AMMONIA SPLITTING FOR RENEWABLE ENERGY CONVERSION By Reza Ghazfar A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Chemistry – Doctor of Philosophy 2023 ABSTRACT Due to the intermittent nature of renewable energy sources like wind and solar, efficient energy storage and distribution is essential when these sources are dormant. Among potential candidates for chemical energy storage, ammonia is gaining more attention, owing to its zero-carbon footprint, relatively efficient synthesis on a global scale, and well-established transportation infrastructure. If ammonia is synthesized via renewable energy sources, the efficient conversion of NH3 to N2 and H2 would complete an energy cycle where H2 stored as NH3 can fuel hydrogen/hybrid vehicles that are being commercialized. The efficiency of ammonia electrolysis to N2 and H2 can be improved by catalysts designed to lower the high overpotentials for oxidation and reduction at conventional electrodes. This work first describes the synthesis of tris and mono(ammine) iron complexes with tridentate phosphine ligands and their role in increasing current densities for NH3 oxidation relative to current densities generated using standard anodes. Electrocatalytic ammonia oxidation by a mononuclear ruthenium ammine complex supported by an isoindole-based tridentate ligand has been investigated next and its oxidation potential has been compared to previously reported mononuclear ruthenium ammine catalysts. At the end, a dinuclear polypyridine ruthenium bis(ammine) complex was reported in which ruthenium centers are held in close proximity by a bridging ligand in a way two NH3 ligands have a syn relationship, allowing the possibility of intramolecular oxidative N–N coupling. Copyright by REZA GHAZFAR 2023 NULLIUS IN VERBA iv ACKNOWLEDGMENTS First, I would like to thank Prof. Milton R. Smith for accepting me into his research group and supporting me during the Ph.D. program. I need to acknowledge Prof. Thomas W. Hamann for his valuable suggestions on the electrochemical aspects of my project. Also, I would like to thank Dr. Amrendra K. Singh for his contributions to the synthesis of some iron complexes. I am thankful to Dr. Daniel Holmes for his assistance in NMR experiments, Dr. Richard J. Staples for X-Ray crystallography, and Prof. John McCracken for EPR spectroscopy. I owe a lot of gratitude to my lab mates – Mona Maleka Ashtiani, Po-Jen Hsiao, Alex O’Connell, Pauline Mansour, Tim Shannon – and ammonia project members – Dan Little, Susanne Miller, Geletu Quing, Chenjia Mi, Arianna Savini – for being always happy to help. v TABLE OF CONTENTS LIST OF ABBREVIATIONS ................................................................................................... vii Chapter 1. AMMONIA AS A CARBON-NEUTRAL FUEL ....................................................... 1 Chapter 2. GENERAL EXPERIMENTAL PROCEDURES ........................................................ 9 Chapter 3. IRON AMMINE COMPLEXES CONTAINING A TRIPODAL PHOSPHINE LIGAND FOR ELECTROCATALYTIC AMMONIA OXIDATION ........................................ 22 Chapter 4. CATALYTIC AMMONIA OXIDATION BY A MONONUCLEAR RUTHENIUM COMPLEX SUPPORTED BY AN ISOINDOLE–BASED LIGAND ........................................ 69 Chapter 5. DINUCLEAR RUTHENIUM BIS(AMMINE) COMPLEXES FOR INTRAMOLECULAR N–N COUPLING ................................................................................... 89 Chapter 6. SUMMARY AND FUTURE DIRECTIONS ...........................................................102 REFERENCES...........................................................................................................................110 SUPPORTING INFORMATION (A) FOR CHAPTER 3 .....................................................119 SUPPORTING INFORMATION (B) FOR CHAPTER 4 .....................................................189 SUPPORTING INFORMATION (C) FOR CHAPTER 5 .....................................................200 vi LIST OF ABBREVIATIONS AO Ammonia Oxidation BDD Boron-doped diamond BE Bulk electrolysis C Columb CE Counter electrode cm Centimeters (1 × 10-2 meters) CV Cyclic Voltammogram, or Cyclic Voltammetry D Diffusion coefficient (cm2s-1) DCM Dichloromethane DFT Density Functional Theory dmpe 1,2-Bis(dimethylphosphino)ethane DMSO Dimethyl sulfoxide e– Electron E1/2 Half-wave potential Ep Peak potential F Faraday constant (96485.3321233100184 C mol-1) Fc Ferrocene Fc+ Ferrocenium Fc* Decamethylferrocene Fc*+ Decamethylferrocenium GC Glassy carbon J Joule vii K Kelvin kg Kilogram (1 × 103 grams) L Liters mA Milliamps (1 × 10-3 amps) mL Milliliters (1 × 10-3 Liters) n Integer number of electrons NHE Normal Hydrogen Electrode NMR Nuclear magnetic resonance M Molar (moles solute divided by liters solvent) MeCN Acetonitrile mM Millimolar (millimoles of solute divided by liters of solvent) mV Millivolts (1 × 10-3 volts) NBO Natural Bond Orbital OTf– Trifluoromethanesulfonate (triflate) anion pH Negative base ten logarithm of the molar concentration of protons ppm Part Per Million RE Reference Electrode s Seconds T Temperature TBAOTf Tetrabutylammonium trifluoromethanesulfonate TBAPF6 Tetrabutylammonium hexafluorophosphate THF Tetrahydrofuran Triflate Trifluoromethanesulfonate viii Trpy 2,2’:6’,2”–terpyridine UV Ultraviolet V Volts Vis Visible vs. Versus WE Working Electrode XPS X-ray Photoelectron Spectroscopy °C Degrees Celsius 298.15 K Grxn Change in Gibbs free energy for a chemical reaction at 298.15 K ix Abbreviation Chemical Structure bid bpp bpy bpy' dmpe trpy t SiP3 x Chapter 1. AMMONIA AS A CARBON-NEUTRAL FUEL 1 1.1. Consumption of Fossil Fuels and Global Warming Global warming due to anthropogenic greenhouse gasses released into the atmosphere remains one of the challenges for today’s world to overcome.1 On the other side, a progressive increase in the planet’s human population results in high demand for energy production. Fossil fuels constitute a major part of the world’s energy consumption. According to BP’s annual statistical review of world energy, fossil fuels provided 83.2% of the global energy supply in 2020 whereas for renewable energies this number reaches 5.7% (Figure 1.1).2 Oil Natural gas Coal 27.2% Nuclear energy Hydroelectricity Renewables 4.3% 24.7% 6.8% 5.7% 31.3% Figure 1.1 World’s energy consumption by fuel in 2020 The emission of CO2 is an inevitable consequence of fossil fuel combustion that is still rising worldwide, mainly due to the high energy demand in transportation sectors and increases in the standard of living (Figure 1.2).3 Although oil, natural gas, and coal resources are limited, new technology continues to make unexploited deposits accessible reserves, and by adding up the discoveries of new reservoirs, fossil fuels will sustain for centuries. 2 40 Total 35 Natural gas Petroleum Billion tons of CO2 / year 30 Coal Cement Production 25 Gas Flaring 20 15 10 5 0 1820 1840 1860 1880 1900 1920 1940 1960 1980 2000 Year Figure 1.2 Increase in CO2 emission due to the consumption of fossil fuels Extraction and employment of new carbon-based fuel deposits regardless of considering the consequences of CO2 emission into the atmosphere will eventuate in a 1.5°C temperature increase for the northern hemisphere by 2050 based on the current trend of energy consumption. The accelerated rise of sea level, changes in precipitation patterns, droughts, and heat waves are the most significant outcomes of global warming.4 1.2. Hydrogen Economy: Removing Carbon Out of Fuel Cycle An alternative for decreasing the emission from the combustion of fossil fuels can be achieved by utilizing non-carbon-based energy carriers. Hydrogen, as a sustainable and green energy carrier, has gained extensive consideration around the world due to its high energy density per mass and taking carbon out of the fuel cycle. Natural gas reforming and coal gasification are the most common industrial processes for massive hydrogen production that generate CO2 emission.5 So, 3 H2 can be regarded as a renewable fuel only if produced directly from renewable energy sources such as wind, solar, and hydropower.6 H2 has higher energy density per mass but lower energy density per volume among the common fuels which makes its transportation costly. Liquefaction of H2 increases its volumetric energy density from 4.5 MJ/L (at 700 bar) to 8.5 MJ/L but consumes ~35% of hydrogen’s energy content.7 Even the volumetric energy density of liquefied hydrogen is 1/4 of gasoline, and this is not a solution due to the high-energy-consuming process of hydrogen liquefaction. Moreover, even a cryogenic liquid hydrogen tank with good thermal insulation has a continuous boil-off at a rate of up to 1% per day which makes long-term storage challenging.8 All of these difficulties and safety concerns regarding hydrogen storage suggest an alternative hydrogen carrier. 1.3. Ammonia as Hydrogen Carrier Among the compounds that store hydrogen chemically, ammonia has attracted a significant amount of consideration due to its low-cost, large-scale production, and carbon-free emission when being used as a fuel. In 2016, 175 million metric tons of ammonia were produced worldwide and more than half of it was used in agriculture.9 Unlike hydrogen which requires carbon fiber- reinforced composite tanks for storage at 700 bar, ammonia liquefies at 10 bar and thus can be stored in stainless steel tanks. Moreover, there is a safe storage and transportation infrastructure due to the large-scale industrial production of ammonia, which is one of the prerequisites for a chemical to be considered as a worldwide fuel. Vehicles can be powered by ammonia in three different ways: internal combustion engines (ICEs), ammonia fuel cells, and onboard conversion of NH3 to N2 and H2, by thermal cracking or NH3 electrolysis and then feeding the produced H2 into a hydrogen fuel cell. Since the activity of ammonia toward combustion is 1/6 of gasoline10, in internal combustion engines ammonia must 4 be mixed with some portion of a combustion enhancer like gasoline, diesel, or H2 to burn and release enough energy to propel the car.11,12 There are some challenges to be solved regarding ammonia ICEs such as a relatively high NH3:gasoline ratio (3:7) in 1400 rpm engine speeds, ammonia’s high auto-ignition (651°C compared to 440 °C for gasoline), low flame temperature, corrosion of the engine parts and potential NOx emissions.13 The prototype of an ammonia fuel cell was first examined in the late 1960s based on alkaline fuel cells (AFCs) using a KOH electrolyte with an operating temperature range of 50-200 °C.14 The net reaction in an alkaline fuel cell is shown in the Scheme 1.1: Scheme 1.1 Overall reactions in an ammonia alkaline fuel cell Based on today’s technology, it is hard to develop a good low-temperature direct ammonia fuel cell with high power density. They are not suitable for transport applications due to the slow start- up and brittleness of their ceramic components.15 Low-temperature proton exchange membrane fuel cells (PEMFCs) using hydrogen as fuel have been developed for various applications including electric vehicles. Toyota Mirai which is commercially available to purchase uses PEMFC as an energy source. The car stores 6 kg of hydrogen at 700 bar in two tanks which can provide enough energy to drive 845 miles. 5 Utilizing ammonia in PEMFCs poisons the Pt/C anode catalyst and reacts with the acidic Nafion membrane; therefore, it is not a suitable fuel for PEMFCs by itself. However, ammonia can be used as a source for hydrogen production and the produced hydrogen can be fed into PEMFC for power generation.16 1.4. Splitting NH3 to H2 and N2 Two methods can be considered for NH3 splitting to H2 and N2: thermal cracking and electrolysis. Thermal cracking of ammonia occurs at temperatures ~500 °C in the presence of a heterogeneous catalyst. Ru is the most used catalyst for this process due to its high activity toward NH3 decomposition. The catalytic activity of Ru is support-dependent, which means the support facilitates the electron transfer and helps the recombination desorption of N atoms from the Ru surface, the rate-determining step in the heterogeneous catalytic cycle. It also enhances the dispersion and increases the effective area of the active catalyst.17 Utilizing carbon materials such as activated carbon, carbon nanotubes (CNTs), and CNTs-MgO as support shows the highest catalytic activity in ammonia decomposition. However, the reaction to produce H2 with a carbon support at high temperatures eventuates in the production of methane and decreases the efficiency of the catalyst over time.18 Investigating the new supports for Ru-based catalysts is the subject of ongoing research. Ru/graphene nanocomposites and Ru catalysts on non-carbon-based supports (MgO, SiO2, Al2O3, TiO2, ZrO2, and Cr2O3) have demonstrated high activity for ammonia dehydrogenation.19 In addition to Ru-based heterogeneous catalysts, the catalytic activity of main group compounds, including LiNH2 and NaNH2, also have been investigated, and it was shown that they could be effective ammonia decomposition catalysts.20 Electrolysis or electro-oxidation is another method for splitting ammonia into H2 and N2. The scalability and ability to operate in on-demand mode and at moderate temperature are the 6 advantages of electrolysis over thermal cracking methods.21 The thermodynamic potential for ammonia electrolysis in aqueous alkaline media is –0.06 V compared with –1.223 V for the electrolysis of water. The theoretical thermodynamic energy consumption is 1.55 Wh/g of H2 from the electrolysis of NH3 compared to 33 Wh/g of H2 from the electrolysis of H2O assuming that there are no kinetic limitations. This means that, theoretically, ammonia electrolysis consumes 95% less energy to produce the same quantity of hydrogen than water electrolysis.21 Electrolysis of ammonia is favorable thermodynamically; however, kinetics limits the rate of reaction requiring higher voltages to be applied. 1.5. Catalysis for Reducing the Overpotential of NH3 Oxidation In the search for electrodes that reduce the overpotential of NH3 oxidation in aqueous media, Pt alloys show effective catalytic properties. Pt/Ir, Pt/Ru, and Pt/Rh alloys are the most efficient catalysts for the dehydrogenation of NH3 in lower potentials. The electrocatalytic activity of Pt alloys decreases with the trend Pt/Ir > Pt/Rh > Pt/Ru.22 A significant amount of effort has been devoted to the development of electrocatalysts for the anode since ammonia oxidation has been identified as the limiting reaction.23 The efficiency of Pt/Ir electrodes for ammonia oxidation range from 80% at 10 mA/cm2 to 60% at 400 mA/cm2 in alkaline media.24 There are still some challenges with ammonia–alkaline electrolytic cells. The commercialization of the technology demands the development of more efficient electrodes with low–cost metals for future large–scale production. Moreover, the hydrogen capacity is limited to 6.1 mass%, because the ammonia concentration in saturated ammonia aqueous solution is 34.2 mass% at 20 °C.25 Unlike the electrolysis of NH3 in aqueous media, electrolysis of liquid NH3 has attracted less attention. In 2010 Hanada et al., reported the current density of 7.2 mA/cm2 from the electrolysis of liquid ammonia with 2.0 V applied potential between two Pt electrodes using 1 M KNH2 as the supporting electrolyte.26 They 7 proposed that oxidation of amide at the anode and reduction of ammonia at the cathode generates nitrogen and hydrogen respectively. Because of the application of a potential between the working and counter electrodes and not using a reference electrode for measuring the exact potential of each electrode, the overpotential of the anodic and cathodic reactions was not reported. In 2015 a revised mechanism was proposed.27 It was suggested that the cathodic reaction proceeds via initial one–electron reduction of NH4+ to NH4• rather than NH4+ dissociation to NH3 and H+ followed by H+ reduction. Also, by using a three-electrode system, overpotentials at the anode and cathode were determined to be 1000 mV and 600 mV, respectively. Finally, it was shown that Pt electrodes get poisoned in NH3(l) due to nitride formation. Even after resolving the problem of electrode poisoning, employing a suitable catalyst that lowers the overpotential of ammonia oxidation for efficient H2 generation is necessary. One possible solution is the use of a homogeneous catalyst. A catalyst should be designed to be oxidized, and then, in turn, oxidize NH3 in the bulk solution. In this project, we focus on the synthesis of Fe and Ru complexes to reduce the anodic overpotential of ammonia oxidation. 8 Chapter 2. GENERAL EXPERIMENTAL PROCEDURES 9 2.1. Electrochemistry 2.1.1. Instrument All electrochemical experiments were performed with a Metrohm Autolab PGSTA128N potentiostat using the Nova 2.1 software package. 2.1.2. Reference Electrode For accurate and precise control of the potential of a working electrode and obtaining reliable electrochemical data, an ideally non-polarized electrode, i.e., reference electrode, should be utilized. The potential of the reference electrode should remain practically constant upon current flow through the electrochemical cell. In previously reported papers, silver-based non-aqueous reference electrodes (Ag/AgNO3, Ag/AgOTf, Ag/AgClO4) have been used for electrochemical measurement in the presence of NH3. It will be shown here that the potential of these reference electrodes shifts due to the reaction of NH3 with silver to yield [Ag(NH3)2]+. This potential shift can result in serious experimental errors, especially during controlled potential electrolysis. Figure 2.1 shows the CV of 1 mM decamethylferrocene (Fc*) in THF containing 2 M of NH4OTf as a supporting electrolyte. Glassy carbon (GC) was used as the working electrode, Pt disk as the counter electrode and for preparing the reference electrode a 1/4 inch diameter glass tube with a Pt fused tip was filled with THF containing 5 mM of AgOTf. Although the CV of Fc* in the absence of NH3 was stable and no potential drift was observed after 200 scans, upon bubbling NH3 to make a saturated solution (3.24 M NH4 measured by titration and NMR spectroscopy), reference electrode potential started to drift. After ~150 scans, potential of the reference electrode became stable, however, ~100 mV drift was observed. Due to the assumption that the formation of [Ag(NH3)2]+ in the reference electrode after introducing NH3 to the solution results in drifting its potential, it was devised that instead of 10 Ag(OTf)2 in the reference electrode, [Ag(NH3)2]+to be used. For this purpose, [Ag(NH3)2]+was easily prepared by bubbling NH3 into AgOTf solution in THF. Upon removing the solvent, white solids of [Ag(NH3)2][OTf] were collected. E = ~100 mV 2/200 Scans 50/200 Scans 100/200 Scans 150/200 Scans 200/200 Scans 10µ 10µ Current (A) Current (A) 0 0 -10µ -10µ -20µ -20µ -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.1 Potential (V vs ref electrode) Potential (V vs ref electrode) Figure 2.1 Left: CV of 1 mM of Fc* in THF with 200 scans containing 2 M NH4OTf as supporting electrolyte right after bubbling NH3 for 5 min to get a saturated NH3 solution (3.24 M of NH3). Right: Same CV in which only scans #2, 50, 100, 150, and 200 are plotted for clarity. Working electrode: GC; counter electrode: Pt disk; reference electrode: Ag wire immersed in THF containing 2 M NH4OTf, and 5 mM AgOTf with Pt fused tip; scan rate: 100 mV/s Although Ag/[Ag(NH3)2]+ reference electrode was stable after bubbling NH3 and no potential drifting was observed, E1/2 of ferrocene (Fc) shifted upon the addition of 1 mM of Fc* into the solution (Figure 2.2). This pattern was observed in different solvents with different supporting electrolytes, so it was assigned to the intrinsic property of the reference electrode. In addition to Fc, E1/2 of other inorganic complexes was also shifted upon the addition of Fc*. For these reasons, a new reference electrode was realized for electrochemical experiments in presence of NH3 to eliminate silver. A previously reported reference electrode with a Pt wire immersed in a 1:1 ratio of Fc*/Fc*+ was chosen for this purpose due to the inertness of Fc* toward NH3.28–31 11 E = ~368 mV 50.0µ Fc Fc Fc* Current (A) 0.0 -50.0µ CV of 1 mM Fc CV of 1 mM Fc after addition of 1 mM Fc* -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 Potential (V vs ref electrode) Figure 2.2 E1/2 of 1 mM Fc after the addition of 1 mM of Fc* with Ag/[Ag(NH3)2]+ reference electrode in THF containing 2 M of NH4OTf shifts 368 mV to more anodic potential. Working electrode: GC; counter electrode: Pt disk; reference electrode: Ag wire immersed in THF containing 2 M NH4OTf, and 5 mM [Ag(NH3)2][OTf] with Pt fused tip; scan rate: 100 mV/s Fc*/Fc*+ reference electrode was prepared by immersing a Pt wire into a 1/4 inch glass tube with CoralPor tip containing THF, 1 M of NH4OTf as supporting electrolyte, 3 mM of Fc*, and 3 mM of Fc*OTf. Because Fc*OTf is not soluble in THF but dissolves in the presence of NH4OTf and Fc* is highly soluble in THF but dissolves very slowly in the presence of NH4OTf, it’s better to prepare Fc*OTf in THF with 1 M NH4OTf and Fc* in pure THF and then mix both. Due to the oxidation of Fc* solutions upon exposure to air, the reference electrode should be prepared under an inert atmosphere (glovebox) and sealed properly. To test the leaking of Fc* and/or Fc*OTf into the solution from the reference electrode through CoralPor membrane, CV of blank THF containing 1 M of NH4OTf was taken with the Fc*/Fc*+ reference electrode. A very small peak at 0 V vs Fc*/Fc*+ reference electrode can be observed in 12 the SQW which indicates a small leak of Fc* and/or Fc*OTf into the solution from the reference electrode through CoralPor membrane, which is reasonable (Figure 2.3). 5.0µ 5.0µ 0.0 Current (A) Current (A) 4.0µ -5.0µ 3.0µ -10.0µ 2.0µ -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 Potential (V vs Fc*/Fc*OTF ref electrode) Potential (V vs Fc*/Fc*OTf ref electrode) Figure 2.3 Left: CV of blank THF containing 1 M of NH4OTf as supporting electrolyte. Right: corresponding SQW at same condition right after taking CV. Working electrode: GC; counter electrode: Pt disk; reference electrode: Pt wire immersed in THF containing 1 M NH4OTf, 3 mM of Fc*, and 3 mM of Fc*OTf with CoralPor fused tip; scan rate: 100 mV/s To test the stability of the reference electrode in the absence of NH3, CV of 1 mM Fc* was taken which shows no drifting throughout 50 scans with the E1/2 of 0 V (Figure 2.4). 15.0µ Scan 2 Scan 50 5.0µ 10.0µ Current (A) Current (A) 0.0 5.0µ -5.0µ 0.0 -0.4 -0.2 0.0 0.2 0.4 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 Potential (V vs Fc*/Fc*OTf ref electrode) Potential (V vs Fc*/Fc*OTf ref electrode) Figure 2.4 Left: CV of 1 mM Fc* in THF with 1 M NH4OTf as supporting electrolyte. Right: corresponding SQW at same condition right after taking CV. Working electrode: GC; counter electrode: Pt disk; reference electrode: Pt wire immersed in THF containing 1 M NH4OTf, 3 mM of Fc*, and 3 mM of Fc*OTf with CoralPor fused tip; scan rate: 100 mV/s 13 After bubbling NH3 (2.8 M NH3) into the 1 mM solution of Fc* in THF, 50 consecutive CVs were taken again. No potential drift was observed after saturating the solution with NH3 and also after 50 consecutive scans (Figure 2.5). These data suggest that Fc*/Fc*+ is the best reference electrode for CV measurements in solutions containing NH3. 10.0µ Scan 2 Scan 50 25.0µ 5.0µ 20.0µ Current (A) Current (A) 15.0µ 0.0 10.0µ -5.0µ 5.0µ -10.0µ 0.0 -0.4 -0.2 0.0 0.2 0.4 0.6 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 Potential (V vs Fc*/Fc*OTf ref electrode) Potential (V vs Fc*/Fc*OTf ref electrode) Figure 2.5 Left: CV of 1 mM Fc* in THF with 1 M NH4OTf as supporting electrolyte after bubbling NH3 (2.8 M NH3). Right: corresponding SQW at same condition right after taking CV. Working electrode: GC; counter electrode: Pt disk; reference electrode: Pt wire immersed in THF containing 1 M NH4OTf, 3 mM of Fc*, and 3 mM of Fc*OTf with CoralPor fused tip; scan rate: 100 mV/s 2.1.3. E1/2 of Fc vs. Fc* in THF and MeCN with NH4OTf and TBAPF6 Electrolytes In most of the literature,32,33 the E1/2 of complexes in non–aqueous media were reported vs Fc. For converting the E1/2 of Fc to Fc* for making an accurate comparison between E1/2 of complexes, CVs of Fc were taken in THF and MeCN in the presence of Fc*, and the results are summarized in Table 2.1. Table 2.1 E1/2 of Fc vs. Fc* in MeCN and THF THF MeCN 2 M NH4OTf E1/2 Fc = 449 mV (vs. Fc*) – 0.1 M TBAPF6 E1/2 Fc = 459 mV (vs. Fc*) E1/2 Fc = 508 mV (vs. Fc*) 14 Different references report –810 V32,34,35 and –0.826 V31,33 for converting NHE to Fc/Fc+ in THF at room temperature. Based on Table 2.1 it can be straightforward to convert the reported E1/2 vs. Fc/Fc+ or NHE in literature to Fc*/Fc*+. 60.0µ Fc THF (2 M NH4OTf) THF (0.1 M TBAPF6) 40.0µ E1/2 Fc = 449 mV vs. Fc* E1/2 Fc = 459 mV vs. Fc* Fc* 30.0µ Fc Fc* 20.0µ Current (A) Current (A) 0.0 0.0 -30.0µ -20.0µ -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 Potential (V vs. Fc*) Potential (V vs. Fc*) 60.0µ MeCN (0.1 M TBAPF6) Fc 40.0µ E1/2 Fc = 508 mV vs. Fc* Fc* Current (A) 20.0µ 0.0 -20.0µ -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 Potential (V vs. Fc*) Figure 2.6 CV of Fc in the presence of Fc* in MeCN and THF 15 2.1.4. Bulk Electrolysis Bulk electrolysis experiments have been conducted in a 60 mL–sized pear shape electrochemical cell with four fused 14/20 thermometer adapters cap and Viton O-ring. Regarding electrodes, a GC plate as the working electrode, Pt mesh as the counter electrode, and an Fc* reference electrode were used. GC plate is dangling from a Pt wire which is connected to a copper wire inside a 1/4-inch diameter glass tube with silver epoxy covered with Loctite EA 1CTM (HYSOL®) adhesive. Pt mesh was also connected to a copper wire in the same manner. The cell was filled with solvent in such a way that the Pt wire which the GC plate is hanging would not be in touch with the solution. The top and bottom parts of the glass tubes were sealed with a glue gun and Loctite EA 1CTM (HYSOL®) adhesive, respectively. A stir bar also was put in the cell for stirring the solution during bulk electrolysis. Figure 2.7 Designed electrochemical cell for bulk electrolysis experiments 16 For the gas sampling port, a 3-layer, 1/2-inch diameter HAMILTON 76006 septum was used. Gas analysis was performed by the HIDEN HPR-20 R&D benchtop gas analysis system. Figure 2.8 BE experiment accompanied by gas analysis For gas analysis, because the instrument only reports the percentage of gases, a certain volume of ultra-pure helium was injected into the cell before the experiment, and moles of produced gases were calculated based on moles of injected helium. 17 2.2. Synthesis 2.2.1. Synthesis of Decamethylferrocenium triflate (Fc*OTf) According to the previously described procedure,32 decamethylferrocene first was sublimed at 140 °C at 0.01 torr as a yellow solid. In the air, 108 mg (0.996 mmol, 0.65 equiv) of p- benzoquinone was dissolved in 35 mL of THF and while stirring, 0.3 mL (3.524 mmol, 2.3 equiv) of triflic acid was added. The pale-yellow color of the p-benzoquinone turns orange. Then immediately decamethylferrocene (500 mg, 1.532 mmol, 1 equiv) was added as a solid. The color changed to green and Fc*OTf crashed out. After stirring for 5 min, the green solid was collected and washed with THF, and dried overnight under vacuum. Yield: 684 mg, 94%. 1H NMR (500 MHz, MeCN-d3) 𝛿 –37.42 (s, br). 19F NMR (470 MHz, MeCN–d3) 𝛿 –79.46 (s). 18 Figure 2.9 1H NMR of Fc*OTf in MeCN-d6 19 Figure 2.10 19F NMR of Fc*OTf in MeCN-d6 20 2.2.2. Synthesis of tris(4-bromophenyl)amine radical cation Following the previously reported procedure,36,37 in the glovebox 6.361 g (13.19 mmol, 1 equiv) of tris(4–bromophenyl)amine was dissolved in 25 mL of degassed and dry DCM. While stirring, 2.54 g (14.51 mmol, 1.1 equiv) of NOPF6 was added slowly as a solid. The color immediately changes from pale yellow to dark blue. Once the addition is complete, the solution was stirred for 10 minutes. Upon addition of 75 mL of Et2O a brown–purple solid crashes out. Solids were filtered and washed with Et2O until the washings were colorless. Solids were collected and dried under vacuum and stored in the freezer of the glovebox at –33 °C. Yield: 4.4 g, 53%. No 1 H or 13C NMR signals were observed for the product. 2.3. DFT Calculations All the free energies (G) were calculated at 1 atm and 298.15 K by density functional theory (DFT) using Gaussian 16 software package,38 and natural bond orbital (NBO) analysis with version 7.0.8.39 All the geometries of complexes were fully optimized with the B3LYP functional40 and def2-TZVP basis set.41 The SMD solvation model42 was applied to the gas-phase optimized structures to calculate the effect of solvation in THF on the free energies. The visualization of orbitals and optimized structures were done using GaussView software.43 21 Chapter 3. IRON AMMINE COMPLEXES CONTAINING A TRIPODAL PHOSPHINE LIGAND FOR ELECTROCATALYTIC AMMONIA OXIDATION 22 3.1. Introduction Iron coordination complexes have been studied extensively for catalytic N2 reduction to NH3. The reverse reaction, where NH3 is oxidized to N2, protons, and electrons, has recently garnered renewed interest because it completes a cycle for storage and distribution of renewable hydrogen as NH3, using the most abundant gas in Earth’s atmosphere as the feedstock. Chart 1.1 summarizes various Fe-based catalysts for NH3 oxidation reported by 2022. In 2019, Peters and co-workers reported electrochemical NH3 oxidation facilitated by a polypyridyl iron complex.44 Due to the high onset potential of NH3 oxidation relative to the thermodynamic limit in their iron system (0.7 V vs Fc/Fc+ in MeCN with boron-doped diamond (BDD) as the working electrode), they reported another iron complex in 2021 that shifted the onset of AO to lower (∼250 mV) potential compared to previous one at same conditions.45 In 2022, Wang and co-workers reported a ferric ammine system that decreases the onset of AO to –0.04 V vs Fc/Fc+ in THF with glassy carbon as the working electrode.46 While the AO mechanism for Peters’ polypyridyl Fe systems is ambiguous, Wang and co- workers have proposed a mechanism that is shown in Scheme 3.1. According to this mechanism, Fe(III) first gets oxidized to Fe(IV) followed by deprotonation of ammine ligand by NH3 to yield the corresponding Fe(IV) amido complex. One–electron oxidation of Fe(IV) amide to Fe(V) amide accompanied by nucleophilic attack of NH3 to amide forms a hydrazido Fe(III) complex in which the NH group of ancillary ligand gets protonated to NH2. Further oxidation of N2H4 to N2, protons, and electrons completes the mechanism cycle. 23 Chart 1.1 Previously reported Fe catalysts for NH3 oxidation The coordination chemistry of first–row transition metals containing tripodal phosphine ligands and their catalytic reactions are well established.47,48 Phosphines have been utilized for synthesizing a large number of iron complexes due to their decent electron donor ability and production of diamagnetic complexes amenable for further NMR studies. Scheme 3.2 highlights some examples of iron ammine complexes with Fe–P bonds.49–52 Four-coordinated complex A reported by Sellman in 1975 is one of the first examples of diamagnetic Fe(II) mono(ammine) complexes containing phosphine ligands. Complex B, however, is a low-spin, tris(ammine) 24 phosphine complex of Fe(II) synthesized by Winfield and co-workers in 1988 in which the PMe3 ligands have both fac and mer configurations. Scheme 3.1 Proposed mechanism by Wang and co-workers46 Complex C reported by Peters is an unusual paramagnetic five-coordinate Fe(I) mono(ammine) complex (S = 3/2) supported by a tris(phosphine)borane ligand with weak Fe–B bond that is yielded by decomposition of its N2H4 counterpart in benzene at room temperature. Another example of isolating Fe ammine complex containing phosphine ligands by N–N bond cleavage of N2H4 has been reported by Umehara et al.51 (complex D), which bears a pyrazole- based ligand. Since hydrazine complexes can be regarded as possible intermediates in catalytic NH3 oxidation reactions,53,54 reproduction of Fe-NH3 by cleaving the N–N bond of bound hydrazine can close the catalytic cycle. 25 In this chapter, we report a Fe(II) tris(ammine) AO catalyst supported by a tripodal phosphine ligand. The electrochemical behavior of this six–coordinated diamagnetic complex has been evaluated in the absence and presence of ammonia. Scheme 3.2 Selected examples of Fe ammine complexes containing phosphine ligands The effect of different electrolytes and electrodes on the catalysis and stability of the catalyst has been accessed as well. For investigating and isolation the possible intermediates of the catalytic ammonia oxidation mechanism, deprotonation, and chemical oxidation of the Fe(II) tris(ammine) complex have been evaluated. All the complexes have been isolated and characterized by various spectroscopic methods. 26 In the end, for evaluating the effect of ancillary ligands on the oxidation potential of the catalyst, two of the ammine ligands of Fe(II) tris(ammine) complex have been displaced by 1,2– bis(dimethylphosphino)ethane (dmpe) to yield a mono(ammine) complex. Cyclic voltammograms of mono and tris(ammine) complexes have been assessed and compared to each other. Moreover, Density Functional Theory (DFT) calculations have been employed for calculating the theoretical oxidation potentials. 27 3.2. Results and discussion The first attempts for evaluating the electrocatalytic activity of Fe(II) ammine complexes supported by phosphine ligands started with synthesizing Fe phosphine complexes with labile ligands, such as MeCN, which are amenable to displacement with NH3. Abstraction of the bridged chlorides of the previously reported Fe dimer 155 by silver triflate in acetonitrile at room temperature yields tris(acetonitrile) complex 2 in which all chlorides are substituted by acetonitrile. Scheme 3.3 Synthesis of Fe(II) ammine complex 3 Substitution of the MeCN ligands of 2 with NH3 in methanol saturated with ammonia in the course of 1 h and precipitation by Et2O affords 3 in 78% yield. MeCN rapidly displaces the NH3 ligands at ambient temperature in the absence of excess NH3 therefore MeCN-d6 was avoided as 31 the NMR solvent. P{1H} NMR of 3 in MeNO2-d3 shows one singlet at 38.07 ppm, which compared to 2 shifted 4.17 ppm downfield. The 31P{1H} NMR spectrum of complex 3 in DMSO- d6 shows a singlet resonance at 41.79 ppm, and the 1H NMR spectrum has a resonance at 2.02 ppm for the NH3 protons, which suggests that the NH3 ligands are still bound to Fe. Even though the NH3 ligands in 3 are displaced yielding tri-solvate compound 2 when it is dissolved in MeCN, the fact that 3 is stable in DMSO-d6 suggests that the kinetics and thermodynamics of NH3 binding in this system are nuanced. These features are discussed in more depth when we describe efforts to 28 prepare dicationic mono(ammine) species tSiP3Fe(L2)(NH3) where L2 is a neutral bidentate chelating ligand. Even though the lability of N–donor ligands made it impossible to synthesize some target molecules, that lability is advantageous for an ammonia oxidation sequence. Single crystals of 2 and 3 suitable for crystallography were obtained from slow diffusion of Et2O into the acetonitrile and methanol, respectively, at ambient temperature (Figure 3.1). Figure 3.1 Molecular structures of complex 2 (left) and 3 (right) with thermal ellipsoids at the 50% probability level (Hydrogen atoms of carbons, triflate counter ions, and solvent molecules were omitted for clarity). Selected bond lengths (Å) and angles (deg) for complex 2 (left): Fe–P1 2.2298(10), Fe–P2 2.2374(10), Fe–P3 2.2468(10), Fe–N1 1.972(3), Fe–N2 1.976(3), Fe–N3 1.963(3); P1–Fe–P2 91.50(3), P1–Fe–P3 92.03(3), P2–Fe–P3 92.46(4), N1–Fe–P1 176.11(9), N1– Fe–P2 92.22(8), N1–Fe–P3 88.93(8). Complex 3 (right): Fe–P1 2.228(3), Fe–P2 2.225(3), Fe–P3 2.228(3), Fe–N1 2.09(1), Fe–N2 2.10(1), Fe–N3 2.07(1); P1–Fe–P2 91.6(1), P1–Fe–P3 91.9(1), P2–Fe–P3 91.6(1), N1–Fe–P1 175.2(3), N1–Fe–P2 91.2(3), N1–Fe–P3 91.9(3) The electrochemical behavior of 0.5 mM of 2 was evaluated by cyclic voltammetry (CV) in THF containing 0.1 M of TBAPF6 as the supporting electrolyte and GC as the working electrode (Figure 3.2). The CV of 2, exhibits one quasi-reversible wave with anodic and cathodic peaks at 1.15 and 1.04 V, respectively, and a peak-to-peak separation of 0.11 V at E1/2 = 1.09 V vs Fc*/Fc*+. The theoretical E1/2 value (1.44 V vs Fc*/Fc*+) was calculated by an SMD-THF solvent model relative to the following reaction 2 + Fc*+ → 2+ + Fc* (G°298K = 33.4143 kcal/mol). 29 6.0µ 4.0µ Current (A) 2.0µ 100 mV/s 0.0 -2.0µ 0.5 1.0 1.5 2.0 Potential (V vs. Fc*) Figure 3.2 CV of 0.5 mM 2 in THF containing 0.1 M TBAPF6 as supporting electrolyte. WE: GC, CE: Pt disk, scan rate: 100 mV/s 8.0µ 6.0µ 4.0µ Current (A) 2.0µ 100 mV/s 0.0 -2.0µ 0.0 0.5 1.0 1.5 2.0 Potential (V vs. Fc*) Figure 3.3 CV of 0.5 mM 3 in THF containing 0.1 M TBAPF6 as supporting electrolyte. WE: GC, CE: Pt disk, scan rate: 100 mV/s 30 In contrast, the CV of 3 under the same conditions shows five anodic redox waves between 0.5 and 1.8 V and four peaks in the cathodic return (Figure 3.3). The first anodic wave appearing at 0.55 V vs Fc*/Fc*+ can be compared to the theoretical E1/2 = 1.17 V for oxidation of 3 obtained by electron transfer free energy from complex 3 to Fc*+ ( 3 + Fc*+ → 3+ + Fc*, G°298K = 27.008 kcal/mol). By changing the supporting electrolyte to 2 M NH4OTf, the CV of 3 shows one anodic and one cathodic peak at 0.63 V and 0.50 V respectively with E1/2 of 0.56 V vs Fc*/Fc*+. To examine the effect of [TBA]+ cation and [PF6]– anion on the CV of 3, 0.5 M of TBAPF6 and 0.5 M of TBAOTf was added to the 1 mM solution of 3 in THF containing 2 M of NH4OTf. 10.0µ 12.0µ 10.0µ 5.0µ 8.0µ Current (A) 0.0 Current (A) 6.0µ 4.0µ -5.0µ 2.0µ 0.0 0.4 0.8 1.2 0.0 0.4 0.8 1.2 Potential (V vs. Fc*) Potential (V vs. Fc*) Figure 3.4 Left: CV of 1 mM 3 in THF containing 2 M NH4OTf as supporting electrolyte. WE: GC, CE: Pt disk, scan rate: 100 mV/s. Right: SQW taken after CV As shown in Figure 3.5, by adding TBAPF6 and TBAOTf to 1 mM solutions of 3 in THF with 2 M NH4OTf as the electrolyte, the shape of CV changes significantly. Both TBA salts show the same behavior. This suggests reaction of 3 with TBA upon oxidation on the surface of the electrode. 31 10.0µ 1 mM 3 10.0µ 1 mM 3 1 mM 3 + 0.5 M TBAOTf 1 mM 3 + 0.5 M TBAPF6 5.0µ 5.0µ Current (A) 0.0 Current (A) 0.0 -5.0µ -5.0µ 0.0 0.4 0.8 1.2 0.0 0.4 0.8 1.2 Potential (V vs. Fc*) Potential (V vs. Fc*) Figure 3.5 Left: CV of 1 mM 3 in THF containing 2 M NH4OTf as supporting electrolyte (green); same CV after adding 0.5 M of TBAOTF (red). Right: CV of 1 mM 3 in THF containing 2 M NH4OTf as supporting electrolyte (green); same CV after adding 0.5 M of TBAPF6 (orange). WE: GC, CE: Pt disk, scan rate: 100 mV/s In addition to the electrochemical oxidation of 3 on the surface of the electrode, an attempt was made to chemically oxidize it by tris(4-bromophenyl)amine radical cation (Scheme 3.4). In the glovebox, 1 equiv (349 mg, 0.163 mmol) of BArF24 salt of 3 was dissolved in 15 mL THF. The oxidant was dissolved in THF (0.183 mmol, 115 mg) and added dropwise to the solution of 3. Color changes from orange to purple. Pentane was added to further precipitation until the purple solids crashed out, and the THF became colorless. Solids were collected on a frit and washed with pentane (yield: 97%). Scheme 3.4 Chemical oxidation of 3 by tris(4-bromophenyl)amine radical 32 The EPR spectrum of one electron oxidized 3 in THF is shown in Figure 3.6 at two different temperatures. The spectrum features two resolved peaks suggestive of low and high spin Fe(III) complexes with axial symmetry. 2.5 EPR at 130 K 2.0 EPR at 10 K 1.5 1.0 Intensity 0.5 0.0 -0.5 -1.0 -1.5 -2.0 0 50 100 150 200 250 300 350 400 450 500 magnetic field (mT) Figure 3.6 EPR spectrum of one electron oxidized 3 in THF at 130 and 10 K Given that the pKa values of NH3 protons typically decrease upon coordination to metal cations,56 we attempted to deprotonate 3 by a hindered base. To achieve this, complex 3 was treated with 1.6 equiv of LiN(SiMe3)2, which resulted in a new species being formed (Scheme 3.5). Single crystals of the product were grown by slow diffusion of diethyl ether into THF solution. The X- ray structure reveals a diferrrous cation with the formula [(tSiP3Fe)2(μ-NH2)3][OTf] (4) (Figure 3.7). This is the first structurally characterized dinuclear iron complex with three NH2 ligands in the bridging positions. Transition metal dinuclear complexes bridged by three μ-NH2 ligands have been reported for Cr,57 Co,58,59 Mo,60 and Pt.61 Solid state structures have been determined for all but the Mo example. 33 Scheme 3.5 Deprotonation of complex 3 with LiN(SiMe3)2 The average Fe–NH2 bond length of 2.026(8) Å in complex 4 is longer than the average Co– N bond length of 1.939(5) Å for the bridging NH2 ligands in [{(NH3)3Co)2(μ-NH2)3][ClO4]3 and identical to the Cr–NH2 distance (2.027(2) Å) in [{(NH3)3Cr)2(μ-NH2)3][I3]. The Fe–Fe distance of 2.840(1) Å is longer than the Co–Co (2.60 Å) and Cr–Cr (2.649(1) Å) distances in the aforementioned complexes.57–59 The shorter distance for the Co is expected since both have d6 low-spin configurations, and Co(III) has a higher effective nuclear charge. Cr(II) d4 complexes typically adopt high-spin configurations, and the Cr–Cr distance is comparable to those in dinuclear Cr(III)Cr(III) compounds where Cr–Cr bonding has been invoked.62,63 The most closely related Fe complex is the low-spin diiron(II) complex {[PhB(CH2P(CH2Cy)2)3]Fe}2(μ–𝜂1:𝜂1– N2H2)(μ–NH2)2 synthesized by Peters and co-workers.64 The average Fe–NH2 (2.041 Å) and Fe– Fe (3.087 Å) distances for this complex are longer than those in 4. 34 Figure 3.7 Molecular structures of complex 4 with thermal ellipsoids at the 50% probability level (Hydrogen atoms of carbons, triflate counter ion, and solvent molecules were omitted for clarity). Selected bond lengths (Å) and angles (deg) for complex 4: Fe–P1 2.1900(15), Fe–P2 2.1895(16), Fe–P3 2.1938(16), Fe–N1 2.029(5), Fe–N2 2.029(5), Fe–N3 2.022(4); P1–Fe–P2 92.37(6), P1– Fe–P3 92.54(6), P2–Fe–P3 92.37(6), N1–Fe–P1 168.49(17), N1–Fe–P2 93.2(2), N1–Fe–P3 97.3(2), Fe–N2–Fe 45.60(14), Fe–N1–Fe 45.59(14), Fe–N3–Fe 45.41(11) The 31P{1H} NMR spectrum of 4 in DMSO-d6 shows one singlet peak at 43.75 ppm, which is shifted 1.96 ppm down-field compared to the tSiP3 resonance in compound 3. In addition to the characteristic resonances of tSiP3 ligand, the protons of the bridging NH2 ligands appear at –2.79 ppm in the 1H NMR spectrum of the compound, which is 4.82 ppm up-field of the resonance for the NH3 ligands in compound 3. Upon chemical oxidation of complex 4 with 2 equiv of AgOTf in CH2Cl2, a solid precipitated over the course of the reaction. Upon recrystallization, a pure compound formulated as [(tSiP3)2Fe][OTf]2 (5) was obtained in 35% yield (Scheme 3.6). Characterization was based on the observation of single resonances for CH2 and PMe2 ligand protons in the 1H NMR spectrum, and one singlet in the 31P{1H} spectrum. The chemical shifts of these resonances are very similar to [L2Fe][BF4]2, L = bis(1,1,1–tris(dimethylphosphinomethyl)ethane, which is closely related to 5, where the CH2PMe2 arms of the tSiP3 ligand are connected to CMe instead of tBuSi.65 The Fe required for mass balance in the oxidation of 4 in Scheme 3.6 is presently unaccounted for. 35 Scheme 3.6 Ligand redistribution promoted by chemical oxidation of 4 The four–coordinate Fe(II) imido complex tSiP3Fe=NAryl (Aryl = 2,6–(iPr2)2C6H3), has been oxidized to a stable Fe(III) cation with AgSbF6 by Odom and co-workers.66 Thus, the ligand redistribution via comproportionation triggered by the oxidation of compound 4 was unexpected. We turned to theory to understand why attempted 2 e– oxidation of 4 triggered ligand redistribution. Density Functional Theory (DFT) calculations of 4 gave Fe–Fe and average Fe–NH2 bond lengths of 2.859 and 2.047 Å that are close to the respective experimental values of 2.840(1) and 2.026(8) Å. Natural Bond Orbital (NBO) theory predicts that the HOMO is comprised of 67.3% of Fe 3dx2–y2 atomic orbitals (AOs) and 25.7% of 2px N AOs of the μ-NH2 ligands, which have 46% non–bonding and 54% anti–bonding character.39,67 Shortening of the Fe–NH2 bonds would be expected after a 2 e– oxidation of 4 to 42+ since the anti–bonding character in HOMO of 4 would be lost. This is consistent with the theory where the Fe–NH2 bonds in 42+ contract ~0.10 Å relative to 4. The theoretical Fe–Fe distance decreases from 2.85 Å in 4 to 2.40 Å in 42+. NBO theory indicates an Fe–Fe bond in 42+—an expected consequence of 2 e– oxidation—comprised of Fe 3dx2–y2 (72%) and 3dz2 (24%) AOs, and a quantum theory of atoms in molecules (QTAIM) analysis reveals a bond critical point between the Fe centers,68,69 with an electron density of 𝜌= 34 kcal/mol. 36 Along with a decrease in Fe–N and Fe–Fe distances, DFT predicts the lengthening of Fe–P bonds by ~0.13 Å upon 2 e– oxidation, which is consistent with nN ––> σFeP* 2nd–order interactions from NBO theory analysis. Populating an orbital with P–C anti-bonding character is consistent with Fe–P lengthening in Fe–NH2 bonds in 42+. Metal π-back bonding to P–C σ* orbitals also contributes to the stabilization of M–P bonds, which would diminish upon oxidation of Fe(II) to Fe(III). Indeed, NBO theory estimates that 18 kcal/mol stabilization in compound 4 arises from second-order interactions between Fe(II) lone pairs and anti-bonding P–C NBOs. In the 2 e– oxidized compound 42+, no interactions of this type exceeded the 0.05 kcal/mol threshold. Hence, loss of Fe–P π-back bonding accompanies Fe–P elongation when 4 is oxidized. Also, the Fe–Fe distance decreases from 2.85 Å in 4 to 2.40 Å in 42+ due to the shortening of the Fe–NH2 bond, which leads to the formation of the Fe–Fe bond. The Fe–Fe bond in 42+ constitutes 83.35% of HOMO–6 with 91.6% bonding character (Figure 3.8). These calculations may explain the decomposition of 4 upon oxidation. Figure 3.8 Top: visualized HOMO of complex 4. Bottom: visualized HOMO–6 of 2 e– oxidized complex 4 37 Wolczanski has used CO, and other ligands, as reporters for charge distribution between ligands and metals in transition metal complexes. We applied his Charge Distribution Via Reporters (CDVR) method for insight into the phosphine redistribution reaction that occurs when complex 4 reacts with AgOTf.70 The CDVR method relies on Equation 1, where CML is the total charge on the metal complex, CM is the charge on the metal center, xL(i) is the number of ligands L(i) bound to M, and cL(i) is the charge on ligand L(i). Equation 1 CML = CM + xL1cL1 + xL2cL2 + … + xL(i)cL(i) 59 Since the Fe Mössbauer isomer shifts for [Fe(CO)6]2+, Fe(CO)5, and [Fe(CO)4)]2– span a narrow range of 0.17 mm/s, the charge on Fe in these complexes is assumed to be identical and CFe is set to 2.0. With this assumption, the cCO values calculated from Equation 1 for [Fe(CO)6]2+, Fe(CO)5, and [Fe(CO)4)]2– are 0, –0.4, and –1, respectively. Plots of cCO vs. νCO of these Fe complexes, and analogous plots for Ru and Os congeners, are linear, which makes it straightforward to calculate cCO from νCO values according to Equation 2. Equation 2 cCO = {νCO – 2207} cm–1/475 cm–1 With a value for cμNH2, CDVR can be applied to calculate charges on P of the tSiP3 ligand. The complex Fe2(CO)6(μ-NiPrC(H)(CO2Me)C(H)(CO2Me)-μ-NiPr) has μ–NR2 and CO as the only ligands. From the average of the reported νCO values (νCO = 2006 cm–1) for this complex and Equation 1, cCO = –0.42. Setting cFe = 2.0, and making the recommended correction for an Fe–Fe bond, the application of Equation 1 gives cμNR2 = –0.43 for the bridging amides. 38 Assuming that cμNR2 ~ cμNH2 and cFe = 2.0, the application of Equation 1 gives cP = –0.28 for the coordinated phosphorus atoms in complex 4. For [(tSiP3Fe)2(μ-NH2)3]2+ (4+) and [(tSiP3Fe)2(μ- NH2)3]3+ (42+), the calculated cP values for 4+ and 42+ are –0.12 and 0.05, respectively. Fe–P π- back bonding diminishes as cP increases, which synergistically weakens Fe–P σ-bonding. With simple arithmetic and qualitative molecular orbital arguments, CDVR captures the essence of more rigorous quantum chemical analyses that rationalize the ligand redistribution triggered by oxidizing complex 4. Given that the pKa values of NH3 protons typically decrease upon coordination to metal cations,56 we hypothesized that oxidation of compound 3 from Fe(II) to Fe(III) can trigger deprotonation, particularly if dissociation of bound NH3 generates a proton acceptor. We examined this hypothesis by theoretically calculating the pKa value of the bound NH3 to Fe(III) center (pKa = 8.36) by a THF protonated (THF––H+––THF) model.56 The Grxn for deprotonation of NH3 on Fe(III) by THF (11.41 kcal/mol) is endergonic. Since weak bases (e.g. H2O) catalyze the rapid exchange of coordinated NH3 protons, the barriers for endoergic reactions are close to the thermodynamic limit. Thus, the intermediate [(tSiP3)Fe(NH3)2(NH2)]2+ should be accessible in THF. Based on NBO analysis, singly occupied molecular orbital (SOMO) in deprotonated form of 3a (3b) resides ~ 40% on NH2 (94% 2pz of N) and 23% on Fe (67% 3dyz, 22% 3dxy). This arises the possibility of intermolecular N–N coupling between two 3b intermediates to form a bridged N2H4 dinuclear complex which its further oxidation yields N2 and H2 (Scheme 3.7). 39 Scheme 3.7 Proposed mechanism of catalytic NH3 oxidation by complex 3 The electrocatalytic behavior of 3 was examined in THF in the presence of NH3 containing 1 and 2 M concentrations of NH4OTf on the surface of BDD and GC electrodes. Figure 3.9 shows the catalytic behavior of 3 in THF with 2 M of NH4OTf as the supporting electrolyte, 3.24 M NH4, and GC as the working electrode in the presence of 1 mM Fc* as the internal standard. For testing the stability of the catalyst on decomposition and passivation of the surface of the GC electrode, 51 consecutive scans were cycled to test any current drop during the catalysis (Figure 3.10). As shown in Figure 3.10 , no current drop or electrode poisoning was observed during 51 scans. It is worth noting that in the presence of TBAPF6 and/or TBAOTf as supporting electrolytes, the catalytic current drops significantly due to the reaction of oxidized forms of 3 with TBA followed by electrode passivation. 40 Blank 3.24 M NH3 120.0µ 1 mM of 3 1 mM of 3 + 3.24 M NH3 80.0µ Current (A) 40.0µ 0.0 -0.4 0.0 0.4 0.8 1.2 Potential (V vs. Fc*) Figure 3.9 Green: CV of 1 mM of Fc* in THF with 3.24 M NH3. Red: CV of 1 mM of 3 in THF in the presence of 1 mM Fc*. Blue: CV of 1 mM of 3 in THF in the presence of 1 mM Fc* with 3.24 M NH4. Supporting electrolyte: 2 M NH4OTf, WE: GC, CE: Pt disk, scan rate: 100 mV/s 200.0µ Scan 2 Scan 51 160.0µ 120.0µ Current (A) 80.0µ 40.0µ 0.0 -0.4 0.0 0.4 0.8 1.2 Potential (V vs. Fc*) Figure 3.10 Cycling 1 mM of 3 in THF containing 2 M NH4OTf, and 3.24 M NH3 for 51 scans (CV was plotted starting form 2nd scan) with overlay of 2nd and 51st scans. WE: GC, CE: Pt disk, scan rate: 100 mV/s 41 In addition to the GC electrode, the electrocatalytic behavior of complex 3 was investigated with the BDD electrode. NH3 gets oxidized on the surface of the BDD electrode in ~ 360 mV higher potential than GC (Figure 3.11); therefore, BDD is suitable for studying the electrocatalytic behavior of complexes with relatively high oxidation potential. 40.0µ GC BDD 20.0µ Fc* Current (A) 0.0 -20.0µ -40.0µ -0.4 0.0 0.4 0.8 1.2 Potential (V vs. Fc*) Figure 3.11 Comparison of the potential of NH3 oxidation in THF in the presence of 3.24 M NH3 and 1 mM Fc* on the surface of GC (blue) and BDD (red) electrodes. Note that the current obtained by the BDD electrode is normalized for overlay by multiplying by a factor of 6.5. Supporting electrolyte: 2 M NH4OTf, WE: GC (blue) and BDD (red), CE: Pt disk, scan rate: 100 mV/s Figure 3.12 shows the CV of complex 3 with and without the presence of NH3 in the presence of 1 mM of Fc* as the internal standard. Also, 50 consecutive scans were cycled in the presence of NH3 to evaluate the stability of the catalyst (Figure 3.13). No current passivation was observed in the course of these multiple scans. 42 Blank 3.24 M NH3 1mM of 3 20µ 1mM of 3 + 3.24 M NH3 Current (A) 10µ 0 -0.4 0.0 0.4 0.8 1.2 Potential (V vs. Fc*) Figure 3.12 Green: CV of 1 mM of Fc* in THF with 3.24 M NH3. Red: CV of 1 mM of 3 in THF in the presence of 1 mM Fc*. Blue: CV of 1 mM of 3 in THF in the presence of 1 mM Fc* with 3.24 M NH4. Supporting electrolyte: 2 M NH4OTf, WE: BDD, CE: Pt disk, scan rate: 100 mV/s 20µ Scan 2 Scan 50 10µ Current (A) 0 -0.4 0.0 0.4 0.8 Potential (V vs. Fc*) Figure 3.13 Cycling 1 mM of 3 in THF containing 2 M NH4OTf, and 3.24 M NH3 for 50 scans (CV was plotted starting from the 2nd scan) with an overlay of 2nd and 50th scans in the presence of 1 mM of Fc* as internal standard. WE: BDD, CE: Pt disk, scan rate: 100 mV/s 43 For identifying the amount of produced N2(g) on the anode in the presence of the catalyst, controlled potential electrolysis (CPE), i.e., bulk electrolysis (BE) was conducted by applying a constant potential vs. Fc*/Fc*+ reference electrode. For better mass transport during the BE, the solution was vigorously stirred. BE was performed in THF with 1 M of NH4OTf as supporting electrolyte, 2.8 M NH3, and 1 mM of complex 3. A GC plate was used as the working electrode, and a Pt mesh was used as the counter electrode. A controlled BE experiment was carried out for 1 h without the presence of 3 at an applied potential of 0.8 V (vs. Fc*) (Figure 3.14). 0.8 4.0m 0.6 Electric Charge (A.s) Current (A) 0.4 2.0m 0.2 0.0 0.0 0 1000 2000 3000 4000 0 1000 2000 3000 4000 Time (s) Time (s) Figure 3.14 Left: control BE of THF solution with 2.8 M NH3 in the absence of catalyst for 1 h by applying 0.8 V (vs. Fc*). Right: the amount of the charge passed (C) during 1 h of BE. Supporting electrolyte: 1 M NH4OTf, WE: GC plate, CE: Pt mesh During 1 h, 0.738417 C charge was passed. Analysis of the headspace of BE cell, reveals the formation of ~ 1.3 µM N2(g). After BE of blank, 1 mM of complex 3 was added quickly to the same cell by an airtight syringe, and 0.8 V (vs. Fc*) was applied to the WE again for 1 h. At this time 2.88861 C charge was passed in the course of 1 h (Figure 3.15). Analysis of the BE’s cell headspace shows the formation of ~ 4.56 µM N2(g) with faradic efficiency of 91.4%. 44 3 4.0m 2 Electric Charge (A.s) Current (A) 2.0m 1 0.0 0 0 1000 2000 3000 4000 0 1000 2000 3000 4000 Time (s) Time (s) Figure 3.15 Left: BE of THF solution with 2.8 M NH3 in the presence of 1 mM complex 3 for 1 h by applying 0.8 V (vs. Fc*). Right: the amount of the charge passed (C) during 1 h of BE. Supporting electrolyte: 1 M NH4OTf, WE: GC plate, CE: Pt mesh After the BE, the GC plate was rinsed with THF and then its surface was analyzed by XPS for detection of any iron species that can elevate the possibility of heterogeneous catalysis. Table 3.1 shows the results of the XPS surface analysis of the GC plate before and after the BE. Table 3.1 XPS data of the surface of the GC plate before and after the BE Element Before BE After BE O 1s 7.68% 7.79% C 1s 78.72% 66.28% N 1s 12.78% 21.40% F 1s 0.35% 1.57% Fe 2p – 0.26% S 2p – 0.96% Si 2p 0.46% 0.59% P 2p – 1.16% According to the XPS data, Fe was deposited on 0.26% of the surface of the electrode. To test the possible heterogeneous catalysis with this low percentage of Fe on the surface of the GC 45 electrode, the following experiment was designed. The potential of 0.8 V (vs. Fc*) was applied to the GC electrode in THF containing 1 M NH4OTf, 2.8 M NH4 and 1 mM 3. After 1 h, the GC electrode was taken out, and rinsed with THF, and inserted into a fresh solution of THF with 1 M NH4OTf and 2.8 M NH4. No catalytic current was observed after this rinse test. Figure 3.16 shows the correlation between the measured current at potential 0.68 V (vs. Fc*) and the concentration of added complex 3 in THF solution with 2 M NH4OTf, BDD as a working electrode, and 3.24 M of NH3. As shown in Figure 3.16, the catalytic current has a linear dependence (1st order ) on the concentration of complex 3. 0 mM 25.0µ 0.40 mM 0.79 mM 20.0µ 1.19 mM 1.55 mM 15.0µ 1.94 mM Current (A) 2.29 mM 10.0µ 2.67 mM 3.00 mM 5.0µ 3.37 mM 3.68 mM 0.0 -5.0µ -0.2 0.0 0.2 0.4 0.6 Potential (V vs. Fc*) 14.0µ 0.008 0 mM 0.40 mM y = 0.0017 x + ( 4.33837  10-4 ) SQW Current at 0.68 V vs. Fc* (mA) 12.0µ 0.79 mM 0.006 R2 = 0.99779 1.19 mM 10.0µ 1.55 mM 1.94 mM 8.0µ 2.29 mM Current (A) 0.004 2.67 mM 6.0µ 3.00 mM 3.37 mM 4.0µ 3.68 mM 0.002 2.0µ 0.000 0.0 -0.2 0.0 0.2 0.4 0.6 0.8 0 1 2 3 4 Potential (V vs. Fc*) Concentraions (mM) of Complex 3 Figure 3.16 Top: CV of THF solution containing 2.4 M NH3, and 1 mM Fc* as internal standard with different concentrations of complex 3. Bottom left: corresponding SQW taken instantly after each CV. Bottom right: SQW current at 0.68 V (vs. Fc*) vs. different concentrations of complex 3. WE: BDD, CE: Pt disk, scan rate: 100 mV/s 46 The dependence of catalytic current on NH3 concentration, however, does not show any linear correlation. Upon the first addition of a low concentration of NH3, the current slightly increases but by addition of other small consecutive NH3 concentrations, no current increase is observed. Saturation of solution with NH3 (2 – 3 M NH3) results in a catalytic current. 0 mM 0 mM 32 mM 20.0µ 22 mM 20.0µ 63 mM 44 mM 94 mM 88 mM 15.0µ 125 mM 15.0µ 220 mM 154 mM sat. NH3 (2.19 M) 308 mM Current (A) Current (A) 10.0µ 10.0µ sat. NH3 (3.24 M) 5.0µ 5.0µ 0.0 0.0 -5.0µ -5.0µ -0.2 0.0 0.2 0.4 0.6 0.8 1.0 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 Potential (V vs. Fc*) Potential (V vs. Fc*) Figure 3.17 Left: CV of 1 mM complex 3 with varying concentrations of NH3 in THF with 2 M of NH4OTf, 1 mM Fc* as internal standard, and GC as working electrode. Right: CV of 1 mM complex 3 with varying concentrations of NH3 in THF with 1 M of LiOTf, 1 mM Fc* as internal standard, and BDD as working electrode. In both cases, the scan rate is 100 mV/s and the Pt disk was used as CE. Figure 3.17 shows the CVs of complex 3 with NH4OTf (left) and LiOTf (right) as supporting electrolytes taken with GC and BDD as working electrodes respectively in the presence of different concentrations of NH3. Reactions of 2 and 3 with bidentate chelating ligands were next explored to prepare mono(ammine) complexes and investigate the effect of chelation on the oxidation potential of the iron ammine complexes. Our initial attempts to displace NH3 and acetonitrile from complexes 2 and 3, respectively, with bipyridine, gave recovered starting materials (Scheme 3.8). 47 Scheme 3.8 Attempted Syntheses of [(tSiP3)Fe(bpy)(L)][OTf]2 Complexes The latter case was surprising since compound 3 was synthesized by displacing acetonitrile ligands from 2 with NH3. The reaction of tSiP3 with [(bpy)Fe(NCMe)4][OTf]2 was tested as an alternate route to a monoacetonitrile compound. Instead, compound 2 and bipyridine were formed, (Scheme 3.8) which can be attributed to the strong trans-influence imparted by phosphine ligands that deter the bpy ligand from binding to the Fe center.71 Indeed the only reported complex of this kind is zero-valent, five-coordinate Fe complex bearing bpy and PhP(CH2CH2PPh2)2 ligands reported by Mukhopadhyay et al.72 We performed an NBO analysis to test the feasibility of forming any bond between 2,2´– bipyridine (bpy) and Fe with the formula [tSiP3Fe(bpy)(NH3)][OTf]2. NBO analysis indicates a 3– center–4–electron hyper bond (3CHB) between phosphines trans to bpy, Fe, and nitrogens of bpy. 48 The resonance hybrid shows a higher P–Fe bonding character (63%) than Fe–N (37%). Also according to the other segments of NBO analysis, the high occupancy of σFeP* (0.25 e) and low occupancy of nN (1.73 e), and large estimated nN→σFeP* 2nd–order interaction energy (62 kcal/mol) are all consistent with strong P–Fe–N hyberbond character. Similar results were obtained for [tSiP3Fe(bpy)(MeCN)][OTf]2 as well. Since the bpy complexes could not be prepared, we targeted [tSiP3Fe(dmpe)(L)][OTf]2 (L = MeCN, NH3, dmpe = 1,2–bis(dimethyl–phosphino)ethane) expecting that that dmpe would bind more strongly than bpy. Compound 2 reacted with dmpe provided [tSiP3Fe(dmpe)(MeCN)][OTf]2 (6) in good yield. Attempts to prepare [tSiP3Fe(dmpe)(NH3)][OTf]2 (7) by displacing MeCN from 6 with NH3 gave low yields of 7. Displacing two NH3 ligands from compound 3 with dmpe required heating to 60 °C, generating a reaction mixture where 7 was a minor species. We circumvented MeCN complexes by reacting the previously reported pentaphospine Fe(II) complex [tSiP3Fe(dmpe)(Cl)][Cl] (8)55 with AgOTf in 1:1 ratio of THF:CH2Cl2 yielding the violet triflate complex [tSiP3Fe(dmpe)(OTf)][OTf] 9. Preliminary data from 2 different X–ray crystallography data collections produce the triflate bond complex. However, data worthy of publication could not be obtained. The 31P{1H} NMR spectrum of 9 shows three moderately broad resonances at 39.82, 27.29, and 16.23 ppm in THF-d8, and the 19F NMR spectrum reveals a sharp singlet peak at –61.59 and a broad peak at –75.34 ppm. The latter 19F chemical shift is similar to unbound triflate ions in the other complexes we have prepared, and the downfield shift upon coordination is typical for triflate binding to Fe(II).73,74 49 Scheme 3.9 Synthesis of Fe(II) mono(ammine) complex 7 The spectroscopic data are consistent with our formulation for compound 9, which reacts rapidly when NH3 is added to a THF solution, accompanied by a color change from violet to orange. [tSiP3Fe(dmpe)(NH3)][OTf]2 (7) was isolated by removing the THF solvent and washing the solid material with Et2O to give an orange powder with 94% yield. The 31P{1H} NMR spectrum of 7 shows coupling typical for an AA´BB´C spin system where AA´ and BB´ refer to chemically equivalent P nuclei of respective dmpe and tSiP3 ligands, and C is the tSiP3 P that lies in the mirror plane containing Fe and the NH3 ligand. The 1H NMR spectrum of 7 (DMSO-d6) has resonances 50 for phosphine ligand protons at similar chemical shifts seen for compound 6 and an additional resonance integrating to three protons at 0.83 ppm for the NH3 ligand. Compounds 6 and 7 were also characterized by single-crystal X-ray diffraction experiments. The crystal structures and selected distances and angles are shown in Figure 3.18. Figure 3.18 Molecular structures of complex 7 (left) and 6 (right) with thermal ellipsoids at the 50% probability level (Hydrogen atoms of carbons, triflate counter ions and solvent molecules were omitted for clarity). Selected bond lengths (Å) and angles (deg) for complex 6 (left): Fe–P1 2.2708(12), Fe–P2 2.2814(12), Fe–P3 2.2911(12), Fe–P4 2.3143(13), Fe–P5 2.3143(13), Fe–N1 2.113(3); P1–Fe1–P2 89.28(4), P1–Fe1–P3 90.16(4), P1–Fe1–P4 91.69(4), P1–Fe1–P5 101.10(5), P2–Fe1–P3 90.07(4), P2–Fe1–P4 96.59(5), P2–Fe1–P5 169.60(5), P3–Fe1–P4 173.11(5), P3– Fe1–P5 89.25(5), P4–Fe1–P5 83.88(5), N1–Fe1–P1 174.48(10), N1–Fe1–P2 85.66(10), N1–Fe1– P3 92.05(10), N1–Fe1–P4 86.70(10), N1–Fe1–P5 83.99(10). Complex 6 (right): Fe–P1 2.273(5), Fe–P2 2.286(4), Fe–P3 2.304(4), Fe–P4 2.305(4), Fe–P5 2.344(4), Fe–N1 1.944(16); P1–Fe1–P2 90.37(17), P1–Fe1–P3 89.76(16), P1–Fe1–P4 100.89(18), P1–Fe1–P5 95.13(16), P2–Fe1–P3 90.44(16), P2–Fe1–P4 91.24(17), P2–Fe1–P5 172.99(17), P3–Fe1–P4 169.20(18), P3–Fe1–P5 93.93(15), P4–Fe1–P5 83.46(16), N1–Fe1–P1 175.0(4), N1–Fe1–P2 88.4(5), N1–Fe1–P3 85.4(4), N1–Fe1–P4 84.0(4), N1–Fe1–P5 86.5(5) The CV of complex 6 is shown in Figure 3.19. Complex 6 has anodic and cathodic peaks at 1.43 and 1.35 V (vs. Fc*), respectively, with a peak–to–peak separation of 0.08 V and E1/2 = 1.39 V. The theoretical E1/2 value for 6 was found to be 1.29 V vs Fc*/Fc*+ according to the following reaction (6 + Fc*+ → 6b + Fc*, G°298K = 29.7340 kcal/mol) in which 6b is the one-electron oxidized form of 6. 51 20µ 10µ Current (A) 0 -10µ 0.8 1.2 1.6 2.0 Potential (V vs. Fc*) Figure 3.19 CV of 1 mM 6 in THF containing 2 M NH4OTf as supporting electrolyte. WE: GC, CE: Pt disk, scan rate: 100 mV/s The CV of compound 7 is shown in Figures 8a and 8b, respectively. Complex 7 has anodic and cathodic peaks at 1.20 and 1.10 V (vs. Fc*), respectively, with a peak–to–peak separation of 0.10 V and E1/2 = 1.15 V. The theoretical E1/2 value for 6 was found to be 1.09 V vs Fc*/Fc*+ according to the following reaction (7 + Fc*+ → 7b + Fc*, G°298K = 25.1345 kcal/mol) in which 7b is the one-electron oxidized form of 7. 30µ 25.0µ 20µ 20.0µ 10µ 15.0µ Current (A) Current (A) 0 10.0µ -10µ 5.0µ -20µ 0.0 -30µ 0.0 0.4 0.8 1.2 1.6 0.0 0.4 0.8 1.2 Potential (V vs. Fc*) Potential (V vs. Fc*) Figure 3.20 Left: CV of 1 mM 7 in THF containing 2 M NH4OTf as supporting electrolyte. WE: GC, CE: Pt disk, scan rate: 100 mV/s. Right: SQW taken after CV 52 Table 3.2 summarizes the theoretical and experimental E1/2 potentials of complexes 2, 3, 6, and 7 vs. Fc*. Acetonitrile complexes show higher E1/2 potentials compared to their ammine analogs both in theoretical and experimental values. The calculated E1/2 potential of complex 3 shows a higher deviation from its experimental value. Also, theoretical calculations suggest that ligation of Fe by dmpe ligand should decrease the E1/2 of the complex; however, experimental values do not match with the theoretical predictions. Indeed, E1/2 of complex 6 is higher than the onset potential of NH3 oxidation on the surface of the GC electrode and hence its electrocatalytic behavior cannot be accessed by GC as the working electrode. Table 3.2 Comparison of the theoretical and experimental E1/2 (V vs. Fc*) values of Complexes 2, 3, 6, and 7 Redox reaction Theoretical Theoretical E1/2 Exp. E1/2 G°298K (kcal.mol–1) (V vs. Fc*) (V vs. Fc*) 2 + Fc*+ 2b + Fc* 33.414 1.4489 1.09 3 + Fc*+ 3b + Fc* 27.008 1.1711 0.56 6 + Fc*+ 6b + Fc* 29.7340 1.2893 1.39 7 + Fc*+ 7b + Fc* 25.1345 1.0899 1.15 Attempts to deprotonate 7 with hindered bases like LiN(SiMe3)2 and LiN(iPr)2 were unsuccessful. This can be explained by the steric effects of the methyl groups of phosphine ligand that protect bound NH3 from hindered bases. A similar observation was reported for [Fe(dppe)(cp)NH3]+ complex (dppe = 1,2-bis(diphenylphosphino)ethane, cp = cyclopentadiene) in THF by Sellman et al. in which deprotonation with NaN(SiMe3)2 and BuLi did not yield the desired product. Utilizing LiAlH4 by Sellman et al., however, provides the corresponding hydrido complex.75 53 There is precedent for the formation of (dmpe)5Fe2 and trans-Fe(dmpe)2Cl2 complexes as byproducts during the reduction of complex 8.55 Also structural interconversion of similar Fe(II) systems with reversible κ2/κ3-coordination modes of flexible tripodal phosphine ligand has been established.76 These structural dynamics should also be taken into account during the evaluation of the electrocatalytic behavior of these types of complexes. Since hydrazine complexes are plausible intermediates in Fe-catalyzed N2 reductions and NH3 oxidation,53 syntheses of hydrazine complexes with a tSiP3Fe platform were explored. Reactions of hydrazine with tris(ammine) complex 3 gave a mixture of products, including the tris(hydrazine) product (vide infra). When hydrazine was added to THF solutions of 6, no reaction was observed. However, when 6 was heated to 60 °C in neat hydrazine, compound 10 was isolated in 57% yield (Scheme 3.10). In addition to a κ1-N2H4 ligand, 10 has a bidentate κ2–acetamidrazone ligand that results from the formal addition of hydrazine to acetonitrile. The 31P{1H} NMR spectrum of 10 in DMSO–d6 shows one doublet at 42.13 ppm (J = 49.9 Hz), and one triplet at 36.22 ppm (J = 49.4 Hz) with a ratio of 2:1 indicative of breaking the C3v symmetry of complex. The crystal structure of 10 shows the formation of κ2–acetamidrazone complex [tSiP3Fe(N2H4)(MeC(NH)NH2NH)][OTf]2 (10). Fe κ2–acetamidrazone ligands have been prepared previously by reacting L4Fe(II)(κ2–N2H4)(NCR) (L = P(OR)3) complexes with excess hydrazine.77 The κ2–acetamidrazone ligand in 10 might arise from the intramolecular nucleophilic attack of η1–N2H4 from the putative intermediate [tSiP3Fe(NCMe)(N2H4)2)][OTf]2. Alternatively, free hydrazine could first attack coordinated acetonitrile in complex 6, since the nucleophilic attack at nitrile carbons is facile when nitrile coordinates to Lewis acids like BF3. 54 Scheme 3.10 Reaction of hydrazine with complex 6 and 9 Compound 10 is not relevant to attempted electrocatalysis in THF solutions. We explored the reactivity of [tSiP3(dmpe)Fe(II)L)] complexes (L ≠ NCMe) with hydrazine. The reaction of 9 with hydrazine in THF, where bound triflate is labile and more prone to substitution, was explored. Triflate gets displaced by DMSO to yield 11 judged by a singlet peak in 19F NMR spectrum at – 77.79 ppm and the solid–state structure (Figure 3.22). The addition of 1 equiv of N2H4 to a solution of 9 in THF or DMSO did not form the corresponding mono(hydrazine) complex even with longer 55 reaction times. By adding 15 equiv of N2H4 to the DMSO-d6 solution of 9 in an NMR tube, the color changes instantly from blue-violet to orange. 31P{1H} NMR spectrum in DMSO-d6 after 10 min shows three multiplets at 44.22, 18.81, and 17.05 ppm with the ratio of 2:1:2 which suggest the formation of the mono(hydrazine) complex 12. Over time, a new singlet peak at 38.06 ppm in 31 P{1H} NMR spectrum emerges accompanied by free dmpe at –46.23 ppm (Figure 3.21). Figure 3.21 31P{1H} NMR spectrum of reaction of complex 9 in DMSO–d6 with 15 equiv of N2H4 in 10 min (green), 1 h (blue), and 12 h (red). Increasing the intensity of the peak at 38.06 ppm is indicative of the formation of the trishydrazine complex by displacing the dmpe ligand. Addition of THF to this DMSO solution with a ratio of 2:1 and slow diffusion of Et2O in 3 days, yields orange crystals suitable for X-ray crystallography. The X-ray structure shows the formation of tris(hydrazine) 10 (Figure 3.22). To the best of our knowledge, this is the first reported 56 tris(hydrazine) Fe complex. It is surprising that dmpe gets displaced by N2H4 in the presence of excess hydrazine but not excess ammonia; even though NH3 is a stronger base than N2H4. Figure 3.22 Molecular structures of complex 10 (top left) with thermal ellipsoids at the 50% probability level (Hydrogen atoms of carbons, counter ions and solvent molecules were omitted for clarity). Selected bond lengths (Å) and angles (deg) for complex 10 (top left): Fe–P1 2.225(1), Fe–P2 2.219(1), Fe–P3 2.231(1), Fe–N1 2.092(3), Fe–N3 2.071(4), Fe–N5 1.994(4), N1–N2 1.451(5), N3–N4 1.433(5); P1–Fe–P3 92.85(4), P1–Fe–P2 91.30(4), P1–Fe–N1 177.4(1), P1–Fe– N3 91.7(1), P1–Fe–N5 88.5(1), P3–Fe–P2 92.38(4), P3–Fe–N1 88.7(1), P3–Fe–N3 93.8(1), P3– Fe–N5 172.9(1), P2–Fe–N1 90.7(1), P2–Fe–N3 173.0(1), P2–Fe–N5 94.6(1), N1–Fe–N3 86.1(1), N1–Fe–N5 89.6(1), N3–Fe–N5 79.2(1), Fe–N3–N4 109.5(3). Complex 13 (top right): Fe–P1 2.219(2), Fe–P2 2.224(2), Fe–P3 2.224(2), Fe–N1 2.072(6), Fe–N3 2.082(6), Fe–N5 2.075(6); P1– Fe–P2 92.04(7), P1–Fe–P3 92.09(7), P1–Fe–N1 175.7(2), P1–Fe–N3 92.4(2), P1–Fe–N5 91.1(2), P2–Fe–P3 92.54(7), P2–Fe–N1 90.9(2), P2–Fe–N3 174.5(2), P2–Fe–N5 90.5(2), P3–Fe–N1 90.9(2), P3–Fe–N3 90.5(2), P3–Fe–N5 175.5(2), N1–Fe–N3 84.5(2), N1–Fe–N5 85.7(2), N3–Fe– N5 86.2(2). Complex 11 (bottom): Fe–P1 2.2497(6), Fe–P2 2.3096(5), Fe–P3 2.2953(6), Fe–P4 2.3126(6), Fe–P5 2.3331(5), Fe–O 2.149(2), S–O 1.519(2); P2–Fe–P3 89.99(2), P2–Fe–P1 88.21(2), P2–Fe–P5 175.07(2), P2–Fe–P4 96.87(2), P2–Fe–O1 91.10(4), P3–Fe–P1 92.45(2), P3– Fe–P5 89.58(2), P3–Fe–P4 169.26(2), P3–Fe–O1 86.79(4), P1–Fe–P5 96.71(2), P1–Fe–P4 96.01(2), P1–Fe–O1 178.97(5), P4–Fe–P5 82.88(2), P5–Fe–O1 83.98(4), P4–Fe–O1 84.83(4), Fe– O–S 138.8(1) 57 3.3. Conclusion An Fe(II) tris(ammine) complex supported by a tridentate phosphine ligand, [tSiP3Fe(NH3)3][OTf]2 (3), was synthesized, and its electrochemical behavior in the absence of NH3 was investigated. The CV of 3 in THF containing 0.1 M of TBAPF6 as supporting electrolyte shows multiple redox features. The same pattern was observed by changing the electrolyte to 0.1 M of TBAOTf as well. While electrocatalytic current decreases during AO in the presence of TBA- based electrolytes, no electrode passivation was observed by changing the supporting electrolyte to NH4OTf, which suggests the reaction of TBA with possible Fe intermediates during the catalysis. Analysis of headspace gases of BE of solutions containing 1 mM of 3 in the presence of NH3 and NH4OTf as supporting electrolyte reveals the generation of N2 as the anodic product. Rinse test and XPS analysis of the working electrode’s surface, obviate the possibility of heterogeneous AO catalysis. Chemical oxidation of 3 yields a paramagnetic complex, which upon addition of NH3 regenerates 3. Moreover, upon deprotonation of 3 with a hindered base, a diferrrous cation with the formula [(tSiP3Fe)2(μ-NH2)3][OTf] (4) was isolated, which is the first structurally characterized dinuclear iron complex with three NH2 ligands on the bridging position. To investigate the effect of ancillary ligands on the oxidation potential of 3, a mono(ammine) Fe(II) complex [tSiP3Fe(dmpe)(NH3)][OTf]2 (7) was synthesized and its electrochemical properties were compared to 3. Displacing two NH3 ligands in 3 with dmpe shifts the oxidation potential of the complex to higher potentials. 58 3.4. Synthesis 3.4.1. Synthesis of (dimethylphosphino)methyllithium The previously reported procedure was employed for the synthesis.78,79 Inside the glovebox, 12.2 g (16.27 ml, 160 mmol, 0.84 equiv) of PMe3 was added to a 100 mL one-neck flask charged with a stir bar and sealed with a rubber septum quickly. 100 mL of tBuLi (1.9 M in pentane, 190 mmol, 1 equiv) was added by syringe at room temperature, and the rubber septum was displaced by a glass stopper with a Teflon sleeve and sealed with a Teflon tape. During stirring, a white precipitate slowly forms. After 6 days, stirring was turned off, and the flask was allowed to stand overnight in the box, waiting for the solid to settle at the bottom of flask. The yellowish supernatant on top was removed by a pipette. The white solids were dried under reduced pressure. (11.28 g, 86%). Caution: (Dimethylphosphino)methyllithium is an extremely pyrophoric solid. It will catch fire upon exposure to air. In addition to solids, supernatant contains unreacted tBuLi and PMe3, which are highly flammable upon air exposure. Proper quenching procedures should be considered. Moreover, filtration of (dimethylphosphino)methyllithium inside the glove box should be avoided due to high static of solids, which will fly around the box’s atmosphere and will result in serious contamination. Also, the mass of solids should be calculated based on the weight difference of empty flask to avoid taking the solids out of flask. All of the synthesized (dimethylphosphino)methyllithium was used for the next reaction step due to its high reactivity. Using a rubber septum should be avoided for sealing due to its reaction with the compound. 59 3.4.2. Synthesis of ((tert-butylsilane)tris(methylene))tris(dimethylphosphane) The previously reported procedure was employed for the synthesis.55,78,79 In the glovebox, to a suspension of 11.27 g (137.42 mmol, 3.1 equiv) (dimethylphosphino)methyllithium in 250 mL Et2O was added a solution of 8.49 g (44.33 mmol, 1 equiv) tBuSiCl3 in Et2O dropwise with vigorous stirring over 20 min. After stirring for 15 more min, the solution was filtered over dry and activated neutral alumina to remove LiCl. The solvent was evaporated to yield a slightly yellow-tinted oil. (9.157 g, 66%). 1H NMR (500 MHz, C6D6) 𝛿 1.16 (br s, 9H), 1.01 (br s, 18H), 0.81 (s, 6H). 31P{1H} NMR (202 MHz, C6D6) 𝛿 –54.88. 60 3.4.3. Synthesis of [(tSiP3Fe)2(μ–Cl)3][Cl] (1) The procedure described by Boncella80 and McNeil55 was employed to synthesize [(tSiP3Fe)2(μ–Cl)3][Cl]. In the glove box, at room temperature, 4.445 g (35.076 mmol, 1.2 equiv) of FeCl2 was suspended in CH2Cl2 and, while stirring, 9.157 g (29.23 mmol, 1 equiv) of ((tert- butylsilane)tris(methylene))tris(dimethylphosphane) was added dropwise. Upon addition, the color changed to deep purple. After stirring for 1 h, the solution was passed through activated neutral alumina to remove excess FeCl2. After removing the solvent, the purple solid was washed first with 10 mL Et2O and then 10 mL pentane and dried under reduced pressure (11.83 g, 92%). 31 P{1H} NMR (202 MHz, MeCN-d3) 𝛿 43.58. 1H NMR (500 MHz, MeCN–d3) 𝛿 1.53 (m, 36H), 0.8 (s, 18H), 0.62 (m, 12H). 61 3.4.4. Synthesis of [tSiP3Fe(NCMe)3][OTf]2 (2) In the glove box, at room temperature, 1 (5.00 g, 5.718 mmol, 1 equiv) was dissolved in MeCN (200 mL) and a solution of AgOTf (6.024 g, 23.446 mmol) in MeCN was added dropwise. After 5 h of stirring at room temperature, the solution was filtered through activated neutral alumina using a fine glass frit to remove precipitated AgCl which yields a clear yellow solution. The MeCN solution was reduced in volume in vacuo and ether was added to precipitate a yellow fine powder. The yellow powder was collected on a frit and washed with ether, and was dried under vacuum (7.509 g, 83%). 1H NMR (500 MHz, MeCN-d3) 𝛿 1.96 (s, 9H), 1.48 (m, 18H), 0.90 (s, 9H), 0.90– 0.87 (m, 6H). 1H NMR (500 MHz, MeNO2-d3) 𝛿 2.49 (s, 9H), 1.59 (m, 18H), 0.98 (m, 6H), 0.92 (s, 9H). 13C{1H} NMR (126 MHz, MeCN-d3) 𝛿 130.80 (s), 26.03 (s), 18.77 (m), 17.32 (q, J = 5.7 31 Hz), 6.30 (dd, J = 7.2, 3.8 Hz), 4.57 (s). P{1H} NMR (202 MHz, MeCN-d3) 𝛿 34.24. 31P{1H} NMR (202 MHz, MeNO2–d3) 𝛿 33.90. 19F NMR (470 MHz, MeCN-d3) 𝛿 –79.09 (s). Anal. Calcd for C21H42F6FeN3O6P3S2Si.Et2O: C, 34.85; H, 6.08; N, 4.88. Found: C, 35.03; H, 5.58; N, 4.94. 62 3.4.5. Synthesis of [tSiP3Fe(NH3)3][OTf]2 (3) A 250 mL Schenck flask was charged with a stir bar and 2 in the glovebox and was sealed by a rubber septum. Then, the flask was taken out of the box and put under an N2 atmosphere. 40 mL of degassed and dry methanol was added by syringe. After dissolving all the 2 in methanol, under stirring NH3 was bubbled into the solution. The solution became hot, and the color changed from yellow to deep orange. After stirring for 15 min under continuous NH3 bubbling, the NH3 needle was taken out of flask. The N2 pressure was increased, and the septum of the flask was removed. 200 mL of Et2O was added at once under vigorous stirring. Orange microcrystalline powder precipitated, and the septum was replaced. The stirring was turned off, and the flask was put in dry ice/acetone bath to precipitate additional product, then the solvent was removed by cannula filtration. Orange solids were dried under vacuum and transferred to the glovebox (2.15 g, 78%). 1 H NMR (500 MHz, DMSO-d6) 𝛿 2.03 (s, 9H), 1.38 (m, 18H), 0.82 (s, 9H), 0.76 (m, 6H). 1H NMR (500 MHz, MeNO2-d3) 𝛿 1.95 (s, 9H), 1.56 (m, 18H), 0.97 (m, 6H), 0.92 (s, 9H). 13C{1H} NMR (126 MHz, DMSO-d6) δ 25.60 (s), 17.96 (dd, J = 12.1, 6.4 Hz), 16.16 (q, J = 5.6 Hz), 7.53 (s). 31 P{1H} NMR (202 MHz, MeNO2-d3) 𝛿 38.07. 31P{1H} NMR (202 MHz, DMSO-d6) 𝛿 41.79. 19F NMR (470 MHz, DMSO-d6) 𝛿 –77.73 (s). Anal. Calcd for C15H42F6FeN3O6P3S2Si: C, 25.18; H, 5.92; N, 5.87. Found: C, 25.31; H, 5.89; N, 5.43. 63 3.4.6. Synthesis of [(tSiP3)2Fe2(μ–NH2)3][OTf] (4) 3 (219 mg, 0.306 mmol, 1 equiv) was suspended in 4 mL THF at room temperature, and 1 M solution of LiN(SiMe3)2 in THF (0.5 mL, 0.490 mmol, 1.6 equiv) was added slowly to yield a brown solution. After stirring for 10 min, the mixture was filtered, and the filtrate was layered with pentane and kept in the glove box’s freezer at –32 °C. After 3 d, the orange precipitate was collected and washed with ether until the washings were colorless. Then, the powder was dried under reduced pressure (76 mg, 53%). 1H NMR (500 MHz, THF-d6) 𝛿 1.42 (s, 36H), 0.80 (s, 18H), 0.64 (s, 12H), –2.63 (s, 6H). 1H NMR (500 MHz, DMSO-d6) 𝛿 1.34 (s, 36H), 0.74 (s, 18H), 0.56 (s, 12H), –2.79 (s, 6H). 13C{1H} NMR (126 MHz, THF-d8) δ 26.38 (s), 19.68 (s), 17.29 (s), 10.51 (s). 31P{1H} NMR (202 MHz, DMSO–d6) 𝛿 43.75 (s). 31P{1H} NMR (202 MHz, THF-d8) 𝛿 44.48 (s). 19 F NMR (470 MHz, DMSO-d6) 𝛿 –77.79 (s). Anal. Calcd for C27H72F3Fe2N3O3P6SSi2.LiCF3SO3: C, 30.98; H, 6.69; N, 3.87. Found: C, 31.10; H, 6.69; N 3.40. 64 3.4.7. Synthesis of [tSiP3Fe(dmpe)(NCMe)][OTf]2 (6) In a glove box, 2 (0.53 g, 0.677 mmol), dmpe (0.10 g, 0.677 mmol) and 10 mL of CH2Cl2 were placed in a pressure tube containing a magnetic stir bar and taken out of the box. The mixture was stirred at 60 °C for 12 h. Removing the solvent under reduced pressure afforded a yellow powder that was washed with Et2O (2 × 3 mL) and dried under vacuum (0.475 g, 82%). 1H NMR (500 MHz, MeCN-d3) 𝛿 2.02–1.95 (m, 4H), 1.75 (d, J = 6.3 Hz, 6H), 1.72 (d, J = 7.9 Hz, 6H), 1.57 (d, J = 7.1 Hz, 6H), 1.46 (d, J = 7.3 Hz, 6H), 1.39 (d, J = 8.1 Hz, 6H), 1.21 (m, 4H), 0.96 (s, 9H), 0.87 (d, J = 12.8 Hz, 2H). 31P{1H} NMR (202 MHz, MeCN-d3) 𝛿 45.89 (m, 2P), 19.07 (m, 1P), 15.92 (m, 2P). 13C{1H} NMR (126 MHz, MeCN-d3) δ 136.37(s), 30.57 (dd, J = 23.4, 18.3 Hz), 26.57 (d, J = 24.1 Hz), 25.83 (s), 24.79 (m), 19.97 (m), 17.08 (q, J = 5.1 Hz), 15.96 (m), 12.63 (d, J = 7.6 Hz), 10.10 (s), 5.65 (s). 19 F NMR (470 MHz, MeCN-d3) 𝛿 –78.98 (s). Anal. Calcd for C23H52F6FeNO6P5S2Si: C, 32.29; H, 6.13; N, 1.64. Found: C, 32.42; H, 6.13; N, 1.72. 65 3.4.8. [tSiP3Fe(dmpe)(Cl)][Cl] (8) Following the previously reported procedure,55 in a glove box, 1 (0.85 g, 0.965 mmol), dmpe (0.14 g, 0.965 mmol) and 25 mL of CH2Cl2 were placed in a pressure tube containing a magnetic stir bar and taken out of the box. The mixture was stirred at 60 °C for 12 h. Removing the solvent under reduced pressure afforded a crimson powder that was washed with Et2O (2 × 3 mL) and dried under vacuum. (0.481 g, 85%). Spectroscopic data were consistent with the literature. 66 3.4.9. Synthesis of [tSiP3Fe(dmpe)(OTf)][OTf] (9) 8 (204 mg, 0.347 mmol, 1 equiv) was dissolved in 8 mL of 1:1 THF:DCM, and AgOTf (179 mg, 0.694 mmol, 2 equiv) was added to the solution. After 10 min of vigorous stirring at room temperature in which color changes from crimson to blue-purple, the solution was filtered over Celite to yield a deep violet solution. Removing solvent in vacuo yields 9 (249 mg, 88%). 31P{1H} NMR (202 MHz, THF-d8) 𝛿 39.82 (br, 2P), 27.29 (br, 1P), 16.23 (br, 2P). 19F NMR (470 MHz, THF-d8) 𝛿 –61.59 (s), –75.34 (br). Anal. Calcd for C21H49F6FeO6P5S2Si: C, 30.97; H, 6.06. Found: C, 31.81; H, 5.90. Due to peak broadness (presumably because of solvent exchange with bound OTf) no satisfactory 1H and 13C NMR spectra were taken. 67 3.4.10. Synthesis of [tSiP3Fe(dmpe)(NH3)][OTf] (7) To a solution of 9 in THF, NH3 was bubbled. The Color changed from blue-purple to orange. The solvent was removed in vacuo, and orange solids were washed with Et2O and dried under reduced pressure (0.251 g, 94%). 1H NMR (500 MHz, DMSO-d6) 𝛿 2.01 (m, 2H), 1.88 (m, 2H), 1.72 (d, J = 5.6 Hz, 6H), 1.67 (d, J = 7.8 Hz, 6H), 1.53 (m, 6H), 1.36 (m, 6H), 1.29 (d, J = 7.6 Hz, 6H), 1.24 – 1.17 (m, 2H), 1.17 – 1.10 (m, 2H), 0.91 (s, 9H), 0.82 (t, J = 3.8 Hz, 3H), 0.77 (d, J = 31 12.0 Hz, 2H). P NMR (202 MHz, DMSO-d6) 𝛿 48.78 (m, 2P), 20.92 (m, 1P), 19.49 (m, 2P). 13 C{1H} NMR (126 MHz, DMSO-d6) δ 29.02 (m), 26.76 (d, J = 22.1 Hz), 25.40 (s), 23.88 (dd, J = 15.9, 9.8 Hz), 20.10 (m), 18.03 (m), 16.19 (q, J = 5.0 Hz), 14.45 (m), 12.45 (d, J = 5.4 Hz), 9.53 (s). Anal. Calcd for C21H52F6FeNO6P5S2Si: C, 30.33; H, 6.30; N, 1.68. Found: C, 29.49; H, 5.92; N 1.33. 68 Chapter 4. CATALYTIC AMMONIA OXIDATION BY A MONONUCLEAR RUTHENIUM COMPLEX SUPPORTED BY AN ISOINDOLE-BASED LIGAND 69 4.1. Introduction Homogeneous electrocatalytic ammonia oxidation (AO) by mononuclear mono(ammine) Ru complexes has been reported previously.81,82 Scheme 4.1 summarizes some of the selected Ru- based AO electrocatalysts.81,82 Among the various proposed mechanisms for catalytic AO, hydrazine/hydrazido and nitride pathways have attracted the most attention. As shown in (Scheme 4.2), in the hydrazine/hydrazido cycle, nucleophilic attack of NH3 on the imido complex (which is an en route intermediate of Ru(III) ammine disproportionation to Ru(II)–NH3 and Ru(IV)–NH) results in N-N bond formation to yield a Ru(II)–N2H4 complex. Hydrazine complexes can be oxidized to their diazene counterparts, while further oxidation can generate dinitrogen. In the nitride pathway, consecutive 3 e– oxidation and proton transfer from Ru(II)–NH3 yields a Ru(IV) nitride intermediate. Intermolecular coupling of two Ru(IV) nitride complexes forms a dinitrogen bridge intermediate. Further displacement of N2 by NH3 completes the catalytic cycle. Scheme 4.1 Ru mononuclear AO electrocatalysts 70 Scheme 4.2 Proposed electrocatalytic AO by hydrzine/hydrazido mechanism Smith and co-workers proposed a hydrazine/hydrazido mechanism based on the previous works of Meyer83–86 for a Ru(II)–NH3 complex supported by polypyridyl ligands (Scheme 4.2). Indeed, in 1981, Meyer and co-workers published a paper on the oxidation of coordinated NH3 in [Ru(trpy)(bpy)NH3]2+ to nitrate by nucleophilic attack of H2O on the Ru(IV) imido complex in aqueous media. So far, the Ru(IV) imido intermediate in these Ru ammine polypyridyl systems has not been isolated; however, a related [OsIV(trpy)(bpy)(NNR)]2+ intermediate has been trapped in the presence of a secondary amine in place of NH3.87 71 In 2019, Nishibayashi and co-workers proposed the nitride pathway for a Ru AO catalyst supported by a 2,2 ′-bipyridyl-6,6′-dicarboxylate ligand (Scheme 4.3).81 The N2–bridged bimetallic Ru complex was synthesized and characterized independently by oxidizing the Ru ammine complex with tris(4-bromophenyl)amine radical cation in the presence of NH4PF6 and 2,4,6– collidine (as the base) in MeCN at –40 °C. In addition to the coupling of two Ru nitride complexes, there are some reported examples of the formation of the N–N bond via nucleophilic attack on Ru nitride, which are shown in Scheme 4.4.88–90 For stabilizing the high oxidation states of Ru in its nitride complexes, the supporting ligands should be good electron donors. Salen ligands (Scheme 4.4), for example, are good candidates, which have been used widely in oxidation chemistry of transition metals.89–92 Scheme 4.3 Proposed electrocatalytic AO by Nishibayashi and co-workers81 72 Scheme 4.4 Examples of the formation of an N–N bond via nucleophilic attack on Ru nitride In this chapter, the synthesis, characterization, and electrocatalytic activity of a Ru ammine complex supported by an isoindole-based ligand will be discussed. The catalytic onset potential of the complex toward NH3 oxidation in different organic solvents has been accessed and compared to other reported AO Ru electrocatalysts. Chemical oxidation of the Ru(II) ammine complex has also been investigated. For characterizing the products of NH3 oxidation in the presence of the catalyst, bulk electrolysis has been conducted, and the release of dinitrogen gas as the anodic product of AO has been monitored by gas analysis of the electrochemical cell’s headspace. 73 4.2. Results and Discussion One strategy to lower the overpotential of electrocatalytic NH3 oxidation is the synthesis of complexes with a reduced net charge. Despite 2,2':6',2''-terpyridine (trpy) which is a rigid and neutral tridentate ligand, (1-Z,3-Z)-N1,N3-bis(pyridin-2-yl)isoindolin-1,3-diimine (Hbid) bears a negative charge once deprotonated and also has a larger chelate bite angle compare to trpy. Also, the formation of Ru(bid)2n+ is less likely than Ru(trpy)2n+ during catalysis due to trans-influence effects. The Hbid ligand was prepared by the previously reported reaction by Siegl (Scheme 4.5).93,94 Scheme 4.5 One-pot Synthesis of Hbid ligand The reaction of Hbid with RuCl3•3H2O in ethanol yields [Ru(bid)Cl3]H+ as a brown color powder which has been used as the precursor for the synthesis of Ru catalysts.95,96 Ru(bid)(bpy’)Cl (1) was prepared by previously reported procedure by llobet and co-workers via the reaction of [Ru(bid)Cl3]H+ with 1 equiv of 4,4'-dimethoxy-2,2'-bipyridine (bpy’).97 Displacement of Cl– with NH3 in 1 followed by anion exchange to PF6 gives blue-purple [Ru(bid)(bpy’)NH3][PF6] (2) in 83% isolated yield (Scheme 4.6). 74 Scheme 4.6 Preparation of [Ru(bid)(bpy’)NH3][PF6] (2) Single crystals of 2 suitable for X-ray crystallography were prepared by slow vapor diffusion of diethyl ether into the dichloromethane solution of 2. Figure 4.1 exhibits the solid-state structure of 2, and its caption summarizes its selected bond lengths and bond angles. Figure 4.1 Crystal structure of complex 2 with thermal ellipsoids at the 50% probability level. Solvent molecules and hydrogens on carbon atoms have been removed for clarity. Selected bond lengths (Å) and angles (deg) for complex 2: Ru–N1 2.156(3), Ru–N2 2.048(2), Ru–N3 2.098(2), Ru–N4 2.087(3), Ru–N6 2.013(2), Ru–N8 2.093(2), N5–C17 1.381(3), N5–C18 1.298(3), N6– C18 1.380(3), N6–C25 1.367(3), N7–C25 1.301(4), N7–C26 1.377(3). Bond angles: N1–Ru–N2 178.00(9), N1–Ru–N3 100.65(9), N2–Ru–N6 94.44(9), N4–Ru–N8 177.26(9), N3–Ru–N4 91.35(9), N6–Ru–N8 89.05(9) 75 40µ 20µ 30µ 10µ Current (A) Current (A) 20µ 0 10µ -10µ 0 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 Potential (V vs. Fc*) Potential (V vs. Fc*) 1 V/s 500 mV /s 60.0µ 400 mV /s 300 mV /s 200 mV /s 40.0µ 100 mV /s 50 mV /s 20.0µ 10 mV /s Current (A) 0.0 -20.0µ -40.0µ -60.0µ -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 Potential (V vs Fc*) Figure 4.2 Left: CV of 1 mM of Complex 2 in the presence of 0.2 mM of Fc* as the internal standard in THF containing 1 M of NH4OTF as supporting electrolyte. Right: correspondent SQW has taken instantly after CV. WE: GC, CE: Pt disk, scan rate: 100 mV/s The cyclic voltammogram of complex 2 is shown in Figure 4.2. One anodic peak at the potential of Epa = 0.49 V (vs. Fc*) and one reverse cathodic peak at the potential of Epc = 0.41 V (vs. Fc*) with the E1/2 = 0.45 V (vs. Fc*) were observed. The diffusion coefficient (D∘) of 2 in THF containing 1 M of NH4OTf, was calculated by Randles–Sevcik equation: 1 2 𝑛𝐹𝜈𝐷∘ 2 𝑠𝑙𝑜𝑝𝑒 𝑅𝑇 𝑖𝑝 = 0.446𝑛𝐹𝐴 ( ) 𝐷∘ = ( ° ) 𝑅𝑇 0.4463𝑛𝐹𝐴𝐶 𝑛𝐹 76 which n is the number of electrons transferred in the redox event (n = 1), F is the Faradic constant (F = 96485.3321 C.mol-1), ν is the scan rate (V.s-1), A is the electrode’s geometrical surface area (A = 0.07068 cm2), C∘ is the bulk concentration of analyte (1 mol.cm-3), R is the gas constant (R = 8.3144621 J.K-1.mol-1), and T is the temperature (T = 295.37 K). By measuring the CV of 2 with different scan rates (Figure 4.3), D∘ was calculated to be 9.19 × 10-6 cm2s-1 for oxidation of 2 and 4.08 × 10-6 cm2s-1 for reduction of [Ru(bid)(bpy’)NH3][PF6][OTf]. 0 60 50 -10 40 ip Cathode (A) ip Anode (A) -20 30 slope = -38.54106 slope = 57.85372 R2 = 0.99974 R2 = 0.99989 -30 20 10 -40 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 1/2 (V1/2s-1/2) 1/2 (V1/2s-1/2) Figure 4.3 Plotting cathodic (left) and anodic (right) peak currents (ip) versus square root of scan rate (ν1/2) for complex 2 in THF containing 1 M of NH4OTf result in a linear line that from their slops diffusion coefficients were calculated to be 9.19 × 10-6 cm2s-1 for oxidation of 2 and 4.08 × 10-6 cm2s-1 for reduction of [Ru(bid)(bpy’)NH3][PF6][OTf]. WE: GC, CE: Pt disk By bubbling NH3 through the solution ([NH3] = 2.8 M) after taking the CV of 2 in THF with 1 M NH4OTf as the supporting electrolyte and glassy carbon (GC) as the working electrode, a catalytic current was obtained with an onset of 0.32 V (vs. Fc*) (Figure 4.4). To confirm that the catalysis is homogeneous and not related to the formation of Ru species on the surface of the GC electrode, a rinse test was performed by rinsing the GC electrode with THF and inserting it back into the solution of THF with 2.8 M NH3 and 1 M NH4OTf with no catalyst present. No catalytic current was observed after rinsing, which suggests that the catalysis is homogeneous. The peak 77 shape of the catalytic current resembles the KT-type electrocatalysis operating by an EC’ mechanism.98,99 Due to the fast kinetics of electrocatalysis in KT, the oxidized form of electrocatalyst consumes all available substrate, in this case NH3, within the reaction diffusion layer. Because all substrate is consumed and there is no available substrate left for the electrocatalyst to react with, the oxidized electrocatalyst is regenerated at potentials anodic to the catalysis-initiating redox event.98 This may explain the nature of the second wave in Figure 4.4, which occurs at a potential of 0.6 V (vs. Fc*). The mechanism of electrocatalytic NH3 oxidation, however, may be more complicated than the simplified KT mechanism as it is a 6 e–, and 6-proton multistep reaction. 100.0µ 1 mM 2 1 mM 2 + 2.8 M NH3 60.0µ 80.0µ 60.0µ 40.0µ Current (A) Current (A) 40.0µ 20.0µ 20.0µ 0.0 0.0 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 Potential (V vs. Fc*) Potential (V vs. Fc*) Figure 4.4 Left: CV (blue line) of 1 mM solution of complex 2 in THF with 1 M of NH4OTf as supporting electrolyte and 0.2 M of Fc* as internal standard; CV (orange line) after bubbling NH3 ([NH3] = 2.8 M] through a 1 mM solution of complex 2 in THF with 1 M NH4OTf as supporting electrolyte and 0.2 M Fc*. Right: SQW of 1 mM solution of complex 2 in THF with 1 M NH4OTf as supporting electrolyte and 0.2 M Fc* in the presence of 2.8 M NH3. WE: GC, CE: Pt disk, scan rate: 100 mV/s 78 200µ 1 V/s 500 mV/s 400 mV/s 300 mV/s 200 mV/s 100 mV/s 100µ 50 mV/s Current (A) 10 mV/s 0 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 Potential (V vs. Fc*) Figure 4.5 CV of 1 mM solution of complex 2 with different scan rates in THF with 1 M NH4OTf as supporting electrolyte and 0.2 M Fc* in the presence of 2.8 M NH3. WE: GC, CE: Pt disk 0 M NH3 0 M NH3 0.028 M NH3 0.028 M NH3 100.0µ 0.055 M NH3 0.055 M NH3 0.082 M NH3 0.082 M NH3 80.0µ 60.0µ 0.108 M NH3 0.108 M NH3 0.133 M NH3 0.133 M NH3 60.0µ 0.158 M NH3 0.158 M NH3 Current (A) Current (A) 0.183 M NH3 40.0µ 0.183 M NH3 40.0µ 0.207 M NH3 0.207 M NH3 0.231 M NH3 0.231 M NH3 20.0µ 0.255 M NH3 0.255 M NH3 20.0µ 2.8 M M NH3 2.8 M NH3 0.0 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 Potential (V vs. Fc*) Potential (V vs. Fc*) Figure 4.6 Left: CV of complex 2 in THF with 1 M NH4OTf and different concentrations of NH3in the presence of 0.2 mM Fc* as internal standard. Right: corresponding SQW has taken instantly after each CV. WE: GC, CE: Pt disk, scan rate: 100 mV/s 79 70µ 0.00 M 2 0.00 M 2 90.0µ 0.09 M 2 0.09 M 2 0.19 M 2 60µ 0.19 M 2 0.29 M 2 0.29 M 2 0.38 M 2 50µ 0.38 M 2 60.0µ 0.47 M 2 0.47 M 2 0.56 M 2 0.56 M 2 Current (A) Current (A) 40µ 0.65 M 2 0.65 M 2 0.74 M 2 30µ 0.74 M 2 30.0µ 0.82 M 2 0.82 M 2 0.91 M 2 0.91 M 2 20µ 0.0 10µ 0 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 -0.2 0.0 0.2 0.4 0.6 0.8 Potential (V vs. Fc*) Potential (V vs. Fc*) 70.0m 40.0m y = 0.064 x + 0.005 y = 0.0391 x + 0.004 SQW Current at 0.39 V vs. Fc* (mA) SQW Current at 0.62 V vs. Fc* (mA) 60.0m R2 = 0.99537 35.0m R2 = 0.99949 50.0m 30.0m 40.0m 25.0m 30.0m 20.0m 20.0m 15.0m 10.0m 10.0m 5.0m 0.0 0.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 Concentration (mM) of Complex 2 Concentration (mM) of Complex 2 Figure 4.7 Top left: CV of various concentrations of complex 2 in THF containing 2.8 M NH3, 1 M NH4OTf, and 0.2 mM Fc* as internal standard. Top right: corresponding SQW taken after each CV. Bottom left: SQW current of the first peak at 0.39 V (vs. Fc*) plotted vs. concentration of complex 2. Bottom right: SQW current of the second peak at 0.62 V (vs. Fc*) plotted vs. concentration of complex 2. WE: GC, CE: Pt disk, scan rate: 100 mV/s Figure 4.5 depicts the CV with different scan rates of 1 mM 2 in THF in the presence of 2.8 M NH3 and 1 M NH4OTf as supporting electrolyte. The second peak was observed at all scan rates starting from 1 V/s to 10 mV/s. The CV of 2 was also taken in THF with different concentrations of NH3 (Figure 4.6). By adding a small concentration of NH3 to the solution of 1 mM 2 in THF, two anodic peaks were observed. 80 The CV of 2 at different concentrations was also examined in THF with 2.8 M NH3 and 1 M NH4OTf. As shown in Figure 4.7, both peaks at 0.39 V (vs. Fc*) and 0.62 V (vs. Fc*) have a linear dependence on the concentration of 2. 7.5m 30 20 Electric Charge (A.s) Current (A) 7.0m 10 0 6.5m 0 1000 2000 3000 4000 0 1000 2000 3000 4000 Time (s) Time (s) Figure 4.8 Left: BE of THF solution with 2.8 M NH3 in the presence of 1 mM complex 2 for 1 h by applying 0.8 V (vs. Fc*). Right: the amount of the charge passed (C) during 1 h of BE. Supporting electrolyte: 1 M NH4OTf, WE: GC plate, CE: Pt mesh For quantifying the amount of produced N2 on the anode, controlled potential electrolysis (bulk electrolysis–BE) was conducted in THF with 2.8 M NH3, 1 M NH4OTf, and 1 mM 2 (Figure 4.8). The potential was fixed at 0.8 V (vs. Fc*), and a GC plate and Pt mesh were used as working and counter electrodes, respectively. During 1 h of BE, 25.5374 Columb of charge was passed. Analyzing the headspace gas reveals the formation of 41 µmol of N2 with 94% faradic efficiency. BE of 2 was also done in MeCN under the same conditions. In the first minutes of BE, the color of the solution changed from blue-purple to red-violet. For characterization of the formed species during BE, water was added to the BE solution for precipitation of solid and removal of dissolved NH4OTf. After filtering the red-violet solids and taking 1H NMR in MeCN-d3, the product was characterized as [Ru(bid)(bpy’)(MeCN)][X], (X = OTf– or PF6–). This was confirmed 81 by comparing the 1H NMR of independently synthesized [Ru(bid)(bpy’)(MeCN)][ PF6] (Figure 4.9). Figure 4.9 Orange: 1H NMR of independently synthesized [Ru(bid)(bpy’)(MeCN)][PF6]; Blue: 1 H NMR of product form in the first minutes of BE of complex 2 in acetonitrile Complex 2 was chemically oxidized by tris(4-bromophenyl)amine radical cation (Scheme 4.7). Upon addition of the oxidant to the blue-purple solution of 2 in THF, the color changes to brown. The oxidized form of 2 (complex 3) was isolated as a brown powder with 91% yield. 82 Scheme 4.7 Chemical oxidation of 2 with tris(4-bromophenyl)amine radical cation Upon addition of NH3 to the brown solution of 3 in THF, the color changed instantly to blue- purple to regenerate 2. Figure 4.10 shows the UV-vis spectra of 2 and 3 before and after addition of NH3. After the addition of excess NH3 to THF solution of 3, the UV–vis spectrum of product matches with the spectrum of 2. Interestingly, the brown color of the solution of 3 in THF slowly changes to green in hours. Slow evaporation of THF yields some green crystals suitable for X-ray crystallography. The structure of complex reveals the protonation of the imine nitrogen of the bid ligand to give ([Ru(Hbid)(bpy’)NH3][PF6]2) (4) (Figure 4.11). Because the ammine ligand coordinated to a Ru(III) center is acidic56, its deprotonation by a basic imine nitrogen of bid ligand is not surprising. The UV–vis spectrum of 4 is shown in Figure 4.10. Upon addition of excess NH3 to the THF solution, the color changes from green to blue-purple. The UV-vis spectrum of the blue–purple solution matches well with the spectrum of 2. The other products of self-deprotonation of 3 can be characterized by 15N NMR of its correspondent 15NH3 labeled complex. Scheme 4.8 summarizes the overall oxidation, reduction, protonation, and deprotonation of the Ru complexes. 83 25000 complex 2 25000 complex 3 20000 20000 Extinction Coefficient (M-1.cm-1) Extinction Coefficient (M-1.cm-1) 15000 15000 10000 10000 5000 5000 0 0 200 300 400 500 600 700 800 200 300 400 500 600 700 800 Wavelength (nm) Wavelength (nm) 25000 25000 complex 3 (after 72 h) complex 3 + NH3 20000 Extinction Coefficient (M-1.cm-1) 20000 Extinction Coefficient (M-1.cm-1) 15000 15000 10000 10000 5000 5000 0 0 200 300 400 500 600 700 800 200 300 400 500 600 700 800 Wavelength (nm) Wavelength (nm) 25000 complex 3 (after 72 h) + NH3 20000 Extinction Coefficient (M-1.cm-1) 15000 10000 5000 0 200 300 400 500 600 700 800 Wavelength (nm) Figure 4.10 Top left: UV-vis spectrum of complex 2. Top right: UV-vis spectrum of complex 3. Middle left: UV-vis spectrum of complex 3 after addition of excess amount of NH3. Middle right: UV-vis spectrum of complex 3 in THF taken after 72 h. Bottom left: UV-vis spectrum after addition of NH3 to complex 3 which has been in THF for 72 h 84 Figure 4.11 Crystal structure of complex 4 with thermal ellipsoids at the 50% probability level. Solvent molecules and hydrogens on carbon atoms have been removed for clarity. Selected bond lengths (Å) and angles (deg) for complex 4: Ru–N1 2.147(5), Ru–N2 2.066(5), Ru–N3 2.063(4), Ru–N4 2.100(4), Ru–N6 2.020(4), Ru–N8 2.078(4), N5–C17 1.390(7), N5–C18 1.343(7), N6– C18 1.323(6), N6–C25 1.406(6), N7–C25 1.279(7), N7–C26 1.384(7). Bond angles: N1–Ru–N2 172.9(2), N1–Ru–N3 96.4(2), N2–Ru–N6 100.8(2), N4–Ru–N8 177.5(2), N3–Ru–N4 92.5 (2), N6–Ru–N8 89.1(2) Scheme 4.8 Overall oxidation, reduction, protonation, and deprotonation of Ru Bid complex 85 Figure 4.12 1H NMR of complex 3 in THF-d6 in different time intervals at room temperature. Crimson: t = 0 h; green: t = 20 h; blue: t = 96 h. In Figure 4.12, room temperature 1H NMR of 3 in THF-d6 at t = 0 h shows broad peaks in the chemical shift range of –40 to 40 ppm, which is the characteristic of a paramagnetic Ru(III) complexe. Over hours, the intensity of the broad peaks decreases while new peaks emerge in the diamagnetic range of 0 to 14 ppm. After 96 h, no broad peaks corresponding to paramagnetic species were observed. The growth of a peak at 12.8 ppm can be assigned to the protons of the protonated imine of the bid ligand in 4. 86 4.3. Conclusion An isoindole-based Ru ammine complex, [Ru(bid)(bpy’)NH3][PF6] (2), was prepared and its electrocatalytic activity was evaluated in the presence of NH3 in THF. The cyclic voltammogram of 2 in THF containing 1 M of NH4OTf as the supporting electrolyte shows one peak with the E1/2 = 0.45 V (vs. Fc*). The addition of 2 to the NH3-saturated THF solution shifted the anodic current of NH3 oxidation to 0.32 V vs. Fc*. Chemical oxidation of 2 by tris(4-bromophenyl)amine radical cation yielded a paramagnetic Ru(III) complex (3), which upon addition of NH3 regenerated the Ru(II) ammine complex (2). Analysis of the headspace gases of the electrochemical cell after bulk electrolysis of the NH3-saturated THF solutions showed the release of N2 gas as the anodic product. In the absence of NH3, the Ru(III) ammine complex (3) undergoes self–deprotonation, suggested by X-ray crystallography, NMR, and UV spectroscopy. Conducting bulk electrolysis experiments of 1 mM solutions of 2 in MeCN saturated with NH3 resulted in the displacement of MeCN after one cycle of the electrocatalysis and confirmed by NMR spectroscopy. 87 4.4. Synthesis 4.4.1. General Synthesis Hbid,93,94 [Ru(bid)Cl3][H+],95,96 and Ru(bid)(bpy’)Cl97 (1) were synthesized according to previously reported procedures. 4.4.2. Synthesis of [Ru(bid)(bpy’)NH3]PF6, (complex 2) A pressure tube was charged with 1 (1.015 g, 1.5589 mmol), a stir bar, and 50 mL of NH4OH. The pressure tube was sealed and heated to 98 °C for 1 h under stirring. The tube was cooled to room temperature and opened to air. An H2O-saturated solution of NH4PF6 was added to the solution to precipitate the PF6 salt of 2. Solids were collected and washed with ether to give blue- purple powder (970 mg, 83%). 1 H NMR (500 MHz, MeCN-d3) δ 9.04 (d, 1H), 8.19 (dd, 2H), 8.05 (d, 1H), 7.76 (d, 1H), 7.74 – 7.67 (m, 4H), 7.63 – 7.57 (m, 2H), 7.55 (d, 1H), 7.54 – 7.50 (m, 2H), 7.03 (d, 1H), 6.56 (td, 2H), 6.48 (dd, 1H), 4.17 (s, 3H), 3.77 (s, 3H), 1.83 (s, 3H). Anal. Calcd for C30H27F6N8O2PRu.(Et2O)1/2: C, 47.18; H, 3.96; N, 13.75. Found: C, 47.59; H, 3.68; N, 13.48. 4.4.3. Synthesis of [Ru(bid)(bpy’)MeCN]PF6 A solution of TlPF6 (48 mg, 0.1381 mmol, 1.1 equiv) in 5 mL of MeCN was added to a solution of 1 (100 mg, 0.1256 mmol, 1 equiv) in 10 mL of MeCN under stirring. The mixture was heated to 50 °C for 1 h. During the reaction, the color changed from green-blue to red-purple. The solution was filtered over Celite to remove TlCl. MeCN was removed in vacuo and the solids were washed with water to remove any excess TlPF6. The solids were dried under reduced pressure (97 mg, 96%). 1H NMR (500 MHz, MeCN-d3) δ 9.30 (d, 1H), 8.17 – 8.10 (m, 2H), 8.06 (d, 1H), 7.76 (d, 1H), 7.73 – 7.66 (m, 2H), 7.65 – 7.57 (m, 4H), 7.56 – 7.48 (m, 3H), 7.19 (d, 1H), 6.61 (m, 2H), 6.53 (dd, 1H), 4.17 (s, 3H), 3.77 (s, 3H), 2.06 (s, 3H). 88 Chapter 5. DINUCLEAR RUTHENIUM BIS(AMMINE) COMPLEXES FOR INTRAMOLECULAR N–N COUPLING 89 5.1. Introduction Intramolecular N-N coupling of coordinated NH3 ligands in a dimeric Ru complex containing a Ru-O-Ru motif has been reported by Meyer and co-workers.84,85 Electrochemical oxidation of a 15 1:1 mixture of N labeled and unlabeled cis,cis-[(bpy)2(H3N)RuIIIORuIII(NH3)(bpy)2]4+ in 15 14 aqueous solution resulted in the formation of N2 and N2 in equimolar ratios; this result establishes that intramolecular dinitrogen formation is the sole pathway for this system.85 Scheme 5.1 Intramolecular N–N coupling by cis,cis-[(bpy)2(H3N)RuIIIORuIII(NH3)(bpy)2]4+ Although this experiment suggests the intramolecular N–N coupling of NH3 ligand, it should be noted that there was no free NH3 in the solution to assess the possibility of N–N coupling of free NH3 and bound ammine. 90 We have begun investigating the possible electrocatalytic behavior of cis,cis- [(bpy)2(H3N)RuIIIORuIII(NH3)(bpy)2]4+ toward NH3 oxidation in nonaqueous solvents saturated with NH3. Preliminary results indicate that cis,cis-[(bpy)2(H3N)RuIIIORuIII(NH3)(bpy)2]4+ does not show any catalytic behavior, which we attribute to the presence of a fragile 𝜇-oxo bridge prone to fragmentation. Instability of the reduced cis,cis-[(bpy)2(H3N)RuIIIORuIII(NH3)(bpy)2]4+ toward loss of the 𝜇-oxo bridge is already established.100 Taking these challenges into account, we focused on the synthesis of dimeric Ru systems bearing more rigid bridging ligands with high stability toward oxidation and reduction. Thus, starting from previously reported 𝜇-chloro dinuclear ruthenium complex [Ru2II(𝜇-Cl)(bpp)(trpy)2]2+,101 bpp: bis(2-pyridyl)-3,5-pyrazolate, we synthesized [Ru2II(NH3)2(bpp)(trpy)2]3+ in which the chloride bridge is displaced by NH3 ligands. In [Ru2II(NH3)2(bpp)(trpy)2]3+, the Ru centers are held in close proximity by the bpp bridging ligand, which in turn orients the ammine ligands toward each other to allow a facile intramolecular coupling. Also, the oxidation state of the Ru centers is +2 which in Meyer’s complex is +3. Scheme 5.2 Structure of [Ru2II(NH3)2(bpp)(trpy)2][PF6]3 (3) 91 5.2. Results and Discussion Heating a solution of 𝜇-chloro dinuclear ruthenium complex [Ru2II(𝜇-Cl)(bpp)(trpy)2]2+ ( 1),101 bpp: bis(2-pyridyl)-3,5-pyrazolate, in methanol saturated with NH3 in pressure tube at 70 °C for 30 min yields mono(ammine) complex [RuIINH3RuIICl(bpp)(trpy)2]2+ (2). Abstraction of chloride by AgNO3 in water with 0.1 M Na3PO4 buffer and excess (NH4)2SO4 yields [Ru2II(NH3)2(bpp)(trpy)2]3+ (3) in 36% isolated yield. The Ru centers are held in close proximity by the bpp bridging ligand, which in turn orients the ammine ligands toward each other to allow a facile intramolecular coupling. CV of 3 in MeCN shows two major anodic peaks at 0.28 and 0.59 V vs. Ag/AgNO3, which can tentatively be assigned to the RuII/RuIII couples of the two Ru centers, and one small peak at 0.76 vs. Ag/AgNO3, which was not observed in 1,2–difluorobenzene (DFB) solvent. Scheme 5.3 Synthesis of Ru dinuclear mono and bis(ammine) complexes 92 30µ 8.0µ 4.0µ 20µ Current (A) Current (A) 0.0 10µ -4.0µ 0 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 Potential (V vs. Ag/AgNO3) Potential (V vs. Ag/AgNO3) Figure 5.1 Left: CV of 3 in MeCN containing 0.1 M TBAPF6 as supporting electrolyte Right: corresponding SQW taken instantly after CV. WE: GC, CE: Pt mesh, RE: Ag/AgNO3, scan rate: 10 mV/s 3.2µ 2.0µ 3.0µ 2.8µ 0.0 Current (A) Current (A) 2.6µ 2.4µ -2.0µ 2.2µ 2.0µ -4.0µ 1.8µ -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 Potential (V vs. Ag/AgNO3) Potential (V vs. Ag/AgNO3) Figure 5.2 Left: CV of 3 in 1,2–difluorobenzene containing 0.1 M TBAPF6 as supporting electrolyte Right: corresponding SQW taken instantly after CV. WE: GC, CE: Pt mesh, RE: Ag/AgNO3, scan rate: 10 mV/s 93 1.2m MeCN (sat. NH3) MeCN (sat. NH3) + 1 mM 2 1.0m 800.0µ Current (A) 600.0µ 400.0µ 200.0µ 0.0 -0.4 0.0 0.4 0.8 1.2 1.6 Potential (V vs. Ag/AgNO3) Figure 5.3 Blue: CV of MeCN saturated with NH3. Orange: CV of MeCN saturated with NH3 after the addition of 1 mM of 3. WE: GC, CE: Pt mesh, RE: Ag/AgNO3, scan rate: 100 mV/s, 0.1 M TBAPF6 as supporting electrolyte Figure 5.3 shows the CV of MeCN saturated with NH3 (blue trace). 1 mM 3 was added right after taking the CV of the blank. A catalytic current was observed at the potential of 0.07 V vs. Ag/AgNO3 and the onset of NH3 oxidation was shifted by 440 mV to less anodic potentials. The onset potential of the catalytic plateau suggests that catalysis starts with the oxidation of the first Ru center. Presumably, upon oxidation of Ru(II) to Ru(III), excess NH3 deprotonates the ammine ligand. One possible mechanism might be the disproportionation of Ru(III) amido to Ru(II) ammine and Ru(IV) imido followed by either intramolecular or intermolecular N–N coupling. 94 Scheme 5.4 Intermolecular vs. intramolecular N–N coupling In the case of intramolecular N–N coupling, 3 can be envisioned as one of the intermediates en route to N2 formation. The possibility of the formation of the hydrogen-bridged dinuclear Ru complex was examined by attempting to synthesize it independently. In fact [Ru2II(𝜇- N2H4)(bpp)(trpy)2]3+ (4) was synthesized by stirring 1 in 98% hydrazine at room temperature, followed by addition of a hydrazine solution to a saturated aqueous solution of NH4PF6 to precipitate the product (Scheme 5.5). 95 The 1H NMR spectrum of 4 shows a singlet resonance for protons of bridged hydrazine at 5.74 ppm in (CD3)2CO and 4.37 ppm in MeCN-d3 Scheme 5.5 Synthesis of dinuclear Ru hydrazine complex Hydrazine-bridged dinuclear Ru complexes have been reported previously. In 1992, Collman and co-workers reported the NH3 oxidation by a bis(ammine) cofacial dinuclear Ru diporphyrin complex.102 Oxidation of this complex yielded intramolecular N–N bond formation and ultimately a bridged dinitrogen complex through bridged hydrazine and diazene intermediates. Complex 3 was chemically oxidized by tris(4-bromophenyl)amine radical cation to yield complex 3b (Scheme 5.6). Relatively sharp peaks in the 1H NMR spectrum of 3b in a wide chemical shift range (–60 to 40 ppm) are indicative of the electronic coupling of two Ru(III) centers through the bpp bridging ligand. This 1H NMR spectrum pattern has also been observed on previously reported cis,cis–[(bpy)2(NH3)RuIIIORuIII(NH3)(bpy)2]4+ complex.103 The addition of NH3 or a hindered base like 2,6–dimethylpyridine to the acetonitrile solution of 3b regenerates complex 3 (Scheme 5.6) along with the disappearance of all paramagnetic peaks. A new set of peaks in the diamagnetic region also appears in the 1H NMR spectrum, which can be 96 15 correlated to the other product(s) of the reaction. More experiments including NH3 isotope labeling can be envisioned for the characterization of the possible intermediates en route to intramolecular N–N coupling of catalytic NH3 oxidation. Scheme 5.6 Chemical oxidation of Ru dimer bis(ammine) 3 Finally, introducing electron-donating groups on the polypyridyl ligands can shift the oxidation potential of the Ru(II) centers to more cathodic potentials, reducing the overpotential of the NH3 oxidation. In addition, the possibility of the AO by heteronuclear bis(ammine) complexes can be examined with this system. For example, displacing one of the Ru centers with Fe may provide versatile information about the effect of Fe on the potential and mechanism of N–N formation in the case of catalysis. 97 5.3. Conclusion For investigating the intramolecular N–N coupling for catalytic AO, a dinuclear Ru bis(ammine) complex (3) was synthesized in which two NH3 ligands are in close proximity. CV of the 3 in MeCN in the absence of NH3 shows two major peaks assigned to the oxidation of Ru(II) centers to Ru(III). A catalytic current was observed when 1 mM of 3 was added to the MeCN solution saturated with NH3. Hydrazine–bridged analogous complex 4 which may be considered as one of the possible intermediates in AO was also prepared and characterized. The bis(ammine) complex 3 was chemically oxidized as well. The addition of NH3 or a hindered base to the oxidized species (3b) regenerated the diamagnetic dinuclear Ru(II) bis(ammine) 3. 98 5.4. Synthesis 5.4.1. General Synthesis Hbpp, and [Ru2II(𝜇-Cl)(bpp)(trpy)2]2+ (1) were synthesized according to previously reported procedures.101 5.4.2. Synthesis of [Ru2II(NH3)Cl(bpp)(trpy)2][PF6]2 (2) Inside a glove bag filled with N2, a pressure tube was charged with 1 (450 mg, 0.370 mmol), 50 mL dry and degassed MeOH saturated with NH3, and a stir bar. After sealing, the pressure tube was taken out and heated for 30 min at 70 °C with stirring. After cooling to room temperature, the pressure tube was opened to air, and N2 was bubbled into the solution to remove the dissolved NH3. MeOH was removed on the rotovap to yield brown-purple solids. The solids were washed with Et2O and dried under vacuum (433 mg, 95%). 1H NMR (500 MHz, acetone-d6) δ 8.68 (dd, J = 8.1, 7.4 Hz, 4H), 8.58 (ddt, J = 14.8, 8.1, 1.1 Hz, 4H), 8.45 (s, 1H), 8.40 (ddd, J = 5.5, 1.6, 0.7 Hz, 2H), 8.29 (ddd, J = 5.4, 1.5, 0.7 Hz, 2H), 8.20 – 8.09 (m, 4H), 7.99 (tdd, J = 7.9, 2.7, 1.5 Hz, 4H), 7.76 (dtd, J = 12.8, 7.7, 1.4 Hz, 2H), 7.43 (ddd, J = 7.6, 5.5, 1.3 Hz, 2H), 7.37 (ddd, J = 7.6, 5.5, 1.3 Hz, 2H), 7.34 (dt, J = 5.7, 1.2 Hz, 1H), 7.26 (dt, J = 5.6, 1.2 Hz, 1H), 6.94 (ddd, J = 7.3, 5.7, 1.4 Hz, 1H), 6.90 (ddd, J = 7.4, 5.8, 1.4 Hz, 1H), 3.06 (s, 3H). 99 5.4.3. Synthesis of [Ru2II(NH3)2(bpp)(trpy)2][PF6]3 (3) Inside a glove bag filled with N2, a pressure tube was charged with 2 (1 equiv, 400 mg, 0.324 mmol), 50 mL degassed 0.1 M Na3PO4, AgNO3 (1 equiv, 55 mg, 0.324 mmol), (NH4)2SO4 (40 equiv, 1.71 g, 12.98 mmol), and a stir bar. After sealing, the pressure tube was taken out, and the suspension was stirred at 100 °C for 1 h. After cooling to room temperature, the pressure tube was opened to air, and the solution was filtered through Celite to remove AgCl. Excess NH4PF6 was added to precipitate the solids to yield a mixture of 3 and [Ru(trpy)2][PF6]2 as by-product, which can be separated based on the insolubility of 3 and solubility of [Ru(trpy)2][PF6]2 in CH2Cl2. 3 can be further purified by crystallization with slow diffusion of Et2O into acetone solution (158 mg, 36%). 1H NMR (500 MHz, acetone-d6) δ 8.75 (d, J = 8.1 Hz, 4H), 8.64 (d, J = 8.1 Hz, 4H), 8.53 (s, 1H), 8.33 (d, J = 5.1 Hz, 4H), 8.26 (t, J = 8.1 Hz, 2H), 8.19 (d, J = 8.0 Hz, 2H), 8.06 (td, J = 7.9, 1.5 Hz, 4H), 7.84 (td, J = 7.8, 1.5 Hz, 2H), 7.48 (ddd, J = 7.4, 5.7, 1.3 Hz, 4H), 7.38 (d, J = 5.7 Hz, 2H), 6.97 (ddd, J = 7.3, 5.7, 1.4 Hz, 2H), 2.84 (s, 6H). 13C NMR (126 MHz, acetone-d6) δ 160.83, 160.54, 156.17, 156.14, 154.33, 151.05, 138.74, 137.59, 135.79, 129.43, 125.21, 124.00, 123.91, 121.09, 105.59. 100 5.4.4. Synthesis of [Ru2II(𝜇-N2H4)(bpp)(trpy)2][PF6]3 (4) 1 (350 mg, 0.288 mmol) was stirred in 15 mL of 98% hydrazine at room temperature for 15 min. The solution was poured into the 40 mL of water saturated with NH4PF6 to precipitate the product as brown solids, which were collected, washed with Et2O, and dried under reduced pressure (293 mg, 75%) 1H NMR (500 MHz, acetone-d6) δ 8.69 (d, J = 8.1 Hz, 4H), 8.59 (s, 1H), 8.58–8.53 (m, 8H), 8.30 (d, J = 8.0 Hz, 2H), 8.21 (t, J = 8.1 Hz, 2H), 8.05 (t, J = 7.9 Hz, 4H), 7.85 (t, J = 7.8 Hz, 2H), 7.64 (t, J = 6.6 Hz, 4H), 7.32 (d, J = 5.7 Hz, 2H), 6.90 (t, J = 6.7 Hz, 2H), 5.74 (s, 4H). 101 Chapter 6. SUMMARY AND FUTURE DIRECTIONS 102 The initial work in ammonia oxidation by molecular systems began with synthesizing Fe ammine coordination complexes, examining their electrochemistry, and evaluating them as electrocatalysts for NH3 oxidation. We began with the preparation of a low-spin Fe tris(ammine) complex, supported by a tripodal phosphine ligand. Cyclic voltammograms (CVs) for anodic scans of NH3-saturated THF solutions when the Fe complex is present and absent were taken. When the Fe complex was present, a catalytic current was observed, and further bulk electrolysis experiments revealed the production of N2 at the anode electrode. Moving toward the more accessible and stable iron complexes, Fe phthalocyanine complexes may be considered good candidates for this purpose. The preliminary results of CVs in THF saturated with NH3 (3.24 M) utilizing a boron-doped diamond (BDD) as the working electrode showed a catalytic onset in the presence of 1 mM of Fe phthalocyanine (Figure 6.1). Fe phthalocyanine 60µ Uncatalyzed NH3 oxidation Fe phthalocyanine + NH3 2.0µ 50µ 40µ Current (A) Current (A) 0.0 30µ 20µ -2.0µ 10µ 0 -0.2 0.0 0.2 0.4 0.6 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 1.2 Potential (V vs. Fc*) Potential (V vs. Fc*) Figure 6.1 Left: CV of 1mM of Fe phthalocyanine in THF in the presence of Fc* as internal standard shows a peak with E1/2 of 0.27 V (vs. Fc*). Right: CV of 1mM of Fe phthalocyanine in THF saturated with NH3 (3.24 M) in the presence of Fc* as internal standard shows a catalytic current with the onset of 0.79 V (vs. Fc*). WE: BDD, CE: Pt disk, 2 M NH4OTf as supporting electrode, scan rate: 100 mV/s 103 Iron(II) phthalocyanine bis(ammine) complexes have already been synthesized and their oxidation chemistry has been reported in the literature.104–108 By introducing an axial ligand the oxidation potential of catalysis can be tuned; also, by anchoring groups the complex can be immobilized on the surface of the electrode for heterogeneous electrochemical NH3 oxidation (Scheme 6.1).109 Furthermore, the synthesis of phthalocyanines through the cyclotetramerization of various phthalic acid derivatives are well established and thus can be utilized to prepare various ligands with different binding affinities and specificities.110 Scheme 6.1 Fe phthalocyanines as candidates for the electrochemical NH3 oxidation In chapter 4, an isoindole-based Ru ammine complex was prepared. Its electrocatalytic activity was evaluated in the presence of NH3 in THF which shifted the anodic current of NH3 oxidation to 0.32 V vs. Fc*. For shifting the potential to more cathodic values, more electron donor substituent groups e.g., NMe2, can be put on both bpy and bid ligands (Scheme 6.2). Also, variant temperature NMR studies with 15N labeled Ru ammine complexes may reveal more details on the mechanism of the catalysis by characterizing the possible intermediates. 104 Scheme 6.2 More electron donor substituent groups on ligands In chapter 5, the synthesis of a dinuclear Ru bis(ammine) complex was described. Non- catalytic intramolecular N–N coupling by an oxo-bridged dinuclear Ru bis(ammine) complex has been reported by Meyer and co-workers.84,85 Unlike the oxo-bridged Ru dimers, the rigid pyrazole- based bridging ligand is not prone to fragmentation.100 The cyclic voltammogram of an acetonitrile solution of this complex shows two quasi-reversible peaks which can be attributed to sequential oxidations of the Ru(II) centers to Ru(III). CVs of an NH3-saturated, 1 mM solution of the complex in acetonitrile exhibit an electrocatalytic current at a potential that is ~300 mV lower than the onset of oxidation of an NH3-saturated acetonitrile solution at the glassy carbon anode. This is the first example of a dinuclear Ru bis(ammine) complex capable of catalytic NH3 oxidation. To make Ru centers more electron efficient and therefore shift the oxidation potential of the catalyst to less anodic potentials, both trpy and Hbpp ligands were functionalized by NMe2 groups. Scheme 6.3 summarizes the multistep synthesis of the NMe2 functionalized Hbpp ligand. In addition to the substitution of pyridyls by NMe2 groups, the 1,3-diketone can also be functionalized by anchoring groups for heterogeneous catalysis before the last step of the reaction. Figure 6.2 compares the 1H NMR of Hbpp and NMe2-substituted Hbpp bridging ligands in DMSO-d6. 105 Scheme 6.3 Synthesis of the bridging ligand with more electron donor groups Figure 6.2 1H NMR of Hbpp (green) and NMe2-substituted Hbpp (red) in DMSO-d6 106 The NMe2-Hbpp bridging ligand was used to synthesize the corresponding Cl-bridged dinuclear Ru complex (Scheme 6.4). The solubility of this complex compared to its unsubstituted counterpart is much lower in most of the organic solvents. Scheme 6.4 Synthesis of the Cl-bridged dinuclear Ru complex supported by NMe2-bpp Figure 6.3 Comparison of 1H NMR of bpp and NMe2-bpp bridged Ru dimers in acetone-d6 107 NMe2-substituted terpyridine tridentate ligand was also synthesized according to Scheme 6.5. Scheme 6.5 Synthesis of NMe2-substituted trpy ligand Figure 6.4 1H NMR of trpy and NMe2-substituted trpy in CDCl3 108 Synthesis of the dinuclear Ru bis(ammine) complexes with bid ligand can also be envisioned. The two bid ligands will reduce the overall charge of the complex by 2 and thus can lower the oxidation potential of the complex to much lower potentials (Scheme 6.6). Scheme 6.6 Dinuclear Ru bis(ammine) complexes supported by bid ligands Last but not least, a wide range of heterodinuclear bis(ammine) complexes can be prepared via these systems to investigate the synergistic effects of various metals on catalytic NH3 oxidation. 109 REFERENCES (1) Diffenbaugh, N. S.; Field, C. B. Changes in Ecologically Critical Terrestrial Climate Conditions. Science 2013, 341, 486–492. (2) BP Statistical Review of World Energy_70th Edition https://www.bp.com/content/dam/bp/business-sites/en/global/corporate/pdfs/energy- economics/statistical-review/bp-stats-review-2021-full-report.pdf (accessed May 25, 2022). (3) Carbon Dioxide Information Analysis Center (CDIAC) Database https://cdiac.ess- dive.lbl.gov/trends/emis/meth_reg.html (accessed May 25, 2022). (4) Fawcett, A. A.; Iyer, G. C.; Clarke, L. E.; Edmonds, J. A.; Hultman, N. E.; McJeon, H. C.; Rogelj, J.; Schuler, R.; Alsalam, J.; Asrar, G. R.; Creason, J.; Jeong, M.; McFarland, J.; Mundra, A.; Shi, W. CLIMATE POLICY. Can Paris Pledges Avert Severe Climate Change? Science 2015, 350, 1168–1169. (5) Jacobson, M. Z.; Colella, W. G.; Golden, D. M. Cleaning the Air and Improving Health with Hydrogen Fuel-Cell Vehicles. Science 2005, 308, 1901–1905. (6) Turner, J. A. Sustainable Hydrogen Production. Science 2004, 305, 972–974. (7) Sherif, S. A.; Barbir, F.; Veziroglu, T. N. Principles of Hydrogen Energy Production, Storage and Utilization. Journal of Scientific & Industrial Research 2003, 62, 46–63. (8) Satyapal, S.; Petrovic, J.; Read, C.; Thomas, G.; Ordaz, G. The U.S. Department of Energy’s National Hydrogen Storage Project: Progress towards Meeting Hydrogen-Powered Vehicle Requirements. Catal. Today 2007, 120, 246–256. (9) Wang, L.; Xia, M.; Wang, H.; Huang, K.; Qian, C.; Maravelias, C. T.; Ozin, G. A. Greening Ammonia toward the Solar Ammonia Refinery. Joule 2018, 2, 1055–1074. (10) Li, J.; Lai, S.; Chen, D.; Wu, R.; Kobayashi, N.; Deng, L.; Huang, H. A Review on Combustion Characteristics of Ammonia as a Carbon-Free Fuel. Frontiers in Energy Research 2021, 9. (11) Boretti, A.; Castelletto, S. NH3 Prospects in Combustion Engines and Fuel Cells for Commercial Aviation by 2030. ACS Energy Lett. 2022, 7, 2557–2564. (12) Kobayashi, H.; Hayakawa, A.; Somarathne, K. D. K. A.; Okafor, E. C. Science and Technology of Ammonia Combustion. Proc. Combust. Inst. 2019, 37, 109–133. (13) Woo, Y.; Jang, J. Y.; Lee, Y. J.; Kim, J.-N. Recent Progress on the Ammonia-Gasoline and the Ammonia-Diesel Dual Fueled Internal Combustion Engines in Korea. 11th NH3 FUEL CONFERENCE 2014. (14) Lan, R.; Tao, S. Ammonia as a Suitable Fuel for Fuel Cells. Front. Energy Res. 2014, 2, 1– 4. 110 (15) Afif, A.; Radenahmad, N.; Cheok, Q.; Shams, S.; Kim, J. H.; Azad, A. K. Ammonia-Fed Fuel Cells: A Comprehensive Review. Renewable Sustainable Energy Rev. 2016, 60, 822– 835. (16) Boggs, B. K.; Botte, G. G. On-Board Hydrogen Storage and Production: An Application of Ammonia Electrolysis. J. Power Sources 2009, 192, 573–581. (17) Li, G.; Kanezashi, M.; Tsuru, T. Catalytic Ammonia Decomposition over High- Performance Ru/Graphene Nanocomposites for Efficient COx-Free Hydrogen Production. Catalysts 2017, 7, 23. (18) Li, L.; Wang, Y.; Xu, Z. P.; Zhu, Z. Catalytic Ammonia Decomposition for CO-Free Hydrogen Generation over Ru/Cr2O3 Catalysts. Appl. Catal. A 2013, 467, 246–252. (19) Chung, D. B.; Kim, H. Y.; Jeon, M.; Lee, D. H.; Park, H. S.; Choi, S. H.; Nam, S. W.; Jang, S. C.; Park, J.-H.; Lee, K.-Y.; Yoon, C. W. Enhanced Ammonia Dehydrogenation over Ru/La(x)–Al2O3 (x = 0–50 Mol%): Structural and Electronic Effects of La Doping. Int. J. Hydrogen Energy 2017, 42, 1639–1647. (20) David, W. I. F.; Makepeace, J. W.; Callear, S. K.; Hunter, H. M. A.; Taylor, J. D.; Wood, T. J.; Jones, M. O. Hydrogen Production from Ammonia Using Sodium Amide. J. Am. Chem. Soc. 2014, 136, 13082–13085. (21) Vitse, F.; Cooper, M.; Botte, G. G. On the Use of Ammonia Electrolysis for Hydrogen Production. J. Power Sources 2005, 142, 18–26. (22) Palaniappan, R.; Ingram, D. C.; Botte, G. G. Hydrogen Evolution Reaction Kinetics on Electrodeposited Pt–M (M = Ir, Ru, Rh, and Ni) Cathodes for Ammonia Electrolysis. J. Electrochem. Soc. 2014, 161, E12–E22. (23) Bunce, N. J.; Bejan, D. Mechanism of Electrochemical Oxidation of Ammonia. Electrochim. Acta 2011, 56, 8085–8093. (24) Zhong, C.; Hu, W. B.; Cheng, Y. F. Recent Advances in Electrocatalysts for Electro- Oxidation of Ammonia. J. Mater. Chem. A Mater. Energy Sustain. 2013, 1, 3216–3238. (25) Nagakura, S.; Iguchi, H.; Ezawa, H.; Iwamura, H.; Sato, F.; Kubo, R. Physics and Chemistry Dictionary. Iwanami Shoten 1988, 1234. (26) Hanada, N.; Hino, S.; Ichikawa, T.; Suzuki, H.; Takai, K.; Kojima, Y. Hydrogen Generation by Electrolysis of Liquid Ammonia. Chem. Commun. 2010, 46, 7775–7777. (27) Little, D. J.; Smith, M. R., III; Hamann, T. W. Electrolysis of Liquid Ammonia for Hydrogen Generation. Energy Environ. Sci. 2015, 8, 2775–2781. (28) Handbook of Reference Electrodes; Inzelt, G., Lewenstam, A., Scholz, F., Eds.; Springer, Berlin, Heidelberg, 2013. 111 (29) Bashkin, J. K.; Kinlen, P. J. Oxygen-Stable Ferrocene Reference Electrodes. Inorg. Chem. 1990, 29, 4507–4509. (30) Su, B.; Hatay, I.; Ge, P. Y.; Mendez, M.; Corminboeuf, C.; Samec, Z.; Ersoz, M.; Girault, H. H. Oxygen and Proton Reduction by Decamethylferrocene in Non-Aqueous Acidic Media. Chem. Commun. 2010, 46, 2918–2919. (31) Noviandri, I.; Brown, K. N.; Fleming, D. S.; Gulyas, P. T.; Lay, P. A.; Masters, A. F.; Phillips, L. The Decamethylferrocenium/Decamethylferrocene Redox Couple: A Superior Redox Standard to the Ferrocenium/Ferrocene Redox Couple for Studying Solvent Effects on the Thermodynamics of Electron Transfer. J. Phys. Chem. B 1999, 103, 6713–6722. (32) Connelly, N. G.; Geiger, W. E. Chemical Redox Agents for Organometallic Chemistry. Chem. Rev. 1996, 96, 877–910. (33) Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Fundamentals and Applications, 2nd Edition; Wiley Global Education, 2000. (34) Pavlishchuk, V. V.; Addison, A. W. Conversion Constants for Redox Potentials Measured versus Different Reference Electrodes in Acetonitrile Solutions at 25°C. Inorganica Chim. Acta 2000, 298, 97–102. (35) Sawyer, D. T.; Sobkowiak, A.; Roberts, J. L. Electrochemistry for Chemists, 2nd Edition; John Wiley & Sons, Inc., 1995. (36) Jahn, U.; Aussieker, S. Triarylaminium Salt Induced Oxidative Cyclizations of Tertiary Amines. Convenient Access to 2-Substituted Pyrrolidinium Salts. Org. Lett. 1999, 1, 849– 852. (37) Barham, J. P.; John, M. P.; Murphy, J. A. Contra-Thermodynamic Hydrogen Atom Abstraction in the Selective C–H Functionalization of Trialkylamine N-CH3 Groups. J. Am. Chem. Soc. 2016, 138, 15482–15487. (38) M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, G. A. Petersson, H. Nakatsuji, X. Li, M. Caricato, A. V. Marenich, J. Bloino, B. G. Janesko, R. Gomperts, B. Mennucci, H. P. Hratchian, J. V. Ortiz, A. F. Izmaylov, J. L. Sonnenberg, D. Williams-Young, F. Ding, F. Lipparini, F. Egidi, J. Goings, B. Peng, A. Petrone, T. Henderson, D. Ranasinghe, V. G. Zakrzewski, J. Gao, N. Rega, G. Zheng, W. Liang, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, K. Throssell, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. J. Bearpark, J. J. Heyd, E. N. Brothers, K. N. Kudin, V. N. Staroverov, T. A. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. P. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, J. M. Millam, M. Klene, C. Adamo, R. Cammi, J. W. Ochterski, R. L. Martin, K. Morokuma, O. Farkas, J. B. Foresman, and D. J. Fox. Gaussian 16, Revision A.03; Gaussian, Inc., Wallingford CT, 2016. 112 (39) E. D. Glendening, J. K. Badenhoop, A. E. Reed, J. E. Carpenter, J. A. Bohmann, C. M. Morales, P. Karafiloglou, C. R. Landis, and F. Weinhold. NBO 7.0; University of Wisconsin, Madison, WI, 2018. (40) Becke, A. D. Density‐functional Thermochemistry. III. The Role of Exact Exchange. J. Chem. Phys. 1993, 98, 5648–5652. (41) Weigend, F.; Ahlrichs, R. Balanced Basis Sets of Split Valence, Triple Zeta Valence and Quadruple Zeta Valence Quality for H to Rn: Design and Assessment of Accuracy. Phys. Chem. Chem. Phys. 2005, 7, 3297–3305. (42) Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. Universal Solvation Model Based on Solute Electron Density and on a Continuum Model of the Solvent Defined by the Bulk Dielectric Constant and Atomic Surface Tensions. J. Phys. Chem. B 2009, 113, 6378–6396. (43) Dennington, R.; Keith, T. A.; Millam, J. M. GaussView, Version 6; Semichem Inc. Shawnee Mission KS, 2019. (44) Zott, M. D.; Garrido-Barros, P.; Peters, J. C. Electrocatalytic Ammonia Oxidation Mediated by a Polypyridyl Iron Catalyst. ACS Catal. 2019, 9, 10101–10108. (45) Zott, M. D.; Peters, J. C. Enhanced Ammonia Oxidation Catalysis by a Low-Spin Iron Complex Featuring Cis Coordination Sites. J. Am. Chem. Soc. 2021, 143, 7612–7616. (46) Li, Y.; Chen, J.-Y.; Miao, Q.; Yu, X.; Feng, L.; Liao, R.-Z.; Ye, S.; Tung, C.-H.; Wang, W. A Parent Iron Amido Complex in Catalysis of Ammonia Oxidation. J. Am. Chem. Soc. 2022, 144, 4365–4375. (47) Neumeyer, F.; Lipschutz, M. I.; Tilley, T. D. Group 8 Transition Metal Complexes of the Tripodal Triphosphino Ligands PhSi(CH2PR2)3 (R = Ph, IPr). Eur. J. Inorg. Chem. 2013, 2013, 6075–6078. (48) Aloisi, A.; Berthet, J.-C.; Genre, C.; Thuéry, P.; Cantat, T. Complexes of the Tripodal Phosphine Ligands PhSi(XPPh2)3 (X = CH2, O): Synthesis, Structure and Catalytic Activity in the Hydroboration of CO2. Dalton Trans. 2016, 45, 14774–14788. (49) Sellmann, D.; Kleinschmidt, E. Einfache Synthese von [C5H5Fe(CO)dppe]+ und photochemische Überführung in den Distickstoff-Komplex [C5H5Fe(dppe)2N2]2+. Angew. Chem. Weinheim Bergstr. Ger. 1975, 87, 595–596. (50) McGhee, L.; Siddique, R. M.; Winfield, J. M. Reactions of Hexakis(Acetonitrile)Iron(II) Hexafluorophosphate in Acetonitrile. Ligation of Ammonia, Pyridine, and Trimethylamine to Iron(II) and the Behaviour of Trimethyl Phosphite and Trimethylphosphine towards Iron(II), Nitrogen Donor Ligand Complex Cations. J. Chem. Soc. Dalton Trans. 1988, No. 5, 1309–1314. 113 (51) Umehara, K.; Kuwata, S.; Ikariya, T. N–N Bond Cleavage of Hydrazines with a Multiproton-Responsive Pincer-Type Iron Complex. J. Am. Chem. Soc. 2013, 135, 6754– 6757. (52) Anderson, J. S.; Moret, M.-E.; Peters, J. C. Conversion of Fe–NH2 to Fe–N2 with Release of NH3. J. Am. Chem. Soc. 2013, 135, 534–537. (53) Chalkley, M. J.; Drover, M. W.; Peters, J. C. Catalytic N2-to-NH3 (or -N2H4) Conversion by Well-Defined Molecular Coordination Complexes. Chem. Rev. 2020, 120, 5582–5636. (54) Gu, N. X.; Oyala, P. H.; Peters, J. C. Hydrazine Formation via Coupling of a Nickel(III)– NH 2 Radical. Angew. Chem. Weinheim Bergstr. Ger. 2021, 133, 4055–4059. (55) Thoreson, K. A.; Follett, A. D.; McNeill, K. Synthesis and Characterization of Pentaphosphino Zero-Valent Iron Complexes and Their Corresponding Iron(II)-Chloride and -Hydride Complexes. Inorg. Chem. 2010, 49, 3942–3949. (56) Najafian, A.; Cundari, T. R. Computational Mechanistic Study of Electro-Oxidation of Ammonia to N2 by Homogenous Ruthenium and Iron Complexes. J. Phys. Chem. A 2019, 123, 7973–7982. (57) Zachwieja, U.; Jacobs, H. Tri-µ-Amido-Bis-[Triamminchrom(III)]-Iodid, [Cr2(NH2)3(NH3)6]I3, Ein Neuer „hexagonaler Perowskit“. Zeitschrift für Kristallographie - Crystalline Materials 1993, 206, 247–254. (58) Frank, W.; Hoffmann, K.; Heck, L. The Tri-μ-Amidobis[Triamminecobalt(III)] Cation: The First Triply Amido-Bridged Complex. Angew. Chem. Int. Ed Engl. 1990, 29, 1158–1159. (59) Andersen, P.; Glerup, J.; Gumm, A.; Hansen, S. K.; Magnussen, M. Dinuclear 1,4,7- Triazacyclononane (Tacn) Complexes of Cobalt(Iii) with Amido and Tacn Bridges. Synthesis, Characterization and Reversible Acid-Accelerated Bridge Cleavage. Dalton Transactions 2004, No. 18, 2929. (60) Dell, W.; Ziegler, M. L. Darstellung Und Charakterisierung von Dreifach NR2-Verbrückten Molybdänkomplexen Des Typs η3-C7H7(CO)2Mo(μ–NR2)3Mo-η7-C7H7 / Synthesis and Characterization of Triply NR2-Bridged Molybdenum Complexes of the Type η3- C7H7(CO)2Mo(μ-NR2)3Mo-η7-C7H7. Zeitschrift für Naturforschung B 1982, 37, 7–12. (61) Frank, W.; Heck, L.; Müller-Becker, S.; Raber, T. The Tri-μ-Amido- Bis[Triammineplatinum(IV)] Cation; Preparation, Structure and Acidity. Inorganica Chimica Acta 1997, 265, 17–22. (62) Noh, S. K.; Sendlinger, S. C.; Janiak, C.; Theopold, K. H. [Cp*(Me)Cr(μ-Me)]2, an Electron-Deficient Chromium(III) Alkyl with Bridging Methyl Groups and a Chromium- Chromium Bond. Journal of the American Chemical Society 1989, 111, 9127–9129. (63) Ganesan, M.; Gabbaï, F. P. [Cp*Cr(C6F5)(Me)(Py)] as a Living Chromium(III) Catalyst for the “Aufbaureaktion.” Organometallics 2004, 23, 4608–4613. 114 (64) Saouma, C. T.; Müller, P.; Peters, J. C. Characterization of Structurally Unusual Diiron N(x)H(y) Complexes. J. Am. Chem. Soc. 2009, 131, 10358–10359. (65) Kashiwabara, K.; Ozeki, Y.; Kita, M.; Fujita, J.; Nakajima, K. Synthesis and Crystal Structure of Bis(1,1,1-Tris(Dimethylphosphinomethyl)Ethane)Iron(II) Tetrafluoroborate Dihydrate. Bull. Chem. Soc. Jpn. 1995, 68, 3453–3457. (66) Aldrich, K. E.; Fales, B. S.; Singh, A. K.; Staples, R. J.; Levine, B. G.; McCracken, J.; Smith, M. R., III; Odom, A. L. Electronic and Structural Comparisons between Iron (II/III) and Ruthenium (II/III) Imide Analogs. Inorg. Chem. 2019, 58, 11699–11715. (67) Chaquin, P.; Canac, Y.; Lepetit, C.; Zargarian, D.; Chauvin, R. Estimating Local Bonding/Antibonding Character of Canonical Molecular Orbitals from Their Energy Derivatives. The Case of Coordinating Lone Pair Orbitals. Int. J. Quantum Chem. 2016, 116, 1285–1295. (68) Bader, R. F. W. A Quantum Theory of Molecular Structure and Its Applications. Chemical Reviews. 1991, pp 893–928. (69) Cortesguzman, F.; Bader, R. Complementarity of QTAIM and MO Theory in the Study of Bonding in Donor?Acceptor Complexes. Coord. Chem. Rev. 2005, 249, 633–662. (70) Wolczanski, P. T. Flipping the Oxidation State Formalism: Charge Distribution in Organometallic Complexes As Reported by Carbon Monoxide. Organometallics 2017, 36, 622–631. (71) Pinter, B.; Van Speybroeck, V.; Waroquier, M.; Geerlings, P.; De Proft, F. Trans Effect and Trans Influence: Importance of Metal Mediated Ligand-Ligand Repulsion. Phys. Chem. Chem. Phys. 2013, 15, 17354–17365. (72) Mukhopadhyay, T. K.; Feller, R. K.; Rein, F. N.; Henson, N. J.; Smythe, N. C.; Trovitch, R. J.; Gordon, J. C. Investigation of Formally Zerovalent Triphos Iron Complexes. Chem. Commun. 2012, 48, 8670–8672. (73) Britovsek, G. J. P.; England, J.; White, A. J. P. Iron(II), Manganese(II) and Cobalt(II) Complexes Containing Tetradentate Biphenyl-Bridged Ligands and Their Application in Alkane Oxidation Catalysis. Dalton Trans. 2006, No. 11, 1399–1408. (74) Britovsek, G. J. P.; England, J.; White, A. J. P. Non-Heme Iron(II) Complexes Containing Tripodal Tetradentate Nitrogen Ligands and Their Application in Alkane Oxidation Catalysis. Inorg. Chem. 2005, 44, 8125–8134. (75) Sellmann, D.; Weber, W. Reaktionen an Komplexgebundenen Liganden: XXXVIII. [CpFe(Dppe)L]- Und [CpFe(Cdpe)L]-Komplexe Mit L = CO-, N2-, NH3-, Sowie N3−- Liganden (Dppe = 1,2-Bis(Diphenylphosphino)Ethan, Cdpe = Cis-1,2- Bis(Diphenylphosphino)Ethen). J. Organomet. Chem. 1986, 304, 195–205. 115 (76) Petuker, A.; Merz, K.; Merten, C.; Apfel, U.-P. Controlled Flexible Coordination in Tripodal Iron(II) Phosphane Complexes: Effects on Reactivity. Inorg. Chem. 2016, 55, 1183–1191. (77) Albertin, G.; Antoniutti, S.; Bordignon, E.; Pattaro, S. Synthesis, Characterisation and Reactivity of Hydrazine Complexes of Iron(II). J. Chem. Soc. Dalton Trans. 1997, No. 22, 4445–4454. (78) Gardner, T. G. Low-Valent Early Transition Metal Complexes Stabilized by the Chelating Tripod Phosphine Ligand (t-Butyl)Tris[(Dimethylphosphino)Methyl]Silane (“Trimpsi”). Ph.D. Thesis. University of Illinois, Urbana-Champaign, 1989. (79) Karsch, H. H.; Appelt, A. Funktionelle Trimethylphosphanderivate, XVIII [1]. Methyl(Phosphinomethyl)Silane Und -Stannane / Functional Trimethylphosphane Derivatives, XVIII [1]. Methyl(Phosphinomethyl)Silanes and -Stannanes. Zeitschrift für Naturforschung B 1983, 38B, 1399–1405. (80) Boncella, J. M.; Green, M. L. H. The Chemistry of Iron- Tris(Dimethylphosphinomethyl)Methylsilane Complexes: Crystal Structure of the Compound [Fe((Me2PCH2)3SiMe)2(Mu-H)2(Mu-C=CH2)] with a Vinylidene Group Which Is Derived from Ethylene. J. Organomet. Chem. 1987, 325, 217–231. (81) Nakajima, K.; Toda, H.; Sakata, K.; Nishibayashi, Y. Ruthenium-Catalysed Oxidative Conversion of Ammonia into Dinitrogen. Nat. Chem. 2019, 11, 702–709. (82) Habibzadeh, F.; Miller, S. L.; Hamann, T. W.; Smith, M. R., III. Homogeneous Electrocatalytic Oxidation of Ammonia to N2 under Mild Conditions. Proc. Natl. Acad. Sci. U. S. A. 2019, 116, 2849–2853. (83) Coia, G. M.; Demadis, K. D.; Meyer, T. J. Oxidation of Ammonia in Osmium Polypyridyl Complexes. Inorg. Chem. 2000, 39, 2212–2223. (84) Ishitani, O.; White, P. S.; Meyer, T. J. Formation of Dinitrogen by Oxidation of [(Bpy)2(NH3)RuORu(NH3)(Bpy)2]4+. Inorg. Chem. 1996, 35, 2167–2168. (85) Ishitani, O.; Ando, E.; Meyer, T. J. Dinitrogen Formation by Oxidative Intramolecular N--- N Coupling in Cis,Cis-[(Bpy)2(NH3 )RuORu(NH3)(Bpy)2]4+. Inorg. Chem. 2003, 42, 1707– 1710. (86) Thompson, M. S.; Meyer, T. J. Oxidation of Coordinated Ammonia to Nitrate. J. Am. Chem. Soc. 1981, 103, 5577–5579. (87) Coia, G. M.; White, P. S.; Meyer, T. J.; Wink, D. A.; Keefer, L. K.; Davis, W. M. Preparation of Osmium Hydrazido Complexes by Interception of an Osmium(IV) Imido Intermediate. J. Am. Chem. Soc. 1994, 116, 3649–3650. 116 (88) Walstrom, A.; Pink, M.; Fan, H.; Tomaszewski, J.; Caulton, K. G. Radical (NO) and Nonradical (N2O) Reagents Convert a Ruthenium(IV) Nitride to the Same Nitrosyl Complex. Inorganic Chemistry. 2007, pp 7704–7706. (89) Man, W.-L.; Tang, T.-M.; Wong, T.-W.; Lau, T.-C.; Peng, S.-M.; Wong, W.-T. Highly Electrophilic (Salen)Ruthenium(VI) Nitrido Complexes. Journal of the American Chemical Society. 2004, pp 478–479. (90) Man, W.-L.; Xie, J.; Pan, Y.; Lam, W. W. Y.; Kwong, H.-K.; Ip, K.-W.; Yiu, S.-M.; Lau, K.-C.; Lau, T.-C. C–N Bond Cleavage of Anilines by a (Salen)Ruthenium(VI) Nitrido Complex. J. Am. Chem. Soc. 2013, 135, 5533–5536. (91) Keener, M.; Peterson, M.; Hernández Sánchez, R.; Oswald, V. F.; Wu, G.; Menard, G. Towards Catalytic Ammonia Oxidation to Dinitrogen: A Synthetic Cycle Using a Simple Manganese Complex. Chemistry-A European Journal 2017. (92) Clarke, R. M.; Storr, T. Tuning Electronic Structure To Control Manganese Nitride Activation. J. Am. Chem. Soc. 2016, 138, 15299–15302. (93) Siegl, W. O. Metal Ion Activation of Nitriles. Syntheses of 1,3-Bis(Arylimino)Isoindolines. J. Org. Chem. 1977, 42, 1872–1878. (94) Siegl, W. O. Metal-Chelating 1, 3-Bis (2′-Pyridylimino) Isoindolines. J. Heterocycl. Chem. 1981, 18, 1613–1618. (95) Gagne, R. R.; Marritt, W. A.; Marks, D. N.; Siegl, W. O. Mononuclear and Binuclear Metal Complexes of 1,3-Bis(2-Pyridylimino)Isoindolines. Inorg. Chem. 1981, 20, 3260–3267. (96) Marks, D. N.; Siegl, W. O.; Gagne, R. R. Synthesis and Characterization of Mononuclear and Binuclear Ruthenium Complexes of 1,3-Bis(2-Pyridylimino)Isoindolines. Inorg. Chem. 1982, 21, 3140–3147. (97) Planas, N.; Ono, T.; Vaquer, L.; Miró, P.; Benet-Buchholz, J.; Gagliardi, L.; Cramer, C. J.; Llobet, A. Carbon Dioxide Reduction by Mononuclear Ruthenium Polypyridyl Complexes. Phys. Chem. Chem. Phys. 2011, 13, 19480. (98) Savéant, J.-M. Elements of Molecular and Biomolecular Electrochemistry: An Electrochemical Approach to Electron Transfer Chemistry; John Wiley & Sons, 2006. (99) Lee, K. J.; Elgrishi, N.; Kandemir, B.; Dempsey, J. L. Electrochemical and Spectroscopic Methods for Evaluating Molecular Electrocatalysts. Nature Reviews Chemistry 2017, 1, 0039. (100) Jurss, J. W.; Concepcion, J. J.; Butler, J. M.; Omberg, K. M.; Baraldo, L. M.; Thompson, D. G.; Lebeau, E. L.; Hornstein, B.; Schoonover, J. R.; Jude, H.; Thompson, J. D.; Dattelbaum, D. M.; Rocha, R. C.; Templeton, J. L.; Meyer, T. J. Electronic Structure of the Water Oxidation Catalyst Cis,Cis-[(Bpy)2(H2O)Ru(III)ORu(III)(OH2)(Bpy)2]4+, the Blue Dimer. Inorg. Chem. 2012, 51, 1345–1358. 117 (101) Sens, C.; Romero, I.; Rodríguez, M.; Llobet, A.; Parella, T.; Benet-Buchholz, J. A New Ru Complex Capable of Catalytically Oxidizing Water to Molecular Dioxygen. J. Am. Chem. Soc. 2004, 126, 7798–7799. (102) Collman, J. P.; Hutchison, J. E.; Ennis, M. S.; Lopez, M. A.; Guilard, R. Reduced Nitrogen Hydride Complexes of a Cofacial Metallodiporphyrin and Their Oxidative Interconversion. An Analysis of Ammonia Oxidation and Prospects for a Dinitrogen Electroreduction Catalyst Based on Cofacial Metallodiporphyrins. J. Am. Chem. Soc. 1992, 114, 8074–8080. (103) Arakawa, R.; Kubota, N.; Fukuo, T.; Ishitani, O.; Ando, E. Study of the Ligand Substitution Reactions of Cis,Cis-[(Bpy)2(L)RuORu(L')(Bpy)2]N+ (L, L' = H2O, OH-, NH3) Using Electrospray Ionization Mass Spectrometry and 1H NMR. Inorg. Chem. 2002, 41, 3749–3754. (104) Ough, E. A.; Stillman, M. J. Analysis of the Absorption and Magnetic Circular Dichroism Spectra of Iron(II) Phthalocyanine. Inorg. Chem. 1994, 33, 573–583. (105) Oña-Burgos, P.; Casimiro, M.; Fernández, I.; Navarro, A. V.; Fernández Sánchez, J. F.; Carretero, A. S.; Gutiérrez, A. F. Octahedral Iron(II) Phthalocyanine Complexes: Multinuclear NMR and Relevance as NO(2) Chemical Sensors. Dalton Trans. 2010, 39, 6231–6238. (106) Huang, J.-S.; Yu, G.-A.; Xie, J.; Wong, K.-M.; Zhu, N.; Che, C.-M. Primary and Secondary Phosphine Complexes of Iron Porphyrins and Ruthenium Phthalocyanine: Synthesis, Structure, and P-H Bond Functionalization. Inorg. Chem. 2008, 47, 9166–9181. (107) Fernández, I.; Pregosin, P. S.; Albinati, A.; Rizzato, S.; Spichiger-Keller, U. E.; Nezel, T.; Fernández-Sánchez, J. F. Solution NMR and X-Ray Structural Studies on Phthalocyaninatoiron Complexes. Helv. Chim. Acta 2006, 89, 1485–1496. (108) Damos, F. S.; de Cássia Silva Luz, R.; Tanaka, A. A. Electroanalysis of Hydrazine and Related Compounds by Oxidation Promoted with MN4 Macrocyclics. In Electrochemistry of N4 Macrocyclic Metal Complexes: Volume 2: Biomimesis, Electroanalysis and Electrosynthesis of MN4 Metal Complexes; Zagal, J. H., Bedioui, F., Eds.; Springer International Publishing: Cham, 2016; pp 201–223. (109) Beiler, A. M.; Denisiuk, A.; Holub, J.; Sánchez-Baygual, F.-J.; Gil-Sepulcre, M.; Ertem, M. Z.; Moonshiram, D.; Piccioni, A.; Llobet, A. Heterogeneous Electrochemical Ammonia Oxidation with a Ru-Bda Oligomer Anchored on Graphitic Electrodes via CH−π Interactions. ACS Energy Lett. 2022, 172–178. (110) Sorokin, A. B. Phthalocyanine Metal Complexes in Catalysis. Chem. Rev. 2013, 113, 8152–8191. 118 SUPPORTING INFORMATION (A) FOR CHAPTER 3 SA.1 NMR Spectra Figure SA.1 31P{1H} NMR of tSiP3 in C6D6 119 Figure SA.2 1H NMR of tSiP3 in C6D6 120 Figure SA.3 31P{1H} NMR of [(tSiP3Fe)2(μ–Cl)3][Cl] (1) in MeCN-d3. The peak at 𝛿 34 ppm is assigned to displacement of bridged Cl with MeCN-d3 to yield [tSiP3Fe(NCMe)3][Cl]2 121 Figure SA.4 1H NMR of [(tSiP3Fe)2(μ–Cl)3][Cl] (1) in MeCN-d3 122 Figure SA.5 1H NMR of [tSiP3Fe(NCMe)3][OTf]2 (2) in MeCN-d3 123 Figure SA.6 13C NMR of [tSiP3Fe(NCMe)3][OTf]2 (2) in MeCN-d3 124 Figure SA.7 31P{1H} NMR of [tSiP3Fe(NCMe)3][OTf]2 (2) in MeCN-d3 125 Figure SA.8 19F NMR of [tSiP3Fe(NCMe)3][OTf]2 (2) in MeCN-d3 126 Figure SA.9 1H NMR of [tSiP3Fe(NCMe)3][OTf]2 (2) in MeNO2-d3 127 Figure SA.10 31P{1H} NMR of [tSiP3Fe(NCMe)3][OTf]2 (2) in MeNO2-d3 128 Figure SA.11 31P{1H} NMR of [tSiP3Fe(NCMe)3][OTf]2 (2) in DMSO-d6 129 Figure SA.12 1H NMR of [tSiP3Fe(NH3)3][OTf]2 (3) in DMSO-d6 130 Figure SA.13 13C NMR of [tSiP3Fe(NH3)3][OTf]2 (3) in DMSO-d6 131 Figure SA.14 31P{1H} NMR of [tSiP3Fe(NH3)3][OTf]2 (3) in DMSO-d6 132 Figure SA.15 19F NMR of [tSiP3Fe(NH3)3][OTf]2 (3) in DMSO-d6 133 Figure SA.16 1H NMR of [tSiP3Fe(NH3)3][OTf]2 (3) in MeNO2-d3 134 Figure SA.17 31P{1H} NMR of [tSiP3Fe(NH3)3][OTf]2 (3) in MeNO2-d3 135 Figure SA.18 1H NMR of [(tSiP3)2Fe2(μ–NH2)3][OTf] (4) in THF-d6 136 Figure SA.19 31P{1H} NMR of [(tSiP3)2Fe2(μ–NH2)3][OTf] (4) in THF-d6 137 Figure SA.20 13C NMR of [(tSiP3)2Fe2(μ–NH2)3][OTf] (4) in THF-d6 138 Figure SA.21 1H NMR of [(tSiP3)2Fe2(μ–NH2)3][OTf] (4) in DMSO-d6 139 Figure SA.22 31P{1H} NMR of [(tSiP3)2Fe2(μ–NH2)3][OTf] (4) in DMSO-d6 140 Figure SA.23 19F NMR of [(tSiP3)2Fe2(μ–NH2)3][OTf] (4) in DMSO-d6 141 Figure SA.24 31P{1H} NMR of [tSiP3Fe(dmpe)(Cl)][Cl] (8) in MeCN-d3 142 Figure SA.25 1H NMR of [tSiP3Fe(dmpe)(Cl)][Cl] (8) in MeCN-d3 143 Figure SA.26 1H NMR of [tSiP3Fe(dmpe)(NCMe)][OTf]2 (6) in MeCN-d3 144 Figure SA.27 13C NMR of [tSiP3Fe(dmpe)(NCMe)][OTf]2 (6) in MeCN-d3 145 Figure SA.28 31P{1H} NMR of [tSiP3Fe(dmpe)(NCMe)][OTf]2 (6) in MeCN-d3 146 Figure SA.29 19F NMR of [tSiP3Fe(dmpe)(NCMe)][OTf]2 (6) in MeCN-d3 147 Figure SA.30 31P{1H} NMR of [tSiP3Fe(dmpe)(NCMe)][OTf]2 (6) in DMSO-d6 148 25 °C –30 °C Figure SA.31 1H NMR of [tSiP3Fe(dmpe)(OTf)][OTf] (9) in THF-d6 at 25 °C (top) and –30 °C (bottom) 149 Figure SA.32 31P{1H} NMR of [tSiP3Fe(dmpe)(OTf)][OTf] (9) in THF-d6 150 Figure SA.33 19F NMR of [tSiP3Fe(dmpe)(OTf)][OTf] (9) in THF-d6 151 Figure SA.34 31P{1H} NMR of [tSiP3Fe(dmpe)(DMSO)][OTf]2 (11) in DMSO-d6 152 Figure SA.35 19F NMR of [tSiP3Fe(dmpe)(DMSO)][OTf]2 (11) in DMSO-d6 153 Figure SA.36 1H NMR of [tSiP3Fe(dmpe)(NH3)][OTf]2 (7) in DMSO-d6 154 Figure SA.37 13C NMR of [tSiP3Fe(dmpe)(NH3)][OTf]2 (7) in DMSO-d6 155 Figure SA.38 31P{1H} NMR of [tSiP3Fe(dmpe)(NH3)][OTf]2 (7) in DMSO-d6 156 Figure SA.39 1H NMR of [tSiP3Fe(N2H4)(MeC(NH)NH2NH)][OTf]2 (10) in DMSO-d6 157 Figure SA.40 31P{1H} NMR of [tSiP3Fe(N2H4)(MeC(NH)NH2NH)][OTf]2 (10) in DMSO-d6 158 Figure SA.41 31P{1H} NMR of [(tSiP3)2Fe][OTf]2 (5) in DMSO-d6 159 Figure SA.42 1H NMR of [(tSiP3)2Fe][OTf]2 (5) in DMSO-d6 160 SA.2 Crystallographic Data SA.2.1 Crystal Data and Experimental for Complex 2 Crystal structure from crystals provide, they appear twinned, but refinement using the 2 components of the twin is reported here. The second domain is rotated from first domain by 179.9 degrees about the reciprocal axis 0.003 1.000 -0.001 and real axis -0.067 1.000 -0.030. BASF value 0.2829(7). Compound MRS1018ATW CCDC 1871069 Formula C21H42F6FeN3O6P3S2Si Dcalc./ g cm-3 1.490 /mm-1 0.789 Formula Weight 787.54 Colour orange Shape block Size/mm3 0.17×0.12×0.11 T/K 173(2) Crystal System triclinic Space Group P-1 a/Å 10.1234(16) b/Å 10.4065(17) Experimental. Single orange block-shaped crystals of c/Å 16.720(3) MRS1018ATW were used as received. A suitable crystal / ° 87.180(2) 0.17×0.12×0.11 mm3 was selected and mounted on a None on an / ° 89.315(2) Bruker APEX-II CCD diffractometer. The crystal was kept at a /° 86.376(2) 3 steady T = 173(2) K during data collection. The structure was V/Å 1755.7(5) Z 2 solved with the ShelXT (Sheldrick, G.M. (2015). Acta Cryst. Z' 1 A71, 3-8) structure solution program using the Intrinsic Phasing Wavelength/Å 0.710730 solution method and by using Olex2 (Dolomanov et al., 2009) as Radiation type MoK the graphical interface. The model was refined with version min/° 1.219 2018/3 of ShelXL (Sheldrick, Acta Cryst. A64 2008, 112-122) max/° 25.900 using Least Squares minimisation. Measured Refl. 12496 Independent 12496 Crystal Data. C21H42F6FeN3O6P3S2Si, Mr = 787.54, triclinic, Refl. P-1 (No. 2), a = 10.1234(16) Å, b = 10.4065(17) Å, c = Reflections with 9842 16.720(3) Å,  = 87.180(2)°,  = 89.315(2)°,  = 86.376(2)°, V = I > 2(I) Rint 0.0460 1755.7(5) Å3, T = 173(2) K, Z = 2, Z' = 1, (MoK) = 0.789, Parameters 401 12496 reflections measured, 12496 unique (Rint = 0.0460) which Restraints 0 were used in all calculations. The final wR2 was 0.1223 (all data) Largest Peak 0.830 and R1 was 0.0460 (I > 2(I)). Deepest Hole -0.318 GooF 1.038 wR2 (all data) 0.1223 wR2 0.1113 R1 (all data) 0.0618 R1 0.0460 161 SA.2.2 Crystal Data and Experimental for Complex 3 Crystal Structure has two Fe(L)(NH3)3, 4 triflate anions, two waters and one Nitromethane moieties in the asymmetric cell. One of the triflates is disordered. Compound MRS820D Formula C31H89F12Fe2N7O15P6S4Si2 CCDC 2023779 Dcalc./ g cm-3 1.502 /mm-1 7.170 Formula Weight 1510.03 Colour orange Shape plate Size/mm3 0.35×0.30×0.06 T/K 100.01(10) Crystal System triclinic Space Group P-1 a/Å 14.3915(3) b/Å 16.3759(3) c/Å 16.8383(3) /° 72.9448(18) / ° 64.938(2) /° 70.772(2) Experimental. Single orange plate crystals of MRS820D V/Å3 3338.88(14) Used as received. A suitable crystal with dimensions 0.35 × Z 2 0.30 × 0.06 mm3 was selected and mounted on a nylon loop Z' 1 Wavelength/Å 1.54184 with paratone oil on a XtaLAB Synergy, Dualflex, HyPix Radiation type Cu K diffractometer. The crystal was kept at a steady T = min/° 2.905 100.01(10) K during data collection. The structure was solved max/° 77.296 with the ShelXT (Sheldrick, 2015) solution program using Measured Refl's. 49696 dual methods and by using Olex2 (Dolomanov et al., 2009) as Indep't Refl's 13679 the graphical interface. The model was refined with ShelXL Refl's I≥2 (I) 11971 2018/3 (Sheldrick, 2015) using full matrix least squares Rint 0.0702 minimisation on F2. Parameters 807 Restraints 30 Largest Peak 0.978 Crystal Data. C31H89F12Fe2N7O15P6S4Si2, Mr = 1510.03, Deepest Hole -0.832 triclinic, P-1 (No. 2), a = 14.3915(3) Å, b = 16.3759(3) Å, c = GooF 1.073 16.8383(3) Å,  = 72.9448(18)°,  = 64.938(2)°,  = wR2 (all data) 0.1683 70.772(2)°, V = 3338.88(14) Å3, T = 100.01(10) K, Z = 2, Z' = wR2 0.1617 R1 (all data) 0.0660 1, (Cu K) = 7.170, 49696 reflections measured, 13679 R1 0.0588 unique (Rint = 0.0702) which were used in all calculations. The final wR2 was 0.1683 (all data) and R1 was 0.0588 (I≥2 (I)). 162 SA.2.3 Crystal Data and Experimental for Complex 4 Crystal Structure from crystal provided. The ether co-solvent shown at 50 percent occupancy resides on mirror plane (1 diethyl ether molecule per Fe dimer). The triflate is disordered across the mirror plane; there is one per molecule of triflate per Fe dimer. Compound MRS1017B Formula C31H82F3Fe2N3O4P6S Si2 Dcalc./ g cm-3 1.320 /mm-1 0.898 Formula Weight 1003.75 Colour red Shape parallelogram Size/mm3 0.16×0.10×0.04 T/K 173(2) Crystal System orthorhombic Space Group Pnma a/Å 19.3780(14) b/Å 25.3037(18) c/Å 10.3017(7) /° 90 /° 90 /° 90 Experimental. Single red parallelogram-shaped crystals V/Å3 5051.3(6) of (MRS1017B) were used as received. A suitable crystal Z 4 (0.16×0.10×0.04) mm3 was selected and mounted on a nylon Z' 0.5 loop with paratone oil on a Bruker APEX-II CCD Wavelength/Å 0.710730 Radiation type MoK diffractometer. The crystal was kept at T = 173(2) K during min/° 1.610 data collection. Using Olex2 (Dolomanov et al., 2009), the max/° 25.413 structure was solved with the XT (Sheldrick, 2015) structure Measured Refl. 31969 solution program, using the Intrinsic Phasing solution method. Independent Refl. 4760 The model was refined with version of XL (Sheldrick, 2008) Reflections Used 2655 using Least Squares minimisation. Rint 0.1284 Parameters 306 Crystal Data. C31H82F3Fe2N3O4P6SSi2, Mr = 1003.75, Restraints 118 orthorhombic, Pnma (No. 62), a = 19.3780(14) Å, b = Largest Peak 0.560 Deepest Hole -0.522 25.3037(18) Å, c = 10.3017(7) Å,  =  =  = 90°, V = GooF 1.011 5051.3(6) Å3, T = 173(2) K, Z = 4, Z' = 0.5, (MoK) = 0.898, wR2 (all data) 0.1538 31969 reflections measured, 4760 unique (Rint = 0.1284) which wR2 0.1208 were used in all calculations. The final wR2 was 0.1538 (all R1 (all data) 0.1241 data) and R1 was 0.0569 (I > 2(I)). R1 0.0569 163 SA.2.4 Crystal Data and Experimental for Complex 6 Crystal structure was determined even though the crystal turned out to be twinned. Compound MRS1018B CCDC 1872543 Formula C25H55F6FeN2O6P5S2Si Dcalc./ g cm-3 1.460 /mm-1 0.763 Formula Weight 896.62 Colour yellow Shape needle Size/mm3 0.26×0.06×0.04 T/K 173(2) Crystal System monoclinic Experimental. Single yellow needle-shaped crystals Flack Parameter 0.48(5) of MRS1018B were used as received. A suitable crystal Hooft Parameter 0.477(14) 0.26×0.06×0.04 mm3 was selected and mounted on a None Space Group Pc on an Bruker APEX-II CCD diffractometer. The crystal a/Å 10.6915(16) was kept at a steady T = 173(2) K during data collection. b/Å 10.0541(15) c/Å 38.082(6) The structure was solved with the ShelXT (Sheldrick, G.M. /° 90 (2015). Acta Cryst. A71, 3-8) structure solution program /° 94.929(2) using the Intrinsic Phasing solution method and by using /° 90 Olex2 (Dolomanov et al., 2009) as the graphical interface. V/Å3 4078.4(11) The model was refined with version 2018/3 of ShelXL Z 4 (Sheldrick, Acta Cryst. A64 2008, 112-122) using Least Z' 2 Squares minimisation. Wavelength/Å 0.710730 Radiation type MoK Crystal Data. C25H55F6FeN2O6P5S2Si, Mr = 896.62, min/° 1.912 monoclinic, Pc (No. 7), a = 10.6915(16) Å, b = max/° 26.069 Measured Refl. 47249 10.0541(15) Å, c = 38.082(6) Å,  = 94.929(2)°,  =  = Independent Refl. 15781 90°, V = 4078.4(11) Å3, T = 173(2) K, Z = 4, Z' = 2, Reflections with I > 12478 (MoK) = 0.763, 47249 reflections measured, 15781 2(I) unique (Rint = 0.0694) which were used in all calculations. Rint 0.0694 The final wR2 was 0.2379 (all data) and R1 was 0.0933 (I > Parameters 812 2(I)). Restraints 68 Largest Peak 1.399 Deepest Hole -1.168 GooF 1.057 wR2 (all data) 0.2379 wR2 0.2235 R1 (all data) 0.1135 R1 0.0933 164 SA.2.5 Crystal Data and Experimental for Complex 7 Crystal structure of compound shows one THF solvent co-crystallized with the molecule of interest. Compound tSiP3dmpeFeNH3 CCDC 2000513 Formula C25H60F6FeNO7P5S2Si Dcalc./ g cm-3 1.460 /mm-1 0.759 Formula Weight 903.65 Colour orange Shape chunk Size/mm3 0.18×0.10×0.08 T/K 173(2) Crystal System triclinic Space Group P-1 a/Å 11.3958(13) b/Å 12.2488(13) c/Å 14.9672(16) /° 82.5200(14) Experimental. Single orange chunk crystals of /° 87.2950(14) tSiP3dmpeFeNH3 used as received. A suitable crystal with /° 83.0945(14) dimensions 0.18 × 0.10 × 0.08 mm3 was selected and mounted V/Å3 2055.3(4) on a nylon loop with paratone oil on a Bruker SMART APEX Z 2 Z' 1 CCD area detector diffractometer. The crystal was kept at a Wavelength/Å 0.71073 steady T = 173(2) K during data collection. The structure was Radiation type MoK solved with the ShelXT (Sheldrick, G.M. (2015). Acta Cryst. min/° 1.373 A71, 3-8) solution program using SHELXT and by using Olex2 max/° 25.415 (Dolomanov et al., 2009) as the graphical interface. The model Measured Refl's. 29469 was refined with ShelXL 2018/3 (Sheldrick, 2015) using full Ind't Refl's 7559 matrix least squares minimisation on F2. Refl's with I > 5218 2(I) Crystal Data. C25H60F6FeNO7P5S2Si, Mr = 903.65, triclinic, Rint 0.0767 Parameters 447 P-1 (No. 2), a = 11.3958(13) Å, b = 12.2488(13) Å, c = Restraints 0 14.9672(16) Å,  = 82.5200(14)°,  = 87.2950(14)°,  = Largest Peak 1.060 83.0945(14)°, V = 2055.3(4) Å3, T = 173(2) K, Z = 2, Z' = 1, Deepest Hole -0.431 (MoK) = 0.759, 29469 reflections measured, 7559 unique GooF 1.018 (Rint = 0.0767) which were used in all calculations. The final wR2 wR2 (all data) 0.1595 was 0.1595 (all data) and R1 was 0.0556 (I > 2(I)). wR2 0.1361 R1 (all data) 0.0870 R1 0.0556 165 SA.2.6 Crystal Structure and Experimental for Complex 10 Crystal structure of needles crystallized in vial. Compound MRS1218A CCDC 1883822 Formula C21H56F6FeN5O8P3S4Si Dcalc./ g cm-3 1.482 /mm-1 0.781 Formula Weight 925.79 Color orange Shape needle Size/mm3 0.44×0.16×0.07 T/K 173(2) Crystal System monoclinic Space Group P21/c a/Å 24.469(3) b/Å 13.1139(14) c/Å 26.041(3) /° 90 /° 96.7260(10) /° 90 Experimental. Single orange needle-shaped crystals of V/Å3 8298.6(15) MRS1218A were used as received. A suitable crystal Z 8 Z' 2 0.44×0.16×0.07 mm3 was selected and mounted on a nylon Wavelength/Å 0.710730 loop with paratone oil on a Bruker APEX-II CCD Radiation type MoK diffractometer. The crystal was kept at a steady T = 173(2) K min/° 1.575 during data collection. The structure was solved with the max/° 25.412 ShelXT (Sheldrick, G.M. (2015). Acta Cryst. A71, 3-8) Measured Refl. 66418 structure solution program using the Intrinsic Phasing Independent 15304 solution method and by using Olex2 (Dolomanov et al., 2009) Refl. Reflections with 10808 as the graphical interface. The model was refined with I > 2(I) version 2018/3 of ShelXL (Sheldrick, Acta Cryst. A64 2008, Rint 0.0661 112-122) using Least Squares minimization. Parameters 923 Restraints 0 Crystal Data. C21H56F6FeN5O8P3S4Si, Mr = 925.79, Largest Peak 1.569 monoclinic, P21/c (No. 14), a = 24.469(3) Å, b = Deepest Hole -0.833 13.1139(14) Å, c = 26.041(3) Å,  = 96.7260(10)°,  =  = GooF 1.032 wR2 (all data) 0.1651 90°, V = 8298.6(15) Å3, T = 173(2) K, Z = 8, Z' = 2, wR2 0.1442 (MoK) = 0.781, 66418 reflections measured, 15304 R1 (all data) 0.0869 unique (Rint = 0.0661) which were used in all calculations. R1 0.0578 The final wR2 was 0.1651 (all data) and R1 was 0.0578 (I > 2(I)). 166 SA.2.7 Crystal Structure and Experimental for Complex 11 Triflate salt of Complex 11 did not give suitable single crystals. Instead triflate was exchanged by PF6– to prepare its corresponding PF6– salt which gave suitable single crystals for structure determination. Compound MRS720A Formula C21H55F12FeOP7SSi CCDC 2020298 Dcalc./ g cm-3 1.582 /mm-1 7.681 Formula Weight 884.44 Colour red Shape plate Size/mm3 0.21×0.10×0.02 T/K 99.99(10) Crystal System monoclinic Space Group P21/n a/Å 11.09300(10) b/Å 14.85970(10) Experimental. Single red plate crystals of MRS720A c/Å 22.7442(2) recrystallised from a mixture of DMSO and THF by solvent /° 90 layering. A suitable crystal with dimensions 0.21 × 0.10 × /° 97.9050(10) 0.02 mm3 was selected and mounted on a nylon loop with /° 90 paratone oil on a XtaLAB Synergy, Dualflex, HyPix V/Å3 3713.50(5) diffractometer. The crystal was kept at a steady T = 99.99(10) K Z 4 Z' 1 during data collection. The structure was solved with the ShelXT Wavelength/Å 1.54184 (Sheldrick, G.M. (2015). Acta Cryst. A71, 3-8) solution program Radiation type Cu K using dual methods and by using Olex2 (Dolomanov et al., 2009) min/° 3.563 as the graphical interface. The model was refined with ShelXL max/° 77.404 (Sheldrick, Acta Cryst. A64 2008, 112-122) using full matrix Measured Refl's. 52959 least squares minimisation on F2. Indep't Refl's 7758 Refl's I≥2 s(I) 7285 Crystal Data. C21H55F12FeOP7SSi, Mr = 884.44, Rint 0.0462 monoclinic, P21/n (No. 14), a = 11.09300(10) Å, b = Parameters 412 Restraints 0 14.85970(10) Å, c = 22.7442(2) Å,  = 97.9050(10)°,  =  = Largest Peak 0.604 90°, V = 3713.50(5) Å3, T = 99.99(10) K, Z = 4, Z' = 1, (Cu Deepest Hole -0.837 K) = 7.681, 52959 reflections measured, 7758 unique (Rint = GooF 1.034 0.0462) which were used in all calculations. The final wR2 was wR2 (all data) 0.0829 0.0829 (all data) and R1 was 0.0328 (I≥2 s(I)). wR2 0.0816 R1 (all data) 0.0350 R1 0.0328 167 SA.2.8 Crystal Structure and Experimental for Complex 13 Crystal structure of the best data set is presented here. the data shows the complex and disordered Triflate along with a 50:50 disorder of DMSO/THF solvent molecule. Not all atoms were refined anisotropic. Compound MRS820C Formula C18H52F6FeN6O7P3S2.5Si CCDC 2026779 Dcalc./ g cm-3 1.554 /mm-1 7.036 Formula Weight 835.65 Colour orange Shape needle Size/mm3 0.12×0.08×0.06 T/K 100(2) Crystal System orthorhombic Flack Parameter 0.147(10) Hooft Parameter 0.143(3) Space Group Pna21 a/Å 26.0800(2) Experimental. Single orange needle crystals of b/Å 13.96248(12) MRS820C used as received. A suitable crystal with c/Å 9.81043(8) dimensions 0.12 × 0.08 × 0.06 mm3 was selected and /° 90 mounted on a nylon loop with paratone oil on a XtaLAB /° 90 Synergy, Dualflex, HyPix diffractometer. The crystal was /° 90 kept at a steady T = 100(2) K during data collection. The V/Å3 3572.38(5) structure was solved with the ShelXT (Sheldrick, 2015) Z 4 Z' 1 solution program using dual methods and by using Olex2 Wavelength/Å 1.54184 (Dolomanov et al., 2009) as the graphical interface. The Radiation type Cu K model was refined with ShelXL 2018/3 (Sheldrick, 2015) min/° 3.389 using full matrix least squares minimisation on F2. max/° 77.510 Measured Refl's. 35587 Crystal Data. C18H52F6FeN6O7P3S2.5Si, Mr = 835.65, Indep't Refl's 5917 orthorhombic, Pna21 (No. 33), a = 26.0800(2) Å, b = Refl's I≥2 s(I) 5776 13.96248(12) Å, c = 9.81043(8) Å,  =  =  = 90°, V = Rint 0.0618 3572.38(5) Å3, T = 100(2) K, Z = 4, Z' = 1, (Cu K) = Parameters 272 Restraints 15 7.036, 35587 reflections measured, 5917 unique (Rint = Largest Peak 1.576 0.0618) which were used in all calculations. The final wR2 Deepest Hole -0.858 was 0.1697 (all data) and R1 was 0.0648 (I≥2 s(I)). GooF 1.028 wR2 (all data) 0.1697 wR2 0.1683 R1 (all data) 0.0660 R1 0.0648 168 SA.3 Cartesian Coordinates of the Optimized Iron Complexes Sum of electronic and thermal Free Energies = –3490.742126 Hartree Fe 1.17812300 0.00055300 -0.00001800 P -0.14531200 1.19573800 1.51621000 P -0.14496400 -1.91168500 0.27650100 P -0.14644800 0.71543200 -1.79359600 Si -2.64359000 -0.00176700 0.00036100 C -0.21432900 2.53304400 -2.04148900 H -0.64306800 3.04023600 -1.17945200 H -0.82478800 2.76424400 -2.91535100 H 0.78861700 2.92305700 -2.20704800 C -0.20926500 0.50108300 3.21402700 H -0.63248100 -0.50133500 3.22192500 H -0.82279200 1.13898800 3.85148200 H 0.79413700 0.45466100 3.63410200 C 0.41051100 0.16908700 -3.45335200 H 1.41244900 0.55050500 -3.65144400 H -0.26632400 0.54823800 -4.22005300 H 0.43518100 -0.91761500 -3.51486300 C 0.40915300 2.90701100 1.87311300 H 1.41171800 2.88979700 2.30105600 H -0.26781900 3.37920700 2.58619800 H 0.43139700 3.50448100 0.96332600 C -0.21159600 -3.03263300 -1.17551300 H -0.64100100 -2.53795000 -2.04436400 H -0.82106800 -3.90616800 -0.94059400 H 0.79168200 -3.36903800 -1.43151300 C -1.92291000 -1.63372500 0.65978700 H -2.04252700 -1.64427600 1.74603100 H -2.49241800 -2.48854800 0.28669300 C -1.92417200 0.24422100 -1.74250900 H -2.04398300 -0.69085200 -2.29510700 H -2.49478300 0.99462100 -2.29523400 C -1.92408300 1.38482700 1.08471000 H -2.04613800 2.33169000 0.55302400 169 H -2.49334100 1.48604800 2.01200100 C -4.54315000 -0.00066500 0.00086300 C 0.41314700 -3.07794500 1.57702400 H 1.41480700 -3.43989000 1.34389300 H -0.26389400 -3.93138400 1.63112400 H 0.43905500 -2.58976800 2.54966000 C -5.06302000 1.39983000 -0.38017400 H -4.73447100 1.71131000 -1.37469200 H -4.75616000 2.16586500 0.33536700 H -6.15616500 1.39572200 -0.39456700 C -5.06368400 -0.36979800 1.40439700 H -4.73292900 0.33482800 2.17129000 H -4.75945300 -1.37334600 1.70969000 H -6.15679700 -0.35235500 1.40853100 C -5.06565700 -1.03033200 -1.02083200 H -4.73689400 -2.04765500 -0.79494000 H -4.76140600 -0.79291000 -2.04260000 H -6.15878700 -1.04045400 -1.00740500 N 2.38739500 -0.54870700 1.48691400 N 2.38863600 -1.01073500 -1.21878300 N 2.38667100 1.56358400 -0.26687100 C 3.15447800 -1.55678900 -1.87674600 C 3.15387400 -0.84609600 2.28791800 C 3.15211400 2.40686200 -0.40997500 C 4.11142100 3.48147500 -0.58961500 H 4.02530400 3.89802000 -1.59408200 H 3.92643700 4.27334300 0.13762300 H 5.12552600 3.10473600 -0.45035700 C 4.11542300 -1.23062600 3.30536600 H 4.02643900 -0.57586600 4.17341900 H 3.93554600 -2.25954200 3.62004400 H 5.12900000 -1.15484800 2.90974500 C 4.11540000 -2.24941000 -2.71611100 H 4.08163300 -3.32196700 -2.52015000 H 3.88676300 -2.07511200 -3.76846900 H 5.12286800 -1.88614000 -2.50937700 170 Sum of electronic and thermal Free Energies = –3490.525941 Hartree Fe -1.17018000 0.00032200 -0.00011900 P 0.17053100 -1.85157800 0.72043000 P 0.17056500 1.54999300 1.24370100 P 0.17003000 0.30293700 -1.96369800 Si 2.66502100 0.00002500 -0.00030700 C 0.20270500 -1.16018500 -3.05415300 H 0.61805800 -2.03412000 -2.55767400 H 0.82564300 -0.93008600 -3.92066000 H -0.80037000 -1.39380300 -3.40432500 C 0.20550000 -2.06198600 2.53301700 H 0.62277300 -1.19482600 3.03966600 H 0.82773700 -2.92787000 2.76734100 H -0.79715700 -2.24652200 2.91243900 C -0.44258000 1.62002200 -3.06587200 H -1.44742800 1.38316800 -3.41409400 H 0.22075000 1.68681900 -3.93052700 H -0.45678900 2.58424400 -2.56223800 C -0.44239700 -3.46519700 0.13299900 H -1.44675700 -3.64831600 0.51349000 H 0.22162500 -4.24685800 0.50742000 H -0.45752200 -3.51186500 -0.95386300 C 0.20486900 3.22548700 0.52105300 H 0.62407900 3.23205700 -0.48243400 H 0.82537600 3.86135900 1.15548500 H -0.79817400 3.64533900 0.48986400 C 1.92302900 1.06532800 1.40002600 H 2.04708900 0.52995400 2.34431100 H 2.50070600 1.98886000 1.49976800 C 1.92276500 0.67965700 -1.62292000 H 2.04754300 1.76497800 -1.63247800 H 2.49990700 0.30321000 -2.47249500 C 1.92268000 -1.74512900 0.22176600 H 2.04590300 -2.29454200 -0.71446600 H 2.50070900 -2.29409600 0.97089000 C 4.55497500 -0.00032200 0.00001800 C -0.44260400 1.84700400 2.93501100 171 H -1.44738700 2.26705700 2.90338600 H 0.22065500 2.56278400 3.42474400 H -0.45694200 0.92918800 3.51899400 C 5.06958400 -1.01482900 -1.04284600 H 4.74475200 -0.77751500 -2.05894900 H 4.76968100 -2.04007400 -0.81472400 H 6.16197100 -1.00052200 -1.05254900 C 5.06873800 -0.39693800 1.40009600 H 4.74114300 -1.39460100 1.70295500 H 4.77064600 0.31432300 2.17360400 H 6.16108300 -0.41535900 1.39263500 C 5.07030300 1.40992500 -0.35668700 H 4.74486900 2.17143800 0.35638200 H 4.77190400 1.72487900 -1.35908900 H 6.16267300 1.41091300 -0.33802000 N -2.36918300 -0.30006900 1.60736800 N -2.36936000 1.54230300 -0.54400200 N -2.36793100 -1.24264000 -1.06376700 C -3.16133300 2.32796700 -0.81789400 C -3.16071400 -0.45593100 2.42510500 C -3.15908400 -1.87380800 -1.60709300 C -4.14879000 -2.67827200 -2.29409200 H -4.12640000 -2.46670200 -3.36492000 H -3.94305200 -3.73877300 -2.13537600 H -5.14423600 -2.44901100 -1.90839300 C -4.15308500 -0.65045500 3.46242100 H -4.12796500 -1.68285700 3.81661900 H -3.95252900 0.01957300 4.30087700 H -5.14808400 -0.43556700 3.06740300 C -4.15254000 3.32429000 -1.16934600 H -4.17877300 4.11058900 -0.41242200 H -3.90856200 3.76799200 -2.13673300 H -5.13846800 2.85923200 -1.23179800 172 Sum of electronic and thermal Free Energies = –3262.080341 Hartree Fe 1.81746800 0.00048100 -0.00053500 P 0.48932800 1.85471400 -0.53892000 P 0.49024100 -0.46000800 1.87402200 P 0.49093500 -1.39557500 -1.33465000 Si -2.01330200 -0.00021900 -0.00109800 N 3.13610300 0.34357500 -1.65993700 H 3.47073900 -0.51158200 -2.09834800 H 3.97653100 0.86759500 -1.42307300 H 2.69617900 0.87041900 -2.40899400 C 0.43656700 -0.95592100 -3.12096400 H 0.03074300 0.04246900 -3.27517200 H -0.19396700 -1.66458900 -3.65943200 H 1.43231500 -0.99793000 -3.56392600 C 0.43121700 3.17705000 0.73974300 H 0.02096700 2.80766200 1.67799200 H -0.19684400 3.99975800 0.39556300 H 1.42633200 3.57971200 0.93217500 C 1.01759600 -3.15637700 -1.45810900 H 2.02622400 -3.23591700 -1.86849600 H 0.34427600 -3.70243800 -2.12018200 H 0.99361700 -3.64882000 -0.48513300 C 1.01677400 2.84797400 -1.99788500 H 2.02434500 3.24461500 -1.85860500 H 0.34154400 3.69318300 -2.13783500 H 0.99644500 2.25552700 -2.91352300 C 0.43368300 -2.22807000 2.38176400 H 0.02196100 -2.85670500 1.59408200 H -0.19278300 -2.34079800 3.26753500 H 1.42932000 -2.59552500 2.63286600 C -1.29099000 -0.01585600 1.75891400 H -1.41859800 0.98272000 2.18469000 H -1.85800100 -0.69043600 2.40517400 C -1.29095100 -1.51679900 -0.89422700 H -1.42016000 -2.38512000 -0.24301100 H -1.85703900 -1.73831200 -1.80232100 C -1.29146400 1.53203900 -0.86761100 173 H -1.41797900 1.40113500 -1.94539600 H -1.85953800 2.42844200 -0.60696000 C -3.91120200 -0.00000100 0.00002500 C 1.01718000 0.30945800 3.46254200 H 2.02445500 -0.00964300 3.73770000 H 0.34195800 0.00968600 4.26509300 H 0.99643300 1.39846400 3.40428600 C -4.43250200 0.19151600 -1.43853800 H -4.10387100 -0.60354200 -2.11235800 H -4.12950400 1.14936500 -1.86713100 H -5.52545700 0.17404900 -1.43855400 C -4.43015300 1.15073800 0.88570400 H -4.09989300 2.13136000 0.53433900 H -4.12691600 1.04201500 1.92934000 H -5.52309100 1.16136800 0.87102300 C -4.43142100 -1.34165000 0.55424200 H -4.10039400 -1.52812100 1.57883600 H -4.12999400 -2.19155400 -0.06200700 H -5.52433600 -1.33276800 0.57194600 N 3.14380200 1.26398800 1.12157000 H 4.01803800 0.81314500 1.38502300 H 2.73901100 1.59067200 1.99449500 H 3.41926100 2.10825500 0.62502000 N 3.14334300 -1.60182900 0.53528700 H 2.73136600 -2.52084400 0.40041000 H 3.43104100 -1.58049500 1.51102200 H 4.01102100 -1.61763600 0.00230300 174 Sum of electronic and thermal Free Energies = –3261.877617 Hartree Fe -1.83027400 0.00003600 -0.00027100 P -0.46986600 1.84781800 -0.76457700 P -0.47036700 -1.58592800 -1.21753700 P -0.47113500 -0.26185100 1.98229400 Si 2.02670100 -0.00017900 0.00026800 N -3.11360200 1.31262800 1.09723400 H -3.39118500 0.96103300 2.01223300 H -3.99121400 1.50871300 0.61541500 H -2.70533600 2.22708800 1.27628400 C -0.44982700 1.22378400 3.04921500 H -0.04431700 2.09278800 2.53490000 H 0.18543600 1.01809000 3.91335100 H -1.44403800 1.46242400 3.42683400 C -0.44782000 2.02753900 -2.58470000 H -0.04434800 1.14626800 -3.07927500 H 0.18902300 2.87731300 -2.83948200 H -1.44165800 2.23667800 -2.98036000 C -1.05280600 -1.56317300 3.13114000 H -2.06265200 -1.35974400 3.49152200 H -0.38964500 -1.57870900 3.99901800 H -1.01886100 -2.55464800 2.67872300 C -1.05161300 3.49378100 -0.21309800 H -2.06166100 3.70404900 -0.56894000 H -0.38852400 4.25275900 -0.63442100 H -1.01717600 3.59860700 0.87167200 C -0.44859000 -3.25242700 -0.46372200 H -0.04358400 -3.24096900 0.54618300 H 0.18696800 -3.89810300 -1.07343800 H -1.44270200 -3.69895900 -0.44590600 C 1.28275700 -1.10731500 -1.36873500 H 1.41443600 -0.60013000 -2.32786200 H 1.86090800 -2.03355900 -1.43919500 C 1.28240800 -0.63138700 1.64391600 H 1.41496000 -1.71543600 1.68555700 H 1.85998800 -0.22788800 2.48099600 C 1.28312600 1.73932900 -0.27393500 175 H 1.41437300 2.31578000 0.64529700 H 1.86160700 2.26394300 -1.04028100 C 3.91527500 -0.00027000 0.00024100 C -1.05145000 -1.93095300 -2.91905000 H -2.06155700 -2.34417500 -2.92376000 H -0.38851100 -2.67550700 -3.36556700 H -1.01626600 -1.04376400 -3.55197600 C 4.42811200 1.05027700 1.00822000 H 4.10194100 0.84971400 2.03186200 H 4.13212000 2.06765000 0.74279300 H 5.52035800 1.03390100 1.02060200 C 4.42799700 0.34759600 -1.41352500 H 4.10137800 1.33422100 -1.75169200 H 4.13232700 -0.39106500 -2.16188800 H 5.52023300 0.36698500 -1.40549300 C 4.42846800 -1.39836600 0.40585800 H 4.10197900 -2.18464800 -0.27941300 H 4.13307200 -1.67715000 1.41981100 H 5.52070200 -1.40072400 0.38482200 N -3.11448000 0.29509900 -1.68440200 H -4.00029600 -0.20438400 -1.60251600 H -2.71576200 -0.02499400 -2.56393400 H -3.37659800 1.26581600 -1.84805900 N -3.11525900 -1.60640800 0.58565100 H -2.71327800 -2.21369300 1.29605100 H -3.38393300 -2.22801500 -0.17545000 H -3.99770900 -1.28629700 0.98523900 176 Sum of electronic and thermal Free Energies= -4146.139853 Hartree Fe -0.86633700 0.01522800 0.06430500 P 0.53977700 1.94518100 -0.02653400 P 0.49515000 -1.05122600 -1.60568400 P 0.56100100 -0.96288400 1.72324200 P -2.54243300 -1.74031900 0.12327600 P -2.51172400 1.09845000 -1.35356000 Si 3.02613900 0.02182200 -0.00766300 C 0.65617200 0.01662500 3.27858900 H 0.94996600 1.04754300 3.09439000 H 1.38423300 -0.43628200 3.95225900 H -0.31238200 0.01605200 3.77579500 C 0.68026600 2.80787600 -1.64665200 H 1.01641300 2.13714700 -2.43395700 H 1.39390500 3.62875200 -1.56557700 H -0.27980500 3.22550600 -1.94218800 C 0.15496700 -2.60632500 2.44333700 H -0.83454100 -2.60011300 2.89848000 H 0.88518600 -2.85349100 3.21533300 H 0.18601400 -3.38647800 1.68584000 C 0.05631800 3.35476800 1.05062800 H -0.94233800 3.71306900 0.80838800 H 0.76299300 4.17436200 0.91459800 H 0.07401500 3.05813300 2.09810600 C -3.07159000 2.79192500 -0.90268900 H -3.25118400 2.87193200 0.16708800 H -4.00163900 3.01557900 -1.42723500 H -2.33824200 3.54319700 -1.19038200 C 0.76512200 -2.86731000 -1.44042500 H 1.00725800 -3.16925200 -0.42554600 H 1.58544900 -3.16594700 -2.09374000 H -0.12232000 -3.40861300 -1.75503400 C -4.09274300 0.15266900 -1.17421700 H -4.56627200 0.47056800 -0.24277600 H -4.77510800 0.42620300 -1.98142100 C 2.24358800 -0.46062900 -1.66701200 H 2.28996000 0.40713100 -2.32820000 177 H 2.83217900 -1.23892000 -2.15836300 C 2.34240800 -1.17800100 1.29186400 H 2.48162800 -2.19761600 0.92642200 H 2.91945900 -1.11099800 2.21751300 C 2.31161500 1.72202800 0.42702600 H 2.40906000 1.87488500 1.50423400 H 2.88501100 2.52313100 -0.04542000 C 4.92433800 0.01783200 -0.07061800 C 0.04687000 -1.01592000 -3.38842000 H -0.92323000 -1.48361600 -3.55506100 H 0.79559800 -1.57104300 -3.95530800 H 0.01609200 0.00217900 -3.76823400 C 5.48918700 0.58173100 1.24847500 H 5.19117300 -0.01037700 2.11713700 H 5.18617000 1.61680200 1.42156200 H 6.58192000 0.56993100 1.21658100 C -2.39766600 1.27143000 -3.18011100 H -1.49615400 1.80439300 -3.47557500 H -3.26185800 1.83230100 -3.53851100 H -2.39711700 0.29771800 -3.66384300 C -3.56267900 -1.76587400 1.65654700 H -2.95014300 -1.96833100 2.53354300 H -4.32276700 -2.54373800 1.58037800 H -4.06255000 -0.81080100 1.80231400 C -3.81952700 -1.34600800 -1.15371400 H -3.43913100 -1.68482900 -2.12070300 H -4.72972400 -1.91663200 -0.95685900 C 5.40660500 0.89357000 -1.24428300 H 5.08239800 1.93289900 -1.15058400 H 5.06664800 0.51666100 -2.21146800 H 6.49946600 0.90394700 -1.27176600 C 5.43762200 -1.42237200 -0.26783800 H 5.07637200 -1.87119400 -1.19641000 H 5.15935400 -2.07950400 0.55906800 H 6.52959200 -1.42080900 -0.32077300 C -2.33161800 -3.54466500 -0.15907700 H -2.03205700 -3.74300700 -1.18538000 H -3.28781500 -4.04070600 0.01218400 H -1.59599000 -3.98070300 0.51223000 N -1.84751800 0.86185900 1.54037000 C -2.40125500 1.36282000 2.41481100 C -3.09448300 1.99154000 3.52494400 H -3.58392000 2.90929300 3.19510000 H -3.85060400 1.31594900 3.92844300 H -2.38683500 2.23897600 4.31801400 178 Sum of electronic and thermal Free Energies = –4145.929533 Hartree Fe -0.86745500 0.02115800 0.07654800 P 0.60116700 1.99431300 -0.22161600 P 0.51974300 -1.27220500 -1.44859300 P 0.64049700 -0.78938000 1.85935100 P -2.62538400 -1.77700900 0.22977600 P -2.58433300 1.01961400 -1.45504700 Si 3.07604000 0.01788900 -0.02645300 C 0.70547900 0.36458200 3.28081500 H 1.03285500 1.36019000 2.99067000 H 1.41009300 -0.02752200 4.01610100 H -0.27140000 0.43700500 3.75448500 C 0.68842800 2.65246400 -1.93021700 H 1.01975300 1.90260800 -2.64444100 H 1.40096500 3.47925100 -1.94611400 H -0.27701500 3.03781400 -2.24900000 C 0.15874600 -2.34523400 2.69684700 H -0.81338600 -2.24410500 3.17664100 H 0.89768400 -2.56389700 3.47033700 H 0.13095400 -3.18482900 2.00649200 C 0.08041600 3.46260000 0.73706300 H -0.92031700 3.78639500 0.45979200 H 0.77896300 4.27566100 0.53054000 H 0.10481200 3.25513300 1.80521800 C -3.08321400 2.75458500 -1.14899600 H -3.25577400 2.94115000 -0.09197200 H -4.01280800 2.94592400 -1.68866000 H -2.33780300 3.45687000 -1.51739700 C 0.79248200 -3.03897700 -1.03900700 H 1.08280600 -3.20278900 -0.00627100 H 1.59141800 -3.41103500 -1.68212700 H -0.09783700 -3.62281700 -1.24787900 C -4.14824000 0.09089000 -1.15329200 H -4.59740700 0.47427300 -0.23506100 H -4.84820200 0.31864900 -1.96056700 C 2.23265900 -0.64149700 -1.60142100 H 2.24772000 0.14526600 -2.35799100 179 H 2.82366900 -1.46152900 -2.01848700 C 2.39951400 -1.03030400 1.41096700 H 2.55510600 -2.08513300 1.17713600 H 2.98330000 -0.83556900 2.31494600 C 2.35451800 1.75913000 0.23806900 H 2.47306500 2.03287300 1.28874200 H 2.93207900 2.49299300 -0.33108900 C 4.96377100 0.00458600 -0.12898000 C -0.03450900 -1.42367400 -3.18570000 H -1.01188300 -1.90080300 -3.24970900 H 0.68423300 -2.04782000 -3.72015000 H -0.07421100 -0.45528300 -3.67755400 C 5.55049200 0.71834400 1.10692700 H 5.27748600 0.22897200 2.04506400 H 5.25304300 1.76779100 1.16586400 H 6.64152500 0.70217200 1.05108700 C -2.42681200 0.98406600 -3.27854200 H -1.51780300 1.47747000 -3.61605000 H -3.28063300 1.51446100 -3.70442200 H -2.43529600 -0.03523700 -3.65634600 C -3.57364000 -1.66681100 1.79547600 H -2.93879700 -1.85331400 2.65947900 H -4.36032300 -2.42274800 1.78150600 H -4.03772900 -0.68991400 1.90547200 C -3.89184700 -1.40697400 -1.05480100 H -3.53536700 -1.81018200 -2.00538500 H -4.80872700 -1.94945500 -0.81150200 C 5.41338000 0.74005000 -1.40876400 H 5.09465000 1.78524100 -1.42824700 H 5.05683900 0.25362900 -2.31955300 H 6.50476200 0.74497700 -1.45970300 C 5.46944900 -1.45248000 -0.16884900 H 5.09147300 -2.00685500 -1.03158200 H 5.21669000 -2.00850600 0.73671700 H 6.55915800 -1.45626200 -0.24769600 C -2.39412100 -3.58293400 0.05059000 H -2.11501100 -3.84438200 -0.96774900 H -3.34835200 -4.06618300 0.26916400 H -1.65080400 -3.97358100 0.74163100 N -1.88205600 1.03388800 1.47473300 C -2.43954900 1.60180500 2.30473300 C -3.14118900 2.31742000 3.35229800 H -4.06335300 2.75212300 2.96104000 H -3.39213700 1.63459900 4.16703200 H -2.51169900 3.11864000 3.74513000 180 Sum of electronic and thermal Free Energies = –4069.912723 Fe 0.96493100 -0.00001300 -0.32705900 P -0.53402100 1.76302900 -0.94705900 P -0.18007600 -0.23398100 1.76594100 P -0.56011400 -1.61697100 -1.23905900 P 2.72388700 -1.58206100 0.22394100 P 2.66197500 1.61494000 0.30294100 Si -2.90406900 0.01009400 0.13994100 N 1.83293300 0.08896300 -2.28905900 H 2.01290800 -0.83304200 -2.67205900 H 2.71894700 0.58293900 -2.35105900 H 1.21994600 0.55298000 -2.95205900 C -0.85810800 -1.41296200 -3.04905900 H -1.21308100 -0.41495300 -3.29705900 H -1.61112800 -2.13094200 -3.37605900 H 0.04888600 -1.61198700 -3.62105900 C -0.56997900 3.25803000 0.12694100 H -0.78998600 3.00503600 1.16194100 H -1.33596000 3.94805100 -0.22905900 H 0.38703500 3.77500300 0.09794100 C -0.13616400 -3.40798200 -1.22405900 H 0.79983100 -3.59600800 -1.74805900 H -0.92517900 -3.97396100 -1.72305900 H -0.04517400 -3.77998500 -0.20605900 C -0.23099900 2.56202000 -2.58205900 H 0.77401300 2.97499300 -2.65105900 H -0.94297600 3.37704000 -2.72305900 H -0.38201800 1.86002400 -3.40405900 C 3.11201200 2.93792800 -0.90205900 H 3.23000100 2.56192500 -1.91805900 H 4.06002400 3.38790200 -0.60405900 H 2.36103300 3.72594900 -0.91905900 C -0.36412300 -1.94197600 2.43494100 H -0.65814300 -2.66596800 1.68094100 H -1.11912300 -1.93795500 3.22194100 H 0.57086800 -2.27100200 2.87794100 C 4.27295000 0.70589600 0.43094100 181 H 4.68094700 0.58788500 -0.57705900 H 4.99296700 1.31487600 0.98094100 C -1.94506100 0.31106800 1.74794100 H -1.97503100 1.38106800 1.96494100 H -2.44407400 -0.17491900 2.58994100 C -2.27911500 -1.63292300 -0.57205900 H -2.32713500 -2.38792200 0.21594100 H -2.94212400 -1.98690500 -1.36505900 C -2.32603200 1.35407800 -1.06705900 H -2.53504100 1.01908400 -2.08505900 H -2.89500700 2.27709400 -0.93005900 C -4.78506900 0.01814600 0.39994100 C 0.41994700 0.58000200 3.30294100 H 1.42793700 0.24997400 3.54994100 H -0.23906100 0.31002000 4.12894100 H 0.41697600 1.66300200 3.20494100 C -5.49907100 -0.05983400 -0.96405900 H -5.25209600 -0.97184100 -1.51405900 H -5.27004700 0.79515900 -1.60405900 H -6.58207100 -0.06480400 -0.81405900 C 2.68100200 2.59494000 1.85794100 H 1.77701900 3.19096500 1.96694100 H 3.53902100 3.26791600 1.84494100 H 2.77098400 1.94293700 2.72394100 C 3.63586800 -2.25708600 -1.23305900 H 2.98785100 -2.87606900 -1.85405900 H 4.46085100 -2.88310900 -0.89105900 H 4.05989000 -1.46709800 -1.85305900 C 4.06791300 -0.65309800 1.08894100 H 3.77091600 -0.53709000 2.13394100 H 4.98889600 -1.24112400 1.08694100 C -5.20403300 1.31715800 1.11794100 H -4.94700800 2.21215000 0.54594100 H -4.75603100 1.40714500 2.10994100 H -6.28803300 1.32918800 1.25394100 C -5.19810200 -1.19284300 1.25994100 H -4.72910200 -1.18385600 2.24694100 H -4.96312900 -2.14384900 0.77794100 H -6.27910200 -1.17581300 1.42394100 C 2.63284500 -3.09605900 1.26194100 H 2.41785200 -2.84405300 2.29694100 H 3.59983100 -3.60108600 1.23394100 H 1.87482600 -3.79003800 0.90594100 182 Sum of electronic and thermal Free Energies = –4069.709733 Hartree Fe 0.97400400 0.01338800 -0.35042600 P -0.59577300 1.86587600 -0.86774800 P -0.21263900 -0.42328800 1.75376200 P -0.63893500 -1.55664000 -1.37653300 P 2.81171200 -1.60848400 0.19606800 P 2.74676200 1.63972200 0.37354600 Si -2.95285100 0.01273900 0.15367300 N 1.83647900 0.21112900 -2.27375000 H 2.00189400 -0.69220700 -2.70997600 H 2.73197700 0.69038600 -2.30550300 H 1.23142400 0.72021700 -2.91232500 C -0.90058000 -1.20978100 -3.16117200 H -1.28331600 -0.20650000 -3.33661100 H -1.63358900 -1.92010400 -3.54780700 H 0.01371900 -1.34818900 -3.73828000 C -0.56209100 3.26188500 0.31911600 H -0.76579100 2.93889300 1.33728000 H -1.32707400 3.98457400 0.02958500 H 0.40021500 3.76796100 0.29797000 C -0.14327100 -3.31976800 -1.43950700 H 0.78029100 -3.45183400 -2.00231900 H -0.92891800 -3.88429100 -1.94568500 H -0.01414600 -3.73611600 -0.44320200 C -0.24862500 2.71334200 -2.45852600 H 0.76205800 3.11751300 -2.49619000 H -0.94802300 3.54523200 -2.56418300 H -0.40757900 2.05323200 -3.31233100 C 3.14482300 3.01344600 -0.77790000 H 3.26986600 2.68640200 -1.80982400 H 4.08762800 3.46465500 -0.46193800 H 2.38203700 3.78956000 -0.75067100 C -0.41635900 -2.17783500 2.24509600 H -0.80272200 -2.80655100 1.44938000 H -1.12055800 -2.20925800 3.07806800 H 0.52478800 -2.59152800 2.59169700 C 4.33613100 0.70171800 0.43776600 183 H 4.73209100 0.61835900 -0.57735600 H 5.06535100 1.29091600 0.99870100 C -1.93787900 0.18894800 1.75727100 H -1.93331300 1.24039300 2.05110700 H -2.44558900 -0.34030900 2.56864700 C -2.33598700 -1.57847500 -0.69049400 H -2.40804100 -2.40312700 0.02111100 H -3.00515700 -1.83865200 -1.51554400 C -2.36791100 1.43224000 -0.97201500 H -2.60090000 1.17538300 -2.00787200 H -2.93379500 2.34204300 -0.75321100 C -4.81879500 0.01769200 0.45419900 C 0.49578800 0.27020100 3.29109200 H 1.50314200 -0.10488400 3.46843400 H -0.13488100 -0.04078200 4.12631100 H 0.51417200 1.35664000 3.26762200 C -5.55845600 0.04544100 -0.89980000 H -5.33641800 -0.82686500 -1.51935000 H -5.33998100 0.94529400 -1.47923800 H -6.63666900 0.03793200 -0.72359000 C 2.71195900 2.50184600 1.98789900 H 1.80032100 3.08039900 2.12018800 H 3.56064200 3.18727200 2.02656700 H 2.80534600 1.79688200 2.81030600 C 3.65499400 -2.21127400 -1.32272600 H 2.98861600 -2.81420600 -1.93934000 H 4.49185100 -2.84667400 -1.02793200 H 4.06031600 -1.39871900 -1.92435100 C 4.14049300 -0.67333300 1.06337200 H 3.86203200 -0.59183500 2.11666900 H 5.06392700 -1.25665700 1.03098300 C -5.20160900 1.26680300 1.27549800 H -4.95325300 2.20009500 0.76400200 H -4.73490300 1.27938700 2.26311300 H -6.28210800 1.27688300 1.43697300 C -5.21922900 -1.25269700 1.23310300 H -4.73069100 -1.32172000 2.20841000 H -5.01090100 -2.16908600 0.67634100 H -6.29498400 -1.23704500 1.42320400 C 2.69247600 -3.14334000 1.18484500 H 2.51037800 -2.91923000 2.23364500 H 3.65025400 -3.66280600 1.11745800 H 1.91728600 -3.81181700 0.81764900 184 Sum of electronic and thermal Free Energies = –2043.920375 Hartree Fe -0.00008700 -0.00001100 -0.00000300 C 1.70164700 1.16130700 0.36370500 C 1.70136900 0.70350500 -0.99249900 C 1.70055700 -0.72780200 -0.97617200 C 1.69965500 -1.15469200 0.39005300 C 1.70041900 0.01289300 1.21819200 C 1.80741400 2.58722100 0.80947400 C 1.80751400 1.56720000 -2.21126800 C 1.80632900 -1.61966700 -2.17468900 C 1.80223900 -2.57057700 0.86728800 C 1.80541500 0.03046600 2.71200700 C -1.70080500 -0.81354100 -0.90613300 C -1.70034200 0.61011000 -1.05494400 C -1.70040600 1.19168700 0.25309200 C -1.70139400 0.12754200 1.21021000 C -1.70146000 -1.11181500 0.49382500 C -1.80499400 -1.81307000 -2.01636700 C -1.80509500 1.35961700 -2.34732800 C -1.80363400 2.65297700 0.56392200 C -1.80663000 0.28288800 2.69597700 C -1.80805400 -2.47661800 1.10130700 H 2.85490200 0.03546500 3.02875600 H 1.33652400 0.91553600 3.14364900 H 1.33863400 -0.84555500 3.16428400 H 2.85114400 -2.87265000 0.96906300 H 1.33331100 -2.70809800 1.84237500 H 1.33363300 -3.26928200 0.17284500 H 2.85600700 -1.81130500 -2.42615100 H 1.33567100 -2.58884200 -2.00494200 H 1.34223300 -1.17530300 -3.05624600 H 2.85716900 1.74490000 -2.47276900 H 1.33428400 1.10690600 -3.07953500 H 1.34539600 2.54375500 -2.06083400 H 2.85701500 2.88998400 0.90140900 H 1.33737200 3.27204600 0.10271100 H 1.34216500 2.74569900 1.78331600 H -2.85789700 -2.76653200 1.22605600 H -1.33711300 -3.23880100 0.47921400 185 H -1.34433900 -2.52115800 2.08771600 H -2.85420400 -2.02472000 -2.25329000 H -1.33516700 -1.45533600 -2.93328600 H -1.33767200 -2.76331800 -1.75418000 H -2.85455500 1.52026300 -2.62054700 H -1.33595900 2.34242500 -2.28769600 H -1.33878900 0.81939400 -3.17230500 H -2.85265900 2.96428500 0.63026400 H -1.33481000 2.90156900 1.51678900 H -1.33520400 3.26745500 -0.20616200 H -2.85608600 0.31537100 3.01112100 H -1.33809700 -0.54673600 3.22675700 H -1.33908900 1.20507300 3.04378700 186 Sum of electronic and thermal Free Energies = – 2043.757440 Hartree Fe 0.00023600 0.00014700 0.00005200 C 1.73983500 0.46668300 -1.11705100 C 1.74986400 -0.92259000 -0.76565800 C 1.73593000 -1.01795000 0.66430400 C 1.71882400 0.31279300 1.19687300 C 1.72088000 1.23065700 0.09572600 C 1.83586300 1.01994100 -2.50104800 C 1.86083000 -2.06694300 -1.71932000 C 1.82550600 -2.27959200 1.45929000 C 1.78578800 0.67971300 2.64328300 C 1.79032400 2.71982700 0.19466100 C -1.73199100 -0.92448200 0.79642600 C -1.71799500 -1.06615500 -0.62981500 C -1.72402600 0.24615900 -1.20565400 C -1.74292100 1.19900800 -0.13535000 C -1.74694900 0.47588400 1.10175000 C -1.81885200 -2.04050500 1.78573200 C -1.78501000 -2.35466400 -1.38217800 C -1.79911300 0.56471000 -2.66297800 C -1.84308700 2.68187600 -0.28395100 C -1.85350200 1.07537000 2.46568600 H 2.83678800 3.04193300 0.19704700 H 1.30313100 3.20899000 -0.64804300 H 1.33616600 3.08896800 1.11374900 H 2.83177500 0.77396700 2.95313900 H 1.30114300 1.63430200 2.84538500 H 1.32733400 -0.07763100 3.27822800 H 2.87612800 -2.52085400 1.65162600 H 1.33203500 -2.19073200 2.42641800 H 1.39095100 -3.12701300 0.93003600 H 2.91647600 -2.30095600 -1.89250300 H 1.38858700 -2.96937800 -1.33288900 H 1.41846700 -1.83589400 -2.68780700 H 2.88791600 1.10831200 -2.79140900 H 1.35095700 0.37638400 -3.23424900 H 1.39567300 2.01386000 -2.57392500 H -2.90713000 1.15141800 2.75406100 187 H -1.35438400 0.46952000 3.22102700 H -1.43528100 2.08054100 2.50207600 H -2.86900000 -2.29178500 1.96751800 H -1.32830600 -2.94437200 1.42601100 H -1.38040700 -1.77150600 2.74609000 H -2.83054900 -2.61656200 -1.57490400 H -1.28407300 -2.29258100 -2.34765800 H -1.34423900 -3.17721700 -0.81983900 H -2.84691200 0.63515300 -2.97295200 H -1.32696200 1.51802700 -2.89756000 H -1.33254300 -0.20748900 -3.27388300 H -2.89624100 2.98182800 -0.28947700 H -1.36066200 3.21004700 0.53783600 H -1.40417800 3.02844300 -1.21887500 188 SUPPORTING INFORMATION (B) FOR CHAPTER 4 SB.1 NMR Spectra Figure SB.1 1H NMR of Hbid in CDCl3 189 Figure SB.2 1H NMR of Ru(bid)(bpy’)Cl (1) in DMF-d7 190 Figure SB.3 1H NMR of [Ru(bid)(bpy’)NH3][PF6] (2) in MeCN-d3 191 Figure SB.4 1H NMR of [Ru(bid)(bpy’)NH3][PF6]2 (3) in THF-d6 at t = 0 h (vertically zoomed out spectrum) 192 Figure SB.5 1H NMR of [Ru(bid)(bpy’)NH3][PF6]2 (3) in THF-d6 at t = 0 h (vertically zoomed in spectrum) 193 Figure SB.6 1H NMR of [Ru(bid)(bpy’)NH3][PF6]2 (3) in THF-d6 at t = 20 h (vertically zoomed out spectrum) 194 Figure SB.7 1H NMR of [Ru(bid)(bpy’)NH3][PF6]2 (3) in THF-d6 at t = 20 h (vertically zoomed in spectrum) 195 Figure SB.8 1H NMR of [Ru(bid)(bpy’)NH3][PF6]2 (3) in THF-d6 at t = 96 h (vertically zoomed out spectrum) 196 Figure SB.9 1H NMR of [Ru(bid)(bpy’)NH3][PF6]2 (3) in THF-d6 at t = 96 h (vertically zoomed in spectrum) 197 SB.2 Crystallographic Data SB.2.1 Crystal Data and Experimental for Complex 2 Compound Complex 2 CCDC 1993356 Formula C31H29Cl2F6N8O2PRu Dcalc./ g cm-3 1.712 /mm-1 6.413 Formula Weight 862.56 Colour blue Shape plate Size/mm3 0.26×0.13×0.04 T/K 100.00(10) Crystal System triclinic Space Group P-1 a/Å 10.7072(2) b/Å 12.3824(3) c/Å 14.0815(3) /° 107.399(2) Experimental. Single blue-plate crystals of Complex 2 used /° 102.504(2) as received. A suitable crystal with dimensions 0.26 × 0.13 × /° 100.944(2) V/Å3 1673.05(7) 0.04 mm3 was selected and mounted on a nylon loop with Z 2 paratone oil on a XtaLAB Synergy, Dualflex, HyPix Z' 1 diffractometer. The crystal was kept at a steady T = 100.00(10) K Wavelength/Å 1.54184 during data collection. The structure was solved with the ShelXT Radiation type Cu K min/° 3.439 (Sheldrick, G.M. (2015). Acta Cryst. A71, 3-8) solution program max/° 71.029 using dual methods and by using Olex2 (Dolomanov et al., 2009) Measured Refl's. 26574 as the graphical interface. The model was refined with ShelXL Ind't Refl's 6310 (Sheldrick, Acta Cryst. A64 2008, 112-122) using full matrix Refl's with I > 2(I) 5901 Rint 0.0526 least squares minimisation on F2. Parameters 474 Restraints 0 Crystal Data. C31H29Cl2F6N8O2PRu, Mr = 862.56, triclinic, Largest Peak 0.732 P-1 (No. 2), a = 10.7072(2) Å, b = 12.3824(3) Å, c = Deepest Hole -1.171 14.0815(3) Å,  = 107.399(2)°,  = 102.504(2)°,  = 100.944(2)°, GooF 1.065 V = 1673.05(7) Å3, T = 100.00(10) K, Z = 2, Z' = 1, (Cu Ka) = wR2 (all data) 0.1062 wR2 0.1010 6.413, 26574 reflections measured, 6310 unique (Rint = 0.0526) R1 (all data) 0.0390 which were used in all calculations. The final wR2 was 0.1062 R1 0.0360 (all data) and R1 was 0.0360 (I > 2(I)). 198 SB.2.2 Crystal Data and Experimental for Complex 4 Compound Complex 4 Formula C34H36F12N8O3P2Ru CCDC 2101754 Dcalc./ g cm-3 1.708 /mm-1 5.023 Formula Weight 995.72 Color red Shape needle-shaped Size/mm3 0.11×0.06×0.02 T/K 99.9(3) Crystal System monoclinic Flack Parameter -0.011(5) Hooft Parameter -0.014(4) Space Group P21 a/Å 9.35985(18) Experimental. Single red needle-shaped crystals of b/Å 15.3882(4) c/Å 13.4460(3) complex 4 used as received. A suitable crystal with /° 90 dimensions 0.11 × 0.06 × 0.02 mm3 was selected and /° 91.9032(17) mounted on a nylon loop with paratone oil on a XtaLAB /° 90 Synergy, Dualflex, HyPix diffractometer. The crystal was V/Å3 1935.58(7) kept at a steady T = 99.9(3) K during data collection. The Z 2 Z' 1 structure was solved with the ShelXS (Sheldrick, 2008) Wavelength/Å 1.54184 solution program using direct methods and by using Olex2 Radiation type Cu K 1.3 (Dolomanov et al., 2009) as the graphical interface. The min/° 3.289 model was refined with ShelXL 2018/3 (Sheldrick, 2015) max/° 80.105 Measured Refl's. 21661 using full matrix least squares minimisation on F2. Indep't Refl's 7474 Refl's I≥2 s(I) 7185 Crystal Data. C34H36F12N8O3P2Ru, Mr = 995.72, Rint 0.0486 monoclinic, P21 (No. 4), a = 9.35985(18) Å, b = Parameters 544 15.3882(4) Å, c = 13.4460(3) Å,  = 91.9032(17)°,  =  = Restraints 1 90°, V = 1935.58(7) Å3, T = 99.9(3) K, Z = 2, Z' = 1, (Cu Largest Peak 0.438 Deepest Hole -0.765 K) = 5.023, 21661 reflections measured, 7474 unique GooF 1.068 (Rint = 0.0486) which were used in all calculations. The final wR2 (all data) 0.0889 wR2 was 0.0889 (all data) and R1 was 0.0348 (I≥2 s(I)). wR2 0.0880 R1 (all data) 0.0365 R1 0.0348 199 SUPPORTING INFORMATION (C) FOR CHAPTER 5 SC.1 NMR Spectra Figure SC.1 1H NMR of Hbpp in DMSO-d6 200 Figure SC.2 1H NMR of [Ru2II(𝜇-Cl)(bpp)(trpy)2][PF6]2 (1) in acetone-d6 201 Figure SC.3 1H NMR of [Ru2II(𝜇-Cl)(bpp)(trpy)2][PF6]2 (1) in acetone-d6 (aromatic region) 202 Figure SC.4 1H NMR of [Ru2II(NH3)Cl(bpp)(trpy)2][PF6]2 (2) in acetone-d6 203 Figure SC.5 1H NMR of [Ru2II(NH3)Cl(bpp)(trpy)2][PF6]2 (2) in acetone-d6 (aromatic region) 204 Figure SC.6 1H NMR of [Ru2II(NH3)2(bpp)(trpy)2][PF6]3 (3) in acetone-d6 205 Figure SC.7 1H NMR of [Ru2II(NH3)2(bpp)(trpy)2][PF6]3 (3) in acetone-d6 (aromatic region) 206 Figure SC.8 1H NMR of [Ru2II(NH3)2(bpp)(trpy)2][PF6]3 (3) in MeCN-d3 207 Figure SC.9 13C NMR of [Ru2II(NH3)2(bpp)(trpy)2][PF6]3 (3) in acetone-d6 208 Figure SC.10 gCOSY NMR of [Ru2II(NH3)2(bpp)(trpy)2][PF6]3 (3) in acetone-d6 209 Figure SC.11 gHSQC NMR of [Ru2II(NH3)2(bpp)(trpy)2][PF6]3 (3) in acetone-d6 210 Figure SC.12 gHMBC NMR of of [Ru2II(NH3)2(bpp)(trpy)2][PF6]3 (3) in acetone-d6 211 Figure SC.13 DEPT NMR of [Ru2II(NH3)2(bpp)(trpy)2][PF6]3 (3) in acetone-d6 212 Figure SC.14 1H NMR of [Ru2II(µ-N2H4)(bpp)(trpy)2][PF6]3 (4) in acetone-d6 213 Figure SC.15 1H NMR of [Ru2II(µ-N2H4)(bpp)(trpy)2][PF6]3 (4) in MeCN-d3 214 Figure SC.16 1H NMR of [Ru2III(NH3)2(bpp)(trpy)2][PF6]5 (3b) in MeCN-d3 215 Figure SC.17 1H NMR of the products of the reaction of 3b with 2,6–dimethylpyridine in MeCN- d3 216 Figure SC.18 1H NMR of the products of the reaction of 3b with 2,6–dimethylpyridine in MeCN- d3 (aromatic region). Blue dots are the peaks of the unidentified product(s) 217