Transition Metal and Lanthanide Chemistry of Dipyridylpyrrolide Ligands

Home , Fulvalene, Hemilability

TRANSITION METAL AND LANTHANIDE

CHEMISTRY OF DIPYRIDYLPYRROLIDE

LIGANDS

James McPherson

A thesis presented in fulfilment of the requirements for the degree of Doctor of Philosophy

School of Chemistry

Faculty of Science

June 2019

Thesis/Dissertation Sheet

Surname/Family Name : McPherson Given Name/s : James Abbreviation for degree as give in the University calendar : PhD Faculty : Science School : Chemistry Thesis Title : Transition metal and lanthanide chemistry of dipyridylpyrrolide ligands

Abstract

Pincer ligands attract great current interest because they bestow exceptional stability on their metal complexes. Modification of the pincer scaffold can be used to tune the electron density at the metal ion and, therefore, the physical and chemical properties. This thesis reports studies of 2,5-di(2- pyridyl)pyrrolide (dpp–) NNN-pincer ligands and their transition and rare-earth metal complexes. Homoleptic complexes were prepared and screened for useful physical (e.g., magnetic or photophysical) or chemical (e.g., redox or reactive) properties. Predictable control of the properties through the substitution of the pyrrolide ring was a particular goal, and was achieved.

The critical review of the literature on the chemistry of dpp– ligands and related pyridine-pyrrolide pincers, Chapter 1, concluded that a range of dpp– derivatives with flexible coordination modes was accessible, but with underdeveloped coordination chemistry.

– Chapter 2 provides a theoretical quantification (B3LYP-GD3BJ/6-31G** calculations) of the internal stain (∆EL(strain)) that builds when a dpp or related pyridine-pyrrolide pincer ligand binds a metal ion. A relationship was discovered correlating ∆EL(strain) and the change in interatomic separation of the two terminal donor atoms when bound (ρL(coord)) and metal-free (ρL(relax)). The relationship was employed to explain, a posteriori, the occurrence of spin crossover activity in Fe(II)-pincer complexes.

R1,R2 A series of five structurally similar cobalt(II) complexes, [Co(dpp )2], bearing different pyrrolide 3,4-substituents are discussed in Chapter 3. The oxidation and spin states of the cobalt ion were switched by appropriate external stimuli, with the CoIII/CoII couple predictably sensitive to the Swain- Lupton electronic parameters for the pyrrolide substituents.

Chapter 4 describes the first lanthanide(III) complexes of dpp– ligands. Ligand-sensitised, lanthanide-centred emission was observed for CO Me,Me CO Me,Me [Eu(dpp 2 )3] (red emission) and [Yb(dpp 2 )3] (near-infrared emission).

2+ – Chapter 5 reports enhanced reactivity for [Ru(dpp)2] complexes compared to the ubiquitous [Ru(tpy)2] . Hemilability of the dpp ligand results from its large ρL(relax), and this was exploited for catalytic transformations of small molecules bound at the ruthenium(II) centre.

Four small preliminary studies are included in Chapter 6, each providing an entry point into more detailed future investigations.

Chapter 7 provides a summary of the results, highlights advances made and suggests how these could be used into the future.

Declaration relating to disposition of project thesis/dissertation

I hereby grant to the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or in part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all property rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation.

I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstracts International (this is applicable to doctoral theses only).

…………………………………………………………… ……………………………………..……………… ……….……………………...…….… Signature Witness Signature Date The University recognises that there may be exceptional circumstances requiring restrictions on copying or conditions on use. Requests for restriction for a period of up to 2 years must be made in writing. Requests for a longer period of restriction may be considered in exceptional circumstances and require the approval of the Dean of Graduate Research.

FOR OFFICE USE ONLY Date of completion of requirements for Award:

‘I hereby grant the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all proprietary rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation. I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstract International (this is applicable to doctoral theses only). I have either used no substantial portions of copyright material in my thesis or I have obtained permission to use copyright material; where permission has not been granted I have applied/will apply for a partial restriction of the digital copy of my thesis or dissertation.'

Signed ……………………………………………......

Date ……………………………………………......

AUTHENTICITY STATEMENT

‘I certify that the Library deposit digital copy is a direct equivalent of the final officially approved version of my thesis. No emendation of content has occurred and if there are any minor variations in formatting, they are the result of the conversion to digital format.’

Signed ……………………………………………......

Date ……………………………………………...... ORIGINALITY STATEMENT

I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial portions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis.

Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project’s design and conception or style, presentation and linguistic expression is acknowledged.

James McPherson

March 2019 INCLUSION OF PUBLICATIONS STATEMENT

UNSW is supportive of candidates publishing their research results during their candidature as detailed in the UNSW Thesis Examination Procedure.

Publications can be used in their thesis in lieu of a Chapter if: • The student contributed greater than 50% of the content in the publication and is the “primary author”, ie. the student was responsible primarily for the planning, execution and preparation of the work for publication • The student has approval to include the publication in their thesis in lieu of a Chapter from their supervisor and Postgraduate Coordinator. • The publication is not subject to any obligations or contractual agreements with a third party that would constrain its inclusion in the thesis

Please indicate whether this thesis contains published material or not.

This thesis contains no publications, either published or submitted for publication ☐ Some of the work described in this thesis has been published and it has been ☐ documented in the relevant Chapters with acknowledgement This thesis has publications (either published or submitted for publication) ☒ incorporated into it in lieu of a chapter and the details are presented below

CANDIDATE’S DECLARATION I declare that: • I have complied with the Thesis Examination Procedure • where I have used a publication in lieu of a Chapter, the listed publication(s) below meet(s) the requirements to be included in the thesis. Name Signature Date (dd/mm/yy) James McPherson

Postgraduate Coordinator’s Declaration I declare that: • the information below is accurate • where listed publication(s) have been used in lieu of Chapter(s), their use complies with the Thesis Examination Procedure • the minimum requirements for the format of the thesis have been met. PGC’s Name PGC’s Signature Date (dd/mm/yy) Prof. Naresh Kumar Details of publication #1: Full title: Tridentate pyridine-pyrrolide chelate ligands: An under-appreciated ligand set with an immensely promising coordination chemistry Authors: James N. McPherson, Biswanath Das and Stephen B. Colbran Journal or book name: Coordination Chemistry Reviews Volume/page numbers: 375, 285-332 Date accepted/ published: 28/3/2018 Status Published X Accepted and In press In progress (submitted) The Candidate’s Contribution to the Work Prepared and compiled the manuscript for submission. Wrote the major section describing the complexes of 2,5-di(2-pyridyl)pyrrolide ligands and their pyrrole precursors. Location of the work in the thesis and/or how the work is incorporated in the thesis: Chapter 1: Introduction Primary Supervisor’s Declaration I declare that: • the information above is accurate • this has been discussed with the PGC and it is agreed that this publication can be included in this thesis in lieu of a Chapter • All of the co-authors of the publication have reviewed the above information and have agreed to its veracity by signing a ‘Co-Author Authorisation’ form. Supervisor’s name Supervisor’s signature Date (dd/mm/yy) A/Prof. Stephen Colbran

Details of publication #2: Full title: A Strain-Deformation Nexus within Pincer Ligands: Application to the Spin States of Iron(II) Complexes Authors: James N. McPherson, Timothy E. Elton and Stephen B. Colbran Journal or book name: Inorganic Chemistry Volume/page numbers: 57, 12312–12322 Date accepted/ published: 10/9/2018 Status Published X Accepted and In press In progress (submitted) The Candidate’s Contribution to the Work Assisted with the experimental design, performed the calculations and the analysis of the crystallographic database. Wrote the draft manuscript and prepared the ESI, wrote a draft response to reviewer comments. Location of the work in the thesis and/or how the work is incorporated in the thesis: Chapter 2: Geometric analysis of strain in pincer ligands Primary Supervisor’s Declaration I declare that: • the information above is accurate • this has been discussed with the PGC and it is agreed that this publication can be included in this thesis in lieu of a Chapter • All of the co-authors of the publication have reviewed the above information and have agreed to its veracity by signing a ‘Co-Author Authorisation’ form. Supervisor’s name Supervisor’s signature Date (dd/mm/yy) A/Prof. Stephen Colbran

Details of publication #3: Full title: Predictable Substituent Control of CoIII/II Redox Potential and Spin Crossover in Bis(dipyridylpyrrolide)cobalt Complexes Authors: James N. McPherson, Ross W. Hogue, Folaranmi Sunday Akogun, Luca Bondì, Ena T. Luis, Jason R. Price, Anna L. Garden, Sally Brooker and Stephen B. Colbran Journal or book name: Inorganic Chemistry Volume/page numbers: 58, 2218–2228 Date accepted/ published: 23/1/2019 Status Published X Accepted and In press In progress (submitted) The Candidate’s Contribution to the Work Experimental design and coordination. Synthesis of all ligands and all complexes. NMR, UV-vis-NIR and electrochemical characterisation. Analysis of all solid-state data post-refinement. Analysis of structural and magnetic properties. Analysis of ligand electronics and sterics against the magnetic properties of the complexes. Correlation of ligand electronics with redox chemistry of the complex. Wrote draft manuscript except for sections on DFT calculations of Co(II)/Co(III) couples and the solid-state magnetic susceptibility measurements; prepared ESI. Location of the work in the thesis and/or how the work is incorporated in the thesis: Chapter 3: Predictable substituent control of CoIII/CoII redox potential and spin crossover in bis (dipyridylpyrrolides)cobalt complexes Primary Supervisor’s Declaration I declare that: • the information above is accurate • this has been discussed with the PGC and it is agreed that this publication can be included in this thesis in lieu of a Chapter • All of the co-authors of the publication have reviewed the above information and have agreed to its veracity by signing a ‘Co-Author Authorisation’ form. Supervisor’s name Supervisor’s signature Date (dd/mm/yy) A/Prof. Stephen Colbran

Details of publication #4: Full title: Synthesis and Structural, Redox and Photophysical Properties of Tris-(2,5-di(2-pyridyl)pyrrolide) Lanthanide Complexes Authors: James N. McPherson, Laura Abad Galan, Hasti Iranmanesh, Max Massi and Stephen B. Colbran Journal or book name: Dalton Transactions Volume/page numbers: N/A Date accepted/ published: N/A Status Published Accepted and In press In progress X (submitted) The Candidate’s Contribution to the Work Experimental design and coordination, prepared and characterised ligands and complexes except for collection of the SC-XRD data (H. Iranmanesh) and the photophysical data (L. Abad Galan). Wrote draft manuscript except for the photophysical section (L. Abad Galan). Location of the work in the thesis and/or how the work is incorporated in the thesis: Chapter 4: Neutral Lanthanide Complexes Primary Supervisor’s Declaration I declare that: • the information above is accurate • this has been discussed with the PGC and it is agreed that this publication can be included in this thesis in lieu of a Chapter • All of the co-authors of the publication have reviewed the above information and have agreed to its veracity by signing a ‘Co-Author Authorisation’ form. Supervisor’s name Supervisor’s signature Date (dd/mm/yy) A/Prof. Stephen Colbran

Transition metal and lanthanide chemistry of dipyridylpyrrolide ligands James McPherson – June-2019

ABSTRACT

(e.g., magnetic or photophysical) or chemical (e.g., redox or reactive) properties. Predictable control of the properties through the substitution of the pyrrolide ring was a particular goal, and was achieved.

The critical review of the literature on the chemistry of dpp– ligands and related pyridine- pyrrolide pincers, Chapter 1, concluded that a range of dpp– derivatives with flexible coordination modes was accessible, but with underdeveloped coordination chemistry.

Chapter 2 provides a theoretical quantification (B3LYP-GD3BJ/6-31G** calculations)

– of the internal stain (∆EL(strain)) that builds when a dpp or related pyridine-pyrrolide pincer ligand binds a metal ion. A relationship was discovered correlating ∆EL(strain) and the change in interatomic separation of the two terminal donor atoms when bound (ρL(coord)) and metal-free

(ρL(relax)). The relationship was employed to explain, a posteriori, the occurrence of spin crossover activity in Fe(II)-pincer complexes.

R1,R2 A series of five structurally similar cobalt(II) complexes, [Co(dpp )2], bearing different pyrrolide 3,4-substituents are discussed in Chapter 3. The oxidation and spin states of the cobalt ion were switched by appropriate external stimuli, with the CoIII/CoII couple predictably sensitive to the Swain-Lupton electronic parameters for the pyrrolide substituents.

i Transition metal and lanthanide chemistry of dipyridylpyrrolide ligands James McPherson – June-2019

Chapter 4 describes the first lanthanide(III) complexes of dpp– ligands. Ligand-sensitised,

CO Me,Me lanthanide-centred emission was observed for [Eu(dpp 2 )3] (red emission) and

CO Me,Me [Yb(dpp 2 )3] (near-infrared emission).

Chapter 5 reports enhanced reactivity for [Ru(dpp)2] complexes compared to the

2+ – ubiquitous [Ru(tpy)2] . Hemilability of the dpp ligand results from its large ρL(relax), and this was exploited for catalytic transformations of small molecules bound at the ruthenium(II) centre.

Four small preliminary studies are included in Chapter 6, each providing an entry point into more detailed future investigations.

Chapter 7 provides a summary of the results, highlights advances made and suggests how these could be used into the future.

ACKNOWLEDGEMENTS

This work would not have been possible without the love and support of friends, family and colleagues. Thank you to you all.

I must thank Steve Colbran for his support and guidance over the last five years. I have thoroughly enjoyed working on this project and have learnt more than a little along the way.

Thank you for always having time to listen to me work through things for myself, and for the encouragement to follow my own ideas and intuition. I wish you all the best in retirement, and

I look forward to seeing more paintings and strong opinions.

Next, I have been most fortunate to have collaborated with some brilliant scientists. To

Max and Sally – my fondest memories during this research, hands down, have been when you each approached me following a presentation and offered your expertise. Thank you. To the other students who assisted, thank you also, for your contributions to our work. To Dr Jason

Price, thanks for battling away with my cobalt structure, and more generally for your work maintaining and improving the Synchrotron facilities, which have been so important during this work.

To the academic department of the School of Chemistry at UNSW – thank you to everyone who, at various stages, has shown interest in my work. Whether it was chatting at a poster, sitting on a panel, or even acknowledgement after a talk – don’t underestimate how much it means to a student for a member of academic staff to show interest in what we are working on. I’d like to particularly thank Dr Jon Beves, my co-supervisor, who has always been prepared to provide letters of support or words of advice on the few occasions Steve has not been available (often at short notice). Thanks also must go to Dr Scott Sulway, A/Prof Graham

Ball and A/Prof Jason Harper for making their expert advice available as required.

iii

Thanks also go to the analytical staff at MWAC, particularly Mohan at the SSEAU, and

Adelle, Doug, Don and Jim of the NMR Facility, for the excellent services provided, particularly for their training and assistance with troubleshooting.

I have very much enjoyed attending many meetings of the local Inorganic and

Supramolecular Chemistry communities, and particularly found the OZOM events inspiring and refreshing. I’ve made lifelong friendships at these meetings and I hope to remain an active member of these communities for many years. To those who have accompanied me while travelling, whether short trips to Melbourne, or longer trips to Japan or around the US, thanks for sharing the fantastic memories.

My life has been centred around UNSW the last 4 years, but it hasn’t all been hard work.

Without naming names (there are too many) – thank you to the Round Table/Rege Pizza crew for making work a place I look forward to showing up to. Thanks also to the Beves group, for including me in your Christmas parties and camping getaways. But most importantly: to the members of the Colbran Group, my partners in crime, thanks for being the legends that you are.

I’m looking forward to catching up with all my friends and family outside UNSW. Thanks to you all for your patience and for showing interest in the work I’ve been doing. To Mum and

Dad, Ian, Cath, Anna (Happy Birthday, looks like I’ll miss dinner tonight) and Lucy: thanks for your unconditional love and support, even though you do all follow the wrong footy team.

To you, the reader, I hope this thesis helps with whatever question has prompted you to flick through it.

Ena, for helping me find love and joy in the minutia, thank you.

LIST OF PUBLICATIONS

(1) McPherson, J. N.; Abad Galan, L.; Iranmanesh, H.; Massi, M.; Colbran, S. B.

“Synthesis and Structural, Redox and Photophysical Properties of Tris-(2,5-di(2-

pyridyl)pyrrolide) Lanthanide Complexes” Dalton Trans., 2019, DOI:

10.1039/c9dt01262a.

(2) McPherson, J. N.; Hogue, R. W.; Akogun, F. S.; Bondì, L.; Luis, E. T.; Price, J. R.;

Garden, A. L.; Brooker, S.; Colbran, S. B. “Predictable Substituent Control of

CoIII/II Redox Potential and Spin Crossover in Bis(dipyridylpyrrolides)cobalt

Complexes” Inorg. Chem. 2019, 58 (3), 2218.

(3) McPherson, J. N.; Elton, T. E.; Colbran, S. B. “A Strain-Deformation Nexus within

Pincer Ligands: Application to the Spin States of Iron(II) Complexes” Inorg.

Chem. 2018, 57 (19), 12312.

(4) McPherson, J. N.; Das, B.; Colbran, S. B. “Tridentate pyridine-pyrrolide chelate

ligands: An under-appreciated ligand set with an immensely promising

coordination chemistry” Coord. Chem. Rev. 2018, 375, 285.

(5) Das, B.; McPherson, J. N.; Colbran, S. B. “Oligomers and macrocycles with

[m]pyridine[n]pyrrole (m + n ≥ 3) domains: Formation and applications of anion,

guest molecule and metal ion complexes” Coord. Chem. Rev. 2018, 363, 29.

CONFERENCE PRESENTATIONS

Oral Presentations

2018 RACI NSW INORGANIC SYMPOSIUM University of Sydney. 19th November 2018 Coordination chemistry of the dipyridylpyrrolide ligands

INTERNATIONAL CONFERENCE ON COORDINATION CHEMISTRY (ICCC2018) Sendai International Center. Japan. 30th July–4th August 2018 The dipyridylpyrrolide ligand – an overlooked analogue of terpyridine

11TH AUSTRALASIAN ORGANOMETALLICS MEETING (OZOM11) The University of Western Australia. 16th–19th January 2018 Dipyridylpyrrolide anion analogues of terpyridine metal complexes

6TH ASIAN CONFERENCE ON COORDINATION CHEMISTRY (ACCC6) Melbourne Convention and Exhibition Centre. 24th–28th July 2017 Dipyridylpyrrolide anion analogues of terpyridine metal complexes

10TH AUSTRALASIAN ORGANOMETALLICS MEETING (OZOM10) The University of Otago, New Zealand. 10th–13th January 2017 Dipyridylpyrrolide anion analogues of terpyridine metal complexes

2016 RACI NSW INORGANIC SYMPOSIUM UNSW Sydney. 24th November 2016 Dipyridylpyrrolato anion analogues of terpyridine metal complexes

Poster Presentations

INORGANIC CHEMISTRY GORDON RESEARCH CONFERENCE AND SEMINAR The University of New England, Maine USA. 16th–22nd June 2018 The dipyridylpyrrolide ligand –more than just anionic terpyridine

6TH ASIAN CONFERENCE ON COORDINATION CHEMISTRY (ACCC6) Melbourne Convention and Exhibition Centre. 24th–28th July 2017 Dipyridylpyrrolide anion analogues of terpyridine metal complexes

27TH INTERNATIONAL CONFERENCE ON ORGANOMETALLIC CHEMISTRY (ICOMC27) Melbourne Convention and Exhibition Centre. 17th–22nd July 2016 Dipyridylpyrrolato anion analogues of terpyridine metal complexes

9TH AUSTRALASIAN ORGANOMETALLICS MEETING (OZOM9) The University of Sydney. 8th–11th December 2015 Dipyridylpyrrolato anion analogues of terpyridine metal complexes

CONTENTS

1 INTRODUCTION ...... 1

1.1 PRIOR PUBLICATION STATEMENT ...... 1

1.2 CONTRIBUTIONS ...... 1

1.3 FOREWORD ...... 1

2 GEOMETRIC ANALYSIS OF STRAIN IN PINCER LIGANDS ...... 51

2.1 PRIOR PUBLICATION STATEMENT ...... 51

2.2 CONTRIBUTIONS ...... 51

2.3 FOREWORD ...... 51

3 PREDICTABLE SUBSTITUENT CONTROL OF COBALT (III/II) REDOX

POTENTIAL AND SPIN CROSSOVER IN COBALT COMPLEXES...... 65

3.1 PRIOR PUBLICATION STATEMENT ...... 65

3.2 CONTRIBUTIONS ...... 65

3.3 FOREWORD ...... 66

4 NEUTRAL LANTHANIDE COMPLEXES ...... 79

4.1 PRIOR PUBLICATION STATEMENT ...... 79

4.2 CONTRIBUTIONS ...... 79

4.3 FOREWORD ...... 79

5 HEMILABILE RUTHENIUM(II) CATALYSTS ...... 93

5.1 INTRODUCTION ...... 93

5.2 RESULTS AND DISCUSSION ...... 94

5.2.1 Synthesis and Characterisation ...... 94

5.2.2 Electrochemistry ...... 98

5.2.3 Reversible Solvent Binding ...... 98

5.2.4 Activation of Small Molecules for Catalysis ...... 112

vii

5.3 CONCLUSIONS ...... 126

5.4 EXPERIMENTAL ...... 127

5.4.1 General Methods ...... 127

5.4.2 Synthesis ...... 129

5.4.3 DFT Calculations ...... 129

5.4.4 Small Scale Stoichiometric Reactions ...... 130

5.4.5 Catalytic Screening Experiments ...... 133

5.5 REFERENCES ...... 134

6 MISCELLANEA...... 138

6.1 HOMOLEPTIC TRANSITION METAL COMPLEXES ...... 138

6.1.1 Synthesis and Characterisation ...... 139

6.1.2 NMR Spectroscopy ...... 140

6.1.3 Crystallography ...... 143

6.1.4 Electrochemistry...... 148

6.1.5 Conclusions ...... 152

6.2 [M(DPP)2] METALLOLIGANDS ...... 152

6.2.1 Introduction ...... 152

6.2.2 Ligand Synthesis ...... 155

CO ,CO + 6.2.3 [Co(H2dpp 2 2)2] ...... 158

6.2.4 Conclusions ...... 164

6.3 ORGANOMETALLIC PIANO-STOOL [CP*M(κ2-DPP)CL] COMPLEXES FOR MULTI-

ELECTRON REDUCTION ...... 164

6.3.1 Synthesis and Characterisation ...... 167

6.3.2 Electrochemistry...... 175

6.3.3 Conclusions ...... 180

viii

6.4 3-PYRIDIN-2-YL-N-[(1E)-PYRIDIN-2-YLMETHYLIDENE]IMIDAZO[1,5-A]

PYRIDINAMINE (LIMPY) – A SIDE-PRODUCT WITH PROMISING COORDINATION CHEMISTRY 180

6.4.1 Ligand Synthesis and Characterisation ...... 182

6.4.2 Synthesis and Characterisation of Metal Complexes ...... 186

6.4.3 Electrochemistry ...... 193

6.4.4 Conclusions ...... 197

6.5 EXPERIMENTAL ...... 197

6.5.1 General Methods ...... 197

6.5.2 Homoleptic Transition Metal Complexes ...... 199

6.5.3 [M(dpp)2] Metalloligands ...... 203

6.5.4 Organometallic Piano-stool [Cp*M(κ2-dpp)Cl] complexes ...... 206

6.5.5 3-Pyridin-2-yl-N-[(1E)-pyridin-2-ylmethylidene]imidazo[1,5-a] pyridinamine

(LImPy) – A Side-product With Promising Coordination Chemistry...... 211

6.6 REFERENCES ...... 216

7 CONCLUSIONS AND OUTLOOK ...... 221

7.1 REFERENCES ...... 230

8 APPENDICES ...... 8-231

LIST OF ABBREVIATIONS AND ACRONYMS

ADP Anisotropic Displacement Parameter ATR FTIR Attenuated Total Reflectance Fourier Transform Infrared Spectroscopy bpy 2,2′-bipyridine CIP Cahn-Ingold-Prelog COSY Correlation Spectroscopy Cp* 1,2,3,4,5-pentamethylcyclopentadienyl CV Cyclic Voltammogram DFT Density Functional Theory DME 1,2-dimethoxyethane DMF N,N-dimethylformamide DMSO Dimethylsulfoxide dpp–/Hdpp Dipyridylpyrrolide/dipyridylpyrrole EXSY Exchange Spectroscopy Fc/Fc+ Ferrocene/ferrocenium HMBC Heteronuclear Multiple Bond Correlation HSQC Heteronuclear Single Quantum Coherence HR-NSI-MS High-Resolution Nanospray Ionisation Mass Spectrometry LImPy 3-Pyridin-2-yl-N-[(1E)-pyridin-2-ylmethylidene]imidazo[1,5-a] pyridinamine LR-ESI-MS Low-Resolution Electrospray Ionisation Mass Spectrometry MLCT Metal-ligand charge transfer MR(I) Magnetic Resonance (Imaging) NADH/NAD+ Nicotinamide adenine dinucleotide NADPH/NADP+ Nicotinamide adenine dinucleotide phosphate NIR Near-Infrared NMR Nuclear Magnetic Resonance nOe nuclear Overhauser effect nOeSY nuclear Overhauser effect Spectroscopy OHD/OHD+ Organic Hydride Donor/Acceptor Phen 1,10-phenanthroline py pyridyl pyr pyrrolide/pyrrolyl SC-XRD Single Crystal X-Ray Diffraction SCO Spin Crossover TFE 2,2,2-trifluoroethanol tpy 2,2′:6′,2′′-terpyridine UV-vis Ultraviolet-visible Spectroscopy VT Variable Temperature

LIST OF APPENDICES

APPENDIX 1 – SUPPORTING INFORMATION FOR CHAPTER 2 ...... 8-233 APPENDIX 2 – SUPPORTING INFORMATION FOR CHAPTER 3 ...... 8-273 APPENDIX 3 – SUPPORTING INFORMATION FOR CHAPTER 4 ...... 8-333 APPENDIX 4 – SUPPORTING INFORMATION FOR CHAPTER 5 ...... 8-355 APPENDIX 5 – CRYSTALLOGRAPHIC DATA...... 8-365 APPENDIX 6 – RESPONSES TO EXAMINER COMMENTS ...... 8-367

Chapter 1: Introduction

1 INTRODUCTION

1.1 Prior Publication Statement

The work in this chapter was published by Elsevier in Coordination Chemistry Reviews

in the 2018 Special Issue on Coordination Chemistry in Australia. This article has been

reprinted with permission from McPherson, J. N.; Das, B.; Colbran, S. B., Tridentate pyridine-

pyrrolide chelate ligands: An under-appreciated ligand set with an immensely promising

1.2 Contributions

JNM wrote the major section, section 2, commented in detail on the drafts of all other

sections, compiled the whole manuscript, numbered all compounds and Figures, Schemes and

Tables, and wrote a draft response to reviewer comments.

JNM and SBC discussed the general introduction and conclusions, sections 1 and 6, and

SBC wrote them.

BD prepared section 4 and proofread the manuscript. SBC wrote sections 1, 3, 5 and 6.

SBC mentored JNM and BD.

1.3 Foreword

This PhD research investigated the coordination chemistry of 2,5-di(2-pyridyl)pyrrolide

(dpp–) ligands, a tridentate and meridional “NNN pincer” ligand, which bears striking

resemblance to the ubiquitous 2,2′:6′,2″-terpyridine (tpy). This chapter reviews the chemistry

of metal complexes of all anionic tridentate NNN pyridine-pyrrolide pincers (i.e. those ligands

formed when one or more of the pyridine rings in tpy are replaced by a pyrrolide). Synthetic

James McPherson – June 2019 1 Chapter 1: Introduction protocols for ligands and complexes, structural and physical properties, and applications are discussed, with future opportunities also identified.

The dpp– ligand (the focus of this PhD research and labelled B– in this Chapter) is discussed in section 2 of the following publication, which includes a short summary at its conclusion. 43 ligand precursors and 195 M-dpp complexes are described. Other pyridine- pyrrolide pincers, are included in this report (Sections 3 and 4), and metallo-supramolecular constructs with 2,5-di(4-pyridyl)pyrrolide ligands are also discussed in Section 5. While these sections (3–5) do not directly relate to the compounds prepared and studied during this PhD research, they do inform the conclusions and opportunities for future development, reported in

Section 6.

A complementary review, to which JNM also contributed (along with BD and SBC), describes the syntheses and metal complexes of pyridine-pyrrolide chains and macrocycles.

Das, B.; McPherson, J. N.; Colbran, S. B. Oligomers and macrocycles with

[m]pyridine[n]pyrrole (m + n ≥ 3) domains: Formation and applications of anion, guest molecule and metal ion complexes, Coord. Chem. Rev. 2018, 363, 29.

James McPherson – June 2019 2 Chapter 1: Introduction

Coordination Chemistry Reviews 375 (2018) 285–332

Contents lists available at ScienceDirect

Coordination Chemistry Reviews

journal homepage: www.elsevier.com/locate/ccr

Review Tridentate pyridine–pyrrolide chelate ligands: An under-appreciated ligand set with an immensely promising coordination chemistry ⇑ James N. McPherson, Biswanath Das, Stephen B. Colbran

School of Chemistry, University of New South Wales, NSW 2052, Australia article info abstract

Article history: This review covers all aspects of the chemistry of metal complexes of tridentate j3-N pyridine–pyrrolide Received 8 November 2017 ligands, from the syntheses of the ligands to what is known about the metal complexes – their structural Received in revised form 8 January 2018 and physical properties, and their reactivities, including use in catalytic processes. Applications of the Accepted 9 January 2018 complexes range from the switching elements in molecular devices, to luminophores in electrolumines- Available online 28 March 2018 cent devices and photosensitisers in light-driven catalyses, to anticancer therapeutics and to efﬁcient catalysts for organic transformations. The general properties of the ligands and metal complexes are Keywords: deduced from the available literature, and areas of j3-N pyridine–pyrrolide metal complex chemistry j3-N pincer ligand ripe for development are pinpointed. Pyridine–pyrrolide metal complexes Ó X-ray crystal structures 2018 Elsevier B.V. All rights reserved. Electronic properties Reactions Applications

Contents

1. Introduction ...... 286 1.1. An obvious analogy between 2,20:60,200-terpyridine and tridentate pyridylpyrrolide anions ...... 286 1.2. Two obvious differences between 2,20:60,200-terpyridine and a meridional tridentate pyridine–pyrrolide ligand...... 287 1.2.1. Electronics...... 287 1.2.2. Ligand bite angles and internal strain...... 287 1.3. Structure of this review ...... 288 1.4. Nomenclature and numbering system for pyridine–pyrrole substances and their metal complexes ...... 288 2. The 2,5-di(2-pyridyl)pyrrolide anion (B) ...... 288 2.1. Syntheses and properties of 2,5-di(2-pyridyl)pyrroles ...... 288 2.1.1. Paal–Knorr pyrrole condensations...... 289 2.1.2. Historical: Hein’s pioneering synthesis of 2,5-di-(2-pyridyl)pyrroles ...... 289 2.1.3. Other routes to Paal–Knorr pyrrole condensation precursors ...... 290 2.1.4. The Stetter reaction ...... 290 2.1.5. Owsley’s 1,4-di-(2-pyridyl)butan-1,4-dione synthesis ...... 290 2.1.6. Direct one-pot synthesis ...... 290 2.1.7. Reductive ring contraction of pyridazines...... 292 2.1.8. Miscellaneous syntheses ...... 295 2.1.9. Other dipyridylpyrrole compounds ...... 295 2.2. Coordination compounds of 2,5-di(2-pyridyl)pyrrolide anions (B)...... 296 2.2.1. Early studies of metal complexes ...... 296 2.2.2. Other early complexes ...... 296 2.2.3. Modern studies of homoleptic first-row transition metal complexes ...... 297

Abbreviations: CT, charge transfer; CV, cyclic voltammetry; DFT, density functional theory; EL, electroluminescence/electroluminescent; Fc+/0, ferrocenium/ferrocene; Py, pyridine; Pyr, pyrrole/pyrrolide; RMS, root mean squared; SC-XRD, single crystal X-ray diffraction; SCE, saturated calomel electrode. ⇑ Corresponding author. E-mail address: [email protected] (S.B. Colbran). https://doi.org/10.1016/j.ccr.2018.01.012 0010-8545/Ó 2018 Elsevier B.V. All rights reserved.

James McPherson – June 2019 3 Chapter 1: Introduction

286 J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332

2.2.4. Other first-row transition metal complexes ...... 301 2.2.5. Second- and third-row transition metal complexes ...... 304 2.3. Summary ...... 313 3. The 2,6-di(2-pyrrolide)pyridine dianion (C2)...... 313

3.1. Syntheses and properties of 2,6-di(2-pyrrolyl)pyridines (H2C) ...... 313 3.1.1. Fischer indole synthesis ...... 313 3.1.2. Metal-assisted couplings ...... 314 3.1.3. Miscellaneous ...... 317 3.1.4. Other dipyrrole–pyridine compounds ...... 317 3.2. Coordination compounds of 2,6-di(2-pyrrolide)pyridine dianions (C2)...... 317 3.3. Summary ...... 324 4. The 2-(2,20-bipyridin-6-yl)pyrrolide (D) and 5-(2-pyridyl)-2,20-bipyrrolide (E2) anions ...... 324 4.1. Syntheses and properties of 2-([2,20-bipyridin]-6-yl)pyrroles ...... 324 4.2. A coordination compound of a 2-(2,20-bipyridin-6-yl)pyrrolide anion (D)...... 325 5. Pyridine–pyrrolide ligands in metal-directed self-assembly of supramolecular cages and polymeric structures ...... 326 5.1. Syntheses and properties of ligands...... 327 5.2. Coordination compounds ...... 327 6. Conclusion ...... 328 6.1. Exploitable routes to ligand precursors ...... 328 6.2. Rich coordination chemistry...... 328 6.3. Geometrical strain...... 329 6.4. Electronic effects ...... 329 6.5. Applications ...... 329 6.5.1. Molecular electronics ...... 329 6.5.2. Electroluminescence and photoredox catalysis ...... 329 6.5.3. Functional supramolecular constructs...... 329 6.5.4. Medicinal chemistry and sensing ...... 329 6.5.5. Molecular catalysis ...... 329 6.6. Scope ...... 330 Declarations of interest ...... 330 Acknowledgements ...... 330 References ...... 330

1. Introduction 2,20:60,200-Terpyridine (tpy or terpy) is the prototypal meridional metal-binding j3-N-ligand, A in Fig. 1 [17–39]. The coordination We provide a ﬁrst and comprehensive review of tridentate pyr- chemistry of terpyridine is rich and varied, extending as it does idine–pyrrolide ligands with cores B–E2 (Fig. 1), their prepara- throughout the Periodic Table from the s-block metal cations, tions and properties, and their coordination chemistry, with an through the variable oxidation state transition metal ions to the emphasis on the uses or potential uses of their coordination com- p-block metal cations, and on to the lanthanide and actinide ions plexes, all in order to raise awareness of these under-appreciated [17]. Numerous terpyridyl metal complexes have been discovered ligands and complexes. that exhibit diverse and very useful photophysical, redox, magnetic, catalytic and medicinal properties [17–39]. Given the diver- sity and usefulness of simple monometallic terpyridyl complexes, 1.1. An obvious analogy between 2,20:60,200-terpyridine and tridentate it is not surprising that the terpyridyl ligand has been regularly pyridylpyrrolide anions exploited to construct metal-containing supramolecular assemblies or macromolecular polymers, with those containing bis(ter- At the cutting edge of contemporary ligand design are pincer pyridyl)metal photo and/or redox centres being of particular ligands. Innovation in pincer ligands has provided a veritable cor- interest [21,22,24,28,31,33,40,41]. The reader is referred to the nucopia of novel metal complexes, which have bolstered many many excellent and comprehensive reviews for further informa- advances, for example, in molecular sensing [1–4], as potential tion on this now voluminous body of ever-expanding research therapeutic agents [1,2,5–7] and, most of all, in catalysis [1,2,8– [17–39]. 12] including in cutting-edge molecular devices for (light-driven) The focus of this review is ligands formed by replacement of one energy conversion and storage [1,2,13–16]. The term ‘‘pincer or more the pyridine rings in terpyridine (A) by the pyrrolide anion ligand” is of course just an affectatious modernism; it is an to afford three-ring, pyridine-pyrrolide ligands B–E2.In avant-garde ’nom de guerre’ for the coordination chemists’ com- comparison with terpyridine, the coordination chemistry of the monplace meridional ligand. pyridine–pyrrolide ligands B–E2 is sparse. Why: is there a reason

N N N N N N N N N N N N N N N

ABCDE

Fig. 1. Drawings of the three ring, 18 p-electron, meridional ligands with pyridine and pyrrolide donors; namely commonplace 2,20:60,200-terpyridine (A), 2,5-di(2-pyridyl) pyrrolide anion (B), 2,6-di-(2-pyrrolide)pyridine dianion (C), and the asymmetric ligands, 2-(2,20-bipyridin-6-yl)pyrrolide anion (D) and 5-(2-pyridyl)-2,20-bipyrrolide dianion (E) (not shown is the 2,20:50,200-terpyrrolide trianion as a genuine metal complex of this is yet to be reported).

James McPherson – June 2019 4 Chapter 1: Introduction

J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332 287

Scheme 1. p-Conjugation in the dipyridylpyrrolide anion.

.A .B the LDPs for pyrrolide and fluoride are similar (13.6 and 13.4 1 120° 126° kcal mol respectively) [51]. Moreover, provided the pyridyl–pyr- 120° 108° rolide ligand is close to planar allowing electronic delocalisation, the adjacent pyridine rings should exhibit some amide character 120° 126° due to p-conjugation with the pyrrolide ring (e.g., as illustrated Fig. 2. A. Overlay of the ligand skeletons with the nitrogen atoms (drawn as dots) of for B in Scheme 1) [45–47]. Although pyridine donors are often the central rings eclipsed for terpyridine (A in red), dipyridylpyrrolide anion (B in redox non-innocent, potential electron-acceptor sites in their blue), and dipyrrolidepyridine dianion (C2 in green). B. Idealised interior and metal complexes,1 such behaviour is less likely for the flanking pyr- exterior bond angles about the ipso-atoms of linked six- and five-membered rings. idine donors in the a pyridyl–pyrrolide ligand because of their amide character. Therefore, overall, a pyridyl–pyrrolide ligand should be capable of stabilising its higher oxidation state complexes by p- for so little study of B–E2 as ligands, or are these ligands that are donation and also of increasing the reducing power of its lower oxida- genuinely under utilised? Part of the motivation for writing this tion state complexes. Also, the pyrrolide anion is difficult to reduce, review is the authors’ conviction that these are ligands with con- but relatively easy to oxidise. Therefore, in metal complexes of tri- siderable promise, that they afford metal complexes with valuable dentate pyridine–pyrrolide ligands, it is anticipated that reduction utility, and that these ligands have been somewhat neglected for of the pyridine–pyrrolide ligand will be more difficult than for the historical reasons, most simply that the ligands are at present easily reduced terpyridine ligand, but it is possible that new not immediately and visibly available. ligand-centred oxidation processes will become feasible. Replacing the central ring of terpyridine by a pyrrole ring and then deprotonating it affords the 2,5-di(2-pyridyl)pyrrolide anion (B), which is the most studied of the meridional tridentate pyri- 1.2.2. Ligand bite angles and internal strain dine–pyrrolide anions. A simple viewing of B suggests an obvious Fig. 2 serves to emphasize that replacement of one or two 6- 0 0 00 analogy with terpyridine. Both have three 6p-electron N- membered pyridine rings in 2,2 :6 ,2 -terpyridine (A)by5- heterocyclic rings and, therefore, it is anticipated that the B anion membered pyrrolide ring(s) to afford one of the tridentate pyri- 2 3 should also act as a meridional metal-binding j3-N-ligand. And, as dine–pyrrolide ligands B –E must increase strain in any j -N we shall see, so it does. It is an anionic analogue of the terpyridine metal complex that these form. The strain introduced by pyrrolide ligand (A), Fig. 1. Where, for example, terpyridine forms homolep- for pyridine substitution is least for an outer ring to afford ligands 2+ C2 and D, and far greater upon substitution of the inner ring to tic [M(tpy)2] complex salts with divalent metal cations, B forms, 2 or at least should form—analogues are often not yet known (Sec- afford ligands B and E . This is because the two ideal exterior bond angles from the 5-membered ring to a second ring are 126° tion 2.2 below) — neutral homoleptic [M(B)2] complexes. Apart from increasing the visibility of pyridine–pyrrolide compared to 120° from a 6-membered ring to a second ring, ligands B–E2, the other purpose of this review is to tease out Fig. 2B. To efficiently bind a metal ion, therefore, it is anticipated exactly how they differ in ligand properties from the terpyridine that the pyridine–pyrrolide ligands will exhibit ’pinched in’, thus ligand and to highlight how these differences impact on their less strained, geometries. Contrariwise, the escalating strain within understood coordination chemistries. Yes, the introduction of a the pyridine–pyrrolide ligand as it binds to a metal ion will pyrrolide donor affords a ligand that is different. Novel coordina- decrease the strength of binding. Overall, the internal strain within 3 tion chemistry results, leading to metal complexes for which a j -N bound pyridine–pyrrolide ligand coupled with the p-donor terpyridyl-metal analogues are unknown or to metal complexes character of pyrrolide groups implies these should be weak field with vastly different character and properties to the better- ligands. Complexes exhibiting high spin states are therefore antic- known terpyridyl-metal analogues. New opportunities result. ipated, especially for the first row transition metal ions. First, however, before discussing the specific chemistry of tri- The above conclusions are further confirmed by simple mod- ⁄⁄ z+ dentate pyridine–pyrrolide ligands, we begin with a first princi- elling (B3LYP/6-31G ) of hypothetical gas-phase [Zn(ligand)] 2 10 ples, back-of-the-envelope, hand-waving, examination of what species, where the ligand is one of A–E and the d zinc(II) ion effects the introduction of a pyrrolide donor might be expected was chosen without co-ligands in order to minimise electronic to have. and steric influences on the resultant structures, Table 1 [59]. The unbound ligand with only the central nitrogen donor (N2) protonated was also modelled. In this case, intramolecular N...H–N2 1.2. Two obvious differences between 2,20:60,200-terpyridine and a interactions direct the ligand to mimic its conformation when meridional tridentate pyridine–pyrrolide ligand coordinated with a metal ion. For all ligands, significant strain builds upon coordination to the Zn(II) ion as indicated by the 1.2.1. Electronics decrease in the distance between the two outer nitrogen donor Consider a tridentate, meridional ligand comprised of linked atoms (N1...N3), and the decrease of the ligand ‘‘bend angle”, pyridine and pyrrolide rings; that is any one of the anionic ligands which is defined as the angle between the two inter-ring CAC B–E2 in Fig. 1. The relative pK (dmso) values of the (unsubsti- a bonds. The largest deformations are predicted for the two ligands tuted, N-protonated) pyridinium and pyrrole rings are 3.4 and 23.0, respectively [42]. The electron rich, anionic and 6p-electron 1 aromatic pyrrolide ring should certainly be a good r- and Bipyridine and terpyridine are prototypal examples of redox non-innocent, p r electron-acceptor ligands; a common misconception, however, is that pyridine (py) -donor to any metal ion it binds [43–51]. The combined - and p p donors are good -acceptor ligands and, therefore, high in the spectrochemical series. -donor ability of a ligand to high valent metal centres is This is not true: pyridine is a weak p-acceptor ligand and lies toward the middle of quantified by Odom’s ligand donor parameter (LDP), and indeed the spectrochemical series between H2O and NH3 ligands [52–58]. James McPherson – June 2019 5 Chapter 1: Introduction

288 J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332

Table 1 Theoretical (B3LYP/6-31G**) optimised geometries of ligands A–E and their mono ligated zinc(II) complexes [59].

Central ring Ligand-H [Zn(ligand)]z+ %Deformation A: Bend angle (°) Pyridyl 105.1 98.2 6.6 RMS deviation (Å) 0 0 N1...N3 (Å) 4.128 3.936 4.7 B: Bend angle (°) Pyrrolyl 129.7 105.6 19 RMS deviation (Å) 0 0.225 N1...N3 (Å) 4.918 3.980 19 C2: Bend angle (°) Pyridyl 108.3 93.9 13 RMS deviation (Å) 0 0 N1...N3 (Å) 4.462 3.782 15 D: Bend angle (°) Pyridyl 107.5 96.3 10.4 RMS deviation (Å) 0 0 N1...N3 (Å) 4.312 3.857 10.6 E2: Bend angle (°) Pyrrolyl 131.2 102.4 22 RMS deviation (Å) 0 0.255 N1...N3 (Å) 5.073 3.873 24

(a) RMS deviation is the root mean squared deviation for all atoms in the ligand from the plane deﬁned by all atoms in the ligand. (b) %Deformation is a measure of the effect of metal binding on the ligand deﬁned as: deformation ¼fjpropertyjligand jpropertyjcomplex g=jpropertyjligand ð100=1Þ%:

counts as core C). The number of acidic protons is given before this

N descriptor (e.g.,H2C or HD) and the charge as a superscript after it H N O O N N N (e.g., C2 or D), as appropriate. The speciﬁc organic systems are sequentially numbered using a superscript in bold immediately HB Q 1 9 2 following the core designator {e.g.,H2C or (C ) }. Metal complexes of these ligands are sequentially numbered in bold (ie. 1, Scheme 2. Retrosynthetic analysis—dipyridylpyrroles HB can be accessed from 2, 3, 4...). Hopefully, no confusion will arise for the reader. dipyridylbutan-1,4-diones (Q).

2. The 2,5-di(2-pyridyl)pyrrolide anion (B)

2 with five-membered pyrrolide cores (B and E )—approximately At the date of submission, a search of the CAS SciFinderTM data- 20% and 25%, respectively. Coordination is also predicted to result base for the B core (charge and protonation unspecified), excluding in puckering of rings in the ligand from coplanarity as indicated by all fused, oligomeric, macrocyclic or N-derivatized systems, returns the root mean squared deviations of the ring carbon and nitrogen 250 compounds described in 50 publications (including correc- + atoms from the mean ligand plane: 0.225 Å for [Zn(B)] and tions and conference papers, which are not cited in this review). 0.255 Å for [Zn(E)], while the free ligands and other zinc complexes Narrowing the search to only include compounds containing met- all remain coplanar (RMS deviation = 0). The puckering presumably als returned 195 compounds (86 of which are hypothetical rhe- + 2 relieves excess strain in species [Zn(B)] and [Zn(E)]. Ligands C nium(I) complexes proposed in a patent, see Section 2.2) across and D with the six-membered pyridyl cores remain coplanar 32 publications. and show significantly smaller deformations, 15% and <10%, respectively, upon coordination to the Zn(II) ion. N N N 1.3. Structure of this review

The review is divided into sections according to each unique B ligand core; so B–D are discussed in Sections 2–4, respectively. Lar- ger, linked pyridyl-pyrrole systems, be they linear or cyclic, will be reviewed elsewhere [60] and are therefore excluded. Section 5 provides description of the tridentate pyridine–pyrrole ligand family Hein prepared the first examples of HB substances, and in the in assembly of metallo-supramolecular constructs. Each of these same work, demonstrated their deprotonation and use as anionic section opens with a thorough review of the ligands and the syn- chelating ligands [61]. For fifty years, the coordination chemistry thetic strategies to them, before the coordination chemistry of each of these ligands lay dormant, until Tour and co-workers explored particular class of ligand is comprehensively reviewed. The final their use in molecular devices in the mid 2000s [62]. In the mean- closing section of the review highlights some key features of the time, organic chemistry protocols have progressed dramatically, coordination chemistry and identifies areas for development. and there are now a wide range of HB derivatives available.

1.4. Nomenclature and numbering system for pyridine–pyrrole 2.1. Syntheses and properties of 2,5-di(2-pyridyl)pyrroles substances and their metal complexes This section summarises the different synthetic strategies Throughout this review, we use the descriptors A–E for the which have been employed to prepare HB derivatives since Hein’s ligand cores depicted in Fig. 1 irrespective of substitution and/or original reports [61,63]. The syntheses of 39 different derivatives further ring annulation (so, for example, a 2,6-di(indolyl)pyridine are described. James McPherson – June 2019 6 Chapter 1: Introduction

J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332 289

KMnO4

N 65 % HOOC N

50 %

COOEt COOEt EtO O I2 CH3CO2Et

NaHC NaOEt EtO 65 % N N N O O N 75 % O O

O NH3 O O O O EtOOC COOEt HOOC COOH O N H+ N H N 4 N N -2CO HB H N H N N N 91 % 2 2 3 HB H3B N N H N HB1

Scheme 3. Hein and co-workers’ classic syntheses of dipyridylpyrrole (HB1) and precursors [61,63]. Reported yields are shown, except for quantitative transformations, with an overall yield for HB2 of 16%, or 14% for HB1.

N+

- CHO HO Cl O S + + NaOAc a) N S EtOH N O O N O 17 %

N+

CHO O HO Cl- S b) ++S NEt3 N O 1,4-dioxane N O O N

28 %

N+

CHO N HO Br- S c) + N + NEt3 N 1,4-dioxane N O O N O 36 %

NH4OAc d) N N O O N 90 % N H N Q HB1

Scheme 4. Access to parent, unsubstituted HB1 through two steps using the Stetter reaction reported by Jones et al. (highest yield: 32% overall) [68].

2.1.1. Paal–Knorr pyrrole condensations 2.1.2. Historical: Hein’s pioneering synthesis of 2,5-di-(2-pyridyl) The Paal–Knorr pyrrole condensation is one of the classical pyr- pyrroles role syntheses and is condensation of a 1,4-diketone (butadione) The classic dipyridylpyrrole synthesis and chemistry was ﬁrst with an amine to afford the corresponding pyrrole [64,65]. The reported by Hein and Melichar in 1954 [61], with some improve- use of ammonia or NH4OAc, as originally reported by Knorr, affords ments reported in 1957 [63]. Scheme 3 summarises the organic the N-unsubstituted pyrrole, and thus HB derivatives can be pre- synthesis. Pinner’s dipyridoyl succinate ester [67] was identiﬁed pared from the appropriate dipyridylbuta-1,4-diones (Q) as an entry point to the desired dipyridylpyrrolide ligand. By per- (Scheme 2) [66]. forming the oxidative coupling of the sodium pyridoyl acetoester

James McPherson – June 2019 7 Chapter 1: Introduction

290 J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332

Subsequently, Wayland, Zdilla and co-workers improved the yield of butadione Q to 53% through optimisation of the stoichiometric ratios in route (a) [69], and confirmed the structure of the parent dipyridylpyrrole HB1 by single crystal X-ray diffraction (SC-XRD), Fig. 3. Notably, pyrrole N2-H...N(pyridyl) lone pair interactions stabilise a planar structure with all rings facing inward. Xiao-Yi Yi and co-workers have also employed the Stetter reac- HB1 tion to prepare di(5-bromopyridyl)pyrroles, from 5-bromopyridyl- 2-carboxaldehyde, divinyl sulfone and NaOAc with thiazolium Fig. 3. SC-XRD structure of 2,5-di(2-pyridyl)pyrrole (HB1) reported by Wayland, catalysis, Scheme 5 [70]. The target butan-1,4-dione precipitated Zdilla and co-workers (CCDC reference code: QOBREF01) [69]. Key intramolecular on cooling the reaction mixture, and was condensed with NH4OAc interactions are shown in green, and selected interatomic distances (Å) and angles without further purification to afford the di(5-bromo-2-pyridyl) (°): N1...H2 2.571, N10...H2 2.467, and N1...H2...N10 165.42. pyrrole, HB5, in 44% overall yield. Finally, Liu and co-workers reported dual thiazolium/palladium with iodine in 1,4-dioxane, the decomposition of the solvent sys- (0) catalysis for the Stetter reaction gives much improved yields for tem that Pinner reported was avoided and the yield was improved a diverse range of substituted 1,4-di-(2-pyridyl)-butan-1,4-diones, sufficiently (to 65%) to be practical. The final step, Paal–Knorr pyr- but described the ring closure of only one example with NH4OAc 6 role condensation of Pinner’s dipyridoyl succinate ester afforded which afforded dipyridylpyrrole HB in 76% overall yield, Scheme 6 the symmetric 3,4-diethyl dipyridylpyrrole HB2 in quantitative [71]. 2 3 yield. Dipyridylpyrrole HB was transformed to the di-acid H3B , anhydride HB4 and, upon decarbonylation, parent dipyridylpyrrole 2.1.5. Owsley’s 1,4-di-(2-pyridyl)butan-1,4-dione synthesis HB1. The other prominent route exploited to dipyridylbutan-1,4- Although Hein’s synthesis requires several steps culminating in dione precursors has been Owley’s treatment of N,N,N,N- modest overall yields for HB1–HB4, none of the steps are particu- tetramethylsuccinamide with a lithio-pyridine, Scheme 7 top larly challenging and tens of grams of functionalised dipyridylpyr- [72]. Although Owsley never performed the subsequent condensa- roles can be accessed within a week by a trained chemist. While tion to afford the pyrrole, several other groups have since exploited many subsequent papers have acknowledged her pioneering work, and optimised this route. By slowly adding n-butyl lithium (over 2 most have focussed on accessing the unsubstituted HB1, and few h) to the reaction mixture at 100 °C, Tour’s group nearly tripled provide more convenient access to functionalised ligands ready the yield of Q to 57%, Scheme 7 bottom [62]. Tour’s group also for further modification. functionalised the 3- and 3,4-pyrrole positions of parent HB1 to afford the thiocyanate-derivatised HB7 and HB8 for immobilisation 2.1.3. Other routes to Paal–Knorr pyrrole condensation precursors of metal complexes onto gold surfaces (see Section 2.2). The struc- 8 Recently, several groups have also identified dipyridylbuta-1,4- ture of HB was determined from SC-XRD data, Scheme 7 (inset), diones as key synthetic precursors for the Paal–Knorr pyrrole con- showing the same inward facing ring structure as found for the 1 densation with an ammonium salt to afford the corresponding parent HB (see Fig. 3). dipyridylpyrrole derivative. The Stetter reaction and Owlsey’s syn- Finally, the Sessler and Xiao-Yi groups employed Owsley’s route 9 thesis were the two most commonly employed routes to the to access the 6-bromopyridyl derivative HB , Scheme 8, which is an dipyridylbutadione intermediates. excellent synthon for further transformations to macrocyclic ligands [73], or for installing other functions such as 2-thienyl or 2.1.4. The Stetter reaction phenyl groups via Suzuki coupling with the appropriate boronic The Stetter reaction is the thiazolium catalysed addition of an acid, Scheme 8 [70]. aldehyde to an appropriately activated Michael acceptor. Jones et al. reported [68] three different routes to dipyridylbutadiones, 2.1.6. Direct one-pot synthesis which included 2-pyridyl, 3-pyridyl and 4-pyridyl derivatives, by In 2014, the Colbran group reported a convenient, one-pot employing the umpolung activation of pyridyl carboxaldehydes direct condensation route to a range of dipyridylpyrroles [74]. They with a thiazolium catalyst and using divinyl sulfone (Scheme found that condensation of two equivalents of a 2- or 4-pyridine 4a,b) or a Mannich base (Scheme 4c) as the Michael acceptors. carboxaldehyde and one equivalent of a methylene ketone in the

Straightforward ring closure with NH4OAc afforded the corre- presence of NH4OAc afforded the dipyridylpyrrole in moderate sponding dipyridylpyrrole (HB1) in 15–32% yield (Scheme 4d). yield, which could be recovered by cooling the reaction mixture

Br Br N+ N CHO - HO Cl NH OAc N O S 4 + + NaOAc O N S NH EtOH O EtOH O Br 78 % 78 % HB5 N N

Br Br

Scheme 5. Two-step preparation of di(5-bromo-2-pyridyl)pyrrole (HB5) using the Stetter reaction [70].

James McPherson – June 2019 8 Chapter 1: Introduction

J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332 291

S I- N+ OH R O N CHO Pd(PPh3)4 OAc + N DMSO, NEt3 N O R R

OMe

O N O N O N

N O N O N O

83 % 80 % OMe 61 %

S Br Ph

O N O N O N

N O N O N O

Br 77 % Ph 84 % 79 % S

Br Ph O N O N O N

N O N O N O Br Ph 23 % 76 % 71 % OH

O N O N O N

N O N O N O

78 % 68 % 72 %

N O N NH OAc 4 N HN HOAc N O 92 % HB6

Scheme 6. Improved access to 2-methyl-1,4-di-(2-pyridyl)-butan-1,4-dione derivatives using palladium/NHC catalysis [71].

and ﬁltration or by chromatography. The syntheses enable con- molecule (O1W), which acted as a hydrogen bond donor to the struction of diverse substituted dipyridylpyrroles directly from phenanthrolyl N atoms (N2, N3, N4 and N5), and a hydrogen bond commercially available ketones, and also accommodate various acceptor for the pyrrolic NAH atom (N1), Fig. 4 [74]. substituted pyridyl precursors (e.g., including quinoline and Recently, Fuchs and Zeitler have reported another one-pot phenanthroline donors), Scheme 9. Although 4-pyridine carbox- route to polysubstituted pyrroles, including HB22 and HB23, from aldehyde successfully formed di-(4-pyridyl)pyrroles, the reaction commercially available starting materials as described in did not proceed with 3-pyridine carboxaldehydes, which points Scheme 10 [75]. The nitroalkene serves as a latent 1,2-bis to the pyridyl group playing an electronic role in the condensation (electrophile), and sequential nitro-Stetter and Stetter reactions reactions. More recent work reveals that this route also provides (see above) afford the di(2-pyridyl)-1,4-butandione in the pres- direct access to dipyridylpyrroles with two electron rich pyrrole ence of an NHC catalyst and potassium carbonate. Subsequent substituents, e.g., HB20, albeit in lower yield. addition of an amine (ammonia for 1-H pyrroles) in acetic acid Interestingly, the SC-XRD structure of HB18 was determined and affords the substituted dipyridylpyrroles without the need to iso- revealed the compound to be a strong host for water guest late the 1,4-diketone Q.

James McPherson – June 2019 9 Chapter 1: Introduction

292 J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332

O Br X = H, 20 % n-BuLi N + + N X=Br,71% N THF N O O N O X X X

Li O O O N N OH DMF, P2O5 N HO N O reflux O N O 57 % 89 %

83 % NH4OAc

NCS (SCN)2 N N NH NH 9:8 CH Cl :THF 7 2 2 1 N HB N HB 62 %

(SCN)2 88 % 2:5 CH2Cl2:THF

SCN NCS N NH 8 N HB HB8

Scheme 7. Top: Owsley’s original synthesis of 1,4-di-(6-bromo-2-pyridyl)butan-1,4-dione [72]. Bottom: Optimised synthesis by Tour et al. of HB8 in 42% overall yield and subsequent derivatisation with thiocyanate groups [62]. Inset, SC-XRD structure of HB8 (CCDC reference code: UCIHAQ) [62]. Key intramolecular interactions are shown in green. Selected interatomic distances (Å) and angles (°): N52...H1 2.373, N22...H1 2.372, and N52...H1...N22 161.65.

O N Br O n-BuLi N N N + N N H N THF N O O Br 9 Br HB Br Br

HO OH S B B HO OH

N K2CO3 N K2CO3 N N H N N H N N H N Pd(PPh ) Pd(PPh ) 3 4 Br Br 3 4 82 % 79 % S S HB10 HB9 HB11

Scheme 8. Owsley’s route to access 2,5-di(6-bromo-2-pyridyl)pyrrole (HB9) [72], and subsequent post synthetic transformations of HB9 into phenyl (HB10) and thienyl (HB11) derivatives [70].

2.1.7. Reductive ring contraction of pyridazines dinitrogen, and electrochemical reductive ring contraction of the Dubreuil’s group developed an alternative route to pyridyl- pyridine-substituted pyridazine intermediates afforded the pyr- pyrrole moieties, through construction of pyridyl-pyridazine pre- roles HB1 and HB24–HB30 in good to excellent yields. The synthesis cursors, using well established [4+2] cycloaddition of a dienophile also worked well for di(4-pyridyl)pyrrole derivatives and tolerated (in this case a terminal alkyne) and the appropriate r-tetrazine, a range of substituents, which were accessed from tetrazine in Scheme 11 [76–79]. Subsequent retro-Diels–Alder elimination of good yields using an appropriate dienophile via the Diels-Alder

James McPherson – June 2019 10 Chapter 1: Introduction

J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332 293

R1 R2 R1 R2

H O H NH4OAc + N EtOH H N O O N N N

Me COOMe EtOOC COOEt Me C(O)Me

N N N N H N N H N N H N HB12 HB2 HB13 %64 %22 37 %

COOEt CN

N N N H N N N H N N H N HB14 HB15 HB16 %22 %13 %94

Me COOMe N Me COOMe COOMe N N H N N N H N N N N N H N HB17 HB18 HB19 58 % 46 % 22 %

Compounds accessed post synthetically: I- N+

Me Me COOMe

N N N H N N H N HB20 HB21 17 % 50 %

Scheme 9. A direct, one-pot, catalyst-free, condensation route to dipyridylpyrroles (overall yields are relative to the starting ketone) [74].

The SC-XRD structures of 2,5-di(4-pyridyl)pyrrole [80,81] and HB1 (CCDC reference code: QOBREF) [80] were determined for the first time by Dubreuil and co-workers. The former structure is interestingly held together by a network of hydrogen bonds between lattice water and di(4-pyridyl)pyrrole molecules, Fig. 5. The SC-XRD structures reported of HB1 by Dubreuil and co- workers and that described four years later by Wayland, Zdilla and co-workers (see text surrounding Fig. 3) are polymorphs: Dubreuil’s group crystallised HB1 in the orthorhombic Pbcn space group with 8 molecules in the unit cell [80], whereas Zdilla et al. 1 crystallised HB in the orthorhombic space group P212121 with 4 Fig. 4. SC-XRD structure of HB18 (CCDC reference code: LOGKID) [74]. Key molecules in the unit cell [69]. 31 intramolecular interatomic separations (Å): N1...O1W 2.852(4), N2...O1W 3.019 The Dubreuil group also prepared a diacid derivative (HB ) fol- (4), N3...O1W 2.822(4), N4...O1W 2.816(5), N5...O1W 2.973(4), O1W...O1ET 2.685 lowing a similar protocol, however it could not be prepared (8), O1W...O1E0 (not shown) 2.61(2). directly from the 6,60-(pyridazine-3,6-diyl)-bis-2-pyridylcar boxylic acid due to solubility issues [82]. Instead, solubilizing n- reaction, Scheme 11. However, electron-withdrawing groups octyl ester groups were introduced (via the diacid chloride and (R = CO2Et) consumed significantly more than the 4e per octanol) and electrochemical reduction at 1.15 V vs. SCE afforded mole of pyridazine, due competing formation of a 1,4- octyl 6,60-(pyrrol-2,5-diyl)-bis-2-pyridinecarboxylate (HB32), dihydropyridazine intermediate. Scheme 12. Saponification of HB32 with KOH in methanol resulted James McPherson – June 2019 11 Chapter 1: Introduction

294 J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332

F F N F N+ N

F F - BF4 1. NHC catalyst (1 mol %) K2CO3 (100 mol %) room temperature for 16 h O N O NO 2. 2-py-CHO (1 equiv.) 3. NH (3 equiv.) in AcOH H N 2 N O 3 N N + 50 °C for 7 h. 70 °C, 36 h R R rehtelyhteid rehtelyhteid R N

Q HB22 and HB23

Scheme 10. Zeitler’s one-pot route to substituted dipyridylpyrroles HB22 (R = Ph), and HB23 (R = furyl) with overall yields of 98% and 70% respectively [75].

[4 + 2] R R Py Py Py Py N N N H ·xH O NaNO R +4e- +4H+ 2 4 2 N 2 N Py C N NH N N NH AcOH (aq) -N N N 2 N Py H Py Py Py

N N N N N H N N H N N H N N H N HB1 HB24 HB25 HB26 82 % 87 % 92 % 68 % -1.05 V, 4.6 h -1.08 V, 6 h -1.10 V, 5.3 h -1.18 V, 5.2 h

O O O O

N N N H N H N N N N H N N N H N

HB27 HB28 HB29 HB30 53 % 77 % 72 % 37 % -1.00 V, 3.2 h -1.05 V, 3.2 h -1.05 V, 4.2 h -1.05 V, 9 h

Scheme 11. The route of Dubreuil et al. to HB1 and new derivatives HB24–HB30 with electrochemical reduction of 3,6-di(2-pyridyl)pyridazine precursors as the key ﬁnal step; listed are the reported cathode potentials relative to SCE for the electrolyses and overall yields across all four steps [80,81].

31 0 in H3B in 38% overall yield from the original 6,6 -(pyridazine-3,6- diyl)-bis-2-pyridylcarboxylic acid. Huc and Dubreuil had previously shown that derivatives of di (pyridyl)pyridazines with helical conformations (such as the synthetic receptor R1 shown in Scheme 13b) were excellent hosts for 6 1 D/L-tartaric acid (Ka >10 L mol ), and experiments with pure L- or D-tartaric acid conﬁrmed each is bound by the M or P helix, respectively [83]. In the oligomeric sequence of R1, the quinoline trimers (Q in Scheme 13) act as caps to close the cavity, while the larger units of the sequence, the pyridines (P), napthyridines (N) and the di-pyridylpyridazine create a large internal cavity that accommodates the tartaric acid guest. Dubreuil and Huc proposed Fig. 5. View from the SC-XRD structure of di(4-pyridyl)pyrrole hydrate reported by that ring contraction of the pyridazine to the pyrrole would result Dubreuil and co-workers (CCDC reference code QOBRAB) [80]. Three crystallo- in major changes to the curvature of the helix backbone, reducing graphically equivalent di(4-pyridyl)pyrrole molecules form three H-bonds with each lattice water molecule shown in green; key interatomic separations (Å): N1...O the afﬁnity of the host, and triggering release of the guest [82]. 33 31 2.824(1), N9...O 2.831(1), N15...O 2.801(1). HB was therefore prepared from H3B as shown in Scheme James McPherson – June 2019 12 Chapter 1: Introduction

J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332 295

however chemical reduction with zinc in acetic acid (Scheme 13b) afforded HB33 in 50% unoptimized yield. Despite this success, N CO H 2 N the conditions required to ensure ring contraction are incompati- N NH CO2H N ble with guest binding, and so improvements are required to 31 N N HB ensure host–guest chemistry and ring contraction can occur in the same medium [82]. CO2H CO2H The Dubreuil group have also constructed larger alternating 1. SOCl2 2. octanol, NEt 3 KOH, MeOH, H2O pyridyl-pyrrolyl molecular strands using the same synthetic 88 % 85 % methodology [84].

2.1.8. Miscellaneous syntheses 1 N CO2Oct Nair and Kim accessed the parent unsubstituted HB in 12% N N -1.15 V vs SCE NH CO Oct yield in two steps from vinyl pyridine, Scheme 14; however the N 2 majority of the reaction product was reported as intractable mate- N 32 51 % N HB rials thus presenting problems of waste and for purification [85]. CO Oct In the mid-to-late 1990s, Lehuede et al. prepared a number of 2 CO Oct 2 tetraarylpyrroles including a di(2-pyridyl)pyrrole with two pyra- 34 31 zine substituents (HB ) and the di(4-pyridyl)pyrrole analogue Scheme 12. Preparation of H3B via installation of solubilizing n-octyl ester groups in 38% overall yield by reductive ring contraction as reported by Lautrette [86]. As shown in Scheme 15a, methylazine was lithiated with et al. [82]. lithium diisopropylamine, before condensation with aromatic nitriles (in this case 2-pyridine carbonitrile) to afford the moisture sensitive imine-enamine. Oxidative coupling of the imine-enamine a) by lead(IV) acetate afforded tetraarylpyrroles, such as HB34 1. H-X N N X obtained in 29% overall yield. The tetraarylpyrroles were screened CO H 2. DIPEA, NH 2 PyBOP NH as antioxidants, however they displayed very little activity [87]. CHCl , 45 °C, 24 h O N 3 N The coordination chemistry of these compounds has not yet been 44 % 31 33 investigated, but from first observation, they seem good candidates H3B HB CO2H O as back-to-back, ditopic ligands. Di-quinolyl and di-quinolyl isoin- X doles (HB35 and HB36) [88] and pyrroles (HB37 and HB38) [89] have b) X also been prepared as shown in Scheme 15(b,c), and again remain N to be investigated as ligands. N O Zn/AcOH N X Recently, Chen and coworkers have reported a high yield, single N NH step route to a di(2-pyridyl)pyrrole HB39, which incorporates a G- N reflux O 1 N quadruplex binding substituent off the pyrrolide core, Scheme 16 R 50 % 33 39 X O HB [90]. Although HB was well-characterised, including by a SC- O XRD study (inset to Scheme 16), there was no discussion of sub- X strate scope or reaction mechanism for this strange synthesis, OiBu OiBu and no other precedents are mentioned. If applicable to a broad range of substrates, this route could provide a most convenient 39 X = H H route to new ligands. The binding properties of HB to telomerase N N N N N N N were investigated through molecular docking predicting a 9.76 H O O NHAc kcal/mol binding energy to human telomerase RNA, and inhibited 2 3 telomerase activity in human tumour SK-OV-3 cell models (IC50 N P Q = 39.5 lM) [91]. The medicinal chemistry of a homoleptic, cobalt (III) complex of HB39 is summarised in Section 2.2. Scheme 13. Alternate preparations of capsule HB33 reported by Lautrette et al. [82]. 31 First, under standard peptide coupling conditions from the diacid H3B , and second via chemical reductive ring contraction of capsule R1. 2.1.9. Other dipyridylpyrrole compounds Fig. 6 depicts ligand precursors which have not yet been reported, despite metal complexes of their deprotonated ions being 13a, and its affinity towards tartaric acid was determined to be at known. Ligands (B40) and (B41) were prepared directly around 1 least two orders of magnitude lower (Ka = 16000 L mol ) than a iron(II) [92] and ruthenium(II) [93] centres, respectively, and are 1 6 1 derivative of R (Ka >10 L mol see above) in a 99:1 mixture of discussed in Section 2.2. While five platinum(II) complexes were deuterated chloroform and dimethyl sulfoxide [82]. Attempts to claimed by Sotoyama and co-workers for each (B42) and (B43), 1 33 electrochemically convert R to HB in situ were unsuccessful, no preparative details were reported for either ligand [94].

Br N N N NN N N Br N NaN H 2 3 N +

CCl4 Br DMSO HB1 %3.21 %9.2

Scheme 14. The 1976 two-step route to HB1 by Nair and Kim [85].

James McPherson – June 2019 13 Chapter 1: Introduction

296 J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332

N H N N N N N N NH NH2 N Li N CN Li Pb(OAc)4 a) N N N N THF THF MeCN N N N N 38 % N N 77 % HB34

O O

Cl Cl

b) N H N N quinoline isoquinoline N N H N O N %11 H %71 HB35 HB36

quinoline isoquinoline AcCl AcCl N N N H N H c) 1,4-dioxane N 1,4-dioxane N N H 50 % 13 % O O O O

DDQ DDQ 43 % 35 % 1,4-dioxane 1,4-dioxane

N N N H N N H N HB37 HB38

Scheme 15. (a) Lehuede’s synthesis of di(pyrazine)-substituted di-pyridylpyrrole HB34 [86]. (b) von Dobeneck’s synthesis of 1,3-di-(2-quinolyl)-isoindole (HB35) and 1,3-di- (2-isoquinolyl)-isoindole (HB36) [88]. (c) Lavilla’s synthesis of 2,5-di-(2-quinolyl)-pyrrole (HB37) and 2,5-di-(1-isoquinolyl)-pyrrole (HB38) [89].

2.2. Coordination compounds of 2,5-di(2-pyridyl)pyrrolide anions (B) divalent metal precursors, (Table 2 entries 3–5) [61]. This series was extended (Table 2 entries 1 and 2) by Beierlein and Hein in Hein not only pioneered the synthesis of dipyridylpyrrolide 1957 to include air-sensitive iron(II) and chromium(II) complexes, ligands, but also reported a series of first-row transition metal 1 and 2, which oxidised completely within several hours in solu- complexes in the 1950s, which were characterised on the basis tion or gained mass steadily when exposed to air as crystalline 2+ of their elemental composition, magnetism, melting points and solids [63], which contrasts with stable [Fe(tpy)2] . The neutral 1 colour [61,63]. For 50 years, the coordination chemistry of these homoleptic complexes [M(B )2](6–9: M = Fe, Co, Ni, Zn) were also ligands lay dormant and unexplored, until Tour and Natelson prepared from unsubstituted parent HB1 (Table 2 entries 6–9) and investigated their use in molecular devices [62,95–98]. Much of the magnetic properties of the iron(II), cobalt(II) and nickel(II) the chemistry reported since has confirmed Hein’s original pro- complexes investigated [63]. The data were consistent with other posed structures with modern characterisation techniques. high-spin, octahedral complexes of these divalent ions. Recently, Xiao-Yi Yi’s group have reported unusual binding modes of these ligands with copper(I) [99–101], copper(II) [102] and silver(I) [70,99,101] centres, and demonstrated promising results 2.2.2. Other early complexes for these complexes as catalysts in useful organic transformations Hein observed several other neutral complexes during her work 1 1 [70,93]. with the (B ) anion. Heating two equivalents of HB with an alcoholic solution of Cu(OAc)2 (1 equiv.) gave a complex of 1:3 Cu(II): B1 empirical composition (determined by partial analysis for C, H, 2.2.1. Early studies of metal complexes N and Cu) in approximately 80% yield [63]. The complex displayed II 2.2.1.1. Neutral homoleptic first-row transition metal ion complexes, similar solubility to the neutral [M (B)2] complexes (soluble in CCl4 M(B)2. Hein, in her first dipyridylpyrrole paper with Melichar in but not in water), and an octahedral structure 10 was proposed 1954, reported the syntheses of neutral, homoleptic complexes, with one HB1 ligand retaining its acidic proton, Scheme 17.On 2 2 [M(B )2](3–5: M = Co, Ni, Cu) of the diester-substituted HB from heating, complex 10 decomposed. In contrast, after hot alcoholic James McPherson – June 2019 14 Chapter 1: Introduction

J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332 297

to later investigations” [63]. A report on these investigations has NH2 never appeared. O N O N Br KI 2.2.3. Modern studies of homoleptic ﬁrst-row transition metal NH 1-pentanol complexes To the present day. Many of the complexes that Hein originally N N 91 % N synthesised and proposed formulations for have now been characterised and investigated employing modern techniques. Tour’s group was the ﬁrst to re-examine dipyridylpyrrolide 39 HB ligands since Hein, and published a comprehensive summary of their investigations of neutral homoleptic Fe(II), Co(II), Ni(II), Cu 39 HB (II) and Zn(II) complexes in 2006 [62]. The complexes [M(B)2](B = B1,M=Co(7), Zn (9); B = B8,M=Fe(15), Co (16), Ni (17), Cu (18), Zn (19)) were obtained from the metal(II) halide and ligand anion (formed by deprotonation with NaH). Tour’s group only par- 8 tially characterised the iron(II) complex, [Fe(B )2](15) due to its ‘‘meta-stability” in agreement with Hein’s comments regarding the instability of these iron(II) species. SC-XRD studies of 7, 9 and 15–19 were undertaken and reveal these complexes adopt compressed octahedral structures, Fig. 7. The MAN(py) bonds average 0.2–0.3 Å longer than the MAN(pyr) bonds, and all acute intra- ligand NAMAN angles are between 73° and 79° (Table 3). In the

Scheme 16. A high yield route to a functionalised di(2-pyridyl)isopyrrole (HB34) structure of zinc(II) complex 9, one molecule exhibits a very long described by Chen and co-workers [90]. Inset: view from the SC-XRD structure of Zn...N(py) interaction (2.754(5) Å) suggesting that tridentate HB39 (CCDC reference code ETUCON). Key interatomic distances (Å) and angles (°): chelation about the small zinc(II) centre is not sufficiently favour- N3...H 2.470, N4...H 2.487, and N3...H...N4 164.5. able to overcome completely the strain associated with the geometric requirements of the ligand. Tour and co-workers commented that this zinc(II) complex was unstable to light and 2 solutions of HB and CuBr2 were mixed, green crystalline needles acid [62]. grew on cooling that were formulated as [Cu(B2)Br] (11) from their EPR studies of the d9 copper(II) complex 18 confirmed the 2 elemental composition [61]. unpaired electron lies in the dz orbital, with gzz = 2.020 and larger 2 Similar to the preparation of 11, moss green needles of [Fe(B ) gxx = gyy = 2.22. By combining the observed spin–orbit coupling 2 2 Cl2](12) were obtained by cooling a mixture of hot alcoholic HB constant (k’) with the observed dxy?dz transition from the UV– and FeCl3 solutions. Complex 12 was only moderately soluble in vis absorption, where dxy is the lowest lying d-orbital, the degree alcohols, and completely insoluble in water [61]. When the same of delocalisation of this spin onto the ligand can be measured by conditions were attempted with HB1, only inhomogeneous solids the ratio of k’/k = 0.57, indicating a large degree of delocalisation 1 were obtained [63]. The desired [Fe(B )Cl2](13) complex, however, [62]. Cyclic Voltammetry studies also demonstrated that the redox was prepared by heating an alcoholic solution of FeCl2 in air, before behaviours of the metal centres are sensitive to the electronic nat- filtration into a boiling solution of HB1, and cooling to afford black ure of substituents on the pyrrolide core, for example the reduction crystals of 13 [63]. Complexes 12 and 13 were characterised by potential of the cobalt(II) centre in the unsubstituted complex 7 their Cl, Fe and N composition, and a dimeric structure with two was 0.38 V vs. Ag/Ag+, while the cobalt(II) centre is less reducing bridging chlorido ligands was proposed to avoid formulation as in complex 16 (0.01 V vs. Ag/Ag+) with the electron withdrawing what then seemed to be ‘‘unusual” five-coordinate iron(III) centres thiocyanide substituted ligands [62]. It is worth noting that both [61,63]. Complexes 12 and 13 have yet to be re-examined, and these complexes are much more easily oxidised than the known 2+ could provide convenient entry to novel heteroleptic iron species. [Co(tpy)2] analogue, with E1/2(Co(II)/Co(III)) = 0.30 V [103]. 8 Hein also prepared a silver complex by heating alcoholic solu- Magnetic data was collected for [Fe(B )2](15) and the effective 1 tions of silver acetate and HB , which was insoluble in alcohol magnetic moment above 6 K was 5.9 lB consistent with a high and benzene [63]. Partial elemental analysis for C, H, N and Ag spin, octahedral iron(II) centre [62]. The data below 6 K was more led to formulation as ‘‘[Ag(B1)]” (14). Hein and Beierlein also dis- complex and was interpreted in terms of zero field splitting (zfs) cussed the synthesis of a homoleptic cadmium(II) complex, the becoming significant or, alternatively, a transition to a low spin data for which was not given as the compound would be ‘‘subject state [62].

N N N N Cl NH NH NH NH Cl N N N N

HB40 HB41 HB42 HB43

Fig. 6. Ligands made about a metal centre, HB40 [92] and HB41 [93], or for which no preparative details have been given, HB42–HB43 [94].

James McPherson – June 2019 15 Chapter 1: Introduction

298 J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332

Table 2 II Homoleptic [M (B)2] complexes reported by Hein and co-workers in the 1950s [61,63].

R R N R N R MX2 N N N M N H EtOH R R N N N HB [M(B)2]

Entry Ligand (HB) MX2 Yield (%) Characterisation data

a 1 EtOOC CrCl2 40 [63] N, Cr% [63] a COOEt 2 Fe(OAc)2 70 [63] MP 208 °C N, Fe% 5.0 ± 0.1 m [63] N N B 3 H Co(OAc)24H2O Not reported MP 216 °C N N, Co% [61] 2 ° 4 HB NiBr2 60 [61] MP 220 C C, H, N, Ni% [61]

5 Cu(OAc)2 Not reported N, Cu% [61]

a 6 H Fe(OAc)2 Not reported MP 168 °C H C, H, N, Fe%

4.5 ± 0.1 mB [63]

7 Co(OAc)2.4H2O 40 [63] MP 273–274 °C N N H C, H, N, Co% N 4.8 ± 0.1 m [63] 1 B 8 HB Ni(OAc)2 Not reported MP 318 °C C, H, N, Ni%

3.2 ± 0.1 mB [63] b 9 Zn(OAc)2 Not reported N, Zn% [63]

a Sensitive to oxygen [63]. b Also reported synthesis with zinc powder in molten naphthalene (with H2 (g) evolution) [63].

H H

N EtOOC COOEt N N R R Cu(OAc) CuBr /EtOH N N 2 2 N Cu N EtOH EtOH N N Cu N H H N N 1 2 Br H H HB N HB N N 80 % 11 H H

R R R R R: Fe precursor: Fe precursor 12 COOEt FeCl N 3 N N EtOH N 13 HFeCl H Fe N 2 N Cl Cl

AgOAc N N N H EtOH N N Ag N HB1 67 % 14

Scheme 17. Top: reaction of HB precursors with Cu(II) salts [61,63], middle: preparation of neutral, iron(III) complexes [61,63], and bottom: preparation of a neutral complex 14 formulated as ‘‘Ag(B1)” [63].

James McPherson – June 2019 16 Chapter 1: Introduction

J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332 299

1 Fig. 7. SC-XRD structures determined by Tour and co-workers of neutral, homoleptic [M(B)2] complexes with H-atoms (and a solvent CH2Cl2 molecule for [Zn(B )2] top right) 1 1 omitted for clarity [62]. Note: for both [Co(B )2] and [Zn(B )2], two unique molecules are present in the asymmetric unit.

Table 3

Average MAN bond distances (Å) and NAMAN bond angles (°) from the SC-XRD structures of [M(B)2] complexes 7, 9 and 15–19.

Complex (CCDC reference) MAN(py) MAN(pyr) \N(py)AMAN(py0) \N(py)AMAN(pyr)

1 y 7 [Co(B )2] (UCIFUI) 2.299(3) 1.944(3) 147.3(1) 73.7(1) 1 y 9 [Zn(B )2]A (UCIFOC) 2.418(7) 1.914(6) 147.6(3) 73.8(1) 1 à 9 [Zn(B )2]B (UCIFOC) 2.206(5) 1.923(6) 145.4(2) 79.0(2) 8 15 [Fe(B )2] (UCIGUJ) 2.275(8) 2.207(8) 146.2(3) 73.1(3) 8 16 [Co(B )2] (UCIGOD) 2.240(4) 1.971(3) 148.3(1) 74.1(1) 8 17 [Ni(B )2] (UCIGIX) 2.22(2) 1.96(1) 151.3(5) 75.7(4) 8 18 [Cu(B )2] (UCIGET) 2.265(7) 1.892(7) 150.3(3) 75.2(3) 8 19 [Zn(B )2] (UCIGAP) 2.297(6) 1.960(7) 147.9(2) 74.0(2) y The unit cell contains two unique complex molecules. à Tetrahedral Zn(II): only N atoms bound to Zn(II) centre measured (i.e., Zn1A...N51A excluded, except for bite angle).

Grohmann, Müller and co-workers also investigated the mag- available, namely the SC-XRD structure of the di-thiocyanate 1 netic properties of Hein’s homoleptic iron(II) complex, [Fe(B )2] derivative 15: average NpyAFeANpy0 = 146.2(3)° and average (6), in particular looking for evidence of spin crossover. In agree- FeANpy = 2.275(7) Å versus average FeANpyr 2.027(8) Å (a differ- ment with Tour and co-workers’ previous ﬁndings for the thio- ence of 0.25(2) Å) [62]. It is noteworthy that the predicted distor- cyanate derivative 15, complex 6 was apparently high spin over tion from idealised octahedral geometry for 6 is more than for 2+ the temperature range (5–330 K) however, very little experimen- the analogous [Fe(tpy)2] species (bite angle of 161° and 0.08 Å tal detail was provided [104]. Moreover, only an image and a mean average FeAN bond difference) [105]. FeAN bond distance of 2.213(3) Å are reported for a SC-XRD struc- The reaction of (bis(trimethylsilyl)amido)(1,3-di-(2-pyridyl)-2- ture that has never been deposited in a database, and there are no azaallyl)iron(II) with a high ratio (4 equivalents) of di-p-tolyl- preparative details, for either the complex or crystal [104]. acetylene afforded a new diamagnetic product with high symme- Jakubikova and her group have subsequently calculated (DFT) try that Wolczanski and co-workers tentatively assigned as [Fe 40 theoretical geometric and electronic structures for several poly- (B )2](20), Scheme 18 (top). However, the formulation seems 1 (pyridyl)iron(II) species, including Hein’s [Fe(B )2](6) complex likely incorrect as switching the spin state from a high spin quintet 1 [105]. Their calculations predict 6 should be high spin, consistent in [Fe(B )2](6) to a low spin singlet in 20 simply by substitution of with the previously acquired experimental observations discussed the central pyrrolide ring with tolyl groups would be remarkable. above. They also predict signiﬁcant distortion for 6 from the ideal The result certainly is worthy of a follow up investigation [92]. 2+ octahedral [Fe(NH3)6] reference complex: NpyAFeANpy0 = 154°, Also, amongst the products of the reaction of (bis(trimethylsi and the equatorial FeANpy bonds 0.20 Å longer than the axial lyl)amido)(1,3-di-(2-pyridyl)-2-azaallyl)iron(II) with only 2 equiv- FeANpyr bonds on average [105]. These predictions seem reason- alents di-p-tolyl-acetylene that Wolczanski and co-workers able when compared with the only experimental data which is obtained was red crystalline dimer 21 involving the reduced

James McPherson – June 2019 17 Chapter 1: Introduction

300 J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332

H H 4 N tol N tol N N Fe N ++ N Fe N tol tol C6D6 N N N

Me3Si SiMe3 20

tol tol 2

N C6H6 N Fe N N Me3Si SiMe3

Scheme 18. Preparation of a diamagnetic iron(II) complex 20 of (B40), and a dinuclear iron(II) complex (21 – CCDC code QIBBAG) involving its 2-electron reduced 2,5- dihydropyrrolide analogue from a di-(p-tolyl)acetylene and 1,3-di-(2-pyridyl)-2-azaallyl as reported by Wolczanski and co-workers [92]. Selected average bond lengths (Å) and angles (°): FeAN(py) 2.154(2), FeAN(Pyr) 2.162(1), Fe0AN(Pyr) 2.107(1), FeAFe0 3.0539(4), N(py)AFeAN(py0) 35.96(6), N(py)AFeAN(pyr) 77.27(6), FeAN(pyr)AFe0 91.34 (5), plane(py)Aplane(pyr) 54.73(7) and plane(py)Aplane(py0) 27.54(6).

gold source/drain electrodes

SiO2 gate oxide

SCN

NCS N N N N NCS SCN Au S S N NH N M N N M N NCS SCN S S 8 N N N N N HB

8 p+ Si(100) wafer gate electrode [M(B )2]

DS DS E

Scheme 19. Top: molecular transistors investigated by Tour and Natelson’s groups, where M are divalent Cu(II) or Co(II) centres [62,95–98]. Bottom, a generic energy diagram for Kondo resonant conduction of unpaired spins between a source and drain electrodes (S and D respectively) at zero bias.

40 2,5-dihydropyrrolide core (H2B ) , Scheme 18 (bottom), which subunits, Scheme 19, would provide several advantages for such was fully characterised, including by SC-XRD and SQUID magne- applications. First, substituted dipyridylpyrrolide anions should tometry. The magneto properties of 21 were somewhat form neutral octahedral complexes with divalent metal centres, complicated: leff (300 K) 5.3 lB, compared with 6.93 lB for thereby avoiding complications due to counter ions. Second, non-interacting high spin Fe(II) centres, and declined steadily to p-conjugation through pyrrolide rings should facilitate electron a modest plateau at 2.4 lB beginning at 25 K before onset of transfer from one set of electrode-binding S-atoms through the ZFS. The data remain open to interpretation, but are consistent metal core to the other set of electrode-binding S-atoms. Third, with interacting iron centres that are separated by 3.0539(4) Å in the tridenticity of the ligands should minimise dissociation and the 173 K crystal structure [92]. other related decomplexation reactions, while also preventing sul- The Tour groups’ interest in dipyridylpyrrolide metal complexes fur coordination to the metal core [62]. Scheme 19 (top) presents a was for their promise as componentry for molecular transistors generic depiction of the construction of such devices. A typical

[62,95–98]. They anticipated that thio-substituted [M(B)2] device consisted of: two 15 nm Au ﬁlms as the source and drain

James McPherson – June 2019 18 Chapter 1: Introduction

J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332 301

+ a)

O N O N N N

- CoCl2·4H2O CoCl4 NH N Co N MeOH/H2O

N N N N N O 88 %

HB39 22

39 + Scheme 20. (a) Synthetic route to the anticancer Co(III) complex 22 [90]; and (b) view of the [Co(B )2] cation from SC-XRD structure of 22 (CCDC code ETUCIH), with H atoms omitted for clarity. Key average bond lengths (Å) and angles (°): MAN(py) 1.963(8), MAN(pyr) 1.810(8), N(py)AMAN(py0) 160.0(3), N(py)AMAN(pyr) 80.0(3).

39 collected in good yield directly from the reaction of HB and CoCl2 and the SC-XRD structure obtained, Scheme 20b [90,107]. Notably, although not commented on by Chen et al., the extension off the ‘‘back” of the central isoindolide serves to highlight the puckering Z+ inherent in the [M(B)2] structures. The complex inhibited tumour growth, likely through mitochondrial dysfunction. Although higher doses than cisplatin were required to achieve the same level of growth inhibition (5.0 mg/kg for 22 versus 2.0 mg/kg for cisplatin), 23 22 maintained a better in vivo safety proﬁle. An alternate cobalt complex without the dipyridylisoindole group bound directly to cobalt(II) centre had far poorer activity [90]. The synthetic method 1 Fig. 8. Crystal structure from SC-XRD data of [Cu(B )Cl] (23, CCDC code KANXIJ) for the preparation of 22 was also protected by the authors in 2015 reported [110] by Zhang and co-workers. Key average bond lengths (Å) and angles (one year prior to their report of its SC-XRD structure and activity) (°): Cu1AN(py) 2.145(3), Cu1AN1 1.863(3), Cu1ACl1 2.222(1), N2ACu1AN3 153.7 (1), and N(py)ACu1AN1 77.0(1), N(py)ACu1ACl1 102.69(8) and N1ACu1ACl1 under patent, albeit for the preparation of the cobalt(II) analogue, 39 174.34(8). that is for neutral ‘‘[Co(B )2]”, which now seems incorrect [107]. The only modern example of complexes of the B anion with 8 nickel in the literature since Hein’s work is Tour’s [Ni(B )2](17) electrodes atop 200 nm SiO2 as the gate oxide with a degenerately [62]. In the electronic spectrum, the three main sets of transitions 8 p-doped Si(1 0 0) wafer as the underlying gate electrode [96]. expected for the octahedral (Oh) d centre, which correspond to 3 3 3 3 Fabrication of these devices resulted in a statistical distribution spin allowed transitions ( A2g to T2g, T1g and T1g (P)) are each 3 of electrode pairs, of which about 15% were bridged by a single split into two bands due to S4 symmetry. The transitions to B2g 3 3 molecule [97]. and Eg ( T2g in Oh) were deconvolved from the spectrum and Kondo resonance is non-linear conduction relayed by unpaired observed at 10,550 cm1 and 9550 cm1 respectively [62]. Mag- spins, Scheme 19 (bottom), and was observed in devices made with netic susceptibility measurements gave a magnetic moment of both the Co(II) [62,95–97] and Cu(II) [62,96,98] systems with 4.3 lB. Cyclic voltammetry of complex 17 revealed a reversible Ni Kondo temperatures >50 K. The high Kondo temperatures are com- (II)/Ni(III) couple at ca. +0.82 V vs. Ag/AgNO3, which compares with 2+ parable to those observed in scanning tunnelling microscopy mea- +1.65 V vs. Ag/AgCl for the Ni(II)/Ni(III) couple of [Ni(tpy)2] [109]. 1 8 surements of transition metals directly bonded to gold surfaces Finally, Tour’s zinc complexes, [Zn(B )2](9) and [Zn(B )2](19) and were explained by the large degree of spin delocalisation onto are also the only modern reports of zinc complexes with this class the ligand observed in magnetic studies of these complexes [62].At of ligand [62]. As expected, these complexes are diamagnetic, with ⁄ the same time as the experimental work, Seminiaro and Yan only charge transfer (CT) and p p transitions observed in the UV reported theoretical studies on the same Co(II) device, which also spectra. No metal-centred redox processes were observed by cyclic predicted good hole conductivities [106]. voltammetry, however reduction of the thiocyanate group to thio-

Another homoleptic cobalt complex has appeared in the recent late was observed for 19 at ca. 2.2 V vs. Ag/AgNO3 (compared to medicinal literature. Chen and co-workers installed a large, planar 1.96 V for HB8). G-quadruplex (G4-DNA) binding structure at the back of a di(2- pyridyl)isoindole (HB39) as shown in Scheme 20a [90,107]. The 2.2.4. Other first-row transition metal complexes authors had previously identified G4-DNA as specific tumour- Despite the promise offered by Hein’s syntheses of heteroleptic selective targets for chemotherapy, with stabilisation of G4-DNA complexes of B derivatives and first row metal ions such as [Fe(B) 1 2 2 triggering telomerase inhibition, an enzyme active in 80–90% of Cl2](12: B = B ; 13: B = B ) or [Cu(B )Br] (11), in the modern era 39 human tumour cells [108]. Crystals of [Co(B )2][CoCl4](22) were only Yi and co-workers have reported such complexes and then, James McPherson – June 2019 19 Chapter 1: Introduction

302 J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332

N N N NaN3 in thf N N N Cu NNN N N N Cu N N N N N N 24

n 24

NO3 N NaNO3 N in thf N Cu N N N Cu N Cu Cl N 23 N O3N 25 25

N N AgOTf N Cu N in thf OTf- N N Cu

N N N 26

Scheme 21. Synthesis and structures from SC-XRD data of multinuclear, helical copper(II) complexes 24–26 reported by Yi, Zhang and coworkers from 23 [102]. Two repeat units are shown for polymeric 24, and H atoms, anions and solvent molecules are omitted for clarity.

Table 4 Selected bond lengths (Å) and inter-plane angles (°) for complexes 24–26 [102].

Complex (CCDC reference) CuAN(py) CuAN(pyr) CuAN(pyrbr) Cu...Cu \plane(py)Aplane(pyr) \plane(py)Aplane(py0)

1 24 [Cu2(l-B )2(N3)2]n (LOYTOK) 2.037(2) – 2.002(2) 3.0625(5) 14.6(1) 44.0(1) 2.055(2) 2.534(2) 5.1634(5) 32.1(1) 1 25 [Cu2(l-B )2(NO3)2] (LOYTUQ) 2.003(1) – 1.988(2) 2.9234(3) 13.35(6) 46.26(6) 2.019(1) 2.371(1) 35.71(7) 1 26 [Cu2(l-B )3]OTf (LOYVAY) 2.063(3) 1.952(6) 2.155(5) 2.815(1) 10.5(3) 36.4(3) 3.152(6) 2.341(5) 26.2(3)

exclusively, only copper(I) and copper(II) examples. There are no and was conﬁrmed by studies with hydroxyl radical scavengers modern reports of other heteroleptic species. (DMSO), which diminished the activity, while singlet oxygen scav- Green [Cu(B1)Cl] (23) was simply prepared by combining HB1 engers (histidine and tryptophan) had no effect. and copper(I) chloride with triethylamine as base in air [110].In Reactions of complex 23 with sodium azide and sodium nitrate the SC-XRD structure of 23, Fig. 8, each Cu(II) centre adopts a dis- and silver(I) triﬂate the resulted in the novel, multinuclear copper torted square planar geometry with the sum of all LACuAL angles (II) complexes, 24–26, with bridging B ligands, Scheme 21 [102]. being 359.42° whereas the angle N(pyr)ACuAN(pyr0) is 153.7° (vs. The steric demands associated with dinuclear bridging were pro- 360° and 180°, respectively, for an ideal square planar geometry). posed to disrupt the co-planarity of the pyridyl–pyrrolide rings As expected, the CuAN(pyr) distance at 1.863(3) Å is short com- resulting helical structures. Complexes 24–26 were characterised pared to the CuAN(py) distances which average 2.145(3) Å. Weak by SC-XRD (insets, Scheme 21), IR and elemental analysis. The 1 intermolecular face-to-face p-stacking arranges the monomers magnetic properties of polymeric [Cu2(B )2(N3)2]n (24) were also into dimers, with the chloride acting as a weak bridging ligand investigated from 1.8 to 300 K at 1000 G [102]. The data revealed 0 linking to the neighbouring Cu(II) centre (Cu1...Cl1 2.841(1) Å). ferromagnetic coupling between the two copper ions (J1 = 7.36 5 1 1 Complex 23 weakly binds (Ka = 2.26 10 L mol ) calf thymus cm ) attributed to the bridging pyrrolide ligand, while the l1,3- DNA via groove binding rather than classical intercalation, and is azido ligand led to weak antiferromagnetic interactions (J2 = also capable of inducing oxidative cleavage of supercoiled DNA in 0.75 cm1), with the material displaying overall very weak anti- the presence of ascorbate reductant [110]. Zhang et al. proposed ferromagnetic coupling (zJ0 = 0.017 cm1) between the repeating that 23 reacted with DNA to form a Cu(II)–DNA complex, and that pairs of copper(II) centres. reduction to a Cu(I)–DNA intermediate led to the subsequent gen- Key average bond lengths, interatomic separations, and angles eration of hydroxyl radicals through reaction with molecular oxy- between the ring planes are summarised in Table 4, with the CCDC gen. Attack of DNA by hydroxyl radical then lead to strand scission, reference codes. The intra-ligand twisting is highlighted by the

James McPherson – June 2019 20 Chapter 1: Introduction

J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332 303

N N N N N NaH, CuI THF H Cu Na a) N THF THF N N N HB1 27 27

Ph P PPh R R 3 3 R X R i) NaH Cu Cu N N ii) Cu2X2(PPh3)2 H N N b) N N

HB 28-33 28

R R R P P i) NaH, CuCl N Cu N H N c) N N ii) 1 eq. Bis(phosphine) or 2 eq. PPh3 N R

HB 34-38 35 Bis(phosphine):

O O

PPh2 PPh2 PPh2 PPh2

XANTPhos DPEPhos

1 9 1 1 Scheme 22. Preparation of novel copper(I) complexes of (B ) (R = H) and (B ) (R = Br): (a) [Cu(l-B )2Na(thf)2](27) [101]; (b) [{Cu(PPh3)}2(l-B)(l-X)] (for B = B , 28: X = Cl; 29: X = Br; 30: X = I; and for B = B9, 31: X = Cl; 32, X = Br; 33:X=I)[100] and (c) [Cu(j2-B)(PP)] (34: B = B1,PP = XANTPhos; 35: B = B1,PP = DPEPhos; 36: B = B9, 9 9 PP = XANTPhos; 37: B = B ,PP = DPEPhos, 38: B = B ,PP = 2PPh3 [99]. Insets: representative SC-XRD structures.

Table 5 Selected bond lengths (Å) and angles (°) for dinuclear complexes of (B1) and (B9) [100,101].

Complex (CCDC reference) MAN(py) MAN(pyr) M...M \plane(py)Aplane(pyr) \plane(py)Aplane(py0)

1 27 [Cu(l-B )2Na(thf)2](RIRQAM) 2.004(1) 2.058(1) 2.893(1) 15.80(6) 40.60(6) 2.437(1)y 2.675(1)y 26.50(6)y 1 28 [{Cu(PPh3)}2(l-B )(l-Cl)] (CUJSEH) 2.036(5) 2.226(5) 2.698(1) 19.9(2) 34.6(2) 2.048(5) 2.168(5) 19.8(2) 1 29 [{Cu(PPh3)}2(l-B )(l-Br)] (CUJSIL) 2.048(8) 2.155(8) 2.671(1) 20.9(4) 28.8(4) 2.039(7) 2.177(8) 10.6(4) 1 30 [{Cu(PPh3)}2(l-B )(l-I)] (CUJSOR) 2.058(4) 2.190(5) 2.6946(8) 20.7(3) 29.3(2) 2.058(4) 2.166(5) 11.5(3) 9 31 [{Cu(PPh3)}2(l-B )(l-Cl)] (CUJSUX) 2.093(3) 2.284(3) 2.7960(7) 26.9(1) 48.3(1) 2.108(3) 2.165(3) 21.5(1) 9 32 [{Cu(PPh3)}2(l-B )(l-Br)] (CUJTAE) 2.066(5) 2.313(4) 2.7389(8) 20.2(2) 33.4(2) 2.095(5) 2.180(3) 13.3(2) 9 33 [{Cu(PPh3)}2(l-B )(l-I)] (CUJTEI) 2.100(7) 2.185(5) 2.701(1) 13.7(2) 33.7(2) 2.059(6) 2.290(5) 20.1(2) y M=Na+. angles between the normal to each ring plane: about 45° for the Xiao-Yi and co-workers prepared copper(I) complexes 27–38 1 two [Cu2(l-B )2] motifs in 24 and 25, and about 36° for the with centres that involve ‘unusual’ (non-terdentate) binding 1 + 1 [Cu2(l-B )3] helix in 26. Comparing the angles between the planes modes, Scheme 22 [99–101]. Deprotonation of HB with sodium through each of the pyridyl arms to that passing through the cen- hydride, followed by addition of copper(I) iodide in tetrahydrofu- 1 tral pyrrolide shows that the pyridyl coordinating the second metal ran afforded the multinuclear [Cu(l-B )2Na(THF)2](27) which centre is twisted more, with the other pyridyl–pyrrolide unit rela- involve a sodium and copper(I) centre bridged by two (B1) anions tively co-planar. [101]. Yi and co-workers also prepared six bridged dicopper(I)

James McPherson – June 2019 21 Chapter 1: Introduction

304 J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332

Table 6 Selected bond lengths (Å) and angles (°) for bidentate Cu(I) complexes of B ligands [99].

Complex (CCDC reference) CuAN(py) CuAN(pyr) NACuAN \plane(py)Aplane(pyr) \plane(py)Aplane(py0) 34 [Cu(j2-B1)(XANTPhos)] (JAGJUZ) 2.083(8) 2.052(8) 80.9(4) 9.0(3) 170.8(3) 35 [Cu(j2-B1)(DPEPhos)] (JAGJOT) 2.113(1) 2.054(1) 81.48(6) 12.55(7) 156.2(7) 36 [Cu(j2-B9)(XANTPhos)] (JAGKAG) 2.238(4) 2.027(4) 80.8(1) 3.4(2) 174.4(2) 2 9 38 [Cu(j -B )(PPh3)2](JAGKEK) 2.221(3) 2.057(2) 80.7(1) 10.4(2) 162.1(2)

*py0 is the unbound pyridine.

CuAN(py) bond distance is 2.067(2) Å, significantly shorter than 34 34 the average CuAN(pyr) distance at 2.208(2) Å. Complexes 28–33 ab em were readily hydrolysed in solution, but were stable in the solid 35ab 35em state. At 298 K, in the solid state these complexes are non- luminous, however freshly prepared solutions in dichloromethane 3 36ab 36em display weak MLCT luminescence on excitation at 420 nm. Yi’s group also prepared mononuclear copper(I) complexes 34– 37ab 37em 38, starting from a chelating phosphine ligands, XANTPhos (4,5-bis (diphenylphosphino)-9,9-dimethylxanthene) and DPEPhos (oxydi- 2,1-phenylene)bis(diphenylphosphine) [99], the appropriate dipyridylpyrrole precursor (HB1 or HB9), copper(I) chloride and sodium hydride in tetrahydrofuran (Scheme 22c) [99]. Complexes 34–38 are air stable and the structures of 34–36 and 38 were determined by SC-XRD studies (Scheme 22 insets); selected bond lengths and angles are shown in Table 6. Of most note is the bidentate coordination of the dipyridylpyrrolide ligand in each of 34–38. The average CuAN(py) bond distance at 2.1638(9) Å is considerably longer than the average CuAN(pyr) bond distance at 2.0475(9) Å. The absorption and emission properties of complexes 34–37 Fig. 9. Absorption and emission spectra of dichloromethane solutions of mononu- were investigated [99]. In the solid state, the complexes were not clear copper(I) complexes 34–37; reprinted (adapted) with permission from Ref. [99], copyright (2015) Elsevier. luminescent. When the dichloromethane solutions were excited at 420 nm (MLCT region) however, strong luminescence was observed, with the emission from the bromo-substituted (B9) halide complexes by again deprotonating the appropriate HB1 (28– complexes blue shifted (460 and 468 nm for 36 and 37 respec- 9 30)orHB (31–33) precursor with sodium hydride, and then add- tively) relative to the unsubstituted (B1) complexes (both 34 l ing dimeric copper(I) precursors [Cu2( -X)2(PPh3)2] (X = Cl, Br, I) to and 35 at 472 nm), Fig. 9. afford [{Cu(PPh3)}2(l-B)(l-X)] complexes 28–33 (28: X = Cl, 1 1 1 9 B = B ; 29: X = Br, B = B ; 30:X=I,B = B ; 31: X = Cl, B = B ; 32: 2.2.5. Second- and third-row transition metal complexes 9 9 X = Br, B = B ; 33:X=I,B = B )(Scheme 2b and Table 5) [100]. Sotoyama and co-workers claimed the use of platinum(II) com- These complexes involve two distorted tetrahedral copper(I) plexes 39–53 as luminescent materials for their use in electrolumi- centres, bound to the flanking pyridyl nitrogen atom and a triph- nescent devices [94]. These complexes were prepared as shown enylphosphine ancillary ligand, bridged by both the pyrrolide core Scheme 23, by refluxing the precursors HB1, di(2-pyridyl) and a halide anion (Cl ,Br and I ). In dimers 28–33, the average isoindole (HB37) or di(2-pyridyl)napthopyrrole (HB38) in glacial

X-:

N N N -O K2PtCl4 NaX NH N Pt Cl N Pt X glacial acetic acid, 130 °C acetone/methanol 42-44 1 42 N N N HB ,HB 39-41 N N- 43 and HB N 45-47

S HS H N N N S NaOH CuI (cat) N Pt S N Pt CH Cl /NEt N acetone/dmso, reflux 2 2 3 N 48-50 N 51-53

Scheme 23. Preparations of electroluminescent platinum(II) complexes 39–53 reported by Sotoyama and co-workers [94].

James McPherson – June 2019 22 Chapter 1: Introduction

J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332 305

Cl i) NEt3 ii) (COD)PtCl Pt N H N 2 N N N N

HB1 39 39

+ Cl OH2 i) LiN(SiMe3)2 ii) PdCl (NCPh) Pd Pd N H N 2 2 N N AgOTf N N OTf- N N N

HB1 45 55

54 55

Scheme 24. Preparation of square planar platinum(II) and palladium(II) complexes reported by Wayland, Zdilla et al. in 2012 [69]; Insets: SC-XRD structures of [Pt(B1)Cl] 1 1 (39), [Pd(B )Cl] (54) and [Pd(B )(OH2)]OTf (55) (OTf anion omitted for clarity); for CCDC reference codes, see Table 7.

Table 7 Selected average bond lengths (Å) and inter-plane angles (°) for complexes 39, 54 and 55 [69].

Complex (CCDC reference) MAN(py) MAN(pyr) \N(py)AMAN(py0) \N(py)AMAN(pyr) 39 [Pt(B1)Cl] (TELQEI) 2.057(4) 1.892(4) 156.4(2) 78.2(1) 54 [Pd(B1)Cl] (TELPUX) 2.074(6) 1.886(5) 155.4(2) 77.7(1) 1 55 [Pd(B )(OH2]OTf (TELQAE) 2.081(6) 1.852(5) 156.1(2) 78.1(2)

[Pd(NCPh) Cl ] Pd N H N 2 2 N N N NEt3 N

56 and 57 HB R1 R2 R1 R2

[Ru(dmso)4Cl2] [Ru(tpy)Cl3] [Ru(dmso)4Cl2], bpy NEt3 NEt3 NEt3

z+ z+

N N N N N N MeOOC R1 N Ru N N Ru N N Ru N COOMe R COOMe N N N N 2 L N

58 59-62 63 and 64

12 Scheme 25. Synthesis of palladium(II) and ruthenium(II) complexes of dipyridylpyrrolide ligands reported by the Colbran group [74]. [Pd(B)Cl] complexes are 56: B = B (R1 20 z+ 12 20 17 21 = Me, R2 = COOMe); 57: B = B (R1 =R2 = Me); [Ru(B)(tpy)] complexes are 59: B = B ,z=1;60: B = B ,z=1;61: B = B (R1 = 4-py, R2 = COOMe), z = 1; 62: B = B (R1 =4- + 12 z+ Mepy ,R2 = COOMe), z = 2; [Ru(B )(bpy)L] complexes are 63: L = Cl, z = 0; 64: L = MeCN, z = 1).

James McPherson – June 2019 23 Chapter 1: Introduction

306 J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332

56 58 59 63

laedi devresbo

N N N N 126° N 112° N 110°

126° 140° 143° O O

Fig. 10. Top: SC-XRD structures of a palladium(II) complex (56) and three ruthenium(II) complexes (58, 59 and 63)of(B12) reported by the Colbran group (H-atoms, anion and lattice solvent molecules are omitted for clarity; for CCDC reference codes, see Table 8) [74]. Bottom: Illustration of the exterior angles associated about an ideal 5- membered ring versus the mean exterior angles from the SC-XRD structures of the four complexes as reported by Colbran et al. [74].

Table 8 Selected average bond lengths and inter-ring angles for complexes 56, 58, 59 and 63 [74].

Complex (CCDC code) MAN(py) MAN(pyr) \N(py)AMAN(py0) \N(py)AMAN(pyr) 56 [Pd(B12)Cl] (LOGKOJ) 2.057(2) 1.878(2) 156.24(7) 78.12(8) 12 58 [Ru(B )2] (LOGLEA) 2.097(4) 1.954(4) 152.3(2) 76.1(2) 12 y 59 [Ru(B )(tpy)]PF6 (LOGKUP) 2.105(7) 1.965(6) 151.5(3) 75.8(3) 63 [Ru(B12)(bpy)Cl] (LOGLAW) 2.095(3) 1.932(3) 153.3(1) 76.8(1) y For tpy ligand in 59:RuANouter 2.057(8) Å, RuANinner 1.948(8) Å, NouterARuANouter 158.6(3)° and NouterARuANinner 79.3(3)°.

acetic acid with K2PtCl4 to afford the complexes [Pt(B)Cl] (39: B = B1; 40: B = B42; and 41: B = B43) in moderate yield. The chloro ligand was exchanged for phenoxide (42: B = B1; 43: B = B42; and 44: B = B43), 1,2,4-triazol-1-ide (45: B = B1; 46: B = B42; and 47: B = B43), 2-benzothiazolethiolate (48: B = B1; 49: B = B42; and 50: B = B43), or phenylacetylide (51: B = B1; 52: B = B42; and 53: B = B43). These complexes were doped (2%) into 4,40-bis(9-carbazolyl) biphenyl (kem = 380 nm) and deposited on a quartz substrate to form luminescent thin films, 50 nm thick. Excitation at k = 365 nm resulted in emission with photoluminescent quantum yields (U) greater than 80%. Complexes of (B1) had the highest U (between 85% for 39 and 92% for 48), with emission at 490 nm. Extension of the pyrrolyl aromatic system resulted in red shifting of the emission at the cost of U: the indolyl complexes emitting between 550 and 560 nm (U 85%) and the napthopyrrolyl complexes at 610 nm (U 82%). Electroluminescent (EL) devices consisting of 30 nm layers of these luminescent thin films between a positive indium tin oxide (ITO) electrode and an aluminium nega- 12 12 + tive electrode were constructed, and were luminescent (at the Fig. 11. CVs of [Ru(B )(bpy)Cl] (63) (purple traces) and [Ru(B )(bpy)(MeCN)] (64) (pink traces) in 0.1 M [NBu4][PF6] MeCN under N2 (dotted traces) and CO2 same wavelengths as previously described) when a voltage was (solid bold traces) atmospheres. Reproduced (adapted) with permission from Ref. 2 applied with a current density of 5 A/m . The pyrrolide systems [74], copyright (2014) WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim. (39, 42, 45, 48 and 51) were again the most efficient (with EL efficiencies between 40 and 42 cd/A), but also required the highest voltages (6.9–7.0 V), in comparison to 34–36 cd/A at 6.2–6.4 V for Cl] (39: M = Pt; 54: M = Pd), from HB1 and an appropriate metal the isoindolides (40, 43, 46, 49 and 52) and 14–17 cd/A at 5.8– source under basic conditions, Scheme 24 [69]. Treatment of 54 1 6.1 V for the napthopyrrolides (41, 44, 47, 50 and 53). The inven- with silver triflate gave [Pd(B )(H2O)])OTf) (55). The SC-XRD struc- tors conclude by claiming that materials derived from 39 to 53 tures of 39, 54 and 55 were determined, Scheme 24. The average are phosphorescent, and may be used to construct luminescent MAN bond length and NAMAN0 angle data for the complexes are devices [94]. presented in Table 7, and the central MAN(pyr) distances are Wayland, Zdilla and co-workers prepared square planar plat- shorter than the terminal MAN(py) distances by ca. 0.2 Å, similar inum(II) and palladium(II) complexes of the (B1) anion, [M(B1) to Yi’s square planar copper(II) complex 23.

James McPherson – June 2019 24 Chapter 1: Introduction

J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332 307

COOH +

COOH N 1. RuCl3 3hat80°C N 1 2. HB , NBu3 8 h at 140 °C N Ru N COOH Cl- HOOC N N,N-dimethylacetamide N N N N N N 65

Scheme 26. Preparation of complex 65 reported by Nomura and co-workers at Fujiﬁlm Corporation as one of a series of ruthenium(II) based dyes claimed for dye-sensitized solar cells [113,114].

The report of Wayland, Zdilla and co-workers is the origin of tances (by 0.048(8) Å) [74]. While the SC-XRD structure of the some misconceptions about the bonding of pyrrolide donors in HB12 ligand precursor was not obtained, Colbran et al. illustrated the subsequent literature. They state that pyrrolide donors have the ‘pinched in’ nature of the pyridyl rings of the (B12) anion by ‘‘flexible p-properties”, that is they can act as both p-donors and considering the ideal external angles about the five-membered p-acceptors [69], and their report has been frequently cited since ring, as depicted in Fig. 10. From their analysis of the structural to support such statements [93,100–102]. While there is literature parameters for a hypothetical undistorted dipyridylpyrrolide precedence of ‘‘flexible” (weak) p-donation from pyrrolide ligands ligand versus those found in the complexes of non-symmetrically (with the nitrogen p-orbital participating in the aromatic p- substituted (B12) ligand, it is clear that the steric demands of system) [111,112,45], the authors only cite one example in support the 3,4-pyrrolide substituents play a significant role in the geome- of pyrrolide acting as a p-acceptor: but that article [112] actually try of its complexes. As one would expect, the more congested pyr- only states that for a series of square planar pyrrolide complexes, idyl adjacent to the ester group is that ‘‘pushed” furthest from ideal the pyrrolide p-orbitals are the HOMOs (and so pyrrolide acts as geometry. a p-acceptor). Yi and Zhang have also claimed that p- Only broad and very weak ligand-centred fluorescence at 405 backbonding is responsible for the Cu(II)ANpyr distance being nm was observed for ruthenium(II) complexes, 58 and 59, in ace- 1 shorter than Cu(II)ANpy for [Cu(B )Cl] (23) (see above) [110]. How- tonitrile at 298 K when excited at 330 nm and 310 nm respectively ever, no attempt was made to deconvolute the geometric con- [74]. Electrochemical analysis of the four [Ru(B)(tpy)][PF6] com- straints associated with tridentate coordination of the plexes (59–62) prepared by the Colbran group confirmed Tour’s dipyridylpyrrolide ligand (see Section 1.2). In point of fact there observation that metal-centred redox processes, in this case the is very little evidence of pyrrolide acting as a p-acceptor within Ru(II)/Ru(III) couple, could be tuned depending on the electronic the work presented by Wayland, Zdilla and co-workers [69]. Struc- nature of the substituents installed at the pyrrolide core. Complex tural analyses are presented that draw conclusions from bond 62 with the electron poor, zwitterionic B21 ligand exhibited the lengths within the pyridyl and pyrrolide rings that are not statisti- highest Ru(II)/Ru(III) couple (+0.35 V vs. Fc+/0); exchanging di- cally different (when ESDs are propagated to 3r), and the authors’ methyl substituted (B20) for B21 to give complex 60 caused the own modelling suggests that any metal-to-ligand back-donation is Ru(II)/Ru(III) couple to shift 390 mV lower (to 0.04 V). The redox predominantly centred on the pyridyl p⁄-orbitals. non-innocence of the dipyridylpyrrolide ligands was also particu- McSkimming et al. also reported the facile incorporation of larly evident in these complexes. Oxidation of the pyrrolide was square planar palladium(II) and, for the first time, octahedral observed between +0.92 V for 60 and +1.47 V for the electron poor ruthenium(II) centres with dipyridylpyrrolide anions (B), 62, while reduction of the pyridyl ring was very much less sensitive Scheme 25 [74]. The palladium(II) complexes [Pd(B)Cl] (56: B = to the nature of pyrrolide substitution, with reduction potentials 12 20 B ; 57: B = B ) were prepared from [Pd(NCPh)2Cl2], the appropri- observed between 2.85 V and 2.96 V. 12 z+ ate ligand precursor HB and NEt3. The homoleptic ruthenium(II) Finally, Colbran et al. showed [Ru(B )(bpy)L] (63 and 64)to 12 complex [Ru(B )2](58) was prepared by heating [Ru(dmso)4Cl2], be active electrocatalysts for the reductive disproportionation of 12 HB and NEt3 in methanol. A series of [Ru(B)(tpy)][PF6]x com- carbon dioxide into carbon monoxide and carbonate ion; the prod- plexes (59: B = B12, x =1;60: B = B20, x =1;61: B = B17, x =1;62: ucts were identified by GC–MS and NMR spectroscopy post- 21 B = B , x = 2) were prepared by heating HB, [Ru(tpy)Cl3] and electrolysis [74]. Catalysis occurs at the second reduction process 12 NEt3 in refluxing ethanol overnight, followed by anion metathesis (see Fig. 11) – that is [Ru(B )(bpy)] is the active catalyst. It was 12 0 with K[PF6]. Reaction of [Ru(dmso)4Cl2], HB and NEt3 with 2,2 - presumed that ejection of the chloride ion was required for [Ru bipyridine in hot toluene overnight produced [Ru(B12)(bpy)Cl] (B12)(bpy)Cl] (63) prior to carbon dioxide reduction.

(63), which upon treatment with K[PF6] in hot acetonitrile afforded Nomura and co-workers at Fujifilm Corporation claimed a series 12 [Ru(B )(bpy)(MeCN)][PF6](64). of ruthenium(II) complexes, very similar to complexes 59–62 The SC-XRD structure of [Pd(B12)Cl] (56) [74] is consistent with reported by McSkimming et al., as dyes for sensitized solar cells, the structures of 54 and 55 [69], see above. For the ruthenium(II) including complex 65, which was prepared as shown in Scheme 26, complexes, the RuAN(pyr) bonds were significantly shorter than and was characterised by mass spectrometry [113,114]. The RuAN(py) bonds for the (B12) ligands (1.95(1) Å versus 2.098(9) patented dyes all comprised Ru(II) ion bound to a donor ligand, Å respectively), which was attributed to the contribution of geo- (B1) in the case of 65, and to an acceptor ligand with surface- metric and electrostatic constraints. In 59, the RuAN(pyr) distance adsorptive groups (the carboxylic acid-derivatised terpyridine in was not significantly different to the RuAN(py)tpy distance for the 65) to allow immobilisation on semi-conductor electrodes such

N-atom of the central, transoid pyridine (1.965(6) Å versus 1.948 as TiO2. (8) Å, respectively), however the mean RuAN(py)B distance was Recently Yi and group have also prepared ruthenium(II) com- slightly longer than the corresponding ﬂanking RuAN(py)tpy dis- plexes of B anions, which have shown promising catalytic activity

James McPherson – June 2019 25 Chapter 1: Introduction

308 J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332

N X

N Ru

N i) NaH N Cl ii) [Ru(COD)Cl2]n N H N N Ru 67-69 N 95 % N

PhICl2 1 HB 91 % 66 N Cl Cl N Ru Cl N

Scheme 27. Synthesis of [Ru(B1)(COD)X]z+ complexes 66–70 [93].

66 76

86 96 70

Fig. 12. SC-XRD structures of [Ru(B)(COD)L]z+ complexes 66–70 reported by Yi and co-workers [93]; C-bound H-atoms, counter-anion, and solvent molecules are omitted for clarity, and, for 68, only one of the two crystallographically unique molecules is shown; for CCDC reference codes, see Table 9.

Table 9 Selected average bond lengths (Å) and inter-plane angles (°) for complexes 66–70 [93].

Complex (CCDC code) RuAN(py) RuAN(pyr) N(py)ARuAN(py0) N(py)ARuAN(pyr) 66 [Ru(B1)(COD)Cl] (AYOSUE) 2.177(3) 1.964 146.38(9) 74.26(9) 1 67 [Ru(B )(COD)(NO3)] (AYOWUI) 2.185(3) 1.964(3) 146.0(1) 74.4(1) 68 [Ru(B1)(COD)(OTf)] (AYOVUH) molecule 1 2.186(4) 1.960(4) 146.1(2) 74.4(2) molecule 2 2.177(4) 1.958(4) 145.9(2) 74.5(2) 1 69 [Ru(B )(COD)(OH2)]BF4 (AYOWES) 2.191(5) 1.960(6) 146.0(2) 74.4(2) 70 [Ru(B41)(COD)Cl] (AYOWAO) 2.188(3) 1.971(3) 146.7(1) 74.3(1)

in the oxidative cleavage of olefins [93]. Reaction of HB1 with NaH (Scheme 27). The later reaction is nice example of post- 1 and [Ru(COD)Cl2]n (COD = 1,5-cyclooctadiene) gave [Ru(B )(COD) coordination modification of the coordinated, but unadorned, Cl] (66) which was treated with silver salts to afford [Ru(B1) (B1) ligand. The structures of 66–70 were determined by SC- (COD)X] (67:X=NO3 ; 68: X = OTf ). Alternatively, treating 66 XRD studies, Fig. 12, and confirm the octahedral geometry about with Ag[BF4], followed by aqueous ammonium hexafluorophos- the ruthenium(II) centre in each complex. Again, the RuAN(py) 1 phate, furnished the aquo complex, [Ru(B )(COD)(OH2)][PF6](69), bonds are longer than the RuAN(pyr) bonds (ca. 0.2 Å), and the and reaction of 66 with iodobenzene dichloride cleanly formed average N(py)ARuAN(py0) wingtip angle of ca. 146° is slightly nar- the dichloride-substituted pyrrolide (B36) ligated complex (70) rower than observed for Colbran’s 58, 59 and 63.

James McPherson – June 2019 26 Chapter 1: Introduction

J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332 309

Overall:

R2 NaIO4 (2.5 equiv.) R3 66 (3 mol %) R1 R2 R3 R4 R1 + EtOAc/MeCN/H O (3:1:2) R4 2 O O 30°C, 2 h

O Cl O N Ru N N R2 R3 1 R1 R2 R3 R4 R + R4 O O NaIO4

N Cl

N Ru R1 R2 R3 R2 N R1 R3 O R4 O 66 O 4 Cl Cl O R N Ru N N Ru N N N

Scheme 28. Catalytic oxidation of oleﬁns by [Ru(B1)(COD)Cl] (66), and a proposed mechanism [93].

N N AgOTf N Ag N NEt3 N N Ag Ag

N N N 71

N N N N AgOTf PPh3 Ph P Ag Ag Ag PPh OTf- NH 3 3 N N N N 72

HB1 72 -

N Li(NSiMe3)2, AgOTf HP(O)(OEt)2 N N 2Li+ Ag Ag OTf- EtO OEt P P O O EtO OEt 73 73

Scheme 29. Preparation of multinuclear silver(I) complexes 71–73 from HB1 [101]; insets: SC-XRD structures (H-atoms and unbound anion are omitted for clarity; for CCDC reference codes, see Table 10).

James McPherson – June 2019 27 Chapter 1: Introduction

310 J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332

Table 10 Selected bond lengths (Å) and angles (°) from complexes 71–73 [101].

Complex (CCDC reference) AgAN(py) AgAN(pyr) AgAN(pyrbr)a Ag...Ag plane(py)Aplane(pyr) plane(py)Aplane(py0)

1 71 [Ag(l-B )]3 (RIRQIU) 2.41(1) – 2.33(1) 2.901(1) 19.69(5) 41.46(5) 2.47(1) 25.50(5) 1 72 [Ag{Ag(B )(PPh3)}2]OTf (RIRPOZ) 2.365(2) 2.216(1) – 3.1283(2) 10.80(8) 48.18(8) 2.128(2) 38.31(9) 1 73 Li2[Ag2(l-B )(PO(OEt)2)2]OTf (RIRPUF) 2.269(4) – 2.223(2) 2.8850(5) 12.8(2) 42.1(2) 30.2(2)

a pyrbr = bridging pyrrolide ligand.

2+ 74 H N AgOTf Br N N Br 2OTf- Ag Ag Br N N Br N H

+ Br 75 N O i) NaH, AgOTf ii) DPEPhos Ph2P PPh2 OTf- NH Br Br Ag Ag N N N Br N

HB9 75

i) AgOTf + Ph3P PPh3 ii) NEt3 iii) PPh Br Br 3 Ag Ag N N OTf- N 76 76

+ Ph3P PPh3 i) AgOTf ii) NEt3 R Ag Ag R iii) PPh 3 N N OTf- N N N H N R R HB 77-79

9 9 9 Scheme 30. Top. Preparation of [Ag2(l-HB )2](OTf)2 dimer (74) [101], [Ag2(l-B )(l-DPEPhos)]OTf (75) [99] and [{Ag(PPh3)}2(l-B )]OTf (76) [70]; insets: SC-XRD structures of 74–76 (with counter-anions and C-bound H-atoms omitted for clarity; for CCDC codes, see Table 11). Bottom: preparation of analogous [{Ag(PPh3)}2(l-B)] complexes (77: B = B10 (R = 6-phenyl); 78: B = B11 (R = 6-thien-2-yl); 79: B = B5 (R = 5-bromo)) [70].

Postulating that the 1,5-cyclooctadiene ligand would be labile left to continue, the yields improved significantly (from 38% after and so function as a protecting group for vacant coordination sites 2 h to 69% after 24 h for cyclohexane), indicating that the catalyst at the ruthenium(II) centre, the Yi and co-workers screened com- remained active over this time. ESI-MS spectra of the solutions plex 66 as a catalyst for the oxidation of a range of olefin substrates after catalysis detected [Ru(B1)(MeCN)(EtOAc)]+ as the parent [93] following the protocol established by Bera and co-workers for ion, indicating that the (B1) ligand remained bound to the ruthe- a similar ruthenium(II) system [115], Scheme 28. Styrene nium centre throughout the catalytic experiments. A cis-dioxo-RuVI derivatives with electron-donating groups performed better (95% intermediate was detected by ESI-MS spectrometry after combin- conversion after 1 h for 4-methoxystyrene) than those with ing [Ru(B1)(COD)Cl] with 80 equivalents of sodium periodate. electron-withdrawing groups (85% conversion after 1 h for 4- The mechanism proposed by Yi and co-workers is shown in fluorostyrene). When the reaction of less reactive substrates was Scheme 28 (bottom) [93], and follows Bera’s earlier proposal [115].

James McPherson – June 2019 28 Chapter 1: Introduction

J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332 311

Table 11 Selected bond lengths (Å) and inter-ring angles (°) for complexes 74–76, 78 and 79 [70,99,101].

Complex (CCDC reference) AgAN(py) AgAN(pyrbr)a Ag...Ag \plane(py)Aplane(pyr) \plane(py)Aplane(py0)

9 74 [Ag(l-HB )]2(OTf)2 (RIRQEQ) Ag1 2.277(3) – – 146.7(1) 23.2(1) 2.282(3) 147.1(1) 9 74 [Ag(l-HB )]2(OTf)2 (RIRQEQ) Ag3 2.246(3) – – 154.4(1) 19.0(1) 2.260(2) 170.9(1) 9 75 [Ag2(l-B )(l-DPEPhos)]OTf (JAGKIO) 2.320(4) 2.295(4) 2.859(1) 25.2(3) 48.5(4) 2.320(4) 2.295(4) 25.2(3) 9 76 [{Ag(PPh3)}2(l-B )]OTf (QECFEM) 2.369(3) 2.278(4) 2.9747(5) 14.9(2) 39.8(2) 2.257(4) 2.450(3) 28.3(2) 11 78 [{Ag(PPh3)}2(l-B )]OTf (QECFIQ) 2.393(2) 2.246(2) 3.0010(5) 9.0(1) 38.9(1) 2.391(2) 2.295(2) 29.9(1) 5 79 [{Ag(PPh3)}2(l-B )]OTf (QECFAI) 2.212(6) 2.550(4) 3.0668(7) 28.2(2) 50.3(2) 2.304(4) 2.267(4) 23.7(2)

a pyrbr = bridging pyrrolide.

R2 R3 N

79 (1.0 mol %) O R1 + R2 R3 + N H R1 H CH2Cl2,35°C

Scheme 31. A3-coupling of aldehydes and amines with phenylacetylene catalysed by the dinuclear silver(I)–B complexes as reported by Xiao-Yi Yi and group [70].

for their dipyridylpyrrolide ligands. Crystallographic metrics are presented in Table 10. For complexes 71 and 73, which involve a N N bridging pyrrolide group similar to in copper(I) complexes 27–33 (where the CuAN(py) bonds are signiﬁcantly shorter than CuAN N Ir N N Ir N A N N (pyr) bonds, see Table 5), the average of Ag N(pyr) bond lengths A N are the same as the Ag N(py) (2.36 vs. 2.37 Å, respectively), while N N N the twisting of the ring planes within the ligands is more pronounced (the average angle between the normals to each pyridyl plane are approximately 41.6°). In complex 72, for the trigonal pla- 08 18 nar Ag(I) centre the AgAN(pyr) bond is shorter than the AgAN(py) bond by 0.149 Å, whereas for the linear Ag(I) centre the AgAN(py) bond is shorter still (by a further 0.088 Å). Fig. 13. Iridium(III) complexes 80 and 81 of (B1) claimed for use in EL devices by Ikemizu and co-workers at Konica Minolta Holdings Inc [116]. Dinuclear silver(I) complexes 74–76 were also prepared with HB9, Scheme 30. First, treatment of HB9 with AgOTf afforded the 9 dimer, [Ag(l-HB )]2(OTf)2 (74), in which the two outer pyridine donors each bind a bridging silver(I) ion and the central pyrrole group is protonated and unbound [101]. When HB9 was ﬁrst N Z deprotonated with sodium hydride [99] (or when triethylamine is added to accept the pyrrolyl proton) [70], then dinuclear silver N Re Z 9 (I) phosphine complexes [Ag2(l-B )(l-DPEPhos)] (75) [99] and 9 [{Ag(PPh3)}2(l-B )] (76) [70] were prepared by adding the appro- N Z priate phosphine ligand to the mixture. This latter protocol was

employed to access three other analogous dimers [{Ag(PPh3)}2(l- B)] (77–79) with other B ligands as shown in Scheme 30, bottom Fig. 14. Generic structure of neutral bis(tridentate) rhenium(I) complexes of the [70]. Key metrics from the SC-XRD structures of 74–76, 78 and 79 1 (B ) anion, where Z is a neutral donor atom (C, N or P) claimed by Xia and Lin for are presented in Table 11.In74, the unbound pyrrole has twisted their use in electroluminescent materials [118]. away from the binding cavity (the angle between the normals to the pyridyl and pyrrole planes is ca. 155°), while complexes 75, Yi and co-workers have also prepared and characterised a range 76, 78 and 79 show data consistent with the trends already dis- of multinuclear silver(I) complexes, similar to the copper(I) com- cussed above for 71 and 73, the other Ag(I) complexes involving pounds 27–38 discussed above, shown in Scheme 29 [101]. Reac- bridging pyrrolide groups. For example, the average AgAN(py) 1 1 A tion of silver(I) triﬂate and HB , with NEt3 gave a [Ag(l-B )]3 and Ag N(pyr) bond distances, 2.32 Å and 2.35 respectively, are trimer (71) which, in all likelihood is the same compound as 14, similar.

first synthesised by Hein from silver(I) acetate [63]. With triph- Yi showed that the [{Ag(PPh3)}2(l-B)] complexes 76–79 are 3 enylphosphine, instead of NEt3, the reaction furnished the unusual robust, and are efficient and recyclable catalysts for so-called A - 1 trimer [Ag{Ag(B )(PPh3)}2](OTf) (72). Tweaking the conditions coupling reactions of aldehydes, amines and alkynes under mild again, now with HB1, AgOTf, diethylphosphite and lithium hexam- conditions, including open to air, in short reaction times, Scheme 31 1 ethylsilazide yielded Li2[{Ag(PO(OEt)2)}2(l-B )](OTf) (73). [70]. Modest substrate screening with 79 showed successful con- Complexes 71–73, along with the Cu(I) analogues, 27–33 above, version of phenylacetylene with a range of aliphatic secondary constituted the first examples to display ‘‘unusual” binding modes amines and aldehydes in less than one hour, while reactions

James McPherson – June 2019 29 Chapter 1: Introduction

312 J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332

M N N N N N N N N N M M M I II III

M N N N M N N H N N N N M M M M IV V VI

Fig. 15. Possible coordination modes for di(2-pyridyl)pyrrolide (B) ligands.

Table 12 Aggregate average MAL bond lengths (Å) and LAMAL0 bond angles (°) for Type I dipyridylpyrrolide (B)-transition metal ion coordination complexes grouped according to metal ion and coordination geometry.a

Metal ion Geometryb MAN(py) MAN(pyr) N(py)AMAN(py0) N(pyr)AMAN(py) Complex(es) & Reference(s) Cu(II) SPl 2.145(3) 1.863(3) 153.7(1) 77.0(1) 23 [110] Pd(II) SPl 2.09 1.87 156 78.0 54 and 55 [69]; 56 [74] Pt(II) SPl 2.057(4) 1.892(4) 156.4(2) 78.2(2) 39 [69]

Zn(II) Td 2.206(5) 1.923(6) 145.4(2) 79.0(2) 9 [62]

Fe(II) Oh 2.275(8) 2.207(8) 146.2(3) 73.1(3) 15 [62]

Co(II) Oh 2.28 1.95 148 73.9 7 and 16 [62]

Co(III) Oh 1.963(8) 1.810(8) 160.0(3) 80.0(3) 22 [90]

Ni(II) Oh 2.22(2) 1.96(1) 151.3(5) 75.7(4) 17 [62]

Cu(II) Oh 2.265(7) 1.892(7) 150.3(3) 75.2(3) 18 [62]

Zn(II) Oh 2.36 1.94 148 73.9 9 and 19 [62]

Ru(II) Oh 2.16 1.96 148 75.0 58, 59 and 63 [74]; 66–70 [93]

a ESDs are listed where the average is for one SC-XRD structure only. b Geometry codes: SPl = square planar, Td = tetrahedral, and Oh = octahedral.

Table 13 Average MAL bond lengths (Å) and LAMAL0 angles (°) in a mononuclear complex with a Type II, bidentate B ligand.a

Metal ion Geometryb N(py)AM (Å) N(pyr)AM (Å) N(py)AMAN(pyr) (°) plane(py)Aplane(pyr) (°) plane(py)Aplane(py0)(°) Reference(s)

Cu(I) Td 2.16 2.05 91.0 8.8 166 34–38 [99]

a ESDs are listed where the average is for one SC-XRD structure only. b Td = tetrahedral.

Table 14 Aggregate average MAL and M...M distances (Å) and inter-plane angles (°) for multinuclear complexes with bridging B ligands grouped by metal ion and ligand coordination mode (see Fig. 15).a

Metal ion Coordination MAN(py) MAN(pyr) M...M/ plane(py)Aplane(pyr) plane(py)Aplane(py0) Reference(s) Mode Cu(II) III 2.103(6) 1.952(6) 2.815(1) 8.2(3) 33.9(3) 26 [102] Ag(I) III 2.365(2) 2.216(1) 3.1283(2) 10.80(8) 48.18(8) 72 [101] Ag(I)b III 2.128(2) – – 38.31(9) – 72 [101] Ag(I) IV 2.266(3) – 6.842(2) 157.8(1) 21.1(1) 74 [101] Cu(I) V 2.06 2.20 2.72 18.3 34.7 27 [101]; 28–33 [100] Cu(I)/Na+ V 2.437(1) 2.675(1) 2.893(1) 21.15(6) 40.60(6) 27 [101] Cu(II) V 2.01 2.23 2.93 16.9 43.9 24–26 [102] Ag(I) V 2.35 2.35 2.94 22.7 43.0 71 and 73 [101]; 75 [99] and 76, 78 and 79 [70] Fe(II) VI 2.154(2) 2.135(1) 3.0539(4) 77.27(6) 35.96(6) 21 [92]

a ESDs are listed where the average is for one SC-XRD structure only. b Bridging (linear) silver(I) centre.

involving benzaldehyde or N,N-dibenzylamine required signiﬁ- for their use as materials for electroluminescent (EL) displays cantly longer times (1.5–2 h) to achieve more than 90% conversion. [116]. While there are some experimental data for devices con- Ikemizu and co-workers included two complexes of (B1), 80 structed with 80, unfortunately no preparative details are available and 81, Fig. 13, along with thirty-two other compounds in a claim for either 80 or 81. In a previous ﬁling of the patent, Tamaru et al.

James McPherson – June 2019 30 Chapter 1: Introduction

J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332 313 claimed many other examples for similar iridium(III) complexes Yi’s group have also reported several mononuclear examples of involving a combination of a monoanionic ligand and a dianionic B binding in a bidentate, type II, fashion, and a summary of the tridentate ligand (such as B and C2, respectively) [117]; see also crystallographic data available is presented in Table 13. The Cu Section 3.2. (I)AN(pyr) bonds are again shorter than Cu(I)AN(py), however this Finally, Xia and Lin have claimed more than eight thousand difference is less exaggerated than for the type I complexes. These hypothetical rhenium(I) complexes as electroluminescent materi- complexes were air stable and luminescent in dichloromethane als [118]. A generic structure for these complexes is shown in solution. Yi’s multinuclear Ag(I) complexes all display longer Fig. 14, and has an anionic tridentate ligand and one neutral triden- MAN bond lengths than the Cu(I) and Cu(II) analogues of the same tate ligand. Ninety-five different anionic ligands were proposed, binding type, which is accommodated by more pronounced twist- including the (B1) anion, along with eighty-six different neutral ing of the ring planes within each ligand in the Ag(I) species ligands to give a total of 8170 combinations of rhenium(I) com- (Table 14). For the type III mode non-bridging pyrrolide, the plexes. The patent claims that current rhenium(I) phosphors in MAN(pyr) bonds are shorter than MAN(py) bonds, but for type V the literature contain monodentate ligands (usually carbonyls), binding modes with bridging pyrrolide rings, the MAN(pyr) dis- and so their stabilities do not meet commercial requirements. tances are longer than the MAN(py) bonds for the Cu complexes The proposed bis(tridentate)rhenium(I) complexes are expected but about the same length for the Ag(I) complexes. Finally, Wol- to be more stable. Other than DFT calculations performed on czanski’s iron(II) dimer 21, with the two electron reduced 40 selected complexes (however, none of those involving B ), no (H2B ) displayed a new binding mode, type VI. Presumably the experimental data is available for the complexes proposed by the large twisting of the ligand rings (the angle between the normals inventors. to the pyridyl and dihydropyrrolide planes is 77.27(6)°) is necessi- tated by the reduced 2,5-dihydropyrrolide core. 2.3. Summary Where physical properties are described for the transition metal–dipyridylpyrrolide complexes, they are always consistent p A diverse assortment of HB derivatives is now available, with with the tridentate B anion being a strong -donor, weak-field substitution at either the pyrrole or pyridyl (or both) possible. A ligand. For example, the experimental studies of the homoleptic number of routes are available—predominantly Owsley’s route to iron(II) complexes reveal these are high spin at all temperatures dipyridylbutan-1,4-diones and subsequent Paal–Knorr pyrrole [62,63,104], and theoretical predictions accord with this result condensation have been preferred by the Tour and Yi groups. Jones [105]. Metal-centred redox processes when observed occur at has demonstrated that the Stetter reaction also provides a short, lower potential than the corresponding terpyridyl complexes and efficient route to these precursors. Two one-pot routes are also the potentials are sensitive to the electronic nature of substituents available: Colbran’s direct condensation which affords a range of at the pyrrole [62,74]; this accords with strong donor character for substituted HB from commercial starting materials in modest the pyrrolide ring [51] tempered by its substitution. yields, and Zeitler’s sequential nitro-Stetter and Stetter reactions followed by Paal–Knorr condensation, which has achieved good 3. The 2,6-di(2-pyrrolide)pyridine dianion (C2) to excellent yields for two derivatives. Finally, reductive ring contraction (particularly by electrochemical reduction) of Up to the time of submission, searches of the CAS SciFinderTM dipyridylpyridazines to dipyridylpyrroles has been reported by database for core C (substitution and charge unspecified) found Dubreuil’s group – and has promising supramolecular applications 201 substances described in 71 publications (which includes cor- in and of itself. rections and conference papers not cited in this review). Some of Coordination complexes of the B anion with a range of transi- these substances are oligomers or macrocycles, and are excluded tion metals have been described. While most complexes involve from this review (they will be the subject of a separate article conventional N3-binding of the metal centre (type I in Fig. 15), [60]). other coordination modes have been reported and are summarised in Fig. 15. These can be grouped as follows: Types I and II represent mononuclear species; types III and IV are dinuclear species with N N each donor N-atom only coordinated to (at most) one metal centre; N and types V and VI are dinuclear species in which the central pyrrolide N-atom bridges both metal ions. Other coordination modes (for example, monodentate B binding only to pyrrolide-N, or C cyclometalated B) have not yet been reported. Conventional, tridentate type I binding of the B anion is most widely encountered to date. A summary of the crystallographic data available for such complexes is presented in Table 12. The C wider bend angle induced by the five-membered pyrrolide ring 3.1. Syntheses and properties of 2,6-di(2-pyrrolyl)pyridines (H2 ) of a B ligand compared to the bend angle for terpyridine with its six-membered central pyridyl ring results in significantly 3.1.1. Fischer indole synthesis shorter MAN(pyr) than MAN(py) bonds. This is most pronounced The Fischer indole synthesis is the condensation of a substituted in the square planar complexes (MAN(pyr) 1.88 Å vs. MAN(py) phenylhydrazine and a carbonyl compound under acidic 2.10 Å on average). Tour’s octahedral iron(II) complex 15 stands conditions to afford an indole. The first compound with a out with a particularly long FeAN(pyr) bond at 2.207(8) Å as antic- pyrrole–pyridine–pyrrole skeleton prepared using Fischer route 1 ipated for the high-spin centre, although it is worth noting that this was the parent 2,6-(diindolyl)pyridine H2C first described by complex was unstable and only partially characterised. For octahe- BuHoi et al. in 1966, which was obtained by cyclisation of the dral complexes of divalent metal ions, Ru(II) complexes are the dihydrazone formed from 2,6-diacetylpyridine, Scheme 32 [119]. 1 only examples with second- or third-row metal ions characterised Thummel et al. also prepared H2C (89% yield) and extended by SC-XRD structures and these all have shorter MAN(py) bonds this route to the syntheses of annulated di(indolyl)pyridines 2 4 (2.16 Å) compared to their high-spin Fe(II) (>2.2 Å). H2C –H2C , Scheme 32 [120]. This route has also been exploited

James McPherson – June 2019 31 Chapter 1: Introduction

314 J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332

N N NH HN + O O PhNHNH2, H N N PPA, ∆ N N 1 H H H2C

(CH2)n (CH )n (CH2)n (CH2)n (CH2)n (CH2)n 2 N N N + NH HN O O PhNHNH2, H N N PPA, ∆ N N H H 2 (n = 1) H2C 3 (n = 2) H2C 4 (n = 3) H2C

Scheme 32. Fischer syntheses of 2,6-di(indolyl)pyridines [119–121].

CO (40 atm), toluene, 160 °C N Pd (mesitylCO ) N 2 2 2 NH HN

NO2 O2N 1 H2C N N

1 Scheme 33. Pd-catalysed ring closure under CO of a nitrostyryl precursor as a route to diindolyl-pyridine H2C [123].

PhSO2Cl, NaOH, 1. LiN(i-Pr)2, THF, °C 2. ZnCl2, THF, 25 °C NBu4Br ZnCl N N N N H N N SO2Ph SO2Ph

Br N Br N NH HN Pd(PPh3)2Cl2, DIBAH; N N then NaOH, EtOH 5 H2C

5 Scheme 34. Negishi cross-coupling route to di(7-azaindolyl)pyridine, H2C [124].

by Thummel’s group to prepare larger, oligomeric linked pyridyl- cles [125–127]. In a report on the use of Pd-catalysed cross- pyrrolyl systems [121,122]. coupling to provide bis(pyrrolyl)arenes, Setsune et al. reveal Suzuki coupling of a pyrrole boronic ester with 2,6-dibromopyridine using

3.1.2. Metal-assisted couplings Pd(OAc)2/PPh3/K2CO3 catalysis affords the target dipyrrolepyridine, 6 Tollari et al. reported the ﬁrst such syntheses in 1998: pyridine- H2C , in 74% yield, Scheme 35 [125]. Zhang et al. subsequently 1 6 centred H2C (87% yield) was prepared from readily accessible reported an improvement of the yield of H2C to 85% when using o-nitrostyryl precursors by closing the rings with CO (40 atm) Pd(OAc)2/K2CO3 catalysis [127]. Hydrolysis of the ester using aque- 7 using Pd2(mesitoylate)2 and 3,4,7,8-tetramethyl-1,10-phenanthro ous NaOH, then neutralisation, afforded diacid-substituted H4C in 7 line as catalyst at 160 °C, Scheme 33 [123]. 88% yield. H4C is a useful precursor in the manufacture of larger, Wang et al. employed a Negishi cross-coupling to obtain self-assembled coordination cages [127]. Setsune has also 5 7 di(7-azaindolyl)pyridine H2C from indole and dibromopyridine, described the smooth decarbonylation of H2C in ethylene glycol 8 6 Scheme 34 [124]. Studies of luminescent tin and lead complexes at 180 °C to afford H2C (91% overall yield from H2C ) [126]. The 5 2 9 8 of the (C ) anion are summarised in Section 3.2. dialdehyde H2C is available from H2C under Vilsmeier conditions 8 Suzuki coupling of a bromo-pyridine and boryl ester- (94% yield) [126]. Treatment of H2C with I2 and KI afforded diiodo- 10 substituted pyrroles is another obvious route to dipyrrolepyri- substituted H2C (68% yield), and Suzuki coupling of this with 11 dines. Sessler and Setsune have often employed this methodology pinacolborane gave H2C (91% yield), Scheme 35, which is ready to obtain intermediates in their elegant syntheses of the macrocy- for subsequent cross-coupling reactions [127]. James McPherson – June 2019 32 Chapter 1: Introduction

J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332 315

Pd(OAc)2, K2CO3 O + N EtO C B 2 N Br N Br NH HN H O EtO2C 6 CO2Et H2C NaOH (aq)

180 °C, (CH2OH)2 I2/KI N N N NH HN NH HN NH HN HO H I I 2C CO2 8 7 H2C 10 H2C H2C O BH O POCl , DMF 3 Pd(PPh3)2Cl2, NEt3

N N NH HN NH HN

OHC CHO O B B O 9 O 11 O H2C H2C

Scheme 35. Dipyrrolepyridines from palladium-catalysed cross-coupling reactions [125–127].

R O 12 H C OCH I 1. NEt3, CuI, PdCl2(PPh3)2 N 2 3 + 2 O 13 H C OH N 2. CuI, NMP, 190 °C R NH HN R 4 14 H N C 2 H4 NH(CH2)3NMe2 O O 15 H4C NH(H2C)3N

Scheme 36. Syntheses of DNA G-quadruplex binders [128,129].

O HN S CF3 F3C O

CF 1. p-toluenesulfonamide 3 N toluene 1. Pd(PPh3)2Cl2, K2CO3 DOWEX, Br N Br 120 °C, 18 h, 81 % DMF, 80 °C, 24 h NH HN HN N 2. Pd(PPh3)2Cl2, CuI, piperidine Br O S O 2. addition of NaOH, DMF, 40 °C, 24 h, 84 % DMF, 80 °C, 3 h CHO activated Zn, THF 14 % over 2 steps 0-25 °C, 5h, 95 % F C CF 3 16 3 O H2C F3C HN S O

16 Scheme 37. Hill and co-workers synthetic route to H2C . Adapted from Ref. [131].

12 Balasubramanian and co-workers employed Songashira cou- diindolylpyridine H2C (Scheme 36) [128,129]. Straightforward pling of 2,6-diethynylpyridine with 4-methylacetoxy-2- hydrolysis with acid afforded the carboxylic acid-substituted, 13 iodoaniline then CuI-catalysed ring closure of the dialkynyl pyri- H4C , and standard peptide coupling protocols were employed 12 dine intermediate (an intramolecular Ullmann reaction) to obtain to couple H2C with a range of amines to provide carboxamides, 14 15 14 15 good yields of the corresponding methylacetoxy-substituted H4C and H4C , in good yield. Carboxamides H4C and H4C

James McPherson – June 2019 33 Chapter 1: Introduction

316 J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332

H H EtO2C EtO2C Br N N H H EtO2C NH2 O O N N H H pyridinium tribromide H NH2 EtO2C N EtO2C N EtO2C N EtO2C N THF, 12 h THF/NEt3 H H H O O N N H H Br N N

EtO2C H EtO2C H 17 H6C

4 -1 Ka = 2.72 x 10 M 1,3-cyclohexanedionate

CO2Et

EtO2C N CO2Et N N H N H H N H H O O H + - 1:1 [H7C17] :picrate

17 17 Scheme 38. Preparation of H6C , the proposed 1:1 host:guest complex formed with 1,3-cyclohexanedionate anion, and the SC-XRD structure of the 1:1 complex of H6C and picric acid (inset: CCDC code YUGVOL) with intermolecular H-bonding interactions shown, and C-bound H atoms omitted for clarity [133].

were found to bind tightly to and stabilise DNA G-quadruplexes O NH [128,129]. Subsequently, Dash and co-workers cleverly demonstrated several logic operations such as XNOR, NOR, AND, NAND N N N and NOT based on the pH-dependent binding to the c-kit2 G- EtO2C CO2Et 14 quadruplex DNA promotor sequence by H4C [130]. In the exper- N O NH iments, pH was used as an external modulator and thiazole orange NH HN (TO) and the c-kit2 promotor sequence as two inputs and the fluo- NC 14 H C18 H C19 rescence of H4C and TO as the outputs. 2 2 More recently, Hill and co-workers have reported a study of 16 18 19 arylpyrrole oligomers, including dipyrrolepyridine H2C ,as Fig. 16. Dipyrrolepyridines, H2C prepared by Corriu et al. [134] and H2C patented by Griffith et al. [135]. receptors for various anions. Scheme 37 summarises their 16 synthetic route to H2C . DOWEX assisted condensation of p-trifluoromethylbenzaldehyde with p-sulfonamide and the subse- H H quent Barbier reaction afforded the required homopropargyl LiOH N sulfonamide for Songashira coupling with dibromopyridine. N DMSO 140 °C N N Cyclisation with Pd(PPh3)2Cl2 as catalyst with K2CO3 as base then H H 16 N N yielded the targeted dipyrrolepyridine H2C . Whereas analogues HO OH 17 % O 20 with a central aryl group were good anion receptors, H C16 was a S H2C ·dmso 2 poor anion receptor which was attributed to the destabilising

20 interaction of the N(py) lone pair with the anion [131]. Scheme 39. Synthesis of H2C dmso by Troﬁmov and co-workers [137].

O Ph Ph Ph Ph Ph R NH4OAc N N N EtOH thiazolium salt O O NH HN O O t-BuOK/EtOH R R O O R R 21 H2C (R=Ph) 22 H2C (R=Me)

21 Scheme 40. Stetter condensation of 2,6-di(carboxaldehyde)pyridine with chalcone followed by ring closure of the 1,4-diketone with ammonium acetate to affordH2C (67% 22 overall yield) [139] or H2C (31% yield in one pot) [49].

James McPherson – June 2019 34 Chapter 1: Introduction

J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332 317

O O HOTs (60 mol%) + 2 N N tBu tBu mesitylene ,∆, 4 days NH2 NH2 NH HN

23 H2C

23 Scheme 41. Synthesis of H2C adapted from Ref. [141].

N F F F C CF 3 3 NH HN

N N N N NH NH HN NH HN NH HN NH HN 24 25 26 27 28 H2C H2C H2C H2C H3C

24 25 26 27 28 Fig. 17. Structures of H2C and H2C [142]; and H2C and H2C [117]; and H3C [143].

R adducts of this compound, and the SC-XRD structure of the 1:1 20 complex of H2C with dmso were also obtained. Nagata and Tanaka prepared the tetraphenyl-substituted 2,6-di 21 N (pyrrole)pyridine, H2C (Scheme 40), as part of their study of a ruthenium(II) complex of (D1) (see Section 4.2), however they NH N Re 21 H do not discuss any investigations into the use of H2C as a ligand (CO)3Br 21 [139]. Their preparation of H2C serves to demonstrate the gener- (R = H, Ph, py) ality of their synthetic protocol, and shows that the Stetter reaction 82-84 (see Section 2.1, above) also provides a useful entry to substituted- dipyrrole pyridines. Indeed, their synthetic route was successfully 20 24 25 22 Fig. 18. Putative [Re(CO)3(H2C)Br] (82:H2C (R = H); 83:H2C (R = Ph); 84:H2C adapted by Milsmann et al. to prepare H2C from the correspond- (R = 4-py)) complexes studied by Moya et al. [142]. ing enone (Scheme 40) [49]. Alternatively, Caulton’s group adapted the single-pot pyrrole synthesis reported by McNeill and co-workers [140] to obtain the 3.1.3. Miscellaneous tetra-(t-butyl) substituted derivative H C23 in 40% yield Various other routes to molecule with a di(pyrrole/indole)pyri- 2 (Scheme 41) [141]. Although the yield is moderate, the ready avail- dine core appear in the literature. Anslyn and co-workers reported ability of a wide range of diketones [140] and di(aminomethyl)pyr- in 1991 the synthesis of H C17, Scheme 38, using a very different 6 idine precursors suggests the McNeill route could provide synthetic strategy than discussed in this article so far. The synthe- convenient access to many dipyrrolylpyridines. Indeed, modular sis started with the bromination of ethyl 4,5-dioxo-octahydroacri syntheses of dipyrrolylpyridine pincers with inequivalent arms dine-9-carboxylate using pyridinium bromide perbromide, fol- can be imagined. lowed by alkylation with ethyl 3,3-diaminoacrylate to provide 17 17 H6C in 58% overall yield. H6C was found to stabilise enolate anions by up to 1 pKa unit in acetonitrile through host:guest com- 3.1.4. Other dipyrrole–pyridine compounds 24 25 plex formation [132]. The binding affinities of several enolates Despite reporting rhenium(I) complexes of H2C and H2C were subsequently quantified, and 1,3-cyclohexanedionate was (Fig. 17, see Section 3.2), Moya et al. included no details on the 4 1 found to bind most tightly (Ka = 2.72 l0 M in acetonitrile with preparation of the ligand precursors themselves [142]. Similarly, + 17 Na @[2.2.1]-cryptand as counterion) [133]. Although, a H6C host- Tamaru and co-workers provided no preparative details for any 26 27 enolate guest complex structure was never crystallographically of the complexes or ligands (such as H2C and H2C , Fig. 17) that determined, it was presumed that in each case a hydrogen bonding they claimed for use in electroluminescent devices [117]. Finally, 28 2 network formed similar to that found in the complex between Mindiola et al. prepared (HC ) directly bound in a cobalt(II) 17 H6C and picric acid, Scheme 38 (inset) [133]. complex (108, see Section 3.2) [143]. Also in 1991, Corriu et al. revealed the reaction of lithium „ methyl(hexynyl)copper with (Me3Si)2NCH2C CCO2Et followed 3.2. Coordination compounds of 2,6-di(2-pyrrolide)pyridine dianions by treatment with pyridine-2,6-dicarbonyl chloride provided a (C2) 18 facile entry point to the 2,6-dipyrrolylpyridine H2C , which was employed for subsequent syntheses of macrocycles and metal In this section, metal complexes of the 2,6-di(2-pyrrolido) 19 2 complexes [134]. In a 2001 patent, Griffith et al. claim H2C , pyridine dianion (C ) or close derivatives are described. Pyri- Fig. 16, among 575 other compounds, as an inhibitor of a1b2- dine–pyrrole oligomers or macrocyles with a C subunit are mediated cell adhesion [135]. In 2005 Luybov and co-workers excluded, as they will be reviewed elsewhere [60]. Since there 20 reported H2C is available upon heating 2,6-diacetylpyridine diox- are few complexes of ligand C, we start at the beginning, at the first ime with MOH/dmso (M = Li, K) at 80–140 °C under 20–30 atm of appearance and consider them in the order they appeared until a acetylene in low (<20%) yields, Scheme 39 [136–138]. N-Vinyl clear reason to buck chronology arises. James McPherson – June 2019 35 Chapter 1: Introduction

318 J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332

N N Pd N 1 [Pd(py)2]+Li2C N

N N Pt N 1 [PtMe2(SMe2)2] +Li2C S 87

py 86

N N Pt N N

1 1 Scheme 42. Preparation of palladium(II) and platinum(II) complexes of H2C ; Inset: SC-XRD structure of the [Pt(C )(py)] (86, CCDC reference code: XOVFUJ) [144].

Table 15 Key average bond lengths (Å) and angles (°) from SC-XRD structures of Pd(II) and Pt(II) complexes reported by Wang and co-workers [144].

Complex (CCDC reference) MAN(ind) MAN(py) \N(ind)AMAN(ind0) \N(ind)AMAN(py) 85 [Pd(C1)(py)] (XOVFIX) 2.010(3) 1.964(3) 161.6(1) 80.9(1) 86 [Pt(C1)(py)] (XOVFUJ) 2.03(3) 1.99(3) 162(2) 81(1) 1 87 [Pt(C )(SMe2)] (XOVFOD) 2.06(1) 1.97(1) 160.3(4) 80.3(4)

N i) n-BuLi, –78 °C N

NH HN ii) MCl2Ph2 N N M X X X X 88 88:M=Sn,X=CH 89:M=Sn,X=N 90:M=Pb,X=CH 91:M=Pb,X=N

1 Scheme 43. Synthesis of complexes 88–91; Inset: SC-XRD structure of [Sn(C )Ph2](88, CCDC reference code: ALILEM) [124].

The ﬁrst mention of 2,6-di(2-pyrrolido)pyridine metal com- couple, and the primary oxidation observed at +0.91 V was plexes in the literature was in 1999. Moya et al. reported measure- assumed Re-centred. 20 24 ments for [Re(CO)3(H2C)Br] (82:R=H,H2C ; 83: R = Ph, H2C ; In 2002, Wang and co-workers described a detailed preparative 25 1 and 84: R = 4-py, H2C ) complexes [142] in a study of correlations and photophysical study of bis(indolyl)pyridine (H2C ) and the between the oxidation potentials and carbonyl stretching frequen- complexes [M(C1)(py)] (85: M = Pd; 86: M = Pt) and [Pt(C1) 8 cies for a series of polypyridyl complexes bound to the tricarbonyl (SMe2)] (87); Scheme 42 [144]. As anticipated for d Pd(II) or Pt bromorhenium fragment. No preparative or characterisation (II) ions, SC-XRD structures of the three complexes (Table 15) details were given for these species, which is unfortunate, as the revealed square planar metal centres; e.g., Inset, Scheme 42. In con- denticity and protonation of the di(pyrrole)pyridine ligands is trast to the square planar complexes of B (39, 54–56) discussed uncertain in each case, and is unlikely to be as presented in the earlier, the MAN(ind) bonds for 85–87 are on average ca. 0.06 Å paper, e.g., Fig. 18. In acetonitrile solution, the primary reduction longer than the MAN(py) bonds and, as a further consequence of +/0 0/ 0 at 1.23 V vs. Fc was attributed to the [Re(CO)3(H2C)Br] the central pyridyl ring, the average wingtip N(ind)AMAN(ind )

James McPherson – June 2019 36 Chapter 1: Introduction

J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332 319

Table 16 Key average metric data from the SC-XRD structures of Sn(IV) and Pb(IV) complexes reported by Wang and co-workers [124].

Complex (CCDC reference) MAN(ind) (Å) MAN(py) (Å) \N(ind)AMAN(ind0)(°) \N(ind)AMAN(py) (°)

1 88 [Sn(C )Ph2](tol) (ALILEM) 2.163(5) 2.161(6) 148.2(2) 74.4(2) 2.180(5) 73.8(2) 5 89 [Sn(C )Ph2](CH2Cl2)(ALILIQ) 2.157(9) 2.224(6) 145.8(2) 72.5(2) 2.176(7) 73.3(2) 1 y 90 [Pb(C )Ph2](CH2Cl2) (ALILOW) 2.295(3) 2.294(4) 142.4(1) 71.2(1) 2.317(3) 71.1(1) 2.316(4) 2.299(4) 141.6(1) 70.9(1) 2.303(5) 70.7(1) 1 y 90 [Pb(C )Ph2](tol) (ALILUC) 2.297(5) 2.310(6) 141.8(2) 71.2(2) 2.284(6) 70.6(2) 2.298(5) 2.298(3) 141.2(2) 70.7(2) 2.312(6) 70.5(2) 5 y 91 [Pb(C )Ph2](H2O) (ALIMAJ) 2.39(1) 2.37(1) 138.3(4) 70.5(4) 2.365(8) 68.0(4) 2.37(1) 2.34(1) 138.5(4) 69.3(4) 2.37(1) 69.2(4) y Two unique [Pb(C)Ph2] molecules in each structure.

CF3 F F N N N N N N N N N CF3 F Ir F Ir Ir N N N N N N N N N

F F 29 39 49

Fig. 20. Patented electroluminescent iridium complexes [117].

Table 17 Key bond length (Å) and angle (°) data from the SC-XRD structures of the Ti and Zr complexes 95–97 [49].

Complex (CCDC reference) MAN(pyr) MAN(py) \N(pyr)AMAN(pyr0) \N(pyr)AMAN(py)

22 y 95 [Ti(C )2](THF) (YAHCAO) 2.058(3) 2.122(2) 147.6(1) 74.2(1) 2.020(3) 74.1(1) 22 96 [Zr(C )2](CH2Cl2)(YAHBIV) 2.150(3) 2.288(3) 140.5(1) 70.3(1) 2.183(3) 70.6(1) 2.143(3) 2.299(3) 139.5(1) 70.0(1) 2.172(2) 70.3(1) 22 97 [Li(THF)4][Ti(C )2](toluene) (YAHBUH) 2.115(3) 2.127(2) 149.3(1) 74.75(9) 2.110(3) 74.91(9) 2.119(2) 2.131(3) 149.5(1) 74.10(9) 2.107(2) 75.45(9) y Both ligands are equivalent by crystallographic symmetry.

James McPherson – June 2019 37 Chapter 1: Introduction

320 J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332

22 Fig. 21. Milsmann’s Earth abundant SET photosensitizer [Zr(C )2](96), showing in the left panel its cyclic voltammogram, in the centre a view from its SC-XRD structure (CCDC reference code: YAHBIV also shown on the right), around the centre a cycle illustrating its use as a SET photocatalyst for oleﬁn reduction (and coupling – not shown), and in the right panel its absorption and emission spectra; reprinted (adapted) with permission from Ref. [49], copyright (2017) American Chemical Society.

(C7)2; 90: M = Pb, C =(C1)2; 91: M = Pb, C =(C5)2), Scheme 43 and Table 16 [124]. Again, the metric parameters show the effect of a central pyridyl donor, with little difference in the terminal MAN(ind) lengths (2.29 Å on average) and the central MAN(py) lengths (also 2.29 Å on average), while the wingtip N(ind)A MAN(ind0) angles (138–148°) being slightly narrower than for the square planar complexes 85–87 discussed above. Interesting solid-state inter-ring p-stacking was observed. At ambient temperature, the Sn complexes showed strong ligand centred fluorescence whereas the Pb complexes were not luminescent, which was expected since the heavier element should lead to more efficient fluorescence quenching. At 77 K, both the Sn and Pb complexes max exhibited blue-green fluorescence (kem 460–490 nm, s < 7 ns) max and red-orange phosphorescence (kem 550–560 nm, s 10–20 ⁄ ms) that involved ligand centred p n transitions, e.g., Fig. 19. The group 14 elements thus enhanced the phosphorescent emission of the ligands, which combines with the high chemical stabil- 23 Fig. 22. The SC-XRD structure of [(C )Pt(NCPh)] (99, CCDC reference code: ity of the complexes to suggest potential use for the complexes in VIWTEC) [141]. Key bond lengths and angles are summarised in Table 18. devices such as as emitters in OLEDS. Related to the theme of the previously described work, a 2007 angle of 161° is 5° wider than for complexes 39, 54–56 with cen- patent claimed that a number of iridium complexes, including tral pyrrolide donors. In emission studies, the Pd complex 85 was the 2,6-di(pyrrolido)pyridine iridium complexes 92–94 depicted found to be non emissive, whereas the Pt congener 86 was a bright in Fig. 20, are highly phosphorescent and well suited for use as orange emitter, both in solution and in a polymer (PVK) matrix (86 electroluminescent materials for optical displays and lighting at 585–590 nm depending on conditions). On the basis of devices [117]. extended Hückel calculations, the luminescence was attributed to Much more recently, Milsmann et al. reported the synthesis and ⁄ 22 ligand centred p p transitions with a significant contribution characterisation of neutral Ti(IV) and Zr(IV) [M(C )2](95: M = Ti; from the dp(Pt) to the p(L) level. Several EL devices using 86 as 96: M = Zr) complexes bearing two 2,6-di(pyrrolido)pyridine the emitter were fabricated on ITO substrate by combination of ligands [49]. The metric parameters from SC-XRD studies of 95– spin-casting and vacuum deposition of layers. The double-layer 97, which are summarised in Table 17, show shorter terminal optimal device had the structure ITO/PVK (hole transport & host) MAN(pyr) bonds than the central MAN(py) (by 0.01–0.2 Å). The + 86 (50 nm)/PBD (30 nm)/LiF (1.5 nm)/Al (150 nm) with the wingtip N(pyr)AMAN(pyr0) angles (139–150°) and acute N(pyr)A active device area being 1.0 5.0 mm2. The device emitted in the MAN(py) angles (70–75°) are similar to those of 88–91 (Table 16). 1 22 red-orange region with the highest efficiency being 0.31 cd A Complex 95 was prepared via [Li(THF)4][Ti(C )2](97), which was 2 achieved at 10 V and 20 mA cm . then oxidised with half an equivalent of I2. Complexes 95 and 96 In 2003, Wang and co-workers detailed the preparation and SC- showed intense LMCT bands in the visible region and underwent XRD structures of the five-coordinate phenyl-tin and phenyl-lead at least three reversible one-electron reductions under highly 1 2 complexes, [M(C)Ph2](88: M = Sn, C =(C ) ; 89: M = Sn, C = reducing conditions; see for example Fig. 21. The Zr(IV) complex

Table 18 Key bond length (Å) and angle (°) data from the SC-XRD structures of four-coordinate Pt(II) and Zn(II) complexes, 99 and 102, respectively [141].

Complex(CCDC reference) MAN(pyr) MAN(py) \N(pyr)AMAN(pyr0)(°) \N(pyr)AMAN(py)

99 [(C23)Pt(NC-Ph)](benzene) (VIWTEC) 2.061(2) 1.950(2) 161.1(1) 80.78(9) 2.053(2) 80.3(1) 102 [(C23)Zn(dmap)] (VIWSIF) 2.004(1) 1.988(1) 146.79(5) 81.19(5) 1.987(1) 81.89(5)

James McPherson – June 2019 38 Chapter 1: Introduction

J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332 321

Fig. 23. The SC-XRD structures of [(HC23)ZnEt] (101, CCDC reference code: VIWTAY) and [(C23)Zn(dmap)] (102, CCDC reference code: VIWSIF) [141]. Key distances (Å) and angles (°) for 101:ZnAN(pyr) 1.946(1), ZnAN(py) 2.074(1), N(pyr)AZnAN(py) 80.79(6), N(pyr)AZnAC(ethyl) 151.9(3), N(py)AZnAC(ethyl) 124.2(3); data for 102 are in Table 18.

and benzyl halide reductive coupling reactions. These reactions demonstrate the ability of 96, which is a complex of the relatively Earth-abundant zirconium ion, to act as a reliable replacement for the more commonly employed precious metal photosensitizers 2+ such as [Ru(bpy)3] . To close this section, some coordination chemistry for 2,6-di (pyrrolide)pyridine ligands where the reactivity and electrochemistry, not the photophysical properties, of the metal complexes were to the fore. First Caulton et al. have probed the steric and electronic charac- 23 teristics of the new 2,6-di(pyrrole)pyridine H2C through looking at its reactions with Fe, Zn, Pd and Pt sources and electrochemical 24 studies on selected isolated complexes [141]. The free base H2C was readily doubly deprotonated with KH to yield free-ﬂowing 23 K2C . Treatment of the later salt with [MCl2(NCPh)2] (M = Pd, Pt) 23 23 afforded the corresponding C2-symmetric [(C )M(NCPh)] com- Fig. 24. The SC-XRD structure of [Fe(HC )2](103) (CCDC reference code: VIWSUR) [141]. Key distances (Å) and angles (°): FeAN(pyr) 2.011(2) and 2.012(2), FeAN(pyr) plexes (98: M = Pd; 99: M = Pt), e.g., Fig. 22. The structure of 99 2.055(2) and 2.077(2); N(pyr)AFeAN(py) 83.43(9) and 82.11(9), N(pyr)AFeAN was determined by SC-XRD, and the key metric parameters (pyr0) 115.26(9), N(py)AFeAN(py0) 133.80(9), N(pyr)AFeAN(py0) 121.09(9) and (Table 18) are consistent with those discussed earlier for 86 and 125.02(9). 87 (Table 15). Both Pd(II) and Pt(II) complexes 98 and 99 showed redox couples at ca. +0.13 and +1.10 V that were attributed to 96 was strongly luminescent and it was demonstrated that ligand-centred oxidations due to their independence of metal quenching of the excited state with mild reductants generated a ion. Reaction of [(C23)Pd(NCPh)] 98 with CO gave a carbonyl com- powerful electron transfer reagent with a ground state potential plex 100 with an exceedingly high carbonyl frequency of 2197 of 2.16 V vs. the Fc+/0 couple. This photoreactivity of 96 was cm1 (which should signal caution as it compares with 2143 1 23 exploited in light-driven organic dehalogenation, oleﬁn reduction cm for free CO!). Direct reaction of H2C and ZnEt2 (1:1 mol

t-Bu t-Bu t-Bu t-Bu

N [Fe{N(SiMe3)2}2], Et2O N NH HN N N -2 HN(SiMe3)2 Fe 104 t-Bu t-Bu t-Bu t-Bu

OEt2 23 H2C 104

N3Ad, toluene ClCPh3, toluene/pentane -N2 -'•CPh3' 105 t-Bu t-Bu t-Bu t-Bu

N N N N N N Fe Fe t-Bu t-Bu t-Bu t-Bu Cl NAd 105 106 106

23 23 23 Scheme 44. Preparation of [Fe(C )(OEt2)] (104), [Fe(C )Cl] (105) and [Fe(C )(=NAd)] (106). Inserts: SC-XRD structures for 104 (CCDC reference code: JOQYOF), 105 (CCDC reference code: JOQZAS) and 106 (CCDC reference code: JOQYUL) [145].

James McPherson – June 2019 39 Chapter 1: Introduction

322 J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332

Table 19 Key bond length (Å) and angle (°) data from the SC-XRD structures of iron complexes 104–106 [145].

Complex (CCDC reference) FeAN(pyr) FeAN(py) \N(pyr)AFeAN(pyr0) \N(pyr)AFeAN(py)

23 104 [Fe(C )(OEt2)] (JOQYOF) 1.987(1) 2.021(2) 144.17(7) 78.96(4) 1.987(1) 78.96(4) 105 [Fe(C23)Cl] (JOQZAS) 1.991(2) 1.981(2) 142.70(9) 80.43(9) 1.998(2) 79.71(9) 106 [Fe(C23)(=N-Ad)](toluene) (JOQYUL) 1.910(2) 1.867(2) 147.6(2) 82.1(8) 1.910(2) 82.1(8)

t Bu t Bu t Bu t Bu t Bu t Bu

N + N h , C D N N3Ad ν 6 6 N N N N N N II II – II Co – solv Co N2 Co t Bu t Bu t Bu t Bu solv N NH solv = Et2O 107: Ad N Ad 108: solv = toluene 109 110 N

23 23 Scheme 45. Reactions of the solvato adducts [Co(C )(solv)] (107 and 108) with 1-adamantyl azide afford the unusual azide adduct [Co(C )(N3-Ad)] (109), photolysis of 28 which results (formally) in extrusion of N2 and CAH insertion to afford product [Co(HC )] (110) with a bound pendant amine [143].

ratio) led to a single product, tri-coordinate [(HC23)ZnEt] (101), In subsequent research, Fortier, Meyer, Mindiola, Caulton and Fig. 23. The complex was remarkably stable to elimination of colleagues described the synthesis of the novel four-coordinate 23 ethane, for example completely surviving heating at 60 °C for 8 h, iron(II)–ether complex [Fe(C )(OEt2)] (104) from [Fe(N(SiMe3)2)2] 23 and NMR experiments showed there is no transfer of the NAH pro- and H2C in ethereal solution. Oxidations of 104 were pursued ton on the pendant pyrrole to the coordinated pyrrolide donor. (Scheme 44) [145]. Single electron oxidation was accomplished

Complex 101, however, bound 4-dimethylaminopyridine (dmap) with Ph3CCl and gave the corresponding iron(III) monomer and quickly eliminated ethane upon heating to 50 °C to afford [Fe(C23)Cl] 105 (along with Gomberg’s radical and dimer). With 23 the C2-symmetric adduct [(C )Zn(dmap)] 102. The geometry at 1-adamantyl azide (Ad-N3), clean and direct conversion of 104 to Zn is perhaps best described as between cis-divacant octahedral the unusual iron(IV) imide [Fe(C23)(@NAd)] 106. SC-XRD structures and distorted tetrahedral. The main distortion from tetrahedral is reveal all three complexes exhibit a cis-divacant octahedral struc- between the pyrrolido donors: N3AZn1AN1 = 146.80(6)°. Other ture and the metric parameters for 104 (Table 19) are akin to that reactions of 101 with bases were also interrogated. Overall, it previously found for [Zn(C23)(dmap)] 102. The central FeAN(py) appeared that a reactant with sufficient Bronsted basicity to depro- shortens slightly with increasing oxidation state of the metal cen- tonate the pendant pyrrole was requisite for ethane elimination, tre (from 2.021(2) Å for 104, to 1.981(2) Å for 105 and 1.910(2) Å suggesting that attack by pyrrolide on the metal centre in concert for 106). SQUID magnetisation measurements (dc mode) revealed with transfer of a proton from the conjugate of base led to loss of 104 was a typical high-spin (S = 2) system with meff = 5.09–5.25 ethane and the product complexes. Caulton et al. also reacted FeCl2 mB over 30–300 K, (S = 5/2) species with negligibly temperature 23 23 with a 1:1 mixture of K2C and H2C in THF; a single iron(II) pro- dependent magnetism—meff = 5.83–5.73 mB over 30–300 K, and 23 duct formed: [Fe(HC )2], 103 [141]. A view from the SC-XRD struc- 106 was a low spin Fe(IV) (S = 0) complex (SQUID magnetisation ture of 103 is presented in Fig. 24. It was noted that the two measurements from 30 to 300 K were made to confirm this fact). pendant pyrrole NAH protons in the complex could potentially Zero-field 57Fe Mössbauer spectra of 104, 105 and 106 showed be exploited in further reactions. characteristic quadrupole doublets with isomer shifts, d,of At this juncture it is worth citing some of the conclusions Caul- +0.86(1), +0.39(1), and 0.09(1) mms1, and quadrupole splitting 1 ton and co-workers make in their paper on complexes from H2C values, DEQ, of 1.12(1), 2.38(1), and 2.78(1) mms , respectively. [141]. They reinforce the concept that pyridine-pyrrolido ligands The consistent shift in isomer shift of 0.48 mms1 indicates that are ‘‘push/pull ligands” stating: ‘‘pyrrolide... ring is an electron the three complexes do indeed exhibit different metal oxidation donating substituent by its Hammett parameters and is certainly states. DFT calculations concurred with this conclusion, revealed also a p donor to a metal due to its amide character” and that the largest component of the electric field vector lies not along

‘‘...pyrroles are powerful ancillaries to increasing the reducing the Fe@Nimido bond but along the FeANpyrrolido bond vector. The power of metals”, but ‘‘By coupling the pyrrolide ligand to a redu- calculations also predicted the same unusual cis-divacant octahe- cible ligand, pyridyl, (whose p⁄ orbitals permit reduction), we cre- dral structure for analogues with the t-butyl substituents removed, ate a push/pull ligand of enhanced potential redox character in thus paving the way for reactivity studies with more open ana- comparison to that of neutral bipyridyl. The push/pull character logues. Exciting prospects! Finally, the authors note the unprece- makes the pyridylpyrrolide capable of being reduced or of being dented geometry and electronic structure of 106 depend entirely oxidised, hence a useful ancillary to metal complexes under either on the (C23)2 pincer being a robust, strongly electron donating, reducing or oxidizing conditions” [141]. Unfortunately, the sub- and, in this case, a redox innocent ligand. tlety that has eluded some workers that have published research Most recently, Mindiola (from the previous group) et al. have on metal complexes of these ligands is that their Janus character extended the previous C2 iron chemistry to cobalt. Different reac- 23 arises from the combination of individual ring characters over tivity was observed, Scheme 45 [143]. The [Co(C )(OEt2)] species the whole ligand, not just the pyrrolide ring. (107) could be prepared similarly to its iron analogue 104 from

James McPherson – June 2019 40 Chapter 1: Introduction

J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332 323

107

110 109

108

23 23 Fig. 25. SC-XRD structures of the solvato complexes [Co(C )(OEt2)] (107, CCDC reference code: ASUDEZ) and [Co(C )]toluene (108, CCDC reference code: ASUDOJ), the 23 28 azide adduct [Co(C )(N3-Ad)] (109, CCDC reference code: ASUDID), and the photolysis product [Co(C )] with a pendant amine (110, CCDC reference code: ASUDUP) [143].

Table 20 Key bond length (Å) and angle (°) data from the SC-XRD structures of Co(II) complexes 107–110 [143].

Complex (CCDC reference) CoAN(pyr) (Å) CoAN(py) (Å) \N(pyr)ACoAN(pyr0)(°) \N(pyr)ACoAN(py) (°)

23 107 [Co(C )(OEt2)](toluene) (ASUDEZ) 1.950(2) 1.974(3) 145.8(1) 80.83(7) 1.950(2) 80.83(7) 108 [Co(C23)](toluene)y (ASUDOJ) 1.945(4) 1.967(7) 152.4(5) 81.5(2) 1.945(4) 81.5(2) 23 109 [Co(C )(N3Ad)] (ASUDID) 1.964(3) 1.971(4) 151.3(1) 81.5(1) 1.958(3) 81.5(1) 110 [Co(C28)] (ASUDUP) 1.843(2) 1.857(2) 158.41(8) 82.28(8) 1.933(2) 83.32(8) y Cobalt centre is disordered over 3 sites – so averages are reported here.

Table 21 Aggregated average bond lengths (A) and bond angles (°) in complexes of the C2 core grouped by metal ion and coordination geometry.a

Metal ion Geometryb MAN(pyr) (Å) MAN(py) (Å) \N(pyr)AMAN(pyr0)(°) \N(pyr)AMAN(py) (°) Reference(s) Co(II) T 1.945 1.967 152.4 81.5 [143] Co(II) SPl 1.888(2) 1.857(2) 158.41(8) 82.80(8) [143] Pd(II) SPl 2.010(3) 1.964(3) 161.6 80.9(1) [144] Pt(II) SPl 2.05 1.97 161 81 [141,144]

Fe(II) cis-diV-Oh 1.987 2.021 144.17 78.96 [145]

Fe(III) cis-diV-Oh 1.995(2) 1.981(2) 142.70(9) 80.07(9) [145]

Fe(IV) cis-diV-Oh 1.910 1.867 147.6 82.1 [145]

Co(II) cis-diV-Oh 1.96 1.973 148.5 81.2 [143]

Zn(II) cis-diV-Oh 1.996(1) 1.988(1) 146.79(5) 81.54(5) [141] Sn(IV) TBP 2.170 2.193 147.0 73.5 [124] Pb(IV) TBP 2.33 2.32 140.6 70.4 [124]

Ti(III) Oh 2.113(3) 2.129(3) 149.4(1) 74.80(9) [49]

Ti(IV) Oh 2.039 2.122 147.6 74.2 [49]

Zr(IV) Oh 2.162(3) 2.294(3) 140.0(1) 70.3(1) [49]

a ESDs are listed where the average is for one SC-XRD structure only. b Geometry codes: T = T-shaped (mer-tri-vacant-octahedral), SPl = square planar, cis-diV-Oh = cis-di-vacant-octahedral, TBP = trigonal bipyramidal, Oh = octahedral.

23 23 23 H2C and [Co(N(SiMe3)2)2] in ether. In toluene, however, H2C molecule sitting interstitial between two [Co(C )] units), however 23 and [Co(N(SiMe3)2)2] afforded [Co(C )] (108), which co- the CoAN bond lengths and angles (Table 20) ﬁt with those crystallised with toluene in the solid state, Fig. 25. The geometric observed in four coordinated cis di-vacant complexes of C2. The parameters of 107 and 109 (Table 20) are similar to the other cis coordination geometry of 110 is described as distorted square pla- di-vacant octahedral complexes 102 and 104 described previously, nar, and the wingtip angle N(pyr)ACoAN(pyr0) of 158.41(8)° is however with slightly wider wingtip N(pyr)ACoAN(pyr) angles wider than observed in other cis di-vacant complexes (142– (145.8(1) and 151.3(1)° for 107 and 109). The three coordinate 152°). Magnetic susceptibility measurements on 107 and 108 in complex 108 is described as T-shaped geometry (with a toluene C6D6 suggested both are high spin Co(II) (S = 3/2) species. Reaction

James McPherson – June 2019 41 Chapter 1: Introduction

324 J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332

N N

Br HCl (2 M) N N NH N N CH 2Cl2 N Boc 0 °C HD1 93% Pd(PPh 3)2Cl2, Li K2CO3 (aq) N B(OMe)3 THF / H2O Boc 80 °C N N HD2 Br N NH N

Scheme 46. Syntheses of HD1 and HD2 using Suzuki coupling [146].

of 107 with 1-adamantyl azide produced the green azide adduct 4. The 2-(2,20-bipyridin-6-yl)pyrrolide (D) and 5-(2-pyridyl)- 23 0 2 [Co(C )(N3-Ad)] (109) rather than the targeted imide. The azide, 2,2 -bipyrrolide (E ) anions uniquely, is bound through the c-N atom, Fig. 25. Photolysis of the azide did not give the targeted imide, but generated the pro- At the time of submission, a final structure search of the CAS duct [Co(HC28)] (110) that formally arises from insertion of an N- SciFinderTM database for core D with charge and substitution Ad group into the CAH bond of a flanking t-butyl group, Fig. 25. unspecified found 59 substances in 18 references. Likewise, a Intriguingly, the magnetic moment of 110 (leff = 3.9 lB) was much search for core E, again with charge and substitution unspecified, lower than the other complexes, perhaps as a consequence of it found 31 substances across 14 references. The vast majority of exhibiting a flatter, nearer square planar structure, and so conse- the substances are longer, linked or macrocyclic systems and were quential differences in spin–orbit coupling. excluded from this review (as they will described elsewhere [60]). Of the remaining substances, some are N-derivatized at the pyrrole 3.3. Summary so that this group cannot form a coordinate bond with a metal ion. Therefore, these substances are not considered. This leaves only There are several synthetic routes that could be readily three (potential) tridentate ligands and one complex, all with core D and none with core E, in this section. exploited to afford derivatives with a dipyrrolepyridine (H2C) core. Noteworthy among these, the Fisher indole synthesis pro- 1 vides a traditional safe route to H2C and other derivatives of N N diindolylpyridines, palladium cross-couplings offer a short route N N N to dipyrrolepyridines, and the Stetter reaction as exploited by N Nagata and Tanaka or the McNeill synthesis as employed by Komine et al. also appear to offer straightforward routes to DE new examples. A variety of p-block and transition metal complexes of tridentate dipyrrolidepyridine (C2) ligands have been prepared. One 4.1. Syntheses and properties of 2-([2,20-bipyridin]-6-yl)pyrroles noteworthy feature is that many of the metal complexes are luminescent, and the favourable photophysical properties of some The routes to cores B and C, Sections 2 and 3 respectively, were these complexes have been exploited in electroluminescent also employed here. Johannes and co-workers used Suzuki cou- devices or as photosensitizers for light-driven catalyses. Opportu- pling of a N-Boc-protected trimethyl-pyrrol-2-ylboronate with 6- nities abound. bromobipyridine with Pd(PPh3)2Cl2 as catalyst and K2CO3 as the A summary of the key metrical information for the metal com- base to obtain, after acid hydrolysis, 2,20-bipyridine-pyrrole (HD1) plexes of the tridentate C2 core, averaged for particular metal ion in 83% overall yield, Scheme 46 [146]. With 2-bromo-1,10- and oxidation state, is presented in Table 21. Noteworthy is that phenanthroline, it was possible to directly isolate the these ligands support a variety of metal coordination geometries, phenanthroline-pyrrole analogue (HD2), albeit in lower yield 39%. for example, ranging in extremes from three-coordinate T-shaped Anions (D1) and (D2) were targeted as potential sensitising species (of surprise, for Co(II)) through to regular octahedral com- ligands for Eu3+ ion (the extended p-conjugation pushes the n,p– plexes (for Ti(III), Ti(IV) and Zr(IV)). The MAN(pyr) distances range p bands into the visible); however stable complexes could not be 1.95 Å (for three coordinate Co(II)) to 2.33 Å (for five-coordinate Pb obtained.

(IV)), and the MAN(py) distances range 1.97 Å to 2.32 Å (for the Classical Paal–Knorr condensation of a diketone with NH4OAc same species, respectively). That the MAN bond distances are soft to afford a pyrrole was employed to good effect by Nagata and (variable according to metal ion) is reﬂected in the N(pyr)AMAN Tanaka [139]. These researchers used the Stetter condensation of (py) bite angles. The bite angles show some small variability, and 6-carboxyaldehyde-2,20-bipyridine with chalcone to obtain the cluster into two groupings 70–75° and 79–83°, the former being requisite bipyridine-substituted 1,4-diketone intermediate for ring 3 found in the more crowded ﬁve and six coordinate species and closure with NH4OAc to afford HD (67% overall yield), Scheme 47. the latter in the species exhibiting four or less coordination The chemistry of a ruthenium(II) complex of the (D3) anion is environments. described next. James McPherson – June 2019 42 Chapter 1: Introduction

J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332 325

N+ Cl-

S CHO Ph Ph Ph OH NH OAc N + 4 N N t-BuOK/EtOH N N O EtOH N HN O Ph O Ph Ph HD3

Scheme 47. Synthesis of HD3 [139].

(a) (b) N N 111 N Ru N N N

[Ru(tpy) ](ClO ) 111 2 4 2

3 2+ Fig. 26. (a) Complex [Ru(D )2] 111 and (b) cyclic voltammograms of 111 and [Ru(tpy)2] in DMF0.1 M Et4NClO4 (reproduced with permission from Ref. [139]).

S S SMe S S HN HN NH S S SMe S S

pyridine Tf 86 % triflic anhydride 46 % Tf Tf CH Cl ,243K N 2 2 N N 15 min

S S SMe S S HN HN NH S S SMe S S

N N N tBuOK Tf THF, r.t. 81 % Tf Tf 78 % 45 min

N N N

S S SMe S S HN HN NH S S SMe S S

N N N ftt-ppd ftt-ppdb

Scheme 48. Sallé and co-workers’ preparation of di(4-pyridyl)pyrrolyl-substituted tetrathiofulvalenes, dpp-ttf and bdpp-ttf [147,148].

0 3 4.2. A coordination compound of a 2-(2,2 -bipyridin-6-yl)pyrrolide precipitated from a mixture of HD , RuCl3xH2O and NEt3 in etha- anion (D) nol that was heated at reﬂux over 17 h, then cooled. A SC-XRD structure is reported for complex 111, but the data is unavailable. Nagata and Tanaka reported the only metal complex with Complex 111 is reported to adopt a similar coordination geometry 2+ asymmetric tridentate pyrrolide-pyridine ligand system, namely about the Ru(II) centre to in [Ru(tpy)2] , but the pyridine and 3 [Ru(D )2](111), Fig. 26a, in 2002 [139]. Deep-green 111 was pyrrole rings are not co-planar (the dihedral angles between

James McPherson – June 2019 43 Chapter 1: Introduction

326 J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332

N Br 1) HO B N O HO (2.01 eq) (2.3 eq) Br Br N N N THF [Pd(PPh 3)4] (0.1 eq.) N N Boc -78 °C, 1 h Boc Na2CO3 (8 eq.), KCl (3 eq.) H 0-4 °C, 18 h toluene, ethanol, H2O 70 % 90 °C, 18 h 2) 4 M HCl in 1,4-dioxane 40 °C, 2h 20 %

Scheme 49. Fujita and co-workers preparation of 2,5-di(4-pyridyl)pyrrole in 14% overall yield [149].

A. B.

112 113

7 Fig. 27. SC-XRD structures found for ligand H4C upon crystallisation from dmf-methanol with diverse additives [127]. (A) No additive and crystallisation from dmf alone: H bonded chains of dmf solvate (CCDC reference code: ILORAI). (B) with [(n-Bu)4N](OAc): H bonded dimers of monoanion.hydrate (CCDC reference code: ILOREI). (C) with Zn 7 m 7 m m (acac)2: discrete [(Zn8(C )8( 2-OH)2](112: cage 1; CCDC reference code: ILORIM). (D) with Zn(OAc)2: 1-D polymeric {[(Zn4(C )3( 2-OH)( -OAc)](H2O)}1 (113: strands of linked cages 2; CCDC reference code: ILOROS). To assist visualisation in views C and D, the carbon atoms of ’forward’ and ’backward’ facing (C7)2 ligands are coloured pale green and pale tan, respectively. It is noteworthy that exclusive carboxylato coordination with the Zn2+ ions, not pyridine or pyrrolide binding, was observed in the reactions with the basic Zn(II) salts.

neighbour rings are 8.5–19.4°), presumably to avoid steric clashes 5. Pyridine–pyrrolide ligands in metal-directed self-assembly of with the phenyl substituents. The RuAN(pyr) distances are 0.024– supramolecular cages and polymeric structures 0.069 Å longer than the RuAN(py)outer distances. Cyclic voltammograms of 111 revealed a reversible Ru(II)/Ru(III) couple at 0.29 V In the coordination chemistry reviewed thus far there are some +/0 2+ (vs. Fc ), which is 1.1 V more negative than that for [Ru(tpy)2] , notable examples in which coordination of pyridine–pyrrolide Fig. 26b. Complex 111 also showed a ligand-centred couple at ligands with metal precursors led to self-assembly of supramolec- 2.17 V (vs. Fc+/0) which was 0.51 V more negative than observed ular structures, such as linear coordination polymers. In this sec- 2+ for [Ru(tpy)2] , Fig. 26b. At higher potentials (+0.5–0.7 V), the irre- tion, the ambit of this review is widened slightly to encompass versible waves suggested pyrrolide oxidation but in this case is self-assembly of supramolecular structures from closely related accompanied by degradation or decomplexation of the ruthe- pyridine–pyrrolide ligands with a combination of interlinked 4- nium(III) ion. These results again point to the strong donor charac- pyridine and pyrrolide rings or where the 2-pyridine–pyrrolide ter of the pyrrolide ring. skeleton does not itself bind with a metal centre. Importantly, in

James McPherson – June 2019 44 Chapter 1: Introduction

J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332 327

A. B.

114

116

Fig. 28. Use of electro-active bis(dipyridylpyrrolo)tetrathiafulvalene (bdpp-ttf) ligands for platinum cage self-assembly. (A) SC-XRD structure of the [Pt6(bdpp-ttf)3(PEt3)12] cage 114 (CCDC reference code: FEQTOM [148]; for clarity, only the ipso-C atoms of the PEt3 ligands are depicted). (B) UV–vis spectra (and Job plot, inset) for titration of Pt6 cage with tcnq; reprinted (adapted) with permission from [148], copyright (2012) American Chemical Society. (C) Complex cation from the SC-XRD structure of (dppp) II Pt (dithiolene) complex 116 (CCDC reference code: FEQTOM) formed on decomposition of Pt6 cage 115 in hot dmso [147]. doing so, an increased scope for three-ring pyridine–pyrrolide ligands becomes apparent.

5.1. Syntheses and properties of ligands

The syntheses of most ligands employed in the assembly of cages and other supramolecular structures have already been summarised. The few exceptions are mentioned here. Sallé and coworkers reported the synthesis of two di(4-pyridyl) pyrrolyl substituted tetrathiofulvalenes, dpp-ttf and bdpp-ttf,in very good yields starting from the pyrrolyl-substituted tetrathio- fulvalene precursors as summarised in Scheme 48 [147,148]. Fujita and co-workers prepared 2,5-di(4-pyridyl)pyrrole from θ =127° θ =135° θ =143° θ =147° θ =149° Boc (butoxycarbonyl)-protected pyrrole, via bromination and Suzuki coupling, Scheme 49. Jones and Dubreuil offer alternate routes to this compound, via the Stetter reaction (27% over 2 steps) [68] or reductive ring contraction (55% over 4 steps) [80] respectively, which have previously been discussed in Section 2.1. Fig. 29. Self-assembly of Pd12L24 and Pd24L48 clusters (a) from M = [Pd(MeCN)4] (BF4)2 and L = di(4-pyridyl)C4H2X (X = O, NH, NMe, S) ligands with different bend angles, noteworthily including three di(4-pyridyl)pyrrole ligands (b). Reproduced 5.2. Coordination compounds (adapted) with permission from [149], copyright (2012) Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim. Sessler and co-workers demonstrated that bis(carboxylpyrrole) 7 pyridine H4C (see Section 3.1) undergoes environmentally so-called environmentally sensitive self-assembly. From dmf, a 7 sensitive self-assembly [127]. SC-XRD structure studies revealed self-assembled (H-bonded) polymer of H4C dmf was obtained, 7 that H4C crystallised differently under diverse conditions—the whereas with [NBu4]OAc dimers of mono-deprotonated James McPherson – June 2019 45 Chapter 1: Introduction

328 J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332

7 [(n-Bu)4N)](H3C )H2O crystallised in which the three N atoms upon crystallisation. In search of larger structures still, Fujita’s point inward, a conformation stabilised by H-bonding between group have extended the linkers between the five membered core 7 the pyrrole NAH and the ancillary water molecule. Ligand H4C and pyridyl arms to form the Pd30L60 icosidodecahedron [155],or was also crystallised in the presence of two basic Zn(II) salts. Crys- explored five membered heterocycles with wider still bend angles 7 tallisation of Zn(acac)2 and H4C in 5:1 dmf-methanol resulted in (h = 152° about a selenophene, X = Se) [156]. This selenophene formation of metallo-mediated assembly of discrete centrosym- ligand (L) assembled with Pd2+ cleanly into an unexpected 7 metric Zn8(H2C )8(m2-OH)2 cages, complex 112, designated ‘‘cage Pd30L60 cage, with different topology to the previously reported 1” in the paper (Fig. 27). Noteworthy the pyridine and pyrrole Archimedean solids. Analysis of the topology for MnL2n suggested donors of ligand C are not involved in binding in cage 1. In cage that this structure should be metastable, and by changing reaction 2+ 7 2 1, the Zn ions are bound by carboxylato groups from (H2C ) conditions Fujita and team were able to obtain crystals of a giant 7 and divide into two Zn4(H2C )3(m-OH) tetramers in which pairs of Pd48L96 cage [156]. Collection of single crystal X-ray diffraction carboxylato-bridged Zn2+ ions are further bridged the hydroxido data for more than 10 crystals and refinement gave 2.85 Å best res- ligand. Each tetramer linked to the other by binding at the two olution for the Pd48L96 cage (as anticipated for such large com- apex Zn2+ ions to one carboxylato-group of one of two tetramer- pounds of high symmetry and large solvent filled void spaces).2 7 2 7 linking (H2C ) ligands. Crystallisation of Zn(OAc)2 and H4C from the same solvent mixture afforded a 4:3 Zn-H C7 coordination 2 6. Conclusion polymer, compound 113, which was designated ‘‘cage 2”. Cage 2 resembles cage 1 in that it also is formed from Zn (H C7) (m-OH) 4 2 3 This is the first review on the coordination chemistry of triden- tetramers. Whereas in cage 1 the two tetramers are spanned by 2 7 2 tate pyridine–pyrrolides B –E , Fig. 1, and of their pyridine–pyr- two ligands of (H2C ) to form the discrete cage, in cage 2 the api- role precursors, HB–H2E. There are several overall take-home cal sites of the tetramers are linked together to form, upon repeti- messages, learnings to be shared with the reader. tion, infinite linear strands. Dynamic light scattering experiments on the solutions prior to crystallisation of the cages gave a mean 7 6.1. Exploitable routes to ligand precursors particle diameter of 2.7 nm for the Zn(acac)2–H4C solution and 7 825 nm for the Zn(OAc)2–H4C solution concurring with expecta- tions of larger particle size in solution as cage 2 develops. These are relatively easily prepared ligands, which is surprising Salle and co-workers reported metal-directed self-assembly of given how little they have been used. In total, this review describes cages from di- and tetratopic ligands with electro-active dipyridyl- 75 precursors to potentially tridentate meridional pyridine–pyrro- pyrrolo tetrathiafulvalene skeletons [148]. A short synthesis (see lide ligands. The routes available are identified in the above sec- Section 5.1 above) afforded the new bis(dipyridylpyrrole)tetrathia tions, and range from very straightforward, one-pot preparations fulvalene (bdpp-ttf) ligand in good yield. This ditopic ligand cleanly in moderate yield through short syntheses involving several steps in excellent overall yield to elaborate concoctions. It is noteworthy assembled with di(phosphine)Pt(II) precursors (either cis-(Et3P)2 that many of the syntheses reported to date provide functionalised Pt(OTf)2 or (dppp)Pt(OTf)2) into trigonal prismatic Pt6 cages, pyridine and pyrrole rings, which should facilitate both the optimi- [Pt6(bdpp-ttf)3(PEt3)12](114) or [Pt6(bdpp-ttf)3(dppp)6](115), which were crystallised and characterised by SC-XRD (Fig. 28A). sation of their electronic and/or steric properties and the integration of innovative functional complexes into next-generation UV–vis titrations (Fig. 28B) with the electron acceptor tcnqF4 showed isosbestic replacement of the bands at 355 and 442 nm molecular assemblies and devices. ox for the electron donating (Et3P)12Pt6(bdpp-ttf)3 cage (E 0.43 0/+ and 0.62 V vs. Fc ) by a band at 690 nm for [(Et3P)12Pt6(bdpp- 6.2. Rich coordination chemistry + ttf)3] and bands at 752 and 849 nm for tcnqF4 and at suggesting 2 a clean 1:1 transformation to a CT complex. Modelling revealed the The coordination chemistry of ligand cores B and C is at the

Pt6 cage was a suitable size for incorporation of one internal tcnqF4 same time rich and diverse, and still barely scratched, re- guest, and titrations and a Job plot also indicated 1:1 complex for- discovered, relatively new and with vast unexplored regions mation thereby corroborating incorporation of precisely one mole- throughout the Periodic table. And, in comparison, nothing, well 2 cule of the electron poor guest into the trigonal prismatic electron almost nought, is known about cores D and E : there is one com- 2 rich Pt6 cage. A follow-up report gave further synthetic details, and plex of core D and none of E ! But these are good ligands. That revealed the ditopic bdpp-ttf ligand, and so the Pt6 cages, were sus- should not be a revelation; all are tridentate chelates with linked ceptible to fragmentation with the corresponding di(phosphine) pyridine and pyrrolide donors. Thus it is unsurprising that genuine PtII(dithiolene) complexes, e.g., 116, being the ultimate products examples of metal ion coordination complexes are sprinkled (Fig. 28C) [147]. throughout Periodic table. For instance, for the dipyridylpyrrolides Fujita has studied in detail the remarkable self-assembly of (core B) – the most studied class of these ligands – well- giant coordination polyhedra from Pd2+ ions and various di(4- characterised complexes have been reported for Na, Fe, Co, Ni, pyridyl)C4H2X (X = O, NH, NMe, S, Se) ligands, Fig. 29 [149–154]. Cu, Zn, Ru, Pd, Ag and Pt. Conversely, it is of note that across all 2 The five-membered C4H2X heterocyclic ring does not bind to a ligands B –E , complexes of second and third row transition metal ion in any of the cages. The metallo-supramolecular assem- and s/p-block metal ions are in general under-represented, and bly is exquisitely sensitive to the bend angle of the ditopic ligand. complexes of lanthanide and actinide ions are completely missing. DFT calculations (B3LYP/6-31G⁄) reveal bend angles (h) anticipated Making compounds for the sake of it, just to fill in gaps, ‘‘stamp- for various 2,5-di(4-pyridyl)C4H2X ligands are X = O: 127°; NH: collecting” to use Rutherford’s vernacular from a much earlier era 135°; NMe: 147°; S: 149° [149]. Selective assembly of Pd12 was [160], however, is no longer a justification. Society fairly demands observed for the furan (X = O)-bridged ligand that has the smallest bend angle (127°). Likewise, criticality of self organisation was 2 We note that much of the popular commentary about the Fujita cages, giant and observed to switch completely from Pd12 to Pd24 cages when the spectacular as they are, has been over-hyped [157]. As they do not approach the ratio of thiophene:furan-bridged ligand was between 3:7 (mean largest self assembled metallo-supramolecular assembly in terms of size or the number of species assembled: molecularly pure 2.2 nm gold-thiolate nanoparticles bridge angle 133.6°) and 2:8 (129.1°). With the wider bend angles have been self assembled by Jin and co-workers (Au246(S-p-toluenyl)80 [158]) and (135°–149°) of the pyrrole- (X = NH and NMe) and thiophene-(X = more recently by Dass’ group (Au279(SPh-tBu)84 [159]), with the structures deter- S) bridged ligands, only M24 cages are observed in solution and mined by SC-XRD at resolutions of 0.96 Å and 0.90 Å respectively. James McPherson – June 2019 46 Chapter 1: Introduction

J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332 329 eventual recompense through advances in useful knowledge and complexes. Milsmann’s recent invention of strongly luminescent IV 22 technology [161]. Reasons for why we think these complexes [Zr (C )2] complex (96) provides exemplar behaviour: it is an should be targeted are summarised below. excellent photosensitiser and superior photoredox catalyst in several light-driven organic transformations [49]. Not described so far, 6.3. Geometrical strain but obvious targets, are lanthanide complexes of the tridentate pyridine–pyrrolide anions. Lanthanide ions are large and should 2 2 At the outset, we presented a ‘‘handwaving, back-of-the- bind ligands B –E strongly; therefore ligands B –E should be envelope” argument about the geometrical effect of a five- excellent sensitizers of lanthanide emission, particularly in the membered pyrrolide ring in a tridentate ligand, especially as the near infrared region, which is important for biomedical and central donor, Section 1.2. Strain is always introduced, but is min- telecommunication applications [162–167]. imised with a central pyridine; i.e., geometric strain within the metal complexes will be greatest for ligands B and E and will 6.5.3. Functional supramolecular constructs be minimised for ligands C2 and D. The metric data from the Clearly these complexes could be built into functional metallo- SC-XRD structures of the complexes summarised in the review supramolecular materials, including discrete cages, coordination strongly support this conclusion. polymers and other macromolecules that exploit their unique photophysical and redox properties [21,22,24,28,31,33,40,41]. 6.4. Electronic effects 6.5.4. Medicinal chemistry and sensing All evidence is that the tridentate pyridine–pyrrolide ligands, The medicinal applications of most of the substances described B–E2, are strong r/p-donors (as fore-shadowed in Section 1.2). in this review have not been considered. They should be. In this III 39 + The p-donor character of these ligands leads to high spin states regard, Chen’s [Co (B )2] complex (22) is conspicuous in its in their first-row transition metal ion complexes, and combines DNA G-quadruplex binding and human tumour cell anti- 1 with the inherent negative ligand charge to confer exceptional sta- proliferative abilities [90]. Even the simple [Cu(B )Cl] (23) binds bility to metal ions in higher oxidation states with a limitation: the and cleaves DNA in the presence of ascorbate, suggesting potential pyrrolide ring is redox non-innocent, an electron donor at high footprinting applications [110]. Several of the pyridine–pyrroles potentials (e.g., between +0.85 and +1.47 V vs. Fc+/0, depending are also noteworthy in strongly binding G-quadruplex DNA on substituents, at a Ru3+ centre [74]). Naive statements such as [91,128,129]. We believe that there are opportunities for turn- ‘‘the flexible p-properties of pyrrolate donors that permits versatile on/off luminescent sensors for bioimaging based on fluorescent p-donor and p-acceptor responses to the metal site p-bonding pyridine–pyrrolide metal complexes [162–165]. There is insuffi- properties” [69,74,101] have no basis in fact. The notion appears cient data to opine further. to have originated with a mis-reading of the literature and apart from the muddling interpretations in one dazing outlier, there is 6.5.5. Molecular catalysis no evidence across a vast swathe of literature for coordinated pyr- To date only three reports describe use of pyridine–pyrrolide rolide anion displaying p-acceptor character; it is always a strong metal ion complexes as catalysts [70,74,93]. Each involves an r/p-donor to the metal ions it binds. This does not negate that a uncomplicated, benign substitution of a dipyridylpyrrolide-metal ion complex (M = Ru or Ag) into an already established catalysis tridentate pyridine–pyrrolide ligand as a whole displays flexible p- recipe that used a very similar complex of the same metal ion as donor/p-acceptor character. In this regard, meridional tridentate the catalyst (see Section 2.2). The good news: these new catalyses ligands D are particularly noteworthy as they combine a familiar worked. The bad news: there is nothing really novel and exciting in electron-acceptor 2,20-bipyridine chelate with an electron-donor the catalyses reported to date, they do not represent major pyrrolide ’on the side’: at a minimum, enhanced redox chemistry advances, the processes are not innovative in any way nor are there should result (as was found for the one example complex in the lit- dramatic improvements in yield or turnover or catalyst stabilities. erature to date [139]). We identify the following properties inherent to pyridine–pyrrolide metal ion complexes as indicating they should be particu- 6.5. Applications larly efficacious molecular catalysts:

The relatively recently re-discovered, now expanding miniverse (i) Stabilisation of high oxidation states: This feature of these of tridentate pyridine–pyrrolide metal ion complexes already con- complexes has already been commented on, and has an obvi- tains tantalising demonstrations of practical promise. Tangible ous implication, these complexes should be effective cata- applications described in this review ripe for exploration and lysts of oxidation/oxygenation reactions. An example is the development include: replication by Yi et al. using [RuII(B1)(COD)Cl] (66) [93] II instead of Bera’s original [Ru (COD)(py-NHC)Br2] catalyst in 6.5.1. Molecular electronics the original oxidative scission of alkenes to aldehydes by

The stand-out example here is Tour’s ’tour de force’, the Kondo- NaIO4 oxidant [115]: in both sets of reactions, Ru(VI)-dioxo resonance-enabled switching of conductivity demonstrated for species were proposed as the active catalysts, see Scheme 28. molecular transistors comprised of single [M(B)2] (M = Co, Cu) (ii) Hemilability: Inherent coordination strain for ligands with a molecules sandwiched between gold-plated source and drain elec- central pyrrolide donor has already been emphasised; thus trodes [62,95–98]. we anticipate ligands B and E2 may be hemilabile in many of their metal coordination complexes, with de-coordination 6.5.2. Electroluminescence and photoredox catalysis of an outer donor group relieving strain and, at the same Many of the metal complexes of ligands C2 were found to be time, opening a vacant site facilitating catalysis [168–171]. highly luminescent, well suited for applications in light-emitting (iii) Oxygen-atom carrier ability: In high-valent metal-oxo com- devices and optical displays, see Section 3.2. This is not unex- plexes of hemilabile ligands B, a pyridine donor could pected. Ligands C2 and Dintroduce electron-donating, strong potentially act as oxygen atom carrier, accepting and carry- r/p-donor pyrrolide rings with minimal geometric strain and, ing an oxo-ligand from the metal as the pyridine-N-oxide therefore, should activate usefully long-lived LMCT states in their through catalytic steps, then donating it back at the James McPherson – June 2019 47 Chapter 1: Introduction

330 J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332

z + (z +1)+ (z +2)+

H+ N N N H H + + R ± H R ± H R N MLx N MLx N MLx R R R H N H+ N N

z+ Scheme 50. A proposal for reversible proton-uptake by a dipyridylpyrrolide-metal complex, [M(B)Lx] .

appropriate time to an organic substrate activated at the References metal centre [172] thus enabling new oxygenation chemistry. (iv) Proton-carrier ability: Likewise, the pyridine donor(s) in [1] M. Albrecht, G. van Koten, Angew. Chem. Int. Ed. 40 (2001) 3750–3781. [2] D. Morales-Morales, C.M. Jensen, The Chemistry of Pincer Compounds, hemilabile complexes of ligands B or the pyrrolide donors Elsevier, Amsterdam, 2007. 2 in complexes of ligands E could potentially act as [3] M. Albrecht, R.A. Gossage, M. Lutz, A.L. Spek, G. van Koten, Chem. Eur. J. 6 proton-acceptors, accepting and carrying a proton through (2000) 1431–1445. the crucial catalytic steps, thereby assisting catalyses involv- [4] S. Shanmugaraju, P.S. Mukherjee, Chem. Eur. J. 21 (2015) 6656–6666. [5] A. Kascatan-Nebioglu, M.J. Panzner, C.A. Tessier, C.L. Cannon, W.J. Youngs, ing proton transfer steps [173–177]. Coord. Chem. Rev. 251 (2007) 884–895. (v) Pyrrolide multi-proton non-innocence: The bound pyrrolide [6] K. Li, T.T. Zou, Y. Chen, X.G. Guan, C.M. Che, Chem. Eur. J. 21 (2015) 7441– donor(s) are also potential carriers of protons without de- 7453. [7] J.L.L. Tsai, T.T. Zou, J. Liu, T.F. Chen, A.O.Y. Chan, C. Yang, C.N. Lok, C.M. Che, coordination from the metal centre as the hydropyrrolide Chem. Sci. 6 (2015) 3823–3830. and dihydropyrrolide anions. Scheme 50 depicts the uptake [8] E. Peris, R.H. Crabtree, Coord. Chem. Rev. 248 (2004) 2239–2246. of a single proton for a complex of a ligand B. Such multi- [9] D. Pugh, A.A. Danopoulos, Coord. Chem. Rev. 251 (2007) 610–641. [10] C. Gunanathan, D. Milstein, Acc. Chem. Res. 44 (2011) 588–602. proton non-innocence for a pyrrolide donor is known for a [11] N. Selander, K.J. Szabo, Chem. Rev. 111 (2011) 2048–2076. bis(phenolato)pyrrolide-titanium(IV) species [178]. Again, [12] C. Gunanathan, D. Milstein, Chem. Rev. 114 (2014) 12024–12087. such behaviour should promote catalyses involving proton [13] D. Benito-Garagorri, M. Puchberger, K. Mereiter, K. Kirchner, Angew. Chem. Int. Ed. 47 (2008) 9142–9145. transfer steps. [14] R. Tanaka, M. Yamashita, K. Nozaki, J. Am. Chem. Soc. 131 (2009) 14168– (vi) Hydride carrier ability: Finally, we note a dihydropyrrolide 14170. species such as depicted right-side of Scheme 50 could also [15] B. Schulze, D. Escudero, C. Friebe, R. Siebert, H. Gorls, U. Kohn, E. Altuntas, A. Baumgaertel, M.D. Hager, A. Winter, B. Dietzek, J. Popp, L. Gonzalez, U.S. act as a donor of a hydride ion and a proton (with the overall Schubert, Chem. Eur. J. 17 (2011) 5494–5498. charge of the complex unchanged) in a transformation catal- [16] A.M. Prokhorov, T. Hofbeck, R. Czerwieniec, A.F. Suleymanova, D.N. ysed at the metal centre; thus the bound pyrrolide groups Kozhevnikov, H. Yersin, J. Am. Chem. Soc. 136 (2014) 9637–9642. also have potential as acceptors/carriers/donors of hydride [17] E.C. Constable, Adv. Inorg. Chem. Rad. 30 (1986) 69–121. [18] J.P. Sauvage, J.P. Collin, J.C. Chambron, S. Guillerez, C. Coudret, V. Balzani, F. ion [179]. Barigelletti, L. De Cola, L. Flamigni, Chem. Rev. 94 (1994) 993–1019. [19] G. Chelucci, R.P. Thummel, Chem. Rev. 102 (2002) 3129–3170. [20] D.R. McMillin, J.J. Moore, Coord. Chem. Rev. 229 (2002) 113–121. 6.6. Scope [21] U.S. Schubert, C. Eschbaumer, Angew. Chem. Int. Ed. 41 (2002) 2892–2926. [22] P.R. Andres, U.S. Schubert, Adv. Mater. 16 (2004) 1043–1068. So what’s next? Where might the coordination chemistry of tri- [23] E. Baranoff, J.P. Collin, L. Flamigni, J.P. Sauvage, Chem. Soc. Rev. 33 (2004) 147–155. dentate pyridine–pyrrolide ligands go? Clearly, each of the above [24] H. Hofmeier, U.S. Schubert, Chem. Soc. Rev. 33 (2004) 373–399. applications is very worthy and important. However, in all cases, [25] E.A. Medlycott, G.S. Hanan, Chem. Soc. Rev. 34 (2005) 133–142. the cited chemistry provides only a starting point, a prototypal pyr- [26] T. Pintauer, K. Matyjaszewski, Coord. Chem. Rev. 249 (2005) 1155–1184. idine–pyrrolide complex made and a practical function demon- [27] K.M.C. Wong, V.W.W. Yam, Coord. Chem. Rev. 251 (2007) 2477–2488. [28] E.C. Constable, Coord. Chem. Rev. 252 (2008) 842–855. strated, but nothing else. There is little depth of knowledge. [29] J.J. Concepcion, J.W. Jurss, M.K. Brennaman, P.G. Hoertz, A.O.T. Patrocinio, N.Y. Further studies are demanded. Coogee Beach [180] is 5 min from M. Iha, J.L. Templeton, T.J. Meyer, Acc. Chem. Res. 42 (2009) 1954–1965. UNSW [181,182]: like a Coogee swimmer beside the cool ocean [30] S.D. Cummings, Coord. Chem. Rev. 253 (2009) 1495–1516. [31] A. Wild, A. Winter, F. Schlutter, U.S. Schubert, Chem. Soc. Rev. 40 (2011) on a hot day, so far only the toes are wet. The bracing plunge, 1459–1511. rewarding swim, and a cold beer – it’s Sydney after all – awaits. [32] A. Winter, M.D. Hager, G.R. Newkome, U.S. Schubert, Adv. Mater. 23 (2011) Go on, get involved, make some tridentate pyridine–pyrrolide 5728–5748. [33] K.M.C. Wong, V.W.W. Yam, Acc. Chem. Res. 44 (2011) 424–434. coordination complexes. Our hope is that this review is helpful [34] Z.F. Chen, J.J. Concepcion, M.K. Brennaman, P. Kang, M.R. Norris, P.G. Hoertz, T. and stimulates an uptake of this interesting coordination chem- J. Meyer, P. Natl, Acad. Sci. USA 109 (2012) 15606–15611. istry. We feel certain that pyridine–pyrrolide metal coordination [35] R. Sakamoto, S. Katagiri, H. Maeda, H. Nishihara, Coord. Chem. Rev. 257 (2013) 1493–1506. complexes with deep function and awesome utility lie just ahead; [36] A.K. Pal, G.S. Hanan, Chem. Soc. Rev. 43 (2014) 6184–6197. one of us has bet his PhD on it. [37] C. Wang, X.Q. Hao, M. Wang, C.L. Guo, B.Q. Xu, E.N. Tan, Y.Y. Zhang, Y.H. Yu, Z. Y. Li, H.B. Yang, M.P. Song, X.P. Li, Chem. Sci. 5 (2014) 1221–1226. [38] R. Sakamoto, K.H. Wu, R. Matsuoka, H. Maeda, H. Nishihara, Chem. Soc. Rev. Declarations of interest 44 (2015) 7698–7714. [39] K.J. Young, B.J. Brennan, R. Tagore, G.W. Brudvig, Acc. Chem. Res. 48 (2015) 567–574. None. [40] I. Eryazici, C.N. Mooreﬁeld, G.R. Newkome, Chem. Rev. 108 (2008) 1834– 1895. Acknowledgements [41] E.C. Constable, Chem. Soc. Rev. 36 (2007) 246–253. [42] F.G. Bordwell, Acc. Chem. Res. 21 (1988) 456–463. [43] E.R. King, T.A. Betley, J. Am. Chem. Soc. 131 (2009) 14374–14380. We thank the Australian Research Council for funding for B.D.’s [44] G.T. Sazama, T.A. Betley, Inorg. Chem. 49 (2010) 2512–2524. salary (Grant No. DP160104383). J.N.M. is grateful for an Australian [45] J.A. Flores, J.G. Andino, N.P. Tsvetkov, M. Pink, R.J. Wolfe, A.R. Head, D.L. Lichtenberger, J. Massa, K.G. Caulton, Inorg. Chem. 50 (2011) 8121–8131. Postgraduate Award. S.B.C. is grateful for his co-authors’ impetus [46] K. Searles, A.K. Das, R.W. Buell, M. Pink, C.H. Chen, K. Pal, D.G. Morgan, D.J. and thankful the job is done. Mindiola, K.G. Caulton, Inorg. Chem. 52 (2013) 5611–5619. James McPherson – June 2019 48 Chapter 1: Introduction

J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332 331

[47] N. Komine, R.W. Buell, C.H. Chen, A.K. Hui, M. Pink, K.G. Caulton, Inorg. Chem. [96] L.H. Yu, Z.K. Keane, J.W. Ciszek, L. Cheng, J.M. Tour, T. Baruah, M.R. Pederson, 53 (2014) 1361–1369. D. Natelson, Phys. Rev. Lett. 95 (2005) 256803. [48] G.T. Sazama, T.A. Betley, Inorg. Chem. 53 (2014) 269–281. [97] D. Natelson, L.H. Yu, J.W. Ciszek, Z.K. Keane, J.M. Tour, Chem. Phys. 324 (2006) [49] Y. Zhang, J.L. Petersen, C. Milsmann, J. Am. Chem. Soc. 138 (2016) 13115– 267–275. 13118. [98] G.D. Scott, Z.K. Keane, J.W. Ciszek, J.M. Tour, D. Natelson, Phys. Rev. B 79 [50] T. Ghorui, S. Roy, S. Pramanik, K. Pramanik, Dalton Trans. 45 (2016) 5720– (2009) 165413. 5729. [99] Y.P. Wang, X.H. Hu, Y.F. Wang, J. Pan, X.Y. Yi, Polyhedron 102 (2015) 782–787. [51] B.S. Billow, T.J. McDaniel, A.L. Odom, Nat. Chem. 9 (2017) 837–842. [100] Y.-P. Wang, J.-J. Xiao, X.-H. Hu, X.-Y. Yi, Inorg. Chim. Acta 435 (2015) 125– [52] M. Wang, J. England, T. Weyhermuller, K. Wieghardt, Eur. J. Inorg. Chem. 130. (2015) 1511–1523. [101] X.-H. Hu, Y. Liang, C. Li, X.-Y. Yi, Dalton Trans. 43 (2014) 2458–2464. [53] J. England, E. Bill, T. Weyhermuller, F. Neese, M. Atanasov, K. Wieghardt, [102] W.-Z. Fang, Y.-P. Wang, Y.-F. Wang, S.-C. Zhang, X.-Y. Yi, RSC Adv. 5 (2015) Inorg. Chem. 54 (2015) (2018) 12002–12018. 8996–9001. [54] M. Wang, J. England, T. Weyhermuller, K. Wieghardt, Inorg. Chem. 53 (2014) [103] J.M. Rao, M.C. Hughes, D.J. Macero, Inorg. Chim. Acta 16 (1976) 231–236. 2276–2287. [104] M.S. Alam, M. Stocker, K. Gieb, P. Mueller, M. Haryono, K. Student, A. [55] A.C. Bowman, J. England, S. Sproules, T. Weyhermuller, K. Wieghardt, Inorg. Grohmann, Angew. Chem. Int. Ed. 49 (2010) 1159–1163. Chem. 52 (2013) 2242–2256. [105] D.N. Bowman, A. Bondarev, S. Mukherjee, E. Jakubikova, Inorg. Chem. 54 [56] J. England, C.C. Scarborough, T. Weyhermuller, S. Sproules, K. Wieghardt, Eur. (2015) 8786–8793. J. Inorg. Chem. (2012) 4605–4621. [106] L. Yan, J.M. Seminario, J. Phys. Chem. A 109 (2005) 6628–6633. [57] C.C. Scarborough, K. Wieghardt, Inorg. Chem. 50 (2011) 9773–9793. [107] Y. Liu, H. Liang, Z. Chen, Q. Qin, Q. Xu, Y. Jiang, Guangxi Normal University, P. [58] C.C. Scarborough, S. Sproules, T. Weyhermuller, S. DeBeer, K. Wieghardt, Rep. China, 2015, CN104370908A,12pp. Inorg. Chem. 50 (2011) 12446–12462. [108] Z.F. Chen, Q.P. Qin, J.L. Qin, Y.C. Liu, K.B. Huang, Y.L. Li, T. Meng, G.H. Zhang, Y. [59] J. McPherson, T. Elton, S. Colbran, in preparation. Peng, X.J. Luo, H. Liang, J. Med. Chem. 58 (2015) 2159–2179. [60] B. Das, J. McPherson, S. Colbran, Coord. Chem. Rev., submitted for publication. [109] C. Arana, S. Yan, M. Keshavarz-K, K.T. Potts, H.D. Abruna, Inorg. Chem. 31 [61] F. Hein, F. Melichar, Pharmazie 9 (1954) 455–460. (1992) 3680–3682. [62] J.W. Ciszek, Z.K. Keane, L. Cheng, M.P. Stewart, L.H. Yu, D. Natelson, J.M. Tour, [110] R. Min, X.-H. Hu, X.-Y. Yi, S.-C. Zhang, J. Cent. South Univ. (Engl. Ed.) 22 (2015) J. Am. Chem. Soc. 128 (2006) 3179–3189. 1619–1625. [63] F. Hein, U. Beierlein, Pharm. Zentralhalle Dtschl. 96 (1957) 401–421. [111] R.V. Bynum, W.E. Hunter, R.D. Rogers, J.L. Atwood, Inorg. Chem. 19 (1980) [64] L. Kürti, B. Czakó, Strategic Applications of Named Reactions in Organic 2368–2374. Synthesis, 1 ed., Elsevier Academic Press, 2005. [112] M. Mathis, W. Harsha, T.W. Hanks, R.D. Bailey, G.L. Schimek, W.T. Pennington, [65] D. St.C. Black, 1H-Pyrroles, Chapt. 9:13, in: W. Ando, U. Bergsträßer, D. Black, Chem. Mater. 10 (1998) 3568–3575. H. Heydt, B. König (Eds.), Science of Synthesis: Houben-Weyl Methods of [113] K. Nomura, H. Sato, K. Kobayashi, K. Watanabe, in, Fujifilm Corporation, Molecular Transformations, Vol. 9: Fully Unsaturated Small-Ring Japan, 2014, WO2014050528A1, 100 pp. Heterocycles and Monocyclic Five-Membered Hetarenes with One [114] K. Nomura, H. Sato, K. Kobayashi, Fujifilm Corporation, Japan, Heteroatom, Thieme, Stuttgart, 2014. WO2014050578A1, 2014, 122 pp. [66] L. Knorr, Berichte 18 (1885) 299–311. [115] P. Daw, R. Petakamsetty, A. Sarbajna, S. Laha, R. Ramapanicker, J.K. Bera, J. Am. [67] A. Pinner, Berichte 34 (1901) 4234–4253. Chem. Soc. 136 (2014) 13987–13990. [68] R.A. Jones, M. Karatza, T.N. Voro, P.U. Civcir, A. Franck, O. Ozturk, J.P. Seaman, [116] H. Ikemizu, T. Oshiyama, M. Nishizeki, E. Kato, Konica Minolta Holdings Inc, A.P. Whitmore, D.J. Williamson, Tetrahedron 52 (1996) 8707–8724. Japan, 2011, JP2011129744A, 75pp. [69] G.H. Imler, Z. Lu, K.A. Kistler, P.J. Carroll, B.B. Wayland, M.J. Zdilla, Inorg. [117] J. Tamaru, E. Kato, M. Nishizeki, N. Sekine, Konica Minolta Holdings Inc, Japan, Chem. 51 (2012) 10122–10128. 2007, JP2007169541A, 72pp. [70] J.-J. Chen, Z.-L. Gan, Q. Huang, X.-Y. Yi, Inorg. Chim. Acta 466 (2017) 93–99. [118] C. Xia, C. Lin, Universal Display Corporation, USA, 2015, US20150155501A1, [71] Y. Bai, S. Xiang, M.L. Leow, X.-W. Liu, Chem. Commun. 50 (2014) 6168– 217pp. 6170. [119] N.P. BuHoi, F. Pärin, P. Jacquignon, Chem. Zentralbl. 137 (1966) 5971–5972. [72] D.C. Owsley, J.M. Nelke, J.J. Bloomfield, J. Org. Chem. 38 (1973) 901–903. [120] R.P. Thummel, V. Hegde, J. Org. Chem. 54 (1989) 1720–1725. [73] Z. Zhang, J.M. Lim, M. Ishida, V.V. Roznyatovskiy, V.M. Lynch, H.-Y. Gong, X. [121] V. Hegde, P. Madhukar, J.D. Madura, R.P. Thummel, J. Am. Chem. Soc. 112 Yang, D. Kim, J.L. Sessler, J. Am. Chem. Soc. 134 (2012) 4076–4079. (1990) 4549–4550. [74] A. McSkimming, V. Diachenko, R. London, K. Olrich, C.J. Onie, M.M. [122] V. Hegde, C.Y. Hung, P. Madhukar, R. Cunningham, T. Hopfner, R.P. Thummel, Bhadbhade, M.P. Bucknall, R.W. Read, S.B. Colbran, Chem. Eur. J. 20 (2014) J. Am. Chem. Soc. 115 (1993) 872–878. 11445–11456. [123] S. Tollari, S. Cenini, A. Rossi, G. Palmisano, J. Mol. Catal. A: Chem. 135 (1998) [75] P.J.W. Fuchs, K. Zeitler, J. Org. Chem. 82 (2017) 7796–7805. 241–248. [76] G. Clavier, P. Audebert, Chem. Rev. 110 (2010) 3299–3314. [124] W.-L. Jia, Q.-D. Liu, R. Wang, S. Wang, Organometallics 22 (2003) 4070–4078. [77] D.L. Boger, Chem. Rev. 86 (1986) 781–793. [125] J.-I. Setsune, M. Toda, K. Watanabe, P.K. Panda, T. Yoshida, Tetrahedron Lett. [78] W.S. Wilson, R.N. Warrener, Tetrahedron Lett. 11 (1970) 4787–4790. 47 (2006) 7541–7544. [79] R.A. Carboni, R.V. Lindsey, J. Am. Chem. Soc. 81 (1959) 4342–4346. [126] J.-I. Setsune, K. Watanabe, J. Am. Chem. Soc. 130 (2008) 2404–2405. [80] H. Bakkali, C. Marie, A. Ly, C. Thobie-Gautier, J. Graton, M. Pipelier, S. [127] H. Zhang, J. Lee, A.D. Lammer, X. Chi, J.T. Brewster, V.M. Lynch, H. Li, Z. Zhang, Sengmany, E. Leonel, J.-Y. Nedelec, M. Evain, D. Dubreuil, Eur. J. Org. Chem. J.L. Sessler, J. Am. Chem. Soc. 138 (2016) 4573–4579. (2008) 2156–2166. [128] J. Dash, R. Nath Das, N. Hegde, G.D. Pantos, P.S. Shirude, S. Balasubramanian, [81] G.T. Manh, R. Hazard, A. Tallec, J.P. Pradere, D. Dubreuil, M. Thiam, L. Toupet, Chem. Eur. J. 18 (2012) 554–564. Electrochim. Acta 47 (2002) 2833–2841. [129] J. Dash, P.S. Shirude, S. Balasubramanian, Chem. Commun. (2008) 3055–3057. [82] G. Lautrette, C. Aube, Y. Ferrand, M. Pipelier, V. Blot, C. Thobie, B. Kauffmann, [130] S. Bhowmik, R.N. Das, B. Parasar, J. Dash, Chem. Commun. 49 (2013) 1817– D. Dubreuil, I. Huc, Chem. Eur. J. 20 (2014) 1547–1553. 1819. [83] Y. Ferrand, A.M. Kendhale, B. Kauffmann, A. Grelard, C. Marie, V. Blot, M. [131] W. Van Rossom, T.G. Terentyeva, K. Sodeyama, Y. Matsushita, Y. Tateyama, K. Pipelier, D. Dubreuil, I. Huc, J. Am. Chem. Soc. 132 (2010) 7858–7859. Ariga, J.P. Hill, Org. Biomol. Chem. 12 (2014) 5492–5499. [84] A. Tabatchnik-Rebillon, C. Aube, H. Bakkali, T. Delaunay, G.T. Manh, V. Blot, C. [132] A.M. Kelly-Rowley, L.A. Cabell, E.V. Anslyn, J. Am. Chem. Soc. 113 (1991) Thobie-Gautier, E. Renault, M. Soulard, A. Planchat, J.-Y. Le Questel, R. Le 9687–9688. Guevel, C. Guguen-Guillouzo, B. Kauffmann, Y. Ferrand, I. Huc, K. Urgin, S. [133] A.M. Kelly-Rowley, V.M. Lynch, E.V. Anslyn, J. Am. Chem. Soc. 117 (1995) Condon, E. Leonel, M. Evain, J. Lebreton, D. Jacquemin, M. Pipelier, D. 3438–3447. Dubreuil, Chem. Eur. J. 16 (2010) (1889) 11876–11889. [134] R.J.P. Corriu, B. Geng, J.J.E. Moreau, C. Vernhet, Chem. Commun. (1991) 211– [85] V. Nair, K.H. Kim, J. Heterocycl. Chem. 13 (1976) 873–876. 213. [86] J. Lehuede, Y. Mettey, J.M. Vierfond, Synth. Commun. 26 (1996) 793–802. [135] I. Sircar, P. Furth, B.R. Teegarden, M. Morningstar, N. Smith, R.C. Griffith, [87] J. Lehuede, B. Fauconneau, L. Barrier, M. Ourakow, A. Piriou, J.-M. Vierfond, Tanabe Seiyaku Co., Ltd, Japan. 2001, WO2001030781A2, 195 pp. Eur. J. Med. Chem. 34 (1999) 991–996. [136] Y.Y. Rusakov, L.B. Krivdin, E.Y. Schmidt, A.I. Mikhaleva, B.A. Trofimov, Magn. [88] H. Von Dobeneck, D. Wolkenstein, H. Deubel, H. Reinhard, Chem. Ber. 102 Reson. Chem. 44 (2006) 692–697. (1969) 3500–3507. [137] B.A. Trofimov, A.M. Vasil’tsov, E.Y. Schmidt, N.V. Zorina, A.V. Afonin, A.B.I. [89] R. Lavilla, T. Gotsens, M. Guerrero, C. Masdeu, M.C. Santano, C. Minguillon, J. Mikhaleva, K.B. Petrushenko, I.A. Ushakov, L.B. Krivdin, V.K. Belsky, L.I. Bosch, Tetrahedron 53 (1997) 13959–13968. Bryukvina, Eur. J. Org. Chem. (2005) 4338–4345. [90] Q.-P. Qin, J.-L. Qin, T. Meng, W.-H. Lin, C.-H. Zhang, Z.-Z. Wei, J.-N. Chen, Y.-C. [138] A.V. Afonin, I.A. Ushakov, D.E. Simonenko, E.Y. Shmidt, N.V. Zorina, A.I. Liu, H. Liang, Z.-F. Chen, Eur. J. Med. Chem. 124 (2016) 380–392. Mikhaleva, B.A. Trofimov, Russ. J. Org. Chem. 41 (2005) 1516–1521. [91] Z.Z. Wei, Q.P. Qin, J.N. Chen, Z.F. Chen, Molecules 21 (2016) 1534. [139] T. Nagata, K. Tanaka, Bull. Chem. Soc. Jpn. 75 (2002) 2469–2470. [92] B.A. Frazier, V.A. Williams, P.T. Wolczanski, S.C. Bart, K. Meyer, T.R. Cundari, E. [140] J.J. Klappa, A.E. Rich, K. McNeill, Org. Lett. 4 (2002) 435–437. B. Lobkovsky, Inorg. Chem. 52 (2013) 3295–3312. [141] N. Komine, R.W. Buell, C.-H. Chen, A.K. Hui, M. Pink, K.G. Caulton, Inorg. [93] Y.-Q. Zhong, H.-Q. Xiao, X.-Y. Yi, Dalton Trans. 45 (2016) 18113–18119. Chem. 53 (2014) 1361–1369. [94] W. Sotoyama, T. Satou, H. Sato, A. Matsuura, N. Sawatari, Fuji Photo Film Co., [142] S.A. Moya, R. Pastene, H. Le Bozec, M. Fernandez, A.J. Pardey, P. Baricelli, R. Ltd, Japan. 2006, WO2006025124A1, 74 pp. Lopez, Bol. Soc. Chil. Quim. 44 (1999) 423–426. [95] L.H. Yu, Z.K. Keane, J.W. Ciszek, L. Cheng, M.P. Stewart, J.M. Tour, D. Natelson, [143] L.N. Grant, M.E. Carroll, P.J. Carroll, D.J. Mindiola, Inorg. Chem. 55 (2016) Phys. Rev. Lett. 93 (2004) 266802. 7997–8002.

James McPherson – June 2019 49 Chapter 1: Introduction

332 J.N. McPherson et al. / Coordination Chemistry Reviews 375 (2018) 285–332

[144] Q. Liu, L. Thorne, I. Kozin, D. Song, C. Seward, M. D’Iorio, Y. Tao, S. Wang, J. [161] G.M. Whitesides, J. Deutch, Nature 469 (2011) 21–22. Chem. Soc., Dalton Trans. (2002) 3234–3240. [162] D. Parker, Coord. Chem. Rev. 205 (2000) 109–130. [145] K. Searles, S. Fortier, M.M. Khusniyarov, P.J. Carroll, J. Sutter, K. Meyer, D.J. [163] J.C.G. Bunzli, Acc. Chem. Res. 39 (2006) 53–61. Mindiola, K.G. Caulton, Angew. Chem. Int. Ed. 53 (2014) 14139–14143. [164] C.P. Montgomery, B.S. Murray, E.J. New, R. Pal, D. Parker, Acc. Chem. Res. 42 [146] M. Bottger, B. Wiegmann, S. Schaumburg, P.G. Jones, W. Kowalsky, H.H. (2009) 925–937. Johannes, Beilstein J. Org. Chem. 8 (2012) 1037–1047. [165] E.G. Moore, A.P.S. Samuel, K.N. Raymond, Acc. Chem. Res. 42 (2009) [147] S. Bivaud, S. Goeb, J.-Y. Balandier, M. Chas, M. Allain, M. Salle, Eur. J. Inorg. 542–552. Chem. 2014 (2014) 2440–2448. [166] L.D. Carlos, R.A.S. Ferreira, V.D. Bermudez, B. Julian-Lopez, P. Escribano, Chem. [148] S. Bivaud, J.-Y. Balandier, M. Chas, M. Allain, S. Goeb, M. Salle, J. Am. Chem. Soc. Rev. 40 (2011) 536–549. Soc. 134 (2012) (1970) 11968–11970. [167] X.Y. Huang, S.Y. Han, W. Huang, X.G. Liu, Chem. Soc. Rev. 42 (2013) 173–201. [149] J. Bunzen, J. Iwasa, P. Bonakdarzadeh, E. Numata, K. Rissanen, S. Sato, M. [168] P. Braunstein, F. Naud, Angew. Chem. Int. Ed. 40 (2001) 680–699. Fujita, Angew. Chem. Int. Ed. 51 (2012) 3161–3163. [169] E. Poverenov, M. Gandelman, L.J.W. Shimon, H. Rozenberg, Y. Ben-David, D. [150] K. Harris, D. Fujita, M. Fujita, Chem. Commun. 49 (2013) 6703–6712. Milstein, Organometallics 24 (2005) 1082–1090. [151] D. Fujita, H. Yokoyama, Y. Ueda, S. Sato, M. Fujita, Angew. Chem. Int. Ed. 54 [170] R. Lindner, B. van den Bosch, M. Lutz, J.N.H. Reek, J.I. van der Vlugt, (2015) 155–158. Organometallics 30 (2011) 499–510. [152] Q.F. Sun, J. Iwasa, D. Ogawa, Y. Ishido, S. Sato, T. Ozeki, Y. Sei, K. Yamaguchi, [171] J.I. van der Vlugt, Eur. J. Inorg. Chem. (2012) 363–375. M. Fujita, Science 328 (2010) 1144–1147. [172] T. Kojima, K. Nakayama, M. Sakaguchi, T. Ogura, K. Ohkubo, S. Fukuzumi, J. [153] M. Tominaga, K. Suzuki, M. Kawano, T. Kusukawa, T. Ozeki, S. Sakamoto, K. Am. Chem. Soc. 133 (2011) 17901–17911. Yamaguchi, M. Fujita, Angew. Chem. Int. Ed. 43 (2004) 5621–5625. [173] D.B. Grotjahn, Chem. Eur. J. 11 (2005) 7146–7153. [154] K. Suzuki, M. Tominaga, M. Kawano, M. Fujita, Chem. Commun. (2009) 1638– [174] C.Q. Li, B. Villa-Marcos, J.L. Xiao, J. Am. Chem. Soc. 131 (2009) 6967–6969. 1640. [175] M. Rueping, R.M. Koenigs, I. Atodiresei, Chem. Eur. J. 16 (2010) 9350–9365. [155] D. Fujita, Y. Ueda, S. Sato, H. Yokoyama, N. Mizuno, T. Kumasaka, M. Fujita, [176] B.G. Zhao, Z.B. Han, K.L. Ding, Angew. Chem. Int. Ed. 52 (2013) 4744–4788. Chem 1 (2016) 91–101. [177] M. Raynal, P. Ballester, A. Vidal-Ferran, P. van Leeuwen, Chem. Soc. Rev. 43 [156] D. Fujita, Y. Ueda, S. Sato, N. Mizuno, T. Kumasaka, M. Fujita, Nature 540 (2014) 1660–1733. (2016) 563–566. [178] S.S. Nadif, M.E. O’Reilly, I. Ghiviriga, K.A. Abboud, A.S. Veige, Angew. Chem. [157] S. Borman, ‘‘The biggest self-assembled supramolecular cage ever” in Chem. Int. Ed. 54 (2015) 15138–15142. Eng. News: https://cen.acs.org/articles/94/web/2016/12/biggest-self- [179] A. McSkimming, S.B. Colbran, Chem. Soc. Rev. 42 (2013) 5439–5488. assembled-supramolecular-cage.html (accessed 30 Oct. 2017). [180] Coogee Beach: https://www.sydney.com/destinations/sydney/sydney- [158] C. Zeng, Y. Chen, K. Kirschbaum, K.J. Lambright, R. Jin, Science 354 (2016) east/coogee/attractions/coogee-beach (accessed 30 October 2017). 1580–1584. [181] School of Chemistry, UNSW: http://www.chemistry.unsw.edu.au/ (accessed [159] N.A. Sakthivel, S. Theivendran, V. Ganeshraj, A.G. Oliver, A. Dass, J. Am. Chem. 30 October 2017). Soc. 139 (2017) 15450–15459. [182] UNSW-Australia’s Global University: https://www.unsw.edu.au (accessed 30 [160] C.P. Snow, The Physicists, Little Brown & Co, 1981. October 2017).

James McPherson – June 2019 50 Chapter 2: Geometric Analysis of Strain in Pincer Ligands

2 GEOMETRIC ANALYSIS OF STRAIN IN PINCER LIGANDS

2.1 Prior Publication Statement

The work in this chapter was published by the American Chemical Society in Inorganic

Chemistry in 2018. This article has been reprinted with permission from McPherson, J. N.;

Elton, T. E.; Colbran, S. B., A Strain-Deformation Nexus within Pincer Ligands: Application

to the Spin States of Iron(II) Complexes, Inorg. Chem. 2018, 57 (19), 12312–12322. Copyright

2.2 Contributions

JNM designed the experiments, performed calculations and analysis of all data, including

the crystallographic database; wrote the manuscript and ESI; and responded to reviewer

comments.

TEE assisted with the design of theoretical calculations and the analyses of structural

data, and wrote the theoretical section of the experimental methods. SBC oversaw experimental

design and direction, and refined the manuscript. SBC mentored JNM and TEE.

2.3 Foreword

This work quantifies the geometric compatibilities of three-ringed pincer ligands—

including 2,5-di(2-pyridyl)pyrrolides (dpp–, or D– in this chapter), the focus of this PhD

research—with the metal ions that they bind, by considering the strain which builds within each

ligand as it deforms from its “unbound” geometry into a “coordinated” geometry. A relationship

between the change in separation of the two terminal donor atoms (∆ρ) and the energetic

James McPherson – June 2019 51 Chapter 2: Geometric Analysis of Strain in Pincer Ligands

destabilisation (∆EL(strain)) was discovered, and thus ρL(relax) was suggested as a useful steric parameter that was considered in subsequent work during this PhD research.

James McPherson – June 2019 52 Chapter 2: Geometric Analysis of Strain in Pincer Ligands

Article

Cite This: Inorg. Chem. 2018, 57, 12312−12322 pubs.acs.org/IC

A Strain-Deformation Nexus within Pincer Ligands: Application to the Spin States of Iron(II) Complexes James N. McPherson, Timothy E. Elton, and Stephen B. Colbran*

School of Chemistry, University of New South Wales, Sydney, New South Wales 2052, Australia

*S Supporting Information

ABSTRACT: The substitution of a pyrrolide ring for one (or more) pyridyl rings within the ubiquitous terpyridine (tpy, A) scaffold results in more open geometries of the pyridine− pyrrolide chelate ligands. DFT calculations (B3LYP-GD3BJ/ 6-31G**) demonstrate that the more open geometries of the unbound ligands are mismatched with the “pinched in” geometries required to chelate transition metal ions (e.g., Zn2+). The strain which builds within these ligands Δ ( EL(strain)) as they bind transition metal ions can be related to changes in a single geometric parameter: the separation between the two terminal N atoms (ρ). This relationship applies more generally to other three-ringed tridentate pincer ligands, including those with different donor groups. The approach was applied to homoleptic iron(II) complexes to investigate the contribution of the steric effects operating within the ligands to the different magnetic properties, including spin crossover (SCO) activities, of these systems.

his work focuses on the geometric consequences of catalytic reactions compared to their terpyridine−metal substitution of one (or more) of the six-membered rings analogues. For example, homoleptic iron(II) complexes with T − in 2,2’:6′,2″-terpyridine (tpy, scaffold A in Figure 1), the derivatives of ligand scaffold D are known to be unstable in prototypal meridional metal-binding κ3-N ligand, by a five- air and as such have only been partially characterized.11,12As membered pyrrolide ring. It is anticipated that geometric strain predicted in a theoretical study,13 these complexes were will build in a metal complex upon the substitution of a experimentally determined11,12 to be high-spin (HS) config- pyridine ring by a pyrrolide ring with consequences of urations of iron(II), unlike terpyridyl analogues which are fundamental importance for the electronic and, therefore, all usually low-spin (LS) and stable.14 There is a single report of a physicochemical properties and the reactivity. Herein, a LS [Fe(D)2] complex, which was tentatively assigned to a straightforward parametrization of the ligand strain within a symmetric product observed in solution by 1H NMR studies, metal complex of a three-ring, tridentate meridional ligand is but the species was never isolated, nor otherwise charac- developed that is demonstrated to correlate well, and most terized.15 usefully, with structure determined from single crystal X-ray The novel chemistry for metal complexes of ligands with

Downloaded via UNIV OF NEW SOUTH WALES on March 27, 2019 at 01:30:16 (UTC). ff di raction (SC-XRD) studies. Application to the spin states of scaffolds B−E is not surprising given the electronic and iron(II) complexes is explored. fi See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles. ff geometric properties of the ve-membered pyrrolide N-donor Metal complexes of ligands based on the terpyridine sca old ring differ from those of the six-membered pyridyl N-donor A, Figure 1, are numerous and extensively studied, and their σ 1−5 − ring. Both the pyridine and pyrrolide rings are good -donor properties are well-understood. Metal terpyridine com- ligands. Odom recently introduced a new tool for quantitative plexes are especially sought after for their functional utility, determination of ligand donation effects to a high-valent metal especially as photo, magnetic, redox, and catalytic centers that ion, the ligand donor parameter (LDP), where larger LDP can be purposefully assembled into strategic molecular, values correlate with weaker combined σ-andπ-donor macromolecular, or supramolecular architectures for advanta- 16 6−9 strength. The LDP assists prediction of useful chemical geous, targeted properties. properties such as the rates of a reaction catalyzed by a series Anionic analogues of the meridional, planar terpyridine σ ligand scaffold A in which one or more of the neutral N-donor, of metal catalysts in which an ancillary -donor ligand is varied. 6 π-electron pyridyl rings are substituted for anionic N-donor, The LDP value for the N-donor pyrrolide ring is 13.6 kcal mol−1, about the same as fluoride ion (13.4 kcal mol−1) but 6 π-electron, pyrrolide rings, scaffolds B−E, are also depicted − − ff larger that than for phenolate (PhO : 12.0 kcal mol 1), t-butyl in Figure 1. The chemistry of ligands with these sca olds was − −1 recently reviewed.10 Metal complexes of these ligands, by alkoxide (t-BuO : 10.6 kcal mol ), and dimethylamide comparison with those with terpyridine, are sparse. However, enough is the known to demonstrate that examples of the Received: July 19, 2018 complexes display novel physicochemical properties and Published: September 10, 2018

Inorganic Chemistry Article

Figure 1. Drawings of ligand scaffolds comprised of three meridionally disposed, 6 π-electron, pyridine and/or pyrrolide donor rings: 2,2’:6′,2″- terpyridine scaffold (A), 2-(2,2′-bipyridin-6-yl)pyrrolide scaffold (B), 2,6-di(2-pyrrolide)pyridine scaffold (C), 2,5-di(2-pyridyl)pyrrolide scaffold (D), and 5-(2-pyridyl)-2,2′-bipyrrolide scaffold (E) (not shown is the 2,2’:5′,2″-terpyrrolide scaffold as a genuine metal complex of a ligand with this scaffold is yet to be reported).

(Me N−: 9.3 kcal mol−1)ions.Terpyridineisaredox might be exploited during ligand design to target a particular 2 − noninnocent ligand that may act as an electron acceptor,17 23 function. whereas pyrrolide in distinct contrast and in keeping with its The strategy which we employ in this work to probe the charge and raised electron density is a redox noninnocent steric effects within these pyridine−pyrrolide pincer ligands − ligand that may act as an electron donor.24 26 was inspired by the work of Comba et al. that explored the The differences in electronic structure upon substitution of a preorganization of tetrathiamacrocyclic ligands for chelating pyridine N-donor by a pyrrolide N-donor are exacerbated by nickel(II) or palladium(II) ions.31 Comba et al. found an the fact that the former is a six-membered ring whereas the empirical correlation between the changes in a structural latter is a five-membered ring. The geometric consequences are displacement parameter from the unbound to bound ligand severe. For example, as suggested in Figure 2, such substitution geometries (as determined by SC-XRD) with the difference in energies of these forms calculated using molecular mechanics (i.e., ligands which deformed less on chelation with an ion were also at similar energies in their bound or unbound conformations).31 Comba’s displacement parameter involved the sum of six different interatomic separations, and while predictive accuracy may be improved by considering additional Figure 2. (A) Overlay of the ligand skeletons with the nitrogen atoms parameters, the fewer parameters involved, the more accessible (depicted by dots) of the central rings eclipsed for terpyridine (A in − and general and therefore more useful the prediction is likely red), dipyrrolidepyridine dianion (C2 in green), and dipyridylpyrrolide anion (D− in blue). (B) Idealized interior and exterior bond to be. angles about the ipso-atoms of linked six- and five-membered rings. At the outset, it was not clear that a single parameter would characterize ligand strain across representative ligands from all families of meridional ligand scaffolds A−E and usefully of a pyrrolide for pyridyl ring should always lead to an increase ff in the separation of the two terminal N donor atoms and an correlate with found (or predicted) structure. The e ects of fi the differing ring donicities, geometries, and different increase in the bend angle of the ligand (here de ned as the ff angle between the two inter-ring bond vectors). Upon binding substituents appended to a sca old, for example, were possible to a metal ion, the ligand will then “pinch in” to optimize the sources of confounding compromise. To construct a view of M−N distances. The minimum system energy, and the final ligand strain, we have used density functional theory (DFT) observed structure, corresponds to the compromise that calculations to optimize and compare the energies and − structures of free ligands (L) and their hypothetical zinc(II) reaches best M N bonding with least ligand strain. z+ Stresses and strains within coordination complexes have complex, Zn(L) , in the vacuum phase. We develop a useful been of much interest, particularly in the context of the entatic parametrization of ligand strain in a three-ring meridional state principle (i.e., the energization due to a misfit between ligand and correlation with structure that allows its estimation ligands and metal ions which results in a net lowering of the from the Cambridge Structural Database (CSD) data for a − relative transition state energy) which has been reviewed metal ligand complex. comprehensively.27,28 Most studies have focused on examining We then consider ligand strain within bis(pincer)iron(II) the immediate coordination geometry of a given metal center complexes that exhibit SCO activity. Understanding and and its deviations from the geometry “preferred” by the metal predicting SCO in metal complexes is still an important ion (i.e., by examining the M−L bond lengths and angles and challenge, with recent reports focusing on the electronic effects comparing them to some reference datum), although such of the ligands and avoiding steric considerations, by system- 32,33 approaches are complicated by the selection of such a atically modifying ligands that are structurally similar. It “preferred” reference geometry.27,29 Computational methods has long been known (and recently exploited by Shatruk and 34,35 have been used to probe the overall energies of complex co-workers) that the magnetic properties of iron(II) complexes can be tuned through steric effects (i.e., molecules, for example, Kroll et al. demonstrated by DFT and − spectroscopic methods that the geometric differences (strain) strain).14,36 38 Others have considered the strain within the alone between a pair of pentadentate ligands resulted in a first coordination sphere of iron(II) complexes with tridentate weakening of the ligand field by ∼20−30 kJ mol−1 in their pincer ligands and the magnetic properties which result. iron(II) complexes, with the more flexible (and stronger field) Halcrow, in particular, has proposed four geometric parameters ligand forming a spin crossover (SCO) active iron(II) from the L−Fe−L bond angles within the complexes to complex.30 Such approaches are often complicated by solvent characterize the distortion,39 and therefore strain, of the or co-ligand interactions, which must be deconvoluted from coordination geometry around each iron(II) center, and these − the overall energies in order to arrive at key parameters that have now been widely accepted.39 46 Our work is related but

12313 DOI: 10.1021/acs.inorgchem.8b02038 Inorg. Chem. 2018, 57, 12312−12322 James McPherson – June 2019 54 Chapter 2: Geometric Analysis of Strain in Pincer Ligands

Inorganic Chemistry Article tells the other half of the story: what happens to the ligand as it A, z =2;L=B, z =1;L=C, z =0;L=D, z =1;L=E, z = 0) were 48−50 binds these centers. performed using the B3LYP functional with the 6-31G** basis set.51 Dispersion corrections were included by adding the D3 version 52 ■ EXPERIMENTAL SECTION of Grimme’s dispersion with Becke−Johnson damping. The geometry optimizations were followed by calculation of vibrational Parameterization of Ligand Geometric Structure. The fi frequencies at the B3LYP level. Positive vibrational frequencies following parameters are used to de ne the deformation upon indicated that the coordinates for each structure were energy minima. binding of a tridentate κ3-N ligand to a metal ion: θ ° Previous studies have shown that this combination of theory and basis Bend angle ( / ): the angle between the two bond vectors from set provides accurate structural information with complexes of similar − the innermost ring to its two outer substituent rings, as shown in blue size and composition.53 57 Calculations without an empirical in Figure 3. dispersion correction returned similar results (see Table S1). Single-point energy calculations (B3LYP/6-31G**)forthe strained geometries were performed after the Cartesian coordinates of the ligand atoms were harvested from the theoretical [Zn(L)]z+ structures described above. The geometry of the ligand was then allowed to relax, although the inter-ring N−C−C−N torsion angles were constrained to 0.0°. The same method was employed for structures from experimental SC-XRD data; however, prior to the single-point energy calculation of the strained (coordinated) geometry, the atomic positions of the H atoms were first optimized (B3LYP/6-31G**) with the heavy atom (N, C, O, P, and S) positions fixed. Solid-State Structural Analysis. The Cartesian coordinates of the ligand atom positions from experimental SC-XRD structures were harvested from searches of the CSD and are listed with their Cambridge Crystallographic Data Centre (CCDC) identifier code and a citation to the original report (Tables S3 and S4). H atoms are not directly observed during SC-XRD structure determinations; their positions are typically fixed at positions inferred from neighboring heavier atoms. If left untreated then the crystallographically fixed H atom positions within the ligand structures contribute significant error in subsequent energy calculations. To find more appropriate H atom positions, prior to the initial single-point energy calculations of the “strained” geometries, the positions of all heavy atoms (i.e., non-H atoms) were fixed, and those of the H atoms were optimized (B3LYP/6-31G**). θ ··· Figure 3. Top: The bend angles ( , blue) and terminal N N For analysis of the homoleptic iron(II) complexes (Table S5), the separations (ρ, red) as measured in this study for terpyridine (tpy, A) ρ − L(coord) parameter was measured directly from the SC-XRD solid- above and dipyridyl pyrrolide (D ) below as examples. Bottom: The state structure prior to a single DFT geometry optimization (B3LYP/ μ deviation from planarity ( rms) was measured by the root-mean- 6-31G**) calculation for each ligand (again with N−C−C−N inter- square of the distances (μ, values as shown in Å) of each ring C or N ° ff ρ ring torsion angles constrained to 0 )toa ord L(relax). atom to the mean plane through them (shown in violet) as shown, for the exemplar theoretical [Zn(D)]+ complex, with the Zn(II) and H atoms omitted for clarity. ■ RESULTS AND DISCUSSION A zinc(II) ion was used to template ligands A−E2− into Terminal N···N donor separation (ρ/Å): the interatomic strained theoretical coordination geometries. The zinc(II) ion separation of the two terminal N donor atoms, as shown in red in was removed, and the ligand’s energy was calculated (B3LYP/ Figure 3. ** “ ” μ 6-31G ) before the ligand was allowed to relax. The Twisting, torsion and/or puckering out of the ligand plane ( rms/ Å): the root mean squared (rms) deviation (Å) of all ring C and N changes in geometric parameters were compared to the energy atoms from the mean plane through them, as shown in Figure 3. stabilization, and trends observed from this training set were DFT Calculations. DFT calculations were performed with the then compared to solid state data obtained from the CSD. Gaussian 2009 software package.47 Vacuum-phase geometry opti- DFT Geometry Optimizations. A hypothetical complex mizations for free ligand HD, and the zinc complexes [Zn(L)]z+ (L = of each parent ligand (all substituents are H atoms) with the

Table 1. Predicted (B3LYP+GD3BJ/6-31G**)Zn−N Distances (Å) and N−Zn−N Bond Angles (deg) for the Hypothetical a [Zn(L)]z+ Species

Zn−N(t) [py/pyr] Zn−N(c) [py/pyr] N(t)−Zn−N(c)N(t)−Zn−N′(t) [Zn(A)]2+ 1.978 [py] 1.993 [py] 83.41 166.83 1.995 [py]b 1.929 [py] 81.71b 167.53 [Zn(B)]1+ 1.878 [pyr] 85.82 [Zn(C)] 1.899 [pyr] 1.946 [py] 84.13 168.26 [Zn(D)]1+ 2.016 [py] 1.864 [pyr] 83.85 158.77 2.013 [py]b 1.889 [pyr] 83.03b 158.86 [Zn(E)] 1.912 [pyr] 86.78 aL is the parent ligand from scaﬀolds A−E, Figure 1. bFor nonsymmetric ligands B− and E2−, distances and angles involving pyridyl (py) N atoms are given ﬁrst, followed by those involving the pyrrolide (pyr) N atoms.

12314 DOI: 10.1021/acs.inorgchem.8b02038 Inorg. Chem. 2018, 57, 12312−12322 James McPherson – June 2019 55 Chapter 2: Geometric Analysis of Strain in Pincer Ligands

Inorganic Chemistry Article Δ zinc(II) ion without coligands was chosen for modeling to state of the free ligand (no constraints). Thus, EL(strain) is an minimize the electronic (symmetric d10 ion)and steric (no underestimation of the true strain energy, except when the other ligands) influences on the structure. The Zn−N and N− global minimum energy conformation of the ligand exactly Zn−N bond lengths and angles obtained are summarized in matches the constrained syn,syn conformation (see Figure S1). Table 1. All Zn−N(pyrrolide) bonds were shorter (1.864− Estimation of Geometric Deformation Parameters. 1.912 Å) than Zn−N(pyridyl) bonds (1.929−2.016 Å). The The three geometric parameters used to define the longest Zn−N(t)(t = terminal, outer ring) and the shortest deformation of a ligand upon binding to a metal ion (θ/°, − ρ μ Zn N(c)(c = central, inner ring) bonds were 2.016 and 1.864 /Å and rms/Å) are summarized in Figure 3, while Scheme 1 Å, respectively, predicted for [Zn(D)]1+ with a central presents a graphical summary of the methodology employed to pyrrolide donor. The three ligands with six-membered pyridyl obtain these geometric parameters. The changes in geometric cores (A, B−, and C2−) had N(t)−Zn−N′(t) bond angles parameters (Δθ and Δρ) were normalized against the values ± ° − 2− fi θ ρ within 167.5 0.8 , while those for D and E with ve- for the relaxed ligand ( L(relax) or L(relax) respectively) such that membered pyrrolide cores were less obtuse, within 158.8 ± the relative deformations are independent from the size of the 0.1°. ligand. The geometric parameters for relaxed syn,syn con- Calculation of the Internal Ligand Strain. Ligand strain formers of ligand scaffolds A−E which were studied in this energies are here defined as the rise in energy of a ligand as a work are summarized in Tables S2 and S3 for the coordination consequence of binding to a metal ion. The ligand strain complexes of these ligand scaffolds. energies were calculated using a strategy similar to those Hypothetical ZnL Complexes and Ligands. As reported by Comba et al.,31 Combariza and Vachet,58 and examples, Figure 4 presents overlaid views of the DFT- Wolney et al.,59,60 and are summarized graphically in Scheme 1. Single-point energy calculations (B3LYP/6-31G**) were

Scheme 1. Graphical Representation of the Methodology Employed for Estimation of the Internal Ligand Strain Energy and Geometric Parameters Describing Ligand Deformation for Ligands in Hypothetical [Zn(L)]z+ Complexes (Top) and for Ligands in Complexes Harvested from Experimental SC-XRD Structural Data (Bottom) Figure 4. Binding with a metal ion (divalent 3d10 zinc) is predicted to induce various degrees of strain within the ligands: overlaid images of the DFT-optimized geometries for the complex ([Zn(L)]z+, pink) and relaxed ligand (syn,syn-L, blue) for L = A, z =2;L=C2−, z =0;L= D−, z = 1. Note the considerable deviation from planarity in the complex of the ligand with a central ﬁve-membered ring (D−).

optimized structures for the relaxed geometries for ligands A, C2−, and D− (shown in blue) and their hypothetical zinc(II) complexes (shown in pink). The geometric parameters and strain energies (the difference between the energy of each ligand in its coordination geometry and that of the same ligand after relaxation, as described above) are summarized in Table 2 (see Table S2 for the geometric parameters for the relaxed ligands). Except for the [Zn(A)]2+ (A = tpy) structure, the ··· ≤ Δρ ρ ≤ relative changes in N N separations (13.0 / L(relax) 31.4%) were greater than the bend angle deformations (13.3 ≤ Δθ θ ≤ ̈ / L(relax) 22.2%), and the latter are naive to changes in the performed on the ligand L having the geometry found in the external ring angles about the ipso atoms. We therefore [Zn(L)]z+ complex, i.e., all ligand atoms in the complex were determine ρ to be the most sensitive “catch-all” parameter to fixed in position, the Zn2+ ion removed, and the energy, describe geometric change within the ligand after binding to a Δ 2− EL(coord), calculated. Next, the ligand was allowed to relax metal ion. Ligands A−C with central six-membered pyridyl − − − Δρ ρ during a geometry optimization with the inter-ring N C C rings were generally predicted to deform less ( / L(relax) fi ° − μ N torsion angles xed at 0 so that the meridional, tridentate between 13.0 25.5%) and remain planar ( rms = 0, see Figure − 2− Δρ ρ (syn,syn) conformations were optimized. The calculation 4) than ligands D or E with pyrrolide cores ( / L(relax) (B3LYP-6-31G**) of the electronic energy of the resulting between 25.5−31.4%), which were also predicted to twist and ff Δ Δμ − geometry a ords an energy, EL(relax), for the relaxed, but bend out of plane (with rms between 0.236 0.266 Å, see syn,syn-conformer of the ligand L. Thus, the “internal ligand Figure 4) when coordinated to a zinc(II) center. strain energy” is given by eq 1. For ligands A−C2−, both the relative degree of deformation Δρ ρ ΔEEE=Δ −Δ ( / L(relax) = 13.0, 19.2, and 25.5%) and ligand strain energy L(strain) L(coord) L(relax) (1) Δ −1 ( EL(strain) = 38, 67, and 113 kJ mol ) associated with fi Δ As de ned, the ligand strain energy, EL(strain), is not the entire chelation of the zinc(II) center increases with the number of strain energy that builds within each ligand as it distorts and flanking pyrrolide rings (0, 1, and 2 for A, B− and C2−, binds to a metal ion. The true strain energy is the difference in respectively) in the ligand scaffold. We offer a 2-fold − global energy minima for the coordinated and entirely relaxed explanation. First, as discussed above, shorter Zn Npyr bonds

12315 DOI: 10.1021/acs.inorgchem.8b02038 Inorg. Chem. 2018, 57, 12312−12322 James McPherson – June 2019 56 Chapter 2: Geometric Analysis of Strain in Pincer Ligands

Inorganic Chemistry Article Δθ θ Δρ ρ Δμ Table 2. Relative Deformations ( / L(relax) / L(relax) and rms) and Calculated Coordination Strain Energies (B3LYP/6- 31G**) for Ligand Backbones from Vacuum-Phase [Zn(L)]z+ Geometries (B3LYP+GD3BJ/6-31G**)

θ Δθ θ ρ Δρ ρ Δμ Δ −1 complex ligand (deg) / L(relax) (%) (Å) / L(relax) (%) rms (Å) EL(strain) (kJ mol ) [Zn(A)]2+ A 98.1 13.3 3.929 13.0 0 38 [Zn(B)]+ B− 96.3 17.3 3.850 19.2 0 67 [Zn(C)] C2− 94.0 21.7 3.777 25.5 0 113 [Zn(D)]+ D− 105.0 18.9 3.963 25.5 0.236 118 [Zn(E)] E2− 101.9 22.2 3.858 31.4 0.266 170

− z+ were predicted than Zn Npy across all theoretical [Zn(L)] complexes studied, which results in greater contraction of θ and ρ. This is compounded by the more “open” relaxed ligand θ ° − geometries ( L(relax): 113.3, 116.4, and 120.0 for A, B , and C2−, respectively) which result from alleviation of steric demands at the back of the ligands as the six-membered pyridyl rings are replaced by five-membered pyrrolide rings. We note θ 2− ° that L(relax) for C was equal to the ideal value (120 ) for a generic six-membered ring. These trends also hold for ligands with pyrrolide cores (relative deformations and ligand strain energies increase from 25.5 to 31.4% and from 118 to 170 kJ mol−1, respectively, for D− and E2−) although at much wider θ (and therefore larger ρ) due to the wider ideal bend angle (144°) for a five- membered core. For these ligands, additional modes of deformation are also predicted, such as twisting and puckering Δμ − 2− of the rings ( rms = 0.236 and 0.266 Å for D and E , respectively) due to the geometric consequences of the short − θ central Zn Npyr bonds (<1.9 Å, see Table 1) and the wide Figure 5. Fitting of ligand strain energy against relative ligand (>100°) associated with five-membered cores already dis- Δρ ρ z+ deformation ( / L(relax)) for vacuum-phase theoretical [Zn(L)] cussed above. complexes (training set) in black. Data from selected solid state SC- Correlation of Ligand Strain Energy and Deforma- XRD structures of complexes (accessed from the CSD, see Table S3) tion. A simple relationship between the coordination strain are shown as empty triangles: L = A (red); L = (C2)2− (purple), L = 3 2− − 2 − 2 and changes in our easily measured geometric parameters was (C ) (mustard); L = D (green) and L = (D ) (blue); where H2C 68 3 sought. The goodness of fits was evaluated by the residual root- = 2,6-bis(5-methyl-3-phenyl-pyrrol-2-yl)pyridine; H2C = 2,6-bis- 67 2 mean-square value (rms = (indol-2-yl)pyridine; and HD = methyl 4-methyl-2,5-bis(pyrid-2- residual yl)pyrrole-3-carboxylate.26 { ∑ (actual value−} predicted value)2 /n , where n = 5 is the number of observations in this case). The strain energies −1 Δθ θ are presented in Figure 5 (as squares) against the relative poorer correlations (rmsresidual= 29.9 kJ mol for / L(relax) ··· Δρ ρ −1 Δμ change in terminal N N donor separation ( / L(relax)). alone, or 16.1 kJ mol with rms included). Fitting these data by linear regression analysis (and enforcing Complexes and Ligands from the CSD. To test the that 0 deformation equates to 0 strain energy) results in the generality of the trends observed for the data on the trend defined by eq 2, which is also shown as the black hypothetical [Zn(L)]z+ species (eqs 1 and 2), “experimental” trendline in Figure 5, with standard error in the slope of just data was sought. Twenty solid-state structures of metal under 10%. complexes of tridentate pyridine−pyrrolide ligands were Δρ selected from the CSD. To challenge our methodology, a ΔE =±(460 37) kJ mol−1 range of metal coordination geometries (octahedral, trigonal L(strain) ρ L(relax) (2) bipyramidal, and square planar), and metal centers (mostly −1 transition metals of various oxidation states but also two The rmsresidual was improved from 17.7 kJ mol for eq 2 to lead(IV) structures from the same unit cell) were deliberately −1 Δμ 11.9 kJ mol when the rms parameter was included (eq 3), selected. No structures were available through the CSD for albeit with a large increase in error. It can be concluded that ligand scaffolds B− or E2−,10 so our “experimental” study was the out-of-plane deformations (twisting, puckering, and 2− − fi limited to derivatives of A, C , and D . The ligand strain torsion) make signi cant but somewhat unpredictable (given energies and the ligand deformation parameters were the large error) contributions to the buildup of strain within determined as graphically summarized in Scheme 1. All data these ligands. obtained are summarized in Table S3. Δρ The calculated ligand strain energy from the solid state Δ=±E (395 44) L(strain) ρ structural data was in poorer agreement to eq 3 (rmsresidual = L(relax) 15.0 kJ mol−1) than for the theoretical vacuum-phase −1 L z+ +±Δ(127 66)μ kJ mol (3) [Zn( )] training set. This was not particularly surprising. rms The largely unpredictable intermolecular interactions inherent Analyses involving the relative bend angle deformations (Δθ/ in crystal packing61 may also contribute to deviations from θ L(relax)) were also investigated (see Figure S2), but resulted in coplanarity of the aromatic ligand ring systems. For example,

12316 DOI: 10.1021/acs.inorgchem.8b02038 Inorg. Chem. 2018, 57, 12312−12322 James McPherson – June 2019 57 Chapter 2: Geometric Analysis of Strain in Pincer Ligands

Inorganic Chemistry Article neither of the solid-state structures of HD in the literature are metal binding therefore involves a smaller relative geometric μ coplanar ( rms = 0.112 and 0.169 Å for the structures with deformation. CCDC codes QOBREF62 and QOBREF01,63 respectively), Beyond Pyridyl−Pyrrolide Pincer Ligands. To push the μ 10 unlike the vacuum-phase optimized geometry ( rms = 0). In generality of the approximation of ligand strain energy by the contrast, the correlation of eq 2 to the solid-state data terminal donor atom separation, nine tridentate pincer ligands, −1 ff (rmsresidual = 12.0 kJ mol ) was an improvement over our with di erent donor atoms and ring systems, were selected theoretical vacuum-phase training set. As shown in Figure S2, from the CSD (Figure 6 and Table S4). Other 5−6−5 and 6− Δ Δθ θ the relationship between EL(strain) and / L(relax) (rmsresidual −1 Δμ = 14.2 kJ mol ) was also stronger than that when rms was −1 included (rmsresidual = 17.4 kJ mol ). Therefore, we present the relative change in terminal N···N donor separations (Δρ/ ρ L(relax)) as the preferred metric parameter for approximating ligand coordination strain as defined herein, due to both ease fi of measurement and goodness-of- t (with lowest rmsresidual). The calculated coordination strain from the solid-state data are Δρ ρ plotted against / L(relax) in Figure 5 (triangles). Going forward, it emphasized that our analysis suggests an ∼10% error for eq 2. Complexes of D− (green and blue triangles, Figure 5) developed greater ligand strain energy (75−121 kJ mol−1) than those of A (red triangles in Figure 5,25−59 kJ mol−1) with those of C2− intermediate (purple and mustard triangles in Figure 5,57−103 kJ mol−1), consistent with the trends observed in the vacuum-phase training set discussed above. Within the octahedral, homoleptic [M(A)2](ClO4)2 complexes (CCDC codes: M = Fe(II), DANMOU;64 M = Ru(II), BENHUZ;65 and M = Os(II), GOGDOV66), the terpyridyl chelate ligands are the least strained in the ruthenium(II) − complexes (∼25 kJ mol 1), while the osmium(II) and iron(II) ** − Figure 6. Calculated coordination strain energies (B3LYP/6-31G ) structures are strained to similar degrees (41 and 52 kJ mol 1 for other “pincer”-type ligands against the relative change in terminal −1 Δρ ρ for iron, 47 and 58 kJ mol for osmium). Interestingly, the donor atom separations ( / L(relax)) from structures obtained from terpyridyl (A) ligand in Colbran’s heteroleptic octahedral the CSD (shown with metal centers labeled and donor atoms colored 2 26 blue, nitrogen; gray, carbon; orange, phosphorus; yellow, sulfur). The [Ru(A)(D )][PF6] complex (CCDC code: LOGKUP) is − Δ Δρ ρ ∼ 1 trend-line, EL(strain) = 460( / L(relax)), obtained from results for the also more strained ( 60 kJ mol ) than that in the homoleptic − − 2− 2+ 2+ 65 training set of pyridyl pyrrolide pincers A E and their Zn [Ru(A)2] system (CCDC code: BENHUZ). 2− complexes, is shown in black. Predicted strain energies that agree The C ligand derivatives had similar deformations and with the model are illustrated in black, while those showing clear error strains in high valent trigonal bipyramidal lead(IV) complexes are shown in red. Δρ ρ − Δ − −1 ( / L(relax) = 13.1 13.5%, EL(strain) =7276 kJ mol , 67 ’ CCDC code: ALILUC) to those in Milsmann s octahedral − Δρ ρ − 6 6 ring systems with imine (pyridyl or pyrazolyl), carbenes, zirconium(IV) photosensitizer ( / L(relax) = 13.4 14.1%, Δ − −1 68 or cyclometalated phenyl rings agreed well with eq 2 (with EL(strain) =57 66 kJ mol , CCDC code: YAHBIV). The residual error values <15 kJ mol−1), and afforded ligand strain lower coordination number square planar palladium(II) energies typical of the pyridyl-pyrrolide pincers already Δρ ρ Δ − − complex was more strained ( / L(relax) = 20.7%, EL(strain) − 1 71 75 − 69 described (between 50 115 kJ mol ). It was somewhat = 103 kJ mol 1, CCDC code: XOVFIX). Δρ ρ − surprising that the relationship between / relax and ligand Within the set of complexes of derivatives of D ,an strain energy (eq 2) observed for all these systems also applied additional observation is worth noting. The octahedral to Thummel’s fused 6−6−6 system (CCDC code: PUH- ruthenium(II) and square planar palladium(II) complexes of 76 −1 − KOS, with a residual error of 12 kJ mol ). We had unsubstituted D (green triangles, Figure 5) were both more anticipated disagreement due to the extra contribution to the Δρ ρ ̈ deformed and more strained ( / L(relax) = 21.7 and 23.9% strain of the fused system, to which our model would be naive. Δ −1 and EL(strain) = 88 and 120 kJ mol for complexes with 70 63 Internal ligand strain energies for LS (CCDC code: IZAFOG) CCDC codes AYOSUE and TELPUX, respectively) than and HS (CCDC code: IZAFOG01) configurations of the SCO similar complexes of Colbran’s substituted (D2)− derivative · active iron(II) complex, [Fe(pybox)2][PF6]2 MeCN (pybox = Δρ ρ − 75 (blue triangles in Figure 5, / L(relax) = 13.7 16.1%; − pyridine-2,6-bis(oxazo-line), were also predicted with rea- Δ − 1 −1 EL(strain) =7591 kJ mol for complexes with CCDC sonable accuracy from eq 2 (with a residual error of 5 kJ mol 26 codes LOGKUP, LOGLEA and LOGKOJ). Since binding of for the LS complex and 23 kJ mol−1 for the HS complex). these dipyridylpyrrolide ligands to a metal typically brings the However, our simple approach does have limitations, as ρ N-donor atoms of the outer rings closer (i.e., L(coord) < highlighted by the data for complexes of PNP (CCDC code: ρ 77 −1 L(relax)), the introduction of bulky inner ring substituents (the XAMQUZ, with residual error of −26 kJ mol ) and SNS 3-methyl ester and 4-methyl pyrrolide substituents for D2) will (CCDC code: ACAROM,78 with residual error of 30 kJ mol−1) serve to “pre-organize” the ligand into its coordinated pincer ligands that only involve one central ring system. Our ρ − “ ” geometry. Thus, L(relax) = 5.322 Å for unsubstituted D model was not trained to handle the additional steric versus 4.879 Å for Colbran’s(D2)− (see Table S2), and interactions associated with the iso-propyl groups substituted

12317 DOI: 10.1021/acs.inorgchem.8b02038 Inorg. Chem. 2018, 57, 12312−12322 James McPherson – June 2019 58 Chapter 2: Geometric Analysis of Strain in Pincer Ligands

Inorganic Chemistry Article directly onto the donor P atoms in the PNP system.77 For the SNS system (CCDC code: ACAROM),78 we note the following. First, the complex has the thioamides twisted out of the pyridine plane due to Jahn−Teller distortion in the d9 Cu(II) complex; hence, this system resulted in a (small) negative ligand strain energy (−17 kJ mol−1). Second, at such Δρ ρ small relative deformation ( / L(relax) < 5%), the error is larger than the actual strain predicted. II Spin States of Homoleptic [Fe (L)2] Systems. With a method to estimate ligand strain now developed for tridentate pincer ligands, we were interested to explore the internal ligand strains within homoleptic iron(II) complexes of various spin states, including those with spin crossover (SCO) activity. Our motivation in doing so was provided by recent reports from Shatruk and co-workers, who found that the spin states of II ∧ z+ homoleptic [Fe (N N)3] complexes could be empirically predicted from the N···N separations for the unbound bidentate diimine ligands (e.g., bpy = 2,2′-bipyridine).34 Shatruk’s method included a trigonometric approximation of the N···N separation in the unbound syn-ligands (with 0° N− C−C−N torsion) from their lower energy anti-configurations (with 180° N−C−C−N torsion). Optimization of the tridentate ligand systems in this work often resulted in nonplanar geometries (with N−C−C−N torsions anywhere between 0 and 180°), and we reiterate that we therefore use the syn,syn-conformations where the N−C−C−N torsions are constrained to 0° as appropriate approximations for the relaxed ligand geometries. II z+ Nineteen homoleptic [Fe (L)2] complexes were selected for study (Table S5): nine SCO inactive HS (five) or LS (four) fi fi fi II z+ con gurations and ve each of HS and LS con guration for Figure 7. Ligand structures from the homoleptic [Fe (L)2] complexes which displayed SCO activity. All the ligands complexes selected from the CSD for investigation in this work. In selected (shown in Figure 7) bound iron(II) centers through blue at the top are ligands that formed LS iron(II) complexes. In red three N donor atoms within either five- or six-membered rings. at the bottom are ligands that formed HS iron(II) complexes, and in A single DFT geometry optimization calculation (B3LYP/6- the middle in black are ligands which formed SCO-active iron(II) ** ρ complexes. 31G ) was performed for each ligand to determine its L(relax) Δ parameter; thus, the internal ligand strain energies ( EL(strain)) ρ ligand initially in its unbound “relaxed” conformation as its were predicted by eq 2 (with L(coord) easily measured from the SC-XRD structures, Table S5). For all iron(II) complexes, as outer N-donors are pinched closer to form a homoleptic ρ ρ complex with the small LS iron(II) ion. anticipated, L(coord) were shorter than L(relax) (Table S5). The Δ predicted EL(strain) for the ligands within these complexes are |±(3.926 0.036) −ρ | ρ L(relax) −1 presented in Figure 8 against the L(relax) parameters calculated ΔLSE L(strain) = 460 kJ mol for each ligand. ρL(relax) Although the LS iron(II) configurations in the data set (4) exhibited shorter Fe−N distances than their HS analogues (by ρ Similarly, the range of L(coord) parameters observed in the HS 0.21 Å on average, Figure S5) as expected, no correlation was ± ρ states of the SCO-active complexes, which was 4.203 0.057 found between L(relax) (calculated as described above) and Å, leads to eq 5 for the internal ligand strain (Δ E ) that spin state. HS and LS complexes were found for ligands across HS L(strain) ρ − builds upon binding the larger HS iron(II) ion: the range of L(relax) obtained (between 4.5 5.1 Å, Table S5) “ ff” |±(4.203 0.057) −ρ | without a geometric cut-o . For example, the SCO-active L(relax) − 75 ρ Δ E = 460 kJ mol 1 complex, [Fe(pybox)2][BF4]2, exhibits an extreme L(relax) HS L(strain) ρ ρ L(relax) value of 5.02 Å, which is comparable to the L(relax) values for ligands with five-membered central rings in HS iron(II) (5) complexes (4.9−5.1 Å, Table S5, entries 29−32). The absence Equations 4 and 5 are presented as blue and red zones, of a geometric cutoff distinctly contrasts with the aforemen- respectively, in Figure 8, which shows the predicted internal Δ Δρ ρ tioned observation of Shatruk et al. that the spin state of ligand strain energies ( EL(strain) estimated from / L(relax) II ∧ z+ ··· ρ homoleptic [Fe (N N)3] complexes correlate with the N N using eq 2) against L(relax) for the 18 homoleptic iron(II) separations in the bidentate diimine ligands.34 complexes selected from the CSD. (Depending on crystal ρ The L(coord) parameters for the SCO-active LS complexes symmetry, each complex gives one or two points.) The quality were within the range 3.926 ± 0.036 Å. This can be substituted of these fits to the data for the ligands within SCO active Δ into eq 2 to generate eq 4, which predicts that the LSEL(strain) complexes was good, which supports the validity of our ρ for any ligand from L(relax): i.e., eq 4 gives an estimate of the approximation of internal ligand strain by eq 2. This analysis internal strain that would build within a tridentate pincer not only demonstrates the obvious, that LS iron(II) complexes,

12318 DOI: 10.1021/acs.inorgchem.8b02038 Inorg. Chem. 2018, 57, 12312−12322 James McPherson – June 2019 59 Chapter 2: Geometric Analysis of Strain in Pincer Ligands

Inorganic Chemistry Article

ble,45,59,60,79,80 are highly challenging81,82 and completely lose the simplicity of our approach. ■ CONCLUSIONS Tridentate pyridine−pyrrolide chelate ligands, as with all pincer ligands, must deform as they bind metal ions. For hypothetical monoligand zinc(II) complexes, the ligands with central pyridyl rings (A−C) were predicted to deform less than those with central pyrrolide cores (D and E, which also were predicted to exhibit additional, out-of-plane deformation), and within these complexes, the more flanking pyrrolide rings in the ligands, the greater the predicted deformations. The relative deformations are proportional to the strain energy built up within each ligand backbone, and the theoretical predictions were validated by comparison to selected structures from the CSD. Furthermore, the relationships found for this series of tridentate pyridine−pyrrolide ligands also held for a more Figure 8. Plot of the predicted internal ligand strain energies general selection of meridional pincer ligands in which each of Δ Δρ ρ ρ ( EL(strain))from / L(relax) values versus L(relax) values for the three N or C donor atoms were housed within either five homoleptic iron(II) complexes. Quintet (HS) configurations are fi or six-membered rings. Across any series of ligands, those with red, while singlet (LS) con gurations are blue. SCO-active complexes Δρ are represented by triangles that notably lie only within the darker greater separations between the outer donor atoms ( values) will result in greater strain within the ligand, and we shaded bands described by eqs 4 (LS) and 5 (HS), whereas ff complexes reported as SCO-inactive are represented by squares. For expect di erent physical properties (e.g., spin state, or excited 46 the identity of all complexes, and references to them, see the state lifetimes ) and chemical reactivities (e.g., hemilability Supporting Information. leading to catalysis) for the more strained structures. For example, our model predicts that LS iron(II) configurations involve at least ∼25 kJ mol−1 more strain within each with shorter Fe−N bond lengths on average than their HS tridentate ligand scaffoldthanHSconfigurations to be counterparts, will build up greater strain within the tridentate compensated for by the ligand field stabilization energy. ligands (Figure S3c), but also gives an approximation of the In summary, we propose the interatomic separation of the magnitude of this destabilization within each ligand of ∼25 kJ terminal donor atoms (ρ) as a straightforward and easily mol−1. The finding revealed by Figure 8 that this additional measured parameter to describe the ligand strain energy ligand destabilization for the LS state is relatively constant developed within a three-ring pincer ligand upon coordination ρ irrespective of L(relax) is unexpected. with a metal ion. The general approach described in this work The strain predicted for all LS iron(II) complexes was in could easily be adapted to other multidentate chelate ligand Δ ff excellent agreement with LSEL(strain) predicted by eq 4. Across sca olds. ρ − the range of L(relax) (4.5 5.0 Å) the ligands deform only as much as was necessary to chelate the smaller LS iron(II) ion. ■ ASSOCIATED CONTENT The ligand strains predicted for HS configurations were more *S Supporting Information ρ varied. Particularly at higher L(relax) (>4.9 Å) for SCO-inactive The Supporting Information is available free of charge on the HS iron(II) complexes (Table S5, entries 29−32), where the ACS Publications website at DOI: 10.1021/acs.inorg- metal−ligand bonding was insufficient to overcome the chem.8b02038. restrictions imposed by ligand geometry, the Fe−N bonds DFT and SC-XRD analyses; diagrams of all pincer lengthen to minimize the strain within the ligand well below that predicted by eq 5. In contrast, the HS configurations of ligands investigated; Cartesian coordinates from the ρ ∼ output of DFT calculations (PDF) the two SCO-active complexes studied with L(relax) 5.0 Å (Table S5, entries 22, 25, and 26) were in good agreement with eq 5. That these particular iron(II) complexes are SCO- ■ AUTHOR INFORMATION ρ active despite long L(relax) values is indicative of stronger Corresponding Author metal−ligand bonding; indeed, the difference ligand field *E-mail: [email protected]. stabilization energy is at the cusp of compensation for the ORCID internal ligand strain for the LS state. Therefore, their HS James N. McPherson: 0000-0003-0628-7631 states follow eq 5. Stephen B. Colbran: 0000-0002-1119-4950 Finally, although the application of eqs 4 and 5 followed by a comparison with experimental data generates plots such as Notes fi illustrated in Figure 8 that well rationalize spin state and SCO The authors declare no competing nancial interest. behavior, this explanation all comes after the fact, a posteriori,it is not predictive. A prediction of the spin state or SCO ■ ACKNOWLEDGMENTS behavior for a particular complex demands a considerably We thank the Australian Research Council for funding (Grant more sophisticated methodology involving the full and difficult No. DP160104383) and the School of Chemistry, UNSW calculation of the optimum geometry, energy and entropy for Sydney for supporting this research. J.N.M. and T.E.E. are both spin states. Such methodologies, while possi- grateful for the receipt of Australian Postgraduate Awards.

12319 DOI: 10.1021/acs.inorgchem.8b02038 Inorg. Chem. 2018, 57, 12312−12322 James McPherson – June 2019 60 Chapter 2: Geometric Analysis of Strain in Pincer Ligands

Inorganic Chemistry Article ■ REFERENCES (19) England, J.; Scarborough, C. C.; Weyhermüller, T.; Sproules, S.; Wieghardt, K. Electronic Structures of the Electron Transfer Series (1) Sauvage, J. P.; Collin, J. P.; Chambron, J. C.; Guillerez, S.; [M(bpy) ]n[M(tpy) ]n and [Fe(tbpy) ]n (M = Fe, Ru; n = 3+, 2+, Coudret, C.; Balzani, V.; Barigelletti, F.; De Cola, L.; Flamigni, L. 3 2 3 1+, 0, 1−): A Mössbauer Spectroscopic and DFT Study. Eur. J. Inorg. Ruthenium(II) and Osmium(II) Bis(terpyridine) Complexes in Chem. 2012, 2012, 4605−4621. Covalently-Linked Multicomponent Systems: Synthesis, Electro- (20) Bowman, A. C.; England, J.; Sproules, S.; Weyhermüller, T.; chemical Behavior, Absorption Spectra, and Photochemical and Wieghardt, K. Electronic Structures of Homoleptic [Tris(2,2′- Photophysical Properties. Chem. Rev. 1994, 94, 993−1019. bipyridine)M]n Complexes of the Early Transition Metals (M = Sc, (2) Wong, K. M.-C.; Yam, V. W.-W. Luminescence platinum(II) Y, Ti, Zr, Hf, V, Nb, Ta; n = 1+, 0, 1−,2−,3−): An Experimental and terpyridyl complexesFrom fundamental studies to sensory Density Functional Theoretical Study. Inorg. Chem. 2013, 52, 2242− functions. Coord. Chem. Rev. 2007, 251, 2477−2488. 2256. (3) Cummings, S. D. Platinum complexes of terpyridine: Interaction (21) Wang, M.; England, J.; Weyhermüller, T.; Wieghardt, K. and reactivity with biomolecules. Coord. Chem. Rev. 2009, 253, 1495− Molecular and Electronic Structures of the Members of the Electron 1516. n − m (4) Chen, Z.; Concepcion, J. J.; Brennaman, M. K.; Kang, P.; Norris, Transfer Series [Mn(bpy)3] (n = 2+, 1+, 0, 1 ) and [Mn(tpy)2] (m = 4+, 3+, 2+, 1+, 0). An Experimental and Density Functional M. R.; Hoertz, P. G.; Meyer, T. J. Splitting CO2 into CO and O2 by a − single catalyst. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 15606−15611. Theory Study. Inorg. Chem. 2014, 53, 2276 2287. (22) Wang, M.; England, J.; Weyhermüller, T.; Wieghardt, K. (5) Young, K. J.; Brennan, B. J.; Tagore, R.; Brudvig, G. W. “ ” 0 Electronic Structures of Low-Valent Neutral Complexes [NiL2] (S Photosynthetic Water Oxidation: Insights from Manganese Model − Chemistry. Acc. Chem. Res. 2015, 48, 567−574. = 0; L = bpy, phen, tpy) An Experimental and DFT Computational ′ ′ ″ Study. Eur. J. Inorg. Chem. 2015, 2015, 1511−1523. (6)Constable,E.C.2,2:6 ,2 -Terpyridines: From chemical ̈ obscurity to common supramolecular motifs. Chem. Soc. Rev. 2007, (23) England, J.; Bill, E.; Weyhermuller, T.; Neese, F.; Atanasov, M.; − Wieghardt, K. Molecular and Electronic Structures of Homoleptic Six- 36, 246 253. ′ ′ ″ ′ (7) Constable, E. C. Expanded ligandsAn assembly principle for Coordinate Cobalt(I) Complexes of 2,2 :6 ,2 -Terpyridine, 2,2 - supramolecular chemistry. Coord. Chem. Rev. 2008, 252, 842−855. Bipyridine, and 1,10-Phenanthroline. An Experimental and Computa- − (8) Wild, A.; Winter, A.; Schlutter, F.; Schubert, U. S. Advances in tional Study. Inorg. Chem. 2015, 54, 12002 12018. the field of π-conjugated 2,2′:6′,2″-terpyridines. Chem. Soc. Rev. 2011, (24) Flores, J. A.; Andino, J. G.; Tsvetkov, N. P.; Pink, M.; Wolfe, R. 40, 1459−1511. J.; Head, A. R.; Lichtenberger, D. L.; Massa, J.; Caulton, K. G. ′ (9) Sakamoto, R.; Wu, K.-H.; Matsuoka, R.; Maeda, H.; Nishihara, Assessment of the Electronic Structure of 2,2 -Pyridylpyrrolides as − H. π-Conjugated bis(terpyridine)metal complex molecular wires. Ligands. Inorg. Chem. 2011, 50, 8121 8131. Chem. Soc. Rev. 2015, 44, 7698−7714. (25) Komine, N.; Buell, R. W.; Chen, C.-H.; Hui, A. K.; Pink, M.; (10) McPherson, J. N.; Das, B.; Colbran, S. B. Tridentate pyridine− Caulton, K. G. Probing the Steric and Electronic Characteristics of a − pyrrolide chelate ligands: An under-appreciated ligand set with an New Bis-Pyrrolide Pincer Ligand. Inorg. Chem. 2014, 53, 1361 1369. immensely promising coordination chemistry. Coord. Chem. Rev. (26) McSkimming, A.; Diachenko, V.; London, R.; Olrich, K.; Onie, 2018, DOI: 10.1016/j.ccr.2018.01.012. C. J.; Bhadbhade, M. M.; Bucknall, M. P.; Read, R. W.; Colbran, S. B. (11) Hein, F.; Melichar, F. Synthesis of 2,5-di(α-pyridyl)pyrrole and An Easy One-Pot Synthesis of Diverse 2,5-Di(2-pyridyl)pyrroles: A of several complex derivatives of 2,5-di(α-pyridyl)-3,4-dicarbethox- Versatile Entry Point to Metal Complexes of Functionalised, Meridial ypyrrole. Pharmazie 1954, 9, 455−460. and Tridentate 2,5-Di(2-pyridyl)pyrrolato Ligands. Chem. - Eur. J. (12) Ciszek, J. W.; Keane, Z. K.; Cheng, L.; Stewart, M. P.; Yu, L. 2014, 20, 11445−11456. H.; Natelson, D.; Tour, J. M. Neutral Complexes of First Row (27) Comba, P. Coordination compounds in the entatic state. Coord. Transition Metals Bearing Unbound Thiocyanates and Their Chem. Rev. 2000, 200−202, 217−245. Assembly on Metallic Surfaces. J. Am. Chem. Soc. 2006, 128, 3179− (28) Stanek, J.; Hoffmann, A.; Herres-Pawlis, S. Renaissance of the 3189. entatic state principle. Coord. Chem. Rev. 2018, 365, 103−121. (13) Bowman, D. N.; Bondarev, A.; Mukherjee, S.; Jakubikova, E. (29) Comba, P. Strains and stresses in coordination compounds. Tuning the Electronic Structure of Fe(II) Polypyridines via Donor Coord. Chem. Rev. 1999, 182, 343−371. Atom and Ligand Scaffold Modifications: A Computational Study. (30)Kroll,N.;Theilacker,K.;Schoknecht,M.;Baabe,D.; Inorg. Chem. 2015, 54, 8786−8793. Wiedemann, D.; Kaupp, M.; Grohmann, A.; Horner, G. Controlled (14) Goodwin, H. A. Spin Crossover in Iron(II) Tris(diimine) and ligand distortion and its consequences for structure, symmetry, Bis(terimine) Systems. In Spin Crossover in Transition Metal conformation and spin-state preferences of iron(II) complexes. Dalton Compounds I;Gütlich, P., Goodwin, H. A., Eds.; Springer Berlin Trans. 2015, 44, 19232−19247. Heidelberg: Berlin, Heidelberg, 2004; pp 59−90. (31) Comba, P.; Fath, A.; Kuhner, A.; Nuber, B. Preorganization of (15) Frazier, B. A.; Williams, V. A.; Wolczanski, P. T.; Bart, S. C.; tetrathiamacrocyclic ligands: implications from computed and Meyer, K.; Cundari, T. R.; Lobkovsky, E. B. C−C Bond Formation experimentally determined structures. J. Chem. Soc., Dalton Trans. and Related Reactions at the CNC Backbone in (smif)FeX (smif = 1997, 1889−1898. 1,3-Di-(2-pyridyl)-2-azaallyl): Dimerizations, 3 + 2 Cyclization, and (32) Cook, L. J. K.; Kulmaczewski, R.; Mohammed, R.; Dudley, S.; Nucleophilic Attack; Transfer Hydrogenations and Alkyne Trimeriza- Barrett, S. A.; Little, M. A.; Deeth, R. J.; Halcrow, M. A. A Unified tion (X = N(TMS)2 dpma = (Di-(2-pyridyl-methyl)-amide)). Inorg. Treatment of the Relationship Between Ligand Substituents and Spin Chem. 2013, 52, 3295−3312. State in a Family of Iron(II) Complexes. Angew. Chem., Int. Ed. 2016, (16) Billow, B. S.; McDaniel, T. J.; Odom, A. L. Quantifying ligand 55, 4327−4331. effects in high-oxidation-state metal catalysis. Nat. Chem. 2017, 9, 837. (33) Rodríguez-Jimenez,́ S.; Yang, M.; Stewart, I.; Garden, A. L.; (17) Scarborough, C. C.; Wieghardt, K. Electronic Structure of 2,2′- Brooker, S. A Simple Method of Predicting Spin State in Solution. J. Bipyridine Organotransition-Metal Complexes. Establishing the Am. Chem. Soc. 2017, 139, 18392−18396. Ligand Oxidation Level by Density Functional Theoretical Calcu- (34) Phan, H.; Hrudka, J. J.; Igimbayeva, D.; Lawson Daku, L. M.; lations. Inorg. Chem. 2011, 50, 9773−9793. Shatruk, M. A Simple Approach for Predicting the Spin State of (18) Scarborough, C. C.; Sproules, S.; Weyhermüller, T.; DeBeer, S.; Homoleptic Fe(II) Tris-diimine Complexes. J. Am. Chem. Soc. 2017, Wieghardt, K. Electronic and Molecular Structures of the Members of 139, 6437−6447. n the Electron Transfer Series [Cr(tbpy)3] (n = 3+, 2+, 1+, 0): An X- (35) Hrudka, J. J.; Phan, H.; Lengyel, J.; Rogachev, A. Y.; Shatruk, ray Absorption Spectroscopic and Density Functional Theoretical M. Power of Three: Incremental Increase in the Ligand Field Strength Study. Inorg. Chem. 2011, 50, 12446−12462. of N-Alkylated 2,2′-Biimidazoles Leads to Spin Crossover in

12320 DOI: 10.1021/acs.inorgchem.8b02038 Inorg. Chem. 2018, 57, 12312−12322 James McPherson – June 2019 61 Chapter 2: Geometric Analysis of Strain in Pincer Ligands

Inorganic Chemistry Article

Homoleptic Tris-Chelated Fe(II) Complexes. Inorg. Chem. 2018, 57, (54) Coates, R. A.; McNary, C. P.; Boles, G. C.; Berden, G.; 5183−5193. Oomens, J.; Armentrout, P. B. Structural characterization of gas-phase (36) Olguín, J.; Brooker, S. Spin crossover active iron(II) complexes cysteine and cysteine methyl ester complexes with zinc and cadmium of selected pyrazole-pyridine/pyrazine ligands. Coord. Chem. Rev. dications by infrared multiple photon dissociation spectroscopy. Phys. 2011, 255, 203−240. Chem. Chem. Phys. 2015, 17, 25799−808. (37) Gütlich, P.; Gaspar, A. B.; Garcia, Y. Spin state switching in iron (55) Duan, T.; Chang, T.-K.; Chi, Y.; Wang, J.-Y.; Chen, Z.-N.; coordination compounds. Beilstein J. Org. Chem. 2013, 9, 342−391. Hung, W.-Y.; Chen, C.-H.; Lee, G.-H. Blue-emitting heteroleptic Ir (38) Gütlich, P. Spin Crossover − Quo Vadis? Eur. J. Inorg. Chem. (III) phosphors with functional 2, 3′-bipyridine or 2-(pyrimidin-5-yl) 2013, 2013, 581−591. pyridine cyclometalates. Dalton Trans. 2015, 44, 14613−14624. (39) Holland, J. M.; McAllister, J. A.; Kilner, C. A.; Thornton-Pett, (56) Xiao, S.; Liu, Z.; Zhao, J.; Pei, M.; Zhang, G.; He, W. A novel M.; Bridgeman, A. J.; Halcrow, M. A. Stereochemical effects on the fluorescent sensor based on imidazo [1, 2-a] pyridine for Zn2+. RSC 2+ spin-state transition shown by salts of [FeL2] [L = 2,6-di(pyrazol-1- Adv. 2016, 6, 27119−27125. yl)pyridine]. Dalton Trans. 2002, 548−554. (57) Li, W.; Wang, Y.-B.; Yang, L.-Y.; Shan, X.-F.; Cai, X.; (40) Halcrow, M. A. Structure:function relationships in molecular Szeghalmi, A.; Ye, Y.; Ma, J.-S.; Luo, M.-D.; Hu, J.; Kiefer, W. − spin-crossover complexes. Chem. Soc. Rev. 2011, 40, 4119 4142. Spectroscopic and Computational Studies on the Coordination- (41) Halcrow, M. A. The foundation of modern spin-crossover. Driven Self-Assembly Complexes (ZnL) and (NiL) [L= Bis(2,4- − 2 2 Chem. Commun. 2013, 49, 10890 10892. dimethyldipyrrin-3-yl)methane]. J. Phys. Chem. B 2006, 110, 21958− (42) Craig, G. A.; Roubeau, O.; Aromí, G. Spin state switching in 21965. 2,6-bis(pyrazol-3-yl)pyridine (3-bpp) based Fe(II) complexes. Coord. (58) Combariza, M. Y.; Vachet, R. W. Gas-phase reactions of − Chem. Rev. 2014, 269,13 31. divalent Ni complex ions with acetonitrile: chelate ring size, inductive, (43) Kershaw Cook, L. J.; Mohammed, R.; Sherborne, G.; Roberts, and steric effects. J. Am. Soc. Mass Spectrom. 2004, 15, 1128−1135. T. D.; Alvarez, S.; Halcrow, M. A. Spin state behavior of iron(II)/ (59) Wolny, J. A.; Paulsen, H.; McGarvey, J. J.; Diller, R.; dipyrazolylpyridine complexes. New insights from crystallographic − − Schunemann, V.; Toftlund, H. Fe(II) complex with the octadentate and solution measurements. Coord. Chem. Rev. 2015, 289 290,2 12. btpa ligand: a DFT study on a spin-crossover system that reveals two (44) Halcrow, M. The Effect of Ligand Design on Metal Ion Spin −  distinct high-spin states. Phys. Chem. Chem. Phys. 2009, 11, 7562 State Lessons from Spin Crossover Complexes. Crystals 2016, 6, 58. 7575. (45) Capel Berdiell, I.; Kulmaczewski, R.; Halcrow, M. A. Iron(II)  (60) Matouzenko, G. S.; Borshch, S. A.; Schunemann, V.; Wolny, J. Complexes of 2,4-Dipyrazolyl-1,3,5-triazine Derivatives The Influ- A. Ligand strain and conformations in a family of Fe(II) spin ence of Ligand Geometry on Metal Ion Spin State. Inorg. Chem. 2017, crossover hexadentate complexes involving the 2-pyridylmethyl-amino 56, 8817−8828. moiety: DFT modelling. Phys. Chem. Chem. Phys. 2013, 15, 7411− (46) Fatur, S. M.; Shepard, S. G.; Higgins, R. F.; Shores, M. P.; 7419. Damrauer, N. H. A Synthetically Tunable System To Control MLCT (61) Price, S. L. Predicting crystal structures of organic compounds. Excited-State Lifetimes and Spin States in Iron(II) Polypyridines. J. Chem. Soc. Rev. 2014, 43, 2098−2111. Am. Chem. Soc. 2017, 139, 4493−4505. (62) Bakkali, H.; Marie, C.; Ly, A.; Thobie-Gautier, C.; Graton, J.; (47) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Pipelier, M.; Sengmany, S.; Leonel, E.; Nedelec, J.-Y.; Evain, M.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. Dubreuil,D.Functionalized2,5-dipyridinylpyrroles by electrochemical reduction of 3,6-dipyridinylpyridazine precursors. Eur. J. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; − Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Org. Chem. 2008, 2008, 2156 2166. Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; (63) Imler, G. H.; Lu, Z.; Kistler, K. A.; Carroll, P. J.; Wayland, B. B.; Zdilla, M. J. Complexes of 2,5-Bis(α-pyridyl)pyrrolate with Pd(II) Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; π Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, and Pt(II): A Monoanionic Iso- -Electron Ligand Analog of Terpyridine. Inorg. Chem. 2012, 51, 10122−10128. R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, ′ ′ ″ S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, (64) Baker, A.; Goodwin, H. Crystal Structure of Bis(2,2 :6 ,2 - terpyridine)iron(II) Bis(perchlorate) Hydrate. Aust. J. Chem. 1985, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; − Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; 38, 207 214. (65) Kozlowska, M.; Rodziewicz, P.; Brus, D. M.; Breczko, J.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; ′ ′ ″ Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. Brzezinski,K.Bis(2,2:6 ,2 -terpyridine)ruthenium(II) bis- D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. (perchlorate) hemihydrate. Acta Crystallogr., Sect. E: Struct. Rep. − Gaussian 09, revision B.01; Gaussian, Inc.: Wallingford, CT, 2009. Online 2012, 68, m1414 m1415. (48) Lee, C.; Yang, W.; Parr, R. G. Development of the Colle- (66) Craig, D. C.; Scudder, M. L.; McHale, W.-A.; Goodwin, H. A. Salvetti correlation-energy formula into a functional of the electron Structural Studies of Complexes of Tridentate Terimine Systems. ′ ′ ″ density. Phys. Rev. B: Condens. Matter Mater. Phys. 1988, 37, 785−789. Crystal Structure of Bis(2,2 :6 ,2 -terpyridine)ruthenium(II) Per- ′ ′ ′′ (49) Becke, A. D. Density-functional thermochemistry. III. The role chlorate Hydrate, Bis(2,2 :6 ,2 -terpyridine)- osmium(II) Perchlorate of exact exchange. J. Chem. Phys. 1993, 98, 5648−5652. Hemihydrate and Bis((1,10-phenanthrolin-2-yl)(pyridin-2-yl)amine)- (50) Stephens, P.; Devlin, F.; Chabalowski, C.; Frisch, M. J. Ab initio iron(II) Tetrafluoroborate Dihydrate. Aust. J. Chem. 1998, 51, 1131− calculation of vibrational absorption and circular dichroism spectra 1140. using density functional force fields. J. Phys. Chem. 1994, 98, 11623− (67) Jia, W.-L.; Liu, Q.-D.; Wang, R.; Wang, S. Novel 11627. Phosphorescent Cyclometalated Organotin(IV) and Organolead(IV) (51) Petersson, G.; Al-Laham, M. A. A complete basis set model Complexes of 2,6-Bis(2’-indolyl)pyridine and 2,6-Bis[2’-(7- chemistry. II. Open-shell systems and the total energies of the first- azaindolyl)]pyridine. Organometallics 2003, 22, 4070−4078. row atoms. J. Chem. Phys. 1991, 94, 6081−6090. (68) Zhang, Y.; Petersen, J. L.; Milsmann, C. A Luminescent (52) Grimme, S.; Ehrlich, S.; Goerigk, L. Effect of the damping Zirconium(IV) Complex as a Molecular Photosensitizer for Visible function in dispersion corrected density functional theory. J. Comput. Light Photoredox Catalysis. J. Am. Chem. Soc. 2016, 138, 13115− Chem. 2011, 32, 1456−1465. 13118. (53) Boles, G. C.; Owen, C. J.; Berden, G.; Oomens, J.; Armentrout, (69) Liu, Q.; Thorne, L.; Kozin, I.; Song, D.; Seward, C.; D’Iorio, P. B. Experimental and theoretical investigations of infrared multiple M.; Tao, Y.; Wang, S. New red-orange phosphorescent/electro- photon dissociation spectra of glutamic acid complexes with Zn2+ and luminescent cycloplatinated complexes of 2,6-bis(2’-indolyl)pyridine. Cd2+. Phys. Chem. Chem. Phys. 2017, 19, 12394−12406. Dalton Trans. 2002, 3234−3240.

12321 DOI: 10.1021/acs.inorgchem.8b02038 Inorg. Chem. 2018, 57, 12312−12322 James McPherson – June 2019 62 Chapter 2: Geometric Analysis of Strain in Pincer Ligands

Inorganic Chemistry Article

(70) Zhong, Y.-Q.; Xiao, H.-Q.; Yi, X.-Y. Synthesis, structural characterization and catalysis of ruthenium(II) complexes based on 2,5-bis(2’-pyridyl)pyrrole ligand. Dalton Trans. 2016, 45, 18113− 18119. (71) Zhang, X.; Suzuki, S.; Kozaki, M.; Okada, K. NCN Pincer−Pt Complexes Coordinated by (Nitronyl Nitroxide)-2-ide Radical Anion. J. Am. Chem. Soc. 2012, 134, 17866−17868. (72) Peris, E.; Loch, J. A.; Mata, J.; Crabtree, R. H. A Pd complex of a tridentate pincer CNC bis-carbene ligand as a robust homogenous Heck catalyst. Chem. Commun. 2001, 201−202. (73) McGuinness, D. S.; Suttil, J. A.; Gardiner, M. G.; Davies, N. W. Ethylene Oligomerization with Cr−NHC Catalysts: Further Insights into the Extended Metallacycle Mechanism of Chain Growth. Organometallics 2008, 27, 4238−4247. (74) Xiao, X. S.; Kwong, W. L.; Guan, X.; Yang, C.; Lu, W.; Che, C. M. Platinum(II) and Gold(III) Allenylidene Complexes: Phosphor- escence, Self-Assembled Nanostructures and Cytotoxicity. Chem. - Eur. J. 2013, 19, 9457−9462. (75) Zhu, Y.-Y.; Li, H.-Q.; Ding, Z.-Y.; Lu, X.-J.; Zhao, L.; Meng, Y.- 2+ S.; Liu, T.; Gao, S. Spin transitions in a series of [Fe(pybox)2] complexes modulated by ligand structures, counter anions, and solvents. Inorg. Chem. Front. 2016, 3, 1624−1636. (76) Groen, J. H.; de Zwart, A. d.; Vlaar, M. J. M.; Ernsting, J. M.; van Leeuwen, P. W. N. M.; Vrieze, K.; Kooijman, H.; Smeets, W. J. J.; Spek, A. L.; Budzelaar, P. H. M.; Xiang, Q.; Thummel, R. P. Insertion Reactions into Palladium−Carbon Bonds of Complexes Containing Terdentate Nitrogen Ligands; Experimental and Ab initio MO Studies. Eur. J. Inorg. Chem. 1998, 1998, 1129−1143. (77) Tanaka, R.; Yamashita, M.; Nozaki, K. Catalytic Hydrogenation of Carbon Dioxide Using Ir(III)−Pincer Complexes. J. Am. Chem. Soc. 2009, 131, 14168−14169. (78) Kapoor, R.; Kataria, A.; Venugopalan, P.; Kapoor, P.; Corbella, M.; Rodríguez, M.; Romero, I.; Llobet, A. New Tetranuclear Cu(II) Complexes: Synthesis, Structure, and Magnetic Properties. Inorg. Chem. 2004, 43, 6699−6706. (79) Zhang, Y. Predicting critical temperatures of iron(II) spin crossover materials: Density functional theory plus U approach. J. Chem. Phys. 2014, 141, 214703. (80) Stock, P.; Wiedemann, D.; Petzold, H.; Hörner, G. Structural Dynamics of Spin Crossover in Iron(II) Complexes with Extended- Tripod Ligands. Inorganics 2017, 5, 60. (81) Mortensen, S. R.; Kepp, K. P. Spin Propensities of Octahedral Complexes From Density Functional Theory. J. Phys. Chem. A 2015, 119, 4041−4050. (82) Kepp, K. P. Theoretical Study of Spin Crossover in 30 Iron Complexes. Inorg. Chem. 2016, 55, 2717−2727.

12322 DOI: 10.1021/acs.inorgchem.8b02038 Inorg. Chem. 2018, 57, 12312−12322 James McPherson – June 2019 63 Chapter 2: Geometric Analysis of Strain in Pincer Ligands

James McPherson – June 2019 64 Chapter 3: Predictable Substituent Control of Cobalt (III/II) Redox Potential and Spin Crossover in Cobalt Complexes

3 PREDICTABLE SUBSTITUENT CONTROL OF COBALT (III/II) REDOX POTENTIAL AND SPIN CROSSOVER IN COBALT COMPLEXES

3.1 Prior Publication Statement

The work in this chapter has been published by the American Chemical Society in

Inorganic Chemistry in 2019. This article has been reprinted with permission from McPherson,

J. N.; Hogue, R. W.; Akogun, F. S.; Bondi, L.; Luis, E. T.; Price, J. R.; Garden, A. L.; Brooker,

S.; and Colbran, S. B., Predictable Substituent Control of CoIII/II Redox Potential and Spin

Crossover in Bis(dipyridylpyrrolide)cobalt Complexes, Inorg. Chem. 2019, 58 (3), 2218–2228.

3.2 Contributions

JNM assisted with experimental design, coordinated all experiments, prepared and

characterised (except magnetism and SC-XRD) the complexes, analysed all data, investigated

correlations between the properties of the ligands and resulting complexes, wrote the draft

manuscript (except for the sections listed below), prepared the supporting information, and

reworked the manuscript during review.

RWH and FSA collected magnetic susceptibility data, made initial fits to thermodynamic

parameters, and wrote the first draft of this part of the research. L Bondì performed DFT

calculations of the Co(III)/Co(II) couples and wrote the first draft of this section of the study.

ETL collected the SC-XRD data and solved and refined the SC-XRD structures for all

James McPherson – June 2019 65 Chapter 3: Predictable Substituent Control of Cobalt (III/II) Redox Potential and Spin Crossover in Cobalt Complexes

CO Et,CO Et complexes. JRP mentored ETL and co-refined the 100 K structure of [Co(dpp 2 2 )2].

ALG mentored LB, and designed and supervised DFT calculations of Co(III)/Co(II) couples.

SB mentored RWH, FSA and LB, designed and supervised the magnetic studies and DFT calculations of Co(III)/Co(II) couples, and re-worked manuscript. SBC mentored JNM, designed and supervised the project, and refined the manuscript and corrections.

3.3 Foreword

This chapter presents an in-depth study of the influence of the R1 and R2 pyrrolide substituents of 2,5-di(2-pyridyl)-3-R1-4-R2-pyrrolide (dppR1,R2)– ligands on the physical

R1,R2 properties (magnetism and redox) of the cobalt centre of homoleptic [Co(dpp )2] complexes.

This work was in collaboration with Prof. Sally Brooker’s group at the University of Otago,

Dunedin, New Zealand.

The Supporting Information (included in Appendix 2) contains more detailed descriptions of the synthetic protocols (and characterisation data), SC-XRD structures and their refinements

CO Et,CO Et (particularly for the 100 K structure of the [Co(dpp 2 2 )2] complex), fitting of the magnetic data to thermodynamic parameters, correlations of ligand electronics and SCO activities, NMR studies, and the DFT study of CoIII/CoII redox processes.

James McPherson – June 2019 66 Chapter 3: Predictable Substituent Control of Cobalt (III/II) Redox Potential and Spin Crossover in Cobalt Complexes

Article

Cite This: Inorg. Chem. 2019, 58, 2218−2228 pubs.acs.org/IC

Predictable Substituent Control of CoIII/II Redox Potential and Spin Crossover in Bis(dipyridylpyrrolide)cobalt Complexes † ‡ ‡ ‡ † James N. McPherson, Ross W. Hogue, Folaranmi Sunday Akogun, Luca Bondì, Ena T. Luis, Jason R. Price,§ Anna L. Garden,‡ Sally Brooker,*,‡ and Stephen B. Colbran*,†

† School of Chemistry, The University of New South Wales, Kensington, NSW 2052, Australia ‡ Department of Chemistry and MacDiarmid Institute for Advanced Materials and Nanotechnology, University of Otago, P.O. Box 56, Dunedin 9054, New Zealand § ANSTO, Australian Synchrotron, Clayton, VIC Australia

*S Supporting Information

ABSTRACT: Afamilyofﬁve easily prepared tridentate monoanionic 2,5-dipyridyl-3-(R1)-4-(R2)-pyrrolide anions (dppR1,R2)−, varying in the nature of the R1 and R2 substituents 1 2 [R ,R = CN, Ph; CO2Et, CO2Et; CO2Me, 4-Py; CO2Me, Me; Me, Me], has been used to generate the analogous family of neutral II R1,R2 [Co (dpp )2] complexes, two of which are structurally characterized at both 100 and 298 K. Both the oxidation and spin states of these complexes can be switched in response to appropriate external stimuli. All complexes, except II Me,Me [Co (dpp )2], exhibit gradual spin crossover (SCO) in the solid state, and SCO activity is observed for three complexes in CDCl3 solution. The cobalt(II) centers in the low spin (LS) complexes are Jahn−Teller tetragonally compressed along the pyrrolide-Co-pyrrolide axis. The complexes in their high spin (HS) states are more distorted than in the LS states, as is also usually the case for SCO active iron(II) complexes. The reversible CoIII/II redox potentials are predictably tuned by choice of substituents R1 and R2, from −0.95 (Me,Me) to −0.45 (CN,Ph) V vs Fc+/Fc, with a linear correlation observed between III/II − E1/2(Co ) and the Swain Lupton parameters of the pyrrolide substituents.

■ INTRODUCTION 2) systems. Another reason is that Co(II) SCO is less often studied. Cobalt complexes with physical properties that switch in  response to external stimuli continue to be of great interest as Cobalt(II) systems are also redox active oxidation to − centers in molecular devices.1 8 Octahedral complexes of the Co(III) is typically reversible and (usually) results in d7 Co(II) ion are paramagnetic, and they can exist in either diamagnetic complexes. Redox processes at metal centers are Downloaded via UNIV OF NEW SOUTH WALES on March 27, 2019 at 01:31:31 (UTC). high (HS) or low spin (LS) conﬁgurations. When the ligand known to be dependent on ligand electronics, typically 20−32

See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles. field stabilization energy (LFSE) is similar to the pairing measured by substituent Hammett parameters or by energy, spin crossover (SCO, i.e., spin state switching) often Lever’s electronic parameter (LEP).27,28,30,33,34 Thus, the III II ′ III II results. Thermal SCO is triggered by changes in temperature, redox potential of the Co /Co couple, E1/2 (Co /Co ), is 35 so the switching temperature (T1/2, the temperature at which another obvious design parameter. The possible impact of the populations of the LS and HS states are equal) is a key this couple is nicely illustrated in a recent study by New and design parameter. Recent studies of solution phase SCO Fe(II) co-workers, who have developed a series of MRI contrast systems have uncovered relationships which allow T1/2 in agents based on Co(tris(pyridylmethyl)amine) species [Co(II) solution to be predictably tuned through appropriate ligand 9−13 is active, and Co(III) inactive] to probe hypoxia within living selection. Correlations for solid state SCO are less 36,37 ′ III fl cells and cancer spheroids. Modulation of E1/2 (Co / common, due to the often confounding in uence of crystal II ff fi 14−16 Co ) could allow other cobalt-based systems to be applied to packing e ects, but have also been identi ed. These ff methods have only been applied to Fe(II) to date. No cells under oxidative stress (as turn-o sensors). Cobalt(II) predictor of the SCO behavior within Co(II) systems has yet complexes capable of both SCO and reversible redox are rare 23,38 been discovered. One reason is that the physical changes on but are also of considerable interest. spin state switching for Co(II) SCO (ΔS = 1) systems are − generally more subtle and less symmetric,5,6,17 19 making Received: December 11, 2018 prediction a more difficult challenge than in Fe(II) SCO (ΔS = Published: January 23, 2019

© 2019 American Chemical Society 2218 DOI: 10.1021/acs.inorgchem.8b03457 Inorg. Chem. 2019, 58, 2218−2228 James McPherson – June 2019 67 Chapter 3: Predictable Substituent Control of Cobalt (III/II) Redox Potential and Spin Crossover in Cobalt Complexes

Inorganic Chemistry Article

II R1,R2 ﬁ ﬁ − 34 Figure 1. (Left) Five bis(dipyridylpyrrolide)cobalt(II) complexes, [Co (dpp )2], and (right) ve speci c dpp ligands (Py = 2-pyridyl), employed in this study.

Figure 2. ORTEP Views from SC-XRD structures of the [Co(dpp)2] complexes studied in this work (with 50% thermal ellipsoids and H atoms − ⟨λ ⟩ omitted for clarity) and annotated with the average Co N bond distances, quadratic elongation parameter Oh , and XLS for each structure. For CO2Et,CO2Et clarity, only one residue (residue 12) of the [Co(dpp )2] structure at 100 K is shown, and the metric data is the average of all data in the unit cell (weighted by % occupancy). Additional metric data are listed in Tables S2 and S3 in the Supporting Information. †For CO2Et,CO2Et ⟨λ ⟩ ﬁ [Co(dpp )2] at 100 K, the Oh for the residue which was most stable during re nement of the SC-XRD data (residue 12, XLS = 95%) is given, rather than the cell average, due to diﬃculties associated with disordered residues.

The di(2-pyridyl)-3-(R1)-4-(R2)-pyrrolide anion, pyrrolide donor (R1 and R2, respectively) are systematically − (dppR1,R2) , shown in Figure 1,isananionic analogue of varied. Remarkable and predictable tuning of the reversible terpyridine (tpy), with the central pyridyl ring replaced by a CoIII/CoII redox couples is demonstrated, and the different pyrrolide σ- and π-donor which can bear two substituents (R1, magnetic responses across the series of Co(II)complexes to R2).39 These changes result in different geometric and changing temperature, in both solid and solution state, are also 40 electronic properties from those offered by tpy, and also reported. offer an obvious advantage over the widely used tpy ligands: II [Co (dpp)2]complexesareneutral and so avoid the ■ RESULTS AND DISCUSSION complications due to counterions that the analogous tpy fi complexes exhibit.18 In addition, homoleptic [CoII(dpp) ] Syntheses of Complexes. The ve targeted [Co- 2 R1,R2 complexes are easily prepared, with the first (R = H or CO Et) (dpp )2] complexes, Figure 1, were easily prepared by a 2 R1,R2 derivatives obtained in the 1950s by Hein and co-workers.41,42 2:1 reaction of Hdpp ligand:cobalt dichloride hexahydrate, R,R with excess triethylamine, in refluxing methanol under an inert In 2006, Tour and co-workers used [Co(dpp )2](R=Hor SCN) complexes to construct single molecule transistors with atmosphere. The neutral complexes were poorly soluble in Kondo temperatures above 50 K.43 methanol, so the dark red-brown powders were collected in In 2014, Colbran and co-workers invented a one-pot good yields (71−88%). Characterization was by elemental and condensation protocol that provided easy entry to a range of thermogravimetric analysis, high-resolution mass spectrometry, Hdpp ligand precursors with diverse substitution at the C3/C4 paramagnetic 1H NMR spectroscopy, UV−vis−NIR electronic positions of the pyrrole.34 This has greatly facilitated the spectroscopy, variable temperature magnetic susceptibility fi II R1,R2 present investigation of a series of ve [Co (dpp )2] measurements in both the solid and solution phase, and cyclic Me,Me complexes, Figure 1, in which the 3- and 4-substituents of the voltammetry. The complex [Co(dpp )2] was air sensitive

2219 DOI: 10.1021/acs.inorgchem.8b03457 Inorg. Chem. 2019, 58, 2218−2228 James McPherson – June 2019 68 Chapter 3: Predictable Substituent Control of Cobalt (III/II) Redox Potential and Spin Crossover in Cobalt Complexes

Inorganic Chemistry Article

1 II Me,Me Figure 3. Assignment of the H NMR spectrum of [Co (dpp )2] through intermolecular electron transfer (chemical exchange) with the III Me,Me 1 diamagnetic [Co (dpp )2][PF6] complex. The H NMR (500 MHz, acetone-d6) spectrum of the mixture was acquired (left), and then a series of 1H NMR spectra were collected following presaturation at the marked frequencies (B3−B6). During the presaturation at a particular marked frequency, electron transfer interconverts the Co(II) and Co(III) complexes, and so the intensity of the corresponding signal in the diamagnetic Co(III) complex is also reduced. The right-side shows the resulting difference spectra over the region for the pyridyl protons in the Co(III) 1 III Me,Me complex, above the assigned H NMR (500 MHz, acetone-d6) spectrum of [Co (dpp )2][PF6]. The signals attributed to diamagnetic III Me,Me * [Co (dpp )2][PF6], residual protio-solvent and water in the left-side spectrum are denoted as . in solution, which accords with its low CoIII/II redox couple between the outer donor atoms of a pincer ligand when metal- ρ “ ” (see below). bound, L(coord), and when free in its coordination ready ρ Solid State Structures. Crystals suitable for single crystal planar syn;syn-conformation, L(relax)) is an easily measured X-ray diffraction (SC-XRD) structural analyses were obtained parameter that can be used to approximate the strain energy R1,R2 − for [Co(dpp )2] (R1,R2 = Me,Me; CO2Me,Me; and which builds within a pincer ligand, such as dpp , upon ff Δ 50 CO2Me,Me CO2Et,CO2Et) by the slow di usion of pentane into chelation ( EL(strain)). For [Co(dpp )2], the DFT CO2Me,Me CO2Me,Me − ρ chloroform solutions, and for Hdpp (pentane into a optimized geometry of (dpp ) gives L(relax) of 4.899 Å ρ ≈ benzene solution); for detailed descriptions of each SC-XRD (see Table S1, Supporting Information) and L(coord) 4.22 Å structure, see the Supporting Information. Figure 2 presents upon complexation with a cobalt(II) ion, which predicts ≈75 −1 − views and salient metric data for the [Co(dpp)2] structures. kJ mol strain within each bound dpp ligand in the − ⟨ − ⟩ CO2Me,Me For each complex, the average Co N bond length, Co N , [Co(dpp )2] species. Me,Me ⟨ − ⟩ scales with the experimental solid state magnetic susceptibility [Co(dpp )2] at 100 K exhibited Co Npyr = 1.97 Å fi ≈ (discussed later), with a slightly better quality of t found (XLS 0), consistent with increased population of the eg − ⟨ − when only the average axial Co Npyrrolide bond length, Co antibonding orbitals associated with the HS state of the ⟩ Npyr , was examined (see Figure S5, Supporting Information). cobalt(II) center. The coordination octahedra about the HS ≈ ⟨λ ⟩ It was therefore possible to determine the relative fraction of [Co(dpp)2] species were more distorted (with XLS 0: Oh − Σ − ° LS cobalt(II) centers (XLS) within each structure from the = 1.071 1.073; and = 154.0 157.5 ) than for the LS ⟨ − ⟩ ≥ ⟨λ ⟩ − Σ Co Npyr value (see eq S5, Supporting Information). A analogues (with XLS 90%: Oh = 1.044 1.045; and = number of geometric parameters have been used to describe 114.8−114.9°) as is the norm in SCO,6 despite the LS state of the distortion of coordination octahedra around SCO active Co(II) experiencing increased Jahn−Teller distortion18 (see iron(II) systems (e.g., φ, θ, and Σ in Table S2, Supporting above). The reduced steric demands of the resulting more Information);44,45 however, only Σ and Robinson’s quadratic open, so less strained, dpp− ligands bound to the HS Co(II) ⟨λ ⟩ 46 fl ρ elongation parameter, Oh , show any trend with relation to centers are re ected in L(coord) = 4.38 Å and the estimated ρ ≈ ff Δ ≈ XLS of the [Co(dpp)2] complexes studied in this work. L(relax) 5.047 Å, which a ord a reduced EL(strain) 65 kJ CO2Me,Me −1 50 The molecules of [Co(dpp )2] in the structures mol . CO2Et,CO2Et determined at both 298 and 100 K were predominantly LS The structures of [Co(dpp )2] were less symmet- ≈ ≈ (XLS 65% at 298 K shifting to XLS 96% at 100 K). The ric. In the structure determined at 298 K, the asymmetric unit ⟨ − ⟩ ⟨ − ⟩ shorter Co Npyr values (and other structural parameters, see is occupied by two unique molecules: Co1 with Co Npyr = ≈ ⟨ − ⟩ Table S2, Supporting Information) are consistent with those 1.9079(15) Å thus XLS 46%, and HS Co2 with Co Npyr = ≈ − observed for other LS bis(terdentate)cobalt(II) com- 1.959(2) Å thus XLS 0. The dpp ligands around Co1 are 5,7,18,38,47−49 ρ ρ plexes. The octahedral environments around more strained ( L(coord) = 4.216(3) Å and estimated L(relax) = fi Δ ≈ −1 these LS cobalt(II) centers are signi cantly distorted (with 4.969 Å, EL(strain) 70 kJ mol ), and the coordination ⟨λ ⟩ 46 Σ ⟨λ ⟩ Σ ° quadratic elongation parameters, Oh , of 1.045 and = octahedron less distorted (with Oh = 1.053 and = 130.9 ) ° ρ Δ 114.9 at 100 K) due to the combination of the ligand imposed than in the HS Co2 complex ( L(coord) = 4.302(3) Å, EL(strain) − − ≈ −1 ⟨λ ⟩ Σ ° tetragonal contraction along the Npyr Co Npyr axis and the 60 kJ mol , Oh = 1.073, and = 157.5 ). At 100 K, the − ff complementary strong Jahn Teller e ect associated with the symmetry is broken further with 16 independent [Co(dpp)2] 18 singly occupied eg orbital. We have recently shown that the molecules occupying the asymmetric unit, with a quadrupling Δρ ff relative ligand deformation L (the di erence in the distance of the b axis (see Figures S6, S7 and S8 and Tables S3 and S4,

2220 DOI: 10.1021/acs.inorgchem.8b03457 Inorg. Chem. 2019, 58, 2218−2228 James McPherson – June 2019 69 Chapter 3: Predictable Substituent Control of Cobalt (III/II) Redox Potential and Spin Crossover in Cobalt Complexes

Inorganic Chemistry Article

Supporting Information for more details) and an overall ⟨Co− ⟩ ≈ ff Npyr = 1.86 Å and XLS 88% across the 16 di erent (some disordered) residues. NMR Studies. Broad paramagnetically shifted peaks were 1 fi II observed in the H NMR spectra of all ve [Co (dpp)2] complexes in acetone-d6 at 298 K (Table S5). The spectrum of II Me,Me the exemplar complex [Co (dpp )2] was unambiguously assigned using a methodology similar to that reported by Constable et al.51 First, the 1H NMR spectrum of the diamagnetic Co(III) analogue complex III Me,Me [Co (dpp )2][PF6] was assigned (Figures 3 (bottom right) and S16−S20 and Table S5, Supporting Information). II Me,Me Then a mixture of paramagnetic [Co (dpp )2]and III Me,Me diamagnetic [Co (dpp )2][PF6] complexes was prepared in acetone-d6. These two complexes interconvert through ffi 1 Figure 4. NIR region of electronic spectra for the [Co(dpp)2] electron transfer, which is su ciently slow on the H NMR fi time scale at 298 K for both species to be observed complexes studied in CH2Cl2 solution at 298 K, tted to Gaussian simultaneously in the mixture (Figure 3, left). Rather than peaks (see Figure S12, Supporting Information). sensitizing the diamagnetic signals through chemical exchange − − [Co(dppCO2Me,Me) ] (10 400 cm 1; fwhh = 2177 cm 1). In spectroscopy as Constable et al. reported,51 a less subtle 2 these cases there are equilibrium mixtures of HS/LS at 298 K approach was employed. A series of 1H NMR spectra were in CDCl (see below). Thus, the electronic spectra provide an collected following presaturation at the frequency of each 3 II Me,Me indicator of SCO activity in solution. signal assigned to the paramagnetic [Co (dpp )2] species. Magnetic Susceptibility Studies. Solid State Measure- The signal for the corresponding 1H environment in the ments. Solid-state magnetic susceptibility measurements were diamagnetic [CoIII(dppMe,Me) ][PF ] was reduced during the 2 6 performed on each [Co(dpp) ] complex from 50−400 K presaturation (due to chemical exchange), and thus the 2 (Figure 5). As the samples converted to hydrates, C, H and N identity of each presaturated signal was deduced from the difference spectra, as shown on the right in Figure 3. 1 Me,Me With the H NMR spectrum for [Co(dpp )2] assigned, the remaining paramagnetic signals were assigned by analogy through comparisons of the chemical shifts and by line width analyses (see Table S5, and the Supporting Information). The proton environments at the B5 and B4 positions (with similar fwhh: 12−21 Hz) were discriminated by inspection of the 1H−1H COSY NMR spectra (with correlations for 3J(B3,B4) observed in all cases, see the Supporting Information). The signals in the 1H NMR spectra due to pyrrolide substituents were generally assigned by integration. These spectra confirm that the complexes are present as [Co(dpp)2] in solution. Electronic Spectra. The UV−vis spectra of the [Co- (dpp)2] complexes in dichloromethane at 298 K were dominated by π* ← π and charge transfer (CT) bands between 250−500 nm (ε = 20 000−65 000 L mol−1 cm−1); see Figure S10. The vis-NIR regions of the electronic spectra show up to three broad features from 550−1100 nm; Figure S11. The d7 cobalt(II) center has 19 terms in a simple octahedral (Oh) environment, and the tetragonal distortion imposed on the cobalt(II) center by the dpp− ligand further splits these terms. Overlap of these d−d bands plus CT transitions results in the broad features observed throughout the visible−NIR region.43 Figure 5. Temperature dependence of the magnetism of the −1 The lowest energy feature at 10 400−10 850 cm observed [Co(dpp)2] complexes in the solid state. for all [Co(dpp) ] complexes was unobscured by nearby 2 − bands; Figure 4. Absorption maxima at ∼10 800 cm 1, with microanalysis and thermogravimetric assay were used to slightly lower ε values, were observed for the complexes that determine the water content immediately prior to each were HS in solution at all temperatures (see below), namely susceptibility measurement. CN,Ph ν̃ −1 Me,Me ·1 χ − 3 [Co(dpp )2](max: 10 820 cm ; peak full width at half [Co(dpp ) ] / H O with T values of 2.4 2.9 cm K − 2 2 2 M height (fwhh) = 1700 cm 1) and [Co(dppMe,Me) ] (10 780 mol−1 over 50−300 K is consistent with a HS Co(II) species. − − 2 cm 1; fwhh = 1390 cm 1). In contrast, the SCO-active The other complexes exhibit gradual SCO transitions without complexes in solution had slightly more intense and broader any hint of complicating discontinuities such as have been 2+ 52 features that tailed to lower energy. This was observed for observed for some [Co(tpy)2] systems. [Co- CO2Et,CO2Et −1 −1 CO2Et,CO2Et ·1 [Co(dpp )2] (10 420 cm ; fwhh = 3130 cm ), (dpp )2] /3H2O underwent the most complete CO2Me,4‑Py −1 −1 [Co(dpp )2] (10 470 cm ; fwhh = 2580 cm ) and SCO transition observed within the accessible temperature

2221 DOI: 10.1021/acs.inorgchem.8b03457 Inorg. Chem. 2019, 58, 2218−2228 James McPherson – June 2019 70 Chapter 3: Predictable Substituent Control of Cobalt (III/II) Redox Potential and Spin Crossover in Cobalt Complexes

Inorganic Chemistry Article

χ − 3 −1 − range: the MT value of 0.6 0.7 cm K mol between 50 100 K is consistent with LS Co(II), and steadily increased above 100 K, reaching 3.0 cm3 K mol−1 at 400 K, consistent with CO2Me,Me ·1 transitioning to the HS state. [Co(dpp )2] /4H2O and CO2Me,4‑Py · [Co(dpp )2] H2O exhibit similar susceptibility plots to CO2Et,CO2Et ·1 one another: like [Co(dpp )2] /3H2O, they too are χ 3 −1 LS at 50 K, with MT values of 0.7 and 0.8 cm K mol , respectively. However, the transitions for [Co- CO2Me,Me ·1 CO2Me,4‑Py · (dpp )2] /4H2O and [Co(dpp )2] H2Oare CO2Et,CO2Et ·1 more gradual than for [Co(dpp )2] /3H2O and are incomplete at 400 K, the upper limit of our instrument, with χ 3 −1 both reaching a MT value of 2.5 cm K mol at 400 K, consistent with a residual LS population. The magnetic responses on heating the samples were identical to those observed during cooling, with no thermal hysteresis observed (see Figure S13). CO2Et,CO2Et SC-XRD for [Co(dpp )2], which underwent the more abrupt SCO transition, revealed a change in space-group and quadrupling of the crystallographic b-axis as the crystal was Figure 6. Variable temperature magnetic susceptibility in CDCl3 of χ 3 −1 CN,Ph CO2Et,CO2Et cooled from 298 K ( MT = 2.55 cm K mol , C2/c, b = [Co(dpp )2](black),[Co(dpp )2] (orange), [Co- − CO2Me,4‑Py CO2Me,Me χ 3 1 (dpp )2] (purple), and [Co(dpp )2] (green). Solid 14.077(3) Å) to 100 K ( MT = 0.729 cm K mol , Pn, b = fi 56.061(11) Å). Two wave-like perturbations appear along the lines represent the results of least-squares tting of the data to eq 2; b-axis indicative of cooperative effects operating during the the resulting parameters are summarized in Table 1. SCO in the solid state.53,54 At highest frequency, the adjacent K, consistent with the literature values for HS Co(II) Co···Co separations alternate ca. 7.19 and 6.83 Å, whereas at compounds.57 lower frequency the adjacent Co···Co separations parallel to Thermodynamic parameters for the SCO transitions were the a-axis trend closer over each cell from 10.263(3) Å (max.) obtained from fits of the magnetic susceptibility versus to 9.834(3) Å (min.); see Figure S8. The change in the temperature data to eq 1,58 which reduces to eq 2 upon octahedral distortion parameters (ΔΣ = 11.8° and Δφ = 3.3°) substitution of realistic values for the maximum (χ T) (= 3.0 are comparable to those measured recently in similar cobalt(II) − M HS − cm3 K mol 1) and minimum (χ T) (= 0.65 cm3 K mol 1) complexes which displayed cooperative SCO.55 M LS CO2Me,Me values taken from the solid-state data. In contrast, [Co(dpp )2]showedonlyslight contraction of the unit cell (in all dimensions) from 298 K ()χχM TTHS − ()M LS 3 3 χM T = + ()χM T LS (V = 3014.1(10) Å ) to 100 K (V = 2900.2(10) Å ). ΔΔSCOH SCOS CN,Ph · χ 3 −1 1exp+− [Co(dpp )2] 3/4H2O has a MT value of 1.5 cm K mol ()RT R (1) at 50 K, with an approximate 60% LS population at this temperature. It then undergoes a gradual and incomplete SCO 2.35 − χM T = + 0.65 χ 3 1 ΔΔSCOH SCOS to fully HS at 400 K ( MT = 3.3 cm K mol ). Gradual and/or 1exp+− ()RT R (2) incomplete SCO, as observed for these [Co(dpp)2] complexes, is commonplace among SCO-active cobalt(II) complexes.5,7,18 While the fits to eq 1 (with two additional variables) resulted Solution State Susceptibility Measurements. The magnetic in tighter correlations than for those to eq 2, we can provide no ff χ susceptibilities of the four complexes that were SCO active in reasonable explanation for the di erent ( MT)HS values CN,Ph the solid state (see above), namely [Co(dpp )2], [Co- predicted by the fits to eq 1 for the structurally similar CO2Et,CO2Et CO2Me,4‑Py χ 3 −1 (dpp )2], [Co(dpp )2]and[Co- complexes {e.g., using eq 1,(MT)HS is 2.1 cm K mol for CO2Me,Me 1 CO2Me,Me 3 −1 (dpp )2], were also investigated using the Evans H [Co(dpp )2]but2.6cm Kmol for [Co- 56 CO2Et,CO2Et NMR method, in CDCl3 solution between 233 and 323 K (dpp )2]}. We therefore present the thermodynamic (the temperature window accessible in CDCl3 solution). parameters from the fits of these data to eq 2 (see Figure 6)in Figure 6 presents the variable temperature magnetic Table 1. fi Δ Δ susceptibility data that was obtained, as well as the ts to The thermodynamic parameters, SCOH and SCOS, these data sets. Three of the four complexes exhibited SCO in obtained from fitting the data sets, Table 1, are consistent CO2Et,CO2Et chloroform solution. Complexes [Co(dpp )2] and with known examples of cobalt(II) complexes in solution CO2Me,4‑Py Δ − −1 Δ − −1 [Co(dpp )2] exhibited very similar behavior to one ( SCOH =523 kJ mol and SCOS =31 81 J mol another, undergoing gradual and incomplete spin transitions K−1),57,59,60 as well as for the much larger number of iron(II) χ 3 −1 Δ − from mostly HS at 323 K, with MT values of 2.4 cm K mol complexes which have been studied in solution ( SCOH =4 CO2Et,CO2Et 3 −1 −1 Δ − −1 −1 11,57,59−61 for [Co(dpp )2] and 2.5 cm K mol for [Co- 41 kJ mol and SCOS =22 146 J mol K ). The CO2Me,4‑Py CO2Me,Me (dpp )2], to about 50:50 LS:HS, with both having a transition temperature obtained for the [Co(dpp )2] χ 3 −1 MT value of 1.9 cm Kmol at 233 K. Complex complex, T1/2 = 314 K, is the highest of the three SCO-active CO2Me,Me [Co(dpp )2] is already at about 50:50 LS:HS at 318 compounds assayed, and is consistent with having a greater LS χ 3 −1 3 −1 − K( MT = 1.9 cm K mol ), and this drops to 1.1 cm K mol population than the other two between 233 323 K. at 233 K, again consistent with a gradual and incomplete SCO That the magnetic behavior of complexes [Co- CO2Et,CO2Et CO2Me,4‑Py within the measured temperature range. In contrast, [Co- (dpp )2] and [Co(dpp )2] are very similar CN,Ph ff (dpp )2] remained HS across the available temperature in solution (Figure 6) but di erent in the solid state (Figures 5 χ − 3 −1 − fl  range, with MT being 2.92 2.75 cm K mol from 233 323 and S13) illustrates that subtle in uences such as crystal

2222 DOI: 10.1021/acs.inorgchem.8b03457 Inorg. Chem. 2019, 58, 2218−2228 James McPherson – June 2019 71 Chapter 3: Predictable Substituent Control of Cobalt (III/II) Redox Potential and Spin Crossover in Cobalt Complexes

Inorganic Chemistry Article

Δ H −1 Δ S −1 −1 Table 1. Thermodynamic Parameters ( SCO /kJ mol ; SCO /J K mol ) for the SCO-Active [Co(dpp)2] Complexes in T CDCl3 Obtained from Fits of the Observed Data to Eq 2, along with the 1/2/K Values

CO2Et,CO2Et CO2Me,4‑Py CO2Me,Me [Co(dpp )2] [Co(dpp )2] [Co(dpp )2] Δ ± ± ± SCOH 6.2 0.2 8.1 0.1 9.1 0.3 Δ ± ± ± SCOS 28.1 0.6 35.7 0.4 29 1 ± ± ± T1/2 221 9 227 5 314 15

ν −1 III II R1,R2 Figure 7. Left: Cyclic voltammograms ( = 100 mV s ) showing the Co /Co couples of the [Co(dpp )2] complexes in dichloromethane III II with 0.1 M [Bu4N][PF6] at a glassy carbon minidisk electrode and 298 K. Right: Plot of the potential of the Co /Co couple for each [Co(dpp R1,R2 σ σ α − α ’ ﬁ − ﬁ )2] complex against , where = F R; = 2.69, and F and R are the sum of Hansch, Leo, and Taft s modi ed Swain Lupton eld and resonance terms for the R1 and R2 pyrrolide substituents in each ligand.21,63 The bars in the data for each reversible process represent the anodic III II R1,R2 and cathodic peak potentials; the predicted Co /Co couples for four hypothetical [Co(dpp )2] species are shown as blue crosses (+).

R1,R2 vs E ′ + Table 2. Redox Potentials (V) Observed for the [Co(dpp )2] Complexes in CH2Cl2 with 0.1 M [NBu4][PF6], . 1/2 (Fc / Fc) at 298 K, along with Theoretical CoIII/CoII Redox Potentials, E°[LS-CoIII/LS-CoII] and E°[LS-CoIII/HS-CoII], Estimated ΔG III II ΔG Using Free Energies ( LS‑Co /Co ) and Solvation Energies ( solv) Determined by DFT (See the Supporting Information) irrev ′ III II ′ − ′ − ° III II ° III II complex Ep,c E1/2 (Co /Co ) E1/2 (dpp /dpp) E1/2 (dpp /dpp) E [LS-Co /LS-Co ] E [LS-Co /HS-Co ] CN,Ph − −  − − [Co(dpp )2] 2.24 0.445 0.493 0.976 CO2Et,CO2Et − −  − − [Co(dpp )2] 2.25 0.605 0.551 1.440 CO2Me,4‑Py − −  − − [Co(dpp )2] 2.30 0.668 0.613 0.958 CO2Me,Me − − − − [Co(dpp )2] 2.34 0.829 +1.02 +1.12 0.752 1.100 Me,Me  − − − [Co(dpp )2] 0.949 +0.70 +0.84 0.843 1.514 packing and interion interactions in the solid state or solvent Supporting Information). No correlations were found between polarity,62 dielectric constant, or donicity in the solution the δ 15N values of the donor atoms of the Hdpp ligands and  ff state may profoundly a ect the magnetic behavior of a the magnetic properties of their [Co(dpp)2] complexes, Figure particular species. It is not practical to completely describe S32. these environments, either experimentally or theoretically. The We also attempted to correlate the SCO activity of the Co best that can be done is to minimize environmental influences complexes with a linear combination of electronic and our (or at least attempt to make them constant throughout a recently reported steric strain parameter for the ligands. A ff study), and so susceptibility measurements in solution o er the meaningfulpredictiverelationship could not be obtained. best hope for giving results from which correlations can be Electrochemistry. The electrochemical responses of the drawn. fi R1,R2 ve [Co(dpp )2] complexes were studied by cyclic We therefore searched for correlations between the solution voltammetry in dichloromethane. Cyclic voltammograms state SCO activity (T1/2 values) found for the [Co(dpp)2] (CVs) of each complex showed a reversible CoIII/CoII couple systems against those parameters previously reported to 9−13 at a potential exquisitely sensitive to the pyrrolide substituents predict SCO activity of iron(II) systems. No correlation 1 2 III II σ σ σ + (R ,R), Figure 7 and Table 2. The Co /Co couples were was found to Hamnett M, p or p parameters, Figure S14,or to the more general Swain and Lupton’s field (F) and completely free of the confounding electrode- and electrolyte- resonance (R) terms, Figure S15.21,63 Slattery and co-workers dependent irreversibility reported by Fontecave and co- workers in their study of the redox chemistry of [Co(tpy- likewise found little correlation between the inductive 2+ 35 fl ′ R)2] complexes. CVs of the more electron rich complexes in uence of the 4 -substituents and the T1/2 values for a 23 Me,Me CO2Me,Me range of SCO-active homoleptic tpy cobalt(II) complexes. [Co(dpp )2] and [Co(dpp )2] also revealed con- Brooker et al. recently reported a predictive correlation secutive reversible oxidations of the two redox noninnocent − 34 between the switching temperatures (T1/2) and the found or dpp ligands, separated by 100 mV, and centered at ca. 0.77 DFT calculated 15N chemical shift of the N donor atoms and 1.07 V, respectively (Table 2 and Figure S43). All Me,Me within the azine rings of unbound bi- and tridentate ligands complexes, except [Co(dpp )2], also showed an irrever- δ 12 15 δ 15 − ( NA). Hence the N chemical shifts ( N) for each Hdpp sible reduction process at very negative potential, ca. 2.3 V irrev ligand were measured in chloroform-d solution (Table S6, (Ep,c , Table 2 and Figure S44).

2223 DOI: 10.1021/acs.inorgchem.8b03457 Inorg. Chem. 2019, 58, 2218−2228 James McPherson – June 2019 72 Chapter 3: Predictable Substituent Control of Cobalt (III/II) Redox Potential and Spin Crossover in Cobalt Complexes

Inorganic Chemistry Article

The ionization energy of a molecule is related to the HOMO approach has been used previously to identify the MOs from eigenvalue, as is the electron affinity to its LUMO which an electron is removed during oxidation.33,65,69 eigenvalue.64,65 Views of the Kohn−Sham HOMOs and Unsurprisingly, a very strong correlation (R2 = 0.98) was ′ III/II LUMOs of all the [Co(dpp)2] complexes are presented in also found between the observed E1/2 (Co ) and the LUMO fi III + the Supporting Information to distinguish between ligand- and energies calculated for the ve (LS) [Co (dpp)2] complexes − metal-centered MOs.66 68 Inspection of the redox-active (Figure 9), consistent with the reduction event involving this frontier orbitals (the LS Co(II) HOMO and the Co(III) orbital, and confirming the reversibility of the CoIII/CoII redox σ* σ − LUMO) reveals their predominant (N( sp) Co(dx2−y2)) processes as observed experimentally. E ′ III II character (see Figure 8, and the Supporting Information), in Predictable Tuning of 1/2 (Co /Co ). The reversible contrast to the HOMOs for the HS Co(II) species with CoIII/CoII redox potentials of the five complexes varied by 500 − − CN,Ph exclusive dpp ligand character (see Figures S46−S50, mV (Figure 7), ranging from 0.445 V for [Co(dpp )2]to − Me,Me Supporting Information). 0.949 V for [Co(dpp )2]. Lever electronic parameters (LEP) are not available for pyrrolide donors,27,28,30,33,34 and the calculation of the LEP for the (dppCO2Me,Me)− ligand from III/II CO2Me,Me z+ the Ru couples of known [Ru(dpp )Lx] complexes − 34 (Lx: tpy, z = 1; bpy and Cl , z = 0; or bpy and MeCN, z = 1) resulted in a large uncertainty (LEP = −0.01 ± 0.06 V). Therefore, the more general Swain−Lupton field (F) and resonance (R) parameters21,63 were used to characterize the electronic contributions of the two substituents on the five- memberedpyrrolidering,andnotthemorecommon Hammett parameters that are derived from six-membered 20,21 ′ III II fi ring systems. Thus, the E1/2 (Co /Co ) values were tted against the sum of F and R parameters for the two substituents R1 and R2 (then doubled, to account for the two ligands coordinated to the cobalt center) reported by Hansch et al.,21 to eq 3, giving eq 4 (for the line shown in the right-side plot, Figure 7) with an r2 of 0.887. III II EE1/2′=++(Co /Co ) 0 fFrR (3) Figure 8. An exemplar redox-active frontier orbital for the [Co(dpp) ] complexes (the LUMO of [CoIII(dppCO2Et,CO2Et) ]+ in III II 2 2 EF1/2′=−+−(Co /Co ) 1.12 0.487 0.181R(4) σ* σ − 2 2 this case) revealing appreciable (N( sp) Co(dx −y )) character (drawn with isosurface value = 0.03 e/Å3). It is helpful to define a single parameter, σ, to quantify the electronic contribution of the four substituents on the cobalt ′ III/II center, eq 5: The experimental E1/2 (Co ) values are linearly correlated (Figure 9) with the HOMO energy for the LS states of the five σα=−FR,where α = fr / = 2.69 (5) II 2  [Co (dpp)2] complexes (R = 0.98) far more strongly than with the HS states (R2 = 0.76)which again supports the The inductive force term (F)ineq 5 dominates by greater than α preferential oxidation of the LS state in these systems. This a factor of 2 ( = 2.69), while the resonance term (R)is opposite in sign. A similar approach, but for the RuIII/RuII couple of our three previously reported [Ru(dppR1,R2)(tpy)]+ complexes,34 for which Swain−Lupton parameters are available 1 2 21 (R ,R =CO2Me,4-Py; CO2Me,Me; and Me,Me) returned a similar dependencethat is the RuIII/RuII redox potential increased (became less negative) with increasing F, but increasing R lowered the potential (but with F dominating by more than a factor of 3 in these casessee Figure S52). Equation 5 allows eq 4 to be simplified to eq 6: III II E1/2′=−+(Co /Co ) 1.12 0.181σ (6) Larger values of σ represent more electron withdrawing substituents, and eq 6 shows that, as anticipated, these ′ III II correlate to higher E1/2 (Co /Co ) values. This correlation (Figure 7, right-side plot) allows us to ′ III II predict E1/2 (Co /Co ) values for some other accessible homoleptic [Co(dpp)2] complexes, and in so doing also ′ III II demonstrated that a range of E1/2 (Co /Co ) values from about −1.7 to 0 V should be accessible for this ligand class. II H,H The analysis predicts that [Co (dpp )2] should be even II Me,Me more easily oxidized than [Co (dpp )2], which we III/II II SCN,SCN Figure 9. Frontier orbital energies (DFT) versus the observed Co observed to slowly oxidize in air, while [Co (dpp )2] ′ III/II fi II CN,Ph couples (E1/2 (Co )) for the ve [Co(dpp)2] complexes. should be more stable than [Co (dpp )2], which is air

2224 DOI: 10.1021/acs.inorgchem.8b03457 Inorg. Chem. 2019, 58, 2218−2228 James McPherson – June 2019 73 Chapter 3: Predictable Substituent Control of Cobalt (III/II) Redox Potential and Spin Crossover in Cobalt Complexes

Inorganic Chemistry Article

43 stable. Tour et al. have already studied these complexes, CoCl2.6H2O (approximately 0.25 mmol) was added, and the mixture albeit without a standard reference couple specified, but our was stirred at reflux for 2 h. After cooling to room temperature, the R1,R2 fi predictions and their findings about sensitivity to molecular crude product, [Co(dpp )2], was collected by ltration, and oxygen are entirely consistent. Future studies will include washed with cold MeOH three times. Yields and the characterization data for each complex are presented in the Supporting Information. preparing fresh samples of those literature complexes, along Magnetic Susceptibility Measurements. Variable temperature with the new complexes shown predictively in Figure 7,aswe magnetic susceptibility data were collected at the University of Otago look to prove the robustness of this correlation. on a Quantum Design Versalab, a cryogen-free PPMS susceptometer, equipped with a vibrating sample mount (VSM) under an applied fi CN,Ph ·3 ■ CONCLUSIONS eld of 1000 Oersteds. Data on [Co(dpp )2] /4H2O, [Co- (dppCO2Et,CO2Et) ]·1/ H O and [Co(dppCO2Me,Me) ]·1/ H Owere Families of (dppR1,R2)− ligands, with systematic variation of the 2 3 2 2 4 2 1 2 collected in settle mode at 5 K intervals, while [Co- R ,R substituents present on the pyrrolide ring, are easily (dppCO2Me,4‑Py) ]·H O were collected in sweep mode at 5 K min−1, II 2 2 prepared. The resulting neutral [Co (dpp)2] complexes of the in the temperature range 400−50 K. Data on [Co- − Me,Me ·1 − tridentate, anionic dpp ligands comprise severely tetragonally (dpp )2] /2H2O were collected in the temperature range 300 compressed octahedral cobalt(II) centers, due to a combina- 50 K in settle mode at 5 K intervals. The susceptibility−temperature tion of Jahn−Teller and strained tridentate ligand bite effects. data for cooling and heating overlaid. Samples were mounted in a This occurs for both the HS and LS states of cobalt(II), but polyethylene capsule for which a background diamagnetic correction − was applied. Data were also corrected for the diamagnetism of the with the HS states exhibiting a longer average Co N length χ dia − × × sample according to the approximation that M = 0.5 MW and somewhat more distortion than the LS state. The strained 10−6 cm3 mol−1 (where MW = molecular weight of the complex).58 bite is a consequence of the wider ligand bend angle, due to CN,Ph Solution magnetic susceptibility data on [Co(dpp )2] the five-membered pyrrolide core, over that provided by the −3 CO2Et,CO2Et −3 (0.006243 g cm ), [Co(dpp )2] (0.006645 g cm ), CO2Me,4‑Py −3 CO2Me,Me pyridine core of the related tpy ligands. We anticipate the [Co(dpp )2] (0.006585 g cm ), and [Co(dpp )2] − − −3 resulting longer Co Ndistal (Co Npy) bonds will result in (0.006742 g cm ) were measured in CDCl3 in the temperature range different reactivities (e.g., hemilability) than the more common 233−323 K, by 1H NMR spectroscopy on a Varian 500 MHz 2+ ’ 56 [Co(tpy)2] systems, thus potentially leading to interesting VNMRS spectrometer using the Evans method. Pure CDCl3 was placed in a capillary NMR tube insert and the paramagnetic solution applications in catalysis. ± Importantly these complexes also exhibit useful properties was placed in the outer tube. Temperatures are accurate to 1 K. The the oxidation and/or spin states of the cobalt centers can be mass susceptibility of the complexes was calculated from the shift of the residual CHCl3 peak in the paramagnetic solution compared to reversibly switched in response to external stimuli. Variable that in pure CDCl , Δf in Hz, using eq 7: temperature SC-XRD studies reveal a phase transition in the 3 CO2Et,CO2Et 3Δf dd− structure of the [Co(dpp )2] complex with the most χ = ++χχ0 s abrupt spin crossover in the solid state, with two wave-like g 4πmf 00m (7) perturbations along the b crystallographic axis indicative of −3 fi where m is the concentration of the paramagnetic solution in g cm cooperativity. Three complexes show signi cant SCO activity and this was corrected for the temperature dependence of the density around room temperature in solution. of chloroform, f is the spectrometer frequency in Hz, and d0 and ds are Predictable tuning of the redox potential, but not of solution χ the densities of the solvent and solution, respectively. 0 is the mass ′ III 3 −1 SCO, is demonstrated. The former correlation, of E1/2 (Co / susceptibility of the solvent in cm g . This equation was simplified II Co ) values with the electronic nature of the pyrrolide by taking an approximation that ds = d0 + m, which is reasonable for substituents, σ (determined by use of the modified Swain− dilute solutions.70,71 This approximation leads to the second and third Lupton parameters for R1,R2), allows us to predict that the terms canceling out, giving eq 8: ′ III II E1/2 (Co /Co ) can be tuned, using accessible substituents, 3Δf by up to ca. 1.7 V. χ = g 4πmf (8) The DFT results suggest that the CoIII/CoII couples proceed − χ through a LS Co(II) LS Co(III) electron transfer pathway, Multiplying g by the molecular weight (MW) gave the molar χ χ thus minimizing the reorganization energy for the electron susceptibility M. These M values were then corrected for the χ dia transfer. The energies of the redox active σ*(N(σ )− diamagnetic contributions of each sample according to M (sample) sp − × × −6 3 −1 2 2 7 = 0.5 MW 10 cm mol (where MW = molecular weight of Co(dx −y ) orbital (the HOMO in the LS d -Co(II) state and 58 χ 6 the complex). Note: The MT values, for two separate VT runs on the LUMO in the LS d -Co(III) state) also correlate with CN,Ph ′ III II the [Co(dpp )2] complex, were averaged at each temperature; the experimental E1/2 (Co /Co ) values, thus a quick DFT χ 3 −1 MT value was 2.8 cm K mol at 298 K in both runs. calculation allows prediction of the oxidation potential for a Solution magnetic susceptibility data were fitted to eqs 1 and 2 by R1,R2 − Co(dpp )2 complex in those cases where Swain Lupton least-squares fitting performed using OriginPro version 9.1.0 from parameters for the substituents R1, R2 are unavailable. OriginLab Corporation. fi We expect these and similar [Co(dpp)2] complexes will nd Electrochemistry. Solutions of compounds were of ca.1mM use as redox or magnetic centers in sophisticated molecular concentration in dichloromethane with tetrabutylammonium hexa- assemblies with targeted function. fluorophosphate (0.1 M) as the supporting electrolyte and were sparged with high purity nitrogen. CVs were recorded at 100 mV s−1 ■ EXPERIMENTAL SECTION under a nitrogen atmosphere using a GAMRY Reference 600 potentiostat using a standard three-electrode system comprising General Synthetic Procedure for [Co(dpp)2] Complexes. glassy carbon minidisk working electrode (1.0 mm diameter, freshly Methanol was dried over activated 3 Å molecular sieves for 24 h prior polished), a platinum wire counter electrode and an eDAQ leakless to use. Reactions were performed under a dry, oxygen free Ag/AgCl reference electrode. A background CV was always obtained atmosphere of dinitrogen using standard Schlenk techniques. Ligand between anodic and cathodic solvent discharge limits prior to adding precursors Hdpp were prepared according to the literature method.34 sample. Ferrocene (Fc) was added as an internal standard at the end R1,R2 Hdpp (approximately 0.5 mmol) was combined with excess NEt3 of each set of CV experiments and all potentials are quoted relative to fl + Δ ≈ (0.3 mL, 2 mmol) in MeOH (10 mL) and stirred at re ux for 15 min. the Fc /Fc couple {E1/2 =0V, Ep 80 mV in dichloromethane}.

2225 DOI: 10.1021/acs.inorgchem.8b03457 Inorg. Chem. 2019, 58, 2218−2228 James McPherson – June 2019 74 Chapter 3: Predictable Substituent Control of Cobalt (III/II) Redox Potential and Spin Crossover in Cobalt Complexes

Inorganic Chemistry Article DFT Calculations. All dpp− ligand DFT calculations were ■ AUTHOR INFORMATION performed with the Gaussian 2009 software package.72 Vacuum phase dpp− ligand geometry optimizations were performed using the Corresponding Authors − B3LYP functional73 75 with the 6-31G** basis set.76 Dispersion *(S.B.C.) E-mail: [email protected]. corrections were included by adding the D3 version of Grimme’s *(S.B.) E-mail: [email protected]. empirical dispersion correction with Becke−Johnson damping.77 The ORCID inter-ring N−C−C−N torsion angles were constrained to 0.0° by the modredundant command. James N. McPherson: 0000-0003-0628-7631 z+ All [Co(dpp)2] (z = 0, 1) complex DFT calculations were Folaranmi Sunday Akogun: 0000-0001-7613-5911 78 performed using the ORCA 4.0.1 code, using the B3LYP functional Ena T. Luis: 0000-0002-3191-8944 ’ coupled with the D3 version of Grimme s functional with BJ damping Anna L. Garden: 0000-0002-8069-3243 correction to account for dispersion forces.77,79 The experimental SC- II Me,Me II CO2,Me,Me Sally Brooker: 0000-0002-5878-8238 XRD structures of [Co (dpp )2], [Co (dpp )2]and II CO2Et,CO2Et Stephen B. Colbran: 0000-0002-1119-4950 [Co (dpp )2] provided the initial Cartesian coordinates for II CO2Me,Me the respective complexes, and the structure of [Co (dpp )2] Notes was modified as required to generate input coordinates for the CO2Me,4‑Py CN,Ph The authors declare no competing financial interest. remaining [Co(dpp )2] and [Co(dpp )2] complexes. The [Co(dpp)2] complexes were preoptimized with a def2-SVP basis set,80 to calculate the Hessian matrix faster and confirm the obtained ■ ACKNOWLEDGMENTS structure as a minimum by the absence of imaginary frequencies, before a final optimization was performed using the def2-TZVPP basis We thank the Australian Government (PhD scholarship to set.80 Both optimizations were supported by a def2/J auxiliary basis J.N.M. and E.T.L.), the University of Otago (Ph.D. scholarship set81 to take advantage of the RI-J approximation.78 Then a single to R.W.H, F.S.A. and L.B.), the Marsden Fund (RSNZ), point calculation on each of the final optimized structures was Australian Research Council (DP160104383) and UNSW performed to calculate the solvation energies in CH2Cl2 with a Sydney for supporting this research. Access to the Australian 82 conductor-like polarizable continuum solvation model CPCM. Synchrotron was made possible by the Collaborative Access X-ray Crystallography. Pale yellow hexagonal plates of the CO2Me,Me Program (MXCAP12368/13234). We gratefully acknowledge Hdpp ligand suitable for SC-XRD were grown by slow the Mark Wainwright Analytical Centre (MWAC) at UNSW diffusion of pentane vapor into a solution of the compound in benzene and the X-ray diffraction measurements were carried out on a Sydney (NMR and BMSF), the Campbell Microanalytical Bruker kappa-II CCD diffractometer at 150 K using IμS Incoatec Laboratory at the University of Otago, the NeSI high Microfocus Source with Mo−Kα radiation (λ = 0.710723 Å). performance computing (HPC) facilities (NZ), and Prof. R1,R2 1 2 − Crystals of the [Co(dpp )2] complexes (R ,R :CO2Et,CO2Et Federico Totti (Florence) for providing access to the LaMM − − brown blocks; CO2Me,Me red-brown plates; and Me,Me red- group HPC facilities. brown plates) suitable for SC-XRD studies were grown by the slow diffusion of pentane vapor into a solution of the compound in 1 2 ff ■ REFERENCES chloroform (benzene for R ,R =CO2Me, Me). Di raction measurements were carried out using Si⟨111⟩ monochromated (1) Letard,́ J.-F.; Guionneau, P.; Goux-Capes, L. Towards Spin synchrotron X-ray radiation (λ = 0.71073 Å) on the MX1 Beamline Crossover Applications. Spin Crossover in Transition Metal Compounds 83 CO2Et,CO2Et − at the Australian Synchrotron. For [Co(dpp )2]and III 2004, 235, 221 249. CO2Me,Me [Co(dpp )2], the single crystal was mounted and a full data (2) Sato, O. Dynamic molecular crystals with switchable physical set was first collected at 298 K, and then the sample was cooled and a properties. Nat. Chem. 2016, 8, 644. second set of data for each was collected at 100 K. (3) Yao, Z.-S.; et al. Anisotropic Change in the Magnetic Crystallographic data have been deposited as entries CCDC Susceptibility of a Dynamic Single Crystal of a Cobalt(II) Complex. CO2Me,Me Me,Me · − 1873690 (Hdpp ), 1873691 {[Co(dpp )2] 2(C5H12)}, Angew. Chem., Int. Ed. 2017, 56, 717 721. CO2Me,Me 1873692 {[Co(dpp )2]at298K},1873693 {[Co- (4) Senthil Kumar, K.; Ruben, M. Emerging trends in spin crossover CO2Me,Me CO2Et,CO2Et (dpp )2] at 100 K}, 1873694 {[Co(dpp )2] at 298 (SCO) based functional materials and devices. Coord. Chem. Rev. CO2Et,CO2Et K}, and 1873695 {[Co(dpp )2] at 100 K}. 2017, 346, 176−205. (5) Krivokapic, I.; Zerara, M.; Daku, M. L.; Vargas, A.; Enachescu, ■ ASSOCIATED CONTENT C.; Ambrus, C.; Tregenna-Piggott, P.; Amstutz, N.; Krausz, E.; Hauser, A. Spin-crossover in cobalt(II) imine complexes. Coord. *S Supporting Information Chem. Rev. 2007, 251, 364−378. The Supporting Information is available free of charge on the (6) Miller, R. G.; Narayanaswamy, S.; Tallon, J. L.; Brooker, S. Spin ACS Publications website at DOI: 10.1021/acs.inorg- crossover with thermal hysteresis in cobalt(II) complexes and the chem.8b03457. importance of scan rate. New J. Chem. 2014, 38, 1932−1941. (7)Drath,O.;Boskovic,C.Switchablecobaltcoordination Experimental details and analytical data for complexes; polymers: Spin crossover and valence tautomerism. Coord. Chem. computational studies; single-crystal X-ray structures; Rev. 2018, 375, 256−266. magnetic susceptibility studies; NMR studies; additional (8) Hayami, S.; Komatsu, Y.; Shimizu, T.; Kamihata, H.; Lee, Y. H. electrochemistry; correlations and data fits; coordinates Spin-crossover in cobalt(II) compounds containing terpyridine and its − of DFT optimized structures (PDF) derivatives. Coord. Chem. Rev. 2011, 255, 1981 1990. (9) Prat, I.; Company, A.; Corona, T.; Parella, T.; Ribas, X.; Costas, Accession Codes M. Assessing the Impact of Electronic and Steric Tuning of the CCDC 1873690−1873695 contain the supplementary crys- Ligand in the Spin State and Catalytic Oxidation Ability of the FeII(Pytacn) Family of Complexes. Inorg. Chem. 2013, 52, 9229− tallographic data for this paper. These data can be obtained 9244. free of charge via www.ccdc.cam.ac.uk/data_request/cif,orby (10) Lin, H.-J.; Siretanu, D.; Dickie, D. A.; Subedi, D.; Scepaniak, J. emailing [email protected], or by contacting The J.; Mitcov, D.; Clerac,́ R.; Smith, J. M. Steric and Electronic Control Cambridge Crystallographic Data Centre, 12 Union Road, of the Spin State in Three-Fold Symmetric, Four-Coordinate Iron(II) Cambridge CB2 1EZ, UK; fax: +44 1223 336033. Complexes. J. Am. Chem. Soc. 2014, 136, 13326−13332.

2226 DOI: 10.1021/acs.inorgchem.8b03457 Inorg. Chem. 2019, 58, 2218−2228 James McPherson – June 2019 75 Chapter 3: Predictable Substituent Control of Cobalt (III/II) Redox Potential and Spin Crossover in Cobalt Complexes

Inorganic Chemistry Article

(11) Kershaw Cook, L. J.; Kulmaczewski, R.; Mohammed, R.; (30) Vanicek, S.; Kopacka, H.; Wurst, K.; Müller, T.; Hassenrück, Dudley, S.; Barrett, S. A.; Little, M. A.; Deeth, R. J.; Halcrow, M. A. A C.; Winter, R. F.; Bildstein, B. Monofunctionalized Cobaltocenium Unified Treatment of the Relationship Between Ligand Substituents Compounds by Dediazoniation Reactions of Cobaltoceniumdiazo- and Spin State in a Family of Iron(II) Complexes. Angew. Chem., Int. nium Bis(hexafluorophosphate). Organometallics 2016, 35, 2101− Ed. 2016, 55, 4327−4331. 2109. (12) Rodriguez-Jimenez, S.; Yang, M.; Stewart, I.; Garden, A. L.; (31) Therrien, J. A.; Wolf, M. O. The Influence of para Substituents Brooker, S. A Simple Method of Predicting Spin State in Solution. J. in Bis(N-Heterocyclic Carbene) Palladium Pincer Complexes for Am. Chem. Soc. 2017, 139, 18392−18396. Electrocatalytic CO2 Reduction. Inorg. Chem. 2017, 56, 1161−1172. (13) Kimura, A.; Ishida, T. Spin-Crossover Temperature Predictable (32) Muck, C. S.; Linseis, M.; Welte, H.; Weickert, S.; Drescher, M.; from DFT Calculation for Iron(II) Complexes with 4-Substituted Winter, R. F. Electrochemical and Spectroscopic Studies on σ-Phenyl Pybox and Related Heteroaromatic Ligands. ACS Omega 2018, 3, Ruthenium Complexes Ru(CO)Cl(C6H4R-4)(PiPr3)2. Organome- 6737−6747. tallics 2018, 37, 2111−2122. (14) Nakano, K.; Suemura, N.; Yoneda, K.; Kawata, S.; Kaizaki, S. (33) Lever, A. B. P. Electrochemical parametrization of metal Substituent effect of the coordinated pyridine in a series of pyrazolato complex redox potentials, using the ruthenium(III)/ruthenium(II) bridged dinuclear diiron(II) complexes on the spin-crossover couple to generate a ligand electrochemical series. Inorg. Chem. 1990, behavior. Dalton Trans. 2005, 740−743. 29, 1271−1285. (15) Park, J. G.; Jeon, I.-R.; Harris, T. D. Electronic Effects of Ligand (34) McSkimming, A.; Diachenko, V.; London, R.; Olrich, K.; Onie, Substitution on Spin Crossover in a Series of Diiminoquinonoid- C. J.; Bhadbhade, M. M.; Bucknall, M. P.; Read, R. W.; Colbran, S. B. Bridged FeII2 Complexes. Inorg. Chem. 2015, 54, 359−369. An Easy One-Pot Synthesis of Diverse 2,5-Di(2-pyridyl)pyrroles: A (16) Phan, H.; Hrudka, J. J.; Igimbayeva, D.; Lawson Daku, L. M.; Versatile Entry Point to Metal Complexes of Functionalised, Meridial Shatruk, M. A Simple Approach for Predicting the Spin State of and Tridentate 2,5-Di(2-pyridyl)pyrrolato Ligands. Chem. - Eur. J. Homoleptic Fe(II) Tris-diimine Complexes. J. Am. Chem. Soc. 2017, 2014, 20, 11445−11456. 139, 6437−6447. (35) Aroua, S.; Todorova, T. K.; Hommes, P.; Chamoreau, L.-M.; (17) Gütlich, P.; Goodwin, H. A. Spin CrossoverAn Overall Reissig, H.-U.; Mougel, V.; Fontecave, M. Synthesis, Characterization, Perspective. Top. Curr. Chem. 2004, 233,1−47. and DFT Analysis of Bis-Terpyridyl-Based Molecular Cobalt (18) Goodwin, H. A. Spin Crossover in Cobalt(II) Systems. Top. Complexes. Inorg. Chem. 2017, 56, 5930−5940. Curr. Chem. 2004, 234,23−47. (36) O’Neill, E. S.; Kolanowski, J. L.; Yin, G. H.; Broadhouse, K. M.; (19) Brooker, S. Spin crossover with thermal hysteresis: practicalities Grieve, S. M.; Renfrew, A. K.; Bonnitcha, P. D.; New, E. J. Reversible and lessons learnt. Chem. Soc. Rev. 2015, 44, 2880−2892. and front magnetogenic cobalt complexes. RSC Adv. 2016, 6, 30021−30027. cover feature. (37) O’Neill, E. S.; Kaur, A.; Bishop, D. P.; Shishmarev, D.; Kuchel, (20) Hammett, L. P. The Effect of Structure upon the Reactions of P. W.; Grieve, S. M.; Figtree, G. A.; Renfrew, A. K.; Bonnitcha, P. D.; Organic Compounds. Benzene Derivatives. J. Am. Chem. Soc. 1937, New, E. J. Hypoxia-Responsive Cobalt Complexes in Tumor 59,96−103. Spheroids: Laser Ablation Inductively Coupled Plasma Mass (21) Hansch, C.; Leo, A.; Taft, R. W. A survey of Hammett Spectrometry and Magnetic Resonance Imaging Studies. Inorg. substituent constants and resonance and field parameters. Chem. Rev. Chem. 2017, 56, 9860−9868. 1991, 91, 165−195. (38) Cowan, M. G.; Olguín, J.; Narayanaswamy, S.; Tallon, J. L.; (22) Maestri, M.; Armaroli, N.; Balzani, V.; Constable, E. C.; Brooker, S. Reversible switching of a cobalt complex by thermal, Thompson, A. M. W. C. Complexes of the Ruthenium(II)-2,2’:6’,2’’- pressure and electrochemical stimuli: abrupt, complete, hysteretic spin terpyridine Family. Effect of Electron-Accepting and -Donating crossover. J. Am. Chem. Soc. 2012, 134, 2892−2894. and front cover Substituents on the Photophysical and Electrochemical Properties. feature. Inorg. Chem. 1995, 34, 2759−2767. (39) McPherson, J. N.; Das, B.; Colbran, S. B. Tridentate pyridine− (23) Chambers, J.; Eaves, B.; Parker, D.; Claxton, R.; Ray, P. S.; pyrrolide chelate ligands: An under-appreciated ligand set with an Slattery, S. J. Inductive influence of 4′-terpyridyl substituents on redox immensely promising coordination chemistry. Coord. Chem. Rev. and spin state properties of iron(II) and cobalt(II) bis-terpyridyl 2018, 375, 285−332. complexes. Inorg. Chim. Acta 2006, 359, 2400−2406. (40) Bowman, D. N.; Bondarev, A.; Mukherjee, S.; Jakubikova, E. (24) Sjödin, M.; Gatjens,̈ J.; Tabares, L. C.; Thuery,́ P.; Pecoraro, V. Tuning the Electronic Structure of Fe(II) Polypyridines via Donor L.; Un, S. Tuning the Redox Properties of Manganese(II) and Its Atom and Ligand Scaffold Modifications: A Computational Study. Implications to the Electrochemistry of Manganese and Iron Inorg. Chem. 2015, 54, 8786−8793. Superoxide Dismutases. Inorg. Chem. 2008, 47, 2897−2908. (41) Hein, F.; Melichar, F. Synthesis of 2,5-di(α-pyridyl)pyrrole and (25) Ostermeier, M.; Berlin, M.-A.; Meudtner, R. M.; Demeshko, S.; of several complex derivatives of 2,5-di(α-pyridyl)-3,4-dicarbethox- Meyer, F.; Limberg, C.; Hecht, S. Complexes of Click-Derived ypyrrole. Pharmazie 1954, 9, 455−460. Bistriazolylpyridines: Remarkable Electronic Influence of Remote (42) Hein, F.; Beierlein, U. Preparation and chemistry of the Substituents on Thermodynamic Stability as well as Electronic and complexes of 2,5-di-(α-pyridyl)pyrrole. Pharm. Zentralhalle Dtschl. Magnetic Properties. Chem. - Eur. J. 2010, 16, 10202−10213. 1957, 96, 401−421. (26) Hewage, J. S.; Wanniarachchi, S.; Morin, T. J.; Liddle, B. J.; (43) Ciszek, J. W.; Keane, Z. K.; Cheng, L.; Stewart, M. P.; Yu, L. Banaszynski, M.; Lindeman, S. V.; Bennett, B.; Gardinier, J. R. H.; Natelson, D.; Tour, J. M. Neutral Complexes of First Row Homoleptic Nickel(II) Complexes of Redox-Tunable Pincer-type Transition Metals Bearing Unbound Thiocyanates and Their Ligands. Inorg. Chem. 2014, 53, 10070−10084. Assembly on Metallic Surfaces. J. Am. Chem. Soc. 2006, 128, 3179− (27) Conradie, M. M.; Conradie, J. Electrochemical behaviour of 3189. Tris(β-diketonato)iron(III) complexes: A DFT and experimental (44) Holland, J. M.; McAllister, J. A.; Kilner, C. A.; Thornton-Pett, study. Electrochim. Acta 2015, 152, 512−519. M.; Bridgeman, A. J.; Halcrow, M. A. Stereochemical effects on the 2+ (28) Freitag, R.; Conradie, J. Electrochemical and Computational spin-state transition shown by salts of [FeL2] [L = 2,6-di(pyrazol-1- Chemistry Study of Mn(β-diketonato)3 complexes. Electrochim. Acta yl)pyridine]. J. Chem. Soc., Dalton Trans. 2002, 548−554. 2015, 158, 418−426. (45) Halcrow, M. A. Structure:function relationships in molecular (29) Sato, Y.; Takizawa, S.-y.; Murata, S. Substituent Effects on spin-crossover complexes. Chem. Soc. Rev. 2011, 40, 4119−4142. Physical Properties and Catalytic Activities toward Water Oxidation in (46) Robinson, K.; Gibbs, G. V.; Ribbe, P. H. Quadratic Elongation: Mononuclear Ruthenium Complexes. Eur. J. Inorg. Chem. 2015, 2015, A Quantitative Measure of Distortion in Coordination Polyhedra. 5495−5502. Science 1971, 172, 567−570.

2227 DOI: 10.1021/acs.inorgchem.8b03457 Inorg. Chem. 2019, 58, 2218−2228 James McPherson – June 2019 76 Chapter 3: Predictable Substituent Control of Cobalt (III/II) Redox Potential and Spin Crossover in Cobalt Complexes

Inorganic Chemistry Article

(47) Raston, C. L.; White, A. H. Crystal structures of bis(terpyridyl)- of Bimetallic Ru(II) and Os(II) Complexes Derived from Imidazole- cobalt(II) thiocyanate dihydrate: an X-ray-induced phase transition? J. 4,5-Dicarboxylic Acid and 2,2′-Bipyridine. Inorg. Chem. 2013, 52, Chem. Soc., Dalton Trans. 1976,7−12. 6860−6879. (48) Kremer, S.; Henke, W.; Reinen, D. High-spin-low-spin (68) Nemykin, V. N.; Purchel, A. A.; Spaeth, A. D.; Barybin, M. V. equilibriums of cobalt(2+) in the terpyridine complexes Co(terpy)2X2 Probing the Electronic Properties of a Trinuclear Molecular Wire − nH2O. Inorg. Chem. 1982, 21, 3013 3022. Involving Isocyanoferrocene and Iron(II) Phthalocyanine Motifs. (49) Figgis, B.; Kucharski, E.; White, A. Crystal structure of Inorg. Chem. 2013, 52, 11004−11012. Bis(2,2’:6’,2’’-terpyridyl)cobalt(II) perchlorate c. 1.3 hydrate. Aust. J. (69) Dance, I. The Correlation of Redox Potential, HOMO Energy, Chem. 1983, 36, 1537−1561. and Oxidation State in Metal Sulfide Clusters and Its Application to (50) McPherson, J. N.; Elton, T. E.; Colbran, S. B. A Strain- Determine the Redox Level of the FeMo-co Active-Site Cluster of Deformation Nexus within Pincer Ligands: Application to the Spin Nitrogenase. Inorg. Chem. 2006, 45, 5084−5091. States of Iron(II) Complexes. Inorg. Chem. 2018, 57, 12312−12322. (70) Piguet, C. Paramagnetic Susceptibility by NMR: The ″Solvent (51) Chow, H. S.; Constable, E. C.; Housecroft, C. E.; Kulicke, K. J.; Correction″ Removed for Large Paramagnetic Molecules. J. Chem. Tao, Y. When electron exchange is chemical exchange-assignment of Educ. 1997, 74, 815. 1H NMR spectra of paramagnetic cobalt(ii)-2,2′:6′,2″-terpyridine (71) Turner, J. W.; Schultz, F. A. Solution Characterization of the complexes. Dalton Trans. 2005, 236−237. Iron(II) Bis(1,4,7-Triazacyclononane) Spin-Equilibrium Reaction. (52) Kilner, C. A.; Halcrow, M. A. An unusual discontinuity in the Inorg. Chem. 2001, 40, 5296−5298. thermal spin transition in [Co(terpy)2][BF4]2. Dalton Trans. 2010, (72) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; 39, 9008−9012. Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, (53) Shatruk, M.; Phan, H.; Chrisostomo, B. A.; Suleimenova, A. B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. Symmetry-breaking structural phase transitions in spin crossover P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; complexes. Coord. Chem. Rev. 2015, 289−290,62−73. Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; (54) Collet, E.; Guionneau, P. Structural analysis of spin-crossover Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; materials: From molecules to materials. C. R. Chim. 2018, 21, 1133. Montgomery, J. A.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, (55) Nakaya, M.; Ohtani, R.; Shin, J. W.; Nakamura, M.; Lindoy, L. J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; F.; Hayami, S. Abrupt spin transition in a modified-terpyridine Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. cobalt(II) complex with a highly-distorted [CoN6] core. Dalton Trans. 2018, 47, 13809−13814. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; (56) Evans, D. F. 400. The determination of the paramagnetic Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; susceptibility of substances in solution by nuclear magnetic resonance. Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; J. Chem. Soc. 1959, 2003−2005. Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. (57) Schweinfurth, D.; Weisser, F.; Bubrin, D.; Bogani, L.; Sarkar, B. D.; Farkas; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Cobalt Complexes with “Click”-Derived Functional Tripodal Ligands: Gaussian 09, Revision B.01; Gaussian Inc.: Wallingford CT, 2009. Spin Crossover and Coordination Ambivalence. Inorg. Chem. 2011, (73) Lee, C.; Yang, W.; Parr, R. G. Development of the Colle- 50, 6114−6121. Salvetti correlation-energy formula into a functional of the electron − (58) Kahn, O. Molecular Magnetism. VCH Publishers Inc.: New density. Phys. Rev. B: Condens. Matter Mater. Phys. 1988, 37, 785 789. York, 1993. (74) Becke, A. D. Density functional thermochemistry. III. The role − (59) Clerac,́ R.; Cotton, F. A.; Dunbar, K. R.; Lu, T.; Murillo, C. A.; of exact exchange. J. Chem. Phys. 1993, 98, 5648 5652. Wang, X. A new linear tricobalt compound with di(2-pyridyl)amide (75) Stephens, P.; Devlin, F.; Chabalowski, C.; Frisch, M. J. Ab initio calculation of vibrational absorption and circular dichroism spectra (dpa) ligands: two-step spin crossover of [Co3(dpa)4Cl2][BF4]. J. Am. − Chem. Soc. 2000, 122, 2272−2278. using density functional force fields. J. Phys. Chem. 1994, 98, 11623 (60) Enamullah, M.; Hasegawa, M.; Fukuda, Y.; Linert, W.; Hoshi, 11627. T. Synthesis, characterization and spin-crossover behaviors of (76) Petersson, G.; Al Laham, M. A. A complete basis set model chemistry. II. Open shell systems and the total energies of the first [Co(hydroxycarboxylato)(triphos)] complexes. Bull. Chem. Soc. Jpn. − 2002, 75, 2449−2453. row atoms. J. Chem. Phys. 1991, 94, 6081 6090. (61) Toftlund, H. Spin equilibrium in solutions. Monatsh. Chem. (77) Grimme, S.; Ehrlich, S.; Goerigk, L. Effect of the damping − function in dispersion corrected density functional theory. J. Comput. 2001, 132, 1269 1277. − (62) Rodríguez-Jimenez,́ S.; Barltrop, A. S.; White, N. G.; Feltham, Chem. 2011, 32, 1456 1465. (78) Neese, F. The ORCA program system. Wiley Interdiscip. Rev. H. L. C.; Brooker, S. Solvent Polarity Predictably Tunes Spin − Crossover T1/2 in Isomeric Iron(II) Pyrimidine Triazoles. Inorg. Comput. Mol. Sci. 2012, 2,73 78. Chem. 2018, 57, 6266−6282. (79) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. A consistent and (63) Swain, C. G.; Lupton, E. C. Field and resonance components of accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu. J. Chem. Phys. 2010, substituent effects. J. Am. Chem. Soc. 1968, 90, 4328−4337. 132, 154104. (64) Tsuneda, T.; Song, J.-W.; Suzuki, S.; Hirao, K. On Koopmans’ (80) Güell, M.; Luis, J. M.; Sola,̀ M.; Swart, M. Importance of the theorem in density functional theory. J. Chem. Phys. 2010, 133, Basis Set for the Spin-State Energetics of Iron Complexes. J. Phys. 174101. Chem. A 2008, 112, 6384−6391. (65) Haslinger, S.; Kück, J. W.; Hahn, E. M.; Cokoja, M.; Pöthig, A.; (81) Weigend, F. Accurate Coulomb-fitting basis sets for H to Rn. Basset, J.-M.; Kühn, F. E. Making Oxidation Potentials Predictable: Phys. Chem. Chem. Phys. 2006, 8, 1057−1065. Coordination of Additives Applied to the Electronic Fine Tuning of (82) Takano, Y.; Houk, K. N. Benchmarking the Conductor-like an Iron(II) Complex. Inorg. Chem. 2014, 53, 11573−11583. Polarizable Continuum Model (CPCM) for Aqueous Solvation Free (66)Blusch,L.K.;Craigo,K.E.;Martin-Diaconescu,V.; Energies of Neutral and Ionic Organic Molecules. J. Chem. Theory McQuarters, A. B.; Bill, E.; Dechert, S.; DeBeer, S.; Lehnert, N.; Comput. 2005, 1,70−77. Meyer, F. Hidden Non-Innocence in an Expanded Porphyrin: ’ (83) Cowieson, N. P.; Aragao, D.; Clift, M.; Ericsson, D. J.; Gee, C.; Electronic Structure of the Siamese-Twin Porphyrin s Dicopper Harrop, S. J.; Mudie, N.; Panjikar, S.; Price, J. R.; Riboldi-Tunnicliffe, Complex in Different Oxidation States. J. Am. Chem. Soc. 2013, 135, − A.; Williamson, R.; Caradoc-Davies, T. MX1: a bending-magnet 13892 13899. crystallography beamline serving both chemical and macromolecular (67) Das, S.; Karmakar, S.; Saha, D.; Baitalik, S. A Combined crystallography communities at the Australian Synchrotron. J. Experimental and DFT/TD-DFT Investigation of Structural, Synchrotron Radiat. 2015, 22, 187−190. Electronic, and Cation-Induced Switching of Photophysical Properties

2228 DOI: 10.1021/acs.inorgchem.8b03457 Inorg. Chem. 2019, 58, 2218−2228 James McPherson – June 2019 77 Chapter 3: Predictable Substituent Control of Cobalt (III/II) Redox Potential and Spin Crossover in Cobalt Complexes

James McPherson – June 2019 78 Chapter 4: Neutral Lanthanide Complexes

4 NEUTRAL LANTHANIDE COMPLEXES

4.1 Prior Publication Statement

The work in this chapter has been accepted for publication by the Royal Society of

Chemistry in Dalton Transactions. This article has been reprinted with permission from

McPherson, J. N.; Abad Galan, L.; Iranmanesh, H.; Massi, M.; and Colbran S. B., Synthesis

and Structural, Redox and Photophysical Properties of Tris-(2,5-di(2-pyridyl)pyrrolide)

Royal Society of Chemistry.

4.2 Contributions

JNM assisted with experimental design, coordinated all experiments, prepared and

characterised (except photophysical studies) all complexes, analysed data, wrote the draft

manuscript (except for the sections on photophysical studies), compiled the manuscript and

prepared the supporting information.

LAG collected all photophysical data and wrote a first draft of the photophysical sections

of the manuscript. HI collected the X-ray crystallographic data and refined the structure for two

Me,Me CO Me,Me) complexes: [La(dpp )3] and [Sm(dpp 2 3]. MM mentored LAG, designed the

photophysical studies, and commented on the manuscript. SBC mentored JNM, conceived and

supervised the project, and refined the manuscript.

4.3 Foreword

The first rare-earth metal complexes of 2,5-di(2-pyridyl)pyrrolide (dpp–) ligands were

prepared, and their physical properties are described in this chapter. This work was performed

James McPherson – June 2019 79 Chapter 4: Neutral Lanthanide Complexes in collaboration with A/Prof. Max Massi and Dr Laura Abad Galan at Curtin University, Perth,

WA, Australia.

James McPherson – June 2019 80 Chapter 4: Neutral Lanthanide Complexes Dalton Transactions

View Article Online PAPER View Journal

Synthesis and structural, redox and photophysical properties of tris-(2,5-di(2-pyridyl)pyrrolide) Cite this: DOI: 10.1039/c9dt01262a lanthanide complexes†

James N. McPherson, a Laura Abad Galan, b Hasti Iranmanesh, a Massimiliano Massi *b and Stephen B. Colbran *a

R1,R2 A ﬁrst series of lanthanide complexes of tris(dipyridyl)pyrrolide ligands has been prepared. The [Ln(dpp )3] R1,R2 1 2 complexes (Ln = La(III), Sm(III), Eu(III), Gd(III)andYb(III); and dpp = 2,5-di(2-pyridyl-3-(R )-4-(R )) pyrrolide) have been isolated and their structures and photophysical and redox properties characterised, both in the solid-state and in solution. In the complexes, the three dpp− ligands form a distorted tricapped trigonal prismatic coordination geometry about the lanthanide ions, with the antiparallel isomer observed in the solid state − for non-symmetric (dppCO2Me,Me) .However,1H NMR spectroscopy of the diamagnetic and paramagnetic R1,R2 [Ln(dpp )3]complexesind6-benzene solution reveal evidence for a statistical distribution of all possible − isomers. Time-resolved luminescence studies suggest that the dpp ligand (with triplet excited state T1 energy

−1 CO2Me,Me Received 25th March 2019, at 18 622 cm ) sensitises red emission from [Eu(dpp )3] and near-infrared emission from

Accepted 24th May 2019 CO2Me,Me [Yb(dpp )3] through the antenna eﬀect. Cyclic voltammetry reveals three consecutive, reversible, one- DOI: 10.1039/c9dt01262a R1,R2 − electron oxidation processes for each [Ln(dpp )3] complex, corresponding to oxidations of each dpp ligand

+/0 CO2Me,Me III/II +/0 rsc.li/dalton between 0.3–0.8 V vs. E1/2 (Fc ), and for [Eu(dpp )3]theEu couple was −2.099 V vs. E1/2 (Fc ).

Introduction A recent review of the chemistry of meridional tridentate pyridine–pyrrolide ligands, anionic analogues of 2,2′;6′,2″-ter- The luminescence of trivalent lanthanide (Ln) cations results pyridine (tpy), revealed that no complexes with lanthanide in sharp emission bands with long lifetimes characteristic of ions have been prepared with the 2,5-di(2-pyridyl)pyrrolide − f–f transitions.1 These properties are relevant across a range of (dpp ) family of ligands.13 In contrast, there are many reports Published on 05 June 2019. Downloaded by UNSW Library 6/11/2019 12:14:40 AM. – applications,2 such as optical fibre lasers and amplifiers,3 or of terpyridine–lanthanide complexes.14 18 – imaging and sensing of biological samples.4 7 In biological We envisaged that the wider bite angle and anionic character − samples, the luminescence from long-lived Ln(III) excited- of the dpp ligands might result in lanthanide complexes that states overcome complications due to shorter-lived biological were less labile, particularly with respect to ligand exchange reac- auto-fluorescence, and near-infrared (NIR) emission from tions (e.g., with water), than their terpyridine–lanthanide ana- − Ln(III) cations, such those based on Yb(III), is particularly logues. Hence, we report the first complexes of the dpp anion

desirable for its optimal penetration of tissues. with rare-earth elements, namely the complexes [Ln(dpp)3](Ln= On the downside, however, Ln(III) ions have inherently La, Sm, Eu, Gd and Yb). These neutral complexes were easily pre- small molar absorptivities as the f–f transitions are Laporte, pared, and have been characterised in solution by mass spec- and often also spin, forbidden. Direct excitation of these ions trometry and 1H NMR spectroscopy, and in the solid state by is therefore impractical. Decoration of these ions with anten- single crystal X-ray diﬀraction (SC-XRD) analyses. Their photo- nae chromophores can lead to sensitised emission with much physical properties were also studied, revealing sensitised Ln(III) – improved luminescence properties.8 12 emission due to the antenna eﬀect. Finally, cyclic voltammetry (CV) showed enhanced redox chemistry due to ligand centred pyrrolide oxidation couples. aSchool of Chemistry, University of New South Wales, UNSW, Sydney, NSW 2052, Australia. E-mail: [email protected] bDepartment of Chemistry, Curtin University, WA 6102, Australia. E-mail: [email protected] Results and discussion †Electronic supplementary information (ESI) available: SC-XRD data, NMR and Synthesis and characterisation HR mass spectra, additional photophysical data and cyclic voltammetry data. CCDC 1904928–1904930. For ESI and crystallographic data in CIF or other elec- Rigorously dry solvents and glassware had to be used for suc- tronic format see DOI: 10.1039/c9dt01262a cessful preparations. The yellow [Ln(dpp)3] complexes were

This journal is © The Royal Society of Chemistry 2019 Dalton Trans. James McPherson – June 2019 81 View Article Online Paper Chapter 4: Neutral Lanthanide Complexes Dalton Transactions

precipitated directly when acetonitrile solutions of the appropriate Hdpp precursor and triethylamine and of the requisite lanthanide(III) nitrate hydrate salt were combined; Scheme 1.

The [Ln(dpp)3] complexes were separated from insoluble inorganic impurities by extracting the crude residue with benzene, and the microcrystalline yellow products were collected following the addition of n-pentane to the benzene solution. Solution state NMR and time-resolved emission data (see − below) show that the dpp ligands dissociate from these

[Ln(dpp)3] complexes in solution on timescales of minutes to CO2Me,Me hours, with [Yb(dpp )3] dissociating the fastest.

Fig. 1 Left: View of the complex molecule from the single crystal X-ray SC-XRD structures Me,Me diffraction structure of [La(dpp )3]·1.5C6H6 (50% thermal ellipsoids at Me,Me 100 K with benzene solvent molecules and H-atoms omitted for clarity). Crystals suitable for SC-XRD studies of [La(dpp )3], CO2Me,Me CO2Me,Me Right: Illustration of the tricapped trigonal prismatic geometry of the La [Sm(dpp )3] and [Eu(dpp )3] were grown in the (III) centre. The distorted trigonal prism is outlined in red dashed lines, ff − dark by the slow di usion of pentane into refrigerated solu- and the N donor atoms from each dpp ligand are connected by black tions of the appropriate [Ln(dpp)3] complex in benzene. lines. Key bond lengths and angles are summarised in Table 1. CO2Me,Me Attempts to grow crystals of [Yb(dpp )3] suitable for SC-XRD studies under similar conditions were unsuccessful, − consistent with its faster dpp ligand dissociation kinetics. CO2Me,Me CO2Me,Me Both [Sm(dpp )3] and [Eu(dpp )3] crystallised ligands are relatively unstrained, with a mean N1⋯N3 separ- ρ in the monoclinic space group, P21/n, with very similar unit ation ( L(coord)) of 4.81 Å, from which the internal strain that cells (see ESI, Table S1†) and extremely similar molecular geo- builds within each ligand as it binds to the La3+ ion may be − metries. Overlaying the molecules by matching the Ln(III) ion estimated to be ∼45 kJ mol 1.20 – – and all 9 coordinating N donor atoms results in an RMSD of As expected, the average Ln Npy (2.76 Å) and Ln Npyrr † Me,Me only 0.0119 Å (see ESI, Fig. S1 ). Complex [La(dpp )3] co- (2.52 Å) bond lengths are longer than in transition metal com- crystallised with lattice benzene molecules, while the Sm(III) plexes (M = FeII,CoII,CoIII,NiII,CuII,ZnII and RuII) with the and Eu(III) analogues crystallised with half a water molecule in same ligand scaffolds (by more than 0.3 Å and up to 0.8 Å for the asymmetric unit, that formed an incomplete hydrogen CoIII).13 The N1B pyridyl rings of neighbouring molecules are bonded chain along the b-axis (see Fig. ESI, S2†). involved in π-stacking interactions (plane–plane distance of − The N-donor atoms of the three dpp ligands in the 3.76 Å, and a shift of 1.75 Å), as are the N3A rings (3.83 and Me,Me complex, [La(dpp )3], constitute a distorted tricapped tri- 1.96 Å). gonal prismatic geometry, with a continuous shape measure- The tricapped trigonal prismatic geometries in the

Published on 05 June 2019. Downloaded by UNSW Library 6/11/2019 12:14:40 AM. CO2Me,Me ment (CShM) of 2.60 against an ideal spherical tricapped trigo- [Ln(dpp )3] structures (Fig. 2 and Table 1) are less nal prism (CShM = 0 for the ideal geometry).19 The triangular distorted: CShMs of 1.85 for Sm and 1.80 for Eu, with the tri- faces of the tricapped trigonal prism are defined by a pyridyl angular faces closer to parallel (α = 4.07(11)° for Sm and − N-donor from each dpp ligand, and the pyrrolide N-donors 4.39(19)° for Eu) and less twisted (β = 16° for both Sm and “ ” Me,Me form the caps over each rectangular face; see Fig. 1. The tri- Eu), than in the structure of [La(dpp )3] discussed CO2Me,Me − angular faces are not parallel (the angle between the normals above. The (dpp ) ligands are also wider (ρL(coord) = to each plane, α, is 9.20(6)°), and the faces have twisted (with a 4.70 Å for Sm, 4.72 Å for Eu) and, therefore, less strained – – – −1 mean Npy centroid(Npy,Npy,Npy) centroid(Npy′,Npy′,Npy′) Npy′ (at ∼15 kJ mol ) than transition metal analogues of the same torsion, β, of 20°) so that they do not eclipse each other. Both ligand.22

enantiomers are present in the centrosymmetric unit cell. The There are four stereoisomers possible for a [Ln(dpp)3] − complex with a non-symmetric ligand such as (dppCO2Me,Me) . The four stereoisomers are depicted diagramatically in Fig. 3 with the non-symmetric ligand represented by blue arrows. Each ligand in the parallel enantiomers (M↑↑↑ and P↑↑↑)is chemically equivalent, while in the anti-parallel (M↑↑↓ and P↓↑↑) isomers each ligand is chemically unique. Although all four isomers are present in solution (see the NMR section below), the crystals characterised by SC-XRD studies of these complexes were of the more statistically probable “antiparallel” enantiomers, which are shown in Fig. 3 with the “B” ligand opposite in orientation to the A and C ligands. Both enantio-

Scheme 1 Preparation of the neutral [Ln(dpp)3] complexes. mers are present in the achiral crystals.

Dalton Trans. This journal is © The Royal Society of Chemistry 2019 James McPherson – June 2019 82 View Article Online Dalton Transactions Chapter 4: Neutral Lanthanide Complexes Paper

CO2Me,Me Fig. 2 Top: Views of the complex molecules from the single crystal X-ray diﬀraction structures of [Sm(dpp )3]·0.5(H2O) (left) and CO2Me,Me [Eu(dpp )3]·0.5(H2O) (right) (50% thermal ellipsoids at 100 K with the second position for the disordered ester groups in ligand C and the

ligand H-atoms omitted for clarity). OW is the water O-atom. Bottom: Illustrations of the tricapped trigonal prismatic geometries, revealing the − antiparallel isomer. The (dppCO2Me,Me) ligands are represented by vectors pointing from N3, the pyridine nearest the methyl substituent.

Table 1 Selected bond lengths (Å) and angles (°)

Me,Me CO2Me,Me CO2Me,Me [La(dpp )3] [Sm(dpp )3] [Eu(dpp )3]

Ligand A Ligand B Ligand C Ligand A Ligand B Ligand C Ligand A Ligand B Ligand C

Ln–N1 2.697(2) 2.790(2) 2.777(2) 2.678(3) 2.676(4) 2.642(4) 2.683(5) 2.685(5) 2.658(5) Ln–N2 2.523(2) 2.530(2) 2.518(2) 2.426(3) 2.426(3) 2.423(3) 2.429(3) 2.425(5) 2.412(5) Ln–N3 2.766(2) 2.790(2) 2.737(2) 2.630(3) 2.640(3) 2.639(3) 2.635(5) 2.640(6) 2.640(6) ρ ⋯ L(coord) {N1 N3} 4.770(3) 4.854(3) 4.819(3) 4.709(4) 4.708(4) 4.687(5) 4.727(8) 4.720(8) 4.716(8)

N1–Ln–N2 61.12(5) 60.12(5) 60.75(5) 62.2(1) 61.6(1) 62.8(1) 62.47(18) 61.71(17) 63.12(19) N3–Ln–N2 60.67(5) 60.99(5) 61.25(5) 62.8(1) 63.1(1) 62.4(1) 62.97(18) 63.24(18) 62.80(19) N1–Ln–N3 121.60(5) 120.87(5) 121.85(5) 125.03(9) 124.6(1) 125.1(1) 125.44(18) 124.85(17) 125.8(2)

α 9.20(6) 4.07(11) 4.39(19) β 21 ± 2 16 ± 1 16 ± 1 CShM (s-TCTPR) 2.60 1.85 1.80

Published on 05 June 2019. Downloaded by UNSW Library 6/11/2019 12:14:40 AM. α β – – – = the angle between the normals to each triangular face; = average Npy centroid(Npy,Npy,Npy) centroid(Npy′,Npy′,Npy′) Npy′ torsion. CShM: continuous shape measurement against a spherical tricapped trigonal prismatic (s-TCTPR) geometry; higher values indicate greater deviation from ideal geometry.19,21

− Mass spectrometry signals grew with time, and may be attributed to dpp ligand CO2Me,Me CO2Me,Me dissociation. [Gd(dpp )3] was essentially silent by NMR (+)-NSI-MS spectra of the [Ln(dpp )3] (Ln = La, Sm and Yb) complexes aspirated from acetonitrile solution showed low spectroscopy, exhibiting only a single very broad signal ≥ δ ≥− † intensity peak envelopes for the singly-charged protic or between 200 100 ppm (see ESI, Fig. S21 ). 1 Me,Me sodium adducts of the complex as expected for neutral The H NMR spectrum of [La(dpp )3] exhibited four δ species. The predominant peaks corresponded to the sodium aromatic multiplets at 8.13 (A6), 7.39 (A3), 6.80 (A4), 6.07 δ adducts of Hdpp and its dimer; e.g., ESI, Fig. S23–25.† The (A5), and a methyl singlet at 2.36 as expected for this sym- − 3+ † 13 1 results suggest protonation and dissociation the dpp ligands metric La complex; see ESI, Fig. S3. The C{ H} NMR spec- is facile under electrospray conditions. For [La(dppMe,Me) ] trum was likewise straightforward showing 8 signals as 3 † 1 however, with no obvious coordination sites for protons or expected for the complex; ESI, Fig. S7. The H NMR spectrum CO2Me,Me for [La(dpp )3] was far more complicated, with four sodium ions, only monomeric, dimeric and trimeric [(Hdpp)x + Na]+ species were observed in the spectrum; ESI, Fig. S26.† equal intensity signals observed for each methyl environment; see Fig. 4, top. This observation is consistent with a 1 : 3 statistical distribution of the parallel to antiparallel isomers NMR spectroscopy (Fig. 3) in solution, with overlapping signals for the 32 Signals from unbound Hdpp ligands were observed in all NMR diﬀerent pyridyl 1H environments complicating the assign- 1 spectra of [Ln(dpp)3] complexes in d6-benzene solutions. These ment. As Fig. 4 displays, the H NMR spectra for the paramag-

This journal is © The Royal Society of Chemistry 2019 Dalton Trans. James McPherson – June 2019 83 View Article Online Paper Chapter 4: Neutral Lanthanide Complexes Dalton Transactions

1 CO2Me,Me Fig. 5 H NMR (600 MHz, C6D6) spectrum of [Sm(dpp )3].

Fig. 3 Representation of the possible helical isomers with three non- CO2Me,Me − symmetric planar tridentate ligands, such as (dpp ) (each rep- the less probable parallel isomers, and three unique ligand resented by a blue arrow), about an idealised tricapped trigonal prism. environments in the more abundant antiparallel isomers). Two of these four environments are coincidental for each type of proton, and so a 1 : 1 : 2 distribution is observed; however close inspection of the 1H–1H COSY NMR spectrum confirms that there are four different overlapping spin systems (e.g., A6/ C6 at δ 2.87), and not three unique 1 : 1 : 2 environments. The 1H NMR spectra at different concentrations of CO2Me,Me [Sm(dpp )3] are superimposable and independent of field strength (ESI, Fig. S13†). Although the eight H6 protons show the most evidence of 5 paramagnetic shifting due to the 4f Sm(III) ion, all signals are broadened sufficiently that the multiplicity is obscured. The major scalar coupling constants can be approximated, either directly as in the case of C3 or by measuring the linewidth of the broadened multiplets (e.g., fwhh ∼13 Hz for H5), and agree Fig. 4 The methyl and methyl carboxylate regions from the 1H NMR with those typical for the free ligand or in other diamagnetic CO2Me,Me (400 MHz, C6D6) spectra of [Ln(dpp )3] (from top to bottom: complex species, although the H6 signals (fwhh ∼14 Hz) show Published on 05 June 2019. Downloaded by UNSW Library 6/11/2019 12:14:40 AM. Ln = La, Sm, Eu or Yb), with expansions shown in the insets. significant additional broadening due to the paramagnetic Sm(III) centre. 13 1 CO2Me,Me The 151 MHz C{ H} spectrum of [Sm(dpp )3]was 13 CO Me,Me also acquired, but with 85 expected C signals (17 signals for netic [Ln(dpp 2 ) ] (Ln = Sm, Eu and Yb) complexes simi- − 3 each dpp ligand environment of which there are four, plus 17 larly reveal a mixture of isomers. On the timescale of the NMR for the unbound Hdpp which dissociated during the experi- experiments, no equilibration is observed between isomers or ment) it was not assigned other than to inspect the 13C signals with the free Hdpp ligand. for the methyl and methyl ester groups and, with the aid of a Fig. 5 shows the 600 MHz 1H NMR spectrum of 1H–13C HSQC experiment, to confirm the presence of five [Sm(dppCO2Me,Me) ]ind -benzene. The signals for the complex 3 6 environments each; e.g., ESI, Fig. S17.† were sufficiently shifted by the paramagnetic Sm(III) centre that assignment was possible. The H3, H4, H5 and H6 signals from − each pyridyl arm of the bound dpp ligands can be confidently Photophysical investigations assigned from 1H–1H COSY experiments (Fig. S15, ESI†), and The absorption and emission spectra of the ligand precursor 1 –13 CO2Me,Me CO2Me,Me the rings can be discriminated through H C HMBC and Hdpp and of the [Gd(dpp )3] complex were 1H–1H NOESY experiments particularly due to the C3–H- recorded in air-equilibrated dichloromethane solutions at methyl correlations (ESI, Fig. S16†). During the solution phase room temperature and at 77 K, to characterise the ligand experiments (∼4 h), some ligand dissociation occurred with centred excited states. The photophysical properties of the CO2Me,Me CO2Me,Me the signals attributed to the unbound Hdpp ligand increasing [Eu(dpp )3] and [Yb(dpp )3] complexes were over time (ESI, Fig. S14†). The statistical distribution of then studied in air-equilibrated dichloromethane solutions −5 isomers results in four unique Sm(III)-bound ligand environ- (ca. 10 M) at room temperature and at 77 K, as well as in the ments of almost equal intensity (three equivalent ligands in solid state (as powders). The stability of the complexes in solu-

Dalton Trans. This journal is © The Royal Society of Chemistry 2019 James McPherson – June 2019 84 View Article Online Dalton Transactions Chapter 4: Neutral Lanthanide Complexes Paper

tion was limited and, therefore, fresh solutions were prepared and measurements performed within 30 min. The photophysical data, including excited state lifetime

decay (τobs), calculated radiative lifetime decay (τR) and intrin- ΦLn sic quantum yield ( Ln) for these complexes are reported in Table 2. The absorption profile of the ligand and the corresponding −5 Gd(III) complex in dichloromethane (at ca. 10 M) were found to be comparable with the characteristic red shift of the ligand band at 350 nm upon coordination (see Fig. 6), in accordance with previous data reported for terpyridine ligands and their binding to the lanthanide centres.24 CO2Me,Me − The singlet and triplet excited states of the (dpp ) CO2Me,Me Fig. 7 Emission spectra (λexcitation = 300 nm) of [Gd(dpp )3] −1 −1 −5 ligand were estimated to be at 26 257 cm and 18 622 cm , complex in CH2Cl2 (ca. 10 M) at 298 K (black trace) and 77 K (red respectively, from measurements of the fluorescent and phos- trace). CO2Me,Me phorescent emission of [Gd(dpp )3]inCH2Cl2 solution at 298 and 77 K (Fig. 7). The lifetime values of these excited states were measured to be 2.1 ns at 450 nm and 2.3 ms at 550 nm, respectively, at 77 K. The long lifetime value found at 550 nm is in agreement with the triplet spin multiplicity of this − excitated state. The energy of the triplet state of (dppCO2Me,Me) −1 5 at 18 622 cm is slightly above the energy of the D0 state of − CO2Me,Me 1 25 Table 2 Photophysical data for the [Ln(dpp )3] complexes Eu(III)(at∼17 200 cm , 581 nm), suggesting energy transfer might not be optimal, as a minimum energy difference (ΔE)of λ τ −1 em R 3500 cm is generally required to minimise back energy trans- Complex Medium (nm) τ (ms) ΦLn obs Ln fer.26 The triplet state energy should, however, be sufficient to —— 2 ∼ −1 [Eu CH2Cl2 400 1.3 ns sensitise the F5/2 state of Yb(III)(at 10 200 cm ), possibly via CO2Me,Me (dpp )3] (298 K) intermediate charge transfer excited states.27 CH2Cl2 612 1.2 ms 1.3 0.94 (77 K) In each case, the lanthanide-centred emission originated as Powder 500 3.6 ns (74%) —— a consequence of the antenna effect as the excitation spectra 12.6 ns (26%) CO Me,Me were analogous to the absorption profile of the Hdpp 2 612 70.1 ns —— ligand (see ESI, Fig. S27–S32†). —— CO2Me,Me [Yb CH2Cl2 400 1.2 ns The emission spectra of [Eu(dpp )3] in solution and CO2Me,Me a ∼ (dpp )3] (298 K) 980 11.9 µs 1.2 0.01 —— in the solid state are overlaid in Fig. 8. The emission spectra in CH2Cl2 550 2.2 ms a (77 K) 980 12.1 μs 1.2 ∼0.01 aCH2Cl2 solution at 298 K showed almost exclusively ligand-

Published on 05 June 2019. Downloaded by UNSW Library 6/11/2019 12:14:40 AM. Powder 500 2.9 ns (42%) —— centred emission from the singlet state at 395 nm. This might 9.3 ns (58%) be due to rapid dissociation of the ligand coupled with ineﬃ- 980 12.1 μs 1.2a ∼0.01 cient energy transfer to the Eu(III) centre. In contrast, in the a 23 Literature value for Yb(III) ion. frozen matrix at 77 K, where thermally activated back energy transfer should be suppressed, characteristic Eu(III) emission

CO2Me,Me Fig. 8 Emission spectra (λexcitation = 300 nm) of [Eu(dpp )3](ca. CO2Me,Me −5 −5 Fig. 6 Absorption proﬁle of Hdpp (ca. 10 M) (black trace) and 10 M) in CH2Cl2 at 298 K (black trace) and 77 K (red trace) and in the CO2Me,Me −5 [Gd(dpp )3](ca.10 M) in CH2Cl2 at 298 K. solid state at 298 K (blue trace).

This journal is © The Royal Society of Chemistry 2019 Dalton Trans. James McPherson – June 2019 85 View Article Online Paper Chapter 4: Neutral Lanthanide Complexes Dalton Transactions

was observed in the region 580–750 nm corresponding to the time values of 11.9 µs for the CH2Cl2 solution at 298 K, 12.1 µs 5 → 7 – 5 → 7 D0 FJ (J =06) transitions. The fact that the D0 F0 in frozen matrix at 77 K and 12.1 µs for the powdered sample. − band at 585 nm has a FWHM of 40 cm 1 is suggestive of a The similar values suggest similar levels of multiphonon relax- 25 single unique emitting Eu(III) centre. The lifetime decay from ation in all three cases. If a standard radiative lifetime for the 5 23 the D0 state of Eu(III) was found to be monoexponential, with Yb(III) ion in solution is considered (1.2 ms), the intrinsic a value of 1.2 ms. The radiative decay lifetime calculated from quantum yield can be approximated to be ∼0.01, which is the emission spectrum was 1.3 ms, indicating an almost quan- comparable value to similar complexes previously reported in titative intrinsic quantum yield of 0.95. When the same experi- the literature.24,31 ments were performed in the solid state at ambient tempera- Simultaneously, emission arising from the ligand excited ture, the recorded spectrum displayed predominantly ligand states was also observed. Fluorescent emission from a singlet

centred emission overlaid to very weak Eu(III) emission. The excited state was observed from the fresh CH2Cl2 solution of CO2Me,Me ligand centred emission, whose lifetime decay was fitted to a [Yb(dpp )3] at room temperature. In the frozen matrix bi-exponential function with values of 3.6 ns (74%) and at 77 K, phosphorescent triplet emission was also seen at 2 12.6 ns (26%) seems to be a combination of emission from 550 nm, indicative of incomplete energy transfer to the F5/2 monomeric and aggregated singlet states, with potentially state of the Yb(III) ion. The lifetime of the triplet emission some contribution from triplet excited states.28,29 On the other (2.2 ms) was only slightly shorter than that found for triplet CO2Me,Me hand, the excited state lifetime decay for the Eu(III) peak at emission from [Gd(dpp )3] (see above). Eqn (1), where 612 nm is 70.1 ns, suggesting again efficient back energy τq and τu represent the lifetime of the phosphorescent ligand transfer. emission in the presence and absence of Yb(III) ions, respect- CO2Me,Me ffi The measurements of [Eu(dpp )3] in solution were ively, describes the e ciency of the energy transfer, which was repeated after one hour in order to assess the stability of the calculated to be 0.04. complex. The results at room temperature and frozen matrix at τ Φ q 77 K were comparable to those found for fresh solutions. The ET ¼ 1 ð1Þ τu result indicates that the ligand-centred emission results from poor sensitisation processes rather than from free ligand Finally, the broad visible emission band seen for the pow- present in solution following decomposition. dered sample at ambient temperature again is suggestive for CO2Me,Me III Fig. 9 presents the emission spectra of [Yb(dpp )3] an aggregate state as was found for the Eu( ) complex. The bi- in solution at 298 and 77 K and in the solid state at 298 K. In exponential decay was fitted to two values, of 2.9 ns (42%) and each case, characteristic NIR emission was observed corres- 9.3 ns (58%), as is commonly observed for aggregation 2 2 32 ponding to the spin-allowed F5/2 → F7/2 transition for the states. 30 Yb(III) ion. This transition is split into three distinguishable When these measurements were repeated on the solution CO2Me,Me bands in the range of 960–1100 nm due to crystal field effects. of [Yb(dpp )3] held at ambient temperature after The slight differences in the fine-structure are attributed to the 30 min, no Yb(III) ion emission was found while the intensity varying degrees of distortion of the Yb(III) centres in the three of the ligand-centred emission band was highly increased. The CO2Me,Me Published on 05 June 2019. Downloaded by UNSW Library 6/11/2019 12:14:40 AM. different media. The band at 940 nm is assigned to emission results confirm that [Yb(dpp )3] has limited stability in from a hot state as it is completely suppressed at 77 K, when it solution. is not thermally accessible. The observed emission lifetime decays were fitted to monoexponential functions giving life- Cyclic voltammetry CO2Me,Me The CV of [Gd(dpp )3] (blue trace, Fig. 10) is representa- CO2Me,Me tive for other [Ln(dpp )3] complexes (Ln = La, Sm, Eu and Gd, see ESI, Fig. S33†), with three consecutive, but closely- spaced, reversible oxidation couples attributed to one-electron − pyrrolide-centred oxidation of the anionic dpp ligands; − • Table 3. That the pyr /pyr couples are consecutive, separated by 130–160 mV, indicates that the effect of each oxidation is − relayed through the central Ln(III) ion to other dpp ligand(s). CO2Me,Me [Eu(dpp )3] alone of all the complexes, displays a metal- centred couple at negative potentials. The EuIII/EuII couple at +/0 −2.099 V was electrochemically reversible {ΔEp = ΔEp(Fc )=

68 mV} but exhibited poor chemical reversibility (ip,a/ip,c = 0.7) suggesting that the Eu(II) species was unstable and consumed CO2Me,Me on the electrochemical timescale. The [Yb(dpp )3] complex displayed diﬀerent electrochemical responses (ESI, λ CO2Me,Me Fig. 9 Emission spectra ( exc = 300 nm) of the [Yb(dpp )3] † complex in CH Cl (ca. 10−5 M) at 298 K (black trace) and 77 K (red trace) Fig. S34 ), with only one reversible oxidation process (E1/2 = 2 2 +/0 and the solid state (blue trace) in the visible region (350–650 nm) and 0.274 V vs. E1/2(Fc )) followed by two irreversible processes (at – +/0 NIR (850 1150 nm). 0.435 and at 0.709 V vs. E1/2(Fc )). Presumably, this complex

Dalton Trans. This journal is © The Royal Society of Chemistry 2019 James McPherson – June 2019 86 View Article Online Dalton Transactions Chapter 4: Neutral Lanthanide Complexes Paper

complexes precludes biological (but not solid-state) appli- − cations. The dpp core can be built into macrocycles,33,34 which may similarly sensitise Ln(III) emission but form complexes of far greater stability. The complexes were redox active with consecutive oxidations of the three pyrrolide cores − observed indicative for interaction between the dpp ligands

within the Ln(dpp)3 complexes. Notably, the europium(III) II/III − +/0 complex also exhibited an Eu couple at 2.10 V vs. E1/2(Fc ) in dichloromethane.

Experimental General methods All NMR spectra were recorded on Bruker Avance III DPX-400 Fig. 10 CVs of [Gd(dppCO2Me,Me) ] (blue traces) and [Eu(dppCO2Me,Me) ] 3 3 and 600 spectrometers operating at 400 and 600 MHz fre- (red traces) in CH2Cl2–0.1 M [NBu4][PF6]. quency for 1H NMR experiments, and at 100.6 and 150.9 MHz for 13C NMR experiments respectively. All chemical shifts were calibrated against residual solvent signals and Fig. 11 depicts Table 3 Potential data (V vs. E (Fc+/0)) for couples from CVs of the 1/2 the atom numbering schemes employed.35 All coupling con- [Ln(dpp)3] complexes stants (J) are reported in hertz. Signals in NMR spectra are reported as broad (br), singlets (s), doublets (d), triplets (t), E1/2 (R1)/V E1/2 (I)/V E1/2 (II)/V E1/2 (III)/V quartets (q) or unclear multiplets (m). Unless otherwise stated, — a [La(dpp)3] 0.373 0.520 0.742 — all spectra were obtained at 298 K. NMR spectra were pro- [Sm(dpp)3] 0.323 0.498 0.605 [Eu(dpp)3] –2.099 0.356 0.488 0.620 cessed with MestReNova 11.0.1 software. FT-IR experiments — [Gd(dpp)3] 0.355 0.511 0.638 were performed with an ATR FTIR Spectrometer on solid a The third oxidation process occurs at slightly higher potential in this samples. Signals in FT-IR spectra were reported as broad (br), case and is irreversible, so the anodic potential is reported. weak (w), medium (m), or strong (s). High resolution mass spectrometry experiments were performed on a Thermo Fisher Scientific Orbitrap LTQ XL ion trap mass spectrometer using a undergoes ligand dissociation during the course of the electro- nanospray ionisation source. Elemental analyses were per- +/0 chemical experiment, with the process at 0.709 V vs. E1/2(Fc ) formed by the Chemical Analysis Facility at Macquarie close to that observed for precursor HdppCO2Me,Me (ESI, University. Continuous shape measurements (CShMs) were Fig. S34†). performed using the SHAPE 2.1 program developed by Alvarez Published on 05 June 2019. Downloaded by UNSW Library 6/11/2019 12:14:40 AM. et al.21 against a spherical tricapped trigonal prismatic (s-TCTPR) reference geometry.19 Conclusions UV-vis and emission spectra The synthesis of neutral Ln(dpp) complexes is straight- 3 Absorption spectra were recorded at room temperature using a forward. Mixing solutions of the Hdpp precursor, a base and PerkinElmer Lambda 35 UV/Vis spectrometer. Uncorrected metal precursor in a solvent where the complex was poorly steady-state emission and excitation spectra were recorded soluble, quickly lead to saturation and precipitation of the using an Edinburgh FLSP980-stm spectrometer equipped with complex. The methodology should be easily employed to the preparation of other dipyridylpyrrolide–lanthanide complexes, especially as the Hdpp precursor space expands. Complexes with the larger Ln(III) ions (La, Sm, Eu) were less susceptible to ligand exchange than the smaller ions {Gd(III), Yb(III)}, and the Yb(III) complex was the least stable and most diﬃcult complex to handle of those prepared in this work. It was notably unstable to hydrolysis from adventitious water even in the most carefully dried ‘anhydrous’ solvent. − Encouragingly, the dpp ligand sensitises NIR emission in − two cases. The triplet excited state of the (dppCO2Me,Me) ligand − at 18 622 cm 1 is suﬃcient to sensitise NIR emission from III III Yb( ), and also, red emission from Eu( ) at low temperature Fig. 11 Numbering schemes for NMR assignments of [Ln(dpp)3] (77 K). Disappointingly, however, the facile hydrolysis of the species.

This journal is © The Royal Society of Chemistry 2019 Dalton Trans. James McPherson – June 2019 87 View Article Online Paper Chapter 4: Neutral Lanthanide Complexes Dalton Transactions

a 450 W xenon arc lamp, double excitation and emission Synthesis and reagents monochromators, a Peltier-cooled Hamamatsu R928P photo- Unless specified, solvents were taken from an Innovative – multiplier (185 850 nm) and a Hamamatsu R5509-42 photo- Technology Pure Solvent Dispenser immediately prior to use – multiplier for detection of NIR radiation (800 1400 nm). and commercially available reagents were used as received. Emission and excitation spectra were corrected for source Reactions were carried out under inert atmospheres. The intensity (lamp and grating) and emission spectral response HdppMe,Me and HdppCO2Me,Me ligand precursors were prepared (detector and grating) by a calibration curve supplied with the as reported previously.22 instrument. Excited-state decays (τ) for the lanthanide emission were recorded using a microsecond flashlamp while the lifetimes General synthesis of [Ln(dpp)3] complexes

for ligand-centred emission were determined with the single A solution of Ln(NO3)3·xH2O (0.05 mmol) in MeCN (5 mL) was photon counting technique (TCSPC) using pulsed picosecond added to a pale yellow solution of Hdpp (∼50 mg, 0.17 mmol)

LEDs (EPLED 320, FWHM < 800 ps) as the excitation source, and NEt3 (0.3 mL, 2 mmol) in MeCN (5 mL) and stirred at with repetitions rates between 10 kHz and 1 MHz, on the same room temperature for 2 hours. The yellow precipitate was col- Edinburgh FLSP980-stm spectrometer. The goodness-of-fit was lected on Celite™, and washed several times with MeCN, and assessed by minimizing the reduced χ2 function and by visual then re-dissolved in the minimum amount of benzene

inspection of the weighted residuals. Experimental uncertain- required. Pentane was added, and the [Ln(dpp)3] product col- ties are estimated to be ±10%. lected as a yellow precipitate. Small signals in the 1Hor13C To record the luminescence spectra at 77 K, the samples NMR spectra from Hdpp ligand, which appeared during NMR were placed in quartz tubes (2 mm diameter) and inserted in a characterisation for all complexes, were observed and have not special quartz Dewar filled with liquid nitrogen. All the sol- been recorded below—except where they overlap with

vents used in the preparation of the solutions for the photo- [Ln(dpp)3] signals, see ESI.† Me,Me 1 physical investigations were of spectrometric grade. [La(dpp )3]. Yield: 29.3 mg, 43%. H NMR (400 MHz, The radiative decay lifetime was calculated according to C6D6) δ 8.13 (ddd, J = 5.2, 1.9, 0.9 Hz, 1H, A6), 7.39 (ddd, J = eqn (2):25 8.4, 1.1, 0.9 Hz, 1H, A3), 6.80 (ddd, J = 8.4, 7.3, 1.9 Hz, 1H, A4), 6.07 (ddd, J = 7.3, 5.2, 1.1 Hz, 1H, A5), 2.36 (s, 3H, Me). 13C 3 1=τr ¼ AMD;0n ðIT=IMDÞð2Þ NMR (101 MHz, C6D6) δ. 158.1 (A2), 149.9 (A6), 138.8 (B2),

where AMD,0 is the spontaneous emission probability of the 136.9 (A4), 121.0 (B3), 119.2 (A3), 117.6 (A5), 12.6 (Me). FT-IR: 7 ← 5 −1 ˜ν F1 D0 transition with a constant value of 14.65 s , IT/IMD = 2914.3 brw, 2850.8 brw, 1584.1 s, 1502.1 m, 1425.6 s, 1343.3 s, is the ratio of the total integrated area and the magnetic dipole 1286.4 w, 1191.5 w, 1154.3 m, 1054.8 w, 984.0 m, 936.5 m, −1 transition of the Eu3+ emission spectrum,36 and n is the refrac- 866.8 w, 780.7 m, 739.0 m, 689.8 m cm . Elemental analysis tive index of the medium, considered 1.5 in the solid state and calc. for C48H42N9La·1(H2O)·1(C6H6): C, 66.19; H, 5.14; and frozen matrix.37 N 12.86%; found: C, 65.80; H 4.64; and N, 12.55%. Crystals suitable for SC-XRD analysis were grown by the slow diﬀusion

Published on 05 June 2019. Downloaded by UNSW Library 6/11/2019 12:14:40 AM. Me,Me of pentane into a solution of [La(dpp )3] in benzene. CO2Me,Me 1 [La(dpp )3]. Yield: 61.5 mg, 63%. H NMR (400 MHz, Electrochemistry C6D6) δ 8.35–8.25 (m, 1H, A3), 8.16–8.02 (m, 2H, A6 + C6), High-purity dinitrogen was used to deoxygenate the solution 7.38–7.29 (m, 1H, C3), 6.89–6.75 (m, 2H, A4 + C4), 6.23–6.02

before each electrochemical experiment. Solutions of com- (m, 2H, A5 + C5), 3.74–3.61 (4 × s, 3H, CO2Me), 2.74–2.62 (4 × s,

pounds were of ca. 1.0 mM concentration in the CH2Cl2–0.1 M 3H, Me). FT-IR: ˜ν = 2943.5 brw, 1685.7 m, 1588.0 s, 1509.2 m,

[NBu4][PF6] solvent electrolyte system used for the CV experi- 1430.5 s, 1350.8 m, 1287.3 m, 1189.1 w, 1145.5 s, 1072.1 s, − ments. The solvent was taken from an Innovative Technology 974.1 m, 898.1 w, 783.3 s, 742.6 s, 693.7 w cm 1.

Pure Solvent Dispenser immediately prior to use and electro- (+)-HR-NSI-MS found: m/z 1038.2202 (calc. [C51H42N9O6La lyte was oven dried at 105 °C for 1 h prior to use. (M) + Na]+: m/z 1038.2214). Elemental analysis calc. for

CVs were recorded using a GAMRY Reference-600 potentio- C51H42N9O6La·0.5(H2O)·0.5(C6H6): C, 60.96; H, 4.36; and N stat. A three-electrode system was used, consisting of a glassy 11.85%; found: C, 60.82; H 4.04; and N, 11.67%. – CO2Me,Me 1 carbon working electrode (1 mm diameter eDAQ), a platinum [Sm(dpp )3]. Yield: 20.7 mg, 27%. H NMR (600 MHz, wire counter electrode, and a Ag/AgCl (leakless – eDAQ) refer- C6D6) δ 9.29–9.08 (m, 1H, C3), 8.14–7.94 (m, 1H, A3), 6.87–6.73 ence electrode. Immediately prior to recording each CV, the (m, 1H, C4), 6.73–6.60 (m, 1H, A4), 5.36–5.02 (m, 2H, C5 & A5), – – ′ – working electrode was polished with increasingly fine 3.98 (s, 3H, CO2Me), 3.61 3.51 (m, 3H, Me ), 2.96 2.38 (m, alumina, rinsed with water then solvent to be used in the 2H, C6 & A6). (+)-HR-NSI-MS found: m/z 1029.2509 (calc. + experiment, and dried by rubbing with a clean tissue. Prior to [C51H42N9O6Sm (M) + H] : m/z 1029.2528), m/z 1051.2327 (calc. recording CVs of the compounds, a blank CV of the solvent- [M + Na]+: m/z 1051.2348). Elemental analysis calc. for

electrolyte background was always obtained between anodic C51H42N9O6Sm·0.5(H2O)·0.5(C6H6): C, 60.31; H, 4.31; and and cathodic solvent discharge limits. Unless otherwise stated, N 11.72%; found: C, 60.41; H 4.02; and N, 11.60%. Crystals − all CVs were acquired at a scan rate, ν, of 100 mV s 1. suitable for SC-XRD analysis were grown in the dark by

Dalton Trans. This journal is © The Royal Society of Chemistry 2019 James McPherson – June 2019 88 View Article Online Dalton Transactions Chapter 4: Neutral Lanthanide Complexes Paper

slow diﬀusion of pentane into a refrigerated solution of ments were carried out using SHELXL-2015 41 through Olex2 42 CO2Me,Me [Sm(dpp )3] in benzene. suite of software. The non-hydrogen atoms were refined aniso- CO2Me,Me 1 [Eu(dpp )3]. Yield: 41.0 mg, 54%. H NMR (400 MHz, tropically. One and half molecules of solvent benzene could be

C6D6) δ 33.5–33.1 (vbrs, 1H), 33.1–32.7 (vbrs, 1H), 32.4–32.0 identified from the diﬀerence map and was included in the (brs, 1H), 32.0–31.7 (brs, 1H), 31.7–31.3 (brs, 1H), 31.0–30.4 least-squares refinement. CO2Me,Me (brs, 1H), 28.8–28.4 (brs, 1H), 28.4–27.9 (brs, 1H), 9.9 (d, J = 2{[Sm(dpp )3]·0.5(H2O)} (CCDC 1904929†). Yellow CO2Me,Me 7 Hz, 1H), 9.8 (d, J = 7 Hz, 1H), 9.4 (d, J = 7 Hz, 1H), 9.4 (d, J = cube like crystals of 2{[Sm(dpp )3]·0.5(H2O)}, were 7 Hz, 1H), 9.3 (d, J = 7 Hz, 1H), 9.2 (d, J = 7 Hz, 1H), 8.5 (d, J = grown in the dark by slow diﬀusion of pentane into a refriger- CO2Me,Me 7 Hz, 1H), 8.5–8.4 (m, 1.5H, Eu(dpp)3 + Hdpp), 6.8 (t, J = 7 Hz, ated solution of [Sm(dpp )3] in benzene. The crystal

1H), 6.7 (t, J = 7 Hz, 1H), 6.6–6.4 (m, 5H, Eu(dpp)3 + Hdpp), with dimensions of 0.023 × 0.02 × 0.02 mm was coated in

6.3 (t, J = 7 Hz, 1H), 6.2 (t, J = 7 Hz, 1H), 1.8 (s, 3H, CO2Me), immersion oil type NVH and transferred to the goniometer 1.8 (s, 3H, CO2Me), 1.7 (s, 3H, CO2Me), 1.6 (s, 3H, CO2Me), under the flow of the nitrogen cold stream (100 K). Diﬀraction

1.4–1.1 (m, 5H), 0.6 (d, J = 6 Hz, 1H), 0.5 (J = 6 Hz, Eu(dpp)3 + measurements were carried out using Si <111> monochro- − − λ H2O), 0.3 (d, J = 6 Hz, Eu(dpp)3 +H2O), 2.5 (s, 3H, Me), 2.6 mated synchrotron X-ray radiation ( = 0.71023 Å) (MX1 (s, 6H, Me), −2.7 (s, 3H, Me). FT-IR: ν = 2943.0 brw, 1682.0 m, Beamline at Australian Synchrotron).38 Data collection was 1589.2 s, 1511.6 m, 1433.0 s, 1357.8 m, 1288.8 m, 1208.6 w, carried out using BluIce39 suite of software and unit cell refine- − 1148.2 s, 1071.8 m, 973.4 w, 783.7 m, 745.0 m, 702.6 w cm 1. ment, data reduction and processing was carried out with

Elemental analysis calc. for C51H42N9O6Eu·0.5(H2O): C, 59.02; program XDS. The structure was solved using dual space H, 4.18; and N 12.15%; found: C, 59.51; H 4.23; and N, methods with SHELXT.40 The full-matrix least-square refine- 11.63%. Crystals suitable for SC-XRD anaylsis were grown in ments were carried out using SHELXL-2015 41 through Olex2 42 the dark by slow diﬀusion of pentane into a refrigerated solu- suite of software. The non-hydrogen atoms were refined aniso- CO2Me,Me tion of [Eu(dpp )3] in benzene. tropically. Half a molecule of water was identified from the CO2Me,Me 1 ﬀ [Gd(dpp )3]. Yield: 33.1 mg, 40%. H NMR (400 MHz, di erence map and was included in the least-squares refine- C6D6) δ 3.9 (vbrs). FT-IR: ˜ν = 2943.7 brw, 1682.2 m, 1589.1 s, ment. One of the ester groups has disorder in two positions, 1511.0 m, 1432.8 s, 1358.9 m, 1288.9 m, 1208.8 w, 1147.9 s, which was defined as part 1 and part 2 using EADP. They were − 1071.6 s, 973.3 m, 876.3 w, 783.5 s, 744.8 s, 701.2 m cm 1. restrained to have similar ADPs (using RIGU) and similar bond

Elemental analysis calc. for C51H42N9O6Gd·0.5(H2O)·0.5(C6H6): lengths (using DFIX). CO2Me,Me C, 59.93; H, 4.28; and N 11.65%; found: C, 59.57; H 3.91; and [Eu(dpp )3]·0.5(H2O) (CCDC 1904930†). Yellow crys- CO2Me,Me N, 11.56%. talline plates of [Eu(dpp )3]·0.5(H2O) were grown in the CO2Me,Me 1 [Yb(dpp )3]. Yield: 21.9 mg, 30%. H NMR (400 MHz, dark by slow diﬀusion of pentane into a refrigerated solution CO2Me,Me C6D6) δ 27.5 (brs, 1H), 26.2 (brs, 1H), 26.0–25.4 (brs, 4H), 24.4 of [Eu(dpp )3] in benzene. A suitable single crystal, (brs, 1H), 23.0 (brs, 1H), 9.9–9.5 (m, 3H), 9.2 (s, 1H), 8.9 (s, measuring of 0.1 × 0.04 × 0.01 mm, was selected under 1H), −3.9 (s, 3H), −4.2 (s, 3H), −4.3 (s, 3H), −4.5 (s, 3H), the polarizing microscope (Leica M165Z), mounted on a −9.1–−9.8 (m, 4H), −10.0 (brs, 1H), −10.1 (brs, 1H), −10.8 MicroMount (MiTeGen, USA) consisting of a thin polymer

Published on 05 June 2019. Downloaded by UNSW Library 6/11/2019 12:14:40 AM. (brs, 1H), −11.2 (brs, 1H), −16.0 (s, 3H), −16.3 (s, 3H), −16.6 tip with a wicking aperture. The single crystal was mounted (s, 3H), −16.6 (s, 3H). FT-IR: ν = 3428 w, 1687.7 m, 1585.1 m, on the goniometer using cryo loops for intensity measure- 1510.3 w, 1435.0 s, 1364.9 m, 1292.6 s, 1153.1 m, 1073.8 m, ments, was coated with paraﬃnoilandthenquicklytrans- − 983.2 w, 943.6 m, 787.3 s, 743.0 m, 670.7 s cm 1. ferred to the cold stream using an Oxford Cryo stream 800

(+)-HR-NSI-MS found: m/z 1051.2707 (calc. [C51H42N9O6Yb attachment. The X-ray diffraction measurements were (M) + H]+: m/z 1051.2719), m/z 1073.2522 (calc. [M + Na]+: m/z carried out on a Bruker D8 Quest Single Crystal diffract- 1073.2539). ometer with a Photon II detector at 100 K by using a IμS3.0 Microfocus Source with Mo-Kα radiation (λ = 0.710723 Å). X-Ray crystallography Symmetry related absorption corrections using the program Me,Me † 43 [La(dpp )3]·1.5(benzene) (CCDC 1904928 ). Yellow plate- SADABS were applied and the data were corrected for Me,Me like crystals of [La(dpp )3]·1.5(benzene), were grown by Lorentz and polarisation effects using Bruker APEX3 soft- Me,Me 43 slow diffusion of pentane into a solution of [La(dpp )3]in ware. The structure was solved using SHELXT (intrinsic benzene. The crystal with dimensions of 0.02 × 0.01 × 0.01 mm phasing)40 and the full-matrix least-square refinement was was coated in immersion oil type NVH and transferred to the carried out using SHELXL-2015 41 through the Olex2 42 suite goniometer under the flow of the nitrogen cold stream (100 K). of software. The non-hydrogen atoms were refined anisotro- Diffraction measurements were carried out using Si <111> pically. One of the ester groups has disorder over two posi- monochromated synchrotron X-ray radiation (λ = 0.71023 Å) tions, which was defined as part 1 and part 2 by using EADP. (MX1 Beamline at Australian Synchrotron).38 Data collection They were restrained to have similar ADPs (using RIGU) and 39 – – was carried out using the BluIce suite of software and unit similar bond lengths (using DFIX), while the C O CH3 bond cell refinement, data reduction and processing was carried out angle was also restrained (using DANG). The disordered qua- with program XDS. The structure was solved using dual space ternary carbon atoms were constrained to have the same methods with SHELXT.40 The full-matrix least-square refine- ADPs.

This journal is © The Royal Society of Chemistry 2019 Dalton Trans. James McPherson – June 2019 89 View Article Online Paper Chapter 4: Neutral Lanthanide Complexes Dalton Transactions Conﬂicts of interest 17J.C.Berthet,Y.Miquel,P.B.Iveson,M.Nierlich,P.Thuery, C. Madic and M. Ephritikhine, Dalton Trans., 2002, 3265–3272. There are no conflicts to declare. 18 E. G. Moore, M. Benaglia, G. Bergamini and P. Ceroni, Eur. J. Inorg. Chem., 2015, 2015, 414–420. 19 A. Ruiz-Martinez, D. Casanova and S. Alvarez, Chem. – Eur. J., 2008, 14, 1291–1303. Acknowledgements 20 J. N. McPherson, T. E. Elton and S. B. Colbran, Inorg. Chem., 2018, 57, 12312–12322. We thank the Australian Government (PhD scholarship to 21 M. Llunell, D. Casanova, J. Cirera, P. Alemany and J. N. M.), Australian Research Council (DP160104383) and S. Alvarez, SHAPE (2.1), Universitat de Barcelona, UNSW for supporting this research. Access to the Australian Barcelona, Spain, 2013. Synchrotron was possible by a Collaborative Access Program 22 A. McSkimming, V. Diachenko, R. London, K. Olrich, (MXCAP12368/13234). We gratefully acknowledge the Mark C. J. Onie, M. M. Bhadbhade, M. P. Bucknall, R. W. Read Wainwright Analytical Centre (MWAC) at UNSW (NMR, BMSF and S. B. Colbran, Chem. – Eur. J., 2014, 20, 11445–11456. and SSEAU), and Dr Remi Rouquette at the Chemical Analysis 23 J. C. G. Bünzli and S. V. Eliseeva, in Comprehensive Facility, Macquarie University. Inorganic Chemistry II: From Elements to Applications, ed. J. Reedijk and K. Poeppelmeier, Elsevier B.V., Amsterdam, 2nd edn, 2013, vol. 8, pp. 339–398. References 24 P. J. Wright, J. L. Kolanowski, W. K. Filipek, Z. Lim, E. G. Moore, S. Stagni, E. J. New and M. Massi, Eur. J. Inorg. 1 S. Cotton, Lanthanide and Actinide Chemistry: Electronic and Chem., 2017, 2017, 5260–5270. Magnetic Properties of the Lanthanides, John Wiley & Sons, 25 K. Binnemans, Coord. Chem. Rev., 2015, 295,1–45. West Sussex, U.K., 2006, pp. 61–87. 26 F. J. Steemers, W. Verboom, D. N. Reinhoudt, 2 J. C. G. Bunzli, Coord. Chem. Rev., 2015, 293,19–47. E. B. Vandertol and J. W. Verhoeven, J. Am. Chem. Soc., 3 V. Bulach, F. Sguerra and M. W. Hosseini, Coord. Chem. 1995, 117, 9408–9414. Rev., 2012, 256, 1468–1478. 27 W. D. Horrocks, J. P. Bolender, W. D. Smith and 4 S. Pandya, J. Yu and D. Parker, Dalton Trans., 2006, 2757– R. M. Supkowski, J. Am. Chem. Soc., 1997, 119, 5972–5973. 2766. 28 J. R. Lakowicz, Principles of Fluorescence Spectroscopy, 5 S. J. Butler and D. Parker, Chem. Soc. Rev., 2013, 42, 1652– Springer US, 3rd edn, 2006. 1666. 29 F. Wurthner, T. E. Kaiser and C. R. Saha-Moller, Angew. 6 D. Kovacs, X. Lu, L. S. Meszaros, M. Ott, J. Andres and Chem., Int. Ed., 2011, 50, 3376–3410. K. E. Borbas, J. Am. Chem. Soc., 2017, 139, 5756–5767. 30 J.-C. G. Bünzli and S. V. Eliseeva, Basics of lanthanide 7 I. Martinic, S. V. Eliseeva, T. N. Nguyen, F. Foucher, photophysics, in Lanthanide luminescence: photophysical, D. Gosset, F. Westall, V. L. Pecoraro and S. Petoud, Chem. analytical and biological aspects, ed. P. Hänninen and

Published on 05 June 2019. Downloaded by UNSW Library 6/11/2019 12:14:40 AM. Sci., 2017, 8, 6042–6050. H. Härmä, Springer series on fluorescence, Springer, 8 S. I. Weissman, J. Chem. Phys., 1942, 10, 214–217. Berlin, 2010, Ch. 1, vol. 7, pp. 1–45. 9 M. J. Beltran-Leiva, P. Cantero-Lopez, C. Zuniga, 31 A. de Bettencourt-Dias, Dalton Trans., 2007, 2229–2241. A. Bulhoes-Figueira, D. Paez-Hernandez and R. Arratia- 32 M. Más-Montoya and R. A. J. Janssen, Adv. Funct. Mater., Perez, Inorg. Chem., 2017, 56, 9200–9208. 2017, 27, 1605779. 10 F. Braun, P. Comba, L. Grimm, D.-P. Herten, B. Pokrandt 33 A. McSkimming, S. Shrestha, M. M. Bhadbhade, and H. Wadepohl, Inorg. Chim. Acta, 2019, 484, 464–468. P. Thordarson and S. B. Colbran, Chem. – Asian J., 2014, 9, 11 Y. Hirai, Assembled Lanthanide Complexes with Advanced 136–145. Photophysical Properties: General Introduction, Springer, 34 B. Das, J. N. McPherson and S. B. Colbran, Coord. Chem. Singapore, 2018, pp. 1–14. Rev., 2018, 363,29–56. 12 D. M. Roitershtein, L. N. Puntus, A. A. Vinogradov, 35 G. R. Fulmer, A. J. M. Miller, N. H. Sherden, H. E. Gottlieb, K. A. Lyssenko, M. E. Minyaev, M. D. Dobrokhodov, A. Nudelman, B. M. Stoltz, J. E. Bercaw and K. I. Goldberg, I. V. Taidakov, E. A. Varaksina, A. V. Churakov and Organometallics, 2010, 29, 2176–2179. I. E. Nifant’ev, Inorg. Chem., 2018, 57, 10199–10213. 36 S. I. Klink, G. A. Hebbink, L. Grave, P. G. B. Oude Alink, 13 J. N. McPherson, B. Das and S. B. Colbran, Coord. Chem. F. C. J. M. van Veggel and M. H. V. Werts, J. Phys. Chem. A, Rev., 2018, 375, 285–332. 2002, 106, 3681–3689. 14 E. C. Constable, in Adv. Inorg. Chem, ed. H. J. Emeléus and 37 N. M. Shavaleev, S. V. Eliseeva, R. Scopelliti and J.-C. A. G. Sharpe, Academic Press, 1986, vol. 30, pp. 69–121. G. Bünzli, Inorg. Chem., 2014, 53, 5171–5178. 15 C. Mallet, R. P. Thummel and C. Hery, Inorg. Chim. Acta, 38 N. P. Cowieson, D. Aragao, M. Clift, D. J. Ericsson, C. Gee, 1993, 210, 223–231. S. J. Harrop, N. Mudie, S. Panjikar, J. R. Price, A. Riboldi- 16 L. I. Semenova, A. N. Sobolev, B. W. Skelton and Tunnicliﬀe, R. Williamson and T. Caradoc-Davies, A. H. White, Aust. J. Chem., 1999, 52, 519–530. J. Synchrotron Radiat., 2015, 22, 187–190.

Dalton Trans. This journal is © The Royal Society of Chemistry 2019 James McPherson – June 2019 90 View Article Online Dalton Transactions Chapter 4: Neutral Lanthanide Complexes Paper

39 T. M. McPhillips, S. E. McPhillips, H. J. Chiu, A. E. Cohen, 41 G. M. Sheldrick, Acta Crystallogr., Sect. C: Struct. Chem., A. M. Deacon, P. J. Ellis, E. Garman, A. Gonzalez, 2015, 71,3–8. N. K. Sauter, R. P. Phizackerley, S. M. Soltis and P. Kuhn, 42 O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. K. Howard J. Synchrotron Radiat., 2002, 9, 401–406. and H. Puschmann, J. Appl. Crystallogr., 2009, 42, 339–341. 40 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Adv., 2015, 43 Bruker, APEX3 and SADABS, Bruker AXS Inc., Madison, 71,3–8. Wisconsin, USA, 2016. Published on 05 June 2019. Downloaded by UNSW Library 6/11/2019 12:14:40 AM.

This journal is © The Royal Society of Chemistry 2019 Dalton Trans. James McPherson – June 2019 91 Chapter 4: Neutral Lanthanide Complexes

James McPherson – June 2019 92 Chapter 5: Hemilabile Ruthenium(II) Catalysts

5 HEMILABILE RUTHENIUM(II) CATALYSTS

5.1 Introduction

“Pincer ligands” (meridional tridentate ligands) have gained much attention recently,

particularly in the field of homogenous molecular catalysis.1-4 This attention is due not only to

the high thermal stability typical for pincer-metal complexes, but also to the electronic and

steric influences of the pincer on bound transition metal ions, which can be used to modulate

and fine tune reaction outcomes. Inner sphere catalytic pathways involve binding and activation

of substrates at metal centres, and so molecular transition metal catalyst design often involves

a “stable” pincer-metal unit that is either coordinatively unsaturated or bound at the active

site(s) by substitutionally labile co-ligands, typically halides or solvent molecules. Another

strategy to access vacant sites at metal centres has been to incorporate weak donors in

multidentate ligands, resulting in hemilability,5 or the reversible release of one donor group

without irreversible ligand dissociation.

The prototypal pincer 2,2′:6′,2″-terpyridine (tpy) has rich and varied coordination

chemistry, and has been regularly exploited to construct supramolecular assemblies containing

M(tpy) units, with well understood function and geometry.6-9 As discussed in Chapter 2, tpy is

preorganised to bind transition metal ions.10 For example, the ruthenium(II) complex

11 [Ru(tpy)2][ClO4]2 is essentially inert to ligand substitution (i.e., is coordinatively saturated),

with only 25 kJ mol–1 coordination strain in each bound ligand.10 While these homoleptic

2+ [Ru(tpy)2] complexes have been of much interest for their photophysical and redox properties,

they are thus unsuitable for catalytic applications that operate through inner sphere pathways.

James McPherson – June 2019 93 Chapter 5: Hemilabile Ruthenium(II) Catalysts

The electronic and geometric properties of 2,5-di(2-pyridyl)pyrrolides (dpp–) complement the rich coordination chemistry known for tpy.12 In contrast to tpy, the electron- rich pyrrolide ring generally stabilises higher oxidation states of bound metal ions. Significantly more strain (up to 90 kJ mol–1) was estimated10 in each (dppCO2Me,Me)– ligand from the solid

CO Me,Me 13 2+ state structure of Colbran’s [Ru(dpp 2 )2] complex, than the strain in tpy for [Ru(tpy)2]

(see Figure 5.1, and as discussed in Chapter 2).

Figure 5.1: Generic depictions of metal-complexes of the prototypal NNN pincer, tpy, on the left and the focus of this work, the dpp– anion, on the right. The strain energies calculated for examples of these ligands found in the literature are included,10 with the ideal geometry for interconnected six- and five-membered rings shadowed behind each illustration. The electron density at the metal-centre can be modulated through variation in the R group substituents as indicated.

The research described in this chapter investigated the effect of strain within dpp– ligands bound in homoleptic ruthenium(II) complexes, [Ru(dpp)2]. Hemilability was predicted and has been exploited for the reversible binding, and catalytic transformations, of small molecules.

5.2 Results and Discussion

5.2.1 Synthesis and Characterisation

Colbran’s previously reported one-pot Hdpp condensation protocol was employed to access 2,5-di(2-pyridyl)-3-methyl carboxylate-4-methylpyrrole (HdppCO2Me,Me),13 while Hein’s original synthesis of 2,5-di(2-pyridyl)-3,4-di(ethylcarboxylate)pyrrole (HdppCO2Et,CO2Et)14-15

R1,R2 was a more practical method for the symmetric ligand precursor. Both [Ru(dpp )2]

James McPherson – June 2019 94 Chapter 5: Hemilabile Ruthenium(II) Catalysts

1 2 13 complexes (R , R = CO2Me, Me; or CO2Et, CO2Et) were prepared as previously reported with the appropriate HdppR1,R2 precursor (Scheme 5.1). Solution state characterisation of

CO Et,CO Et [Ru(dpp 2 2 )2] was by NMR and UV-vis spectroscopy, high resolution mass spectrometry

CO Me,Me and cyclic voltammetry, with analogous behaviours observed for [Ru(dpp 2 )2] (see

Appendix 4).

R1,R2 13 Scheme 5.1: Synthesis of [Ru(dpp )2] complexes described in this work.

CO Et,CO Et In the mass spectrum of [Ru(dpp 2 2 )2], a peak envelope centred at m/z 830.1635

CO Et,CO Et + corresponds to the singly charged species, [Ru(dpp 2 2 )2] (calc. m/z 830.1633). The experimental and the simulated isotopic distributions were in good agreement, as is shown in

Figure 5.2.

Figure 5.2: The observed peaks from the high resolution mass spectrum of the crude reaction mixture from the CO Et,CO Et synthesis of [Ru(dpp 2 2 )2] (top) versus the simulated isotopic distribution of peaks for the singly charged

CO Et,CO Et + species [Ru(dpp 2 2 )2] (bottom).

James McPherson – June 2019 95 Chapter 5: Hemilabile Ruthenium(II) Catalysts

Only one set of signals for pyridyl (HB3–HB6) and ethyl ester (HA3″ and HA3′′′) 1H

1 CO Et,CO Et environments were evident in the H NMR spectrum of [Ru(dpp 2 2 )2] in benzene-d6

(Figure 5.3, top), consistent with S4 symmetry about the ruthenium(II) centre. Assignment of the signals associated with the ethyl ester pyrrolide substituents (HA3″ and HA3′′′) was trivial from inspection of the relative integrals and splitting patterns due to scalar coupling. The four pyridyl signals HB3–HB6 were assigned on the basis of the 3J coupling constants (see Table 5.1), which have been shown in thousands of other polypyridyl complexes of d6 ruthenium(II), iron(II) and

3 cobalt(III) ions to closely resemble those in the free ligands, and with J5,6 always significantly

3 16 lower than J3,4.

Table 5.1: Chemical shifts (δ) and coupling constants (J /Hz) for the signals assigned to the pyridyl 1H B3 B6 1 CO Et,CO Et environments (H –H ) in the 300 MHz H NMR spectrum of [Ru(dpp 2 2 )2] in benzene-d6. In bold are the major 3J coupling constants which aided assignment.

1 H Assignment: δ J1 /Hz J2 /Hz J3 /Hz HB3 8.86 8.2 1.5 0.8 HB4 6.61 8.2 7.3 1.6 HB5 5.59 7.3 5.5 1.5 HB6 6.95 5.5 1.6 0.8

These assignments, and the assignment of the 101 MHz 13C{1H} NMR spectrum of

CO Et,CO Et [Ru(dpp 2 2 )2] in benzene-d6 (middle panel of Figure 5.3) were confirmed using conventional 2-D NMR techniques, with the key 1H-13C HSQC and HMBC NMR correlations annotated in the spectra presented in the bottom panel of Figure 5.3.

James McPherson – June 2019 96 Chapter 5: Hemilabile Ruthenium(II) Catalysts

1 13 1 CO Et,CO Et Figure 5.3: Assignment of the H and C{ H} NMR spectra of [Ru(dpp 2 2 )2] in benzene-d6. Top panel: the

1 CO Et,CO Et 300 MHz H NMR spectrum of [Ru(dpp 2 2 )2] in benzene-d6, with numbering scheme and assignments

13 1 CO Et,CO Et indicated. Middle panel: the assigned 101 MHz C{ H} NMR spectrum of [Ru(dpp 2 2 )2] in benzene-d6. Bottom panels: the 1H-13C HSQC (left) and HMBC (right) NMR spectra (at 400 MHz and 101 MHz for 1H and 13 CO Et,CO Et C respectively) of [Ru(dpp 2 2 )2] in benzene-d6 showing the key correlations used for the assignment of the 1D spectra above.

James McPherson – June 2019 97 Chapter 5: Hemilabile Ruthenium(II) Catalysts

5.2.2 Electrochemistry

CO Et,CO Et The redox behaviour of [Ru(dpp 2 2 )2] was ascertained by cyclic voltammetry experiments conducted in dichloromethane using tetrabutylammonium hexafluorophosphate

(0.1 M) as the supporting electrolyte. As is shown in Figure 5.4, the potential of the RuIII/II redox

+/0 CO Et,CO Et couple was shifted to a higher potential by ~200 mV {–0.15 V vs. Fc for [Ru(dpp 2 2 )2]

+/0 CO Me,Me compared to –0.36 V vs. Fc for the more electron rich [Ru(dpp 2 )2]}. The sensitivity of the metal centred redox potential to the pyrrolide substituents is similar to that seen for the

17 – [Co(dpp)2] systems discussed in Chapter 3. Oxidation of the dpp ligand was not observed

CO Et,CO Et within the solvent discharge limits for complex [Ru(dpp 2 2 )2].

CO Et,CO Et CO Me,Me Figure 5.4: Cyclic voltammograms of [Ru(dpp 2 2 )2] (blue) and [Ru(dpp 2 )2] (black) recorded in dichloromethane with tetrabutylammonium hexafluorophosphate (0.1 M) supporting electrolyte at 100 mV s–1, a 1 mm glassy carbon disk working electrode, and 298 K under an atmosphere of dinitrogen.

5.2.3 Reversible Solvent Binding

1 CO Et,CO Et New species appear in the H NMR spectra when [Ru(dpp 2 2 )2] is dissolved in more coordinating solvents such as acetonitrile-d3 or DMSO-d6 as shown in Figure 5.5 below.

James McPherson – June 2019 98 Chapter 5: Hemilabile Ruthenium(II) Catalysts

1 CO Et,CO Et Figure 5.5: Stacked 400 MHz H NMR spectra of [Ru(dpp 2 2 )2] in solvents as labelled.

1 15 The release of pyridyl donors in acetonitrile-d3 was confirmed by H- N HMBC NMR experiments (acquired at 600 MHz and 61 MHz for 1H and 15N, respectively, and at 298 K).

The 1H-15N HMBC NMR spectrum, panel (c) in Figure 5.6, reveals two overlapping correlations between the HB6 signal and a 15N peak with chemical shift (300 < δ < 350 ppm) typical for a pyridyl N-atom with a non-bonding pair of electrons.18 The remaining four signals assigned to different HB6 environments all correlate with 15N peaks having chemical shifts (at

δ ≈ 250 ppm) typical for pyridyl N-atoms that are coordinated to transition metals or protons.18

The binding of solvent is reversible: when the acetonitrile-d3 solvent was removed and the residue dissolved in benzene-d6, only the signals attributed to the symmetric

CO Et,CO Et [Ru(dpp 2 2 )2] species were observed (panel (d) in Figure 5.6).

James McPherson – June 2019 99 Chapter 5: Hemilabile Ruthenium(II) Catalysts

CO Et,CO Et 1 Figure 5.6: Reversible solvent binding at ruthenium by [Ru(dpp 2 2 )2] in coordinating solvents. Stacked H

CO Et,CO Et B6 NMR spectra of [Ru(dpp 2 2 )2], with signals assigned to H environments marked by triangles: (a) in

1 15 benzene-d6 at 400 MHz; (b) acetonitrile-d3 at 600 MHz; (c) the H- N HMBC NMR spectrum of CO Et,CO Et 1 15 [Ru(dpp 2 2 )2] in acetonitrile-d3 (at 600 MHz and 61 MHz for H and N, respectively); (d) after the removal of acetonitrile-d3 from sample (b/c), re-dissolved in benezene-d6 at 400 MHz.

Between 0.95 and 1.5 ppm, a total of six triplets, each assigned to unique HA3′′′ environments, were sufficiently resolved such that the peaks did not overlap one another in the

1 600 MHz H NMR spectrum acquired in acetonitrile-d3 as shown in Figure 5.7. In addition to the single peak at δ 1.44 (black in Figure 5.7) assigned to the S4 symmetric parent complex,

CO Et,CO Et [Ru(dpp 2 2 )2], three signals are consistent with the formation of a mono-solvato

CO Et,CO Et [Ru(dpp 2 2 )2(MeCN)] complex (blue in Figure 5.7) with three unique environments in a

2:1:1 ratio due to the Cs symmetry.

1 CO Et,CO Et Figure 5.7: The 600 MHz H NMR spectrum of [Ru(dpp 2 2 )2(MeCN)x] (x = 0 in black, 1 in blue, and 2 in red) in acetonitrile-d3 showing the signals corresponding to six unique methyl environments.

James McPherson – June 2019 100 Chapter 5: Hemilabile Ruthenium(II) Catalysts

The remaining two peaks, in a 1:1 ratio relative to each other and shown in maroon in

CO Et,CO Et Figure 5.7, were assigned to a single bis-solvato species, [Ru(dpp 2 2 )2(MeCN)2].

Although numerous geometric isomers of the bis-solvato species are possible, only formation of any one of the four isomers with two-fold symmetry shown in Figure 5.8 is consistent with

1 the number of environments observed in the H NMR spectrum. From these, either the C2h

1 trans-isomer or C2 cis-racemate (the ∆ and Λ propellers are indistinguishable by H NMR spectroscopy) appear the most likely, as they have the minimum steric congestion and require

CO Et,CO Et least reorganisation to form from the mono-solvato complex, [Ru(dpp 2 2 )2(MeCN)].

CO Et,CO Et Figure 5.8: Geometric isomers with two-fold symmetry of the bis-solvato complex, [Ru(dpp 2 2 )2(L)2] (L = MeCN), coloured by symmetry equivalence. The IUPAC recommended configuration indices for the octahedral geometries are also listed, with the Cahn-Ingold-Prelog priority sequence Npy > Npyr > NMeCN.

5.2.3.1 Vacuum Phase Molecular Geometry

2 The molecular geometries of the possible bis-solvato complexes, [Ru(κ -dpp)2(MeCN)2], were thus investigated by DFT calculations. The electronic effects of –CO2R are all very similar for R = H, Me or Et, with identical Swain-Lupton F and R terms (0.34 and –0.11,

19 2 CO H,CO H respectively), and so the geometries of [Ru(κ -dpp 2 2 )2(MeCN)2] were calculated to minimise computational cost. The cartesian coordinates from the previously reported13 SC-

CO Me,Me XRD structure of [Ru(dpp 2 )2] were used to generate the initial coordinates for the three

2 CO H,CO H [Ru(κ -dpp 2 2 )2(MeCN)2] isomers shown in Figure 5.8, which were then optimised in

James McPherson – June 2019 101 Chapter 5: Hemilabile Ruthenium(II) Catalysts vacuo by DFT with Grimme’s empirical dispersion correction (BP86-GD3) using the split valence def2-SVP basis set.

2 CO H,CO H These calculations confirm the cis-[Ru(κ -dpp 2 2 )2(MeCN)2] as the most energetically favoured configuration for the bis-solvato complex, by more than 35 kJ mol–1 as shown in Figure 5.9.

Figure 5.9: Ball-and-stick views and the relative energies of the optimised geometries (BP86-GD3/def2-SVP) of

2 CO H,CO H the bis-solvato complexes, [Ru(κ -dpp 2 2 )2(MeCN)2], with C-bound H-atoms omitted for clarity, and C- atoms in acetonitrile ligands coloured black for clarity.

2 With the cis-[Ru(κ -dpp)2(MeCN)2] thus confirmed as the preferred configuration for the

CO Et,CO Et bis-solvato complexes, the molecular geometries of [Ru(dpp 2 2 )2(MeCN)x] (x = 0, 1 or

2) were calculated and optimised in vacuo, with the ethyl fragment restored to account for possible steric influences, using the larger, triple-ζ def2-TZVP basis set.

James McPherson – June 2019 102 Chapter 5: Hemilabile Ruthenium(II) Catalysts

CO Et,CO Et The optimised geometry of [Ru(dpp 2 2 )2], presented in Figure 5.10 and with key metric parameters summarised in Table 5.2, was in good agreement with the previously

13 CO Me,Me reported solid-state geometry of [Ru(dpp 2 )2] obtained from SC-XRD data. The six N- donor atoms form a distorted octahedral coordination geometry about the ruthenium(II) centre,

20 with quadratic elongation parameter, 〈λOh〉, of 1.040, with the equatorial Ru–Npy bonds ~ 0.14

Å longer than the two Ru–Npyr bonds. The two terminal pyridyl N-donors of each

CO Et,CO Et – (dpp 2 2 ) ligand are separated (ρL(coord), see Chapter 2) by ~4.05 Å, and as ρL(relax) = 4.969

Å for (dppCO2Et,CO2Et)– was calculated in Chapter 3, equation 5.1 can be derived (see Chapter 2) to thus predict ~85 kJ mol–1 of strain within each (dppCO2Et,CO2Et)– ligand in this

CO Et,CO Et [Ru(dpp 2 2 )2] complex.

| ( ) . Å| ( ) = (460 ± 37) kJ mol (eq. 5.1) 𝐿𝐿 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐. Å 𝜌𝜌 −4 969 −1 𝐿𝐿 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 4 969 In vacuo∆𝐸𝐸, the octahedral coordination spheres of the solvato complexes,

CO Et,CO Et [Ru(dpp 2 2 )2(MeCN)x] (x = 1 or 2; Figure 5.10 and Table 5.2), were predicted to be less distorted (with 〈λOh〉 = 1.0257 and 1.0155 for x = 1 and 2, respectively) than the unsolvated

CO Et,CO Et parent complex, [Ru(dpp 2 2 )2]. For the solvato complexes with x = 1 and 2, the

Npyr–Ru–Npyr′ angles (170.3º and 167.6º, respectively) deviated from the ideal (180º) whereas

CO Et,CO Et it did not in [Ru(dpp 2 2 )2] (179.5º). The unbound pyridyl groups in the

κ2-(dppCO2Et,CO2Et)– ligands twist away from the ruthenium(II) centre, with inter ring torsions

(N–C–C–N) between 46.5 and 57.0º along the C–C bond connecting the pyrrolide and unbound pyridyl rings, but nevertheless provide steric bulk which elongates the Ru–Npyr bond to these

CO Et,CO Et bidentate ligands by 0.13 Å from 1.94 Å in [Ru(dpp 2 2 )2].

James McPherson – June 2019 103 Chapter 5: Hemilabile Ruthenium(II) Catalysts

Figure 5.10: Ball-and-stick views of the optimised geometries (BP86-GD3/def2-TZVP) of CO Et,CO Et [Ru(dpp 2 2 )2(MeCN)x] (x = 0, 1 or 2), with hydrogen atoms omitted and carbon atoms in acetonitrile ligand

CO Et,CO Et coloured black for clarity. The left view is of the parent molecule, [Ru(dpp 2 2 )2]. In

CO Et,CO Et [Ru(dpp 2 2 )2(MeCN)] (centre view), the inter-ring Npy–C–C–Npyr torsion for the unbound pyridyl group is

CO Et,CO Et 46.5º. In cis-[Ru(dpp 2 2 )2(MeCN)2] (right view), the transoid Npy–Ru–NMeCN angles are 176.9º and 177.1º and the inter-ring Npy–C–C–Npyr torsions for the unbound pyridyl group are 53.6º and 57.0º.

Table 5.2: Key metric data from the optimised (BP86-GD3/def2-TZVP) geometries of CO Et,CO Et [Ru(dpp 2 2 )2(MeCN)x] (x = 0, 1 or 2), in vacuo, and from the analogous solid-state (ss) SC-XRD structure

CO Me,Me 13 of [Ru(dpp 2 )2] reported previously by the Colbran group. The ∆EL(strain) parameter was predicted by equation 5.1 from ρL(coord) for each ligand as described in Chapter 2, and 〈λOh〉 is Robinson’s quadratic elongation parameter.20

CO Me,Me ref 13 CO Et,CO Et CO Et,CO Et CO Et,CO Et ss-[Ru(dpp 2 )2] [Ru(dpp 2 2 )2] [Ru(dpp 2 2 )2(MeCN)] cis-[Ru(dpp 2 2 )2(MeCN)2] κ3-dpp κ2-dpp Ru–Npyr /Å 1.953(4)–1.954(4) 1.9414–1.9415 1.9252 2.0874 2.0557–2.0650 Ru–Npy /Å 2.090(4)–2.107(4) 2.0823–2.0854 2.0949–2.0997 2.0615 2.0666–2.0685 Ru–NMeCN /Å - - 1.962 1.9649–1.9671 Npy–Ru–Npy′ /º 151.9(1)–152.7(2) 153.13–153.14 154.65 - - Npy–Ru–Npyr /º 75.7(1)–76.8(2) 76.52–76.62 77.27–77.41 77.45 78.01–78.50 Npyr–Ru–Npyr′ /º 175.4(2) 179.45 170.32 167.59 ρL(coord) /Å 4.066(6)–4.077(5) 4.0529–4.0537 4.0923 - - –1 ∆EL(strain) /kJ mol 75 85 80 - - 〈λOh〉 1.0435 1.0400 1.0257 1.0155

CO Et,CO Et On binding a molecule of acetonitrile to form [Ru(dpp 2 2 )2(MeCN)x] (x = 1 or 2), the strain within each (dppCO2Et,CO2Et)– ligand is relieved with the release of a pyridyl ring. The

3 CO Et,CO Et – CO Et,CO Et two Ru–Npy bonds in the tridentate κ -(dpp 2 2 ) ligand in [Ru(dpp 2 2 )2(MeCN)]

James McPherson – June 2019 104 Chapter 5: Hemilabile Ruthenium(II) Catalysts

–1 elongate by 0.02 Å, and so ρL(coord) increases to 4.09 Å: a relaxation of ~5 kJ mol by equation

CO Et,CO Et 5.1 compared with the ligands in [Ru(dpp 2 2 )2]. Equation 5.1 is not appropriate for the

2 CO Et,CO Et – CO Et,CO Et bidentate κ -(dpp 2 2 ) ligands in [Ru(dpp 2 2 )2(MeCN)x] (x = 1 or 2). The cartesian coordinates for each bound ligand were therefore harvested from the DFT optimised geometries, and single point energy calculations were performed at the same level of theory

(B3LYP-GD3BJ/6-31G**) as discussed in Chapter 2.10 The relative energies of each ligand conformation are shown in Figure 5.11, with the average energy of the two tridentate ligands

CO Et,CO Et in the [Ru(dpp 2 2 )2] complex as the reference, and this analysis predicts that the release of a pyridyl ring from the ruthenium centre results in a relaxation of ca. 60 kJ mol–1 strain within the dpp– ligand.

Figure 5.11: Relative energies (B3LYP-GD3BJ/6-31G**) of the different (dppCO2Et,CO2Et)– ligand conformations

CO Et,CO Et from the optimised geometries of [Ru(dpp 2 2 )2(MeCN)x] (x = 0 in black, 1 in blue, or 2 in red). The average

CO Et,CO Et energy of the two strained ligands in the structure of [Ru(dpp 2 2 )2] was chosen as the reference.

0 5.2.3.2 Equilibrium compositions, binding constants, and ∆G in acetonitrile-d3

CO Et,CO Et The relative concentration of each [Ru(dpp 2 2 )2(MeCN)x] complex (x = 0, 1 or 2) at a given temperature (T in K), χx, T, is given by equation 5.2:

James McPherson – June 2019 105 Chapter 5: Hemilabile Ruthenium(II) Catalysts

χx, T = ΣIx, T / Itotal, T (eq. 5.2)

The binding of acetonitrile at ruthenium is reversible (see Figure 5.6), and so can be considered as the two sequential equilibria shown in Scheme 5.2.

Scheme 5.2: Binding of acetonitrile solvent by [Ru(dpp)2].

The equilibrium (binding) constants (Kx, x = 1 or 2, see Scheme 5.1) and free energies

–1 (∆xG/kJ mol ) are given by equations 5.3 and 5.4 respectively—where R is the ideal gas constant and T is the temperature in K:

Kx = χx / χ(x–1), (eq. 5.3)

0 ∆xG = – RT ln(K) (eq. 5.4)

CO Et,CO Et The equilibrium compositions of each [Ru(dpp 2 2 )2(MeCN)x] complex (x = 0, 1

1 or 2) in acetonitrile–d3 were measured between the range of 288–338 K by H NMR at 400

MHz, Figure 5.12.

James McPherson – June 2019 106 Chapter 5: Hemilabile Ruthenium(II) Catalysts

1 CO Et,CO Et Figure 5.12: Stacked 400 MHz H NMR spectra of [Ru(dpp 2 2 )2(MeCN)x] in acetonitrile-d3 acquired at the

CO Et,CO Et temperatures indicated. The signals assigned to the methyl protons in [Ru(dpp 2 2 )2] are shown in black,

CO Et,CO Et CO Et,CO Et [Ru(dpp 2 2 )2(MeCN)] in blue and [Ru(dpp 2 2 )2(MeCN)2] in red. At 351 K, the signal-to-noise (I / σ) for the mono- and bis-solvato species was not sufficient for quantification.

From these, the standard free energies were calculated (Table 5.3), and plotted against T

0 0 (in K), Figure 5.13, to determine the thermodynamic parameters ∆xH and ∆xS (Table 5.4) according to equation 5.5:

0 0 0 ∆xG = ∆xH – T∆xS (eq. 5.5)

CO Et,CO Et Table 5.3: The thermal equilibria data for [Ru(dpp 2 2 )2(MeCN)x] (x = 0, 1 or 2) in acetonitrile-d3 from variable temperature, 400 MHz 1H NMR spectra.

288 K 298 K 313 K 328 K 338 K χ0 0.34 0.45 0.55 0.63 0.70 χ1 0.35 0.34 0.31 0.28 0.24 χ2 0.30 0.21 0.13 0.09 0.05 K1 1.0 0.77 0.56 0.45 0.35 0 –1 ∆1G / kJ mol –0.05 +0.66 +1.50 +2.18 +2.97 K2 0.87 0.61 0.43 0.33 0.22 0 –1 ∆2G / kJ mol +0.35 +1.23 +2.22 +3.04 +4.21

James McPherson – June 2019 107 Chapter 5: Hemilabile Ruthenium(II) Catalysts

1 Figure 5.13: Free energies vs. T for K1 and K2 from VT 400 MHz H NMR experiments in acetonitrile-d3.

Table 5.4: The thermodynamic parameters (∆H0 in kJ mol–1 and ∆S0 in J K–1 mol–1) for the reversible binding of CO Et,CO Et acetonitrile by [Ru(dpp 2 2 )2].

∆H0/kJ mol–1 ∆S0/J K–1 mol–1

K1: –15.8 ± 1.0 –55.2 ± 3.2

K2: –20.4 ± 1.3 –72.1 ± 4.1

These data reveal that there is a large entropic cost associated with release of a pyridyl ring to bind solvent molecules, with ∆S0 ≈ –55 J K–1 mol–1 for the first binding event, and an additional –70 J K–1 mol–1 for the second binding event, consistent with the chelate effect. A consequence of the temperature dependences of these equilibria, equations 5.4 and 5.5, is that

CO Et,CO Et the solvato complexes are less favoured than the parent [Ru(dpp 2 2 )2] at higher temperatures (as observed).

5.2.3.3 Behaviour in chloroform

CO Et,CO Et The behaviour of [Ru(dpp 2 2 )2] in chloroform-d was more complicated. In

CO Et,CO Et addition to the signals attributed to the symmetric [Ru(dpp 2 2 )2] complex (shown in black in Figure 5.14), numerous broad signals were also observed in the 1H NMR spectrum, which suggests a mixture of products in dynamic exchange on the timescale of the NMR

James McPherson – June 2019 108 Chapter 5: Hemilabile Ruthenium(II) Catalysts experiment. No evidence of dpp– ligand dissociation was observed in the 1H NMR spectrum as shown in Figure 5.14.

1 CO Et,CO Et Figure 5.14: Top: 300 MHz H NMR spectrum of [Ru(dpp 2 2 )2] in CDCl3 at 298 K, with the signals

CO Et,CO Et 1 attributed to a symmetric [Ru(dpp 2 2 )2] species shown in black. The 600 MHz H NMR spectrum of dppCO2Et,CO2EtH at 298 K is shown on the bottom as a reference to confirm its absence from the spectrum above.

To investigate whether the dynamic behaviour observed in chloroform was due to the presence of chloride ions, excess tetrabutylammonium chloride ([Bu4N]Cl) was added to

1 samples of [Ru(dpp)2] in acetone-d6 or acetonitrile-d3 and monitored by H NMR spectroscopy.

1 CO Me,Me No changes in the H NMR spectrum of [Ru(dpp 2 )2] in acetone-d6 were observed after the addition of a five-fold excess of chloride ion as [Bu4N]Cl (see Figure 5.15). Nor did the

CO Et,CO Et addition of [Bu4N]Cl affect the equilibrium distribution of [Ru(dpp 2 2 )2(MeCN)x] (x =

0, 1 or 2) species observed in acetonitrile-d3 (see Figure 5.16).

James McPherson – June 2019 109 Chapter 5: Hemilabile Ruthenium(II) Catalysts

1 CO Me,Me Figure 5.15: Stacked 400 MHz H NMR spectra of [Ru(dpp 2 )2] (black) in acetone-d6 before (top) and after

(bottom) the addition of 5 equivalents of [Bu4N]Cl (blue).

1 CO Et,CO Et Figure 5.16: Stacked H NMR spectra of [Ru(dpp 2 2 )2] in acetonitrile-d3 before (at 300 MHz, top) and following the addition of [Bu4N]Cl (at 400 MHz, bottom).

James McPherson – June 2019 110 Chapter 5: Hemilabile Ruthenium(II) Catalysts

Concluding that chloride ion (alone) was not responsible for the dynamic behaviour of

[Ru(dpp)2] observed in chloroform, attention was turned to the possible influence of acidic impurities present in chloroform.21 Two equivalents of acetic acid were added to a sample of

CO Et,CO Et 1 [Ru(dpp 2 2 )2] in benzene-d6, with no change in the H NMR spectrum of the ruthenium(II) complex observed, even after the solution was heated at reflux for twenty four hours, as shown in Figure 5.17.

1 CO Et,CO Et Figure 5.17: Stacked 300 MHz H NMR (298 K) spectra of [Ru(dpp 2 2 )2] (black) (top) and after a mixture

CO Et,CO Et of [Ru(dpp 2 2 )2] and two equivalents of acetic acid (red) in benzene-d6 was heated at reflux for 24 hours (bottom).

CO Me,Me On the other hand, the addition of HCl (12 M, aq) to a solution of [Ru(dpp 2 )2] also in benzene-d6 resulted only in a broadening of the signals attributed to the complex in the 400

MHz 1H NMR spectrum as shown in Figure 5.18, although the composition of the aqueous phase of the mixture was not examined.

James McPherson – June 2019 111 Chapter 5: Hemilabile Ruthenium(II) Catalysts

1 CO Me,Me Figure 5.18: Stacked 400 MHz H NMR spectra of [Ru(dpp 2 )2] in benzene-d6 acquired before (top) and after (bottom) the solution was washed with HCl (12 M, aq).

It is also possible that other decomposition products of chloroform (e.g., radical species,

21 1 Cl2, and/or phosgene), or simply chloroform itself, contribute to the complicated H NMR spectra of [Ru(dpp)2] complexes observed in chloroform-d, however these were not investigated directly.

5.2.4 Activation of Small Molecules for Catalysis

It was envisaged that the reversible opening of vacant sites in the [Ru(dpp)2] complexes could be exploited to bind and activate small molecules for catalysis. To further investigate the scope of substrates capable of binding these vacant sites at ruthenium(II), solutions of

1 [Ru(dpp)2] complexes were treated with various small molecules and monitored in situ by H

NMR spectroscopy. Description of these preliminary studies follow.

James McPherson – June 2019 112 Chapter 5: Hemilabile Ruthenium(II) Catalysts

Carbon monoxide (CO) is a strong σ-donor and π-acceptor ligand, and metal carbonyl complexes have important applications as synthetic precursors, catalysts and therapeutics. The

CO Me,Me [Ru(dpp 2 )2] complex was dissolved in benzene-d6 in an NMR tube, and CO (g) was bubbled through the sample for five minutes. No additional species were detected in the 1H

NMR spectrum of the complex acquired following the sparging of CO (g), presented as the middle trace in Figure 5.19, and no further attempts at CO binding were pursued.

1 CO Me,Me Figure 5.19: Stacked 400 MHz H NMR spectra of [Ru(dpp 2 )2] in benzene-d6 (peaks highlighted in black) before (top trace) and after the addition of various reagents (lower traces), with the signals assigned to CO Me,Me [Ru(dpp 2 )2pyx] species in the bottom trace highlighted in blue.

Free pyridine was an obvious candidate to access the vacant sites at ruthenium(II). It was of interest to determine whether there was sufficient steric clearance at the ruthenium centre for the planar aromatic pyridine to bind and relieve strain in the dpp– ligand. The addition of

CO Me,Me pyridine (in large excess) to a solution of [Ru(dpp 2 )2] in benzene-d6 resulted in new signals with low intensity in the 1H NMR spectrum (highlighted in blue in Figure 5.19), similar to the spectra observed when the complexes were dissolved in coordinating solvents

(DMSO-d6 or acetonitrile-d3) discussed above. However there was no distinguishable change

James McPherson – June 2019 113 Chapter 5: Hemilabile Ruthenium(II) Catalysts

CO Et,CO Et in the UV-vis spectrum of [Ru(dpp 2 2 )2] dissolved in THF following the addition of pyridine (see Figure 5.20).

CO Et,CO Et Figure 5.20: The UV-visible absorption spectra (in THF) of [Ru(dpp 2 2 )2] before (red) and after the addition of pyridine (in blue) at 61.5 µM (solid lines) and 6.15 µM (dashed lines).

Next, a bidentate chelate ligand, 2,2′-bipyridine (bpy), was investigated. Binding of such bidentate ligands should be less entropically disfavoured than the mono-dentate ligands previously discussed, and bpy is only capable of bidentate-binding of ruthenium(II) through two sites in a cis conformation. Assuming that access to these sites is via release of a pyridyl

– ring in the dpp ligand as observed in acetonitrile-d3, there are three pairs of ∆/Λ propellers

CO Me,Me CO Me,Me – possible from the parent complex, [Ru(dpp 2 )2], with the non-symmetric (dpp 2 ) ligand as shown in Figure 5.21. These isomers are not necessarily iso-energetic—for example, differences in the steric bulk of the methyl carboxylate substituent versus that of the methyl substituent might result in the selective formation of only one pair of C2 enantiomers.

James McPherson – June 2019 114 Chapter 5: Hemilabile Ruthenium(II) Catalysts

2 CO Me,Me Figure 5.21: Three pairs of ∆/Λ enantiomers of [Ru(κ -dpp 2 )2(bpy)] are possible from racemic

CO Me,Me [Ru(dpp 2 )2] as shown. For each pair of the C2 symmetric enantiomers, the ligand scaffolds are coloured by symmetry equivalence (for the non-symmetric C1 enantiomers, all pyrrolide and pyridyl environments are unique). CO Me,Me Beside each clockwise (C, top) and anticlockwise (A, bottom) [Ru(dpp 2 )2] enantiomer is shown the IUPAC recommended configuration indices for the octahedral geometries and a diagram depicting the Cahn-Ingold-Prelog (CIP) priority sequence (blue labels) for each donor atom.22

1 CO Me,Me No change was observed in the H NMR spectrum of the complex, [Ru(dpp 2 )2], in benzene-d6 for two hours after two equivalents of bpy were added. After heating the mixture at

70 ºC for an additional two hours, however, at least eight new methyl or methyl carboxylate signals were observed at low intensity in the 1H NMR spectrum, highlighted in blue in Figure

5.22. This number of new signals is inconsistent with selective formation of only one of the C2 symmetric enantiomeric pairs. From the sum of the relative integrals of these eight new signals

James McPherson – June 2019 115 Chapter 5: Hemilabile Ruthenium(II) Catalysts

(Σ Inew CO2Me/Me = 0.41 versus Σ Itotal CO2Me/Me = 4.38), less than 10% of the ruthenium complex formed bpy-adducts during this experiment.

1 CO Me,Me Figure 5.22: Stacked 400 MHz H NMR spectra (acquired at 298 K) of [Ru(dpp 2 )2] in benzene-d6 before (top spectrum) and after (middle spectrum) adding bpy (2 equiv.) and stirring for 2 hours at room temperature and CO Me,Me an additional 2 hours at 70 ºC. A mixture of [Ru(dpp 2 )2] (black signals), bpy (red signals) and signals for new bpy-adducts (blue signals) are seen in the reaction mixture. Below is an expansion of both spectra superimposed on one another, showing more clearly (in blue) at least eight new methyl/methyl carboxylate environments at low intensity.

Under similar conditions, but instead in acetone-d6 and without requiring any heating to accelerate ligand exchange kinetics, at least 70% of the [Ru(dpp)2] species in solution after four hours of stirring were a form of bpy-adduct (Σ Inew CO2Me/Me = 4.71 versus Σ Itotal CO2Me/Me = 6.75).

James McPherson – June 2019 116 Chapter 5: Hemilabile Ruthenium(II) Catalysts

The six new methyl/methyl carboxylate environments observed in the 1H NMR spectrum of the mixture (middle spectrum and inset, Figure 5.23) were again inconsistent with selective

1 formation of only one pair of C2 symmetric enantiomers. Additional H environments may be obscured by the nearby peaks assigned to water (H2O and HOD) and residual protic solvent

– (acetone-d5). It is also important to note that during these experiments, no unbound dpp ligand was detected by 1H NMR.

1 CO Me,Me Figure 5.23: 500 MHz H NMR spectra of [Ru(dpp 2 )2] in acetone-d6 before (top spectrum) and after (middle spectrum) adding bpy (2 equiv.) and stirring for four hours at room temperature. A mixture of CO Me,Me [Ru(dpp 2 )2] (black signals), bpy (red signals) and signals for new bpy-adducts (blue signals) are seen in the reaction mixture. The inset shows an expansion of the middle spectrum over the signals assigned to the different methyl/methyl carboxylate environments that are not obscured by the nearby signals due to residual water and protic solvent. The bottom spectrum is of HdppCO2Me,Me (green signals) and reveals that this species is not present in the reaction mixture (middle spectrum).

Analysis of this mixture of ruthenium species in acetone by (+)-ESI-MS confirmed the

CO Me,Me + presence of the parent complex, [Ru(dpp 2 )2] (at m/z = 686.50, [M] calc. m/z = 686.13),

CO Me,Me and both the proton- and sodium-adducts of [Ru(dpp 2 )2(bpy)] (at m/z = 843.25, relative

James McPherson – June 2019 117 Chapter 5: Hemilabile Ruthenium(II) Catalysts abundance ~90% for [M + bpy + H]+ calc. m/z = 843.20; and at m/z = 865.25, relative abundance

~ 60% for [M + bpy + Na]+ calc. m/z = 865.18), with good agreement between the observed and theoretical isotopic distributions for each species as shown insets in Figure 5.24.

CO Me,Me Figure 5.24: The full low resolution mass spectrum (positive mode ESI) of a mixture of [Ru(dpp 2 )2] (M) and bpy after it was stirred at room temperature for 12 hours. Shown inset are the experimental versus simulated isotopic distributions for [M]+ in black (obs: m/z 686.50; calc. m/z 686.13), [M + bpy + H]+ in blue (obs: m/z 843.25; calc. m/z 843.20) and [M + bpy + Na]+ in pink (obs: m/z 865.25; calc. m/z 865.18).

5.2.4.1 Catalytic screening – Hydration of nitriles

Benzonitrile (PhCN) also binds to [Ru(dpp)2] complexes. For example, after heating

CO Et,CO Et + [Ru(dpp 2 2 )2] ([M + H] calc. m/z = 831.17) and PhCN in 1,2-dimethoxyethane (DME) at 160 °C for 2 h, the major species detected by (+)-ESI-MS (at m/z = 934.33) corresponds to

CO Et,CO Et + the proton adduct of the mono-solvato [Ru(dpp 2 2 )2(PhCN)] ([M + PhCN + H] calc. m/z = 934.21), with the proton-adducts of the unsolvated species (at m/z 831.33, relative abundance ~15%) and bis-solvato species at m/z = 1037.33 ([M + 2PhCN + H]+ calc. m/z =

1037.26, relative abundance ~5%) also observed in the gas phase (see Figure 5.25).

James McPherson – June 2019 118 Chapter 5: Hemilabile Ruthenium(II) Catalysts

CO Et,CO Et Figure 5.25: The full low resolution mass spectrum (positive mode ESI) of [Ru(dpp 2 2 )2] (M) and benzonitrile (PhCN) in 1,2-dimethoxyethane (DME) after heating at 160 °C in a sealed tube for 2 h. Inserts show the experimental (black) vs. simulated isotopic distributions for [M + H]+ in blue (obs: m/z 831.33; calc. m/z 831.17), [M + PhCN + H]+ in red (obs: m/z 934.33; calc. m/z 934.21) and [M + 2PhCN + H]+ in green (obs: m/z 1037.33; calc. m/z 1037.26).

The hydration of nitriles (Scheme 5.3) is a desirable route for the synthesis of amides, which are key intermediates in the fine chemical and polymer industries, with excellent atom economy. Transition metal-mediated or -catalysed nitrile hydration avoids the possibility of over-hydrolysis and the poor functional group compatibility generally associated with traditional amide syntheses which rely on the use strong acids or bases, and as such has received much attention.23-30

Scheme 5.3: Hydration of nitriles to form amides.

Transition metal-catalysed nitrile hydration relies on the activation of a nitrile to nucleophilic attack through binding a transition metal centre, such as a ruthenium(II) ion.

Additional cooperativity has been achieved29-34 through the construction of a ligand bearing a

James McPherson – June 2019 119 Chapter 5: Hemilabile Ruthenium(II) Catalysts basic hydrogen bond accepting group (E in Figure 5.26) in proximity to the bound nitrile to increase the nucleophilicity of water in the secondary coordination sphere, with two recent examples of such catalysts presented in Figure 5.26. It was envisaged that the hemilabile pyridyl donors in the [Ru(dpp)2] complexes may serve such function, and so the activity of

CO Et,CO Et [Ru(dpp 2 2 )2] as a catalyst for the hydration of benzonitrile was investigated as follows.

Figure 5.26: Cooperative strategy to facilitate nitrile hydration. Nitrile binding at metal centres increases its electrophilicity, while H-bonding interactions between the unbound pyridyl (in proximity to the bound substrate) and water increase the nucleophilicity of the water molecule. This strategy has been previously exploited successfully, for example presented on the left are Oshiki’s ruthenium(II)32 and Bera’s rhodium(I)33 catalysts.

The activity of [Ru(dpp)2] as a catalyst for the hydration of nitriles was initially screened under mild conditions, by heating benzonitrile in a mixture of ethanol and water (2:1 v/v) at

CO Et,CO Et reflux with a catalytic amount of [Ru(dpp 2 2 )2] (0.5 mol%) present, and the progress of the reaction was monitored by 1H NMR spectroscopy. After four days, 35% conversion of benzonitrile to benzamide was achieved, as determined from the relative integrals of each

1 species in the 400 MHz H NMR spectrum acquired in DMSO-d6 (Figure 5.27, bottom spectrum).

James McPherson – June 2019 120 Chapter 5: Hemilabile Ruthenium(II) Catalysts

Figure 5.27: Stacked 400 MHz 1H NMR spectra of aliquots from a mixture of benzonitrile (peaks highlighted in CO Et,CO Et blue) and [Ru(dpp 2 2 )2] (0.5 mol%) in ethanol/water (2:1 v/v) which were dissolved in DMSO-d6, taken before (top) and after the reaction mixture was heated at reflux for 2 h (middle) or 4 days (bottom). Signals assigned to benzamide are highlighted in red. The conversion to benzamide after 4 days is estimated to be ~35%.

The reaction mixture was then evaporated to dryness to remove solvent and unreacted benzonitrile. The light red solid residue was determined to be a clean mixture of the desired

CO Et,CO Et 1 benzamide product and intact [Ru(dpp 2 2 )2] catalyst by H NMR spectroscopy (see

Figure 5.28), with no trace of the over-hydrolysed product benzoic acid, nor any evidence of decomposed or dissociated dpp– ligand.

James McPherson – June 2019 121 Chapter 5: Hemilabile Ruthenium(II) Catalysts

Figure 5.28: 400 MHz 1H NMR spectrum of the pale red solid residue remaining after benzonitrile and CO Et,CO Et [Ru(dpp 2 2 )2] (0.5 mol%) were heated at reflux in ethanol/water (2:1 v/v) for four days and the mixture evaporated to dryness (at 78 mbar and 40 ºC), and then re-dissolved in chloroform-d. Signals assigned to CO Et,CO Et benzamide are highlighted in red, and a magnified spectrum showing the signals attributed to [Ru(dpp 2 2 )2] in black is presented above.

Attempts to improve the rate of conversion through solvent systems with increased boiling points (e.g., 1,2-dimethoxyethane/water) were unsuccessful, possibly due to the entropic penalties associated with nitrile binding discussed in Section 5.2.2.2.

5.2.4.2 Oxidative cleavage of styrene

High-valent ruthenium-oxo species, such as ruthenium tetroxide (RuO4), have been widely used as oxidants for a variety of organic species.35-36 Alkenes often undergo oxidative cleavage to form carbonyls or carboxylic acids in the presence of RuO4, which can be generated in situ from catalytic amounts of RuCl3 or RuO2 and a stoichiometric oxidant such as NaIO4.

This reactivity can be modulated through either the choice of solvent system or the addition of ancillary ligands (typically poly-pyridyls such as bpy or 1,10-phenanthroline derivatives, or porphyrins) to favour instead cis–dihydroxylation or epoxidation, for example, as shown in

Scheme 5.4.

James McPherson – June 2019 122 Chapter 5: Hemilabile Ruthenium(II) Catalysts

Scheme 5.4: Olefin oxidation by high-valent ruthenium-oxo species formed in situ. A. Oxidative cleavage affords carboxylic acids,37 while cis-dihydroxylated products are also possible reaction outcomes under appropriate conditions as shown in B.38 The reactivity of the ruthenium catalyst can be further modulated through ancillary ligands, such as 3,4,7,8-tetramethyl-1,10-phenanthroline, to favour epoxidation as shown in C.39

Two ruthenium(II) complexes, with labile 1,4-cyclooctadiene (COD) ligands, have been recently shown to selectively cleave olefins into aldehydes.40-41 As shown in Scheme 5.5, the mechanisms proposed suggest that these complexes were precatalysts, with COD dissociation required to activate the complex for catalytic turnover in the presence of an oxidant (periodate).

As we have shown above in Section 5.2.2, vacant sites at ruthenium can be accessed by small molecules without complete (irreversible) ligand dissociation of [Ru(dpp)2] complexes. The reactivities of these complexes was therefore studied under the conditions reported originally by Bera’s group.40

James McPherson – June 2019 123 Chapter 5: Hemilabile Ruthenium(II) Catalysts

Scheme 5.5: The oxidation of styrene to benzaldehyde is catalysed selectively by ruthenium complexes reported by Bera’s40 and Yi’s groups.41 The first step in the proposed mechanisms for each complex involves the dissociation of the labile COD ligand from the isolated ruthenium(II) precatalyst. The [Ru(dpp)2] complexes studied in this work have vacant sites at ruthenium masked by hemilabile pyridyl donors, which allow formation of an analogous cis-oxo active ruthenium intermediate without irreversible ligand dissociation/degradation.

Within two hours of stirring a mixture of styrene with two equivalents of sodium

CO Et,CO Et periodate and catalytic [Ru(dpp 2 2 )2] (0.5 mol%), the styrene had been completely consumed, with no signals associated with the alkene observed in the 1H NMR spectrum. A large signal was observed at δ 10.03 which corresponds to the formation of an aldehyde functional group (Figure 5.29). Comparison of the 1H NMR spectrum of the reaction mixture to that of an authentic sample of benzaldehyde (shown at the bottom of Figure 5.29) confirms benzaldehyde as the major product.

James McPherson – June 2019 124 Chapter 5: Hemilabile Ruthenium(II) Catalysts

Figure 5.29: Stacked 400 MHz 1H NMR spectra in chloroform-d of (top to bottom): styrene (blue); an aliquot CO Et,CO Et from the reaction mixture of styrene (blue), NaIO4 and [Ru(dpp 2 2 )2] after 45 minutes at room temperature; an aliquot from the same reaction mixture after 2 h at room temperature; and bottom is a sample of authentic benzaldehyde (red signals).

Signals associated with the key [Ru(dpp)2Lx] intermediates analogous to those proposed for Bera’s system40 were observed in the high resolution mass spectrum of the crude reaction mixture acquired after the conclusion of the reaction (with excellent agreement between observed and simulated isotopic peak distributions, as is shown in Scheme 5.5). These intermediates are the cis-dioxoruthenium(VI) [Ru(dpp)2O2] (A) and its [3 + 2] cycloadduct with styrene, B. Therefore, the similar mechanism shown in Scheme 5.6 is proposed for this reaction.

CO Et,CO Et + The ruthenium complex [Ru(dpp 2 2 )2] ([M] : calc. m/z = 830.1670) is oxidised by periodate, with overlapping signals centred at m/z = 862.1512 and 863.1594 (relative abundance

~ 1%) corresponding to formation of the reactive cis-dioxoruthenium(VI) species A ([M +

2O]+: calc. m/z = 862.1569; and [M + 2O + H]+: calc. m/z = 863.1642). Studies of stoichiometric reactions of other cis-dioxoruthenium(VI) species with styrene reveal benzaldehyde is the major product with small amounts (up to 5%) of styrene oxide also observed.42-44 Consequently,

James McPherson – June 2019 125 Chapter 5: Hemilabile Ruthenium(II) Catalysts therefore, intermediate A then undergoes [3 + 2] cycloaddition with styrene, and the most abundant ruthenium-containing signal in the mass spectrum at m/z = 967.2219 (relative abundance ~15%) is consistent with intermediate B (corresponding to [M + 2O + styrene + H]+: calc. m/z = 967.2268). Oxidative cleavage of the C–C bond prior to release of the products would afford species C with the same mass as intermediate B. The release of the aldehyde

CO Et,CO Et products regenerates [Ru(dpp 2 2 )2] to close the cycle, and peaks at m/z = 830.1615 at

~5% relative abundance were detected confirming the integrity of the [Ru(dpp)2] complex

([M]+: calc. m/z = 830.1670).

Scheme 5.6: The proposed mechanism for the selective oxidation of styrene to benzaldehyde catalysed by CO Et,CO Et [Ru(dpp 2 2 )2] (M), with the observed (black) vs. simulated (green and orange) isotopic distributions for each intermediate shown, from the high resolution mass spectrum of the reaction mixture.

5.3 Conclusions

Chelation of dpp– ligands to a ruthenium(II) ion induces ca. 80 kJ mol–1 strain within each ligand, and release of a pendant pyridyl donor group relieves ca. 75% of this strain. Under ambient conditions, mono- and bis-solvato species were observed during 1H NMR spectroscopy

James McPherson – June 2019 126 Chapter 5: Hemilabile Ruthenium(II) Catalysts experiments in coordinating solvents, without irreversible ligand dissociation. Gas-phase

2 theoretical studies suggest the cis-[Ru(κ -dpp)(MeCN)2] geometry is preferred for the bis- solvato acetonitrile species, and as such that the two “vacant” sites protected by the dpp– ligands are cis to one another. The binding of small molecules, e.g. nitriles, sulfoxides and aromatic heterocycles (pyridyl groups), is reversible. This strain-induced hemilability has been exploited for the catalytic transformation of small molecules which access these vacant sites at the ruthenium centre.

5.4 Experimental

5.4.1 General Methods

All NMR spectra were recorded on Bruker Avance III DPX-300, 400, 500 or 600 spectrometers operating at 300, 400, 500 or 600 MHz for 1H NMR experiments (75.5, 100.6,

125.8 or 150.9 MHz for 13C NMR; 30.4, 40.6, 50.7 or 60.8 MHz for 15N NMR), respectively.

All chemical shifts in 1H or 13C NMR spectra were calibrated against residual solvent signals.45

A 1-dimensional (1D) 1H NMR spectrum was acquired immediately prior to each 1H-15N

HMBC NMR experiment. Calibration of the residual solvent in the 1H NMR spectrum allowed calibration of the f1 (15N) and f2 (1H) dimensions in the following 2D experiment (Ξ =

10.132912).46 The scalar coupling constant of 11 Hz for 2J(15N, 1H) was used for all 1H-15N

HMBC NMR experiments.47 All coupling constants (J) are reported in hertz. Signals in NMR spectra are reported as broad (br) singlets (s), doublets (d), triplets (t), quartets (q) or unclear multiplets (m). Unless otherwise stated, all spectra were obtained at 298 K. For variable temperature (VT) 1H NMR studies, two experiments were performed at each temperature, at least 15 minutes apart, to confirm that the composition of the solution had reached thermal equilibrium. NMR spectra were processed with MestReNova 12.0 software.48 FT-IR experiments were performed with an Agilent Cary 630 ATR FTIR spectrometer. Signals in FT-

James McPherson – June 2019 127 Chapter 5: Hemilabile Ruthenium(II) Catalysts

IR spectra are reported as broad (br), weak (w), medium (m) or strong (s). UV-Vis experiments were performed on a Varian Cary 60 spectrophotometer. High resolution mass spectra were run on a Thermo Fisher Scientific Orbitrap LTQ XL ion trap mass spectrometer using a nanospray ionisation source. Low-resolution mass spectrometry experiments were performed on a Thermo

Fisher LCQ mass spectrometer. Elemental micro-analyses were performed by the Chemical

Analysis Facility at Macquarie University.

5.4.1.1 Synthesis and Reagents

Unless specified, solvents were taken from an Innovative Technology Pure Solvent

Dispenser immediately prior to use and commercially available reagents were used as received.

2-Formyl pyridine was dissolved in diethyl ether, filtered through 60-200 µm silica gel, and the ether removed by rotary evaporation immediately prior to use. Flash and gravity chromatographies were carried out on 60-200 µm silica gel. Reactions were carried out under

CO Me,Me 13 CO Et,CO Et 14-15 CO Me,Me 13 inert atmospheres. Reagents dpp 2 H, dpp 2 2 H, [Ru(dpp 2 )2] and

49 [Ru(dmso)4Cl2] were prepared according to literature methods.

5.4.1.2 Electrochemistry

High purity dinitrogen was used to deoxygenate the solution before each electrochemical experiment. Solutions of compounds were of ca. 1.0 mM concentration for cyclic voltammetry experiments with tetrabutylammonium hexafluorophosphate (0.1 M) as the supporting electrolyte. Cyclic voltammograms (CVs) were recorded using a GAMRY Reference 600 potentiostat. A three-electrode system with a glassy carbon working electrode (1 mm diameter), platinum wire counter electrode and an EDAQ leakless Ag/AgCl reference electrode was used.

Immediately prior to recording each CV, the working electrode was polished with increasingly fine alumina, rinsed with water then solvent, and dried with a Kimwipe. Prior to recording CVs, a blank background was always obtained between the anodic and cathodic solvent discharge

James McPherson – June 2019 128 Chapter 5: Hemilabile Ruthenium(II) Catalysts limits. Ferrocene (Fc) was added as an internal standard at the end of each set of cyclic voltammetry experiments, and all potentials are quoted relative to the measured Fc+/0 couple.

5.4.2 Synthesis

CO Et,CO Et [Ru(dpp 2 2 )2]:

The reported literature method was adapted as follows.13 dppCO2Et,CO2EtH (100.2 mg,

0.2742 mmol) and Et3N (0.3 mL, 2 mmol) were combined in degassed methanol (5 mL). Upon addition of [Ru(dmso)4Cl2] (68.9 mg, 0.142 mmol), the yellow solution was stirred and heated at 50 ºC overnight. After 1 h, the solution had changed colour to a deep red/brown. On

CO Et,CO Et completion of the reaction, the mixture was cooled to 0 ºC, and [Ru(dpp 2 2 )2] (24.7 mg,

1 25%) was collected by filtration as red/brown micro needles. H NMR (300 MHz, benzene-d6)

δ 8.86 (ddd, J = 8.2, 1.5, 0.8 Hz, 4H, HB3), 6.95 (ddd, J = 5.5, 1.6, 0.8 Hz, 4H, HB6), 6.61 (ddd,

J = 8.2, 7.3, 1.6 Hz, 4H, HB4), 5.59 (ddd, J = 7.3, 5.5, 1.5 Hz, 4H, HB5), 4.49 (q, J = 7.1 Hz, 8H,

A3″ A3′′′ 13 1 A3′ H ), 1.28 (t, J = 7.1 Hz, 12H, H ). C{ H} NMR (101 MHz, benzene-d6) δ 166.62 (C ),

160.35 (CB2), 152.82 (CB6), 141.00 (CA2), 135.04 (CB4), 121.56 (CB5), 121.40 (CB3), 118.72

A3 A3″ A3′′′ –6 –1 –1 (C ), 60.52 (C ), 14.67 (C ). UV-Vis (THF, 6.15 x 10 M): λmax /nm (ε / L mol cm )

335 (3.7 × 104), 364 (1.7 × 104), 382 (1.5 × 104), 418 (7.0 × 103), 462 (5.6 × 103), 545 (2.8 ×

3 + 10 ). (+)-HR-NSI-MS found: m/z 830.1635 (calc. [C40H36N6O8Ru (M)] : m/z 830.1633). Anal.

Calc. for C40H36N6O8Ru C: 57.90 H: 4.37 N: 10.13%. Found: C: 56.82 H: 4.36 N: 9.76%.

5.4.3 DFT Calculations

DFT calculations were performed with the Gaussian09 software package.50 Geometry optimisations for the ruthenium(II) complexes were performed using the gradient-corrected

BP86 exchange correlation functional with either of the split valence def2-SVP or def2-TZVP

James McPherson – June 2019 129 Chapter 5: Hemilabile Ruthenium(II) Catalysts basis sets for all atoms, with pseudo potentials for the core electrons of the ruthenium(II) ion.

Dispersion corrections were included by adding the D3 version of Grimme’s dispersion. The optimisations in vacuo were followed by the calculation of vibrational frequencies to ensure the located coordinates were energy minima with no imaginary frequencies.

To calculate the relative energies of each (dppCO2Et,CO2Et)– ligand conformation, their cartesian coordinates were harvested from the BP86-GD3/def2-TZVP geometries of the

CO Et,CO Et [Ru(dpp 2 2 )2(MeCN)x] (x = 0, 1 or 2) complexes, and single point energy calculations performed using B3LYP-GD3BJ/6-31G**, as discussed in Chapter 2.10

5.4.4 Small Scale Stoichiometric Reactions

R1,R2 1 2 A red solution of [Ru(dpp )2] complex (R ,R = CO2Me,Me or CO2Et,CO2Et; ca. 1 mg, ~1 µmol) in deuterated solvent (0.6 mL) was prepared and analysed by 1H NMR spectroscopy. The solution was transferred to a small vial charged with a magnetic stir bar and a reagent (ca. 2–5 µmol). The mixture was then stirred under conditions as specified below, and then transferred to an NMR tube and analysed by 1H NMR spectroscopy, and in some cases by (+)-ESI-MS. All 1H NMR spectra were acquired at 298 K.

5.4.4.1 Tetrabutylammonium chloride {[Bu4N]Cl}

CO Me,Me A five-fold excess of [Bu4N]Cl was added to a solutions of [Ru(dpp 2 )2] in acetone-

CO Et,CO Et d6 and of [Ru(dpp 2 2 )2] in acetonitrile-d3, and each of these solutions were stirred under

+ ambient conditions for 1 hour. In both cases, other than the presence of [Bu4N] , no changes in

1 CO Me,Me the H NMR spectra {acquired at 300 MHz for [Ru(dpp 2 )2] and at 400 MHz for

CO Et,CO Et [Ru(dpp 2 2 )2]} were observed.

James McPherson – June 2019 130 Chapter 5: Hemilabile Ruthenium(II) Catalysts

5.4.4.2 Acetic Acid (AcOH)

CO Me,Me A two-fold excess of AcOH was added to a solution of [Ru(dpp 2 )2] in benzene-d6.

Other than the presence of AcOH, no change in the 300 MHz 1H NMR spectrum of

CO Me,Me [Ru(dpp 2 )2] was observed, neither after the solution was stirred under ambient conditions for 2 hours, nor after the solution was heated at reflux for 24 hours.

5.4.4.3 Hydrochloric Acid (HCl)

CO Me,Me HCl (12 M, 0.05 mL, aq) was added to a solution of [Ru(dpp 2 )2] in benzene-d6, which was stirred under ambient conditions for 15 minutes. In the 400 MHz 1H NMR spectrum, a large, very broad signal centred at δ 7.9 was attributed to acidic protons, while the signals

CO Me,Me assigned to the [Ru(dpp 2 )2] complex appeared broader (with fwhh ≈ 15 Hz) than those observed prior to the addition of HCl (with fwhh ≤ 3 Hz), such that splitting due to scalar coupling was no longer observable.

5.4.4.4 Carbon monoxide (CO)

CO Me,Me CO (g) was bubbled through a solution of [Ru(dpp 2 )2] in benzene-d6 for 15 minutes, then the NMR tube was sealed and analysed immediately by 1H NMR spectroscopy.

No additional species were detected in the 400 MHz 1H NMR spectrum.

5.4.4.5 Pyridine (py)

CO Me,Me A twenty-fold excess of pyridine was added to a solution of [Ru(dpp 2 )2] in benzene-d6, and the solution stirred under ambient conditions for 15 minutes. Six new singlets were observed in the 400 MHz 1H NMR spectrum at δ 3.81, 3.73, 3.72, 3.64, 2.85, and 2.82 and these were attributed to the methyl carboxylate/methyl environments of pyridine adducts of the ruthenium(II) complex. Note that additional signals were also observed in the aromatic

James McPherson – June 2019 131 Chapter 5: Hemilabile Ruthenium(II) Catalysts region, but were obscured in many cases by signals assigned to free pyridine or the parent

CO Me,Me [Ru(dpp 2 )2] complex.

CO Et,CO Et Pyridine (0.05 mL, 0.6 mmol) was added to solutions of [Ru(dpp 2 2 )2] (61.5 µM or 6.15 µM) in THF (3 mL). No change in the UV-vis region of the electronic spectra were observed in either case.

5.4.4.6 2,2′-Bipyridine (bpy)

A two-fold excess of bpy (1.32 mg, 8.45 µmol) was added to a solution of

CO Me,Me [Ru(dpp 2 )2] (3.16mg, 4.37 µmol) in benzene-d6 (0.6 mL), and the solution was stirred under ambient conditions for 2 h and then at 70 ºC for an additional 2 h. The eight new singlets observed in the 400 MHz 1H NMR spectrum at δ 3.66, 3.65, 3.28 (2 singlets), 3.05, 3.04, and

2.41 (2 singlets), were attributed to the methyl carboxylate/methyl environments of bpy adducts of the ruthenium(II) complex. Additional signals were also observed in the aromatic region, however these were difficult to identify due to the low signal/noise ratio.

CO Me,Me A two-fold excess of bpy was added to a solution of [Ru(dpp 2 )2] in acetone-d6, and the solution was stirred under ambient conditions for 4 h. The six new singlets observed in the 500 MHz 1H NMR spectrum at δ 3.77 (2 singlets), 3.17 (2 singlets), 2.71, and 1.86, were attributed to the methyl carboxylate/methyl environments of bpy adducts of the ruthenium(II) complex. Additional signals were also observed in the aromatic region, e.g. at δ 9.16, 8.16,

8.04, 7.87 – 7.81 (m), 7.65 – 7.55 (m), 7.31 – 7.19 (m), 6.93 – 6.76 (m), and 6.36 – 6.18 (m), however additional signals were obscured by signals assigned to bpy or unreacted complex, and so these were not assigned. This mixture was also analysed by (+)-ESI-MS after it was stirred

+ under ambient conditions for a total of 12 h: found m/z 686.50 (calc. [C34H28N6O4Ru (M)] : m/z

686.13), m/z 843.25 (calc. [M + bpy + H]+: m/z 843.20) and m/z 865.25 (calc. [M + bpy + Na]+: m/z 865.18).

James McPherson – June 2019 132 Chapter 5: Hemilabile Ruthenium(II) Catalysts

5.4.4.7 Benzonitrile (PhCN)

CO Et,CO Et Benzonitrile (100 µL, 1 mmol) was added to a solution of [Ru(dpp 2 2 )2] (1.7 mg,

2 µmol) in 1,2-dimethyoxyethane (DME, 0.5 mL), and the solution was then heated at 160 ºC for 2 hours. The (+)-ESI-MS showed peak envelopes at m/z 831.33 (calc. [C40H36N6O8Ru (M)

+ H]+: m/z 831.17), m/z 934.33 (calc. [M + PhCN + H]+: m/z 934.21), and m/z 1037.33 (calc.

[M + 2PhCN + H]+: m/z 1037.26).

5.4.5 Catalytic Screening Experiments

5.4.5.1 Hydration of Benzonitrile

CO Et,CO Et A red solution of benzonitrile (206.0 µL, 2.0 mmol) and [Ru(dpp 2 2 )2] (6.4 mg,

0.0077 mmol) in ethanol (10 mL) and water (5 mL) was heated at reflux.51 The reaction was monitored by 1H NMR spectroscopy: aliquots of the reaction mixture were sampled, which were dissolved in DMSO-d6 then the spectrum recorded. After four days, the pale red mixture was cooled to room temperature, and solvent removed under vacuum to afford a mixture of

CO Et,CO Et 1 benzamide and [Ru(dpp 2 2 )2] (combined mass 21.0 mg). H NMR (400 MHz, CDCl3) δ

B3 CO Et,CO Et 8.41 (ddd, J = 8.2, 1.5, 0.8 Hz, 4H, H [Ru(dpp 2 2 )2]), 8.15 – 8.06 (ddd, J = 5.6, 1.7,

B6 CO Et,CO Et B4 0.8 Hz, 4H, H [Ru(dpp 2 2 )2]), 7.36 (ddd, J = 8.2, 7.4, 1.7 Hz, 4H, H

CO Et,CO Et B5 CO Et,CO Et [Ru(dpp 2 2 )2]), 6.55 (ddd, J = 7.4, 5.6, 1.5 Hz, 4H, H [Ru(dpp 2 2 )2]), 4.50 (q, J

A3″ CO Et,CO Et A3′′′ CO Et,CO Et = 7.2 Hz, 8H, H [Ru(dpp 2 2 )2]), 1.50 (t, J = 7.1 Hz, 12H, H [Ru(dpp 2 2 )2])

7.87 – 7.77 (m, 88H), 7.58 – 7.50 (m, 44H), 7.50 – 7.41 (m, 88H), 6.04 (brs, 44H), 5.67 (brs,

44 H).

James McPherson – June 2019 133 Chapter 5: Hemilabile Ruthenium(II) Catalysts

5.4.5.2 Oxidation of Styrene

The procedure was adapted from that reported by Bera et al.40 as follows. Styrene (115

CO Et,CO Et µL, 1.00 mmol) was added to a red solution of [Ru(dpp 2 2 )2] (3.1 mg, 0.0037 mmol) in

EtOAc (2.0 mL) and MeCN (2.0 mL) and the mixture stirred under ambient conditions for 5 minutes. NaIO4 (0.4535 g, 2.120 mmol) and water (1.0 mL) were then sequentially added, and the solution stirred for 3 days. Aliquots of the mixture were filtered through Celite® 545

(particle size 0.02–0.1 mm), extracted with chloroform-d, and the 1H NMR spectra recorded.

(+)-HR-NSI-MS was performed on the crude reaction mixture after 3 days, after filtration through Celite® 545 to first remove insoluble impurities (NaI).

5.5 References

1. Organometallic Pincer Chemistry. Berlin, Heidelberg : Springer: Berlin, Heidelberg, 2013. 2. Pincer Compounds: Chemistry and Applications. 1st ed.; Elsevier Inc.: Amsterdam, 2018; p 754. 3. Peris, E.; Crabtree, R. H., Key factors in pincer ligand design. Chem. Soc. Rev. 2018, 47 (6), 1959-1968. 4. Mukherjee, A.; Milstein, D., Homogeneous Catalysis by Cobalt and Manganese Pincer Complexes. ACS Catal. 2018, 8 (12), 11435-11469. 5. Jeffrey, J. C.; Rauchfuss, T. B., Metal complexes of hemilabile ligands. Reactivity and structure of dichlorobis(o-(diphenylphosphino)anisole)ruthenium(II). Inorg. Chem. 1979, 18 (10), 2658-2666. 6. Attwood, M.; Turner, S. S., Back to back 2,6-bis(pyrazol-1-yl)pyridine and 2,2′:6′,2″- terpyridine ligands: Untapped potential for spin crossover research and beyond. Coord. Chem. Rev. 2017, 353, 247-277. 7. Chakraborty, S.; Newkome, G. R., Terpyridine-based metallosupramolecular constructs: tailored monomers to precise 2D-motifs and 3D-metallocages. Chem. Soc. Rev. 2018, 47 (11), 3991-4016. 8. Gao, Z.; Han, Y.; Gao, Z.; Wang, F., Multicomponent Assembled Systems Based on Platinum(II) Terpyridine Complexes. Acc. Chem. Res. 2018, 51 (11), 2719-2729. 9. Mede, T.; Jäger, M.; Schubert, U. S., “Chemistry-on-the-complex”: functional RuII polypyridyl-type sensitizers as divergent building blocks. Chem. Soc. Rev. 2018, 47 (20), 7577- 7627. 10. McPherson, J. N.; Elton, T. E.; Colbran, S. B., A Strain-Deformation Nexus within Pincer Ligands: Application to the Spin States of Iron(II) Complexes. Inorg. Chem. 2018, 57 (19), 12312-12322. 11. Ludlow III, J. M.; Guo, Z.; Schultz, A.; Sarkar, R.; Moorefield, C. N.; Wesdemiotis, C.; Newkome, G. R., Group 8 Metallomacrocycles – Synthesis, Characterization, and Stability. Eur. J. Inorg. Chem. 2015, 2015 (34), 5662-5668.

James McPherson – June 2019 134 Chapter 5: Hemilabile Ruthenium(II) Catalysts

12. McPherson, J. N.; Das, B.; Colbran, S. B., Tridentate pyridine–pyrrolide chelate ligands: An under-appreciated ligand set with an immensely promising coordination chemistry. Coord. Chem. Rev. 2018, 375, 285-332. 13. McSkimming, A.; Diachenko, V.; London, R.; Olrich, K.; Onie, C. J.; Bhadbhade, M. M.; Bucknall, M. P.; Read, R. W.; Colbran, S. B., An easy one-pot synthesis of diverse 2,5- di(2-pyridyl)pyrroles: a versatile entry point to metal complexes of functionalised, meridial and tridentate 2,5-di(2-pyridyl)pyrrolato ligands. Chem. Eur. J. 2014, 20 (36), 11445-56. 14. Hein, F. M., F., Synthesis of 2,5-di(α-pyridyl)pyrrole and of several complex derivatives of 2,5-di(α-pyridyl)-3,4-dicarbethoxypyrrole. Die Pharmazie 1954, 9, 455-460. 15. Hein, F.; Beierlein, U., Preparation and chemistry of the complexes of 2,5-di-(α- pyridyl)pyrrole. Pharm. Zentralhalle Dtschl. 1957, 96, 401-21. 16. Chow, H. S.; Constable, E. C.; Housecroft, C. E.; Kulicke, K. J.; Tao, Y., When electron exchange is chemical exchange–assignment of 1H NMR spectra of paramagnetic cobalt(II)- 2,2′:6′,2″-terpyridine complexes. Dalton Trans. 2005, (2), 236-237. 17. McPherson, J. N.; Hogue, R. W.; Akogun, F. S.; Bondi, L.; Luis, E. T.; Price, J. R.; Garden, A. L.; Brooker, S.; Colbran, S. B., Predictable Substituent Control of Co(III/II) Redox Potential and Spin Crossover in Bis(dipyridylpyrrolide)cobalt Complexes. Inorg. Chem. 2019, 58 (3), 2218-2228. 18. Mason, J., Nitrogen nuclear magnetic resonance spectroscopy in inorganic, organometallic, and bioinorganic chemistry. Chem. Rev. 1981, 81 (3), 205-227. 19. Hansch, C.; Leo, A.; Taft, R. W., A survey of Hammett substituent constants and resonance and field parameters. Chem. Rev. 1991, 91 (2), 165-195. 20. Robinson, K.; Gibbs, G. V.; Ribbe, P. H., Quadratic elongation: a quantitative measure of distortion in coordination polyhedra. Science 1971, 172 (3983), 567-70. 21. Armarego, W. L. F., Chapter 3 - Purification of Organic Chemicals. In Purification of Laboratory Chemicals (Eighth Edition), Armarego, W. L. F., Ed. Butterworth-Heinemann: 2017; pp 95-634. 22. Nomenclature of Inorganic Chemistry IUPAC Recommendations 2005. RSC Publishing: Cambridge UK, 2005. 23. Parkins, A. W., Catalytic Hydration of Nitriles to Amides. Platin. Met. Rev. 1996, 40 (4), 169-174. 24. Kukushkin, V. Y.; Pombeiro, A. J. L., Additions to Metal-Activated Organonitriles. Chem. Rev. 2002, 102 (5), 1771-1802. 25. Kukushkin, V. Y.; Pombeiro, A. J. L., Metal-mediated and metal-catalyzed hydrolysis of nitriles. Inorganica Chim. Acta 2005, 358 (1), 1-21. 26. Ahmed, T. J.; Knapp, S. M. M.; Tyler, D. R., Frontiers in catalytic nitrile hydration: Nitrile and cyanohydrin hydration catalyzed by homogeneous organometallic complexes. Coord. Chem. Rev. 2011, 255 (7), 949-974. 27. García-Álvarez, R.; Crochet, P.; Cadierno, V., Metal-catalyzed amide bond forming reactions in an environmentally friendly aqueous medium: nitrile hydrations and beyond. Green Chem. 2013, 15 (1), 46-66. 28. Downs, E. L.; Tyler, D. R., Nanoparticle catalysts for nitrile hydration. Coord. Chem. Rev. 2014, 280, 28-37. 29. Singh, K.; Sarbajna, A.; Dutta, I.; Pandey, P.; Bera, J. K., Hemilability-Driven Water Activation: A NiII Catalyst for Base-Free Hydration of Nitriles to Amides. Chem. Eur. J. 2017, 23 (32), 7761-7771. 30. Xing, X.; Xu, C.; Chen, B.; Li, C.; Virgil, S. C.; Grubbs, R. H., Highly Active Platinum Catalysts for Nitrile and Cyanohydrin Hydration: Catalyst Design and Ligand Screening via High-Throughput Techniques. J. Am. Chem. Soc. 2018, 140 (50), 17782-17789.

James McPherson – June 2019 135 Chapter 5: Hemilabile Ruthenium(II) Catalysts

31. Fung, W. K.; Huang, X.; Man; ManNg, S. M.; Hung, M. Y.; Lin, Z.; Lau, C. P., Dihydrogen-Bond-Promoted Catalysis: Catalytic Hydration of Nitriles with the 5 Indenylruthenium Hydride Complex (η -C9H7)Ru(dppm)H (dppm = Bis(diphenylphosphino)methane). J. Am. Chem. Soc. 2003, 125 (38), 11539-11544. 32. Oshiki, T.; Yamashita, H.; Sawada, K.; Utsunomiya, M.; Takahashi, K.; Takai, K., Dramatic Rate Acceleration by a Diphenyl-2-pyridylphosphine Ligand in the Hydration of Nitriles Catalyzed by Ru(acac)2 Complexes. Organometallics 2005, 24 (26), 6287-6290. 33. Daw, P.; Sinha, A.; Rahaman, S. M. W.; Dinda, S.; Bera, J. K., Bifunctional Water Activation for Catalytic Hydration of Organonitriles. Organometallics 2012, 31 (9), 3790-3797. 34. González-Fernández, R.; Crochet, P.; Cadierno, V.; Menéndez, M. I.; López, R., Phosphinous Acid-Assisted Hydration of Nitriles: Understanding the Controversial Reactivity of Osmium and Ruthenium Catalysts. Chem. Eur. J. 2017, 23 (60), 15210-15221. 35. Sharpless, K. B.; Akashi, K.; Oshima, K., Ruthenium catalyzed oxidation of alcohols to aldehydes and ketones by amine-n-oxides. Tetrahedron Lett. 1976, 17 (29), 2503-2506. 36. Murahashi, S.; Komiya, N., Ruthenium-Catalyzed Oxidation for Organic Synthesis. In Modern Oxidation Methods, Bäckvall, J., Ed. 2010. 37. Carlsen, P. H. J.; Katsuki, T.; Martin, V. S.; Sharpless, K. B., A greatly improved procedure for ruthenium tetroxide catalyzed oxidations of organic compounds. J. Org. Chem. 1981, 46 (19), 3936-3938. 38. Shing, T. K. M.; Tai, V. W.-F.; Tam, E. K. W., Practical and Rapid Vicinal Hydroxylation of Alkenes by Catalytic Ruthenium Tetraoxide. Angew. Chem. Int. Ed. 1994, 33 (22), 2312-2313. 39. Eskénazi, C.; Balavoine, G.; Meunier, F.; Rivière, H., Tuning of reactivity in epoxidation of alkenes by iron and ruthenium complexes associated to non-porphyrinic ligands. Chem. Commun. 1985, (16), 1111-1113. 40. Daw, P.; Petakamsetty, R.; Sarbajna, A.; Laha, S.; Ramapanicker, R.; Bera, J. K., A highly efficient catalyst for selective oxidative scission of olefins to aldehydes: abnormal-NHC- Ru(II) complex in oxidation chemistry. J. Am. Chem. Soc. 2014, 136 (40), 13987-90. 41. Zhong, Y. Q.; Xiao, H. Q.; Yi, X. Y., Synthesis, structural characterization and catalysis of ruthenium(II) complexes based on 2,5-bis(2'-pyridyl)pyrrole ligand. Dalton Trans. 2016, 45 (45), 18113-18119. 42. Griffith, W. P.; Jolliffe, J. M.; Ley, S. V.; Williams, D. J., A new ruthenium(VI) oxidant: preparation, X-ray crystal structure, and properties of (Ph4P)[RuO2(OAc)Cl2]. Chem. Commun. 1990, (18), 1219-1221. 43. Li, C.-K.; Che, C.-M.; Tong, W.-F.; Tang, W.-T.; Wong, K.-Y.; Lai, T.-F., Synthesis, structure, reactivity and electrochemistry of cis-dioxoruthenium-(VI) and -(V) complexes containing N,N,N′,N′,3,6-hexamethyl-3,6-diazaoctane-1,8-diamine. Dalton Trans. 1992, (13), 2109-2116. 44. Cheng, W.-C.; Yu, W.-Y.; Cheung, K.-K.; Che, C.-M., A novel cis-dioxoruthenium(VI) complex of N,N′,N″-trimethyl-1,4,7-triazacyclononane (Me3tacn) for organic oxidation. Chem. Commun. 1994, (9), 1063-1064. 45. Fulmer, G. R.; Miller, A. J. M.; Sherden, N. H.; Gottlieb, H. E.; Nudelman, A.; Stoltz, B. M.; Bercaw, J. E.; Goldberg, K. I., NMR Chemical Shifts of Trace Impurities: Common Laboratory Solvents, Organics, and Gases in Deuterated Solvents Relevant to the Organometallic Chemist. Organometallics 2010, 29 (9), 2176-2179. 46. Harris, R. K.; Becker, E. D.; De Menezes, S. M. C.; Granger, P.; Hoffman, R. E.; Zilm, K. W., Further Conventions for NMR Shielding and Chemical Shifts (IUPAC Recommendations 2008). Magn. Reson. Chem. 2008, 46 (6), 582-598. 47. Battistin, F.; Balducci, G.; Demitri, N.; Iengo, E.; Milani, B.; Alessio, E., 15N NMR spectroscopy unambiguously establishes the coordination mode of the diimine linker 2-(2′-

James McPherson – June 2019 136 Chapter 5: Hemilabile Ruthenium(II) Catalysts pyridyl)pyrimidine-4-carboxylic acid (cppH) in Ru(II) complexes. Dalton Trans. 2015, 44 (35), 15671-15682. 48. MestReNova, 12.0.0-20080; Mestrelab Research S.L.: 2017. 49. Bratsos, I.; Alessio, E.; Ringenberg, M. E.; Rauchfuss, T. B., Ruthenium Complexes. In Inorganic Syntheses, John Wiley & Sons, Inc.: 2010; Vol. 35, pp 148-163. 50. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, Revision A01, Wallingford CT, 2009. 51. Jiang, X. B.; Minnaard, A. J.; Feringa, B. L.; de Vries, J. G., Platinum-catalyzed selective hydration of hindered nitriles and nitriles with acid- or base-sensitive groups. J. Org. Chem. 2004, 69 (7), 2327-31.

James McPherson – June 2019 137 Chapter 6: Miscellanea

6 MISCELLANEA

During the course of this project, various other curiosities were pursued. Homoleptic first

row transition metal complexes of 2,5-di(2-pyridyl)-3-R1-4-R2-pyrrolide ligands,

R1,R2 1-2 3 [M(dpp )2], analogous to those originally prepared by Hein and later by Tour, were also

prepared and their redox properties surveyed. Organometallic piano-stool Rh(III) and Ir(III)

complexes were prepared with dpp– ligands bound in a bidentate fashion with a pendant

unbound pyridyl, and tested as electrocatalysts for proton-dependent reduction. 3-Pyridin-2-yl-

N-[(1E)-pyridin-2-ylmethylidene]imidazo[1,5-a]pyridinamine (LImPy) was identified and

isolated as a useful side-product of Colbran’s one-pot Hdpp synthesis,4 and its coordination

chemistry was briefly probed. These preliminary studies are presented here to inspire—and to

provide the foundations for—future work.

6.1 Homoleptic Transition Metal Complexes

In addition to the [Co(dpp)2] complexes described in Chapter 3 and [Ru(dpp)2] complexes

R1,Me II II II 1 in Chapter 5, six other neutral homoleptic [M(dpp )2] (where M = Ni , Cu or Zn and R

= CO2Me or Me) were prepared. These are analogous to some of the complexes originally

reported by Hein, Melichar and Beierlein in the 1950s,1-2 which were more recently studied by

Tour and co-workers.3 These complexes were prepared and characterised: in the solid-state by

SC-XRD or in solution by 1H NMR spectroscopy, primarily for the zinc(II) and nickel(II)

complexes. The redox properties of these complexes were of particular interest, and were

surveyed by cyclic voltammetry. In general, the study of these complexes complements the

James McPherson – June 2019 138 Chapter 6: Miscellanea more in-depth study, including of magnetic properties, on their cobalt analogues that is discussed in Chapter 3.

6.1.1 Synthesis and Characterisation

R1,Me 1 The [M(dpp )2] (M = Ni(II), Cu(II) or Zn(II), R = CO2Me or Me) complexes were

5 prepared as reported for the [Co(dpp)2] analogues described in Chapter 3. The divalent metal chloride salts were added to methanolic solutions of the ligand precursor HdppR1,Me and excess triethylamine. The neutral complex products were poorly soluble in methanol, and so were collected in reasonable to good yields (46–91%) after the mixture was heated at reflux for two hours under an inert atmosphere. The nickel(II) and copper(II) complexes were recrystallised from hot methanol. Characterisation was by 1H NMR spectroscopy and high-resolution mass spectrometry.

Me,Me In the high resolution NSI-mass spectra of [M(dpp )2] aspirated in methanol, peak

Me,Me + envelopes corresponding to the singly charged proton-adducts [M(dpp )2 (M) + H] were observed at m/z 555.1801 for M = Ni (calc. m/z 555.1802), m/z 560.1742 for M = Cu (calc. m/z

560.1744) and at m/z 561.1741 for M = Zn (calc. m/z 561.1740). In the (+)-HR-NSI-MS spectra

CO Me,Me of the [M(dpp 2 )2] complexes, peak envelopes corresponding to the singly charged

CO Me,Me + proton-adducts [M(dpp 2 )2 (M) + H] were observed at m/z 643.1599 for M = Ni (calc. m/z 643.1598), m/z 648.1531 for M = Cu (calc. m/z 648.1541) and at m/z 649.1526 for M = Zn

(calc. m/z 649.1536). Peak envelopes corresponding to singly charged [M – OMe]+ species were also observed at m/z 611.1334 for M = Ni (calc. m/z 611.1336), m/z 616.1269 for M = Cu (calc. m/z 616.1279) and at m/z 617.1263 for M = Zn (calc. m/z 617.1274), which suggests heterolytic cleavage of the methyl carboxylate ester group during nanospray ionisation. In all cases, the observed and simulated isotopic distributions were in excellent agreement.

James McPherson – June 2019 139 Chapter 6: Miscellanea

6.1.2 NMR Spectroscopy

Only a single dpp– ligand environment was observed in the 1H NMR spectra of the

R1,Me 1 diamagnetic [Zn(dpp )2] (R = CO2Me or Me) complexes. The unambiguous assignment of

1 CO Me,Me the signals in the 300 MHz H NMR spectrum of [Zn(dpp 2 )2] in benzene-d6 (Figure 6.1) was somewhat hindered due to overlapping of the signals attributed to the H6 and H5 environments in the two non-equivalent pyridyl rings of the (dppCO2Me,Me)– ligand. As is shown in Figure 6.1, the two different H3 environments were discriminated from the nOe correlation observed between HC3 and HMe, and the signals attributed to HB4 and HC4 were then identified from the 1H-1H COSY NMR spectrum.

CO Me,Me 1 Figure 6.1: 300 MHz NMR spectra of the [Zn(dpp 2 )2] complex in benzene-d6. Top: H NMR spectrum including an image of the complex showing the numbering scheme and the key nOe correlation distinguished in the 1H-1H nOeSY NMR spectrum in blue. The 1H-1H nOeSY NMR spectrum is shown on the bottom right, and the 1H-1H COSY NMR spectrum is shown on the bottom left.

James McPherson – June 2019 140 Chapter 6: Miscellanea

1 Me,Me The signals in the aromatic region of the 400 MHz H NMR spectrum of [Zn(dpp )2] in acetonitrile-d3 shown in Figure 6.2, were primarily assigned from their major J coupling

3 3 3 constants ( J3,4 ≈ 8 Hz, J4,5 ≈ 7 Hz, and J5,6 ≈ 5 Hz) which were typical for those observed throughout this work and, also more generally, for other metal-coordinated pyridyl rings with substituents at the 2-position.6 These assignments were confirmed by conventional 2D NMR techniques.

1 Me,Me Figure 6.2: The 400 MHz H NMR spectrum of the diamagnetic [Zn(dpp )2] complex in acetonitrile-d3, with an expansion of the aromatic region shown inset with assignments for the signals observed.

1 Me,Me The 400 MHz H NMR spectrum of [Zn(dpp )2] in acetonitrile-d3 acquired at 298 K

(Figure 6.2 and Figure 6.3, top trace) is consistent with a single species with S4 symmetry in solution. Even on cooling the solution to 233 K (see Figure 6.3), there was no evidence for solvato species, similar to those observed for [Ru(dpp)2] discussed in Chapter 5, with lower

CO Me,Me symmetry. This was also the case for [Zn(dpp 2 )2]. This was somewhat unexpected. The first-row d10 zinc(II) ion generally displays faster ligand substitution kinetics than the second-

6 7 row d ruthenium(II) ion. Also, the terminal M–Npy bond lengths observed in solid state structures of this (see below) and other octahedral Zn-dpp complexes across the literature are, on average, approximately 10% longer than those for ruthenium(II) ions (see Chapter 1).8

Hence, ligand hemilability and solvent binding was anticipated.

James McPherson – June 2019 141 Chapter 6: Miscellanea

1 Me,Me Figure 6.3: Stacked 400 MHz H NMR of [Zn(dpp )2] in acetonitrile-d3 acquired at temperatures as indicated, as the solution was cooled from 298 K to 233 K. Note that a break in scale occurs between the aromatic signals (6.70 ≤ δ ≤ 7.68) and the aliphatic region (2.0 ≤ δ ≤ 2.5) of the spectra.

While no direct evidence for the reversible binding of solvent was obtained through these

VT 1H NMR or experiments, the more likely explanation for the symmetric spectra observed is

Me,Me the rapid interconversion of the solvato species with the parent [Zn(dpp )2] complex on the timescale of the 400 MHz 1H NMR experiment, even at 233 K. Experiments with shorter timescales (vibrational or electronic spectroscopy) were not undertaken.

1 R1,Me 1 No signals were observed in the H NMR spectra of the [Cu(dpp )2] (R = CO2Me or

Me) complexes, even when the region –100 ≤ δ ≤ 300 was surveyed, due to the efficient relaxation of nuclear magnetisation via the paramagnetic d9 copper(II) ion. Ten broad, paramagnetically shifted signals (labelled a–j in Figure 6.4) were observed (0 ≤ δ ≤ 130) in the

1 CO Me,Me 300 H NMR spectrum of [Ni(dpp 2 )2] in benzene-d6, consistent with two inequivalent pyridyl rings and two methyl/methyl carboxylate substituents. Only four broad signals were

1 Me,Me observed (5 ≤ δ ≤ 110) in the 400 H NMR spectrum of [Ni(dpp )2] in acetone-d6;

James McPherson – June 2019 142 Chapter 6: Miscellanea presumably the fifth signal was obscured by those of either residual water or protic solvent present in the sample.

1 CO Me,Me Figure 6.4: The 300 MHz H NMR spectrum of the paramagnetic [Ni(dpp 2 )2] complex in benzene-d6, with the signals attributed to the ten 1H environments within the complex labelled a–j. A magnification is also presented above the spectrum showing the very broad signals, a–d, more clearly.

6.1.3 Crystallography

CO Me,Me Me,Me The molecular structures of [Ni(dpp 2 )2], [Cu(dpp )2]·0.25(MeCN) and

Me,Me [Zn(dpp )2]·0.25(MeCN) were investigated by SC-XRD.

CO Me,Me Pale orange needles of [Ni(dpp 2 )2] were obtained from the slow diffusion of pentane into a dichloromethane solution of the complex. A view of the structure of the complex, with salient metric data, is presented in Figure 6.5, and the key metal-ligand bond lengths and angles are summarised in Table 6.1. Both enantiomers of the C2 symmetric complex were

CO Me,Me present in the monoclinic P21/c unit cell. From the structure of [Ni(dpp 2 )2], the averages of the Ni–Npyr and Ni–Npy bond lengths (1.952(3) and 2.298(4) Å, respectively) and of the intra- ligand Npy–Ni–Npyr bond angles (74.1(1)º) are entirely consistent with those data reported for

SCN,SCN 3 Tour’s [Ni(dpp )2] complex. The tetragonal contraction along the Npyr–Ni–Npyr′ axis is a consequence of the dpp– ligand geometry, and results in the distorted octahedral geometry around the nickel(II) ion, with T = 0.849 (for an idea octahedron, T = 1).9 Approximately 45 kJ mol–1 of coordination strain within each (dppCO2Me,Me)– ligand can be predicted from the

James McPherson – June 2019 143 Chapter 6: Miscellanea interatomic separation of the pyridyl N-donor atoms (ρ), according to the relationship discussed

10 5 in Chapter 2, and using ρL(relax) = 4.899 Å calculated as described in Chapter 3.

CO Me,Me Figure 6.5: ORTEP view from the SC-XRD structure of the [Ni(dpp 2 )2] molecule, showing 50% thermal

10 ellipsoids at 100 K with H atoms omitted for clarity. The average Ni–N bond lengths, ∆EL(strain) and a measure of

9 relative tetragonal distortion, Τ = M–Npyr / M–Npy .

〈 〉 〈 〉 Me,Me Brown needles of [Cu(dpp )2]·0.25(MeCN) grew after a hot solution of

Me,Me [Cu(dpp )2] in acetonitrile was cooled and refrigerated at 4 ºC for one week. These needles

Me,Me were suitable for SC-XRD studies which revealed two [Cu(dpp )2] complex molecules in the asymmetric unit of the triclinic P-1 cell, and one disordered acetonitrile molecule at a centre of inversion within the structure. Separate views of each complex molecule are presented in

Figure 6.6, which show one of the (dppMe,Me)– ligands binding Cu1 is 2:1 disordered across two

Me,Me positions. The M–Npyr bonds in these [Cu(dpp )2] complexes (1.89(2) Å on average) were

CO Me,Me shorter than those for the [Ni(dpp 2 )2] complex and, as a consequence, the terminal M–

Me,Me Npy bonds were longer in the [Cu(dpp )2] complexes (2.32(2) Å on average). Thus the more distorted octahedral geometry about the copper(II) complex (T = 0.816) is consistent with the

Me,Me – 5 more “open” geometry of the unbound (dpp ) (ρL(relax) = 5.047 Å) compared to

(dppCO2Me,Me)–, and also is due to the Jahn-Teller distortions associated with octahedral

James McPherson – June 2019 144 Chapter 6: Miscellanea complexes of the d9 copper(II) ion.9, 11-12 The coordination strain (~55 kJ mol–1 average) within

Me,Me – Me,Me –1 the (dpp ) ligands in [Cu(dpp )2] is slightly increased compared to that (~45 kJ mol

CO Me,Me – CO Me,Me average) in the (dpp 2 ) ligands of [Ni(dpp 2 )2] (see above), consistent with the increased distortion for the Cu(II) species.

Me,Me Figure 6.6: ORTEP views of each [M(dpp )2] molecule from the SC-XRD structures, showing 50% thermal ellipsoids, at 100 K for M = Cu(II) (orange) on top, and at 150 K for M = Zn(II) (grey) on the bottom. Acetonitrile solvent molecules and H-atoms are omitted for clarity. Key metric data are presented alongside each complex, with additional data provided in Table 6.1.

James McPherson – June 2019 145 Chapter 6: Miscellanea

R1,R2 Table 6.1: Key bond lengths (Å) and angles (º) from SC-XRD studies of [M(dpp )2] complexes.

CO Me,Me Me,Me Me,Me [Ni(dpp 2 )2] [Cu(dpp )2] [Zn(dpp )2] Distances (Å) ·0.25(MeCN) ·0.25(MeCN)a or Angles (°) 100 K 100 K 150 K Cu1b Cu2c Zn1 Zn2c M–N(pyr) M–N1A /Å 1.957(3) 1.876(6) 1.879(5) 1.923(6) 1.924(5) M–N(py) M–N1B /Å 2.290(3) 2.238(6) 2.255(6) 2.399(6) 2.297(5) M–N(py′) M–N1C /Å 2.283(4) 2.216(7) 2.188(7) 2.278(5) 2.373(7) M–N(pyr) M–N1D /Å 1.947(3) 1.91(2) 1.908(5) 1.937(6) 1.924(5) M–N(py) M–N1E /Å 2.244(3) 2.41(2) 2.234(8) 2.444(6) 2.229(7) M–N(py′) M–N1F /Å 2.375(3) 2.37(2) 2.623(7) 2.306(7) 2.538(7) N1B–M–N1A /° 74.0(1) 75.7(2) 74.8(2) 72.0(2) 75.0(2) Intra-ligand N1C–M–N1A /° 74.7(1) 76.0(3) 76.1(3) 75.4(2) 72.7(2) N(py)–M–N(pyr) N1E–M–N1D /° 74.5(1) 72.7 77.0(3) 72.5(2) 76.7(2) N1F–M–N1D /° 73.3(1) 74.0 70.2(2) 74.8(3) 70.3(2) N1B–M–N1E /° 94.3(1) 94.1 94.0(2) 92.2(2) 95.5(2) Inter-ligand N1B–M–N1F /° 97.8(1) 93.7 95.3(2) 92.9(2) 98.5(2) N(py)–M–N(py′) N1C–M–N1E /° 97.7(1) 94.2 91.7(2) 95.2(2) 89.8(2) N1C–M–N1F /° 87.3(1) 94.1 95.2(2) 97.7(2) 94.0(2) N1B–M–N1D /° 106.0(1) 109.4 111.3(2) 97.3(2) 117.8(2) Inter-ligand N1C–M–N1D /° 105.0(1) 98.9 97.8(3) 115.2(2) 94.5(2) N(py)–M–N(pyr) N1E–M–N1A /° 113.3(1) 108.7 100.2(3) 101.3(2) 103.0(2) N1F–M–N1A /° 100.9(1) 104.8 112.6(2) 111.0(2) 109.5(2) N(pyr)–M–N(pyr′) N1A–M–N1D /° 174.2(1) 174.7 173.3(3) 167.8(2) 167.2(2) N(py)–M–N(py′) N1B–M–N1C /° 148.7(1) 151.7(2) 150.9(2) 147.4(2) 147.6(2) N1E–M–N1F /° 147.6(1) 146.5 147.2(2) 147.2(2) 146.9(2) N(py)…N(py′) ρ1 /Å 4.404(5) 4.319(9) 4.301(9) 4.489(7) 4.485(8) separation ρ2 /Å 4.437(5) 4.58 4.66(1) 4.558(9) 4.571(9) –1 Intra-ligand ∆E1 /kJ mol 45 65 70 50 50 d –1 strain ∆E2 /kJ mol 45 45 35 45 45 M–Npyr /Å 1.952(3) 1.89(2) 1.894(5) 1.930(6) 1.924(5) M–Npy /Å 2.298(4) 2.31(2) 2.325(8) 2.357(7) 2.359(7) 〈 〉 Octahedral M–N /Å 2.183(4) 2.23(2) 2.242(8) 2.273(7) 2.272(7) 〈 〉 Fittinge φ /° 174.2(1) 174.7 173.3(3) 167.8(2) 167.2(2) 〈 〉 θ /° 86.2 89.8 88.4 89.8 89.0 Σ /° 151.2 149.1 140 148.1 148.3 T 0.849 0.818 0.815 0.819 0.816

Me,Me a: Weak diffraction of the [Zn(dpp )2]·0.25(MeCN) crystals resulted in poor quality data (Rint > 27%), and so the distances and angles from the model are listed only to provide approximations and inform qualitative comparisons to the data for other [M(dpp)2] complexes. b: One ligand bound to Cu1 was disordered across two sites. The centroid of each pair of donor atoms was therefore calculated and used for the measurements in this table. c: For Cu2 and Zn1: N1A = N1G, N1B = N1H, N1C = N1I, N1D = N1J, N1E = N1K, and N1F = N1L. d:

10 CO Me,Me – the intra-ligand strain, ∆EL(strain) = 460 × (∆ρ/ρL(relax)), where ρL(relax) = 4.924 Å for (dpp 2 ) or 5.047 Å for

(dppMe,Me)–.5 e: φ and θ are Halcrow’s angular Jahn-Teller distortion parameters; Σ is the sum of |90º – α| for all

9, 13 cis N–M–N angles (α), and Τ = M–Npyr / M–Npy is a measure of relative tetragonal distortion.

〈 〉 〈 〉

James McPherson – June 2019 146 Chapter 6: Miscellanea

Me,Me Pale yellow needles of [Zn(dpp )2]·0.25(MeCN) were also grown from cooling a hot

Me,Me solution of the [Zn(dpp )2] complex in acetonitrile. Only weak diffraction was observed during SC-XRD experiments with these crystals. Nonetheless, the observed diffraction data

Me,Me was refined to give a structure for the [Zn(dpp )2] complex molecule very similar to that

Me,Me found for its copper analogue in the structure of [Cu(dpp )2]·0.25MeCN (see Figure 6.6 and

Tables 6.1 and 6.2).

R1,R2 Table 6.2: SC-XRD data and refinement parameters for [M(dpp )2] complexes

CO Me,Me Me,Me Me,Me Complex [Ni(dpp 2 )2] [Cu(dpp )2]·0.25MeCN [Zn(dpp )2]·0.25MeCN Empirical formula C34H28N6NiO4 C65H57.5Cu2N12.5 C130H115N25Zn4 Formula weight 643.33 1140.81 2288.94 Temperature/K 100.15 100.15 150.09 Crystal system monoclinic triclinic triclinic Space group P21/c P-1 P-1 a /Å 9.7340(19) 8.8030(18) 8.9208(13) b /Å 18.478(4) 17.270(4) 17.156(3) c /Å 16.494(3) 18.424(4) 18.468(3) α /° 90 98.79(3) 99.058(10) β /° 97.40(3) 98.96(3) 98.372(10) γ /° 90 92.68(3) 92.914(10) Volume /Å3 2942.0(10) 2727.1(10) 2753.6(8) Z 4 2 1 -3 ρcalc /g cm 1.452 1.389 1.380 Μ /mm-1 0.711 0.835 0.926 F(000) 1336.0 1186.0 1190.0 0.08 × 0.02 × Crystal size /mm3 0.2 × 0.05 × 0.05 0.5 × 0.05 × 0.05 0.02 MoKα (λ = Radiation MoKα (λ = 0.7107) MoKα (λ = 0.71073) 0.71073) 2θ range for data collection /° 3.326 to 63.716 2.268 to 52.742 5.468 to 50.054 -11 ≤ h ≤ 11 -11 ≤ h ≤ 11, -10 ≤ h ≤ 10, Index ranges -24 ≤ k ≤ 24 -19 ≤ k ≤ 19, -20 ≤ k ≤ 20, -21 ≤ l ≤ 21 -23 ≤ l ≤ 23 -21 ≤ l ≤ 21 Reflections collected 54972 38781 39468 7518 10162 9722 Independent reflections [Rint = 0.1071, [Rint = 0.0265, [Rint = 0.2562, Rsigma = 0.0529] Rsigma = 0.0207] Rsigma = 0.2960] Data/restraints/parameters 7518/0/411 10162/160/721 9722/111/737 Goodness-of-fit on F2 1.039 1.173 0.941 R1 = 0.0624, R1 = 0.0845, R1 = 0.0728, Final R indexes [I ≥ 2σ (I)] wR2 = 0.1399 wR2 = 0.2662 wR2 = 0.1239 R1 = 0.1094, R1 = 0.0876, R1 = 0.2241, Final R indexes [all data] wR2 = 0.1662 wR2 = 0.2674 wR2 = 0.1743 Largest diff. peak/hole / e Å–3 0.96/-1.00 1.80/-0.88 0.40/-0.55

James McPherson – June 2019 147 Chapter 6: Miscellanea

6.1.4 Electrochemistry

R1,Me 1 The redox properties of [M(dpp )2] (M = Zn or Cu; R = CO2Me or Me) in dichloromethane were surveyed by cyclic voltammetry with tetrabutylammonium hexafluorophosphate (0.1 M) as the supporting electrolyte. The cyclic voltammograms (CVs) are presented in Figure 6.7 with key redox-potential data listed in Table 6.3. Two sequential and reversible processes, labelled I and II in Figure 6.7, were observed on scanning anodically

R1,Me 10 in the [Zn(dpp )2]. The d zinc(II) ion is redox-inert and, therefore, processes I and II are attributed to oxidation of each of the dpp– anions. Oxidation of the more electron rich ligands

1 +/0 (R = Me) occurred at lower potentials (E1/2 (I) = 0.084 V vs. Fc , and E1/2 (II) = 0.255 V vs.

+/0 1 Fc ) than those for the ligands with electron-withdrawing ester group (R = CO2Me: E1/2 (I) =

+/0 +/0 0.455 V vs. Fc , and E1/2 (II) = 0.602 V vs. Fc ). The potentials of these dpp-centred processes were significantly lower than was observed for the same processes in the analogous

R1,Me +/0 [Co(dpp )2] complexes (with E1/2 (I) = 0.70 or 1.02 V vs. Fc and E1/2 (II) = 0.84 or 1.12

+/0 1 V vs. Fc , for R = Me or CO2Me, respectively) under the same conditions, as discussed in

Chapter 3.5 This is consistent with the different oxidation states of the metal centres, with oxidation of the cobalt(II) ion to cobalt(III) (E1/2 {Co(II)/Co(III)} = –0.949 or –0.829 V vs.

+/0 1 5 – Fc , for R = Me or CO2Me, respectively) preceding oxidation of the dpp ligands, and thus increasing the potentials of the ligand centred oxidation couples, relative to those for the ligands bound to the divalent zinc(II) ion.

James McPherson – June 2019 148 Chapter 6: Miscellanea

R1,Me 1 Figure 6.7: CVs for [M(dpp )2] complexes (M = Zn on the left, M = Cu on the right; and R = CO2Me in blue,

Me in red) in dichloromethane, with [Bu4N][PF6] supporting electrolyte at 298 K. Other conditions: 1.0 mm

+/0 diameter glassy carbon minidisk working electrode, ∆Ep (Fc ) = 74 mV. Redox potentials are summarised in Table 6.3.

R1,Me Table 6.3: Redox potentials (V) observed for the [M(dpp )2] complexes (M = Zn on the left, M = Cu on the 1 +/0 right; and R = CO2Me in blue, Me in red) in CH2Cl2 with 0.1 M [Bu4N][PF6], vs. E1/2 Fc at a 1.0 mm glassy carbon minidisk electrode, 100 mV s–1 and 298 K.

Me,Me CO Me,Me Me,Me CO Me,Me [Zn(dpp )2] [Zn(dpp 2 )2] [Cu(dpp )2] [Cu(dpp 2 )2] E1/2 (I) 0.084 0.455 0.043 0.402 E1/2 (II) 0.255 0.602 0.321 0.674 Ep, c {Cu(II)/Cu(I)} - - –1.59 –1.69

– R1,Me The dpp ligand-centred oxidation processes were also observed for the [Cu(dpp )2] complexes (Figure 6.7 on the right), centred at similar potentials (average E1/2 = 0.182 or 0.538

+/0 1 V vs. Fc for R = Me or CO2Me, respectively) but with much larger separation (E1/2(II) –

1 R1,Me E1/2(I) = 0.278 or 0.272 V; for R = Me or CO2Me, respectively) compared to the [Zn(dpp )2]

+/0 complexes (with average E1/2 = 0.170 or 0.529 V vs. Fc ; and E1/2(II) – E1/2(I) = 0.171 or 0.147

1 V; for R = Me or CO2Me, respectively). The increased E1/2(II) – E1/2(I) values for the Cu(II) complex over the Zn(II) complex is suggestive for stronger binding to the Cu(II) ion (so consistent with the Irving-William series14 and the SC-XRD structural data, Table 6.1) leading to better communication between the bound ligands. The Cu(II)/Cu(I) couple for each

James McPherson – June 2019 149 Chapter 6: Miscellanea

R1,Me +/0 1 [Cu(dpp )2] complex was typically broad, at ca. –1.59 or –1.69 V vs. Fc for R = Me or

CO2Me, respectively. The broad reduction couples are indicative for electron transfer concurrent and dependent on slow structural reorganisation; i.e. the Cu(I) and Cu(II) centres differ in structure as anticipated (Cu(I) is expected to exhibit lower coordination numbers).15-17

Why the Cu(II)/Cu(I) couple should be less reversible, i.e. the gating structural change slower,15

Me,Me for [Cu(dpp )2] is unknown.

Me,Me The electrochemical response of [Cu(dpp )2] under CO2 was also investigated. No significant current enhancement was observed at the potentials of the Cu(II)/Cu(I) processes when the CVs were recorded after purging the solutions with CO2 for 15 minutes (e.g., see the

Me,Me teal traces in Figure 6.8 for [Cu(dpp )2]). However, the addition of 2,2,2-trifluoroethanol

(TFE, 0.22 M) under CO2 resulted in two new reduction processes, R1 at –0.85 and R2 at –1.22

V vs. Fc+/0, followed by a large reduction feature suggestive for a catalytic wave with

red +/0 E onset ~ –1.6 V vs. Fc . The large reduction wave was not observed when the Cu complex was omitted. Further experiments are required to confirm the reaction products, including

+ whether H or CO2 is reduced, and the active catalytic species (which could be homogenous or heterogenous and molecular or non-molecular).

Me,Me Figure 6.8: CVs of [Cu(dpp )2] in 0.1 M [Bu4N][PF6]–CH2Cl2 at a glassy carbon minidisk electrode and 298

K, under atmospheres of: N2 (g) in red, CO2 (g) in teal, and with added 2,2,2,-trifluoroethanol (TFE, 0.22 M) under under CO2 (g) in purple. The response of TFE (0.22 M) in 0.1 M [Bu4N][PF6]–CH2Cl2 under CO2 (g) in the absence Me,Me of [Cu(dpp )2] is also presented in black in each panel. An expansion is shown on the right.

James McPherson – June 2019 150 Chapter 6: Miscellanea

R1,Me The CVs for the [Ni(dpp )2] complexes are presented in Figure 6.9, and show the

+/0 1 reversible Ni(II)/Ni(III) couples (E1/2 = –0.142 or 0.080 V vs. Fc for R = Me or CO2Me,

SCN,SCN 3 respectively) similarly to as reported by Tour et al. for [Ni(dpp )2]. Scanning to higher potentials than surveyed by Tour et al. reveals a second, reversible single-electron oxidation

+/0 1 process, O2 (E1/2 = 0.530 or 0.988 V vs. Fc for R = Me or CO2Me, respectively). These potentials are appreciably lower than was observed for the first oxidation of the (dppR1,Me)–

+/0 1 ligands bound to the trivalent cobalt(III) ion (E1/2 = 0.70 or 1.02 V vs. Fc for R = Me or

5 R1,Me R1,Me CO2Me, respectively) as discussed in Chapter 3. All other [M(dpp )2] and [Ln(dpp )3]

1 complexes (where R = CO2Me or Me) studied during this PhD research have displayed sequential, and reversible, single-electron oxidations for each of the bound dpp– ligands. It is therefore likely that O2 is a nickel-centred process, i.e. the Ni(III)/Ni(IV) couple, with the second of the ligand-centred processes moved anodically outside the solvent discharge limit.

+/0 The third irreversible single-electron oxidation process, O3 (at Ep, a = 1.148 or 1.288 V vs. Fc

1 – for R = Me or CO2Me, respectively), is likely the first of the dpp centred oxidations. Scanning through process O3 reduces the cathodic current at process O2, with emergence of a new cathodic process, R*, for the reduction of a (decomposition) product species.

R1,Me 1 Figure 6.9: CVs for [Ni(dpp )2] complexes (R = CO2Me in blue, Me in red) in dichloromethane, with

[Bu4N][PF6] supporting electrolyte at 298 K. Other conditions: 1.0 mm diameter glassy carbon minidisk working

+/0 –1 electrode, ∆Ep (Fc ) = 74 mV, and ν = 100 mV s . Redox potentials are summarised in Table 6.4.

James McPherson – June 2019 151 Chapter 6: Miscellanea

R1,Me 1 Table 6.4: Redox potentials (V) observed for the [Ni(dpp )2] complexes (R = CO2Me in blue, Me in red) in +/0 –1 CH2Cl2 with 0.1 M [Bu4N][PF6], vs. E1/2 Fc at a 1.0 mm glassy carbon minidisk electrode, 100 mV s and 298 K.

Me,Me CO Me,Me [Ni(dpp )2] [Ni(dpp 2 )2] E1/2 –0.142 0.080 {Ni(II)/Ni(III)} E1/2 (O2) 0.530 0.988 Ep, a (O3) 1.148 1.288

6.1.5 Conclusions

R1,Me 1 The preparations of [M(dpp )2] (M = Ni(II), Cu(II) or Zn(II), R = CO2Me or Me)

H,H complexes were facile, and, in contrast to Tour’s report for [Zn(dpp )2] which was unstable to light and acid,3 the complexes prepared in this work were all stable under ambient conditions in both the solid and solution states. This is possibly due to the substituents at the 3- and

4-positions at the pyrrolide which block oxidative polymerisation, consistent with the reversible ligand centred oxidation chemistry observed in the CVs. These complexes retain the redox processes at the metal centres were influenced by the pyrrolide substituents, with the more electron rich methyl group more stabilising higher nickel oxidation states (i.e., Ni(III) and

Ni(IV)). The [Cu(dpp)2] complexes appear to function as electrocatalysts for the electroreduction of protons or CO2, however the exact product(s) and the active species are unknown. Further studies are suggested.

6.2 [M(dpp)2] Metalloligands

6.2.1 Introduction

Hein’s pioneering report of the synthesis of HdppCO2Et,CO2Et also included several post- synthetic modifications to the pyrrolide substituents. Of particular interest was hydrolysis of the two ethyl ester substituents to carboxylic acid groups. The electronic properties of

James McPherson – June 2019 152 Chapter 6: Miscellanea carboxylic acids change dramatically depending on protonation state: when protonated, the

Swain-Lupton parameters are identical to those of the ethyl ester (with F = 0.34 and R = 0.11), but the deprotonated carboxylates are more electron rich (with F = –0.10 and R = 0.10).18 This should result in redox potentials for metal ions bound by this ligand which depend on the protonation state of the acid groups. The relationship discovered in Chapter 3 relating the

R1,R2 potential of the Co(III)/Co(II) couple for [Co(dpp )2] complexes and a linear combination of F and R for the R1 and R2 substituents5 thus predicts up to five possible such potentials

CO ,CO n – 4 (across a range of 0.8 V) for the different protonation states of the Hn[Co(dpp 2 2)2]

(n = 0, 1, 2, 3, or 4) species shown in Figure 6.10. Such redox control through environmental pH (or pH control through redox environment) could lead to interesting molecular devices and applications—e.g., a stoichiometric, single-electron reductant that can be switched on or modulated through careful control of pH. The magnetic properties (i.e. spin state and/or SCO) of the cobalt(II) ions might also be sensitive to the protonation state of these acid groups.

Moreover, supramolecular chemists have also used peripheral carboxylate substituents extensively to construct large molecular cages and networks. In particular, metal organic frameworks (MOFs) often feature metal-oxide structural building units linked by organic molecules bearing carboxylate groups,19 and so it was envisaged that metal complexes of the di-carboxylate (dppCO2,CO2)3– trianion may be useful metalloligands for supramolecular construction, as suggested in Figure 6.10. The geometry of these 4-connecting carboxylate

CO ,CO z– – [M(dpp 2 2)2] linking units are unique: the angles between the two CO2 units within each ligand are predicted to be approximately 72º due to the 5-membered pyrrolide ring, and the two ligand planes are perpendicular to one another due to the octahedral coordination geometry about the metal. It is not at all clear what framework geometries may emerge when using these linkers. Rigid frameworks may present opportunities to achieve more abrupt SCO in the

[Co(dpp)2] units discussed in Chapter 3, while well defined cavities within the frameworks

James McPherson – June 2019 153 Chapter 6: Miscellanea

might modulate the reaction outcomes for the hemilabile [Ru(dpp)2] catalysts discussed in

Chapter 4.

CO ,CO n – 4 Figure 6.10: Top: the five different protonation states of Hn[Co(dpp 2 2)2] (n = 0, 1, 2, 3, or 4), with the

+/0 potentials (versus Fc in 0.1 M [Bu4N][PF6]–CH2Cl2) of the Co(III)/Co(II) couple predicted by 5 CO ,CO E1/2 = –1.12 + 0.487 F – 0.181 R, as derived in Chapter 3. Bottom: an illustration of [M(dpp 2 2)2] as a supramolecular building block.

CO ,CO n – 3 At the outset, it was not clear whether the (Hndpp 2 2) ligand would bind transition metal ions: its coordination chemistry had not been previously investigated. This work describes modification of Hein’s post synthetic transformations of the di-ester precursor,

CO Et,CO Et III CO ,CO Hdpp 2 2 , and the preparation of the first [Co (dpp 2 2)2] species bearing unbound carboxylate groups.

James McPherson – June 2019 154 Chapter 6: Miscellanea

6.2.2 Ligand Synthesis

The di-ester substituted pyrrole, HdppCO2Et,CO2Et, was prepared either directly by the

McSkimming et al. one pot synthesis,4 or on larger scale, across several steps, as reported by

1 CO Et,CO Et Hein et al. The two ethyl ester groups in Hdpp 2 2 were hydrolysed in refluxing H2SO4

(9 M, aq).1 On cooling to room temperature, the yellow solution was partially neutralised with

CO ,CO NaHCO3 (sat, aq) to pH 3, and the yellow solid, NaδH3–δdpp 2 2, product collected. As is

CO ,CO clear from the ESI-MS of NaδH3–δdpp 2 2, and the analysis of the subsequent cobalt complex prepared with this ligand (discussed below), some sodium (δ) has been carried through from

(CO) O this work-up. The carboxylic groups could be converted to the anhydride, NaδH1–δdpp 2 , by

CO ,CO dissolving NaδH3–δdpp 2 2 in acetic anhydride, and heating at reflux for one hour;

(CO) O 1 NaδH1–δdpp 2 was collected from the mixture after it was cooled to room temperature.

These products were poorly soluble in water and methanol, but sufficiently so for ESI-MS analyses when aspirated in methanol, and analysis by 1H and 13C{1H} NMR spectroscopy was also possible in DMSO-d6.

Scheme 6.1: Post-synthetic modification of HdppCO2Et,CO2Et, as originally reported by Hein.1

CO ,CO In the mass spectrum of NaδH3–δdpp 2 2 (Figure 6.11), peak envelopes corresponding

CO ,CO + to the singly charged proton- or sodium-adducts [H3dpp 2 2 (M) + H] (calc. m/z 310.08) and [M + Na]+ (calc. m/z 332.06) were observed at m/z 310.08 and m/z 332.08, respectively. A

James McPherson – June 2019 155 Chapter 6: Miscellanea peak envelope consistent with the sodium-adduct of a singly charged dimer [2M + Na]+ (calc. m/z 641.14) was also observed at m/z 641.25.

CO ,CO Figure 6.11: Low resolution ESI mass spectrum for H3dpp 2 2 aspirated in methanol.

Peak envelopes observed at m/z 324.08 and m/z 346.08 in the ESI-MS of the anhydride

(CO) O derivative, NaδH1–δdpp 2 (Figure 6.12), were consistent with proton- and sodium-adducts of the pyrrole with a molecule of methanol [Hdpp(CO)2O (M) + MeOH + H]+ (calc. m/z 324.10) and [M + MeOH + Na]+ (calc. m/z 346.08) respectively, which suggests ring-opening methanolysis of the anhydride during ESI. Peaks at m/z 669.25, consistent with the sodium- adduct of the dimeric [2(M + MeOH) + Na]+ (calc. m/z 669.17), were also observed.

James McPherson – June 2019 156 Chapter 6: Miscellanea

Figure 6.12: Low resolution ESI mass spectrum for Hdpp(CO)2O aspirated in methanol.

1 CO Et,CO Et CO ,CO The 400 MHz H NMR spectra for Hdpp 2 2 , NaδH3–δdpp 2 2 and NaδH1–

(CO) O δdpp 2 are presented in Figure 6.13. Each spectrum is consistent with a different (Na/H)dpp species, with two-fold symmetry. The signals in the aromatic region of the spectra were

1 3 3 assigned to pyridyl H environments from their major JH,H coupling constants ( J3,4 ≈ 8 Hz,

3 3 6 CO ,CO J4,5 ≈ 7 Hz, and J5,6 ≈ 5 Hz). In the spectrum of NaδH3–δdpp 2 2 (middle trace, Figure

6.13), cleavage of the ester was confirmed by the absence of any signals in the aliphatic region of the spectrum. Although a signal at δ 12.35 assigned to the pyrrole NH was observed, the protonation state of the carboxylic acids is uncertain due to the presence of, and exchange with,

1 (CO) O water in the solution. The H NMR spectrum of anhydride NaδH1–δdpp 2 in DMSO-d6

CO ,CO (bottom trace, Figure 6.13) is similar to that of NaδH3–δdpp 2 2, but with the largest difference for the signal assigned to HB3, which was shifted downfield by 0.20 ppm, and no signal for the pyrrole NH environment was observed.

James McPherson – June 2019 157 Chapter 6: Miscellanea

1 CO Et,CO Et CO ,CO Figure 6.13: Stacked 400 MHz H NMR spectra of Hdpp 2 2 (top trace), NaδH3–δdpp 2 2 (middle trace)

(CO) O and NaδH1–δdpp 2 (bottom trace) in DMSO-d6.

No evidence for hydrogen-bonded O–H groups (including water) were detected in the

CO ,CO FT-IR spectrum of NaδH3–δdpp 2 2, while the strong signals observed at 1810.7 and

–1 (CO) O 1755.7 cm in the FT-IR spectrum of NaδH1–δdpp 2 are characteristic for the symmetric and asymmetric stretching modes, respectively, of the carbonyl groups in cyclic carboxylic anhydrides.

CO2,CO2 + 6.2.3 [Co(H2dpp )2]

CO ,CO Reaction of the diacid, NaδH3–δdpp 2 2, with Co(OAc)2·4H2O in refluxing methanol resulted in a brown solid, as is shown in Scheme 6.2.

James McPherson – June 2019 158 Chapter 6: Miscellanea

CO ,CO + Scheme 6.2: Synthesis of [Co(H2dpp 2 2)2]

Analysis of the product by 1H NMR spectroscopy (Figure 6.14, top panel) revealed a

6 diamagnetic species, therefore a LS d Co(III) species, and consistent with S4 molecular

CO ,CO + z+ symmetry for [Co(NaδH2–δdpp 2 2)2] . As with the other [M(dpp)2] complexes prepared

1 CO ,CO + during this project, the signal in the H NMR spectrum of [Co(NaδH2–δdpp 2 2)2] that was assigned to the HB3 environment was observed downfield at δ 9.39, while the signal assigned to HB6 at δ 6.86 was more shielded, relative to the signals assigned to the same environments in the spectrum of the unbound ligand (at δ 8.13 and δ 8.68 for HB3 and HB6, respectively).

In the 13C{1H} NMR spectrum (Figure 6.14, middle trace), assignment of the signals to the CB3, CB4, CB5, and CB6 environments were trivial from the 1H-13C HSQC NMR spectrum

(not shown). Correlations between the signal at δ 157.73 in the 13C dimension and the signals assigned to HB3, HB4 and HB6 in the 1H-13C HMBC NMR spectrum (Figure 6.14, bottom left panel) were consistent with its assignment as CB2. A weak correlation in the 1H-13C HMBC

NMR spectrum (Figure 6.14, bottom left panel) was also observed between the signal assigned to HB3 and the signal at δ 142.66, which was thus assigned to CA2. While no correlations involving the signals at δ 166.53 or at δ 126.35 were observed in the 1H-13C HMBC NMR spectrum, these could be assigned to the CA3′ and CA3 environments, respectively, from their chemical shifts.

James McPherson – June 2019 159 Chapter 6: Miscellanea

1 13 1 CO ,CO + Figure 6.14: Assignment of the H and C{ H} spectra of [Co(NaδH2–δdpp 2 2)2] in DMSO-d6. Top panel: the 400 MHz 1H NMR spectrum with numbering scheme and assignments indicated. Middle panel: the 151 MHz 13C{1H} NMR spectrum. Bottom panels: the 1H-13C HMBC (left) and 1H-15N HMBC (right) spectra (at 600 MHz, 151 MHz and 61 MHz, for 1H, 13C and 15N, respectively), with key correlations identified.

James McPherson – June 2019 160 Chapter 6: Miscellanea

In the 1H-15N HMBC NMR spectrum (Figure 6.14, bottom right panel), a correlation observed between the signal assigned to HB6 and a 15N environment at δ ~199, consistent with

NB1 coordinated to a transition metal ion.

In the (+)-ESI-MS (Figure 6.15), a number of singly-charged, mono-nuclear species were

CO ,CO + + observed. The molecular ion, [Co(H2dpp 2 2)2] ([M] calc. m/z 675.07) at m/z 675.17 with relative abundance ~65%, was the second most abundant species observed, with the most

+ abundant peak envelope centred at m/z 631.17 consistent with a decarboxylated [M – CO2] species (calc. m/z 631.08). A peak envelope at m/z 587.17 with relative abundance ~50% was

+ consistent with a twice-decarboxylated [M – 2CO2] species (calc. m/z 587.09). Other low intensity signals (with relative abundances below 15%) consistent with the substitution of a sodium atom for a hydrogen in the molecular ion or decarboxylated ions were also observed, as is shown in Figure 6.15.

CO ,CO + Figure 6.15: The (+)-LR-ESI-MS of the [Co(H2dpp 2 2)2] complex aspirated in methanol.

James McPherson – June 2019 161 Chapter 6: Miscellanea

CO ,CO Long, fine needles of polymeric {Na(MeOH)[Co(Hdpp 2 2)2]}n·7(DMF) were grown

CO ,CO + by slow diffusion of diethyl ether into a solution of the [Co(NaδH2–δdpp 2 2)2] complex in

N,N-dimethylformamide (DMF). Only very weak diffraction (I/σ = 2.8) was observed during

SC-XRD experiments with these crystals. While the data were insufficient to quantify geometric properties (Rint = 40.9%), a model was refined (to R1 = 15.4% for I ≥ 2σ(I), wR2 = 45.2% for all data) to confirm the connectivity within the triclinic P-1 unit cell. As shown

CO ,CO + in Figures 6.16 and 6.17, the refined structure confirms two different [Co(Hdpp 2 2)2] complex monomers linked through interactions between the carboxylate groups and sodium

CO ,CO ions, to form a 2-dimensional polymer, {Na(MeOH)[Co(Hdpp 2 2)2]}n. Each sodium ion is bound to a solvent molecule, modelled in the refined structure as methanol. Due to the weak diffraction data, H-atoms were not included in the model during refinement, and the protonation state of the carboxylate groups remains uncertain.

The short Co–Npyr and Co–Npy bonds measured in both molecules (ca. 1.7 Å and 1.9 Å,

III + respectively) were consistent with those measured for the one other [Co (dpp)2] structure

20 previously reported (with 〈Co–Npyr〉 = 1.810(8) Å and 〈Co–Npy〉 = 1.963(8) Å), while the same

II bonds in [Co (dpp)2] complexes are typically much longer (Co–Npyr > 1.9 Å and

5, 8 Co–Npy > 2.2 Å).

CO ,CO As shown in Figure 6.17, these networks of {Na(MeOH)[Co(Hdpp 2 2)2]}n polymer form pseudo two-dimensional sheets formed of complex molecules interconnected through carboxylate-sodium ion interaction, that stack along the crystallographic c axis. A large, solvent

3 accessible void (with Vvoid = 1187 Å , 32% of Vcell) was masked during refinement and contained 284 electrons—equivalent to approximately seven DMF molecules (280 electrons).

James McPherson – June 2019 162 Chapter 6: Miscellanea

CO ,CO Figure 6.16: Two ball-and-stick views from the SC-XRD structure for {[Co(Hdpp 2 2)2]Na}n showing each

CO ,CO + {[Co(Hdpp 2 2)2] unit and neighbouring sodium ions bound to the carboxylates. The averages of the Co–Npyr

z+ and Co–Npy bond lengths are presented to inform qualitative comparisons to those for other [Co(dpp)2] (z = 0 or 1) geometries only.

Figure 6.17: From left to right: views down the a, b and c crystallographic axes, respectively, from the SC-XRD CO ,CO structure of {Na(MeOH)[Co(Hdpp 2 2)2]}n. The octahedra around Co1 are shown in blue, and those around Co2 in pink, and the mean plane through each polymeric 2D sheet is shown in grey. The separation of each “plane” (in Å) shown in the view down the b axis.

James McPherson – June 2019 163 Chapter 6: Miscellanea

6.2.4 Conclusions

As reported more than 50 years ago,1 the hydrolysis of the ester-substituted

CO Et,CO Et CO ,CO Hdpp 2 2 under acidic conditions is facile, and affords the diacid H3dpp 2 2. The diacid

CO ,CO – pyrrolide, (H2dpp 2 2) , binds a cobalt(III) ion through the three N-donor atoms selectively

+ over the carboxylate O-donor atoms, forming a [Co(dpp)2] complex bearing peripheral carboxylate groups, and further studies on the pH dependence of the Co(III)/Co(II) redox couple are of interest. The peripheral carboxylate groups resulted in a porous polymeric material, with

+ [Co(dpp)2] units linked through sodium ion-carboxylate interactions. Further studies will seek to employ more rigid linkages, to form frameworks that don’t collapse on desolvation, and also investigate other transition metal ions {e.g., ruthenium(II)} at the centre of the [M(dpp)2] metalloligand. Such materials may have interesting physical properties, e.g. cooperative SCO in rigid cobalt(II) frameworks.

CO ,CO (CO) O Ring closure of the diacid H3dpp 2 2 affords the anhydride H1dpp 2 , which represents a useful synthon for further post-synthetic transformations on the Hdpp scaffold, and may also be an interesting ligand precursor in its own right (e.g., as “protected” carboxylate groups).

6.3 Organometallic Piano-stool [Cp*M(κ2-dpp)Cl] Complexes for Multi- electron Reduction

Increasing global energy requirements demand new catalysts capable of multi-electron processes. Current industrial multi-electron processes use molecular hydrogen as a source of two electrons, which is produced from the steam reformation of methane. Although efficient, these current processes are energy intensive, and rely upon the consumption of fossil carbon reserves with the release of carbon dioxide. Alternative multi-electron reducing agents, or new

James McPherson – June 2019 164 Chapter 6: Miscellanea strategies to produce molecular hydrogen, are of considerable interest. Nature offers attractive insights. Rather than molecular hydrogen, Nature transfers hydride (formally a proton and two electrons) to unsaturated substrates. As shown in Figure 6.18, Nature mostly uses NADH as a proton-dependent source of two electrons and NADPH mostly as a direct hydride donor (i.e. the two cofactors have distinct roles).21

Figure 6.18: Top left, Nature’s multi-electron redox cofactors, NAD(P)H/NAD(P)+, which store and deliver hydride (H– = H+ + 2e–) to substrates bound and activated at metal centres in metalloenzymes. Top right, Steckhan’s 2+ prototypal [Cp*Rh(bpy)(H2O)] complex which artificially—electrochemically—regenerates NADH. Middle right are two Organic Hydride Donors: 1,3-dimethyl-2,3-dihydrobenzoimidazoline (biH) and Hantzsch’s Ester (HEH) in blue, and their conjugate hydride acceptors, bi+ and HE+ in red. Bottom are two recent examples of complexes prepared within the Colbran group incorporating organic hydride acceptors (OHD+ shown in red) tethered directly to [Cp*Rh(N,N)Cl]+ units, which have demonstrated activities as catalysts for the electroreduction of protons and for the transfer of hydrogen (H+ + H–) from formate to imines.22-23

James McPherson – June 2019 165 Chapter 6: Miscellanea

Simple transition metal complexes of the form [Cp*M(diimine)X]z+

(Cp* = 1,2,3,4,5-pentamethylcyclopentadienyl; M = Rh(III) or Ir(III); diimine is a diimine ligand, typically 2,2′-bipyridine; and X is either a solvent or halide co-ligand) have been used as electrocatalysts for transfer hydride ion (H+ + 2e–) to oxidised NAD(P)+ to thus regenerate reduced NAD(P)H non-naturally.

The Colbran group have recently been interested in tethering organic hydride donor

(OHD-H) mimics of NAD(P)H (e.g. Hantzsch ester or dihydrobenzo-[d]-imidazoline) units— or in their conjugate hydride acceptor forms (OHD+) as shown in Figure 6.18—directly to diimine groups that bind Cp*M domains in the pursuit of new (electro)catalysts for multi- electron reduction of challenging substrates, such as CO2 (g).

The facial Cp* arene ligand is incompatible with the meridional tridentate binding mode of the dpp– ligand. This was seen as a potential advantage. It was anticipated that the dpp– ligand would be forced to bind in a bidentate fashion, as has been observed for similar complexes with

24 terpyridyl ligands and for the hemilabile [Ru(dpp)2] complexes, see Chapter 5 above. Such a binding mode leaves a pyridyl group unbound, which could impart interesting redox or proton responsive activity close to the substrate binding site at the metal, as summarised in Figure 6.19.

Figure 6.19: The key design elements of the neutral [Cp*M(κ2-dpp)Cl] complexes described in this work.

James McPherson – June 2019 166 Chapter 6: Miscellanea

6.3.1 Synthesis and Characterisation

The [Cp*M(κ2- dppR1,R2)Cl] complexes (M = Rh(III) or Ir(III), and R1,R2 = Me,Me;

CO2Me,Me; or CO2Et,CO2Et) described in this work were prepared from the straightforward reaction of the appropriate Hdpp precursor and [Cp*MCl2]2 dimer in dichloromethane, with triethylamine present to accept the acidic pyrrolyl proton, as summarised in Scheme 6.3. The symmetric dpp– ligands yielded racemic mixtures of the enantiomeric [Cp*M(κ2-dppR1,R2)Cl] products, while the non-symmetric (dppCO2Me,Me)– afforded both possible structural isomers of

[Cp*M(κ2-dppCO2Me,Me)Cl], shown in Figure 6.20.

1 2 Scheme 6.3: The preparation of [Cp*M(κ2-dppR ,R )Cl] complexes.

Figure 6.20: Two structural isomers of [Cp*M(κ2-dpp)Cl] complexes are formed with non-symmetric dpp– ligands such as (dppCO2Me,Me)– shown.

These new [Cp*M(κ2-dpp)Cl] complexes were characterised by 1H and 13C NMR spectroscopy and high-resolution mass spectrometry.

James McPherson – June 2019 167 Chapter 6: Miscellanea

6.3.1.1 NMR Spectroscopy

The 1H and 13C{1H} NMR spectra (at 600 MHz and 151 MHz, respectively) of

[Cp*Rh(κ2-dppMe,Me)Cl] in chloroform-d were unambiguously assigned using conventional 2D experiments (applying 1H-1H COSY and nOeSY, 1H-13C HSQC and HMBC NMR

– spectroscopy) as shown in Figure 6.21. The C2v symmetry of the dpp ligand is broken upon bidentate binding and the two different pyridyl spin systems in the (κ2-dppMe,Me)– ligand (B and

B′) were discriminated through either 1H-1H COSY NMR or, as shown in Figure 6.21 (bottom right panel), 1H-1H nOeSY NMR correlations. The signals assigned to H6 in the bound (B) and unbound (B′) pyridyl rings were identified through 1H-15N HMBC NMR correlations as shown in Figure 6.21 (bottom left panel). The signal at δ 8.36 in the 1H dimension correlated with a

15N environment at δ ~228—typical25 for pyridyl N atoms coordinated to transition metal ions—assigned to (HB6, NB1), while the signal at δ 8.64 in the 1H dimension correlated to a 15N environment at δ ~305 as expected for the HB6′ proton environment in the unbound pyridyl ring.

These assignments were further confirmed through inspection of the 1H-1H nOeSY NMR spectrum, Figure 6.21 (bottom right panel), with through space nOe correlations observed between the signals assigned to HB6 in the bound pyridyl ring and to HD1′ in the Cp* ring, and with nOe correlations between the signals for HB3 and HA3′ also observed.

Although correlations due to chemical exchange were observed in the 1H-1H nOeSY

NMR spectrum of [Cp*Rh(κ2-dppMe,Me)Cl] in chloroform-d acquired at 400 MHz and 328 K, this chemical exchange is slow on the timescale of the 1H NMR experiment, with no coalescence observed across the spectra at temperatures up to 328 K (see Figure 6.22). This is in stark contrast to [Cp*Rh(κ2-tpy)Cl] complexes,24, 26-27 which undergo rapid exchange at room temperature, and must be cooled to 213 K for the different signals in the 1H NMR spectra

24 (200 MHz, methanol-d4) to be resolved.

James McPherson – June 2019 168 Chapter 6: Miscellanea

Figure 6.21: Example of NMR spectroscopic characterisation: spectra of the [Cp*Rh(κ2-dppMe,Me)Cl] complex in chloroform-d. Top: 600 MHz 1H NMR spectrum, including an expansion over the aromatic region and an image from the SC-XRD structure inset showing the numbering scheme and the key correlations distinguished in the 2D spectra below (1H-15N correlations in purple, and 1H-1H nOeSY in blue and 1H-1H EXSY in red). The 1H-15N HMBC spectrum (acquired at 600 MHz and 61 MHz for 1H and 15N, respectively) is shown in the bottom left, and the 400 MHz 1H-1H nOeSY spectrum is on the bottom right, with through space (nOe) correlations indicated in blue whereas correlations due to chemical exchange (distinguished by identical phases for the diagonal and cross- peak signals) are indicated in red.

James McPherson – June 2019 169 Chapter 6: Miscellanea

Figure 6.22: Stacked 400 MHz 1H NMR spectra (of the [Cp*Rh(κ2-dppMe,Me)Cl] complex in chloroform-d at variable temperatures as marked. Note that a break in scale occurs between the aromatic and aliphatic regions of the 1H NMR spectra.

From these data an upper limit can be placed on the rate of exchange, kex, for the

Me,Me – (dpp ) ligand as follows. The rate, kex, must be slower than the rate required for any two exchanging signals to appear to coalesce on the NMR timescale, kcoalesce, given by equations 6.1 and 6.2:

kex ≤ kcoalesce (eq. 6.1)

= (eq. 6.2) 𝜋𝜋 ∆ν 𝑘𝑘𝑐𝑐𝑐𝑐 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 √2 where ∆ν is the separation of the two interconverting signals in Hz. The minimum value

A3′ A4′ of kcoalesce is that derived for H / H , the signals closest in energy (i.e., with minimum

–1 ∆ν = 34 Hz). From this value, an upper limit of kex ≤ 76 s is obtained. A lower limit for the free energy of activation, ∆G‡, can now be derived using the Eyring Equation (equation 6.3):

James McPherson – June 2019 170 Chapter 6: Miscellanea

‡

−∆𝐺𝐺 (eq. 6.3) 𝐵𝐵 𝑘𝑘 𝑇𝑇 𝑅𝑅𝑅𝑅 𝑐𝑐𝑐𝑐 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 ℎ which can be rearranged to give𝑘𝑘 equation≥ 6.4: 𝑒𝑒

‡ ln (eq. 6.4) ℎ 𝑘𝑘𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 ∆𝐺𝐺 ≥ −𝑅𝑅𝑅𝑅 � 𝑘𝑘𝐵𝐵 𝑇𝑇 � where kB, h¸ and R are the Planck, Boltzmann and gas constants, respectively, and T is

–1 the temperature of the experiment. The smallest value of kcoalesce (76 s ) and largest T (328 K) gives the minimum free energy of activation, ∆G‡ ≥ 69 kJ mol–1 for the intramolecular exchange. This lower limit of ∆G‡ for the [Cp*Rh(κ2-dppMe,Me)Cl] complex is significantly larger than the barrier calculated using the Eyring equation, ∆G‡ ≈ 51 kJ mol–1, with the data

J K 24 reported by Aneetha et al. (coalescence of H /H : ∆ν = 68 Hz at 253 K in methanol-d4) for their [Cp*Rh(κ2-tpy)Cl] analogue.

The 1H and 13C{1H} NMR spectra of the other [Cp*M(κ2-dppR1,R2)Cl] complexes with symmetric dpp– ligands (R1 = R2) were similarly assigned. Different solvents were chosen to minimise the overlap of signals in the 1H spectra. For example, for [Cp*Ir(κ2-dppMe,Me)Cl], benzene-d6 resulted in the best separation of all peaks, which were assigned from inspection of the 400 MHz 1H-1H nOeSY NMR spectrum as shown in Figure 6.23. No correlations due to chemical exchange were observed in the nOeSY NMR spectrum for [Cp*Ir(κ2-dppMe,Me)Cl].

This result is in keeping with stronger (or less labile) coordinate bonds formed by third-row iridium(III) ions over second-row rhodium(III) ions, as expected.7

James McPherson – June 2019 171 Chapter 6: Miscellanea

1 2 Me,Me Figure 6.23: Top: 600 MHz H NMR spectrum of the [Cp*Ir(κ -dpp )Cl] complex in benzene-d6 at 298 K, including an expansion over the aromatic region. Bottom: on the left is a diagram showing the numbering scheme with key nOe interactions observed in the 400 MHz 1H-1H nOeSY NMR spectrum which is shown on the right.

Both structural isomers were observed in the 600 MHz 1H NMR spectrum of

2 CO Me,Me [Cp*Rh(κ -dpp 2 )Cl] in benzene-d6, as shown in Figure 6.24. The isomer of

[Cp*Rh(κ2-dppCO2Me,Me)Cl] with the “B”: pyridyl ring—the ring closest to the methyl ester pyrrolide substituent—bound to the rhodium(III) centre was ca. four times more abundant than the alternative isomer.

James McPherson – June 2019 172 Chapter 6: Miscellanea

1 2 CO Me,Me Figure 6.24: The 600 MHz H NMR spectrum for [Cp*Rh(κ -dpp 2 )Cl] in benzene-d6, with expansions across the aromatic region for each of the two different isomers (at different magnifications for clarity). Inset are diagrams identifying each isomer with the numbering schemes for assignment of the 1H NMR spectrum and with key nOe correlations indicated.

6.3.1.2 Crystallography

Red needles of [Cp*Rh(κ2-dppMe,Me)Cl] and [Cp*Rh(κ2-dppCO2Et,CO2Et)Cl] suitable for

SC-XRD studies were grown either from the slow diffusion of pentane into a solution of

[Cp*Rh(κ2-dppMe,Me)Cl] in acetone and from the slow evaporation of a solution of

[Cp*Rh(κ2-dppCO2Et,CO2Et)Cl] in a 1:1 mixture of dichloromethane/hexanes, respectively. Views of the molecular structures of [Cp*Rh(κ2-dppMe,Me)Cl] and [Cp*Rh(κ2-dppCO2Et,CO2Et)Cl] are presented in Figure 6.25 with key metric data in Table 6.5. The Rh–Npyrrolide bond lengths

{2.087(2) Å for [Cp*Rh(κ2-dppMe,Me)Cl] and 2.081(4) Å for [Cp*Rh(κ2-dppCO2Et,CO2Et)Cl]} are

James McPherson – June 2019 173 Chapter 6: Miscellanea

not statistically different to those of the Rh–Npyridyl bonds (2.091(2) Å for

[Cp*Rh(κ2-dppMe,Me)Cl] and 2.089(4) Å for [Cp*Rh(κ2-dppCO2Et,CO2Et)Cl]} for the bidentate binding dpp– ligands, while tridentate chelation of dpp– ligands to octahedral transition metal

8 ions forces the central M–Npyrrolide bonds to be ≥ 0.2 Å shorter than terminal M–Npyridyl bonds.

– The Npyridyl–Rh–Npyrrolide bite angles in the bidentate dpp ligands at 76.88(6)º for

[Cp*Rh(κ2-dppMe,Me)Cl] and 75.9(1)º for [Cp*Rh(κ2-dppCO2Et,CO2Et)Cl] are similar to each other and within the previously reported range for tridentate dpp– ligands around octahedral metal centres (at 73–80º).8 The bound pyridyl and pyrrolide rings are co-planar in the bidentate dpp– structures, while the unbound pyridyl ring has twisted away, with N1C–C2C–C5A–N1A inter- ring torsions of –144.3(2)º for [Cp*Rh(κ2-dppMe,Me)Cl] and 155.6(4)º for

[Cp*Rh(κ2-dppCO2Et,CO2Et)Cl]. The difference in the twist likely reflects the differing intramolecular pyrrolide ring substituent to pyridyl ring interactions and intermolecular crystal packing forces.

1 2 Figure 6.25: ORTEP views from SC-XRD structures of [Cp*Rh(κ2-dppR ,R )Cl] (R1 = R2 = Me on the left, and R1 2 = R = CO2Et on the right) showing 50% thermal ellipsoids at 100 K on the left and 150 K on the right. H atoms—

2 CO Et,CO Et and a CH2Cl2 solvent molecule in the structure of [Cp*Rh(κ -dpp 2 2 )Cl] on the right—have been omitted for clarity. Key bond lengths and angles are summarised in Table 6.5.

James McPherson – June 2019 174 Chapter 6: Miscellanea

1 2 Table 6.5: Key metric data from the SC-XRD structures of [Cp*Rh(κ2-dppR ,R )Cl] (R1 = R2 = Me and R1 = R2 =

CO2Et).

[Cp*Rh(κ2-dppMe,Me)Cl] [Cp*Rh(κ2-dppCO2Et,CO2Et)Cl] Rh–N1A /Å 2.087(2) 2.081(4) Rh–N1B /Å 2.091(2) 2.089(3) Rh–Cl /Å 2.4179(9) 2.403(1) 〈Rh–CCp〉 /Å 2.167(2) 2.156(6) N1A…N1B /Å 2.598(2) 2.564(5) N–Rh–N /º 76.88(6) 75.9(1) N1C–C2C–C5A–N1A /º –144.3(2) 155.6(4)

6.3.2 Electrochemistry

The redox properties of [Cp*M(κ2-dppR1,R2)Cl] (R1 = R2 = Me, M = Rh or Ir;

1 2 R = R = CO2Et, M = Rh) in acetonitrile were surveyed by cyclic voltammetry—shown in

Figures 6.26, 6.27, 6.28 with potential data for the redox processes summarised in Table 6.6— with tetrabutylammonium hexafluorophosphate (0.1 M) as the supporting electrolyte. For

[Cp*Rh(κ2-dppR1,R2)Cl] (Figure 6.26), an irreversible reduction process (R1) was observed at

–1.76 V vs. Fc+/0, which gives rise to a small anodic peak (O*) at –1.26 V vs. Fc+/0 in the reverse sweep. [Cp*RhIII(diimine)Cl]+ complexes typically exhibit a primary 2-electron reduction, which is irreversible due to concerted M–Cl bond cleavage, to afford the corresponding RhI species and Cl– ion (equation 6.5).22-23, 28-33 The behaviour at R1 is consistent with this precedent; i.e., the primary reduction is a 2-electron process and affords a [Cp*RhI(dpp)]– species. On the return scan, oxidation of the [Cp*RhI(dpp)]– species occurs at O*, and is accompanied by the binding of either chloride ion or acetonitrile solvent (equation 6.6). As shown on the right of Figure 6.26, O* was not observed until after the sample was first scanned through R1. On subsequent CV cycles, the potential of R1 reduction process shifts to –1.69 V vs. Fc+/0 (R1′), consistent with subsequent reduction of [Cp*RhIII(κ2-dpp)(MeCN)]+ formed in situ following the first scan.

James McPherson – June 2019 175 Chapter 6: Miscellanea

2 Me,Me Figure 6.26: Cyclic voltammograms for [Cp*Rh(κ -dpp )Cl] in acetonitrile, with [Bu4N][PF6] supporting electrolyte at 298 K. Other conditions: 1.0 mm diameter glassy carbon minidisk working electrode,

+/0 ∆Ep (Fc ) = 64.8 mV. Redox potentials are summarised in Table 6.6.

(eq. 6.5)

(eq. 6.6)

The first reduction (R1) is followed by a reversible reduction process (R2/O2) centred at

+/0 –1.97 V vs. Fc (∆Ep = 60 mV cf. ∆Ep = 65 mV for Fc standard and with ip, a/ip, c ≥ 0.95). The current and narrow peak separation for this process are suggestive for a 2-electron process, which is unusual for [Cp*RhIII(diimine)Cl]+ complexes. Without detailed further studies (e.g., using spectro-electrochemistry backed by DFT calculations), this process cannot be confidently assigned. A third reversible reduction process, R3, appears at –2.70 V vs. Fc+/0 with low reversibility (∆Ep = 115 mV and ip, a/ip, c ≈ 0.46) and with a current half that of processes R1 and R2. If R1 and R2 are 2-electron processes, then the current for R3 is consistent for a 1- electron process. However, the underlying redox chemistry cannot be confidently assigned.

In the anodic regime, a poorly reversible oxidation process, O4 (with ∆Ep = 91 mV and

+/0 ip, c/ip, a ≈ 0.27), was observed at 0.23 V vs. Fc . Such a process has not been reported for other diimine ligands, and so this is attributed to oxidation of the pyrrolide core which is unique to the dpp– ligand, presumably followed by some degree of ligand dissociation.

James McPherson – June 2019 176 Chapter 6: Miscellanea

2 Table 6.6: Redox potentials (V) observed for the [Cp*M(κ -dpp)Cl] complexes in MeCN with 0.1 M [Bu4N][PF6], +/0 –1 vs. E1/2 Fc at a 1.0 mm glassy carbon minidisk electrode, 200 mV s and 298 K.

[Cp*Rh(κ2-dppMe,Me)Cl] [Cp*Rh(κ2-dppCO2Et,CO2Et)Cl] [Cp*Ir(κ2-dppMe,Me)Cl][a] Ep, c (R1) [Ep, c (R1′)] –1.76 [–1.69] –1.61 [–1.53] –2.18 [b] E1/2 (R2) –1.97 (∆Ep = 60 mV) –1.70 (∆Ep = 59 mV) –2.75 Ep, c (R3) –2.70 –2.46 –2.87 Ep, a (O*) –1.26 –1.17 - Ep, a (O4) 0.23 0.84 0.30 Ep, a (O5) - - 0.77 –1 [a]: scan rate = 100 mV s . [b]: Ep, c (R2) for the irreversible process observed.

All the redox processes observed for [Cp*Rh(κ2-dppMe,Me)Cl] described above were also observed for [Cp*Rh(κ2-dppCO2Et,CO2Et)Cl], but at higher potentials due to the electronic influence of the ester substituents on the pyrrolide ring, as shown in the top left panel of Figure

6.27 and listed in Table 6.6. While the reduction events observed and attributed to either metal- or pyridyl-centred processes (R1, R2 and R3) shifted modestly (by 150–300 mV), the oxidation of the pyrrolide ring—with a shift of 610 mV—was far more sensitive to the electronics of the pyrrolide substituents.

The electrochemistry of [Cp*Rh(κ2-dppCO2Et,CO2Et)Cl] changes dramatically in the presence of water (Figure 6.27, top right panel) or other Brønsted acids {e.g.

2,2,2-trifluoroethanol (TFE) Figure 6.27, bottom left panel}. Potential data for the new processes are summarised in Table 6.7. The observations are consistent with the pendant pyridyl hydrogen bonding with water and/or enabling new proton-coupled electron transfer (PCET)

+/0 processes. A large catalytic wave (Rcat) was observed at –2.37 V vs. Fc in the presence of water with a catalytic current (icat) proportional to the concentration of water, which is consistent with the electrocatalytic reduction of protons to afford molecular hydrogen. The addition of TFE (to a concentration of 0.07 M) shifted the potential of Rcat slightly to –2.33 V

+/0 vs. Fc and caused a large increase in icat consistent with the increased acidity of TFE (over water) facilitating proton reduction.

James McPherson – June 2019 177 Chapter 6: Miscellanea

2 CO Et,CO Et Figure 6.27: Cyclic voltammograms for [Cp*Rh(κ -dpp 2 2 )Cl] in acetonitrile at 298 K, with [Bu4N][PF6] supporting electrolyte. Other conditions: 200 mV s–1 scan rate, at a 1.0 mm diameter glassy carbon minidisk

+/0 working electrode, and as indicated in each panel, ∆Ep (Fc ) = 72.9 mV. Shown in grey in each of the three panels showing electrocatalytic responses are the CVs recorded under the same conditions as indicated, but without [Cp*Rh(κ2-dppCO2Et,CO2Et)Cl]. Redox potentials are summarised in Table 6.7. Top left: CVs for

[Cp*Rh(κ2-dppMe,Me)Cl] under the same conditions are shown in black; top right: shown inset is the linear plot of icat vs. [H2O] observed under different water concentrations.

When the solution of [Cp*Rh(κ2-dppCO2Et,CO2Et)Cl] in acetonitrile, water (2.2 M) and TFE

(0.07 M) was saturated with CO2, different redox behaviour was again observed (bottom right panel, Figure 6.27). The catalytic response observed under the inert N2 (g) atmosphere (R1) was suppressed, with a new process, R1′ instead observed at –2.17 V vs. Fc+/0, with a smaller icat (10.6 µA). It was not established whether the changes in redox processes were due to selective CO2 reduction, or to the change in conditions (CO2 in MeCN-H2O is acidic) further enabling proton reduction. More experiments are clearly required to establish the active species, reaction products and catalytic efficiencies, and the mechanistic pathways.

James McPherson – June 2019 178 Chapter 6: Miscellanea

Table 6.7: Electrochemical potentials (in V vs. Fc+/0) for the redox events observed for

2 CO Et,CO Et [Cp*Rh(κ -dpp 2 2 )Cl] in acetonitrile at 298 K, with [Bu4N][PF6] supporting electrolyte in the presence of various substrates.

H2O (2.2 M) & TFE H2O (2.2 M) & TFE dry MeCN, N2 (g): H2O (2.2 M), N2 (g) (0.07 M), N2 (g) (0.07 M), CO2 (g) Ep, c (R1a) - –1.24 –1.24 - Ep, c (R1b) - –1.48 –1.48 - Ep, c (R1) –1.61 - - –1.58 Ep, c (R2) –1.73 –1.66 –1.59 - Ep, c (Rcat) –2.46[a] –2.37 [13.5] –2.33 [23.4] –2.17 [10.6] [icat / µA] [a]: In the absence of substrate the peak potential for reduction process R3 is reported.

A CV of [Cp*Ir(κ2-dppMe,Me)Cl] is presented in Figure 6.28. The primary reduction process, R1, at –2.18 V vs. Fc+/0 is irreversible and occurs at 420 mV lower potential than observed for the analogous rhodium complex. The potential of this wave does not change on subsequent scans (unlike for the rhodium analogue, see above). Reduction R1 is followed by two reduction processes at low potential (R2 at –2.75 V vs. Fc+/0 and R3 at –2.87 V vs. Fc+/0) that are merged with the onset of the solvent discharge (so peak currents cannot be determined).

The reversible process that immediately followed the primary reduction in the rhodium complexes, see above, is missing in the CVs of this iridium species. To positive potentials, CVs of [Cp*Ir(κ2-dppMe,Me)Cl] reveal two consecutive oxidation processes at 0.33 V and at 0.77 V vs. Fc+/0, each with lower current than for R1 and thus suggestive for 1-electron processes.

2 Me,Me Figure 6.28: Cyclic voltammogram for [Cp*Ir(κ -dpp )Cl] in acetonitrile at 298 K, with [Bu4N][PF6]

+/0 supporting electrolyte. Other conditions: 1.0 mm diameter glassy carbon minidisk working electrode, ∆Ep (Fc ) = 64.0 mV. Redox potentials are summarised in Table 6.7. Also shown in black is a CV for [Cp*Rh(κ2-dppMe,Me)Cl] recorded under the same conditions except at 200 mV s–1 as a reference.

James McPherson – June 2019 179 Chapter 6: Miscellanea

6.3.3 Conclusions

The facial Cp* ancillary ligand is incompatible with the tridentate binding mode of dpp– anions, and so instead [Cp*M(κ2-dpp)Cl] (M = Rh(III) or Ir(III)) complexes were prepared.

The unbound pyridyl ring thus represents an internal base, close to the labile halide ligand. The kinetics of the intramolecular exchange between the two pyridyl groups in

[Cp*Rh(κ2-dppMe,Me)Cl] is slow on the timescale of 400 MHz 1H NMR experiments in chloroform-d, which is in contrast to similar [Cp*Rh(κ2-tpy)Cl]+ species. The redox chemistry observed in the CVs for these new [Cp*Rh(κ2-dpp)Cl] complexes was unusual for

[Cp*Rh(diimine)X] species, particularly the secondary, reversible reduction process R2, and bears further investigation. Preliminary electrocatalysis screening experiments in the presence of water suggest proton coupled electroreduction, with a linear dependence between the catalytic current (icat) and the concentration of water. Further experiments are needed to determine the reaction products and active catalytic species, and also to confirm that the behaviour observed under CO2 is indeed a different process to that under inert N2. Also of interest for future investigations is whether these new [Cp*Rh(κ2-dpp)Cl] complexes are capable of electrocatalytic regeneration of NAD(P)H from NAD(P)+.

6.4 3-Pyridin-2-yl-N-[(1E)-pyridin-2-ylmethylidene]imidazo[1,5-a] pyridinamine (LImPy) – A Side-product With Promising Coordination Chemistry

Cooperative behaviour can emerge when multiple metal centres are incorporated within the same complex.34-38 Catalyses involving multinuclear complexes not only demonstrate dramatic increases in efficiencies and selectivities, but can also lead to reactions not possible with only one metal centre.35 For example, the urease metalloenzyme isolated from Klebsiella

James McPherson – June 2019 180 Chapter 6: Miscellanea aerogenes bacteria contains two nickel centres separated by 3.5 Å, and their cooperativity increases the rate of urea hydrolysis by a factor of 1014 versus the un-catalysed reaction.34

Cooperativity within bimetallic systems can also be achieved through the incorporations of two complex units with different functionality, for example Ishitani’s diads for photocatalytic

2+ CO2 reduction which consist of a [Ru(bpy)3] photosensitiser tethered directly to the catalytic

[Re(CO)3Cl] unit, with catalytic performances (e.g. Φ and TON) improved by an order of magnitude compared to the mixed mononuclear species.39

When Colbran’s one-pot Hdpp condensation was adapted to prepare HdppMe,Me directly,

3-pyridin-2-yl-N-[(1E)-pyridin-2-ylmethylidene]imidazo[1,5-a]pyridinamine (LImPy) was also isolated. LImPy had been previously reported as an undesired side-product from the trimeric condensation of 2-formyl pyridine and ammonium acetate by Kulhánek et al. who were

40 preparing terpene derivatives. As shown in Figure 6.29, the E isomer of LImPy has two obvious bidentate metal-binding domains (the pyridyl-imidazo[1,5-a]pyridine site I, and the pyridyl- imine site II), and may also be capable of chelating a single metal ion with all four N-donors.

The Z isomer (which has not yet been reported/observed) may bind metal ions through a tridentate mode for the Z isomer (see Figure 6.29). The delocalised heteroaromatic imdazo[1,5-a]pyridyl ring may also provide functionality (e.g., redox non-innocence), while the imino-pyridine moiety can also serve as a redox-active and proton-responsive unit.

James McPherson – June 2019 181 Chapter 6: Miscellanea

Figure 6.29: LImPy may bind transition metal ions at different sites, with four generic examples of possible different binding modes shown beneath.

6.4.1 Ligand Synthesis and Characterisation

3-Pyridin-2-yl-N-[(1E)-pyridin-2-ylmethylidene]imidazo[1,5-a]pyridinamine (LImPy) was isolated as a side-product in approximately 10% yield when Colbran’s one-pot direct condensation protocol4 was adapted to prepare HdppMe,Me from butanone and 2-formyl pyridine in 17% yield, as shown in Scheme 6.4a. Kulhánek et al. had previously reported40 the preparation of LImPy in 65% yield, from the condensation of three equivalents of 2-formyl pyridine with two equivalents of ammonium acetate in DMSO at 100 ºC after four hours,

Scheme 6.4b. Both synthetic protocols were employed to prepare the LImPy ligand used in this work.

James McPherson – June 2019 182 Chapter 6: Miscellanea

Me,Me Scheme 6.4: Preparation of LImPy, either as a side-product of the synthesis of dpp H by Colbran’s direct condensation protocol (a),4 or as reported by Kulhánek et al. (b).40

1 13 1 The H and C{ H} NMR spectra (at 600 MHz and 151 MHz respectively) of LImPy in

1 1 acetone-d6 were unambiguously assigned using conventional 2D experiments ( H- H COSY and nOeSY, 1H-13C HSQC and HMBC NMR) as illustrated in Figure 6.30. Only one species

(with two isomers possible due to the stereochemistry of the imine double bond) was observed.

The three spin systems were identified through 1H-1H COSY NMR. Correlations in the 1H-13C

HMBC NMR spectrum involving the signal at δ 9.38 in the 1H dimension assigned to the imine proton HB2′ unambiguously identified the 13C signal which was assigned to CB2 at δ 156.84, and hence the rest of the B ring system. The signal in the 1H NMR spectrum assigned to HA7 was identified through 1H-1H nOeSY NMR correlations with the signal assigned to HB3 (with

A7 B3 H …H separated by 3.80 Å in the SC-XRD structure of LImPy, see below), and correlations between the signals assigned to HA4 and HC6 (HA4…HC6 separated by 3.68 Å in the solid state) confirmed the remaining spin system. The E stereochemistry of the imine double bond in solution was confirmed by the absence of any nOe correlations between spins from the C ring with any of the spins in the B ring, which would be expected for the Z configuration.

James McPherson – June 2019 183 Chapter 6: Miscellanea

1 13 1 Figure 6.30: The 600 MHz H NMR (top) and 151 MHz C{ H} NMR (bottom) spectra of LImPy in acetone-d6.

A view of the molecular geometry from the SC-XRD structure of LImPy is shown in the inset with the assignment numbering and key 2D correlations (from the 1H-13C HMBC NMR spectrum in purple and 1H-1H nOeSY NMR spectrum in black) shown.

Single crystals of LImPy were grown as colourless needles when a concentrated solution of LImPy in ethanol was cooled at –22 ºC for a week. The structure of LImPy was determined at

100 K (Figure 6.31) and is consistent with the 173 K structure previously reported by Bureš

40 and co-workers, with the planar LImPy molecules packed in parallel π-π stacked layers separated by 3.4201(6) Å and offset by 1.6354(8) Å.

James McPherson – June 2019 184 Chapter 6: Miscellanea

Figure 6.31: ORTEP view (50% thermal ellipsoids at 100 K) of the molecular geometry of LImPy. Other key bond lengths (Å) and angles (º): N1C–C2C 1.346(3), C2C–C3A 1.465(2), C3A–N2A 1.330(2), N1AA–C2BB 1.288(3), C2BB–C2B 1.455(2), C2B–N1B 1.350(3), N1C–C2C–C3A 118.7(1), C2C–C3A–N2A 123.3(1), N1AA–C2BB–C2B 120.4(1) and C2BB–C2B–N1B 116.1(1).

The unbound N…N separations for each bidentate binding site {d(N…N) = 2.84 Å for binding site I (N1C…N2A), and 2.70 Å for site II (N1AA…N1B)} were approximated from the structure with the N-donor sites in the anti-configuration as shown in Figure 6.32. Shatruk and co-workers have showed that such approximation is a useful ligand metric, which is indicative of the relative ligand field strength (shorter = stronger field),41 and may result in selectivity due to the different compatibilities of the binding sites for particular metal ions. For example, the intermediate d(N…N) value for site I is in the 2.78–2.93 Å range that was identified as being compatible with forming spin crossover active iron(II) species, while the narrower site II is more compatible with low spin iron(II).41

Figure 6.32: The calculation of Shatruk’s d(N…N) parameter for the bidentate binding domains from the anti- configuration of the unbound ligand.41

James McPherson – June 2019 185 Chapter 6: Miscellanea

6.4.2 Synthesis and Characterisation of Metal Complexes

6.4.2.1 A Bimetallic rhenium(I) catalyst

Mononuclear [Re(N—N)(CO)3Cl] complexes (where N—N is a diimine ligand, e.g., bpy) have shown useful activities, particularly as electrocatalysts for the reduction of carbon dioxide.39, 42-49 These useful catalysts are simply prepared from pentacarbonyl rhenium(I) precursors, and their mechanism of action has been well studied including recently within our

50 group, and so the homo-bimetallic [LImPy{Re(CO)3Cl}2] complex was identified as an initial target. As shown in Scheme 6.5, this complex was easily collected in good yield from the straightforward reaction of LImPy and Re(CO)5Cl in refluxing toluene.

Scheme 6.5: Preparation of the homo bimetallic [LImPy{Re(CO)3Cl}2] complex.

In the high resolution NSI-mass spectrum of [LImPy{Re(CO)3Cl}2] aspirated in methanol, a peak envelope centred at m/z 931.9218 corresponds to the singly charged sodium-adduct,

+ [LImPy{Re(CO)3Cl}2 (M) + Na] (calc. m/z 931.9228). Peak envelopes consistent with loss of a single chloride ion were also observed, at m/z 873.9650 ([M – Cl]+: calc. m/z 873.9641) and m/z 905.9912 ([M – Cl + MeOH]+: calc. m/z 905.9904). The experimental and the simulated isotopic distributions were in excellent agreement as shown in Figure 6.33.

James McPherson – June 2019 186 Chapter 6: Miscellanea

Figure 6.33: (+)-HR-NSI-MS of [LImPy{Re(CO)3Cl}2]. Inset the spectrum is expanded to show the signals of interest in black against simulated isotopic distributions of the molecular ions as indicated below.

Two species were observed in the 600 MHz 1H NMR spectrum of the

[LImPy{Re(CO)3Cl}2] complex in DMSO-d6. The structures of these two species were confirmed by SC-XRD experiments (see below) as the homochiral (form I) and the heterochiral

(form II) isomers of [LImPy{Re(CO)3Cl}2], due to the chirality at each rhenium(I) centre.

1 Integration of the H NMR spectrum of [LImPy{Re(CO)3Cl}2] in DMSO-d6 at 298 K revealed roughly a 5:3 ratio of these isomers, however the identity of the major isomer was not confirmed.

These isomers are not interconverting on the timescale of the NMR experiment at 298 K, and the signals for fifteen of the twenty six (two sets of thirteen) unique 1H environments overlap one another and result in complicated multiplets (the purple signals in Figure 6.34), and so the spectrum was not assigned any further.

James McPherson – June 2019 187 Chapter 6: Miscellanea

1 Figure 6.34: The 600 MHz H NMR spectrum of [LImPy{Re(CO)3Cl}2] in DMSO-d6. Signals assigned to the major isomer are coloured blue, those assigned to the minor isomer are red, and in purple are signals for both species that overlap.

Two distinct crystalline forms of [LImPy{Re(CO)3Cl}2] were observed when pentane vapour slowly diffused into a solution of the complex in acetone: dark red needles (form I) and dark red plates (form II). The structure of form I was determined by SC-XRD at 150 K and a view is shown in Figure 6.35, with key metric parameters summarised in Table 6.8. The two binding domains are no longer co-planar, with ~58º torsion about the C1A–N1AA bond. Each facial rhenium(I) centre is chiral, and the clockwise (C) or anticlockwise (A) convention recommended by IUPAC was used to describe the chirality of each octahedral centre.51 Both homochiral enantiomers (i.e. C,C as shown in Figure 6.35 and A,A not shown) were present in the unit cell of form I. The d(N…N) distances of each binding domain have contracted to ~2.63

Å, with bite angles (N–Re–N) of 74º.

James McPherson – June 2019 188 Chapter 6: Miscellanea

Figure 6.35: An ORTEP view (50% thermal ellipsoids at 150 K) of the molecular geometry of form I of the homochiral bimetallic [LImPy{Re(CO)3Cl}2] complex, which crystallised as dark red needles, is shown on the left. The H atoms were omitted for clarity and key bond distances and angles summarised in Table 6.8 below. On the right are shown additional ball and stick views of each rhenium(I) centre in the asymmetric unit, with the IUPAC recommended configuration indices for the octahedral geometries and the Cahn-Ingold-Prelog priority sequence for each donor atom.51 Both enantiomers (i.e. C,C shown and A,A not shown) are present in the unit cell.

Table 6.8: Key metric data from the SC-XRD structures of [LImPy{Re(CO)3Cl}2], where Σ is the sum of |90 – α| for all 12 cis L–Re–L angles (α).

Form I Form I Form IIa Form IIa Re1 Re2 Re1 Re2 Re–N /Å 2.173(8) 2.182(7) 2.13(3) 2.18(2) Re–NPy /Å 2.191(7) 2.181(8) 2.20(3) 2.17(2) Re–Cl /Å 2.489(3) 2.465(3) 2.470(7) 2.466(9) 〈Re–CCO〉 /Å 1.91(1) 1.94(1) 1.89(3) 1.89(3) d(N…N)bound /Å 2.630(7) 2.62(1) 2.58(3) 2.63(3) Re…Re /Å 5.2304(7) 5.280(2) N–Re–N /º 74.1(3) 74.0(3) 73.0(9) 74.2(7) Σ /º 61.2 63.4 63.7 61.6 N2A–C1A–N1AA–C2BB /º 57.830(3) –55.438(15) a: Only weak diffraction (I/σ = 3.1) was observed from the dark red plates (Form II), which resulted in poor quality data (Rint = 24%, R1 = 8.5% for I ≥ 2σ(I), and wR2 = 22.2% for all data), and so the distances and angles are listed only to provide approximations and to inform qualitative comparisons.

The structure of the dark red plates (form II) was also investigated by SC-XRD. Several crystals were studied, and although data of sufficient quality to accurately measure interatomic distances or angles were not obtained—presumably due to evaporation of volatile solvent molecules—the data (Rint = 24%) and refinement (R1 = 8.5% for I ≥ 2σ(I), and wR2 = 22.2% for all data) were sufficient to confirm the connectivity in the molecular geometry of the heterochiral [LImPy{Re(CO)3Cl}2] isomer to be that shown in Figure 6.36. Within each

James McPherson – June 2019 189 Chapter 6: Miscellanea

[LImPy{Re(CO)3Cl}2] molecule in form II, the rhenium(I) centres are of opposite chirality (the

C,A enantiomer is shown in Figure 6.36, while A,C is not shown, but also present in the unit cell).

Figure 6.36: A ball and stick view of the molecular geometry of form II of the heterochiral bimetallic

[LImPy{Re(CO)3Cl}2] complex is shown on the left, which crystallised as dark red plates. The H atoms were omitted for clarity. On the right are shown additional ball and stick views of each rhenium(I) centre in the asymmetric unit, with the IUPAC recommended configuration indices for the octahedral geometries and the Cahn-Ingold-Prelog priority sequence for each donor atom.51 Both enantiomers (i.e. Re1,Re2 = C,A as shown or A,C not shown) are present in the unit cell.

6.4.2.2 Selective binding of iridium(III) at site II

With metal binding in the two bidentate domains thus demonstrated, it was of interest whether either of the sites could be bound selectively, and so one equivalent of LImPy was combined with half an equivalent (so 1:1 ligand to metal) of [Cp*IrCl2]2 dimer

(Cp* = pentamethylcyclopentadienyl). Following anion metathesis with K[PF6] (aq), the mononuclear [LImPy{Cp*IrCl}][PF6] complex was collected, with the iridium(III) bound at the more flexible pyridyl-imine site II and the more rigid pyridyl-imidazo[1,5-a]pyridine site I, with wider d(N…N) (2.84 Å, see section 6.2.4.1), unbound as shown in Scheme 6.6.

James McPherson – June 2019 190 Chapter 6: Miscellanea

Scheme 6.6: Synthesis of [LImPy{Cp*IrCl}][PF6].

1 The 600 MHz H NMR spectrum of [LImPy{Cp*IrCl}][PF6] in acetonitrile-d3, shown in

Figure 6.37, was sufficiently resolved that it could be assigned (along with the 151 MHz

13C{1H} NMR spectrum) using conventional 2D NMR characterisation techniques, as

1 15 described for LImPy above. The H- N HMBC NMR spectrum (at 600 MHz and 61 MHz, respectively) unambiguously confirms that the {Cp*IrCl}+ species is bound at the pyridyl-imine

1 C6 binding site II in LImPy. The correlation observed between the H signal assigned to the H environment and a 15N environment at δ ~300 which is typical for pyridyl N atoms with a non- bonding pair of electrons confirms it is unbound, while the 1H signals assigned to the HB2′ and

HB6 environments correlated with 15N environments (δ ≈ 230–235) typical for pyridyl N atoms which are coordinated to transition metals.25 More evidence for the iridium(III) centre binding at site II was observed in the 400 MHz 1H-1H nOeSY NMR spectrum with through space correlations observed between the signals assigned to the Cp-methyl protons and those assigned to HA7 and HB6.

James McPherson – June 2019 191 Chapter 6: Miscellanea

1 Figure 6.37: Top: the 600 MHz H NMR spectrum of [LImPy{Cp*IrCl}][PF6] in acetonitrile-d3. Inset is the molecular geometry with key 2D correlations shown (from the 1H-13C HMBC spectrum in purple, from the 1H-1H nOeSY spectrum in black and from the 1H-15N HMBC spectrum in teal). On the bottom left is shown the 1H-15N HMBC NMR spectrum (at 600 MHz and 61 MHz, respectively) and the 400 MHz 1H-1H nOeSY spectrum on the bottom left recorded on the complex in acetonitrile-d3, and with key correlations highlighted.

The iridium complex, [LImPy{Cp*IrCl}][PF6], crystallised with half a molecule of acetonitrile as dark red blocks from the slow diffusion of diethyl ether into a solution of

[LImPy{Cp*IrCl}][PF6] in acetonitrile. Several of these crystals were examined by SC-XRD, but

James McPherson – June 2019 192 Chapter 6: Miscellanea small quantities of polycrystalline material were observed as streaking of reflections at low angle in all cases, possibly due to decomposition of the complex, or evaporation of lattice solvent. Nonetheless, the data and refinement were of sufficient quality to confirm connectivity

+ + within the [LImPy{Cp*IrCl}] complex, specifically, that the [Cp*IrCl] unit was bound to LImPy at the imino-pyridine site II, with the pyridyl-imidazo[1,5-a]pyridine moiety unbound, as is shown in Figure 6.38.

Figure 6.38: A ball and stick view of the molecular structure of [LImPy{Cp*IrCl}][PF6]·0.5MeCN from SC-XRD data, confirming the connectivity within the molecule, and binding of [Cp*IrCl]+ at the imino-pyridine binding site II of LImPy. Half a molecule of acetonitrile solvent (not shown) was identified in the difference map. Both enantiomers of the complex were present in the unit cell.

6.4.3 Electrochemistry

The redox properties of LImPy and [LImPy{Re(CO)3Cl}2] in acetonitrile solution were surveyed by cyclic voltammetry; see Figures 6.39 and 6.40 and Table 6.9. Cyclic

+/0 Voltammograms (CVs) of LImPy exhibit an irreversible oxidation (O1, Ep, a at 0.44 V vs. Fc ),

+/0 and reduction (R1 at E1/2 = –2.12 V vs. Fc with ip, a/ip, c = 0.42). To more negative potentials, further reductions (R2 and R3) are observed at –2.83 and –3.01 V vs. Fc+/0, respectively; these

James McPherson – June 2019 193 Chapter 6: Miscellanea are merged with the rising cathodic solvent discharge and so it is not possible to accurately determine peak currents or process reversibilities. Switching the scan direction immediately after each process had little effect on the electrochemical (ir)reversibility observed.

Figure 6.39: Cyclic voltammograms of LImPy in acetonitrile with 0.1 M [Bu4N][PF6] at a glassy carbon minidisk

+/0 electrode and 298 K, ∆Ep (Fc ) = 67.3 mV.

+/0 Table 6.9: Redox potentials (/V vs. Fc ) and peak currents (in µA) for LImPy and [LImPy{Re(CO)3Cl}2] (ca. 1mM) in dry, deoxygenated acetonitrile with [Bu4N][PF6] (0.1 M) supporting electrolyte at a glassy carbon minidisk electrode and 298 K.

LImPy [LImPy{Re(CO)3Cl}2] [LImPy{Re(CO)3Cl}2] Substrates: None None CO2 (sat) Ep, a (O1) +0.44 - - [ip, a (O1)] [4] E1/2 (R1) –2.12 –1.19 –1.19 [ip, c (R1)] [–4] [-0.8] [–0.8] Ep, c (R2) –2.83 –2.24 –2.19 [ip, c (R2)] [–11] [–1.6] [–3.7] Ep, c (R3) –3.01 –2.74 –2.73 [ip, c (R3)] [–13] [–3.3] [–12.2] Ep, c (*) –1.71 –1.68 - [ip, c (*)] [–0.7] [–1.0] Ep, c (**) –2.48 - - [ip, c (**)] [–7.3]

The CV of [LImPy{Re(CO)3Cl}2] as the mixture of both isomers (see above) in dry acetonitrile with 0.1 M [Bu4N][PF6] under an atmosphere of dinitrogen is shown on the left in

Figure 6.40. The ligand-centred oxidation process (O1 in Figure 6.39) was not observed within the solvent discharge limits for [LImPy{Re(CO)3Cl}2]. CVs of the dimer isomers instead

James McPherson – June 2019 194 Chapter 6: Miscellanea

+/0 exhibited a poorly reversible reduction process (R1 at E1/2 = –1.19 V vs. Fc with ip, a/ip, c = 0.35) followed a second poorly reversible reduction process (R2 at ca. –2.23 V vs.

Fc+/0) and a third reduction (R3 at –2.73 V vs. Fc+/0) near the tail of the cathodic solvent discharge. There was also a small feature, marked with an asterisk (*) at –1.70 V vs. Fc+/0.

Figure 6.40: CVs of [LImPy{Re(CO)3Cl}2] in acetonitrile with 0.1 M [Bu4N][PF6] at a glassy carbon minidisk

+/0 electrode and 298 K, ∆Ep (Fc ) = 67.3 mV, under the conditions indicated. Shown in grey in the panel on the right is the CV recorded under the same conditions, but without [LImPy{Re(CO)3Cl}2].

The typical electrochemistry of mononuclear [Re(N—N)(CO)3Cl] complexes (where

N—N is a bidentate, diimine ligand, e.g., bpy or phen) is discussed below.44-45, 47, 49, 52 The primary single-electron reduction of [Re(N—N)(CO)3Cl] is typically reversible, with the added

•– electron residing principally on the diimine ligand in the 19-electron [Re(N—N)(CO)3Cl] radical anion (equation 6.7).

(eq. 6.7)

•– The subsequent 1-electron reduction of the [Re(N—N)(CO)3Cl] radical anion is electrochemically irreversible, with the rapid cleavage of the Re–Cl bond to afford the

– 18-electron anion, [Re(N—N)(CO)3] , and a chloride ion, equation 6.8.

James McPherson – June 2019 195 Chapter 6: Miscellanea

(eq. 6.8)

– Oxidation of the [Re(N—N)(CO)3] anion by one-electron then typically results in the formation of a facial Re–Re bonded dimer, equation 6.9.

(eq. 6.9)

The electrochemistry observed for the [LImPy{Re(CO)3Cl}2] dimers (Figure 6.40 and

Table 6.9) is more complicated than that for the mononuclear [Re(N—N)(CO)3Cl] complexes, as discussed above. For example, it is worth noting that the geometry of the planar LImPy ligand is incompatible with formation of a facial Re–Re bonded dimer. The processes observed for the

[LImPy{Re(CO)3Cl}2] dimer cannot be definitively assigned without further experiments, including spectro-electrochemistry to characterise the redox intermediates and DFT calculations to predict where successive electrons added to the dimer should localise and, also, to predict the spectroscopic properties of the putative redox intermediates.

In the CV recorded after the sample of [LImPy{Re(CO)3Cl}2] in dry acetonitrile with 0.1

M [Bu4N][PF6] was purged with CO2 (g) for 20 minutes (Figure 6.40, right panel), large increases in currents of R2 and R3 were observed that are indicative of electrocatalysis. The

42-45, 47-49, 52 mononuclear [Re(N—N)(CO)3Cl] catalysts are amongst the most studied for electroreduction of CO2 to CO and, presumably, that is the process here. More experiments, including bulk electrolyses with GC-monitoring of the head-space gases, would be required to determine this.

James McPherson – June 2019 196 Chapter 6: Miscellanea

6.4.4 Conclusions

LImPy was easily prepared, and has been investigated as a ligand for bimetallic complexes, through two diimine binding sites. Cyclic voltammetry studies of LImPy reveal that the ligand may be capable of storing and supplying electrons, with irreversible reductions (e.g., R1 at

–2.12 V vs. Fc+/0) and oxidation (O1 at 0.44 V vs. Fc+/0) observed. A homo-bimetallic rhenium(I) complex, [LImPy{Re(CO)3Cl}2] was prepared, as a mixture of two isomers, with the two rhenium(I) ions separated by ~5.2–5.3 Å in the solid state. These [LImPy{Re(CO)3Cl}2] complexes showed complicated and unusual electrochemical responses during CV studies, and require further studies to characterise and assign the electrical and chemical steps involved. In the presence of CO2, enhanced cathodic currents were observed, and so these complexes should be investigated further as electrocatalysts for the reduction of CO2. Binding site II (a pyridyl- imine moiety) selectively binds [Cp*IrIII]2+ units over site I (a pyridyl-imidazopyridine), and offers a potential route to hetero-bimetallic species, and thus cooperativity, which remains to be explored in future work.

6.5 Experimental

6.5.1 General Methods

All NMR spectra were recorded on Bruker Avance III DPX-300, 400, 500 or 600 spectrometers operating at 300, 400, 500 or 600 MHz for 1H NMR experiments (75.5, 100.6,

125.8 or 150.9 MHz for 13C NMR; 30.4, 40.6, 50.7 or 60.8 MHz for 15N NMR), respectively.

All chemical shifts were calibrated against residual solvent signals.53 A 1-dimensional (1D) 1H

NMR spectrum was acquired immediately prior to each 1H-15N HMBC NMR experiment.

Calibration of the residual solvent in the 1H NMR spectrum allowed calibration of the f1 (15N) and f2 (1H) dimensions in the following 2D experiment (Ξ = 10.132912).54 The scalar coupling

James McPherson – June 2019 197 Chapter 6: Miscellanea constant of 11 Hz for 2J(15N, 1H) was used for all 1H-15N HMBC NMR experiments.55 All coupling constants (J) are reported in hertz. Signals in NMR spectra are reported as broad (br) singlets (s), doublets (d), triplets (t), quartets (q) or unclear multiplets (m). Unless otherwise stated, all spectra were obtained at 298 K. For variable temperature (VT) 1H NMR studies, two experiments were performed at each temperature, at least 15 minutes apart, to confirm that the composition of the solution had reached thermal equilibrium. NMR spectra were processed with MestReNova 11.0.1 software.56 FT-IR experiments were performed with an Agilent Cary

630 ATR FTIR spectrometer. Signals in FT-IR spectra were reported as very broad (vbr), broad

(br), weak (w), medium (m) or strong (s). UV-Vis experiments were performed on a Varian

Cary 60 spectrophotometer. High resolution mass spectra were run on a Thermo Fisher

Scientific Orbitrap LTQ XL ion trap mass spectrometer using a nanospray ionisation source.

Elemental micro-analyses were performed by the Chemical Analysis Facility at Macquarie

University.

6.5.1.1 Synthesis and Reagents

Unless specified, solvents were taken from an Innovative Technology Pure Solvent

Dispenser immediately prior to use and commercially available reagents were used as received.

Reactions were carried out under inert atmospheres using standard Schlenk line techniques.

6.5.1.2 Crystallography

Single crystal X-ray diffraction (SC-XRD) data were collected on either a Bruker kappa-

II CCD diffractometer using a IµS Incoatec Microfocus Source with Mo-Kα radiation

(λ = 0.71073 Å) and an APEX-II CCD detector or a Bruker D8 Quest Single Crystal diffractometer using an IµS 3.0 Microfocus Source with Mo-Kα radiation (λ = 0.71073 Å) and a Photon II detector. Datasets from the Australian Synchrotron, using Si<111> monochromated

James McPherson – June 2019 198 Chapter 6: Miscellanea synchrotron X-ray radiation (λ = 0.71073 Å) on the MX1 beamline,57 were collected by Dr

Hasti Iranmanesh.

6.5.1.3 Electrochemistry

Solutions of compounds were of ca. 1 mM concentrations in either dichloromethane or acetonitrile with tetrabutylammonium hexafluorophosphate (0.1 M) as the supporting electrolyte. High-purity dinitrogen was used to deoxygenate the solution before each electrochemical experiment. Unless otherwise stated, CVs were recorded at 100 mV s–1 under an atmosphere of dinitrogen using a GAMRY Reference 600 potentiostat using a standard three-electrode system comprising of a glassy carbon minidisk working electrode (1.0 mm diameter, freshly polished), a platinum wire counter electrode and an eDAQ leakless Ag/AgCl reference electrode. A background CV was always obtained between anodic and cathodic solvent discharge limits prior to adding sample. Ferrocene (Fc) was added as an internal standard at the end of each set of CV experiments and all potentials are quoted relative to the

Fc+/0 couple.

6.5.2 Homoleptic Transition Metal Complexes

6.5.2.1 Synthesis

6.5.2.1.1 General Procedure

R1,R2 Hdpp (approx. 1 mmol) was combined with NEt3 (0.3 mL, 2 mmol) in dry, deaerated

MeOH (10 mL) and stirred at reflux for 15 min. The appropriate MCl2·xH2O metal source

(approx. 0.5 mmol) was added, and the mixture stirred at reflux for 2 h under an inert

R1,R2 atmosphere. After cooling to room temperature, the crude product, [M(dpp )2] was collected by filtration and washed with cold MeOH three times. Where required, further purification was

James McPherson – June 2019 199 Chapter 6: Miscellanea by recrystallisation from boiling methanol. Yields were calculated against metal chloride hydrate as follows with the characterisation data for each complex.

CO2Me,Me 6.5.2.1.1.1 [Ni(dpp )2]

1 Yield: 101.3 mg, 75% as a yellow solid. H NMR (300 MHz, benzene-d6) δ 113.7 (vbrs),

108.5 (vbrs), 64.7 (brs), 59.0 (brs), 43.4 (brs), 42.0 (brs), 9.8 (brs), 9.4 (brs), 4.1 (brs), 3.9 (brs).

+ (+)-HR-NSI-MS found m/z 643.1599 (calc. [C34H28N6O4Ni (M) + H] : m/z 643.1598) m/z

+ 611.1334 (calc. [M – OMe] : m/z 611.1336). Anal. Calc. for C34H28N6O4Ni C: 63.48 H: 4.39

N: 13.06%. Found: C: 63.25 H: 4.27 N: 13.05%. Pale orange needles suitable for SC-XRD

CO Me,Me experiments were grown from the diffusion of pentane into a solution of [Ni(dpp 2 )2] in dichloromethane followed by cooling at 4 ºC for 1 week.

CO2Me,Me 6.5.2.1.1.2 [Cu(dpp )2]

Yield: 151.2 mg, 78% as brown needles. (+)-HR-NSI-MS found m/z 648.1531 (calc.

+ + [C34H28N6O4Cu (M) + H] : m/z 648.1541), m/z 616.1269 (calc. [M – OMe] : m/z 616.1279).

CO2Me,Me 6.5.2.1.1.3 [Zn(dpp )2]

1 Yield: 423.1 mg, 91% as a yellow solid. H NMR (300 MHz, benzene-d6) δ 9.31 (dt, J =

8.3, 1.0 Hz, 1H, HB3), 7.65–7.59 (m, 2H, HB6 and HC6), 7.47 (dt, J = 8.2, 1.1 Hz, 1H, HC3), 7.03

(ddd, J = 8.3, 7.4, 1.8 Hz, 1H, HB4), 6.84 (ddd, J = 8.2, 7.4, 1.8 Hz, 1H, HC4), 6.10–5.98 (m,

2H, HB5 and HC5), 3.78 (s, 3H, HCO2Me), 3.00 (s, 3H, HMe). (+)-HR-NSI-MS found m/z 649.1526

+ + (calc. [C34H28N6O4Zn (M) + H] : m/z 649.1536), m/z 617.1263 (calc. [M – OMe] : m/z

617.1274). Anal. Calc. for C34H28N6O4Zn C: 62.83 H: 4.34 N: 12.93%. Found: C: 62.49 H:

4.15 N: 12.44%.

James McPherson – June 2019 200 Chapter 6: Miscellanea

Me,Me 6.5.2.1.1.4 [Ni(dpp )2]

1 Yield: 25.2 mg, 46% yield as orange needles. H NMR (400 MHz, acetone-d6) δ 105.7

(vbrs), 56.1 (brs), 43.8 (brs), 9.4 (brs). (+)-HR-NSI-MS found m/z 555.1801 (calc. [C32H28N6Ni

+ (M) + H] : m/z 555.1802). Anal. Calc. for C32H28N6Ni C: 69.21 H: 5.08 N: 15.13%. Found: C:

68.90 H: 4.92 N: 15.07%.

Me,Me 6.5.2.1.1.5 [Cu(dpp )2]

Yield: 28.0 mg, 85% as brown needles. (+)-HR-NSI-MS found m/z 560.1742 (calc.

+ Me,Me [C32H28N6Cu (M) + H] : m/z 560.1744). Brown needles of [Cu(dpp )2]·0.25(MeCN)

Me,Me suitable for SC-XRD experiments were grown from a saturated solution of [Cu(dpp )2] in boiling acetonitrile, followed by cooling at 4 ºC for 1 week.

Me,Me 6.5.2.1.1.6 [Zn(dpp )2]

1 Yield: 39.5 mg, 64% as a yellow solid. H NMR (400 MHz, acetonitrile-d3) δ 7.65 (ddd,

J = 8.2, 1.4, 1.1 Hz, 1H, HB3), 7.60 (ddd, J = 8.2, 7.0, 1.8 Hz, 1H, HB4), 7.55 (ddd, J = 5.0, 1.8,

1.4 Hz, 1H, HB6), 6.75 (ddd, J = 7.0, 5.0, 1.4 Hz, 1H, HB5), 2.42 (s, 3H, HMe). (+)-HR-NSI-MS

+ found m/z 561.1741 (calc. [C32H28N6Zn (M) + H] : m/z 561.1740). Pale yellow needles of

Me,Me [Zn(dpp )2]·0.25(MeCN) suitable for SC-XRD experiments were grown from a saturated

Me,Me solution of [Zn(dpp )2] in boiling acetonitrile, followed by cooling at 4 ºC for 1 week.

6.5.2.2 Crystallography

CO2Me,Me 6.5.2.2.1 [Ni(dpp )2]

CO Me,Me The crystal of [Ni(dpp 2 )2] with dimensions 0.08 × 0.02 × 0.02 mm was coated in immersion oil type NVH and transferred to the goniometer under the flow of the nitrogen cold stream (100 K). Diffraction measurements were carried out using Si<111> monochromated synchrotron radiation (λ = 0.71073 Å) on the MX1 Beamline at the Australian Synchrotron.57

James McPherson – June 2019 201 Chapter 6: Miscellanea

Data collection was with the BluIce software suite,58 and the unit cell refinement, data reduction

59 and processing was with XDS. The structure was solved in the P21/c spacegroup by

SHELXT60 (with intrinsic phasing), and the full-matrix least-square refinements carried out using SHELXL-201561 through Olex262 software. The non-hydrogen atoms were refined anisotropically.

Me,Me 6.5.2.2.2 [Cu(dpp )2]·0.25(MeCN)

Me,Me The crystal of [Cu(dpp )2]·0.25(MeCN) with dimensions 0.2 × 0.05 × 0.05 mm was coated in immersion oil type NVH and transferred to the goniometer under the flow of the nitrogen cold stream (100 K). Diffraction measurements were carried out using Si<111> monochromated synchrotron radiation (λ = 0.71073 Å) on the MX1 Beamline at the Australian

Synchrotron.57 Data collection was with the BluIce software suite,58 and the unit cell refinement, data reduction and processing was with XDS.59 The structure was solved in the

60 P21/c spacegroup by SHELXT (with intrinsic phasing), and the full-matrix least-square refinements carried out using SHELXL-201561 through Olex262 software. The non-hydrogen atoms were refined anisotropically. A disordered molecule of acetonitrile was identified from the difference map at a point of inversion and included in the least-squares refinement. One ligand at Cu1 was disordered over two sites. The atoms disordered over the two sites were modelled isotropically, and the bond- and 1,3- distances within the connectivity list for the non- hydrogen atoms for each disordered dpp– ligand were restrained to be the same as those in the other ligand at Cu1, which was most stable during refinement, using the SAME function.

Me,Me 6.5.2.2.3 [Zn(dpp )2]·0.25(MeCN)

Me,Me The crystal of [Zn(dpp )2]·0.25(MeCN) with dimensions 0.5 × 0.05 × 0.05 mm was selected under a polarizing microscope (Leica M165Z), and placed on a MicroMount

(MiTeGen, USA) consisting of a thin polymer tip with a wicking aperture. The X-ray diffraction

James McPherson – June 2019 202 Chapter 6: Miscellanea measurements were carried out on a Bruker Kappa-II CCD diffractometer at 150 K using a IµS

Incoatec Microfocus Source with Mo-Kα radiation (λ = 0.71073 Å). The single crystal was coated with immersion oil type NVH and then quickly transferred to the cold nitrogen stream generated by an Oxford Cryostream 700 series. Symmetry related absorption corrections using the program SADABS63 were applied and the data were corrected for Lorentz and polarisation effects using Bruker APEX363 software. The structure was solved in the triclinic P-1 spacegroup by SHELXT60 (with intrinsic phasing) and the full-matrix least-square refinements were carried out using SHELXL-201561 through Olex262 software. Weak diffraction was observed (I/σ = 3.4), and the quality of the data obtained (Rint= 25.62%) was insufficient to accurately model interatomic bond lengths or angles. The non-hydrogen atoms were refined anisotropically. The data confirms the connectivity of the zinc(II) complex. One ligand bound to Zn1 appears disordered, with prolate ADPs indicating thermal motion, but due to the poor quality of the data, further modelling of this disorder was not completed. The carbon and nitrogen atoms within each pyridyl or pyrrolide ring of the disordered ligand were modelled as rigid bodies using RIGU. A disordered molecule of acetonitrile solvent was identified from the difference map at a point of inversion and included in the least-squares refinement.

6.5.3 [M(dpp)2] Metalloligands

CO2,CO2 6.5.3.1 NaδH3–δdpp synthesis

CO Et,CO Et A solution of Hdpp 2 2 (164.3 mg, 0.4497 mmol) in H2SO4 (9 M, aq, 10 mL) was heated at reflux for 1 hour. After cooling to room temperature, the yellow solution was partially

CO ,CO neutralised to pH 3 with NaHCO3 (sat, aq). Yellow product NaδH3–δdpp 2 2 (75.1 mg, 55%) was then collected, washed with several portions of water and then diethyl ether, and dried

1 NH under vacuum. H NMR (400 MHz, DMSO-d6) δ 12.35 (brs, 1H, H ), 8.68 (ddd, J = 4.9, 1.4,

James McPherson – June 2019 203 Chapter 6: Miscellanea

0.7 Hz, 1H, HB6), 8.13 (dt, J = 8.2, 0.7 Hz, 1H, HB3), 8.02 (ddd, J = 8.2, 7.0, 1.4 Hz, 1H, HB4),

B5 13 1 7.46 (dd, J = 7.0, 4.9, 0.7 Hz, 1H, H ). C{ H} NMR (101 MHz, DMSO-d6) δ 165.67, 147.93,

147.84 (CB6), 138.27 (CB4), 130.53, 123.16 (CB5), 121.71 (CB3), 119.40. FT-IR: ν = 3104.6 brw,

1671.4 w, 1572.9 w, 1442.5 s, 1351.6 m, 1235.9 m, 1145.9 m, 1062.1 w, 984.6 w, 881.7 w,

–1 + 772.1 s cm . (+)-LR-ESI-MS found m/z 310.08 (calc. [C16H11N3O4 (M) + H] : m/z 310.08), m/z 332.08 (calc. [M + Na]+: m/z 332.06), m/z 641.25 (calc. [2M + Na]+: m/z 641.14).

(CO)2O 6.5.3.2 NaδH1–δdpp synthesis

CO ,CO NaδH1–δdpp 2 2 (22.8 mg, 0.074 mmol) was dissolved in acetic anhydride (2 mL) and the solution was heated at reflux for 1 h. The resulting brown solution was cooled to room

(CO) O temperature. White solid NaδH1–δdpp 2 (13.4 mg, 62%) was collected by filtration and washed with several portions of water and then ether, and then dried under vacuum. 1H NMR

B6 (400 MHz, DMSO-d6) δ 8.76 (ddd, J = 4.9, 1.5, 0.8 Hz, 1H, H ), 8.33 (dt, J = 8.0, 0.8 Hz, 1H,

HB3), 8.07 (ddd, J = 8.0, 7.6, 1.5 Hz, 1H, HB4), 7.42 (ddd, J = 7.6, 4.9, 0.8 Hz, 1H, HB5). 13C{1H}

B6 B4 B5 NMR (101 MHz, DMSO-d6) δ 159.26, 155.05, 149.85 (C ), 137.88 (C ), 124.56 (C ), 122.31

(CB3), 117.41. FT-IR: ν = 3388.0 m, 1810.7 s (C=O symmetric stretch), 1755.7 s (C=O asymmetric stretch), 1568.1 m, 1450.6 s, 1295.0 w, 1127.2 m, 1050.3 m, 864.5 s, 776.0 w,

–1 741.4 s, 704.5 m cm . (+)-LR-ESI-MS found m/z 324.08 (calc. [C16H9N3O3 (M) +MeOH +

H]+: m/z 324.10), m/z 346.08 (calc. [M + MeOH + Na]+: m/z 346.08), m/z 669.25 (calc. [2(M +

MeOH) + Na]+: m/z 669.17).

CO2,CO2 + 6.5.3.3 [Co(H2dpp )2] synthesis

CO ,CO Co(OAc)2·4(H2O) (21.7 mg, 0.0871 mmol) was added to a solution of H3dpp 2 2 (48.0 mg, 0.155 mmol) in methanol (20 mL) and the mixture was stirred at reflux for 3 h. The brown solid product (20.5 mg) was collected and washed with several portions of water. 1H NMR (400

James McPherson – June 2019 204 Chapter 6: Miscellanea

B3 B4 MHz, DMSO-d6) δ 9.39 (d, J = 8.2 Hz, 1H, H ), 7.94 (t, J = 7.8 Hz, 1H, H ), 7.10 (t, J = 6.6

B5 B6 13 1 Hz, 1H, H ), 6.86 (d, J = 5.7 Hz, 1H, H ). C{ H} NMR (151 MHz, DMSO-d6) δ 166.53

(CA3′), 157.73 (CB2), 150.48 (CB6), 142.66 (CA2), 141.20 (CB4), 126.35 (CA3), 125.01 (CB5),

124.34 (CB3). FT-IR: ν = 3353.1 vbrw, 3114.5 w, 1687.1 w, 1598.0 m, 1495.4 m, 1437.3 s,

1369.0 m, 1294.4 w, 1159.3 m, 1049.6 w, 993.5 w, 905.6 w, 770.3 s, 688.3 m cm–1. (+)-LR-

+ ESI-MS Found: m/z 675.17 (calc. [C32H20N6O8Co (M)] : m/z 675.07), m/z 697.08 (calc.

[M – H + Na]+: m/z 697.05), m/z 719.08 (calc. [M – 2H + 2Na]+: m/z 719.03), m/z 631.17 (calc.

+ + [M – CO2] : m/z 631.08), m/z 653.08 (calc. [M – CO2 – H + Na] : m/z 653.06), m/z 587.17

+ (calc. [M – 2CO2] : m/z 587.09).

6.5.3.4 Crystallography

CO2,CO2 6.5.3.4.1 {Na(MeOH)[Co(Hdpp )2]}n·7(DMF)

Brown needles suitable for SC-XRD experiments were grown from the slow diffusion of

CO ,CO + diethylether into a solution of [Co(Hdpp 2 2)2] in a refrigerated (at 4 ºC) solution of N,N-

CO ,CO dimethylformamide over 1 week. The crystal of {Na(MeOH)[Co(Hdpp 2 2)2]}n·7(DMF) with dimensions 0.5 × 0.04 × 0.04 mm was selected under a polarizing microscope (Leica

M165Z), and placed on a MicroMount (MiTeGen, USA) consisting of a thin polymer tip with a wicking aperture. The X-ray diffraction measurements were carried out on a Bruker Quest D8

Single Crystal diffractometer with a Photon II detector at 100 K using a IµS 3.0 Microfocus

Source with Mo-Kα radiation (λ = 0.71073 Å). The single crystal was coated with immersion oil type NVH and then quickly transferred to the cold nitrogen stream generated by an Oxford

Cryostream 800 series. Several crystals were examined, but in all cases only weak diffraction was observed. Symmetry related absorption corrections using the program SADABS63 were applied and the data were corrected for Lorentz and polarisation effects using Bruker APEX363 software. Although the cell dimensions (α = β = 90º) suggest monoclinic metric symmetry,

James McPherson – June 2019 205 Chapter 6: Miscellanea there were a large number (> 100–1000) of exceptions to the systematic absences expected for all monoclinic spacegroups, and so the structure was solved in the triclinic P-1 space group by

SHELXT60 (with intrinsic phasing) and the full-matrix least-square refinements were carried out using SHELXL-201561 through Olex262 software. Weak diffraction was observed

(I/σ = 3.4), and the quality of the data obtained (Rint = 40.89%) was insufficient to accurately model interatomic bond lengths or angles. Neither H-atoms nor disorder were modelled within this structure. All atoms were modelled anisotropically, and the C- and N-atoms within pyridyl or pyrrole rings were constrained to regular hexagonal- or pentagonal- geometries using AFIX.

The carboxylate group {C7G, O3G and O4G} was unstable during refinement, and so the position of C7G restrained (against the position of C4G) using DFIX, and the 1,3-bond distance between O3G and O4G restrained with DANG. Two solvent molecules bound to the sodium ions were identified from the difference map, and modelled as methanol. A very large, solvent

3 accessible void (Vvoid = 1187 Å ,and ~284 electrons) was identified and masked by Olex2 during

62, 64 refinement, which could correspond to seven DMF solvent molecules. As a result, R1 was improved by 9.5%.

6.5.4 Organometallic Piano-stool [Cp*M(κ2-dpp)Cl] complexes

6.5.4.1 Synthesis

General method: The literature method was adapted as follows. [Cp*MCl2]2 precursor

(~50 mg, ~80 µmol) was combined with Hdpp (~180 µmol) and Et3N (500 µmol) in dichloromethane and stirred overnight at room temperature. The solvent was then removed under a stream of dinitrogen, and the crude solid was washed with cold acetone. The resulting

[Cp*M(κ2–dpp)Cl] complex was recrystallised from dichloromethane/pentane.

Characterisation data follow.

James McPherson – June 2019 206 Chapter 6: Miscellanea

6.5.4.1.1 [Cp*Rh(κ2-dppMe,Me)Cl]

Yield: 21.0 mg, 24%. 1H NMR (600 MHz, chloroform-d) δ 8.64 (ddd, J = 4.8, 1.9, 0.9

Hz, 1H, HB6′), 8.47 (ddd, J = 8.0, 1.2, 0.9 Hz, 1H, HB3′), 8.36 (ddd, J = 5.8, 1.6, 0.8 Hz, 1H,

HB6), 7.71 (ddd, J = 8.0, 7.4, 1.9 Hz, 1H, HB4′), 7.53 (ddd, J = 8.3, 7.1, 1.6 Hz, 1H, HB4), 7.46

(ddd, J = 8.3, 1.4, 0.8 Hz, 1H, HB3), 7.08 (ddd, J = 7.4, 4.8, 1.2 Hz, 1H, HB5′), 6.84 (ddd, J =

7.1, 5.8, 1.4 Hz, 1H, HB5), 2.36 (s, 3H, HA3′), 2.28 (s, 3H, HA4′), 1.25 (s, 15H, HD1′). 13C{1H}

NMR (151 MHz, chloroform-d) δ 157.61 (CB2), 157.20 (CB2′), 150.89 (CB6), 148.14 (CB6′),

143.82 (CA5), 137.25 (CA2), 137.00 (CB4), 136.07 (CB4′), 126.48 (CB3′), 124.05 (CA3/4), 124.03

(CA4/3), 119.98 (CB5′), 118.01 (CB5), 117.63 (CB3), 94.42 (d, J = 7.9 Hz, CD1), 11.81 (CA3′), 10.83

(CA4′), 8.80 (CD1′).

Red needles suitable for SC-XRD experiments were grown from the slow diffusion of pentane into a solution of [Cp*Rh(κ2-dppMe,Me)Cl] in acetone.

6.5.4.1.2 [Cp*Ir(κ2-dppMe,Me)Cl]

1 Yield: 19.1 mg, 19%. H NMR (600 MHz, benzene-d6) δ 8.98 (ddd, J = 8.0, 1.2, 0.9 Hz,

1H, HB3′), 8.72 (ddd, J = 4.8, 1.9, 0.9 Hz, 1H, HB6′), 8.18 (ddd, J = 5.8, 1.6, 0.8 Hz, 1H, HB6),

7.38 (ddd, J = 8.0, 7.3, 1.9 Hz, 1H, HB4′), 7.19 – 7.17 (m, 1H, HB3), 6.81 (ddd, J = 8.6, 7.2, 1.6

Hz, 1H, HB4), 6.69 (ddd, J = 7.3, 4.8, 1.2 Hz, 1H, HB5′), 6.21 (ddd, J = 7.2, 5.8, 1.4 Hz, 1H, HB5),

2.68 (s, 3H, HA4′), 2.44 (s, 3H, HA3′), 0.96 (s, 15H, HD1′). 13C{1H} NMR (151 MHz,

B2 B2′ B6 B6′ A5 benzene-d6) δ 158.38 (C ), 157.75 (C ), 150.97 (C ), 148.31 (C ), 144.20 (C ), 137.26

(CA2), 136.63 (CB4), 135.65 (CB4′), 127.90 (CB3′), 123.76 (CA3/4), 122.97 (CA4/3), 120.19 (CB5′),

116.98 (CB5), 116.70 (CB3), 85.86 (CD1), 11.40 (CA3′ and CA4′), 8.35 (CD1′).

6.5.4.1.3 [Cp*Rh(κ2-dppCO2Me,Me)Cl]

Was isolated as a 4:1 mixture of diastereomers. Yield: 54.5 mg, 38%. (+)-HR-NSI-MS

+ Found: m/z 566.1063 (calc. [C27H29N3O2RhCl (M) + H] : m/z 566.1076), m/z 588.0881 (calc.

James McPherson – June 2019 207 Chapter 6: Miscellanea

+ + [C27H29N3O2RhCl (M) + Na] : m/z 588.0896), m/z 530.1299 (calc. [M – Cl] : m/z 530.1309), m/z 1153.1879 (calc. [2M + Na]+: m/z 1153.1899), m/z 1095.2303 (calc. [2M – Cl]+: m/z

1095.2313).

Major Isomer:

1 B3 H NMR (600 MHz, benzene-d6) δ 9.22 (dt, J = 8.4, 1.0 Hz, 4H, H ,

major), 9.09 (dt, J = 8.0, 1.1 Hz, 4H, HC3, major), 8.69 (ddd, J = 4.8,

1.9, 0.9 Hz, 4H, HC6, major), 8.27 (ddd, J = 5.6, 1.7, 0.7 Hz, 4H, HB6, major), 7.43 (ddd, J = 8.0, 7.4, 1.9 Hz, 4H, HC4, major), 7.00 (ddd, J = 8.6, 7.2, 1.7 Hz, 4H,

HB4, major), 6.70 (ddd, J = 7.4, 4.8, 1.2 Hz, 4H, HC5, major), 6.35 (ddd, J = 7.1, 5.7, 1.4 Hz,

4H, HB5, major), 3.65 (s, 12H, HA3″, major), 3.12 (s, 12H, HA4′, major), 0.82 (s, 60H, HD1′, major).

Minor Isomer:

1 B3 H NMR (600 MHz, benzene-d6) δ 8.89 (dt, J = 7.9, 1.1 Hz, 1H, H ,

minor), 8.66 (ddd, J = 4.8, 1.9, 0.9 Hz, 1H, HB6, minor), 8.20 (dt, J = 5.5,

1.3 Hz, 1H, HC6, minor), 7.37 (td, J = 7.6, 1.9 Hz, 1H, HB4, minor), 7.09 (d,

J = 8.3 Hz, 1H, HC3, minor), 6.82 (ddd, J = 8.4, 7.2, 1.6 Hz, 1H, HC4, minor), 6.76 (ddd, J = 7.5, 4.8, 1.2 Hz, 1H, HB5, minor), 6.30 (ddd, J = 7.1, 5.7, 1.3 Hz, 1H,

HC5, minor), 3.61 (s, 3H, HA3′′, minor), 2.76 (s, 3H, HA4′, minor), 0.83 (s, 15H, HD1′, minor).

6.5.4.1.4 [Cp*Ir(κ2-dppCO2Me,Me)Cl]

Yield: 95.9 mg, 40%.(+)-HR-NSI-MS found: m/z 656.1632 (calc. [C27H29N3O2IrCl (M)

+ H]+: m/z 656.1627), m/z 620.1873 (calc. [M – Cl]+: m/z 620.1866), m/z 1333.3013 (calc. [2M

+ Na]+: m/z 1333.3006), m/z 1275.3446 (calc. [2M – Cl]+: m/z 1275.3420).

James McPherson – June 2019 208 Chapter 6: Miscellanea

6.5.4.1.5 [Cp*Rh(κ2-dppCO2Et,CO2Et)Cl]

1 Yield: 40.9 mg, 40%. H NMR (300 MHz, benzene-d6) δ 9.17 – 8.98 (m, 1H), 8.87 (d, J

= 8.2 Hz, 1H), 8.60 (d, J = 4.8 Hz, 1H), 8.20 (d, J = 5.6 Hz, 1H), 7.34 (d, J = 7.8 Hz, 1H), 6.91

(d, J = 10.2 Hz, 2H), 6.69 (dd, J = 7.4, 4.8 Hz, 1H), 6.43 – 6.25 (m, 1H), 4.52–4.22 (m, 4H),

1.23 (t, J = 7.1 Hz, 3H), 1.17 (t, J = 7.0 Hz, 3H), 0.76 (s, 15H). (+)-HR-NSI-MS Found: m/z

+ + 638.1288 (calc. [C30H33N3O4RhCl (M) + H] : m/z 638.1287), m/z 602.1520 (calc. [M – Cl] : m/z 602.1521), m/z 1275.2515 (calc. [2M + H]+: m/z 1275.2502). Red needles suitable for SC-

XRD experiments were grown from the slow evaporation of a solution of

[Cp*Rh(κ2-dppCO2Et,CO2Et)Cl] in a 1:1 (v/v) mixture of dichloromethane/n-hexane.

6.5.4.1.6 [Cp*Ir(κ2-dppCO2Et,CO2Et)Cl]

Yield: 25.2 mg, 28%. (+)-HR-NSI-MS found: m/z 728.1861 (calc.

+ + [C30H33N3O4IrCl (M) + H] : m/z 728.1862), m/z 692.2101 (calc. [M – Cl] : m/z 692.2095).

6.5.4.2 Crystallography

6.5.4.2.1 [Cp*Rh(κ2-dppMe,Me)Cl]

Red needles suitable for SC-XRD experiments were grown from the slow diffusion of pentane into a solution of [Cp*Rh(κ2-dppMe,Me)Cl] in acetone. The crystal of

[Cp*Rh(κ2-dppMe,Me)Cl] with dimensions 0.05 × 0.01 × 0.01 mm was coated in immersion oil type NVH and transferred to the goniometer under the flow of the nitrogen cold stream (100

K). Diffraction measurements were carried out using Si<111> monochromated synchrotron radiation (λ = 0.71073 Å) on the MX1 Beamline at the Australian Synchrotron.57 Data collection was with the BluIce software suite,58 and the unit cell refinement, data reduction and

59 60 processing was with XDS. The structure was solved in the P21/n spacegroup by SHELXT

(with intrinsic phasing), and the full-matrix least-square refinements carried out using

James McPherson – June 2019 209 Chapter 6: Miscellanea

SHELXL-201561 through Olex262 software. The non-hydrogen atoms were refined anisotropically.

2 CO2Et,CO2Et 6.5.4.2.2 [Cp*Rh(κ -dpp )Cl]·CH2Cl2

Red needles suitable for SC-XRD experiments were grown from the slow evaporation of a solution of [Cp*Rh(κ2-dppCO2Et,CO2Et)Cl] in a 1:1 (v/v) mixture of dichloromethane/n-hexane.

2 CO Et,CO Et The crystal of [Cp*Rh(κ -dpp 2 2 )Cl]·CH2Cl2 with dimensions 0.1 × 0.05 × 0.05 mm was selected under a polarizing microscope (Leica M165Z), and placed on a MicroMount

(MiTeGen, USA) consisting of a thin polymer tip with a wicking aperture. The X-ray diffraction measurements were carried out on a Bruker Kappa-II CCD diffractometer at 150 K using a IµS

RIGU. A molecule of dichloromethane solvent was identified from the difference map and included in the least-squares refinement.

James McPherson – June 2019 210 Chapter 6: Miscellanea

6.5.5 3-Pyridin-2-yl-N-[(1E)-pyridin-2-ylmethylidene]imidazo[1,5-a] pyridinamine (LImPy) – A Side-product With Promising Coordination Chemistry

6.5.5.1 Synthesis

6.5.5.1.1 3-Pyridin-2-yl-N-[(1E)-pyridin-2-ylmethylidene]imidazo[1,5-a] pyridinamine (LImPy)

Colbran’s previously reported Hdpp synthesis4 was adapted as follows to prepare

Me,Me dpp H, which also resulted in the formation of LImPy as a side-product. 2-Formyl pyridine

(4.6 mL, 48 mmol) was added to a refluxing solution of 2-butanone (2.4 mL, 27 mmol) and ammonium acetate (2.7916 g, 36.22 mmol) in ethanol (30 mL) and the reaction mixture was heated at reflux overnight. After cooling to room temperature, the solvent was removed under

Me,Me 4 reduced pressure and dpp H (1.0385 g, 17%) and LImPy (419.4 mg, 9%) were separated from the crude products by flash column chromatography using a 1:1 ethyl acetate/hexanes mobile phase. Characterisation data for dppMe,MeH: 1H NMR (300 MHz, chloroform-d) δ 10.53

(s, 1H, NpyrH), 8.57 (ddd, J = 4.9, 1.8, 1.0 Hz, 2H, H6), 7.69 (ddd, J = 8.1, 7.2, 1.8 Hz, 2H, H4),

7.62 (ddd, J = 8.1, 1.4, 1.0 Hz, 2H, H3), 7.07 (ddd, J = 7.2, 4.9, 1.4 Hz, 2H, H5), 2.38 (s, 6H).

1 Characterisation data for LImPy: H NMR (600 MHz, acetone-d6) δ 10.02 (ddd, J = 7.5, 1.2, 0.9

Hz, 1H, HA4), 9.38 (s, 1H, HB2ʹ), 8.73 (ddd, J = 4.9, 1.8, 0.9 Hz, 1H, HC6), 8.70 (ddd, J = 4.9,

1.8, 1.0 Hz, 1H, HB6), 8.46 (ddd, J = 8.1, 1.1, 0.9 Hz, 1H, HC3), 8.37 (ddd, J = 7.8, 1.2, 1.0 Hz,

1H, HB3), 8.01 – 7.95 (m, 2H, HA7 and HC4), 7.91 (ddd, J = 7.8, 7.4, 1.8 Hz, 1H, HB4), 7.42 (ddd,

J = 7.4, 4.8, 1.2 Hz, 1H, HB5), 7.39 (ddd, J = 7.4, 4.8, 1.1 Hz, 1H, HC5), 7.16 (ddd, J = 9.0, 6.4,

0.9 Hz, 1H, HA6), 6.98 (ddd, J = 7.5, 6.4, 1.3 Hz, 1H, HA5). 13C{1H} NMR (151 MHz, acetone-

B2 B2ʹ C2 B6 C6 A1 d6) δ 156.84 (C ), 155.14 (C ), 151.77 (C ), 150.61 (C ), 149.33 (C ), 139.00 (C ), 137.81

(HC4), 137.21 (CB4), 134.27 (CA3), 130.02 (CACq), 127.30 (CA4), 125.19 (CB5), 123.21 (CC5),

123.05 (CA6), 122.92 (CC3), 121.20 (CB3), 118.26 (CA7), 115.63 (CA5). (+)-HR-NSI-MS found:

+ + m/z 300.1238 (calc. [C18H13N5 (M) + H] m/z 300.1244), m/z 322.1057 (calc. [(M) + Na]

James McPherson – June 2019 211 Chapter 6: Miscellanea

m/z 322.1069). Colourless crystalline needles of LImPy, suitable for single crystal X-ray diffraction studies, were grown by dissolving LImPy in the minimum amount of hot ethanol, which was then stored at –22 ºC for one week.

40 Alternatively, LImPy was prepared directly as reported by Kulhánek et al.

6.5.5.1.2 [LImPy{Re(CO)3Cl}2]

Re(CO)5Cl (39.4 mg, 0.109 mmol) was added to a pale orange, refluxing solution of LImPy

(20.8 mg, 0.0695 mmol) in toluene (15 mL). The resulting cherry red solution was stirred at reflux for an hour, and then cooled to room temperature. Dark red [LImPy{Re(CO)3Cl}2] was collected by filtration, and washed with diethyl ether, as a mixture of two isomers in a 5:3 ratio.

1 H NMR (600 MHz, DMSO-d6) δ 9.68 (s, 5H), 9.53 (s, 3H), 9.34 (d, J = 7.3 Hz, 3H), 9.27 (dd,

J = 7.3, 1.2 Hz, 5H), 9.23 – 9.18 (m, 5H + 3H), 9.14 – 9.09 (m, 3H), 9.06 (dt, J = 5.4, 1.2 Hz,

5H), 8.84 (d, J = 9.2 Hz, 3H), 8.81 (d, J = 8.5 Hz, 3H), 8.77 – 8.74 (m, 5H), 8.72 (dt, J = 9.2,

1.2 Hz, 5H), 8.51 (d, J = 7.7 Hz, 3H), 8.46 (dtd, J = 15.3, 7.6, 1.5 Hz, 5H + 5H + 3H),

8.40 – 8.32 (m, 5H + 3H), 7.96 (ddd, J = 7.4, 5.4, 1.6 Hz, 5H + 3H), 7.69 (ddt, J = 8.0, 6.0, 3.0

Hz, 5H + 3H), 7.57 (dd, J = 9.3, 6.7 Hz, 5H + 3H), 7.51 – 7.39 (m, 5H + 3H). (+)-HR-NSI-MS

+ found: m/z 931.9218 (calc. [C24H13Cl2N5O6Re2 (M) + Na] m/z 931.9222), m/z 873.9650 (calc.

[M – Cl]+ m/z 873.9636), m/z 905.9912 (calc. [M – Cl + MeOH]+ m/z 905.9898). Anal. Calc. for C24H13N5O6ReCl2 C: 31.65 H: 1.44 N: 7.69%. Found: C: 30.46 H: 1.90 N: 7.20%.

6.5.5.1.3 [LImPy{Cp*IrCl}][PF6]

[Cp*IrCl2]2 (71.4 mg. 0.0896 mmol) dimer was added to a pale brown solution of LImPy

(90.9 mg, 0.203 mmol) in CH2Cl2 (15 mL) and the resulting dark red solution was stirred overnight at room temperature. The solvent was then removed, and the dark red residue re- dissolved in the minimum amount of methanol required. K[PF6] (aq, saturated) was added, and the dark red [LImPy{Cp*IrCl}][PF6] (94.1 mg, 65%) was collected by filtration and washed first

James McPherson – June 2019 212 Chapter 6: Miscellanea

1 with water then diethyl ether. H NMR (600 MHz, acetonitrile-d3) δ 10.11 (ddd, J = 7.3, 1.1,

1.0 Hz, 1H, HA4), 9.37 (s, 1H, HB2ʹ) 8.86 (ddd, J = 4.8, 1.4, 1.0, 1H, HB6), 8.76 (ddd, J = 4.8,

1.8, 0.9 Hz, 1H, HC6), 8.52 (dt, J = 9.1, 1.3, 1.0 Hz, 1H, HA7), 8.34 (ddd, J = 8.1, 1.2, 0.9 Hz,

1H, HC3), 8.27 (dt, J = 7.8, 1.0 Hz, 1H, HB3),8.19 (td, J = 7.8, 1.4 Hz, 1H, HB4), 7.98 (ddd, J =

8.1, 7.5, 1.8 Hz, 1H, HC4), 7.83 (ddd, J = 7.8, 4.8, 1.0 Hz, 1H, HB5), 7.42 (ddd, J = 7.5, 4.8, 1.2

Hz, 1H, HC5), 7.29 (ddd, J = 9.1, 6.5, 1.1 Hz, 1H, HA6), 7.06 (ddd, J = 7.3, 6.5, 1.3 Hz, 1H, HA5),

Cp-Me 13 1 B2ʹ B2 1.42 (s, 15H, H ). C{ H} NMR (151 MHz, acetonitrile-d3) δ 165.40 (C ), 156.29 (C ),

153.35 (CB6), 150.89 (CC2), 149.73 (CC6), 141.59 (CB4), 138.33 (CC4), 137.30 (CA1), 134.84

(CA3), 130.88 (CB5), 129.95 (CB3), 128.01 (CA4), 127.77 (CACq), 125.22 (CA6), 124.12 (CC5),

123.16 (CC3), 119.51 (CA7), 116.46 (CA5), 91.34 (CCp), 8.83 (CCp-Me). Anal. Calc. for

C28H28N5IrClPF6 C: 41.66 H: 3.50 N: 8.68%. Found: C: 41.77 H: 4.07 N: 9.02%.

6.5.5.2 Crystallography

6.5.5.2.1 [LImPy{Re(CO)3Cl}2] – Form I

Dark red needles of [LImPy{Re(CO)3}2]·0.5pentane were collected after the slow diffusion of pentane into a solution of [LImPy{Re(CO)3}2] in acetone. The crystal of

[LImPy{Re(CO)3}2]·0.5pentane with dimensions 0.1 × 0.04 × 0.04 mm was selected under a polarizing microscope (Leica M165Z), and placed on a MicroMount (MiTeGen, USA) consisting of a thin polymer tip with a wicking aperture. The X-ray diffraction measurements were carried out on a Bruker Kappa-II CCD diffractometer at 150 K using a IµS Incoatec

Microfocus Source with Mo-Kα radiation (λ = 0.71073 Å). The single crystal was coated with immersion oil type NVH and then quickly transferred to the cold nitrogen stream generated by an Oxford Cryostream 700 series. Symmetry related absorption corrections using the program

SADABS63 were applied and the data were corrected for Lorentz and polarisation effects using

Bruker APEX363 software. The structure was solved in the triclinic P-1 space group by

James McPherson – June 2019 213 Chapter 6: Miscellanea

SHELXT60 (with intrinsic phasing) and the full-matrix least-square refinements were carried out using SHELXL-201561 through Olex262 software. The non-hydrogen atoms were refined anisotropically. A solvent mask was used in Olex2,62 which identified a very large

3 3 (Vvoid = 449.5 Å ) solvent accessible void in the unit cell (Vcell = 1626.4(3) Å ) accounting for

134 electrons, which could correspond to up to four solvent molecules (of either acetone or pentane).

6.5.5.2.2 [LImPy{Re(CO)3Cl}2] – Form II

Dark red plates of the hetero-chiral racemate, [LImPy{Re(CO)3}2], were collected after the slow diffusion of pentane into a solution of [LImPy{Re(CO)3}2] in acetone. The crystal of

[LImPy{Re(CO)3}2] with dimensions 0.1 × 0.1 × 0.05 mm was selected under a polarizing microscope (Leica M165Z), and placed on a MicroMount (MiTeGen, USA) consisting of a thin polymer tip with a wicking aperture. The X-ray diffraction measurements were carried out on a Bruker Kappa-II CCD diffractometer at 150 K using a IµS Incoatec Microfocus Source with

Mo-Kα radiation (λ = 0.71073 Å). The single crystal was coated with immersion oil type NVH and then quickly transferred to the cold nitrogen stream generated by an Oxford Cryostream

700 series. Symmetry related absorption corrections using the program SADABS63 were applied and the data were corrected for Lorentz and polarisation effects using Bruker APEX363 software. The structure was solved in the triclinic P-1 space group by SHELXT60 (with intrinsic phasing) and the full-matrix least-square refinements were carried out using SHELXL-201561 through Olex262 software. Weak diffraction was observed (I/σ = 5.2), and the quality of the data obtained (Rint= 24.01%) was insufficient to accurately model interatomic bond lengths or angles. The non-hydrogen atoms were refined anisotropically, except for the carbon and oxygen atoms of the carbonyl ligands. The data confirms the connectivity of the hetero-chiral

[LImPy{Re(CO)3}2] molecules, with large anisotropic displacement parameters (ADPs) for the

James McPherson – June 2019 214 Chapter 6: Miscellanea chloride ligands if the chirality of either rhenium(I) centre was inverted, as shown in Figure

6.41, below.

Figure 6.41: ORTEP diagram of the SC-XRD structure of the hetero-chiral [LImPy{Re(CO)3}2] isomer as modelled from the SC-XRD data collected from form II (plates). Shown in the middle, and on the right, are the same views but modelled with the chirality at Re1 (middle) or Re2 (on the right) inverted to give either homochiral enantiomer. The large ADPs associated with the chloride ligands—not to mention the short C–O, long Re–C, and Re–C–O angles of ca. 90º for the carbonyl ligands—in both homochiral models indicate that these models are poorer fits to the data than the heterochiral model on the left, despite similar refinement statistics (R1 = 8.5% for the heterochiral model, cf. 8.7-9.0% for the homochiral models).

6.5.5.2.3 [LImPy{Cp*IrCl}][PF6]·0.5MeCN

Dark red blocks of [LImPy{Cp*IrCl}][PF6]·0.5MeCN were collected after the slow diffusion of diethyl ether into a solution of [LImPy{Cp*IrCl}][PF6] in acetonitrile. The crystal of [LImPy{Cp*IrCl}][PF6]·0.5MeCN with dimensions 0.07 × 0.153 × 0.193 mm was selected under a polarizing microscope (Leica M165Z), and placed on a MicroMount (MiTeGen, USA) consisting of a thin polymer tip with a wicking aperture. The X-ray diffraction measurements were carried out on a Bruker Kappa-II CCD diffractometer at 150 K using a IµS Incoatec

SADABS63 were applied and the data were corrected for Lorentz and polarisation effects using

Bruker APEX363 software. The structure was solved in the monoclinic P2/n space group by

SHELXT60 (with intrinsic phasing) and the full-matrix least-square refinements were carried out using SHELXL-201561 through Olex262 software. The non-hydrogen atoms were refined

James McPherson – June 2019 215 Chapter 6: Miscellanea anisotropically. A half molecule of acetonitrile solvent was identified from the difference map and included in the least-squares refinement. The carbon atoms in the Cp* ring (C1D, C2D, …,

C10D) were modelled as a rigid body using RIGU.

During data collection, some decomposition of the crystal was observed as streaking of reflections at low angles. Several crystals were examined, and all displayed this same behaviour.

6.6 References

1. Hein, F. M., F., Synthesis of 2,5-di(α-pyridyl)pyrrole and of several complex derivatives of 2,5-di(α-pyridyl)-3,4-dicarbethoxypyrrole. Die Pharmazie 1954, 9, 455-460. 2. Hein, F.; Beierlein, U., Preparation and chemistry of the complexes of 2,5-di-(α- pyridyl)pyrrole. Pharm. Zentralhalle Dtschl. 1957, 96, 401-21. 3. Ciszek, J. W.; Keane, Z. K.; Cheng, L.; Stewart, M. P.; Yu, L. H.; Natelson, D.; Tour, J. M., Neutral Complexes of First Row Transition Metals Bearing Unbound Thiocyanates and Their Assembly on Metallic Surfaces. J. Am. Chem. Soc. 2006, 128 (10), 3179-3189. 4. McSkimming, A.; Diachenko, V.; London, R.; Olrich, K.; Onie, C. J.; Bhadbhade, M. M.; Bucknall, M. P.; Read, R. W.; Colbran, S. B., An easy one-pot synthesis of diverse 2,5- di(2-pyridyl)pyrroles: a versatile entry point to metal complexes of functionalised, meridial and tridentate 2,5-di(2-pyridyl)pyrrolato ligands. Chem. Eur. J. 2014, 20 (36), 11445-56. 5. McPherson, J. N.; Hogue, R. W.; Akogun, F. S.; Bondi, L.; Luis, E. T.; Price, J. R.; Garden, A. L.; Brooker, S.; Colbran, S. B., Predictable Substituent Control of Co(III/II) Redox Potential and Spin Crossover in Bis(dipyridylpyrrolide)cobalt Complexes. Inorg. Chem. 2019, 58 (3), 2218-2228. 6. Chow, H. S.; Constable, E. C.; Housecroft, C. E.; Kulicke, K. J.; Tao, Y., When electron exchange is chemical exchange–assignment of 1H NMR spectra of paramagnetic cobalt(II)-2,2 ′:6′,2″-terpyridine complexes. Dalton Trans. 2005, (2), 236-237. 7. Lincoln, S. F., Mechanistic Studies of Metal Aqua Ions: A Semi-Historical Perspective. Helv. Chim. Acta 2005, 88 (3), 523-545. 8. McPherson, J. N.; Das, B.; Colbran, S. B., Tridentate pyridine–pyrrolide chelate ligands: An under-appreciated ligand set with an immensely promising coordination chemistry. Coord. Chem. Rev. 2018, 375, 285-332. 9. Procter, I. M.; Hathaway, B. J.; Nicholls, P., The electronic properties and stereochemistry of the copper(II) ion. Part I. Bis(ethylenediamine)copper(II) complexes. J. Chem. Soc. A 1968, (0), 1678-1684. 10. McPherson, J. N.; Elton, T. E.; Colbran, S. B., A Strain-Deformation Nexus within Pincer Ligands: Application to the Spin States of Iron(II) Complexes. Inorg. Chem. 2018, 57 (19), 12312-12322. 11. Jahn, H. A.; Teller, E., Stability of polyatomic molecules in degenerate electronic states - I—Orbital degeneracy. Proc. R. Soc. London, Ser. A 1937, 161 (905), 220-235.

James McPherson – June 2019 216 Chapter 6: Miscellanea

12. Housecroft, C. E.; Sharpe, A. G., Inorganic Chemistry. Fourth edition. ed.; Harlow, England : New York : Pearson: 2012. 13. Halcrow, M. A., Structure:function relationships in molecular spin-crossover complexes. Chem. Soc. Rev. 2011, 40 (7), 4119-4142. 14. Irving, H.; Williams, R. J. P., 637. The stability of transition-metal complexes. J. Chem. Soc. 1953, (0), 3192-3210. 15. Vande Linde, A. M. Q.; Westerby, B. C.; Ochrymowycz, L. A.; Rorabacher, D. B., Applicability of the Marcus relationship to copper(II/I) electron transfer. Comparison of NMR self-exchange relaxation and reduction and oxidation cross-reaction kinetics for a macrocyclic amino tetrathiaether-copper(II/I) complex in aqueous solution. Inorg. Chem. 1993, 32 (3), 251- 257. 16. Villeneuve, N. M.; Schroeder, R. R.; Ochrymowycz, L. A.; Rorabacher, D. B., Cyclic Voltammetric Evaluation of Rate Constants for Conformational Transitions Accompanying Electron Transfer. Effect of Varying Structural Constraints in Copper(II/I) Complexes with Dicyclohexanediyl-Substituted Macrocyclic Tetrathiaethers. Inorg. Chem. 1997, 36 (20), 4475- 4483. 17. Ambundo, E. A.; Deydier, M.-V.; Grall, A. J.; Aguera-Vega, N.; Dressel, L. T.; Cooper, T. H.; Heeg, M. J.; Ochrymowycz, L. A.; Rorabacher, D. B., Influence of Coordination Geometry upon Copper(II/I) Redox Potentials. Physical Parameters for Twelve Copper Tripodal Ligand Complexes. Inorg. Chem. 1999, 38 (19), 4233-4242. 18. Hansch, C.; Leo, A.; Taft, R. W., A survey of Hammett substituent constants and resonance and field parameters. Chem. Rev. 1991, 91 (2), 165-195. 19. Batten, S. R.; Neville, S. M.; Turner, D. R., Coordination Polymers: Design, Analysis and Application. Royal Society of Chemistry: Cambridge, UK, 2009. 20. Qin, Q.-P.; Qin, J.-L.; Meng, T.; Lin, W.-H.; Zhang, C.-H.; Wei, Z.-Z.; Chen, J.-N.; Liu, Y.-C.; Liang, H.; Chen, Z.-F., High in vivo antitumor activity of cobalt oxoisoaporphine complexes by targeting G-quadruplex DNA, telomerase and disrupting mitochondrial functions. Eur. J. Med. Chem. 2016, 124, 380-392. 21. Berg, J. M., Biochemistry. 6th ed. ed.; New York, N.Y. : W. H. Freeman: New York, N.Y., 2007; p 565-566. 22. McSkimming, A.; Bhadbhade, M. M.; Colbran, S. B., Bio-Inspired Catalytic Imine Reduction by Rhodium Complexes with Tethered Hantzsch Pyridinium Groups: Evidence for Direct Hydride Transfer from Dihydropyridine to Metal-Activated Substrate. Angew. Chem. Int. Ed. 2013, 52 (12), 3411-3416. 23. McSkimming, A.; Chan, B.; Bhadbhade, M. M.; Ball, G. E.; Colbran, S. B., Bio- Inspired Transition Metal–Organic Hydride Conjugates for Catalysis of Transfer Hydrogenation: Experiment and Theory. Chem. Eur. J. 2015, 21 (7), 2821-2834. 24. Aneetha, H.; Zacharias, P. S.; Srinivas, B.; Lee, G. H.; Wang, Y., Synthesis and characterization of Cp*Rh(III) and Ir(III) polypyridyl complexes: fluxional behavior of Rh(III) 5 complexes and molecular structure of [Cp*Rh(Ph-terpy)Cl]BF4 complex (Cp* = η -C5Me5). Polyhedron 1998, 18 (1-2), 299-307. 25. Mason, J., Nitrogen nuclear magnetic resonance spectroscopy in inorganic, organometallic, and bioinorganic chemistry. Chem. Rev. 1981, 81 (3), 205-227. 26. Govindaswamy, P.; Kollipara, M. R., Syntheses and characterization of η5- pentamethylcyclopentadienyl rhodium(III) and iridium(III) complexes containing polypyridyls. J. Coord. Chem. 2006, 59 (6), 663-669. 27. Cheng, J.; Zhu, M.; Wang, C.; Li, J.; Jiang, X.; Wei, Y.; Tang, W.; Xue, D.; Xiao, J., Chemoselective dehydrogenative esterification of aldehydes and alcohols with a dimeric rhodium(II) catalyst. Chem. Sci. 2016, 7 (7), 4428-4434.

James McPherson – June 2019 217 Chapter 6: Miscellanea

28. Kölle, U.; Grützel, M., Organometallic Rhodium(III) Complexes as Catalysts for the Photoreduction of Protons to Hydrogen on Colloidal TiO2. Angew. Chem. Int. Ed. 1987, 26 (6), 567-570. 29. Ruppert, R.; Herrmann, S.; Steckhan, E., Efficient indirect electrochemical in-situ regeneration of nadh:electrochemically driven enzymatic reduction of pyruvate catalyzed by d- ldh. Tetrahedron Lett. 1987, 28 (52), 6583-6586. 30. Kölle, U.; Kang, B.-S.; Infelta, P.; Comte, P.; Grätzel, M., Elektrochemische und pulsradiolytische Reduktion von (Pentamethylcyclopentadienyl)(polypyridyl)rhodium- Komplexen. Chem. Ber. 1989, 122 (10), 1869-1880. 31. Ladwig, M.; Kaim, W., Spectroscopic and electrochemical properties of the isomeric bidiazine complexes [(C5Me5)ClRh(bdz)]+ and (C5Me5)Rh(bdz) and their relevance to the + catalysis of the 2 H → H2 reaction by 2,2′-bipyridine analogues. J. Organomet. Chem. 1991, 419 (1), 233-243. 32. Caix, C.; Chardon-Noblat, S.; Deronzier, A.; Moutet, J.-C.; Tingry, S., (Pentamethylcyclopentadienyl)(polypyridyl) rhodium and iridium complexes as electrocatalysts for the reduction of protons to dihydrogen and the hydrogenation of organics. J. Organomet. Chem. 1997, 540 (1), 105-111. 33. Caix, C.; Chardon-Noblat, S.; Deronzier, A., Electrocatalytic reduction of CO2 into 5 + formate with [(η -Me5C5)M(L)Cl] complexes (L = 2,2′-bipyridine ligands; M = Rh(III) and Ir(III)). J. Electroanal. Chem. 1997, 434 (1), 163-170. 34. Steinhagen, H.; Helmchen, G., Asymmetric Two-Center Catalysis—Learning from Nature. Angew. Chem. Int. Ed. 1996, 35 (20), 2339-2342. 35. Kamijo, S.; Yamamoto, Y., Organic Synthesis with Bimetallic Systems. In Multimetallic Catalysts in Organic Synthesis, Shibasaki, M.; Yamamoto, Y., Eds. 2005. 36. Broussard, M. E.; Juma, B.; Train, S. G.; Peng, W.-J.; Laneman, S. A.; Stanley, G. G., A Bimetallic Hydroformylation Catalyst: High Regioselectivity and Reactivity Through Homobimetallic Cooperativity. Science 1993, 260 (5115), 1784-1788. 37. Bosnich, B., Cooperative Bimetallic Redox Reactivity. Inorg. Chem. 1999, 38 (11), 2554-2562. 38. Nakane, T.; Tanioka, Y.; Tsukada, N., Synthesis of Multinuclear Copper Complexes Bridged by Diquinolylamidinates and Their Application to Copper-Catalyzed Coupling of Terminal Alkynes and Aryl, Allyl, and Benzyl Halides. Organometallics 2015, 34 (7), 1191- 1196. 39. Kuramochi, Y.; Ishitani, O.; Ishida, H., Reaction mechanisms of catalytic photochemical CO2 reduction using Re(I) and Ru(II) complexes. Coord. Chem. Rev. 2018, 373, 333-356. 40. Kulhánek, J.; Bureš, F.; Šimon, P.; Bernd Schweizer, W., Utilizing terpene derivatives in the synthesis of annulated terpene-imidazoles with application in the nitroaldol reaction. Tetrahedron: Asymmetry 2008, 19 (21), 2462-2469. 41. Phan, H.; Hrudka, J. J.; Igimbayeva, D.; Lawson Daku, L. M.; Shatruk, M., A Simple Approach for Predicting the Spin State of Homoleptic Fe(II) Tris-diimine Complexes. J. Am. Chem. Soc. 2017, 139 (18), 6437-6447. 42. Hawecker, J.; Lehn, J.-M.; Ziessel, R., Efficient photochemical reduction of CO2 to CO 2+ 2+ by visible light irradiation of systems containing Re(bipy)(CO)3X or Ru(bipy)3 –Co combinations as homogeneous catalysts. Chem. Commun. 1983, (9), 536-538. 43. Hawecker, J.; Lehn, J.-M.; Ziessel, R., Electrocatalytic reduction of carbon dioxide mediated by Re(bipy)(CO)3Cl (bipy = 2,2′-bipyridine). Chem. Commun. 1984, (6), 328-330. 44. Sullivan, B. P.; Bolinger, C. M.; Conrad, D.; Vining, W. J.; Meyer, T. J., One- and two- electron pathways in the electrocatalytic reduction of CO2 by fac-Re(bpy)(CO)3Cl (bpy = 2,2 ′-bipyridine). Chem. Commun. 1985, (20), 1414-1416.

James McPherson – June 2019 218 Chapter 6: Miscellanea

45. Hawecker, J.; Lehn, J.-M.; Ziessel, R., Photochemical and Electrochemical Reduction of Carbon Dioxide to Carbon Monoxide Mediated by (2,2 ′ - Bipyridine)tricarbonylchlororhenium(I) and Related Complexes as Homogeneous Catalysts. Helv. Chim. Acta 1986, 69 (8), 1990-2012. 46. Sahara, G.; Ishitani, O., Efficient Photocatalysts for CO2 Reduction. Inorg. Chem. 2015, 54 (11), 5096-5104. 47. Elgrishi, N.; Chambers, M. B.; Wang, X.; Fontecave, M., Molecular polypyridine-based metal complexes as catalysts for the reduction of CO2. Chem. Soc. Rev. 2017, 46 (3), 761-796. 48. Francke, R.; Schille, B.; Roemelt, M., Homogeneously Catalyzed Electroreduction of Carbon Dioxide—Methods, Mechanisms, and Catalysts. Chem. Rev. 2018, 118 (9), 4631-4701. 49. Clark, M. L.; Cheung, P. L.; Lessio, M.; Carter, E. A.; Kubiak, C. P., Kinetic and Mechanistic Effects of Bipyridine (bpy) Substituent, Labile Ligand, and Brønsted Acid on Electrocatalytic CO2 Reduction by Re(bpy) Complexes. ACS Catal. 2018, 8 (3), 2021-2029. 50. Ezzedinloo, L. Transition metal-organic hydride electrocatalysts for reduction of CO2. PhD, UNSW Sydney, Sydney, NSW, 2018. 51. Nomenclature of Inorganic Chemistry IUPAC Recommendations 2005. RSC Publishing: Cambridge UK, 2005. 52. Smieja, J. M.; Kubiak, C. P., Re(bipy-tBu)(CO)3Cl−improved Catalytic Activity for Reduction of Carbon Dioxide: IR-Spectroelectrochemical and Mechanistic Studies. Inorg. Chem. 2010, 49 (20), 9283-9289. 53. Fulmer, G. R.; Miller, A. J. M.; Sherden, N. H.; Gottlieb, H. E.; Nudelman, A.; Stoltz, B. M.; Bercaw, J. E.; Goldberg, K. I., NMR Chemical Shifts of Trace Impurities: Common Laboratory Solvents, Organics, and Gases in Deuterated Solvents Relevant to the Organometallic Chemist. Organometallics 2010, 29 (9), 2176-2179. 54. Harris, R. K.; Becker, E. D.; De Menezes, S. M. C.; Granger, P.; Hoffman, R. E.; Zilm, K. W., Further Conventions for NMR Shielding and Chemical Shifts (IUPAC Recommendations 2008). Magn. Reson. Chem. 2008, 46 (6), 582-598. 55. Battistin, F.; Balducci, G.; Demitri, N.; Iengo, E.; Milani, B.; Alessio, E., 15N NMR spectroscopy unambiguously establishes the coordination mode of the diimine linker 2-(2′- pyridyl)pyrimidine-4-carboxylic acid (cppH) in Ru(II) complexes. Dalton Trans. 2015, 44 (35), 15671-15682. 56. MestReNova, 12.0.0-20080; Mestrelab Research S.L.: 2017. 57. Cowieson, N. P.; Aragao, D.; Clift, M.; Ericsson, D. J.; Gee, C.; Harrop, S. J.; Mudie, N.; Panjikar, S.; Price, J. R.; Riboldi-Tunnicliffe, A.; Williamson, R.; Caradoc-Davies, T., MX1: a bending-magnet crystallography beamline serving both chemical and macromolecular crystallography communities at the Australian Synchrotron. J. Synchrotron Radiat. 2015, 22 (1), 187-190. 58. McPhillips, T. M.; McPhillips, S. E.; Chiu, H.-J.; Cohen, A. E.; Deacon, A. M.; Ellis, P. J.; Garman, E.; Gonzalez, A.; Sauter, N. K.; Phizackerley, R. P.; Soltis, S. M.; Kuhn, P., Blu- Ice and the Distributed Control System: software for data acquisition and instrument control at macromolecular crystallography beamlines. J. Synchrotron Radiat. 2002, 9 (6), 401-406. 59. Kabsch, W., XDS. Acta Cryst. D. 2010, 66 (2), 125-132. 60. Sheldrick, G., SHELXT - Integrated space-group and crystal-structure determination. Acta Cryst. A. 2015, 71 (1), 3-8. 61. Sheldrick, G., Crystal structure refinement with SHELXL. Acta Cryst. C. 2015, 71 (1), 3-8. 62. Dolomanov, O. V.; Bourhis, L. J.; Gildea, R. J.; Howard, J. A. K.; Puschmann, H., OLEX2: a complete structure solution, refinement and analysis program. J. Appl. Crystallogr. 2009, 42 (2), 339-341. 63. APEX3, SAINT and SADABS, Bruker AXS Inc.: 2016.

James McPherson – June 2019 219 Chapter 6: Miscellanea

64. Rees, B.; Jenner, L.; Yusupov, M., Bulk-solvent correction in large macromolecular structures. Acta Cryst. D. 2005, 61 (9), 1299-1301.

James McPherson – June 2019 220 Chapter 7: Conclusions and Outlook

7 CONCLUSIONS AND OUTLOOK

This thesis describes an exploration of the coordination chemistry of the 2,5-di(2-

pyridyl)pyrrolide (dpp–) anion aimed at providing new functional units with interesting physical

(e.g., magnetic or photophysical) or chemical (e.g. redox or reactive) properties. The purpose

of this chapter is to tie the conclusions which followed each of the previous chapters together

in the context of the overall aim—to unveil the rich chemistry that these ligands potentially

present.

A diverse range of Hdpp precursors are available with functionalisation possible from

one to all of the pyrrolide and pyridyl rings.1 Elaborate devices are thus achievable: for example,

Tour and co-workers had demonstrated well before this work began that [M(dpp)2] complexes

{M = Co(II) or Cu(I)}, sandwiched between two gold electrodes through thiocyanide

functionalisation at each pyrrolide, were excellent single molecule transistors, with high Kondo

Temperatures.2 The development of new devices demands better understanding of M-dpp

species, so that opportunities can be identified and solutions designed rationally. This PhD

research has predominantly focussed on the contributions of dpp– ligands to the useful

properties of their simple homoleptic [M(dpp)x] (M = transition metal(II) ion, x = 2; M =

lanthanide(III) ion, x = 3) complexes, and the ability to tune these properties through

appropriate substitution of the pyrrolide ring.

Chapter 2 describes the geometric consequences of a 5-membered central pyrrolide ring

in the dpp– ligand. Much greater internal strain was predicted (by DFT calculation) from

theoretical (DFT optimised) and solid state (SC-XRD) structures within dpp– ligands bound to

transition metal ions than for tpy ligands. A correlation was discovered between the strains

predicted by DFT calculations and the relative changes in separation of terminal donor atoms

James McPherson – June 2019 221 Chapter 7: Conclusions and Outlook from the unbound to bound ligand geometries (∆ρ) scaled for the original separation in the free ligand (i.e., ∆EL(strain) ∝ ∆ρ/ρL(relax)). This correlation was found to not only apply to pyridine- pyrrolide pincer ligands, but more generally to other three-ringed tridentate pincers. These geometric ρ parameters are easily measured from structures determined either experimentally

(e.g., from SC-XRD data), or estimated from theoretical calculations. Thus, ∆ρ/ρL(relax) provides a very useful proxy for internal strain built when a pincer ligand binds a metal ion.

This is important, as the strain will contribute to the physical properties of metal complexes. For example, eighteen different homoleptic iron(II) pincer complexes were discussed in Chapter 2, and the internal ligand strain explains, a posteriori, their different spin states and SCO activities/inactivities. The prediction of SCO activity has received much attention recently, however most studies attempt to homogenise (and thus ignore) the contribution of internal ligand strain by selecting structurally similar ligands and attempting to only vary electronic contributions.3-5 Shatruk’s group have recently demonstrated that ligand strain plays an important role in SCO activity for homoleptic bidentate iron(II) systems,6-7 and so the work described in Chapter 2 of this thesis provides an excellent tool to describe similar effects, but in the context of tridentate pincer ligands where they have been previously avoided.

Figure 7.1 depicts some generic strategies by which the ρL(relax) parameter, and more importantly ∆EL(strain), can be manipulated.

Figure 7.1: Generic strategies to influence ρL(relax) (shown in purple), and hence the internal ligand strain which builds on chelating metal ions, in three-ringed pincer systems. Elongated ρL(relax) generally result in increased

∆EL(strain).

James McPherson – June 2019 222 Chapter 7: Conclusions and Outlook

The analysis of ligand strain described in Chapter 2 predicts increased reactivity for

–1 [Ru(dpp)2] species (∆EL(strain) ≈ 75 kJ mol for each ligand) relative to the ubiquitous

2+ –1 [Ru(tpy)2] complexes (∆EL(strain) ≈ 25 kJ mol for each ligand). The increased strain in the bound dpp– ligands weakens binding of an outer pyridyl donor such that it is easily substituted; hemilability is observed. The successful prediction and exploitation of the internal ligand strain to achieve catalytic transformations of small molecules, as described in Chapter 5, is one of the key successes reported in this PhD thesis.

Alkylation of the pyridyl rings released due to the strain within the dpp– ligand (e.g., by

8 reacting [Ru(dpp)2] with MeI) was not attempted during this thesis, but would block the re- coordination of the unbound ring. As shown in red in Scheme 5.1, each resulting pyridinium ring is analogous to an organic hydride acceptor (OHD+), each capable of storing a proton and two electrons to generate the reduced dihydropyridine conjugates of the OHD+ cation which is shown in blue. These redox active units would be close to the labile coordination sites of the ruthenium ion, and so are well positioned for multi-electron proton coupled electron transfer

(PCET) reactions with small molecules bound to ruthenium.

Scheme 7.1: Alkylation of the pendant pyridyl ring, released due to the strain in the tridentate dpp– ligand, would block its re-coordination. Each pyridinium would be capable of storing a up to two electrons (with a proton), and thus proton coupled two or four electron redox reactions may be possible for bound substrates.

R1,R2 Chapter 3 describes the preparation of a family of five [Co(dpp )2] complexes and investigations of the influence exerted by the R1 and R2 substituents onto the physical properties

James McPherson – June 2019 223 Chapter 7: Conclusions and Outlook

(redox and magnetism) of the complexes. Different magnetic properties were observed for the

R1,R2 five [Co(dpp )2] complexes studied, including gradual SCO—as solids for four complexes, and also in chloroform-d solution for three complexes—however no useful ligand parameter

(or combination of parameters) could be found that accounted for the differences in magnetic properties. However, the potential of the Co(III)/Co(II) redox couple in these complexes was shown to respond predictably to the combined electronic nature of R1 and R2 (as described by

Swain and Lupton’s F and R parameters).

Paramagnetic ions, such as the d7 cobalt(II) ion, have been widely used as magnetic resonance imaging (MRI) contrast agents, which are generally restricted to discriminating different physical aspects of the biological sample (e.g., water content of different tissues or fluid pooling).9 The desire to probe chemical changes in biological systems has fuelled the development of responsive MRI contrast agents.10 New and co-workers have demonstrated that reversible switching between cobalt(III) (diamagnetic, “off”) and cobalt(II) (paramagnetic,

“on”) allows off-on monitoring of hypoxic tissue through MRI.11 The sensitivity of the

R1,R2 1 Co(III)/Co(II) couples in the [Co(dpp )2] complexes to the electronic influences of R and

R2 discussed in Chapter 3 thus represents an opportunity for responsive MR contrast agents that report the redox environment of a sample across a range of potentials (theoretically by more than 1.5 V). There is also the opportunity to install additional function (e.g., a targeting group)

1 III II 2 at R , and tune E1/2(Co /Co ) through R .

James McPherson – June 2019 224 Chapter 7: Conclusions and Outlook

Figure 7.2: [Co(dpp)2] complexes as MR contrast agents which report their local redox environment. As III II demonstrated in Chapter 3, the switching potential for the Co /Co couple (E1/2) can be tuned through the electronic influences of R1 R2.

The dpp– anion not only binds transition metal ions, but also rare-earth ions, as

CO Me,Me demonstrated for the first time by the series of 4f-lanthanide(III) [Ln(dpp 2 )3] complexes that are described in Chapter 4. These complexes were structurally characterised in the solid- state and in solution. The redox properties of the dpp– anions were retained on complexation with Ln(III) ions. The useful emissive properties of the red-emitting Eu(III) ion and the

NIR-emitting Yb(III) ion were enhanced, with sensitised emission caused by excitation of the

(dppCO2Me,Me)– ligand.

The sensitisation of the lanthanide(III) excitation was limited by the photophysical properties of the (dppCO2Me,Me)– ligand. Rather than relying on the dpp– ligand to both bind and sensitise the lanthanide(III) ion, the installation of an organic (or inorganic) dye onto the Hdpp scaffold (e.g., pyrene as shown in Figure 7.3) would improve the photophysical response of the antenna. A modular synthesis would allow a range of modified ligands to be prepared so as to match the different photophysical requirements of targeted lanthanide(III) ions. The lability of the dpp– ligand to exchange for water is also a challenge that must be addressed if these complexes are to find use as dyes for biological applications. Conversion of the pyridyl rings

James McPherson – June 2019 225 Chapter 7: Conclusions and Outlook

to pyridinium N-oxides, as shown at the bottom of Figure 7.2, would reduce ρL(relax), and so

∆EL(strain) and substitutional lability, for the oxophilic lanthanide(III) ions.

Figure 7.3: Strategies to improve the performance of the [Ln(dpp)3] complexes described in Chapter 4. Tethering an organic (or inorganic) dye, such as pyrene as shown, to the dpp– ligand may enhance the emission from the Ln(III). A modular synthesis would represent a strategy to manipulate the antennae excited states. Ligand exchange kinetics may be reduced for complexes of the pyridinium N-oxides.

Chapter 6 on miscellanea describes four small preliminary studies, each worthy of more complete further investigations. First-row [M(dpp)2] complexes (M = Ni, Cu or Zn) were described in Chapter 6.1, and the response of the [Cu(dpp)2] complexes under reducing

+ conditions in the presence of CO2 and H requires further investigation. The dicarboxylate

CO ,CO 3– (z+) CO ,CO z–6 pyrrolide, (dpp 2 2) , represents an entry point to [M (dpp 2 2)2] metallo-ligands for supramolecular elaboration, and a porous polymeric cobalt(III) species,

CO ,CO {[Co(Hdpp 2 2)2]Na}n, has been described in Chapter 6.2. Incorporation of rigid structural building units with these 4-connecting carboxylate metallo-linkers could result in well-defined

(z+) CO ,CO z–6 pores with unprecedented geometry imposed by the [M (dpp 2 2)2] metallo-ligand.

Facial ancillary ligands, e.g. pentamethyl cyclopentadienyl (Cp*), are incompatible with the

James McPherson – June 2019 226 Chapter 7: Conclusions and Outlook tridentate binding mode of the dpp– ligand, and Chapter 6.3 describes new [Cp*M(κ2-dpp)Cl]

(M = Rh or Ir) complexes with unbound pyridyl rings. The CVs recorded for

[Cp*Rh(κ2-dpp)Cl] complexes display unusual behaviour, and preliminary experiments in the presence of water indicate that these complexes facilitate proton coupled electroreduction.

Finally, Chapter 6.4 describes some preliminary studies on LImPy, a different ligand scaffold altogether, that was isolated as a side-product during several Hdpp syntheses. A homo bimetallic rhenium(I) electrocatalyst, [LImPy{Re(CO)3Cl}2], was prepared, with promising

+ preliminary electrocatalytic activity observed in CVs recorded in the presence of CO2 and H .

2+ The two binding bidentate binding sites at LImPy can be addressed selectively by [Cp*IrCl] units, and so cooperative heterobimetallic species bridged by LImPy await discovery.

Overall, the work described in this thesis reveals that the chemistry of M-dpp complexes is ripe for development.

For example, as summarised pictorially in Figure 7.4, all studies of M-dpp complexes to date have focussed on late-transition metal ions (Groups 8–12), and the work described in

Chapter 4 adds the 4f lanthanide ions, with their interesting emissive properties. Other pyridine- pyrrolide ligands (particularly 2,6-di-(2-pyrrolide)pyridyl dianions, C2–, as discussed in section

3 in Chapter 1) have been shown to bind early-transition metal ions (e.g., Ti and Zr) and main- group metal ions (e.g., Pb and Sn) to produce stable complexes with favourable photophysical properties.1 Pyridine-pyrrolide macrocycles have also been shown to bind 5f actinide ions (e.g.,

U, Np and Pu).12 Similar studies have yet to be explored for the dpp– ligand, and await an intrepid chemist.

James McPherson – June 2019 227 Chapter 7: Conclusions and Outlook

Figure 7.4: Periodic table of the elements revealing the “stamp collection” of known dpp– complexes. Metal ions that have been bound by dpp– ions and described in this work are coloured in blue, while those not prepared in this work and reported by others are coloured red. Other metal ions for which no known dpp– complex is known, but complexes with related ligands have been reported are coloured green (for di-(2-pyrrolide)pyridyl derivatives) or yellow (for pyridine-pyrrolide macrocycles). Image adapted from www.iupac.org (accessed March 2019).

With the properties of the [M(dpp)2] complexes now well understood, their incorporation into useful devices requires some organic synthesis. It would be most desirable to develop a strategy by which additional function could be connected to a particular [M(dpp)2] species. The carboxylic anhydride pyrrole, Hdpp(CO)2O, which is readily prepared by adapting Hein’s original recipes,13 may represent a suitable starting material for such transformations, as shown in

Scheme 7.2. It is expected that treatment with primary amines will result in conversion of the cyclic anhydride to the imide, and some examples of interesting dpp– ligand targets from readily accessible primary amines are shown at the bottom of Scheme 7.2.

James McPherson – June 2019 228 Chapter 7: Conclusions and Outlook

Scheme 7.2: Treatment of Hdpp(CO)2O with primary amines should result in conversion to imides. Some examples are shown below. Left to right: two Hdpp moieties linked by a pyrene dye from 1,6-diaminopyrene; conjugation of two metal binding moieties, 2,2'-bipyridine (bpy) with dpp– for hetero-bimetallics; the zwitterionic ammonium- sulfonate enhances aqueous solubility;14 and the alkyne represents an additional synthetic handle, e.g. for Huisgen azide-alkyne (“click”) 1,3-cycloadditions.15

In closing, the coordination chemistry of the dpp– family of ligands has matured as a result of this PhD research. Useful properties for simple complexes were influenced predictably through modifications to the dpp– scaffold. The groundwork has been laid that enables the design of specific M-dpp complexes to meet specific constraints, and supramolecular elaborations of these complexes offer additional tantalising possibilities. It is hoped that this work will inspire others to apply their own creativity to these underappreciated ligands.

James McPherson – June 2019 229 Chapter 7: Conclusions and Outlook

7.1 References

1. McPherson, J. N.; Das, B.; Colbran, S. B., Tridentate pyridine–pyrrolide chelate ligands: An under-appreciated ligand set with an immensely promising coordination chemistry. Coord. Chem. Rev. 2018, 375, 285-332. 2. Ciszek, J. W.; Keane, Z. K.; Cheng, L.; Stewart, M. P.; Yu, L. H.; Natelson, D.; Tour, J. M., Neutral Complexes of First Row Transition Metals Bearing Unbound Thiocyanates and Their Assembly on Metallic Surfaces. J. Am. Chem. Soc. 2006, 128 (10), 3179-3189. 3. Kershaw Cook, L. J.; Kulmaczewski, R.; Mohammed, R.; Dudley, S.; Barrett, S. A.; Little, M. A.; Deeth, R. J.; Halcrow, M. A., A Unified Treatment of the Relationship Between Ligand Substituents and Spin State in a Family of Iron(II) Complexes. Angew. Chem. Int. Ed. 2016, 55 (13), 4327-4331. 4. Rodríguez-Jiménez, S.; Yang, M.; Stewart, I.; Garden, A. L.; Brooker, S., A Simple Method of Predicting Spin State in Solution. J. Am. Chem. Soc. 2017, 139 (50), 18392-18396. 5. Kimura, A.; Ishida, T., Spin-Crossover Temperature Predictable from DFT Calculation for Iron(II) Complexes with 4-Substituted Pybox and Related Heteroaromatic Ligands. ACS Omega 2018, 3 (6), 6737-6747. 6. Phan, H.; Hrudka, J. J.; Igimbayeva, D.; Lawson Daku, L. M.; Shatruk, M., A Simple Approach for Predicting the Spin State of Homoleptic Fe(II) Tris-diimine Complexes. J. Am. Chem. Soc. 2017, 139 (18), 6437-6447. 7. Hrudka, J. J.; Phan, H.; Lengyel, J.; Rogachev, A. Y.; Shatruk, M., Power of Three: Incremental Increase in the Ligand Field Strength of N-Alkylated 2,2′-Biimidazoles Leads to Spin Crossover in Homoleptic Tris-Chelated Fe(II) Complexes. Inorg. Chem. 2018, 57 (9), 5183-5193. 8. Amoroso, A. J.; Banu, A.; Coogan, M. P.; Edwards, P. G.; Hossain, G.; Malik, K. M. + A., Functionalisation of terpyridine complexes containing the Re(CO)3 moiety. Dalton Trans. 2010, 39 (30), 6993-7003. 9. De León-Rodríguez, L. M.; Martins, A. F.; Pinho, M. C.; Rofsky, N. M.; Sherry, A. D., Basic MR relaxation mechanisms and contrast agent design. J. Magn. Reson. Imag. 2015, 42 (3), 545-565. 10. Hingorani, D. V.; Bernstein, A. S.; Pagel, M. D., A review of responsive MRI contrast agents: 2005–2014. Contrast Media Mol. Imaging 2015, 10 (4), 245-265. 11. O'Neill, E. S.; Kolanowski, J. L.; Yin, G. H.; Broadhouse, K. M.; Grieve, S. M.; Renfrew, A. K.; Bonnitcha, P. D.; New, E. J., Reversible magnetogenic cobalt complexes. RSC Adv. 2016, 6 (36), 30021-30027. 12. Das, B.; McPherson, J. N.; Colbran, S. B., Oligomers and macrocycles with [m]pyridine[n]pyrrole (m + n ≥ 3) domains: Formation and applications of anion, guest molecule and metal ion complexes. Coord. Chem. Rev. 2018, 363, 29-56. 13. Hein, F. M., F., Synthesis of 2,5-di(α-pyridyl)pyrrole and of several complex derivatives of 2,5-di(α-pyridyl)-3,4-dicarbethoxypyrrole. Die Pharmazie 1954, 9, 455-460. 14. García, K. P.; Zarschler, K.; Barbaro, L.; Barreto, J. A.; O'Malley, W.; Spiccia, L.; Stephan, H.; Graham, B., Zwitterionic-Coated “Stealth” Nanoparticles for Biomedical Applications: Recent Advances in Countering Biomolecular Corona Formation and Uptake by the Mononuclear Phagocyte System. Small 2014, 10 (13), 2516-2529. 15. Huisgen, R., 1,3-Dipolar Cycloadditions. Proc. Chem. Soc. 1961, (October), 357-396.

James McPherson – June 2019 230 Chapter 8: Appendices

8 APPENDICES

James McPherson – June 2019 8-231 Chapter 8: Appendices

James McPherson – June 2019 8-232 Chapter 8: Appendices

APPENDIX 1 – SUPPORTING INFORMATION FOR CHAPTER 2

James McPherson – June 2019 8-233 Chapter 8: Appendices

Supporting Information

A strain-deformation nexus within pincer ligands: application to the spin states of iron(II) complexes

James N. McPherson, Timothy E. Elton and Stephen B. Colbran*

School of Chemistry, UNSW Sydney, NSW, 2052, Australia

*Author for correspondence: [email protected]

James McPherson – June 2019 8-234 Chapter 8: Appendices

Table S1. Geometric parameters for hypothetical [Zn(L)]2+ complexes optimised without empirical dispersion correction (B3LYP/6-31G**)

Zn–N(t) Zn–N(c) θ ρ μ rms N(t)–Zn–N(c) N(t)–Zn–N′(t) [py/pyr] [py/pyr] /° /Å /Å [Zn(A)]2+ 1.981 [py] 1.932 [py] 83.54 167.08 98 3.94 0 2.001† [py] 81.78† [Zn(B)]1+ 1.927 [py] 167.8 96.3 3.857 0 1.879 [pyr] 86.02 [Zn(C)] 1.901 [pyr] 1.944 [py] 84.26 168.53 94 3.78 0 [Zn(D)]1+ 2.024 [py] 1.860 [pyr] 83.76 159.31 106 3.98 0.225 82.89† 2.022 [py] [Zn(E)] 1.884 [pyr] 159.4 102.4 3.873 0.255 1.915 [pyr] 86.87 †for non-symmetric ligands B– and E2–, the terminal Zn–N(pyridyl) distances and angles are given first and are followed by the terminal Zn–N(pyrrolide) distances.

Fig. S1. Chelate strain compared to coordination strain as defined in this work for the exemplar terpyridyl (A) ligand scaffold.

Table S2. Geometric parameters (ρL(relax), θL(relax) and μrms–L(relax)) and Energy (ΔEL(relax)) calculated (B3LYP/6- 31G**) for relaxed ligand backbones (with inter-ring N-C-C-N torsions fixed at 0°).

θ L(relax) ρ L(relax) μ rms–L(relax) ΔEL(relax) Entry ligand /° /Å /Å /Eh 1 A 113.3 4.514 0 –742.466 2 B– 116.4 4.766 0 –703.761 2– 3 C 120.0 5.071 0 –664.943 2– 4 (C2) 115.0 4.711 0.060 –1205.778 2– 5 (C3) 119.6 5.004 0 –972.331 – 6 D 139.3 5.322 0 –703.760 – 7 (D2) 128.6 4.879 0.044 –971.007 2– 8 E 143.7 5.625 0 –664.907

2 James McPherson – June 2019 8-235 Chapter 8: Appendices

Fig. S2. Fits of DFT-calculated ligand strain energy (B3LYP/6-31G**) for ligands (A–E2–) in the hypothetical [Zn(L)]z+ complexes against their different relative deformation parameters (data points represented as black squares, with the equation and correlation of the fit presented). DFT-calculated strain energies (empty triangles) for experimental, solid-state structures from the CSD (see Table S3) have been plotted on top of the fits as empty triangles: A red, (C2)2– purple, (C3)2– mustard, D– green and (D2)– in blue.

3 James McPherson – June 2019 8-236 Chapter 8: Appendices

Table S3. Relative deformations (Δρ/ρL(relax), Δθ/θL(relax) and Δμrms) and calculated strain energies (B3LYP/6- 31G**) for ligand backbones from vacuum phase [Zn(L)]z+ geometries (B3LYP-GD3BJ/6-31G**) and SC- XRD structures from the CSD database (with CCDC identifier codes as listed).

Complex θ Δθ / θ L(relax) ρ Δρ / ρ L(relax) Δμ rms Strain Entry ligand -1 (CCDC code) /° /% /Å /% /Å /kJ mol 2+ [Zn(A)] 1 A 98.1 13.3 3.929 13.0 0 38

+ [Zn(B)] – 2 B 96.3 17.3 3.850 19.2 0 67

[Zn(C)] 2– 3 C 94.0 21.7 3.777 25.5 0 113

+ [Zn(D)] – 4 D 105.0 18.9 3.963 25.5 0.236 118

[Zn(E)] 2– 5 E 101.9 22.2 3.858 31.4 0.266 170

[Fe(A)2] 2[ClO4] 6 1 A 100.9(6) 10.9 3.925(9) 13.0 0.039 41 (DANMOU) [Fe(A)2] 2[ClO4] 7 1 A 100.6(6) 11.1 3.918(8) 13.2 0.054 52 (DANMOU) [Ru(A)2] 2[ClO4] 8 2 A 104.4(3) 7.8 4.065(4) 9.9 0.043 27 (BENHUZ) [Ru(A)2] 2[ClO4] 9 2 A 103.8(2) 8.3 4.066(4) 9.9 0.075 25 (BENHUZ) [Ru(D2)(A)] 10 3 A 103.1(7) 8.9 4.04(1) 10.5 0.054 59 (LOGKUP) [Os(A)2] 2[ClO4] 11 4 A 101.2(9) 10.6 4.047(11) 10.3 0.049 47 (GOGDOV) [Os(A)2] 2[ClO4] 12 4 A 100.4(9) 11.3 4.035(11) 10.6 0.035 58 (GOGDOV) 2 [Zr(C )2] 2 2– 13 5 (C ) 104.9(4) 8.8 4.078(4) 13.4 0.073 66 (YAHBIV) 2 [Zr(C )2] 2 2– 14 5 (C ) 103.6(4) 9.9 4.048(4) 14.1 0.140 57 (YAHBIV) 3 [Pb(C )(Ph)2] 3 2– 15 6 (C ) 108.2(6) 9.5 4.328(8) 13.5 0.013 76 (ALILUC) 3 [Pb(C )(Ph)2] 3 2– 16 6 (C ) 109.7(4) 8.3 4.348(8) 13.1 0.082 72 (ALILUC) 3 [Pd(C )(py)] 3 2– 17 7 (C ) 101.5(3) 15.1 3.967(5) 20.7 0.052 103 (XOVFIX) [Pd(D)Cl] – 18 8 D 110.1(7) 21.0 4.052(8) 23.9 0.013 120 (TELPUX) 2 [Pd(D )Cl] 2 – 19 3 (D ) 108.68(18) 15.5 4.025(3) 17.5 0.016 87 (LOGKOJ) [Pt(D)Cl] – 20 8 D 108.5(5) 22.1 4.027(5) 24.3 0.020 121 (TELQEI) [Cu(D)Cl] – 21 9 D 114.7(3) 17.7 4.176(4) 21.5 0.072 80 (KANXIJ) [Ru(D)(COD)Cl] – 22 10 D 113.6(2) 18.4 4.169(3) 21.7 0.090 88 (AYOSUE) 2 [Ru(D )(A)] 2 – 23 3 (D ) 111.0(7) 13.7 4.08(1) 16.4 0.088 79 (LOGKUP) 2 [Ru(D )2] 2 – 24 3 (D ) 109.3(4) 15.7 4.066(6) 16.7 0.070 91 (LOGLEA) 2 [Ru(D )2] 2 – 25 3 (D ) 109.6(4) 16.1 4.077(5) 16.4 0.112 75 (LOGLEA)

Fig. S3. The structures of the derivatives of C2– and D– in Table S3.

4 James McPherson – June 2019 8-237 Chapter 8: Appendices

Table S4. Relative deformations (Δρ/ρL(relax), Δθ/θL(relax) and Δμrms) and calculated strain energies (B3LYP/6- 31G**) for alternate ligand backbones from SC-XRD structures from the CSD database (with CCDC identifier codes as listed). The residual error is defined as the difference remaining when the calculated value is subtracted from the predicted value.

Donors Δθ / θ Δρ / Strain Complex θ ρ Δμ rms - Residual Entry Ring size L(relax) ρ L(relax) /kJ mol –1 (CCDC code) /° /Å /Å 1 /kJ mol /% /% [Pt(Pz2Ph)(NN)] N-C-N 1 11 100.2(5) 19.3 3.976(7) 21.9 0.046 114 –13 (KEPSUV) 5-6-5 [Pd(Im2Py)Br]Br C-N-C 2 12 101.3(4) 11.1 4.006(7) 14.5 0.013 69 –2 (WOXFEU) 5-6-5 [Cr(Im2Py)Cl3] C-N-C 3 13 103.3(3) 9.4 4.067(5) 13.2 0.018 59 2 (YOKGUB) 5-6-5 [Au(Ph2Py)(C≡CR)] C-N-C 4 14 103.4(5) 12.0 4.080(8) 16.9 0.019 90 –12 (CEXYIP) 6-6-6 N-N-N [Pd(NNN)(Me)]BAr′4 5 15 6-6-6 106.4(3) 10.1 4.134(6) 12.5 0.027 46 12 (PUHKOS) (fused) LS [Fe(pybox)2] 2[PF6] 6 16 5-6-5 96.3(5) 16.8 3.956(6) 21.7 0.036 92 5 (IZAFOG) HS [Fe(pybox)2] 7 2[PF6] 5-6-5 101.4(3) 12.4 4.205(4) 16.2 0.054 52 23 (IZAFOG01)16 [Ir(PNP)H3] P-N-P 8 17 115.5(7) 2.2 4.474(4) 4.3 0.169 46 –26 (XAMQUZ) 6 [Cu(SNS)Cl]2[Cu2Cl6] S-N-S 9 18 109.2(2) 2.5 4.5122(11) 2.8 0.166 –17 30 (ACAROM) 6

Fig. S4. The structures of the pincer ligands in Table S4.

5 James McPherson – June 2019 8-238 Chapter 8: Appendices

Table S5. Metric parameters (ρL(coord) and Fe–N bond distances measured directly from solid state structure; ρL(relax) from B3LYP/6-31G** optimised geometry) and predicted strain energies for alternate ligand backbones from SC-XRD structures of iron(II) complexes from the CSD database (with CCDC identifier codes as listed). Predicted strain was calculated according to equation 2. The structure of each ligand is represented in Figure 7.

Complex ρ L(coord) ρ L(relax) Fe–Nt Fe–Nc Fe–Nav Strain Predicted Entry Ring Sizes Spin State: -1 (CCDC code) /Å /Å /Å /Å /Å /kJ mol [Fe(A)2] 2[ClO4] 1 1 6-6-6 LS 3.925(9) 4.514 1.990(6) 1.890(6) 1.956(6) 60 (DANMOU) [Fe(A)2] 2[ClO4] 2 1 6-6-6 LS 3.918(8) 4.514 1.986(6) 1.891(6) 1.954(6) 61 (DANMOU) [Fe(thiaz2Py)2] 2[ClO4] 3 19 5-6-5 LS 3.899(4) 4.655 1.973(3) 1.903(3) 1.950(3) 75 (DIXDIX) [Fe(thiaz2Py)2] 2[ClO4] 4 19 5-6-5 LS 3.897(4) 4.655 1.974(3) 1.910(3) 1.952(3) 75 (DIXDIX) Ph [Fe(bpyt )2] 2[BF4] 5 20 6-6-6 LS 3.911(2) 4.774 1.986(2) 1.864(1) 1.945(2) 83 (BUKDUH) Ph [Fe(bpyt )2] 2[BF4] 6 20 6-6-6 LS 3.907(2) 4.774 1.984(2) 1.870(1) 1.946(2) 84 (BUKDUH) [Fe(phen-bi)2] 6-6-5 7 21 LS 3.946(4) 4.922 2.002(3) 1.891(3) 1.965(3) 91 (DOMQIH) (fused) [Fe(phen-bi)2] 6-6-5 8 21 LS 3.931(4) 4.922 1.995(2) 1.892(2) 1.960(2) 93 (DOMQIH) (fused) F [Fe(bpzp )2] 2[BF4] 9 22 5-6-5 LS 3.89(1) 4.559 1.974(8) 1.904(7) 1.951(8) 78 (EKAKIM01) F [Fe(bpzp )2] 2[BF4] 10 22 5-6-5 LS 3.90(1) 4.559 1.981(8) 1.903(7) 1.955(8) 66 (EKAKIM01) F [Fe(bpzp )2] 2[BF4] 11 22 5-6-5 HS 4.150(5) 4.559 2.165(4) 2.113(4) 2.147(4) 41 (EKAKIM) F [Fe(bpzp )2] 2[BF4] 12 22 5-6-5 HS 4.198(5) 4.559 2.189(4) 2.126(4) 2.168(4) 36 (EKAKIM) Me [Fe(bpz p)2] 2[BF4] 13 23 5-6-5 LS 3.929(4) 4.561 1.997(3) 1.891(2) 1.962(3) 64 (AFANEB01) Me [Fe(bpz p)2] 2[BF4] 14 23 5-6-5 LS 3.931(4) 4.561 1.992(3) 1.891(3) 1.958(3) 64 (AFANEB01) Me [Fe(bpz p)2] 2[BF4] 15 23 5-6-5 HS 4.160(4) 4.561 2.158(3) 2.084(3) 2.133(3) 40 (AFANEB01) Me [Fe(bpz p)2] 2[BF4] 16 23 5-6-5 HS 4.146(4) 4.561 2.150(3) 2.079(3) 2.126(3) 42 (AFANEB01) Me [Fe(bpz p)2] 2[BF4] 17 23 5-6-5 HS 4.182(5) 4.561 2.189(3) 2.125(3) 2.167(3) 38 (AFANEB01) Me [Fe(bpz p)2] 2[BF4] 18 23 5-6-5 HS 4.197(4) 4.561 2.188(3) 2.131(2) 2.169(3) 37 (AFANEB01) SAc [Fe(bpyt )2] 2[BF4] 19 20 6-6-6 LS 3.931(4) 4.772 1.998(3) 1.870(3) 1.955(3) 81 (BUKDIV) SAc [Fe(bpyt )2] 2[BF4] 20 20 6-6-6 LS 3.930(5) 4.772 1.998(4) 1.867(3) 1.954(4) 81 (BUKDIV) [Fe(phen-tetrazolide)2] 6-6-5 21 24 LS 3.938(2) 4.992 2.0017(15) 1.9064(14) 1.9699(15) 97 (QIDJET01) (fused) [Fe(phen-tetrazolide)2] 6-6-5 22 24 HS 4.259(3) 4.992 2.2163(19) 2.1204(18) 2.1843(19) 68 (QIDJET) (fused) [Fe(pybox)2] 2[PF6] 23 16 5-6-5 LS 3.927(6) 5.018 2.000(4) 1.935(5) 1.978(5) 100 (IZAFOG) [Fe(pybox)2] 2[PF6] 24 16 5-6-5 LS 3.956(6) 5.018 2.016(5) 1.938(5) 1.990(5) 97 (IZAFOG) [Fe(pybox)2] 2[PF6] 25 16 5-6-5 HS 4.205(4) 5.018 2.186(3) 2.117(3) 2.163(3) 75 (IZAFOG01) [Fe(pybox)2] 2[PF6] 26 16 5-6-5 HS 4.224(5) 5.018 2.195(3) 2.108(3) 2.166(3) 73 (IZAFOG01) NMe2 [Fe(bpzp )2] 2[BF4] 27 22 5-6-5 HS 4.207(3) 4.565 2.200(2) 2.112(2) 2.170(2) 36 (EKAJUX) NMe2 [Fe(bpzp )2] 2[BF4] 28 22 5-6-5 HS 4.162(3) 4.565 2.184(2) 2.113(2) 2.160(2) 41 (EKAJUX) NMe2 [Fe(bpzt )2] 2[BF4] 29 25 5-6-5 HS 4.257(4) 4.795 2.244(3) 2.108(3) 2.199(3) 52 (FEDRUE) 3 [Fe(D )2] 30 26 6-5-6 HS 4.375(11) 4.918 2.286(8) 2.016(8) 2.196(8) 51 (UCIGUJ) 3 [Fe(D )2] 31 26 6-5-6 HS 4.331(10) 4.918 2.265(7) 2.038(7) 2.189(7) 55 (UCIGUJ) [Fe(Py2thiaz)2] 2[BF4] 32 27 6-5-6 HS 4.35(2) 5.064 2.283(15) 2.047(6) 2.204(15) 65 (PUBCAQ) [Fe(bppz)2] 2[BF4] 33 28 6-5-6 HS 4.349(8) 5.075 2.308(6) 2.020(8) 2.212(8) 66 (ZAJGOG)

6 James McPherson – June 2019 8-239 Chapter 8: Appendices

II z+ Fig. S5. Plots of selected Fe-N distances from homoleptic [Fe (L)2] complexes selected from the CSD (see Table S4) against the calculated (B3LYP/6-31G**) terminal N…N donor separations for the relaxed

(unbound) ligands (ρL(relax)). (a) mean Fe–Nterminal distances; (b) mean Fe–Ncentral distances; (c) mean all Fe– N distances. Red symbols represent HS iron(II) complexes, while blue symbols are LS configurations. SCO inactive complexes are represented by squares, and those with SCO activity are indicated by triangles.

References 1. Baker, A.; Goodwin, H., Crystal Structure of Bis(2,2':6',2"-terpyridine)iron(II) Bis(perchlorate) Hydrate. Aust. J. Chem. 1985, 38, 207-214. 2. Kozlowska, M.; Rodziewicz, P.; Brus, D. M.; Breczko, J.; Brzezinski, K., Bis(2,2':6',2''- terpyridine)ruthenium(II) bis(perchlorate) hemihydrate. Acta Cryst. E 2012, 68, m1414-m1415. 3. McSkimming, A.; Diachenko, V.; London, R.; Olrich, K.; Onie, C. J.; Bhadbhade, M. M.; Bucknall, M. P.; Read, R. W.; Colbran, S. B., An Easy One-Pot Synthesis of Diverse 2,5-Di(2-pyridyl)pyrroles: A Versatile Entry Point to Metal Complexes of Functionalised, Meridial and Tridentate 2,5-Di(2- pyridyl)pyrrolato Ligands. Chem. Eur. J. 2014, 20, 11445-11456. 4. Craig, D. C.; Scudder, M. L.; McHale, W.-A.; Goodwin, H. A., Structural Studies of Complexes of Tridentate Terimine Systems. Crystal Structure of Bis(2,2′:6′,2′′-terpyridine)ruthenium(II) Perchlorate Hydrate, Bis(2,2′:6′,2′′-terpyridine)- osmium(II) Perchlorate Hemihydrate and Bis((1,10-phenanthrolin-2- yl)(pyridin-2-yl)amine)iron(II) Tetrafluoroborate Dihydrate. Aust. J. Chem. 1999, 51, 1131-1140. 5. Zhang, Y.; Petersen, J. L.; Milsmann, C., A Luminescent Zirconium(IV) Complex as a Molecular Photosensitizer for Visible Light Photoredox Catalysis. J. Am. Chem. Soc. 2016, 138 (40), 13115-13118. 6. Jia, W.-L.; Liu, Q.-D.; Wang, R.; Wang, S., Novel Phosphorescent Cyclometalated Organotin(IV) and Organolead(IV) Complexes of 2,6-Bis(2′-indolyl)pyridine and 2,6-Bis[2′-(7-azaindolyl)]pyridine. Organometallics 2003, 22, 4070-4078. 7. Liu, Q.; Thorne, L.; Kozin, I.; Song, D.; Seward, C.; D'Iorio, M.; Tao, Y.; Wang, S., New red-orange phosphorescent/electroluminescent cycloplatinated complexes of 2,6-bis(2[prime or minute]- indolyl)pyridine. Dalton Trans. 2002, 3234-3240. 8. Imler, G. H.; Lu, Z.; Kistler, K. A.; Carroll, P. J.; Wayland, B. B.; Zdilla, M. J., Complexes of 2,5- Bis(α-pyridyl)pyrrolate with Pd(II) and Pt(II): A Monoanionic Iso-π-Electron Ligand Analog of Terpyridine. Inorg. Chem. 2012, 51, 10122-10128. 9. Min, R.; Hu, X.-h.; Yi, X.-y.; Zhang, S.-c., Synthesis, structure, DNA binding and cleavage activity of a new copper(II) complex of bispyridylpyrrolide. J. Cent. South Univ. (Engl. Ed.) 2015, 22, 1619-1625. 10. Zhong, Y.-Q.; Xiao, H.-Q.; Yi, X.-Y., Synthesis, structural characterization and catalysis of ruthenium(II) complexes based on 2,5-bis(2'-pyridyl)pyrrole ligand. Dalton Trans. 2016, 45, 18113-18119. 11. Zhang, X.; Suzuki, S.; Kozaki, M.; Okada, K., NCN Pincer–Pt Complexes Coordinated by (Nitronyl Nitroxide)-2-ide Radical Anion. J. Am. Chem. Soc. 2012, 134, 17866-17868. 12. Peris, E.; Loch, J. A.; Mata, J.; Crabtree, R. H., A Pd complex of a tridentate pincer CNC bis- carbene ligand as a robust homogenous Heck catalyst. Chem. Commun. 2001, 201-202. 13. McGuinness, D. S.; Suttil, J. A.; Gardiner, M. G.; Davies, N. W., Ethylene Oligomerization with Cr−NHC Catalysts: Further Insights into the Extended Metallacycle Mechanism of Chain Growth. Organometallics 2008, 27, 4238-4247. 7 James McPherson – June 2019 8-240 Chapter 8: Appendices

14. Xiao, X. S.; Kwong, W. L.; Guan, X.; Yang, C.; Lu, W.; Che, C. M., Platinum(II) and Gold(III) Allenylidene Complexes: Phosphorescence, SelfAssembled Nanostructures and Cytotoxicity. Chem. Eur. J. 2013, 19, 9457-9462. 15. Groen, J. H.; Zwart, A. d.; Vlaar, M. J. M.; Ernsting, J. M.; Leeuwen, P. W. N. M. v.; Vrieze, K.; Kooĳman, H.; Smeets, W. J. J.; Spek, A. L.; Budzelaar, P. H. M.; Xiang, Q.; Thummel, R. P., Insertion Reactions into Palladium–Carbon Bonds of Complexes Containing Terdentate Nitrogen Ligands; Experimental and Ab initio MO Studies. Eur. J. Inorg. Chem. 1998, 1129-1143. 16. Zhu, Y.-Y.; Li, H.-Q.; Ding, Z.-Y.; Lu, X.-J.; Zhao, L.; Meng, Y.-S.; Liu, T.; Gao, S., Spin transitions 2+ in a series of [Fe(pybox)2] complexes modulated by ligand structures, counter anions, and solvents. Inorg. Chem. Front. 2016, 3, 1624-1636. 17. Tanaka, R.; Yamashita, M.; Nozaki, K., Catalytic Hydrogenation of Carbon Dioxide Using Ir(III)−Pincer Complexes. J. Am. Chem. Soc. 2009, 131, 14168-14169. 18. Kapoor, R.; Kataria, A.; Venugopalan, P.; Kapoor, P.; Corbella, M.; Rodríguez, M.; Romero, I.; Llobet, A., New Tetranuclear Cu(II) Complexes: Synthesis, Structure, and Magnetic Properties. Inorg. Chem. 2004, 43, 6699-6706. 19. Baker, A.; Goodwin, H., Iron(II) and Nickel(II) Complexes of 2,6-Di(Thiazol-4-Yl)Pyridine and Related Ligands: Magnetic, Spectral and Structural Studies. Aust. J. Chem. 1986, 39 (2), 209-220. 20. Hain, S. K.; Heinemann, F. W.; Gieb, K.; Müller, P.; Hörner, G.; Grohmann, A., On the Spin Behaviour of Iron(II)–Dipyridyltriazine Complexes and Their Performance as Thermal and Photonic Spin Switches. Eur. J. Inorg. Chem. 2010, 221-232. 21. Seredyuk, M.; Znovjyak, K. O.; Kusz, J.; Nowak, M.; Munoz, M. C.; Real, J. A., Control of the spin state by charge and ligand substitution: two-step spin crossover behaviour in a novel neutral iron(ii) complex. Dalton Trans. 2014, 43, 16387-16394. 22. Cook, L. J. K.; Kulmaczewski, R.; Mohammed, R.; Dudley, S.; Barrett, S. A.; Little, M. A.; Deeth, R. J.; Halcrow, M. A., A Unified Treatment of the Relationship Between Ligand Substituents and Spin State in a Family of Iron(II) Complexes. Angew. Chem. Int. Ed. 2016, 55, 4327-4331. 23. Money, V. A.; Carbonera, C.; Elhaïk, J.; Halcrow, M. A.; Howard, J. A. K.; Létard, J. F., Interplay Between Kinetically Slow Thermal SpinCrossover and Metastable HighSpin State Relaxation in an Iron(II) Complex with Similar T1/2 and T(LIESST). Chem. Eur. J. 2007, 13, 5503-5514. 24. Zhang, W.; Zhao, F.; Liu, T.; Yuan, M.; Wang, Z.-M.; Gao, S., Spin Crossover in a Series of Iron(II) Complexes of 2-(2-Alkyl-2H-tetrazol-5-yl)-1,10-phenanthroline: Effects of Alkyl Side Chain, Solvent, and Anion. Inorg. Chem. 2007, 46, 2541-2555. 25. Capel Berdiell, I.; Kulmaczewski, R.; Halcrow, M. A., Iron(II) Complexes of 2,4-Dipyrazolyl-1,3,5- triazine Derivatives—The Influence of Ligand Geometry on Metal Ion Spin State. Inorg. Chem. 2017, 56, 8817-8828. 26. Ciszek, J. W.; Keane, Z. K.; Cheng, L.; Stewart, M. P.; Yu, L. H.; Natelson, D.; Tour, J. M., Neutral Complexes of First Row Transition Metals Bearing Unbound Thiocyanates and Their Assembly on Metallic Surfaces. J. Am. Chem. Soc. 2006, 128, 3179-3189. 27. Childs, B. J.; Craig, D. C.; Scudder, M. L.; Goodwin, H. A., Structural and electronic studies of the coordination of 6-(thiazol-2-yl)-2,2′-bipyridine and related systems to Fe(II), Co(II) and Ni(II). Inorganica Chim. Acta 1998, 274, 32-41. 28. Baker, A.; Craig, D.; Dong, G.; Rae, A., Metal Complexes of 1,3-Bis(Pyridin-2-yl)pyrazole: Spectral, Magnetic and Structural Studies. Aust. J. Chem. 1995, 48, 1071-1079.

8 James McPherson – June 2019 8-241 Chapter 8: Appendices DFT Calculation output files

A. Hypothetical zinc complexes [ZnIIL]z+ (B3LYP-GD3BJ/6-31G** Table 1):

+ [Zn(B)] : ΔE = –2482.52321816 Eh

C 2.48976 0.21802 -0.00030 N 2.07432 -1.12514 -0.00047 C -2.06897 0.43932 0.00005 C -3.39318 0.86140 0.00013 C -4.41929 -0.07775 0.00018 2+ [Zn(A)] : ΔE = –2521.22908487 Eh C -4.10326 -1.43353 0.00013 C -2.76656 -1.80018 0.00006 C 1.18813 -1.35739 -0.00007 N -1.76937 -0.89909 0.00003 C 1.22577 -2.75262 -0.00000 H -3.61649 1.92159 0.00015 C -0.00001 -3.43319 0.00006 H -5.45456 0.24706 0.00023 C -1.22578 -2.75260 0.00007 H -4.87213 -2.19701 0.00014 N 0.00000 -0.75258 -0.00011 H -2.47526 -2.84520 0.00002 C -1.18814 -1.35738 -0.00003 C 3.20016 -1.88940 -0.00001 C -2.31516 -0.37965 -0.00003 C 4.32974 -1.08133 0.00089 C -3.65122 -0.75412 0.00013 H 3.14050 -2.96917 -0.00010 C -4.64733 0.22733 0.00013 C 3.88128 0.25225 0.00001 C -4.28193 1.57050 -0.00003 H 5.35551 -1.42094 0.00163 C -2.92958 1.89501 -0.00020 H 4.49712 1.14139 0.00009 N -1.96464 0.95354 -0.00020 C -0.91160 1.37177 -0.00002 C 2.31515 -0.37966 -0.00007 C -0.89326 2.75743 0.00016 C 3.65121 -0.75413 -0.00018 C 0.37739 3.37102 0.00007 C 4.64731 0.22733 -0.00009 C 1.55716 2.64090 -0.00013 C 4.28191 1.57050 0.00012 C 1.47055 1.23244 -0.00020 C 2.92955 1.89500 0.00023 N 0.23685 0.70332 -0.00021 N 1.96462 0.95352 0.00015 H -1.79419 3.35829 0.00037 H -2.60447 2.92922 -0.00032 H 0.43115 4.45506 0.00016 H -5.02300 2.36115 -0.00003 H 2.51894 3.13997 -0.00022 H -5.69325 -0.06061 0.00026 Zn 0.19924 -1.22530 -0.00013 H -3.91994 -1.80361 0.00025 H -2.15638 -3.30655 0.00015 H -0.00002 -4.51809 0.00013 H 2.15635 -3.30658 0.00003 H 3.91994 -1.80362 -0.00032 H 5.02298 2.36115 0.00020 H 5.69323 -0.06061 -0.00017 H 2.60444 2.92921 0.00041 Zn 0.00003 1.18027 0.00001

[Zn(C)]: ΔE = –2444.02493861 Eh

C -1.19819 -1.27993 -0.00000 C -1.22915 -2.67990 0.00000 C 0.00000 -3.34741 0.00000 C 1.22915 -2.67990 0.00000 C 1.19819 -1.27993 -0.00000 9 James McPherson – June 2019 8-242 Chapter 8: Appendices N 0.00000 -0.68853 -0.00000 H 1.34191 3.48894 -0.62365 H -2.16431 -3.22718 0.00000 H -1.34191 3.48894 -0.62365 H 0.00000 -4.43408 0.00000 Zn 0.00000 -1.05598 0.23897 H 2.16431 -3.22718 0.00000 N -1.88868 1.06309 0.00000 C -3.03787 1.79213 0.00000 C -4.14209 0.95153 0.00000 C -3.64747 -0.37083 0.00000 C -2.26188 -0.28839 0.00000 H -3.00780 2.87363 0.00000 H -5.17920 1.25782 0.00000 H -4.23005 -1.28246 -0.00000 N 1.88868 1.06309 0.00000 C 2.26188 -0.28839 0.00000 C 3.64747 -0.37083 0.00000 C 4.14209 0.95153 0.00000 [Zn(E)]: ΔE = –2444.26367302 Eh C 3.03787 1.79213 0.00000 H 4.23005 -1.28246 0.00000 N -0.25272 0.76668 -0.68769 H 5.17920 1.25782 0.00000 C -1.35113 1.40484 -0.22872 H 3.00780 2.87363 0.00000 C -0.93806 2.67433 0.26326 Zn 0.00000 1.25720 -0.00000 C 0.46576 2.70374 0.19259 C 0.87894 1.45714 -0.35257 C -2.47307 0.48921 -0.03233 N -2.12295 -0.87590 0.10270 C 2.00414 0.59738 -0.13287 C 3.33112 1.03435 0.08062 C 4.34084 0.12625 0.31058 C 4.05243 -1.25369 0.33011 C 2.74452 -1.64470 0.13916 N 1.73051 -0.77434 -0.06395 H 3.53100 2.09897 0.03701 H 5.35844 0.47211 0.46307 H 4.82418 -1.99459 0.49789 + [Zn(D)] : ΔG = -2482.80597731 Eh H 2.46780 -2.69414 0.16581 H 1.10532 3.49354 0.56519 N 0.00000 0.76578 0.63249 H -1.57375 3.43809 0.68977 C -1.11021 1.44551 0.26315 Zn -0.25223 -1.06966 -0.24325 C -0.70597 2.70290 -0.24119 C -3.28218 -1.57122 0.32537 C 0.70598 2.70290 -0.24119 C -4.35807 -0.69966 0.35660 C 1.11021 1.44551 0.26315 H -3.26673 -2.64271 0.47164 C -2.26760 0.57459 0.09118 C -3.84275 0.60597 0.13640 C -3.58943 1.01168 -0.03965 H -5.39178 -0.96817 0.52654 C -4.60922 0.09315 -0.24526 H -4.40644 1.52807 0.09394 C -4.30842 -1.27184 -0.31614 C -2.98685 -1.66299 -0.19146 N -1.98130 -0.78110 -0.01073 H -3.79613 2.07236 0.04110 H -5.63525 0.43263 -0.33925 H -5.07960 -2.01628 -0.47050 H -2.70503 -2.70830 -0.25474 C 2.26760 0.57459 0.09118 C 3.58943 1.01168 -0.03965 C 4.60922 0.09315 -0.24526 C 4.30842 -1.27184 -0.31614 C 2.98685 -1.66299 -0.19146 N 1.98130 -0.78110 -0.01073 H 3.79613 2.07236 0.04110 H 5.63525 0.43263 -0.33925 H 5.07960 -2.01628 -0.47050 H 2.70503 -2.70830 -0.25474 10 James McPherson – June 2019 8-243 Chapter 8: Appendices HD (B3LYP-GD3BJ/6-31G**) C 1.18889 -1.35780 -0.00005 C 1.22624 -2.75423 0.00004 C 0.00000 -3.43451 0.00008 C -1.22624 -2.75423 0.00005 N 0.00000 -0.75339 -0.00011 C -1.18889 -1.35780 -0.00004 C -2.31834 -0.37813 -0.00003 C -3.65572 -0.75305 -0.00004 C -4.65278 0.22800 -0.00001 C -4.28822 1.57173 0.00002 C -2.93538 1.89632 0.00002

N -1.96876 0.95542 -0.00001 HD: ΔE = –704.459140840 Eh C 2.31834 -0.37813 -0.00005 N 0.00000 0.00074 0.00000 C 3.65572 -0.75305 -0.00005 C -1.12529 0.77744 -0.00000 C 4.65278 0.22800 -0.00001 C -0.70488 2.11019 -0.00000 C 4.28822 1.57173 0.00004 C 0.70488 2.11019 -0.00000 C 2.93538 1.89632 0.00003 C 1.12529 0.77744 -0.00000 N 1.96876 0.95542 -0.00001 C -2.44082 0.15767 -0.00000 H -2.61171 2.93145 0.00005 C -3.62323 0.91767 0.00000 H -5.02962 2.36259 0.00005 C -4.84427 0.25731 0.00000 H -5.69885 -0.06093 -0.00001 C -4.86279 -1.13886 0.00000 H -3.92589 -1.80258 -0.00006 C -3.63947 -1.80828 -0.00000 H -2.15618 -3.31009 0.00011 N -2.45515 -1.19268 -0.00000 H 0.00000 -4.51985 0.00015 H -3.57453 2.00083 0.00001 H 2.15618 -3.31009 0.00010 H -5.77131 0.82266 0.00001 H 3.92589 -1.80258 -0.00009 H -5.79467 -1.69342 0.00000 H 5.02962 2.36259 0.00007 H -3.60906 -2.89619 -0.00000 H 5.69885 -0.06093 -0.00001 C 2.44082 0.15767 -0.00000 H 2.61171 2.93145 0.00007 C 3.62323 0.91767 0.00000 Zn -0.00000 1.17833 0.00001 C 4.84427 0.25731 0.00000 C 4.86279 -1.13886 0.00000 C 3.63947 -1.80828 0.00000 N 2.45515 -1.19268 -0.00000 H 3.57453 2.00083 0.00000 H 5.77131 0.82266 0.00001 H 5.79467 -1.69342 0.00000 H 3.60906 -2.89619 -0.00000 H 0.00000 -1.00994 0.00000 H 1.34780 2.97839 -0.00000 H -1.34780 2.97839 -0.00000

+ [Zn(B)] : ΔE = –2482.45250784 Eh

C 2.49177 0.21430 -0.00028 II z+ N 2.07786 -1.12779 -0.00046 [Zn L] (B3LYP/6-31G** Table S1): C -2.07210 0.43813 0.00006 C -3.39751 0.86079 0.00013 C -4.42471 -0.07775 0.00015 C -4.10965 -1.43397 0.00010 C -2.77246 -1.80072 0.00005 N -1.77350 -0.90060 0.00003 H -3.62216 1.92111 0.00015 H -5.46003 0.24821 0.00020 H -4.87885 -2.19765 0.00010 H -2.48284 -2.84667 0.00002 C 3.20512 -1.88899 -0.00004 C 4.33379 -1.08061 0.00088 2+ [Zn(A)] : ΔE = -2521.15367696 Eh H 3.14865 -2.96937 -0.00013 11 James McPherson – June 2019 8-244 Chapter 8: Appendices C 3.88382 0.25119 -0.00002 H 5.36000 -1.41986 0.00161 H 4.49983 1.14064 0.00004 C -0.91215 1.37220 -0.00000 C -0.89255 2.75889 0.00015 C 0.37905 3.37132 0.00006 C 1.55878 2.64076 -0.00014 C 1.47147 1.23099 -0.00018 N 0.23669 0.70349 -0.00017 H -1.79228 3.36227 0.00034

H 0.43356 4.45575 0.00013 + H 2.52022 3.14134 -0.00024 [Zn(D)] : ΔE = –2482.73721109 Eh Zn 0.20102 -1.22336 -0.00012 N 0.00000 0.75952 0.60872 C -1.10864 1.44223 0.24828 C -0.70543 2.70686 -0.23814 C 0.70543 2.70686 -0.23813 C 1.10864 1.44223 0.24828 C -2.27226 0.57333 0.08710 C -3.59594 1.01371 -0.02734 C -4.62105 0.09760 -0.21990 C -4.32470 -1.26844 -0.29441 C -3.00199 -1.66251 -0.18516 N -1.99080 -0.78351 -0.01647 H -3.80104 2.07492 0.05582 [Zn(C)]: ΔE = –2443.96053656 Eh H -5.64759 0.44015 -0.30101 H -5.09952 -2.01158 -0.43920 C -1.19907 -1.27930 -0.00000 H -2.72495 -2.70942 -0.25105 C -1.22971 -2.68053 0.00000 C 2.27226 0.57333 0.08710 C 0.00000 -3.34765 0.00000 C 3.59594 1.01371 -0.02733 C 1.22971 -2.68053 0.00000 C 4.62105 0.09759 -0.21990 C 1.19907 -1.27930 -0.00000 C 4.32470 -1.26844 -0.29441 N 0.00000 -0.68879 -0.00000 C 3.00199 -1.66251 -0.18516 H -2.16401 -3.23007 0.00000 N 1.99080 -0.78351 -0.01647 H 0.00000 -4.43476 0.00000 H 3.80104 2.07492 0.05582 H 2.16401 -3.23007 0.00000 H 5.64759 0.44015 -0.30101 N -1.89137 1.06518 0.00000 H 5.09951 -2.01158 -0.43920 C -3.04168 1.79143 0.00000 H 2.72495 -2.70942 -0.25105 C -4.14532 0.95105 0.00000 H 1.34034 3.50026 -0.60798 C -3.64985 -0.36964 0.00000 H -1.34034 3.50026 -0.60798 C -2.26385 -0.28508 0.00000 Zn 0.00000 -1.05898 0.22037 H -3.01453 2.87345 0.00000 H -5.18270 1.25746 0.00000 H -4.23293 -1.28137 -0.00000 N 1.89137 1.06518 0.00000 C 2.26385 -0.28508 0.00000 C 3.64985 -0.36964 0.00000 C 4.14532 0.95105 0.00000 C 3.04168 1.79143 0.00000 H 4.23293 -1.28137 -0.00000 H 5.18270 1.25746 0.00000 H 3.01453 2.87345 0.00000 Zn 0.00000 1.25520 -0.00000 [Zn(E)]: ΔE = –2444.17220333 Eh

N -0.25068 0.76152 -0.66846 C -1.34994 1.40148 -0.22043 C -0.94111 2.67728 0.25826 C 0.46218 2.70749 0.19230 C 0.87731 1.45457 -0.33589 12 James McPherson – June 2019 8-245 Chapter 8: Appendices C -2.47663 0.48573 -0.03268 C -2.31511 0.08946 -0.00009 N -2.13142 -0.87887 0.10457 C -3.65121 0.46386 -0.00019 C 2.00869 0.59670 -0.12602 C -4.64731 -0.51755 -0.00009 C 3.33900 1.03559 0.07063 C -4.28190 -1.86075 0.00011 C 4.35347 0.12876 0.28786 C -2.92960 -2.18524 0.00021 C 4.06733 -1.25177 0.31163 N -1.96460 -1.24374 0.00011 C 2.75691 -1.64433 0.13637 H 2.60450 -3.21943 -0.00029 N 1.73791 -0.77598 -0.05471 H 5.02300 -2.65132 0.00001 H 3.53906 2.10044 0.02368 H 5.69320 -0.22962 0.00031 H 5.37277 0.47657 0.42681 H 3.91989 1.51337 0.00031 H 4.84208 -1.99219 0.46992 H 2.15639 3.01637 0.00011 H 2.48357 -2.69507 0.16604 H -0.00002 4.22786 0.00011 H 1.09846 3.50459 0.55591 H -2.15641 3.01636 0.00001 H -1.57853 3.44715 0.67199 H -3.91991 1.51335 -0.00029 Zn -0.25492 -1.07061 -0.22842 H -5.02300 -2.65135 0.00021 C -3.29579 -1.56959 0.31323 H -5.69320 -0.22965 -0.00019 C -4.36947 -0.69569 0.33292 H -2.60440 -3.21944 0.00041 H -3.28714 -2.64174 0.45870 C -3.84825 0.60760 0.11985 H -5.40597 -0.96232 0.49060 H -4.40953 1.53130 0.07107

B. Calculations of ΔEL(strain) (B3LYP/6- - B : ΔEL(coord) = –703.7611521 Eh 31G**) C 2.53982 -0.14257 -0.00034 Strained ligand geometries from N 2.10274 -1.47881 -0.00054 theoretical [ZnIIL]z+ (Table S3 1-5) C -2.01482 0.15213 0.00006 C -3.33205 0.59551 0.00006 C -4.37315 -0.32704 0.00016 C -4.07903 -1.68765 0.00006 C -2.74841 -2.07584 0.00006 N -1.73682 -1.19092 -0.00004 H -3.53824 1.65917 0.00006 H -5.40308 0.01450 0.00016 H -4.86003 -2.43867 0.00006 H -2.47398 -3.12539 -0.00004 C 3.21618 -2.26114 -0.00004 C 4.35855 -1.47134 0.00086 H 3.13910 -3.33984 -0.00014 E A: ΔEL(coord) = –742.466359709 h C 3.93169 -0.13069 -0.00004 H 5.37875 -1.82742 0.00156 C -1.18811 1.06716 -0.00009 H 4.56174 0.74837 0.00006 C -1.22581 2.46236 0.00001 C -0.84255 1.06587 -0.00004 C -0.00001 3.14296 0.00011 C -0.80193 2.45099 0.00016 C 1.22579 2.46237 0.00011 C 0.47849 3.04405 0.00006 N -0.00001 0.46236 -0.00009 C 1.64637 2.29504 -0.00014 C 1.18809 1.06717 0.00001 C 1.53700 0.88811 -0.00024 C 2.31519 0.08937 0.00001 N 0.29493 0.37896 -0.00024 C 3.65119 0.46387 0.00011 H -1.69304 3.06633 0.00036 C 4.64729 -0.51753 0.00011 H 0.54975 4.12714 0.00016 C 4.28190 -1.86073 0.00001 H 2.61599 2.77858 -0.00024 C 2.92960 -2.18523 -0.00019 N 1.96460 -1.24373 -0.00019 13 James McPherson – June 2019 8-246 Chapter 8: Appendices C 1.55633 -0.19053 4.30840 C 1.94213 -0.05010 2.98690 N 1.05360 0.09463 1.98130 H -1.79965 0.03017 3.79610 H -0.14571 -0.28313 5.63520 H 2.30650 -0.31448 5.07960 H 2.98914 -0.07067 2.70500 C -0.30513 0.14123 -2.26760 C -0.73654 -0.00736 -3.58940 C 0.18968 -0.17538 -4.60920

2- C 1.55633 -0.19053 -4.30840 C : ΔEL(coord) = –664.9434942 Eh C 1.94213 -0.05010 -2.98690 N 1.05360 0.09463 -1.98130 C 0.00000 1.19820 0.93068 H -1.79965 0.03017 -3.79610 C 0.00000 1.22920 2.33068 H -0.14571 -0.28313 -5.63520 C 0.00000 0.00000 2.99818 H 2.30650 -0.31448 -5.07960 C 0.00000 -1.22920 2.33068 H 2.98914 -0.07067 -2.70500 C 0.00000 -1.19820 0.93068 H -3.18789 -0.69177 -1.34190 N 0.00000 0.00000 0.33928 H -3.18789 -0.69177 1.34190 H 0.00000 2.16430 2.87798 H 0.00000 0.00000 4.08488 H 0.00000 -2.16430 2.87798 N 0.00000 1.88870 -1.41232 C 0.00000 3.03790 -2.14132 C 0.00000 4.14210 -1.30072 C 0.00000 3.64750 0.02158 C 0.00000 2.26190 -0.06082 H 0.00000 3.00780 -3.22282 H 0.00000 5.17920 -1.60702 H 0.00000 4.23000 0.93328 N 0.00000 -1.88870 -1.41232 2- C 0.00000 -2.26190 -0.06082 E : ΔEL(coord) = –664.906998324 Eh C 0.00000 -3.64750 0.02158 C 0.00000 -4.14210 -1.30072 N 0.32835 0.49941 0.73341 C 0.00000 -3.03790 -2.14132 C 1.43810 1.09749 0.24855 H 0.00000 -4.23000 0.93328 C 1.04694 2.34885 -0.30404 H 0.00000 -5.17920 -1.60702 C -0.35648 2.40404 -0.23948 H 0.00000 -3.00780 -3.22282 C -0.79107 1.19130 0.36241 C 2.54566 0.15579 0.09921 N 2.17379 -1.20836 0.02745 C -1.92945 0.34034 0.17967 C -3.24866 0.78790 -0.05859 C -4.27231 -0.11363 -0.24896 C -4.00620 -1.49746 -0.20252 C -2.70529 -1.89994 0.01094 N -1.67788 -1.03730 0.17622 H -3.43144 1.85654 -0.06582 H -5.28379 0.24086 -0.42096 H -4.78944 -2.23299 -0.33783

- H -2.44552 -2.95373 0.03469 D : ΔEL(coord) = –703.7600816 Eh H -0.98217 3.18555 -0.65098 H 1.69590 3.08143 -0.76390 N -0.51822 0.67429 0.00000 C 3.32227 -1.93187 -0.15841 C -1.18231 0.27762 1.11020 C 4.41219 -1.08016 -0.22701 C -2.41812 -0.27758 0.70600 H 3.28983 -3.00868 -0.25406 C -2.41812 -0.27758 -0.70600 C 3.91750 0.24255 -0.07023 C -1.18231 0.27762 -1.11020 H 5.44185 -1.37290 -0.38063 C -0.30513 0.14123 2.26760 H 4.49582 1.15648 -0.06931 C -0.73654 -0.00736 3.58940 C 0.18968 -0.17538 4.60920 14 James McPherson – June 2019 8-247 Chapter 8: Appendices Strained Py-Pyr- ligand geometries from SC-XRD structures (Table S3 6-25; Table S5 1 & 2).

DANMOU: ΔEL(coord) = –742.46112876 Eh

N -1.96184 -1.20330 -0.09174 C -2.30920 0.09597 -0.01279 C -3.63760 0.48665 0.11073 C -4.60906 -0.50098 0.10329 DANMOU: ΔEL(coord) = –742.46535110 Eh C -4.26136 -1.80351 -0.01264 C -2.92956 -2.14438 -0.10275 N 1.95217 -1.21702 -0.06774 N 0.00668 0.35859 -0.02428 C 2.30794 0.08667 -0.00708 C -1.19173 1.02774 -0.01017 C 3.62973 0.48461 0.05514 C -1.18525 2.40174 -0.01214 C 4.63338 -0.47676 0.08026 C -0.01017 3.07906 -0.02352 C 4.26800 -1.82316 -0.01622 C 1.18383 2.40113 -0.01038 C 2.92857 -2.14658 -0.06223 C 1.17874 1.01791 -0.02411 N 0.00568 0.35241 -0.01331 N 1.95565 -1.21287 -0.01244 C 1.19000 1.00391 0.00092 C 2.30629 0.08438 -0.02033 C 1.22028 2.40775 0.00474 C 3.64577 0.48924 -0.02858 C 0.01340 3.07827 -0.01429 C 4.63647 -0.47494 0.03460 C -1.18750 2.39234 -0.00377 C 4.26234 -1.81324 0.06475 C -1.17355 1.02828 -0.00586 C 2.92055 -2.14852 0.04357 N -1.97238 -1.20712 -0.02364 H -3.91402 1.53213 0.20409 C -2.31195 0.08100 -0.03919 H -5.65575 -0.22049 0.19175 C -3.63620 0.48771 -0.02181 H -5.02303 -2.57959 -0.02391 C -4.64597 -0.47611 0.01025 H -2.59646 -3.17666 -0.17970 C -4.28557 -1.79726 0.06434 H -2.11999 2.95915 -0.01697 C -2.93366 -2.12464 0.03641 H -0.01012 4.16528 -0.02784 H 3.89034 1.53725 0.11419 H 2.11606 2.96071 -0.00580 H 5.67779 -0.18695 0.14130 H 3.90654 1.54300 -0.05645 H 5.02525 -2.60224 -0.01706 H 5.68509 -0.19320 0.04586 H 2.59774 -3.18201 -0.11443 H 5.01888 -2.59226 0.11369 H 2.15342 2.96269 0.00274 H 2.58905 -3.18444 0.05730 H 0.00568 4.16475 -0.02280 H -2.11874 2.95291 -0.00842 H -3.89596 1.54195 -0.04138 H -5.68977 -0.17627 0.01691 H -5.03542 -2.58220 0.11329 H -2.60993 -3.16395 0.05893

BENHUZ: ΔEL(coord) = –742.47067189 Eh

N 2.03023 -1.27151 -0.01719 N -2.03475 -1.26029 -0.07684 C -3.04642 -2.14257 -0.07808 H -2.76427 -3.19178 -0.14037 15 James McPherson – June 2019 8-248 Chapter 8: Appendices C 3.04781 -2.15426 0.06007 H -5.72227 -0.09331 0.06007 H 2.76616 -3.20448 0.09281 C 0.01221 3.05729 -0.03096 C -1.17902 0.98124 -0.00956 H 0.01147 4.14393 -0.03414 C -2.34513 0.08571 -0.01266 C 1.21580 2.37847 -0.03589 C 2.33631 0.07014 -0.02921 H 2.15099 2.92929 -0.04439 C -3.66338 0.51715 0.05317 N -2.03041 -1.26063 -0.02423 H -3.89368 1.57737 0.10521 N 2.03373 -1.25772 -0.13275 C 4.37652 -1.75683 0.06788 C 2.33549 0.07408 -0.01223 H 5.16944 -2.49801 0.11847 C 3.65135 0.51413 -0.03489 H 3.87863 1.57523 -0.06639 C 0.01325 3.05676 0.00761 H 0.01776 4.14308 0.01953 C -1.19636 2.37962 0.00420 H -2.12734 2.93824 0.01139 C -4.37358 -1.75844 0.00230 H -5.15792 -2.51076 0.01058 C 4.67693 -0.40759 -0.00095 H 5.71137 -0.07606 -0.01057 C -4.69279 -0.41202 0.08223 H -5.72676 -0.08808 0.15004 LOGKUP A: ΔEL(coord) = –742.45838104 Eh N 0.00247 0.31933 -0.01766 C 1.18014 0.97542 -0.02179 N -2.00944 -1.24530 -0.01708 C 1.21303 2.36887 -0.00749 N -0.00520 0.32180 0.02773 H 2.14907 2.91858 -0.00589 N 2.03354 -1.24373 0.02093 C -3.02029 -2.15954 -0.08483 H -2.71675 -3.20112 -0.14580 C -4.34060 -1.77018 -0.06896 H -5.12773 -2.51767 -0.13153 C -4.66200 -0.44401 0.03018 H -5.70751 -0.14573 0.05475 C -3.67505 0.52692 0.09859 H -3.92001 1.58001 0.17142 C -2.32423 0.06850 0.04794 C -1.22511 0.98290 0.02712 C -1.21229 2.35665 0.02521 H -2.14427 2.91850 0.04829 BENHUZ: ΔEL(coord) = –742.47116739 Eh C -0.04602 3.04532 -0.03871 H -0.06227 4.13128 -0.05982 C 1.18078 0.98264 -0.03031 C 1.21549 2.37769 -0.07489 C 3.04449 -2.14472 -0.13992 H 2.13476 2.94983 -0.13476 H 2.76267 -3.18958 -0.25266 C 1.18029 0.99106 -0.02617 C 3.64452 0.50494 0.14832 C 2.34924 0.10749 -0.01862 H 3.87007 1.55998 0.27083 C 3.65967 0.51238 -0.00457 C -3.66481 0.51361 0.00320 H 3.90826 1.57075 -0.00784 H -3.89790 1.57423 -0.00924 C 4.68514 -0.42201 -0.00217 C -1.19824 2.37943 -0.02942 H 5.72099 -0.09695 -0.01196 H -2.12843 2.93936 -0.02358 C 4.36318 -1.75578 0.01357 N 0.00064 0.32518 -0.03900 H 5.14538 -2.51111 0.01865 C -1.17915 0.98622 -0.02924 C 3.03503 -2.13776 0.05636 C 4.37044 -1.76129 -0.00002 H 2.74211 -3.18494 0.09725 H 5.16108 -2.50645 -0.00514 C -3.03736 -2.14797 0.04834 H -2.74913 -3.19714 0.06196 C 4.66415 -0.42145 0.15783 H 5.69358 -0.09776 0.28429 C -2.34052 0.07900 -0.02148 C -4.36715 -1.76357 0.08260 H -5.15142 -2.51411 0.13273 C -4.68539 -0.41606 0.04437 16 James McPherson – June 2019 8-249 Chapter 8: Appendices C -4.36095 -1.75346 0.06271 C -4.65118 -0.40845 -0.00005 C -3.64669 0.49874 -0.01666 C -2.30145 0.05281 -0.02880 C -1.21072 0.98114 0.00204 C -1.22282 2.37285 0.02691 C 0.00052 3.05938 -0.01344 C 1.20371 2.37018 -0.04201 C 1.18460 1.02280 0.00299 C 2.32599 0.09254 0.00405 C 3.64290 0.51549 0.04203 GOGDOV: ΔEL(coord) = –742.46285148 Eh C 4.66225 -0.43109 0.07041 C 4.34626 -1.77337 0.00530 N -2.04137 -1.26546 -0.09371 C 3.02952 -2.12226 -0.06237 N 0.00009 0.32106 -0.05275 H -2.74482 -3.19337 0.02030 N 2.00415 -1.26474 -0.00157 H -5.14761 -2.50056 0.10647 C -3.03602 -2.13593 -0.09117 H -5.68666 -0.07641 0.00281 C -4.36812 -1.76847 0.00877 H -3.86725 1.56140 -0.03909 C -4.65231 -0.40861 0.10563 H -2.14904 2.93897 0.05107 C -3.64192 0.53774 0.07133 H -0.00201 4.14504 -0.02357 C -2.32410 0.06754 -0.02840 H 2.13446 2.92911 -0.08207 C -1.18720 1.00215 -0.00003 H 3.88992 1.57218 0.08330 C -1.16784 2.38856 -0.00640 H 5.69833 -0.10850 0.12432 C 0.02331 3.08627 0.01164 H 5.12389 -2.53161 0.01055 C 1.21583 2.34847 -0.03321 H 2.73753 -3.17073 -0.12101 C 1.19271 0.98188 -0.02659 C 2.30187 0.07046 -0.01449 C 3.65559 0.51452 -0.01258 C 4.67052 -0.44950 0.01727 C 4.35245 -1.76148 0.06161 C 2.99517 -2.15270 0.05115 H -2.75185 -3.18572 -0.16145 H -5.16015 -2.51022 0.02481 H -5.68662 -0.08484 0.19648 H -3.86748 1.59611 0.13930 H -2.10025 2.94948 0.01204 H 0.03964 4.17012 0.01964 H 2.15848 2.89033 -0.03219 YAHBIV: ΔEL(coord) = –1205.75313190 Eh H 3.90111 1.57044 -0.03533 H 5.70974 -0.13085 0.01481 N 2.07080 2.21581 -0.01131 H 5.12669 -2.52358 0.09807 N 0.00912 0.70215 -0.01742 H 2.71086 -3.20267 0.07287 N -2.07962 2.22103 -0.11218 C 3.31302 2.81597 0.01525 C 4.32789 1.89606 0.05878 H 5.40084 2.08118 0.04276 C 3.70690 0.61562 0.00734 C 2.34010 0.84647 0.00528 C 1.17523 0.00052 0.05991 C 1.19653 -1.37108 0.26189 H 2.13256 -1.90152 0.38588 C 0.00169 -2.04840 0.34953 H -0.00407 -3.12444 0.52958 C -1.19328 -1.36369 0.22909 H -2.13277 -1.89327 0.32463 GOGDOV: ΔEL(coord) = –742.45873930 Eh C -1.16023 0.00348 0.03103 C -2.32607 0.85669 -0.06918 N -2.01453 -1.27380 -0.03001 C -3.69741 0.61024 -0.10116 N 0.01246 0.30230 0.00647 C -4.32657 1.86585 -0.10589 N 2.01994 -1.25562 -0.05298 H -5.40120 2.03721 -0.15174 C -3.02058 -2.13991 0.01399 C -3.32446 2.80540 -0.14162 17 James McPherson – June 2019 8-250 Chapter 8: Appendices C 4.43543 -0.66516 -0.00944 C 1.14939 0.02646 -0.33168 C 5.45110 -0.94320 0.90973 C 2.31485 0.85948 -0.12271 H 5.65940 -0.20507 1.68008 C 3.66887 0.57794 -0.00728 C 6.17533 -2.11759 0.85462 C 4.30904 1.84078 0.06642 H 6.95997 -2.30341 1.58828 H 5.37519 2.00544 0.21593 C 5.89744 -3.06478 -0.11535 C 3.31798 2.79479 0.04263 H 6.46225 -3.99512 -0.15637 C -4.41452 -0.67872 0.04524 C 4.89807 -2.81071 -1.02005 C -5.61759 -0.85230 -0.62591 H 4.66828 -3.54137 -1.79547 H -5.98688 -0.03539 -1.24105 C 4.18538 -1.62930 -0.97350 C -6.33327 -2.02979 -0.53246 H 3.40915 -1.43522 -1.70748 H -7.27141 -2.13050 -1.07925 C -4.41466 -0.68117 -0.04369 C -5.86991 -3.06828 0.23090 C -5.36099 -0.88524 0.95555 H -6.43239 -3.99878 0.29990 H -5.50227 -0.09809 1.69097 C -4.68713 -2.91645 0.91222 C -6.10919 -2.05022 1.03030 H -4.31006 -3.72495 1.53799 H -6.83967 -2.17280 1.83014 C -3.97545 -1.73174 0.83367 C -5.91937 -3.06418 0.10722 H -3.05637 -1.61195 1.39758 H -6.49545 -3.98622 0.16243 C 4.37048 -0.72315 0.08635 C -4.96381 -2.88249 -0.90153 C 3.90500 -1.79995 0.84566 H -4.79989 -3.66453 -1.64152 H 2.94941 -1.70872 1.35125 C -4.23639 -1.70719 -0.96869 C 4.64180 -2.97330 0.96344 H -3.50502 -1.56633 -1.75956 H 4.24140 -3.79176 1.56115 C 3.46417 4.29240 -0.03623 C 5.86758 -3.09777 0.35621 H 3.34719 4.70425 -1.05296 H 6.44804 -4.01424 0.45171 H 4.45344 4.60968 0.32323 C 6.35260 -2.02885 -0.38173 H 2.70354 4.78917 0.57983 H 7.32313 -2.10672 -0.87290 C -3.51004 4.28077 -0.23837 C 5.61796 -0.87801 -0.51008 H -2.78391 4.81379 0.38921 H 6.00272 -0.05272 -1.10408 H -4.51872 4.57869 0.08086 C -3.37245 4.28392 0.40587 H -3.37069 4.66987 -1.26137 H -3.41978 4.58077 1.46801 H -4.25453 4.72954 -0.07940 H -2.47547 4.75342 -0.01144 C 3.52393 4.27175 0.14067 H 3.10434 4.80797 -0.72376 H 4.59309 4.51630 0.19669 H 3.03737 4.70473 1.02777

YAHBIV: ΔEL(coord) = –1205.75662947 Eh

N -2.05119 2.21845 0.06911 N -0.00729 0.70452 -0.11802 N 2.06940 2.22634 -0.07798 C -3.29460 2.81006 0.20934 C -4.30155 1.88314 0.17253 ALILUC: ΔEL(coord) = –972.30176586 Eh H -5.36785 2.05832 0.31279 C -3.69118 0.60789 0.04664 N -2.16266 -0.53376 -0.04644 C -2.33070 0.86047 -0.04818 N 2.16560 -0.55340 -0.01437 C -1.17437 0.02728 -0.30077 N 0.02453 1.01730 -0.01686 C -1.21370 -1.27595 -0.76306 C -3.40593 -1.06233 -0.02425 H -2.16011 -1.76597 -0.96071 C -3.79721 -2.38447 -0.02336 C -0.02457 -1.93179 -1.01223 H -3.02877 -3.15771 -0.04808 H -0.03236 -2.95063 -1.40149 C -5.12624 -2.72799 0.02156 C 1.17781 -1.28098 -0.79501 H -5.40969 -3.78152 0.03193 H 2.11839 -1.77099 -1.01876 C -6.10837 -1.74790 0.05901 18 James McPherson – June 2019 8-251 Chapter 8: Appendices H -7.16203 -2.02828 0.09395 H 2.14449 3.61438 -0.02398 C -5.75449 -0.43443 0.02933 C 0.01986 3.73850 -0.03083 H -6.54007 0.32797 0.04975 H 0.02381 4.83137 -0.01948 C -4.41639 -0.04063 -0.00632 C -1.18854 3.08044 -0.05973 C -3.67933 1.13419 -0.01900 H -2.12974 3.62453 -0.06107 H -4.10705 2.13718 -0.00097 C -1.16380 1.67970 -0.06594 C -2.34321 0.83517 -0.02976 C -2.35852 0.85672 -0.06098 C -1.16915 1.65982 -0.02398 C -3.68844 1.15970 0.12632 C -1.19787 3.04732 -0.01974 H -4.12725 2.14320 0.28342 H -2.14750 3.57967 -0.02669 C -4.39834 -0.05507 0.10600 C -0.00724 3.74270 0.02603 C -5.73398 -0.44229 0.23935 H -0.01167 4.83304 0.04454 H -6.51683 0.30643 0.39030 C 1.19080 3.06630 0.02192 C -6.07898 -1.76499 0.15677 H 2.12883 3.62001 0.05049 H -7.12593 -2.05527 0.26331 C 1.20346 1.65305 0.00319 C -5.10519 -2.76499 -0.01760 C 2.34845 0.79882 0.00250 H -5.41069 -3.81075 -0.06580 C 3.68141 1.13457 0.00257 C -3.79161 -2.41892 -0.16739 H 4.11012 2.13406 0.01403 H -3.01513 -3.16726 -0.31975 C 4.40114 -0.06882 0.00768 C -3.42754 -1.07419 -0.09422 C 5.73087 -0.40366 0.02354 H 6.49781 0.37639 0.04197 C 6.10346 -1.71299 0.02131 H 7.15868 -1.98925 0.02965 C 5.12020 -2.73339 -0.01177 H 5.44206 -3.77603 -0.01502 C 3.78990 -2.42731 -0.00457 H 3.02410 -3.20145 -0.01631 C 3.41121 -1.09220 -0.00681

XOVFIX: ΔEL(coord) = –972.29139833 Eh

N -1.97710 -0.41558 -0.10590 N -0.00491 1.22577 0.01159 N 1.99014 -0.41186 -0.12285 C -3.17109 -1.08143 -0.04806 C -3.43737 -2.46157 -0.08720 H -2.61384 -3.16798 -0.17078 ALILUC: ΔEL(coord) = –972.30306890 Eh C -4.74094 -2.87349 -0.01835 H -4.96769 -3.94140 -0.04534 N 2.15965 -0.52560 -0.12299 C -5.80945 -1.96185 0.09024 N -2.18714 -0.49950 -0.20402 H -6.83038 -2.34377 0.14148 N -0.00666 1.01784 -0.07778 C -5.57122 -0.62472 0.12594 C 3.40775 -1.08036 -0.04978 H -6.40118 0.08192 0.20813 C 3.78032 -2.43768 -0.09736 C -4.24494 -0.15618 0.05908 H 3.02305 -3.20950 -0.21101 C -3.65570 1.12528 0.07221 C 5.11869 -2.72776 -0.00644 H -4.18836 2.07028 0.15590 H 5.42784 -3.77620 -0.03760 C -2.30847 0.93631 -0.02908 C 6.08744 -1.76544 0.16642 C 3.18267 -1.07061 -0.06594 H 7.13433 -2.05853 0.25581 C 3.42926 -2.46041 -0.08954 C 5.74210 -0.43883 0.18631 H 2.59163 -3.15031 -0.16758 H 6.51568 0.32446 0.30475 C 4.72229 -2.88860 -0.00023 C 4.39015 -0.06476 0.06306 H 4.94267 -3.95786 -0.00755 C 3.67374 1.14484 0.06346 C 5.78398 -1.98044 0.10650 H 4.10574 2.13883 0.15408 H 6.80355 -2.36485 0.17712 C 2.36334 0.82055 -0.04476 C 5.56377 -0.63050 0.11785 C 1.18461 1.66944 -0.05153 H 6.40323 0.06362 0.20279 C 1.19856 3.07624 -0.04103 C 4.24879 -0.15785 0.03132 19 James McPherson – June 2019 8-252 Chapter 8: Appendices C 3.67385 1.12808 0.05384 H 4.21319 2.06805 0.14646 C 2.32066 0.94099 -0.02011 C -1.19666 1.85725 -0.02834 C -1.20483 3.24692 -0.04218 H -2.13818 3.80877 -0.07265 C 0.00610 3.90679 -0.03509 H 0.01236 4.99843 -0.04375 C 1.21092 3.23219 -0.00909 H 2.14929 3.78673 -0.01298 C 1.19283 1.85051 -0.01564 LOGKOJ: ΔEL(coord) = –970.97396599 Eh

O 3.25187 -1.40912 -0.03033 O 1.97999 -3.22801 0.05623 N 0.86039 2.64704 -0.01630 N -0.48460 0.57400 -0.03122 N -2.89769 1.20597 0.01760 C 1.46846 3.84263 -0.01036 H 0.80806 4.71189 -0.01807 C 2.83572 3.98334 0.01602 H 3.29270 4.97209 0.02960 C 3.61374 2.84401 0.02620 TELPUX: ΔEL(coord) = –703.75923319 Eh H 4.69961 2.91985 0.04788 C 3.01386 1.59862 0.00016 N 2.02588 -1.05196 0.00964 H 3.60190 0.68934 -0.00143 N 0.00000 0.39150 0.00000 C 1.62931 1.50004 -0.01923 C 3.04609 -1.93826 0.01707 C 0.81815 0.26463 -0.03437 C 4.37498 -1.52546 0.00261 C 0.89206 -1.14549 -0.02620 C 4.65794 -0.18319 -0.01049 C -0.44525 -1.64872 -0.03075 C 3.62641 0.73550 -0.01813 C -1.27308 -0.52231 -0.01981 C 2.31577 0.29437 0.01383 C -2.68777 -0.15386 0.00323 C 1.11791 1.13211 0.01066 C -3.77423 -1.03085 0.01272 C 0.69949 2.47034 0.01429 H -3.60740 -2.10358 0.02117 H 2.77015 -2.99304 0.02105 C -5.05615 -0.52588 0.01128 H 5.17881 -2.26109 0.00719 H -5.91079 -1.19957 0.00877 H 5.69175 0.15893 -0.02661 C -5.24575 0.83854 0.02743 H 3.83578 1.80310 -0.03207 H -6.25591 1.24795 0.03402 H 1.31066 3.37289 0.01951 C -4.16262 1.67739 0.03682 N -2.02588 -1.05196 -0.00964 H -4.27030 2.76202 0.05578 C -3.04610 -1.93826 -0.01707 C 2.15805 -1.90768 -0.01106 C -4.37499 -1.52546 -0.00261 C 3.18557 -4.02145 0.06678 C -4.65794 -0.18318 0.01048 H 3.77172 -3.85575 -0.84202 C -3.62641 0.73550 0.01813 H 2.85381 -5.05975 0.12338 C -2.31577 0.29438 -0.01383 H 3.81006 -3.77070 0.92904 C -1.11791 1.13211 -0.01066 C -0.93308 -3.06766 -0.07192 C -0.69949 2.47034 -0.01429 H -0.20093 -3.74384 -0.51753 H -2.77016 -2.99303 -0.02105 H -1.85604 -3.14388 -0.65840 H -5.17882 -2.26109 -0.00719 H -1.15992 -3.46998 0.92848 H -5.69175 0.15894 0.02661 H -3.83578 1.80310 0.03207 H -1.31066 3.37289 -0.01951

20 James McPherson – June 2019 8-253 Chapter 8: Appendices C -4.72295 -0.06721 -0.30520 H -5.69838 0.36623 -0.51845 C -3.60983 0.75264 -0.23902 H -3.73447 1.82193 -0.40951 C -2.34025 0.26602 0.01726 C -1.10317 1.03336 0.09893 C -0.68554 2.37608 0.17977 H -1.29964 3.27122 0.25694 C 0.70874 2.35891 0.17904 H 1.33739 3.24485 0.25320 C 1.10602 1.03128 0.09521 TELQEI: ΔEL(coord) = –703.75892247 Eh C 2.33682 0.25288 0.01449 C 3.60826 0.75566 -0.18055 N -0.00858 2.01370 -1.05785 H 3.73411 1.83214 -0.29408 N 0.00000 0.00000 0.41424 C 4.73277 -0.05868 -0.25129 C -0.02614 3.02688 -1.94056 H 5.71244 0.38735 -0.41195 H -0.04841 2.74210 -2.99387 C 4.58040 -1.42826 -0.12152 C 0.00000 4.35744 -1.54899 H 5.42257 -2.11400 -0.17025 H -0.01081 5.15886 -2.28578 C 3.28987 -1.88285 0.08268 C 0.03542 4.63594 -0.19936 H 3.10078 -2.95228 0.20429 H 0.06039 5.67242 0.13655 C 0.04464 3.62561 0.72507 H 0.07126 3.84969 1.78974 C 0.00835 2.29663 0.30255 C 0.00473 1.12076 1.14894 C -0.00533 0.70013 2.49723 H -0.00164 1.31127 3.39924 N 0.00858 -2.01370 -1.05785 C 0.02614 -3.02688 -1.94056 H 0.04841 -2.74210 -2.99387 C 0.00000 -4.35744 -1.54899 H 0.01081 -5.15886 -2.28578 AYOSUE: ΔEL(coord) = –703.77144698 Eh C -0.03542 -4.63594 -0.19936 H -0.06039 -5.67242 0.13655 N -0.00578 0.28008 0.00637 C -0.04464 -3.62561 0.72507 N 2.08736 -1.10549 -0.03833 H -0.07126 -3.84969 1.78974 N -2.08394 -1.11027 0.05946 C -0.00835 -2.29663 0.30255 C -3.17562 -1.88527 0.05382 C -0.00473 -1.12076 1.14894 H -2.97588 -2.95957 0.09883 C 0.00533 -0.70013 2.49723 C -4.47877 -1.44138 0.00256 H 0.00164 -1.31127 3.39924 H -5.30718 -2.14607 0.00401 C -4.69003 -0.07757 -0.04594 H -5.70223 0.32347 -0.08408 C -3.61305 0.77177 -0.04544 H -3.75359 1.84959 -0.08479 C -2.31569 0.25832 0.00027 C -1.11369 1.04601 0.01146 C -0.69660 2.39581 0.00713 H -1.31619 3.29094 0.01022 C 0.69689 2.39301 0.00705 H 1.32103 3.28476 0.00734 C 1.10258 1.04074 0.00881 KANXIJ: ΔEL(coord) = –703.77453085 Eh C 2.31290 0.25396 -0.00793 C 3.60262 0.77304 0.02372 N -0.00878 0.27257 0.04975 H 3.73787 1.85202 0.05287 N -2.21241 -1.08824 0.23994 C 4.69165 -0.06735 0.01798 N 2.19945 -1.10471 0.16052 H 5.70010 0.34309 0.04251 C -3.30842 -1.86511 0.16077 C 4.48902 -1.43841 -0.01615 H -3.12627 -2.92714 0.34333 H 5.32231 -2.13722 -0.02092 C -4.57282 -1.42144 -0.10634 C 3.18858 -1.88770 -0.04237 H -5.41569 -2.10743 -0.15040 H 2.98555 -2.96105 -0.06829 21 James McPherson – June 2019 8-254 Chapter 8: Appendices

LOGLEA: ΔEL(coord) = –970.97243597 Eh LOGKUP D: ΔEL(coord) = –970.97734902 Eh O 3.24131 1.49269 0.05724 O 3.01287 -1.66032 0.33574 O 1.82123 3.26318 0.03578 O 1.75323 -3.44808 -0.29772 N 0.92856 -2.68112 0.11406 N 0.98235 2.51268 0.04033 N -0.45604 -0.62508 -0.11699 N -0.52218 0.54489 -0.15009 N -2.88333 -1.26568 0.07990 N -2.90359 1.28241 0.21576 C 1.59752 -3.83350 0.19199 C 1.69726 3.65607 0.01667 H 0.99511 -4.72973 0.35250 H 1.10869 4.57050 0.11452 C 2.99410 -3.91026 0.06731 C 3.08608 3.70731 -0.07652 H 3.51201 -4.86432 0.13966 H 3.63143 4.64712 -0.07254 C 3.68278 -2.72800 -0.14441 C 3.72967 2.46938 -0.16324 H 4.76717 -2.75639 -0.25234 H 4.81713 2.43249 -0.22556 C 3.03719 -1.54327 -0.22221 C 3.02502 1.30831 -0.18300 H 3.57547 -0.61380 -0.36660 H 3.54846 0.36407 -0.26596 C 1.65366 -1.48697 -0.07441 C 1.62499 1.29128 -0.11162 C 0.83416 -0.30665 -0.11052 C 0.75043 0.12696 -0.12186 C 0.85482 1.16219 -0.07117 C 0.72233 -1.28216 -0.12541 C -0.45468 1.61342 -0.07261 C -0.62391 -1.72119 -0.13331 C -1.27340 0.48215 -0.08301 C -1.38159 -0.48510 -0.11472 C -2.67770 0.09620 -0.02862 C -2.77401 -0.08975 -0.02737 C -3.78440 0.95550 -0.09612 C -3.88557 -0.89126 -0.18530 H -3.64698 2.02634 -0.20067 H -3.75774 -1.94676 -0.40847 C -5.05658 0.42820 -0.04333 C -5.13072 -0.35214 -0.09367 H -5.90686 1.10859 -0.09687 H -6.01002 -0.98008 -0.22841 C -5.27093 -0.91634 0.06907 C -5.28735 1.00626 0.18662 H -6.27310 -1.33623 0.11011 H -6.27644 1.44928 0.27737 C -4.15287 -1.72230 0.12056 C -4.14267 1.77503 0.32489 H -4.24870 -2.80644 0.20569 H -4.21131 2.84543 0.52928 C 2.11554 1.94167 -0.00537 C 1.82367 -2.24669 -0.06332 C 2.96250 4.11931 0.22298 C 4.11951 -2.55178 0.48399 H 3.47153 3.89335 1.16516 H 4.32245 -3.09694 -0.44355 H 2.57394 5.13974 0.23884 H 3.93151 -3.28624 1.27454 H 3.68332 4.00095 -0.59183 H 4.97446 -1.92653 0.75054 C -0.90546 3.06993 -0.03733 C -1.19365 -3.08806 -0.01156 H -0.55405 3.62830 -0.91231 H -1.55930 -3.47912 -0.97464 H -0.51561 3.59708 0.83977 H -2.04972 -3.09524 0.67495 H -1.99540 3.14087 -0.01084 H -0.45139 -3.80558 0.34413

22 James McPherson – June 2019 8-255 Chapter 8: Appendices

Relaxed Py-Pyr- ligand geometries (Table S2; Table S3; Table S5 1 & 2, 29 & 30)

LOGLEA: ΔEL(coord) = –970.97859669 Eh

O 1.86433 -3.35547 -0.47868 O 2.98630 -1.67195 0.42784 N 0.93591 2.53665 0.10366 N -0.52440 0.52331 0.06921 A: ΔEL(relax) = –742.4808698 Eh N -2.92860 1.24206 0.21869 C 1.63271 3.66886 -0.01127 C -1.15933 0.83517 0.00000 H 1.05300 4.59106 0.06889 C -1.20037 2.24041 0.00001 C 3.00397 3.71316 -0.19710 C 0.00000 2.94135 0.00002 H 3.51803 4.66805 -0.29181 C 1.20037 2.24040 0.00002 C 3.69445 2.52082 -0.27904 N 0.00000 0.16594 0.00000 H 4.77339 2.51743 -0.42516 C 1.15933 0.83517 0.00001 C 3.01441 1.33080 -0.15160 C 2.41348 0.00960 0.00001 H 3.53835 0.38517 -0.20530 C 3.69020 0.59775 0.00003 C 1.62097 1.32343 0.00562 C 4.81829 -0.21801 0.00003 C 0.76038 0.15479 0.04782 C 4.65119 -1.60010 0.00003 C 0.74763 -1.28378 -0.05671 C 3.34493 -2.09437 0.00002 C -0.59716 -1.68776 -0.08551 N 2.25683 -1.32528 0.00001 C -1.35860 -0.53225 0.01179 C -2.41348 0.00960 -0.00001 C -2.75245 -0.11234 0.04301 C -3.69020 0.59775 -0.00002 C -3.87123 -0.94425 -0.12235 C -4.81829 -0.21801 -0.00004 H -3.73287 -2.01167 -0.26786 C -4.65119 -1.60010 -0.00005 C -5.14191 -0.39866 -0.12033 C -3.34493 -2.09437 -0.00003 H -6.00923 -1.04187 -0.25737 N -2.25683 -1.32528 -0.00002 C -5.29948 0.96373 0.05895 H 3.16647 -3.16889 0.00001 H -6.29002 1.41479 0.06570 H 5.49986 -2.27653 0.00003 C -4.18121 1.72924 0.22220 H 5.81077 0.22337 0.00004 H -4.25961 2.80840 0.36882 H 3.81535 1.67391 0.00003 C 1.86836 -2.21371 -0.08458 H 2.13721 2.78385 0.00002 C 4.12931 -2.55006 0.45869 H 0.00000 4.02743 0.00002 H 3.93399 -3.41779 1.09534 H -2.13721 2.78385 0.00001 H 4.94663 -1.95321 0.86789 H -3.81536 1.67391 -0.00002 H 4.37935 -2.90958 -0.54355 H -5.49986 -2.27653 -0.00006 C -1.13462 -3.10304 -0.13224 H -5.81077 0.22336 -0.00005 H -1.87088 -3.27362 0.66480 H -3.16647 -3.16889 -0.00004 H -0.33267 -3.83551 -0.03471 H -1.64611 -3.31044 -1.08384

23 James McPherson – June 2019 8-256 Chapter 8: Appendices H 0.00000 2.14839 2.49885 H 0.00000 0.00000 3.75346 H 0.00000 -2.14839 2.49885 N 0.00000 2.53570 -1.56168 C 0.00000 3.85381 -1.84350 C 0.00000 4.65930 -0.68435 C 0.00000 3.75014 0.38737 C 0.00000 2.45027 -0.19052 H 0.00000 4.19732 -2.88022 H 0.00000 5.74711 -0.62818

– H 0.00000 4.00695 1.44317 B : ΔEL(relax) = –703.7867049 Eh N 0.00000 -2.53570 -1.56168 C 0.00000 -2.45027 -0.19052 C 2.67577 -0.23060 0.00006 C 0.00000 -3.75014 0.38737 N 2.63598 -1.60880 -0.00007 C 0.00000 -4.65930 -0.68435 C -2.15495 0.03464 -0.00003 C 0.00000 -3.85381 -1.84350 C -3.38944 0.71823 -0.00007 H 0.00000 -4.00695 1.44317 C -4.58045 0.00154 -0.00012 H 0.00000 -5.74711 -0.62818 C -4.53293 -1.39196 -0.00012 H 0.00000 -4.19732 -2.88022 C -3.27021 -1.98588 -0.00008 N -2.12067 -1.31192 -0.00004 H -3.42221 1.80087 -0.00007 H -5.53310 0.52633 -0.00015 H -5.43537 -1.99666 -0.00015 H -3.17945 -3.07302 -0.00008 C 3.91701 -1.99835 -0.00007 C 4.82566 -0.90823 0.00027 H 4.17323 -3.05740 -0.00008 C 4.02532 0.23399 -0.00012 H 5.91095 -0.95282 0.00050 H 4.37318 1.26195 -0.00019 2 2– C -0.84131 0.76575 0.00002 (C ) : ΔEL(relax) = –1205.778176110 Eh C -0.81145 2.17072 0.00004 C 0.44725 2.78585 0.00007 N 2.35548 2.32629 -0.12987 C 1.58867 2.01274 0.00009 N -0.00001 0.84585 0.02505 C 1.48868 0.58681 0.00013 N -2.35547 2.32633 -0.12966 N 0.25730 0.00492 0.00006 C 3.62492 2.79629 -0.10682 H -1.71003 2.77627 0.00002 C 4.55427 1.76625 0.04964 H 0.52214 3.87214 0.00006 H 5.63808 1.84618 0.02977 H 2.56660 2.48158 0.00007 C 3.80899 0.55342 0.08553 C 2.42676 0.97564 0.01251 C 1.17683 0.18326 0.09543 C 1.20392 -1.20224 0.38661 H 2.13956 -1.71887 0.54827 C -0.00001 -1.87798 0.53605 H -0.00003 -2.93735 0.79501 C -1.20395 -1.20228 0.38638 H -2.13958 -1.71900 0.54776 C -1.17686 0.18325 0.09533 C -2.42678 0.97564 0.01252 C -3.80900 0.55346 0.08540

2– C -4.55428 1.76627 0.04952 C : ΔEL(relax) = –664.9865477 Eh H -5.63809 1.84620 0.02960 C -3.62488 2.79633 -0.10669 C 0.00000 1.17655 0.54554 C 4.49551 -0.73666 0.03717 C 0.00000 1.20268 1.96636 C 5.67777 -0.94773 0.79305 C 0.00000 0.00000 2.66066 H 5.99343 -0.16013 1.47162 C 0.00000 -1.20268 1.96636 C 6.42160 -2.12141 0.69878 C 0.00000 -1.17655 0.54554 H 7.31992 -2.23560 1.30542 N 0.00000 0.00000 -0.12740 C 6.01852 -3.15663 -0.15055 24 James McPherson – June 2019 8-257 Chapter 8: Appendices H 6.59273 -4.07833 -0.21955 H 7.72210 -1.39221 -0.00056 C 4.85655 -2.97644 -0.90987 C 6.06427 -0.01986 -0.00020 H 4.52777 -3.76019 -1.59122 H 6.70362 0.86527 -0.00019 C 4.12072 -1.79743 -0.82612 C 4.65834 0.11088 -0.00006 H 3.24079 -1.66890 -1.44759 C 3.73652 1.17831 0.00040 C -4.49552 -0.73664 0.03708 H 3.97891 2.23476 -0.00138 C -5.67761 -0.94780 0.79318 C 2.44987 0.57735 0.00032 H -5.99315 -0.16027 1.47188 C -1.16956 1.32263 -0.00016 C -6.42144 -2.12150 0.69895 C -1.19950 2.73678 -0.00042 H -7.31961 -2.23577 1.30578 H -2.14598 3.26543 -0.00101 C -6.01851 -3.15662 -0.15056 C 0.00001 3.43618 0.00019 H -6.59271 -4.07833 -0.21953 H 0.00000 4.52668 0.00036 C -4.85671 -2.97633 -0.91012 C 1.19950 2.73678 0.00070 H -4.52808 -3.76000 -1.59163 H 2.14599 3.26541 0.00169 C -4.12089 -1.79731 -0.82642 C 1.16955 1.32262 0.00028 H -3.24111 -1.66868 -1.44806 C 3.90314 4.26886 -0.24027 H 3.40783 4.69456 -1.12378 H 4.98041 4.46315 -0.33085 H 3.53581 4.84434 0.62337 C -3.90308 4.26893 -0.23988 H -3.53734 4.84411 0.62466 H -4.98022 4.46316 -0.33221 H -3.40627 4.69502 -1.12234

– D : ΔEL(relax) = –703.8050017 Eh

N 0.00000 0.25022 0.00000 C 0.00000 -0.56692 1.08348 C 0.00000 -1.94494 0.69244 C 0.00000 -1.94494 -0.69244 C 0.00000 -0.56692 -1.08348 C 0.00000 -0.06105 2.44736 C -0.00000 -0.95349 3.55474 3 2– (C ) : ΔEL(relax) = –972.330651272 Eh C -0.00000 -0.46227 4.84902 C 0.00000 0.92236 5.05629 N -2.50191 -0.78646 0.00023 C 0.00000 1.72798 3.91638 N -0.00000 0.64684 0.00014 N 0.00000 1.28211 2.66122 N 2.50191 -0.78647 0.00022 H -0.00000 -2.02394 3.37949 C -3.82623 -1.08365 0.00006 H -0.00000 -1.14960 5.69342 C -4.43782 -2.35468 0.00040 H 0.00000 1.35689 6.05205 H -3.81138 -3.24530 0.00075 H 0.00000 2.81530 4.02740 C -5.82551 -2.45109 0.00031 C 0.00000 -0.06105 -2.44736 H -6.30101 -3.43229 0.00057 C -0.00000 -0.95349 -3.55474 C -6.63640 -1.28769 -0.00007 C -0.00000 -0.46227 -4.84902 H -7.72209 -1.39222 -0.00014 C 0.00000 0.92236 -5.05629 C -6.06427 -0.01987 -0.00035 C 0.00000 1.72798 -3.91638 H -6.70363 0.86525 -0.00064 N 0.00000 1.28211 -2.66122 C -4.65834 0.11088 -0.00034 H -0.00000 -2.02394 -3.37949 C -3.73652 1.17831 -0.00054 H -0.00000 -1.14960 -5.69342 H -3.97892 2.23476 -0.00125 H 0.00000 1.35689 -6.05205 C -2.44988 0.57736 -0.00011 H 0.00000 2.81530 -4.02740 C 3.82622 -1.08366 -0.00001 H 0.00000 -2.81915 -1.33446 C 4.43782 -2.35468 -0.00006 H 0.00000 -2.81915 1.33446 H 3.81140 -3.24531 0.00011 C 5.82551 -2.45108 -0.00030 H 6.30103 -3.43227 -0.00037 C 6.63641 -1.28768 -0.00040 25 James McPherson – June 2019 8-258 Chapter 8: Appendices

2 2– 3 2– (D ) : ΔEL(relax) = –971.007250295 Eh (D ) : ΔEL(relax) = –1684.665547790 Eh

O 2.82920 -1.99356 -0.62927 S 1.67885 2.21762 -0.88976 O 1.30920 -3.22844 0.46768 S -1.67903 2.21755 -0.88937 N 1.70626 2.64536 0.08192 N 2.45874 -2.49098 0.21028 N -0.41645 0.99136 -0.03572 N -0.00007 -1.31180 -0.00633 N -3.07578 1.67848 0.00296 N -2.45899 -2.49079 0.21017 C 2.71961 3.50311 0.15420 N 2.57233 3.45567 1.55360 H 2.44025 4.55606 0.23997 N -2.57115 3.45597 1.55431 C 4.06896 3.14484 0.12920 C 3.63521 -3.10242 0.32728 H 4.85140 3.89611 0.19717 H 3.59109 -4.17315 0.53434 C 4.35222 1.78475 0.00561 C 4.87289 -2.46856 0.20552 H 5.38190 1.43355 -0.03108 H 5.79967 -3.02457 0.31395 C 3.31102 0.86822 -0.08039 C 4.85914 -1.09871 -0.05360 H 3.51664 -0.18456 -0.20918 H 5.78764 -0.54169 -0.15420 C 1.97132 1.31794 -0.03019 C 3.64194 -0.43996 -0.17767 C 0.79259 0.42384 -0.09351 H 3.62487 0.62240 -0.37494 C 0.66542 -1.02847 -0.11450 C 2.43565 -1.16241 -0.04613 C -0.73442 -1.28375 -0.08902 C 1.10028 -0.54136 -0.16004 C -1.34918 -0.00507 -0.03730 C 0.71101 0.80096 -0.42677 C -2.77103 0.35273 -0.01674 C -0.71096 0.80104 -0.42674 C -3.82055 -0.60241 0.02937 C -1.10037 -0.54124 -0.16011 H -3.60060 -1.65896 0.05706 C -2.43580 -1.16223 -0.04626 C -5.14525 -0.18755 0.05782 C -3.64204 -0.43970 -0.17773 H -5.94211 -0.92828 0.09067 H -3.62490 0.62268 -0.37494 C -5.44053 1.17620 0.05332 C -4.85928 -1.09834 -0.05368 H -6.46118 1.54764 0.07519 H -5.78775 -0.54124 -0.15425 C -4.35136 2.05075 0.03221 C -4.87315 -2.46821 0.20543 H -4.52555 3.12937 0.04401 H -5.79998 -3.02415 0.31385 C 1.69976 -2.04843 -0.14875 C -3.63551 -3.10215 0.32716 C 2.26909 -4.27698 0.40509 H -3.59147 -4.17288 0.53421 H 2.52339 -4.53318 -0.62889 C 2.16867 2.92352 0.59587 H 1.80692 -5.13399 0.90156 C -2.16806 2.92369 0.59642 H 3.19572 -4.00052 0.91856 C -1.40778 -2.62348 -0.23639 H -0.68788 -3.40539 -0.47121 H -2.15803 -2.60660 -1.03675 H -1.92556 -2.93904 0.68148

2– E : ΔEL(relax) = –664.9069983 Eh

N 0.30762 -0.38111 0.00000 C 1.40862 0.39232 0.00000 26 James McPherson – June 2019 8-259 Chapter 8: Appendices C 1.04483 1.79272 0.00000 N 1.98029 -1.41327 0.01016 C -0.33537 1.84285 0.00000 N 2.29091 -0.07464 -0.03033 C -0.78160 0.47621 0.00000 C 3.62618 0.09583 -0.02065 C 2.79813 -0.08413 0.00000 H 4.07165 1.08078 -0.04665 N 3.16427 -1.40919 0.00000 C 4.20381 -1.14800 0.04937 C -2.14184 0.04936 0.00000 H 5.26763 -1.35763 0.07532 C -3.22405 1.00233 0.00000 C 3.13350 -2.05621 0.06271 C -4.53807 0.59068 0.00000 H 3.14791 -3.13853 0.10141 C -4.84105 -0.78871 0.00000 C -3.73870 -1.65139 0.00000 N -2.45986 -1.29728 0.00000 H -2.99161 2.06317 0.00000 H -5.33865 1.33333 0.00000 H -5.86174 -1.16514 0.00000 H -3.92019 -2.73440 -0.00000 H -0.94703 2.74187 0.00000 H 1.72801 2.63726 0.00000 C 4.51922 -1.41064 0.00000 C 5.06510 -0.11259 0.00000 H 5.07089 -2.35476 0.00000 KEPSUV: ΔEL(relax) = –681.620400283 Eh C 3.94697 0.74889 0.00000 H 6.11860 0.16795 0.00000 N -2.54625 -1.52464 0.00047 H 3.96437 1.83627 -0.00000 N -2.46242 -0.17512 0.00019 C -3.70389 0.39329 -0.00046 H -3.84566 1.46137 -0.00093 C -4.63662 -0.62769 -0.00056 Strain within pincer ligand geometries H -5.71384 -0.53077 -0.00090 from SC-XRD structures (Table S4; Table C -3.85052 -1.79808 0.00003 H -4.17521 -2.83189 0.00016 S5 23–26) C -0.00002 -0.26489 0.00026 C -1.16093 0.51373 0.00007 C -1.20656 1.91859 0.00005 H -2.13298 2.49086 0.00010 C 0.00004 2.61742 0.00017 H 0.00007 3.70550 0.00025 C 1.20661 1.91853 0.00026 H 2.13306 2.49077 0.00056 C 1.16094 0.51369 0.00015 N 2.54624 -1.52468 0.00042 N 2.46239 -0.17516 0.00024 C 3.70386 0.39328 -0.00044 KEPSUV: ΔEL(coord) = –681.57708049 Eh H 3.84560 1.46136 -0.00077 C 4.63660 -0.62768 -0.00055 N -1.99473 -1.39905 -0.08869 H 5.71382 -0.53072 -0.00101 N -2.29861 -0.05695 0.00632 C 3.85052 -1.79808 -0.00010 C -3.63523 0.08410 0.09243 H 4.17522 -2.83189 -0.00001 H -4.09260 1.06072 0.18024 C -4.19769 -1.14068 0.03981 H -5.26099 -1.35143 0.08924 C -3.13321 -2.05703 -0.07191 H -3.14435 -3.13736 -0.13845 C 0.01133 0.15909 -0.02128 C -1.21155 0.84643 -0.00671 C -1.22687 2.23174 0.01953 H -2.13618 2.84450 0.05451 C 0.01100 2.88778 0.00227 H 0.01464 3.97811 0.01286 C 1.24069 2.22136 -0.03499 H 2.15734 2.82308 -0.05384 C 1.19971 0.84312 -0.03672 27 James McPherson – June 2019 8-260 Chapter 8: Appendices N -3.25009 -1.51652 0.00792 C -1.15087 1.22904 -0.00861 C -1.21152 2.60304 -0.01926 C 0.00000 3.27629 0.00000 C -2.03355 -0.98808 0.02073 C -3.61654 0.64057 -0.04592 C -4.22495 -0.52829 -0.03529 C -3.51850 -2.94952 0.07173 H -2.14950 3.14740 -0.03226 H 0.00000 4.36292 0.00000 H -4.02900 1.63732 -0.07862 WOXFEU: ΔEL(coord) = –776.83661624 Eh H -5.28509 -0.74250 -0.05506 H -4.07488 -3.19904 0.98090 C 1.99662 -0.99519 0.00616 H -4.09442 -3.27572 -0.80025 N 3.21098 -1.53591 -0.02121 H -2.55072 -3.44929 0.08349 C 3.51661 -2.96734 0.00942 N 2.25105 0.35814 0.00720 H 4.11618 -3.24910 -0.86139 N 3.25008 -1.51652 -0.00792 H 2.56486 -3.49768 -0.00968 C 1.15088 1.22904 0.00861 H 4.06357 -3.22241 0.92217 C 1.21154 2.60304 0.01927 C 4.19617 -0.55337 -0.01725 C 2.03354 -0.98808 -0.02073 H 5.25437 -0.77802 -0.03436 C 3.61654 0.64057 0.04591 C 3.59974 0.62917 0.00270 C 4.22496 -0.52831 0.03528 H 4.02838 1.61982 0.00744 C 3.51849 -2.94953 -0.07172 N 2.23543 0.36774 0.02176 H 2.14952 3.14740 0.03226 C 1.15256 1.25363 0.01554 H 4.02900 1.63732 0.07862 N 0.00681 0.60068 0.00772 H 5.28509 -0.74252 0.05505 C -1.16370 1.25866 0.00134 H 4.07485 -3.19905 -0.98090 C -1.20677 2.63077 -0.01112 H 4.09440 -3.27573 0.80025 H -2.13673 3.19051 -0.02433 H 2.55070 -3.44930 -0.08348 C 0.01146 3.29708 -0.00406 H 0.01800 4.38328 -0.01200 C 1.20971 2.61508 0.00700 H 2.14910 3.15924 0.01087 N -2.25246 0.36454 -0.01401 C -2.00950 -1.00103 -0.02559 N -3.23053 -1.53346 -0.00628 C -3.49835 -2.97092 0.00293 H -4.08340 -3.26046 -0.87557 H -4.04457 -3.25536 0.90769 H -2.52976 -3.46890 -0.01650 C -4.20679 -0.54770 0.01640 H -5.26579 -0.76830 0.03198 WOXFEU/YOKGUB: ΔEL(relax) = –776.8629151 Eh C -3.60446 0.62773 0.00060 H -3.88560 1.64282 0.01325 N 0.00000 0.34678 0.00000 N -2.33824 0.24581 0.00014 N -3.68054 -1.40854 -0.00018 C -1.14541 1.02055 0.00015 C -1.20572 2.42273 0.00037 C 0.00000 3.11427 0.00000 C -2.34343 -1.14564 -0.00001 C -3.63321 0.78652 0.00029 C -4.48021 -0.26564 0.00018 C -4.19942 -2.76639 -0.00037 H -2.14221 2.96380 0.00093 H 0.00000 4.20018 0.00000 H -3.85459 1.84022 0.00020 YOKGUB: ΔEL(coord) = –776.84060739 Eh H -5.55977 -0.29298 -0.00000 H -4.80917 -2.95179 0.89016 N -0.00000 0.55420 0.00000 H -4.80886 -2.95176 -0.89126 N -2.25104 0.35814 -0.00720 H -3.34169 -3.43778 -0.00027 28 James McPherson – June 2019 8-261 Chapter 8: Appendices N 2.33824 0.24581 -0.00014 N 3.68054 -1.40854 0.00018 C 1.14541 1.02055 -0.00015 C 1.20572 2.42273 -0.00037 C 2.34343 -1.14564 0.00001 C 3.63321 0.78652 -0.00029 C 4.48021 -0.26564 -0.00018 C 4.19942 -2.76639 0.00037 H 2.14221 2.96380 -0.00093 H 3.85459 1.84022 -0.00020 H 5.55977 -0.29298 0.00000 H 4.80917 -2.95179 -0.89016 CEXYIP: ΔEL(relax) = –708.969073512 Eh H 4.80886 -2.95176 0.89126 H 3.34169 -3.43778 0.00027 N 0.00001 0.03231 -0.00015 C 3.75213 -2.10673 0.00010 H 3.83848 -3.20741 0.00022 C 2.45331 -1.52985 0.00016 C -1.18056 0.69097 -0.00004 C -2.47943 -0.09706 -0.00023 C -2.45335 -1.52984 -0.00053 C 2.47944 -0.09707 0.00000 C 4.93323 -0.00916 0.00015 H 5.85518 0.57583 0.00035 C -0.00003 2.80433 0.00001 H -0.00001 3.89639 0.00007 C -1.19985 2.10851 0.00050 CEXYIP: ΔEL(coord) = –708.93490374 Eh H -2.13039 2.66352 0.00128 C 4.96488 -1.40841 -0.00010 N 0.01205 0.33341 0.00825 H 5.93021 -1.93201 -0.00039 C 3.11010 -2.23048 -0.02297 C -3.75211 -2.10673 -0.00018 H 2.95210 -3.32366 -0.04387 H -3.83849 -3.20744 -0.00075 C 2.04140 -1.34990 -0.01069 C -4.96488 -1.40840 0.00058 C -1.20545 0.98861 -0.00779 H -5.93027 -1.93201 0.00122 C -2.35562 0.06880 -0.02714 C -3.69546 0.63487 -0.00037 C -2.03818 -1.32130 -0.01259 H -3.72120 1.72689 -0.00117 C 2.35295 0.07161 0.02164 C 1.18061 0.69100 -0.00016 C 4.69614 -0.41800 0.01689 C 3.69549 0.63487 0.00035 H 5.73077 -0.06570 0.01907 H 3.72117 1.72687 0.00097 C 0.01113 3.04624 -0.01019 C 1.19984 2.10850 -0.00053 H 0.01630 4.13885 -0.02258 H 2.13033 2.66358 -0.00120 C -1.19893 2.37567 -0.03461 C -4.93323 -0.00917 0.00029 H -2.12883 2.94556 -0.06385 H -5.85518 0.57584 0.00031 C 4.42842 -1.76962 -0.02619 H 5.28497 -2.46732 -0.04673 C -3.11186 -2.23217 0.01179 H -2.94193 -3.32097 0.02966 C -4.44297 -1.76964 0.02409 H -5.29943 -2.46670 0.04013 C -3.67435 0.49132 -0.00082 H -3.93680 1.56169 0.00394 C 1.20861 0.96243 0.02277 C 3.66012 0.49961 0.02443 H 3.92650 1.56791 0.05038 C 1.20813 2.36815 0.01665 H 2.14223 2.93159 0.03250 PUHKOS: ΔEL(coord) = –894.93190061 Eh C -4.70493 -0.41117 0.00266 H -5.73843 -0.05600 0.01610 N -2.06758 -1.77121 -0.03249 N -0.00011 -0.19752 -0.01889 N 2.06632 -1.76749 -0.04207 C -3.13420 -2.54922 -0.01614 29 James McPherson – June 2019 8-262 Chapter 8: Appendices C -4.43945 -2.05408 0.03336 C 2.41181 -0.47555 -0.00007 C -4.66261 -0.69900 0.03854 C 3.64165 0.24285 0.00002 C -3.55664 0.17734 0.00318 C 4.83735 -0.50525 0.00001 C -2.29174 -0.41105 -0.02051 C 4.77541 -1.88410 0.00001 C -3.64503 1.61577 0.00426 C 3.50582 -2.49310 0.00015 C -2.54785 2.41022 0.01800 H -3.42579 -3.57937 -0.00002 C -1.14212 0.44165 -0.01359 H -5.67480 -2.49168 -0.00018 C -1.21825 1.84990 0.00329 H -5.79225 0.01420 -0.00002 C -0.00082 2.54484 -0.00494 H -4.60157 2.19560 0.00029 C 1.21756 1.85377 -0.01213 H -2.48996 3.47046 0.00007 C 1.15539 0.44717 -0.01669 H 0.00006 3.49065 0.00002 C 2.53812 2.41523 -0.01794 H 2.48989 3.47044 0.00012 C 3.63529 1.61976 -0.00158 H 4.60156 2.19566 0.00013 C 2.29576 -0.41397 -0.02287 H 5.79228 0.01413 -0.00009 C 3.56202 0.17358 0.01551 H 5.67483 -2.49162 -0.00009 C 4.65703 -0.69457 0.05802 H 3.42575 -3.57936 0.00001 C 4.44141 -2.04756 0.04432 C 3.13983 -2.55943 -0.02271 H -2.95349 -3.62324 -0.03344 H -5.27679 -2.74632 0.05996 H -5.67320 -0.29935 0.06970 H -4.63469 2.06624 0.01279 H -2.66290 3.49156 0.02829 H -0.00132 3.63286 0.00038 H 2.65605 3.49625 -0.02841 H 4.62340 2.07428 0.00364 H 5.66802 -0.29606 0.09940 H 5.28467 -2.73280 0.07519 H 2.95766 -3.63184 -0.04957 IZAFOG: ΔEL(coord) = –740.44345037 Eh

O -3.59281 0.27248 0.07071 O 3.57001 0.28435 -0.01659 N -1.97746 -1.31525 -0.05939 N -0.00246 0.21733 -0.01912 N 1.97837 -1.35306 -0.02134 C -3.25521 -2.11168 -0.04274 H -3.25354 -2.77748 0.82596 H -3.31048 -2.72601 -0.94646 C -4.32224 -1.02981 0.02229 H -4.97038 -0.98459 -0.85718 H -4.94785 -1.06422 0.91806 PUHKOS: ΔEL(relax) = –894.949534156 Eh C -2.28410 -0.07067 0.01661 C -1.19580 0.89733 0.00866 N -2.36240 -1.82461 0.00000 C -1.20001 2.30928 0.02805 N -0.00003 -0.39947 -0.00007 H -2.14643 2.84220 0.06234 N 2.36240 -1.82458 0.00002 C -0.00121 3.01342 -0.00326 C -3.50578 -2.49308 0.00007 H 0.00841 4.09723 0.00264 C -4.77544 -1.88408 -0.00004 C 1.20819 2.28309 -0.04396 C -4.83737 -0.50527 0.00001 H 2.15928 2.80737 -0.06672 C -3.64160 0.24282 0.00005 C 1.17226 0.90972 -0.03609 C -2.41184 -0.47552 -0.00005 C 2.28011 -0.08824 -0.02952 C -3.64409 1.68127 0.00010 C 4.34347 -0.99718 0.06909 C -2.48180 2.38369 0.00004 H 4.90625 -0.95818 1.00491 C -1.15330 0.27499 -0.00010 H 5.03848 -0.99764 -0.77334 C -1.21213 1.70976 -0.00005 C 3.26701 -2.11910 0.02082 C -0.00002 2.40282 -0.00004 H 3.35000 -2.75082 -0.86931 C 1.21214 1.70968 -0.00005 H 3.28474 -2.76255 0.90545 C 1.15337 0.27503 -0.00012 C 2.48176 2.38367 0.00003 C 3.64409 1.68128 0.00004 30 James McPherson – June 2019 8-263 Chapter 8: Appendices H 4.19904 -2.41302 -0.87912 H 4.19833 -2.40920 0.88401 C 4.64934 -0.42759 -0.00171 H 5.25998 -0.24258 0.88688 H 5.25511 -0.24409 -0.89402 C 2.40275 -0.24465 0.00018 C 1.14557 0.54320 0.00017 C 1.19933 1.94586 0.00019 H 2.15715 2.45074 0.00024 C 0.00001 2.65120 0.00017 H 0.00001 3.73696 0.00017 IZAFOG01: ΔEL(coord) = –740.45869954 Eh C -1.19931 1.94586 0.00010 H -2.15714 2.45075 0.00009 O 3.53488 0.33372 -0.10781 C -1.14556 0.54320 0.00005 O -3.54011 0.35113 -0.02824 C -2.40275 -0.24465 -0.00004 N 2.09405 -1.38814 0.03569 C -4.64935 -0.42759 -0.00025 N -0.00541 0.13721 -0.03324 H -5.25737 -0.24336 -0.89084 N -2.11081 -1.37686 0.01241 H -5.25771 -0.24333 0.89010 C 3.43617 -2.01306 0.07349 C -3.94657 -1.81113 -0.00012 H 3.54218 -2.71444 -0.76151 H -4.19871 -2.41102 0.88151 H 3.54548 -2.58922 0.99863 H -4.19868 -2.41117 -0.88165 C 4.41099 -0.83358 -0.01251 H 5.03870 -0.70315 0.87362 H 5.04917 -0.83909 -0.90026 C 2.29359 -0.13869 -0.04267 C 1.15671 0.79455 -0.03320 C 1.20581 2.18635 0.00311 H 2.15403 2.71340 0.00050 C 0.00436 2.85592 0.07454 H 0.00688 3.94202 0.12482 C -1.18117 2.19268 0.06312 H -2.12619 2.72546 0.10740 C -1.15872 0.80236 -0.00052 C -2.29745 -0.12793 -0.01230 XAMQUZ: ΔEL(coord) = –1482.59857081 Eh C -4.39027 -0.84703 -0.09248 H -4.89309 -0.82516 -1.06515 P 2.23414 -0.49677 0.11905 H -5.14874 -0.74350 0.68802 N -0.00001 1.36674 0.00046 C -3.45571 -2.00126 0.07683 C 2.33945 1.25681 0.73270 H -3.56233 -2.51347 1.04067 C 1.12614 2.06420 0.32336 H -3.55505 -2.75892 -0.70682 C 1.14920 3.45986 0.33096 C -0.00006 4.15998 -0.00031 C 3.16165 -1.44444 1.39040 C 2.96471 -2.95027 1.26661 C 4.63685 -1.07451 1.52001 C 3.29460 -0.42204 -1.39405 C 3.36680 -1.78696 -2.08050 C 2.76483 0.64702 -2.34528 H 3.34781 -3.45842 2.15983 H 2.35612 1.20352 1.83084 H -0.00010 5.24659 -0.00053 H 2.06226 3.98612 0.59598 H 4.78107 0.00028 1.67180 IZAFOG/IZAFOG01: ΔEL(relax) = –740.4786437 Eh H 3.49953 -3.36767 0.40808 H 3.26075 1.77818 0.44462 O 3.54960 0.51332 0.00047 H 2.65417 -1.13572 2.31743 O -3.54960 0.51332 -0.00005 H 4.32206 -0.13011 -1.12449 N 2.50903 -1.51311 0.00012 H 5.19987 -1.35939 0.62431 N 0.00000 -0.14654 0.00009 H 5.10924 -1.58905 2.36757 N -2.50904 -1.51310 -0.00009 H 1.90743 -3.20862 1.15531 C 3.94657 -1.81113 0.00101 H 3.89271 -2.52869 -1.47420 31 James McPherson – June 2019 8-264 Chapter 8: Appendices H 1.72276 0.45367 -2.62396 H -0.00002 5.34229 0.00002 H 2.36521 -2.17822 -2.29158 H 2.14969 4.06930 0.11579 H 3.35858 0.66121 -3.26748 H 5.24280 0.53956 -0.24721 H 3.90294 -1.70826 -3.03436 H 4.17321 -3.02672 0.71507 H 2.80835 1.65118 -1.91495 H 3.14184 1.69737 -0.60189 P -2.23404 -0.49691 -0.11857 H 4.04849 -0.18098 1.84629 C -2.33900 1.25633 -0.73325 H 3.48367 -0.45077 -1.95783 C -1.12597 2.06399 -0.32357 H 5.35936 -1.15003 -0.75645 C -1.14909 3.45964 -0.33195 H 6.19635 -0.60852 0.69753 C -3.16069 -1.44531 -1.38999 H 3.29814 -2.52341 2.16369 C -2.96377 -2.95106 -1.26525 H 3.52129 -2.87059 -1.34416 C -4.63581 -1.07549 -1.52076 H 0.42374 -0.70180 -2.09414 C -3.29548 -0.42137 1.39381 H 1.75618 -2.94166 -1.44252 C -3.36808 -1.78592 2.08095 H 1.51103 -0.62994 -3.49793 C -2.76635 0.64822 2.34479 H 2.71721 -2.65920 -2.90090 H -3.34627 -3.45971 -2.15844 H 1.36296 0.75641 -2.41174 H -2.35497 1.20244 -1.83138 P -2.32537 -0.51511 -0.22590 H -2.06209 3.98573 -0.59752 C -2.45544 1.36783 -0.18996 H -4.77997 -0.00080 -1.67324 C -1.15731 2.14866 -0.07189 H -3.49913 -3.36802 -0.40685 C -1.19745 3.54968 -0.06412 H -3.26050 1.77783 -0.44605 C -4.05042 -0.84845 -0.97225 H -2.65262 -1.13708 -2.31686 C -4.14818 -2.28334 -1.51816 H -4.32277 -0.12963 1.12343 C -5.27357 -0.49290 -0.11527 H -5.19941 -1.35990 -0.62528 C -2.54843 -0.91351 1.61478 H -5.10764 -1.59052 -2.36835 C -2.64513 -2.43242 1.83030 H -1.90656 -3.20931 -1.15312 C -1.39237 -0.33576 2.44739 H -3.89358 -2.52800 1.47472 H -5.06679 -2.40891 -2.10437 H -1.72445 0.45505 2.62425 H -2.93697 1.65669 -1.13326 H -2.36662 -2.17704 2.29290 H -2.14972 4.06930 -0.11577 H -3.36070 0.66290 3.26661 H -5.24280 0.53954 0.24729 H -3.90485 -1.70672 3.03443 H -4.17324 -3.02672 -0.71505 H -2.80962 1.65215 1.91388 H -3.14186 1.69735 0.60189 H -4.04855 -0.18098 -1.84625 H -3.48361 -0.45079 1.95785 H -5.35933 -1.15005 0.75652 H -6.19638 -0.60854 -0.69743 H -3.29821 -2.52341 -2.16370 H -3.52124 -2.87061 1.34416 H -0.42367 -0.70181 2.09405 H -1.75613 -2.94167 1.44246 H -1.51092 -0.62996 3.49787 H -2.71711 -2.65923 2.90088 H -1.36289 0.75639 2.41168

XAMQUZ: ΔEL(relax) = –1482.615905080 Eh

P 2.32535 -0.51511 0.22588 N -0.00001 1.47505 -0.00000 C 2.45542 1.36783 0.18995 C 1.15729 2.14866 0.07189 C 1.19743 3.54968 0.06413 C -0.00001 4.25574 0.00001 C 4.05039 -0.84844 0.97228 C 4.14813 -2.28334 1.51819 C 5.27356 -0.49289 0.11535 ACAROM: ΔEL(coord) = –1546.08474654 Eh C 2.54848 -0.91350 -1.61481 C 2.64519 -2.43240 -1.83033 S 2.12981 2.16371 0.36921 C 1.39245 -0.33575 -2.44744 S -2.31785 2.14241 -0.39013 H 5.06673 -2.40891 2.10443 N -0.04717 0.29053 -0.04998 H 2.93694 1.65669 1.13326 N 3.47900 -0.09968 0.50235 32 James McPherson – June 2019 8-265 Chapter 8: Appendices N -3.54517 -0.13670 0.23792 H -2.22883 1.98857 -0.92502 C 1.14954 -0.30141 -0.27638 C -0.10346 2.23733 -1.03569 C 1.21835 -1.47516 -1.01320 H -0.15988 3.18136 -1.56969 H 2.16763 -1.90951 -1.29877 C 1.12891 1.65396 -0.77090 C 0.05438 -2.06918 -1.44841 H 2.03938 2.11962 -1.12457 H 0.10067 -2.96968 -2.05444 C 1.15949 0.40462 -0.11734 C -1.16835 -1.50296 -1.14673 C -2.37750 -0.53907 0.25428 H -2.08222 -1.95261 -1.52047 C -3.88094 1.33858 0.86162 C -1.18892 -0.32160 -0.42326 H -2.93202 1.77076 1.17783 C 2.32825 0.47413 0.20951 H -4.41619 1.07671 1.78641 C 3.69306 -1.54574 0.74170 C -4.69726 2.36492 0.07554 H 2.72968 -2.05133 0.75559 H -4.18441 2.68659 -0.83582 H 4.10944 -1.64554 1.75186 H -4.87199 3.24896 0.69733 C 4.63799 -2.17452 -0.26655 H -5.67374 1.97072 -0.21935 H 4.28645 -2.04213 -1.29404 C -4.81098 -0.79143 0.04364 H 4.73162 -3.24890 -0.07419 H -5.67882 -0.15433 0.22948 H 5.64006 -1.74293 -0.20330 H -4.77037 -1.54548 0.83769 C 4.66335 0.73398 0.80110 C -4.95658 -1.45779 -1.32298 H 5.34740 0.11143 1.38678 H -5.02472 -0.70898 -2.11802 H 4.33704 1.57058 1.42101 H -5.86690 -2.06563 -1.34237 C 5.34793 1.23615 -0.45341 H -4.10990 -2.11446 -1.53297 H 5.69907 0.41625 -1.08602 C 2.45270 -0.37399 0.01873 H 6.21481 1.84789 -0.18119 C 3.87247 1.55683 0.61429 H 4.66749 1.86239 -1.03418 H 2.92833 2.06496 0.80382 C -2.43549 0.45449 -0.14762 H 4.50073 2.24072 0.03136 C -3.64086 -1.53779 0.72440 C 4.53247 1.23988 1.96225 H -2.72325 -2.06241 0.47782 H 3.91112 0.55615 2.54747 H -4.45547 -2.03146 0.18052 H 4.65810 2.16548 2.53384 C -3.86508 -1.61445 2.21336 H 5.51922 0.78483 1.84243 H -3.05169 -1.12276 2.75541 C 4.84401 -0.38319 -0.57228 H -3.89444 -2.66340 2.52820 H 5.05601 -1.16273 0.16801 H -4.80551 -1.15629 2.52984 H 5.65988 0.34526 -0.53634 C -4.82723 0.62424 0.27906 C 4.77033 -0.98551 -1.97456 H -4.57621 1.67176 0.43714 H 3.97646 -1.73185 -2.04422 H -5.39315 0.26867 1.14378 H 5.71774 -1.47911 -2.21363 C -5.60252 0.42452 -0.98737 H 4.58731 -0.20883 -2.72333 H -5.04779 0.81586 -1.84521 H -6.55674 0.95990 -0.93563 H -5.82922 -0.63088 -1.17510 Relaxed ligand geometries from iron(II)

complexes (Table S5)

ACAROM: ΔEL(relax) = –1546.078130720 Eh

S -2.18860 -2.11583 0.74110 DANMOU: ΔEL(coord) = –742.480870195 Eh S 2.44617 -1.96300 0.49479 N 0.03055 -0.25065 0.16342 C -1.18811 1.06716 -0.00009 N -3.61652 0.07424 0.13134 C -1.22581 2.46236 0.00001 N 3.61561 0.35106 -0.21227 C -0.00001 3.14296 0.00011 C -1.16118 0.32764 -0.01430 C 1.22579 2.46237 0.00011 C -1.26664 1.57545 -0.65929 33 James McPherson – June 2019 8-266 Chapter 8: Appendices N -0.00001 0.46236 -0.00009 H -4.00889 1.53059 -0.00016 C 1.18809 1.06717 0.00001 H -3.69351 -2.98200 -0.00029 C 2.31519 0.08937 0.00001 C 3.65119 0.46387 0.00011 C 4.64729 -0.51753 0.00011 C 4.28190 -1.86073 0.00001 C 2.92960 -2.18523 -0.00019 N 1.96460 -1.24373 -0.00019 C -2.31511 0.08946 -0.00009 C -3.65121 0.46386 -0.00019 C -4.64731 -0.51755 -0.00009 C -4.28190 -1.86075 0.00011 C -2.92960 -2.18524 0.00021 N -1.96460 -1.24374 0.00011 H 2.60450 -3.21943 -0.00029 BUKDUH: ΔEL(relax) = –1005.64335514 Eh H 5.02300 -2.65132 0.00001 H 5.69320 -0.22962 0.00031 N -2.38855 2.75721 0.00005 H 3.91989 1.51337 0.00031 N -0.00080 1.40692 -0.00002 H 2.15639 3.01637 0.00011 N 2.38545 2.75987 -0.00001 H -0.00002 4.22786 0.00011 N -1.18307 -0.65370 -0.00009 H -2.15641 3.01636 0.00001 N 1.18380 -0.65238 -0.00006 H -3.91991 1.51335 -0.00029 C -3.54593 3.41898 0.00010 H -5.02300 -2.65135 0.00021 H -3.47375 4.50526 0.00021 H -5.69320 -0.22965 -0.00019 C -4.79753 2.79638 0.00010 H -2.60440 -3.21944 0.00041 H -5.70636 3.38988 0.00021 C -4.83900 1.40394 -0.00002 H -5.78792 0.87571 -0.00004 C -3.63940 0.69737 -0.00010 H -3.61450 -0.38482 -0.00018 C -2.43511 1.41581 -0.00008 C -1.12826 0.68724 -0.00012 C 0.00071 -1.28527 -0.00006 C 1.12748 0.68851 -0.00005 C 2.43351 1.41853 -0.00004 C 3.63861 0.70143 -0.00006 H 3.61490 -0.38079 -0.00013 C 4.83741 1.40933 0.00002 DIXDIX: ΔEL(relax) = Eh H 5.78692 0.88217 0.00004 C 4.79439 2.80173 0.00009 N 2.32735 -1.38155 0.00012 H 5.70257 3.39622 0.00021 C 3.50590 -1.91469 -0.00008 C 3.54210 3.42293 0.00003 S 4.84194 -0.77775 0.00007 H 3.46870 4.50913 0.00005 C 3.68234 0.50170 0.00017 C 0.00156 -2.76872 -0.00000 C 2.40109 0.00000 -0.00016 C -1.20851 -3.47973 0.00005 N -0.00001 0.12937 -0.00000 H -2.14004 -2.92639 0.00007 C 1.15610 0.80694 -0.00022 C -1.20584 -4.87210 0.00008 C 1.19991 2.21118 -0.00012 H -2.14746 -5.41315 0.00013 C 0.00003 2.91295 0.00001 C 0.00314 -5.57161 0.00006 C -1.19988 2.21119 0.00003 H 0.00375 -6.65781 0.00006 C -1.15614 0.80697 0.00016 C 1.21132 -4.87074 0.00003 C -2.40108 0.00006 0.00001 H 2.15355 -5.41072 0.00005 C -3.68236 0.50153 -0.00013 C 1.21243 -3.47836 0.00000 S -4.84210 -0.77764 0.00021 H 2.14333 -2.92398 -0.00001 C -3.50559 -1.91494 -0.00016 N -2.32730 -1.38161 -0.00028 H 3.69358 -2.98176 -0.00003 H 4.00904 1.53062 0.00029 H 2.14381 2.74457 -0.00015 H 0.00003 3.99894 -0.00009 H -2.14375 2.74460 -0.00002 34 James McPherson – June 2019 8-267 Chapter 8: Appendices

N 0.00001 -0.33520 0.00019 C 1.14158 0.34483 0.00015 C 1.21443 1.74510 0.00017 H 2.13746 2.30686 0.00019 C 0.00003 2.41235 0.00007 C -1.21442 1.74505 0.00008 H -2.13739 2.30694 -0.00015 C -1.14164 0.34485 0.00019 N 2.33043 -0.42240 0.00016 N 2.27957 -1.78401 0.00020 DOMQIH: ΔEL(relax) = –949.749856724 Eh C 3.54630 -2.16565 -0.00000 H 3.78094 -3.22175 0.00005 N -2.63577 1.87796 -0.00009 C 4.44218 -1.06438 -0.00040 N -0.54626 0.06046 -0.00031 H 5.52170 -1.07729 -0.00061 N 2.20069 0.96380 -0.00028 C 3.62708 0.03845 -0.00026 N 2.86288 -1.26907 -0.00000 H 3.86861 1.08874 -0.00040 C -3.63900 2.73865 0.00003 N -2.33042 -0.42242 0.00012 H -3.36582 3.79440 -0.00006 N -2.27955 -1.78403 0.00005 C -4.99925 2.36800 0.00019 C -3.54629 -2.16566 0.00016 H -5.77654 3.12698 0.00035 H -3.78095 -3.22175 0.00052 C -5.30350 1.02315 0.00024 C -4.44217 -1.06439 -0.00023 H -6.33642 0.68079 0.00035 H -5.52168 -1.07732 -0.00061 C -4.26265 0.06800 0.00010 C -3.62706 0.03843 -0.00049 C -2.91606 0.55341 0.00000 H -3.86858 1.08874 -0.00052 C -4.52501 -1.33874 0.00009 F -0.00006 3.75436 -0.00001 H -5.55821 -1.67814 0.00010 C -3.49138 -2.23031 0.00003 H -3.68839 -3.30089 -0.00001 C -2.13233 -1.79503 -0.00001 C -1.80933 -0.40442 -0.00011 C -1.04395 -2.70411 0.00005 H -1.24570 -3.77400 0.00017 C 0.23914 -2.22782 -0.00002 H 1.10935 -2.87304 0.00010 C 0.46722 -0.81520 -0.00036 C 1.85331 -0.34664 -0.00035 C 3.97130 -0.48364 0.00001 C 3.55975 0.89937 -0.00016 AFANEB/AFANEB01: ΔEL(relax) = –776.951804 Eh C 5.34110 -0.81904 0.00030 H 5.65264 -1.86162 0.00053 N -2.28064 1.05715 -0.00001 C 6.27210 0.21053 0.00035 N -2.32968 -0.30574 -0.00008 H 7.33526 -0.02418 0.00053 C -3.87996 2.91346 -0.00001 C 5.86816 1.57121 0.00015 H -2.96019 3.50139 0.00026 H 6.63132 2.34764 0.00018 H -4.46891 3.18707 0.88252 C 4.52622 1.92613 -0.00008 H -4.46841 3.18712 -0.88285 H 4.21673 2.96888 -0.00024 C -3.54581 1.45379 0.00004 C -4.43786 0.34150 0.00010 H -5.51805 0.35465 0.00002 C -3.62652 -0.76235 0.00022 H -3.87140 -1.81213 0.00044 N 0.00002 -0.39665 -0.00011 C -1.14203 -1.07588 -0.00013 C -1.20657 -2.47763 -0.00009 H -2.14600 -3.01408 -0.00013 C -0.00006 -3.16848 -0.00004 H -0.00006 -4.25410 0.00001 C 1.20650 -2.47767 -0.00002 H 2.14589 -3.01421 0.00000 EKAKIM/EKAKIM01: ΔEL(relax) = –797.5346654 Eh C 1.14208 -1.07594 -0.00005 35 James McPherson – June 2019 8-268 Chapter 8: Appendices N 2.32970 -0.30579 -0.00005 H 1.59019 1.85354 -0.35791 N 2.28066 1.05711 -0.00011 S 5.77564 -0.75537 -0.68766 C 3.62656 -0.76236 0.00013 O 5.82601 1.01823 1.33846 H 3.87146 -1.81213 0.00026 C 6.48051 0.29479 0.63070 C 4.43788 0.34152 0.00023 C 7.98287 0.13076 0.73846 H 5.51807 0.35469 0.00024 H 8.19678 -0.62165 1.50537 C 3.54583 1.45379 -0.00004 H 8.41942 1.07980 1.05680 C 3.87992 2.91348 -0.00004 H 8.43348 -0.19777 -0.20084 H 2.96014 3.50138 -0.00105 H 4.46948 3.18695 -0.88220 H 4.46773 3.18732 0.88317

QIDJET: ΔEL(relax) = –828.145023125 Eh

N -1.76874 -1.78697 0.00006 BUKDIV: ΔEL(relax) = –1556.48288288 Eh N 0.57405 -0.31305 0.00004 N 3.21985 -1.60726 -0.00002 N -4.47983 -2.10823 0.19198 N 4.54399 -1.72566 0.00003 N -2.91226 0.13697 0.03698 N 5.11261 -0.50452 -0.00002 N -4.02990 2.64146 0.08329 N 4.16283 0.42199 -0.00012 N -0.97930 -1.23827 -0.08789 C -2.89200 -2.48348 0.00004 N -0.75633 1.11809 -0.14266 H -2.78405 -3.56860 0.00008 C -5.24720 -3.19574 0.26786 C -4.17932 -1.90830 -0.00004 H -6.31849 -3.01857 0.34626 H -5.06374 -2.53916 -0.00009 C -4.74865 -4.50169 0.25129 C -4.27389 -0.53283 -0.00007 H -5.42450 -5.34854 0.31639 H -5.24211 -0.03619 -0.00024 C -3.37060 -4.67804 0.14955 C -3.09874 0.25129 -0.00003 H -2.93691 -5.67357 0.13284 C -3.14187 1.68176 0.00000 C -2.55471 -3.55276 0.06935 H -4.11062 2.17592 0.00000 H -1.47826 -3.63331 -0.01108 C -1.98387 2.40380 0.00001 C -3.15287 -2.28471 0.09422 H -2.01379 3.49189 0.00004 C -2.30554 -1.05469 0.01016 C -0.70782 1.76413 0.00002 C -0.24190 -0.12027 -0.16206 C 0.50746 2.49515 0.00002 C -2.09300 1.19136 -0.04171 H 0.47307 3.58324 -0.00001 C -2.69384 2.56102 -0.01717 C 1.70212 1.82814 0.00003 C -1.86802 3.69158 -0.09603 H 2.65842 2.33782 0.00005 H -0.79588 3.56487 -0.17454 C 1.70982 0.39611 0.00004 C -2.45759 4.95257 -0.06968 C -0.60298 0.34129 0.00001 H -1.84454 5.84704 -0.12869 C -1.84309 -0.43507 0.00003 C -3.84379 5.04299 0.03397 C 2.99929 -0.27476 0.00001 H -4.34816 6.00381 0.05836 C -4.57885 3.85609 0.10710 H -5.66398 3.88717 0.18915 C 1.23034 -0.26218 -0.27271 C 1.82331 -1.53307 -0.28683 H 1.19072 -2.40929 -0.21133 C 3.20442 -1.66617 -0.38345 H 3.65849 -2.65179 -0.37783 C 4.01388 -0.52735 -0.48439 C 3.43008 0.74597 -0.48171 H 4.05214 1.62886 -0.56143 C 2.05023 0.87276 -0.36669 36 James McPherson – June 2019 8-269 Chapter 8: Appendices IZAFOG/IZAFOG01: ΔEL(relax) = -740.47864374 Eh N 2.28226 -2.35544 0.00068 C 3.55007 -2.73789 0.00088 O 3.54960 0.51332 0.00047 H 3.78430 -3.79417 0.00146 O -3.54960 0.51332 -0.00005 C 4.44373 -1.63713 0.00026 N 2.50903 -1.51311 0.00012 H 5.52344 -1.64732 0.00015 N 0.00000 -0.14654 0.00009 C 3.62467 -0.53518 -0.00034 N -2.50904 -1.51310 -0.00009 H 3.86672 0.51482 -0.00088 C 3.94657 -1.81113 0.00101 N 0.00009 3.27309 -0.00339 H 4.19904 -2.41302 -0.87912 C -1.25728 4.00303 0.00231 H 4.19833 -2.40920 0.88401 H -1.86513 3.77567 -0.88339 C 4.64934 -0.42759 -0.00171 H -1.05020 5.07319 -0.00077 H 5.25998 -0.24258 0.88688 H -1.85666 3.77861 0.89468 H 5.25511 -0.24409 -0.89402 C 1.25754 4.00287 0.00201 C 2.40275 -0.24465 0.00018 H 1.85686 3.77887 0.89453 C 1.14557 0.54320 0.00017 H 1.05059 5.07306 -0.00175 C 1.19933 1.94586 0.00019 H 1.86539 3.77494 -0.88353 H 2.15715 2.45074 0.00024 C 0.00001 2.65120 0.00017 H 0.00001 3.73696 0.00017 C -1.19931 1.94586 0.00010 H -2.15714 2.45075 0.00009 C -1.14556 0.54320 0.00005 C -2.40275 -0.24465 -0.00004 C -4.64935 -0.42759 -0.00025 H -5.25737 -0.24336 -0.89084 H -5.25771 -0.24333 0.89010 C -3.94657 -1.81113 -0.00012 H -4.19871 -2.41102 0.88151 H -4.19868 -2.41117 -0.88165 FEDRUE: ΔEL(relax) = –864.400166097 Eh

N 0.00002 -1.00950 -0.00000 C 1.10660 -0.26828 0.00001 N 1.19246 1.05983 0.00004 C -0.00001 1.69574 0.00007 N -1.19248 1.05981 0.00006 C -1.10659 -0.26830 0.00004 N 2.33496 -0.94822 -0.00001 N 2.39772 -2.31167 -0.00005 C 3.69213 -2.57712 -0.00002 H 4.02107 -3.60800 -0.00004 C 4.48956 -1.39769 0.00003 EKAJUX: ΔEL(relax) = –832.281429214 Eh H 5.56640 -1.31991 0.00001 C 3.58465 -0.36987 0.00003 N -0.00003 -0.92542 -0.00039 H 3.69592 0.70110 0.00011 C -1.13119 -0.22851 -0.00069 N -2.33493 -0.94826 -0.00001 C -1.20548 1.16442 -0.00130 N -2.39767 -2.31171 -0.00010 H -2.15687 1.67210 -0.00138 C -3.69207 -2.57718 -0.00004 C 0.00002 1.89800 -0.00195 H -4.02100 -3.60807 -0.00003 C 1.20550 1.16437 -0.00121 C -4.48952 -1.39776 0.00011 H 2.15690 1.67205 -0.00099 H -5.56636 -1.32001 0.00021 C 1.13117 -0.22855 -0.00067 C -3.58463 -0.36993 -0.00003 N -2.33024 -0.99552 -0.00016 H -3.69590 0.70104 -0.00002 N -2.28238 -2.35535 0.00060 N -0.00004 3.05406 0.00010 C -3.55021 -2.73776 0.00089 C 1.23160 3.82785 -0.00005 H -3.78447 -3.79403 0.00152 H 1.27451 4.46995 0.88850 C -4.44381 -1.63696 0.00030 H 1.27416 4.47012 -0.88848 H -5.52352 -1.64708 0.00028 H 2.08549 3.15605 -0.00027 C -3.62470 -0.53504 -0.00043 C -1.23172 3.82779 -0.00009 H -3.86673 0.51497 -0.00093 H -2.08557 3.15594 0.00004 N 2.33018 -0.99560 -0.00010 H -1.27448 4.46976 -0.88872 37 James McPherson – June 2019 8-270 Chapter 8: Appendices H -1.27451 4.47014 0.88826 N -2.58823 -1.36279 -0.00024 N 2.47550 -1.47529 -0.00012 C -3.79365 -1.93638 0.00002 C -4.99544 -1.22775 0.00032 C -4.92835 0.16502 0.00027 C -3.68098 0.77852 0.00001 C -2.52418 -0.02082 -0.00017 C -1.16953 0.58355 -0.00023 C -0.89605 1.93343 0.00009 C 1.05030 0.45369 -0.00030 H -3.79900 -3.02527 -0.00005 H -5.94635 -1.75046 0.00042 H -5.83327 0.76570 0.00039 UCIGUJ: ΔEL(relax) = –1684.66554780 Eh H -3.61017 1.86081 -0.00022 H -1.58314 2.76645 0.00098 S 1.67885 2.21762 -0.88976 N -0.05614 -0.22539 -0.00035 S -1.67903 2.21755 -0.88937 S 0.80281 2.21570 -0.00001 N 2.45874 -2.49098 0.21028 C 4.88157 -1.32047 0.00025 N -0.00007 -1.31180 -0.00633 C 4.80784 0.07074 0.00025 N -2.45899 -2.49079 0.21017 C 3.55552 0.67704 0.00006 N 2.57233 3.45567 1.55360 C 2.40819 -0.13295 -0.00018 N -2.57115 3.45597 1.55431 C 3.68353 -2.03939 0.00006 C 3.63521 -3.10242 0.32728 H 3.47465 1.75947 0.00008 H 3.59109 -4.17315 0.53434 H 5.70893 0.67666 0.00041 C 4.87289 -2.46856 0.20552 H 5.83538 -1.83772 0.00038 H 5.79967 -3.02457 0.31395 H 3.69747 -3.12799 0.00008 C 4.85914 -1.09871 -0.05360 H 5.78764 -0.54169 -0.15420 C 3.64194 -0.43996 -0.17767 H 3.62487 0.62240 -0.37494 C 2.43565 -1.16241 -0.04613 C 1.10028 -0.54136 -0.16004 C 0.71101 0.80096 -0.42677 C -0.71096 0.80104 -0.42674 C -1.10037 -0.54124 -0.16011 C -2.43580 -1.16223 -0.04626 C -3.64204 -0.43970 -0.17773 H -3.62490 0.62268 -0.37494 C -4.85928 -1.09834 -0.05368 ZAJGOG: ΔEL(relax) = –720.396353421 Eh H -5.78775 -0.54124 -0.15425 C -4.87315 -2.46821 0.20543 N 0.00753 -0.19577 0.00003 H -5.79998 -3.02415 0.31385 N -2.57289 -1.25629 0.00009 C -3.63551 -3.10215 0.32716 C -2.42538 0.08103 0.00003 H -3.59147 -4.17288 0.53421 N 1.08592 0.62329 -0.00003 C 2.16867 2.92352 0.59587 C -0.64558 1.98520 0.00012 C -2.16806 2.92369 0.59642 C 0.72276 1.94953 0.00016 C -1.04903 0.61566 0.00000 H 1.44325 2.75096 0.00013 C 2.40278 0.08765 -0.00004 N 2.50179 -1.23862 0.00000 C 3.72696 -1.77135 0.00001 C 4.90135 -1.02099 0.00004 C 4.78036 0.36869 -0.00008 C 3.51516 0.94458 -0.00015 H 3.40873 2.02207 -0.00028 H 5.66138 1.00334 -0.00012 H 5.87078 -1.50720 0.00018 H 3.76597 -2.85890 0.00008 PUBCAQ: ΔEL(relax) = –1063.246660590 Eh H -1.26118 2.87167 0.00027 C -3.52826 0.95163 -0.00004 38 James McPherson – June 2019 8-271 Chapter 8: Appendices C -4.81330 0.41811 -0.00011 C -4.96737 -0.96661 -0.00009 C -3.81148 -1.75064 0.00007 H -3.88755 -2.83707 0.00009 H -5.94910 -1.42907 -0.00023 H -5.67881 1.07446 -0.00021 H -3.38363 2.02645 -0.00004

39 James McPherson – June 2019 8-272 Chapter 8: Appendices

APPENDIX 2 – SUPPORTING INFORMATION FOR CHAPTER 3

James McPherson – June 2019 8-273 Chapter 8: Appendices Predictable Substituent Control of CoIII/II Redox Potential and Spin Crossover in Bis(dipyridylpyrrolide)cobalt Complexes James N. McPherson,a Ross W. Hogue,b Folaranmi Sunday Akogun,b Luca Bondì,b Ena T. Luis,a Jason R. Price,c Anna L. Garden,b Sally Brooker*b,d and Stephen B. Colbran*a,e a) School of Chemistry, The University of New South Wales, NSW 2052, Australia b) Department of Chemistry and MacDiarmid Institute for Advanced Materials and Nanotechnology, University of Otago, P.O. Box 56, Dunedin 9054, New Zealand c) ANSTO, Australian Synchrotron, Clayton, VIC, Australia d) email: [email protected] e) email: [email protected]

SUPPORTING INFORMATION

Contents 1. General Methods ...... 8-275 2. Preparations of the Cobalt Complexes ...... 8-275 3. Vacuum Phase Ligand Geometry Optimisations by DFT ...... 8-277 4. Single Crystal X-ray Diffraction (SC-XRD) Studies ...... 8-278 5. Electronic Spectra ...... 8-287 6. Magnetic Susceptibility Studies ...... 8-288 7. Correlations between Ligand Electronic Effects and SCO Activity ...... 8-290 8. NMR Studies ...... 8-292 9. Additional Electrochemistry ...... 8-307

III II 10. DFT Study of Co /Co redox processes in [Co(dpp)2] complexes ...... 8-308

III II + 11. Fit of E1/2(Ru /Ru ) vs Swain-Lupton parameters for [Ru(dpp)(tpy)] complexes ...... 8-316 12. References ...... 8-317 13. Coordinates for DFT Optimised Geometries ...... 8-319

James McPherson – June 2019 8-274 Chapter 8: Appendices 1. General Methods All NMR spectra were recorded on Bruker Avance III 300, 400, 500 and 600 spectrometers operating at 300, 400, 500 and 600 MHz frequency for 1H NMR experiments respectively. All chemical shifts were calibrated against residual solvent signals.1 NMR spectra were processed with MestReNova 11.0.1 software. Signals in NMR spectra are reported as very broad (vbr), broad (br), singlets (s), doublets (d), triplets (t), quartets (q) or multiplets. All coupling constants (J) are reported in Hertz, and for broad, paramagnetic signals, the linewidth at half height is reported in Hertz. UV-Vis experiments were performed on a Varian Cary 60 spectrophotometer, and broad peaks in the NIR region were fit to Gaussian functions using OriginPro 2016, OriginLab Corporation. Mass spectra were run on a Thermo Fisher Scientific Orbitrap LTQ XL ion trap mass spectrometer using a nanospray ionisation source. TGA were recorded on a TA instruments Q50 with the samples in a platinum pan heated at 5°C min−1. Nitrogen gas flow was set to 20 mL min-1 over the balance and 20 mL min−1 over the sample. Microanalysis was performed by the Campbell Microanalytical Laboratory at the University of Otago.

2. Preparations of the Cobalt Complexes

Figure S1. Preparations of the cobalt(II) complexes studied in this work and the numbering scheme used.

General Procedure, and Characterization data for [Co(dpp)2] Complexes

R1,R2 Hdpp (approx. 0.5 mmol) was combined with NEt3 (0.3 mL, 2 mmol) in dry, deaerated MeOH (10 mL) and stirred at reflux for 15 min. CoCl2.6H2O (approx. 0.25 mmol) was added, and the mixture was stirred at R1,R2 reflux for 2 h. After cooling to room temperature, the crude product, [Co(dpp )2], was collected by filtration, and washed with cold MeOH three times. Yields were calculated against CoCl2·6H2O as follows, with the characterisation data for each complex.

CN,Ph 1 [Co(dpp )2]. Yield: 182.9 mg, 86 % as a dark red/brown solid. H NMR (500 MHz, benzene-d6) δ 77.7 (brs, 46 Hz, 1H, B4), 76.0 (brs, 56 Hz, 1H, C4), 43.1 (brs, 31 Hz, 2H, H2Ph), 41.3 (vbrs, 354 Hz, 1H, B6/C6), 32.1 (vbrs, 289 Hz, 1H, C6/B6), 22.5 (d, J = 7.5 Hz, 2H, H3Ph), 17.1 (t, J = 7.5 Hz, 1H, H4Ph), 12.8 (brs, 19 Hz, 1H, B5/C5), 10.8 (brs, 18 Hz, 1H, C5/B5), –2.0 (brs, 18 Hz, 1H, B4), –4.9 (brs, 17 Hz, 1H, C4). 4 4 4 4 2 UV-vis (CH2Cl2): λmax (ɛmax): 259 (2.8 x10 ), 301 (3.2 x10 ), 332 (3.6 x10 ), 374 (2.4 x10 ), 505 (2.3 x 10 ), –1 –1 651 (~ 40), 781 (~ 20) 924 nm (~30 L mol cm ). (+)-HR-ESI-MS found: m/z 701.1609 (calc. [C42H26N8Co (M)]+: m/z 701.1607), m/z 702.1683 (calc. [M + H]+: m/z 702.1685), m/z 724.1505 (calc. [M + Na]+: m/z

James McPherson – June 2019 8-275 Chapter 8: Appendices 724.1505). Anal. Calc. for the hydrate C42H26N8Co·¾H2O C: 70.54, H: 3.88, N: 15.67 %. Found: C: 70.55, H: 3.76, N: 15.62 %.

CO Et,CO Et 2 1 [Co(dpp 2 2 )2]. Yield 86.5 mg, 74 % as dark red/brown needles. M.p. 218.8°C (lit. 216°C). H NMR

(400 MHz, benzene-d6) δ 63.5 (brs, 39 Hz, 1H, B3), 51.6 (vbrs, 284 Hz, 1H, B6), 17.1 (q, J = 7 Hz, 2H, - OCH2CH3), 16.8 (brs, 15 Hz, 1H, B5), 10.5 (t, J = 7 Hz, 3H, -OCH2CH3), 1.2 (brs, 13 Hz, 1H, B4). UV-vis 4 4 4 –1 –1 (CH2Cl2): λmax (ɛmax): 259 (3.0 x10 ), 291 (3.2 x10 ), 335 (4.9 x10 ), 960 nm (~40 L mol cm ). (+)-HR-ESI- + + MS found: m/z 787.1910 (calc. [C40H36N6O8Co (M)] : m/z 787.1921), m/z 810.1806 (calc. [M + Na] : m/z + 810.1819) m/z 1597.3721 (calc. [2M + Na] : m/z 1597.3745). Anal. Calc. for the hydrate C40H36N6O8Co·⅓H2O C: 60.54, H: 4.66, N: 10.59%. Found: C: 60.24, H: 4.64, N: 10.94 %. TGA: 0.4% mass loss, calc. for ⅓H2O = 0.7%.

CO Me,4-Py 1 [Co(dpp 2 )2]. Yield: 186.6 mg, 88 % as a dark red/brown solid. H NMR (400 MHz, chloroform-d) δ 68.9 (brs, 46 Hz, 1H), 59.7 (vbrs, 322 Hz, 1H), 58.3 (brs, 45 Hz, 1H), 47.4 (vbrs, 287 Hz, 1H), 28.9 (brs, 45 Hz, 2H), 19.1 (brs, 20 Hz, 1H), 17.7 (brs, 16 Hz, 2H), 17.2 (brs, 21 Hz, 1H), 16.9 (brs, 14 Hz, 3H, -OCH3), 4 4 4.0 (brs, 19 Hz, 1H), -0.2 (brs, 18 Hz, 1H). UV-vis (CH2Cl2): λmax (ɛmax): 273 (3.0 x10 ), 299 (2.6 x10 ), 330 (3.6 x104), 380 (2.1 x104), 525 (7.2 x102), 802 (~40), ~955 nm (~50 L mol–1 cm–1). (+)-HR-ESI-MS found: m/z + + 769.1709 (calc. [C42H30N8O4Co (M)] : m/z 769.1717), m/z 770.1779 (calc. [M + H] : m/z 770.1795), m/z 2+ 385.5934 (calc. [M + H] : m/z 385.5934). Anal. Calc. for the hydrate C42H30CoN8O4·H2O C: 64.04, H: 4.09, N: 14.23 %. Found: C: 64.17, H: 4.06, N: 14.19 %.

CO Me,Me 1 [Co(dpp 2 )2]. Yield: 130.5 mg, 77 % as a dark red/brown solid. M.p. 350.5°C. H NMR (500 MHz,

benzene-d6) δ 62.6 (brs, 43 Hz, 1H, B3), 62.3 (vbrs, 330 Hz, 1H, B6/C6), 50.5 (brs, 35 Hz, 1H, C3), 45.1 (vbrs, 268 Hz, 1H, C6/B6), 30.9 (brs, 16 Hz, 3H, Me/-CO2Me), 19.7 (brs, 17 Hz, 1H, C5), 17.2 (brs, 21 Hz, 1H, B5), 14.2 (brs, 8 Hz, 3H, -CO2Me/Me), 5.6 (brs, 12 Hz, 1H, B4), 0.8 (brs, 12 Hz, 1H, C4). UV-vis (CH2Cl2): 4 4 4 4 –1 –1 λmax (ɛmax): 247 (3.7 x10 ), 305 (5.5 x10 ), 335 (6.1 x10 ), 387 (4.2 x10 ), 962 nm (~50 L mol cm ). (+)-HR- + ESI-MS found: m/z 643.1484 (calc. [C34H28N6O4Co (M)] : m/z 643.1499). Anal. Calc. for the hydrate C34H28N6O4Co·¼H2O C: 63.01, H: 4.43, N: 12.97 %. Found: C: 62.66, H: 4.34, N: 13.27 %. TGA found 0.5%, calc. for ¼H2O = 0.7%.

COOMe,Me When the above reaction was performed in air, [Co(dpp )2] was obtained in 68 % yield.

Me,Me 1 [Co(dpp )2]. Yield: 80.3 mg, 71 % as a dark red/brown solid. M.p. 336.8°C. H NMR (500 MHz, acetone-d6) δ 75.7 (brs, 51 Hz, 1H, B3), 71.6 (brs, 36 Hz, 3H, Me), 31.1 (vbrs, 272 Hz, 1H, B6), 11.7 (brs, 15 4 4 Hz, 1H, B5), –3.5 (brs, 15 Hz, 1H, B4). UV-vis (CH2Cl2): λmax (ɛmax): 246 (2.0 x 10 ), 303 (sh), 328 (3.2 x10 ), 417 (3.3 x104), 676 (~50), ~928 nm (~30 L mol–1 cm–1). (+)-HR-ESI-MS found: m/z 555.1695 (calc. + [C32H28N6Co (M)] : m/z 555.1702). Anal. Calc. for the hydrate C32H28CoN6·½H2O C: 68.08, H: 5.18, N: 14.89 %. Found: C: 68.15, H: 5.30, N: 14.92 %.

Me,Me Me,Me [Co(dpp )2][PF6]. When the above reaction is performed in air: Hdpp (41.6 mg, 0.167 mmol) and Et3N (0.3 mL, 2 mmol) were methanol (20 mL) and the pale yellow/brown solution was stirred at room temperature in air for 15 minutes, after which CoCl2·6H2O (19.4 mg, 0.0815 mmol) was added. The green/brown solution was stirred at room temperature, in air, overnight, and the diamagnetic low spin cobalt(III) species was isolated by first removing excess solvent and then redissolving in the minimum amount Me,Me of methanol. On addition of saturated K[PF6] (aq), [Co(dpp )2][PF6] (38.8 mg, 68 %) precipitates as a yellow/green solid, which was collected by filtration and washed with water three times. 1H NMR (600 MHz, acetone-d6) λ 7.78 (ddd, J = 8.1, 7.3, 1.5 Hz, 1H, B4), 7.71 (ddd, J = 8.1, 1.5, 0.8 Hz, 1H, B3), 6.87 (ddd, J = 7.3, 5.8, 1.5 Hz, 1H, B5), 6.79 (ddd, J = 5.8, 1.5, 0.8 Hz, 1H, B6), 2.72 (s, 3H, A3ʹ). 13C NMR δ 160.0 (B2),

James McPherson – June 2019 8-276 Chapter 8: Appendices 152.3 (B6), 141.3 (B4), 141.0 (A2), 126.7 (A3), 123.0 (B5), 119.6 (B3) and 10.8 (A3ʹ) ppm. 1H-15N HMBC 4 4 NMR (600 MHz, acetone-d6) δ 190.0 ppm (Npyridyl). UV-vis (CH2Cl2): λmax (ɛmax): 257 (2.7 x10 ), 295 (3.3 x10 ), 327 (2.3 x104), 350 (1.7 x104), 443 nm (2.3 x104 L mol–1 cm–1). (+)-HR-ESI-MS found: m/z 555.1700 (calc. + [C32H28N6Co (M)] : m/z 555.1702).

3. Vacuum Phase Ligand Geometry Optimisations by DFT

3 Some of us have previously shown that the strain energy, ∆EL(strain), which builds up within a tridentate pincer – ligand (including dpp ), ∆EL(relax), on chelating to a metal ion, ∆EL(coord), can be approximated by considering the associated change in the distal N···N donor separation, ∆ρ (equations 1-3). The value of ∆ρ is the

difference between the distal N···N separation observed by solid-state SC-XRD after coordination, ρL(coord), and that calculated by DFT for the “relaxed/free” deprotonated ligand in its planar syn;syn-conformation, 3 ρL(relax).

∆EL(strain) = ∆EL(coord) – ∆EL(relax) (S1)

∆ρ = ρL(coord) – ρL(relax) (S2)

∆EL(strain) = 460 × (∆ρ) / ρL(relax) (S3) The optimised geometries for each deprotonated ligand (Figure S2) were calculated using B3LYP-GD3BJ/6- 31G** level DFT calculations (with the inter ring N-C-C-N torsions constrained to 0°), in the vacuum phase.3 COOMe,Me – CN, Ph – This gave calculated ρL(relax) values in the range from 4.899 Å for (dpp ) to 5.058 Å for (dpp ) (Table 1). We can therefore predict that the complexes with the more open ligands, (dppCN,Ph)– and (dppMe,Me)– , will involve more strain within each ligand than those with shorter ρL(relax) values, given that M-N distances of the order of 1.9-2.3 Å, and intraligand Npy-M-Npy angles approaching 180°, are anticipated.

Table S1. N···N separations (Å) from vacuum phase geometry optimisations (B3LYP-GD3BJ/6-31G**) – of the five dpp ligands (ρL(relax))

1 2 R ,R CN,Ph CO2Et,CO2Et CO2Me,4-Py CO2Me,Me Me,Me ρL(relax) 5.058 4.969 4.924 4.899 5.047

Figure S2. Vacuum phase optimised (B3LYP-GD3BJ/6-31G**) geometries for, from left to right, top to CN,Ph – COOEt,COOEt – COOMe,Py – COOMe,Me – Me,Me – bottom: (dpp ) , (dpp ) , (dpp ) , (dpp ) and (dpp ) .

James McPherson – June 2019 8-277 Chapter 8: Appendices 4. Single crystal X-ray diffraction (SC-XRD) studies

HdppCO2Me,Me (CCDC 1873690)

Pale yellow hexagonal plates of the HdppCO2Me,Me ligand precursor suitable for single crystal X-ray diffraction were grown by slow vapour diffusion of pentane into a solution of the compound in benzene. The crystal of HdppCO2Me,Me, with dimensions of 0.1 x 0.3 x 0.3 mm selected under the polarizing microscope (Leica M165Z), was picked up on a MicroMount (MiTeGen, USA) consisting of a thin polymer tip with a wicking aperture. The X-ray diffraction measurements were carried out on a Bruker kappa-II CCD diffractometer at 150 K using IµS Incoatec Microfocus Source with Mo-Kα radiation (λ = 0.710723 Å). The single crystal, mounted on the goniometer using a cryo loop for intensity measurements, was coated with immersion oil type NVH and then quickly transferred to the cold nitrogen stream generated by an Oxford Cryostream 700 series. Symmetry related absorption corrections using the program SADABS4 were applied and the data were corrected for Lorentz and polarisation effects using Bruker APEX3 software.4 The structure was solved by SHELXT5 (with intrinsic phasing) and the full-matrix least-square refinements were carried out using SHELXL-20146 through Olex27 software. The non-hydrogen atoms were refined anisotropically. The N1…H2 2.375 Å and N3…H2 2.321 Å are both shorter than the sum of the van der Waals radii (Figure S3). The HdppCO2Me,Me ligand stacks in layers separated by 3.476(2) Å, with the pyrrole rings involved in offset face-face π interactions,8 as shown in Figure S4.

Figure S3. ORTEP view from SC-XRD data (50 % thermal ellipsoids at 150 K) of the HdppCO2Me,Me ligand precursor. Key inter-atomic separations (Å) and angles (°): N1···H2: 2.375, N3···H2: 2.321, N1–H2–N3: 159.4, ρ(N1···N3): 4.621(3).

James McPherson – June 2019 8-278 Chapter 8: Appendices

Figure S4. Solid state packing of HdppCO2Me,Me in layers, from the SC-XRD structure determined at 150 K, shown as 50 % thermal ellipsoids with C-bound H-atoms omitted for clarity. Each layer in (a) is separated by 3.476(2) Å, with the pyrrole rings (shaded grey) are offset as shown in (b).

James McPherson – June 2019 8-279 Chapter 8: Appendices [Co(dpp)2] Complexes

Table S2. Key bond lengths (Å) and angles (°) from SC-XRD studies, with magnetic susceptibilities at the temperature of data acquisition from variable temperature magnetic susceptibility studies (see below) also presented.

COOMe,Me COOEt,COOEt Me,Me Distances (Å) or [Co(dpp )2] [Co(dpp )2] [Co(dpp )2]·2(C5H12) Angles (°) @ 298 K @ 100 K @ 298 K @100 K @ 100 K Co1 Co2 (residue 12)‡ Co–N(py)A Co–N1A /Å 2.188(4) 2.174(3) 2.1902(17) 2.246(2) 2.171(10) 2.289(3) Co–N(py′)A Co–N3A /Å 2.171(4) 2.160(3) 2.1656(18) 2.2392(16) 2.162(10) 2.289(3) Co–N(pyr)A Co–N2A /Å 1.888(3) 1.863(3) 1.9079(15) 1.959(2) 1.863(8) 1.969(4) Co–N(py)B Co–N1B /Å 2.164(4) 2.087(4) 2.1902(17) 2.246(2) 2.068(12) 2.289(3) Co–N(py′)B Co–N3B /Å 2.165(4) 2.083(4) 2.1656(18) 2.2392(16) 2.058(11) 2.289(3) Co–N(pyr)B Co–N2B /Å 1.884(3) 1.842(3) 1.9079(15) 1.959(2) 1.844(9) 1.969(4) N1A–Co–N2A /° 75.76(15) 76.82(13) 75.41(6) 73.59(8) 76.3(4) 73.16(7) Intra-ligand N3A–Co–N2A /° 75.38(14) 75.79(13) 75.50(6) 73.94(7) 76.8(4) 73.16(7) N(py)–Co–N(pyr) N1B–Co–N2B /° 76.65(15) 78.21(15) 75.41(6) 73.59(8) 77.1(4) 73.16(7) N3B–Co–N2B /° 76.11(15) 77.53(15) 75.50(6) 73.94(7) 78.3(4) 73.16(7) N1A–Co–N1B /° 90.72(12) 89.89(12) 97.19(9) 98.41(10) 94.2(4) 94.81(4) Inter-ligand N1A–Co–N3B /° 94.92(13) 94.31(13) 91.26(7) 87.81(7) 91.6(4) 94.81(4) N(py)–Co–N(py′) N3A–Co–N1B /° 93.12(12) 92.70(13) 91.26(7) 87.81(7) 90.4(4) 94.81(4) N3A–Co–N3B /° 94.73(13) 94.49(13) 94.79(10) 104.31(8) 95.1(4) 94.81(4) N1A–Co–N2B /° 101.64(14) 99.82(14) 103.38(6) 109.73(8) 105.1(4) 106.84(7) Inter-ligand N3A–Co–N2B /° 107.29(14) 107.62(14) 105.74(6) 103.00(7) 101.8(4) 106.84(7) N(py)–Co–N(pyr) N1B–Co–N2A /° 104.63(13) 102.58(13) 103.38(6) 109.73(8) 102.4(4) 106.84(7) N3B–Co–N2A /° 102.60(13) 101.64(13) 105.74(6) 103.00(7) 102.3(4) 106.84(7) N(pyr)–Co–N(pyr′) N2A–Co–N2B /° 177.05(15) 176.50(15) 178.22(8) 175.17(8) 178.5(4) 180 N(py)–Co–N(py′) N1A–Co–N3A /° 150.92(13) 152.39(12) 150.85(6) 147.13(8) 153.1(3) 146.32(14) N1B–Co–N3B /° 152.76(14) 155.74(14) 150.85(6) 147.13(8) 155.4(4) 146.32(14)

N(py)…N(py′) ρA /Å 4.219(5) 4.209(5) 4.216(3) 4.302(3) 4.214(15) 4.382(6)

separation ρB /Å 4.208(5) 4.077(5) 4.216(3) 4.302(3) 4.033(16) 4.382(6) –1 Intra-ligand ∆EA /kJ mol 64 65 70 62 70 61 3 –1 strain ∆EB /kJ mol 65 77 70 62 87 61 3 VOh /Å 11.15 10.58 11.29 11.96 10.49 12.60

Co–Npyr /Å 1.886(3) 1.853(3) 1.9079(15) 1.959(2) 1.854(9) 1.969(4)

〈Co–Npy 〉 /Å 2.172(4) 2.126(4) 2.1779(18) 2.243(2) 2.115(12) 2.289(3)

〈 Co–N 〉 /Å 2.077(4) 2.035(4) 2.0879(15) 2.148(2) 2.028(12) 2.183(4) φ Octahedral 〈 〉 177.05(15) 176.50(15) 178.22(8) 175.17(8) 178.5(4) 180 fitting† θ 88.65(5) 88.53(4) 82.54(3) 77.20(4) 97.48(10) 90 Σ 125.8 114.9 130.9 157.5 114.4 154.0 T 0.868 0.871 0.876 0.874 0.877 0.860

λOh 1.051 1.045 1.053 1.073 1.044 1.071 Χ 〈 LS〉 0.65 0.96 0.46 0 0.95 0 † Where VOh is the volume of the octahedron defined by the six N donor atoms; φ and θ are Halcrow’s angular Jahn-Teller distortion parameters for iron(II) systems; Σ is the sum of |90 – α| for all twelve cis N–Co–N angles (α); T is the tetragonal distortion (T = Co– 9 2 10 Nax / Co–Neq ); λOh = (1/6)Σ(li / l0) is the quadratic elongation parameter proposed by Robinson et al.; and ΧLS is the LS fraction 〈 determined from Co–Npyr and equation S5 below. The fraction of LS cobalt(II) centres (ΧLS) was determined from Co–Npyr according〉 〈 to 〉eq 〈S5. 〉‡Residue 12 presented as the most stable residue during refinement of the SC-XRD data. More detail for all CO Et,CO Et residues and disordered〈 fragments〉 within the [Co(dpp 2 2 )2] structure obtained at 100 K is provided in Table S3 below.〈 〉

James McPherson – June 2019 8-280 Chapter 8: Appendices Correlation of average Co–N bond length with χMT and ΧLS As shown in Figure S5 below, correlations were observed between Co–Npyr (black), Co–Npy (blue) and Co–N (pink) and the χMT measurements on solid samples of the [Co(dpp)2] samples studied in this work. 〈 〉 〈 〉 〈 〉

Figure S5. Correlations between the Co–Npyr (pentagons, black trendline), Co–Npy (hexagons, blue trendline) and Co–N (squares, pink trendline) from the [Co(dpp)2] structures described in this work. The 〈 〉 CO Et,CO Et 〈 〉 averages for all the sixteen [Co(dpp)2] molecules in the [Co(dpp 2 2 )2] structure determined at 100 K, 〈 〉 including disorder, have been represented by error bars around the total mean for the unit cell.

The fraction of LS cobalt(II) centres (ΧLS) within a sample can be correlated directly with χMT (equation S4), and so the trends observed in Figure SX can be used to determine ΧLS directly from the average Co–N bond lengths in SC-XRD structures. In this work, the Co–Npyr data gave the tightest trend (smallest error), and so equation S2 was used to determine ΧLS from all crystallographic data (by combining the trend observed in 〈 〉 3 –1 3 –1 Figure SX with equation S5 and assuming (χMT)LS = 0.65 cm mol K and (χMT)HS = 3.0 cm mol K , see the magnetic susceptibility section).

χMT = ΧLS (χMT)LS + (1 – ΧLS) (χMT)HS (S4)

ΧLS = {1.958 – Co–Npyr } / 0.1097 (S5)

〈 〉

CO Et,CO Et [Co(dpp 2 2 )2] (CCDC 1873694 @ 298 K, CCDC 1873695 @ 100 K) CO Et,CO Et Brown blocks of [Co(dpp 2 2 )2] suitable for single crystal X-ray diffraction were grown by slow diffusion of pentane into a solution of the compound in chloroform. The crystal with dimensions 0.08 x 0.08 x 0.04 mm was coated in Paratone and transferred to the goniometer. Two data sets were collected for the crystal; the first was obtained at room temperature, after which the crystal was placed in the cryostream at 100 K for the second collection. Diffraction measurements were carried out using Si<111> monochromated synchrotron X- ray radiation (λ = 0.71073 Å) on the MX1 Beamline at the Australian Synchrotron.11 Data collection was carried out using Australian Synchrotron QEGUI software and unit cell refinement, data reduction and processing were carried out with XDS,12 with clarification of lattice parrameters of the 100K data collection with CrysAlisPro.13 The structures were solved using dual space methods with SHELXT.5 The least-squares

James McPherson – June 2019 8-281 Chapter 8: Appendices refinement was carried out with SHELXL-20146 through Olex27 and ShelXle14 graphical interfaces. In the 298 K structure the non-hydrogen atoms were refined anisotropically. The structure at room temperature crystallised in the base-centred monoclinic C2/c space group. The ethyl substituent in one of the ethyl groups, on O2B, was disordered across two positions (atoms C9 and C10 in ligand B). This disorder was modelled over two positions with occupancies of 0.63 and 0.37, and refined with similar anisotropic displacement parameters. The lower residual electron density used to assign the two positions resulted in C–O and C–C bond lengths that were too short to make chemical sense. To correct this, the C9–O2 bond lengths were restrained to be similar to the corresponding bonds in ligand A, and the C9– C10 bonds were restrained to be similar across the two disordered positions. At 100 K, the structure underwent an incomplete phase transition and was best modeled to data with a lowering of symmetry of the 298 K lattice from C-centered to primitive monoclinic, along with a commensurate quadrupulling of the b-cell axis (see Figure S6, S7 and S8). The correct unit cell parameters were obtained by indexing using CrysAlisPro.13 Structure solution was carried out using dual space methods with SHELXT5 and was explored in multiple space groups with the Pn space group offering the most rational model. Due to significant correlations between the molecules and extensive disorder, the resolution of the model is poor. The asymmetric unit contains sixteen independent complexes which were given numerical identifiers using the RESI command. Residue 12 was stable to refinement, and so was used for extensive geometric restraints on disordered molecules with all partial occupancy non-hydrogen atoms held isotropic. All disordered parts were refined freely before being fixed.

CO Et,CO Et Figure S6. The asymmetric units (and cell axes) from the SC-XRD structures of [Co(dpp 2 2 )2] at 298 K on the left and at 100 K on the right.

CO Et,CO Et The [Co(dpp 2 2 )2] molecules are more densely packed (as measured by the Co…Co separations) in the 100 K structure than in the structure obtained at 298 K. Going down the b axis in the structure obtained CO Et,CO Et at 298 K, the Co1…Co2 separations of neighbouring [Co(dpp 2 2 )2] molecules alternate between 7.0011(15) Å and 7.0759(16) Å. In the structure at 100 K, however, these distances instead alternate between 6.822(3)–6.838(3) Å and 7.181(3)–7.198(3) Å (see Fig. S7, supporting info). At 298 K, the Co1…Co2 separation between neighbouring molecules along the a axis is 13.796(3) Å, which is slightly longer than the range of separations (13.436(3)–13.622(3) Å) along the equivalent c axis in the 100 K structure. The separations along the a axis at 100 K occur in two different regimes as shown in Fig S7, either between 9.830(3)–9.983(3) Å or between 10.020(3)–10.264(3) Å, which are shorter than along the equivalent c axis in the 298 K structure (11.5980(19) Å).

James McPherson – June 2019 8-282 Chapter 8: Appendices

l l

k k 298 K 100 K

l l

h h 298 K 100 K

k k

h h 298 K 100 K

Figure S7. Comparison of the reciprocal space precession images from SC-XRD data collected on a single CO Et,CO Et crystal of the [Co(dpp 2 2 )2] complex at 298 K shown on the left, which was cooled to 100 K shown on the right. Quadrupling of the b axis (k in reciprocal space) was observed (reflections closer together in k) during the phase transition.13

James McPherson – June 2019 8-283 Chapter 8: Appendices CO Et,CO Et Table S3. Co–N bond lengths and LS fraction (ΧLS) for all 16 residues in the [Co(dpp 2 2 )2] structure determined at 100 K.

Residue: Whole Cell: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1.93– 2.010– 2.041– 2.096– 2.048– 2.046– 2.003– 1.930– 2.010– 1.990– 1.990– 2.090– 2.058– 2.057– 2.030– 2.047– 2.050– Co–Npy range /Å 2.42 2.300 2.266 2.128 2.420 2.241 2.310 2.280 2.270 2.309 2.360 2.173 2.171 2.233 2.294 2.223 2.201 Co–Npy /Å 2.13 2.135 2.129 2.105 2.132 2.140 2.144 2.130 2.137 2.142 2.142 2.126 2.115 2.135 2.133 2.130 2.117 1.74– 1.740– 1.834– 1.837– 1.810– 1.855– 1.777– 1.828– 1.837– 1.777– 1.838– 1.800– 1.844– 1.821– 1.790– 1.837– 1.836– Co–Npyr range /Å 〈 〉 2.01 2.010 1.893 1.879 1.871 1.890 1.933 1.883 1.874 1.940 1.920 1.897 1.863 1.897 1.960 1.908 1.910 Co–Npyr /Å 1.86 1.863 1.864 1.858 1.841 1.873 1.865 1.851 1.852 1.865 1.863 1.846 1.854 1.859 1.879 1.873 1.873

Co–N /Å 2.04 2.045 2.041 2.023 2.035 2.051 2.051 2.037 2.042 2.050 2.048 2.033 2.028 2.043 2.048 2.044 2.035 〈 〉 ΧLS 0.88 0.87 0.86 0.91 1 0.78 0.85 0.97 0.96 0.84 0.86 1 0.95 0.90 0.72 0.78 0.77 〈 〉

Figure S8. Schematic arrangement of the different residues and intermolecular Co…Co separations that repeat in the SC-XRD structures of CO Et,CO Et [Co(dpp 2 2 )2] determined at 298 K (top) and 100 K (bottom).

James McPherson – June 2019 8-284 Chapter 8: Appendices CO Me,Me [Co(dpp 2 )2] (CCDC 1873692 @ 298 K, CCDC 1873693 @ 100 K) CO Me,Me Red-brown plates of [Co(dpp 2 )2] suitable for single crystal X-ray diffraction were grown by slow diffusion of pentane into a solution of the compound in benzene. The crystal with dimensions 0.03 x 0.02 x 0.01 mm was coated in Paratone and transferred to the goniometer. Two data sets were collected for the crystal; the first was obtained at room temperature, after which the crystal was placed in the cryostream at 100 K for the second collection. Diffraction measurements were carried out using Si<111> monochromated synchrotron X-ray radiation (λ = 0.71073 Å) on the MX1 Beamline at the Australian Synchrotron.11 Data collection was carried out using Australian Synchrotron QEGUI software and unit cell refinement, data reduction and processing were carried out with XDS.12 The structure was solved using dual space methods with SHELXT.5 The least-squares refinement was carried out with SHELXL-20146 through Olex27 software. The non-hydrogen atoms were refined anisotropically.

Me,Me [Co(dpp )2]·2(C5H12) (CCDC 1873691) Me,Me Red-brown plates of [Co(dpp )2] suitable for single crystal X-ray diffraction were grown by slow diffusion of pentane into a solution of the compound in chloroform. The crystal with dimensions 0.03 x 0.03 x 0.01 mm was coated in Paratone and transferred to the goniometer at 100 K. Diffraction measurements were carried out using Si<111> monochromated synchrotron X-ray radiation (λ = 0.71073 Å) on the MX1 Beamline at the Australian Synchrotron.11 Data collection was carried out using Australian Synchrotron QEGUI software and unit cell refinement, data reduction and processing were carried out with XDS.12 The structure was solved using dual space methods with SHELXT.5 The least-squares refinement was carried out with SHELXL-20146 through Olex27 software. The non-hydrogen atoms of the complex were refined anisotropically. The complex crystallised in a highly symmetric space group, forming solvent-accessible square channels (7.996(1) Å across) along one dimension of the lattice which accounted for 37% of the unit cell volume (see Figure S8). Individual solvent molecules could not be identified from the difference map; however, residual electron density as well as the shape of the channels indicated the likelihood of low-occupancy pentane molecules filling the space. One molecule of pentane in the asymmetric unit was modelled isotropically using Ilia’s fragment library with occupancy of 0.25, which lowered the R1 from 9.04 % to 7.55 %.

Me,Me Figure S9. Two views of the molecular packing of [Co(dpp )2]·2(C5H12) from SC-XRD data collected at Me,Me 100 K, down the (left) b axis and (right) c axis. The atoms in the [Co(dpp )2] complex are shown as thermal ellipsoids (50 %), while the volumes occupied by disordered pentane C (dark grey) and H (light grey) atoms are represented by spacefill models.

James McPherson – June 2019 8-285 Chapter 8: Appendices Table S4. X-ray crystallographic data and refinement parameters for structures.

CO2Et,CO2Et CO2Et,CO2Et CO2Me,Me CO2Me,Me Me,Me CO2Me,Me [Co(dpp )2] @ 298 K [Co(dpp )2] @ 100 K [Co(dpp )2] @ 298 K [Co(dpp )2] @ 100 K [Co(dpp )2] Hdpp CCDC 1873694 CCDC 1873695 CCDC 1873692 CCDC 1873693 CCDC 1873691 CCDC 1873690

Empirical formula C40H36CoN6O8 C40H36CoN6O8 C34H28CoN6O4 C34H28CoN6O4 C42H52CoN6 C17H15N3O2 Formula weight 787.68 787.68 643.55 643.55 699.82 293.32 Temperature/K 298 100 298 100 100 150.01 Crystal system monoclinic Monoclinic monoclinic monoclinic tetragonal Triclinic

Space group C2/c Pn P21/c P21/c I41/amd P-1 a/Å 27.592(6) 19.890(4) 9.3630(19) 9.1960(18) 15.992(2) 9.1032(11) b/Å 14.077(3) 56.061(11) 18.802(4) 18.756(4) 15.992(2) 9.98667(13) c/Å 20.130(4) 27.055(5) 17.145(3) 16.840(3) 14.261(3) 9.9600(12) α/° 90 90 90 90 90 62.551(3) β/° 104.38(3) 105.95(3) 93.00(3) 93.13(3) 90 63.555(3) γ/° 90 90 90 90 90 83.562(4) Volume/Å3 7574(3) 29006(11) 3014.1(11) 2900.2(10) 3647.3(13) 705.91(15) Z 8 32 4 4 4 2

3 ρcalcg/cm 1.382 1.443 1.418 1.474 1.274 1.380 μ/mm-1 0.514 0.537 0.619 0.644 0.509 0.093 F(000) 3272 13088.0 1332 1332 1492 308.0 Crystal size/mm3 0.08 × 0.08 × 0.04 0.08 × 0.08 × 0.04 0.03 × 0.02 × 0.01 0.03 × 0.02 × 0.01 0.04 × 0.02 × 0.01 0.3 × 0.3 × 0.1 Radiation MoKα (λ = 0.71073) MoKα (λ = 0.71073) MoKα (λ = 0.71073) MoKα (λ = 0.71073) MoKα (λ = 0.71073) MoKα (λ = 0.71073) 2Θ range for data collection/° 3.27 to 63.656 1.452 to 55 4.758 to 63.982 4.844 to 63.384 3.826 to 63.194 4.684 to 54.424 -39 ≤ h ≤ 39, -18 ≤ k ≤ 18, -24 -24 ≤ h ≤ 24, -72 ≤ k ≤ 72, - -13 ≤ h ≤ 13, -26 ≤ k ≤ 25, - -13 ≤ h ≤ 13, -26 ≤ k ≤ 26, - -23 ≤ h ≤ 23, -23 ≤ k ≤ 23, -11 ≤ h ≤ 11, -12 ≤ k ≤ 12, Index ranges ≤ l ≤ 24 35 ≤ l ≤ 35 23 ≤ l ≤ 23 23 ≤ l ≤ 23 -20 ≤ l ≤ 20 -12 ≤ l ≤ 12 Reflections collected 63200 474504 49669 47836 30226 19671

9474 [Rint = 0.0283, Rsigma = 125709 [Rint = 0.0479, 7962 [Rint = 0.0490, Rsigma = 7526 [Rint = 0.0343, Rsigma = 1495 [Rint = 0.0479, 3145 [Rint = 0.0910, Independent reflections 0.0170] Rsigma = 0.0518] 0.0296] 0.0194] Rsigma = 0.0161] Rsigma = 0.0654] Data/restraints/parameters 9474/20/521 125709/20578/7559 7962/0/410 7526/1/410 1495/0/72 3145/0/201 Goodness-of-fit on F2 1.048 1.347 1.039 1.055 1.224 1.052

Final R indexes [I>=2σ (I)] R1 = 0.0496, wR2 = 0.1474 R1 = 0.0961, wR2 = 0.3850 R1 = 0.0814, wR2 = 0.2303 R1 = 0.0818, wR2 = 0.2125 R1 = 0.0755, wR2 = 0.1898 R1 = 0.0512, wR2 = 0.1065

Final R indexes [all data] R1 = 0.0572, wR2 = 0.1549 R1 = 0.1705, wR2 = 0.4468 R1 = 0.1197, wR2 = 0.2650 R1 = 0.0947, wR2 = 0.2230 R1 = 0.0797, wR2 = 0.1921 R1 = 0.0909, wR2 = 0.1251 Largest diff. peak/hole / e Å-3 0.46/-0.40 0.05/-0.07 0.55/-0.76 0.73/-1.36 0.78/-0.59 0.19/-0.23

James McPherson – June 2019 8-286 Chapter 8: Appendices

5. Electronic Spectra

Figure S10. UV-vis-NIR spectra of dilute [Co(dpp)2] complexes.

Figure S11. Vis-NIR spectra of concentrated [Co(dpp)2] complexes.

James McPherson – June 2019 8-287 Chapter 8: Appendices

Figure S12. Fitting (shown in blue) of the NIR regions from electronic spectra (CH2Cl2 at 298 K) of the [Co(dpp)2] complexes studied.

6. Magnetic Susceptibility Studies

Figure S13. Left: The solid state magnetic data for [Co(dpp)2] complexes over two full thermal cycles each: cooling from 300 K to 50 K (blue), then heating to 400 K (red), cooling back to 50 K for a second cycle (in blue) and heating up to 300 K (red). Right: comparisons of the curves of the variable temperature magnetic R1,R2 susceptibility data in CDCl3 solution (crosses) and solid state (dots) for the [Co(dpp )2] complexes.

COOEt,COOEt COOMe,Me Fitting CDCl3 solution data to Eq. 1 and 2 for [Co(dpp )2] and [Co(dpp )2]

The spin equilibrium follows where xHS is the mole fraction of the HS state:

James McPherson – June 2019 8-288 Chapter 8: Appendices [ ] [ ]

𝐾𝐾 1 𝐻𝐻𝐻𝐻 𝐿𝐿𝐿𝐿 ⇌ The thermodynamic expression of the equilibrium𝑥𝑥𝐻𝐻𝐻𝐻 constant− 𝑥𝑥is:𝐻𝐻𝐻𝐻

= exp ( ) ∆𝐻𝐻 ∆𝑆𝑆 𝐾𝐾 − So the equilibrium constant for this is: 𝑅𝑅𝑅𝑅 𝑅𝑅 = → = → = ( ) 1−𝑥𝑥𝐻𝐻𝐻𝐻 1 1 ∆𝐻𝐻 ∆𝑆𝑆 𝐻𝐻𝐻𝐻 𝐻𝐻𝐻𝐻 𝐻𝐻𝐻𝐻 𝐾𝐾 𝑥𝑥 𝑥𝑥 𝐾𝐾+1 𝑥𝑥 1+exp 𝑅𝑅𝑅𝑅− 𝑅𝑅 The χMT product can be expressed by:

= ( ) + (1 )( )

𝑀𝑀 𝐻𝐻𝐻𝐻 𝑀𝑀 𝐻𝐻𝐻𝐻 𝐻𝐻𝐻𝐻 𝑀𝑀 𝐿𝐿𝐿𝐿 Where (χMT)HS and (χMT)LS are the𝜒𝜒 magnetic𝑇𝑇 𝑥𝑥 susceptibilities𝜒𝜒 𝑇𝑇 − of 𝑥𝑥 the 𝜒𝜒[HS𝑇𝑇-HS] and [HS-LS] states respectively. Combined with expressions for K:

( ) ( ) = + ( ) 1𝜒𝜒𝑀𝑀 +𝑇𝑇exp𝐻𝐻𝐻𝐻 (− 𝜒𝜒𝑀𝑀𝑇𝑇 𝐿𝐿)𝐿𝐿 𝜒𝜒𝑀𝑀𝑇𝑇 𝜒𝜒𝑀𝑀𝑇𝑇 𝐿𝐿𝐿𝐿 ∆𝐻𝐻 ∆𝑆𝑆 − Least squares fitting couldn’t converge to this equation,𝑅𝑅𝑅𝑅 𝑅𝑅given 11 data points and 4 parameters [∆H, ∆S, (χMT)HS and (χMT)LS]. From the solid state magnetic data, the maximum ΧMT is 3.0, so it is sensible to fix 3 -1 (χMT)HS at 3 cm K mol to improve the data fitting, i.e. equation S6.

( ) = + ( ) (S6) ( ) 3− 𝜒𝜒𝑀𝑀𝑇𝑇 𝐿𝐿𝐿𝐿 𝑀𝑀 ∆𝐻𝐻 ∆𝑆𝑆 𝑀𝑀 𝐿𝐿𝐿𝐿 𝜒𝜒 𝑇𝑇 1+exp 𝑅𝑅𝑅𝑅− 𝑅𝑅 𝜒𝜒 𝑇𝑇 Fitting to this equation we get sensible values for ∆H and ∆S (see table 1), although large errors, and hence even larger errors in predicted T½ (= ∆H/∆S). This model leaves (χMT)LS as a free variable. For COOEt,COOEt 3 -1 [Co(dpp )2] the best fit of (χMT)LS is 1.74 cm K mol which suggests possibly only an approximate 3 -1 50% conversion to LS in solution at the low temperature limit (χMT would be ≈1.8 cm K mol for xHS = 0.5). COOMe,Me 3 -1 For [Co(dpp )2] the best fit of (χMT)LS is 1.05 cm K mol which is more consistent with an almost complete transition to LS at the low temperature limit.

3 -1 Alternatively we can also fix (χMT)LS to 0.65 cm K mol , based on the lower limit of χMT achieved for these two complexes in solid state, and use equation S7:

. = + 0.65 (S7) ( ) 2 35 𝑀𝑀 ∆𝐻𝐻 ∆𝑆𝑆 𝜒𝜒 𝑇𝑇 1+exp 𝑅𝑅𝑅𝑅− 𝑅𝑅 Unsurprisingly, this greatly improves the accuracy of the parameters ∆H and ∆S, and hence T½. The COOMe,Me [Co(dpp )2] complex clearly has the higher T½ which agrees with it being more LS than COOEt,COOEt [Co(dpp )2] at all measured temperatures.

James McPherson – June 2019 8-289 Chapter 8: Appendices 7. Correlations between Ligand Electronic Effects and SCO Activity

Initially, the observed χMT value at 298 K for each complex was used as indicative of the relative populations of HS:LS at this (ambient) temperature. None of the correlations with the Hammett parameters (σM, σp or + σp ), as suggested by Deeth, Halcrow and co-workers, predicted the magnetic responses of these cobalt(II) complexes in solution (e.g., see Figure S13). This is not surprising as the Hammett parameters are for substituents to six-membered rings (and we note that the weakest correlation of r2 = 0.61 reported by Deeth and Halcrow was for σM as a predictor of substituent effects on five-membered pyrazolyl rings). Hence we turned to the field (F) and resonance (R) parameters proposed by Swain and Lupton, as these provide a more general measure of substituent effects,15 so are more appropriate for these five-membered pyrrolide systems. The resulting plot (Figure S14) of χMT values (at 298 K) for the three SCO-active [Co(dpp)2] complexes versus the sum of the modified Swain-Lupton parameters (reported by Hansch et al.16) for the pyrrolide R1 and R2 substituents (assuming the contributions from the 2-pyridyl rings to be constant), CN,Ph incorrectly predicts that [Co(dpp )2] should be SCO active—with a relative LS population {(χMT)predicted ≈ 3 –1 CO2Me,Me 3 –1 2.0 cm K mol at 298 K}, between that of [Co(dpp )2] (χMT = 1.75 cm K mol at 298 K) and CO2Et,CO2Et 3 –1 3 [Co(dpp )2] (χMT = 2.3 cm K mol at 298 K)—when in fact it remained fully HS (χMT = 2.8 cm K mol–1 at 298 K). Slattery and co-workers studied a range of SCO-active homoleptic tpy cobalt(II) complexes 17 and likewise found little correlation between the inductive influence of the 4′-substituents on the T1/2.

Figure S14. Switching temperatures calculated from solution state data (a–c), or χMT at 298 K (d–f) for R1,R2 + 16, 18 1 2 [Co(dpp )2] complexes against: Hammett σ constants, from left to right: σM, σp and σp . R , R are: CN, Ph (black); CO2Et, CO2Et (orange); CO2Me, 4-Py (purple); or CO2Me, Me (green). Solution state data Me,Me were not collected for [Co(dpp )2] however it was observed at HS for all T (50–400 K) in the solid state 3 –1 measurements (with χMT = 2.9 cm K mol at 298 K), and is therefore represented as half-filled red symbols.

James McPherson – June 2019 8-290 Chapter 8: Appendices

Figure S15. The magnetic susceptibilities (χMT in solution at 298 K) of the three [Co(dpp)2] complexes which were SCO active in the solution state were fitted to the modified16 Swain-Lupton15 parameters of the pyrrolide Me,Me substituents, as shown in blue. Also plotted are the χMT values in solution (except for [Co(dpp )2] which was only measured in the solid state, shown in red) at 298 K for all complexes studied.

James McPherson – June 2019 8-291 Chapter 8: Appendices 8. NMR Studies

1 Table S5. Summary of the H chemical shifts (δ /ppm) and line widths at half height (in parentheses /Hz) for paramagnetic [Co(dpp)2] NMR experiments, ordered from highest line width to lowest. Unless otherwise noted, conditions: 298 K, in deuterated benzene at 500 MHz.

CN,Ph CO Et,CO Et [a] CO Me,Py [b] CO Me,Me Me,Me [c] Me,Me [d] [Co(dpp )2] [Co(dpp 2 2 )2] [Co(dpp 2 )2] [Co(dpp 2 )2] [Co(dpp )2] [Co(dpp )2]

41.3 (354) 32.1 (289) 51.6 (284) 59.7 (322) 47.4 (287) 62.3 (330) 45.1 (268) 31.1 (272) 32.0 (395) δ (1H) 77.7 (46) 76.0 (56) 63.5 (39) 68.9 (46) 58.3 (45) 62.6 (43) 50.5 (35) 75.7 (36) 73.0 (55) py: 12.8 (19) 10.8 (18) 16.8 (15) 17.2 (21) 19.1 (20) 17.2 (21) 19.7 (17) 11.7 (15) 10.8 (15) -2.0 (18) -4.9 (17) 1.2 (13) 3.9 (19) -0.2 (18) 5.6 (12) 0.8 (12) -3.5 (15) -4.2 (15)

δ (1H) 43.1 (31) 17.1 (8) 28.9 (45) 16.9 (14) 71.6 (36) 14.2 (8) 71.6 (36) 70.7 (37) R1,R2: 22.5 (11) 10.5 (7) 17.7 (16) 17.1 (11) [a]: 1H NMR spectrum acquired at 500 MHz. [b]: Due to solubility, the 1H NMR spectrum was acquired in deuterated chloroform solution, and spin systems were not characterised by 1H-1H COSY NMR. [c] The 1H NMR spectrum was acquired in deuterated acetone at 500 MHz, and assigned in a mixture with III Me,Me [Co (dpp )2][PF6]. [d] in benzene-d6 at 400 MHz.

James McPherson – June 2019 8-292 Chapter 8: Appendices III Me,Me Characterisation of [Co (dpp )2][PF6]

1 III Me,Me Figure S16. H NMR (600 MHz, acetone-d6) spectrum of [Co (dpp )2][PF6].

13 1 III Me,Me Figure S17. C{ H} NMR (600 MHz, acetone-d6) spectrum of [Co (dpp )2][PF6].

James McPherson – June 2019 8-293 Chapter 8: Appendices

1 13 III Me,Me Figure S18. H- C HSQC NMR (600 MHz, acetone-d6) spectrum of [Co (dpp )2][PF6].

1 13 III Me,Me Figure S19. H- C HMBC NMR (600 MHz, acetone-d6) spectrum of [Co (dpp )2][PF6].

James McPherson – June 2019 8-294 Chapter 8: Appendices

1 15 III Me,Me Figure S20. H- N HMBC NMR (600 MHz, acetone-d6) spectrum of [Co (dpp )2][PF6].

1 II H NMR spectra for paramagnetic [Co (dpp)2] complexes

C6HD5 H2O

1 CN,Ph Figure S21. H NMR (500 MHz, benzene-d6) of [Co(dpp )2].

James McPherson – June 2019 8-295 Chapter 8: Appendices

1 1 CN,Ph Figure S22. H- H COSY NMR (500 MHz, benzene-d6) of [Co(dpp )2].

C6HD5

H2O

1 CO Et,CO Et Figure S23. H NMR (400 MHz, benzene-d6) of [Co(dpp 2 2 )2].

James McPherson – June 2019 8-296 Chapter 8: Appendices

1 1 CO Et,CO Et Figure S24. H- H COSY NMR (400 MHz, benzene-d6) of [Co(dpp 2 2 )2].

CHCl3

H2O

1 CO Me,4-Py Figure S25. H NMR (400 MHz, CDCl3) spectrum of [Co(dpp 2 )2].

James McPherson – June 2019 8-297 Chapter 8: Appendices

C6HD5

H2O

1 CO Me,Me Figure S26. H NMR (500 MHz, benzene-d6) of [Co(dpp 2 )2].

1 1 CO Me,Me Figure S27. H- H COSY NMR (500 MHz, benzene-d6) of [Co(dpp 2 )2].

James McPherson – June 2019 8-298 Chapter 8: Appendices

acetone

H2O

1 Me,Me Figure S28. H NMR (500 MHz, acetone-d6) spectrum of [Co(dpp )2].

1 1 Me,Me Figure S29. H- H COSY NMR (500 MHz, acetone-d6) spectrum of [Co(dpp )2].

James McPherson – June 2019 8-299 Chapter 8: Appendices

C6HD5 H2O

1 Me,Me Figure S30. H NMR (500 MHz, benzene-d6) of [Co(dpp )2].

Ligand 1H and 15N investigations: For SCO-active iron(II) complexes, some of us recently reported a predictive correlation between the found or DFT calculated 15N chemical shift of the N donor atoms within azine rings of unbound bi- and tri-dentate 19 ligands (δNA) and the switching temperatures (T1/2). As N-donor ligands are by far the most commonly used to obtain SCO-active complexes, this simple approach, or the closely related use of calculated partial charge 20 on the N donor (ρNA), has great potential for wide application. Indeed, even for families of mixed donor systems, if the changes to the ligand mostly affect the N-donor then this approach should still prove useful. The 1H and 13C chemical shifts for the ligands were therefore acquired. A correlation between chemical shift and SCO T1/2 values was not found (Table S6 and Figure S32).

Figure S31. 1H–15N HSQC (left) and HMBC (right) NMR (600 MHz, d-chloroform, 298 K) spectra of mixtures of dppCN,PhH (black), dppCO2Et,CO2EtH (orange), dppCO2Me,4-PyH (purple), dppCO2Me,MeH (green) and dppMe,MeH (red), with superimposed 1H NMR (600 MHz, d-chloroform, 298 K) traces of each individual ligand above. The correlations in the HSQC have been assigned (with dppCO2Et,CO2EtH and dppCO2Me,4-PyH overlapping); there is not enough separation in the HMBC experiment to assign the individual species within the mixture (see

James McPherson – June 2019 8-300 Chapter 8: Appendices ESI for experiments of the individual species). Note that the 15N environment for the 4-pyridyl substituent in dppCO2Me,4-PyH is also observed (shown in purple in the figure on the right) at 304.0 ppm through correlation with the proton at the 2′ position at 8.68 ppm.

Table S6. Solution state switching temperatures (T1/2) calculated from the magnetic data for [Co(dpp)2] complexes and 1H and 15N NMR data for the corresponding dppH ligand precursors.

1 15 1 15 R1,R2 – χT (CDCl3 @298 K) δ( Hpyrrole) δ( Npyrrole) J(H,Npyrrole) δ( Npy) (dpp ) T1/2 /K /cm3 K mol-1 /ppm /ppm /Hz /ppm (dppCN,Ph)– inactive 2.8 11.06 150.5 99.4 299.8, 300.0 (dppCO2Et,CO2Et)– 210 ± 20 2.3 11.07 154.1 98.8 302.0 (dppCO2Me,4-Py)– 230 ± 30 2.4 11.08 154.2 99.1 300.7, 303.0 (dppCO2Me,Me)– 250 ± 10 1.75 10.71 153.1 98.6 300.9, 300.9 (dppMe,Me)– - 2.9 10.35 142.3 98.2 299.0

Figure S32. Switching temperatures calculated from solution state magnetic studies in deuterated chloroform R1,R2 1 15 15 of [Co(dpp )2] complexes vs: δ Hpyrrole (top left), δ Npyrrole (top right), δ Npyridyl (bottom right) and pyrrole 1 R1,R2 1 2 J(NH) (bottom left) for dpp H precursors; where R , R = CN, Ph (black); CO2Et, CO2Et (orange); CO2Me, 4-Py (purple); or CO2Me, Me (green).

James McPherson – June 2019 8-301 Chapter 8: Appendices

The individual 1H-15N HSQC and HMBC spectra follow.

1 15 CN,Ph Figure S33. H– N HSQC (600 MHz, CDCl3) of dpp H.

1 15 CN,Ph Figure S34. H– N HMBC (600 MHz, CDCl3) of dpp H.

James McPherson – June 2019 8-302 Chapter 8: Appendices

1 15 CO Et,CO Et Figure S35. H– N HSQC (600 MHz, CDCl3) of dpp 2 2 H.

1 15 CO Et,CO Et Figure S36. H– N HMBC (600 MHz, CDCl3) of dpp 2 2 H.

James McPherson – June 2019 8-303 Chapter 8: Appendices

1 15 CO Me,4-Py Figure S37. H– N HSQC (600 MHz, CDCl3) of dpp 2 H.

1 15 CO Me,Py Figure S38. H– N HMBC (600 MHz, CDCl3) of dpp 2 H.

James McPherson – June 2019 8-304 Chapter 8: Appendices

1 15 CO Me,Me Figure S39. H– N HSQC (600 MHz, CDCl3) of dpp 2 H.

1 15 CO Me,Me Figure S40. H– N HMBC (600 MHz, CDCl3) of dpp 2 H.

James McPherson – June 2019 8-305 Chapter 8: Appendices

1 15 Me,Me Figure S41. H– N HSQC (600 MHz, CDCl3) of dpp H.

1 15 Me,Me Figure S42. H– N HMBC (600 MHz, CDCl3) of dpp H.

James McPherson – June 2019 8-306 Chapter 8: Appendices 9. Additional Electrochemistry

CO Me,Me Further to the discussion of the full width CV of exemplar [Co(dpp 2 )2], shown in Figure 10: An irreversible reduction wave (indicated by an asterisk (*) in Figure S44) was also observed in the tail the c cathodic discharge limit (E p = –2.34 V). This process occurred at slightly higher potentials for the more 1 2 CN,Ph electron withdrawing R , R groups (Table 2 and Figure S44), for example at –2.24 V for [Co(dpp )2], while Me,Me the process was not observed within the cathodic discharge limits for the more electron rich [Co(dpp )2]. Further analysis was complicated by the encroaching discharge limit.

CO Me,Me Me,Me Figure S43. The cyclic voltammograms of [Co(dpp)2] (dpp = dpp 2 in green and dpp in red) and Me,Me [Co(dpp )2][PF6] (dashed red line). The current was normalised to give equal peak currents for each CoIII/CoII couple.

CN,Ph CO Et,CO Et CO Me,4- Figure S44. The cyclic voltammograms of [Co(dpp)2] (dpp = dpp black, dpp 2 2 orange, dpp 2 Py purple and dppCO2Me,Me green) showing the CoIII/CoII couples and a multi-electron reduction process for each species. The current was normalised to give equal peak currents for the CoIII/CoII couples.

James McPherson – June 2019 8-307 Chapter 8: Appendices III II 10. DFT Study of Co /Co Redox Processes in the [Co(dpp)2] Complexes

Geometry optimizations and solution free energies

II Solution free energies for each DFT optimised [Co (dpp)2] complex (B3LYP-GD3BJ/def2-TZVPP) in both LS (S = 1/2) or HS (S = 3/2) electronic configurations were obtained from single point energy calculations using a conductor-like polarizable continuum solvation model (CPCM) in CH2Cl2, the solvent used for all experimental electrochemical measurements undertaken in this work. The solution free energies were also III + calculated for each of the one electron oxidised [Co (dpp)2] species in their LS (S=0) state only, as HS 21, 22 II Co(III) is extremely rare. The difference between solution free energies (∆Gsol) of each [Co (dpp)2] III + complex and its oxidised [Co (dpp)2] analogue can be used to predict the oxidation potential (E°) in V against the standard Fc+/0 couple by equation S8:23

° = 4.9 (S8) ∆𝐺𝐺𝑠𝑠𝑠𝑠𝑠𝑠 where n is the number of electrons𝐸𝐸 −transferred𝑛𝑛𝑛𝑛 − (n = 1 in this case) and F is the Faraday constant.

The Co–N bond lengths, and twelve cis N–Co–N angles from the optimized geometries of all five [Co(dpp)2] complexes are summarised below in Table S7.

James McPherson – June 2019 8-308 Chapter 8: Appendices N1 N2 N4 Co N6 N5 3 N Figure S45. The nitrogen labelling scheme for the distances and angles presented in Table S7. Note N2 and N5 represent the two pyrrolide N donor atoms.

Table S7. Reported structural data for calculated systems. Coordination sphere is analysed through Co-N bond distance (reported in Å) and N-Co-N angle (reported in degrees).

1 5 1 3 1 2 1 6 3 5 3 4 3 2 3 6 5 4 4 2 2 6 6 5 1 2 3 4 5 6

II CN,Ph Co (dpp )2[LS] 103,07 93 75,94 94,53 106,14 92,04 74,85 92,43 78,16 102,02 101,92 77,93 2,24 1,89 2,33 2,09 1,83 2,08 II COOEt,COOEt Co (dpp )2[LS] 106,02 91,27 75,71 94,77 101,87 94,4 76,4 91 78 101,9 102 78,1 2,24 1,88 2,22 2,09 1,84 2,08 II COOMe,4-Py Co (dpp )2[LS] 103,38 93,3 75,85 93,06 105,19 92,41 75,59 92,7 78,69 101,2 102,19 77,93 2,27 1,89 2,23 2,08 1,84 2,06 II COOMe,Me Co (dpp )2[LS] 103,91 93,08 76,05 92,41 104,28 93,73 75,74 92,19 77,86 103,49 100,11 78,53 2,27 1,89 2,21 2,08 1,84 2,07 II Me,Me Co (dpp )2[LS] 103,36 93,02 76,65 93,1 103,33 93,09 76,66 93 76067 103,33 103,32 76,67 2,18 1,86 2,18 2,18 1,86 2,18

III CN,Ph + Co (dpp )2 99,83 91,77 79,59 89,9 101,36 94,3 79,19 91,78 79,2 101,52 99,66 79,6 2 1,82 1,99 1,99 1,82 2 III COOEt,COOEt + Co (dpp ) 2 100,4 90,84 79,44 92,84 100,56 92,25 79,59 91,67 79,47 99,54 101,45 79,53 1,99 1,82 1,99 1,99 1,82 1,99 III COOMe,4-Py + Co (dpp ) 2 100,34 92,15 79,9 92,68 100,6 91,35 79,15 91,45 79,25 101,4 99,69 79,66 1,99 2 1,98 1,99 1,82 2 III COOMe,Me + Co (dpp ) 2 101,09 91,82 79,47 92,12 101,09 91,82 97,47 92,12 79,3 100,05 100,83 79,82 1,99 1,82 1,98 1,98 1,82 1,99 III Me,Me + Co (dpp ) 2 101 92,24 79,19 92,31 100,59 91,72 79,12 91,77 79,2 100,76 100,83 79,21 2 1,82 2 2 1,82 2

James McPherson – June 2019 8-309 Chapter 8: Appendices Frontier Orbital Analyses

II CN,Ph II CN,Ph III CN,Ph + Co (dpp )2[HS] Co (dpp )2[LS] Co (dpp )2

HOMO HOMO HOMO

LUMO LUMO LUMO

Figure S46. The frontier MOs (HOMOs above and LUMOs below) of all the electronic derivatives of CN,Ph [Co(dpp )2].

James McPherson – June 2019 8-310 Chapter 8: Appendices II COOEt,COOEt II COOEt,COOEt III COOEt,COOEt + Co (dpp )2[HS] Co (dpp )2[LS] Co (dpp )2

HOMO HOMO HOMO

LUMO LUMO LUMO

Figure S47. The frontier MOs (HOMOs above and LUMOs below) of all the electronic derivatives of CO Et,CO Et [Co(dpp 2 2 )2].

James McPherson – June 2019 8-311 Chapter 8: Appendices II COOMe,4-Py II COOMe,4-Py III COOMe,4-Py + Co (dpp )2[HS] Co (dpp )2[LS] Co (dpp )2

HOMO HOMO HOMO

LUMO LUMO LUMO

Figure S48. The frontier MOs (HOMOs above and LUMOs below) of all the electronic derivatives of CO Me,4-Py [Co(dpp 2 )2].

James McPherson – June 2019 8-312 Chapter 8: Appendices II COOMe,Me II COOMe,Me III COOMe,Me + Co (dpp )2[HS] Co (dpp )2[LS] Co (dpp )2

HOMO HOMO HOMO

LUMO LUMO LUMO

Figure S49. The frontier MOs (HOMOs above and LUMOs below) of all the electronic derivatives of CO Me,Me [Co(dpp 2 )2].

James McPherson – June 2019 8-313 Chapter 8: Appendices II Me,Me II Me,Me III Me,Me + Co (dpp )2[HS] Co (dpp )2[LS] Co (dpp )2

HOMO HOMO HOMO

LUMO LUMO LUMO

Figure S50. The frontier MOs (HOMOs above and LUMOs below) of all the electronic derivatives of Me,Me [Co(dpp )2].

James McPherson – June 2019 8-314 Chapter 8: Appendices Structural Reorganization Analysis

II III + The structural rearrangement due to oxidation of the [Co (dpp)2] complexes to [Co (dpp)2] were quantified by the root mean square deviation (RMSD) of the atomic positions within the optimised structures (Figure SX R1,R2 and Table S7). To make systematic comparisons in the rearrangements across the five [Co(dpp )2] complexes, the R1 and R2 substituents were removed prior to these analyses.

II CN,Ph II COOEt,COOEt II COOMe,4-Py II COOMe,Me II Me,Me Co (dpp )2 Co (dpp )2 Co (dpp )2 Co (dpp )2 Co (dpp )2

Figure S51. Reported superimposed structure for RMSD calculations (see Table S8) of the Cobalt II R1,R2 coordination sphere distortion along the series of five complexes. Co (dpp )2[HS] are reported in blue, II R1,R2 III R1,R2 + Co (dpp )2[LS] are reported in red, and [Co (dpp )2] are reported in black.

Table S8. The structural reorganization for the LS (first column) and HS (second column) configurations of II R1,R2 III R1,R2 + the five [Co (dpp )2] complexes as they are oxidized to [Co (dpp )2] (as RMSD values in Å) from the overlaying the optimized (B3LYP-GD3BJ/def2-TZVPP) structures (after first removing the R1 and R2 substituents).

II R1,R2 II R1,R2 RMSD [Co (dpp )2] [LS] RMSD [Co (dpp )2] [HS] III CN,Ph + [Co (dpp )2] 0.193 0.343 III CO Et,CO Et + [Co (dpp 2 2 )2] 0.169 0.321 III CO Me,4-Py + [Co (dpp 2 )2] 0.171 0.330 III CO Me,Me + [Co (dpp 2 )2] 0.169 0.319 III Me,Me + [Co (dpp )2] 0.167 0.291

As measured, the closer the RMSD is to zero, the smaller the reorganization. Table 7 shows that all LS configurations are associated with smaller reorganisation (RMSD ~ 0.17 Å), while unsurprisingly oxidation of II III + HS [Co (dpp)2] centres to LS [Co (dpp)2] involves much larger reorganization (with RMSD of 0.29–0.34 Å).

James McPherson – June 2019 8-315 Chapter 8: Appendices III II + 11. Fit of E1/2(Ru /Ru ) vs Swain-Lupton parameters for [Ru(dpp)(tpy)] complexes

III II R1,R2 1 2 The Ru /Ru redox potentials for [Ru(dpp )(tpy)][PF6] (R , R = COOMe, 4-Py; COOMe, Me; and Me, Me) +/0 24 were reported against Fc measured in acetonitrile–0.1 M [Bu4N][PF6] by Colbran et al. Table S2 below III II shows the E1/2(Ru /Ru ) with the sum modified Swain-Lupton force (F) and resonance (R) parameters for each dpp– pyrrolide substitiuent.16 Note that although Colbran et al. also reported data for COOMe,PyMe [Ru(dpp )(tpy)][PF6]2, Swain-Lupton parameters for N-methyl-pyridinium ions are not currently available.

+/0 Table S9. Ruthenium(III)/ruthenium(II) redox potentials against Fc in acetonitrile–0.1 M [Bu4N][PF6] as reported by Colbran et al.24 and modified Swain-Lupton parameters for the respective [Ru(dpp)(tpy)]+ complexes.

III II 24 16 16 Complex: E1/2 (Ru /Ru ) ΣF ΣR /V vs Fc+/0 [Ru(dppCO2Me,4-Py)(tpy)]+ 0.27 0.55 0.34 [Ru(dppCO2Me,Me)(tpy)]+ 0.19 0.35 -0.07 [Ru(dppMe,Me)(tpy)]+ -0.04 0.02 -0.36

Fitting these data by linear regression results in equation (S9) below, which describes the deviation of the RuIII/RuII redox potential as a function of the electronic parameters of the dpp– pyrrolide substituents, and is shown graphically in Figure S39.

III II E1/2(Ru /Ru ) = –0.150 + 0.920F – 0.254R (S9)

Figure S52. Fitting of the RuIII/RuII redox couple for Colbran’s [Ru(dpp)(tpy)]+ complexes24 against the modified Swain-Lupton field (F) and resonance (R) parameters for the dpp– pyrrolide substituents.16

James McPherson – June 2019 8-316 Chapter 8: Appendices

Figure S53. Magnetic susceptibility at 298 K (squares on the left), and switching temperatures (triangles on R1,R2 III II +/0 the right) calculated from solution state data for [Co(dpp )2] complexes against E1/2 (Co /Co ) in V vs Fc . 1 2 R , R are: CN, Ph (black); CO2Et, CO2Et (orange); CO2Me, 4-Py (purple); or CO2Me, Me (green). Solution Me,Me state data were not collected for [Co(dpp )2] however it was observed at HS for all T (50–400 K) in the 3 –1 solid state measurements, with χMT = 2.9 cm K mol at 298 K, and is therefore represented as half filled red symbols.

12. References

1. Fulmer, G. R.; Miller, A. J. M.; Sherden, N. H.; Gottlieb, H. E.; Nudelman, A.; Stoltz, B. M.; Bercaw, J. E.; Goldberg, K. I. NMR Chemical Shifts of Trace Impurities: Common Laboratory Solvents, Organics, and Gases in Deuterated Solvents Relevant to the Organometallic Chemist. Organometallics 2010, 29, 2176- 2179. 2. Hein, F.; Melichar, F. Synthesis of 2,5-di(α-pyridyl)pyrrole and of several complex derivatives of 2,5- di(α-pyridyl)-3,4-dicarbethoxypyrrole. Pharmazie 1954, 9, 455-60. 3. McPherson, J. N.; Elton, T. E.; Colbran, S. B. A Strain-Deformation Nexus within Pincer Ligands: Application to the Spin States of Iron(II) Complexes. Inorg. Chem. 2018, 57, 12312-12322. 4. Xiao, S.; Liu, Z.; Zhao, J.; Pei, M.; Zhang, G.; He, W. A novel fluorescent sensor based on imidazo [1, 2-a] pyridine for Zn2+. RSC Adv. 2016, 6, 27119-27125. 5. Sheldrick, G. SHELXT - Integrated space-group and crystal-structure determination. Acta Crystallogr. A 2015, 71, 3-8. 6. Sheldrick, G. Crystal structure refinement with SHELXL. Acta Crystallogr. C 2015, 71, 3-8. 7. Dolomanov, O. V.; Bourhis, L. J.; Gildea, R. J.; Howard, J. A. K.; Puschmann, H. OLEX2: a complete structure solution, refinement and analysis program. J. Appl. Crystallogr. 2009, 42, 339-341. 8. Hunter, C. A.; Sanders, J. K. M. The nature of π-π interactions. J. Am. Chem. Soc. 1990, 112, 5525- 5534. 9. Ciszek, J. W.; Keane, Z. K.; Cheng, L.; Stewart, M. P.; Yu, L. H.; Natelson, D.; Tour, J. M. Neutral Complexes of First Row Transition Metals Bearing Unbound Thiocyanates and Their Assembly on Metallic Surfaces. J. Am. Chem. Soc. 2006, 128, 3179-3189. 10. Robinson, K.; Gibbs, G. V.; Ribbe, P. H. Quadratic Elongation: A Quantitative Measure of Distortion in Coordination Polyhedra. Science 1971, 172, 567-570. 11. Cowieson, N. P.; Aragao, D.; Clift, M.; Ericsson, D. J.; Gee, C.; Harrop, S. J.; Mudie, N.; Panjikar, S.; Price, J. R.; Riboldi-Tunnicliffe, A.; Williamson, R.; Caradoc-Davies, T. MX1: a bending-magnet crystallography beamline serving both chemical and macromolecular crystallography communities at the Australian Synchrotron. J. Synchrotron Radiat. 2015, 22, 187-190. 12. Kabsch, W. XDS. Acta Crystallogr. D 2010, 66, 125-132. 13. CrysAlisPro, Oxford Diffraction Ltd: Abingdon, Oxfordshire, England, 2006. 14. Hubschle, C. B.; Sheldrick, G. M.; Dittrich, B. ShelXle: a Qt graphical user interface for SHELXL. J. Appl. Crystallogr. 2011, 44, 1281-1284.

James McPherson – June 2019 8-317 Chapter 8: Appendices 15. Swain, C. G.; Lupton, E. C. Field and resonance components of substituent effects. J. Am. Chem. Soc. 1968, 90, 4328-4337. 16. Hansch, C.; Leo, A.; Taft, R. W. A survey of Hammett substituent constants and resonance and field parameters. Chem. Rev. 1991, 91, 165-195. 17. Chambers, J.; Eaves, B.; Parker, D.; Claxton, R.; Ray, P. S.; Slattery, S. J. Inductive influence of 4′- terpyridyl substituents on redox and spin state properties of iron(II) and cobalt(II) bis-terpyridyl complexes. Inorganica Chim. Acta 2006, 359, 2400-2406. 18. Hammett, L. P. The Effect of Structure upon the Reactions of Organic Compounds. Benzene Derivatives. J. Am. Chem. Soc. 1937, 59, 96-103. 19. Rodriguez-Jimenez, S.; Yang, M.; Stewart, I.; Garden, A. L.; Brooker, S. A Simple Method of Predicting Spin State in Solution. J. Am. Chem. Soc. 2017, 139, 18392-18396. 20. Kimura, A.; Ishida, T. Spin-Crossover Temperature Predictable from DFT Calculation for Iron(II) Complexes with 4-Substituted Pybox and Related Heteroaromatic Ligands. ACS Omega 2018, 3, 6737-6747. 21. Bersuker, I. B. Electronic Structure and Properties of Transition Metal Compounds: Introduction to the Theory, Second Edition. 2010. 22. Albright, T. A.; Burdett, J. K.; Whangbo, M. H. Orbital Interactions in Chemistry: Second Edition. 2013; p 1-819. 23. Namazian, M.; Lin, C. Y.; Coote, M. L. Benchmark Calculations of Absolute Reduction Potential of Ferricinium/Ferrocene Couple in Nonaqueous Solutions. J. Chem. Theory Comput. 2010, 6, 2721-2725. 24. McSkimming, A.; Diachenko, V.; London, R.; Olrich, K.; Onie, C. J.; Bhadbhade, M. M.; Bucknall, M. P.; Read, R. W.; Colbran, S. B. An Easy One-Pot Synthesis of Diverse 2,5-Di(2-pyridyl)pyrroles: A Versatile Entry Point to Metal Complexes of Functionalised, Meridial and Tridentate 2,5-Di(2-pyridyl)pyrrolato Ligands. Chem. Eur. J. 2014, 20, 11445-11456.

James McPherson – June 2019 8-318 Chapter 8: Appendices 13. Coordinates for DFT Optimised H -3.71021 -1.29816 -0.14719 Geometries H -6.12055 -0.84556 0.23755 H -6.86398 1.53242 0.65936 (dppCN,Ph)– H -5.10851 3.30940 0.66948 C 2.17218 4.41122 -0.09820 Energy: -1027.22666844 Eh C 3.52581 4.11806 -0.26422 N 0.89273 1.24559 0.03458 C 3.86228 2.77552 -0.44710 C -0.41660 0.86995 -0.01731 C 2.86255 1.81494 -0.43298 C -0.53861 -0.55081 -0.05533 C 1.51603 2.19680 -0.21812 C 0.79732 -1.03219 -0.07699 N 1.19574 3.50637 -0.08189 C 1.64149 0.13166 -0.00630 H 1.85373 5.44834 0.02213 C 3.10578 0.19292 0.03401 H 4.27415 4.90490 -0.26752 C 3.90548 -0.97246 -0.01153 H 4.89560 2.47862 -0.61123 C 5.28784 -0.85868 0.02721 H 3.12104 0.78245 -0.61242 C 5.86955 0.40726 0.11089 C -1.23266 -2.04005 -0.30373 C 5.00444 1.50140 0.15084 C 1.65528 -1.04249 -0.45318 N 3.67618 1.41904 0.11427 O -0.16607 -2.82394 0.07302 H 3.44602 -1.94965 -0.07549 O -2.28430 -2.57538 -0.64046 H 5.90512 -1.75318 -0.00756 C -0.26597 -4.21552 -0.22834 H 6.94640 0.54430 0.14393 O 2.02639 -1.62020 -1.45848 H 5.41186 2.51195 0.21622 O 2.37724 -1.08450 0.70626 C -1.97842 4.15334 -0.07953 C 3.62406 -1.78623 0.64650 C -3.35240 3.93568 -0.18471 C 4.42252 -1.37929 1.87141 C -3.77575 2.60748 -0.25627 H 4.14354 -1.53222 -0.28353 C -2.83692 1.58819 -0.19849 H 3.43214 -2.86531 0.62469 C -1.46368 1.89804 -0.04911 H 4.59344 -0.29887 1.86893 N -1.05930 3.19205 -0.02448 H 5.39207 -1.88907 1.88427 H -1.59303 5.17418 -0.04701 H 3.88268 -1.63454 2.78789 H -4.05065 4.76641 -0.22520 C 1.13352 -4.79340 -0.10759 H -4.83014 2.36491 -0.36723 H -0.67091 -4.34429 -1.23724 H -3.15704 0.56017 -0.28200 H -0.96856 -4.69538 0.46456 C 1.14646 -2.39855 -0.10892 H 1.80641 -4.27629 -0.79643 H -4.74492 -3.89573 0.28878 H 1.12960 -5.86370 -0.34183 C -3.90141 -3.21464 0.21481 H 1.51784 -4.66308 0.90919 C -2.99667 -3.32406 -0.84155 H -3.13219 -4.09421 -1.59641 (dppCO2Me,4-Py)– C -1.91643 -2.44825 -0.93845

H -1.21542 -2.53653 -1.76140 Energy: -1178.89373851 Eh C -1.71002 -1.44101 0.01927 N -0.85812 1.53175 -0.00310 C -2.62095 -1.35783 1.08885 C 0.45222 1.15846 -0.00451 H -2.46129 -0.59777 1.84699 C 0.56892 -0.25672 0.00703 C -3.70616 -2.22536 1.18153 C -0.76476 -0.75521 0.07664 H -4.39546 -2.13716 2.01752 C -1.60933 0.42203 0.06587 N 1.43391 -3.53242 -0.14846 C -3.07773 0.56718 0.01435 C -3.96952 -0.52471 0.08750 (dppCO2Et,CO2Et)– C -5.33848 -0.29929 0.01758 C -5.81742 1.00374 -0.11242 Energy: -1238.31244406 Eh C -4.86541 2.02334 -0.15887 N -0.87169 1.65109 -0.04502 N -3.55038 1.83293 -0.10040 C 0.41487 1.23009 -0.19673 H -3.58129 -1.52405 0.22189 C 0.46431 -0.18081 -0.29839 H -6.02612 -1.14050 0.07191 C -0.88151 -0.63241 -0.24732 H -6.87932 1.22599 -0.16849 C -1.67088 0.57285 -0.09064 H -5.18772 3.06263 -0.24940 C -3.12111 0.76482 0.10388 C 2.00324 4.44620 0.04275 C -4.05271 -0.29463 0.06333 C 3.38213 4.23526 0.07314 C -5.40123 -0.03001 0.26669 C 3.81540 2.90885 0.09143 C -5.81980 1.27987 0.49762 C 2.88006 1.88468 0.06466 C -4.83223 2.26619 0.50345 C 1.49873 2.18826 0.00694 N -3.53513 2.03817 0.31668 N 1.08787 3.48055 0.01850

James McPherson – June 2019 8-319 Chapter 8: Appendices H 1.61134 5.46512 0.04495 H -1.92701 2.92386 0.68091 H 4.07728 5.06933 0.09283 H -0.69531 3.38591 -0.48045 H 4.87560 2.67015 0.13322 H 3.21398 0.85880 0.09721 (dppMe,Me)– C -1.08998 -2.17192 0.10065 N 4.11501 -2.72285 0.02637 Energy: -782.507010209 Eh C 3.32074 -2.60719 1.10152 N 0.00000 -0.52326 -0.00005 H 3.62336 -3.17724 1.97989 C 1.08909 0.28271 0.00002 C 2.17476 -1.81959 1.14813 C 0.69924 1.67222 0.00001 H 1.57410 -1.76909 2.05037 C -0.69925 1.67221 -0.00001 C 1.78589 -1.08861 0.01541 C -1.08909 0.28271 -0.00000 C 2.61488 -1.20741 -1.11076 C -2.42426 -0.29887 0.00002 H 2.36554 -0.66830 -2.01856 C -3.61058 0.48418 0.00002 C 3.74350 -2.01861 -1.05403 C -4.85502 -0.12694 0.00002 H 4.38644 -2.11503 -1.92880 C -4.93923 -1.52081 -0.00000 O -0.12425 -2.92339 -0.52818 C -3.72831 -2.21778 -0.00003 O -2.05804 -2.73730 0.60095 N -2.52373 -1.65820 -0.00002 C -0.25959 -4.33075 -0.38139 H -3.55270 1.56189 0.00003 H 0.60530 -4.76430 -0.88705 H -5.75635 0.48330 0.00003 H -1.18812 -4.69086 -0.83544 H -5.89104 -2.04393 -0.00000 H -0.26214 -4.62179 0.67406 H -3.73791 -3.31062 -0.00005 C 3.72831 -2.21777 -0.00007 (dppCO2Me,Me)– C 4.93923 -1.52081 -0.00001 C 4.85503 -0.12694 0.00010

Energy: -971.085530870 Eh C 3.61058 0.48419 0.00012 N -0.41256 -1.00437 -0.03749 C 2.42426 -0.29886 0.00002 C -1.35031 -0.01031 -0.03563 N 2.52373 -1.65819 -0.00005 C -0.73860 1.26971 -0.08670 H 3.73791 -3.31062 -0.00012 C 0.66213 1.01930 -0.11421 H 5.89104 -2.04393 -0.00003 C 0.79563 -0.43166 -0.09510 H 5.75636 0.48330 0.00018 C 1.97883 -1.31228 -0.03141 H 3.55271 1.56190 0.00024 C 3.31118 -0.84425 -0.07913 C -1.56919 2.89686 -0.00000 C 4.36324 -1.74673 0.00893 H -0.96858 3.81216 0.00000 C 4.09673 -3.10997 0.13235 H -2.22381 2.95319 0.88142 C 2.75265 -3.48625 0.15546 H -2.22382 2.95320 -0.88143 N 1.72847 -2.64204 0.08143 C 1.56918 2.89687 -0.00006 H 3.50140 0.21086 -0.20802 H 2.22391 2.95307 -0.88141 H 5.38813 -1.38274 -0.02585 H 2.22369 2.95336 0.88144 H 4.88842 -3.85082 0.20191 H 0.96856 3.81217 -0.00029 H 2.48742 -4.54240 0.24149 C -4.35302 -2.05774 0.02578 (dppCF3, CF3)– C -5.43805 -1.17880 0.05104 C -5.13809 0.18362 0.06326 Energy: -1377.96010868 Eh C -3.81269 0.59334 0.03654 N -0.00004 1.33334 0.00050 C -2.76856 -0.36591 -0.01691 C -1.09391 0.53934 0.00242 N -3.07625 -1.69060 -0.00216 C -0.70832 -0.83802 -0.03640 H -4.53181 -3.13523 0.03319 C 0.70861 -0.83788 0.03660 H -6.45979 -1.54625 0.07143 C 1.09394 0.53952 -0.00151 H -5.93201 0.92668 0.10150 C 2.42798 1.16349 -0.03403 H -3.58687 1.64762 0.07249 C 3.62650 0.42620 -0.17605 C 1.68630 2.04507 -0.14847 C 4.84177 1.09130 -0.25664 C -1.41375 2.60538 -0.23798 C 4.86698 2.48556 -0.22484 O 1.28561 3.21619 0.47579 C 3.63530 3.13478 -0.13938 O 2.81338 2.00437 -0.63578 N 2.45916 2.51780 -0.06010 C 2.22871 4.27759 0.40020 H 3.60859 -0.64928 -0.24890 H 1.76071 5.12805 0.90175 H 5.76215 0.52126 -0.35984 H 2.46507 4.53477 -0.63764 H 5.79457 3.04708 -0.28596 H 3.16571 4.01348 0.90078 H 3.59511 4.22562 -0.14638 H -2.16755 2.58008 -1.03416 C -3.63529 3.13466 0.13950

James McPherson – June 2019 8-320 Chapter 8: Appendices C -4.86708 2.48550 0.22379 O 2.02212 -2.82019 -0.7078 C -4.84198 1.09124 0.25529 C -3.62670 0.42605 0.17560 C -2.42799 1.16329 0.03490 N -2.45913 2.51760 0.06101 H -3.59499 4.22549 0.14678 H -5.79470 3.04706 0.28415 H -5.76251 0.52126 0.35748 H -3.60901 -0.64944 0.24814 C 1.55135 -2.00587 0.39082 F 0.85521 -3.04704 0.90754 F 2.47791 -1.69856 1.35087 F 2.28225 -2.52295 -0.64993 C -1.55123 -2.00581 -0.39080 F -2.47739 -1.69822 -1.35107 F -2.28256 -2.52259 0.64972 F -0.85527 -3.04734 -0.90724

(dppNO2,Ph)–

Energy: -1139.47502866 Eh N 0.91653 1.25372 -0.05070 C -0.40734 0.90944 -0.00959 C -0.56585 -0.49378 0.03514 C 0.75533 -1.00757 -0.03487 C 1.64836 0.13289 -0.07250 N 1.00382 -2.39939 -0.11614 C -1.43392 1.95912 -0.02850 C -2.82033 1.67613 -0.00248 C -3.73737 2.71699 -0.02648 C -3.28045 4.03449 -0.07970 C -1.89844 4.22262 -0.10932 N -0.99940 3.24116 -0.08709 H -3.16930 0.65520 0.03426 H -4.80239 2.49810 -0.00554 H -3.96164 4.88001 -0.10040 H -1.48995 5.23376 -0.15556 C 3.12086 0.25851 -0.01834 C 4.00477 -0.83996 0.02655 C 5.37248 -0.61832 0.13021 C 5.85701 0.68731 0.18290 C 4.91159 1.71309 0.13617 N 3.59812 1.52509 0.04967 H 3.61084 -1.84295 -0.03379 H 6.05335 -1.46558 0.16719 H 6.91808 0.90672 0.25994 H 5.23710 2.75420 0.17874 C -1.81633 -1.28546 0.05688 C -2.39200 -1.75068 -1.13330 H -1.88352 -1.54943 -2.07092 C -3.59814 -2.45055 -1.12037 H -4.02818 -2.80143 -2.05498 C -4.25457 -2.69552 0.08599 H -5.19692 -3.23671 0.09733 C -3.68890 -2.24143 1.27948 H -4.18985 -2.43128 2.22532 C -2.48135 -1.54762 1.26364 H -2.03726 -1.19321 2.18811 O 0.16961 -3.18353 0.37425

James McPherson – June 2019 8-321 Chapter 8: Appendices

R1,R2 Table S10. Cartesian coordinates for the DFT optimized structures of each [Co(dpp )2] complex in each electronic state.

II CN,Ph II CN,Ph III CN,Ph + Co (dpp )2[HS] Co (dpp )2[LS] Co (dpp )2 Co 0.00000 0.00000 0.00000 Co 0.00000 0.00000 0.00000 Co 0.00000 0.00000 0.00000 N 0.06379 -2.26185 -0.64970 N -0.07265 2.18069 -0.52684 N 0.06424 -1.95889 -0.36369 N 0.02105 -0.05038 -1.96956 N -0.00348 0.01594 -1.88765 N 0.48252 -0.00550 -1.75976 N -2.31029 -0.10482 0.43482 N 2.04587 0.05965 0.43252 N -1.99563 0.03316 -0.19219 N 0.06379 2.15908 -0.64970 N 0.11332 -2.23970 -0.62770 N 0.06424 1.97023 -0.36369 N -0.21142 -0.31316 1.93653 N -0.00348 0.01594 1.83467 N -0.49940 0.02878 1.75345 C -0.00834 1.05064 -2.72813 C 0.07057 -1.06892 -2.66923 C 0.53745 1.10610 -2.49052 N 2.06869 0.14120 0.87951 N -2.02643 -0.13203 0.43144 N 1.78109 0.04418 0.89554 C -0.00393 -1.16143 -2.74239 C -0.05907 1.14427 -2.64191 C 0.58891 -1.12600 -2.50104 C -0.04722 -0.77548 -4.09789 C -0.03042 0.78344 -4.00488 C 0.76577 -0.72924 -3.84916 N -2.82693 -1.72835 5.60509 N 2.31228 0.19460 5.93001 N -3.86942 0.62616 4.96131 C -0.06124 0.65318 -4.08456 C 0.06408 -0.63887 -4.02011 C 0.72356 0.70250 -3.83512 C -0.05281 -2.41376 -1.99294 C -0.04151 2.37846 -1.86980 C 0.34186 -2.29022 -1.67076 C -1.08504 -0.99389 3.87083 C 0.70244 0.06996 3.93697 C -1.75159 0.34798 3.54165 C 0.02474 2.30616 -1.99964 C 0.13728 -2.34538 -1.97931 C 0.32455 2.28226 -1.67603 C -1.36704 -0.60506 2.53880 C 1.10327 0.05977 2.57628 C -1.76716 0.16500 2.13733 C -2.55836 -0.49334 1.71020 C 2.30900 0.08837 1.77440 C -2.66452 0.14062 1.00327 C -2.02810 -1.40415 4.83542 C 1.56852 0.14291 5.04678 C -2.89956 0.49554 4.34973 N -0.10562 2.23714 -6.10227 N 0.14469 -2.17392 -6.07419 N 0.96519 2.35225 -5.78377 C 2.11755 -0.14153 2.20832 C -2.31041 -0.13538 1.77099 C 1.73007 0.14562 2.26734 C -3.85840 -0.76622 2.14067 C 3.61691 0.13961 2.25052 C -4.05217 0.22228 1.03199 H -4.03550 -1.07959 3.15886 H 3.79664 0.16523 3.31569 H -4.55705 0.30961 1.98342 C 0.33388 -0.93698 4.03048 C -0.72755 0.04038 3.95273 C -0.38221 0.31815 3.96094 C 0.00024 -3.32534 0.14429 C -0.03874 3.22480 0.29420 C -0.20198 -2.91345 0.52872 H 0.08796 -3.12781 1.20620 H -0.06373 2.99365 1.35252 H -0.41123 -2.57784 1.53417 C 0.82757 -0.49427 2.78563 C -1.11944 -0.00106 2.59459 C 0.37155 0.12978 2.77729 C 0.11045 3.24071 0.12247 C 0.17268 -3.34154 0.11083 C -0.15398 2.94460 0.51915 H 0.14217 3.05942 1.18977 H 0.15118 -3.19397 1.18420 H -0.35115 2.63587 1.53585 C -3.32010 0.01663 -0.41931 C 3.05924 0.07785 -0.42804 C -2.68135 -0.00530 -1.33434 H -3.05610 0.32357 -1.42435 H 2.78841 0.05280 -1.47541 H -2.09921 -0.09304 -2.24060 C 0.02782 3.57344 -2.58705 C 0.22279 -3.59044 -2.60745 C 0.35196 3.60775 -2.09535 H -0.00345 3.66715 -3.66258 H 0.24052 -3.65203 -3.68561 H 0.55561 3.83013 -3.13300 C -0.24761 -3.68205 -2.54739 C 0.04563 3.67199 -2.39422 C 0.32236 -3.62626 -2.06351 H -0.35943 -3.79184 -3.61423 H 0.09216 3.81824 -3.46147 H 0.52374 -3.87492 -3.09379 C 0.11833 4.53071 -0.38859 C 0.25837 -4.61476 -0.43771 C -0.13551 4.28175 0.15735 H 0.15635 5.37994 0.27777 H 0.30436 -5.48297 0.20356 H -0.31948 5.04042 0.90289 C -0.08195 1.50846 -5.20553 C 0.10283 -1.46870 -5.15933 C 0.86588 1.58380 -4.92854 C -0.04159 -1.61042 -5.30612 C -0.11021 1.64835 -5.18892 C 0.97592 -1.57816 -5.02437 C 3.32117 -0.03131 2.91095 C -3.62370 -0.29752 2.21191 C 2.90970 0.28826 2.99395 H 3.34287 -0.23609 3.96955 H -3.83118 -0.31818 3.27031 H 2.85792 0.38619 4.06721

James McPherson – June 2019 8-322 Chapter 8: Appendices

C -4.90326 -0.63229 1.24019 C 4.66216 0.15757 1.34004 C -4.75467 0.18822 -0.16254 H -5.91695 -0.83943 1.55599 H 5.68429 0.19683 1.69138 H -5.83386 0.25157 -0.15625 C 1.07086 -1.33076 5.23698 C -1.57423 0.08770 5.14821 C 0.10760 0.42947 5.33697 C 0.07450 4.69048 -1.77004 C 0.28408 -4.73126 -1.82347 C 0.11806 4.61268 -1.16969 H 0.07821 5.67971 -2.20779 H 0.35118 -5.70448 -2.29096 H 0.13572 5.64796 -1.48043 C -4.63960 -0.23343 -0.06560 C 4.38601 0.12533 -0.02228 C -4.06453 0.06947 -1.36440 H -5.43003 -0.11882 -0.79286 H 5.17581 0.13703 -0.75884 H -4.58168 0.03731 -2.31140 C 3.17677 0.51149 0.24327 C -3.01298 -0.26765 -0.45049 C 2.95606 0.05801 0.26479 H 3.06615 0.72504 -0.81276 H -2.71704 -0.25717 -1.49110 H 2.91802 -0.02608 -0.81163 C -0.17669 -4.61494 -0.33637 C 0.03198 4.53544 -0.15481 C -0.22072 -4.25489 0.18997 H -0.22065 -5.45166 0.34528 H 0.05697 5.35417 0.54942 H -0.44102 -4.99620 0.94302 C -0.30596 -4.78413 -1.71067 C 0.08115 4.75164 -1.52814 C 0.03856 -4.60906 -1.13000 H -0.45913 -5.76994 -2.12931 H 0.15109 5.75660 -1.92266 H 0.01910 -5.64744 -1.43067 C 4.40543 0.62853 0.87453 C -4.33904 -0.42184 -0.07535 C 4.15548 0.18381 0.94304 H 5.27947 0.93140 0.31685 H -5.10577 -0.53103 -0.82801 H 5.08485 0.19182 0.39409 C 4.46667 0.35465 2.23743 C -4.64031 -0.44175 1.28170 C 4.12398 0.30924 2.32818 H 5.40188 0.44769 2.77337 H -5.66115 -0.57253 1.61466 H 5.04254 0.42359 2.88690 C 0.92882 -2.60203 -5.48789 C -1.13744 2.58964 -5.31946 C 1.86896 -2.65475 -4.97009 C 0.94117 -3.38328 -6.63492 C -1.21355 3.41067 -6.43587 C 2.07810 -3.45797 -6.08196 C -0.01298 -3.18297 -7.62706 C -0.26824 3.29941 -7.45035 C 1.40169 -3.19534 -7.26851 C -0.97377 -2.19165 -7.46509 C 0.74642 2.35544 -7.34168 C 0.52233 -2.12068 -7.33775 C -0.98768 -1.41151 -6.31650 C 0.82393 1.53573 -6.22333 C 0.31063 -1.31662 -6.22644 C 2.16350 -2.20196 5.15449 C -2.66890 0.95812 5.21240 C 1.09002 -0.44754 5.81122 C 2.84822 -2.59659 6.29511 C -3.46877 1.01498 6.34499 C 1.55181 -0.35442 7.11628 C 2.44729 -2.13647 7.54506 C -3.18485 0.20831 7.44198 C 1.03635 0.61470 7.97055 C 0.67493 -0.88044 6.50075 C -1.29428 -0.71105 6.26224 C -0.41146 1.39248 6.20832 C 1.35657 -1.28040 7.64240 C -2.09233 -0.65044 7.39665 C 0.05170 1.48345 7.51365 H 1.68456 -2.74533 -4.72726 H -1.88516 2.66474 -4.54124 H 2.42829 -2.83782 -4.06211 H 1.70275 -4.14173 -6.75974 H -2.01742 4.13019 -6.51945 H 2.78055 -4.27856 -6.02778 H -1.71508 -2.02399 -8.23497 H 1.48086 2.25562 -8.12968 H -0.00042 -1.90623 -8.25962 H -1.73879 -0.64338 -6.19517 H 1.61733 0.80560 -6.14372 H -0.37902 -0.48661 -6.28902 H 2.46159 -2.58437 4.18758 H -2.87396 1.61095 4.37476 H 1.46487 -1.22821 5.16251 H 3.68664 -3.27528 6.20983 H -4.30594 1.69988 6.37669 H 2.30219 -1.04764 7.47156 H -0.17025 -0.21194 6.58557 H -0.44999 -1.38534 6.23426 H -1.17415 2.07563 5.86174 H 1.03328 -0.92212 8.61058 H -1.85966 -1.27722 8.24708 H -0.35827 2.23436 8.17477 H 2.97558 -2.44950 8.43551 H -0.32746 3.93712 -8.32202 H 1.39350 0.68586 8.98856 H -0.00239 -3.78948 -8.52258 H -3.80509 0.25482 8.32687 H 1.56650 -3.81839 -8.13659

James McPherson – June 2019 8-323 Chapter 8: Appendices

II COOEt,COOEt II COOEt,COOEt III COOEt,COOEt + Co (dpp )2[HS] Co (dpp )2[LS] Co (dpp )2 Co 0.00000 0.00000 0.00000 Co 0.00000 0.00000 0.00000 Co 0.00000 0.00000 0.00000 N -0.16316 -0.23073 1.95928 N -0.01282 -0.06891 1.83750 N -0.01592 -0.00232 -1.82385 N -2.24820 -0.08223 0.47334 N -2.04024 -0.07082 0.41280 N -0.00954 1.95864 -0.36245 O -1.60235 -0.21336 6.17310 O -1.14208 0.22190 6.16180 O -1.41367 1.18011 -5.89434 O 0.66963 -1.84518 6.03620 O 1.13773 -1.36284 5.94884 O 0.16290 -1.12201 -6.15189 O -3.00328 -1.52559 5.00790 O -2.75526 -0.88644 5.06067 O -0.05260 2.76768 -5.06503 N 2.07500 0.13491 0.79912 N 2.03966 0.05822 0.45112 N -0.00954 -1.95951 -0.36245 C -2.50924 -0.41707 1.76227 C -2.32757 -0.21094 1.74166 C -0.05528 2.28959 -1.69579 C -1.31464 -0.48375 2.60909 C -1.12775 -0.22488 2.57197 C -0.09953 1.11776 -2.55640 C 0.88657 -0.27449 2.79995 C 1.09602 -0.07743 2.59769 C -0.14766 -1.11814 -2.55419 -1.01352 -0.70373 3.96953 -0.73559 -0.33358 3.92690 -0.29160 0.71879 -3.89729 C 2.17354 -0.04276 2.14287 C 2.30691 0.02248 1.79065 C -0.11017 -2.29101 -1.69327 C 0.40566 -0.58328 4.09182 C 0.69333 -0.25383 3.94020 C -0.30441 -0.71154 -3.89974 C -3.81616 -0.68332 2.17524 C -3.64655 -0.33539 2.17361 C -0.03974 3.62459 -2.08600 C -3.99832 -0.98419 3.19438 C -3.84088 -0.48062 3.22440 C -0.04925 3.85347 -3.14074 H -1.97414 -0.88564 5.06612 H -1.64811 -0.38623 5.07212 H -0.54190 1.65948 -5.00629 C 1.20905 -0.86039 5.28688 C 1.60374 -0.45184 5.07212 C -0.36431 -1.64171 -5.04208 C -3.24199 0.00149 -0.40392 C -3.02987 -0.03441 -0.47412 C 0.04178 2.91616 0.56317 C -4.56665 -0.23012 -0.06191 C -4.36461 -0.13287 -0.10933 C 0.04972 4.26002 0.22977 C -4.84651 -0.58016 1.25341 C -4.66842 -0.28912 1.23728 C 0.00957 4.61212 -1.11417 C 3.41466 0.03330 2.77915 C 3.61832 0.09905 2.25653 C -0.18046 -3.62603 -2.07638 C 3.46373 -0.07989 3.85022 C 3.79701 0.10100 3.32022 C -0.29539 -3.85576 -3.12426 H 3.16771 0.38262 0.08360 H 3.04232 0.15831 -0.41636 H 0.03729 -2.91658 0.56375 C 4.54676 0.28191 2.01999 C 4.65377 0.19534 1.33921 C -0.12526 -4.61320 -1.10447 C -2.40747 -0.41821 7.34812 C -1.92596 0.13013 7.36564 C -1.68796 2.00711 -7.05262 C -2.45380 -1.48767 7.55787 C -2.16105 -0.91802 7.55298 C -0.74890 2.18010 -7.57804 H -3.42361 -0.07793 7.14599 H -2.86900 0.65677 7.21355 H -2.06255 2.97098 -6.70956 H 4.43246 0.46229 0.64626 H 4.37016 0.22507 -0.02055 H -0.01119 -4.26070 0.23486 C -1.76452 0.35384 8.47739 C -1.11414 0.73857 8.48618 C -2.69535 1.27171 -7.90319 C -0.75060 -0.00087 8.66263 C -0.18017 0.19581 8.62976 C -2.30408 0.30830 -8.22924 H -2.34651 0.22756 9.39131 H -1.68127 0.69834 9.41714 H -2.93007 1.86356 -8.78816 H -1.71685 1.41754 8.24444 H -0.87512 1.78065 8.27407 H -3.62004 1.10027 -7.35261 H 2.25409 -0.31522 5.57865 H 2.68426 0.08879 5.19608 H -0.80544 -2.76932 -4.95926 O -0.08073 -0.03300 -1.97755 O 0.01829 0.01362 -1.88031 O -0.01531 -0.00257 1.82252 N 0.07434 -2.21477 -0.63713 N 0.00589 -2.15460 -0.53774 N 1.95226 0.02049 0.37785 N 0.27099 -1.16193 -6.27855 N 1.14869 -1.12296 -6.01916 N 1.10345 -0.76219 6.09626 O -1.44218 1.04352 -6.09248 O -0.64009 1.02937 -6.18246 O -1.24885 0.74716 6.05290 O -1.05030 -2.68826 -5.29621 O -0.35650 -2.64199 -5.32954 O 2.73577 0.36693 5.04119 O 0.07434 2.12792 -0.63713 O 0.00589 2.17751 -0.53774 O -1.96375 -0.02495 0.34784 N -0.16671 -2.39289 -1.96066 N -0.03668 -2.34646 -1.88068 N 2.27514 0.04792 1.71445 C -0.23218 -1.14287 -2.72362 C -0.00273 -1.10107 -2.64319 C 1.09680 0.02821 2.56924 C -0.11814 1.07326 -2.74292 C 0.09199 1.12303 -2.64388 C -1.13871 -0.05475 2.55071

James McPherson – June 2019 8-324 Chapter 8: Appendices

C -0.37254 -0.75177 -4.07200 C 0.07118 -0.70676 -3.99446 C 0.68274 -0.00930 3.92150 C -0.00973 2.31746 -1.98038 C 0.08778 2.36958 -1.87966 C -2.30427 -0.07160 1.67935 C -0.31061 0.67624 -4.08476 C 0.11846 0.71909 -3.99918 C -0.74746 -0.04503 3.90901 C -0.35183 -3.67489 -2.48194 C -0.12751 -3.63385 -2.41420 C 3.60772 0.10222 2.10879 C -0.44833 -1.63611 -5.24272 C 0.22580 -1.59226 -5.15589 C 1.61203 -0.09116 5.06311 C -0.53239 1.55555 -5.23726 C 0.07463 1.58530 -5.18047 C -1.70286 0.01869 5.03186 C 0.14148 -3.26655 0.17117 C -0.02628 -3.20016 0.28019 C 2.91400 0.03145 -0.54438 C 0.32263 -3.04873 1.21693 C 0.00473 -2.97393 1.33931 C 2.58721 0.01204 -1.57393 H -0.01447 -4.57015 -0.27813 H -0.09940 -4.50837 -0.17683 H 4.25596 0.06889 -0.20544 C 0.05229 -5.39747 0.41338 C -0.12140 -5.33050 0.52383 C 5.00498 0.07490 -0.98292 H -0.26734 -4.76562 -1.63035 H -0.15302 -4.71727 -1.54997 H 4.60019 0.10752 1.14043 C 0.03495 3.60551 -2.51880 C 0.17946 3.65791 -2.41085 C -3.64125 -0.14581 2.05543 C 0.19804 3.17622 0.17126 C 0.00694 3.22474 0.27842 C -2.91449 -0.03676 -0.58634 C 0.26797 2.95180 1.22859 C -0.05287 2.99956 1.33656 C -2.58001 -0.00356 -1.61301 H 0.15298 4.68966 -1.66428 H 0.17037 4.74257 -1.54777 H -4.62151 -0.15257 1.07510 C 0.15690 -1.88329 -7.51873 C 1.30356 -1.85121 -7.25067 C 1.91799 -0.82577 7.29401 C -0.89815 -1.95068 -7.78729 C 0.33858 -1.88896 -7.75818 C 2.21844 0.18751 7.55819 H 0.52555 -2.89926 -7.37356 H 1.59546 -2.87669 -7.02226 H 2.81987 -1.39411 7.06725 H 0.23665 4.48247 -0.29207 H 0.08198 4.53380 -0.17669 H -4.26015 -0.09466 -0.26538 C 0.33511 5.30423 0.40219 C 0.07791 5.35598 0.52428 C -4.99814 -0.10238 -1.05325 H 0.95882 -1.13147 -8.55598 H 2.35085 -1.13406 -8.07139 H 1.08837 -1.47683 8.37444 C 0.57443 -0.11980 -8.68618 C 2.04094 -0.11297 -8.29384 C 0.18998 -0.89570 8.57938 H 0.90121 -1.64840 -9.51482 H 2.50841 -1.66000 -9.01397 H 1.67111 -1.54001 9.29367 H 2.00683 -1.06381 -8.26433 H 3.30065 -1.09360 -7.53815 H 0.78989 -2.48510 8.08865 H -0.00445 2.63474 -5.41350 H 0.58169 2.68471 -5.27226 H -2.80259 -0.49291 4.99448 O -1.67545 1.77548 -7.31101 O -0.68304 1.74816 -7.42996 O -2.10825 0.86498 7.21559 C -0.73371 2.20300 -7.65157 C 0.28155 2.22476 -7.59699 C -2.62960 -0.07870 7.35989 H -2.00716 1.01950 -8.02102 H -0.84058 0.97616 -8.18164 H -1.41701 1.02817 8.03937 H -2.72801 2.84996 -7.12114 H -1.80709 2.76528 -7.44113 H -3.07502 2.02004 7.06418 C -3.65485 2.41815 -6.74266 C -2.76262 2.28392 -7.23134 C -2.53991 2.95385 6.89219 H -2.94044 3.33317 -8.07646 H -1.86965 3.24008 -8.42193 H -3.66265 2.12700 7.97691 H -2.37969 3.60834 -6.42202 H -1.63014 3.53884 -6.69575 H -3.76095 1.84515 6.23678 H 1.31799 -2.14412 7.28718 H 1.91281 -1.57168 7.14508 H 0.13048 -1.93955 -7.34967 C 0.52121 -2.53403 7.91877 C 1.18335 -1.89710 7.88529 C 0.13890 -1.21635 -8.16212 H 1.70051 -1.21969 7.71681 H 2.34400 -0.62149 7.45587 H -0.80517 -2.49443 -7.36651 H 2.42463 -3.16451 7.10735 H 2.98766 -2.61976 6.93629 H 1.32885 -2.86271 -7.41424 C 2.03927 -4.07178 6.64140 C 2.54982 -3.55816 6.59514 C 2.25795 -2.29530 -7.36168 H 2.84488 -3.43043 8.07893 H 3.50923 -2.80717 7.87660 H 1.31687 -3.41150 -8.35688 H 3.22222 -2.75964 6.48660 H 3.71537 -2.28045 6.20102 H 1.30572 -3.58490 -6.59997 H -2.95813 0.26038 -1.41691 H -2.73170 0.07382 -1.50862 H 0.07879 2.58613 1.59118 H -5.34536 -0.14576 -0.80595 H -5.13596 -0.09560 -0.86466 H 0.09242 5.00503 1.00994 H -5.86376 -0.78255 1.56149 H -5.69715 -0.38131 1.55911 H 0.02392 5.65317 -1.40591 H -0.57981 -3.79252 -3.52905 H -0.19681 -3.76186 -3.48266 H 3.83003 0.15423 3.16346

James McPherson – June 2019 8-325 Chapter 8: Appendices

H -0.40887 -5.76417 -2.02224 H -0.22376 -5.72065 -1.94852 H 5.63928 0.14779 1.43665 H 0.18864 5.69210 -2.06988 H 0.24068 5.74616 -1.94591 H -5.66354 -0.20968 1.35772 H -0.00040 3.72906 -3.58923 H 0.27760 3.78416 -3.47698 H -3.87777 -0.21589 3.10582 H 3.01510 0.52202 -0.97963 H 2.76160 0.18754 -1.46103 H 0.11048 -2.58638 1.58976 H 5.29311 0.66445 0.02550 H 5.15175 0.30520 -0.76185 H 0.03079 -5.00475 1.01592 H 5.51349 0.34392 2.50190 H 5.67640 0.25535 1.68688 H -0.17666 -5.65403 -1.39257 H H H

II COOMe,Py II COOMe,Py III COOMe,Py + Co (dpp )2[HS] Co (dpp )2[LS] Co (dpp )2 Co 0.00000 0.00000 0.00000 Co 0.00000 0.00000 0.00000 Co 0.00000 0.00000 0.00000 N 0.03911 2.26032 -0.62985 N -2.03470 -0.01773 0.41101 N -0.10525 1.95921 -0.35359 N 0.01617 0.02900 -1.97068 N 0.00337 0.02467 1.83586 N -0.01018 -0.00448 -1.82304 N 2.13928 -0.00358 0.63017 N 0.00215 -2.16122 -0.55563 N 1.94968 0.09772 0.38209 N 0.03911 -2.12086 -0.62985 N 2.01034 -0.00789 0.42601 N 0.12214 -1.94582 -0.36929 O 0.65276 -2.65486 -5.22870 O 2.75175 0.48732 4.96648 O -0.25221 -2.79945 -4.91225 N 0.01694 0.35403 1.95245 N -0.00733 0.00029 -1.88844 N -0.01018 -0.00448 1.82290 C 0.08671 -1.06432 -2.74762 C 1.10369 0.04223 2.59737 C 0.02836 -1.11054 -2.57241 N -2.18358 -0.11237 0.70583 N 0.00215 2.20462 -0.55563 N -1.96314 -0.13775 0.34753 C 0.05895 1.15441 -2.71977 C -1.12565 0.04656 2.57447 C -0.12121 1.12157 -2.54897 O 0.84348 1.64277 6.22021 O -0.57734 -0.84208 -6.29510 O 0.89562 0.07261 6.24944 C 0.14178 0.79541 -4.06937 C -0.76082 0.07008 3.93259 C -0.14388 0.74961 -3.90774 O 2.77873 1.27780 5.15079 O 0.15285 -2.70023 -5.27519 O 2.76878 -0.05669 5.01801 C 0.16646 -0.63099 -4.10587 C 0.66971 0.06938 3.96174 C -0.04794 -0.67791 -3.93167 C 0.05213 2.40916 -1.97674 C -2.31889 -0.01800 1.74810 C -0.14458 2.28687 -1.68743 C 0.71059 0.93157 4.00234 C -0.06867 -0.67588 -4.02311 C 0.66305 -0.08115 3.92891 C 0.03730 -2.31230 -1.97722 C 2.30544 -0.01151 1.76408 C 0.13263 -2.28025 -1.70500 C 1.11697 0.57897 2.67703 C -0.00693 -1.09780 -2.66047 C 1.09278 0.00332 2.57281 C 2.35447 0.39268 1.91335 C 0.01360 -2.35027 -1.90130 C 2.27257 0.07679 1.72034 C 1.56298 1.28394 5.13244 C -0.13671 -1.52137 -5.21159 C 1.55742 -0.02734 5.09112 O 0.13034 -0.85875 -6.46577 O 1.01305 0.05605 6.31697 O -0.04425 -1.03113 -6.28136 C -2.34296 0.25568 2.00339 C 0.00168 2.37218 -1.90145 C -2.29575 -0.19357 1.68009 3.65552 0.58023 2.38406 0.04970 -3.64059 -2.43475 3.60583 0.11571 2.11411 C -0.71760 0.88851 4.00142 C -0.10002 0.75119 -4.00023 C -0.76639 -0.15164 3.90643 0.04809 3.33284 0.15394 -3.02662 -0.09038 -0.47250 -0.10638 2.91898 0.57231 C 0.03988 3.13793 1.22001 C -2.72962 -0.08794 -1.51297 C -0.07211 2.58840 1.60031 -1.09520 0.52341 2.70135 -0.05367 1.12384 -2.64811 -1.13955 -0.10040 2.54846 C C C -0.01130 -3.16420 0.19140 2.99026 -0.06165 -0.47095 0.21164 -2.89135 0.56517 H 3.16979 -0.21701 -0.18058 H 0.02193 -3.20843 0.25958 H 2.91035 0.16128 -0.53863 2.92082 -0.52866 -1.18785 0.01456 -2.98216 1.31937 2.58385 0.17310 -1.56849 -0.03170 -3.60799 -2.49537 3.63365 -0.09102 2.18043 0.24870 -3.61741 -2.07234

James McPherson – June 2019 8-326 Chapter 8: Appendices

C -0.03486 -3.74706 -3.56134 C 3.84954 -0.10106 3.23351 C 0.25750 -3.86248 -3.11965 0.06977 3.68380 -2.55167 -3.63931 -0.10572 2.18982 -0.17918 3.62420 -2.07425 C 0.07728 3.79307 -3.62458 C -3.84923 -0.11725 3.24768 C -0.20388 3.86804 -3.12503 -0.07402 -4.47472 -0.25671 4.32901 -0.13022 -0.11675 0.32325 -4.23382 0.24781 C 0.29365 -1.35099 -5.37195 C 1.44406 0.18466 5.19298 C -0.10441 -1.47682 -5.16266 0.23125 1.73015 -5.20822 -1.70911 0.14215 5.05653 -0.28907 1.69798 -5.02625 H H H -3.62127 0.33960 2.56582 0.07298 3.65357 -2.45817 -3.62698 -0.34813 2.05902 C C C -3.73571 0.63134 3.59784 0.08544 3.77770 -3.52921 -3.87348 -0.39708 3.10864 H H H 4.72197 0.35568 1.52742 0.06387 -4.72579 -1.57188 4.59718 0.18230 1.14641 C C C -1.64422 1.13617 5.12321 -0.12801 1.67375 -5.14925 -1.70838 -0.27541 5.03214 H H H -0.08755 -4.68779 -1.62956 4.64563 -0.15033 1.23506 0.34507 -4.59368 -1.09345 C C C 4.48744 -0.05205 0.21983 0.04980 -4.51847 -0.19817 4.25236 0.20653 -0.19931 C C C 5.29575 -0.23788 -0.47243 0.06278 -5.34183 0.50138 5.00018 0.25774 -0.97629 C C C -3.25153 -0.39237 -0.03559 0.06138 3.26591 0.24085 -2.91398 -0.22319 -0.58384 C C C -3.05196 -0.67707 -1.06172 0.06206 3.05805 1.30451 -2.58188 -0.17505 -1.61081 H H H 0.07107 4.62780 -0.34418 -4.35923 -0.17093 -0.09808 -0.14491 4.26181 0.24047 C C C 0.08317 5.47271 0.32880 -5.13136 -0.22903 -0.85101 -0.14370 5.00809 1.02041 C C C 0.08092 4.79501 -1.72496 -4.66206 -0.18243 1.25898 -0.17845 4.61330 -1.10525 C C C 0.09834 5.78674 -2.15701 -5.68889 -0.25375 1.59163 -0.20152 5.65386 -1.39801 C C C 0.81229 -3.38277 -6.45294 3.56608 0.61531 6.13724 -0.32375 -3.64445 -6.07608 H H H -0.11497 -3.38589 -7.02354 3.18622 1.40313 6.78517 -1.14857 -3.33896 -6.71550 C C C 1.08542 -4.39284 -6.15974 4.55947 0.86815 5.77624 -0.48593 -4.64911 -5.69745 H H H 1.59776 -2.93846 -7.06175 3.58826 -0.31979 6.69501 0.60542 -3.59115 -6.63975 C C C -4.54697 -0.33084 0.45496 0.12417 4.56659 -0.23880 -4.25089 -0.37320 -0.25980 H H H -5.38518 -0.57019 -0.18299 0.17079 5.40113 0.44568 -4.98921 -0.44062 -1.04438 C C C -4.72403 0.04307 1.78397 0.13251 4.75233 -1.61740 -4.60599 -0.43928 1.08386 H H H -5.71755 0.10285 2.20812 0.18914 5.74835 -2.03609 -5.64145 -0.56259 1.36951 C C C 1.59366 1.98984 7.38686 -0.62060 -1.57520 -7.52240 1.69955 0.13030 7.44121 H H H 0.85426 2.20639 8.15233 -0.98403 -0.87296 -8.26680 0.99210 0.16297 8.26306 H H H 2.21503 2.86433 7.19814 -1.29388 -2.42703 -7.43851 2.32164 1.02308 7.43293 H H H 2.23297 1.16371 7.69460 0.37168 -1.93580 -7.79026 2.33553 -0.74953 7.51244 C C C -0.90954 2.20132 -5.85340 -2.61539 1.19995 5.15021 -1.40316 2.53585 -5.09156 H H H -0.77558 3.12851 -6.87663 -3.54166 1.21800 6.18460 -1.49197 3.46925 -6.11680 C C C 0.39778 3.61144 -7.28922 -3.62682 0.26763 7.11491 -0.55816 3.61935 -7.05393 H H H 1.48865 3.15446 -6.67420 -2.75405 -0.73936 7.02908 0.50206 2.81193 -6.99477 C C C 1.45826 2.22732 -5.64089 -1.79019 -0.84533 6.03710 0.68085 1.84471 -6.01485 H H H -2.30391 2.35482 5.26695 -1.24721 2.45705 -5.42738 -2.66019 0.71290 5.27824 H H H -3.19099 2.52747 6.32166 -1.21772 3.32931 -6.50765 -3.55976 0.54166 6.32414 H H H -3.46216 1.58221 7.22221 -0.16575 3.47292 -7.31462 -3.57163 -0.52978 7.11472 C C C -1.92950 0.14586 6.06142 0.97289 1.82445 -5.99193 -1.71796 -1.39516 5.86202 C C C -2.83260 0.41422 7.08022 0.90538 2.72284 -7.04712 -2.65985 -1.47422 6.87823 N N N -1.88700 1.84347 -5.56211 -2.59370 2.00223 4.42498 -2.19692 2.45413 -4.36084 C C C -1.65476 3.50313 -7.38926 -4.24825 2.03691 6.26746 -2.35377 4.12364 -6.18554

James McPherson – June 2019 8-327 Chapter 8: Appendices

C 2.43655 3.55174 -7.02073 C -2.82860 -1.50305 7.79523 C 1.24768 2.94262 -7.77031 C 2.37612 1.89698 -5.17425 C -1.10403 -1.67863 6.02996 C 1.55469 1.21102 -6.02735 C -2.12176 3.15898 4.56730 C -2.13258 2.38009 -4.81119 C -2.68990 1.61043 4.67514 N -3.70800 3.47249 6.44859 N -2.08411 3.94051 -6.73647 N -4.30136 1.30380 6.53527 C -1.44581 -0.81899 5.99760 C 1.86957 1.24388 -5.82424 C -1.00110 -2.19131 5.71916 C -3.06188 -0.34645 7.81857 C 1.75505 2.84827 -7.70948 C -2.68378 -2.33845 7.53209 H -0.00346 -2.93130 1.24912 H 2.67820 -0.05558 -1.50678 H 0.19611 -2.55155 1.59048 H -0.14279 -5.69258 -2.02739 H 5.67696 -0.21283 1.55581 H 0.43707 -5.63256 -1.37869 H -0.11375 -5.29266 0.44771 H 5.09096 -0.17185 -0.88106 H 0.39414 -4.96835 1.03560 H 3.79915 0.89193 3.40562 H 0.07761 -3.76413 -3.50491 H 3.82613 0.08935 3.16993 H 5.73371 0.49893 1.88334 H 0.09000 -5.72991 -1.97415 H 5.63605 0.21438 1.44451 H H H H H H H H H H H H H H H H H H H H H H H H

II COOMe,Me II COOMe,Me III COOMe,Me + Co (dpp )2[HS] Co (dpp )2[LS] Co (dpp )2 Co 0.00000 0.00000 0.00000 Co 0.00000 0.00000 0.00000 Co 0.00000 0.00000 0.00000 N -0.31136 2.23498 -0.54529 N -0.01234 2.20019 -0.54547 N 1.95801 0.00802 0.36403 N -0.15444 0.08170 -1.96954 N -0.02197 0.00077 -1.88596 N -0.01112 -0.01870 1.82337 N 2.06477 0.14765 0.75848 N 2.02348 0.01053 0.46063 N -0.01539 1.95175 -0.36182 N 0.33319 -2.06695 -0.72343 N -0.01234 -2.14376 -0.54547 N -1.95145 -0.01341 0.35644 O 0.39652 -2.36796 -5.37198 O 0.31985 -2.66131 -5.15449 O -2.79439 -0.50024 4.91620 N -0.15444 0.08170 1.97280 N -0.02199 0.00034 1.83683 N -0.00947 0.00083 -1.82358 C -0.08187 -0.97996 -2.78280 C -0.06968 -1.09585 -2.65843 C -1.12062 -0.09497 2.56070 N -2.24093 -0.31705 0.51338 N -2.03462 -0.00519 0.38745 N -0.01539 -1.96151 -0.36182 C -0.47258 1.19751 -2.66552 C -0.10980 1.12531 -2.64068 C 1.11772 -0.06335 2.56031 O 0.37885 0.79072 6.42826 O 0.78299 -0.08397 6.30512 O -0.41880 0.95303 -6.23199 C -0.63435 0.86495 -4.02126 C -0.22888 0.76126 -3.99280 C 0.73811 -0.16526 3.91354 O 2.35464 0.34663 5.47289 O 2.69394 0.03183 5.13780 O -0.35983 2.80706 -4.96715 C -0.37695 -0.53482 -4.10963 C -0.19652 -0.66178 -4.01832 C -0.69050 -0.18341 3.91997 C -0.58121 2.40654 -1.86592 C -0.09703 2.36711 -1.89078 C 2.28222 -0.01885 1.70019 C 0.37165 0.27872 4.13995 C 0.61829 0.01163 3.96867 C -0.19584 0.68729 -3.91635 C 0.16378 -2.23259 -2.06359 C -0.09559 -2.34087 -1.88832 C -2.28995 -0.06449 1.68982 C 0.88272 0.21189 2.80518 C 1.06445 0.00895 2.60808 C -0.08566 1.10864 -2.55867 C 2.17319 0.26000 2.11082 C 2.28442 0.01243 1.80496 C -0.06354 2.28359 -1.69686

James McPherson – June 2019 8-328 Chapter 8: Appendices

C 1.14842 0.45642 5.35942 C 1.47768 -0.00668 5.14011 C -0.32672 1.59641 -5.05586 O -1.01619 -0.94016 -6.36842 O -0.72219 -1.04840 -6.30964 O -1.02126 -0.20894 6.26637 C -2.50652 -0.09874 1.82779 C -2.34140 -0.02054 1.72147 C -0.06736 -2.29042 -1.69687 C 3.43139 0.39485 2.70380 C 3.59925 0.01853 2.26797 C -0.08267 3.62161 -2.07775 H 3.49478 0.46641 3.77684 H 3.77074 0.02301 3.33219 H -0.12616 3.85037 -3.13100 C -1.04900 0.15241 4.03479 C -0.81228 0.01433 3.93218 C -0.18469 -0.74362 -3.91535 C -0.39835 3.26465 0.28894 C 0.00190 3.26040 0.25269 C 2.91723 0.05103 -0.56114 C -1.32093 0.04316 2.66168 C -1.16278 0.00110 2.57029 C -0.07546 -1.12710 -2.56254 C 0.54860 -3.11920 0.05825 C -0.04369 -3.17799 0.28535 C -2.89786 0.02505 -0.58061 C 3.15582 0.16097 0.00011 C 3.03023 0.01024 -0.40646 C 0.01403 2.90521 0.56772 H 2.99033 0.05953 -1.06562 H 2.74947 0.00587 -1.45155 H 0.04858 2.56918 1.59395 C 0.20163 -3.51734 -2.61376 C -0.23024 -3.63814 -2.39232 C -3.63353 -0.07132 2.05326 H 0.06840 -3.63978 -3.67365 H -0.29470 -3.78728 -3.45484 H -3.88098 -0.11153 3.09952 C -0.94712 3.66853 -2.35002 C -0.17506 3.65368 -2.43875 C 3.62245 0.00775 2.08276 H -1.17188 3.80073 -3.39607 H -0.25085 3.78270 -3.50643 H 3.87445 0.00316 3.13131 C 0.60000 -4.41817 -0.42365 C -0.17110 -4.48980 -0.14748 C -4.24604 0.01865 -0.26748 C -0.39837 -1.26273 -5.37562 C -0.25288 -1.43033 -5.25786 C -1.47722 -0.28713 5.15004 C -0.96578 1.79807 -5.14245 C -0.32873 1.69277 -5.15931 C 1.63904 -0.25412 5.10091 H -2.00080 2.14544 -5.07342 H -1.26770 2.25339 -5.13503 H 2.65785 -0.49636 4.80607 H -0.85827 1.30758 -6.10412 H -0.30221 1.14843 -6.09697 H 1.29018 -1.00932 5.80153 H -0.32400 2.68142 -5.12442 H 0.48432 2.42248 -5.15343 H 1.65588 0.68619 5.65483 C -3.83317 0.01052 2.26291 C -3.67818 -0.07062 2.12510 C -0.12254 -3.63320 -2.06837 H -4.04817 0.22821 3.29642 H -3.92360 -0.09815 3.17441 H -0.16770 -3.89467 -3.11339 C 4.56008 0.41720 1.90054 C 4.63879 0.02085 1.35017 C -0.04932 4.60588 -1.10140 H 5.53714 0.52356 2.35351 H 5.66278 0.02698 1.69971 H -0.06150 5.64778 -1.39050 C -2.04936 0.07020 5.14543 C -1.75094 0.05006 5.09677 C -0.26246 -1.65386 -5.09858 H -1.61973 -0.38981 6.03046 H -1.70501 -0.87169 5.67936 H -1.19354 -1.51436 -5.64764 H -2.91250 -0.52123 4.84438 H -2.77776 0.19339 4.77106 H -0.19725 -2.69680 -4.80008 H -2.40705 1.05855 5.44755 H -1.49934 0.85924 5.78106 H 0.54419 -1.45203 -5.80258 C 0.41741 -4.60795 -1.78842 C -0.26945 -4.71019 -1.51608 C -4.61101 -0.02847 1.07199 H 0.44258 -5.60399 -2.21073 H -0.37604 -5.71528 -1.90280 H -5.65494 -0.03116 1.35383 C 4.43338 0.29693 0.52192 C 4.36014 0.01489 -0.01057 C -0.00100 4.25088 0.24104 H 5.29278 0.30489 -0.13271 H 5.14606 0.01513 -0.75173 H 0.02443 4.99307 1.02467 C -3.23645 -0.44618 -0.35591 C -3.00778 -0.03159 -0.51889 C -0.01321 -2.91621 0.56931 C -0.74752 4.54303 -0.12294 C -0.06676 4.56376 -0.22021 C 4.26068 0.06995 -0.23071 C -1.02450 4.73660 -1.47226 C -0.15791 4.75179 -1.59532 C 4.61170 0.04900 1.11537 C 0.42317 -3.11088 -6.59522 C 0.29617 -3.45259 -6.34685 C -3.61987 -0.62409 6.08813 H -0.57636 -3.44716 -6.86761 H -0.72545 -3.60226 -6.69323 H -3.25691 -1.42772 6.72472 H 1.07588 -3.95950 -6.40897 H 0.75279 -4.40160 -6.07878 H -4.61672 -0.84880 5.72072 H 0.81872 -2.50008 -7.40523 H 0.86647 -2.96907 -7.13847 H -3.61963 0.30670 6.65186 C -4.57273 -0.37558 0.01128 C -4.35081 -0.07480 -0.17808 C -0.06577 -4.26074 0.24823 H -5.35162 -0.49273 -0.72788 H -5.10675 -0.09586 -0.94904 H -0.06456 -5.00234 1.03249 C -4.86398 -0.13623 1.34986 C -4.68220 -0.09544 1.17179 C -0.12194 -4.61733 -1.09505

James McPherson – June 2019 8-329 Chapter 8: Appendices

H -5.89117 -0.05297 1.67996 H -5.71806 -0.13482 1.48149 H -0.16618 -5.65874 -1.38254 C 1.07874 0.96548 7.66283 C 1.57513 -0.11050 7.49535 C -0.55948 1.78802 -7.39545 H 0.31977 1.22649 8.39509 H 0.86676 -0.18271 8.31603 H -0.63856 1.10409 -8.23420 H 1.81632 1.76244 7.57928 H 2.17039 0.79733 7.58518 H 0.30981 2.43315 -7.50615 H 1.58816 0.04738 7.95296 H 2.24531 -0.96902 7.49440 H -1.45315 2.40330 -7.31459 H 0.77197 -5.24541 0.24927 H -0.19445 -5.30228 0.56418 H -4.98110 0.05213 -1.05728 H 0.67673 -2.90375 1.11216 H 0.03134 -2.93719 1.33928 H -2.55515 0.06299 -1.60439 H -2.94349 -0.60712 -1.38685 H -2.68705 -0.01997 -1.55223 H 0.02673 -2.58037 1.59548 H -0.18026 3.04827 1.32826 H 0.06964 3.05077 1.31392 H 2.58439 0.07116 -1.58890 H -0.80106 5.35294 0.58979 H -0.05078 5.39742 0.46664 H 5.00634 0.10427 -1.01042 H -1.30385 5.71579 -1.83852 H -0.21607 5.75006 -2.00882 H 5.65208 0.06906 1.40915

II Me,Me II Me,Me III Me,Me + Co (dpp )2[HS] Co (dpp )2[LS] Co (dpp )2 Co 0.00000 0.00000 0.00000 Co 0.00000 0.00000 0.00000 Co 0.00000 0.00000 0.00000 N 0.00003 0.00017 -1.96675 N 0.00007 -0.00035 -1.85543 N 0.00140 0.01680 -1.81974 N -0.00637 -2.18731 -0.63170 N -0.00821 -2.12245 -0.50306 N -0.02223 -1.96028 -0.39297 C -0.00306 -1.09974 -2.73483 C -0.00499 -1.10554 -2.62190 C -0.01573 -1.08820 -2.57546 C -0.00137 -0.70405 -4.10320 C -0.00346 -0.70598 -3.98648 C -0.01701 -0.67131 -3.93558 C -0.00689 -2.33956 -1.98843 C -0.01021 -2.32855 -1.85360 C -0.02952 -2.26261 -1.73769 C -0.01081 -3.62649 -2.54609 C -0.01703 -3.63115 -2.36647 C -0.04942 -3.59640 -2.14442 H -0.01149 -3.74504 -3.61852 H -0.01886 -3.78676 -3.43456 H -0.05562 -3.82684 -3.19909 C -0.00922 -3.25785 0.15522 C -0.01234 -3.16161 0.32211 C -0.03249 -2.93584 0.51416 H -0.00841 -3.05621 1.22016 H -0.01030 -2.92115 1.37855 H -0.02543 -2.62058 1.54804 C -0.01386 -4.73089 -1.71391 C -0.02141 -4.70514 -1.49293 C -0.06066 -4.60161 -1.19308 H -0.01692 -5.72647 -2.13844 H -0.02673 -5.71584 -1.87968 H -0.07623 -5.63755 -1.50275 C -0.01299 -4.55637 -0.33204 C -0.01896 -4.47795 -0.12002 C -0.05161 -4.27395 0.15964 H -0.01521 -5.39654 0.34673 H -0.02213 -5.29249 0.58935 H -0.05937 -5.03379 0.92615 N 0.00645 2.18741 -0.63131 N 0.00804 2.12230 -0.50394 N 0.02098 1.96750 -0.35609 C 0.00323 1.10021 -2.73463 C 0.00484 1.10452 -2.62237 C 0.01033 1.13594 -2.55440 C 0.00174 0.70477 -4.10307 C 0.00286 0.70439 -3.98677 C -0.00052 0.74510 -3.92218 C 0.00704 2.33990 -1.98801 C 0.00992 2.32785 -1.85457 C 0.02395 2.29442 -1.69497 C 0.01103 3.62693 -2.54543 C 0.01642 3.63023 -2.36800 C 0.03886 3.63555 -2.07717 H 0.01176 3.74568 -3.61784 H 0.01813 3.78539 -3.43616 H 0.04152 3.88505 -3.12749 C 0.00928 3.25780 0.15581 C 0.01203 3.16183 0.32078 C 0.03165 2.92638 0.56886 H 0.00841 3.05595 1.22071 H 0.01012 2.92181 1.37732 H 0.02824 2.59339 1.59724 C 0.01406 4.73118 -1.71304 C 0.02063 4.70460 -1.49491 C 0.05013 4.62323 -1.10776 H 0.01718 5.72683 -2.13739 H 0.02569 5.71513 -1.88209 H 0.06190 5.66470 -1.39848 C 0.01312 4.55640 -0.33121 C 0.01834 4.47798 -0.12191 C 0.04627 4.27079 0.23875 H 0.01533 5.39646 0.34772 H 0.02139 5.29282 0.58712 H 0.05443 5.01644 1.01906 N 0.00006 0.00012 1.96695 N -0.00021 -0.00006 1.85505 N -0.00248 -0.01054 1.82020

James McPherson – June 2019 8-330 Chapter 8: Appendices

N 2.18809 -0.01108 0.63206 N 2.12102 -0.00653 0.50264 N 1.96347 -0.01551 0.37695 C 1.10004 -0.00583 2.73498 C 1.10478 -0.00418 2.62172 C 1.10904 0.01201 2.56626 C 0.70453 -0.00301 4.10339 C 0.70481 -0.00396 3.98619 C 0.70405 0.06955 3.92871 C 2.33996 -0.01309 1.98880 C 2.32762 -0.00822 1.85319 C 2.27660 -0.00122 1.71905 C 3.62675 -0.02204 2.54679 C 3.63038 -0.01378 2.36557 C 3.61378 0.00292 2.11503 H 3.74505 -0.02408 3.61924 H 3.78633 -0.01564 3.43362 H 3.85267 0.01400 3.16776 C 3.25882 -0.01726 -0.15457 C 3.15991 -0.00954 -0.32290 C 2.93179 -0.02780 -0.53798 H 3.05749 -0.01514 -1.21956 H 2.91906 -0.00783 -1.37924 H 2.60868 -0.03868 -1.56942 C 4.73136 -0.02840 1.71491 C 4.70408 -0.01695 1.49164 C 4.61141 -0.00869 1.15579 H 5.72680 -0.03534 2.13972 H 5.71491 -0.02122 1.87803 H 5.64991 -0.00583 1.45716 C 4.55720 -0.02595 0.33300 C 4.47639 -0.01466 0.11882 C 4.27282 -0.02501 -0.19418 H 5.39754 -0.03068 -0.34557 H 5.29068 -0.01694 -0.59084 H 5.02652 -0.03487 -0.96670 N -2.18805 0.01118 0.63220 N -2.12125 0.00633 0.50228 N -1.96446 0.02726 0.37190 C -1.09986 0.00654 2.73505 C -1.10533 0.00301 2.62153 C -1.11514 0.03762 2.56336 C -0.70427 0.00447 4.10343 C -0.70560 0.00112 3.98607 C -0.71246 0.08592 3.92685 C -2.33983 0.01363 1.98895 C -2.32806 0.00714 1.85280 C -2.28050 0.05114 1.71317 C -3.62658 0.02284 2.54702 C -3.63089 0.01182 2.36499 C -3.61818 0.08856 2.10576 H -3.74481 0.02523 3.61947 H -3.78700 0.01294 3.43302 H -3.85928 0.10757 3.15787 C -3.25882 0.01718 -0.15437 C -3.16002 0.00946 -0.32341 C -2.93055 0.03547 -0.54544 H -3.05757 0.01472 -1.21937 H -2.91901 0.00851 -1.37971 H -2.60541 0.01473 -1.57609 C -4.73124 0.02902 1.71521 C -4.70446 0.01507 1.49090 C -4.61343 0.09901 1.14405 H -5.72666 0.03617 2.14007 H -5.71536 0.01867 1.87715 H -5.65229 0.12802 1.44278 C -4.55717 0.02612 0.33328 C -4.47656 0.01377 0.11812 C -4.27200 0.07073 -0.20501 H -5.39755 0.03069 -0.34523 H -5.29075 0.01616 -0.59166 H -5.02380 0.07652 -0.97941 C 1.61047 -0.00709 5.29434 C 1.61446 -0.00785 5.17367 C 1.61511 0.10451 5.11372 H 2.25280 -0.89192 5.31342 H 2.25775 -0.89204 5.18985 H 2.27352 0.97593 5.08784 H 1.04588 -0.00004 6.22564 H 1.05292 -0.00123 6.10675 H 2.25066 -0.78293 5.15751 H 2.26650 0.86769 5.30868 H 2.27046 0.86696 5.18532 H 1.05715 0.14891 6.04643 C -1.61013 0.00920 5.29444 C -1.61545 0.00355 5.17341 C -1.62556 0.14048 5.10954 H -2.25246 0.89405 5.31308 H -2.25894 0.88758 5.19042 H -2.27104 1.02128 5.07737 H -1.04548 0.00265 6.22571 H -1.05404 -0.00394 6.10656 H -1.06919 0.18186 6.04334 H -2.26617 -0.86556 5.30929 H -2.27125 -0.87142 5.18404 H -2.27421 -0.73732 5.15663 C 0.00351 1.61086 -5.29391 C 0.00649 1.61383 -5.17443 C 0.00430 1.66851 -5.09812 H -0.00085 1.04637 -6.22530 H 0.00443 1.05205 -6.10738 H -0.00671 1.12035 -6.03763 H 0.88612 2.25626 -5.31222 H 0.88852 2.26008 -5.18838 H 0.89181 2.30549 -5.10437 H -0.87353 2.26386 -5.30882 H -0.87053 2.26688 -5.18862 H -0.86848 2.32558 -5.09571 C -0.00297 -1.60992 -5.29421 C -0.00748 -1.61594 -5.17374 C -0.03200 -1.57222 -5.12872 H 0.00150 -1.04525 -6.22549 H -0.00540 -1.05459 -6.10695 H -0.03025 -1.00633 -6.05770 H -0.88557 -2.25533 -5.31275 H -0.88968 -2.26198 -5.18728 H -0.91900 -2.20989 -5.13876 H 0.87408 -2.26290 -5.30913 H 0.86937 -2.26922 -5.18774 H 0.84127 -2.22839 -5.14684

James McPherson – June 2019 8-331 Chapter 8: Appendices

James McPherson – June 2019 8-332 Chapter 8: Appendices

APPENDIX 3 – SUPPORTING INFORMATION FOR CHAPTER 4

James McPherson – June 2019 8-333 Dalton Transactions Chapter 8: Appendices Synthesis and Structural, Redox and Photophysical Properties of Tris-(2,5-di(2- pyridyl)pyrrolide) Lanthanide Complexes

James N. McPherson, a Laura Abad Galan, b Hasti Iranmanesh, a Max Massi b* and Stephen B. Colbran a* aSchool of Chemistry, University of New South Wales, NSW 2052 Australia; bDepartment of Chemistry, Curtin University, WA 6102 Australia

Electronic Supplementary Information

James McPherson – June 2019 8-334 Dalton Transactions Chapter 8: Appendices SC-XRD structures

CO Me,Me CO Me,Me Fig S1. Overlaid SC-XRD structures of [Sm(dpp 2 )3] (pink) and [Eu(dpp 2 )3] (light blue) by matching the Ln(III) centre and 9 coordinating N atoms (RMSD = 0.0119 Å).

CO Me,Me Fig S2. H-bonding network with the lattice water molecule linking neighbouring [Ln(dpp 2 )3] complex molecules (Ln = Sm, Eu) down the b axis. Key distances and angles: Ln = Sm: O2C…O1 2.90(1) Å, O1…O1B 2.91(1) Å, and O2C–O1–O1B 92.8(4); Ln = Eu: O1C′…O1 2.91(3) Å, O1…O1B 2.92(2) Å, and O1C′–O1– O1B 98.3(7)º.

James McPherson – June 2019 8-335 Dalton Transactions

Table S1. Experimental details Chapter 8: Appendices

Me,Me CO2Me,Me CO2Me,Me Complex [La(dpp )3] [Sm(dpp )3] [Eu(dpp )3] CIF name p1bar_a james_sm1_a James_Eu_June18_0m_a Crystal data

Chemical formula C48H42LaN9·1.5(C6H6) 2(C51H42N9O6Sm)·H2O C51H42EuN9O6·0.5(H2O)

Mr 1000.98 2072.58 1066.48

Crystal system, space Triclinic, P¯1 Monoclinic, P21/n Monoclinic, P21/n group Temperature (K) 100 100 99.95 a, b, c (Å) 12.928 (3), 13.980 (3), 17.177 (3), 13.222 (3), 17.3254(12), 13.3245(9), 14.100 (3) 19.402 (4) 19.4895(13) , ,  (°) 80.97 (3), 83.88 (3), 70.72 90.85 (3) 90.717(2) (3) V (Å3) 2371.5 (9) 4406.0 (15) 4498.8(5) Z 2 2 4 Radiation type Mo K Mo K Mo K  (mm-1) 0.95 1.40 1.458 Crystal size (mm) 0.02 × 0.01 × 0.01 0.02 × 0.02 × 0.02 0.1 × 0.04 × 0.01

Data collection Diffractometer ADSC Quantum 210r ADSC Quantum 210r Bruker D8 Quest Absorption correction – – SADABS No. of measured, 39500, 10228, 9722 65218, 9559, 8469 146595, 7935 independent and observed [I > 2(I)] reflections

Rint 0.018 0.029 0.1555

-1 (sin /)max (Å ) 0.658 0.641 0.595

Refinement R[F2 > 2(F2)], wR(F2), 0.024, 0.061, 1.07 0.039, 0.103, 1.14 0.0479, 0.1249, 1.076 S No. of reflections 10228 9559 7935 No. of parameters 611 651 653 No. of restraints – 11 49 H-atom treatment H-atom parameters H-atom parameters H-atom parameters constrained constrained constrained

-3 max, min (e Å ) 0.51, -1.40 0.92, -1.16 0.94, -0.85

Computer programs: BluIce (McPhillips, 2002), XDS (Kabsch, 1993), SHELXT (Sheldrick, 2015), SHELXL (Sheldrick, 2015), Olex2 (Dolomanov et al., 2009).

James McPherson – June 2019 8-336 Dalton Transactions Chapter 8: Appendices NMR Spectra

1 Me,Me Fig. S3. H NMR (400 MHz, C6D6) spectrum of [La( dpp)3].

1 Me,Me Fig. S4. H NMR (400 MHz, C6D6) spectra of [La(dpp )3]: after 20 minutes in solution (bottom) and after 6 hours in solution (top). The signals, which correspond to unbound dppMe,MeH—and increase over time— have been highlighted by grey regions.

James McPherson – June 2019 8-337 Dalton Transactions Chapter 8: Appendices

1 1 Me,Me Fig S5. H- H COSY NMR (400 MHz, C6D6) spectrum of [La( dpp)3].

1 1 Me,Me Fig S6. H- H nOeSY NMR (400 MHz, C6D6) spectrum of [La( dpp)3]. The key interaction (A3 – Me) has been highlighted.

James McPherson – June 2019 8-338 Dalton Transactions Chapter 8: Appendices

13 1 Me,Me Fig. S7. C{ H} NMR (101 MHz, C6D6) spectrum of [La( dpp)3].

1 13 1 Me,Me Fig S8. H- C{ H} HSQC NMR (400 MHz, 101 MHz, C6D6) spectrum of [La( dpp)3].

James McPherson – June 2019 8-339 Dalton Transactions Chapter 8: Appendices

1 13 1 Me,Me Fig S9. H- C{ H} HMBC NMR (400 MHz, 101 MHz, C6D6) spectrum of [La( dpp)3].

1 CO Me,Me Fig. S10. H NMR (400 MHz, C6D6) spectrum of [La(dpp 2 )3].

James McPherson – June 2019 8-340 Dalton Transactions Chapter 8: Appendices

1 1 CO Me,Me Fig S11. H- H COSY NMR (400 MHz, C6D6) spectrum of [La(dpp 2 )3].

1 1 CO Me,Me Fig S12. H- H nOeSY NMR (400 MHz, C6D6) spectrum of [La(dpp 2 )3].

James McPherson – June 2019 8-341 Dalton Transactions Chapter 8: Appendices

1 CO Me,Me Fig. S13. Superimposed H NMR (C6D6) spectra of solutions of [Sm(dpp 2 )3] prepared separately at different fields (600 MHz in blue, 400 MHz in black).

1 CO Me,Me Fig. S14. From bottom to top, H NMR (C6D6) spectra of [Sm(dpp 2 )3]: after 20 minutes in solution (600 CO Me,Me MHz, bottom), [Sm(dpp 2 )3] after 4 hours in solution (600 MHz, second from the bottom), after 12 hours in solution (600 MHz, second from the top); and after 9 days in solution (400 MHz, top). Selected signals, which correspond to unbound HdppCO2Me,Me and increase over time, have been highlighted by grey regions, CO Me,Me while the signals assigned to the [Sm(dpp 2 )3] complex (and decrease over time) are coloured red and blue according to their assignment (Fig. X) for clarity. Note – a break in scale occurs between 4–5 ppm.

James McPherson – June 2019 8-342 Dalton Transactions Chapter 8: Appendices

1 1 CO Me,Me Fig. S15. H- H COSY NMR (600 MHz, C6D6) of [Sm(dpp 2 )3]. In black are signals corresponding to free HdppCO2Me,Me which has dissociated during the course of the solution state NMR studies.

1 1 CO Me,Me Fig. S16. H- H nOeSY NMR (600 MHz, C6D6) of [Sm(dpp 2 )3] highlighting the key nOe interactions (opposite phase to diagonal peaks) between C3 and the 4-methyl pyrrolide proton environments.

S10

James McPherson – June 2019 8-343 Dalton Transactions Chapter 8: Appendices

13 1 CO Me,Me 13 Fig. S17. C{ H} NMR (151 MHz, C6D6) spectrum for [Sm(dpp 2 )3]. Shown inset are the five C environments associated with the pyrrole methyl substituent.

1 13 CO Me,Me Fig. S18. H- C HSQC NMR (600 MHz, 151 MHz, C6D6) spectrum of [Sm(dpp 2 )3].

S11

James McPherson – June 2019 8-344 Dalton Transactions Chapter 8: Appendices

1 13 CO Me,Me Fig. S19. H- C HMBC NMR (600 MHz, 151 MHz, C6D6) spectrum of [Sm(dpp 2 )3].

1 CO Me,Me Fig. S20. H NMR (400 MHz, C6D6) spectrum of [Eu(dpp 2 )3].

S12

James McPherson – June 2019 8-345 Dalton Transactions Chapter 8: Appendices

1 CO Me,Me Fig. S21. H NMR (400 MHz, C6D6) spectrum of [Gd(dpp 2 )3].

1 CO Me,Me Fig. S22. H NMR (400 MHz, C6D6) spectrum of [Yb(dpp 2 )3].

S13

James McPherson – June 2019 8-346 Dalton Transactions Chapter 8: Appendices Mass Spectra JM-2-19-1 #1-35 RT: 0.01-1.00 AV: 35 NL: 1.99E8 T: FTMS + c NSI Full ms [150.00-2000.00] 609.2214 100 316.1051 95

1038.2202 NL: 80 100 2.98E7 JM-2-19-1#1-35 RT: 90 0.01-1.00 AV: 35 T: FTMS + c NSI Full ms 75 80 [150.00-2000.00] 70 70 60 1039.2231 50 65 40 RelativeAbundance 30

60 20 1040.2262 10 1035.7133 1041.2291 1028.0613 1032.2335 1047.1777 1054.1939 1058.3608 0 1038.2214 NL: 55 100 5.48E5 90 C 51 H42 LaN9 O6 +Na: C 51 H42 La 1 N9 O6 Na 1 50 80 pa Chrg 1 70

45 60 1039.2247 50 RelativeAbundance 40 40 30

20 35 1040.2281 10 1041.2314 1037.2221 1046.2457 0 30 1025 1030 1035 1040 1045 1050 1055 m/z 25

20 568.2773 1038.2202 15 363.2133 10 284.0789

5 918.3031 459.6545 723.1215 1096.2825 1347.3048 1640.4186 1816.2904 0 200 400 600 800 1000 1200 1400 1600 1800 2000 m/z CO Me,Me Fig. S23. (+)-HR-NSI-MS of [La(dpp 2 )3] in acetonitrile. Shown inset is the observed isotopomers (top) CO Me,Me + against the simulated isotopic distribution for [La(dpp 2 )3 + Na] .

1051.2327 NL: 100 7.74E6 1053.2353 Sm(dpp)3#13-33 RT: 80 1048.2305 0.36-0.94 AV: 21 T: FTMS + p NSI Full ms 1046.2276 [150.00-2000.00] 60 1029.2509 1055.2415 1026.2487 40 1024.2459 1033.2599 20 1043.2247 1021.2434 1001.6759 1009.5044 1038.5847 1060.7440 1069.2097 1074.1964 1084.2396 1091.4861 0 1029.2528 NL: 100 1.47E5

1031.2553 C 51 H42 N9 O 6 Sm +H: 80 C 51 H43 N9 O 6 Sm 1 pa Chrg 1

60 1024.2480

40 RelativeAbundance 20 1021.2451 1036.2721 0 1051.2348 NL: 100 1.47E5

1053.2372 C 51 H42 N9 O 6 Sm +Na: 80 C 51 H42 N9 O 6 Sm 1 Na 1 pa Chrg 1

60 1046.2299

20 1043.2270 1058.2540 0 1000 1010 1020 1030 1040 1050 1060 1070 1080 1090 m/z CO Me,Me Fig. S24. (+)-HR-NSI-MS of [Sm(dpp 2 )3] in acetonitrile (top), against the simulated isotopic distribution CO Me,Me + CO Me,Me + for [Sm(dpp 2 )3 + H] (middle) and [Sm(dpp 2 )3 + Na] (bottom).

S14

James McPherson – June 2019 8-347 Dalton Transactions

JM-2-63-1 #1-35 RT: 0.01-0.99 AV: 35 NL: 1.18E9 Chapter 8: Appendices T: FTMS + c NSI Full ms [150.00-2000.00] 609.2208 100

1073.2522 NL: 100 4.40E7 95 JM-2-63-1#1-35 RT: 80 0.01-0.99 AV: 35 T: 1071.2498 FTMS + c NSI Full ms 1051.2707 [150.00-2000.00] 90 60 316.1048 1049.2682 1075.2557 40 1069.2474 1053.2742 85 1047.2657 20 1077.2614 1055.2802 1039.1984 1061.7002 1089.2256 0 80 1051.2719 NL: 100 1.74E5

C 51 H42 YbN9 O6 +H: 80 C 51 H43 Yb 1 N9 O6 75 1049.2695 pa Chrg 1 60

70 40

RelativeAbundance 20 65 1045.2670 1058.2924 0 1073.2539 NL: 100 1.74E5 60 C 51 H42 YbN9 O6 +Na: 80 C 51 H42 Yb 1 N9 O6 Na 1 1071.2514 pa Chrg 1 55 60

50 20

1067.2489 1080.2744 0 45 1040 1050 1060 1070 1080 1090 m/z

RelativeAbundance 40

15 667.1789 284.0787 10

5 494.2036 1073.2522 758.1539 393.2964 978.5499 1271.6661 1605.4365 1816.2926 0 200 400 600 800 1000 1200 1400 1600 1800 2000 m/z

CO Me,Me Fig. S25. (+)-HR-NSI-MS of [Yb(dpp 2 )3] in acetonitrile. Shown inset is the observed isotopomers (top) CO Me,Me + CO Me,Me + against the simulated isotopic distribution for [Yb(dpp 2 )3 + H] (middle) and for [Yb(dpp 2 )3 + Na] (bottom).

S15

James McPherson – June 2019 8-348 Dalton Transactions

JM-2-21-1 #1-35 RT: 0.01-0.99 AV: 35 NL: 7.42E8 Chapter 8: Appendices T: FTMS + c NSI Full ms [150.00-2000.00] 521.2413 100

70 250.1332 65

60 288.1099

RelativeAbundance 40

10 770.3675 642.4227 393.2965 5 853.4348 1058.3461 1183.6868 1464.9105 1677.8908 1816.2896 0 200 400 600 800 1000 1200 1400 1600 1800 2000 m/z

Me,Me Fig. S26. (+)-HR-NSI-MS of [La(dpp )3] in acetonitrile.

S16

James McPherson – June 2019 8-349 Dalton Transactions Chapter 8: Appendices Photophysical studies

[Eu(dpp)3]:

Fig. S27. Normalised excitation (em = 612 nm) and emission (exc = 300 nm) spectra of [Eu(dpp)3] in CH2Cl2 at room temperature.

Fig. S28. Normalised excitation (em = 612 nm) and emission (exc = 300 nm) spectra of [Eu(dpp)3] in CH2Cl2 at 77K.

S17

James McPherson – June 2019 8-350 Dalton Transactions Chapter 8: Appendices

Fig. S29. Normalised excitation (em = 612 nm) and emission (exc = 300 nm) spectra of [Eu(dpp)3] in the solid state.

[Yb(dpp)3]:

Fig. S30. Normalised excitation (em = 980 nm) and emission (exc = 300 nm) spectra of [Yb(dpp)3] in CH2Cl2 at room temperature.

S18

James McPherson – June 2019 8-351 Dalton Transactions Chapter 8: Appendices

Fig. S31. Normalised excitation (em = 980 nm) and emission (exc = 300 nm) spectra of [Yb(dpp)3] in CH2Cl2 at 77K.

Fig. S32. Normalised excitation (em = 980 nm) and emission (exc = 300 nm) spectra of [Yb(dpp)3] in the solid state.

S19

James McPherson – June 2019 8-352 Dalton Transactions Chapter 8: Appendices Cyclic Voltammetry

CO Me,Me Fig. S33. Overlay of CVs for [Ln(dpp 2 )3] (Ln = La(III), black trace; Sm(III), orange trace; Eu(III), red trace; and Gd(III), blue trace).

CO Me,Me COOMe,Me Fig. S34. CV of [Yb(dpp 2 )3] (green trace) overlayed onto those for [Gd(dpp )3] (transparent blue trace) and HdppCOOMe,Me (dashed black trace).

S20

James McPherson – June 2019 8-353 Chapter 8: Appendices

James McPherson – June 2019 8-354 Chapter 8: Appendices

APPENDIX 4 – SUPPORTING INFORMATION FOR CHAPTER 5

James McPherson – June 2019 8-355 Chapter 8: Appendices

Cartesian Coordinates from DFT Calculations

CO2H,CO2H BP86-GD3/def2-SVP Optimisation [Ru(dpp )2(MeCN)2]

CO H,CO H CO H,CO H cis-[Ru(dpp 2 2 )2(MeCN)2] trans-[Ru(dpp 2 2 )2(MeCN)2] OC-6-33′ OC-6-1′2′ E = –2520.94809375 Ha E = –2520.93413243 Ha ZPE correction: 0.560121 Ha ZPE correction: 0.560294 Ha Ru -0.00419 0.03258 -0.01298 Ru 0.0039 -0.01454 -0.00441 O 5.94904 1.38861 -0.69052 O 6.21962 -1.10543 -0.51165 O 6.54408 -0.7793 -0.40624 O 6.49409 0.56767 0.99245 N 2.03472 -0.13738 0.19072 N 2.07683 0.21415 -0.1325 N 0.15177 -1.49312 1.3736 N 0.22096 1.86798 -0.82752 C 3.11502 0.54016 -0.27098 C 3.20611 -0.55568 -0.08921 C 4.29876 -0.18286 0.05106 C 4.34381 0.29114 0.05221 C 3.88904 -1.34406 0.78203 C 3.8687 1.63748 0.0517 C 2.46872 -1.27694 0.85523 C 2.45409 1.54901 -0.09697 C 1.4208 -2.0213 1.5391 C 1.42333 2.48137 -0.53714 C 1.59939 -3.16865 2.34655 C 1.58491 3.86016 -0.81334 H 2.60278 -3.6018 2.43174 H 2.51199 4.3559 -0.50253 C 0.50331 -3.74561 2.99439 C 0.57635 4.55518 -1.48799 H 0.64353 -4.64038 3.62089 C -0.57932 3.86801 -1.90646 C -0.77389 -3.17576 2.8351 C -0.71198 2.52364 -1.54833 H -1.66365 -3.59463 3.3267 C 5.75037 -0.16281 0.11227 C -0.90243 -2.05469 2.00997 H 7.41677 0.25213 0.87174 H -1.87803 -1.57911 1.8299 C 4.75407 2.81996 0.02931 C 5.64508 0.24577 -0.36381 O -6.45889 -0.42061 -1.064 H 7.40014 -0.35873 -0.64009 O -6.19795 1.11883 0.57781 C 4.82182 -2.27704 1.45923 N -2.06717 -0.23814 0.12996 O -5.88584 1.3783 0.54438 N -0.21454 -1.90497 0.80763 O -6.528 -0.79148 0.68601 C -3.19023 0.54549 0.09072 N -2.03816 -0.18242 -0.21945 C -4.33843 -0.28731 -0.01795 N -0.1291 -1.5019 -1.39463 C -3.87905 -1.64133 0.0062 C -3.1328 0.47709 0.24793 C -2.45719 -1.56515 0.11842 C -4.30298 -0.26284 -0.07541 C -1.41965 -2.51185 0.51357 C -3.87225 -1.41254 -0.8168 C -1.57313 -3.90242 0.72329 C -2.45032 -1.32477 -0.88303 H -2.50442 -4.38063 0.39061 C -1.38375 -2.06883 -1.54133 C -0.5539 -4.6175 1.3622 C -1.53603 -3.26082 -2.28602 H -0.66851 -5.69951 1.53339 H -2.53314 -3.72099 -2.3287 C 0.60063 -3.942 1.80075 C -0.42509 -3.82954 -2.91768 H 1.40998 -4.45378 2.33961 H -0.54091 -4.75789 -3.49879 C 0.72844 -2.58432 1.49119 C 0.83393 -3.21245 -2.79671 H 1.62208 -2.00884 1.76726 H 1.73255 -3.62377 -3.27897 C -5.73156 0.20991 -0.09469 C 0.93714 -2.05697 -2.01563 H -7.36656 -0.05373 -0.97482 H 1.90061 -1.55 -1.85715 C -4.71229 -2.85342 0.04777 C -5.67028 0.03 0.40057 O -5.94198 -2.63032 0.59402 H -6.78716 1.44619 0.92954 O 4.24558 3.88619 0.73853 C -4.74267 -2.40653 -1.47381 H 4.91937 4.5944 0.62837 O -5.92975 -1.87657 -1.86906 O 5.83543 2.89377 -0.53119 O 4.47646 -3.59725 1.28704 H -6.41358 -3.48821 0.51277 H 5.15655 -4.0985 1.79155 O -4.38324 -3.97725 -0.32818 O 5.80022 -1.95546 2.10998 N -0.04963 -0.81271 -1.79138 H -6.45285 -2.63611 -2.20745 C 0.02883 -1.36051 -2.82257 O -4.46364 -3.58363 -1.69417 C -3.16188 2.02035 0.18431 N 0.11149 1.4308 1.36556 C -3.99045 2.81929 -0.64516 C 0.40879 2.2954 2.09701 C -3.92267 4.21232 -0.54054 C -2.99475 1.76302 0.97601 C -3.02031 4.78185 0.37547 C -3.37157 1.86829 2.33444 C -2.24387 3.91024 1.15734 C -3.18126 3.08399 3.00278 N -2.31324 2.57229 1.08485 C -2.62095 4.16151 2.29412 H -4.65777 2.34247 -1.37728 C -2.28741 3.96154 0.94271 H -4.55571 4.84956 -1.17849 N -2.47015 2.80335 0.2888 H -2.91936 5.87269 0.48303 H -3.79503 0.99239 2.84845 H -1.51847 4.31969 1.88551

James McPherson – June 2019 8-356 Chapter 8: Appendices

H -3.45992 3.19023 4.06348 N 0.05485 0.78274 1.78257 H -2.45416 5.1391 2.77362 C -0.02819 1.32957 2.81401 H -1.84797 4.78849 0.35346 C 3.19275 -2.02885 -0.21545 N -0.13853 1.43571 -1.38292 C 4.02642 -2.83968 0.59669 C -0.42975 2.30385 -2.11251 C 3.96774 -4.23056 0.46183 C 2.97386 1.83901 -0.97551 C 3.07101 -4.78641 -0.468 C 3.38409 1.98175 -2.32036 C 2.29023 -3.90337 -1.23224 C 3.18689 3.20758 -2.96631 N 2.34974 -2.56686 -1.12952 C 2.5887 4.25916 -2.24909 H 1.56942 -4.30207 -1.97081 C 2.22728 4.02407 -0.91119 H 4.69401 -2.37492 1.33562 N 2.41441 2.85418 -0.27892 H 2.97789 -5.87529 -0.59976 H 1.7617 4.8306 -0.31356 H 4.60526 -4.87698 1.08591 H 3.84775 1.13009 -2.8391 C 0.28966 -2.11212 -4.03824 H 2.41838 5.24477 -2.7107 H -0.41082 -2.96809 -4.12726 H 3.49534 3.34381 -4.01513 H 1.33052 -2.49843 -3.9901 C 0.94848 3.39358 2.88028 H 0.18254 -1.47395 -4.9394 H 1.88244 3.72952 2.37908 C -0.29373 2.08158 4.02846 H 1.18789 3.06884 3.91374 H 0.43871 2.9065 4.14811 H 0.22672 4.23391 2.93085 H -0.24107 1.43121 4.92578 C -0.94681 3.40886 -2.90144 H -1.31589 2.51125 3.95332 H -1.88645 3.75582 -2.41925 H -1.60118 1.93789 -1.81724 H -1.17022 3.08885 -3.94 H -1.38082 4.3613 -2.47336 H -0.21473 4.24104 -2.93694 H 0.70059 5.62733 -1.70739 CO H,CO H CO H,CO H trans-[Ru(dpp 2 2 )2(MeCN)2] trans-[Ru(dpp 2 2 )2(MeCN)2] OC-6-1′2′ OC-6-2′2 E = –2520.92489487 Ha E = –2520. 92666007 Ha ZPE correction: 0.560064 Ha ZPE correction: 0.559919 Ha Ru 0.03082 0.07433 0.52714 Ru 0.17058 -1.47147 0.11839 O -5.60883 1.47847 -1.64469 O 3.61469 3.84029 -0.43266 O -6.57131 -0.30508 -0.63105 O 3.21343 3.36901 -2.61301 N -2.07833 -0.05263 0.34545 N 1.5891 -0.15159 -0.58193 N -0.32278 -1.93817 0.90113 N 1.87358 -2.65693 -0.06885 C -3.06106 0.72222 -0.1989 C 1.58627 1.11227 -1.10683 C -4.23105 -0.05974 -0.41544 C 2.90368 1.62822 -1.03187 C -3.92096 -1.39133 -0.00684 C 3.74013 0.61598 -0.47527 C -2.5767 -1.35134 0.44551 C 2.87896 -0.50558 -0.2451 C -1.62761 -2.37426 0.8434 C 3.05751 -1.93201 -0.0245 C -1.94485 -3.70769 1.20152 C 4.29133 -2.61499 0.10493 H -2.99194 -4.03015 1.15461 H 5.20737 -2.01867 0.19372 C -0.94167 -4.56482 1.65962 C 4.32304 -4.01176 0.09575 H -1.18511 -5.60335 1.93378 C 3.12045 -4.72921 -0.06023 C 0.36862 -4.06959 1.80595 C 1.92617 -4.00557 -0.13586 H 1.18858 -4.68807 2.19743 C 3.30394 3.03567 -1.2998 C 0.62139 -2.7484 1.43353 H 3.44597 4.32421 -2.65481 H 1.61954 -2.30568 1.54422 C 5.17934 0.83623 -0.29885 C -5.49104 0.4603 -0.97095 O -3.06414 3.70005 0.58865 H -7.32765 0.09328 -1.1144 O -5.00752 2.66989 1.12718 C -4.79494 -2.56137 -0.28052 N -1.4698 -0.23959 0.43004 O 5.33834 -1.51814 -2.07818 N -1.37053 -2.80804 0.57829 O 6.5978 -0.58789 -0.43876 C -1.74435 1.05611 0.78945 N 2.09046 0.11747 0.17998 C -3.10981 1.33126 0.49031 N 0.4686 2.11629 0.67975 C -3.68982 0.11603 0.00408 C 3.00858 -0.76199 -0.33964 C -2.64895 -0.86028 0.07448 C 4.24154 -0.08947 -0.5446 C -2.61024 -2.31434 0.19344 C 4.04643 1.27294 -0.16493 C -3.7277 -3.17817 0.11764 C 2.68671 1.36115 0.2645 H -4.68869 -2.74779 -0.20145 C 1.80605 2.45832 0.62972 C -3.59443 -4.51678 0.50057 C 2.22291 3.76989 0.96112 H -4.46293 -5.19209 0.44883 H 3.29852 3.99403 0.91808 C -2.35545 -4.97762 0.98763 C 1.27858 4.7186 1.36319 H -2.21901 -6.00783 1.34823 H 1.59903 5.74155 1.61645 C -1.27765 -4.08562 1.00796 C -0.07093 4.33637 1.47489 H -0.28915 -4.39157 1.38232 H -0.85286 5.03032 1.81545 C -3.84436 2.58543 0.76189 C -0.41967 3.02549 1.1396 H -3.63434 4.44871 0.87192 H -1.45357 2.67345 1.23883 C -5.01345 -0.03449 -0.62181 C 5.51265 -0.71768 -0.98263 O -5.42119 1.11223 -1.23618 H 6.21721 -1.93098 -2.23235 O 5.78519 -0.03333 0.58312

James McPherson – June 2019 8-357 Chapter 8: Appendices

C 5.03535 2.35245 -0.32762 H 6.7278 0.24533 0.58608 O 5.92435 2.08834 -1.3254 O 5.82546 1.72644 -0.83708 O -5.01077 -3.35082 0.8228 H -6.33517 0.92705 -1.54471 H -5.5879 -4.07869 0.49726 O -5.69864 -1.05516 -0.67334 O -5.29363 -2.82745 -1.35968 N 0.70074 -1.05241 1.95673 H 6.5847 2.8139 -1.28123 C 1.01218 -0.66026 3.01635 O 5.1034 3.39894 0.31546 C -0.83636 1.87886 1.62678 N -0.12474 0.04268 -1.43602 C -1.34963 2.53382 2.77875 C -0.35804 0.12358 -2.57624 C -0.48756 3.31996 3.55598 C 2.73163 -2.18215 -0.65332 C 0.8648 3.43299 3.18205 C 3.60746 -3.20045 -0.197 C 1.29288 2.71705 2.04869 C 3.31045 -4.53584 -0.49372 N 0.46975 1.95994 1.30969 C 2.14924 -4.82353 -1.23306 H -2.40602 2.41039 3.06077 C 1.35327 -3.74442 -1.65629 H -0.86968 3.84131 4.44872 N 1.63105 -2.45965 -1.388 H 1.57068 4.05613 3.7527 H 4.48964 -2.93335 0.40495 H 2.33758 2.77966 1.68835 H 3.96848 -5.34594 -0.1399 N -0.33397 -1.79586 -1.74924 H 1.8662 -5.85853 -1.47943 C -0.62006 -1.86298 -2.88111 H 0.43423 -3.92893 -2.24288 C 0.37332 1.80122 -1.59924 N 0.13792 0.17559 2.4754 C 0.18523 3.18763 -1.36409 C 0.2199 0.25598 3.6449 C -1.00633 3.79289 -1.77386 C -2.92204 2.1606 -0.54312 C -1.99113 3.00939 -2.39916 C -3.74343 3.12499 0.08902 C -1.70592 1.65268 -2.61227 C -3.56986 4.47803 -0.22494 N -0.55593 1.06322 -2.24738 C -2.5861 4.8316 -1.16501 H -2.45854 0.9983 -3.09089 C -1.84365 3.80002 -1.76416 H 0.93645 3.76001 -0.80247 N -2.00425 2.49764 -1.47605 H -2.96577 3.42849 -2.69002 H -1.07273 4.03851 -2.51976 H -1.18623 4.85862 -1.56514 H -4.49525 2.79887 0.82271 C 1.36106 -0.03582 4.28137 H -2.40244 5.88234 -1.43733 H 0.68275 0.82671 4.45674 H -4.19016 5.24895 0.25961 H 2.40118 0.34872 4.24009 C -0.74593 0.32033 -3.96066 H 1.27345 -0.74812 5.12763 H 0.12105 0.64457 -4.57182 C -0.96338 -1.71757 -4.2848 H -1.52885 1.10826 -3.98341 H -2.03691 -1.94177 -4.45243 H -1.15733 -0.61447 -4.39369 H -0.35522 -2.38739 -4.92662 C 0.322 0.36211 5.0928 H -0.7655 -0.66048 -4.56591 H -0.10455 -0.54122 5.57693 H 0.9589 -4.51502 -0.25638 H -0.23502 1.24957 5.46085 H 3.09992 -5.82725 -0.11776 H 1.3826 0.46033 5.40775 H 5.28437 -4.54136 0.18783

BP86-GD3/def2-TZVP Optimisation

CO2Et,CO2Et CO2Et,CO2Et [Ru(dpp )2] [Ru(dpp )2MeCN] E = –2572.57369866 Ha E = -2705.39176521 Ha ZPE correction: 0.684392 Ha ZPE correction: 0.729968 Ha Ru 0.02595 -0.04371 0.07635 Ru 0.1528 0.34932 0.00536 O -4.94853 2.35233 2.25337 O 5.4153 2.49223 -1.67744 O -6.2164 0.71912 1.31011 O 6.41311 0.54191 -1.07363 N -0.31633 1.09633 1.78767 N 0.64531 1.7397 -1.48217 N -1.90339 0.17209 0.09787 N 2.07157 0.44367 0.13137 N -0.59253 -1.06725 -1.62831 N 0.58002 -1.03305 1.52692 C 0.62812 1.54034 2.64286 C -0.21462 2.40847 -2.27306 H 1.65165 1.24957 2.4002 H -1.26956 2.21534 -2.0728 C 0.33302 2.31752 3.75778 C 0.20961 3.31456 -3.24401 H 1.13995 2.64585 4.41307 H -0.53083 3.82438 -3.86103 C -1.00176 2.66347 4.00146 C 1.57907 3.55044 -3.39436 H -1.26718 3.28024 4.86173 H 1.94304 4.25843 -4.14124 C -1.99382 2.22082 3.13143 C 2.48448 2.88342 -2.5708 H -3.04334 2.48252 3.26434 H 3.55899 3.05565 -2.6277

James McPherson – June 2019 8-358 Chapter 8: Appendices

C -1.65444 1.42797 2.02533 C 2.01487 1.97267 -1.61593 C -2.56925 0.89124 1.03295 C 2.835 1.20273 -0.6861 C -3.95893 0.82722 0.6953 C 4.19689 0.87836 -0.39434 C -4.05872 0.04119 -0.4976 C 4.17327 -0.11342 0.64169 C -2.73163 -0.36061 -0.84023 C 2.80132 -0.36993 0.93649 C -1.97268 -1.06708 -1.85779 C 1.94233 -1.19496 1.77876 C -2.48542 -1.71043 -2.99388 C 2.36095 -2.09732 2.76593 H -3.56264 -1.70534 -3.14404 H 3.42777 -2.21544 2.94024 C -1.63076 -2.34792 -3.88777 C 1.42301 -2.83535 3.48446 H -2.03566 -2.84618 -4.77017 H 1.75395 -3.53883 4.2505 C -0.25275 -2.34643 -3.64009 C 0.06219 -2.67182 3.21015 H 0.45206 -2.83778 -4.3107 H -0.70376 -3.23717 3.74074 C 0.21799 -1.69801 -2.50376 C -0.31128 -1.76074 2.22461 H 1.28174 -1.66607 -2.26164 H -1.35744 -1.59684 1.96012 C -5.05243 1.39574 1.49071 C 5.36754 1.40945 -1.10141 C -7.40227 1.31478 1.89688 C 7.67722 1.04691 -1.57629 H -7.36794 1.17394 2.98953 H 7.62295 1.11071 -2.67531 C -8.59721 0.62959 1.26186 C 8.75271 0.08725 -1.10431 H -7.39168 2.39753 1.69981 H 7.83222 2.06592 -1.1905 C -5.28668 -0.17464 -1.2865 C 5.35529 -0.64886 1.34526 O 5.43755 1.98721 -0.90702 O -5.97357 0.84333 0.07226 O 6.22161 -0.04293 -1.57113 O -6.10785 -1.13281 1.19613 N 0.67077 1.60572 -1.01908 N -1.8744 -0.08385 -0.23978 N 1.95531 -0.25603 0.03657 N 0.19084 -1.25561 -1.28794 N 0.34235 -1.80762 1.14288 C -3.05741 0.42857 0.20606 C -0.12595 2.55349 -1.5561 C -4.08384 -0.5293 0.03499 H -1.19379 2.42626 -1.36971 C -3.49444 -1.66672 -0.57935 C 0.36265 3.62267 -2.29875 C -2.11622 -1.34745 -0.74164 H -0.33185 4.35626 -2.70789 C -0.98242 -1.97985 -1.37914 C 1.74493 3.72359 -2.49777 C -1.00079 -3.19899 -2.08036 H 2.16385 4.54918 -3.07553 H -1.92376 -3.77591 -2.08892 C 2.58527 2.76159 -1.9466 C 0.14802 -3.65063 -2.71759 H 3.66509 2.8165 -2.06336 H 0.13187 -4.59559 -3.2635 C 2.05493 1.69713 -1.20294 C 1.31955 -2.88597 -2.64858 C 2.79855 0.62116 -0.57114 H 2.24173 -3.19838 -3.13833 C 4.1208 0.10404 -0.40979 C 1.29835 -1.70575 -1.91809 C 4.00096 -1.11451 0.33194 H 2.19175 -1.0901 -1.81296 C 2.60654 -1.3093 0.58573 C -5.47886 -0.35998 0.4919 C 1.67719 -2.21493 1.23714 C -7.25227 1.23314 0.6349 C 2.00148 -3.40732 1.90061 H -8.03411 0.56035 0.24794 H 3.0494 -3.70598 1.92418 C -7.48363 2.67953 0.24033 C 0.99676 -4.17642 2.4803 H -7.21505 1.10012 1.72836 H 1.24978 -5.10417 2.99618 C -4.21662 -2.87546 -0.99414 C -0.33494 -3.75343 2.3875 O -5.52223 -2.5928 -1.25395 H -1.15135 -4.32798 2.82532 H 9.73913 0.42331 -1.457 C -0.61446 -2.57174 1.70946 H 8.57345 -0.92625 -1.49154 H -1.63485 -2.20094 1.59842 H 8.76568 0.04564 -0.00608 C 5.36292 0.62006 -1.01643 O 5.22605 -1.99998 1.55987 C 6.60828 2.59206 -1.52172 C 6.33375 -2.63951 2.25444 H 7.51064 2.16767 -1.05486 H 5.88561 -3.53136 2.71575 C 6.50401 4.09018 -1.30868 H 6.70292 -1.96516 3.04147 H 6.63287 2.32076 -2.58934 C 7.44116 -3.00795 1.27967 C 5.08484 -1.97224 0.82429 H 7.05512 -3.65883 0.48232 O 6.23826 -1.27661 0.99499 H 8.24657 -3.54135 1.80716 H -9.53083 1.04564 1.66878 H 7.86119 -2.10158 0.82486 H -8.58343 -0.4515 1.46266 O 6.31972 -0.00416 1.71762 H -8.58627 0.78395 0.1737 H -7.51252 2.78729 -0.85322 O -5.34652 -1.46535 -1.75648 H -6.6756 3.31681 0.62832 C -6.51749 -1.79149 -2.55739 H -8.4405 3.03509 0.65068 H -6.1941 -2.64033 -3.17699 C -6.40666 -3.72893 -1.4277 H -6.76023 -0.93595 -3.20514 C -7.82335 -3.19008 -1.36165 C -7.69446 -2.1615 -1.66835 H -6.20343 -4.46174 -0.63167 H -7.43236 -2.99926 -1.00655 H -6.18314 -4.21269 -2.39245 H -8.55497 -2.46015 -2.28623 H -7.99132 -2.69018 -0.39729 H -7.98782 -1.30161 -1.05234 H -8.54566 -4.01327 -1.46699 O -6.13904 0.6617 -1.52864 H -8.00454 -2.46415 -2.16755 H 6.46606 4.33125 -0.23695 O -3.75141 -4.0065 -1.09558

James McPherson – June 2019 8-359 Chapter 8: Appendices

H 5.60025 4.49756 -1.78488 N -0.06781 1.7714 1.3385 H 7.37909 4.59211 -1.74715 C -0.18021 2.58035 2.16954 C 7.42712 -2.06139 1.27112 C -0.40001 3.5824 3.19377 C 8.61228 -1.12943 1.10452 H -0.27191 4.59158 2.77467 H 7.46301 -2.90844 0.56896 H 0.30934 3.45206 4.02494 H 7.35476 -2.47457 2.29025 H -1.42801 3.49924 3.57836 H 8.62869 -0.71862 0.08524 C -3.18364 1.79939 0.71422 H 9.55005 -1.67655 1.28222 C -3.90267 2.08381 1.89286 H 8.55806 -0.29394 1.8175 C -4.04572 3.4073 2.30021 O 4.98089 -3.16833 1.08021 C -3.44969 4.41764 1.5365 C -2.73626 4.04099 0.39438 MeCN N -2.60717 2.77485 -0.02299 E = –132.805509312 Ha H -4.33983 1.26635 2.4687 ZPE correction: 0.043808 Ha H -4.60898 3.64947 3.20419 C 1.17860 -0.00001 0.0000 H -3.53562 5.46904 1.81583 C -0.27795 0.00001 0.00001 H -2.24785 4.79821 -0.22852 N -1.44066 -0.00000 -0.00000 H 1.56026 0.51248 0.89489 H 1.56021 0.51878 -0.89127 H 1.56017 -1.03128 -0.00365

CO2Et,CO2Et cis-[Ru(dpp )2(MeCN)2] E = –2838.20756495 Ha ZPE correction: 0.774977 Ha Ru 0.01849 0.36344 0.02229 O -5.86339 2.12402 0.83776 O -6.55423 0.01604 0.32693 N -2.03168 0.45346 -0.2079 N -0.29683 -1.0732 -1.43218 C -3.05392 1.16918 0.31514 C -4.28592 0.51948 0.01691 C -3.96219 -0.63088 -0.75452 C -2.55961 -0.63486 -0.88514 C -1.61248 -1.43527 -1.6242 C -1.9295 -2.47255 -2.51999 H -2.97077 -2.74901 -2.66708 C -0.91739 -3.12916 -3.20847 H -1.16238 -3.93488 -3.90269 C 0.41386 -2.74289 -3.0069 H 1.23864 -3.22967 -3.52647 C 0.67739 -1.71238 -2.11315 H 1.69448 -1.37621 -1.91192 C -5.60741 0.99727 0.43091 C -7.91725 0.43328 0.59009 H -8.01117 0.70157 1.65515 C -8.82025 -0.72325 0.20688 H -8.13266 1.3391 0.00226 C -4.90971 -1.58627 -1.37679 O 6.03725 0.70759 -0.36416 O 6.22246 -1.4669 -1.01803 N 1.99342 -0.16536 0.23654 N -0.10109 -1.21242 1.35393 C 3.17808 0.29299 -0.24651 C 4.18684 -0.66823 -0.00464 C 3.57202 -1.7486 0.69181 C 2.2039 -1.3963 0.82722 C 1.04643 -1.96685 1.48591 C 1.01971 -3.15965 2.22939 H 1.923 -3.76563 2.27121 C -0.15364 -3.55194 2.86347 H -0.1758 -4.47858 3.43994 C -1.29926 -2.75555 2.75082 H -2.23869 -3.02573 3.23269 C -1.23068 -1.60069 1.98089 H -2.09838 -0.95737 1.8349

James McPherson – June 2019 8-360 Chapter 8: Appendices

C 5.5719 -0.57314 -0.5002 C 7.29407 0.99981 -1.02373 H 8.10312 0.44983 -0.51689 C 7.48491 2.50362 -0.95554 H 7.24652 0.63223 -2.06162 C 4.26522 -2.93934 1.20437 O 5.55322 -2.64847 1.52471 H -9.87243 -0.45199 0.37957 H -8.5908 -1.61687 0.80499 H -8.69238 -0.97232 -0.85609 O -4.67361 -2.85471 -0.91545 C -5.50075 -3.90635 -1.49023 H -4.89726 -4.81712 -1.37116 H -5.64912 -3.70108 -2.56156 C -6.83204 -4.00949 -0.76385 H -6.67455 -4.1736 0.31172 H -7.41667 -4.85183 -1.16407 H -7.40853 -3.08604 -0.90012 O -5.75769 -1.31742 -2.20713 H 7.52979 2.84516 0.08839 H 6.64894 3.01913 -1.45063 H 8.42193 2.78958 -1.45619 C 6.42133 -3.77962 1.79338 C 7.84405 -3.25469 1.75531 H 6.24314 -4.55124 1.02889 H 6.15531 -4.21032 2.77246 H 8.05301 -2.80855 0.77288 H 8.55403 -4.07638 1.9324 H 8.00076 -2.48806 2.52801 O 3.78408 -4.06022 1.33458 N 0.14552 1.79809 -1.31749 C 0.04007 2.6858 -2.05957 C 3.29591 1.59665 -0.92398 C 3.81283 1.67964 -2.23007 C 3.91806 2.92493 -2.84258 C 3.49859 4.05698 -2.13602 C 2.994 3.87644 -0.84484 N 2.89319 2.68634 -0.23641 H 4.12194 0.76811 -2.74417 H 4.31726 3.01345 -3.85513 H 3.56675 5.05624 -2.57038 H 2.65429 4.73964 -0.26185 N 0.26599 1.66452 1.47374 C 0.55137 2.41407 2.31467 C -2.81438 2.43569 1.03353 C -3.30022 2.63385 2.3374 C -3.02736 3.83236 2.98912 C -2.27478 4.80442 2.32139 C -1.83731 4.52063 1.02528 N -2.09541 3.37403 0.38017 H -1.24885 5.25882 0.46899 H -3.89015 1.85214 2.8154 H -2.04379 5.7647 2.78688 H -3.3977 4.00934 4.00093 C -0.17858 3.83418 -2.91434 H -1.04937 4.39503 -2.53958 H -0.38078 3.51689 -3.94814 H 0.7071 4.48646 -2.91386 C 0.96476 3.36412 3.32719 H 1.92323 3.81919 3.03316 H 1.0985 2.85999 4.29603 H 0.20613 4.15235 3.44055

James McPherson – June 2019 8-361 Chapter 8: Appendices

B3LYP-GD3BJ/6-31G(d,p) Single-point ligand energies

CO2Et,CO2Et CO2Et,CO2Et [Ru(dpp )2] – L1 [Ru(dpp )2] – L2 E = –1238.28314932 Ha E = –1238.283899 Ha O 2.93077 0.57795 -0.29084 O 0.83788 -3.32215 0.58477 O 2.07914 1.95839 1.30489 O -1.21659 -2.81336 -0.24281 N 0.33079 -3.42813 -0.02302 N 3.72982 0.45934 -0.1825 N -0.89195 -1.25217 0.00242 N 1.28629 0.93498 -0.00541 N -3.33861 -1.71277 -0.18385 N 0.95267 3.40876 -0.06134 C 0.88235 -4.6598 -0.03691 C 5.06727 0.31853 -0.29141 H 0.17675 -5.4862 -0.13896 H 5.62801 1.24922 -0.39361 C 2.2512 -4.87851 0.07175 C 5.70096 -0.91931 -0.27314 H 2.63674 -5.89781 0.05502 H 6.78594 -0.96924 -0.36623 C 3.09824 -3.77092 0.19927 C 4.91963 -2.07203 -0.12662 H 4.17773 -3.90472 0.28707 H 5.38547 -3.0585 -0.09846 C 2.55145 -2.49173 0.20824 C 3.53803 -1.95251 -0.00839 H 3.17835 -1.60654 0.28635 H 2.88914 -2.81687 0.13116 C 1.16503 -2.31104 0.09616 C 2.93768 -0.68542 -0.04447 C 0.44957 -1.04693 0.09209 C 1.51812 -0.39779 0.0646 C 0.64541 0.36294 0.21738 C 0.2433 -1.03544 0.19711 C -0.6528 0.9655 0.18598 C -0.73987 0.00566 0.21416 C -1.60026 -0.0988 0.05808 C -0.03634 1.24105 0.07681 C -3.02479 -0.3557 -0.05551 C -0.23514 2.68009 0.06377 C -4.04934 0.6018 -0.03041 C -1.45721 3.36095 0.16609 H -3.76523 1.64548 0.1034 H -2.3673 2.77156 0.25162 C -5.37735 0.20312 -0.15071 C -1.49251 4.75164 0.14114 H -6.17401 0.94864 -0.13089 H -2.44633 5.27562 0.22178 C -5.67844 -1.15755 -0.28817 C -0.29619 5.46659 0.00721 H -6.7052 -1.51137 -0.38244 H -0.27811 6.55595 -0.02218 C -4.63345 -2.07526 -0.2951 C 0.89533 4.75673 -0.09115 H -4.81169 -3.14781 -0.38928 H 1.85652 5.26269 -0.19757 C 1.91222 1.06907 0.48889 C 0.0177 -2.48466 0.21948 C 4.23257 1.18213 -0.0572 C -1.61385 -4.19954 -0.08131 H 4.1577 2.26454 -0.24453 H -1.06188 -4.81527 -0.81008 C 5.21613 0.50577 -0.99302 C -3.11436 -4.25358 -0.29617 H 4.50353 1.05069 1.00267 H -1.32646 -4.53733 0.92597 C -0.97496 2.39694 0.19461 C -2.18092 -0.17489 0.4742 O 0.046 3.14493 -0.29689 H -3.4748 -5.28672 -0.182 H 4.91402 0.63954 -2.04135 H -3.38035 -3.90344 -1.3041 H 5.28155 -0.57295 -0.78909 H -3.62418 -3.61789 0.4414 H 6.21758 0.9414 -0.86153 O -2.93111 0.64859 -0.33171 C -0.07548 4.58204 -0.1373 C -4.37343 0.5598 -0.15618 C 1.28536 5.17188 -0.4553 H -4.74982 1.52999 -0.51125 H -0.39076 4.79808 0.89505 H -4.59954 0.44827 0.91475 H -0.86427 4.95191 -0.81227 C -4.94621 -0.59438 -0.96386 H 2.04472 4.75439 0.22075 H -4.69146 -0.48728 -2.02788 H 1.26106 6.26469 -0.33076 H -6.04243 -0.61525 -0.86728 H 1.57817 4.94758 -1.49129 H -4.54459 -1.54803 -0.59754 O -2.0377 2.8815 0.57237 O -2.67159 -0.92409 1.30036

CO2Et,CO2Et CO2Et,CO2Et [Ru(dpp )2MeCN] – L1 [Ru(dpp )2MeCN] – L2 E = –1238.28567717 Ha E = –1238.30750033 Ha O 0.97873 -3.28821 0.61785 O 2.51888 0.92038 -0.44782 O -1.07377 -2.84638 -0.25174 O 1.67444 2.12227 1.29242 N 3.71182 0.61277 -0.199 N -0.76351 -1.62775 0.25409 N 1.25076 0.99482 0.13941 N -3.24419 -2.36453 0.05389 N 0.77826 3.46253 -0.06016 C 0.48886 -1.08799 0.28627 C 5.04622 0.52794 -0.35531 C 0.39663 0.32282 0.32384 H 5.56097 1.48575 -0.44465 C -0.98471 0.6517 0.27612 C 5.72489 -0.68988 -0.35651 C -1.67644 -0.59196 0.22564 H 6.8066 -0.70026 -0.49317 C -3.05466 -0.99578 0.05486 C 4.99614 -1.86859 -0.17404 C -4.15127 -0.13629 -0.13766 H 5.49867 -2.83755 -0.16397 H -3.98291 0.93732 -0.07641 C 3.61618 -1.80026 0.01037 C -5.41432 -0.66454 -0.37308

James McPherson – June 2019 8-362 Chapter 8: Appendices

H 3.00815 -2.68709 0.18715 H -6.26403 0.00344 -0.52467 C 2.97348 -0.55592 -0.007 C -5.58347 -2.05449 -0.40972 C 1.54365 -0.32443 0.17164 H -6.55453 -2.51243 -0.5974 C 0.29743 -1.01946 0.26643 C -4.47805 -2.8627 -0.18215 C -0.73252 -0.02131 0.28794 H -4.56453 -3.9493 -0.17413 C -0.08245 1.24487 0.19171 C 1.54788 1.23907 0.46011 C -0.35882 2.67547 0.1217 C 3.79067 1.59667 -0.27818 C -1.61653 3.2871 0.21121 H 3.66048 2.66696 -0.50447 H -2.49122 2.65483 0.34382 C 4.77594 0.92125 -1.21325 C -1.73177 4.67166 0.10915 H 4.09708 1.51831 0.77762 H -2.71374 5.14328 0.17675 C -1.5305 2.01361 0.23166 C -0.58411 5.44501 -0.08726 O -0.63194 2.87639 -0.31624 H -0.63062 6.52963 -0.18349 H 4.44703 1.00893 -2.25844 C 0.647 4.7976 -0.16538 H 4.86606 -0.14722 -0.96866 H 1.5761 5.34709 -0.32685 H 5.76745 1.3888 -1.11981 C 0.13372 -2.47728 0.25058 C -0.93522 4.28913 -0.19252 C -1.41303 -4.25238 -0.13453 C 0.33281 5.03864 -0.55592 H -0.81722 -4.82408 -0.86463 H -1.25721 4.49376 0.84017 C -2.90413 -4.36719 -0.38724 H -1.77719 4.53463 -0.86001 H -1.13441 -4.60326 0.87069 H 1.1515 4.74032 0.11406 C -2.17017 -0.27512 0.50614 H 0.17115 6.12272 -0.46038 H -3.22 -5.4181 -0.30887 H 0.63274 4.82244 -1.59158 H -3.16242 -4.00209 -1.39185 O -2.63362 2.37262 0.63185 H -3.45913 -3.77534 0.3542 C 1.70456 -1.90732 0.22304 O -2.9367 0.51985 -0.31165 C 2.79611 -1.67171 1.08334 C -4.37743 0.36066 -0.17965 C 3.95087 -2.43638 0.94273 H -4.7889 1.31907 -0.52771 C 3.98491 -3.43513 -0.03739 H -4.6282 0.21725 0.88197 C 2.84743 -3.61325 -0.83092 C -4.87143 -0.8027 -1.02513 N 1.73685 -2.87237 -0.72295 H -4.59118 -0.66265 -2.07885 H 2.72463 -0.89338 1.84492 H -5.96776 -0.8769 -0.96182 H 4.81188 -2.26158 1.59174 H -4.43576 -1.74348 -0.66468 H 4.86823 -4.05863 -0.18529 O -2.64497 -1.05922 1.3087 H 2.82783 -4.38789 -1.60518

CO2Et,CO2Et CO2Et,CO2Et cis-[Ru(dpp )2(MeCN)2] – L1 cis-[Ru(dpp )2(MeCN)2] – L2 E = –1238.30293411 Ha E = –1238.30752863 Ha O -5.86339 2.12402 0.83776 O 6.03725 0.70759 -0.36416 O -6.55423 0.01604 0.32693 O 6.22246 -1.4669 -1.01803 N -2.03168 0.45346 -0.2079 N 1.99342 -0.16536 0.23654 N -0.29683 -1.0732 -1.43218 N -0.10109 -1.21242 1.35393 C -3.05392 1.16918 0.31514 C 3.17808 0.29299 -0.24651 C -4.28592 0.51948 0.01691 C 4.18684 -0.66823 -0.00464 C -3.96219 -0.63088 -0.75452 C 3.57202 -1.7486 0.69181 C -2.55961 -0.63486 -0.88514 C 2.2039 -1.3963 0.82722 C -1.61248 -1.43527 -1.6242 C 1.04643 -1.96685 1.48591 C -1.9295 -2.47255 -2.51999 C 1.01971 -3.15965 2.22939 H -2.97077 -2.74901 -2.66708 H 1.923 -3.76563 2.27121 C -0.91739 -3.12916 -3.20847 C -0.15364 -3.55194 2.86347 H -1.16238 -3.93488 -3.90269 H -0.1758 -4.47858 3.43994 C 0.41386 -2.74289 -3.0069 C -1.29926 -2.75555 2.75082 H 1.23864 -3.22967 -3.52647 H -2.23869 -3.02573 3.23269 C 0.67739 -1.71238 -2.11315 C -1.23068 -1.60069 1.98089 H 1.69448 -1.37621 -1.91192 H -2.09838 -0.95737 1.8349 C -5.60741 0.99727 0.43091 C 5.5719 -0.57314 -0.5002 C -7.91725 0.43328 0.59009 C 7.29407 0.99981 -1.02373 H -8.01117 0.70157 1.65515 H 8.10312 0.44983 -0.51689 C -8.82025 -0.72325 0.20688 C 7.48491 2.50362 -0.95554 H -8.13266 1.3391 0.00226 H 7.24652 0.63223 -2.06162 C -4.90971 -1.58627 -1.37679 C 4.26522 -2.93934 1.20437 H -9.87243 -0.45199 0.37957 O 5.55322 -2.64847 1.52471 H -8.5908 -1.61687 0.80499 H 7.52979 2.84516 0.08839 H -8.69238 -0.97232 -0.85609 H 6.64894 3.01913 -1.45063 O -4.67361 -2.85471 -0.91545 H 8.42193 2.78958 -1.45619 C -5.50075 -3.90635 -1.49023 C 6.42133 -3.77962 1.79338 H -4.89726 -4.81712 -1.37116 C 7.84405 -3.25469 1.75531

James McPherson – June 2019 8-363 Chapter 8: Appendices

H -5.64912 -3.70108 -2.56156 H 6.24314 -4.55124 1.02889 C -6.83204 -4.00949 -0.76385 H 6.15531 -4.21032 2.77246 H -6.67455 -4.1736 0.31172 H 8.05301 -2.80855 0.77288 H -7.41667 -4.85183 -1.16407 H 8.55403 -4.07638 1.9324 H -7.40853 -3.08604 -0.90012 H 8.00076 -2.48806 2.52801 O -5.75769 -1.31742 -2.20713 O 3.78408 -4.06022 1.33458 C -2.81438 2.43569 1.03353 C 3.29591 1.59665 -0.92398 C -3.30022 2.63385 2.3374 C 3.81283 1.67964 -2.23007 C -3.02736 3.83236 2.98912 C 3.91806 2.92493 -2.84258 C -2.27478 4.80442 2.32139 C 3.49859 4.05698 -2.13602 C -1.83731 4.52063 1.02528 C 2.994 3.87644 -0.84484 N -2.09541 3.37403 0.38017 N 2.89319 2.68634 -0.23641 H -1.24885 5.25882 0.46899 H 4.12194 0.76811 -2.74417 H -3.89015 1.85214 2.8154 H 4.31726 3.01345 -3.85513 H -2.04379 5.7647 2.78688 H 3.56675 5.05624 -2.57038 H -3.3977 4.00934 4.00093 H 2.65429 4.73964 -0.26185

James McPherson – June 2019 8-364 Chapter 8: Appendices

APPENDIX 5 – CRYSTALLOGRAPHIC DATA

CIF files and CheckCIF reports are included separately.

James McPherson – June 2019 8-365 Chapter 8: Appendices

James McPherson – June 2019 8-366 Chapter 8: Appendices

APPENDIX 6 – RESPONSES TO EXAMINER COMMENTS

The following queries were raised during examination of this thesis regarding the published work described in Chapters Three and Four.

Chapter 3

One query about the SCO-related work in this Chapter concerns the solution state magnetic susceptibility measured by the Evans method. These measurements were performed in CDCl3 up to a temperature of 323 K. It seems strange that at the highest temperatures, the

χMT profiles of three SCO complexes appear to level off at a value that is lower than expected, meanwhile the profile for the HS-Co(II) complex decreases by about 10 % by the highest temperature. I wonder if this might reflect evaporation of the solvent as it approaches its boiling temperature, causing an increase of the concentration of the metal complex in solution.

3 –1 Response: As is explicitly stated within this published work, the χMT of 2.8 cm K mol

CN,Ph at 298 K was reproduced over two separate VT runs for [Co(dpp )2], and these solution state measurements were carried out within sealed NMR tubes. The change in concentration due to evaporation of CDCl3 at 323 K within a sealed NMR tube is negligible and trivial to

‡ approximate , and cannot account for the unusual χMT values observed at higher populations of HS-Co(II) species in the solution state measurements. Therefore, as stated in the published work: “we can provide no reasonable explanation for the different (χMT)HS values predicted by the fits to eq 1 for the structurally similar complexes” and no further change (speculation) is warranted.

‡At the boiling point of a solvent, the partial pressure of the gas-phase solvent is 1 atm, which limits the maximum amount of CDCl3 evaporated in this case to n = pV/RT = 1 atm × 2 × 10–6 m3 / (8.21 m3 atm K–1 mol–1 × 334 K) = 0.73 nmol cf. the 0.5 mL sample solution contains 3 -3 –1 n = 0.5 cm × 1.5 g cm / 120 g mol = 6.3 mmol CDCl3. The amount of chloroform evaporated to reach 1 atm is 7 orders of magnitude smaller, i.e. the concentration change with temperature is negligible throughout these experiments as is widely known and accepted throughout the chemistry community (and thus deemed superfluous and unnecessary to explicitly state).

James McPherson – June 2019 8-367 Chapter 8: Appendices

A second query concerns the measurement of the redox process assigned as Co(III/II). As synthesized, the complexes are in the Co(II) oxidation state, but the CVs (Figure 7 left) appear to be measured sweeping the potential from more positive to more negative, as if the process is a reduction instead of an oxidation. The current values have been normalized and the "rest potential"/"position of zero current" is not clear. Normally the CV of an oxidation process would be measured starting at a more negative potential ("position of zero current") and scanning to a more positive switch potential and then back to the initial potential. This is a small point of convention and the E1/2 values should be the same for a reversible process measured either way.

Response: As the examiner states, the CVs for these reversible processes were identical regardless of whether the experiment started at higher or lower potentials relative to E1/2. The

Me,Me Me,Me CVs for [Co(dpp )2] and [Co(dpp )2][PF6] shown in Figure S43 (Appendix 2) were identical. As the key result described in this work was the correlation of E1/2 to electronic parameters of the dpp– ligand substituents, the currents were normalised to show the relative positions of the E1/2 values most clearly.

Chapter 4

The next section focuses mainly on the 1H NMR spectra of the complexes. These spectra clearly indicate resonances due to free ligand following dissociation of the complexes, which occurs at different rates depending on the size of the lanthanide ions. The discussion focuses on the assignment of resonances due to the different stereoisomers. However, given the significant quantity of free ligands evident by NMR, the bis(dipyridylpyrrolide) aqua complexes that result from ligand dissociation should also be considered in this discussion.

Response: The mono- or bis-(dipyridylpyrrolide) aqua species [Ln(dpp)x(H2O)y] (x = 1 or 2) were considered, however no evidence for them was observed in the results from any of the 1H or 13C NMR, photophysical or voltammetric experiments undertaken during this work.

James McPherson – June 2019 8-368