Controlling the Physical Properties of Organic ... · Controlling the Physical Properties of Organic Semiconductors through Siloxane Chemistry and ... Physical Properties of Organic

Controlling the Physical Properties of Organic Semiconductors through Siloxane Chemistry and other

Organic Electronic Materials

by

Brett A. Kamino

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy

Chemical Engineering and Applied Chemistry University of Toronto

© Copyright by Brett A. Kamino 2013

ii

Controlling the Physical Properties of Organic Semiconductors

through Siloxane Chemistry and other Organic Electronic

Materials

Brett A. Kamino

Doctor of Philosophy

Department of Chemical Engineering and Applied Chemistry University of Toronto

2013

Abstract

Triarylamine type materials with vastly altered physical properties are synthesized by

hybridizing organic semiconducting structures with silicone and siloxane groups. By altering the

silicon content of these materials, we can tune their physical composition from free flowing

liquids, to amorphous glasses, to cross-linked films. Much of this modification is enabled by the

unique use of a metal-free Si-H activation chemistry; the Piers-Rubinsztajn reaction. This

chemistry is demonstrated to be a general and rapid way to build up hybrid semiconducting

structures. Key to the utility of these materials in electronic devices, it is shown that

hybridization with silicon groups has a negligible effect on the useful electrochemical properties

of the base materials. Building on this, it is shown that charge carrier mobility through a

prototypical liquid organic semiconductor is similar to the base materials and transport is

described by existing dispersive transport theories. Finally, two side projects in the area of

organic electronics are discussed. New phthalonitrile based fluorophores are characterized and

their utility as deep-blue emitting dopants in organic light emitting diodes is demonstrated. These

same π-extended phthalonitriles can also be used as precursors to new red-shifted BsubPcs which

display exceptional electrochemical stability and tuning.

iii

Acknowledgments

I would like to thank all of my friends and family for their constant support and understanding as

I embarked on a longer than planned educational adventure. In particular, I need to thank my

Mother and Father for pretending to understand my project and legitimately being proud of my

accomplishments. I thank all of my friends outside of grad school and especially Lana for

helping to keep me sane throughout the years.

I would also like to thank all of the members of the Bender lab past and present for being such a

warm group of friends and colleagues. The friendly atmosphere maintained in the laboratory and

at the pub after the laboratory has made my PhD experience all the better and easier to deal with

over the years. As well, I’d like to thank everyone for providing a patient ear to any frustrations

I’ve had over the years.

Lastly, I’d like thank Professor Bender for originally taking me into the group and giving me a

chance to earn my PhD. You’ve been a patient supervisor and have enabled to have a great deal

of freedom to explore my own scientific path. Finally, I’d to thank you for putting up with my

constant rebellion and for your belief in me.

iv

Table of Contents

Acknowledgments .......................................................................................................................... iii

Table of Contents ........................................................................................................................... iv

List of Tables ................................................................................................................................. ix

List of Figures and Schemes ........................................................................................................... x

Chapter 1: Introduction ............................................................................................................. 1

1.1 Organic Electronics Background ........................................................................................ 1

1.2 Thesis Overview ................................................................................................................. 3

1.3 Thesis Statement ................................................................................................................. 4

1.4 Background on Triarylamines ............................................................................................. 4

1.5 Engineering the Physical Properties of Organic Semiconductors ...................................... 6

1.6 References ........................................................................................................................... 9

Chapter 2: The Use of Siloxanes, Silsesquioxanes, and Silicones in Organic Semiconducting Materials ........................................................................................................ 11

2.1 Executive Summary .......................................................................................................... 11

2.2 Statement of Contributions ............................................................................................... 11

2.3 Paper ................................................................................................................................. 11

2.3.1 Abstract ................................................................................................................. 11

2.3.2 Introduction ........................................................................................................... 12

2.3.3 Chemistries for Silicone and Siloxane Incorporation ........................................... 14

2.3.4 Examples of Hybrid Materials .............................................................................. 18

2.3.5 Conclusions and Outlook ...................................................................................... 32

2.4 References ......................................................................................................................... 32

Chapter 3: Effect of Triarylamine Structure on the Photoinduced Electron Transfer to Boron Subphthalocyanine ........................................................................................................ 37


v


3.3 Paper ................................................................................................................................. 38

3.3.1 Abstract ................................................................................................................. 38

3.3.2 Introduction ........................................................................................................... 38

3.3.3 Results and Discussion ......................................................................................... 40

3.3.4 Conclusions ........................................................................................................... 50

3.3.5 References ............................................................................................................. 50

3.3.6 51

Chapter 4: Controlling the Physical and Electrochemical Properties of Arylamines Through the Use of Simple Silyl Ethers: Liquid, Waxy and Glassy Arylamines .................... 52



4.3 Paper ................................................................................................................................. 53

4.3.1 Abstract ................................................................................................................. 53

4.3.2 Introduction ........................................................................................................... 53


4.3.4 Conclusions ........................................................................................................... 65

4.3.5 References ............................................................................................................. 66

Chapter 5: Siloxane-Triarylamine Hybrids: Discrete Room Temperature Liquid Triarylamines via the Piers-Rubinsztajn Reaction ................................................................... 68



5.3 Paper ................................................................................................................................. 69

5.3.1 Abstract ................................................................................................................. 69

5.3.2 Body ...................................................................................................................... 69

5.3.3 References ............................................................................................................. 74

vi

Chapter 6: Liquid Triarylamines: The Scope and Limitations of Piers-Rubsinsztajn Conditions for Obtaining Triarylamine-Siloxane Hybrid Materials ........................................ 76



6.3 Paper ................................................................................................................................. 77

6.3.1 Abstract ................................................................................................................. 77

6.3.2 Introduction ........................................................................................................... 77


6.3.4 Conclusions ........................................................................................................... 91

6.3.5 References ............................................................................................................. 93

Chapter 7: Hole Mobility of a Liquid Organic Semiconductor .............................................. 95



7.3 Paper ................................................................................................................................. 96

7.3.1 Abstract ................................................................................................................. 96

7.3.2 Body ...................................................................................................................... 96

7.3.3 References ........................................................................................................... 104

Chapter 8: Crosslinked Triarylamine-Siloxane Films using Piers-Rubinsztajn Chemistry . 106

8.1 Executive Summary ........................................................................................................ 106

8.2 Statement of Contributions ............................................................................................. 106

8.3 Paper Draft ...................................................................................................................... 107

8.3.1 Abstract ............................................................................................................... 107

8.3.2 Introduction ......................................................................................................... 107

8.3.3 Results and Discussion ....................................................................................... 109

8.3.4 Conclusions ......................................................................................................... 116

8.3.5 References ........................................................................................................... 117

vii

Personal Interest Projects 1: Design of Deep-Blue Emitting Materials for OLEDs ................... 120

Chapter 9: ............................................................................................................................... 120



9.6 Paper ............................................................................................................................... 121

9.6.1 Abstract ............................................................................................................... 121

9.6.2 Introduction ......................................................................................................... 121

9.6.3 Results and Discussion ....................................................................................... 122

9.6.4 Conclusions ......................................................................................................... 131

9.7 References ....................................................................................................................... 131

Chapter 10: Personal Interest Projects 2: Colour Tuning Boron Subphthalocyanine ............ 134



10.3 Paper ............................................................................................................................... 134

10.3.1 Body .................................................................................................................... 134

10.3.2 References ........................................................................................................... 144

Chapter 11: Concluding Remarks and Future Work ................................................................... 147

Chapter 11: ............................................................................................................................. 147

11.1 Summary ......................................................................................................................... 147

11.2 Future Directions ............................................................................................................ 150

11.3 References ....................................................................................................................... 152

Appendices .................................................................................................................................. 153

Chapter 12: ............................................................................................................................. 153

12.1 Additional Information for Chapter 3 ............................................................................. 153

12.1.1 Experimental Information ................................................................................... 153

12.1.2 Supplemental Information of Merit .................................................................... 157

viii







12.4.2 Supplemental Information of Merit .................................................................... 184



12.5.2 Supplemental Information for Chapter 7 ............................................................ 191


12.7 Additional Information for Chapter 10 ........................................................................... 197

12.7.1 11.7.1 Experimental Information ........................................................................ 197

12.8 Appendices References for Chapter 10 ........................................................................... 199

12.8.1 General Information ............................................................................................ 199

12.8.2 Synthetic Details and Compounds Characterization ........................................... 200

12.8.3 NMR Study of Phthalonitrile 10-2 ...................................................................... 202

12.8.4 UV-Vis and PL Plots .......................................................................................... 205

12.9 DFT Calculated Molecular Orbitals ................................................................................ 209

12.10 Cyclic Voltammetry ............................................................................................ 214

ix

List of Tables

Table 3-1: Electrochemical oxidation potentials, fluorescence quenching efficiency, and free energy change upon photoinduced electron transfer reaction with 3,4-DMPhO-BsubPc for triarylamines (3-1a-i, 3-2a-c, 3-5b and 3-6b). 41

Table 4-1. Comparison of the physical and electrical properties of compounds 4-4a, 5 and 6 with previously reported analogous compounds. 54

Table 4-2. Comparison of the physical and electrical properties of compounds 4-4a-d including hydrolytic stability. 57

Table 4-3. Characterization data for multi-nitrogen centred triarylamine series, compounds 4-7, 4-8a-b, 4-9 and 4-11 (see Scheme 4-1 for chemical structures of TIPS and TBDPS). 61

Table 5-1: Thermal and electrochemical information for precursor triarylamine compounds 5-1a-c and their siloxane functionalized counterparts, 5-2a-c. 69

Table 6-1: DSC and CV results from silicone-hybridized triarylamines. 79

Table 8-1: Collected experimental information for different formulation conditions. 108

Table 9-1. Photophysical, electrochemical and thermal properties of compounds 9-1 and 9-2 120

Table 9-2. Performance of OLEDs incorporating 9-1 and 9-2 as dopants 127

Table 10-1: Calculated and experimentally determined properties of compounds F5BsubPc, 10-3a-b and 10-4a-c. 138

Figure 10-3: Cyclic voltammogram of BsubPc compound 10-4b in DCM with 0.1M tetrabutylammonium perchlorate and decamethylferrocene as an internal standard. 139

x

List of Figures and Schemes

Figure 1-1: Examples of common triarylamine materials utilized in organic electronics. Semiconducting…. 5 Figure 2-1: The molecular fragments of a silicone/siloxane organic electronic hybrid material (using a triarylamine as a representative example)… 11 Figure 2-2. Example of siloxane containing triarylamie applied in a xerographic… 22

Figure 2-3: Example of siloxane containing material used as an electron … 23

Figure 2-4. Structure of common silsesquioxane-T8 synthetic precursors. 24

Figure 2-5: Examples of several POSS-T8 derivatives functionalized with organic semiconducting groups. 25 Figure 3-1: Structures of 3,4-DMPhO-BsubPc and the triarylamines used in this study (containing either one (3-1a-i) or two nitrogen centers (3-2a-c)). 37 Figure 3-2: Size exclusions chromatograms of the triarylamine dendrimers (3-5a and 3-6a) and… 40 Figure 3-3: Solution electrochemistry of triarylamine dendrimers (3-5b, 3-6b) and a representative single triarylamine (3-1a). 43 Figure 3-4: Experimentally determined Stern-Volmer constants (K) plotted against half-wave oxidation potentials (E1/2,ox) of the triarylamine donor. 44 Figure 3-5: Experimentally determined Stern Volmer constant (K) plotted against the free energy change estimated by a modified Rehm-Weller equation (Eq. 3-2). 46 Figure 4-1 Two previously reported liquid arylamines. 50 Figure 6-1: Steady-state solution absorbance spectroscopy in toluene of a) compound 1 b) compound 6-4d with 0, 0.25, 0.5, and 1 equivalents of BCF. 85 Figure 7-1: Example photogenerated transients through (a) 2TIPS in a poly(styrene) matrix (9 um) and (b) neat 2TIPS (50 um) 94 Figure 7-2: Field dependent hole mobility as a function of temperature for 2TIPS in a (a) polystyrene matrix and (b) as a neat liquid. 96 Figure 7-3: Temperature dependence on hole mobility for 2TIPS doped in polystyrene (50 wt%) at 555 kV/cm and neat 2TIPS at 100 kV/cm. 99 Figure 8-1: IR spectra of QM4 and films F, E, and D (top to bottom). Film D prepared in a

matrix of KBr and the remainder studied by ATR. 109

Figure 8-2: (a) UV-Vis absorption (black) and photoluminescence spectra of

compound 4 in a THF solution (red) and a neat film (blue). (b) Photoluminescence

spectra of films A, D, and F on glass. 111

Figure 8-3: Electrochemistry with decamethylferrocene internal reference. 112

xi

Figure 9-1. Geometry optimized structure for (a) compound 9-1 (top) and (b) compound 2 (bottom) and their predicted HOMO and LUMO distributions. 119

Figure 9-2. Normalized absorption (black line) and emission spectra. 121

Figure 9-3a. (i) Electroluminescent spectra of compound 9-1 125

Figure 9-3b. (i) Electroluminescent spectra of compound 9-2 126

Figure 10-1: Thermal ellipsoid plot of (a) 10-3b·(THF)2 (CCDC deposition 910746) and (b)

10-4a·(CHCl3)2 (CCDC deposition 910747). 133

Figure 10-2: Normalized UV-Vis absorbance spectra of (a) F5BsubPc, 10-4a and 10-3b and (b) 10-4a-c. 136

Scheme 2-1: Summary of some common coupling techniques to join organic semiconductors with silicone/siloxane components 13 Scheme 2-2: Summary of synthetic strategies towards side-chain polymeric organic semiconductors. 16 Scheme 2-3. Cross section of an OLED device and examples of siloxane containing materials applied at either the ITO/hole transport layer interface… 19 Scheme 3-1: Synthetic pathway towards triarylamine dendrimers (3-5b and 3-6b). Conditions… 38 Scheme 4-1: Synthesis of triarylamines containing silyl ethers 53 Scheme 4-2 Synthesis of silyl ether containing arylamines with multiple nitrogen centres. 59 Scheme 5-1: Several synthetic transformations accessible by using the Piers-Rubinsztajn reaction. 66 Scheme 5-2: Synthesis of single nitrogen centered triarylamines 5-2a-c 68 Scheme 6-1: Synthesis of siloxane functionalized arylamines. 76 Scheme 6-2: Synthesis of spiro core triarylamine 6-9 81 Scheme 6-3: Synthesis of triarylamine 6-11 81 Scheme 6-2: Synthesis of siloxane functionalized carbazoles. 83 Scheme 8-1: Synthesis of an arylamine monomer for ring-opening under Piers-Rubinsztajn conditions. 106 Scheme 8-2: Reagents used to achieve functional cross-linked films using Piers-Rubinsztajn chemistry. 107 Scheme 9-1. Synthetic pathways towards phthalonitriles 9-1 and 9-2. 118

Scheme 10-1: Synthesis of π extended BsubPcs 10-3a-b and 10-4a-c and their precursor phthalonitriles. 132

xii

List of Abbreviations

OLED – organic Light Emitting Diode

OFET – organic Field-Effect Transistor

OPV – organic photovoltaic

DSC – differential Scanning Calorimetry

PDMS – poly(dimethylsiloxane)

BCF – tris(pentafluorophenyl)borane

PVK – poly(vinylcarbazole)

PSX – See Scheme 2-2

Tg – glass Transition Temperature

ITO – indium Tin Oxide

P3HT – poly(3-hexylthiophene)

PPV – Poly(phenylenevinylene)

POSS – silsesquioxane

AlQ3 - tris(8-hydroxyquinolinato)aluminium

PET – photoinduced electron transfer

BsubPc – boron subphthalocyanine

SEC – size exclusion chromatography

MM – hexamethyldisiloxane

TMEPA – tris(4-methoxyethoxyphenyl)amine

DSSC – dye sensitized solar cell

TIPS – triisopropylsilane

TBDMS – tert-butyldimethylsilane

THDMS – tert-hexyldimethylsilane

xiii

DPTBS – diphenyl-t-butylsilane

CV – cyclic voltammetry

MDHM – 1,1,1,3,5,5,5-heptamethyltrisiloxane

MMH – 1,1,3,3,3-heptamethyldisiloxane

HRMS – high resolution mass spectroscopy

LOS – liquid organic semiconductor

2TIPS - 3,4-dimethyl-N,N-bis(4-((triisopropylsilyl)oxy)phenyl)aninline (See Figure 7-11)

TPD - N,N`-diphenyl-N,N`-bis(3-methylphenyl)-(1,1`-biphenyl)-4,4`-diamine

TAPC - 1,1-bis(di-4-tolylaminophenyl)cyclohexane

QM4 - tetrakis(dimethylsilyl)orthosilicate

TGA - thermal gravimetric analysis

CIE - International Commission on Illumination

RGB – Red, Green, Blue

NTSC – National Television Standards Council

CBP - 4,4`-di(9-carbazolyl)-biphenyl

TPBi - 2,2`,2``-(1,3,5-benzinetriyl)tris(1-phenyl-1-H-benzimidazole

1

1

Chapter 1: Introduction

1.1 Organic Electronics Background

Organic electronic devices are a diverse group of technologies which use organic

semiconducting materials as a primary functional element. These devices include organic light

emitting diodes (OLEDs),1 organic field-effect transistors (OFETs),2 organic xerographic

photoreceptors,3 organic photovoltaics (OPVs),4 and many other devices.

All of these devices are based on semiconducting materials made up of organic small molecules,

oligomers, or polymers.5 This is in contrast to traditionally inorganic semiconductors based on

semimetals such as silicon. Much like silicon, organic semiconductors can be designed to

selectively move positive charge (holes) or negative charges (electrons). As well, they can be

designed to interact with light in a number of useful ways. These include the absorption of light

for exciton generation or quenching of excitons to generate light. By designing appropriate

materials and combining them with complementary materials in a full device structure,

remarkable electronic functions can be realized.

Such materials are gaining a great deal of attention due to several advantages inherent to using

organic semiconductors over their inorganic counterparts. Using organic semiconductors in

electronic devices has a number of potential advantages for widespread commercialization.

Among these advantages is the ability to process devices at substantially lower temperatures

using solution casting techniques. This has led to prototype roll-to-roll printing processes to be

investigated by a number of different groups.6 As well, organic semiconductors have

significantly lower purity requirements as compared to inorganic semiconductors in most cases.7

All of these aspects can lower the overall cost of processing organic electronic devices as

compared to those made with inorganic materials. This potentially lower cost of production has

produced a great deal of interest from both academic labs and industrial companies.

Beyond cost, organic material have another significant advantage over inorganic materials.

Through a diverse synthetic tool box and a healthy imagination, an almost infinite number of

organic materials can be produced to achieve an equally diverse set of electronic, optical, and

2

2

physical properties. This is something that is not possible with inorganic semiconductors which

are limited greatly by the real-estate of the periodic table.

Perhaps because of this immense potential for material optimization, a number of these device

types have already reached maturity and have achieved commercialization. The oldest example

of this is xerographic photoreceptors which have been the standard component in most

photocopying devices for well over a decade. More recently, we have seen the

commercialization of OLED based displays for televisions and personal electronics such as

smartphones.8 At the time of this writing, a number of industrial companies are in the process of

developing prototype devices for both OPVs9 and OFETs10 which are on the cusp of widespread

deployment.

Despite this success, a number of significant gaps remain in realizing the ultimate promise of

organic semiconductors for inexpensive and flexible electronic devices. Among these gaps,

many of them are on the processing side of organic semiconductors.

Organic electronics have long been hyped to allow for highly flexible and mechanically robust

devices build upon flexible plastic substrates. Indeed, this has been shown in a number of

prototype devices and even elastomeric devices which can be squished and stretched have been

demonstrated.11 While the benefits to this kind of device are obvious, the reality is that upon

repeated mechanical stresses, many of these devices rapidly degrade.12 This is due to a large

number of reasons including delamination and cracking of the brittle crystalline layers used in

these devices. While this drawback can be corrected by the careful design of specific device

architectures on extremely thin substrates,13 such substrates are not likely to be widely applicable

due to their poor mechanical properties. Given this interest in creating mechanically robust

flexible and stretchable devices, it seems peculiar that no one is interested in producing

semiconducting materials that are inherently flexible and stretchable.

Another unfulfilled goal of organic electronics has been the development of high-speed printing

processes such as roll-to-roll solution printing processes or ink-jet printing on an industrial scale.

These processes are envisioned to be similar to conventional printing where devices can be

rapidly constructed by printing successive layers of organic semiconductors from solution. This

goal would preclude expensive and slow vacuum processing and greatly help to bring down the

3

3

overall cost of organic electronic devices. Again, realization of this goal is prevented by the

current limitations of organic semiconductors and processing properties. Many organic

semiconductors are very poorly soluble in organic solvents and require the use of highly toxic

and environmentally damaging chlorinated solvents.14 This poses a number of downstream waste

issues as well as corrosion issues for a roll-to-roll printing apparatus. A second issue with this

goal is the difficulty of producing multi-layered devices. Most organic electronics devices

require successive layers of different materials for optimized function. However, printing a

second organic layer on top of a first layer will dissolve and mix the two layers together. This

can be avoided with orthogonal solvent processing,15 but few materials are amenable to this

strategy and it places severe restrictions on the materials chosen. To solve some of these

deficiencies, the development of high soluble materials with differential solubility would be

ideal.

I believe that further advances in the field of organic electronics will require the development of

organic semiconductors with highly tailored physical properties to help overcome some of

remaining challenges for organic electronics in the interest of industrial scale processibility.

However, inclusion of these specific physical properties must not occur at the cost of the

electronic and photonic properties that these materials make these materials useful. This

particular point posses a unique challenge in the design of new organic semiconducting materials

which will require new synthetic strategies and chemical processes.

1.2 Thesis Overview

This thesis is a compilation of published or to-be-submitted papers produced over the course of

my PhD thesis. Before each chapter, there is an executive summary detailing how the work in

that chapter relates to my thesis statement. After that, there is a statement of contributions

outlining what each of the authors on the paper worked on. For each chapter and publication, I

have done a large majority of the experimental work and the bulk of the writing of the paper

itself.

Included in this thesis are two chapters which are not related to my primary thesis statement

(Chapters 9 and 10). These chapters represent side projects that I initiated during my PhD thesis.

Both projects are related to the broader overall goals of the Bender Laboratory and the field of

4

4

Organic Electronics. Because these projects were completed during the course of my thesis, I

decided to include them as two ‘Personal Interest Projects’.

1.3 Thesis Statement

Selective incorporation of silicone and siloxane components into nitrogen-based hole

transporting compounds can be an effective tool to control the physical properties of organic

semiconductors. This can be done to produce materials which are morphologically stable glasses,

free-flowing liquids, and cross-linked films. For many of these materials, a novel use of the

Piers-Rubinsztajn reaction is demonstrated to be a facile and convenient method to hybridize

siloxane components with triarylamine components. We show that the introduction of siloxane

or silicone groups has a negligible effect on the useful electrochemical properties of the

materials. As well, we demonstrate that there is no significant difference in the charge carrier

mobility of a model liquid material as compared to the analogous solid compound.

1.4 Background on Triarylamines

Parts of this section have been taken and paraphrased from a paper published and discussed in

Chapter 3 (Brett A. Kamino, Graham E. Morse, Timothy P. Bender, 2011, Journal of Physical

Chemistry C, 115 (42), 20716-20723).

Throughout this thesis, we have chosen to focus our efforts on a single family of organic

semiconductors as a model group of compounds: triarylamines. Single and multi-nitrogen

centered triarylamines are an important class of functional materials in the area of organic

electronics. Owing to their well-behaved chemical and electrochemical oxidation, this class of

organic semiconductors represents one of the most frequently studied electron-donating materials

(hole-transporting materials) in the field.5 This makes them an attractive target for developing

synthetic methodologies to alter the physical properties of organic semiconductors. Several

popular examples of this motif are illustrated below in Figure 1-1.

5

5

N N N N

N N NN

OLED Materials

Xerography Materials

Figure 1-1: Examples of common triarylamine materials utilized in organic electronics

Beyond their basic ability to function as stable and reversible electron donors, the adoption and

study of these materials is aided by the demonstrated ability to modify their electronic and

physical properties over a wide range. Fine control over their oxidation potentials16 and access to

relatively stable polycations can be achieved by the appropriate use of electron-withdrawing/-

donating groups and by the construction of large molecules containing multiple conjugated

triarylamine centers. 1 †,17 Their physical properties can ranges between crystalline solids18 and

morphologically stable glasses.19 Because of their highly tunable properties, triarylamines have

become standard materials in some organic electronic devices such as xerographic

photoreceptors, light-emitting diodes, field effect transistors, bulk heterojunction solar cells,20

and solid-state dye-sensitized solar cells.21 As well, triarylamine moieties have been incorporated

into the molecular structure of light-absorbing oligomers, polymers,22 and photosensitizers.23

† We will use the term triarylamines to refer to all triarylamine structures regardless of number of nitrogen centers

and will include carbazole structures

6

6

1.5 Engineering the Physical Properties of Organic Semiconductors

One of the primary goals of this thesis was designing and synthesizing materials that were

liquids or elastomers at room temperature. We will define such compositions as soft materials.

Since soft organic semiconductors are not typical organic electronic materials, it is worth

exploring some of the design elements that needed to successfully design soft material

properties.

Because the design constraints for producing soft materials is quite different than that of typical

organic electronic materials, it is instructive to first discuss more typical materials as a point of

contrast. Surveying the organic electronic literature will reveal a vast array of different

compounds with different functions. These compounds can be roughly divided into two groups:

glass forming compounds and crystalline materials.

Crystalline materials are organic molecules whose solid state arrangement is highly ordered and

dominated by strong intermolecular interactions. These materials will typically form

polycrystalline films upon deposition from solution or vapour. Occasionally, careful processing

and favourable crystal packing can lead these kinds of materials to form single crystals.

However, this is often the exception. As well, the packing arrangement and degree of

crystallinity in these materials can have a profound effect on the final properties of the film and

performance of the completed device. Because of this, controlling intermolecular interactions is

key to optimizing the performance of many of these materials. In the field of organic electronics,

the most dominant intermolecular interaction is π-π stacking between adjacent molecules. This

should come as no surprise given that basically all organic electronic materials contained

extended-π systems in some form or another.

The second category of organic electronic materials, glass-forming compounds, has a very

different set of design principals compared to crystalline materials. Glass-forming compounds

are designed so that they form completely amorphous and homogeneous solid-state films upon

solution or vapour deposition. Such materials are ideal for applications where relatively low

charge-carrier mobilities are required and crystal defects can negatively affect device

performance. The primary application for such materials is for organic light-emitting diodes,

7

7

although xerographic photoreceptors often use amorphous materials as well. These materials

benefit greatly from a processing point of view because they are homogeneous. Therefore, there

are no issues with controlling crystal-grain size and orientation.

In designing these compounds, there are two specific design criteria which must be followed.

The first is designing a material without the ability to undergo strong intermolecular interactions

in the solid-state. And the second is designing a material with a very high glass transition

temperature to ensure morphological stability over time.

The first of these is critical to ensure that thin-film deposition will result in an amorphous

material instead of an ordered solid. As such, this instantly rules out compounds which are

highly planar and that contain strongly polar functionalities. Compounds with these elements

require significant structural modification in order to form amorphous films. Several examples

are discussed in chapter 2. Unfortunately, it is extremely difficult to design effective organic

electronic materials that do not contain many planar sections. Extended π-conjugation is one of

the defining characteristics of organic semiconductors and is often required to produce materials

with usable redox potentials. Because of this, the effective design of amorphous materials

requires a more insightful approach. These approaches all revolve around preventing π-π

stacking through increased steric hindrance and by lower overall symmetry of the molecule. Both

of these facets frequently involve producing molecules which extend into three dimensional-

spaces. Triarylamines are an excellent example of a material which can form glassy films. While

a simple triarylamine does not form glassy films well, larger derivatives such as those shown in

Figure 1-1 do. One of the key elements about triarylamines which help with forming amorphous

films is their propeller shape. Each phenyl ring around the nitrogen core is slightly twisted and

extends the molecule into three-dimensional space. Having several of these groups together on

the same molecule has been a successful strategy to produce glassy films.

The second design criterion for good glass forming compounds is to ensure a high glass

transition temperature for improved morphological stability. The glass transition temperature of a

material is defined as the temperature where a material transitions between being an amorphous

glass and a melt. This idea has strong roots in the polymer sciences and is most easily

characterized using differential scanning calorimetry (DSC). In a typical DSC experiment, the

glass transition temperature can be detected as a strong change in the slope of the heat flow. The

8

8

reason why the glass transition temperature is so critical for an amorphous material is because a

glassy-state is never the most thermodynamically stable state for a material. A crystalline-

compound takes advantage of intermolecular interactions which can energetically stabilize a

material in the solid-state. Therefore, glassy materials are only kinetically stable because they are

‘frozen’ in their particular arrangement. This energetic situation becomes a problem when glassy

semiconductors experience heating near or above their glass transition temperature. When this

happens, the material can rearrange to a more thermodynamically favourable crystalline state.

This will change the optoelectronic properties of the film. Therefore, a morphologically stable

glassy organic semiconductor requires a glass transition temperature that is much higher than the

typical operating temperatures of the intended devices. This is typically achieved by increasing

molecular size and by ensuring a very rigid structure. Larger molecules tend to have high glass

transition temperatures because of the increase thermal energy required for them to form a melt.

Rigid molecular structures similarly require more thermal energy to allow for molecular rotation.

For the design of soft organic semiconducting materials, we require materials with glass

transition temperatures below the operating and processing temperatures of the material but we

cannot allow the material to crystallize in the melt-state. These design criteria are quite difficult

and quite different from those of the previous sets of materials. However, from the previous

discussions we can discuss how one may achieve these properties. The first step is to prevent

crystallization of the molecule. This can be done by either designing a molecule with very weak

intermolecular interactions or by providing significant steric repulsion to prevent crystallization.

The second and, perhaps more important step, is to lower the glass transition temperature of the

material. This can be done by using very flexible elements in the molecule; molecular fragments

with very low conformational energy barriers.

9

9

1.6 References

1. a) Grimsdale, A.C.; Chan, K.L.; Martin, R.E.; Jokisz, P.G.; Holmes, A.B. Chem. Rev.

2009, 109, 987-1091., b)Xiao, L.; Chen, Z.; Qu, B.; Luo, J.; Kong, S.; Gong, Q.;

Kido, J. Adv. Mater. 2011, 23, 926-952.

2. Allard, S.; Forster, M.; Souharce, B.; Thiem, H.; Scherf, U. Angew. Chemie. Inter. Ed.

2008, 47, 4070-4098.

3. P. M. Borsenberger and D. S. Weiss, Organic Photoreceptors for Xerography, Marcel

Dekker Inc., New York, 1998.

4. Kippelen, B.; Brédas, J.-L. Energy Environ. Sci. 2009, 2, 251-261.

5. Shirota, Y.; Kageyama, H. Chem. Rev. 2007, 107, 953–1010.

6. Søndergaard, R.R.; Hösel, M.; Krebs, F.C. J. Polym. Sci. B.2013, 51 (1), 16-34.

7. Karl, N. Mol. Cryst. Liq. Cryst. 1989, 171, 157-177.

8. Rajeswaran, G.; Itoh, M.; Boroson, M.; Barry, S.; Hatwar, T.K.; Kahen, K.B.; Yoneda,

K.; Yokoyama, R.; Yamada, T.; Komiya, N.; Kanno, H.; Takahashi, H. SID

Symposium Digest of Technical Papers 2000, 31(1), 974-977.

9. Yue, D.; Khatav, P.; You, F.; Darling, S.B. Energy Environ. Sci. 2011, 4, 1434.

10. Bobbert, P.A.; Sharma, A.; Mathijssen, S.G.J.; Kemerink, M.; de Leeuw, D.M. Adv.

Mater. 2012, 24, 1146-1158.

11. Lipomi, D.J.; Chong, H.; Vosgueritchian, M.; Mei, J.; Bao, Z. Sol. Energy Mater. Sol.

Cells, 2012, 107, 355-365.

12. Sokolov, A.N.; Cao, Y.; Johnson, O.B.; Bao, Z. Adv. Mater. 2012, 22, 175-183.

13. Yi, H.T.; Payne, M.M.; Anthony, J.E.; Podzorov, V. Nature Comm. 2013, 3, ASAP.,

Kaltenbrunner, M.; White, M.S.; Glowacki, E.D.; Sekitani, T.; Someya, T.; Sariciftci,

N.S.; Bauer, S. Nat. Comm. 2012, 3, 770

10

10

14. Galagan, Y.; Vries, I.G.; Langen, A.P.; Andriessen, R.; Verhees, W.J.H.; Veenstra, S.C.;

Kroon, J.M. Chemical Engineering and Processing 2011, 50, 454-461.

15. Ahmed, E.; Earmme, T.; Jenekhe, S.A. Adv. Func. Mater. 2011, 21, 3889-3899. Tung,

C.V.; Kim, J.; Cote, L.J.; Huang, J. J. Am. Chem. Soc. 2011, 133, 9262-9265.

16. (a) Bender, T. P.; Graham, J. F.; Duff, J. M. Chem. Mater. 2001, 13, 4105–4111. (b)

Amthor, S.; Noller, B.; Lambert, C. Chem. Phys. 2005, 316, 141–152.

17. Thelakkat, M. Macromol. Mater. Eng. 2002, 287, 442–461.

18. Song, Y.; Di, C.; Yang, X.; Li, S.; Xu, W.; Liu, Y.; Yang, L.; Shuai, Z.; Zhang, D.; Zhu,

D. J. Am. Chem. Soc. 2006, 128, 15940–15941.

19. Gagnon, E.; Maris, T.; Wuest, J. D. Org. Lett. 2010, 12, 404–407.

20. Sommer, M.; Lindner, S. M.; Thelakkat Adv. Funct. Mater. 2007, 17, 1493–1500.

21. Li, B.; Wang, L.; Kang, B.; Wang, P.; Qiu, Y. Sol. Energ. Mater. Sol. Cells 2006, 90 (5),

549–573.

22. Brabec, C.; Dyakonov, V.; Sherf, U. Organic Photovoltaics: Materials, Device Physics,

and Manufacturing Technologies; Wiley Verlag GmbH: Weinheim, 2008.

23. Ning, Z.; Tian, H. Chem. Commun. 2009, 5484–5495.

11

11

Chapter 2: The Use of Siloxanes, Silsesquioxanes, and Silicones in Organic Semiconducting Materials

2.1 Executive Summary

This article is a tutorial review that was submitted to Chemical Society Reviews. This paper

covers how non-conductive silicon elements can be used to optimize the physical properties of

organic semiconductors. More importantly, this also places the work performed in my thesis

within context of the field.

This paper covers how silicone and siloxane chemistry has been previously used in the literature

with organic semiconductors. These examples range from small molecules, to polymers, to

cross-linked films. I believe that these examples help to convey the power of silicon chemistry to

control the physical properties of organic semiconducting materials. As a side benefit, it acts as a

concise and topical literature review on the integration of various silicon chemistries into organic

semiconducting materials.

2.2 Statement of Contributions

The idea, content, organization, and writing of this paper were entirely done by me with writing

input from Prof. Bender as corresponding author.

2.3 Paper

2.3.1 Abstract

Optimization of the physical and electronic properties of organic semiconductors is a key step in

improving the performance of organic light emitting diodes, organic photovoltaics, organic field

effect transistors, and other electronic devices. Separate tuning of the physical and electronic

properties of these organic semiconductors can be achieved by the hybridization of organo-

silicon structures (silicones, siloxanes, silsesquioxanes) with organic semiconductors. Common

chemical means to achieve this hybridization are summarized while a large range of literature

examples are covered to demonstrate the range and flexibility of this synthetic strategy.

12

12

2.3.2 Introduction

The study and application of organic semiconducting materials has emerged as an area of intense

interest in both the academic and industrial communities. The proven utility of organic

semiconducting materials in light emitting diodes (OLED),1,2 photovoltaics (OPV),3,4

xerography,5 field effect transistors (OFET),6 and many other device types,7 has prompted

significant study into both the discovery of new materials (novel compositions) and optimizing

known classes of materials and their chemical structures. To do this, chemists and engineers have

relied on the ability to modify the electronic and physical properties of these semiconducting

materials through rational chemical design of the semiconductors’ structure.8 Broadly stated,

changes to a chemical structure can be used to tune the electronic properties of materials in order

to alter its energy levels, interactions with light, chemical stability and other properties. These

changes can be subtle, such as an addition of a single electron withdrawing or electron donating

group. Or, they can be more substantial like dimerization or extension of a π conjugated system.

Chemical modifications can also alter the physical properties of these materials, properties which

impact the processibility of the material as well as its performance. These properties can be tuned

to improve such characteristics as device processing conditions, thin film formation, solubility,

and intermolecular interactions. Absolute control over both the electronic and physical properties

is critical when designing organic semiconducting structures. And, care must be taken not only to

how the material acts on its own, but how it acts when combined with other complementary

materials in a complete device.

While significant changes can be made to semiconductor properties through molecular

engineering, trying to tune certain properties while not affecting others can be quite difficult. For

example, one may want to increase conjugation length in a planar organic semiconductor to

improve charge transport. Such changes can result in a decrease in solubility and changes in the

intermolecular interactions in the solid state. Indeed, both the physical and electronic properties

of many organic semiconductors are interrelated and chemical modifications are rarely entirely

selective to one set of properties or the other. Therefore, it is of immense utility to be able to

engineer materials with independent control over both the physical and electronic properties.

One method which has been successful in decoupling physical and electronic properties has been

the incorporation of siloxanes and silicones into organic semiconductor molecular structures. It is

13

13

this method which is the focus of this tutorial review.

Silicones and siloxanes are a ubiquitous class of materials in modern society and are found in

many different applications including anti-foaming agents, lubricants, cosmetics, and coatings.9

This broad group of materials are generally considered quite thermally stable, electronically inert

and rather non-toxic. Most industrial applications use poly(dimethylsiloxane) (PDMS) or cross-

linked varients of this same polymer. But, silicon chemistry is an extremely diverse subject

touching many aspects of biology, organic synthesis, and materials chemistry. An authoritative

book on silicon in many different roles may be referenced for more information.10 Much of

silicon chemistry is dominated by its bonding with oxygen atoms and different classes of silicon

containing materials can be defined by their bonding with this element. Therefore, from a

nomenclature standpoint we will consider all silicon compositions with a basic formula of

[R2SiO2]n to be a siloxane or silicone. Furthermore, we will roughly define siloxanes as small

molecule fragments with this empirical formula and silicones to be their macromolecular

equivalents. Silicon compounds with the empirical formula RSiO2/3 are known as

silsesquioxanes. These cage like structures share some properties and common chemistry with

siloxanes and silicones and thus will be included in this tutorial. As well, silicon can take on

other empirical formulae resulting in materials that are electronically active. Examples include,

silicon atoms included in conjugated aromatic structures such as siloles and silica nano-crystals

(with an formula of [SiO2]n). While certainly interesting, these additional classes of materials

will not be considered within the scope of siloxanes and silicones and this tutorial.

At first glance, the incorporation of silicon(e) into an organic electronic material is a rather

curious strategy. Siloxanes and silicones are known for their high dielectric constants and are

considered to be quite electronically inert. However, it is this exact property that allows

independent tuning of physical and electronic properties in these hybrid materials. Because

added siloxane functionalities have a small impact on the electronic properties of a

semiconductor, their function can purely be to engineer the physical attributes of the final

material. This idea is illustrated in Figure 2-1.

In this tutorial review, the use of silicones and siloxanes in organic electronic materials will be

14

14

discussed and explored. Specifically, the methods of incorporating silicones and siloxanes into

the chemical structure of existing classes of organic semiconducting materials will be briefly

covered. This will include popular and efficient coupling strategies used to join silicon

containing precursor materials to organic semiconductors as well as some fundamental

considerations into their use. Examples of hybrid materials will be discussed in detail with a

focus on the specific impact that silicone and siloxane incorporation has on the final physical and

electronic properties of the materials. For the purpose of this article, these materials are divided

into several broad groups based on the nature of materials and include: discrete small molecules

and dendritic structures, polymers, and cross-linked films. This tutorial review is written to serve

as a tour into how traditional silicon chemistry can be integrated into the area of organic

semiconducting materials to produce new and exciting forms of applied matter and materials.

Figure 2-1: The molecular fragments of a silicone/siloxane organic electronic hybrid material (using a triarylamine as a representative example) and the relative effects of each fragment on

the overall properties of the hybrid.

2.3.3 Chemistries for Silicone and Siloxane Incorporation

While there may be many methods for chemically bonding siloxane and organic semiconductor

functionalities, several considerations must be made before a synthetic pathway is chosen. The

first must be the availability of different siloxane and silicone precursors and their associated

chemical functionalities. The availability of these reagents greatly reflects the industrial uses of

these chemicals. Because of this, certain silicon functionalities may be acquired quite

inexpensively while others inexplicably difficult or impossible to find. The second consideration

15

15

is the functionality present in the organic semiconducting fragment. In particular, some thought

must be made into how the organic semiconducting group is to be coupled with the silicon

containing fragment. Because most organic semiconducting structures have more synthetic

variability than do silicone or siloxane precursors, this half of the coupling strategy is often more

flexible. Below we outline several coupling strategies; each requiring a different, yet commonly

encountered functional group to be present on each of the organic semiconducting segment and

silicon containing fragment.

The last consideration and the one that limits the number of potential reagents/reaction

conditions used is the potential chemical instability of siloxanes and silicones. While stable to

many conditions, siloxane and silicone structures can often be hydrolytically unstable at any pH

other than neutral. Reaction or workup conditions involving acidic or basic aqueous conditions

will often result in hydrolysis ultimately leading to redistribution and/or metathesis of the

siloxane component.10, 11 This will lead to a loss of discrete structure and the generation of poorly

defined silicone oligomers in most cases, a problem which may be of some concern depending

on the needs of the final material. As will be seen in the examples to come, these issues can be

avoided by simply choosing reaction and purification conditions appropriately.

Finally, while other ways to form silicon/siloxane bonds or couple silicones to carbon do clearly

exist, we have chosen to focus on three primary reactions for this tutorial based on their

generality and frequency of use in the literature.

16

16

Scheme 2-1: Summary of some common coupling techniques to join organic semiconductors with silicone/siloxane components.

2.3.3.1 Hydrosilylation

Hydrosilylation is one of the most common methods for creating hybrid materials from silicones

and perhaps one of the most general. Hydrosilylation is the addition of a silane (Si-H) to an

unsaturated carbon-carbon bond (alkene or alkyne) resulting in the formation of a carbon-silicon

bond typically following anti-Markovnikov addition (Scheme 2-1).12 This transformation is most

often mediated by platinum-based catalysts, although free-radical initiated methods as well as

rhodium, nickel, and other transition metals can be utilized.10 To effect this reaction, the two

most common platinum catalysts are either Pt0 species such as Karstedt’s catalyst (Scheme 2-1)13

or H2PtCl6, a PtIV species known as Speier’s Catalyst (Scheme 2-1).14 For practically all of the

examples discussed below, these catalysts are sufficiently active to achieve clean reactions while

also being commercially available. For most systems containing few functional groups (most

organic semiconducting species), platinum catalysts are effective and a good starting point for

experimentation. Systems which contain strong coordinating functionalities such as amines or

phosphines are not compatible with this chemistry due to competitive binding of the platinum

catalyst. Many other common silicone/siloxane functionalities are compatible with this chemistry

17

17

including chloro-silanes and alkoxy-silanes which can be further reacted in subsequent processes

under orthogonal conditions. For the unsaturated carbon-carbon bond, vinyl or allyl groups are

most often used and easily incorporated into many organic semiconducting structures usually by

using suitably functionalized precursor materials during synthesis.

2.3.3.2 Piers-Rubinsztajn Reaction

The Piers-Rubinsztajn reaction15,16 is a relatively new process to join siloxane/silicone

functionalities with organic molecules, including organic semiconducting materials.17-20 This

reaction takes place between a silane and an aryl or alkyl hydroxyl or alkoxy group (Scheme 2-1)

and is catalysed by tris(pentafluorophenyl)borane (B(C6F5)3 or simply BCF), a strong and water

tolerant Lewis acid. Like hydrosilylation, the Piers-Rubinsztajn reaction proceeds with low

catalyst loadings and there are no significant by-products beyond gaseous hydrocarbons which

are easily dealt with on a laboratory scale. Because Piers-Rubinsztajn reactions can result in the

rapid generation of flammable gases (H2, CH4, etc.), some degree of caution must be exercised

when dealing with this process. Despite this minor downside, the reaction is compatible with a

variety of different silanes including small molecules18-20 and polymers.17 It should be noted that

alkoxy-silanes will also react under these conditions.21 Many other chemical functionalities are

tolerated so long as they cannot coordinate with the highly Lewis acidic BCF catalyst.22

Molecules with even moderate donor capacity (carbonyls for example) can either poison the

catalyst or undergo unwanted side-reactions such as hydrosilylation of the carbonyl group.23

While the Piers-Rubinsztajn reaction may be less tolerant of other functional groups than

hydrosilylation, it has the added benefit of using simple coupling partners such as hydroxyl or

alkoxy functionalities. These functionalities can easily be incorporated into the chemical

structures of some organic semiconductors. As well, unlike the functional groups required for

hydrosilylation, these functionalities are not potentially heat sensitive. Given these two points,

Piers-Rubinsztajn conditions may result in a more straightforward synthetic pathway in certain

situations to silicone containing organic electronic materials.

2.3.3.3 Mizoroki-Heck Reaction

The Mizoroki-Heck reaction is a palladium catalysed cross-coupling reaction between an

unsaturated carbon-carbon bond (typically an alkene) and an aryl halide resulting in the

formation of a carbon-carbon single bond (Scheme 2-1).24,25 This reaction is most often catalysed

18

18

by palladium metal ligated by an organophosphine in the presence of stoichiometric amounts of a

weak base; although many variations exist.26 Because of the widespread use of hydrosilylation

chemistry in industry, many vinyl-silicone functionalized starting materials exist. These vinyl

groups can be used as reagents in the Mizoroki-Heck reaction making them convenient starting

materials. Similar to hydrosilylation, this reaction has a wide scope and is compatible with many

organic functional groups. As well, it has the advantage of requiring an aryl-halide functionality

for reaction. Halides can be easily introduced in many chemical structures through the use of

easily handled halogenating reagents such as N-halosuccinimides. Aryl-halides are also more

heat stable than many vinyl-functionalities used in hydrosilylation conditions thus making

purification of organic electronic materials containing halides more desirable.

2.3.4 Examples of Hybrid Materials

2.3.4.1 Polymeric Materials

Polymeric materials represent the first disclosed examples of siloxane-organic semiconductor

hybrid structures. The concept was first shown by Strohriegl in a report from 1986.27 Drawing

from interest in poly(vinylcarbazole) (PVK) as a photoconductor for xerographic photoreceptors

and a precursor to conductive polymers, Strohriegl and co-workers synthesized a side-chain

carbazole polymer with a siloxane main chain as a structural analogue to PVK. This was

achieved by reacting poly(methylhydrosiloxane) with an N-vinyl functionalized carbazole under

standard hydrosilylation conditions, the product of which was given the general term PSX

(Scheme 2-2). Over the course of this work and several follow up publications,28-30 Strohriegl

and co-workers demonstrated that this is a facile method to producing siloxane-organic

semiconductor hybrid materials.

19

19

Scheme 2-2: Summary of synthetic strategies towards side-chain polymeric organic semiconductors.

The PSX class of materials was found to have significantly lower glass transition temperatures

(Tg) than PVK (which generally has a Tg > 200 °C). This is expected given the greater

conformational freedom found in poly(siloxane)s as compared to typical polyalkanes. This

increase in conformational freedom along the main polymer chain also results in a complete

absence of carbazole excimer formation under photoluminescence. In contrast, significant

excimer formation is observed in PVK thin solid films and in solution. This had already been

studied in detail for PVK and is understood to occur due to interactions between neighbouring

carbazole pendant units.31 The absence of these interactions in the analogous poly(siloxane)

hybrid materials is a good example of how siloxane hybridization is capable of controlling

intermolecular interactions between organic semiconducting moieties. Finally, these materials

were found to be effective photoconductors as shown by a series of time-of-flight mobility

experiments.30 Comparing a series of different polymers with varying side chain lengths, it was

found that these polymers have comparable charge mobility and photogeneration efficiencies to

PVK. Considering the already proven utility of PVK for organic electronic devices at the time,

these initial results showed that siloxane hybrid materials also had the potential to be

successfully used in such devices.

2-2 2-3

2-1

(PSX)

20

20

Most of the hybrid macromolecular materials share a common synthetic pathway to post-

polymerization functionalization, namely the utilization of hydrosilylation chemistry. However,

an alternate pathway to similar materials has recently been demonstrated by our laboratory

(Scheme 2-2, Compound 2-2).17 This approach relies on Piers-Rubinsztajn conditions to react an

aryl methoxy functionalized triarylamine with poly(methylhydro-co-methylphenyl)siloxane. This

alternate synthetic methodology has the benefit of avoiding the need for reactive allyl or vinyl

groups to be present on the triarylamine units. Instead it utilizes aryl-methyl ether groups as

reaction partners (Scheme 2-2) which are chemically stable prior to Piers-Rubinsztajn conditions

and easily incorporated into triarylamines. As well, it does not require expensive platinum based

catalysts and alleviates concerns about their subsequent removal from the final product.

Of all the potential applications of these materials, the area of photorefractive composites has

been impacted the most by the availability of siloxane hybrid polymers. Devices based on these

composites respond to incident light by changing their refractive index. Because of these

properties, photorefractive devices based on organic polymers are of interest for potential

application in optical data storage media and optical security systems.32 Traditional polymeric

systems based on blends of PVK as a photosensitizer and non-linear optical chromophores show

good initial performance but suffer from several drawbacks. Crystalline components in these

blends are not morphologically stable and many blends suffer from crystallization and phase

separation over time. The other serious problem is due to the orientational enhancement effect in

photorefractive polymers.33 This effect is the enhancement of the photorefractive properties with

increasing orientational mobility of the photorefractive components within the blend. Because of

this, optimal performance for these devices is achieved in blends with a great deal of

conformational freedom and a low glass transition temperature (Tg). This requirement

exacerbates the issues of morphological instability, an issue which can be partially eliminated by

the use of additional plasticizers. But, these plasticizers have the additional downside of diluting

the photorefractive effect.

Because of their inherently low Tg and amorphous nature, siloxane modified polymeric host

materials have been shown to be a solution to the aforementioned issues when acting as hosts in

21

21

photorefractive applications. The first example of this application used a copolymer that

incorporated both carbazole units and a non-linear chromophore pendant to the same siloxane

polymer chain.34 This approach allowed for tuning of the component ratios by modifying the

polymer stoichiometry without fear of phase separation or crystallization issues. Using this

approach, Bratcher et al. produced polymers with both good morphological stability and

moderate photorefractive gain coefficients (2.8 cm-1). It was discovered later that polymers

containing only hole-transport groups could be used when doped with small molecule

photosensitizers and non-linear optical chromophores forming a simplified photorefractive

system. Such doped systems were first described by Moon et al. using PSX as a host material to

achieve a photorefractive systems with exceptional gain coefficients (>390 cm-1) and stable

morphological properties.35,36 Building on these studies, other photorefractive systems

incorporating hole transporting moieties other than carbazole have been reported.37-41 These

alternate hole transporting polymers carry over many of the benefits of PSX while offering

improved response time due to the greater photoconductivity of many hole transport materials as

compared to carbazole.

More complex silicone structures have also been functionalized with organic semiconducting

groups. Ladder type polysilsesquioxanes hybrid materials have been developed by reacting a

trialkoxy substituted organic semiconductor under specific hydrolytic condensation conditions.

Using this method, a carbazole substituted polysilsesquioxane has been reported (2-3) and

characterized.42 This material shows many similar properties to PSX and its derivatives

including good solution processibility and minimized excimer formation upon photoexcitation.

Uniquely, this ladder type polymer has a glass transition temperature of 95 °C. This value is

significantly higher than that found for PSX derivatives and is attributed to the more rigid

ladder-type silicone backbone. The polymer readily forms glassy films from solution and was

demonstrated to function as a hole transporting polymer and host in a simple single layer OLED.

This strategy has also been applied to perylene bisimides, a common electron transporting

material.43 Upon condensation, the alkoxysilane functionalized monomers produces a thermally

stable, film forming polymer. This polymer is found to have an exceedingly high glass transition

of ~310 °C and stable electrochemistry that resembles the monomeric perylene bisimide.

22

22

2.3.4.2 Cross-linked Films

Based on the results and chemistry of silicones outlined above, it should come as no surprise that

the idea of silicone hybrid material evolved towards the use of cross linked films. Cross-linked

films are occasionally more desirable as semiconducting layers in organic electronic devices than

their soluble counter parts for several reasons. The principal reason is due to the insolubility

found in most cross-linked films. This allows solution processing of additional functional layers

on top of a cross-linked layer without the potential for interlayer mixing. By incorporating cross-

linked layers into a complex device structure, sequential solution processing steps can used to

deliver a multi-layered structure. These kinds of multi-layered solution processed devices can

normally only be achieved by using orthogonal solvent systems for different layers, a strategy

which is inherently limited. Cross-linked films also can result in system morphology that is

significantly stabilized over time. Because molecular orientations are locked in by the cross

linking process, crystallization or phase separation observed in other organic semiconductor

types can be avoided for increased device stability.

Scheme 2-3. Cross section of an OLED device and examples of siloxane containing materials applied at either the ITO/hole transport layer interface or in the bulk of the functional layers (as indicated by color coding).

2-6

2-5

2-4

2-7 R = -SiCl3

2-8 R = -Si(OMe)3

23

23

The first example of using a polycondensation process to cross-link organic semiconductor films

relied on the sol-gel reaction of alkoxysilane functionalized organic semiconductors.44 By

preparing analogues of a known emissive material (2-4) and electron (2-5) and hole transporting

(a carbazole derivative) layers (Scheme 2-3) with triethoxysilane groups, homogeneous films

could be achieved by depositing the molecules under typical sol-gel conditions. Using these

conditions, alkoxysilane groups hydrolyze, self-condense and cross-link with each other11 to

form stable films. The utility of such cross-linked films was demonstrated by depositing multiple

layers on top of one another to achieve a multilayer OLED structure. Because each organic layer

is cross-linked and insoluble, subsequent layers do not disturb the interface of the previous layer.

OLEDs produced using this method had high yet reasonable turn on voltages (~13 V) and rather

average external quantum efficiencies (<0.5 %, for unoptimized devices).44

Another approach to cross-linked layers relied on the use of trichlorosilane groups as

polymerizable functionalities incorporated into triarylamine molecules (such as 2-6, 2-7 or 2-8 in

Scheme 2-3).45 The hydrolytically unstable trichlorosilane groups rapidly hydrolyze, condense

and react/crosslink with each other upon contact with atmospheric moisture in a sol-gel

process.11 These materials were found to yield high quality homogeneous cross-linked films

when spin coated onto a transparent indium-tin oxide (ITO) substrates under ambient conditions.

The resulting films were found to have very good mechanical and thermal properties as would be

expected for a highly cross-linked film. More importantly, the films retained the useful

electrochemical properties of the triarylamine moieties. Practical utility of this approach was

demonstrated by constructing simple bilayer OLEDs using this cross-linked layer as a hole-

transport layer. The resulting devices displayed low turn on voltages (~6 V) and typical external

quantum efficiencies for such device architectures (~0.2%). Combined with the results

highlighted in the preceding paragraphs, these devices again demonstrate that cross-linking

triarylamines using sol-gel chemistry does not greatly diminish their hole transporting properties

and are a way to produce insoluble hole transporting layers.

The use of cross-linked organic semiconductor layers of triarylamines has also been

demonstrated as a way to increase the performance of fluorescent OLED devices through

improved hole injection and exciton confinement. Several studies based on these interlayers have

24

24

yielded optimized fluorescent devices with excellent performance based on both solution

processed46 and multi-layered vacuum deposited OLED architectures.47,48 This increase in device

performance is the result of modifying the electronic properties of the electrode to better work

with adjacent organic semiconducting materials. In particular, the use of these crosslink organic

semiconducting layers can modify the energy levels of the electrode to improve charge

injection/collection while blocking unwanted opposite charges. Similar results have also been

achieved using interlayers derived from the more hydrolytically stable alkoxysilanes.49 The use

of such alkoxysilanes allows for easier synthesis, purification, and handling of the final materials

while achieving similar results. A highly optimized vacuum deposited device incorporating a

cross-linked interlayer yielded a maximum external quantum efficiency of ~4.4% and a low turn-

on voltage of ~4.5 V.50

Recently, phosphorescent OLEDs have also been produced using cross-linked triarylamines as

hole transport layers.51 In this example by Lim et al, the alkoxysilane functionalized triarylamine

was deposited as a hole-transport layer and cross-linked. An emitting layer containing a

phosphorescent dopant in a polymeric host material was solution deposited on top followed by

several vacuum layers. As expected, the performance of this phosphorescent device exceeded

that of the aforementioned fluorescent devices with higher external quantum efficiencies (up to

6.4%) and similarly good luminance and turn-on voltages.

The utility of such crosslinked films prepared using sol-gel chemistry has also been

demonstrated in the area of xerographic photoreceptors. In this case a triarylamine containing

two reactive diisopropyl silane groups was prepared by reaction of a carboxylated precursor with

3-iodopropyl diisopropyl silane (2-9, Figure 2-2). Subsequent sol-gel processing of the

triarylamine in the presence of a second electronically inert material (derived from

hydrosilylation of technical grade divinyl benzene) following by coating onto a photoreceptor

device yielded high quality, hard hole conducting films. When compared to typical non-

crosslinked coatings for xerographic photoreceptors, these hybrid films were found to have

superior mechanical properties and better resisted abrasive wear while maintaining printed image

quality.52-54

25

25

Figure 2-2. Example of siloxane containing triarylamie applied in a xerographic photoreceptor.

Another application for these kinds of materials is as a thin interlayer on electrodes for organic

electronic devices. Transparent conducting electrodes for OPV and OLED devices are useful for

obvious reasons. However, these metal oxide electrodes have the disadvantage of having a

hydrophilic surface and energy levels that may not match complementary organic materials

contacting the electrode within a OPV or OLED. Their hydrophilic surface properties can cause

issues when attempting to coat these electrodes with hydrophobic organic semiconductors. This

mismatch in surface energies can result in surface wetting issues and delamination of organic

layers upon heating. It has been shown that silane functionalized triarylamines can mitigate these

issues by acting as a compatibilizing layer for other organics.55, 56 Monolayers of triarylamines

can be prepared by coating hydroxyl-functionalized indium tin oxide (ITO) substrates with the

chloro or alkoxy silane (such as 2-6, 2-7 or 2-8 in Scheme 2-3) functionalized material under

inert conditions. This procedure results in the formation of a monolayer of hole-transporting

molecules that is self-limiting in its thickness from a lack of atmospheric moisture needed for

further reaction. Using this modified ITO layer as an anode for a bilayer OLED, greatly

enhanced thermal stability and charge injection into the organic layer were found as compared to

both unmodified ITO and ITO modified with a typical organic hole injection layer such as

PEDOT:PSS.56 This enhanced stability was attributed to better interfacial adhesion between the

layers of the device.

This approach has also been extended to cross-linkable interlayers for organic photovoltaics.

Much like for OLEDs, OPVs benefit from electrode interlayers to improve both surface wetting

and electronic properties for better compatibility with subsequent layers. Silyl chloride

functionalized triarylamines (such as 2-6, 2-7 or 2-8 in Scheme 2-3) have been used as

2-9

26

26

interlayers for prototypical bulk heterojunction OPVs based on conjugated polymers and

fullerenes as electron donors and electron acceptors, respectively. For OPVs with either

polyphenylenevinylene (MDMO-PPV)57,58 or poly-3-hexylthiophene (P3HT)59 as donor

polymers, it was found that the cross-linkable triarylamines blended with a high mobility

polymer functioned as a suitable replacement with superior performance when compared to the

prototypical interlayer poly(ethylenedioxythiophene):poly(styrene sulfonate) (PEDOT:PSS) in

the case of OPVs based on MDMO-PPV. For P3HT based devices, almost identical performance

was achieved.

Figure 2-3: Example of siloxane containing material used as an electron selective interlayer for OPV.

Electron selective interlayers based on siloxane chemistry have also been demonstrated to act as

interlayers for organic photovoltaics.60 Conceptually this differs from the above example as this

interlayer needs to selectively accept negative charges instead of positive charges. Hains et al

showed a functionalized perylene bisimide derivative (2-10, Figure 2-3) could form an electron

selective interlayer on ITO in an inverted P3HT/fullerene based device. The resulting devices

showed fairly mediocre performance which improved upon doping the layer with ionic

components and a high mobility conjugated polymer. Despite the performance of these initial

results, we feel the idea of developing electron selective cross-linking systems should be of great

interest to those interested in the development of inverted OLEDs and OPVs.

2.3.4.3 Discrete Molecules

2.3.4.3.1 Silsesquioxane Based Semiconductors

Despite the propensity for siloxane functionalities to undergo polymerization and cross-linking

reactions, there are many examples of discrete siloxane structures in organic electronic materials.

The most prominent of which are silsesquioxane hybrid materials. Silsesquioxanes are a class of

silicon based materials with an empirical formula of RSiO2/3.61 This is a ratio of silicon to

2-10

27

27

oxygen between that of pure silica (SiO2) and silicones. This unique bonding arrangement results

in a cage structure that can be a discrete assembly, a partially defined unit, or a macromolecular

structure. These are all commonly referred to as polysilsesquioxanes (POSSs). Several discrete

assemblies are commonly encountered and the naming of different units follows a Tn naming

convention, where n is the number of silicon atoms. The most common unit, T8, has 8 silicon

atoms comprising a cubic structure that allows for up to 8 functionalities to be present at each of

the vertices of the cube (Figure 2-4).

Figure 2-4. Structure of common silsesquioxane-T8 synthetic precursors.

POSSs are unique compared to other silicone classes in that the discrete T8 unit is

conformationally rigid. Consequently, their use in organic electronic semiconductors is primarily

to act as a rigid inert central unit which can be functionalized with many different active

semiconducting moieties. This hybridization generally results in materials with better thermal

stability, solution processibility, improved film formation, and reduced intermolecular

interactions between the semiconducting moieties. This strategy has also been used to engineer

small molecules which have poor film forming properties into hybrid materials which can be

readily solution processed. The functionalization of POSS cores with organic semiconducting

structures has emerged as a way to produce novel and useful materials, primarily for OLED

applications.

28

28

Figure 2-5: Examples of several POSS-T8 derivatives functionalized with organic semiconducting groups.

The first examples of hybrid POSS strategy involved complete or partial coupling of hole

transporting groups onto the periphery of the T8 unit. In work by Sellinger et al.,62 an octa-vinyl

functionalized T8 was poly-functionalized with a two nitrogen centred triarylamine by the

Mizoroki-Heck reaction (see Scheme 2-1) resulting in a mixture of hybrid materials with 3 to 10

substitutions. This surprisingly large range of substitutions suggests very poor selectivity with

this synthetic strategy or an inability to push the reaction to completion. Regardless, the final

triarylamine functionalized T8 material was found to form stable amorphous glasses from

solution casting. Films of the material showed identical photoluminescence to samples in

2-13

2-14

2-12

2-11

29

29

solution indicating a lack of aggregation in the solid state. The new material and the

bare/precursor triarylamine were used as hole transport layers with tris(8-

hydroxyquinolinato)aluminium (AlQ3) in a standard 2-layer OLED configuration. Both devices

performed similarly indicating that POSS-functionalization had a negligible impact on the

electronic properties of the triarylamine while altering its physical properties.

Complete substitution of the POSS core has been achieved using the hydrosilylation approach

(see Scheme 2-1). In this strategy, N-vinylcarbazole was reacted with

octakis(dimethylsiloxy)silsequioxane using Karstedt’s catalyst under hydrosilylation conditions

to yield a hybrid POSS (Figure 2-5, 2-11).63 The resulting compound is very thermally stable (up

to 400 °C) and readily forms stable amorphous glasses from a melt. Much like PSX polymers

and the aforementioned example by Sellinger et al, photoluminescent spectra shows no excimer

formation. This again suggests that this approach results in a material with negligible interactions

between neighbouring carbazole units in the solid state. Hydrosilylation appears to be far more

selective and controllable approach (yielding 8 substitutions) as compared to the Mizoroki-Heck

approach where partial and/or over-functionalization appears to be common.64 Other examples of

the functionalization of POSS using hole transporting units have been reported including using

other carbazoles,65 and a variety of triarylamines.66

The functionalization of POSS cores can also be used to obtain novel light emitting materials for

use in OLEDs. The benefit to this approach relies on the known ability of POSS

functionalization to minimize interactions between semiconducting molecules (as highlighted in

the preceding paragraphs). By functionalizing a POSS with fluorescent units, the intermolecular

interactions between these fluorescent units can be limited which in turn improves the

fluorescent efficiency of these materials in the solid-state. This is very desirable as high solid-

state photoluminescent yields are a critical parameter for the production of highly efficient

OLEDs.67

The successful application of this idea was first shown by the synthesis of a POSS core

functionalized with oligofluorenes and applied in a blue emitting OLED.68 Despite poor

performance for unoptimized devices, the hybrid POSS materials performed better than their

30

30

polymeric or small molecule counterparts. The reason for this difference was again ascribed to

the reduced intermolecular interactions between the flat, crystalline oligofluorenes that normally

quench luminescence in the solid-state. Another striking example of this of this idea was shown

by the Sellinger group where the highly planar and crystalline pyrene fluorophore was

hybridized with a POSS core.64 The hybrid pyrene materials easily formed amorphous films with

excellent morphological stability and high photoluminescence quantum yields. OLEDs made

from this material performed quite well with external quantum efficiencies peaking at 2.63%,

high luminescence (36000 cd/m2), and low turn-on voltages (3.1 V). From these studies, it is

clear that hybrid POSS materials can produce highly luminescent films containing chromophores

that normally would not emit strongly in the solid-state.

Building on this strategy, POSS cores with up to three different fluorophores have been

synthesized and studied (Figure 2-5, 2-12).69 By using various combinations of blue, yellow, and

orange emitters, OLEDs with electroluminescence from each of the different chromophores was

demonstrated. Unfortunately, Forehlich et al showed the small spatial separation and energy

level overlap of the multiple emitters resulted in a great deal of energy transfer from the higher

energy blue emitter to the lower energy orange or yellow emitters. This resulted in unbalanced

emission which may limit the further exploration of POSS materials with multiple emitters.

Higher OLED efficiencies have been achieved by incorporating phosphorescent emitters onto

POSS cores. The first such effort used an analogue of the well-known Ir(ppy)2(acac) phosphor to

produce a fully derivatized POSS derivative with 8 emitting Ir(ppy)2(acac) moieties.70 Quantum

yields of the hybrid POSSs were moderately higher than the lone phosphors indicating the utility

of incorporation of the rigid POSS core. Single layered OLEDs showed very modest

performance with high turn-on voltages (12 V) and low brightness (max of 1172 cd/m2). Yang et

al improved on this result by synthesizing additional POSS derivatives with three different

analogues of known iridium based emitting complexes.71 Single layer OLEDs with high

efficiencies were achieved by blending the POSS derivatives into a mixture of PVK and an

electron transporting material. Using this approach, high efficiency red, green and blue devices

were demonstrated with external quantum efficiency values of up to 8.4%, high luminance and

low driving voltages. Blending of the three emitters was shown to produce white light OLEDs.

31

31

In the same study, even higher efficiencies were found when the hydrid POSS derivatives

containing 7 carbazole groups and 1 iridium emitter were synthesized. By chemically combining

hole transporting groups and emitting groups onto the same POSS core, multi-functional

materials were realized. Devices based on these mixed emitting and hole transporting derivatives

showed better performance which is attributed to decreased interaction between iridium

complexes and improved charge transport. Most significantly, single layered devices using this

multifunctional hybrid POSS material in the absence of additional hole or electron transporting

materials produced a device with moderate efficiency (1.3%). This demonstrated that

incorporating mixed functionalities onto a single POSS core is a promising strategy to efficient

OLEDs with significantly simplified structures. Analogous materials using platinum based

emitters have also been demonstrated (Figure 2-5, 2-13).72 These materials showed similar

function and efficiency to those made with iridium emitters.

POSS functionalization with other organic semiconducting functionalities has been demonstrated

but not explored as thoroughly as the above examples. POSS derivatives with electron

transporting 2,5-diphenyl-1,3,4-oxadiazole groups (Figure 2-5, 2-14) have been synthesized

although no electron transporting capability was demonstrated.73 Derivatives containing boron

dipyrromethene dyes have also been reported.74 Both of these cases show better thermal stability

and photoluminescent efficiency for the POSS functionalized molecules as compared to their

small molecule equivalents.

The ability of POSS cores to be decorated with different functional groups may open the

possibility of even more complex materials bearing multiple transporting groups and emitting

materials with minimal synthetic effort. Given that adding functional groups to POSSs seems to

only improve thermal properties and solution processibility, this approach may serve as a general

and facile way to improve the physical properties of many other known organic semiconductors

for a variety of device applications. In our opinion, further exploration of this strategy is

certainly warranted.

32

32

2.3.5 Conclusions and Outlook

In summary, the use of electronically inert silicone and siloxane fragments in organic

semiconducting materials has been outlined. Both the common chemistries and synthetic

concerns needed to obtain these materials have been reviewed as well as different examples of

polymeric, cross-linked and discrete hybrid materials comprised of silicones, silsesquioxanes and

siloxanes. Despite the large range of materials used for a variety of applications, all of these

materials can be tied together with a common theme: the use of silicone and siloxanes to control

the physical properties of the final material without altering the inherently useful electronic

properties of the organic semiconducting fragment. This has been demonstrated across a variety

of different material types proving that this strategy is a powerful synthetic tool available to

engineer new and potentially better performing materials for organic electronic devices.

Finally, this field has been slowly growing over a long period of time resulting in a number very

interesting and novel uses of these hybrid materials in various organic electronic devices from

various independent research streams. But, many of these research streams have not overlapped

greatly despite some of the common goals and techniques used between groups. As such, it is

hoped that this tutorial review has been able to put these various research efforts into context

with one another and may potentially be used to inspire subsequent efforts within the field.

2.4 References

1. A. C. Grimsdale, K. L. Chan, R. E. Martin, P. G. Jokisz and A. B. Holmes, Chem. Rev., 2009,

109, 897-1091.

2. L. Xiao, Z. Chen, B. Qu, J. Luo, S. Kong, Q. Gong and J. Kido, Adv. Mat., 2011, 23, 926-952.

3. A. W. Hains, Z. Liang, M. A. Woodhouse and B. A. Gregg, Chem.l Rev., 2010, 110, 6689-

6735.

4. Y.-J. Cheng, S.-H. Yang and C.-S. Hsu, Chem. Rev., 2009, 109, 5868-5923.

5. P. M. Borsenberger and D. S. Weiss, Organic Photoreceptors for Xerography, Marcel Dekker

Inc., New York, 1998.

6. S. Allard, M. Forster, B. Souharce, H. Thiem and U. Scherf, Angew. Chemie Inter. Ed., 2008,

47, 4070-4098.

7. S. R. Forrest and M. E. Thompson, Chem. Rev., 2007, 107, 923-925.

8. Y. Shirota and H. Kageyama, Chem. Rev., 2007, 107, 953-1010.

33

33

9. M. Androit, S. H. Chao, A. Colas, S. Cray, F. de Buyl, J. V. Degroot, D. A., T. Easton, J. L.

Garaud, E. Gerlach, F. Gubbels, M. Jungk, S. Lealey, J. P. Lecomte, B. Lenoble, R. Meeks,

A. Mountney, G. Shearer, S. Stassen, C. Stevens, X. Thomas and A. T. Wolf, in Inorganic

Polymers, eds. R. D. Jaeger and M. Gleria, Nova Science Pub Inc New York, 2007, pp. 61-

161.

10. M. A. Brook, Silicon in Organic, Organmetallic, and Polymer Chemistry, John Wiley &

Sons, New York, 2000.

11. L. L. Hench and J.K. West, Chem. Rev., 1990, 90, 33-72.

12. B. Marciniec, Comprehensive Handbook on Hydrosilylation, 1st edn., Pergamon, 1992.

13. B.D. Karstedt, US 3,775,452. November 27,1973.

14. J. L. Speier, J. A. Webster and G. H. Barnes, J. Am. Chem. Soc., 1957, 79, 974-979.

15. S. Rubinsztajn and J. Cella, Macromolecules, 2005, 38, 1061-1063.

16. M. A. Brook, J. B. Grande and F. Ganachaud, Adv. Polym. Sci., 2010, 235, 161-183.

17. M. J. Gretton, B. A. Kamino and T. P. Bender, Macromolecules, 2011.

18. B. A. Kamino, J. Castrucci and T. P. Bender, Silicon, 2011, 3, 125-137.

19. B. A. Kamino, J. B. Grande, M. A. Brook and T. P. Bender, Org. Lett., 2011, 13, 154-157.

20. B. A. Kamino, B. Mills, C. Reali, M. J. Gretton, M. A. Brook and T. P. Bender, J. Org.

Chem., 2011.

21. S. Rubinsztajn and J. A. Cella, Macromolecules, 2005, 38, 1061-1063.

22. J. B. Grande, D. B. Thompson, F. Gonzaga and M. A. Brook, Chem. Comm., 2010, 27,

4988-4990.

23. D. J. Parks, J. M. Blackwell and W. E. Piers, J. Org. Chem. , 2000, 65, 3090-3098.

24. T. Mizoroki, K. Mori and A. Ozaki, Bulletin of the Chemical Society of Japan, 1971, 44,

581.

25. R. F. Heck and J. P. Nolley, J. Org. Chem., 1972, 37, 2320-2322.

26. I. P. Beletskaya and A. V. Cheprakov, Chem. Rev., 2000, 100, 3009-3066.

27. P. Strohriegl, Makromol. Chem., Rapid Commun., 1986, 7, 771-775.

28. M. Lux and P. Strohriegl, Makromol. Chem., Rapid Commun., 1987, 188, 811-820.

29. M. Hennecke and P. Strohriegl, Makromol. Chem., Rapid Commun., 1988, 189, 2601-2609.

30. H. Domes, R. Fischer, D. Haarer and P. Strohriegl, Makromol. Chem., Rapid Commun.,

1989, 190, 165-174.

34

34

31. M. Yokoyama, T. Tamamura, M. Atsumi, M. Yoshimura, Y. Shirota and H. Mikawa,

Macromolecules, 1975, 8, 101-104.

32. K. Meerholz, B. L. Volodin, Sandalphon, B. Kippelen and N. Peyghambaraian, Nature,

1994, 371, 497-500.

33. W. E. Moerner, S. M. Silence, F. Hache and G. C. Bjorklund, J. Opt. Soc. Am. B., 1994, 11,

320-330.

34. M. S. Bratcher, M. S. DeClue, A. Grunnet-Jepsen, D. Wright, B. R. Smith, W. E. Moerner

and J. S. Siegel, J. Am. Chem. Soc., 1998, 120, 9680-9681.

35. I. K. Moon, D. Yoo, H. S. Moon, D.-H. Shin, W. S. Jahng, H. Chun and N. Kim, Molecular

Crystals and Liquid Crystals, 2000, 349, 43-46.

36. H. Chun, I. K. Moon, D.-H. Shin and N. Kim, Chem. Mater, 2001, 13, 2813-2817.

37. D. Wright, U. Gubler, W. E. Moerner, M. S. DeClue and J. S. Siegel, J. Phys. Chem. B,

2003, 107, 4732-4737.

38. S. H. Lee, W. S. Jahng, K. H. Park, N. Kim, W.-J. Joo and D. H. Choi, Macromolecular

Research, 2003, 11, 431-436.

39. I. K. Moon, J.-W. Oh and N. Kim, Journal of Photochemistry and Photobiology A:

Chemistry, 2008, 194, 327-332.

40. I. K. Moon, J.-W. Oh and N. Kim, Journal of Photochemistry and Photobiology A:

Chemistry, 2008, 194, 351-355.

41. I. Moon, C. Choi and N. Kim, Journal of Photochemistry and Photobiology A: Chemistry,

2009, 202, 57-62.

42. S.-S. Choi, H. S. Lee, S. S. Hwang, D. H. Choi and K.-Y. Baek, J. Mat. Chem., 2010, 20,

9852.

43. W. Fu, C. He, S. Jiang, Z. Chen, J. Zhang, Z. Li, S. Yan and R. Zhang, Macromolecules,

2011, 44, 203-207.

44. T. Dantas de Morais, F. Chaput, K. Lahlil and J. P. Boilot, Adv. Mater., 1999, 22, 107-112.

45. W. Li, W. Q., J. Cui, H. Chou, S. E. Shaheen, G. E. Jabbour, J. D. Anderson, P. Lee, B.

Kippelen, N. Peyghambaraian, N. Armstrong and T. J. Marks, Adv. Mater., 1999, 11, 730-

734.

46. H. Yan, P. Lee, N. R. Armstrong, A. Graham, G. A. Evmenenko, P. Dutta and T. J. Marks,

J. Am. Chem. Soc., 2005, 127, 3172-3183.

35

35

47. Q. Huang, J. Cui, J. G. C. Veinot, H. Yan and T. J. Marks, Appl. Phys. Lett., 2003, 82, 331.

48. Q. Huang, G. A. Evmenenko, P. Dutta, P. Lee, N. R. Armstrong and T. J. Marks, J. Am.

Chem. Soc., 2005, 127, 10227-10242.

49. J. Li and T. J. Marks, Chem. Mater, 2008, 20, 4873-4882.

50. Q. Huang, J. Li, T. J. Marks, G. A. Evmenenko and P. Dutta, J. Appl. Phys., 2007, 101,

093101.

51. Y. Lim, Y.-S. Park, Y. Kang, D. Y. Jang, J. H. Kim, J.-J. Kim, A. Sellinger and D. Y. Yoon,

J. Am. Chem. Soc., 2011, 133, 1375-1382.

52. T. P. Bender, US 7,390,599. June 24, 2008.

53. T. P. Bender, J.F. Graham, Y. Gagnon, C.K. Hsiao, N.X. Hu and Y. Qi, US 7,348,447.

March 25, 2008.

54. T. P. Bender and N.X. Hu, US 7,262,314. August 28, 2007.

55. J. Cui, Q. Huang, Q. Wang and T. J. Marks, Langmuir, 2001, 17, 2051-2054.

56. J. Cui, Q. Huang, J. G. C. Veinot, H. Yan, Q. Wang, G. R. Hutchinson, A. Richter, G., G. A.

Evmenenko, P. Dutta and T. J. Marks, Langmuir, 2002, 18, 9958-9970.

57. A. W. Hains and T. J. Marks, Appl. Phys. Lett., 2008, 92, 023504.

58. A. W. Hains, J. Liu, A. B. F. Martinson, M. D. Irwin and T. J. Marks, Adv. Funct. Mat.,

2010, 20, 595-606.

59. A. W. Hains, C. Ramanan, M. D. Irwin, J. Liu, M. R. Wasielewski and T. J. Marks, ACS

Appl. Mat. Inter., 2010, 2, 175-185.

60. A. W. Hains, H.-Y. Chen, T. H. Reilly and B. A. Gregg, ACS Appl. Mat. Inter., 2011.

61. M. I. Ronald H. Baney, Akihito Sakakibara, Toshio Suzuki, Chem. Rev. , 1995, 95, 1409-

1430.

62. A. Sellinger, R. Tamaki, R. M. Laine, K. Ueno, H. Tanabe, E. Williams and G. E. Jabbour,

Chem. Comm., 2005, 3700.

63. I. Imae, 2005, 5937, 59371N-59371N-59378.

64. M. Y. Lo, C. Zhen, M. Lauters, G. E. Jabbour and A. Sellinger, J. Am. Chem. Soc., 2007,

129, 5808-5809.

65. C.-C. Cheng, C.-H. Chien, Y.-C. Yen, Y.-S. Ye, F.-H. Ko, C.-H. Lin and F.-C. Chang, Acta

Materialia, 2009, 57, 1938-1946.

66. M. Y. Lo, K. Ueno, H. Tanabe and A. Sellinger, The Chemical Record, 2006, 6, 157-168.

36

36

67. S. Nowy, B. C. Krummacher, J. r. Frischeisen, N. A. Reinke and W. Brütting, J. Appl. Phys.,

2008, 104, 123109.

68. H.-J. Cho, D.-H. Hwang, J.-I. Lee, Y.-K. Jung, J.-H. Park, J. Lee, S.-K. Lee and H.-K. Shim,

Chem. Mater., 2006, 18, 3780-3787.

69. J. D. Froehlich, R. Young, T. Nakamura, Y. Ohmori, S. Li and A. Mochizuki, Chem. Mater,

2007, 19, 4991-4997.

70. K. Chen, Y. Chang, S. Yang and C. Hsu, Thin Solid Films, 2006, 514, 103-109.

71. X. Yang, J. D. Froehlich, H. S. Chae, S. Li, A. Mochizuki and G. E. Jabbour, Adv. Funct.

Mat., 2009, 19, 2623-2629.

72. X. Yang, J. D. Froehlich, H. S. Chae, B. T. Harding, S. Li, A. Mochizuki and G. E. Jabbour,

Chem. Mater., 2010, 22, 4776-4782.

73. X. Wang, S. Guang, H. Xu, X. Su and N. Lin, J. Mat. Chem., 2011, 21, 12941-12948.

74. M. Liras, M. Pintado-Sierra, F. Amat-Guerri and R. Sastre, J. Mat. Chem., 2011, 21, 12803-

12811.

37

37

Chapter 3: Effect of Triarylamine Structure on the Photoinduced Electron Transfer to Boron Subphthalocyanine


This chapter contains a paper published in the Journal of Physical Chemistry C.

Brett A. Kamino, Graham E. Morse, Timothy P. Bender, 2011, Journal of Physical Chemistry

C, 115 (42), 20716-20723. Please note that the experimental section and supplemental information can be found in the appendices in section 12.1. Figure and Schemes are reprinted with permission. Copyright 2011 American Chemical Society

In this first chapter, we investigate how triarylamines interact electronically with an electron

acceptor, boron subphthalocyanine (BsubPc). BsubPc was chosen because it is actively being

studied in our laboratory as an electron acceptor in organic photovoltaics. This work was

originally performed in order to develop optimized photovoltaic devices using these two classes

of materials. To help with this process we needed to understand how triarylamines and BsubPcs

work together using basic photochemical techniques. Specifically, we were interested in the

photoinduced electron transfer reaction between these two materials, a key step in the generation

of power from an organic photovoltaic device.

More broadly, this paper gave me an opportunity to explore how to tune the electrochemical

properties of triarylamines over a wide range. This exploration contrasts somewhat to the bulk of

my thesis in that we were not interested in physical state of the triarylamines. Because of this,

I’ve opted to include it as the first research chapter of my thesis to illustrate how triarylamines

can be tuned over very wide range electrochemical properties. The ability to precisely control

triarylamine electrochemistry learned in this paper was applied over the remainder of my thesis

when trying to develop novel materials with unique physical properties. As well, it provides

some interesting design criteria and thoughts on how silicone-hybridized triarylamines might

eventually be incorporated into future organic photovoltaic devices.

38

38


The authorship of this paper is as follows: Brett A. Kamino, Graham Morse, Timothy P. Bender.

Graham Morse contributed by performing fluorescence quenching experiments on a number of

off-the-shelf triarylamines (3-1a, 3-1e-i). The remainder of the fluorescence quenching

experiments was performed by me. All synthesis and characterization of new materials, and use

of theoretical framework was performed by me. The article was entirely written by me with input

from Prof. Bender as corresponding author.

3.3 Paper

3.3.1 Abstract

The photoinduced electron transfer (PET) reaction between a phenoxy-boronsubphthalocyanine

derivative and a series of triarylamine electron donors was investigated. A series of triarylamines

ranging in oxidation potentials and number of redox centers was prepared to study the effect of

triarylamine structure on the photoinduced electron transfer (PET) reaction. In the case of

multiple nitrogen centers, the triarylamines were dendritic in nature and were synthesized by a

convergent strategy relying on successive C-N coupling and thermolytic deprotection steps. The

efficiency of the PET reaction was found to be exponentially dependent on the oxidation

potential of the triarylamine beyond a certain threshold. The free-energy change of the PET

reaction was estimated using the Rehm-Weller equation, and this framework could be used to

adequately explain the observed behavior of the system. We have concluded that the specific

structure of the triarylamine is not important in the PET reaction and that efficiency of electron

transfer is almost solely dictated by the oxidation potential of the triarylamine donor.

3.3.2 Introduction

Triarylamines are a very important class of materials for a number organic electronic

applications (see Chapter 1.3). Before we begin to explore how to tune and alter their physical

properties, more work is needed on understanding how their electronic properties can be altered.

In achieving this goal we are also granted the opportunity to study how different triarylamines

may interact electronically with other useful classes of material. Despite all of the work

performed on producing new triarylamine structures, there has been comparatively little done to

understand the effects of specific triarylamine molecular structures and their associated

39

39

substituents on electron transfer processes with complementary materials. One very important

factor affecting the ultimate performance of these devices is how well the triarylamines are able

to donate an electron into a complementary acceptor material upon photoexcitation of the

acceptor. Such an interaction can be an important factor in optimizing the charge separation and

charge extraction processes in an organic solar cell (for example) and thus improving device

efficiencies1 for a selected group or pairing of materials.2

In this paper, we study the effect of triarylamine chemical structure on photoinduced electron

transfer efficiencies to a light-absorbing electron acceptor. This was done by studying the

fluorescence quenching in solution of a model fluorescent electron acceptor with various

triarylamines acting as electron donors. For the fluorophore and electron acceptor, a soluble

boron subphthalocyanine (BsubPc) derivative was chosen: 3,4-

dimethylphenoxyboronsubphthalocyanine (3,4-DMPhO-BsubPc, Figure 3-1).3 Beyond its

pleasing magenta color, this acceptor was chosen because BsubPc derivatives are currently of

interest for application in both organic photovoltaics4 and organic light emitting diode devices.5

As well, the established position of its HOMO allows a wide range of triarylamine donors to be

used as fluorescent quenchers. While chloroboronsubphthalocyanine (Cl-BsubPc) is typically

used as an electron-donating material, recent studies have shown the potential of phenoxy-

substituted BsubPc derivatives to act as electron acceptors/n-type charge transporting materials.6

A series of triarylamines that spanned both a range of oxidation potentials and a variety of

conjugated molecular structures were paired with 3,4-DMPhO-BsubPc. Obtaining triarylamines

that include one or two nitrogen centers was facilitated by our previous work in the area and by

the straightforward synthesis of triarylamines with two nitrogen centers (such molecules are

commonly known in the literature). However, to access triarylamines with a higher number of

nitrogen centers and as a consequence very low oxidation potentials, we purposefully

synthesized dendritic triarylamines for this study. Such dendrimers possess a high degree of

conjugation and associated charge stabilization while maintaining reasonable solubility due to

their nonplanar structures. These unique attributes result in materials that contain very small

energy gradients7 and stable electrochemistry.8 They have been studied as models for charge

transfer9 as well as the generation of high-spin polycations.10 By synthesizing and adding these

40

40

dendritic structures to a series of more conventional structures, we hoped to better understand the

range of structural and electronic properties that may affect the photoinduced electron transfer

reaction between a triarylamine and 3,4-DMPhO-BsubPc.

3.3.3 Results and Discussion

A series of triarylamine donors were used or purposefully prepared to study their ability to act as

electron donors to 3,4-DMPhO-BsubPc upon photoexcitation. This series incorporates single

triarylamines(3-1a-i, Figure 3-1) bearing various combinations of electron-donating groups as

well as two nitrogen-centered triarylamines which are constructed with different molecular

fragments separating the two nitrogen centers (3-2a-c, Figure 3-1).

NNN

NNN

OMe

OMeMeO

MeOOMeOMe

OMeMeOOMe

OMeMeO

MeO

N

OMeMeO

OMe

N

CH3MeO

OMe

CH3

N

CH3H3C

OMe

CH3CH3

N

CH3H3C

CH3

CH3CH3

2a 2b 2c

N

CH3

CH3CH3

H3C

N

CH3H3C

CH3

CH3

N

CH3H3C

F

CH3

N

F

F

F

1a 1b 1c 1d

1e 1f 1g 1i

N

CH3H3C

CH3

1h

CH3 CH3

F

3,4-DMPhO-BsubPc

O

N

N

N

N NN

B

CH3

CH3

Figure 3-1: Structures of 3,4-DMPhO-BsubPc and the triarylamines used in this study (containing either one (3-1a-i) or two nitrogen centers (3-2a-c)).

The chemistry utilized to synthesize single or two-nitrogen centered triarylamines relies on either

Buchwald-Hartwig11 or Ullman amination to construct the necessary C-N linkages. The detailed

syntheses of compounds 3-1a-i are described elsewhere (Figure 3-1).12 Triarylamines with two

nitrogen centers (3-2a-c) were synthesized in a single step using conventional methods. Their

3-1a 3-1b 3-1c 3-1d

3-1e 3-1f 3-1g 3-1h 3-1i

3-2a 3-2b 3-2c

41

41

synthesis is illustrated in the Supporting Information accompanying this article (Figure S3-1). To

gain access to triarylamines containing multiple nitrogen centers as well as materials with very

low oxidation potentials, dendritic triarylamines were synthesized. The dendrimers were

prepared in a convergent approach utilizing an alternating C-N coupling and deprotection

sequence to systematically increase generation size (Scheme 3-1).

Scheme 3-1: Synthetic pathway towards triarylamine dendrimers (3-5b and 3-6b). Conditions (i) sodium tert-butoxide, bis(dibenzylideneacetone)palladium or palladium acetate (see experimental), tri-tert-butylphosphine, toluene, reflux. (ii) 1,2,3,4-tetrahydronaphthalene, 200 °C, overnight.

Catalytic hydrogenolysis of benzyl8 or diphenyl methyl7 protecting groups or the acid promoted

cleavage of a t-butylcarbamate (tBOC) group by trifluoroacetic acid have been used elsewhere to

deprotect diphenylamine moieties.9,10 We however, developed a simplified deprotection strategy

which avoided hydrogenolysis and the use of strong organic acids. This was necessitated by our

observation that the use of trifluoroacetic acid, which while considered to be a very weakly

oxidizing acid compared to mineral acids such as H2SO4 or HNO3 resulted in the oxidation of the

triarylamine dendrimers as evidenced by a bright green color present in the reaction medium

characteristic of the radical cation. We found that removal of the tBOC group was facilitated by

simple heating (thermolysis) at 200 °C.13 This reaction proceeds cleanly resulting in no

42

42

observable oxidized product and is performed best when a small amount of a high boiling

solvent (in this case tetralin) is added to help melt the substrate and lower the viscosity of the

melt. Using this synthetic strategy, dendrons up to the second generation were prepared in good

yields. Attempts to produce dendrons of a higher generation by this strategy resulted in partial

substitution, a limitation that is likely a result of steric factors. For each generation dendron, the

free amine was finally capped with a p-tolyl group resulting in dendrimers 3-5b and 3-6b. Both

dendrimers were isolated as fine white powders, and their purity and composition were

established by 1H NMR, mass spectroscopy, and size exclusion chromatography (SEC). 1H

NMR analysis of both the free amine and p-tolyl-capped dendrimers (3-5a-b and 3-6a-b,

respectively) resulted in many overlapping resonances in the aromatic region of the spectra. We

were unable to acquire satisfactory 13C NMR spectra for all but compound 3-6b. Solubilities in

benzene-d6 were too low to achieve a good signal-to-noise ratio. Attempts to obtain spectra in

chlorinated solvents (chloroform-d and dichloromethane-d2) quickly resulted in oxidation of the

dendrimers as evidenced by the evolution of a characteristic bright green color which is

attributable to the presence of the radical cation. The presence of the paramagnetic radical

cations prevented NMR analysis in these solvents. The dendrimers were run through a low

molecular weight size exclusion column, and the chromatograms show well-resolved peaks for

each molecule and illustrate the progression of molecular size between each generation as well as

the purity of each (Figure 3-2).

43

43

Figure 3-2: Size exclusion chromatograms of the triarylamine dendrimers (3-5a and 3-6a) and a

single triarylamine analogue (3-1a) as detected by UV-Vis absorbance.

With this broad series of triarylamines in hand, each previously unreported compound was

characterized by solution cyclic voltammetry to determine their relative oxidation potentials

(Table 3-1). The electrochemistry was performed in a dichloromethane solution with

tetrabutylammonium perchlorate as the supporting electrolyte at a scan rate of 50 mV/s. A

platinum disk working electrode, platinum wire counter electrode, and saturated Ag/AgCl

pseudo reference electrode were also used. A small amount of decamethylferrocene was added to

each sample as an internal standard. All oxidation potentials are corrected to the accepted half

wave oxidation potential of decamethylferrocene (-0.012 V vs. Ag/AgCl).13 Several

triarylamines used in this study (3-1a-i) have been previously characterized under identical

conditions, and the literature values were incorporated into our data set. All of the studied

triarylamines displayed at least a single well-defined and reversible oxidation. This allowed for

an accurate determination of the half-wave oxidation potentials of the entire series. With respect

to the range of half-wave oxidation potentials, triarylamines with single nitrogen centers ranged

3-1a

3-5b

3-6b

44

44

from 1138 to 654 mV vs. Ag/AgCl, while the two nitrogen-centered amines ranged from 693

mV for the weakly conjugated 3-2b to 417 mV vs. Ag/AgCl for the more conjugated 3-2a. The

triarylamine dendrimers extended this range of half-wave oxidation potentials down to 273 mV

for the second-generation dendrimer (3-6b), whereas the first-generation dendrimer (3-5b) was

found to have an oxidation potential of 471 mV vs. Ag/AgCl. The first generation dendrimer (3-

5b) was found to undergo two fully reversible one-electron oxidations which is consistent with

similar structures in the literature (Figure 3-3).9 Increasing the potential further resulted in a third

irreversible oxidation. The second-generation dendrimer (3-6b) undergoes three distinct and

reversible oxidation events (Figure 3-3) likely attributable to the large number of conjugated

nitrogen centers present in this molecule. Full voltammagrams for 3-2a-c, 3-5b, and 3-6b are

illustrated in Figures S3-2 and S3-3.

45

45

Table 3-2: Electrochemical oxidation potentials, fluorescence quenching efficiency, and free energy change upon photoinduced electron transfer reaction with 3,4-DMPhO-BsubPc for triarylamines (3-1a-i, 3-2a-c, 3-5b and 3-6b).

Compound

Number of

Redox

Centers

E1ox (mV vs.

Ag/AgCl) K (mol-1) ∆G∆G∆G∆GPETPETPETPET (eV)(eV)(eV)(eV)

3-1a 1 820 10.0 ± 1.82 -0.098

3-1b 1 735 14.4 ± 0.48 -0.187

3-1c 1 690 18.7 ± 0.80 -0.228

3-1d 1 654 23.4 ± 1.16 -0.260

3-1e 1 814 5.2 ± 0.21 -0.127

3-1f 1 844 4.9 ± 0.15 -0.132

3-1g 1 912 2.8 ± 0.23 -0.064

3-1h 1 981 0.7 ± 0.07 0.005

3-1i 1 1138 0* 0.045

3-2a 2 417 71.4 ± 4.28 -0.569

3-2b 2 693 18.4 ± 0.52 -0.284

3-2c 2 643 26.4 ± 0.17 -0.333

3-5b 3 471 44.6 ± 5.56 -0.561

3-6b 7 273 187.7 ± 6.89 -0.843

* Value obtained is less than experimental error, considered to be 0.

46

46

Figure 3-3: Solution electrochemistry of triarylamine dendrimers (3-5b, 3-6b) and a representative single triarylamine (3-1a). Voltammagrams are corrected to the internal standard decamethylferrocene (E1/2,red = -0.012 V vs. Ag/AgCl).

The ability of the complete series of triarylamines to undergo an electron transfer event with a

photoexcited BsubPc was investigated using a standard steady state fluorescence quenching

experiment in solution. The BsubPc (3,4-DMPhO-BsubPc) was prepared as a 1x 10-5 M solution

in anhydrous toluene having varying amounts of triarylamine present. The solution was

photoexcited at 550 nm, and the intensity of the emission peak at 578 nm was recorded. The

relative efficiency of the photoinduced electron transfer reaction was determined by the Stern-

Volmer (equation 3-1)

][10 QKF

F⋅+= (3-1)

where F0 is the fluorescence intensity without any quencher added; F is the fluorescence intensity

at quencher concentration [Q]; and K is the quenching coefficient. For each triarylamine, a plot

3-1a

3-5b

3-6b

47

47

of F0/F vs. [Q] (a Stern-Volmer plot) was found to be linear and gave a good correlation when

fitted with a linear regression. The fluorescence quenching efficiencies (K) are summarized in

Table 3-1. The quenching coefficients range from those where no significant quenching was

observed (3-1i) to relatively large quenching coefficients (3-6b). To determine the effect of

triarylamine structure and oxidation potential on the photoinduced electron transfer (PET)

process, the quenching coefficients were plotted against the electrochemical half-wave oxidation

potentials (Figure 3-4). From this plot, it is apparent that there is a correlation between oxidation

potential and quenching efficiencies. Triarylamines with stronger electron-donating substituents

or lower oxidation potentials had higher quenching efficiencies than triarylamines with high

potentials. Perhaps more surprisingly, the majority of the triarylamines showed an exponential

dependence of quenching coefficient with oxidation potential. This relationship deviates only for

the triarylamines with the highest oxidation potentials. Fitting a line through the nine best

quenchers results in a particularly good fit with R2 = 0.98. As the oxidation potential increases

past 825 mV (vs. Ag/AgCl), the quenching constant quickly drops off. For example, triarylamine

1i has the highest oxidation potential and did not quench 3,4-DMPhO-BsubPc significantly.

Figure 3-4: Experimentally determined Stern-Volmer constants (K) plotted against half-wave oxidation potentials (E1/2,ox) of the triarylamine donor. Error bars are not included as they are small relative to the size of the data marker point.

48

48

To better understand these observations and to confirm whether our system is behaving as a

normal photoinduced electron transfer system, the free energy change of the electron transfer

process was estimated using the Rehm-Weller equation.15 This correlation was developed for

measurements in polar solvents. Because our measurements are carried out in a nonpolar solvent

(toluene, ε = 2.38), a modified version of the Rehm-Weller equation is used to take into account

the measured redox potential differences in solvents used which are of different dielectric

constants as follows:16

+−

−+⋅+−−=∆

−+−+RR

e

RRR

eEAEDEG

DCMDAtoluene

exDCMredoxPET 21

21

41

21

21

4))()((

0

2

0

2

επεεπε (3-2)

where Eox(D) is this oxidation potential of the electron donor (triarylamine); Ered(A) is the

reduction potential of the acceptor (3,4-DMPhO-BsubPc); 3 Eex is the energy of the excitation; e

is the elementary charge; ε0 is the vacuum permittivity; εtouene and εDCM are the dielectric

constants of toluene and dichloromethane, respectively; R+ and R- are the average ion radii of the

donor and acceptor, respectively; and RDA is the average donor-acceptor distance (estimated as 6

Å).16a The average ion radii of the donor and acceptor molecules were roughly estimated using

calculated values for molecular volume assuming that the molecules are spheres. These

molecular volumes were obtained through molecular modeling calculations on each of the

molecules studied and are included in the Supporting Information (Table S3-1).17 The average

donor-acceptor distance was estimated from molecular dynamics simulations using molecular

mechanics force fields (MM+). Center to center separation distances typically varied between 5.5

and 6.5 Å; a median estimate of 6 Å was used for these calculations. The resulting ∆GPET values

(Table 3-1) were plotted against the measured Stern-Volmer constants (Figure 3-5). The plot

shows an expected trend; as the free energy change becomes more favorable, the fluorescence

quenching efficiency increases. As well, a decrease in quenching efficiency occurs near ∆GPET =

0 eV, thus confirming that our system is well behaved and fits well into this theoretical

framework. This observation is expected given the lack of thermodynamic driving force for the

PET to the photoexcited 3,4-DMPhO-BsubPc for ∆GPET > 0 eV and indicates that our system is

well behaved and can be described by the Rehm-Weller model over the range studied. Looking

at the quenching efficiencies of the materials with the most favorable energy changes, it is

interesting to note that no plateau in quenching efficiency is observed. A diffusion limited

plateau in quenching efficiency at high values of ∆GPET is predicted by Rehm-Weller theory and

49

49

is observed in most systems.18 In our case, it may simply be that no materials have a sufficiently

high enough ∆GPET value to result in diffusion limited electron transfer.

Figure 3-5: Experimentally determined Stern Volmer constant (K) plotted against the free energy change estimated by a modified Rehm-Weller equation (Eq. 3-2). Error bars are not included as they are small relative to the size of the data marker point.

When looking at this analysis, it should be emphasized that several assumptions were made in

the calculation of the Coulombic attraction energy term, particularly in regards to the ion radii

(R+ and R-) and the donor-acceptor distance (RDA). Regardless of the assumptions associated

with this calculation, several important observations can be drawn from the data. Looking first at

the effect of triarylamine structure on quenching efficiencies, there is no relation between

structure and the fluorescence quenching efficiency. Triarylamines with single or two nitrogen

centers of similar oxidation potentials have equivalent quenching efficiencies. Therefore, a

conclusion can be drawn that there is no need to synthesize dendritic triarylamines to act as

electron donors to BsubPcs. It can also be seen that once the PET reaction is energetically

favorable, the quenching efficiency very closely follows an exponential relationship with

solution oxidation potential. The one deviation from this behavior is for compound 3-1e which

has a similar oxidation potential to 3-1a but a much lower quenching efficiency. This may be

50

50

explained by the much bulkier substitution pattern (t-butyl vs methyl) intuitively suggesting that

bulky non-conjugated substituents inhibit the electron transfer process via a steric effect.

3.3.4 Conclusions

In summary, the effect of triarylamine structure and oxidation potential on the photoinduced

electron transfer (PET) reaction to 3,4-DMPhO-BsubPc was investigated. This was performed

using a series conventional triarylamines and novel triarylamine dendrimers which spanned a

large range of oxidation potentials, number of conjugated redox centers, and structures. Relying

on steady-state fluorescence quenching, it was found that two regimes of PET reactions could be

found. Under a certain oxidation potential, the quenching efficiency was found to scale

exponentially with oxidation potential of the triarylamine regardless of it chemical structure.

This implies that oxidation potential is the primary factor in determining the efficiency of

electron transfer in this system and that triarylamine structure is not particularly important for

this process. By placing our data into the Rehm-Weller theoretical framework we have

determined that this system is well behaved including the expected deviation from the

exponential dependence of quenching on the oxidation potential of the donor coinciding with a

loss of thermodynamic driving at values for ∆GPET > 0 eV. The results of this study will directly

aid in the design of triarylamines as effective electron donating materials to BsubPc derivatives.

It also suggests a general methodology by which other donor/acceptor materials can be selected

and optimized for PET.

3.3.5 References

1. Sariciftci, N. S.; Smilowitz, L.; Heeger, A. J.; Wudl, F. Science 1992, 258 (5087), 1474–1476.

2. Guchhait, A.; Pal, A. J. J. Phys. Chem. C 2010, 114, 19294 19298.

3. Morse, G. E.; Helander, M. H.; Stanwick, J.; Sauks, J. M.; Paton, A. S.; Lu, Z. H.; Bender, T. P. J. Phys. Chem. C 2011, 115, 11709–11718.

4. (a) Mutolo, K. L.; Mayo, E. I.; Rand, B. P.; Forrest, S. R.; Thompson, M. E. J. Am.

Chem. Soc. 2006, 128, 8108–8109. (b) Sullivan, P.; Durand, A.; Hancox, I.; Beaumont, N.; Mirri, G.; Tucker, J. H. R.; Hatton, R. A.; Shipman, M.; Jones, T. S. Adv. Energy

Mater. 2011, 1(3), 352-355. (c) Ma, B.; Miyamoto, Y.; Woo, C. H.; Frechet, J. M. J.; Zhang, F.; Liu, Y. Proc. SPIE 2009, 74161E.

51

51

5. Helander, M. G.; Morse, G. E.; Qiu, J.; Castrucci, J. S.; Bender, T. P.; Lu, Z. ACS Appl.

Mater. Interfaces 2010, 2, 3147–3152.

6. Morse, G. E.; Helander, M. G.; Maka, J. F.; Lu, Z.; Bender, T. P. ACS Appl. Mater.

Interfaces 2010, 2, 1934–1944.

7. (a) Ranasinghe, M. I.; Varnavski, O. P.; Pawlas, J.; Hauck, S. I.; Louie, J.; Hartwig, J. F.; Goodson, T. J. Am. Chem. Soc. 2002, 124, 6520–6521. (b) Hagedorn, K. V.; Varnavski, O.; Hartwig, J.; Goodson, T. J. Phys. Chem. C 2008, 112, 2235–2238.

8. Louie, J.; Hartwig, J. F. J. Am. Chem. Soc. 1997, 119, 11695–11696.

9. Hirao, Y.; Ito, A.; Tanaka, K. J. Phys. Chem. A 2007, 111, 2951–2956.

10. Hirao, Y.; Ino, H.; Ito, A.; Tanaka, K.; Kato, T. J. Phys. Chem. A 2006, 110, 4866–4872.

11. (a) Hartwig, J. F.; Kawatsura, M.; Hauck, S. I.; Shaughnessy, K. H.; Alcazar Roman, L. M. J. Org. Chem. 1999, 64, 5575. (b) Jiang, L.; Stephan, B. L. Palladium-Catalyzed

Aromatic Carbon-Nitrogen Bond Formation. Metal-Catalyzed Cross-Coupling

Reactions, 2nd ed.; de Meijere, A., Diederich, F., Eds.; Wiley-VCH: New York, 2004; Chapter 13.

12. Bender, T. P.; Graham, J. F.; Duff, J. M. Chem. Mater. 2001, 13, 4105–4111.

13. Rawal, V. H.; Jones, R. J.; Cava, M. P. J. Org. Chem. 1987, 52, 19.

14. Noviandri, I; Brown, K. N.; Fleming, D. S.; Fulvas, P. T.; Lay, P. A.; Masters, A. F.; Phillips, L. J. Phys. Chem. B 1999, 103, 6713.

15. Rehm, D.; Weller, A. Isr. J. Chem. 1970, 8, 259–271.

16. (a) Weller, A. Z. Phys. Chem. 1982, 133, 93–98. (b) Heitele, H.; Finckh, P.; Weeren, S.; Pöllinger, F.; Michel-Beyerle, M. E. J. Phys. Chem. 1989, 93, 5173–5179. (c) Perez, L.; Garcia-Martinez, J. C.; Diez-Barra, E.; Atienzar, P.; Garcia, H.; Rodriguez-Lopez, J.; Langa, F. Chem. Eur. J. 2006, 12, 5149–5157.

17. Molecule volumes were calculated by Spartan ’06 utilizing molecular mechanics with a MMFF force field.

18. (a) Dosche, C.; Mickler, W.; Löhmannsröben, H.-G.; Agenet, N.; Vollhardt, K. P. C. J. Photochem. Photobiol. A 2007, 188, 371–377. (b) Nad, S.; Pal, H. J. Phys. Chem. A 2000, 104, 673–680.

52

52

Chapter 4: Controlling the Physical and Electrochemical Properties of Arylamines Through the Use of Simple Silyl Ethers: Liquid, Waxy and Glassy Arylamines


This chapter was published as a full paper in Silicon.

Brett A. Kamino, Jeffrey Castrucci, Timothy P. Bender 2011, Silicon, Vol. 3, No. 3, 125-137.

Figure and Schemes are reprinted with permission. Copyright 2011 Springer.

The work described in this chapter details our very first attempts at controlling the physical

properties of triarylamines through incorporation of silicon chemistry. Specifically, we were

interested in developing methods to produce liquid organic semiconductors, a group of materials

with very antithetical properties to traditional triarylamines. In this work, we incorporated bulky

silyl ethers onto the periphery of triarylamine structures to alter their properties. These new

materials were studied by differential scanning calorimetry and solution electrochemistry.

Different silyl ether groups were explored as were different triarylamine structures. At the time,

silyl ethers seemed like an ideal group to modify physical properties. They could be easily

introduced onto common triarylamine building blocks and their large steric bulk should have

prevented strong intermolecular interactions, thus minimizing crystallization. As well, they are

not known to be conjugated or electronically active, therefore their presence shouldn`t have a

large effect on the electronic properties of the base material.

Overall, this strategy proved marginally successful. Of the series of triarylamine, only a single

compound (4-5) would be isolated as a freely flowing liquid. As it turned out, any triarylamine

compound with more than one nitrogen centre remained a solid at room temperature. Some

compounds were even found to be highly crystalline despite the presence of these bulky groups.

While this was a somewhat disappointing outcome, several important discoveries were made that

impacted the direction of this project and of my thesis. Most importantly, we found a well-

behaved and easily synthesized liquid triarylamine (4-5). This compound later goes on to be used

as a model liquid organic semiconductor for our charge transport studies (Chapter 7). Secondly,

one multi-nitrogen centered compound (4-7) was found to be soluble in hexamethyldisiloxane

53

53

(MM). This was very interesting as MM is an inert and non-toxic fluid, as well it is a non-solvent

for practically everything but silicones. This is highly interesting as it may allow the solution

deposition of this molecule on top of other soluble derivatives on the basis of orthogonal solvent

processing. At the time we attempted to build up discrete layers through solution processing.

Unfortunately, these layers were difficult to study due to the very low glass transition

temperature of the compound. Future work in this area could be used to develop more

triarylamines and organic semiconductors that are soluble in MM fluid.


The authorship of this paper is as follows: Brett A. Kamino, Jeffrey Castrucci, Timothy P.

Bender. Jeffrey Castrucci helped to synthesize and characterize four compounds (4-4a-d) as an

undergraduate student under my guidance. The remainder of the compounds were synthesized

and characterized by me. The paper was written by me and Prof. Bender as corresponding

author.

4.3 Paper

4.3.1 Abstract

The effect of silyl ether substitution on the physical and electrochemical properties of single and

two-nitrogen centered arylamines was explored. It was found that this substitution can

significantly lower the Tg and suppress crystallization of these compounds. This resulted in

arylamines that were isolated as liquids, waxes, glasses, and in some cases crystalline solids at

ambient conditions. Additionally, these silyl ether groups were found to be relatively strong

electron donating groups with similar donating potentials to the well-known methyl ether group.

It is concluded that silyl ether substitution is a synthetic handle to greatly alter the physical

properties of arylamines without substantially changing their basic electronic properties.

4.3.2 Introduction

A large number of triarylamine derivatives have been reported and the effects of certain

molecular fragments on the electronic and physical properties of triarylamines are well known.1

However, the vast majority of this effort has been applied to the study of crystalline or glassy

triarylamines. Such materials have very high glass transition temperatures (for solution or

54

54

vacuum deposition of stable amorphous films) or are highly crystalline for applications where a

polycrystalline morphology may be beneficial (OPVs, OFETs).2 On the other hand, there is

comparatively little work appearing in the literature regarding "soft" triarylamines, which we will

define as triarylamines which are liquids or waxes at ambient conditions. There are only two

examples within the literature of arylamine-type compounds that are liquids at room temperature:

tris(4-methoxyethoxyphenyl)amine (TMEPA) and N-(2-ethylhexyl)carbazole (Fig. 4-1).

N

OO

O

O

O

O

N

2-ethylhexyl carbazole

Figure 4-1 Two previously reported liquid arylamines. N-2-ethylhexylcarbazole (left) and

TMEPA (right).

TMEPA is a triarylamine with 2-methoxyethoxy groups in each of the three para-positions. It is a

viscous oil which shows both crystalline and glassy states below room temperature. TMEPA has

been applied in dye sensitized solar cells (DSSCs) showing an efficiency of 2.4% under AM1.5

illumination.3 This is in contrast to the typical use of liquid electrolytes as a hole transporting

medium in DSSCs. Concerns about leakage and long term stability of the liquid electrolyte have

prompted the search for alternatives which has included solid state arylamines.5-9 While TMEPA

shows utility, the synthetic methodology used leaves little room for synthetic control over the

physical and electronic properties through molecular variation.

The second case of a liquid arylamine, N-(2-ethylhexyl) carbazole, is an N-alkylated carbazole.

Carbazoles have been used as building blocks for main chain conducting polymers10 and small

molecule hosts for OLEDs.11 There are several studies exploring the charge transport properties

of N-(2-ethylhexyl)carbazole.12,13 Most importantly, Wada et al. have shown it to have good hole

carrier mobility (greater than poly(N-vinylcarbazole)),14 while Adachi et al. have shown it to

have utility as a liquid host material in OLEDs.15 Once again, this compound shows utility in

several applications but lacks a synthetic handle to tune its electronic and physical properties.

55

55

Given the demonstrated utility of soft arylamines, the development of new materials is necessary

to facilitate further study in this area. Towards this end, the goal of this study was to develop a

simple synthetic methodology which yields liquid, waxy, or low melting triarylamines with

tunable physical and electronic properties. Along with such a methodology, it is also desirable to

use readily accessible molecular fragments which are amenable to assembly via simple and

established chemistry. Silyl ethers where chosen due to their well-defined chemical structures

and because of the unique physical properties of organosilicones.16


Initial efforts were focused on simple triarylamines containing one nitrogen-center. A series of

three triarylamines with triisopropylsilyl ether (−OTIPS) substituents in the para position were

prepared. Each is illustrated in Scheme 4-1. The –OTIPS group was chosen initially because of

its hydrolytic stability,17,18 ease of introduction, and its lipophilic structure. Moreover, there is a

consensus that the related –OTIPS groups when added to a conjugated π- electron system by a σ-

bond to carbon is an acceptable functional group for organic electronic materials.19-20 3,4-

Dimethylphenyl substitutes were utilized on the remainder of the triarylamine molecule (in the

case of 4-4a and 4-5) in order to add an additional element of asymmetry and to allow for an

easy comparison of the physical and electronic properties to other triarylamines previously

characterized.21

56

56

NH2

i

HN

OH

Br

OR

Br

N

OR

1 2

2 + 3a-d

Si Si Si SiPh

Ph

3a (R=TIPS)3b (R=TBDMS)3c (R=THDMS)3d (R=DPTBS)

4a (R=TIPS)4b (R=TBDMS)4c (R=THDMS)4d (R=DPTBS)

TIPS = TBDMS = THDMS = DPTBS =

N

OTIPS

OTIPS

LiNH2 + 3a N

OTIPS

OTIPSTIPSO

1 + 3a

5

6

∗ ∗ ∗ ∗

ii

iii

iii

iii

Scheme 4-1: Synthesis of triarylamines containing silyl ethers (Conditions: (i) AlCl3, CaCl2, tetralin, 220 °C. (ii) R-Cl, imidazole, DMF, rt (iii) Pd(OAc)2, P(t-butyl)3, Na(t-butoxide), toluene, 110 °C.).

Our synthetic route to the targeted triarylamines uses p-(triisopropyl)silyloxy bromobenzene (4-

3a) as a simple and basic molecular building block (which was prepared by a known patented

procedure).22 To produce 4-4a, 4-3a was coupled with the diphenylamine derivative 4-2 using

palladium catalyzed Buchwald-Hartwig amination conditions.23-29 Compound 4-2 itself was

4-6

4-5

4-2

4-1

4-1 + 4-3a

4-2 + 4-3a-d

LiNH2 + 4-3a

4-4a (R=TIPS) 4-4b (R=TBDMS) 4-4c (R=THDMS)

4-4d (R=DPTBS)

4-3a (R=TIPS) 4-3b (R=TBDMS) 4-3c (R=THDMS) 4-3d (R=DPTBS)

57

57

synthesized by the self-condensation of 3,4-dimethylaniline (4-1) as adapted from the patented

procedure.30 Compound 4-5 was synthesized using the same amination conditions as for 4-4a

except 3,4-dimethylaniline was used as the starting material. Finally, for triarylamine 4-6,

lithium amide was used as a nitrogen source and was coupled with three equivalents of 4-3a to

form 4-6. In each case, syntheses were performed reliably and reproducibly on the gram and tens

of grams scales.

Table 4-1. Comparison of the physical and electrical properties of compounds 4-4a, 4-5 and 4-6

with previously reported analogous compounds.

Compound R Tg (°C) Tm (°C) E1/2 (mV vs.

Ag/AgCl)

N/A N/A 134a N/A

-OMeb 4 122 735

-OTIPS (4-4a) -14 96 732

-OMeb 1 86 690

-OTIPS (4-5) -28 N/A 692

-OMeb 8 98 654

-OTIPS (4-6) -21 N/A 659

a – Data taken from Goodbrand, H.B. and Hu, N.X. 32

b – Data taken from Bender, T.P. et al. 21 c – Half wave potential measured in acetonitrile by cyclic voltammetry.

Triarylamines 4-4a, 4-5 and 4-6 were characterized by differential scanning calorimetry (DSC)

and cyclic voltammetry (CV) to determine their physical and electronic properties. From DSC

analysis it was found that the inclusion of the –OTIPS group into these compounds significantly

58

58

lowers both the melting point (if present at all) and the Tg when compared to the analogous –

OMe substituted triarylamines (Table 4-1).21,31,32 Compound 4-4a, which is a white solid at room

temperature, was found to exhibit both crystalline and glassy states. In contrast, 4-5 is a room

temperature liquid having a viscosity similar to a PDMS standard with a weight averaged

molecular weight of 79,100 g/mol and a glass transition temperature of −28 °C. Triarylamine 4-6

appears to the eye as a waxy non-flowing solid at room temperature and has a glass transition

temperature of −21 °C. When gently heated with a heat gun, 4-6 begins to flow. Neither 4-5 nor

4-6 showed recrystallization or other phase changes except a Tg during heating or cooling.

Inclusion of the –OMe group has little effect on the physical properties of the resulting

triarylamine26 while inclusion of the –OTIPS group has a marked effect on the physical

properties. For example, the change in melting point from tris(3,4-dimethylphenyl)amine to

bis(3,4-dimethylphenyl)- 4-methoxyphenyl amine is a decrease of 11 °C, whereas from the same

to 4-4a, the melting point is decreased by 37 °C. Cyclic voltammetry measurements show that

each compound undergoes a reversible one-electron oxidation between 600 mV and 700 mV (vs.

Ag/AgCl). This analysis is summarized in Table 4-1. Interestingly, the half wave oxidation

potentials of the three compounds decreases as the number of –OTIPS groups on the triarylamine

are increased. This indicates that the –OTIPS group is electron donating. While it may be

generally understood that silyl ethers are electron donating groups, no direct comparison to

analogous compounds has been made to determine the magnitude of electronic donation into π-

conjugated systems of this type. The electrochemistry of the direct methoxy analogues is

available in the literature.21 A direct comparison between the two sets of molecules reveals that

the –OTIPS group has comparable electron donating ability to the more widely used –OMe

group.

Triarylamines 4-4a, 4-5 and 4-6 all have high solubility in a variety of solvents including non-

polar, aromatic and chlorinated solvents. For example, 4-5 and 4-6 are miscible in chlorinated

solvents (DCM, CHCl3, and chlorobenzenes) as well as common non polar solvents (petroleum

ethers, ethers) and aromatic solvents (toluene, benzene, chlorobenzene). Perhaps somewhat

surprisingly, 4-5 and 4-6 are also practically miscible with hexamethyldisiloxane, silicone fluids

and commercial mixtures of cyclic dimethylsiloxanes. These liquids that are generally not

considered solvents in the field of organic electronics. 4-4a also showed limited solubility in

59

59

hexamethyldisiloxane, solutions up to 4 wt. % could be prepared by gently heating the

triarylamine in hexamethyldisiloxane.

One additional parameter which may be relevant to the application of silyl ether containing

triarylamines is their hydrolytic stability, something which is not an issue with methoxy groups.

Due to their widespread use in protecting group chemistry, other silyl chlorides with a variety of

substituents are readily available and their relative hydrolytic stability is documented. However it

is not known how that knowledge translates to their incorporation into triarylamines. Moreover,

their incorporation may have a different effect on the final physical electronic and physical

properties depending on the structure of the silyl ether.

In order to study these considerations, a series of singly substituted triarylamines with different

silyl ether groups were synthesized. Singly substituted triarylamines were chosen to simplify the

analysis of the hydrolytic stability as only one product would be formed and the hydrolysis

process would be expected to exhibit simple kinetic behaviour (first order). The silyl ethers

chosen for this study were trimethylsiloxy (−OSiMe3), tert-butyldimethylsiloxy (−OSitBuMe2),

tert-hexyldimethylsiloxy (−OSitHexylMe2), and diphenyl-tert-butylsiloxy (−OSitBuPh2) groups.

These different silyl ethers are all commonly used in protecting group chemistry and are known

to span across a range of hydrolytic stabilities.17,18,33 All of the reactions to produce the resulting

triarylamines (4-4b-d, Scheme 4-1) and their intermediates proceeded smoothly except the

coupling of diarylamine 4-2 with the trimethylsilyl ether of 4-bromophenol (not shown). This

reaction did not yield the desired product under a variety of conditions.

Triarylamines 4-4b and 4-4c were isolated as white crystalline powders, similar to 4a in

appearance. DSC analysis shows that 4-4b and 4-4c differ in Tm and Tg with both products

having a higher Tg than 4-4a and a Tm that straddles that of 4-4a. Interestingly, 4-4d was isolated

as a non-crystalline glass which was optically transparent and began to flow upon gentle heating.

DSC analysis showed only a glass transition indicating that this compound forms an amorphous

glass at room temperature. This difference in the observed physical states is likely a consequence

of the larger nature of the –OSitBuPh2 group. Very large aryl substituents are known to increase

the glass transition temperature of triarylamines.34 Finally, comparing the electrochemical data

between the different silyl ether groups, it can be seen that there is little effect on the oxidation

60

60

potential of the compounds. Each compound showed reversible electrochemical oxidation events

that occurred at approximately the same half wave potential as 4-4a (Table 4-2).

Table 4-2. Comparison of the physical and electrical properties of compounds 4-4a-d including hydrolytic stability.

O

N

R

Si Si Si Si

Ph

Ph

TIPS = TBDMS = THDMS = DPTBS =

∗ ∗ ∗ ∗4a 4b 4c 4dR =

Compound Tg (°C) Tm (°C) E1/2 (mV vs. Ag/AgCl)

Acidic Hydrolysis

Halflife (hrs)

Basic Hydrolysis

Halflife (hrs) 4-4a -14a 96a 732 15 260 4-4b 2 86 731 13 34 4-4c -8 100 743 53 200 4-4d 28 N/A 733 770 130

a Repeated from Table 4-1.

In order to evaluate hydrolytic stability a 0.1 mmol solution of each triarylamine (4-4a-d) was

prepared in 95 vol% 1,4-dioxane/5 vol% 1 M HCl or 10 M NaOH aqueous solution. Each sample

was then stirred at room temperature in a sealed vial in the dark and sampled at regular intervals.

It is important to note that the hydrolysis experiments were conducted not as a gauge of real

world stability, but as a metric to confirm expected trends in silyl ether hydrolytic stability. The

triarylamines hydrolyzed exponentially with the expected first order kinetics allowing their half-

lives to be extracted. The results of which are summarized in Table 4-2. Under the acidic test

conditions it was observed that the hydrolytic stability, as measured by the half-life of 4-4a-d,

varies according to: –OSitBuPh2 >> −OSitHexylMe2 > −OTIPS ≈ −OSitBuMe2 and under basic

conditions –OTIPS >> −OSitHexylMe2 > −OSitBuPh2 > −OSitBuMe2. Neither of these trends is

consistent with that which is observed for the analogous phenyl ethers.33 The reason for these

disparities is unclear.

Encouraged by the ability of the –OTIPS group to prevent crystallization and effectively lower

the glass transition temperatures of simple triarylamines, the synthetic strategy was extended to

multi-nitrogen centered arylamines. These molecules typically have higher hole transport

4-4b 4-4c 4-4d 4-4a

61

61

mobilities and more utility in organic electronic devices.2 Taking all of the results outlined above

into account, we felt that the –OTIPS group had the best balance of lowering the Tg and Tm while

maintaining hydrolytic stability. While the very bulky –OSitBuPh2 group prevented

crystallization on the mono substituted triarylamine and possessed good hydrolytic stability, we

were concerned about the relatively high Tg that this group imparted on 4-4d (as compared with

–OTIPS on 4-4a). We believe this would have the effect of raising Tg proportionally if this group

were applied to larger structures as it is has been documented that very bulky substituents on

triarylamine groups can enhance glass formation and raise the Tg value.34

Two well-known classes of two-nitrogen centred arylamines were targeted: those based on the

phenylene diamine (4-7 and 4-8a) and benzidine cores (4-9, 4-10). The synthesis of these

compounds again relied on the use of p-triisopropylsiloxy bromobenzene (4-3a) as a common

reagent (Scheme 4-2). All of these materials with exception of the 4-11 were prepared in a single

step and all were isolated using column chromatography in reasonable yields. Compound 4-11

was prepared in two steps by the selective amination of 4,4`-dibromobiphenyl with excess 3,4-

dimethylaniline followed by further coupling with 4-3a under Buchwald-Hartwig conditions.

62

62

i

NH2

NH2

OTIPS

Br

N N

OTIPS

TIPSO

OTIPS

OTIPS

7

NH2

NH2

OR

Br

N N

RO

RO OR

OR

8a (R = TIPS)8b (R = DPTBS)

i

NH2

NH2

OTIPS

Br

N N

TIPSO

TIPSO OTIPS

OTIPS

9

i

N N

TIPSO OTIPS

11

NHHN

Br

Br

NH3 i

ii

OTIPS

Br

Scheme 4-2 Synthesis of silyl ether containing arylamines with multiple nitrogen centres.

Conditions: (i) Pd(dba)2, P(t-butyl)3, Na(t-butoxide), toluene, 110 °C. (ii) Pd(OAc)2, P(t-butyl)3,

Na(t-butoxide), toluene, 110°C.

4-11

4-9

4-8a (R = TIPS) 4-8b (R = DPTBS)

4-7

63

63

Unlike the triarylamine series, none of the target compounds were isolated as free flowing

liquids at room temperature. Instead it was found that the molecules were either purely

amorphous materials or highly crystalline solids. Both compounds 4-7 and 4-11 formed optically

clear glasses when isolated. Neither of these materials showed crystallization over 2 months

stored at room temperature. DSC analysis only showed glass transitions on heating and cooling,

no other thermal transitions were observed. In contrast, compound 4-9 was isolated as a hard

white powder. DSC showed a melting point upon the first heating cycle at 83 °C. A strong

complementary crystallization peak was observed while the sample was rapidly cooled to −50

°C. Despite this, upon the second heating cycle a weak glass transition was detected at 29 °C,

followed by a melting point at 83 °C. Compound 4-8a was isolated as soft crystalline solid. DSC

did not detect any melting transitions up to 220 °C (the upper limit of our DSC) on the first

heating cycle, while the second heating cycle revealed an endothermic peak at 11 °C. A

complementary exothermic peak at 2 °C was detected on the cooling cycle. Since no thermal

activity was noticed during the first heating scan, which rules out compound degradation, it can

be assumed that this extra set is due to another polymorph of the compound.

The non-crystalline compounds (4-7 and 4-11) both showed excellent solubility in all non-polar

organic solvents. Compound 4-11 was found to be sparingly soluble in hexamethyldisiloxane

while 4-7 was found to be miscible with hexamethyldisiloxane. The crystalline members of this

family (4-8a and 4-9) showed more limited solubility with neither compound being soluble in

hexamethyldisiloxane while remaining soluble in aromatic or chlorinated solvents.

64

64

Table 4-3. Characterization data for multi-nitrogen centred triarylamine series, compounds 4-7,

4-8a-b, 4-9 and 4-11 (see Scheme 4-1 for chemical structures of TIPS and TBDPS).

Compound Tg

(°C)

Tm

(°C)

E1/2 (mV vs

Ag/AgCl)

4-7

N N

OTIPS

TIPSO

OTIPS

OTIPS

23 N/A 702, 986

4-8a

N N

TIPSO

TIPSO OTIPS

OTIPS

N/A 11 421, 868

4-8b

N N

TBDPSO

TBDPSO OTBDPS

OTBDPS

60 159 451, 908

4-9

N N

TIPSO

TIPSO OTIPS

OTIPS

29 83 638, 865

4-11

N N

TIPSO OTIPS

50 N/A 661, 886

65

65

On examination of this series it is observed that the physical properties are strongly symmetry

dependent. Both 4-8a and 4-9 show crystalline domains in the solid state occurring at fairly low

temperatures. While the less symmetric 4-7 and 4-11 form optically transparent glasses upon

heating and cooling. This series demonstrates that the general strategy to use the bulky –OTIPS

group to impede crystallization and impart liquidity fails when moving to larger molecules.

From the experimentation with other silyl ether groups outlined above, it is known that the –

OSitBuPh2 group was sufficiently bulky to disrupt crystallization of a singly substituted

triarylamine (4-4d) while simultaneously having no effect on its oxidation potential. It was

decided to test to see if the –OSitBuPh2 group would be bulky enough to disrupt crystallization

of an analogous compound to the highly crystalline 4-8a. Thus compound 4-8b was synthesized

using a similar synthetic pathway to 4-8a in moderate yield and isolated as small white needles.

CV analysis showed similar electrochemical behaviour as 4-8a while DSC analysis revealed very

complex phase transition behaviour. During the first heating cycle multiple endothermic and

exothermic transitions were observed. Upon the second heating curve only a single glass

transition was found which had a complementarily event on cooling curve. No further

crystallization or melting was observed during the remainder of the analysis. The appearance of a

glass transition temperature during DSC analysis indicates that the use of bulkier silyl ether

substitutes is partially successful at inhibiting crystallization.

4.3.4 Conclusions

In summary, triarylamines functionalized with silyl ethers were synthesized and studied. It was

found that incorporation of –OTIPS group(s) into a triarylamine could drastically lower the glass

transition temperature and inhibit crystallization of the resulting compound. Electrochemical

characterization of these new materials showed that this substitution had a very minor effect on

oxidation behaviour of the compound. When compared to other common silyl ether groups, the –

OTIPS group was found to have the greatest effect on the physical properties of the resulting

material while maintaining moderate hydrolytic stability. When this strategy was extended to

larger multi-nitrogen centered triarylamines, the effect of molecular symmetry was found to be

dominant over the effect of –OTIPS substitution. For less symmetric substitution patterns,

morphologically stable amorphous glasses could be obtained. While for more symmetric

substitution patterns, highly crystalline materials were obtained. These unique triarylamines have

66

66

been made using simple and readily available chemical building blocks and were assembled

using well established chemistry. This study demonstrates the utility of simple silyl ethers in

controlling or altering the physical properties of triarylamines without significantly changing

their useful electronic attributes.

4.3.5 References

1. Thelakkat, M. Macromol. Mater. Eng. 2002, 287, 442–461.

2. Shirota, Y.; Kageyama, H. Chem. Rev. 2007, 107, 953–1010.

3. Snaith, H.J.; Zakeeruddin, S.M.; Wang, Q.; Péechy, P.; Grätzel, M. Nano. Lett. 2006, 6,

2000.

4. Bach, U.; Lupo, D.; Compte, P.; Moser, J.E; Weissörtel, F.; Salbeck, J.; Spreitzer, H.;

Grätzel, M. Nature 1998, 395, 583.

5. Yum, J.; Chen, P.; Grätzel, M.; Nazeeruddin, M.K. Chem. Sus. Chem. 2008, 1, 699.

6. Salbeck, J.; Bauer, J.; Weissörtel, F.; Bestgen, H. Synth. Met. 1997, 91, 209.

7. Snaith, H.J.; Moule, A.J.; Klein, C.; Meerholz, K.; Friend, R.H.; Grätzel, M. Nano. Lett.

2007, 7, 3372.

8. Ding, I-.K.; Tétreault, N.; Brillet, J.; Hardin, B.E.; Smith, E.H.; Rosenthal, S.J.; Sauvage,

F.; Grätzel, M.; McGehee, M.D. Adv. Func. Mater. 2009, 19, 2431.

9. Snaith, H.J., Humphry-Baker, R.; Chen, P.; Cesar, O.; Zakeeruddin, S.M.; Grätzel, M.

Nanotechnology 2008, 424003.

10. Blouin, N.; Leclerc, M. Acc. Chem. Res. 2008, 41, 1110.

11. Brunner, K.; Dijken, A.; Börner, H.; Bastiannsen, J.J.A.M.; Kiggen, N.M.M.; Langeveld,

B.M.W J. Am. Chem. Soc.2004, 126, 6035.

12. Hecdrickx, E.; Guenther, B.D.; Zhang, Y.; Wang, J.F.; Staub, K.; Zhang, W.; Marder,

S.R.; Kippelen, B.; Peyghambarian, N. Chem. Phys. 1999, 245, 407.

13. Ribierre, J.; Aoyama, T.; Kobayashi, T.; Sassa, T.; Muto, T.; Wada, T. J. Appl. Phys.

2007, 102, 033106.

14. Ribierre, J.C.; Aoyama, T.; Muto, T.; Imase, Y.; Wada, T. Org. Electr. 2008, 9, 396.

15. Xu, D.; Adachi, C.; Appl. Phys. Lett. 2009, 95, 053304.

16. Noviadri, I.; Brown, K.N.; Flemings, D.S; Gulyas, P.T.; Lay, P.A.; Masters, A.F.;

Phillips, L. J. Phys. Chem. B 1999, 103, 6713.

67

67

17. Brook, M.A. Silicon in Organic, Organometallic, and Polymer Chemistry, Wiley, New

York, NY, 2000.

18. Wuts, P.G.; Greene, T.W. Greene’s Protective Groups in Organic Synthesis 4th Ed.,

Wiley, Hoboken, NJ, 2007.

19. Anthony, J. Angew. Chem. Int. Ed. 2008, 47, 452.

20. Lehnherr, D.; Gao, J.; Hegmann, F.A.; Tywinsky, R. Org. Lett. 2008, 10, 4779.

21. Bender, T.P.; Graham, J.F.; Duff, J.M.; Chem. Mater. 2001, 13, 4105.

22. Manley, P.J.; Miller, W.H.; Uzinskas, I.N. Vintronectin Receptor Antagonists, European

Patent, 1218005, 2004

23. Hartwig, J.F.; Kawatsura, M.; Hauck, S.I.; Shaughnessy, K.H.; Alcazar Roman, L.M. J.

Org. Chem. 1999, 64, 5575.

24. Jiang, L.; Buchwald, S.L. Palladium-Catalyzed Aromatic Carbon-Nitrogen Bond

Formation, In: de Meijere, A.; Diedereich, F. (Editors) Metal Catalyzed Cross-Coupling

Reactions 2nd

Ed.; Wiley-VCH, NY, 2004

25. Bender, T.P.; Coggan, J.A.; McGuire, G.; Murphy, L.D.; Toth, A.E.J., US Pat. 7408085,

2008.

26. Bender, T.P.; Coggan, J.A. US Pat. 7402700, 2008.

27. Bender, T.P.; Goodbrand, H.B.; Hu, N.X. US Pat. 7402699, 2008.

28. Bender, T.P. Coggan, J.A. US Pat. 7345203, 2008.

29. Coggan, J.A.; Bender, T.P., US Pat. 7332630, 2008.

30. Shintou, T.; Fujii, S.; Kubo, S., US Pat. 6218576, 2001.

31. DSC plots showing temperature heat flow profiles are included in the supporting

information (Section 9.2.2)

32. Goodbrand, H.B.; Hu, N. J. Org. Chem. 1999, 64, 670.

33. Davies, J.S.; Higginbotham, C.L.; Tremeer, E.J.; Brown, C.; Treadgold, R.C. J. Chem.

Soc. Perkin Trans. 1992, 1, 3043.

34. Gagnon, E.; Maris, T.; Weust, J.D. Org. Lett. 2010, 12, 404.

68

68

Chapter 5: Siloxane-Triarylamine Hybrids: Discrete Room Temperature Liquid Triarylamines via the Piers-Rubinsztajn Reaction


Sections of this chapter have been published as a letter in Organic Letters as well as a full paper

in the Journal of Organic Chemistry:

Brett A. Kamino, John B. Grande, Mike Brook, Timothy P. Bender (2011), Organic Letters, Vol. 13, No. 1, 154-157. Figure and Schemes are reprinted with permission. Copyright 2011 American Chemical Society This chapter is a conceptual follow up to the work discussed in chapter 4. In chapter 4, the use

of bulky silyl ether groups was found to be ineffective in transforming normally crystalline

triarylamines into liquids at room temperature. One improvement we could think of at the time

was to use oligosiloxanes instead of bulky silyl ether groups. Such oligosiloxanes would have a

great deal more conformational freedom then bulky silyl ethers. And, we hypothesized that this

would be more effective in producing liquid organic semiconductors. Unfortunately, these

groups are somewhat more sensitive to chemical reactions and did not survive the catalytic

coupling reactions used to synthesize triarylamines. In this chapter, we disclose a method where

triarylamines are synthesized with stable –OMe groups and then functionalized with an

oligosiloxane using the Piers-Rubinsztajn chemistry. This unique chemistry is crucial in the

synthesis of these materials and its use on organic functional materials is very novel.


The authorship of the first paper is as follows: Brett A. Kamino, John B. Grande, Mike Brook,

Timothy P. Bender. All synthesis and characterization was performed by me. John B. Grande

performed initial tests, the results of which were not included in this publication. Mike Brook is

Mr. Grande’s supervisor. The paper was written by me with guidance from Prof. Tim Bender.

69

69

5.3 Paper

5.3.1 Abstract

A series of room temperature liquid siloxane-triarylamine hybrids were synthesized using the

Piers-Rubinsztajn reaction. These materials displayed well behaved electrochemical oxidation

and low Tg’s and were free-flowing liquids. The interaction between the Lewis acidic

tris(pentafluorophenyl)borane catalyst and the Lewis basic starting triarylamine substrates were

investigated by steady state UV-Vis spectroscopy and 19F NMR to reveal a unique and non-

productive charge transfer reaction between substrate and catalyst.

5.3.2 Body

The Piers-Rubinsztajn reaction has been shown to be a powerful way to construct complex

discrete siloxane architectures in an efficient manner.1 This reaction uses the strong Lewis acid

tris(pentafluorophenyl)borane (B(C6F5)3) to catalyze the reaction between Si-H and Si-O-R

groups (where R = H, Me or other alkyl, R3SiH + R′OSiR′′3 → R3SiOSiR′′3 + R′H, Scheme 5-

1A).2 This chemistry typically occurs very rapidly and is done under non-aqueous conditions.

Crucial to the utility of the process is the fact that silicones do not undergo

metathesis/redistribution in the presence of this Lewis acid.1b As well, the borane catalyst is

generally easy to remove and the byproduct is either hydrogen or volatile hydrocarbon gases

(such as methane) either of which rapidly leave the solution during reaction or under gentle

vacuum. Using this chemistry, the synthesis of many complicated and otherwise inaccessible

siloxane structures and other chemical derivatives can be achieved.3

Scheme 5-1: Two synthetic transformations accessible by using the Piers-Rubinsztajn reaction.

70

70

In addition to the synthesis of Si-O-Si bonds for discrete siloxane architectures, this chemistry

has also been shown to work between Si-H bonds and aryl-hydroxyl groups and aryl-methoxy

groups to form aryl-O-Si bonds (Scheme 5-1B).4

In this chapter we describe a series of free-flowing room temperature liquid siloxane-

triarylamine hybrid compounds that were prepared using the Piers-Rubinsztajn reaction. For this

chapter a reactive oligosiloxanes was used: 1,1,1,3,3-pentamethyldisiloxane (-MM). This was

chosen as prototypical reactive silicone because it is a pure and discrete compound, a liquid at

room temperature, inexpensive and readily available.

We begin our study by investigating the use of this chemistry on simple, single nitrogen centered

triarylamines with p-methoxy functionalization. Each precursor p-methoxy-triarylamine can be

synthesized in a single step using well established Buchwald-Hartwig coupling conditions5 by

the reaction between 4-bromoanisole and bis(3,4-dimethylphenyl)amine, 3,4-dimethylaniline, or

p-anisidine giving triarylamines 5-1a-c, respectively. Each methoxy functionalized triarylamine

was subsequently reacted with -MM in the presence of tris(pentafluorophenyl)borane. In a

typical procedure, the triarylamine was dissolved in toluene (10 wt. %) which contained a

catalytic amount of tris(pentafluorophenyl)borane (1 mol. %) at room temperature in an open

vessel. To this, -MM was added drop wise. There is a short induction time following which the

rapid evolution of gas occurs (methane in this case): Safety Note - the evolution can be vigorous,

and the addition rate of the silane should be adjusted accordingly. Reactions were worked up by

the addition of ∼0.5 g of basic alumina, which was allowed to stir for an additional 20 min

within the reaction vessel to capture the borane catalyst. The reaction solution was filtered, and

the solvent and excess 1,1,1,3,3-pentamethyldisiloxane were simply removed by rotary

evaporation. A general reaction is illustrated in Scheme 5-2. The isolated yields for these

reactions typically exceeded 90%. We found that no further purification of these compounds was

required after removal of excess pentamethyldisiloxane and boron catalyst (as shown by HPLC

and 1H/19F NMR analysis). All three siloxane-triarylamine hybrid compounds were isolated as

pale yellow, free-flowing liquids. Compounds 5-2b and 5-2c had viscosities similar to that of a

5130 g/mol weight averaged PDMS standard, whereas 5-2a was similar to a 24 800 g/mol

standard. Precise measurements are underway.

71

71

Scheme 5-2: Synthesis of single nitrogen centered triarylamines 5-2a-c

Each of the compounds 5-2a-c was characterized by differential scanning calorimetry (DSC) to

establish the effect of siloxane substitution on the arylamine and to observe any thermal

transitions which may exist (Table 5-1). DSC analysis was performed by first rapidly cooling

each liquid to -80 °C. Each sample was then subsequently heated to room temperature, back

down to -80 °C and finally back to room temperature all at a rate of 10 °C/min. Sharp glass

transitions (Tg) were observed in all cases in each of the heating cycles. No other thermal

transitions were observed for these compounds. It can be seen that addition of the -MM group

has a strong effect on the physical properties of the triarylamines. The normally crystalline

starting methoxy-triarylamines (5-1a-c) are converted to liquids with very low glass transition

temperatures ranging from -45 to -63 °C. It is observed that increasing the number of -MM

chains decreases the glass transition temperature. In no cases were other thermal transitions

including crystalline transitions observed.

Each siloxane-triarylamine hybrid was also characterized by cyclic voltammetry (CV) to

determine the effect of -MM substitution on the electrochemistry of the compounds. Each was

run under identical conditions to those for compounds 5-1a-c, and their results were compared.6

CV was performed in dichloromethane with (Bu)4NClO4 as a supporting electrolyte. Each

compound was scanned from -300 to +1100 mV and back to -300 mV at 50 mV/s. Each was

cycled through this range four times in total. Decamethylferrocene was used as an internal

standard and all data is corrected to its published half wave potential.7 The results are

summarized in Table 5-1. Siloxane-triarylamine hybrids (5-2a-c) have very similar

electrochemical behavior to that of the methoxy substituted triarylamines 5-1a-c with a single

1a 1b 1c

2a

2b

2c

72

72

reversible oxidation event. We observed that the halfwave oxidation potential (E1/2) for

compounds 5-2a-c decreases with increasing -MM substitution. This implies that -MM is an

electron-donating group.6 However, based on the comparison to the E1/2 for compounds 5-1a-c

we can conclude that the -MM moiety is a weaker electron-donating group than methoxy.

Table 5-1: Thermal and electrochemical information for precursor triarylamine compounds 5-

1a-c and their siloxane functionalized counterparts, 5-2a-c.

The accepted mechanism for the Piers-Rubinsztajn reaction initially involves the formation of a

reversible complex between the Si-H of the silane and the boron center of the Lewis acid catalyst

as a key intermediate step.1b It was anticipated that the strong Lewis acid used herein8 could

competitively bind to the substrate triarylamine, which is a weak Lewis base.9

The efficiency of the process described herein suggests there is little interaction between

tris(pentafluorophenyl)borane and the triarylamine substrate. However, if a single electron

trasnfer interaction took place, it would likely involve the removal of an electron from the

triarylamine resulting in the brightly colored radical cation of the triarylamine thereby allowing

for detection of even trace amounts of interactions.10,11 Thus UV-Vis spectra of triarylamine 5-1c

were taken between 300 and 2000 nm with varying equivalents of tris(pentafluorophenyl)borane

in dilute solutions of toluene (0.0298 mol/L). Looking at the visible region of the spectrum, a

very weak absorbance centered at 742 nm can be observed (Figure 5-1).

73

73

Figure 5-1: Steady-state UV-Vis absorbance spectra of 1c upon additiona of 0.25, 0.5, and 1

equiv. of tris(pentafluorophenyl)borane. Also shown is 1c oxidized with 0.5 equiv. of SbCl5 for

reference.

This peak increased in intensity in a nonlinear fashion with increasing amounts of

tris(pentafluorophenyl) borane. A chemically oxidized solution of triarylamine 5-1c with 0.5

equiv of antimony(V) chloride (SbCl5) was also prepared. It is well-known that the oxidative

action of SbCl5 on a (tri)arylamine produces the radical cation via a single electron transfer.12The

UV-Vis spectrum of the mixture of 5-1c and SbCl5 overlaps with the position and has the same

peak shape observed for the mixtures of 5-1c and tris(pentafluorophenyl)borane (Figure 5-1). It

can thus be concluded that there is some level of interaction between

tris(pentafluorophenyl)borane and the triarylamine and that the interaction results in the

formation of a radical cation/radical anion pair. The absorbance at 742 nm was extremely weak

compared to its primary absorption band at 298 nm. Given the extinction coefficient of a

(tri)arylamine radical cation is known to be very high13 relative to the neutral compound, the

spectra suggest that 5-1c and tris(pentafluorophenyl)borane are in an equilibrium shifted far

toward dissociation. To confirm this, 19F NMR was performed on an equimolar mixture of 5-1c

and tris(pentafluorophenyl)borane. No change in the 19F chemical shifts of

tris(pentafluorophenyl)borane was observed thereby confirming the equilibrium is shifted far

toward dissociation.

74

74

In summation, the functionalization of triarylamines with discrete siloxane chains under Piers-

Rubinsztajn conditions (tris(pentafluorophenyl)borane catalysis) has been shown to drastically

change their physical properties. The result is a sample of three free-flowing room temperature

liquid siloxane-triarylamine hybrids. Any Lewis acid/Lewis base interaction between the

arylamine and borane catalyst is weak such that free catalyst is available to interact with the

hydrosilane and initiate the Piers-Rubinsztajn process. We believe that this synthetic strategy

will facilitate further application of liquid arylamines in optoelectronic devices.

5.3.3 References

1. (a) Grande, J. B.; Thompson, D. B.; Gonzaga, F.; Brook, M. A. Chem. Commun. 2010,

46, 4988. (b) Brook, M. A.; Grand, J. B.; Ganachaud, F. Adv. Polym. Sci. 2010, 1.

2. (a) Piers, W. E. The Chemistry of Perfluoroaryl Boranes. In Advances in Organometallic

Chemistry; West, R., Hill, A. F., Eds.; Elsevier Academic Press: San Diego, 2005; Vol.

52, 1. (b) Chojnowski, J.; Rubinsztajn, S.; Cella, J. A.; Fortuniak, W.; Cypryk, M.;

Kurjata, J.; Kazmierski, K. Organometallics 2005, 24, 6077.

3. Thompson, D. B.; Brook, M. A. J. Am. Chem. Soc. 2008, 130, 32.

4. Cella, J.; Rubinsztajn, S. Macromolecules 2008, 41, 6965.

5. (a) Hartwig, J. F.; Kawatsura, M.; Hauck, S. I.; Shaughnessy, K. H.; Alcazar Roman, L.

M. J. Org. Chem. 1999, 64, 5575. (b) Jiang, L.; Buchwald, S. L. Palladium-Catalyzed

Aromatic Carbon-Nitrogen Bond Formation. Metal-Catalyzed Cross-Coupling Reactions,

2nd ed.; de Meijere, A., Diederich, F., Eds.; Wiley-VCH: 2004; Chapter 13. (c) Bender,

T. P.; Coggan, J. A.; McGuire, G.; Murphy, L. D.; Toth, A. E. J. US Patent 7,408,085,

2008. (d) Bender, T. P.; Coggan, J. A. US Patent 7,402,700, 2008. (e) Bender, T. P.;

Goodbrand, H. B.; Hu, N. X.; US Patent 7,402,699, 2008. (f) Bender, T. P.; Coggan, J.

A.; US Patent 7,345,203, 2008. (g) Coggan, J. A.; Bender, T. P. US Patent 7,332,630,

2008.

6. Bender, T. P.; Graham, J. F.; Duff, J. M. Chem. Mater. 2001, 13, 4105

7. Noviandri, I.; Brown, K. N.; Fleming, D. S.; Gulyas, P. T.; Lay, P. A.; Masters, A. F.;

Phillips, L. J. Phys. Chem. B 1999, 103, 6713.

8. Brook, M. A. Silicon in Organic, Organometallic, and Polymer Chemistry; John Wiley &

Sons, Inc.: New York, 2000.

9. Piers, W. E.; Chivers, T. Chem. Soc. Rev. 1997, 26, 345.

75

75

10. Park, M. H.; Park, J. H.; Do, Y.; Lee, M. H. Polymer 2010, 51, 4735.

11. Blackwell, J. M.; Sonmer, E. R.; Scoccitti, T.; Piers, W. E. Org. Lett. 2000, 2, 3921.

12. Amthor, S.; Noller, B.; Lambert, C. Chem. Phys. 2005, 316, 141–152.

13. Zhou, G.; Baumbarten, M.; Müllen, K. J. Am. Chem. Soc. 2007, 129, 12211.

76

76

Chapter 6: Liquid Triarylamines: The Scope and Limitations of Piers-Rubsinsztajn Conditions for Obtaining Triarylamine-Siloxane Hybrid Materials


This chapter was published as a full paper in the Journal of Organic Chemistry.

Brett A. Kamino, Bridget Mills, Christopher Reali, Michael J. Gretton, Mike Brook, Timothy P.

Bender, 2012, Journal of Organic Chemistry, 77 (4), 1663-1674.

Figure and Schemes are reprinted with permission. Copyright 2011 American Chemical Society

This paper is a follow up publication to Chapter 6 where the previously mentioned concepts are

greatly expanded on. Because of the success of our synthetic strategy for producing liquid

organic semiconductors, we sought to expand our prior work and determine the limits of this

particular strategy for obtaining liquid semiconductors. As well, there was a strong desire to

explore the nature of the catalyst substrate interactions found in the previous chapter.

In this study, we attempted to produce liquid versions of a large range of nitrogen containing

hole transport materials found across the literature. This was done to show the utility of our

methodology as well as to build up a library of potential compounds to study in device

applications. This study was fairly successful and we were able to produce liquid versions of a

large number of organic semiconductors while learning a number of key molecular design rules.

In concluding this study and the previous one, I can safely say that we have extensively

demonstrated a flexible strategy to easily produce liquid organic semiconductors.


The author list for this paper is as follows: Brett A. Kamino, Bridget Mills, Christopher Reali,

Michael J. Gretton, Mike Brook, Timothy P. Bender. Bridget Mills and Christopher Reali were

undergraduate students working under me on this project. They contributed towards the synthesis

of chemical intermediates and several of the final products using procedures developed by me.

77

77

Michael J. Gretton provided input on the project. Mike Brook originally consulted on the use of

Piers-Rubinsztajn chemistry for triarylamines. The large bulk of synthesis was performed by me.

As well, all characterization and studies into catalyst interaction was performed by me. The

paper was written by me with input from Prof. Bender.

6.3 Paper

6.3.1 Abstract

New liquid triarylamine−siloxane hybrid materials are produced using the Piers−Rubinsztajn

reaction. Under mild conditions, liquid analogues of conventional and commonly crystalline

triarylamines are easily synthesized from readily available or accessible intermediates. Using a

diverse selection of triarylamines, we explored the effects of siloxane group and substitution

pattern on the physical properties of these materials, and we have demonstrated that relatively

large molecular liquids with desirable electrochemical properties can be produced. The

interactions between the strongly Lewis acidic catalyst used for this transformation,

tris(pentafluorophenyl)borane (BCF), and the Lewis basic triarylamine substrates were studied.

Through UV−vis−NIR and 19F NMR spectroscopy, we have proposed that the catalyst undergoes

a reversible redox reaction with the substrates to produce a charge transfer complex. The

formation of this charge transfer complex is sensitive to the oxidation potential of the

triarylamine and can greatly affect the kinetics of the Piers−Rubinsztajn reaction.

6.3.2 Introduction

We have recently shown that the incorporation of short, discrete disiloxane units around the

periphery of triarylamines is a versatile strategy to produce liquid triarylamines (see Chapter 5).

Central to this synthetic strategy is the use of the Piers−Rubinsztajn reaction,1 which uses

tris(pentafluorophenyl)borane (BCF) as a catalyst) to activate stable and easily handled silanes.

The reaction is performed under very mild and ambient conditions and requires low catalyst

loadings. The only byproduct of the reaction is methane, and the substitution proceeds without

metathesis or redistribution of the siloxane component, which commonly occur for many types

of siloxane chemistries and limit the ability to construct discrete structures. Rapid conversion

was observed despite the use of the strongly Lewis acidic catalyst, which on first examination

may have been assumed to complex with the Lewis basic triarylamine substrate,2 preventing the

78

78

reaction from proceeding. An investigation using UV−Vis spectroscopy into the potential

complexation showed a small amount of charge transfer between BCF and the triarylamine

indicated by the presence of an absorption, which could be directly attributed to the oxidized

arylamine (radical cation). Despite the detectable presence of this absorption, the relatively weak

signal observed coupled with the known high extinction coefficient for radical cations of

triarylamines suggests that the equilibrium between the charge transfer state and the neutral state

lies far toward the latter.

Piers−Rubinsztajn conditions have previously been shown to enable the reaction of silanes with a

number of different coupling partners including carbonyl groups,3 aryl hydroxyl/alkoxy groups,4

alkyl hydroxyl groups,5 siloxy groups, and silanols.6 Piers−Rubinsztajn conditions have been

shown to have a wide scope and have been used for the synthesis of structured and functional

materials such as siloxane dendrimers,7 polysiloxanes and copolymers thereof,8 and surface

active siloxane ambiphiles.9 While this chemistry has good functional group tolerance,10 there

are cases where Lewis basic substrates react poorly because of competitive catalyst binding to

these Lewis basic functionalities.11 The strength of the Lewis basic functionality is thus an

important factor in the reaction kinetics and whether the reaction can proceed at all.

Our original study (Chapter 5) was limited to performing this reaction on simple triarylamines

with a single nitrogen center. However, simple triarylamines have a limited range of

electrochemical oxidation potentials and applications in functional devices.12 Therefore, we

sought out to broaden the scope of the Piers−Rubinsztajn process and its intersection with

triarylamine−siloxane hybrid materials with an aim at producing a larger number of liquid

triarylamines. We have focused our attention on multinitrogen centered triarylamines as well as

molecules containing the related carbazole moeity. Triarylamines containing multiple conjugated

nitrogen centers are known to possess a wider range of electrochemical properties and can

exhibit better stability upon oxidation (see Chapter 3), leading to improved device performance.

Conversely, they also have the potential to be significantly more Lewis basic than simple

triarylamines because of their multiple nitrogen centers. Therefore, these substrates provide an

opportunity to probe the scope and limitation of the reaction of methoxy-functionalized

triarylamines with discrete silanes under Piers−Rubinsztajn conditions. The limitation of most

interest is whether the decrease in oxidation potential on moving to triarylamines with multiple

79

79

centers will result in a significant shift in charge transfer equilibrium between BCF and the

triarylamine substrate, thus inhibiting the reaction.


While there are many different triarylamine structures from which to choose for this study, we

began by choosing two of the more common motifs: triarylamines with a phenylene diamine core

and those with a benzidine core (Scheme 6-1). Several variants on these backbones were

synthesized so as to understand the effect of molecular symmetry and degree of siloxane

functionalization on the physical and electrochemical properties of the resulting materials.

Furthermore, the structural changes in each variant are also accompanied by inherent electronic

variations affecting the Lewis basicity of the molecule.

Triarylamines based on a phenylene diamine core (compounds 6-2a−b) were synthesized in two

steps (Scheme 6-1). 1,4-Phenylene diamine was first reacted with 4-bromoanisole under standard

Buchwald−Hartwig amination conditions13 to yield an intermediate triarylamine 6-1. This aryl-

methoxy-substituted triarylamine was then functionalized with discrete siloxane groups

(1,1,1,3,3-pentamethyldisiloxane (MMH) or 1,1,1,3,5,5,5-heptamethyltrisiloxane (MDHM)) by a

reaction with the corresponding silanes in the presence of a catalytic amount of

tris(pentafluorophenyl)borane (BCF, the so-called Piers−Rubinsztajn conditions) under ambient

and open air conditions. In our previous work (Chapter 5), we found that the functionalization of

methoxy-containing triarylamines with MMH under these conditions proceeded rapidly.

However, in this case, complete conversion took approximately 7 h to achieve. Gentle heating

(50 °C) was found to speed up the process considerably. Caution: As the temperature in the flask

increased, the reaction would proceed rapidly with the rapid evolution of methane gas.

80

80

NH2H2N

OMe

Br

Pd(dba)2P(t-butyl)3Na(t-butoxide)

toluene, reflux

NN

MeO

MeO OMe

OMe

B(C6F5)3

toluene, 50 oC

MMH or MDHM

NN

R

R R

R

6-1 6-2a R = -OMM

6-2b R = -OD(M)2

NH2

R1

R2

MeO

Br


toluene, reflux

HN

R1

R2

OMe

6-3a R1 = -H, R2 = -H

6-3b R1 = -CH3, R2 = -H

6-3c R1 = -CH3, R2 = -CH3

Pd(OAc)2P(t-butyl)3Na(t-butoxide)

toluene, reflux

Br Br

N N

R1R2R1 R2

MeO OMe

B(C6F5)3toluene, rt

MMH or MDHM

6-4a R1 = -H, R2 = -H

6-4b R1 = -CH3, R2 = -H

6-4c R1 = -CH3, R2 = -CH36-4d R1 = -OMe, R2 = -H

N N

R1R2R1 R2

R3 R3

6-5a R1 = -H, R2 = -H, R3 = -OMM

6-5b R1 = -CH3, R2 = -H, R3 = -OMM

6-5c R1 = -CH3, R2 = -CH3, R3 = -OMM

6-5d R1 = -OMM, R2 = -H, R3 = -OMM

6-5e R1 = -CH3, R2 = -H, R3 = -OD(M)26-5f R1 = -OD(M)2, R2 = -H, R3 = -OD(M)2

Scheme 6-1: Synthesis of siloxane functionalized arylamines.

Triarylamines with a benzidine core (6-5a−h) were synthesized in three steps (Scheme 1).

Diarylamines were first prepared by Buchwald−Hartwig coupling of the corresponding aryl

bromides with an excess of the appropriate aniline. In our case, each diarylamine was easily

81

81

purified by an acidic aqueous extraction followed by recrystallization from nonpolar solvents.

They were then reacted with 4,4′-dibromobiphenyl to achieve the methoxy-functionalized

triarylamines (6-4a−e). Finally, the triarylamines were functionalized with discrete siloxane

chains by reaction with the corresponding silane and catalytic amounts of BCF to give

compounds 6-5a−h. Unlike the reaction with the 1,4-phenylene-based materials (6-2a−b), the

introduction of the siloxane groups proceeded quickly at room temperature, requiring up to a

minute of induction time before reacting rapidly.

Generally speaking, all reactions proceeded to high conversion, and we did not see any

correlation between triarylamine molecular structure and crude yields. We also did not see a

correlation between the equivalent amount of silane used (in some cases we used 2 equiv. and in

others 5 equiv.) and the crude yields. NMR analysis of the crude reaction mixtures does suggest

that the reaction generates small amounts of polymeric silicone species as evidenced by a

corresponding and characteristic CH3 resonance in the 1H NMR spectrum. The generation of this

byproduct is likely due to the reaction of ambient water with silane in the reaction mixture

moderated by BCF.14 Column chromatography was found to be effective in removing this

byproduct. Clearly, conducting the P−R process under anhydrous conditions would preclude the

formation of the silicone species. However, as column chromatography is generally necessary to

produce triarylamine suitable for study in organic electronic devices, its use to remove the

produced silicones is balanced against the ease of operating under ambient vs. anhydrous

conditions.

All of the final triarylamine−siloxane hybrid materials were isolated by column chromatography

(SiO2, toluene/cyclohexane) as viscous liquids (compounds 6-2b, 6-5a−c, e−f,h) or as crystalline

solids (compounds 6-2a, 6-5d, and 6-5g), and their structure and purity (and those of the

intermediates) were unambiguously confirmed by high resolution mass spectrometry (HRMS)

and NMR analysis. Yields for the complete unoptimized process ranged significantly between 48

and 90% after column chromatography. Overall, compounds with the bulkier −OD(M)2 group

gave higher isolated yields than those with the less bulky −OMM group.

Each compound was characterized by differential scanning calorimetry (DSC) to determine the

effect of silicone substitution and molecular symmetry on the melting point and/or glass

transition temperatures of the materials (Table 6-1). Consistent with our observations regarding

82

82

the analogous single nitrogen centered triarylamine−siloxane hybrids (Chapter 5), adding

discrete silicone groups to the arylamine cores had a significant effect on the glass transition

temperature (Tg) and the presence of a crystalline state. Using pentamethyldisiloxane (−OMM)

as a siloxane, the most symmetrically substituted arylamines (6-2b,6-5d, 6-5g) were isolated as

crystalline solids and exhibited well-defined and sharp melting points by DSC. Upon heating,

compound 6-5d exhibited two exothermic transitions at 52 and 58 °C, possibly indicating

polymorphism. For each compound that showed a melting transition on the first DSC scan, a

glass transition temperature well below room temperature was detected on the second heating

cycle. These low glass transition temperatures suggested that if crystallization could be inhibited,

the materials would behave as liquids at room temperature. This hypothesis was tested by

utilizing the bulkier and branched 1,1,1,3,5,5,5-heptamethyltrisiloxane (−OD(M)2) group in

place of the linear –OMM group. The resulting compounds (6-2b, 6-5f, 6-5h) were isolated as

free-flowing oils, and no detectable crystallization was observed over several months of storage

at ambient conditions. The bulkier siloxane groups appear to prevent the π−π stacking

interactions that are found in the crystal structures of these molecules.15 Additionally, we found

that the presence of asymmetry in the substitution pattern of the triarylamine along the axis

bisecting the two nitrogen atoms resulted in compounds that were isolated as highly viscous oils

at room temperature (6-5a, 6-5b, 6-5c, 6-5e). For example, compounds 6-5a−c containing two

−OMM groups and various methyl group substituents on the adjacent phenyl ring are liquids,

whereas their symmetric counterparts are not. Increasing the number of methyl groups raised the

glass transition temperature, whereas replacing the linear −OMM siloxane with a branched

−OD(M)2 siloxane (6-5e) lowered the glass transition temperature by 15 °C as compared to its

structural counterpart (6-5b).

83

83

Table 6-1: DSC and CV results from silicone-hybridized triarylamines.

Compound Appearance Tm (0C) Tg (0C)

Eox 1 (mV vs.

Ag/AgCl)

Eox 2 (mV vs.

Ag/AgCl)

6-2a Crystalline 87 -49§ 443 902

6-2b Oil - -47 436 906

6-5a Oil - 6 749 994

6-5b Oil - 11 725 942

6-5c Oil - 14 673 935

6-5d Crystalline 52, 58 -23§ 661 892

6-5e Oil - -4 699 942

6-5f Oil - -31 682 921

6-5g Crystalline 88 -1 701 907

6-5h Oil - -9 704 929

6-9 Crystalline 150 28 616 772

6-11 Crystalline 131 20 366 696

6-15a Oil - -44 1495 † -

6-15b Oil - -50 1320 † -

6-15c Crystal 47 -48§ 1485 † -

6-15d Oil - -48 1510 † -

§ - detected on second heating cycle. † - Irreversible, peak potential on first scan reported. Tm – melting temperature; Tg – glass transition temperature; Eox,1 – first half wave oxidation

potential; Eox2 – second half wave oxidation potential.

84

84

From these observations, along with those in the previous chapter, we can make several

conclusions about the effect of siloxane substitution on the physical state of arylamines. We find

that the incorporation of larger siloxane chains lowers the glass transition temperature of the

material. Furthermore, as the base molecule increases in size, the number of siloxane chains

needed to achieve a room temperature liquid arylamine−silicone hybrid increases. A secondary

consideration is the addition of asymmetry to the molecule, which is shown to decrease the

chances of crystallization in the molecule.

The electrochemical behavior of this initial group of triarylamine−siloxane hybrids was studied

using solution cyclic voltammetry in dichloromethane with 0.1 M tetrabutylammonium

perchlorate as a supporting electrolyte. A small amount of decamethylferrocene was added to all

solutions and used as an internal standard.16 The results of the electrochemical analysis are

included in Table 6-1. All of the compounds underwent two reversible 1-electron oxidation

events, which is characteristic of triarylamines containing two nitrogen centers. This

demonstrates that the siloxane functionalization of arylamines does not have an undesirable

impact on their electrochemical behavior. Variations in the position of the oxidation potentials

can be explained by number and strength of electron-donating silicone groups as well as degree

of conjugation between each arylamine redox center. There does not seem to be a consistent

difference in the electron-donating potential of the −OD(M2) versus the −OMM group.

Encouraged by the success of this chemistry on phenylenediamine and benzidine-based

substrates, we further extended this study to larger (higher molecular weight) multinitrogen

centered triarylamines. Two well-known triarylamine structures were chosen as model

compounds: one based on the spiro-TAD core17 and one based on the TDATA motif, each

containing four nitrogen centers (Schemes 6-2 and 6-3, respectively). The spiro-TAD derivative

(6-9) was synthesized using a slightly modified literature procedure,18 resulting in a structure

containing four aryl-methoxy groups for substitution (Scheme 6-2). When reacted with MDHM

under the Piers−Rubinsztajn conditions, the substrate reacted rapidly and cleanly to yield a

molecular glass upon isolation, which ultimately crystallized over several weeks. The TDATA

structure (6-11) was synthesized in two steps from tris(p-bromophenyl) amine (Scheme 6-3) and

isolated as a crystalline solid. In this case, the reaction proceeded very slowly requiring

approximately 16 h for completion. Both structures showed typical electrochemical behavior

85

85

relative to their respective classes, and their synthesis demonstrates the ability of

Piers−Rubinsztajn conditions to successfully install −OD(M)2 units on large multi-nitrogen

centered substrates.

Br

O1) MgTHF

2)

3) HClAcetic Acid

Br2/FeCl3DCM, rt

Br

Br

Br

Br

HN

OMe

Pd(OAc)2P(t-butyl)3Na(t-butoxide)

Toluene, reflux

N

N

N

N

O

O

O

O

Si

Si

Si

Si

O

O

O

O

OO

OO

Si

Si

Si

Si

Si

Si

Si

Si

NN

NN

OMe

MeOOMe

OMe

B(C6F5)3Toluene, rt

MDHM

6-6 6-7

6-86-9

Scheme 6-2: Synthesis of spiro core triarylamine 6-9

Scheme 6-3: Synthesis of triarylamine 6-11

86

86

Next, we chose to extend this chemistry to a related group of arylamine semiconductors:

carbazoles. To this end, a number of N-phenylcarbazoles with methoxy-substituents in various

positions were synthesized and subjected to the Piers−Rubinsztajn reaction conditions.

Carbazoles, like triarylamines, are common materials in organic electronics and have been used

as both p-type semiconductors and ambipolar host materials.19 Four N-phenylcarbazole

precursors were synthesized. For the structures with methoxy groups present on the carbazole

ring itself (6-14a−b, Scheme 6-4), Suzuki coupling of 1-bromo-2-nitro-4-methoxybenzene and

an arylhalide followed by a reductive ring-closing reaction promoted by triethylphosphite

resulted in the desired carbazoles (6-13a−b). These carbazoles were then N-arylated under

Ullman coupling conditions to yield the final methoxy-functionalized precursors (6-14a−b), of

which 6-14b has been previously described. Those with methoxy groups on the phenyl ring (6-

14c−d) were prepared in a one-step procedure under Ullman coupling conditions using carbazole

and the appropriate halo-anisole. Because of the planar molecular structure of compounds 6-

14a−d, they were further functionalized exclusively with −OD(M)2 groups under

Piers−Rubinsztajn conditions in order to inhibit any possible crystallization. In each case, the

reaction proceeded extremely rapidly, and the resulting siloxane-functionalized carbazoles (6-

15a−d) were isolated in good yields. For all but compound 6-15d, the resulting materials were

isolated as clear free-flowing liquids. Electrochemical analysis of the carbazoles 6-15a−d

showed that each underwent an irreversible one electron oxidation typical for carbazoles.19

Interestingly, the di-substituted 6-14b appears to undergo two oxidations, which are poorly

resolved under standard cyclic voltammetry conditions. It is unclear at this time whether these

events represent two subsequent electrons being removed or that another chemical change is

occurring. Full electrochemical and calorimetry data for these compounds is included in Table 6-

1. The effect of substitution pattern on the physical properties of the carbazoles is in line with our

previous observations in that the most symmetric substitution patterns (6-15d) result in a

derivative that was crystalline, whereas the asymmetrically substituted structures are isolated as

liquids.

87

87

Scheme 6-2: Synthesis of siloxane functionalized carbazoles.

Effect of Substrate on Catalyst Efficacy. In our approach to the synthesis of

arylamine−siloxane hybrid materials, the use of Piers−Rubinsztajn conditions is critical to the

installation of discrete siloxane groups while avoiding unwanted metathesis and redistribution

reactions of the silicones. The previously studied simple triarylamines were found to react

rapidly at room temperature (often reacting as quickly as silane could be added to the mixture).

However, in this study, it was observed that the efficacy of this reaction (as indicated by its time

to completion at room temperature) varied depending on the structure of the triarylamine

substrate. The rate of the reaction for this series of compounds could be qualitatively observed by

the evolution of methane gas. Compared to the triarylamines used in our previous study, the

benzidine-based substrates reacted somewhat more slowly, with reactions typically finishing

several minutes after complete silane addition. Substrates 6-1 and 6-10 stood out as the only

substrates requiring heat to proceed rapidly; leaving the substrates to react at room temperature

necessitated 7 and 16 h of reaction time for complete substitution, respectively. We also

observed rapid reaction in the formation of triarylamines 6-9 and 6-11 and also for all carbazole

derivatives (6-14a−d).

In our previous work, we found an interaction between the strongly Lewis acidic

tris(pentafluorophenyl)borane (BCF) and the weakly Lewis basic tris(p-methoxyphenyl)amine

did occur but that the level of interaction was quite weak. This interaction between catalyst and

substrate was shown to result in the formation of the triarylamine radical cation. However, we

88

88

did not directly prove the existence of the corresponding radical anion of BCF. While such

reactions are uncommon, several examples of BCF acting as an oxidant are noted in the

literature.20 A complete understanding of the strength of BCF as an oxidant has not been

established owing to the instability of the this particular borane anion.21 Since the radical cation

of an arylamine can be easily detected by its characteristic UV−vis−NIR absorption, we set out

to see if the observed differences in reactivity between triarylamines could be explained by

differing interactions between the precursor and the BCF catalyst.

Solutions of arylamines 6-1 and 6-4d were prepared in toluene (our reaction solvent) along with

varying molar equivalents of BCF. Compounds 6-1 and 6-4d were chosen as representative

compounds because their oxidized cations have been previously studied in the literature.22

Furthermore, the observed reaction kinetics with these substrates varied greatly under the

described conditions. To study this interaction under simulated reaction conditions, solutions of

both BCF and substrate were prepared under ambient conditions and allowed to stand in open air

for 15 min. These solutions were studied by UV−Vis−NIR absorbance measurements, which

found large changes in the visible and near-infrared region of the spectra upon addition of BCF

(see Figure 6-1). Compounds 6-1 and 6-4d both exhibited broad absorption bands in the NIR as

increasing amounts of BCF were added. These broad absorption bands can be assigned to

intervalence charge transfer (IVCT) bands of the radical cations of the arylamines (6-1•+ and 6-

4d•+, respectively), which result from the partial delocalization of the charge between two

conjugated redox centers. The UV−Vis−NIR absorbance spectra of 6-1•+ and 6-4d•+ are well-

studied as model charge transfer systems, and their molar extinction coefficients in

dichloromethane are known.22 A comparison between the published spectra of 6-1•+ and 6-4d•+

and the spectra generated by mixing the neutral triarylamines with BCF reveals that they are

practically the same. Small differences in the λmax found in the literature and in our study are

likely due to differences in the solvents used for each measurement. Such charge transfer bands

are inherently sensitive to solvent polarity.23 The presence of atmospheric conditions can

potentially complicate this experiment, as it is well-known that BCF readily forms adducts with

water under the conditions of this study.24 To rule out any possibility of water playing a role in

the observed reaction, solutions were prepared under glovebox conditions. Under these

anhydrous conditions, the exact same color changes were noted, and the samples were

spectroscopically identical.

89

89

Figure 6-1: Steady-state solution absorbance spectroscopy in toluene of a) compound 6-1 b) compound 6-4d with 0, 0.25, 0.5, and 1 equivalents of BCF. Arrows indicate increasing absorption with increasing equivalents of BCF.

On the basis of these spectra and the known molar extinction coefficients of 6-1•+ and 6-4d•+

(albeit in different solvents) we can estimate the amount of arylamine that is oxidized (see tables

S6-1 and S6-2). If an equimolar mixture of 6-1 and BCF is considered, the percentage of 6-1

oxidized is 5 mol % (corresponding to 5 mol % BCF reduced). If the same consideration is given

to a mixture of 6-1 with catalytic amounts of BCF (0.01 equiv to 6-1), the amount of 6-1

oxidized is 0.2 mol % (corresponding to 20 mol % of BCF reduced). In a similar manner for

90

90

compound 6-4d, 2 mol % is oxidized at equal molar amounts (corresponding to 2 mol % BCF

reduced), and 0.03 mol % is oxidized at catalytic amounts (corresponding to 3 mol % BCF

reduced). We can therefore conclude that while the presence of 6-1•+ and 6-4d•+ is detectable,

even by eye, the result is not the quantitative reduction of BCF to BCF•−, rather it is only a partial

reduction. Furthermore, given the partial reduction and the absence of the formation of a

precipitate,25 we hypothesize that the system is in a dynamic equilibrium between the charge

transfer couple and the freely dissociated species. Taking into consideration our previous

observation of the very slight oxidation of tris(p-methoxyphenyl)amine by BCF, we propose that

the extent of oxidation of the arylamine by BCF is directly related to the oxidation potential of

the arylamine. That is, the equilibrium shifts toward formation of the radical cation/radical anion

pair for arylamines with lower oxidation potentials. The oxidation potentials of the tris(p-

methoxyphenyl)amine, 6-4d, and 6-1 are 654,26 605,22 and 375 mV vs. Ag/AgCl, respectively.

Additionally, these values are proportional to the observed kinetic rates of the substrates under

these conditions. This result suggests that the nature of interaction between triarylamines and

BCF is a redox reaction where the BCF is acting as an oxidant. Furthermore, we can conclude

that the redox reaction between certain triarylamines and BCF results in partial sequestration of

the catalyst and retardation of the rate of the desired reaction.

To further support this hypothesis, mixtures of 6-1 and 6-4d along with 1 mol % equiv of BCF

were dissolved in toluene-d8 and studied by 19F and 1H NMR. In each case, the 1H spectra

obtained were significantly broadened because of the presence of the paramagnetic arylamine

radical cation (see section 12.4, Figures S6-3 and S6-5). 19F NMR spectra were difficult to obtain

with a reasonable signal-to-noise ratio because of the highly dilute nature the BCF and the

relatively low solubility of arylamines 6-1 and 6-4d. The 19F NMR of the mixture of compound

6-1 and 1 mol % BCF displayed a large number of resonances, with several corresponding to the

free BCF (Section 12.4, Figure S6-2). Under the same conditions, the mixture of compound 6-4d

and 1 mol % BCF resulted in a cleaner spectrum comprising mostly of free BCF (Section 12.4,

Figure S6-4). Because of the poor signal-to-noise ratio obtained and the paramagnetic nature of

the solution, concrete comparisons between the amount of free BCF in solution in the

UV−Vis−NIR experiment and the NMR experiment are difficult to make.

91

91

We therefore considered mixtures of arylamines 6-1 and 6-4d and BCF in a 1:1 molar ratio (in

C6D6; Section 12-4, Figures S6-6 and S6-7). Under these conditions, all three fluorine resonances

of the free BCF are visible and make up the majority of the detected signal, although moderately

broadened. This was particularly evident for compound 6-1, and the difference in observed

amount of free BCF between the two concentrations supports the idea of a dynamic chemical

equilibrium between the free and bound state.

6.3.4 Conclusions

In summary, the scope and limitations of Piers−Rubinsztajn conditions for the synthesis of

silicone−arylamine hybrids has been explored with an emphasis on obtaining liquid triarylamines

having multiple nitrogen centers. The catalytic functionalization of a series of two nitrogen

centered triarylamines was successful, and this substitution was found to have a significant effect

on the physical properties of the arylamines, resulting in a number of new liquid triarylamines.

We observed that by using the bulkier and branched −OD(M)2 group, normally crystalline bis-

arylamines could be made into free-flowing liquids by effectively lowering the glass transition

temperature to below room temperature and inhibiting crystallization. The flexibility of this

synthetic strategy was further explored by demonstrating that this chemistry can be used on

larger multinitrogen centered arylamines as well as on carbazole-based precursors. The siloxane-

functionalized triarylamines retained the electrochemical behavior of their parent compounds.

Furthermore, the interaction between the Lewis acidic BCF (tris(pentafluorophenyl)borane) and

the Lewis basic arylamine was further studied. It is proposed that the BCF catalyst can act as a

one-electron oxidant, forming a radical cation/radical anion pair that is in equilibrium with the

freely dissociated arylamine and BCF. This redox reaction sequesters the catalyst and retards the

rate of the desired reaction in some cases. This mechanism is supported by UV−vis−NIR and 19F

NMR spectra. From studying these compounds, we hypothesize that the extent of the charge

transfer interaction is dependent on the triarylamine oxidation potential. More easily oxidized

triarylamines appear to achieve an equilibrium state further favoring the charge transfer pair than

those triarylamines that are more electron-poor. This observation directly relates to the

qualitative kinetics of the Piers−Rubinsztajn reaction on these substrates.

Through the work presented in both our preliminary study and this one, we have successfully

increased the number of known liquid organic semiconductors. These novel materials emulate

92

92

the basic electrochemical properties of their widely used solid counterparts while spanning a

wide range of oxidation potentials. Perhaps more importantly, we have shown that this synthetic

strategy is quite general, which may also extend to other popular classes of organic

semiconductors and possibly be used to produce other liquid organic semiconductor types in the

future. This work will hopefully help further the study of organic electronic devices

incorporating liquid layers and allow for the synthesis of device specific liquid materials.

93

93

6.3.5 References

1. a) Piers, W.E. The chemistry of perfluoroaryl boranes. In Advances in Organometallic

Chemistry; West, R., Hill, A.F., Eds.; Elsevier Academic Press: San Diego, 2005; Vol. 52,

1. b) Brook, M.A.; Grande, J.B.; Ganachaud, F. Adv. Polym. Sci. 2010, 235, 161-183.

2. Parks, D.J.; Piers, W.E. J. Am. Chem. Soc. 1996, 118, 9440-9441.

3. Rubinsztajn, S.; Cella, J. A. Macromolecules 2005, 38 (4), 1061-1063

4. Blackwell, J.M.; Foster, K.L.; Beck, V.H.; Piers, W.E. J. Org. Chem. 1999, 64, 4887-

4892.

5. Zhou, D.; Kawakami, Y. Macromolecules 2005, 16, 6902-6908.

6. Brook, M.A. Silicon in Organic, Organometallic, and Polymer Chemistry, John Wiley

and Sons, New York, 2000.

7. Thompson, D. B.; Brook, M. A. J. Am. Chem. Soc. 2008, 1, 32-33.

8. a) Chojnowski, J.; Fortuniak, W.; Kurjata, J.; Rubinsztajn, S.; Cella, J. A.

Macromolecules 2006, 11, 3802-3807. b) Chojnowski, J.; Rubinsztajn, S.; Fortuniak, W.;

Kujata, J. Inorg. Organomet. Polym. Mater. 2007, 17, 173-187 c) Cella, J.; Rubinsztajn,

S. Macromolecules 2008, 41, 6965-6971.

9. Grande, J.B.; Gonzaga, F.; Brook, M.A. Dalton Trans. 2010, 39, 9369-9378.

10. Grande, J. B.; Thompson, D. B.; Gonzaga, F.; Brook, M. A. Chem. Comm. 2010, 27,

4988-4990.

11. a) Parks, D.J.; Blackwell, J.M.; Piers, W.E. J. Org. Chem. 2000, 65, 3090-3098. b)

Blackwell, J.M.; Sonmor, E.R.; Scoccitti, T.; Piers, W.E. Org. Lett. 2000, 2, 3921-3923.

12. Thelakkat, M. Macromol. Mater. Eng. 2002, 287, 442-461.

13. (a) Hartwig, J. F.; Kawatsura, M.; Hauck, S. I.; Shaughnessy, K. H.; Alcazar Roman, L.

M. J. Org. Chem. 1999, 64, 5575. (b) Jiang, L.; Stephan, B. L. Palladium-Catalyzed

Aromatic Carbon-Nitrogen Bond Formation. Metal-Catalyzed Cross-Coupling Reactions,

2nd ed.; de Meijere, A., Diederich, F., Eds.; Wiley-VCH: 2004; Chapter 13. (c) Bender, T.

P.; Coggan, J. A.; McGuire, G.; Murphy, L. D.; Toth, A. E. J. US Patent 7,408,085, 2008.

(d) Bender, T. P.; Coggan, J. A. US Patent 7,402,700, 2008. (e) Bender, T. P.; Goodbrand,

94

94

H. B.; Hu, N. X.; US Patent 7,402,699, 2008. (f) Bender, T. P.; Coggan, J. A.; US Patent

7,345,203, 2008. (g) Coggan, J. A.; Bender, T. P. US Patent 7,332,630, 2008.

14. (a) Janiak, C.; Braun, L.; Scharmann, T.G.; Girgsdies, F. Acta. Crystallogr. Sect. C 1998,

C54 (11), 1722-1724. (b) Di Saverio, A.; Focante, F.; Isabella, C.; Resconi, L.;

Beringhelli, T.; D’Alfonso, G.; Donghi, D.; Maggioni, D.; Mercandelli, P.; Sironi, A.

Inorg. Chem. 2005, 44 (14), 5030-5041.

15. Szeghalmi, A.V.; Erdmann, M.; Engel, V.; Schmitt, M.; Amthor, S.; Kriegisch, V.; Nöll,

G.; Stahl, R.; Lambert, C.; Leusser, D.; Stalke, D.; Zabel, M.; Popp, J. J. Am. Chem. Soc.

2004, 126, 7834-7845.

16. Noviandri, I; Brown, K.N.; Fleming, D.S.; Fulvas, P.T.; Lay, P.A.; Masters, A.F.; Phillips,

L. J. Phys. Chem. B 1999, 103, 6713.

17. Saragi, T.P.I.; Spehr, T.; Siebert, A.; Fuhrmann-Lieker, T.; Salbeck, J. Chem. Rev. 2007,

107 (4), 1011-1065.

18. Wu, R.; Schumm, J.S.; Pearson, D.L.; Tour, J.M. J. Org. Chem. 1996, 61 (20), 6906-6921.

19. Brunner, K.; Dujken, A.; Börner, H.; Bastiaansen, J.J.A.M.; Kiggen, N.M.M.; Langeveld,

B.M.W. J. Am. Chem. Soc. 2004, 126 (19), 6035-6042.

20. Piers, W.E. The Chemistry of Perfluoroaryl Boranes. In Advances in Organometallic

Chemistry; West, R.; Hill, A.F. Eds.; Elsevier Inc. USA, 2005; Vol. 52, 1-76.

21. Kwaan, R.J.; Harlan, C.J.; Norton, J.R. Organometallics 2001, 20, 3818-3820.

22. Lambert, C.; Nöll, G. J. Am. Chem. Soc. 1999, 121, 8434-8442.

23. a) Nelsen, S.F.; Trieber, D.A.; Ismagilov, R.F.; Teki, Y. J. Am. Chem. Soc. 2001, 123,

5684-5694. b) Nelsen, S.F.; Tran, H.Q. J. Phys. Chem. A 1999, 103, 8139-8144.

24. Di Saverio, A.; Focante, F.; Camurati, I.; Resconi, L.; Beringhelli, T.; D`Alfonso, G.;

Donghu, D.; Maggioni, D.; Mercandelli, P.; Sironi, A. Inorg. Chem. 2005, 44 (14), 5030-

5041.

25. It is worth noting that attempts to generate 1•+ and 4d•+ quantitatively by the use of strong

oxidizing agents (SbCl5 or N(C6H4Br)3SbCl6) in toluene at ambient conditions resulted in

that the rapid formation of a precipitate.. Such species generally need to be formed in

coordinating/polar solvents to remain soluble. See Connelly, N.G.; Geiger, W.E. Chem.

Rev.1996, 96, 877-910

26. Bender, T.P.; Graham, J.F.; Duff, J.M. Chem. Mater. 2001, 13, 4105.

95

95

Chapter 7: Hole Mobility of a Liquid Organic Semiconductor


This chapter was published as a letter in the Journal of Physical Chemistry Letters.

Brett A. Kamino, Timothy P. Bender, Richard A. Klenkler (2012), Journal of Physical

Chemistry Letters, 3 (8), 1002-1006. Figure and Schemes are reprinted with permission. Copyright 2011 American Chemical Society Over the course of the previous chapters (4-6), we demonstrated several methods to produce

liquid organic semiconductors from traditional semiconductors while retaining their useful

electrochemical properties. However, in order to evaluate their potential use in electronic

devices, more characterization was necessary. One of the most critical parameters in organic

semiconductors is the ability of the material to move either positive or negative charge. This

parameter is known as the charge carrier mobility and its value has many implications for

organic electronic devices. For liquid organic semiconductors, this value had only been measured

once in the literature and the measurements were not performed in depth. This left many

questions about the mechanism of transport and whether or not this complied with current

theories.

While working on this paper, I worked at the Xerox Research Centre of Canada to develop a

technique to measure the charge carrier mobility of our one of our materials and to investigate

how charge transport through the liquid state behaved when put in current theoretical models.

This project was very successful and I managed to develop a simple and practical method to

measure the charge carrier mobility of our liquids by modifying an existing setup developed by

Xerox. I was able to show that our materials do indeed move charge throughout the liquid state

and that the material behaved very similarly to conventional materials. As well, this behavior fit

very nicely into existing charge transport theories.

From this study, we can surmise that our liquid organic semiconductor materials retain their

useful charge transporting properties despite the large change in physical properties. This bodes

well for the eventual use of liquid organic semiconductors in organic electronic devices.

96

96


The authorship of this paper is as follow: Brett A. Kamino, Timothy P. Bender, Richard

Klenkler. All experiments in their entirety were performed by me as all as all data analysis and

data interpretation. Richard Klenkler is a researcher at the Xerox Research Centre of Canada

familiar with the analytical technique used. The paper was written by me with input from both

T.P. Bender and R. Klenkler.

7.3 Paper

7.3.1 Abstract

The first detailed study of charge transport through a liquid organic semiconductor (LOS) is

reported with the goal of elucidating the effects of molecular motion on charge transport through

molecular liquids. Using a liquid, silyl ether-substituted triarylamine, hole transport mobilities

were obtained over a wide range of temperatures above the glass transition temperature of the

material. Analysis of this data reveals that molecular motion(s) have a negligible effect on

macroscopic charge transport through a molecular liquid. The results strongly resemble transport

behavior found in conventional, disordered solids and suggest that silyl ether-substituted LOSs

may be good candidates for integration into electronic devices, by those who are familiar with

the application of traditional triarylamines, where their unique physical state can or could be

exploited.

7.3.2 Body

Liquid organic semiconductors (LOSs) are an emerging class of materials for organic electronic

devices. While conventional organic semiconducting materials are designed to form highly

crystalline solids or morphologically stable glasses,1 LOSs are specifically intended to be free-

flowing liquids at room temperature. Such materials present several unique processing

advantages over their more typical counterparts, including solvent-free device processing and the

presumed ability to easily achieve intimate contact with nanostructured or mesoporous surfaces.

The utility of LOSs has been demonstrated in dye-sensitized solar cells,2 photorefractive

devices,3 as host materials in liquid active layer light emitting diodes,4,5 and several other device

types.6,7 However, this class of materials remains understudied. Until recently, only two LOSs

were known in the literature: tris(4-methoxyethoxyphenyl)amine (TMEPA)2 and N-(2-

97

97

ethylhexyl)carbazole.3−5 Our group has worked to expand the available set of LOSs by

developing several general synthetic methods that, by incorporation of siloxane or silyl ether

molecular fragments into conventional arylamine molecular structures (triarylamines and

carbazoles), yield LOSs.8−10 We have shown that these silicon−arylamine hybrid materials retain

their characteristic electrochemical properties while having vastly different physical properties

from their parent materials, including presenting themselves as liquids and waxes. Despite the

success of this strategy in producing a number of new LOSs, it remained unclear what effect the

substitution of the siloxane or silyl ether groups would have on the hole-transporting properties

of these materials, as charge transport through the liquid phase is poorly understood for organic

semiconductors. We are aware of only a single example of a charge transport measurement

through an LOS.11 In that report, Ribierre et al. show that the liquid semiconductor, N-(2-

ethylhexyl)carbazole, acts as a p-type semiconductor and has an order of magnitude higher hole

mobility than its equivalent polymeric, solid-state analogue (poly(N-vinylcarbazole)).

Additionally, the transport characteristics of several solid molecular semiconductors have been

studied above their glass transition, but this type of study is limited by the possibility of

crystallization of the material above the glass transition temperature (Tg) and the relatively high

temperatures required to achieve a liquid state. In this work, we study the hole-transporting

ability of a silyl ether-substituted triarylamine LOS: N,N-bis(4-(triisopropylsilyloxy)phenyl)-3,4-

dimethylaniline (2TIPS, Figure 7-1a), a material that is a liquid at room temperature (Tg = −28

°C). Time-of-flight photocurrent measurements12 were performed on samples of 2TIPS doped

into a solid polymeric matrix and of 2TIPS as a neat liquid. The goals were to validate the

charge transporting capability of a silyl ether-substituted triarylamine and to begin to understand

whether charge transport in the liquid phase adheres to conventional theories explaining charge

transport through amorphous molecular solids. In particular, we were interested to see whether

the charge transport characteristics of an LOS can be described by the disorder formalism model

using 2TIPS as a model system. Developed by Bässler and co-workers,13 this model has been

used to successfully describe the transport behavior seen in many conventional organic

semiconducting materials, including those based on triarylamines. 2TIPS (Figure 7-1a) was

chosen as a model compound, because we have previously shown that it has predictable

electrochemical behavior8 and is a member of the well-studied triarylamine family of hole

transport compounds.14

98

98

a)

5.00E-04 5.00E-03 5.00E-02

Vo

lta

ge

(a

.u.)

Time (s)

ttransit

CG

L

Al

Al

2T

IPS

/Po

ly(s

tyre

ne

)

Laser Pulse

+-

5 x 10-3 5 x 10-25 x 10-4

(b)

2.00E-03 2.00E-02 2.00E-01

Vo

lta

ge

(a

.u.)

Time (s)

ttransit

CG

L

Al

Ka

pto

n

Sp

ace

r

Ka

pto

n

Sp

ace

r

Al

2T

IPSLaser Pulse

+-

TP

D

2 x 10-22 x 10-3 2 x 10-1

Figure 7-1: Example photogenerated transients through (a) 2TIPS in a poly(styrene) matrix (9 um) and (b) neat 2TIPS (50 um) with the calculated transit time shown (Note: y-axis is linearly scaled while x-axis is logarithmic). Device structures are illustrated next to their respective transients and the chemical structure of 2TIPS is shown.

99

99

We began our study by investigating the basic transport properties of 2TIPS doped into an inert

polymer matrix. Polymers commonly used include polystyrene and polycarbonate-A.

Polystyrene was chosen as the inert matrix as it formed higher quality films when doped with

2TIPS than did polycarbonate-A. The time-of-flight cell (Figure 7-1a, inset) was assembled in a

similar manner to what we have previously described.15 Specifically, a Mylar substrate metalized

with an aluminum electrode was sequentially coated with a silane blocking layer, a charge

generator layer (CGL, composed of a hydroxygallium phthalocyanine dispersion in a polymer

binder) and a blade coated hole-transport layer consisting of 2TIPS and polystyrene at a 1:1 ratio

(by weight). The cell was completed by pressure contact with a second metalized Mylar

electrode. The time-of-flight measurements were performed as described by Melnyk and Pai,12

using a nitrogen pulse laser with a dye attachment (Laser Science VSL-337ND-S & DUO-220).

The output laser wavelength was tuned to 650 nm, so as to be only absorbed by the CGL, and

laser intensity was adjusted such that less than 1/10 CV (where, C is the device capacitance and

V the applied bias) of charge was injected; this ensured that charge transport would not be space

charge limited.12 Signal transients were collected at different field strengths, transport layer

thicknesses, and temperatures. These measurements showed characteristics typical of

triarylamines blended with an inert polymer (Figure 7-1a).16 In order to validate our technique,

transients were measured as a function of transport layer thickness over a range of 5−37 µm.

Transit times (and hence mobility) were found to scale linearly with thickness (Figure S7-1,

section 12.5), confirming that the transient signals are indeed due to transport across the entire

thickness of the transport layer and that the calculated mobilities were not an artifact of the

system. Field-dependent mobility over a temperature range of −30 to 65 °C was measured and is

shown in Figure 7-2a for a 9 µm thick hole-transport layer. The transients were non-dispersive as

indicated by a plateau of current followed by a well-defined drop in current defining the transit

time. Hole-transport mobilities were found to be weakly field dependent across the temperature

range (Figure 7-1a). Comparing measurements at variable temperatures, we find that the mobility

is strongly temperature activated until we reach the glass transition temperature of the polymer

blend (measured to be 48 °C)17 where this dependence plateaus. This behavior at the glass

transition temperature has been previously observed in doped polymers,18 molecular glasses,19

and polymeric semiconductors.20 In order to see whether 2TIPS behaved similarly to other

triarylamines, the results were analyzed within the context of the disorder formalism:

100

100

Σ−

−= 2/12

22

0 exp3

2exp),( E

kTC

kTET

σσµµ 7-1

where µ is the mobility, µ0 is the mobility at infinite temperature and zero field, σ is the width of

the density-of-states in eV, k is the Boltzmann constant, T is the absolute temperature, Σ

describes the positional disorder, C is an empirical constant, and E is the applied electric field.

This equation describes the dependence of charge carrier mobility on temperature and electric

field and predicts that log µ should scale linearly with the inverse square of temperature and the

square of the electric field. Excluding the points at and above the glass transition temperature,

the data collected for the 2TIPS/polystyrene films shows the expected dependence (Figure 7-2a).

By extrapolating the mobilities to an electric field of zero, the values of µ0 and σ were found to

be 1.92 × 10−3 cm2 V−1 S−1 and 0.104 eV, respectively. When compared to other triarylamines

found in the literature, these values are in line with other well-known, non-liquid triarylamines,

including TPD and TAPC.21,22

a)

1.00E-07

1.00E-06

1.00E-05

300 500 700

Ho

le M

ob

ilit

y (

cm

-2v

-1s

-1)

E1/2 (V/cm)1/2

+ 65 °C

- 30 °C

+ 20 °C

1 x 10-6

1 x 10-5

1 x 10-7

101

101

(b)

1.00E-07

1.00E-06

1.00E-05

1.00E-04

150 250 350

Ho

le M

ob

ilit

y (

cm

2V

-1s

-1)

E1/2 (V/cm)1/2

1 x 10-5

1 x 10-7

1 x 10-4

- 40 °C

+ 60 °C

+ 20 °C

1 x 10-6

Figure 7-2: Field dependent hole mobility as a function of temperature for 2TIPS in a (a) polystyrene matrix and (b) as a neat liquid.

For the time-of-flight measurement of 2TIPS as a neat liquid, a different cell design was

required. Empty cells were assembled by sandwiching two strips of Kapton (polyimide) film

between the same two Mylar substrates described above and gently clipping them together

(Figure 7-1b, inset). Once prepared, the empty cell was heated to 50 °C, and the neat 2TIPS was

drawn into the cell by capillary action. Unfortunately, this setup produced weak photocurrents

and highly dispersive transients, making accurate determination of the mobilities impossible. It

was found that adding a thin (275 nm) layer of TPD on top of the hydroxy gallium

phthalocyanine layer by physical vapor deposition increased photocurrent and improved the

resolution of the transient so that a sharp and distinct transit time could be observed and defined.

The cell design and an example transient are illustrated in Figure 7-1b. The additional TPD layer

seems to improve charge injection between the phthalocyanine layer and 2TIPS, but the rising

plateau in the transient is indicative of the slow release of a well of charge in the CGL. This

manifests itself in a measured current that steadily increases over time until the transit time

(Ttransit, Figure 7-1b) is reached, possibility indicating a charge injection barrier. If a charge

injection barrier exists on a time scale similar to the transit time, the results of this experiment

would certainly be invalidated.23 To investigate this possibility and validate the fidelity of the

102

102

technique, transients were measured as a function of transport layer thickness over a range of

50−125 um (as defined by the thickness of the Kapton spacer). Again, transit times and mobility

were found to scale linearly with thickness of the 2TIPS layer (Figure S7-2, section 12.5),

thereby confirming that the mobility values being measured are independent of any charge

injection issues.

As a point of comparison, we measured the mobility of N-(2-ethylhexyl)carbazole, an LOS with

a known hole transport mobility.11 The compound was synthesized by the method outlined in the

9.5.1, and a time-of-flight cell with this material was prepared in the same manner as described

above (Figure 7-1b), including the additional 275 nm TPD interlayer. Transients were obtained

over a variety of fields, and the field-dependent mobilities agreed well with those previously

reported for this material, albeit at a lower field strength (Figures S7-3 and S7-4, section 12.5).

This confirmation of the known hole-transport mobility of N-(2-ethylhexyl)carbazole provides

further validation of our experimental setup.

Using a 2TIPS cell with a 50 um Kapton spacer and the TPD interlayer, field-dependent

transients were collected between −40 and 60 °C (Figure 7-2b). However, it was only possible to

collect transients through a narrow range of field strengths. For most samples, increasing the

field strength over 1 × 105 V/cm resulted in shorting the device. At very low temperatures, this

effect was not observed, and higher field strengths were found to be necessary to achieve

measurable signals. Much like the 2TIPS/polystyrene samples, results for the neat 2TIPS cells

were interpreted within the context of the disorder formalism (eq. 7-1). From Figure 7-2b, we

can see that log µ scales with E1/2 showing a weak dependence on field strength like that which

was observed for the solid sample. Plotting the temperature dependent mobility at different field

strengths again shows a linear relationship between log µ versus T−2 similar to the data observed

for the solid 2TIPS/polystyrene system (Figure 7-3). Interestingly, the majority of the data points

obtained on this line correspond to values above the Tg of 2TIPS (Tg = −28 °C).9 This

relationship implies that charge transport through liquid 2TIPS is governed by temperature-

activated hopping, as predicted by the disorder formalism and as is seen in most disordered

solids. This relationship holds true for the entire range of values above the Tg of this compound,

a span of approximately 90 °C. This represents the first investigation into the effect of

temperature on charge mobility in an organic molecular liquid and invites discussion into the role

103

103

of molecular motion on the charge transport process. Analysis of the extrapolated zero field

mobilities gives µ0 and σ values of 8.89 × 10−2 cm2 V−1 s−1 and 0.115 eV, respectively.

Comparing these parameters to those determined for the solid 2TIPS/ polystyrene system, we

find that the prefactor mobility (µ0) scales with the increased hole-transport mobilities observed

in this sample while the width of the density-of-states (σ) is somewhat larger than what was

determined for the solid sample. This is contrary to the expected trend found in typical materials.

In most solid-state examples, it is expected that the density of states decreases in energy when

going from a dilute sample (2TIPS doped into polystyrene) to a more concentrated sample (neat

2TIPS).24 In this case, the observed increase in the density of states can be explained by the

increased local motions of individual hole transport molecules when going from a solid to a

liquid.25

1.00E-07

1.00E-06

1.00E-05

1.00E-04

1.00E-03

6 11 16(1000/T)2 (K-2)

Neat 2TIPS

2TIPS/poly(styrene)

Tg1 x 10-6

1 x 10-7

1 x 10-5

1 x 10-4

1 x 10-3

Mo

bilit

y (

cm

2V

-1s

-1)

Figure 7-3: Temperature dependence on hole mobility for 2TIPS doped in polystyrene (50 wt%) at 555 kV/cm and neat 2TIPS at 100 kV/cm. The glass transition temperature of the polymer blend is indicated as Tg at 48 °C.

Temperature-driven changes in the density-of-states due to molecular motion and solvation

energy have been proposed to describe deviations in temperature-dependent mobility behavior

above the glass transition temperature in solid systems.26 However, this is still a poorly

understood phenomenon that produces very different behaviors for various material types. For

most doped polymer systems (as observed for 2TIPS/polystyrene in this letter), the glass

104

104

transition is characterized by the onset of temperature-independent mobility. Other systems have

shown an inversion in the mobility dependence20 or even a continuation of normal behavior well

above the glass transition temperature.19 Our LOS was studied well above its glass transition and

displays quite pedestrian behavior throughout this range. In fact, the only difference found was a

slight increase in the density of states compared to the solid polymeric sample. If molecular

motions were important to charge transport through a LOS, we would expect a deviation from

normal temperature-dependent behavior due to an increase in physical diffusion at higher

temperatures. The absence of such an effect strongly suggests that molecular motions have a

negligible effect on the hole-transport mobility of our system.

In summary, the charge transporting properties of a silyl ether-substituted liquid triarylamine,

2TIPS, was studied. As both a dopant in an inert polymer matrix and as a neat liquid, the charge

transport molecule exhibited conventional hole transporting properties when studied using a

time-of-flight technique. Most interestingly, the charge mobility of neat 2TIPS closely followed

the temperature dependence predicted by the disorder formalism model for solid materials. This

behavior is observed over a large range of temperatures above the Tg, and we conclude that

molecular motion has a negligible effect on macroscopic charge transport in this material save

for a small increase in the density of states. The conventional charge mobility behavior of these

materials suggests that silyl ether-functionalized LOSs may be excellent targets for integration

into novel, liquid electronic devices by those who are familiar with traditional solid triarylamines

but who might want to exploit the novel liquid state of LOSs.

7.3.3 References

1. Shirota, Y.; Kageyama, H. Chem. Rev. 2007, 107, 953,-1010.

2. Snaith, H.J.; Zakeeruddin, S.M.; Wang, Q.; Péechy, P.; Grätzel, M. Nano. Lett. 2006, 6, 2000.

3. Ribierre, J.; Aoyama, T.; Kobayashi, T.; Sassa, T.; Muto, T.; Wada, T. J. Appl. Phys. 2007, 102, 033106.

4. Hirata, S.; Kubota,, K.; Jung, H.H.; Hirata, O.; Goushi, K.; Yahiro, M.; Adachi, C. Adv.

Mater. 2011, 23 (7), 889-893.

5. Xu, D.; Adachi, C. Appl. Phys. Lett. 2009, 95, 053304.

6. Hendricks, E.; Guenther, B. D.; Zhang, Y.; Wang, J. F.; Staub, K.; Zhang, Q.; Marder, S. R.; Kippelen, B.L.; Peyghambarian, N. Chem. Phys. 1999, 245, 407.

105

105

7. Ribierre, J.C.; Aoyama, T.; Muto, T.; André, P. Org. Electron. 2011, 12 (11), 1800-1805.

8. Kamino, B.K.; Grande, J.B.; Brook, M.A.; Bender, T.P. Org. Lett. 2011, 13, 154-157

9. Kamino, B.K.; Castrucci, J.; Bender, T.P. Silicon, 2011, 3 (3), 125-137.

10. Kamino, B.K.; Mills, B.M.; Reali, C.; Bender, T.P. J. Org. Chem. 2011, 77 (4), 1663-

1674.

11. Ribierre, J.-C.; Aoyama, T.; Muto, T.; Imase, Y.; Wada, T. Org. Electron. 2008, 9, 396-400.

12. Melnyk, A.R.; Pai, D.M. Physical Methods of Chemistry, 2nd ed., Vol. VIII, Rossiter, B.W.; Baetzold, R.C., Eds.; Wiley, New York, 1993; Chapter 5, 321-386.

13. Auweraer, M.V.; Schryver, F.C.D.; Borsenberger, P.M.; Bässler, H. Adv. Mater. 1994, 6 (3), 199-213.

14. Borsenberger, P.M.; Weiss, D.S. Organic Photoreceptors for Xerography; Marcel Dekker, Inc.: New York, 1998.

15. Klenkler, R.A.; Xu, G.; Graham, J.F.; Popovic, Z.D. App. Phys. Lett. 2006, 88, 102101

16. Borsenberger, P.M. J. Appl. Phys. 1990, 68, 6263-6273.

17. Glass transition of polymer blend determined through differential scanning calorimetry.

18. Abkowitz, M.; Stolka, M.; Morgan, M. J. Appl. Phys. 1981, 52(5), 3453-3457.

19. Bässler, H.; Borsenberger, P.M. Chem. Phys. 1993, 177, 763-771.

20. Abkowitz, M.A.; McGrane, K.M.; Knier, F.E.; Stolka, M. Mol. Cryst. Liq. Cryst. 1990, 183, 157-169.

21. Borsenberger, P.M.; Gruenbaum, W.T.; Magin, E.H.; Sorriero, L.J. Chem. Phys. 1995, 195, 435-442.

22. N,N`-diphenyl-N,N`-bis(3-methylphenyl)-(1,1`-biphenyl)-4,4`-diamine (TPD) and 1,1-bis(di-4-tolylaminophenyl)cyclohexane (TAPC).

23. Chen, I. Jpn. J. App. Phys. 1989, 28, 21.

24. Schein, L.B.; Borsenberger, P.M. Chem. Phys. 1993, 177, 773-781.

25. Borsenberger, P.M.; Pautmeier, L.; Bässler, H. J. Chem. Phys. 1991, 95 (2), 1258-1265.

26. Bässler, H. Adv. Mater. 1993, 5 (9), 662-665.

106

106

Chapter 8: Crosslinked Triarylamine-Siloxane Films using Piers-Rubinsztajn Chemistry


This chapter is a draft of paper that will be submitted as a communication to Macromolecular

Rapid Communications.

The author list will be as follows:

Brett A. Kamino, Anjuli Szawiola, Ishan Gupta, Michael J. Gretton, Timothy P. Bender.

This work represents a conceptual extension to the work done in the previous chapters on

creating siloxane-triarylamine hybrid materials. In this work, we try to exploit the previously

developed chemistry and methods to create cross-linked organic semiconductor films instead of

liquid organic semiconductors. Such cross-linked films have the potential to be more easily

incorporated into organic semiconducting devices than liquid organic semiconductors. Such

materials have already been demonstrated successfully in a number of devices. This chapter is

the final project on my main thesis project and demonstrates that our synthetic approach to

siloxane-triarylamine hybrid materials can be extended to other forms of matter relatively easily.

This chapter is a proof of concept of this particular approach and further optimization and

integration into devices will be one of the challenges left to future graduate students.


The author list for this paper is as follows: Brett A. Kamino, Anjuli Szawiola, Ishan Gupta,

Michael J. Gretton, Timothy P. Bender. I came up with the primary concept for this project and

the designed of all materials and processes used. As well, I led the development of experiments,

performed most of the analytical technqiues myself, and solely analyzed and processed all data.

Anjuli Szawiola and Inshan Gupta assisted with synthesis of compounds and elaboration of

synthetic techniques developed by me. Additionally, Anjuli Szawiola performed a portion of the

IR and photoluminescence expeirments under my direction. Writing of this chapter was

performed by me with input from both Anjuli Szawiola and Michael J. Gretton.

107

107

8.3 Paper Draft

8.3.1 Abstract

A cross-linked thin film with a triarylamine monomer has been synthesized by Piers-Rubinsztajn

chemistry from readily available materials to be used as a charge-transport layer in organic

electronic devices. The triarylamine compound was synthesized and functionalized with a cyclic

ether that upon ring-opening, participates in cross-linking with tetrakis(dimethylsiloxy)silane

forming smooth, glassy films. Various catalyst loadings and curing temperatures were tested to

determine their effect on film quality. It was found that the films were electrochemically active

and amorphous, with low glass transition temperatures. In addition, while curing temperature

was found to have little effect on monomer aggregation, catalyst loading had a substantial

influence, presumably due to the higher cross-link density formed with a higher catalyst loading.

8.3.2 Introduction

Silane activation by tris(pentafluorophenyl)borane,1 or the Piers-Rubinsztajn reaction,2 is a

powerful method for the construction of silicon heteroatom bonds. Using this chemistry, a

number of organic reactions such as hydrosilylation of carbonyl groups,3 silyl ether formation

through dehydrocarbonative condensation,4 and metal-free reductions5 can be achieved using

mild reaction conditions with good functional group tolerance.6 Additionally, this chemistry

benefits from the use of easily handled and commercially available silanes as a silicon source.

Beyond simple organic transformations, the Piers-Rubinsztajn reaction has been used to produce

a number of functional materials and polymers. Among these, Rubinsztajn and coworkers have

shown how this chemistry can be used to rapidly build well-defined siloxane-organic copolymers

from simple starting materials.7,8 Other examples of this reaction’s use with silicones include the

synthesis of explicit siloxane dendrimers9 and silicone surfactants.10 More recently, our group

has found that the Piers-Rubinsztajn reaction is a convenient method for the functionalization of

charge transporting materials with silicone or siloxane groups. Using this strategy, we have been

able to prepare a range of different siloxane/charge transporting hybrid materials with extremely

varied physical properties ranging from polymeric glasses11 to low Tg molecular liquids.12,13

Despite the large change in physical properties that results from this transformation, we have

shown that the useful electronic properties of the charge transporting materials remain

unchanged.14

108

108

Given the proven utility of Piers-Rubinsztajn chemistry for producing functional organic

semiconducting materials, one may consider the possibility of using this chemistry to produce

cross-linked films incorporating organic charge transporting molecules. Cross-linked films

containing charge transporting layers have many applications in organic electronic devices and

many advantages over conventional small molecules and polymers. These advantages include

improved morphological stability,15 and the ability to solution deposit additional layers on top of

the cross-linked layer.16 Such chemically cross-linked layers have been successfully

demonstrated in organic photovoltaics,17 organic light emitting diodes,18 and organic field effect

transistors.19 Interestingly, several silicone based cross-linking systems have been previously

reported and shown to be immensely useful as charge transporting layers in organic electronic

devices.20-22 In these previously reported silicone systems, charge transporting compounds were

functionalized with trichloro or trialkoxy silanes and polymerized by sol-gel chemistry.23 Thin

films made with these systems have been successfully used in OLEDs as hole transporting

compounds as well as emitting layers.

While crosslinking formulations based on silicone polycondensation methods have shown to be

very effective as layers in electronic devices, there are some synthetic disadvantages inherent to

this chemistry. First, this chemistry requires the introduction of water and an acid into the

formulation. Water has been shown to be one of the chief factors in device degradation and even

miniscule amounts can limit the life span of organic electronic devices.24,25 The second major

disadvantage is the need to functionalize the charge-transporting molecule with a reactive silane

group. This additional functionalization step increases the total number of synthetic steps

required and adds difficulty to the purification and isolation of the final monomer.

In this communication, we demonstrate how Piers-Rubinsztajn chemistry can be used to produce

high quality cross-linked films from very stable and easily synthesized reagents under non-

hydrolytic conditions. The key to this achievement is the use of cyclic ethers as substrates rather

than hydroxyl alkoxy groups. Film formulation is explored by studying the thermal,

photophysical, and electrochemical properties of these films.

109

109


Before we began the development of a cross-linking system based on Piers-Rubisnztajn

chemistry, we had to deal with aspects of this chemistry that were not favourable towards thin-

film formation. Specifically, our previous conditions11-13 for the addition of siloxane fragments

to triarylamine structures involved reacting aryl methoxy groups with silanes using

tris(pentafluorophenyl)borane (BCF) to promote the reaction. Under these conditions, the

reaction results in the rapid release of stoichiometric amounts of hydrocarbons and a noticeable

amount of heat. This uncontrollable reaction speed at room temperature and rapid release of

gases poses certain problems when trying to reproducibly obtain smooth cross-linked films. The

vigorous nature of this chemistry has in fact been recently shown to produce elastomeric foams

under certain conditions.26

In order to develop a cross-linking system based on Piers-Rubinsztajn chemistry incorporating

charge transporting triarylamines, we needed a formulation which reacts more controllably and

does not release any volatile gases. To achieve this, we chose to synthesize a triarylamine27

based monomer incorporating two 2,3-dihydrobenzofuran groups to react in our cross-linking

system. Using dihydrofuran groups instead hydroxyl or alkoxy groups to react with a silane

offers several key advantages. Firstly, when reacted with a silane under Piers-Rubinsztajn

conditions, cyclic ethers ring-open instead of displacing a volatile hydrocarbon.5 This eliminates

the problem of bubble formation inside the film. The second advantage is a slower rate of

reaction due to non-productive binding of the catalyst to the Lewis-basic dihydrofuran and

decreased nucleophilicity of the cyclic ether as compared to an alkoxy or hydroxyl group.†

110

110

N N

O

Si

O

Si O

Si

O

Si

O O

Br

8-1

(i)

(iii)

8-4

(ii)

N

N

O

O

8-3

(iv)

NH2

O

HN

8-2

8-2

Scheme 8-1: Synthesis of an arylamine monomer for ring-opening under Piers-Rubinsztajn

conditions. (i) NBS, THF, rt, 16hrs, 89% (ii) Pd(OAc)2, P(t-butyl)3, NaO(t-butyl), toluene, 110°C

70%. (iii) Pd(OAc)2, P(t-butyl)3, NaO(t-butyl), toluene, 110°C, 85%. (iv) 1,1,1,3,3-

pentamethyldisiloxane, tris(pentafluorophenyl)borane, toluene, rt, 78%.

2,3-Dihydrobenzofuran was incorporated into the structure of the arylamine by first brominating

it using NBS and then using successive Buchwald-Hartwig28 coupling steps to form the

arylamine monomer (8-3) (Scheme 8-1). Before attempting any cross-linking and film formation

with this monomer, we reacted it with a mono-functional silane (1,1,1,3,3-

pentamethyldisiloxane) under standard Piers-Rubinsztajn conditions to test its reactivity. Under

these conditions, the desired ring-opening reaction proceeded smoothly over the course of

several minutes releasing a noticeable exotherm and no gaseous byproducts to yield compound

8-4 as confirmed by 1H, 29Si, and 13C NMR and HRMS. This small molecular analogue was

isolated as a pale yellow glass which crystallized on sitting over several months.

With confirmation of the desired reactivity in our monomer, we set about synthesizing cross-

linked films with a number of poly-functional silanes. Our initial attempts to use commercial

111

111

poly(methylhydrosiloxane-co-dimethylsiloxane) and poly(methylhydrosiloxane-co-

phenylmethylsiloxane) as comonomers failed to produce homogeneous films under a number of

conditions. These films displayed significant phase separation, poor surface quality, and

somewhat temperamental curing conditions often resulting in unreacted polymer remaining. We

had greater success using tetrakis(dimethylsiloxy)silane (QM4) which consistently resulted in

high quality films upon drying of the solvent (Scheme 8-2). We suspect that this success is due

to better phase compatibility of the two components and more favourable kinetics resulting from

a less sterically hindered silane group. Using these components, smooth glassy films were

produced by drop casting the formulations on glass substrates. The resulting films remained

optically clear over 2 months of storage in ambient conditions and did not appreciably swell

when soaked in common non-polar solvents such as toluene and chloroform.

Scheme 8-2: Reagents used to achieve functional cross-linked films using Piers-Rubinsztajn

chemistry. (v) tris(pentafluorophenyl)borane, toluene.

112

112

Table 8-1: Collected experimental information for different formulation conditions.

To explore the scope and versatility of this crosslinking system, six formulation conditions were

studied which varied in curing temperature and catalyst loading (Table 8-1). Under all

conditions, smooth, homogenous, and glassy films were obtained by drop casting the solutions

onto glass slides and curing the films for 30 minutes in air.‡ To understand the extent of the

reaction, IR spectroscopy was performed on the films and compared to QM4 to determine if

residual Si-H groups remained in the film. The films using 1-2 mol% catalyst relative to QM4

(Films A-E) all looked very similar by IR and did not show any residual Si-H stretching peak

(see section 12.6). The film with the lowest catalyst loading (Film F) however, did show a small

amount of residual Si-H peak in the IR (Figure 8-1). From these spectra we can conclude that

under most of our film formation conditions, virtually quantitative consumption of Si-H groups

is observed. At first, this observation is somewhat curious given the small excess of Si-H groups

in each formulation (~4 mol%). However, since the cross-linking reaction is performed in air, it

is likely that adventitious water in the atmosphere is able to react with Si-H to condense two

QM4 groups together.1 From this IR study we can conclude that 1 mol% BCF catalyst is required

for full consumption of Si-H groups under these conditions.

113

113

65011501650215026503150

Tra

nsm

itta

nce

Wavenumber (cm-1)

0.5 mol%

1 mol%

2 mol%

QM4

Si-H

F

E

D

Figure 8-1: IR spectra of QM4 and films 6, 5, and 4 (top to bottom). Film D prepared in a matrix

of KBr and the remainder studied by ATR.

The films were then characterized by differential scanning calorimetry (DSC) and thermal

gravimetric analysis (TGA) to gain further insight into their morphology and structure. Using

DSC, we found that all of the films displayed a single glass transition just above room

temperature on both the heating and cooling cycles (Tg, Table 8-1) with no crystalline transitions

observed. TGA performed on the films revealed a bimodal decomposition pattern for all films.

In each case, five percent mass loss occurred at fairly low temperatures (~140 to 170 °C, Table

8-1) followed by a plateau at around 90-93% mass. Decomposition accelerated again above ~400

°C. As a point of comparison, TGA was also run on our model compound (8-4) which showed

only a single, sharp decrease in mass with 5 wt% loss at 326 °C. The significant mass loss in the

films at relatively low temperatures can be explained by the likely presence of small amounts of

oligomers and trapped solvent within the film itself. Minimizing this loss of material is an

important consideration in the eventual use of these films in organic electronic devices.

Interestingly, we can correlate the glass transition temperature of the films to the thermal

stability. Films with higher thermal stability also have higher glass transition temperatures. We

114

114

suspect that this correlation is reflective of the effective cross-link density in the films and that

films with the highest Tg or Td have the highest cross-link density. From these data, several

trends in the formulation parameters can be identified. Catalyst concentration has a positive

effect on cross-link density of the films which agrees well with the previously mentioned IR

results. Looking at the effect of temperature, we find that films cured at 50 °C have highest

cross-link density compared to films cured at higher or lower temperatures. This can be

explained by two competing temperature effects during film formation. Higher reaction

temperatures likely increase the overall curing rate, but higher temperatures also increase the rate

of solvent evaporation. Because of the need for catalyst mass transport in this system, solvent

inclusion is probably necessary for a high cross-linking density.

To help understand the conformation of the redox active triarylamine monomers within the

films, both the films themselves and the model compound (8-4) were studied by

photoluminescence (PL) spectroscopy. PL spectra of compound 8-4 was obtained in dilute THF

solutions and in the solid-state (Figure 8-2a). A significant red shift was observed between the

two samples with the solid state sample having a λmaxex of 449 nm versus 410 nm for the same

compound in solution. This red shift is indicative of intermolecular aggregation affects and can

serve as a good tool to probe the intermolecular interactions of the cross-linked films.29,30 All six

films were studied in the solid-state. Films A, B and C were found to have near identical spectra

with a λmaxex of ~432 nm (Figure 8-2b, Table 8-1) with a smaller shoulder at shorter wavelengths.

A peak shape comparison to the model compound reveals that some intermolecular aggregation

is occurring but curing temperature in this range has a small impact on this effect. Interestingly,

films at a higher temperature (D, E and F) were found to have a more red shift spectrum with F

having the greatest amount of red shifting. From the thermal characterization experiments, we

suspect that these films have lower cross-linking densities. This may allow for more aggregation

between specific components in the solid-state and result in a more red-shifted

photoluminescence spectrum. Spectra for the remaining films are illustrated in section 12.6.

115

115

0

0.2

0.4

0.6

0.8

1

0

0.2

0.4

0.6

0.8

1

285 335 385 435 485 535 585

No

rma

lize

d P

L I

nte

ns

ity

Ab

so

rba

nc

e (

a.u

).

Wavelength (nm)

0

0.2

0.4

0.6

0.8

1

385 435 485 535 585

No

rmali

zed

PL

In

ten

sit

y

Wavelength (nm)

A D F

Figure 8-2: (a) UV-Vis absorption (black) and photoluminescence spectra of compound 4 in a

THF solution (red) and a neat film (blue). (b) Photoluminescence spectra of films A, D, and F on

glass.

Finally, electrochemical characterization was performed on both the small molecule model

compound (4) and film D. Compound 8-4 was studied by solution cyclic voltammetry in DCM

(Figure 8-3) and displayed two reversible oxidations at 655 mV and 910 mV vs. Ag/AgCl. This

behaviour is fairly typical for bis-arylamines. Film D was studied as a thin film on ITO in both

acetonitrile and aqueous solutions. The films studied in acetonitrile yielded two poorly resolved

oxidation waves similar to compound 8-4 (Figure 3). This oxidation was found to be only

partially reversible as subsequent scans yielded smaller and smaller currents. Differential pulse

voltammetry on a fresh film was performed to help resolve each oxidation peak and yielded two

oxidations with peaks at 790 mV and 905 mV vs. Ag/AgCl. Films studied in aqueous solutions

only showed a single, fully irreversible oxidation peak (see section 12.6). Subsequent voltage

sweeps on these films yielded zero redox activity. Regardless of this instability, the behaviour of

these films on ITO confirms that the cross-linked films remain electrochemically active after

reaction. The differences in redox potentials between the films in acetonitrile and the model

compound (8-4) can be attributed to differences in working electrode material, solvent, and

differences in molecular conformation in the solid-state.

116

116

-0.4 -0.2 0 0.2 0.4 0.6 0.8 1 1.2

Cu

rre

nt

(a.u

.)

Voltage (vs. Ag/AgCl)

Reference

(a)

(b)

(c)

Figure 8-3: Electrochemistry with decamethylferrocene internal reference. (a) cyclic voltammetry of 8-4 in DCM. (b) cyclic voltammetry of film D on ITO in acetonitrile. (c) differential pulse voltammetry of film D on ITO in acetonitrile.

8.3.4 Conclusions

We have successfully demonstrated the first use of Piers-Rubinsztajn chemistry to achieve cross-

linked silicone films with a redox active component. The use of cyclic ethers proved to be

critical in formulating compositions that reproducibly and controllably reacted to give smooth

glassy films. Complete consumption of all Si-H groups was found to be achieved with low

catalyst loadings and the resulting films were found to be amorphous with low glass transition

temperatures. Photophysical characterization of the films showed that minimal aggregation of

the redox active triarylamine monomers could be achieved by altering the formulation

parameters. Electrochemical characterization of the films confirmed that they retained the

electrochemical activity of the monomer. Future work will include exploring how other film

properties can be further tuned using new monomers and integration of the current films in

organic electronic devices.

117

117

8.3.5 References

† This decrease in reactivity can be evidenced by the reaction of 3-hydroxytetrahydrofuran

where the hydroxy group is reacted in excellent yields over the hydrofuran group.4

‡ Films made at room temperature (entry 4) were cured for 2 hours.

1. Piers, W. E.; Marwitz, A. J. V.; Mercier, L. G. Inorg. Chem. 2011, 50, 12252.

2. Brook, M. A.; Grande, J. B.; Ganachaud, F. Adv. Polym. Sci. 2011, 235, 161.

3. Parks, D. J.; Piers, W. E. J. Am. Chem. Soc. 1996, 118, 9440.

4. Blackwell, J. M.; Foster, K. L.; Beck, V. H.; Piers, W. E. J. Org. Chem. 1999, 64, 4887.

5. Gevorgyan, V.; Liu, J.-X.; Rubin, M.; Benson, S.; Yamamoto, Y. Tett. Lett. 1999, 40, 8919.

6. Grande, J. B.; Thompson, D. B.; Gonzaga, F.; Brook, M. A. Chem. Comm. 2010, 27, 4988.

7. Rubinsztajn, S.; Cella, J. A. Macromolecules 2005, 38, 1061.

8. Cella, J.; Rubinsztajn, S. Macromolecules 2008, 41, 6965.

9. Thompson, D. B.; Brook, M. A. J. Am. Chem. Soc. 2007, 130, 32.

10. Grande, J. B.; Gonzaga, F.; Brook, M. A. Dalton Trans. 2010, 39, 9369.

11. Gretton, M. J.; Kamino, B. A.; Bender, T. P. Macromolecules 2011, 45, 723-728.

12. Kamino, B. A.; Grande, J. B.; Brook, M. A.; Bender, T. P. Org. Lett. 2010, 13, 154.

13. Kamino, B. A.; Mills, B.; Reali, C.; Gretton, M. J.; Brook, M. A.; Bender, T. P. J. Org.

Chem. 2012, 77, 1663.

14. Kamino, B. A.; Bender, T. P.; Klenkler, R. A. J. Phys. Chem. Lett. 2012, 3, 1002.

15. Helgesen, M.; Sondergaard, R.; Krebs, F. C. J. Mater. Chem. 2010, 20, 36.

118

118

16. Ma, B.; Lauterwasser, F.; Deng, L.; Zonte, C. S.; Kim, B. J.; Fréchet, J. M. J.; Borek, C.;

Thompson, M. E. Chem. Mater. 2007, 19, 4827.

17. Hsieh, C.-H.; Cheng, Y.-J.; Li, P.-J.; Chen, C.-H.; Dubosc, M.; Liang, R.-M.; Hsu, C.-S. J.

Am. Chem. Soc. 2010, 132, 4887.

18. Nuyken, O.; Bacher, E.; Braig, T.; Fáber, R.; Mielke, F.; Rojahn, M.; Wiederhirn, V.;

Meerholz, K.; Müller, D. Designed Monomers and Polymers 2002, 5, 195.

19. Kim, C.; Wang, Z.; Choi, H.-J.; Ha, Y.-G.; Facchetti, A.; Marks, T. J. J. Am. Chem. Soc.

2008, 130, 6867.

20. Li, W.; Wang, Q.; Cui, J.; Chou, H.; Shaheen, S. E.; Jabbour, G. E.; Anderson, J.; Lee, P.;

Kippelen, B.; Peyghambarian, N.; Armstrong, N. R.; Marks, T. J. Adv. Mater. 1999, 11, 730.

21. Li, J.; Marks, T. J. Chem. Mater. 2008, 20, 4873.

22. Morais, T. D. d.; Chaput, F.; Lahlil, K.; Boilot, J.-P. Adv. Mater. 1999, 11, 107.

23. Brook, M. A. Silicon in Organic, Organmetallic, and Polymer Chemistry; John Wiley &

Sons: New York, 2000.

24. Schaer, M.; Nüesch, F.; Berner, D.; Leo, W.; Zuppiroli, L. Adv. Func. Mater. 2001, 11, 116.

25. Kawano, K.; Pacios, R.; Poplavskyy, D.; Nelson, J.; Bradley, D. D. C.; Durrant, J. R. Solar

Energy Materials and Solar Cells 2006, 90, 3520.

26. Grande, J. B.; Fawcett, A. S.; McLaughlin, A. J.; Gonzaga, F.; Bender, T. P.; Brook, M. A.

Polymer 2012, 53, 3135.

27. Thelakkat, M. Macro. Mat. Engineering 2002, 287, 442.

28. Hartwig, J. F.; Kawatsura, M.; Hauck, S. I.; Shaughnessy, K. H.; Alcazar-Roman, L. M. J.

Org. Chem. 1999, 64, 5575.

29. Teetsov, J.; Anne Fox, M. J. Mater. Chem. 1999, 9, 2117.

119

119

30. Chen, W.-H.; Wang, K.-L.; Hung, W.-Y.; Jiang, J.-C.; Liaw, D.-J.; Lee, K.-R.; Lai, J.-Y.;

Chen, C.-L. J. Polym. Sci. A 2010, 48, 4654.

120

120

Personal Interest Projects 1: Design of Deep-Blue Emitting Materials for OLEDs


This chapter has been previously published as a full paper in Organic Electronics.

Brett A. Kamino, Yi-Lu Chang, Zheng-Hong Lu and Timothy P. Bender (2012), Organic

Electronics, 13 (8), 1479-1485. Figure and Schemes are reprinted with permission. Copyright 2012 Elsevier

The work described in this paper does not relate to my primary thesis statement and the projects

previously discussed. However, this work did fit within the overall goals of our laboratory and

that of a close collaborator. Specifically, this goal was the design of new materials for organic

light emitting diodes with a particular focus on white lighting and display technology. When this

project began, I was very interested in learning more about this area of research that was being

actively pursued nearby. Because of this motivation, when the opportunity to study materials for

this project arose, I quickly took it up.

The genesis of this project relates very closely with the materials covered in Chapter 10. As a

prelude to the next chapter, we were very interested in designing new light absorbing materials

for organic solar cells that were related to boron subphthalocyanines. This material was

mentioned earlier in Chapter 4. While working on this project, I noticed that one of the precursor

materials had excellent photoluminescent properties with strong emission in the deep-blue part of

the spectrum. From this observation and several discussions with my co-author (Yi-Lu Chang)

we began this project.


The authorship of this paper is as follows: Brett A. Kamino, Yi-Lu Chang, Zheng-Hong Lu,

Timothy P. Bender. This project was instigated by me and all compound synthesis and

characterization was performed by me. Yi-Lu Chang integrated my materials into a standard

OLED design and performed all device testing. Zheng-Hong Lu is Yi-Lu Chang’s supervisor and

all device work was performed in his laboratory. The paper was written primarily by me with

input from Yi-Lu Chang and Prof. Timothy P. Bender.

121

121

9.6 Paper

9.6.1 Abstract

Unprecedented phthalonitrile based fluorophores are reported and studied for incorporation into

organic light emitting diodes. The phthalonitriles were obtained using a very simple synthetic

procedure and were found to be highly fluorescent with quantum yields approaching unity; they

are also thermally stable and electrochemically active. When incorporated into OLEDs as

fluorescent dopants, the resulting devices have good device efficiencies and emit in the deep-

blue area of the spectrum with CIE coordinates that are close to the NTSC standard for blue.

9.6.2 Introduction

Organic light emitting diodes (OLEDs) are a promising technology for full-colour, large area

display technology.1 Such displays typically rely on the use of the red, green, blue (RGB) colour

space which requires separate emission from red, green, and blue materials/OLEDs. Currently,

highly efficient phosphorescent OLEDs emitting red and green light are well-known and can

achieve internal quantum efficiencies approaching unity. However, the development of high-

efficiency blue OLEDs with Commission Internationale de l’éclairage (CIE)2 coordinates

matching the NTSC colour standard for blue (CIEx = 0.14, CIEy = 0.08)3 has proven more

difficult and remains a current challenge for display technology. Blue phosphorescent devices

have been studied extensively and a number of nonsaturated blue or sky-blue devices are well

known.4,5 While high efficiencies have been achieved these technologies still have many

disadvantages for display technologies including poor colour saturation and extremely short

operational lifetimes.6 Improvements in colour saturation for these devices has recently been

developed through the engineering of better host materials7,8 and phosphorescent emitting

materials.9 These devices have impressive performance but phosphorescent devices are still

largely limited to CIEy coordinates of <0.15 and issues with the stability of high energy

phosphorescent emitters remain. Therefore, there is still room for further improvement in colour

saturation and stability by engineering materials that can more closely match the standard for

blue colour in display technology.

One alternative approach to this problem has been to combine blue fluorescent dopant with red

and green phosphorescent dopants to achieve full colour displays.10,11 While fluorescent blue

122

122

emitting materials are inherently less efficient due to only being able to harvest singlet

excitons/electron–hole pairs, they are better able to achieve deep-blue emission while

maintaining high photoluminescent efficiencies and chemical stability. Because of this, there is a

strong interest in developing high-efficiency, deep-blue fluorescent OLEDs that can approach the

NTSC colour standard for blue. In regards to material design however, there are relatively few

classes of materials that satisfy these requirements. The majority of these are based on a limited

number of structural themes, including: styrylbiphenyls,12-15 acenes,16,17 group XIII metal

complexes18 and phenylquinolines.19 Given this limited pool of deep-blue emitting groups, there

exists a need to identify additional molecular structures which emit deep-blue light. In this paper,

we disclose an unprecedented pair of highly-emissive phthalonitrile based blue emitters and

establish their utility in deep-blue OLED devices that closely match the NTSC standard for blue.


The inspiration for this study came from the description of 9-1 (Scheme 9-1) in a patent

concerning new phthalonitriles for the synthesis of phthalocyanines20 wherein no mention of the

fluorescent or other optical properties was included. Upon repeating the described synthesis, we

discovered that compound 9-1 was highly fluorescent even when photo-excited with a simple

laboratory UV lamp. We also decided to synthesize a structural variant in the form of compound

9-2 as the reagent was commercially available (Scheme 9-1). Both phthalonitriles contain an

extended π-electron system through two [1,4] benzodioxin linkages to yield highly conjugated

and planar compounds (as shown by DFT calculations, Figure 9-1).

Scheme 9-1. Synthetic pathways towards phthalonitriles 9-1 and 9-2.

123

123

HOMO

LUMO

LUMO

HOMO

Figure 9-1. Geometry optimized structure for (a) compound 9-1 (top) and (b) compound 9-2 (bottom) and their predicted HOMO and LUMO distributions.

The syntheses rely on nucleophilic aromatic substitution of tetrachlorophthalonitrile with the

appropriate aromatic 1,2-diol in dimethylformamide along with stoichiometric quantities of

potassium carbonate. Both syntheses proceeded rapidly and cleanly from commercially available

starting materials to produce the desired products in high yields. Purification was rapidly

achieved by simply filtering the pure product and washing with solvent. The resulting materials

were isolated as insoluble white powders and we did not detect the formation of regioisomers.

124

124

Despite the facile workup procedure, exceptionally high purities were achieved for both

compounds without the need for subsequent purification steps. Structural determination and

purity was achieved through elemental analysis and high-resolution mass spectrometry.

Additionally, 1H NMR of the compounds is reported. Unfortunately, the poor solubility of the

compounds inhibited us from acquiring 13C NMR data.

Table 9-1. Photophysical, electrochemical and thermal properties of compounds 1 and 2

Compound λmax

sol emission

(nm) a

λmax, film

emission (nm) Φsol (%) a,b

Eoxpeak

(onset)

(V)c

Eredpeak (V)c

EHOMOd

[calculated]

(eV)

ELUMOf

e[calculated]e

(eV)

Td (°C)g

9-1 410 421 ~1 1.76 (1.56) -1.62 -6.14

[-6.07] -2.93 [-2.14] 277

9-2 405 434 0.9 1.66 (1.53) - -6.11

[-6.02] -2.86 [-2.15] 396

a Measured in tetrahydrofuran; λexcitation = 340nm. b Measured relative to 9,10-diphenylanthracene in cyclohexane21 c Peak and onset potential versus Ag/AgCl. d Estimated from onset potential.22 e As determined from DFT calculations f Estimated from HOMO and optical bandgap. g Defined as 5% mass loss from TGA

125

125

Figure 9-2. Normalized absorption (black line) and emission spectra (solid color for solution,

dashed for thin film) of 9-1 (a) and 9-2 (b). Solution spectra were collected in THF.

The photophysical properties of each compound were explored through solution UV–VIS

absorbance and photoluminescence measurements in tetrahydrofuran (Figure 9-2). Additionally,

solid-state photoluminescence measurements were taken from vacuum deposited films. All of

this data is summarized in Table 9-1. From this data, we observe that each compound emits in

126

126

the deep-blue area of the spectrum in both solution and in the solid-state. Interestingly, both the

onset of absorbance and the solution emission peak for 1 are slightly red shifted compared to 9-2

(~5 nm) in solution. This is somewhat surprising considering the greater conjugation length of 9-

2 versus 1 which would normally facilitate the opposite trend as seen in a series of acenes (e.g.

benzene, naphthalene, anthracene). As well, ground-state conjugation throughout each structure

is confirmed by the DFT modelling results (Figure 9-1) further supporting this prediction. We

believe that this deviation from the expected behaviour may be due to the nature of conjugation

throughout each molecule. The presence of planar oxygen atoms in these structures leads to

hyperconjugation between the peripheral phenyl units and the central phthalonitrile core through

the oxygen atoms. This less than perfect delocalization across the structure may prevent the

expected lowering of the optical band gap between 9-1 and 9-2. That being said, we do not have

a more thorough understanding of this phenomenon at this time. Solid-state photoluminescence

shows moderate red shifting of the emission peak for each compound compared to solution

photoluminescence with the effect being greater for 9-2 than 9-1. This larger change in the

photoluminescent peak wavelength could be attributable to an increase in intermolecular

interactions in the solid state afforded by the greater degree of conjugation.

Relative fluorescent quantum yield measurements were performed in THF and referenced to

9,10-diphenylanthracene in a cyclohexane solution (Φyield ~ 0.9).21 The difference in solvents

was necessitated by the low solubility of 9-1 and 9-2 in cyclohexane. 9,10-Diphenylanthracene

was chosen as a reference as it has similar absorbance and emission wavelengths as our

compounds. Solutions were degassed prior to each experiment and sufficiently low

concentrations were used to minimize reabsorption errors. Both compounds are highly

photoluminescent with 9-1 having a quantum yield of near unity while 9-2 having a quantum

yield of 0.9. The high photoluminescent efficiencies were surprising as phthalonitriles are not

known to be particularly efficient fluorophores. To the best of our knowledge this is the first

example of a strongly fluorescent phthalonitrile derivatives.

In order to explore the potential of these compounds to be used as emitters in OLEDs, each was

characterized by solution electrochemical techniques and thermogravimetric analysis. The

electrochemical behaviour of 9-1 and 9-2 was determined through cyclic voltammetry in

dichloromethane with 0.1 M of tetrabutylammonium perchlorate as a supporting electrolyte.

127

127

Both compounds displayed fully irreversible oxidation waves with peak potentials at 1.76 and

1.66 V vs. Ag/AgCl for 9-1 and 9-2, respectively (Table 9-1). Additionally, an irreversible

reduction peak is observed for compound 1 at -1.62 V vs. Ag/AgCl while no clear reduction

activity was observed for 2. From this data, absolute molecular orbital energies were estimated22

from the onset oxidations waves for the HOMO energy, while the LUMO energy values were

estimated from the HOMO energy added to the energy of the optical bandgap (Table 9-1).

Most OLED layers are deposited using physical vapour deposition requiring candidate

compounds to possess some degree of thermal stability in order to be processed into devices. To

evaluate the compatibility of phthalonitriles 9-1 and 9-2 with modern OLED manufacturing

techniques, the thermal stability of each was measured using thermogravimetric analysis (TGA)

under N2 with a ramp rate of 10 °C/min. For compounds 9-1 and 9-2 we observed a 5% mass

loss at 277 and 396 °C (Table 9-1), respectively. A complete loss of initial mass was observed

for compound 1 indicating no formation of ash. This implies that the compound simply sublimed

upon heating. In the case of phthalonitrile 9-2, ash was formed and thus the 5% weight loss

indicates the onset of decomposition of the compound. Regardless of the nature of the 5% weight

loss, these relatively high values confirm the stability of these structural motifs and suggest that

they would be well-suited for physical vapour deposition.

Alluded to above, geometry optimized DFT calculations were performed not only to understand

the molecular shape and conformation of 9-1 and 9-2 but also to aid the understanding of

electronic structure of this structural motif. Because compounds 9-1 and 9-2 represent a

completely new class of fluorophore, an understanding of the electronic structure is important for

further elaboration on the molecular structure. Visualizations of the calculated HOMO and

LUMO are given in Figure 9-1. In each case, the HOMO is evenly distributed throughout the

entire conjugated structure of the molecules. In contrast, the LUMO is isolated to the

phthalonitrile core of each molecule. This distribution of orbital densities suggests that tuning of

the HOMO could easily be accomplished by the addition of electron donating groups to the

periphery of the molecules whereas little variation would be possible in the LUMO level.

Finally, each molecule was incorporated as a fluorescent dopant in a series of four simple OLED

devices23 at both 2 and 4 wt.% with respect to the host material (4,4`-di(9-carbazolyl)-biphenyl,

CBP) (Figure 9-3, Table 9-2). The device structure used was: ITO/MoO3/CBP (35 nm)/emitting

128

128

layer (15 nm)/TPBi (65 nm)/LiF (1 nm)/Al (100 nm) where the emitting layer is composed of the

fluorescent dopant and CBP (TPBi = 2,2`,2``-(1,3,5-benzinetriyl)tris(1-phenyl-1-H-

benzimidazole)). All four devices achieved deep-blue emission with CIE coordinates that are

exceptionally close to standard blue (CIEx = 0.14, CIEy = 0.08) with moderate luminance values

and good turn-on voltages (Figure 9-3). In particular, devices made using 9-1 showed blue

emission that comes very close to the ideal value with CIE coordinates of CIEx = 0.16, CIEy =

0.08 (for the 2 wt.% doped device at 100 cd/m2). Looking at the electroluminescent spectra for

compound 9-1 (Figure 9-3), we can conclude that there is almost exclusively emission from the

dopant with a small shoulder at wavelengths < 400 nm. For devices using 9-2 however,

significant emission occurred below 400 nm and a broadened peak shape was observed. This

shoulder is consistent with electroluminescence from the CBP host material24 and either implies

poor charge balance within the device, or inefficient energy transfer from the host to the dopant.

Moreover, when comparing the estimated energy levels of the fluorophores to that of the host

material (CBP) using the same methods as above, CBP has a significantly higher HOMO level at

-5.61 eV.25 This large mismatch in the HOMO energies of the host and guest may impede direct

energy trapping by the guest molecule resulting in endothermic energy transfer and ultimately

limit the efficiency of the devices. The observation that the electroluminescence from the host

material appears more prominent for compound 9-2 compared to compound 1 could be

attributable to the higher Elumo of -2.86 eV for compound 9-2 as compared to that of compound

9-1 (-2.93 eV, Table 9-1). This could lead to a lower capability for the CBP host to confine

charges to compound 9-2, leading to a higher probability of energy back transfer from the dopant

to the host thereby resulting in higher host emission.

129

129

Figure 9-3a. (i) Electroluminescent spectra of compound 9-1 (solid lines are for 2 wt% doping in the

emission layer and hashed lines are for 4 wt% doping) at 100 cd/m2; (ii) Current density (solid lines) and

luminance (hashed lines) as a function of voltage for devices using 2 wt% (black lines) and 4 wt% (red

lines) of compound 9-1 as a dopant in the emission layer; (iii) External quantum efficiency versus

luminance for devices using 2 wt% (black line) and 4 wt% (red line) compound 9-1 as a dopant in the

emission layer; (iv) Current efficiency (hashes lines) and power efficiency (solid lines) as a function of

current density in devices using 2 wt% (black lines) and 4 wt% (red lines) compound 9-1 as a dopant in

the emission layer. Device architecture for all devices is illustrated (Figure 9-3a(i), inset).

130

130

Figure 9-3b. (i) Electroluminescent spectra of compound 9-2 (solid lines are for 2 wt% doping in the

emission layer and hashed lines are for 4 wt% doping) at 100 cd/m2; (ii) Current density (solid lines) and

luminance (hashed lines) as a function of voltage for devices using 2 wt% (black lines) and 4 wt% (blue

lines) of compound 9-2 as a dopant in the emission layer; (iii) External quantum efficiency versus

luminance for devices using 2 wt% (black line) and 4 wt% (blue) compound 9-2 as a dopant in the

emission layer; (iv) Current efficiency (hashes lines) and power efficiency (solid lines) as a function of

current density in devices using 2 wt% (black lines) and 4 wt% (blue lines) compound 9-2 as a dopant in

the emission layer. Device architecture for all devices is illustrated (Figure 9-3a(i), inset).

131

131

Table 9-2. Performance of OLEDs incorporating 1 and 2 as dopants

Compound Concentration

(wt%) Vturn on (V)a EQEmax (%) λmax

EL (nm) b CIExb CIEy

b

9-1 2 3.4 2.14 434 0.16 0.08

4 3.4 2.12 440 0.16 0.09

9-2 2 3.8 1.66 434 0.16 0.11

4 3.8 1.90 444 0.16 0.12

a As defined by a luminance of >0.5 cd/m2. b Determined at a luminance of 100 cd/m2

Despite the potential challenges in device optimization, these initial prototype devices performed

admirably well. A maximum external quantum efficiency of 2.14% was observed for the 2 wt.%

doped device with compound 9-1 while devices made with compound 9-2 showed slightly lower

values with a maximum of 1.90%. Both values are extremely promising and we anticipate that

further engineering of the devices and the compounds themselves can yield even higher

performance. Lifetime measurements were prevented by the low morphological stability of the

host material, CBP with low glass transition temperature of 62 °C.26 Future development of more

stable hosts and transport layers are in progress to study device life times.

9.6.4 Conclusions

In summary, we have synthesized and characterized two novel phthalonitrile based fluorophores

for use in deep blue OLEDs that approach the NTSC standard for blue. The compounds possess

many ideal qualities for use in OLED devices including good emission coordinates/colour and

thermal stability. When incorporated into prototype OLEDs, these compounds undergo

electroluminescence with high external efficiencies and a colour near the NTSC standard for

blue. The high efficiency and good colour rendering of these compounds suggests that they may

be a promising material for application in a full colour OLED display.

9.7 References

1. Forrest, S.R., Nature 2004, 428, 911.

2. CIE Commission Internationale de l’Eclairage Proceedings,Cambridge University Press:

Cambridge, 1932.

132

132

3. D.G. Fink, Color Television Standards: Selected Papers and Records, McGraw-Hill, New

York, 1955.

4. Sasabe, H.; Gonmori, E.; Chiba, T.; Li, Y.-J.; Tanaka, D.; Su, S.-J.; Takeda, T.; Pu, Y.-J.;

Nakayama, K.-I.; Kido, J. Chem. Mater. 2008, 20, 5951.

5. Chopra, N.; Lee, J.; Zheng, Y.; Eom, S,-H.; Xue, J.; So, F. Appl. Phys. Lett 2008, 93,

143307.

6. Moraes, I.R.D.; Scholz, S.; Lussem, B.; Leo, K. Org. Electron. 2011, 12, 341.

7. Jeon, S.O.; Yook, K.S.; Joo, C.W.; Lee, J.Y. Adv. Mater. 2010, 22, 1872.

8. Jeon, S.O.; Jang, S.E.; Son, H.S.; Lee, J.Y. Adv. Mater. 2011, 23, 1411-1436.

9. Sasabe, H.; Takamatsu, J.; Motoyama, T.; Watanabe, S.; Wagenblast, G.; Langer, N.;

Molt, O.; Fuchs, E.; Lennartz, C.; Kido, J. Adv. Mater. 2010, 22, 5003.

10. Sun, Y.; Giebink, N.C.; Kanno, H.; Ma, B.; Thompson, M.E.; Forrest, S.R. Nature 2006,

440, 908.

11. Rosenow, T.C.; Furno, M.; Reineke, S.; Olthof, S.; Lussem, B.R.; Leo, K. J. Appl. Phys.

2010, 108, 113113.

12. Lee, M.T.; Liao, C.H.; Tsai, C.H.; Chen, C.H. Adv. Mater. 2005, 17, 2493.

13. Ho, M.-H.; Wu, Y.-S.; Wen, S.-W.; Chen, T.-M.; Chen, C.H. Appl. Phys. Lett. 2007, 91,

083515.

14. Kim, S.-K.; Yang, B.; Ma, Y.; Lee, J-H.; Park, J.-W. J. Mater. Chem. 2008, 18, 3376.

15. Kim, S.O.; Lee, K.H.; Kim, G.Y.; Seo, J.H.; Kim, Y.K.; Yoon, S.S. Synth. Met. 2010, 160,

1259.

16. Ho, M.-H.; Wu, Y.-S; Wen, S.-W.; Lee, M.-T.; Chen, T.-M.; Chen, C.H.; Kwok, K.-C.;

So, S.-K.; Yeung, K.-T.; Cheng, Y.-K.; Gao, Z.-Q. Appl. Phys. Lett. 2006, 89, 252903.

133

133

17. Tao, S.; Zhou, Y.; Lee, C.-S.; Zhang, X.; Lee, S.-T. Chem. Mater. 2010, 22, 2138.

18. Liao, S.-H.; Shiu, J.-R.; Liu, S.-W.; Yeh, S.J.; Chen, Y.-H.; Chen, C.-T.; Chow, T.J.; Wu,

C.-I. J. Am. Chem. Soc. 2009, 131, 763.

19. Lee, S.J.; Park, J.S.; Yoon, K.-J.; Kim, Y.-I.; Jin, S.-H.; Kang, S.K.; Gal, Y.-S.; Kang, S.;

Lee, J.Y.; Kang, J.-W.; Lee, S.-H.; Park, H.D.; Kim, J.-J. Adv. Func. Mater. 2008, 18,

3922.

20. Gregory, P.; Reynolds, S.J. US Patent 6,335,442, 2002.

21. Eaton, D.F.; Pure. Appl. Chem. 1988, 60, 1107.

22. Cardona, C.M.; Li, W.; Kaifer, A.E.; Stockdale, D.; Bazan, G.C. Adv. Mater. 2011, 23,

2367.

23. Wang, Z.B.; Helander, M.G.; Qiu, J.; Puzzo, D.P.; Greiner, M.T.; Liu, Z.W.; Lu, Z.H.

Appl. Phys. Lett. 2011, 98, 073310.

24. Zou, L.; Savvate’ev, V.; Booher, J.; Kim, C.H.; Shinar, J. Appl. Phys. Lett. 2001, 79,

2282.

25. Hu, D.; Lu, P.; Wang, C.; Lie, H.; Wang, H.; Wang, Z.; Fei, T.; Gu, X.; Ma, Y. J. Mater.

Chem. 2009, 19, 6143.

26. Xiao, L.; Chen, Z.; Qu, B.; Luo, J.; Kon, S.; Gong, Q.; Kido, J. Adv. Mater. 2011, 23, 926.

134

134

Chapter 10: Personal Interest Projects 2: Colour Tuning Boron Subphthalocyanine


This chapter represents a draft of a paper that is intended to be submitted to Chemical

Communications.

As with Chapter 9, the work described heirin does not relate to my primary thesis statement.

However, the contents of this project were of great interest to me and are intimately related to

other objectives in our laboratory. In this paper, I’ve synthesized and characterized a number of

new boron subphthalocyanine (BsubPc) derivatives with drastically altered photophysical and

electrochemical properties compared to base materials. These compounds are based on the π-

extended phthlaonitrile derivatives disclosed in Chapter 9. This work represents one of the few

published methods to tune BsubPc derivatives over a wide range and complements other

materials developed within the laboratory. Materials discussed in this chapter will hopefully go

on to be incorporated into organic photovoltaic devices in the future.


The author list of this paper draft is as follows: Brett A. Kamino, Timothy P. Bender

The concept of this work and the design of experiments were done solely by me. All synthesis,

purification and characterization of these materials was done solely by me. The paper was

written primarily by me with input from Prof. Tim Bender.

10.3 Paper

10.3.1 Body

First reported in 1972 by Meller and Ossko,1 boron subphthalocyanine2 (BsubPc) is a ring

contracted cousin of the well-known family of phthalocyanine dyes and pigments. BsubPcs are

characterized by a striking magenta color, strong orange fluorescence, and a unique bowl-like

molecular shape. Beyond their basic molecular properties, they have recently received a great

deal of attention for their application as functional organic materials in non-linear optics,3 and a

variety of organic electronic devices including organic field-effect transistors (OFETs),4 organic

135

135

light emitting diodes (OLEDs)5 and organic photovoltaics (OPVs). As an electron donor layer in

OPVs, BsubPcs have been shown to pair well with C60 as an electron acceptor layer, producing

devices with efficiencies up to 5.4%.6 As well, BsubPcs7 and their halogenated derivatives8 have

also been shown to work effectively as electron acceptor layers in OPVs. The unique properties

of BsubPcs, their flexible roles within a number of organic electronic devices of BsubPcs and

their auspicious performance warrants further investigation into their development as functional

materials.9

Engineering of BsubPcs to further improve device performance requires the development of new

synthetic tools and strategies to allow for the tailoring of molecular structure and the resulting

optoelectronic properties. Specifically, altering their electrochemical properties and their

absorption and interaction with light of varying wavelenghts is desirable. Several strategies to

achieve variations in the optoelectronic properties of BsubPcs have been reported. These

methods include simple peripheral substitution of the isoindoline units which comprise the

BsubPc itself,10 and the dimerization of the BsubPc either with itself11 or with other

chromophores.12 However, the most dramatic method to alter the basic properties of BsubPcs has

been in modification of the core macrocyclic subPc fragment through the use of modified

phthalonitrile-type precursors. This can be accomplished either by incorporating modified

phthalonitriles symmetrically around the ring13 or through the use of mixtures of different

phthalonitriles during synthesis. The mixed phthalonitrile approach results in the formation of a

statistical mixture of products from which individual compounds can often be difficult to isolate

in anything other than a low yield.14 Despite this, in each case the use of modified or mixed

phthalonitriles has been shown to greatly alter the optoelectronic properties of the resulting

BsubPcs. Widespread exploration of this strategy has been limited due to a lack of phthalonitrile

precursor materials which are easily prepared.

The synthesis of phthalonitriles 10-1 and 10-2 have been previously described by Gregory in the

patent literature, along with the preparation of zinc phthalocyanines from each .15 Phthalonitriles

10-1 and 10-2 are made from the reaction of tetrachlorophthalonitrile with catechol or 3,5-di-t-

butylcatechol (respectively, Scheme 10-1) in DMF at moderate temperatures. The reaction is

surprisingly selective and the workup is facile. We have previously shown that phthalonitrile 10-

1 is itself an efficient deep-blue light emitter when incorporated into OLEDs.16 What is unique

136

136

about phthalonitrile 10-1 is that it contains a π extended system despite the presence of the four

sp3 oxygen atoms. Using DFT modeling we have shown that the HOMO of phthalonitrile 10-1 is

equally distributed throughout the molecule while having density resident on the oxygen atoms.

The extended conjugation is further evident in the significantly red-shifted λmax of phthalonitrile

10-1 from that of phthalonitrile itself.16 Moreoever, the zinc phthalocyanines reported by

Gregory have a λmax 80-95 nm red shifted compared to standard zinc phthalocyanine.15

Scheme 10-1: Synthesis of π extended BsubPcs 10-3a-b and 10-4a-c and their precursor

phthalonitriles.

10-1 10-2

10-3a

10-3b

10-4a: X = H 10-4b: X = F 10-4c: X = Cl

137

137

In our laboratory, symmetrically substituted BsubPcs 10-3a and 10-3b were obtained by a

standard cyclotrimerization reaction of phthalonitrile 10-1 with BCl3 followed by axial

substitution by 4-t-butylphenol (10-3a) or pentafluorophenol (10-3b, Scheme 10-1).17 Attempts

to use BBr3 in the cyclotrimerization reaction were unsuccessful, likely due competitive ether

cleavage by BBr3 of a molecular system enabled for SNAr reactions. The resulting BsubPcs 10-3a

and 10-3b were isolated as dark green powders in comparable yields to that of typical BsubPcs.

Their respective identity and purity was established using 1H, 11B, 19F NMR, low molecular

weight GPC, and HRMS. Single crystals of 10-3b solvated with THF were grown and the

structure solved by x-ray diffraction. Analysis confirms not only the molecular structure of

BsubPc 10-3b but that it retains the bowl shaped configuration typical of a normal BsubPc with

minimal distortion from the π-extension (Figure 10-1a).

Owing to the successful synthesis of the symmetrically substituted BsubPcs (10-3a-b), the

reaction of 10-1 mixed with unmodified phthalonitrile under the same reaction conditions was

performed. As expected, analysis of the reaction mixture by low molecular weight SEC and

HPLC revealed a statistical mixture of compounds including regular Cl-BsubPc, the axially

chlorinated intermediate to 10-3a and 10-3b and a number of compounds with various

absorbance profiles indicative of BsubPcs and with varying molecular sizes. Unfortunately, this

mixture proved recalcitrant to separation using a number of different purification strategies.

Chromatographic separation techniques consistently resulted in co-elution of a number of the

components including unreacted phthalonitrile 10-1, a common issue with the separation of these

types of materials.18

(a)

138

138

(b)

Figure 10-1: Thermal ellipsoid plot of (a) 10-3b·(THF)2 (CCDC deposition 910746) and (b)

10-4a·(CHCl3)2 (CCDC deposition 910747). Thermal ellipsoids are set at the 50% probability

level. Hydrogen atoms and included solvent have been omitted for clarity. Colors: boron – pink;

nitrogen – blue; carbon – grey; oxygen – red; fluorine – magenta.

In an effort to circumvent the isolation issue and obtain unsymmetrically substituted BsubPcs,

phthalonitrile 10-2 containing four t-butyl groups was synthesized. It was hypothesized that the

substantially increased steric bulk of the phthalonitrile would prevent the formation of BsubPcs

entirely derived from phthalonitrile 10-2 (homo-cyclotrimerization). Thus, simplifying the

mixture of products produced when used in a mixture with phthalonitrile. As well, we felt the

presence of t-butyl groups would assist in separation of the final products and prevent co-elution

of products upon chromatographic separation.

Simple molecular mechanics calculations were performed on the hypothetical BsubPc made

entirely from phthalonitrile 10-2.This structure was found to have significant distortions from

planarity within the BsubPc ring system caused by steric crowding of neighboring dioxy-t-butyl-

phenylene fragments. Therefore the overall molecule would have an unreasonable amount of

steric strain and likely not be stable.

Phthalonitrile 10-2 was isolated as a mixture of three structural isomers. Using high-field 1H

NMR spectroscopy (700 MHz), the three isomers of 10-2 were positively identified and their

139

139

relative abundance was established (see section 12.8.3). Phthalonitrile 10-2 was placed alone

under standard reaction conditions with BCl3 and indeed no BsubPc products were detected by

HPLC (Scheme 10-1).

When cyclotrimerizations where performed with a mixture of phthalonitrile 10-2 (component A)

and either phthalonitrile, tetrafluorophthalonitrile, or tetrachlorophthalonitrile (component B), a

mixture of regular Cl-BsubPc, one major and one minor product were observed using low

molecular weight GPC analysis with a photodiode array detector. Each product were found to

have a red shifted absorption spectrum with a profile similar to normal BsubPc. The crude

mixture from each reaction was reacted with pentafluorophenol under standard conditions

(Scheme 10-1) and a single product (AB2) was successfully isolated from the reaction mixture by

chromatography over silica gel. The purity of this product was determined by the presence of a

single spot on TLC and a single peak using low molecular weight GPC analysis. Because

phthalonitrile 2 is a mixture of three isomers, the new BsubPc products are also expected to exist

as a mixture of structural isomers. Indeed this was found to be the case for 10-4a and 10-4c

(determined by 1H NMR spectroscopy). Surprisingly, product 10-4b was isolated as a single

structural isomer. It is not known why this particular isomer was able to be isolated. A small

amount of a single isomer of 10-4a was isolated by vapour diffusion crystallization. The crystals

proved suitable for x-ray diffraction and on analysis confirmed the structure of 10-4a (as its

chloroform solvate, Figure 10-1b).

140

140

(a)

(b)

Figure 10-2: Normalized UV-Vis absorbance spectra of (a) F5BsubPc, 10-4a and 10-3b and (b)

10-4a-c.

10-4a

10-3b

141

141

Presumably due to the large difference in polarity afforded by the t-butyl groups, the desired

products 10-4a-c were easily separated from regular pentafluorophenoxy-BsubPc (F5BsubPc)

made up of only phthalonitrile (B3). The second minor product with red shifted absorption was

found to have a larger molecular size (by GPC analysis) and a small red-shift in its absorption

spectrum relative to 10-4a-c. We suspected that this was the product made up of two π-extended

phthalonitriles 10-2 and a single phthalonitrile (A2B). However, we could not isolate this minor

product by chromatography over silica gel. It appeared to degrade on-column and despite

multiple attempts we could not observe the compound eluting. For the A2B system, simple

molecular mechanics calculations show steric repulsion between the adjacent t-butyl groups

resulting in a distortion of the dioxy-t-butyl-phenylene fragment from planarity. This effect is

similar to that seen in the A3 system, albeit not to the same level. This would seem to support the

experimental observations of poor chemical stability of the minor A2B product.

Each new BsubPc was characterized by UV-Vis absorbance spectroscopy and cyclic

voltammetry. Solution absorbance spectra (Figure 10-2, Table 10-1) show that incorporation of

phthalonitriles 10-1 and 10-2 are effective in red-shifting the absorbance of the base

chromophore relative to a reference BsubPc compound: F5BsubPc (Figure 10-2).5d BsubPcs 10-

3a and 10-3b possess an intense Q-band with a similar shape to F5BsubPc but red-shifted

approximately 100 nm to a λmax of 662 nm (Figure 10-2a) approximately equivalent to that of

boron subnaphthalocyanine (λmax = 663 nm).13b Large extinction coefficents were also measured

(ε = 9.5 x 105 L mol-1 cm-1 for 10-3a and 8.3 x 105 L mol-1 cm-1 for 10-3b) which are in the range

of typical BsubPcs and one order of magnitude higher than that reported for boron

subnaphthalocyanine (ε = 7.94 × 104 L mol-1 cm-1).13b Compound 10-4a, containing only a single

π-extended ligand, shows a smaller red shift in the Q-band of 48 nm. The other non-symmetric

BsubPcs (10-4b and 10-4c) also display broadened and red-shifted Q-bands which are found at a

λmax of 625 nm and 631 nm, respectively. Extinction coefficients for BsubPcs 10-4a-c were

smaller in magnitude than 10-3a-b, and were found to be between 4.4 x 105 L mol-1 cm-1 and 6.1

x 105 L mol-1 cm-1. Photoluminescent spectra for each compound were also obtained, but in all

cases only very weak fluorescence signals were detected.

142

142

Table 10-1: Calculated and experimentally determined properties of compounds F5BsubPc, 10-

3a-b and 10-4a-c.

Compound λmax

(nm)b ε x 105 (L

mol-1 cm-1)b λPL

(nm)b E1/2

ox

(mV)c,d E1/2

red 1 (mV ) c,d

[E1/2red 2 (mV)]

EHOMOcalc

(eV) ELUMO

calc (eV)

F5BsubPc 561 9.0 592a N/A -878a -5.31a -2.59a

10-3a 652 9.5 664 839 -935 -4.62 -2.34

10-3b 659 8.3 671 877 -886 -4.77 -2.47

10-4a 609 6.1 626 989 -990 -4.95 -2.50

10-4b 625 4.4 672 1163 -655 [-1263] -5.29 -3.04

10-4c 631 5.3 680 1197 -560 [-1255] -5.41 -3.17

a - data taken from Morse et al.5d, b data collected in THF solution, c data collected in DCM solution, d potentials referenced to

Ag/AgCl

The electrochemistry of 10-3a-b and 10-4a-c was explored by solution cyclic voltammetry in

DCM with 0.1M tetrabutylammonium perchlorate as the electrolyte and decamethylferrocene as

an internal reference19 (see section 12.10 for full voltammograms). Under these conditions, each

compound was found to undergo reversible oxidation and reduction reactions. The oxidative

behaviour of 10-3a-b and 10-4a-c was fairly typical for BsubPc derivatives. However, the

partially fluorinated and chlorinated (10-4b and 10-4c) BsubPcs each displayed two reversible

reductions (Figure 10-3). Typical BsubPcs rarely show reversible reductions5c and 10-4b-c are

the first known examples of a BsubPc derivative to undergo two reversible reductions. We may

therefore conclude that the BsubPcs reported herein have superior electrochemical stability when

compared to most other BsubPc derivatives. Additionally, throughout this series of BsubPcs, a

significant range of oxidation and reduction potentials can be achieved depending on the number

and type of phthalonitriles used in their syntheses.

143

143

Figure 10-3: Cyclic voltammogram of BsubPc compound 10-4b in DCM with 0.1M

tetrabutylammonium perchlorate and decamethylferrocene as an internal standard.

The increased electrochemical stability and the trends seen in the UV-Vis absorbance spectra

were further investigated using geometry optimized DFT calculations (see section 12.9). For the

symmetric BsubPcs (10-3a and 10-3b), both the HOMO and LUMO are spread out over the

entire bowl structure. From this, we can presume that the increase in electrochemical stability is

due to increased delocalization of charge throughout the extended π conjugated structure

effectively stabilizing the radical cation/anion. For the non-symmetric products (10-4a-c), the

HOMO is distributed evenly throughout the bowl structure while the LUMO is mostly localized

over the isoindoline units originating from the normal phthalonitrile and the boron centre. This

partial charge separation between the two halves of the bowl is reminiscent of donor-acceptor

chromophores where the optoelectronic properties can be tuned by altering the strength of the

electron donors/acceptors.20

This argument is supported by the change in position of the Q-band on increasing of the electron

withdrawing potential of the isoindoline units (compounds 10-4a-c, Figure 10-2b). As we go

from hydrogen to strong electron withdrawing substituents (-H < -F < -Cl), we see a narrowing

in the band gap as indicated by a red-shift in the absorption spectrum of the material. This is a

144

144

characteristic sign of an electron push-pull or donor-acceptor electronic system. To the best of

our knowledge, this is first example of a push-pull system within the conjugated bowl structure

of a BsubPc derivative. The intra-bowl push-pull system also greatly alters the electrochemical

oxidation/reduction potentials. Comparing F5BsubPc, 10-4a, and 10-3b which have 0, 1, or 3

electron rich isoindoline units, we observe an increase in reduction potential with increasing

electron density (Table 10-1). Similarly, comparing compounds 10-4a-c, we observe a lowering

of reduction potential and an increase in oxidation potential with stronger electron withdrawing

fragments. Thus, we can state that altering the electron donating or electron withdrawing

potentials of isoindoline units with the bowl of a BsubPc can be an effective tool in tuning not

only the band gap of the material but also its electrochemical potentials.

In summary, we have synthesized and characterized a series of π-extended BsubPc derivatives

using two electron rich π extended phthalonitrile precursors (phthalonitriles 10-1 and 10-2). The

resulting BsubPcs all display exceptional electrochemical stability as well as intense, red-shifted

absorbance spectra relative to normal BsubPcs. We have also synthesized a series of BsubPcs

from mixtures of phthalonitrile 10-2 with normal phthalonitriles and demonstrated that these

same properties can be readily tuned by varying the stoichiometry of the π-extended

phthalonitrile 10-2 and normal phthalonitriles. By setting up an intra-bowl push-pull system

within an unsymmetric BsubPcs, the optoelectronic properties can effectively be fine-tuned

across a range of absorption wavelengths and oxidation and/or reduction potentials.

10.3.2 References

1. Meller, A.; Ossko, A., Monatsh. Chem. 1972, 103 (1), 150-155.

2. Claessens, C. G.; Gonzalez-Rodriguez, D.; Torres, T., Chem. Rev. 2002, 102, 835-853.

3. (a) Dini, D.; Vagin, S.; Hanack, M.; Amendola, V.; Meneghetti, M., Chem. Comm. 2005,

(30), 3796-3798; (b) Claessens, C. G.; González-Rodríguez, D.; Torres, T.; Martín, G.;

Agulló-López, F.; Ledoux, I.; Zyss, J.; Ferro, V. R.; García de la Vega, J. M., J. Phys. Chem.

B 2005, 109 (9), 3800-3806.

4. Yasuda, T.; Tsutsui, T., Mol. Cryst. Liq. Cryst. 2006, 462 (1), 3-9.

145

145

5. (a) Díaz, D. D.; Bolink, H. J.; Cappelli, L.; Claessens, C. G.; Coronado, E.; Torres, T.,

Tetrahedron Lett. 2007, 48 (27), 4657-4660; (b) Helander, M. G.; Morse, G. E.; Qiu, J.;

Castrucci, J. S.; Bender, T. P.; Lu, Z.-H., ACS Appl. Mater. Interfaces 2010, 2 (11), 3147-

3152; (c) Morse, G. E.; Castrucci, J. S.; Helander, M. G.; Lu, Z.-H.; Bender, T. P., ACS Appl.

Mater. Interfaces 2011, 3 (9), 3538-3544; (d) Morse, G. E.; Helander, M. G.; Maka, J. F.; Lu,

Z.-H.; Bender, T. P., ACS Appl. Mater. Interfaces 2010, 2 (7), 1934-1944.

6. (a) Mutolo, K. L.; Mayo, E. I.; Rand, B. P.; Forrest, S. R.; Thompson, M. E., J. Am. Chem.

Soc. 2006, 128 (25), 8108-8109; (b) Pandey, R.; Holmes, R. J., Adv. Mater. 2010, 22 (46),

5301-5305; (c) Pandey, R.; Zou, Y.; Holmes, R. J., Appl. Phys. Lett. 2012, 101 (3), 033308.

7. Beaumont, N.; Cho, S. W.; Sullivan, P.; Newby, D.; Smith, K. E.; Jones, T. S., Adv. Func.

Mater. 2012, 22 (3), 561-566.

8. (a) Gommans, H.; Aernouts, T.; Verreet, B.; Heremans, P.; Medina, A. s.; Claessens, C. G.;

Torres, T., Adv. Func. Mater. 2009, 19 (21), 3435-3439; (b) Sullivan, P.; Duraud, A.; Hancox,

l.; Beaumont, N.; Mirri, G.; Tucker, J. H. R.; Hatton, R. A.; Shipman, M.; Jones, T. S., Adv.

Energy Mater. 2011, 1 (3), 352-355.

9. Morse, G. E.; Bender, T. P., ACS Appl. Mater. Interfaces 2012, 4 (10), 5055-5068.

10.(a) del Rey, B.; Keller, U.; Torres, T.; Rojo, G.; Agulló-López, F.; Nonell, S.; Martí, C.;

Brasselet, S.; Ledoux, I.; Zyss, J., J. Am. Chem. Soc. 1998, 120 (49), 12808-12817; (b)

González-Rodríguez, D.; Torres, T.; Guldi, D. M.; Rivera, J.; Herranz, M. Á.; Echegoyen, L.,

J. Am. Chem. Soc. 2004, 126 (20), 6301-6313.

11.(a) Yamasaki, Y.; Mori, T., Bull. Chem. Soc. Jpn. 2011, 84 (11), 1208-1214; (b) Eckert, A.

K.; Rodríguez-Morgade, M. S.; Torres, T., Chem. Comm. 2007, (40), 4104; (c) Claessens, C.

G.; Torres, T., Angew. Chem. Int. Ed. 2002, 41 (14), 2561-2565; (d) Fukuda, T.; Stork, J. R.;

Potucek, R. J.; Olmstead, M. M.; Noll, B. C.; Kobayashi, N.; Durfee, W. S., Angew. Chem.

Int. Ed. 2002, 41 (14), 2565-2568.

12.(a) Xu, H.; Ng, K. P., Inorg. Chem. 2008, 47, 7921-7927; (b) Zhao, Z.; Cammidge, A. N.;

Cook, M. J., Chem. Comm. 2009, (48), 7530; (c) Mauldin, C. E.; Piliego, C.; Poulsen, D.;

146

146

Unruh, D. A.; Woo, C.; Ma, B.; Mynar, J. L.; Fréchet, J. M. J., ACS Appl. Mater. Interfaces

2010, 2 (10), 2833-2838.

13.(a) Shimizu, S.; Miura, A.; Khene, S.; Nyokong, T.; Kobayashi, N., J. Am. Chem. Soc. 2011,

133 (43), 17322-17328; (b) Nonell, S.; Rubio, N.; del Rey, B.; Torres, T., J. Chem. Soc.

Perkin Trans. 2 2000, (6), 1091-1094.

14.(a) Shimizu, S.; Otaki, T.; Yamazaki, Y.; Kobayashi, N., Chem. Comm. 2012, 48 (34), 4100;

(b) Shimizu, S.; Nakano, S.; Hosoya, T.; Kobayashi, N., Chem. Comm. 2011, 47 (1), 316; (c)

Zhu, H.; Shimizu, S.; Kobayashi, N., Angew. Chem. Int. Ed. 2010, 49 (43), 8000-8003; (d)

Zyskowski, C. D.; Kennedy, V. O., J. Porphyrins Phthalocyanines 2000, 4, 707-712.

15.Gregory, P.; Reynolds, S. J. Phthalocyanines. US 6335442 B1, 2002.

16.Kamino, B. A.; Chang, Y.-L.; Lu, Z.-H.; Bender, T. P., Org. Electron. 2012, 13 (8), 1479-

1485.

17.Morse, G. E.; Paton, A. S.; Lough, A.; Bender, T. P., Dalton Trans. 2010, 39 (16), 3915.

18.González-Rodríguez, D.; Claessens, C. G.; Torres, T., J. Porphyrins Phthalocyanines 2009,

13 (02), 203-214.

19.(a) Noviandri, I.; Brown, K. N.; Fleming, D.S.; Fulvas, P.T.; Lay, P.A.; Masters, A.F.;

Philips, L. J. Phys. Chem. B 1999, 103, 6713. (b) Bender, T. P.; Graham, J. F.; Duff, J. M.,

Chem. Mater. 2001, 13 (11), 4105-4111.

20.Meier, H., Angew. Chem. Int. Ed. 2005, 44 (17), 2482-2506.

147

147

Chapter 11: Concluding Remarks and Future Work

11.1 Summary

In summary, we have explored various ways to alter the physical properties of organic

semiconductors without adversely affecting their basic electrochemical properties. These studies

have all focused on arylamine based compounds which are derivatives of well-known

compounds used throughout the literature.

This began with a study where we produced a series of triarylamine based compounds with a

wide range of structures and electrochemical properties. The photoinduced electron transfer

reaction with a fluorescent electron acceptor was studied to determine the effects of triarylamine

structure on its potential use in a organic photovoltaic cell. From this study we can conclude that

the actual structure of a triarylamine is unimportant when studying this specific process. The

efficiency of this electron transfer reaction was almost entirely determined by the

thermodynamics of the system studied.

The primary objective of this thesis was to alter the physical properties of arylamines so that they

would be freely flowing liquids at room temperature. We have shown two successful strategies

to accomplish this: the use of bulky silyl ether groups which can be installed with conventional

chemistry and the use of oligosiloxane groups which can be installed with the Piers-Rubinsztajn

reaction. Using bulky silyl ether groups was moderately successful on single triarylamine

substrates. We were able to generate materials which had glass transition temperatures well

below room temperature and in one case a free flowing liquid triarylamine. However, when

applied to larger substrates, this strategy failed to prevent crystallization and to lower the glass

transition low enough to allow for liquid materials. Despite this, we showed that this

modification strategy had a very minor impact on the useful electrochemical properties of the

base materials. As well, time-of-flight photocurrent mobility measurements on a liquid

triarylamine in this series confirms that these materials retain their favourable charge

transporting properties. Analysis of these results reveals that the charge transport behavior

through this liquid organic semiconductor is similar to what is found in conventional materials as

well.

148

148

Our second strategy to develop liquid organic semiconductors involved the novel application of a

Si-H activation chemistry known colloquially as the Piers-Rubinsztajn reaction. This synthetic

strategy allowed the use of readily available silanes and easily-purified methoxy functionalized

triarylamines as substrates. This chemistry resulted in the development of a wide library of

various triaryalmine structures, many of which are freely flowing liquids at room temperature.

And, much like the triarylamines functionalized with silyl ether groups, these materials all

displayed very predicatable electrochemical behavior. Uniquely, we have shown that some

substrates functionalize very poorly with this chemistry due to a non-productive outer-sphere

charge transfer reaction between highly electron rich triarylamines substrates and the very

electron poor borane catalyst. Despite this, the chemistry remains usable on a wide range of

different substrates.

This synthetic strategy was also extended beyond making liquid triaryalmines. By using cyclic

ethers instead of alkoxides as reaction partners, it is shown that cross-linked films can also be

synthesized by this Piers-Rubinsztajn chemistry. In a proof of concept experiment, we have

shown that triarylamines can be easily cross-linked into amorphous, glassy films with low

catalyst loadings. Much like we observed for the liquid triaryalmines, the glassy films seem to

retain their electrochemical properties despite the change in physical state.

Finally, two side projects were pursued: the synthesis of novel deep-blue fluorescent emitters for

OLEDS and the synthesis of boron subphthalocyanines with varied optoelectronic properties.

Novel deep-blue fluorescent emitters for OLEDs were explored by using a very simple synthesis

of π-extended phthalonitriles. These structures were shown to be highly fluorescent and non-

optimized OLEDs using them as functional elements displayed moderate efficiencies and

excellent colour renderings close to NTSC standards for blue displays. Using these same

phthalonitrile type materials, new boron subphthalocyanine derivatives were synthesized which

displayed widely tuneable electrochemical and photophysical properties. We show that by

combining these electron-rich phthalonitriles with electron poor phthalonitriles can allow fine-

tuning of the optical bandgap to produce multi-coloured subphthalocyanine derivatives. As well,

highly-stable electrochemistry was observed of a wide range suggesting that such materials may

have a future in organic electronic devices.

149

149

To put this work in a bit of perspective, the overall goal and theme of this work was really to

explore how far the physical properties of organic semiconductors could be altered and

engineered. As previously stated throughout this document, basically all organic semiconductors

in the literature are designed to be high glass transition temperature amorphous solids or highly

crystalline solids. The work in this thesis differentiates inself from other semiconductor

development because we are interested in materials with low glass transition temperatures and

minimized intermolecular interactions.

While current semiconductor designs have indeed been successful for a variety or applications,

they can be lacking in certain instances. The most important area where current materials fail is

in the development of flexible and elastomeric devices. Flexible and stretchable devices are a

talking point on almost any introductory slide about organic electronics. And indeed, given the

promise of numerous applications of such devices, this is a laudable goal. However, it is

somewhat irritating and suprising that no one ever botheedr to think about designing active layer

materials with flexibility or stretchability in mind. Proof of concept demonstrations of such

devices in the literature ultimately rely on building conventional devices on flexible or

elastomeric substrates. This approach is ultimately limited as brittle and crystalline materials can

crack or delaminate under physical strain. I believe that it is clear that further development

towards flexible or stretchable device will require a focus on improving the active layer materials

used with an emphasis on their physical properties.

The materials developed in this thesis including the liquid and near elastomeric materials were

designed to overcome these problems by virtue of the materials being inherently able to respond

to physical stress in a nondestructive manner. One can imagine bending an organic electronic

device with multiple layers of different materials. At the point of bending, the materials can be

under considerable stress. Having a material that can readily deform to this stress can partially

relieve fatigue experienced by the device and improve the overall mechanical durability fot he

device. Similarly, having elastomeric organic semiconductors on top of an elastomeric substrate

may allow for the development of an elastomeric device. Of course, these lofty claims require

further material development and device integration to prove the concept. Regardless, these new

materials really represent a new approach to thinking about organic electronic materials by

designing them with the final physical state in mind instead of just electronic properties.

150

150

I believe that the synthetic strategies developed within this thesis have the potential to allow the

development of truly flexible or stretchable electronic devices. This is enabled by relatively

simple and scaleable chemistry and the use of silicone components. The Piers-Rubinsztajn

reaction used throughout this thesis is very easy to perform and our demonstration of using it to

polymerize samples in-situ further hints at its utility. As well, the use of silicones derived from

silanes allows the use of a wide array of already commercialized starting materials.

Understanding the full scope of this chemical strategy will require only time and creativity.

All of this potential on the chemistry side is further encouraged by the preliminary findings that

silicone functionalization has very little detrimental effect on the charge transporting properties

of triarylamines. A great deal of further work will be required to make a general case and other

classes of materials will be needed as well.

11.2 Future Directions

The following paragraphs are some ideas for future projects based on the work presented here in

this thesis.

As stated above, I believe that there is a great deal of future potential in this project. This will

require new perspectives on the work and motivated researchers. There are two basic areas for

future work on this project: application of soft materials in devices and further exploration of the

chemistry.

In terms of application of these materials in devices, the key challenge will be in exploiting the

unique physical properties of these materials in devices. This obviously poses difficulties in

adapting device architectures to suite the unique processing requirements that these materials

present. Focusing on the liquid materials, the first major obstacle in application involves finding

a suitable architecture that allows for relatively thick semiconducting layers. Due to surface

tension of a liquid, it is unrealistic to utilize liquid organic semiconductors in devices which are

optimal with layer thicknesses under 1 µm. As well, it would be relatively difficult to find a way

to define a layer this thick using conventional spacers. This instantly rules out conventional

excitonic photovoltaic structures and OLEDs. Due to the intentionally low intermolecular order

151

151

found in these materials, tradition dictates that high performance organic field-effect transistors

are out of the question as well. But all is not lost. There are many potentially useful device types

that may not be as popular as the aforementioned examples, but may be more amenable to liquid

organic semiconductors.

The easiest place to start would be in devices which have already been shown to work with

liquid organic or low Tg semiconductors. Photorefractive devices had a boom in publications in

early 2000s but no commercialization or common application has been realized. Application of

our liquid organic semiconducting materials into these devices seems obvious as optimal

performance has already been achieved with amorphous, low Tg semiconductors. Typically, the

low Tg requirement was achieved by the use of plasticizers or low Tg semiconducting polymers.

Using a low Tg semiconductor could lead to a simplified cell structure. As well, a liquid organic

semiconductor has already been proven to be a viable option for these devices.1 More

background here can be found in Chapter 2. Application of our materials into photorefractive

formulations may be an immediate opportunity.

The other area which I believe may be of most interest to future graduate students is using these

materials as redox mediators for dye sensitized solar cells. A liquid triarylamine was used as a

non-volatile liquid mediator in a dye sensitized solar cell previously.2 Moreover, through

collaboration with a researcher at York university, we have already shown that silicon hybridized

triarylamines can function as efficient redox mediators.3 This example however, relies on a solid

materials in a solvent solution. Provided that any formulation issues can be solved, I believe that

superior efficiencies can be achieved using our new synthetic methodologies as a basis for

optimized materials.

Another relatively straightforward project for the liquid organic semiconductors would be

extension beyond triarylamines to other functional materials. We initially chose triarylamines as

they were simple to produce and could be studied fairly easily. But, other materials with new

functions should also be amenable to this chemistry and may open up the door to further

application.

The cross-linked systems that we demonstrated have much more potential to be applied in

conventional devices if time is taken to properly investigate them. We have already shown that

152

152

very thin layers can be produced with this system so that integration into OLEDs or organic

photovoltaics should be possible. However, the biggest unknown remaining in this project is the

effect of residual tris(pentafluorophenyl)borane (BCF) catalyst in the film itself. Because this

catalyst is highly electrophilic and electrochemically active, there is the possibility that it may

interact with the functional elements within the cross-linked films. The extent of this interaction

would need to be immediately studied before further integration is realized.

The other major avenue for the cross-linked films is the exploitation of longer chain siloxane

components to further lower the Tg of the final film. Currently, we have films with glass

transition temperatures just about room temperature. Including large sections of flexible siloxane

groups should be effective in lowering the glass transition and result in films that are elastomeric

in nature. I believe that such elastomeric films would be a great interest to the community and be

one of the first examples of a solid-state semiconductor with inherent mechanical flexibility. As

well, using this strategy with non-triarylamine based monomers should be an obvious step

forward to produce multi-functional films.

11.3 References

1. Ribierre, J.; Aoyama, T.; Kobayashi, T.; Sassa, T.; Muto, T.; Wada, T. J. Appl. Phys.

2007, 102, 033106

2. Snaith, H. J.; Zakeeruddin, S. M.; Wang, Q.; Péchy, P.; Grätzel, M. Nano Lett. 2006, 6,

2000.

3. Sepehrifard, A.; Kamino, B.A.; Bender, T.P.; Morin, S. ACS App. Mater. Int. 2012,, 4,

6211-6215.

153

153

Appendices

12.1 Additional Information for Chapter 3

12.1.1 Experimental Information

Materials. All reagents and starting materials were used as received. All solvents were

purchased from Caledon Laboratories (Ontario, Canada) and used as received except toluene

which was purified through a commercial solvent purification system prior to use. Deuterated

NMR solvents were purchased from Cambridge Isotopes. External standards were prepared by

sealing the standard in a melting point tube with a flame and placing it freely within the NMR

tube. NMR spectra were acquired on a Varian 400 NMR system with a field strength of 400

MHz. Size exclusion chromatography was performed using Waters Styragel HR0.5 and a Waters

Styragel HR1 placed in series, each having a column size of 4.6 x 300 mm. GPC was operated

with THF as the mobile phase at a rate of 0.75 mL/min. Detection was achieved by a UV-Vis

photodiode array. Fluorescence spectroscopy for the quenching measurements was performed in

a dimly lit room with a Perkin-Elmer L55 spectrometer. Mass spectroscopy for the dendritic

triarylamines was acquired on an Accu-TOF mass spectrometer (JEOL USA Inc. Peabody, MA)

with a DART-SVP ion source (Ionsense Inc., Saugus, MA) using He Gas at 300-500 °C.

Samples were dissolved in CH2Cl2 and introduced into the sampling region using glass melting-

point capillaries. Compounds 3-6a and 3-6b required volatilization using a butane torch directly

on the sample. Mass spectroscopy for samples 3-2a, 3-2b, and 3-2c was achieved using an

AB/Sciex QStar mass spectrometer with an ESI source (50:50 methanol and water). Cyclic

voltammetry was performed with a Bioanalytical Systems C3 electrochemical cell setup. The

working electrode was a 1 mm platinum disk with a platinum wire used as a counter electrode.

The reference electrode was a Ag/AgCl saturated salt solution. All electrochemistry was done in

“Spectro” grade dichloromethane from Caledon Laboratories. Decamethylferrocene was added

to the solutions as an internal reference, and all electrochemical half-wave potentials are

corrected to its published potential of -0.012 V (vs Ag/AgCl). The synthesis of 3,4-DMPhO-

BsubPc and the syntheses of compounds 3-1a to 3-1i have been previously reported in our

group.

154

154

Synthesis of 3-2a. 1,4-Phenylene diamine (1.000 g, 9.25 mmol), sodium tert-butoxide (5.332 g,

55.5 mmol), and bis-(dibenzylideneacetone)palladium (106 mg, 0.18 mmol) were added to a

round-bottom flask. This flask was sealed under an argon atmosphere. Anhydrous toluene

(50mL), 4-bromoanisole (7.611 g, 40.69 mmol), and tri-tert-butylphosphine (29.9 mg, 0.15

mmol, added as a stock solution in toluene) were added. This mixture was refluxed under an

inert atmosphere for 2 h. Upon cooling, acidic clay (10 g, montmorillonite K10) and acidic

alumina (1 g, standard basic) were added to the mixture. This slurry was filtered, washing with

toluene to yield a clear, pale yellow solution. This solution was precipitated into methanol to

yield 4.237 g of a fine pale white powder (86.3% yield). 1H NMR (400 MHz, C6D6): δ7.14 (d, J

= 9.08 Hz, integration obscured by solvent peak), 7.08 (s, 4H), 6.73 (d, J = 9.08 Hz, 8H), 3.30 (s,

12H). 13C NMR (100 MHz, C6D6): 156.34, 143.84, 142.55, 126.35, 124.17, 115.45, 53.39.

HRMS [M+] calculated 532.2356, found 532.2372.

Synthesis of 3-2b. 1,3-Phenylene diamine (1.000 g, 9.25 mmol), sodium tert-butoxide (5.332 g,

55.5 mmol), and bis-(dibenzylideneacetone)palladium (106 mg, 0.18 mmol) were added to a

round-bottom flask. This flask was sealed under an argon atmosphere. Anhydrous toluene

(50mL), 4-bromoanisole (7.611 g, 40.69 mmol), and tri-tert-butylphosphine (29.9 mg, 0.15

mmol, added as a stock solution in toluene) were added. This mixture was refluxed under an

inert atmosphere for 2 h. Upon cooling, acidic clay (10 g, montmorillonite K10) and acidic

alumina (1 g, standard basic) were added to the mixture. This slurry was filtered washing with

toluene to yield a clear, pale yellow solution. This solution was concentrated under vacuum and

precipitated into methanol. A light yellow powder was collected and dried under vacuum to

obtain 3.728 g of dry material (75.7% yield). 1H NMR (400 MHz, C6D6): δ 7.11 (d, J = 9.08 Hz,

8H), 6.73 (dd, J1 = 7.91 Hz, J2 = 2.34 Hz, 2H), 7.05 (t, J = 2.34 Hz, 1H), 7.02 (d, J = 7.92 Hz,

1H), 6.67 (d, J = 9.08 Hz, 8H), 3.28(s, 12H). 13C NMR (100 MHz, C6D6): 156.56, 150.48,

141.92, 130.30, 127.04, 115.36, 115.30, 114.71, 53.31. HRMS [M+H] calculated 533.2434,

found 533.2429.

Synthesis of 3-2c. Bis(4-methoxyphenyl)amine (3.824 g, 16.8 mmol), 4,40-dibromobiphenyl

(2.500 g, 8.0 mmol), sodium tertbutoxide (1.922 g, 20 mmol), palladium(II) acetate (72 mg, 0.3

mmol), tri-tert-butylphosphine (52 mg, 0.3 mmol, added from stock solution in anhydrous

toluene), and anhydrous toluene (20 mL) were reacted for 3 h at reflux under an inert

155

155

atmosphere. After cooling, acidic clay (2 g, montmorillonite K10) and acidic alumina (1 g,

standard basic) were added, and the mixture was filtered. The compound was precipitated into

methanol and recrystallized from EtOAc. Slightly crystalline yellow flakes were collected (4.880

g, 70%). 1H NMR (400 MHz, C6D6): δ 7.46 (d, J = 8.61 Hz, 4H), 7.17 (m, obscured by residual

solvent), 7.12 (d, J = 9.00 Hz, 8H), 6.74 (d, J = 9.00 H, 8H), 3.31 (s, 12H). 13C NMR (100 MHz,

C6D6): δ 156.43, 148.16, 141.72, 133.98, 127.56, 126.82, 122.04, 55.06.

Synthesis of N-tBOC-bis(4-bromophenyl)amine (3-4). Bis(4-bromophenyl)amine (25 g, 76.43

mmol), di-tert-butyl dicarbonate (18.32 g, 84.07 mmol), 4-dimethylaminopyridine (1.872 g,

15.29 mmol), and tetrahydrofuran (125 mL) were added to an oven-dried flask. The flask was

refluxed for 24 h and allowed to cool. Approximately 75 mL of solvent was removed by rotary

distillation, and the flask was cooled to 5 °C overnight. Large white crystals were isolated from

the mother liquor and washed lightly with cold methanol. The final pure product was dried

overnight under vacuum (26.14 g, 80% yield). 1H NMR (400 MHz, CDCl3): δ 7.43 (d, J = 8.86

Hz, 4H), 7.06 (d, J = 8.86 Hz, 4H), 1.44 (s, 12H). 13C NMR (100 MHz, CDCl3): δ 153.35,

141.93, 132.15, 128.68, 119.52, 82.16, 28.35.

Synthesis of Generation 1 Dendron (3-5a). Palladium acetate (146 mg, 0.65 mmol) and tri-tert-

butylphosphine (105.25 mg, 0.52 mmol) were added to an oven-dried flask under a flow of argon

and allowed to stir for 30 min. 4 (27.77 g, 65.01 mmol), bis(3,4-dimethylphenyl)amine (3-3)

(30.766 g, 136.54 mmol), sodium tert-butoxide (16.66 g, 173.36 mmol), and 90 mL of toluene

were then added to the flask. The solution was refluxed for 2 h and allowed to cool. When the

solution was at room temperature, an additional 100 mL of toluene was added, and the solution

was treated with 5 g of basic alumina and 5 g of montmorillonite K10 clay. The solution was

filtered, and the yellow mother liquor was collected and concentrated under vacuum. The now

concentrated and viscous solution was precipitated into 50 mL of methanol and allowed to stir.

The collected solid was dried and placed into a vessel with 10 mL of tetralin (1,2,3,4-

tetrahydronaphthalene) and heated at 200 °C under an argon atmosphere overnight. The still

yellow solution was allowed to cool and precipitated into 50 mL of methanol to yield a white

solid (40.721 g, 78% yield). Mass Spectroscopy [M+H]: 616.3690. Expected [M+H]: 616.3691.

Synthesis of Generation 1 Dendrimer (3-5b). Bis(dibenzylideneacetone) palladium (14 mg,

0.024 mmol) and tri-tertbutylphosphine (4 mg, 0.02 mmol, added as a stock solution in toluene)

156

156

were added to an oven-dried flask under a flow of argon and allowed to stir for 30 min.

Compound 3-5a (300 mg, 0.49 mmol), 4-bromotoluene (13 mg, 0.73 mmol), sodium tertbutoxide

(94 mg, 1.0 mmol), and 2 mL of anhydrous toluene were added, and the solution was refluxed

overnight. Upon completion, the solution was allowed to cool and then treated with basic

alumina (0.25 g) andmontmorillonite K10 clay (0.25 g). The solids were filtered, washing with

additional toluene, and reduced to a viscous liquor. This liquor was precipitated by dropwise

addition into 50 mL of stirring methanol. A white, fluffy powder was collected (305 mg, 89%

yield). The compound was characterized by NMR, mass spectroscopy, and gel permeation

chromatography. 1H NMR (400 MHz, CDCl3): δ 7.15-7.09 (m, 14H), 7.03 (dd, J1 = 8.19 Hz, J2

= 2.14 Hz, 4H), 6.92 (d, J = 7.99 Hz, 4H), 6.87 (d, J = 8.19 Hz, 2H), 2.07 (s, 3H), 2.00 (s, 12H),

1.91 (s, 12H). Mass Spectroscopy [M+H]: 706.4143. Expected [M+H]: 706.4161.

Synthesis of Generation 2 Dendron (3-6a). Palladium acetate (217 mg, 0.97 mmol) and tri-tert-

butylphosphine (156 mg, 0.77 mmol) were added to an oven-dried flask under a flow of argon

and allowed to stir for 30 min. 3-5a (20 g, 32.47 mmol), 4 (6.866 g, 16.08 mmol), sodium tert-

butoxide (4.057 g, 42.22 mmol), and 100mL of toluene were then added to the flask. The

solution was refluxed for 18 h and allowed to cool. When the solution was at room temperature,

an additional 150 mL of toluene was added, and the solution was treated with 5 g of basic

alumina and 5 g of montmorillonite K10 clay. The solution was filtered, and the yellow mother

liquor was collected and concentrated under vacuum. The now concentrated and viscous solution

was precipitated into 50 mL of methanol and allowed to stir. The collected solid was dried and

placed into a vessel with 10 mL of tetralin (1,2,3,4-tetrahydronaphthalene) and heated at 200 °C

under an argon atmosphere overnight. The still yellow solution was allowed to cool and

precipitated into 50 mL of methanol to yield a white solid (22.455 g, 91% yield). Mass

Spectroscopy [M+H]: 1396.8. Expected [M+H]: 1396.8.

Synthesis of Generation 2 Dendrimer (3-6b). Bis(dibenzylideneacetone) palladium (6 mg, 0.01

mmol) and tri-tert-butylphosphine (1.74 mg, 0.009 mmol, added as a stock solution in toluene)

were added to an oven-dried flask under a flow of argon and allowed to stir for 30 min.

Compound 3-6a (300 mg, 0.215 mmol), 4-bromotoluene (55 mg, 0.322 mmol), sodium tert-

butoxide (41 mg, 0.430 mmol), and 2 mL of anhydrous toluene were added, and the solution was

refluxed overnight. Upon completion, the solution was allowed to cool and then treated with

157

157

basic alumina (0.25 g) and montmorillonite K10 clay (0.25 g). The solids were filtered, washing

with additional toluene, and reduced to a viscous liquor. This liquor was precipitated by adding

dropwise into 50 mL of stirring methanol. A yellow, fluffy powder was collected (295 mg, 92%

yield). Compound was characterized by NMR, mass spectroscopy, and gel permeation

chromatography. 1H NMR (400 MHz, C6D6 �): δ 7.14 6.86 (m, 52), 2.09 (s, 3H), 2.00 (s, 24H),

1.90 (s, 24H). 13C NMR (100 MHz, C6D6): δ 147.11, 144.23, 143.57, 143.55, 143.14, 137.94,

132.30, 131.19, 131.16, 131.00, 130.58, 128.27, 126.47, 125.45, 125.39, 125.01, 124.85, 122.80,

21.12, 20.14, 19.42. Mass Spectroscopy [M+H]: 1486.8. Expected [M+H]: 1486.8.

12.1.2 Supplemental Information of Merit

NN

NH2

NH2


Toluene

OMe

Br

N

NH2


Toluene

OMe

Br

MeO OMe

OMeMeO

NH2

N

OMe

MeO OMe

OMe

Br BrNH

OMe

MeO Pd(OAc)2P(t-butyl)3Na(t-butoxide)

TolueneN N

MeO

MeO OMe

OMe

2a

2b

2c

Figure S3-1: Synthetic scheme for the synthesis of two nitrogen centered triarylamines (3-2a-

c)

3.0 Electrochemistry of New Compound

158

158

-0.4 -0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4

Cu

rre

nt

(µA

)

Voltage (V)

2a

2c

2b

2b to1100 mV

Figure S3-2: Cyclic voltammetry of compounds 3-2a-c in dichloromethane with 0.1 M

tetrabutylammonium perchlorate scanned at a rate of 50 mV/s.

-0.4 0.1 0.6 1.1 1.6

Voltage (V)

5b

5b to 1400 mV

6b

6b to 1400 mV

Figure S3-3: Cyclic voltammetry of compounds 3-5b and 3-6b in dichloromethane with 0.1 M

tetrabutylammonium perchlorate scanned at a rate of 50 mV/s.

159

159

Table S3-1: Collects the calculated molecular volumes and radii from molecular mechanics calculations and the square of the residual values from the Stern Volmer plots.

Compound

Results from MM calculations Quality of Stern

Volmer Fit

Molecular Volume

(Å3) Spherical Radii (Å) R2

3-1a 369.96 4.453369398 0.9381

3-1b 379.56 4.491560731 0.9967

3-1c 370.53 4.455655337 0.9946

3-1d 361.5 4.419161746 0.9927

3-1e 424.56 4.662478281 0.997784

3-1f 365.28 4.434511256 0.994205

3-1g 357.09 4.401118138 0.992294

3-1h 357.06 4.400994885 0.97302

3-1i 294 4.124958406 -

3-2a 569.62 569.62 0.9893

3-2b 569.61 569.61 0.9976

3-2c 653.47 653.47 0.9999

3-5b 804.95 653.47 0.9699

3-6b 1673.57 653.47 0.9973

160

160



12.2.1.1 General Procedures

All silicon containing materials were purchased from Gelest Inc. and used without further

purification. Palladium catalysts were purchased from Strem Chemicals Inc. and stored in an

inert atmosphere glovebox. Reagent grade solvents were purchased from Caledon Laboratories.

Toluene and DMF were dried and degassed using a solvent purification system prior to use. All

other solvents were used as received. Sodium tert-butoxide was stored either in an inert

atmosphere glovebox or in a desiccator. All Buchwald-Hartwig aminations were performed in a

glove box with an argon atmosphere. Other reactions were preformed under an argon atmosphere

using standard procedures. Reactions were monitored by reverse phase HPLC (Waters PAH C18

5 µm, acetonitrile mobile phase 1.2 mL/min, photodiode array detector) or by low molecular

weight GPC (Waters Styragel HR-1 and Styragel HR-2 in series, eluting with THF at 0.75

mL/min). All NMR spectra were collected on a Varian Mercury 400 spectrometer. Chemical

shifts are reported in parts per million referenced to either tetramethylsilane (TMS) internal

standard or relative to residual C-H solvent peaks. Coupling constants (J) are reported in Hz.

Mass spectroscopy was taken with an AB/Sciex QStar mass spectrometer. Samples were

introduced with an ESI source in solution (50:50 methanol and water) via an HPLC pump.

Cyclic voltammetry was performed with a Bioanalytical Systems C3 electrochemical cell setup.

The working electrode was a 1 mm platinum disc with a platinum wire used as a counter

electrode. The reference electrode was a Ag/AgCl saturated salt solution. All electrochemistry

was done in spectro grade dichloromethane from Caledon Laboratories. Decamethylferrocene

was added to the solutions as an internal reference. All electrochemical half wave potentials are

corrected to the decamethylferrocene half wave potential of −0.012 V (vs. Ag/AgCl).

Differential scanning calorimetry was performed with a TA instruments DSC2920 equipped with

a refrigerated cooling system. All samples were heated from 30 °C to 200 °C or 110 °C at 10

°C/min with a ±0.5 °C modulation every 30 s. The samples were rapidly cooled to −50 °C then

heated again to 200 °C or 110 °C at 10 °C/min with the same temperature modulation.

161

161

12.2.1.2 Synthesis

bis(3,4-dimethylphenyl)amine (4-2): 3,4-Dimethylaniline (50.3 g, 0.415 mol) was added to a

round bottom flask was heated to 60 °C under argon. After the aniline was completely melted,

anhydrous calcium chloride (12.0 g, 0.108 mol), aluminum chloride (15.4 g, 0.115 mol), and 30

mL of tetrahydronaphthalene (tetralin) were added. The flask was heated to 220 °C for 18 h. The

flask was allowed to cool and diluted with 60 mL of toluene. The organic phase was added to

~200 mL of ice water and allowed to separate. The organic phase was washed with 3×75 mL of

5% aqueous hydrochloric acid solution, 75 mL of 5% aqueous sodium bicarbonate and 100 mL

of an aqueous saturated brine solution. The organic phase was dried over MgSO4 and distilled

under vacuum to remove remaining solvent. After the majority of the tetralin had been removed

the organic phase was allowed to cool to ~60 °C, at this point 40 mL of methanol was quickly

added and the solution was stored at 3-5 °C overnight. Slightly red crystals were collected and

washed with cold methanol and dried under vacuum. 21.52 g was collected (46% Yield). 1H

NMR (400 MHz, CDCl3): δ 7.00 (d, J=7.95 Hz, 2 H), 6.83 (s, 2 H), 6.79 (d, J=7.95 Hz, 2 H),

5.28 (s, 1 H), 2.21 (s, 6 H), 2.19 (s, 6 H). 13C NMR (100 MHz, CDCl3): δ 141.80, 137.30,

130.51, 129.07, 119.67, 115.43, 20.19, 19.15.

p-triisopropylsiloxy-bromobenzene (4-3a): Triisopropylchlorosilane (25.0 g, 0.1297 mol), 4-

bromophenol (21.3 g, 0.1232 mol), imidazole (16.8 g, 0.2464 mol), and 125 mL of N,N-

dimethylformamide were stirred at room temperature overnight. The solution was added to a

separatory funnel where 65 mL of water was added. The product was extracted with hexanes

(3x100mL). The hexanes phase was dried over magnesium sulphate and removed under vacuum

to yield a clear oil. 40.57 g was recovered (100% yield). Trace amounts of DMF can be detected

in the 1 H NMR (resonances not indicated). 1H NMR (400 MHz, CDCl3): δ 7.31 (d, J= 8.848 Hz,

2 H), 6.77 (d, J=8.848 Hz, 2 H), 1.25 (s, 3 H), 1.1 (d, J=7.16 Hz, 18 H). 13C NMR (100 MHz,

CDCl3): δ 155.51, 132.47, 121.92, 113.44, 18.11, 12.86. Theoretical MS (ESI) of C15H25BrOSi

[M]: 328.1. Mass found: 328.1

p-tertbutyldimethylsiloxy-bromobenzene (4-3b): The same general procedure for 4-3a was

repeated, using tertbutyldimthylsilyl chloride. The product was isolated as a clear oil in

quantitative yield. The product was used without further purification. Trace amounts of DMF can

be detected in the 1H NMR (resonances not indicated). 1H NMR (400 MHz, CDCl3): δ7.33 (d,

162

162

J=8.6 Hz, 2 H), 6.73 (d, J=8.59 Hz, 2 H), 0.99 (s, 9 H), 0.19 (s, 6 H). 13C NMR (100 MHz,

CDCl3): δ 154.66, 132.15, 121.73, 113.50, 25.52, 18.02, - 4.61. Theoretical MS (ESI) of

C12H19BrOSi [M]: 286.0. Mass found: 286.0

p-terthexylsiloxy-bromobenzene (4-3c): The same general procedure for 4-3a was repeated,

using tert-hexyl-di-methylsilyl chloride. The product was isolated as a slightly yellow clear oil in

quantitative yield. The product was used without further purification. 1H NMR (400 MHz,

CDCl3): δ7.32 (d, J=8.59 Hz, 2 H), 6.71 (d, J= 8.59 Hz, 2 H), 1.73 (m, J=7.03 Hz, 1 H), 0.94 (d,

J= 7.03 Hz), 0.94 (s, integration with previous peak comes 12 H), 0.22 (s, 6 H). 13C NMR (100

MHz, C6D6): δ 155.27, 132.96, 122.55, 114.30, 34.85, 25.64, 20.72, 19.22, -2.01. Theoretical

MS (ESI) of C14H23OBrSi [M]: 314.1. Mass found: 314.1

p-diphenyltertbutylsiloxy-bromobenzene (4-3d): The same general procedure for 4-3a was

repeated, using di-phenyl-tert-butylsilyl chloride. The product was isolated as a white soft

crystalline solid in quantitative yield. The product was used without further purification. 1H

NMR (400 MHz, CDCl3): δ7.74-7.67 (m, 4 H), 7.48-7.36 (m, 6 H), 7.19 (d, J=9.18 Hz, 2 H),

6.64 (d, 9.2 Hz, 2 H), 1.1 (s, 9 H). 13C NMR (100 MHz, CDCl3): δ 154.71, 135.43, 132.39,

132.05, 130.02, 127.83, 121.45, 113.32, 26.42, 19.40. Theoretical MS (ESI) of C22H23BrOSi

[M]: 410.1 Mass found: 410.1

4-4a: Palladium acetate (102 mg, 0.45 mmol) and tri-tertbutylphosphine (73 mg, 0.36 mmol, in

0.01 g/mL stock solution in anhydrous toluene) were added to a round bottom flask and stirred at

room temperature for 30 min. To this flask, p-triisopropylsiloxy-bromobenzene (4-3a) (2.98 g,

9.048 mmol), bis(3,4-dimethylphenyl)amine (2) (2.24 g, 10 mmol), and sodium tert-butoxide

(0.961 g, 10 mmol) were added and the solution was refluxed for 4 h. The solution was taken out

of the glove box where it was treated with 10 g of Montmorillonite clay (Montmorillonite K10

from Sigma) for 15 min and subsequently filtered to remove solids leaving a slightly yellow

solution which was washed with additional toluene. Solvent was removed under vacuum leaving

an off white fluffy solid. This was dispersed in 100 mL of methanol and allowed to stir in the

dark overnight. The methanol slurry was filtered yielding a pure white solid. 3.60 g was

recovered (84% yield). 1H NMR (400 MHz, CDCl3): δ 6.99–6.90 (m, 4H), 6.84–7.70 (6H), 2.19

(s, 6 H), 2.14 (s, 6H), 1.24 (m, 3H), 1.10 (d, J= 7.1 Hz, 18H). 13C NMR (100 MHz, CDCl3): δ

163

163

151.74, 146.28, 141.65, 137.13, 130.07, 129.82, 126.32, 124.41, 120.60, 120.49, 19.89, 19.02,

17.91, 12.62. Theoretical MS (ESI) of C31H44NOSi [M+H]: 474.3. Mass found: 474.3.

4-4b: Same procedure as 4-4a except 4-3b was used as the aryl bromide. The completed reaction

was purified by column chromatography on silica eluting with cyclohexane and toluene. The

product was isolated as a white crystalline solid in a 74% yield. 1H NMR (400 MHz, CDCl3): δ

7.15 (d), 7.11 (d, J=2.3 Hz, 2H), 7.04 (dd, J1=8.0 Hz, J2=2.3 Hz, 2H), 6.93 (d, J=8.0 Hz, 2H),

6.79 (d, J=8.96 Hz, 2H), 2.01 (s, 6H), 1.92 (s, 6H), 0.99 (s, 9H), 0.11 (s, 6H). 13C NMR (100

MHz, C6D6): δ 151.82, 147.50, 143.32, 137.84, 131.10, 130.75, 126.48, 125.93, 122.25, 121.27,

26.22, 20.16, 19.39, 12.75, -4.01. Theoretical MS (ESI) of C30H42NOSi [M+H]: 432.3 found:

432.3.

4-4c: Same procedure as 4-4a except 4-3c was used as an aryl bromide. The completed reaction

was purified by precipitating into 100 mL of methanol to yield a white crystalline solid. The

product was isolated as white crystalline powder in an 85% yield. 1H NMR (400 MHz, C6D6): δ

7.16, (d, J=8.77 Hz, integration obscured by solvent peak), 7.11 (d, J=2.3 Hz, 2H), 7.04 (dd,

J1=8.19 Hz, J2=2.3 Hz, 2H), 6.93 (d, J= 8.19 Hz, 2H), 6.78 (d, J=8.77 Hz, 2H), 2.01 (s, 6H),

1.92 (s, 6H), 1.69 (m, J=6.82 Hz, 1H), 0.96 (d, J=6.82 Hz), 0.94 (s, integration with previous

peak is 12H), 0.16 (s, 6H,). 13C NMR (100 MHz, C6D6): δ 151.66, 147.51, 143.29, 137.814,

131.10, 130.73, 126.52, 125.91, 122.23, 121.38, 34.90, 25.64, 20.75, 20.17, 19.39, 19.16, -2.02.

Theoretical MS (ESI) of C30H41NOSi [M+H]: 460.3 found: 460.3.

4-4d: Same procedure as 4-4a except 4-3d was used as an aryl bromide. The completed reaction

was purified by column chromatography on silica eluting with cyclohexane and toluene. The

product was isolated as light orange glass in 64% yield. 1H NMR (400 MHz, CDCl3): δ 7.82–

7.77 (m, 4H), 7.16–7.14 (m), 6.99 (d, J= 2.1 Hz, 2H), 6.93 (dd, J1=8.10 Hz, J2=2.53 Hz), 6.92

(d, J= 8.96 Hz, this and previous peak integrate to 4H), 6.86 (d, J=8.19 Hz, 2H), 6.78 (d, J=8.96

Hz, 2H), 1.99 (s, 6H), 1.89, (s, 6H), 1.18 (s, 9H). 13C NMR (100 MHz, C6D6): δ 151.75, 147.36,

141.16, 137.70, 136.33, 133.87, 131.01, 130.64, 130.49, 128.46, 126.08, 125.94, 122.26, 120.78,

27.15, 20.13, 20.04, 19.38. Theoretical MS (ESI) of C38H42NOSi [M+H]: 556.3 found: 556.3.

4-5: Palladium acetate (46 mg, 0.2 mmol) and tri-tertbutylphosphine (33 mg, 0.16 mmol, in 0.01

g/mL stock solution in anhydrous toluene) were added to a round bottom flask and stirred at

164

164


9.048 mmol), 3,4-dimethylaniline (4-1) (0.5 g, 4 mmol), and sodium tert-butoxide (0.990 g, 10

mmol) were added and the solution was refluxed for 4 h. The reacted solution was filtered

washing with toluene and the resulting oil was purified by column chromatography over silica

gel eluting with cyclohexane. The compound was isolated as a yellow oil. 1.15 g was collected

(45% yield). 1H NMR (400 MHz, CDCl3): δ 6.97–6.86 (5 H), 6.78–6.69 (6H), 2.18 (s, 3H), (2.12

(s, 3H), 1.23 (m, 6H), 1.09 (d, J=7.08 Hz, 36H). 13C NMR (100 MHz, CDCl3): δ 151.33, 146.59,

141.85, 137.05, 130.03, 129.37, 125.38, 123.75, 120.37, 119.89, 19.88, 18.87, 17.92, 12.63.

Theoretical MS (ESI) of C38H60NO2Si2 [M+H+]: 618.4. Mass found: 618.4.

4-6: Palladium acetate (46 mg, 0.2 mmol) and tri-tertbutylphosphine (33 mg, 0.16 mmol, in 0.01

g/mL stock solution in anhydrous toluene) were added to a round bottom flask and stirred at


9.048 mmol), lithium amide (0.064 g, 2.8 mmol), and sodium tert-butoxide (0.990 g, 9.048

mmol) was added. The solution was refluxed for 16 h. The reacted solution was filtered to

washing with additional toluene. The filtrate was purified by column chromatography over silica

gel eluting with cyclohexane. This yielded a yellow clear waxy solid. 0.925 g of product was

recovered (43.5% yield). 1H NMR (400 MHz, CDCl3): δ 6.86 (d, J=4.56 Hz, 6 H) 6.73 (d, J=4.56

Hz, 6 H), 1.23 (m, 9 H), 1.09 (d, J=7.18 Hz, 54 H). 13C NMR (100 MHz, CDCl3): δ 150.95,

142.11, 124.55, 120.29, 17.93, 12.63. Theoretical MS (ESI) of C45H76NO3Si3 [M+H]: 762.5.

Mass found: 762.5.

4-7: Bis(dibenylideneacetone)palladium(0) (71 mg, 0.124 mmol) and tri-tert-butylphosphine (33

mg, 0.16 mmol, in 0.01 g/mL stock solution in anhydrous toluene) were added to a round bottom

flask and stirred at room temperature for 30 min. To this solution, 1,3-phenylenediamine (671

mg, 6.21 mmol), p-triisopropylsiloxy-bromobenzene (4-3a) (9 g, 27.3mmol), and sodium tert-

butoxide (2.98 g, 31 mmol) were added and the solution was refluxed overnight. The solution

was then filtered eluting with additional toluene. The filtrate was purified by column

chromatography over silica gel eluting with cyclohexane then a gradation to 3:1

cyclohexane:toluene. The product was isolated as a slightly yellow perfectly clear glass. 3.119 g

of product was obtained (46% yield). 1H NMR (400 MHz, C6D6): δ7.13 (t, J=2.14 Hz, 1 H), 7.07

(d, J= 8.96 Hz, 8 H), 6.99 (t, J=8.18 Hz, 1 H), 6.77 (d, 8.96 Hz), 6.75 (dd, J1=8.18 Hz, J2=2.14

165

165

Hz, integration with previous peak is 10 H), 1.14 (m), 1.10 (d, J=5.65 Hz, integration over both

peaks is 84 H). 13C NMR (100 MHz, C6D6): δ152.48, 150.33, 142.39, 130.40, 126.64, 121.19,

116.48, 115.91, 18.51, 13.33. Theoretical MS (ESI) of C66H105N2O4Si4 [M+H]: 1101.7. Mass

found: 1101.7.

4-8a: Bis(dibenylideneacetone)palladium(0) (212 mg, 0.37 mmol) and tri-tert-butylphosphine

(60 mg, 0.296 mmol, in 0.01 g/mL stock solution in anhydrous toluene) were added to a round

bottom flask and stirred at room temperature for 30 min. To this solution, 1,4- phenylenediamine

(2 g, 18.5 mmol), p-triisopropylsiloxy- bromobenzene (4-3a) (26.80 g, 81.4 mmol), and sodium

tert-butoxide (8.88 g, 92.5 mmol) were added and refluxed overnight. The solution was then

filtered washing with additional toluene. This filtrate was purified by column chromatography

over silica gel eluting with cyclohexane and a gradation to 1:1 cyclohexane and toluene. The

purity of this compound was assessed by low molecular weight GPC and TLC analysis. The

product was isolated as a white crystalline solid. 14.460 g of product was obtained (71% yield).

A crystal suitable for X-ray diffraction was grown by taking up 50 mg of compound in 5 mL of

toluene, this vial was placed in a sealed container with 100 mL of acetonitrile. 1H NMR (400

MHz, C6D6): δ7.10 (d, J=8.96 Hz, 8 H), 7.06 (s, 4 H), 6.80 (d, J=8.96 Hz, 8 H), 1.14 (m), 1.11

(d, J=6.04 Hz). 13C NMR (100 MHz, C6D6): δ 152.19, 143.82, 142.84, 126.06, 124.60, 121.18,

18.51, 13.34. Theoretical MS (ESI) of C66H104N2O4Si4 [M]: 1100.7. Mass found: 1100.7.

4-8b: Bis(dibenylideneacetone)palladium (0) (38 mg, 0.066 mmol) and tri-tert-butylphosphine


bottom flask and stirred at room temperature for 30 min. To this solution, 1,4-phenylenediamine

(90 mg, 0.83 mmol), p-diphenyltertbutylsiloxybromobenzene (4-3d) (2 g, 4.98 mmol), and

sodium tertbutoxide 400 mg, 4.16 mmol) were added and was refluxed overnight. The solution

was filtered washing with and the filtrate was purified by column chromatography over silica gel

eluting with 5:1 cyclohexane:toluene and then a gradation to 1:1 cyclohexane:toluene. 0.881 g of

small white needles was obtained (71% yield). 1H NMR (400 MHz, CDCl3): δ 7.86–7.79 (m, 16

H), 7.19–7.14 (m, integration obscured by solvent), 6.84–6.71 (m, 20 H), 1.18 (s, 36 H). 13C

NMR (100 MHz, C6D6): δ 151.14, 142.86, 142.25, 135.80, 133.32, 130.02, 127.99, 125.26,

123.95, 120.45, 26.62, 19.51. Theoretical MS (ESI) of C96H96N2O4Si4 [M]: 1428.6. Mass found:

1428.6

166

166

4-9: Bis(dibenylideneacetone)palladium (0) (125 mg, 0.217 mmol) and tri-tert-butylphosphine


bottom flask and stirred at room temperature for 30 min. To this solution, benzidine (1.000 g,

5.428 mmol), p-triisopropylsiloxy-bromobenzene (4-3a) (10.726 g, 32.57 mmol), and sodium

tert-butoxide (3.129 g, 32.57 mmol) were added with 30 mL of toluene to wash down the solid.

The reaction was refluxed for 35 h and allowed to cool. The solution was treated with 10 g of

Montmorillonite clay (Montmorillonite K10 from Sigma) and filtered to remove solid by

washing with toluene. The product was purified by column chromatography over silica gel

eluting with 10:1 cyclohexane:toluene and then a gradation to 4:1 cyclohexane:toluene. 5.280 g

of white hard solid was obtained (83% yield). 1H NMR (400 MHz, CDCl3): δ7.42 (d, J=8.57 Hz,

4 H), 7.17 (d, J=8.77 Hz), 7.08 (d, J=8.77 Hz, 8 H), 6.83 (d, J=8.57 Hz, 4 H), 1.15 (m) 1.11 (d,

J=6.04 Hz). 13C NMR (100 MHz, C6D6): δ 152.64, 148.35, 142.45, 134.54, 127.89, 126.78,

123.02, 121.27, 18.51, 13.36. Theoretical MS (ESI) of C72H108N2O4Si4 [M]: 1176.7. Mass found:

1176.7.

4-10: Bis(dibenylideneacetone)palladium (0) (92 mg, 0.16 mmol) and tri-tert-butylphosphine

(0.026 g, 0.128 mmol, in 0.01 g/mL stock solution in anhydrous toluene) were added to a round

bottom flask and stirred at room temperature for 30 min. To this solution, 4,4`-dibromobiphenyl

(5 g, 16 mmol), 3,4-dimethylaniline (4.078 g, 33.6 mmol), and sodium tert-butoxide (4.620 g, 48

mmol) were added with 25 mL of toluene to wash down the solid. The reaction was refluxed for

3 h. The resulting solution was treated with 10 g of montmorillonite clay (Montmorillonite K10

from Sigma) and filtered to remove solids by washing with toluene. The toluene phase was

washed 3 times with 100 mL of a 10% hydrochloric acid solution, dilute sodium carbonate and

finally dried over magnesium sulfate. The organic phase was reduced yielding a yellow powder

that was used as is. 1.136 g obtained (18% Yield). The product was used as prepared without

further purification. Theoretical HRMS for C28H28N2 [M+H] 393.2, found 393.2.

4-11: Palladium (II) acetate (6 mg, 0.0254 mmol) and tri-tertbutylphosphine (4 mg, 0.0203

mmol, in 0.01 g/mL stock solution in anhydrous toluene) were added to a round bottom flask and

stirred at room temperature for 30 min. To this solution, 10 (0.5 g, 1.27 mmol), p-

triisopropylsiloxybromobenzene (4-3a) (0.923 g, 2.802 mmol) and sodium tertbutoxide (0.305 g,

3.175 mmol) were added and the solution was refluxed for 48 h. The mixture was filtered

167

167

washing with toluene. The filtrate was purified using column chromatography over silica gel

eluting with 5:1 cyclohexane:toluene. The product was isolated as a yellow glass. 0.743 g of

material was obtained (66% yield). 1H NMR (400 MHz, CDCl3): δ 7.42 (d, J=8.77 Hz, 4 H),

7.23 (d, J=8.77 Hz, 4 H), 7.11 (d, J=8.96), 7.09 (d, J=2.34 Hz), 7.03 (dd, J1=8.19 Hz, J2=2.34 Hz,

2 H), 6.83 (d, J= 8.19 Hz, 2 H), 6.84 (d, J=8.96 Hz, 4 H), 2.02 (s, 6 H), 1.93 (s, 6 H), 1.15 (m),

1.11 (d, 5.85 Hz). 13C NMR (100 MHz, C6D6): δ 152.76, 148.27, 146.87, 142.44, 138.00, 134.70,

131.46, 131.21, 127.95, 127.20, 126.52, 123.38, 122.88, 121.28, 20.15, 19.44, 18.50, 13.34.

Theoretical MS (ESI) of C58H76N2O2Si2 [M]: 888.5. Mass found: 888.5

12.2.1.3 Hydrolytic Stability Testing

A 0.02 M solution of arylamine in 4.75 mL of 1,4-dioxane and 0.25 mL of a 1 M HCl or 10 M

NaOH aqueous solution was prepared. For an internal standard, 4,4`-dibromobiphenyl was

utilized. The solutions were stirred in the dark monitoring the decomposition by HPLC analysis.

Decomposition was fitted to a first order exponential decay and the half-life was extracted from

this curve. Further details are included in the supporting information.

168

168



Materials: Toluene was purified using a PureSolv solvent purification system just prior to use.

Tris(pentafluorophenyl)borane was obtained from the Sigma-Aldrich and used without further

purification. Pentamethyldisiloxane was obtained from Gelest Inc. and was used without further

purification. Deuterated benzene (C6D6) was purchased from Cambridge Isotopes and used

without further purification. Reactions were monitored by reverse phase HPLC (Waters PAH

C18 5µm, acetonitrile mobile phase 1.2 mL/min), a photo diode array (200nm-500nm) was used

for detection. All NMR spectra were collected on a Varian Mercury 400 spectrometer in C6D6.

Chemical shifts are reported in parts per million referenced relative to residual C-H solvent

peaks. Coupling constants (J) are reported in Hz. High resolution mass spectroscopy was taken

with an AB/Sciex QStar mass spectrometer. Samples were introduced with an ESI source in

solution (50:50 methanol and water) via an HPLC pump. Cyclic voltammetry was performed

with a Bioanalytical Systems C3 electrochemical cell setup. The working electrode was a 1mm

platinum disc with a platinum wire used as a counter electrode. The reference electrode was

Ag/AgCl2 saturated salt solution. All electrochemistry was done in ‘Spectro’ grade

dichloromethane from Caledon Laboratories. Decamethylferrocene was added to the solutions as

an internal reference. All DSC half wave potentials are corrected to the decamethylferrocene half

wave potential of -0.012 V (vs. Ag/AgCl). Differential scanning calorimetry was preformed with

a TA instruments Q2000 with a refrigerated cooling system. Tests were performed under a

blanket of nitrogen. Triarylamines 5-1a-c were synthesized as previously reported (Bender, T.P.;

Graham, J.F.; Duff, J.M.; Chem. Mater. 2001, 13, 4105.)

Compound 5-2a. In a round bottom flask, tris(pentafluorophenyl)borane (15 mg, 0.029 mmol, 1

mol%) was dissolved in toluene (10 mL) to which the arylamine 5-1a (1.0 g, 3.21 mmol) was

added. To this pentamethyldisiloxane (1.428 g, 9.63 mmol) was slowly added drop wise.

CAUTION under these reaction conditions, methane gas is vigorously generated in solution and

expelled by bubbling. A strong exotherm has also been observed. These considerations should be

taken into account when performing or scaling up this chemistry. Solutions were warm to the

touch after reaction. After bubbling ceased, the solutions were allow to stir for 20 minutes. At

169

169

this point 0.5 g of basic standard alumina was added to the reaction and allowed to stir for an

additional 20 minutes. The solution was filtered. Solvent and excess silane were removed by

rotary evaporation over several hours. 1.329g of product was collected (2.87 mmol, 95%). 1H

NMR (400 MHz, C6D6): δ 7.16 (d, J = 8.96 Hz, integration obscured by residual solvent peak),

7.09 (d, J = 2.14 Hz, 2H), 7.03 (dd, J1 = 7.99 Hz, J2 = 2.14 Hz, 2H), 6.94 (d, J = 8.96 Hz), 6.92

(d, J = 8.01 Hz, integration over this and last peak is 4H), 2.02 (s, 6H), 1.92 (s, 6H), 0.21 (s, 6H),

0.11 (s, 9H) 13C NMR (100 MHz, C6D6): 150.92, 147.48, 143.62, 137.82, 131.10, 130.76,

126.39, 125.98, 122.31, 121.22, 20.15, 19.39, 2.16, 0.21. Theoretical HRMS for C33H57NO6Si6

[M+H] 732.2874 found 732.2876.

Compound 5-2b: Same general procedure as for compound 5-2a using: arylamine 5-1b (1g,

3.00 mmol), tris(pentafluorophenyl)borane (15 mg, 0.029 mmol, 1 mol%),

pentamethyldisiloxane (2.670 g, 18 mmol). 1.644 g of product was collected (2.75 mmol, 92%). 1H NMR (400 MHz, C6D6): δ 7.10 (d, J= 8.96 Hz, 4H), 7.04 (d, J= 2.14 Hz, 1H), 6.97 (dd, J1

=8.19 Hz, J2 = 2.14 Hz, 1H), 6.92 (d, J= 8.96 Hz), 6.90 (d, J= 8.19 Hz, integration over this and

last peak is 5H), 2.02 (s, 3H), 1.92 (s, 3H), 0.22 (s, 12H), 0.11 (s, 18H). 13C NMR (100 MHz,

C6D6): 150.81, 147.47, 143.60, 137.79, 131.08, 130.56, 126.11, 125.56, 121.89, 121.21, 20.15,

19.39, 2.16, 0.19. Theoretical HRMS for C30H47NO4Si4 [M+H] 598.2654 found 598.2650.

Compound 5-2c: Same general procedure as for compound 5-2a using: arylamine 5-1c (0.500 g,

1.49 mmol), tris(pentafluorophenyl)borane (8 mg, 0.015 mmol, 1 mol%), pentamethyldisiloxane

(1.990 g, 13 mmol). 1.047 g of product was collected (1.43 mmol, 96%). 1H NMR (400 MHz,

C6D6): δ 7.04 (d, J= 8.96 Hz, 6H), 6.90 (d, J= 8.96 Hz, 6H), 0.22 (s, 18H), 0.11 (s, 27H). 13C

NMR (100 MHz, C6D6): 150.71, 143.65, 125.78, 121.20, 2.15, 0.18. Theoretical HRMS for

C27H37NO2Si2 [M+H] 464.2435 found 464.2425.



Safety note: For reactions involving Piers−Rubinsztajn chemistry (coupling of silanes using

tris(pentaf luorophenyl)borane), the reaction can proceed extremely rapidly, evolving flammable

gases with a noticeable exotherm. In our opinion, care should be exercised in scaling this

170

170

chemistry. The caref ul dropwise addition of silane(s) at a moderate rate is recommended, and if

no reaction is detected after several minutes, the addition of silane should be halted.

General Piers−Rubinsztajn Procedure (General P−R procedure).

The aryl-methoxy-functionalized substrate, tris-(pentafluorophenyl)borane, and toluene were

added to a generously sized round-bottom flask with a magnetic stirrer under atmospheric

conditions. The flask should be at least four times the volume of the reagents to prevent any

material from spilling out upon reaction. To maximize the reaction yield, the P−R reaction

should be done at anhydrous conditions using predried reagents and solvent. Silane was added

dropwise to this stirring solution, taking care to control the rate of methane evolving. Upon

completion, standard activity basic alumina (∼0.5 g) was added, and the solution was allowed to

stir for 20 min. After this time, the solution was filtered to remove the added alumina and

reduced under vacuum for purification.

Compound 6-1. 1,4-Phenylene diamine (1.000 g, 9.25 mmol), sodium tert -butoxide (5.33 g,

55.5 mmol), and bis- (dibenzylideneacetone)palladium (106 mg, 0.184 mmol) were added to a

round-bottom flask. This flask was sealed under an argon atmosphere. Anhydrous toluene (50.0

mL), 4-bromoanisole (7.61 g, 40.7 mmol), and tri-tert-butylphosphine (29.9 mg, 0.148 mmol,

added as a stock solution in toluene) were added. This mixture was refluxed under an inert

atmosphere for 2 h. Upon cooling, acidic clay (10.0 g, Montmorillonite K10) and acidic alumina

(1.00 g, standard basic) were added to the mixture. This slurry was filtered, washing with toluene

to yield a clear, pale yellow solution. This solution was concentrated under vacuum and

precipitated into methanol. 4.24 g of a fine pale white powder was obtained (86% yield): 1H

NMR (400

MHz, C6D6) δ 7.14 (d, J = 9.08 Hz, integration obscured by solvent peak), 7.08 (s, 4H), 6.73 (d, J

= 9.08 Hz, 8H), 3.30 (s, 12H); 13C NMR (100 MHz, C6D6) δ 156.3, 143.8, 142.6, 126.4, 124.2,

115.5, 53.4; HRMS (ESI) [M+] calcd for C34H32N2O4 532.2356, found 532.2372.

Compound 6-2a. Using the general P−R procedure, 6-1 (500 mg, 0.939 mmol), anhydrous

toluene (10.0 mL), tris(pentafluorophenyl)-borane (5 mg, 0.01 mmol), and

pentamethyldisiloxane (3.14 g, 21.3 mmol) were reacted. After 20 min, the solution was

171

171

immersed in an oil bath at 50 °C. The solution vigorously evolved methane and was allowed to

continue to stir for 20 min. The product was purified by column chromatography over silica gel

eluting with 2:1 cyclohexane/toluene. 661 mg of a flakey crystalline solid was isolated (yield

66%): 1H NMR (400 MHz, C6D6) δ 7.09 (d, J = 8.98 Hz, 8H), 7.01 (s, 4H), 6.91 (d, J = 8.98 Hz,

8H), 0.22 (s, 24H), 0.11 (s, 36H); 13C NMR (100 MHz, C6D6) δ 151.0, 143.8, 143.3, 126.2,

124.6, 121.3, 2.1, 0.2; HRMS (ESI) [M + H] calcd for C50H81N2O8Si8 1061.4141, found

1061.4146.

Compound 6-2b. Using the general P−R procedure, compound 6-1 (0.500 g, 0.939 mmol),

1,1,1,3,5,5,5-heptamethyltrisiloxane (2.51 g, 11.3 mmol), tris(pentafluorophenyl)borane (5 mg,

0.01 mmol), and anhydrous toluene (5.00 mL) were reacted. The product was isolated by column

chromatography over silica gel eluting with 2:1 cyclohexane/toluene. A viscous yellow oil was

isolated (1.034 g, 81% yield): 1H NMR (400 MHz, C6D6) δ 7.10 (d, J = 8.77 Hz, 8H), 7.07 (s,

4H), 6.97 (d, J = 8.77 Hz, 8H), 0.27 (s, 12H), 0.15 (s, 72H); 13C NMR (100 MHz, C6D6) δ 150.2,

143.4, 143.1, 125.9, 124.1, 120.9, 1.7, −3.0; HRMS (ESI) [M+] calcd for C58H104N2O12Si12

1356.4821, found 1356.4814.

Compound 6-3a. Aniline (5.98 g, 64.2 mmol), 4-bromoanisole (10.0 g, 53.5 mmol), sodium tert-

butoxide (7.71 g, 80.2 mmol), bis(dibenzylideneacetone)palladium (154 mg, 0.268 mmol), tri-

tertbutylphosphine (43.0 mg, 0.212 mmol, as a 10 g/L solution in toluene), and anhydrous

toluene (75.0 mL) were added to a round bottom flask. The flask was refluxed for 2 h under

argon gas. Once cool, acid-washed clay (montmorillonite K10 from Sigma Aldrich, 10.0 g) and

standard basic alumina (1.00 g) were added to the slurry and stirred for 30 min. The slurry was

filtered, and the clear light yellow mother liquor was collected. This organic phase was washed

with 10% HCl solution three times and with brine once and then dried over magnesium sulfate.

The organic phase was concentrated under vacuum and recrystallized from boiling heptanes to

yield small silver needles. 7.43 g of product was isolated (67% yield): 1H NMR (400 MHz,

C6D6) δ 7.23 (t, J = 7.99 Hz, 2H), 7.09 (d, J = 8.57 Hz, 2H), 6.95−6.82 (m, 5H), 5.50 (s, broad,

1H), 3.81 (s, 3H).

172

172

Compound 6-3b. Using the same general procedure as for compound 6-3a above, p-toluidine

(6.88 g, 64.2 mmol), 4-bromoanisole (10.0 g, 53.5 mmol), sodium t-butoxide (7.71 g, 80.2

mmol),

bis(dibenzylideneacetone)palladium (154 mg, 0.268 mmol), tri-tertbutylphosphine (43 mg, 0.212

mmol), and anhydrous toluene (75.0 mL) were reacted at reflux for 2 h. Product recrystallized

from heptanes to yield light brown flakes. 7.26 g of pure product was isolated (64% yield): 1H

NMR (400 MHz, CDCl3) δ 7.06−7.00 (m, 4H), 6.88−6.82 (m, 4H), 5.39 (s, broad, 1H), 3.80 (s,

3H), 2.28 (s, 3H).

Compound 6-3c. Using the same general procedure as for compound 6-3a above, 3,4-

dimethylaniline (3.56 g, 29.4 mmol), 4-bromoanisole (5.00 g, 26.7 mmol), sodium t-butoxide

(3.84 g, 40.0 mmol), bis(dibenzylideneacetone)palladium (154 mg, 0.268 mmol), tri-tert-

butylphosphine (43 mg, 0.212 mmol), and anhydrous toluene (35.0 mL) were refluxed for 2 h.

Product recrystallized from heptanes to yield silver needles. 4.50 g of pure product was isolated

(67% yield): 1H NMR (400 MHz, CDCl3) δ 7.05−6.95 (m, 3H), 6.83 Hz (d, J = 8.96 Hz, 2H),

6.76−6.64 (m, 2H), 5.33 (s, broad, 1H), 3.78 (s, 3H), 2.19 (s, 6H).

Compound 6-3d. Using the same general procedure as for compound 6-3a above, 2-bromo-6-

methoxynaphthalene (10.0 g, 42.2 mmol), para-anisidine (6.23 g, 50.6 mmol), sodium t-butoxide

(5.95 g, 61.9 mmol), bis(dibenzylideneacetone)palladium (119 mg, 0.207 mmol), tri-tert-

butylphosphine (33 mg, 0.163 mmol), and anhydrous toluene (50.0 mL) were refluxed for 1 h.

Product recrystallized from heptanes to yield fine gray needles (5.54 g, 47% yield): 1H NMR

(400 MHz, CDCl3) δ 7.53 (d, J = 9.00 Hz, 1H), 7.39 (d, J = 0.99 Hz, 1H), 7.18 (dd, J1 = 11.35

Hz, J2 = 2.74 Hz, integration obscured by residual solvent peak), 7.02−6.91 (m, 4H), 6.79 (d, J =

8.61 Hz, 2H), 5.01 (s, broad, 1H), 3.44 (s, 3H), 3.37 (s, 3H). NMR (100 MHz, C6D6) δ 156.81,

156.09, 141.91, 137.09, 131.20, 130.49, 128.60, 128.50, 122.48, 120.45, 119.88, 115.42, 111.25,

106.85, 55.48, 55.19; HRMS (ESI) [M+ H] calcd for C18H18NO2 280.1338, found 280.1346.

Compound 6-4a. Compound 6-3a (3.00 g, 15.1 mmol), 4,4′-dibromobiphenyl (2.24 g, 7.17

mmol), sodium tert-butoxide (2.07 g, 21.5 mmol), palladium(II) acetate (16 mg, 0.0713 mmol),

tri-tertbutylphosphine (16 mg, 0.0791 mmol, added as a stock solution in toluene), and

173

173

anhydrous toluene (25 mL) were added to a round bottom flask. This solution was refluxed for 1

h under an argon atmosphere. Upon cooling, acid-washed clay (montmorillonite K10 from

Sigma Aldrich, 2.00 g) and standard activity basic alumina (0.500 g) were added and allowed to

stir into the mixture for 10 min. The solids were filtered out, washing with additional toluene.

The mother

liquor was collected and concentrated under vacuum until the solution began to become a

viscous oil. This oil was precipitated into rapidly stirring methanol (50 mL) to yield a fine white

powder. This powder was collected by filtration and washed with cold methanol. 2.635 g of

product was isolated (67% yield): 1H NMR (400 MHz, C6D6) δ 7.41 (d, J = 8.61 Hz, 4H),

7.19−7.13 (m, peak obscured by solvent peak), 7.10 (d, J = 7.41 Hz, 4H), 7.06 (d, J = 9.00 Hz,

4H), 6.85 (t, J = 7.04 Hz, 2H), 6.71 (d, J = 9.00 Hz, 4H), 3.29 (s, 6H).

Compound 6-4b. Using the same general procedure as for compound 6-4a, compound 6-3b (3

g, 14.1 mmol), 4,4′-dibromobiphenyl (2.09 g, 6.67 mmol), sodium tert-butoxide (1.93 g, 20.1

mmol),

palladium(II) acetate (15 mg, 0.0668 mmol), tri-tert-butylphosphine (11 mg, 0.0544 mmol,

added from stock solution in anhydrous toluene), and anhydrous toluene (25.0 mL) were reacted

for 2 h at reflux. 7.06 g of isolated material was collected (87% yield): 1H NMR (400 MHz,

C6D6) δ 7.43, (d, J = 8.77 Hz, 4H), 7.11 (d, J = 8.77 Hz, integration obscured by solvent),

7.15−7.08 (m, integration obscured by solvent), 6.94 (d, J = 7.99 Hz, 4H) 6.72 (d, J = 8.96 Hz,

4H), 3.30 (s, 6H), 2.12 (s, 6H); 13C NMR (100 MHz, C6D6) δ 156.7, 147.9,146.3, 141.5, 134.4,

132.1, 130.3, 127.6, 127.4, 124.4, 123.0, 115.2, 55.0, 20.8; HRMS (EI) [M+] calcd for

C40H36N2O2 576.2777, found 576.2769.

Compound 6-4c. Using the same general procedure as for compound 6-4a, compound 6-3c

(6.41 g, 28.2 mmol), 4,4′-dibromobiphenyl (4.00 g, 12.8 mmol), sodium tert-butoxide (3.08 g,

32.0 mmol), palladium(II) acetate (115 mg, 0.512 mmol), tri-tert-butylphosphine (83 mg, 0.410

mmol, added from stock solution in anhydrous toluene), and anhydrous toluene (25.0 mL) were

reacted for 3 h at reflux. The compound precipitated poorly in methanol and was purified by

column chromatography over silica gel eluting with 1:1 cyclohexane/toluene. A fine white

powder was collected (5.21 g, 67%): 1H NMR (400 MHz, C6D6) δ 7.44 (d, J = 8.96 Hz, 4H),

174

174

7.22 (d, J = 8.77 Hz, 4H), 7.15 (d, integration and coupling obscured by residual solvent peak),

7.11 (d, J = 2.14 Hz, 2H), 7.03 (dd, J1 = 8.18 Hz, J2 = 2.14 Hz, 2H), 6.94 (d, J = 8.18 Hz, 2H),

6.73 (d, J = 8.96 Hz, 4H), 3.30 (s, 6H), 2.03 (s, 6H), 1.93 (s, 6H). 13C NMR (100 MHz, C6D6) δ

156.6, 148.0, 146.6, 141.7, 137.7, 134.3, 131.0, 130.9, 127.6, 127.3, 126.0, 122.8, 122.4, 115.2,

55.0, 19.8, 19.1; HRMS (ESI) [M+] calcd for C42H40N2O2 604.3090, found 604.3100.

Compound 6-4d. Using the same general procedure as for compound 6-4a, bis(4-

methoxyphenyl)amine (3.82 g, 16.8 mmol), 4,4′-dibromobiphenyl (2.50 g, 8.01 mmol), sodium t-

butoxide (1.92 g, 20.0 mmol), palladium(II) acetate (72 mg, 0.321 mmol), tri-tertbutylphosphine

(52 mg, 0.257 mmol, added from stock solution in anhydrous toluene), and anhydrous toluene

(20.0 mL) were reacted for 3 h at reflux. The compound was precipitated into methanol and

recrystallized from EtOAc. Slightly yellow flakes were collected (3.33 g, 70%): 1H NMR (400

MHz, C6D6) δ 7.46 (d, J = 8.61 Hz, 4H), 7.17 (m, obscured by residual solvent), 7.12 (d, J = 9.00

Hz, 8H), 6.74 (d, J = 9.00 H, 8H), 3.31 (s, 12H); 13C NMR (100 MHz, C6D6) δ 156.4, 148.2,

141.7, 134.0, 127.6, 126.8, 122.0, 115.17, 55.0.

Compound 6-4e. Using the same general procedure as for compound 6-4a, 6-3d (2.43 g, 8.71

mmol), 4,4′-dibromobiphenyl (1.35 g, 4.34 mmol), sodium t-butoxide (1.29 g, 13.4 mmol),

palladium(II)

acetate (13 mg, 0.0579 mmol), tri-tert-butylphosphine (9 mg, 0.046 mmol), and anhydrous

toluene (15.0 mL) were refluxed for 1 h. Compound was precipitated into methanol to yield a

fine white powder. The product was further purified by recrystallization from toluene/EtOAc to

yield small white crystals (2.66 g, 86% yield): 1H NMR (400 MHz, C6D6) δ 7.56 (d, J = 1.96 Hz,

2H), 7.52 (d, J = 9.00 Hz, 2H), 7.47 (d, J = 8.61 Hz, 4H), 7.42 (dd, J1 = 9.00 Hz, J2 = 2.35

Hz, 2H), 7.27−7.22 (m, 6H), 7.17−7.09 (m, integration obscurd by residual solvent peak), 6.94

(d, J = 2.35 Hz, 2H), 6.75 (d, J = 9.00 Hz, 4H), 3.40 (s, 6H), 3.32 (s, 6H); 13C NMR (100 MHz,

C6D6) δ 157.8, 157.1, 148.2, 144.8, 144.8, 135.0, 131.9, 130.9, 129.3, 128.5, 128.1, 127.8, 125.6,

123.6, 121.1, 119.8, 115.7, 106.6, 55.4, 55.2; HRMS (ESI) [M + H] calcd for C48H41N2O4

709.3066, found 709.3048.

175

175

Compound 6-5a. Using the general P−R procedure, 6-4a (0.500 g 0.937 mmol),

tris(pentafluorophenyl)borane (5 mg, 0.00977 mmol), and 5.00 mL of toluene were stirred, and

pentamethyldisiloxane (1.35 g, 9.11 mmol) was added dropwise. The reaction proceeded

vigorously and was allowed to stir for an additional 20 min after bubbling had ceased. The

product was purified by column chromatography over silica gel eluting with 5:2

cyclohexane/toluene. The product was isolated as a viscous pale yellow oil (0.391 g, 51% yield): 1H NMR (400 MHz, C6D6) δ 7.38 (d, J = 8.77 Hz, 4H), 7.16−7.12 (m, integration obscured by

solvent), 7.12−7.02 (8H), 6.92 (d, J = 9.16 Hz, 4H), 6.84 (t, J = 7.21 Hz, 2H), 0.23 (s, 12H), 0.12

(s, 18H); 13C NMR (100 MHz, C6D6) δ 151.5, 148.7, 147.6, 142.2, 135.0, 129.6, 127.7, 127.3,

123.9, 123.9, 122.5, 121.1, 1.8, −0.2; HRMS (ESI) [M+] calcd for C46H56N2O4Si4 812.3317,

found 812.3300.

Compound 6-5b. Using the general P−R procedure, 6-4b (0.500 g, 0.867 mmol),

pentamethyldisiloxane (1.29 g, 8.67 mmol), tris(pentafluorophenyl)borane (5 mg, 0.00977

mmol), and anhydrous toluene (5.00 mL) were reacted at room temperature; the reaction

proceeded vigorously. The compound was purified by column chromatography eluting with 5:2

cyclohexane/toluene. The product was isolated as a viscous and pale yellow oil (0.350 g, 48%

yield): 1H NMR (400 MHz, C6D6) δ 7.40 (d, J = 8.77 Hz, 4H), 7.18 (m, obscured by solvent),

7.13−7.06 (m, 8H), 6.97−6.89 (m, 8H), 2.11 (s, 6H), 0.23 (s, 12H), 0.12 (s, 18H); 13C NMR (100

MHz, C6D6) δ 151.2, 147.8, 146.2, 142.5, 134.6, 132.2, 130.3, 127.6, 126.9, 124.7, 123.3, 121.0,

20.8, 1.8, −0.1; HRMS (ESI) [M+] calcd for C48H60N2O4Si4 840.3630, found 840.3639.

Compound 6-5c. Using the general P−R procedure, 6-4c (0.500 g, 0.827 mmol),


mmol), and anhydrous toluene (5.00 mL) were reacted. The product was purified by column

chromatography eluting with 5:2 cyclohexane/toluene. The product was isolated as a viscous and

pale yellow oil (0.467, 65% yield): 1H NMR (400 MHz, C6D6) 7.41 (d, J = 8.96 Hz, 4H), 7.2 (d,

J = 8.57 Hz, 4H), 7.13 (d, 8.96 Hz, 4H), 7.08 (d, J = 2.14 Hz, 2H), 7.01 (dd, J1 = 5.65 Hz, J2 =

2.14 Hz, 2H), 6.97−6.91 (m, 6H), 2.02 (s, 6H), 1.92 (s, 6H), 0.23 (s, 12H), 0.12 (s, 18H); 13C

NMR (100 MHz, C6D6) δ 151.1, 147.9, 146.5, 142.7, 137.7, 134.5, 131.2, 130.9, 127.6, 126.8,

176

176

126.3, 123.2, 122.7, 121.0, 19.8, 19.1, 1.9, −0.1; HRMS (ESI) [M+] calcd for C50H64N2O4Si4

868.3943, found 868.3930.

Compound 6-5d. Using the general P−R procedure, 6-4d (0.500 g, 0.821 mmol),


mmol), and 10.0 mL of anhydrous toluene were reacted. The product was purified by passing the

compound through a plug of silica gel eluting with 1:1 hexanes/toluene. A clear oil was obtained,

which slowly crystallized to soft white crystals after several days (0.860 g, 92% yield): 1H NMR

(400 MHz, C6D6) δ 7.40 (d, J = 8.77 Hz, 4H), 7.14 (partially obscured by solvent peak), 7.08 (d,

J = 8.96 Hz, 8H), 6.93 (d, J = 8.96 Hz, 8H), 0.23 (s, 24H), 0.12 (s, 36H); 13C NMR (100 MHz,

C6D6) δ 151.0, 148.0, 142.6, 134.3, 127.6, 126.5, 122.8, 121.0, 1.8, −0.2; HRMS (ESI) [M+]

calcd for C56H84N2O8Si8 1136.4382, found 1136.4368.

Compound 6-5e. Using the general P−R procedure, 6-4b (0.500 g, 0.867 mmol), 1,1,1,3,5,5,5-

heptamethyltrisiloxane (1.930 g, 8.67 mmol), tris(pentafluorophenyl)borane (4 mg, 0.00781

mmol), and anhydrous toluene (5 mL) were reacted at room temperature; the reaction proceeded

rapidly. The compound was purified by column chromatography eluting with 5:2

cyclohexane/toluene. The final product was isolated as a viscous clear oil (0.592 g, 69% yield): 1H NMR (400 MHz, C6D6) δ 7.41 (d, J = 8.77 Hz, 4H), 7.18 (d, J = 8.96 Hz, integration obscured

by solvent), 7.13−7.07 (m, 8H), 7.00 (d, J = 8.96 Hz, 4H), 6.92 (d, J = 8.77 Hz, 4H), 2.11 (s, 6H),

0.28 (s, 6H), 0.17 (s, 36H); 13C NMR (100 MHz, C6D6) δ 150.8, 147.8, 146.2,

142.6, 134.6, 132.2, 130.3, 127.7, 126.9, 124.6, 123.2, 121.0, 20.8, 1.7, −3.0; HRMS (ESI) [M+]


Compound 6-5f. Using the general P−R procedure, 6-4d (0.500 g, 0.821 mmol), 1,1,1,3,5,5,5-

heptamethyltrisiloxane (1.46 g, 6.57 mmol), tris(pentafluorophenyl)borane (4 mg, 0.00781

mmol), and anhydrous toluene (10.0 mL) were reacted. Compound was purified by column

chromatography eluting with 1:1 cyclohexane/toluene. The final product was isolated as a pale

yellow oil (1.06 g, 90% yield): 1H NMR (400 MHz, C6D6) δ 7.43 (d, J = 8.77 Hz, 8H), 7.16

(multiplicity and integration obscured by solvent peak), 7.09 (d, J = 8.77 Hz, 8H), 7.00 (d, J =

8.96 Hz, 4H), 0.28 (s, 12H), 0.16 (s, 72H); 13C NMR (100 MHz, C6D6) δ 150.6, 148.0, 142.8,

177

177

134.3, 127.6, 128.5, 122.7, 121.0, 1.7, −3.0; HRMS (ESI) [M+] calcd for C64H108N2O12Si12

1432.5134, found 1432.5153.

Compound 6-5g. Using the general P−R procedure, 6-4e (0.500 g, 0.705 mmol),


mmol), and toluene (10.0 mL) were reacted. Compound was purified by column chromatography

eluting with 1:1 cyclohexane/toluene to yield a slightly green powder (0.426 g, 49% yield): 1H

NMR (400 MHz, C6D6) δ 7.54 (d, J = 2.14 Hz, 2H), 7.47 (d, J = 9.2 Hz, 2H), 7.44 (d, J = 8.77

Hz, 4H), 7.42 (d, J = 2.34 Hz, 2H), 7.34 (dd, J1 = 8.96 Hz, J2 = 2.34 Hz, 2H), 7.26 (d, J = 8.96

Hz, 2H), 7.20 (d, J = 8.77 Hz, 4H), 7.16 (obscured by solvent peak), 7.11 (d, J = 8.77 Hz, 4H),

6.97 Hz (d, J = 8.96 Hz, 4H), 0.27 (s, 12H), 0.24 (s, 12H), 0.14 (s, 12H), 0.13 (s, 12H); 13C NMR

(100 MHz, C6D6) δ 152.5, 151.7, 148.1, 145.0, 142.7, 135.2, 132.1, 131.4, 129.3, 127.4, 125.8,

124.0, 122.8, 121.4, 121.4, 115.6, 2.2, 2.2, 0.2, 0.1; HRMS (ESI) [M + H] calcd for

C64H89N2O8Si8 1237.4773, found 1237.4715.

Compound 6-5h. Using the general P−R procedure, 6-4e (400 mg, 0.564 mmol), 1,1,1,3,5,5,5-

heptamethyltrisiloxane (748 mg, 3.36 mmol), tris(pentafluorophenyl)borane (3 mg, 0.00587

mmol), and toluene (10.0 mL) were reacted. Compound purified by column chromatography

eluting with 1:1 cyclohexane/toluene to yield a viscous clear oil (803 mg, 93% yield): 1H NMR

(400 MHz, C6D6) δ 7.53−7.47 (m, 6H), 7.43 (d, J = 8.61 Hz, 4H), 7.34 (dd, J1 = 9.00 Hz,

J2 = 1.96 Hz, 2H), 7.30 (d, J = 9.00 Hz, 2H), 7.22 (dd, J1= 9.00 Hz, J2 =2.35 Hz, 2H), 7.19 (d, J

= 8.61 Hz, 4H), 7.07 (d, J = 9.00 Hz, 4H), 7.01 (d, J = 9.00 Hz, 4H), 0.32 (s, 6H), 0.29 (s, 6H),

0.17 (s, 72H); 13C NMR (100 MHz, C6D6) δ 152.0, 151.3, 148.0, 145.0, 142.8, 135.2, 132.0,

131.4, 129.2, 127.5, 125.7, 124.0, 122.6, 121.5, 121.1, 115.5, 2.12, 2.10, −2.6, −2.7; HRMS

(ESI) [M + H] calcd for C72H113N2O12Si12 1533.5519, found 1533.5583.

Compound 6-6. 2-Bromobiphenyl (5.00 g, 21.4 mmol) in anhydrous THF (11.0 mL) was

reacted with magnesium (0.567 g, 23.3 mmol) under inert gas at room temperature. Upon

formation of the Grignard reagent, 9-fluorenone (3.90 g, 21.6 mmol in 5.00 mL of THF) was

added, and the solution was refluxed for 4 h and allowed to cool. Upon cooling, a yellow

precipitate was formed and collected, washing with cold methanol. This solid was stirred into a

178

178

5% HCl solution (22.0 mL) for 2 h at room temperature and washed with additional methanol.

Finally, the solid was dissolved in acetic acid (22.0 mL) and refluxed for 40 min. Upon cooling,

large white crystals

of the pure product were obtained (3.73 g, 56% yield).

Compound 6-7. Compound 6-6 (1.00 g, 3.16 mmol), FeCl3 (2 mg, 0.0123 mmol), and 6 mL of

chloroform were mixed under an inert atmosphere. Molecular bromine (2.02 g, 12.6 mmol) in

2.00 mL of chloroform was added, and the reaction was allowed to proceed at room temperature

in the absence of light for 5 days. Upon completion, the reaction was quenched with aqueous

ammonium hydroxide and recrystallized from chloroform/ethanol (50/50 to yield a pure white

crystalline product (1.10 g, 55% yield).

Compound 6-8. Palladium(II) acetate (5 mg, 0.0223 mmol), 7(200 mg, 0.316 mmol), 6-3b (340

mg, 1.59 mmol), sodium tert-butoxide (145 mg, 1.51 mmol), and 2.30 mL of anhydrous toluene

were added to a round-bottom flask under an atmosphere of argon. Once sealed, tri-tert-

butylphosphine (3.75 mg, 0.0185 mmol) was added as a toluene stock solution. The mixture was

refluxed for 5 h under argon and allowed to cool. The solution was filtered through a short plug

of silica to yield a clear yellow liquid; this solution was concentrated and precipitated into

stirring methanol to yield the product as a fine white powder (258 mg, 67% yield): 1H NMR (400

MHz, C6D6) δ 7.12 (dd, J1 = 7.82 Hz, J2 = 2.93 Hz, 8H), 7.09−7.01 (m, 16H), 7.00 (dd, J1 =

8.22 Hz, J2 = 2.35 Hz, 4H), 6.95 (d, J = 8.22, 8H), 6.71 (d, J = 9.39 Hz, 8H), 3.25 (s, 12H), 2.05

(s, 12H); 13C NMR (100 MHz, C6D6) δ 156.7, 151.1, 148.2, 146.9, 141.94, 135.6, 131.7, 130.3,

127.2, 124.0, 123.5, 121.1, 119.2, 115.4, 66.8, 55.3, 21.1. MS (ESI) [M+] calcd for

C81H68N4O8 1160.5, found 1160.5

Compound 6-9. Using the general P−R procedure, 11 (200 mg, 0.172 mmol), 1,1,1,3,5,5,5-

heptamethyltrisiloxane (445 mg, 2.00 mmol), 2.00 mL of toluene, and

tris(pentafluorophenyl)borane (1 mg, 0.00195 mmol) were reacted at room temperature.

Compound was purified by column chromatography on silica gel eluting with hexanes/toluene.

Product was isolated as a glass (294 mg, 86% yield): 1H NMR (400 MHz, C6D6) δ 7.18 (signal

obscured by solvent), 7.09 (d, J = 2.35 Hz, 4H), 7.05−6.95 (m, 36H), 2.05 (s, 12H), 0.28 (s,

179

179

12H), 0.14 (s, 72H). 13C NMR (100 MHz, C6D6) δ 151.0, 150.9, 148.1, 146.6, 143.0, 136.4,

132.2, 130.4, 126.8, 124.3, 123.5, 121.2, 120.8, 118.8, 66.8, 21.2, 2.1, −2.6; MS (ESI) [M + 3H]


Compound 6-10. Tris(4-bromophenyl)amine (0.500 g, 1.04 mmol), compound 6-3c (0.786 g,

3.46 mmol), sodium t-butoxide (0.444 g, 4.63 mmol), palladium(II) acetate (5 mg, 0.00223

mmol), tri-tertbutylphosphine (3 mg, 0.00148 mmol), and anhydrous toluene (10.0 mL) were

refluxed for 16 h. Upon cooling, the mixture was filtered, and the resulting liquor was

concentrated and precipitated into methanol to yield a light yellow fine powder (742 mg, 78%

yield): 1H NMR (400 MHz, C6D6) δ 7.16−7.12 (m, integration obscured by residual solvent),

7.10−7.06 (m, 9H), 7.02 (dd, J1 = 8.02 Hz, J2 = 2.54 Hz, 3H), 6.92 (d, J = 8.22 Hz, 3H), 6.71 (d,

J = 9.00 Hz, 6H); 13C NMR (100 MHz, C6D6) δ 156.6, 147.2, 144.3, 143.1, 142.3, 137.9, 131.1,

130.8, 127.2, 125.7, 125.5, 124.5, 122.0, 115.5, 55.4, 20.2, 19.4; HRMS (ESI) [M + H] calcd for

C63H61N4O3 921.4744, found 921.4765.

Compound 6-11. Using the general P−R procedure, compound 6-10 (0.349 g, 0.379 mmol),

1,1,1,3,5,5,5-heptamethyltrisiloxane (0.871 g, 3.91 mmol), tris(pentafluorophenyl)borane (2 mg,

0.00391 mmol), and 5.00 mL of toluene were reacted for 16 h under ambient conditions. Product

was purified by column chromatography eluting with hexanes/toluene and isolated as a white

crystalline solid (269 mg, 46% yield): 1H NMR (400 MHz, C6D6) δ 7.14−7.09 (m, 12H), 7.08−

7.04 (m, 9H), 7.02−6.95 (m, 9H), 6.91 (d, J = 8.22 Hz, 3H), 2.01 (s, 9H), 1.92 (s, 9H), 2.07 (s,

9H), 0.15 (s, 54H); 13C NMR (100 MHz, C6D6) δ 150.7, 147.1, 144.2, 143.3, 143.1, 137.9, 131.1,

130.9, 126.7, 126.0, 125.4, 124.7, 122.3, 121.3, 20.15, 19.4, 2.1, −2.7; HRMS (ESI)

[M + H] calcd for C81H115N4O9Si9 1539.6587, found 1539.6546.

Compound 6-12a. In an inert atmosphere glovebox, 4-bromo-3-nitroanisole (3.00 g, 12.9

mmol), phenylboronic acid (1.73 g, 14.2 mmol ) , cesium fluoride (3.93 g, 25.8 mmol), bis-

(dibenzylideneacetone)palladium (149 mg, 0.258 mmol), tri-tertbutylphosphine (52 mg, 0.258

mmol, added as a 10 g/L stock solution in anhydrous toluene), and anhydrous tetrahydrofuran

(60.0 mL) were added to a stirred round-bottom flask. This flask was allowed to stir at ambient

temperature for 24 h. The resulting solution was removed from the glovebox, and the solids were

180

180

filtered, eluting with THF. This light brown solution was dried under vacuum. The remaining

solids were taken up in toluene and washed with water and brine. The organic toluene phase was

dried over magnesium sulphate and filtered through a short plug of silica gel, eluting with

toluene. This light-yellow liquid was concentrated under vacuum until it resembled a viscous oil.

Hexanes (5.00 mL) were added to this resulting oil to precipitate light yellow crystals. These

crystals were collected and washed with cold hexanes and then dried under vacuum (2.44 g

collected, 85% yield): 1H NMR (400 MHz, CDCl3) δ 7.43−7.32 ppm (m, 5H), 7.30−7.25 (m,

2H), 7.15 (dd, J1 = 8.6 Hz, J2 = 2.7 Hz, 1H), 3.90 (s, 3H); 13C NMR (100 MHz, CDCl3) δ 159.2,

137.4, 132.9, 128.80, 128.76, 128.2, 128.0, 118.8, 109.2, 56.1; HRMS (EI) [M+] calcd for

C13H11NO3 229.0733, found 229.0739.

Compound 6-12b. In an inert atmosphere glovebox, 4-bromo-3-nitroanisole (3.47 g, 15.0

mmol), 4-methoxyphenylboronic acid (2.50 g, 16.5 mmol), cesium fluoride (7.52 g, 49.5 mmol),

bis- (dibenzylideneacetone)palladium (259 mg, 0.45 mmol), tri-tertbutylphosphine (91 mg, 0.450

mmol, added as a 10 g/L stock solution in anhydrous toluene), and anhydrous tetrahydrofuran

(30.0 mL) were added to a stirred round-bottom flask. This flask was allowed to stir at ambient

temperature for 24 h. The resulting solution was removed from the glovebox, and the solids were

filtered, eluting with THF. This light brown solution was dried under vacuum. The remaining

solids were taken up in toluene and washed with water and brine. The organic toluene phase was

dried over magnesium sulphate and filtered through a short plug of silica gel, eluting with

toluene. This light yellow liquid was then concentrated under vacuum until it resembled a

viscous oil. Hexanes (5.00 mL) were added to this resulting oil, and the solution was cooled in a

refrigerator overnight. Light-yellow crystals were collected and washed with cold hexanes and

then dried under vacuum (2.55 g collected, 65% yield). This product was used without further

purification: 1H NMR (400 MHz, CDCl3) δ 7.34−7.30 (m, 2H), 7.21 (d, J = 8.96 Hz, 2H), 7.13

(dd, J1 = 8.57 Hz, J2 = 2.73 Hz, 1H), 6.96 (d, J = 8.96 Hz, 2H), 3.89 (s, 3H), 3.84 (s, 3H); 13C

NMR (100 MHz, CDCl3) δ 159.5, 159.0, 132.9, 129.6, 129.3, 128.4, 118.8, 114.3, 109.0, 104.9,

56.1, 55.4; HRMS (ESI) [M+] calcd for C14H13NO4 259.0845, found 259.0845.

Compound 6-13a. Compound 6-12a (2.20 g, 9.56 mmol) was added to a stirred round-bottom

flask, and the flask was flushed with argon gas and sealed under a positive pressure of gas.

181

181

Triethylphosphite (7.70 mL) was added via syringe, and the solution was heated at 160 °C for 16

h. The reaction was allowed to cool, and 10.0 mL of methanol was added to the solution. This

was allowed to rest in a refrigerator overnight to yield white square crystals. These crystals were

collected by filtration and washed sparingly with cold methanol. The collected crystals were then

dried under vacuum (1.69 g was isolated, 56% yield): 1H NMR (400 MHz, (CD3)2SO) δ 11.10 (s,

broad, 1H), 7.98 (m, 2H), 7.41 (d, J = 7.82 Hz, 1H), 7.28 (td, J1 = 7.63 Hz, J2 = 1.17 Hz, 1H),

7.10 (td, J1 = 7.82 Hz, J2 = 1.17 Hz, 1H), 6.96 (d, J = 2.35, 1H), 6.76 (dd, J1 =8.61 Hz, J2 = 2.35

Hz, 1H), 3.84 (s, 3H); 13C NMR (100 MHz, (CD3)2SO) δ 158.4, 141.0, 139.7, 124.0, 122.6,

120.9, 119.2, 118.5, 116.1, 110.6, 107.6, 94.4, 55.2; HRMS (ESI) [M + H] calcd for C13H12NO

198.0913, found 198.0906.

Compound 6-13b. Compound 6-12b (2.55 g, 9.81 mmol) and triethylphosphite (8.90 mL) were

heated at 160 °C for 16 h under argon. The reaction was allowed to cool, and 10.0 mL of

methanol was added to the solution. This was allowed to rest in a refrigerator overnight to yield

small white needles. These crystals were collected by filtration and washed sparingly with cold

methanol. The resulting white crystals were then dried under vacuum (1.83 g, 82% yield): 1H

NMR (400 MHz, (CD3)2SO) δ 10.94 (s, broad, 1H), 7.83 (d, J = 8.38 Hz, 2H), 6.92 (d, J = 2.14

Hz, 2H), 6.72 (dd, J1 = 8.38 Hz, J2 = 2.14 Hz, 2H); 13C NMR (100 MHz, (CD3)2SO) δ 157.5,

141.0, 119.9, 116.4, 107.3, 94.6, 55.2; HRMS (ESI) [M + H] calcd for C14H14NO2 228.1019,

found 228.1011

Compound 6-14a. Iodobenzene (1.241 g, 6.08 mmol), copper(I) iodide (77 mg, 0.404 mmol),

compound 6-13a (0.800 g, 4.06 mmol), L-proline (94 mg, 0.816 mmol), and potassium carbonate

(1.12 g, 8.10 mmol) were added to a stirred round-bottom flask. DMSO (7.00 mL) was added to

the flask, and the solution was sparged with argon gas for 30 min. The solution was heated to

180 °C for 1 h. After the solution was cool, toluene was added (25.0 mL), and any solids were

removed by filtration. The light brown liquor was concentrated under high vacuum to completely

remove any solvent and excess iodobenzene. The resulting solids were taken up in toluene and

passed through a short plug of silica, eluting with toluene. Solvent was removed under vacuum,

resulting in large flakey white crystals. These crystals were recrystallized from boiling heptanes

and dried under vacuum. 794 mg was isolated (72% yield): 1H NMR (400 MHz, CDCl3) δ 8.04

182

182

(dt, J1 = 7.83 Hz, J2 = 1.17 Hz, 1H), 8.00 (d, J = 8.22 Hz, 1H), 7.64−7.53 (m, 4H), 7.47 (m, 1H),

7.36−7.28 (m, 2H), 7.25 (m, 1H), 6.88 (td, J1 = 10.96 Hz, J2 = 2.35 Hz, 2H), 3.83 (s, 3H); 13C

NMR (100 MHz, CDCl3) δ 159.1, 142.2, 141.1, 137.7, 129.90, 127.5, 127.1, 124.6, 123.5, 121.0,

112.0, 119.4, 117.2, 109.5, 108.5, 94.0, 55.6; HRMS (ESI) [M + H] calcd for C19H16NO

274.1226, found 274.1227.

Compound 6-14b. Same general procedure as for compound 6-14a was used. Iodobenzene (1.0

g, 4.90 mmol), compound 13b (0.750 g, 3.27 mmol), copper(I) iodide (63 mg, 0.331 mmol), L-

proline (76 mg, 0.660 mmol), potassium carbonate (912 mg, 6.60 mmol), and 7.00 mL of DMSO

were reacted for 2 h. Product recrystallized from heptanes to yield white flakes. 821 mg of

product was isolated (82% yield): 1H NMR (400 MHz, CDCl3) δ 7.89 (d, J = 8.57 Hz, 2H), 7.61

(t, J = 7.41 Hz, 2H), 7.54 (d, J = 8.38 Hz, 2H), 7.47 (t, J = 7.41, 1H), 6.87 (dd, J1 = 8.38 Hz, J2 =

2.34 Hz, 2H), 6.82 (d, J = 2.34 Hz, 2H), 3.82 (s, 6H); 13C NMR (100 MHz, CDCl3) δ 158.3,

142.3, 137.6, 130.0, 127.5, 127.1, 120.1, 117.4, 108.2, 94.3, 55.7; HRMS (ESI) [M + H] calcd

C20H18NO2 304.1332, found 304.1340.

Compound 6-14c. Same general procedure as for compound 6-14a was used. 3-Iodoanisole

(10.92 g, 46.6 mmol), carbazole, (6.00 g, 35.9 mmol), copper(I) iodide (684 mg, 3.59 mmol), L-

proline (827 mg, 7.18 mmol), potassium carbonate (9.92 g, 71.8 mmol), and 60.0 mL of DMSO

were reacted for 3 h. Product recrystallized from heptanes and isolated as white flakes (7.75 g,

79% yield): 1H NMR (400 MHz, CDCl3) δ 8.14 (d, J = 7.75 Hz, 2H), 7.50 (t, J = 8.09 Hz, 1H),

7.45 (d, J = 7.99 Hz, 2H), 7.41 (t, d, J1 = 6.63 Hz, J2 = 1.36 Hz, 2H), 7.28 (ddd, J1 = 7.75 Hz, J2

= 6.63 Hz, J3 = 1.36 Hz, 2H), 7.16 (dq, J1 = 7.80 Hz, J2 = 0.97 Hz, 1H), 7.11 (t, J = 2.24 Hz,

1H), 7.01 (ddd, J1 = 8.38 Hz, J2 = 2.53 Hz, J3 = 0.97 Hz, 1H), 3.86 (s, 3H); 13C NMR (100

MHz, CDCl3) δ 160.8, 140.8, 138.8, 130.5, 125.9, 123.3, 120.3, 119.9, 119.3, 113.2, 112.6,

109.9, 55.5; HRMS (ESI) [M + H] calcd for C19H15NO 274.1226, found 274.1220.

Compound 6-14d. Same general procedure as for compound 6-14a was used. Carbazole (1.00 g,

5.98 mmol), 4-bromoanisole (1.34 g, 7.18 mmol), copper(I) iodide (114 mg, 0.599 mmol), L-

proline (138 mg, 1.20 mmol), potassium carbonate (1.653 g, 12.0 mmol), and DMSO (10.0 mL)

were added to a round-bottom flask. This solution was sparged with argon gas for 30 min and

183

183

then heated to 180 °C for 16 h. The reaction was diluted with toluene, and the solids were filtered

out. The mother liquor was passed through a plug of silica and then concentrated under high

vacuum and heat to remove the solvents, residual 4-bromoanisole and residual carbazole. The

resulting solids were recrystallized twice from boiling heptanes to yield long thin white needles.

685 mg were collected (42% yield): 1H NMR (400 MHz, CDCl3) δ 8.14 (d, J1 = 7.70 Hz 2H),

7.45 (d, 8.96 Hz, 2H), 7.40 (m, 2H), 7.32 (d, J = 8.38 Hz, 2H), 7.27 (t, J = 7.02 Hz, 2H), 7.11 (d,

J = 8.96 Hz, 2H), 3.92 (s, 3H); 13C NMR (100 MHz, CDCl3) δ 158.8, 141.4, 130.3, 128.6, 125.8,

123.1, 120.2, 119.6, 115.0, 109.7, 55.6; HRMS (ESI) [M+] calcd for C19H16NO 273.1148, found

273.1146.

Compound 6-15a. Using the general P−R procedure, tris(pentafluorophenyl)borane (10 mg,

0.0195 mmol), compound 6-14a (0.500 g, 1.82 mmol), and anhydrous toluene (5.00 mL) were

added to a open stirred vessel. 1,1,1,3,5,5,5-Heptamethyltrisiloxane (1.623 g, 7.32 mmol) was

added dropwise to this solution and allowed to stir for 20 min. Upon evaporation, a colorless oil

was obtained. This oil was diluted with hexanes and loaded onto a short plug of silica gel. This

plug was washed with hexanes (100 mL) and then washed with toluene. The toluene fraction was

collected and concentrated to yield a colorless oil (834 mg, 95% yield): 1H NMR (400 MHz,

C6D6) δ 8.00 (m, 1H), 7.93 (d, J = 8.61 Hz, 1H), 7.28−7.10 (m, integration obscured by solvent

peak), 7.04 (t, J = 7.43 Hz, 1H), 0.30 (s, 3H), 0.12 (s, 18H); 13C NMR (100 MHz, C6D6) δ 154.0,

142.8, 141.9, 138.2, 123.0, 127.5, 127.5, 125.2, 124.2, 121.4, 120.5, 120.1, 118.7, 113.8, 110.0,

101.0, 1.7, −3.1; HRMS (ESI) [M + H] calcd for C25H34NO3Si3 480.1841, found 480.1863.

Compound 6-15b. Using the general P−R procedure, compound 6-14b (0.500 g, 1.65 mmol),

tris(pentafluorophenyl)borane (9 mg, 0.176 mmol), 1,1,1,3,5,5,5-heptamethyltrisiloxane (2.90 g,

13.2 mmol), and anhydrous toluene (5.00 mL) were reacted. Product was purified by loading

onto a silica plug, washing first with hexanes and then washing with toluene. Product isolated as

a clear oil (1.00 g, 85% yield): 1H NMR (400 MHz, C6D6) δ 7.85 (d, J = 8.22 Hz, 2H), 7.21 (d, J

= 8.22 Hz, 2H), 7.19−7.11 (m, integration obscured by solvent), 7.04 (t, J = 7.43 Hz), 0.30 (s,

6H), 0.12 (s, 36H); 13C NMR (100 MHz, C6D6) δ 153.2, 143.1, 138.2, 130.1, 127.6, 127.6, 120.7,

118.9, 113.7, 101.1, 1.7, −3.1; HRMS (ESI) [M+] calcd for C32H53NO6Si6 715.2488, found

715.2486.

184

184

Compound 6-15c. Using the general P−R procedure, compound 6-14c (4.00 g, 14.6 mmol),

tris(pentafluorophenyl)borane (75 mg, 0.146 mmol), 1,1,1,3,5,5,5-heptamethyltrisiloxane (6.51

g, 29.2 mmol), and toluene (40.0 mL) were reacted. Product was purified by loading onto a silica

plug, washing first with hexanes and then washing with toluene. Product was isolated as a clear

oil (6.04 g, 86% yield): 1H NMR (400 MHz, CDCl3) δ 8.17 (d, J = 7.79 Hz, 2H), 7.50−7.40 (m,

5H), 7.00 (t, J = 7.00 Hz, 2H), 7.21 (dq, J1 = 7.80 Hz, J2 = 1.00 Hz, 1H), 7.17 (t, J = 1.95 Hz,

1H), 7.06 (ddd, J1 = 7.60 Hz, J2 = 2.34 Hz, J3 = 0.80 Hz,1H); 13C NMR (100 MHz, CDCl3) δ

155.5, 140.8, 138.6, 130.3, 125.9, 123.3, 120.3, 120.2, 119.8, 119.0, 118.7, 109.9, 1.6, −3.3;

HRMS (ESI) [M+] calcd for C25H33NO3Si3 479.1768, found 479.1755.

Compound 6-15d. Using the general P−R procedure, compound 6-14d (0.500 g, 1.83 mmol),

tris(pentafluorophenyl)borane (10 mg, 0.195 mmol), 1,1,1,3,5,5,5-heptamethyltrisiloxane (0.814

g, 3.66 mmol), and anhydrous toluene (5.00 mL) were reacted. Product was purified by loading

onto a silica plug, washing first with hexanes and then washing with toluene. Product was

isolated as a clear oil which crystallized into soft white needles after 2 months of storage

(728 mg, 83% yield): 1H NMR (400 MHz, CDCl3) δ 8.16 (d, J = 7.43 Hz, 2H), 7.45−7.38 (m,

4H), 7.34 (d, J = 7.82, 2H), 7.27 (t, J = 7.43 Hz, integration obscured by solvent peak), 7.14 (d, J

= 8.61 Hz, 2H), 0.28 (s, 3H), 0.15 (s, 18H); 13C NMR (100 MHz, CDCl3) δ 153.6, 141.3, 131.1,

128.4, 125.81, 123.1, 121.1, 120.2, 119.6, 109.7, 1.6, −3.2; HRMS (ESI) [M+] calcd for

C25H33NO3Si3 479.1768, found 479.1776.

12.4.2 Supplemental Information of Merit

Table S6-1: Lambda Max Comparison between Reference and Our Data (DCM vs Toluene)

Lambda Max for 1 (nm) Lambda Max for 4d (nm)

This Work (Toluene) 1056 1666

Reference Work (DCM) 1049 1572

Table S6-2: Estimated amount of oxidized arylamine from UV-VIS-NIR data.

185

185

Equivalents BCF Percent of Arylamine Oxidized Percent of BCF Reduced*

6-1 6-4d 6-1 6-4d

0.01 0.20 0.03 20.38 2.84

0.25 1.06 0.41 4.23 1.62

0.5 2.20 0.86 4.40 1.73

1.0 5.27 1.92 5.27 1.92

* Calculated assuming a 1:1 redox reaction with the arylamine

tris(pentafluorphenyl)borane (δ: -135.73, -154.66, -163.15

)

In Toluene TFA STd.esp

-135 -140 -145 -150 -155 -160 -165Chemical Shift (ppm)

0.0005

0.0010

0.0015

0.0020

0.0025

0.0030

0.0035

0.0040

0.0045

0.0050

0.0055

0.0060

0.0065

0.0070

Norm

aliz

ed Inte

nsity

Figure S6-1: 19F NMR of tris(pentafluorophenyl)borane in Toluene-d8.

NMR Spectra of 6-1 with 1mol% BCF 19F NMR, 25 °C

B

F

F

F

F

F

F

F

F

F

F

F

F

F

F

F

186

186

rxn 11


0

0.00005

0.00010

0.00015

Norm

aliz

ed I

nte

nsity

Figure S6-2: 19F NMR at 25 °C of compound 6-1 with 1 mol% tris(pentafluorophenyl)borane

in toluene-d8

1H NMR, 25 0C

proton

8 7 6 5 4 3 2 1 0Chemical Shift (ppm)

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.10

0.11

0.12

0.13

0.14

0.15

0.16

Norm

aliz

ed Inte

nsity

0.4

0

Figure S6-3: 1H NMR at 25 °C of compound 6-1 with 1 mol% tris(pentafluorophenyl)borane in toluene-

d8

187

187

NMR Spectra of 6-4d with 1mol% BCF 19F NMR, 25 0C

f19

-135 -140 -145 -150 -155 -160 -165 -170Chemical Shift (ppm)

-0.00005

0

0.00005

0.00010N

orm

aliz

ed

In

ten

sity

Figure S6-4: 19F NMR at 25 °C of compound 6-4d with 1 mol% tris(pentafluorophenyl)borane in

toluene-d8

188

188

1H NMR, 25 0C proton

8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5Chemical Shift (ppm)

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.10

0.11

0.12

0.13

0.14

0.15

0.16

Norm

aliz

ed Inte

nsity

0.4

0

Figure S6-5: 1H NMR at 25 °C of compound 6-4d with 1 mol% tris(pentafluorophenyl)borane in

toluene-d8

Compound 6-1 and BCF at 1 to 1 molar mixture 19F NMR at 250C in C6D6:

19f rt


0

0.0005

0.0010

0.0015

0.0020

0.0025

No

rma

lize

d I

nte

nsity

-16

4.1

4

-15

7.1

5

-13

5.7

0-1

35

.10

Figure S6-6: 19F NMR at 25 °C of compound 6-1 with 1 mol% tris(pentafluorophenyl)borane

in benzene-d6

189

189

Compound 6-4d and BCF at 1 to 1 molar mixture 19 f benzene.esp


0

0.0005

0.0010

0.0015

No

rma

lize

d In

ten

sity

-16

4.0

4

-15

9.9

1

-15

6.6

1

-13

5.6

8-1

35

.25

Figure S6-7: 19F NMR at 25 °C of compound 6-4d with 1 mol tris(pentafluorophenyl)borane in

benzene-d6



Time of flight transients were acquired by applying a specific electric field across the two

electrodes with a positive potential on the electrode with the charge generation layer (CGL).

Photogeneration was achieved by exposing the CGL to a 10 ns laser pulse through the device

from a dye laser. The device was terminated at resistor of value R and the voltage over time was

recorded on an oscilloscope. Before each test, the capacitance of the cells was measured by three

terminal measurement and a value of R was chosen to ensure that the transit time (tt) was less

than 10 times the RC time constant. Transit times were obtained by fitting moving average lines

to the flat regions of the voltage trace in order to obtain an intersection where the voltage begins

to rapidly drop. The mobility value was calculated using:

190

190

(1)

where µ is the charge carrier mobility, L is the thickness of the device, E is the electric applied

across the device and tt is the transit time calculated as described above. Variable temperature

measurements were achieved by placing the assembled cell into an oven equipped with a glass

window. For each temperature setting, the cell was allowed to equilibrate for 30 minutes before

measurements.

The solid samples were prepared first by mixing 250 mg of 2TIPS, 250 mg of polystyrene, and 7

g of spectro grade dichloromethane. The resulting lacquer was blade coated onto the CGL and

dried at 120 °C for 30 minutes. Film thicknesses were measured using a contact profilometer.

The cell was completed by sandwiching the coated CGL and a counter electrode together at a

pressure greater than 1 MPa.

The polystyrene used was obtained as commercial beads with Mw ~ 350,000 and Mn ~170,000

and a Tg of 95 °C. Prior to use, the beads were dissolved into methylene chloride and precipitated

into methanol followed by drying.

The neat liquid samples were prepared by first modifying the CGL with a 275 nm layer of N,N`-

diphenyl-N,N`-bis(3-methylphenyl)-(1,1`-biphenyl)-4,4`-diamine or TPD by physical vapour

deposition. The layer was deposited at a base pressure of <1 x10-5 Torr at a rate of 10 Å/s. The

cell was then assembled by placing two strips of polyimide film with a gap of ~0.5 cm between

them and a counter electrode strip across the entire layer. Care was taken to avoid direct physical

contact between the two electrodes. The cell was gently clipped together between two

microscope slides using office paper clips and the liquid organic semiconductor was introduced

at the edge of the open cell. Capillary filling of the cell was encouraged by heating the cell to

50°C for several minutes and the filled cells were allowed to cool before measurements began.

Synthesis of N-(2-ethylhexyl)carbazole: Carbazole (2g, 11.96 mmol), 2-ethylhexylbromide

(3.46 g, 17.94 mmol), sodium hydroxide (0.717 g, 17.94 mmol), tetrabutylammonium bromide

(100 mg, 0.120 mmol), and acetone (10 mL) were reacted at reflux for 48 hours. After cooling,

the reaction was filtered and the eluent was first purified by column chromatography (10:1

Hexanes/Toluene) over silica gel and then by vacuum distillation. The final product was isolated

191

191

as a clear oil (2.186 g, 7.822 mmol). 1H NMR (400 MHz, C6D6): δ 8.07 (dq, J1 = 7.80 Hz, J2

=0.65 Hz, 2H) 7.44 (tt, J1 = 7.80 Hz, J2 = 1.36 Hz, 2H), 7.29 (dt, J1 = 8.19 Hz, J2 = 0.78 Hz, 2H),

7.24 (tt, J1 = 8.19 Hz, J2 = 0.78 Hz, 2H), 3.77 (dd, J1 = 7.41 Hz, J2 = 2.14 Hz, 2H), 1.93 (m, 1H),

1.11 (m, 8H), 0.77 (t, J = 7.02 Hz, 3H), 0.68 (t, J = 7.02 Hz, 3H). 13C NMR (100 MHz, C6D6): δ

141.7, 126.2, 123.9, 121.2, 119.6, 109.6, 47.7, 39.8, 31.6, 29.3, 25.0, 23.6, 14.5, 11.3.

12.5.2 Supplemental Information for Chapter 7

Hole Mobilities of 2TIPS in Polystyrene (50 wt%) at Multiple Thicknesses

1.00E-06

1.00E-05

300 400 500 600 700 800 900

Ho

le M

ob

ilit

y (

cm

2v

-1s

-1)

E1/2 (V/cm)1/2

37 um Film 9 um Film 5 um Film

Figure S7-1: Extracted hole mobilities of 2TIPS in a polystyrene matrix (50 wt%) with different

film thicknesses.

Hole Mobilities in Neat 2TIPS using Multiple Thicknesses

192

192

1.00E-05

1.00E-04

150 200 250 300 350

Mo

bilit

y (

cm

2/V

s)

E1/2 (V/cm)1/2

125 um spacer

50 um spacer

75um

Figure S7-2: Extracted hole mobilities of neat 2TIPS using different sample thicknesses.

Example Transient of transport through N-(2-ethylhexyl)carbazole

1.50E-03

1.50E-02

1.50E-03 1.50E-02 1.50E-01

vo

ltag

e (

V)

time (s)

ttransit

Figure S7-3: An example of a time of flight transient obtained through neat N-(2-

ethylhexyl)carbazole.

193

193

Hole Mobility of N-(2-ethylhexyl)carbazole

1.00E-07

1.00E-06

1.00E-05

1.00E-04

260 270 280 290 300 310 320

Ho

le M

ob

ilit

y (

cm

2/V

s)

E1/2 (V/cm)1/2

Figure S7-4: Hole mobilities of neat N-(2-ethylhexyl)carbazole

194

194


65011501650215026503150

Tra

nsm

itta

nc

e

Wavenumber (cm-1)

QM4

Si-H

A

B

C

Figure S8-1: Infrared spectra of films A, B, and C prepared in a KBr matrix.

195

195

0

0.2

0.4

0.6

0.8

1

385 435 485 535 585

No

rma

lize

d P

L I

nte

ns

ity

Wavelength (nm)

B C E

Figure S8-2: Solid-state photoluminescence of films B, C, and E. Excitation wavelength 365

nm.

196

196

-0.1 0.1 0.3 0.5 0.7 0.9 1.1 1.3

Cu

rre

nt

(a.u

.)

Voltage (vs. Ag/AgCl) Fi

Figure S8-3: Electrochemistry of D on ITO in water with 0.1M NaCl as the supporting

electrolyte. (black) Cyclic voltammetry, (red) Differential pulse voltammetry.

197

197



General methods and procedures

Chemicals were purchased and used without further purification using standard laboratory

methods. NMR spectra were collected at 25 °C at a field strength of 400 MHz. Chemical shifts

are referenced to residual solvent signals. High resolution mass spectroscopy (HRMS) was

obtained using an AccuTOF mass spectrometer (JEOL USA Inc., Peabody, MA) with a DART-

SVP ion source (Ionsense Inc., Saugus, MA) using He gas. Electrochemistry was performed

using a standard three electrode setup. A platinum disk was used as the working electrode while

a platinum wire was used as a counter electrode. All data is corrected to the internal redox

standard of decamethylferrocene and numbers are referenced to Ag/AgCl. Thermogravimetric

analysis (TGA) was performed at ramp rate of 10 °C/min under N2.

Density functional theory calculations were performed using Spartan ’06 for windows. Geometry

optimizations were carried out using the Becke–Lee–Yang–Parr exchange correlation function

with a 6-31G(D) basis set. The OLEDs were fabricated in a cluster tool (Kurt J. Lesker

LUMINOS) under a base pressure of <10-8 Torr on pre-patterned indium tin oxide (ITO) coated

glass with a thickness of 1.1 mm. Prior to loading, the ITO was cleaned using standard solvents.

Subsequently, a MoO3 layer was deposited on top to obtain a high work function and facilitate

hole injection into CBP. All organic layers were deposited in a dedicated chamber, whereas the

cathode, consisting of LiF(1 nm)/Al(100 nm), was evaporated in a separate chamber without

breaking vacuum. The active area of each device was 2 mm2 as measured by an optical

microscope. The EQE and power efficiency were measured using an integrating sphere with a

silicon photodiode with NIST traceable calibration. The electroluminance (EL) spectra were

measured using an Ocean Optics USB4000 spectrometer. All measurements were done in air

with 3 s dwell time between each data point.

Synthesis of 9-1 and 9-2

Synthesis of 9-1: Tetrachlorophthalonitrile (1.00 g, 3.76 mmol), catechol (1.03 g, 9.38 mmol),

potassium carbonate (1.30 g, 9.38 mmol), and dimethylformamide (20 mL) were heated to 100

°C under argon for 1 h. After cooling to room temperature, the solids were filtered and washed

198

198

with water (100 mL) then methanol (50 mL). The collected solids were dried under vacuum to

yield a fine white powder (1.20 g, 94% yield). 1H NMR (400 MHz, CDCl3): δ 7.08–7.01

(m). HRMS (DART) [M+NH4] calcd for C20H12N3O4 358.0828, found 358.0822. Elemental

Analysis calcd for (%) C20H8N2O4: C 70.59, H 2.37, N 8.23. Found: C 70.56, H 2.33, N 8.31.

Synthesis of 9-2: Tetrachlorophthalonitrile (3.00 g, 11.3 mmol), 2,3-dihydroxynaphthalene (4.52

g, 28.2 mmol), potassium carbonate (3.90 g, 28.2 mmol), and dimethylformamide (60 mL) were

heated to 100 °C under argon for 5 h. The cooled solids were collected by filtration and washed

with water (200 mL), methanol (200 mL), then dichloromethane (50 mL), in that order. The

resulting solids were dried under vacuum to yield a fine white powder (4.74 g, 95% yield). 1H

NMR (400 MHz, CDCl3): δ 7.74 (q, J = 3.52 Hz, 4H), 7.51 (d, J = 7.91 Hz, 4H), 7.45 (q, J =

3.22 Hz, 4H). HRMS (DART) [M+H] calcd for C28H13N2O4 441.0875, found 441.0876.

Elemental Analysis calcd for (%) C28H12N2O4: C 76.36, H 2.75, N 6.36. Found: C 75.98, H 3.00,

N 6.29.

References

1. I. Noviandri, K.N. Brown, D.S. Fleming, P.T. Gulyas, P.A. Lay, A.F. Masters, P.

Leonidas, J. Phys. Chem. B 103 (1999) 6713–6722.

2. A.D. Becke, Phys. Rev. A 38 (1988) 3098–3100.

199

199

12.8 Appendices References for Chapter 10

12.8.1 General Information

Starting materials and solvents were purchased from various companies and used without further

purification.

NMR Spectra were collected at 25 °C at a field-strength of 400 MHz. In order to resolve the

isomers of compound 2, NMR spectra at a field strength of 700 MHz was collected. All 1H and 13C spectra are referenced to residual solvent or TMS and chemical shifts are reported in parts

per million while coupling constants are reported in Hz. 19F and 11B NMR are referenced to BF3-

OEt2 which we arbitrarily assigned to 0 ppm for both nuclei. External standards were used with

each NMR experiment.

High resolution mass spectroscopy (HRMS) was acquired with either DART or ESI ionization

techniques. For those using DART, the spectra were acquired using an AccuTOF mass

spectrometer (JEOL USA Inc. Peabody, MA) with a DART-SVP ion source (Ionsense Inc.,

Saugus, MA) using He Gas.

Electrochemistry was performed in a solution of DCM with 0.1 M tetraammonium perchlorate as

a supporting electrolyte. A 1 mm platinum disc was used as a working electrode with a platinum

wire counter electrode and Ag/AgCl reference electrode. All cyclic voltammetry experiments

were run with an internal standard of decamethylferrocene at a scanning rate of 100 mV/s. All

half wave potentials are corrected to the published halfwave potential of decamethylferrocene (-

0.012 V vs. Ag/AgCl).2 All half wave potentials are reported relative to Ag/AgCl.

Density functional theory (DFT) calculations were implemented using Spartan ’06 for windows. Structures were geometry optimized using the Becke-Lee-Yang-Parr exchange correlation function3 with a 6-31G(D) basis set.

2 Noviandri, I.; Brown, K.N.; Fleming, D.S.; Gulyas, P.T.; Lay, P.A.; Masters, A.F.; Leonidas, P. J. Phys. Chem. B

1999, 103 (32), 6713-6722. 3 Becke, A.D. Phys. Rev. A 1988, 38 (6), 3098-3100

200

200

12.8.2 Synthetic Details and Compounds Characterization

Compound 10-2

Tetrachlorophthalonitrile (1.000 g, 3.76 mmol), 3,5-di-t-butylcatechol (1.755 g, 7.90 mmol),

potassium carbonate (1.091 g, 7.90 g), and N,N-dimethylformamide (20 mL) were heated to 100

°C for 6 hours under an atmosphere of argon gas. Upon cooling, water (20 mL) was added to the

solution to form a fine slurry. The solids were collected by filtration, washing with water (3 x 50

mL) and methanol (3 x 50 mL) resulting in the pure product as a fine white powder (1.927 g,

91% Yield). HRMS (DART) [M+H] cald for C36H41N2O4 565.30663, found 565.30659.

Compound 10-3a

Compound 10-1 (11.690 g, 34.4 mmol) was dissolved in 1,2-dichlorobenzene (300 mL) under an

inert atmosphere. Boron trichloride (81 mL of a 1M solution in heptanes, 81 mmol) was added to

this solution and the heptanes were distilled off and the mixture was refluxed for 2 hours. Upon

cooling, the dichlorobenzene was removed under vacuum and the resulting black solids were

continuously extracted with methanol (48 Hrs) then acetonitrile (24 Hrs) using a soxhlet

apparatus. The remaining solids were dried under vacuum resulting in a fine green/black powder

assumed to be the chloro substituted subphthalocyanine product (8.01 g, 66 % crude yield).

The crude -Cl substituted subphthalocyanine (500 mg, ~0.47 mmol), 4-t-butylphenol (352 mg,

2.34 mmol), and chlorobenzene (5 mL) were heated at reflux for 48 hours. Upon cooling, the

chlorobenzene was removed under vacuum and the resulting green solids were loaded onto a

plug of alumina (basic, standard activity) and extracted continuously with dichloromethane using

a Kaufman apparatus. The extracted, dark green liquor was then concentrated under vacuum

resulting in a dark green powder (319 mg, 58 % Yield). HRMS (ESI) [M+] calcd 1180.2506,

found 1180.2770. 1H (400 MHz, CD2Cl2): δ 7.21 (d, J = 7.8 Hz, 6 H), 6.96-6.79 (m, 20H), 5.47

(d, J = 8.6 Hz, 2H), 1.06 (s, 9H). 11B (CD2Cl2): δ -14.8

Compound 10-3b

Using the procedure for 10-3a: The crude Cl substituted subphthalocyanine (500 mg, ~0.47

mmol), pentafluorophenol (431 mg, 2.34 mmol), and chlorobenzene (5 mL) were heated at

reflux for 18 hours. The product was isolated as a dark green powder (207 mg, 36 % yield). MS

201

201

(ESI) [M+] calcd 1215.1, found 1215.1. MS signal insufficient for high resolution mass

spectroscopy. 1H (400 MHz, CD2Cl2): δ 7.29 (dd, J1 = 8.0 Hz, J2 = 1.4 Hz, 6 H), 7.16-7.00 (m,

20H). 19F (CD2Cl2, referenced to BF3-O(Et)2): δ -5.35 (d, J = 20.6 Hz, 2F), -10.3 (t, J = 20.6 Hz,

2F), -11.7 (t, J = 20.6 Hz, 1F). 11B (CD2Cl2, referenced to BF3-O(Et)2): δ -14.5

Compound 10-4a

10-2 (2.258 g, 4.00 mmol), phthalonitrile (256 mg, 2 mmol), 20 mL 1,2-dichlorobenzene were

stirred under an inert atmosphere. Boron trichloride (10 mmol, 10 mL of heptanes solution) was

added to the mixture and the heptanes were distilled off were reacted for 2 hours under and inert

atmosphere. Upon cooling, the solution was dried under vacuum and the blue solids were

continuously extracted with methanol for 18 hours. The remaining solids resembled a dark blue

powder (1.554 g, 90% crude yield).

The crude above product (1.000 g, ~1.15 mmol), pentafluorophenol (1.000 g, 5.43 mmol), and

chlorobenzene (10 mL) were refluxed under an inert atmosphere for 18 hours. After removal of

the chlorobenzene under vacuum, the remaining blue solids were purified by column

chromatography over silica gel eluting with 3/2 hexanes/toluene. A fraction containing a single

blue spot was isolated (258 mg, 22% Yield) and shown to be a single product on low molecular

weight GPC. HRMS (DART) [M+H] calcd for C58H48BF5N6O5 1015.3778, found 1015.3718. 1H

(400 MHz, CDCl3): δ 8.80 (dm, J1 = 20.7 Hz, 4H), 7.91 (q, J = 3.1 Hz, 4H), 7.15 (d, J = 2.3 Hz,

1.7H), 7.07-7.05 (m, 2H), 6.95 (d, J = 2.3 Hz, 0.3H), 1.94 (s, 15.1H), 1.46 (s, 2.9), 1.36 (s, 15.1),

1.29 (s, 2.9). 11B (CDCl3, referenced to BF3-O(Et)2): δ -14.6. 19F (CDCl3, referenced to BF3-

O(Et)2): δ -5.3 (d, J = 21.4 Hz, 2F), -11.1 (t, J = 21.4 Hz, 2F), -12.0 (t, J = 21.4 Hz, 1F).

Compound 10-4b

Compound 10-2 (4.00 mmol, 2.258 g) tetrafluorophthalonitrile (4.00 mmol, 800 mg), 20 mL 1,2-

dichlorobenzene, and BCl3 (10.0 mmol, 10 mL of a 1M solution in heptanes) were mixed

together under argon. The heptanes were then distilled off and the reagents heated at reflux for 2

hours. After cooling the solvent was removed under vacuum and the dark green/blue solids were

continuously extracted with methanol using a soxhlet apparatus for 16 hours. The extracted

solids were dried and the crude, -Cl substituted product collected (1.511 g, 75% crude yield).

202

202

The above crude product (800 mg, ~0.790 mmol), pentafluorophenol (800 mg, 4.35 mmol), and

chlorobenzene (8 mL) were refluxed under argon for 10 hours. After removal of the solvent, the

solids were purified by flash chromatography over silicon eluting with 5/1 hexanes to toluene.

Two green fractions with identical retention times by lmw GPC and UV-VIS absorbance spectra

were isolated (fraction 1: 113 mg, 12% Yield, fraction 2: 107 mg, 12% yield). Unfortunately, the

second fraction quickly degraded upon heating under vacuum and only the first fraction was

characterized. HRMS (DART) [M+] calcd for C58H40BF13N6O5 1155.3060, found 1158.2968. 1H

(400 MHz, CDCl3): δ 7.18 (2.3 Hz, 2H), 7.08 (2,3 Hz, 2H), 1.83 (s, 18H), 1.37 (s, 18H). 11B

(CDCl3, referenced to BF3-O(Et)2): δ -14.8. 19F (CDCl3, referenced to BF3-O(Et)2): δ 17.6 (t, J =

18.3 Hz, 2F), 14.9 (t, J = 18.3 Hz, 2F), 4.3 (t, 18.3 Hz, 2F), 3.4 (t, J = 18.3 Hz, 2F), -5.5 (2, J =

22.9 Hz, 2F), -9.9 (t, J =20.6 Hz, 2F), -10.3 (t, J = 22.9 Hz, 1F)

Compound 10-4c

Compound 10-2 (4.00 mmol, 2.258 g) tetrachlorophthalonitrile (4.00 mmol, 1.063 g), 20 mL 1,2-

dichlorobenzene, and BCl3 (10 mmol, 10 mL of a 1M solution in heptanes) were mixed together

under argon. The heptanes were then distilled off and the reagents heated at reflux for 2 hours.

After cooling, the solvent was removed under vacuum and the dark green/blue solids were

continuously extracted with methanol using a soxhlet apparatus for 16 hours. After extraction,

the solids were dried and the crude, -Cl substituted product collected (652 g, 29% crude yield).

The above crude product (500 mg, ~0.438 mmol), pentafluorophenol (500 mg, 2.72 mmol), and

chlorobenzene (5 mL) were refluxed under argon for 18 hours. After drying, the resulting dark

blue/green powder was purified twice successively over silica gel eluting with a gradient from

10/1 to 3/1 Hexanes/Toluene. HRMS (DART) [M+H] calcd for C58H40BCl8F5N6O5 1287.0660,

found 1287.0719. 1H (400 MHz, CDCl3): δ 7.22-7.19 (m, 1.9H), 7.14 (d, J = 2.3 Hz, 0.8H), 7.10

(d, J = 2.3 Hz, 0.8H), 6.95 (d, J = 2.3 Hz, 0.5H), 1.83 (s, 12H), 1.55 (s, 5.0H), 1.40-1.38 (m,

19H). 11B (43 MHz, CDCl3, referenced to BF3-O(Et)2): δ -14.7. 19F (376 MHz, CDCl3, referenced

to BF3-O(Et)2): δ -5.4 (m, 2F), -9.9 (m, 2F), -10.6 (m, 1F).

12.8.3 NMR Study of Phthalonitrile 10-2

Due to the nature of the substitution reaction, the following three isomers are expected:

203

203

Figure S10-1: Structures of three expected isomers for compound 10-2

HPLC analysis was not able to isolate and quantify the ratios of each isomer. 1H NMR at 400

MHz showed some separation of the aromatic protons for each isomer but the resolution was

insufficient. To resolve these peaks, 1H NMR and gCOSY (1H-1H) analysis was performed at

700 MHz (Figures S2-4). The enhanced resolution afforded by the higher field strength allowed

enough resolution for accurate integration of a number of peaks. Through ring coupling detected

by gCOSY and knowledge of the integration values in the 1D spectrum allowed for

quantification of each isomer.

204

204

cdcl3 proton 700 mhz

7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0Chemical Shift (ppm)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

No

rma

lize

d In

ten

sity

0.0

0

7.2

6

Figure S10-2: 1H NMR spectrum at 700 MHz in CDCl3 of compound 10-2. Inset: Close up of

alkyl region of spectrum.


7.05 7.00 6.95 6.90 6.85 6.80Chemical Shift (ppm)

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.10

0.11

0.12

0.13

Norm

alized Inte

nsity

Figure S10-3: Zoom of aromatic region of 1H NMR spectrum at 700 MHz in CDCl3 of

compound 10-2. Coloured lines show coupled spin systems from gCOSY experiment.


1.50 1.45 1.40 1.35 1.30 1.25Chemical Shift (ppm)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

Norm

aliz

ed I

nte

nsity

205

205

7.05 7.00 6.95 6.90 6.85 6.80 6.75F2 Chemical Shift (ppm)

6.75

6.80

6.85

6.90

6.95

7.00

7.05

F1

Ch

em

ica

l S

hift

(pp

m)

Figure S10-4: gCOSY (1H, 1H) spectrum at 700 MHz in CDCl3 of compound 10-2.

1H NMR (700 MHz, CDCl3): δ 7.07-7.05 (m, 4.1 H ), 7.05 (d, J = 2.4 Hz, 1.6 H), 7.03 (d, J = 2.4

Hz, 1 H), 6.96 (d, J = 2.4 Hz, 2.6 H), 6.93 (d, J = 2.4 Hz, 1.6 H), 6.91 (d, J = 2.4 Hz, 1 H), 6.79

(d, J = 2.4 Hz, 1.6 H), 1.48-1.45 (m, 61 H), 1.31-1.29 (m, 61 H).

12.8.4 UV-Vis and PL Plots

0

0.2

0.4

0.6

0.8

1

0

20000

40000

60000

80000

100000

300 400 500 600 700 800 900

Ph

oto

lum

ine

sce

nt

Inte

ns

ity

(a.u

.)

Ex

tin

cti

on

Co

eff

icie

nt

(L-1

mo

l-1c

m-1

)

Wavelength (nm)

206

206

Figure S10-5: UV-VIS absorbance spectrum of compound 10-3a in THF (left axis) and

photoluminescence emission spectrum of 10-3a in THF at an excitation wavelength of 650 nm.

0

0.2

0.4

0.6

0.8

1

0

10000

20000

30000

40000

50000

60000

70000

80000

90000

300 400 500 600 700 800 900

Ph

oto

lum

ine

sce

nt

Inte

ns

ity

(a.u

.)

Ex

tin

cti

on

Co

eff

icie

nt

(L-1

mo

l-1c

m-1

)

Wavelength (nm)

Figure S10-6: UV-VIS absorbance spectrum of compound 10-3b in THF (left axis) and

photoluminescence emission spectrum of 10-3b in THF at an excitation wavelength of 650 nm.

0

0.2

0.4

0.6

0.8

1

0

10000

20000

30000

40000

50000

60000

300 400 500 600 700 800

Ph

oto

lum

ine

sce

nt

Inte

ns

ity

(a.u

.)

Ex

tin

cti

on

Co

eff

icie

nt

(L-1

mo

l-1c

m-1

)

Wavelength (nm)

Figure S10-7: UV-VIS absorbance spectrum of compound 10-4a in THF (left axis) and

photoluminescence emission spectrum of 10-4a in THF at an excitation wavelength of 601 nm.

207

207

0

0.2

0.4

0.6

0.8

1

0

5000

10000

15000

20000

25000

30000

35000

40000

45000

300 400 500 600 700 800 900

Ph

oto

lum

ine

sce

nt

Inte

ns

ity

(a.u

.)

Ex

tin

cti

on

Co

eff

icie

nt

(L-1

mo

l-1c

m-1

)

Wavelength (nm)

Figure S10-8: UV-VIS absorbance spectrum of compound 10-4b in THF (left axis) and

photoluminescence emission spectrum of 10-4b in THF at an excitation wavelength of 621 nm.

0

0.2

0.4

0.6

0.8

1

0

10000

20000

30000

40000

50000

300 400 500 600 700 800 900

Ph

oto

lum

ine

sce

nt

Inte

nsit

y (

a.u

.)

Ex

tin

cti

on

Co

eff

icie

nt

(L-1

mo

l-1cm

-1)

Wavelength (nm)

Figure S10-9: UV-VIS absorbance spectrum of compound 10-4c in THF (left axis) and

photoluminescence emission spectrum of 10-4c in THF at an excitation wavelength of 626 nm.

208

208

0

10000

20000

30000

40000

50000

60000

70000

80000

90000

300 350 400 450 500 550 600 650

Ex

tin

cti

on

Co

eff

icie

nt

(L-1

mo

l-1c

m-1

)

Wavelength (nm)

Figure S10-10: UV-VIS absorbance spectrum of compound F5-BsubPc in THF.

209

209

12.9 DFT Calculated Molecular Orbitals

N

N

N

NN N

B

O

O

OO

O

O O

O

OO

O

O

O

HOMO LUMO

Figure S10-11: Geometry optimized DFT structures for compound 10-3a showing HOMO and

LUMO distributions.

210

210

HOMO LUMO

Figure S10-12: Geometry optimized DFT structures for compound 10-3b showing HOMO and

LUMO distributions.

211

211

HOMO LUMO

Figure S10-13: Geometry optimized DFT structures for one isomer of compound 10-4a with

HOMO and LUMO distributions.

212

212

HOMO LUMO

Figure S10-14: Geometry optimized DFT structures for one isomer of compound 10-4b with


213

213

HOMO LUMO

Figure S10-15: Geometry optimized DFT structures for one isomer of compound 10-4c with


214

214

12.10 Cyclic Voltammetry

-2 -1.5 -1 -0.5 0 0.5 1 1.5

Voltage (V)

E1/2ox

E1/2red1

E1/2reference

Figure S10-16: Cyclic voltammogram of compound 10-3a in DCM with 0.1M


-1.5 -1 -0.5 0 0.5 1 1.5

Voltage (V)

E1/2ox

E1/2red1

E1/2reference

Figure S10-17: Cyclic voltammogram of compound 10-3b in DCM with 0.1M


215

215

-1.5 -1 -0.5 0 0.5 1 1.5

Voltage (V)

E1/2ox

E1/2red

E1/2reference

Figure S10-18: Cyclic voltammogram of compound 10-4a in DCM with 0.1M


-2 -1.5 -1 -0.5 0 0.5 1 1.5 2

Voltage (V)

E1/2ox

E1/2red1

E1/2red2

E1/2reference

Figure S10-19: Cyclic voltammogram of compound 10-4b in DCM with 0.1M


216

216

-2 -1.5 -1 -0.5 0 0.5 1 1.5 2

Voltage (V)

E1/2ox

E1/2red1

E1/2red2

E1/2reference

Figure S10-20: Cyclic voltammogram of compound 10-4c in DCM with 0.1M


Documents

Controlling the Physical Properties of Organic ... · Controlling the Physical Properties of Organic Semiconductors through Siloxane Chemistry and ... Physical Properties of Organic