L particles; these molecules inherently prevent small molecules from entering or
exiting the channels. They also facilitate energy transfer to outside the channels of the
zeolite-L molecule. Molecular stopcocks are designed with three parts: the head, the
spacer and the label (Figure 3-4).
16

The head of the stopcock molecule is larger than
the channel entrance, therefore preventing it from
entering the channel, and acts as a plug. These can be
either reactive or non-reactive moieties. The spacer is
between the head and the label, and enters the channel. This controls the length of the
tail, and can elongate the stopcock molecule to improve solubility and flexibility. The
label enters the channel and hooks to it. This can be done with a non-reactive label or a
reactive label. The method with which to attach the stopcock molecule to the channel

Figure 3-4: Stopcock molecule
in zeolite pore

72

depends on the interaction between them. Stopcock molecules can either bind reversibly
(by van der Waals forces), electrostatically or irreversibly (covalently bound) (Figure 3-5,
a).
17

The selectivity of stopcock position of the molecules on the zeolite is characterized by
fluorescence (confocal) microscopy. This is done by either utilizing fluorescent heads, or
marking the stopper with a luminescent material.
14
This is the most popular method of
stopper selectivity, though not quantitative. Another method for characterizing the
amount of stopcock molecules present on the surface is to measure the absorption of the
starting solution before adsorption to the pore, and the absorption after the stopcock has
bound to the surface.
1 ,14
Quantitative values are not often given, but adsorption is also a
function of the solvent used. Binding to the
pore entrance selectively is a precise process,
and depends on the solubility of the molecule
in solution and the concentration of the
material in solution (Figure 3-5,b).
17

The majority of stopcock molecules have
been utilized to inject or extract electronic
excitation energy in or out of zeolite-L
molecules.
18
Inserting stopcocks is dependent
on the solution they are in, which is a mixture
of a solubilizing solvent as well as one in which the stopcock is insoluble. The most
strongly bound stopcock molecules utilize electrostatic binding, and covalent bonds to
bind to zeolite; this study used electrostatic binding.
3.2.4 Energy and Electron Transfer in Zeolite-L
(a)
(b)
Figure 3-5: (a)Reversible binding by van
der Waals forces, electrostatic binding
and irreversible binding by covalent
forces and (b) Dependence of binding
selectivity on solution

73

Energy and electron transfer have been
studied between dyes and stopcocks
encapsulated in zeolite. Energy transfer
(FRET) was discussed in Chapter 2.
Electron transfer is the process by which an
excited molecule donates an electron to a
ground state molecule in close proximity;
this creates a charge transfer state (Figure
3-6). This charge transfer state will then
either recombine to the original donor and
acceptor ground states, or it will charge separate to create a radical anion (or radical
cation). Self-absorption is a process that suppresses energy transfer; this occurs when a
neighboring identical molecule reabsorbs the emission of a photon from a molecule. A
small Stokes shift, the distance between the peak absorption and peak emission, is one
indicator that self-absorption can occur. This is more likely to occur in condensed phase
matter.

Figure 3-6: Schematic relationship among
electron, hole, and energy transfer. R* is
an electronically excited molecule after
light absorption. M is a second molecular
species it is interacting with.

74

Extensive research into FRET in donor
dye molecules loaded in zeolite-L to acceptor
stopcock molecules at the pore openings has
been performed (Figure 3-7). These systems
use low dye loading in the pores to prevent
self-quenching. FRET is optimized by
aligning the electronic transition dipole
moment (ETDM) parallel to the channel,
and to the stopcock, as well as ensuring the
stopcock is within proximity of the loaded dyes, and significant spectral overlap between
the emission of the donor and the absorption of the acceptor. Energy transfer has been
proven in multiple systems through steady state photophysical studies.
19
A wide variety
of molecules have been loaded into zeolite-L and studied for their capability to either
donate or accept energy from another molecule. This is covered in depth in a review by
Calzaferri and Lutkouskaya.
1
Energy transfer between two dyes encapsulated in zeolite
has also been studied, with donor molecules, pyronine (Py
+
), and oxonine (Ox
+
) as an
acceptor being the standard.
6
Other molecules that are both neutral and cationic have also
been investigated, including methylviologen, naphthalene, anthracene, stilbene,
fluorenone, perylene derivatives and others.
1 8,20,21
Different stopcock molecules designed

Figure 3-7: (A) Schematic representation
of energy transfer and (B) possible
electronic transition dipole moments for a
stopcock molecule
18

75

to either donate or accept electronic energy
from the loaded dyes include ruthenium
derivatives, BODIPY derivatives,
phthalocyanine and others. One way in
which FRET has been characterized between
dyes and stopcocks is through confocal
microscopy. An example of this by Fabio
Cucinotta et al. uses an iridium donor, and
pyronine acceptor loaded in the channels
(Figure 3-8).
20
Confocal images show that in
3 μm long zeolite crystals, the dye and
stopcock are isolated at the pore opening.
The emission of the energy donor stopcock
overlaps with the absorption of the energy
acceptor emission. When the energy
acceptor is loaded in zeolite without the energy donor, there is not significant emission.
Once loaded with the energy acceptor, the emission increases significantly.
Electron transfer in zeolite has not been investigated in as much detail, but was
demonstrated by Thomas Mallouk in the early 1990’s based on Ru(bpy)2
+
/BV
2+
.

Diffuse
reflectance transient absorption spectrscopy was
used to measure the rates of electron transfer in
the system. Prabir Dutta later investigated a
Ir(ppy)3/PVS D/A system.
2,22

Figure 3-8: Top: Schematic representation
and fluorescence microscopy images of 3
μm long zeolite crystals, showing stopper 1
at the channel entrances; a PyY-loaded
zeolite; and a crystal loaded with both
donor 1 and acceptor; the borders of the
crystal are indicated along with the
positions of the maximum dye distribution
and the inflection point. Bottom: Emission
spectra of 1-PyY-Z (black), 1-Z (blue), and
PyY -Z (red); the absorption spectrum of
PyY -Z (dark yellow) is also reported to
show the overlap with the 1-Z emission.
20

Figure 3-9: Long-lived light-
induced charge separation in a
zeolite L-based molecular triad

76

3.3 Results
3.3.1 Energy Transfer in Zeolite-L
Tetracene was chosen for this study for a variety of reasons: it’s a well-known and
characterized electron-donor, it’s ability to π- stack allows efficient charge transport, it
has the appropriate dimensions to fit in zeolite-L pores (3Åx 12Å), and its ability to
sublime. Tetracene can only load along the small axis, and fit in a channel along the long
axis. It displays strong absorption and emission properties, so energy transfer can be
readily characterized.
In solution, tetracene has a small Stokes shift
(4 nm). The solution phase lifetime and quantum
yield are given in Table 3-1.
As tetracene encapsulated in zeolite will
pack in various formations, the photophysical markers of various solid-state properties is
important to understand. In the solid state, the absorption and emissive properties
change due to varying amounts of structural disorder and Davydov splitting (splitting of
the electronic bands due to the interaction of multiple molecules interacting in the unit
cell) (Figure 3-10 a,b).
23

Table 3-1: Lifetime (τ ), quantum
yield (Φ) and radiative rate (kr) of
tetracene in methanol
τ (ns) Φ
kr
(x10
7
s
-1
)
5.5 0.14 2.55

77

(a) (b)
Figure 3-10: (a) Absorption spectra of dilute tetracene in THF (solid line), tetracene
nanoparticles in aqueous solution (dashed line), and a vacuum- deposited film of tetracene
(dotted line). Inset: tetracene molecular structure and orientation of its transition dipole;
(b)Fluorescence spectra for samples in (a). The labels 0-0 and 0-1 denote the vibronic peaks.
23

Optical effects, such as reabsorption, are often a concern when looking at packed, thin
film samples. This effect was demonstrated by J. Burdett et al. by studying photophysics
in polycrystalline films of tetracene (Figure 3-11,a).
24
Three films were made of varying
thicknesses, and the one in which the absorption is greater than 0.5 demonstrates how
optical effects affect fluorescence. As the optical density in this film is high, photons that
have been emitted from tetracene are reabsorbed. This results in a red- shifted and
broadened emission. Peter et al. reported on excimers in tetracene films, and measured

(a) (b) (c)
Figure 3-11: (a) The absorption spectra of 40 nm (black), 80 nm (red), and 215 nm
(blue) thick vacuum evaporated films, showing shape changes due to optical effects
(top) and the normalized fluorescence spectra of the same films (bottom);
24
(b) Gauss-
analysis of the emission spectrum of a low temperature tetracene film and (c) a
crystalline tetracene layer, where I and II are monomer and monomer-defect band
(respectively), III and IV are the fitting components of the excimer band
25

78

their emission (see Figure 3-11 (b) and (c)). There were two prevalent species, the
monomer (τ1 =6.8±1.2 ns) at higher energy and the excimer state at lower energy
(τ2=21.3±7.1 ns).
25

3.3.1.1 Tetracene Encapsulation in Zeolite
While there are no reports of tetracene loaded in zeolite-L, it has been loaded in the
three-dimensional pores of zeolite-Y (13Å spherical pores with 7.6Å entrances) and
ZSM-5 zeolite.
26 –28
It is found that one end of tetracene in zeolite-Y will coordinate to a
sodium counterion by its π-electron system, pointing the other end into an adjacent
supercage where it interacts with the zeolite lattice. When considering zeolite-L, the most
readily accessible counterion is at the D position, which is located at the pore opening. It
is likely that one end of tetracene will coordinate here, with the other end of the tetracene
along the pore length. The other possibility is that for the two ends to coordinate to sites
C/E, however, as tetracene will barely fit in the widest section of zeolite-L, the first
scenario is the most probable. This will allow one tetracene molecule per cage, and orient
the tetracene’s electronic transitional dipole moment parallel to the c-axis of zeolite-L.
FRET will occur along the channel, so will be funneled toward the stopcock molecules.
The process by which dyes were intercalated was optimized to ensure water in the
channels was evacuated before dye insertion, as well as total dye insertion along the length
of the pore (see section 3.5.2). The samples are denoted herein as Z-Tx, wherein x is the
percent loading of tetracene. The absorption of Z-T samples was measured in the solid
state by diffuse reflectance UV-vis spectroscopy to eliminate scattering from zeolite-L,
and the emission was also measured in the solid state.
3.3.1.2 Photophysical Characterization of Z-T Samples
The absorption and emission of the Z-T samples are compared with monomeric and
79

polycrystalline tetracene, which have different ground and excited state electronic
structures. The absorption spectra of the Z-T samples are bathochromically shifted and
broadened compared to the monomer; this is due to Davydov splitting in the crystalline
state (Figure 3-12, a).
At low loading (Z-T25), tetracene has more monomeric characteristics; as loading
increases, it becomes more polycrystalline in the zeolite: this is evident from the
fluorescence spectra (Figure 3-12, b). The emission of Z-T samples red- shifts and
broadens. At higher loading, it is similar to solid-state tetracene, as there are no longer
distinct vibronic features, such as those in monomeric tetracene. This is evidence of
tetracene aggregate formation in the pores of zeolite-L.
300 350 400 450 500 550
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0.16

Kubelka- Munk (a.u.)
Wavelength (nm)
(a)
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)

500 550 600 650 700 750 800
0.0
0.2
0.4
0.6
0.8
1.0

Kubelka- Munk (a.u.)
Wavelength (nm)
(b)

Figure 3-12: (a) Normalized absorption of tetracene in solution (black, right axis) and Kubelka-
Munk of Z-T25 (cyan), Z-T50 (light blue), Z-T75 (blue) and Z-T1 00 (dark blue) as powders and (b)
emission (excited at 445 nm) of the same samples.

The broad, bathochromic shift of tetracene emission with increased loading could
be caused by: (1) tetracene aggregation, (2) tetracene excimer emission or (3) optical
effects due to self-absorption. As the optical density was kept below 0.2, and the
absorption and excitation spectra do not show optical effects, it is not option (3).
Tetracene excimers are known to emit at 600 nm, with a lifetime of 21 ns. The lifetime of
the species investigated was measured, and there was one long lived component of ~20
80

ns, but it was only 4% of the total lifetime (Table 3-2). Excimer species are present, but
in a small population. Reports of tetracene dimers with face-to-face stacking have
emission that is 70 nm red shifted from the monomeric species.
29
This could be possible
if two tetracene are stacked in a single channel. The prevalent species in Z-T samples is
about 85 nm red- shifted from the monomer, so it is unlikely to be dimers. Therefore,
tetracene aggregates is the likely majority orientation.
The lifetime and quantum yield were measured (see Table 3-2). The quantum yield
was consistently 0.04 across all the samples. This is reduced from tetracene in solution,
which is 0.14; therefore the emission is quenched in zeolite-L. Zeolite-L is known to act
as a donor and acceptor, due to its ionic nature, and depending on the encapsulated dye.
Excited-state tetracene can donate its energy to the zeolite framework, which
nonradiatively decays. Three lifetimes were measured by fitting the decays; the fastest
lifetime is 0.4 ns, a lifetime from 4-8 ns, and one between 15-30 ns (see Table 3-2). The 4
ns lifetime is from the monomer species, and the 15-30 ns lifetime is either the excimer
lifetime or delayed fluorescence.
Table 3-2: Lifetime (in ns), lifetime
contribution (A) and quantum yield
(ex. 440 nm, measured at 550nm)

τ1 (A1) τ2 (A2) τ3 (A3) Φ
ZT25 0.4 (58) 3.8 (38) 15 (4) 0.04
ZT50 0.4 (7 4) 5 (23) 21 (3) 0.04
ZT75 0.5 (68) 7 (28) 28 (4) 0.04
ZT1 00 1 (7 3) 8 (22) 30 (4) 0.03

0 20 40 60 80 100
10
100
1000
10000
Counts
Time (ns)
Figure 3-13: Lifetime decay of tetracene in
zeolite at various loadings; Z-T25 (cyan), Z-
T50 (light blue), Z-T75 (blue) and Z-T100 (dark
blue) as powders

81

Plots of the lifetime decays exemplify that as loading (and aggregation) increases,
the lifetime also increases; this is counterintuitive, as in most dye molecules, more
aggregated states would induce more quenching (Figure 3-13). The increase in lifetime is
possibly due to delayed fluorescence in tetracene. Delayed fluorescence occurs when two
chromophores in the triplet-excited state are in close enough proximity that the
delocalized energy between them will combine to a singlet exciton. Delayed fluorescence
has been measured on the order of 20-100 ns for polycrystalline tetracene samples, and
varies depending on the degree of crystallinity.
24
As tetracene loading increases, there are
more crystalline domains and delayed fluorescence is favored.
After tetracene was loaded in zeolite-L and the photophysics characterized, the
zeolite-L pore entrances were exchanged with a stopcock BODIPY molecule.
3.3.1.3 Pore Capping with BODIPY-dipyridinium
Design of a stopcock molecule included
such parameters as: (1) red-shifted
fluorescence from the donor, (2) large
molecule (>7.2Å), and (3) include a label
and cationic tethering group. Boron-
dipyrromethane (BODIPY) appended with
a phenyl- diether(dipyridinium) was
synthesized for this purpose (Figure 3-14,
inset). The synthesis of BODIPY-
dipyridinium was carried out by Dr. Jessica
Golden, and is given in section 3.5.4.
BODIPY is a well-known fluorescent emitter, with a large molar extinction coefficient,
400 450 500 550 600 650 700
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity
Wavelength (nm)

Figure 3-14: Normalized absorption
(dashed) and fluorescence (solid) in acetone
(excited at 480 nm) Inset: BODIPY-
dipyridinium molecular structure

Table 3-3: Lifetime (τ ), quantum yield (Φ)
and radiative rate (kr) of BODIPY-
dipyridinium in acetonitrile
τ (ns) Φ kr (x10
7
s
-1
)
6.1 0.59 9.7

82

a high quantum yield(>50%), and an easily tunable
emission depending on the functional groups added.
It is ideal for studying energy transfer, through steady
state photophysics. It is too large to fit in the zeolite-
L pores, which will prevent small molecules from
entering, as well as the inhibit tetracene from leaving
the pores. The pyridinium units enter the pores and
exchange with the K
+
ions in the pores so they will not be removed. Pyridinium was chosen
as the binding unit as it is a strong electron acceptor. DFT calculated HOMO/LUMO
energies (using method B3LVP, with a basis set of LACVP and ECP fit-LACVP) for
BODIPY-dipyridinium show the energy density of the HOMO located on the pyridinium
units, and the LUMO on BODIPY.
The absorption spectra of BODIPY is red-shifted from that of tetracene, and overlaps
with tetracene’s emission - this will facilitate FRET (see Figure 3-14). The lifetime and
quantum yield of BODIPY in acetonitrile are given in Table 3-3.
kr is the radiative rate of fluorescence (f), calculated from equation (3-3):
(3-3) 𝑘 𝑟 =
Φ
𝑓 𝜏 𝑓
BODIPY-dipyridinium was fully loaded into the pore openings by the methods given
in section 3.5.3. A sample of empty zeolite-L
was exchanged with BODIPY-dipyridinium as
a reference (Z-BD). BODIPY loading resulted
in symmetric functionalization of BODIPY on
zeolite-L (Figure 3-16).

Figure 3-16: BODIPY-dipyridinium
symmetric capping on zeolite-L

Figure 3-15: DFT calculated
HOMO/LUMO electron density
on BODIPY-dipyridinium

83

The photophysics of the system were studied; the absorption profile is the same as
that for Z-T, but with a peak for BODIPY’s absorption at
λ=525 nm (see Figure 3-17 (a)). The emission from tetracene and BODIPY were measured
by exciting at wavelengths at which only BODIPY absorbs (λex=500 nm), or only tetracene
absorbs (λex=445 nm) (Figure 3-17, b,c, respectively). Exciting BODIPY results in BODIPY
emission, though the spectra are broadened and slightly red-shifted from the monomer
(Figure 3-17, b). The shoulder at 575 nm increases with increased loading of tetracene;
this is due to tetracene aggregates emitting (Figure 3-11, b, vide supra).
Emission from exciting tetracene was hypsochromically shifted and narrowed as
compared to that of Z-T (Figure 3-12, b) although the emission still broadens with
increased loading (Figure 3-17, c). This is due to BODIPY contribution, as evidenced by
the shoulder from BODIPY at 540 nm. This indicates that FRET is occuring in this system.
The quantum yield of emission from both tetracene and BODIPY in Z-T-BD was
measured (Figure 3-17, d and Table 3-4). The quantum yield from both tetracene and
BODIPY was suppressed, making it is difficult to analyze if FRET is occuring this way.
The low quantum yield is likely due to self-quenching in the solid state.

84

350 400 450 500 550
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0.16

Kubelka-Munk (a.u.)
Wavelength (nm)
(a)
550 600 650 700 750
0.0
0.2
0.4
0.6
0.8
1.0

Normalized Intensity (a.u.)
Wavelength (nm)
(b)
500 550 600 650 700 750 800
0.0
0.2
0.4
0.6
0.8
1.0

Normalized Intensity (a.u.)
Wavelength (nm)
(c)
400 410 420 490 500 510
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
Quantum Yield
Wavelength (nm)
(d)
0.00
0.01
0.02
0.03
0.04
0.05
0.06
Kubelka Munk (a.u.)

Figure 3-17: (a) Kubelka Munk function of tetracene in zeolite at various loadings, capped with
BODIPY-dipyridinium; 0% (red), 25% (cyan), 50% (light blue), 75% (blue) and 100% (dark blue)
(b) emission of the same samples, excited at 490 nm. (c) emission of the same samples, excited
at 445 nm (d) quantum yield of 25% (cyan) and 100% (dark blue) loaded Z-T-BD as a function of
wavelength, absorbance of Z-BD (red)

This study indicates that as loading
increases, more tetracene aggregates and
crystalline states are formed. FRET from
tetracene to BODIPY occurs in the Z-T-
BD system. FRET can aid in light
harvesting in an artificial photosynthetic system, but electron transfer is also necessary .
To observe electron transfer to aid in photocatalytic water splitting, FRET should be
eliminated. FRET is caused by the overlap between the emission of tetracene and
absorption of BODIPY. Blue-shifting the dye’s emission, or red-shifting the acceptor’s
absorption, can eliminate this overlap. Therefore, anthracene was studied as a donor dye
Table 3-4: Quantum yield (Φ) of Z-T-BD
samples

Φ
tetracene
Φ
BODIPY
Z-T25 -BD 0.03 0.07
Z-T50 -BD 0.03 0.07
Z-T75 -BD 0.03 0.07
Z-T100-BD 0.03 0.05
Z-BD - 0.04

85

molecule in zeolite-L.
3.3.2 Electron Transfer in LHUs
Electron transfer in LHU’s was investigated between anthracene (D) and bodipy-
dipyridinium (A). The spectral overlap between these two
3.3.2.1 Anthracene Loading in Zeolite-L
Anthracene is a well characterized small molecule donor. Its small size (9.8Å x 5.6Å
x 3.3Å) allows it to fit in the channel of zeolite-L, it π-stacks for improved energy transfer,
and it has a large extinction coefficient and strong emission (Figure 3-18 and Table 3-5).
It is not as a good of an electron donor as
tetracene, but it absorbs in the UV and its
emission does not overlap with the absorption
of BODIPY (Figure 3-18, a).

Anthracene has been extensively studied in many systems, including zeolite-dye
assemblies. A previous study by Hashimoto et al. investigated the packing of anthracene
in the channels of zeolite-L, wherein they determined that the cation exchanged in the
channels of zeolite can impact how anthracene stacks (see Figure 3-19).
30
At high loading
300 350 400 450 500 550 600 650
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
(a)
-6 -5 -4 -3 -2
-3E-5
-2E-5
-1E-5
0
1E-5
2E-5
3E-5
-4.039
Energy (eV)

-5.31
-2.28
-2.385
-2.98
-3.076
-5.41
Potential (A)
(b)

Figure 3-18: Absorption spectra (dashed line) and fluorescence spectra (solid line) in ethanol
and (b) cyclic voltammetry in DMF of anthracene (green) and BODIPY dipyridinium (red).
Inset: Molecular structure of anthracene.
Table 3-5: Lifetime, quantum yield and
radiative rate of anthracene in ethanol,
λex=345 nm
τ (ns) Φ kr (x10
7
s
-1
)
2.2 0.27 12.3

86

in zeolite NaKL they observed an intermolecular excimer structure; whereas in zeolite KL
the excimer emission at 530 nm had a fluorescence lifetime of 50 ns, which suggests that
the two anthracene molecules adopt a co-facial arrangement. They determined that the
cation exchanged in the channels as well as the size of the molecule can affect dye packing
in the channels.

Anthracene was loaded in dehydrated zeolite-L (Z-A) with the same loading as that of
tetracene (25%, 50%, 75% and 100% of pore volume) and the photophysics of the system
were investigated (see Figure 3-20).
As with Z-T, the absorption and emission of Z-A were broadened and red- shifted
from the monomeric species. The absorption does not change with increased loading,
unlike Z-T, where Davydov splitting caused a shoulder to come in at longer wavelengths
with higher loading. The line shape of emission of anthracene appears similar to the
monomer, except for a broadening at longer wavelengths with increased loading. This is
indicative of aggregate formation, and not excimer emission, as a distinct excimer peak

(a) (b) (c)
Figure 3-19: (a) Absorption and corrected emission spectra of anthracene adsorbed in zeolite
NaKL at two loading levels: (A) 2.0 x 10−6 mol g
−1
; (B) 6.0 x 10
−5
mol g
−1
; (b) 4.0 x 10
−6
mol
g
−1
adsorbed in zeolite KL; (c) Geometry of associated anthracene molecules in the channel
depending on the cation (top) KL and (bottom) NaKL.

87

was not observed (such as that measured by Hashimoto et. al).
30

325 350 375 400 425 450
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0.16
Kubelka- Munk (a.u.)
Wavelength (nm)
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
(a)
400 450 500 550 600 650
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
(b)

Figure 3-20: (a) Kubelka-Munk function of anthracene in solution (green) and adsorbed in
zeolite-L at various loadings: ZA25 (light grey), ZA50 (grey), ZA75 (dark grey) and ZA100 (black);
(b) emission spectra of the same samples, excited at 360 nm.

The lifetime and quantum yield were measured, and as loading increased the
quantum yield of fluorescence from anthracene decreased (Table 3-6). This is caused by
anthracene aggregates self-quenching. There are yet again three lifetimes: a very short
lifetime, one that is around that of anthracene’s monomer and one long lifetime. The
decays are plotted to better understand the difference in samples (see Figure 3-21).

Table 3-6: Lifetime, lifetime contribution (A) and quantum yield of ZA

τ1 (A1) (ns) τ2 (A2) (ns) τ3 (A3) (ns) Φ
ZA25 0.3 (65) 5 (31) 15 (4) 0.09
ZT50 0.5 (54) 4 (43) 15 (3) 0.06
ZT75 0.5 (61) 2 (36) 12 (3) 0.04
ZT
100
0.3 (68) 2 (29) 11 (3) 0.03
88

Anthracene’s lifetime decay has the opposite
trend as that of tetracene; as loading is
increased, the lifetime is quenched. The
aggregate formation is causing self-
quenching. Whereas in tetracene when
aggregates were formed delayed fluorescence
dominated the emission, in anthracene this is
not the case. The short lifetime is attributed
to interaction with the zeolite-L framework. τ2
is caused by anthracene monomers, and the
long lived lifetime can be attributed to charge transfer states between anthracene and
zeolite.
3.3.2.2 Z-A-BD Photophysics
BODIPY-dipyridinium was fully loaded on the surface of the Z-A samples (Z- A-BD)
with the same technique as that used for the Z-T samples. The photophysics were
measured and compared to the Z-A samples. The absorbance is similar to that of Z-A,
except for additional absorbance from BODIPY at 525 nm (Figure 3-22, a). The
fluorescence spectra from exciting BODIPY resulted in only BODIPY emission, which was
broadened compared to Z-BD (Figure 3-22, b). This indicates the presence of anthracene
impacts the excited state of BODIPY, as broadening can be the result of a charge-transfer
state.
0 5 10 15 20 25 30 35
100
1000
10000

Counts
Time (ns)

Figure 3-21: Lifetime decay of Z-A at
various loading: ZA25 (light grey), ZA50
(grey), ZA75 (dark grey) and ZA100 (black).
Samples were excited at 375 nm and
decay measured at 409 nm

89

350 400 450 500 550 600
0.00
0.02
0.04
0.06
0.08
0.10
0.12
Kubelka Munk (a.u.)
Wavelength (nm)
(a)

550 600 650 700 750
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
(b)

400 450 500 550 600 650
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
(c)

0 5 10 15 20 25
100
1000
10000

Counts
Time Delay (ns)
(d)

Figure 3-22: (a) Absorption spectra of anthracene adsorbed in zeolite-L and capped with
BODIPY-dipyridinium at various loadings ZA25BD (light grey), ZA50BD (grey), ZA75BD (dark
grey) and ZA1 00BD (black); (b) emission spectra of the same samples, excited at 500 nm; (c)
excited at 360 nm and (d) lifetime decay of the same samples, excited at 375 nm.

Anthracene emission (λex=375 nm) from Z-A-BD samples was similar to that of the
Z-A samples, except for an additional peak at 540 nm, from BODIPY emission (Figure
3-22, c). As BODIPY was not excited, this indicates anthracene transferred excited state
energy to BODIPY, which then emitted. This was especially prevalent in Z-A25-BD. The
quantum yield and lifetime were measured (Figure 3-22, d and Table 2.10). Both the
quantum yield of Z-A-BD were quenched from the Z-A samples, and BODIPY’s quantum
yield is quenched with increased loading of anthracene.
Photoinduced electron transfer (PET) is occurring in this system, as FRET was
minimized and both anthracene and BODIPY’s emission were quenched with increased
loading of anthracene. While there was still some orbital overlap between anthracene and
90

BODIPY due to the aggregated states broadening and red shifting the emission of
anthracene, this was not contributing to energy transfer, but actually quenching it. This
is evidenced by the quantum yield of anthracene, which is reduced as the loading
increases. BODIPY’s quantum yield was most indicative of PET, since it decreased with
increased loading of anthracene. As anthracene loading increases, the polycrystalline
states formed facilitate electron transfer to BODIPY. This electron transfer forms radical
anions, which non-radiatively decay.
Unfortunately, the change in quantum yield is 0.02, which is very low. BODIPY
itself has a high quantum yield, but it is self-quenched on the surface of zeolite-L. To
regain the quantum yield, BODIPY was diluted on the surface of zeolite-L.
Table 3-7: Quantum yield (φ) of Z-A-BD samples
Further examination of the dynamics of electron transfer in this system would require
transient absorption spectroscopy. This could measure the spectra of any charge transfer
complexes between anthracene and BODIPY that are being formed in the ps time regime.
However, this is very difficult to measure in this system, as zeolite-L is a solid-state
powder that scatters light effectively. Transient absorption measurements on zeolite-L
systems have only been done using diffuse reflectance transient absorption spectroscopy,
which is a rare and specialized technique.
Another method to determine whether electron transfer is occuring, is to utilize the
LHU’s as a photosensitizer in a photocatalytic hydrogen evolution reaction cell. These
experiments were set up in Prof. Peter Brüggeller’s lab at the Universität Innsbruck, and
Φ anthracene Φ BODIP Y
Z-A25-BD 0.05 0.04
Z-A50-BD 0.05 0.04
Z-A75-BD 0.03 0.03
Z-A1 00-BD 0.02 0.02
Z-BD - 0.04
91

executed by myself and Dr. Christof Strabler.
3.3.3 Water Splitting in LHUs
3.3.3.1 Hydrogen Evolution Reaction (HER)
HER was studied using the developed LHUs as a photosensitizer, an HER catalyst,
and a sacrificial donor. The turnover number (TON) was calculated, which is a ratio
between the moles of product, and the moles of reactant studied (hydrogen gas and
BODIPY) (see equation (3-4)):
(3-4) 𝑇𝑂𝑁 =
𝑚𝑜𝑙𝑠 𝑜𝑓 𝐻 2
𝑝𝑟𝑜𝑑𝑢𝑐𝑒𝑑 𝑚𝑜𝑙𝑠 𝑜𝑓 𝑠𝑡𝑜𝑝𝑐𝑜𝑐𝑘

3.3.3.2 Molecular HER Catalysts
To determine the optimal HER catalys to use with the LHU’s developed, a series of
hydrogen evolution catalysts were synthesized that had previously demonstrated efficient
water splitting with various photosensitizers. The HER molecular catalysts investigated
were [Pd(PPh3 )Cl2]2, [Ni(pyS)3]
−
, Co-N4 and Co-S4 (see Figure 3-23). [Pd(PPh3 )Cl2]2 is
the only hydrogen evolution reaction molecular catalyst that had been sensitized by a
BODIPY and undergone water splitting.
31
[Ni(pyS)3 ]
−
has demonstrated very high
catalytic activity for a noble-metal free catalyst, and so was chosen.
32
Co-N4 and Co-S4 are
also both catalysts made of earth abundant materials, and operate under basic and acidic
conditions, respectively;
33,34
this will determine what conditions the system works best
under. HER activity that has been previously demonstrated is given in Table 3-8.

92

(a) (b) (c) (d)

Figure 3-23: HER catalysts (a) [Pd(PPh3)Cl2]2 (b) Co-N4 (c) [Ni(pyS)3]
−
(d) Co-S4

Table 3-8: Catalyst, photosensitizer, conditions and TON from literature
Catalyst Photosensitizer pH Hours TON
[Pd(PPh3)Cl2]2 BODIPY 4.5 20 304
CoS4 Ru(bpy)3
2+
4.5 20 2,700
CoN4 [Ir(ppy)2(bpy)][PF6] 11.5 4 307
[Ni(pyS)3 ]
−
Fluorescein 11.5 40 5,500

HER catalysts were screened with Z-A100-BD to determine which catalyst was most
active with the LHU (Table 3-9). Basic conditions worked best for the system, as CoN4
and [Ni(pyS)3]
−
were the most active catalysts, with [Ni(pyS)3 ]
−
being the most active
with a TON of 465 after 20 hours.
Z-A100-BD was used as it showed the greatest quenching in BODIPY, which is
assumed to be due to PET. Reference samples were subsequently compared to Z-A100-BD
by measuring the catalytic activity of the individual components of the system (BODIPY-
dipyridinium in solution and Z-A100). The TON of these systems were effectively null,
thus, it only works with the complete system (Table 3-9). This experiment demonstrated
the LHU designed and synthesized works as an effective electron donor to promote the
hydrogen evolution reaction.
93

Table 3-9: Catalyst, photosensitizer, hours and TON measured

3.3.3.3 Catalytic Palladium Nanoparticles
Palladium nanoparticles (Pd-NP) were synthesized and used as an electrode catalyst
for HER. The efficiency of energy transfer between the LHU and Pd nanoparticle was
investigated by intra- and inter- molecular energy transfer (see Figure 3-24).
Intramolecular energy transfer was studied by binding Pd nanoparticles to the surface of
Z-A-BD. The synthesis and functionalization of Pd-NPs on zeolite-L was made following
the procedure by Mandal et. al (see experimental section).
35
The intermolecular Pd-NP’s,
synthesized in Peter Brüeggler’s lab, were 4-5 nm in size, stabilized by polyethyleneglycol
and modified with pyridine to activate the Pd
center. Intermolecular energy transfer was
investigated by adding the Pd NP to solution with
Z-A100-BD. The HER experiment was done in 1:1
H2O:ACN, with 5% asorbic acid and a pH of 4.5.
The TON was measured with both systems as well
as Pd- NPs and Z-A system (Table 3-10).

Catalyst Photosensitizer Hours TON
[Pd(PPh3)Cl2]2 Z-A100-BD 20 87
CoS4-catalyst Z-A100-BD
20 0
CoN4 Z-A100-BD
20 127
[Ni(pyS)3 ]
−
Z-A100-BD 20 465
[Ni(pyS)3 ]
−
Z-A100 20 0
[Ni(pyS)3 ]
−
BODIPY 20 3.5

(a) (b)
Figure 3-24: (a) Intra- and (b)
intermolecular Pd nanoparticle: LHU
system

94

Table 3-10: Catalyst, photosensitizer, hours and TON measured
Catalyst: Pd nanoparticles
Turnover Number
20 h 40 h 60 h
Intramolecular
Z-Pd NP 0 - -
Z-A100-Pd NP 90 103 0
Z-A100-BD-Pd NP 122 163 180
Intermolecular
Z-Pd NP 0 - -
Z-A100-Pd NP 0 - -
Z-A100-BD-Pd NP 114 125 0

Both systems prove inter- and intra- molecular catalytic systems can be made with Pd
nanoparticles and the LHUs developed. Intramolecular HER was more efficient, as this is
not a diffusion-limited reaction. Intramolecular HER occurred with the Z-A system,
though it was more efficient with the BODIPY. As anthracene is an electron donor, it likely
photosensitized the Pd nanoparticles. Diffusion of Pd nanoparticles to the surface of the
LHU in intermolecular HER limits electron transfer, so hydrogen evolution was only
observed in the case of the entire LHU.
These promising results led to further investigation of optimizing electron transfer in
LHU’s. As previously mentioned, self-quenching of BODIPY on the surface of zeolite-L
leads to non-radiative decay. To prevent this, BODIPY was diluted with other stopcock
molecules.
3.3.4 Optimizing Electron Transfer in LHUs
Diluent stopcock molecules were synthesized to investigate prevention of self-
quenching of BODIPY-dipyridinium on the surface of zeolite-L. Stopcock molecules
investigated were fluorescent naphthalene-pyridinium and benign diether- pyridinium.
3.3.4.1 Diluent Stopcock Molecules
Naphthalene-pyridinium was chosen because naphthalene is a wide-bandgap
material, so it does not interact electronically with BODIPY. Naphthalene has been well
95

characterized; its photophysical properties are found in Table 3-11.
36

Table 3-11: Photophysical properties of naphthalene and naphthalene-pyridinium

Interestingly, naphthalene-
pyridinium has very different
photophysics from the parent
compound, naphthalene. The
absorbance and emission are strongly
bathochromically- shifted and
broadened (see Figure 3-25). This is
likely caused by twisted
intramolecular charge transfer
(TICT) in the excited state, which has been observed in a similar molecule, 4-[2-(4
dimethylaminophenyl)ethenyl]- 1-methylpyridinium iodide.
37
TICT is a photoinduced
process in D-A dyads, wherein the excited state favors a different conformer that is
twisted about a bond. This creates an excited state, P*, that is different from the locally
excited state, R*. The TICT state is stabilized because of charge transfer from the electron
donor (naphthalene) to the electron acceptor (pyridinium). This causes a hypsochromic
shift in the excitation spectrum as compared to the ground state absorption; this was
observed for naphthalene-pyridinium (see Figure 3-25). The photophysical properties are
given in Table 3-11.
The radiative rate for naphthalene pyridinium is two orders of magnitude greater
λabs (nm) λem (nm) τ (ns) Φ
k r
(x10
7
s
-1
)
knr
(x10
7
s
-1
)
Naphthalene 275 325 100 0.2 0.2 0.8
Naphthalene-
py ridinium
394 57 5 3.8 0.94 25 1.6
300 400 500 600 700 800
0
1E4
2E4
3E4
4E4
5E4
6E4
7E4
8E4
Molar Absorptivity (M
-1
cm
-1
)
Wavelength (nm)
0.0
5.0E3
1.0E4
1.5E4
2.0E4
2.5E4
Intensity (a.u.)

Figure 3-25: Naphthalene- pyridinium (inset)
absorbance (solid dark blue), emission (solid
blue) and excitation (dashed blue). Emission
was excited at 400 nm, excitation spectra was
measured at 575 nm, in acetonitrile.

96

than naphthalene. TICT can be further investigated by measuring emission in solvents of
varying polarity, as the charge transfer state will be destabilized in polar solvents.
Diether-pyridinium was
synthesized as a precursor to BODIPY-
pyridinium, and was found to be a great
diluent molecule for its size and lack of
photophysical properties. It is an ideal
candidate for stopcock dilution as it is not
emissive and it absorbs in the UV (see
Figure 3-26).
3.3.4.2 Diluting BODIPY-dipyridinium on zeolite
Samples were prepared in which the zeolite-L pores were fully stoppered, and the
percentage of diluent and BODIPY-dipyridinum was varied (denoted Z-NPX:BD Y,
wherein X and Y are the percentages of naphthalene-pyridinium and BODIPY-
dipyridinium, respectively). The absorption spectra were measured using diffuse
reflectance UV-vis, and the emission spectra of BODIPY were measured by exciting at 490
nm (Figure 3-27, a, b). The absorbance spectra had two distinct peaks from BODIPY (520
nm) and naphthalene (375 nm). The emission was measured by exciting BODIPY
(λex=490 nm), which confirmed BODIPY was the only species emitting, (see Figure 3-27
(b)). BODIPY was diluted from NP1 00:BD0 to NP99:BD01 and the quantum yield measured
(Figure 3-27 (c)). The self-quenching of BODIPY is observed; at low concentrations of
BODIPY, the quantum yield is almost the same as that in solution (0.59). As the
concentration increases, the quantum yield decreases. The lifetime decay also approaches
solution phase BODIPY lifetime when decreasing the concentration of BODIPY on the
275 300 325 350 375 400 425 450
0
1E4
2E4
3E4
4E4
5E4
6E4
7E4
8E4
9E4
Molar Absorptivity (L mol
-1
cm
-1
)
Wavelength (nm)

Figure 3-26: Diether- pyridinium (inset)
absorbance in DCM:MeOH

97

surface (Figure 3-27, d).
300 350 400 450 500 550 600
0.0
0.2
0.4
0.6
0.8
1.0
Kubelka-Munk (a.u.)
Wavelength (nm)
(a)

525 550 575 600 625 650 675 700
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
(b)

20 40 60 80 100
0.0
0.1
0.2
0.3
0.4
0.5
0.6
Quantum Yield
BODIPY on Surface (%)
(c)

15 20 25 30 35 40 45 50 55
0.01
0.1
1
NP
0
:BD
100
NP
25
:BD
75
NP
50
:BD
50
NP
75
:BD
25
NP
80
:BD
20
NP
85
:BD
15
BODIPY (soln)
Normalized Decay (a.u.)
Time (ns)
(d)

Figure 3-27: (a) Kubelka-Munk and (b) emission spectra excited at 490 nm of 0NP:100BD (red),
25NP:75BD (orange), 50NP:50BD (light green), 75NP:25BD (dark green) and 100NP:0BD
(blue) (c) quantum yield of BODIPY with decreasing concentration on the surface of zeolite-L;
(d) lifetime decay of the same samples excited at 405 nm, as well as NP80:BD20 (blue), NP85: BD 15
(dark blue) and BODIPY-dipyridinium in acetonitrile (black).

BODIPY quantum yield was regained on the surface of zeolite-L through dilution.
However, naphthalene-pyridinium absorbs at the same wavelength as anthracene, and
emits at the same wavelength of BODIPY, making it difficult to distinguish. Thus, a wide-
bandgap, non-emissive stopcock molecule is ideal to determine if naphthalene-
pyridinium is also participating in electron or energy transfer.
BODIPY-dipyridinium was diluted with diether-pyridinium (DP) on the surface of
empty zeolite-L. This was to determine if the BODIPY radiative rate could be recovered,
such as was done with naphthalene-pyridinium. Samples were prepared with calculated
98

100:0, 75:25, 50:50, 25:75 and 0:100 BODIPY:DP loading on the surface of empty zeolite.
The emission spectra of these samples indicate that as loading is decreased, the BODIPY
emission starts to resemble that of the monomer, as the 25:75 sample is narrowed and
blue-shifted compared with those in which the BODIPY is more packed (see Figure 3-28).
300 325 350 375 400 425 450 475 500 525 550
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
(a)
525 550 575 600 625 650 675 700
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
(b)

Figure 3-28: BODIPY diluted on empty zeolite-L with diether pyridinium (DP) in 100:0
BODIPY:DP (red), 75:25 (orange), 50:50 (green) and 25:75 (dark green) (a) excitation spectra
(emission λ=580 nm) and (b) emission spectra (excitation λ=490 nm)

3.3.4.3 Energy and Electron Transfer in LHUs
The complete LHU was assembled using zeolite loaded with anthracene, and capped
with BODIPY-dipyridinium diluted with diether-pyridinium at various concentrations
(100:0, 75:25, 50:50, 25:75 and 0:100 BODIPY:DP). The steady state photophysics were
measured (see Figure 3-31 and Table 3-12).
An important note here is that the samples of zeolite loaded with anthracene had
been assembled 24 months prior to stopcock loading and the subsequent experiments.
The intense electrostatic fields in zeolite (due to the ionic framework) has been found to
99

stabilize polar adsorbates. This has proven to
create electronic confinement of anthracene in
zeolite pores, which increase the MO energies,
especially the HOMO.
38
Anthracene
encapsulated into zeolite for over a year results
in a strong confinement effect as it interacts
with the counter-cations.
39
This stabilizes the
radical cation formed with anthracene when it
charge transfers to the zeolite after
irradiation.
3 9 ,40
These effects have been
observed in ZSM-5, wherein the pore sizes are less than 6 nm, occluding more than one
anthracene molecule per pore (which prevents photodimerization). The pores of zeolite-
L are large enough to allow more than one molecule per cage, so photodimerization is
possible. Anthracene dimers and excimer states have been observed in zeolite Y
nanocages.
41
The lifetime of CT complexes in zeolite with anthracene and an acceptor
molecule has been observed on the order of μs, as the polar environment of zeolite
stabilizes the cation formed and inhibits back electron recombination.
42

The photophysics of the samples with varying concentrations of BODIPY on the
surface are all similar, except that there is a notable peak from BODIPY in Figure 3-31 (b),
when anthracene is excited. This indicates that there is energy transfer from anthracene
to BODIPY, which is then emitting. While energy transfer is occuring, it is not the
dominant mechanism, as this would result in a net increase in the radiative rate. As shown

Figure 3-29: Room temperature FT ‐
Raman spectra (1064 nm excitation)
recorded as a function of time after
mixing solid ANT and dehydrated
Rb3.4ZSM ‐5 zeolite with 1
molecule/unit cell loading: a) ANT
powder; b) 20 days; c) 84 days; d) 135
days; e) 340 days. I=Raman intensity
in arbitrary units.
39

100

in Figure 3-30, the radiative rate decreased
with decreasing concentration. As it is
quenched in the presence of anthracene, this
is likely due to either trap states being
formed, or photoinduced electron transfer.
Of particular note is how the lifetime of these
samples has changed from those in section
3.3.2.2; it is three orders of magnitude longer.
A microsecond lifetime indicates that when BODIPY is excited, it charge transfers to
anthracene, and a charge separated state is formed. Zeolite’s strongly polarizing
environment stabilizes charge separated states, which is eventually able to recombine and
radiatively decay. Hydrogen evolution reaction experiments were performed to
determine how well it acted as a photosensitizer.
Table 3-12: Z-A-BDX-DPY samples photophysical properties
(lifetime: λex=473 nm, measured λem=540 nm, Φ: λex=480 nm)
τ1 (µs) Φ kr (x10
4
s
-1
)
ZA-BD1 00-DP0 1.21 0.08 7.1
ZA-BD75-DP25 1.20 0.06 5.3
ZA-BD50-DP50 1.81 0.10 6.4
ZA-BD25-DP75 1.24 0.09 6.3

0 25 50 75 100
0.00
0.05
0.10
0.15
0.20
0.25
Quantum Yield
% BODIPY Loading

Figure 3-30: Quantum yield as a function
of BODIPY concentration for Z-BD (blue)
and Z-A-BD (red)
101

300 325 350 375 400
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
(a)

400 450 500 550 600
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
(b)

300 350 400 450 500 550
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
(c)

550 600 650 700
0.0
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
(d)

Figure 3-31: ZA-BD1 00-DP0 (green), ZA-BD75-DP25 (light blue), ZA-BD50-DP50 (dark blue), ZA-
BD25-DP75 (purple) and ZA-BD0-DP100 (pink) (a) excitation spectra of anthracene component,
measured at emission wavelength of 440 nm, (b) emission spectra of anthracene component,
excited at 375 nm, (c) excitation spectra of BODIPY component, measured at emission
wavelength of 580 nm and (d) emission spectra of BODIPY component, excited at 490 nm.

3.3.5 Optimizing LHU Electron Transfer
BODIPY was diluted on the surface of both zeolite and zeolite-anthracene with
[Ni(pyS)3]-. The pH was consistently kept at 11.5, with 5% TEA and the concentration of
[Ni(pyS)3]- is in 10x+ excess of the photosensitizer.

102

Table 3-13: Experimental Conditions and TON measured
Sample Light Source Solvent TON
ZABD1 00
Hg EtOH 7
525/365 EtOH 4
Hg Acetonitrile 18
525/365 Acetonitrile 21
ZABD75
Hg EtOH 8
525/365 EtOH 4
ZABD50 Hg EtOH 2
ZABD25
Hg EtOH 14
525/365 EtOH 6
Hg Acetonitrile 35
525/365 Acetonitrile 22
ZABD0
Hg EtOH 11
Hg Acetonitrile 14
525/365 Acetonitrile 7

At first glance, these results demonstrate that the Hg lamp works best, and the
sample with BODIPY diluted to 25% had the best performance. This is expected, as this
sample prevents BODIPY from self-quenching. However, the TON’s from these
experiments are significantly reduced from the previous experiments. Further
investigation was necessary.
These experiments were repeated for longer times (Table 14). After roughly 20h
of irradiation hydrogen is being produced, with TON values related to the chromophore
between 200 and 500. It was found that the ZABD25DP75 sample had the best TON of 573
after 89 hours. ZABD0DP1 00 also had a high TON of 543.6. This indicates that the diether-
103

pyridinium is participating in
electron transfer, and acting as
an electron acceptor. Thus, in
sample ZABD25DP75, both BD
and DP are acting as acceptors
and contributing to the HER. In
samples where hydrogen was produced, a a black precipitate formed. The nature of this
precipitate was not investigated (eg. By TEM or DLS) but the formation of an active Ni(0)
species is known in literature.
43,44

Background measurements of the [Ni(pyS)3 ]TEA catalyst were done, with no PS.
It is evident that at least after 20 hours of irradiation an intense breakdown of the catalyst
takes place (formation of black precipitate and discoloration of the green/yellow Ni(II)
solution). This breakdown seems to go along with a stoichiometric production of
hydrogen. There is an initiation phase of the reaction evident, but the time of this
initiation is not dependent on the concentration of the WRC or the chromophore.
Therefore the hydrogen production cannot be clearly assigned to a photosensitized
reaction of the employed chromophore. However, as there was hydrogen produced from
the LHU’s employed, they are likely active with Ni(0) species. Further investigations are
necessary to confirm these results.
The previous results in section 3.3.3.2 did not consider the photoinduced
breakdown of the WRC, which was likely also occuring. Another explanation of the
different results may due to the occlusion of anthracene in zeolite-L over time, which has
formed different species, including a long lived charge transfer state.
Through a series of experiments, energy and electron transfer have been observed in
Table 14: TON of HER with WRC, [Ni(pyS)3]
−
at a pH
of 11, irradiated with 150W Hg

TON
16 hours 19 hours 89 hours
ZABD0DP1 00 1.6 48.4 543.6
ZABD25DP75 2.1 25.5 573.1
ZABD50DP50 249.3 2.2 448.7
ZABD75DP25 0 2.5 257.3
ZABD1 00DP0 - - 119.3
No PS 0 0.3 1.1

104

the developed LHU’s. HER experiments were performed that prove LHUs are a viable
tool for water splitting. The next steps further pursue the dominant photophysical
methancism, FRET or electron- transfer. BODIPY’s quantum yield was regained by
diluting it with stopcock molecules.
3.4 Conclusions and Future Outlooks
Molecular organization of dye molecules in zeolite-L as a light harvesting unit for
HER water splitting was demonstrated. It was found that increased loading formed
aggregated states that quenched the stopcock emission, which could be regained through
dilution. PET most likely occurred from BODIPY-dipyridinium to anthracene in zeolite-
L, as hydrogen evolution occurred in the presence of various WRC. Intramolecular HER
through covalently bonding Pd catalysts to the surface of the LHU was more efficient than
intermolecular HER. Synthesis and characterization of various stopcock molecules
demonstrated electron transfer could be optimized by diluting a non-interacting stopcock
into the active electron acceptor layer. Water splitting through the hydrogen evolution
reaction occurred best when using the LHU in conjunction with [Ni(pyS)3 ]
−
as the HER
catalyst and
triethylamine as the sacrificial donor. This is likely due to decomposition of [Ni(pyS)3]
−

to Ni(o), which is an active catalyst and participated in conjunction with the LHU.
3.4.1 Future Recommendations
Next steps for this project would look into the viability of the LHU, Z-A-BD, as a
compatible photosensitizer with Ni nanoparticles. HER can be investigated both
intramolecularly and intermolecularly. The role of the diluent molecule in HER
experiments can also be further probed, first by measuring the redox properties through
electrochemistry. This will identify if it is capable of accepting or donating electrons from
105

the anthracene and BODIPY, respectively. Future work can also include encapsulating
and stoppering various molecules in zeolite. Making a monolayer of LHU’s on an
electrode surface can also help measure the charge extraction capabilities of this system.
In this case, a thinner zeolite would be ideal (on the order of 10’s of nm), as well as
functionalizing the surface with the stopcock molecule after forming the monolayer, such
that it is assymetrically functionalized on zeolite.
3.4.2 Challenges with LHUs
Results show promise for successful application of microparticulate LHUs in HER,
however, this is a challenging system to work with for multiple reasons:
• Light scattering of zeolite-L: Zeolite-L inherently scatters light, which is why
diffuse reflectance measurements are necessary for measurement. However, scattering is
not completely eliminated, and thus can skew the
data.
• Multiple states: It was hypothesized that a polycrystalline state may form in the
zeolite-L, but as the zeolite-L has defects itself, the dye molecules
do not organize perfectly. Thus, there are multiple states, monomers and aggregates, and
these all impact the overall system.
• Solid-state measurements: As zeolite-L is an alumino-silicate scaffolding that
organizes the dye molecules, it is a solid-state system. This makes characterization
difficult, as most instrumentation is optimized for solution phase characterization.
There are various approaches to making donor-acceptor systems for light harvesting
for water splitting. This approach was to use supramolecular organization of D-A
molecules for electron transfer; a molecular system is proposed.

106

3.5 Experimental Methods
3.5.1 Synthesis of Zeolite-L
Hydrothermal synthesis was used to make zeolite-L (see Figure 3-32). The size and
morphology of zeolite-L varies depending on the gel composition, specifically the
H2O/SiO2 and K2O/SiO2 ratios. The type of silica source and whether or not the
crystallization was done dynamically or statically can also affect the zeolite-L size and
shape.
5

Figure 3-32: Experimental scheme for zeolite-L synthesis

To obtain disc-shaped zeolite-L that is 1 m in diameter and 200 nm thick, the
aluminate component (solution A) is prepared by dissolving 9.32 g of KOH (Fluka, purum
p.a. 85%) and 5.87 g of sodium hydroxide (~98%, Fluka) in
59.5 g of doubly distilled water. Aluminum hydroxide (2.1 g) is added once the hydroxides
107

are fully dissolved. The reaction mixture is refluxed for 3 h in order to dissolve the
aluminum hydroxide. The clear solution is cooled to room temperature and water loss is
compensated. The silicate component (solution
B) is prepared by mixing 58.9 g of Ludox HS-40 (40% SiO2, Aldrich) with 4.8
g of ethanol (Honeywell, p.a.) and then sonicating the mixture for 30 minutes at room
temperature. Solution A is then poured quickly into solution B under strong stirring and
the gel is aged for 10 min under stirring. The white gel is then evenly split in two PTFE
pressure vessels (40 mL volume). Crystallization
takes place at 160 °C for 72 h in a rotating oven at a speed of 20 rpm. Once the reaction is
finished, the vessels are cooled in an ice bath for 1 hour before
opening them. The milky white suspension is then centrifuged for 15 min at 3100 rpm.
The white residue is then washed with boiling doubly distilled water until the pH of the
supernatant is 7.
The cation sites in the pores of the zeolite-L are at this point exchanged with both
K+ and Na
+
. To fully exchange with K
+
, 1 g of zeolite-L was dispersed in 1M KNO3 and
stirred for 3 hours. The K
+
exchanged material is
then washed three times with doubly distilled
water to remove excess salt. Sample is dried in
the oven overnight.
The final shape and thickness of zeolite-L is
measured by SEM, and was found to be 200 nm
in diameter and 1 m thick (Figure 3-33).

Figure 3-33: SEM of zeolite-L used
(200 nm thick, 1 um diameter)

108

3.5.2 Pore Loading
To load densely packed neutral dyes
in zeolite-L, gas phase adsorption has
been utilized. Approximately 100-200
mg of potassium exchanged zeolite L
was mixed with n-dye (moles) of pure dye in a glass ampoule (wherein n-dye depends on
the level of loading, p, desired).
The mixture was dried on a high vacuum line for 6 hours at 1 x 10
−5
mbar before sealing
the ampoule. This removed water molecules bound inside the zeolite-L. Dye insertion
then took place in a rotating oven at the sublimation temperature (see Table 3-16) for 1-4
days. The dye-aloaded material was then washed with 15 mL n-butanol, which was
centrifuged at 2900 rpm for 15 min. This was repeated until the supernatant solution does
not show absorption of the dye molecules by UV-vis spectroscopy. This was to ensure
removal of dyes adsorbed on the external surface. The effective loading degrees were
determined by HF analysis, TGA or XPS.
Table 3-16: Sublimation temperature of dye molecules used
Dye Sublimation Temperature
Anthracene 180°C
Tetracene 200°C

Tetracene was loaded into zeolite-L, and the total amount loaded was calculated by
TGA (Table 3-15).
To determine that tetracene is being loaded into the pores of the zeolite, and not just
adsorbed to the exterior, a sample in which tetracene was crushed on the outside of
zeolite-L was prepared. This was compared to that of tetracene sublimed in zeolite, and
Table 3-15: Mass Loss of Tetracene in Zeolite-L
Sample Mass Loss (%) Mass Loss (mg)
ZT25 1.2 1.23
ZT50 2.4 2.47
ZT75 3.6 3.7
ZT
1 00
4.7 4.95

109

there was a clear difference in color resulting from the changes in molecular packing (see
Figure 3-34 (a)). The photophysical properties of this sample was compared to that of
tetracene powder (Figure 3-34 (b) and (c)). The absorption of the adsorbed samples has
a peak at 500 nm with a shoulder at 525 nm, which is similar to that of a tetracene film
and aggregates (Figure 3-10), as well as monomeric defects (Figure 3-11). The emission
has one peak at 550 nm, indicative of aggregates (Figure 3-10). However, the sample of
tetracene that is sublimed into zeolite has absorption similar to that of monomeric
tetracene, albeit a shoulder into the red that could be monomer defects or excimers in the
zeolite. The emission is one broad peak, which has been attributed to the excimer state of
tetracene.
25

350 400 450 500 550
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0

Normalized Intensity
Wavelength (nm)
500 550 600 650 700 750 800
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0

Normalized Intensity
Wavelength (nm)

(a) (b) (c)
Figure 3-34: (a) Tetracene adsorbed on zeolite, before sublimation (top), tetracene sublimed in
zeolite (bottom); (b) excitation and (c) emission of tetracene powder (red), tetracene absorbed
on (blue) and encapsulated in (green) zeolite (excitation measured at 580 nm, emission excited
at 445 nm).

110

The loading of anthracene in zeolite was also confirmed by TGA (see Table 3-17)
Table 3-17: Mass Loss of Anthracene in Zeolite-L
Sample Mass Loss (%) Mass Loss (mg)
ZA25 1.3 1.29
ZA50 2.5 2.57
ZA75 3.7 3.86
ZA1 00 4.9 5.14

111

3.5.3 Pore Stoppering
Electrostatic binding of the cationic stopper molecules synthesized was carried out
through solution phase exchange.
17
Zeolite-L was sonicated in 10 mL toluene. The
stopcock molecule was added to DCM:MeOH in a 1:1 proportion relative to the number
of channel openings, ne (moles) (eq. (3-5):
(3-5) 𝑛 𝑒 = 5.21𝑥 10
−7
(
𝑋 𝑧 𝑙 𝑧 )
wherein Xz is the mass of zeolite in mg, and lz is the average length of zeolite in nm. The
two solutions were added together, sonicated and then stirred overnight. The resulting
stopcock-modified zeolite-L was centrifuged off, washed with DCM:MeOH, then dried.
Washing is done to remove any excess dye.
BODIPY-dipyridinium was fully loaded by calculating the number of pore channel
openings and adding that amount of BODIPY-dipyridinium to the solution of zeolite-L
loaded with tetracene (Z-T). The final solution had no BODIPY in it, and when the final
zeolite-L loaded with tetracene and capped with BODIPY -dipyridinium (Z-T-BD) was
washed with DCM, no BODIPY- dipyridinium dissolved in solution. Thus, it was
confirmed that BODIPY- dipyridinium was exchanged with the zeolite-L pore openings
and not adsorbed on the surface.
3.5.4 Synthesis of BODIPY-dipyridinium
The synthesis of BODIPY-dipyridinium was carried out by Jessica Golden, from Prof.
Mark Thompson’s lab, according to the procedure shown in Figure 3-35, and described
below.
(1) 2,8-diethyl-1,3,7,9-tetramethyl-5-(4-cyanophenyl)dipyrromethane:
3-ethyl-2,4-dimethyl-pyrrole (3.11 g, 38.3 mmol) and 4-formylbenzonitrile (2.25 g, 17.2
mmol) were combined in CH2Cl2 (100 mL) and the solution was degassed by sparging
112

with N2 for 15 min. Trifluoroacetic acid (190 µL) was added in one portion, causing a color
change to orange then brown, and the reaction was shielded from light and stirred for 3
h. Sodium hydroxide (0.2 M aq, 100 mL) was added to quench the reaction and stirring
was continued for 30 min, causing a color change in the organic layer to bright yellow.
The organics were removed and the remaining aqueous layer extracted with ethyl acetate
(2 × 40 mL). The combined organics were washed with water (2 × 50 mL) and brine (2 ×
50 mL), then dried (MgSO4) and pumped down to an orange oil (6.25 g) that solidified
upon standing overnight. This material was further purified by eluting through a short
SiO2 gel column with DCM, affording the desired product as a yellow solid.
(2) 2,8-diethyl-1,3,7,9-tetramethyl-5-(4-formylphenyl)dipyrromethane:
2,8-diethyl-1,3,7,9-tetramethyl-5-(4-cyanophenyl) dipyrromethane (1.133 g,
4.114 mmol) was added to an oven-dried 200-mL, 1-neck roundbottom flask equipped
with a stir bar. An addition funnel with septum was placed on top and the entire system
placed under N2 by three pump–backfill cycles. CH2 Cl2 (60 mL Dri-Solv) was added by
syringe, then diisobutylaluminum hydride (1.0 M solution in hexanes, 13.0 mL, 13 mmol)
was added to the funnel by syringe and subsequently added dropwise to the stirring
solution over a period of 30 min. The reaction was allowed to continue overnight (16 h).
A saturated aqueous NH4Cl solution (50 mL) was added to the mixture and, after stirring
vigorously for 3 h, the resulting gooey emulsion was filtered and the organic layer
removed.
The organics were washed with NaOH (10% aq, 1 × 50 mL) and, after separating the
organic and aqueous layers, the aqueous layers were further extracted
with CH2Cl2 (25 mL). The combined organic layers were dried (MgSO4) and
concentrated to a red-orange film. A short (∼4”) column (SiO2 gel, DCM eluent) was run
113

and monitored by TLC until all of the product had eluted (Rf
0.5). The combined fractions were concentrated by rotary evaporation to afford the title
product as a peach solid.
(3) 4,4-difluoro-2,6-diethyl-1,3,5,7-tetramethyl-8-(4-formylphenyl)-4-bora-3a,4a-
diaza-s-indacene (BODIPY-COH):
2,8-diethyl-1,3,7,9-tetramethyl-5-(4-formylphenyl) dipyrromethane (1.178 g,
4.25 mmol) was placed in an oven-dried 3-neck roundbottom flask, and the atmosphere
in the flask was exchanged for N2 by 3 pump–backfill cycles. Dry CH2Cl2 (50 mL) was
added and the dipyrromethane dissolved with stirring to give a yellow solution. 2,3-
dichloro-5,6-dicyano-1,4-benzoquinone (DDQ) (1.06 g, 4.67 mmol, 1.1 equiv) was added
under an N2 purge, causing a color change to dark red as solids crash out. After 1.5 h,
diisopropylethylamine (DIPEA) (3.0 mL) was added, and the solution stirred for 15 min,
then boron trifluoride diethyl-etherate (BF3-OEt2) (3.6 mL) was added slowly and the
reaction allowed to proceed for 1 h. After the reaction was complete, NaHCO3 (5% aq, 200
mL) was added and the mixture stirred vigorously for 18 h to kill excess BF 3 and
decompose any amine–borane adducts. The resulting mixture was filtered and the
organic layer removed and washed with Na2SO3 (10% aq, 2 × 100 mL), HCl (5% aq, 2 ×
50 mL), and brine (2 × 100 mL). The combined organics were dried (Na2SO4), filtered
and concentrated to a dark film. The product was purified by 2 rounds of chromatography
on SiO2 gel (DCM first, then CHCl3), and concentrated to greenish crystals.
(4) 4,4-difluoro-2,6-diethyl-1,3,5,7-tetramethyl-8-(4-(1,1-diyl) bis(1-methylpyridin-
1-ium)phenyl)-4-bora-3a,4a-diaza-s-indacene (BODIPY-dipyridinium):
Following the procedure outlined by Phillips,
45
BODIPY-COH (0.12 g, 0.29 mmol) and 1
4-dimethylpyridinium iodide (0.35 g, 1.47 mmol) were dissolved in 15 mL hot methanol.
114

Piperidine (1 mL) was added and the solution refluxed for 3 hours. Methanol was removed
in vacuo and the orange powered residue was recrystallized from hot methanol.
115

Figure 3-35: BODIPY-dipyridinium synthesis

116

3.5.4.1 Synthesis of Naphthalene-Pyridinium
Naphthalene-pyridinium was synthesized by the procedure outlined in Figure 3-36 (a),
and the H-NMR is given in in Figure 3-36 (b).

(a) (b)
Figure 3-36: (a) Naphthalene-pyridinium synthesis and (b) H-NMR of naphthalene-pyridinium

(1) 4-(naphthalene-1-yl)benzaldehyde:
A Suzuki-Miyaura cross coupling reaction was used to synthesize the starting
naphthalene-aldehyde.
26
1-bromonaphthalene (207 mg, 1 mmol), 4-formylphenylboronic
acid (180 mg, 1.2 mmol) and 2mol% Pd(OAc)2 were combined in 12 mL DMF at r.t. under
air. A separate solution of K2 CO3 (194 mg, 1.4 mmol) in 12 mL DI water was prepared and
then added dropwise to the DMF solution. This was stirred and heated at 100°C overnight.
(2) The same condensation reaction from step 4 of making BODIPY -dipyridinium
was used to make the pyridinium analog. This only formed one pyridinium substituent.
3.5.4.2 Diether-pyridinum synthesis
Diether pyridinium was made following the procedure by Phillips (see Figure 3-37).
45

117

Figure 3-37: Diether-pyridinium synthesis

3.5.4.3 HER experiments
Usually about 5 mg of chromophore and water reduction catalyst were put into a
quartz glass vial with a stirring bar. After evacuation and flushing with Ar (3 times), 5ml
of degassed irradiation solvent (EtOH:H2O (1:1), 10% TEA, adjusted to pH 11 with HCl,
or MeCN:H2O (1:1), 10% TEA, adjusted to pH 11 with HCl in case of catalysts other than
[Ni(pyS)3]TEA) were added. The irradiation was conducted at ambient temperature and
the vials were placed 8 cm away from a water cooled, cylindrical, medium pressure, 150
Watts Hg-lamp from Heraeus. The temperature of the glass vials remained at ambient
temperature during the course of irradiation. The vials were equipped with a ground glass
valve and a silicon septum. For hydrogen measurement via micro-GC the valve was
opened and a needle, from the GC, was inserted through the septum into the headspace
of the vial. Via external calibration, the hydrogen content was quantified. The overall
error of the procedure is estimated to be below 10%. The experimental conditions for
studying intramolecular HER is given in Table 3-18.

118

Table 3-18: Experimental conditions for intramolecular HER
Catalyst Photosensitizer Solvent Sac. Donor pH Time (h)
CoS4
-
Z-A100-BD
1:1 mL H2O:
ACN
5% ascorbic acid 4.5 24
CoN4
-
Z-A100-BD
1:1 mL H2O:
ACN
5% TEA 11.5 24
[Pd(PPh3)Cl2]2
-
Z-A100-BD
1:1 mL H2O:
ACN
5% ascorbic acid 4.5 24
[Ni(pyS)3]
−
Z-A100-BD
1:1 mL H2O:
EtOH
5% TEA 11.5 24
[Ni(pyS)3]
−
Z-A100
1:1 mL H2O:
EtOH
5% TEA 11.5 24
[Ni(pyS)3]
−
BODIPY
1:1 mL H2O:
EtOH
5% TEA 11.5 24

119

3.6 Bibliography
1. Calzaferri, G. & Lutkouskaya, K. Mimicking the antenna system of green plants.
Photochem. Photobiol. Sci. 7, 879–910 (2008).
2. Kim, Y. Il & Mallouk, T. E. Dynamic electron-transfer quenching of the tris(2,2’-
bipyridyl)ruthenium(II) MLCT excited state by intrazeolitic methylviologen ions.
J. Phys. Chem. 96, 2879–2885 (1992).
3. Yonemoto, E. H. et al. Photoinduced Electron Transfer Reactions in Zeolite-Based
Donor- Acceptor and Donor-Donor- Acceptor Diads and Triads. J. Am. Chem.
Soc. 10557–10563 (1994).
4. Meier, W. & Baerlocher, C. Zeolite type frameworks: connectivities, configurations
and conformations. in Molecular Sieves: Structures and Structure Determination
2, 141–160 (1999).
5. Ruiz, A. Z., Brühwiler, D., Ban, T. & Calzaferri, G. Synthesis of Zeolite L. Tuning
Size and Morphology. Monatshefte für Chemie 136, 77–89 (2005).
6. Megelski, S. & Calzaferri, G. Tuning the Size and Shape of Zeolite L-Based
Inorganic–Organic Host–Guest Composites for Optical Antenna Systems. Adv.
Funct. Mater. 11, 277–286 (2001).
7. Gigli, L. et al. Close-Packed Dye Molecules in Zeolite Channels Self-Assemble into
Supramolecular Nanoladders. J. Phys. Chem. C 118, 15732–15743 (2014).
8. Fois, E., Tabacchi, G. & Calzaferri, G. Orientation and Order of Xanthene Dyes in
the One-Dimensional Channels of Zeolite L: Bridging the Gap between
Experimental Data and Molecular Behavior. J. Phys. Chem. C 116, 16784–16799
(2012).
9. Devaux, A. et al. Self-Absorption and Luminescence Quantum Yields of Dye-
120

Zeolite L Composites. J. Phys. Chem. C 117, 23034–23047 (2013).
10. Gigli, L. et al. Thionine Dye Confined in Zeolite L: Synthesis Location and Optical
Properties. J. Phys. Chem. C 119, 16156–16165 (2015).
11. Hashimoto, S., Ikuta, S., Asahi, T. & Masuhara, H. Fluorescence Spectroscopic
Studies of Anthracene Adsorbed into Zeolites: From the Detection of Cation−π
Interaction to the Observation of Dimers and Crystals. Langmuir 14, 4284–4291
(1998).
12. Brühwiler, D., Gfeller, N. & Calzaferri, G. Resorufin in the Channels of Zeolite L. J.
Phys. Chem. B 102, 2923–2929 (1998).
13. Devaux, A. et al. Efficient and Robust Host–Guest Antenna Composite for Light
Harvesting. Chem. Mater. 26, 6878–6885 (2014).
14. Schulz-Ekloff, G., Wöhrle, D., van Duffel, B. & Schoonheydt, R. A. Chromophores
in porous silicas and minerals: preparation and optical properties. Microporous
Mesoporous Mater. 51, 91–138 (2002).
15. Hashimoto, S., Uwada, T., Masuhara, H. & Asahi, T. Fabrication of Gold
Nanoparticle-Doped Zeolite L Crystals and Characterization by Optical
Microscopy: Laser Ablation- and Crystallization Inclusion-Based Approach. J.
Phys. Chem. C 112, 15089–15093 (2008).
16. Brühwiler, D. & Calzaferri, G. Molecular sieves as host materials for
supramolecular organization. Microporous Mesoporous Mater. 72, 1–23 (2004).
17. López-Duarte, I. et al. On the Significance of the Anchoring Group in the Design
of Antenna Materials Based on Phthalocyanine Stopcocks and Zeolite L. Chem. –
A Eur. J. 17, 1855–1862 (2011).
18. Dieu, L.-Q. et al. Novel phthalocyanine-based stopcock for zeolite L. Chem.
121

Commun. 1187–1189 (2008). doi:10.1039/B714995F
19. Alberto, C. Supramolecular Photochemistry. Controlling Photochemical
Processes. Edited by V. Ramamurthy and Yoshihisa Inoue. Angew. Chemie Int.
Ed. 51, 6311–6312
20. Cucinotta, F. et al. Energy Transfer at the Zeolite L Boundaries: Towards Photo-
and Electroresponsive Materials. Chempluschem 79, 45–57 (2014).
21. Calzaferri, G. et al. Photonic antenna system for light harvesting, transport and
trapping. J. Mater. Chem. 12, 1–13 (2002).
22. Xu, C., Mochizuki, D., Maitani, M. M. & Wada, Y. Electron transfer from excited
[Ir(2-phenylpyridyl)3] through a coexisting electron relay in zeolite. Eur. J. Inorg.
Chem. (2014). doi:10.1002/ejic.201301431
23. Lim, S.-H., Bjorklund, T. G., Spano, F. C. & Bardeen, C. J. Exciton Delocalization
and Superradiance in Tetracene Thin Films and Nanoaggregates. Phys. Rev. Lett.
92, 107402 (2004).
24. Burdett, J. J., Müller, A. M., Gosztola, D. & Bardeen, C. J. Excited state dynamics
in solid and monomeric tetracene: The roles of superradiance and exciton fission.
J. Chem. Phys. 133, 144506 (2010).
25. Peter, G. & Bässler, H. Fluorescence and energy transfer in disordered solid
tetracene. Chem. Phys. 49, 9–16 (1980).
26. Baehtz, C. & Fuess, H. Naphthalene, anthracene and 2,3-benzanthracene in zeolite
Y. Phys. Chem. Chem. Phys. 4, 4543–4548 (2002).
27. Kochi, J. K. & Yoon, K. B. Stepwise assembly of charge-transfer complexes within
zeolite supercages as visual probes for shape selectivity. J. Phys. Chem. 95, 3780–
3790 (1991).
122

28. Marquis, S., Moissette, A., Hureau, M., Vezin, H. & Brémard, C. Spontaneous
charge separation and recombination induced by tetracene incorporation in pores
of acidic HnZSM-5 zeolites. J. Phys. Chem. C 111, 17346–17356 (2007).
29. Liu, H. et al. Synthesis and photophysical properties of a ‘face-to-face’ stacked
tetracene dimer. Phys. Chem. Chem. Phys. 17, 6523–6531 (2015).
30. Hashimoto, S., Hagiri, M., Matsubara, N. & Tobita, S. Photophysical studies of
neutral aromatic species confined in zeolite L: Comparison with cationic dyes.
Phys. Chem. Chem. Phys. 3, 5043–5051 (2001).
31. Dura, L. et al. Design of BODIPY Dyes as Photosensitisers in Multicomponent
Catalyst Systems for Light-Driven Hydrogen Production. Chem. - A Eur. J. 21,
13549–13552 (2015).
32. Han, Z., McNamara, W. R., Eum, M.-S. S., Holland, P. L. & Eisenberg, R. A nickel
thiolate catalyst for the long-lived photocatalytic production of hydrogen in a
noble-metal-free system. Angew. Chem. Int. Ed. Engl. 51, 1667–70 (2012).
33. Zhang, P. et al. Photochemical hydrogen production with molecular devices
comprising a zinc porphyrin and a cobaloxime catalyst. Sci. China Chem. 55,
1274–1282 (2012).
34. McNamara, W. R. et al. Cobalt-dithiolene complexes for the photocatalytic and
electrocatalytic reduction of protons in aqueous solutions. Proc. Natl. Acad. Sci.
109, 15594–15599 (2012).
35. Mandal, S., Roy, D., Chaudhari, R. V & Sastry, M. Pt and Pd Nanoparticles
Immobilized on Amine-Functionalized Zeolite: Excellent Catalysts for
Hydrogenation and Heck Reactions. Chem. Mater. 16, 3714–3724 (2004).
36. Turro, N., Ramamurthy, V. & Scaiano, J. C. Modern Molecular Photochemistry of
123

Organic Molecules. (University Science Books, 2010).
37. Cao, X., Tolbert, R. W., McHale, J. L. & Edwards, W. D. Theoretical Study of
Solvent Effects on the Intramolecular Charge Transfer of a Hemicyanine Dye. J.
Phys. Chem. A 102, 2739–2748 (1998).
38. Márquez, F., García, H., Palomares, E., Fernández, L. & Corma, A. Spectroscopic
evidence in support of the molecular orbital confinement concept: Case of
anthracene incorporated in zeolites [17]. J. Am. Chem. Soc. 122, 6520–6521
(2000).
39. Marquis, S., Moissette, A. & Brémard, C. Incorporation of Anthracene into
Zeolites: Confinement Effect on the Recombination Rate of Photoinduced Radical
Cation-Electron Pair. ChemPhysChem 7, 1525–1534 (2006).
40. Kuchi, V., Oliver, A. M., Paddon-Row, M. N. & Howe, R. F. Long-lived charge
separation in a donor-acceptor dyad adsorbed in mesoporous MCM-41. Chem.
Commun. 1, 1149–1150 (1999).
41. Kwon, O., Yu, H. & Jang, D. Formation Mechanism of Anthracene Dimers and
Excimers in NaY Zeolitic Nanocavities. 3970–3974 (2004).
doi:10.1021/jp0309163
42. García, H. et al. Generation and reactions of organic radical cations in zeolites.
Chem. Rev. 102, 3947–4007 (2002).
43. Chen, L. et al. Dual Homogeneous and Heterogeneous Pathways in Photo- and
Electrocatalytic Hydrogen Evolution with Nickel(II) Catalysts Bearing
Tetradentate Macrocyclic Ligands. ACS Catal. 5, 356–364 (2015).
44. Yamada, Y., Miyahigashi, T., Kotani, H., Ohkubo, K. & Fukuzumi, S.
Environmental Science Photocatalytic hydrogen evolution with Ni nanoparticles
124

by using. Energy Environ. Sci. 5, 6111–6118 (2012).
45. PHILLIPS, A. P. CONDENSATION OF AROMATIC ALDEHYDES WITH γ-
PICOLINE METHIODIDE. J. Org. Chem. 14, 302–305 (1949).

125

4 Chirality-Driven Assembly of
Push-Pull Molecules

126

Table of Contents

4.1 Introduction ..................................................................................................... 128
4.1.1 Molecular Strategies for Solar Water Splitting ....................................................................... 128
4.1.2 Selectivity of Bisoxazoline (BOX) ............................................................................................ 129
4.1.3 Project Description ................................................................................................................. 133
4.2 Results ............................................................................................................. 134
4.2.1 mPM D/A ................................................................................................................................ 134
4.2.2 Ph-BOX Zinc Complexes .......................................................................................................... 135
4.2.3 Homochiral, Homoleptic Complexes ...................................................................................... 137
4.2.3.1 DPAR,2-{Zn} .............................................................................................................................. 142
4.2.3.2 NaphS,2-{Zn} ............................................................................................................................. 144
4.2.3.3 AnR,2-{Zn} ................................................................................................................................. 145
4.2.4 AnR,2-{Zn}: LE and CT states ..................................................................................................... 145
4.2.5 Intramolecular Energy Transfer .............................................................................................. 148
4.2.6 Intramolecular Electron Transfer............................................................................................ 154
4.2.6.1 PET in MDPAS-{Zn}-AnR ........................................................................................................... 156
4.2.6.2 PET in MDPAS-{Zn}-DPAR ......................................................................................................... 158
4.3 Conclusions ...................................................................................................... 159
4.3.1 Future Work ............................................................................................................................ 161
4.3.1.1 Intramolecular FRET ............................................................................................................... 161
4.3.1.2 Intramolecular PET ................................................................................................................. 161
4.3.1.1 CT Exciton Coupling ................................................................................................................ 162
4.3.1.2 Photolysis ................................................................................................................................ 162
4.4 Experimental Procedures .................................................................................. 163
4.4.1 General Methods .................................................................................................................... 163
4.4.2 Synthesis of diethyl malonate derivatives .............................................................................. 163
4.4.2.1 Synthesis of diethyl 2-(4-bromophenyl)malonate .................................................................. 163
4.4.2.2 Synthesis of diethyl 2-(4-(phenylethynyl)phenyl)malonate ................................................... 164
4.4.2.3 Synthesis of diethyl 2-(4-methyl(phenyl)amino)phenyl)malonate ......................................... 165
4.4.2.4 Synthesis of diethyl 2-(9,10-diphenylanthracene) malonate (An-DM): .................................. 166
4.4.2.5 Synthesis of 2-(4-naphthalen-1-yl) diethyl malonate (NaPh-DM) .......................................... 167
4.4.3 General Synthesis of bis(oxazoline) (BOX) .............................................................................. 168
4.4.3.1 Synthesis of
Ph
BOX-Ph2 ............................................................................................................ 169
4.4.3.2 Synthesis of
DPA
BOX-Ph2 .......................................................................................................... 170
4.4.3.3 Synthesis of
MDPA
BOXSS-Ph2 ..................................................................................................... 170
4.4.3.4 Synthesis of
Anth
BOX-Ph2 ......................................................................................................... 170
4.4.3.5 Synthesis of
NaPh
BOXSS-Ph2 ...................................................................................................... 170
4.4.4 General preparation of homochiral zinc complexes, .............................................................. 171
(
R
BOX-Ph2)2Zn ................................................................................................................................................... 171
4.4.4.1 Preparation of (
Ph
BOX-Ph2)2Zn ................................................................................................ 171
4.4.4.2 Preparation of (
DPA
BOX-Ph2)2Zn .............................................................................................. 172
4.4.4.3 Preparation of (
MDPA
BOXSS-Ph2)2Zn .......................................................................................... 172
4.4.4.4 Preparation of (
An
BOX-Ph2)2Zn ................................................................................................ 173
4.4.4.5 Preparation of (
NaPh
BOXSS-Ph2)2Zn ........................................................................................... 173
4.4.4.6 General preparation of heterochiral Zn complex, (
R
BOXRR)Zn(
R1
BOXSS) ................................. 173
4.4.4.7 Preparation of (
Ph
BOXRR-Ph2)Zn(
Ph
BOXSS-Ph2) (R=R1=Ph) ...................................................... 174
4.4.4.8 Preparation of (
DPA
BOXRR-Ph2)Zn(
DPA
BOXSS-Ph2) (R=R1=DPA) .................................................. 174
4.4.4.9 Preparation of (
DPA
BOXRR-Ph2)Zn(
Ph
BOXSS-Ph2) (R=DPA, R1=Ph) .............................................. 174
127

4.4.4.10 Preparation of (
DPA
BOXRR-Ph2)Zn(
MDPA
BOXSS-Ph2) (R=DPA, R1=MDPA) .................................... 175
4.4.4.11 Preparation of (
An
BOXRR-Ph2)Zn(
An
BOXSS-Ph2) (R=R1=An) ........................................................ 175
4.4.4.12 Preparation of (
An
BOXRR-Ph2)Zn(
Ph
BOXSS-Ph2) (R=An, R1=Ph) .................................................. 175
4.4.4.13 Preparation of (
An
BOXRR-Ph2)Zn(
MDPA
BOXSS-Ph2) (R=An, R1=MDPA) ........................................ 175
4.4.4.14 Preparation of (
An
BOXRR-Ph2)Zn(
NaPh
BOXSS-Ph2) (R=An, R1=NaPh) ........................................... 176
4.5 Bibliography ..................................................................................................... 177

128

4.1 Introduction
4.1.1 Molecular Strategies for Solar Water Splitting
The abundance of solar energy makes it a premier alternative to fossil fuels as an
energy source. Developing artificial photosynthetic systems that can convert solar energy
into chemical energy is one ideal alternative energy.
1,2
Facile molecular designs for donor
(D)-acceptor (A) dyads is critical for light absorption and energy transfer in
photocatalytic systems. Assembly of large molecular arrays through covalent linkages is
expensive and synthetically demanding; thus, self-assembly of the building blocks for
higher order architectures is ideal.
2,3
Artificial photosynthtic systems mimic plant
photosynthesis by replicating the processes involved. There are three basic parts: light
collection, energy tunneling to the catalyst, then electron or hole transfer to the catalyst
for hydrogen or oxygen evolution (Figure 4-1).
1

Figure 4-1: General concept for artificial photosynthesis.

Forming dyads, and higher order assemblies often involves complicated synthetic
schemes, and multiple ligands (Figure 4-2). The building blocks of such systems need to
be covalently bound to facilitate and control distances for electron and charge transfer.
Self-assembly of these building blocks creates hierarchal organization of these molecules
129

for larger scale energy transfer at the supramolecular level.

Donor-acceptor molecules are used to study the underlying electronic principles for
photoinduced energy and electron transfer. One pitfall of these dyads is that they do not
necessarily stabilize the charge separated state (CSS) long enough to be kinetically
competent to carry out redox reactions with catalysts or other species. Thus, systems
developed today are donor-sensitizer-acceptors, such as the one developed by Gust et al.
composed of a carotenoid donor, a porphyrin sensitizer and a fullerene acceptor (Figure
4-2, a). The porphyrin has the lowest excited S1, which first reduces the fullerene, and
subsequentially oxidizes the carotenoid. The excited state lifetime between a covalently
bound porphyrin and fullerene would decay in 3.3 ns, whereas the CSS lifetime is
extended to 57 ns in the case with the carotenoid.
2,4

This study is the preliminary work to examine the feasibility of utilizing chirality to
drive selective assembly of metal-centered donor-acceptor molecules. Demonstrating
basic principles of energy and electron transfer in D-A dyads can lead to facile assembly
of D-S-A motifs.
4.1.2 Selectivity of Bisoxazoline (BOX)
This study is based on selective metal-centered assembly of ligands bearing either D-

(a) (b)
Figure 4-2: Artificial photosynthesis and charge separation in a (a) non- assembled and (b)
self-assembled molecular system.
1,2

130

or A- moieties. This is done through the coordination of chiral bis(oxazoline) (BOX) to
zinc metal centers. BOX has been extensively studied in coordination chemistry and
asymmetric catalysis; thus a strong synthetic foundation has been set.
5,6
The inherent
chirality of the 4-position group allows for selective binding. This makes it ideal in
asymmetric catalysis. BOX for luminescent applications has only recently been
investigated by select groups, as a wide band gap ligand that undergoes circularly
polarized luminescence.
7–9
BOX has a wide HOMO-LUMO gap, therefore enabling charge
transfer between the D-A chromophores, even across the d
10
electron configuration of
Zn(II). Trinh et al. reported on fast interligand charge transfer (CT) in bis-dipyrrin Zn
complexes, indicating effective electron coupling of ligands across zinc.
10

Previous reports have demonstrated chiral discrimination between enantiomers of
the BOX ligands upon formation of zinc complexes.
11,12
Crosignani et al. first introduced
the thermodynamically stable heterochiral Zn(II)-BOX complex. This was investigated
further by Takacs, proving that when Zn
2+
is treated with a mixture of the R,R and S,S
versions of the BOX-Ph2 ligand, only the heterochiral complex, (BOXRR-Ph2)Zn(BOXSS-
Ph2), is formed (Figure 4-3, a).
11–13
This can be easily monitored by
1
H-NMR (Figure 4-3,
b).
131

Crystal structures confirm the homochiral complex is slightly distorted from
tetrahedral coordination to minimize the steric interaction between the meso-position
phenyl groups (Figure 4-3 (c)).
12
The heterochiral complex has near perfect tetrahedron
geometry, making it a more stable complex. Thus, when the RR and SS homochiral
complexes are mixed at an equimolar concentration, the resulting complex is solely the
heterochiral complex (Figure 4-4, a).

(a) (b)

(c)
Figure 4-3: (a) Synthesis of (H-BOX
Ph2
RR) 2Zn and (H-BOX
Ph2
SS)Zn(H-BOX
Ph2
RR),
(b) Alkyl region of H-NMR of the homochiral complex, heterochiral complex and free
BOX and (c) crystal structure of (H-BOX SS) 2Zn (left) and (H-BOX SS)Zn(H-BOX RR)
(right). Reprinted from Takacs, 2005.
12

v
132

The ease of heterochiral formation was confirmed through a simple
1
H-NMR study,
using PhR,2-{Zn} and PhS,2-{Zn} (see Figure 4-4, b). An NMR sample was prepared in d-
chloroform containing 1.2 M PhS,2-{Zn}. A similar sample was prepared with PhR,2-{Zn},
which was titrated into to the PhS,2-{Zn} solution, instantly forming the heterochiral
complex.
1
H-NMR was monitored at various titrations until the solution was equimolar.
If the moiety appended at the meso- position (denoted herein as the meso-position
moiety, or mPM) on the BOXRR ligand is a D and an A is appended on the same position
in the BOXSS enantiomer, a mixture of the two ligands with Zn
2+
will give solely (D-
BOXRR)Zn(BOXSS-A) (Figure 4-5, a). Atkins et. al. report on a series of (D-
BOXRR)Zn(BOXSS-A) complexes they synthesized. They postulated a resonant quinoidal
form, wherein Zn would act as a π-bridge to facilitate electron transfer, which would be a
CT state (Figure 4-5, b).
11
The D- and A- mPM’s chosen did not have emissive properties,
and they did not report electron transfer within the complexes (Figure 4-5, a).
11
The study
herein examines various fluorescent (D-BOXRR)Zn(BOXSS-A) complexes to determine
the mechanism of energy and electron transfer as well as BOX’s role in these processes.

(a) (b)

Figure 4-4: (a) Exchange reaction between (Ph-BOX SS) 2Zn and (Ph-BOX RR) 2Zn to form
(Ph-BOX SS)Zn(Ph-BOX RR) and (b) Monitoring the exchange reaction through
1
H-NMR

133

(a) (b)

Figure 4-5: (a) (top) D- BOX RR and A- BOX SS, (bottom) heterochiral (Et 2N- BOX RR)Zn(BOX SS-
CN) made by Atkins et al. in which Et 2N is the donor and CN is the acceptor (b)
Pseudoenantiomeric donor/acceptor-substituted aryl-substituted box ligands forming single
enantiomer, chiral donor-acceptor zinc complexes
11

4.1.3 Project Description
This study is the seminal work to develop chirality- assembled, light harvesting dyads,
taking advantage of the ease of synthesis to study multiple photoinduced energy
pathways. This was done through various mPM D-A on BOX, ligated to Zn
2+
to form
molecular dyads. The steady-state photophysical characterization of these complexes has
been investigated herein.

134

4.2 Results
BOX ligands were appended with mPM: phenyl (Ph), naphthalene (Naph),
diphenylanthracene (An), diphenylacetylene (DPA), and methyldiphenylamine (MDPA).
Various zinc complexes were prepared with these BOX ligands (Figure 4-6).

Figure 4-6: Chemical structures of BOX ligands and complexes in this study

4.2.1 mPM D/A
Ph, Naph and An were chosen as increasing π- conjugation bathochromically shifts
the absorption and emission. These chromophores have been studied in energy transfer
applications due to the extended π- conjugation, which allows significant orbital overlap.
Excitation of naphthalene has demonstrated singlet energy transfer to anthracene.
14

Synthetically, they are readily appended to BOX through Suzuki coupling.
DPA is known to facilitate charge transfer dynamics by creating a charge separated
state (CSS) when appended with a D or A.
15–17
The CSS of DPA is expected to facilitate
electron transfer to a strong donor, such as MDPA. Amines are excellent donors with a
135

HOMO lower than BOX, to ensure there is electron transfer directly from MDPA to DPA.
Energy transfer dynamics between D/A mPM are the main focus of this study, but
prior to that, the properties of BOX complexes without a D/A were investigated.
4.2.2 Ph-BOX Zinc Complexes
Simple ligands containing phenyl
groups at the 4- and meso- positions were
initially synthesized and studied (Figure
4-7).
These simple complexes, Ph-
BOXRR, PhR,2-{Zn}, and PhR-{Zn}-PhS
provide reference for the dynamics
between the ligand, the homochiral and the
heterochiral, homoleptic complexes.
DFT calculations (geometry
optimized using the method B3LYP, and a
basis LACVP with ECP fit-LACVP),
determined the approximate distribution
of electron density of PhR, PhR,2-{Zn} and PhR-{Zn}-PhS, in the highest occupied
molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO), in a
vacuum (Figure 4-7 and Table 4-1). The zinc complex stabilizes the HOMO and LUMO of
BOX.
Table 4-1:DFT calculated HOMO and LUMO energies of Ph R,2-{Zn} and Ph R-{Zn}-Ph S

HOMO-2 HOMO-1 HOMO LUMO LUMO+1 LUMO+2
Ph R,2-{Zn} -6.2 -4.93 -4.9 -0.49 -0.41 -0.33
Ph R-{Zn}-Ph S -6.2 -4.79 -4.79 -0.49 -0.35 -0.27

Figure 4-7: (top) Ph- BOX RR (left) and Ph-
BOX SS (right) synthesized: both tautomers
(diimine and iminoenamine, respectively)
are shown. (bottom) DFT calculated HOMO
and LUMO energies of (from left to right):
Ph-BOX SS, Ph R,2-{Zn} and Ph R-{Zn}-Ph S.
136

The electron density of the HOMO in PhR,2-{Zn} is delocalized over the complex,
whereas the HOMO and HOMO-1 are nearly degenerate states, localized on one ligand in
PhR-{Zn}-PhS. The LUMO is delocalized over the BOX core and the -phenyl’s in both
complexes. The heterochiral complexes are near perfect tetrahedral coordination, while
the homochiral complexes deviate slightly due to steric hindrance between the phenyls
located on the oxazoline rings, as confirmed by crystal structures measured by Takacs
(Figure 4-3, c). This results in a slight destabilization of the HOMO in the homochiral,
compared to the heterochiral.
Intramolecular coupling is observed in the absorption spectra of Ph-BOXRR, PhR,2-
{Zn} and PhR-{Zn}-PhS in nonpolar cyclohexane (CH) and polar methylene chloride
(DCM) (Figure 4-8, a and b, respectively). The ligand absorbs at higher energy than the
zinc complexes, which in cyclohexane (CH), they have two peaks, one resembling the
ligand at high energies and another at lower energies. The band at low energy is identified
as absorption from the complex itself. In polar media, the complex is stabilized, as the
higher energy transition is destabilized, resulting in solely absorption from the complex
(Figure 4-8, b). The zinc complexes are bathochromically shifted from the free ligand, as
the bis-complex has electronic influence from zinc, reduced movement of the free ligand,
and coupling is enabled between the two ligands. Comparing differences from the
homochiral to the heterochiral, it is noticed that the absorbance spectra of the
heterochiral complex, PhR-{Zn}-PhS, is red shifted from the homochiral, PhR,2-{Zn} (by
8nm in CH and 5 nm in DCM). This provides evidence of stronger coupling in the
heterochiral conformation.
18
This is ideal for efficient interligand electron transfer in a
heterochiral, heteroleptic complex.
137

250 275 300 325 350
0.2
0.4
0.6
0.8
1.0
Normalized Absorbance (a.u.)
Wavelength (nm)
(a)
250 275 300 325 350
0.2
0.4
0.6
0.8
1.0
Normalized Absorbance (a.u.)
Wavelength (nm)
(b)

Figure 4-8: Absorbance spectra of free ligand Ph-BOX RR (grey), homoleptic Ph R,2-{Zn} (black),
and heteroleptic Ph R-{Zn}-Ph S (red) in (a) CH and (b) DCM.

The Ph-{Zn} complexes provide evidence for coupling in the ground state; to further
probe this mechanism, emissive species were synthesized. The homochiral, homoleptic
derivatives with various D/A mPM were investigated to observe the intraligand interaction
between BOX and the mPM. To fully understand the processes the complexes undergo without
energy transfer, homochiral, homoleptic complexes were first investigated.
4.2.3 Homochiral, Homoleptic Complexes
Naphthalene, anthracene, phenylacetylene and methylphenylamine were
appended to phenyl-BOX for investigation in photoinduced energy and electron transfer.
DFT calculations of homochiral, homoleptic complexes, NaphS,2-{Zn}, AnR,2-{Zn} and
DPAR,2-{Zn}, helps elucidate the electronic distribution of the HOMO and LUMO in the
ground and excited state, respectively. These calculations were done using the same
geometry optimized conditions described in section 4.2.2. DFT calculates the HOMO and
HOMO-1 for these complexes localized on BOX and the mPM, and very close in energy,
indicating no particular preferance for either ligand (Table 4-2). The LUMO of NaphS,2-
{Zn}, AnR,2-{Zn} and DPAR,2-{Zn} is localized on the mPM (Figure 4-9). Emission from
the excited state of these complexes is expected to resemble that of the mPM D/A.
138

LUMO

HOMO

DPA R,2-{Zn} Naph S,2-{Zn} An R,2-{Zn} MDPA S,2-{Zn}
Figure 4-9: HOMO and LUMO electron density on homoleptic BOX 2Zn complexes (left to right)
Naph S,2-{Zn}, An R,2-{Zn} and DPA R,2-{Zn}.

Table 4-2: DFT calculated HOMO and LUMO energies (eV) for BOX 2Zn complexes

HOMO-2 HOMO-1 HOMO LUMO LUMO+1 LUMO+2
Naph S,2-{Zn} -4.93 -3.92 -3.51 -1.58 -1.33 -1.23
An R,2-{Zn} -5.03 -4.90 -4.87 -1.44 -1.41 -0.57
MDPA S,2-{Zn} -5.20 -4.41 -4.35 -0.41 -0.33 -0.27
DPA R,2-{Zn} -5.69 -4.90 -4.87 -0.95 -0.93 -0.63

139

The molar absorptivity of the
homochiral, homoleptic complexes, PhR,2-
Zn, NaphS,2-Zn, AnR,2-Zn and DPAS,2-Zn in
DCM were compared (Figure 4-10). The
transition at 315 nm found in PhR,2-{Zn} is
observed in all the complexes, though with a
long-band edge. This is attributed to ground
state coupling between the mPM and BOX.
AnR,2-{Zn} also has transition states from anthracene, from 350-425 nm.
Diphenylacetlyene absorbs at 275 nm, so the new band at 365 nm in DPAR,2-{Zn}, is due
to a charge transfer (CT) state. This is typical for DPA D/A molecules and has been
extensively studied with various D/A, such as NEt2 and CN (Figure 4-13).
16,17,19

Steady-state photophysics were measured in various dry solvents
(methylcyclohexane (CH), toluene (TOL), tetrahydrofuran (THF) and acetonitrile (ACN))
(Figure 4-11 and Table 4-3). The radiative rate (kr) and nonradiative rate (knr) were
calculated from equations 4-1, 4-2, respectively:
20

4-1 𝑘 𝑟 =
𝜙 𝜏
4-2 𝑘 𝑛𝑟
=
1
𝜏 − 𝑘 𝑟

where Φ is the quantum yield, and τ is the lifetime of the chromophore.
The absorption spectra of NaphS,2-{Zn}, AnR,2-{Zn} and MDPAS,2-{Zn} did not
change significantly with solvent. That of DPAR,2-{Zn} bathochromically shifted with
increasing solvent polarity, which has been observed when appending diphenylacetylene
with an electron donor.
21

250 275 300 325 350 375 400 425
0.0
2.0E4
4.0E4
6.0E4
8.0E4
1.0E5
1.2E5
1.4E5
Molar Absorptivity (M
-1
cm
-1
)
Wavelength (nm)

Figure 4-10: Molar absorptivity of Ph R,2-
{Zn} (black), DPA R,2-{Zn} (green),
Naph S,2-Zn (blue), An R,2-{Zn} (red), and
MDPA S,2-{Zn} (purple) in DCM

140

0.2
0.4
0.6
0.8
1.0
250 275 300 325 350 375 400 425
CH
TOL
THF
ACN
Normalized Intensity (a.u.)
(a)
400 450 500 550 600 650 700
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
(b)
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
(e)
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
(d)

0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
(c)
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
(d)

250 275 300 325 350 375 400 425
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
(g)
400 450 500 550 600 650 700
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
(h)

Figure 4-11: Left- absorbance and right- emission of homochiral, homoleptic complexes, (a,b)
DPA R,2-{Zn} (blue), (c,d) Naph S,2-{Zn} (green), (e,f) An R,2-{Zn} (red), and (g,h) MDPA S,2-{Zn}
in CH (solid), TOL (dashed), THF (dash-dot) and ACN (dot)

141

Table 4-3: Emission maximum, quantum yield (Φ), lifetime (τ) and radiative rate (k r) of
homochiral, homoleptic zinc complexes in various solvents
Solvent
λem
(nm)
ϕ τ (ns)
k r
(x10
7
s
-1
)
k nr
(x10
7
s
-1
)
DPA R,2-{Zn}
CH 394 0.02
1.9 (71)
- -
3.9 (29)
TOL 400 0.19 2.3 8.4 35.1
THF 422 0.44 1.9 23.3 29.4
ACN 450 0.71 3.3 21.4 8.9
Naph S,2-{Zn}
CH 400 0.03 n.d. - -
TOL 418 0.13 0.3 40.4 272
THF 438 0.56 2.3 24.6 19.3
ACN 480 0.64 5.8 11.1 6.6
An R,2-{Zn}

CH 442 0.88 2.6 33.7 4.8
TOL 466 0.75 4.7 16.0 5.3
THF 524 0.48 14.1 3.4 3.7
ACN
410
0.05 3.9 1.3 24.1
610
MDPA S,2-{Zn}
CH 394 0.03
1.8 (62)
- -
6.0 (38)
TOL 398 0.11
1.0 (93)
- -
5.0 (7)
THF 388 0.09
2.5 (92)
- -
9.2 (8)
ACN 378 0.11
3.4 (48)
- -
12.3 (52)

The homochiral, homoleptic complexes demonstrate a bathochromic shift in their
emission with increasing solvent polarity, a typical behavior of CT states (see Figure
4-11).
16,20,22
The DFT calculations had LUMO density directly on the mPM D/A with
minimal influence from BOX. The expected emission from DFT calculations contradicts
the results; however, DFT assumes the molecules in a vacuum, and did not estimate the
electron density in the presence of solvent. The presence of the solvent induces the CT
state.
142

This solvatochromic behavior in emission is an inherent property of the BOX2Zn
molecules, as all subsequent complexes studied demonstrate it. The mechanism of
formation of the CT states are dramatically different between complexes, as evidenced by
the trends in photophysical data. This is clarified by plotting the radiative and
nonradiative rates versus solvent dielectric constant (Figure 4-12). The nature of the CT
state is due to intraligand charge transfer between BOX (D) and the mPM (A). The CT
state is localized on a single ligand. The A in each of these complexes dictates the
photophysical behavior, which is why they all behave differently. MDPAS,2-{Zn} is the
only complex in which CT is not observed, and it is also the only one with a mPM D. DFT
calculations of MDPAS,2-{Zn} show it is the only complex in which the electron density of
the LUMO is located on BOX and not MDPA; thus it is possible that a CT state is not
formed between the two.
0 5 10 15 20 25 30 35 40
0
5
10
15
20
25
30
35
40
45

Radiative Rate (10
7
s
-1
)
Solvent Dielectric Constant
0 5 10 15 20 25 30 35 40
0
5
10
15
20
25
30
35
40
270
275

Nonradiative Rate (10
7
s
-1
)
Solvent Dielectric Constant

Figure 4-12: (a) Radiative rate and (b) nonradiative rate as a function of solvent dielectric
constant of Naph S,2-{Zn} (blue), An R,2-{Zn} (red) and DPA R,2-{Zn} (green)

4.2.3.1 DPAR,2-{Zn}
The radiative rate of DPAR,2-{Zn} increases with polarity, while the nonradiative
rate decreases (Table 4-3 and Figure 4-12, a). In MeCH, two lifetimes are observed,
indicating the presence of both the LE (short lifetime) and CT state (long lifetime).
22
In
143

polar solvents, the LE emission is suppressed,
as internal conversion to the CT state prevails.
The mechanism by which DPA undergoes
charge separation has been well studied.
15,23,24

As DPA is excited it undergoes twisted
intramolecular charge transfer (TICT) and
from the ππ
*
it bends into an intermediate
state, πσ
*
, before charge separating completely (Figure 4-13).
15
This has been
demonstrated experimentally with D/A appended to DPA, as well as theoretically
explained for DPA.
15,25,26
It has been confirmed that the addition of an electron donating
group to DPA enhances the state switch from the ππ* to the πσ* state, which is also
enhanced in polar solvents. As BOX is an electron donor, the charge transfer state is being
stabilized. The increasing quantum yield with increasing polarity is attributed to
stabilization of the CT state. An intermediate, bent πσ
C ≡ C
∗
state is formed before charge
separation.
15,19,25

Szyskowska et. al. synthesized a similar D-A
molecule with DPA, N,N-
dimethyl(phenylethynyl)aniline (denoted
DMAPhacPh) (Figure 4-14).
21
They measured a decrease in quantum yield from CH
(0.35) to THF (0.24) to ACN (0.10), with similar lifetimes (though biexponential in THF
and ACN). This is quite different from DPAR,2-{Zn}, which had increasing quantum yield
with polarity. This indicates that there is another underlying mechanism that varies from
a traditional D-A molecule in DPAR,2-{Zn}. This could be due to the fact that DPAR,2-{Zn}
is a dyad, with two D-A units bound together by Zn.

Figure 4-13: Charge separation
dynamics in a D-DPA-A molecule.
15

Figure 4-14: Structure of
DMAPhacPh
144

4.2.3.2 NaphS,2-{Zn}
The radiative and nonradiative rate of NaphS,2-{Zn}
decrease with increasing polarity (Table 4-3). This complex
was compared with a simple D-A molecule, 1-(4-
dimethylaminophenyl)naphthalene (denoted NA1), studied
by Herbich et al (Figure 4-15).
27
In polar solvents, they
observed a similar trend in radiative and nonradiative rates
for NA1, with both decreasing. However, whereas the
radiative decay in NaphS,2-{Zn} was effectively quenched in CH, with a quantum yield of
0.03 (Table 4-3), they observed a quantum yield of 0.57 for NA1. They attribute the
photophysical behavior of the CT state in NA1 to a flattening of the excited state in
nonpolar solvents, and no significant conformational change in polar solvents. This is
possibly similar for NaphS,2-{Zn}, but there is another mechanism in play as the behavior
of NaphS,2-{Zn} in nonpolar solvents (CH) is dramatically different from NA1.
Nonpolar medium prevents the formation of a quinoid structure, and so favors the
benzenoid structure (Figure 4-16). This results in a high nonradiative rate, as fast
interconversion in the conjugated
system creates torsional vibrations.
This could account for the quenching of
radiative decay of NaphS,2-{Zn} in CH.
As solvent polarity increases, the
electron density redistributes to favor
the quinoid form of NaphS,2-{Zn}. Polar
CT states are expected to be better

Figure 4-16: Normal benzenoidal (left) and
charge transfer quinoidal (right) form of
Naph S,2-{Zn}

Figure 4-15: Structure
of 1-(4-
dimethylaminophenyl)
naphthalene

145

solvated in polar solvents, leading to an increase in lifetime, and resulting in decreased
radiative and nonradiative rates.
28–31
This was observed for NaphS,2-{Zn}.
4.2.3.3 AnR,2-{Zn}
The radiative rate of AnR,2-{Zn} diminishes as the nonradiative rate increases with
increasing solvent polarity (Figure 4-12 a, b). This behavior is typical for D-A dyads and
zinc complexes that undergo charge transfer.
32–34
Whereas Naphs,2-{Zn} and DPAR,2-{Zn}
form a stabilized, emissive, CT state with quidinoidal character (Figure 4-16), AnR,2-{Zn} is
most stable in the benzenoid form. This is evidenced by the high radiative rate in
nonpolar solvents. Studies have shown that, compared with naphthalene and benzene,
the sterics and strong aromatic localization of anthracene inhibits delocalization and
formation of the quinoid structure.
35,36
Forcing AnR,2-{Zn} into this conformation results
in a destabilized CT state and a reduced radiative rate with increasing solvent polarity.
AnR,2-{Zn} also exhibits dual fluorescence in polar solvents, THF and ACN. This was
further investigated through concentration and solid state studies.
Overall, the homoleptic complexes have demonstrated the formation of a CT state
formed between BOX and the chromophore, which is ideal for energy transfer processes.
This is investigated further through the formation of heterochiral, heteroleptic complexes
(section 4.2.5).
4.2.4 AnR,2-{Zn}: LE and CT states
To further probe the dual fluorescence of AnR,2-{Zn}, excited state dynamics were
studied. AnR,2-{Zn} has distinct emission spectra from the locally excited (LE) state of
anthracene, centered at 410 nm, and the CT state, centered at 545 nm, in DCM (Figure
4-17). Comparing the emission spectrum of AnR,2-{Zn} in ACN (Figure 4-11, d) with that
146

in DCM, the LE emission does not shift with solvent polarity, whereas the CT state does.
This confirms the nature of the two individual states. The CT state emission dominated
when AnR,2-{Zn} was excited at 310 nm (where BOX absorbs), whereas LE emission
dominated when it was excited at 360 nm (anthracene’s absorption) (see Figure 4-17, b).
250 275 300 325 350 375 400 425
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
(a)
400 450 500 550 600 650 700 750
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
(b)

Figure 4-17: (a) Excitation of homochiral, homoleptic An R,2-{Zn} from emission wavelength 530
nm (black dash) and 440 nm (grey dash); (b) emission from exciting An R,2-{Zn} at 310 nm
(black solid) and 360 nm (grey solid)
Charge from exciting BOX is being donated to the anthracene moiety; however, when
anthracene is excited, this mechanism is supressed. Further evidence of two excited
states present is found in the excitation spectra of these two states, with the LE resembling
that of the ligand absorption and the CT overlapping with the complex absorption (Figure
4-17 a and Figure 4-18, a). The lifetime of these individual species also differ, with τLE=6.9
ns and τCT=13 ns. Simultaneous emission from a LE and CT state is commonly found in
charge transfer dyads, but the excitation spectrum still matches the absorption spectrum.

37–40
It is unusual that a single complex undergoing charge transfer has two different
excitation spectra. The nature of these two states was investigated further through
concentration studies.

The absorption and emission of the free ligand, An-BOXRR, and AnR,2-{Zn} at various
concentrations were measured in DCM (see Figure 4-18 a, b). The absorbance spectra
were normalized to the peak at 260 nm. As concentration decreases, the contribution of
147

BOX (λabs=315nm) decreases, approaching that of the free ligand. This phenomena is
enhanced in the emission spectra, as the CT state emission decreases at solute
concentrations below 0.21 μm. This indicates that at low concentrations, the complexes
are dissociating in polar media. These experiments were carried out in DCM using the
homoleptic, homochiral complex. To determine if it is a product of the solvent, it was
repeated in acetonitrile where the same conclusions were reached. As the homochiral
complex is known to be less thermodynamically stable than the heterochiral complex, the
concentration studies were also repeated with AnR-{Zn}-AnS, with yet again similar
results.
250 275 300 325 350 375 400 425
0.2
0.4
0.6
0.8
1.0
Normalized Absorbance (a.u.)
Wavelength (nm)
(a)
400 450 500 550 600 650 700
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
(b)

400 450 500 550 600 650 700
5.0E4
1.0E5
1.5E5
2.0E5
2.5E5
Intensity (a.u.)
Wavelength (nm)
(d)

Figure 4-18: (a) Absorbance and (b) emission of An R,2-{Zn} at concentrations of: 0.32 μM
(red), 0.21 μM (orange), 0.10 μM (yellow), 0.05 μM (light green), 0.03 μM (green), 0.01 μM
(blue), 5.4x10
-3
μM (navy) and An-BOX RR (black); (c) :
1
H-NMR of An R,2-{Zn} and An R-{Zn}-An S
in d-acetonitrile, measured at t=0 and at t=24hours and (d) Successive measurements of the
emission of An R,2-{Zn} at 1wt% in polystyrene matrix dissolved in toluene (λ ex=360nm).

148

To confirm dissociation in polar media,
1
H-NMR in d-acetonitrile was measured
for both the homochiral and heterochiral complex (Figure 4-18, c). After one day, no
dissociation was observed, so it is an excited state phenomena. AnR,2-{Zn} was frozen in
a polystyrene matrix at 1wt%, and the emission spectra was measured (Figure 4-18, d).
As polystyrene is non-polar, the CT state is hyspochromically shifted from that in DCM,
and it has strong LE character. Ten measurements were taken in succession and the CT
energy gradually diminished as the LE emission grew in. After bond dissociation, the
rigid matrix prevents the ligands from rebonding with zinc. This confirms that the
complexes are photodissociating. Photodissociation has not previously been mentioned
in BOX literature, though this is unsurprising as no previous, in depth, photophysical
analysis has been done with these ligands.
5

In designing D-A complexes, photodissociation is not ideal. However, as observed in
Figure 4-18 (a), the ratio of the LE and CT emission is the same in solutions at
concentrations of 0.21 μM and greater. An equilibrium is reached at these concentrations,
allowing for further studies to be performed at or above these concentrations. This
phenomenon of dual fluorescence in polar media was only observed in An-{Zn}
complexes. Concentrated samples were utilized for the following measurements.
4.2.5 Intramolecular Energy Transfer
Intramolecular FRET occurs through dipole-dipole interactions, the two fluorophores
must be within 10 nm of each other, and the absorbance spectra of the A must overlap
with the emission of the D. Naph (D) and An (A) have been previously studied as a
donor/acceptor pair for energy transfer.
14,41,42
Naph has been found to efficiently energy
transfer to An, as the molecules electronic spectra have substantial overlap. They were
investigated for intramolecular FRET within the BOX2Zn complexes by synthesizing
149

NaphS-{Zn}-AnR.
DFT calculations were done with the same conditions as those found in section
4.2.2. The HOMO for these complexes is localized on BOX and the meso- position
chromophore (Figure 4-1). The PhS lowers the HOMO, allowing a comparison of a single
fluorescent ligand participation versus two. The HOMO and LUMO energies are nearly
identical for NaphS-{Zn}-AnR and AnS-{Zn}-AnR, as the electron density is localized on
the AnR ligand (Figure 4-19). While AnS-{Zn}-AnR has near degnerate HOMO/HOMO-1
and LUMO/LUMO+1 states, the NaphS destabilizes the LUMO+1 and HOMO-1 in NaphS-
{Zn}-AnR. All three complexes have similar LUMO energies, so they should emit the
same, should there be not significant solvent effects.
LUMO

HOMO

Ph S-{Zn}-An R Naph S-{Zn}-An R An S-{Zn}-An R

Figure 4-19: DFT calculated HOMO and LUMO electron density on heteroleptic BOX 2Zn
complexes, Ph S-{Zn}-An R, Naph S-{Zn}-An R, An S-{Zn}-An R and MDPA S-{Zn}-An R.
150

Table 4-4: DFT calculated HOMO and LUMO energies (eV) for An-{Zn} complexes

HOMO-2 HOMO-1 HOMO LUMO LUMO+1 LUMO+2
An R-{Zn}-Ph S -4.93 -4.87 -4.84 -1.39 -0.57 -0.44
An R-{Zn}-Naph S -5.01 -4.84 -4.79 -1.41 -0.82 -0.6
An R-{Zn}-An S -5.01 -4.79 -4.79 -1.41 -1.41 -0.63

The emission of NaphS,2-{Zn} overlaps the absorption of AnR,2-{Zn}, enabling
FRET from naphthalene to anthracene (Figure 4-20). CH and TOL, significant overlap is
observed, whereas in more polar solvents (THF and ACN), the emission of NaphS,2-{Zn}
shifts to minimally overlap with AnR,2-{Zn} absorption. It is also important to note that
the lifetime of NaphS,2-{Zn} was not detected in CH, and the quantum yield was effectively
quenched so a radiative rate was not quantified. FRET is dependent on the radiative rate
of the donor, kD, as given in the equation for the rate of electron transfer, kET, eqn 4-3:
20

4-3 𝑘 𝐸𝑇
=
𝑘 𝐷 𝛼 𝜅 2
𝐽 ( 𝜀 𝐴 )
𝑅 𝐷𝐴
6

where  is a proportionality constant, κ takes into account the orientation of the dipoles
of the donor and acceptor, J( A) is the
spectral density integral (a measure of the
overlap between the A absorption and D
emission) and RDA is the distance between
the donor and acceptor. Given the quench
kD in CH, and minimal overlap in THF and
ACN, efficient FRET is only expected to be
observed in TOL.
PhS-{Zn}-AnR, NaphS-{Zn}-AnR and AnS-{Zn}-AnR were prepared and the steady state
photophysics studied (see Figure 4-21 and Table 4-5). The absorbance spectra of all the
300 350 400 450 500 550 600
0.0
0.2
0.4
0.6
0.8
1.0
CH
TOL
THF
ACN
Normalized Intensity (a.u.)
Wavelength (nm)

Figure 4-20: Absorbance spectra of An R,2-
{Zn} in CH (red) and emission of Naph S,2-
{Zn} in various solvents (blue)

151

complexes have a long-wavelength edge of the low energy band from anthracene,
indicating a CT state. The emission spectra of all three complexes are similar in nonpolar
solvents, with emission from the CT state formed on AnR. The radiative rates decrease as
the CT state is formed in polar solvents, similar to AnR,2-{Zn}, indicating efficient charge
separation.
PhS-{Zn}-AnR, and AnS-{Zn}-AnR have similar absorption and emission spectra
while NaphS-{Zn}-AnR is different (Figure 4-21). Despite having similar electronic
spectra, the quantum yield of PhS-{Zn}-AnR is consistently lower than AnS-{Zn}-AnR,
AnR,2-{Zn} and NaphS-{Zn}-AnR (Table 4-5). PhS-{Zn}-AnR is a reference complex to
determine how one AnR ligand would act on Zn, as the PhS ligand does not contribute to
excited state emission. Should the species with two emissive mPM D/A (ie. AnS-{Zn}-
AnR) not interact with each other in the excited state, their photophysical properties
would be identical to PhS-{Zn}-AnR As this is not the case, there is potentially exciton
coupling between the CT states of the two mPM D/A.

152

0.2
0.4
0.6
0.8
1.0
250 275 300 325 350 375 400 425
Normalized Intensity (a.u.)
(a)
400 450 500 550 600 650 700
0.2
0.4
0.6
0.8
1.0
CH
TOL
THF
ACN
Normalized Intensity (a.u.)
(b)

0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
(c)
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
(d)

250 275 300 325 350 375 400 425
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
(e)
400 450 500 550 600 650 700
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
(f)

Figure 4-21: Normalized absorption (left) and emission (excited at 330 nm) (right) of (a,b) Ph S-
{Zn}-An R (black), (c,d) Naph S-{Zn}- An R (blue) and (e,f) An S-{Zn}-An R (red) in CH (solid), TOL
(dash), THF (dash-dot) and ACN (dot)

153

Table 4-5: Emission maximum, quantum yield (Φ), lifetime (τ) and radiative rate (k r) of Naph S-
{Zn}-An R (λ ex=330nm)
Solvent λem (nm) ϕ τ (ns)
k r
(x10
7
s
-1
)
k nr
(x10
7
s
-1
)
Ph S-{Zn}-An R
CH 446 0.55 3.2 17.3 14.0
TOL 468 0.77 5.0 15.4 4.5
THF 526 0.31 14.4 2.2 4.8
ACN
410
0.01
6.8 (92)
- - 12.8 (8)
614 3.2
Naph S-{Zn}-An R
CH 446 0.70 3.2 22.0 9.5
TOL 470 1.0 4.9 20.4 0
THF
432
0.41
2.0 (88)
- - 5.2 (12)
522 14.2
ACN 480 0.21
3.0 (23)
- -
6.7 (77)
An S-{Zn}-An R
CH 446 0.72 3.2 22.8 8.9
TOL 470 0.83 4.8 17.2 3.5
THF 524 0.44 14.1 - -
ACN
410
0.05
7.0 (94)
- -
12.8 (6)
618 3.4
15.1 (22)

FRET from NaphS to AnR was investigated by comparing, the results in TOL of
NaphS-{Zn}-AnR to AnS-{Zn}-AnR. The quantum yield of NaphS-{Zn}-AnR in TOL is unity,
whereas AnS-{Zn}-AnR was 0.83. The radiative rate of NaphS-{Zn}-AnR also increased
from that of AnS-{Zn}-AnR (Table 4-5). This is evidence of intramolecular energy transfer
from NaphS to AnR.
In nonpolar CH and slightly polar TOL, NaphS-{Zn}-AnR has a similar emission
profile to the other An-{Zn} heteroleptic complexes, but in THF and ACN there is a
distinct dual emission (Figure 4-22). Dual fluorescence in An-{Zn} complexes typically
attributes the higher energy emission to the LE state, and the low energy to AnR. In this
154

case, the high energy wavelength in NaphS-
{Zn}-AnR overlaps significantly with
NaphS,2-{Zn}, with a shoulder overlapping
with the LE of AnR,2-{Zn} (Figure 4-22, c).
The lifetimes measured at the two λem, 432
and 522, also correspond to the lifetimes of
the homochiral, homoleptic complexes,
NaphS,2-{Zn}, and AnR,2-{Zn}, respectively.
In ACN, the majority of emission is from
NaphS, with a band edge corresponding to
the LE state of AnR (Figure 4-22, d). This
dual fluorescence is attributed to incomplete
FRET from NaphS and AnR. As solvent
polarity increases, the spectral overlap
between NaphS and AnR is reduced, such
that NaphS is no longer transferring energy
to AnR, but emitting itself.
FRET has been shown to occur in
NaphS-{Zn}-AnR, though only efficiently in
TOL.
4.2.6 Intramolecular Electron Transfer
Intramolecular photoinduced electron transfer (PET) between the two ligands on Zn
can occur with properly tuned energy levels of the mPM D/A. PET is observed in steady
state photophysics by an increase in the non-radiative rate, as the radical ions formed are
0.2
0.4
0.6
0.8
1.0
350 400 450 500 550 600 650
Naph
S,2
-{Zn}
An
R,2
-{Zn}
Naph
S
-{Zn}-An
R
Normalized Intensity (a.u.)
(a)

0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
(b)

0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
(c)

350 400 450 500 550 600 650
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
(d)

Figure 4-22: Emission of Naph S,2-{Zn}
(blue), An R,2-{Zn} (red) and Naph S-{Zn}-
An R (purple) in (a) CH, (b) TOL, (c) THF
and (d) ACN
155

nonemissive. MDPAS-{Zn}-AnR and MDPAS-{Zn}- DPAR and were designed to be push-
pull, A-{Zn}-D complexes. AnR,2-{Zn} and DPAR,2-{Zn} have demonstrated a stabilized
charge transfer state on the AnR and DPAR ligands (respectively), which will facilitate
electron transfer to a strong donor, MDPA. PhS-{Zn}-DPAR was also made as a reference
complex, to compare with MDPAS-{Zn}- DPAR.
DFT calculated HOMO/LUMO energies (using the same conditions given in section
4.2.2) show that complexes MDPAS-{Zn}-AnR and MDPAS-{Zn}-DPAR should undergo
electron transfer from MDPAS to AnR and DPAR (Figure 4-23). PET is a competing
nonradiative decay pathway, resulting in a lower radiative rate, which should be
measured if PET is occuring in these complexes.
LUMO

HOMO

Ph S-{Zn}-DPA R MDPA S-{Zn}-DPA R MDPA S-{Zn}-An R

Figure 4-23: DFT calculated HOMO and LUMO electron density on heteroleptic BOX2Zn
complexes, Ph S-{Zn}-DPA R, DPA S-{Zn}-DPA R, and MDPA S-{Zn}-DPA R.

156

Table 4-6: DFT calculated HOMO and LUMO energies (eV) for DPA-{Zn} complexes

HOMO-2 HOMO-1 HOMO LUMO LUMO+1 LUMO+2
DPA R-{Zn}-Ph S -5.63 -4.87 -4.79 -0.9 -0.6 -0.46
DPA R-{Zn}-MDPA S -5.28 -4.84 -4.35 -0.9 -0.57 -0.46
An R-{Zn}-MDPA S -4.95 -4.73 -4.16 -1.36 -0.49 -0.35

4.2.6.1 PET in MDPAS-{Zn}-AnR
The most obvious case of PET is observed in MDPAS-{Zn}-AnR, as evidenced by
the increase in non-radiative rate in CH and TOL, as compared with that of PhS-{Zn}- AnR
and AnS-{Zn}-AnR (Table 4-7 and Table 4-5, vide supra). Should there be a nonradiative
decay pathway other than PET in heteroleptic complexes, PhS-{Zn}- AnR would have the
same non-radiative rate (Table 4-5, vide supra). Electron transfer is difficult to qualify
for MDPAS-{Zn}-AnR in ACN, as the quantum yield of all of the An-{Zn} complexes is
quenched to 0.01-0.05 (Table 4-5, vide supra).
Increasing solvent polarity leads to a prolonged lifetime, likely due to the charge
separated state. In THF, there is a biexponential lifetime; one short lifetime from the
decay of the CT state after excitation, and a long lifetime of 66 ns. This can be attributed
to a long-lived CSS formed between the ligands after electron transfer from MDPAS to
AnR, which then recombines to radiatively decay. These are promising results, as a long
charge separated lifetime is required for charge transfer to a catalyst to occur.
2,4
The
biexponential lifetime indicates that the S1 and CT are close in energy, so that electron
transfer and back electron transfer are occuring, and there is emission from both states.
43

Further investigation into the nature of electron transfer in this complex can be done
through electrochemistry measurement (cyclic voltammetry and differential pulse
voltammetry) and transient absorption spectroscopy.

157

0.2
0.4
0.6
0.8
1.0
250 275 300 325 350 375 400 425
Normalized Intensity (a.u.)
(a)
400 450 500 550 600
0.2
0.4
0.6
0.8
1.0
CH
TOL
THF
ACN
Normalized Intensity (a.u.)
(b)

0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
(c)
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
(d)

250 275 300 325 350 375 400 425
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
(e)
400 450 500 550 600 650 700
0.2
0.4
0.6
0.8
1.0
Normalized Intensity (a.u.)
Wavelength (nm)
(f)

Figure 4-24: Normalized absorption (left) and emission (right) of (a,b) Ph S-{Zn}-DPA R (black) ,
(c,d) MDPA S-{Zn}-DPA R (blue) and (e,f) MDPA S-{Zn}-An R (purple) in CH (solid), TOL
(dashed), THF (dash-dot) and ACN (dot)

158

Table 4-7: Emission maximum, quantum yield (Φ), lifetime (τ) and radiative rate (k r) of DPA-
BOX complexes

Solvent
λ em
(nm)
ϕ τ (ns)
k r
(x10
7
s
-1
)
k nr
(x10
7
s
-1
)
DPA R-{Zn}-Ph S
CH 380 0.07 2.6 42.4 600
TOL 400 0.54 1.0 57.2 48.0
THF 424 0.33 1.9 28.5 24.1
ACN 454 0.66 3.8 17.5 9.2

DPA R-{Zn}-MDPA S
CH 380 0.12
1.3 (55)
- -
10.9 (45)
TOL 400 0.43
0.9 (98)
- -
3.8 (2)
THF 422 0.41
1.0 (35)
- -
5.9 (65)
ACN 454 0.04
2.8 (52)
- -
4.6 (48)
MDPA S-{Zn}-An R
CH 448 0.37 3.3 11.4 19.4
TOL 470 0.44 5.0 8.9 11.1
THF 524 0.08
12.4 (58)
- -
65.7 (42)
ACN 410 0.01
5.8 (78)
- -
15.1 (22)

4.2.6.2 PET in MDPAS-{Zn}-DPAR
To investigate electron transfer in MDPAS-{Zn}-DPAR, the heterochiral complexes,
PhS-{Zn}-DPAR, and MDPAS-{Zn}-DPAR, were prepared and the photophysical properties
measured (Figure 4-24 and Table 4-7). Exciton coupling has been extensively studied in
diphenylacetylene complexes by appending DPA with either a donor or acceptor
molecule. DPA appended with various donor molecules has demonstrated a
solvatochromic shift, due to charge redistribution in the excited state.
16,21,23

The contribution to the absorption of the CT band at 360 nm is halved in DPAR-{Zn}-
PhS and DPAR-{Zn}-MDPAS as compared with DPAR,2-{Zn}, as there is only one DPA-
BOX ligand. Similar conclusions about the observed effect of exciton coupling between
CT states in the An-{Zn} complexes can be made with the DPA-{Zn} complexes. PhS-{Zn}-
159

DPAR has only one DPA-BOX ligand, so exciton coupling between CT states formed on
the BOX ligands will not be as strong as that in homochiral, homoleptic complex, DPAR,2-
{Zn} (confirmed by the reduced radiative rate of PhS-{Zn}-DPAR from DPAR,2-{Zn}).
Electron transfer in MDPAS-{Zn}-DPAR was less obvious than in MDPAS-{Zn}-AnR.
While intramolecular electron transfer was observed across the solvents studied for
MDPAS-{Zn}-AnR, there was only a noticeable drop in radiative rate for MDPAS-{Zn}-
DPAR in polar ACN (as compared with PhS-{Zn}-DPAR, Table 4-7) . The CT state of DPAR
is the most stable in this solvent, allowing for effiecient charge separation (Table 4-3, vide
supra).
Electron transfer in MDPAS-{Zn}-DPAR can be improved upon by making DPAR a
stronger acceptor. This can be done by appending a cyano (CN) group to the mPM, DPA.
4.3 Conclusions
This study has demonstrated that chirality-driven assembly of BOX2Zn complexes is
a simple method to form push-pull dyads, allowing study of various intramolecular
photoinduced energy transfer processes. All complexes (except MDPAS,2-{Zn})
demonstrate a CT state formed between BOX and the mPM, which was stabilized by Zn.
The nature of the CT state was determined by the mPM. It was also found that An2-{Zn}
complexes under photolysis, which can be avoided at concentrations higher than 0.2 μM.
Intramolecular energy transfer via FRET was observed in NaphS-{Zn}-AnR in TOL.
The bathochromic shift of the CT state of NaphS,2-{Zn} reduced the spectral overlap with
AnR,2-{Zn}, leading to inefficient energy transfer in NaphS-{Zn}-AnR in more polar
solvents.
Intramolecular PET from MDPA to An and DPA in heterochiral, heteroleptic zinc
complexes, AnR-{Zn}-MDPAS and DPAR-{Zn}-MDPAS, respectively, was observed
160

through steady-state photophysical measurments. PET was more efficient in AnR-{Zn}-
MDPAS than in DPAR-{Zn}-MDPAS, as AnR has a stabilized LUMO compared with DPAR.
PET was undergone in AnR-{Zn}-MDPAS in CH and TOL.
This preliminary work has demonstrated multiple electronic pathways in BOX2-{Zn}
complexes, opening an avenue for further study. The ease of chirality driven-assembly of
light harvesting molecular arrays makes this an attactive strategy. Further understanding
of the photolysis mechanism is required to ensure these are stable complexes for
photocatalytic reactions.

161

4.3.1 Future Work
This was the seminal study of a new class of light harvesting molecules. There are
many directions of research this study can take, depending on the mechanism involved.
Further investigation into FRET, PET, CT exciton coupling, and photolysis are discussed
below.
4.3.1.1 Intramolecular FRET
FRET in these complexes can be investigated further through improving the spectral
overlap. This can be done by utilizing the D, NaphS with an A that has a red-shifted
absorption from that of anthracene, such as tetracene. Conversely, a D with a
hypsochromically shifted emission can be utilized with A, AnR. The rate at which energy
is transferred can be determined through time-resolved fluorescence studies.
44

4.3.1.2 Intramolecular PET
PET in these complexes can be further investigated first through electrochemistry
measurements to measure the reduction and oxidation potential. These measurements
will confirm PET, and the free energy change for electron transfer can be calculated
(equation 4-4).
20

4-4 Δ 𝐺 𝑒𝑡
= ( 𝐼𝑃 )
𝐷 − ( 𝐸 𝐴 )
𝐴

where (IP)D is the ionization potential of the donor and (EA)A is the electron affinity of
the acceptor.
These can be further improved upon through other complexes with stronger donor or
acceptor character, such as the addition of a cyano group on the acceptor, or utilizing a
stronger donor, such as one with more N atoms. DFT calculations can approximate the
S1 energy. Raising this energy will ideally reduce back electron transfer from the CT state.
162

The electron transfer processes can be measured through transient absorption
spectroscopy, to measure the spectra of the radical ion species formed.
4.3.1.1 CT Exciton Coupling
The potential for exciton coupling between CT states formed on the ligands is not
investigated in this study. However, a comparison of the photophysical properties of PhS-
{Zn}-AnR and PhS-{Zn}-DPAR to their homochiral, homoleptic counterparts (AnR,2-{Zn}
and DPAR,2-{Zn}, respectively) provides evidence of exciton coupling between two
emissive mPM, through the enhanced radiative rate. Circular dichroism will identify if
exciton coupling is occuring, through a spectral signature of exciton couplets.
18
This can
be a stepping stone for further investigation into this unique process.
4.3.1.2 Photolysis
Photolysis was investigated in AnR,2-{Zn}, and confirmed results were found for PhS-
{Zn}-AnR. This can be further investigated through materials characterization after
photolysis, such as H-NMR, C-NMR and MALDI. Ligating BOX to other divalent metal
ions (e.g. Mg, Ca, Cd and Sn) and determining their photostability will probe if this is an
inherent property to BOX2Zn, or just to BOX.
This work has shown that chirality-driven light harvesting dyads have an
interesting and bright future!

163

4.4 Experimental Procedures
4.4.1 General Methods
All reactions were carried out using standard Schlenk techniques using dried and
degassed solvents. Ethyl 2-(4-bromophenyl)acetate was purchased from TCI, ethyl
cyanoformate and zinc acetate were purchased from Sigma Aldrich, (S)-(+)-2-
phenylglycinol and (R)-(-)-2-phenylglycinol were purchased from Matrix Scientific.
NMR spectra were recorded on a Varian 400 NMR spectrometer and referenced to the
residual proton resonance of chloroform (CDCl3) solvent at 7.26 ppm or acetonitrile
(CD3CN) at 1.94 ppm. The UV-visible spectra were recorded on a Hewlett-Packard 8453
diode array spectrometer. Photoluminescent emission measurements were performed
using a Photon Technology International Quantamaster Model C-60 Fluorimeter.
Fluorescent lifetimes were measured by time-correlated single-photon counting using an
IBH Fluorocube instrument equipped with an LED excitation source. Quantum yield
measurements were carried out on the fluorimeter with diphenylanthracene in
cyclohexane as a reference, with a quantum yield of 100%.
4.4.2 Synthesis of diethyl malonate derivatives
4.4.2.1 Synthesis of diethyl 2-(4-bromophenyl)malonate
Diethyl 2-(4-bromophenyl)malonate was synthesized according to the procedure by
Aube.
45
The crude product was purified by column chromatography with ethyl acetate
and hexane (20:80). The final product was a colorless oil. Yield: 78%, 6g.
1
H NMR (400 MHz, Chloroform-d) δ 7.45 (d, 2H), 7.28 (d, 2H), 4.58 (s, 1H), 4.25 – 4.10
(m, 4H), 1.22 (t, J = 7.1, 0.7 Hz, 7H).
164

Figure 4-25: Synthesis of Diethyl 2-(4-bromophenyl)malonate

4.4.2.2 Synthesis of diethyl 2-(4-(phenylethynyl)phenyl)malonate
Diethyl 2-(4-(phenylethynyl)phenyl)malonate was synthesized by Sonogashira
coupling by modifying the procedure given by Korovina et al.
46
In a 100 mL dry round
bottom flask was added 1 eq diethyl 2-(4-bromophenyl)malonate (2.4g, 7.62 mmol), 0.05
eq Pd(PPh3)4 (0.44 g, 0.38 mmol) and 0.06 eq CuI (0.07 g, 0.38 mmol). It was backfilled
w/N2 three times, then 50 mL dry THF was added, and heated to 60°C. 2.7 eq
phenylacetylene (1.31 g, 12.83 mmol) was dissolved in 10 mL dry triethylamine (TEA),
which was sparged with N2 for 15 minutes. This was added dropwise to the THF solution
while vigorously stirring. The resulting brown mixture was stirred overnight at 60°C.
Upon cooling to room temperature, the reaction was quenched with saturated NH4Cl and
extracted with DCM. The combined organic layers were dried over sodium sulfate, and
the solvent was remove in vacuo. The product was purified by flash chromatography on
silica gel using ethyl acetate:hexanes (20:80). Yield: 82% (2.1g).
1
H NMR (400 MHz, Chloroform-d) δ 7.55 – 7.50 (m, 4H), 7.42 – 7.37 (m, 2H), 7.37 – 7.32
(m, 2H), 7.26 (m, 2H), 4.61 (s, 1H), 4.31 – 4.14 (m, 4H), 1.27 (t, J = 7.1 Hz, 6H).
13
C NMR
(101 MHz, Chloroform-d) δ 167.83, 132.78, 131.75, 131.62, 129.32, 128.33, 123.30, 123.13,
165

88.88, 61.94, 57.82, 13.99.

Figure 4-26: Synthesis of diethyl 2-(4-(phenylethynyl)phenyl)malonate

4.4.2.3 Synthesis of diethyl 2-(4-
methyl(phenyl)amino)phenyl)malonate
Synthesized the precursor, ethyl 2-(4-methyl(phenyl)amino)phenyl)acetate,
following the procedure given by Miyake et al.
47
The final product was purified by column
chromotography on silica, with 5% ethyl acetate in hexanes. Yield: 75%, 1.04 g.
1
H NMR
(400 MHz, Chloroform-d) δ 7.33 – 7.23 (m, 2H), 7.28 – 7.16 (m, 2H), 7.08 – 6.98 (m,
3H), 7.02 – 6.92 (m, 2H), 4.17 (q, J = 7.1, 0.6 Hz, 2H), 3.57 (s, 2H), 3.31 (s, 3H), 1.28 (t, J
= 7.1, 0.6 Hz, 3H).

Figure 4-27: Synthesis of ethyl 2-(4-methyl(phenyl)amino)phenyl)acetate

Synthesized diethyl 2-(4-methyl(phenyl)amino)phenyl)malonate (MDPA-DM)
166

following the procedure given by Aube et al. for the bromo-derivative (see section 4.4.2.1),
using the precursor synthesized above. The final product was purified by column
chromotography on silica, with 20% ethyl acetate in hexanes. Product was a green-blue
oil (Figure 4-28). Yield: 76%, 963 mg.
1
H NMR (400 MHz, Chloroform-d) δ 7.36 – 7.22
(m, 4H), 7.14 – 6.90 (m, 5H), 4.56 (s, 1H), 4.32 – 4.15 (m, 4H), 3.32 (s, 3H), 1.28 (t, J =
7.1 Hz, 6H).
13
C NMR (101 MHz, Chloroform-d) δ 168.52, 148.85, 148.62, 129.98, 129.30,
124.31, 122.38, 122.10, 118.67, 61.69, 57.25, 40.18, 14.06.

Figure 4-28: Synthesis of diethyl 2-(4-methyl(phenyl)amino)phenyl)malonate

4.4.2.4 Synthesis of diethyl 2-(9,10-diphenylanthracene) malonate (An-
DM):
A general procedure for Suzuki coupling was modified from Tu. et al.
48

In an oven-dry, 50 mL RBF, added diethyl 2-(4-bromophenyl)malonate (1 g, 1 eq), 10-
(phenylanthracen-9-yl)-boronic acid (995 mg, 1.05 eq)and K2CO3 (877 mg, 2 eq).
Pumped and backfilled three times. Added anhydrous toluene (10 mL) to dissolve. Added
DI water (5 mL). Purged with N2 for 15 minutes. Added Pd(P(Ph)3)4 (367 mg, 0.1 eq)
catalyst, then refluxed for 24 hours. After cooling down to r.t., water (40.0 mL) was
added. Then, the resulting mixture was extracted with CH2Cl2 (3x30 mL) and the
167

combined organic layers were dried with NaSO4. Removal of the solvents in vacuo gave a
residue, which was subjected to column chromatography on silica gel, 20% ethyl acetate
in hexanes resulted in a white-yellow solid (Figure 4-29). Yield=49%, 0.75 g.
1
H NMR
(400 MHz, Chloroform-d) δ 7.74 – 7.66 (m, 4H), 7.70 – 7.62 (m, 2H), 7.66 – 7.58 (m,
2H), 7.61 – 7.51 (m, 2H), 7.55 – 7.43 (m, 4H), 7.39 – 7.28 (m, 4H), 4.81 (s, 1H), 4.42 –
4.24 (m, 4H), 1.42 – 1.31 (m, 5H). MALDI-TOF: m/z calculated: 488.58 [M]+; found:
[M]+

Figure 4-29: Synthesis of diethyl 2-(9,10-diphenylanthracene) malonate

4.4.2.5 Synthesis of 2-(4-naphthalen-1-yl) diethyl malonate (NaPh-DM)
The synthesis for 2-(4-naphthalen-1-yl) diethyl malonate was done with the same
procedure as that of 9,10-diphenylanthracene diethylmalonate, except with 4-
(naphthalen-1-yl)boronic acid as the boronic acid (Figure 4-30). Yield: 95% (1.1 g)
1
H
NMR (400 MHz, Chloroform-d) δ 7.95 – 7.82 (m, 3H), 7.58 – 7.40 (m, 9H), 4.72 (s, 1H),
4.36 – 4.19 (m, 4H), 1.32 (t, J = 7.1 Hz, 6H).
168

Figure 4-30: Synthesis of 2-(4-naphthalen-1-yl) diethyl malonate

4.4.3 General Synthesis of bis(oxazoline) (BOX)
The general synthesis of
R,H
BOX-Ph2 was taken from literature.
49
A stirred mixture
of the respective diethyl malonate (5 g, 21 mmol) with 2.00 eq (R)- or (S)- phenylglycinol
(5.81 g, 42 mmol) and 0.02 eq sodium hydride (17 mg, 0.42 mmol, 60 % dispersion in
mineral oil) was heated in an Ace Glass heavy-wall pressure tube under a nitrogen
atmosphere to 160°C for 3 to 4 hours. The product was dissolved in methylene chloride
and transferred to a oven-dried schlenk-round bottom flask to remove the organic
solvents in vacuo. The respective malonamides were yielded quantitatively and used
without further purification.
In the same RBF, 5.00 eq triethylamine (10.7g, 106 mmol) was added to the diamide
(8.9 g, 21 mmol) and dissolved in ~300 mL dry methylene chloride. This was cooled to
0°C before treating with 2.50 eq methanesulfonyl chloride (6 g, 53 mmol). The solution
was stirred for 1 h at room temperature and then washed with saturated ammonium
chloride solution (3 x 350 mL). The solvent was removed in vacuo and colourless to
orange oils were obtained.
The respective mesylate (12.2 g, 21 mmol) was redissolved in 400 mL methanol/water
(1 : 1) (without further workup). 5.00 eq of sodium hydroxide (4.2 g, 106 mmol) were
added and the solution was heated under reflux for 2 h. It was then concentrated in vacuo
and the aqueous phase extracted with methylene chloride (3 x 300 mL). The combined
169

organic phase was washed well with water and dried over sodium sulfate and the solvent
was removed in vacuo.
The steps for the general procedure are given in Figure 4-31, with a meso- position
phenyl.

Figure 4-31: Ph-BOX S synthesis
(a) (b)
Figure 4-32:H-NMR of (a) Ph-BOXSS and (b) Ph-BOXRR

4.4.3.1 Synthesis of
Ph
BOX-Ph2
The synthesis of
Ph
BOXSS-Ph2 and
Ph
BOXRR-Ph2 followed the general procedure for
BOX, from diethyl phenylmalonate; the crude product was purified by crashing out in
170

methanol as an off-white powder (Yield: 65%, 5.26 g). Characterization of
Ph
BOX-Ph2 has
been previously reported by Atkins et al.
11

1
H-NMR (400 MHz, Chloroform-d) δ 7.46-7.23 (m, 14H), 7.22-7.10 (m, 1H), 5.14 (dd, J=
8.9, 7.2 Hz, 2H), 4.63 (dd, J= 8.6, 8.6 Hz, 2H), 4.09 (dd, J = 8.4, 7.3 Hz, 2H).
4.4.3.2 Synthesis of
DPA
BOX-Ph2
Diphenylacetylene (DPA) was appended to the meso- position of BOX to form compound
DPA
BOX-Ph2.
DPA
BOXSS-Ph2 and
DPA
BOXRR-Ph2 were synthesized by the general
procedure for bis(oxazoline), utilizing the synthesized diethyl 2-(4-
(phenylethynyl)phenyl)malonate as the precursor. The malonamide was a greenish-
yellow oil, and the corresponding mesitylate was a rust colored oil. The final product is a
dark yellow-orange oily-solid, and was used directly in the formation of the zinc complex
without further purification. Yield: 78% (1.1 g).
4.4.3.3 Synthesis of
MDPA
BOXSS-Ph2
MDPA
BOXSS-Ph2 was synthesized by the general procedure for BOX, utilizing the
synthesized MDPA-DM was used as the precursor. The malonamide was an orange oil,
and the corresponding mesitylate was a dark red oil. The final product is an orange oily-
solid, and was used directly in the formation of the zinc complex without further
purification. Yield: 55% (0.5g).
4.4.3.4 Synthesis of
Anth
BOX-Ph2
An
BOXSS-Ph2 and
An
BOXRR-Ph2were synthesized by the general procedure for
bis(oxazoline), utilizing the synthesized An-DM as the precursor. The final product is a
dark yellow-orange oily- solid, and was used directly in the formation of the zinc complex
without further purification. Yield: 55% (0.49 g).
4.4.3.5 Synthesis of
NaPh
BOXSS-Ph2
171

NaPh
BOXSS-Ph2 was synthesized by the general procedure for bis(oxazoline), utilizing the
synthesized NaPh-DM as the precursor. The malonamide was a green oil, and the
corresponding mesitylate was a yellow oil. The final product is a light lavender-grey solid,
and was used directly in the formation of the zinc complex without further purification.
4.4.4 General preparation of homochiral zinc complexes,
(
R
BOX-Ph2)2Zn
The general procedure for preparing the homochiral zinc complexes was taken from
Takacs et al.
12
Zinc acetate dihydrate (29 mg, 0.13 mmol) was added in one portion to a
stirred solution (rt, air) containing 2 eq. of the respective BOX (100mg, 0.26 mmol) to a
mixture of CH2Cl2:MeOH (1:1, 20 mL). The resulting clear solution was stirred for 1 hour,
and then concentrated in vacuo. The crude product was purified by chromatography on
silica (20% ethyl acetate in hexanes) to give the zinc complex. The general procedure for
synthesizing the homochiral zinc complex is given in Figure 4-33.

Figure 4-33: Synthesis of homochiral Ph S,2-{Zn} complex.

4.4.4.1 Preparation of (
Ph
BOX-Ph2)2Zn
PhR,2-{Zn} and PhS,2-{Zn} were prepared by
the general procedure. Column
chromatography on silica with eluent 20%
ethyl acetate in hexanes isolated the product.
The final product was triturated in methanol,
and the powder was isolated by filtration as a
172

light lavender powder (Yield: 84%, 840 mg). These complexes were previous
characterized by Takacs et al.
11

1
H NMR (400 MHz, Chloroform-d) δ 7.38 – 7.29 (m, 4H), 7.29 – 7.13 (m, 7H), 7.00 –
6.91 (m, 4H), 4.67 (dd, J = 8.5, 8.5 Hz, 2H), 4.34 (dd, J = 8.5, 8.5 Hz, 2H), 3.82 (dd, J =
8.4, 8.4 Hz, 2H).
4.4.4.2 Preparation of (
DPA
BOX-Ph2)2Zn
DPAR,2-{Zn} and DPAS,2-{Zn} was prepared by the general procedure. Column
chromatography on silica with eluent 20% ethyl acetate in hexanes isolated the product.
The final product was triturated in methanol, and the powder was isolated by filtration as
a light yellow powder (Yield: 65%).
1
H NMR (400 MHz, Chloroform-d) δ 7.56 – 7.50 (m, 2H), 7.50 – 7.44 (m, 2H), 7.40 –
7.27 (m, 5H), 7.25 – 7.14 (m, 6H), 7.00 – 6.91 (m, 4H), 4.67 (dd, J = 8.5 Hz, 2H), 4.36
(dd, J = 8.5 Hz, 2H), 3.84 (dd, J = 8.4 Hz, 2H).
13
C NMR (101 MHz, Chloroform-d) δ
170.12, 140.88, 131.77, 131.54, 130.54, 128.62, 128.25, 127.93, 127.83, 126.74, 123.83,
119.28, 90.33, 73.68, 66.83.
4.4.4.3 Preparation of (
MDPA
BOXSS-Ph2)2Zn
MDPAS,2-{Zn} was prepared by the general procedure. Column chromatography on silica
with eluent 20% ethyl acetate in hexanes isolated the product. The final product was
triturated in methanol, and the powder was isolated by filtration as a light tan powder
(Yield: 75%, 80mg).
1
H NMR (400 MHz, Chloroform-d) δ 7.36 – 7.21 (m, 2H), 7.25 – 7.13 (m, 6H), 7.13 – 6.99
(m, 4H), 6.99 – 6.86 (m, 5H), 6.86 – 6.81 (m, 2H), 4.65 (dd, J = 8.6 Hz, 2H), 4.35 (dd, J
= 8.5 Hz, 2H), 3.82 (dd, J = 8.4 Hz, 2H), 3.34 (s, 3H).

173

4.4.4.4 Preparation of (
An
BOX-Ph2)2Zn
AnR,2-{Zn} and AnS,2-{Zn} were prepared by the general procedure. Column
chromatography on silica with eluent 20% ethyl acetate in hexanes isolated the product.
The final product was triturated in methanol, and the powder was isolated by filtration as
a light yellow powder (Yield: 80%, 260mg).
1
H NMR (400 MHz, Chloroform-d) δ 7.99 – 7.82 (m, 2H), 7.76 – 7.61 (m, 4H), 7.63 – 7.49
(m, 4H), 7.50 – 7.27 (m, 14H), 7.10 – 7.03 (m, 4H), 4.72 (dd, J = 8.8 Hz, 2H), 4.50 (dd, J
= 8.5 Hz, 2H), 3.96 (dd, J = 8.6 Hz, 2H).
13
C NMR (101 MHz, Chloroform-d) δ 170.44,
140.90, 133.12, 131.73, 131.39, 131.36, 131.29, 130.72, 130.53, 130.32, 130.04, 129.90,
129.30, 128.82, 128.67, 128.38, 128.00, 127.89, 127.64, 127.55, 127.38, 127.08, 126.96,
126.83, 126.54, 126.31, 125.40, 125.01, 124.95, 124.66, 73.59, 66.96, 64.14, 60.39, 14.19,
1.00. MALDI-TOF: m/z calculated: 1332.92 [M]+; found: 634.5, 1329.9 [M]+
4.4.4.5 Preparation of (
NaPh
BOXSS-Ph2)2Zn
NaphS,2-{Zn} was prepared by the general procedure. Column chromatography on silica
with eluent 20% ethyl acetate in hexanes isolated the product. The final product was
triturated in methanol, and the powder was isolated by filtration as a light lavender
powder (Yield: 75%, 80mg).
1
H NMR (400 MHz, Chloroform-d) δ 8.15 (ddt, J = 8.3, 1.6,
0.8 Hz, 1H), 7.95 – 7.80 (m, 2H), 7.59 – 7.33 (m, 7H), 7.33 – 7.18 (m, 6H), 7.05 – 6.97
(m, 4H), 4.71 (dd, J = 8.6 Hz, 2H), 4.43 (dd, J = 8.5 Hz, 2H), 3.91 (dd, J = 8.5 Hz, 2H).
13
C NMR (151 MHz, DMSO-d6) δ 170.07, 141.60, 140.37, 137.12, 136.07, 133.96, 132.38,
131.39, 128.87, 128.75, 127.99, 127.21, 127.11, 126.64, 126.29, 126.09, 125.93, 73.55.
4.4.4.6 General preparation of heterochiral Zn complex,
(
R
BOXRR)Zn(
R1
BOXSS)
The general procedure for preparing the heterochiral zinc complexes was taken from
174

Atkins et al.
11
An equimolar mixture of (
R
BOXRR-Ph2)2Zn and (
R1
BOXSS-Ph2)2Zn in
CH2Cl2:MeOH (1:1, 5 mL) was stirred for 2 hours. Afterwards, the reaction mixture was
concentrated in vacuo to a volume of ~1 mL, in which a precipitate had formed. The
resulting precipitate was isolated by filtration, washed with methanol and dried in vacuo
to give the final product, (
R
BOXRR-Ph2)Zn(
R1
BOXSS-Ph2).
4.4.4.7 Preparation of (
Ph
BOXRR-Ph2)Zn(
Ph
BOXSS-Ph2) (R=R1=Ph)
PhS-{Zn}-Ph R was prepared by the general procedure. The resulting product was yielded
quantitatively as a lavender powder. Full
characterization is given by Atkins et al.
11

1
H NMR (400 MHz, Chloroform-d) δ 7.40 –
7.29 (m, 8H), 7.33 – 7.23 (m, 2H), 7.27 – 7.15
(m, 2H), 7.19 – 7.09 (m, 4H), 4.08 (dd, J = 8.4,
5.6, 1.0 Hz, 2H), 3.91 (dd, 2H), 3.41 (dd, J = 9.2,
5.5 Hz, 2H).
4.4.4.8 Preparation of (
DPA
BOXRR-Ph2)Zn(
DPA
BOXSS-Ph2) (R=R1=DPA)
DPAS-{Zn}-DPAR was prepared by the general procedure. The resulting product was
yielded quantitatively as a yellow powder.
1
H NMR (400 MHz, Chloroform-d) δ 7.52 (ddd, J = 20.6, 7.2, 1.8 Hz, 6H), 7.42 – 7.30 (m,
9H), 7.16 – 7.05 (m, 5H), 4.11 (dd, J = 8.4, 5.6 Hz, 2H), 3.94 (dd, J = 8.8 Hz, 2H), 3.41
(dd, J = 9.2, 5.5 Hz, 2H).
13
C NMR (101 MHz, Chloroform-d) δ 168.49, 143.36, 131.84,
131.54, 130.47, 128.55, 128.24, 127.61, 127.21, 77.18, 72.89, 65.46.
4.4.4.9 Preparation of (
DPA
BOXRR-Ph2)Zn(
Ph
BOXSS-Ph2) (R=DPA,
R1=Ph)
DPAR-{Zn}-PhS was prepared by the general procedure. The resulting product was
175

yielded quantitatively as a yellow powder.
1
H NMR (400 MHz, Chloroform-d) δ 7.62 – 7.46 (m, 3H), 7.45 – 7.30 (m, 11H), 7.29 –
7.17 (m, 16H), 7.17 – 7.09 (m, 6H), 4.08 (ddt, J = 11.9, 5.8, 3.1 Hz, 3H), 3.92 (dd, J = 8.7
Hz, 2H), 3.41 (dt, J = 9.1, 6.1 Hz, 3H).
13
C NMR (101 MHz, Chloroform-d) δ 132.19, 131.56,
130.49, 128.55, 128.53, 128.26, 127.30, 127.24, 72.91, 65.48.
4.4.4.10 Preparation of (
DPA
BOXRR-Ph2)Zn(
MDPA
BOXSS-Ph2) (R=DPA,
R1=MDPA)
DPAR-{Zn}-MDPAS was prepared by the general procedure. The resulting product was
yielded quantitatively as a yellow powder.
4.4.4.11 Preparation of (
An
BOXRR-Ph2)Zn(
An
BOXSS-Ph2) (R=R1=An)
(
An
BOXRR-Ph2)Zn(
An
BOXSS-Ph2) was prepared by the general procedure. The resulting
product was yielded quantitatively as a yellow powder.
4.4.4.12 Preparation of (
An
BOXRR-Ph2)Zn(
Ph
BOXSS-Ph2) (R=An, R1=Ph)
PhS-{Zn}-AnR was prepared by the general procedure. The resulting product was yielded
quantitatively as a yellow powder.
1
H NMR (400 MHz, Chloroform-d) δ 8.01 – 7.91 (m, 3H), 7.75 – 7.67 (m, 3H), 7.67 – 7.24
(m, 44H), 7.37 (s, 12H), 7.19 (tt, J = 6.6, 1.3 Hz, 15H), 4.19 (dd, J = 8.3, 5.7 Hz, 3H), 4.11
(dd, J = 8.4, 5.4 Hz, 3H), 4.04 (dd, J = 9.2, 8.4 Hz, 3H), 3.94 (dd, J = 9.2, 8.4 Hz, 3H).
4.4.4.13 Preparation of (
An
BOXRR-Ph2)Zn(
MDPA
BOXSS-Ph2) (R=An,
R1=MDPA)
MDPAS-{Zn}-AnR was prepared by the general procedure. The resulting product was
yielded quantitatively as a yellow powder.
1
H NMR (400 MHz, Chloroform-d) δ 8.01 –
7.94 (m, 3H), 7.74 – 7.43 (m, 18H), 7.42 – 7.27 (m, 23H), 7.22 – 7.15 (m, 11H), 7.12 – 7.03
(m, 4H), 4.19 (dd, J = 8.3, 5.7 Hz, 3H), 4.12 (dd, J = 8.4, 5.5 Hz, 3H), 4.04 (t, J = 8.7 Hz,
176

2H), 3.96 (dd, J = 8.8 Hz, 3H), 3.37 (s, 3H)
4.4.4.14 Preparation of (
An
BOXRR-Ph2)Zn(
NaPh
BOXSS-Ph2) (R=An,
R1=NaPh)
NaphS-{Zn}-AnR was prepared by the general procedure. The resulting product was
yielded quantitatively as a yellow powder.
1
H NMR (400 MHz, Chloroform-d) δ 8.17 (dd,
J = 20.4, 8.3 Hz, 1H), 8.02 – 7.81 (m, 5H), 7.72 – 7.64 (m, 1H), 7.63 – 7.56 (m, 3H), 7.56
– 7.45 (m, 9H), 7.45 – 7.28 (m, 14H), 7.26 (s, 6H), 7.25 – 7.18 (m, 9H), 7.01 (dd, J = 7.3,
2.2 Hz, 1H), 4.20 (ddd, J = 10.5, 8.4, 5.6 Hz, 4H), 4.11 – 3.97 (m, 4H), 3.51 (dd, J = 9.3,
4.7 Hz, 4H).

177

4.5 Bibliography
1. Frischmann, P. D., Mahata, K. & Würthner, F. Powering the future of molecular
artificial photosynthesis with light-harvesting metallosupramolecular dye
assemblies. Chem. Soc. Rev. 42, 1847–70 (2013).
2. Gust, D., Moore, T. A. & Moore, A. L. Solar fuels via artificial photosynthesis. Acc.
Chem. Res. 42, 1890–1898 (2009).
3. Wasielewski, M. R. Self-Assembly Strategies for Integrating Light Harvesting and
Charge Separation in Artifical Photosynthetic Systems. Acc. Chem. Res. 42, 1910–
1921 (2009).
4. Kodis, G., Liddell, P. A., Moore, A. L., Moore, T. A. & Gust, D. Synthesis and
photochemistry of a carotene–porphyrin–fullerene model photosynthetic reaction
center. J. Phys. Org. Chem. 17, 724–734 (2004).
5. Desimoni, G., Faita, G. & Jørgensen, K. A. C2 -Symmetric Chiral Bis(Oxazoline)
Ligands in Asymmetric Catalysis. Chem. Rev. 106, 3561–3651 (2006).
6. Rechavi, D. & Lemaire, M. Enantioselective Catalysis Using Heterogeneous
Bis(oxazoline) Ligands: Which Factors Influence the Enantioselectivity? Chem.
Rev. 102, 3467–3494 (2002).
7. Chorazy, S., Nakabayashi, K., Ohkoshi, S. I. & Sieklucka, B. Green to red
luminescence switchable by excitation light in cyanido-bridged Tb
III
-W
V

ferromagnet. Chem. Mater. 26, 4072–4075 (2014).
8. Chorazy, S. et al. Multifunctionality in bimetallic LnIII[WV(CN) 8]3- (Ln=Gd,
Nd) coordination helices: Optical activity, luminescence, and magnetic coupling.
Chem. - A Eur. J. 20, 7144–7159 (2014).
9. Kögel, J. F. et al. Heteroleptic [Bis(oxazoline)](dipyrrinato)zinc(II) Complexes:
178

Bright and Circularly Polarized Luminescence from an Originally Achiral
Dipyrrinato Ligand. Angew. Chemie - Int. Ed. 55, 1377–1381 (2016).
10. Trinh, C. et al. Symmetry-breaking charge transfer of visible light absorbing
systems: Zinc dipyrrins. J. Phys. Chem. C 118, 21834–21845 (2014).
11. Atkins, J. M., Moteki, S. A., DiMagno, S. G. & Takacs, J. M. Single enantiomer,
chiral donor-acceptor metal complexes from bisoxazoline pseudoracemates. Org.
Lett. 8, 2759–2762 (2006).
12. Takacs, J. M., Hrvatin, P. M., Atkins, J. M., Reddy, D. S. & Clark, J. L. The
selective formation of neutral, heteroleptic zinc(ii) complexes via self-
discrimination of chiral bisoxazoline racemates and pseudoracemates. New J.
Chem. 29, 263 (2005).
13. Crosignani, S. et al. Chiral amplification in Diels-Alder and 1,3-dipolar
cycloadditions catalyzed by bis(oxazoline)-Zn(II)-based chiral complexes.
Tetrahedron Lett. 40, 7007–7010 (1999).
14. Zhao, C. C. et al. Inner-assembly singlet energy transfer in naphthalenea-
anthracene system linked by 2-ureido-4{1H}-pyrimidinone binding module.
Tetrahedron Lett. 45, 6807–6811 (2004).
15. Zgierski, M. Z. & Lim, E. C. On the mechanism of intramolecular charge transfer
in para-disubstituted diphenylacetylenes containing electron-donating and
electron-accepting groups: Role of πσ * state in electron-transfer dynamics. Chem.
Phys. Lett. 393, 143–149 (2004).
16. Szyszkowska, M., Czaplewski, C. & Wiczk, W. Photophysical and theoretical
studies of diphenylacetylene derivatives with restricted rotation. J. Mol. Struct.
1138, 81–89 (2017).
179

17. Benniston, A. C., Harriman, A. & Rostron, J. P. The effect of solvent polarity on
the photophysical properties of 4-cyano-(4’-methylthio)diphenylacetylene: A
prototypic donor-acceptor system. Phys. Chem. Chem. Phys. 7, 3041–3047
(2005).
18. Telfer, S. G., McLean, T. M. & Waterland, M. R. Exciton coupling in coordination
compounds. Dalt. Trans. 40, 3097–3108 (2011).
19. Saltiel, J. & Kumar, V. K. R. Photophysics of diphenylacetylene: Light from the
‘dark state’. J. Phys. Chem. A 116, 10548–10558 (2012).
20. Turro, N., Ramamurthy, V. & Scaiano, J. C. Modern Molecular Photochemistry of
Organic Molecules. (University Science Books, 2010).
21. Szyszkowska, M., Bylińska, I. & Wiczk, W. Influence of an electron-acceptor
substituent type on the photophysical properties of unsymmetrically substituted
diphenylacetylene. J. Photochem. Photobiol. A Chem. 326, 76–88 (2016).
22. Singh, A., Pati, A. K. & Mishra, A. K. Photoinduced intramolecular charge transfer
in a cross-conjugated push–pull enediyne: implications toward photoreaction.
Phys. Chem. Chem. Phys. 20, 14889–14898 (2018).
23. Hirata, Y., Okada, T. & Nomoto, T. Photoinduced intramolecular charge
separation of p-N , N-dimethylamino-p ’ -cyano-diphenylacetylene in polar
solvents. Chem. Phys. Lett. 278, 133–138 (1997).
24. Wierzbicka, M., Bylińska, I., Sikorski, A., Czaplewski, C. & Wiczk, W.
Experimental and theoretical studies of spectroscopic properties of simple
symmetrically substituted diphenylbuta-1,3-diyne derivatives. Photochem.
Photobiol. Sci. 14, 2251–2260 (2015).
25. Amatatsu, Y. & Hosokawa, M. Theoretical study on the photochemical behavior of
180

diphenylacetylene in the low-lying excited states. J. Phys. Chem. A 108, 10238–
10244 (2004).
26. Menning, S. et al. Bridged tolanes: A twisted tale. J. Org. Chem. 79, 6571–6578
(2014).
27. Herbich, J. & Kapturkiewicz, A. Electronic structure and molecular conformation
in the excited charge transfer singlet states of 9-acridyl and other aryl derivatives
of aromatic amines. J. Am. Chem. Soc. 120, 1014–1029 (1998).
28. Chou, P.-T., Chang, C.-P., Clements, J. H. & Meng-Shin, K. Synthesis and
Spectroscopic Studies of 4-Formyl-4’-N,N-dimethylamino-1,1’-biphenyl: The
Unusual Red Edge Effect and Efficient Laser Generation. J. Fluoresc. 5, 369–375
(1995).
29. Maus, M., Rettig, W., Bonafoux, D. & Lapouyade, R. Photoinduced intramolecular
charge transfer in a series of differently twisted donor-acceptor biphenyls as
revealed by fluorescence. J. Phys. Chem. A 103, 3388–3401 (1999).
30. Sasaki, S., Niko, Y., Klymchenko, A. S. & Konishi, G. I. Design of donor-acceptor
geometry for tuning excited-state polarization: Fluorescence solvatochromism of
push-pull biphenyls with various torsional restrictions on their aryl-aryl bonds.
Tetrahedron 70, 7551–7559 (2014).
31. Sumalekshmy, S. & Gopidas, K. R. Intramolecular charge transfer processes in
donor-acceptor substituted vinyltetrahydropyrenes. Photochem. Photobiol. Sci. 4,
539–546 (2005).
32. Kitagawa, Y., Yachi, R., Nakanishi, T., Fushimi, K. & Hasegawa, Y. J-Type
Heteroexciton Coupling Effect on an Asymmetric Donor-Acceptor-Donor-Type
Fluorophore. J. Phys. Chem. A (2017). doi:10.1021/acs.jpca.7b02783
181

33. Benniston, A. C., Harriman, A. & Rostron, J. P. The effect of solvent polarity on
the photophysical properties of 4-cyano-(4 0 -methylthio)diphenylacetylene: A
prototypic donor–acceptor system. doi:10.1039/b506776f
34. Kim, T. et al. Symmetry-breaking charge transfer in the excited state of directly
linked push-pull porphyrin arrays. Phys. Chem. Chem. Phys. 19, 13970–13977
(2017).
35. Fraind, A. M. & Tovar, J. D. Comparative survey of conducting polymers
containing benzene, naphthalene, and anthracene cores: Interplay of localized
aromaticity and polymer electronic structures. J. Phys. Chem. B 114, 3104–3116
(2010).
36. Fraind, A. M. et al. Charge delocalization through benzene, naphthalene, and
anthracene bridges in π-conjugated oligomers: An experimental and quantum
chemical study. J. Phys. Chem. B 117, 6304–6317 (2013).
37. Wan, Y. et al. Intramolecular charge transfer induced emission from
triphenylamine-o-carborane dyads. RSC Adv. 7, 35543–35548 (2017).
38. Hu, R. et al. Twisted Intramolecular Charge Transfer and {Aggregation-Induced}
Emission of {BODIPY} Derivatives. 113, 15845–15853 (2009).
39. Grabowski, Z., Rotkiewicz, K. & Rettig, W. Structural Changes Accompanying
Intramolecular Electron Transfer: Focus on Twisted Intramolecular Charge-
Transfer States and Structures. Chem. Rev. 103, 3899–4031 (2003).
40. Wang, Z. et al. Insights into the Efficient Intersystem Crossing of Bodipy-
Anthracene Compact Dyads with Steady State and Time-Resolved
Optical/Magnetic Spectroscopies and Observation of the Delayed Fluorescence. J.
Phys. Chem. C 128, 265–274 (2019).
182

41. Scholes, G. D., Ghiggino, K. P., Oliver, A. M. & Paddon-Row, M. N. Intramolecular
electronic energy transfer between rigidly linked naphthalene and anthracene
chromophores. J. Phys. Chem. 97, 11871–11876 (1993).
42. Albelda, M. T. et al. Switching from intramolecular energy transfer to
intramolecular electron transfer by the action of pH and Zn 2 þ co-ordination.
Chem. Phys. Lett. 353, 63–68 (2002).
43. Veldman, D., Meskers, S. C. J. & Janssen, R. A. J. The Energy of Charge-Transfer
States in Electron Donor–Acceptor Blends: Insight into the Energy Losses in
Organic Solar Cells. Adv. Funct. Mater. 19, 1939–1948 (2009).
44. Rhee, H. et al. Intramolecular and intermolecular energy transfers in donor-
acceptor linear porphyrin arrays. J. Chem. Phys. 125, (2006).
45. Katz, C. E. & Aubé, J. Unusual Tethering Effects in the Schmidt Reaction of
Hydroxyalkyl Azides with Ketones: Cation-?? and Steric Stabilization of a
Pseudoaxial Phenyl Group. J. Am. Chem. Soc. 125, 13948–13949 (2003).
46. Korovina, N. V. et al. Singlet Fission in a Covalently Linked Cofacial
Alkynyltetracene Dimer. J. Am. Chem. Soc. 138, 617–627 (2016).
47. Miyake, Y., Nakajima, K. & Nishibayashi, Y. Visible Light-Mediated Oxidative
Decarboxylation of Arylacetic Acids into Benzyl Radicals: Addition to Electron-
Deficient Alkenes by Using Photoredox Catalysts. Chem. Commun. 49, 1–52
(2013).
48. Tu, D. et al. Highly Emissive Organic Single-Molecule White Emitters by
Engineering o-Carborane-Based Luminophores. Angew. Chemie - Int. Ed. 298,
11370–11374 (2017).
49. Walli, A., Dechert, S. & Meyer, F. Tautomerism in bis(oxazoline)s. European J.
183

Org. Chem. 7044–7049 (2013). doi:10.1002/ejoc.201301282

Abstract (if available)

Abstract The work herein incorporates multiple approaches to developing alternative energy technologies through energy and electron management. The first solution works to optimizing an organic photovoltaic (OPV) bilayer device structure by incorporating an electron transport layer of zinc oxide nanoparticles in an inverted device structure. This was found to have superior device performance than its conventional structure counterpart. Another approach is development of a novel supramolecular light harvesting unit (LHU) used in the hydrogen evolution reaction (HER) to achieve solar water splitting. The LHU was designed as a microparticulate OPV, to be used in conjunction with an oxygen or hydrogen evolution electrode catalyst. The LHU harvests light with inexpensive materials to improve energy and electron transfer to a HER catalyst. The final solution takes a molecular approach towards artificial photosynthesis, by using chirality to drive assembly of donor-acceptor molecules. Intramolecular charge transfer states were found to be an inherent property of the dyads synthesized, allowing for intramolecular electron transfer. These are ideal candidates as light harvesting complexes in artificial photosynthesis.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Energy management in organic photovoltaics

PDF

Utilizing n-heterocyclic chromophores for solar energy harvesting.

PDF

Molecular and polymeric donors for bulk heterojunction organic solar cells

PDF

Improving the sustainability of conjugated polymer synthesis via direct arylation polymerization

PDF

Hydrogen peroxide vapor for small satellite propulsion

PDF

Chemical vapor deposition of graphene: synthesis, characterization, and applications

PDF

Singlet fission in covalently-linked tetracenes

PDF

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

PDF

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

PDF

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

PDF

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

PDF

Electrochemical pathways for sustainable energy storage and energy conversion

PDF

Molecular design strategies for blue organic light emitting diodes

PDF

Fabrication, deposition, and characterization of size-selected metal nanoclusters with a magnetron sputtering gas aggregation source

PDF

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

PDF

Solution processing of chalcogenide functional materials using thiol–amine “alkahest” solvent systems

Asset Metadata

Creator Wilson, Rebecca Jean (author)

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

School Viterbi School of Engineering

Degree Doctor of Philosophy

Degree Program Chemical Engineering

Publication Date 08/14/2019

Defense Date 02/25/2019

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

Tag electron transfer,energy transfer,hydrogen evolution reaction,light absorption,OAI-PMH Harvest,organic photovoltaic,solar water splitting,zeolite

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Thompson, Mark (committee chair), Armani, Andrea (committee member), Thompson, Barry (committee member)

Creator Email beckywilson22@gmail.com,rebeccjw@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-213958

Unique identifier UC11663210

Identifier etd-WilsonRebe-7780.pdf (filename),usctheses-c89-213958 (legacy record id)

Legacy Identifier etd-WilsonRebe-7780.pdf

Dmrecord 213958

Document Type Dissertation

Format application/pdf (imt)

Rights Wilson, Rebecca Jean

Type texts

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

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA