Cubane and Triptycene As Scaffolds in the Synthesis of Porphyrin Arrays

Home , Cubane, Favorskii rearrangement, Prelog strain

Cubane and Triptycene as Scaffolds in the Synthesis of Porphyrin Arrays

Submitted by

Gemma M. Locke

B.A. (Mod.) Medicinal Chemistry

Trinity College Dublin, Ireland

A thesis submitted to the University of Dublin, Trinity College for the degree

Doctor of Philosophy

Under the Supervision of Prof. Dr. Mathias O. Senge

University of Dublin, Trinity College September 2020

Declaration

I declare that this thesis has not been submitted as an exercise for a degree at this or any other university and it is entirely my own work.

I agree to deposit this thesis in the University’s open access institutional repository or allow the Library to do so on my behalf, subject to Irish Copyright Legislation and Trinity College Library conditions of use and acknowledgement.

I consent to the examiner retaining a copy of the thesis beyond the examining period, should they so wish.

Furthermore, unpublished and/or published work of others, is duly acknowledged in the text wherever included.

Signed: ______

March 2020 Trinity College Dublin

Summary

The primary aim of this research was to synthesise multichromophoric arrays that are linked through rigid isolating units with the capacity to arrange the chromophores in a linear and fixed orientation. The electronically isolated multichromophoric systems could then ultimately be tested in electron transfer studies for their applicability as photosynthesis mimics.

Initially, 1,4-diethynylcubane was employed as the rigid isolating scaffold and one to two porphyrins were reacted with it in order to obtain the coupled product(s). Pd-catalysed Sonogashira cross-coupling reactions were used to try and achieve these bisporphyrin complexes. After extensive optimisation attempts, of both copper and copper-free Sonogashira reaction conditions, the desired alkynylcubane-linked bisporphyrin was detected, but unfortunately not isolated. The instability of the alkynylcubane-linked systems prevented efficient purification and inhibited further reactions, and while the cubane porphyrin dimer was identified, its insolubility prevented further analysis. Overall, five new alkynylcubane-linked monoporphyrin compounds were synthesised, albeit in very low yields. 1,4-Bis(phenylethynyl)cubane was obtained suggesting that coupling with the large porphyrin systems may have been the issue rather than solely the instability of 1,4- diethynylcubane.

Following on from the work with 1,4-diethynylcubane, 9,10-diethynyltriptycene was investigated for its uses as a rigid isolating unit. Initially, Sonogashira cross-coupling conditions that were developed from the previous work with cubane, were utilised with various porphyrins and boron dipyrromethenes (BODIPYs) and worked seamlessly. Although there are previous examples of triptycene porphyrin complexes, this was the first example of a linear porphyrin dimer that was connected to triptycene through the bridgehead carbons. Symmetric and unsymmetric examples of these complexes were obtained. The unsymmetric porphyrin dimer proved more difficult to access, owing to the unreliability of the Diels–Alder reaction of benzyne with 9-[(triisopropylsilane)ethynyl]-10- [(trimethylsilane)ethynyl]-anthracene to access the triptycene starting material. In a slightly different vein, a series of porphyrin-cubane/bicyclo[1.1.1]pentane-porphyrin arrays were synthesised via amide coupling reactions. The utilisation of semi-rigid amide bonds to attach the porphyrin skeletons to a rigid scaffold introduced a controlled conformational flexibility into porphyrin dyad/s. This allowed significant modulation of the photophysical properties in the porphyrin dyad/s through the coordination of transition metal(II) ions. These reversible photophysical changes suggest that the dimeric systems are acting like switchable porphyrin tweezers. The (1-[bis(dimethylamino)methylene]-1H- 1,2,3-triazolo[4,5-b]pyridinium

iii

3-oxid hexafluorophosphate) (HATU) amide coupling procedure employed was robust and versatile, allowing access to the very first porphyrin-cubane/ bicyclo[1.1.1]pentane- porphyrin arrays, representing the largest non-polymeric structures available for cubane/bicyclo[1.1.1]pentane derivatives. These reactions demonstrated considerable substrate scope, from utilisation of small phenyl moieties to large porphyrin rings, with varying lengths and different angles. Depending on the orientation of the substituents around the amide bond of the cubane/bicyclo[1.1.1]pentane units different intermolecular interactions were identified through single crystal X-ray analysis. Moreover, X-ray structural analysis revealed non-covalent interactions that are the first-of-their-kind, including a unique iodine-oxygen interaction between cubane molecules.

Lastly on a different note, a cationic dimer with a phenylene linker and an anionic porphyrin dimer with a conjugated butadiene linker were synthesised. The central aim of this work was to introduce water-soluble moieties into porphyrin dimers as there are very limited examples in the literature of porphyrin dimers with application in photodynamic therapy. It was hoped that by introducing cation/anionic moieties to a dimer that these water-soluble moieties would allow it to be suitable for biological application. The porphyrin dimer was successfully obtained, but tests to see its efficiency as a photodynamic therapy agent are still underway.

Publications

G. M. Locke, M. O. Senge, “Towards Electron Transfer Compounds with Rigid Resistor Units” ECS Trans. 2016, 16, 1–11. doi:10.1149/07216.0001ecst

S. S. R. Bernhard, G. M. Locke, S. Plunkett, A. Meindl, K. J. Flanagan, M. O. Senge, “Cubane Cross-Coupling and Cubane-Porphyrin Arrays” Chem. Eur. J. 2018, 24, 1026– 1030. doi:10.1002/chem.201704344

Front Cover: Cubane Cross‐Coupling and Cubane– Porphyrin Arrays (Chem. Eur. J. 2018, 24, 997).

doi:10.1002/chem.201705227

Invited review: G. M. Locke, S. S. R. Bernhard, M. O. Senge, “Nonconjugated Hydrocarbons as Rigid‐Linear Motifs: Isosteres for Material Sciences and Bioorganic and Medicinal Chemistry” Chem. Eur. J. 2019, 25, 4590–4647. doi:10.1002/chem.201804225

N. Grover‡, G. M. Locke‡, K. J. Flanagan, M. H. R. Beh, A. Thompson, M. O. Senge, “Bridging and Conformational Control of Porphyrin Units through Non-Traditional Rigid Scaffolds” submitted for publication.

Conference Abstracts

G. M. Locke, M. Roucan, S.Plunkett, M. O. Senge. “Functionalisation of cubane scaffolds- Towards the use of cubane as a rigid linker in energy transfer compounds” in the Centre for Synthesis and Chemical Biology (CSCB) Recent Advances in Synthesis and Chemical Biology XVI, (04.12.15), Dublin, Ireland. This poster won one of three prizes for the best poster award.

G. M. Locke, S. S. R. Bernhard, S. Plunkett and M. O. Senge “Advances in the Functionalisation of Cubane, for Use as a Scaffold in the Synthesis of Multiporphyrin Arrays” in the Tetrapyrrole Discussion Group Meeting (21.03.16–22.03.16), Liverpool John Moores University, Liverpool, England. Abstract book page 17.

G. M. Locke, S. S. R. Bernhard, S. Plunkett and M. O. Senge “Advances in the Functionalisation of Cubane, for Use as a Scaffold in the Synthesis of Multiporphyrin Arrays” in the ORCHEM 2016 conference (05.09.16–07.09.16), Weimar, Germany. Abstract-ID: 6243_15208.

G. M. Locke, S. S. R. Bernhard, S. Plunkett and M. O. Senge “Advances in the Functionalisation of Cubane, for Use as a Scaffold in the Synthesis of Multiporphyrin Arrays” in the CSCB conference (09.12.16) Trinity College Dublin, Dublin, Ireland. Abstract P023.

G. M. Locke, S. S. R. Bernhard, S. Plunkett, K. J. Flanagan and M. O. Senge “The Synthesis of Multiporphyrin Arrays with non-Heteroatom Linkers” in the GDCh Scientific Forum Chemistry 2017 (10.09.17–14.09.17), Berlin, Germany. Abstract-ID: 7210-14218.

G. M. Locke, S. S. R. Bernhard and M. O. Senge, “Rolling the Die for Dyes” in 16th Belgian Organic Synthesis Symposium, (08.07.18–13.07.18), Brussels, Belgium, Abstract book P152.

G. M. Locke, A. F. Syeda and M. O. Senge, “Triptycene and its use as a scaffold in the synthesis of multiporphyrin arrays” in Centre for Synthesis and Chemical Biology (CSCB) Recent Advances in Synthesis and Chemical Biology XVI, (07.12.18), Dublin, Ireland.

Acknowledgements

First of all, I would like to thank Prof. Dr. Mathias O. Senge for giving me the opportunity to pursue this research and offering me a position in his research group. Also for his encouragement, belief in my capabilities and supervision which encourages self-thought and the constant pursuit of new knowledge. I would also like to thank the Irish Research Council for their funding support.

A big thank you goes out to my valued colleagues Alina, Bhavya, Elisabeth, Harry, Ganapathi, Jess, Karolis, Keith, Marc, Marie, Mikhail, Nitika, Stefan, Susan, Piotr and Zoi who helped make our workplace a safe, fun and encouraging environment. A place where laughter was a part of everyday life and there was never a shortage of advice or people who wanted a coffee break. All the brave international souls must also be noted who were always up for an after work swim in the Irish sea, sometimes regardless of the weather. I want to additionally thank Keith for his X-ray crystallography work, Ganapathi and Alina for supplying some of the starting materials for my work and for everyone’s help proof-reading this thesis. I especially want to thank our two brilliant post-Docs Dr. Stefan Bernard and Dr. Nitika Grover (the referencing Queen) for their unwavering patience and consistent help in all things chemistry and for being a part of team cubane/BCP.

I am also indebted to Dr. Gary Hessman, Dr. Martin Feeney, Dr. John O’Brien and Dr. Manuel Rüther for their help in analytical questions and I would like to express my gratitude to everyone working behind the curtains at Trinity College Dublin.

Finally, I want to thank my family and friends, who have listened to my failures, cheered me up when I was down, consistently prayed for me and my chemistry and celebrated with me in my successes. I want to thank God who is my constant strength and encouragement, who gave me hope at times when I could have despaired and who has taught me what “His mercies are new every morning” truly means over the past four years.

vii

Table of Contents

Declaration...... ii Summary ...... iii Publications ...... v Conference Abstracts ...... v Acknowledgements ...... vii List of Abbreviations ...... xi Chapter 1. Introduction ...... 1 1.1. Nonconjugated Rigid Hydrocarbons ...... 1 1.2. Physical Properties ...... 2 1.2.1. Size and Dimension ...... 2 1.2.2. Bond Considerations, Electronic Communication, and Ring Strain ...... 2 1.2.3. Electronic Communication ...... 3 1.2.4. Ring Strain ...... 4 1.2.5. Bridgehead Cations, Anions and Radicals ...... 5 1.3. Cubane – a Rigid Synthetic Scaffold...... 9 1.3.1. The History of Cubane ...... 9 1.3.2. Properties and Applications of Cubane ...... 11 1.3.3. Reactivity and Manipulation of the Bridgehead Carbon Centre ...... 16 1.3.4. Cubane Chemistry ...... 18 1.4. Triptycene...... 22 1.4.1. Structure and Properties of Triptycene ...... 24 1.4.2. Applications...... 26 1.4.3. Triptycene Chemistry ...... 29 1.5. An Introduction to Porphyrins ...... 30 1.5.1. Structure of Porphyrins ...... 31 1.5.2. Photosynthetic Mimics ...... 33 1.5.3. Cubane and Triptycene as Potential Coupling Partners ...... 33 1.5.4. Merging Porphyrins with Cubanes and Triptycenes ...... 35 1.6. Applications of Rigid Linear Linkers in Electron Transfer Systems ...... 36 1.6.1. Small Organic Acceptor Donor Systems ...... 36 1.6.2. Porphyrins for Use in Electron Transfer Systems ...... 38 1.6.3. Non-linear Triptycene Porphyrin Arrays ...... 43 1.7. Pd-catalysed Cross-coupling Reactions ...... 45

viii

Objectives ...... 49 Chapter 2. Cubane Cross-coupling and Cubane–Porphyrin Arrays ...... 51 2.1. Background on the History of Cubane Functionalisation ...... 51 2.2. Method Development and Overview of Synthesised New Compounds ...... 52 2.3. Synthesis of Starting Materials for Sonogashira Cross-coupling reactions ...... 58 2.3.1. Synthesis of Dimethyl Cubane-1,4-dicarboxylate ...... 58 2.3.2. Porphyrin Synthesis...... 61 2.3.3. Synthesis of BODIPYs ...... 64 2.4. Cubane Cross-coupling Reactions via the Sonogashira Reaction ...... 66 2.4.1. Sonogashira Reaction with 1,4-Diethynylcubane ...... 66 2.4.2. Sonogashira Reactions with 1,4-Di[(trimethylsilyl)ethynyl]cubane ...... 70 2.4.3. Optimisation of the Sonogashira Reaction using Free Base Porphyrins ...... 71 2.4.4. Optimisation of the Sonogashira Reaction using Metalloporphyrins ...... 73 2.5. Conclusion and Outlook ...... 78 Chapter 3. Triptycene Cross-coupling and Cubane–Porphyrin Arrays ...... 80 3.1. Background on the History and Application of Triptycene as a Scaffold ...... 80 3.2. Synthesis of Functionalised Triptycenes ...... 82 3.3. Synthesis of Triptycene-linked Porphyrin Dimers ...... 89 3.4. Towards the Synthesis of Benzene-linked Porphyrin Dimers ...... 93 3.5. Single Crystal X-ray Structures ...... 98 3.6. Photophysical Studies of Triptycene Complexes ...... 102 3.7. Conclusion and Outlook ...... 108 Chapter 4. Bridging and Conformational Control of Porphyrin Units through Non- Traditional Rigid Scaffolds ...... 110 4.1. Background on the Application of Cubane/BCP amide linkers ...... 110 4.2.1. Porphyrin Synthesis...... 113 4.2.2. Cubane Synthesis ...... 117 4.2.3. BCP Synthesis – conducted by Dr. Nitika Grover ...... 124 4.3. Single Crystal X-ray Analysis ...... 127 4.4. Spectroscopy Studies ...... 133 4.5. Tweezer Studies with Metal Ions ...... 133 4.6. Conclusions and Outlook ...... 137 Chapter 5. Synthesis of Covalently linked Water-soluble Dimers for PDT Treatment. ... 138 5.1. Background on the Basic Principles of PDT Treatment ...... 138 5.2. Synthesis ...... 140 ix

5.2.1. Sulfonation of Phenyl Porphyrin Dimers ...... 140 5.2.2. Synthesis of Carboxylic Acid Porphyrin Dimers ...... 146 5.2.3. Synthesis of Nitrogen-centred Cationic Porphyrin Dimers ...... 151 5.3. Singlet Oxygen Measurements ...... 153 5.4. Conclusions and Outlook ...... 156 Chapter 6. Experimental ...... 157 6.1. General methods, Instrumentation and Considerations ...... 157 6.2. Ch. 2 – Cubane Cross-coupling and Cubane–Porphyrin Arrays ...... 158 6.2.1. Synthesis of Cubane Precursors ...... 158

6.2.2. Synthesis of A2BC-Porphyrins ...... 159 6.2.3. Synthesis of cubanyl-porphyrin complexes ...... 165 6.3. Ch. 3 –Triptycene Cross-coupling and Triptycene–Porphyrin Arrays ...... 173 6.3.1. Synthesis of Triptycene Precursors ...... 173 6.3.2. Synthesis of Porphyrin Precursors ...... 174 6.3.3. Synthesis of Triptycene-linked Chromophores ...... 176 6.3.4. Synthesis of Ethynylbenzene Compounds ...... 181 6.4. Ch. 4 – Bridging and Conformational Control of Porphyrin Units through Non- Traditional Rigid Scaffolds ...... 183 6.4.1. Synthesis of Cubane Precursors ...... 183 6.4.2. Synthesis of Dipyrromethane and Porphyrin Precursors ...... 183 6.4.3. Synthesis of Porphyrins for Cubanyl Amide-linked Complexes...... 185 6.4.4. Synthesis of Cubane Linkers ...... 193 6.4.5. Synthesis of Cubane-linked Porphyrin Monomers ...... 197 6.4.6. Synthesis of Cubane-linked Porphyrin Dimers ...... 202 6.5. Ch. 5 – Synthesis of Covalently-linked Water-soluble Dimers for PDT Treatment...... 207 6.5.1. Synthesis of Porphyrin Precursors ...... 207 6.5.2. Synthesis of Porphyrin Dimer Complexes ...... 211 6.6. Single Crystal X-ray Data ...... 214 References ...... 216

List of Abbreviations

Ac acetyl acac acetylacetonate

AuP gold porphyrin

BB 1,4-diethynyl benzene

BBN borabicyclo[3.3.1]nonane

BCO bicyclo[2.2.]octane

BCP bicyclo[1.1.1]pentane

BODIPY boron dipyrromethene

BDE bond dissociation energy

Bmb 5,5''-((1E,1'E)-(1,4-dimethoxy-9,10-[1,2]benzenoanthracene-9,10- diyl)bis(ethene-2,1-diyl))di-2,2'-bipyridine

Bpy 2,2'-bipyridine

Bqb 9,10-bis((E)-2-([2,2'-bipyridin]-5-yl)vinyl)-9,10-dihydro-9,10- [1,2]benzenoanthracene-13,16-dione

Btb 9,10-bis((E)-2-([2,2'-bipyridin]-5-yl)vinyl)-9,10-triptycene cat catalyst calcd calculated

CT charge transfer

CS charge separation

δ chemical shift

DAniF N,N’-di-p-anisylformamidinate

DABCO 1,4-diazabicyclo[2.2.]octane

DCE dichloroethane

DDQ 2,3-dichloro-5,6-dicyanobenzoquinone

DME dimethoxyethane

DMF dimethylformamide

xi dec decomposition

DIC N,N′-diisopropylcarbodiimide

DMF dimethyl formamide

DPBF 1,3-diphenylisobenzofuran

DPM dipyrromethane

D–B–A donor-bridge-acceptor

EDG electron donating group

EWG electron withdrawing group

ECF ethylchloroformate

Et ethyl

EtOH ethanol eq equivalent

FeP iron porphyrin

∆ heat

HATU (1-[bis(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium

3-oxid hexafluorophosphate)

HOAt 1-hydroxy-7-azabenzotriazole

HRMS high resolution mass spectrometry

H2P metal-free porphyrin

IBX 2-Iodoxybenzoic acid

IFV internal free volume

IP ion pair i-Pr iso-propyl

Pyr pyridine

λ wavelength

LiTMP lithium tetramethylpiperidide

xii

M molarity

[M]+ molecular cation m multiplet

Me methyl

MeOH methanol mol-% mole fraction m.p. melting point m/z mass-to-charge ratio

NBO natural bond orbital

NB 1,4-diethynylnaphthalene

NBS N-bromosuccinimide n-Bu n-butyl

NGF nerve growth factor

NMR nuclear magnetic resonance n.d. not detected

NIm 1,4:5,8-naphthalenetetracarboximide

NRHs nonconjugated rigid hydrocarbons

PEP phenylene-ethynylene-phenylene

Ph phenyl

PIFA [bis(trifluoroacetoxy)iodo]benzene ppm parts per million ppt precipitate

Prod product

RC reaction centre

Rf retention factor

ROS reactive oxygen species

xiii rt room temperature (20 °C) s singlet

STM scanning tunnelling microscope

SET single-electron-transfer

Solv solvent

T temperature t time t-Bu tert-butyl

TEMPO (2,2,6,6-tetramethylpiperidin-1-yl)oxyl

TFA trifluoroacetic acid

THF tetrahydrofuran

TIPS triisopropylsilane

TMS trimethylsilane

TMSA trimethylsilylacetylene

TLC thin layer chromatography

Tol toluene

TPP 5,10,15,20-tetraphenylporphyrin

TPPS 4,4',4'',4'''-(porphyrin-5,10,15,20-tetrayl)tetrabenzenesulfonate p-Tol para-tolyl

Porph porphyrin

Prod product

Solv solvent

UV ultraviolet

Vis visible v/v volume to volume

ZnP zinc porphyrin

xiv

Chapter 1. Introduction

1.1. Nonconjugated Rigid Hydrocarbons

Arranging functionalities in a linear fashion enables the alteration of the distance between the groups without effecting the overall geometry of the molecule, which is of high importance in medical and material sciences. This can be achieved with nonconjugated rigid hydrocarbons (NRHs) which share three main features for their usage in biological and material sciences; i) they are rigid, i.e. all conformational changes are frozen; ii) they are linear, where the arrangement of the different functionalities happens in a defined spatial manner, in this case a 180° fashion; iii) they are nonconjugated, as the formal sp3 hybridised carbon centre interrupts electronic communication (even though this is only partially true, as will be discussed in detail vide infra).

Over the years, linear linking has been a classic field of para-substituted benzenes and alkynes, due to the plethora of well-established cross-coupling reactions.[1] Recently however, certain other hydrocarbons, such as cubane or bicyclo[1.1.1]pentane (BCP) have begun to draw attention, due to their remarkable properties.[2] Besides these highly strained, rather unusual hydrocarbons more common moieties are in use, such as bicyclo[2.2.2]octane (BCO) and triptycene (Figure 1). In particular, the use of triptycene as a molecular rotor has received interest, propelled by the Nobel prize in chemistry for molecular machines awarded in 2016.[3] Another candidate fulfilling all the above mentioned criteria is tricyclo[2.1.0.02,5]pentane; but this compound is rarely used, since the stability of the plain hydrocarbon is too low to be of practical use.[4]

Figure 1. Nonconjugated rigid hydrocarbons (NHRs).

1.2. Physical Properties

To understand the different chemical behaviour during the synthesis and the resulting properties of the NRHs a closer look into some specific physical detail is necessary. This chapter identifies common trends and differences from a physical perspective between the different groups and combines it with additional information about the specific bond situation given for the individual linker motif.

1.2.1. Size and Dimension

The structural properties of the NRHs have been studied by single-crystal X-ray diffraction, electron diffraction and other types of spectroscopy. Even though the spatial geometry for each linker is linear, there is a remarkable variety in size and therefore in the length of the linker unit (Figure 2). While acetylene 1 is rather small with a C–C distance of 1.20 Å,[5] the size increases gradually from BCP 2 (1.85 Å),[6] to BCO 3 (2.60 Å),[7] to triptycene 4 (2.61 Å),[8] to cubane 5 (2.72 Å),[9] which is very close to the largest representative benzene 6 (2.79 Å).[10]

Figure 2. Bond length and internal C–C distance for the bridgehead carbons.

1.2.2. Bond Considerations, Electronic Communication, and Ring Strain

Bridgehead hybridisation and distances are not only important for the electronic communication of linear attached functionalities of the target molecule, they also indicate the reactivity of the compound. 1H and 13C NMR shifts and coupling constants are powerful indicators to determine the formal character of a bond.[11] The chemical shifts for unstrained hydrocarbons like acetylene 1, benzene 6 and cyclohexane 7 are referred to here as standards (Figure 3).[12] Within those molecules the carbon-hydrogen coupling represents the classic hybridisation (50% = sp, 33% = sp2, 25% = sp3). A different situation occurs for 2 the nonconjugated-linear linkers. While the rather unstrained triptycene 4 and BCO 3 show the expected chemical shifts,[13] the degree of s-character is already significantly higher compared to cyclohexane.[14] This effect further increases for BCP 2 and cubane 5[15]. Here the formal bond situation is closer to sp2 then to sp3. The extraordinary s-character of the exocyclic C–H-orbitals in cubane, and therefore the p-rich character of the cage C–C- orbitals, are further reflected in the high acidity of the C–H bond.[16] The kinetic acidity of cubane is 6.6×10-4 that of benzene and 6.3×104 that of cyclohexane.[16c] This result is quite similar to other strained hydrocarbons, e.g., the kinetic acidity for cyclopropane is 7.1×104 that of cyclohexane, and both cyclopropane and cubane have a similar degree of hybridisation (32% vs. 31%).[17]

Figure 3. NMR and hybridisation data for bridgehead carbon atoms; the s-character is calculated via [11] Mueller and Pritchard’s empirical relationship: %s = 1/5 J(C–H).

1.2.3. Electronic Communication

An important characteristic of the different linker units is the efficacy of electronic communication within the carbon scaffold (“through-bond”) and the bridgehead-bridgehead carbons (“through-space”). A highly effective approach to investigate those influences is the acidity of linear-substituted carboxylic acids (Table 1). The acidity of BCP acid 8a (pKa =

5.63) and cubane acid 11a (pKa = 5.94) is significantly higher compared to unstrained BCO acid 9a (pKa = 6.54), as a result of the increased electronegativity (s-character) of the exocyclic carbon bonding orbitals relative to BCO.[18] In addition, there is a clear correlation of bridgehead-bridgehead distance and the influence of the substituent. The short

3 bridgehead-bridgehead distance within the BCP derivatives leads to strong influence of the substituent.

This effect, can be further demonstrated by the position of the substituent: 3-substituted BCO acids show less influence compared to 4-substituted ones, since the effect of the bridgehead-bridgehead interaction is missing.[19] Wiberg could show, that the effects of the different substituents for bicyclic carboxylic acids result from a field effect.[19] This inductive/field effect was shown to have a linear dependence on the acidity of the compound and the C–X dipole. The electronic effects in 4-substituted cubane carboxylic acids can even be utilised to illustrate substituent effects in cyclobutanes, which are otherwise too weak to be detected.[20] Irngartinger et al. could determine the different C–C bond lengths within the cubane skeleton, depending on the substituent present in the 4-position.[21] Again, a closer look into the strong 1,3-nonbonding interactions between bridgehead carbons is necessary for the BCP cage system. As mentioned above, the short interbridgehead distance (1.80–1.98 Å) leads to repulsion of the back lobes of the exocyclic hybrid orbitals, and is considered to be a main contributor to the overall strain energy of the system.[22]

Table 1. Acidity of linearly-substituted hydrocarbons.

Linker

pKa for X 8 9 10 11 a H 5.63 H 6.54 H 5.20 H 5.94

b F 4.84 CH3 6.50 CH3 5.23 CO2H 5.43

c Cl 4.69 Cl 5.72 Cl 4.67 CO2Me 5.40

d CF3 4.75 CF3 5.79 Br 4.67 Br 5.32

[a] [18, 23] Values were determined in EtOH/H2O v/v 1:1).

1.2.4. Ring Strain

Unusual bonding situations are exemplified by the strain energy of the particular systems. Usually the overall ring strain is a combination of various parameters: i) Baeyer strain[24] or angle strain (deformation of bond angle); ii) Pitzer strain[25] or conformational strain (torsional eclipsing interactions); iii) Prelog strain[26] or transannular strain (van der Waals interactions of transannular atoms). For example, the angle strain, the deviation from the tetrahedral angle of 109.5°, is higher in cyclopropane than that in cyclobutane (Table 2). The strain energy of cubane is about six times the amount of cyclobutane, which represents 4 the sum of the six “cyclobutane faces of the cube”. But compared to the ring strain per C– C bond, cubane has double the amount of cyclobutane. The deviation is even higher when the strain energy is calculated per number of carbon atoms (cubane ~20.2 kcal vs. butane ~6.6 kcal). However, one should be aware, that the concept of ring strain has its intrinsic limitations, e.g., in terms of chemical stability. The less strained isomer is not always the most stable one. In addition, cubane and BCP are relative stable compounds, even though highly strained. But nevertheless, the concept of ring strain is still very valuable and can be applied for a better interpretation of chemical reactivity.[27] An interesting example is the correlation of ring strain with the ability of the molecule to act as a hydrogen bond donor; the higher the strain, the more acidic the C–H bond.[28] While many of these strained hydrocarbons are thermodynamically unstable, due to their deviation from standard sp3 geometry, they possess remarkable kinetic stability. For example, cubane is thermodynamically unstable (ΔHf = 144 kcal/mol), and highly strained (SE = 161.5 kcal/mol), but it is capable of withstanding temperatures of up to 220 °C (Table 2).[29] BCP also contains notable kinetic stability, as it is shown to withstand temperatures greater than 280 °C.[30] BCO and triptycene are notably less strained (SE = 7.4 and 7, respectively) with heats of formation of 23.6 and 76.8 kcal/mol.

Table 2. Heat of combustion and strain energy of cycloalkanes.

Compound Heat of Heat of Strain Strain combustion formation energy[31] energy per ΔHc ΔHf (kcal/mol) C–C bond (kcal/mol) (kcal/mol) (kcal/mol)

[32] [32] [31] Cyclopropane (C3H6) 505.8 12.7 27.7 9.1

[20] [33] [31] Cyclobutane (C4H8) 656.0 6.8 26.3 6.6

[34] [34] [31] Cyclohexane (C6H12) 943.8 29.8 0.4 0.1

[35] [31] BCP (C5H8) n.d. 51 (calc.) 68.0 11.3

[36] [36] [37] Cubane (C8H8) 1155.2 142.7 161.5 13.5

[38] [38] BCO (C8H14) 1195.5 23.6 7.4 0.8 Triptycene 2409.1[39] 76.8[39] 7.0[39] 0.8 1.2.5. Bridgehead Cations, Anions and Radicals

In terms of chemical reactivity and stability other important factors have to be considered. How does the rigid scaffold behave in reactive intermediates? How stable are the according bridgehead cations or anions? Is the homolysis of the carbon-hydrogen bond possible, and how stable is the corresponding bridgehead radical? Which intermediates will break the rigid cage-like structure and rearrange to more stable products? In this section, those

5 questions will be addressed from a rather physical point of view; examples concerning the actual reactivity and chemistry will be discussed in detail in Sections 1.3.4. and 1.4.3.

1.2.5.1. Bridgehead Cations

The cubane cage is very stable toward skeletal rearrangement and the solvolysis rate is about 1015 times faster than originally calculated, even though the cubyl cation formed seems to neglect all known cation-rationales, such as the geometry is far from flat and hyperconjugation would result in a very strained cubene-like structure, an extreme example of a pyramidalised olefin (Figure 4).[40] This remarkable effect can be explained by additional stabilisation through delocalisation of the positive charge along the surrounding C–C bonds

(resonance stabilisation). In addition, further hyperconjugation, vicinal through bond σCαCβ→ p* could be the reason for the notable stability of the cubane cation.[41] Furthermore, Eaton et al. pointed out another consideration: the internuclear line between the individual carbon bonds represents a false picture of the actually bent bond (“banana bond”),[42] since the endocyclic interorbital angle is rather 101–104° than 90°,[43] reducing the actual pyramidalisation angle.[41] Further influences on the solvolysis rate are reported for substituted cubanes. While electron-withdrawing groups in γ-position slow down solvolysis, electron-donors in β-position accelerate it[40c, 44] So far no skeletal rearrangement has been observed for the cubane cation, and only recent calculations showed a possible pathway for the cage-opening via the cuneyl cation. The rate determining barrier for ring opening of 25.3 kcal/mol highlights the kinetic persistence of this unusual cation.[45]

Figure 4. Approximate solvolysis rate of brominated bridgeheads and cation stabilisation effect.

Originally, the first synthesis of triptycene by Bartlett et al. was conducted to compare the stabilising effect of benzene rings on the rigid, tied back orientation in triptycene with the triptycenyl cation.[46] In triptycene the π-system is fixed in an orthogonal orientation relative to the developing p-orbital, therefore no overlap/stabilisation is possible. Leaving groups at

6 the bridgehead position do not show the usual substitution chemistry since it cannot be flattened and as the backside is shielded by the other bridgehead carbon.[47] Shielding even occurs at the front side of the molecule, and for the triple benzylic-like position in 9- bromomethyltriptycene, due to the peri-type steric hindrance of the cationic centre and the adjacent aromatic hydrogens (Figure 5).[48]

Figure 5. Nucleophilic substitution on 9-substituted triptycenes.

Another unique property of the cubane core is the instability of the cubylcarbinyl cation, as the positive charge next to the cubane skeleton usually leads to rapid rearrangement to the homocubyl derivatives, as was observed for cubylmethyl alcohols 12. Originally described by T. W. Cole in his Ph.D. dissertation with Eaton 1966 (Scheme 1),[49] modern calculations suggest a mechanism via the homocubyloxonium ion 13a, for the Wagner–Meerwein type rearrangement.[50]

Scheme 1. Cubylcarbinyl cation rearrangement.

1.2.5.2. Bridgehead Anions

In general, the acidities and bond dissociation energies of hydrocarbons are influenced by

[51] the hybridisation of the according orbital. The higher the s-character, or the J(C–H), the higher the kinetic acidity of the hydrocarbon (low values of ΔHa) (Figure 6). This can be seen in the high kinetic acidities of BCP 17 and cubane 5. The high s-character increases

7 the acidity, as seen when compared to BCO 3, CP 15 and t-Bu 16. But due to the higher transannular interaction between the back lobes of the bridgehead carbon orbitals, the effect for BCP is partly mitigated. This effect can be illustrated by adding an electronegative substituent, e.g., chlorine, at position 3 of BCP. The energetically low-lying σ* orbitals of the halogen-carbon bonds can be populated, leading to a longer carbon-halogen bond.[52] The homolytic bond dissociation energy (BDE), reflecting the stability and the ease of formation of the alkyl radical, follows the predicted order for the stability of the radicals. The anionic intermediates are of greater importance for the functionalisation than the cationic counterparts. This is especially due to the wide variety of organometallic coupling and exchange reactions.

Figure 6. Experimental gas-phase acidities (∆Ha), electron affinities (EA), and bond dissociation energies (BDE). All values in kcal/mol; parenthetical values are results of G3 computations.[53]

1.2.5.3. Bridgehead Radicals

As for the bridgehead carbocations, the geometry for bridgehead radicals is quite different from other more flexible systems. The rigid core results in a permanent pyramidal conformation, far from the favoured planar conformation. Also, in comparison to the planar tert-butyl radical, which shows rapid inversion, the conformation is locked within the inflexible cage structure. The orientation of the SOMO and the tied back carbon atoms result in less distinctive stabilisation effects (Figure 7). Hyperconjugative delocalisation or β- scission of bridgehead carbons lead to high-energy bridgehead alkenes (anti-Bredt[54] alkenes). Consequently, the reduced steric shielding of the exposed bridgehead radicals, in combination with the alleviated hyperconjugation, results in a much higher reactivity in comparison to “normal” tertiary carbon radicals.[55] But despite their high reactivity the half-

8 lives for all bridgehead-radicals are long enough to study their properties with spectroscopic methods.

Figure 7. a) Hyperconjugative delocalisation. b) Definition of dihedral angle Θ

1.3. Cubane – a Rigid Synthetic Scaffold

1.3.1. The History of Cubane

Having been a topic of much discussion in the early 1900s and dismissed by many as synthetically implausible, cubane was first synthesised by Phillip E. Eaton and Thomas. W. Cole in 1964.[56] Despite its fascinating simple structure of eight methine units in perfect cubic arrangement, this synthesis is considered a milestone of organic chemistry,[57] challenging the concept of carbon hybridisation. Cubane was considered impossible due to the huge deviation in geometry of each carbon atom from the tetrahedral angle (109.5° of sp3-carbon vs. 90° in a cubic arrangement).[58] Despite its high strain, cubane shows a remarkable high thermal stability, due to the lack of orbital symmetry-allowed ring opening the thermal rearrangement does not occur under 200 °C.[59] Successive syntheses have emerged since then and all, bar two, contain a double Favorskii rearrangement as the vital step required to contract the bishomocubane system to the highly strained cubane scaffold. One important modification of Eaton’s synthesis was performed by Chapman et al.[60] which utilises the ethylene ketal of cyclopentanone 18 as an alternative starting material and consequentially results in a higher yielding reaction with more workable intermediates. In 1997 a further adaptation of cubane synthesis was published by Tsanaktsidis and co- workers[61] which avoids some of the more expensive and corrosive reagents and improves the purification of key intermediates. Additionally, it allows access to an overall process that can now be achieved in five steps performed in the laboratory on multi-gram scales.[62]

Details of the Tsanaktsidis approach are shown in Scheme 2. The tribromo-derivative 20 is achieved via the bromination of 19. Following this, a highly endo-selective [4π+2π] Diels– Alder dimerisation occurs in the presence of excess NaOH and refluxing methanol to afford the bisethylene ketal 21 in approximately 60% yield over the two steps. Ketal 21 is converted into the dienone 22 in concentrated sulfuric acid. A [2π+2π] photocyclisation reaction is then carried out by the irradiation of dienone 22 at room temperature. The resulting caged dione 23 then undergoes a double Favorskii reaction when heated with aqueous NaOH, to

9 produce the diacid. It has been reported that the optimised purification of the diacid is achieved through functional group interconversion to dimethyl cubane-1,4-dicarboxylate 24 and consequent sublimation to give a crystalline product.[61] The yields of 42−47% are reported for the synthesis of 2224 which is a notable improvement on the ca. 20% reported by both Eaton and Chapman.[56, 60] Overall, a yield of ~25% is reported for the entire process.[61] Diester 24 has proven to be the most common and convenient entry point into accessing the cubane scaffold and consequential synthesis.

Scheme 2. Reported Tsanaktsidis method (black) and Eaton’s original synthesis (brown) for the synthesis of dimethyl cubane-1,4-dicarboxylate 24.

Other approaches, to generate substituted cubanes make use of [2+2] cycloadditions of substituted tricyclo[4.2.0.02,5]octa-3,7-diene 29 derivatives, a method not suitable for unsubstituted cubanes, due to the large distance of the involved π-orbitals (3.05 Å) (Scheme 3).[63]

Scheme 3. Persubstituted cubanes.

Alongside cubanes challenging synthesis, an additional topic of much discussion is the reaction of 1,4-diiodocubane 30 with tert-butyl lithium to form cubane 32, three different reactive intermediates, the cubane-diyl 31a, the cubane-propellane 31b and the cubane- diene 31c, seem plausible (Scheme 4).[64] Due to an orbital mismatch the long distance of the internal bond would be rather weak and calculations could rule out the likelihood of a cubane-propellane intermediate and strengthen the possibility of a singlet diradical species.[65] The cubane-diene intermediate 31c, a supposedly excellent dienophile, could not be quenched in the presence of Diels–Alder dienes.

Scheme 4. Reactive Intermediates.

1.3.2. Properties and Applications of Cubane

Due to the unusual structure, cubanes’ physical properties are quite different from other hydrocarbons (Table 3). The perfect cubic arrangement of the carbon atoms results in an

3 [66] octahedral (Oh) point group, with a high density of 1.29 g/cm . The C–C distance of 1.571 Å is slightly longer than in unstrained alkanes (1.54 Å) and in cyclobutane (1.55 Å).[67]

Table 3. Physical properties of cubane

Molecular Formula C8H8

[68] Shape Oh point group

Colour[68a] Transparent, rhombic crystals

Decomposition[29] >220 °C

Density[66] 1.29 g/cm3

Heat of formation[36] 144 kcal/mol

Strain energy[37] 161.5 kcal/mol

C–C bond length[67] 1.571 Å

C–H bond length[67] 1.109 Å

Further information can be deduced in comprehensive reviews[2a, 66, 69] and book chapters,[70] each delivering a unique view onto chemistry`s platonic body par excellence.

Due to the strained geometry of the cubane system a departure from classical sp3 hybridisation can be seen. The ring strain present in the system induces the C–C bonds to adopt more p-character while the C–H bonds compensate by becoming more s-rich. Della 11 et al. have calculated the s-character of the C–H bonds in cubane to be 30.1%.[14] When compared with 26% for adamantane, it is seen that cubane is more closely related to a sp2 hybridised system (33% s) than that of a sp3 system (25% s). This means that the acidity of the hydrogen atoms of cubane has increased with respect to conventional saturated hydrocarbons. Additionally, when compared with cyclohexane, cubane’s kinetic acidity is almost 63,000 times greater.[16b]

Owing to its exceptional structure, pronounced strain and symmetry, cubane has become a compound of much interest since its discovery. Many cubane derivatives have been synthesised, all with varying physical, chemical and biological properties. Currently cubane is credited with being the highest density hydrocarbon known, and coupled with its incredibly high energy strain has led to its application in the fields of high energy fuels and explosives.[71] For example, octanitrocubane 33 is proven to be an exceptionally powerful, while shock insensitive, explosive.[72]

In contrast cubylamines, have shown evidence of antiviral activity.[66] Additionally, in 1996 Cheng et al. published research on N-cubylmethyl substituted morphinoids, which act as novel narcotic antagonists. Cheng showed that N-cubylmethylnormorphine 34 and N- cubylmethylnoroxymorphone have a higher potency at the μ and κ opioid receptors when compared with that of morphine and oxymorphone respectively (Figure 8).[73]

Figure 8. Examples of cubane scaffolds: octanitrocubane 33 and N-cubylmethylnoroxymorphone 34.

Additionally, a series of 4-substituted-1-cubanecarboxylic acids were synthesised with interesting crystallisation patterns that indicate their potential for crystal engineering.[74] Within this family of compounds, the unusual catemers of the syn-anti-conformation are frequently observed, specifically in the cases where X= Cl, Br, I or CO2Me. The reoccurrence of this rare conformation is attributed to the stabilisation between C–H···O and cubanyl C–H bonds caused by the high acidity of the cubanyl protons. In this orientation, the halogens can be seen to fill the centrosymmetric voids that are present due to the rest of the packing, thus stabilising the whole crystal structure. Moreover, when X =

CO2Me cyclic patterns are formed, and in cases where the substituent is too small i.e. X =

H, too big X = Ph, or when it has a specific hydrogen bonding preference of its own (X =

CONH2), the catemer no longer forms and instead dimer formation is seen (Figure 9).

Figure 9. Different hydrogen bonding interactions for 1,4-substituted cubane carboxylic acids.

Due to its very rigid three-dimensional structure cubane offers opportunities to make a variety of new materials, including polymers. The first cubanyl polymer was achieved by the metathesis polymerisation of the 1,4-bis(homoallyl)-cubane 35. Addition of Schrock’s molybdenum catalyst enabled the production of an oligomer with an average of 6.2 repeat units per chain (Figure 10).[75]

Figure 10. Cubane polymers.

Figure 11. Towards transparent MOF and a molecular tunnel.

The singly metal–metal-bonded complex [Rh2(cis-DAniF)2-(CH3CNeq)4(CH3CNax)2](BF4)2 36

+ 2- 2- (DAniF = N,N’-di-p-anisylformamidinate) was assembled with (Et4N )2(Carb ), where Carb is the dicarboxylate anion of BCP and cubane, amongst other groups to form square complexes (Figure 11). In crystal form these complexes stack, forming infinite tunnels which can be closed to allow for the encapsulation of solvent molecules.[76]

Cubane has also found application as a drug bioisotere, which is a popular option to overcome problems with design and development of drug candidates. Due to its ubiquity in most drug molecules benzene is often chosen as a target for bioisosteric manipulation.[77] Chosen for its rigidity, unique electronics and synthetic accessibility, many drugs contain benzene moieties, although the presence of a benzene ring in a drugs structure is known to be one of the leading causes for compound attrition in drug discovery (Figure 12).[78] As cubane is incapable of undergoing π-π stacking interactions it was hypothesised to overcome this problem.

Figure 12. A visual concept of cubane as a benzene bioisostere.

Figure 13. Cubane benzene bioisosteres.

In the early 90’s Eaton postulated that cubane would be an ideal isostere for benzene, due to their similarity in size and shape (Figure 12).[40d, 66] Various molecules were synthesised to test the applicability of this hypothesis with applications in cancer therapeutics, AD medication and pain management The histone deacetylase inhibitor SAHA 37 (suberanilohydroxamic acid),[79] used for the treatment of cutaneous T-cell lymphoma (CTCL) was the first molecule investigated (Figure 13).[10] Tumour cell line inhibition studies showed that both, SAHA and the cubane analogue 37a, had similar potencies in tumour cell inhibition, but SAHA’s toxicity was slightly greater towards NFF primary cells.

Previous studies showed significantly increased neurite outgrowth in PC12 neural precursor cells derived from a rat pheochromocytoma upon administration of the neotrophic drug leteprinim.[80] The treatment of PC12 cells with either leteprinim 38 or the cubane analogue 38a were both ineffective in the absence of nerve growth factor (NGF). But upon administration together the differentiation capacity of the cubane analogue was remarkably better than the parent compound.

The third compound targeted was the non-selective sodium ion channel blocker benzocaine 39,[81] a widely used local anaesthetic. Adult male rats were used to test the effect of the cubane 39a and parent analogues, through injection of the drugs and application of an acute noxious heat stimulus at the source of injection. Results showed the same local anaesthetic efficacy of both drugs in this model. Other examples of cubane as a bioisosteres are outlined in the publication by Chalmers et. al.[10] Overall results showed the applicability of cubane as a benzene bioisostere.

Figure 14. Non-natural peptides.

In an effort to increase the metabolic stability and bioavailability of peptides, additional non- natural amino acids have been incorporated into peptides. Adamantane-substituted peptides have been shown to enhance the peptides’ ability to penetrate biological membranes, and they have been investigated for their anti-tumour[82] and antimicrobial activities.[83] With structural similarities to adamantane, cubane-based derivatives are also a viable option for non-natural amino acids. The first cubane-based amino acid synthesised for its neuroprotective properties was 4-carboxylcubylglycine 40b.[84] As unfunctionalised cubane is a closer mimic of hydrophobic amino acids such as leucine, isoleucine, and

15 phenylalanine, cubane derivative 40a was synthesised, as were the dipeptide derivatives 40b–40d bearing cubane residues in place of side chains (Figure 14).[85] Due to the high sensitivity of the vinylcubane unit, initial attempts to prepare a cubane-based alanine derivative via the corresponding dehydroalanine were unsuccessful. However, the successful synthesis of a cubane-based glycine derivative and cubane-substituted dipeptides 40b–40d in diastereomerically pure form was achieved upon the addition of lithiated cubane to a (R,S)-glyoxylate sulfinimine.

1.3.3. Reactivity and Manipulation of the Bridgehead Carbon Centre

Limitations of the Carbon Skeleton – Stability and Rearrangements of Cubane

As cubane bears a vast amount of strain energy in its cage structure, certain reaction conditions are able to crack the kinetically stable skeleton and lead to thermodynamic more stable products. But as the activation energy (Ea= 43 kcal/mol) of the rate-determining step for a homolytical C–C-bond cleavage, during the rearrangement to cyclooctatetraene 41, provides only limited relief in the overall strain, thermal ring-opening only occurs at higher temperatures (Scheme 5).[86]

Scheme 5. Thermal isomerisation.

Scheme 6. Rearrangement of cubane skeleton.

Another option to release some strain energy is a metal-catalysed σ-bond rearrangement. In the presence of Ag(I) or Pd(II), insertion into the C–C bond of cubane occurs (Scheme

6a).[87] A rearrangement of the cubane skeleton yields the stable cuneane 42 (lat. cuneus = wedge). A similar rearrangement is observed in the presence of Li-cations or protons (Scheme 6b).[88] Rhodium(I) complexes rapidly insert into the C–C bond of cubanes in an oxidative addition style. The intermediate five membered metallacyclus 41a is rearranged to the more staple tricyclodiene 41b (Scheme 6c).[89]

Certain other reaction conditions are able to crack the kinetically stable cubane framework. In the presence of palladium on charcoal Stober et al. observed a stepwise hydrogenation of the strained skeleton.[90] The ring-opening of the first ring leads to the secocubane 44. Every further hydrogenation step, from secocubane to nortwistbrendane 45 and from nortwistbrendane to BCO 3, releases about 50 kcal/mol of strain energy (Scheme 7).[91] The order of hydrogenation proceeds as expected; the least stable/the longest bond is hydrogenated first.

Scheme 7. Stepwise hydrogenation of cubane.

The presence of an anionic charge at the carbon vicinal to the cubane skeleton 47 leads to degradation of the ring system 48. Deprotonation at the α-position of an acceptor system results in a homoallylic rearrangement (Scheme 8).[92] A similar instability is observed for cubylmethylalcohol 52. Slight acidic conditions induce a homoketonisation reaction yielding cyclobutene 54 (Scheme 9).[93]

Scheme 8. Homoallylic rearrangement.

Scheme 9. Cubylmethylalcohol rearrangement.

A very interesting rearrangement of the cubane core was observed during the attempts to polymerise the iodo-vinyl cubane 55.[94] The partial positive charge in the vicinal position 56 leads to a sequence of skeletal rearrangements (Scheme 10). A similar rearrangement was observed under pure thermal conditions and at elevated temperatures during Grubbs- metathesis.[95] The final product after a cascade of cycloreversions and cycloadditions is the stable iodo-vinyl styrene 61.

Scheme 10. Vinyl iodide rearrangement.

1.3.4. Cubane Chemistry

Cubanyl chemistry has evolved over the years in order to increase its applicability in modern day synthesis. Work in the 80s showed how a cubanyl carboxylic acid group 11a could be transformed into an iodide 62 by a hypervalent iodine oxidative decarboxylation mechanism

[96] (Scheme 11). This is achieved through the use of hypervalent Phl(OAc)2-CCl4-I2 and with irradiation of the compound. The iodocubane products were achieved in 80–90%. Phenylcubane 64 was first obtained from the transformation of fluorocubane 63, when it is reacted with benzene and boron trifluoride in toluene in a Friedel–Crafts type reaction.[97] After 30 minutes, a 40% yield of phenylcubane was achieved as the sole product.

Much chemistry has been reported based on work with carbonyl cubanes. Dimethyl cubane- 1,4-dicarboxylate 24 was transformed into a variety of different functional groups from reactions with the ester functional group. Triazole 65a, isoxazole 65b, oxazole 65c, pyrazole 65d, thiazole 65e, benzimidazol 65f, imidazole 65g, pyridine 65h, and imidazole 65i functionalities were all introduced through classic carbonyl transformations.[98]

In order to access cubane for use in palladium cross-coupling reactions a series of chemically distinct, highly strained, activated cubane scaffolds were synthesised by the Senge group (Scheme 12). This was achieved by using iodinated cubane derivatives 67 to

18 optimise lithium−halogen exchange reactions.[99] Boron 68d, phosphorus 68f, tin 68e, silicon 68h, sulfur 68i and alkyl 68a−c groups were attached to the cubane scaffold with this method. The optimum conditions found for the metal-halogen exchange reaction allowed for the generation of the lithiated intermediate through the reaction of cubanyl iodide 67 with two equivalents of t-BuLi at –78 °C in THF for one hour. The reaction mixture was then allowed to warm to room temperature after two equivalents of the relevant R–X reagent were added. These electrophilic cubanes were then investigated for their use in Suzuki– Miyaura, Negishi, and Stille cross-coupling reactions with various halogenated phenyl groups, but all coupling reactions proved unsuccessful.

Scheme 11. Examples of organic transformations with cubane and classic cubane carbonyl chemistry.

Scheme 12. Expanded cubane chemistry.

Scheme 13. Redox-active ester chemistry.

A popular and versatile method has arisen that allows for the attachment of a variety of different functional groups directly to many of the bulky alkyl groups of interest in this work, i.e. BCP, BCO and cubane. These methods, as developed by Baran’s group involve the cross-coupling of redox-active esters, derived from alkyl carboxylic acids, with

20 organometallic species and employ single electron transfer to achieve the desired coupled compounds (Scheme 13). The mechanism for the majority of these reactions involves a single-electron-transfer (SET) mechanism, which circumvents the need for oxidative addition of a transition metal into the carbon-halogen bond, which can often be detrimental for the stability of the strained- alkyl groups under discussion.

The alkyl redox-active ester 70 has been shown to successfully couple with organozinc and organomagnesium species using an Fe-based catalyst system such as Fe(acac)3 and the dppBz ligand to attach phenyl groups to cubane 71, BCO 71a and BCP 71b;[100] while the formation of alkyl boronic acids and esters can also be achieved through the use of the alkyl redox-active esters. The catalytic system of NiCl•6H2O and MgBr2•OEt2 is employed to couple the redox-active ester with lithiated bis(pinacolato)diboron.[101] The boronic acid can then be accessed by reacting the ester product with boron trichloride. This method was successfully applied to synthesise methyl 4-(pinacolboron)cubane-1-carboxylate 72 and methyl 4-(pinacolboron)-BCO-1-carboxylate 72a.

On the other hand, alkyl boronic acids and esters can also be produced through photoinduced decarboxylative borylation of carboxylic acids as developed by Aggarwal and co-workers.[102] This method does not use transition metal catalysis but instead utilises light to initiate the radical combination of the N-hydroxyphthalimide ester derivative with the diboron reagent bis(catecholato)diboron. This reaction was used with both cubane, BCO and BCP scaffolds 72, 73a and 73b, respectively, to form their boronic ester derivatives in moderate to good yields.

Additional nickel-catalysed Barton chemistry with cubane can be observed such as the decarboxylation and Giese radical conjugate addition reactions.[103] These reactions were reinvestigated to simplify the required conditions and widen the scope of the reactions. In each case N-hydroxyphthalimide (NHPI) based redox-active esters were utilised and a thermally initiated Ni-catalysed radical formation was carried out. Subsequent trapping with either a hydrogen atom source (PhSiH3) or an electron-deficient olefin in the case of 74 and 75, respectively, led to the two products of interest. The former route resulted in the decarboxylated cubane product in 77% yield while the latter route gave the alkylated cubanyl product in 56% yield. Additionally, an interrupted Barton decarboxylation reaction can be used to synthesise simple sulfinate salts 77 from readily available carboxylic acids 76.[104] The carboxylic acid is transformed to acyl chloride, then reacted with 1- hydroxypyridine-2-thione, followed by illumination with light, then exposure to ruthenium trichloride and sodium periodate to form the sulfone product. Addition of sodium ethoxide allows for the conversion to the sulfinate salt. The cubanyl sulfone product was achieved in 42% and the salt in 95%.

1.4. Triptycene

Highly related to the BCO system is the triple benzannulated tripycene, the simplest representative of the so-called iptycenes.[105] Triptycene – named after “triptych” (τρίπτυχον greek. “threefold”), a painting or book with three leaves hinged on a common axis – was first synthesised by Bartlett et al. in 1942 (Scheme 14).[46c] A Diels–Alder reaction was used to install the threefold geometry, by reacting anthracene 78 with quinone in refluxing xylene. The final defunctionalisation of the quinone intermediate was obtained by isomerisation, oxidation to the corresponding quinone 80, substitution with hydroxylamine hydrochloride and a subsequent reduction to the diamine 81. The bis-diazoniumtriptycene was afforded through diazotisation with sodium nitrite and reduction of the corresponding chloride in the presence of potassium hydroxide and palladium on calcium carbonate, to yield triptycene 4. This synthetic route enabled access to a series of substituted BCO derivatives, in order to study the nucleophilic substitution at the bridgehead carbon centre and thus investigate the reactivity of structures, which cannot undergo the Walden inversion.[46a, 46b] A further insight into the chemistry of those bridgeheads will be given at a later stage. Up to now, using a Diels–Alder reaction with anthracene derivatives is still the most valid method for the introduction of functionalities in the bridgehead positions, especially with the usage of aryne 82 coupling partners, generating the (functionalised)-triptycene unit in a single step.[106]

Scheme 14. Synthesis of triptycene according to Bartlett et al.[39] i) Quinone, ∆, xylene, 140 °C, 2 h; ii) HBr, AcOH; iii) KBrO3, AcOH; iv) H2NOH•HCl; v) SnCl2, HCl, EtOH; vi) H2SO4, AcOH, NaNO2, HCl; vii) Pd/CaCO3, NH2NH2, KOH, EtOH, 100 °C, 1.5 h; viii) o-fluorobromobenzene or o-diazonium benzoate.

The missing link from BCO to triptycene is bicyclo[2.2.2]octa-2,5,7-triene, the so-called barrelene 87. Its first synthesis was achieved by Zimmerman and Paufler in 1960 (Scheme 15).[107]

Scheme 15. Synthesis of barrelene by Zimmerman et al.; a) Cu, 650 °C; b) methyl vinyl ketone, 160

°C; c) H2NOH, EtOH/H2O; d) p-TsCl, then NaHCO3; e) NaOH, H2O; f) MeI, NaOH; g) Ag2O, then Δ.[107]

Coumalic acid 83 was decarboxylated over copper at 650 °C to -pyrone 84. Afterwards the labile diene was heated under reflux with two equivalents of methyl vinyl ketone, generating the bicyclo[2.2.2]octene 85 skeleton in a double Diels–Alder reaction with loss of carbon dioxide. Generation of the bis-oxime, and a subsequent second-order Beckmann rearrangement under alkaline conditions of the bis-tosylate, yielded the diamine 86. Exhaustive methylation and Hofmann elimination of the quaternary ammonium salts led to the introduction of the two missing double bonds, finalising the synthesis of barrelene 87.

The photochemical isomerisation of barrelene represents one of the most fundamental and general photo-processes; a di-π-methane rearrangement is observed, which initiated different general mechanistic discussions (Scheme 16).[108]

Scheme 16. Reactivity and photochemistry of barrelene.

Different pathways are observed depending on the presence of an additional photosensitiser. Direct irradiation results in a photochemical isomerisation to cyclooctatetraene 41 via the intramolecular [2+2]-cycloadduct 88. In the presence of a suitable photosensitiser, the triplet excited state reacts to the diradical 90a. Ring opening of

23 the cyclopropane intermediate to the stabilised allylradical 90b results in the subsequent formation of semibullvalene 89.[109] Consideration of the bond structure, according to Hückel’s rule, suggests barrelene 87 as an aromatic compound. Although the π-bonds are not coplanar, the p-orbitals show some homoconjugative overlap with a phase shift, a characteristic phenomenon for Möbius-type aromatic structures (Figure 15).[110]

Figure 15. Barrelene 87 as a Möbius aromatic system.

However, the chemistry and reactivity of 87 indicate a rather non-aromatic behaviour. Hydrogenation of the double bonds results in the saturated BCO 3 (Scheme 16). Thereby, comparison to other hydrocarbons suggests a destabilisation of about 10 kcal/mol due to angular strain.[111] The addition of bromine leads to a transannular bond formation yielding 89, showing typical non-aromatic behaviour as well.[110a]

Despite its theoretical interest, little use of barrelene 87 as a building block is reported in the literature.[112] The existence of the even more bizarre propellene analogue, [2.2.2]propellatriene, is topic of speculations and for now remains at the realms of theoretical chemists.[113] The carbon-skeleton of triptycene and BCO, both unstrained linker systems, are nearly indestructible under normal reaction conditions.

1.4.1. Structure and Properties of Triptycene

Over the years many characteristic properties of triptycene have been observed, for

[114] example its rigid Y-shaped structure that has a D3h symmetry. The molecule has a height (H) of 7.0 Å and length (L) of 8.1 Å, as measured from the centre points of the extreme hydrogen atoms. This creates an internal free volume (IFV), defined as an equilateral triangular based prism of H and base L (Figure 16).[115] The IFV, aided by the high energy barrier to the twisting or deformation of the [2.2.2]-bridgehead system, results in three 120° electron-rich clefts between the phenyl rings (Figure 16).[114, 116] Consequently, triptycene has potential applications in host-guest complexes, molecular inclusion compounds and coordination compounds with unusual geometries.[117] Most significantly for our purposes the 120o orientation provided by this framework qualifies it as a useful linker group for multichromophore assemblies.[118] As a result, interest arose in using a triptycene unit as a scaffold for the construction of multiporphyrin arrays with applications in host-guest chemistry. Additionally, several reactive positions for functionalisation and open electron-

24 rich cavities exist around the periphery of triptycene, which all afford several additional prospective applications such as molecular machines,[119] supramolecular chemistry,[120] electron transfer,[106b] anti-cancer[121] and sensor[122] applications.

Figure 16. IMFV of triptycene, hexafunctionalised triptycene.

Due to the 120° rigid void, mono-, di-, tri-, tetra-, penta- and hexafunctionalisations of triptycene, the scaffolds can be displayed in a spatially defined manner.[46c] Thus, a variety of functional groups can be presented in a fixed orientation. Functionalisation of unsubstituted triptycene involves the addition of new functional groups to the triptycene periphery, which can result in altered optical and structural properties. There are two main synthetic pathways for the preparation of functionalised triptycenes. The anthracene A and a benzyne B derivatives can be functionalised prior to the formation of triptycene. A Diels– Alder reaction can then be employed to construct the triptycene scaffold C. Alternatively, triptycene D can be functionalised post-formation. (Scheme 17).[123] Consequently, due to the limited availability of aryne precursors and anthracene derivatives, the second pathway of selective reactions and derivatisations of triptycene became a more efficient and practical route to triptycene derivatives with specific functional groups.[124]

Scheme 17. Synthetic pathways for the functionalisation of triptycene.

1.4.2. Applications

1.4.2.1. Molecular Rotor and Gyroscopes

Molecular rotors are described as molecules containing two parts that can easily rotate relative to each other. The parts can either be interlocked such as rotaxanes and catenanes or held together by chemical bonds. The rotator has the smaller moment of inertia while the stator has the largest.

Work with molecular machines was first pioneered and developed by Jean-Pierre Sauvage, Sir J. Fraser Stoddart and Bernard L. Feringa, who were awarded the 2016 Nobel prize "for the design and synthesis of molecular machines".[125] The interest in functional molecular machines first began in the nineties when the real time rotation of an ATP synthase macromolecular motor was observed.[126] The synthetic aim of this field of study involves the synthesis of miniature man-made mechanical devices on the nanoscale. In order for this to be successful the motions of the molecular machine must be controllable, the surface on which the molecule is on must be suitable, as must be the external stimulus that is necessary to provide the energy for movement.[127]

Figure 17. a) Molecular machines with triptycene wheels; b) Molecular compasses and gyroscopes.

Triptycene enables the entire functionalised molecule to move easily on the surface by its rolling motion. System 92 was the first ever synthesised nanovehicle, with the two triptycene wheels linked along a butadiyne axel that enabled the molecule to have almost free-rotation and maintained a linear geometry (Figure 17).[5, 128] Synthesised over three steps from 9- bromoanthracene, 9-ethynyltriptycene was then dimerised using Glaser-Hay methods to

26 give the diethynyltriptycene wheel dimer.[129] The molecule was sublimed onto a Cu(110) surface and wheel rotation was induced and observed for the first time using a scanning tunnelling microscope (STM) tip. Typically, only one triptycene wheel was observed moving while the other stayed stationary, so further molecules were investigated to see if both wheels could simultaneously be activated.

There is a growing need for organic materials that have tuneable transmittance, refraction, polarisation and colour for use in communication technologies.[130] While current research in this area originates from polymer and liquid crystal chemistry, a new concept has emerged where photonics materials are constructed using dipolar units that can reorient rapidly under the influence of electric, magnetic, and optical stimuli.[131] These new molecular architectures are expected to function similarly to macroscopic compasses and gyroscopes.

A convergent synthesis was reported to prepare molecular gyroscopes where para- phenylene rotors linked by triple bonds to methyl-substituted triptycenes act as pivots and encapsulating frames. Compounds 93a–c were prepared from 2,3-dimethyl-1,3-butadiene using a Diels–Alder cycloaddition and a Sonogashira reaction to attach the three units. Different triptycenes with methyl, propyl, and benzene substituents at the bridgehead C10 position were employed to synthesise a variety of molecular gyroscopes. The best results were observed with small methyl and propyl electron donating substituents at C10 over the larger benzene group. Also, results suggest for free movement around the phenylene axis to occur as low as 100 K, illustrating a relatively efficient gyroscopic motion.[132]

1.4.2.2. Molecular multi-Gear Systems and Turnstiles

A cyclic multi-gear system 94 was synthesised with four 9,10-triptycene units connected to four 1,2-phenylene units via ethynyl linkers. This macrocycle was achieved through successive Sonogashira reactions with 9,10-diethynyltriptycene derivatives and diiodobenzene. DFT calculations showed the triptycene units to be intermeshed with one another via π···π and CH···π interactions. Despite this, all triptycene units in the tetramer rotated in a correlated and frictionless manner, regardless of the continuous interactions between the gear units. 1H NMR showed rapid rotation of the system, even at low temperatures (Figure 18).[133]

Figure 18. a) Quadruple triptycene gears; b) Molecular turnstiles regulated by metal ions.

A family of metal-mediated molecular turnstiles 95 was synthesised with various methyl- substituted triptycene rotors connected to two stators with pyridyl binding sites (Figure 18).[134] The methyl groups were introduced into the triptycene rotor to investigate the effect that increased rotor size has on the closed-state formation. In addition, metal ion coordination was investigated for close-state formation. It was observed that free-rotation occurs in the system at room temperature, but in the presence of either Pd2+ or Ag+ rotation is restricted via metal coordination at the pyridyl sites. The closed-state of the system is only observed at reduced temperatures upon Ag+ coordination showing that it is the binding of metal ions to the pyridyl group that determines the closed-state not the size of the rotor. In this manner, the bistability of molecular turnstiles can be regulated allowing for potential applications in the construction of functional molecular devices.

1.4.3. Triptycene Chemistry

Scheme 18. Functionalisation of the bridgehead carbon in triptycenes. *Solv. = solvent.

9-Bromotriptycene 96 has been utilised by many groups to further functionalise triptycene at the 9-position. This is most commonly executed through halogen-lithium exchange reactions (Scheme 18). This was first carried out in the 1970’s, when trimethyl-9-triptycenyl tin 98a was synthesised by quenching the 9-lithiotriptycene intermediate 97 with trimethyltinchloride.[135] Following this trimethyl-9-triptycenyl tin 98a was utilised to access dimethyl-9-triptycenyl tin hydride 98b. This was achieved via a bromination reaction

[136] followed by the reduction of tin with LiAlH4. Additionally, when 9-lithiotriptycene was quenched with formaldehyde the triptycene carbaldehyde 98c was formed.[137] 9-Triptycene phenethyl selenide 99 was prepared from 9-bromotriptycene again through the use of a

29 lithium−halogen exchange reaction and quenched with diphenyl diselenide to form the product in a 75% yield.[138] Moreover, 9-lithiotriptycene 97 was reacted with a series of metals such as GaCl3, AlCl3 and InBr3 to form the complexes, [(tript)GaCl2(THF)] 100a,

[139] [(tript)AlCl2(OEt2)] 100b and [(tript)InBr(µ-Br)2Li(OEt2)2] 100c, where tript = 9-triptycenyl.

Iodine and chlorine were also introduced at the 9-position of triptycene. Iodination was achieved by Bartlett via the radical-induced decomposition of ditriptycenoyl peroxide 101 at

[114] 80 °C in the presence of I2 to give compound 102. Chlorination was observed when 9- N-(acyloxy)phthalimide triptycene 103 and 1,4-diazaBCO (DABCO) were irradiated for three hours with a 100 W high pressure Hg lamp in a solution of [t-BuOH−CCl4−H2O (53:42:5, v/v)] resulting in the decarboxylated chloride product 104 in 59%.[140]]

1.5. An Introduction to Porphyrins

In contrast to the recent discovery of cubane, the first mention related to a porphyrin can be dated as far back as 460−370 BC when Hippocrates described a study similar to acute porphyria.[141] The term porphyrin originates from the Greek word, πορφυρίνης (porphyus), which means reddish-purple. The reddish colour of blood was first thought to be produced by iron but in the early 1840s this was disproved when iron was liberated from dried blood by washing it with concentrated sulfuric acid. After dissolving the resulting iron free residue in alcohol a reddish purple colour was still observed. In this way so-called iron-free hematin was formed and in-depth porphyrin studies began.[142]

It was not until 1912 that the porphyrin skeleton was first proposed by Küster[143] which outlined the presence of four pyrrolic units linked through methine bridges. 1929 gave the first synthetic proof of the suggested structure, when heme was first synthesised by Fischer and Klarer [144] Since then the area of tetrapyrrole chemistry has gained much momentum, resulting in a large and diverse library of structures with vast and varied applications. Porphyrins have remarkable optical and photophysical properties,[145] conformational flexibility,[146] chemical stability,[147] and biological relevance.[148] There are a vast number of applications for the macrocycle and they have shown their relevance in the fields of optics,[149] light harvesting,[150] surface chemistry[151] and cancer therapy.[152]

Porphyrins make-up a large part of a class of compounds known as tetrapyrrolic macrocycles. They are one of life’s most versatile and important compounds and can been seen throughout nature, from photosynthesis in plants (chlorophyll) and bacteria (bacteriochlorophyll), to oxygen transport in the blood (haemoglobin), enzymatic function

(cytochrome P450) in the liver and brain and nervous system function (vitamin B12) in the human body. While haem 105 is a classic example of a porphyrin consisting of 22 π- electrons, chlorophyll 106, consists of 20 π-electrons and is classified as a chlorin (Figure

19). Porphyrins and chlorins are only two examples of what make up the vast family of tetrapyrrolic macrocycles.

Figure 19. Structures of haem and chlorophyll.

1.5.1. Structure of Porphyrins

The porphyrin system (Figure 20) is made up of four pyrrole rings connected by four methine bridges. This macrocyclic ring consists of 22 π-electrons, of which only 18 constitute the aromatic system while the remaining four π-electrons possess more double bond character.

There are three different forms of carbon atoms present in the porphyrin structure, the Cα- and Cβ-carbons of the pyrrole subunits and the methine-bridged meso Cm carbons. Porphyrins undergo many reactions typical to aromatic systems namely electrophilic and nucleophilic aromatic substitution via halogenation, nitration and acylation reactions.[147b] These reactions all occur at the meso and β-carbons as they are the most reactive carbons.[148b] Moreover, the inner core of the porphyrin can also be modified most commonly through the insertion of a metal ion[153] or protonation[154], but additionally, also through alkylation[155] reactions of one or more of the nitrogen atoms. The type of metal inserted into the porphyrin core has significant effects on the properties of the molecule. For example, divalent ions, such as Zn2+ or Ni2+, activate the porphyrin towards nucleophilic attacks at the meso position, while the β-carbon is activated when electrophilic metals are present.

Figure 20. Porphyrin structure with Cα, Cβ and Cm positions indicated and meso-substituted porphyrins with ABCD nomenclature.

Although naturally occurring porphyrins are not meso-substituted, this class of porphyrin is the synthetically most accessible class of tetrapyrrole with wide ranging applications, in addition to being useful for fundamental studies and synthetic methods development. A classification system, that was coined the “porphyrin alphabet soup”[147b], dictates that when all four meso positions are different the porphyrin is called an ABCD porphyrin,[156] or in the case of TPP where all four R groups are a phenyl ring the porphyrin is labelled as an A4 porphyrin (Figure 20).

There are now numerous methods available to attach different substituents to the porphyrin enabling the functionalisation of the porphyrin core in a defined manner.[147b] Many different functionalities such as alkyl, aryl, heterocyclic and organometallic groups can be quickly generated for meso-substituted porphyrins and just require different aldehydes in a pyrrole condensation.[156] This has made meso-substituted porphyrins the go-to porphyrin of many present-day studies.

Figure 21. Retrosynthetic analysis of meso-substituted ABCD porphyrins.

Essentially, symmetry considerations of the desired porphyrin determine the synthetic route taken (Figure 21). If the target structure is symmetric, only a simple acid-catalysed tetramerisation of the pyrrole precursor with an aldehyde carrying the meso-substituent (route 4) is required such as in Rothemund[157] and Lindsey[158] condensations. On the other hand, while unsymmetric systems can be obtained in a one-step ‘mixed’ condensation reaction with different pyrroles/aldehydes (route 1) if the preferred regioisomer is not the major product difficult chromatographic purification is almost guaranteed. To overcome this, the synthesis of unsymmetric systems often require multiple steps. For the synthesis of less

[159] symmetrical A2 and A2B2 scaffolds, dipyrryl subunits (for 5,15-systems, route 1) or tripyrryl analogues (for 5,10-systems, route 2)[160] have been used in MacDonald [2+2]

32 condensations[161] with the desired aldehyde. Unsymmetric systems can also be accessed via a linear tetrapyrrole (route 3) followed by cyclisation.

1.5.2. Photosynthetic Mimics

There is an increasing need to discover more efficient methods to harvest energy and, as is often the case, we can look to nature as the supreme example of how to maximise this process. As a result many artificial molecular devices have been made for solar energy conversion by acting as photosynthetic mimics.[162] During photosynthesis, in order to absorb and transfer solar energy effectively the antenna chlorophylls in photosynthetic bacteria are organised as non-covalent macrocycles in a spatially defined manner (Figure 22).[41] Therefore, for electron transfer systems, it is crucial to have the correct geometry, conformation and spatial arrangement of the molecular building blocks in these photochemical systems.[163] In order to achieve this, rigid-linear isolating linkers (among others) are required to provide defined structures and fixed regiochemical arrangements similar to those seen in nature.

Figure 22. Basic overview of electron transfer mechanism occurring in photosystems during photosynthesis.

It is of major interest to mimic these processes with porphyrin assemblies, so they can be used as model compounds in electron transfer studies.[146a] The success in this field is dependent on the availability of appropriate scaffold molecules and linker units. While there is an availability of these structures, there is a notable absence of rigid, non-conjugating units with defined geometry. NRHs such as cubane and triptycene are hypothesised to act as rigid isolators, thus opening a new avenue for photosynthetic mimics.

1.5.3. Cubane and Triptycene as Potential Coupling Partners

Although tertiary alkanes are typically problematic when utilised as coupling partners, cubane is predicted to overcome this. Owing to the high s-character of its C–H bonds, cubane reactivity is closer to alkenyl or aryl coupling partners.[14] Tertiary alkyl systems are often “bad” coupling partners as they favour isomerisation over intended cross-coupling reactions.[164] In addition, they have an increased tendency to undergo β-hydride elimination. Again it is assumed that cubane will evade these problems, as the high ring stain of the

33 pyramidalised olefin that will be present in the cubane framework, if β-hydride elimination occurs, is energetically disfavoured.[165] Nevertheless, there have been a few recorded transition metal mediated isomerisation reactions of cubanes so this possibility cannot be fully overlooked.[88a, 89] The final obstacle cubane must overcome for effective cross-coupling is the steric constraints associated with tertiary sp3 systems. As a result of the rigidity of cubane, the 90° C–C–C bond angle present in the scaffold causes its neighbouring carbon atoms to be somewhat ‘held back’ from the reactive centre, and when compared to the C– C–C bond angle in a adamantine or tert-butyl group (~109°), the overall steric demand in cubane is theoretically reduced allowing for successful cross-coupling.[99]

Just two examples of palladium-catalysed cross-coupling reactions of cubane have been reported (Scheme 19). Firstly, a Negishi-style condensation of a zinc–cubane system 108 with benzoyl chloride[166] and secondly a Sonogashira cross-coupling reaction of an alkynylcubane 110.[167] Furthermore, extensive work by the Senge group has shown cubanes’ incompatibility as a direct coupling partner in Pd-catalysed cross-coupling reactions.[99] These findings have led to interest in further accessing alkynylcubanes as coupling partners. Alkynylcubanes were first reported in 1991.[60] Preliminary results from within the Senge group indicated these systems to be suitable Sonogashira cross-coupling partners, but further work is required to prove the general versatility of this scaffold for synthetic chemists.[168] As such, comprehensive studies involving the use of various isomeric alkynylcubanes as rigid isolators may have significant applications beyond the field of tetrapyrrole chemistry.

Scheme 19: Previous examples of cross-coupling reactions with cubane and attempted direct Pd cross-coupling reactions with cubane.

1.5.4. Merging Porphyrins with Cubanes and Triptycenes

In terms of photosynthetic reaction centre mimics the current state-of-the-art systems often incorporate fullerene or other carbon nanostructure as three-dimensional scaffolds for porphyrin arrays.[169] Other advances in the field have incorporated butadiyne-linked porphyrin nano-rings[170] or molecular wires[171] as model compounds and electron transfer systems. What these systems all share is that they are almost invariably linked by electronically conjugating groups[172] or heteroatoms.[173] Due to the lack of applicable subunits; rigid, small molecule, aliphatically-linked bisporphyrins are a class of systems which have yet to be thoroughly investigated in porphyrin chemistry.[174]

Figure 23: Cubane and benzene scaffolds.

As previously discussed, considering the geometry of cubane, it is noteworthy that the distance across the cube (the body diagonal) is 2.72 Å, which is almost equivalent to the distance across a benzene ring, i.e., 2.79 Å (Figures 2 and 23).[67-68] Eaton’s cubane bioisotere hypothesis was confirmed in a recent publication from Tsanaktsidis et. al.[10] This validation implies that, given robust substitution procedures, the cubane scaffold could conceivably be used as a (nontoxic) bioisostere for benzene. Furthermore, while the inherent flexibility of standard sp3 linker units has limited the research into aliphatically linked, multiporphyrin systems, cubane, as an essentially inflexible carbon skeleton, overcomes this problem. Similar to a more rigid adamantine,[175] cubane should also serve as a highly efficient electronic isolator – preventing bound groups from communicating to each other (i.e. conjugation) and potentially opening a unique avenue into scaffold and electron transfer chemistry.[69a]

Triptycene has a distance across its bridgehead carbons of 2.61 Å, which is also close to the 2.79 Å distance observed for benzene. Although, while this distance is close, triptycene will never be a suitable bioisostere for benzene due to its large 3D IFV. A highly desirable property of triptycene lies in the rigid linearity of the bridgehead carbons that, in addition to cubane, allow for the arrangement of substituents in not only a geometrically controlled manner but also in an electronic one that inhibits conjugation between the substituents, regardless of the three aromatic rings that are present on this molecule. This attribute was

35 further supported when the effect of sequential addition of π-bridges in the form of benzene rings to the BCO scaffold (to eventually form the triptycene complex 118) was investigated with respect to the electron transfer rates (Figure 24).[176] A 4-(pyrrolidin-1-yl)phenyl electron donor and 10-cyanoanthracen-9-yl electron acceptor were attached via alkyne linkages to the scaffold and used as the model system 115. Previous studies had proven the existence of the Marcus inverted region through the use of saturated spacer groups.[177] In contrast, the advantage of π-spacers lies in the maintained closer energetic resonance between the π-systems of the D and A groups conveying electronic communication. Unexpectedly, it was shown, by Natural Bond Orbital (NBO) analysis, that the additional π-pathways played no role in the electron transfer.[176] Furthermore, no effect was seen from the changes in the σ-system because of shifted hybridisation, despite the predictions of the photoelectron spectra.

Figure 24. Small molecule chromophores for electron transfer studies.

1.6. Applications of Rigid Linear Linkers in Electron Transfer Systems

1.6.1. Small Organic Acceptor Donor Systems

Work began in the area of electron transfer studies as early as the 1980s with small organic chromophores, and has now progressed to larger multiporphyrin complexes.[178] The majority of systems seen in literature are linked with either BCO, triptycene or BCP. Synthetic methods employed to achieve these complexes ranges from substitution reactions to transition metal catalysed cross-coupling methodologies such as Suzuki, Sonogashira and Stille reactions.

One of the first examples of an electron transfer system, modelled with a linear rigid linker, utilised BCO as the linker.[178] The [2.2.2] rods consisted of one or two BCO moieties with different chromophores attached to the bridgehead carbons. α-Naphthyl 119 and β-naphthyl

120 were used as the donor groups, while the energy acceptor units consisted of acetyl, benzoyl, or cyclohexanecarbonyl moieties (Figure 25).

Emission was observed by both the donor and the acceptor moieties upon excitation of the donor chromophore, with the ratio depending on the system.[178] The rates of intramolecular energy transfer and the decay exhibited by the bridgehead groups were determined using single-photon counting. The most rapid transmission was seen with 120b where the β- naphthyl group was the donor group while benzoyl was the most efficient acceptor. Additionally, the shorter rod length showed quicker transmission, due to the simple dipole- dipole coupling that is aided by excitation transmission through the molecule, thus indicating a correlation between distance and transmission. The molecules were synthesised over a series of steps, starting from 1,4-dichloroBCO. The two chromophores were attached sequentially, the donor moiety via an aromatic alkylation and, more challengingly, the acceptor units via lithium addition reactions.

Following the successful identification of the most efficient donor and acceptor moieties in the rigid linker system seen in 120b, the effect of inserting an aromatic ring between two BCO scaffolds was investigated.[179] α-Naphthalenyl was implemented as the donor group and an acetyl or benzoyl group acted as the acceptor groups 121. Results showed that transmission of singlet excitation proved to be less efficient compared to the previously studied rods.[179] ∆-Density determinations were employed to describe the distribution of electronic excitation in these systems and to see what was controlling energy transmission. These determinations showed that while most of the excitation energy is located in the terminal chromophores, some is distributed onto the BCO units, indicating the presence of through-bond energy transfer. Evidence also showed that energy transfer occurs mainly through-bond transfer for the short rods but when the rod is lengthened Förster through- space transfer can be observed.[179]

Several more examples of electron transfer systems that utilise linear rigid linkers have been synthesised and modelled, and these have been highlighted by us in full, in a comprehensive review.[180]

Figure 25. Naphthalene systems with various small molecule chromophores and BCO linkers.

Alternatively, different triptycene spacers of varying oxidation states were employed to make Ru2+ and Os2+ complexes 122–124, allowing the spacer to act as a redox active switching unit and enabling the quinone spacer to potentially control energy and electron transfer reactions in between two complexed metal centres (Figure 26).[181] Key steps in the synthesis of these complexes were the Diels–Alder reactions to form the triptycene scaffold and the lithium addition reactions used to connect the RuII moieties to the diformyl triptycene. The ratio between the relative emission of (Ru-bmb-Ru), (Ru-btb-Ru) and (Ru-bqb-Ru) was found to be 3.7:2:1. The highest emission was found for the bmb ligand due to the donor effect of the methoxy-substituents, while the lowest emission was seen for the bqb system as the emission is strongly quenched by the quinone moiety. Owing to the quinone’s redox active properties emission tuning can be obtained. The electronic energy level is positive enough for the quinone moiety, in the metal complex Ru-bqb-Os, to act as a quencher for the Ru-based 3CT excited state as is shown by the lack of phosphorescence seen from the Os-based 3CT level.

Figure 26. Triptycene-linked electron transfer systems.

1.6.2. Porphyrins for Use in Electron Transfer Systems

Porphyrins have been employed as more accurate photosynthetic mimics due to their similarity in structure and electronics with the chlorophyll pigments in plants[163] and due to the many advances in their synthesis, structural understanding and functionalisation.[146a, 147b, 148d, 163b] In 1984, the first porphyrin photosynthetic model with a linear rigid linker was published. The link between electron tunnelling and distance was investigated by extending the length between two chromophores with additional BCO linkers (Figure 27).[182] Due to the employment of the rigid BCO linker, the dependence of variables such as distance, orientation, and the energy gap between donor and acceptor molecules in electron-transfer processes can be more effectively studied. The electronic energy for photosynthetic model

38 compounds, linked by BCO, containing a porphyrin and quinone unit 125, was calculated through the use of a semiempirical method. The edge-to-edge distances of the BCO and biBCO systems were found to be 10 Å and 14 Å, respectively. It was predicted that in these systems hole transfer would dominate, resulting in a significant difference between the rate of decay for the forward electron transfer compared with that of the reverse electron transfer over distance. Results indicate that the symmetry of the donor and acceptor orbitals relative to the BCO linker orbitals determines the energy dependence on it to mediate the donor- acceptor interactions. Also, for every additional BCO unit added it was expected that the forward rate from the singlet excited state will slow by a factor of 1500, while the reverse rate will only decrease by a factor of 60.

Figure 27. Monoporphyrin BCO-linked systems as photosynthetic mimics.

To effectively mimic photosynthesis, the ability to control the ratio of the rates for charge separation and recombination is key in creating long-lived charge separated states, which are necessary for efficient electron transfer processes.[169b] Marcus theory was used to further understand these rates in electron transfer reactions on a series of electron donor– acceptor systems with push–pull chromophores as electron acceptors.[183] A ZnP electron donor connected via a rigid phenylene-ethynylene-phenylene (PEP)–BCO linker to different anilino-substituted multicyanobutadienes or extended tetracyanoquinodimethane analogues 126 was synthesised. First reduction potentials were obtained, and coupled with other results, showed that the extent of ZnP fluorescence quenching correlates with the strength of the electron acceptor. This finding indicates that a rational tuning of the photophysical properties by the push–pull chromophores as electron acceptors is possible. The computed Marcus curves showed that the charge-recombination kinetics in the inverted region were greatly affected by enhancing the electron–vibration couplings, due to the conformationally-fixed push–pull acceptor chromophores. X-ray crystal structure data

39 showed well-defined systems holding the acceptor and donor moieties at fixed distances with edge-to-edge distances of around 17 Å with little spectral overlap between the porphyrin and acceptor moieties. It can be deduced from this data, that if electron transfer is occurring via a through-bond mechanism the likelihood of Förster resonance energy transfer is greatly reduced.

It is of great interest to generate long-lived ion pair (IP) states via singlet radical ion pair states in order to effectively mimic efficient charge separation (CS) in a photosynthetic reaction centre (RC). Strategies based on multicomponent donor-acceptor systems that require multistep electron-transfers offer great possibilities in achieving this goal. With this in mind, conformationally constrained triads were synthesised consisting of a metal-free porphyrin (H2P), a zinc porphyrin (ZnP) and 1,4,5,8-naphthalenetetracarboximide (NIm) 127 (Figure 28).[184] The porphyrin moieties are bridged by four different aromatic spacers one of which is BCO-1,4-diylbis(1,4-phenylene). Picosecond excited-state dynamics were studied with these systems, using picosecond time-resolved transient absorption

+ spectroscopy. Results showed that long-lived ion pair states of the triads (ZnP) –(H2P)–

- (NIm) were observed upon photo-excitation via charge separation between the (H2P)* and

+ NIm followed by a hole-transfer reaction from the (H2P) to the ZnP. The rates of this hole transfer were used to determine quantum yields of the formation of long lived ion pair states (Figure 29).

Figure 28. BCO-linked porphyrin dimers.

Figure 29. Artificial photosynthetic bisporphyrin 127 models with an energy level diagram.

To further understand the factors that control the dependence of ET rates on the distance between the electron donor and the electron acceptor, a series of (ZnII–FeIII) 5,10,15,20- tetraarylmetalloporphyrin dimers, with a variety of different linkers, including BCO, were synthesised and the kinetics of their PET reactivity were measured.[185] Through the use of fluorescence lifetime measurements, energy transfer from the electronically excited state of the zinc porphyrin to the bis(imidazole)iron porphyrin cation could be determined. Results showed that when the distance was increased by 13 Å the rate of electron transfer only decreased by a factor of 165 indicating a small reduction of the electronic coupling with distance. Selective nucleophilic aromatic substitution at the para-fluorine on tetraarylporphyrins was employed in the synthesis of the molecular building blocks. This method allowed for a wide variety of systematic modifications such as type and length of spacer, metal centre, and redox-potential difference between donor and acceptor. A single zinc(II) or one iron(III) atom can be inserted after the synthesis of the symmetrical porphyrin dimer (Figure 30).

More studies involving porphyrin-based donor-bridge-acceptor (D–B–A) systems were investigated to find the triplet excited-state deactivation of a gold porphyrin (AuP) (Figure 30).[186] 1,4-DiethynylBCO was utilised as a saturated linker to explore whether electrons/electron holes can be transferred within the dimer 129 between AuP and ZnP. As a comparison 1,4-diethynylbenzene (BB) and 1,4-diethynylnaphthalene (NB) were derived as conjugated linkers. The porphyrins are separated by 19 Å, edge-to-edge, so a direct (through space) exchange mechanism was not predicted. Results showed no quenching of AuP i.e. no hole transfer, when the conjugation in the system is broken due to the linker BCO. While long-range hole transfer from AuP to ZnP occurs on the nanosecond time scale at room temperature in the dimers connected by fully conjugated bridging chromophores (NB and BB).

A similar porphyrin scaffold was used to investigate effects on the photophysical processes in the donor-bridge-acceptor (D–B–A) systems through studies of the acceptor spin state.[187] Again, 1,4-diethynylBCO, -BB and -NB were the linkers investigated. In this system, 130 FeP is acting as the acceptor while ZnP acts as the donor. The FeP is modulated from high-spin iron(II) to low-spin iron(III) by the coordination of an imidazole ligand. In similar previous systems, the high spin Fe(III)P significantly enhances the intersystem crossing in the ZnP, as the dominating deactivation pathway for the singlet excited zinc porphyrin. But this process is only a minor contribution to the quenching of the low-spin iron(III); instead the major photophysical process that is occurring is a long-range electron transfer on the picosecond time-scale allowing intersystem crossing to occur at its “normal” rate. EPR and UV-Vis measurements were used to prove the change from iron’s high spin state to low spin state upon imidazole coordination. Steady-state and time- resolved fluorescence measurements were used to measure total quenching efficiencies for the excited states of the zinc porphyrin donors.

Figure 30. Bisporphyrin BCO-linked systems as photosynthetic mimics.

Again an analogous porphyrin-based D–B–A system 131 with a 1,4-diethynylBCO spacer was investigated, this time to find out the contributions towards singlet energy transfer from

Förster and Dexter mechanisms.[188] A clear distinction between the two mechanisms can be seen when the inert BCO spacer is used. This so-called superexchange mechanism for singlet energy transfer has been shown to make a significant contribution to the energy transfer rates in several D–B–A systems, and its D–A distance as well as D–B energy gap dependencies have been studied.[189] In each system, the acceptor is a free base porphyrin and the donor consists of a zinc porphyrin with/without a coordinated pyridine ligand. Complementary results were obtained for the energy transfer processes in these BCO- bridged porphyrin D–B–A systems, with both experimental and theoretical methods. Results highlighted that the singlet energy transfer contribution was relatively similar for both Coulombic (Förster) and through-bond superexchange mechanisms and that the relative contributions do not vary with the D–A distance. The distance dependence was shown to be approximately exponential for through-bond coupling for singlet energy transfer.

1.6.3. Non-linear Triptycene Porphyrin Arrays

As was discussed in detail in Section 1.4., triptycene 1 has numerous desirable properties including an IFV,[115] three 120° electron-rich clefts between the phenyl rings of the molecule and the ability to arrange its substituents in a rigid and invariable manner (Figure 16).[114, 116] Owing to these properties, the Senge group was led to synthesise multiporphyrin triptycene- linked complexes. Both Suzuki and Sonogashira cross-coupling methods were employed to realise a new class of triptycene-linked porphyrin arrays. The trimeric porphyrin arrays, 133 and 134, were either linked directly to the three aromatic rings of triptycene or through ethynyl linkers (Scheme 20).[190]

Scheme 20: Synthesis of trisubstituted triptycene scaffolds for electron transfer studies. (i)

Pd(PPh3)4, K3PO4, DMF, 100 °C; (ii) Pd2(dba)3, AsPh3 THF, NEt3, 12 h, rt.

Starting from unsubstituted triptycene 1, 2,6,14-triiodotriptycene[191] 132 was synthesised through a nitration reaction (with separation of the two 2,6,14- and 2,7,14-regioisomers), followed by an amination and iodination reaction using Sandmeyer methodology.[192] The iodinated triptycene 132 was then used as a platform to make a series of different compounds. Subsequently, Suzuki cross-coupling with borylated porphyrins afforded the directly-linked triptycene–porphyrin complexes 133a–c. Further incorporation of an ethynyl moiety to the porphyrin allowed for Sonogashira cross-coupling with triptycene 132, resulting in an arm-extended porphyrin system 17, thus increasing the scaffolds IFV (Scheme 20).

Moreover, to expand the library of available multipigment triptycene arrays, BODIPY was attached to triptycene 132, firstly via an ethynyl bond to give complex 134b (Scheme 20). The direct attachment of BODIPY to the triptycene was then achieved through a linear synthetic strategy (Scheme 21). 4-Formylphenyl units were introduced onto triptycene 132 via a Suzuki–Miyaura[193] cross-coupling reaction to yield 2,6,14-tris(4- formylphenyl)triptycene 135. Dipyrromethane 136 was then synthesised through a condensation reaction with pyrrole, followed by an oxidation reaction.[194] Compound 136 was then converted to the BODIPY-substituted triptycene 137 (Scheme 21).

Scheme 21. Synthesis of triptycene with three DPM or BODIPY units. (i) 4-Formylphenylboronic acid, Pd(PPh3)4, Cs2CO3, THF, 70 °C, 24 h; (ii) pyrrole, InCl3, rt., 20 min; (iii) DDQ, DIPEA, BF3•OEt2, rt., 45 min.

These examples of multichromic arrays attach the chromophores to the triptycene scaffold at the aromatic units. However, there have been no examples of multiporphyrin arrays at the bridgehead carbons. Attachment of moieties at the bridgehead positions of triptycene

44 offers a unique opportunity. For example, there is already a plethora of examples of 9,10- difunctionalised triptycenes in the literature that are used in a vast array of applications from MOFs to various molecular rotors or rods.[180] But no examples exist that employ 9,10- difunctionalised triptycene as a rigid isolating scaffold for investigations in electron transfer studies.

1.7. Pd-catalysed Cross-coupling Reactions

Beginning in 1972 with the publication of Heck and Nolley’s ground-breaking paper,[195] where palladium salts were utilised to couple benzyl, aryl and styryl halides with terminal alkenes under mild conditions, the field of transition metal-catalysed C–C bond formation reactions has quickly become a reliable and facile method to access a wide range of C–C coupled products. Other notable named reactions have evolved since then including Kumada coupling[196] (1972), the Sonogashira reaction[197] (1975), Negishi reaction[198] (1977), Stille reaction[199] (1978), Suzuki–Miyaura reaction[193] (1979), Hiyama reaction[200] (1988) and the Buchwald–Hartwig amination[201] (1994). In 2010 the Nobel Prize in Chemistry was awarded “for palladium-catalysed cross-couplings in organic synthesis” to Richard Heck, Eiichi Negishi and Akira Suzuki, recognising the importance of this work.[202]

Pivotal to the work performed in this research is the Sonogashira reaction, involving the reaction of alkenyl or aryl halides or triflates with a terminal alkyne, in the presence of Pd(0)/Cu(I) and a base, thus enabling sp3-sp coupling. The reaction can proceed with/in the absence of a copper co-catalyst and both catalytic cycles can be observed in Figure 31 and Figure 32.[203]

Figure 31. Catalytic cycle for the Cu(I)-catalysed Sonogashira cross-coupling reaction.[204]

It is commonly accepted that there are four main steps in the co-catalysed Sonogashira reaction mechanism. Firstly, oxidative addition occurs with the insertion of an electrophilic carbon-heteroatom bond (such as a halogen, triflate, tosylate or phosphonate) onto the low valent palladium metal, consequently oxidising Pd(0) to Pd(II). Following this, the heteroatom is displaced by the nucleophilic component (also called transmetallation). Thirdly, trans-cis isomerisation occurs, enabling the two functionalities to be in proximity to one another. Finally, the coupled product is released via reductive elimination, resulting in the regeneration of the low valent palladium metal. The role of the Cu(I) salt is widely believed to facilitate the transfer of the alkynyl group to the Pd. Typically, the rate limiting step of the cycle is oxidative addition.[205] But a more recent report established that transmetallation was in fact the rate-limiting step,[206] as it was shown to be a Pd and Cu co- catalysed synergistic process that exhibited a first order kinetic dependence on the Pd and Cu catalysts, respectively. But even now the exact mechanism of the homogeneous Pd- catalysed and Cu co-catalysed Sonogashira reaction is not well understood and has remained a topic of debate.

Figure 32. Catalytic cycles proposed for the Cu-free Sonogashira reaction involving two different pathways for the deprotection cycle.

A frequently encountered shortcoming of the Sonogashira reaction is the formation of homocoupled alkynyl products such as Glaser coupling[207] or Hay coupling[208] due to the presence of oxidising agents, copper salts or air. These homocoupled products can be formed in large quantities and can also prove difficult to separate owing to similar chromatographic mobility as the desired products.[1] Methods have been developed to overcome this unwanted side reaction, one of which is the copper-free Sonogashira reaction.[209] DFT calculations and theoretical mechanistic studies,[210] have shown that the Pd-catalysed Cu-free Sonogashira reaction mechanism can occur via two routes: namely carbopalladation[211] or deprotonation[212] (Figure 32). Oxidative addition of the organohalide R1–X to the low valent Pd complex occurs in both mechanisms, as does the replacement of a ligand by the alkyne.

From the theoretical results it could be seen that the energy barrier for the proposed carbopalladation mechanism was very high,[210b] indicating the unlikelihood of this

47 mechanism under the reaction conditions. In contrast, the calculated Gibbs energy barriers indicated the feasibility of both cationic and anionic mechanisms for the deprotonation mechanism[213] (Figure 32). The cationic mechanism involves a L–X ligand substitution (Figure 32; cationic pathway), the alkyne is then deprotonated by an external base and subsequently reductive elimination occurs of Pd(II) to Pd(0). In contrast, deprotonation of the alkyne occurs first in the anionic pathway (Figure 32; anionic pathway), followed by L– X ligand substitution and reductive elimination step to regenerate Pd(0).

Hammett parameters[214] showed the tendency of alkynes that are connected to electron withdrawing groups (EWG) to favour the anionic mechanism, while alkynes bearing electron donating groups (EDG) favoured the cationic pathway.[213] Mechanistic studies of the Cu- free version of the Sonogashira reaction also showed the importance of the solvent, as it plays a role in stabilising the ionic intermediates of the catalytic cycle, while steric bulk decreases the stabilising ability of the solvent.[215] The ligand size also influences the catalytic activity of the Pd catalyst complexes, as reactivity is increased by the use of bulkier phosphine ligands.[216]

Objectives

Porphyrins are a unique class of compounds that are ubiquitous in nature and are amongst some of the most important cofactors, with functions in a wide variety of roles ranging from oxygen transport, electron transfer and oxidation reactions to photosynthesis.[217] On a molecular level, these effects are related to their chemical properties, namely their photochemical (energy and exciton transfer), redox (electron transfer, catalysis), and coordination properties (metal and axial ligand binding) as well as their conformational flexibility (functional control).[146a, 147b, 163e] By virtue of these properties porphyrins are of significant importance to the fine chemicals industry and are involved in an ever-expanding array of applications ranging from use as pigments and catalysts, to emerging areas such as cancer therapy, solar energy, sensors, nonlinear optics and nanomaterials. As disparate as these topics might appear, they are connected by the unique properties of the porphyrin dye and modern advances in each of the fields requires the synthesis of unsymmetrically substituted porphyrins with different substituents arranged in a very controlled and defined manner. As such the search for novel, structurally defined linking and scaffold units is an ongoing challenge in porphyrinoid chemistry[218] and serves as the motivation for this present work.

It is of major interest to mimic these processes with porphyrin assemblies, so they can be used as model compounds in electron transfer studies.[146a] Consequently, the principal goal of the present work is the synthesis of tailored multiporphyrin arrays that are electronically isolated with defined structures and regiochemical arrangements to satisfy these requirements (Figure 33). The success in this field is limited to the availability of appropriate scaffold molecules and linker units. While there is an availability of these structures, there is a notable absence of rigid, non-conjugating units with defined geometry. NRHs such as cubane and triptycene are hypothesised to act as rigid isolators, thus opening a new avenue for photosynthetic mimics.

Figure 33: Multiporphyrin array with scaffold linker.[162b]

The current state-of-the-art employs phenylene, ethynyl, ethenyl or alkane linkers to connect porphyrin units together.[219] When porphyrin arrays are constructed directly by π- conjugated linkers they exhibit significantly altered UV-Vis spectra, indicating very strong electronic coupling i.e. losing the characteristic of individual units due to delocalisation of π- electrons. Whereas, when porphyrin arrays are connected by flexible aliphatic linkers conformational control over the molecule is lost leading to incomparable results. Cubane and triptycene both contain saturated sp3 bridgehead carbons that have the innate ability to arrange their substituents in a rigid and linear manner, while simultaneously inhibiting through bond communication. For these reasons, they were chosen as the linking scaffolds in this work in order to create a new type of porphyrin array that will hopefully act as a closer photosynthetic mimic to those already observed in the literature.[174]

Cubane, in conjunction with BCP, was chosen as a scaffold to investigate controlled conformational flexibility in porphyrin dyads that are connected to it through semi-rigid amide bonds. The amide bonds allow for significant modulation of the photophysical properties in the porphyrin dyad/s through the coordination of transition metal(II) ions. By varying the distance and angles between the two chromophores it is hoped that the extent of the impact that cation coordination has on the photophysical properties of a multichromophoric systems can be investigated. As cubane and BCP are rigid and relatively inert, this allows the amide bonds to be the only variable in the system and a true measurement of their role in the conformational changes can be undertaken. Ultimately, it is hoped to bridge two porphyrin units through non-traditional BCP/cubane connectors as a test case for multichromophoric and/or electroactive systems in general. These amide- linked cubanes could open-up a new area of research into conformational analysis expanding the application and relevance of these cubanyl porphyrin systems.[220]

The final objective of this work was again to synthesise a porphyrin array but this time for application in photodynamic therapy (PDT). In this instance, how the porphyrins were connected was not the concern, but rather that steps towards water solubility were taken to allow these photoactive compounds to have a biological application. The strategy to achieve water-soluble porphyrin dyads was to introduce cationic/anionic substituents to the molecules. As examples of porphyrin dimers with application in PDT are limited in the literature, it is hoped that introducing water-soluble moieties will open up their applicability in this area.

Chapter 2. Cubane Cross-coupling and Cubane–Porphyrin Arrays

2.1. Background on the History of Cubane Functionalisation

Even 50 years after its first synthesis by Eaton et al. in 1964, the chemistry of cubane still presents a synthetic challenge and is underdeveloped; especially in terms of modern cross- coupling methodologies.[56, 221] Most of the derivatisations arise from various conventional functional group interconversions of the dimethylcubane-1,4-dicarboxylate. Current state of the art is almost solely the use of “classical” carbonyl chemistry,[2a, 98, 222] thus limiting the scope of potential cubane compounds and therefore access to advanced cubane systems. Just recently cubanes began to draw attention again due to their biological properties and their use as a benzene bioisostere.[10, 66] Our interest in cubane arose from two different origins. Firstly, we recognised the lack of useful reactions to install aryl groups directly onto the cubane core. Secondly, we were interested in utilising cubane as a rigid linker unit for applications in materials chemistry. However to achieve this, the right tools had to be developed; tools outside the “carbonyl-world”. Screening the literature revealed only four different approaches, each one with their own significant drawbacks (Scheme 22).[223]

Scheme 22. State-of-the-art of cubane-aryl coupling.

The first example of cubane-aryl systems was mentioned by Moriarty et al. in 1989 describing the radical functionalisation of cubane carboxylic acid in the presence of

[40c] Pb(OAc)4 and benzene. Even though the yield is high, this method is only suitable for a limited number of aryl groups and suffers from poor regioselectivity. Eaton’s group utilised the highly reactive cubane-1,4-diyl diradical for phenylation, in order to prepare cubane rods in “living polymerisation”-like fashion.[224] Here the major disadvantage is the missing functional group tolerance due to harsh conditions. Recently, the Baran group has shown the versatility of their newly developed redox-active ester concept, proving the robustness of their method by preparing mono- and diarylated cubanes shortly after.[100] Thus far, this approach represents the benchmark in terms of flexibility, but again only results in low yields, even for simple aryl substrates.

Our own initial approach for applying palladium-catalysed cross-coupling chemistry on cubanes, failed, unfortunately, owing to the instability of the cubane core in presence of palladium and most likely due to the very slow oxidative addition step during the catalytic cycle (two-electron process).[89, 99] With this knowledge we turned our attention to a single electron transfer mechanism (SET), circumventing the requirement of the oxidative addition of the transition metal to the cubane core. Even though the first approach via an iron- catalysed Kumada coupling proved unsuccessful, we realised the versatility of redox-active esters for cubane-aryl cross-coupling.[225] The reported yield of 25% for the iron-catalysed coupling[100] raised our curiosity to identify the key obstacles for the cubane-aryl coupling. A closer look revealed five major contributing factors i) solvent, ii) temperature, iii) concentration, iv) Ni-source, and v) ligand. The following work involving the application of SET to the cubane scaffold was performed by Dr. Stefan Bernhard.[223]

2.2. Method Development and Overview of Synthesised New Compounds

After thorough investigation, the influence of the ligand was identified as the key parameter for the cubane-aryl coupling (Scheme 23). Ligand design is a powerful tool to modulate the reactivity of transition metal catalysts and to control the reaction pathway. Ligand redox effects play an important role during the catalytic cycle.[226] Within the tested library of bipyridines, bisphosphines and tridentate ligands, the very electron-deficient 4,4`- functionalized bipyridines outperformed other ligand systems; especially the bisphosphine and rigid phenanthroline ligands. According to our findings the ideal ligand possesses some flexibility along the biaryl axis, has no substituent in the α-position to the nitrogen and carries electron withdrawing substituents in the 4,4`-positions. We hypothesise the influence of the electron-poor 4,4`-position to affect the rate determining step of the catalytic cycle, namely the reductive elimination of the cubane and aryl residue from the nickel centre. Bulky ligands and electron withdrawing substituents favour the reductive elimination due to strain release

52 and electronic stabilisation.[227] This assumption is in accordance with previous findings of cubane`s reactivity: i) the cubyl radical is rather stable[55] and is captured in the presence of Michael acceptors or radical scavengers (e.g., TEMPO); ii) the reductive elimination proceeds faster for hydrides and the according hydride-coupling product can be obtained in excellent yield.[103, 228]

Scheme 23. Contributing factors influencing cubane cross-couplings.

Realising the impediments and to test our hypothesis we investigated the scope of our new catalyst/ligand systems (Scheme 24). The coupling of very electron-rich aryl moieties proceeded in good yields of over 50% (140d, 140f, 140i). Likewise, unfunctionalised aryl groups could be coupled smoothly (140a, 140b, 140c). One remaining problem is the coupling of electronic deficient aryl substrates. A weak electron withdrawing substituent in the p-position is tolerated, but only with a significant decrease in yield (140j, 18%).

In case of very electron-poor aryl groups and heteroaromatic systems, a more electron rich ligand (4,4'-di-tert-butyl-2,2'-bipyridine, L1) proved to be beneficial in order to overcome decomposition before the coupling (140m, 140n, 140o). Generally, the yield is lower for these important classes of compounds, but we were pleased to observe the first directly linked cubane-porphyrin system (140m, 140n, 140o), a system which has not been reported before. The preparation of a suitable porphyrin-zinc species was achieved by adapting the iodine-metal exchange developed by the Osuka group.[229]

Notable is the smooth coupling of useful compound 140f, which can be subject of further functionalisation and was used for the preparation of cubane-linked porphyrin-system 148 (vide infra).

In addition, the procedure could be used to prepare difunctionalised aryl-cubane 140h from 140a using a sequence of deprotection of the methyl ester, activation of the carboxylic acid, and Ni-catalysed coupling. A major limitation right now is to overcome the sterical hindrance

53 of ortho-substituents (entry 140e, 25%) and to extend the scope for further heteroaromatic coupling partners.

With the direct-coupled cubane-aryl systems in hand we decided to go back to our initial approach of Pd-catalysed cross-coupling chemistry with cubanes.[99]

Scheme 24. Nickel-catalysed cross-coupling and substrate scope. Where L5 = 2,2'-bipyridine.

Since the direct approach proved to be unfruitful, we decided to apply an arm extension of the cubane core with alkynyl-groups, hoping that the palladium/copper-alkynyl-cubane intermediate is more stable than the directly attached counterparts. So we rolled the dice again and to our delight the initial reaction showed clean conversion of the cubane alkynyl to the substituted product (Table 4).

After optimisation of the general coupling conditions, we were interested to see if the Sonogashira reaction was tolerant to electronic changes in the aryl substituent. The presence of electron-withdrawing groups on the aryl iodides in p- or o-positions (entries 1,3,5) led to product formation in excellent yields (~90%). With the introduction of electron donating groups (entries 2, 4) a slight decrease in yield was observed (76−80%). This behaviour is in accordance with the literature, as oxidative addition of the aryl halide to the

54 active palladium species is believed to be the rate-limiting step.[230] Less reactive aryl bromides required more forcing conditions and only electron-poor aromatic systems (entries 6−8) gave product formation. Particularly product 143g may be highlighted as belonging to a new family of nitro-cubanes. The ease of attachment of two nitro-groups in the direct proximity to the alkynyl-cubane-core could give a new boost to the search for cheaper cubane-based high energy materials.[71]

Table 4. Substrate scope for Sonogashira cross-coupling with alkynyl cubane 141.

Entry X R1 R2 R3 Prod. Yield[%]

1 I CO2Et H H 143a 90

2 I NH2 H H 143b 76

3 I H H Br 143c 90

4 I H OH H 143d 79

5 I H H CHO 143e 88

6 Br H H H 143f 68

7 Br NO2 H NO2 143g 51

8 Br CN H H 143h 64

Reaction conditions: All reactions were performed at 0.1 M cubane concentration using three equivalents of aryl halide, Pd(PPh3)4 (10 mol-%) and CuI (30 mol-%) under argon for 16 h. Where applicable, yields are isolated yields after chromatography on silica gel.

The stability of alkynyl cubanes towards palladium and copper catalysis during Sonogashira-conditions is astonishing as previous findings showed that the presence of Rh(I) leads to a rearrangement of the alkynyl-cubanes to either cyclooctatetraen derivatives or tricyclooctadiene.[224] In addition, the presence of Pd on charcoal either leads to a ring- opening to cyclooctatetraen,[85] or to reduction to bicyclo[2.2.2]-octane under hydrogen atmosphere.[91] However, with this reassuring indication of cubane’s stability, we turned our interest to more delicate cubane-porphyrin systems (Scheme 25).

Scheme 25. Cubane-porphyrin systems via Sonogashira coupling.

The application of the optimised conditions for Sonogashira coupling for normal aryl substrates was not effective in the initial reaction. Next to insertion of copper into the macrocyclic porphyrin core we observed the formation of the Glaser-coupled cubane dimer, rather than the Sonogashira product.[231]

Adjustment of the conditions and utilisation of a Cu-free methodology provided us with arm- extended porphyrin−cubane systems 145a (73%) and 145b (63%) in good yields.[232] Furthermore, this approach was extended to 5,15-dibrominated-10,20-di(4- methylphenyl)porphyrin 144c, to achieve a linear porphyrin with two cubane handles on each side in 145c (51%). Even more delicate porphyrin−cubane systems were achieved by using the bisalkynyl-cubane, either as monoprotected 146a or as the bisalkynyl species 146b (Table 5).

To overcome the low solubility of the triphenylporphyrins 144d−144e, we changed to more soluble dialkylporphyrins 144f−144g, but the yield of the isolated pure compound dropped significantly. A determination of the yield from the crude reaction mixture via 1H-NMR indicated yields comparable to the corresponding arylated-porphyrins (entry 1−4). Hence, decomposition during the purification is responsible for the low yields of the dialkyl-porphyrin systems. Only the reaction with the bisalkynyl 146b achieved the product in good yields again. Due to the high copper concentration during the reaction conditions, the insertion of copper was detected as side reaction yielding porphyrin 147i. Thus far, the limitation for this kind of systems is the preparation of bisporphyrin-alkynyl-cubanes, which could not be isolated. Isolation of the pure product from the reaction mixture was not achieved, either by column chromatography on silica gel, aluminium oxide, celite, size exclusion chromatography or by iterative recrystallisation in various solvents. The crude purple solid appeared to be insoluble in most organic solvents and UV-Vis studies indicated the absence of any porphyrinoid species.

Interestingly, the instability of the material arises from the porphyrin-fragment rather than the cubane-system, since a double Sonogashira coupling of cubanyl 146b with excess of

56 iodobenzene yielded the arylated-product 147h without notable decomposition of the cubane core (entry 8).

Table 5. Sonogashira coupling of bisalkynyl-cubanes with bromoporphyrins.

Entry M R1 R2 R3 Prod. Yield [%]

1 2H Ph Ph TMS 147a[a] 83

2 Zn(II) Ph Ph TMS 147b[a] 83

3 2H Ph Ph H 147c[c] 82

4 Zn(II) Ph Ph H 147d[c] 86

5 2H n-hexyl Ph TMS 147e[b] 12

6 Zn(II) n-hexyl Ph TMS 147f[a] 16

7 2H n-hexyl Ph H 147g[d] 68

8 Coupling with iodobenzene H 147h[a] 20

Reaction conditions: 0.33 mmol of cubane 146a/146b and 0.29 mmol porphyrin 144d−144g at 65 °C

[a] [b] for 3 h with either Pd(PPh3)4 (10 mol-%, CuI (20 mol-%), THF/NEt3 (3:1, v/v) or Pd2(dba)3 (15 mol-

%), AsPh3 (30 mol-%), THF/NEt3 (3:1, v/v). Deprotection of the according TMS-protected coupling

[c] [d] product were performed with either TBAF/THF or K2CO3/MeOH.

Finally, we were able to apply the methodology presented for the synthesis of a cubane- linked porphyrin 148 (Figure 34). The synthesis was achieved with an overall yield of 20%, starting from building block 140f with deprotection of the alkynyl followed by Pd-catalysed coupling with bromoporphyrin 144e. Porphyrin 148 is the first example of a cubane-linked porphyrin array with a further synthetic handle, which represents a significant step towards using cubane-linked systems as electron transfer compounds with rigid resistor units.[162b] With a dimension of 9.68 Å from one cubane-carbon (C3) to the end of the acetylene-unit

(C18) and with all carbons in line (∡C3C11C18 = 175°) this represents an interesting new class of linker system.

Figure 34. A series of linear linkers: Cubane-linked porphyrin 148 and molecular structure of precursor 140f (thermal displacement 50%).

2.3. Synthesis of Starting Materials for Sonogashira Cross-coupling reactions

2.3.1. Synthesis of Dimethyl Cubane-1,4-dicarboxylate

The synthesis of the cubane skeleton was achieved via Tsanaktsidis’s synthesis of diester 24. Starting from cyclopentanone 18, dimethyl-cubane-1,4-dicarboxylate 24 was synthesised continuously over seven steps. Accumulatively this process requires over four hundred hours of reaction and work-up time.[61] Over the course of the past four years more than 100 g of the cubane starting material 24 was synthesised in overall yields varying from 15−20% (Scheme 26). As the cubane diester is a kinetically stable compound which offers many different synthetic avenues, it was chosen as the cubanyl intermediate to be stockpiled for its ability to be stored at room temperature under atmospheric conditions. Additionally, many of the intermediates observed in this process were black and non- crystalline but the diester 24 was obtained as colourless crystals, after chromatography with silica gel and recrystallisation from methanol.

Scheme 26. Synthesis of dimethyl-cubane-1,4-dicarboxylate 24 from cyclopentanone 18.

Key steps in the synthesis of 1,4-diestercubane 24 include the protection of the carbonyl of cyclopentanone 18 using ethylene glycol and p-toluenesulfonic acid as suitable catalyst, this was achieved in yields usually around 80% (Scheme 27). This step deviates from the original Eaton synthesis and is an important adaption as it provides regioselectivity later on in the Diels–Alder addition step. Dean−Stark conditions were employed in this reaction as water is produced in this ketalisation process and must be removed to mitigate reversal of the carbonyl protection. The 1,4-dioxaspiro[4.4]nonane product 19, a colourless oil, was then brominated using molecular bromine in 1,4-dioxane with a by-product of hydrogen bromide in yields around 53%. The dioxane solvent was essential to the reactions success as early literature published on direct halogenation indicated that the brominating agent was likely 1,4-dioxane dibromide, formed in situ.[233] This point was further validated by additional literature that showed that when other solvents such as N,N’-dimethylformamide, trimethylphosphate and acetonitrile are used with brominating agents such as pyridinium perbromide, N-bromosuccinamide or molecular bromine the desired 6,6,9-tribromo-1,4- dioxaspiro[4.4]nonane species 20 is not produced.[62b] Hence, in spite of 1,4-dioxanes carcinogenic properties it was selected for this reaction.

Scheme 27. Ketal protection and bromination by the dioxane dibromide species.

The tribrominated product 20 was then used in a Diels–Alder cycloaddition (Scheme 28). The highly reactive 2-bromocyclopentadienone ethylene ketal (6-bromo-1,4- dioxaspiro[4.4]nona-6,8-diene) 20a was generated in situ through double elimination with NaOH in methanol.[61] The Diels–Alder dimerisation is a highly endo selective, yielding the bisketal 21 in yields up to 82%.

Scheme 28. Diels–Alder [4π+2π] cycloaddition.

The Diels–Alder adduct 21 was then deprotected at room temperature using H2SO4. The subsequent dione 22 was isolated in yields over 90% and dried in a vacuum desiccator with

P2O5 for complete removal of water (Scheme 29). The beginning of the cubane skeleton is formed in the next step when a [2π+2π] cyclisation occurs irradiating the dione for 72 hours with a medium pressure mercury lamp in the presence of acid to yield an ‘open cage’ dione 23. Three days are required to properly produce the desired product due to the minor emissions, in the necessary region from 300 to 350 nm, produced by the 400 W mercury

[62b] lamp. While the reaction solution is degassed with N2 to avoid radical formation/oxidative decomposition, this process is inevitable and thus reduces the yield of the reaction and purity of the photocyclised product. The yield was usually not determined in this step and the semi-crude product was used in the next reaction.

Scheme 29. Ketal deprotection and [2+2] cycloaddition reaction.

The final cubane skeleton is formed after a Favorskii rearrangement to contract the two 5- membered rings to the cyclobutane faces. The base-driven reaction proceeds through a cyclopropanone intermediate 23b, which opens into the carboxylate anion 23d. The cubane diacid 28 is obtained after an aqueous work-up. For clarity, Scheme 30 shows the ring contraction on only one face of the dione. As purification of the cubane-1,4-dicarboxylic acid 28 is not straightforward the carboxylic acid moieties were converted into ester groups by heating the cubane in the presence of acetyl chloride and MeOH and the product was isolated in an overall yield, from cyclopentanone 18, of 45% yield.

Scheme 30. Proposed mechanism for the Favorskii rearrangement and functional group interconversion to the diester cubane 34.

2.3.2. Porphyrin Synthesis

For the synthesis of meso-substituted porphyrins a wide range of synthetic routes are available. Generally speaking, target compounds are members of the Ax- and ABCD-type series (Figure 35), which is colloquially termed the ‘porphyrin alphabet soup’.[147b] To access

[234] less symmetrical A2BC scaffolds, reactions involving MacDonald [2+2] condensations using dipyrryl subunits (for 5,15-systems)[235] are utilised followed by monosubstitution reactions involving organolithium reagents that have come to be known as Senge reaction conditions (Scheme 31).[236] This method relies on the reactivity of the porphyrin meso- position towards strong nucleophiles.[236a] In a straightforward addition/oxidation procedure many alkyl or aryl residues can be installed onto a preformed porphyrin scaffold in what has become a powerful modification technique.[236a] Halogenation reactions can then be performed to functionalise the remaining free meso-position on the porphyrin. Additionally, metalation of the porphyrin can be performed at any stage of the synthesis once the core of the porphyrin is formed.[236b]

Figure 35. Porphyrin structure with Cα, Cβ and Cm positions indicated and meso-substituted porphyrins with ABCD nomenclature.

Scheme 31. General synthetic approach to the synthesis of tetra-substituted A2BC porphyrins.

2.3.2.1. Porphyrin Syntheses Derived from 5,15-Dihexylporphyrin

Following literature procedures and the method outlined in Scheme 32, six different A2BC porphyrins, derived from 5,15-dihexyl-10-phenylporphyrin 34, were synthesised.[236b]

Scheme 32. Synthesis of six different A2BC porphyrins derived from 5,15-dihexylporphyrin.

Three brominated derivatives 5-bromo-10,20-dihexyl-15-phenylporphyrin 144f, (5-bromo- 10,20-dihexyl-15-phenylporphyrinato)zinc(II) 144q and (5-bromo-10,20-dihexyl-15- phenylporphyrinato)nickel(II) 144i were synthesised. This was followed by three iodinated derivatives: 5,15-dihexyl-10-iodo-20-phenylporphyrin 144r, (5,15-dihexyl-10-iodo-20- phenylporphyrinato)zinc(II) 144o and (5,15-dihexyl-10-iodo-20- phenylporphyrinato)nickel(II) 144j. These two differently halogenated porphyrins were synthesised to investigate their reactivity towards cross-coupling reactions with 1,4- diethynylcubane in the Sonogashira reaction. They were chosen due to their easy handling and high solubility. Additionally, different metals were inserted into the porphyrin core to alter the electronic properties of the porphyrin system for the photosynthetic mimic electronic studies. The single crystal X-ray crystallographic studies of compounds 144f and 144r were kindly performed by Dr. Keith Flanagan (Figure 36 and 37). Planarity is very evident from the crystals. Porphyrin stacking can be observed in the iodo crystal with iodine- π interactions of C−C bond distances of 3.352 Å. Hydrophobic interaction can also be observed at distances of 2.788 Å. In contrast, no hydrophobic interactions were present in the bromoporphyrin crystal and instead a π-π stacking distance of 3.375 Å was noted.

Figure 36. Molecular structure of bromoporphyrin 144f with the top view (a) and side view (b) (thermal displacement 50%).

Figure 37. Molecular structure of iodoporphyrin 144r, with the top view (a) and side view (b) (thermal displacement 50%).

To generate a library of different porphyrins for use in the Sonogashira cross-coupling reactions, several more porphyrins with different substitution patterns were produced. These included 5-bromo-10,15,20-triphenylporphyrin 144d (Scheme 33), 5-bromo-15- phenyl-10,20-di(4-methylphenyl)porphyrin 144a (Scheme 34), 5-bromo-10,20-di(pentan-3-

63 yl)-15-phenylporphyrin 152a (Scheme 35) and their zinc(II) derivatives 144e, 144j and 152b . Again, these porphyrin were synthesised via the procedure outlined in Scheme 31.[236b]

Scheme 33. Synthesis of (5-bromo-10,15,20-triphenylporphyrinato)zinc(II) 144e.

Scheme 34. Synthesis of (5-bromo-15-phenyl-10,20-bis(4-methylphenylporphyrinato)zinc(II) 144j.

Scheme 35. Synthesis of (5-bromo-10,20-di(pentan-3-yl)-15-phenylporphyrinato)zinc(II) 152b.

2.3.3. Synthesis of BODIPYs

BODIPYs are another class of pyrrolic compounds that are essentially half a porphyrin molecule. These highly fluorescent organic compounds are a part of the difluoroboraindacene family. Owing to their versatility, they have become increasingly popular for their use as fluorophores. First synthesised by Treibs and Kreuzer in 1968,[237] widespread interest in these molecules was not received until the 1980s.[238]

Since then, BODIPY-based dyes have found application in many areas including biological labelling and are known photostable substitutes for fluorescein, giving them applications in cell imaging,[239].

Due to the conjugated π-electron system present in BODIPYs, they possess an intense UV- Vis absorption at approximately 400 nm.[238b] They are synthesised from DPM precursors that can be meso-functionalised prior to BODIPY formation.[240] The reaction is a three-step one-pot synthesis, where DDQ (1 eq.) oxidises the dipyrromethane to the dipyrromethene, followed by abstraction of the protons from the pyrrolic units by NEt3, and finally formation of the BODIPY molecule upon complexation with boron trifluoride etherate, (BF3•OEt2) (Scheme 36).

Scheme 36. BODIPY synthesis.

Two different BODIPYs were synthesised for Sonogashira cross-coupling reactions. Both were prepared by the method previously described. The appropriate DPM 153 was formed pre-functionalised with the desired meso-substituent. Neat pyrrole (20 eq.) was reacted with either 4-bromobenzaldehyde or 4-ethynylbenzaldehyde at in the presence of TFA at 80 °C.

NaOH (0.1 M) was added to quench the reaction and the product was extracted with CH2Cl2 and purified via silica gel column chromatography in yields ~25%. The one-pot boron insertion reaction was then executed to form the BODIPY molecule. BODIPY 154a was synthesised in a 65% yield while BODIPY 154b was achieved in a 58% yield (Figure 38). Both compounds were obtained as fluorescent orange-green crystals.

Figure 38. BODIPY starting materials for Sonogashira cross-coupling reactions.

2.4. Cubane Cross-coupling Reactions via the Sonogashira Reaction

2.4.1. Sonogashira Reaction with 1,4-Diethynylcubane

Starting from diester 24, a four-step reaction was required to access 1,4-diethynylcubane 158 for use in Sonogashira cross-coupling reactions. This was achieved through literature procedures (Scheme 37). 1,4-Diester cubane 24 was reduced to the diol 155 using LiAlH4

(3 eq.) in THF at 0 °C and isolated in a 98% yield as white powder. LiBH4 was also implemented as a reducing agent resulting in yields up to 80% of the diol 155. The cubane diol was then oxidised to the aldehyde 156 using Swern oxidation conditions.[224] A solution of DMSO in dry CH2Cl2 was added dropwise to a solution of oxalyl chloride in CH2Cl2 that had been cooled to −78 °C under an inert atmosphere. The cubane 155 was dissolved in a mixture of dry CH2Cl2 and THF (1:6, v/v) and that too was added dropwise to the reaction mixture. The reaction was allowed to stir at −15 °C for 1.5 hours. The solution was re-cooled to −78 °C and NEt3 was added dropwise, resulting in the release of dimethyl sulfide. The reaction was then allowed to warm to room temperature and stirred for an additional ten minutes. The reaction was quenched with water and the product was extracted with CH2Cl2. Long exposure to column chromatography with silica gel resulted in a dramatic decrease in yield, therefore short flash chromatography with CH2Cl2 was applied so the crude mixture could be purified quickly, with minimal exposure. While yields up to 70% were achieved with the Swern oxidation, upon continuous repetition of the reaction it proved to be sensitive to water and especially temperature during the addition of oxalyl chloride. In an effort to find a more reliable and additionally odourless oxidation reaction, a variety of different oxidation procedures were investigated (Table 6). An oxidation with TEMPO was performed using

KBr and 5% NaHCO3 in CH2Cl2. The standard reaction conditions for this reaction require addition of TEMPO at 0 °C to a solution of the substrate-to-be-oxidised, KBr and the

[241] NaHCO3 base, followed by rigorous stirring at room temperature. The rate of stirring of the reaction mixture is important to ensure the combination of the organic and aqueous phases. The cubane diol 155 has poor solubility in CH2Cl2 at lower temperatures, so only trace amounts of the oxidized product was obtained. Literature procedures have also been reported that use EtOAc as a solvent for TEMPO oxidations.[242] Thus, the same conditions were followed with EtOAc as before but solubility was even worse for the cubane in this solvent so no product was observed.

IBX was then investigated for its ability to oxidise the alcohol 155 to the desired cubane aldehyde 156. IBX was synthesised from 2-iodobenzoic acid using 3 equivalents of oxone in water affording yields over 80% of white crystals (Scheme 38).[243] The reaction was heated to 70 °C and stirred for three hours before reducing the temperature to room temperature and stirring the reaction for a further 1.5 hours. IBX was added to a solution of

66 the cubane in EtOAc and heated to 80 °C over three hours. Even at the increased temperature the solubility of cubane was still quite poor in EtOAc so only trace amounts of the product were observed. The solvent was switched to acetone as the ketone solvent theoretically should not interfere with the oxidation. The reaction was heated to 78 °C for three quarters of an hour but no product was observed. This may have been due to the short reaction time that was employed in the literature procedure.[244] As a result, an IBX oxidation in a solvent mixture of EtOAc/acetone (4:1, v/v) was carried out at 80 °C over a period of 18 hours. None of the desired product was obtained. As the cubane aldehyde is extremely sensitive and unstable, it was decided that the extended heating time for this reaction was not conducive to high yields of the product. An additional reaction was then carried out in a solvent mixture of EtOAc/acetone (3:1, v/v) allowing the reaction to stir at 80 °C for four hours. Encouragingly, the desired product was obtained in good to moderate yields, albeit IBX impurities were present in the product so an exact yield cannot be stated. During the course of all the attempted optimisations of the cubane alcohol oxidation one thing remained constant, namely, the aldehyde product’s instability towards purification with silica gel. For this reason, the impure product was carried forward into the next reaction. Unfortunately, the IBX impurities inhibited the completion of the next reaction resulting in the need to find an alternative oxidation route.

Scheme 37. Synthesis of 1,4-diethynylcubane 158 from diester 24.

Scheme 38. IBX synthesis.[243]

Oxidation attempts were next tried with Al(Oi-Pr)3, the cubane 155 was heated to 87 °C with the oxidant in a mixture of acetone/toluene for four hours.[245] Only trace amounts of the

67 product were observed. The Parikh−Doering oxidation was then performed on several different scales as it is similar to the standard Swern conditions but utilises the sulfur trioxide

[246] pyridine complex (SO3•Pyr) instead of oxalyl chloride as the activating agent. DMSO was employed as the oxidant, while NEt3 acts as the base. Using the reagent SO3•Pyr means that milder temperatures can be employed between 0 and 20 °C. Initially the reaction was allowed to stir for one hour at 0 °C, but in this instance, over-oxidation of the cubane diol was observed and the carboxylic acid by-product was obtained, lowering the yield of the reaction significantly to 26%. Shortening the reaction time to 30 minutes allowed for an isolated yield of the aldehyde product of up-to 76% yield. Swern oxidation conditions were also revisited as over-oxidation is not a concern with this reaction. With the use of a cryostat the temperature of the reaction solution could be maintained at exactly −78 and −15 °C and this dramatically increased the reliability of the reaction as temperature and yields ranging from 76−93% were attained.

Table 6. Optimisation reaction of the oxidisation of cubane diol 155 to cubane aldehyde 156.

Reaction Conditions Solvent(s) T (°C) T (h) Yield (%)

(COCl)2 1 Swern THF/CH2Cl2 −78−15 1.5 20−75 DMSO, NEt3 TEMPO, KBr 2 TEMPO CH2Cl2 020 2 Trace NaHCO3 (5%) TEMPO, KBr 3 TEMPO EtOAc 020 2 n.d. NaHCO3 (5%) 4 IBX IBX EtOAc 80 3.25 n.d. 5 IBX IBX Acetone 80 0.45 n.d. EtOAc/ 6 IBX IBX 80 18 n.d. Acetone EtOAc/ Not isolated 7 IBX IBX 80 4 Acetone pure

i 8 Oppenauer Al(O Pr)3 Acetone/Tol 87 4 Trace

Parikh− DMSO, NEt3 9 CH2Cl2 0 1 26% Doering SO3•Py

Parikh− DMSO, NEt3 10 CH2Cl2 0 0.5 76% Doering SO3•Py

Corey–Fuchs conditions were then employed to produce the diethynyl derivative 158

[247] (Scheme 37). Firstly, carbon tetrabromide (2.1 eq.) was dissolved in CH2Cl2 in an inert atmosphere then triphenylphosphine (4.4 eq.) and was added in portions. The liberation of the bromide ions resulted in a gradual darkening of the solution from colourless to brown caused by the formation of molecular bromine. The cubane aldehyde 156 was dissolved separately in CH2Cl2 and added dropwise to the reaction mixture the solution was allowed to stir at room temperature for 45 minutes and the after purification the product was isolated in yields up to 60%. Following this, 1,4-diethynylcubane 158 was obtained upon the slow addition of n-BuLi to a solution of the dibromoalkene 157 in THF at −78 °C. The reaction stirred at this temperature for one hour before being quenched with MeOH and 5% HCl. After purification the desired product 158 was obtained as off-white crystals in 95% yield.

Copper-free Sonogashira methodologies were initially implemented to couple the free base porphyrins 144f and 144r to alkynylcubane 158 (Scheme 39). This method was chosen to avoid the potential Glaser by-product. The Sonogashira reaction was performed using a 2:1 equivalence of the bromo- and iodoporphyrins 144f and 144r, respectively, and the unprotected 1,4-diethynylcubane 158, respectively. In both cases no coupled product was observed and only recovered porphyrin starting material and dehalogenated porphyrin side product were obtained. It is hypothesised that the low reactivity/instability of the alkynylcubane may prohibit its efficient complexation with the catalyst. Owing to this reduced reactivity future experimentation should use excess cubane while limiting the more reactive porphyrin. Additionally, protection of one ethynyl proton may minimise side reactions and allow for the synthesis of an unsymmetric system. Secondly, it is hypothesised that the concentration of the reactants plays an important role in the reaction. It was thought that high concentrations are required for efficient coupling to occur, to allow for maximum exposure of the reagents to the catalyst and encourage complexation.

Scheme 39. Attempted mono- and bisporphyrin cubane-connected synthesis

2.4.2. Sonogashira Reactions with 1,4-Di[(trimethylsilyl)ethynyl]cubane

In order to increase the efficiency of the Sonogashira reaction, one ethynyl residue of cubanyl 158 was protected with trimethylsilane. n-BuLi was added dropwise to a solution of ethynylcubane at −78 °C over a period of one hour. The reaction was stirred for two hours to allow time for the reagent to generate. TMS chloride was then added and the solution was stirred for one hour. The reaction mixture was warmed to room temperature and then quenched with 2 M HCl and extracted with CH2Cl2. Column chromatography with silica gel was employed to purify the desired product with n-hexane. The polarities of the staring material 158, mono-protected product 162 and diprotected products 163 are very similar, hence a long column was required. Additionally the products are UV-inactive so a p- anisaldehyde stain was used to identify the different fractions by TLC. Similarly, cubane 162 was also obtained, via the deprotection of cubane 163 using the complex MeLi•LiBr. MeLi•LiBr was added dropwise over 20 minutes at 15 °C and the solution was stirred for an hour and then quenched with H2O. After column chromatography, the product was isolated in a 30% yield (Scheme 40).[224]

Scheme 40. Synthesis of 1-ethynyltrimethylsilane-4-ethynyl cubane 160 via a TMS deprotection with the complex MeLi•LiBr.

A copper-free Sonogashira reaction was performed with ethynylcubane 162 (Scheme 41).[232] Initially, the procedure was performed on a 1:1 equivalent basis of the cubane and porphyrin but product formation was not observed. Encouragingly, when the number of equivalents of ethynylcubane 162 was increased to 2.5, the desired coupled product 164 was observed in a 7% yield, confirming the necessity to have the cubane in excess.

Scheme 41. Cross-coupling reactions with ethynylcubane 162.

The TMS protecting group was removed using the potassium carbonate base in MeOH in order to re-access the second ethynylcubane proton for another coupling reaction, to afford the cubane-linked porphyrin 165 in an 83% yield. (Scheme 42).[149a]

Scheme 42. TMS-deprotection of compound 164 with K2CO3.

Due to the poor yields obtained via the copper-free Sonogashira reaction, it was necessary to attempt to optimise this process and investigate different Sonogashira conditions in order to improve the yield of the reaction.

2.4.3. Optimisation of the Sonogashira Reaction using Free Base Porphyrins

It can be seen in entries 1–4, in Table 7, that all reactions are copper-free Sonogashira reactions with varying conditions. All reactions were carried out with one equivalent of porphyrin and two equivalents of the cubane precursor. Previously tried conditions with cubanyl 158, Pd2(dba)3, AsPh3 and the bromoporphyrin 144f in THF/NEt3, were trialled again, but in a higher concentration, as it was found that when the solution was too dilute the main product observed was the dehalogenated porphyrin (entry 1, Table 7). The same conditions were used as in entry 1, but, with toluene as the solvent (entry 2). In both cases, an insoluble black-purple metallic solid was obtained; however, a small portion of the product did dissolve and was identified by mass spectrometry as the cubanyl bisporphyrin product 161. Unfortunately, no further reaction or analysis could be conducted due to the negligible yield. The iodoporphyrin 144r gave none of the desired products (entry 3) while reaction with compound 162 gave a 5% yield of porphyrin 164 (entry 4).

Table 7. Optimisation of the Sonogashira reaction with free base porphyrins.

X R Cat. Co-cat. Solvent T. t Result (°C) (h)

1* Br H Pd2(dba)3 AsPh3 THF/NEt3 65 16 Insoluble ppt.

2* Br H Pd2(dba)3 AsPh3 Tol/NEt3 65 16 Insoluble ppt.

3* I H Pd2(dba)3 AsPh3 THF/NEt3 65 16 n.d.

4* Br TMS Pd2(dba)3 AsPh3 THF/NEt3 65 16 A 164 5%

5* Br H Pd2(dba)3 CuI THF/NEt3 65 16 n.d.

6* Br TMS Pd2(dba)3 CuI THF/NEt3 65 16 n.d.

7* Br H Pd(PPh3)4 CuI THF/NEt3 65 4 B 165 Cu 7% C 161a

8* Br H Pd(PPh3)2Cl2 CuI THF/NEt3 20 4 B 165 Cu 12% C 161a * All experiments were performed on a 50 mg or 72 mM scale, in 1–2 mL of solvent. [a] The product was not isolated.

It was surmised that the copper-free Sonogashira coupling reaction is not activating enough to efficiently couple the two desired compounds (Table 7). As a result, in entries 5–8, standard Sonogashira methodologies were implemented using copper iodide as the co-

[248] catalyst. Both entries 5 and 6 show that no products were obtained with Pd2(dba)3 as the catalyst, but when the reaction time and catalyst were changed in entries 7 and 8 both the porphyrin monomer and dimer, 165 and 161, respectively, were observed through mass spectrometry but porphyrin dimer 161, the cubane-linker bisporphyrin product, could not be isolated. Additionally, copper insertion was observed through identification of porphyrin monomer 165 even though only 0.2 equivalence of CuI was used. Some palladium insertion was also detected but none of the desired cubanyl linked free base porphyrin was identified.

The main problem still encountered in these syntheses is the extremely low yields of the products. Initially, this was thought to be due to production of several side products, including different porphyrins and cubane-linked porphyrins, caused by copper and

72 palladium insertions, but it became apparent that the sensitivity of the cubanyl linked porphyrin systems to purification was the largest contributing factor to the poor yields observed. Decomposition of the cubanyl-porphyrin complexes, which was initially assumed to be the catalyst, could be seen on the base line of every TLC performed and every purification technique employed for the cubanyl complexes. To minimise additional products, it was decided to try cross-coupling reactions with nickel and zinc porphyrins, to prevent any unwanted metal insertion reactions. Furthermore, different purification techniques were to be investigated to prevent the decomposition of the cubanyl porphyrin complexes.

2.4.4. Optimisation of the Sonogashira Reaction using Metalloporphyrins

In all the performed reactions, one equivalent of the porphyrin was used on a 50 mg scale with 2.5 equivalents of cubane 158 or 162. The reactions were performed in 1–2 mL of THF and NEt3 (3:1, v/v) at a temperature of 65 °C (Table 8). The desired target A, cubanyl porphyrin 166, was obtained for all reactions with the bromoporphyrin 144q (entries 1–3 & 6). While in the case of the nickel porphyrins; the bromoporphyrin 144i, used in entry 4, showed the formation of bisporphyrin 169 (product C) which could not be isolated due to solubility issues, but was identified by mass spectrometry. In entry 5, coupling with the iodoporphyrin 144j resulted in no product formation and mainly decomposition of the porphyrin starting material was observed. This result may be due to the increased reactivity of this porphyrin caused by the more labile C–I bond.

In entry 1, the formation of product A was observed by TLC analysis and with mass spectrometry after a reaction time of four hours at room temperature, but, a large proportion of starting material was still present. Starting material was detected even after additional catalyst was added and more time was allowed. As a result, the reaction was repeated at a higher temperature of 65 °C (entry 2). The reaction was monitored by TLC analysis and nearly full consumption of the porphyrin starting material was observed after three hours. Following purification, a yield of 12% of product 166 was obtained. The low yield was due to the decomposition of the porphyrin product 166 upon contact with the acidic SiO2 and

Al2O3. No alternative purification method performed was successful. As a compromise, rapid purification with SiO2 and Al2O3 was carried out to isolate the pure products. An increased yield of 16% was achieved for the synthesis of product 168 in entry 3. The increase in yield is likely due the protection of one of the ethynyl bonds, increasing its stability towards purification. Interestingly, in entry 6, it can be seen that coupling with porphyrin 144e gave a much higher yield than with the n-hexyl substituted counterpart 144q. The higher yield obtained of 86% was due to the increased stability of the porphyrin product 167, although, rapid purification with flash column chromatography using aluminium oxide

73 was still required. Achieving the cubanyl porphyrin product 167 in such a high yield opens up the door for the applicability of using this method to access cubane-linked porphyrin systems and subsequently the physiochemical properties they possess.

As previously mentioned, the main issue encountered with these reactions was purification. The reaction outlined in entry 2 was performed five times and typically crude yields of 55– 70% were obtained. Many different purification methods were employed to isolate the pure cubanyl porphyrin target 166. Some of the methods employed were purification using silica, aluminium oxide grade 1 and grade 3, preparative TLC plates, size exclusion chromatography and recrystallisation without any previous column chromatography. In all cases, due to considerable decomposition, only small amounts of the cubane-linked products were obtained which inhibited further reactions to form the cubanyl bisporphyrin product.

Table 8. Optimisation of the Sonogashira reaction with metallated porphyrins.

Porphyrin Meso Cubane Cat. T (h) t Result (%) (M, X ) (R1) (R2) (°C)

[a] 1 Zn, Br n-hexyl H Pd(PPh3)4 20 rt. A 166

2 Zn, Br n-hexyl H Pd(PPh3)4 3 65 A 166 (12%)

3 Zn, Br n-hexyl TMS Pd(PPh3)2Cl2 3 65 B 168 (16%)

[b] 4 Ni, Br n-hexyl H Pd(PPh3)2Cl2 3 65 C 169

5 Ni, I n-hexyl H Pd(PPh3)2Cl2 4 rt. n.d.

6 Zn, Br Ph H Pd(PPh3)2Cl2 3 65 A 167 (86%) [a] Yield not determined. [b] Insoluble.

The conclusions drawn from these reactions were that aluminium oxide grade 1 or grade 3 are preferable over silica gel for purification, but, there is only a marginal difference. Bromoporphyrins are required for coupling to the cubane system while iodoporphyrins give unreliable results. Future work on these system will require reactions on a larger scale to compensate for the decomposition observed, especially as the yield for the second Sonogashira reaction is very low.

2.4.5. Attempted Synthesis of Bisporphyrin-connected Cubane

In an attempt to bypass the purification of the cubane-linked monoporphyrin product, reactions were performed to directly synthesise the cubane-linked zinc porphyrin dimer 170 (Scheme 43). Firstly, the iodoporphyrin 144o was used in two equivalents with one equivalent of cubane 158. The homo-coupled product was detected by mass spectrometry but it was not isolated. Secondly, one equivalent of bromoporphyrin 144q was reacted with two equivalents of cubanyl 158. Once TLC analysis indicated the formation of the cubanyl monoporphyrin product 166 a second Sonogashira reaction in situ was attempted by the addition of a further equivalent of porphyrin and catalysts, but the outcome of this reaction was degradation and loss of the monoporphyrin-connected cubane product.

Scheme 43. Attempted cubane-linked porphyrin dimer synthesis.

Another avenue that was investigated to synthesise the cubane-linked bisporphyrin 171 was executed via a second Sonogashira reaction using copper-free conditions with the crude cubanyl monoporphyrin product 167. But none of the desired cubanyl bisporphyrin product 171 was observed and no monomer starting material was recovered (Scheme 44).

Scheme 44. Cubanyl bisporphyrin synthesis.

In order to investigate whether the porphyrin or the ethynyl cubane 158 was contributing to the unsuccessful coupling a reaction was carried out using the standard Sonogashira cross- coupling conditions with iodobenzene (Scheme 45). The reaction was carried out once and purified using preparative TLC with a solvent system of n-hexane/CH2Cl2 (8:1. v/v). One of the six fractions contained the product 147h and it was isolated as white crystals in a 20% yield. The compound was stable towards purification indicating that instability of the monosubstituted cubane-linked products originated from the unsubstituted ethynyl bond. The formation of this product also suggests that the large porphyrin macrocycle may be what is hindering the successful coupling rather than just 1,4-diethynylcubane 158. Following the successful coupling with iodobenzene, a coupling reaction was performed with cubane-linked porphyrin monomer 167 and either 9-bromoanthracene 172 or BODIPY 154a (Scheme 46). It was hoped that coupling with a smaller moiety, similar in characteristics to benzene would result in a di-functionalised cubane system. Unfortunately, after standard Sonogashira cross-coupling conditions were employed none of the desired cubane-linked porphyrin-anthracene dimer 173a or the cubane-linked porphyrin−BODIPY dimer 173b could be isolated.

Scheme 45. Sonogashira cross-coupling reaction with iodobenzene.

Scheme 46. Attempted synthesis of cubane-linked porphyrin-anthracene dimer and BODIPY dimer.

A Sonogashira cross-coupling reaction was also attempted with BODIPY 154a and diethynylcubane 163 (Scheme 47). The TMS-protected ethynyl bonds of cubane 163 were deprotected in situ with 1M TBAF (3 eq.). The standard Sonogashira cross-coupling conditions were then utilised in an effort to synthesise the cubane-linked BODIPY dimer 174. The reaction was left to react for four hours at 70 °C. A black polymerised precipitate was evident in the reaction mixture along unreacted and dehalogenated starting material after the allotted reaction time.

Scheme 47. Attempted synthesis of cubane-linked BODIPY dimer

An alternative cubane linker 140f, that had been synthesised via the previously discussed

SET procedure, was deprotected using K2CO3 (7 eq.) in a solution of MeOH/THF (1:3, v/v) at room temperature over 18 hours in a 95% yield (Scheme 48). The deprotected cubane

77 linker 140p was then investigated for its use in Sonogashira cross-coupling reactions with halogenated porphyrins. Previous results had shown the reliability and stability of the subsequent cubane-linked products when an unsymmetrically substituted cubane was used as the linker, in addition to when coupling occurs only on one side of the cubane scaffold. As a result, the desired porphyrin-coupled product 148 was obtained in a 61% yield due to its stability towards purification and reduced amount of side-products. Moreover, compound 148 provides a synthetic handle that has the potential to be used in further nickel catalysed cross-coupling reactions in the future.

Scheme 48. Sonogashira cross-coupling with an alternative cubane linker.

2.5. Conclusion and Outlook

In conclusion, a modified cubane-aryl coupling is reported. After systematic investigations the key obstacle of cubane cross-coupling was identified and circumvented by altering the electron density of the transition metal through an adjusted choice of ligand. Furthermore, for the first time not only the preparation of directly linked cubane-porphyrin system was achieved but also the possibility of Sonogashira functionalisation of alkynyl arm-extended cubanes, proving the general compatibility of cubanes and Pd-catalysed cross-coupling reactions.

Several novel alkyl chain porphyrins were synthesised in this work in order to be used in Sonogashira cross-coupling reactions with the alkynyl arm-extended cubanes. After extensive optimisation attempts, of both copper-free and copper Sonogashira reactions, the desired alkynylcubane-linked bisporphyrins were not isolated. The reason for lack of success was mainly due to the stability of the alkynylcubane-linked systems, which prevented efficient purification and inhibited further reactions. However, a coupling reaction with 1,4-diethynylcubane 158 and iodobenzene did in fact result in the formation of the decoupled product 1,4-di(phenylethynyl)cubane 147h indicating that coupling with this

78 alkynylcubane is possible. Nevertheless, exhaustive attempts of multiple cross-coupling reactions with larger compounds such as halogenated-BODIPYs or porphyrins remained unattainable, suggesting that while 1,4-diethynylcubane 158 is capable of cross-coupling it is not suitable for wide spread application. Despite the problems encountered eight new alkynylcubane-linked monoporphyrin complexes 147a–g, 147i were synthesised.

Reactions involving alkyl chain porphyrins resulted in poor yields of <20%, while coupling with aromatically substituted porphyrins led to the formation of the complexes 147a and 147b in yields of >80%. Although regardless of the porphyrins substituents, all of the alkynylcubane-linked monoporphyrin complexes 147a–g, 147i formed were very sensitive to purification and were incapable of further couplings to form the target porphyrin dimers. Interestingly, a robust cross-coupling reaction was observed with methyl 4-(4'- ethynylphenyl)cubane-1-carboxylate 140p and iodoporphyrin 140f, to synthesise compound 148a which was stable towards purification on silica gel, even with long exposure. These results suggest that the instability of diethynylcubane-coupled products may stem from the symmetric alkynyl system as cubane 140p only contains one ethynyl bond and did not show any of the same adverse properties.

The new methodologies outlined in this work allow for the preparation of new classes of cubane compounds, important not only for medicinal but also for materials chemistry. Further investigations on the unique behaviour of cubane during (radical) cross-coupling conditions and the exact mechanism of decomposition are the topic of ongoing investigations. Moreover, the pursuit of a cubane-linked porphyrin dimer is still a topic of much interest. Due to the instability of 1,4-diethynylcubane 158 and the insolubility of the cubane-linked porphyrin dimers formed in this work a different avenue will need to be pursued in an effort to achieve this goal. Aromatic derivatives of cubane 140h, with directly linked phenylene groups at the cubane bridgehead carbons, could be a viable alternative to 1,4-diethynylcubane 158 in order to access the desired cubanyl porphyrin dimers, especially with the advent of the new SET methods outlined in this work. Switching to more reliable chemistry such as amide coupling may also be a viable alternative in order to access to cubane-linked porphyrin dimers.

Chapter 3. Triptycene Cross-coupling and Cubane–Porphyrin Arrays

3.1. Background on the History and Application of Triptycene as a Scaffold

Synthetic chemists are continually seeking to prepare new rigid multi-porphyrin architectures due to their potential applications as organic conducting materials, near- infrared (NIR) dyes, nonlinear optical materials and molecular wires.[147b, 249] A number of synthetic strategies have been employed to access these multi-porphyrin arrays, using synthetic strategies such as (a) connecting porphyrin units via phenylene, ethynyl, ethenyl or alkane linkers[219] or (b) connecting two or more meso–meso-linked porphyrin units via oxidative fusing reactions.[250] However, most of the constructed porphyrin arrays reported in literature encountered problems, such as poor solubility, synthetic inaccessibility and conformational heterogeneity. In meso–meso-linked porphyrin arrays, the porphyrin units are orthogonal to one another, which theoretically can cause a significant energy/charge sink. Furthermore, porphyrin arrays joined directly by π-conjugated linkers exhibit significantly altered UV-Vis spectra, indicating very strong electronic coupling i.e. loss of the characteristics of individual units due to delocalisation of π-electrons. Hence, it is necessary to design molecules, which can predictably achieve an energy- and/or electron-transfer process without causing serious electronic delocalisation and/or an energy sink.[147b, 219, 249- 250]

A key goal of much of the recent research into multiporphyrin arrays has focused on modulating the absorption profile of the porphyrins by increasing the size of the aromatic π- system. In contrast, the approach of this research deviates significantly from previous studies by examining how isolating the individual porphyrin units affects the overall photophysical properties of the system, while maintaining its structural integrity and rigidity. It thus complements the existing research by addressing an area of multiporphyrinoid chemistry that has been neglected thus far.

Triptycene consists of three benzene rings fused to a bicyclo[2.2.2]octane (BCO) skeleton, it is rigid, isolating and amenable to a wide range of chemical transformations. Due to the 120° rigid void, mono-, di-, tri-, tetra-, penta- and hexa-functionalisations of triptycene scaffolds can be achieved in a spatially defined manner.[46c] Thus, a variety of functional groups can be presented in a fixed orientation. The periphery of triptycene was successfully functionalised when multiporphyrin triptycene-linked complexes 133a–c, 137 and 134a−b were synthesised by Senge and co-workers with the purpose of conducting electron transfer studies (Figure 39). Both Suzuki and Sonogashira cross-coupling reactions were employed to realise this new class of triptycene-linked trimeric porphyrins. The three porphyrins, or

80 three BODIPYs, were either linked directly to the periphery of triptycene or via various linkers.[162b, 190, 251]

Figure 39. Synthesis of trisubstituted triptycene scaffolds.

While much work has been done on the functionalisation of the aromatic rings of triptycene,[162b, 190, 251] the attachment of chromophores at the bridgehead positions has yet to be realised. Functionalisation of the bridgehead positions of triptycene allows for the attachment of moieties in 180° linear fashion, subsequently accessing the molecular rotation observed in several other linearly substituted triptycenes (Figure 40).[180] Attaching porphyrins to the bridgehead positions in a linear fashion not only allows for compounds with well-defined spatial arrangements but additionally, the non-conjugated character of the triptycene sp3 carbon atoms prevents through-bond electronic communication and undesirable orbital overlap.[252]. Therefore, linearly bridgehead-linked triptycene porphyrin complexes could lead to compounds with the required conformation, geometry and spatial arrangement necessary for use in electron transfer studies, similar to the concept outlined already for cubane in Chapter 2.

Figure 40. Linear arrangement of bridgehead substituted triptycenes.

3.2. Synthesis of Functionalised Triptycenes

To synthesise linearly linked triptycene porphyrin dimers, the attachment of various linker groups at the triptycene bridgehead carbon atoms was investigated. The initial approach was to functionalise 9,10-dibromotriptycene 176 directly through various Pd-catalysed cross-coupling reactions. Thereby, 9,10-dibromotriptycene 176 itself was synthesised from 9,10-dibromoanthracene 175 via a Diels−Alder reaction with anthranilic acid and isopentyl nitrite in dimethoxyethane (DME).[253] Anthracene was the limiting reagent in the reaction while anthranilic acid and isopentyl nitrite were used in excess in a ratio of 1:4:18. The anthranilic acid was transformed into benzyne over the course of the reaction, which subsequently underwent a [4+2] cycloaddition with the central anthracene ring. The anthracene 175 (1 eq.) was dissolved in a mixture of isopentyl nitrite (6 eq.) and DME and heated to 90 °C. Additional isopentyl nitrite (12 eq.) in DME and anthranilic acid (4 eq.) in DME were added simultaneously dropwise to the reaction vessel over a period of 20 minutes. The rate of addition is very important for the reactions success, as the formation of the benzyne reagent is very sensitive and when the reagents are added too quickly it is not formed. The reaction was stirred for a further three hours at 90 °C, followed by 50 °C for 17 hours. The crude material was purified by flash chromatography using silica gel and n-hexane. The polarity of the anthracene starting material 175 and the triptycene product 176 are very similar so it was important to ensure full transformation of the starting material over the course of the reaction to allow for easier purification. The product was found to be colourless in solution and fluorescent blue when analysed under UV-light during TLC analysis. The desired triptycene product was isolated as a yellow powder in 25% yield. The reason for the low yield obtained is due to the instability of the benzyne reagent that is formed in situ.

The initial goal of this project was to expand the library of 9,10-difunctionalised triptycenes 177 as a platform for further functionalisation (Scheme 49). With 9,10-dibromotriptycene now in hand, various small molecules, such as boronic acids, boronic esters and 82 ethynyltrimethylsilane (TMSA) were reacted with 177 under Sonogashira and Suzuki cross- coupling conditions.[203, 230]

Scheme 49. Synthetic targets to expand the library of linkers available for 9,10-difunctionalised triptycenes.

In an attempt to access 9,10-bis(4-((trimethylsilyl)ethynyl)phenyl)triptycene 177d, dibromotriptycene 176 and the boronic ester 178a were reacted under Suzuki cross- coupling conditions (Scheme 50).[203] The two reagents were dried in a Schlenk tube, along with the base K3PO4. THF was added and the solution was saturated with argon for 5−10 minutes. The catalyst Pd(PPh3)4 was then added and the solution was stirred at 70 °C for 18 hours. The crude product was purified via silica gel column chromatography (n- hexane:CH2Cl2, 1:1, v/v). The various isolated fractions required further purification. Therefore, the solvent was removed in vacuo and the crude material was purified using a

1 preparative TLC plate (Et2O:n-hexane, 2:1, v/v). The main product gave a H NMR spectrum with only the characteristic triptycene peaks observed at 7.83 and 7.14 ppm while no signals were observed for the phenyl ring or TMS unit of the boronic ester 178a indicating that the desired product had not been formed.

Following the failed Suzuki cross-coupling reaction, another experiment was performed to synthesise a different 9,10-difunctionalised triptycene, this time using Sonogashira cross- coupling conditions.[230] 9,10-Dibromotriptycene 176 was reacted with TMSA in order to obtain 9,10-bis((trimethylsilyl)ethynyl)triptycene 177e (Scheme 50). In a Schlenk tube, the anthracene 175 was added and dried for two hours. The reagent was then dissolved in THF and NEt3 and the solution was bubbled with argon for 5−10 minutes. The catalyst

Pd(PPh3)2Cl2 was added alongside the reagent TMSA. The reaction was stirred at 70 °C 83 under argon for 18 hours. TLC analysis showed the presence of a new product with the

1 same colour and Rf as the desired product 177e. But again, H NMR and mass spectrometry did not indicate product formation but rather the retention and decomposition of the starting material 176.

Scheme 50. Suzuki and Sonogashira cross-coupling reactions with 9,10-dibromotriptycene.

Scheme 51. Organocopper cross-coupling reactions with 9,10-dibromotriptycene 176.

Owing to the lack of results produced by the Pd-catalysed cross-coupling route, a different method was sought in order to functionalise 9,10-dibromotriptycene. A recent paper by Uchiyama et al. showed the use of organocopper cross-coupling reactions to functionalise

84 the bridgehead carbon atoms of triptycene by using 9-bromotriptycene.[254] While 9- bromotriptycene was utilised in the paper the aim of this work is to obtain disubstituted triptycene products, consequently the methods outlined in the publication were applied to 9,10-dibromotriptycene 176 (Scheme 51). The reaction was performed with twice the equivalents of the reagents to functionalise both bridgehead positions of the triptycene. In order to form the organocopper reagent, triptycene 176 was reacted with n-BuLi followed by the addition of CuI. The catalyst Pd(OAc)2, along with the ligand tris(2- methoxyphenyl)phosphine and reagent methyl 4-bromobenzoate were then added to the reaction mixture and after 18 hours a brown solution was observed. Purification of the reaction mixture was attempted but numerous products were observed through TLC analysis. Due to the impurity of the sample, inconclusive 1H NMR and mass spectrometer analysis of the product 177f no further purification was carried out.

Following this, as the literature procedure[254] uses a monobrominated triptycene, the reaction was performed again with triptycene 176 but with only one equivalent of n-BuLi, CuI and the phenyl ester (Scheme 51). Initial purification and TLC analysis showed the presence of several different products, but again mass spectrometry analysis did not show the formation of the desired product 177h.

After no product was isolated while performing reactions with 9,10-dibromotriptycene 176, another precursor was sought as a starting point for functionalisation. It was decided to carry out Sonogashira and Suzuki cross-coupling reactions by functionalising 9,10- dibromoanthracene 175 prior to triptycene formation. Dibromoanthracene 175 was reacted with the same boronic acids, boronic esters and TMSA as dibromotriptycene 176, as well as halogenated benzenes 178f and 178g. The anthracene derivatives could then be transformed into triptycene through a Diels−Alder reaction with benzyne that was previously outlined.

Initially, 9,10-dibromoanthracene 175 was reacted under Suzuki cross-coupling conditions with boronic ester 178a, Pd(PPh3)4 and Na2CO3 in a solution of water, toluene and THF by stirring the reaction for 18 hours at 80 °C (Scheme 52). Unfortunately, none of the desired product 179d was observed in 1H NMR and by mass spectrometry.

Scheme 52. Functionalisation of 9,10-dibromoanthracene.

A different boronic ester 178e was then selected in an effort to synthesise 9,10-bis(4- bromophenyl)anthracene 179g (Scheme 53). Suzuki cross-coupling conditions were employed using Pd(PPh3)4 as the catalyst, K3PO4 as the base in a solution of THF and toluene. The reaction was run several times with variations of the catalyst equivalents and of the base used, in addition to varying the temperature and reaction time but the desired product was not observed under any conditions. As a result, a different route to acquire the anthracene product 179g was then taken. Therein, anthraquinone 180 was utilised instead of the dibromoanthracene 175. 1,4-Dibromobenzene 178f was reacted with n-BuLi at 0 °C in diethyl ether and the lithiated reagent was allowed to generate over 1.5 hours. Anthraquinone was then added to the reaction mixture and this was allowed to stir for a further hour. A 1 M solution of HCl was added along with tin(II) chloride in order to oxidise the compound to the anthracene product. Unfortunately, the desired product was not observed through any spectroscopic method. 1-Bromo,4-iodobenzene 178g was also used as substitute for 1,4-dibromobenzene 178f under the same conditions but again, product 179g could not be observed.

Scheme 53. Two attempted methods to synthesise 9,10-bis(4-bromophenyl)anthracene.

A Sonogashira cross-coupling reaction was then carried out between dibromoanthracene

175 and TMSA using Pd(PPh3)2Cl2 and CuI as the catalyst and co-catalyst, respectively, in a solvent mixture of THF and NEt3 (Scheme 54). The reaction mixture was heated to 70 °C and allowed to stir under argon for 24 hours. Notably, colour changes were observed over the course of the reaction from yellow to red and finally black. After purification by silica gel column chromatography using n-hexane as the eluent, the desired product 179e was collected as a fluorescent yellow solution and then isolated as an orange solid in a 95% yield after evaporation of the solvent in vacuo. With the functionalised anthracene in hand, a Diels−Alder cycloaddition reaction was carried out applying the previously outlined reaction conditions. The obtained crude mixture was purified using silica gel column chromatography, again using n-hexane as the eluent. Following this, the product 177e could be isolated as a yellow solid in a 17% yield after removal of the solvent at reduced pressure.

Scheme 54. Synthesis of TMS-protected diethynyltriptycene 177e.

In order to prepare an unsymmetrically bridgehead-functionalised triptycene, two different protecting groups were sought for 1,4-diethynyltriptycene. This was achieved in two steps from anthraquinone starting material 180 (Scheme 55).[133] First, n-BuLi was added to a solution of ethynyltriisopropylsilane (ethynyl-TIPS) in diethyl ether at 0 °C. This was left to stir under argon for 30 minutes at that temperature, followed by stirring for another hour at room temperature. Then, anthraquinone was added and the solution was stirred for one more hour. Alongside this, a ((TMS)-ethynyl)lithium solution was prepared in the same way as previously described when using the ethynyl TIPS reagent (Scheme 55). After this solution had been allowed to stir for the required 90 minutes, it was added dropwise to the anthraquinone-containing reaction vessel. The reaction was left to stir for around 18 hours under an argon atmosphere at room temperature. The solution was then quenched with aqueous NaHCO3, filtered through Celite, washed with water, extracted with CH2Cl2 and purified by silica gel column chromatography using n-hexane. The product TIPS((10-(TMS- ethynyl)anthracen-9-yl)ethynyl)silane 179i eluted as a fluorescent yellow solution as the second fraction. After recrystallisation from ethanol a yellow solid could be obtained in a 58% yield. This reaction was repeated many times throughout the course of this work with different outcomes and it was found that the rate of addition of the second lithiated reagent was important. Otherwise, a low yield was obtained when the addition was too fast, the optimum length of addition was found to be over the course of 30 minutes.

Scheme 55. Synthesis of unsymmetrically substituted triptycene 177j.

The unsymmetrically substituted anthracene 179i was then dried under high vacuum for use in the following Diels−Alder cycloaddition step (Scheme 55). Subsequently, anthracene 179i in DME and isopentyl nitrite was heated to 90 °C under argon. Isopentyl nitrite in DME and anthranilic acid in DME were added simultaneously dropwise over a 30 minutes period, forming the benzyne reagent in situ. This reaction was very sensitive, requiring slow and

88 careful addition of these reagents. The solution was allowed to stir under argon for three hours at this temperature before it was reduced to 50 °C and allowed to stir for 18 hours. The solvent was removed in vacuo and the crude product was purified using silica gel column chromatography with n-hexane as the eluent. The desired triptycene product 177i was obtained as a yellow solid in a 12% yield. The TMS-protected ethynyl unit could then be selectively deprotected by stirring the compound in ethanol at 80 °C in the presence of

KF for 1.5 hours. The solvent was then removed in vacuo and dissolved in CH2Cl2, followed by aqueous work up. The product 177j was obtained after recrystallisation from MeOH as a yellow-brown solid in yields varying from 55−90%.

3.3. Synthesis of Triptycene-linked Porphyrin Dimers

With the introduction of synthetic handles to the bridgehead positions of triptycenes in 177e and 177i, efforts were made to attach various chromophores in order to eventually investigate the effects of triptycene as a linker for electron transfer properties (Scheme 56). With the goal of synthesising various new triptycene-linked porphyrin and BODIPY dimers, ethynyltriptycene 177e was dissolved in THF under an argon atmosphere. To allow for the deprotection of the TMS protecting groups TBAF (1 M) was added dropwise and the solution was allowed to stir for five minutes at room temperature. Following this, the relevant porphyrin/BODIPY was added to the reaction mixture along with the catalyst Pd(PPh3)2Cl2, co-catalyst CuI and NEt3 as a base. The reaction mixture was saturated with argon for 5−10 minutes and then heated to 70 °C for four hours. Once TLC analysis indicated the completion of the reaction, the crude material was purified via column chromatography using silica gel. The BODIPY dimer 181a was purified using a solvent system of CH2Cl2/n- hexane (2:1, v/v). The product was eluted as a fluorescent orange solution after the elution of unreacted and dehalogenated starting materials and was isolated as pink-orange crystals in 30% yield. Similarly, the Pd-catalysed Sonogashira cross-coupling reaction was successful for the reaction with porphyrin 144q, resulting in a yield of 50% of the purple- green product 181b. Column chromatography with silica gel was carried out to purify this dimer with a solvent system of EtOAc/n-hexane (1:6, v/v). An additional reaction was carried out via a Glaser coupling with BODIPY 154b in order to extend the distance between the two BODIPY chromophores in compound 181c. Even though the same conditions as previously followed for 181a were employed, the desired product 181c could not be detected by spectroscopic techniques. The unsuccessful reaction may have been due to formation of Glaser coupling side products.

Scheme 56. Sonogashira cross-coupling reactions to form symmetric porphyrin and BODIPY triptycene-linked dimers 181b and 181a, respectively.

In an effort to synthesise an unsymmetrically substituted porphyrin dimer, triptycene 177j was utilised (Scheme 57). Firstly, a zinc porphyrin 144o was coupled to triptycene 177j using Sonogashira methods that were previously successfully employed for compound 177e. The crude mixture was then purified using silica gel column chromatography

(CH2Cl2:n-hexane, 1:1, v/v) after which 181d could be isolated as a purple solid in 60% yield. Subsequently, an in situ deprotection of the TIPS group was carried out by dropwise addition of TBAF (1 M) to a solution of the triptycene-linked porphyrin 181d. The polarity of the in situ deprotected product and TIPS protected starting material 181d were very similar so TLC analysis was unable to confirm full conversion. For this reason, it may be that the ethynyl bond was not completely deprotected before the next coupling reaction with nickel porphyrin 144i. After the TBAF deprotection step, the second Sonogashira cross-coupling reaction was carried out with the nickel porphyrin 144i under the same reaction conditions

90 as before for 181d. Following this, the reaction mixture was purified using silica gel column chromatography (n-hexane:EtOAc, 6:1, v/v) and a dark purple solid 181e was isolated in 8% yield. Moreover, the crystal structure of the product 181e could be obtained, confirming a linear porphyrin–triptycene–porphyrin axis which can be seen in more detail in Figures 42–44 in Section 3.5.

Scheme 57. Sonogashira cross-coupling reactions to form triptycene substituted porphyrin monomer 181d and an unsymmetrically substituted triptycene-linked porphyrin dimer 181e.

The unsymmetrically substituted triptycene 177j was utilised for several other cross- coupling reactions, including one with the dibrominated porphyrin 182b (Scheme 58). This allowed access to a triptycene−porphyrin−triptycene complex 181f. The reaction was carried out using standard Sonogashira cross-coupling conditions and purified with silica gel column chromatography using n-hexane/CH2Cl2 as the eluent (1:1, v/v). The product was isolated as the second fraction as a green solid in 19% yield. A major side-product of the reaction was the Glaser product, i.e. dimerisation of two ethynyltriptycene molecules 177j. The coupling reaction was initially carried out with dibromo-hexylporphyrin 144u and -diphenylporphyrin 144v but both porphyrins had extremely low solubility (Figure 41). Following this, porphyrin 182b was selected for subsequent Sonogashira cross-coupling reactions.

Figure 41. Dibromo-substituted porphyrins 144u and 144v.

Scheme 58. Synthesis of triptycene−porphyrin−triptycene complex 181f.

Triptycene 177j was also used in a reaction with 2,6,14-triiodotriptycene 132 in order to prepare the quadruple triptycene scaffold 183 but the reaction was unsuccessful (Scheme 59). The same Sonogashira reaction conditions that had been used so far, as for 181e, were implemented. One reason could be the triptycene 177j that was used in the reaction could have been partly decomposed, as it is unstable when kept deprotected for prolonged periods of time, even when stored below 0 °C. An additional and equally contributing factor is the physical consistency of the triptycene starting material used, as deprotection can often result in an oily solid indicating the presence of impurities as triptycene 177j should be a solid. This reaction was only attempted once so it may prove more viable with further optimisation in the future, such as variations in the equivalence of triptycene 177j used in the reaction.

Scheme 59. Attempted synthesis of quadruple triptycene scaffold.

Finally, triptycene 177j was used in a cross-coupling reaction with BCP 184 (Scheme 60). Unfortunately, none of the desired product 185 was observed, but there is still potential for a successful reaction upon optimisation of the reaction conditions such as employing copper-free Sonogashira cross-coupling conditions.

Scheme 60. Attempted synthesis of triptycene−BCP−triptycene linker 185.

3.4. Towards the Synthesis of Benzene-linked Porphyrin Dimers

As a comparison for later electronic transfer studies with isolated cubane-linked and triptycene-linked porphyrin systems, conjugated benzene-linked derivatives were synthesised over a number of steps starting from terephthalaldehyde 186. 1,4- Diethynylbenzene 188 was synthesised according to the literature in an overall yield of 65% (Scheme 61).[255] Corey–Fuchs reaction conditions[247] were employed and the alkene 187 was isolated in a 65% yield after a reaction with triphenylphosphine and carbon tetrabromide. The ethynylbenzene 188 was then obtained after reaction with n-BuLi. n-BuLi was added dropwise at −78 °C to the reaction mixture and was then allowed to stir for 1.5

93 hours, followed by 18 hours of stirring at room temperature. The yellow solution turned to a green-black colour after the addition. Usual purification with column chromatography was followed to afford the diethynylbenzene 188 in a 90% yield. In order to synthesise the unsymmetric bisporphyrin systems, 1-ethynyl-4-(TMS)benzene 189 was prepared. Three different bases were used to selectively mono-protect ethynylbenzene 188 with a TMS group each on a 200 mg scale (Scheme 61). A bulkier base was trialled as in order to be more selective for a single proton and overcome the formation of the high-yielding diprotected side-product. Consequently, it was found that the original method using t-BuLi still gave the highest yield of 48% and thus was used going forward with both the cubane and benzene monoprotection reactions.[256]

Scheme 61. 1,4-Diethynylbenzene 188 and 1-ethynyl-4-(trimethylsilylethynyl)benzene 189 synthesis.

A copper-free Sonogashira reaction with bromoporphyrin 144f on a 90 mg scale afforded the benzene-linked porphyrin product 190 in a 64% yield (Scheme 62). After a TMS deprotection reaction with K2CO3, benzene-linked porphyrin 191 was obtained in a 69% yield.

Scheme 62. Synthesis of ethynylbenzene porphyrin 191. 94

The final reaction to prepare the benzene-linked porphyrin dimer 192 with the benzene- linked monoporphyrin 191 was performed on a 35 mg scale and size-exclusion chromatography was used to purify the crude material, but no product was observed in any of the fractions by mass spectrometry (Scheme 63). This reaction will be performed again in the future and further investigation into the purification and stability of the porphyrin dimer complex 192 will be conducted.

Scheme 63. Attempted benzene-linked porphyrin dimer 192 synthesis.

A series of reactions was performed with 1,4-diethynylbenzene 188. Several different Sonogashira cross-coupling reactions were performed to find the optimum reaction conditions that could then be applied to the triptycene/cubane scaffold (as seen in Chapter 2). There were many conclusions drawn from the reactions conducted. For example, when the reaction solution was too dilute (e.g. 720 μM in 4 mL of solvent verses 2 mL) the main product observed was the dehalogenated porphyrin. Additionally, when toluene was used as a solvent no cross-coupled product was observed. The main deduction drawn was that the original conditions, using brominated porphyrins and a 3:1 ratio of a THF/NEt3 solvent system, gave the best results.

Table 9. Test reactions towards the synthesis of ethynylbenzene porphyrins.

Porph. Cat.[a] Solv. Ratio T. t 188 Result (mg) (mL) (v/v) (°C). (h) (Eq.) 2H–Br A Tol/NEt3 5:1 35 18 1 n.d. 1 (125) (39)

[c] 2H–I A Tol/NEt3 3:1 40 18 1 n.d. 2 (100) (12) 2H–I A THF/NEt3 3:1 65 18 2 n.d. 3 (100) (28) [b] 2H–Br A THF/NEt3 3:1 75 18 2 193+191 4 (250) (12) Zn–Br B THF/NEt3 3:1 70 4 1 n.d. 5 (50) (1.33) Zn–I B THF/NEt3 3:1 70 4 1 n.d. 6 (38) (1.33) Zn–Br B THF/NEt3 3:1 70 4 2.5 n.d. 7 (30) (1.33) [b] Zn–I A THF/NEt3 3:1 70 18 1 194 8 (38) (1.33)

[a] Catalyst A: Pd2(dba)3 (15 mol-%), AsPh3 (2 eq.); catalyst B: Pd(PPh3)2Cl2 (15 mol-%), CuI (30 mol- %). [b]Product was detected through mass spectrometry. [c]An additional aliquot of catalyst was added to the reaction and it was allowed to stir for an additional 18 hours.

A different route to synthesise a benzene-linked dimer was then taken (Scheme 64). The functionalities of the benzene linker and the porphyrin were inverted, as it was suspected that homo-coupling of the 1,4-diethynylbenzene 188 in previous reactions reduced the amount of linker available for coupling with the halogenated porphyrin as this a regular occurrence in standard Sonogashira cross-coupling reactions.[197, 257] To overcome this, excess ethynylporphyrin was utilised in a coupling reaction with 1,4-diiodobenzene 196. TMS-protected porphyrin 195 was deprotected in situ with 1 M TBAF, Sonogashira cross- coupling conditions were then employed and the reaction was allowed to stir at 70 °C for four hours, at which point TLC analysis indicated full consumption of the starting material and a green solution was observed. The crude material was purified by column chromatography using silica gel (n-hexane:CH2Cl2, 3:1, v/v). Unfortunately, none of the desired product 194 was isolated in a pure and significant amount.

Scheme 64. An alternative synthetic route to synthesise the porphyrin dimer 194.

In a further effort to obtain an ethynlbenzene-linked porphyrin dimer to be used for electron transfer comparison studies with triptycene and cubane derivatives, alternative synthetic routes were undertaken (Scheme 65). Firstly, the porphyrin 197 was coupled with 1-bromo- 4-ethynylbenzene 198 but the subsequent products 199 and 199a had low solubility and similar polarities leading to very difficult separation, therefore another synthetic route was chosen. Porphyrin 202 was coupled with the ethynyl bond of the benzene reagent 198, instead of the brominated functionality, as less side products were likely to form with this route (Scheme 66). Reagent 198 was deprotected in situ using 1 M TBAF and the subsequent reagent was coupled with the iodoporphyrin 202 using standard Sonogashira cross-coupling conditions. The compound was successfully purified using silica gel column chromatography (n-hexane: CH2Cl2, 2:1, v/v) and isolated as a green-purple solid in a 78% yield. The nickel ethynylporphyrin 201 was synthesised from iodoporphyrin 200, as a pink solid in an 88% yield. This allowed for a final cross-coupling reaction to be performed between the two reagents 201 and 204. Previously applied Sonogashira cross-coupling conditions were implemented but after purification and analysis none of the desired product 204 was observed.

Scheme 65. Attempted synthesis of ethynylbenzene-linked porphyrin.

Scheme 66. An alternative synthetic route to synthesise porphyrin dimer 204.

Owing to the functionalisation of the bridgehead positions of triptycene with ethynyl bonds various triptycene-linked dimers could be synthesised 181a, 181b and 181d–181f. Single crystal X-ray structures were obtained for triptycene 177j and the unsymmetric dimer 181e which gave further insight into the chemical structures of these triptycene-linked complexes.

3.5. Single Crystal X-ray Structures

Throughout this project we obtained two crystals structures for triptycene 177j and dimer 181e the data for which was collected and refined by Dr. Keith Flanagan. Single crystal X- ray structural analysis revealed several interesting non-covalent interactions between the synthesised triptycene molecules. The crystal structure of triptycene 177j highlights the linear arrangement of the two ethynyl bonds within the molecule (Figure 42a). In addition, an interesting stacking pattern is observed with this crystal structure. Two of the three faces of triptycene rings show π-stacking interaction with two other triptycene molecules (Figure 42c), while the third face of triptycene exhibits C–H···π interactions between the isopropyl group of the TIPS of one triptycene molecule and phenyl moiety of another, at distances of 2.840 Å (Figure 42b). It can be seen that aromatic moieties of the triptycene molecules are

98 not fully overlapping and as a result are arranged as a slip plane. While the usual π- interactions observed are parallel displaced, these are weaker interactions than a full π- overlap. But, in a multi π-system where numerous slip plane interactions are present, collectively these interaction are stronger than a limited number of full π-overlap interactions.[258] After looking at the interlocked pattern observed in the crystal packing of triptycene, it can be understood what first attracted chemists to the idea that triptycene could be used as a molecular rotor or gear.[3]

Figure 42. Single crystal X-ray structures of triptycene 177j. (a) Side view of triptycene 177j; (b) Top view of triptycene 177j, showing C–H···π interactions between the TIPS group of one molecule and the third pocket of another triptycene molecule. The ethynyltriptycene connected to the TIPS-group is omitted for clarity; (c) Top-view of triptycene molecules π-stacking, with two additional triptycene molecules (thermal displacement 50%). Hydrogens atoms and some silyl protecting groups are omitted for clarity. Orange atoms represent Si, grey atoms represent C and white/green atoms represent H.

The single crystal X-ray structure analysis of the dimer 181e also revealed numerous interactions between the porphyrin–triptycene–porphyrin molecules and confirmed the conformation of the two porphyrin subunits with respect to one another (Figure 44). While the ethynyl bonds around the triptycene are linear, the porphyrin planes are arranged with a dihedral angle of 66.84° between them. The two different porphyrin faces of dimer 181e can be seen in Figure 43a and 43b with either the zinc porphyrin tilted forward or the nickel porphyrin tilted forward, respectively. The zinc porphyrin displays a π-stacking interactions with other zinc porphyrins that can be observed at distances of 3.377 Å (Figure 43c). While the nickel porphyrin π-stacks with other nickel porphyrins at a distances of 3.299 Å (Figure 45a). The triptycene-phenyl ring can be seen π-stacking simultaneously with the phenyl ring

99 of the zinc and nickel porphyrins, and can be observed at distances of ~ 3.4 Å (Figure 45b). It is noteworthy, that no π-stacking was observed between the zinc and nickel porphyrins.

Figure 43. Single crystal X-ray structure of triptycene-linked zinc-nickel porphyrin dimer 181e; (a) Top-view of zinc porphyrin; (b) Top-view of nickel porphyrin (thermal displacement 50%). Hydrogens atoms are omitted for clarity; (c) Side-view of triptycene-linked zinc-nickel porphyrin dimer 181e showing the π-stacking interactions of the zinc porphyrins (thermal displacement 50%). Hydrogens atoms are omitted for clarity.

100

Figure 44. Overview of intermolecular interactions between the triptycene-linked zinc-nickel porphyrin dimer molecules 181e (thermal displacement 50%). Hydrogens atoms are omitted for clarity.

Figure 45. Overview of intermolecular interactions between triptycene-linked zinc-nickel porphyrin dimer molecules 181e. (a) Side view of nickel porphyrin, showing π-stacking interactions between the nickel porphyrins; (b) Top view of zinc-nickel dimer 181e showing intermolecular interactions between both porphyrin phenyl rings and the triptycene aromatic face (thermal displacement 50%). Hydrogens atoms are omitted for clarity.

101

The single crystal X-ray structure analysis of the triptycene 177j. and the dimer 181e revealed the apparent linearity and rigidity of the triptycene linker. While the ethynyl bonds around the triptycene are linear, the porphyrin planes in the dimer 181e are arranged with a dihedral angle of 66.84° between them. This dihedral angle indicates that through-space electron/energy transfer should theoretically be possible as an orthogonal geometry is avoided between the neighbouring porphyrin chromophores.[249c]

3.6. Photophysical Studies of Triptycene Complexes

The photophysical properties of the five new triptycene-liked porphyrin/BODIPY compounds 181a, 181b and 181d–181f were investigated in preliminary studies for their suitability as model compounds for electron transfer in the context of artificial photosynthesis. For this, initially, the UV-Vis spectra of the symmetric zinc dimer 181b, the unsymmetric dimer 181e and the zinc monomer 181d were recorded in CHCl3 (Figure 46a). Similar Soret band positions were seen for all three compounds at 431, 432 and 433 nm, respectively. A similar Soret band for both the dimer and monomer 181b and 181d suggests that the two subunits in the dimer were not connected through efficient electronic conjugation/delocalisation. Two Q bands were observed in all cases, which is characteristic of zinc porphyrins due to the increased symmetry in the compound as compared to the free base derivatives, which ordinarily feature four Q bands.[259] While the Q bands observed for the zinc dimer and monomer, 181b and 181d had low full-width at half-maxima (FWHM) of ~26 nm i.e. a narrow band, the zinc-nickel dimer 181e exhibited a high FWHM of ~50 nm, potentially caused by overlapping of the nickel and zinc porphyrin Q bands. The overlapping of the Q bands could be due to the modulation of frontier molecular orbitals. The difference in the FWHM of the Q bands of the symmetric and unsymmetric dimers, 181b and 181e, respectively, is nearly double, which is a significant difference in comparison to the Soret bands that exhibit FWHM of 10 nm for the two dimers and 8 nm for the monomer.

In the emission spectrum, the fluorescence of the zinc-nickel dimer 181e is quenched as compared to the dimer and monomer, 181b and 181d (Figure 46b). The quenched fluorescence of the zinc-nickel dimer 181e may be due to the electronic deficient nature of the nickel porphyrin and the difference in its triplet energy state.[260] In this instance, the excited zinc porphyrin is acting as electron donor while the electronically inactive nickel porphyrin is acting as the acceptor. An electron/energy transfer is occurring between the two porphyrins, therefore, when the molecule is excited at the wavelength of the zinc porphyrin, the fluorescence emission ordinarily observed for the zinc porphyrin does not occur as the energy has been transferred to the nickel porphyrin, which subsequently quenches the fluorescence. The reason for the reduced fluorescence exhibited when the nickel porphyrin is present is due to the heavy atom effect that is characteristic of nickel.[260a]

102

When an atom with a high atomic number, such as nickel, is bound to an excited molecular entity, the rate of the spin-forbidden process is enhanced, resulting in drastically reduced emission.[261] These initial results suggest that an energy/electron transfer is taking place, as the strong emission of the excited zinc porphyrin that was observed for the zinc monomer 181d and dimer 181b does not occur when connected to the nickel porphyrin. The zinc porphyrin emission is not seen as the energy is instead transferred to the nickel porphyrin and subsequently quenched. Due to the differently substituted meso-positions present on all the porphyrins for the three compounds 181d, 181e and 181b, the emission spectra observed have signals that are unequal in intensity, which is characteristic of unsymmetric porphyrin units. In contrast, the emission spectra seen for a symmetric porphyrin like H2TPP has two signals of equal intensity.[262] As previously mentioned, the zinc monomer and dimer, 181d and 181b, show similar fluorescence emissions. This is due to the presence of the non-conjugated triptycene linker in the dimer 181b, causing the two porphyrins to be electronically isolated and akin in environment to the porphyrin of the zinc monomer 181d.

103

Figure 46. (a) UV-Vis of symmetric and unsymmetric triptycene porphyrin dimers 181b and 181e, and a triptycene porphyrin monomer 181d in CHCl3; (b) Emission spectrum of symmetric and unsymmetric triptycene-linked porphyrin dimers 181b and 181e, and a triptycene-linked porphyrin monomer 181d in CHCl3 (all compounds were excited at λ = 430 nm).

To effectively demonstrate the significance and purpose of utilising triptycene as a rigid linear linker, the UV-Vis and fluorescence emission spectra of various different dimers, connected via different linker groups, were recorded (Figures 48 and 49). Both the triptycene-linked zinc porphyrin dimer 181b and BODIPY dimer 181a were compared with a meso–meso-linked dimer 193 and butadiene-linked dimer 194, whose synthesises will be discussed further in Chapter 5 (Figure 47). The UV-Vis spectra seen in Figure 48 highlight

104 the diversity in properties of the four compounds under investigation. The triptycene-linked zinc dimer 181b, previously discussed, has a Soret band at 433 nm and two Q bands at 564 and 610 nm, respectively, while the triptycene-linked BODIPY dimer 181a has a band at 367 nm and another more intense band at 506 nm. In both these instances, the compounds show clear and distinct signals due to the absence of any unwanted electronic delocalisation. The BODIPY unit was selected as a means for comparison with the triptycene-linked porphyrin dimers as they exhibit high fluorescence emissions.[263]

Owing to the orthogonal orientation of the two porphyrins, distinct signals were also observed for the meso–meso-linked dimer 193. The effects of an orthogonal arrangement of the two subunits meso–meso-linked porphyrin dimers is well-known in the literature[249c] and results in two distinct Soret bands seen at 418 and 455 nm, as each porphyrin is electronically isolated due to the lack of orbital overlap between the two units. In stark contrast, the linearly linked butadiene dimer 194 maintains conjugation between the two porphyrin units resulting in very strong electronic coupling i.e. losing the characteristic of individual units due to delocalisation of π-electrons.[264] The Soret bands of the four different porphyrins in the butadiene-linked dimer 194 are barely distinguishable from one another at 422, 439, 447 and 480 nm, due to the electronic coupling present in the system. This effect is even more notable for the dimers Q bands which cannot all be distinguished.

The difference between these four compounds is further seen in the fluorescence emission spectra (Figure 49). The most notable distinction between the compounds is in the emission values observed. The BODIPY dimer 181a shows the lowest emission of the four compounds with only one emission band seen at 527 nm. The symmetric meso–meso- linked dimer 193 exhibits a similar emission at 617 nm with two equally intense bands (a characteristic of symmetric porphyrins). The nearly similar λmax of both these dimers reinforces the lack of extended π-conjugation or communication that is present in these molecules. Comparatively, the zinc dimer 181b has an increased emission seen at 617 nm, while the butadiene-linked dimer 194 distinguishes itself again with a notably enhanced emission at 688 nm. This is again due to the orbital overlap between the two moieties resulting in an enhanced emission. The λmax of butadiene-linked dimer 194 is red-shifted 71 and 75 nm with respect to the zinc porphyrin dimer 181b and the meso–meso-linked dimer 193. The difference in emission is potentially due to the combined effect of the four electron- withdrawing methyl ester groups of the porphyrins of 194, but more likely due to the extended conjugation present between the two linearly-linked porphyrin units. Extended conjugation causes increased planarity and conjugation in the system, along with a reduction in the HOMO-LUMO gap resulting in the red-shifted spectra.[249c] In contrast to the porphyrin emission spectra, BODIPYs exhibit a blue-shifted spectrum with lower Stokes

105

Shift values, enabling a different frequency of light to be accessed. Overall, there is a significant difference in the HOMO-LUMO gap of the three porphyrin dimers. The meso– meso-linked dimer 193 has the largest gap of 2.21 eV, followed by the triptycene-linked dimer 181b with a gap of 2.03 eV and the butadiene-linked dimer 194 with 1.87 eV. These values give an initial indication of the extent of π-conjugation occurring within these systems, as the more conjugation present in the system the lower the HOMO-LUMO gap.

Figure 47. Various porphyrin/BODIPY dimers with different linker groups.

Figure 48. UV-Vis spectra of various porphyrin/BODIPY dimers with different linker groups in CHCl3.

106

Figure 49. Fluorescence emission spectrum of various porphyrin/BODIPY dimers with different linker groups in CHCl3. The excitation for compounds 181b 181a 193 194 was λ = 433, 370, 455 and 450 nm, respectively.

Figure 50. UV-Vis of the zinc porphyrin dimer 181b vs. the triptycene dimer 181f in CHCl3.

107

Figure 51. Emission spectra of the zinc porphyrin dimer 181b vs. the triptycene dimer 181f in CHCl3 (compound 181b and 181f were excited at λ = 433 and 436 nm, respectively).

The difference in the UV-Vis and fluorescence spectra was also investigated for the porphyrin-triptycene-porphyrin dimer 181b and the triptycene-porphyrin-triptycene dimer 181f. The triptycene dimer 181f shows a slight red-shift in the UV-Vis spectrum of 3, 9 and 6 nm for the three bands, respectively, potentially from the combined effect of the two ethynyl moieties and the electron withdrawing CO2Me groups (Figure 50). The red-shift was reversed for the emission spectrum, as a 5 nm hypsochromic shift of the porphyrin dimer 181b can be seen with respect to the triptycene dimer 181f (Figure 51). A reduced intensity is also observed for the porphyrin dimer 181b potentially due to π-π interactions between the porphyrin units of two dimer molecules.

3.7. Conclusion and Outlook

In conclusion, a variety of Sonogashira and Suzuki cross-coupling reactions were performed with 9,10-dibromotriptycene and various small molecules, in an effort to expand the library of available 9,10-difunctionalised triptycenes 177 as a platform for further functionalisation. Although, all attempts of functionalising triptycene 176 were unfruitful, due to the inactivity of the bridgehead carbons. As a result, an alternative synthetic route was then devised to functionalise 9,10-dibromoanthracene 175 prior to the formation of triptycene. Initial attempts to functionalise 175 were also unsuccessful until the synthesis of 9,10-di[(trimethylsilyl)ethynyl]ethynylanthracene 179e was achieved via Sonogashira cross-

108 coupling methods with TMSA. Consequently, the subsequent triptycene 177e enabled opportunities for further functionalisation with the newly introduced ethynyl synthetic handle.

Unlike 1,4-diethynylcubane 158, 9,10-diethynyltriptycene 179e proved to be stable towards cross-coupling reactions with various halogenated chromophores such as BODIPY and porphyrins, thus, enabling access to a series of porphyrin/BODIPY dimers. Several of the triptycene dimers synthesised were symmetric i.e. had the same two chromophores attached. Moreover, the synthesis of one unsymmetric porphyrin dimer 181e and a triptycene dimer 181f was also achieved. A single crystal X-ray structure was obtained for the unsymmetrically-substituted porphyrin dimer 181e, which highlighted the evident linearity in the system and arrangement of the porphyrins at a dihedral angle of 66.84° with respect to one another. Various π-π stacking interactions were also observed between the zinc porphyrins, the nickel porphyrins and a triptycene face with a porphyrin phenyl ring. UV-Vis and fluorescence studies of five of the synthesised complexes were conducted to give preliminary results of triptycenes applicability as a rigid isolating scaffold in electron transfer studies. The results of these measurements showcased the isolating properties of the triptycene scaffold, made apparent when compared to other linker groups such as butadiene- and meso–meso-linked complexes 194 and 193, respectively, where either the individual characteristic of the porphyrin units were diminished due to delocalisation of π- electrons or chemically isolated due to the inherent orthogonality in the system. Preliminary evidence of electron transfer was also observed with the unsymmetric zinc-nickel complex 181e as a drastically reduced emission of the excited zinc porphyrin was observed, when excited at wavelength corresponding to the zinc porphyrin, owing to electron transfer from the zinc porphyrin unit to the nickel porphyrin unit.

Initial studies have shown the promise of triptycene as a linker for electron transfer studies, and an expanded library could lead to further interesting results where the linearity, hybridisation and accessibility of triptycene can be further highlighted. While 9,10- diethynyltriptycene offers an initial stepping stone into obtaining these chromophoric arrays, further investigation into alternative functional groups, such as boronic esters and aromatic units, to functionalise triptycene need to be conducted to widen the scope of this unit as a viable scaffold.

109

Chapter 4. Bridging and Conformational Control of Porphyrin Units through Non-Traditional Rigid Scaffolds

4.1. Background on the Application of Cubane/BCP amide linkers

A straightforward strategy for avoiding any undesirable overlapping of the π-systems in porphyrin array may be to attach two porphyrin skeletons through non-traditional rigid scaffolds such as bicyclo[1.1.1]pentane (BCP) or cubane. This concept was already observed in Chapter 3 where triptycene was used as a linker for porphyrin dimers, and demonstrated the ability to maintain electronically distinct porphyrin environments observed through UV-Vis and emission spectrums. These saturated entities are transparent to UV- Vis light and exhibit specific three-dimensional (3D) arrangements of the bridgehead carbons, which can arrange chromophore units in a rigid linear fashion, minimising electron delocalisation and conjugation potentially solving the previously outlined drawback.[2a, 265]

1,4-Disubstituted cubane is a well-known bioisostere, as previously discussed in Chapter 1 and 2, of para-substituted phenylene rings due to the similar distance across the cube body diagonal of 2.72 Å vs. 2.79 Å for a benzene ring.[10, 66] BCP is the smallest member of the cyclic alkane family and exhibits the shortest non-bonded distance between bridgehead carbon atoms of 1.85 Å, which is closer in similarity to ethyne (1.20 Å) also reported as a bioisostere in the literature.[2b] The 3D, compact, electronically isolating, and saturated structures of cubane and BCP enable them to avoid undesirable π–π stacking which may lead to improved solubility of chromophoric arrays. Despite their advantageous well-defined dimensionalities and rigid-rod geometries the chemistry of these moieties is only beginning to be developed, especially in terms of functionalisation or C–H-activation at the bridgehead carbons.[180, 266]

BCP and cubane are transparent to UV-Vis light and most often their application is restricted to bioisosteres[267] and crystal engineering as was evident in the comprehensive review that our group carried out.[180, 268] The structural pre-organisation and high thermal stability of these compounds make them attractive candidates to link two chromophoric units; which is the central goal of this work.[180, 269] The limited use of these non-traditional scaffolds is due to perceived complex synthetic procedures and limited commercial supply chain of precursors. In addition, appending rigid sp3 linkers as connectors between two chromophoric units can still be synthetically quite demanding.

Recent synthetic developments by Baran’s, Aggarwal’s, and Senge’s (as outlined in Chapter 2) groups, include methods based on decarboxylative sp3 C−C coupling to functionalise the bridgehead carbons of cubane and BCP.[102, 223, 270] Knochel and co- workers have also reported an efficient method to synthesise 1,3-bisaryl substituted

110

BCPs.[271] Additionally, amide bonds have been introduced at the bridgehead carbons of cubane and BCP,[98, 272] although thus far, most of these moieties reported in the literature were used as bioisosteres[267] and in crystal engineering.[268] Yet, these amide-linked compounds also have the potential to be utilised as molecular building blocks. Moreover, due to their structures, cubane/BCP could be implemented as rigid scaffolds linking two chromophores with amide bonds, while simultaneously providing a synthetic handle for molecular recognition of small molecules or ions through coordination at the amide bonds.

The amide bond is crucially important as one of the main chemical linkages found in biologically and pharmaceutically active compounds.[273] Amide bonds exhibit a planar trans configuration of the N−H and C=O moieties and undergo very little rotation or twisting around the bond due to amido-imido tautomerisation. The semi-rigid nature of amide bonds enables them to have conformational control over the molecular architecture of the compound they are part of through hydrogen bonds and through the coordination of metal ions. As cubane and BCP are rigid and relatively inert, this allows the amide bonds to be the only variable in the system and a true measurement of their role in the conformational changes can be undertaken. Herein, the first successful synthesis of porphyrin dimers is reported that utilises either BCP or cubane as a rigid linear scaffold (Figure 52). Consequently, a library of BCP/cubane porphyrin arrays have been synthesised and characterised, resulting in examples of some of the largest non-polymeric structures available for cubane and BCP. The BCP compounds shown in this chapter are thanks to Dr. Nitika Grover who synthesised them in a collaboration on this project.

Figure 52. Schematic representation of target porphyrin arrays.

The utilisation of semi-rigid amide bonds for the attachment of a porphyrin skeleton to a rigid scaffold introduces a controlled conformational flexibility into porphyrin dyad/s. This could enable significant modulation of the photophysical properties in the porphyrin dyad/s through the coordination of transition metal(II) ions to the amide bonds. By varying the distance and angles between the two chromophores it is hoped that the extent of the impact that cation coordination has on the photophysical properties of a multi-chromophoric

111 systems can be investigated. Hence, in this work the very first successful attempt to bridge two porphyrin units through non-traditional BCP/cubane connectors is narrated as a test case for multichromophoric and/or electroactive systems in general.

There were two main strategies employed in order to synthesise the amide-linked BCP/cubane porphyrin dimers. The first strategy was to connect the cubane and porphyrin moieties through Pd-catalysed cross-coupling reactions, while the second strategy connected these two groups by an amide condensation reaction. The first reaction pathway involves an initial amide condensation reaction between cubane 28 and either ortho/meta/para iodo or ethynyl aniline (Scheme 67). A Sonogashira or Suzuki cross- coupling reaction can then be employed to couple the cubane moiety with the porphyrin. Conversely, the second reaction pathway first employs a Pd-catalysed cross-coupling reaction with either ortho/meta/para iodo or ethynyl aniline 207a–f and the corresponding iodo, borylated or ethynylporphyrin, 202, 208 and 209, respectively (Scheme 68). The aminoporphyrin is then reacted with dicarboxylic acid cubane 28 in an amide condensation reaction in order to access the amide-linked cubane porphyrin dimers.

Scheme 67. Reterosynthetic pathway 1, towards accessing the cubane-linked porphyrin dimers.

112

Scheme 68. Reterosynthetic pathway 2, towards accessing the cubane-linked porphyrin dimers.

4.2.1. Porphyrin Synthesis

Several different porphyrins were synthesised in order to access the desired amide-linked dimers (Scheme 69). The iodoporphyrin 202 was used as the starting material for all the required transformations. Initially, for the first reaction pathway with Pd-catalysis in the key coupling step, porphyrins 202, 208 and 209 were synthesised through standard methods previously outlined in Chapter 2. The iodoporphyrin 202 was transformed into the TMS- protected ethynylporphyrin 210 via a Sonogashira cross-coupling reaction. TMSA 178b was coupled with iodoporphyrin 202 using Pd(PPh3)2Cl2 as the catalyst, CuI as a co-catalyst and

NEt3 as the base. The reaction was then stirred in THF at 70 °C for three hours. Subsequently, the compound was purified using flash chromatography with silica gel and a

2:1, n-hexane:CH2Cl2 solvent system, to afford the product 210 as a purple-green powder in a 67% yield. The ethynyl proton of porphyrin 210 was then deprotected using a 1 M TBAF solution in THF under argon, to give the product 209 as a purple-green solid in a 80% yield.

Furthermore, iodoporphyrin 202 was also utilised in several Suzuki cross-coupling reactions. The porphyrin 202 was borylated in a 91% yield using pinacolborane 178h,

Pd(PPh3)2Cl2 as the catalyst and NEt3 as the base in 1,2-DCE. The reaction was allowed to stir at 90 °C for three hours (Scheme 69). Following this, the compound was purified under gravity filtration through a short silica gel column with a 2:1, n-hexane:CH2Cl2 solvent

113 system, to give the borylated product 208 as a bright pink powder. Similarly, another Suzuki cross-coupling reaction was performed to synthesise the porphyrin 211 as a pink solid in a 70% yield. Iodoporphyrin 202 was coupled with the borylated compound 178a using

Pd(PPh3)2Cl2 as the catalyst and K3PO4 as the base. The reaction was stirred for three hours in THF at 70 °C. Afterwards, the ethynyl bond of porphyrin 211 was then deprotected using 1 M TBAF in THF at room temperature and to give the isolated product 212 in a 99% yield.

Scheme 69. Synthesis of various porphyrins derived from iodoporphyrin 202.

Likewise, for use in synthetic route 2 where an amide condensation reaction is the key step, aminoporphyrins 213a, 213b, 213c and 213d were synthesised using previously described Sonogashira cross-coupling conditions (Scheme 70). Iodoporphyrin 202 was coupled with either 4/3/2-ethynylaniline 207a−c, resulting in a para-, meta- or ortho-aminoporphyrin, 213a, 213b or 213c, respectively. Subsequently, as polar amino groups are known to interact strongly with silica, due to its partial acidity, aluminium oxide grade (III) was used

114 to purify these porphyrins by column chromatography (CH2Cl2:n-hexane, 2:1). In spite of this, the meta- and ortho-aminoporphyrin derivatives, 213b and 213c, respectively, could be purified with silica gel due to their reduced affinity for it. The reduced affinity of the porphyrins may be due to the non-linear angle of the meta- and ortho-amino groups which decreases the amount of intermolecular bonding occurring with the silica particles. All the porphyrins appeared as green crystals with yields ranging from 77% to 92%. In contrast, the arm-extended aminoporphyrin 213d, required an additional step in its synthesis. Firstly, iodoporphyrin 202 was coupled with the pinacol ester 178a under previously described Suzuki cross-coupling conditions, and after purification the target porphyrin 214 was isolated in a 95% yield as pink-purple crystals. Following this, porphyrin 214 was then coupled with 4-ethynylaniline 207a under the same conditions as the previous aminoporphyrins. An expected, a lower yield of 58% was achieved for the arm-extended product 213d due to the lower reactivity of the para-halogenated benzene ring of 214 compared with the meso-position of the porphyrin 202.[274]

Scheme 70. Synthesis of ethynyl-linked aminoporphyrins for use in synthetic pathway 2.

115

The “shortest-arm” aminoporphyrin 213e was synthesised via a different route. Starting from pyrrole 215, it was reacted with benzaldehyde 216, in a 1:1 ratio, in propionic acid at 150 oC for 30 minutes to form 5,10,15,20-tetraphenylporphyrin 217 (Scheme 71). Once the reaction mixture was allowed to cool to room temperature MeOH was added and the desired porphyrin 217 precipitated out of the black solution. The purple crystals obtained were washed with additional MeOH and hot water and isolated in a 20% yield. One of the phenyl rings was then nitrated at room temperature with sodium nitrate in TFA, the reaction was allowed to stir for exactly three minutes before the solution was poured on to a solution of

[275] ice-cold water. The product was extracted from the aqueous layer with CH2Cl2 and neutralised with NaHCO3. The reaction was limited to a modest scale, as, while a respectable 65% yield was obtained for 200 mg of TPP 217, an incremental decrease in yield was observed as this reaction was scaled up. In order to reduce the nitroxyl group to the desired amino group, porphyrin 218 was dissolved in a neat solution of HCl with tin(II) chloride. The reaction was stirred at 65 oC for one hour and quenched by its slow addition to ice-water followed by neutralisation with NH3OH. The desired aminoporphyrin 213e was isolated, after filtration through Al2O3 and recrystallisation from MeOH, as purple crystals in an 81% yield or an overall 10% yield. The zinc(II) derivative was also obtained by reacting the aminoporphyrin 213e in the presence of zinc acetate in MeOH and CHCl3.The resulting product 213f had drastically reduced solubility in chlorinated solvents and as a result was difficult to purify while not losing product. It was isolated in a yield of 55% after an aqueous wash with NaHCO3.

Scheme 71. Synthesis of the “shortest-arm” aminoporphyrins used in this project.

In order to access the ortho- and meta- derivatives of the “shortest-arm dimer” Suzuki cross- coupling reactions were conducted with the borylated porphyrin 208 and either 3-/2-

116 iodoaniline 207d and 207e (Scheme 72). In both cases Pd(PPh3)4 was utilised as the catalyst and K3PO4 as the base in THF. Unfortunately, none of the desired products 213g and 213h were obtained. It is suggested that a more classical approach to obtain these porphyrins via [3+1] condensations may give more successful results.

Scheme 72. Alternative route to synthesise meta- and ortho- aminoporphyrins 213g and 213h.

4.2.2. Cubane Synthesis

Cubane diester 24 was used to access the cubane starting materials necessary for the amide coupling reactions. Depending on the equivalence of NaOH used i.e. 0.9 or 4, and the temperature employed the cubane diester 24 was converted to either to the mono- or dicarboxylic acid, 76 and 28, respectively (Scheme 73). The monocarboxylic acid 76 was synthesised in a 73% yield when 0.9 equivalents of NaOH was added to it in a solution of MeOH. The reaction was allowed to stir for 18 hours at room temperature and was isolated as white crystals after an aqueous work up and abstraction with a CH2Cl2/MeOH mixture. On the other hand, the dicarboxylic acid was obtained when the cubane diester 24 was reacted with four equivalents of NaOH and stirred at a temperature of 80 °C for four hours. The white crystals were isolated when the solution was acidified to <3 pH with concentrated HCl causing the cubane crystals to precipitate from of solution and be obtained in an 87% yield.

117

Scheme 73. FGI of diester 24 to either dicarboxylic acid 28 or monocarboxylic acid 76.

Table 10. Test C−H activation reactions for the amide coupling of cubane 76 with porphyrin 213e.

Cubane Activating Base Solv. T. t. Yield (eq.) Agent (eq.) (h) (°C ) (%)

1 2.0 DIC (1.3) DMAP THF 48 65 43

[a] 2 2.0 ECF (1.3) NEt3 CHCl3 120 75 31

3 6.0 DIC (1.3) DMAP THF 48 65 37

4 3.0 HATU/HOAt DIPEA DMF 24 25 79 (1.3/1.3)

[a]ECF = Ethylchloroformate.

The cubane carboxylic acids were then used as the starting point for classical amide bond formation in order to access the porphyrin monomers and dimers.[276] Various conditions were employed for carbonyl activation-based amide coupling at the carboxylic acid of cubane with substituted aryl amines.

The results of this investigation found that the use of HATU/HOAt[277] as an activating agent in presence of DIPEA furnished the cubane amide coupled products in the highest yields; for example, 79% for product 220. However, the use of other activating agents such as

118

[278] [279] DIC and ethylchloroformate in presence of NEt3 or DMAP also resulted in the formation of product 220, albeit in lower yields of 43 and 31%, respectively (Table 10).

The substrate scope was investigated incorporating various combinations of structural motifs such as ortho-/meta-/para-substituted aniline and porphyrin units (Scheme 74 and 76). The cubane aniline-coupled products were all synthesised in the same manner (Scheme 74). Either the mono- or dicarboxylic acid cubane was dissolved in DMF, in a dry microwave vial. HATU, HOAt and DIPEA were all added under an argon atmosphere and the reaction mixture was allowed to stir at room temperature. After 30 minutes the colourless solution becomes yellow and the relevant aniline is added with another aliquot of DMF. The reaction was then stirred for 24 hours under a continuous argon atmosphere. Attempts to purify the crude cubane compounds 221–226 via column chromatography using silica gel were mostly unfruitful, due to degradation of the product on silica gel. Recrystallisation from

CH2Cl2 proved very effective in removing any remaining aniline and other impurities affording pure white crystals in yields ranging from 31% to 67%.

Scheme 74. Amide coupling of cubane moieties 76 & 28 with substrate scope.

While mono- and dianiline coupled cubane products were achieved for the para- and meta- derivatives, all attempts to obtain ortho-substituted aniline cubanes 227–230 proved unfruitful (Figure 53). 2-Ethynylaniline and 2-iodoaniline were both employed in a series of conditions to achieve both the mono- and dicoupled products. Initial experiments were

119 conducted with the ethynyl aniline, firstly with the optimised HATU conditions but none of the desired product was observed. Following the lack of success, the carboxylic acid cubane was reacted with oxalyl chloride (2.2 eq.) in CH2Cl2 (with a drop of DMF), the reaction was allowed to stir for two hours at 40 °C after the careful addition of oxalyl chloride at 0 °C. NEt3 (4 eq.) was added dropwise to the solution once it was re-cooled to 0 °C. Upon addition of the aniline, effervescence was observed and then the reaction was allowed to stir for an additional for two hours at 40 °C. After purification, none of the desired coupled- product was observed. Owing to the proximity of the ethynyl bond to the cubane scaffold, it was hypothesised that a rearrangement reaction might be occurring. To overcome this, reactions with the 2-iodoaniline derivative were carried out. Both HATU and oxalyl chloride conditions were applied to the mono- and dicarboxylic acid cubane 76 and 28, respectively; but again none of the desired product was observed. Due to the apparent problems with this substrate no further efforts were expended to obtain the ortho-substituted aniline cubane products.

Figure 53. ortho-Functionalised synthetic targets.

With the functionalised cubane derivatives in hand after the first step in synthetic route one, Pd-catalysed cross-coupling reactions were employed to attach the porphyrin moieties (Scheme 75). Unfortunately, these reactions showed little or no success. Reactions with the monoaniline cubanes 221 and 222 were successful, but any and all reactions with the difunctionalised cubanes 223–226 did not yield usable amounts of any of the desired products.

Iodoaniline 221 was coupled with ethynylporphyrin 209 using standard Sonogashira cross- coupling conditions. The resulting reaction mixture was filtered through silica, initially with

CH2Cl2 to remove any left-over porphyrin starting material or side-products. The target cubane-linked porphyrin 231 was obtained after switching to a solvent system of CH2Cl2 and 5% MeOH. The product 231 was obtained as a green solid in a 75% yield. Iodoaniline 221 was also coupled with porphyrin 212 using standard Sonogashira cross-coupling conditions. The same purification procedure was applied and the target cubane-linked porphyrin 232 was obtained as a green solid in a 56% yield.

120

Di(iodoaniline) cubane 223 was used in several reactions with various porphyrins to access the different cubane-linked porphyrin dimers. Firstly, it was coupled with the borylated porphyrin 208 in order to achieve the porphyrin dimer 234. Standard Suzuki cross-coupling conditions were employed. The desired dimer was observed by mass spectrometry but was not isolated in a usable amount. Cubane 223 was also used in a coupling reaction with ethynylporphyrin 209 using standard Sonogashira cross-coupling conditions, but none of the desired cubane-linked porphyrin dimer 239 was observed. Alternatively, cubane 223 was also used in a coupling reaction with porphyrin 212 using standard Sonogashira cross- coupling conditions. After purification, only a trace amount of the desired product 240 was observed through mass spectrometry.

Cubane 225 was also used in a Sonogashira cross-coupling reaction with iodoporphyrin 202. TLC analysis indicated that no product formation had occurred, which was collaborated by mass spectrometry results, hence, no further purification or analysis of this reaction was carried out. Evidently, synthetic route 1 is not a feasible route to access the cubane-linked porphyrin dimers. Consequently, the second synthetic route was then investigated.

Scheme 75. Implemenatation of synthetic pathway 1.

121

Due to the unsuitability of synthetic route 1, it was decided to try another avenue towards accessing the desired cubane-linked porphyrin dimers. As the amide condensation reactions at the carboxylic acid functionality of cubane showed fruitful results it was decided to attempt it in the last and most complicated step in the reaction i.e. synthetic route 2. The scope of amide coupling with amine-substituted porphyrins was explored (Scheme 76). The amide coupling of amine-substituted porphyrins 213e and 213f with 4-(methoxycarbonyl)- cubane-1-carboxylic acid 76 to yield cubane-appended porphyrins 220 and 233, gave yields of 79 and 52%, respectively; whereas the reaction of cubane-1,4-biscarboxylic acid 28 with porphyrins (213e, 213f, 213b and 213c) resulted in access to the very first porphyrin- cubane-porphyrin arrays (234–235, 237–238). Additionally, an unsymmetrically-substituted porphyrin dimer 237 was synthesised. The cubane ester moiety of the cubane-linked porphyrin monomers, 220 and 233, was transformed to the carboxylic acid using an optimised procedure with the LiOH base in a solution of H2O, MeOH and THF, to yield the carboxylic acid adducts 220a and 233a in quantitative yields. It was imperative that the amount of water in the reaction exceeded the amount of MeOH, as otherwise full conversion to the desired carboxylic acid was not obtained. Some hydrolysis of the amide bond was also observed if the equivalence of LiOH employed was too high, evidenced by the pink porphyrin with reduced polarity observed by TLC analysis. This functional group interconversion then allowed for a second amide coupling to occur. Cubane-linked porphyrin monomer 220a was subjected to HATU amide coupling conditions with the zinc aminoporphyrin 213f to yield the unsymmetric dimer 236 in a 55% yield. This robust reaction demonstrated considerable scope, from the utilisation of small benzene rings to large porphyrin systems with varying lengths and at different angles with respect to the cubane plane.

Excitingly, these are the first examples of cubane-linked porphyrin dimers and while many different variations of these dimers were obtained, unfortunately, two of the original target para-substituted dimers 239 and 240 could not be accessed in sufficient yields and purity (Figure 54). As previously discussed, synthetic route 1 was not a viable way to access these dimers but owing to the success of synthetic route two, several different attempts of various amide-couplings were investigated further. The optimised HATU amide coupling conditions were utilised for the coupling of the aminoporphyrins 213a and 213d with dicarboxylic acid cubane 28, but an isolatable amount of product was not obtained (Scheme 76). Alternatively, the cubane-linked monoporphyrin 232 was converted into the carboxylic acid derivative 232a using the optimised LiOH conditions in 97% yield (Scheme 77). The monomer 232a was then subjected to an additional amide coupling reaction with aminoporphyrin 213d but none of the desired porphyrin dimer 240 was observed. Owing to time constraints, the synthesis of these cubane-linked dimers was investigated no further.

122

Scheme 76. Implementation of synthetic pathway 2.

123

Figure 54. Inaccessible para-linked porphyrin dimers.

Scheme 77. Towards the synthesis of cubane-linked porphyrin dimer 240.

4.2.3. BCP Synthesis – conducted by Dr. Nitika Grover

Successful functionalisation of cubane motivated us to next attempt functionalisation of the BCP using the same amide coupling method. However, initial attempts to synthesise the required BCP building blocks using HATU/HOAt, EDC or DIC were unsuccessful. This may be due to unstable BCP intermediates capable of undergoing rearrangement to result in ring-opened BCP moieties. Thus, a different method was employed to access the BCP scaffold. The BCP carboxylic acid 241b was reacted with (COCl)2/NEt3, followed by the desired amine at room temperature but with limited success, as the desired product was detected only in a small amounts by 1H NMR spectroscopy and mass spectrometry. To overcome this, the reaction temperature was increased to 40 C for both steps resulting in

124 a significant increase in product yield (49–77%) (Scheme 78). The crude reaction mixture was purified via recrystallisation using a small amount of CH2Cl2 and ample n-hexane to access the white powdered compounds. The synthesised BCP-based building blocks 242– 249 were further subjected to Pd-catalysed cross-coupling reactions to yield porphyrin-BCP conjugates (250–259), as shown in Scheme 79. Unfortunately, amide coupling with ortho- substituted anilines was unsuccessful. Several synthetic attempts were made but in each case degradation of precursor was observed. The ineffective ortho-substituted aniline coupling may be caused by the amine and iodo/ethynyl units’ proximity, enabling H-bonding interactions between the moieties, ultimately reducing the basicity and reactivity of the amine. On the other hand, the cubane porphyrin dimer 238 was accessed in a 50% yield due to the replacement of the small H-bonding moieties with a porphyrin, preventing the reduced amine basicity.

Pd-catalysed cross-coupling reactions are versatile and straightforward approaches for porphyrins to form carbon–carbon bonds with a wide range of functionalities[147b, 173a, 280] In contrast to cubane,[99] the BCP ring is more tolerant towards Pd-catalysed coupling reactions.[281] The very first BCP-porphyrins 250, 253, 256 and 258 were synthesised through Suzuki−Miyaura cross-coupling reactions with borylated porphyrin 208[117] and BCPs 243, 245, 247 and 249, respectively. Whereas, the porphyrin arrays 251 and 252 were synthesised using Sonogashira cross-coupling reactions of porphyrin 209 and porphyrin 212 and BCP 242.

Scheme 78. Amide coupling at bridgehead carbon of BCP moiety and substrate scope.

125

Scheme 79. Pd-catalysed coupling reaction and substrate scope of BCP-linked porphyrins.

126

Sonogashira reactions did not proceed well for the synthesis of meta-derivatives 254, 255 and 259, while a copper-free modified Sonogashira reaction of iodoporphyrin[172c] 202 with BCP 245 and 249 proceeded well to access the porphyrin-BCP arrays 254 and 259, respectively.

4.3. Single Crystal X-ray Analysis

The structure of compounds [5-(2'-aminophenylacetylene)-10,20-bis(4-methylphenyl)-15- phenylporphyrinato]zinc(II) 213c, cubane 221, BCP 243, 345, 248, 251, 252, dimethyl bicyclo[1.1.1]pentane-1,3-dicarboxylate and were determined using single crystal X-ray diffraction analysis.* Structural parameter tables and refinement details (Table 11 and 12) are provided in Chapter 6. * X-ray crystal structure determinations were again thanks to, and performed by Dr. Keith J. Flanagan.

Figure 55. (a) Molecular structure of ortho-substituted aminoporphyrin 213c with all non-hydrogen atoms labelled (thermal displacement 50%); (b) Moiety packing shown with labels omitted; (c) Intermolecular head-to-tail interaction between the Zn(II) metal of porphyrin unit 213c (acceptor) and

N donor atom of NH2 moiety.

The molecular structure of ortho-substituted aminoporphyrin 213c can be seen in its single crystal X-Ray structure (Figure 55a). The crystal packing of this molecule can also be observed in Figure 55b. Interestingly, the crystal structure of porphyrin 213c illustrates a specific orientation of the amino group due to the intermolecular bonding occurring between it and the metal centre of another porphyrin. A head-to-tail interaction can be seen between the Zn(II) metal of one porphyrin unit 213c (acceptor) the N donor atom of NH2 moiety of

127 another porphyrin molecule at a distance of 2.23 Å (Figure 55c). Moreover, this effect can be seen in the 1H NMR of the ortho-substituted aminoporphyrin 213c, as the amino group can be observed at a resonance of −0.58 ppm (Figure 56 & 57). This signal is shifted drastically up-field when compared to the 1H NMR of the para-substituted aminoporphyrin 213a which shows the amino signal at 5.01 ppm (Figure 56 & 57). The para-substituted aminoporphyrin 213a has no opportunity to interact with the zinc centre of an additional porphyrin due to the inherent π–π stacking of porphyrins that does not orientate the amino group near another porphyrin centre. As the ortho-substituted amino group is coordinating to the metal centre of another porphyrin it is experiencing the shielding effect of the porphyrin ring-current, thus causing the up-field shift.[282] In addition the aromatic proton adjacent to the amino group of porphyrin 213c also experiences the shielding effect of the ring current, as evident by its signal seen at 3.20 ppm. This signal is again greatly shifted up-field in comparison to the para-substituted porphyrin 213a which has two proton signals observed at 6.77 ppm. The difference in signals observed for the remaining protons on the amino ring are not as drastic for the two porphyrins 213a and 213c. Porphyrin 213c has the last three signals at 6.22, 6.53 and 6.87 ppm, while the last two protons for porphyrin 213a can be found at 7.73 ppm in a multiplet with another phenyl rings protons.

Figure 56. Structures of para-substituted aminoporphyrin 213a and ortho-substituted aminoporphyrin 213c.

128

1 Figure 57. Comparison of (a) H NMR spectra of ortho-substituted aminoporphyrin 213c in CDCl3 and (b) 1H NMR spectra of para-substituted aminoporphyrin 213a in d-THF.

Figure 58. Comparison of (a) 1H NMR spectra of cubane 221 in d-DMSO and (b) 1H NMR spectra of BCP 243 in CDCl3.

129

An example of the 1H NMR spectra observed for the cubane/BCP linker units can be seen in Figure 58. The spectra of cubane 221 (Figure 58a) and BCP 243 (Figure 58b) show signals at 3.33 and 3.42 ppm for the six cubane protons, while it can be seen that BCP has only one peak representing its six CH2 protons at 2.38 ppm. Furthermore, the crystal structure of cubane scaffold 221 illustrates two types of intermolecular non-covalent interactions (Figure. 59). The structure exhibits a head-to-tail N1–H···O1=C interaction at a distance of 2.853 Å with an angle of 175.4°. Furthermore, the iodo-atom at the para-position of the phenylene moiety exhibits a head-to-tail halogen bond interaction with O(2)=C of the ester group with a distance of 3.074 Å and an angle of 170.1°. The observed halogen and hydrogen bond interactions are nearly orthogonal to each other. Interestingly, the combined and repetitive intermolecular halogen and hydrogen bond interactions result in a supramolecular 3D network between the cubane molecules directed by the substituents at 1,4-bridgehead positions. There are only a few reports available in the literature on oxygen- iodine interactions,[283] but the specific 3D-orientation of cubane potentially favours this unique packing pattern enabling it to access this uncommon interaction.

Figure 59. (a) Molecular structure of cubane 221 in the crystal. (b) Molecular arrangement of compound 221 in the crystal shows the non-covalent interactions between N1–H···O1=C and C=O2···I1.

Figure 60. (a) Molecular structure of compound 243 in the crystal. (b) Molecular arrangement of compound 243 in the crystal shows the non-covalent interaction between N1–H···O1.

130

Similarly, the crystal structures of BCP 243, 245 and 248 exhibit non-covalent interactions between amide N–H donors and C=O acceptors within the crystal lattices. The nature of these interactions are dependent upon the substitution pattern at the phenylene ring. The crystal structure of para-substituted BCP compound 243 reveals repetitive head-to-head N1–H···O1=C hydrogen bond interactions at distances of 2.970 Å (Figure. 60). In contrast to 243, the crystal structure of meta-substituted BCP 245 exhibits head-to-tail N1–H···O2=C interactions at distances of 3.061 Å, leading to the formation of a non-covalently attached inversion-centred dimer. Similarly, the crystal structure of bis-meta-substituted BCP 248 shows head-to-tail interactions at distances of 2.910 Å and forms a supramolecular 3D network/array.

Figure 61. (a) Molecular structure of compound 252 in the crystal. (b) Intermolecular head-to-tail interaction between the Zn metal of the porphyrin unit (acceptor) and the C=O donor moiety of the amide bond.

In nature, the 3D structures of proteins and other biomolecules are controlled using H- bonding interactions between trans N–H and C=O moieties of amino acids and these 3D architectures are responsible for their specific biological functions.[284] Hence, the substituents surrounding the amide bonds direct the non-covalent interactions in all of the above-mentioned crystal structures, and this indicates the possibility to potentially mimic protein architecture with sp3 rigid scaffolds (BCP or cubane) exhibiting the essential conformational space for a protein’s function while simultaneously providing amide bonds for “substrate” coordination.

131

The crystal structure of the BCP-porphyrin 252 illustrates the planar conformation of the macrocyclic core while the crystal packing of this molecule further shows intermolecular head-to-tail non-covalent D···A interactions between the acceptor Zn(II) metal of the porphyrin and donor oxygen atom of the carbonyl group in the amide bond at a distance of 2.191 Å (Figure. 61). This particular interaction supports the proposed mechanism of binding between a transition metal(II) and the C=O moiety of the amide bond (vide infra).

Along with the above mentioned non-covalent interactions, we also observed a unique example of a porphyrin-based ion pair complex. A pair of opposite charges held together by Columbic interactions in the same solvent-shell is known as an ion pair.[285] Charge- separated ion pair complexes are quite common in transition-metal organometallic chemistry. However, while ion pair complexes of phlorins and porphodimethene-based systems have been reported, this type of interaction has not previously been observed for systems with intact porphyrin cores.[286] The crystal structure of porphyrin 251 is unique as

+ it exhibits an ion pair interaction between an axial chloride ligand and a [Et3NH] counter ion in the same unit cell without disturbing the aromatic 18π-electron pathway (Figure. 62). Axial coordination results in displacement of the Zn(II) ion from the 24-atom mean plane by 0.51 Å. The chloride and triethylammonium ions exhibit an ion pair interaction at a distance of 3.043 Å. This is further supported by the 1H and 13C NMR spectra of compound 251

+ where the ratio of the porphyrin derivative and Et3NH was found to be 1:1. To best of our knowledge, it is the first example of a porphyrin-based charge-separated ion pair complex.

Figure 62. (a) Molecular structure of compound 251 in the crystal. (b) Charge separated ion pair + complex of porphyrin 251 and [HNEt3] .

132

Structure elucidation of BCP-porphyrin 251 revealed the nearly coplanar nature of the BCP- appended arm with respect to the porphyrin plane, with only an angle of 9.87° between the two moieties. In contrast, BCP-porphyrin 252, which has a larger distance between the BCP and porphyrin moieties, showed orthogonal rotation of the phenyl rings with respect to the porphyrin plane of 62.43°. Hence, 251 and its analogues show more promise towards the synthesis of cubane/BCP-linked porphyrin systems for electron/energy studies owing to their extended conjugation.

4.4. Spectroscopy Studies

UV-Vis spectra of the chromophore arrays were recorded in CHCl3 or THF at room temperature. The free base dimers 235 and 257 illustrate a typical etio-type porphyrin spectrum, having the Soret and four Q-bands in decreasing intensity. The symmetrical zinc dimers 234, 237, 238, 256, 258 and 259 showed an absorbance maximum at 422 nm. The full width at half maxima (FWHM) of these dimers is nearly equal to that of H2TPP 217 or its zinc(II) complex (ZnTPP), displaying no evidence of exciton coupling between the two porphyrin units.[287] The lack of exciton coupling observed between two porphyrin units suggests that the porphyrins are electronically isolated. Therefore, these results indicate that the dimers are arranged in trans orientation as no communication between the porphyrins is being observed.[288] Absorption spectra of ethynyl-linked porphyrin dimers such as 237, 238 and 289 exhibited a 15–18 nm bathochromic shift compared to the phenylene-linked dimers 234 and 256 due to the π-extended ethynyl or phenylethynyl moieties. The UV-Vis spectra of ethynyl-linked dimers (237, 238 and 259) exhibit nearly the same FWHM and λmax as compared to the precursor amine porphyrin (213b and 213c).

Similar λmax values of monomers and dimers indicate the lack of through space or through bond electronic communication between porphyrin units, i.e. a trans orientation of the synthesised dimers, as was observed for the linearly linked triptycene dimers in Chapter 3. In the case of a trans-configuration, the aromatic cloud of the porphyrin units are far apart, hence, the FWHM is sharp. Conversely, in a cis-configuration the interaction is increased between the porphyrin aromatic clouds thus the FWHM also increases.

4.5. Tweezer Studies with Metal Ions

Due to the remarkable photophysical and electronic properties of porphyrins and their derivatives, they have attracted considerable attention over many decades.[249] The introduction of specific molecular recognition motifs into porphyrin arrays can be investigated to explore their conformational flexibility and molecular recognition processes.[289] In nature, many proteins contain metal ions as a cofactor unit, the selection of a particular metal in a specific oxidation state often being crucial for the biological

133 function. Importantly, coordination of a metal ion dictates the folding/unfolding of the protein skeleton.[290] The coordination of metal cations to molecules containing amide bonds has been reported in the literature, particularly in the active site of enzymes, and they have been shown to coordinate at the carbonyl oxygen as opposed to the NH moiety in the amide bond. Only few cases of NH coordination have been observed and only in the presence of a base.[290]

Initially, unsuccessful zinc metalations of the free base porphyrin dimers (235 and 257) led to the suggestion that an external coordination of the metal ion was taking place. Consequently, the synthesised porphyrin-cubane/BCP-porphyrin arrays featuring a semi- rigid amide linkage were investigated for their suitability as hosts for divalent transition metal ions (Figure 63). In contrast to complex protein structures, conformational changes in simple porphyrin dimers can be easily monitored by UV-Vis spectroscopy and/or distinct colour change. The coordination chemistry of amide-bond-linked porphyrin dimers is explored mainly in terms of binding with N-donating ligands such as anion binding with NHs or for amino acid recognition, but metal ion detection through porphyrin dimers has not yet been reported in the literature.[291]

Figure 63. Proposed coordination mode of porphyrin dimers and metal ions.

The M(II) recognition properties of the synthesised porphyrin dimers 234–238 and 256–259 were studied in CHCl3 with various metal ions, such as Cu(II), Fe(II), Zn(II), Hg(II), Cd(II), Co(II), in the form of perchlorate salts using UV-Vis spectroscopy. The same spectral changes were observed for the sequential addition of all metal ions, bar Co(II) as it did not exhibit any coordination with porphyrin dimers. Porphyrins 234 and 256 represent an illustrative example of the spectral changes induced by the addition of Zn(II) ions, as shown in Figure 64a and Figure 64b. When aliquots of Zn(II) were added to a solution of 234, the absorption spectra changed drastically with an isosbestic point at 431 nm. As the concentration of the Zn(II) ions increased, absorbance at the Soret and Q bands, 422 nm 134 and 549 nm, decreased with a concomitant increment of absorbance at 440 nm and 664 nm. The corresponding data fit to the binding isotherm to show the 1:1 coordination mode of Zn(II) and dimer 234 as shown in the inset of Figure 64a. The change observed in the absorption spectra reflects the combined conformational and electronic effects in the system upon metal ion coordination to the carbonyl oxygen atoms of the amide bond.[292] It can be deduced that the porphyrins in the complexes now have a “cis”-orientation, upon metal ion binding, which brings the porphyrins closer together.

Similar titrations were performed with a set of other transition metal ions such as Fe(II), Hg(II), Cu(II) and Cd(II). From the change in the Soret bands upon titrations, the association

[293] constant Ka of Zn(II) with dimer 234 was calculated using a Benesi–Hildebrand (BH) plot. Association constants were found in the range of 103 to 105. In general, cubane-linked dimers showed stronger affinity towards metal recognition as compared to the BCP-linked dimers. This was thought to be due to the longer distance that is apparent across the cubane face that allows for a binding site for the metal ions that is more akin to their radii than BCP.

Interestingly, titrating the cubane dimer 236 with the M(II) ions demonstrated a 1:2 stoichiometry with the metal ions and resulted in an almost two-fold increase in the value of

Ka, which may arise due to the presence of dual binding modes involving the C=O moieties and the inner core of the porphyrins.[289c] In the unsymmetric dimer 236, the Zn(II) porphyrin unit acts as a donor and the free base unit acts as an acceptor.[294] Hence, the affinity of the free base inner core NH units (Lewis base) increase towards the M(II) ions (Lewis acid). This phenomenon was only observed with dimer 236 as the normal 1:1 binding stoichiometry was still observed for the free base dimer 235 indicating the importance of electronics in the dual binding mode.

Figure 64a and 64b shows that the absorbance of a metal coordinated complex is lower when compared to the parent compound cubane 234 and BCP 256, probably due to an enhanced π–π interaction between the two porphyrin units. The change in FWHM is often related to conformational flexibility.[291] In the present case, an increase in FWHM indicates the enhancement of overlapping vibrational transitions and increased electronic communication between porphyrin units, i.e. a change in orientation of the porphyrin units.

The 115 nm bathochromic shift in the Qx(0,0) band also illustrates the change in conformation of the porphyrin dimers, resulting in reduction of the a HOMO-LUMO gap by 410 meV. To confirm the nature of the weak intermolecular metal interaction, the reversibility of system was tested. The sequential addition of M(II) ions to a porphyrin dimer solution resulted in a colour change from pink to green. Upon washing the complex with water, the original pink solution could be regenerated restoring the initial UV-Vis spectral features of the porphyrin dimer, further proving the proposed site of metal binding, as the porphyrin core remained

135

(c)

Figure 64. (a) UV-Vis spectrum of Zn(II) addition to a solution of cubane 234 in CHCl3. (b) UV-Vis spectrum of Zn(II) addition to a solution of BCP 256 in CHCl3. (c) Structures of cubane 234 and BCP 256. 136 intact during the titration (Figure 63). In addition, insets show a BH plot with 1:1 stoichiometry where R2 = 0.99 for both Figure 64a and 64b. In summary, metal ion association to the porphyrin dimers results in reversible and drastic changes to the UV-Vis absorption spectra indicating that the M(II) ion can act as a template to control the conformation of these dimers.

4.6. Conclusions and Outlook

We have designed, synthesised and characterised bridgehead-substituted BCP and cubane derivatives via amide coupling reactions. This work demonstrates a broad substrate scope with over 35 new derivatives of cubane/BCP that were synthesised in moderate to good yields. The single crystal X-ray structures of small rigid linker motifs (221, 243, 245 and 248) revealed supramolecular 3D networks with combined and repetitive inter- and intramolecular H-bonding interactions. Significantly, the crystal structure of cubane 221 showed an unusual C=O2···I1 interaction along with the usual N1–H···O1=C interaction to result in a 3D cage-like structure.

Titration of M(II) ions with porphyrin dimers (234, 235, 237, 238, 256–258) suggested a 1:1 association of the guest to host. Interestingly, titration of the unsymmetric cubane dimer 236 with M(II) ions revealed the association of two M(II) ions with one porphyrin unit. The association constants are two-times higher as compared to the symmetric dimers, which may be due to dual binding sites in 236 i.e. the amide bonds and the inner core NHs of the free base porphyrin unit. These preliminary results with M(II) ion coordination provide significant groundwork for future applications, focusing on the synthesis of template- directed, controlled and reversible molecular engineering of porphyrin arrays. The complexation behaviour of these porphyrin hosts is potentially dependent on the structurally pre-organised BCP/cubane scaffold in association with the semi-rigid amide moieties, which make them the first-of-their-kind in template induced reversible porphyrin tweezers.

While the alkynylcubane-linked porphyrin dimers discussed in Chapter 2 were unable to be isolated, these cubanyl amide-linked porphyrin dimers can be synthesised in moderate to good yields, and in a straight forward and reliable manner. Consequently, these porphyrin dimers also have potential application in electron transfer studies, as they are linked by an isolating unit and are held together by the semi-rigid amide bonds. The synthesis of various unsymmetric dimers is also possible through the stepwise conversion of the cubanyl ester groups to the prerequisite carboxylic acid in a controllable manner, making the expansion of a diverse library of various dimers a real possibility..

137

Chapter 5. Synthesis of Covalently linked Water-soluble Dimers for PDT Treatment.

5.1. Background on the Basic Principles of PDT Treatment

As was shown in Chapters 1 through 4, porphyrins, and in particular porphyrin dimers, have great potential in a variety of applications, namely as photosynthetic mimics, which can be analysed through electron transfer studies or as molecular tweezers through the coordination of metal ions. Another and extremely rich area of porphyrin research is photodynamic therapy (PDT).[152b] The development of modern photomedicine was pioneered initially by Finsen in 1899[295] and developed through a series of works by von Tappeiner, Jesionek and Raab that refined the principles underpinning all conventional photomedical treatments.[296] Since then, research into PDT has rapidly expanded.[297] PDT involves the use of reactive oxygen species (ROS) to destroy harmful biological targets. It is a safe and reliable treatment for various conditions from dermatological conditions[298] to cancers,[299] and age-related macular degeneration.[300] In addition, antimicrobial PDT has re-emerged as a treatment for infectious disease.[301]

Figure 65. Jablonski diagram showing singlet oxygen generation. Where 1 = absorption, 2 = internal conversion (IC), 3 = fluorescence (FL), 4 = intersystem crossing (ISC), 5 = phosphorescence (PL), 6 = energy transfer, PS = photosystem.[152b]

138

Singlet oxygen, one of the most prominent ROS, is generated by a light-driven photosensitising process. When singlet oxygen is released in cells it ultimately leads to induced apoptosis.[302] Upon absorption of a photon by the photosensitiser (PS) an electron is promoted from the ground state (S0) to an electronically excited singlet state (S2) (Figure 65).[303] A variety of relaxation processes can then occur from the excited singlet state, namely emission of a photon via fluorescence from S1 after internal conversion (IC) has occurred from S2. Alternatively, phosphorescence can occur when the excited energy is transferred to T1 through intersystem crossing (ISC) to generate the excited triplet state

[304] (T1).

When ISC occurs the PS excited triplet state energy can be transferred directly to molecular

3 oxygen in the triplet ground state ( O2), this then leads to the biologically active singlet

1 excited state of molecular oxygen, also called singlet oxygen ( O2). Therefore, the efficiency of the ISC, alongside the lifetime of the T1 state of PS directly affect the quantum yield of

1 [305] O2.

While ample examples of porphyrins that act as PDT agents exist in the literature, there are comparatively very limited instances of porphyrin dimers that are applied in this manner.[152b, 306] One of the only examples of a commercially available multiporphyrin array used for PDT is Photofrin, also known as Porfimer sodium (Figure 66). Photofrin, a derivative of haematoporphyrin, is a mixture of oligomers, which connect up-to-eight porphyrin units via flexible ester and ether linkers and it is used for the treatment of oesophageal, bladder and non-small cell lung cancer, amongst others.[307] The objective of this work is to synthesise simple but rigidly linked water-soluble porphyrin dimers in order to ultimately investigate their applicability in PDT.

Figure 66. Chemical structure of a haematoporphyrin derivative 260.

139

5.2. Synthesis

5.2.1. Sulfonation of Phenyl Porphyrin Dimers

There are several different methods available to the synthetic chemist to introduce water- soluble properties to a porphyrin. Hydrogel formulations and PEGylation are but two examples of a solid-state support and a chemical modification that have been attached or applied to porphyrins to alter their physicochemical properties.[308] Another facile and commonly employed method is the introduction of ionisable groups to the porphyrin periphery, such as carboxylic acids[309] and sulfites[310] to make the porphyrin anionic or alternatively the attachment of methylated pyridine groups in order to form cationic porphyrins.[311]

The introduction of anionic sulfite groups on the phenyl rings of a porphyrin dimer was first pursued in an effort to synthesise a water-soluble porphyrin dimer. Although firstly, as a test reaction, a sulfonation procedure was carried out on TPP. The porphyrin 217 was subjected to a sulfonation reaction by grinding the porphyrin in a pestle and mortar with H2SO4 before transferring the reagents to a reaction vessel (Scheme 80).[310] The reaction mixture was heated to 110 °C for four hours and was then left to stand at room temperature for 18 hours.

The resulting mixture was filtered through silica with CH2Cl2 as eluent and then washed with a small amount of water. The pH of the solution was adjusted to ~8–10 pH by the addition of a 1 M solution of NaOH. The solution was filtered through silica one final time, before isolating the purple-green solid TPPS in a quantitative yield. Functionalisation only occurs at the meso-phenyl rings of TPP, as the core of the porphyrin is protonated in acidic conditions, resulting in substitution on the macrocyclic ring (i.e. at the β-positions) being inhibited.[312]

Scheme 80. Synthesis of TPPS 261.

140

The meso–meso-linked dimer 263 was then synthesised in a 55% yield (Scheme 81). This was achieved through the synthesis of dipyrromethane (DPM) 149 from pyrrole 215 and formaldehyde in a 25% yield. DPM (2 eq.) was used in a condensation reaction with

[158] benzaldehyde 216 (2 eq.) and TFA (1.2 eq.) in CH2Cl2 using Lindsey conditions to obtain 5,15-diphenylporphyrin 262 in a 63% yield. Senge reaction conditions [236a] were then employed using PhLi (6 eq.) in THF to introduce a third phenyl ring to the porphyrin. Instead of quenching the reaction with water and then adding the oxidising agent DDQ, DDQ (4 eq.) was added directly to the reaction mixture resulting in the dimerised product 263 instead of the monomer 144t. The product was purified by filtrating the crude reaction mixture through silica gel using CH2Cl2 as eluent.

Scheme 81. Synthesis of meso–meso-linked porphyrin dimer 263 with PhLi.

With the porphyrin dimer 263 in-hand the same procedure which was utilised for the sulfonation of TPP 217 (Scheme 82). A solution of porphyrin dimer in H2SO4 was heated for four hours and after purification, a brown solid was obtained. Mass spectroscopy and UV-Vis spectrometry did not show any evidence of the desired sulfonated product 264. In addition, 1H NMR spectroscopy revealed no characteristic porphyrin signals of the product or starting material, indicating that decomposition of the porphyrin dimer had occurred. A milder procedure was then sought, that utilises chlorosulfonic acid as the sulfonating agent/solvent.[313] The reaction mixture was allowed to stir at room temperature for one hour instead of the four hours at 110 °C suggested by the sulfonic acid[310] procedure. The reaction mixture was quenched through the careful dropwise addition of the solution to a beaker of ice; even though precautions were taken the reaction was still very violent.

Saturated NaHCO3 was added to neutralise the acid and the product was extracted with

CHCl3. Again, none of the desired product 265 was observed, but starting material was

141 recovered. The reaction time was then increased to allow more time for the starting material to react. This time the reaction mixture was stirred at room temperature for 24 hours, as opposed to the original suggested time in the literature of an hour that was used for the monomer 217.[314] Unfortunately, following the isolation of the reaction product only starting material was observed by analysis with mass spectrometry. The failed sulfonation attempts of the meso–meso-linked dimer 263 were hypothesised to be due the lower oxidation potential of meso–meso linker dimer[315] compared to TPP 217, leading to decomposition of the material in the harsh heated sulfonic acid conditions. In contrast, the chlorosulfonic acid procedure performed at room temperature was too mild resulting in the retention of the starting material. As a result, the pursuit of conjugated porphyrin dimers was then undertaken in an effort to have a porphyrin dimer closer in electronic nature to TPP.

Scheme 82. Sulfonation attempts of porphyrin dimer 263.

Consequently, the sulfonation reaction was performed with the milder chlorosulfonic acid procedure[313] with the dimer 266.[316] The reaction was allowed to stir at room temperature for 24 hours but again the desired product 267 was not observed through mass spectrometry or 1H NMR spectroscopy (Scheme 83).

142

Scheme 83. Attempted sulfonation of dimer 266.

A porphyrin dimer connected by a phenylene linker was then selected as the next synthetic target, as this bears the closest resemblance to the structure of TPP that a dimer can obtain. 5,15-diphenylporphyrin 262 (1 eq.) was reacted under Senge conditions[236a] with PhLi (7 eq.) in THF to attach a third phenyl ring (Scheme 84). This time the reaction was quenched with water before DDQ (5 eq.) was added, enabling access to the porphyrin monomer 144t over the meso–meso-linked dimer 262. Following this, 5,10,15-triphenylporphyrin 144t was brominated with NBS (1.2 eq.) and pyridine in CHCl3 to give the brominated porphyrin 144d in a 75% yield. Followed finally by a zinc insertion with zinc(II) acetate (2.5 eq.) in

MeOH/CHCl3 to allow for the formation of (5-bromo-10,15,20-triphenylporphyrinato)zinc(II) 144e in a 86% yield. A Suzuki cross-coupling reaction was then performed between the porphyrin 144e and 1,4-benzenediboronic acid bis(pinacol) ester 268a using Pd(PPh3)4 (0.3 eq.) as the catalyst, excess K2CO3 (10 eq.) as the base in a mixture of degassed toluene and DMF. The reaction mixture was heated to 90 °C for 24 hours and was quenched with water. The coloured products were extracted with CH2Cl2 and the crude material was purified using column chromatography (SiO2, CH2Cl2/n-hexane, 2:1, v/v). The product dimer 269 was observed when analysed by 1H NMR and mass spectrometry, albeit in a very low yield.

A Suzuki cross-coupling reaction was carried out with the same reagents to synthesise the phenylene-linked dimer 269, but this time the reaction was heated to 110 °C for 30 minutes in a microwave reactor. After initial purification, no product formation was observed.

143

Scheme 84. Synthesis of phenylene-linked zinc(II) dimer 269.

As iodinated porphyrins are known to have better success rates in cross-coupling reactions, a Suzuki cross-coupling reaction was performed with the iodoporphyrin derivative (Scheme 85). The iodinated analogue of the porphyrin was then synthesised from 5,10,15- triphenylporphyrin 144t using molecular iodine (1.2 eq.), PIFA (1 eq.) and pyridine (100 µL) in CHCl3. The reaction mixture was stirred for two hours at room temperature and washed with sodium thiosulfate to remove the molecular iodine. This was followed by filtration through silica gel with CH2Cl2 to give the pure porphyrin product 262b in an 89% yield. A Suzuki reaction with the same conditions as previously stated in Scheme 84 was performed and 1,4-bis(10,15,20-triphenylporphyrin-5-yl)benzene 270 was synthesised in a poor yield of 5%.

A slightly alternative synthetic route was followed in order to synthesise the zinc(II) dimer 269. This was done in order to investigate if a more reliable method, which gives an increased yield of the desired dimer, could be achieved (Scheme 86). A Suzuki cross- coupling reaction was performed with the reactive functionalities inverted i.e. with the borylated porphyrin 262c and 1,4-diiodobenzene linker 268c. The bromoporphyrin 144e was first borylated with pinacolborane (10 eq.) using Pd(PPh3)2Cl2 (0.2 eq.) as the catalyst and NEt3 (10 eq.) as the base in 1,2-dichloroethane. The reaction was stirred at 90 °C for three hours under an argon atmosphere. The crude reaction mixture was then filtered through silica with CH2Cl2 to afford the pure product 262c as purple crystals in a 91% yield. The borylated porphyrin 262c was reacted with the linker 268c using slightly different Suzuki cross-coupling conditions with Pd(PPh3)4 (0.1 eq.) as the catalyst and Cs2CO3 (10 eq.) as the base in toluene. The reaction was heated at 120 °C for 48 hours and purified by standard procedures. While product 269 was detected, a pure and usable amount was not obtained.

Alterations of the reaction conditions were made to use the base K2CO3 and solvent DMF

144 instead of Cs2CO3 and toluene. These changes enabled access to the desired dimer 269 in a 35% yield. Following this, a sulfonation reaction was performed with the dimer 269. The original sulfonation conditions were used, by stirring the porphyrin in concentrated sulfuric acid at 110 °C for four hours, which led to the successful formation of the sulfonated dimer 271 albeit in trace amounts.

Scheme 85. Suzuki cross-coupling reaction with bis(pinacolato)diboron to form the phenylene-linked dimer 269a.

145

Scheme 86. Suzuki cross-coupling reaction with 1,4-diiodobenzene to form the phenylene-linked dimer 269 followed by sulfonation of the dimer to produce compound 271.

5.2.2. Synthesis of Carboxylic Acid Porphyrin Dimers

A different synthetic strategy was employed to acquire anionic porphyrin dimers. The functional group interconversion of ester functionalities to carboxylic acids is a reliable and well-known procedure, ordinarily achieved through the use of hydroxyl-based bases such as potassium or sodium hydroxide.[223] Thus, the synthesis of carboxylic acid porphyrin dimers offers a more viable synthetic route to synthesise charged porphyrin systems when compared to the previous sulfonation strategy. Consequently, a number of different porphyrin dimers were synthesised.

146

Scheme 87. Synthesis of porphyrin 275.

Scheme 88. Suzuki cross-coupling reaction to form phenylene-linked dimer 276.

A condensation was performed with DPM 149 (1 eq.) and methyl 4-formylbenzoate 272 (1 eq.) with TFA (0.6 eq.) in CH2Cl2 at room temperature (Scheme 87). The reaction was allowed to stir for four hours then DDQ (1.5 eq.) was added and the solution was stirred for a further hour. The reaction was quenched with NEt3 and purified through standard methods

147 in order to access porphyrin 273 as a purple solid in a 21% yield. Following this, a zinc(II) metalation reaction was performed with Zn(II)(OAc)2•2H2O (5 eq.) in MeOH/CH2Cl2. The porphyrin starting material 273 had extremely poorly solubility and required high volumes of CHCl3 to dissolve, but nevertheless, the reaction solution was still a suspension. The reaction was left to stir for 24 hours and the desired zinc porphyrin adduct 274 was isolated in a near-quantitative yield. Ordinarily, the next reaction in the porphyrin starting material sequence would have been a reaction with PhLi to protect the other meso-position from halogenation, but owing to the presence of the ester moieties it was decided to by-pass this step so as not to interfere with the ester functionalities. As the iodinated porphyrin was previously shown to be more proficient in the Suzuki cross-coupling reaction, iodination of the porphyrin 274 was attempted. Immediately after the addition of I2 (1.2 eq.) and PIFA (1 eq.) a black colour was observed in the reaction mixture. TLC analysis showed full consumption of the starting material and none of the desired product 275a was detected by mass spectrometer analysis. The obvious alternative to iodination was then trialled and the porphyrin 274 was brominated using 1.1 equivalents of NBS in an attempt to avoid the dibrominated side-product. The desired monobrominated product 275 was successfully isolated after column chromatography with silica gel and a solvent mixture of CH2Cl2/n- hexane (2:1, v/v) to give the product 275 in an 82% yield. The brominated porphyrin 275 was then used in a Suzuki cross-coupling reaction with 1,4-benzenediboronic acid bis(pinacol) ester 286a (Scheme 88). After purification, the target dimer 276 was identified and a 1H NMR spectrum was obtained. Owing to solubility issues of the dimer 276 it was decided that an alternative ester derivative should be synthesised.

Scheme 89. Bromination of various porphyrins with NBS.

Before it was decided to move away from 5,15-bis(4-methoxycarbonylphenyl)porphyrin 273 based-derivatives, the free base and nickel monobrominated derivatives, 278a and 278, were synthesised in 50% and 70% yields, respectively (Scheme 89). A Suzuki cross-

148 coupling reaction was performed with the free base derivative 278 and 1,4- benzenediboronic acid bis(pinacol) ester 286a to form the phenylene-linked porphyrin dimer 279 but extremely low yields were obtained (Scheme 90).

Scheme 90. Suzuki cross-coupling reaction to form porphyrin dimer 279.

Due to the poor solubility encountered with porphyrin 273 and its derivatives it was decided to use its structural isomer porphyrin 281, as the break in symmetry prevents the stacking of the porphyrin molecules in solution, thus increasing its solubility. A condensation reaction was performed with methyl 3-formylbenzoate 280 and DPM 149 to form porphyrin 281 in a 65% yield (Scheme 91). This regioisomer proved to be much more soluble than its porphyrin 273 counterpart and was far easier to work with. In addition, it was also produced in a by far superior yield (three-times higher compared to the para-derivative) of 65%. Although when the scale of this condensation was increased to 14.6 mmol no porphyrin product was seen, so moving forward a 7.3 mmol scale was adhered to. Nickel(II) and zinc(II) insertion reactions were performed to form the zinc and nickel adducts, 284 and 283, in 94% and 84% yields, respectively. The zinc(II) and free base porphyrins were then monobrominated to give the zinc(II) and free base porphyrins, 285 and 282, in 89% and 83% yields, respectively. The bromination of porphyrin 285 proceeded quickly, although purification was cumbersome, while the bromination of the free base porphyrin 281 was very slow and required additional NBS. However, purification of the latter proceeded without difficulties. The compounds were purified using column chromatography with silica gel and a solvent system of CH2Cl2/n-hexane (3:1, v/v). The first fraction retrieved from the column contained the dibrominated side-product 182a that possessed remarkable solubility for a dibrominated porphyrin species. The desired monobrominated porphyrin 282 was acquired in the second fraction, while the unreacted starting material 281 was collected in the third fraction. A bromination reaction was also carried out with the nickel(II) porphyrin 283 but only the dibrominated product was able to be isolated pure in a 10% yield.

149

Scheme 91. The synthesis of various porphyrin ester derivatives.

A Sonogashira cross-coupling reaction was conducted under standard conditions with ethynyl-TMS (2 eq.) in order to convert the bromoporphyrin 285 into the required ethynylporphyrin 286 (Scheme 92). The desired product 286 was obtained in an 86% yield. A Glaser cross-coupling reaction was then performed with the TMS-protected ethynylporphyrin 286. 1 M TBAF was employed to deprotect the ethynyl bond and then standard Sonogashira conditions were implemented to access the desired butadiene-linked zinc dimer 287 in a 32% yield. The product was purified using silica gel column chromatography with a solvent system of EtOAc/n-hexane (2:1, v/v), followed by

150 recrystallisation from CHCl3 and n-hexane which afforded purple-green crystals. A functional group interconversion reaction was then performed with the porphyrin dimer 287 in order to synthesise the tetracarboxylic acid dimer 288. The dimer 287 was reacted with

KOH (350 eq.) in a solution of MeOH/THF/H2O and stirred at a temperature of 80 °C until TLC analysis indicated the completion of the reaction after five hours. The solvents were removed in vacuo and the crude material was re-dissolved in water and neutralised with 1 M HCl until a precipitate was observed. The green solid was isolated through vacuum filtration and was washed with further aliquots of water before being dried under vacuum. The product was isolated in a 95% yield.

Scheme 92. The synthesis of butadiene-linked dimer 288.

5.2.3. Synthesis of Nitrogen-centred Cationic Porphyrin Dimers

A few attempts were made towards the synthesis of cationic porphyrin dimers. Firstly, 5,15- dihexylporphyrin 289 was reacted under Senge reaction conditions[236a] with ArLi 290 to form a meso–meso-linked dimer 291 with two amine groups (Scheme 93). To a solution of 4- bromo-dimethylaniline 290 in Et2O, n-BuLi was added dropwise at 0 °C and the reaction was allowed to stir at this temperature for one hour, where the formation of yellow colour was observed. A purple solution of the porphyrin 289 dissolved in THF was then added to the reaction vessel and the reaction was allowed to stir for 30 minutes at room temperature. Following this, DDQ (4 eq.) was added and the reaction was stirred for a further hour and a

151 brown solution was obtained. The amine groups have the potential to be methylated in an additional step to transform the dimer into a cationic species, although, the final step was never performed. Even though mass spectrometry results did reveal the formation of the desired product 291, the target compound was only isolated in a trace amount after purification. The insufficient yield may have been due to the incomplete formation of the Grignard reagent, as only a faint yellow solution was observed upon its synthesis when a deep yellow colour was expected.[317]

Scheme 93. Synthesis of meso–meso-linked hexyl dimer 291.

Starting from pyrrole 215 the meso-substituted pyridine DPM 293 was synthesised in a 50% yield in a condensation reaction with 4-pyridinecarboxaldehyde 292. 5,15-Bis(pyridin-4- yl)porphyrin 295 was synthesised in a 12% yield after a condensation reaction with trimethyl orthoformate 294. The DPM 293 and trimethyl orthoformate were dissolved in a degassed solution of CH2Cl2. A solution of trichloroacetic acid in CH2Cl2 was then added dropwise and the solution was allowed to stir at room temperature, in darkness, for four hours. Pyridine was added to quench the reaction and it was allowed to stir for a further 17 hours. The crude product was filtered through silica gel with CH2Cl2 and a small amount of MeOH. After recrystallisation from CH2Cl2/n-hexane, product 295 was isolated as purple crystals. This porphyrin 295 can then either be used with Senge reaction conditions using PhLi to form the meso–meso-linked dimer or it can alternatively be linked through a butadiene-linker as that has proven to be the most reliable method of obtaining porphyrin dimers (Scheme 94). Once the dimer has been synthesised, the pyridine groups can then be methylated to form another nitrogen-centred cationic dimer.

152

Scheme 94. Synthesis of porphyrin 295.

5.3. Singlet Oxygen Measurements

1 With the water-soluble dimer 288 in hand, its proficiency at singlet oxygen ( O2) generation

1 was investigated. The results of O2 generation of the dimer were compared with those of the corresponding monomer 286 and commercially available Temoporfin 299 to access its applicability for PDT (Figure 67).[307a] 1,3-Diphenylisobenzofuran (DPBF) 296 was selected

1 as the O2 trapping agent to investigate the amount of singlet oxygen generated (Scheme

[318] 1 95). DPBF is a fluorescent yellow compound that reacts selectively with O2, resulting in reduced absorption. It characteristically has an absorption band at 417 nm observed in its UV-Vis spectrum. The presence of singlet oxygen evokes a [4+2] cycloaddition reaction with the furan ring of DPBF to give the intermediate 297 (Scheme 95).[319] Elimination of water leads to ring-opening of the six-membered ring to form the UV-inactive diketone species 298. Upon exposure to a broad spectrum light source, a sensitizer that is present, such as a porphyrin, generates singlet oxygen. Exposure of the porphyrin solution to a halogen lamp (Philips, 15V-150 W lamp) is directly proportional to the amount of singlet oxygen produced. Solutions of porphyrin and DPBF 296 in DMF were exposed to a light source for three seconds intervals and UV-Vis spectra were acquired after each exposure (Figure 68). Thus, the gradual decrease in DPBFs absorption band could be monitored and corresponds to the amount of singlet oxygen produced.

The time required for the complete quenching of DPBF was 60 and 66 seconds for the two monomers Temoporfin 299 and the monomer 286, respectively (Figure 69). Interestingly, the dimer 287 showed significantly enhanced singlet oxygen generation, as DPBF was quenched after already 21 seconds, which is three-times faster than the commercially available Temoporfin 299. This offers an interesting insight into the use of porphyrin dimers as valuable sensitisers in PDT.

153

Scheme 95. Decomposition pathway of 1,3-diphenylisobenzofuran (DPBF) in the presence of singlet oxygen.[320]

Figure 67. The chemical structures of Temoporfin 299, the porphyrin monomer 286 and dimer 287.

154

Figure 68. DPBF quenching with (a) Temoporfin 299, (b) monomer 286 and (c) dimer 287 in DMF by irradiating the sample in three seconds intervals.

DPBF Quenching

2.5 Foscan 2 Monomer

1.5 Dimer

1 Absorption (a.u.)Absorption 0.5

0 0 10 20 30 40 50 60 70 Time (seconds)

Figure 69. Time vs. absorption of Temoporfin 299, monomer 286 and dimer 287.

155

5.4. Conclusions and Outlook

In an effort to synthesise water-soluble porphyrin dimers for use as photosensitisers in PDT treatments, a series of cationic and anionic dimers were targeted. Initially, sulfonation reactions were attempted to introduce cationic sulfate ions to pre-existing porphyrin dimers. The sulfonation reaction was tried with the traditional route of heating the porphyrin in sulfuric acid but only decomposition of the porphyrin was observed for the meso–meso- linked dimer 263. A milder route employing a chlorosulfonic acid procedure was then trialled with the meso–meso-linked dimer 263 and ethene-linked dimer 266, but only starting materials were observed after initial analysis. After several synthetic attempts, a porphyrin dimer with a phenylene linker was obtained and sulfonation of this dimer 269 led to the successful formation of the water-soluble sulfonated dimer 271 albeit in trace amounts.

Similarly, anionic porphyrin dimers were also a synthetic target for obtaining water-soluble porphyrin dimers. Initial work with 5,15-bis(4-methoxycarbonylphenyl)porphyrin 273 and its derivatives proved difficult due to solubility issues. In order to overcome this, the regioisomer 5,15-bis(3-methoxycarbonylphenyl)porphyrin 281, that employs the ester moieties at the meta position of the porphyrin, was utilised and proved to have drastically superior solubility and hence yields. A series of new compounds were synthesised with this porphyrin, including the homo-coupled porphyrin dimer 288 with four carboxylic acids present that introduce the water-solubility to the molecule.

The synthesis of nitrogen-centred cationic porphyrin dimers was commenced with synthetic targets such as porphyrin dimers with amines and pyridine functionalities that can be methylated to introduce a cationic charge into the dimer. More work is necessary before these dimers can be achieved.

The new porphyrins that were obtained during the synthesis of the carboxylic acid dimer were investigated for their ability to act as porphyrin photosensitisers. The dimer 287, the monomer 286 and commercially available Temoporfin 299 were all used in singlet oxygen measurement experiments with DPBF. Results showed that at the same concentration, the porphyrin dimer 287 was three-times more efficient at generating singlet oxygen as compared to the monomer 286 and Temoporfin 299, supporting the argument for the use of porphyrin dimers in PDT treatment. Much optimisation of this current work is needed in order to further pursue cationic and anionic porphyrin dimers for water-soluble PDT treatment.

156

Chapter 6. Experimental

6.1. General methods, Instrumentation and Considerations

All commercial chemicals used were supplied by Sigma Aldrich, Acros Organics, Fluka Frontier Scientific, Inc. and Fischer and used without further purification unless otherwise stated. Large scale reactions used anhydrous dichloromethane that was distilled from phosphorus pentoxide, while smaller amounts of tetrahydrofuran (THF), diethyl ether, toluene (Tol), and CH2Cl2 were withdrawn from an automated solvent purification system. To protect air- and moisture sensitive compounds, the corresponding reactions were carried out under “Schlenk conditions” using argon as inert gas. Air and residual moisture were removed from the instruments by a hot-air gun under high vacuum and the flasks were purged with argon subsequently.

Reactions were monitored by thin layer chromatography (TLC) carried out on silica gel plates using UV light as a visualizing agent or p-anisaldehyde and heat as a staining/developing agent. Analytical thin layer chromatography was performed using silica gel 60 (fluorescence indicator F254, precoated sheets, 0.2 mm thick, 20 cm × 20 cm; Merck) or aluminium oxide 60 (neutral, F254; Merck) plates and visualized by UV irradiation (λ = 254 nm). Column chromatography was carried out using silica Gel (230–400 mesh; Merck), aluminium oxide (neutral, activated with 6% H2O, Brockman Grade III) or size exclusion chromatography with Bio-Beads™ S-X1 Resin. Mobile phases are given as (v/v). Yields refer to chromatographically and spectroscopically (1H NMR) homogeneous material, unless otherwise noted. Room temperature refers to 20–25 °C.

Melting points are uncorrected and were measured with a Stuart SMP-50 melting point apparatus. NMR spectra were recorded using Bruker DPX400 (400 MHz for 1H NMR, 101 MHz for 13C NMR), Bruker AV 600 (600 MHz for 1H NMR, 151 MHz for 13C NMR) and Bruker AV 400 (400 MHz for 1H NMR, 101 MHz for 13C NMR) instruments. All NMR experiments were performed at room temperature. Chemical shifts are given in ppm and referenced to the residual peak of the deuterated NMR solvent. Signal multiplicities are abbreviated as follows: singlet = s, multiplet = m, doublet = d, doublet of doublets = dd. The assignment of the signals was confirmed by 2D spectra (COSY, HMBC, HSQC) except for those compounds with low solubility. ESI mass spectra were acquired in positive or negative modes as required, using a Micromass time-of-flight mass spectrometer (TOF), or a Bruker mircoOTOF-Q II spectrometer interfaced to a Dionex UltiMate 3000 LC. APCI experiments were carried out on a Bruker microOTOF-Q III spectrometer interfaced to a Dionex Ultimate 3000 C or direct insertion probe in positive or negative modes. IR spectra were recorded on a PerkinElmer Spectrum 100 FT-IR spectrometer. UV−Vis spectra were recorded in

157 solutions using a Specord 250 spectrophotometer from Analytik Jena (1 cm path length quartz cell). Emission, excitation spectra and lifetimes were measured using a Cary Eclipse G9800A fluorescence spectrophotometer and Horiba Jobin Yvon Fluorolog 4. In vacuo, either refers to use of a Büchi Rotavapor R-200; a Büchi Rotavapor R-210; a Büchi Rotavapor R-100 with a Büchi V-491 heating bath and Büchi V-850 vacuum controller.

The stoichiometry and binding constants for protonation were analysed using the Benesi- Hildebrand equation for 1:1 complex formation between porphyrin and guest ion/s

ΔA 1 max = 1 + ( ) ΔA k푎[M]

Ka is the association constant and [M] is the concentration of the ion. A plot of 1/ΔA vs. 1/[M] will yield a straight line with slope 1/ ka. for 1:1 stoichiometry. The association constants were calculated by dividing the intercept of the graph by the slope.

Photo-irradiations were performed in quartz cuvettes (2 × 1 × 1 cm) using a polychromatic light source (Philips, 15V-150 W lamp), equipped with a 400 nm cut-off filter (Schott GG 400). The samples temperature controlled using a Peltier element (Cary Peltier1×1 Cell Holder). All UV-Vis measurements were conducted at a concentration of 1.83x10-6 M for all three porphyrins. In addition, the same concentration of 7.4x10-4 M of DPBF was added to the cuvette. The solutions were exposed to the light source for three seconds until the DPBF absorption band no longer decreased and the porphyrin signals were regenerated.

Single crystal X-ray diffraction data for all compounds were collected on a Bruker APEX 2

DUO CCD diffractometer by using graphite-monochromated MoKα (λ = 0.71073 Å) radiation and Incoatec IμS CuKα (λ = 1.54178 Å) radiation. Crystals were mounted on a MiTeGen MicroMount and collected at 100(2) K by using an Oxford Cryosystems Cobra low- temperature device. Data were collected by using omega and phi scans and were corrected for Lorentz and polarization effects by using the APEX software suite.[321] Using Olex2, the structure was solved with the XT structure solution program, using the intrinsic phasing solution method and refined against │F2│ with XL using least squares minimization.[322] Hydrogen atoms were generally placed in geometrically calculated positions and refined using a riding model.

6.2. Ch. 2 – Cubane Cross-coupling and Cubane–Porphyrin Arrays

6.2.1. Synthesis of Cubane Precursors

1,4-Diester cubane 24 and its precursors (19–23) were synthesised as per Tsanaktsidis’ method.[62b] 1,4-Diethynylcubane 158, its precursors (155–157), and TMS-protected cubane

158

162 were synthesised via Eaton methodology.[224] All compounds and intermediates had analytical data consistent with literature values.

Methyl 4-(4'-ethynylphenyl)cubane-1-carboxylate (140p)

Cubane 144f (46 mg, 140 μmol) was dissolved in THF (90 mL).

A solution of K2CO3 (110.5 mg, 80 μmol), in MeOH (30 mL) was added to the reaction and monitored by TLC. H2O was added and the product was extracted with CH2Cl2 (×3), washed with NaHCO3 and dried over MgSO4 and solvents were removed in vacuo to yield off white crystals (35 mg, 0.13 mmol, 95%). m.p. = 199–201 °C (dec.).

Rf = 0.39 (SiO2, n-hexane:EtOAc, 10:1, v/v).

1 H NMR (600 MHz, CDCl3): δ = 7.48 (d, J = 8.2 Hz, 2H, phenyl-H), 7.15 (d, J = 8.2 Hz, 2H, phenyl-H), 4.27–4.23 (m, 3H, cubane-H), 4.17–4.13 (m, 3H, cubane-H), 3.74 (s, 3H, CH3), 3.06 ppm (s, 1H, ethynyl-H).

13 C NMR (151 MHz, CDCl3): δ = 172.7, 142.9, 132.4, 125.0, 119.4, 83.7, 77.0, 60.2, 56.6, 51.7, 48.9, 46.2, 29.9 ppm.

IR (neat)/cm-1: ν̃ = 3237 (m, alkyne), 2924 (w, cubane C–H), 1711 (s, C=O), 1438 (m, phenyl C–H), 1315 (s, phenyl C–H), 1214 (s), 1091 (s), 907 (w), 822 (s), 673 (m).

+ HRMS (APCI) m/z calcd. for C18H14O2 [M] : 263.1066, found 263.1055.

6.2.2. Synthesis of A2BC-Porphyrins

Dipyrromethane 149 was synthesised according to the Lindsey method[235b] 5,15- Disubstituted porphyrins were synthesised via standard condensation reactions followed by phenyl lithium insertion reactions using Senge reaction conditions[323] to afford 5,15-dihexyl- 10-phenylporphyrin 150, 5,10,15-triphenylporphyrin 144t, 5,15-(4'-methylphenyl)-10- phenylporphyrin and 5,15-(dipentan-3'-yl)-10-phenylporphyrin.[234, 236b]

Bromination reactions[324] resulted in 5-bromo-10,20-bis(4'-methylphenyl)-15- phenylporphyrin 144a, 5,15-dibromo-10,20-bis(4'-methylphenyl)porphyrin 144c, 5-bromo- 10,15,20-triphenylporphyrin 144d and 5-bromo-10,20-(dipentan-3'-yl)-15-phenylporphyrin. Iodination reactions[325] resulted in 5-iodo-10,20-bis(4'-methylphenyl)-15-phenylporphyrin and 5-iodo-10,15,20-triphenylporphyrin. Nickel(II) and zinc(II) insertion reactions were performed following literature procedures.[326] to afford [5-bromo-10,20-bis(4'- methylphenyl)-15-phenylporphyrinato]nickel(II) 144b, (5-bromo-10,20-dihexyl-15- phenylporphyrinato)zinc(II) 144g, (5-bromo-10,15,20-triphenylporphyrinato)zinc(II) 144e,

159

[5-iodo-10,20-bis(4'-methylphenyl)-15-phenylporphyrinato]nickel(II) 144k, [5-iodo- 10,15,20-tris(4'-methylphenyl)porphyrinato]nickel(II) 144l, (5-bromo-10,15,20- triphenylporphyrinato)nickel(II) 144m.[326] All known porphyrins had analytical data consistent with those reported in the literature.

General Procedure 1 (Zn(II) insertion):

Free base porphyrin (1 eq.) was dissolved in CHCl3 while Zn(II)(OAc)2•2H2O (2.5 eq.) was dissolved separately in MeOH and then added to the porphyrin solution to stir at room temperature for two hours. The reaction process was monitored by TLC and once all the starting material was consumed the reaction mixture was washed with NaHCO3, dried over

MgSO4 and filtered through silica gel using CH2Cl2 as the eluent. Removal of the solvents was followed by recrystallisation from CHCl3/MeOH.

General Procedure 2 (Ni(II) insertion):

Free base porphyrin (1 eq.) and Ni(acac)2 (1.5 eq.) were dissolved in toluene and heated to 120 °C. The reaction process was monitored by TLC and once all the starting material was consumed the reaction flask was allowed to cool and filtered through silica gel using CH2Cl2 as eluent. Removal of the solvents was followed by recrystallisation from CHCl3/MeOH.

General Procedure 3 (Bromination of porphyrins):

This procedure was adapted from Boyle and co-workers.[324] The trisubstituted porphyrin was dissolved in CHCl3. At room temperature N-bromosuccinimide (NBS, 3 eq.) and pyridine (0.30 mL) were added. The reaction was stirred at room temperature shielded from light for two hours. The progress of the reaction was monitored by TLC and once all starting material had been consumed the reaction mixture was filtered through silica gel using

CH2Cl2 as eluent. The solvents were removed in vacuo and the crude product was purified by recrystallisation from CHCl3/MeOH.

General Procedure 4 (Iodination of porphyrins):

Following the procedure of Boyle and co-workers[325] the porphyrin (1 eq.) was dissolved in

CHCl3 and purged with argon. Iodine (1.5 eq.) and PIFA (1.1 eq.) were added and the flask. The reaction mixture was left to stir at room temperature until consumption of the starting material was complete (typically ~2 hours). The solution was then filtered through silica gel using CH2Cl2 as eluent, solvents were removed and the product was recrystallised from

CHCl3/MeOH.

160

5-Bromo-10,20-dihexyl-15-phenylporphyrin (144f)

Synthesised via General Procedure 3 from 5,15-dihexyl-10- phenylporphyrin (270 mg, 0.49 mmol), NBS (104 mg, 0.58 mmol),

pyridine in CHCl3 (480 mL). The product was obtained as purple crystals (268 mg, 0.42 mmol, 86%).

m.p. = 118–119 °C (dec.).

Rf = 0.31 (SiO2, n-hexane:CH2Cl2, 2:1, v/v).

1 H NMR (400 MHz, CDCl3): δ = 9.66 (d, J = 4.9 Hz, 2H, Hβ), 9.38 (d, J = 4.9 Hz, 2H, Hβ),

9.30 (d, J = 4.8 Hz, 2H, Hβ), 8.78 (d, J = 4.8 Hz, 2H, Hβ), 8.16–8.12 (m, 2H, phenyl-H), 7.77–

7.70 (m, 3H, phenyl-H), 4.85–4.77 (m, 4H, hexyl-CH2), 2.45 (dt, J = 15.6, 7.9 Hz, 4H, hexyl-

CH2), 1.75 (dt, J = 15.2, 7.5 Hz, 4H, hexyl-CH2), 1.54–1.42 (m, 4H, hexyl-CH2), 1.37 (dq, J

= 14.2, 7.0 Hz, 4H, hexyl-CH2), 0.97-0.87 (m, 6H, hexyl-CH3), –2.77 ppm (s, 2H, NH).

13 C NMR (101 MHz, CDCl3): δ = 142.5, 134.4, 132.7, 128.9, 127.9, 126.7, 121.3, 119.9, 102.2, 38.9, 35.4, 32.0, 30.3, 22.9, 14.3 ppm.

UV-Vis (CH2Cl2): λmax [nm] (log ε) = 419 (5.62), 520 (4.17), 555 (4.02), 601.4 (3.79), 658.0 (3.93).

+ HRMS (MALDI) m/z calcd. for C38H41N4Br [M] : 632.2515, found 632.2539.

5,15-Dihexyl-10-iodo-20-phenylporphyrin (144r)

Synthesised via General Procedure 4 from 5,15-dihexyl-10- phenylporphyrin (51.6 mg, 93 μmol), iodine (21 mg, 95 μmol) and PIFA

(40 mg, 93 μmol) in CHCl3 (25 mL). The product was obtained as purple crystals (76 mg, 0.11 mmol, 73%).

m.p. = 114 °C (dec.).

Rf = 0.26 (SiO2, n-hexane:CH2Cl2, 2:1, v/v).

1 H NMR (400 MHz, CDCl3): δ = 9.71 (d, J = 4.9 Hz, 2H, Hβ), 9.39 (d, J = 4.9 Hz, 2H, Hβ),

9.33 (d, J = 4.8 Hz, 2H, Hβ), 8.81 (d, J = 4.8 Hz, 2H, Hβ), 8.18–8.13 (m, 2H, phenyl-H), 7.83–

7.71 (m, 3H, phenyl-H), 4.88–4.81 (m, 4H, hexyl-CH2), 2.47 (dt, J = 15.5, 7.9 Hz, 4H, hexyl-

CH2), 1.77 (dt, J = 15.1, 7.5 Hz, 4H, hexyl-CH2), 1.54–1.45 (m, 4H hexyl-CH2), 1.38 (dq, J

= 14.1, 7.0 Hz, 4H, hexyl-CH2), 0.93 (t, J = 7.2 Hz, 6H, hexyl-CH3), –2.71 ppm (s, 2H, NH).

161

13 C NMR (101 MHz, CDCl3): δ = 142.4, 137.3, 134.5, 132.0, 127.9, 126.8, 120.8, 120.2, 78.1, 38.9, 35.5, 32.0, 30.3, 22.9, 14.3 ppm.

UV-Vis (CH2Cl2): λmax [nm] (log ε) = 422 (5.64), 523 (4.02), 559 (3.19), 605.5 (3.58), 659.0 (3.82).

+ HRMS (MALDI) m/z calcd. for C38H41N4I [M] : 680.2376, found 680.2370.

(5-Bromo-10,20-dihexyl-15-phenylporphyrinato)nickel(II) (144i)

Synthesised via General Procedure 3 from (5,15-dihexyl-10- phenylporphyrinato)nickel(II) (350 mg, 0.57 mmol), NBS (122 mg,

0.69 mmol), pyridine in CHCl3 (200 mL). The product was obtained as purple crystals (300 mg, 0.43 mmol, 76%).

m.p. = 88 °C (dec.).

Rf = 0.55 (SiO2, n-hexane:CH2Cl2 ,1:1, v/v).

1 H NMR (400 MHz, CDCl3): δ = 9.47 (dd, J = 5.0, 0.8 Hz, 2H, Hβ), 9.27 (d, J = 5.1 Hz, 2H,

Hβ), 9.17 (d, J = 5.0 Hz, 2H, Hβ), 8.68 (dd, J = 5.0, 0.8 Hz, 2H, Hβ), 7.94 (d, J = 7.4 Hz, 2H, phenyl- H), 7.71–7.61 (m, 3H, phenyl-H), 4.53–4.43 (m, 4H, hexyl-CH2), 2.24 (dt, J = 15.6,

7.7 Hz, 4H, hexyl-CH2), 1.61–1.54 (m, 4H, hexyl-CH2), 1.40 (dt, J = 14.8, 6.8 Hz, 4H, hexyl-

CH2), 1.34– 1.24 (m, 4H, hexyl-CH2), 0.88 ppm (t, J = 6.9 Hz, 6H, hexyl-CH3).

13 C NMR (101 MHz, CDCl3): δ = 142.6, 142.4, 141.8, 141.1, 140.7, 133.7, 133.3, 132.7, 130.4, 129.7, 127.8, 127.0, 118.6, 118.4, 101.1, 37.6, 34.1, 31.9, 30.1, 22.8, 14.3 ppm.

UV-Vis (CH2Cl2): λmax [nm] (log ε) = 419 (6.36), 536 (5.18).

+ HRMS (MALDI) m/z calcd. for C38H39N4NiBr [M] : 688.1712, found 688.1719.

(5,15-Dihexyl-10-iodo-20-phenylporphyrinato)nickel(II) (144j)

Synthesised via General Procedure 2 from 5,15-dihexyl-10-iodo-20-

phenylporphyrin 144r (287 mg, 0.42 mmol), Ni(acac)2 (120 mg, 0.63

mmol) in toluene (30 mL). Recrystallisation (CHCl3/MeOH) yielded purple crystals (278 mg, 0.37 mmol, 84%).

m.p. = 87 °C (dec.).

Rf = 0.26 (SiO2, n-hexane: CH2Cl2, 6:1, v/v). 162

1 H NMR (400 MHz, CDCl3): δ = 9.43 (d, J = 4.9 Hz, 2H, Hβ), 9.18 (d, J = 4.9 Hz, 2H, Hβ),

9.13 (d, J = 4.9 Hz, 2H, Hβ), 8.66 (d, J = 4.9 Hz, 2H, Hβ), 7.92 (d, 2H, phenyl-H), 7.73–7.61

(m, 3H, phenyl-H), 4.41 (t, J = 7.4 Hz, 4H, hexyl-CH2), 2.22 (dt, J = 14.2, 7.1 Hz, 4H, hexyl-

CH2), 1.60– 1.49 (m, 4H, hexyl-CH2), 1.39 (dt, J = 14.4, 6.4 Hz, 4H, hexyl-CH2), 1.30 (dt, J

= 14.1, 7.0 Hz, 4H, hexyl-CH2), 0.89 ppm (t, J = 7.1 Hz, 6H, hexyl-CH3).

13 C NMR (101 MHz, CDCl3): δ = 143.4, 143.0, 142.5, 141.7, 140.6, 138.1, 133.6, 132.9, 130.9, 129.9, 127.9, 127.0, 118.8, 118.6, 75.8, 37.6, 34.1, 31.9, 30.1, 22.8, 14.3 ppm.

UV-Vis (CH2Cl2): λmax [nm] (log ε) = 421 (5.36), 538 (4.22).

+ HRMS (MALDI) m/z calcd. for C38H39N4INi [M] : 736.1573, found 736.1594.

[5-Iodo-10,15-bis(4'-methylphenyl)-20-phenylporphyrinato]nickel(II) (144k)

Synthesised via General Procedure 2 from porphyrin 144s (517 mg,

0.75 mmol) and Ni(acac)2 (291 mg, 1.12 mmol) in toluene (300 mL).

After recrystallisation (CHCl3/MeOH) purple crystals were obtained (500 mg, 0.67 mmol, 89%).

m.p. = 250 °C (dec.).

Rf = 0.45 (n-hexane:CH2Cl2, 3:1, v/v).

1 H NMR (400 MHz, CDCl3): δ = 9.48 (d, J = 5.0 Hz, 2H, Hβ), 8.76 (d, J = 5.0 Hz, 2H, Hβ),

8.66–8.70 (m, 4H, Hβ), 7.96 (dd, J = 7.6, 1.6 Hz, 2H, phenyl-H), 7.85 (d, J = 7.9 Hz, 4H, tolyl-H), 7.67 (dd, J = 9.3, 6.2 Hz, 3H, phenyl-H), 7.48 (d, J = 7.9 Hz, 4H, tolyl-H), 2.65 ppm

(s, 6H, CH3).

13 C NMR (100 MHz, CDCl3): δ = 144.5, 143.7, 143.0, 142.9, 140.7, 137.9, 137.7, 137.6, 133.7, 132.8, 132.6, 128.0, 127.8, 127.1, 119.7, 21.6 ppm.

UV-Vis (CH2Cl2): λmax [nm] (log ε) = 420 (6.61), 534 (5.46).

+ HRMS (MALDI) m/z calcd. for C40H27IN4Ni [M] : 748.0634, found 748.0616.

163

(5,15-Dihexyl-10-phenylporphyrinato)zinc(II) (144n)

Synthesised via General Procedure 1 from 5,15-dihexyl-10-

phenylporphyrin (150 mg, 0.27 mmol), Zn(II)(OAc)2•2H2O (118.7 mg, 0.54

mmol) in CHCl3 (150 mL) and MeOH (150 mL). Purple crystals were obtained (164 mg, 0.27 mmol, 98%).

m.p. = 168 °C (dec.).

Rf = 0.23 (SiO2, n-hexane:CH2Cl2, 2:1, v/v).

1 H NMR (400 MHz, CDCl3): δ = 9.35 (d, J = 4.6 Hz, 2H, Hβ), 9.22 (s, 1H, meso-H), 8.99 (d,

J = 4.5 Hz, 2H, Hβ), 8.94 (d, J = 4.6 Hz, 2H, Hβ), 8.73 (d, J = 4.5 Hz, 2H, Hβ), 8.30 (dd, J =

7.2, 1.6 Hz, 2H, phenyl-H), 7.86 (t, J = 6.0 Hz, 3H, phenyl-H), 4.57–4.50 (m, 4H, hexyl-CH2),

2.46–2.36 (m, 4H, hexyl-CH2), 1.86–1.76 (m, 4H, hexyl-CH2), 1.54 (dt, J = 15.4, 6.9 Hz, 4H, hexyl-CH2), 1.49–1.40 (m, 4H, hexyl-CH2), 1.00 ppm (t, J = 7.2 Hz, 6H, hexyl-CH3).

13 C NMR (101 MHz, CDCl3): δ = 149.6, 149.2, 148.6, 148.0, 143.4, 134.5, 131.9, 131.1, 128.3, 127.4, 126.5, 119.6, 104.1, 38.9, 35.2, 32.0, 30.4, 22.8, 14.2 ppm.

UV-Vis (CH2Cl2): λmax [nm] (log ε) = 414 (5.65), 545 (4.22).

+ HRMS (MALDI) m/z calcd. for C38H40N4Zn [M] : 616.2544, found 616.2563.

(5,15-Dihexyl-10-iodo-20-phenylporphyrinato)zinc(II) (144o)

Synthesised via General Procedure 1 from iodoporphyrin 144r (300

mg, 0.44 mmol), Zn(II)(OAc)2•2H2O (144.87 mg, 0.66 mmol) in CHCl3 (150 mL) and MeOH (150 mL). Purple crystals were obtained (327 mg, 0.44 mmol, 99%).

m.p. = 193 °C (dec.).

Rf = 0.62 (SiO2, n-hexane:CH2Cl2, 1:1, v/v).

1 H NMR (400 MHz, CDCl3): δ = 9.40 (d, J = 2.5 Hz, 2H, Hβ), 9.25 (d, J = 4.5 Hz, 2H, Hβ),

9.05 (d, J = 2.5 Hz, 2H, Hβ), 8.83 (d, J = 4.6 Hz, 2H, Hβ), 8.20–8.15 (m, 2H phenyl-H), 7.85–

7.73 (m, 3H phenyl-H), 4.51 (m, 4H hexyl-CH2), 2.42–2.29 (m, 4H hexyl-CH2), 1.76 (dt, J =

14.9, 7.4 Hz, 4H hexyl-CH2), 1.53–1.44 (m, 4H hexyl-CH2), 1.39 (dt, J = 14.4, 7.0 Hz, 4H hexyl-CH2), 0.95 ppm (t, J = 7.2 Hz, 6H hexyl-CH3).

164

13 C NMR (101 MHz, CDCl3): δ = 150.7, 150.3, 149.6, 143.0, 137.7, 134.5, 132.6, 129.7, 128.8, 127.6, 126.7, 121.3, 120.9, 80.0, 39.1, 35.5, 32.0, 30.5, 22.9, 14.3 ppm.

UV-Vis (CH2Cl2): λmax [nm] (log ε) = 425 (6.57), 556 (5.04).

+ HRMS (MALDI) m/z calcd. for C38H39N4IZn [M] : 742.1511, found 742.1548.

(5,15-Dihexyl-10-phenylporphyrinato)nickel(II) (144p)

Synthesised via General Procedure 2 from 5,15-dihexyl-10-

phenylporphyrin (400 mg, 0.74 mmol), Ni(acac)2 (258 mg, 1.11 mmol) in

toluene (40 mL). Recrystallisation (CHCl3/MeOH) yielded purple crystals (290 mg, 0.49 mmol, 66%).

m.p. = 76–79 °C (dec.).

Rf = 0.745 (SiO2, n-hexane:CH2Cl2, 1:1, v/v).

1 H NMR (400 MHz, CDCl3): δ = 9.57 (d, J = 3.8 Hz, 1H, meso-H), 9.37 (t, J = 4.2 Hz, 2H,

Hβ), 9.29 (t, J = 4.3 Hz, 2H, Hβ), 9.09 (t, J = 4.3 Hz, 2H, Hβ), 8.78 (t, J = 4.3 Hz, 2H, Hβ), 8.00 (d, J = 5.4 Hz, 2H, phenyl-H), 7.70 (dd, J = 11.5, 8.5 Hz, 3H, phenyl-H), 4.59 (dd, J =

12.1, 6.5 Hz, 4H, hexyl-CH2), 2.31 (t, J = 11.2 Hz, 4H, hexyl-CH2), 1.62 (m, 4H, hexyl-CH2),

1.49–1.39 (m, 4H, hexyl-CH2), 1.34 (dd, J = 13.7, 6.9 Hz, 4H, hexyl-CH2), 0.92 ppm (dt, J =

10.8, 5.6 Hz, 6H, hexyl-CH3).

13 C NMR (101 MHz, CDCl3): δ = 142.6, 141.8, 141.3, 133.8, 132.4, 129.9, 129.3, 127.7, 126.9, 117.8, 103.5, 37.6, 34.3, 31.9, 30.2, 22.8, 14.3 ppm.

UV-Vis (CH2Cl2): λmax [nm] (log ε) = 410 (6.31), 526 (5.14).

+ HRMS (MALDI) m/z calcd. for C38H4ON4Ni [M] : 610.2606, found 610.2623.

6.2.3. Synthesis of cubanyl-porphyrin complexes

General Procedure 5: (Copper-free Sonogashira cross-coupling of alkynylcubanes with porphyrins):

This procedure was adapted from Lindsey and co-workers.[232] The ethynylcubane and porphyrin were placed in an oven-dried Schlenk flask and heated under vacuum. The flask was purged with argon and anhydrous THF and NEt3 were added by syringe in a 3:1 ratio. The solution was subjected to three freeze-pump-thaw cycles before releasing to argon.

Pd2(dba)3 and AsPh3 were added. The reaction mixture was heated to 65 °C for 18 hours and then diluted with CH2Cl2 (10 mL) before removal of solvents in vacuo. Crude products

165 were purified by column chromatography (SiO2). Alkynylcubane substituted porphyrins typically eluted as the second porphyrin containing fraction after any unreacted bromoporphyrin.

General Procedure 6: (Sonogashira cross-coupling of alkynylcubane and porphyrins):

PdCl2(PPh3)2 and CuI were added. The reaction mixture was heated to 65 °C for three hours and then diluted with CH2Cl2 (10 mL) before removal of solvents in vacuo. The crude material was purified by filtration through a short silica gel column using CH2Cl2 as eluent to remove any unreacted starting material.

General Procedure 7: (TMS deprotection with TBAF)

TMS protected porphyrin was dissolved in THF in a 50 mL RBF at room temperature. TBAF (1 M solution) was added and the reaction stirred for one hour until TLC analyses indicated complete consumption of starting material. The reaction mixture was poured into water and extracted with CH2Cl2.

General Procedure 8: (TMS deprotection with potassium carbonate)

The TMS-protected porphyrin was dissolved in THF, K2CO3 in MeOH was then added under atmospheric conditions and the solution was stirred vigorously for one hour. The reaction mixture was partitioned between H2O and CH2Cl2. The organic layer was washed with H2O

(2 × 100 mL) and dried over MgSO4 before a recrystallisation was carried out.

5,10,15-Triphenyl-20-((4'-((trimethylsilylethynyl)cuban-1-yl)ethynyl)porphyrin (147a)

Synthesised by modified General Procedure 6 from 1-ethynyl-4-(TMS-ethynyl)cubane 162 (165 mg, 0.73 mmol), 5-bromo-10,15,20- triphenylporphyrin 144d (179 mg, 0.29 mmol),

Pd(PPh3)4 (33 mg, 29 μmol), CuI (11 mg, 58 μmol) in THF (10 mL) and NEt3 (3 mL). The reaction was stirred at 65 °C for three hours.

Purification by column chromatography (SiO2; n-hexane:CH2Cl2, 3:1, v/v) followed by recrystallisation (CHCl3/n-hexane) gave the title compound as a purple solid (183 mg, 0.24 mmol, 83%).

166 m.p. >300 °C.

Rf = 0.25 (n-hexane:CH2Cl2, 1:1, v/v).

1 H NMR (400 MHz, CDCl3/C5D5N): δ = 9.61 (d, J = 4.6 Hz, 2H, Hβ), 8.85 (d, J = 4.6 Hz, 2H,

Hβ), 8.74 (s, 4H, Hβ), 8.11–8.15 (m, 6H, Ph-m-CH), 7.65–7.72 (m, 9H, Ph-o-/p-CH), 4.47–

4.49 (m, 3H, cubane-CH), 4.28–4.30 (m, 3H, cubane-CH), 0.21 ppm (s, 9H, Si-CH3).

13 C NMR (100 MHz, CDCl3/C5D5N): δ = 152.1, 150.2, 149.8, 149.6, 143.2, 143.1, 134.4, 132.3, 131.7, 131.4, 130.5, 127.2, 126.3, 126.2, 122.1, 121.3, 105.7, 99.6, 94.8, 94.0, 93.1, 50.0, 49.5, 46.9, 46.0, 0.2 ppm.

UV-Vis (CH2Cl2): λmax [nm] (log ε) = 434 (5.66), 566 (4.24), 604 (4.23).

+ HRMS (Maldi) m/z calcd. for C53H38N4SiZn [M] : 822.2157, found 822.2194.

[5,10,15-Triphenyl-20-((4'-((trimethylsilylethynyl)cuban-1-yl)ethynyl)- porphyrinato]zinc(II) (147b)

Synthesised by modified General Procedure 6 from cubanyl 162 (165 mg, 0.73 mmol), (5-bromo- 10,15,20-triphenylporphyrinato)zinc(II) 144e (197

mg, 0.29 mmol), Pd(PPh3)4 (33 mg, 29 μmol), CuI

(11 mg, 58 μmol) in THF (10 mL) and NEt3 (3 mL). The reaction was stirred at 65 °C for three hours. Purification by column chromatography

(SiO2, n-hexane:CH2Cl2, 3:1, v/v) followed by recrystallisation (CHCl3/n-hexane) gave the title compound as a purple solid (198 mg, 0.24 mmol, 83%). m.p. >300 °C.

Rf = 0.25 (n-hexane:CH2Cl2, 1:1, v/v).

1 H NMR (400 MHz, CDCl3/C5D5N): δ = 9.61 (d, J = 4.6 Hz, 2H, Hβ), 8.85 (d, J = 4.6 Hz, 2H,

Hβ), 8.74 (s, 4H, Hβ), 8.11–8.15 (m, 6H, Ph-m-CH), 7.65–7.72 (m, 9H, Ph-o/p-CH), 4.47-

4.49 (m, 3H, cubane-CH), 4.28–4.30 (m, 3H, cubane-CH), 0.21 ppm (s, 9H, Si-CH3).

UV-Vis (CH2Cl2): λmax [nm] (log ε) = 434 (5.66), 566 (4.24), 604 (4.23).

+ HRMS (Maldi) m/z calcd. for C53H38N4SiZn [M] : 822.2157, found 822.2194.

167

5-((4'-Ethynylcuban-1-yl)ethynyl)-10,15,20-triphenylporphyrin (147c)

Synthesised via General Procedure 7 from compound 147a (100 mg, 0.13 mmol) and TBAF (1 M solution, 0.40 mmol, 0.40 mL) in THF (10 mL). Recrystallisation

(CHCl3/MeOH) gave the title compound as a purple solid (73 mg, 0.11 mmol, 82%). m.p. >300 °C.

Rf = 0.42 (n-hexane:CH2Cl2, 1:1, v/v).

1 H NMR (400 MHz, CDCl3/C5D5N): δ = 9.57 (d, J = 4.6 Hz, 2H, Hβ), 8.82 (d, J = 4.6 Hz, 2H,

Hβ), 8.71 (s, 4H, Hβ), 8.11–8.15 (m, 6H, Ph-m-CH), 7.68–7.72 (m, 9H, Ph-o-/p-CH), 4.49– 4.50 (m, 3H, cubane-CH), 4.29–4.30 (m, 3H, cubane-CH), 2.88 (s, 1H, alkyne-CH), –2.43 ppm (s, 2H, NH).

13 C NMR (100 MHz, CDCl3/C5D5N): δ = 145.7, 141.9, 141.7, 134.4, 134.3, 127.8, 126.7, 126.6, 120.8, 117.6, 100.1, 96.1, 91.8, 83.5, 50.1, 49.1, 47.7, 46.2 ppm.

UV-Vis (CH2Cl2): λmax [nm] (log ε) = 430 (5.63), 532 (4.15), 572 (4.42), 608 (3.76), 666 (4.01).

+ HRMS (Maldi) m/z calcd. for C50H32N4 [M] : 688.2627, found 688.2645.

[5-((4'-Ethynylcuban-1-yl)ethynyl)-10,15,20-triphenylporphyrinato]zinc(II) (147d)

Synthesised via General Procedure 7 from compound 147b (300 mg, 0.36 mmol) and TBAF (1 M solution, 1 mmol, 1.0 mL) in THF (30 mL). Recrystallisation

(CHCl3/n-hexane) gave the title compound as a purple solid (230 mg, 0.31 mmol, 86 %). m.p. >300 °C.

Rf = 0.23 (n-hexane:CH2Cl2, 1:1, v/v).

1 H NMR (400 MHz, CDCl3/C5D5N): δ = 9.57 (d, J = 4.6 Hz, 2H, Hβ), 8.81 (d, J = 4.6 Hz, 2H,

Hβ), 8.70 (s, 4H, Hβ), 8.06–8.08 (m, 6H, Ph-m-CH), 7.60-7.64 (m, 9H, Ph-o/p-CH), 4.42– 4.45 (m, 3H, cubane-CH), 4.20–4.23 (m, 3H, cubane-CH), 2.85 ppm (s, 1H, alkyne-CH).

13 C NMR (100 MHz, CDCl3/C5D5N): δ = 152.0, 150.2, 149.7, 143.2, 143.1, 142.8, 134.5, 134.3, 132.3, 132.0, 131.9, 131.7, 131.4, 130.5, 127.2, 126.3, 126.2, 121.3, 99.5, 94.6, 93.1, 83.6, 50.1, 48.9, 46.1, 45.7 ppm.

168

UV-Vis (CH2Cl2): λmax [nm] (log ε) = 434 (5.66), 566 (4.30), 606 (4.28).

+ HRMS (Maldi) m/z calcd. for C50H30N4Zn[M] : 750.1762, found 750.1763.

5,15-Dihexyl-10-phenyl-20-[(4'-((trimethylsilyl)ethynyl)cuban-1-yl)ethynyl]porphyrin (147e)

Synthesised via General Procedure 5 from cubanyl 162 (73 mg, 325 μmol), porphyrin 144q

(82 mg, 130 μmol), Pd2(dba)3 (17.8 mg, 19.5 μmol)

and AsPh3 (79.6 mg, 260 μmol) in THF (6 mL) and

NEt3 (2 mL). The product 147e was obtained as purple crystals (12 mg, 15 μmol, 12%). m.p. = 119–124 °C (dec.).

Rf = 0.44 (SiO2, n-hexane:CH2Cl2, 1:1, v/v).

1 H NMR (400 MHz, CDCl3): δ = 9.68 (d, J = 4.2 Hz, 2H, Hβ), 9.48 (d, J = 4.1 Hz, 2H, Hβ),

9.41 (d, J = 4.3 Hz, 2H, Hβ), 8.86 (d, J = 4.3 Hz, 2H, Hβ), 8.16 (d, J = 6.1 Hz, 2H, phenyl-H),

7.75 (m, 3H, phenyl-H), 4.90 (d, J = 9.6 Hz, 4H, hexyl-CH2), 4.56 (t, J = 4.8 Hz, 3H, cubane-

CH), 4.39–4.35 (t, J = 4.8 Hz, 3H, cubane-CH), 2.51–2.46 (m, 4H, hexyl-CH2), 1.82–1.77

(m, 4H, hexyl-CH2), 1.48 (s, 4H, hexyl-CH2), 1.40 (d, J = 6.8 Hz, 4H, hexyl-CH2), 0.93 (s,

6H, hexyl-CH3), 0.26 (s, 9H, Si-CH3), –2.31 ppm (s, 2H, NH).

13 C NMR (151 MHz, CDCl3): δ = 151.5, 150.7, 150.1, 149.3, 142.6, 134.4, 132.5, 131.3, 129.6, 128.7, 127.8, 126.7, 120.8, 120.6, 99.0, 95.7, 94.3, 92.4, 50.2, 49.8, 47.8, 47.2, 38.7, 35.3, 32.1, 30.3, 29.9, 22.9, 14.3, 0.4 ppm.

UV-Vis (CH2Cl2): λmax [nm] (log ε) = 431 (5.49), 529 (3.94), 573 (4.2), 671 (3.86).

+ HRMS (MALDI) m/z calcd. for C53H56N4Si [M] : 776.4268, found 776.4274.

(5,15-Dihexyl-10-phenyl-20-((4'-((trimethylsilyl)ethynyl)cuban-1-yl)-ethynyl) porphyrinato)zinc(II) (147f)

Synthesised via General Procedure 6 from porphyrin 144f (50 mg, 72 μmol), cubanyl 162 (55

mg, 180 μmol), PdCl2(PPh3)2 (7.6 mg, 10.8 μmol)

and CuI (2.7 mg, 14.4 μmol) in THF and NEt3 (3:1, v/v). Recrystallisation from n-hexane afforded purple crystals (9 mg, 11 μmol, 16%).

169 m.p. > 300 °C.

Rf = 0.20 (SiO2, n-hexane:CH2Cl2, 2:1, v/v).

1 H NMR (400 MHz, CDCl3): δ = 9.55 (d, J = 4.6 Hz, 2H, Hβ), 9.37–9.33 (m, 4H, Hβ), 8.84 (d,

J = 4.6 Hz, 2H, Hβ), 8.17 (t, J = 5.9 Hz, 2H, phenyl-H), 7.82–7.71 (m, 3H, phenyl-H), 4.74–

4.67 (m, 4H, hexyl-CH2), 4.59–4.55 (m, 3H, cubane-CH), 4.41–4.36 (m, 3H, cubane-CH),

2.45 (dd, J = 15.3, 7.6 Hz, 4H, hexyl-CH2), 1.79 (dt, J = 14.9, 7.5 Hz, 4H, hexyl-CH2), 1.48

(d, J = 7.4 Hz, 4H, hexyl-CH2), 1.40 (td, J = 14.3, 6.9 Hz, 4H, hexyl-CH2), 0.94 (t, J = 7.2

Hz, 6H, hexyl-CH3), 0.27 ppm (s, 9H, Si-CH3).

13 C NMR (101 MHz, CDCl3): δ = 151.2, 150.4, 149.7, 149.1, 143.0, 134.4, 132.4, 131.1, 129.3, 128.6, 127.6, 126.6, 121.8, 121.5, 98.1, 95.1, 90.2, 86.1, 67.5, 50.2, 49.8, 49.5, 38.9, 35.6, 32.1, 30.5, 25.3, 22.9, 14.3, 0.4 ppm.

UV-Vis (CH2Cl2): λmax [nm] (log ε) = 433 (5.32), 566 (4.07), 615 (4.13).

+ HRMS (MALDI) m/z calcd. for C53H54N4SiZn [M] : 838.3409, found 838.3440.

5-((4'-Ethynylcuban-1-yl)ethynyl)-10,20-dihexyl-15-phenylporphyrin (147g)

Synthesised via General Procedure 8 from porphyrin

147e (12 mg, 15 μmol) in THF (10 mL) and K2CO3 (14 mg, 105 μmol) in MeOH (3 mL) and then recrystallisation (n-hexane) yielded purple crystals (7 mg, 10 μmol, 68%).

m.p. = 130 °C (dec.).

Rf = 0.43 (SiO2, n-hexane:CH2Cl2, 1:1, v/v).

1 H NMR (400 MHz, CDCl3): δ = 9.65 (d, J = 4.7 Hz, 2H, Hβ), 9.42 (d, J = 4.6 Hz, 2H, Hβ),

9.30 (d, J = 4.6 Hz, 2H, Hβ), 8.75 (d, J = 4.6 Hz, 2H, Hβ), 8.13 (d, J = 7.4 Hz, 2H, phenyl-H),

7.72 (dd, J = 13.7, 6.6 Hz, 3H, phenyl-H), 4.90–4.84 (m, 4H, hexyl-CH2), 4.59–4.54 (m, 3H, cubane- CH), 4.35 (t, 3H, cubane-CH), 2.92 (s, 1H, alkyne-CH), 2.46 (dt, J = 15.8, 8.1 Hz,

4H, hexyl- CH2), 1.75 (dt, J = 14.9, 7.6 Hz, 4H, hexyl-CH2), 1.46 (dd, J = 15.0, 7.7 Hz, 4H, hexyl-CH2), 1.35 (td, J = 14.5, 7.2 Hz, 4H, hexyl-CH2), 0.90 (t, J = 7.2 Hz, 6H, hexyl-CH3), –2.32 ppm (s, 2H, NH).

13 C NMR (101 MHz, CDCl3): δ = 142.6, 134.4, 131.9, 131.0, 128.6, 127.8, 126.7, 120.8, 120.6, 98.9, 95.5, 92.4, 83.8, 77.6, 50.3, 49.3, 47.9, 46.4, 38.7, 35.3, 32.0, 30.3, 29.9, 22.9, 14.3 ppm.

170

UV-Vis (CH2Cl2): λmax [nm] (log ε) = 431 (6.59), 533 (5.03), 573 (5.4).

+ HRMS (MALDI) m/z calcd. for C50H48N4 [M] : 704.3879, found 704.3909.

1,4-Di(phenylethynyl)cubane (147h)

Synthesised by modified General Procedure 6 from 1,4- diethynylcubane 158 (15.2 mg, 10 μmol), iodobenzene (112

μL, 1 mmol), PdCl2(PPh3)2 (7 mg, 1 μmol) and CuI (2.5 mg, 1.3 μmol) in anhydrous THF and NEt3 (1 mL:0.33 mL). The reaction mixture was heated to 65 °C for four. An aqueous work-up was performed with H2O, NaHCO3 and brine. The crude material was purified by preparative TLC n-hexane:CH2Cl2 in a 1:6 ratio. Recrystallisation (n-hexane) yielded pale yellow crystals (6 mg, 20 μmol, 20%). m.p. = 179–181°C (dec.).

Rf = 0.28 (SiO2, n-hexane: CH2Cl2, 8:1, v/v).

1 H NMR (600 MHz, CDCl3): δ = 7.40 (d, J = 4.2 Hz, 4H, phenyl-H), 7.29 (d, J = 5.0 Hz, 6H, phenyl-H), 4.17 ppm (s, 6H, cubane-CH).

13 C NMR (101 MHz, CDCl3): δ = 131.5, 128.4, 127.9, 123.8, 89.7, 89.0, 49.6, 46.9 ppm.

IR (neat)/cm-1: ν̃ = 2922 (m), 2851 (m), 2204 (w), 1595 (w), 1487 (m), 1441 (m), 1259 (m), 1023 (m), 796 (m), 749 (s), 686 (s), 588 (w).

+ HRMS (APCI) m/z calcd. for C24H17 [M] : 305.1325, found 305.1334.

(5,15-Dihexyl-10-phenyl-20-((4'-((trimethylsilyl)ethynyl)cuban-1'-yl)-ethynyl) porphyrinato)copper(II) (147i)

Synthesised via General Procedure 6 from (5- bromo-10,20-dihexyl-15- phenylporphyrinato)copper(II) (80 mg, 72 μmol),

cubanyl 146b (40 mg, 260 μmol), Pd(PPh3)4 (14.5 mg, 12.6 μmol) and CuI (4.8 mg, 25.2 μmol) in

THF/NEt3 (3 mL:1 mL). Recrystallisation (CHCl3/MeOH) yielded purple crystals (15 mg, 20 μmol, 27%). m.p. = 163 °C (dec.).

Rf = 0.51 (SiO2, n-hexane:CH2Cl2, 1:1, v/v).

171

1 H NMR (400 MHz, CDCl3): δ = 9.55 (d, J = 4.3 Hz, 2H, Hβ), 9.34 (t, J = 5.0 Hz, 4H, Hβ),

8.84 (d, J = 4.6 Hz, 2H, Hβ), 8.17 (d, J = 6.3 Hz, 2H, phenyl-H), 7.76 (dd, J = 14.9, 7.5 Hz,

3H, phenyl-H), 4.77 (s, 4H, hexyl-CH2), 4.62–4.57 (m, 3H, cubane-CH), 4.40–4.35 (m, 3H, cubane- CH), 2.46 (dt, J = 16.0, 7.9 Hz, 1H, alkyne-CH), 1.79 (dt, J = 14.7, 7.5 Hz, 4H, hexyl-CH2), 1.50 (dd, J = 21.8, 7.6 Hz, 4H, hexyl-CH2), 1.44–1.35 (m, 4H, hexyl-CH2), 0.94 ppm (t, J = 7.3 Hz, 6H, hexyl-CH3).

13 C NMR (101 MHz, CDCl3): δ = 150.9, 150.2, 149.4, 149.0, 143.1, 134.4, 132.3, 130.8, 128.9, 128.4, 127.6, 126.6, 121.3, 121.2, 99.2, 94.8, 92.6, 83.9, 77.7, 50.3, 49.3, 38.9, 32.0, 30.5, 22.9, 14.3 ppm.

UV-Vis (CH2Cl2): λmax [nm] (log ε) = 433 (6.74), 566 (5.2), 613 (5.39).

+ HRMS (MALDI) m/z calcd. for C50H47N4Cu [M] : 766.3097, found 766.3076.

Methyl [4-(4'-((10″,15″,20″-triphenylporphyrinato-5″-yl)ethynyl)phenyl)]zinc(II) cubane-1-carboxylate (148a)

Synthesised by modified General Procedure 6 from cubane 140f (35 mg, 130 μmol), porphyrin

141h (45 mg, 67 μmol), PdCl2(PPh3)2 (7 mg, 1 μmol) and CuI (2.5 mg, 1.3 μmol) in THF and

NEt3 (2 mL:0.66 mL). The reaction mixture was heated to 65 °C for four hours. An aqueous work-up was performed with H2O, NaHCO3 and brine. The crude material was purified by filtration through a short silica gel column using n- hexane:CH2Cl2 in a 1:1 ratio. Recrystallisation (CHCl3/MeOH) yielded purple crystals (33 mg, 38 μmol, 57%). m.p. > 300 °C (dec.).

Rf = 0.26 (SiO2, n-hexane:CH2Cl2, 1:2, v/v).

1 H NMR (600 MHz, CDCl3): δ = 9.83 (d, J = 4.5 Hz, 2H, Hβ), 9.00 (d, J = 4.5 Hz, 2H, Hβ),

8.87 (s, 4H, Hβ), 8.22–8.20 (m, 4H, phenyl-H), 8.19–8.17 (m, 2H, phenyl-H), 8.01 (d, J = 8.0 Hz, 2H, phenyl-H), 7.77 (m, 10H, phenyl-H), 7.39 (d, J = 8.0 Hz, 2H, phenyl-H), 4.32–4.30

(m, 3H, cubane-CH), 4.29–4.26 (m, 3H, cubane-CH), 3.74 ppm (s, 3H ester-CH3).

13 C NMR (101 MHz, CDCl3): δ = 172.8, 152.4, 150.7, 150.2, 150.0, 142.8, 142.7, 142.6, 134.6, 134.4, 133.0, 132.4, 132.1, 131.8, 131.1, 127.8, 126.8, 126.7, 125.4, 122.9, 122.2, 122.0, 100.6, 96.7, 92.5, 60.6, 60.5, 56.7, 51.8, 49.0, 46.3, 29.9, 21.2, 14.4 ppm.

UV-Vis (CH2Cl2): λmax [nm] (log ε) = 440 (5.7), 566 (4.34), 611 (4.52);

172

+ HRMS (MALDI) m/z calcd. for C56H36N4O2Zn [M] : 860.2130, found 860.2158.

[(4-{(10',15',20'-Triphenylporphyrinato-5'-yl)ethynyl)phenyl}]zinc(II) cubane-1- carboxylate (148b)

Compound 148a (58 mg, 67 μmol) and NaOH (809 mg, 20.2 mmol) were placed in a 50 mL RBF and THF and MeOH were added. The reaction mixture was heated to 65 °C for 48

hours and then diluted with H2O. Solution was acidified to pH 5 with HCl and extracted with a mixture of CHCl3 and THF, organic phase was dried over MgSO4 and solvents were removed in vacuo. Recrystallisation (CHCl3/n- hexane) yielded purple crystals (55mg, 65 μmol, 98%). m.p. > 300 °C.

Rf = 0 (SiO2, CH2Cl2).

1 H NMR (600 MHz, THF-d8): δ = 10.94 (s, 1H, CO2H), 9.80 (d, J = 4.5 Hz, 2H, Hβ), 8.90 (d,

J = 4.5 Hz, 2H, Hβ), 8.75 (s, 4H, Hβ), 8.20–8.18 (m, 4H, phenyl-H), 8.15 (d, J = 7.2 Hz, 2H, phenyl-H), 8.04 (d, J = 7.2 Hz, 2H, phenyl-H), 7.77–7.72 (m, 9H, phenyl-H), 7.45 (d, J = 7.8 Hz, 2H, phenyl-H), 4.25 ppm (s, 6H, cubane-CH).

13 C NMR (600 MHz, THF-d8): δ = 170.9, 153.3, 151.5, 151.0, 150.8, 144.4, 144.3, 143.9, 135.5, 135.4, 133.2, 132.6, 132.5, 132.4, 132.3, 131.4, 128.4, 127.9, 127.8, 127.4, 127.4, 126.2, 123.5, 123.2, 122.7, 100.5, 94.0, 60.7 , 49.9, 47.1, 30.8, 14.6 ppm.

UV-Vis (CHCl3): λmax [nm] (log ε) = 441 (5.18), 564 (3.96), 610 (4.01).

+ HRMS (MALDI) m/z calcd. for C55H34N4O2Zn [M] : 846.1973, found 846.1995.

6.3. Ch. 3 –Triptycene Cross-coupling and Triptycene–Porphyrin Arrays

6.3.1. Synthesis of Triptycene Precursors

9,10-Diethynyltriptycene 177e and its anthracene precursor 179e were synthesised as per the reported literature procedure.[134] While, 9-[(triisopropylsilyl)ethynyl]-10- ethynyltriptycene 177j, its precursors 179i and 177i were synthesised according to the procedure performed by Toyota and co-workers.[133] All compounds and intermediates had analytical data consistent with literature values.

173

6.3.2. Synthesis of Porphyrin Precursors

5,15-Bis(3'-methoxycarbonylphenyl)-10,20-dibromoporphyrin (182a)

Porphyrin 281 (100 mg, 0.173 mmol) was dissolved in CHCl3 (150 mL) and degassed for 30 minutes. At 0 °C, N-bromosuccinimide (34 mg, 0.19 mmol) and pyridine (0.1 mL) were added. The reaction was stirred at room temperature for three hours. The progress of the reaction was monitored by TLC and once all starting material had been consumed the reaction mixture was filtered through silica gel

using CH2Cl2 as eluent. The solvents were removed in vacuo and the

crude product was purified by recrystallisation from CHCl3/CH3OH. The product was obtained as purple crystals (19 mg, 26 μmol, 15%). m.p. = 298 °C (dec.).

Rf = 0.75 (SiO2, CH2Cl2:n-hexane, 3:1, v/v).

1 H NMR (400 MHz, CDCl3): δ = 9.63 (d, J = 4.6 Hz, 4H, Hβ), 8.85 (s, 2H, phenyl-H), 8.77 (d,

J = 4.6 Hz, 4H, Hβ), 8.52 (d, J = 7.7 Hz, 2H, phenyl-H), 8.34 (d, J = 7.7 Hz, 2H, phenyl-H),

7.88 (t, J = 7.7 Hz, 2H, phenyl-H), 4.01 (s, 6H, ester-CH3), –2.77 ppm (s, 2H, NH).

13 C NMR (101 MHz, CDCl3): δ = 167.3, 141.8, 138.5, 134.9, 129.5, 129.2, 127.2, 120.2, 104.2, 52.6 ppm.

IR (neat)/cm-1: ν̃ = 2923 (m), 2853 (w), 1726 (s, C=O), 1580 (w), 1463 (w), 1436 (m), 1286 (m), 1241 (s), 1190 (m), 1105 (m), 961 (m), 791 (s), 746 (s), 728 (s), 629 (m).

UV-Vis (CHCl3): λmax [nm] (log ε) = 424 (5.51), 523 (4.21), 558 (3.00), 603 (3.72), 660 (3.67).

+ HRMS (MALDI) m/z calcd. for C36H24N4O4Br2 [M] : 734.0164, found 734.0186.

[5,15-Bis(3'-methoxycarbonylphenyl)-10,20-dibromoporphyrinato]zinc(II) (182b)

Synthesised by General Procedure 1 from free base porphyrin 182a

(60 mg, 0.081 mmol), Zn(II)(OAc)2•2H2O (89 mg, 0.41 mmol in CHCl3 and MeOH (50 mL:15 mL) The porphyrin solution to stir at room temperature for four hours. The reaction process was monitored by TLC and once all the starting material was consumed the reaction

mixture was washed with NaHCO3, dried over MgSO4 and filtered

through silica gel using CH2Cl2 as the eluent. Purple crystals were obtained (60 mg, 75 μmol, 92%).

174 m.p. = 305–307 °C.

Rf = 0.22 (SiO2, CH2Cl2:n-hexane, 3:1, v/v).

1 H NMR (400 MHz, CDCl3): δ = 9.65 (d, J = 4.7 Hz, 4H, Hβ), 8.79 (s, 2H, phenyl-H), 8.76 (d,

J = 4.7 Hz, 4H, Hβ), 8.45 (d, J = 7.7 Hz, 2H, phenyl-H), 8.34 (d, J = 7.7 Hz, 2H, phenyl-H),

7.83 (t, J = 7.7 Hz, 2H, phenyl-H), 3.98 ppm (d, 6H, ester-CH3).

13 C NMR (101 MHz, CDCl3): δ = 167.7, 150.6, 150.3, 143.0, 138.5, 134.9, 133.3, 133.1, 128.9, 128.5, 126.8, 120.5, 105.1, 52.5 ppm.

IR (neat)/cm-1: ν̃ = 1697 (m, C=O), 1430 (w), 1286 (m), 1227 (m), 1022 (m), 997 (s), 791 (s), 755 (s), 732 (s), 696 (m).

UV-Vis (CHCl3): λmax [nm] (log ε) = 425 (5.68), 556 (4.34), 596 (3.82).

+ HRMS (MALDI) m/z calcd. for C36H22N4O4Br2Zn [M] : 795.9299, found 795.9333.

[5,15-Dihexyl-10-phenyl-20-((trimethylsilyl)ethynyl)porphyrinato]zinc(II) (195)

Synthesised by modified General Procedure 6 from bromoporphyrin 144q (50 mg, 71 μmol), TMSA (49 μL, 0.36

mmol), PdCl2(PPh3)2 (7.5 mg, 10 μmol) and CuI (2 mg, 20

μmol) in THF/NEt3 (2 mL:0.66 mL). The reaction was allowed to stir at 70 °C for four hours. The crude material was purified by filtration with silica gel column using CH2Cl2:n-hexane (2:1, v/v). Recrystallisation

(CHCl3/n-hexane) yielded purple crystals (35 mg, 49 μmol, 68%). m.p. = 263 °C.

Rf = 0.35 (SiO2, CH2Cl2:n-hexane, 1:1, v/v).

1 H NMR (400 MHz, CDCl3): δ = 9.63 (d, J = 4.6 Hz, 2H, Hβ), 9.35 (dd, 4H, Hβ), 8.84 (d, J =

4.6 Hz, 2H, Hβ), 8.17 (d, J = 6.5 Hz, 2H, phenyl-H), 7.80–7.73 (m, 3H, phenyl-H), 4.79–4.72

(m, 4H, hexyl-CH2), 2.45 (dt, J = 15.1, 7.8 Hz, 4H, hexyl-CH2), 1.79 (dt, J = 15.1, 7.8 Hz,

4H, hexyl-CH2), 1.55–1.51 (m, 4H, hexyl-CH2), 1.40 (dd, J = 14.7, 7.2 Hz, 4H, hexyl-CH2),

0.94 (t, J = 7.2 Hz, 6H, hexyl-CH3), 0.66 ppm (s, 9H, TMS-CH3).

13 C NMR (101 MHz, CDCl3): δ = 151.7, 150.6, 149.8, 149.0, 143.0, 134.4, 132.5, 131.2, 129.5, 128.5, 127.6, 126.6, 121.7, 108.1, 100.6, 98.3, 38.9, 35.6, 32.0, 30.4, 22.9, 14.3, 0.6 ppm.

IR (neat)/cm-1: ν̃ = 2954 (w), 2919 (w), 2852 (w), 2137 (w), 1246 (w), 1218 (w), 1072 (m), 1008 (m), 839 (s), 785 (s), 755 (m), 701 (s), 636 (m).

175

UV-Vis (CHCl3): λmax [nm] (log ε) = 432 (5.78), 565 (4.31), 611 (4.32).

+ HRMS (MALDI) m/z calcd. for C43H48N4SiZn [M] : 712.2940, found 712.2916.

6.3.3. Synthesis of Triptycene-linked Chromophores

General Procedure 9: (in situ deprotection with TBAF and Sonogashira cross-coupling of alkynyltriptycene and porphyrins/BODIPY):

TMS protected triptycene was dissolved in THF in a 25 mL Schlenk tube at room temperature under an inert atmosphere. TBAF (1 M solution) was added dropwise and the reaction stirred for five minutes until TLC analyses indicated complete consumption of starting material. NEt3 was added by syringe in a 3:1 ratio of THF to NEt3. The solution was bubbled with argon for five to ten minutes before PdCl2(PPh3)2 and CuI were added. The reaction mixture was heated to 70 °C for four hours and then diluted with CH2Cl2 (10 mL) before removal of solvents in vacuo. The crude material was purified by filtration through a short silica gel column using CH2Cl2 as eluent to remove any unreacted starting material.

[5,15-Dihexyl-10-phenyl-20-{(10'-(triisopropylsilyl)ethynyl)-9'-ethynyltriptycenyl} porphyrinato]zinc(II) (181d)

Synthesised by modified General Procedure 6 from porphyrin 144q (20 mg, 29 μmol), triptycene 177e

(20 mg, 44 μmol), PdCl2(PPh3)2 (3 mg, 4.35 μmol)

and CuI (1 mg, 8.7 μmol) in THF and NEt3 (0.75 mL:0.25 mL). The reaction was allowed to stir at 70 °C for four hours. The crude material was purified by filtration through a short silica gel column using n-hexane:CH2Cl2 (4:1, v/v). Recrystallisation

(CHCl3/n-hexane) yielded purple crystals (19 mg, 18 μmol, 60%). m.p. = 296 °C.

Rf = 0.65 (SiO2, CH2Cl2:n-hexane, 4:1, v/v).

1 H NMR (600 MHz, CDCl3): δ = 10.07 (d, J = 4.6 Hz, 2H, Hβ), 9.60 (d, J = 4.6 Hz, 2H, Hβ),

9.45 (d, J = 4.6 Hz, 2H, Hβ), 8.92 (d, J = 4.6 Hz, 2H, Hβ), 8.49 (d, J = 7.4 Hz, 3H, triptycene- H), 8.21–8.19 (m, 2H, phenyl-H), 7.98 (d, J = 7.4 Hz, 3H, triptycene-H), 7.82–7.74 (m, 3H, phenyl-H), 7.31 (td, J = 7.3, 1.0 Hz, 3H, triptycene-H), 7.27 (td, J = 7.5, 1.2 Hz, 3H, triptycene-H), 4.94–4.87 (m, 4H, hexyl-CH2), 2.54 (dt, J = 15.2, 7.6 Hz, 4H, hexyl-CH2), 1.85

(dt, J = 15.2, 7.6 Hz, 4H, hexyl-CH2), 1.54 (d, J = 7.6 Hz, 4H, hexyl-CH2), 1.43 (dd, J = 7.6,

2.5 Hz, 4H, hexyl-CH2), 1.40 (dd, 21H, TIPS-CH3), 0.95 ppm (t, J = 7.6 Hz, 6H, hexyl-CH3).

176

13 C NMR (151 MHz, CDCl3): δ = 151.9, 150.8, 150.1, 149.3, 144.7, 144.1, 142.9, 134.4, 132.8, 132.3, 131.3, 130.2, 128.8, 127.7, 126.7, 126.2, 123.0, 122.8, 122.1, 101.6, 98.1, 95.9, 94.4, 89.9, 54.4, 53.9, 39.1, 35.8, 32.1, 30.5, 29.8, 22.9, 19.0, 14.3, 11.7 ppm.

IR (neat)/cm-1: ν̃ = 2921 (m), 2851 (m), 1451 (m), 1305 (m), 1211 (m), 1072 (m), 1008 (s), 939 (m), 787 (s), 750 (s), 708 (s), 639 (s).

UV-Vis (CHCl3): λmax [nm] (log ε) = 430 (5.84), 563 (4.30), 608 (4.35).

+ HRMS (MALDI) m/z calcd. for C71H72N4SiZn [M] : 1072.4818, found 1072.4800.

[5,15-Dihexyl-10-(9'-ethynyl(10'-ethynyltriptycenyl))-20-phenylporphyrinato]zinc(II) (181g)

Synthesised by General Procedure 7 from compound 181d (7 mg, 7 μmol) and TBAF (10 μL, 10 μmol) in THF (2 mL). The reaction was stirred under Ar at room temperature for five minutes. The crude material was purified by filtration through a short silica gel column using

CH2Cl2. Recrystallisation (CHCl3/n-hexane) to yielded purple crystals (5 mg, 5 μmol 87%). m.p. = 252 °C.

Rf = 0.27 (SiO2, CH2Cl2:n-hexane, 1:1, v/v).

1 H NMR (600 MHz, CDCl3): δ = 10.08 (d, J = 4.5 Hz, 2H, Hβ), 9.62 (d, J = 4.6 Hz, 2H, Hβ),

9.46 (d, J = 4.6 Hz, 2H, Hβ), 8.92 (d, J = 4.6 Hz, 2H, Hβ), 8.50 (d, J = 7.0 Hz, 3H, triptycene- H), 8.21–8.18 (m, 2H, phenyl-H), 7.96 (d, 3H, J = 7.0 Hz, triptycene-H), 7.82–7.74 (m, 3H, phenyl-H), 7.32 (td, J = 7.3, 1.0 Hz, 3H, triptycene-H), 7.27 (td, J = 7.3, 1.0 Hz, 3H, triptycene-H), 4.93 (t, 4H, J = 7.8 Hz, hexyl-CH2), 3.43 (s, 1H, alkynyl-H), 2.55 (dt, J = 15.5,

7.8 Hz, 4H, hexyl-CH2), 1.88–1.81 (m, 4H, hexyl-CH2), 1.54 (dt, J = 15.1, 7.4 Hz, 4H, hexyl-

CH2), 1.41 (dt, J = 14.6, 7.4 Hz, 4H, hexyl-CH2), 0.95 ppm (t, J = 7.4 Hz, 6H, hexyl-CH3).

13 C NMR (151 MHz, CDCl3): δ = 151.9, 150.9, 150.1, 149.3, 144.7, 144.1, 142.9, 134.4, 132.8, 132.3, 131.3, 130.2, 128.8, 127.7, 126.7, 126.2, 126.1, 123.0, 122.8, 122.1, 122.0, 101.6, 98.1, 95.9, 94.4, 89.9, 54.4, 53.9, 39.1, 35.8, 32.1, 30.5, 22.9, 19.1, 14.3, 11.6 ppm.

IR (neat)/cm-1: ν̃ = 2922 (m), 2858 (m), 1453 (m), 1259 (m), 1032 (s), 789 (s), 751 (s), 702 (m), 640 (s).

UV-Vis (CHCl3): λmax [nm] (log ε) = 431 (5.57), 565 (4.06), 609 (4.09).

+ HRMS (MALDI) m/z calcd. for C62H52N4Zn [M] : 916.3483, found 916.3505.

177

9,10-Bis-[5'-ethynyl(10″,20″-dihexyl-15″-phenylporphyrinato)zinc(II)]triptycene (181b)

Synthesised by General Procedure 9 from triptycene 177e (23 mg, 50 μmol), TBAF (150 μL, 150 μmol), bromoporphyrin (87 mg, 125 μmol),

PdCl2(PPh3)2 (3.4 mg, 4.8 μmol) and

CuI (1.7 mg, 8.9 μmol) in THF/NEt3 (2 mL:0.66 mL). The reaction was allowed to stir for four hours at 70 °C. The crude material was purified by filtration with a silica gel column using

6:1 n-hexane:EtOAc. Recrystallisation (CHCl3/n-hexane) yielded purple crystals (40 mg, 26 μmol, 50%). m.p. = 332–335 °C.

Rf = 0.54 (SiO2, CH2Cl2:n-hexane, 2:1, v/v).

1 H NMR (400 MHz, CDCl3): δ = 10.11 (d, J = 4.5 Hz, 4H, Hβ), 9.62 (d, J = 4.5 Hz, 4H, Hβ),

9.38 (d, J = 4.6 Hz, 4H, Hβ), 8.79 (d, J = 4.6 Hz, 4H, Hβ), 8.60 (dd, J = 5.5, 3.3 Hz, 6H, triptycene-H), 8.12 (d, J = 6.2 Hz, 4H, phenyl-H), 7.70–7.64 (m, 6H, phenyl-H), 7.37 (dd, J

= 5.5, 3.3 Hz, 6H, triptycene-H), 4.98–4.92 (m, 8H, hexyl-CH2), 2.51–2.48 (m, 8H, hexyl-

CH2), 1.84–1.76 (m, 8H, hexyl-CH2), 1.52–1.44 (m, 8H, hexyl-CH2), 1.39–1.30 (m, 8H, hexyl-CH2), 0.87 ppm (t, J = 7.3 Hz, 12H, hexyl-CH3).

13 C NMR (101 MHz, CDCl3): δ = 151.8, 150.7, 150.0, 148.9, 145.1, 134.3, 132.3, 131.0, 129.9, 128.3, 127.3, 126.3, 126.2, 123.2, 121.6, 96.8, 88.9, 54.639.2, 35.8, 31.9, 30.4, 22.7, 14.0 ppm.

UV-Vis (CHCl3): λmax [nm] (log ε) = 434 (6.06), 564 (4.28), 610 (4.20).

IR (neat)/cm-1: ν̃ = 2921 (m), 2851 (m), 1451 (m), 1305 (m), 1211 (m), 1072 (m), 1008 (s), 939 (m), 787 (s), 750 (s), 708 (s), 639 (s).

+ HRMS (MALDI) m/z calcd. for C100H90N8Zn2 [M] : 1530.5871, found 1530.5919.

9,10-Bis-[4'-ethynyl(phenyl)-1'-borondipyrromethene]triptycene (181a)

Synthesised by General Procedure 9 from triptycene 177e (23 mg, 50 μmol), 1 M TBAF (150 μL, 150 μmol), BODIPY 154a

(43.5 mg, 125 μmol), PdCl2(PPh3)2 (3.4 mg, 4.8 μmol) and CuI (1.7 mg, 8.9 μmol) in THF/NEt3 (2 mL:0.66 mL). The reaction was

178 allowed to stir for four hours at 70 °C. The crude material was purified by filtration with a silica gel column using CH2Cl2:n-hexane (2:1, v/v). Recrystallisation (CHCl3/n-hexane) yielded red crystals (12.5 mg, 15 μmol, 30%). m.p. >300 °C.

Rf = 0.75 (SiO2, CH2Cl2:n-hexane, 3:1, v/v).

1 H NMR (600 MHz, CDCl3): δ = 8.03–7.99 (m, 8H, α-pyrrole-H and phenyl-H), 7.88 (dd, J = 5.2, 3.0 Hz, 6H, triptycene-H), 7.72 (d, 4H, phenyl-H), 7.19 (dd, J = 5.2, 3.0 Hz, 6H, triptycene-H), 7.03 (d, 4H, pyrrole-H), 6.62 ppm (s, 4H, pyrrole-H).

13 C NMR (151 MHz, CDCl3): δ = 146.4, 144.7, 143.5, 135.0, 134.3, 132.3, 131.6, 130.9, 126.3, 126.2, 125.6, 122.5, 119.0, 92.3, 86.4, 29.8 ppm.

11 B NMR (128 MHz, CDCl3): δ = 0.32 (t, J = 28.7 Hz).

19 F NMR (377 MHz, CDCl3): δ = -145.04 (dd, J = 57.5, 28.7 Hz).

IR (neat)/cm-1: ν̃ = 2921 (w), 2851 (w), 1560 (m), 1535 (m), 1412 (m), 1386 (s), 1257 (s), 1154 (m), 1103 (s), 1070 (s), 982 (m), 912 (m), 750 (s), 739 (m), 638 (m), 580 (w)

UV-Vis (CHCl3): λmax [nm] (log ε) = 367 (5.57), 506 (5.97).

+ HRMS (MALDI) m/z calcd. for C54H32N4F4B2 [M] : 834.2749, found 834.2748.

9-((5',15'-Dihexyl-10'-phenylporphyrinato-20'-yl)ethynyl)zinc(II)-10-((5',15'-dihexyl- 10'-phenylporphyrinato-20'-yl)ethynyl)nickel(II) triptycene

Porphyrin 144i (15.6 mg, 2.2 μmol) and triptycene 181g (14 mg, 15 μmol) were placed in an oven-dried Schlenk flask and heated under vacuum. The flask was purged with argon and anhydrous

THF/NEt3 (1 mL:0.33 mL) were added by syringe. Argon was bubbled through the solution for five minutes. PdCl2(PPh3)2 (1.5 mg, 2 μmol) and CuI (1 mg, 4 μmol) were added and the reaction was allowed to stir at 70 °C for 18 hours and then diluted with CH2Cl2 (10 mL) before removal of solvents in vacuo. The crude material was purified by filtration with a silica gel column using 2:1: CH2Cl2:n-hexane.

Recrystallisation (CHCl3/n-hexane) yielded purple crystals (2 mg, 1.3 μmol, 8%). m.p. > 300 °C.

Rf = 0.7 (SiO2, CH2Cl2:n-hexane, 2:1, v/v).

179

1 H NMR (600 MHz, CDCl3) δ = 10.23 (d, J = 4.3 Hz, 4H, Hβ), 9.70 (d, J = 4.3 Hz, 4H, Hβ),

9.49 (d, J = 4.3 Hz, 4H, Hβ), 8.95 (d, J = 4.4 Hz, 4H, Hβ), 8.70 (dd, J = 5.2, 3.3 Hz, 6H, triptycene-H), 8.23 (d, J = 6.7 Hz, 4H, phenyl-H), 7.78 (t, J = 6.7 Hz, 6H, phenyl-H), 7.48

(dd, J = 5.6, 2.4 Hz, 6H, triptycene-H), 5.00–4.95 (m, 8H, hexyl-CH2), 2.63–2.55 (m, 8H, hexyl-CH2), 1.92–1.84 (m, 8H, hexyl-CH2), 1.60–1.55 (m, 8H, hexyl-CH2), 1.48–1.41 (m,

8H, hexyl-CH2), 0.97 (t, J = 7.3 Hz, 12H).

13 C NMR (151 MHz, CDCl3) δ = 152.0, 150.9, 150.2, 149.3, 145.1, 142.9, 134.4, 132.8, 131.4, 130.3, 128.8, 127.7, 126.7, 126.5, 123.4, 122.1, 98.2, 96.2, 90.0, 67.8, 39.2, 35.9, 32.1, 30.6, 22.9, 14.3.

IR (neat)/cm-1: ν̃ = 2920 (w), 2852 (w), 1599 (w), 1451 (m), 1334 (w), 1305 (w), 1210 (m), 1071 (m), 1008 (s), 787 (s), 749 (s), 699 (s), 638 (s).

UV-Vis (CHCl3): λmax [nm] (log ε) = 433 (5.77), 563 (4.32), 609 (4.31).

+ HRMS (MALDI) m/z calcd. for C100H90N8NiZn [M] : 1524.5933, found 1524.5970.

[5,15-Bis{(10'-((triisopropylsilyl)ethynyl)-9'-triptycenyl)ethynyl}-10,20-bis(3'- methoxycarbonylphenyl)porphyrinato]zinc(II) (181f)

Dibromoporphyrin 182b (16 mg, 20 μmol) and triptycene 177j (23 mg, 50 μmol) were placed in an oven- dried Schlenk flask and heated under vacuum. The flask was purged

with argon and anhydrous THF/NEt3 (1 mL :0.33 mL) were added by syringe. Argon was bubbled through the solution for five minutes. PdCl2(PPh3)2 (2 mg, 3 μmol) and CuI (0.7 mg, 6 μmol) were added and the reaction was allowed to stir at 70 °C for 18 hours and then diluted with CH2Cl2 (10 mL) before removal of solvents in vacuo. The crude material was purified by filtration with silica gel column using CH2Cl2:n-hexane (1:1, v/v). Recrystallisation (CHCl3/n-hexane) yielded green crystals (6 mg, 4 μmol, 19%). m.p. = 284 °C.

Rf = 0.15 (SiO2, CH2Cl2:n-hexane, 2:1, v/v).

1 H NMR (400 MHz, CDCl3): δ = 10.12 (d, J = 4.4 Hz, 4H, Hβ), 9.02 (d, J = 4.4 Hz, 4H, Hβ), 8.94 (s, 2H. phenyl-H), 8.49 (dd, J = 18.6, 7.7 Hz, 4H, phenyl-H), 8.41 (d, J = 6.2 Hz, 6H,

180 triptycene-H), 7.96 (d, J = 6.2 Hz, 6H, triptycene-H), 7.91 (d, J = 7.7 Hz, 2H, phenyl-H), 4.00

(s, 6H, ester-CH3), 1.37 ppm (d, 42H, TIPS- CH3).

13 C NMR (101 MHz, CDCl3): δ = 167.3, 152.7, 150.3, 144.3, 143.9, 143.4, 143.1, 142.4, 138.2, 134.7, 133.2, 131.8, 130.9, 129.2, 128.9, 128.8, 127.0, 126.0, 122.7, 101.2, 94.4, 68.5, 54.2, 53.7, 52.4, 37.1, 33.5, 32.2, 31.9, 31.2, 29.7, 29.4, 26.4, 22.6, 19.8, 18.9, 14.4, 14.1, 11.5 ppm.

IR (neat)/cm-1: ν̃ = 2926 (w), 1723 (w, C=O), 1452 (w), 1260 (m), 1017 (s), 1249 (m), 794 (s), 751 (s), 689 (m), 668 (w), 670 (w).

UV-Vis (CHCl3): λmax [nm] (log ε) = 436 (5.92), 573 (4.46), 616 (4.75).

+ HRMS (MALDI) m/z calcd. for C102H88N4O4Si2Zn [M] : 1552.5636, found 1552.5616.

6.3.4. Synthesis of Ethynylbenzene Compounds

5,15-Dihexyl-10-phenyl-20-((4'-((trimethylsilyl)ethynyl)phenyl)ethynyl)porphyrin

Synthesised via General Procedure 5 from ethynylbenzene 189 (30 mg, 150 μmol),

bromoporphyrin 144f (38 mg, 60 μmol), Pd2(dba)3

(8.24 mg, 9 μmol) and AsPh3 (37 mg, 120 μmol) in

THF/NEt3 (3:1, 4 mL). Recrystallisation

(CHCl3/CH3OH) yielded purple crystals (38 mg, 50 μmol, 84%). m.p. = 157–159 °C (dec.).

Rf = 0.43 (SiO2, n-hexane:CH2Cl2, 2:1, v/v).

1 H NMR (400 MHz, CDCl3): δ = 9.67 (d, J = 4.7 Hz, 2H, Hβ), 9.35 (d, J = 4.8 Hz, 2H, Hβ),

9.26 (d, J = 4.8 Hz, 2H, Hβ), 8.76 (d, J = 4.8 Hz, 2H, Hβ), 8.18–8.13 (d, 2H, J = 6.7 Hz, phenyl-H), 7.98 (d, J = 8.2 Hz, 2H, phenyl-H), 7.82–7.68 (m, 5H, phenyl-H), 4.82–4.73 (m,

4H, hexyl-CH2), 2.50–2.39 (m, 4H, hexyl-CH2), 1.75 (dt, J = 15.1, 7.5 Hz, 4H, hexyl-CH2),

1.53–1.42 (m, 4H, hexyl-CH2), 1.38 (dq, J = 14.3, 7.1 Hz, 4H, hexyl-CH2), 0.93 (t, J = 7.2

Hz, 6H, hexyl-CH2), 0.36 (s, 9H, hexyl-CH3), –2.36 ppm (s, 2H, NH).

13 C NMR (101 MHz, CDCl3): δ = 142.8, 134.7, 132.7, 131.7, 128.2, 127.0, 124.8, 123.4, 121.4, 105.4, 97.9, 96.9, 96.3, 95.2, 38.9, 35.5, 32.3, 31.4, 30.6, 30.2, 23.2, 14.6, 0.5 ppm.

UV-Vis (CH2Cl2): λmax [nm] (log ε) = 439 (5.5), 582 (4.52), 675 (4.25).

+ HRMS (MALDI) m/z calcd. for C51H54N4Si [M] : 750.4118, found 750.4138.

181

[5-((4'-Bromophenyl)ethynyl)-10,20-bis(4'-methylphenyl)-15- phenylporphyrinato]zinc(II) (203)

Synthesised by General Procedure 9 from ((4- bromophenyl)ethynyl)trimethylsilane 198 (140 mg, 0.55 mmol), 1 M TBAF (0.8 mL, 0.825 mmol), iodoporphyrin

202 (80 mg, 0.11 mmol), PdCl2(PPh3)2 (11.5 mg, 17 μmol)

and CuI (6.5 mg, 34 μmol) in THF (2 mL) and NEt3 (0.66 mL). The reaction was allowed to stir at 70 °C for three hours. The crude material was purified by filtration through a short silica gel column using n-hexane:CH2Cl2 (2:1, v/v).

Recrystallisation (CHCl3/n-hexane) yielded purple-green crystals (69 mg, 85 μmol, 78%). m.p. = 215 °C.

Rf = 0.67 (SiO2, CH2Cl2:n-hexane, 2:1, v/v).

1 H NMR (400 MHz, CDCl3): δ = 9.69 (d, J = 4.6 Hz, 2H, Hβ), 8.99 (d, J = 4.6 Hz, 2H, Hβ),

8.91 (d, J = 4.6 Hz, 2H, Hβ), 8.87 (d, J = 4.6 Hz, 2H, Hβ), 8.19 (d, J = 6.7 Hz, 2H, phenyl-H), 8.10 (d, J = 7.8 Hz, 4H, tolyl-H), 7.77 (dd, J = 10.2, 7.8 Hz, 4H, phenyl-H), 7.60 (d, J = 7.6

Hz, 7H, tolyl-H and phenyl-H), 2.76 ppm (s, 6H, tolyl-CH3).

13 C NMR (101 MHz, CDCl3): δ = 152.1, 150.7, 150.0, 142.6, 139.5, 137.3, 134.4, 134.3, 132.9, 132.9, 132.2, 132.0, 131.8, 131.7, 131.4, 130.5, 127.4, 126.6, 123.2, 122.8, 122.2, 94.9, 93.8, 21.5 ppm.

IR (neat): ν̃ = 2919 (w), 2190 (w), 1598 (w), 1486 (m), 1440 (w), 1339 (m), 1260 (m), 1065 (m), 1180 (w), 1065 (m), 997 (s), 792 (s), 700 (m), 752 (m).

UV-Vis (CH2Cl2): λmax [nm] (log ε) = 441 (5.84), 567 (4.50), 613 (4.64).

+ HRMS (MALDI) m/z calcd. for C48H31N4BrZn [M] : 806.1024, found 806.1041.

[5,15-Bis(4'-methylphenyl)-10-phenyl-20- ((trimethylsilyl)ethynyl)porphyrinato]nickel(II) (201)

Synthesised by General Procedure 6 from iodoporphyrin 200

(46 mg, 0.06 mmol), TMSA (42 μL, 0.3 mmol), PdCl2(PPh3)2 (6

mg, 9 μmol) and CuI (2 mg, 18 μmol) in THF and NEt3 (2 mL:0.66 mL). The reaction was allowed to stir at 70 °C for three hours. Then the crude material was purified by filtration through a short silica gel column using 2:1 n-hexane: CH2Cl2. Recrystallisation (CHCl3/n-hexane) yielded purple crystals (38 mg, 53 μmol, 88%).

182 m.p. = >265 °C.

Rf = 0.61 (SiO2, CH2Cl2:n-hexane, 2:1, v/v).

1 H NMR (600 MHz, CDCl3): δ = 9.49 (d, J = 4.8 Hz, 2H, Hβ), 8.81 (d, J = 4.8 Hz, 2H, Hβ),

8.67 (dd, J = 12.7, 4.8 Hz, 4H, Hβ), 7.98 (d, 2H, phenyl-H), 7.87 (d, J = 7.7 Hz, 4H, tolyl-H),

7.69–7.64 (m, 3H, phenyl-H), 7.49 (d, J = 7.7 Hz, 4H, tolyl-H), 2.65 (s, 6H, tolyl-CH3), 0.54 ppm (s, 9H, TMS-CH3).

13 C NMR (151 MHz, CDCl3): δ = 145.2, 143.5, 142.7, 142.6, 140.9, 137.8, 137.9, 133.9, 133.7, 133.2, 132.4, 132.2, 131.7, 127.9, 127.8, 127.0, 120.5, 120.0, 102.3, 98.7, 21.6, 0.4 ppm.

IR (neat): ν̃ = 2920 (w), 2851 (w), 2146 (w), 1461 (w), 1350 (m), 1249 (m), 1210 (w), 1180 (w), 1071 (m), 1003 (s), 839 (s), 795 (s), 750 (m), 709 (m), 698 (s).

UV-Vis (CH2Cl2): λmax [nm] (log ε) = 432 (5.70), 563 (4.32), 604 (4.18).

+ HRMS (MALDI) m/z calcd. for C45H36N4SiNi [M] : 718.2063, found 718.2083.

6.4. Ch. 4 – Bridging and Conformational Control of Porphyrin Units through Non-Traditional Rigid Scaffolds

6.4.1. Synthesis of Cubane Precursors

The cubane precursors for the amide condensation, cubane di- and monocarboxylic acid 28 and 76, were also obtained using Tsanaktsidis’ method.[61] All compounds and intermediates had analytical data consistent with literature values.[327]

6.4.2. Synthesis of Dipyrromethane and Porphyrin Precursors

Tetraphenylporphyrin (TPP) 217 was synthesised according to the Adler–Longo method.[328] 5-(4’-Aminophenyl)-10,15,20-triphenylporphyrin 213e and its zinc(II) complex 213f were synthesised by regioselective mono-nitration at the p-position of meso-phenyl followed by a reduction.[275] meso-Free dipyrromethane (DPM) 149 was synthesised according to the procedure reported by Lindsey and co-workers.[235b] 5,15-Bis(4'-methylphenyl)-10- phenylporphyrin was afforded via a standard porphyrin condensation[235c] followed by Senge reaction conditions.[323] 5-Iodo-10,20-bis(4'-methylphenyl)-15-phenylporphyrin was obtained via performing known procedures in literature.[326]

All known compounds had analytical data consistent with those previously reported in the literature.

183

General Procedure 10: (HATU coupling[277] for amide condensation reactions):

4-(Methoxycarbonyl)cubane-1-carboxylic acid 76 (1.0 eq.) or cubane-1,4-dicarboxylic acid 28 (1.0 eq.) was placed in an oven-dried microwave vial and heated under vacuum. The reaction +flask was purged with argon, anhydrous DMF (0.25 mL) was added and the reaction mixture was heated slightly to dissolve the cubane. HATU (1.3/2.6 eq.), HOAt (1.3/2.6 eq.) and DIPEA (4.0/8.0 eq.) were then added and the reaction mixture was left to stir at room temperature. for 30 minutes under argon. The amine (2.0 eq.) was added under an argon flow to the flask alongside additional anhydrous DMF (0.25 mL). The reaction mixture was stirred at room temperature for a further 24 hours and then diluted with H2O.

General Procedure 11: (Oxalyl chloride coupling[272] for amide condensation reactions):

In an oven-dried microwave vial, a drop of DMF was added to the solution of 4- (methoxycarbonyl)cubane-1-carboxylic acid (1.0 eq.) 76 or cubane-1,4-dicarboxylic acid 28 in CH2Cl2. Oxalyl chloride (1.2 eq./2.2 eq.) was added dropwise to above solution at 0 °C. The reaction mixture was warmed to 40 °C and stirred for two hours under inert atmosphere.

The reaction mixture was cooled to 0 °C, NEt3 (3.0/6.0 eq.) followed by substituted aniline (1.1/2.0 eq.) were added slowly to the reaction mixture. The reaction vial was warmed to 40 °C and stirred for two hours under argon. The solvent was evaporated in vacuo, crude reaction mixture was recrystallised from CH2Cl2/n-hexane. The desired product was obtained as a white crystalline material.

General Procedure 12 to synthesise meso-ethynylamine substituted porphyrins (213a‒ 213d)

Iodoporphyrin 202 (1.0 eq.) was placed in an oven-dried Schlenk flask and heated under vacuum. The reaction flask was purged with argon and a mixture of THF/NEt3 (3:1) was added. Argon was bubbled through the solution for 15 minutes then 2-/3-/4-ethynylaniline

(5.0 eq.), PdCl2(PPh3)2 (0.15 eq.) and CuI (0.3 eq.) were added. The reaction mixture was heated to 70 °C and allowed to stir for four hours. The reaction mixture was diluted with

CH2Cl2 (10 mL) followed by removal of solvents in vacuo.

General Procedure 13 for Sonogashira cross-coupling using meso-ethynylporphyrin:

An oven-dried Schlenk tube charged with ethynylporphyrin (1.0 eq.) and iodo-substituted cubane was heated under vacuum. THF (5 mL) followed by NEt3 (2.5 mL) were added to reaction vessel. Argon was bubbled through the solution for 10–15 minutes and

PdCl2(PPh3)2 (0.2 eq.) and CuI (0.3 eq.) were added. The resulting reaction mixture was heated at 40 °C and progress of the reaction was monitored by TLC. Reaction mixture was

184 filtered through a ceilite pad. Solvent was evaporate in vacuo, crude reaction mixture was purified by silica gel column chromatography.

General Procedure 14 for Suzuki cross-coupling:

To an oven-dried Schlenk tube charged with porphyrin (2.1 eq.), cubane linker (1.0 eq.) and

K3PO4 (10.0 eq.) anhydrous DMF (5 mL) was added under inert atmosphere. The above solution was purged with argon for further 15 minutes followed by addition of 0.2 eq. of

Pd(PPh3)4. The reaction mixture was heated to 100 °C and allowed to stir for 18 hours. The solvent was removed in vacuo, crude reaction mixture was dissolved in CH2Cl2 washed with

NaHCO3 followed by brine. The organic layer was extracted with CH2Cl2, extracted organic phases were combined and solvent was evaporated. The resulting crude reaction mixture was purified by silica gel column chromatography.

6.4.3. Synthesis of Porphyrins for Cubanyl Amide-linked Complexes

[5-(4'-Aminophenylacetylene)-10,20-bis(4'-methylphenyl)-15- phenylporphyrinato]zinc(II) (213a)

Synthesised via General Procedure 12 from [5-iodo- 10,20-bis(4'-methylphenyl)-15- phenylporphyrinato]zinc(II) 202 (100 mg, 130 μmol), 4-

ethynylaniline (76 mg, 660 μmol), PdCl2(PPh3)2 (14 mg,

20 μmol) and CuI (7.5 mg, 40 μmol) in THF/NEt3 (2 mL/0.66 mL). The reaction was heated to 70 °C and allowed to stir for four hours. The crude material was purified by column chromatography (Al2O3, activated, Brockmann Grade III),

CH2Cl2:n-hexane, 2:1, v/v). Recrystallisation (CHCl3/n-hexane) yielded green crystals (85 mg, 0.11 mmol, 88%). m.p. >300 °C.

Rf = 0.64 (SiO2, EtOAc:n-hexane, 1:1, v/v).

1 H NMR (600 MHz, THF-d8): δ = 9.75 (d, J = 4.5 Hz, 2H, Hβ), 8.88 (d, J = 4.5 Hz, 2H, Hβ),

8.73 (dd, J = 23.2, 4.5 Hz, 4H, Hβ), 8.14 (d, 2H, phenyl-H), 8.06 (d, J = 7.6 Hz, 4H, tolyl-H), 7.75–7.71 (m, 5H, phenyl-H), 7.57 (d, J = 7.6 Hz, 4H, tolyl-H), 6.77 (d, 2H, phenyl-H), 5.01

(s, 2H, amine-H), 2.69 ppm (s, 6H, tolyl-CH3).

13 C NMR (151 MHz, THF-d8): δ = 153.0, 151.4, 151.0, 150.8, 150.4, 144.6, 141.5, 137.9, 135.4, 135.3, 133.7, 132.8, 132.2, 132.2, 131.3, 128.2, 128.1, 127.3, 122.5, 122.4, 115.1, 112.6, 102.4, 98.6, 91.4, 21.7 ppm.

185

IR (neat)/cm-1: ν̃ = 2919 (w), 2184 (w), 1603 (m), 1509 (m), 1339 (m), 1206 (m), 1055 (w), 996 (s), 946 (w), 793 (s), 714 (m), 569 (w).

UV-Vis (CHCl3): λmax [nm] (log ε) = 452 (5.31), 581 (4.09), 634 (4.52).

+ HRMS (MALDI-TOF) m/z calcd. for C48H33N5Zn [M] 743.2027; found 743.2029.

[5-(4'-Bromophenyl)-10,20-bis(4'-methylphenyl)-15-phenylporphyrinato]zinc(II) (214)

Porphyrin 202 (150 mg, 200 μmol), pinacol borane (566 mg,

2 mmol) and K3PO4 (850 mg, 4 mmol) were placed in an oven- dried Schlenk flask and heated under vacuum. The reaction flask was purged with argon and THF (13 mL) was added. Argon was bubbled through the solution for five minutes then

Pd(PPh3)4 (46 mg, 40 μmol) was added. The reaction was heated to 70 °C and allowed to stir for four hours. The reaction mixture was diluted with CH2Cl2 (10 mL) before removal of solvents in vacuo. The crude material was purified by column chromatography (SiO2,

CH2Cl2:n-hexane, 1:1). Recrystallisation (CHCl3/n-hexane) yielded pink-purple crystals (147 mg, 0.18 mmol, 95%). m.p. >300 °C.

Rf = 0.49 (SiO2, CH2Cl2:n-hexane, 1:1, v/v).

1 H NMR (400 MHz, CDCl3): δ = 8.98 (t, J = 4.7 Hz, 4H, Hβ), 8.92 (dd, J = 11.1, 4.7 Hz, 4H,

Hβ), 8.21 (d, J = 8.3 Hz, 2H, phenyl-H), 8.10 (d, J = 7.6 Hz, 6H, phenyl-H and tolyl-H), 7.89 (d, J = 8.3 Hz, 2H, phenyl-H), 7.79–7.72 (m, 3H, phenyl-H), 7.56 (d, J = 7.6 Hz, 4H, tolyl-

H), 2.72 ppm (s, 6H, tolyl-CH3).

13 C NMR (151 MHz, CDCl3): δ = 150.6, 150.5, 150.4, 149.9, 142.9, 142.0, 139.9, 137.3, 135.9, 134.5, 132.4, 132.3, 132.1, 131.6, 129.9, 127.7, 127.6, 126.7, 122.3, 121.6, 119.4, 21.7 ppm.

IR (neat)/cm-1: ν̃ = 1481 (w), 1337 (m), 1205 (w), 1179 (w), 1067 (m), 997 (s), 847 (w), 794 (s), 746 (m), 720 (m), 699 (m), 661 (w).

UV-Vis (CHCl3): λmax [nm] (log ε) = 401 (4.64), 422 (5.80), 550 (4.38).

+ HRMS (MALDI-TOF) m/z calcd. for C46H31N4BrZn [M] 782.1024; found 782.1027.

186

[5-(4'-(4″-Ethynylaniline)phenyl)-10,20-bis-(4'-methylphenyl)-15- phenylporphyrinato]zinc(II) (10)

Synthesised via General Procedure 12 from porphyrin 214 (117 mg, 150 μmol), 4-

ethynylaniline (88 mg, 750 μmol), PdCl2(PPh3)2 (16 mg, 22 μmol) and CuI (9 mg, 45 μmol) in

THF/NEt3 (3 mL/1 mL). The reaction was heated to 70 °C and allowed to stir for 24 hours. The crude material was purified by column chromatography (Al2O3, activated, Brockmann Grade III, CH2Cl2:n-hexane, 2:1).

Recrystallisation (CHCl3/n-hexane) yielded green crystals (72 mg, 87 μmol, 58%). m.p. >300 °C.

Rf = 0.56 (SiO2, CH2Cl2:n-hexane, 3:1, v/v).

1 H NMR (600 MHz, THF-d8): δ = 8.86 (s, 4H, Hβ), 8.83 (dd, J = 21.6, 4.5 Hz, 4H, Hβ), 8.19 (dd, J = 7.1, 1.3 Hz, 2H, phenyl-H), 8.16 (d, J = 7.9 Hz, 2H, phenyl-H), 8.07 (d, J = 7.7 Hz, 4H, tolyl-H), 7.83 (d, J = 7.9 Hz, 2H), 7.75–7.72 (m, 3H, phenyl-H), 7.56 (d, J = 7.7 Hz, 4H, tolyl-H), 7.34 (d, J = 8.2 Hz, 2H, phenyl-H), 6.62 (d, J = 8.2 Hz, 2H, phenyl-H), 4.92 (s, 2H, amine-H), 2.69 ppm (s, 6H, tolyl-CH3).

13 C NMR (151 MHz, THF-d8): δ = 151.3, 151.3, 151.1, 150.9, 150.3, 144.8, 143.9, 141.8, 137.8, 135.5, 135.4, 133.7, 132.6, 132.4, 132.2, 132.1, 130.1, 128.2, 128.0, 127.3, 124.7, 121.7, 121.6, 120.9, 114.8, 111.5, 92.9, 87.5, 21.7 ppm.

IR (neat)/cm-1: ν̃ = 2921 (w), 2852 (w), 1599 (w), 1515 (m), 1337 (m), 1286 (w), 1179 (w), 1067 (w), 994 (s), 793 (s), 719 (m), 561 (w).

UV-Vis (THF): λmax [nm] (log ε) = 427 (5.81), 559 (4.38), 600 (4.20).

+ HRMS (MALDI-TOF) m/z calcd. for C54H37N5Zn [M] 819.2340; found 819.2344.

[5-(3'-Aminophenylacetylene)-10,20-bis-(4'-methylphenyl)-15- phenylporphyrinato]zinc(II) (213b)

Synthesised via General Procedure 12 from porphyrin 202 (100 mg, 130 μmol), 3-ethynylaniline (74 μL, 660 μmol),

PdCl2(PPh3)2 (14 mg, 20 μmol) and CuI (7.5 mg, 40 μmol)

in THF/NEt3 (2 mL/0.66 mL). The reaction was heated to 70 °C and allowed to stir for four hours. The crude material

187 was purified by column chromatography (SiO2, CH2Cl2:n-hexane, 2:1). Recrystallisation

(CHCl3/n-hexane) yielded green crystals (90 mg, 0.12 mmol, 92%). m.p. = 238–240 °C.

Rf = 0.59 (SiO2, EtOAc:n-hexane, 1:1, v/v).

1 H NMR (600 MHz, THF-d8): δ = 9.76 (d, J = 4.5 Hz, 2H, Hβ), 8.91 (d, J = 4.5 Hz, 2H, Hβ),

8.75 (dd, J = 19.4, 4.5 Hz, 4H, Hβ), 8.15 (dd, 2H, phenyl-H), 8.07 (d, J = 7.6 Hz, 4H, tolyl- H), 7.74–7.70 (m, 3H, phenyl-H), 7.58 (d, J = 7.6 Hz, 4H, tolyl-H), 7.27 (d, J = 7.7 Hz, 2H, phenyl-H), 7.22 (t, J = 7.7 Hz, 1H, phenyl-H), 6.71 (d, J = 7.7 Hz, 1H, phenyl-H), 4.76 (brs,

2H, amine-H), 2.70 ppm (s, 6H, tolyl-CH3).

13 C NMR (151 MHz, THF-d8): δ = 153.2, 151.6, 150.9, 149.9, 144.5, 141.4, 138.0, 135.4, 135.3, 133.1, 133.0, 132.9, 132.4, 132.3, 131.3, 130.1, 129.3, 129.2, 128.3, 127.3, 125.9, 123.2, 122.7, 120.9, 117.8, 115.6, 100.8, 97.7, 92.7, 21.7 ppm.

IR (neat)/cm-1: ν̃ = 3016 (w), 2970 (w), 1738 (s), 1439 (w), 1365 (m), 1228 (m), 1217 (m), 995 (w), 894 (w), 793 (w), 714 (w), 700 (w).

UV-Vis (CHCl3): λmax [nm] (log ε) = 444 (5.64), 566 (4.31), 611 (4.09).

+ HRMS (MALDI-TOF) m/z calcd. for C48H33N5Zn [M] 743.2027; found 743.2024.

[5-(2'-Aminophenylacetylene)-10,20-bis-(4'-methylphenyl)-15- phenylporphyrinato]zinc(II) (213c)

Synthesised via General Procedure 12 from porphyrin 202 (100 mg, 130 μmol), 2-ethynylaniline (148 μL, 660 μmol),

PdCl2(PPh3)2 (14 mg, 20 μmol) and CuI (7.5 mg, 40 μmol) in

THF/NEt3 (2 mL:0.66 mL). The reaction was heated to 70 °C and allowed to stir for four hours. The crude material was purified by column chromatography (SiO2, CH2Cl2:n-hexane, 2:1). Recrystallisation

(CHCl3/n-hexane) yielded green crystals (75 mg, 0.1 mmol, 77%). m.p. = 234–237 °C.

Rf = 0.79 (SiO2, EtOAc:n-hexane, 1:1, v/v).

1 H NMR (600 MHz, CDCl3): δ = 8.87 (d, J = 4.3 Hz, 2H, Hβ), 8.81 (d, J = 4.3 Hz, 2H, Hβ),

8.75 (d, J = 4.3 Hz, 2H, Hβ), 8.37 (brs, 2H, phenyl-H), 8.32 (d, J = 4.3 Hz, 2H, Hβ), 8.08 (d, J = 6.8 Hz, 4H, tolyl-H), 7.82–7.79 (m, 3H, phenyl-H), 7.62 (d, J = 6.8 Hz, 4H, tolyl-H), 6.87 (d, J = 7.1 Hz, 1H, aminophenyl-H), 6.53 (t, J = 7.1 Hz, 1H, aminophenyl-H), 6.22 (t, J = 7.1

188

Hz, 1H, aminophenyl-H), 3.20 (brs, 1H, aminophenyl-H), 2.84 (s, 6H, tolyl-CH3), -0.58 ppm (brs, 2H, amine-H).

13 C NMR (151 MHz, CDCl3): δ = 152.2, 150.8, 150.0, 149.6, 143.5, 140.2, 136.9, 134.8, 134.6, 132.7, 131.8, 129.9, 129.8, 127.5, 127.4, 127.2, 126.7, 122.5, 121.8, 121.1, 115.0, 110.9, 99.2, 98.2, 88.1, 21.8 ppm.

IR (neat)/cm-1: ν̃ = 1596 (w), 1488 (m), 1439 (w), 1339 (m), 1205 (m), 1179 (m), 1064 (w), 996 (s), 946 (w), 825 (w), 792 (s), 741 (s), 708 (s), 662 (w), 609 (w), 568 (w).

UV-Vis (CHCl3): λmax [nm] (log ε) = 441 (5.48), 568 (4.28), 614 (4.35).

+ HRMS (MALDI-TOF) m/z calcd. for C48H33N5Zn [M] 743.2027; found 743.2015.

[5-Iodo-10,20-bis-(4'-methylphenyl)-15-phenylporphyrinato]zinc(II) (202)

Synthesised via General Procedure 1 from 5-Iodo-10,20-bis(4'- methylphenyl)-15-phenylporphyrin (830 mg, 1.2 mmol) and

Zn(II)(OAc)2•2H2O (658 mg, 3.0 mmol) in CHCl3 (180 mL) and MeOH

(180 mL). Recrystallisation from CHCl3/CH3OH gave purple crystals (750 mg, 1 mmol, 83%). m.p. >300 °C

Rf = 0.73 (SiO2, CH2Cl2:n-hexane, 2:1, v/v).

1 H NMR (400 MHz, CDCl3): δ = 9.77 (d, J = 4.7 Hz, 2H), 8.98 (d, J = 4.7 Hz, 2H), 8.91 (dd, J = 11.4, 4.7 Hz, 4H), 8.19–8.17 (m, 2H), 8.07 (d, J = 7.9 Hz, 4H), 7.78–7.71 (m, 3H), 7.57 (d, J = 7.9 Hz, 4H), 2.73 ppm (s, 6H).

13 C NMR (101 MHz, CDCl3): δ = 152.2, 151.5, 150.8, 150.7, 142.7, 139.6, 137.8, 137.5, 134.5, 134.4, 133.8, 132.5, 132.5, 127.7, 127.5, 126.7, 122.2, 122.1, 21.7 ppm.

IR (neat)/cm-1: ν̃ =1490 (w), 1320 (w), 1211 (w), 1074 (w), 996 (s), 795 (s), 782 (s), 755 (m), 722 (m), 707 (m), 567 (w).

UV-Vis (CHCl3): λmax [nm] (log ε) = 426 (5.79), 555 (4.41), 594 (3.96).

+ HRMS (MALDI-TOF) m/z calcd. for C40H27N4ZnI [M] 754.0572; found 754.0590.

[5,15-Bis(4'-methylphenyl)-10-phenyl-20-(4',4',5',5'-tetramethyl-1',3',2'-dioxaborolan- 2'-yl)porphyrinato]zinc(II) (208)

189

Porphyrin 202 (100 mg, 0.13 mmol) was placed in an oven-dried Schlenk tube and heated under vacuum. The flask was purged with argon followed by addition of 10 mL anhydrous 1,2- dichloroethane. Argon was bubbled through the solution for 5–

10 minutes. PdCl2(PPh3)2 (16 mg, 22 μmol) and 4,4,5,5- tetramethyl-1,3,2-dioxaborolane (0.19 mL, 1.3 mmol) were added and the reaction mixture was allowed to stir for 2 hours at 90 °C. The solvent was then removed in vacuo. The crude reaction mixture was purified by silica gel column chromatography using CH2Cl2:n-hexane (2:1) to afford the title compound as a bright pink solid (90 mg, 0.12 mmol, 91%). m.p. >300 °C.

Rf = 0.43 (SiO2, CH2Cl2:n-hexane, 2:1, v/v).

1 H NMR (600 MHz, CDCl3): δ = 9.90 (d, J = 4.6 Hz, 2H, Hβ), 9.10 (d, J = 4.6 Hz, 2H, Hβ),

8.95 (dd, J = 14.8, 4.6 Hz, 4H, Hβ), 8.21 (d, 2H, phenyl-H), 8.10 (d, J = 7.6 Hz, 4H, tolyl-H),

7.77–7.72 (m, 3H, phenyl-H), 7.56 (d, J = 7.6 Hz, 4H, tolyl-H), 2.72 (s, 6H, tolyl-CH3), 1.86 ppm (s, 12H, pinncolborane-CH3).

13 C NMR (151 MHz, CDCl3): δ = 154.5, 150.7, 150.3, 149.4, 143.0, 140.1, 137.2, 134.6, 134.5, 133.2, 132.8, 132.1, 131.7, 127.6, 127.4, 126.6, 122.6, 121.2, 85.4, 25.5, 21.7 ppm.

IR (neat)/cm-1: ν̃ = 1527 (w), 1446 (w), 1367 (w), 1303 (m), 1203 (w), 1143 (m), 1061 (m), 996 (s), 856 (w), 795 (s), 719 (m), 699 (m), 661 (w).

UV-Vis (CHCl3): λmax [nm] (log ε) = 420 (5.72), 549 (4.34).

+ HRMS (MALDI-TOF) m/z calcd. for C46H39N4O2ZnB [M] 754.2458; found 754.2459.

[5,15-Bis(4'-methylphenyl)-10-phenyl-20-((trimethylsilyl)ethynyl)porphyrinato]zinc(II) (210)

Synthesised according to known synthetic procedure[117] using

porphyrin 202 (150 mg, 0.2 mmol), PdCl2(PPh3)2 (21 mg, 30 μmol), ethynyltrimethylsilane (0.28 mL, 2.0 mmol) and CuI

(11.5 mg, 60 μmol) in anhydrous THF and NEt3 (2 mL:0.67 mL). The reaction was heated to 70 °C and allowed to stir for

3 h. The crude material was purified by column chromatography (SiO2, CH2Cl2:n-hexane,

2:1) then Recrystallisation (CHCl3/n-hexane) yielded purple crystals (97 mg, 0.13 mmol, 67%). m.p. >300 °C.

190

Rf = 0.83 (SiO2, CH2Cl2:n-hexane, 2:1, v/v).

1 H NMR (600 MHz, CDCl3): δ = 9.75 (d, J = 4.5 Hz, 2H, Hβ), 9.01 (d, J = 4.5 Hz, 2H, Hβ),

8.87 (dd, J = 21.9, 4.5 Hz, 4H, Hβ), 8.17 (d, 2H, phenyl-H), 8.08 (d, J = 7.6 Hz, 4H, tolyl-H),

7.77–7.72 (m, 3H, phenyl-H), 7.57 (d, J = 7.6 Hz, 4H, tolyl-H), 2.72 (s, 6H, tolyl-CH3), 0.61 ppm (s, 9H, tolyl-CH3).

13 C NMR (151 MHz, CDCl3): δ = 152.8, 150.9, 150.3, 150.0, 142.8, 139.6, 137.4, 134.5, 134.4, 133.2, 132.3, 132.0, 131.1, 127.7, 127.5, 126.7, 122.9, 122.2, 107.8, 101.4, 99.5, 21.7, 0.6 ppm.

IR (neat)/cm-1: ν̃ = 2149 (w), 1489 (w), 1434 (w), 1338 (w), 1248 (w), 1209 (w), 1064 (w), 999 (m), 839 (s), 795 (s), 699 (s), 559 (w).

UV-Vis (CHCl3): λmax [nm] (log ε) = 432 (5.70), 563 (4.32), 604 (4.18).

+ HRMS (MALDI-TOF) m/z calcd. for C45H36N4SiZn [M] 724.2001; found 724.2002.

[5-Ethynyl-10,20-bis(4'-methylphenyl)-15-phenylporphyrinato]zinc(II) (209)

Porphyrin 210 (176 mg, 0.24 mmol) was placed in an oven-dried Schlenk flask and heated under vacuum. The flask was purged with argon and anhydrous THF (7 mL) was added by syringe. 1 M solution of TBAF (0.36 mL, 0.36 mmol) was added dropwise to the solution

for 30 minutes at room temperature and then stirred under N2.

Progress of reaction was monitored by TLC, reaction mixture was diluted with CH2Cl2 (10 mL) solvents was evaporated in vacuo. The crude mixture was purified by column chromatography (SiO2, CH2Cl2:n-hexane, 2:1). Recrystallisation (CHCl3/n-hexane) then yielded purple-green crystals (125 mg, 0.19 mmol, 80%). m.p. >300 °C.

Rf = 0.8 (SiO2, CH2Cl2:n-hexane, 2:1, v/v).

1 H NMR (600 MHz, CDCl3): δ = 9.76 (d, J = 4.5 Hz, 2H, Hβ), 9.03 (d, J = 4.5 Hz, 2H, Hβ),

8.89 (dd, J = 18.8, 4.5 Hz, 4H, Hβ), 8.18 (d, 2H, phenyl-H), 8.08 (d, J = 7.7 Hz, 4H, tolyl-H), 7.77–7.72 (m, 3H, phenyl-H), 7.57 (d, J = 7.7 Hz, 4H, tolyl-H), 4.15 (s, 1H, ethynyl-H), 2.72 ppm (s, 6H, tolyl-CH3).

13 C NMR (151 MHz, CDCl3): δ = 152.9, 151.0, 150.3, 150.0, 142.7, 139.6, 137.5, 134.5, 134.4, 133.3, 132.4, 132.1, 131.1, 127.8, 127.5, 126.7, 123.1, 122.2, 98.3, 86.3, 83.5, 21.7 ppm.

191

IR (neat)/cm-1: ν̃ = 3265 (w), 1493 (w), 1441 (w), 1339 (w), 1209 (w), 1071 (m), 997 (s), 791 (s), 756 (m), 714 (s), 639 (m).

UV-Vis (CHCl3): λmax [nm] (log ε) = 428 (5.69), 560 (4.37), 600 (4.03).

+ HRMS (MALDI -TOF) m/z calcd. for C42H28N4Zn [M] 653.1684; found 653.1669.

[5,15-Bis(4'-methylphenyl)-10-phenyl-20-((4'- trimethylsilyl)ethynylphenyl)porphyrinato] zinc(II) (211)

An oven-dried Schlenk tube was charged with porphyrin 202 (200 mg, 260 μmol), pinacol ester 178a (794 mg,

2.7 mmol), and K3PO4 (1.13 g, 5.3 mmol). The tube was purged with argon and anhydrous THF (17 mL) was

added by syringe. Pd(PPh3)4 (60 mg, 52 μmol) was added to the above reaction mixture and heated at 70 °C and allowed to stir for 18 hours. The solvent was then removed in vacuo. The crude material was purified by column chromatography (SiO2, CH2Cl2:n-hexane, 2:1) then recrystallisation (CHCl3/n-hexane) yielded purple crystals (145 mg, 0.18 mmol, 70%). m.p. >300 °C.

Rf = 0.49 (SiO2, CH2Cl2:n-hexane, 2:1, v/v).

1 H NMR (400 MHz, CDCl3): δ = 8.98 (dd, J = 4.8, 0.9 Hz, 4H, Hβ), 8.92 (dd, J = 13.4, 4.8

Hz, 4H, Hβ), 8.22 (d, 2H, phenyl-H), 8.17 (d, J = 8.0 Hz, 2H, ethynylphenyl-H), 8.09 (d, J = 7.8 Hz, 4H, tolyl-H), 7.87 (d, J = 8.0 Hz, 2H, ethynylphenyl-H), 7.78–7.71 (m, 3H, phenyl-

H), 7.55 (d, J = 7.8 Hz, 4H, tolyl-H), 2.71 (s, 6H, tolyl-CH3), 0.38 ppm (s, 9H, tolyl-CH3).

13 C NMR (101 MHz, CDCl3): δ = 134.6, 134.5, 132.4, 132.2, 132.1, 131.2, 130.4, 127.5, 126.7, 125.9, 105.3, 95.7, 84.1, 25.0, 0.1 ppm.

IR (neat)/cm-1: ν̃ = 2958 (w), 2158 (w), 2204 (w), 1607 (w), 1354 (m), 1249 (m), 1139 (m), 999 (m), 840 (s), 796 (s), 652 (s).

UV-Vis (CHCl3): λmax [nm] (log ε) = 423 (5.30), 551 (3.96).

+ HRMS (MALDI -TOF) m/z calcd. for C51H40N4SiZn [M] 800.2314; found 800.2326.

192

[5-(4'-Ethynylphenyl)-10,20-bis(4'-methylphenyl)-15-phenylporphyrinato]zinc(II) (212)

Porphyrin 211 (128 mg, 160 μmol) was placed in an oven- dried Schlenk flask and heated under vacuum. The flask was purged with argon and anhydrous THF was added by syringe. 1 M solution of TBAF (240 μL, 240 μmol) was added dropwise to the solution at room temperature and then stirred under Argon for five minutes. The solvent was removed in vacuo. The crude material was purified by column chromatography (SiO2, CH2Cl2).

Recrystallisation (CHCl3/n-hexane) yielded purple crystals (116 mg, 0.16 mmol, 99%). m.p. >300 °C.

Rf = 0.74 (SiO2, CH2Cl2:n-hexane, 2:1, v/v).

1 H NMR (400 MHz, CDCl3): δ = 9.03–8.88 (m, 8H, Hβ), 8.21–8.19 (m, 4H, phenyl-H), 8.10 (d, J = 6.2 Hz, 4H, tolyl-H), 7.89 (d, 2H, ethynylphenyl-H), 7.76–7.75 (m, 3H, phenyl-H),

7.56 (d, J = 6.2 Hz, 4H, tolyl-H), 3.31 (s, 1H, ethynyl-H), 2.72 ppm (s, 6H, tolyl-CH3).

13 C NMR (101 MHz, CDCl3): δ = 150.6, 150.5, 150.3, 149.9, 143.7, 142.9, 139.9, 137.3, 134.5, 134.5, 132.4, 132.2, 132.1, 131.7, 130.5, 127.6, 127.5, 126.7, 121.5, 121.4, 120.0, 83.9, 78.2, 29.9, 21.7 ppm.

IR (neat)/cm-1: ν̃ = 2919 (w), 1492 (w), 1338 (w), 1206 (w), 1179 (w), 1068 (w), 996 (s), 795 (s), 720 (m), 700 (m), 652 (w), 616 (w).

UV-Vis (CHCl3): λmax [nm] (log ε) = 422 (5.74), 550 (4.37).

+ HRMS (MALDI-TOF) m/z calcd. for C48H32N4Zn [M] 728.1918; found 728.1904.

6.4.4. Synthesis of Cubane Linkers

Methyl-4-((4'-iodophenyl)carbamoyl)cubane-1-carboxylate (221)

Synthesised via General Procedure 10 from 4- (methoxycarbonyl)cubane-1-carboxylic acid 76 (50 mg, 240 μmol), 4-iodoaniline (160 mg, 730 μmol), HATU (119 mg, 310 μmol), HOAt (42.5 mg, 310 μmol) and DIPEA (17 μL) in anhydrous DMF (0.5 mL). The product was extracted with a mixture of CH2Cl2 /MeOH (×3), dried over MgSO4 and the solvent removed under reduced pressure, the crystals were washed with CH2Cl2 to remove any remaining aniline and dried under reduced pressure. The product was obtained as white crystals (120 mg, 0.29 mmol, 61%).

193 m.p. = 240–245 °C.

Rf = 0.35 (SiO2, EtOAc:n-hexane, 2:3, v/v).

1 H NMR (600 MHz, DMSO-d6): δ = 9.77 (s, 1H, amide-H), 7.63 (d, J = 8.6 Hz, 2H, phenyl- H), 7.50 (d, J = 8.6 Hz, 2H, phenyl-H), 4.29–4.22 (m, 3H, cubane-H), 4.18–4.15 (m, 3H, cubane-H), 3.64 ppm (s, 3H, ester-CH3).

13 C NMR (151 MHz, DMSO-d6): δ = 171.3, 169.5, 138.8, 137.2, 121.8, 86.8, 57.8, 54.9, 51.3, 46.5, 46.0 ppm.

IR (neat)/cm-1: ν̃ = 1720 (m, C=O), 1644 (m), 1581 (m), 1512 (m), 1390 (m), 1322 (m), 1219 (w), 1169 (w), 1088 (m), 824 (s), 792 (s), 710 (m), 598 (m).

- HRMS (APCI) m/z calcd. for C17H13INO3 [M-H] 405.994565, found 405.993716.

Methyl-4-((4'-ethynylphenyl)carbamoyl)cubane-1-carboxylate (222)

Synthesised via General Procedure 10 from 4- (methoxycarbonyl)cubane-1-carboxylic acid 76 (15 mg, 70 μmol), HATU (43.5 mg, 114 μmol), HOAt (15.5 mg, 114 μmol), DIPEA (61 μL), 4-ethynylaniline (25 mg, 210 μmol) in anhydrous DMF (0.5 mL). The product was extracted with a mixture of CH2Cl2/MeOH (×3), dried over MgSO4 and the solvent removed under reduced pressure, the crystals were washed with CH2Cl2 to remove any remaining aniline and dried under reduced pressure to afford white crystals (8 mg, 26 μmol, 37%). m.p. = 243–247 °C.

Rf = 0.67 (SiO2, EtOAc:n-hexane, 1:1, v/v).

1 H NMR (600 MHz, DMSO-d6): δ = 9.86 (s, 1H, amide-H), 7.69 (d, J = 8.5 Hz, 2H, phenyl- H), 7.41 (d, J = 8.5 Hz, 2H, phenyl-H), 4.26 (t, J = 4.6 Hz, 3H cubane-H), 4.17 (t, J = 4.7 Hz,

3H cubane-H), 4.07 (s, 1H, ethynyl-H), 3.64 ppm (s, 3H, ester-CH3).

13 C NMR (151 MHz, DMSO-d6): δ = 171.2, 169.6, 139.6, 132.2, 119.3, 116.1, 83.6, 79.8, 57.8, 54.8, 51.3, 46.6, 46.0 ppm.

IR (neat)/cm-1: ν̃ = 3252 (m, alkyne), 2993 (w), 2950 (w), 1721 (s, C=O), 1645 (s, C=O), 1585 (s), 1509 (s), 1401 (m), 1336 (s), 1288 (m), 1221 (m), 1090 (s), 931 (w), 833 (s), 721 (s), 671 (m).

+ HRMS (APCI) m/z calcd. for C19H16NO3 [M+H] 306.112470; found 306.113336.

194

N1,N4-Bis(4'-iodophenyl)cubane-1,4-dicarboxamide (223)

Synthesised via General Procedure 10 from cubane- 1,4-dicarboxylic acid 28 (46 mg, 240 μmol), 4- iodoaniline (263 mg, 1.2 mmol), HATU (119 mg, 310 μmol), HOAt (42.5 mg, 310 μmol) and DIPEA (17 μL) in anhydrous DMF (0.5 mL). The reaction mixture was diluted with H2O and CH2Cl2 causing a white precipitate to form. The product was collected by vacuum filtration, washed with CH2Cl2 to remove any leftover aniline, and dried under reduced pressure to obtain white crystals (55 mg, 93 μmol, 39%). m.p. = 257–259 °C

Rf = 0.37 (SiO2, n-hexane:EtOAc, 1:3, v/v).

1 H NMR (400 MHz, DMSO-d6): δ = 9.77 (s, 2H, amide-H), 7.64 (d, J = 8.6 Hz, 4H, phenyl- H), 7.52 (d, J = 8.6 Hz, 4H, phenyl-H), 4.24 ppm (s, 6H, cubane-H).

13 C NMR (151 MHz, DMSO-d6): δ = 169.9, 138.9, 137.2, 121.8, 86.7, 57.6, 46.3 ppm.

IR (neat)/cm-1: ν̃ = 3342 (w), 2985 (w), 1738 (w), 1645 (s, C=O), 1588 (m), 1505 (s), 1390 (s), 1324 (m), 1289 (w), 1243 (m), 1217 (w), 1061 (w), 1007 (m), 955 (w), 809 (s), 795 (m), 669 (m), 610 (w).

- HRMS (APCI) m/z calcd. for C22H15I2N2O2 [M-H] 592.922847; found 592.922494.

N1,N4-Bis(3'-iodophenyl)cubane-1,4-dicarboxamide (224)

Synthesised via General Procedure 11 from cubane-1,4- dicarboxylic acid 28 (46 mg, 240 μmol), oxalyl chloride (45

μL, 530 μmol), NEt3 (0.13 mL, 960 μmol) and 3-iodoaniline

(64 μL, 530 μmol) in DMF (0.25 mL) and CH2Cl2 (1.5 mL). Solvents were evaporated in vacuo and resulting solid was washed with CH2Cl2 to give the product as a white crystals (44 mg, 74 μmol, 31%). m.p. = 243–248 °C.

Rf = 0.37 (SiO2, EtOAc:n-hexane, 3:1, v/v).

1 H NMR (600 MHz, DMSO-d6): δ = 9.75 (s, 2H, amide-H), 8.13 (s, 2H, phenyl-H), 7.70 (d, J = 8.2 Hz, 2H, phenyl-H), 7.41 (d, J = 8.2 Hz, 2H, phenyl-H), 7.12 (t, J = 8.2 Hz, 2H, phenyl- H), 4.25 ppm (s, 6H, cubane-H).

13 C NMR (151 MHz, DMSO-d6): δ = 169.9, 140.5, 131.7, 130.7, 127.7, 118.7, 94.4, 57.5, 46.3 ppm.

195

IR (neat)/cm-1: ν̃ = 3200 (w), 2996 (w), 1738 (w), 1642 (s, C=O), 1580 (s, C=O), 1537 (m), 1474 (s), 1406 (m), 1333 (m), 1287 (w), 1242 (w), 1197 (w), 1092 (w), 996 (w), 947 (w), 901 (w), 865 (w), 845 (w), 772 (s), 721 (w), 681 (m), 657 (w), 615 (w), 572 (w).

+ HRMS (APCI) m/z calcd. for C22H17I2N2O2 [M+H] 594.937400; found 594.937107.

N1,N4-Bis(4'-ethynylphenyl)cubane-1,4-dicarboxamide (225)

Synthesised via General Procedure 10 from cubane-1,4-dicarboxylic acid 28 (30 mg, 157 μmol), HATU (156 mg, 410 μmol), HOAt (56 mg, 410 μmol), DIPEA (219 μL) and 4-ethynylaniline (55 mg, 470 μmol) in DMF (0.5 mL). The reaction mixture was diluted with CH2Cl2 causing the product to crash out of the solution.

The product was collected by vacuum filtration, washed with CH2Cl2 to remove any leftover aniline, and dried under reduced pressure to obtain white crystals (39 mg, 0.1 mmol, 67%). m.p. = 247–252 °C.

Rf = 0.74 (SiO2, EtOAc:n-hexane, 3:1, v/v).

1 H NMR (600 MHz, DMSO-d6): δ = 9.84 (s, 2H, amide-H), 7.71 (d, J = 8.5 Hz, 4H, phenyl- H), 7.42 (d, J = 8.5 Hz, 4H, phenyl-H), 4.26 (s, 6H, cubane-H), 4.08 (s, 2H, ethynyl-H) ppm.

13 C NMR (151 MHz, DMSO-d6): δ = 169.9, 139.6, 132.2, 119.3, 116.1, 83.6, 79.8, 57.5, 46.3 ppm.

IR (neat)/cm-1: ν̃ = 3283 (m, alkyne), 3218 (w), 3086 (w), 3001 (w), 1650 (s, C=O), 1590 (s, C=O), 1510 (s), 1404 (s), 1338 (s), 1290 (m), 1252 (s), 1090 (m), 944 (m), 877 (m), 840 (s), 770 (m), 667 (m), 619 (s).

+ HRMS (APCI) m/z calcd. for C26H19N2O2 [M+H] : 391.144104, found 391.143171.

N1,N4-Bis(3'-ethynylphenyl)cubane-1,4-dicarboxamide (226)

Synthesised via General Procedure 10 from cubane-1,4- dicarboxylic acid 28 (15 mg, 78 μmol), HATU (77 mg, 200 μmol), HOAt (27 mg, 200 μmol), DIPEA (108 μL) and 3- ethynylaniline (27 μL , 234 μmol) in anhydrous DMF (1 mL). The reaction mixture was diluted with H2O and CH2Cl2 causing a white precipitate to form. The product was collected by vacuum filtration, washed with CH2Cl2 to remove any leftover aniline, and dried under reduced pressure to obtain white crystals (36 mg, 92 μmol, 50%). m.p. = 297–302 °C (charred).

196

Rf = 0.69 (SiO2, EtOAc:n-hexane, 3:1, v/v).

1 H NMR (600 MHz, DMSO-d6): δ = 9.76 (s, 2H, amide-H), 7.86 (s, 2H, phenyl-H), 7.69 (d, J = 7.9 Hz, 2H, phenyl-H), 7.33 (t, J = 7.9 Hz, 2H, phenyl-H), 7.16 (d, J = 7.9 Hz, 2H, phenyl- H), 4.26 (s, 6H, cubane-H), 4.17 ppm (s, 2H, ethynyl-H).

13 C NMR (151 MHz, DMSO-d6): δ = 169.9, 139.3, 129.1, 126.4, 122.4, 121.8, 120.2, 83.4, 80.4, 57.5, 46.3, 45.9 ppm.

IR (neat)/cm-1: ν̃ = 3312 (w, alkyne), 3218 (w), 3048 (w), 2999 (w), 1649 (m, C=O), 1603 (m, C=O), 1524(m), 1480 (s), 1406 (s), 1331 (s), 1298 (m), 1225 (m), 1086 (w), 950 (m), 859 (m), 782 (s), 683 (s), 599 (s).

+ HRMS (APCI) m/z calcd. for C26H19N2O2 [M+H] 391.144104; found 391.143981.

6.4.5. Synthesis of Cubane-linked Porphyrin Monomers

[5-{4'-(4″-Carbamoyl-1″-methoxycarbonylcubane)ethynylphenyl}-10,20-bis(4'- methylphenyl)-15-phenylporphyrinato]zinc(II) (233)

Synthesised via General Procedure 10 from cubane 76 (38 mg, 80 μmol), porphyrin 213f (38 mg, 55 μmol), HATU (27 mg, 72 μmol), HOAt (10 mg, 72 μmol) and DIPEA (38 μL) in anhydrous DMF (0.5 mL). The product was extracted with a mixture of CH2Cl2/MeOH (×3), dried over MgSO4 and the solvent removed under reduced pressure. The crude material was purified by column chromatography (SiO2,

CH2Cl2:EtOAc, 100 to 98.8:0.02). Recrystallisation (CH2Cl2/MeOH) yielded the product as purple crystals (25 mg, 28 μmol, 52%). m.p. >300 °C.

Rf = 0.65 (SiO2, EtOAc:CH2Cl2, 1:9, v/v).

1 H NMR (600 MHz, DMSO-d6): δ = 10.10 (s, 1H, amide-H), 8.82 (d, J = 4.5 Hz, 2H, Hβ),

8.77 (s, 6H, Hβ), 8.18 (d, J = 6.0 Hz, 6H, phenyl-H), 8.12–8.08 (m, 4H, phenyl-H), 7.80 (d, J = 6.0 Hz, 9H, phenyl-H), 4.42 (t, J = 4.7 Hz, 3H, cubane-H), 4.28 (t, J = 4.7 Hz, 3H, cubane-

H), 3.69 ppm (s, 3H, ester-CH3).

13 C NMR (151 MHz, DMSO-d6): δ = 171.3, 169.7, 149.5, 149.2, 149.2, 142.7, 138.5, 137.6, 134.4, 134.2, 131.6, 131.5, 127.4, 126.6, 120.3, 120.2, 120.2, 117.6, 58.0, 55.0, 51.3, 46.7, 46.1 ppm.

197

IR (neat)/cm-1: ν̃ = 2986 (w), 1722 (w), 1655 (m, C=O), 1594 (w), 1507 (m), 1440 (w), 1203 (m), 1067 (w), 992 (s), 797 (s), 751 (m), 703 (m), 604 (w).

UV-Vis (CHCl3): λmax [nm] (log ε) = 422 (5.69), 549 (4.25), 590 (3.37).

+ HRMS (MALDI) m/z calcd. for C55H37N5O3Zn [M] : 879.2188, found 879.2172.

[5-{4'-(4″-Carbamoylcubane-1″-carboxylate)phenyl}-10,15,20- triphenylporphyrinato]zinc(II) (233a)

Compound 233 (30 mg, 34 μmol) was dissolved in a mixture of THF/MeOH (2 mL:1

mL). LiOH (1 mg, 41 μmol) in H2O (4 mL) was added to above solution and reaction mixture was stirred at room temperature for 18 hours.

Reaction mixture was washed with 1 M HCl and extracted with a mixture of CHCl3/THF, organic layer was dried over MgSO4 and solvents were removed in vacuo. The product was obtained as purple crystals (29 mg, 33 μmol, quantitative). m.p. >300 °C.

Rf = 0.35 (SiO2, CHCl3/MeOH, 6:1, v/v).

1 H NMR (600 MHz, 5:CDCl3/2:(CD3)2CO/0.01:Pyridine-d5): δ = 8.79 (s, 1H, amide-H), 8.63

(d, J = 4.5 Hz, 2H, Hβ), 8.59–8.58 (m, 6H, Hβ), 7.92 (d, J = 6.4 Hz, 6H, phenyl-H), 7.86 (d, J = 8.1 Hz, 2H, phenyl-H), 7.77 (d, J = 8.1 Hz, 2H, phenyl-H), 7.49–7.42 (m, 9H, phenyl-H), 4.26–4.17 (m, 3H, cubane-H), 4.10–4.06 ppm (m, 3H, cubane-H).

13 C NMR (151 MHz, 5:CDCl3/2:(CD3)2CO/0.01:Pyridine-d5): δ = 173.4, 170.1, 149.8, 149.6, 149.6, 143.1, 138.6, 137.6, 134.5, 134.2, 134.1, 131.2, 126.8, 125.9, 120.1, 120.1, 119.8, 117.3, 58.4, 56.0, 46.9, 46.4 ppm.

IR (neat)/cm-1: ν̃ = 3325 (w), 2959(w), 2923 (w), 2285 (w), 1689 (m), 1648 (m), 1595 (w), 1511 (w), 1440 (w), 1400 (w), 1338 (w), 1260 (w), 1069 (s), 1001 (s), 795 (s), 700 (s), 558 (m).

UV-Vis (CHCl3): λmax [nm] (log ε) = 421 (6.54), 549 (5.15).

+ HRMS (MALDI-TOF) m/z calcd. for C54H35N5O3Zn [M] 865.2031; found 865.2042.

198

5-{4′-(4″-Carbamoyl-1″-methoxycarbonylcubane)phenyl}-10,15,20- triphenylporphyrin (220)

Synthesised via General Procedure 10 from cubane 76 (197.4 mg, 956 μmol), porphyrin 213e (200 mg, 318 μmol), HATU (158mg, 413 μmol), HOAt (56 mg, 143 μmol) and DIPEA (222 μL) in anhydrous DMF (1 mL). The product was extracted with CH2Cl2 (×3), washed with H2O (×4), dried over MgSO4 and the solvent removed under reduced pressure. The crude material was purified by column chromatography (SiO2, CH2Cl2:n-hexane, 1:1). Product was obtained as purple crystals (197 mg, 0.24 mmol, 75%). m.p. = 229–232 °C

Rf = 0.88 (SiO2, CH2Cl2:MeOH, 20:1, v/v).

1 H NMR (400 MHz, CDCl3):δ = 8.85 (d, J = 3.9 Hz, 8H, Hβ), 8.20 (t, J = 8.1 Hz, 8H, phenyl- H), 7.97 (d, J = 8.1 Hz, 2H, phenyl-H), 7.79–7.73 (m, 9H, phenyl-H), 7.49 (s, 1H, amide-H),

4.49–4.44 (m, 3H, cubane-H), 4.43–4.38 (m, 3H, cubane-H), 3.78 (s, 3H, ester-CH3), −2.78 ppm (s, 2H, NH).

13 C NMR (101 MHz, CDCl3): δ = 172.1, 169.9, 142.3, 138.4, 137.4, 135.3, 134.6, 127.8, 126.8, 120.3, 118.1, 58.8 , 56.1, 51.9, 47.8, 47.5, 46.9, 46.8 ppm.

IR (neat)/cm-1: ν̃ = 1711 (m), 1662 (m), 1591 (m), 1520 (m), 1440 (m), 1323 (m), 1219 (m), 980 (m), 965 (m), 796 (s), 721 (m), 698 (s).

UV-Vis (CH2Cl2): λmax [nm] (log ε) = 419 (6.75), 516 (5.33), 551 (5.01), 591 (4.83), 647 (4.75).

+ HRMS (MALDI-TOF) m/z calcd. for C55H39N5O3 [M] 817.3052; found 817.3053.

5-{4′-(4″-Carbamoylcubane-1″-carboxylate)phenyl}-10,15,20-triphenylporphyrin (220a)

Compound 220 (63 mg, 77 μmol) was dissolved in a mixture of THF/MeOH (6:2 mL).

LiOH (3.7 mg, 154 μmol) in 3 mL H2O was added dropwise to the solution. The reaction mixture was stirred at room temperature for 18 hours and then diluted with H2O. Solution was acidified to pH 5 with 1 M HCl and extracted

199 with a mixture of CHCl3 and THF, organic phase was dried over MgSO4 and solvents were removed in vacuo. The product was obtained as purple crystals (55 mg, 68 μmol, 98%). m.p. >300 °C;

Rf = 0 (SiO2, CH2Cl2, v/v).

1 H NMR (400 MHz, DMSO-d6): δ = 10.10 (s, 1H, amide-H), 8.89 (d, 2H, Hβ), 8.83 (s, 6H,

Hβ), 8.23–8.21 (m, 6H, phenyl-H), 8.15 (s, 4H, phenyl-H), 7.84–7.72 (m, 9H, phenyl-H), 4.40–4.38 (m, 3H, cubane-H), 4.24–4.21 (m, 3H, cubane-H), −2.94 ppm (s, 2H, NH).

13 C NMR (151 MHz, DMSO-d6): δ = 170.1, 141.2, 139.0, 135.9, 134.6, 134.2, 128.1, 127.0, 120.0, 119.9, 119.9, 117.9, 67.0, 58.0, 46.7, 46.5, 46.1, 46.0 ppm.

IR (neat)/cm-1: ν̃ = 3344 (w), 1691 (m), 1587 (m), 1517 (m), 1400 (m), 1339 (w), 1160 (m), 1094 (m), 966 (s), 801 (s), 729 (s), 697 (s), 618 (m).

UV-Vis (CH3OH): λmax [nm] (log ε) = 415 (5.52), 513 (4.16), 548 (3.88), 591 (3.78), 647 (3.56).

+ HRMS (MALDI-TOF) m/z calcd. for C54H37N5O3 [M] 803.2896; found 803.2894.

[5-{4′-(4″-Carbamoyl-1″-methoxycarbonylcubane)ethynylphenyl}-10,20-bis(4′- methylphenyl)-15-phenylporphyrinato]zinc(II) (231)

Synthesised via General Procedure 13 from porphyrin 209 (60 mg, 90 μmol), cubane 221 (25 mg, 60 μmol),

PdCl2(PPh3)2 (6.3 mg, 9 μmol) and CuI (3.4 mg, 18 μmol) in anhydrous

THF/NEt3 (1 mL:0.33 mL). The reaction mixture was allowed to stir at 70 °C for four hours and then diluted with CH2Cl2 (10 mL) before removal of solvents in vacuo. The crude material was purified by column chromatography (SiO2, CH2Cl2:n-hexane, 2:1).

Recrystallisation (CHCl3/n-hexane) yielded purple crystals (42 mg, 45 μmol, 75%). m.p. >300 °C.

Rf = 0.81 (SiO2, CH2Cl2:EtOAc, 9:1, v/v).

1 H NMR (400 MHz, CDCl3): δ = 9.80 (d, J = 4.5 Hz, 2H, Hβ), 9.02 (d, J = 4.5 Hz, 2H, Hβ),

8.86 (dd, J = 15.8, 4.5 Hz, 4H, Hβ), 8.17 (d, J = 8.3 Hz, 2H, phenyl-H), 8.09 (d, J = 7.8 Hz, 4H, tolyl-H), 7.94 (d, J = 8.3 Hz, 2H, ethynylphenyl-H), 7.76–7.69 (m, 3H, phenyl-H), 7.57 (d, J = 7.8 Hz, 4H, tolyl-H), 7.48 (d, 2H, ethynylphenyl-H), 7.10 (brs, 1H, amide-H), 4.21–

200

4.20 (m, 3H, cubane-H), 4.14–4.13 (m, 3H, cubane-H), 3.71 (s, 3H, ester-CH3), 2.72 ppm

(s, 6H, tolyl-CH3).

13 C NMR (151 MHz, CDCl3): δ = 171.9, 152.3, 150.8, 150.1, 142.8, 139.7, 137.3, 134.5, 134.4, 133.0, 132.5, 132.2, 132.1, 130.9, 127.7, 127.5, 126.7, 122.7, 122.2, 119.5, 100.3, 96.0, 92.7, 65.4, 58.4, 51.8, 47.3, 46.7, 21.7 ppm.

IR (neat)/cm-1: ν̃ = 2970 (w), 2923 (w), 1736 (s, C=O), 1657 (m), 1582 (w), 1511 (s), 1434 (m), 1403 (m), 1339 (m), 1216 (s), 1205 (s), 1181 (m), 1088 (m), 1064 (w), 996 (s), 945 (w), 836 (m), 795 (s), 716 (m), 701 (m), 596 (w).

UV-Vis (CHCl3): λmax [nm] (log ε) = 444 (5.64), 568 (4.29), 614 (4.47).

+ HRMS (MALDI-TOF) m/z calcd. for C59H41N5O3Zn [M] 931.2501; found 931.2469.

[5-{4′-(4″-Ethynylphenyl)carbamoyl)-1″-methoxycarbonylcubane)phenyl}-10,20- bis(4′-methylphenyl)-15-phenylporphyrinato]zinc(II) (232)

Synthesised via General Procedure 13 from porphyrin 212 (25 mg, 60 μmol), cubane 221 (66

mg, 90 μmol) PdCl2(PPh3)2 (6.3 mg, 9 μmol) and CuI (3.4 mg, 18

μmol) in anhydrous THF/NEt3 (1 mL:0.33 mL). The reaction mixture was allowed to stir at

65 °C for four hours and then diluted with CH2Cl2 (10 mL) before removal of solvents in vacuo. The crude material was purified by column chromatography (SiO2,

CH2Cl2:(CH₃)2CO, 100:0 to 98.8:0.02). Recrystallisation (CHCl3/n-hexane) yielded purple crystals (34 mg, 34 μmol, 56%). m.p. >300 °C.

Rf = 0.64 (SiO2, CH2Cl2: EtOAc, 9:1, v/v).

1 H NMR (600 MHz, CDCl3): δ = 9.00–8.93 (m, 8H, Hβ), 8.22 (t, J = 7.6 Hz, 4H, phenyl-H and ethynylphenyl-H), 8.10 (d, J = 7.4 Hz, 4H, tolyl-H), 7.92 (d, J = 7.6 Hz, 2H, ethynylphenyl-H), 7.78–7.74 (m, 3H, phenyl-H), 7.62 (d, J = 8.1 Hz, 2H, aniline-H), 7.56 (d, J = 7.4 Hz, 6H, tolyl-H and aniline-H), 7.17 (s, 1H, amide-H), 4.28 (s, 6H, cubane-H), 3.74

(s, 3H, ester-CH3), 2.72 ppm (s, 6H, tolyl-CH3).

13 C NMR (151 MHz, CDCl3): δ = 171.9, 169.4, 150.6, 150.5, 150.3, 149.9, 143.2, 143.0, 139.9, 137.7, 137.3, 134.6, 134.6, 134.5, 132.7, 132.4, 132.2, 132.1, 131.7, 129.9, 127.6,

201

127.5, 126.7, 122.6, 121.5, 121.3, 120.2, 119.5, 90.2, 89.5, 58.6, 56.0, 51.9, 47.4, 46.8, 21.7 ppm.

IR (neat)/cm-1: ν̃ = 2970 (w), 1738 (s, C=O), 1486 (w), 1435 (w), 1365 (m), 1217 (s), 1204 (s), 1086 (w), 994 (m), 794 (m), 1217 (w), 1061 (w), 1007 (m), 955 (w), 809 (s), 795 (m), 717 (m), 701 (w), 568 (w).

UV-Vis (CHCl3): λmax [nm] (log ε) = 423 (5.72), 550 (4.34), 591 (3.80).

+ HRMS (MALDI-TOF) m/z calcd. for C65H45N5O3Zn [M] 1007.2814; found 1007.2787.

6.4.6. Synthesis of Cubane-linked Porphyrin Dimers

N1,N4-Bis[4′-{(10″,15″,20″-triphenylporphyrinato)zinc(II)-5″-yl}-phenyl]cubane-1,4- dicarboxamides (234)

Synthesised via General Procedure 10 from cubane 28 (15 mg, 80 μmol), porphyrin 213f (110 mg, 160 μmol), HATU (79 mg, 210 μmol), HOAt (28.5 mg, 210 μmol) and DIPEA (111 μL) in anhydrous DMF (0.5 mL). H2O was added and the product was extracted with CH2Cl2/MeOH (×3), washed with

H2O (×4), dried over MgSO4 and the solvent removed under reduced pressure. The crude material was purified by column chromatography (SiO2, CH2Cl2:(CH₃)2CO, 100:0 to 98.8:0.02). The product was obtained as purple crystals (35 mg, 27 μmol, 28%). m.p. >300 °C.

Rf = 0.71 (SiO2, CH2Cl2:(CH₃)2CO, 20:1, v/v).

1 H NMR (600 MHz, CDCl3/THF-d8): δ = 8.90 (dd, J = 14.4, 4.5 Hz, 8H, Hβ), 8.86 (s, 8H, Hβ), 8.21–8.19 (m, 16H, phenyl-H), 7.99 (s, 2H, NH), 7.98 (s, 4H, phenyl-H), 7.75–7.70 (m, 18H, phenyl-H), 4.58 ppm (s, 6H, cubane-H).

13 C NMR (151 MHz, CDCl3/THF-d8): δ = 169.9, 150.2, 150.1, 143.4, 139.6, 137.2, 135.2, 135.2, 134.6, 134.6, 131.7, 127.3, 126.5, 120.8, 120.1, 117.8, 60.5, 58.9, 47.2 ppm.

IR (neat)/cm-1: ν̃ = 1736 (w), 1660 (w), 1594 (w), 1486 (m), 1439 (w), 1398 (w), 1338 (w), 1235 (w), 1203 (w), 1174 (w), 1067 (w), 993 (s), 796 (s), 749 (m), 718 (m), 701 (s), 569 (w).

UV-Vis (CHCl3): λmax [nm] (log ε) = 422 (6.19), 549 (4.80), 589 (4.15).

202

+ HRMS (MALDI-TOF) m/z calcd. for C98H62N10O2Zn2 [M] 1538.3640; found 1538.3662.

N1,N4-Bis[4′-{(10″,15″,20″-triphenylporphyrin-5″-yl}-phenyl]cubane-1,4- dicarboxamides (235)

Synthesised via General Procedure 10 from cubane 28, porphyrin 213e (129 mg, 210 μmol), HATU (34.1 mg, 90 μmol),

HOAt (12.2 mg, 90 μmol) and DIPEA (48 μL) in anhydrous DMF (0.5 mL). H2O was added and the product was extracted with CH2Cl2 (×3), washed with H2O (×4), dried over MgSO4 and the solvent removed under reduced pressure. The crude material was purified by column chromatography (SiO2, CH2Cl2:(CH₃)2CO, 100:0 to 98.8:0.02). The product was obtained as purple crystals (55 mg, 39 μmol, 56%). m.p. = > 350 °C.

Rf = 0.52 (SiO2, CH2Cl2:EtOAc, 40:1, v/v).

1 H NMR (600 MHz, CDCl3): δ = 8.88 (d, J = 3.4 Hz, 8H, Hβ), 8.85 (s, 8H, Hβ), 8.23 (m, 16H, phenyl-H), 8.21 (s, 2H, phenyl-H), 8.02 (d, J = 7.8 Hz, 4H, phenyl-H), 7.77 (m, 18H, phenyl- H), 7.58 (s, 2H, amide-H), 4.63 (s, 6H, cubane-H), –2.76 ppm (s, 4H, NH).

13 C NMR (151 MHz, CDCl3): δ = 189.7, 142.3, 135.4, 134.7, 127.9, 126.8, 120.4, 118.1, 47.3 ppm.

IR (neat)/cm-1: ν̃ = 1674 (w), 1584 (w), 1494 (w), 1397 (w), 1317 (w), 1190 (w), 965 (m), 798 (s), 753 (m), 722 (m), 701 (s).

UV-Vis (CH2Cl2): λmax [nm] (log ε) = (log ε) 422 (6.01), 518 (4.65), 554 (4.40), 593 (4.29), 649 (4.21).

+ HRMS (MALDI-TOF) m/z calcd. for C98H67N10O2 [M] : 1415.5448, found 1415.5514.

203

N1-[4′-{(10″,20″,15″-triphenylporphyrinato)zinc(II)-5′-yl}phenyl]-N4-[4′-{(10″,20″,15″- triphenylporphyrin)-5′-yl}phenyl]cubane-1,4-dicarboxamide (236)

Synthesised via General Procedure 10 from compound 220a (30 mg, 37 μmol), porphyrin 213f (31 mg, 45 μmol), HATU (43.5 mg, 114 μmol), HOAt (15.5 mg, 114 μmol) and DIPEA (61 μL) in anhydrous DMF (0.5 mL). The reaction mixture washed with brine and the products were extracted with a mixture of

CH2Cl2/THF (×3), dried over MgSO4 and the solvent removed under reduced pressure. The crude material was purified by column chromatography (SiO2, CH2Cl2:(CH₃)2CO, 100:0 to

98.8:0.02, v/v). The product was recrystallised from CH2Cl2/CH3OH and obtained as purple crystals (28 mg, 19 μmol, 51%). m.p. >300 °C.

Rf = 0.64 (SiO2, (CH₃)2CO:CH2Cl2, 1:20, v/v).

1 H NMR (600 MHz, DMSO-d6): δ = 10.21 (s, 1H, amide-H), 10.16 (s, 1H, amide-H), 8.92 (s,

2H, Hβ), 8.86–8.84 (m, 8H, Hβ), 8.80–8.78 (m, 6H, Hβ), 8.25–8.15 (m, 20H, phenyl-H), 7.83 (dd, J = 25.5, 5.4 Hz, 18H, phenyl-H), 4.52 (s, 6H, cubane-H), –2.88 ppm (s, 2H, NH).

13 C NMR (151 MHz, DMSO-d6): δ = 170.1, 149.5, 149.3, 149.2, 142.8, 141.2, 134.6, 134.4, 134.2, 134.2, 131.5, 128.1, 127.4, 127.0, 126.6, 120.3, 120.0, 117.9, 117.7, 46.6 ppm.

IR (neat)/cm-1: ν̃ = 1658 (w), 1596 (w), 1489 (m), 1400 (w), 1320 (w), 1179 (w), 1071 (w), 1002 (m), 994 (m), 796 (s), 718 (m), 700 (s), 563 (w).

UV-Vis (CHCl3): λmax [nm] (log ε) = 422 (5.79), 451 (5.66), 550 (4.43), 671 (4.81).

+ HRMS (MALDI-TOF) m/z calcd. for C98H64N10O2Zn [M] 1476.4505; found 1476.4495.

N1,N4-Bis[3′-{(10″,20″-bis(4′-methylphenyl)-15″-phenylporphyrinato)zinc(II)-5″- yl}phenylacetylene]cubane-1,4-dicarboxamide (237)

Synthesised via General Procedure 10 from cubane 28 (10.5 mg, 55 μmol), porphyrin 213b (85

204 mg, 110 μmol), HATU (54.4 mg, 143 μmol), HOAt (19.5 mg, 143 μmol) and DIPEA (76 μL) in anhydrous DMF (0.5 mL). The reaction mixture was then washed with brine and the products were extracted with a mixture of CH2Cl2/MeOH (×3), dried over MgSO4 and the solvent removed under reduced pressure. The crude material was purified by column chromatography (SiO2, CH2Cl2:(CH₃)2CO,100:0 to 98.8:0.02). The product was recrystallised from CH2Cl2/CH3OH and obtained as purple crystals (32 mg, 19 μmol, 36%). m.p. >300 °C.

Rf = 0.33 (SiO2, CH2Cl2:(CH₃)2CO, 20:1, v/v).

1 H NMR (600 MHz, CDCl3): δ = 9.77 (d, J = 4.4 Hz, 4H, Hβ), 8.97 (d, J = 4.4 Hz, 4H, Hβ),

8.80 (dd, J = 21.3, 4.5 Hz, 8H, Hβ), 8.19 (s, 1H, amide-H), 8.16 (d, 4H, phenyl-H), 8.08 (d, J = 7.2 Hz, 8H, tolyl-H), 7.80–7.77 (m, 3H, phenyl-H), 7.74–7.69 (m, 8H, phenyl-H), 7.59 (s, 1H, amide-H), 7.55 (d, J = 7.2 Hz, 8H, tolyl-H), 7.52–7.49 (m, 3H, phenyl-H), 4.50 (s, 6H, cubane-H), 2.71 ppm (s, 12H, tolyl-CH3).

13 C NMR (151 MHz, CDCl3): δ = 169.7, 152.3, 150.7, 150.0, 149.9, 143.2, 140.1, 138.1, 137.1, 134.6, 134.5, 132.8, 131.9, 131.7, 130.5, 129.5, 127.4, 127.3, 126.5, 125.4, 122.7, 122.5, 121.9, 119.7, 98.8, 95.2, 93.9, 67.7, 58.8, 47.1, 21.7 ppm.

UV-Vis (THF): λmax [nm] (log ε) = 443 (6.13), 557 (4.65), 624 (4.94).

IR (neat)/cm-1: ν̃ = 1650 (w), 1598 (w), 1519 (w), 1483 (m), 1401 (w), 1339 (m), 1305 (w), 1207 (m), 1180 (w), 1064 (w), 996 (s), 845 (w), 792 (s), 714 (m), 701 (m), 682 (m), 569 (m).

+ HRMS (MALDI-TOF) m/z calcd. for C106H70N10O2Zn2 [M] 1642.4266; found 1642.4296.

205

N1,N4-Bis[2′-{(10″,20″-bis(4′-methylphenyl)-15″-phenylporphyrinato)zinc(II)-5″- yl}phenylacetylene]cubane-1,4-dicarboxamide (238)

Synthesised via General Procedure 10 from cubane-1,4- dicarboxylic acid 28 (8.4 mg, 44 μmol), porphyrin 213c (65 mg, 87 μmol), HATU (43.5 mg, 114 μmol), HOAt (15.5 mg, 114 μmol) and DIPEA (61 μL) in anhydrous DMF (0.5 mL). The reaction mixture was then washed with brine and the

products were extracted with a mixture of CH2Cl2/MeOH

(×3), dried over MgSO4 and the solvent removed under reduced pressure. The crude material was purified by

column chromatography (SiO2, CH2Cl2:(CH₃)2CO, 100:0 to 98.8:0.02). The product was recrystallised from

CH2Cl2/CH3OH and obtained as purple crystals (36 mg, 22 μmol, 50%). m.p. >300 °C.

Rf = 0.75 (SiO2, CH2Cl2:(CH₃)2CO, 40:1, v/v).

1 H NMR (600 MHz, THF-d8): δ = 9.61 (d, J = 4.4 Hz, 4H, Hβ), 8.91 (d, J = 4.4 Hz, 4H, Hβ),

8.75 (dd, J = 11.1, 4.4 Hz, 10H, Hβ and amide-H), 8.43 (d, J = 8.1 Hz, 2H, aniline-H), 8.15 (d, J = 6.4 Hz, 4H, phenyl-H), 8.03 (d, J = 7.6 Hz, 8H, tolyl-H), 7.95 (d, J = 6.8 Hz, 2H, aniline-H), 7.74 (dt, J = 13.9, 7.0 Hz, 6H, phenyl-H), 7.51 (d, J = 7.5 Hz, 8H, tolyl-H), 7.37 (t, J = 6.8 Hz, 2H, aniline-H), 7.22 (t, J = 6.8 Hz, 2H, aniline-H), 3.81 (s, 6H, cubane-H), 2.60 ppm (s, 12H, tolyl-CH3).

13 C NMR (151 MHz, THF-d8): δ = 170.0, 153.0, 151.8, 150.9, 146.7, 144.3, 141.1, 138.1, 135.4, 135.3, 133.7, 132.6, 132.4, 132.3, 130.8, 130.3, 128.2, 127.4, 124.0, 123.2, 120.7, 114.8, 101.1, 98.11, 91.6, 86.1, 59.7, 47.8, 21.6 ppm.

IR (neat)/cm-1: ν̃ = 3380 (w), 2920 (w), 2187 (w), 1652 (m), 1574 (m), 1510 (m), 1313 (m), 1206 (m), 1063 (m), 995 (s), 944 (m), 794 (s), 755 (s), 715 (m), 596 (m).

UV-Vis (THF): λmax [nm] (log ε) = 439 (5.92), 574 (4.53), 622 (4.72).

+ HRMS (MALDI-TOF) m/z calcd. for C106H70N10O2Zn2 [M] 1642.4266; found 1642.4304.

206

6.5. Ch. 5 – Synthesis of Covalently-linked Water-soluble Dimers for PDT Treatment.

6.5.1. Synthesis of Porphyrin Precursors

[5,15-Bis(3'-methoxycarbonylphenyl)porphyrinato]zinc(II) (284)

Synthesised via General Procedure 1 from porphyrin 281 (200 mg,

0.345 mmol) and Zn(II)(OAc)2•2H2O (379 mg, 1.73 mmol) in CHCl3

(100 mL) and MeOH (30 mL). Recrystallisation from CHCl3/CH3OH gave purple crystals (185 mg, 0.29 mmol, 84%).

m.p. >300°C.

Rf = 0.29 (SiO2, n-hexane:CH2Cl2, 1:2, v/v).

1 H NMR (400 MHz, CDCl3): δ = 10.22 (s, 2H, meso-H), 9.37 (d, J = 4.5 Hz, 4H, Hβ), 9.01

(d, J = 4.5 Hz, 4H, Hβ), 8.92–8.88 (m, 2H, phenyl-H), 8.48 (d, J = 7.9 Hz, 2H, phenyl-H), 8.40 (d, J = 7.5 Hz, 2H, phenyl-H), 7.87 (t, J = 7.7 Hz, 2H, phenyl-H), 3.97 ppm (s, 6H, ester-

CH3).

13 C NMR (101 MHz, CDCl3): δ = 167.5, 150.0, 149.6, 143.0, 138.6, 135.1, 132.4, 132.2, 128.9, 128.9, 126.9, 118.8, 106.5, 52.5 ppm.

IR (neat)/cm-1: ν̃ = 2931 (w), 1724 (m, C=O), 1433 (w), 1288 (m), 1218 (m), 1070 (m), 994 (s), 856 (m), 783 (s), 748 (s), 728 (s), 700 (s).

UV-Vis (CH2Cl2): λmax [nm] (log ε) = 409 (5.91), 536 (4.49).

+ HRMS (ESI) m/z calcd. for C36H25N4O4Zn [M] : 641.1161, found 641.1183.

207

[5,15-Bis(3'-methoxycarbonylphenyl)porphyrinato]nickel(II) (283)

Synthesised via General Procedure 2 from porphyrin 281 (200 mg,

0.35 mmol) and Ni(acac)2 (133 mg, 0.52 mmol) in toluene (20 mL).

Recrystallisation from CHCl3/CH3OH yielded purple crystals (210 mg, 0.33 mmol, 94%).

m.p. >300°C.

Rf = 0.45 (SiO2, n-hexane:CH2Cl2, 1:2, v/v).

1 H NMR (400 MHz, CDCl3): δ = 9.95 (s, 2H, meso-H), 9.20 (d, J = 4.7 Hz, 4H, Hβ), 8.86 (d,

J = 4.7 Hz, 4H, Hβ), 8.77 (s, 2H, phenyl-H), 8.46 (d, J = 7.8 Hz, 2H, phenyl-H), 8.25 (d, J =

7.2 Hz, 2H, phenyl-H), 7.82 (t, J = 7.7 Hz, 2H, phenyl-H), 3.99 ppm (s, 6H, ester-CH3).

13 C NMR (101 MHz, CDCl3): δ = 167.4, 142.9, 141.5, 137.9, 134.4, 132.5, 132.3, 129.3, 129.1, 127.2, 117.3, 105.6, 52.5 ppm.

IR (neat)/cm-1: ν̃ = 2936 (w), 1726 (s, C=O), 1584 (w), 1435 (m), 1390 (w), 1292 (s), 1254 (m), 1223 (s), 1107 (m), 1071 (m), 996 (s), 855 (s), 783 (s), 747 (s), 723 (s), 700 (s).

UV-Vis (CH2Cl2): λmax [nm] (log ε) = 400 (5.35), 516 (4.18), 548 (3.92).

+ HRMS (ESI) m/z calcd. for C38H40N4Ni [M] : 635.1224, found 635.1198.

[5,15-Dibromo-10,20-Bis(3'-methoxycarbonylphenyl)porphyrinato]nickel(II) (283a)

Synthesised by General Procedure 3 with porphyrin 283 (100 mg, 0.157 mmol), N-bromosuccinimide (30.8 mg, 0.173 mmol) and

pyridine (0.15 mL) in CHCl3 (150 mL). The crude material was purified

with a silica gel column using n-hexane:CH2Cl2 in a 1:1 ratio. The product was isolated as purple crystals (13 mg, 16 μmol, 10%).

m.p. >300°C.

Rf = 0.63 (SiO2, n-hexane:CH2Cl2, 1:3, v/v).

208

1 H NMR (400 MHz, CDCl3): δ = δ 9.45 (d, J = 5.0 Hz, 4H, Hβ), 8.64 (d, J = 5.0 Hz, 6H, phenyl-H and Hβ), 8.43 (d, J = 8.0 Hz, 2H, phenyl-H), 8.12 (d, J = 7.5 Hz, 2H, phenyl-H),

7.78 (t, J = 7.7 Hz, 2H, phenyl-H), 3.98 (s, 6H, ester-CH3).

13 C NMR (101 MHz, CDCl3): δ = 167.2, 143.2, 143.1, 140.4, 137.7, 134.2, 134.1, 133.6, 131.0, 129.5, 129.3, 127.4, 118.8, 103.2, 29.9.

IR (neat)/cm-1: ν̃ = 2954 (m), 2921 (m), 2852 (m), 1719 (s), 1459 (w), 1434 (w), 1359 (w), 1290 (m), 1274 (m), 1223 (m), 1105 (m), 1081 (s), 1040 (m), 1003 (s), 789 (s), 747 (s), 729 (s), 706 (s), 692 (s).

UV-Vis (CH2Cl2): λmax [nm] (log ε) = 422 (5.13), 538 (4.00).

+ HRMS (MALDI) m/z calcd. for C36H22N4O4NiBr2 [M] : 789.9361, found 789.9380.

5-Bromo-10,20-bis(3'-methoxycarbonylphenyl)porphyrin (282)

Synthesised by General Procedure 3 with porphyrin 281 (200 mg, 0.346 mmol), N-bromosuccinimide (67.7 mg, 0.38 mmol) and pyridine

(0.15 mL) in CHCl3 (300 mL). The crude material was purified with a

silica gel column using n-hexane:CH2Cl2 in a 1:3 ratio. The product was obtained as purple crystals (190 mg, 0.29 mmol, 83%).

m.p. >300°C.

Rf = 0.45 (SiO2, n-hexane:CH2Cl2, 1:3, v/v).

1 H NMR (400 MHz, CDCl3): δ = 10.21 (s, 1H, meso-H), 9.76 (d, J = 4.8 Hz, 2H, Hβ), 9.32

(d, J = 4.8 Hz, 2H, Hβ), 8.90 (d, J = 4.8 Hz, 4H, Hβ), 8.88 (s, 2H, phenyl-H), 8.54–8.50 (m, 2H, phenyl-H), 8.40 (d, J = 7.7 Hz, 2H, phenyl-H), 7.89 (t, J = 7.7 Hz, 2H, phenyl-H), 4.02

(s, 6H, ester-CH3), –3.01 ppm (s, 2H, NH).

13 C NMR (101 MHz, CDCl3): δ = 167.4, 141.8, 138.7, 135.1, 129.4, 129.2, 127.3, 119.1, 106.0, 104.2, 52.6 ppm.

IR (neat)/cm-1: ν̃ = 1736 (s, C=O), 1441 (w), 1293 (m), 1240 (m), 1083 (m), 956 (m), 792 (s), 750 (s), 730 (s), 693 (m).

UV-Vis (CH2Cl2): λmax [nm] (log ε) = 416 (5.79), 511 (4.43), 548 (3.65), 587 (3.85), 645 (3.65).

209

+ HRMS (ESI) m/z calcd. for C36H26BrN4O4 [M] : 657.1132, found 657.1141.

[5-Bromo-10,20-bis(3'-methoxycarbonylphenyl)porphyrinato]zinc(II) (285)

Synthesised via General Procedure 1 from porphyrin 284 (97 mg, 0.13

mmol) and Zn(II)(OAc)2•2H2O (144 mg, 0.66 mmol) in CHCl3 (50 mL),

and MeOH (15 mL) Recrystallisation from CHCl3/n-hexane gave purple crystals (90 mg, 0.12 mmol, 86%).

m.p. >300°C.

Rf = 0.29 (SiO2, CH2Cl2).

1 H NMR (400 MHz, CDCl3): δ = 10.02 (d, J = 3.2 Hz, 1H, meso-H), 9.73 (dd, J = 4.7, 1.1

Hz, 2H, Hβ), 9.23 (d, J = 4.5 Hz, 2H, Hβ), 8.88 (t, J = 4.8 Hz, 4H, Hβ), 8.83–8.80 (m, 2H, phenyl-H), 8.45 (d, J = 7.9 Hz, 2H, phenyl-H), 8.32 (d, J = 7.2 Hz, 2H, phenyl-H), 7.85 (t, J

= 7.7 Hz, 2H, phenyl-H), 3.94 ppm (s, 6H, ester-CH3).

13 C NMR (101 MHz, CDCl3): δ = 167.4, 150.4, 150.3, 149.5, 142.7, 138.5, 135.0, 133.3, 132.9, 132.8, 132.5, 129.1, 128.9, 126.9, 119.9, 106.7, 52.5 ppm.

IR (neat)/cm-1: ν̃ = 2923 (w), 2853 (w), 1718 (s), 1579 (w), 1517 (w), 1436 (w), 1384 (w), 1280 (s), 1219 (s), 1106 (m), 1061 (s), 990 (s), 848 (m), 785 (s), 749 (s), 729 (s), 696 (s).

UV-Vis (CH2Cl2): λmax [nm] (log ε) = 417 (5.54), 546 (4.27).

+ HRMS (ESI) m/z calcd. for C36H23BrN4O4Zn [M] : 718.0194, found 718.0230.

[5,15-Bis(3'-methoxycarbonylphenyl)-10- ((trimethylsilyl)ethynyl)porphyrinato]zinc(II) (286)

Synthesised via General Procedure 6 from bromoporphyrin 285

(82 mg, 110 μmol), TMSA (32 μL, 230 μmol), Pd(PPh3)Cl2 (7.7

mg, 11 μmol) and CuI (1.2 mg, 11 μmol) in THF and NEt3 (3:1, v/v). The crude material was purified by filtration through a short

silica gel plug using CH2Cl2 as eluent to remove any unreacted

starting material. Recrystallisation (CH2Cl2/n-hexane) yielded purple crystals (70 mg, 90 μmol, 86%).

210 m.p. = 176–179 °C.

Rf = 0.48 (SiO2, CH2Cl2).

1 H NMR (400 MHz, CDCl3): δ = 10.06 (s, 1H, meso-H), 9.79 (d, J = 4.6 Hz, 2H, Hβ), 9.24

(d, J = 4.5 Hz, 2H, Hβ), 8.92 (d, J = 4.6 Hz, 2H, Hβ), 8.88 (d, J = 4.5 Hz, 2H, Hβ), 8.84 (s, 2H, phenyl-H), 8.45 (d, J = 7.7 Hz, 2H, phenyl-H), 8.36 (d, J = 6.4 Hz, 2H, phenyl-H), 7.86

(t, J = 7.7 Hz, 2H, phenyl-H), 3.95 (s, 6H, ester-CH3), 0.63 ppm (s, 9H, TMS-CH3).

13 C NMR (101 MHz, CDCl3): δ = 167.4, 152.4, 150.3, 149.7, 142.7, 138.5, 135.0, 132.6, 132.3, 131.6, 128.9, 127.0, 120.1, 107.8, 107.6, 101.8, 100.6, 52.4, 0.5 ppm.

IR (neat)/cm-1: ν̃ = 2956 (w), 2134 (w), 1722 (s), 1580 (w), 1436 (m), 1281 (s), 1278 (s), 1165 (m), 1107 (m), 1061 (m), 995 (m), 840 (s) 778 (s), 751 (s), 715 (s) 647 (m).

UV-Vis (CHCl3): λmax [nm] (log ε) = 427 (5.61), 557 (4.26), 594 (3.00).

+ HRMS (MALDI) m/z calcd. for C41H32N4O4SiZn [M] : 736.1484, found 736.1501.

6.5.2. Synthesis of Porphyrin Dimer Complexes

1,4-Bis(10',15',20'-triphenylporphyrinato-5'-yl)]zinc(II) benzene (269)

Synthesised according to General Procedure 14 with porphyrin (80 mg, 110 μmol), 1,4-

diiodobenzene (14.5 mg, 44 μmol), Pd(PPh3)4

(38 mg, 33 μmol) and K2CO3 (61 mg, 440 μmol) in DMF (8 mL). The reaction was heated to 100°C for 18 hours under an argon atmosphere. The crude material was purified by filtration through a short silica gel plug using CH2Cl2 as eluent to remove any unreacted starting material. Recrystallisation (n-hexane) yielded purple crystals (20 mg, 16 μmol, 35%). m.p. >300 °C.

Rf = 0.55 (SiO2, CH2Cl2:EtOAc, 1:1, v/v).

1 H NMR (600 MHz, CDCl3:THF-d8, 10:1, v/v): δ = 9.28 (d, J = 4.4 Hz, 4H, Hβ), 8.98 (d, J =

4.4 Hz, 4H, Hβ), 8.83 (d, J = 3.0 Hz, 8H, Hβ), 8.52 (s, 4H, phenylene-H), 8.21 (d, J = 4.9 Hz, 8H, phenyl-H), 8.16 (d, J = 5.7 Hz, 4H, phenyl-H), 7.72–7.65 (m, 18H, phenyl-H).

13 C NMR (151 MHz, CDCl3:THF-d8, 10:1, v/v): δ = 150.2, 150.1, 150.1, 143.4, 142.3, 134.6, 134.5, 132.6, 131.8, 131.7, 131.6, 127.2, 126.5, 126.3, 120.8, 120.7, 120.5.

IR (neat)/cm-1: ν̃ = 1595 (w), 1522 (w), 1483 (w), 1439 (w), 1338 (m), 1203 (m), 1066 (m), 992 (s), 794 (s), 749 (s), 717 (s), 701 (s), 660 (m).

211

UV-Vis (CHCl3): λmax [nm] (log ε) = 419 (5.41), 430 (5.43), 551 (4.34), 591 (3.78).

+ HRMS (MALDI) m/z calcd. for C82H50N8Zn2 [M] : 1274.2741, found 1274.2739.

5,5'-Bis[{10'',20''-bis-(3'-methoxycarbonylphenyl)porphyrinato}zinc(II)]butadiyne (287)

Ethynylporphyrin 286 (50 mg, 68 μmol) was placed in an oven-dried Schlenk flask and heated under vacuum. The flask was purged with argon and anhydrous THF was added by syringe. TBAF (0.1mL, 100 μmol) was added dropwise to the solution at room temperature

and then stirred under N2 for one hour. The

solution was then released to air, PdCl2(PPh3)2 (3.4 mg, 4.8 μmol) and CuI (1.7 mg, 8.9 μmol) were added and the reaction was allowed to stir for a further one hour and then diluted with CH2Cl2 (10 mL) before removal of solvents in vacuo. The crude material was purified by filtration through a short silica gel column using

CH2Cl2:1%MeOH. Recrystallisation (CHCl3/n-hexane) yielded purple crystals (29 mg, 22 μmol, 32%). m.p. >300 °C.

Rf = 0.16 (SiO2, n-hexane:EtOAc, 2:1, v/v).

1 H NMR (600 MHz, THF-d8): δ = 10.27 (s, 2H, meso-H), 10.01 (d, J = 4.5 Hz, 4H, Hβ), 9.38

(d, J = 4.3 Hz, 4H, Hβ), 9.01 (d, J = 4.5 Hz, 4H, Hβ), 8.92 (s, 4H, phenyl-H), 8.88 (d, J = 4.3

Hz, 4H, Hβ), 8.51 (d, J = 7.7 Hz, 4H, phenyl-H), 8.47 (d, J = 7.7 Hz, 4H, phenyl-H), 7.94 (t,

J = 7.7 Hz, 4H, phenyl-H), 3.97 ppm (s, 12H, ester-CH3).

13 C NMR (151 MHz, THF-d8): δ = 167.5, 154.1, 151.7, 151.0, 150.5, 144.3, 139.5, 135.7, 133.5, 133.3, 132.7, 131.6, 130.1, 129.6, 127.8, 121.4, 109.5, 99.1, 88.9, 82.7, 52.5 ppm.

IR (neat)/cm-1: ν̃ = 2950 (w), 2125 (w), 1724 (s, C=O), 1498 (w), 1435 (m), 1379 (w), 1282 (s), 1219 (s), 1107 (s), 995 (m), 970 (m), 868 (w), 785 (s), 749 (s), 702 (s).

UV-Vis (CHCl3): λmax [nm] (log ε) = 422 (5.21), 439 (5.19), 447 (5.22), 480 (5.14), 564 (4.32), 614 (4.35), 663 (4.54), 692 (4.12).

+ HRMS (MALDI) m/z calcd. for C76H46N8O8Zn2 [M] : 1326.2022, found 1326.1996.

212

5,5'-Bis[{10'',20''-bis-(3'-carboxyphenyl)porphyrinato}zinc(II)]butadiyne (288)

Porphyrin 287 (7 mg, 55 μmol) and KOH (108 mg, 1.9 mmol) were dissolved in a solution of

MeOH/THF/H2O (3 mL, 1:1:1 ratio, v/v). The reaction mixture was stirred for 18 hours at 80 °C. Following this, the solvents were removed in vacuo

and the crude material was re-dissolved in H2O and neutralised with 1 M HCl. The observed precipitate was collected by vacuum filtration and washed with further aliquots of H2O and MeOH. The green solid was then dried under vacuum and then isolated in a quantitative yield. m.p. >300 °C.

Rf = 0.59 (SiO2, MeOH:CH2Cl2, 1:3, v/v).

1 H NMR (400 MHz, THF-d8) δ 10.27 (s, 2H, meso-H), 10.02 (d, J = 4.4 Hz, 4H, Hβ), 9.37 (d,

J = 4.4 Hz, 4H, Hβ), 9.03 (d, J = 4.4 Hz, 4H, Hβ), 8.94 (s, 4H, phenyl-H), 8.89 (d, J = 4.4 Hz,

4H, Hβ), 8.51 (d, J = 7.7 Hz, 4H, phenyl-H), 8.46 (d, J = 7.7 Hz, 4H, phenyl-H), 7.92 (t, J = 7.7 Hz, 4H, phenyl-H).

13C NMR (101 MHz, THF) δ 168.1, 154.1, 151.7, 151.0, 150.6, 144.2, 139.3, 136.2, 133.6, 133.2, 132.8, 131.6, 130.7, 129.9, 127.6, 121.7, 109.4, 99.0, 88.9, 82.7.

IR (neat)/cm-1: ν̃ = 2916 (s), 2849 (m), 1701 (s, C=O), 1438 (m), 1260 (m), 1208 (m), 1095 (m), 1018 (s), 791 (s), 753 (m), 674 (m).

UV-Vis (THF): λmax [nm] (log ε) = 424 (5.04), 444 (5.05), 449 (5.08), 482 (4.82), 565 (4.12), 630 (4.22), 681 (4.21).

+ HRMS (MALDI) m/z calcd. for C72H38N8O8Zn2 [M] : 1270.1396, found 1270.1398.

213

6.6. Single Crystal X-ray Data

Table 11. Details of XRD data refinement. Compound 144f 144r 177j 181e Empirical C38H40BrN4 C38H41IN4 C18H21Si C100H90N8NiZn Formula Formula weight 632.65 680.65 265.45 1527.87 Temperature/K 100(2) 100(2) 100(2) 100(2) Crystal System triclinic triclinic triclinic triclinic Space group P1 P1 P1 P1 a/Å 4.8845(6) 4.8665(3) 10.9626(4) 10.5794(9) b/Å 12.1267(12) 24.1992(13) 11.8640(4) 12.0811(10) c/Å 26.324(3) 27.2069(14) 12.8821(5) 31.213(3) α/º 100.441(3) 74.9098(18) 90.244(2) 90.007(2) β/º 91.101(3) 89.9892(19) 93.352(2) 94.017(2 γ/º 90.923(3) 89.030(2) 98.157(2) 90.367(2) Volume/ Å3 1532.8(3) 3093.1(3) 1655.53(10) 3979.4(6) Z 2 4 4 2 3 ρcalc g/cm 1.371 1.462 1.0649 1.275 μ/mm-1 1.375 1.070 1.112 0.593 F(000) 662 1400 572 1608 Crystal 0.120 × 0.300 0.040 × 0.080 0.150 × 0.150 0.300 × 0.300 × size/mm3 × 0.450 ×1.000 × 0.300 0.300 Radiation MoK\a MoK\a CuK\a MoK\a Wavelength/ Å 0.71073 0.71073 1.54178 0.71073 2θ/ º 1.574 to 2.671 to 3.44 to 66.81 0.654 to 25.250 25.631 25.500 Reflections 37088 26342 36162 85838 collected Independent 5710 11377 5788 14446 reflections Rint 0.097 0.0646 0.0696 0.0383 Rsigma 0.0764 0.0909 0.0442 0.279 Restraints 73 39 6 40 Parameters 454 746 403 891 GooF 1.062 1.056 1.058 1.053 R1 [I> 2σ (I)] 0.060 0.0633 0.0808 0.0695 wR2 [I> 2σ (I)] 0.1573 0.1257 0.2360 0.1631 R1 [all data] 0.0956 0.1023 0.0972 0.0920 wR2 [all data] 0.1693 0.1411 0.2531 0.1797 Largest peak/e 0.846 1.434 0.941 0.984 Å-3 Deepest hole/e -0.933 -1.783 -0.446 -0.937 Å-3 Flack parameter - - - -

214

Table 12. Details of XRD data refinement. Compound 221 213c

Empirical Formula C17H14INO3 C96H66N10Zn2 Formula weight 407.19 1490.32 Temperature/K 100(2) 100(2) Crystal System Orthorhombic triclinic Space group Pbca P1 a/Å 9.4896(7) 14.5004(5) b/Å 11.2295(8) 15.7893(6) c/Å 28.340(2) 18.4632(7) α/º 90 105.7570(10) β/º 90 106.3710(10) γ/º 90 93.4260(10) Volume/ Å3 3020.0(4) 3860.6(2) Z 8 2 3 ρcalc g/cm 1.791 1.282 μ/mm-1 16.77 0.677 F(000) 1600 1544.0 Crystal size/mm3 0.6 x 0.5 x 0.03 0.27 × 0.16 × 0.05 Radiation CuKα MoKα Wavelength/ Å 1.54178 0.71073 2θ/ º 6.238 to 2.71 to 52.226 136.72 Reflections collected 50812 234315 Independent reflections 2770 15291 Rint 0.0855 0.0752, Rsigma 0.0313 0.0354 Restraints 54 228 Parameters 200 1049 GooF 1.260 1.033 R1 [I> 2σ (I)] 0.0819 0.0413 wR2 [I> 2σ (I)] 0.2405 0.0938 R1 [all data] 0.0854 0.0654 wR2 [all data] 0.2422 0.1041 Largest peak/e Å-3 1.593 0.87 Deepest hole/e Å-3 -1.649 -0.57 Flack parameter - -

215

References

[1] R. Chinchilla, C. Nájera, Chem. Rev. 2007, 107, 874−922. [2] (a) K. F. Biegasiewicz, J. R. Griffiths, G. P. Savage, J. Tsanaktsidis, R. Priefer, Chem. Rev. 2015, 115, 6719−6745; (b) A. M. Dilmaç, E. Spuling, A. de Meijere, S. Bräse, Angew. Chem. Int. Ed. 2017, 56, 5684–5718. [3] V. Richards, Nat. Chem. 2016, 8, 1090–1090. [4] P. Dowd, H. Irngartinger, Chem. Rev. 1989, 89, 985−996. [5] S. Toyota, Chem. Rev. 2010, 110, 5398−5424. [6] J. F. Chiang, S. H. Bauer, J. Am. Chem. Soc. 1970, 92, 1614–1617. [7] O. Ermer, J. D. Dunitz, Chem. Commun. 1968, 10, 567–567. [8] W. F. Sanjuan-Szklarz, A. A. Hoser, M. Gutmann, A. Ø. Madsen, K. Woźniak, IUCrJ 2016, 3, 61−70. [9] T. Yildirim, P. M. Gehring, D. A. Neumann, P. E. Eaton, T. Emrick, Carbon 1998, 36, 809–815. [10] B. A. Chalmers, H. Xing, S. Houston, C. Clark, S. Ghassabian, A. Kuo, B. Cao, A. Reitsma, C. E. P. Murray, J. E. Stok, G. M. Boyle, C. J. Pierce, S. W. Littler, D. A. Winkler, P. V. Bernhardt, C. Pasay, J. J. De Voss, J. McCarthy, P. G. Parsons, G. H. Walter, M. T. Smith, H. M. Cooper, S. K. Nilsson, J. Tsanaktsidis, G. P. Savage, C. M. Williams, Angew. Chem. Int. Ed. 2016, 55, 3580–3585. [11] N. Muller, D. E. Pritchard, J. Chem. Phys. 1959, 31, 768–780. [12] H. E. Gottlieb, V. Kotlyar, A. Nudelman, J. Org. Chem. 1997, 7512–7515. [13] (a) R. Mello, M. Fiorentino, C. Fusco, R. Curci, J. Am. Chem. Soc. 1989, 111, 6749−6757; (b) A. Yoshimura, J. M. Fuchs, K. R. Middleton, A. V. Maskaev, G. T. Rohde, A. Saito, P. S. Postnikov, M. S. Yusubov, V. N. Nemykin, V. V. Zhdankin, Chem. Eur. J. 2017, 23, 16738–16742. [14] E. W. Della, P. T. Hine, H. K. Patney, J. Org. Chem. 1977, 42, 2940−2941. [15] (a) E. W. Della, E. Cotsaris, P. T. Hine, P. E. Pigou, Aust. J. Chem. 1981, 34, 913−916; (b) C. J. Rhodes, J. C. Walton, E. W. Della, J. Chem. Soc., Perkin Trans. 2 1993, 11, 2125–2125. [16] (a) Z. B. Maksić, M. Eckert-Maksić, Tetrahedron 1969, 25, 5113–5114; (b) T.-Y. Luh, L. M. Stock, J. Am. Chem. Soc. 1974, 96, 3712−3713; (c) R. E. Dixon, A. Streitwieser, P. G. Williams, P. E. Eaton, J. Am. Chem. Soc. 1991, 113, 357–358. [17] A. Streitwieser, G. R. Ziegler, J. Am. Chem. Soc. 1969, 91, 5081−5084. [18] T. W. Cole, C. J. Mayers, L. M. Stock, J. Am. Chem. Soc. 1974, 96, 4555–4557. [19] K. B. Wiberg, J. Org. Chem. 2002, 67, 1613–1617. [20] F. H. Allen, Acta Cryst. B 1984, 40, 64–72. [21] H. Irngartinger, S. Strack, F. Gredel, A. Dreuw, E. W. Della, Eur. J. Org. Chem. 1999, 1253−1257. [22] (a) K. B. Wiberg, Tetrahedron Lett. 1985, 26, 599–602; (b) K. B. Wiberg, C. M. Hadad, S. Sieber, P. V. R. Schleyer, J. Am. Chem. Soc. 1992, 114, 5820–5828.

216

[23] (a) K. Bowden, D. C. Parkin, Can. J. Chem. 1969, 47, 177–183; (b) W. Adcock, A. V. Blokhin, G. M. Elsey, N. H. Head, A. R. Krstic, M. D. Levin, J. Michl, J. Munton, E. Pinkhassik, M. Robert, J.-M. Savéant, A. Shtarev, I. Stibo, J. Org. Chem. 1999, 64, 2618–2625; (c) W. Adcock, Y. Baran, A. Filippi, M. Speranza, N. A. Trout, J. Org. Chem. 2005, 70, 1029–1034. [24] A. Baeyer, Ber. Dtsch. Chem. Ges. 1885, 18, 2269–2281. [25] (a) J. D. Kemp, K. S. Pitzer, J. Chem. Phys. 1936, 4, 749−750; (b) K. S. Pitzer, J. Chem. Phys. 1937, 5, 473–479. [26] J. D. Dunitz, V. Prelog, Angew. Chem. 1960, 72, 896–902. [27] J. F. Liebman, A. Greenberg, Chem. Rev. 1976, 76, 311−365. [28] I. Alkorta, N. Campillo, I. Rozas, J. Elguero, J. Org. Chem. 1998, 63, 7759−7763. [29] Z. Li, S. L. Anderson, J. Phys. Chem. A 2003, 107, 1162−1174. [30] R. Srinivasan, J. Am. Chem. Soc. 1968, 90, 2752−2754. [31] H. A. Bent, Chem. Rev. 1961, 61, 275−311. [32] J. W. Knowlton, F. D. Rossini, J. Res. Natl. Bur. Stand. 1949, 43, 113–115. [33] S. W. Slayden, J. F. Liebman, Chem. Rev. 2001, 101, 1541–1566. [34] E. J. Prosen, W. H. Johnson, F. D. Rossini, J. Res. Natl. Bur. Stand. 1946, 37, 51– 56. [35] K. B. Wiberg, G. Bonneville, R. Dempsey, Isr. J. Chem. 1983, 23, 85–92. [36] B. D. Kybett, S. Carroll, P. Natalis, D. W. Bonnell, J. L. Margrave, J. L. Franklin, J. Am. Chem. Soc. 1966, 88, 626−626. [37] M. V. Roux, G. Martín-Valcarcel, R. Notario, S. Kini, J. S. Chickos, J. F. Liebman, J. Chem. Eng. Data 2011, 56, 1220−1228. [38] W.-K. Wong, E. F. Westrum, J. Phys. Chem. 1970, 74, 1303–1308. [39] D. L. Rodgers, E. F. Westrum, J. T. S. Andrews, J. Chem. Thermodyn. 1973, 5, 733–739. [40] (a) D. A. Hrovat, W. T. Borden, J. Am. Chem. Soc. 1988, 110, 4710−4718; (b) W. T. Borden, Chem. Rev. 1989, 89, 1095−1109; (c) R. M. Moriarty, S. M. Tuladhar, R. Penmasta, A. K. Awasthi, J. Am. Chem. Soc. 1990, 112, 3228−3230; (d) P. E. Eaton, J. P. Zhou, J. Am. Chem. Soc. 1992, 114, 3118−3120; (e) S. Vázquez, P. Camps, Tetrahedron 2005, 61, 5147–5208. [41] P. E. Eaton, C. X. Yang, Y. Xiong, J. Am. Chem. Soc. 1990, 112, 3225−3226. [42] R. J. Doedens, P. E. Eaton, E. B. Fleischer, Eur. J. Org. Chem. 2017, 2627−2630. [43] Y. P. Auberson, C. Brocklehurst, M. Furegati, T. C. Fessard, G. Koch, A. Decker, L. La Vecchia, E. Briard, ChemMedChem. 2017, 12, 590−598. [44] (a) J. M. Lehn, G. Wipff, Chem. Phys. Lett. 1972, 15, 450−454; (b) P. E. Eaton, R. B. Appell, J. Am. Chem. Soc. 1990, 112, 4055−4057; (c) D. N. Kevill, M. J. D'Souza, R. M. Moriarty, S. M. Tuladhar, R. Penmasta, A. K. Awasthi, J. Chem. Soc., Chem. Commun. 1990, 623−624. [45] S. Jalife, S. Mondal, J. L. Cabellos, G. Martinez-Guajardo, M. A. Fernandez-Herrera, G. Merino, Chem. Commun. 2016, 52, 3403–3405.

217

[46] (a) P. D. Bartlett, L. H. Knox, J. Am. Chem. Soc. 1939, 61, 3184−3192; (b) P. D. Bartlett, S. G. Cohen, J. Am. Chem. Soc. 1940, 62, 1183−1189; (c) P. D. Bartlett, M. J. Ryan, S. G. Cohen, J. Am. Chem. Soc. 1942, 64, 2649–2653. [47] C. W. Jefford, R. M. McCreadie, P. Muller, B. Siegfried, J. Chem. Educ. 1971, 48, 708−711. [48] J. W. Wilt, T. P. Malloy, J. Org. Chem. 1972, 37, 2781–2782. [49] T. W. Cole, PhD thesis, University of Chicago (Chicago, IL), 1966. [50] B. J. Smith, J. Tsanaktsidis, J. Org. Chem. 1997, 62, 5709−5712. [51] A. Fattahi, R. E. McCarthy, M. R. Ahmad, S. R. Kass, J. Am. Chem. Soc. 2003, 125, 11746−11750. [52] J. M. Lehn, G. Wipff, Tetrahedron Lett. 1980, 21, 159−162. [53] (a) C. H. DePuy, S. Gronert, S. E. Barlow, V. M. Bierbaum, R. Damrauer, J. Am. Chem. Soc. 1989, 111, 1968−1973; (b) R. A. Seburg, R. R. Squires, Int. J. Mass Spectrom. Ion Process. 1997, 167−168, 541−557; (c) M. Hare, T. Emrick, P. E. Eaton, S. R. Kass, J. Am. Chem. Soc. 1997, 119, 237−238; (d) R. R. Sauers, Tetrahedron 1999, 55, 10013−10026; (e) D. R. Reed, S. R. Kass, K. R. Mondanaro, W. P. Dailey, J. Am. Chem. Soc. 2002, 124, 2790−2795; (f) A. Fattahi, L. Lis, Z. A. Tehrani, S. S. Marimanikkuppam, S. R. Kass, J. Org. Chem. 2012, 77, 1909−1914. [54] J. Bredt, Liebigs Ann. Chem. 1924, 437, 1–13. [55] J. C. Walton, Chem. Soc. Rev. 1992, 21, 105–112. [56] P. E. Eaton, T. W. Cole, J. Am. Chem. Soc. 1964, 86, 3157−3158. [57] M. Francl, Nat. Chem. 2017, 10, 1−2. [58] W. Weltner, J. Am. Chem. Soc. 1953, 75, 4224–4231. [59] W. v. E. Doering, W. R. Roth, R. Breuckmann, L. Figge, H.-W. Lennartz, W.-D. Fessner, H. Prinzbach, Chem. Ber. 1988, 121, 1−9. [60] N. B. Chapman, J. M. Key, K. J. Toyne, J. Org. Chem. 1970, 35, 3860−3867. [61] M. Bliese, J. Tsanaktsidis, Aust. J. Chem. 1997, 50, 189−192. [62] (a) J. R. Griffiths, J. Tsanaktsidis, G. P. Savage, R. Priefer, Thermochim. Acta 2010, 499, 15−20; (b) M. J. Falkiner, S. W. Littler, K. J. McRae, G. P. Savage, J. Tsanaktsidis, Org. Process Res. Dev. 2013, 17, 1503−1509. [63] (a) J. C. Barborak, L. Watts, R. Pettit, J. Am. Chem. Soc. 1966, 88, 1328−1329; (b) L. F. Pelosi, W. T. Miller, J. Am. Chem. Soc. 1976, 98, 4311−4312; (c) R. Gleiter, M. Karcher, Angew. Chem. Int. Ed. 1988, 27, 840–841; (d) R. Gleiter, S. Brand, Tetrahedron Lett. 1994, 35, 4969−4972; (e) A. de Meijere, S. Redlich, D. Frank, J. Magull, A. Hofmeister, H. Menzel, B. König, J. Svoboda, Angew. Chem. Int. Ed. 2007, 46, 4574–4576. [64] P. E. Eaton, J. Tsanaktsidis, J. Am. Chem. Soc. 1990, 112, 876−878. [65] (a) K. Hassenruck, J. G. Radziszewski, V. Balaji, G. S. Murthy, A. J. McKinley, D. E. David, V. M. Lynch, H. D. Martin, J. Michl, J. Am. Chem. Soc. 1990, 112, 873−874; (b) D. A. Hrovat, W. T. Borden, J. Am. Chem. Soc. 1990, 112, 875−876. [66] P. E. Eaton, Angew. Chem. Int. Ed. Engl. 1992, 31, 1421–1436.

218

[67] L. Hedberg, K. Hedberg, P. E. Eaton, N. Nodari, A. G. Robiette, J. Am. Chem. Soc. 1991, 113, 1514−1517. [68] (a) E. B. Fleischer, J. Am. Chem. Soc. 1964, 86, 3889−3890; (b) A. Almenningen, T. Jonvik, H. D. Martin, T. Urbanek, J. Mol. Struct. 1985, 128, 239−247. [69] (a) G. W. Griffin, A. P. Marchand, Chem. Rev. 1989, 89, 997−1010; (b) F. Agapito, R. C. Santos, R. M. Borges dos Santos, J. A. Martinho Simões, J. Phys. Chem. A 2015, 119, 2998−3007. [70] (a) H. Hopf, Classics in Hydrocarbon Chemistry, Wiley-VCH, Weinheim, 2000; (b) H. Hopf, J. F. Liebman, H. M. Perks, in The Chemistry of Cyclobutanes (Eds.: Z. Rappoport, J. F. Liebman), Wiley, Chichester, 2005, p. 1061. [71] (a) P. E. Eaton, R. L. Gilardi, M. X. Zhang, Adv. Mater. 2000, 12, 1143−1148; (b) P. E. Eaton, Explos. Pyrotech. 2002, 27, 1−6. [72] (a) J. Kortus, M. R. Pederson, S. L. Richardson, Chem. Phys. Lett. 2000, 322, 224−230; (b) M.-X. Zhang, P. E. Eaton, R. Gilardi, Angew. Chem. Int. Ed. 2000, 39, 401–404; (c) D. A. Hrovat, W. T. Borden, P. E. Eaton, B. Kahr, J. Am. Chem. Soc. 2001, 123, 1289−1293. [73] C. Y. Cheng, L. W. Hsin, Y. P. Lin, P. L. Tao, T. T. Jong, Bioorg. Med. Chem. 1996, 4, 73−80. [74] S. S. Kuduva, D. C. Craig, A. Nangia, G. R. Desiraju, J. Am. Chem. Soc. 1999, 121, 1936–1944. [75] Y. Chauvin, L. Saussine, Macromolecules 1996, 29, 1163–1166. [76] A. Kuc, A. Enyashin, G. Seifert, J. Phys. Chem. B 2007, 111, 8179–8186. [77] N. A. Meanwell, J. Med. Chem. 2011, 54, 2529−2591. [78] T. J. Ritchie, S. J. F. Macdonald, Drug Discov. Today 2009, 14, 1011−1020. [79] V. M. Richon, Br. J. Cancer 2006, 95, S2–S6. [80] P. J. Middlemiss, A. J. Glasky, M. P. Rathbone, E. Werstuik, S. Hindley, J. Gysbers, Neurosci. Lett. 1995, 199, 131–134. [81] M. J. O’Neil, The Merck Index: an Encyclopedia of Chemicals, Drugs, and Biologicals, Royal Society of Chemistry, Cambridge, 2013. [82] S. Horvat, K. Mlinarić-Majerski, L. Glavas-Obrovac, A. Jakas, J. Veljković, S. Marczi, G. Kragol, M. Roscić, M. Matković, A. Milostić-Srb, J. Med. Chem. 2006, 49, 3136−3142. [83] V. V. Kapoerchan, A. D. Knijnenburg, M. Niamat, E. Spalburg, A. J. Neeling, P. H. Nibbering, R. H. Mars-Groenendijk, D. Noort, J. M. Otero, A. L. Llamas-Saiz, M. J. van Raaij, G. A. van der Marel, H. S. Overkleeft, M. Overhand, Chem. Eur. J. 2010, 16, 12174–12181. [84] R. Pellicciari, G. Costantino, E. Giovagnoni, L. Mattoli, I. Brabet, J.-P. Pin, Bioorg. Med. Chem. Lett. 1998, 8, 1569−1574. [85] Q. I. Churches, R. J. Mulder, J. M. White, J. Tsanaktsidis, P. J. Duggan, Aust. J. Chem. 2012, 65, 690−690. [86] (a) X. Z. Qin, A. D. Trifunac, P. E. Eaton, Y. Xiong, J. Am. Chem. Soc. 1990, 112, 4565–4567; (b) A. Marcinek, J. Rogowski, J. Gȩbicki, G.-F. Chen, F. Williams, J. Phys. Chem. A 2000, 104, 5265–5268; (c) P. R. Schreiner, A. Wittkopp, P. A.

219

Gunchenko, A. I. Yaroshinsky, S. A. Peleshanko, A. A. Fokin, Chem. Eur. J. 2001, 7, 2739–2744; (d) H. Xing, S. D. Houston, X. Chen, S. Ghassabian, T. Fahrenhorst- Jones, A. Kuo, C.-E. P. Murray, K.-A. Conn, K. N. Jaeschke, D.-Y. Jin, C. Pasay, P. V. Bernhardt, J. M. Burns, J. Tsanaktsidis, G. P. Savage, G. M. Boyle, J. J. De Voss, J. McCarthy, G. H. Walter, T. H. J. Burne, M. T. Smith, J.-K. Tie, C. M. Williams, Chem. Eur. J. 2019, 25, 2729–2734. [87] H.-D. Martin, T. Urbanek, P. Pföhler, R. Walsh, J. Chem. Soc., Chem. Commun. 1985, 14, 964−965. [88] (a) P. E. Eaton, L. Cassar, J. Halpern, J. Am. Chem. Soc. 1970, 92, 6366−6368; (b) S. Moss, B. T. King, A. de Meijere, S. I. Kozhushkov, P. E. Eaton, J. Michl, Org. Lett. 2001, 3, 2375−2377. [89] L. Cassar, P. E. Eaton, J. Halpern, J. Am. Chem. Soc. 1970, 92, 3515−3518. [90] R. Stober, H. Musso, Angew. Chem. Int. Ed. Engl. 1977, 16, 415–416. [91] R. Stober, H. Musso, E. Ōsawa, Tetrahedron 1986, 42, 1757−1761. [92] A. J. H. Klunder, B. Zwanenburg, Tetrahedron 1975, 31, 1419−1426. [93] P. E. Eaton, Y. C. Yip, J. Am. Chem. Soc. 1991, 113, 7692−7697. [94] V. M. Carroll, D. N. Harpp, R. Priefer, Tetrahedron Lett. 2008, 49, 2677−2680. [95] A. Fasolini, Master thesis, University of Bologna (Bologna), 2016. [96] R. M. Moriarty, J. S. Khosrowshahi, T. M. Dalecki, J. Chem. Soc., Chem. Commun. 1987, 675−676. [97] G. A. Olah, C. S. Lee, G. K. S. Prakash, R. M. Moriarty, M. S. C. Rao, J. Am. Chem. Soc. 1993, 115, 10728−10732. [98] J. Wlochal, R. D. M. Davies, J. Burton, Org. Lett. 2014, 16, 4094−4097. [99] S. Plunkett, K. J. Flanagan, B. Twamley, M. O. Senge, Organometallics 2015, 34, 1408−1414. [100] F. Toriyama, J. Cornella, L. Wimmer, T.-G. Chen, D. D. Dixon, G. Creech, P. S. Baran, J. Am. Chem. Soc. 2016, 138, 11132−11135. [101] C. Li, J. Wang, L. M. Barton, S. Yu, M. Tian, D. S. Peters, M. Kumar, A. W. Yu, K. A. Johnson, A. K. Chatterjee, M. Yan, P. S. Baran, Science 2017, 356, 1–8. [102] A. Fawcett, J. Pradeilles, Y. Wang, T. Mutsuga, E. L. Myers, V. K. Aggarwal, Science 2017, 357, 283–286. [103] T. Qin, L. R. Malins, J. T. Edwards, R. R. Merchant, A. J. E. Novak, J. Z. Zhong, R. B. Mills, M. Yan, C. Yuan, M. D. Eastgate, P. S. Baran, Angew. Chem. Int. Ed. 2017, 56, 260−265. [104] R. Gianatassio, S. Kawamura, C. L. Eprile, K. Foo, J. Ge, A. C. Burns, M. R. Collins, P. S. Baran, Angew. Chem. Int. Ed. 2014, 53, 9851−9855. [105] (a) J. H. Chong, M. J. MacLachlan, Chem. Soc. Rev. 2009, 38, 3301−3315; (b) L. Zhao, Z. Li, T. Wirth, Chem. Lett. 2010, 39, 658–667. [106] (a) G. Wittig, Org. Synth. 1959, 39, 75–75; (b) A. Wiehe, M. O. Senge, H. Kurreck, Liebigs Ann. 1997, 9, 1951–1963 ; (c) A. Wiehe, M. O. Senge, A. Schäfer, M. Speck, S. Tannert, H. Kurreck, B. Röder, Tetrahedron 2001, 57, 10089–10110. [107] H. E. Zimmerman, R. M. Paufler, J. Am. Chem. Soc. 1960, 82, 1514–1515.

220

[108] (a) H. E. Zimmerman, D. Armesto, Chem. Rev. 1996, 96, 3065–3112; (b) L. M. Frutos, U. Sancho, O. Castaño, Org. Lett. 2004, 6, 1229−1231; (c) R. A. Matute, K. N. Houk, Angew. Chem. Int. Ed. 2012, 51, 13097−13100; (d) X. Li, T. Liao, L. W. Chung, J. Am. Chem. Soc. 2017, 139, 16438−16441. [109] H. E. Zimmerman, G. L. Grunewald, J. Am. Chem. Soc. 1966, 88, 183–184. [110] (a) H. E. Zimmerman, G. L. Grunewald, R. M. Paufler, M. A. Sherwin, J. Am. Chem. Soc. 1969, 91, 2330–2338; (b) H. E. Zimmerman, Acc. Chem. Res. 1971, 4, 272– 280; (c) A. Y. Meyer, R. Pasternak, Tetrahedron 1977, 33, 3233−3237; (d) E. Haselbach, L. Neuhaus, R. P. Johnson, K. N. Houk, M. N. Paddon-Row, Helv. Chim. Acta 1982, 65, 1743−1751; (e) H. S. Rzepa, Chem. Rev. 2005, 105, 3697−3715. [111] R. B. Turner, J. Am. Chem. Soc. 1964, 86, 3586–3587. [112] (a) S. Cossu, S. Battaggia, O. Lucchi, J. Org. Chem. 1997, 62, 4162−4163; (b) M. W. Wagaman, E. Bellmann, M. Cucullu, R. H. Grubbs, J. Org. Chem. 1997, 62, 9076–9082. [113] E. R. Davidson, J. Am. Chem. Soc. 1997, 119, 1449−1449. [114] P. D. Bartlett, F. D. Greene, J. Am. Chem. Soc. 1954, 76, 1088−1096. [115] N. T. Tsui, A. J. Paraskos, L. Torun, T. M. Swager, E. L. Thomas, Macromolecules 2006, 39, 3350–3358. [116] T. M. Long, T. M. Swager, Adv. Mater. 2001, 13, 601−604. [117] A. G. Hyslop, M. A. Kellett, P. M. Iovine, M. J. Therien, J. Am. Chem. Soc. 1998, 120, 12676–12677. [118] L. L. Chng, C. J. Chang, D. G. Nocera, J. Org. Chem. 2003, 68, 4075−4078. [119] T. R. Kelly, H. Silva, R. A. Silva, Nature 1999, 401, 150−152. [120] X.-Z. Zhu, C.-F. Chen, J. Am. Chem. Soc. 2005, 127, 13158–13159. [121] E. M. Perchellet, M. J. Magill, X. Huang, C. E. Brantis, D. H. Hua, J.-P. Perchellet, AntiCancer Drugs 1999, 10, 749–766. [122] J.-S. Yang, T. M. Swager, J. Am. Chem. Soc. 1998, 120, 5321–5322. [123] G. Wittig, W. Stilz, E. Knauss, Angew. Chem. 1958, 70, 166–166. [124] P. F. Li, C. F. Chen, J. Org. Chem. 2012, 77, 9250–9259. [125] D. A. Leigh, Angew. Chem. Int. Ed. 2016, 55, 14506–14508. [126] J. E. Walker, Angew. Chem. Int. Ed. 1998, 37, 2308–2319. [127] F. Chérioux, O. Galangau, F. Palmino, G. Rapenne, ChemPhysChem 2016, 17, 1742–1751. [128] H.-P. Jacquot de Rouville, R. Garbage, R. E. Cook, A. R. Pujol, A. M. Sirven, G. Rapenne, Chem. Eur. J. 2012, 18, 3023−3031. [129] G. Rapenne, G. Jimenez-Bueno, Tetrahedron 2007, 63, 7018−7026. [130] P. S. Bols, H. L. Anderson, Acc. Chem. Res. 2018, 51, 2083−2092. [131] D. Kim, A. Osuka, Acc. Chem. Res. 2004, 37, 735−745. [132] C. E. Godinez, G. Zepeda, C. J. Mortko, H. Dang, M. A. Garcia-Garibay, J. Org. Chem. 2004, 69, 1652−1662.

221

[133] S. Toyota, K. Kawahata, K. Sugahara, K. Wakamatsu, T. Iwanaga, Eur. J. Org. Chem. 2017, 5696–5707. [134] G. Wang, H. Xiao, J. He, J. Xiang, Y. Wang, X. Chen, Y. Che, H. Jiang, J. Org. Chem. 2016, 81, 3364–3371. [135] R. J. Ranson, R. M. G. Roberts, J. Organomet. Chem. 1976, 107, 295−300. [136] V. I. Dodero, M. B. Faraoni, D. C. Gerbino, L. C. Koll, A. E. Zúñiga, T. N. Mitchell, J. C. Podestá, Organometallics 2005, 24, 1992−1995. [137] Y. Kawada, H. Iwamura, J. Org. Chem. 1981, 46, 3357−3359. [138] (a) T. Nakamoto, S. Hayashi, W. Nakanishi, J. Org. Chem. 2008, 73, 9259−9269; (b) Z. Zielinski, N. Presseau, R. Amorati, L. Valgimigli, D. A. Pratt, J. Am. Chem. Soc. 2014, 136, 1570–1578. [139] R. J. Baker, M. Brym, C. Jones, M. Waugh, J. Organomet. Chem. 2004, 689, 781−790. [140] K. Okada, K. Okamoto, M. Oda, J. Chem. Soc., Chem. Commun. 1989, 21, 1636−1636. [141] C. Rimington, Int. J. Biochem. 1993, 25, 1351−1352. [142] (a) J. J. Berzelius, Lehrbuch der Chemie, Vol. 9, Arnoldische Buchhandlung, Dresden & Leipzig, 1840; (b) K. Torben, Int. J. Biochem. 1980, 11, 189−200. [143] W. Küster, Ber. Dtsch. Chem. Ges. 1912, 45, 1935−1946. [144] (a) H. Fischer, J. Klarer, Liebigs Ann. 1926, 450, 181−201; (b) K. M. Smith, Ed., Porphyrins and Metalloporphyrins, Elsevier Science, Amsterdam, 1975. [145] (a) R. Shediac, M. H. B. Gray, H. T. Uyeda, R. C. Johnson, J. T. Hupp, P. J. Angiolillo, M. J. Therien, J. Am. Chem. Soc. 2000, 122, 7017−7033; (b) H. S. Cho, D. H. Jeong, S. Cho, D. Kim, Y. Matsuzaki, K. Tanaka, A. Tsuda, A. Osuka, J. Am. Chem. Soc. 2002, 124, 14642−14654; (c) M. O. Senge, M. Fazekas, E. G. A. Notaras, W. J. Blau, M. Zawadzka, O. B. Locos, E. M. Ní Mhuircheartaigh, Adv. Mater. 2007, 19, 2737−2774. [146] (a) M. O. Senge, Chem. Commun. 2006, 3, 243−256; (b) S. A. MacGowan, M. O. Senge, Chem. Commun. 2011, 47, 11621−11623; (c) S. A. MacGowan, M. O. Senge, Inorg. Chem. 2013, 52, 1228−1237. [147] (a) S. Plunkett, M. O. Senge, ECS Trans. 2011, 35, 147−157; (b) M. O. Senge, Chem. Commun. 2011, 47, 1943–1960. [148] (a) M. Ravikanth, T. K. Chandrashekar, in Coordination Chemistry, Vol. 82, Springer, Berlin, Heidelberg, 1995, p. 105−188; (b) L. R. Milgrom, The Colours of Life, Oxford University Press, New York, 1997; (c) J. A. Shelnutt, X.-Z. Song, J.-G. Ma, S.-L. Jia, W. Jentzen, C. J. Medforth, Chem. Soc. Rev. 1998, 27, 31−42; (d) M. O. Senge, S. A. MacGowan, J. M. O'Brien, Chem. Commun. 2015, 51, 17031−17063. [149] (a) T. E. O. Screen, K. B. Lawton, G. S. Wilson, N. Dolney, R. Ispasoiu, T. Goodson III, S. J. Martin, D. D. C. Bradley, H. L. Anderson, J. Mater. Chem. 2001, 11, 312−320; (b) M. O. Senge, M. Fazekas, M. Pintea, M. Zawadzka, W. J. Blau, Eur. J. Org. Chem. 2011, 5797−5816; (c) E. G. A. Notaras, M. Fazekas, J. J. Doyle, W. J. Blau, M. O. Senge, Chem. Commun. 2007, 21, 2166−2168. [150] (a) D. Gust, T. A. Moore, A. L. Moore, Acc. Chem. Res. 2009, 42, 1890–1898; (b) A. Yella, H.-W. Lee, H. N. Tsao, C. Yi, A. K. Chandiran, M. K. Nazeeruddin, E. W.-

222

G. Diau, C.-Y. Yeh, S. M. Zakeeruddin, M. Grätzel, Science 2011, 334, 629–634; (c) B. E. Hardin, H. J. Snaith, M. D. McGehee, Nat. Photonics 2012, 6, 162−169; (d) L.-L. Li, E. W.-G. Diau, Chem. Soc. Rev. 2013, 42, 291−304; (e) S. Mathew, A. Yella, P. Gao, R. Humphry-Baker, B. F. E. Curchod, N. Ashari-Astani, I. Tavernelli, U. Rothlisberger, M. K. Nazeeruddin, M. Grätzel, Nat. Chem. 2014, 6, 242−247. [151] (a) S. A. Krasnikov, N. N. Sergeeva, Y. N. Sergeeva, M. O. Senge, A. A. Cafolla, Phys. Chem. Chem. Phys. 2010, 12, 6666−6671; (b) S. A. Krasnikov, C. M. Doyle, N. N. Sergeeva, A. B. Preobrajenski, N. A. Vinogradov, Y. N. Sergeeva, A. A. Zakharov, M. O. Senge, A. A. Cafolla, Nano Res. 2011, 4, 376−384; (c) J. P. Beggan, S. A. Krasnikov, N. N. Sergeeva, M. O. Senge, A. A. Cafolla, Nanotechnology 2012, 23, 235606−235616. [152] (a) O. Kazuya, K. Yoshiaki, Anti-Cancer Agents Med. Chem. 2008, 8, 269–279; (b) M. Ethirajan, Y. Chen, P. Joshi, R. K. Pandey, Chem. Soc. Rev. 2011, 40, 340–362; (c) S. Yano, S. Hirohara, M. Obata, Y. Hagiya, S.-i. Ogura, A. Ikeda, H. Kataoka, M. Tanaka, T. Joh, J. Photochem. Photobiol. C 2011, 12, 46–67; (d) M. O. Senge, M. W. Radomski, Photodiagn. Photodyn. 2013, 10, 1–16. [153] S. Okada, H. Segawa, J. Am. Chem. Soc. 2003, 125, 2792−2796. [154] (a) O. S. Finikova, A. V. Cheprakov, P. J. Carroll, S. Dalosto, S. A. Vinogradov, Inorg. Chem. 2002, 41, 6944−6946; (b) A. B. Rudine, B. D. DelFatti, C. C. Wamser, J. Org. Chem. 2013, 78, 6040−6049. [155] M. Roucan, K. J. Flanagan, J. O'Brien, M. O. Senge, Eur. J. Org. Chem. 2018, 6432−6446. [156] J. S. Lindsey, Acc. Chem. Res. 2010, 43, 300−311. [157] (a) P. Rothemund, J. Am. Chem. Soc. 1935, 57, 2010−2011; (b) P. Rothemund, J. Am. Chem. Soc. 1936, 58, 625−627. [158] (a) J. S. Lindsey, H. C. Hsu, I. C. Schreiman, Tetrahedron Lett. 1986, 27, 4969−4970; (b) J. S. Lindsey, I. C. Schreiman, H. C. Hsu, P. C. Kearney, A. M. Marguerettaz, J. Org. Chem. 1987, 52, 827−836; (c) J. S. Lindsey, R. W. Wagner, J. Org. Chem. 1989, 54, 828−836. [159] (a) J. S. Manka, D. S. Lawrence, Tetrahedron Lett. 1989, 30, 6989−6992; (b) C. H. Lee, J. S. Lindsey, Tetrahedron 1994, 50, 11427−11440; (c) C. Brückner, J. J. Posakony, C. K. Johnson, R. W. Boyle, B. R. James, D. Dolphin, J. Porphyr. Phthalocyanines 1998, 2, 455−465; (d) B. J. Littler, M. A. Miller, C.-H. Hung, R. W. Wagner, D. F. O'Shea, P. D. Boyle, J. S. Lindsey, J. Org. Chem. 1999, 64, 1391−1396. [160] (a) S. Taniguchi, H. Hasegawa, M. Nishimura, M. Takahashi, Synlett 1999, 73−74; (b) S. Taniguchi, H. Hasegawa, S. Yanagiya, Y. Tabeta, Y. Nakano, M. Takahashi, Tetrahedron 2001, 57, 2103−2108; (c) C. Ryppa, M. O. Senge, S. S. Hatscher, E. Kleinpeter, P. Wacker, U. Schilde, A. Wiehe, Chem. Eur. J. 2005, 11, 3427−3442. [161] G. P. Arsenault, E. Bullock, S. F. Macdonald, J. Am. Chem. Soc. 1960, 82, 4384−4389. [162] (a) A. A. Ryan, M. O. Senge, Photochem. Photobiol. Sci. 2015, 14, 638–660; (b) G. M. Locke, M. O. Senge, ECS Trans. 2016, 72, 1–11. [163] (a) M. R. Wasielewski, Chem. Rev. 1992, 92, 435–461; (b) H. Kurreck, M. Huber, Angew. Chem. Int. Ed. Engl. 1995, 34, 849−866; (c) G. McDermott, S. M. Prince, A. A. Freer, A. M. Hawthornthwaite-Lawless, M. Z. Papiz, R. J. Cogdell, N. W. Isaacs, Nature 1995, 374, 517−521; (d) D. Gust, T. A. Moore, A. L. Moore, Acc. Chem. Res.

223

2001, 34, 40–48; (e) M. O. Senge, A. A. Ryan, K. A. Letchford, S. A. MacGowan, T. Mielke, Symmetry 2014, 6, 781–843; (f) Q. Yan, Z. Luo, K. Cai, Y. Ma, D. Zhao, Chem. Soc. Rev. 2014, 43, 4199–4221. [164] A. Rudolph, M. Lautens, Angew. Chem. Int. Ed. 2009, 48, 2656−2670. [165] P. E. Eaton, M. Maggini, J. Am. Chem. Soc. 1988, 110, 7230−7232. [166] P. E. Eaton, H. Higuchi, R. Millikan, Tetrahedron Lett. 1987, 28, 1055−1058. [167] P. E. Eaton, D. Stossel, J. Org. Chem. 1991, 56, 5138−5142. [168] S. Plunkett, PhD thesis, Trinity College Dublin (Dublin), 2015. [169] (a) H. Imahori, D. M. Guldi, K. Tamaki, Y. Yoshida, C. Luo, Y. Sakata, S. Fukuzumi, J. Am. Chem. Soc. 2001, 123, 6617−6628; (b) D. M. Guldi, Chem. Soc. Rev. 2002, 31, 22−36. [170] M. C. O’Sullivan, J. K. Sprafke, D. V. Kondratuk, C. Rinfray, T. D. W. Claridge, A. Saywell, M. O. Blunt, J. N. O’Shea, P. H. Beton, M. Malfois, H. L. Anderson, Nature 2011, 469, 72−75. [171] (a) F. C. Grozema, C. Houarner-Rassin, P. Prins, L. D. A. Siebbeles, H. L. Anderson, J. Am. Chem. Soc. 2007, 129, 13370−13371; (b) M. U. Winters, E. Dahlstedt, H. E. Blades, C. J. Wilson, M. J. Frampton, H. L. Anderson, B. Albinsson, J. Am. Chem. Soc. 2007, 129, 4291–4297. [172] (a) J.-H. Chou, M. E. Kosal, H. S. Nalwa, N. A. Rakow, K. S. Suslick, The Porphyrin Handbook, Vol. 8, Academic Press, 2000; (b) I. A. Maretina, Russ. J. Gen. Chem. 2009, 79, 1544−1581; (c) A. A. Ryan, A. Gehrold, R. Perusitti, M. Pintea, M. Fazekas, O. B. Locos, F. Blaikie, M. O. Senge, Eur. J. Org. Chem. 2011, 5817– 5844. [173] (a) L. J. Esdaile, M. O. Senge, D. P. Arnold, Chem. Commun. 2006, 40, 4192–4194; (b) A. A. Ryan, S. Plunkett, A. Casey, T. McCabe, M. O. Senge, Chem. Commun. 2014, 50, 353−355. [174] A. K. Burrell, D. L. Officer, P. G. Plieger, D. C. W. Reid, Chem. Rev. 2001, 101, 2751−2796. [175] (a) R. C. Fort, P. v. R. Schleyer, Chem. Rev. 1964, 64, 277–300; (b) H. Schwertfeger, A. A. Fokin, P. R. Schreiner, Angew. Chem. Int. Ed. 2008, 47, 1022– 1036. [176] R. H. Goldsmith, J. Vura-Weis, A. M. Scott, S. Borkar, A. Sen, M. A. Ratner, M. R. Wasielewski, J. Am. Chem. Soc. 2008, 130, 7659−7669. [177] (a) J. R. Miller, L. T. Calcaterra, G. L. Closs, J. Am. Chem. Soc. 1984, 106, 3047−3049; (b) M. R. Wasielewski, M. P. Niemczyk, W. A. Svec, E. B. Pewitt, J. Am. Chem. Soc. 1985, 107, 5562–5563. [178] H. E. Zimmerman, T. D. Goldman, T. K. Hirzel, S. P. Schmidt, J. Org. Chem. 1980, 45, 3933–3951. [179] H. E. Zimmerman, Y. A. Lapin, E. E. Nesterov, G. A. Sereda, J. Org. Chem. 2000, 65, 7740−7746. [180] G. M. Locke, S. S. R. Bernhard, M. O. Senge, Chem. Eur. J. 2019, 25, 4590−4647. [181] A. Beyeler, P. Belser, Coord. Chem. Rev. 2002, 230, 29−39.

224

[182] (a) D. N. Beratan, J. Am. Chem. Soc. 1986, 108, 4321−4326; (b) J. R. Bolton, T.-F. Ho, S. Liauw, A. Siemiarczuk, C. S. K. Wan, A. C. Weedon, J. Chem. Soc., Chem. Commun. 1985, 9, 559−560; (c) A. D. Joran, B. A. Leland, G. G. Geller, J. J. Hopfield, P. B. Dervan, J. Am. Chem. Soc. 1984, 106, 6090−6092. [183] (a) T. A. Reekie, M. Sekita, L. M. Urner, S. Bauroth, L. Ruhlmann, J.-P. Gisselbrecht, C. Boudon, N. Trapp, T. Clark, D. M. Guldi, F. Diederich, Chem. Eur. J. 2017, 23, 6357–6369; (b) L. M. Urner, M. Sekita, N. Trapp, W. B. Schweizer, M. Wörle, J.-P. Gisselbrecht, C. Boudon, D. M. Guldi, F. Diederich, Eur. J. Org. Chem. 2014, 91– 108. [184] (a) A. Osuka, R.-P. Zhang, K. Maruyama, N. Mataga, Y. Tanaka, T. Okada, Chem. Phys. Lett. 1993, 215, 179−184; (b) A. Osuka, R.-P. Zhang, K. Maruyama, T. Ohno, K. Nozaki, Bull. Chem. Soc. Jpn. 1993, 66, 3773−3782. [185] C. F. Portela, J. Brunckova, J. L. Richards, B. Schöllhorn, Y. Iamamoto, D. Magde, T. G. Traylor, C. L. Perrin, J. Phys. Chem. A 1999, 103, 10540−10552. [186] J. Andréasson, G. Kodis, T. Ljungdahl, A. L. Moore, T. A. Moore, D. Gust, J. Mårtensson, B. Albinsson, J. Phys. Chem. A 2003, 107, 8825−8833. [187] K. Pettersson, K. Kilså, J. Mårtensson, B. Albinsson, J. Am. Chem. Soc. 2004, 126, 6710−6719. [188] K. Pettersson, A. Kyrychenko, E. Rönnow, T. Ljungdahl, J. Mårtensson, B. Albinsson, J. Phys. Chem. A 2006, 110, 310−318. [189] A. Osuka, G. Noya, S. Taniguchi, T. Okada, Y. Nishimura, I. Yamazaki, N. Mataga, Chem. Eur. J. 2000, 6, 33–46. [190] E. M. Finnigan, R. Rein, N. Solladié, K. Dahms, D. C. G. Götz, G. Bringmann, M. O. Senge, Tetrahedron 2011, 67, 1126–1134. [191] C. Zhang, C.-F. Chen, J. Org. Chem. 2006, 71, 6626–6629. [192] (a) S. Patai, The Chemistry of Diazonium and Diazo Groups, Wiley, New York, 1978; (b) C. Galli, Chem. Rev. 1988, 88, 765−792. [193] N. Miyaura, K. Yamada, A. Suzuki, Tetrahedron Lett. 1979, 20, 3437–3440. [194] J. K. Laha, S. Dhanalekshmi, M. Taniguchi, A. Ambroise, J. S. Lindsey, Org. Process Res. Dev. 2003, 7, 799−812. [195] R. F. Heck, J. P. Nolley, J. Org. Chem. 1972, 37, 2320–2322. [196] K. Tamao, K. Sumitani, M. Kumada, J. Am. Chem. Soc. 1972, 94, 4374–4376. [197] K. Sonogashira, Y. Tohda, N. Hagihara, Tetrahedron Lett. 1975, 16, 4467–4470. [198] A. O. King, N. Okukado, E.-i. Negishi, J. Chem. Soc., Chem. Commun. 1977, 19, 683−684. [199] D. Milstein, J. K. Stille, J. Am. Chem. Soc. 1978, 100, 3636–3638. [200] Y. Hatanaka, T. Hiyama, J. Org. Chem. 1988, 53, 918–920. [201] (a) A. S. Guram, S. L. Buchwald, J. Am. Chem. Soc. 1994, 116, 7901–7902; (b) F. Paul, J. Patt, J. F. Hartwig, J. Am. Chem. Soc. 1994, 116, 5969–5970. [202] C. C. C. Johansson Seechurn, M. O. Kitching, T. J. Colacot, V. Snieckus, Angew. Chem. Int. Ed. 2012, 51, 5062–5085. [203] N. Miyaura, A. Suzuki, Chem. Rev. 1995, 95, 2457–2483.

225

[204] M. Karak, L. C. A. Barbosa, G. C. Hargaden, RSC Adv. 2014, 4, 53442−53466. [205] T. Ohe, N. Miyaura, A. Suzuki, J. Org. Chem. 1993, 58, 2201–2208. [206] P. Bertus, F. Fecourt, C. Bauder, P. Pale, New J. Chem. 2004, 28, 12–14. [207] C. Glaser, Ber. Dtsch. Chem. Ges. 1869, 2, 422–424. [208] A. S. Hay, J. Org. Chem. 1962, 27, 3320–3321. [209] T. Fukuyama, M. Shinmen, S. Nishitani, M. Sato, I. Ryu, Org. Lett. 2002, 4, 1691– 1694. [210] (a) L. Sikk, J. Tammiku-Taul, P. Burk, Organometallics 2011, 30, 5656–5664; (b) M. García-Melchor, M. C. Pacheco, C. Nájera, A. Lledós, G. Ujaque, ACS Catal. 2012, 2, 135–144. [211] (a) H. A. Dieck, F. R. Heck, J. Organomet. Chem. 1975, 93, 259–263; (b) C. Amatore, S. Bensalem, S. Ghalem, A. Jutand, J. Organomet. Chem. 2004, 689, 4642–4646. [212] A. Soheili, J. Albaneze-Walker, J. A. Murry, P. G. Dormer, D. L. Hughes, Org. Lett. 2003, 5, 4191–4194. [213] T. Ljungdahl, T. Bennur, A. Dallas, H. Emtenäs, J. Mårtensson, Organometallics 2008, 27, 2490–2498. [214] (a) C. Hansch, A. Leo, R. W. Taft, Chem. Rev. 1991, 91, 165–195; (b) S. Ehrenson, R. T. C. Brownlee, R. W. Taft, Progress in Physical Organic Chemistry, Vol. 10, John Wiley & Sons Inc., Chichester 2007. [215] T. Ljungdahl, K. Pettersson, B. Albinsson, J. Martensson, J. Org. Chem. 2006, 71, 1677–1687. [216] K. H. Shaughnessy, P. Kim, J. F. Hartwig, J. Am. Chem. Soc. 1999, 121, 2123– 2132. [217] (a) G. L. Eichhorn, Inorganic Biochemistry, Elsevier, Amsterdam, 1973; (b) D. Dolphin, Ed., The Porphyrins, Academic Press, New York, 1979; (c) S. J. Lippard, J. M. Berg, Principles of Bioinorganic Chemistry, University Science Books, Mill Valley, California, 1994; (d) A. R. Battersby, Nat. Prod. Rep. 2000, 17, 507–526. [218] S. Horn, K. Dahms, M. O. Senge, J. Porphyr. Phthalocyanines 2008, 12, 1053−1077. [219] (a) S. Prathapan, T. E. Johnson, J. S. Lindsey, J. Am. Chem. Soc. 1993, 115, 7519– 7520; (b) R. Paolesse, R. K. Pandey, T. P. Forsyth, l. Jaquinod, K. R. Gerzevske, D. J. Nurco, M. O. Senge, S. Licoccia, T. Boschi, K. M. Smith, J. Am. Chem. Soc. 1996, 118, 3869–2882; (c) M. O. Senge, W. W. Kalisch, K. Ruhlandt-Senge, Chem. Commun. 1996 2149–2150; (d) P. S. Bols, H. L. Anderson, Acc. Chem. Res. 2018, 51, 2083–2092. [220] (a) M. Mariko, O. Makiko, N. Misaki, K. Yoshihiro, Y. Yoshinori, Y. Tetsu, N. Hiroshi, Chem. Lett. 2008, 37, 404−405; (b) A. Choudhary, T. Raines Ronald, ChemBioChem. 2011, 12, 1801−1807. [221] P. E. Eaton, T. W. Cole, J. Am. Chem. Soc. 1964, 86, 962−964. [222] K. C. Nicolaou, D. Vourloumis, S. Totokotsopoulos, A. Papakyriakou, H. Karsunky, H. Fernando, J. Gavrilyuk, D. Webb, A. F. Stepan, ChemMedChem 2016, 11, 31−37.

226

[223] S. S. R. Bernhard, G. M. Locke, S. Plunkett, A. Meindl, K. J. Flanagan, M. O. Senge, Chem. Eur. J. 2018, 24, 1026–1030. [224] P. E. Eaton, E. Galoppini, R. Gilardi, J. Am. Chem. Soc. 1994, 116, 7588−7596. [225] J. Cornella, J. T. Edwards, T. Qin, S. Kawamura, J. Wang, C.-M. Pan, R. Gianatassio, M. Schmidt, M. D. Eastgate, P. S. Baran, J. Am. Chem. Soc. 2016, 138, 2174−2177. [226] G. D. Jones, J. L. Martin, C. McFarland, O. R. Allen, R. E. Hall, A. D. Haley, R. J. Brandon, T. Konovalova, P. J. Desrochers, P. Pulay, D. A. Vicic, J. Am. Chem. Soc. 2006, 128, 13175−13183. [227] D. A. Culkin, J. F. Hartwig, Organometallics 2004, 23, 3398−3416. [228] J. J. Low, W. A. Goddard, J. Am. Chem. Soc. 1984, 106, 8321−8322. [229] (a) A. Krasovskiy, P. Knochel, Angew. Chem. Int. Ed. 2004, 43, 3333−3336; (b) K. Fujimoto, H. Yorimitsu, A. Osuka, Eur. J. Org. Chem. 2014, 4327−4334. [230] K. Sonogashira, J. Organomet. Chem. 2002, 653, 46−49. [231] (a) C. Glaser, Liebigs Ann. 1870, 154, 137−171; (b) K. S. Sindhu, G. Anilkumar, RSC Adv. 2014, 4, 27867−27887. [232] R. W. Wagner, T. E. Johnson, F. Li, J. S. Lindsey, J. Org. Chem. 1995, 60, 5266−5273. [233] G. M. Kosolapoff, J. Am. Chem. Soc. 1953, 75, 3596−3597. [234] G. P. Arsenault, E. Bullock, S. F. MacDonald, J. Am. Chem. Soc. 1960, 82, 4384−4389. [235] (a) J. S. Manka, D. S. Lawrence, Tetrahedron Lett. 1989, 30, 6989−6992; (b) C.-H. Lee, J. S. Lindsey, Tetrahedron 1994, 50, 11427−11440; (c) C. Brückner, J. J. Posakony, C. K. Johnson, R. W. Boyle, B. R. James, D. Dolphin, J. Porphyr. Phthalocyanines 1998, 2, 455−465; (d) B. J. Littler, M. A. Miller, C.-H. Hung, R. W. Wagner, D. F. O'Shea, P. D. Boyle, J. S. Lindsey, J. Org. Chem. 1999, 64, 1391−1396. [236] (a) M. O. Senge, Acc. Chem. Res. 2005, 38, 733−743; (b) M. O. Senge, Y. M. Shaker, M. Pintea, C. Ryppa, S. S. Hatscher, A. A. Ryan, Y. Sergeeva, Eur. J. Org. Chem. 2010, 237−258. [237] A. Treibs, F.-H. Kreuzer, Liebigs Ann. Chem. 1968, 718, 208−223. [238] (a) E. V. de Wael, J. A. Pardoen, J. A. van Koeveringe, J. Lugtenburg, Recl. Trav. Chim. Pays-Bas 1977, 96, 306−309; (b) D. A. Lightner, M. Reisinger, W. M. D. Wijekoon, J. Org. Chem. 1987, 52, 5391−5395. [239] G. Ulrich, R. Ziessel, A. Harriman, Angew. Chem. Int. Ed. 2007, 47, 1184–1201. [240] H. R. A. Golf, H.-U. Reissig, A. Wiehe, J. Org. Chem. 2015, 80, 5133−5143. [241] A. Gheorghe, T. Chinnusamy, E. Cuevas-Yañez, P. Hilgers, O. Reiser, Org. Lett. 2008, 10, 4171−4174. [242] M. Shibuya, R. Doi, T. Shibuta, S.-i. Uesugi, Y. Iwabuchi, Org. Lett. 2012, 14, 5006−5009. [243] G. D. Vidya, R. D. Sulaksha, Curr. Org. Chem. 2017, 14, 1180−1184.

227

[244] M. Shibuya, M. Tomizawa, I. Suzuki, Y. Iwabuchi, J. Am. Chem. Soc. 2006, 128, 8412–8413. [245] R. V. Oppenauer, Recl. Trav. Chim. Pays-Bas 1937, 56, 137–144. [246] T. T. Tidwell, Org. React. 2004, 297–555. [247] E. J. Corey, P. L. Fuchs, Tetrahedron Lett. 1972, 13, 3769–3772. [248] P. Fackler, C. Berthold, F. Voss, T. Bach, J. Am. Chem. Soc. 2010, 132, 15911−15913. [249] (a) N. Aratani, D. Kim, A. Osuka, Acc. Chem. Res. 2009, 42, 1922–1934; (b) T. Tanaka, B. S. Lee, N. Aratani, M.-C. Yoon, D. Kim, A. Osuka, Chem. Eur. J. 2011, 17, 14400–14412; (c) T. Tanaka, A. Osuka, Chem. Soc. Rev. 2015, 44, 943–969. [250] (a) M. O. Senge, X. Feng, Tetrahedron Lett. 1999, 40, 4165–4168; (b) A. Tsuda, A. Osuka, Science 2001, 293, 79–82; (c) N. K. S. Davis, A. L. Thompson, H. L. Anderson, J. Am. Chem. Soc. 2011, 133, 30–31; (d) W. Auwärter, D. Écija, F. Klappenberger, J. V. Barth, Nat. Chem. 2015, 7, 105–120. [251] G. Emandi, Y. M. Shaker, K. J. Flanagan, J. M. O'Brien, M. O. Senge, Eur. J. Org. Chem. 2017, 6680–6692. [252] A. de Meijere, L. Zhao, V. N. Belov, M. Bossi, M. Noltemeyer, S. W. Hell, Chem. Eur. J. 2007, 13, 2503–2516. [253] W. Adcock, V. S. Iyer, J. Org. Chem. 1988, 53, 5259–5266. [254] M. Oi, R. Takita, J. Kanazawa, A. Muranaka, C. Wang, M. Uchiyama, Chem. Sci. 2019, 10, 6107–6112. [255] D. E. Polyansky, E. O. Danilov, S. V. Voskresensky, M. A. J. Rodgers, D. C. Neckers, J. Am. Chem. Soc. 2005, 127, 13452–13453. [256] Y. Liu, S. Jiang, K. Glusac, D. H. Powell, D. F. Anderson, K. S. Schanze, J. Am. Chem. Soc. 2002, 124, 12412–12413. [257] (a) K. Takahashi, Y. Kuroyama, K. Sonogashira, N. Hagihara, Synthesis 1980, 627– 630; (b) K. Sonogashira, Metal Catalyzed Cross-Coupling Reactions, Wiley-VCH, Weinheim, 1998. [258] C. Janiak, J. Chem. Soc., Dalton Trans. 2000, 3885–3896. [259] (a) M. Gouterman, J. Chem. Phys. 1959, 30, 1139−1161; (b) J. A. Shelnutt, V. Ortiz, J. Phys. Chem. 1985, 89, 4733−4739. [260] (a) K. M. Kadish, M. Lin, E. V. Caemelbecke, G. De Stefano, C. J. Medforth, D. J. Nurco, N. Y. Nelson, B. Krattinger, C. M. Muzzi, L. Jaquinod, Y. Xu, D. C. Shyr, K. M. Smith, J. A. Shelnutt, Inorg. Chem. 2002, 41, 6673–6687; (b) N. Grover, M. Sankar, Y. Song, K. M. Kadish, Inorg. Chem. 2016, 55, 584–597. [261] A. D. McNaught, A. Wilkinson, International Union of Pure and Applied Chemistry (IUPAC): Compendium of Chemical Terminology (the "Gold Book") Blackwell Scientific Publications, Oxford, 1997. [262] J. Karolczak, D. Kowalska, A. Lukaszewicz, A. Maciejewski, R. P. Steer, J. Phys. Chem. A 2004, 108, 4570–4575. [263] N. Boens, V. Leen, W. Dehaen, Chem. Soc. Rev. 2012, 41, 1130–1172.

228

[264] M. D. Peeks, C. E. Tait, P. Neuhaus, G. M. Fischer, M. Hoffmann, R. Haver, A. Cnossen, J. R. Harmer, C. R. Timmel, H. L. Anderson, J. Am. Chem. Soc. 2017, 139, 10461−10471. [265] M. D. Levin, P. Kaszynski, J. Michl, Chem. Rev. 2000, 100, 169–234. [266] P. K. Mykhailiuk, Org. Biomol. Chem. 2019, 17, 2839–2849. [267] (a) K. C. Nicolaou, J. Yin, D. Mandal, R. D. Erande, P. Klahn, M. Jin, M. Aujay, J. Sandoval, J. Gavrilyuk, D. Vourloumis, J. Am. Chem. Soc. 2016, 138, 1698–1708; (b) N. D. Measom, K. D. Down, D. J. Hirst, C. Jamieson, E. S. Manas, V. K. Patel, D. O. Somers, ACS Med. Chem. Lett. 2017, 8, 43–48; (c) Y. L. Goh, Y. T. Cui, V. Pendharkar, V. A. Adsool, ACS Med. Chem. Lett. 2017, 8, 516–520; (d) T. A. Reekie, C. M. Williams, L. M. Rendina, M. Kassiou, J. Med. Chem. 2019, 62, 1078– 1095. [268] (a) A. Rodríguez-Fortea, J. Kaleta, C. Mézière, M. Allain, E. Canadell, P. Wzietek, J. Michl, P. Batail, ACS Omega 2018, 3, 1293−1297; (b) K. J. Flanagan, S. S. R. Bernhard, S. Plunkett, M. O. Senge, Chem. Eur. J. 2019, 25, 6941−6954. [269] A. de Meijere, L. Zhao, V. N. Belov, M. Bossi, M. Noltemeyer, S. W. Hell, Chem. Eur. J. 2007, 13, 2503–2516. [270] J. M. Smith, S. J. Harwood, P. S. Baran, Acc. Chem. Res. 2018, 51, 1807–1817. [271] I. S. Makarov, C. E. Brocklehurst, K. Karaghiosoff, G. Koch, P. Knochel, Angew. Chem. Int. Ed. 2017, 56, 12774–12777. [272] M. J. Soth, G. Liu, K. Le, J. Cross, J. Philip, US Patent WO 2018/107072 A1, 2018. [273] (a) R. B. Merrifield, J. Am. Chem. Soc. 1963, 85, 2149–2154; (b) A. Graul, J. Castaner, Drugs Future 1997, 22, 956–968; (c) J. W. Bode, Curr. Opin. Drug Disc. Dev. 2006, 9, 765–775. [274] C. J. Hondros, K. Aravindu, J. R. Diers, D. Holten, J. S. Lindsey, D. F. Bocian, Inorg. Chem. 2012, 51, 11076–11086. [275] S. Mathew, M. R. Johnston, Chem. Eur. J. 2009, 15, 248–253. [276] E. Valeur, M. Bradley, Chem. Soc. Rev. 2009, 38, 606–631. [277] L. A. Carpino, H. Imazumi, B. M. Foxman, M. J. Vela, P. Henklein, A. El-Faham, J. Klose, M. Bienert, Org. Lett. 2000, 2, 2253−2256. [278] G. A. Sable, J. Park, H. Kim, S.-J. Lim, S. Jang, D. Lim, Eur. J. Org. Chem. 2015, 7043−7052. [279] D. Kumar, K. P. Chandra Shekar, B. Mishra, R. Kurihara, M. Ogura, T. Ito, Bioorg. Med. Chem. Lett. 2013, 23, 3221–3224. [280] (a) S. Hiroto, Y. Miyake, H. Shinokubo, Chem. Rev. 2017, 117, 2910–3043; (b) J. M. O’Brien, E. Sitte, K. J. Flanagan, H. Kühner, L. J. Hallen, D. Gibbons, M. O. Senge, J. Org. Chem. 2019, 84, 6158–6173. [281] J. D. D. Rehm, B. Ziemer, G. Szeimies, Eur. J. Org. Chem. 2001, 1049–1052. [282] H. Kalita, W.-Z. Lee, M. Ravikanth, Dalton Trans. 2013, 42, 14537–14544. [283] K.-S. Shin, M. Brezgunova, O. Jeannin, T. Roisnel, F. Camerel, P. Auban-Senzier, M. Fourmigué, Cryst. Growth Des. 2011, 11, 5337–5345. [284] J. M. Berg, J. L. Tymoczko, L. Stryer, Biochemistry (5th Ed.), W. H. Freeman & Co., New York, 2002.

229

[285] A. Macchioni, Chem. Rev. 2005, 105, 2039–2074. [286] (a) Y. Yamamoto, Y. Hirata, M. Kodama, T. Yamaguchi, S. Matsukawa, K.-y. Akiba, D. Hashizume, F. Iwasaki, A. Muranaka, M. Uchiyama, P. Chen, K. M. Kadish, N. Kobayashi, J. Am. Chem. Soc. 2010, 132, 12627–12638; (b) N. L. Bill, M. Ishida, S. Bähring, J. M. Lim, S. Lee, C. M. Davis, V. M. Lynch, K. A. Nielsen, J. O. Jeppesen, K. Ohkubo, S. Fukuzumi, D. Kim, J. L. Sessler, J. Am. Chem. Soc. 2013, 135, 10852–10862. [287] A. R. Mulholland, P. Thordarson, E. J. Mensforth, S. J. Langford, Org. Biomol. Chem. 2012, 10, 6045–6053. [288] J. Jiang, X. Fang, B. Liu, C. Hu, Inorg. Chem. 2014, 53, 3298–3306. [289] (a) J. L. Sessler, P. A. Gale, W.-S. Cho, Anion Receptor Chemistry, RSC Publishing, Cambridge, 2006; (b) Y. Ding, W.-H. Zhu, Y. Xie, Chem. Rev. 2017, 117, 2203– 2256; (c) M. Kielmann, C. Prior, M. O. Senge, New J. Chem. 2018, 42, 7529–7550; (d) M. Kielmann, M. O. Senge, Angew. Chem. Int. Ed. 2019, 58, 418–441. [290] K. J. Waldron, J. C. Rutherford, D. Ford, N. J. Robinson, Nature 2009, 460, 823– 830. [291] (a) G. Hungerford, M. Van der Auweraer, J.-C. Chambron, V. Heitz, J.-P. Sauvage, J.-L. Pierre, D. Zurita, Chem. Eur. J. 1999, 5, 2089–2100; (b) Y. Li, X. Li, Y. Li, H. Liu, S. Wang, H. Gan, J. Li, N. Wang, X. He, D. Zhu, Angew. Chem. Int. Ed. 2006, 45, 3639–3643; (c) D. P. Cormode, M. G. B. Drew, R. Jagessar, P. D. Beer, Dalton Trans. 2008, 6732–6741. [292] A. Buccolieri, M. Hasan, S. Bettini, V. Bonfrate, L. Salvatore, A. Santino, V. Borovkov, G. Giancane, Anal. Chem. 2018, 90, 6952–6958. [293] H. A. Benesi, J. H. Hildebrand, J. Am. Chem. Soc. 1949, 71, 2703–2707. [294] P. Brodard, S. Matzinger, E. Vauthey, O. Mongin, C. Papamicaël, A. Gossauer, J. Phys. Chem. A 1999, 103, 5858–5870. [295] N. R. Finsen, La Photothérapie, Paris 1899. [296] (a) O. Raab, Z. Biol. 1900, 39, 524–546; (b) H. v. Tappeiner, Münch. Med. Wochenschr. 1900, 47, 5–7; (c) H. v. Tappeiner, A. Jodlbauer, Die Sensibilisierende Wirkung Fluorescierender Substanzen Gesammelte Untersuchungen über die Photodynamische Erscheinung, Vogel, Leipzig, 1907; (d) A. Jesionek, Lichtbiologie. Die Experimentellen Grundlagen der Modernen Lichtbehandlung, Vieweg, Braunschweig, 1910. [297] S. Callaghan, M. O. Senge, Photochem. Photobiol. Sci. 2018, 17, 1490–1514. [298] X. Wen, Y. Li, M. R. Hamblin, Photodiagn. Photodyn. Ther. 2017, 19, 140–152. [299] (a) T. J. Dougherty, C. J. Gomer, B. W. Henderson, G. Jori, D. Kessel, M. Korbelik, J. Moan, Q. Peng, J. Natl. Cancer Inst. 1998, 90, 889–905; (b) S. B. Brown, E. A. Brown, I. Walker, Lancet Oncol. 2004, 5, 497–508. [300] U. Schmidt-Erfurth, T. Hasan, Surv. Ophthalmol. 2000, 45, 195–214. [301] (a) M. R. Hamblin, T. Hasan, Photochem. Photobiol. Sci. 2004, 3, 436–450; (b) A. Wiehe, J. M. O'Brien, M. O. Senge, Photochem. Photobiol. Sci. 2019, DOI: 10.1039/C9PP00211A; (c) B. Khurana, P. Gierlich, A. Meindl, L. C. Gomes-da-Silva, M. O. Senge, Photochem. Photobiol. Sci. 2019, DOI: 10.1039/C9PP00221A. [302] V. Kumar, A. Sharma, Int. Immunopharmacol. 2010, 10, 1325–1334.

230

[303] M. C. DeRosa, R. J. Crutchley, Coord. Chem. Rev. 2002, 233–234, 351–371. [304] (a) D. L. Gilbert, C. A. Colton, Carbohydr. Polym. 2001, 46, 195–200; (b) P. R. Ogilby, Chem. Soc. Rev. 2010, 39, 3181–3209. [305] (a) E. l. G. Azenha, A. C. Serra, M. Pineiro, M. M. Pereira, J. Seixas de Melo, L. G. Arnaut, S. J. Formosinho, A. M. d. A. Rocha Gonsalves, Chem. Phys. 2002, 280, 177–190; (b) A. Gorman, J. Killoran, C. O'Shea, T. Kenna, W. M. Gallagher, D. F. O'Shea, J. Am. Chem. Soc. 2004, 126, 10619–10631; (c) F. M. Pimenta, R. L. Jensen, L. Holmegaard, T. V. Esipova, M. Westberg, T. Breitenbach, P. R. Ogilby, J. Phys. Chem. B 2012, 116, 10234–10246. [306] H.-G. Jeong, M.-S. Choi, Isr. J. Chem. 2016, 56, 110–118. [307] (a) T. J. Dougherty, C. J. Gomer, B. W. Henderson, G. Jori, D. Kessel, M. Korbelik, J. Moan, Q. Peng, J. Natl. Cancer Inst. 1998, 90, 889–905; (b) V. G. Schweitzer, M. L. Somers, Oncol. Rev. 2010, 4, 203–209; (c) M. Panjehpour, B. F. Overholt, Front. Gastrointest. Res. 2010, 27, 128–139; (d) R. Allison, K. Moghissi, G. Downie, K. Dixon, Photodiagn. Photodyn. Ther. 2011, 8, 231–239. [308] (a) J. F. Lovell, A. Roxin, K. K. Ng, Q. Qi, J. D. McMullen, R. S. DaCosta, G. Zheng, Biomacromolecules 2011, 12, 3115–3118; (b) H. Huang, R. Hernandez, J. Geng, H. Sun, W. Song, F. Chen, S. A. Graves, R. J. Nickles, C. Cheng, W. Cai, J. F. Lovell, Biomaterials 2016, 76, 25–32. [309] Y. Li, R. Geyer, D. Sen, Biochemistry 1996, 35, 6911–6922. [310] C. A. Busby, R. K. Dinello, D. Dolphin, Can. J. Chem. 1975, 53, 1554–1555. [311] M. Merchat, G. Bertolini, P. Giacomini, A. Villaneuva, G. Jori, J. Photochem. Photobiol. B 1996, 32, 153–157. [312] K. M. Kadish, K. M. Smith, R. Guilard, (Eds.), The Porphyrin Handbook Vol. 1, Elsevier, Oxford, 1999. [313] A. J. F. N. Sobral, L. L. G. Justino, A. C. C. Santos, J. A. Silva, C. T. Arranja, M. R. Silva, A. M. Beja, J. Porphyr. Phthalocyanines 2008, 12, 845–848. [314] A .M. d'A. Rocha Gonsalves, R. A. W. Johnstone, M. M. Pereira, A. M. P. de SantAna, A. C. Serra, A. J. F. N. Sobral, P. A. Stocks, Heterocycles 1996, 43, 829– 838. [315] N. Fukui, H. Yorimitsu, A. Osuka, Chem. Eur. J. 2016, 22, 18476–18483. [316] O. Locos, B. Bašić, J. C. McMurtrie, P. Jensen, D. P. Arnold, Chem. Eur. J. 2012, 18, 5574–5588. [317] X. Feng, M. O. Senge, J. Chem. Soc., Perkin Trans. 1 2001, 1030–1038. [318] Y. Cakmak, S. Kolemen, S. Duman, Y. Dede, Y. Dolen, B. Kilic, Z. Kostereli, L. T. Yildirim, A. L. Dogan, D. Guc, E. U. Akkaya, Angew. Chem. Int. Ed. 2011, 50, 11937–11941. [319] K. Żamojć, M. Zdrowowicz, P. Rudnicki-Velasquez, K. Krzymiński, B. Zaborowski, P. Niedziałkowski, D. Jacewicz, L. Chmurzyński, Free Radical Res. 2016, 51, 1–32. [320] P. Carloni, E. Damiani, L. Greci, P. Stipa, F. Tanfani, E. Tartaglini, M. Wozniak, Res. Chem. Intermed. 1993, 19, 395−405. [321] (a) in Saint, Version 8.37a ed., Bruker AXS, Inc., Madison, WI, 2013; (b) in SADABS, version 2016/2 ed., Bruker AXS, Inc, Madison, WI,, 2014; (c) in APEX3, Version 2016.9-0 ed., Bruker AXS, Inc., Madison, WI,, 2016.

231

[322] (a) O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. K. Howard, H. Puschmann, J. Appl. Crystallogr. 2009, 42, 339–341; (b) G. Sheldrick, Acta Cryst. Sect. A 2015, 71, 3–8. [323] W. W. Kalisch, M. O. Senge, Angew. Chem. Int. Ed. Engl. 1998, 37, 1107–1109. [324] S. Shanmugathasan, C. K. Johnson, C. Edwards, E. K. Matthews, D. Dolphin, R. W. Boyle, J. Porphyrins Phthalocyanines 2000, 4, 228–232. [325] R. W. Boyle, C. K. Johnson, D. Dolphin, J. Chem. Soc., Chem. Commun. 1995, 527–528. [326] S. Plunkett, K. Dahms, M. O. Senge, Eur. J. Org. Chem. 2013, 1566–1579. [327] C. Beinat, S. D. Banister, J. Hoban, J. Tsanaktsidis, A. Metaxas, A. D. Windhorst, M. Kassiou, Bioorg. Med. Chem. Lett. 2014, 24, 828–830. [328] A. D. Adler, F. R. Longo, J. D. Finarelli, J. Goldmacher, J. Assour, L. Korsakoff, J. Org. Chem. 1967, 32, 476–476.

232