Ferrocenes and Isoindolines As Reagents for Redox Flow Battery Electrolytes and Moieties in Chromophores, Chelates, and Macrocycles

Home , Salt bridge, Semipermeable membrane

@ 2021

Briana R. Schrage

A Dissertation

Presented to

The Graduate Faculty of The University of Akron

In Partial Fulfillment

of the Requirements for the Degree

Doctor of Philosophy

Briana R. Schrage

August, 2021

FERROCENES AND ISOINDOLINES AS REAGENTS FOR REDOX FLOW BATTERY ELECTROLYTES AND MOIETIES IN CHROMOPHORES, CHELATES, AND MACROCYCLES

Briana R. Schrage

Dissertation

Approved: Accepted:

______Advisor Department Chair Dr. Christopher J. Ziegler Dr. Christopher J. Ziegler

______Committee Member Dean of the College Dr. Aliaksei Boika Dr. Joseph Urgo

______Committee Member Dean of the Graduate School Dr. Claire A. Tessier Dr. Marnie M. Saunders

______Committee Member Date Dr. Yi Pang

______Committee Member Dr. Junpeng Wang

iii ABSTRACT

Although rechargeable battery technology has been around as early as the

1800s, redox flow battery (RFB) technology has a little under five decades of research. The most common and well-studied system is the all-vanadium RFB.

To this day there is still no perfect RFB design and many batteries suffer from membrane crossover due to corrosive solvents or electroactive materials.

Additionally, the cost of some electrolyte components are expensive, and the materials themselves may be toxic. Recent studies have investigated the use of metallocenes as potential RFB components, particularly ferrocene. The ferrocene scaffold is easily modified and this organometallic unit undergoes a highly reversible redox reaction. Introducing water solubilizing groups to metallocenes can allow for these materials to be used in aqueous RFB devices.

The second and third chapters of this dissertation investigate the electrochemistry of ferrocene-based compounds, and a practical application involving one of these compounds was investigated in a RFB cell. In chapter II, a series of four all-ferrocene salts were synthesized comprising of a

(ferrocenemethyl)trimethylammonium cation and either carboxylate or sulfonate ferrocene anion. Cyclic voltammograms in water, propylene carbonate (PC), and

DMF result in different potentials of the ions in solution. Chapter III investigates the 1,1’-bis(sulfonate)ferrocene dianion disodium salt as a catholye species in a

iv battery cell paired with anthraquinone-2,7-disulfonic acid disodium salt as the anolyte. The RFB experiments were performed using aqueous solvent in both neutral pH conditions with 1 M NaNO3, as well as acidic pH conditions using either

0.5 M H2SO4 or 2 M acetate buffer.

The second half of this dissertation investigates chelates, chromophores and macrocycles that use 1,3-diiminoisoindoline (DII) as a starting reagent. Since

Elvidge and Linstead discovered DII in 1952, they quickly made progress generating new chelates and macrocycles with this compound. Elvidge and

Linstead’s discovery of chelates called bis(arylimino)isoindoline (BAI) ligands arise from the condensation of DII with aryl amines, and the ligands typically bind to metals in a meridional coordination mode rather than a facial mode. Chapter IV investigates a rare example where three BAI ligands bind to Re(CO)3 in a facial coordination mode. The ligands distort from planarity and the complexes exhibit

MLCT bands in the UV-visible spectra. Chapter V of this dissertation revisits the

Knoevenagel condensation reaction between DII and ethylcyanoacetate, first carried out by Elvidge and Linstead, and extends it to other organic acids. Four

1,3-diylideneisoindolines were synthesized, yielding brightly colored chromophores. The DFT calculations reveal that that the HOMO energies vary depending on the alkene substituent and possess a significant degree of alkene character.

In addition to BAI chelates, Elvidge and Linstead synthesized macrocyclic systems called hemiporphyrazines, where the condensation reaction between DII

and 2,6-diaminopyridine, yields a 20π electron non-aromatic system. Chapter VI of this dissertation studies the synthesis of a hexameric hemiporphyrazine resulting from the condensation of DII with a dimeric form of 2,6-diaminopyridine called bis(6-amino-2-pyridyl)amine. The expanded system does not possess any aromaticity, like hemiporphyrazine, and the X-ray crystal structure of the ligand shows two inverted rings in the backbone.

The last two chapters of this dissertation examine the structure and electronics of new phthalocyanine and subphthalocyanine anologs called biliazine and subbiliazine. Along with BAIs, under certain circumstances singly substituted

DII chelates can be generated. These chelates are called semihemiporphyrazines and a new semihemiporphyrazine was generated with DII and 3-aminopyrazole in chapter VII. Subsequent dimerization of this chelate results in a phthalocyanine anolog called biliazine, where the meso position of the macrocycle is closed by a hydrogen bond. This reaction was carried out in the presence of Zn, Cu, and Co acetate metal templates, and the free base ligand was also isolated. The ring contracted variants of these systems are presented in chapter VIII of this dissertation. Two BAI chelates with aminopyrazole and aminoindazole were synthesized and reacted with BF3 to yield a ring contracted variant of biliazine.

These complexes called subbiliazines have a similar bowl-shape to subphthalocyanine. Reaction of the ligands under non-anhydrous conditions results in hydrolysis products, including the formation of a new dibenzo aza-

BODIPY analog.

vi DEDICATION

To my father, Dr. Dean Schrage

ACKNOWLEDGEMENT

I would first like to thank my advisor Dr. Christopher J. Ziegler. Words cannot express my gratitude for all of his guidance and help. He has formed me into the scientist I am today, and I could not have accomplished as much as I did without his direction, and broad scientific knowledge. His strong work ethic, ability to navigate through problems, teach others, and design unique projects is truly an inspiration. I will always cherish his teachings and his mentoring will stay with me forever.

I would like to thank Dr. Aliaksei Boika for all of his help with the redox flow battery project. His involvement has been incredibly useful and much appreciated.

I would also like to thank Dr. Viktor Nemykin, who has been patient with me and always answers my questions. I am grateful for his involvement in the computational analysis of my molecules. I would also like to thank Dr. Richard

Herrick, who has provided valuable input on projects.

Thank you to the members of my committee: Dr. Claire Tessier, Dr. Yi

Pang, Dr. Aliaksei Boika, and Dr. Junpeng Wang. Their teachings and passion for chemistry have been an inspiration. They have all been incredibly supportive, and

I appreciate their time spent serving on my committee. I would additionally like to thank The University of Akron and the department of Chemistry for accepting me into the program and financially supporting me during my Ph.D. career.

viii I would like to thank all group members in the lab (current and past). Thank you, Sanjay Gaire, and Huayi Wang for creating a supportive, and collaborative work environment. I would especially like to thank Laura Crandall for taking me under her wing. She is a great “Lab Mom” and friend to me. I am extremely grateful to Laura, Kullapa Chanawanno, Allen Osinski, and Dan Morris for accepting me right away into the group and showing me the ropes. They all took the time to help me with my research and have been great role models.

I would also like to thank Dr. Boika’s group members for all their work and support involving the redox flow battery project. The electrochemical measurements and battery tests carried out by Zhiling Zhao, Baosen Zhang, and

Arianna Frkonja-Kuczin have been most essential. I am especially grateful for

Zhiling’s friendship and help over the years.

I would also like to thank other graduate students, staff, and faculty who have provided help throughout these five years. Jason Bella has always been an incredible resource and great friend. Thank you, Mike Stromyer for all the help and training with the X-ray diffractometer. Thank you, Jason O’Neill, for always running my mass spec samples and working hard to evaluate everything with precision and care. I would also like to thank Bart, Simon, Venkat, Jessi, Nancy, and Jean- their help has been essential.

Finally, I would like to thank my family. I could not have completed this journey without their love and care. They have been an incredible support system.

ix TABLE OF CONTENTS

LIST OF FIGURES ...... xiii

LIST OF TABLES ...... xxi

LIST OF SCHEMES ...... xxii

LIST OF ABBREVIATIONS ...... xxiii

CHAPTER

I. INTRODUCTION AND BACKGROUND ...... 1

1.1 Redox Flow Batteries (Chapters II and III) ...... 1

RFB Design ...... 2

All-Vanadium RFB and Other Aqueous Systems ...... 5

Non-Aqueous RFB Systems ...... 7

Drawbacks of known RFBs ...... 10

Metallocenes in RFBs ...... 11

Metallocene Modification ...... 14

Non-Aqueous Metallocene RFB Systems ...... 17

Aqueous Metallocene RFB Systems ...... 20

Biredox RFB Systems ...... 21

RFB Conclusions ...... 24

1.2 1,3-Diiminoisoindoline as a Reagent (Chapters IV-VIII) ...... 25

Phthalocyanines...... 28

Expanded Phthalocyanines ...... 30

x Subphthalocyanines ...... 33

Hemiporphyrazines ...... 36

Bis(arylimino)isoindolines ...... 44

Semihemiporphryazines ...... 48

Phthalazines ...... 53

Conclusions on DII as a Reagent ...... 55

II. EVALUATING FERROCENE IONS AND ALL-FERROCENE SALTS FOR ELECTROCHEMICAL APPLICATIONS ...... 57 Introduction ...... 57

Experimental ...... 60

Results and Discussion ...... 68

Conclusions ...... 88

III. INVESTIGATIONS INTO AQUEOUS REDOX FLOW BATTERIES BASED ON FERROCENE BISULFONATE ...... 89 Introduction ...... 89

Experimental ...... 193

Results and Discussion ...... 100

Conclusions ...... 120

IV. BINDING A MERIDIONAL LIGAND IN A FACIAL GEOMETRY: A SQUARE PEG IN A ROUND HOLE ...... 122 Introduction ...... 122

Experimental ...... 125

Results and Discussion ...... 133

Conclusions ...... 141

V. 1,3-DIYLIDENEISOINDOLINES: SYNTHESIS, STRUCTURE, REDOX, AND OPTICAL PROPERTIES ...... 142 Introduction ...... 142

xi Experimental ...... 145

Results and Discussion ...... 154

Conclusions ...... 163

VI. THE SYNTHESIS OF A HEXAMERIC EXPANDED HEMIPORPHYRAZINE ...... 164 Introduction ...... 164

Experimental ...... 167

Results and Discussion ...... 173

Conclusions ...... 182

VII. BILIAZINE: A RING OPEN PHTHALOCYANINE ANALOG WITH A MESO HYDROGEN BOND ...... 183 Introduction ...... 183

Experimental ...... 186

Results and Discussion ...... 194

Conclusions ...... 205

VIII. SUBBILIAZINE: A CONTRACTED PHTHALOCYANINE ANALOG ...... 206 Introduction ...... 206

Experimental ...... 208

Results and Discussion ...... 219

Conclusions ...... 237

IX. SUMMARY ...... 238

REFERENCES ...... 242

APPENDICES ...... 287

APPENDIX A: PERMISSIONS ...... 288

xii LIST OF FIGURES

Figure Page

1.1.1: Schematic drawing of a redox flow battery ...... 2 1.1.2: Commonly studied RFB systems with their corresponding half-cell reactions and working potentials from reference 28...... 6 1.1.3: The Co2+ complex and redox couples from reference 44 ...... 8 1.1.4: Examples of organic molecules used in RFBs ...... 9 1.1.5: Schematics of an all-organic RFB from reference 53 ...... 10 1.1.6: The redox reactions of ferrocene and cobaltocene ...... 13 1.1.7: The synthesis of ferrocene bis sulfonate and its crystal structure (CCDC: 1487629) from reference 97 ...... 15 1.1.8: The synthesis of ferrocene bis sulfonyl chloride and ferrocene bis sulfonamides from references 97, 100-106 ...... 16 1.1.9: Redox couples and their voltage separations from reference 107 ...... 18 1.1.10: The synthesis of the tetraferrocene species from reference 22 ...... 19 1.1.11: Cationic and anionic ferrocene species from references 115-121 ...... 20 1.1.12: Biredox ferrocene species from references 128, 131 and 132 ...... 23

1.2.1: The synthesis of DII and the structures of its tautomers (CCDC 228419, 228420, 914961) ...... 25 1.2.2: Examples of DII based macrocycles and chelates ...... 27 1.2.3: The X-ray structure of CuPc (CCDC 1133493) from two viewpoints. View 1: perpendicular to least squares plane, and view 2: along least squares plane. Gray, light blue, and orange spheres represent carbon, nitrogen, and copper respectively. Hydrogen atoms have been omitted for clarity...... 29

1.2.4: The X-ray structures of SPcUO2 and the rectangle expanded Pc from reference 176 and 177 (R groups and hydrogen atoms have been omitted for clarity) (CCDC 1125690 and 830031). View 1: side-on view and view 2: top-

xiii down view. Gray, light blue, red, turquoise, and dark blue spheres represent carbon, nitrogen, oxygen, molybdenum, and uranium respectively...... 32

1.2.5: The synthesis and X-ray structure of chloro-SubPc (CCDC 1232697)...... 34 1.2.6: The synthesis of asymmetric Pcs from DII (reference 194) ...... 35 1.2.7: The synthesis of hemiporphyrazine ...... 36

1.2.8: The X-ray structures of NiHp and NiHp(Py)2 (CCDC 1220437 and 1124972). View 1: side-on view and view 2: top-down view. Gray, light blue, and green spheres represent carbon, nitrogen, and nickel respectively. Hydrogen atoms have been omitted for clarity ...... 37 1.2.9: The X-ray structures of the 20π electron hemiporphyrazine, and the oxidized 18π electron system from reference 200 (CCDC 848549 and 848550). View 1: side-on view and view 2: top-down view. Gray, light blue, red, and white spheres represent carbon, nitrogen, oxygen, and hydrogen respectively .... 39 1.2.10: The synthesis of dicarbahemiporphyrazine and benziphthalocyanine from references 205 and 207 ...... 40 1.2.11: The X-ray structures of Ni and Zn benziphthalocyanine complexes from references 207 and 209 (CCDC 726129 and 750085). Hydrogen atoms have been omitted for clarity. Gray, light blue, green and blue-gray spheres represent carbon, nitrogen, nickel, and zinc respectively ...... 41 1.2.12: The X-ray structures of systems synthesized from 3-unit intermediates from references 217 and 221 (CCDC 1953097 and 1496230). Gray and light blue spheres represent carbon and nitrogen respectively ...... 42 1.2.13: The synthesis of an expanded Pc analog from reference 222 ...... 43 1.2.14: The synthesis of an expanded hemiporphyrazine from reference 224. .. 44 1.2.15: The X-ray structures of various BAI chelates (CCDC 218623, 756935, 824820). Gray, light blue, and yellow spheres represent carbon, nitrogen, and sulfur respectively ...... 45 1.2.16: The synthesis of BAI ligands using Siegl conditions ...... 46 1.2.17: The structures of various BAI metal complexes. Top left: homoleptic Co(BPI)2 complex, and heteroleptic Fe (top right), and Cu (bottom) BAI complexes. (CCDC 256231, 939329, 1120882, 831064). Hydrogen atoms have been omitted for clarity. Gray, light blue, green, yellow, purple, orange (large) and orange (small) spheres represent carbon, nitrogen, chlorine, sulfur, cobalt, iron, and copper respectively ...... 47 1.2.18: The X-ray structures of semihemiporphyrazines from reference 156. (CCDC 1527134-1527136). Hydrogen atoms have been omitted for clarity. xiv Gray, light blue, green, yellow, red, and dark blue spheres represent carbon, nitrogen, chlorine, sulfur, oxygen, and rhenium respectively...... 49 1.2.19: The X-ray structures of the aza(dibenzo)dipyrromethene complexes of B (top) and Re (bottom) from references 242 and 243. Gray, light blue, red, pink, and dark blue spheres represent carbon, nitrogen, oxygen, boron, and rhenium respectively ...... 50 1.2.20: The synthesis of the 1:1 product, asymmetric BAI, and Pd(BAI) (along with its X-ray structure) from reference 240 (CCDC 729236). Hydrogen atoms have been omitted for clarity ...... 51 1.2.21: The synthesis of boxmi ligands, and the structure of an exemplary boxmi ligand from reference 245 (CCDC 830007). Hydrogen atoms have been omitted for clarity ...... 52 1.2.22: The ring expansion of BAIs using hydrazine ...... 53 1.2.23: The synthesis of the Ni phthalazine complex from reference 248 ...... 54 1.2.24: The X-ray structures of binuclear phthalazine complexes of Cu (left) and Fe (right). (CCDC 1129164 and 990912). Hydrogen atoms have been omitted for clarity. Hydrogen atoms have been omitted for clarity. Gray, light blue, green, yellow, orange (large) and orange (small) spheres represent carbon, nitrogen, chlorine, sulfur, iron, and copper respectively ...... 55 2.1: The structures of all-ferrocene salts 2.1–2.4...... 59

1 2.2: H NMR (500 MHz) of 2.1 in d6 – DMSO ...... 65

1 2.3: H NMR (500 MHz) of 2.2 in d6 – DMSO ...... 65

1 2.4: H NMR (500 MHz) of 2.3 in d6 – DMSO ...... 66

1 2.5: H NMR (500 MHz) of 2.4 in d6 – DMSO ...... 66 2.6: The structures of compounds 2.1-2.4 with 35% thermal ellipsoids. Hydrogen atom positions have been omitted for clarity ...... 70 2.7: Cyclic voltammograms of single ferrocene salts (2 mM) measured on 2mm dia. Pt at scan rate 60 mV/s in aqueous solution with 0.2M KCl ...... 75 2.8: Cyclic voltammograms of single ferrocene salts (2 mM) measured on 10μm dia. Pt at scan rate 20 mV/s in aqueous solution with 0.2M KCl ...... 76 2.9: Cyclic voltammograms of single ferrocene salts (2 mM) measured on 2mm dia. Pt at scan rate 60 mV/s in PC solution with 0.2M TBAPF6 ...... 77 2.10: Cyclic voltammograms of single ferrocene salts (2 mM) measured on 10μm dia. Pt at scan rate 20 mV/s in PC solution with 0.2M TBAPF6 ...... 78 2.11: Cyclic voltammograms of single ferrocene salts (2 mM) measured on 2mm dia. Pt at scan rate 60 mV/s in DMF solution with 0.2M TBAPF6...... 79 xv 2.12: Cyclic voltammograms of single ferrocene salts (2 mM) measured on 10μm dia. Pt at scan rate 20 mV/s in DMF solution with 0.2M TBAPF6 ...... 80 2.13: Cyclic voltammograms of all-ferrocene salts 2.1 – 2.4 (2 mM) measured on 2mm dia. Pt at scan rate 60 mV/s. Top: aqueous solution with 0.2M KCl; Middle: PC solution with 0.2M TBAPF6; Bottom: DMF solution with 0.2M TBAPF6. Salts 1 and 2: red solid curves; salts 3 and 4: blue dotted curves. Ferrocene methanol has relative potential of -29 mV and -24 mV to ferrocene in PC and DMF respectively ...... 83 2.14: Cyclic voltammograms of all-ferrocene salts 2.11 – 2.4 (2 mM) measured on 10μm dia. Pt at scan rate 20 mV/s. Top: aqueous solution with 0.2M KCl; Middle: PC solution with 0.2M TBAPF6; Bottom: DMF solution with 0.2M TBAPF6. Salts 1 and 2: red solid curves; salts 3 and 4: blue dotted curves. Ferrocene methanol has relative potential of -29 mV and -24 mV to ferrocene in PC and DMF respectively ...... 84 2.15: Cyclic voltammograms of bis sulfonate ammonium salt (2 mM) in DMF-water mixtures with molar fraction of water 1 (red curve), 0.97 (blue), 0.74 (brown), 0.32 (green) and 0 (purple). No supporting electrolyte was added to the mixtures. Working electrode – 10 μm Pt disk, scan rate 20 mV/s ...... 86 2.16: Limiting current (data points) and inverse solution viscosity (curve) as a function of the molar fraction of water (X) in the DMF-water mixtures. The values of the limiting current were determined from the results in Fig. 2.15. The viscosity data were taken from reference 290 ...... 87 3.1: Schematic diagram of an “all-aqueous” RFB ...... 92 3.2: 1H NMR (300 MHz) of 1,1’-FcDS in d6-DMSO ...... 97 3.3: Solubility tests using UV–visible spectrum. A and B: Calibration curves for the relationship between UV–visible absorbance peak and concentration of 1,1’- FcDS in 1 M 1 M NaNO3 with (A) and without (B) the addition of 0.5 M EG. C and D: UV–visible spectra of diluted supernatant of 1,1’-FcDS saturated solutions prepared in in 1 M 1 M NaNO3 with (C: dilute 100 times) and without (D: dilute 200 times) the addition of 0.5 M EG ...... 99 3.4: Anolyte (2,7-AQDS): anthraquinone-2,7-disulfonic acid disodium salt (left). Catholyte (1,1’-FcDS):1,1’-bis(sulfonate)ferrocene disodium salt (right). The corresponding crystal structures are shown at the bottom of the diagram with 35% thermal ellipsoids. Hydrogen atoms have been omitted for clarity .... 101 3.5: CVs of 2 mM 2,7-AQDS (A) or 2 mM 1,1’-FcDS (B) in an aqueous solution containing 1 M NaNO3 as a supporting electrolyte. Working electrode: 3 mm dia. glassy carbon, reference electrode: Ag|AgCl|KCl (2 M), counter electrode: platinum wire. From inner curve to outer one, the scan rate varies from 20 mV/s to 65 mV/s...... 104

xvi 3.6: CVs of 2 mM 1,1’-FcDS in aqueous solution with 0.5 M EG. 1 M NaNO3 was added as supporting electrolyte. Working electrode: 3 mm dia. glassy carbon, reference electrode: Ag|AgCl|KCl (2 M), counter electrode: platinum wire. From inner curve to outer one, the scan rate varies from 20 mV/s to 65 mV/s ...... 105 3.7: CVs of 2 mM 2,7-AQDS (A) or 2 mM 1,1’-FcDS (B) in aqueous solution. 2 M acetate buffer (pH:4.53) was added as supporting electrolyte. Working electrode: 3 mm dia. glassy carbon, reference electrode: Ag|AgCl|KCl (2 M), counter electrode: platinum wire. From inner curve to outer one, the scan rate varies from 20 mV/s to 65 mV/s ...... 106 3.8: CVs of 3 mM 2,7-AQDS (A) or 3 mM 1,1’-FcDS (B) in aqueous solution. 0.5 M H2SO4 was added as supporting electrolyte. Working electrode: 3 mm dia. glassy carbon, reference electrode: Ag|AgCl|KCl (2 M), counter electrode: platinum wire. From inner curve to outer one, the scan rate varies from 20 mV/s to 65 mV/s ...... 107

3.9: 1,1’-FcDS/2,7-AQDS RFB using 1 M NaNO3 as the supporting electrolyte (0.5 M EG added). A: Ten charge and discharge cycles (#2 to #11 cycles) at constant current 25 mA (2.8 mA/cm2); B: capacity vs cycle number; C: Coulombic efficiency (CE), energy efficiency (EE) and voltage efficiency (VE) vs cycle number ...... 110 3.10: 100 charge and discharge cycles at constant current 25 mA for 1,1’- FcDS/2,7-AQDS RFB using 1 M NaNO3 as supporting electrolyte (0.5 M EG added)...... 111

3.11: 1,1’-FcDS/2,7-AQDS RFB using 2 M acetate buffer as supporting electrolyte (0.5 M EG added). A: Ten charge and discharge cycles (#2 to #11 cycles) at constant current 25 mA; B: capacity vs cycling number; C: CE, EE and VE vs cycling number...... 112 3.12: 100 charge and discharge cycles at constant current 25 mA for 1,1’- FcDS/2,7-AQDS RFB using 2 M acetate buffer as supporting electrolyte (0.5 M EG added)...... 113

3.13: 1,1’-FcDS/2,7-AQDS RFB using 0.5 M H2SO4 as supporting electrolyte. A: Ten charge and discharge cycles (#2 to #11 cycles) at constant current 25 mA (2.8 mA/cm2); B: capacity vs cycling number; C: CE, EE and VE vs cycling number...... 115 3.14: UV-visible spectra for 1,1’-FcDS, basic iron(III) acetate, and the 1,1’-FcDS (2 M acetate buffer) decomposition solution in water...... 116

3.15: UV-visible spectra for 1,1’-FcDS, FeCl3 (1M H2SO4), and the 1,1’-FcDS (0.5 M H2SO4) decomposition solution in water ...... 117 3.16: CVs of catholyte (1,1’-FcDS) and anolyte (2,7-AQDS) after 100 cycles of charge/discharge. A: using 1 M NaNO3 as supporting electrolyte, scan rate: 30 xvii mV/s; B: using 2 M acetate buffer as supporting electrolyte, dashed curve: CVs of iron(III) acetate in 2 M acetate buffer, scan rate: 30 mV/s; C: using 0.5 M H2SO4 as supporting electrolyte, dashed curve: CVs of iron(III) sulfate in 0.5 M H2SO4, scan rate: 25 mV/s. In all experiments, working electrode: 3 mm dia. glassy carbon, reference electrode: Ag|AgCl|KCl (2 M), counter electrode: platinum wire...... 119

4.1: Synthesis of Re(CO)3(BAI) complexes 4.1-4.3 ...... 124

1 4.2: H NMR (300 MHz) of 4.1 in CDCl3 ...... 129

1 4.3: H NMR (300 MHz) of 4.2 in CDCl3 ...... 130 4.4: 1H NMR (500 MHz) of 4.3 in d6-DMSO ...... 131 4.5: Structure of compounds 4.1-4.3, with 35% probability ellipsoids. Hydrogen atoms except on nitrogen atom positions have been omitted for clarity. .... 135

4.6: UV-visible spectra for compounds 4.1-4.3 in CHCl3 (DMF for 4.3) ...... 137 4.7: Experimental and TDDFT predicted spectra for compounds 4.1 (top) and 4.3 (bottom)...... 138 4.8: Experimental and TDDFT-predicted molecular orbital energy levels for compounds 4.1 and 4.3. The structures of the HOMOs and LUMOs are shown along with percent metal/ligand composition ...... 140

1 5.1: H NMR of 5.1 in CDCl3...... 149

1 5.2: H NMR of 5.2 in CDCl3...... 150

1 5.3: H NMR of 5.3 in CDCl3 ...... 151

1 5.4: H NMR of 5.4 in CDCl3 ...... 152 5.5: The structures of compounds 5.1, 5.2, and 5.4 with 35% thermal ellipsoids. Non-ionizable hydrogen atoms have been omitted for clarity...... 156 5.6: UV-visible spectra for compounds 5.1-5.4 in DCM ...... 157

5.7: Cyclic voltammograms for compounds 5.1-5.4 recorded in DMF/0.1 TBAPF6 system at room temperature. Redox potentials (V) versus FcH/FcH+ are displayed in the table ...... 158 5.8: DFT-predicted frontier orbitals (top) and energy levels (bottom) for compounds 5.1-5.4 ...... 160 5.9: Experimental and TDDFT-predicted spectra for compounds 5.1-5.4 ...... 162 6.1: Hemiporphyrazine and several expanded macrocycles : hexaphyrin (A), superphthalocyanine (B), hemiporphyrazine (C), thiadiazole-expanded hemiporphyrazine (D), and hexahemiporphyrazine (E) ...... 165

xviii 1 6.2: H NMR (500 MHz) of 6.1 in d6 – DMSO ...... 170

1 6.3: H NMR (500 MHz) of 6.2 in d6 – DMSO ...... 171 6.4: Structure of 2,6-diaminopyridine HCl salt (DAP·HCl), showing 35% probability ellipsoids. Hydrogen atoms on carbon positions have been omitted for clarity ...... 174 6.5: The structure of bis(6-amino-2-pyridyl)amine (6.1, top) and its HCl salt (6.1·HCl, bottom) with 35% thermal ellipsoids. Hydrogen atoms on carbon positions and counterions have been omitted for clarity ...... 176 6.6: The three possible conformations of bis(6-amino-2-pyridyl)amine ...... 178 6.7: 1H NMR spectrum of hexahemiporphyrazine 6.2, with a close up of the isoindoline and pyridine C-H resonances ...... 179 6.8: The structure of hexahemiporphyrazine 6.2 with 35% thermal ellipsoids. Hydrogen atoms on carbon positions have been omitted for clarity ...... 180 6.9: The UV-visible spectra of bis(6-amino-2-pyridyl)amine (6.1), hexahemiporphyrazine (6.2), and diiminoisoindoline in dimethylsulfoxide (DMSO)...... 181 7.1: The structures of phthalocyanine (left), bis-pyridyl hemiporphyrazine (middle) and biliazine (H2BlzH, right) ...... 184 7.2: The synthesis of 7.1 and the four biliazine compounds described in this report. Top right: The structure of compound 7.1 with 35% thermal ellipsoids. Hydrogen atoms on carbon positions and atom labels are not shown ...... 185 7.3: 1H NMR (500 MHz) of 7.1 in d6-DMSO ...... 190 7.4: 1H NMR (500 MHz) of 7.5 in d6-DMSO ...... 191

7.5: The structures of H2BlzH (7.2), Cu(BlzH) (7.3), Co(BlzH)(MeOH)2 (7.4), and Zn(BlzH)(MeOH)2 (7.5) with 35% thermal ellipsoids. Hydrogen atoms on carbon and oxygen positions have been omitted for clarity ...... 195

7.6: The UV-visible spectra of H2BlzH and its metal complexes in DMF solution ...... 198 7.7: UV−Vis and MCD spectra of 7.5 ...... 199

7.8: Cyclic voltammograms in DMF/0.1 TBAPF6 ...... 200 7.9: B3LYP α-spin and β-spin relative energies of the frontier orbitals for compounds 7.2-7.5 ...... 202 7.10: B3LYP DFT-predicted frontier orbitals for compounds 7.2-7.5 ...... 203 7.11: Experimental and B3LYP TDDFT-predicted spectra for compounds 7.2- 7.5 ...... 204 xix 8.1. The structures of phthalocyanine, and biliazine (top), subphthalocyanine, and subbiliazines (bottom)...... 207 8.2: 1H NMR (300 MHz) of 8.1 in d6-DMSO ...... 213 8.3: 1H NMR (300 MHz) of 8.2 in d6-DMSO ...... 214

1 8.4: H NMR (300 MHz) of 8.1BF in CDCl3 ...... 215

1 8.5: H NMR (300 MHz) of 8.2BF in CDCl3 ...... 216 8.6: 1H NMR (300 MHz) of 8.3 in d6-DMSO ...... 217 8.7: The structures of free bases 8.1, and 8.2, subbiliazine 8.1BF, hydrolysis product 8.3, and partial structure elucidation of 8.2BF with 35% thermal ellipsoids. Hydrogen atoms on carbon positions have been omitted for clarity ...... 221

8.8: UV-visible spectra for compounds 8.1BF and 8.2BF in CHCl3...... 224 8.9: UV-visible spectra for compounds 8.1 and 8.2 in DMF ...... 225

8.10: Absorption and excitation spectra for compounds 8.1BF and 8.2BF in CHCl3 ...... 226 8.11: The normalized absorption (solid) and emission (dashed) spectra for compound 8.1BF, and 8.2BF in CHCl3 ...... 228 8.12: The normalized absorption (solid) and emission (dashed) spectra for compound 8.3 in DMF ...... 229 8.13: UV-visible spectra for compound 8.3 in DMF ...... 230 8.14: Cyclic voltammograms of 8.1, 8.2, 8.1BF, and 8.2BF in DMF/0.1 TBAPF6...... 231 8.15: Relative energies of the frontier orbitals for compounds 8.1, 8.2, 8.1BF, and 8.2BF ...... 232 8.16: DFT-predicted frontier orbitals for compounds 8.1, 8.2, 8.1BF and 8.2BF ...... 233 8.17: Experimental and B3LYP TDDFT-predicted spectra for compounds 8.1, 8.2, 8.1BF, and 8.2BF ...... 235 8.18: Experimental and B3LYP TDDFT-predicted excitation and emission spectra for compounds 8.1BF and 8.2BF ...... 236

xx LIST OF TABLES

Table Page

2.1: Crystal data and structure refinement parameters of compounds 2.1-2.4 ... 67 2.2: Solubilities of compounds 2.1-2.4 (units: mM) at 293 K and 1 atm ...... 72 2.3: Half-wave potentials in mV versus ferrocenium methanol/ ferrocene methanol couple using a 2 mm dia. Pt electrode. The values were determined by averaging the peak potentials, unless a clear irreversibility is observed, in which case we estimated the half-wave potential values ...... 73 3.1: X-ray crystal data and structure parameters for compounds 2,7-AQDS and 1,1’-FcDS ...... 98 3.2: Solubility and capacity data of 1,1’-FcDS and 2,7-AQDS ...... 103 3.3: Estimated electrochemical data for 1,1’-FcDS and 2,7-AQDS ...... 108 4.1: X-ray crystal data and structure parameters for compounds 4.1-4.3 ...... 132 5.1: X-ray crystal data and structure parameters for compounds 5.1, 5.2, and 5.4 ...... 153 6.1: X-ray crystal data and structure parameters for compounds DAP·HCl, 6.1, 6.1·HCl, and 6.2 ...... 172 7.1: X-ray crystal data and structure parameters for compounds 7.1 and 7.2 .. 192 7.2: X-ray crystal data and structure parameters for compounds 7.3-7.5 ...... 193 8.1: X-ray crystal data and structure parameters for compounds 8.1, 8.2, 8.1BF, and 8.3 ...... 218

xxi LIST OF SCHEMES

Scheme Page

5.1: Synthetic route for preparation of the compounds 5.1-5.4 ...... 144 6.1: Synthesis of 6.1 and 6.2 ...... 175 8.1. The synthesis of 8.1 and 8.2, the subbiliazine compounds (8.1BF and 8.2BF), and the hydrolysis product (8.3)...... 220

xxii LIST OF ABBREVIATIONS Acac- acetylacetone Acenen- bis(acetylacetone)ethylenediamine ADBM- aza-dibenzodipyrromethene BAI- bis(arylimino)isoindolines BODIPY- boron dipyrromethene BOPHY- bis(difluoroboron)1,2-bis((pyrrol-2-yl)methylene)hydrazine Boxmi- bis(oxazolinylmethylidene)isoindolines BPI- bis(2-pyridylimino)isoindoline Bpy- 2,2’-bipyridine Cp- cyclopentadiene

DCM/ CH2Cl2- dichloromethane DES- deep eutectic solution Dhph-1,4-dihydrazinophthalazine DII- 1,3-diiminoisoindoline DIPEA- N,N-diisopropylethylamine DMF- dimethylformamide DMSO- dimethylsulfoxide HOMO- highest occupied molecular orbital Hp-hemiporphyrazine LUMO- lowest unoccupied molecular orbital MeCN/ACN- acetonitrile MLCT- metal-to-ligand charge transfer PC- propylene carbonate

xxiii Pc- phthalocyanine Py-pyridine RFB- redox flow battery

SPcUO2- superphthalocyanine SubPc- subphthalocyanine TAP- tetraazaporphyrin

TBAPF6- tetrabutylammonium hexafluorophosphate TEMPO- (2,2,6,6-tetramethylpiperidin-1-yl)oxyl Tpy- terpyridine VRB- all-vanadium redox battery

xxiv CHAPTER I

Introduction and Background

1.1 Redox Flow Batteries (Chapters II and III)

Research in the area of renewable energy sources continues to grow as a result of the world’s increased electricity needs. The most common forms of renewable energy are wind and solar power.1,2 However, these methods are intermittent and can have some adverse environmental impacts, along with high initial costs and continuing repair needs.3–5 Energy storage devices such as rechargeable batteries offer a solution to the intermittent nature of some renewable energy forms. The first example of a rechargeable battery was the lead-acid battery developed by Gaston Planté in 1860.6 Eventually, the lithium-ion battery, designed for small electronics and electric vehicles, was developed and commercialized in 1991 by Sony Corporation.7 Since then, battery systems have seen improvements in cost and materials, but the “ideal” battery has yet to be discovered. The need for newer energy storage technology has led to a class of battery called the redox flow battery (RFB). The RFB has great potential in the energy sector as they can be a lower cost alternative to lithium-ion batteries. The

1 design of the RFB can also have large scale capabilities, and greater design flexibility.8–11

1.1.1 RFB Design

Figure 1.1.1 portrays a schematic diagram of a redox flow battery cell.12

Two separate tanks hold the electroactive components dissolved in solution, which can be either aqueous or non-aqueous. A pump then drives the two component solutions into the electrochemical cell and which are kept separate by a membrane. The redox reaction then occurs at the electrodes which are connected to a power source.10,13

Figure 1.1.1: Schematic drawing of a redox flow battery. Reprinted with permission from J. Vac. Sci. Technol. B, Nanotechnol. and Microelectron. Mater. Process.

Society. 2

In a redox flow battery schematic, the two electroactive materials stored in tanks are known as the anolyte and catholyte. The catholyte is stored in the catholyte tank, or the positive electrolyte tank, whereas the analyte is retained in the anolyte tank, or the negative electrolyte tank.14 These materials are pumped into an electrochemical cell, containing two electrodes separated by a membrane.

During the charging process the catholyte solution is oxidized and loses an electron to the cathode, flowing across the membrane to the anolyte half-cell where a reduction occurs. The discharge process is thermodynamically favorable, and during this process the reverse reactions occur.10 In a system where the charge carrier is a proton (H+), the charging process of the battery occurs when the electrolyte solution with H+ generated in the positive half-cell diffuses across the semipermeable membrane to the negative half-cell to achieve an electrical equilibrium of the electrolyte solution. The stored power can be discharged in the reverse process resulting in the battery cycle.13

The membrane separating the two electroactive components is an important element of the battery design. The membrane allows for charge balance between the catholyte and anolyte solutions via migration of ions across the membrane. Ideally, the membrane also prevents crossover of the electroactive components, which leads to mixing of the materials and shortening of the life of the battery. The most common type of membrane used in a RFB is an ion- exchange membrane, which can be either cation-exchange or anion-exchange.15

This type of membrane is semi-permeable and allows for either a specific or a

3 range of charged species to pass through, while repelling the oppositely charged molecules.16,17 Cation-exchange membranes permit small charge carrier species like H+, and Na+ to pass through.15,18,19 Additionally, species like OH-, or Cl-, can pass through an anion-exchange membrane in an analogous manner.20

An alternative method to an ion-exchange membrane is a size-exclusion porous membrane.21–24 Porous membranes are an inexpensive alternative to ion exchange membranes, and are typically made from low cost materials like polypropylene, polyethylene, or cellulose acetate. The pore radii for these membranes are nano-sized and typically range from ~5 to 21 nm. For these systems, the electroactive species must either be in the form of a polymer, or be sterically large molecules to avoid crossover in these membranes.25

The development of flow batteries containing redox active polymeric materials paired with a size-exclusion membrane are becoming an increasingly important area of research. For these systems, it is common to make copolymers with alternating electroactive moieties like (2,2,6,6-tetramethylpiperidin-1-yl)oxyl

TEMPO, or viologens.21 A single material that can act as both electroactive components allows for a symmetric RFB setup as the same material can be used in each side of the electrochemical cell. Since these polymeric RFB systems can utilize cheap membranes, there is the possibility for scalable applications. The components don’t display crossover contamination, and have respectable cycling performances. However, some polymers used in these systems have poor solubilities in comparison to their monomer starting materials, resulting in lowered

4 energy densities.21,26,27 Polymeric systems also show poor performance due to viscosity issues.26

1.1.2 All-Vanadium RFB and Other Aqueous Systems

The all-vanadium redox battery (VRB) is the most studied RFB that is used commercially. The system was developed in 1986, and utilizes vanadium in four different oxidation states (V(V), V(IV), V(III), V(II))28 usually in the form of

29 vandadium oxides such as V2O5. Figure 1.1.2 shows the redox reactions occurring at each electrode where a vanadium species acts as both catholyte and anolyte. By using a single metal component, cell compartment leakage is not a catastrophic flaw, and battery operation can proceed without crossover failure.30,31

The cell contains concentrated sulfuric acid (~ 4 M) that solubilizes the components and stabilizes the vanadium species in solution. A cation-exchange membrane allows for the passage of H+ cations during the battery cycling, and a working voltage of 1.26 V vs. RHE can be achieved.28

Figure 1.1.2: Commonly studied RFB systems with their corresponding half-cell reactions and working potentials from reference 28.

Another aqueous RFB based entirely on metal components is the iron- chromium system (Figure 1.1.2). Developed by NASA in 1975, this battery requires a single-electron transfer, simplifying the charge transfer process.9,32,33

The total capital cost of the iron-chromium battery is cheaper than the all-vanadium system, and this design can achieve higher power densities. However, unlike a single metal component system, the two electroactive materials in the iron- chromium battery often results in crossover contamination and capacity decay.32

An additional example of an aqueous RBF is the vanadium-bromine system

(Figure 1.1.2). The solubility of V2O5 is more limited than in the all-vanadium system, and maximum solubility never reaches > 2 M. With the addition of halide ions (Br- and Cl-), the solubility can be improved. Replacing the V4+/V5+ redox couple with the bromide/polyhalide redox couple results in concentrated solutions up to 4 M in hydrochloric acid.34 Overall, this alteration to the previously known

VFB displays how modifications to the electroactive species can potentially enhance battery performance.

1.1.3 Non-Aqueous RFB Systems

Most battery designs in the literature use water as the solvent and are thus examples of aqueous RFBs. In 1984, Singh first proposed the idea of using nonaqueous solvents in a RFB system.35 The choice of solvent is important in battery design, and can alter factors like the working temperature range, energy density, and cell voltage.36 Furthermore, a wider voltage potential window can be achieved by some organic solvents in contrast to water. Solvents like propylene carbonate

(PC) are of interest due to its high dielectric constant. However, PC has a high viscosity resulting in low conductivity. Therefore, mixing PC with solvents like ethers or acetonitrile (ACN) can improve performance.35

Both metal complexes and organic-based molecules have been used in non-aqueous RFBs.37,38 Some of the first non-aqueous RFBs used metal complexes of Fe,39,40 Ni,40 Mn,41 V,42,43 Ru,39 Co,44 and Cr43 coordinated by chelating ligands such as the bidentate 2,2’-bipyridine (bpy), acetylacetone (acac), and tridentate terpyridine (tpy).36,45 Because the redox reactions occur at the metal site, a large degree of tunability can be achieved by altering the nature of the ligands and metal center.

Zhang, Lan, and Li demonstrated the use of bis(acetylacetone)ethylenediamine cobalt(II) (Co(II)(acacen)) in a battery with

ACN as the solvent.44 The redox couples Co3+/Co2+ and Co2+/Co+ are shown in

Figure 1.1.3. Overall, the cell potential was found to be ~2 V, and the redox reactions were highly reversible within range of temperatures between 0-50 °C.44

Figure 1.1.3: The Co2+ complex and redox couples from reference 44.

Alternatively, purely organic compounds have been used in RFBs.38,46

These devices include both aqueous and non-aqueous solvents, but the latter tends to be more common.47 Unlike the metal-based systems, the redox pair is often in the form of a stable radical. Figure 1.1.4 portrays a series of well-studied organic molecules. Molecules like TEMPO,48–52 organics with the formula

n+ 53,54 55–59 60–62 (C5H4NR)2 (viologens), anthraquinones, and phenothiazines, along with others have been candidates as battery electrolytes.

Figure 1.1.4: Examples of organic molecules used in RFBs.

In 2016, Liu, Wei, Sprenkle, and Wang reported on a battery system that used 4-hydroxy TEMPO as the catholyte paired with methyl viologen as the anolyte.53 Shown in Figure 1.1.5, this provides an example of an “all-organic” based redox flow battery. The battery was cycled 100 times with almost no capacity loss in a non-corrosive electrolyte containing 1-1.5 M NaCl.53 However, the maximum solubility of the TEMPO could only reach 0.3 M in solution, while the methylviologen solubility was much higher at 3 M.

John Wiley and Sons.

1.1.4 Drawbacks of known RFBs

There is currently no “ideal” redox flow battery system on the market today.

There are many drawbacks to each of the aqueous systems previously mentioned including solubility issues, component toxicity, crossover contamination, and high materials costs.63 Solubility of the components continues to be an issue as the all- vanadium battery has a maximum solubility of < 1.7 M, and the vanadium pentoxide can precipitate from solutions if temperature rise above 40 °C.63

Whereas the solubility of vanadium batteries can be increased by adding halide ions, toxicity becomes an additional issue with halides. The vanadium-bromine

10 system runs the risk of forming toxic bromine vapors during operation.

Compartment leakage in multi-component systems like the iron-chromium system is prominent and cannot always be avoided. An ion exchange membrane may only allow for H+ or Cl- to pass, but the corrosive nature of the electrolyte (HCl) often results in degradation of components leading to crossover of iron and chromium ions.10,64

Non-aqueous RFBs suffer from the same issues as their aqueous counterparts. Solubility tends to be less than in aqueous solvents and

46 concentrations typically don’t reach above 1 M. Metal complexes like Mn(acac)3,

Fe(bpy)3, and Ni(bpy)3 can only reach concentrations of 0.6 M, 0.4 M and 0.2 M respectively in non-aqueous solvents.40,41 Additionally, toxicity of both electroactive materials and solvent should also be considered when designing a battery system. The compound methylviologen, also known as paraquat, is a pesticide that has known cytotoxicity.65,66 Chlorinated solvents like dichloromethane and chloroform are highly volatile and pose both environmental hazards and negative human health effects, which is why there is a higher demand for water soluble battery systems.67,68

1.1.5 Metallocenes in RFBs

Metallocenes have been proposed at potential electroactive components for redox flow battery systems. Ferrocene was the first metallocene synthesized, and was discovered in 1951 by Kealy, Paulson, and Miller.69 The initial reactions were

11 carried out by reacting cyclopentadienylmagnesium bromide (II) with FeCl3, or by the direct reaction of reduced iron with cyclopentadiene at 300 °C. The resultant compound from these reactions is an air stable, 18 electron organometallic complex.70 The Fe(II) center is sandwiched between two cyclopentadienyl anions and undergoes a highly stable and reversible redox process. In its redox couple, shown in Figure 1.1.6, the metal undergoes oxidation from Fe2+/Fe3+ and is often used as a reference electrode potential for calibration of voltaic devices.70

Cobaltocene was the next metallocene to be discovered, but unlike ferrocene which is air stable, cobaltocene is reactive to O2 and must be stored and handled an under inert atmosphere. Cobaltocene shown in figure 1.1.6 is an electron rich 19 electron complex, and oxidation of Co2+ to Co3+ in cobaltocenium forms, which is a highly stable 18 electron cation.71 Similar to ferrocene, the redox couple is also highly stable and reversible, and cobaltocene is also often utilized as an internal electrochemical standard, oxidizing at a potential of -1.33 V relative to SHE.72

Figure 1.1.6: The redox reactions of ferrocene and cobaltocene.

Whereas ferrocene and cobaltocene are the most studied metallocenes, similarly structured sandwich complexes of V,73 Cr,74,75 and Ni76,77 and other metals have been synthesized.78 Metallocenes are not limited to classical sandwich complexes and the structures can be bent with additional ligands coordinated to the metal. Ligands can include halides, and in these compounds the halides can be substituted by various nucleophiles like hydrides.79 Additionally, the metallocene scaffold can contain a single Cp ring coordinated to the metal with additional ligands, known as a ‘piano-stool’ structure or half sandwich compound.80,81

The synthesis and applications of metallocenes have been well investigated and these diverse complexes have many promising uses in the areas of anticancer research,80 biosensors,82 and catalysis,83 and the use of metallocenes in RFB

13 devices is beginning to emerge as a popular area of research. Currently, only a few examples of ferrocene and cobaltocene in both aqueous and non-aqueous

RFB setups have been reported, though ferrocene is more prevalent in the literature. Whereas ferrocene demonstrates good performance in non-aqueous solvents, the major drawback to the ferrocene scaffold is its poor solubility in aqueous media, which is most desired for a cost effective and environmentally conscious battery design.

1.1.6 Metallocene Modification

The ferrocene framework can be modified in a variety of ways to increase both aqueous and non-aqueous solubility. The cyclopentadiene (Cp) rings can undergo alkylation,84 acetylation,85 arylation,86 formylation,87 sulfonation,88,89 and other reactions,90 similar to those of benzene.91 Cobaltocene is more limited with regard to aromatic substitution reactions. The Cp rings can be modified before

92 successive reaction with a CoCl2 salt. However, one can also functionalize the oxidized cobaltocenium salt, and this area has generated a lot of interest in recent years. There are still limitations to the reactivity of cobaltocenium, because the deactivating positive charge makes it challenging to modify through traditional reactions used for ferrocene. Nucleophilic substitution reactions are often required to obtain substituted cobaltocenium compounds.93–95

Modifying ferrocene with sulfonate groups is a potential way to enhance water solubility of ferrocene. Chapters II and III utilize the sulfonation chemistry of ferrocene to generate water soluble compounds for RFB applications. In 1958,

Knox and Pauson synthesized ferrocene bis sulfonic acid as shown in Figure

1.1.7.96 Ferrocene was reacted with two equivalents of chlorosulfonic acid in the presence of acetic anhydride to produce the bis sulfonic acid species. Whereas the sulfonic acid species is quite hygroscopic, the conjugate ammonium salt is easily isolated as a crystalline material.

Figure 1.1.7: The synthesis of ferrocene bis sulfonate and its crystal structure

(CCDC: 1487629) from reference 97.

These procedures were later optimized by the Ziegler group resulting in yields of

53 % and 95 % for the sulfonic acid and sulfonate salt respectively.97 The bis ferrocene sulfonate salt is an important precursor to ferrocene sulfonyl chloride, which can be further reacted with primary and secondary amines to generate

15 sulfonamides in the presence of mild, sterically bulky base like N,N- diisopropylethylamine (DIPEA) (Figure 1.1.8).

Figure 1.1.8: The synthesis of ferrocene bis sulfonyl chloride and ferrocene bis sulfonamides from references 97, 100-106.

The mono sulfonated ferrocene is more challenging to synthesize and initial reported reactions exhibited poor yields. Sulfonation with sulfur trioxide pyridine complex,98 sulfuric acid,99 and chlorosulfonic acid96 were all attempted resulting in mixtures of mono and bis products. Slocum and Achermann chose to exchange the acetic anhydride solvent for diethyl ether and cool the solution in an ice bath to reduce the yield of ferrocenium byproduct.99 Unlike the ferrocene bis sulfonic acid, isolation of the mono sulfonic acid was not attempted. The acid was generated in situ and was used to produce the mono ferrocene sulfonyl chloride using PCl3.

Like the bis sulfonyl chloride, the ferrocene mono sulfonyl chloride can be reacted with various amines to produce sulfonamides including bioconjugates,97,100,101 ligand scaffolds for metals,102,103 and redox active ionic liquids.104–106

1.1.7 Non-aqueous Metallocene RFB Systems

Redox flow battery systems that contain metallocenes as electroactive components were initially based on simple synthetic modifications to the metallocene scaffold or relied on completely unmodified molecules. A notable example includes early work by Hwang, Park, and Kim.107 In this report they prepared an “all-metallocene” based RFB.107 Unmodified ferrocene and bromoferrocene were used in the cathodic compartment whereas cobaltocenium, and decamethylcobaltocenium hexafluorophosphate were placed in the anodic compartment of the cell. The resultant systems had moderate solubility ~0.1-0.5 M in acetonitrile (ACN) and propylene carbonate (PC) solvent. Figure 1.1.9 displays the half wave potentials and the cell voltage separations they achieved. The unmodified ferrocene and cobaltocenium couple results in a voltage separation potential of ~1.33 V. To increase the voltage potential separation of the system, simple modifications were made to the metallocenes. An electron withdrawing bromide was added to ferrocene, resulting in an increase in redox potential, and electron donating methyl groups were added to cobaltocenium, resulting in a decrease in redox potential.107 The resultant separation potential is increased to

~2.05 V.

Figure 1.1.9: Redox couples and their voltage separations from reference 107.

This report in 2015 set the stage for metallocene-based battery research and since then, research containing batteries using ferrocene have been reported.108,109 Battery setups with simple synthetic modifications to the ferrocene scaffold can be found including 1,1’-dimethylferrocene,110 1-acetylferrocene,111 brominated ferrocenes,112,113 and ferrocene-containing Schiff bases.114 In 2017

Sisto and co-workers designed an elaborate system that combined a large tetraferrocene and perylene diimide separated by a size exclusion membrane.22

Figure 1.1.10 displays the synthesis of the tetraferrocene species. Pentaerythritol was deprotonated with NaH, and bromohexyl ferrocene was added, to afford the product as a brown oil (66 % yield).

Figure 1.1.10: The synthesis of the tetraferrocene species from reference 22.

The tetraferrocene species in Figure 1.1.10 is one of the first examples of a macromolecular ferrocene used in a RFB with a size exclusion membrane, and a crossover of only 0.05 % of the tetraferrocene species was discovered. Solubility in acetonitrile was limited to ~0.1 M for the tetraferrocene species, but each molecule contains four electron equivalents of redox active ferrocene. Overall, this system displayed long term stability with coulombic efficiency of ~ 99%, and an open circuit potential of 0.85 V.22

1.1.8 Aqueous Metallocene RFB Systems

Synthetic modifications can also be made to the ferrocene scaffold to increase water solubility. These modifications typically involve the generation of quaternary ammonium salts.115–120 Figure 1.1.11 shows various examples of cationic ferrocenes used in RBF devices which utilize an anion-exchange membrane. Additionally, Figure 1.1.11 gives an example of an anionic ferrocene using water soluble sulfonates.121

Figure 1.1.11: Cationic and anionic ferrocene species from references 115-121.

In 2016, a report by Liu and coworkers utilized (ferrocenylmethyl) trimethylammonium chloride in a RBF battery device paired with the methyl viologen cation as the anolyte.122 The (ferrocenylmethyl)trimethylammonium chloride was reported to reach concentrations up to 4 M in H2O, and 3 M in 2 M

NaCl, making it a highly water soluble compound for a RFB system.122 In some aqueous RFB systems such as the sulfonated anthraquinone/HBr system reported by Aziz and coworkers, current densities could reach up to 500 mA/cm2.123

However, their use of corrosive and toxic electrolytes present an additional obstacle, making it a less than ideal RFB system. While Liu and coworkers report moderate energy density in their system (60 mA/cm2), it is important to acknowledge the use of NaCl as a nontoxic electrolyte.122 Additionally, in regard to corrosivity, NaCl is a superior alternative to RFB systems that use corrosive electrolytes like concentrated sulfuric acid.

1.1.9 Biredox RFB Systems

Most RFBs contain two different electrolytes solutions with a membrane separator to prevent mixing. The use of a biredox pair is a potential solution to prevent crossover of materials. A biredox system contains either a single molecule functionalized with two redox-active moieties, or a single solution with both anolyte and catholye.124 When applied to a RFB, the same solution is used in both sides of the battery resulting in a symmetric battery cell. One way to achieve this is to

21 generate a salt where the anion and cation contain the anolyte and catholyte.125,126

These types of systems have yet to be implemented in a battery setup, but many salts have been synthesized and characterized containing moieties like

TEMPO,124 and anthraquinone.125,127

Neutral biredox systems can also be utilized in RFB setups.128 Molecules like BODIPY dyes129 and imidazoline oxides130 can undergo multiple reversible redox reactions within the same molecule, and have been used in single component biredox battery setups. Another way to generate a biredox system is to covalently connect the anolyte and catholyte to each other.

Figure 1.1.12 shows some examples of biredox ferrocenes covalently bound to phthalimides and anthraquinone.128,131,132 The Fc/Fc+ oxidation takes place at the positive electrode whereas the organic radical reductions occur at more negative potentials at the negative electrode. In the ferrocene/anthraquinone

RFB, the ferrocene is readily oxidized at -0.42 V while the anthraquinone is reduced at -1.01 V resulting in an open circuit potential of 1.42 V. The ferrocene/phthalimide systems have a larger open circuit potential of 1.94 V.

Although large voltages >1 V were achieved, the solubilities of the electrolytes are rather poor, reaching maximum concentrations of only 20-30 mM. Crossover contamination was successfully avoided using the same solution in each half of the cell, but an additional issue was present. The redox active components have the potential to react with each other.

In the ferrocene/anthraquinone system, it was discovered that side reactions involving the anthraquinone phenolate ion and electron-deficient molecules lead to an eventual decay in capacity of the battery.128 The determination of these side reactions and decomposition products are seldom discussed in the literature, but addressing these problems is imperative for the future of designing molecular scaffolds and determining working battery conditions.59,128

Figure 1.1.12: Biredox ferrocene species from references 128, 131 and 132.

As previously mentioned, a biredox system does not have to originate from a single molecule containing both redox active components. A biredox setup can contain a mixture of electroactive components in the same solution, producing a symmetric redox flow battery setup. Since low solubility still creates an issue in most covalent biredox RFBs, a deep eutectic solution (DES) is a potential remedy to the solubility problem.133,134 DESs are homogeneous metal-based or organic mixtures with melting points lower than their original components.135

In 2019, Yu and co-workers designed a eutectic ferrocene mixture for a nonaqueous battery cell, using dimethylferrocene (DMFc) and N-butylphthalimide.136

DMFc was first utilized in a RFB in 2017 by Lu and co-workers.110 By itself, DMFc has a melting point of 37-40 °C. Lu and co-workers demonstrated stable battery cycling when DMFc was heated to a constant 50 °C, and they observed that concentrations of 3 M were reached. Combining DMFc and N-butylphthalimide reduces the overall melting point to ~9 °C, creating a eutectic solution with concentrations reaching 3.5 M.136

1.1.10 RFB Conclusions

Redox flow batteries are a relatively new form of battery technology, with a little under five decades of research in the field. There is still no perfect battery electrolyte on the market, and research is spread out with focuses on both organic and inorganic compounds paired with both aqueous and nonaqueous solvent systems. Using metallocenes as components for RFBs offers another scaffold with incredible synthetic versatility and increasing the water solubility of these compounds allows for the development of more environmentally friendly battery systems.

1.2 1,3-Diiminoisoindoline as a Reagent (Chapters IV-VIII)

1,3-Diiminoisoindoline (DII) was first synthesized by Elvidge and Linstead in 1952,137 and this compound quickly became an important synthetic precursor to generate many chelates,138 macrocycles,139 and chromophores.137 As shown in

Figure 1.2.1, DII was first synthesized by reacting phthalonitrile with gaseous ammonia in methanol solvent under pressure. The resultant free base can exist as multiple tautomers and the structures of these different protonated forms have been elucidated by X-ray crystallography.140,141

Figure 1.2.1: The synthesis of DII and the structures of its tautomers (CCDC

228419, 228420, 914961).

Phthalocyanine (Pc) (Figure 1.2.2) was the first DII-based macrocycles to be discovered in 1907. The fully aromatic, 18π electron delocalization of these systems has allowed it to be widely utilized in optoelectronic devices,142 photosensitizers for photodynamic therapies, semiconducting materials,143 and many other applications.143,144 The most studied modifications to Pc systems include the expanded super Pcs and contracted SubPcs.145,146 The aromatic characteristics of these modified systems allow for similar uses as seen in Pcs,147 with the applications more focused in the areas of organic light emitting diodes

(OLEDS),148 energy transfer dyads,149,150 and organic photovoltaics.148

Around the same time Elvidge and Linstead discovered DII, they made rapid progress synthesizing various chelates and macrocyclic systems.137,151 one of the earliest of these discoveries were the bis(arylimino)isoindoline chelates

(Figure 1.2.2).137 These tridentate chelates can be synthesized with virtually any aryl or alkyl amine, and metal complexes of these ligands are often primarily used as catalysts for oxidation reactions.152,153 Shortly after the report of the first report of the BAI bis(pyridylamino)isoindoline (BPI), Elvidge and Linstead developed the hemiporphyrazine (Hp) systems (Figure 1.2.2). Although Hps have structural similarities to Pc, they are non-aromatic 20π electron macrocycles.154,155

Although Elvidge and Linstead pioneered the fields of BAI and Hp chemistry, many related systems have subsequently since been discovered. The reactivity of DII with aryl and alkyl amines typically result in chelates with substitutions of both imine sites. However, a category of chelates called

26 semihemiporphyrazines (Figure 1.2.2) are the result of a single substitution, leaving one unreacted imine.156,157 Additionally, the hydrazine mediated ring expansion of BAI chelates, can result in a phthalazine (Figure 1.2.2).158 This class of chelate is not as well studied as the BAI systems, but are important in the areas of catalysis.159

Figure 1.2.2: Examples of DII based macrocycles and chelates.

1.2.1 Phthalocyanine

Phthalocyanines are a class of macrocycles comprised of four isoindoline units linked via bridging sp2 hybridized nitrogen atoms. Considered to be an analog of porphyrin, phthalocyanines are 18π electron aromatic macrocycles with useful optical and electronic properties.160 The synthesis of Pc was reported as early as 1907 by Braun and Tcherniac, who observed the formation of an unidentified blue product was when refluxing o-cyanobenzamide in ethanol.161

Years later, in 1934, Linstead and Robertson correctly predicted the structure of

Pc, and noted that this compound can bind many metal ions at the center of the ring.162 Since its initial discovery, Pc has been synthesized from large variety of starting materials including DII, phthalonitrile, phthalic anhydride, and phthalimide.163

The structure of phthalocyanine is planar with two ionizable protons at the core, rendering it a useful chelate for metals. Metal Pcs are typically prepared via template reactions of the appropriate metal precursor with phthalonitrile, and complexes of Zn,164 Co,165 Cu,166,167 Ni,164,168 and Fe169 have all been reported.

Unlike the preferred octahedral geometry of metal hemiporphyrazine complexes

(vide infra), Pc prefers square planar complexes as shown in Figure 1.2.3.

Figure 1.2.3: The X-ray structure of CuPc (CCDC 1133493) from two viewpoints.

View 1: perpendicular to least squares plane and view 2: along least squares plane. Gray, light blue, and orange spheres represent carbon, nitrogen, and copper respectively. Hydrogen atoms have been omitted for clarity.

1.2.2 Expanded Phthalocyanines

Unsubstituted Pcs have been well studied over the past few decades and many peripherally subsitituted phthalocyanines have been synthesized, typically for enhanced solubility.170–172 However, the overall size of the core can also be modified, resulting in both ring expanded and contracted phthalocyanines. The use of templating agents like BCl3, and other metal salts are typically required for expanded and contracted macrocycle formation, similar to metal Pcs, although free base Pcs can be synthesized without templates.145,171,173,174 In addition, the literature does contain an example of an expanded Pc synthesized without a metal template.175

The first metal template expanded phthalocyanine was synthesized by

Marks and Stojakovic in 1978 and was called superphthalocyanine (SPcUO2)

(Figure 1.2.4).176 Superphthalocyanine is the result of a template reaction using uranyl salts and phthalonitrile, and the macrocycle contains five DII subunits in comparison to the characteristic four seen in Pc. Although SPcUO2 is a 22π electron system, it displays a lesser degree of aromaticity in comparison to 18π electron Pc. This is attributed to its deviation from planarity (Figure 1.2.4) and buckling of the strained macrocycle, resulting in disruption of the electronic delocalization.145

Expanded Pcs can also be synthesized via template reactions with other large ions such as W and Mo as reported by Kobayashi.177 Pc precursor molecules

30 like phthalonitrile, phthalic anhydride, or DII were combined with excess urea, and molybdenum or tungsten salts in 1-chloronaphthalene. The resultant macrocycles shown in Figure 1.2.4 display a ring-expanded Pc. The four DII subunits are joined by two urea moieties, and two equivalents of metal carbonyl reside in the central cavity. This Pc analog is a 22π electron system with absorbances extending out to

1200-1500 nm.177

Figure 1.2.4: The X-ray structures of SPcUO2 and the rectangle expanded Pc from reference 176 and 177 (R groups and hydrogen atoms have been omitted for clarity) (CCDC 1125690 and 830031). View 1: side-on view and view 2: top-down view. Gray, light blue, red, turquoise, and dark blue spheres represent carbon, nitrogen, oxygen, molybdenum, and uranium respectively.

1.2.3 Subphthalocyanines

The size of the templating agent reacted with phthalonitrile can directly impact the pore size of the macrocycle formed. In 1971, Meller and Ossko reacted phthalonitrile with BCl3 as the templating agent. Instead of the expected boron phthalocyanine cyclotetramerization product, a purple compound was isolated from the reaction and was discovered to be a contracted cyclotrimerization analog of Pc, shown in Figure 1.2.5.178,179 These Pc analogs dubbed subphthalocyanines

(SubPcs) are 14 π-electron systems with a bowl shaped structure due to the small nature of the central boron atom. The boron atom resides above the plane of the three isoindole nitrogen atoms as seen in the X-ray crystal structure (Figure 1.2.5).

These molecules are widely used in organic photovoltaics,148,180,181 self-assembly systems,182,183 and energy transfer dyads.149,150

Figure 1.2.5: The synthesis and X-ray structure of chloro-SubPc (CCDC 1232697).

Modified SubPcs can be obtained from various peripherally modified phthalonitrile precursors, or by exchanging the axial ligands of the boron atom.146,184–186 SubPc itself is unstable under many reaction conditions and post peripheral modification of the macrocycle must be carefully carried out to avoid decomposition.187 Standard reactions of SubPcs involve Buchwald-Hartwig aminations,188 and borylations,188,189 acylations,190 and alkylations.191 Although

SubPc is not synthesized directly from DII, it is known to react with DII to form asymmetric phthalocyanines.192,193 The synthesis of asymmetric Pcs is notably

34 challenging and the reactions between two different peripherally substituted phthalonitriles results in complex mixtures that cannot be separated by chromatography. Kobayashi and coworkers have exploited the ring strain of

SubPc by reacting a tri-tert-butylated SubPc with various DIIs as seen in Figure

1.2.6.194

Figure 1.2.6: The synthesis of asymmetric Pcs from DII (reference 194).

1.2.4 Hemiporphyrazines

After synthesizing DII, Elvidge and Linstead soon discovered DII can react with 2,6-diaminopyridine in equimolar amounts to generate a new macrocycle

(Figure 1.2.7).195 This molecule, later referred to as “hemiporphyrazine” (Hp) by

Campbell, is comprised of four subunits in the macrocycle.196 These subunits contain two isoindole units across from each other, connected by C-N-C bonds to non-isoindole aromatic rings.196 The resultant macrocycle is typically a non- aromatic 20π electron system.

Figure 1.2.7: The synthesis of hemiporphyrazine.

Bis-pyridyl hemiporphyrazines, the most common form of this macrocycle, can readily bind metals in a square planar fashion, with the ligand often saddled around the metal center (Figure 1.2.8).197 The two isoindole protons are ionizable resulting in a dianionic tetradendate ligand. Hps do not display symmetric binding at the metal center. For example, in the NiHp complex of the bis-pyridine Hp, the isoindole nitrogen-metal bonds (~1.86 Å) tend to be shorter than the pyridine-metal bonds (~2.00 Å), and the metal complexes are predisposed toward axial ligation.197 36

Metal Hp complexes of zinc and nickel readily coordinate solvents like H2O, and pyridine (Py), resulting in both square pyramidal and octahedral complexes.155,197–

199 With axial ligation, the Hp ligand becomes increasingly planar with slight

197 ruffling of the macrocycle as seen in NiHp(Py)2 (Figure 1.2.8).

Figure 1.2.8: The X-ray structures of NiHp and NiHp(Py)2 (CCDC 1220437 and

1124972). View 1: side-on view and view 2: top-down view. Gray, light blue, and green spheres represent carbon, nitrogen, and nickel respectively.

The backbone of Hp can also be altered to exhibit tunable aromaticity, as demonstrated by Muranaka and co-workers.200 Hemiporphyrazines with redox active subunits were first synthesized by Ziegler and coworkers.201 They initially 37 synthesized 20π electron hemiporphyrazine with resorcinol or phenol units (Figure

1.2.9). The X-ray structure of the 20π electron macrocycle displays a nonplanar saddled conformation (Figure 1.2.9). When the Hp is oxidized with dichloro-5,6- dicyano-1,4-benzoquinone (DDQ), the color changes from red to dark green. This new product was determined to be the 18π electron oxidized macrocycle, and its structure is shown in Figure 1.2.9. The X-ray structure is planar, and the chemical shift of the internal C-H protons moves upfield to -0.49 ppm. This indicates a diatropic ring current and is confirmation of the aromaticity of the macrocycle.

Additionally, the UV-visible spectrum of the oxidized species displayed two new bands at ~852 and 653 nm. These bands were designated as split Q band type absorptions, similarly seen in low-symmetry tetraazaporphyrin (TAP) and phthalocyanine (Pc) derivatives.202

Figure 1.2.9: The X-ray structures of the 20π electron hemiporphyrazine, and the oxidized 18π electron system from reference 200 (CCDC 848549 and 848550).

View 1: side-on view and view 2: top-down view. Gray, light blue, red, and white spheres represent carbon, nitrogen, oxygen, and hydrogen respectively.

The most common and well-studied alterations to the Hp scaffold involve the modification of the peripheral R groups of DII, and varying the identity of the aryl diamine.203 Since their first report on Hp systems, Elvidge and Linstead characterized other Hps containing aryl amines such as 3,5-diaminopyridine, 2,8-

39 diaminoacridine, 2,7-diaminonaphthalene, and m-phenylenediamine.204

Additionally, Hps can be generated with aliphatic diamines such as the Hp analog called “cyclohexylcyanine”, which uses diamino cyclohexane as reported by

Ziegler.205 After Elvidge and Linstead synthesized a Hp analog with m- phenylenediamine called dicarbahemiporphyrazine (Figure 1.2.10),204 they generated a macrocycle with three isoindole units and one unit of benzene shown in Figure 1.2.10.206 The 3-unit intermediate is an important reagent as it can react further with DII to generate a new macrocycle. This molecule is considered to be three-quarters of a phthalocyanine and the macrocycle was eventually named benziphthalocyanine by Ziegler.199,201

Figure 1.2.10: The synthesis of dicarbahemiporphyrazine and benziphthalocyanine from references 204 and 206. 40

Similar to Pcs, the solubility of benziphthalocyanine is very poor but the metal chemistry of this macrocycle has been explored with lithium, and other first row transition metals. Metalation of the macrocycle with Co and Ni undergoes C-

H activation and a metal-carbon bond formation, resulting in the 2+ oxidation state of the metal (Figure 1.2.11).207,208 Insertion of Li, and Zn results in lower coordination complexes with agostic-type interactions as opposed to direct metal- carbon bonds (Figure 1.2.11).199,209 Similar benziphthalocyanine-like systems have been reported with triazoles,210 1,8-naphthalene211,212 and other aryl groups.213–216

Figure 1.2.11: The X-ray structures of Ni and Zn benziphthalocyanine complexes from references 207 and 209 (CCDC 726129 and 750085). Hydrogen atoms have been omitted for clarity. Gray, light blue, green and blue-gray spheres represent carbon, nitrogen, nickel, and zinc respectively.

The isolation of 3-unit intermediates has been utilized to synthesize various hemiporphyrazine analogs such as ring opened systems,217 asymmetric Hps,218–

220 and expanded macrocyclic systems.221,222 Figure 1.2.12 portrays the structure of a helical system resulting from the reaction of a 3-unit intermediate consisting of DII and 2,6-diaminopyridine with two equivalents of 2-aminopyridine. When complexed with Zn, the helical enantiomers of the Zn complex were separated by chiral chromatography, and the enantiomers displayed circularly polarized luminescence.217 Figure 1.2.12 also provides an example of an expanded Hp macrocycle resulting from the dimerization of two 3-unit intermediates.221

Figure 1.2.12: The X-ray structures of systems synthesized from 3-unit intermediates from references 217 and 221 (CCDC 1953097 and 1496230). Gray and light blue spheres represent carbon and nitrogen respectively.

This dimerization of 3-unit intermediates was first noticed by the Torres group. They synthesized an expanded hemiporphyrazine derivative shown in

Figure 1.2.13.222 Many 3-unit intermediate species were synthesized with

42 substituted diamino triazoles and DII. When heated to 135 °C these compounds underwent dimerization to form expanded Pc derivatives. While these 28 π systems are not aromatic, their UV-Vis spectra show Soret-like bands at 338-402 nm. These bands are red shifted in comparison to Hp systems due to the increased conjugation of the expanded macrocycle.

Figure 1.2.13: The synthesis of an expanded Pc analog from reference 222.

The chemistry of expanded hemiporphyrazine systems is primarily rooted in macrocyclic formation from 3-unit intermediates. However, an expanded hemiporhyrazine system reported by Torres and Islyaikin contains alternating units of DII and substituted diaminothiazoles (Figure 1.2.14).223,224 Refluxing equimolar amounts of these reagents results in the cross condensation “3+3” macrocycle, as opposed to typical “2+2” hemiporphyrazines. The X-ray crystal structures of these macrocycles portrays a relatively planar macrocycle with the sulfur atom facing outward and the N-N unit is part of the macrocycle cavity. 1H NMR spectra show 43 the isoindole NH protons ~11.5-12.5 ppm indicating no diatropic ring current.

Additionally, no Soret-like transitions are observed for any of these macrocycles and the absorption spectra show transitions in the UV and visible region. Strong absorptions are seen between ~250-400 nm with molar absorptivities around 5 x

104 M -1cm-1 and weaker low energy absorptions are observed between ~500-550 nm. 223,224

Figure 1.2.14: The synthesis of an expanded hemiporphyrazine from reference

224.

1.2.5 Bis(arylimino)isoindolines

In 1952 Elvidge and Linstead reacted DII with 2-aminopyridine in boiling alcohol solvents to generate a ligand known as bis(2-pyridylimino)isoindoline (BPI).137

This condensation reaction with DII can be extended to virtually any aryl amine

44 resulting in tridentate chelates. These ligands, known as bis(arylimino)isoindolines

(BAIs), are commonly found in the literature containing substituted aryl groups such as pyridines,137 benzimidazoles,225 and thiazoles226 (Figure 1.2.15).

Figure 1.2.15: The X-ray structures of various BAI chelates (CCDC 218623,

756935, 824820). Gray, light blue, and yellow spheres represent carbon, nitrogen, and sulfur respectively.

Although BAIs were initially synthesized with DII, Walter Siegl presented an alternative method for synthesizing the ligands.227 BAIs can be generated by the reaction of phthalonitrile, with aryl or alkyl amines in the presence of alkali earth metals in alcohol solvents. Yields are often maximized when 10 mole percent

CaCl2 is used as the catalyst with n-butanol as the solvent (Figure 1.2.16). The reaction requires much longer reaction times, but the overall yields tend to be higher than those seen with Linstead-type conditions.138

Figure 1.2.16: The synthesis of BAI ligands using Siegl conditions.

BAI chelates synthesized through Siegl and Linstead conditions are often ideal for binding metals, and this chemistry has extended across the transition152,228 as well as main group elements.229 Typically, BAI ligands are monoanionic and bind to metals in a meridional coordination mode, although a facial mode has been recently observed.230 Both homoleptic 2:1,231 and heteroleptic 1:1232–234 complexes can be synthesized with examples shown in

Figure 1.2.17. The metalation chemistry of BAI’s have generated a lot of interest over the years in the fields of catalysis,235–238 bioinorganic chemistry,231–234 and photoactive materials.239

Figure 1.2.17: The structures of various BAI metal complexes. Top left: homoleptic

Co(BPI)2 complex, and heteroleptic Fe (top right), and Cu (bottom) BAI complexes.

(CCDC 256231, 939329, 1120882, 831064). Hydrogen atoms have been omitted for clarity. Gray, light blue, green, yellow, purple, orange (large) and orange (small) spheres represent carbon, nitrogen, chlorine, sulfur, cobalt, iron, and copper respectively.

1.2.6 Semihemiporphyrazines

Typically, reactions of DII or phthalonitrile with amines results in substitutions at both imine positions.138 Most BAI ligands are symmetric 2:1 products, and the reaction seldom halts at the 1:1 product. Some groups have successfully generated 1:1 arylimine DII adducts.157,240 Several years ago, the

Ziegler group observed monofunctionalized bis(alkylimino)isoindolines when both

Siegl and Linstead conditions were utilized with 1-naphthylamine, and

157 cyclopropylamine. Later in 2016, the Ziegler group used the Re(CO)3 moiety as a templating agent to generate 1:1 condensation products of DII, and a single equivalent of aryl amine.156 The carbonyl groups on the metal are not labile, and prevent the formation of 2:1 compounds. These compounds are referred to as half of a hemiporphyrazine or “semihemiporphyrazines” and examples of pyridine, benzimidazole, and thiazole are shown in Figure 1.2.18. The carbonyls are arranged in a facial coordination mode while the bidentate chelate tilts out of the coordination plane. The terminal carbon-nitrogen bonds lengths range from ~1.29-

1.31 Å, indicating the terminal carbon-nitrogen bond has amine character.156

Figure 1.2.18: The X-ray structures of semihemiporphyrazines from reference 156.

(CCDC 1527134-1527136). Hydrogen atoms have been omitted for clarity. Gray, light blue, green, yellow, red, and dark blue spheres represent carbon, nitrogen, chlorine, sulfur, oxygen, and rhenium respectively.

In addition to synthesizing half hemiporphyrazines with the Re(CO)3 moiety, template reactions with rhenium and boron can be used to generate half phthalocyanines called aza(dibenzopyrro)methenes (ADBMs). In the past

ADBMs have been synthesized by reacting Grignards with phthalonitrile.241 The

Ziegler group utilized DII as the precursor molecule to synthesize ADBMs complexes of boron and rhenium.242,243 DII was refluxed in high boiling solvents

242,243 with either Re(CO)5Cl or BPh3 as the templating agent. Two equivalents of

DII condense resulting in a monoanionic bidentate chelate (Figure 1.2.19). The lack of Re-CO lability, and increased strength of the B-phenyl bond of the

Re(CO)3Cl and B(Ph)3 templates inhibit the formation of Pc and SubPc products.

Additionally, trace water in the reaction solvent results in either bis(oxo), or bis(imine) terminated compounds, as well as the mixed-imine/oxo species.

Figure 1.2.19: The X-ray structures of the aza(dibenzo)dipyrromethene complexes of B (top) and Re (bottom) from references 242 and 243. Gray, light blue, red, pink, and dark blue spheres represent carbon, nitrogen, oxygen, boron, and rhenium respectively.

In 2010, Kleeberg, and Bröring synthesized a semihemiporphyrazine ligand that was utilized to generate an asymmetric BAI ligand containing pyridine and thiazole units, depicted in Figure 1.2.20.240 Although DII was reacted with two equivalents of 2-amino-4-tert-butylthiazole in ethanol, the reaction stopped at the 50

1:1 product. The condensation reaction between DII and amines undergoes an amine-imine equilibrium, and the imine terminal is more reactive, driving the reaction to form BAIs.244 The authors of this study predicted that the reaction can halt at the stable isolable 1:1 material if that product is in the amine form. They then reacted the 1:1 product with a second equivalent of amine, 2-aminopyridine, in ethanol, resulting in an asymmetric BAI ligand. The ligand underwent C-H bond activation to form carbon bonded Pd complexes (Figure 1.2.20).240

Figure 1.2.20: The synthesis of the 1:1 product, asymmetric BAI, and Pd(BAI)

(along with its X-ray structure) from reference 240 (CCDC 729236). Hydrogen atoms have been omitted for clarity.

As previously mentioned, phthalocyanine and its related systems can be readily synthesized from precursors other than DII. The synthesis of isoindoline- based chelates are also not limited to condensation reactions of DII with aryl and alkyl amines, or CaCl2 mediated Siegl reactions. Gade and co-workers reacted various phthalimides with phosphorous Wittig reagents as shown in Figure 1.2.21.

This class of pincer ligands are referred to as bis(oxazolinylmethylidene)isoindolines (boxmi), and contain a C=C bond instead of the C=N bond seen in BAIs.245 Phthalimides are not particularly reactive at the carbonyl and require the use of reactive Wittig reagents to generate the desired boxmi products. Since its discovery, many first row transition metal complexes of boxmi have been synthesized to use as catalysts in the hydrosilylation of ketones.246,247

Figure 1.2.21: The synthesis of boxmi ligands, and the structure of an exemplary boxmi ligand from reference 245 (CCDC 830007). Hydrogen atoms have been omitted for clarity.

1.2.7 Phthalazines

The chemistry of BAI chelates is versatile and can extend beyond peripheral modifications of the aryl amines or isoindoline. When BAIs are reacted with hydrazine, the isoindoline core is modified via expansion of the indole ring, resulting in a phthalazine as shown in Figure 1.2.22.

Figure 1.2.22: The ring expansion of BAIs using hydrazine.

This hydrazine expansion chemistry was first reported in 1969 by Blake,

Ball, and Andew.248 A 1,4-dihydrazinophthalazine (dhph) chelate was synthesized and reacted with various nickel and cobalt salts. They were able to provide direct evidence for the binding modes of the metal complexes (Figure 1.2.23), with the use of X-ray crystallography. The dhpd ligand binds to two Ni centers in a trans- planar arrangement, with four water molecules coordinated to the metal centers resulting in octahedral geometry of Ni.248 The phthalazine chemistry of the dhpd ligand was followed up by Elvidge and Lever that same year, as they synthesized 53 a phthalazine derivative of BPI, and subsequently investigated the metal chemistry of the ligand using Co, Ni, and Cu salts.249 Similarly, they observed the formation of binuclear metal complexes.

Figure 1.2.23: The synthesis of the Ni phthalazine complex from reference 248.

Binuclear transition metal complexes phthalazines often display distorted octahedral geometries around the metal centers, and close proximity of the two metal centers. The binuclear phthalazine complexes of Cu and Fe in Figure 1.2.24 have metal-metal distances of ~3.01 and 3.14 Å.250,251 Whereas these complexes are found to be useful for oxidation catalysis,252,253 they also display interesting metal-metal antiferromagnetic coupling characteristics.159,251,254,255

Figure 1.2.24: The X-ray structures of binuclear phthalazine complexes of Cu (left) and Fe (right). (CCDC 1129164 and 990912). Hydrogen atoms have been omitted for clarity. Hydrogen atoms have been omitted for clarity. Gray, light blue, green, yellow, orange (large) and orange (small) spheres represent carbon, nitrogen, chlorine, sulfur, iron, and copper respectively.

1.2.8 Conclusions on DII as a Reagent

In conclusion, DII is a versatile molecule that undergoes simple Schiff base condensation reactions often resulting in stable, isolable materials. The macrocyclic systems and chelates are easier to isolate than certain compounds like porphyrins, and the metal chemistry of these systems exhibit widely useful optical and electronic properties, and practical uses such as catalysis. While only a few modified systems such as SubPcs, benziphthalocyanines, and phthalazines were discussed, it is important to note the field of DII chemistry is potentially much

55 more expansive and the generation of new macrocyclic systems and chelates continue to be investigated.

56 CHAPTER II

EVALUATING FERROCENE IONS AND ALL-FERROCENE SALTS FOR ELECTROCHEMICAL APPLICATIONS

THE TEXT OF THIS CHAPTER IS ADAPTED FROM THE MATERIAL AS IT APPEARS IN: SCHRAGE, B. R.; ZHAO, Z.; BOIKA, A.; ZIEGLER, C. J. EVALUATING FERROCENE IONS AND ALL-FERROCENE SALTS FOR ELECTROCHEMICAL APPLICATIONS. J. ORGANOMET. CHEM. 2019, 897, 23– 31. COPYRIGHT © 2019 PUBLISHED BY ELSEVIER B.V. ELECTROCHEMICAL MEASUREMENTS WERE CARRIED OUT BY ZHILING ZHAO.

DOI: 10.1016/J.JORGANCHEM.2019.06.023

Introduction

The increasing importance of renewable sources of energy, in particular with regard to solar energy harvesting, has accelerated the development of new energy storage materials. Recently, the redox flow battery (RFB) approach to energy storage has received increased attention. First developed more than thirty years ago, RFB design has been revisited as a potentially low cost and scalable method for storing potential generated by photovoltaics.9,11,12,28,256 In particular, the development of new chemical compounds as the redox active components of

RFB devices has progressed rapidly. A number of new molecular strategies are

57 being developed and tested, ranging from organic compounds to transition metal species for use in both aqueous and organic media.6-16 The ferrocene unit (Figure

2.1) possesses several potential advantages as a molecular component of RFB devices. This species exhibits very reversible and well understood oxidation and reduction. Additionally, ferrocene can be readily functionalized on one or both rings; such modification can optimize solubility and tune redox potentials.

However, in spite of this recent interest in the use of these materials, the structural space of ferrocene derivatives remains largely unexplored for RFB applications.22,108,111,119,131,266,267 In this report, we present several ionic compounds comprised of ferrocene cations and anions, or “all-ferrocene,” salts in several solvent systems. A system comprised of multiple ferrocene units could exhibit multiple redox potentials, due to differing substitutions on the ferrocene units. Thus, it is theoretically possible to achieve a stable mixed-valent state, and to obtain a potential from the reaction between the fully oxidized and fully reduced forms of, for example, a bis-ferrocene system.

We have synthesized several all-ferrocene salts, shown in Figure 2.1, comprised of the ammonium cation and either a carboxylate or sulfate modified ferrocene. Additionally, we can alter the ferrocene alt ion stoichiometry by using either monofunctionalized or bifunctionalized ferrocenes. The ferrocene salts along with the anionic and cationic precursors were investigated by electrochemical methods. Notably, we observed a dependence on hydrogen bonding in the observed redox potentials for the ferrocene anions in this study.

This effect is particularly notable in salts incorporating the bis(sulfonate anion); we previously published a short communication on this chemistry.268 We observed that, despite the current enthusiasm for ferrocene and ferrocene derivatives, not every ferrocene derivative is equally reversible and stable. Additionally, as we expected, solvent effects play a significant role in the performance of these compounds, and solvent choice plays a key role in separating closely spaced redox processes, as well as overall reversibility and stability.

Figure 2.1: The structures of all-ferrocene salts 2.1–2.4.

Experimental

General Information

All reagents and starting materials were purchased from commercial vendors and used without further purification. Bis ferrocene sulfonic acid was prepared according to the previously published procedure.269 The mono potassium ferrocenylsulfonate was prepared by an modification of a previously published procedure.99 Deuterated solvents were purchased from Cambridge Isotope

Laboratories and used as received.

NMR spectra were recorded on a Varian 500 MHz spectrometer and chemical shifts were given in ppm relative to residual solvent resonances (1H NMR and 13C NMR spectra). High resolution mass spectrometry experiments were performed on a Bruker MicroTOF-III instrument. Infrared spectra were collected on Thermo Scientific Nicolet iS5 which was equipped with an iD5 ATR.

X-ray intensity data were measured on a Bruker CCD-based diffractometer with dual Cu/Mo ImuS microfocus optics (Cu Kα radiation, λ = 1.54178 Å, Mo Kα radiation, λ =0.71073 Å). Crystals were mounted on a cryoloop using Paratone oil and placed under a steam of nitrogen at 100 K (Oxford Cryosystems). The detector was placed at a distance of 5.00 cm from the crystal. The data were corrected for absorption with the SADABS program. The structures were refined using the

Bruker SHELXTL Software Package (Version 6.1), and were solved using direct methods until the final anisotropic full-matrix, least squares refinement of F2 converged.270 CCDC numbers 1847169 and 1890052-1890054 contain the

60 supplementary crystallographic data for this paper. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/structures

Electrochemistry measurements were conducted using a CHI 920d potentiostat in a standard three-electrode configuration. A platinum wire was used as an auxiliary electrode. The working electrode used in voltammetry experiments was a platinum disk having a diameter of either 10 µm (microelectrode) or 2 mm

(conventional electrode). The nonaqueous Ag/Ag+ reference electrode was used by immersing silver wire in degassed dimethylformamide (DMF) or propylene carbonate (PC) solution of 0.01 M AgNO3/0.2 M tetrabutylammonium hexafluorophosphate (TBAPF6). All potentials were referred to the ferrocene methanol/ferrocenium methanol couple. The concentration of analyte was 2 mM, and the supporting electrolyte was 0.2 M TBAPF6 dissolved in DMF or propylene carbonate (PC). All solutions were purged with dinitrogen prior to any electrochemical measurements.

Syntheses

Synthesis of 2.1. Ferrocene monosulfonic acid was dissolved in EtOH and neutralized with KOH. The generated mono potassium ferrocenylsulfonate was filtered, and isolated as a yellow solid. The product was washed with cold EtOH, and used without further purification. A sample of 0.25 g (0.82 mmol) of mono potassium ferrocenylsulfonate was dissolved in 2 mL of water and 10 mL of EtOH.

To the stirring solution, 0.15 g (0.86 mmol) of AgNO3 was added and the solution

61 turned green. After the solids were dissolved, 0.33 g (0.86 mmol) of

(ferrocenemethyl)trimethylammonium iodide were added to the mixture. After 1 hour of stirring at room temperature, the AgI precipitate was filtered. The filtrate was collected and the solvent was removed by rotary evaporation. The residue was precipitated with the addition of ether, and the resultant solid was filtered and washed with cold DCM. Crystals suitable for X-ray diffraction were grown by vapor diffusion of MeOH into a hexane solution.

2.1. Yield: 0.37 g (86 %). IR (S-O stretch, cm-1): 891 (w), (SO stretch, cm-

1 ): 1355 (br, m), 1177(s). Hi-res ESI MS (positive mode) calcd C24H29Fe2NO3S

1 523.0568 m/z, found 523.0553. H NMR (500 MHz, d6 - DMSO): δ = 4.49 (t, 2H, H on C5H4), 4.37 (m, 4H, H on C5H4, and CH2), 4.29 (s, 2H, H on C5H4), 4.24 (m, 5H

13 1 on Cp), 4.18 (s, 5H on Cp), 4.05 (s, 2H, H on C5H4), 2.90 (s, 9H, CH3). C{ H}

NMR (125 MHz, d6 - DMSO): δ = 97.1, 73.4, 72.1, 70.1, 69.5, 69.1, 67.2, 67.1,

65.6, 51.5 (CH3).

Synthesis of 2.2-2.4. The procedure for the synthesis of 2.3-2.4 is the same as 2.2 except 0.72 mmol of 1,1’- bis ferrocene sulfonic acid, and 0.91 mmol of 1,1’- bis ferrocene carboxylic acid were used. To generate compound 2.2, 0.25 g (1.09 mmol) of ferrocene carboxylic acid was dissolved in 10 mL of water. To this stirring solution, 0.126 g (0.54 mmol) of Ag2O, and 0.42 g (1.09 mmol) of

(ferrocenemethyl)trimethylammonium iodide were added. The reaction flask was wrapped in foil in the dark and left to stir at room temperature for 24 hours. The

AgI was filtered, and the filtrate was freeze dried. The residue was dissolved in

MeOH and precipitated with the addition of ether. The generated solid was filtered and washed with cold DCM. Crystals suitable for X-ray diffraction were grown by vapor diffusion of hexane into a MeOH solution, THF into a MeOH solution, and water evaporation for 2-4 respectively.

2.2. Yield: 0.40 g (75%). IR (CO stretch, cm-1): 1377(vs), 1571 (vs). Hi-res

1 ESI MS (positive mode) calcd C25H19Fe2NO2 487.0898 m/z, found 487.0897. H

NMR (500 MHz, d6 - DMSO): δ = 4.49 (t, 2H, H on C5H4), 4.38 (m, 4H, H on C5H4, and CH2), 4.34 (t, 2H, H on C5H4), 4.24 (s, 5H on Cp), 4.01 (s, 5H on Cp), 3.99 (s,

13 1 2H, H on C5H4), 2.91 (s, 9H, CH3). C{ H} NMR (125 MHz, d6 - DMSO): δ = 170.6

(C=O), 84.5, 73.3, 71.9, 69.9, 69.6, 68.9, 68.3, 67.6, 65.4, 51.3 (CH3).

2.3. Yield: 0.55 g (89 %). IR (SO stretch, cm-1): 1383 (m), 1175 (s). Hi-res

ESI MS (positive mode) calcd C38H48Fe3N2O6S2 861.1083 m/z, found 861.0950.

1 H NMR (300 MHz, d6 -DMSO): δ = 4.50 (s, 4H, H on C5H4), 4.39 (s, 4H, H on

C5H4), 4.38 (s, 4H, CH2), 4.31 (s, 4H, H on C5H4), 4.24 (s, 10H, Cp), 4.11 (s, 4H,

13 1 H on C5H4), 2.92 (s, 18H, CH3). C{ H} NMR (125 MHz, d6 – DMSO): δ = 50.92,

65.00, 67.50, 68.50, 69.52, 69.91, 71.49, 72.78.

2.4 Yield: 0.52 g (73%). IR (CO stretch, cm-1): 1381 (vs), 1555 (vs). ESI MS

+ (positive mode) calcd C14H21FeN 259.1018 m/z, found 259.1045 [M+H] , calcd

+ 1 C12H10FeNaO4 296.9821 m/z, found 296.9873 [M+Na] . H NMR (500 MHz, d6 -

DMSO): δ = 4.51 (s, 4H, H on C5H4), 4.43 (m, 4H, H on C5H4, and CH2), 4.37 (s,

4H, H on C5H4), 4.30 (m, 4H, H on C5H4), 4.24 (s, 10H on Cp), 3.84 (s, 4H, H on

13 1 C5H4), 2.96 (s, 18H, CH3). C{ H} NMR (125 MHz, d6 - DMSO): δ = 171.7 (C=O),

83.6, 73.3, 71.9, 70.1, 69.9, 68.9, 68.4, 65.4, 51.4 (CH3).

16 2.90

14 water

4.24 10

4.18

Normalized Intensity Normalized

4.37

DMSO 4

4.49

4.29

4.05

1.96 3.99 2.17 4.96 5.09 2.19 9.10

4.5 4.0 3.5 3.0 2.5 Chemical Shift (ppm)

1 Figure 2.2: H NMR (500 MHz) of 2.1 in d6 – DMSO.

1 Figure 2.3: H NMR (500 MHz) of 2.2 in d6 – DMSO.

4.24 2.92

8 DMSO 7

4.38

4.39 Water 6

4.50 5

Normalized Intensity Normalized 4 4.11

4.31 3

4.02 8.04 3.95 3.99

10.07 18.16

5.0 4.5 4.0 3.5 3.0 2.5 Chemical Shift (ppm)

1 Figure 2.4: H NMR (500 MHz) of 2.3 in d6 – DMSO.

20 water

Normalized Intensity Normalized

4.24 6 DMSO

4.37 4 2.96

4.30

3.84

4.51 2

4.43

4.08 4.09 4.09 4.10 3.91

10.18 18.27

5.0 4.5 4.0 3.5 3.0 2.5 Chemical Shift (ppm)

1 Figure 2.5: H NMR (500 MHz) of 2.4 in d6 – DMSO.

Table 2.1: Crystal data and structure refinement parameters of compounds 2.1- 2.4.

Compound 2.1 2.2 2.3 2.4

Emp. form C24H31Fe2NO C25H31Fe2NO C38H48Fe3N2O6 C80H96Fe6N4 4S 3 S2 O21 Form. weight 541.26 505.21 860.45 1784.70 Crystal monoclinic monoclinic monoclinic orthorhombic system Space group P21/n P21/n Cc Pba2 a/ Å 6.0319(2) 5.84690(10) 29.1460(8) 22.6902(11) b/ Å 11.8522(4) 11.4601(2) 15.1870(4) 23.0469(13) c/ Å 32.1489(11) 32.2404(5) 21.1928(6) 7.8856(4) α(°) 90 90 90 90 β(°) 91.460(2) 90.8030(10) 126.0700(10) 90 γ(°) 90 90 90 90 Volume (Å3) 862.63(10) 2159.23(6) 7582.5(4) 4123.7(4) Z 4 4 8 2 Dc (Mg/m3) 1.559 1.554 1.507 1.437 µ (mm-1) 1.385 10.995 1.291 1.100 F(000) 1128 1056 3584 1856 Reflections 21780 12352 68166 74748 collected Data/Restrai 4052 / 0 / 300 3186 / 0 / 291 16450 / 2 / 932 10247 / 1 / nts/Paramet 507 ers GOF on F2 1.035 1.013 1.086 1.044 2 R1 (on Fo , I 0.0344 0.0318 0.0592 0.0363 > 2σ(I)) 2 wR2 (on Fo , 0.0693 0.0725 0.1446 0.0939 I > 2σ(I)) R1 (all data) 0.0476 0.0402 0.0647 0.0445 wR2 (all 0.0745 0.0762 0.1495 0.0988 data)

Results and Discussion

The syntheses of the all-ferrocene salts (compounds 2.1-2.4) are carried out via straightforward ion metatheses reactions. The ammonium and carboxylic acid modified ferrocenes can be readily synthesized271–275 or purchased from chemical vendors. The mono and bis sulfonate modified ferrocenes can be produced via the reaction of chlorosulfonic acid with ferrocene, with the desired products formed by manipulation of reaction conditions.88,99,269 We have used these compounds over the past few years as precursors to synthesize a variety of sulfonamide and sulfonyl compounds.97,101,269 Scheme 2.1 shows the synthetic methods used to generate compounds 2.1-2.4. The all-ferrocene salts are produced via use of silver ferrocene salts as in situ intermediates, which upon reaction with the ferrocene trimethylammonium iodides, afford the desired all- ferrocene salts as pure crystalline materials. The mono-sulfonate salt uses silver nitrate as the source of the metathesis reagent, while the remaining three preparations employ silver oxide in either one half or one equivalent depending on the charge of the anion. These compounds were fully characterized to confirm their composition, including by single crystal X-ray diffraction for all of the compounds, the structures of which are shown in Figure 2.6. The salts exhibit the expected structural features for these ions, in good agreement with previously published structure elucidations.276–283

Scheme 2.1: The syntheses of all ferrocene salts 2.1–2.4.

Figure 2.6: The structures of compounds 2.1-2.4 with 35% thermal ellipsoids.

Hydrogen atom positions have been omitted for clarity.

One of the key properties of any redox active component of RFB is solubility; without sufficient solubility, optimal energy density cannot be achieved.

Table 2.2 lists the solubility of compounds 2.1-2.4 in several organic solvents and in water; solubilities were determined after sonication for ten minutes at 293 K and

1 atm of pressure. All four salts are reasonably soluble in both aqueous and organic media. The best solubilities are observed in aqueous solution, and in organic solvents larger solubilities are seen in the polar solvents like PC, DMF and

DMSO. One concern is the lower solubility in MeCN, which is in contrast with many ferrocene derivatives. However, with regard to aqueous solubility, the salts exhibit significantly increased solubility versus compounds such as ferrocene methanol. Additionally, hydrogen bonding interactions may also be playing a role in solvation, as we have observed a notable effect of these intermolecular forces on the oxidation potential of the anions.

Table 2.2: Solubilities of compounds 2.1-2.4 (units: mM) at 293 K and 1 atm.

We investigated the electrochemistry of solutions of the single ferrocene cations and anions, as well as salts 2.1-2.4. The electrochemical behaviors of the ammonium and carboxylate modified ferrocenes have been explored; in particular the ammonium cation has been used in a wide variety of electrochemical applications.119,122,284 In contrast, the electrochemistries of the mono and bis- sulfonate modified ferrocene have not been as extensively investigated.285,286

Table 2.3 shows the half wave potentials versus ferrocenium methanol/ ferrocene methanol redox couple for the five ions.

Table 2.3: Half-wave potentials in mV versus ferrocenium methanol/ ferrocene methanol couple using a 2 mm dia. Pt electrode. The values were determined by averaging the peak potentials, unless a clear irreversibility is observed, in which case we estimated the half-wave potential values.

Figures 2.7-2.12 show the cyclic voltammograms of the ammonium cation, mono carboxylate, bis carboxylate, mono sulfonate, and bis-sulfonate modified ferrocene ions respectively in water, DMF, and PC solutions. We note several trends in the degree of reversibility and the potentials in these solvent systems.

First, the carboxylate salts clearly exhibit solvent-dependent stability issues in organic media, as seen in the lack of reversibility. Second, all of the ferrocenes undergo significant shifts in redox potential depending on the solvent. In H2O and

PC, the half wave potentials of the ferrocene ammonium cation are 124 and 120 mV versus ferrocenium methanol/ferrocene methanol, while in DMF the potential shifts to the lower value of 78 mV. For the anions, in most cases the potential 73 shifts to more negative values going from water to PC or DMF. This effect is most clearly seen in the bis-sulfonate modified ferrocene; we initially reported this solvent dependent redox behavior in a communication.268 One exception to this trend is the monocarboxylate anion, which increases in potential in organic solvents.

Figure 2.7: Cyclic voltammograms of single ferrocene salts (2 mM) measured on

2mm dia. Pt at scan rate 60 mV/s in aqueous solution with 0.2M KCl.

Figure 2.8: Cyclic voltammograms of single ferrocene salts (2 mM) measured on

10μm dia. Pt at scan rate 20 mV/s in aqueous solution with 0.2M KCl.

Figure 2.9: Cyclic voltammograms of single ferrocene salts (2 mM) measured on

2mm dia. Pt at scan rate 60 mV/s in PC solution with 0.2M TBAPF6.

Figure 2.10: Cyclic voltammograms of single ferrocene salts (2 mM) measured on

10μm dia. Pt at scan rate 20 mV/s in PC solution with 0.2M TBAPF6.

Figure 2.11: Cyclic voltammograms of single ferrocene salts (2 mM) measured on

2mm dia. Pt at scan rate 60 mV/s in DMF solution with 0.2M TBAPF6.

Figure 2.12: Cyclic voltammograms of single ferrocene salts (2 mM) measured on 10μm dia. Pt at scan rate 20 mV/s in DMF solution with 0.2M TBAPF6.

In the salt systems, we observe some changes in the behavior from that of the individual ions themselves. The cyclic voltammograms for 2.1-2.4 are shown

80 in Figures 2.13 and 2.14 (2 mm Pt electrode and 10 μm Pt microelectrode respectively). In compounds 2.1 and 2.4 in water and in compound 2.2 in DMF and PC, we observe simultaneous oxidations in both cation and anion, and thus cannot differentiate the half potentials of each ionic component. For compound

2.1 in water and compound 2.2 in PC and DMF, this lack of separation is clearly expected, as the potentials of the ferrocenes are effectively identical when alone with an inert counterion in solution. However, for salt 2.4 there is a ~45 mV separation between the two ferrocene ions when separate with an inert counterion, and yet two separate redox processes are not observed upon combination. This is explained by the inability of the cyclic voltammetry experiment to separate the two redox processes since the difference of the standard potentials is <100 mV.287

The most striking changes observed in these salt systems are seen in the oxidation potentials of the anionic components. For the sulfonate salts 2.1 and

2.3, the oxidation potentials of the anions shift to negative potentials in the organic solvents PC and DMF. The largest shift is observed for the bis-sulfonate ferrocene anion, which moves approximately 450 mV from H2O to DMF. Initially we ascribed this shift due to the changes in intermolecular interactions with the solvent medium, such as dielectric constant changes or hydrogen bonding interactions. Measuring the redox potential in PC allowed us to help elucidate the nature of these interactions, as PC has a similar dielectric constant as H2O but lacks the ability to be a hydrogen bond donor. Notably, the redox potentials of the sulfonate anions do move but are closer to those of DMF than to those of water. Thus, we

81 hypothesize that hydrogen bonding plays a more significant role in modulating the potential of the sulfonate anions than does the polarity of the solvent. As for the carboxylate anions, the effects of solvent interactions are more difficult to elucidate, due to their lack of reversibility. The lack of reversibility of ferrocene carboxylate anions is well known.288,289 However, we do observe that in the carboxylate salts 2 and 4, the redox potential is most positive in PC, intermediate in DMF, and lowest in H2O.

Figure 2.13: Cyclic voltammograms of all-ferrocene salts 2.1 – 2.4 (2 mM) measured on 2mm dia. Pt at scan rate 60 mV/s. Top: aqueous solution with 0.2M

KCl; Middle: PC solution with 0.2M TBAPF6; Bottom: DMF solution with 0.2M

TBAPF6. Salts 1 and 2: red solid curves; salts 3 and 4: blue dotted curves.

Ferrocene methanol has relative potential of -29 mV and -24 mV to ferrocene in

PC and DMF respectively.

Figure 2.14: Cyclic voltammograms of all-ferrocene salts 2.11 – 2.4 (2 mM) measured on 10μm dia. Pt at scan rate 20 mV/s. Top: aqueous solution with 0.2M

KCl; Middle: PC solution with 0.2M TBAPF6; Bottom: DMF solution with 0.2M

TBAPF6. Salts 1 and 2: red solid curves; salts 3 and 4: blue dotted curves.

Ferrocene methanol has relative potential of -29 mV and -24 mV to ferrocene in

PC and DMF respectively.

Previously, we presented the redox properties of compound 2.3, and we observed that it is possible to tune its redox response to exhibit only one oxidation peak. Figure 2.15 shows the cyclic voltammograms of compound 2.3 at various molar fractions of water and DMF. At a molar fraction of water x = 0.74, we observe only one oxidation wave instead of the two as seen in only water or DMF. There is clearly a change in the values of the limiting currents in these experiments dependent on the solution composition. The main effect is easily attributable to changes in solution viscosity, which is dependent on the solvent composition.

Figure 2.16 shows a plot of the inverse solution viscosity and the limiting current values versus the solvent composition; the viscosity data for the water-DMF mixtures were taken from the literature.290 The inverse viscosity was used since the diffusion coefficients are inversely proportional to the solution viscosity according to the Stokes-Einstein equation. On the graph, the limiting current points line up quite well with the inverse viscosity curve, in support of our hypothesis.

Some disagreement between the two sets of values is expected due to the presence of the migration effect, since no supporting electrolyte was added to these mixtures.

1.00E-09 0.00E+00 -1.00E-09 -2.00E-09

) -3.00E-09

(

t -4.00E-09

r -5.00E-09

C -6.00E-09 -7.00E-09 -8.00E-09 -9.00E-09 0.7 0.5 0.3 0.1 -0.1 -0.3 -0.5 + Potential (V) vs. FcCH2OH /FcCH2OH

Figure 2.15: Cyclic voltammograms of bis sulfonate ammonium salt (2 mM) in

DMF-water mixtures with molar fraction of water 1 (red curve), 0.97 (blue), 0.74

(brown), 0.32 (green) and 0 (purple). No supporting electrolyte was added to the mixtures. Working electrode – 10 μm Pt disk, scan rate 20 mV/s.

1.4 9E-09

8E-09

) 1.2 1

- 7E-09 ·s

1 1 - 6E-09

0.8 5E-09 mPa

0.6 4E-09 3E-09 0.4

2E-09 limiting current (A) current limiting

1/viscosity ( 1/viscosity 0.2 1E-09

0 0 0 0.2 0.4 0.6 0.8 1 1.2 x

Figure 2.16: Limiting current (data points) and inverse solution viscosity (curve) as a function of the molar fraction of water (X) in the DMF-water mixtures. The values of the limiting current were determined from the results in Fig. 2.15. The viscosity data were taken from reference 290.

Conclusions

In conclusion, we present a study into the syntheses and electrochemical activity of a series of all-ferrocene salts. The four salts, comprised of

(ferrocenemethyl)trimethylammonium cations and either carboxylate or sulfonate ferrocene anions, can be readily synthesized via simple metathesis reactions.

Depending on the identity of the anions and cations in compounds 2.1-2.4, we observe different degrees of separation between the oxidation potentials of the ferrocene components. The potentials of the ions in salts 2.1-2.4 are highly dependent on solvent conditions, and comparisons between water, PC, and DMF reveal that hydrogen bonding in addition to other solvent interactions may be playing a key role in modulating the redox potentials on the anionic components of these salts. Clearly, the choices of solvent and ferrocene ions will have a significant role in any redox flow battery design. With regard to RFB materials, the system with the largest difference in potential between ferrocene cation and anion is compound 2.3, and this salt, depending on the solvent choice, could form a stable mixed valent state. It is notable that the ferrocene sulfonates have improved solubility versus ferrocene methanol and are reversible, and this is the first evaluation of these compounds for possible energy storage applications. We are continuing our investigations into these compounds and the effects of solvent and solvation on potential.

88 CHAPTER III

INVESTIGATIONS INTO AQUEOUS REDOX FLOW BATTERIES BASED ON FERROCENE BISULFONATE

THE TEXT OF THIS CHAPTER IS ADAPTED FROM THE MATERIAL AS IT APPEARS IN: ZHAO, Z.; ZHANG, B.; SCHRAGE, B. R.; ZIEGLER, C. J.; BOIKA, A. INVESTIGATIONS INTO AQUEOUS REDOX FLOW BATTERIES BASED ON FERROCENE BISULFONATE. ACS APPL. ENERGY MATER. 2020, 3 (10), 10270–10277. COPYRIGHT © 2020 AMERICAN CHEMICAL SOCIETY. ELECTROCHEMICAL MEASUREMENTS AND BATTERY EXPERIMENTS WERE CARRIED OUT BY ZHILING ZHAO AND BAOSEN ZHANG.

DOI: 10.1021/ACSAEM.0C02259

Introduction

During the past few decades, development of renewable energy sources has attracted much attention due to their potential to reduce the carbon footprint of electricity generation.291 However, these sources, such as solar and wind energy, are intermittent and unpredictable; thus, an energy storage technique is required.292,293 Lithium ion batteries are good candidates for energy storage because of their outstanding performance.294 Recently, an alternate technique, the

89 redox flow battery (RFB) has become commercially viable, which not only has low cost but also offers longer lifetimes and improved safety.10,292,295 The energy and power of an RFB can be decoupled in this technology by controlling the volume of electrolytes and the size of the RFB stack.296 Thus, RFB devices are advantageous for electrical energy storage, and are suitable for large scale applications. RFBs can be classified by the redox species used in the electrolyte solution or the solvent system.292,297 For the purpose of green energy storage, aqueous RFBs have gained more attention than non-aqueous RFBs. A series of molecular strategies ranging from transition metal species to organic compounds have been developed as the redox active species for aqueous RFB applications.122,298–304 Yet, many of these redox active materials exhibit drawbacks, such as expensive and low abundance materials for vanadium RFBs, self-discharging issues due to crossover of free ligand (triethanolamine) for all-soluble all-iron RFBs, and corrosive electrolytes for alkaline quinone RFBs.298,299,301

Since iron is an earth-abundant element, attention has turned to the redox active organometallic compound, ferrocene. Because this compound exhibits very reversible oxidation and reduction and its chemistry is well understood, researchers have showed interest in applying ferrocene derivatives to RFB development.23,107,122,305 Due to the limited solubility of unmodified ferrocene in water, RFBs using ferrocene initially focused on non-aqueous solvent systems.23,107 To meet the requirements of green chemistry, the development of aqueous ferrocene-based RFBs with high working voltage has been initiated and

received some success by utilizing the hydrophilic ammonium functional group.122,305 On account of the positively charged ferrocene units, viologen derivatives have been typically paired as an anolyte.23,122,305 However, a well- known viologen derivative, N,N’-dimethyl-4,4’-bipyridinium dichloride (paraquat), is known to be toxic and may lead to significant health problems after long-term exposure.306,307 RFBs are ideally operated at high concentrations to reach high capacity, thus, large amounts of viologen derivatives present in RFB tanks would present a potential hazard.

To address these problems, we have focused on the development of anionic ferrocenes as highly soluble and reversible redox active materials. In this report, we present an aqueous RFB which consists of sulfate modified ferrocene as the catholyte, and anthraquinone-2,7-disulfonic acid disodium salt (2,7-AQDS) as the anolyte. Both of these compounds are highly soluble in water due to the presence of the sulfonate functional groups. 1,1’-bis(sulfonate)ferrocene dianion disodium salt (1,1’-FcDS) was found to be oxidized at a more positive potential than (ferrocenylmethyl)trimethylammonium cation because of the higher electron withdrawing strength of the sulfonate functional group.308 Thus, RFB using 1,1’-

FcDS as the catholyte is expected to generate a larger working voltage. Moreover, it opens up the possibility of adopting neutral or negatively charged redox species as the anolyte such as, for example, anthraquinones. Compared to highly toxic paraquat (acute oral LD50 75 mg/kg in rats) used as the anolyte in aqueous ferrocene based RFB, anthraquinones are much safer, and 2,7-AQDS has an

309,310 acute oral LD50 (in rats) up to 4000 mg/kg. Anthraquinones are widely used in food, pharmaceutical, and paper industries, and their application to RFBs has been explored extensively.57,300,301 To avoid the formation of unstable free radicals during RFB operation, a slightly acidic low-cost acetate buffer and 0.5 M H2SO4 were tested respectively as supporting electrolyte in this work.311 In addition, ethylene glycol (EG) was added to improve the solubility of both 1,1’-FcDS and

2,7-AQDS in aqueous solutions, which was found to increase the solubility of 2,7-

AQDS.57 We observed that the properties and stability of the RFBs were highly dependent on the identity of the supporting electrolyte.

Figure 3.1: Schematic diagram of an “all-aqueous” RFB.

Experimental

General Information

The catholyte, 1,1’-FcDS was prepared from bis ferrocene sulfonic acid, which can be synthesized based on a previously reported procedure.312 Basic iron(III) acetate was generated by a previously published procedure.313 The anolyte, 2,7-AQDS, was purchased from Pfaltz & Bauer Inc. All other chemicals were of ACS grade and purchased from Sigma-Aldrich or Fisher Scientific and used directly. Nanopure water was obtained from a Milli-Q Integral 5 water purification system (Millipore, Bedford, MA).

X-ray Crystallographic Data Collection and Structure Solution and Refinement

Single crystal data for all structures were collected on a Bruker CCD-based diffractometer with dual Cu/Mo ImuS microfocus optics (Cu Kα radiation, λ

=1.54178 Å or Mo Kα radiation, λ = 0.71073). Crystals were mounted on a cryoloop using Paratone oil and placed under a steam of nitrogen at 100 K (Oxford

Cryosystems). The data were corrected for absorption with the SADABS program.270 The structures were refined using Bruker SHELXTL Software Package

(Version 6.1) and were solved using direct methods until the final anisotropic full- matrix least squares refinement of F2 converged. Electronic Supplementary

Information (ESI) available: CCDC 2009128-2009129 contains the supplementary crystallographic data for this paper (Table 3.1). These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.

Cyclic voltammetry electrochemical characterization

Electrochemical measurements were conducted using a CHI 920d potentiostat in a standard three-electrode configuration. A platinum wire was used as an auxiliary electrode. The working electrode used in voltammetry experiments was a glassy carbon disk with a diameter of 3 mm. A Ag|AgCl|KCl (2 M) reference electrode was used. The supporting electrolyte was 1 M NaNO3, 2 M acetate buffer

(molar ratio between sodium acetate and acetic acid is 1:1, pH 4.53) or 0.5 M

H2SO4. All solutions were purged with nitrogen or argon prior to any electrochemical measurements.

Cyclic voltammograms were collected at scan rates varying from 20 mV/s to 65 mV/s for the measurement of the diffusion coefficients (D) of 1,1’-FcDS and

2,7-AQDS using Randles-Sevcik equation. The cyclic voltammograms were compared (at a scan rate of 65 mV/s) for the redox potentials of 1,1’-FcDS and

2,7-AQDS to evaluate the open circuit potential (OCP) under acidic conditions and

1 M NaNO3.

Flow cell tests

The flow cell (Figure 3.1) was constructed using a PEM fuel cell hardware which accepts liquid at both electrodes, two graphite felt electrodes (G100,

AvCarb, USA) and a piece of cation exchange membrane (Nafion 117). The surface area of graphite felt was 9 cm2. Nafion membrane and graphite felt electrodes were immersed in the supporting electrolyte (1 M NaNO3, 2 M acetate buffer or 0.5 M H2SO4) overnight before the charge-discharge tests. Each

94 electrode chamber was connected with an electrolyte reservoir using Versilon

2001 (Saint-Gobain) tubing. Each reservoir contained 30 mL of the 0.03 M catholyte/anolyte and the supporting electrolyte. A Masterflex L/S peristaltic pump

(Cole-Parmer, USA) was used to circulate the electrolytes through the flow chamber at a rate of 40 mL/min. Both reservoirs were purged with argon to remove oxygen and then sealed in a glove box filled with argon before cell cycling.

Constant-current charge-discharge cycling tests were carried out at 293 K with an

Arbin LBT21084 Battery Cycler in a two-electrode configuration. The full cell was charged/discharged at current 25 mA (2.8 mA/cm2) for 100 cycles.

Solubility test

The solubility of 1,1’-FcDS in 1 M NaNO3 with and without the addition of

EG was measured using UV-visible spectra. Saturated solutions were prepared for both conditions, and supernatant was diluted 100 times (without EG) and 200 times (with EG) for the UV-vis absorbance measurements. The correlation of the peak absorbance (Figure 3.3, C and D) and concentration was calculated using a calibrated relationship of standard samples (Figure 3.3, A and B).

Permeability test

The permeability of 1,1’-FcDS across Nafion 117 membrane was evaluated using the flow cell setup. For the solutions in electrolyte reservoirs, one (donating) side was filled with 30 mL of 0.03 M 1,1’-FcDS and 1 M NaNO3, another (receiving) side was filled with 30 mL of 1 M NaNO3. Peristaltic pump was used to circulate the electrolytes at a rate of 40 mL/min for 7 days, during which samples were taken

95 from the receiving side after 24, 120, 144 and 182 hours. Cyclic voltammograms were collected for the samples so that the concentration of 1,1’-FcDS in each sample could be calculated using the peak current values (by comparing to the

CVs of 1,1’-FcDS with known concentration). The permeability was calculated using the method reported by Kwabi et al.301

Synthesis

Synthesis of 1,1’-FcDS: Ferrocene bis sulfonic acid (5.0 g, 14.5 mmol) was dissolved in 200 mL of ethanol and a solution of concentrated NaOH was slowly added to the mixture and stirred for 30 minutes. The resulting yellow precipitate was filtered, and air-dried to yield a yellow solid. Crystals suitable for

X-ray diffraction were obtained by slow evaporation of a DMF/water mix.

−1 1 1,1’-FcDS: Yield: 5.4 g (96%). IR: 1366, 1163 cm (νSO). H NMR (300

13 1 MHz, DMSO-d6): 4.12 (s, 4H on C5H4), 4.33 (s, 4H on C5H4). C{ H} NMR (125

MHz, d6-DMSO): δ = 96.9, 69.7, 68.5. HRMS (ESI-TOF, negative mode) m/z:

- calcd for C10H8FeNaO6S2 366.9015, found 366.9060 [M-Na] .

water

3.36

Normalized Intensity Normalized 8

DMSO 6

4.12

4.33 4

2.50 2

3.98 4.00

9.5 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 Chemical Shift (ppm)

Figure S1. 1H NMR (300 MHz) of 1,1’-FcDS in d6-DMSO. Figure 3.2: 1H NMR (300 MHz) of 1,1’-FcDS in d6-DMSO.

Table 3.1: X-ray crystal data and structure parameters for compounds 2,7-AQDS and 1,1’-FcDS.

Compound 2,7-AQDS 1,1’-FcDS CCDC 2009129 2009128 Empirical formula C96H98N4Na12O69S12 C10H20FeNa2O12S2 Formula weight 3072.38 498.21 Crystal system Monoclinic Triclinic Space group C2/m P-1 a/ Å 20.109(2) 6.2579(4) b/ Å 20.778(2) 6.7586(4) c/ Å 15.8044(14) 23.6781(15) α(°) 90 93.877(3) β(°) 112.816(5) 96.594(3) γ(°) 90 109.331(3) Volume (Å3) 6087.0(11) 932.77(10) Z 2 2 Dc (Mg/m3) 1.676 1.774 µ (mm-1) 0.370 1.137 F(000) 3156 512 reflns collected 102040 38129 indep. reflns 5544 4630 GOF on F2 1.546 1.071 2 R1 (on Fo , I > 0.1796 0.0400 2σ(I)) 2 wR2 (on Fo , I > 0.4190 0.1284 2σ(I)) R1 (all data) 0.2183 0.0472 wR2 (all data) 0.4383 0.1315

Figure 3.3: Solubility tests using UV–visible spectrum. A and B: Calibration curves for the relationship between UV–visible absorbance peak and concentration of

1,1’-FcDS in 1 M 1 M NaNO3 with (A) and without (B) the addition of 0.5 M EG. C and D: UV–visible spectra of diluted supernatant of 1,1’-FcDS saturated solutions prepared in in 1 M 1 M NaNO3 with (C: dilute 100 times) and without (D: dilute 200 times) the addition of 0.5 M EG.

Results and discussion

Ferrocene derivatives have been widely utilized in materials chemistry.70

Most of the ferrocene compounds have good solubilities in organic solvents but not in water. Therefore, previously they have often been applied in nonaqueous

RFB systems.23,107 Over the past several years, aqueous ferrocene-based RFB systems have been explored using ferrocene ammonium compounds, producing all-cationic RFBs employing anion exchange resins.122 We have explored the synthesis of new ferrocene-based electroactive materials, and have recently presented work on all-ferrocene salts where large potential differences are observed between anolyte and catholyte.312 Unfortunately, those all-ferrocene salts are not practical for RFBs that use ion exchange resins, as either the ferrocene anion or ferrocene cation bind to the oppositely charged ion-exchange resin, thus impeding function.

For the work presented here, we employed a water-soluble ferrocene

– compound functionalized with SO3 groups, 1,1’-FcDS (Figure 3.4). This compound can be readily prepared from commercially available reagents and exhibits high solubility in water.312 Along with this species, we also employed the bis sulfonate modified anthraquinone 2,7-AQDS, which also exhibits good water solubility and is commercially available. Thus, this is an all-anionic approach to the RFB problem, using sulfonates as stable functional groups to drive optimal aqueous solubility. Sulfonate modification is frequently used in drug design to

100 impart water solubility, and this functional group can also be readily applied to the

RFB problem.

Figure 3.4: Anolyte (2,7-AQDS): anthraquinone-2,7-disulfonic acid disodium salt

(left). Catholyte (1,1’-FcDS):1,1’-bis(sulfonate)ferrocene disodium salt (right). The corresponding crystal structures are shown at the bottom of the diagram with 35% thermal ellipsoids. Hydrogen atoms have been omitted for clarity.

The synthesis of 1,1’-FcDS is accomplished via the sulfonation of ferrocene, followed by neutralization of the bis-sulfonic acid derivative to afford the sodium salt. We were able to structurally characterize 1,1’-FcDS, along with the

2,7-AQDS salt, using single crystal X-ray methods. The structures are shown in

Figure 3.4 below their chemical diagrams. As can be seen in the structures, each

101 compound exhibits two peripheral sulfonate groups. The presence of these anionic functional groups is key to transforming the hydrophobic precursors into highly water-soluble compounds.

We determined the solubility of 1,1’-FcDS in various solvents using UV-vis measurements and the results are shown in Table 3.2 as well as Figure 3.3. In 1

M NaNO3, the solubility is 0.3 M and theoretical capacity is 8 Ah/L (assuming 0.3

M 1,1’-FcDS as a catholyte; theoretical capacity = solubility (M) × 96,485 (C/mol)

/ 3,600 (C/Ah)). Previously, it was reported that addition of EG enhanced the solubility of 2,7-AQDS due to the interactions between the polar and nonpolar groups of EG and 2,7-AQDS.57 Therefore, we expected that EG could also increase the solubility of 1,1’-FcDS, since it has both polar (sulfonyl) group and nonpolar (phenyl) group. We found that by adding 0.5 M EG in 1 M NaNO3 solution, the solubility of 1,1’-FcDS increases to 0.6 M. In addition, the theoretical RFB capacity reaches 16.1 Ah/L.

102

Table 3.2: Solubility and capacity data of 1,1’-FcDS and 2,7-AQDS.

a from reference 57

Electrochemical Properties

Cyclic voltammograms were collected to characterize the electrochemical properties of 1,1’-FcDS and 2,7-AQDS. When the compounds were dissolved in 1

M NaNO3, 1,1’-FcDS showed a half-wave potential, E1/2, equal to 0.65 V, and 2,7-

AQDS showed an E1/2 = –0.45 V (Figure 3.5). Therefore, based on these results, an open-circuit potential (OCP) for the RFB composed of 1,1’-FcDS as a catholyte and 2,7-AQDS as an anolyte is expected to be ca. 1.1 V.

103

Figure 3.5: CVs of 2 mM 2,7-AQDS (A) or 2 mM 1,1’-FcDS (B) in an aqueous solution containing 1 M NaNO3 as a supporting electrolyte. Working electrode: 3 mm dia. glassy carbon, reference electrode: Ag|AgCl|KCl (2 M), counter electrode: platinum wire. From inner curve to outer one, the scan rate varies from 20 mV/s to

65 mV/s.

We further analyzed the CV data to determine the diffusion coefficients of the redox species. Using the Randles-Sevcik equation, the diffusion coefficient of

1,1’-FcDS was calculated to be 1.46×10-6 cm2/s (without EG). With EG added, the diffusion coefficient of 1,1’-FcDS decreased to 1.29×10-6 cm2/s (based on the CVs in Figure 3.6), presumably because of the higher viscosity for EG (0.0161 Pa·s for pure EG) vs water (0.0009 Pa·s). The obtained diffusion coefficient in water was comparable with the value published for (ferrocenylmethyl)trimethylammonium in an aqueous RFB.122

104

1.2E-05

7.0E-06

2.0E-06

20 mV/s -3.0E-06 25 mV/s 30 mV/s

Crrent (A) Crrent 35 mV/s -8.0E-06 40 mV/s 45 mV/s 50 mV/s -1.3E-05 55 mV/s 60 mV/s 65 mV/s -1.8E-05 0.9 0.8 0.7 0.6 0.5 0.4 Potential (V) vs. Ag|AgCl

Figure 3.6: CVs of 2 mM 1,1’-FcDS in aqueous solution with 0.5 M EG. 1 M NaNO3 was added as supporting electrolyte. Working electrode: 3 mm dia. glassy carbon, reference electrode: Ag|AgCl|KCl (2 M), counter electrode: platinum wire. From inner curve to outer one, the scan rate varies from 20 mV/s to 65 mV/s.

The electrochemical data for 1,1’-FcDS are summarized in Table 3.3. The electrochemical behavior of 2,7-AQDS was also evaluated here under the same conditions as for 1,1’-FcDS (Table 3.3). Apart from the neutral pH supporting electrolyte, 1 M NaNO3, CVs were also collected in acidic supporting electrolyte solutions: 2 M acetate buffer (pH 4.53) and 0.5 M H2SO4 (Figures 3.7 and 3.8).

105

Figure 3.7: CVs of 2 mM 2,7-AQDS (A) or 2 mM 1,1’-FcDS (B) in aqueous solution.

2 M acetate buffer (pH:4.53) was added as supporting electrolyte. Working electrode: 3 mm dia. glassy carbon, reference electrode: Ag|AgCl|KCl (2 M), counter electrode: platinum wire. From inner curve to outer one, the scan rate varies from 20 mV/s to 65 mV/s.

106

Figure 3.8: CVs of 3 mM 2,7-AQDS (A) or 3 mM 1,1’-FcDS (B) in aqueous solution.

0.5 M H2SO4 was added as supporting electrolyte. Working electrode: 3 mm dia. glassy carbon, reference electrode: Ag|AgCl|KCl (2 M), counter electrode: platinum wire. From inner curve to outer one, the scan rate varies from 20 mV/s to

65 mV/s.

107

It is known that anthraquinone derivatives (AQ) can form a radical anion during electrochemical reduction in neutral and alkaline electrolytes (equation

3.1):311

AQ + AQ2− ↔ 2AQ.− (3.1)

To avoid the degradation of 2,7-AQDS during reduction, acidic supporting electrolytes were adapted and compared to the behavior in 1 M NaNO3. The E1/2 of 2,7-AQDS shifts from -0.458 V in 1 M NaNO3 to -0.234 V in 2M acetate buffer.

This trend is consistent with the finding that AQ’s reduction potential is pH

311 dependent. According to the E1/2 values, a 0.9-1.1 V OCP is expected for the

1,1’-FcDS/2,7-AQDS RFB at various pH conditions.

Table 3.3: Estimated electrochemical data for 1,1’-FcDS and 2,7-AQDS.

108

Flow cell tests under different pH conditions

The 1,1’-FcDS/2,7-AQDS RFB performance at neutral pH, 1 M NaNO3 was evaluated first. The 100-time cycling was performed at 25 mA (Figure 3.10), and the initial ten charge/discharge cycles are shown in Figure 3.9 A. The trends of the coulombic efficiency, voltage efficiency and energy efficiency of the 1,1’-FcDS/2,7-

AQDS RFB are shown in the Figure 3.9 C. The energy efficiency stays at 60% and the coulombic efficiency is stable above 99%. However, a large loss in capacity

(from 0.008 Ah to 0.002 Ah) is observed (Figure 3.9 B). Presumably, the degradation of the capacity results from 2,7-AQDS, which partially decomposes under the basic/neutral conditions.311

109

Figure 3.9: 1,1’-FcDS/2,7-AQDS RFB using 1 M NaNO3 as the supporting electrolyte (0.5 M EG added). A: Ten charge and discharge cycles (#2 to #11 cycles) at constant current 25 mA (2.8 mA/cm2); B: capacity vs cycle number; C:

Coulombic efficiency (CE), energy efficiency (EE) and voltage efficiency (VE) vs cycle number.

110

Figure 3.10: 100 charge and discharge cycles at constant current 25 mA for 1,1’-

FcDS/2,7-AQDS RFB using 1 M NaNO3 as supporting electrolyte (0.5 M EG added).

111

To eliminate the formation of AQ.− radical, an acidic supporting electrolyte using acetate buffer was tested for the flow cell operation. Acetate buffer can also maintain the pH 4.53 of electrolyte during charge/discharge processes. A concentration of 2 M was chosen to match the ionic strength of the buffer to 1 M

NaNO3. The initial ten charge/discharge cycles are shown in Figure 3.11 A, while

100 cycles are included in Figure 3.12.

Figure 3.11: 1,1’-FcDS/2,7-AQDS RFB using 2 M acetate buffer as supporting electrolyte (0.5 M EG added). A: Ten charge and discharge cycles (#2 to #11 cycles) at constant current 25 mA; B: capacity vs cycling number; C: CE, EE and

VE vs cycling number.

112

Figure 3.12: 100 charge and discharge cycles at constant current 25 mA for 1,1’-

FcDS/2,7-AQDS RFB using 2 M acetate buffer as supporting electrolyte (0.5 M EG added).

From the charge/discharge data, the cell voltage decreases to ~0.5 V. This working voltage is much lower than expected; an OCP of ~0.9 V is expected if calculated using E1/2 in Table 3.3. A possible reason is the formation of a new species during the charge/discharge process, due to decomposition of 1,1’-FcDS; this is discussed at the end of the Results and Discussions section. The first supporting evidence is the decrease of the diffusion coefficient (D) for 1,1’-FcDS in the acetate buffer (Table 3.3). There is no report in the literature showing an increase in viscosity of acetate buffer solutions, therefore, the viscosity change from 1 M NaNO3 to 2 M acetate buffer should be negligible. Thus, the change of D

113 is probably an indication for the irreversible electrochemical process involving 1,1’-

FcDS and the formation of a new species, iron(III) acetate (discussed below). It has been reported that electrochemical oxidation product Fc+ can undergo nucleophilic attack and destroy the Fe redox centers.314 With those unusual behaviors observed in the acetate buffer, EE decreased by 10 % compared to the

RFB using 1 M NaNO3 (Figure 3.11 C). In terms of the capacity, the decay under the basic/neutral condition was not observed; it is stabilized around 0.0075 Ah

(Figure 3.11 B). It seems the acidic environment created by the acetate buffer improves the capacity by stabilizing the 2,7-AQDS. However, the potential interaction of acetate ions with 1,1’-FcDS may lead to the additional problems such as the mentioned formation of iron(III) acetate.

Sulfuric acid as a supporting electrolyte was utilized in the flow-cell test as well. The initial ten charge/discharge cycles are shown in Figure 3.13 A. CE, VE,

EE versus cycle number are shown in Figure 3.13 C. The coulombic efficiency is stable above 99%. This indicates that the crossover rate of 1,1’-FcDS and 2,7-

AQDS is very low. For the negatively charged 1,1’-FcDS and 2,7-AQDS, when redox reactions take place, it is difficult to penetrate a Nafion membrane, which is a cation-exchange membrane. This leads to low crossover rate and high CE.

However, the cell voltage of the RFB is sacrificed, because low pH causes reduction potential of 2,7-AQDS shift to a higher value (Table 3.3). Yet, EE stays at 80%, which is a high EE value for a ferrocene-based aqueous RFB. The decay of capacity (Figure 3.13 B) is observed during the first 20 cycles and capacity stays

114 around 0.008 Ah thereafter. This capacity fading situation is improved compared to the flow cell under basic/neutral conditions (Figure 3.9 B). Thus, acidic environment stabilizes the active species during the charge/discharge electrochemical reactions and avoids the severe decrease of the RFB capacity.

Figure 3.13: 1,1’-FcDS/2,7-AQDS RFB using 0.5 M H2SO4 as supporting electrolyte. A: Ten charge and discharge cycles (#2 to #11 cycles) at constant current 25 mA (2.8 mA/cm2); B: capacity vs cycling number; C: CE, EE and VE vs cycling number.

115

Decomposition of electrolyte during cell cycling

In order to determine what chemical process is taking place in the RFB cell in the cycling experiments, we first characterized the 1,1’-FcDS solution in acetate buffer after a typical series of 100 charge/discharge cycles. We isolated a red crystalline material which we determined was basic iron(III) acetate. This compound has a highly stable trimeric cluster structure comprised of six peripheral bridging acetate units binding the triiron core and a central μ-oxo atom.315 The cluster is monocationic and is quite stable in aqueous solution. Additionally, we investigated the UV-visible spectra of the solutions of 1,1’-FcDS in acetate buffer and in H2SO4 upon completion of the cell cycling process. The spectra revealed mixtures of 1,1’-FcDS and either iron acetate or Fe(III) ions for the acetate buffer and H2SO4 conditions respectively (see Figures 3.14 and 3.15).

Figure 3.14: UV-visible spectra for 1,1’-FcDS, basic iron(III) acetate, and the 1,1’-

FcDS (2 M acetate buffer) decomposition solution in water. 116

Figure 3.15: UV-visible spectra for 1,1’-FcDS, FeCl3 (1M H2SO4), and the 1,1’-

FcDS (0.5 M H2SO4) decomposition solution in water.

We also recorded a series of CVs for the catholyte and anolyte species after the 100 charge/discharge cycles had been completed; the results are shown in

Figure 3.16. While the data in neutral 1 M NaNO3 (Figure 3.16 A) did not show the presence of a new redox species, under acidic conditions the CVs indicated the transformation of the catholyte as indicated by the appearance of additional peaks at more negative potentials (Figure 3.16 B and C). Since in the acetate buffer electrolyte the formation of iron(III) acetate had been confirmed by X-ray crystallography, we decided to record a CV for this compound and compare it to the CV for the catholyte solution after the cycling. The comparison of peak

117 potentials confirmed a good agreement as shown in Figure 3.16 B. This fact further confirms our initial argument regarding the decomposition of 1,1’-FcDS due to the nucleophilic attack by the acetate-ions. Interestingly, we were also able to confirm the nature of the additional peaks for the catholyte in Figure 3.16 C (0.5M H2SO4).

Based on the results in Figure 3.16 B, we suspected that the nucleophilic attack by the sulfate-ions could be responsible for the decomposition of the catholyte; a recorded CV for iron(III) sulfate confirmed this (dashed line in Figure 3.16 C).

Solution crossover rate measurements

To investigate the performance of used Nafion117 membrane and the reason for large capacity fade for our RFB systems, we surveyed the membrane by characterizing the permeability of 1,1’-FcDS (see experimental section for the details of the procedure). The membrane shows excellent performance with low permeability of 1.67×10-9 cm2/s, which is lower than the permeability of the

Nafion117 membrane for vanadium ion. These results confirm that the crossover of 1,1’-FcDS is not the main reason for large capacity loss.

118

Figure 3.16: CVs of catholyte (1,1’-FcDS) and anolyte (2,7-AQDS) after 100 cycles of charge/discharge. A: using 1 M NaNO3 as supporting electrolyte, scan rate: 30 mV/s; B: using 2 M acetate buffer as supporting electrolyte, dashed curve: CVs of iron(III) acetate in 2 M acetate buffer, scan rate: 30 mV/s; C: using 0.5 M H2SO4 as supporting electrolyte, dashed curve: CVs of iron(III) sulfate in 0.5 M H2SO4, scan rate: 25 mV/s. In all experiments, working electrode: 3 mm dia. glassy carbon, reference electrode: Ag|AgCl|KCl (2 M), counter electrode: platinum wire.

119

Conclusions

Ferrocene-based catholyte materials are gaining increased popularity, however, little is done to understand the fundamentals of their operation under various conditions. This paper investigates the behavior of one of such materials,

1,1’-FcDS, for potential RFB applications under neutral and acidic conditions. The derivative of anthraquinone, 2,7-AQDS served as the anolyte in all cases. Both compounds are highly soluble due to their ionic nature, leading to the potential fabrication of a device with high energy density. Additionally, these materials are aqueous soluble, which from a green chemistry perspective is more preferable than organic media.

The choice of all-anionic catholyte and anolyte was deemed advantageous in that it would eliminate adsorption of the redox species onto cation-exchange

Nafion membrane. Basic electrochemical characterization (cyclic voltammetry) was performed in three environments: neutral (1 M NaNO3) and acidic (2 M acetate buffer pH 4.53 and 0.5 M H2SO4) conditions, as well as with addition of EG, and the solubilities of the catholyte and anolyte were measured.

By recording charge-discharge curves in neutral solutions containing EG, we observed substantial capacity fading, which we attributed to irreversible electrochemical behavior on the anolyte side (formation of a radical anion). Trying to remedy this issue, we performed the flow experiments in 2 M acetate buffer solution (pH 4.53). To our surprise, we still observed the decrease in capacity, but now the reason was in the catholyte. Even though ferrocene derivatives are widely

120 regarded as highly reversible electrochemical systems, they are prone to chemical transformations in RFB applications; in this case, to a nucleophilic attack by acetate and sulfate ions, while stable in the presence of nitrate. These findings are fundamental to the development of RFB materials, and our future work will address the uncovered instability of ferrocene derivatives, as well as finding alternative anolyte candidates.

121 CHAPTER IV

BINDING A MERIDIONAL LIGAND IN A FACIAL GEOMETRY: A SQUARE PEG IN A ROUND HOLE

THE TEXT OF THIS CHAPTER IS A REPRINT OF THE MATERIAL AS IT APPEARS IN: SCHRAGE, B. R.; VITALE, D.; KELLY, K. A.; NEMYKIN, V. N.; HERRICK, R. S.; ZIEGLER, C. J. BINDING A MERIDIONAL LIGAND IN A FACIAL GEOMETRY: A SQUARE PEG IN A ROUND HOLE. J. ORGANOMET. CHEM. 2020, 919, 121331. COPYRIGHT © 2020 ELSEVIER B.V. ALL RIGHTS RESERVED.

DOI: 10.1016/J.JORGANCHEM.2020.121331

Introduction

Bis(pyridylimino)isoindoline (BPI) (Figure 4.1), first synthesized by Linstead in 1952,137 has been investigated for its metal binding for more than half of a century, This tridentate ligand, comprised of a central isoindoline ring with two peripheral pyridine units, exhibits rich metal coordination chemistry and can be readily modified on all three arene rings. Additionally, the pyridine rings can be replaced with alternative heterocycles, and asymmetric chelates can be readily generated. 240,244 The BPI ligand and its analogous chelates are synthesized either through the method developed by Linstead via primary amines and 1,3- 122 diiminoisoindoline, or through a CaCl2 mediated process from phthalonitrile reported by Siegl. 137,227 After the initial report of the synthesis of BPI by Linstead, the metal chemistry of this ligand system has been explored across the periodic table and has gained attention as a chelate for transition metal catalyzed oxidative organic transformations, including the hydrosilylation of ketones and the oxidation of alcohols and sulfides.235–238,316–318 Since then, metal BAI systems have attracted attention in the field of bioinorganic chemistry for their use as model enzyme systems. Active site models of catechol oxidase,319,320 catalase,321 superoxide dismutases (SODs),322 and other complexes have been pursued to mimic specific enzyme reactions.233,323 The bis(arylimino)isoindoline (BAI) ligands adopt planar conformations, and as a result bind to metal ions via a meridional coordination mode.

In this report, we present an unusual binding mode for this ligand class, driven by the geometrical preference of a transition metal ion. When the BPI chelate binds to the Re(CO)3 unit, which exclusively adopts a facial configuration with nitrogenous donors, the BPI unit deforms and binds in a tridentate facial configuration (Figure 4.1). We have structurally characterized the resultant

Re(CO)3(BPI) complex, and have explored its electronic structure through DFT and

TDDFT methods. Additionally, we extended this chemistry to the structurally related bis(4-methyl-2-pyridylimino)isoindoline, and bis(2-benzimidazolylimino) isoindoline.

123

Figure 4.1: Synthesis of Re(CO)3(BAI) complexes 4.1-4.3.

124

Experimental

General Information

All reagents and starting materials were purchased from commercial vendors and used without further purification, and all reactions were performed aerobically. The bis(arylimino)isoindoline ligands were synthesized according to previously published procedures.152,324,325 Deuterated solvents were purchased from Cambridge Isotope Laboratories and used as received.

NMR spectra were recorded on 300 MHz and 500 MHz spectrometers and chemical shifts were given in ppm relative to residual solvent resonances (1H NMR and 13C NMR spectra). High-resolution mass spectrometry experiments were performed on a Bruker MicroTOF-III and MicroTOF-qIII instruments. Infrared spectra were collected on Thermo Scientific Nicolet iS5 that was equipped with an iD5 ATR. UV-visible spectra were recorded on a Cary 100 Bio UV-visible spectrometer.

Bruker SHELXTL Software Package (Version 6.1), 270 and were solved using direct

125 methods until the final anisotropic full-matrix, least squares refinement of F2 converged.

Computational Details

The starting geometries of compounds 4.1 and 4.3 were optimized using a

B3LYP exchange–correlation functional.326 Energy minima in optimized geometry were confirmed by the frequency calculations (absence of the imaginary frequencies). The solvent effect was modeled using the polarized continuum model

327 (PCM). In all calculations, CHCl3 was used as the solvent. In PCM-TDDFT calculation, the first 50 states were calculated. All atoms were modeled using the

SDD 328,329 basis set. Gaussian 09 software was used in all calculations. 330 The

QMForge program was used for molecular orbital analysis in all cases. 331

Syntheses

Synthesis of 4.1-4.3. The procedure for generating 4.1 is the same as 4.2 and 4.3 except bis(4-methyl-2-pyridylimino)isoindoline (0.045 g, 0.14 mmol), and bis(2-benzimidazolylimino)isoindoline (0.052 g, 0.14 mmol) were used in 4.2 and

4.3. A solution of Re(CO)5Cl (0.050 g, 0.14 mmol), and bis(2- pyridylimino)isoindoline (0.041 g, 0.14 mmol) in toluene (4.00 mL) were heated to reflux for 4 hours. The solution was cooled to room temperature, and the resultant precipitate was filtered and washed with cold toluene. The compounds were isolated as red (4.1, 4.3) and orange (4.2) solids. Crystals suitable for X-ray diffraction were grown by vapor diffusion of heptane into a CHCl3 solution (4.1),

126 hexane into a CHCl3 solution (4.2), and slow evaporation of EtOH (4.3) at room temperature.

−1 1 4.1: Yield: 0.067 g (85.3%). IR: 2012, 1868 cm (νCO). H NMR (300

MHz, CDCl3): δ = 8.68 (d, 7.3 Hz, 2H, H on py), 7.98 (m, 2H, H on isoindoline),

7.84 (t, 7.3 Hz, 2H, H on py), 7.60 (m, 2H, H on isoindoline), 7.57 (d, J = 8.2 Hz,

2H, H on py), 7.14 (t, 7.3 Hz, 2H, H on py). HRMS (ESI-TOF, positive mode) m/z:

+ calcd for C21H13N5O3Re 570.0570, found 570.0590 [M+H] . UV-Visible (λmax nm, ε

× 103M-1 cm-1): 320 (9.6), 476 (2.0).

−1 1 4.2: Yield: 0.073 g (88.7%). IR: 2017, 1869 cm (νCO). H NMR (300

MHz, CDCl3): δ = 8.47 (d, J = 5.0 Hz, 2H, H on py), 8.08 (m, 2H, H on isoindoline),

7.66 (m, 2H, H on isoindoline), 7.31 (s, 2H, H on py), 6.67 (d, J = 5.9 Hz, 2H, H on py), 2.41 (s, 6H, CH3). HRMS (ESI-TOF, positive mode) m/z: calcd for

+ 3 -1 C23H17N5O3Re 598.0883, found 598.0905 [M+H] . UV-Visible (λmax nm, ε × 10 M cm-1): 320 (9.6), 483 (2.0). UV-Visible (nm, ε × 103M-1 cm-1): 294 (13), 378 (14),

470 (4.5).

−1 1 4.3: Yield: 0.071 g (79.9%). IR: 2009, 1864 cm (νCO). H NMR (500

MHz, d6-DMSO): δ = 13.43 (s, 2H, NH), 8.05 (d, J = 8.3 Hz, 2H, H on benzimidazole), 7.94 (m, 2H, H on isoindoline), 7.74 (m, 2H, H on isoindoline),

7.49 (t, 7.3 Hz, 2H, H on benzimidazole), 7.45 (d, 7.8 Hz, 2H, H on benzimidazole),

7.31 (t, 7.3 Hz, 2H, H on benzimidazole). 13C{1H} NMR (125 MHz, d6-DMSO): δ

= 194.3 (CO), 167.2, 153.6, 141.5, 136.0, 132.5, 132.2, 123.6, 123.5, 122.1, 116.8,

111.9. HRMS (ESI-TOF, positive mode) m/z: calcd for C25H15N7O3Re 648.0788,

127

+ 3 -1 -1 found 648.0788 [M+H] . UV-Visible (λmax nm, ε × 10 M cm ): 283 (18), 362 (19),

473 (5.0).

128

CDCl3

7.27

Normalized Intensity Normalized

7.61

7.58

7.14

7.60

7.59

8.69

8.70

8.68

7.57

7.83

8.67

7.14

7.16

7.54

7.16

7.97

7.99 7.81

8.00

7.13

7.80

7.98

7.12

7.86

2 7.11

1.97 2.00 1.97 2.12 1.74 2.11

8.5 8.0 7.5 7.0 Chemical Shift (ppm)

1 Figure 4.2: H NMR (300 MHz) of 4.1 in CDCl3.

129

2.41

12 CDCl3 10

8 7.27

Normalized Intensity Normalized

8.47 4 8.48

7.31

7.67

6.95 6.97 water

8.08 2 8.07

1.85 1.79 2.13 1.98 2.00 6.15

12 11 10 9 8 7 6 5 4 3 2 1 Chemical Shift (ppm)

CDCl3

7.27

3.5

8.47

8.48

3.0

7.31

7.67

7.64 2.5 7.66

6.95

6.97

2.0

Normalized Intensity Normalized

1.5

8.08

8.07

1.0

0.5

1.85 1.79 2.13 1.98 2.00

8.5 8.4 8.3 8.2 8.1 8.0 7.9 7.8 7.7 7.6 7.5 7.4 7.3 7.2 7.1 7.0 6.9 Chemical Shift (ppm)

1 Figure 4.3: H NMR (300 MHz) of 4.2 in CDCl3.

130

0.50

0.45

0.40

0.35

0.30

0.25

7.45 0.20 8.05

8.07

Normalized Intensity Normalized

7.94

7.75

7.47

7.95

7.73

0.15 7.74

7.74

7.31

7.49

0.10 7.33

7.51

7.30

0.05

2.15 1.75 2.13 1.81 2.27 2.00

8.1 8.0 7.9 7.8 7.7 7.6 7.5 7.4 7.3 Chemical Shift (ppm)

Figure 4.4: 1H NMR (500 MHz) of 4.3 in d6-DMSO.

131

Table 4.1: X-ray crystal data and structure parameters for compounds 4.1-4.3.

Compound 4.1 4.2 4.3 Empirical C21H12N5O3Re C23H16N5O3Re C25H14N7O3Re formula Formula 568.56 596.61 646.63 weight Crystal Monoclinic Monoclinic Tetragonal system Space group P21/m P21/m I4/m a/ Å 6.7522(2) 6.595(2) 16.6881(12) b/ Å 15.3489(4) 17.690(6) 16.6881(12) c/ Å 9.0349(3) 9.146(3) 16.5835(15) α(°) 90 90 90 β(°) 106.1680(10) 107.11(3) 90 γ(°) 90 90 90 Volume (Å3) 899.33(5) 1019.8(6) 4618.4(8) Z 2 2 8 Dc (Mg/m3) 2.100 1.943 1.860 µ (mm-1) 13.537 5.995 5.306 F(000) 544 576 2496 reflns 5922 10023 19276 collected 1581 [R(int) = 2619 [R(int) = 2980 [R(int) = indep. reflns 0.0210] 0.1030] 0.2059] Data / 1581 / 12 / 142 2619 / 30 / 152 2980 / 0 / 169 restraints / parameters GOF on F2 1.011 1.103 0.972 2 R1 (on Fo , I 0.0269 0.0666 0.0686 > 2σ(I)) 2 wR2 (on Fo , I 0.0712 0.1343 0.1323 > 2σ(I)) R1 (all data) 0.0270 0.0952 0.1503 wR2 (all data) 0.0713 0.1426 0.1561

132

Results and Discussion

Over the past few years, we have been investigating the coordination

156,243,332 chemistry of isoindoline based chelates with the Re(CO)3 moiety. The rigid nature of the facial geometry of this unit plus the lack of lability of the CO ligands can be used in template ligand synthesis and to prevent macrocycle formation.

This was demonstrated using Re(CO)3 to generate asymmetric bidentate chelates containing an isoindoline fragment and referred to as half of a hemiporphyrazine macrocyle (‘semihemiporphyrazine’).156 In this work we investigated the reaction between three BAI’s with different aryl groups (2-pyridyl, 4-methyl-2-pyridyl, and 2- benzimidazolyl) and Re(CO)5Cl as shown in Figure 4.1. After bringing the solution to reflux, a red or orange precipitate was collected. As seen with other isoindoline- rhenium complexes,156,332 we were initially expecting the formation of a complex with the ligand acting as bidentate ligand. However, we observed formation of a tridentate, facial coordination mode for this ligand. No bidentate product was observed. Reactions of the BAI ligands with Re(CO)5Br were then attempted. The tridentate products were formed to a limited extent, but the decreased lability of the Br anion led to incomplete reactions and complex mixtures. We were able to structurally characterize the products by single crystal X-ray methods, and the structures are shown in Figure 4.5. The structures confirm that the BAI ligands coordinate in a symmetric fashion to three facial positions on the metal center. The ligands are deprotonated at the isoindoline and thus monoanionic, providing charge balance for the Re(I) center. As expected for Re(CO)3 complexes with

133 nitrogenous bases, the carbonyls are facially arranged. The BAI ligand binds more tightly at the isoindole nitrogen, with Re-N bond lengths ranging from ~2.10 to 2.15

Å, while the two aryl nitrogen atoms bind to the metal with longer Re-N lengths ranging from ~2.22 to 2.24 Å. The metal-carbon and carbon oxygen bonds of the carbonyls are typical for Re(CO)3 complexes, but it notable that the M-C bond trans to the isoindoline is slightly longer than those across from the pyridine units for 4.1 and 4.2.

134

Figure 4.5: Structure of compounds 4.1-4.3, with 35% probability ellipsoids.

Hydrogen atoms except on nitrogen atom positions have been omitted for clarity.

In order to accommodate the facial coordination environment, both the angle of bonding of the isoindoline and the orientations of the pyridine units on the ligand have to deform from a planar conformation. The plane defined by the isoindole unit lies at a ~33° (4.1 and 4.2), and ~22° (4.3) angle to the N-Re-C bond, rather than binding straight on as seen in meridional coordination environments. 135

Additionally, the pyridine and benzimidazole units also deviate from the plane of the isoindoline by ~25°, 26°, and 30° for 4.1-4.3 toward their N-Re-C bond ordinates. The planes of the aryl groups, however, are not coincident with these bonds and deviate by ~13°. Thus, there are two types of deformation observed in these complexes.

The three Re(CO)3(BAI) complexes are diamagnetic and could be characterized by NMR spectroscopy (Figures 4.2-4.4). However, compounds 4.1 and 4.2 have limited solubility in CHCl3 and were unstable in other solvents, making it challenging to obtain 13C NMR spectra. In the 1H spectra for 4.1-4.3, we observed the expected resonances for a high symmetric tridentate complex. The isoindoline aromatic protons exhibit AA′BB′ spin system patterns. We also observed the disappearance of the isoindoline N-H proton, as expected upon metal coordination. In the IR spectra, the Re(CO)3 unit shows a1- and e-type CO stretches produced by the pseudo-C3v environment of the facial carbonyl units with frequencies that range from ∼2009 to 2017 cm−1 and from ∼1864 to 1869 cm−1, respectively.

Many Re(CO)3 complexes with bound diimine or similar conjugated ligand systems exhibit metal to ligand charge transfer transitions in their UV-visible spectra. The UV-visible spectra for compounds 4.1-4.3 are shown in Figure 4.6, and the TDDFT experimental and calculated spectra are shown in Figure 4.7 for compounds 4.1 and 4.3. The tridendate complexation of the ligands to the metal center gives rise to bands around ~470 nm that we can characterize as MLCT

136 transitions in the intermediate energy region of the visible spectrum. The extinction coefficients for the these Re(CO)3(BAI) MLCT bands are observed to be between

−1 −1 2000 and 5000 M cm , which is in good agreement with those seen for Re(CO)3 diimine compounds.156,243,332–337 The lower energy transitions observed in these spectra have maxima above 450 nm, which is lower than that seen for many

Re(CO)3 diimine compounds as well as our previously synthesized semihemiporphyrazine Re(CO)3 DII complexes, which exhibit their low energy maxima above 500 nm.

20 1 2 3 15

-1

-1 10

, M

-3

0 300 350 400 450 500 550 600 Wavelength, nm

Figure 4.6: UV-visible spectra for compounds 4.1-4.3 in CHCl3 (DMF for 4.3).

137

Figure 4.7: Experimental and TDDFT predicted spectra for compounds 4.1 (top) and 4.3 (bottom).

To ascertain the nature of these transitions, we carried out DFT and TDDFT calculations on compounds 4.1 and 4.3. The frontier molecular orbitals are shown in Figure 4.8 for these two compounds. The HOMOs for both complexes exhibit significant d orbital character, while the LUMOs are composed of primarily ligand based orbitals. The energy levels of the frontier orbitals for 4.1 and 4.3 are also 138 shown in Figure 4.8; both the HOMO and LUMO energies are readily isolated from other occupied and unoccupied orbitals. Additionally, the HOMO exhibits a combination of metal d orbital and iminoaryl ring character, while the LUMO has a significant amount of isoindoline character. The HOMO−LUMO gaps are at ∼2.63 and ∼2.35 eV for 4.1 and 4.3 respectively. These spacings are much lower than that observed for the semihemiporphyrazine Re(CO)3 complexes, which ranged from 5.2-5.4 eV. TDDFT methods reveal that the low energy optical transitions can be characterized as HOMO to LUMO transitions, and thus we can assign them as MLCT in character.

139

Figure 4.8: Experimental and TDDFT-predicted molecular orbital energy levels for compounds 4.1 and 4.3. The structures of the HOMOs and LUMOs are shown along with percent metal/ligand composition.

140

Conclusions

In conclusion, we were able to readily synthesize a series of tridendate bis(arylimino)isoindoline conjugates with Re(CO)3 using mild conditions. The elucidated X-ray crystal structures of the Re(CO)3(BAI) complexes show a facial binding mode, which is unexpected for this class of ligand. The facial coordination mode distorts the ligand from planarity, which affects the degree of orbital overlap between the iminoaryl groups and the central isoindoline. Compounds 4.1-4.3 exhibit MLCT bands in their UV-visible spectra, which are similar to those observed in Re(CO)3 diimine complexes, however the HOMO-LUMO energy gap is smaller than seen in other isoindoline-Re(CO)3 compounds. We are continuing our work on the fundamental chemistry of rhenium carbonyls, and how the preferred facial coordination geometry can be used to control ligand binding modes.

141 CHAPTER V

1,3-DIYLIDENEISOINDOLINES: SYNTHESIS, STRUCTURE, REDOX, AND OPTICAL PROPERTIES

THE TEXT OF THIS CHAPTER IS ADAPTED FROM THE MATERIAL AS IT APPEARS IN: ZATSIKHA, Y. V; SCHRAGE, B. R.; MEYER, J.; NEMYKIN, V. N.; ZIEGLER, C. J. 1,3-DIYLIDENEISOINDOLINES: SYNTHESIS, STRUCTURE, REDOX, AND OPTICAL PROPERTIES. J. ORG. CHEM. 2019, 84 (10), 6217– 6222. COPYRIGHT © 2019 AMERICAN CHEMICAL SOCIETY.

DOI: 10.1021/ACS.JOC.9B00468

Introduction

1,3-Diiminoisoindoline (DII, Scheme 5.1) has long been an important reagent for the synthesis of chelates and chromophores.195,338–341 First introduced by Linstead in the early 1950s and produced by the reaction of ammonia with phthalonitrile,195 DII can be used to synthesize phthalocyanines as well the related hemiporphyrazines, bis(arylimino)isoindoline chelates, and related chromophores.140,154,157,201,203,332,342–351 One of the most common types of reactivity observed with DII is the formation of Schiff bases upon exposure to primary amines, which is exemplified in the hemiporphyrazines and bis(arylimino)indoline ligands. However, the DII unit can potentially undergo a

142 variety of chemical reactions to potentiate its use as a building block for a variety of structural motifs. This chemistry, outside of the well-known Schiff base chemistry, has been generally unexplored. However, this chemistry could provide a route for generating more complex molecular architectures. One example of such as reaction was explored by Lever and coworkers, who exploited the ring expansion reaction of DII with hydrazine to form phthalazines as a means to generate metal chelating compounds.158,249,352–355

In this report, we present a study into the use of organic CH-acids to prepare 1,3-diylideneisoindolines (DII, Scheme 5.1) from DII. Although several similar chromophores originated from the reaction between DII and organic CH- acids are reported in the patent literature356–361 and commercially available as yellow or orange pigments, their general spectroscopic, redox, and electronic structure properties are yet to be reported. In order to fill this knowledge gap, we synthesized four new compounds 5.1-5.4 as shown in the Scheme 5.1 using a one-step reaction between DII and organic CH-acids chosen from four different organic CH-acids: cyanoethylacetate (5.1), Meldrum’s acid (5.2), 1,3-indanedione

(5.3), and dimethyl barbituric acid (5.4). Compounds 5.1-5.4 exhibit intense UV- visible bands which are highly dependent on the identity of the substituents at the alkene bridging positions. We have probed their electronic structure through spectroscopy, electrochemistry, and Density Functional Theory (DFT) methods, and have elucidated the natures of the π-π* transitions using time-dependent DFT

(TDDFT).

143

Scheme 5.1: Synthetic route for preparation of the compounds 5.1-5.4.

144

Experimental

General Information

NMR spectra were recorded on a 300 MHz spectrometers and chemical shifts were given in ppm relative to residual solvent resonances (1H NMR and 13C

NMR spectra). High-resolution mass spectrometry experiments were performed on a Bruker MicroTOF-III and MicroTOF-qIII instruments. Infrared spectra were collected on Thermo Scientific Nicolet iS5 that was equipped with an iD5 ATR.

UV-visible spectra were recorded on a Hitachi 3010 and Jasco V-770 spectrometers.

Bruker SHELXTL Software Package (Version 6.1),362 and were solved using direct methods until the final anisotropic full-matrix, least squares refinement of F2 converged. CCDC 1886266-1886268 contain the supplementary crystallographic

145 data for this paper. The data can be obtained free of charge from The Cambridge

Crystallographic Data Centre via www.ccdc.cam.ac.uk/structures.

Computational Details

The starting geometries of compounds 5.1-5.4 were optimized using a

TPSSh exchange-correlation functional.326 Energy minima in optimized geometry was confirmed by the frequency calculations (absence of the imaginary frequencies). Solvent effect was modeled using the polarized continuum model

(PCM).327 In all calculations, DCM was used as the solvent. In PCM-TDDFT calculation, the first 30 states were calculated. All atoms were modeled using 6-

311G(d)363 basis set. Gaussian 09 software was used in all calculations.330

QMForge program was used for molecular orbital analysis.331

Syntheses

Synthesis of 5.1-5.4: A mixture of DII (1 mmol, 144 mg) and the corresponding CH- acid (2.1 eq) were dissolved in glacial acetic acid (10 mL) and refluxed for 1-3 min. Resulting solution was cooled down to room temperature, diluted with water, and filtered to give orange (or yellow) powder of pure compounds 1-4. Yields ranged from 71 – 90 %.

5.1. 2,2'-(1H-isoindole-1,3(2H)-diylidene)-bis(cyanoacetic acid ethyl ester):

1 Yield 254 mg (79 %). H NMR (300 MHz, CDCl3) δ: 13.08 (s, 1H), 8.68 – 8.65 (m,

2H), 7.82 – 7.79 (m, 2H), 4.44 (q, J = 7.1 Hz, 4H), 1.43 (t, J = 7.1 Hz, 6H); 13C{1H}

NMR (CDCl3, 125 MHz) δ: 168.7, 155.4, 133.6, 132.2, 126.0, 115.2, 80.0, 62.8,

14.2; IR (cm-1) 3278, 2994, 2979, 2222, 2213, 1699, 1683, 1592, 1469, 1453,

146

1368, 1272, 1221, 1169, 1156, 1141, 1093, 1018, 855, 768; HRMS (ESI-TOF,

+ positive mode) m/z: calcd for C18H16N3O4 338.1110, found 338.1141 [M+H] ; UV- vis, nm (e x 104 M-1cm-1): 372 (2.03), 394 (3.71), 418 (3.98).

5.2. 5,5'-(1H-isoindole-1,3(2H)-diylidene)bis-(2,2-dimethyl-1,3-dioxane-4,6-

1 dione): Yield 283 mg (71 %). H NMR (300 MHz, CDCl3) δ 15.41 (s, 1H), 9.34 –

13 1 9.28 (m, 2H), 7.72 – 7.66 (m, 2H), 3.50 (s, 6H), 3.46 (s, 6H); C{ H} NMR (CDCl3,

125 MHz) δ: 164.3, 161.7, 159.7, 150.7, 134.5, 133.9, 131.2, 99.3, 29.2, 29.0; IR

(cm-1) 3390, 2958, 2219, 2211, 1686, 1512, 1470, 1458, 1433, 1418, 1338, 1307,

1203, 1174, 1152, 1090, 773, 703, 703; HRMS (ESI-TOF, positive mode) m/z:

+ 4 calcd for C20H17NO8Na 422.0852, found 422.0842 [M+Na] ; UV-vis, nm (e x 10

M-1cm-1): 393 (1.91), 416 (2.65), 441 (2.88).

5.3. 2,2'-(1H-isoindole-1,3(2H)-diylidene)bis-(1H-indene-1,3(2H)-dione):

1 Yield 342 mg (85 %). H NMR (300 MHz, CDCl3) δ 14.30 (s, 1H), 9.73 – 9.70 (m,

2H), 8.05 – 8.00 (m, 4H), 7.87 – 7.82 (m, 6H); IR (cm-1) 3295, 3239, 1678, 1637,

1559, 1450, 1354, 1221, 1095, 1018, 881, 813, 730, 690, 658; HRMS (APCI-TOF,

- negative mode) m/z: calcd for C26H12NO4 402.0772, found 402.0785 [M-H] ; UV- vis, nm (e x 104 M-1cm-1): 370 (1.56), 387 (1.58), 425 (2.01), 458 (5.37), 489 (9.36).

5.4. 5,5'-(1H-isoindole-1,3(2H)-diylidene)bis[1,3-dimethyl-2,4,6(1H,3H,5H)-

1 pyrimidinetrione]: Yield 380 mg (90 %) H NMR (300 MHz, CDCl3) δ 8.59 – 8.56

(m, 2H), 7.94 – 7.91 (m, 2H), 1.55 (s, 6H); IR (cm-1) 3015, 1965, 1716, 1662, 1633,

1581, 1512, 1461, 1438, 1414, 1387, 1373, 1328, 1314, 1292, 1259, 1217, 1193,

1169, 1105, 1065, 1053, 1042, 974, 945, 871, 818, 799, 788, 765, 756, 750, 717,

147

687, 682, 671, 619; HRMS (ESI-TOF, positive mode) m/z: calcd for C20H17N5O6Na

446.1070, found 446.1077 [M+Na]+; UV-vis, nm (e x 104 M-1cm-1): 421 (1.77), 447

(3.12), 474 (3.90).

148

1 Figure 5.1: H NMR of 5.1 in CDCl3.

149

1 Figure 5.2: H NMR of 5.2 in CDCl3.

150

1 Figure 5.3: H NMR of 5.3 in CDCl3.

151

1 Figure 5.4: H NMR of 5.4 in CDCl3.

152

Table 5.1: X-ray crystal data and structure parameters for compounds 5.1, 5.2, and 5.4.

Compound 5.1 5.2 5.4 Empirical formula C36H30N6O8 C20H17NO8 C20H17N5O6 Formula weight 674.66 399.35 423.39 Crystal system monoclinic triclinic monoclinic Space group P2/n P-1 C2/c a/ Å 15.0392(17) 6.6482(2) 38.945(2) b/ Å 4.6544(5) 12.1779(3) 4.5998(3) c/ Å 23.336(2) 12.7330(3) 26.6918(16) α(°) 90 62.6070(10) 90 β(°) 93.506(7) 75.839(2) 133.009(2) γ(°) 90 88.593(2) 90 Volume (Å3) 1630.4(3) 882.82(4) 3496.5(4) Z 2 2 8 Dc (Mg/m3) 1.374 1.502 1.609 µ (mm-1) 0.099 0.118 1.029 F(000) 704 416 1760 reflns collected 25568 16682 19562 indep. reflns 3985 4475 3100 GOF on F2 0.954 1.074 0.989 2 R1 (on Fo , I > 0.0589 0.0433 0.0358 2σ(I)) 2 wR2 (on Fo , I > 0.1195 0.1175 0.1158 2σ(I)) R1 (all data) 0.1449 0.0577 0.0425 wR2 (all data) 0.1527 0.1268 0.1251

153

Results and Discussion

The formation of 1,3-diylideneisoindolene was first observed by Elvidge and

Lindstead in 1956 upon the exposure of DII to an excess of ethylcyanoacetate.364

This chemistry was not extensively pursued in the chemical literature after this point, with the exception of some appearances in the dye patent literature.356–361

Recently, we rediscovered this chemistry with 5,5-dimethyl-1,3-cyclohexanedione

(dimedone) as CH-acid,365 and now found that we could extend it to a variety of organic CH-acids with activated methylene groups (pKa < 20). Additionally, the reaction can be optimized to use stoichiometric amounts of the organic acid by using glacial acetic acid as the solvent and the catalyst. The four compounds shown in Scheme 5.1 can be readily prepared in high yield and isolated as crystalline solids. The resultant materials, however can have limited solubility, which challenges characterization by spectroscopic methods such as 13C NMR spectroscopy (compounds 5.3 and 5.4). However, all of the 1H NMR spectra for

5.1-5.4 exhibit the expected resonances (Figures 5.1-5.4), with diagnostic AA’BB’ spin system patterns for the isoindoline aromatic protons as well as protons from the CH-acid fragments.

We were able to elucidate the molecular structures of the ethylcyanoacetate

(5.1), Meldrum’s (5.2), and dimethylbarbaturic (5.4) acid variants of the DYI compounds shown in Figure 5.5. The analogous barbaturic acid compound had been structurally elucidated previously.366 All three compounds characterized in

154 this work are planar, with the exception of the sp3 hybridized carbons on the

Meldrum’s acid derived species. In each case, the imine C=N double bonds of the parent DII has been replaced with alkene-type C=C bonds. The bond lengths for these alkene units are on the range of ~1.36-1.38 Å for the four compounds.

Clearly these distances are slightly longer than a traditional carbon-carbon double bond, and inspection of the C-N bonds of the isoindoline unit (~1.36-1.37 Å) reveal some delocalization onto the ring. The central nitrogen in the isoindoline retains its ionizable hydrogen atom position, which engages in hydrogen bonding interactions with oxygen atoms from the organic acid groups. The hydrogen bond length (as measured from internal nitrogen atoms to oxygen atoms) range between

2.6 and 2.7 Å.

155

Figure 5.5: The structures of compounds 5.1, 5.2, and 5.4 with 35% thermal ellipsoids. Non-ionizable hydrogen atoms have been omitted for clarity.

Compounds 5.1-5.4 exhibit a range of colors from red to yellow; the spectra of the four species are shown in Figure 5.6. All four compounds exhibit similar features in their spectra, with each showing three bands of increasing intensity with decreasing energy. It is tempting to correlate the UV-visible features of compounds

5.1-5.4, with the pKa of the component acid, but inspection of the values indicates that this trend does not exist. The pKa values for ethyl cyanoacetate, 1,3- indandione, Meldrum’s acid, and 5,5-dimethylbarbituric acid are 9.0, 8.9, 7.3, and

4.1 respectively, however the absorption maxima of the lowest energy bands for

156 these compounds are 418, 489, 441, and 469 nm respectively. The extinction coefficients for these compounds are relatively large, on the order of ~2-9 x 104 M-

1 cm-1. None of the four compounds exhibit any appreciable fluorescence.

Figure 5.6: UV-visible spectra for compounds 5.1-5.4 in DCM.

157

Figure 5.7: Cyclic voltammograms for compounds 5.1-5.4 recorded in DMF/0.1

+ TBAPF6 system at room temperature. Redox potentials (V) versus FcH/FcH are displayed in the table.

158

The redox properties of all compounds were investigated using cyclic voltammetry (CV) approach (Figure 5.7). Each compound exhibits two quasi- reversible reduction processes. The first reduction potentials shift to more negative values in the order of 5.1 (most negative) < 5.3 ~ 5.4 < 5.2 (least negative). These values do not agree with the observed energies of transitions in the UV-visible spectra, which rank from 5.1 (highest energy) > 5.2 > 5.4 > 5.3 (lowest energy).

The gap between the first and a second reduction processes is greatest for compound 5.1 (0.54 V), followed by 5.2, while compounds 5.3 and 5.4 show similar gaps (~0.41 V). We were not able to observe any oxidation processes in the electrochemical window.

In order to understand the observed trends in UV-visible spectroscopy and electrochemistry, we further probed the electronic structures and optical spectroscopy of 5.1–5.4 using DFT and TDDFT calculations. The DFT-predicted frontier MOs and molecular energy diagram are presented in Figure 5.8. In all cases, the DFT-predicted HOMO and LUMO orbitals resemble each other. Indeed, the HOMOs of 5.1–5.4 are dominated by the contribution of the isoindole nitrogen atom, carbon atom directly attached to the isoindole fragment, and carbonyl oxygen atoms. The LUMOs for all compounds are dominated by the contributions from isoindole carbon atoms and C-C=O fragments directly attached to the isoindole terminal alkene units (Figure 5.8). It is interesting to note that the DFT- predicted differences in the energies of the LUMOs in 5.1–5.4 (ΔE = 0.171 eV) are

159 very small, but the energies of the HOMOs in these compounds varies more significantly (ΔE = 0.360 eV).

Figure 5.8: DFT-predicted frontier orbitals (top) and energy levels (bottom) for compounds 5.1-5.4. 160

The experimental UV-vis spectra of compounds 5.1–5.4 correlate well with the TDDFT-predicted ones (Figure 5.9). In particular, the well-resolved experimental spectra of compounds 5.1 and 5.3 can be explained as follows. The low-energy, most intense band in these compounds is described as a predominantly HOMO → LUMO single-electron excitation predicted by TDDFT at

422 and 487 nm respectively. This band is complemented in experimental spectra by 0-1 and 0-2 vibronic satellites observed at 419 and 394 (compound 5.1) and

489 and 456 nm (compound 5.3), respectively. Such vibronic satellites are very characteristic for the previously reported isoindole functional dyes and their analogues. According to our TDDFT calculations, the higher-energy regions of the

UV-vis spectra of 5.1 and 5.3 are dominated by the HOMO-n → LUMO single- electron transitions. In the case of dyes 5.2 and 5.4, in addition to the most intense

HOMO → LUMO transitions, TDDFT predicts a large number of the HOMO-n →

LUMO transitions in the visible energy envelope that are responsible for the broader experimental spectra of these compounds.

161

Figure 5.9: Experimental and TDDFT-predicted spectra for compounds 5.1-5.4.

162

Conclusions

In conclusion, we have revisited and expanded the chemistry of Linstead concerning the synthesis of a series of four 1,3-diylideneisoindolines. All of these compounds can be readily prepared in one-step from DII, and the four compounds are intensely colored. X-ray structural elucidation reveals planar structures, with the terminal alkene units clearly conjugated with the central isoindoline unit. These compounds exhibit intense absorption bands in the UV-and visible range, and the energies of these transitions are highly dependent on the structure of the organic acid substituent. DFT calculations reveal that these compounds exhibit similar

LUMO energy levels, but that the HOMO energies vary depending on the alkene substituent and possess a significant degree of alkene character. We are continuing to explore the chemistry of the isoindolines as a means to produce new and novel chromophore systems.

163 CHAPTER VI

THE SYNTHESIS OF A HEXAMERIC EXPANDED HEMIPORPHYRAZINE

THE TEXT OF THIS CHAPTER IS ADAPTED FROM THE MATERIAL AS IT APPEARS IN: SCHRAGE, B. R.; CHANAWANNO, K.; CRANDALL, L. A.; ZIEGLER, C. J. THE SYNTHESIS OF A HEXAMERIC EXPANDED HEMIPORPHYRAZINE. J. PORPHYR. PHTHALOCYANINES 2020, 24 (1–3). COPYRIGHT © 2020 WORLD SCIENTIFIC PUBLISHING COMPANY

DOI: 10.1142/S1088424619500901

Introduction

The synthesis of expanded porphyrins has evolved into a significant area in the field of porphyrin and phthalocyanine analogs. Although examples of expanded porphyrinoids have been around for decades, such as texaphyrin,367,368 and sapphyrin,369–371 the chemistry of this class of macrocycle increased appreciably after 2001, in particular with the contributions of Osuka and coworkers,

372 One of the more heavily studied systems since 2001 has been hexaphryin(1.1.1.1.1.1) (Figure 6.1), which can adopt multiple reduction states (as both (26)hexaphryin(1.1.1.1.1.1) and (28) hexaphryin(1.1.1.1.1.1)), is highly 164 flexible, and often binds more than one equivalent of a metal ion.373–375 These compounds have been used to probe fundamental aspects of aromaticity and its relationship with macrocycle conformation, protonation state, and degree of oxidation/reduction.

Figure 6.1: Hemiporphyrazine and several expanded macrocycles : hexaphyrin

(A), superphthalocyanine (B), hemiporphyrazine (C), thiadiazole-expanded hemiporphyrazine (D), and hexahemiporphyrazine (E).

Although the expansion of porphyrinic macrocycles has been considerably developed, the same is not true for expanded phthalocyanine analogs. The pentameric system superphthalocyanine is a well-studied example, but is only 165

formed when phthalonitrile is condensed with the uranyl cation as a template. 145,164

Hemiporphyrazines, phthalocyanine analogs with one or more aryl groups replacing isoindolines, has been an alternative strategy for forming expanded phthalocyanine-like ring systems,175,222,224,344,376–378 The most commonly investigated hemiporphyrazine has been the bis-pyridine system,379 but macrocycles with One example of such an expanded hemiporphyrazine was achieved using thiadiazole as the alternate aryl group, affording the hexameric macrocycle shown in Figure 6.1.342,380 More recently, a new pentameric hemiporphyrazine has been developed using para phenylene diamine.221

In this report, we present the synthesis of a hexahemiporphyrazine, comprised of four equivalents of pyridine and two equivalents of isoindoline. The precursor to this macrocycle is bis(6-aminopyridiyl)amine, which condenses with one equivalent of diiminoisoindoline to afford the resultant hexahemiporphyrazine macrocycle. As in normal bis-pyridyl hemiporphyrazine and similar macrocycles, this expanded ring lacks cross conjugated aromaticicy, as evidenced by NMR and

UV-visible spectroscopy. As in hexaphyrin and in other expanded porphyrinoids, hexahemiporphyrazine exhibits inversion of two of the ring subunits in the macrocycle; in the present case two pyridines are oriented such that the nitrogen atom position faces the exterior of the ring.

166

Experimental

General Information

All reagents and starting materials were purchased from commercial vendors and used without further purification. 1,3-diiminoisoindoline (DII) was synthesized according to a previously published procedure.157 The di-(6-amino-2- pyridyl)-amine was prepared by a modification of a previously published procedure, described below.381 The 2,6-diaminopyridine hydrochloride starting material was prepared by the reaction of HCl with 2,6-diaminopyridine. Deuterated solvents were purchased from Cambridge Isotope Laboratories and used as received.

NMR spectra were recorded on a 500 MHz spectrometer and chemical shifts were given in ppm relative to residual solvent resonances (1H NMR and 13C

NMR spectra). High resolution mass spectrometry experiments were performed on a Bruker MicroTOF-III instrument. Infrared spectra were collected on Thermo

Scientific Nicolet iS5 which was equipped with an iD5 ATR. Electronic absorption spectra were recorded on Hitachi U-2000 UV–vis spectrophotometer.

167

Bruker SHELXTL Software Package (Version 6.1), and were solved using direct methods until the final anisotropic full-matrix, least squares refinement of F2 converged. CCDC numbers 1917893-1917896 contain the supplementary crystallographic data for this paper. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/structures

Syntheses

Synthesis of di-(6-amino-2-pyridyl)-amine (6.1): Di-(6-amino-2-pyridyl)- amine was prepared by solvent-free melt reaction. 2.00 g of 2,6-diaminopyridine

(18.3 mmol) was fused with 2.67 g of 2,6-diaminopyridine hydrochloride (18.3 mmol) for 48 hours at 190°C. The reaction mixture was cooled to room temperature and 200 mL of DI water was added. The solution was then filtered through a pad of Celite. 1.5 equivalents of NaOH (1.10 g, 27.5 mmol) were added to the filtrate, and left to stir at room temperature overnight. The aqueous solution was extracted with CHCl3. The organic layer was collected, and solvent was removed under vacuum. Excess 2,6-diaminopyridine was sublimed, while the remaining residue was dissolved in CHCl3, and washed 3x with water. The organic fraction was reduced in volume and hexane was added. The precipitate that formed was filtered, and isolated as a beige solid. Single crystals of 6.1 were obtained by evaporation from a chloroform solution. Crystals of the HCl adduct of

6.1 were isolated from a vapor diffusion of hexane into an EtOH solution.

168

Bis(6-amino-2-pyridyl)amine (6.1): Yield: 0.56 g (15.2 %). IR (N-H stretch,

-1 1 cm ): 3448 (m), 3300 (m). H NMR (500 MHz, d6 - DMSO): δ = 5.52 (s, 4H, NH2),

5.90 (d, J = 7.8 Hz, 2H, H on py), 6.87 (d, J = 7.8, 2H, H on py), 7.19 (t, J = 7.8,

13 1 2H, H on py), 8.46 (s, 1H, N-H). C{ H} NMR (125 MHz, d6 - DMSO): δ = 158.1,

153.4, 138.2, 99.0, 98.7. ESI MS (positive mode) calcd C10H12N5 202.1093 m/z, found 202.1086.

Synthesis of hexahemiporphyrazine (6.2): Di-(6-amino-2-pyridyl)-amine

(0.050 g, 0.248 mmol) and DII (0.036 g, 0.248 mmol) were suspended in 5 mL of

EtOH, and BF3O(C2H5)2 (0.0276 mmol) were added. The solution was refluxed overnight and reduced in volume. The crude product was purified by column chromatography (silica, 5% MeOH: CH2Cl2) to yield the target ligand as a yellow solid. Single crystals were isolated from a vapor diffusion of acetone into a solution of 6.2 in chloroform.

Hexahemiporphyrazine (6.2): Yield: 0.011 g (14.2 %). IR (N-H stretch, cm-

1 1 ): 3303 (m). H NMR (500 MHz, d6 - DMSO): δ = 6.72 (d, J = 7.3 Hz, 4H, H on py), 7.12 (d, J = 8.3, 4H, H on py), 7.51 (t, J = 7.8, 4H, H on py), 7.81 (m, 4H, H on isoindoline), 8.05 (m, 4H, H on isoindoline), 9.52 (s, 2H, N-H), 12.19 (s, 2H, N-H).

13 1 C{ H} NMR (125 MHz, d6 - DMSO): δ = 157.8, 152.6, 151.7, 151.5, 140.0, 134.4,

132.3, 122.3, 112.2, 107.2. ESI MS (positive mode) calcd C36H25N12 625.2325 m/z, found 625.2356.

169

water

5.52

3.30 18

7.19

5.91

5.90

12 6.89

6.87

DMSO 8 7.17

Normalized Intensity Normalized

8.46

7.21

2.50

1.13 2.09 2.02 2.00 4.29

9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 Chemical Shift (ppm)

1 Figure 6.2: H NMR (500 MHz) of 6.1 in d6 – DMSO.

170

water

DMSO

3.29 2.50

3.0

2.50

2.5

2.0

Normalized Intensity Normalized 1.5

8.05

9.52

7.82

8.07

7.81 1.0 7.80

7.81

12.19

8.06

7.51

6.73

6.72

7.49

7.52

0.5 7.83

1.95 2.18 4.17 4.21 4.43 4.18 4.03

12 11 10 9 8 7 6 5 4 3 Chemical Shift (ppm)

1 Figure 6.3: H NMR (500 MHz) of 6.2 in d6 – DMSO.

171

Table 6.1: X-ray crystal data and structure parameters for compounds DAP·HCl,

6.1, 6.1·HCl, and 6.2.

Compound DAP·HCl 6.1 6.1·HCl 6.2

Emp. form C5H10ClN3O C10H11N5 C22H32Cl2N10O C38H26Cl6 N12 2 Form. 163.61 201.24 539.47 863.41 weight Crystal Monoclinic Monoclinic Monoclinic Monoclinic system Space P21/c P21/n C2/c P21/c group a/ Å 6.7931(5) 9.7478(3) 28.795(3) 14.7399(11) b/ Å 12.9213(9) 7.6884(2) 11.2435(10) 6.8829(5) c/ Å 9.1285(6) 13.4222(4) 17.7434(16) 19.5268(13) α(°) 90 90 90 90 β(°) 106.644(3) 108.3800(10) 116.627(6) 105.794(5) γ(°) 90 90 90 90 Volume (Å3) 767.69(9) 954.61(5) 5135.3(9) 1906.3(2) Z 4 4 8 2 Dc (Mg/m3) 1.416 1.400 1.396 1.504 µ (mm-1) 0.434 0.092 2.620 4.506 F(000) 344 424 2272 880 Reflections 13903 21955 23392 5791 collected Data/Restrai 1942 / 0 / 2080 / 0 / 136 4427 / 0 / 347 5791 / 6 / 285 nts/Paramet 105 ers GOF on F2 1.092 1.047 1.545 1.561 2 R1 (on Fo , I 0.0256 0.0427 0.1243 0.1696 > 2σ(I)) 2 wR2 (on Fo , 0.0684 0.1131 0.3748 0.3984 I > 2σ(I)) R1 (all data) 0.0292 0.0450 0.1614 0.2023 wR2 (all 0.0710 0.1153 0.4175 0.4279 data)

172

Results and Discussion

The key reagent for the synthesize of the hemiporphyrazines are aryl diamines, such as 2,6-diaminopyridine and o-phenylenediamine. When these reagents are reacted with either diiminoisoindoline or phthalonitrile, ABAB type hemiporphyrazines are formed. In our investigations into new reagents for hemiporphyrazine synthesis, we proposed that we might link two diamine precursors prior to reaction to form the hemiporphyrazine. The reaction scheme used for the formation of hexahemiporphyrazine 2 is shown in Scheme 6.1. We were able to produce the bis(6-amino-2-pyridyl)amine by a melt reaction comprised of equal mole equivalents of 2,6-diaminopyridine and its hydrochloride salt (this latter compound was fully characterized, including via X-ray crystallography, Figure 6.4).

173

Figure 6.4: Structure of 2,6-diaminopyridine HCl salt (DAP·HCl), showing 35% probability ellipsoids. Hydrogen atoms on carbon positions have been omitted for clarity.

Although the reaction itself was quite straightforward, purification of the desired product was challenging. Pure product was eventually isolated by removal of the 2,6-diaminopyridine starting material via sublimation. The bis(6-amino-2- pyridyl)amine starting material was characterized by single crystal X-ray diffraction, and the structure is shown in Figure 6.5. This species can adopt multiple conformations, which are shown in Figure 6.6. In the solid state structure of bis(6- amino-2-pyridyl)amine, orientation A is adopted, although as will be shown below, orientation C is seen in the hexihemiporphyrazine backbone. When the bis(6- amino-2-pyridyl)amine precursor is protonated, orientation B is adopted, as can be

174 seen in Figure 6.5. The two pyridine nitrogen atoms face the same direction due to the presence of a hydrogen bonding interaction, with a N-N spacing of ~2.6 Å.

Scheme 6.1: Synthesis of 6.1 and 6.2.

175

Figure 6.5: The structure of bis(6-amino-2-pyridyl)amine (6.1, top) and its HCl salt

(6.1·HCl, bottom) with 35% thermal ellipsoids. Hydrogen atoms on carbon positions and counterions have been omitted for clarity.

The reaction of a 1:1 ratio of bis(6-amino-2-pyridyl)amine with DII produces a yellow colored product which can be purified by column chromatography. The

1H NMR spectrum of the product is distinctly different from that of either of the starting materials and is shown in Figure 6.7. The spectrum exhibits seven different proton chemical shifts. The isoindoline protons exhibit the typical AA’BB’ spin system pattern, which is indicative of symmetry with regard to this unit in the 176 macrocycle backbone. For the bis(6-amino-2-pyridyl)amine protons, three resonances are observed. Additionally, there are two NH resonances present in the spectrum at ~9.5 and ~12.4 ppm, which can be assigned as the bridging amine hydrogen atom and the internal NH position respectively. The chemical shifts observed in the spectrum are consistent with a non-cross conjugated electronic structure (neither aromatic nor anti-aromatic); a similar lack of ring current effects are seen in normal hemiporphyrazines.

177

Figure 6.6: The three possible conformations of bis(6-amino-2-pyridyl)amine.

178

Figure 6.7: 1H NMR spectrum of hexahemiporphyrazine 6.2, with a close up of the isoindoline and pyridine C-H resonances.

We were also able to characterize the product by single crystal X-ray diffraction. The elucidated structure is shown in Figure 6.8. The ratio of isoindoline to bis(6-amino-2-pyridyl)amine is 1:1, with two equivalents of each unit to produce a hexameric ring structure. The macrocycle is largely planar, with the exception of two of the pyridine rings, one from each of the bis(6-amino-2-pyridyl)amine units.

These two rings, which are at an angle of `34° to the plane of the ring, are also inverted such that the pyridine nitrogen atom faces the outside of the ring.

Examination of the bond distances in the macrocycle itself indicates that there is no cross conjugation across the macrocycle. The C-N bond distances between the pyridine rings and extracyclic nitrogen atoms are clearly single in character

179

(~1.36-1.40 Å). On the diiminoisoindoline unit, the C-N bonds on the outside of the ring are much shorter (imine character) than the internal C-N bonds, which are more single in character. As a result, we can clearly assign a proton to the internal isoindoline nitrogen atom position. The two inverted pyridine rings interact in a π stacking fashion in the solid state, with a distance of ~3.4 Å. However, in solution, the number of pyridine 1H resonances (three) indicates that this inversion is likely fluxional, and that this conformation might be stabilized by packing in the solid state. This flexibility also contributes to the disorder seen in the inverted pyridine rings in the solid state structure.

Figure 6.8: The structure of hexahemiporphyrazine 6.2 with 35% thermal ellipsoids. Hydrogen atoms on carbon positions have been omitted for clarity.

180

The UV-visible spectrum of the hexamer was measured and compared to those of the bis(6-amino-2-pyridyl)amine starting material as well as diiminoisondoline. As can be seen in the Figure 6.9, the spectrum of the hexahemiporphyrazine differs significantly from those of the two starting materials.

The hexamer does exhibit a two broad absorption bands, with a UV peak at 309 nm and a second with a peak at 384 nm that extends into the visible region.

Although the absorption spectrum is different from that of the two precursors, the general spectroscopic characteristics of this new expanded macrocycle resemble those seen in normal bis-pyridyl hemiporphyrazine.

Figure 6.9: The UV-visible spectra of bis(6-amino-2-pyridyl)amine (6.1), hexahemiporphyrazine (6.2), and diiminoisoindoline in dimethylsulfoxide (DMSO).

181

Conclusions

In conclusion, we have synthesized a new hexameric hemiporphyrazine structure using bis(6-amino-2-pyridyl)amine as a starting material. The precursor can be synthesized readily from 2,6-diaminopyridine and its conjugate acid; purification however is challenging and requires separation by sublimation. The bis(6-amino-2-pyridyl)amine unit exhibits three possible conformations, two seen in the free base and protonated form, and one in the macrocycle backbone.

Reaction of this species with diiminoisoindoline results in the titular hexahemiporphyrazine, which was fully characterized by NMR and X-ray crystallographic methods. As in hexaphyrin, hexahemiporphyrazine shows two inverted rings in its backbone, but unlike many expanded porphyrinoids, hexahemiporphyrazine shows no evidence of cross conjugation across the macrocycle. We are continuing our work on hemiporphryazines through the development of new reagents for their synthesis.

182 CHAPTER VII

BILIAZINE: A RING OPEN PHTHALOCYANINE ANALOG WITH A MESO HYDROGEN BOND

THE TEXT OF THIS CHAPTER IS ADAPTED FROM THE MATERIAL AS IT APPEARS IN: SCHRAGE, B. R.; NEMYKIN, V. N.; ZIEGLER, C. J. BILIAZINE: A RING OPEN PHTHALOCYANINE ANALOG WITH A MESO HYDROGEN BOND. CHEM. COMMUN. 2020, 56 (49), 6628–6631. COPYRIGHT © 2020 THE ROYAL SOCIETY OF CHEMISTRY. REPRODUCED BY PERMISSION OF THE ROYAL SOCIETY OF CHEMISTRY.

DOI: 10.1039/D0CC03060K

Introduction

Synthetic efforts in the field of porphyrin and phthalocyanine analog macrocycles continue apace after more than five decades of progress. In addition to ring expansion and contraction, a common strategy in the development of new macrocycle types is the replacement or alteration of atom positions in the porphyrinic skeleton. Modification can take place at the core sites, at the α or β positions, or at a meso bridging atom site.347,382–385 In the latter case, porphyrins analogs have produced with a variety of heteroatoms at the meso position in recent years; less work has been carried out on phthalocyanine variants.346,386,387 One

183 related modification is to remove an atom position entirely from the skeleton, the most well-known example of which is the vacataporphyrin system developed by

Latos-Grażyński.388–391 In this communication, we present the synthesis and characterization of a hemiporphyrazine-type phthalocyanine analog where a meso nitrogen atom position has been removed. In its place resides a hydrogen bond that links two pyrazole nitrogen atoms and thus completes the macrocycle (Figure

7.1). To the best of our knowledge, this is the first example of a phthalocyanine type macrocycle with a hydrogen atom occupying a meso position. Similar chelate closure chemistry is observed in the cobaloxime family of compounds.

Figure 7.1: The structures of phthalocyanine (left), bis-pyridyl hemiporphyrazine

(middle) and biliazine (H2BlzH, right).

The synthesis of this hemiporphyrazine-type system, which we have dubbed “biliazine” (H2BlzH) (7.2), resulted from our work on isoindoline-based chelates, which includes the hemiporphyrazines,199,203,392 bis(arylimino)isoindolines,157,365,393,394 and the semihemiporphyrazines.156,243,332

We were exploring the chemistry of pyrazole-modifed isoindolines, which can be 184

produced via the reaction of 2-aminopyrazole and diiminoisoindoline (DII, Figure

7.2). Unlike the majority of reactions that occur between arylamines and DII, the reaction halts at the 1:1 stoichiometric product under mild reaction conditions, resulting in compound 7.1. We were able to fully characterize this species 7.2, including via X-ray crystallography, as shown in the Figure 7.2. This compound can be considered a semihemiporphyrazine, a class of isoindoline-based bidentate

156 chelates that we were able to synthesize using Re(CO)3 as a template.

Figure 7.2: The synthesis of 7.1 and the four biliazine compounds described in this report. Top right: The structure of compound 7.1 with 35% thermal ellipsoids.

Hydrogen atoms on carbon positions and atom labels are not shown.

185

Experimental

General Information

All reagents and starting materials were purchased from commercial vendors and used without further purification. DII was synthesized according to previously published procedures.157 Deuterated solvents were purchased from

Cambridge Isotope Laboratories and used as received.

Bruker SHELXTL Software Package (Version 6.1),270 and were solved using direct methods until the final anisotropic full-matrix, least squares refinement of F2 converged. Electronic Supplementary Information (ESI) available Synthetic

186 procedures, spectroscopic data, DFT data, and X-ray parameters. CCDC

1988807-1988811 contain the supplementary crystallographic data for this paper.

Electrochemistry measurements were conducted using a CHI 820D potentiostat in a standard three-electrode configuration. Platinum wire was used as the counter electrode. The working electrode used was a 2 mm diameter platinum disk. A nonaqueous Ag/Ag+ reference electrode was used by immersing silver wire in a degassed DMF solution of 0.01 M AgNO3 /0.1 M tetrabutylammonium hexafluorophosphate (TBAPF6). All potentials were referenced to the ferrocene/ferrocenium couple. The concentration of analyte was

1.0 mM, and the supporting electrolyte was 0.1 M TBAPF6 dissolved in DMF.

Computational Details

The starting geometries of compounds 7.2-7.5 were optimized using a

B3LYP exchange–correlation functional.395 Energy minima in optimized geometry were confirmed by the frequency calculations (absence of the imaginary frequencies). The solvent effect was modeled using the polarized continuum model

327 (PCM). In all calculations, CHCl3 was used as the solvent. In PCM-TDDFT calculation, the first 50 states were calculated. All atoms were modeled using the

6-311G(d)363 basis set. Gaussian 09 software was used in all calculations.330 The

QMForge program was used for molecular orbital analysis in all cases.396

Syntheses

Synthesis of 7.1. 1,3-Diiminoisoindoline (1.00 g, 6.89 mmol) and 3-amino pyrazole (0.57 g, 6.89 mmol) were dissolved in methanol (20 mL) and stirred at

187 room temperature for 1 hour. The resultant solid was filtered, washed with cold methanol, and air dried to give a yellow solid. Crystals suitable for X-ray diffraction were grown by slow evaporation from DMF.

1 7.1: Yield: 1.16 g (79.6%). H NMR (300 MHz, d6 - DMSO): δ = 12.66 (s,

1H), 8.70 (s, 2H), 7.90 (d, J = 7.0 Hz, 1H), 7.76 (d, J = 6.7 Hz, 1H), 7.57 (m, 2H),

13 1 7.47 (s, 1H), 6.64 (s, 1H). C{ H} NMR (125 MHz, d6 - DMSO): δ = 170.2, 163.8,

140.7, 134.4, 131.1, 129.9, 121.2, 100.5. HRMS (ESI-TOF, positive mode) m/z:

+ calcd for C11H10N5 212.0931, found 212.0935 [M+H] .

Synthesis of H2BlzH (7.2). Compound 7.1 (0.10 g, 2.37 mmol) was dissolved in n-butanol (7 mL) and heated to a reflux for 1 hour. The resultant solid was filtered, washed with cold n-butanol, and air dried to give a red solid. Crystals suitable for X-ray diffraction were grown by slow evaporation from DMF.

H2BlzH (7.2): Yield: 0.088 g (91.5%). HRMS (ESI-TOF, positive mode) m/z:

+ calcd for C22H15N9Na 428.1343, found 428.1346 [M+Na] .

Synthesis of MBlzH complexes. A solution of M(OAc)2 hydrate (M = Co,

Cu, Zn) (0.12 mmol), and two equivalents of 7.1 (0.05 g, 0.24 mmol) in MeOH (4.00 mL) were heated until all of the solids dissolved. The mixture was then stirred at room temperature for 1 hour. The resultant precipitate was filtered and washed with cold MeOH. The compounds were isolated as brown (Co(BlzH)(MeOH)2)

(7.4), and red (Cu(BlzH) (7.3) and Zn(BlzH)(MeOH)2) (7.5) solids. Crystals suitable for X-ray diffraction were grown by slow evaporation of MeOH

188

(Co(BlzH)(MeOH)2, and Zn(BlzH)(MeOH)2), and slow evaporation of THF

(Cu(BlzH)).

Cu(BlzH) (7.3): Yield: 0.050 g (91.0%). HRMS (ESI-TOF, positive mode)

+ m/z: calcd for C22H14CuN9 467.0663, found 467.0646 [M+H] .

Co(BlzH)(MeOH)2 (7.4): Yield: 0.049 g (78.3%). HRMS (ESI-TOF, positive

+ mode) m/z: calcd for C22H14CoN9 463.0699, found 463.0681 [M+H] .

1 Zn(BlzH)(MeOH)2 (7.5): Yield: 0.052 g (82.2%). H NMR (500 MHz, d6 -

DMSO): δ = 16.73 (s, NH), 8.17 (m, 2H), 7.99 (m, 1H), 7.77 (m, 2H), 7.72 (m, 4H),

7.53 (d, J = 1.3 Hz, 2H), 6.63 (d, J = 1.3 Hz, 2H), 4.06 (q, J = 5.3 Hz, CH3OH), 3.17

13 1 (d, J = 5.0 Hz, CH3OH). C{ H} NMR (125 MHz, d6 - DMSO): δ = 172.8, 157.8,

156.0, 140.4, 138.7, 131.7, 130.6, 122.4, 121.6, 107.1. HRMS (ESI-TOF, positive

+ mode) m/z: calcd for C22H14N9Zn 468.0658, found 468.0665 [M+H] .

189

7.57

Normalized Intensity Normalized 5

7.55

7.90

4 7.92

3 7.76 MeOH MeOH

7.54 2

7.47

8.70 6.64 water DMSO

1 12.66

0.94

1.86

0.98 0.80

0.75

13 12 11 10 9 8 7 6 5 4 3 2 Chemical Shift (ppm)

Figure 7.3: 1H NMR (500 MHz) of 7.1 in d6-DMSO.

190

MeOH

7.72 7.72 7.71 7.54 6.63 6.63

1.1 8.17

8.16 1.0 MeOH

8.00 0.9

0.8

0.7

0.6

Normalized Intensity Normalized 0.5

0.4

0.3

0.2 water 0.1 16.73

2.19 1.94 2.07

17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 Chemical Shift (ppm)

Figure 7.4: 1H NMR (500 MHz) of 7.5 in d6-DMSO.

191

Table 7.1: X-ray crystal data and structure parameters for compounds 7.1 and 7.2.

Compound 7.1 7.2 Empirical formula C11H9N5 C22H15N9 Formula weight 211.23 405.43 Crystal system Monoclinic Monoclinic Space group C2/c P21/n a/ Å 11.6141(5) 4.579(6) b/ Å 12.7038(5) 17.112(13) c/ Å 13.4922(6) 23.460(16) α(°) 90 92 β(°) 99.080(3) 93.02(8) γ(°) 90 90 Volume (Å3) 1965.74(15) 1836(3) Z 8 4 Dc (Mg/m3) 1.427 1.467 µ (mm-1) 0.757 0.772 F(000) 880 840 reflns collected 6403 11216 indep. reflns 1687 3125 GOF on F2 1.046 1.015 2 R1 (on Fo , I > 0.0461 0.0674 2σ(I)) 2 wR2 (on Fo , I > 0.1266 0.1725 2σ(I)) R1 (all data) 0.0530 0.1127 wR2 (all data) 0.1336 0.1976

192

Table 7.2: X-ray crystal data and structure parameters for compounds 7.3-7.5.

Compound 7.3 7.4 7.5 Empirical formula C22H13CuN9 C24H21CoN9O2 C24H21N9O2Zn Formula weight 466.95 526.43 532.87 Crystal system Monoclinic Monoclinic Monoclinic Space group P21/n P21/c P21/c a/ Å 4.6467(4) 6.8635(3) 6.8862(4) b/ Å 16.7271(17) 23.0498(13) 23.1744(14) c/ Å 23.289(2) 14.1576(8) 14.2313(8) α(°) 90 90 90 β(°) 92.885(7) 102.251(3) 102.434(4) γ(°) 90 90 90 Volume (Å3) 1807.9(3) 2188.8(2) 2217.8(2) Z 4 4 4 Dc (Mg/m3) 1.716 1.598 1.596 µ (mm-1) 1.994 6.531 1.153 F(000) 948 1084 1096 reflns collected 8721 11209 21335 indep. reflns 2876 3325 5523 GOF on F2 1.008 1.050 1.048 2 R1 (on Fo , I > 0.0490 0.0468 0.0511 2σ(I)) 2 wR2 (on Fo , I > 0.1239 0.1159 0.1083 2σ(I)) R1 (all data) 0.0851 0.0590 0.0910 wR2 (all data) 0.1454 0.1228 0.1229

193

Results and Discussion

We investigated the metal chemistry of compound 7.1, reacting it with

Co(II), Cu(II), and Zn(II) salts. In all cases, rather than remaining as the bidentate chelate, compound 7.1 reacted with metal acetates in refluxing methanol solution to form the tetradentate biliazine chelate with the metal in the central pore

(M(BlzH)(MeOH)2, M = Co(II) (7.4), Zn(II) (7.5); M(BlzH), M = Cu(II) (7.3)). It is unclear if the metal is acting as a template in these reactions, as the free ligand

H2BlzH can be produced upon heating of 7.1 in refluxing butanol solution. All four compounds can be isolated as crystalline solids, and we were able to structurally elucidate all of them using single crystal X-ray methods (Figure 7.5). The chelate is comprised of two isoindoline units and two pyrazole groups bridged by three nitrogen atom positions. The structure can be characterized as similar to that of a hemiporphyrazine, however it can be characterized as having an AABB type structure rather than the typical ABAB configuration seen in normal hemiporphyrazines. The free base has two ionizable protons on nitrogen atom positions in the core of the ring, and all three metal derivatives have the metal ion in the +2 oxidation state, resulting in neutral complexes. The chelates are planar, with very small average deviations from the plane of the four interior nitrogen atom positions (~1.3-6.6°).

194

Figure 7.5: The structures of H2BlzH (7.2), Cu(BlzH) (7.3), Co(BlzH)(MeOH)2 (7.4), and Zn(BlzH)(MeOH)2 (7.5) with 35% thermal ellipsoids. Hydrogen atoms on carbon and oxygen positions have been omitted for clarity.

The free base 7.2 and the three metal derivatives 7.3-7.5 exhibit similar structural parameters with regard to the degree of electronic delocalization and the strength of the meso hydrogen bond. The degree of electronic delocalization can be discerned from the bond lengths in the ring skeleton of biliazine and its metal complexes. In the chelate ring localized on the isoindoline side of the macrocycles, 195 the C-N bonds lengths are indicative of electronic delocalization (~1.33-1.36 Å), however in the terminal C-N bonds of this fragment (to the bridging nitrogen atom adjacent to the pyrazoles), there is clear double bond character (~1.27-1.30 Å).

Thus, while there is clear conjugation across the macrocycle, there is some degree of polyene (inequivalent bonds) rather than full cyanine (equivalent bonds) character. All four compounds exhibit the meso hydrogen bond, although the length does change depending on the identity of the metal ion. In the freebase, the N···H-N length is ~2.8 Å, which is quite strong for a nitrogen donor-acceptor hydrogen bond. Interestingly, insertion of the metal ions reduces the heteroatom distance to ~2.5-2.6 Å in the cobalt and copper complexes, and to ~2.7 Å in the zinc compound, which also strengthen the hydrogen bond. In all cases, at least part of this reduction can be attributed to the metal contracting the size of the core of the ligand. The core size reduction can be measured in by the distance between the internal nitrogen atom positions. For example, in Co(BlzH)(MeOH)2 7.4 the three interior non-H bond distances measure 2.70, 2.74 and 2.69 Å, while the same distances in the free base 7.2 are 2.78, 2.85, and 2.79 Å.

The solubility of the biliazine compounds can be limited; the freebase is sparingly soluble. However, we can readily investigate the NMR spectra of

Zn(BlzH) 7.5 due to the solubilizing effect of the axial ligands. The C-H resonances show distinctive changes versus those of the precursor 7.1, and reflect the symmetry of the molecule. The isoindoline resonances reveal an ABCD type splitting pattern, and we observe two doublet resonances for the pyrazole C-H

196 positions. The meso hydrogen bond appears at a downfield resonance of 16.73 ppm, which is clearly indicative of the strength of the N···H-N hydrogen bond. The homonuclear N···H-N hydrogen bonds in 1,8-bis(dimethylamino)naphthalene

(Proton-sponge) derivatives display similar hydrogen bonding strengths, with the

NH proton resonance show up on the range of 15-20 ppm.397 There is little evidence for aromaticity in this hydrogen bond closed macrocycle, as the proton resonances on the carbon positions show reveal no global ring current effects.

All new compounds are strongly absorbing in the visible region, and the UV visible spectra of H2BlzH 7.2 and the MBlzH 7.3-7.5 complexes are shown in

Figure 7.6. The spectra of these four compounds lack clear Soret-type transitions; instead lower intensity (ε ≈ 1.5 x 104) bands between 380 and 580 nm are observed. We do not observe clear Soret like behavior in the optical spectra of new compounds, which might be indicative of the lack of conjugation in this structural motif. In agreement with this hypothesis, the MCD spectrum of

Zn(BlzH)(MeOH)2 7.5 consists of very weak Faraday B-terms (Figure 7.7).

197

Figure 7.6: The UV-visible spectra of H2BlzH and its metal complexes in DMF solution.

198

ZnBlzH 2.0 UV-Vis 1.5

Abs. 1.0 0.5 0.0 0.5 MCD 0.0 -0.5 -1.0 300 350 400 450 500 550 600

MCD Intensity (MDegs) MCD Intensity Wavelength (nm)

Figure 7.7: UV−Vis and MCD spectra of 7.5.

We also probed the electrochemistry of free base biliazine and its metal complexes (Figure 7.8). As expected, the free base and zinc complex exhibit similar properties, with irreversible reductions below 1 V vs. Fc/Fc+, and a single irreversible oxidation near 400 mV. The Cu(BlzH) 7.3 species exhibits reductions similar to that of Zn(BlzH)(MeOH)2 7.5 but the voltammogram exhibits additional features at -1.55 V that could be attributed to a copper center. In contrast, the cobalt complex Co(BlzH)(MeOH)2 7.4 shows clear reversible redox processes that can be attributed to a Co(II)/Co(I) reduction at -1.11 V and a Co(II)/Co(III) oxidation at 0.063 V.

199

Figure 7.8: Cyclic voltammograms in DMF/0.1 TBAPF6.

In order to probe the electronic structures of the biliazine free base and the three metal complexes presented in this report, we carried out DFT and TDDFT calculations using B3LYP exchange correlation functional. Figure 7.9 shows the energy levels of the frontier orbitals for the free base and M(BlzH) complexes. The

LUMO and LUMO+1 energy levels have a higher degree of isolation from other unoccupied orbitals while the HOMO energies have a lesser degree of segregation from the other occupied orbitals. The HOMO-LUMO gap is at ~2.86 eV for the free

200 base biliazine, and range between 2.68-2.80 eV for the M(BlzH) complexes. The

DFT-predicted frontier orbitals are shown in Figure 7.10. The HOMOs of the

M(Blz)s show an asymmetric distribution of orbitals on one half of the molecule which clearly results from the asymmetry of the hydrogen bond. The LUMOs are more symmetric with orbitals generally on the isoindole moieties, and pyrrole nitrogen atoms. Regardless, when compared to the frontier molecular orbitals of phthalocyanines, hemiporphyrazines, and porphyrins, the biliazine compounds display a higher degree of asymmetric orbital distribution.167,398–402 The experimental and B3LYP TDDFT-predicted spectra for the biliazine compounds can be found in Figure 7.11. The lower energy transitions around ~500-550 nm can be mainly classified as HOMO-LUMO transitions. The higher energy transitions around ~425-450 nm are broadly attributed to a mixture of HOMO-1-

LUMO, HOMO-2-LUMO, and HOMO-LUMO+1 transitions. The experimental and theoretical spectra have similar general features with peaks around 450 and 550 nm. However, it is clear the experimental data is more complex and displays splitting of the transitions that we do not observe in our calculations. These can be at least in part, attributed to the usual vibronic transitions that are characteristic for phthalocyanine-like compounds.

201

Figure 7.9: B3LYP α-spin and β-spin relative energies of the frontier orbitals for compounds 7.2-7.5.

202

Figure 7.10: B3LYP DFT-predicted frontier orbitals for compounds 7.2-7.5.

203

Figure 7.11: Experimental and B3LYP TDDFT-predicted spectra for compounds

7.2-7.5.

204

Conclusions

In conclusion, we have synthesized a new phthalocyanine analog where the ring is effectively closed by a hydrogen bond. The biliazine ring can be condensed as the free base from a singly pyrazole substituted diiminiosoindoline, or can be produced in the presence of a metal ion to afford the metalated adducts under templating conditions. X-ray crystallography and NMR spectroscopy show the presence of a strong hydrogen bond bridging the two halves of the molecule, although spectroscopy does not support the presence of a strong diatropic ring current in these compounds. We are continuing to explore this family of compounds, including expanding the metal chemistry of this chelate.

205 CHAPTER VIII

SUBBILIAZINE: A CONTRACTED PHTHALOCYANINE ANALOG

THE TEXT OF THIS CHAPTER IS ADAPTED FROM THE MATERIAL AS IT APPEARS IN: SCHRAGE, B. R.; NEMYKIN, V. N.; ZIEGLER, C. J. SUBBILIAZINE: A CONTRACTED PHTHALOCYANINE ANALOG. ORG. LETT. 2021, 23 (3), 1076–1080. COPYRIGHT © 2021 AMERICAN CHEMICAL SOCIETY.

DOI: 10.1021/ACS.ORGLETT.0C04291

Introduction

Modified phthalocyanines have been heavily investigated for both their fundamental properties as well as for applications in functional materials.170,403–408

Contracted phthalocyanines, in particular the subphthalocyanine (subPc) class of compounds, comprise some of the most studied of the phthalocyanine analogs.144,171,180,409–412 Work on these compounds has spurred synthetic efforts toward other contracted ring systems, including subporphyrin, subporphyrazine, and benzosubporphyrin.173,187,413 Recently we reported the synthesis of a new hemiporphyrazine-type phthalocyanine analog, which we have named biliazine.414

206

This tetradentate chelate mimics the structure of phthalocyanine, but has a hydrogen bond replacing one of the meso nitrogen positions. In this report, we present a contracted biliazine, or subbiliazine (subBlzH) (Figure 8.1), where a boron is used to contracts a bis(pyrazolylimino)isoindoline to form a hydrogen bond as a meso linkage. These compounds are structurally similar to the subphthalocyanines, and we have produced both pyrazole and indazole variants.

As in subphthalocyanine, contraction by boron forces the chelate to adopt a nonplanar conformation. This reaction produced the 2:1 pyazole:isoindoline adduct

8.1, and furthermore could be extended to indazole to afford the -expanded analog 8.2.

Figure 8.1. The structures of phthalocyanine, and biliazine (top), subphthalocyanine, and subbiliazines (bottom). 207

Experimental

General Information

All reagents and starting materials were purchased from commercial vendors and used without further purification. Deuterated solvents were purchased from Cambridge Isotope Laboratories and used as received.

NMR spectra were recorded on 300 MHz and 500 MHz spectrometers and chemical shifts were given in ppm relative to residual solvent resonances (1H NMR and 13C NMR spectra). 19F NMR spectra were referenced and corrected to an external reference trifluoro acetic acid (TFA) standard (-76.55 ppm). High- resolution mass spectrometry experiments were performed on a Bruker MicroTOF-

III and MicroTOF-qIII instruments. Infrared spectra were collected on Thermo

Scientific Nicolet iS5 that was equipped with an iD5 ATR. UV-visible spectra were recorded on a Shimadzu UV-2600 UV-Visible spectrophotometer. Fluorescence emission data in solution were recorded on a Horiba Jobin-Yvon FluoroMax-4 fluorescence spectrophotometer using fluorescein in 0.1 M NaOH as a standard.

All slit widths were held constant at 2 nm. The quantum yields in solution were

2 퐺푟푎푑푋 휂푋 calculated using the following equation:Φ푋 = Φ푆푇 ( ) ∗ ( 2 ); ηST = 1.333, ΦST 퐺푟푎푑푆푇 휂푆푇

= 0.79; ηX = 1.445 (8.1BF and 8.2BF) and 1.431 (8.3), and Grad the gradient from the plot of integrated fluorescence intensity vs absorbance.415

X-ray intensity data were measured on a Bruker CCD-based and

PHOTON II CPAD-based diffractometer with dual Cu/Mo ImuS microfocus optics

(Cu Kα radiation, λ = 1.54178 Å, Mo Kα radiation, λ =0.71073 Å). Crystals were

208 mounted on a cryoloop using Paratone oil and placed under a steam of nitrogen at 100 K (Oxford Cryosystems). The detector was placed at a distance of 5.00 cm from the crystal. The data were corrected for absorption with the SADABS program. The structures were refined using the Bruker SHELXTL Software

Package (Version 6.1),270 and were solved using direct methods until the final anisotropic full-matrix, least squares refinement of F2 converged. The disordered toluene solvent in structure 8.1BF was squeezed out by PLATON.416

1.0 mM, and the supporting electrolyte was 0.1 M TBAPF6 dissolved in DMF.

Computational Details

The starting geometries of compounds 8.1, 8.2, 8.1BF, and 8.2BF were optimized using a B3LYP exchange–correlation functional.395 Energy minima in optimized geometry were confirmed by the frequency calculations (absence of the imaginary frequencies). The solvent effect was modeled using the polarized continuum model (PCM).327 In all calculations, DMF was used as the solvent. In

PCM-TDDFT calculation, the first 50 states were calculated. All atoms were

209 modeled using the 6-311G(d)363 basis set. Gaussian 09 software was used in all calculations.330 The QMForge program was used for molecular orbital analysis in all cases.396

Syntheses

Synthesis of 8.1 and 8.2. The procedure for generating 8.1 is the same as

8.2 except 1H-indazole-3-amine (2.08 g, 15.61 mmol) was used in

8.2. Phthalonitrile (1.00 g, 7.80 mmol) and 3-amino pyrazole (1.30 g, 15.61 mmol were mixed together and heated on a sand bath at 180 °C for 1 hour. The residue was cooled to room temperature, and dissolved in a minimal amount of DMF. DI water was added to induce precipitation of the product. The resultant solid was filtered and air dried to give a yellow powder. Crystals of 8.1 and 8.2 suitable for

X-ray diffraction were grown from slow evaporation from DMF.

−1 −1 8.1: Yield: 1.56 g (72%). MP: 240-247 °C. IR: 3132 cm (νNH), 1632 cm

1 (νCN(imine)). H NMR (300 MHz, d6 - DMSO): δ = 12.79 (s, 2H, NH), 11.91 (s, 1H,

NH), 7.93 (m, 2H, H on isoindoline), 7.77 (s, 2H, H on pyrazole), 7.68 (m, 2H, H

13 1 on isoindoline), 6.33 (s, 2H, H on pyrazole). C{ H} NMR (125 MHz, d6 - DMSO):

δ = 155.6, 148.3, 134.9, 131.4, 129.8, 121.8, 102.6. HRMS (ESI-TOF, positive

+ mode) m/z: calcd for C14H12N7 278.1149, found 278.1151 [M+H] .

−1 −1 8.2: Yield: 2.24 g (76%). MP: 307-312 °C. IR: 3183 cm (νNH), 1635 cm

1 (νCN(imine)). H NMR (300 MHz, d6 - DMSO): δ = 13.13 (s, 2H, NH), 12.33 (s, 1H,

NH), 8.13 (m, 2H, H on isoindoline), 7.97 (d, J = 8.20 Hz, 2H, H on indazole), 7.78

(m, 2H, H on isoindoline), 7.57 (d, J = 8.49 Hz, 2H, H on indazole), 7.46 (t, J = 7.90

210

Hz, 2H, H on indazole), 7.24 (t, J = 7.90 Hz, 2H, H on indazole). 13C{1H} NMR

(125 MHz, d6 - DMSO): δ = 149.4, 148.9, 140.9, 134.8, 134.0, 133.9, 131.6, 127.0,

122.1, 120.7, 119.9, 119.8, 110.4. HRMS (ESI-TOF, positive mode) m/z: calcd for

+ C22H16N7 378.1462, found 378.1460 [M+H] .

Synthesis of 8.1BF and 8.2BF. The procedure for generating 8.1BF is the same as 8.2BF except ligand 8.2 (1.36 g, 3.6 mmol) was used in 8.2BF and dry

DMF was used as the solvent. Triethylamine (10 mL, 36.1 mmol), and BF3·OEt2

(3.2 mL, 25.2 mmol) were added to ligand 8.1 (1.00 g, 3.6 mmol) in dry toluene (15 mL). The reaction was refluxed overnight under N2. The solution was then cooled to room temperature and DI water was added to induce precipitation of the product.

The crude materials was recrystallized in toluene to yield crystals of pure material that were filtered and rinsed with cold toluene, to give a yellow solid (red solid for

8.2BF). Crystals of 8.1BF and 8.2BF suitable for X-ray diffraction were grown from toluene.

−1 1 8.1BF: Yield: 0.59 g (54%). IR: 1601 cm (νCN(imine)). H NMR (300 MHz,

CDCl3): δ = 12.53 (s, NH), 7.96 (m, 2H, H on isoindoline), 7.56-7.53 (m, 4H, H on

13 1 isoindoline and pyrazole), 6.32 (s, 2H, H on pyrazole). C{ H} NMR (125 MHz, d6

- DMSO): δ = 155.8, 151.0, 135.3, 131.1, 129.7, 128.9, 122.3, 122.1, 102.5. 19F

(282 MHz, CDCl3): δ = -149.93. HRMS (ESI-TOF, negative mode) m/z: calcd for

- C14H8BFN7 304.0924, found 304.0918 [M-H] .

−1 1 8.2BF: Yield: 0.98 g (67%). IR: 1569 cm (νCN(imine)). H NMR (300 MHz,

CDCl3): δ = 13.87 (s, NH), 8.24 (m, 2H, H on isoindoline), 8.19 (d, J = 8.25 Hz, 2H,

211

H on indazole), 7.74 (m, 2H, H on isoindoline), 7.63 (d, J = 8.46 Hz, 2H, H on indazole), 7.50 (t, J = 7.25 Hz, 2H, H on indazole), 7.29 (t, J = 7.25 Hz, 2H, H on

13 1 indazole). C{ H} NMR (125 MHz, CDCl3): δ = 151.2, 150.5, 141.0, 134.9, 130.9,

126.6, 122.6, 122.3, 121.1, 120.5, 120.4, 117.6, 116.0, 110.1. 19F (282 MHz,

CDCl3): δ = -149.22. HRMS (ESI-TOF, negative mode) m/z: calcd for C22H12BFN7

404.1237, found 404. 1234 [M-H]-.

Synthesis of 8.3. Ligand 8.2 (0.25 g, 0.66 mmol) was combined with

BF3·OEt2 (1 mL) in a round bottom flask (no base) and heated to 100 °C for 1 hr.

During this time the initial deep red solution eventually yielded a tan precipitate.

The solution was brought to room temperature and the solids were filtered and washed with water. The crude material was recrystallized in acetone to yield crystals of pure material that were filtered and rinsed with cold acetone. Crystals suitable for X-ray diffraction were grown from acetone.

−1 1 8.3: Yield: 0.11 g (53%). IR: 1723 cm (νCO). H NMR (300 MHz, d6 -

DMSO): δ = 8.13 (d, J = 8.25 Hz, 1H), 8.08 (d, J = 7.05 Hz, 1H), 7.93-7.75 (m, 4H),

7.64 (d, J = 8.66 Hz, 1H, H on indazole), 7.45 (t, J = 7.45 Hz, 1H, H on

13 1 indazole). C{ H} NMR (125 MHz, d6 - DMSO): δ = 170.5, 158.8, 144.9, 141.0,

135.4, 134.3, 133.6, 132.4, 131.2, 123.8, 123.6, 122.8, 121.4, 115.3, 111.6. 19F

(282 MHz, CDCl3): δ = -144.65, 144.59. HRMS (ESI-TOF, negative mode) m/z:

- calcd for C15H8BF2N4O 309.0765, found 309.0757 [M-H] .

212

Figure 8.2: 1H NMR (300 MHz) of 8.1 in d6-DMSO.

213

Figure 8.3: 1H NMR (300 MHz) of 8.2 in d6-DMSO.

214

1 Figure 8.4: H NMR (300 MHz) of 8.1BF in CDCl3.

215

1 Figure 8.5: H NMR (300 MHz) of 8.2BF in CDCl3.

216

Figure 8.6: 1H NMR (300 MHz) of 8.3 in d6-DMSO.

217

Table 8.1: X-ray crystal data and structure parameters for compounds 8.1, 8.2,

8.1BF, and 8.3.

Compound 8.1 8.2 8.1BF 8.3 CCDC 2052188 2052189 2052190 2052191 Empirical C31H29N15O C22H15N7 C20H24BFN8 C15H9BF2N4O formula Formula 627.69 377.41 406.28 310.07 weight Crystal system Monoclinic Monoclinic Monoclinic Monoclinic Space group P21/c P21/c Pn C2/c a/ Å 10.2918(2) 14.1029(7) 9.327(3) 14.7987(12) b/ Å 18.5703(3) 23.6368(13) 22.100(5) 14.1332(14) c/ Å 15.8997(2) 11.1420(6) 24.878(6) 6.9968(7) α(°) 90 90 90 60 β(°) 98.5810(10) 93.727(3) 92.46(2) 116.300(3) γ(°) 90 90 90 90 Volume (Å3) 3004.76(9) 3706.3(3) 5123(3) 1311.9(2) Z 4 8 8 4 Dc (Mg/m3) 1.388 1.353 1.054 1.574 µ (mm-1) 0.759 0.690 0.072 0.122 F(000) 1312 1568 1712 632 reflns collected 18150 23861 48943 11909 indep. reflns 5250 6300 18565 1629 GOF on F2 1.033 1.011 1.072 1.225 2 R1 (on Fo , I > 0.0543 0.0416 0.0709 0.0556 2σ(I)) 2 wR2 (on Fo , I 0.1483 0.0941 0.1819 0.1206 > 2σ(I)) R1 (all data) 0.0660 0.0657 0.1073 0.0674 wR2 (all data) 0.1583 0.1059 0.1991 0.1238

218

Results and Discussion

We recently reported on the reaction between diiminoisoindoline and aminopyrazole.414 Unlike with many arylindolines, this reaction halts at the 1:1 product rather than the typical 2:1 species. In order to generate the bis(pyrazolylimino)isoindoline, we used Siegl type conditions starting with phthalonitrile, as shown in Scheme 8.1 (72% and 76% for 8.1 and 8.2).417

Previously, the chemistry of boron and bis(pyridylimino)isoindoline has been explored by Bender,229 but in the case of pyrazole, we surmised that a ring contracted form of biliazine (a subbiliazine) would be formed upon boron insertion.

Reaction of both 8.1 and 8.2 with BF3 and base (triethylamine) under dry conditions resulted in the formation of 8.1BF (54%) and 8.2BF (67%) (Scheme 8.1). These two compounds were fully characterized, and the structure of 8.1BF was elucidated by X-ray crystallography (Figure 8.7). Subbiliazine 2BF also formed a crystalline product, but we were unable to fully elucidate its structure by X-ray diffraction, although connectivity could be discerned from a partial solution (Figure

8.7).

219

Scheme 8.1. The synthesis of 8.1 and 8.2, the subbiliazine compounds (8.1BF and 8.2BF), and the hydrolysis product (8.3).

220

Figure 8.7: The structures of free bases 8.1, and 8.2, subbiliazine 8.1BF, hydrolysis product 8.3, and partial structure elucidation of 8.2BF with 35% thermal ellipsoids. Hydrogen atoms on carbon positions have been omitted for clarity.

As in the biliazines, the structure of 8.1BF shows a macrocycle like structure with a hydrogen bond where a nitrogen would be in a meso position. This hydrogen bond is less strong than seen in the biliazines, and we observe a N-N

221 distance of ~3.2-3.4 Å, which is indicative of a weak hydrogen bond. The presence of this hydrogen bond is confirmed in the 1H NMR spectrum, which reveals a resonance at 12.53 ppm. The B-F bond distances in 8.1BF are ~1.39-1.44 Å, and are comparable to the axial B-F distances seen in fluorinated subPcs (~1.38-1.41

Å).418–420 In many subPcs, the isoindole N-B distances range from ~1.47-1.51

Å.418–421 Compound 8.1BF has slightly longer bond lengths, with the isoindole N-

B distances of ~1.50-1.54 Å, and the pyrazole N-B distances ranging from ~1.49-

1.56 Å. The structure of 8.1BF also has a bowl shape, similar to that of Cl-subPc.

The depth of the bowl can be measured by defining the plane of the six outer carbon atoms and measuring its distance to the central boron.421 The bowl depth of 8.1BF is ~1.25 Å, making the molecule twice as shallow in comparison to Cl- subPc (~2.45 Å). The coordination of the bis(pyrazolylimino)isoindoline ligand to boron results in the deformation of a typically planar chelate, and the open nature of the ligand allows for a less constrained coordination sphere. This deviation from planarity can similarly be seen in our recent work in which bis(pyridylimino)isoindoline (BPI) was coordinated to Re(CO)3 in a facial coordination mode, resulting in a similar distortion in ligand planarity.230

When ligands 8.1 and 8.2 are reacted under non-anhydrous conditions in the presence of a large excess of BF3 and no base, we do not observe the formation of 8.1BF and 8.2BF. We surmise that due to the strain of the bowl-like shape, the two compounds are predisposed to hydrolysis. Two different products are observed. When ligand 8.1 is subjected to similar reaction conditions,

222 hydrolysis produces phthalimide, as confirmed by X-ray crystallography and product analysis. For ligand 8.2, a new product forms from the decomposition reaction mixture. The solution yields a tan crystalline solid that rapidly forms with additional water, which was identified as hydrolysis product 8.3 (53%) (Scheme

8.1). Compound 8.3 is a dibenzo aza-BODIPY analog,422–429 comprised of an indazole and an iminooxoisoindoline bound to a BF2 unit. Compound 8.3 can also be considered as having a semihemiporphyrazine-type structure, which we

156 reported on previously in several Re(CO)3 compounds. This compound was structurally elucidated, and is shown in Figure 8.7. The ligand is bound to the boron in a bidentate fashion and has rigid planarity in the solid state. The CO bond length is ~1.22 Å, which is indicative of a carbonyl rather than an imine.

Compounds 8.1BF and 8.2BF as well as the free base ligands absorb strongly in the UV and visible regions of the spectrum, as shown in Figure 8.8 (free bases in Figure 8.9). The free base ligands 8.1 and 8.2 exhibit π to π* transitions with vibronic fine structure similar to those seen in other substituted isoindoline systems.430–432 Insertion of boron in both ligands results in an appearance of a red- shifted shoulder in the spectra. Extinction coefficients range between 104 to 2.5 x

104 M-1cm-1 for these compounds. The excitation spectra of compounds 8.1BF and

8.2BF were collected (Figure 8.10) and correlate well with the absorption spectra, varying in intensities of the bands.

223

Figure 8.8: UV-visible spectra for compounds 8.1BF and 8.2BF in CHCl3.

224

Figure 8.9: UV-visible spectra for compounds 8.1 and 8.2 in DMF.

225

Figure 8.10: Absorption and excitation spectra for compounds 8.1BF and 8.2BF in CHCl3.

226

While the free base compounds 8.1 and 8.2 are not emissive, 8.1BF and

8.2BF do fluoresce as also shown in Figure 8.11. The quantum yields of emission for these two compounds are 16 and 22% respectively. It is notable that the emission bands are not mirror images of the absorbance bands, which implies possible vibronic progression involvement in the excited state, as seen in the

BOPHY fluorophore.433 Decomposition product 8.3 is also strongly absorbing and emits with a high quantum yield (74%), as can be seen in Figure 8.12 (UV-visible spectrum in Figure 8.13). Notably, we observed large Stokes shifts in emission for all three compounds, with values of 137 and 233 nm for compounds 8.1BF and

8.2BF respectively when exited at their maximum absorbance. Dibenzo aza-

BODIPY analog 8.3 also shows a significant Stokes shift value of 102 nm.

Photophysical parameters are summarized in the supplementary information.

227

1.0 1.0 8.1BF 8.1BF 8.2BF 0.8 0.8 8.2BF

0.6 0.6

0.4 0.4

0.2 0.2

Normalized Absorbance

Normalized Emission Intensity 0.0 0.0 400 500 600 700 800 Wavelength (nm)

Figure 8.11: The normalized absorption (solid) and emission (dashed) spectra for compound 8.1BF, and 8.2BF in CHCl3.

228

1.0 1.0 8.3 8.3 0.8 0.8

0.6 0.6

0.4 0.4

0.2 0.2

Normalized Absorbance

Normalized Emission Intensity

0.0 0.0 400 500 600 Wavelength, nm

Figure 8.12: The normalized absorption (solid) and emission (dashed) spectra for compound 8.3 in DMF.

229

Figure 8.13: UV-visible spectra for compound 8.3 in DMF.

We also investigated the electrochemistry of the free bases and the subBlzH complexes (Figure 8.14). None of the compounds exhibit cleanly reversible redox events, and two additional irreversible reductions are observed in the BF complexes at ~-0.88V and ~-1.24V for 8.1BF, and ~-0.88V, and ~-1.31V for 8.2BF.

230

8.1 8.2 8.1BF 8.2BF

Current

-0.5 -1.0 -1.5 -2.0 Potential (V vs. Fc/Fc+)

Figure 8.14: Cyclic voltammograms of 8.1, 8.2, 8.1BF, and 8.2BF in DMF/0.1

TBAPF6.

In order to probe the electronic structures of these compounds, DFT and

TDDFT calculations using the B3LYP exchange correlation functional were performed on the two free base ligands 8.1 and 8.2 and their corresponding subbiliazine complexes 8.1BF and 8.2BF. Figure 8.15 displays the frontier orbital energy levels for these compounds. For all compounds, the LUMOs are readily isolated from other unoccupied orbitals. The HOMO-LUMO gap for pyrazole- based ligand 8.1 is ~3.67 eV, and decreases to ~3.23 eV with the expansion of the

π-system in the indazole-based ligand 8.2. Upon coordination with boron, the 231 energy levels of the LUMO stabilizes resulting in energy gaps of ~3.45 and ~2.73 eV for 8.1BF and 8.2BF respectively.

-2

Virtual

-4

E, eV E,

-6

Occupied

8.11 8.22 8.1BF3 8.2BF4

Figure 8.15: Relative energies of the frontier orbitals for compounds 8.1, 8.2,

8.1BF, and 8.2BF.

The DFT-predicted frontier orbitals for the free base ligands and subBlzH complexes are also shown in Figure 8.16. As seen in the biliazine complexes of zinc, cobalt, and copper, the subbiliazines similarly show an asymmetric distribution of orbitals on one half of the molecule. The LUMOs are more evenly distributed with major orbital distribution on the isoindole.

232

Figure 8.16: DFT-predicted frontier orbitals for compounds 8.1, 8.2, 8.1BF and

8.2BF.

233

TDDFT calculations were carried out to ascertain the nature of the transitions in the free bases and subBlzH complexes and can be found in Figure

8.17. The broad absorbances in 8.1 and 8.2 are both comprised of transitions that are a mixture of HOMO- LUMO and HOMO-1-LUMO transitions. The subBlzH complexes have higher energy bands at ~358-401 nm are comprised of a mixture of HOMO-1-LUMO, HOMO-2-LUMO, and HOMO-LUMO+1 transitions.

Additionally, the lower energy band at ~419 and 546 nm are π to π* HOMO-LUMO transitions. We have also carried out TDDFT calculations for compounds 8.1BF and 8.2BF using excited state geometries to model the emission (Figure 8.18).

The calculated fluorescence spectra agree well with the observed experimental data and are supportive of the large Stokes shifts seen in 8.1BF and 8.2BF. The significant change in DFT predicted excited state geometries also correlate well with the large Stokes shifts in 8.1BF and 8.2BF.

234

Figure 8.17: Experimental and B3LYP TDDFT-predicted spectra for compounds

8.1, 8.2, 8.1BF, and 8.2BF.

235

Figure 8.18: Experimental and B3LYP TDDFT-predicted excitation and emission spectra for compounds 8.1BF and 8.2BF.

236

Conclusions

In conclusion, two new bis(arylimino)isoindoline chelates 8.1 and 8.2 were synthesized by classic Siegl type conditions. The insertion of boron into these ligands results in the complexes 8.1BF and 8.2BF, which have a similar bowl- shape to subphthalocyanine. These macrocycles are closed by a weak hydrogen bond, and are considered to be a ring contracted form of biliazine, which we call subbiliazine. When the ligands are reacted with BF3 under non-anhydrous conditions, hydrolysis products can be observed. Ligand 8.2 yields the hydrolysis product 8.3 which can be described as a dibenzo aza-BODIPY analog. The subbiliazines have moderate quantum yields that are comparable to that of Cl- subPc. We are continuing to investigate the chemistry of the biliazine scaffold, as well as further study compound 8.3. We plan to continue investigating the chemistry of 8.3 and related systems by extending our efforts to the pyrazole variant, as well as the imine terminal analogs.

237 CHAPTER IX

SUMMARY

In this dissertation, two main topics were investigated. The first topic examines the electrochemical behavior of ferrocene salts, and their incorporation in redox flow battery design. Later chapters probe the use of DII as a reagent for the synthesis of new macrocycles, chelates, and chromophores.

In chapter II, the synthesis of four new all-ferrocene salts was described.

The ion potentials in these salts are highly dependent on solvent conditions. The solvents water, PC, and DMF were compared. The salts containing sulfonate anions displayed greater reversibility and stability in PC and DMF than the carboxylate salts. Additionally, effects such as hydrogen bonding were found to play an important role in the variation of redox potentials of the salt’s anions.

Chapter III utilizes the 1,1’-bis(sulfonate)ferrocene dianion with sodium cations in a RFB device as the catholyte. Anthraquinone-2,7-disulfonic acid disodium salt was chosen as the anolyte in the battery design. Three aqueous battery tests were performed under varying conditions, resulting in different battery performance outcomes. The first experiment employed neutral conditions, using 1 M NaNO3.

The ferrocene component remained stable, but the basic conditions allowed for the formation of radical anions in the anolyte cell, resulting in irreversible electrochemical behavior. Battery tests under acidic conditions with 2 M acetate 238 buffer and 0.5 M H2SO4 stabilize the anthraquinone component previously degraded under basic conditions. However, the ferrocene salt was prone to nucleophilic attack, and the ferrocene degraded into iron acetate and sulfate salts.

In the future the Ziegler group will investigate the performance of 1,1’- bis(sulfonate)ferrocene salts with various cations. Cations containing organic units such as imidazolium and pyridinium are possible candidates for increasing the stability of ferrocene salts. Additionally, synthesizing a ferrocene sulfonate derivative with more steric bulk may result in a more robust catholyte scaffold.

Octamethylferrocene can be readily sulfonated, and this chemistry has been explored. The methyl groups may protect the iron from nucleophilic attack and enhance battery performance.

Chapters IV-VIII revolve around the chemistry of DII, which was first explored by Elvidge and Linstead. In chapter IV, previously known BAI ligands were complexed to the Re(CO)3 unit. BAIs bind to metals in a meridional coordination mode. However, three Re complexes with BAIs bound in a facial coordination mode were synthesized. The compounds exhibit MLCT transitions in their UV-visible spectra as confirmed by DFT and TDDFT calculations.

In chapter V, the syntheses of several 1,3-diylideneisoindolines were presented. The reaction of DII and four organic CH-acids resulted in highly colored planar chromophores with terminal alkenes. The strong absorbing bands in the

UV and visible range were dependent on the nature of the CH-acid substituent.

239

Chapter VI is about the synthesis of a new hexameric expanded hemiporphyrazine. Bis(6-amino-2-pyridyl)amine was utilized as the starting material, synthesized by the melt reaction of 2,6-diaminopyridine and its hydrochloride salt. Reaction of bis(6-amino-2-pyridyl)amine with DII resulted in the expanded hexahemiporphyrazine macrocycle. The lack of aromaticity of the macrocycle is evident in the 1H NMR spectrum, and the crystal structure shows a non-planar ring with the pyridine rings inverted in the backbone.

Chapters VII and VIII presented the chemistry of the biliazine and subbiliazine systems. Both systems are phthalocyanine and subphthalocyanine analogs that lack a meso atom position and the macrocycles are closed by a hydrogen bond. In chapter VII, a metal-free semihemiporphyrazine analog was synthesized via the reaction of DII with 3-aminopyrazole. Two equivalents of the ligand were condensed either without or in the presence of a metal acetate, resulting in a new chelate (biliazine). 1H NMR spectroscopy and X-ray crystallography indicate the presence of a strong H bond in the meso position of the macrocycle. The UV-visible spectra of the new compounds were compared with DFT and TDDFT calculations, and their electronic structures were probed.

In chapter VIII, the syntheses of two new BAI chelates were presented.

Reaction of the BAIs with BF3 and mild base (triethylamine), resulted in the formation of new boron complexes with similar bowl-shape to SubPcs. The macrocycles are closed by a weak hydrogen bond and are a ring contracted form of biliazine. Hydrolysis of one of the BAIs resulted in a new product which is a

240 dibenzo aza-BODIPY analog. The new compounds were all fluorescent and the

UV-visible spectra of the new subbiliazines were compared with DFT and TDDFT calculations, and their electronic structures were probed.

The synthesis of new molecules containing DII is a rich area of chemistry.

In the future, the closure of the biliazine ring could be attempted. The macrocycle can be closed with either organic linkages, or metal ions, resulting in the possible induction of aromaticity of the system. Additionally, the chemistry of iminoisoindolinone seen in hydrolysis chemistry of chapter VIII can be explored and utilized in a similar fashion to DII. The condensation reaction of DII and CH- acids can be extended to iminoisoindolinone resulting in ylideneisoindolinones, and the semihemiporphyrazine chelates can also be synthesized, and reacted with moieties like BF3 and Re(CO)3.

241 REFERENCES

(1) Bird, L.; Lew, D.; Milligan, M.; Carlini, E. M.; Estanqueiro, A.; Flynn, D.; Gomez-Lazaro, E.; Holttinen, H.; Menemenlis, N.; Orths, A.; et al. Wind and Solar Energy Curtailment: A Review of International Experience. Renew. Sustain. Energy Rev. 2016, 65, 577–586.

(2) Díaz-González, F.; Sumper, A.; Gomis-Bellmunt, O.; Villafáfila-Robles, R. A Review of Energy Storage Technologies for Wind Power Applications. Renew. Sustain. Energy Rev. 2012, 16 (4), 2154–2171.

(3) Saidur, R.; Rahim, N. A.; Islam, M. R.; Solangi, K. H. Environmental Impact of Wind Energy. Renew. Sustain. Energy Rev. 2011, 15 (5), 2423–2430.

(4) Hernandez, R. R.; Easter, S. B.; Murphy-Mariscal, M. L.; Maestre, F. T.; Tavassoli, M.; Allen, E. B.; Barrows, C. W.; Belnap, J.; Ochoa-Hueso, R.; Ravi, S.; et al. Environmental Impacts of Utility-Scale Solar Energy. Renew. Sustain. Energy Rev. 2014, 29, 766–779.

(5) Wang, S.; Wang, S. Impacts of Wind Energy on Environment: A Review. Renew. Sustain. Energy Rev. 2015, 49, 437–443.

(6) Kurzweil, P. Gaston Planté and His Invention of the Lead-Acid Battery-The Genesis of the First Practical Rechargeable Battery. J. Power Sources 2010, 195, 4424–4434.

(7) Li, M.; Lu, J.; Chen, Z.; Amine, K. 30 Years of Lithium-Ion Batteries. Adv. Mater. 2018, 30 (33), 1800561.

(8) Xia, L.; Yu, L.; Hu, D.; Chen, G. Z. Electrolytes for Electrochemical Energy Storage. Mater. Chem. Front. 2017, 1 (4), 584–618.

(9) Weber, A. Z.; Mench, M. M.; Meyers, J. P.; Ross, P. N.; Gostick, J. T.; Liu, Q. Redox Flow Batteries: A Review. J. Appl. Electrochem. 2011, 41 (10), 1137–1164.

242

(10) Leung, P.; Li, X.; De León, C. P.; Berlouis, L.; Low, C. T. J.; Walsh, F. C. Progress in Redox Flow Batteries, Remaining Challenges and Their Applications in Energy Storage. Rsc Adv. 2012, 2 (27), 10125–10156.

(11) Alotto, P.; Guarnieri, M.; Moro, F. Redox Flow Batteries for the Storage of Renewable Energy: A Review. Renew. Sustain. Energy Rev. 2014, 29, 325– 335.

(12) Qi, Z.; Koenig, G. M. Review Article: Flow Battery Systems with Solid Electroactive Materials. J. Vac. Sci. Technol. B 2017, 35 (4), 40801.

(13) Tokuda, N.; Kumamoto, T.; Shigematsu, T.; Deguchi, H.; Ito, T.; Yoshikawa, N.; Hara, T. Development of a Redox Flow Battery System. SEI Tech Rev 1998, 88–94.

(14) Shigematsu, T. The Development and Demonstration Status of Practical Flow Battery Systems. Curr. Opin. Electrochem. 2019, 18, 55–60.

(15) Gubler, L. Membranes and Separators for Redox Flow Batteries. Curr. Opin. Electrochem. 2019, 18, 31–36.

(16) Tanaka, Y.; Editor. Ion Exchange Membranes: Fundamentals and Applications, 2nd Edition.; Elsevier, 2015.

(17) Hagesteijn, K. F. L.; Jiang, S.; Ladewig, B. P. A Review of the Synthesis and Characterization of Anion Exchange Membranes. J. Mater. Sci. 2018, 53 (16), 11131–11150.

(18) Hosseiny, S. S.; Wessling, M. 13 - Ion Exchange Membranes for Vanadium Redox Flow Batteries. In Woodhead Publishing Series in Energy; Basile, A., Nunes, S. P. B. T.-A. M. S. and T. for S. E. and E. A., Eds.; Woodhead Publishing, 2011; pp 413–434.

(19) Luo, T.; Abdu, S.; Wessling, M. Selectivity of Ion Exchange Membranes: A Review. J. Memb. Sci. 2018, 555, 429–454.

(20) Xu, T. Ion Exchange Membranes: State of Their Development and Perspective. J. Memb. Sci. 2005, 263 (1), 1–29.

(21) Lai, Y. Y.; Li, X.; Zhu, Y. Polymeric Active Materials for Redox Flow Battery 243

Application. ACS Appl. Polym. Mater. 2020, 2 (2), 113–128.

(22) Milton, M.; Cheng, Q.; Yang, Y.; Nuckolls, C.; Hernández Sánchez, R.; Sisto, T. J.; Hernandez Sanchez, R.; Sisto, T. J. Molecular Materials for Nonaqueous Flow Batteries with a High Coulombic Efficiency and Stable Cycling. Nano Lett. 2017, 17 (12), 7859–7863.

(23) Montoto, E. C.; Nagarjuna, G.; Moore, J. S.; Rodríguez-López, J. Redox Active Polymers for Non-Aqueous Redox Flow Batteries: Validation of the Size-Exclusion Approach. J. Electrochem. Soc. 2017, 164 (7), A1688– A1694.

(24) Janoschka, T.; Martin, N.; Martin, U.; Friebe, C.; Morgenstern, S.; Hiller, H.; Hager, M. D.; Schubert, U. S. An Aqueous, Polymer-Based Redox-Flow Battery Using Non-Corrosive, Safe, and Low-Cost Materials. Nature 2015, 527 (7576), 78–81.

(25) Zhong, F.; Yang, M.; Ding, M.; Jia, C. Organic Electroactive Molecule-Based Electrolytes for Redox Flow Batteries: Status and Challenges of Molecular Design. Front. Chem. 2020, 8, 451.

(26) Hendriks, K. H.; Robinson, S. G.; Braten, M. N.; Sevov, C. S.; Helms, B. A.; Sigman, M. S.; Minteer, S. D.; Sanford, M. S. High-Performance Oligomeric Catholytes for Effective Macromolecular Separation in Nonaqueous Redox Flow Batteries. ACS Cent. Sci. 2018, 4 (2), 189–196.

(27) Cao, J.; Ding, F.; Chen, H.; Wang, H.; Wang, W.; Chen, Z.; Xu, J. A New Redox-Active Conjugated Polymer Containing Anthraquinone Pendants as Anode Material for Aqueous All-Organic Hybrid-Flow Battery. J. Power Sources 2019, 423, 316–322.

(28) Soloveichik, G. L. Flow Batteries: Current Status and Trends. Chem. Rev. 2015, 115 (20), 11533–11558.

(29) Kear, G.; Shah, A. A.; Walsh, F. C. Development of the All-Vanadium Redox Flow Battery for Energy Storage: A Review of Technological, Financial and Policy Aspects. Int. J. Energy Res. 2012, 36 (11), 1105–1120.

(30) Yuan, X.-Z.; Song, C.; Platt, A.; Zhao, N.; Wang, H.; Li, H.; Fatih, K.; Jang, D. A Review of All-Vanadium Redox Flow Battery Durability: Degradation Mechanisms and Mitigation Strategies. Int. J. Energy Res. 2019, 43 (13), 244

6599–6638.

(31) Lourenssen, K.; Williams, J.; Ahmadpour, F.; Clemmer, R.; Tasnim, S. Vanadium Redox Flow Batteries: A Comprehensive Review. J. Energy Storage 2019, 25, 100844.

(32) Zeng, Y. K.; Zhao, T. S.; An, L.; Zhou, X. L.; Wei, L. A Comparative Study of All-Vanadium and Iron-Chromium Redox Flow Batteries for Large-Scale Energy Storage. J. Power Sources 2015, 300, 438–443.

(33) Thaller, L. H. Electrically Rechargeable Redox Flow Cell., August 22, 1975.

(34) Cunha, Á.; Martins, J.; Rodrigues, N.; Brito, F. Vanadium Redox Flow Batteries: A Technology Review. Int. J. Energy Res. 2014, 39.

(35) Singh, P. Application of Non-Aqueous Solvents to Batteries. J. Power Sources 1984, 11 (1), 135–142.

(36) Gong, K.; Fang, Q.; Gu, S.; Li, S. F. Y.; Yan, Y. Nonaqueous Redox-Flow Batteries: Organic Solvents, Supporting Electrolytes, and Redox Pairs. Energy Environ. Sci. 2015, 8 (12), 3515–3530.

(37) Shin, S.-H.; Yun, S.-H.; Moon, S.-H. A Review of Current Developments in Non-Aqueous Redox Flow Batteries: Characterization of Their Membranes for Design Perspective. RSC Adv. 2013, 3 (24), 9095–9116.

(38) Winsberg, J.; Hagemann, T.; Janoschka, T.; Hager, M. D.; Schubert, U. S. Redox-Flow Batteries: From Metals to Organic Redox-Active Materials. Angew. Chemie Int. Ed. 2017, 56 (3), 686–711.

(39) Chakrabarti, M. H.; Dryfe, R. A. W.; Roberts, E. P. L. Evaluation of Electrolytes for Redox Flow Battery Applications. Electrochim. Acta 2007, 52 (5), 2189–2195.

(40) Mun, J.; Lee, M.-J.; Park, J.-W.; Oh, D.-J.; Lee, D.-Y.; Doo, S.-G. Non- Aqueous Redox Flow Batteries with Nickel and Iron Tris(2,2ʹ-Bipyridine) Complex Electrolyte. Electrochem. Solid-State Lett. 2012, 15 (6), A80.

(41) AES, S.; AA, S.; QH, L.; YD, L.; CW, M.; LT, T. Non-aqueous manganese acetylacetonate electrolyte for redox flow batteries. J. Power Sources 2011, 245

196 (13), 5742–5745.

(42) Liu, Q.; Sleightholme, A. E. S.; Shinkle, A. A.; Li, Y.; Thompson, L. T. Non- Aqueous Vanadium Acetylacetonate Electrolyte for Redox Flow Batteries. Electrochem. commun. 2009, 11 (12), 2312–2315.

(43) Suttil, J. A.; Kucharyson, J. F.; Escalante-Garcia, I. L.; Cabrera, P. J.; James, B. R.; Savinell, R. F.; Sanford, M. S.; Thompson, L. T. Metal Acetylacetonate Complexes for High Energy Density Non-Aqueous Redox Flow Batteries. J. Mater. Chem. A 2015, 3 (15), 7929–7938.

(44) Zhang, D.; Lan, H.; Li, Y. The Application of a Non-Aqueous Bis(Acetylacetone)Ethylenediamine Cobalt Electrolyte in Redox Flow Battery. J. Power Sources 2012, 217, 199–203.

(45) Hogue, R. W.; Toghill, K. E. Metal Coordination Complexes in Nonaqueous Redox Flow Batteries. Curr. Opin. Electrochem. 2019, 18, 37–45.

(46) Chen, H.; Cong, G.; Lu, Y.-C. Recent Progress in Organic Redox Flow Batteries: Active Materials, Electrolytes and Membranes. J. Energy Chem. 2018, 27 (5), 1304–1325.

(47) Luo, J.; Hu, B.; Hu, M.; Zhao, Y.; Liu, T. L. Status and Prospects of Organic Redox Flow Batteries toward Sustainable Energy Storage. ACS Energy Lett. 2019, 4 (9), 2220–2240.

(48) Yuan, J.; Zhang, C.; Zhen, Y.; Zhao, Y.; Li, Y. Enhancing the Performance of an All-Organic Non-Aqueous Redox Flow Battery. J. Power Sources 2019, 443, 227283.

(49) Hagemann, T.; Winsberg, J.; Grube, M.; Nischang, I.; Janoschka, T.; Martin, N.; Hager, M. D.; Schubert, U. S. An Aqueous All-Organic Redox-Flow Battery Employing a (2,2,6,6-Tetramethylpiperidin-1-Yl)Oxyl-Containing Polymer as Catholyte and Dimethyl Viologen Dichloride as Anolyte. J. Power Sources 2018, 378, 546–554.

(50) Singh, V.; Kim, S.; Kang, J.; Byon, H. R. Aqueous Organic Redox Flow Batteries. Nano Res. 2019, 12 (9), 1988–2001.

(51) Breslin, D. T.; Schuster, G. B. Anthraquinone Photonucleases: Mechanisms for GG-Selective and Nonselective Cleavage of Double-Stranded DNA. J. 246

Am. Chem. Soc. 1996, 118 (10), 2311–2319.

(52) Liu, Y.; Goulet, M.-A.; Tong, L.; Liu, Y.; Ji, Y.; Wu, L.; Gordon, R. G.; Aziz, M. J.; Yang, Z.; Xu, T. A Long-Lifetime All-Organic Aqueous Flow Battery Utilizing TMAP-TEMPO Radical. Chem 2019, 5 (7), 1861–1870.

(53) Liu, T.; Wei, X.; Nie, Z.; Sprenkle, V.; Wang, W. A Total Organic Aqueous Redox Flow Battery Employing a Low Cost and Sustainable Methyl Viologen Anolyte and 4-HO-TEMPO Catholyte. Adv. Energy Mater. 2016, 6 (3), 1501449.

(54) Hu, B.; Tang, Y.; Luo, J.; Grove, G.; Guo, Y.; Liu, T. L. Improved Radical Stability of Viologen Anolytes in Aqueous Organic Redox Flow Batteries. Chem. Commun. (Cambridge, United Kingdom) 2018, 54 (50), 6871–6874.

(55) Lantz, A. W.; Shavalier, S. A.; Schroeder, W.; Rasmussen, P. G. Evaluation of an Aqueous Biphenol- and Anthraquinone-Based Electrolyte Redox Flow Battery. ACS Appl. Energy Mater. 2019, 2 (11), 7893–7902.

(56) Lee, W.; Permatasari, A.; Kwon, Y. Neutral PH Aqueous Redox Flow Batteries Using an Anthraquinone-Ferrocyanide Redox Couple. J. Mater. Chem. C 2020, 8 (17), 5727–5731.

(57) Lee, W.; Permatasari, A.; Kwon, B. W.; Kwon, Y. Performance Evaluation of Aqueous Organic Redox Flow Battery Using Anthraquinone-2,7-Disulfonic Acid Disodium Salt and Potassium Iodide Redox Couple. Chem. Eng. J. 2019, 358, 1438–1445.

(58) Gerhardt, M. R.; Tong, L.; Gómez-Bombarelli, R.; Chen, Q.; Marshak, M. P.; Galvin, C. J.; Aspuru-Guzik, A.; Gordon, R. G.; Aziz, M. J. Anthraquinone Derivatives in Aqueous Flow Batteries. Adv. Energy Mater. 2017, 7 (8), 1601488.

(59) Huang, J.; Yang, Z.; Vijayakumar, M.; Duan, W.; Hollas, A.; Pan, B.; Wang, W.; Wei, X.; Zhang, L. A Two-Electron Storage Nonaqueous Organic Redox Flow Battery. Adv. Sustain. Syst. 2018, 2 (3), 1700131.

(60) Kowalski, J. A.; Casselman, M. D.; Kaur, A. P.; Milshtein, J. D.; Elliott, C. F.; Modekrutti, S.; Attanayake, N. H.; Zhang, N.; Parkin, S. R.; Risko, C.; et al. A Stable Two-Electron-Donating Phenothiazine for Application in Nonaqueous Redox Flow Batteries. J. Mater. Chem. A 2017, 5 (46), 24371– 247

24379.

(61) Attanayake, N. H.; Kowalski, J. A.; Greco, K. V; Casselman, M. D.; Milshtein, J. D.; Chapman, S. J.; Parkin, S. R.; Brushett, F. R.; Odom, S. A. Tailoring Two-Electron-Donating Phenothiazines To Enable High-Concentration Redox Electrolytes for Use in Nonaqueous Redox Flow Batteries. Chem. Mater. 2019, 31 (12), 4353–4363.

(62) Zhang, C.; Niu, Z.; Peng, S.; Ding, Y.; Zhang, L.; Guo, X.; Zhao, Y.; Yu, G. Phenothiazine-Based Organic Catholyte for High-Capacity and Long-Life Aqueous Redox Flow Batteries. Adv. Mater. 2019, 31 (24), 1901052.

(63) Li, B.; Liu, J. Progress and Directions in Low-Cost Redox-Flow Batteries for Large-Scale Energy Storage. Natl. Sci. Rev. 2017, 4 (1), 91–105.

(64) Ponce de León, C.; Frías-Ferrer, A.; González-García, J.; Szánto, D. A.; Walsh, F. C. Redox Flow Cells for Energy Conversion. J. Power Sources 2006, 160 (1), 716–732.

(65) Fukushima, T.; Tanaka, K.; Lim, H.; Moriyama, M. Mechanism of Cytotoxicity of Paraquat. Environ. Health Prev. Med. 2002, 7 (3), 89–94.

(66) Tsai, W.-T. A Review on Environmental Exposure and Health Risks of Herbicide Paraquat. Toxicol. Environ. Chem. 2013, 95 (2), 197–206.

(67) Tobiszewski, M.; Namieśnik, J.; Pena-Pereira, F. Environmental Risk-Based Ranking of Solvents Using the Combination of a Multimedia Model and Multi- Criteria Decision Analysis. Green Chem. 2017, 19 (4), 1034–1042.

(68) Schlosser, P. M.; Bale, A. S.; Gibbons, C. F.; Wilkins, A.; Cooper, G. S. Human Health Effects of Dichloromethane: Key Findings and Scientific Issues. Environ. Health Perspect. 2015, 123 (2), 114–119.

(69) Kealy, T. J.; Pauson, P. L. A New Type of Organo-Iron Compound. Nat. (London, United Kingdom) 1951, 168, 1039–1040.

(70) Astruc, D. Why Is Ferrocene so Exceptional? Eur. J. Inorg. Chem. 2017, 2017 (1), 6–29.

(71) Metallocenes and Sandwich Complexes BT - Organometallic Chemistry and 248

Catalysis; Astruc, D., Ed.; Springer Berlin Heidelberg: Berlin, Heidelberg, 2007; pp 251–288.

(72) Rogers, E. I.; Silvester, D. S.; Poole, D. L.; Aldous, L.; Hardacre, C.; Compton, R. G. Voltammetric Characterization of the Ferrocene|Ferrocenium and Cobaltocenium|Cobaltocene Redox Couples in RTILs. J. Phys. Chem. C 2008, 112 (7), 2729–2735.

(73) Lorber, C. Vanadium Organometallics. In Compr. Organomet. Chem. III; Elsevier Ltd., 2007; Vol. 5, pp 1–60.

(74) Kalousova, J.; Holecek, J.; Votinsky, J.; Benes, L. Reaction Behavior of Chromocene. Zeitschrift fuer Chemie 1983, 23 (9), 327–331.

(75) Shapiro, P. J. The Elusive Chemistry of Bent-Sandwich Chromocene. Curr. Org. Chem. 2007, 11 (17), 1482–1490.

(76) Barnett, K. W. Chemistry of Nickelocene. J. Organomet. Chem. 1974, 78 (2), 139–163.

(77) Pasynkiewicz, S.; Pietrzykowski, A. From Nickelocene to Novel Organonickel Compounds. Coord. Chem. Rev. 2002, 231 (1–2), 199–206.

(78) Warren Jr., L. F.; Hawthorne, M. F. Metallocene Analogs of Copper, Gold, and Palladium Derived from the (3)-1,2-Dicarbollide Ion. J. Am. Chem. Soc. 1968, 90 (18), 4823–4828.

(79) C. Green, J. Bent Metallocenes Revisited. Chem. Soc. Rev. 1998, 27 (4), 263–272.

(80) Cross, J. M.; Blower, T. R.; Gallagher, N.; Gill, J. H.; Rockley, K. L.; Walton, J. W. Anticancer Ru(II) and Rh(III) Piano-Stool Complexes That Are Histone Deacetylase Inhibitors. Chempluschem 2016, 81 (12), 1276–1280.

(81) Liu, H.-K.; Sadler, P. J. Metal Complexes as DNA Intercalators. Acc. Chem. Res. 2011, 44 (5), 349–359.

(82) Wang, B.; Takahashi, S.; Du, X.; Anzai, J.-I. Electrochemical Biosensors Based on Ferroceneboronic Acid and Its Derivatives: A Review. Biosensors 2014, 4 (3), 243–256. 249

(83) Luft, G. Use of Metallocene Catalysts in the Polymerization under High Pressure. In High Pressure Chemical Engineering; von Rohr, P. R., Trepp, C. B. T.-P. T. P., Eds.; Elsevier, 1996; Vol. 12, pp 73–78.

(84) Ahn, S.; Song, Y.-S.; Yoo, B. R.; Jung, I. N. Lewis Acid-Catalyzed Friedel−Crafts Alkylation of Ferrocene with Allylchlorosilanes. Organometallics 2000, 19 (14), 2777–2780.

(85) Rinehart, K. L.; Motz, K. L.; Moon, S. Organic Chemistry of Ferrocene. I. The Acetylation of Dialkylferrocenes. J. Am. Chem. Soc. 1957, 79 (11), 2749– 2754.

(86) Rosenblum, M.; Howells, W. G.; Banerjee, A. K.; Bennett, C. The Structure and Chemistry of Ferrocene. VI. Mechanism of the Arylation Reaction. J. Am. Chem. Soc. 1962, 84 (14), 2726–2732.

(87) Broadhead, G. D.; Osgerby, J. M.; Pauson, P. L. Ferrocene Derivatives. V. Ferrocenealdehyde. J. Chem. Soc. 1958, 650–656.

(88) Pauson, P. L. Ferrocene. In e-EROS Encycl. Reagents Org. Synth.; John Wiley & Sons, Ltd., 2001; pp 1–5.

(89) Slocum, D. W.; Achermann, W. The Preparation of Several N-Substituted and N,N-Disubstituted Ferrocenesulfonamides. Synth. React. Inorg. Met. Chem. 1982, 12 (4), 397–405.

(90) Sethi, S.; Das, P. K.; Behera, N. The Chemistry of Aminoferrocene, Fe{(η5- C5H4NH2)(η5-Cp)}: Synthesis, Reactivity and Applications. J. Organomet. Chem. 2016, 824, 140–165.

(91) Ouellette, R. J.; Rawn, J. D. Electrophilic Aromatic Substitution; Academic Press, 2018; pp 375–407.

(92) Pagano, J. K.; Sylvester, E. C.; Warnick, E. P.; Dougherty, W. G.; Piro, N. A.; Kassel, W. S.; Nataro, C. Electrochemical Parameterization of 1,1′- Disubstituted Cobaltocenium Compounds. J. Organomet. Chem. 2014, 750, 107–111.

(93) Wolter-Steingrube, A.; Cordsen, K.; Heck, J. Nucleophilic Substitution in the Nitrocobaltocenium Ion. Eur. J. Inorg. Chem. 2017, 2017 (10), 1314–1319.

250

(94) Lauck, M.; Förster, C.; Heinze, K. N-Cobaltocenium Amide as Reactive Nucleophilic Reagent for Donor–Acceptor Bimetallocenes. Organometallics 2017, 36 (24), 4968–4978.

(95) Vanicek, S.; Kopacka, H.; Wurst, K.; Müller, T.; Hassenrück, C.; Winter, R. F.; Bildstein, B. Monofunctionalized Cobaltocenium Compounds by Dediazoniation Reactions of Cobaltoceniumdiazonium Bis(Hexafluorophosphate). Organometallics 2016, 35 (12), 2101–2109.

(96) Knox, G. R.; Pauson, P. L. 134. Ferrocene Derivatives. Part VII. Some Sulphur Derivatives. J. Chem. Soc. 1958, No. 0, 692–696.

(97) Chanawanno, K.; Holstrom, C.; Crandall, L. A.; Dodge, H.; Nemykin, V. N.; Herrick, R. S.; Ziegler, C. J. The Synthesis and Structures of 1,1’- Bis(Sulfonyl)Ferrocene Derivatives. Dalton Trans. 2016, 45 (36), 14320– 14326.

(98) Weinmayr, V. Hydrogen Fluoride as a Condensing Agent. V. Reactions of Dicyclopentadienyliron in Anhydrous Hydrogen Fluoride. J. Am. Chem. Soc. 1955, 77 (11), 3009–3011.

(99) Slocum, D. W.; Achermann, W. The Preparation of Several N-Substituted and N,N-Substituted Ferrocenesulfonamides. Synth. React. Inorganic, Met. Chem. 1982, 12 (4), 397–405.

(100) Chanawanno, K.; Holstrom, C.; Nemykin, V. N.; Herrick, R. S.; Ziegler, C. J. New 1,1’-Ferrocene Bis(Sulfonyl) Reagents. ChemistrySelect 2016, 1 (20), 6438–6441.

(101) Chanawanno, K.; Blesener, T. S. S.; Schrage, B. R. R.; Nemykin, V. N. N.; Herrick, R. S. S.; Ziegler, C. J. J. Amino Acid Ferrocene Conjugates Using Sulfonamide Linkages. J. Organomet. Chem. 2018, 870, 121–129.

(102) Almendras, I.; Huentupil, Y.; Novoa, N.; Roussel, P.; Melis, D. R.; Smith, G. S.; Arancibia, R. Trinuclear Ni(II), Pd(II) and Cu(II) Complexes Containing the 2-Hydroxy-Benzaldehyde-Ferrocenyl-Sulfonylhydrazone Ligand: Synthesis, Structural Characterization and Antiplasmodial Evaluation. Inorganica Chim. Acta 2019, 496, 119050.

(103) Huentupil, Y.; Chung, P.; Novoa, N.; Arancibia, R.; Roussel, P.; Oyarzo, J.; Klahn, A. H.; Silva, C.; Calvis, C.; Messeguer, R.; et al. Novel Multifunctional 251

and Multitarget Homo- (Fe2) and Heterobimetallic [(Fe,M) with M = Re or Mn] Sulfonyl Hydrazones. Dalton Trans. 2020, 49 (35), 12249–12265.

(104) Gelinas, B.; Das, D.; Rochefort, D. Air-Stable, Self-Bleaching Electrochromic Device Based on Viologen- and Ferrocene-Containing Triflimide Redox Ionic Liquids. ACS Appl. Mater. Interfaces 2017, 9 (34), 28726–28736.

(105) Xie, H. J.; Gelinas, B.; Rochefort, D. Electrochemical and Physicochemical Properties of Redox Ionic Liquids Using Electroactive Anions: Influence of Alkylimidazolium Chain Length. Electrochim. Acta 2016, 200, 283–289.

(106) Gelinas, B.; Natali, M.; Bibienne, T.; Li, Q. P.; Dolle, M.; Rochefort, D. Electrochemical and Transport Properties of Ions in Mixtures of Electroactive Ionic Liquid and Propylene Carbonate with a Lithium Salt for Lithium-Ion Batteries. J. Phys. Chem. C 2016, 120 (10), 5315–5325.

(107) Hwang, B.; Park, M.; Kim, K. Ferrocene and Cobaltocene Derivatives for Non‐Aqueous Redox Flow Batteries. ChemSusChem 2015, 8 (2), 310–314.

(108) Ding, Y.; Zhao, Y.; Li, Y.; Goodenough, J. B.; Yu, G. A High-Performance All-Metallocene-Based, Non-Aqueous Redox Flow Battery. Energy Environ. Sci. 2017, 10 (2), 491–497.

(109) Armstrong, C. G.; Hogue, R. W.; Toghill, K. E. Characterisation of the Ferrocene/Ferrocenium Ion Redox Couple as a Model Chemistry for Non- Aqueous Redox Flow Battery Research. J. Electroanal. Chem. 2020, 872, 114241.

(110) Cong, G.; Zhou, Y.; Li, Z.; Lu, Y.-C. A Highly Concentrated Catholyte Enabled by a Low-Melting-Point Ferrocene Derivative. ACS Energy Lett. 2017, 2 (4), 869–875.

(111) Kim, H.; Yoon, T.; Kim, Y.; Hwang, S.; Ryu, J. H.; Oh, S. M. Increase of Both Solubility and Working Voltage by Acetyl Substitution on Ferrocene for Non- Aqueous Flow Battery. Electrochem. commun. 2016, 69, 72–75.

(112) Huang, Q.; Li, H.; Grätzel, M.; Wang, Q. Reversible Chemical Delithiation/Lithiation of LiFePO4: Towards a Redox Flow Lithium-Ion Battery. Phys. Chem. Chem. Phys. 2013, 15 (6), 1793–1797.

(113) Park, K.; Cho, J. H.; Shanmuganathan, K.; Song, J.; Peng, J.; Gobet, M.; 252

Greenbaum, S.; Ellison, C. J.; Goodenough, J. B. New Battery Strategies with a Polymer/Al2O3 Separator. J. Power Sources 2014, 263, 52–58.

(114) Xie, C.; Xu, W.; Zhang, H.; Hu, X.; Li, X. A Multi-Electron Transfer Ferrocene Derivative Positive Redox Moiety with Improved Solubility and Potential. Chem. Commun. 2018, 54 (60), 8419–8422.

(115) Zhu, Y.; Yang, F.; Niu, Z.; Wu, H.; He, Y.; Zhu, H.; Ye, J.; Zhao, Y.; Zhang, X. Enhanced Cyclability of Organic Redox Flow Batteries Enabled by an Artificial Bipolar Molecule in Neutral Aqueous Electrolyte. J. Power Sources 2019, 417, 83–89.

(116) Hu, B.; Seefeldt, C.; DeBruler, C.; Liu, T. L. Boosting the Energy Efficiency and Power Performance of Neutral Aqueous Organic Redox Flow Batteries. J. Mater. Chem. A 2017, 5 (42), 22137–22145.

(117) DeBruler, C.; Hu, B.; Moss, J.; Liu, X.; Luo, J.; Sun, Y.; Liu, T. L. Designer Two-Electron Storage Viologen Anolyte Materials for Neutral Aqueous Organic Redox Flow Batteries. Chem 2017, 3 (6), 961–978.

(118) Kim, S.; Kim, D.; Hwang, G.; Jeon, J. A Bromide-Ligand Ferrocene Derivative Redox Species with High Reversibility and Electrochemical Stability for Aqueous Redox Flow Batteries. J. Electroanal. Chem. 2020, 869, 114131.

(119) Cosimbescu, L.; Wei, X.; Vijayakumar, M.; Xu, W.; Helm, M. L.; Burton, S. D.; Sorensen, C. M.; Liu, J.; Sprenkle, V.; Wang, W. Anion-Tunable Properties and Electrochemical Performance of Functionalized Ferrocene Compounds. Sci. Rep. 2015, 5, 14117.

(120) Laramie, S. M.; Milshtein, J. D.; Breault, T. M.; Brushett, F. R.; Thompson, L. T. Performance and Cost Characteristics of Multi-Electron Transfer, Common Ion Exchange Non-Aqueous Redox Flow Batteries. J. Power Sources 2016, 327, 681–692.

(121) Yu, J.; Salla, M.; Zhang, H.; Ji, Y.; Zhang, F.; Zhou, M.; Wang, Q. A Robust Anionic Sulfonated Ferrocene Derivative for PH-Neutral Aqueous Flow Battery. Energy Storage Mater. 2020, 29, 216–222.

(122) Hu, B.; DeBruler, C.; Rhodes, Z.; Liu, T. L. Long-Cycling Aqueous Organic Redox Flow Battery (AORFB) toward Sustainable and Safe Energy Storage. 253

J. Am. Chem. Soc. 2017, 139 (3), 1207–1214.

(123) Huskinson, B.; Marshak, M. P.; Suh, C.; Er, S.; Gerhardt, M. R.; Galvin, C. J.; Chen, X.; Aspuru-Guzik, A.; Gordon, R. G.; Aziz, M. J. A Metal-Free Organic–Inorganic Aqueous Flow Battery. Nature 2014, 505 (7482), 195– 198.

(124) Mourad, E.; Coustan, L.; Freunberger, S. A.; Mehdi, A.; Vioux, A.; Favier, F.; Fontaine, O. Biredox Ionic Liquids: Electrochemical Investigation and Impact of Ion Size on Electron Transfer. Electrochim. Acta 2016, 206, 513–523.

(125) Mourad, E.; Coustan, L.; Lannelongue, P.; Zigah, D.; Mehdi, A.; Vioux, A.; Freunberger, S. A.; Favier, F.; Fontaine, O. Biredox Ionic Liquids with Solid- like Redox Density in the Liquid State for High-Energy Supercapacitors. Nat. Mater. 2017, 16 (4), 446–453.

(126) Vioux, A.; Coasne, B. From Ionogels to Biredox Ionic Liquids: Some Emerging Opportunities for Electrochemical Energy Storage and Conversion Devices. Adv. Energy Mater. 2017, 7 (22), 1700883.

(127) Pham-Truong, T.-N.; Randriamahazaka, H.; Ghilane, J. Electrochemistry of Bi-Redox Ionic Liquid from Solution to Bi-Functional Carbon Surface. Electrochim. Acta 2020, 354, 136689.

(128) Zhen, Y.; Zhang, C.; Yuan, J.; Zhao, Y.; Li, Y. Ferrocene/Anthraquinone Based Bi-Redox Molecule for Symmetric Nonaqueous Redox Flow Battery. J. Power Sources 2020, 480, 229132.

(129) Kosswattaarachchi, A. M.; Friedman, A. E.; Cook, T. R. Characterization of a BODIPY Dye as an Active Species for Redox Flow Batteries. ChemSusChem 2016, 9 (23), 3317–3323.

(130) Duan, W.; Vemuri, R. S.; Milshtein, J. D.; Laramie, S.; Dmello, R. D.; Huang, J.; Zhang, L.; Hu, D.; Vijayakumar, M.; Wang, W.; et al. A Symmetric Organic-Based Nonaqueous Redox Flow Battery and Its State of Charge Diagnostics by FTIR. J. Mater. Chem. A 2016, 4 (15), 5448–5456.

(131) Hwang, S.; Kim, H. seung; Ryu, J. H.; Oh, S. M. N-Ferrocenylphthalimide; A Single Redox Couple Formed by Attaching a Ferrocene Moiety to Phthalimide for Non-Aqueous Flow Batteries. J. Power Sources 2018, 395 (May), 60–65. 254

(132) Hwang, S.; Kim, H.; Ryu, J. H.; Oh, S. M. N-(α-Ferrocenyl)Ethylphthalimide as a Single Redox Couple for Non-Aqueous Flow Batteries. J. Power Sources 2019, 421, 1–5.

(133) Cong, G.; Lu, Y.-C. Organic Eutectic Electrolytes for Future Flow Batteries. Chem 2018, 4 (12), 2732–2734.

(134) Wang, Y.; Niu, Z.; Zheng, Q.; Zhang, C.; Ye, J.; Dai, G.; Zhao, Y.; Zhang, X. Zn-Based Eutectic Mixture as Anolyte for Hybrid Redox Flow Batteries. Sci. Rep. 2018, 8 (1), 5740.

(135) Zhang, C.; Zhang, L.; Ding, Y.; Peng, S.; Guo, X.; Zhao, Y.; He, G.; Yu, G. Progress and Prospects of Next-Generation Redox Flow Batteries. Energy Storage Mater. 2018, 15, 324–350.

(136) Zhang, C.; Qian, Y.; Ding, Y.; Zhang, L.; Guo, X.; Zhao, Y.; Yu, G. Biredox Eutectic Electrolytes Derived from Organic Redox-Active Molecules: High- Energy Storage Systems. Angew. Chemie, Int. Ed. 2019, 58 (21), 7045– 7050.

(137) Elvidge, J. A.; Linstead, R. P. Heterocyclic Imines. I. Imino Derivatives of Isoindoline and Their Reaction with Primary Bases. J. Chem. Soc. 1952, 5000–5007.

(138) Csonka, R.; Speier, G.; Kaizer, J. Isoindoline-Derived Ligands and Applications. RSC Adv. 2015, 5 (24), 18401–18419.

(139) Fernandez-Lazaro, F.; Torres, T.; Hauschel, B.; Hanack, M. Hemiporphyrazines as Targets for the Preparation of Molecular Materials: Synthesis and Physical Properties. Chem. Rev. (Washington, D. C.) 1998, 98 (2), 563–575.

(140) Engle, J. T.; Allison, A. N.; Standard, J. M.; Tamgho, I.; Ziegler, C. J. The Structures of Several Modified Isoindolines, the Building Blocks of Phthalocyanines. J. Porphyrins Phthalocyanines 2013, 17 (8–9), 712–721.

(141) Zhang, Z.-Q.; Njus, J. M.; Sandman, D. J.; Guo, C.; Foxman, B. M.; Erk, P.; van Gelder, R. Diiminoisoindoline: Tautomerism, Conformations, and Polymorphism. Chem. Commun. 2004, No. 7, 886–887.

(142) de la Torre, G.; Vázquez, P.; Agulló-López, F.; Torres, T. Role of Structural 255

Factors in the Nonlinear Optical Properties of Phthalocyanines and Related Compounds. Chem. Rev. 2004, 104 (9), 3723–3750.

(143) Claessens, C. G.; Hahn, U.; Torres, T. Phthalocyanines: From Outstanding Electronic Properties to Emerging Applications. Chem. Rec. 2008, 8 (2), 75– 97.

(144) De La Torre, G.; Bottari, G.; Torres, T. Phthalocyanines and Subphthalocyanines: Perfect Partners for Fullerenes and Carbon Nanotubes in Molecular Photovoltaics. Adv. Energy Mater. 2017, 7 (10), 1601700.

(145) Day, V. W.; Marks, T. J.; Wachter, W. A. Large Metal Ion-Centered Template Reactions. Uranyl Complex of Cyclopentakis(2-Iminoisoindoline). J. Am. Chem. Soc. 1975, 97 (16), 4519–4527.

(146) Geyer, M.; Plenzig, F.; Rauschnabel, J.; Hanack, M.; Rey, B. D.; Sastre, A.; Torres, T. Subphthalocyanines: Preparation, Reactivity and Physical Properties. Synthesis (Stuttg). 1996, 1996, 1139–1151.

(147) Dowds, M.; Nielsen, M. B. Controlling the Optical Properties of Boron Subphthalocyanines and Their Analogues. Mol. Syst. Des. Eng. 2021, 6 (1), 6–24.

(148) Morse, G. E.; Bender, T. P. Boron Subphthalocyanines as Organic Electronic Materials. ACS Appl. Mater. Interfaces 2012, 4 (10), 5055–5068.

(149) Shibata, N.; Das, B.; Tokunaga, E.; Shiro, M.; Kobayashi, N. Trifluoroethoxy- Coating Improves the Axial Ligand Substitution of Subphthalocyanine. Chem. – A Eur. J. 2010, 16 (25), 7554–7562.

(150) González-Rodríguez, D.; Claessens, C. G.; Torres, T.; Liu, S.; Echegoyen, L.; Vila, N.; Nonell, S. Tuning Photoinduced Energy- and Electron-Transfer Events in Subphthalocyanine–Phthalocyanine Dyads. Chem. – A Eur. J. 2005, 11 (13), 3881–3893.

(151) Sripothongnak, S.; Barone, N. V; Çetin, A.; Wu, R.; Durfee, W. S.; Ziegler, C. J. The Structures of Free-Base Carbahemiporphyrazines. J. Porphyr. Phthalocyanines 2010, 14 (02), 170–177.

(152) Czaun, M.; Kothandaraman, J.; Goeppert, A.; Yang, B.; Greenberg, S.; May, R. B.; Olah, G. A.; Prakash, G. K. S. Iridium-Catalyzed Continuous Hydrogen 256

Generation from Formic Acid and Its Subsequent Utilization in a Fuel Cell: Toward a Carbon Neutral Chemical Energy Storage. ACS Catal. 2016, 6 (11), 7475–7484.

(153) Saha, S.; Sahil, S. T.; Mazumder, M. M. R.; Stephens, A. M.; Cronin, B.; Duin, E. C.; Jurss, J. W.; Farnum, B. H. Synthesis, Characterization, and Electrocatalytic Activity of Bis(Pyridylimino)Isoindoline Cu(II) and Ni(II) Complexes. Dalton Trans. 2021, 50 (3), 926–935.

(154) Fernandez-Lazaro, F.; De Mendoza, J.; Mo, O.; Rodriguez-Morgade, S.; Torres, T.; Yanez, M.; Elguero, J. Phthalocyanine Analogs. Part 1. Synthesis, Spectroscopy, and Theoretical Study of 8,18- Dihydrodibenzo[b,l]-5,7,8,10,15,17,18,20-Octa-Azaporphyrin and MNDO Calculations on Its Related Hueckel Heteroannulene. J. Chem. Soc. Perkin Trans. 2 Phys. Org. Chem. 1989, No. 7, 797–803.

(155) Agostinelli, E.; Attanasio, D.; Collamati, I.; Fares, V. Hemiporphyrazine, a Porphyrin-Related Macrocycle That Induces Rhombically Compressed Stereochemistries: Structure and Properties of Bis(Pyridine)(Hemiporphyrazinato)Nickel(II). Inorg. Chem. 1984, 23 (8), 1162–1165.

(156) Osinski, A. J.; Blesener, T.; Hasheminasab, A.; Holstrom, C.; Nemykin, V. N.; Herrick, R. S.; Ziegler, C. J. Re(CO)3-Templated Synthesis of Semihemiporphyrazines. Inorg. Chem. 2016, 55 (24), 12527–12530.

(157) Tamgho, I.-S.; Engle, J. T.; Ziegler, C. J. The Syntheses and Structures of Bis(Alkylimino)Isoindolines. Tetrahedron Lett. 2013, 54 (45), 6114–6117.

(158) Elvidge, J. A.; Redman, A. P. Heterocyclic Imines and Amines. Part XV. Reactions of Hydrazines with 1, 3-Di-Iminoisoindoline and Related Compounds. J. Chem. Soc. Perkin Trans. 1 Org. Bio-Organic Chem. 1972, No. 22, 2820–2826.

(159) Mandal, S. K.; Woon, T. C.; Thompson, L. K.; Newlands, M. J.; Gabe, E. J. Binuclear Copper(II) Complexes of Some Tetradentate Pyridyl Phthalazine Ligands. Structure, Electrochemistry and Magnetism - Crystal-Structure of Aqua-µ-[1,4-Di-(4',6'-Dimethylpyridin-2'-Ylamino)-Phthalazine-N1',µ-N2 ,µ- N3. Aust. J. Chem. 1986, 39 (7), 1007–1021.

(160) Lu, H.; Kobayashi, N. Optically Active Porphyrin and Phthalocyanine

257

Systems. Chem. Rev. 2016, 116 (10), 6184–6261.

(161) Braun, A.; Tcherniac, J. Über Die Produkte Der Einwirkung von Acetanhydrid Auf Phthalamid. Berichte der Dtsch. Chem. Gesellschaft 1907, 40 (2), 2709– 2714.

(162) Linstead, R. P. 212. Phthalocyanines. Part I. A New Type of Synthetic Colouring Matters. J. Chem. Soc. 1934, No. 0, 1016–1017.

(163) McKeown, N. B. Phthalocyanine Materials: Synthesis, Structure and Function; Bruce Dunn; Cambridge University Press, 1998.

(164) Marks, T. J.; Stojakovic, D. R. Macrocycle Contraction Reactions of 5,35:14,19-Diimino-7,12:21,26:28,33- Trinitrilopentabenzo[c,h,m,r,w][1,6,11,16,21]Pentaazacyclopentacosinatodi oxouranium(VI). J. Chem. Soc. Chem. Commun. 1975, No. 1, 28–29.

(165) Brumboiu, I. E.; Prokopiou, G.; Kronik, L.; Brena, B. Valence Electronic Structure of Cobalt Phthalocyanine from an Optimally Tuned Range- Separated Hybrid Functional. J. Chem. Phys. 2017, 147 (4), 44301.

(166) Brown, C. J. Crystal Structure of β-Copper Phthalocyanine. J. Chem. Soc. A Inorganic, Phys. Theor. 1968, No. 0, 2488–2493.

(167) Marom, N.; Hod, O.; Scuseria, G. E.; Kronik, L. Electronic Structure of Copper Phthalocyanine: A Comparative Density Functional Theory Study. J. Chem. Phys. 2008, 128 (16), 164107/1-164107/6.

(168) Gürek, A. G.; Durmuş, M.; Ahsen, V. Synthesis and Mesomorphic Properties of Tetra- and Octa-Substituted Phthalocyanines. New J. Chem. 2004, 28 (6), 693–699.

(169) Åhlund, J.; Nilson, K.; Schiessling, J.; Kjeldgaard, L.; Berner, S.; Mårtensson, N.; Puglia, C.; Brena, B.; Nyberg, M.; Luo, Y. The Electronic Structure of Iron Phthalocyanine Probed by Photoelectron and X-Ray Absorption Spectroscopies and Density Functional Theory Calculations. J. Chem. Phys. 2006, 125 (3), 34709.

(170) Urbani, M.; de la Torre, G.; Nazeeruddin, M. K.; Torres, T. Phthalocyanines and Porphyrinoid Analogues as Hole- and Electron-Transporting Materials for Perovskite Solar Cells. Chem. Soc. Rev. 2019, 48 (10), 2738–2766. 258

(171) Gonzalez-Rodriguez, D.; Bottari, G. Phthalocyanines, Subphthalocyanines and Porphyrins for Energy and Electron Transfer Applications. J. Porphyr. Phthalocyanines 2009, 13 (4–5), 624–636.

(172) O’Flaherty, S. M.; Hold, S. V; Cook, M. J.; Torres, T.; Chen, Y.; Hanack, M.; Blau, W. J. Molecular Engineering of Peripherally and Axially Modified Phthalocyanines for Optical Limiting and Nonlinear Optics. Adv. Mater. (Weinheim, Ger. 2003, 15 (1), 19–32.

(173) Remiro-Buenamañana, S.; Díaz-Moscoso, A.; Hughes, D. L.; Bochmann, M.; Tizzard, G. J.; Coles, S. J.; Cammidge, A. N. Synthesis of Meso- Substituted Subphthalocyanine–Subporphyrin Hybrids: Boron Subtribenzodiazaporphyrins. Angew. Chemie Int. Ed. 2015, 54 (26), 7510– 7514.

(174) Shimizu, S.; Kobayashi, N. Structurally-Modified Subphthalocyanines: Molecular Design towards Realization of Expected Properties from the Electronic Structure and Structural Features of Subphthalocyanine. Chem. Commun. 2014, 50 (53), 6949–6966.

(175) Furuyama, T.; Sato, T.; Kobayashi, N. A Bottom-up Synthesis of Antiaromatic Expanded Phthalocyanines: Pentabenzotriazasmaragdyrins, i.e. Norcorroles of Superphthalocyanines. J. Am. Chem. Soc. 2015, 137 (43), 13788–13791.

(176) Marks, T. J.; Stojakovic, D. R. Large Metal Ion-Centered Template Reactions. Chemical and Spectral Studies of the “Superphthalocyanine” Dioxocyclopentakis(1-Iminoisoindolinato)Uranium(VI) and Its Derivatives. J. Am. Chem. Soc. 1978, 100 (6), 1695–1705.

(177) Matsushita, O.; Derkacheva, V. M.; Muranaka, A.; Shimizu, S.; Uchiyama, M.; Luk’yanets, E. A.; Kobayashi, N. Rectangular-Shaped Expanded Phthalocyanines with Two Central Metal Atoms. J. Am. Chem. Soc. 2012, 134 (7), 3411–3418.

(178) Kietaibl, H. Die Kristall- Und Molekülstruktur Eines Neuratigen Phthalocyaninähnlichen Borkomplexes. Monatshefte für Chemie / Chem. Mon. 1974, 105 (2), 405–418.

(179) Meller, A.; Ossko, A. Phthalocyaninartige Bor-Komplexe. Monatshefte für Chemie / Chem. Mon. 1972, 103 (1), 150–155.

259

(180) Mori, S.; Shibata, N. Synthesis and Application of Trifluoroethoxy- Substituted Phthalocyanines and Subphthalocyanines. Beilstein J. Org. Chem. 2017, 13, 2273–2296.

(181) Pakhomov, G. L.; Travkin, V. V; Stuzhin, P. A. Hexachlorinated Boron(III) Subphthalocyanine as Acceptor for Organic Photovoltaics: A Brief Overview. In Recent Adv. Boron-Containing Mater.; IntechOpen Ltd., 2020; pp 1–18.

(182) Mayoral, M. J.; Torres, T.; Gonzalez-Rodriguez, D. Polar Columnar Assemblies of Subphthalocyanines. J. Porphyr. Phthalocyanines 2020, 24 (1/3), 33–42.

(183) Medina, A.; Claessens, C. G. Subphthalocyanines:: Synthesis, Self- Organization, Properties and Applications. J. Porphyr. Phthalocyanines 2009, 13 (4–5), 446–454.

(184) Hanack, M.; Geyer, M. Synthesis and Separation of Structural Isomers of Tri-Tert-Butylsubphthalocyaninatophenylboron(III). J. Chem. Soc. Chem. Commun. 1994, No. 19, 2253–2254.

(185) Kasuga, K.; Idehara, T.; Handa, M.; Ueda, Y.; Fujiwara, T.; Isa, K. Structure and Some Properties of (Alkoxo)(Subphthalocyaninato)Boron(III). Bull. Chem. Soc. Jpn. 1996, 69 (9), 2559–2563.

(186) Potz †, R.; Göldner, M.; Hückstädt, H.; Cornelissen, U.; Tutaß, A.; Homborg, H. Synthese Und Strukturelle Charakterisierung von Borsubphthalocyaninaten. Zeitschrift für Anorg. und Allg. Chemie 2000, 626 (2), 588–596.

(187) Claessens, C. G.; González-Rodríguez, D.; Rodríguez-Morgade, M. S.; Medina, A.; Torres, T. Subphthalocyanines, Subporphyrazines, and Subporphyrins: Singular Nonplanar Aromatic Systems. Chem. Rev. 2014, 114 (4), 2192–2277.

(188) Wang, Z.; Fu, X. Synthesis, Characterization, and Reactivity Studies of Subphthalocyanine Boron Triflate. Organometallics 2017, 36 (2), 285–290.

(189) González-Rodríguez, D.; Torres, T. Peripheral Functionalization of Subphthalocyanines. European J. Org. Chem. 2009, 2009 (12), 1871–1879.

(190) Tejerina, L.; Martínez-Díaz, M. V.; Torres, T. One-Pot Synthesis of π- 260

Extended Fluorenone-Fused Subphthalocyanines. Org. Lett. 2019, 21 (8), 2908–2912.

(191) Roy, I.; Shetty, D.; Hota, R.; Baek, K.; Kim, J.; Kim, C.; Kappert, S.; Kim, K. A Multifunctional Subphthalocyanine Nanosphere for Targeting, Labeling, and Killing of Antibiotic-Resistant Bacteria. Angew. Chemie Int. Ed. 2015, 54 (50), 15152–15155.

(192) Sakamoto, K.; Yoshino, S.; Takemoto, M.; Sugaya, K.; Kubo, H.; Komoriya, T.; Kamei, S.; Furukawa, S. Synthesis of Arylsulfanyl-Subphthalocyanines and Their Ring Expansion Reaction. J. Porphyr. Phthalocyanines 2015, 19 (05), 688–694.

(193) Cogal, S.; Erten-Ela, S.; Ocakoglu, K.; Oksuz, A. U. Asymmetric Phthalocyanine Derivatives Containing 4-Carboxyphenyl Substituents for Dye-Sensitized Solar Cells. Dye. Pigment. 2015, 113, 474–480.

(194) Kobayashi, N.; Kondo, R.; Nakajima, S.; Osa, T. New Route to Unsymmetrical Phthalocyanine Analogs by the Use of Structurally Distorted Subphthalocyanines. J. Am. Chem. Soc. 1990, 112 (26), 9640–9641.

(195) Elvidge, J. A.; Linstead, R. P. 976. Conjugated Macrocycles. Part XXIV. A New Type of Cross-Conjugated Macrocycle, Related to the Azaporphins. J. Chem. Soc. 1952, No. 0, 5008.

(196) Campbell, J. B. Macrocyclic Coloring Compounds., October 2, 1956.

(197) Peng, S.-M.; Wang, Y.; Ho, T.-F.; Chango, I.-C.; Tang, C.-P.; Wang, C.-J. Structural Relationships between the Hemiporphyrazine Macrocyclic Ligand and Its Metal Complexes. I.: Saddle Shaped Neutral Ligand Hydrate, C26H16N8·-H2O, and Nickel Complex, [Ni(C26H14N8)]. J. Chinese Chem. Soc. 1986, 33 (1), 13–21.

(198) Ruf, M.; Durfee, W. S.; Pierpont, C. G. Synthesis of a Quinone- Functionalized Macrocyclic Ligand and the Intense Fluorescence of Its Zinc Complex. Chem. Commun. 2004, No. 8, 1022–1023.

(199) Sripothongnak, S.; Pischera, A. M.; Espe, M. P.; Durfee, W. S.; Ziegler, C. J. Synthesis and Characterization of Lithium Hemiporphyrazines. Inorg. Chem. 2009, 48 (4), 1293–1300.

261

(200) Muranaka, A.; Ohira, S.; Hashizume, D.; Koshino, H.; Kyotani, F.; Hirayama, M.; Uchiyama, M. [18]/[20]π Hemiporphyrazine: A Redox-Switchable Near- Infrared Dye. J. Am. Chem. Soc. 2012, 134 (1), 190–193.

(201) Costa, R.; Schick, A. J.; Paul, N. B.; Durfee, W. S.; Ziegler, C. J. Hydroxybenziphthalocyanines: Non-Aromatic Phthalocyanine Analogues That Exhibit Strong UV-Visible Absorptions. New J. Chem. 2011, 35 (4), 794–799.

(202) Miwa, H.; Ishii, K.; Kobayashi, N. Electronic Structures of Zinc and Palladium Tetraazaporphyrin Derivatives Controlled by Fused Benzo Rings. Chem. – A Eur. J. 2004, 10 (18), 4422–4435.

(203) Barone, N.; Costa, R.; Sripothangnok, S.; Ziegler, C. J. Dihydroxy- and Tetrahydroxydicarbahemiporphyrazine: Phthalocyanine Analogues with Phenol and Resorcinol Units. Eur. J. Inorg. Chem. 2010, No. 5, 775–780.

(204) Clark, P. F.; Elvidge, J. A.; Linstead, R. P. Conjugated Macrocycles. Part XXV. Cross-Conjugated Macrocycles with Inner Great Rings of 16, 20, and 24 Atoms. J. Chem. Soc. 1954, No. 0, 2490–2497.

(205) Costa, R.; Ziegler, C. J. Cyclohexylcyanine: A New Aliphatic Phthalocyanine Analogue. Chem. Commun. 2011, 47 (3), 982–984.

(206) Elvidge, J. A.; Golden, J. H. Conjugated Macrocycles. XXVIII. Adducts from Diiminoisoindoline and m-Arylenediamines, and a New Type of Cross- Conjugated Macrocycle with Three-Quarters of the Chromophore of Phthalocyanine. J. Chem. Soc. 1957, 700–709.

(207) Sripothongnak, S.; Barone, N.; Ziegler, C. J. C–H Bond Activation and Ring Oxidation in Nickel Carbahemiporphyrazines. Chem. Commun. 2009, No. 30, 4584–4586.

(208) Fields, K. B.; Engle, J. T.; Sripothongnak, S.; Kim, C.; Zhang, X. P.; Ziegler, C. J. Cobalt Carbaporphyrin-Catalyzed Cyclopropanation. Chem. Commun. 2011, 47 (2), 749–751.

(209) Bonner, E. S.; Engle, J. T.; Sripothongnak, S.; Ziegler, C. J. Zinc Complexes of the Carbahemiporphyrazines. Dalton Trans. 2010, 39 (8), 1932–1934.

(210) Rodríguez-Morgade, M. S.; Cabezón, B.; Esperanza, S.; Torres, T. 262

Expanded Phthalocyanine Analogues: Synthesis and Characterization of New Triazole-Derived Annulenes Containing Six Heterocyclic Subunits. Chem. – A Eur. J. 2001, 7 (11), 2407–2413.

(211) Shimizu, S.; Zhu, H.; Kobayashi, N. Azaphenalene Phthalocyanines: Phthalocyanine Analogues with Six-Membered-Ring Units Instead of Five- Membered-Ring Units. Chem. - Eur. J 2010, 16 (36), 11151–11159, S11151/1-S11151/6.

(212) Kobayashi, N.; Sasaki, N.; Konami, H. A New Phthalocyanine Analog Containing One 1,8-Naphthalene Unit. Inorg. Chem. 1997, 36 (25), 5674– 5675.

(213) Shimizu, S.; Uemura, K.; Zhu, H.; Kobayashi, N. Core-Modified Phthalocyanine Analogues with a Seven-Membered Ring Unit in Place of a Five-Membered Ring Unit. Tetrahedron Lett. 2012, 53 (5), 579–581.

(214) Shimizu, S.; Zhu, H.; Kobayashi, N. Azepiphthalocyanine: An Unprecedented Large Twist of a π-Conjugation System upon Core- Modification with a Seven-Membered Ring Unit. Chem. Commun. (Cambridge, United Kingdom) 2011, 47 (11), 3072–3074.

(215) Qi, D.; Zhang, Y.; Cai, X.; Jiang, J.; Bai, M. Inner Hydrogen Atom Transfer in Benzo-Fused Low Symmetrical Metal-Free Tetraazaporphyrin and Phthalocyanine Analogues: Density Functional Theory Studies. J. Mol. Graph. Model. 2009, 27 (6), 693–700.

(216) Elvidge, J. A.; Golden, J. H.; Linstead, R. P. Conjugated Macrocycles. XXIX. Tribenzotetrazaporphin Metal Derivatives and Dibromotribenzotetrazaporphin. J. Chem. Soc. 1957, 2466–2472.

(217) Tanaka, Y.; Murayama, T.; Muranaka, A.; Imai, E.; Uchiyama, M. Ring- Opened Hemiporphyrazines: Helical Molecules Exhibiting Circularly Polarized Luminescence. Chem. – A Eur. J. 2020, 26 (8), 1768–1771.

(218) Fernandez-Lazaro, F.; Rodriguez-Morgade, S.; Torres, T. Synthesis and Electrical Properties of Metallotriazolehemiporphyrazines. Synth. Met. 1994, 62 (3), 281–285.

(219) Berezina, G. R.; Kolesnikov, N. A.; Smirnov, R. P. Solubility and Thermodynamics of Solution of Conjugate Macroheterocycles and Their 263

Complexes with Metals in Dimethylformamide. Izv. Vyss. Uchebnykh Zaved. Khimiya i Khimicheskaya Tekhnologiya 1989, 32 (9), 42–45.

(220) Smirnov, R. P.; Gnedina, V. A.; Borodkin, V. F. Reaction of Diamines with Cyanuric Chloride. Unsymmetrical Macroheterocyclic Compounds and Their Metal Complexes. In Tr. Vses. Mezhvuz. Nauch.-Tekh. Konf. Vop. Sin. Primen. Krasitelei; 1970; pp 17–21.

(221) Iizuka, F.; Kage, Y.; Kobayashi, N.; Furuta, H.; Shimizu, S. Phenylene- Bridged Expanded Porphyrazines. Chempluschem 2017, 82 (7), 1021– 1024.

(222) Rodriguez-Morgade, M. S.; Cabezon, B.; Esperanza, S.; Torres, T. Expanded Phthalocyanine Analogues: Synthesis and Characterization of New Triazole-Derived Annulenes Containing Six Heterocyclic Subunits. Chem. - Eur. J 2001, 7 (11), 2407–2413.

(223) Islyaikin, M. K.; Danilova, E. A.; Yagodarova, L. D.; Rodríguez-Morgade, M. S.; Torres, T. Thiadiazole-Derived Expanded Heteroazaporphyrinoids. Org. Lett. 2001, 3 (14), 2153–2156.

(224) Trukhina, O. N.; Rodriguez-Morgade, M. S.; Wolfrum, S.; Caballero, E.; Snejko, N.; Danilova, E. A.; Gutierrez-Puebla, E.; Islyaikin, M. K.; Guldi, D. M.; Torres, T. Scrutinizing the Chemical Nature and Photophysics of an Expanded Hemiporphyrazine: The Special Case of [30]Trithia- 2,3,5,10,12,13,15,20,22,23,25,30-Dodecaazahexaphyrin. J. Am. Chem. Soc. 2010, 132 (37), 12991–12999.

(225) Kripli, B.; Baráth, G.; Balogh-Hergovich, É.; Giorgi, M.; Simaan, A. J.; Párkányi, L.; Pap, J. S.; Kaizer, J.; Speier, G. Correlation between the SOD- like Activity of Hexacoordinate Iron(II) Complexes and Their Fe3+/Fe2+ Redox Potentials. Inorg. Chem. Commun. 2011, 14 (1), 205–209.

(226) Martić, G.; Engle, J. T.; Ziegler, C. J. Complexes of 1,3-Bis(2- Thiazolylimino)Isoindoline with Middle and Late First Row Transition Metals. Inorg. Chem. Commun. 2011, 14 (11), 1749–1752.

(227) Siegl, W. O. Metal Ion Activation of Nitriles. Syntheses of 1,3- Bis(Arylimino)Isoindolines. J. Org. Chem. 1977, 42 (11), 1872–1878.

(228) Siegl, W. O. Metal-Chelating 1,3-Bis(2′-Pyridylimino)Isoindolines. J. 264

Heterocycl. Chem. 1981, 18 (8), 1613–1618.

(229) Dang, J. D.; Bender, T. P. Boron, Aluminum, Gallium, and Indium Complexes of 1,3-Bis(2-Pyridylimino)Isoindoline (BPI). Inorg. Chem. Commun. 2013, 30, 147–151.

(230) Schrage, B. R.; Vitale, D.; Kelly, K. A.; Nemykin, V. N.; Herrick, R. S.; Ziegler, C. J. Binding a Meridional Ligand in a Facial Geometry: A Square Peg in a Round Hole. J. Organomet. Chem. 2020, 919, 121331.

(231) Tamil Selvi, P.; Stoeckli-Evans, H.; Palaniandavar, M. Synthesis, Structure and DNA Interaction of Cobalt(III) Bis-Complexes of 1,3-Bis(2- Pyridylimino)Isoindoline and 1,4,7-Triazacyclononane. J. Inorg. Biochem. 2005, 99 (11), 2110–2118.

(232) Pap, J. S.; Bányai, V.; Szilvási, D. S.; Kaizer, J.; Speier, G.; Giorgi, M. Influence of Meridional N3-Ligands on Supramolecular Assembling and Redox Behavior of Carboxylatocopper(II) Complexes. Inorg. Chem. Commun. 2011, 14 (11), 1767–1772.

(233) Varadi, T.; Pap, J. S. J. S.; Giorgi, M.; Parkanyi, L.; Csay, T. T. T.; Speier, G. G.; Kaizer, J. J. J.; Váradi, T.; Pap, J. S. J. S.; Giorgi, M.; et al. Iron(III) Complexes with Meridional Ligands as Functional Models of Intradiol- Cleaving Catechol Dioxygenases. Inorg. Chem. 2013, 52 (3), 1559–1569.

(234) Bereman, R. D.; Shields, G. D.; Dorfman, J. R.; Bordner, J. Stereoelectronic Properties of Metalloenzymes. 10. a Refined Model That Mimics the Type II Copper(II) Site in Galactose Oxidase. J. Inorg. Biochem. 1983, 19 (1), 75– 93.

(235) Tseng, K.-N. T.; Kampf, J. W.; Szymczak, N. K. Base-Free, Acceptorless, and Chemoselective Alcohol Dehydrogenation Catalyzed by an Amide- Derived NNN-Ruthenium(II) Hydride Complex. Organometallics 2013, 32 (7), 2046–2049.

(236) Gagne, R. R.; Marks, D. N. Ruthenium Complexes of 1,3-Bis(2- Pyridylimino)Isoindolines as Alcohol Oxidation Catalysts. Inorg. Chem. 1984, 23 (1), 65–74.

(237) Bakthavachalam, K.; Reddy, N. D. Synthesis of Aluminum Complexes of Triaza Framework Ligands and Their Catalytic Activity toward 265

Polymerization of ε-Caprolactone. Organometallics 2013, 32 (11), 3174– 3184.

(238) Mueller, A. L.; Wadepohl, H.; Gade, L. H. Bis(Pyridylimino)Isoindolato (BPI) Osmium Complexes: Structural Chemistry and Reactivity. Organometallics 2015, 34 (12), 2810–2818.

(239) Sauer, D. C.; Melen, R. L.; Kruck, M.; Gade, L. H. Chromophores, Fluorophores and Robust Ancillary Ligands for Molecular Catalysts: 1,3- Bis(2-Pyridylimino)Isoindolines. Eur. J. Inorg. Chem. 2014, 2014 (28), 4715- N4725.

(240) Kleeberg, C.; Broering, M. Palladium Complexes of Unsymmetrical 1,3- Bis(Arylimino)Isoindoline (BAI) Ligands: A Novel Class of Complexes Exhibiting Unusual Structural Features. Polyhedron 2010, 29 (1), 507–513.

(241) Donyagina, V. F.; Shimizu, S.; Kobayashi, N.; Lukyanets, E. A. Synthesis of N,N-Difluoroboryl Complexes of 3,3′-Diarylazadiisoindolylmethenes. Tetrahedron Lett. 2008, 49 (42), 6152–6154.

(242) Crandall, L. A.; Rhoda, H. M.; Nemykin, V. N.; Ziegler, C. J. Boron Templated Synthesis of a BODIPY Analogue from a Phthalocyanine Precursor. New J. Chem. 2016, 40 (7), 5675–5678.

(243) Crandall, L. A.; Bogdanowicz, C. A.; Hasheminasab, A.; Chanawanno, K.; Herrick, R. S.; Ziegler, C. J. Re(CO)3-Templated Formation of Aza(Dibenzo)Dipyrromethenes. Inorg. Chem. 2016, 55 (7), 3209–3211.

(244) Broering, M.; Kleeberg, C.; Koehler, S. Palladium(II) Complexes of Unsymmetrical CNN Pincer Ligands. Inorg. Chem. 2008, 47 (14), 6404– 6412.

(245) Deng, Q.-H.; Wadepohl, H.; Gade, L. H. The Synthesis of a New Class of Chiral Pincer Ligands and Their Applications in Enantioselective Catalytic Fluorinations and the Nozaki–Hiyama–Kishi Reaction. Chem. – A Eur. J. 2011, 17 (52), 14922–14928.

(246) Bleith, T.; Wadepohl, H.; Gade, L. H. Iron Achieves Noble Metal Reactivity and Selectivity: Highly Reactive and Enantioselective Iron Complexes as Catalysts in the Hydrosilylation of Ketones. J. Am. Chem. Soc. 2015, 137 (7), 2456–2459. 266

(247) Blasius, C. K.; Wadepohl, H.; Gade, L. H. NNN-Cobalt(II) Pincer Complexes: Paramagnetic NMR Spectroscopy in Solution and Application as Hydrosilylation Catalysts. Eur. J. Inorg. Chem. 2020, 2020 (24), 2335–2342.

(248) Andrew, J. E.; Ball, P. W.; Blake, A. B. Binuclear Cobalt(II) and Nickel(II) Complexes of Dihydrazinophthalazine and Dipyridylpyridazine: The Dependence of Superexchange on Electronic Configuration. J. Chem. Soc. D Chem. Commun. 1969, No. 4, 143–144.

(249) Thompson, L. K.; Chacko, V. T.; Elvidge, J. A.; Lever, A. B. P.; Parish, R. V. Binuclear Copper(II) Complexes of the Tetradentate Ligand 1,4-Di(2’- Pyridyl)Aminophthalazine: Some Novel Five-Coordinate Derivatives. Can. J. Chem. 1969, 47 (22), 4141–4152.

(250) Szávuly, M.; Szilvási, S. D.; Csonka, R.; Klesitz, D.; Speier, G.; Giorgi, M.; Kaizer, J. Catalytic Oxidation of Alcohols and Sulfides with Hydrogen Peroxide Using Isoindoline and Phthalazine-Based Diiron Complexes. J. Mol. Catal. A Chem. 2014, 393, 317–324.

(251) Thompson, L. K.; Hanson, A. W.; Ramaswamy, B. S. Magnetic Interactions in Copper Complexes of Tetradentate Binucleating Phthalazine Ligands. Crystal and Molecular Structure of Binuclear .Mu.-Hydroxo-Bridged Copper(II) Complexes of 1,4-Bis(2-Pyridylamino)Phthalazine and 1,4-Bis((4- Methyl-2-Pyridyl)Amino). Inorg. Chem. 1984, 23 (16), 2459–2465.

(252) Gaire, S.; Ortiz, R. J.; Schrage, B. R.; Lozada, I. B.; Mandapati, P.; Osinski, A. J.; Herbert, D. E.; Ziegler, C. J. (8-Amino)Quinoline and (4- Amino)Phenanthridine Complexes of Re(CO)3 Halides. J. Organomet. Chem. 2020, 921, 121338.

(253) Yuvaraj, H.; Gayathri, D.; Kalkhambkar, R. G.; Gupta, V. K. (E)-Ethyl 2- Cyano-3-(1H-Pyrrol-2-Yl)Acrylate. Acta Crystallogr. Sect. E Struct. Reports Online 2011, 67 (8), o2135.

(254) Tandon, S. S.; Thompson, L. K.; Miller, D. O. A Novel Tetranuclear Copper(II) Complex with Three Different Types of Azide, Exhibiting Weak Antiferromagnetic Exchange. J. Chem. Soc. Chem. Commun. 1995, No. 18, 1907–1908.

(255) Tandon, S. S.; Mandal, S. K.; Thompson, L. K.; Hynes, R. C. Tetranuclear, Spin-Coupled, Copper(II) Complexes of a Novel Octadentate Bis-Diazine

267

Ligand. J. Chem. Soc. Chem. Commun. 1991, No. 21, 1572–1573.

(256) Chakrabarti, M. H.; Hajimolana, S. A.; Mjalli, F. S.; Saleem, M.; Mustafa, I. Redox Flow Battery for Energy Storage. Arab. J. Sci. Eng. 2013, 38 (4), 723– 739.

(257) Janoschka, T.; Hager, M. D.; Schubert, U. S.; Janoschka, T.; Hager, M. D.; Schubert, U. S.; Martin, N. An Aqueous Redox-Flow Battery with High Capacity and Power: The TEMPTMA/MV System. Angew. Chem. Int. Ed. Engl. 2016, 55 (46), 14427–14430.

(258) Rajpurohit, J.; Upadhyay, A.; Das, C.; Dubey, R.; Vaidya, S.; Krishnan, V.; Kumar, A.; Shanmugam, M. Unusual Methylenediolate Bridged Hexanuclear Ruthenium(III) Complexes: Syntheses and Their Application. Inorg. Chem. 2018, 57 (23), 14967–14982.

(259) Petty II, A.; Mann, S.; Dumitrascu, A.; Olson, K.; Guarr, T. F. Multifunctional Pyridinium Systems for Nonaqueous Redox Flow Batteries. ECS Trans. 2017, 80 (10, Selected Proceedings: ECS Meeting, 232nd, National Harbor, 2017), 1241–1255.

(260) Duarte, G. M.; Braun, J. D.; Giesbrecht, P. K.; Herbert, D. E. Redox Non- Innocent Bis(2,6-Diimine-Pyridine) Ligand-Iron Complexes as Anolytes for Flow Battery Applications. Dalton Trans. 2017, 46 (47), 16439–16445.

(261) Shimizu, A.; Takenaka, K.; Handa, N.; Nokami, T.; Itoh, T.; Yoshida, J. Liquid Quinones for Solvent-Free Redox Flow Batteries. Adv. Mater. 2017, 29 (41), 1606592.

(262) Navalpotro, P.; Palma, J.; Anderson, M.; Marcilla, R. A Membrane-Free Redox Flow Battery with Two Immiscible Redox Electrolytes. Angew. Chemie, Int. Ed. 2017, 56 (41), 12460–12465.

(263) Li, Y.; Sniekers, J.; Malaquias, J.; Li, X.; Schaltin, S.; Stappers, L.; Binnemans, K.; Fransaer, J.; Vankelecom, I. F. J. A Non-Aqueous All- Copper Redox Flow Battery with Highly Soluble Active Species. Electrochim. Acta 2017, 236, 116–121.

(264) Armstrong, C. G.; Toghill, K. E. Cobalt(II) Complexes with Azole-Pyridine Type Ligands for Non-Aqueous Redox-Flow Batteries: Tunable Electrochemistry via Structural Modification. J. Power Sources 2017, 349, 268

121–129.

(265) Zhao, Y.; Ding, Y.; Li, Y.; Peng, L.; Byon, H. R.; Goodenough, J. B.; Yu, G. A Chemistry and Material Perspective on Lithium Redox Flow Batteries towards High-Density Electrical Energy Storage. Chem. Soc. Rev. 2015, 44 (22), 7968–7996.

(266) Hwang, B.; Park, M.-S. S.; Kim, K. Ferrocene and Cobaltocene Derivatives for Non-Aqueous Redox Flow Batteries. ChemSusChem 2015, 8 (2), 310– 314.

(267) Friedl, J.; Lebedeva, M. A.; Porfyrakis, K.; Stimming, U.; Chamberlain, T. W. All-Fullerene-Based Cells for Nonaqueous Redox Flow Batteries. J. Am. Chem. Soc. 2018, 140 (1), 401–405.

(268) Schrage, B. R.; Zhao, Z.; Ziegler, C. J.; Boika, A. Cation-Anion Redox Switching in an All-Ferrocene Salt. ChemElectroChem 2018, 5 (23), 3624– 3627.

(269) Chanawanno, K.; Holstrom, C.; Crandall, L. A.; Dodge, H.; Nemykin, V. N.; Herrick, R. S.; Ziegler, C. J. The Synthesis and Structures of 1,1′- Bis(Sulfonyl)Ferrocene Derivatives. Dalton Trans. 2016, 45 (36), 14320– 14326.

(270) Sheldrick, G. M. A Short History of SHELX. Acta Crystallogr. Sect. A Found. Crystallogr. 2008, 64 (1), 112–122.

(271) Wen, Y.; Cheng, Z.; Li, W.; Li, Z.; Liao, D.; Hu, X.; Pan, N.; Wang, D.; Hull, T. R. A Novel Oligomer Containing DOPO and Ferrocene Groups: Synthesis, Characterization, and Its Application in Fire Retardant Epoxy Resin. Polym. Degrad. Stab. 2018, 156, 111–124.

(272) Chernyy, S.; Wang, Z.; Kirkensgaard, J. J. K.; Bakke, A.; Mortensen, K.; Ndoni, S.; Almdal, K. Synthesis and Characterization of Ferrocene Containing Block Copolymers. J. Polym. Sci. Part A Polym. Chem. 2017, 55 (3), 495–503.

(273) Peng, X.; He, H.; Xia, J.; Lou, Z.; Chang, G.; Zhang, X.; Wang, S. Visual Detection and Removal of Mercury Ions by a Ferrocene Derivative. Tetrahedron Lett. 2014, 55 (25), 3541–3544.

269

(274) Radulovic, N. S.; Mladenovic, M. Z.; Stojanovic-Radic, Z.; Bogdanovic, G. A.; Stevanovic, D.; Vukicevic, R. D. Synthesis, Characterization, and Antimicrobial Evaluation of a Small Library of Ferrocene-Containing Acetoacetates and Phenyl Analogs: The Discovery of a Potent Anticandidal Agent. Mol. Divers. 2014, 18 (3), 497–510.

(275) Federman Neto, A.; Miller, J.; Faria de Andrade, V.; Fujimoto, S. Y.; Afonso, M. M. D. F.; Archanjo, F. C.; Darin, V. A.; Andrade e. Silva, M. L.; Borges, A. D. L.; Del Ponte, G. New Synthesis of Ferrocene Monocarboxylic Acid and Systematic Studies on the Preparation of Related Key-Intermediates. Zeitschrift fuer Anorg. und Allg. Chemie 2002, 628 (1), 209–216.

(276) Sahoo, P.; Puranik, V. G.; Patra, A. K.; Sastry, P. U.; Dastidar, P. Ferrocene Based Organometallic Gelators: A Supramolecular Synthon Approach. Soft Matter 2011, 7 (7), 3634–3641.

(277) Blake, A. J.; Caltagirone, C.; Lippolis, V.; Schroeder, M.; Wilson, C. (Ferrocenylmethyl)Trimethylammonium Triiodide. Acta Crystallogr. Sect. E Struct. Reports Online 2004, 60 (1), m20–m21.

(278) Xie, J.; Abrahams, B. F.; Zimmermann, T. J.; Mukherjee, A.; Wedd, A. G. Ferrocene Mono- and Di-Sulfonates as Building Blocks in Hydrogen-Bonded Networks. Aust. J. Chem. 2007, 60 (8), 578–582.

(279) Ferguson, G.; Gallagher, J. F.; Glidewell, C.; Zakaria, C. M. Structures of (Ferrocenylmethyl)Trimethylammonium Iodide and Hexa-N- Methylferrocene-1,1’-Diylbis(Methylammonium Iodide). Acta Crystallogr. Sect. B Struct. Sci. 1994, B50 (2), 146–150.

(280) Zakaria, C. M.; Ferguson, G.; Lough, A. J.; Glidewell, C. Ferrocene-1,1’- Dicarboxylic Acid as a Building Block in Supramolecular Chemistry: Supramolecular Structures in One, Two and Three Dimensions. Acta Crystallogr. Sect. B Struct. Sci. 2002, B58 (5), 786–802.

(281) Cooke, G.; Duclairoir, F. M. A.; Kraft, A.; Rosair, G.; Rotello, V. M. Pronounced Stabilization of the Ferrocenium State of Ferrocenecarboxylic Acid by Salt Bridge Formation with a Benzamidine. Tetrahedron Lett. 2004, 45 (3), 557–560.

(282) Russell, V. A.; Ward, M. D. Two-Dimensional Hydrogen-Bonded Assemblies: The Influence of Sterics and Competitive Hydrogen Bonding on

270

the Structures of Guanidinium Arenesulfonate Networks. J. Mater. Chem. 1997, 7 (7), 1123–1133.

(283) Wang, Y.; Wang, Y. Supporting Information. Acta Crystallogr. Sect. E 2012, 68, 197.

(284) Reynes, O.; Moutet, J.-C.; Pecaut, J.; Royal, G.; Saint-Aman, E. (Ferrocenylmethyl)Trimethylammonium Cation: A Very Simple Probe for the Electro-Chemical Sensing of Dihydrogen Phosphate and ATP Anions. New J. Chem. 2002, 26 (1), 9–12.

(285) Tustin, G. J.; Lafitte, V. G. H.; Banks, C. E.; Jones, T. G. J.; Smith, R. B.; Davis, J.; Lawrence, N. S. Synthesis and Characterisation of Water Soluble Ferrocenes: Molecular Tuning of Redox Potentials. J. Organomet. Chem. 2007, 692 (23), 5173–5182.

(286) Lawrence, N. S.; Tustin, G. J.; Faulkner, M.; Jones, T. G. J. Ferrocene Sulfonates as Electrocatalysts for Sulfide Detection. Electrochim. Acta 2006, 52 (2), 499–503.

(287) Bard, A. J.; Faulkner, L. R. Electrochemical Methods : Fundamentals and Applications, 2nd ed.; John Wiley: New York ; Chichester [England], 2001.

(288) Singh, A.; Chowdhury, D. R.; Paul, A. A Kinetic Study of Ferrocenium Cation Decomposition Utilizing an Integrated Electrochemical Methodology Composed of Cyclic Voltammetry and Amperometry. Anal. (Cambridge, United Kingdom) 2014, 139 (22), 5747–5754.

(289) Vijh, A. K.; Conway, B. E. Electrode Kinetic Aspects of the Kolbe Reaction. Chem. Rev. 1967, 67 (6), 623–664.

(290) David R. Lide, E. CRC Handbook of Chemistry and Physics; CRC Press, Ed.; Boca Raton, 2005.

(291) Boyle, G. Renewable Energy.; Oxford University Press, 2004.

(292) Wang, W.; Luo, Q.; Li, B.; Wei, X.; Li, L.; Yang, Z. Recent Progress in Redox Flow Battery Research and Development. Adv. Funct. Mater. 2013, 23 (8), 970–986.

271

(293) Skyllas-Kazacos, M.; Chakrabarti, M. H.; Hajimolana, S. A.; Mjalli, F. S.; Saleem, M. Progress in Flow Battery Research and Development. J. Electrochem. Soc. 2011, 158 (8), R55–R79.

(294) Marom, R.; Amalraj, S. F.; Leifer, N.; Jacob, D.; Aurbach, D. A Review of Advanced and Practical Lithium Battery Materials. J. Mater. Chem. 2011, 21 (27), 9938–9954.

(295) Parasuraman, A.; Lim, T. M.; Menictas, C.; Skyllas-Kazacos, M. Review of Material Research and Development for Vanadium Redox Flow Battery Applications. Electrochim. Acta 2013, 101, 27–40.

(296) Perry, M. L.; Weber, A. Z. Advanced Redox-Flow Batteries: A Perspective. J. Electrochem. Soc. 2016, 163 (1), A5064–A5067.

(297) Liu, W.; Lu, W.; Zhang, H.; Li, X. Aqueous Flow Batteries: Research and Development. Chem. Eur. J. 2019, 25 (7), 1649–1664.

(298) Skyllas‐Kazacos, M.; Grossmith, F. Efficient Vanadium Redox Flow Cell. J. Electrochem. Soc. 1987, 134 (12), 2950–2953.

(299) Gong, K.; Xu, F.; Grunewald, J. B.; Ma, X.; Zhao, Y.; Gu, S.; Yan, Y. All- Soluble All-Iron Aqueous Redox-Flow Battery. ACS Energy Lett. 2016, 1 (1), 89–93.

(300) Lin, K.; Chen, Q.; Gerhardt, M. R.; Tong, L.; Kim, S. B.; Eisenach, L.; Valle, A. W.; Hardee, D.; Gordon, R. G.; Aziz, M. J.; et al. Alkaline Quinone Flow Battery. Science 2015, 349 (6255), 1529–1532.

(301) Kwabi, D. G.; Lin, K.; Ji, Y.; Kerr, E. F.; Goulet, M.-A.; De Porcellinis, D.; Tabor, D. P.; Pollack, D. A.; Aspuru-Guzik, A.; Gordon, R. G. Alkaline Quinone Flow Battery with Long Lifetime at pH 12. Joule 2018, 2 (9), 1894– 1906.

(302) Yan, W.; Wang, C.; Tian, J.; Zhu, G.; Ma, L.; Wang, Y.; Chen, R.; Hu, Y.; Wang, L.; Chen, T.; et al. All-Polymer Particulate Slurry Batteries. Nat. Commun. 2019, 10 (1), 1–11.

(303) Wang, C.; Li, X.; Yu, B.; Wang, Y.; Yang, Z.; Wang, H.; Lin, H.; Ma, J.; Li, G.; Jin, Z. Molecular Design of Fused-Ring Phenazine Derivatives for Long- Cycling Alkaline Redox Flow Batteries. ACS Energy Lett. 2020, 5 (2), 411– 272

417.

(304) Wang, C.; Yang, Z.; Wang, Y.; Zhao, P.; Yan, W.; Zhu, G.; Ma, L.; Yu, B.; Wang, L.; Li, G.; et al. High-Performance Alkaline Organic Redox Flow Batteries Based on 2-Hydroxy-3-Carboxy-1,4-Naphthoquinone. ACS Energy Lett. 2018, 3 (10), 2404–2409.

(305) Beh, E. S.; De Porcellinis, D.; Gracia, R. L.; Xia, K. T.; Gordon, R. G.; Aziz, M. J. A Neutral PH Aqueous Organic–Organometallic Redox Flow Battery with Extremely High Capacity Retention. ACS Energy Lett. 2017, 2 (3), 639– 644.

(306) Bus, J. S.; Gibson, J. E. Paraquat: Model for Oxidant-Initiated Toxicity. Environ. Health Perspect. 1984, 55, 37–46.

(307) Tanner, C. M.; Kamel, F.; Ross, G. W.; Hoppin, J. A.; Goldman, S. M.; Korell, M.; Marras, C.; Bhudhikanok, G. S.; Kasten, M.; Chade, A. R. Rotenone, Paraquat, and Parkinson’s Disease. Environ. Health Perspect. 2011, 119 (6), 866–872.

(308) Schrage, B. R.; Zhao, Z.; Ziegler, C. J.; Boika, A. Cation-Anion Redox Switching in an All-Ferrocene Salt. ChemElectroChem 2018, 5 (23).

(309) Melchiorri, D.; Reiter, R. J.; Sewerynek, E.; Hara, M.; Chen, L.; Nisticò, G. Paraquat Toxicity and Oxidative Damage: Reduction by Melatonin. Biochem. Pharmacol. 1996, 51 (8), 1095–1099.

(310) Montoya, S. C. N.; Agnese, A. M.; Pérez, C.; Tiraboschi, I. N.; Cabrera, J. L. Pharmacological and Toxicological Activity of Heterophyllaea Pustulata Anthraquinone Extracts. Phytomedicine 2003, 10 (6–7), 569–574.

(311) Kelsall, G. H.; Thompson, I. Redox Chemistry of H 2 S Oxidation by the British Gas Stretford Process Part III: Electrochemical Behaviour of Anthraquinone 2, 7 Disulphonate in Alkaline Electrolytes. J. Appl. Electrochem. 1993, 23 (4), 296–307.

(312) Schrage, B. R.; Zhao, Z.; Boika, A.; Ziegler, C. J. Evaluating Ferrocene Ions and All-Ferrocene Salts for Electrochemical Applications. J. Organomet. Chem. 2019, 897, 23–31.

(313) Liu, J.; Bo, X.; Li, M.; Yin, D.; Guo, L. Contrastive Study on Porphyrinic Iron 273

Metal-Organic Framework Supported on Various Carbon Matrices as Efficient Electrocatalysts. J. Colloid Interface Sci. 2018, 513, 438–447.

(314) Cuartero, M.; Acres, R. G.; Bradley, J.; Jarolimova, Z.; Wang, L.; Bakker, E.; Crespo, G. A.; De Marco, R. Electrochemical Mechanism of Ferrocene- Based Redox Molecules in Thin Film Membrane Electrodes. Electrochim. Acta 2017, 238, 357–367.

(315) Sato, T.; Ambe, F.; Endo, K.; Katada, M.; Maeda, H.; Nakamoto, T.; Sano, H. Mixed-Valence States of [Fe3O(CH2XCO2)6(H2O)3]·nH2O (X = H, Cl, and Br) Characterized by X-Ray Crystallography and 57Fe-Moessbauer Spectroscopy. J. Am. Chem. Soc. 1996, 118 (14), 3450–3458.

(316) Sauer, D. C.; Wadepohl, H.; Gade, L. H. Cobalt Alkyl Complexes of a New Family of Chiral 1,3-Bis(2-Pyridylimino)Isoindolates and Their Application in Asymmetric Hydrosilylation. Inorg. Chem. 2012, 51 (23), 12948–12958.

(317) Zhang, P.; Liao, H.; Wang, H.; Li, X.; Yang, F.; Zhang, S. Cis-1,4- Polymerization of Isoprene Catalyzed by 1,3-Bis(2-Pyridylimino)Isoindoline- Ligated Rare-Earth-Metal Dialkyl Complexes. Organometallics 2017, 36 (13), 2446–2451.

(318) Szavuly, M.; Szilvasi, S. D.; Csonka, R.; Klesitz, D.; Speier, G.; Giorgi, M.; Kaizer, J. Catalytic Oxidation of Alcohols and Sulfides with Hydrogen Peroxide Using Isoindoline and Phthalazine-Based Diiron Complexes. J. Mol. Catal. A Chem. 2014, 393, 317–324.

(319) Csay, T.; Kripli, B.; Giorgi, M.; Kaizer, J.; Speier, G. A Flexible Hydroxy- Bridged Dicopper Complex as Catechol Oxidase Mimic. Inorg. Chem. Commun. 2010, 13 (2), 227–230.

(320) Kaizer, J.; Barath, G.; Csonka, R.; Speier, G.; Korecz, L.; Rockenbauer, A.; Parkanyi, L. Catechol Oxidase and Phenoxazinone Synthase Activity of a Manganese(II) Isoindoline Complex. J. Inorg. Biochem. 2008, 102 (4), 773– 780.

(321) Kaizer, J.; Kripli, B.; Speier, G.; Párkányi, L. Synthesis , Structure , and Catalase-like Activity of a Novel Manganese ( II ) Complex : Polyhedron 2009, 28 (5), 933–936.

(322) Pap, J. S.; Kripli, B.; Banyai, V.; Giorgi, M.; Korecz, L.; Gajda, T.; Arus, D.; 274

Kaizer, J.; Speier, G. Tetra-, Penta- and Hexacoordinate Copper(II) Complexes with N3 Donor Isoindoline-Based Ligands: Characterization and SOD-like Activity. Inorganica Chim. Acta 2011, 376 (1), 158–169.

(323) Balogh-Hergovich, E.; Kaizer, J.; Speier, G.; Huttner, G.; Jacobi, A. Preparation and Oxygenation of (Flavonolato)Copper Isoindoline Complexes with Relevance to Quercetin Dioxygenase. Inorg. Chem. 2000, 39 (19), 4224–4229.

(324) Kaizer, J.; Kripli, B.; Speier, G.; Parkanyi, L. Synthesis, Structure, and Catalase-like Activity of a Novel Manganese(II) Complex: Dichloro[1,3- Bis(2’-Benzimidazolylimino)Isoindoline]Manganese(II). Polyhedron 2009, 28 (5), 933–936.

(325) Bautista, D. V; Dewan, J. C.; Thompson, L. K. Copper Complexes of Polyfunctional Phthalazines. Mononuclear and Polynuclear Derivatives. Can. J. Chem. 1982, 60 (20), 2583–2593.

(326) Tao, J.; Perdew, J. P.; Staroverov, V. N.; Scuseria, G. E. Climbing the Density Functional Ladder: Nonempirical Meta-Generalized Gradient Approximation Designed for Molecules and Solids. Phys. Rev. Lett. 2003, 91 (14), 146401/1-146401/4.

(327) Tomasi, J.; Mennucci, B.; Cammi, R. Quantum Mechanical Continuum Solvation Models. Chem. Rev. 2005, 105 (8), 2999–3093.

(328) Fuentealba, P.; Preuss, H.; Stoll, H.; Von Szentpaly, L. A Proper Account of Core-Polarization with Pseudopotentials: Single Valence-Electron Alkali Compounds. Chem. Phys. Lett. 1982, 89 (5), 418–422.

(329) Andrae, D.; Haeussermann, U.; Dolg, M.; Stoll, H.; Preuss, H. Energy- Adjusted Ab Initio Pseudopotentials for the Second and Third Row Transition Elements. Theor. Chim. Acta 1990, 77 (2), 123–141.

(330) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, Ma.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A. Gaussian 09, Revision D. 01, Gaussian. Inc. Wallingford, CT 2009.

(331) Tenderholt, A. L. QMForge, Version 2.4.

(332) Osinski, A. J.; Morris, D. L.; Herrick, R. S.; Ziegler, C. J. Re(CO)3-Templated 275

Synthesis of α-Amidinoazadi(Benzopyrro)Methenes. Inorg. Chem. 2017, 56 (24), 14734–14737.

(333) Hasheminasab, A.; Engle, J. T.; Bass, J.; Herrick, R. S.; Ziegler, C. J. The Synthesis of Dimeric ReI-Phenylenediimine Conjugates: Spectroscopic and Electrochemical Studies. Eur. J. Inorg. Chem. 2014, 2014 (16), 2643–2652.

(334) Ng, C.-O.; Cheng, S.-C.; Chu, W.-K.; Tang, K.-M.; Yiu, S.-M.; Ko, C.-C. Luminescent Rhenium(I) Pyridyldiaminocarbene Complexes: Photophysics, Anion-Binding, and CO2-Capturing Properties. Inorg. Chem. 2016, 55 (16), 7969–7979.

(335) Chanawanno, K.; Rhoda, H. M.; Hasheminasab, A.; Crandall, L. A.; King, A. J.; Herrick, R. S.; Nemykin, V. N.; Ziegler, C. J. Using Hydrazine to Link Ferrocene with Re(CO)3: A Modular Approach. J. Organomet. Chem. 2016, 818, 145–153.

(336) Schrage, B. R.; Herrick, R. S.; Ziegler, C. J. The Synthesis and Characterization of Re(CO)3 Pyca-Anthraquinone Conjugates. J. Organomet. Chem. 2019, 880, 170–174.

(337) Gomez-Iglesias, P.; Guyon, F.; Khatyr, A.; Ulrich, G.; Knorr, M.; Martin- Alvarez, J. M.; Miguel, D.; Villafane, F. Luminescent Rhenium(I) Tricarbonyl Complexes with Pyrazolylamidino Ligands: Photophysical, Electrochemical, and Computational Studies. Dalton Trans. 2015, 44 (40), 17516–17528.

(338) Kadish, K. M.; Smith, K. M.; Guilard, R.; Editors. Handbook of Porphyrin Science with Applications to Chemistry, Physics, Materials Science, Engineering, Biology and Medicine: Volume 20; Chlorophylls and Related Systems.; World Scientific Publishing Co. Pte. Ltd., 2012.

(339) Clark, P. F.; Elvidge, J. A.; Linstead, R. P. 722. Heterocyclic Imines and Amines. Part II. Derivatives of Iso Indoline and Iso Indolenine. J. Chem. Soc. 1953, 3593.

(340) Clark, P. F.; Elvidge, J. A.; Golden, J. H. 799. Heterocyclic Imines and Amines. Part VII. N-Substituted Phthalic Imidine Derivatives and Their Reactions with Amines. J. Chem. Soc. 1956, 4135.

(341) Elvidge, J. A.; Golden, J. H. 138. Conjugated Macrocycles. Part XXVIII. Adducts from Di-Iminoisoindoline and Arylene-m-Diamines, and a New Type 276

of Crossconjugated Macrocycle with Three-Quarters of the Chromophore of Phthalocyanine. J. Chem. Soc. 1957, 700.

(342) de la Torre, G.; Torres, T. Stepwise Synthesis of Substituted Dicyanotriazolehemiporphyrazines. A Regioselective Approach to Unsymmetrically Substituted Hemiporphyrazines. J. Org. Chem. 1996, 61 (18), 6446–6449.

(343) Dini, D.; Calvete, M. J. F.; Hanack, M.; Amendola, V.; Meneghetti, M. Large Two-Photon Absorption Cross Sections of Hemiporphyrazines in the Excited State: The Multiphoton Absorption Process of Hemiporphyrazines with Different Central Metals. J. Am. Chem. Soc. 2008, 130 (37), 12290–12298.

(344) Kobayashi, N.; Inagaki, S.; Nemykin, V. N.; Nonomura, T. A Novel Hemiporphyrazine Comprising Three Isoindolediimine and Three Thiadiazole Units. Angew. Chemie, Int. Ed. 2001, 40 (14), 2710–2712.

(345) Nemykin, V. N.; Lukyanets, E. A. Synthesis of Substituted Phthalocyanines. ARKIVOC 2010, No. 1, 136.

(346) Nemykin, V. N.; Dudkin, S. V; Dumoulin, F.; Hirel, C.; Gurek, A. G.; Ahsen, V. Synthetic Approaches to Asymmetric Phthalocyanines and Their Analogues. ARKIVOC 2014, No. I, 142.

(347) Lukyanets, E. A.; Nemykin, V. N. The Key Role of Peripheral Substituents in the Chemistry of Phthalocyanines and Their Analogs. J. Porphyrins Phthalocyanines 2010, 14, 1.

(348) Tamgho, I.-S.; Crandall, L. A.; Engle, J. T.; Ziegler, C. J. The Synthesis of a Silver Hexagon Using a Diiminoisoindoline-Based Ligand. Inorg. Chem. Commun. 2017, 76, 122–124.

(349) Tamgho, I.-S.; Engle, J. T.; Ziegler, C. J. Phthalocrowns: Isoindoline-Crown Ether Macrocycles. J. Org. Chem. 2012, 77 (24), 11372–11376.

(350) Igarashi, T.; Watanabe, K.; Ichijima, S.; Ise, T. Organic Electroluminescent Device and Platinum Compound., November 18, 2004.

(351) Engle, J. T.; Martic, G.; Ziegler, C. J. Investigations into the Coordination Chemistry of 1,3-Bis(2’-Benzimidazolylimino)Isoindoline. Makrogeterotsikly 2013, 6 (4), 353–359. 277

(352) Lever, A. B. P.; Ramaswamy, B. S.; Pickens, S. R. Catecholase Activity in a Model Binuclear Copper Phthalazine Derivative. Inorganica Chim. Acta 1980, 46 (3), L59–L61.

(353) Thompson, L. K.; Niel, V.; Grove, H.; Miller, D. O.; Newlands, M. J.; Bird, P. H.; Wickramasinghe, W. A.; Lever, A. B. P. Cobalt and Nickel Complexes of a Tetradentate Phthalazine Ligand - Mononuclear, Dinuclear and Trinuclear Derivatives. Polyhedron 2004, 23 (7), 1175–1184.

(354) Lever, A. B. P.; Thompson, L. K.; Reiff, W. M. Antiferromagnetic Exchange Interactions and Delocalization in Some Tetrahedral Binuclear Cobalt(II) Complexes of Phthalazine and Pyrazine Ligands. Inorg. Chem. 1972, 11 (1), 104–109.

(355) Lever, A. B. P.; Thompson, L. K.; Reiff, W. M. Caveat on Exchange Interactions. Reply to Comments. Inorg. Chem. 1972, 11 (9), 2292.

(356) Wang, G.; Wang, F.; Chen, L.; Wang, L.; Tian, H. Process for Preparation of C.I. Pigment Yellow 139., July 18, 2012.

(357) Hainz, R.; Allaz, J.; Schroeder, D. Isometric Isoindoline Yellow Pigment., December 13, 2011.

(358) Oda, K.; Togai, M.; Hamaki, H. Coloring Composition for Photoresist with Low Phase Difference., March 12, 2018.

(359) Yauchi, H. Organic Transistors Provided with Isoindole Derivative Semiconductor Materials., May 1, 2008.

(360) Shibata, T.; Suzuki, T.; Hayata, H.; Kinoshita, A. Electrophotographic Photoreceptor with Good Repeating Durability., June 20, 1997.

(361) Lv, D.; Chen, D.; Zhan, X.; Chen, Z. Method for Preparation of Isoindoline Pigment., February 24, 2010.

(362) Sheldrick, G. M. A Short History Of SHELX. Acta Crystallogr., Sect. A Found. Crystallogr. 2008, 64, 112.

(363) McLean, A. D.; Chandler, G. S. Contracted Gaussian Basis Sets for Molecular Calculations. I. Second Row Atoms, Z = 11-18. J. Chem. Phys. 278

1980, 72 (10), 5639–5648.

(364) Elvidge, J. A.; Fitt, J. S.; Linstead, R. P. 50. Heterocyclic Imines and Amines. Part VI. Condensation Products from Di-Imino Iso Indoline and Succinimidine with Cyanoacetates. J. Chem. Soc. 1956, 235.

(365) Makarova, E. A.; Zatsikha, Y. V; Newman, K. M. E.; Paidi, V. K.; Beletsky, V. A.; Lierop, J. Van; Lukyanets, E. A.; Nemykin, V. N.; van Lierop, J.; Lukyanets, E. A.; et al. Direct Synthesis of an Unprecedented Stable Radical of Nickel(II) 3,5-Bis(Dimedonyl)Azadiisoindomethene with Strong and Narrow Near-Infrared Absorption at λ ∼ 1000 Nm. Inorg. Chem. 2017, 56 (11), 6052–6055.

(366) Erk, P.; Hengelsberg, H.; Haddow, M. F.; van Gelder, R. The Innovative Momentum of Crystal Engineering. CrystEngComm 2004, 6, 474–483.

(367) Sessler, J. L.; Hemmi, G.; Mody, T. D.; Murai, T.; Burrell, A.; Young, S. W. Texaphyrins: Synthesis and Applications. Acc. Chem. Res. 1994, 27 (2), 43– 50.

(368) Preihs, C.; Arambula, J. F.; Magda, D.; Jeong, H.; Yoo, D.; Cheon, J.; Siddik, Z. H.; Sessler, J. L. Recent Developments in Texaphyrin Chemistry and Drug Discovery. Inorg. Chem. 2013, 52 (21), 12184–12192.

(369) Bauer, V. J.; Clive, D. L. J.; Dolphin, D.; Paine III, J. B.; Harris, F. L.; King, M. M.; Loder, J.; Wang, S. W. C.; Woodward, R. B. Sapphyrins: Novel Aromatic Pentapyrrolic Macrocycles. J. Am. Chem. Soc. 1983, 105 (21), 6429–6436.

(370) Broadhurst, M. J.; Grigg, R.; Johnson, A. W. Synthesis of 22 π-Electron Macrocycles. Sapphyrins and Related Compounds. J. Chem. Soc. Perkin Trans. 1 Org. Bio-Organic Chem. 1972, No. 16, 2111–2116.

(371) Sessler, J. L.; Cyr, M. J.; Burrell, A. K. Sapphyrins: New Life for an Old “Expanded Porphyrin”. Synlett 1991, No. 3, 127–134.

(372) Saito, S.; Osuka, A. Expanded Porphyrins: Intriguing Structures, Electronic Properties, and Reactivities. Angew. Chemie, Int. Ed. 2011, 50 (19), 4342– 4373.

(373) Sankar, J.; Mori, S.; Saito, S.; Rath, H.; Suzuki, M.; Inokuma, Y.; Shinokubo, 279

H.; Suk Kim, K.; Yoon, Z. S.; Shin, J.-Y.; et al. Unambiguous Identification of Mobius Aromaticity for Meso-Aryl-Substituted [28]Hexaphyrins(1.1.1.1.1.1). J. Am. Chem. Soc. 2008, 130 (41), 13568–13579.

(374) Jux, N. The Porphyrin Twist: Hueckel and Moebius Aromaticity. Angew. Chemie, Int. Ed. 2008, 47 (14), 2543–2546.

(375) Tanaka, Y.; Saito, S.; Mori, S.; Aratani, N.; Shinokubo, H.; Shibata, N.; Higuchi, Y.; Yoon, Z. S.; Kim, K. S.; Noh, S. B.; et al. Metalation of Expanded Porphyrins: A Chemical Trigger Used to Produce Molecular Twisting and Mobius Aromaticity. Angew. Chem. Int. Ed. Engl. 2008, 47 (4), 681–684.

(376) Eckert, A. K.; Trukhina, O. N.; Rodriguez-Morgade, M. S.; Danilova, E. A.; Islyaikin, M. K.; Torres, T. Copper(II)-Template Synthesis of Hexaphyrin Meso-Hexaaza Analogues Containing Four Thiadiazole Moieties. Mendeleev Commun. 2010, 20 (4), 192–194.

(377) Cirera, B.; Bottari, G.; Martin-Jimenez, A.; Otero, R.; Miranda, R.; Torres, T.; Ecija, D.; Trukhina, O.; Bottari, G.; Torres, T.; et al. Long-Range Orientational Self-Assembly, Spatially Controlled Deprotonation, and Off- Centered Metalation of an Expanded Porphyrin. J. Am. Chem. Soc. 2017, 139 (40), 14129–14136.

(378) Danilova, E. A.; Melenchuk, T. V; Trukhina, O. N.; Zakharov, A. V; Islyaikin, M. K. Thiadiazole Containing Macrocyclic Compounds as Core-Modified Analogs of Phthalocyanine. Makrogeterotsikly 2010, 3 (1), 33–37.

(379) Elvidge, J. A.; Linstead, R. P. Conjugated Macrocycles. XXIV. A New Type of Cross-Conjugated Macrocycle, Related to the Azaporphines. J. Chem. Soc. 1952, 5008–5012.

(380) Trukhina, O. N.; Zhabanov, Y. A.; Krasnov, A. V; Danilova, E. A.; Islyaikin, M. K. Synthesis and Thermal Behavior of Unsubstituted [30]Trithia- 2,3,5,10,12,13,15,20,22,23,25,30-Dodecaazahexaphyrin. J. Porphyrins Phthalocyanines 2011, 15 (11–12), 1287–1291.

(381) Bernstein, J.; Stearns, B.; Shaw, E.; Lott, W. A. Derivatives of 2,6- Diaminopyridine. J. Am. Chem. Soc. 1947, 69, 1151–1158.

(382) Nardis, S. Synthetic Routes to Unsymmetrical Porphyrins. Top. Heterocycl. Chem. 2014, 33 (Synthesis and Modifications of Porphyrinoids), 203–229. 280

(383) Liu, D.; Cheng, L.; He, J.; Guo, Y.; Liu, Y. Research Advances in Synthesis and Properties of Porphyrin Derivatives. Huaxue Tongbao 2018, 81 (9), 780–791.

(384) Bruckner, C. The Breaking and Mending of Meso-Tetraarylporphyrins: Transmuting the Pyrrolic Building Blocks. Acc. Chem. Res. 2016, 49 (6), 1080–1092.

(385) Smith, K. M. Development of Porphyrin Syntheses. New J. Chem. 2016, 40 (7), 5644–5649.

(386) Mack, J.; Kobayashi, N. Low Symmetry Phthalocyanines and Their Analogues. Chem. Rev. (Washington, DC, United States) 2011, 111 (2), 281–321.

(387) Chen, J.; Zhu, C.; Xu, Y.; Zhang, P.; Liang, T. Advances in Phthalocyanine Compounds and Their Photochemical and Electrochemical Properties. Curr. Org. Chem. 2018, 22 (5), 485–504.

(388) Pacholska-Dudziak, E.; Gaworek, A.; Latos-Grazynski, L. Iron(II) Vacataporphyrins: A Variable Annulene Conformation inside a Regular Porphyrin Frame. Inorg. Chem. 2011, 50 (21), 10956–10965.

(389) Pacholska-Dudziak, E.; Szterenberg, L.; Latos-Grazynski, L. A Flexible Porphyrin-Annulene Hybrid: A Nonporphyrin Conformation for Meso- Tetraaryldivacataporphyrin. Chemistry 2011, 17 (12), 3500–3511.

(390) Pacholska-Dudziak, E.; Skonieczny, J.; Pawlicki, M.; Szterenberg, L.; Latos- Grazynski, L. Cadmium(II), Nickel(II), and Zinc(II) Complexes of Vacataporphyrin: A Variable Annulene Conformation inside a Standard Porphyrin Frame. Inorg. Chem. 2005, 44 (24), 8794–8803.

(391) Pacholska, E.; Latos-Grazynski, L.; Ciunik, Z. A Direct Link between Annulene and Porphyrin Chemistry--21-Vacataporphyrin. Chemistry 2002, 8 (23), 5403–5406.

(392) Schrage, B. R.; Chanawanno, K.; Crandall, L. A.; Ziegler, C. J. The Synthesis of a Hexameric Expanded Hemiporphyrazine. J. Porphyr. Phthalocyanines 2019, 24 (1/3), 1–6.

(393) Zatsikha, Y. V; Schrage, B. R.; Meyer, J.; Nemykin, V. N.; Ziegler, C. J. 1,3- 281

Diylideneisoindolines: Synthesis, Structure, Redox, and Optical Properties. J. Org. Chem. 2019, 84 (10), 6217–6222.

(394) Zatsikha, Y. V; Shamova, L. I.; Blesener, T. S.; Kuzmin, I. A.; Germanov, Y. V; Herbert, D. E.; Nemykin, V. N. Development of a Class of Easily Scalable, Electron-Deficient, Core-Extended Benzo-Fused Azadipyrromethene Derivatives. J. Org. Chem. 2019, 84 (22), 14540–14557.

(395) Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. Ab Initio Calculation of Vibrational Absorption and Circular Dichroism Spectra Using Density Functional Force Fields. J. Phys. Chem. 1994, 98 (45), 11623– 11627.

(396) Tenderholt, A. QMForge. Stanford University: Stanford, CA, USA.

(397) Hansen, P. E.; Spanget-Larsen, J. NMR and IR Investigations of Strong Intramolecular Hydrogen Bonds. Molecules 2017, 22 (4), 552/1-552/21.

(398) Parusel, A. B. J.; Ghosh, A. Density Functional Theory Based Configuration Interaction Calculations on the Electronic Spectra of Free-Base Porphyrin, Chlorin, Bacteriochlorin, and Cis- and Trans-Isobacteriochlorin. J. Phys. Chem. A 2000, 104 (11), 2504–2507.

(399) Baerends, E. J.; Ricciardi, G.; Rosa, A.; van Gisbergen, S. J. A. A DFT/TDDFT Interpretation of the Ground and Excited States of Porphyrin and Porphyrazine Complexes. Coord. Chem. Rev. 2002, 230 (1–2), 5–27.

(400) Shao, J.; Steene, E.; Hoffman, B. M.; Ghosh, A. EPR, ENDOR, and DFT Studies on (β-Octahalo-Meso-Tetraarylporphyrin)Copper Complexes: 2 2 Characterization of the Metal(Dx -Y )-Porphyrin(a2u) Orbital Interaction. Eur. J. Inorg. Chem. 2005, No. 8, 1609–1615.

(401) Zakharov, A. V; Stryapan, M. G.; Islyaikin, M. K. Structure, Electronic and Vibrational Spectra and Aromaticity of Hemiporphyrazine and Its Hydrates: A Density Functional Theory Study. J. Mol. Struct. THEOCHEM 2009, 906 (1–3), 56–62.

(402) Rintoul, L.; Harper, S. R.; Arnold, D. P. A Systematic Theoretical Study of the Electronic Structures of Porphyrin Dimers: DFT and TD-DFT Calculations on Diporphyrins Linked by Ethane, Ethene, Ethyne, Imine, and Azo Bridges. Phys. Chem. Chem. Phys. 2013, 15 (43), 18951–18964. 282

(403) Santos, K. L. M.; Barros, R. M.; da Silva Lima, D. P.; Nunes, A. M. A.; Sato, M. R.; Faccio, R.; de Lima Damasceno, B. P. G.; Oshiro-Junior, J. A. Prospective Application of Phthalocyanines in the Photodynamic Therapy against Microorganisms and Tumor Cells: A Mini-Review. Photodiagnosis Photodyn. Ther. 2020, 32, 102032.

(404) Matsuo, Y.; Ogumi, K.; Jeon, I.; Wang, H.; Nakagawa, T. Recent Progress in Porphyrin- and Phthalocyanine-Containing Perovskite Solar Cells. RSC Adv. 2020, 10 (54), 32678–32689.

(405) Rak, J.; Pouckova, P.; Benes, J.; Vetvicka, D. Drug Delivery Systems for Phthalocyanines for Photodynamic Therapy. Anticancer Res. 2019, 39 (7), 3323–3339.

(406) Miletin, M.; Zimcik, P.; Novakova, V. Photodynamic Properties of Aza- Analogues of Phthalocyanines. Photochem. Photobiol. Sci. 2018, 17 (11), 1749–1766.

(407) Gobo, N. R. S.; Brocksom, T. J.; de Oliveira, K. T. Soluble and Non- Aggregated Phthalocyanines: Synthesis, Mechanistic Aspects and Their Main Building Blocks. Curr. Org. Synth. 2017, 14 (8), 1132–1155.

(408) Brogdon, P.; Cheema, H.; Delcamp, J. H. Near-Infrared-Absorbing Metal- Free Organic, Porphyrin, and Phthalocyanine Sensitizers for Panchromatic Dye-Sensitized Solar Cells. ChemSusChem 2018, 11 (1), 86–103.

(409) Inokuma, Y.; Osuka, A. Subporphyrins: Emerging Contracted Porphyrins with Aromatic 14π-Electronic Systems and Bowl-Shaped Structures: Rational and Unexpected Synthetic Routes. Dalton Trans. 2008, No. 19, 2517–2526.

(410) Araujo, A. R. L.; Tome, A. C.; Santos, C. I. M.; Faustino, M. A. F.; Neves, M. G. P. M. S.; Simoes, M. M. Q.; Moura, N. M. M.; Abu-Orabi, S. T.; Cavaleiro, J. A. S. Azides and Porphyrinoids: Synthetic Approaches and Applications. Part 2-Azides, Phthalocyanines, Subphthalocyanines and Porphyrazines. Molecules 2020, 25 (7), 1749.

(411) Grant, T. M.; Josey, D. S.; Sampson, K. L.; Mudigonda, T.; Bender, T. P.; Lessard, B. H. Boron Subphthalocyanines and Silicon Phthalocyanines for Use as Active Materials in Organic Photovoltaics. Chem. Rec. 2019, 19 (6), 1093–1112.

283

(412) Rhoda, H. M.; Kayser, M. P.; Wang, Y.; Nazarenko, A. Y.; Belosludov, R. V; Kiprof, P.; Blank, D. A.; Nemykin, V. N. Tuning Up an Electronic Structure of the Subphthalocyanine Derivatives toward Electron-Transfer Process in Noncovalent Complexes with C60 and C70 Fullerenes: Experimental and Theoretical Studies. Inorg. Chem. 2016, 55 (19), 9549–9563.

(413) Azarias, C.; Pawelek, M.; Jacquemin, D. Structural and Optical Properties of Subporphyrinoids: A TD-DFT Study. J. Phys. Chem. A 2017, 121 (22), 4306–4317.

(414) Schrage, B. R.; Nemykin, V. N.; Ziegler, C. J. Biliazine: A Ring Open Phthalocyanine Analog with a Meso Hydrogen Bond. Chem. Commun. 2020, 56 (49), 6628–6631.

(415) Williams, A. T. R.; Winfield, S. A.; Miller, J. N. Relative Fluorescence Quantum Yields Using a Computer-Controlled Luminescence Spectrometer. Analyst 1983, 108 (1290), 1067–1071.

(416) Spek, A. L. CheckCIF Validation ALERTS: What They Mean and How to Respond. Acta Crystallogr. Sect. E 2020, 76 (1), 1–11.

(417) Nozari, M.; Addison, A. W.; Reeves, G. T.; Zeller, M.; Jasinski, J. P.; Kaur, M.; Gilbert, J. G.; Hamilton, C. R.; Popovitch, J. M.; Wolf, L. M.; et al. New Pyrazole- and Benzimidazole-Derived Ligand Systems. J. Heterocycl. Chem. 2018, 55 (6), 1291–1307.

(418) Bukuroshi, E.; Vestfrid, J.; Gross, Z.; Bender, T. P. Fluorinated Boron Subphthalocyanines: Lewis Acid Based Templating Chemistry Facilitates Random Halide Exchange, and Fluoride versus Chloride Affects the Basic Photophysical Properties and the Solid-State Arrangement. New J. Chem. 2019, 43 (42), 16730–16737.

(419) Zhang, C.; Guo, Y.; He, D.; Komiya, J.; Watanabe, G.; Ogaki, T.; Wang, C.; Nihonyanagi, A.; Inuzuka, H.; Gong, H.; et al. A Design Principle for Polar Assemblies with C3-Sym Bowl-Shaped π-Conjugated Molecules. Angew. Chemie Int. Ed. 2020, 60 (6).

(420) Rodríguez-Morgade, M. S.; Claessens, C. G.; Medina, A.; González- Rodríguez, D.; Gutiérrez-Puebla, E.; Monge, A.; Alkorta, I.; Elguero, J.; Torres, T. Synthesis, Characterization, Molecular Structure and Theoretical Studies of Axially Fluoro-Substituted Subazaporphyrins. Chem. – A Eur. J.

284

2008, 14 (4), 1055.

(421) Virdo, J. D.; Lough, A. J.; Bender, T. P. Redetermination of the Crystal Structure of Boron Subphthalocyanine Chloride (Cl-BsubPc) Enabled by Slow Train Sublimation. Acta Crystallogr. Sect. C 2016, 72 (4), 297–307.

(422) Clarke, R. G.; Hall, M. J. Recent Developments in the Synthesis of the BODIPY Dyes. Adv. Heterocycl. Chem. 2019, 128, 181–261.

(423) Jiang, X.-D.; Li, S.; Guan, J.; Fang, T.; Liu, X.; Xiao, L.-J. Recent Advances of the Near-Infrared Fluorescent Aza-BODIPY Dyes. Curr. Org. Chem. 2016, 20 (16), 1736–1744.

(424) Bodio, E.; Goze, C. Investigation of B-F Substitution on BODIPY and Aza- BODIPY Dyes: Development of B-O and B-C BODIPYs. Dye. Pigment. 2019, 160, 700–710.

(425) Boens, N.; Verbelen, B.; Ortiz, M. J.; Jiao, L.; Dehaen, W. Synthesis of BODIPY Dyes through Postfunctionalization of the Boron Dipyrromethene Core. Coord. Chem. Rev. 2019, 399, 213024.

(426) Chinna Ayya Swamy, P.; Sivaraman, G.; Priyanka, R. N.; Raja, S. O.; Ponnuvel, K.; Shanmugpriya, J.; Gulyani, A. Near Infrared (NIR) Absorbing Dyes as Promising Photosensitizer for Photo Dynamic Therapy. Coord. Chem. Rev. 2020, 411, 213233.

(427) Wang, J.; Gong, Q.; Wang, L.; Hao, E.; Jiao, L. The Main Strategies for Tuning BODIPY Fluorophores into Photosensitizers. J. Porphyr. Phthalocyanines 2020, 24 (5/7), 603–635.

(428) Poddar, M.; Misra, R. Recent Advances of BODIPY Based Derivatives for Optoelectronic Applications. Coord. Chem. Rev. 2020, 421, 213462.

(429) Zhang, W.; Ahmed, A.; Cong, H.; Wang, S.; Shen, Y.; Yu, B. Application of Multifunctional BODIPY in Photodynamic Therapy. Dye. Pigment. 2021, 185 (Part A), 108937.

(430) Hanson, K.; Patel, N.; Whited, M. T.; Djurovich, P. I.; Thompson, M. E. Substituted 1,3-Bis(Imino)Isoindole Diols: A New Class of Proton Transfer Dyes. Org. Lett. 2011, 13 (7), 1598–1601.

285

(431) Wen, H.-M.; Wu, Y.-H.; Xu, L.-J.; Zhang, L.-Y.; Chen, C.-N.; Chen, Z.-N. Luminescent Square-Planar Platinum(II) Complexes with Tridentate 3-Bis(2- Pyridylimino)Isoindoline and Monodentate N-Heterocyclic Ligands. Dalton Trans. 2011, 40 (26), 6929–6938.

(432) Csay, T.; Barath, G.; Kripli, B.; Kaizer, J.; Speier, G. Synthesis and Catalase- like Activity of Manganese(II) Complexes with Isoindoline-Based Ligands. Monogr. Ser. Int. Conf. Coord. Bioinorg. Chem. held Period. Smolenice Slovakia 2009, 9 (Insights into Coordination, Bioinorganic, and Applied Inorganic Chemistry), 120–132.

(433) Wang, L.; Tamgho, I.-S.; Crandall, L. A.; Rack, J. J.; Ziegler, C. J. Ultrafast Dynamics of a New Class of Highly Fluorescent Boron Difluoride Dyes. Phys. Chem. Chem. Phys. 2015, 17 (4), 2349–2351.

286

APPENDICES

287

APPENDIX A

PERMISSIONS

288

289

290

291

292

293

294

295

296

297