University of Southern California Dissertations and Theses

Catalysis for solar fuels production: molecular and materials approaches

(USC Thesis Other)

Catalysis for solar fuels production: molecular and materials approaches

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Catalysis for Solar Fuels Production:
Molecular and Materials Approaches

by

Nicholas M. Orchanian

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(CHEMISTRY)

May 2020

ii

“Learn why the world wags and what wags it.
That is the only thing which the mind can never exhaust,
never alienate, never be tortured by, never fear or distrust,
and never dream of regretting.”

-T.H. White, “The Once and Future King”

iii

ACKNOWLEDGEMENTS

The completion of this thesis would not have been possible without the encouragement and
support of a great number of people. I have had the incredible fortune to be surrounded by a
network of wonderful individuals—family and friends, mentors and labmates, classmates and
colleagues—and I wouldn’t be the person I am today without each of them. The completion of a
dissertation is a long and difficult journey, but it has been a pleasure to spend the past several years
with the people I’ve met along the way.
I first want to thank my advisor, Prof. Smaranda C. Marinescu. The freedom and flexibility
that Smaranda offered during my graduate studies was invaluable for my growth as a chemist.
Smaranda allowed me to pursue my scientific interests and to follow my research down unexpected
paths. I set out to develop my skills in synthetic chemistry, and have learned so much more in the
process. I’ve had the opportunity to present my work at conferences, develop new research
directions, interact with collaborators and industry contacts, and learn the interminable art of
proposal writing. I’m very grateful for the opportunities that Smaranda gave me and for her
encouragement and support.
I also want to thank and acknowledge my undergraduate research advisor, Prof. Jahan M.
Dawlaty. Working with Jahan provided me with the confidence to pursue graduate studies, and I
certainly would not have considered a future in research had I not been fortunate enough to explore
my scientific interests as an undergraduate researcher in the Dawlaty Lab. Jahan’s infectious
enthusiasm for his work has been a constant source of inspiration to me, and Jahan has always had
an open door for me when I wanted to discuss my research or career path. I’ve been very lucky to
have his support throughout my undergraduate and graduate careers.
iv

I want to thank my screening and qualifying exam committee members: Prof. Mark E.
Thompson, Prof. Sri R. Narayan, Prof. Ralf Haiges, Prof. Jongseung Yoon, and Prof. Moh El-
Naggar. My experience in Mark’s undergraduate “Advanced Inorganic Chemistry” course is what
excited me to pursue inorganic chemistry in graduate school, and my framework for understanding
coordination chemistry was built in his course. The opportunity to discuss my research with Sri
was a great pleasure, and learning the nearly-wizardly craft of crystallography from Ralf was
essential to my work. Thank you to Prof. Brent C. Melot for our collaborative work and for your
courses in materials chemistry. As my graduate studies were built on the foundations that I
developed as an undergraduate, I also want to thank all the faculty at USC who encouraged and
inspired me during that time. I’d like to thank Prof. Richard L. Brutchey in particular, who sparked
my interest in nanoscience when I was an undergraduate student in his course, and who supported
my interest in pursuing graduate studies. I also would like to thank Prof. Michael Inkpen and Prof.
Megan Fieser, who have taken the time while building their own labs to help me land my dream
job.
A special thank you to the members of the Marinescu Group, current and former, without
whom this work would not be possible. Thank you to Dr. Courtney Downes for training me as a
budding first-year graduate student and for showing us the path to a successful PhD. Thank you to
Dr. Andrew Clough for spending many hours training and helping me with XPS, AFM, and SEM
studies. Thank you to Dr. Alon Chapovetsky for showing us how to conduct labwork with élan
and encouraging me to pursue after-work beers. Thank you to Dr. Damir Popov for gifting me a
ligand that enabled many of my studies. Thank you to Geovanni Rangel for being a fantastic safety
officer and helping our group run smoothly. Thank you to my former undergraduate researchers,
v

Jacques Esterhuizen and Jack Skrainka, for helping me with the laborious ground work on which
I built my studies.
Of the current group, I first want to thank my undergraduate researcher Lorena Hong for
her incredible work, which allowed me to broaden our scope to multiple projects. I want to thank
Eric Johnson for his friendship and constant devotion to this group—without Eric, the equipment
that enables our research would all come to a screeching halt. I want to thank Ashely Hellman for
being a great friend throughout graduate school and providing feedback on my often addled
scientific writing. I want to thank Keying Chen for providing me with the unparalleled opportunity
to run t-BuLi reactions, for providing the lab with a constant source of olfactory bliss in the form
of organoselenium chemistry, and for maintaining our electrochemistry instrumentation which
enabled my research. I want to thank Jeremy Intrator for supporting our group as safety officer and
keeping our lab running safely, and for single-handedly fighting the brave fight against uncapped
sharps. I want to thank Dr. Jeffrey J. Liu for bringing his knowledge and skills to our group as a
postdoctoral researcher; we have all learned a lot from Jeff and he has been a wonderful addition
to the Marinescu team. Finally, I want to thank our newest recruits: Adam Samuel and David
Velazquez. I wish you both the best in your studies, and a particular thanks to David for keeping
my projects alive after I leave USC.
I have a great number of friends to thank for bringing me immense joy over the past many
years. From new friends to old, the relationships I’ve built over the years have made my journey a
happy one. I want to thank my friends from Waltham who have supported me from middle school
to now. Thank you to the self-titled “Hometown Hunnies”—Katie McGuire, Maureen O’Meara,
Gianna Scaltreto, and Lauren Tierney—as well as Andrea Lewis and Nadja Kern, for making
vi

Middle School and High School tolerable. While I’m sad to leave my LA friends, I am excited to
be closer to these amazing people on the East Coast.
I want to thank all the wonderful friends I’ve made during my near-decade at USC,
although there are far too many to name. In a somewhat-chronological order of our meeting: Thank
you to my very first friend at USC, Bonamico Jacobs, for being a loyal and constant friend, and to
all the boys of Phi Delta Theta who made my undergraduate experience a truly special one. I
couldn’t imagine a better group of people to spend my college days with. I want to thank Amanda
Levy for our many years of friendship, my life is brighter with you in it. Thank you to Sheridan
Watson and Peter Nordahl-Hansen for being amazing friends to me and welcoming me into their
lives. Thank you to Sydney Goldman for being an extraordinary big sis, opera pal, and lifelong
friend. Thank you to Theresa Huffman for your kindness and friendship, and for housing me during
my many trips to New York. Thank you to Daniel LeBlanc for being a wonderful friend, and for
spending many evenings of discourse with me at Saddles. Thank you to Jack Wangerin, Sarah
Ostresh, Jameson Burke, and Mariel Stolarski for keeping my life full of surprises and for our
many adventures, and I hope for many more in the future. Thank you to Dimiana Saad for
preparing a social calendar for me which I couldn’t possibly keep up with. Thank you to the many
remarkable people, including Mattie Hanson, Molly Smith, Adam Silverman, and countless more,
who have been great friends to me throughout the years. I also want to thank all the new friends
I’ve met during graduate school. Thank you to Abegail Tadle, Caroline Black, Savannah Kapper,
JoAnna Milam, Robert Pankow, Sahar Roshandel, Ryan Hunt, Sara Smock, Sean O’Connell, and
countless others who have trudged through the mire and muck of graduate school with me.
I owe a great thank-you to my entire extended family. Thank you first to my great-aunt
Beatrice Orchanian and my grandmother Haigouhie Orchanian for their boundless love and for the
vii

tremendous time and care they have taken to make me happy. Thank you to my grandfather Marty
for encouraging my interest in science. Thank you to my many aunts and uncles on the East coast
for their love and support: Zohrab and Andrea Orchanian, Azniv and Jack Nigoghosian, Hagop
and Nora Orchanian, Rhoda and Mark Belemjian, Peter and Melissa Manoogian, Mark and Leesa
Manoogian, and many more. Thank you to all my cousins who made my childhood anything but
quiet, although, as hinted by the previous list, there are far too many to name. A particular thank
you is owed to my cousin Shant Orchanian for taking any opportunity to visit me in LA. Thank
you to all my family in California for making my move here feel effortless. Thank you especially
to my aunt Suzy Orchanian for her love and support, and for helping me move into my freshman
dorm in the hot LA summer with no elevator (and the many moves after that, and painting my
apartment, and offering me a place to stay in San Diego, and hosting Thanksgiving dinner, among
countless other things). Thank you to Garine and Hagop Abajian and to Angel and Greg for always
checking in on me and for making LA a second home. Thank you to all my amazing aunts, uncles,
and cousins in LA, of whom there are, again, too many to name.
Above all, I want to thank my parents and sister: Zareh, Lori, and Stephanie Orchanian. I
am very lucky to have such remarkable parents who have supported and loved me at every step of
my journey. Their relentless optimism, tireless work ethic, and unwavering love have been an
unending source of inspiration to me throughout my life. I want to thank my sister, Stephanie
Orchanian, for being my constant friend in life. I would not be here today without these three
amazing people, and couldn’t imagine my life with anyone else. I love you and thank you for
everything you’ve done for me.

viii

TABLE OF CONTENTS

Epigraph ......................................................................................................................................... ii
Acknowledgements ......................................................................................................................... iii
Table of Contents ......................................................................................................................... viii
List of Figures ................................................................................................................................ xi
List of Schemes ............................................................................................................................. xvi
List of Tables ................................................................................................................................ xvi
Chapter 1. General Introduction ...................................................................................................1
1.1 Global Energy Outlook ..................................................................................................2
1.2 Carbon Emissions and Renewable Electricity ...............................................................3
1.3 Solar Fuels .....................................................................................................................5
1.4 Hydrogen Evolution Reaction ........................................................................................6
1.5 Carbon Dioxide Reduction Reaction .............................................................................8
1.6 Catalyst Design ............................................................................................................10
1.7 Heterogeneous and Homogeneous Catalysts ...............................................................13
1.8 Cobalt Dithiolene Catalysts for Hydrogen Evolution ..................................................15
1.9 Rhenium Bipyridine Catalysts for Carbon Dioxide Reduction ...................................17
1.10 Prospects in Molecular and Materials Systems for Solar Fuels .................................18
1.11 References ..................................................................................................................22
Chapter 2. Electrocatalytic Syngas Generation with a Redox Non-Innocent Cobalt 2-
Phosphinobenzenethiolate Complex..................................................................................................................................... 28
2.1 Abstract ........................................................................................................................29
2.2 Introduction ..................................................................................................................29
2.3 Results and Discussion ................................................................................................32
2.4 Conclusions ..................................................................................................................56
2.5 Experimental Methods .................................................................................................57
2.5.1 Materials and Synthesis ................................................................................57
2.5.2 Synthesis of Co(PiPr2SC6H4)2 (CoPS) ..........................................................58
2.5.3 X-ray Crystallography .................................................................................59
2.5.4 Electrochemistry ...........................................................................................60
2.5.5 Nuclear Magnetic Resonance .......................................................................60
2.5.6 Electrochemical Analysis..............................................................................61
2.5.7 Gas Chromatography ....................................................................................62
2.5.8 Product Detection..........................................................................................62
2.5.9 UV-Vis Spectroscopy ...................................................................................63
2.5.10 Computational Methods ..............................................................................63
2.5.11 Nuclear Coordinates for Optimized Geometries .........................................64
2.6 References ....................................................................................................................69
Chapter 3. Influence of Transition Metal Ion on Electrocatalytic CO2 Reduction by
Aminopyridine Macrocycles ....................................................................................................................................................... 74
3.1 Abstract ........................................................................................................................75
3.2 Introduction ..................................................................................................................66
ix

3.3 Results and Discussion ................................................................................................77
3.4 Conclusions ................................................................................................................107
3.5 Experimental Methods ...............................................................................................108
3.5.1 Materials and Synthesis .............................................................................108
3.5.2 Synthesis of Complexes ..............................................................................109
3.5.3 X-ray Crystallography ...............................................................................110
3.5.4 Electrochemistry .........................................................................................110
3.5.5 Nuclear Magnetic Resonance Spectroscopy ...............................................111
3.5.6 Gas Chromatography ..................................................................................112
3.5.7 X-ray Photoelectron Spectroscopy .............................................................112
3.5.8 UV-Vis and FTIR Spectroscopy .................................................................113
3.5.9 Computational Methods ..............................................................................113
3.5.10 Nuclear Coordinates for Optimized Geometries .......................................114
3.6 References ..................................................................................................................118
Chapter 4. Surface-Immobilized Conjugated Polymers Incorporating Rhenium
Bipyridine Motifs for Electrocatalytic and Photocatalytic CO2 Reduction .........................121
4.1 Abstract ......................................................................................................................122
4.2 Introduction ................................................................................................................122
4.3 Results and Discussion ..............................................................................................127
4.3.1 Synthesis of [2,2'-bipyridine]-5,5'-bis(diazonium) rhenium complex ........127
4.3.2 Electropolymerization of 2 .........................................................................132
4.3.3 Characterization of Electropolymerized Films ...........................................136
4.3.3.1 XPS ..............................................................................................136
4.3.3.2 SEM and AFM .............................................................................140
4.3.3.3 IRRAS and UV-Vis .....................................................................145
4.3.3.4 Electrochemistry of Films ............................................................149
4.3.4 Catalytic Studies .........................................................................................157
4.3.4.1 Cyclic Voltammetry .....................................................................157
4.3.4.2 Controlled Potential Electrolysis .................................................159
4.3.4.3 Photocatalysis ..............................................................................162
4.3.5 Post-Catalysis Electrode Characterization ..................................................166
4.4 Conclusions ................................................................................................................168
4.5 Experimental Methods ...............................................................................................169
4.5.1 Materials and Synthesis ..............................................................................169
4.5.2 Synthesis of 2 ..............................................................................................170
4.5.3 Electrochemistry ........................................................................................170
4.5.4 Electrochemical analysis .............................................................................171
4.5.5 Gas Chromatography ..................................................................................173
4.5.6 Product Detection........................................................................................173
4.5.7 UV-Vis Spectroscopy .................................................................................174
x

4.5.8 SEM ............................................................................................................174
4.5.9 Atomic Force Microscopy ..........................................................................174
4.5.10 ICP-OES ...................................................................................................175
4.5.11 Photocatalytic Studies ...............................................................................175
4.5.12 Computational Methods ............................................................................175
4.6 References ..................................................................................................................177
Chapter 5. Immobilized Molecular Wires on Carbon Cloth Electrodes facilitate CO2
Electrolysis ..................................................................................................................................184
5.1 Abstract ......................................................................................................................185
5.2 Introduction ................................................................................................................185
5.3 Results and Discussion ..............................................................................................187
5.4 Conclusions ................................................................................................................197
5.5 Experimental Methods ...............................................................................................198
5.5.1 Materials and Synthesis ..............................................................................198
5.5.2 Synthesis of diamino bipyridine complex...................................................198
5.5.3 Synthesis of bis(diazonium) monomer .......................................................199
5.5.4 Modification of CCE Samples ....................................................................199
5.5.5 Electrochemistry .........................................................................................200
5.5.6 Electrochemical Analysis............................................................................201
5.5.7 XPS .............................................................................................................202
5.5.8 High Resolution Scanning Electron Microscopy ........................................203
5.5.9 ICP-OES .....................................................................................................203
5.6 References ..................................................................................................................204
Bibliography ................................................................................................................................207

xi

LIST OF FIGURES

Figure 1.1. End-use energy consumption by sector and electricity consumption by source .........3
Figure 1.2. Carbon dioxide concentration at Mauna Loa Observatory ..........................................4

Figure 1.3. Schematic representation of a grid architecture which incorporates catalysts for the
production of solar fuels. .....................................................................................................6
Figure 1.4. Schematic depiction of an electrolysis cell for water splitting ....................................8
Figure 1.5. Abundance of the elements in Earth’s crust ..............................................................12
Figure 1.6. The active site structure of [NiFe] CO-dehydrogenase and the proposed
mechanism of reactivity .....................................................................................................13
Figure 1.7. Proposed mechanism for hydrogen evolution with the [Co(bdt)2]
–
catalyst ..............17
Figure 1.8. Proposed mechanistic scheme for carbon dioxide reduction with the
[Re(bpy)(CO)3Cl] catalyst ................................................................................................18
Figure 2.1. Chemdraw and solid-state structure of CoPS .............................................................32
Figure 2.2.
1
H NMR spectrum of 1 mM CoPS in a solution of benzene-d6 ................................34
Figure 2.3. Calculated spin density distribution of CoPS ............................................................34
Figure 2.4. UV-Vis of CoPs and spin localization plot ................................................................35
Figure 2.5. Measured absorbance for observed electronic transitions as a function of CoPS
concentration as determined by UV-Vis spectroscopy. .....................................................36
Figure 2.6. UV-Vis study of 1 mM CoPS in acetonitrile solution ................................................37
Figure 2.7. Molecular orbital diagram for CoPS ..........................................................................38
Figure 2.8. Cyclic voltammograms of 1 mM CoPS in an acetonitrile solution under N2
atmosphere .........................................................................................................................40
Figure 2.9. Cyclic voltammetry of CoPS (1 mM) in acetonitrile with 0.1 M [nBu4N][PF6]
supporting electrolyte (scan rate = 100 mV/s) from 0 V vs Fc/Fc
+
to
(a) -2.3 V vs Fc/Fc
+
and (b) -2.7 V vs Fc/Fc
+
. ...................................................................40
Figure 2.10. Cyclic voltammograms of 1 mM CoPS in an acetonitrile solution under N2
atmosphere at various scan rates ........................................................................................42
Figure 2.11. Randles-Sevcik analysis based on variable scan rate experiments. .........................42
Figure 2.12. Calculated geometries for CoPS and reduced states ................................................43
Figure 2.13. Calculated spin densities for CoPS and reduced states ............................................43
Figure 2.14. Cyclic voltammograms of 1 mM CoPS in an acetonitrile solution under CO2
atmosphere with no acid, phenol, 2,2,2-trifluoroethanol, or water ...................................44
Figure 2.15. Cyclic voltammograms of 1 mM CoPS in an acetonitrile solution under N2 and
CO2 atmosphere ................................................................................................................45
Figure 2.16. Cyclic voltammograms of 1 mM CoPS in an acetonitrile solution with 0.1 M
[nBu4N][PF6] electrolyte with scan rate = 250 mV/s under N2 atmosphere and
under CO2 with 0.5 M H2O (blue)......................................................................................45
Figure 2.17. Cyclic voltammetry of 1mM CoPS in acetonitrile under CO2 atmosphere
with 0.1 M [nBu4N][PF6] supporting electrolyte at various scan rates .............................46
Figure 2.18. Controlled potential electrolysis traces for CoPS studies under CO2 ......................47
xii

Figure 2.19. Summary of CPE results illustrating Faradaic efficiencies for H2 and CO
production. .........................................................................................................................48
Figure 2.20. TONtotal presented as a function of the pKa of the proton source. ............................48
Figure 2.21. Summary of H2:CO ratio calculations from CPE experiments ...............................49
Figure 2.22. Controlled potential electrolysis traces for CoPS studies under N2 ........................51
Figure 2.23. Rinse test traces measured under catalytic conditions. .............................................52
Figure 2.24. Cyclic voltammetry of a glassy carbon working electrode in acetonitrile
with 0.1 M [nBu4N][PF6] supporting electrolyte (scan rate = 100 mV/s) in
the presence of various acid additives at a concentration of 0.5 M ...................................53
Figure 2.25. UV-Vis study of 1 mM CoPS in acetonitrile solution, protonation study
with 3 eq trifluoroethanol, and subsequent deprotonation study
with excess triethylamine ..................................................................................................54
Figure 2.26.
1
H-NMR spectra of 1 mM CoPS in acetonitrile-d3 without acid additive,
with 0.5 M PhOH, with 0.5 M TFE, and with 0.5 M H2O. ...............................................54
Figure 3.1. Solid-state structures of FeL
1
and NiL
1
, and calculated geometries .........................79
Figure 3.2. Calculated geometries for the ML
1
(M = Fe, Co, Ni) series .....................................81
Figure 3.3. Transmittance FTIR spectrum of FeL
1
.......................................................................83
Figure 3.4. Transmittance FTIR spectrum of CoL
1
......................................................................83
Figure 3.5. Transmittance FTIR spectrum of NiL
1
.......................................................................84
Figure 3.6. Calculated N-H stretching frequencies ......................................................................84
Figure 3.7. UV-Vis studies for the ML
1
series ............................................................................85
Figure 3.8. Absorbance values at varying concentrations of FeL
1
in DMF ..................................85
Figure 3.9. Absorbance values at varying concentrations of FeL
1
in DMF .................................86
Figure 3.10. Absorbance values at varying concentrations of FeL
1
in DMF ...............................86
Figure 3.11. Predicted UV-Vis transitions based on TD-DFT calculations ................................87
Figure 3.12. Frontier molecular orbital diagram for the ML
1
series ............................................88
Figure 3.13. Molecular orbital images for the ML
1
series. ..........................................................89
Figure 3.14. Spin density localization plots for CoL
1
and NiL
1
..................................................89
Figure 3.15. Electrochemical studies of the FeL
1
and NiL
1
complexes under N2 and CO2
atmosphere in DMF with a scan rate of 100 mV/s ............................................................93
Figure 3.16. Cyclic voltammetry of NiL
1
at various scan rates ...................................................94
Figure 3.17. Plot showing the peak current densities for the NiL
1
reductions as a
function of the scan rate .....................................................................................................94
Figure 3.18. Cyclic voltammetry of FeL
1
in DMF at a scan rate of 100 mV/s .............................95
Figure 3.19. Cyclic voltammetry of FeL
1
at various scan rates ...................................................95
Figure 3.20. Plot showing the peak current densities for the NiL
1
reductions as a
function of the scan rate ....................................................................................................96
Figure 3.21. Cyclic voltammetry of FeL
1
under CO2 atmosphere ...............................................97
Figure 3.22. Cyclic voltammetry of NiL
1
under CO2 atmosphere ...............................................97
Figure 3.23. Sequential cyclic voltammetry scans of FeL
1
under N2 atmosphere ........................98

xiii

Figure 3.24. Sequential cyclic voltammetry scans of NiL
1
under N2 atmosphere .......................98
Figure 3.25. Titration studies of the various complexes with 2,2,2-trifluoroethanol in DMF ......99
Figure 3.26. Plots of the estimated rate constants kobs vs. [TFE] in DMF for FeL
1
...................100
Figure 3.27. Plots of the estimated rate constants kobs vs. [TFE] in DMF for NiL
1
...................100
Figure 3.28. Controlled potential electrolysis of 0.5 mM FeL
1
performed at -2.68 V ..............101
Figure 3.29. Controlled potential electrolysis of 0.5 mM NiL
1
performed at -2.90 V ..............101
Figure 3.30. Controlled potential electrolysis of 0.5 mM FeL
1
performed at -2.40 V ..............102
Figure 3.31. SEM images for FeL
1
studies .................................................................................103
Figure 3.32. SEM images for NiL
1
studies ................................................................................104
Figure 3.33. SEM images for CoL
1
studies ...............................................................................104
Figure 3.34. XPS studies for post-CPE electrodes......................................................................104
Figure 3.35. Controlled potential electrolysis of a FeL
1
-GCE performed at -2.4 V ..................106
Figure 3.36. Controlled potential electrolysis of a NiL
1
-GCE performed at -2.9 V ..................106
Figure 4.1.
1
H NMR spectrum of 1 ............................................................................................129
Figure 4.2.
1
H NMR spectrum of 2 ............................................................................................129
Figure 4.3.
19
F NMR spectrum of 2 ............................................................................................129
Figure 4.4. ATR FTIR spectrum of 2..........................................................................................129
Figure 4.5. Calculated vibrational spectrum of 2 ........................................................................130
Figure 4.6. UV-Vis spectrum of 2...............................................................................................130
Figure 4.7. XPS Survey spectrum of 2........................................................................................130
Figure 4.8. High-resolution XPS of the Re 4f region for complex 2 ..........................................131
Figure 4.9. High-resolution XPS of the Cl 2p region for complex 2 ..........................................131
Figure 4.10. High-resolution XPS of the N 1s region for complex 2 .........................................131
Figure 4.11. High-resolution XPS of the B 1s region for complex 2 .........................................132
Figure 4.12. High-resolution XPS of the F 1s region for complex 2 ..........................................132
Figure 4.13. Cyclic voltammograms for electropolymerization of 2 on glassy carbon ..............133
Figure 4.14. Cyclic voltammograms for electropolymerization of 2 for nine cycles .................133
Figure 4.15. Cyclic voltammograms for electropolymerization of 2 on FTO ............................134
Figure 4.16. Cyclic voltammograms for electropolymerization of 2 on FTO for
ten cycles with switching potential of -2.5 V ..................................................................134
Figure 4.17. Cyclic voltammograms for electropolymerization of 2 on FTO for
ten cycles with switching potential of -2 V .....................................................................134
Figure 4.18. High-resolution Re 4f XPS spectrum for a modified FTO electrode (n=20).
SEM image of a modified FTO electrode (n = 20) at 25,000x magnification. AFM
topology of modified FTO electrode (n = 20). ...............................................................135
Figure 4.19. XPS survey scan for modified FTO electrodes. .....................................................138
Figure 4.20. High-resolution XPS of the Re 4f region for modified FTO electrodes ................138
Figure 4.21. High-resolution XPS of the Sn 2p region for modified FTO electrodes,
with n = 1 and n = 20. ......................................................................................................139
Figure 4.22. High-resolution XPS of the Cl 2p region for a modified
FTO electrode (n = 20) ....................................................................................................139
Figure 4.23. High-resolution XPS of the P 2p region for a modified FTO electrode (n = 20). ..139
xiv

Figure 4.24. High-resolution XPS of the F 1s region for a modified FTO electrode (n = 20) ...140
Figure 4.25. High-resolution XPS of the C 1s region for a modified FTO electrode (n = 20). ..140
Figure 4.26. SEM images of FTO films with varying thickness ................................................142
Figure 4.27. SEM images of interfacial regions for FTO films ..................................................143
Figure 4.28. AFM data collected for FTO films with varying thickness ....................................144
Figure 4.29 3D Projections of AFM data ...................................................................................145
Figure 4.30. PM-IRRAS results measured for a modified Au substrate (n = 10)
under both p- and s- polarization, with corresponding carbonyl stretching vectors.
UV-Vis absorption spectrum measured for a modified FTO substrate (n = 10). ............146
Figure 4.31. PM-IRRAS of a modified FTO electrode (n = 20) with s- and p-polarization. .....148
Figure 4.32. PM-IRRAS of modified FTO electrodes with p polarization and PM-IRRAS
peak height for the carbonyl stretching modes as a function of the
electroactive surface coverage .........................................................................................148
Figure 4.33. UV-Vis spectra of FTO films with varying thickness. ...........................................149
Figure 4.34. Cyclic voltammetry of electrodes modified with varying numbers of
grafting scans in acetonitrile solutions with 0.1 M TBAPF6 electrolyte. Catalyst
loading as a function of the number of grafting scans applied as determined by
cyclic voltammetry and ICP-OES. Double-layer capacitance (Cdl) measurement
for a modified electrode (n = 5). Cdl of deposited films as a function of the
number of grafting scans ..................................................................................................150
Figure 4.35. Cyclic voltammetry of a modified glassy carbon electrode (n = 5) in
acetonitrile with 0.1 M TBAPF6 supporting electrolyte .................................................151
Figure 4.36. Cyclic voltammetry of a modified glassy carbon electrode (n = 1) in
acetonitrile with 0.1 M TBAPF6 supporting electrolyte .................................................151
Figure 4.37. Cyclic voltammetry of a modified glassy carbon electrode (n = 10) in
acetonitrile with 0.1 M TBAPF6 supporting electrolyte ..................................................152
Figure 4.38. Cyclic voltammetry of a modified glassy carbon electrode (n = 20) in
acetonitrile with 0.1 M TBAPF6 supporting electrolyte. .................................................152
Figure 4.39. Three sequential cyclic voltammetry scans of a modified FTO
electrode (n = 5), and a fourth scan beginning with an anodic sweep ............................153
Figure 4.40. Images of modified glassy carbon and gold electrodes ..........................................154
Figure 4.41. Double-layer charging current density at the open-circuit potential for
a modified glassy carbon electrode (n = 1) as a function of scan rate. ............................155
Figure 4.42. Double-layer charging current density at the open-circuit potential for
a modified glassy carbon electrode (n = 5) as a function of scan rate. ............................155
Figure 4.43. Double-layer charging current density at the open-circuit potential for
a modified glassy carbon electrode (n = 10) as a function of scan rate. ..........................155
Figure 4.44. Double-layer charging current density at the open-circuit potential for
a modified glassy carbon electrode (n = 20) as a function of scan rate. ..........................156
Figure 4.45. Cyclic voltammograms of modified electrodes in acetonitrile solution
under inert atmosphere and 1 atm of CO2. Cyclic voltammograms of
modified electrodes in acetonitrile solution under 1 atm of CO2 with 0.5 M
trifluoroethanol added ......................................................................................................158

xv

Figure 4.46. Foot-of-the-wave analysis for modified graphite rod electrode and modified
Nafion-MWCNT electrode ..............................................................................................158
Figure 4.47. Cyclic voltammograms in acetonitrile solutions under N2 and CO2
atmosphere with modified graphite rod electrode and modified Nafion-MWCNT ........159
Figure 4.48. Catalyst loading for modified graphite rod electrodes as determined by cyclic
voltammetry and ICP .......................................................................................................160
Figure 4.49. Results of controlled potential electrolysis experiments with modified
graphite rod electrodes .....................................................................................................161
Figure 4.50. Catalyst loading for modified TiO2 electrodes as determined by
cyclic voltammetry and ICP .............................................................................................163
Figure 4.51. TON as a function of catalyst loading for photocatalytic studies with
modified TiO2 electrodes. ................................................................................................165
Figure 4.52. High-resolution XPS of the Re 4f region for a modified FTO substrate
before and after a 1 hour controlled potential electrolysis experiment at -2.25 V in
acetonitrile solution with 0.1 M TBAPF6 supporting electrolyte under saturated
CO2 atmosphere. ..............................................................................................................167
Figure 4.53. IRRAS studies of a modified FTO substrate before and after a 1 hour
controlled potential electrolysis experiment at -2.25 V in acetonitrile solution
with 0.1 M TBAPF6 supporting electrolyte under saturated
CO2 atmosphere. ..............................................................................................................168
Figure 4.54. UV-Vis studies of a modified FTO substrate before and after a 1 hour
controlled potential electrolysis experiment at -2.25 V in acetonitrile solution
with 0.1 M TBAPF6 supporting electrolyte under saturated CO2 atmosphere ................168
Figure 5.1. High-resolution scanning electron microscopy (HR-SEM) imaging of
the carbon cloth electrode and modification procedure via
directed electropolymerization ........................................................................................187
Figure 5.2. Cyclic voltammetry of Re-CCE in acetonitrile solvent under
N2 atmosphere and CO2 atmosphere ................................................................................188

Figure 5.3. Randles-Sevcik analysis of Re-CCE, indicating a linear
dependency of J (mA/cm
2
) on the scan rate.. ...................................................................189
Figure 5.4. Controlled potential electrolysis study with Re-CCE under H-cell conditions ........190
Figure 5.5. High-resolution Re 4f and Cl 2p X-ray photoelectron spectroscopy
of Re-CCE before and after two hours of electrolysis. ....................................................192
Figure 5.6. Kinetic isotope effect (KIE) study of Re-CCE ........................................................194
Figure 5.7. Catalytic Tafel plots collected in under 1 atmosphere CO2 in
acetonitrile solvent ...........................................................................................................194

xvi

LIST OF SCHEMES

Scheme 2.1. Mechanistic hypothesis for observed reactivity for CoPS. .......................................53
Scheme 3.1. Synthesis of complexes .............................................................................................78
Scheme 3.2. Frontier molecular orbitals for the [ML
1
] series ......................................................83
Scheme 4.1. Syntheses of complex 2 and electropolymerization methodology .........................127
Scheme 4.2. Device architecture and energy diagram for photocatalytic CO2 reduction ...........152
Scheme 5.1. Proposed mechanism of catalysis for Re-CCE ......................................................197

LIST OF TABLES

Table 2.1. X-ray crystallography data and structure refinement for CoPS ...................................33
Table 2.2. Selected bond lengths (Å) for CoPS ............................................................................33
Table 2.3. Selected bond angles (°) for CoPS ...............................................................................34
Table 2.4. Calculation of molar extinction coefficients for CoPS based on
dilution studies by UV-Vis spectroscopy ..........................................................................36
Table 2.5. Calculated oscillator strengths and wavelengths of electronic
transitions by TD-DFT and comparison to experimentally-determined
transition wavelengths .......................................................................................................39
Table 2.6. Summary of results from CPE studies. ........................................................................47
Table 2.7. Summary of CPE results under CO2 ............................................................................47
Table 2.8. Summary of FE calculations from CPE experiments under CO2 ................................48
Table 2.9. Summary of CPE results under N2 ...............................................................................52
Table 2.10. Summary of FE calculations from CPE experiments under N2 .................................52
Table 2.11. Thermodynamic potential for CO2 reduction to CO in the presence of
various proton sources, and overpotential associated with CoPS. .....................................55
Table 2.12. Nuclear coordinates for [CoPS]
0
................................................................................64
Table 2.13. Nuclear coordinates for [CoPS]
-
................................................................................65
Table 2.14. Nuclear coordinates for [CoPS]
2-
...............................................................................66
Table 3.1. X-ray crystallography data and structure refinement for FeL
1
....................................79
Table 3.2. X-ray crystallography data and structure refinement for NiL
1
....................................80
Table 3.3. Selected calculated bond distances for FeL
1
and NiL
1
based on DFT geometry
optimizations. .....................................................................................................................81
Table 3.4. Summary of electrochemical behavior of the ML
1
complexes ....................................93
Table 3.5. Summary of the electrocatalytic behavior of the ML
1
complexes ............................102
Table 3.6. Optimized structure for FeL
1
....................................................................................114
Table 3.7. Optimized structure for CoL
1
....................................................................................115
Table 3.8. Optimized structure for NiL
1
....................................................................................116
Table 4.1. RMS surface roughness of modified FTO electrodes. ...............................................145
Table 4.2. Catalyst loading for FTO electrodes ..........................................................................154
Table 4.3. Summary of controlled potential electrolysis studies ...............................................161
xvii

Table 4.4. Summary of photocatalytic studies ...........................................................................165
Table 5.1. Summary of results from controlled-potential electrolysis (CPE) studies. ...............191
Table 5.2. Calculation of catalytic parameters (TON, TOF) based on ICP loading ..................191
Table 5.3. Comparison to selected literature examples ..............................................................195

1

CHAPTER 1

General Introduction

2

1.1 GLOBAL ENERGY PROJECTIONS
Global energy demand is projected to increase as a function of growth in both global
standards of living and population.
[1]
In particular, population increases and dramatic increases in
industrialization in Asian and African nations is predicted to dominate this growing demand for
energy.
[1-2]
Non-OECD Asian nations, including China and India, are projected to account for
greater than half of the global increase in energy demand and are predicted to nearly double their
2018 energy consumption by the year 2050.
[3]
While energy consumption in Africa is projected to
increase by 110% and energy consumption in the Middle East is projected to increase by 55%
from 2018 to 2050, China and India remain the dominant contributors to growth in energy
consumption. The current global demand for energy is projected at ~635 quadrillion British
thermal units (quad) for 2020, and estimates indicate that this may grow to ~900 quad by the year
2050. This growth in demand is predicted to be predominantly driven by growth in the industrial
sector, which encompasses mining and refining operations, manufacturing and construction, as
well as agriculture, though increases in end-use energy consumption are projected for the
transportation, residential, and commercial sectors as well. Energy use in the industrial sector alone
is predicted to increase by over 30% between 2018 and 2050. This growth is localized primarily
in non-OECD countries, with a major shift in manufacturing operations from China to India taking
place over this time period.

3

Figure 1.1 End-use energy consumption by sector (left) and electricity consumption by source
(right). Reprinted with permission from Reference 1.

1.2 CARBON EMISSIONS AND RENEWABLE ELECTRICITY
Historically, U.S. energy consumption has been largely satisfied by fossil fuel sources,
particularly coal and oil.
[4]
Recently, natural gas has emerged as a dominant fuel, outpacing growth
in coal and petroleum.
[5]
Each of these fuels leads to the emission of carbon dioxide into Earth’s
atmosphere, resulting in a notable increase in atmospheric CO2 concentrations through the
industrial era. Measurements at Mauna Loa Observatory have revealed this increasing trend in
CO2 concentration, which has since been termed the “Keeling Curve.”
[6]
The correlation between
CO2 concentration and global average temperature increase has raised concerns regarding the
future consumption of fossil fuels to satisfy increasing global energy demand.
[7]
As such, there is
a growing demand to explore renewable and carbon-neutral sources of energy to power global
development.
[8]

4

Figure 1.2 Carbon dioxide concentration at Mauna Loa Observatory. Reprinted with permission
from Reference 6.

In comparison to fossil fuel sources, solar and wind energy result in reduced CO2 emissions
and could enable distributed energy generation.
[9]
Distributed production of electricity would
subsequently enable a departure from large, centralized grid architectures towards microgrid
architectures.
[10,11]
Microgrids could be of particular benefit to regions with no pre-established grid
and low political stability, where the infrastructural cost and instability of a centralized power
system hinder the availability of reliable electricity.
[12]
However, penetration of renewables into
established grid infrastructures has not yet been realized on a global scale. While fossil fuels
provide high energy density, are conveniently transported, and can be burned on-demand to
manage demand-side variability, solar and wind energy lead to a supply-side variability problem.
The variability of sunlight results in efficient electricity generation during the daylight hours, but
results in a supply-side shortage during hours of peak demand in the early morning and evening.
In order to facilitate grid-scale renewable energy penetration, it will be necessary to incorporate
energy storage technologies into grid architectures. This would allow for excess solar- or wind-
derived electricity to be stored during times of high production and low demand, and to be
consumed during times of low production and high demand. Traditionally, pumped hydroelectric
storage has served as the dominant technology for large-scale energy storage, but increasing
5

attention is being given to the generation of chemical fuels with renewable electricity. In analogy
to the natural process of photosynthesis, whereby solar energy is stored in the form of chemical
bonds, the chemicals generated from renewable electricity through small molecule transformations
have been termed “solar fuels.”
[13]

1.3 SOLAR FUELS
The term “solar fuels” encompasses a broad range of chemicals that can be produced
through the transformation of abundant small molecules using energy derived from solar
electricity.
[13-15]
As CO2 concentrations have increased dramatically with fossil fuel combustion,
the transformation of this waste product into value-added chemicals offers a promising pathway
both to enable efficient storage of solar energy and to mitigate the environmental impact of fossil
fuel consumption by closing the carbon cycle.
[16]
A generalized scheme for the incorporation of
solar fuels into a grid architecture is depicted below in Figure 1.3.
[17]
Initially, electricity is
generated either at photovoltaic arrays, wind turbines, or other renewable sources. If demand is
high, this electricity can be directed to consumers for immediate use. If demand is low, this
electricity can be redirected to catalytic platforms to drive chemical transformations. These
catalytic platforms can involve electrocatalysts, which are activated directly by electrical current
and facilitate multi-proton and multi-electron transfer to abundant substrates, like CO2, H2O, or
N2, to generating value-added products, including CO, methanol, hydrogen, and ammonia. These
platforms can also incorporate photoelectrocatalysts, whereby the collection of solar irradiation
further reduces the energetic cost of these chemical transformations.
[18]
Following production,
these fuels can be stored for future electricity generation during peak-demand hours, or distributed
to the industrial sector for consumption as feedstock chemicals to produce higher-value chemicals
and materials. As the cost for renewable energy generation has decreased dramatically, the barrier
6

to implementation of a grid architecture of this type requires the development of high-efficiency
catalysts to facilitate chemical transformations.
[19]

Figure 1.3 Schematic representation of a grid architecture which incorporates catalysts for the
production of solar fuels. Reprinted with permission from Reference 17.

1.4 HYDROGEN EVOLUTION REACTION
One promising pathway for the generation of solar fuels is the conversion of water into
gaseous hydrogen and oxygen, a process termed “water splitting.”
[20]
In comparison to traditional
fossil fuels, such as crude oil and natural gas, hydrogen gas carries a larger energy density per unit
mass.
[21]
In units of Wh/kg, hydrogen provides nearly double the energy density of methane,
making it a promising fuel to replace fossil fuels. Traditionally, storage of hydrogen has been cited
as a challenge to wide-scale implementation of this fuel, though technological advances in gas
storage are increasingly alleviating this challenge.
[22]
Beyond its use as a fuel, hydrogen is also a
vital commodity chemical in widespread use for the production of ammonia and methanol on an
industrial scale.
[23]
However, the majority of hydrogen currently generated is produced through the
process of steam-methane reforming.
[24]

7

CH4 + 2H2O → CO2 + 4H2 Eq. 1.1

As shown in Equation 1.1, this process involves the conversion of methane (CH4) and water into
carbon dioxide and hydrogen. The generation of carbon dioxide as a result of this transformation
results in significant greenhouse gas emissions, and the dependence of this transformation on
methane requires continued extraction of fossil fuels to maintain hydrogen production. In order to
facilitate the sustainable production of hydrogen, alternative transformation are necessary to
displace steam-methane reforming. An especially promising route to hydrogen production is the
water splitting reaction, shown in Equation 1.2. Unlike steam-methane reforming, water splitting
begins with water as a feedstock, which is abundant on Earth and circumvents the need to extract
methane from Earth’s crust. In further contrast to steam-methane reforming, the byproduct of
hydrogen production though water splitting is gaseous oxygen rather than carbon dioxide, which
imparts carbon-neutrality to this process.

2H2O → O2 + 2H2 Eq. 1.2

Splitting water into hydrogen and oxygen, as shown in Equation 1.2, requires a
thermodynamic potential of 1.23 V.
[25]
While natural photosynthetic systems satisfy this energetic
requirement directly with solar irradiation by photoexcitation, a promising route for the
development of artificial systems is to collect solar irradiation at a photovoltaic array and drive an
electrocatalytic system with solar-derived current.
[26-28]
A schematic representation of such a
system is depicted in Figure 1.4.
[29]
In a typical electrolysis cell, two electrodes are separated by
a semipermeable membrane in a conductive solution. By driving current through this system with
a power supply, electrons are directed from the anode to the cathode. At the anode, where electrons
are collected from the solution, an oxidative half-reaction is driven by the applied potential. At the
8

cathode, where electrons are directed, a reductive half-reaction is driven. In the case of water
splitting, water oxidation to generate O2 occurs at the anode and the electrons removed from water
in this compartment are driven to the cathodic cell where they are consumed in the hydrogen
evolution reaction. The resulting gaseous products can then be collected from their respective cells.
Subsequent consumption of H2 in a fuel cell configuration allows for efficient release of the energy
stored in the hydrogen-hydrogen bond, with clean water generated as the only byproduct.
[30]

Figure 1.4 Schematic depiction of an electrolysis cell for water splitting. Reprinted with
permission from Reference 29.

1.5 CARBON DIOXIDE REDUCTION REACTION
While hydrogen evolution can facilitate a carbon-neutral infrastructure for the storage of
electrical energy, this process does not impact the concentration of CO2 already present in the
atmosphere, nor does it allow for the generation of carbon-based chemicals and materials. As such,
significant attention has also been directed towards the carbon dioxide reduction reaction
(CO2RR).
[31]
By reducing carbon dioxide with proton and electron equivalents, a number of
valuable carbon-containing products can be generated.
[32]
These products include carbon
monoxide (CO), formic acid (HCOOH), and methanol (CH3OH), among others. Each of these
chemicals represents a vital commodity for industrial chemical production. Carbon monoxide is
9

widely applied as syngas—a mixture of CO and H2—for the production of diesel fuels and
methanol, and methanol is generated on a scale of >100 million metric tons per year for the
production of a wide variety of chemicals.
[33]
Methanol has also been proposed as an energy carrier
for a “methanol economy,” in which this liquid fuel could displace traditional gasoline in the
transportation sector.
[34-38]
The reduction of carbon dioxide to carbon monoxide, formic acid, and
methanol are depicted in Equations 1.3-1.5.

CO2 + 2H
+
+ 2e
-
→ CO + H2O Eq. 1.3
CO2 + 2H
+
+ 2e
-
→ HCOOH Eq. 1.4
CO2 + 6H
+
+ 6e
-
→ CH3OH + H2O Eq. 1.5

Unlike the water splitting reaction, in which hydrogen is the only possible reductive
product, the reduction of carbon dioxide can lead to the generation of many possible products (only
three of which are depicted above). As more proton and electron equivalents are involved in the
transformation, the thermodynamic potential for the reduction of CO2 is reduced drastically from
the one-electron reduction of CO2 to CO2
-
(-1.9 V).
[39]
However, this thermodynamic benefit
comes with a kinetic cost, as additional reaction steps and multi-reactant mechanisms are required.
This presents a challenge, as one product must be selectively generated over the remaining possible
products in order to prevent increased costs associated with product separation. Additionally, as
these reactions also involve protons as reactants, another layer of competition exists between
reduction of carbon dioxide and reduction of protons to generate hydrogen.
[40]
Successful
implementation of these reactions into a grid-scale energy infrastructure therefore requires the
design of active and selective catalysts to carry out these transformations.

10

1.6 CATALYST DESIGN
In order to develop efficient catalytic platforms for solar fuels production, a number of
complex criteria must be simultaneously satisfied. Not only must the cost of the catalyst be
sufficiently low to enable large-scale implementation, but the catalyst must perform with high
efficiency, high selectivity, high activity, and low overpotential.
[41]
For this reason, catalyst
performance is described by a number of different parameters.
[42]
For electrocatalytic systems, the
efficiency of a catalyst is measured in terms of a Faradaic efficiency and overpotential.
[43]
The
Faradaic efficiency represents the ratio between the quantity of electrons stored as chemical bonds
and the quantity of electrons directed to the catalytic system. An ideal electrocatalyst exhibits a
Faradaic efficiency of 100%, which indicates that every electron supplied to the catalyst is stored
in the form of a chemical bond in the resulting products. The overpotential of a catalyst is the
measured difference between the thermodynamic potential of a reaction and the potential applied
during catalyst operation. As discussed previously, the thermodynamic potential of water splitting
is 1.23 V. An ideal catalyst would operate at this potential with high rates. In most practical cases,
a potential larger than this must be applied to drive the reaction with sufficiently useful catalytic
rates, resulting in a overpotential. Because the overpotential represents an additional energetic cost,
a reduction in overpotential results in increased energetic efficiency for the overall transformation.
The selectivity of an electrocatalyst is measured in terms of the ratio of the Faradaic efficiencies
for the different products measured. For an ideally selective system, only one product is generated
with high Faradaic efficiency. The activity of a catalyst is measured in terms of catalytic turnover
numbers (TONs) and turnover frequencies (TOFs).
[44-46]
The turnover number is a measure of the
quantity of product generated per quantity of catalyst, and the turnover frequency is a measure of
the turnover numbers generated per unit time. While a large TON indicates long-term stability of
11

the catalyst over multiple catalytic cycles, a high TOF indicates rapid catalysis. Development of
practical catalysts with industrial applications requires that these various metrics be maximized.

To satisfy the first criteria of cost, it is of paramount importance to develop catalytic
systems based on earth-abundant elements and to develop methodologies to improve catalyst
lifetime and reusability. As shown in Figure 1.5, transition metal elements typically employed for
catalytic transformations exhibit a wide range of abundances in the Earth’s crust.
[47]
In the case of
hydrogen evolution, platinum has been widely shown to be the most efficient and active catalyst,
but platinum is of particularly low abundance.
[48]
Similarly for carbon dioxide reduction, silver
and gold catalysts have been shown to exhibit very high activity and selectivity for conversion of
carbon dioxide, but these elements are of low abundance.
[49,50]
As such, the development of large
scale electrocatalytic platforms for energy storage will require the development of catalysts based
on first-row transition metals.
[51]
In cases where precious metals are incorporated, methods must
be developed which enable reuse and recycling of these catalysts to improve catalyst lifetime and
reduce the need for continued extraction of precious metals.
[52]

12

Figure 1.5 Abundance of the elements in Earth’s crust. Reprinted with permission from Reference
47.

To design catalysts which satisfy the requirements of high activity, selectivity, and
efficiency, it is of value to first look to nature where catalytic transformations have been optimized
through evolutionary forces.
[53]
In nature, enzymatic systems function as catalysts to facilitate a
broad range of chemical transformations. A large family of enzymes, termed “metalloenzymes,”
incorporate transition metal atoms ligated with amino acid residues or coenzymes within a protein
scaffold.
[54]
The CO-dehydrogenase family of enzymes, for example, facilitates the reversible
interconversion of CO2 and CO.
[55]
A representation of the active site structure of [NiFe] CO-
dehydrogenase is depicted in Figure 1.6a.
[56]
This system involves a number of key motifs,
including a bimetallic active site, chelating thiolate ligands, redox-active iron-sulfur clusters, and
pendant hydrogen bonding residues, which operate in concert to enable high activity and
selectivity. As depicted in Figure 1.6b, the proposed mechanism of activity for this system
involves the formation of a hydrogen-bonded intermediate wherein a pendant histidine residue
forms a hydrogen bonding interaction with the bound CO2 substrate, which stabilizes this
intermediate for subsequent proton transfer. The incorporation of redox-active iron-sulfur clusters
provides an electron transport pathway for the shuttling of electron equivalents to the active site,
13

as well as providing an electron reservoir which enables multi-electron transformations.
[57]
A broad
range of additional enzymatic systems are known, including [NiFe] hydrogenase for reversible
hydrogen evolution, which similarly incorporate thiolate moieties—capable of protonation or
hydrogen-bond formation—and redox-active iron-sulfur clusters which facilitate electron
transport.
[58]
As such, the incorporation of these motifs into synthetic molecular and materials
systems has been a growing area of research.

Figure 1.6 The active site structure of [NiFe] CO-dehydrogenase (a) and the proposed mechanism
of reactivity (b). Reprinted with permission from Reference 56.

1.7 HETEROGENEOUS AND HOMOGENEOUS CATALYSIS
Traditionally, chemical transformations have been carried out on industrial scales
predominantly through the use of heterogeneous catalysts.
[59]
These heterogeneous, or “surface,”
systems typically involve a solid-state catalyst and a liquid or gaseous substrate interacting with
the surface of the catalytic material. As the surface of a material is highly complex—involving
defect sites, edges, crystal facets, and surface ligands—distinguishing between various possible
catalytic mechanisms and various possible active sites is often not feasible.
[60]
Without a clear
understanding of the underlying mechanism of activity, rational improvement of heterogeneous
systems is a significant challenge. Despite this disadvantage, the phase boundary inherent in this
system allows for simple separation of the catalyst from the reaction medium, as well as convenient
separation of the catalyst from the generated products.
[61]
Additionally, these systems typically
14

exhibit long-term stability which allows for multi-cycle reuse of the catalyst which reduces the
associated lifetime cost.

Molecular systems offer a number of advantages relative to heterogeneous systems.
[62]

Unlike the complexity of a material surface, molecular systems have well-defined structures. The
active site of the catalyst can therefore be rationally modified through synthetic methodologies.
Additionally, intermediates in the catalytic cycle can be more easily isolated and characterized,
allowing for a clear understanding of the underlying mechanisms at play. Together, these two key
advantages allow for the rational tuning of a molecular catalysts activity, selectivity, and stability.
These advantages come at the cost of increased separation cost, as molecular catalysts are typically
employed in homogeneous systems where the catalyst and substrate are both solubilized.
[63]
In the
case of electrocatalysis, the diffusion of these molecular species through the solvent matrix to the
electrochemical double-layer serves as a limitation to the rate of reaction.
[64]
Molecular systems
also typically exhibit lower stability relative to surface systems, which reduces the lifetime of these
catalysts and increases their cost of implementation.

A promising method for combining the advantages of heterogeneous and homogeneous
systems while minimizing their respective drawbacks is the heterogenization of molecular
complexes.
[65,66]
Molecular catalysts can be immobilized on a broad range of surfaces through
various approaches. These approaches have included appending alkylthiol anchoring groups onto
molecular species for attachment to gold surfaces as well as attachment of silane-derivatives to
metal-oxide surfaces through hydration of chlorosilanes.
[67,68]
A wide range of other anchoring-
groups, including phosphonic acids, carboxylic acids, diazonium salts, and π-conjugated systems
(particularly pyrene), have been shown to effectively attach molecular species to surfaces, though
15

each carries with it certain advantages and disadvantages.
[69-72]
These tethering moieties differ in
the surfaces onto which they are able to bind, the strength with which they bind to those surfaces,
and the interactions they have with other species in the system (including their propensity to
dimerize or polymerize on the surface). These different surface-molecule interactions have been
shown to impact the rate of charge transport across the molecular linker, although transport is
dominated by tunneling.
[73,74]
In addition to surface immobilization, molecular species can also be
heterogenized through incorporation into extended materials.
[75]
The advancement of research in
the fields of conductive polymers and coordination networks have opened new doors in molecular
catalysis.
[76]
By applying molecular building blocks to generate polymers, metal-organic
frameworks (MOFs), or covalent organic frameworks (COFs), molecular materials can be
constructed with extended structure and permanent porosity.
[77,80]
These properties both allow for
site isolation of the molecular units and for substrate diffusion through the active material. Site
isolation has been shown to inhibit bimolecular deactivation pathways and therefore improve
catalyst lifetimes, while porosity increases the accessible number of active sites available for
conversion.
[81]

1.8 COBALT DITHIOLENE CATALYSTS FOR HYDROGEN EVOLUTION
In order to supplant the application of platinum in water electrolysis, it is necessary to
develop catalytic platforms based on non-precious, first-row transition metals. Unlike platinum,
transition metals of the first row tend to undergo one-electron redox events, rather than two-
electron redox events.
[82]
Based on this distinction, researchers in first-row transition metal
catalysis have explored redox-active ligand platforms in analogy to the redox-active motifs present
in enzymatic systems.
[83]
Redox activity refers to the ability of the ligand to undergo changes in
redox state, thereby mitigating high or low oxidation states at the metal center.
[84]
By supporting a
16

first-row transition metal complex with a redox-active ligand framework, two-electron transfers
from the catalyst to bound substrate can be enabled without necessitating a two-electron redox
change at the metal center.
[85]
Further inspired by the widespread incorporation of thiolate motifs
in enzymatic systems, as in the active site of [FeFe]-hydrogenase, dithiolate ligand scaffolds have
been widely explored for hydrogen evolution catalysis.
[86-89]
Of these systems, a cobalt complex
with the benzenedithiolate ligand, [Co(bdt)2][TBA] (where bdt = benzene-1,2-dithiolate, TBA =
tetrabutylammonium), has been shown to efficiently catalyze the hydrogen evolution reaction
under both photocatalytic and electrocatalytic conditions.

In the presence of a molecular sensitizer, Ru(bpy)3
2+
(where bpy = bipyridine), a TON of
2700 was reported for this system under irradiation in a 1:1 mixture of water:acetonitrile at pH 4.0
(with ascorbic acid as a sacrificial electron donor).
[90]
Under electrocatalytic conditions, a Faradaic
efficiency of >99% was reported for controlled electrolysis experiments at -1.0 V vs SCE. This
reactivity was further explored through computational studies to develop a mechanistic
understanding. Density functional theory (DFT) calculations predict that the anionic complex first
undergoes a one-electron reduction to generate the dianionic species, [Co(bdt)2]
2–
. Following this,
an initial protonation occurs at one of the chelating thiolate ligands to generate a singly-protonated
species. A second protonation then occurs at a different thiolate site, which is followed by a second
one-electron reduction. The catalyst then undergoes a proton transfer from one of the thiolate sites
to generate a cobalt hydride species. Subsequent coupling of this cobalt hydride with the proximal
protonated thiolate generates H2 which is released to regenerate the active catalyst.

17

Figure 1.7 Proposed mechanistic scheme for hydrogen evolution with the [Co(bdt)2]
–
catalyst.
Reprinted with permission from Reference 91.

1.9 RHENIUM BIPYRIDINE CATALYSTS FOR CARBON DIOXIDE REDUCTION
The rhenium bipyridine system, [Re(bpy)(CO)3Cl], is one of the most well-studied
molecular catalysts to-date for the selective conversion of carbon dioxide to carbon monoxide.
[92-
94]
First reported by Lehn et al., this system has been widely explored due to its selectivity for CO2
reduction over hydrogen evolution with high activity.
[95]
Notably, this system is not only active as
an electrocatalyst, but also as a single-component photocatalyst.
[96]
The unique selectivity for this
system for CO2 binding over metal-hydride formation has been attributed to the redox non-
innocent nature of the bipyridine ligand.
[97]
The frontier orbitals of the active catalyst exhibit mixed
metal-ligand character, providing appropriate symmetry for both σ- and π-type interactions with
axially-bound substrates at the rhenium site. While both protons and CO2 are capable of σ-type
interactions, only CO2 benefits from the additional stabilization through π-type interactions with
the catalyst.

Further computational studies have been explored to provide insight into the mechanism
of this system.
[98]
These studies suggest that an initial one-electron reduction of the catalyst results
in dissociation of the axial chloride ligand to generate the five-coordinate rhenium species,
18

[Re(bpy)(CO)3], and a second reduction event generates the anionic active catalyst,
[Re(bpy)(CO)3]
2–
. Binding of CO2 at the active catalyst generates a metal-carboxylate species,
[Re(bpy)(CO)3(CO2)], which undergoes a protonation event to form a rhenium hydroxycarbonyl
species, [Re(bpy)(CO3)(CO2H)]. A one-electron reduction and protonation facilitate release of
H2O to generate a tetracarbonyl intermediate, [Re(bpy)(CO4)], and a second reduction event leads
to the loss of CO and regeneration of the active catalyst to close the catalytic cycle.

Figure 1.8 Proposed mechanistic scheme for carbon dioxide reduction with the [Re(bpy)(CO)3Cl]
catalyst. Reprinted with permission from Reference 98.

1.10 PROSPECTS IN MOLECULAR AND MATERIALS SYSTEMS FOR SOLAR FUELS
A broad range of catalysts are now known both for water splitting and for the
transformation of carbon dioxide to value-added products. Ongoing research at the frontier of
molecular catalysis for solar fuels production seeks to explore structure-activity relationships
through synthetic modification and subsequent characterization. By probing these fundamental
relationships, new design principles are established for the development of next-generation
catalytic platforms. Basic research into the mechanism of activity and the influence of electronic
19

structure on mechanism and reactivity is therefore key for the development of catalysts with low
overpotential, high catalytic rates, and long catalyst lifetimes. As the sustainable development of
large-scale systems benefits from the reuse and recycling of catalysts, the development of
advanced methodologies for the heterogenization of molecular catalysts offers promising
opportunities in green chemical transformations. Fundamental insights into the relationships
between heterogenization methodology and catalyst activity and lifetime are expected to provide
additional design principles for the development of large-scale catalytic platforms. The availability
of such systems could enable grid-scale solar fuels production with potential impacts in the energy
and chemical manufacturing sectors.

In this thesis, we seek to address both the design of molecular systems for solar fuels
production and the integration of molecular systems into extended heterogeneous materials.
Chapter 2 describes the design, synthesis, and characterization of a cobalt complex with the 2-
(diisopropylphosphaneyl)benzenethiol ligand. In analogy to the bis(dithiolene) complexes of
cobalt previously studied, this novel system exhibits ligand non-innocence which is predicted by
density functional theory calculations. Catalytic studies reveal the generation of syngas mixtures
under CO2 atmosphere, with a maximum TOFCPE of 250(25) s
-1
measured in the presence of
phenol. When water is applied as a proton source syngas mixtures with H2:CO ratios of ~3 were
generated, and this ratio is influenced by the identity of the acid additive. A mechanistic hypothesis
is proposed based on stoichiometric studies, electrochemical results, and density functional
calculations, which together indicate that protonation of the pendant thiolate moieties plays a key
role in the catalytic cycle. This study provides insight into the development of a catalytic system
incorporating non-innocent ligands with pendant base moieties for electrochemical syngas
production. Chapter 3 describes the development of macrocyclic catalysts for CO2 reduction based
20

on iron and nickel as earth-abundant elements. Prior studies regarding a azacalix[4](2,6)-pyridine
framework incorporating cobalt indicated high selectivity and activity towards CO2 reduction to
CO. By translating this behavior to more abundant elements, the costs of this catalyst could be
reduced for practical implementation. Upon switching the metal center from cobalt to iron or
nickel, however, a drastic departure in catalytic behavior was observed. Both the iron and nickel
systems exhibit poor selectivity for CO over H2 production, and both result in the deposition of
heterogeneous, nanoparticulate material onto the electrode surface. Density functional theory
calculations were explored to rationalize this departure in behavior. Molecular orbital pictures
suggest that, of the three transition metals discussed, only cobalt positions the frontier orbital such
that both σ- and π-type interactions with an axially-bound CO2 species are facilitated. These results
broaden the scope of our prior studies by demonstrating the significance of the transition metal ion
in determining the electrochemical behavior of these macrocyclic systems.

Chapters 4 and 5 explore a novel methodology for the immobilization of molecular
catalysts to electrode surfaces. This methodology is based on prior established work in diazonium
grafting, where the reduction of aryldiazonium salts at an electrode surface leads to surface
functionalization by the corresponding aryl-radical species. Traditional techniques for surface
modification with diazonium salts results in the formation of amorphous films with no long-range
order and poor control over the loading of surface species. In our improved methodology,
bis(diazonium) salts of a rhenium bipyridine complex are employed to enable directed growth of
rigid films with extended conjugation. An optimized electropolymerization methodology was
developed, enabling the growth of oriented molecular films with nanomolar control over surface
loadings. The resulting films were shown to be active as both electrocatalysts and single-
component photocatalysts for the selective conversion of CO2 to CO. Electrocatalytic studies using
21

modified graphite rod electrodes performed at -2.25 V vs Fc/Fc
+
for 2 hours reveal CO production
with Faradaic efficiencies and turnover numbers up to 99% and 3606, respectively. Photocatalytic
studies of modified TiO2 devices demonstrate enhanced activity at low catalyst loadings, with
turnover numbers up to 70 during 5 hours of irradiation. In Chapter 5, this methodology is extended
to carbon fabric electrodes which offer high surface area and low substrate cost. CO2 electrolysis
studies with these electroactive cloths reveal rapid (kcat ~ 40 s
-1
) and selective (Faradaic efficiency
>99%) conversion to CO with turnover numbers (TON) per rhenium site reaching ~290,000 and
catalytic currents (icat) >10 mA/cm
2
. This represents a substantial improvement over prior systems,
with an 80-fold increase in turnover number relative to the graphite rod studies presented in
Chapter 4. Kinetic isotope effect measurements and electrochemical data indicate a mechanism
involving charging of the conjugated backbone, followed by chloride dissociation to open an
accessible coordination site at the rhenium centers of the metallopolymer. As numerous metal-
bipyridine complexes are known for a broad scope of electrocatalytic transformations, these
integrated carbon cloth devices are anticipated to serve as a platform for future studies.

22

1.11 REFERENCES
1. U.S. Energy Information Administration, Internation Energy Outlook. 2019.
2. IEA. 2020 World Energy Outlook: Executive Summary. Oecd/Iea 2020, 11.
3. IEA. 2019 World Energy Outlook: Executive Summary. Oecd/Iea 2019, 11.
4. U.S. Energy Information Administration, Annual Energy Outlook. 2020.
5. U.S. Energy Information Administration, Short-Term Energy Outlook. 2020.
6. Keeling, Ralph F.; Keeling, Charles D. (2017). Atmospheric Monthly In Situ CO 2 Data -
Mauna Loa Observatory, Hawaii. In Scripps CO2 Program Data. UC San Diego Library
Digital Collections. https://doi.org/10.6075/J08W3BHW Accessed on May 6, 2020.
7. Joos et al. Atmospheric Chemistry and Physics. 2013, 13(5), 2793-2825.
8. Turner, J. A. Science 2004, 305, 972-974.
9. Chu, S.; Majumdar, A. Nature 2012, 488, 294-303.
10. Alanne, K.; Saari, A. Renew. Sustain. Energy Rev. 2006, 10 (6), 539–558.
11. Thiam, D.-R. Renew. Energy 2010, 35 (8), 1615–1623.
12. Lacey-Barnacle, M.; Robison, R.; Foulds, C. Energy Sustain. Dev. 2020, 55, 122–138.
13. Nocera, D. G. Acc. Chem. Res. 2017, 50 (3), 616–619.
14. Graves, C.; Ebbesen, S. D.; Mogensen, M.; Lackner, K. S. Renew. Sustain. Energy Rev.
2011, 15 (1), 1–23.
15. Detz, R. J.; Reek, J. N. H.; Van Der Zwaan, B. C. C. Energy Environ. Sci. 2018, 11 (7),
1653–1669.
16. Olah, G. A.; Prakash, G. K. S.; Goeppert, A. J. Am. Chem. Soc. 2011, 133 (33), 12881–
12898.
17. Seh, Z. W.; Kibsgaard, J.; Dickens, C. F.; Chorkendorff, I.; Nørskov, J. K.; Jaramillo, T.
F. Science 2017, 355, eaad4998.
18. House, R. L.; Iha, N. Y. M.; Coppo, R. L.; Alibabaei, L.; Sherman, B. D.; Kang, P.;
Brennaman, M. K.; Hoertz, P. G.; Meyer, T. J. J. Photochem. Photobiol. C Photochem.
Rev. 2015, 25, 32–45.
23

19. Kim, D.; Sakimoto, K. K.; Hong, D.; Yang, P. Angew. Chemie Int. Ed. 2015, 54 (11), 3259–
3266.
20. G. Walter, M.; L. Warren, E.; R. McKone, J.; W. Boettcher, S.; Mi, Q.; A. Santori, E.; S.
Lewis, N. Chem. Rev. 2010, 110 (11), 6446–6473.
21. Züttel, A.; Remhof, A.; Borgschulte, A.; Friedrichs, O. Philos. Trans. R. Soc. A Math. Phys.
Eng. Sci. 2010, 368 (1923), 3329–3342.
22. Rowsell, J. L. C.; Millward, A. R.; Park, K. S.; Yaghi, O. M. J. Am. Chem. Soc. 2004, 126
(18), 5666–5667.
23. Rostrup-Nielsen, J. R. Catal. Today 1993, 18 (4), 305–324.
24. Rostrup-Nielsen, J. R. Catal. Today 2002, 71, 243–247.
25. Meyer, K.; Ranocchiari, M.; van Bokhoven, J. A. Energy Environ. Sci. 2015, 8 (7), 1923–
1937.
26. Luna, P. De; Hahn, C.; Higgins, D.; Jaffer, S. A.; Jaramillo, T. F.; Sargent, E. H. Science
2019, 364, 6438.
27. Zeng, M.; Li, Y.; Houlachi, G. J. Mater. Chem. A 2015, 3 (29), 14942–14962.
28. A. Fujishima, K. Honda, Nature 1972, 238, 37–38.
29. US Department of Energy's Office of Energy Efficiency & Renewable Energy.
30. Demirdoven, N.; Deutch, J.; Golden, D. M. Science. 2004, 305 (5686), 974–976.
31. Verma, S.; Kim, B.; Jhong, H.-R. M.; Ma, S.; Kenis, P. J. A. ChemSusChem 2016, 9 (15),
1972–1979.
32. Qiao, J.; Liu, Y.; Hong, F.; Zhang, J.; Koper, M. T. M. M. Chem. Soc. Rev. 2014, 43 (2),
631–675.
33. M. Berggren, ‘Global Methanol: Demand Grows as Margins Atrophy’, 19th IMPCA Asian
Methanol Conference, Singapore, 1st–3rd November, 2016
34. Olah, G. A. Angew. Chemie - Int. Ed. 2005, 44 (18), 2636–2639.
35. A. Olah, G.; Goeppert, A.; K. Surya Prakash, G. J. Org. Chem. 2008, 74 (2), 487–498.
36. Surampudi, S.; Narayanan, S. R. R.; Vamos, E.; Frank, H.; Halpert, G.; LaConti, A.; Kosek,
J.; Prakash, G. K. S.; Olah, G. A. A. J. Power Sources 1994, 47 (3), 377–385.
24

37. Olah, G. A. Catalysis Letters, 2004 93(1-2), 1-2.
38. Alberico, E.; Nielsen, M. Chem. Commun. 2015, 51 (31), 6714–6725.
39. Takeda, H.; Ishitani, O. Coord. Chem. Rev. 2010, 254 (3–4), 346–354.
40. Barlow, J. M.; Yang, J. Y. ACS Cent. Sci. 2019, 5 (4), 580–588.
41. Artero, V.; Saveant, J. M. Energy Environ. Sci. 2014, 7 (11), 3808–3814.
42. Costentin, C.; Savéant, J.-M. ChemElectroChem 2014, 1 (7), 1226–1236.
43. Costentin, C.; Passard, G.; Savéant, J. M. J. Am. Chem. Soc. 2015, 137 (16), 5461–5467.
44. Costentin, C.; Savéant, J.-M. J. Am. Chem. Soc. 2017, 139 (24), 8245–8250.
45. Bligaard, T.; Morris Bullock, R.; T. Campbell, C.; G. Chen, J.; C. Gates, B.; J. Gorte, R.;
W. Jones, C.; D. Jones, W.; R. Kitchin, J.; L. Scott, S. ACS Catal. 2016, 6 (4), 2590–2602.
46. Bligaard, T.; Morris Bullock, R.; T. Campbell, C.; G. Chen, J.; C. Gates, B.; J. Gorte, R.;
W. Jones, C.; D. Jones, W.; R. Kitchin, J.; L. Scott, S. ACS Catal. 2016, 6 (4), 2590–2602.
47. U.S. Department of the Interior, U.S. Geological Survey—USGS Fact Sheet. 2002.
48. Morales-Guio, C. G.; Stern, L.-A.; Hu, X. Chem. Soc. Rev. 2014, 43, 6555-6569.
49. Qiao, J.; Liu, Y.; Hong, F.; Zhang, J. Chem. Soc. Rev. 2014; Vol. 43.
50. Kortlever, R.; Shen, J.; Schouten, K. J. P.; Calle-Vallejo, F.; Koper, M. T. M. J. Phys.
Chem. Lett. 2015, 6 (20), 4073–4082.
51. Su, B.; Cao, Z.-C.; Shi, Z.-J. Acc. Chem. Res. 2015, 48 (3), 886–896.
52. Bairagya, M. Das; Bujol, R. J.; Elgrishi, N. Chem. - A Eur. J. 2020, 1–11.
53. Artero, V. Nat. Energy 2017, 2, 17131.
54. Wang, H. Chem. Soc. Rev. 2013, 42, 8360–8375.
55. Holm, R. H.; Kennepohl, P.; Solomon, E. I. Chem. Rev. 1996, 96, 2239−2314.
56. Appel, A. M.; Bercaw, J. E.; Bocarsly, A. B.; Dobbek, H.; Dubois, D. L.; Dupuis, M.;
Ferry, J. G.; Fujita, E.; Hille, R.; Kenis, P. J. A.; Kerfeld, C. A.; Morris, R. H.; Peden, C.
H. F.; Portis, A. R.; Ragsdale, S. W.; Rauchfuss, T. B.; Reek, J. N. H.; Seefeldt, L. C.;
Thauer, R. K.; Waldrop, G. L. Chem. Rev. 2013, 113 (8), 6621–6658.
25

57. Beinert, H. JBIC 2000, 5, 2-15.
58. Ogata, H.; Nishikawa, K.; Lubitz, W. Nature 2015, 520 (7548), 571–574.
59. Armor, J. N. A Catal. Today 2011, 163 (1), 3–9.
60. Zhou, Z. Y.; Tian, N.; Li, J. T.; Broadwell, I.; Sun, S. G. Chem. Soc. Rev. 2011, 40 (7),
4167–4185.
61. Mizuno, N.; Misono, M. Chem. Rev. 1998, 98 (1), 199–217.
62. Thoi, V. S.; Sun, Y.; Long, J. R.; Chang, C. J. Chem. Soc. Rev. 2013, 42, 2388-2400.
63. Morris, A. J.; Meyer, G. J.; Fujita, E. Acc. Chem. Res. 2009, 42 (12), 1983–1994.
64. Costentin, C.; Saveant, J.-M. J. Phys. Chem. C 2015, 119, 22, 12174-12182.
65. Bullock, R. M.; Das, A. K.; Appel, A. M. Chem. - A Eur. J. 2017, 23 (32), 7626–7641.
66. Louis, M. E.; Fenton, T. G.; Rondeau, J.; Jin, T.; Li, G. Comments Inorg. Chem. 2016, 36
(1), 38–60.
67. Love, C. J.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Chem. Rev.
2005, 105, 1103−1169
68. Ulman, A. Chem. Rev. 1996, 96, 1533–1554.
69. Meyer, T. J.; Sheridan, M. V.; Sherman, B. D. Chem. Soc. Rev. 2017, 46 (20), 6148–6169.
70. Materna, K. L.; Crabtree, R. H.; Brudvig, G. W. Chem. Soc. Rev. 2017, 46 (20), 6099–
6110.
71. Mahouche-Chergui, S.; Gam-Derouich, S.; Mangeney, C.; Chehimi, M. M. Chem. Soc.
Rev. 2011, 40 (7), 4143–4166.
72. Blakemore, J. D.; Gupta, A.; Warren, J. J.; Brunschwig, B. S.; Gray, H. B. J. Am. Chem.
Soc. 2013, 135 (49), 18288–18291.
73. Bowers, C. M.; Rappoport, D.; Baghbanzadeh, M.; Simeone, F. C.; Liao, K. C.; Semenov,
S. N.; Zaba, T.; Cyganik, P.; Aspuru-Guzik, A.; Whitesides, G. M. J. Phys. Chem. C 2016,
120 (21), 11331–11337.
74. Chen, F.; Li, X.; Hihath, J.; Huang, Z.; Tao, N. J. Am. Chem. Soc. 2006, 128 (49), 15874–
15881.
26

75. Furukawa, H.; Cordova, K. E.; O’Keeffe, M.; Yaghi, O. M. Science 2013, 341 (6149),
1230444–1230444.
76. Downes, C. A.; Marinescu, S. C. ChemSusChem, 2017, 10, 4374–4392
77. O’Toole, T. R.; Margerum, L. D.; Westmoreland, T. D.; Vining, W. J.; Murray, R. W.;
Meyer, T. J. J. Chem. Soc. Chem. Commun. 1985, 20, 1416.
78. Andrieux, C. P.; Dumas-Bouchiat, J. M.; Saveant, J. M. J. Electroanal. Chem. 1980, 114
(1), 159–163.
79. Wang, J. L.; Wang, C.; Lin, W. ACS Catal. 2012, 2 (12), 2630–2640.
80. Wang, S.; Wang, X. Small 2015, 11 (26), 3097–3112.
81. Drake, T.; Ji, P.; Lin, W. Acc. Chem. Res. 2018, 51 (9), 2129–2138.
82. Wang, M.; Chen, L.; Sun, L. Energy Environ. Sci. 2012, 5 (5), 6763–6778.
83. George, G. N.; Costa, C.; Moura, J. J. G.; Moura, I. J. Am. Chem. Soc. 1999, 121 (11),
2625–2626.
84. Luca, O. R.; Crabtree, R. H. Chem. Soc. Rev. 2013, 42 (4), 1440–1459.
85. Arévalo, R.; Chirik, P. J. J. Am. Chem. Soc. 2019, 141, 23, 9106-9123.
86. Fogeron, T.; Todorova, T. K.; Porcher, J. P.; Gomez-Mingot, M.; Chamoreau, L. M.;
Mellot-Draznieks, C.; Li, Y.; Fontecave, M. A ACS Catal. 2018, 8 (3), 2030–2038.
87. Letko, C. S.; Panetier, J. A.; Head-Gordon, M.; Tilley, T. D. J. Am. Chem. Soc. 2014, 136
(26), 9364–9376.
88. Lee, K. J.; McCarthy, B. D.; Rountree, E. S.; Dempsey, J. L Inorg. Chem. 2017, 56 (4),
1988–1998.
89. Eisenberg, R.; Gray, H. B. Inorg. Chem. 2011, 50 (20), 9741–9751.
90. McNamara, W. R.; Han, Z.; Yin, C.-J.; Brennessel, W. W.; Holland, P. L.; Eisenberg, R.
Proc. Natl. Acad. Sci. 2012, 109 (39), 15594–15599.
91. Solis, B. H.; Hammes-Schiffer, S. J. Am. Chem. Soc. 2012, 134 (37), 15253–15256.
92. Smieja, J. M.; Kubiak, C. P. Inorg. Chem. 2010, 49 (20), 9283–9289.
93. Kumar, B.; Smieja, J. M.; Kubiak, C. P. J. Phys. Chem. C 2010, 114 (33), 14220–14223.
27

94. Clark, M. L.; Cheung, P. L.; Lessio, M.; Carter, E. A.; Kubiak, C. P. ACS Catal. 2018, 8
(3), 2021–2029.
95. Hawecker, J.; Lehn, J.-M.; Ziessel, R. J. Chem. Soc., Chem. Commun. 1984, 984 (6), 328–
330.
96. Hawecker, J.; Lehn, J.-M.; Ziessel, R. J. Chem. Soc., Chem. Commun. 1983, No. 9, 536–
538.
97. Smieja, J. M.; Benson, E. E.; Kumar, B.; Grice, K. A.; Seu, C. S.; Miller, A. J. M.; Mayer,
J. M.; Kubiak, C. P Proc. Natl. Acad. Sci. U.S.A. 2012, 109 (39), 15646–15650.
98. Keith, J. A.; Grice, K. A.; Kubiak, P.; Carter, E. A. J. Am. Chem. Soc. 2013, 135, 15823–
15829.
28

CHAPTER 2

Electrocatalytic Syngas Generation with a Redox Non-Innocent
Cobalt 2-Phosphinobenzenethiolate Complex

29

2.1 ABSTRACT
A cobalt complex supported by the 2-(diisopropylphosphaneyl)benzenethiol ligand was
synthesized and its electronic structure and reactivity were explored. Density functional theory
calculations and electronic spectroscopy measurements suggest a non-negligible degree of ground-
state thiyl-radical character, in analogy to previously studied dithiolene and diselenolene systems.
Electrochemical studies in the presence of 1 atm of CO2 and Brønsted acid additives indicate that
the cobalt complex generates syngas, a mixture of H2 and CO, with Faradaic efficiencies up to
>99%. The ratios of H2:CO generated vary based on the additive. A H2:CO ratio of ~3:1 is
generated when H2O is used as the Brønsted acid additive. A maximum TOFCPE of 250(25) s
-1
and
TONCPE of 9.06  10
5
was measured in the presence of 0.5 M PhOH. A mechanistic hypothesis is
proposed whereby competitive binding between a proton and CO2 dictates selectivity. This study
provides insight into the development of a catalytic system incorporating non-innocent ligands
with pendant base moieties for electrochemical syngas production.

Key words: Carbon Dioxide Fixation, Homogeneous Catalysis, Sustainable Chemistry, Solar
Fuels, Green Chemistry

2.2 INTRODUCTION
Displacing fossil fuels in the electricity generation, transportation, and chemical production
sectors with sustainable alternatives would allow for continued global advancement while
mitigating the resulting environmental impact.
1,2
While sunlight provides a renewable source of
energy, its spatio-temporal variability demands distributed energy storage to meet peak demands.
3

An ideal solution is to redirect excess solar-derived electricity towards electrocatalytic cells at
times of low demand to convert abundant small molecules into liquid fuels and industrial
chemicals.
4
Of particular interest for solar energy conversion are the fundamental reactions of
30

photosynthesis: water splitting and CO2 reduction.
5
Storing solar energy through the chemical
transformation of these substrates offers a strategy to recycle captured CO2.
6
Synthesis gas
(syngas), a combination of CO and H2 in ratios of 1:1 to 1:3, is a valuable feedstock derived from
coal or natural gas in widescale use for the synthesis of hydrocarbons for diesel fuel (through
Fischer-Tropsch chemistry) and the production of methanol.
7,8,9,10
A sustainable energy future will
require a non-fossil fuel method for the formation of syngas, and the electrolysis of CO2 using
solar-derived electricity is a viable pathway to do so.
11
In particular, it is vital to develop solar-to-
fuel technologies based on complexes containing only non-precious elements that catalyze water
splitting and CO2 reduction with high selectivity.
11
Fundamental studies into the electronic
structure and reactivity of molecular systems offers a promising path to develop mechanistic
understandings of these transformations, which may ultimately lead to new design principles.
The development of molecular catalysts allows for synthetically-tunable active sites with
rapid and selective reactivity.
12,13
Numerous molecular systems have been reported for the
selective conversion of CO2 to CO, as well as for the reduction of protons to H2.
14,15
However,
fewer systems have been reported which allow for tunable generation of syngas mixtures and fewer
yet which use only non-precious metals.
16
One literature example describes photoelectrocatalytic
syngas production with a homogeneous rhenium bipyridine complex at a p-Si photoelectrode.
17
It
is proposed that the homogeneous rhenium catalyst produces CO, while the surface of the p-Si
photoelectrode facilitates H2 evolution. Single-component systems have also been reported, such
as nickel and cobalt complexes, which generate tunable syngas mixtures by varying the applied
potential.
18,19
A ruthenium(II) polypyridyl carbene complex was reported to facilitate both syngas
production and water oxidation in aqueous media.
20
In both cases, the competitive binding of CO2
or H
+
as substrate in the catalytic cycle was identified as a key step in determining the overall
31

product selectivity. As such, we considered the need to include structural motifs to facilitate this
competitive binding.
Ligand non-innocence and hydrogen-bonding interactions have been shown to lower the
activation barrier of reactive intermediates and to direct proton-coupled electron transfer (PCET)
reactions.
21
Many synthetic systems have been devised to exploit these structural motifs, in
analogy to CO-dehydrogenase and [FeFe]-hydrogenase for the reversible conversion of CO2 to
CO and protons to H2, respectively.
22,23,24
Thiolate and selenolate ligands in cobalt bis(dithiolene)
and bis(diselenolene) complexes have been shown to serve as proton relays, proximal to a cobalt
hydride, which facilitate rapid H2 evolution.
25,26,27,28
These cobalt chalcogenide complexes have
been primarily optimized for H2 evolution over CO2 reduction. In order to develop a system with
competitive CO2 binding, prior studies suggest that stronger donor ligands may be necessary to
increase the nucleophilicity of the cobalt center.
29,30,31

Hybrid phosphine-thiolate (PS) chelating systems offer both a strong π-acidic phosphine
for the stabilization of low-valent metal centers as well as a π-basic thiolate, which may serve to
facilitate hydrogen-bond networks. It was shown that ligand protonation in a cobalt complex with
thiolate and phosphine donors facilitates CO2 reduction with high turnover frequency (1559 s
-1
),
low overpotential (70 mV), and high selectivity for CO (~95%).
32
Analogous systems based on a
2-phosphaneylbenzenethiol ligand have shown promising activity towards H2 evolution and
oxidation.
33
Electronic structure calculations of a rhenium complex with the 2-
(diphenylphosphaneyl)benzenethiol ligand have indicated redox non-innocence behavior,
suggesting that this ligand platform facilitates similar orbital interactions as those facilitated by
dithiolenes. Catalytic studies of nickel and rhenium complexes supported by PS ligands have
provided insight into the behavior of this ligand class, suggesting both metal-centered and ligand-
32

centered reactivity.
33,34
Herein, we describe a cobalt complex supported by the 2-
(diisopropylphosphaneyl)benzenethiol ligand, [Co
II
(P(iPr)2SC6H4)2], designated CoPS. Along
with structural characterization, density functional theory (DFT) and time-dependent DFT
calculations are explored to model the electronic structure of this system. The reactivity of CoPS
is studied in the presence of CO2 and various Brønsted acid additives, including water.

2.3 RESULTS AND DISCUSSION

Figure 2.1 Left: Chemdraw structure of CoPS. Right: Solid state structure of CoPS, top view with
color designations are as follows: Co = pink, S = yellow, P = orange, C = gray, H = white.

Single crystal X-ray diffraction studies of CoPS indicate a square planar geometry around
the cobalt center with a trans arrangement of the phosphine ligands (Figure 2.1, details tabulated
in Tables 2.1-2.3). The complex exhibits pseudo C2h symmetry with eclipsed isopropyl
substituents, though the orientation of these substituents may be attributed to packing interactions
in the solid state. This geometry generates a sterically-crowded cobalt center with an accessible
binding pocket at cobalt and sterically-accessible thiolate moieties. The trans arrangement of the
phosphine ligands is attributed to steric repulsion of the isopropyl groups, as was shown previously
for the analogous nickel and tin complexes.
33,35
To probe this structure further, DFT calculations
were performed at the 6-31G(d)/PBE level of theory (see Experimental Methods for details) to
ensure convergence. The resulting optimized geometry was confirmed as a stable minimum with
frequency calculations at the same level of theory. The calculated geometry predicted by DFT is
consistent with the solid-state structure obtained for CoPS, and is similar to that of the analogous
33

nickel complex.
33
The measured Co–S and Co–P bond lengths of 2.1625(4) and 2.2163(5) Å,
respectively, are also consistent with the optimized DFT geometry. No counter-ions are observed
in the outer-sphere, consistent with a neutral complex, with a single benzene solvent molecule per
unit cell. Elemental analysis of the complex further confirms the predicted chemical formula (see
Experimental Methods).

Table 2.1 X-ray crystallography data and structure refinement for CoPS.
Chemical Formula C30H42CoP2S2
Formula weight 587.62
Crystal system Monoclinic
Space group P 1 21/n 1
a (Å) 11.658(3)
b (Å) 9.739(3)
c (Å) 14.049(4)
α (°) 90
β (°) 111.120(3)
γ (°) 90
V (Å
3
) 1487.9(7)
Z 2
Dcalc (g/cm
3
) 1.312
μ (Mo Κα) (mm
-1
) 0.842
F (000) 622
Reflections collected 21559
Independence reflections 4494 [Rint = 0.0267]
R1 (I > 2σ(I)) 0.0208
wR2 (all data) 0.0538
Goodness-of-fit (GOF) on F
2
1.015

Table 2.2 Selected bond lengths (Å) for CoPS.
Co1-S1 2.1625(4)
Co1-P1 2.2163(5)
S1-C10 1.7735(10)
P1-C6 1.8542(10)
P1-C9 1.8233(10)
C9-C10 1.4041(13)

34

Table 2.3 Selected bond angles (°) for CoPS.
S1-Co1-S1 180.0
P1-Co1-P1 180.0
S1-Co1-P1 90.95(2)
C10-S1-Co1 107.70(4)
C9-P1-Co1 107.49(3)

Figure 2.2 400 MHz
1
H NMR spectrum of 1 mM CoPS in a solution of benzene-d6 (left). Pink
asterisks are attributed to the molecular complex. Additional resonances are attributed to residual
solvent. Also shown is a 400 MHz
1
H NMR spectrum of 1 mM CoPS in a solution of acetonitrile-
d3 (right), exhibiting analogous resonances.

Figure 2.3 Calculated spin density distribution of CoPS at the def2-TZVP/PBE level of theory.
Plotted with isovalue = 0.05 for clarity.

1
H NMR spectroscopy studies of CoPS in benzene-d6 (selected for increased solubility)
reveal four broad paramagnetic resonances at  –22.5, –16, 11.2, and 24.9 ppm, which each
35

integrate to two protons and are attributed to the aromatic protons of the benzene backbone.
Additional resonances at  –0.5, 2.5, and 4.8 ppm are observed and integrate to a total of 28 protons
in a 3:1:3 ratio. These resonances are assigned to the distinct isopropyl environments (Figure 2.2).
The molecular nature of this complex was also confirmed in acetonitrile through NMR
spectroscopy studies, which show analogous resonances in the paramagnetic spectra.
Paramagnetic susceptibility measurements were performed in benzene-d6 according to Evan’s
method to determine the effective magnetic moment of CoPS.
36
Based on these studies, a eff of
1.77 B was determined, which corresponds to one unpaired electron. These results are supported
by unrestricted single-point energy calculations with a larger basis set at the def2-TZVP/PBE level
of theory, which predict the neutral complex to exhibit ground state doublet character with non-
negligible spin density localized on the thiolate ligands ( ρ = 0.2), suggesting a degree of thiyl-
radical character as illustrated in Figures 2.3 and 2.4.

Figure 2.4 UV-Vis spectroscopy study of 0.2 mM CoPS in a benzene solution. Inset highlights a
low-energy transition at ~830 nm. Also shown is the calculated spin density for the ground state,
indicating ~20% thiyl-radical character.

36

Figure 2.5 Measured absorbance for observed electronic transitions as a function of CoPS
concentration as determined by UV-Vis spectroscopy.

Table 2.4 Calculation of molar extinction coefficients for CoPS based on dilution studies by UV-
Vis spectroscopy.
[CoPS] (mM) A (834 nm) A (480 nm) A (388 nm) A (358 nm) A (288 nm)
0.40 0.015 1.164 1.616 1.001 -
0.20 0.007 0.545 0.757 0.488 3.1583
0.16 0.006 0.452 0.628 0.391 2.5787
0.08 0.003 0.219 0.303 0.179 1.2344
0.04 0.001 0.112 0.154 0.086 0.6276
Slope
(M
-1
cm
-1
)
3.8 2862.8 3975.3 2491.0 15871.0

37

Figure 2.6 UV-Vis study of 1 mM CoPS in acetonitrile solution

The electronic structure of this complex was probed experimentally through UV-Vis
spectroscopy in a benzene solution. Upon dissolution in benzene, the intense orange crystals
produce a vivid, yellow solution. As seen in Figure 2.4, four dominant electronic transitions were
observed with one additional shoulder. Dilution studies were also performed to determine the
molar extinction coefficients for these transitions. The results of these studies reveal transitions at
λ ( ε) [nm (M
-1
cm
-1
)] = 834 (3.8), 480 (2862.8), 388 (3975.3), 358 (2491.0, shoulder), and 288
(15871.0) (see Figure 2.5 and Table 2.4). The observed electronic spectrum is analogous to that
reported for a cobalt bis(dithiolene) complex, suggesting similar thiyl-radical character and a
mixed metal-ligand ground state. As such, assignment of the oxidation state of cobalt is
ambiguous. Analogous transitions were observed in acetonitrile, suggesting a coordination
environment analogous to that in benzene (Figure 2.6).

To further probe the electronic structure of CoPS, unrestricted time-dependent density
functional theory (TD-DFT) calculations were performed at the def2-TZVP/B3LYP level of theory
in the gas phase. The B3LYP functional was selected based on literature precedent, as this
functional provides an appropriate description of a broad range of molecular transition metal
38

complexes.
37
The resulting Kohn-Sham molecular orbital picture is depicted in Figure 2.7. The
predicted UV-Vis spectrum parallels the experimental spectra with five dominant transitions,
though the absolute energies of these predictions vary due to the gas-phase nature of these
calculations (Table 2.5). As such, these calculations offer qualitative predictions for the character
of the electronic transitions. Based on these calculations, the ground state electronic configuration
of the complex is predicted to have a singly occupied 3bg orbital, which displays π
*
(Co-S)
character with significant contribution from the S(p) and Co(dxz) orbitals. This qualitative SOMO
character parallels that predicted for both the cobalt bis(dithiolene) and bis(diselenolene)
complexes.
21
The lowest-lying unoccupied one-electron orbital corresponds to the 3bg β orbital,
which exhibits similar character as the α SOMO with more substantial metal contribution.

Figure 2.7 Molecular orbital diagram based on TD-DFT calculations at the def2-TZVP/B3LYP
level of theory. Donor orbitals shown with negative regions colored red and positive regions
colored blue, acceptor orbitals are shown with negative regions colored green and positive regions
colored pink. Vertical lines indicate dominant transitions (solid) and additional contributions
(dashed), with transitions identified by color (λ1 = red, λ2 = orange, λ3 = green, λ4 = blue, and λ5 =
violet).

39

Table 2.5 Calculated oscillator strengths and wavelengths of electronic transitions by TD-DFT
and comparison to experimentally-determined transition wavelengths. Only transitions with fcalc ≥
7 x 10
-5
are shown.
Transition fcalc λcalc (nm) λexp (nm) |Difference| (nm)
λ1 7.0  10
-5
721 834 113
λ2 5.9  10
-4
482 480 2.0
λ3 2.7  10
-3
416 388 28.8
λ4 1.8  10
-2
394 358 36.6
λ5 1.2  10
-2
390 288 103

Avg |Difference| (nm) 56.6

As illustrated in Figure 2.7, the lowest-energy transition observed at 834 nm ( λ1) is
predicted to correspond to the singlet-to-singlet excitation of a β spin in an orbital of ag symmetry
with Co(dxy) and S(py) orbital contribution to the mixed metal-ligand 3bg orbital. The next lowest-
energy transition ( λ2 = 480 nm) is predicted to correspond to a ligand-to-metal charge transfer
(LMCT) excitation of an α electron from the 3bg mixed metal-ligand character SOMO with
predominant ligand-character to the 2ag LUMO, which exhibits σ
*
(Co-S) character with
contributions from S(px) and Co(dx2-y2) orbitals. The electronic transition observed at 388 nm ( λ
3) is predicted to exhibit metal-to-ligand charge transfer (MLCT) character with both α and β
contribution. An additional shoulder appearing at 358 nm ( λ4) is attributed to excitation of an
electron from occupied orbitals with S(pz) and Co(dz2) contribution to the 2ag LUMO. Finally, the
highest-energy observed transition ( λ5 = 288 nm) is predicted to correspond to a ligand-to-metal
charge transfer (LMCT) transition from the 1bu orbital exhibiting S(py) contributions to the 2ag
LUMO of Co(dx2-y2) character.
40

Figure 2.8 Cyclic voltammograms of 1 mM CoPS in an acetonitrile solution under N2 atmosphere
with 0.1 M [nBu4N][PF6] with scan rate = 100 mV/s.

Figure 2.9 Cyclic voltammetry of CoPS (1 mM) in acetonitrile with 0.1 M [nBu4N][PF6]
supporting electrolyte (scan rate = 100 mV/s) from 0 V vs Fc/Fc
+
to (a) -2.3 V vs Fc/Fc
+
and (b) -
2.7 V vs Fc/Fc
+
.

The electrochemical behavior of CoPS was analyzed by cyclic voltammetry (CV) studies
in an acetonitrile solution with 0.1 M [nBu4N][PF6] electrolyte using a glassy carbon working
electrode.
38
Acetonitrile was selected as the electrochemical solvent in analogy to the studies
performed on the cobalt bis(dithiolene) and bis(diselenolene) complexes. Studies in 1:1
water:acetonitrile mixtures were not feasible due to the poor solubility of CoPS. Acetonitrile also
proved convenient due to its broad electrochemical window. Under N2 atmosphere, the complex
41

displays one oxidation feature with E1/2 at –0.21 V, and two reductive features at –2.08 V, and –
2.36 V, respectively (Figures 2.8 and 2.9, all potentials are referenced versus Fc
0/+
). The features
at -2.08 V and -2.36 V appear quasi-reversible, and no change in the reduction event at -2.08 V is
observed if the potential is reversed before reaching the second reduction. While the features at -
0.21 V and -2.08 V are similar in size, the feature at -0.21 V appears smaller indicating less charge
passed (Figure 2.9). This observation suggests the formation of a minor species following the
reduction event at -2.08 V. Based on the predicted molecular orbital diagram (Figure 2.7), the
anodic feature at -0.21 V is assigned to the one-electron oxidation of CoPS to [CoPS]
+
, which
results in the depopulation of the mixed metal-ligand SOMO. Following this event, the cathodic
feature at -2.08 V is assigned to the one-electron reduction of CoPS to [CoPS]
–
, resulting in a
diamagnetic complex with the occupation of the mixed metal-ligand β LUMO level. The complex
behavior observed for the feature at -2.36 V can be rationalized by the predicted character of the
next-lowest lying orbital for the square planar configuration of CoPS. As shown in Figure 2.7, a
two-electron reduction of CoPS to [CoPS]
2–
is predicted to result in the occupation of an α orbital
with σ
*
(Co-S) character. The population of this orbital can be expected to result in a change in
geometry at the cobalt center, possibly leading to metal-ligand bond elongation, bond cleavage, or
isomerization away from a square planar environment. The cathodic feature at -2.36 V can be
tentatively assigned to the one-electron reduction of the species resulting from this reduction-
initiated chemical change.

42

Figure 2.10 Cyclic voltammograms of 1 mM CoPS in an acetonitrile solution under N2
atmosphere with 0.1 M [nBu4N][PF6] and 1 mM ferrocene at various scan rates.

Figure 2.11 Randles-Sevcik analysis based on variable scan rate experiments. Redox features are
labelled by formal oxidation state of cobalt.

To further probe the electrochemical behavior of this system, cyclic voltammetry
experiments were performed at various scan rates (Figure 2.10). The feature at -0.21 V appears
reversible at all scan rates, as confirmed by Randles-Sevcik analysis (see Figure 2.11). More
complex behavior is observed for the subsequent two events at -2.08 V and -2.36 V. At low scan
rates, both reduction events are irreversible. Increasing the scan rate causes the first feature (at -
2.08 V) to exhibit reversibility while diminishing the second redox event. This behavior suggests
that a chemical event follows the first reduction of the complex. This chemical step occurs more
slowly than the CV timescale at fast scan rates. The feature at -2.36 V also approaches the size of
43

the features at -0.21 V and -2.08 V as the scan rate is decreased. This observation indicates that
the feature at -2.36 V may correspond to a new species generated following reduction of the
complex at -2.08 V, possibly an isomer. As the scan rate is decreased, a longer timescale is allowed
for the generation of this new species to occur and for reduction of this species to be detected as a
feature in the CV profile. An additional oxidation feature appears at -0.56 V following reduction,
indicating the formation of a new species following reduction of the complex.

Figure 2.12 Calculated geometries for [CoPS]
0
, [CoPS]
–
, and [CoPS]
2–
at the 6-31G(d)/PBE level
of theory. All geometries were confirmed as stable minima with frequency calculations at the same
level of theory.

Figure 2.13 Calculated spin densities for the paramagnetic species, [CoPS]
0
and [CoPS]
-2
, at the
def2-TZVP/PBE level of theory. Both species (rows) are depicted in three different views
(columns) for clarity.

To probe this behavior, gas-phase geometry optimizations were performed at the 6-
31G(d)/PBE level of theory for the one- and two-electron-reduced forms of the complex. The one-
44

electron reduced form, [CoPS]
–
, is predicted to be diamagnetic with a singlet ground state and
isomerizes towards a tetrahedral geometry upon relaxation, as shown in Figure 2.12. This result
indicates that isomerization of the complex is predicted to be more energetically favorable than
metal-ligand bond elongation in a square planar geometry or metal-ligand bond cleavage. An
additional one-electron reduction generates the doubly-anionic species, [CoPS]
2–
, which is
predicted to be paramagnetic with a doublet ground state. This complex distorts further towards a
tetrahedral geometry, and the spin density for this species is predicted to localize predominantly
on the metal center (Figure 2.13). Based on these calculations, the quasi-reversibility of the
observed reduction features is attributed to this isomerization from a square planar geometry
towards a tetrahedral geometry at the cobalt center. Further, the localization of spin density
predicted by DFT for the reduced species predicts that the first reduction (E1/2 = -2.08 V) is mixed
metal- and ligand-based, while the second reduction event (E1/2 = -2.36 V) appears predominantly
metal-based.

Figure 2.14. Cyclic voltammograms of 1 mM CoPS in acetonitrile solutions with 0.1 M
[nBu4N][PF6] and scan rate = 100 mV/s under N2 (black), CO2 (red), and under CO2 in the presence
of either 0.5 M PhOH (orange), 0.5 M TFE (green), or 0.5 M H2O (blue).

45

Figure 2.15. Cyclic voltammograms of 1 mM CoPS in an acetonitrile solution with 0.1 M
[nBu4N][PF6] electrolyte with scan rate = 100 mV/s under N2 atmosphere (black) and CO2 (red).

Figure 2.16 Cyclic voltammograms of 1 mM CoPS in an acetonitrile solution with 0.1 M
[nBu4N][PF6] electrolyte with scan rate = 250 mV/s under N2 atmosphere (black) and under CO2
with 0.5 M H2O (blue).

46

Figure 2.17 Cyclic voltammetry of 1mM CoPS in acetonitrile under CO2 atmosphere with 0.1 M
[nBu4N][PF6] supporting electrolyte at various scan rates.

Upon switching the atmosphere from N2 to CO2, an irreversible increase in current density
is observed at a potential corresponding to the first reduction event (E1/2 = -2.08 V), as shown in
Figure 2.14 and Figure 2.15. This suggests reduction-induced reactivity with CO2. Upon scanning
to more cathodic potentials, a plateau-like shape is reached at –2.18 V. On the return scan, a small
return oxidation feature is present at -1.58 V. This return feature does not appear under N2 alone,
which suggests that it may indicate the oxidation of an intermediate in the CO2 reduction cycle.
Larger current densities are measured in the presence of Brønsted acid additives, including phenol
(PhOH), 2,2,2-trifluoroethanol (TFE), or H2O (Figure 2.14). While experiments in the presence
of PhOH and TFE result in pseudo-plateau waveshapes at 100 mV/s, faster scan rates are required
to reach plateau currents in the presence of water (Figure 2.16). This pseudo-plateau region has
previously been attributed to entering a kinetic regime.
39
To further probe the behavior of this
complex in the presence of CO2, variable scan rate experiments were conducted in the absence of
an acid additive. These studies indicate a change in waveshape from a plateau shape to a peak
shape (Figure 2.17) at faster scan rates. This behavior suggests an “EC”-type mechanism, where
a chemical step (C) follows an electrochemical reduction step (E) and has insufficient time to
proceed as the scan rate increases. This result further suggests that pre-association of CO2 to the
47

neutral CoPS complex (a “CE”-type mechanism) is unlikely, as a reduction event appears to
precede chemical steps.

Figure 2.18. Controlled potential electrolysis traces measured under CO2 atmosphere. In all cases
a solution of 1 mM CoPS in acetonitrile (with 0.1 M [nBu4N][PF6] supporting electrolyte) was
held at a potential of -2.18 V for 1 hour in the absence of acid (red), presence of 0.5 M PhOH
(orange), presence of 0.5 M TFE (green), or presence of 0.5 M water (blue).

Table 2.6 Summary of results from CPE studies.
Acid

Total FE (%) H2:CO iCPE (mA/cm
2
)
[a]
TONCPE
[b]

None - - 1.1 -
PhOH >99 4.1±0.5 4.8 9.06  10
5

TFE 91±9 10.2±1 2.9 3.21  10
5

H2O 99±7 2.7±0.3 1.6 1.04  10
5

[a] Average current density measured over 1 hour. [b] Details in Experimental Methods

Table 2.7 Summary of CPE results under CO2. All data presented are averaged quantities of
multiple experiments, as reflected in the determined error bars. Charge passed was determined by
integration of the CPE trace, and gaseous products were determined by gas chromatography.

Potential
(V)
Duration
(h)
Charge (C) mmol H2 mmol CO
TOFCPE (s
-1
)
No acid -2.18 1 10.23±0.51 - - -
PhOH -2.18 1 43.10±2.1 1.83±0.2E-04 4.58±0.4E-05 250(25)
TFE -2.18 1 25.66±1.3 1.10±0.1E-04 1.05±0.1E-05 89(9)
H2O -2.18 1 14.71±0.74 5.48±0.6E-05 2.02±0.2E-05 29(6)
H2O -2.18 3 35.88±1.8 0.144±0.01 0.0436±0.004 24(5)

48

Table 2.8 Summary of FE calculations from CPE experiments under CO2. All data presented are
averaged quantities of multiple experiments, as reflected in the determined error bars. Charge
passed was determined by integration of the CPE trace, and gaseous products were determined by
gas chromatography.

Potential
(V)
Duration
(h)
FE (H2) FE (CO) FE (Total) H2/CO
No acid -2.18 1 - - - -
PhOH -2.18 1 82±8% 20±2% >99% 4.1±0.5
TFE -2.18 1 83±8% 8±1% 91±9% 10.2±1
H2O -2.18 1 72±7% 29±3% 99±8% 2.7±0.3
H2O -2.18 3 75±7% 23±2% 98±7% 3.3±0.5

Figure 2.19 Summary of CPE results illustrating Faradaic efficiencies for H2 (blue) and CO (red)
production.

Figure 2.20 TONtotal presented as a function of the pKa of the proton source. All pKa estimates
based on literature precedent.
10

49

Figure 2.21 Summary of H2:CO ratio calculations from CPE experiments. All data presented are
averaged quantities of multiple experiments, as reflected in the determined error bars. Gaseous
products were determined by gas chromatography. All pKa estimates based on literature
precedent.
10

Following CV studies, controlled potential electrolysis (CPE) experiments were performed
with CoPS under CO2 atmosphere to determine product selectivity and catalyst stability (Figure
2.18). Results for these studies are summarized in Tables 2.6-2.8 and Figures 2.19-2.21. In a
typical experiment, a glassy carbon working electrode was held at a potential of –2.18 V for one
hour in a 1 mM solution of CoPS. A sample of the headspace was collected via syringe and
analyzed by gas chromatography for gaseous products. TONCPE and TOFCPE were calculated based
on established equations (see Experimental Methods). These calculations account for the diffusion
kinetics of a homogeneous molecular species, and assume that only the concentration of catalyst
in the diffusion layer near the electrode is active. Under CO2 atmosphere in the absence of additive,
a stable current density of 1.1 mA/cm
2
was measured for the duration of the experiment. Neither
H2 nor CO were detected, consistent with the proton-dependence of these reactions. Additional
products were also tested for, including formate, methane, methanol, carbonate, or other organics
(acids or aldehydes), but none of these species were detected by either NMR spectroscopy or gas
chromatography measurements. Prior studies related to an analogous nickel complex with both
50

thiolate and phosphine donors suggest reduction-initiated decomposition may proceed in the
absence of substrate.
40
Additionally, no products were detected under N2 in the absence of an acid
additive.
In the presence of Brønsted acid additives (PhOH, TFE, or H2O), increased currents were
measured with a concurrent increase in the quantity of gaseous products and total FE. The addition
of 0.5 M PhOH generates a stable current density of 4.8 mA/cm
2
. A H2:CO ratio of 4.1±0.5 in the
presence of phenol was measured, with a total FE of >99% (Figure 2.19). In the presence of 0.5
M TFE, an average current density of 2.9 mA/cm
2
was observed, with a H2:CO ratio of 10.2±1
and a total FE of 91±9%. Upon addition of 0.5 M H2O, a current density of 1.6 mA/cm
2
was
measured with a total FE of >99% and a H2:CO ratio of 2.7±0.3. It was observed that the choice
in Brønsted acid impacts both the measured TONCPE and the H2:CO ratio of the resulting syngas
mixture, with the highest preference for CO production measured in the presence of H2O (Figures
2.20-2.21). The measured TONCPE is observed to increase as a function of increasing acid pKa
(Figure 2.20), with the highest TONCPE of 9.06  10
5
measured in the presence of phenol. The
high FE and favorable H2:CO ratio measured with water could be maintained for a 3-hour CPE
experiment (Tables 2.7-2.8). Notably, this time-dependent product quantification indicates
negligible change in the H2:CO ratio produced over time, as both the 1 hour and 3 hour
measurements in the presence of water generate a similar H2:CO ratio. Neither H2 nor CO was
detected in the absence of catalyst, and no formate or additional products (including methane,
methanol, carbonate, or other organics) were detected under any of the conditions studied. As the
total FE is near unity in all cases, the absence of additional side products is consistent with the
quantities of H2 and CO detected. No H2 or CO was detected under N2, and upon the addition of
acid only H2 was detected in all cases. Additionally, higher currents were generated in the presence
51

of an acid additive under N2 relative to the currents measured under CO2 atmosphere (see Figure
2.22 and Tables 2.9-2.10). This suggests a competition exists between CO2 coordination and
protonation, which hinders HER in the presence of CO2. Rinse tests were performed following
each experiment to confirm that the solubilized complex was responsible for the observed activity
(Figure 2.23). Lower current densities with sharp onsets were observed in the absence of catalyst,
consistent with low background activity for hydrogen evolution by the glassy carbon working
electrode at the operating potential of catalysis (Figure 2.24). Catalytic parameters were calculated
based on these CPE studies, which indicate a maximum TOFCPE of 250(25) s
-1
in the presence of
PhOH (Table 2.7). For all acid additives studied, TONCPE > 1  10
5
were measured (Table 2.6).

Figure 2.22 Controlled potential electrolysis traces measured under nitrogen atmosphere. In all
cases a solution of 1 mM CoPS in acetonitrile (with 0.1 M [nBu4N][PF6] supporting electrolyte)
was held at a potential of -2.18 V for 1 hour in the absence of acid (red), presence of 0.5 M PhOH
(orange), presence of 0.5 M TFE (green), or presence of 0.5 M water (blue).

52

Table 2.9 Summary of CPE results under nitrogen atmosphere. All data presented are averaged
quantities of multiple experiments, as reflected in the determined error bars. Charge passed was
determined by integration of the CPE trace, and gaseous products were determined by gas
chromatography.

Potential
(V)
Duration
(h)
Charge (C) mmol H2 mmol CO
No acid -2.18 1 2.72±0.2 - -
PhOH -2.18 1 94.66±4.7 5.34±0.5E-04 -
TFE -2.18 1 38.62±1.9 2.08±0.2E-04 -
H2O -2.18 1 3.42±0.74 4.67±0.6E-05 -

Table 2.10. Summary of results from CPE experiments under nitrogen atmosphere. All data
presented are averaged quantities of multiple experiments, as reflected in the determined error
bars. Charge passed was determined by integration of the CPE trace, and gaseous products were
determined by gas chromatography.

Potential
(V)
Duration
(h)
FE (H2)
No acid -2.18 1 -
PhOH -2.18 1 >99%
TFE -2.18 1 >99%
H2O -2.18 1 26±2%

Figure 2.23 Rinse test traces measured under catalytic conditions. In all cases, the CPE solutions
were removed from the electrolysis cell by syringe and the cell was rinsed with acetonitrile. A
typical CPE blank trace in the absence of catalyst is shown for comparison (black). Slightly higher
currents are generated than in the blank experiment, due to minor concentrations of solubilized
catalyst, though no detectable products were generated.

53

Figure 2.24 Cyclic voltammetry of a glassy carbon working electrode in acetonitrile with 0.1 M
[nBu4N][PF6] supporting electrolyte (scan rate = 100 mV/s) in the presence of various acid
additives at a concentration of 0.5 M.

Scheme 2.1 Mechanistic hypothesis for observed reactivity.

54

Figure 2.25 UV-Vis study of 1 mM CoPS in acetonitrile solution (blue), protonation study with 3
eq trifluoroethanol (red), and subsequent deprotonation study with excess triethylamine (black).

Figure 2.26
1
H-NMR spectra of 1 mM CoPS in acetonitrile-d3 without acid additive (red), with
0.5 M PhOH (orange), with 0.5 M TFE (green), and with 0.5 M H2O (blue). Pink asterisks mark
characteristic
1
H-NMR signatures of CoPS, and black asterisks mark the corresponding shifted
features associated with protonation.

55

Table 2.11 Thermodynamic potential for CO2 reduction to CO in the presence of various proton
sources, and overpotential associated with CoPS.
Acid pKa
E°
(CO2/CO)
V vs Fc/Fc
+

η (V)
None N/A –0.12 2.06
H2CO3 17.03 –1.13 1.05
TFE 20.55 –1.33 0.85
PhOH 29.1 –1.84 0.34

Based on the experimental observations, a mechanistic hypothesis can be constructed to
model the measured reactivity, as shown in Scheme 2.1. Based on electrochemical studies, no
reactivity is observed between CO2 and CoPS prior to reduction, suggesting that an initial CO2-
binding step to the neutral species is unfavorable. In contrast, a color change was observed upon
treatment of an acetonitrile solution of CoPS with trifluoroacetic acid (TFA), with a corresponding
return to the original UV-Vis spectrum upon the addition of triethylamine base (Figure 2.25). This
result suggests that the neutral complex may undergo reversible protonation prior to reduction. As
TFA is a strong acid with a pKa of 12.8 in acetonitrile,
1
H-NMR spectra were collected in
acetronitrile-d3 in the presence of 0.5 M phenol, TFE, and H2O to test whether protonation is
facilitated by these acids (Figure 2.26). In the presence of 0.5 M TFE, additional resonances are
observed which we associate with the formation of a protonated species. As these resonances are
not observed in the presence of phenol or H2O, the results of this study suggest that only TFE is
capable of protonating the neutral complex prior to reduction. These results are consistent with the
measured H2:CO ratios as TFE is observed to generate the highest H2:CO ratio of 10.2±1. Due to
the bulkiness of the phosphine ligands and the literature precedent established for cobalt
bis(dithiolene) and bis(diselenolene) catalysts, we propose that this protonation event takes place
at the thiolate ligand. This is further predicted by our DFT results, which indicate non-negligible
thiyl-radical character. The presence of strong acids is expected to follow this protonation-first
56

pathway (“CE”-type mechanism), while weaker or sterically-bulky acids may follow a reduction-
first pathway (“EC”-type mechanism). As such, multiple competing mechanistic pathways may be
available. As CPE studies are conducted at a lower potential than the second reduction of the
complex, a second reduction step (E) prior to chemical step (C) is ruled out, further suggesting
EC- rather than EE-type activation. Following reduction of the neutral complex, a competition
exists between CO2-adduct formation and protonation. This step is expected to determine the
overall selectivity of the catalyst, as binding of either substrate allows the catalyst to enter either
the CO2RR or HER cycles. Following liberation of CO or H2 to regenerate the neutral species, the
complex can either receive another electron from the electrode to regenerate the reduced complex
or become protonated to regenerate the protonated species.
The thermodynamic potentials for the two-electron reduction of CO2 to CO in acetonitrile
in the presence of various acid additives are also provided in Table 2.11, alongside the calculated
overpotentials.
41
Also presented are the calculated overpotentials, as determined based on an
applied potential of -2.18 V for all studies. As carbonic acid is formed by the reaction of CO2 with
water, the overpotential ( η) for CO2 reduction in the presence of carbonic acid is also provided
for comparison. Additionally, the operation of concerted proton-coupled electron transfer
pathways is suggested by cyclic voltammetry studies, and more complex mechanistic schemes are
therefore likely contributing to the observed reactivity. Ongoing stoichiometric studies are in work
to further distinguish between these competing pathways.

2.4 CONCLUSION
In conclusion, we report here a cobalt bis(2-[diisopropylphosphaneyl]benzenethiol)
complex, CoPS, exhibiting ligand-non innocence and activity towards electrocatalytic syngas
production. Electronic structure calculations predict non-negligible ground state thiyl-radical
57

character, and electronic spectroscopy reveals a low-energy charge transfer transition. In the
presence of 0.5 M phenol, the highest catalytic activity is observed with FE of >99%, and H 2:CO
ratio of 4.1±0.5. Excellent FE (>99%) and a favorable H2:CO ratio (2.7±0.3) were also measured
for the co-electrolysis of CO2 and H2O. For all acid additives studies, TONCPE > 1  10
5
were
measured. A mechanistic hypothesis is proposed, in which pKa-dependent protonation by external
acid and competitive binding between CO2 and protons determines overall selectivity.
Furthermore, the modular nature of this ligand system enables the potential for broad tunability at
the phosphine site for future studies to optimize the syngas mixture produced.

2.5 EXPERIMENTAL METHODS
2.5.1 Materials and Synthesis
All manipulations of air- and moisture-sensitive materials were conducted under nitrogen
atmosphere in a Vacuum Atmospheres glovebox or on a dual manifold Schlenk line with oven-
dried glassware. Water was deionized with the Millipore Synergy system (18.2 MΩ·cm). All other
solvents used were degassed with nitrogen, passed through activated alumina columns, and stored
over 4Å Linde-type molecular sieves. Proton NMR spectra were acquired using a Varian 400-MR
2-Channel spectrometer at room temperature and referenced to the residual
1
H resonances of the
deuterated solvent (
1
H: benzene-d6, δ 7.16 ppm). The desired ligand was synthesized according to
a reported literature procedure.
35
CoPS was synthesized using an analogous procedure to those
reported for similar complexes.
35
All other chemical reagents were purchased from commercial
vendors and used without further purification. Elemental analyses were performed by Robertson
Microlit Laboratories, Ledgewood, New Jersey.

58

2.5.2 Synthesis of Co(PiPr2SC6H4)2 (CoPS)
The desired 2-(diisopropylphosphaneyl)benzenethiol ligand was prepared according to the
literature procedure.
35
In an inert-atmosphere glovebox, a Schlenk flask was charged with n-BuLi
(15.8 mL), tetramethylethylenediamine (5.9 mL), and cyclohexane (12.0 mL). This reaction
mixture was transferred to a Schlenk line and cooled to 0 °C. A separate flask was charged with
thiophenol (1.9 mL) and cyclohexane (4.8 mL), sparged with N2, then cooled to 0 °C. The
thiophenol solution was transferred dropwise via cannula to the Schlenk flask charged with n-
BuLi, and the reaction mixture was maintained at 0 °C for 30 min. The reaction mixture was then
allowed to warm to room temperature with stirring over-night. After overnight reaction the
dilithiated product, Li2SC6H4, was obtained as a white suspension, which was subsequently
isolated by air-free filtration. The resulting solid was washed with three portions of 10 mL pentane
before drying under reduced pressure for one hour to afford a white solid. The white solid was
dissolved in THF (20.0 mL), yielding a red-orange solution. A separate flask was charged with
iPr2PCl (2.0 mL) and THF (2.0 mL), and both flasks were cooled to -78 °C. Once cooled, the
phosphine solution was transferred dropwise via cannula to the red-orange solution of Li2SC6H4.
The reaction mixture was maintained at -78 °C for 30 min., before it was allowed to warm to room
temperature with stirring overnight. Following overnight stirring, a yellow solution was observed,
which was acidified with a N2-sparged solution of HCl (1.5 mL) in DI water (18.0 mL) at 0 °C.
The organic layer of the resulting biphasic solution was transferred to a clean flask via cannula
and the aqueous portion was extracted with three portions of 5 mL diethyl ether. The combined
organics were dried with MgSO4 and isolated by air-free filtration. The resulting yellow-brown
solution was reduced to a brown oil under vacuum and purified by vacuum distillation. The
distilled product was diluted with methanol (8 mL) and treated with triethylamine (0.7 mL). Excess
59

anhydrous cobalt(II) chloride (580 mg) was added in one portion to the pale-yellow ligand
solution, causing an immediate color change to vivid yellow. This solution was allowed to stir
overnight. The crude CoPS product was collected by filtration and dried under reduced pressure.
Crystals suitable for crystallographic and electrochemical characterization were prepared by
recrystallization in air through layering of methanol over a concentrated solution of CoPS in
benzene at 6 °C. (Yield: 80%)
1
H NMR (400 MHz, benzene-d6) δ 24.9 (s, 2H, PhH), 11.2 (s, 2H,
PhH), 4.8 (br s, 12H, -CH(CH3)2), 2.5 (br s, 4H, -CH(CH3)2), -0.5 (br s, 12H, -CH(CH3)2), -16.1
(s, 2H, PhH), -22.5 (s, 2H, PhH). Anal. calcd (C24H36CoP2S2): C, 56.57; H, 7.12; Co, 11.56; Found:
C, 56.38; H, 6.89; Co, 10.88.

2.5.3 X-Ray Crystallography
A clear orange prism-like specimen was mounted for the X-ray crystallographic analysis,
with approximate dimensions of 0.198 mm  0.214 mm  0.368 mm. The X-ray intensity data were
measured on a Bruker APEX DUO system equipped with a fine-focus tube (MoKα, λ = 0.71073
Å) and a TRIUMPH curved-crystal monochromator, with a theta range of 1.96 to 30.47°. A total
of 2520 frames were collected during a 7 hour exposure time. The frames were integrated with the
Bruker SAINT software package using a SAINT V8.38A (Bruker AXS, 2013) algorithm. Data
were corrected for absorption effects using the multi-scan method (SADABS). The SHELXTL XT
2014/5 (Bruker AXS, 2014) Software Package was used to determine the structure solution with
direct methods.
11
The SHELXTL XT 2014/5 (Bruker AXS, 2014) Software Package was used for
refinement by full-matric least-squares on F2.
49
Additional details are provided in Table 2.1.

60

2.5.4 Electrochemistry
Electrochemistry experiments were carried out in acetonitrile solution with 0.1 M
[nBu4N][PF6] electrolyte using a Pine potentiostat. All cyclic voltammetry and controlled potential
electrolysis experiments were carried out with a three electrode configuration in an
electrochemical H-cell under a nitrogen or CO2 atmosphere using glassy carbon working and
counter electrodes. The working and counter compartments of the cell were separate by a porous
glass frit. Working and counter electrodes were polished using Micropolish powder (purchased
from CH Instruments, Inc.) prior to electrochemical studies. Six-sided polished glassy carbon
electrodes with working areas of 2.5 cm
2
were used for all studies. All current densities presented
are normalized to this geometric surface area. The pseudo-reference electrode used was a silver
wire purchased from VWR and was isolated from solution with a glass capillary and Vycor frit.
All experiments were referenced relative to ferrocene (Fc) with the Fc
+
/Fc couple defined as 0 V.
Ohmic drop was compensated using positive feedback compensation as implemented in the
Aftermath software suite.

2.5.5 Nuclear Magnetic Resonance
Proton NMR spectra were acquired using a Varian S-2 400-MR 2-Channel spectrometer at
room temperature and referenced to the residual
1
H resonances of the deuterated solvent (
1
H:
benzene-d6, δ 7.16 ppm; acetonitrile-d3, δ 1.94 ppm). Evan’s method was applied according to
literature precedent.
1
A 1 mM solution of catalyst in benzene-d6 was prepared and a sealed
capillary containing protio-benzene was inserted into the NMR sample tube. The splitting between
the solvent peaks was measured with a
1
H NMR spectroscopy study and reported equations were
applied to calculate μeff.
1

61

2.5.6 Electrochemical Analysis
Calculation of TOFCPE (Eq. 2.1), and TONCPE (Eq. 2.2) were performed according to
established protocol.
1
This method assumes Nernstian electron transfer to the catalyst. As this
method does not directly consider the quantity of gaseous products generated, the resulting TOFCPE
and TONCPE represent the overall activity towards two-electron-reduced products (both CO and
H2). In Eq. 2.1, i represents the average current during CPE (i = Charge*F.E./time, C/s), F
represents Faraday’s constant (F = 96 485 C/mol), A represents the surface area of the working
electrode (2.5 cm
2
), D represents the diffusion coefficient (determined to be 1 × 10
-5
cm
2
/s based
on Randles-Sevcik analysis of the oxidation feature at -0.21 V) and [cat] represents the bulk
catalyst concentration ([cat] = 1 mM = 1 × 10
-6
mol/cm
3
). In Eq. 2.2, t represents the duration of
the electrolysis experiment (s
-1
).

𝑇 𝑂 𝐹 𝐶𝑃𝐸 =
𝑖 2
𝐹 2
𝐴 2
𝐷 [ 𝑐𝑎 𝑡 ]
2
Eq. 2.1
𝑇 𝑂 𝑁 𝐶𝑃𝐸 = 𝑇 𝑂 𝐹 𝐶𝑃𝐸 × 𝑡 Eq. 2.2

Wash tests were performed by removing the catalyst solution from the H-cell via syringe
under a positive pressure of CO2, and rinsing the working cell three times with acetonitrile (Figure
2.14). The cell was maintained under 1 atm of CO2, and the electrode was not removed from the
cell during these washings to prevent O2-exposure. As such, a small amount of catalyst remains in
the cell despite these washings, due to the presence of small amounts of catalyst trapped in the
pores of the glass frit that separates the counter and working compartments of the electrolysis cell.
The additional current measured during the wash tests relative to the blank CPE study is
62

representative of this small amount of solubilized catalyst. No visible material is present on the
electrode following CPE, and much lower current densities are observed in the wash test relative
to any of the preparative CPE studies. As such, these studies strongly suggest that the observed
reactivity is due to solubilized catalyst, as no gaseous products are observed following three wash
cycles.

2.5.7 Gas Chromatography
Gaseous products were quantified using a Shimadzu GC-2010-Plus instrument equipped
with a BID detector and a Restek ShinCarbon ST Micropacked column. In a typical experiment, 2
mL of gas were withdrawn from the headspace of the electrochemical cell with a gas-tight syringe
and injected into the instrument. Calibration plots were prepared with multiple injections of syngas
standards purchased from Praxair, Inc. Faradaic efficiencies were determined by dividing the
amount of gaseous product produced as measured by gas chromatography by the amount of gas
expected based on the total charge measured during controlled potential electrolysis. Multiple runs
were performed for each condition studied, leading to similar behavior. The reported μmol of gas
produced (and subsequently FE and TON) are averaged values and error bars are determined from
multiple experiments.

2.5.8 Product Detection
Detection of formate was performed according to literature precedent.
2
Following CPE
studies, a 5 mL aliquot of each electrolysis solution was collected and extracted with 2 mL D2O.
The aqueous portion was acidified with one drop HCl and tested for formate by
1
H NMR
spectroscopy. Formate was not detected in any of the experiments performed. Detection of
63

methanol, aldehydes, and other alcohols was carried out through NMR spectroscopy studies, and
detection of methane, carbon monoxide, and hydrogen was carried out through gas
chromatography.

2.5.9 UV-Vis Spectroscopy
UV-Vis spectra were collected using a UV-1800 Shimadzu UV spectrophotometer.
Samples were studied in transmittance mode with a 1 cm quartz cuvette, and the spectrum
measured for a blank benzene or acetonitrile sample was subtracted as background.

2.5.10 Computational Methods
All calculations were run using the Q-CHEM program package.
42
Geometry optimizations
were run with unrestricted DFT calculations at the PBE level of theory with the Pople 6-31G(d)
basis set for all atoms.
43-46
All optimized geometries were verified as stable minima with frequency
calculations at the same level of theory. The PBE functional was used throughout this study, as it
provided minimal spin contamination for the reduced complexes. The range-separated hybrid ω-
B97xD functional as well as the PBE0 and BP86 functionals were also tested, but all yielded >20%
spin contamination for the singly- and doubly-reduced complexes. Single point energy calculations
were run with a larger triple-zeta def2-TZVP basis set for all atoms.
7
Time-dependent density
functional theory calculations were run at the B3LYP level of theory with the def2-TZVP basis set
for all atoms. Kohn-Sham orbital images are presented with isovalues of 0.05 for clarity. Basis sets
were retrieved from the Basis Set Exchange.
47-48

64

2.5.11 Nuclear Coordinates for Optimized Geometries
Table 2.12 Nuclear coordinates for [CoPS]
0
.
I Atom X Y Z
1 P -0.0581365 1.4105284 -1.6631213
2 C 0.3164792 1.0575953 -4.51082
3 H 0.1488916 2.1321935 -4.6482454
4 C 0.5561038 0.2505782 -5.6317084
5 H 0.5856512 0.6945876 -6.6317293
6 C 0.7569329 -1.1289419 -5.460891
7 H 0.9479966 -1.764518 -6.3323773
8 C 0.7139462 -1.702368 -4.1847644
9 H 0.8682751 -2.7782047 -4.0528856
10 C 0.4740188 -0.8922299 -3.0571127
11 S 0.4105619 -1.6091933 -1.4245911
12 Co 0.1668392 0.0006665 0.0056878
13 C 0.4578328 0.8980962 3.0690565
14 C 0.6861136 1.7112492 4.1969762
15 H 0.8309909 2.7883674 4.0648538
16 C 0.7297138 1.139135 5.4736519
17 H 0.9115288 1.7770507 6.3454083
18 C 0.5413317 -0.2421275 5.6446789
19 H 0.5715759 -0.6851637 6.6451097
20 C 0.3133115 -1.052128 4.5235305
21 H 0.1554057 -2.1281712 4.6611363
22 C 0.275197 -0.4896463 3.2354209
23 P -0.050695 -1.4101178 1.6749594
24 C -1.8200298 2.0573828 -1.9161794
25 H -1.7236667 2.8721307 -2.662839
26 C 1.0614954 2.9121686 -1.8884827
27 H 0.943786 3.1785149 -2.9572683
28 C -1.807954 -2.0706887 1.9249585
29 H -1.7067185 -2.8831896 2.6734241
30 C 1.0803098 -2.9023643 1.9046763
31 H 0.9601303 -3.1697412 2.9729383
32 C 0.2794222 0.4939567 -3.2232109
33 C -2.7230346 0.9558366 -2.4951662
34 H -2.7752016 0.089405 -1.811348
35 H -3.7491932 1.3468055 -2.6203866
36 H -2.3699592 0.5933666 -3.4731029
37 C -2.4273356 2.6199642 -0.6209783
38 H -3.3775961 3.1352962 -0.8515217
39 H -2.6481923 1.8034141 0.0874017
40 H -1.7647788 3.3272921 -0.0994658
65

41 C 0.6882698 4.141969 -1.0502235
42 H -0.3137952 4.5287332 -1.2999033
43 H 0.721977 3.9250455 0.029654
44 H 1.4129737 4.9517715 -1.2534997
45 C 2.5172071 2.4760621 -1.6567297
46 H 2.7814054 1.6009621 -2.2745061
47 H 3.2051222 3.30047 -1.9188195
48 H 2.6842685 2.2122912 -0.5987827
49 S 0.3949092 1.6131884 1.43573
50 C -2.7218953 -0.9759475 2.4995603
51 H -3.7443797 -1.3764285 2.624753
52 H -2.3732036 -0.6073236 3.4767555
53 H -2.7816791 -0.1122938 1.8128882
54 C -2.4069806 -2.6400459 0.6289247
55 H -3.3546051 -3.1613453 0.8569565
56 H -2.6306058 -1.8258579 -0.081225
57 H -1.7379408 -3.3438015 0.1108923
58 C 0.7213341 -4.1354495 1.0648952
59 H -0.2792311 -4.5294599 1.3090511
60 H 0.7595295 -3.919387 -0.015028
61 H 1.4508609 -4.9396757 1.2730424
62 C 2.533256 -2.4537759 1.6792806
63 H 2.7028323 -2.1882454 0.6221722
64 H 2.7875224 -1.5766464 2.298269
65 H 3.2271377 -3.272259 1.9441539
Nuclear Repulsion Energy = 4607.7682230237 hartrees

Table 2.13 Nuclear coordinates for [CoPS]
–
.

I Atom X Y Z
1 P -0.0303905 1.3400189 -1.5814805
2 C -0.1645212 0.9598697 -4.4655509
3 H -0.5402877 1.9879027 -4.5610368
4 C 0.0294181 0.1900945 -5.6237588
5 H -0.1896305 0.6102096 -6.6120458
6 C 0.5129415 -1.1269951 -5.4962236
7 H 0.6737158 -1.7386491 -6.392968
8 C 0.7848638 -1.666397 -4.2355824
9 H 1.1528393 -2.6945905 -4.1394429
10 C 0.5830394 -0.8982409 -3.0638602
11 S 0.9062815 -1.5531273 -1.4502872
66

12 Co 0.4578341 0.0047601 0.0082007
13 C 0.562781 0.9054808 3.0812964
14 C 0.7516125 1.6745328 4.2546081
15 H 1.1067907 2.7074308 4.1607993
16 C 0.4829271 1.1300373 5.5137663
17 H 0.6334549 1.7423551 6.4118242
18 C 0.0154177 -0.1931502 5.6381106
19 H -0.2013037 -0.6173133 6.6251925
20 C -0.1657096 -0.9636375 4.4783433
21 H -0.5296821 -1.9961383 4.5712305
22 C 0.1059728 -0.4329604 3.2058953
23 P -0.021661 -1.3376976 1.5941802
24 C -1.7635874 2.1136798 -1.6617508
25 H -1.7178458 2.9093025 -2.4349219
26 C 1.0949442 2.8186687 -2.0241815
27 H 0.8540248 3.0820205 -3.0737729
28 C -1.7492807 -2.1243012 1.6676635
29 H -1.7010491 -2.9197587 2.4408242
30 C 1.1131128 -2.8086332 2.0390427
31 H 0.8703307 -3.0746421 3.0875503
32 C 0.1097811 0.4342196 -3.1915557
33 C -2.8033319 1.0638844 -2.0759351
34 H -2.7774024 0.2178774 -1.3655151
35 H -3.8214086 1.4989255 -2.053709
36 H -2.6142851 0.6625757 -3.0843342
37 C -2.1597143 2.7373122 -0.3141419
38 H -3.0987152 3.3147595 -0.4234264
39 H -2.3254427 1.9463093 0.4374988
40 H -1.3832489 3.397404 0.099758
41 C 0.9231116 4.0714656 -1.1540201
42 H -0.0683704 4.5387566 -1.2852184
43 H 1.0539765 3.8366091 -0.0837984
44 H 1.6845867 4.8276299 -1.4302082
45 C 2.5415948 2.3029704 -1.9710178
46 H 2.6615319 1.3811678 -2.5659602
47 H 3.2456763 3.066252 -2.3563329
48 H 2.821327 2.0618592 -0.9305302
49 S 0.8812719 1.5668876 1.4693731
50 C -2.79716 -1.0810999 2.0779973
51 H -3.8130133 -1.5208125 2.0477449
52 H -2.6167564 -0.6819311 3.0888317
53 H -2.7701961 -0.2326573 1.3705631
54 C -2.1350555 -2.7503104 0.318281
55 H -3.0734319 -3.3297879 0.4219308
56 H -2.2967202 -1.9609566 -0.4360222
67

57 H -1.3540718 -3.40935 -0.0886414
58 C 0.9541844 -4.0624663 1.1677574
59 H -0.0333046 -4.5389181 1.2963318
60 H 1.085429 -3.8257974 0.0980191
61 H 1.7218708 -4.8118001 1.4453865
62 C 2.5561473 -2.2823219 1.9920765
63 H 2.8398793 -2.0416799 0.9525773
64 H 2.6658078 -1.3582615 2.5855578
65 H 3.2637382 -3.0395081 2.3829528
Nuclear Repulsion Energy = 4624.6743801470 hartrees

Table 2.14 Nuclear coordinates for [CoPS]
2–
.

I Atom X Y Z
1 P -0.1510368 1.0827326 -1.6374724
2 C 0.7097716 1.0307457 -4.4347808
3 H 0.0024507 1.853167 -4.6188167
4 C 1.4774011 0.5449273 -5.5085652
5 H 1.3805732 0.9855728 -6.5101617
6 C 2.3731231 -0.5169702 -5.272525
7 H 2.9836516 -0.9161469 -6.0957335
8 C 2.489105 -1.0694629 -3.9924044
9 H 3.1808803 -1.9025024 -3.8117051
10 C 1.7340217 -0.5754894 -2.8961252
11 S 1.8916188 -1.2747525 -1.2832844
12 Co 0.6018986 0.0761318 0.053662
13 C 1.5975346 0.6357106 3.05739
14 C 2.3269816 1.087734 4.1882019
15 H 2.9590203 1.9781185 4.0752793
16 C 2.2635076 0.4200607 5.4166593
17 H 2.8578759 0.7857909 6.2668463
18 C 1.4439489 -0.7161922 5.5662044
19 H 1.3911587 -1.2468975 6.5266263
20 C 0.6958705 -1.1588959 4.460079
21 H 0.0402136 -2.0348349 4.5771054
22 C 0.7661979 -0.5194179 3.208761
23 P -0.1820586 -1.0263911 1.6701495
24 C -1.8991134 0.7985361 -2.4377207
25 H -1.9038804 1.3091414 -3.4233782
68

26 C -0.0502536 2.996869 -1.8665999
27 H -0.6262808 3.3004314 -2.7689535
28 C -1.9327528 -0.7668133 2.4696726
29 H -1.9335877 -1.3008318 3.4430215
30 C -0.0977747 -2.9501205 1.770802
31 H -0.6893912 -3.3089374 2.6423417
32 C 0.8245116 0.5037155 -3.1348288
33 C -2.077279 -0.7075706 -2.6586583
34 H -2.0306547 -1.243003 -1.6915407
35 H -3.054377 -0.9297889 -3.1381442
36 H -1.2711897 -1.1169424 -3.2913269
37 C -3.038412 1.3504084 -1.5753717
38 H -4.0244791 0.9552251 -1.9002795
39 H -2.8827591 1.0524761 -0.5198003
40 H -3.0956868 2.4524222 -1.6112242
41 C -0.633706 3.6988084 -0.6297715
42 H -1.6808653 3.4131309 -0.4370245
43 H -0.0526912 3.4207748 0.2679801
44 H -0.5932866 4.803136 -0.7453795
45 C 1.41677 3.418048 -2.0390802
46 H 1.8498429 3.0375749 -2.9791896
47 H 1.5158508 4.5240469 -2.0279995
48 H 2.0196292 3.0123014 -1.2056026
49 S 1.6931667 1.4729835 1.5045094
50 C -2.1047788 0.7348482 2.7264186
51 H -3.0932238 0.9527076 3.1841923
52 H -1.313986 1.1180037 3.3936172
53 H -2.0221226 1.2971399 1.7764939
54 C -3.0725114 -1.2987131 1.5964442
55 H -4.057768 -0.9075105 1.9283795
56 H -2.9165192 -0.9814195 0.5470248
57 H -3.1319462 -2.4012492 1.6111595
58 C -0.6739075 -3.5421035 0.4738312
59 H -1.7446752 -3.3140958 0.3421931
60 H -0.1362974 -3.1143131 -0.3926534
61 H -0.5539328 -4.6463978 0.4555153
62 C 1.3613193 -3.4052465 1.9268566
63 H 1.9709702 -2.997626 1.1002886
64 H 1.8062369 -3.0589678 2.8745649
65 H 1.4338027 -4.5128964 1.8888107
Nuclear Repulsion Energy = 4577.4152397201 hartrees

69

2.6 REFERENCES
1. IEA. 2018 World Energy Outlook: Executive Summary. Oecd/Iea 2018, 11.
2. Nocera, D. G. Solar Fuels and Solar Chemicals Industry. Acc. Chem. Res. 2017, 50, 616–
619.
3. Detz, R. J.; Reek, J. N. H.; Van Der Zwaan, B. C. C. The Future of Solar Fuels: When
Could They Become Competitive? Energy Environ. Sci. 2018, 11, 1653–1669.
4. House, R. L.; Iha, N. Y. M.; Coppo, R. L.; Alibabaei, L.; Sherman, B. D.; Kang, P.;
Brennaman, M. K.; Hoertz, P. G.; Meyer, T. J. Artificial Photosynthesis: Where Are We
Now? Where Can We Go? J. Photochem. Photobiol. C Photochem. Rev. 2015, 25, 32–45.
5. Kim, D.; Sakimoto, K. K.; Hong, D.; Yang, P. Artificial Photosynthesis for Sustainable
Fuel and Chemical Production. Angew. Chemie Int. Ed. 2015, 54 (11), 3259–3266.
6. Graves, C.; Ebbesen, S. D.; Mogensen, M.; Lackner, K. S. Sustainable Hydrocarbon Fuels
by Recycling CO2 and H2O with Renewable or Nuclear Energy. Renew. Sustain. Energy
Rev. 2011, 15, 1–23.
7. U.S. Energy Information Administration – EIA. “New Methanol Plants Expected to
Increase Industrial Natural Gas Use through 2020” 2019, https://www.eia.gov/.
8. Olah, G. A. Beyond Oil and Gas: The Methanol Economy. Angew. Chem. Int. Ed. 2005,
44, 2636-2639.
9. Rostrup-Nielsen, J. R. Production of Synthesis Gas. Catal. Today 1993, 18, 305–324.
10. Qiao, J.; Liu, Y.; Hong, F.; Zhang, J. A Review of Catalysts for the Electroreduction of
Carbon Dioxide to Produce Low-Carbon Fuels. Chem. Soc. Rev. 2014, 43, 631-675.
11. Su, B.; Cao, Z.-C.; Shi, Z.-J. Exploration of Earth-Abundant Transition Metals (Fe, Co,
and Ni) as Catalysts in Unreactive Chemical Bond Activations. Acc. Chem. Res. 2015, 48,
886–896.
12. Collin, J. P.; Sauvage, J. P. Electrochemical Reduction of Carbon Dioxide Mediated by
Molecular Catalysts. Coord. Chem. Rev. 1989, 93, 245–268.
13. Barlow, J. M.; Yang, J. Y. Thermodynamic Considerations for Optimizing Selective CO 2
Reduction by Molecular Catalysts. ACS Cent. Sci. 2019, 5, 580–588.
14. Windle, C. D.; Perutz, R. N. Advances in Molecular Photocatalytic and Electrocatalytic
CO2 Reduction. Coord. Chem. Rev. 2012, 256, 2562–2570.
70

15. Francke, R.; Schille, B.; Roemelt, M. Homogeneously Catalyzed Electroreduction of
Carbon Dioxide - Methods, Mechanisms, and Catalysts. Chem. Rev. 2018, 118, 4631–
4701.
16. Zhou, X.; Micheroni, D.; Lin, Z.; Poon, C.; Li, Z.; Lin, W. Graphene-Immobilized fac-
Re(Bipy)(CO)3Cl for Syngas Generation from Carbon Dioxide. ACS Appl. Mater.
Interfaces 2016, 8, 4192–4198.
17. Kumar, B.; Smieja, J. M.; Sasayama, A. F.; Kubiak, C. P. Tunable, Light-Assisted Co-
Generation of CO and H2 from CO2 and H2O by Re(Bipy-tBu)(CO)3Cl and p-Si in Non-
Aqueous Medium. Chem. Commun. 2012, 48, 272–274.
18. (a) Wang, J. W.; Huang, H. H.; Sun, J. K.; Zhong, D. C.; Lu, T. B. Syngas Production with
a Highly-Robust Nickel(II) Homogeneous Electrocatalyst in a Water-Containing System.
ACS Catal. 2018, 8, 7612–7620. (b) Elgrishi, N.; Chambers, M. B.; Artero, V.; Fontecave,
M. Terpyridine Complexes of First Row Transition Metals and Electrochemical Reduction
of CO2 to CO. Phys. Chem. Chem. Phys. 2014, 16, 13635–13644.
19. Froehlich, J. D.; Kubiak, C. P. Homogeneous CO2 Reduction by Ni(Cyclam) at a Glassy
Carbon Electrode. Inorg. Chem. 2012, 51, 3932–3934.
20. (a) Chen, Z.; Concepcion, J. J.; Brennaman, M. K.; Kang, P.; Norris, M. R.; Hoertz, P. G.;
Meyer, T. J. Splitting CO2 into CO and O2 by a Single Catalyst. Proc. Natl. Acad. Sci.
2012, 109, 15606–15611. (b) Kang, P.; Chen, Z.; Nayak, A.; Zhang, S.; Meyer, T. J. Single
Catalyst Electrocatalytic Reduction of CO2 in Water to H2+CO Syngas Mixtures with
Water Oxidation to O2. Energy Environ. Sci. 2014, 7, 4007–4012. (c) Chen, Z.; Kang, P.;
Zhang, M. T.; Meyer, T. J. Making Syngas Electrocatalytically Using a Polypyridyl
Ruthenium Catalyst. Chem. Commun. 2014, 50, 335–337.
21. Eisenberg, R.; Gray, H. B. Noninnocence in Metal Complexes: A Dithiolene Dawn. Inorg.
Chem. 2011, 50, 9741–9751.
22. Ogata, H.; Nishikawa, K.; Lubitz, W. Hydrogens Detected by Subatomic Resolution
Protein Crystallography in a [NiFe] Hydrogenase. Nature 2015, 520, 571–574.
23. Chapovetsky, A.; Do, T. H.; Haiges, R.; Takase, M. K.; Marinescu, S. C. Proton-Assisted
Reduction of CO2 by Cobalt Aminopyridine Macrocycles. J. Am. Chem. Soc. 2016, 138,
5765–5768.
24. Nichols, E. M.; Derrick, J. S.; Nistanaki, S. K.; Smith, P. T.; Chang, C. J. Positional Effects
of Second-Sphere Amide Pendants on Electrochemical CO2 Reduction Catalyzed by Iron
Porphyrins. Chem. Sci. 2018, 9, 2952–2960.
25. McNamara, W. R.; Han, Z.; Yin, C.-J.; Brennessel, W. W.; Holland, P. L.; Eisenberg, R.
Cobalt-Dithiolene Complexes for the Photocatalytic and Electrocatalytic Reduction of
Protons in Aqueous Solutions. Proc. Natl. Acad. Sci. 2012, 109, 15594–15599.
71

26. Lee, K. J.; McCarthy, B. D.; Rountree, E. S.; Dempsey, J. L. Identification of an Electrode-
Adsorbed Intermediate in the Catalytic Hydrogen Evolution Mechanism of a Cobalt
Dithiolene Complex. Inorg. Chem. 2017, 56, 1988–1998.
27. Letko, C. S.; Panetier, J. A.; Head-Gordon, M.; Tilley, T. D. Mechanism of the
Electrocatalytic Reduction of Protons with Diaryldithiolene Cobalt Complexes. J. Am.
Chem. Soc. 2014, 136, 9364–9376.
28. Downes, C. A.; Yoo, J. W.; Orchanian, N. M.; Haiges, R.; Marinescu, S. C. H2 Evolution
by a Cobalt Selenolate Electrocatalyst and Related Mechanistic Studies. Chem. Commun.
2017, 53, 7306–7309.
29. Jeletic, M. S.; Helm, M. L.; Hulley, E. B.; Mock, M. T.; Appel, A. M.; Linehan, J. C. A
Cobalt Hydride Catalyst for the Hydrogenation of CO2: Pathways for Catalysis and
Deactivation. ACS Catal. 2014, 4, 3755–3762.
30. Fogeron, T.; Todorova, T. K.; Porcher, J. P.; Gomez-Mingot, M.; Chamoreau, L. M.;
Mellot-Draznieks, C.; Li, Y.; Fontecave, M. A Bioinspired Nickel(Bis-Dithiolene)
Complex as a Homogeneous Catalyst for Carbon Dioxide Electroreduction. ACS Catal.
2018, 8, 2030–2038.
31. Arévalo, R.; Chirik, P. J. Enabling Two-Electron Pathways with Iron and Cobalt: From
Ligand Design to Catalytic Applications. J. Am. Chem. Soc. 2019. 141, 23, 9106-9123.
32. Dey, S.; Ahmed, M. E.; Dey, A. Activation of Co(I) State in a Cobalt-Dithiolato Catalyst
for Selective and Efficient CO2 Reduction to CO. Inorg. Chem. 2018, 57, 5939–5947.
33. Jain, R.; Mashuta, M. S.; Buchanan, R. M.; Grapperhaus, C. A. Electrocatalytic Hydrogen
Evolution and Hydrogen Oxidation with a Ni(PS)2 Complex. Eur. J. Inorg. Chem. 2017,
2017, 3714–3719.
34. Haddad, A. Z.; Kumar, D.; Ouch Sampson, K.; Matzner, A. M.; Mashuta, M. S.;
Grapperhaus, C. A. Proposed Ligand-Centered Electrocatalytic Hydrogen Evolution and
Hydrogen Oxidation at a Noninnocent Mononuclear Metal-Thiolate. J. Am. Chem. Soc.
2015, 137, 9238–9241.
35. Barry, B. M.; Stein, B. W.; Larsen, C. A.; Wirtz, M. N.; Geiger, W. E.; Waterman, R.;
Kemp, R. A. Metal Complexes (M = Zn, Sn, and Pb) of 2-Phosphinobenzenethiolates:
Insights into Ligand Folding and Hemilability. Inorg. Chem. 2013, 52, 9875–9884.
36. Chapovetsky, A.; Welborn, M.; Luna, J. M.; Haiges, R.; Miller, T. F.; Marinescu, S. C.
Pendant Hydrogen-Bond Donors in Cobalt Catalysts Independently Enhance CO2
Reduction. ACS Cent. Sci. 2018, 4, 397–404.
37. (a) Russo, T. V.; Martin, R. L.; Hay, P. J. Application of Gradient-Corrected Density
Functional Theory to the Structures and Thermochemistries of ScF3, TiF4, VF5, and CrF6.
72

J. Chem. Phys. 1995, 102, 8023–8028. (b) Li, Z.; Liu, W. Critical Assessment of TD-DFT
for Excited States of Open-Shell Systems: I. Doublet-Doublet Transitions. J. Chem. Theory
Comput. 2016, 12, 238–260.
38. Costentin, C.; Robert, M.; Savéant, J. M. Catalysis of the Electrochemical Reduction of
Carbon Dioxide. Chem. Soc. Rev. 2013, 42, 2423–2436.
39. Costentin, C.; Savéant, J.-M. Multielectron, Multistep Molecular Catalysis of
Electrochemical Reactions: Benchmarking of Homogeneous Catalysts. ChemElectroChem
2014, 1, 1226–1236.
40. McCarthy, B. D.; Donley, C. L.; Dempsey, J. L. Electrode Initiated Proton-Coupled
Electron Transfer to Promote Degradation of a Nickel(II) Coordination Complex. Chem.
Sci. 2015, 6, 2827–2834.
41. (a) Clark, M. L.; Cheung, P. L.; Lessio, M.; Carter, E. A.; Kubiak, C. P. Kinetic and
Mechanistic Effects of Bipyridine (Bpy) Substituent, Labile Ligand, and Brønsted Acid on
Electrocatalytic CO2 Reduction by Re(Bpy) Complexes. ACS Catal. 2018, 8, 2021–2029.
(b) Lam, Y. C.; Nielsen, R. J.; Gray, H. B.; Goddard III, W. A. A Mn Bipyrimidine Catalyst
Predicted To Reduce CO2 at Lower Overpotential. ACS Catal. 2015, 5, 2521-2528. (c)
Kütt, A.; Leito, I.; Kaljurand, I.; Sooväli, L.; Vlasov, V. M.; Yagupolskii, L. M.; Koppel,
I. A. A Comprehensive Self-Consistent Spectrophotometric Acidity Scale of Neutral
Brønsted Acids in Acetonitrile J. Org. Chem. 2006, 71, 2829-2838. (d) Pegis, M. L.;
Roberts, J. A.; Wasylenko, D. J.; Mader, E. A.; Appel, A. M.; Mayer, J. M. Standard
Reduction Potentials for Oxygen and Carbon Dioxide Couples in Acetonitrile and N,N-
Dimethylformamide. Inorg. Chem. 2015, 54, 11883−11888.
42. Shao, Y.; Molnar, L.; Jung, Y.; Kussmann, J.; Ochsenfeld, C.; Brown, S.; Gilbert, A.;
Slipchenko, L.; Levchenko, S.; O'Neill, D.; DiStasio, R.; Lochan, R.; Wang, T.; Beran, G.;
Besley, N.; Herbert, J.; Lin, C.; Van Voorhis, T.; Chien, S.; Sodt, A.; Steele, R.; Rassolov,
V.; Maslen, P.; Korambath, P.; Adamson, R.; Austin, B.; Baker, J.; Byrd, E.; Dachsel, H.;
Doerksen, R.; Dreuw, A.; Dunietz, B.; Dutoi, A.; Furlani, T.; Gwaltney, S.; Heyden, A.;
Hirata, S.; Hsu, C.; Kedziora, G.; Khalliulin, R.; Klunzinger, P.; Lee, A.; Lee, M.; Liang,
W.; Lotan, I.; Nair, N.; Peters, B.; Proynov, E.; Pieniazek, P.; Rhee, Y.; Ritchie, J.; Rosta,
E.; Sherrill, C.; Simmonett, A.; Subotnik, J.; Woodcock, H.; Zhang, W.; Bell, A.;
Chakraborty, A.; Chipman, D.; Keil, F.; Warshel, A.; Hehre, W.; Schaefer, H.; Kong, J.;
Krylov, A.; Gill, P.; Head-Gordon, M. Advances in Methods and Algorithms in a Modern
Quantum Chemistry Program Package. Phys. Chem. Chem. Phys. 2006, 8, 3172-3191.
43. Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient Approximation Made
Simple. Phys. Rev. Lett. 1996, 77, 3865–3868.
44. Ditchfield, R.; Hehre, W. J.; Pople, J. A. Self-Consistent Molecular-Orbital Methods. IX.
An Extended Gaussian-Type Basis for Molecular-Orbital Studies of Organic Molecules. J.
Chem. Phys. 1971, 54, 724–728.
73

45. Hehre, W. J.; Ditchfield, K.; Pople, J. A. Self-Consistent Molecular Orbital Methods. XII.
Further Extensions of Gaussian-Type Basis Sets for Use in Molecular Orbital Studies of
Organic Molecules. J. Chem. Phys. 1972, 56, 2257–2261.
46. 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.
47. Feller, D. The role of databases in support of computational chemistry calculations J.
Comput. Chem. 1996, 17, 1571-1586.
48. Schuchardt, K. L.; Didier, B. T.; Elsethagen, T; Sun, L.; Gurumoorthi, V.; Chase, J.; Li, J.;
Windus, T. L. Basis Set Exchange: A Community Database for Computational Sciences J.
Chem. Inf. Model. 2007, 47, 1045-1052.
49. (a) Sheldrick, G. M. Phase Annealing in SHELX-90: Direct Methods for Larger Structures.
Acta Cryst. 1990, A46, 467. (b) Sheldrick, G. M. A Short History of SHELX. Acta Cryst.
2008, A64, 112. (c) Müller, P. Practical Suggestions for Better Crystals. Cryst. Rev. 2009,
15, 57.

74

CHAPTER 3

Influence of Transition Metal Ion on Electrocatalytic CO2 Reduction
with Aminopyridine Macrocycles

75

3.1 ABSTRACT
We report here the synthesis and characterization of iron and nickel complexes, FeL
1
and
NiL
1
, based on the azacalix[4](2,6)-pyridine framework, and compare their reactivity with that of
the structurally-analogous CoL
1
¸ which we have previously shown to efficiently catalyze the
electrochemical conversion of CO2 to CO. The structures of FeL
1
and NiL
1
were characterized by
1
H-NMR spectroscopy, FTIR spectroscopy, and single crystal X-ray diffraction studies,
confirming a similar coordination environment to the cobalt analogue. The electronic structure of
these complexes are probed through UV-Vis studies and density functional theory (DFT)
calculations, revealing a change in frontier orbital character as a function of the transition metal
identity. Finally, the electrochemical behavior of these complexes is studied via cyclic
voltammetry and controlled potential electrolysis in the presence of CO2. Both FeL
1
and NiL
1

exhibit lower selectivity for CO than CoL
1
, instead generating syngas mixtures as a result of
concomitant H2 evolution. Unlike CoL
1
, both FeL
1
and NiL
1
decompose during electrolysis to
deposit nanoparticulate material onto the glassy carbon electrode surface. The deposited materials
(ML
1
-GCE) are characterized by SEM and XPS analysis, and electrochemical studies show that
these materials generate predominantly H2. This divergence from the behavior of CoL
1
is
rationalized through the subtle differences in electronic structure resulting from the influence of
the metal ion, as predicted by DFT calculations. These results broaden the scope of our prior
studies by demonstrating the significance of the transition metal ion in determining the
electrochemical behavior of these macrocyclic systems.

Key words: Solar Fuels, Carbon Dioxide Reduction, Hydrogen Evolution, Electrocatalysis,
Syngas, Macrocycle

76

3.2 INTRODUCTION
The catalytic conversion of carbon dioxide (CO2) into value-added products is a promising
pathway to a carbon-neutral energy economy.
1-4
Due to the many possible products of CO2
reduction—including carbon monoxide (CO), formate, and methanol—catalyst selectivity remains
a challenge. Furthermore, competitive hydrogen evolution presents an additional barrier to
selective CO2 reduction in the presence of proton donors. In nature, the enzyme NiFe carbon
monoxide dehydrogenase (NiFe-CODH) catalyzes the selective reversible conversion of CO2 to
CO via a two proton and two electron transfer.
5
X-ray diffraction studies suggest that NiFe-CODH
converts CO2 to CO via a bifunctional activation by two metal centers in the NiFe cluster, where
CO2 binding is stabilized by H-bonding interactions from neighboring amino acid residues. These
studies indicate that a transition metal center with pendant proton donor motifs makes for an
effective design for CO2 reduction catalysts.
Molecular catalysts are of interest in this regard because they allow for fine-tuning of the
catalyst structure, which allows for synthetic control over the steric and electronic environment at
the active site. Many examples exist of molecular electrocatalysts displaying widely different
activity and selectivity upon variation of the identity of the transition metal center. It has been
shown that the product selectivity for CO2 reduction with complexes bearing a pentadentate
macrocyclic ligand (2,13-dimethyl-3,6,9,12,18-pentaazabicyclo-[12.3.1]octadeca-
1(18),2,12,14,16-pentaene) is highly-dependent on the identity of the metal center.
6
When a
cobalt(II) ion is incorporated into the ligand framework, the resulting complex converts CO2 to
CO with a faradaic efficiency (FE) of 97 % and a turnover number (TON) of 270 after 22 hours
under photochemical conditions. Upon switching the metal ion to iron(III), the resulting complex
selectively convert CO2 to formic acid (HCOOH) with FE of 80 % and TON of 1260 after 3 hours
77

under electrochemical conditions. This change in selectivity is attributed to the weakly π-donating
nature of iron(III), hindering C-O bond cleavage in the Fe-CO2 and resulting in facile isomerization
of the adduct to selectively generate HCOOH. Similarly, studies of cobalt(II) and iron(II)
complexes bearing quaterpyridine (qpy) ligands have demonstrated drastic changes in reactivity
as a function of transition metal identity.
7,8
The cobalt(II) complex efficiently catalyzes the
reduction of CO2 to CO with FECO of 97% at an overpotential (η) of 140 mV under electrochemical
conditions, while the iron(II) complex exhibits a FECO of 48% under analogous conditions. This
notable decrease in faradaic efficiency was attributed to a closing of the catalytic cycle due to
generation of the inactive [Fe
(0)
qpy(CO)] from an electrochemically-generated [Fe
(I)
qpy(CO)]
species. These examples highlight the impact of the transition metal center on determining the
activity and selectivity of structurally-analogous catalysts.
Our group has previously reported a series of cobalt(II) complexes bearing macrocyclic
ligands with varying numbers (0-4) of secondary and tertiary pendant amine groups incorporated
in the ligand scaffold.
9,10
These prior studies demonstrate that when all four pendant amines are
secondary (CoL
1
, Scheme 3.1), the complex makes for an efficient electrocatalyst for the selective
reduction of CO2 to CO with high rates.
9
A linear trend can be seen between the number of
secondary amine groups and the activity of the catalysts, with additional protonated amine
substituents yielding enhanced rate constants. Conversely, methylation of all four amines to
generate tertiary sites (CoL
2
, Scheme 3.1) results in a 300-fold decrease in activity.
Electrochemical studies and density functional theory (DFT) studies suggest a mechanism in
which pendant secondary amines facilitate a hydrogen-bonding network that increases the local
proton activity in the vicinity of the metal center and enables proton transfer from the acid additive
(trifluoroethanol, TFE) to the metal-coordinated CO2 substrate.
10
In this current work, we
78

investigate a series of ML
1
complexes (Scheme 3.1) with iron(II) or nickel(II) metal centers to
study the influence of transition metal identity on the reactivity of this system. Both FeL
1
and NiL
1

are structurally-characterized and their electronic structures are compared to that of CoL
1
.
Following these studies, the electrochemical behavior of FeL
1
and NiL
1
are explored through
cyclic voltammetry and controlled potential electrolysis studies in the presence of CO 2. The
electronic structure of these complexes is explored through DFT studies and the differences in
reactivity for these complexes are rationalized with a frontier molecular orbital picture.

3.3 RESULTS AND DISCUSSION

Scheme 3.1 Synthesis of complexes.

The desired L
1
ligand was prepared according to reported literature procedure.
11
The
iron(II) complex, FeL
1
, was prepared via addition of stoichiometric [Fe(H2O)6][BF4]2 in 9:1
pyridine:DMF to a pyridine solution of L
1
. The resulting brown solid was collected and
recrystallized to yield red crystals suitable for single crystal X-ray diffraction (XRD) studies. The
nickel(II) complex, NiL
1
, was similarly prepared via addition of stoichiometric
[Ni(MeCN)6][BF4]2 in 9:1 pyridine:DMF to a pyridine solution of L
1
, generating a brown powder
which was recrystallized to yield in amber crystals suitable for single crystal XRD studies. The
formation of the both metal complexes, FeL
1
and NiL
1
, was achieved in moderately high yields
of 70% and 86%, respectively (Scheme 3.1). Both complexes are synthesized and handled in air
79

with no evidence of oxidation or decomposition by
1
H-NMR spectroscopy (see Experimental
Methods for detailed synthesis).

Figure 3.1 Side (top) and top (bottom) views of (a) FeL
1
and (b) NiL
1
as determined by single
crystal XRD, and of (a) FeL
1
and (b) NiL
1
as determined by DFT calculations. Hydrogen atoms,
noncoordinating anions, solvent molecules, and axial pyridine ligands omitted from (a) and (b) for
clarity.

Table 3.1 Crystal data and structure refinement for FeL1.
Identification code GR0235FeNH
Chemical formula C30H26BF4FeN10
Formula weight 669.27 g/mol
Temperature 100(2) K
Wavelength 0.71073 Å
Crystal size 0.104 x 0.135 x 0.460 mm
Crystal habit clear yellow plate
Crystal system monoclinic
Space group P 1 21/m 1
Unit cell dimensions a = 10.7530(16) Å α = 90°

b = 13.695(2) Å β = 91.719(3)°

c = 15.596(2) Å γ = 90°
Volume 2295.7(6) Å
3

Z 2
Density (calculated) 0.968 g/cm
3

Absorption coefficient 0.371 mm
-1

F(000) 686
Diffractometer Bruker APEX II CCD Bruker APEX DUO
Radiation source fine-focus tube (MoKα , λ = 0.71073 Å)
Theta range for data collection 1.71 to 21.97°
Index ranges -11<=h<=11, -14<=k<=14, -24<=l<=24
Reflections collected 43704
Independent reflections 4301 [R(int) = 0.0685]
Coverage of independent reflections 100.0%
Absorption correction multi-scan
Structure solution technique direct methods
80

Structure solution program SHELXTL XT 2014/5 (Bruker AXS, 2014)
Refinement method Full-matrix least-squares on F
2

Refinement program SHELXTL XL 2017/1 (Bruker AXS, 2017)
Function minimized Σ w(Fo
2
- Fc
2
)
2

Data / restraints / parameters 4301 / 703 / 543
Goodness-of-fit on F
2
1.095
Δ/σmax 0.001
Final R indices 3636 data; I>2σ(I) R1 = 0.0853, wR2 = 0.2019

all data R1 = 0.0967, wR2 = 0.2085
Weighting scheme w=1/[σ
2
(Fo
2
)+(0.0628P)
2
+42.4197P]
where P=(Fo
2
+2Fc
2
)/3
Largest diff. peak and hole 1.314 and -0.700 eÅ
-3

R.M.S. deviation from mean 0.118 eÅ
-3

Table 3.2 Crystal data and structure refinement for NiL1
Identification code GR02XXX_Ni_NH
Chemical formula C30H26B2F8N10Ni
Formula weight 758.94 g/mol
Temperature 100(2) K
Wavelength 0.71073 Å
Crystal system monoclinic
Space group P 1 21/n 1
Unit cell dimensions a = 10.920(3) Å α = 90°

b = 13.874(3) Å β = 95.981(4)°

c = 23.300(6) Å γ = 90°
Volume 3510.8(15) Å
3

Z 4
Density (calculated) 1.436 g/cm
3

Absorption coefficient 0.633 mm
-1

F(000) 1544
Diffractometer Bruker APEX II CCD Bruker APEX DUO
Radiation source fine-focus tube (MoKα , λ = 0.71073 Å)
Theta range for data collection 1.71 to 21.97°
Index ranges -11<=h<=11, -14<=k<=14, -24<=l<=24
Reflections collected 43704
Independent reflections 4301 [R(int) = 0.0685]
Coverage of independent reflections 100.0%
Absorption correction multi-scan
Structure solution technique direct methods
Structure solution program SHELXTL XT 2014/5 (Bruker AXS, 2014)
Refinement method Full-matrix least-squares on F
2

Refinement program SHELXTL XL 2017/1 (Bruker AXS, 2017)
Function minimized Σ w(Fo
2
- Fc
2
)
2

Data / restraints / parameters 4301 / 703 / 543
Goodness-of-fit on F
2
1.095
81

Δ/σmax 0.001
Final R indices 3636 data; I>2σ(I) R1 = 0.0853, wR2 = 0.2019
all data R1 = 0.0967, wR2 = 0.2085
Weighting scheme w=1/[σ
2
(Fo
2
)+(0.0628P)
2
+42.4197P]
where P=(Fo
2
+2Fc
2
)/3
Largest diff. peak and hole 1.314 and -0.700 eÅ
-3

R.M.S. deviation from mean 0.118 eÅ
-3

Figure 3.2 Calculated geometries for the ML
1
(M = Fe, Co, Ni) series at the 6-31+G(d)/B3LYP
level of theory. Geometries were confirmed as local minima with frequency calculations at the
same level of theory.

Table 3.3 Selected calculated bond distances for FeL
1
and NiL
1
based on DFT geometry
optimizations.

Fe Ni
M-N2 1.98761 2.02831
M-N4 1.99403 2.02877
M-N6 1.98803 2.02872
M-N8 1.99396 2.02901
N1-H 1.0141 1.01445
N3-H 1.01421 1.0144
N5-H 1.01414 1.01446
N7-H 1.01417 1.01422
M-N(avg) 1.990908 2.028703
N(avg)-H 1.014155 1.014383
N-M-N 179.4132 179.9199
N-M-N 177.1722 179.7926
N-M-N(avg) 178.2927 179.8562

82

Single-crystal XRD analysis (Figure 3.1a,b) reveals FeL
1
and NiL
1
bear the formulae
FeL
1
∙(pyridine)2 and NiL
1
∙(pyridine)2, respectively (Tables 3.1-3.2). FeL
1
and NiL
1
adopt similar
coordination environments to CoL
1
; the pyridine nitrogen atoms of the macrocycle bind to the
metal in a square planer fashion, with an average bond distance of 1.95(0) Å for FeL
1
and 2.05(9)
Å for NiL
1
. The metal-NH distances average to 3.10(9) Å for FeL
1
and 3.14(5) Å for NiL
1
. The
macrocyclic ligand exhibits a saddle-like conformation, as previously shown for the CoL
1

complex.
9
These distances closely match the bond distances in CoL
1
as well as those reported for
related iron(II) and nickel(II) polypyridine complexes.
12,13
The solid state structures indicate that
the metal centers exhibit octahedral coordination with two axial pyridine ligands, and two BF4
-

counterions outside the coordination sphere. To further confirm these similarities in coordination
environment, DFT calculations were performed on FeL
1
, CoL
1
, and NiL
1
at the 6-
31+G(d)/B3LYP level of theory (see experimental methods for details). The B3LYP functional
was selected based on the prior success of applying this functional towards electronic structure
and thermochemical calculations of CoL
1
.
10
Geometry optimizations were performed in the
absence of axial solvent ligands and all calculated structures were confirmed as stable minima via
frequency calculations at the same level of theory. Frequency calculations for FeL
1
and NiL
1
do
not exhibit imaginary modes, as is consistent with a stable local minimum structure. The calculated
structures for FeL
1
, CoL
1
, and NiL
1
closely match the coordination environments determined
through single-crystal XRD studies. In all cases, the macrocyclic ligand is predicted to exhibit a
saddle-like conformation and the four pyridine nitrogen atoms of the macrocycle bind to the metal
in a square planer fashion (Figure 3.2). The average bond distances between the metal ion and
coordinating nitrogen atoms are predicted to be of 1.99(1) Å for FeL
1
and 2.02(9) Å for NiL
1
,
similar to the crystallographic bond distances (Table 3.3). The minor bond elongation predicted
83

for FeL
1
and bond contraction predicted for NiL
1
are attributed to geometric distortions in the
solid state packing which is expected to deviate from the gas-phase optimized geometry.

Figure 3.3 Transmittance FTIR spectrum of FeL
1
.

Figure 3.4 Transmittance FTIR spectrum of CoL
1
.

84

Figure 3.5 Transmittance FTIR spectrum of NiL
1
.

Figure 3.6 Predicted N-H stretching frequencies as calculated by DFT at the 6-31+G(d)/B3LYP
level of theory.

FTIR studies of the complexes reveal N-H stretching frequencies of 3349 cm
-1
for FeL
1
,
3343 cm
-1
for CoL
1
, and 3346 cm
-1
for NiL
1
(Figures 3.3-3.5). The stretching frequency for CoL
1

exhibits a bathochromic shift by 6 cm
-1
with respect to FeL
1
and by 3 cm
-1
with respect to NiL
1
.
This trend is further supported by the predicted vibrational modes based on DFT calculations at
the 6-31+G(d)/B3LYP level of theory (see Experimental Methods for details). The calculated N-
H stretching modes for CoL
1
is predicted to exhibit a bathochromic shift by 5 cm
-1
with respect to
FeL
1
and by 3 cm
-1
with respect to NiL
1
(Figure 3.6). These results indicate that the N-H bond
strength in CoL
1
is weaker than that of FeL
1
or NiL
1
. As our prior studies on CoL
1
have
85

demonstrated that the acidity of these N-H groups is key to the promotion of hydrogen-bond
networks which facilitate catalysis, it is expected that these subtle electronic changes are likely to
impact the electrocatalytic activity of FeL
1
and NiL
1
.
10

Figure 3.7 UV-VIS spectra of (a) FeL
1
, (b) CoL
1
, and (c) NiL
1
. Spectra acquired in DMF with a
1 cm quartz cuvette. a) A single absorbance is observed at 344 nm with ε = 19,200 M
-1
cm
-1
. b)
Absorbances are observed at 336 nm (ε = 34,400 M
-1
cm
-1
) and 434 nm (ε = 2,011 M
-1
cm
-1
). c) A
single absorbance is observed at 336 nm (ε = 27,500 M
-1
cm
-1
).

Figure 3.8 Absorbance values at varying concentrations of FeL
1
in DMF. Absorbance values
measured at 344 nm.

86

Figure 3.9 Absorbance values at varying concentrations of CoL
1
in DMF. Absorbance values
measured at 336 nm.

Figure 3.10 Absorbance values at varying concentrations of NiL
1
in DMF. Absorbance values
measured at 336 nm.

87

Figure 3.11 Predicted UV-Vis transitions based on TD-DFT calculations at the 6-31+G(d)/B3LYP
level of theory. Transitions are visualized with the initial orbital state depicted with green/orange
mesh lobes and final orbitals depicted with red/blue solid lobes. All orbitals are depicted with
isovalue = 0.05 for clarity.

To probe the electronic structure of these complexes, UV-Vis spectroscopy studies were
performed for FeL
1
, CoL
1
, and NiL
1
in DMF solution (Figure 3.7-3.10). All three complexes
exhibit a strong absorption peak in the UV region. The measured spectra of FeL
1
reveal a broad
transition at 334 nm (1.92 × 10
4
M
-1
cm
-1
). For CoL
1
, a sharper UV transition similarly appears at

334 nm (3.44 × 10
4
M
-1
cm
-1
). NiL
1
also displays a sharp transition at 334 nm (2.75 × 10
4
M
-1
cm
-
1
). As the wavelength of this transition is not influenced by the identity of the transition metal, it
can be inferred that this corresponds to an intraligand π-π
*
transition. Additionally, while FeL
1
and
NiL
1
both exhibit only a single dominant transition, CoL
1
exhibits an additional feature at 442
nm (2.01 × 10
3
M
-1
cm
-1
). To identify the character of these transitions, time-dependent DFT (TD-
DFT) calculations were performed at the 6-31+G(d)/B3LYP level of theory (Figure 3.11). The
B3LYP functional was selected for TD-DFT studies based on literature precedent for the
88

applicability of this functional towards the prediction of excited states of transition metal
systems.
15,16
All three complexes are predicted to exhibit intraligand π-π
*
transitions in the UV
region. While the electronic spectra of CoL
1
and NiL
1
are each predicted to show a single
transition in this region, FeL
1
is predicted to show several closely-spaced transitions with
intraligand π-π
*
character. This prediction is consistent with the broader linewidth of the UV
transition observed experimentally for FeL
1
. Additionally, while FeL
1
and NiL
1
are both predicted
to exhibit ligand-to-metal charge transfer (LMCT) transitions in the near-IR region (at wavelengths
>800 nm), CoL
1
is predicted to undergo a LMCT transition in the visible region as is consistent
with the experimentally-measured spectra.

Figure 3.12 Frontier molecular orbital diagrams for FeL
1
, CoL
1
, and NiL
1
at the 6-
31+G(d)/B3LYP level of theory (see Experimental Methods for details).

89

Figure 3.13 Molecular orbital images as calculated by DFT at the 6-31+G(d)/B3LYP level of
theory.

Figure 3.14 Spin density localization plots for paramagnetic complexes as calculated by DFT at
the 6-31+G(d)/B3LYP level of theory.

The electronic structure of these complexes was further explored jointly through
1
H-NMR
spectroscopy studies and unrestricted single-point energy calculations. Studies of FeL
1
in pyridine-
d5 reveal a diamagnetic
1
H-NMR spectrum, suggesting a low-spin d
6
complex. In contrast, NiL
1

displays a paramagnetic
1
H-NMR spectrum with a μeff of 3.01 μB, corresponding to two unpaired
electrons, suggesting a high-spin d
8
complex.
14
Similarly, CoL
1
is also paramagnetic with a μeff of
3.63 μB, corresponding to three unpaired electrons.
11
These experimental results are supported by
DFT predictions at the 6-31+G(d)/B3LYP level of theory, and the predicted molecular orbital
90

diagrams for these complexes are depicted in Figure 3.12-3.13. Single-point energy calculations
predict a singlet ground state for FeL
1
, consistent with a closed-shell d
6
complex as was determined
by
1
H-NMR spectroscopy. The calculated highest-occupied molecular orbital (HOMO) for FeL
1

is predicted to exhibit predominantly ligand-based character associated with the π-system of the
macrocycle. The corresponding lowest-unoccupied molecular orbital (LUMO) is predicted to
exhibit predominantly metal-based character associated with the Fe(dyz) orbital. The next-lowest
unoccupied orbital (LUMO+1) is predicted to exhibit predominantly ligand-based character.
Together, these results suggest that the first one-electron reduction of FeL
1
to [FeL
1
]
1-
is a metal-
based event and leads to the occupation of the Fe(dyz) α-orbital, and the second reduction of
[FeL
1
]
1-
to [FeL
1
]
2-
is another metal-based event and leads to the occupation of the Fe(dyz) β-
orbital. A third reduction event would be predicted to populate the α-LUMO+1 orbital, leading to
the occupation of a π
*
orbital of the macrocycle system. Analogous calculations for CoL
1
predict
a ground-state quartet configuration, consistent with three unpaired electrons as determined by
1
H-
NMR spectroscopy. The calculated highest-occupied molecular orbital (HOMO) for CoL
1
is
predicted to exhibit ligand-based character associated with the π-system of the macrocycle,
analogous to that of FeL
1
. The first singly-occupied molecular orbital (SOMO) is predicted to
exhibit predominantly metal-based character associated with the Co(dz2) orbital. The next-lowest
singly-occupied orbital (SOMO+1) is also predicted to exhibit metal-based character associated
with a superposition of the Co(dxz) and Co(dyz) orbitals. The highest singly-occupied orbital
(SOMO+2) is predicted to exhibit mixed metal-ligand character associated with both the Co(dx2-
y2) orbital and the ligand π-system. These results suggest that, unlike FeL
1
, the cobalt-based system
is predicted to undergo two metal-based reduction events followed by a mixed metal-ligand
reduction event. Calculations for NiL
1
predict a ground-state triplet configuration possessing two
91

unpaired electrons, consistent with
1
H-NMR spectroscopy. The calculated highest-occupied
molecular orbital (HOMO) for NiL
1
is analogous to that of the previous two complexes, exhibiting
ligand-based character associated with the π-system of the macrocycle. The SOMO of NiL
1
is
predicted to exhibit metal-based character associated with the Ni(dyz) orbital, similar to that of
FeL
1
. The next-lowest singly-occupied orbital (SOMO+1) is also predicted to exhibit metal-based
character associated the Ni(dx2-y2) orbital. As such, this suggests that NiL
1
is predicted to undergo
two metal-based reduction events. The predominantly metal-based character of the singly-
occupied orbitals for CoL
1
and NiL
1
is further supported by spin-localization calculations (Figure
3.14). For both for CoL
1
and NiL
1
, visualization of the spin density indicates that the unpaired
spins of these complexes are localized primarily on the transition metal ion.

Scheme 3.2 Frontier molecular orbitals for the [ML
1
] series.

These distinctions in predicted molecular orbital character highlight the impact of the
transition metal ion on modulating the electronic structure of these complexes. While the HOMO
for this system remains ligand-based as the metal ion is exchanged between iron, cobalt, and nickel,
the ordering of the higher-lying frontier orbitals is strongly influenced by transition metal identity.
As illustrated in Scheme 3.2, iron is predicted to position the frontier orbitals to facilitate a two-
electron reduction of the metal center with the occupation of Fe(dyz)-symmetry orbitals, and a third
reduction event localized on the ligand. In contrast, cobalt and is predicted to position the frontier
92

orbitals to facilitate a two-electron reduction of the metal center, occupying orbitals of Co(dz2) and
Co(dxz)+Co(dyz) symmetry. Nickel is also predicted to facilitate two metal-based reduction events,
occupying orbitals of Ni(dyz) and Ni(dx2-y2) symmetry. As CO2 reduction requires a two-electron
transfer to a bound CO2 molecule, principles of orbital symmetry suggest that the occupation of
metal-based orbitals with appropriate symmetry for interaction with an axially-bound bent CO2
substrate would facilitate facile electron transfer from the catalyst to the substrate.
17
Prior studies
have indicated that selectivity for CO2 reduction over proton reduction is best facilitated by
complexes capable of both σ-type and π-type interactions with the bent CO2 molecule.
18
The σ-
type interaction facilitates activation of CO2 through the formation of a metal-carbon bond, while
the π-type interaction selectively stabilizes an interaction with CO2 over protons, hindering
competitive hydrogen evolution. Based on the predicted molecular orbital pictures depicted in
Scheme 3.2, these calculations suggest that only CoL
1
is capable of both σ-type and π-type
interactions with the bent CO2 molecule. The occupation of a Co(dz2) orbital facilitates a σ-type
interaction with CO2, while occupation of a Co(dxz)-type orbital would facilitate a π-type
interaction. In contrast, FeL
1
is predicted to undergo two metal-based reduction events with the
occupation of Fe(dyz)-symmetry orbitals. And while NiL
1
is predicted to undergo two metal-based
reductions, the second reduction is predicted to occupy a Ni(dx2-y2) orbital with no spatial
projection along the z-axis and therefore improper symmetry for orbital overlap with an axially-
bound CO2 substrate. With this in mind, the cobalt system is predicted to present the most-optimal
orbital configuration for selective CO2 activation.

93

Figure 3.15 Electrochemical studies of the ML
1
complexes under N2 (blue) and CO2 (red)
atmosphere with a scan rate of 100 mV/s. (a) FeL
1
, (b) NiL
1
.

The electrochemical behavior of FeL
1
and NiL
1
was characterized through cyclic
voltammetry (CV) studies. Both complexes were studied at a concentration of 0.5 mM in DMF
solution with 0.1 M [nBu4N][PF6] supporting electrolyte, as we previously reported for the CoL
1

system. A three-electrode configuration was applied with a glassy carbon working electrode,
platinum counter electrode, and silver wire pseudo-reference electrode separated from solution by
a glass frit, and all potentials are reported vs Fc
+
/Fc.

Table 3.4 Summary of electrochemical behavior of the ML
1
complexes. Reduction potentials
determined by cyclic voltammetry with 0.5 mM of ML
1
in 0.1 M [nBuN4][PF6] in DMF at 100
mV/s with a glassy carbon electrode.
ML
1
ML
1
/[ML
1
]
1-

(V vs Fc
+
/Fc)
[ML
1
]
1-
/[ML
1
]
2-

(V vs Fc
+
/Fc)

Reference
FeL
1
-1.94 (irrev) -2.35 (irrev) This work
CoL
1
-1.65 -2.46 (irrev) 9,10
NiL
1
-1.51 -2.77 (irrev.) This work

94

Figure 3.16 Cyclic voltammograms of 0.5 mM NiL1 in a DMF solution containing 0.1 M
[nBu4N][PF6] under an atmosphere of N2 displaying the reversible one-electron reduction,
corresponding to the Ni
II/I
couple, (left) and the irreversible one-electron reduction (right),
corresponding to the Ni
I/0
reduction. Scan rates vary from 0.025 to 2 V/s.

Figure 3.17 Plot showing peak current densities for both NiL1 reductions in the CVs of 0.5 mM
NiL1 in DMF containing 0.1 M [nBu4N][PF6] under an atmosphere N2. The cathodic and anodic
peak current densities increase linearly with the square root of the scan rate. This is indicative of
a freely-diffusing species, in which the electrode reaction is controlled by mass transport.

95

Figure 3.18 Cyclic voltammogram of 0.05 mM of FeL1 in a DMF solution containing 0.1 M
[nBu4N][PF6] under an atmosphere of N2

Figure 3.19 Cyclic voltammograms of FeL1 (0.05 mM left, 0.5 mM right) in a DMF solution
containing 0.1 M [nBu4N][PF6] under an atmosphere of N2 displaying the irreversible reductions
corresponding to the Fe
II/I
reduction (left, -1.94 V) and the Fe
I/0
reduction (right -2.35 V). Scan
rates vary from 0.025 to 2 V/s.

96

Figure 3.20 Plot showing peak current densities for both FeL1 reductions in the CVs of FeL1 (0.05
mM left, 0.5 mM right) in DMF containing 0.1 M [nBu4N][PF6] under an atmosphere N2. The
cathodic peak current densities increase linearly with the square root of the scan rate. This is
indicative of a freely-diffusing species, in which the electrode reaction is controlled by mass
transport.

Cyclic voltammetry of FeL
1
and NiL
1
reveals an initial irreversible reduction of FeL
1
at -
1.94 V and a reversible reduction for NiL
1
at -1.51 V attributed to the Fe
(II)
/Fe
(I)
and

Ni
(II)
/Ni
(I)
reductions, respectively (Figure 3.15-3.21 and Table 3.4). Scanning more cathodically reveals
irreversible reductions for FeL
1
and NiL
1
at -2.35 V and -2.77 V, respectively, corresponding to
the Fe
(I)
/Fe
(0)
and Ni
(I)
/Ni
(0)
reduction events. In this regard, the electrochemical behavior of NiL
1

more closely resembles that of CoL
1
, as both exhibit one reversible and one irreversible reduction,
whereas FeL
1
deviates from this behavior as no reversible features are observed (Table 3.4). A
trend can be seen in the reduction potentials of the ML
1
/[ML
1
]
1-
and [ML
1
]
1-
/[ML
1
]
2-
series as a
function of the transition metal ion: the ML
1
/[ML
1
]
1-
reductions occur at more positive potentials
across the series, whereas the [ML
1
]
1-
/[ML
1
]
2-
reductions occur at more negative potentials across
the series. This indicates that as the metal ion is changed from iron to nickel, the energetic
separation between the two reduction events increases.

97

Figure 3.21 Cyclic voltammogram of 0.5 mM of FeL1 in a DMF solution containing 0.1 M
[nBu4N][PF6] under an atmosphere of CO2 after repetitive scans displays irreproducible current
densities. Scan rate is 100 mV/s.

Figure 3.22 Cyclic voltammogram of 0.5 mM of NiL1 in a DMF solution containing 0.1 M
[nBu4N][PF6] under an atmosphere of CO2 after repetitive scans displays irreproducible current
densities. Scan rate is 100 mV/s.

98

Figure 3.23 Cyclic voltammogram of 0.5 mM of FeL1 in a DMF solution containing 0.1 M
[nBu4N][PF6] under an atmosphere of N2 after repetitive scans displays reproducible current
densities. Scan rate is 100 mV/s.

Figure 3.24 Cyclic voltammogram of 0.5 mM of NiL1 in a DMF solution containing 0.1 M
[nBu4N][PF6] under an atmosphere of N2 after repetitive scans displays reproducible current
densities. Scan rate is 100 mV/s.

When CV experiments are performed in the presence of a CO2 atmosphere, enhanced
currents are observed near the second reduction events for FeL
1
and NiL
1
. Additionally, an anodic
shift in the second reduction potential by 0.1 V is measured (Figure 3.15). Similar behavior is
observed for CoL
1
,
9
which suggests that the doubly-reduced FeL
1
and NiL
1
complexes are
likewise capable of interacting with CO2. Unlike the CoL
1
system, the waveshape of the CV traces
for FeL
1
and NiL
1
bear a peak-like shape, rather than a plateau shape. This is indicative of a
99

reaction that is limited by substrate diffusion.
19
Furthermore, sequential scans of FeL
1
and NiL
1

under CO2 (Figures 3.21-3.22) demonstrate a decrease in current with further scanning. This
behavior suggests possible decomposition of the catalyst at cathodic potentials and subsequent
deposition of material as a passivating film onto the working electrode. As this behavior is not
observed under N2 atmosphere (Figures 3.23-3.24), these results indicate that passivation of the
electrode surface results from the interaction of the reduced complexes with CO2 substrate.

Figure 3.25 Titration studies of the various complexes with 2,2,2-trifluoroethanol (TFE) in 0.1 M
[nBu4N][PF6] in DMF under CO2. (a) CVs of FeL
1
, (b) CVs of CoL
1
(reference 10), and (c) CVs
of NiL
1
.

100

Figure 3.26 Linear (A) and Log-Log (B) plots of the observed rate constants kobs vs. [TFE] in a
DMF solution containing 0.1 M [nBu4N][PF6], FeL1 (0.5 mM), and CO2 (0.20 M). The slope of
the linear plot corresponds to an overall rate constant of 268 M
–1
s
–1
. The slope of the log-log plot
is not equal to one, indicating the dependence on protons is not first order.

Figure 3.27 Linear (A) and Log-Log (B) plots of the observed rate constants kobs vs. [TFE] in a
DMF solution containing 0.1 M [nBu4N][PF6], NiL1 (0.5 mM), and CO2 (0.20 M). The slope of
the linear plot corresponds to an overall rate constant of 14.3 M
–1
s
–1
. The slope of the log-log plot
is not equal to one, indicating the dependence on protons is not first order.

We have previously reported that the addition of 2,2,2-trifluoroethanol (TFE) to CoL
1

results in larger catalytic currents, resulting from a noncooperative mechanism wherein the
pendant amines facilitate a hydrogen-bonding network which enables direct proton transfer from
external acid to the activated CO2 adduct.
10
To probe the possible electrocatalytic behavior of FeL
1

and NiL
1
, titration studies were similarly performed with TFE for these complexes. Under a CO 2
101

atmosphere, the peak current potentials for FeL
1
and NiL
1
are measured as -2.78 and -2.90 V,
respectively. Titrating the complexes with TFE (Figure 3.25-3.27) results in enhanced current
densities, reaching 4.58 mA/cm
2
for FeL
1
at -2.68 V and 1.6 mA/cm
2
for NiL
1
at -2.90 V. These
values are substantially lower than that measured for CoL
1
, which reaches at a current density of
~11 mA/cm
2
. This indicates lower catalytic activity for the iron and nickel analogues, as the current
density measured under catalytic conditions is proportional to the rate of the electrocatalytic
reaction.
19

Figure 3.28 Controlled potential electrolysis of 0.5 mM FeL1 performed at -2.68 V, corresponding
to the potential of max current observed by cyclic voltammetry, containing 0.1 M [nBu4N][PF6]
and 1.4 M TFE.

Figure 3.29 Controlled potential electrolysis of 0.5 mM NiL1 performed at -2.9 V, containing 0.1
M [nBu4N][PF6] and 1.4 M TFE.
102

Figure 3.30 Controlled potential electrolysis of 0.5 mM FeL1 performed at -2.4 V, containing 0.1
M [nBu4N][PF6] and 1.4 M TFE.

Table 3.5 Summary of the electrocatalytic behavior of the ML
1
complexes and ML
1
-GCE
materials. Experiments were performed in 0.1 M [nBuN4][PF6] in DMF with 1.4 M TFE with a
glassy carbon working electrode.
ML
1
CPE
Potential (V)
ML
1

CO/H2 FE (%)

ML
1
-GCE
CO/H2 FE (%)
Reference
FeL
1
-2.40 19/16 3/93 This work
CoL
1
-2.80 99/0 N/A 9,10
NiL
1
-2.90 55/45 47/14 This work

In order to study the selectivity and stability of these systems, both FeL
1
and NiL
1
were
further studied by controlled potential electrolysis (CPE) studies. Our prior CPE studies of CoL
1
demonstrate this system as a competent catalyst for CO2-to-CO reduction, able to selectively
generate CO with a faradaic efficiency (FE) of 98% at a potential of -2.8 V in DMF in the presence
of 1.4 M TFE.
9
Furthermore, it was shown that this complex exhibits no decrease in measured
current during a 2 hour electrolysis study, and no evidence of catalyst degradation or deposition
was observed. CPE experiments were performed for FeL
1
and NiL
1
under similar conditions at
their respective potentials of maximum current (-2.68 V for FeL
1
and -2.90 V for NiL
1
). CPE of
FeL
1
at -2.68 V (Figure 3.28) reveals a rapid decrease in current within 15 min of electrolysis.
After 1 h, no gaseous products were detected in the headspace and no formate or additional
103

products (including methanol, carbonate, or aldehydes) were observed in solution. Conversely,
CPE of NiL
1
at -2.90 V (Figure 3.29) results in the production of 19.7 μmol of CO and 16.6 μmol
of H2, corresponding to faradaic efficiencies of 55% and 45%, respectively (Table 3.5). CV studies
of NiL
1
before and after CPE reveal a large increase in current density following the CPE
experiment (Figure 3.29). These results suggest decomposition of the material under catalytic
conditions, which was confirmed visually by the formation of a brown film on the electrode
surface. Performing CPE of FeL
1
at a more positive potential of -2.40 V (Figure 3.30) shows
behavior similar to NiL
1
, although at much lower efficiency. After 1 h, electrolysis in the presence
of FeL
1
generates 7.4 μmol of CO and 6.4 μmol of H2, resulting in Faradaic efficiencies of 19%
and 17%, respectively (Table 3.5). Analysis of the cell solution reveals that no formate or
additional products (including methanol, carbonate, or aldehydes) were produced. Similarly, CV
traces performed before and after electrolysis reveal an increase in current density and the
formation of a brown film on the electrode surface, suggesting FeL
1
also deposits onto the working
electrode (Figure 3.30).

Figure 3.31 Scanning Electron Microscope (SEM) images (25 kV accelerating voltage) of the
DMF-washed (left), unwashed (middle), and clean (right) surfaces of a GCE used for a CPE
experiment of 0.5 mM FeL1 with 0.1 M [nBu4N][PF6] and 1.4 M TFE in DMF. Both the DMF-
washed and unwashed surfaces display similar morphologies of deposited FeL1 material.

104

Figure 3.32 SEM images (25 kV accelerating voltage) of the DMF-washed (left), unwashed
(middle), and clean (right) portions of a GCE used for a CPE experiment of 0.5 mM NiL1 with 0.1
M [nBu4N][PF6] and 1.4 M TFE in DMF. Both the DMF-washed and unwashed surfaces display
similar morphologies of deposited NiL1 material.

Figure 3.33 SEM images (25 kV accelerating voltage) of the DMF-washed (left), unwashed
(middle), and clean (right) portions of a GCE used for a CPE experiment of 0.5 mM CoL1 with
0.1 M [nBu4N][PF6] and 1.2 M TFE in DMF. The DMF-washed surface shows no deposition of
material. The unwashed portion shows a morphology indicative of the electrolyte.

Figure 3.34 XPS analysis of DMF-washed ML
1
-GCE surfaces. (a) Fe 2p core level XPS
spectrum of FeL
1
-GCE; (b) Co 2p core level XPS spectrum of CoL
1
-GCE; (c) Ni 2p core level
XPS spectrum of NiL
1
-GCE.

The glassy carbon electrode (GCE) surface following CPE of FeL
1
, CoL
1
, and NiL
1
was
analyzed by scanning electron microscopy (SEM) and X-ray photoelectron spectroscopy (XPS) to
characterize the possible decomposition products generated during electrolysis studies. These
post-CPE electrodes are referred to as ML
1
-GCE. Two portions of the electrode were analyzed by
105

SEM: one portion which was examined as-is following CPE, and another portion which was first
washed with DMF to remove any soluble species. XPS experiments were only performed on the
DMF-washed surface. SEM images of both the as-prepared and DMF-washed FeL
1
-GCE surfaces
reveal a morphology of nanoparticles of about 100 nm in diameter (Figure 3.31). Similar results
were obtained for NiL
1
-GCE with the observance of nanoparticulate material on the DMF-washed
portion of the electrode (Figure 3.32). In contrast, SEM studies were performed for CoL
1
-GCE
(Figure 3.33), and no deposition of CoL
1
onto the GCE surface was observed. SEM images of the
as-prepared CoL
1
-GCE surface reveal morphologies indicative of the electrolyte [nBu4N][PF6],
whereas images of the washed surface appear indistinguishable from that of the clean surface of a
freshly-washed GCE electrode. These results suggest that rinsing the FeL
1
-GCE and NiL
1
-GCE
with DMF simply removes deposited electrolyte, but the deposited film is not removed, whereas
CoL
1
exhibits no evidence of decomposition or film deposition. XPS studies of FeL
1
-GCE
(Figure 3.34) reveal two iron signals with binding energies of 723 eV and 710 eV, corresponding
to the 2p3/2 and 2p1/2 levels respectively. The measured spin-orbit splitting of 13 eV is characteristic
of the Fe 2p region.
20
Additionally, the measured binding energies and lack of complex multiplet
splitting or sharp satellite features are indicative of a low-spin iron(II) material. XPS studies of
NiL
1
-GCE (Figure 3.34) reveal signals with binding energies of 873 eV and 856 eV,
corresponding to the 2p3/2 and 2p1/2 levels respectively. This large spin-orbit splitting of 17 eV is
characteristic of the Ni 2p region.
21
An additional broad feature is observed at 862 eV. These
features appear at higher binding energies than expected for nickel metal, and are more consistent
with Ni(OH)2. Together with the broad nature of these features, these results indicate the
deposition of a nickel-containing material with incorporation of oxygen, possibly from CO2 or
TFE. This is further supported by the observation that catalyst deposition does not occur in the
106

absence of CO2 and TFE. XPS analysis of the DMF-washed CoL
1
-GCE surface (Figure 3.34)
does not reveal any prominent signals indicative of the presence of cobalt.

Figure 3.35 Controlled potential electrolysis of a FeL
1
-GCE performed at -2.4 V, containing 0.1
M [NBu4][PF6] and 1.4 M TFE.

Figure 3.36 Controlled potential electrolysis of a NiL
1
-GCE performed at -2.9 V, containing 0.1
M [nBu4N][PF6] and 1.4 M TFE.

In order to test the possible electrocatalytic behavior of the FeL
1
-GCE and NiL
1
-GCE
materials, CPE studies were performed with these electrodes in the absence of molecular catalyst.
CPE experiments of FeL
1
-GCE at -2.4 V in DMF containing [nBu4N][PF6] (0.1 M) and TFE (1.4
M) under CO2 atmosphere reveal the production of 6.3 μmol of H2 with a FE of 94% and 0.3 μmol
of CO with a FE of 5% after 1 hour of electrolysis (Table 3.5 and Figure 3.35). A similar
107

experiment was performed with NiL
1
-GCE, producing 2.3 μmol of H2 and 0.7 μmol of CO after
1 hour, resulting in FE values of 47% and 14 %, respectively (Table 3.5 and Figure 3.36). These
studies suggest that, unlike the cobalt system, both FeL
1
and NiL
1
exhibit electrochemical
instability under catalytic conditions. Both complexes lead to the deposition of insoluble,
catalytically-active materials onto the working electrode whereas CoL
1
has been shown to remain
stable and solubilized throughout CPE studies. These results are consistent with the predicted
molecular orbital pictures of these complexes, as shown in Scheme 3.2. As discussed previously,
only CoL
1
is predicted to undergo two metal-based reduction events with the occupation of orbitals
of proper symmetry for interaction with an axial CO2 substrate. The poor performance of FeL
1

and NiL
1
is therefore likely to originate from these subtle differences in electronic structure. As
catalyst deposition is not observed in the presence of N2, our results indicate that FeL
1
and NiL
1

are capable of interacting with CO2 but ineffective at facilitating the two-electron transfer and
protonation necessary for catalytic turnover. Instead, this interaction with CO2 leads to degradation
of the complex and the generation of insoluble materials.

3.4 CONCLUSION
In conclusion, we have prepared iron and nickel analogues of an efficient cobalt-based
catalyst, CoL
1
. These complexes were characterized by single-crystal XRD,
1
H-NMR, UV-VIS,
and FTIR spectroscopy, and their electrochemical behavior was explored through CV and CPE
studies. The results of electrochemical studies indicate a periodic trend upon changing the metal
center from Fe to Co to Ni, with an anodic shift in reduction potentials for the M
2+/+
reductions
and a cathodic shift in reduction potentials for the M
+/0
reductions. Of the three complexes, CoL
1

is the best catalyst for CO2 reduction, exhibiting the highest rate, faradaic efficiency, and
selectivity for the conversion of CO2 to CO. The FeL
1
and NiL
1
complexes both produce syngas
108

mixtures of CO and H2 and operate with substantially lower current densities. Additionally, both
FeL
1
and NiL
1
were shown to exhibit instability under cathodic conditions in the presence of CO2,
leading to decomposition and deposition onto the electrode surface during electrolysis. The
deposited materials, FeL
1
-GCE and NiL
1
-GCE, were also shown to serve as competent catalysts
generating predominantly H2 in the case of FeL
1
-GCE and syngas mixtures in the case of NiL
1
-
GCE. These results are rationalized through DFT calculations, which indicate that only CoL
1

exhibits an electronic structure conducive to both σ-type and π-type interactions with CO2. The
two-electron reduction of CoL
1
is predicted to lead to the occupation of a Co(dz2) orbital and an
orbital of Co(dxz)+Co(dyz) character, and together these occupied orbitals facilitate productive
interactions with CO2 and allow for facile electron transfer and catalytic turnover. This work
broadens the scope of our efficient CoL1 catalyst design by extending these studies to earth-
abundant metal centers. The results of these studies highlights the crucial role played by the
transition metal ion in modulating the electronic structure and subsequent reactivity of these
molecular systems.

3.5 EXPERIMENTAL METHODS
3.5.1 Materials and Synthesis
All manipulations of air and moisture sensitive materials were conducted under a nitrogen
atmosphere in a Vacuum Atmospheres drybox or on a dual manifold Schlenk line. The glassware
was oven-dried prior to use. All solvents were degassed with nitrogen and passed through activated
alumina columns and stored over 4Å Linde-type molecular sieves. Deuterated solvents were dried
over 4Å Linde-type molecular sieves prior to use. Proton NMR spectra were acquired at room
temperature using Varian (Mercury 400 2-Channel, VNMRS-500 2-Channel, VNMRS- 600 3
Channel, and 400-MR 2-Channel) spectrometers and referenced to the residual
1
H resonances of
109

the deuterated solvent (
1
H: CDCl3 δ 7.26 and pyridine-d5 δ 7.22, 7.58, 8.74) and are reported as
parts per million relative to tetramethylsilane. Elemental analyses were performed by Robertson
Microlit Laboratories, Ledgewood, New Jersey.

3.5.2 Synthesis of Complexes
The ligand L1 and the complex CoL1 were prepared according to the previously reported
literature procedures.
9,10,30

FeL1. [Fe(H2O)6][BF4]2 (27.1 mg, 0.070 mmol) in 9:1 pyridine:DMF (1 mL) was added to a
solution of L1 (16.7 mg, 0.071 mmol) in pyridine (1 mL) giving rise to a brown solution. The
mixture was allowed to stir for 15 min and filtered through a microfiber filter. Slow diffusion with
diethyl ether produced red crystals, 70 % yield.
1
H NMR (500 Hz, pyridine-d5) δ 10.74 (s, 4H,
NH), 7.75 (t, J = 7.94 Hz, 4H, p-NC5H3), and 7.06 (d, J = 7.94 Hz, 8H, m-NC5H3). Anal. calcd for
[FeL1·2py·1DMF·1H2O]: C, 46.79; H, 4.16; N, 18.19. Found: C, 46.70; H, 4.13; N, 17.79

NiL1. [Ni(MeCN)6][BF4]2 (27.1 mg, 0.070 mmol) in 9:1 pyridine:DMF (1 mL) was added to a
solution of L1 (16.7 mg, 0.071 mmol) in pyridine (1 mL) giving rise to a brown solution. The
mixture was allowed to stir for 30 minutes. The solution was filtered through a microfiber filter.
Slow diffusion with diethyl ether produced amber crystals, 86 % yield.
1
H NMR (500 Hz, pyridine
d5) δ 49.86 (s, 4H, p-NC5H3) and 14.96 (s, 4H, NH). Anal. calcd for [NiL1·2py·1DMF·1H2O]: C,
46.63; H, 4.15; N, 18.13. Found: C, 46.71; H, 4.16; N, 17.76.

110

3.5.3 X-Ray Crystallography
The X-ray intensity data were measured on a Bruker APEX DUO system equipped with a
fine-focus tube (MoKα, λ = 0.71073 Å) and a TRIUMPH curved-crystal monochromator. The
frames were integrated with the Bruker SAINT software package using a SAINT V8.38A (Bruker
AXS, 2013) algorithm. Data were corrected for absorption effects using the multi-scan method
(SADABS). The SHELXTL XT 2014/5 (Bruker AXS, 2014) Software Package was used to
determine the structure solution with direct methods. The SHELXTL XT 2014/5 (Bruker AXS,
2014) Software Package was used for refinement by full-matric least-squares on F2. Additional
details are provided in Table 3.1-3.2.

3.5.4 Electrochemistry
Electrochemistry experiments were carried out using a Pine potentiostat. All experiments
in this paper were referenced relative decamethylferrocene (Fc*) with the Fe
3+/2+
couple at -0.48
V as an internal standard. All electrochemical experiments were performed with 0.1 M
tetrabutylammonium hexafluorophosphate as supporting electrolyte, which was recrystallized
before use. ML
1
complex concentrations were generally at 0.5 mM and experiments with CO2
were performed at gas saturation in dimethylformamide (DMF, 0.2 M).

Cyclic voltammetry (CV) experiments were performed in a single compartment
electrochemical cell under nitrogen or CO2 atmosphere using a 3 mm diameter glassy carbon
electrode as the working electrode, a platinum wire as auxiliary electrode and a silver wire as the
reference electrode. Controlled-potential electrolysis (CPE) measurements were conducted in a
two-chambered H cell. The first chamber held the working and reference electrodes in 40 mL of
111

0.1 M tetrabutylammonium hexafluorophosphate and 1.4 M TFE in DMF. The second chamber
held the auxiliary electrode in 20 mL of 0.1 M tetrabutylammonium hexafluorophosphate in DMF.
The two chambers were separated by a fine porosity glass frit. The reference electrode was
separated from solution by a Vycor tip. Glassy carbon plate electrodes (GCE) (6 cm × 1 cm × 0.3
cm; Tokai Carbon USA) were used as the working and auxiliary electrodes. After CPE
experiments were performed for the ML
1
complexes, the GCE electrode was saved either as is or
rinsed with DMF and dried under vacuum atmosphere until further use. These GCE electrodes
were used against in CPE experiments to test the performance of the deposited materials. Gas
analysis for CPE experiments were performed using 2 mL sample aliquots taken from the
headspace of the electrochemical cell and injected on a Shimadzu BID-2010 plus series gas
chromatograph with a 2m × 1mm ID micropacked column. Faradaic efficiencies were determined
by dividing he measured gas produced by the amount of gas expected based on the charge passed
during the bulk electrolysis experiment

3.5.5 Nuclear Magnetic Resonance Spectroscopy
Evan’s method was used to determine the total spin (S) of a metal complex by
1
H NMR
spectroscopy.
29
Equation 3.1 is used to determine the MMs (Measured Molar Susceptibility). ΔHz
is the difference in hertz between the peaks of the solvent in contact with the complex and the ones
in the capillary tube, and M is the molarity of the sample (in units of mol/L), and HzNMR is the
spectrometer frequency, in hertz (500,000,000).
MMs=
3000 x ΔHz
4π x M x Hz
NMR
Eq. 3.1
Subsequently, Eq. 3.2-3.3 were used to determine the number of unpaired electrons.
X
P
= MMs− X
D
, 𝑋 𝐷 =
mM
2
x 10
−6
Eq. 3.2
112

μ
eff
= 2.84√T x X
P
Eq. 3.3
XP indicates the corrected molar susceptibility, mM is the molar mass of the complex
(g/mol), and T is the temperature in Kelvin (298 K).

3.5.6 Gas Chromatography
Gaseous products were quantified using a Shimadzu GC-2010-Plus instrument equipped
with a BID detector and a Restek ShinCarbon ST Micropacked column. In a typical experiment, 2
mL of gas were withdrawn from the headspace of the electrochemical cell with a gas-tight syringe
and injected into the instrument. Calibration plots were prepared with multiple injections of syngas
standards purchased from Praxair, Inc. Faradaic efficiencies were determined by dividing the
amount of gaseous product produced as measured by gas chromatography by the amount of gas
expected based on the total charge measured during controlled potential electrolysis. Multiple runs
were performed for each condition studied, leading to similar behavior. The reported μmol of gas
produced (and subsequently FE and TON) are averaged values and error bars are determined from
multiple experiments.

3.5.7 X-Ray Photoelectron Spectroscopy
XPS data were collected using a Kratos AXIS Ultra instrument. The monochromatic X-ray
S2 source was the Al K α line at 1486.7 eV, and the hybrid lens and slot mode were used. Low
resolution survey spectra were acquired between binding energies of 1–1200 eV. Higher resolution
detailed scans, with a resolution of 0.1 eV, were collected on individual XPS regions of interest.
The sample chamber was maintained at < 9 × 10
–9
Torr. The XPS data were analyzed using the
CasaXPS software.
113

3.5.8 UV-Vis and FTIR Spectroscopy
UV-Vis spectra were collected using a UV-1800 Shimadzu UV spectrophotometer.
Samples were studied in transmittance mode with a 1 cm quartz cuvette, and the spectrum
measured for a blank dimethylformamide sample was subtracted as background. FTIR spectra
were acquired using a Bruker Vertex 80v spectrometer on a KBr pressed pellet.

3.5.9 Computational Methods
All calculations were run using the Q-CHEM program package.
22
Gas-phase calculations
were performed and axially-bound solvent molecules were neglected for computational simplicity.
Geometry optimizations were run with unrestricted DFT calculations at the B3LYP level of theory
with the Pople 6-31+G(d) basis set for all atoms.
23-28
All optimized geometries were verified as
stable minima with frequency calculations at the same level of theory. The B3LYP functional was
used throughout this study based on the prior successful application of this functional to
mechanistic modelling of the cobalt complex.
10
Time-dependent density functional theory
calculations were run at the B3LYP level of theory with the Pople 6-31+G(d) basis set for all
atoms. Only transitions with predicted oscillator strengths >0.005 were considered for a qualitative
picture of the character of the electronic transitions for these systems. Kohn-Sham orbital images
are presented with isovalues of 0.05 for clarity. Basis sets were retrieved from the Basis Set
Exchange.

114

3.5.10 Nuclear Coordinates for Optimized Geometries
Table 3.6 Optimized structure for FeL
1

I Atom X Y Z
1 Fe -0.0181593 -0.0000034 -0.0310541
2 N 1.5440321 1.8724235 -1.8913117
3 N 1.2349857 -0.4375572 -1.519021
4 N 0.600926 1.3190074 2.6974382
5 N -1.0833952 -1.6248409 -0.4503009
6 N -1.2142752 0.3847834 1.5172963
7 N 1.0596001 1.6139647 0.4000108
8 N 0.8748043 -2.7400656 -1.1567626
9 N -3.0238377 -0.5165082 0.2985664
10 C -0.743442 0.9615649 2.6584024
11 C -0.505238 -2.7557636 -0.9272812
12 C 1.8343917 0.5532736 -2.239869
13 C 1.2322262 2.038994 1.6765691
14 C 1.526285 -1.7261643 -1.8515506
15 C 1.6519974 2.3443319 -0.5779611
16 C -2.4273898 -1.6505618 -0.2656879
17 C -1.5528062 1.2007721 3.7691315
18 C 2.4431519 -2.0564399 -2.8502341
19 C 2.7007033 0.2927786 -3.2991161
20 H 1.8843373 2.5400217 -2.5747254
21 C -2.5434607 0.0852215 1.461336
22 C -2.9103998 0.9132658 3.6792111
23 H 0.9006497 1.610565 3.6213632
24 C -3.2223001 -2.7521117 -0.5834596
25 C -1.2234482 -3.9181652 -1.2126022
26 H -4.0369181 -0.555633 0.2711112
27 C 3.0214074 -1.0307684 -3.5894075
28 C -2.6042915 -3.9058469 -1.0530702
29 H 1.2455166 -3.6600083 -1.3681329
30 C 1.9883431 3.1636162 2.01064
31 C -3.4220268 0.3669836 2.5055471
32 C 2.3744233 3.5118566 -0.3302241
33 C 2.5521161 3.9176656 0.9876848
34 H -1.133847 1.6509157 4.6630399
35 H 2.6562609 -3.0966956 -3.0730779
36 H 3.1500089 1.1114644 -3.851383
37 H -3.5682889 1.1166164 4.5188867
38 H -4.2943642 -2.7168856 -0.4197425
39 H -0.7113731 -4.7959824 -1.5928184
40 H 3.7168276 -1.2605739 -4.3911585
41 H -3.1924504 -4.7875504 -1.2886755
42 H 2.0933168 3.461188 3.0488067
115

43 H -4.4745341 0.1196122 2.4153562
44 H 2.8235121 4.0599005 -1.1520326
45 H 3.1266779 4.810311 1.2154648
Nuclear Repulsion Energy = 3300.9075895661 hartrees
Zero point vibrational energy: 219.484 kcal/mol

Table 3.7 Optimized structure for CoL1

I Atom X Y Z
1 Co -0.0833029 0.0624784 -0.09964
2 N 1.6107048 1.8794504 -1.8812637
3 N 1.2489643 -0.4354455 -1.5929908
4 N 0.5272828 1.4386859 2.6985873
5 N -1.1247889 -1.6769331 -0.4732242
6 N -1.2742594 0.4292907 1.5423873
7 N 1.0900756 1.6794561 0.4117902
8 N 0.8293134 -2.7437544 -1.2790691
9 N -3.0430739 -0.5639096 0.3345551
10 C -0.8100372 1.0331754 2.662782
11 C -0.5393261 -2.782279 -0.994615
12 C 1.8896385 0.5555172 -2.2628683
13 C 1.2110572 2.120836 1.6871354
14 C 1.508928 -1.7181518 -1.9428063
15 C 1.739447 2.3582974 -0.566791
16 C -2.456756 -1.7151666 -0.2185801
17 C -1.6197036 1.2601373 3.7795346
18 C 2.4324689 -2.0488661 -2.938952
19 C 2.7873892 0.3050244 -3.2982889
20 H 1.9514873 2.5493142 -2.56279
21 C -2.5806968 0.0650253 1.5036165
22 C -2.9599101 0.8969547 3.7151369
23 H 0.8096543 1.7435221 3.6234994
24 C -3.2424906 -2.8341738 -0.4882176
25 C -1.2533997 -3.9562022 -1.2499044
26 H -4.0564136 -0.6044489 0.3033576
27 C 3.0658306 -1.0203318 -3.626715
28 C -2.6215241 -3.9702344 -1.0027145
29 H 1.1972374 -3.6594111 -1.5121966
30 C 1.9962847 3.226663 2.0240498
31 C -3.4589466 0.2990017 2.5596043
32 C 2.5054365 3.4947889 -0.3133944
33 C 2.6376961 3.9229307 1.0056102
34 H -1.2151667 1.7477939 4.6605426
35 H 2.6167813 -3.0875316 -3.1934434
36 H 3.2859022 1.124319 -3.8058724
37 H -3.6147861 1.0773897 4.5623329
38 H -4.3041686 -2.8290522 -0.2644322
116

39 H -0.7531754 -4.82234 -1.6707972
40 H 3.7753204 -1.2487613 -4.4164064
41 H -3.2032613 -4.8639448 -1.2079444
42 H 2.0679353 3.5549197 3.0558847
43 H -4.4958363 -0.0135731 2.4934375
44 H 3.015771 4.0068385 -1.1225654
45 H 3.2406723 4.7956723 1.2378413
Nuclear Repulsion Energy = 3296.3577340910 hartrees
Zero point vibrational energy: 218.762 kcal/mol

Table 3.8 Optimized structure for NiL1

I Atom X Y Z
1 Ni 0.0051055 -0.0210537 -0.0063185
2 N 1.5702873 1.8725978 -1.868561
3 N 1.2549758 -0.4422113 -1.5481836
4 N 0.5853632 1.3239384 2.7100981
5 N -1.0867095 -1.6739557 -0.4420712
6 N -1.2400065 0.3967014 1.5399955
7 N 1.0995346 1.630714 0.4291223
8 N 0.8868027 -2.7407931 -1.1708493
9 N -3.0171136 -0.5456793 0.3088431
10 C -0.7756213 0.9888761 2.6660322
11 C -0.4956756 -2.7855261 -0.9419466
12 C 1.8442737 0.5522364 -2.2544885
13 C 1.2372534 2.056477 1.7073421
14 C 1.5268895 -1.7233723 -1.8923331
15 C 1.7027024 2.3411168 -0.553655
16 C -2.4253625 -1.6948268 -0.2360427
17 C -1.5970237 1.2694676 3.7586148
18 C 2.4074744 -2.0516699 -2.9240827
19 C 2.696925 0.3039872 -3.3304396
20 H 1.8955804 2.5480699 -2.5516147
21 C -2.5573242 0.0894543 1.4715788
22 C -2.9529236 0.9694818 3.6632227
23 H 0.8818466 1.6000245 3.6401482
24 C -3.2136929 -2.8047194 -0.542064
25 C -1.2096705 -3.9498086 -1.2280331
26 H -4.0300312 -0.5793859 0.2656738
27 C 2.9871453 -1.0184796 -3.6546464
28 C -2.5879305 -3.9469556 -1.0336086
29 H 1.2721773 -3.656759 -1.3748032
30 C 1.9921514 3.1800435 2.0457218
31 C -3.4500143 0.3816506 2.5034754
32 C 2.4359761 3.5007349 -0.298905
33 C 2.58701 3.9124344 1.0220925
117

34 H -1.1902996 1.7484587 4.643258
35 H 2.6002207 -3.0901491 -3.1726899
36 H 3.1535772 1.1265169 -3.8711202
37 H -3.620073 1.1928759 4.4903548
38 H -4.2832216 -2.7864183 -0.3600268
39 H -0.7027342 -4.8203423 -1.6312021
40 H 3.6622878 -1.243033 -4.474957
41 H -3.1727342 -4.8324841 -1.2639846
42 H 2.0766698 3.4940321 3.0809197
43 H -4.4987077 0.117758 2.4142787
44 H 2.9085871 4.0416056 -1.1122415
45 H 3.1663081 4.8013411 1.2532981

Nuclear Repulsion Energy = 3346.5095360160 hartrees
Zero point vibrational energy: 218.991 kcal/mol

118

3.6 REFERENCES
1. Benson, E. E.; Kubiak, C. P.; Sathrum, A. J.; Smieja, J. M. Electrocatalytic and
Homogeneous Approaches to Conversion of CO2 to Liquid Fuels. Chem. Soc. Rev. 2009,
38 (1), 89–99.
2. Costentin, C.; Robert, M.; Savéant, J. M. Current Issues in Molecular Catalysis Illustrated
by Iron Porphyrins as Catalysts of the CO2-to-CO Electrochemical Conversion. Acc. Chem.
Res. 2015, 48 (12), 2996–3006.
3. Francke, R.; Schille, B.; Roemelt, M. Homogeneously Catalyzed Electroreduction of
Carbon Dioxide - Methods, Mechanisms, and Catalysts. Chem. Rev. 2018, 118 (9), 4631–
4701.
4. Artz, J.; Müller, T. E.; Thenert, K.; Kleinekorte, J.; Meys, R.; Sternberg, A.; Bardow, A.;
Leitner, W. Sustainable Conversion of Carbon Dioxide: An Integrated Review of Catalysis
and Life Cycle Assessment. Chem. Rev. 2018, 118 (2), 434–504.
5. Jeoung, J.-H.; Dobbek, H. Carbon Dioxide Activation at the Ni,Fe-Cluster of Anaerobic
Carbon Monoxide Dehydrogenase. Science. 2010, 330, 1461–1465.
6. Chen, L.; Guo, Z.; Wei, X. G.; Gallenkamp, C.; Bonin, J.; Anxolabéhère-Mallart, E.; Lau,
K. C.; Lau, T. C.; Robert, M. Molecular Catalysis of the Electrochemical and
Photochemical Reduction of CO2 with Earth-Abundant Metal Complexes. Selective
Production of CO vs HCOOH by Switching of the Metal Center. J. Am. Chem. Soc. 2015,
137 (34), 10918–10921.
7. Guo, Z.; Cheng, S.; Cometto, C.; Anxolabéhère-Mallart, E.; Ng, S.-M.; Ko, C.-C.; Liu, G.;
Chen, L.; Robert, M.; Lau, T.-C. Highly Efficient and Selective Photocatalytic CO2
Reduction by Iron and Cobalt Quaterpyridine Complexes. J. Am. Chem. Soc. 2016, 138
(30), 9413–9416.
8. Cometto, C.; Chen, L.; Lo, P. K.; Guo, Z.; Lau, K. C.; Anxolabéhère-Mallart, E.; Fave, C.;
Lau, T. C.; Robert, M. Highly Selective Molecular Catalysts for the CO 2-to-CO
Electrochemical Conversion at Very Low Overpotential. Contrasting Fe vs Co
Quaterpyridine Complexes upon Mechanistic Studies. ACS Catal. 2018, 8 (4), 3411–3417.
9. Chapovetsky, A.; Do, T. H.; Haiges, R.; Takase, M. K.; Marinescu, S. C. Proton-Assisted
Reduction of CO2 by Cobalt Aminopyridine Macrocycles. J. Am. Chem. Soc. 2016 (138),
5765–5768.
10. Chapovetsky, A.; Welborn, M.; Luna, J. M.; Haiges, R.; Miller, T. F.; Marinescu, S. C.
Pendant Hydrogen-Bond Donors in Cobalt Catalysts Independently Enhance CO2
Reduction. ACS Cent. Sci. 2018, 4 (3), 397–404.
11. Zhang, E. X.; Wang, D. X.; Huang, Z. T.; Wang, M. X. Synthesis of (NH)m(NMe)4-m-
Bridged Calix[4]Pyridines and the Effect of NH Bridge on Structure and Properties. J. Org.
Chem. 2009, 74 (22), 8595–8603.
119

12. Elgrishi, N.; Chambers, M. B.; Wang, X.; Fontecave, M. Molecular Polypyridine-Based
Metal Complexes as Catalysts for the Reduction of CO2. Chem. Soc. Rev. 2017, 46 (3),
761–796.
13. Pun, S.; Chung, W.; Lam, K.; Guo, P.; Chan, P.; Wong, K.; Che, C.; Chen, T.; Peng, S.
Iron(I) Complexes of 2,9-Bis(2-Hydroxyphenyl)-1,10-Phenanthroline (H2dophen) as
Electrocatalysts for Carbon Dioxide Reduction. X-Ray Crystal Structures of
[Fe(Dophen)Cl]2_2HCON(CH3)2 and [Fe(Dophen)(N-MeIm)2]ClO4 (N-MeIm 1-
Methylimidazole). J. Chem. Soc., Dalton Trans. 2002, 575–583.
14. Piguet, C. Paramagnetic Susceptibility by NMR : The “ Solvent Correction ” Removed for
Large Paramagnetic Molecules. J. Chem. Ed. 1997, 74 (7), 815–816.
15. Russo, T. V.; Martin, R. L.; Hay, P. J. Application of Gradient-Corrected Density
Functional Theory to the Structures and Thermochemistries of ScF3, TiF4, VF5, and CrF6.
J. Chem. Phys. 1995, 102, 8023–8028.
16. Li, Z.; Liu, W. Critical Assessment of TD-DFT for Excited States of Open-Shell Systems:
I. Doublet-Doublet Transitions. J. Chem. Theory Comput. 2016, 12, 238–260.
17. Mealli, C.; Hoffman, R.; Stockis, A. A Molecular Orbital Analysis of the Bonding
Capabilities of Carbon Disulfide and Carbon Dioxide toward Transition-Metal Fragments
Inorg. Chem. 1984, 23, 65-74.
18. Smieja, J. M.; Benson, E. E.; Kumar, B.; Grice, K. A.; Seu, C. S.; Miller, A. J. M.; Mayer,
J. M.; Kubiak, C. P. Kinetic and Structural Studies, Origins of Selectivity, and Interfacial
Charge Transfer in the Artificial Photosynthesis of CO. Proc. Natl. Acad. Sci. 2012, 109
(39), 15646–15650.
19. Saveant, J.-M. Molecular Catalysis of Electrochemical Reactions. Mechanistic Aspects.
Chem. Rev. 2008, 108, 2348–2378.
20. Grosvenor, A. P.; Kobe, B. A.; Biesinger, M. C.; McIntyre, N. S. Investigation of multiplet
splitting of Fe 2p XPS spectra and bonding in iron compounds. Surf. Interface Anal. 2004,
36, 1564-1574.
21. Grosvenor, A. P.; Biesinger, M. C.; Smart, R. St.C.; McIntyre, N. S. New interpretations
of XPS spectra of nickel metal and oxides. Surface Science 2006, 600, 1771–1779.

120

22. Shao, Y.; Molnar, L.; Jung, Y.; Kussmann, J.; Ochsenfeld, C.; Brown, S.; Gilbert, A.;
Slipchenko, L.; Levchenko, S.; O'Neill, D.; DiStasio, R.; Lochan, R.; Wang, T.; Beran, G.;
Besley, N.; Herbert, J.; Lin, C.; Van Voorhis, T.; Chien, S.; Sodt, A.; Steele, R.; Rassolov,
V.; Maslen, P.; Korambath, P.; Adamson, R.; Austin, B.; Baker, J.; Byrd, E.; Dachsel, H.;
Doerksen, R.; Dreuw, A.; Dunietz, B.; Dutoi, A.; Furlani, T.; Gwaltney, S.; Heyden, A.;
Hirata, S.; Hsu, C.; Kedziora, G.; Khalliulin, R.; Klunzinger, P.; Lee, A.; Lee, M.; Liang,
W.; Lotan, I.; Nair, N.; Peters, B.; Proynov, E.; Pieniazek, P.; Rhee, Y.; Ritchie, J.; Rosta,
E.; Sherrill, C.; Simmonett, A.; Subotnik, J.; Woodcock, H.; Zhang, W.; Bell, A.;
Chakraborty, A.; Chipman, D.; Keil, F.; Warshel, A.; Hehre, W.; Schaefer, H.; Kong, J.;
Krylov, A.; Gill, P.; Head-Gordon, M. Advances in Methods and Algorithms in a Modern
Quantum Chemistry Program Package. Phys. Chem. Chem. Phys. 2006, 8, 3172-3191.
23. Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient Approximation Made
Simple. Phys. Rev. Lett. 1996, 77, 3865–3868.
24. Ditchfield, R.; Hehre, W. J.; Pople, J. A. Self-Consistent Molecular-Orbital Methods. IX.
An Extended Gaussian-Type Basis for Molecular-Orbital Studies of Organic Molecules. J.
Chem. Phys. 1971, 54, 724–728.
25. Hehre, W. J.; Ditchfield, K.; Pople, J. A. Self-Consistent Molecular Orbital Methods. XII.
Further Extensions of Gaussian-Type Basis Sets for Use in Molecular Orbital Studies of
Organic Molecules. J. Chem. Phys. 1972, 56, 2257–2261.
26. 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.
27. Feller, D. The role of databases in support of computational chemistry calculations J.
Comput. Chem. 1996, 17, 1571-1586.
28. Schuchardt, K. L.; Didier, B. T.; Elsethagen, T; Sun, L.; Gurumoorthi, V.; Chase, J.; Li, J.;
Windus, T. L. Basis Set Exchange: A Community Database for Computational Sciences J.
Chem. Inf. Model. 2007, 47, 1045-1052.
29. Piguet, C. Paramagnetic Susceptibility by NMR : The “ Solvent Correction ” Removed for
Large Paramagnetic Molecules. J. Chem. Ed. 1997, 74 (7), 815–816.
30. Zhang, E. X.; Wang, D. X.; Huang, Z. T.; Wang, M. X. Synthesis of (NH)m(NMe)4-m-
Bridged Calix[4]Pyridines and the Effect of NH Bridge on Structure and Properties. J. Org.
Chem. 2009, 74 (22), 8595–8603.
121

CHAPTER 4

Surface-Immobilized Conjugated Polymers Incorporating Rhenium Bipyridine Motifs for
Electrocatalytic and Photocatalytic CO2 Reduction

A portion of this chapter has appeared in print:

Orchanian, N. M.; Hong, L. E.; Skrainka, J. A.; Esterhuizen, J. A.; Popov, D. A.; Marinescu, S.
C.* “Surface-Immobilized Conjugated Polymers Incorporating Rhenium Bipyridine Motifs for
Electrocatalytic and Photocatalytic CO2 Reduction”, ACS Appl. Energy Mater., 2019, 2 (1), 110-
23.
122

4.1 ABSTRACT
The solar-driven conversion of CO2 to value-added products provides a promising route
for solar energy storage and atmospheric CO2 remediation. In this report, a variety of supporting
electrode materials were successfully modified with a [2,2'-bipyridine]-5,5'-bis(diazonium)
rhenium complex through a surface-localized electropolymerization method. Physical
characterization of the resulting multilayer films confirms that the coordination environments of
the rhenium bipyridine tricarbonyl sites are preserved upon immobilization and that the
polymerized catalyst moieties exhibit long-range structural order with uniform film growth. UV-
Vis studies reveal additional absorption bands in the visible region for the polymeric films that are
not present in the analogous rhenium bipyridine complexes. Electrochemical studies with modified
graphite rod electrodes show that the electrocatalytic activity of these films increases with catalyst
loading up to an optimal value, beyond which electron and mass transport through the material
become rate-limiting. Electrocatalytic studies performed at -2.25 V vs Fc/Fc
+
for 2 hours reveal
CO production with faradaic efficiencies and turnover numbers up to 99% and 3606, respectively.
Photocatalytic studies of the modified TiO2 devices demonstrate enhanced activity at low catalyst
loadings, with turnover numbers up to 70 during 5 hours of irradiation.

Keywords: Solar Energy Conversion, Electrocatalysis, Photocatalysis, Rhenium Bipyridine,
Surface Modification, Metallopolymers

4.2 INTRODUCTION
As the rapid growth in global population continues, the demands for sustainable energy
sources and energy storage technologies are expected to increase.
1
The annual availability of solar
energy (23,000 TW per year) provides a potentially inexhaustible source of renewable, clean, and
distributed power for the planet.
2,3
However, the intermittency of solar energy requires the
123

development of technologies to harness, store, and transport it efficiently. Storing this energy in
the form of chemical bonds would serve as an ideal solution to the global energy challenges.
4
This
transformation could be accomplished either by directly driving chemical reactions with solar
photons (photocatalysis), by using solar-derived electricity to drive electrolysis (electrocatalysis),
or a combination of both (photoelectrocatalysis).
5
In particular, the catalytic conversion of carbon
dioxide (CO2) to carbon monoxide (CO) presents an opportunity to transform an abundant small
molecule to a value-added C1 chemical feedstock. However, this reaction suffers from a broad
product distribution and competitive hydrogen evolution. As such, specialized catalysts are
required for the selective conversion of CO2 to a single value-added product.
Both homogeneous and heterogeneous catalysts have been developed for the CO2 reduction
reaction (CO2RR), each with their respective benefits and drawbacks.
6
The well-defined active
sites of homogeneous catalysts enable thorough mechanistic studies and subsequent synthetic
optimization, though these systems typically suffer from limited stability, diffusion-dependent
kinetics, and poor recyclability.
7,8
In contrast, heterogeneous catalysts allow for increased
lifetimes, diminished diffusion limitations, improved recyclability, and the elimination of catalyst
separation cost, though their ill-defined surface structures prohibit rational improvement and
typically yield low selectivity for CO2 to CO conversion.
9
An alternative to these pathways is the
heterogenization of molecular catalysts by anchoring molecular species to a surface support.
10
This
has been shown to improve the activity of these molecular species and effectively combine the
benefits of homogeneous and heterogeneous systems.
Of the known molecular catalysts for CO2RR, the rhenium (I) bipyridine tricarbonyl
chloride complex (Re(bpy)(CO)3Cl) is one of the most well-studied to date, offering high activity
and selectivity for CO production under both electrocatalytic and unsensitized photocatalytic
124

conditions.
11,12
The heterogenization of Re(bpy)(CO)3Cl has been successfully applied to a wide
range of surfaces through both non-covalent and covalent interactions.
13,14
In some cases, these
architectures perform efficiently with high activity and orders of magnitude lower usage of
expensive metals relative to their homogeneous counterparts.
15
However, the functional groups
selected for surface attachment limit the substrate scope of these methods. This limitation hinders
catalyst development by preventing parallel studies of the heterogenized species across various
supporting electrodes, which would allow for the translation of these devices between electrode
architectures amenable to electrocatalysis and those amenable to photocatalysis.
Electropolymerization presents a promising opportunity to generate nanostructured
molecular films with broad substrate scope and tunable catalyst coverages.
16
In particular, the
electropolymerization of rhenium bipyridine complexes has been studied using a variety of
polymerization methods, monomer units, and substrates.
17
Following a report on the
electropolymerization of a rhenium 4-vinyl-4'-methyl-2,2'-bipyridine complex, Re(vbpy)(CO)3Cl,
onto platinum disk electrodes, this procedure was subsequently applied to the modification of a
variety of semiconductor materials.
18,19
These modified electrodes were shown to exhibit high
turnover numbers (TON) and faradaic efficiencies (FE) for the electrocatalytic CO2RR to CO, with
TON = 516 and FE = 90% after 80 min. at -1.55 V vs SCE for a vinyl-substituted rhenium complex
immobilized onto a platinum electrode. It was reported that diluting the rhenium complexes with
a ruthenium tris(bipyridine) photosensitizer generated materials with improved stability and
activity, with a TON = 3800 at -1.55 V vs SCE for a 30% Re and 70% Ru copolymer film (FE not
reported).
20
Although this technique was applied to semiconducting substrates for
photoelectrocatalytic studies, unbiased photocatalytic studies were not reported.
21
Despite
promising activity, the flexibility imparted by the methylene spacers and the generation of vinyl
125

radicals led to undesirable side-reactions during grafting, including the formation of Re-Re and
Re-C bonds via radical-radical coupling. Consequently, these films were shown to exhibit limited
stability, as anodic polarization led to film degradation, which was attributed to Re-C bond
cleavage.
In contrast to the flexibility of methylene-spaced polymers generated by vinyl
polymerization, polymers with conjugated backbones display structural rigidity.
22
Several reports
on rigid metallopolymers suggest that this class of materials displays promising properties for
photocatalytic applications.
23
Studies on the conjugated poly([2,2'-bipyridine]-5,5'-diyl) and
related metallopolymers indicate that these structures present unique photophysical properties,
including intraligand π-π* transitions and metal-to-ligand charge transfer (MLCT) bands, which
facilitate photocatalytic H2-evolving activity.
24,25
Rigid rhenium bipyridine metallopolymers with
aryleneethynylene architectures were reported to exhibit π-π* transitions as well as dπ(Re)-πpolymer
charge transfer bands, though the catalytic applications of these materials have not been reported.
26

An analogous polymeric material generated from an alkyne-substituted rhenium bipyridine
complex was shown to generate rigid polymers with electrocatalytic activity for CO2RR, but these
materials performed poorly (FE for CO production = 33%) relative to the vinyl-polymerized
materials and no studies under illumination were reported.
27
Due to the promising photophysical
properties of poly([2,2'-bipyridine]-5,5'-diyl) materials and their underexplored catalytic
applications, we sought to develop a method for the growth of surface-immobilized polymers with
a poly(Re(CO)3Cl[2,2'-bipyridine]-5,5'-diyl) structure.
Diazonium chemistry offers a promising route for surface modification with a broad
substrate scope, improved stability, and aryl-bond formation.
28,29
A previous report on the
electrochemical grafting of p-bis(diazonium)benzene has shown that bis(diazonium) salts offer
126

several advantages over the typically-employed mono(diazonium) analogues for the generation of
fully-conjugated films.
30
The p-bis(diazonium)benzene unit was covalently anchored to the
supporting electrode surface following the electrochemical reduction of a single diazonium group,
while preserving the second diazonium group for subsequent chemical functionalization. The
diazonium-terminated film could be directed either toward aryl- or azo-bond coupling reactions,
though polymerization studies were not reported. We sought to modify this method of surface
functionalization to generate surface-immobilized conjugated polymers incorporating rhenium
bipyridine motifs. By targeting the synthesis of a [2,2'-bipyridine]-5,5'-bis(diazonium) rhenium
complex (2), we sought to explore a 5'-directed electropolymerization mechanism for improved
control over film structure and morphology. Grafting of this bis(diazonium) species is expected to
generate a diazonium-terminated film, facilitating continued film growth. The structural rigidity
and potential long-range order of the resulting films are expected to hinder dimer formation. The
substrate scope of this method is investigated by modifying a variety of supporting electrode
surfaces. Following grafting studies, the structure, morphology, and electrochemical behavior of
these modified electrodes are investigated. Finally, the electrocatalytic and unbiased photocatalytic
CO2RR activity of these devices are explored.

127

4.3 RESULTS AND DISCUSSION
4.3.1 Synthesis of [2,2'-bipyridine]-5,5'-bis(diazonium) rhenium complex (2)

Scheme 4.1 Synthesis of 2 with (i) 2.4 eq. NOBF4 in anhydrous acetonitrile solutions at -40 °C.
(ii) Electrochemical grafting of bis(diazonium) complex 2 by cyclic voltammetry in a 0.5 mM
solution of 2 in acetonitrile with 0.1 M TBAPF6 supporting electrolyte. (iii) Electropolymerization
of 2 by subsequent cyclic voltammetry scans.

The rhenium [2,2'-bipyridine]-5,5'-diamine complex 1 (Scheme 4.1) was synthesized
according to previous reports for similar complexes, by metalation of the [2,2'-bipyridine]-5,5'-
diamine ligand with rhenium (I) pentacarbonyl chloride in refluxed toluene solution.
31
The
1
H
NMR spectrum of 1 in acetonitrile-d3 (Figure 4.1) displays peaks at  8.71, 7.64 and 7.25 ppm,
attributed to the aromatic protons, as well as a broad peak at  4.82 ppm which corresponds to the
amine (–NH2) protons. Treatment of the rhenium complex 1 with nitrosonium tetrafluoroborate
(2.4 equivalents) in anhydrous acetonitrile solutions at -40 °C, leads to an immediate color change
from pale-yellow to dark blue. Addition of diethyl ether results in the formation of a dark blue
precipitate, which was collected by filtration, and stored in the dark at -27 °C. The dark blue
precipitate was characterized by a variety of spectroscopic studies, such as
1
H and
19
F NMR, ATR-
FTIR, UV-Vis, and XPS. The
1
H NMR spectrum of the dark blue precipitate displays peaks that
are slightly shifted from that of the diamine complex 1; the amine peak is no longer present and
the aromatic peaks are shifted downfield (  10.10, 9.29 and 9.02 ppm), which indicates the clean
formation of the desired rhenium bis(diazonium) complex 2 with strongly electron-withdrawing
128

diazonium groups (Figure 4.2). The
19
F spectrum exhibits a peak at -151.6 ppm, which is
characteristic of the tetrafluoroborate anion (Figure 4.3). The ATR-FTIR spectrum of this
complex contains characteristic carbonyl stretches for a fac-Re(CO)3 species at 2033 cm
-1
, 1957
cm
-1
, and 1929 cm
-1
, with an additional diazonium stretch appearing at 2310 cm
-1
(Figure 4.4).
These experimental IR frequencies are consistent with the gas-phase calculated frequencies using
M06 level of theory with the LANL2DZ basis set and ECP, as implemented in the Q-Chem
software package (Figure 4.5, computational methods detailed in experimental methods).
32,33,34

The UV-Vis spectrum of complex 2 maintains the MLCT band for the parent complex 1 at 326
nm, with an additional broad feature at 612 nm (Figure 4.6). XPS survey results confirm the
presence of Re, C, O, N, Cl, B, and F in solid 2 (Figure 4.7). The high-resolution Re 4f region is
consistent with that of complex 1, with features at binding energies of 43.4 and 40.9 eV, which
correspond to the Re 4f5/2 and 4f7/2 levels, respectively. These binding energies are similar with
those of other reported rhenium (I) bipyridine tricarbonyl moieties,
13
suggesting no change in the
oxidation state of rhenium upon addition of NOBF4 (Figure 4.8). The Cl 2p region of complex 2
displays features at 198.9 eV and 197.1 eV, corresponding to the Cl 2p1/2 and Cl 2p3/2 levels
(Figure 4.9), which are consistent with other reported rhenium species. The presence of the Cl 2p
features confirms that the chloride ligand is not displaced upon diazotization. In contrast to
complex 1, the N 1s region of complex 2 exhibits a peak at 404 eV, indicative of diazonium
moieties (Figure 4.10). The high-resolution B 1s and F 1s regions of 2 display features with
binding energies at 192.1 and 682.9 eV, respectively, which correspond to the tetrafluoroborate
anion (BF4
-
) (Figure 4.11-4.12).

129

Figure 4.1
1
H NMR spectrum of 1 in acetonitrile-d3.

Figure 4.2
1
H NMR spectrum of 2 in acetonitrile-d3.

Figure 4.3
19
F NMR spectrum of 2 in acetonitrile-d3.

Figure 4.4 ATR FTIR spectrum of complex 2.

130

Figure 4.5 Calculated vibrational spectrum of 2 at the LANL2DZ/M06 level of theory. Only
stretching modes with calculated intensity > 200 are included for clarity.

Figure 4.6 UV-Vis spectrum of 2 in an acetonitrile solution (0.5 mM 2).

Figure 4.7 XPS survey scan for complex 2.

131

Figure 4.8 High-resolution XPS of the Re 4f region for complex 2.

Figure 4.9 High-resolution XPS of the Cl 2p region for complex 2.

Figure 4.10 High-resolution XPS of the N 1s region for complex 2.

132

Figure 4.11 High-resolution XPS of the B 1s region for complex 2.

Figure 4.12 High-resolution XPS of the F 1s region for complex 2.

4.3.2 Electropolymerization of 2
Figure 4.13a presents consecutive cyclic voltammetry (CV) scans (ν = 100 mV/s) for a
glassy carbon (GC) working electrode immersed in a 0.5 mM solution of 2 with 0.1 M TBAPF6
supporting electrolyte in acetonitrile. On the initial scan, a broad irreversible reduction feature is
observed at –0.71 V vs Fc/Fc
+
(all subsequent potentials are referenced to the
ferrocene/ferrocenium reversible couple). This broad reduction feature appears at potentials much
more positive than those of the rhenium bipyridine species (-1.74 V and -2.12 V vs Fc/Fc
+
for 1),
and it is assigned to the one-electron reduction of a diazonium group. This reduction leads to
liberation of dinitrogen and subsequent formation of an aryl radical in close proximity to the
133

electrode surface, as is established in the literature for electrochemical diazonium reduction.
28

Additional grafting scans cause this feature to shift to more negative potentials, indicating that
charge transport from the substrate to the solution is impeded, as is consistent with the formation
of a molecular film.
35

Figure 4.13 Cyclic voltammograms for electropolymerization of 2 on a glassy carbon working
electrode immersed in 0.5 mM 2 in acetonitrile solution with 0.1 M TBAPF6 electrolyte. a) ν =
100 mV/s, Ps = -1.60 V b) ν = 100 mV/s, Ps = -2.60 V.

Figure 4.14 Cyclic voltammetry of a glassy carbon electrode in 0.5 mM 2 in acetonitrile with 0.1
M TBAPF6 supporting electrolyte (ν = 100 mV/s).

134

Figure 4.15 Cyclic voltammetry of a FTO electrode in 0.5 mM 2 in acetonitrile with 0.1 M
TBAPF6 supporting electrolyte (ν = 100 mV/s).

Figure 4.16 Cyclic voltammetry of a FTO electrode in 0.5 mM 2 in acetonitrile with 0.1 M
TBAPF6 supporting electrolyte (ν = 100 mV/s).

Figure 4.17 Cyclic voltammetry of a FTO electrode in 0.5 mM 2 in acetonitrile with 0.1 M
TBAPF6 supporting electrolyte (ν = 100 mV/s).

135

Following this initial reduction feature, a quasi-reversible feature at –2.01 V and an
irreversible feature at –2.24 V appear (Figure 4.13b). These features

are characteristic of
Re(bpy)(CO)3Cl reduction events, previously described as predominantly bipyridine-based.
36
The
increase in current for these features with consecutive grafting scans provides evidence for
electropolymerization, as more redox-active rhenium bipyridine species are available at the
electrode surface with continued film growth.
37
Subsequent scans reveal the absence of these
reduction features and a decrease in current density across the CV window as early as the ninth
voltammogram, which we attribute to hindered electrode kinetics as a result of the electrodeposited
film (Figure 4.14).
38
If the CV switching potential (Ps) is limited to -1.6 V, as in Figure 4.13a, the
diazonium reduction feature shifts at a slower rate and appears at -1.2 V

on the fifth scan, in
contrast to Figure 4.13b, in which the diazonium feature has shifted to -1.8 V

by the fifth
voltammogram. This behavior illustrates that the potential window selected for grafting can tune
the rate of film deposition. Similar characteristics were observed in analogous experiments with
FTO working electrodes which suggests that grafting proceeds similarly (Figures 4.15-4.17).
To prepare films on various electrode materials (FTO, TiO2, gold, glassy carbon, graphite
rod, or carbon nanotubes) for structural characterization and catalytic studies, a standardized
grafting methodology was employed. Cyclic voltammetry scans were performed between an initial
potential of -0.6 V (Pi = -0.6 V) and a switching potential of -1.6 V (Ps = -1.6 V). Cyclic
voltammetry, rather than controlled potential electrolysis or linear sweep voltammetry, was
selected to minimize the trapping of charged species in the film as the return sweep serves to repel
these species from the electrode surface between successive grafting scans.
39
Fast scan rates (ν =
1 V/s) were employed to minimize the Helmholtz layer during electropolymerization and
subsequently minimize the trapping of electrolyte in the film, facilitating uniform film growth. An
136

increasing number of scans, n, were run across a series of electrode samples to produce modified
electrodes with varying catalyst loadings. Each of these samples exhibits a distinct orange film
with increasing coloration across the series. These modified substrates were rinsed with
acetonitrile and acetone to remove potentially physisorbed diazonium species or polymerization
products. No visible change in the coloration of these films was observed after extensive rinsing
or sonication in acetone, suggesting that the coloration was due to a robustly-immobilized material.

4.3.3 Characterization of Electropolymerized Films
4.3.3.1 XPS
After grafting and rinsing, modified FTO slides were analyzed by high-resolution X-ray
photoelectron spectroscopy (Figure 4.18-4.19). The rhenium 4f region is illustrated in Figure
4.18a. The bare FTO substrate displays no peaks in the rhenium 4f region, as expected. Following
electrodeposition of 2 (n = 20), the modified substrate (solid traces) displays two features at 43.4
eV and 41.0 eV, corresponding to the Re 4f5/2 and 4f7/2, respectively. Similar values were reported
for the Re 4f signals of analogous rhenium bipyridine complexes.
13
These features are present and
unchanged following additional sonication cycles. This provides evidence that the rhenium species
is robustly immobilized to the substrate, pointing to successful reduction of the diazonium salts
and subsequent covalent attachment to the surface. These Re 4f features increase in intensity for
films deposited with a greater number of grafting scans, providing a strong initial indication for
CV-controlled film thickness (Figure 4.20). This can also be inferred from the decrease in
intensity for the high-resolution Sn 2p features. The bare FTO substrate exhibits features at 493.9
eV and 485.4 eV, corresponding to the Sn 3d1/2 and 3d5/2, respectively, which are clearly present
for a film modified with n = 1 (Figure 4.21). These features are completely passivated as the
catalyst loading is increased (up to n = 20), demonstrating that film thickness has exceeded the
137

XPS sampling depth (~5-10 nm). For films grafted with Ps = -1.60 V, the high-resolution Cl 2p
spectra show two characteristic features for the Cl 2p1/2 (198.3 eV) and 2p3/2 (196.7 eV) levels,
consistent with Re(bpy)(CO)3Cl moieties in the film (Figure 4.22). For films grafted with Ps = -
2.60 V, no chlorine was detected by XPS, consistent with the literature precedent for chloride
dissociation associated with a one-electron reduction of the bipyridine ligand with subsequent
ligand-to-metal charge transfer.
40,31
This further corroborates our assignment of the redox features
at -2.01 V and -2.24 V as the characteristic Re(bpy)(CO)3Cl reduction events. Analysis of the P
2p region reveals features at 135.8 eV and 134.9 eV, corresponding to the 2p1/2 and 2p3/2 levels,
respectively, and a feature appears at 684.4 eV in the F 1s region (Figure 4.23-4.24). These results
suggest that PF6
-
anions, introduced during grafting as electrolyte, are trapped in the film during
electropolymerization. Additionally, the high-resolution C 1s spectra exhibit a π-π* shakeup
feature at 292 eV, a characteristic feature in carbon-based materials with conjugation (Figure
4.25).
41
This provides evidence for the growth of conjugated polymers during
electropolymerization.

Figure 4.18 a) High-resolution Re 4f XPS spectrum for a modified FTO electrode (n=20). b) SEM
image of a modified FTO electrode (n = 20) at 25,000  magnification. c) AFM topology of
modified FTO electrode (n = 20).

138

Figure 4.19 XPS survey scan for modified FTO electrodes.

Figure 4.20 High-resolution XPS of the Re 4f region for modified FTO electrodes, a) n = 1, b) n
= 5, c) n = 10, d) n = 20.

139

Figure 4.21 High-resolution XPS of the Sn 2p region for modified FTO electrodes, n = 1 (left)
and n = 20 (right).

Figure 4.22 High-resolution XPS of the Cl 2p region for a modified FTO electrode (n = 20).

Figure 4.23 High-resolution XPS of the P 2p region for a modified FTO electrode (n = 20).

140

Figure 4.24 High-resolution XPS of the F 1s region for a modified FTO electrode (n = 20).

Figure 4.25 High-resolution XPS of the C 1s region for a modified FTO electrode (n = 20).

4.3.3.2 Scanning Electron Microscopy and Atomic Force Microscopy
A scanning electron microscopy (SEM) image acquired at 25,000 magnification for a
modified FTO electrode prepared with 20 grafting scans (n = 20) is presented in Figure 4.18b.
This image illustrates that film growth is uniform across the substrate. Images acquired for films
of varying catalyst loadings show that the observed morphology evolves from that of the
underlying FTO substrate to a uniform and smooth surface (Figure 4.26). Analysis of the interface
between the bare region and the modified region of the FTO electrodes highlights the applicability
of this deposition method to masking and patterned growth, as well-defined boundaries were
formed even with the simple Teflon-tape masks employed here (Figure 4.27).
141

For quantitative data regarding the morphology of the modified electrode materials, atomic
force microscopy (AFM) was applied to the analysis of FTO samples. Figure 4.18c presents the
results from an AFM tapping mode experiment on a modified FTO electrode (n = 20), which was
performed on a series of electrodes with increasing catalyst loading (Figure 4.28-4.29). As
illustrated above, these films grow as dense, uniform films. The root-mean-square surface
roughness (Sq) was measured for four modified FTO films (n = 1, 5, 10, 20), with results tabulated
in Table 4.1. These measurements reveal a decreasing trend from Sq = 11.25 (n = 1) to Sq = 9.88
(n = 20) as catalyst loading is increased. This trend demonstrates that film morphology is uniform
across the electrode surface and that smooth electrode coverage is facilitated during
electropolymerization.

142

Figure 4.26 SEM images of FTO films with varying thickness.

143

Figure 4.27 SEM images of interfacial regions for FTO films.

144

Figure 4.28 AFM data collected for FTO films with varying thickness.

145

Figure 4.29 3D projections of AFM data.

Table 4.1 RMS surface roughness of modified FTO electrodes.
n Sq
1 11.25
5 10.46
10 10.32
20 9.88

4.3.3.3 Infrared Reflection Absorption and UV-Vis Spectroscopy
Infrared reflection absorption spectroscopy (IRRAS) was employed as a surface-sensitive
method to confirm the integrity of the fac-tricarbonyl coordination environment of the rhenium
centers after electropolymerization. Gold substrates were selected for this study, as their high
reflectivity allowed for spectra with increased signal-to-noise ratio. It has been well-documented
146

that rhenium fac-tricarbonyl complexes exhibit three distinct carbonyl stretching modes in the
1900-2100 cm
-1
frequency range (though two of these modes often coalesce to one broad feature),
corresponding to one high-energy, fully-symmetric mode (a1') and two nearly-degenerate lower-
energy modes (a'' and a2'), illustrated in Figure 4.30a.
42
Due to π-backbonding between the
rhenium center and its carbonyl ligands, these stretching modes exhibit shifts in frequency related
to the electron density of the metal cation and therefore function as convenient reporter ligands.
43

The observed spectra for the electropolymerized films are consistent with fac-tricarbonyl rhenium,
with one sharp high-energy band at 2030 cm
-1
and a broad lower-energy band at 1929 cm
-1
. The
small redshift in the stretching frequency (Δν = 3 cm
-1
for the high-energy mode) indicates
increased electron density at rhenium compared to the bis(diazonium) complex 2, as is expected
for the loss of the electron-withdrawing diazonium substituents.

Figure 4.30 a) PM-IRRAS results measured for a modified Au substrate (n = 10) under both p-
(blue trace) and s- (red trace) polarization. The a1' (pink), a'' (blue), and a2' (green) are illustrated
with their corresponding carbonyl stretching vectors and identified by colored asterisks. b) UV-
Vis absorption spectrum measured for a modified FTO substrate (n = 10).

To determine whether these films display long-range order, PM-IRRAS studies were
conducted by introducing a ZnSe polarizing lens. Spectra collected for a modified gold electrode
(n = 10) are presented in Figure 4.30a. The spectrum obtained under s-polarized irradiation
147

(perpendicular to surface normal) displays increased absorption for the carbonyl stretching modes
relative to the corresponding spectrum obtained under p-polarization (parallel to surface normal).
This can be quantified as the ratio of the peak height (ΔR) for the carbonyl stretching mode at 2030
cm
-1
under s- and p-polarization (ΔRs/ΔRp). For a modified Au electrode (n = 10), this corresponds
to a value of 74. The large polarization-dependence of these stretching modes demonstrates that
the majority of the rhenium tricarbonyl moieties are oriented with respect to one another,
confirming the presence of long-range order in the deposited films. This suggests that the vertical
growth mechanism proposed is predominant. Similar results are observed for modified FTO
electrodes (Figure 4.31). Further, plotting the peak height for the carbonyl stretching modes as a
function of the number of grafting scans reveals a linear trend, as is consistent with increasing
surface concentration of Re(bpy)(CO)3Cl moieties with sequential grafting scans (Figure 4.32).
The UV-Vis spectrum of a modified FTO electrode (n = 20) is shown in Figure 4.30b.
Unlike the molecular complex 2 or the unsubstituted rhenium bipyridine complex,
Re(bpy)(CO)3Cl, the films grown in this study exhibit features in the visible region with transitions
at 398 nm, 496 nm, and 644 nm, in addition to the characteristic MLCT band at 336 nm. These
new transitions are consistent with those reported previously for conjugated rhenium bipyridine
polymers, and have been attributed to metal-ligandπ-backbone charge transfer and intraligand π-π*
transitions.
26
The red-shift of these bands relative to the MLCT transition of complex 2
demonstrate that electrons are promoted to an optically-excited state with extensive
delocalization.
43,44
As the photocatalytic behavior of the unsubstituted rhenium bipyridine
complex is hindered by its limited absorption cross section, broadening this into the visible range
is greatly beneficial for applications in artificial photosynthesis. Similar UV-Vis spectra were
recorded for a series of films with varying catalyst loadings, and all exhibit similar transitions
148

(Figure 4.33), suggesting that oligomeric species are formed even for electrodes modified with n
= 1.
45

Figure 4.31 PM-IRRAS of a modified FTO electrode (n = 20) with s- (red) and p- (black)
polarization.

Figure 4.32 a) PM-IRRAS of modified FTO electrodes with p polarization b) PM-IRRAS peak
height for the carbonyl stretching modes as a function of the electroactive surface coverage.

149

Figure 4.33 UV-Vis spectra of FTO films with varying thickness.

4.3.3.4 Electrochemistry of Films
After grafting, modified electrodes were washed vigorously with acetone for 1 minute,
dried under nitrogen, and immersed in an acetonitrile solution for electrochemical studies. Cyclic
voltammetry experiments for a series of modified GC electrodes (n = 1, 5, 10, and 20) in
acetonitrile with 0.1 M TBAPF6 supporting electrolyte are presented in Figure 4.34a. An
irreversible cathodic peak appears at -1.62 V (for n = 20), which has previously been attributed to
the buildup of excess electrolyte in the polymer matrix and subsequent discharge of these
species.
37,46,47
As catalyst loading is increased, this feature shifts to more cathodic potentials and
grows, indicting hindered charge transport and increased electrolyte buildup for thicker films. The
two rhenium bipyridine-based cathodic waves of 2 have converged into a single broad feature at -
1.95 V, now corresponding to a 2-electron reduction of the deposited films with retained quasi-
reversibility. The broadening of this feature illustrates the hindered electron transfer kinetics
resulting from growth of the polymer film.
48
For thin films (n = 5), there is a 13.2 mV peak
separation between the cathodic and anodic peak potentials, as expected for a non-diffusing,
surface-immobilized species (Figure 4.35).
48
As catalyst loading increases (n = 10), the peak
separation increases as well (ΔE = 63.3 mV), which suggests that the diffusion-like transport of
electrons is more substantial for longer polymeric chains. Randles-Sevcik analyses were
150

performed on the cyclic voltammetry data experiments conducted with variable scan rates, and in
all cases, the slopes of the logarithmic plots are near to the ideal value of 1 for an immobilized
species, with the exception of the thinnest film (n = 1), which is dominated by passive charging
current (Figures 4.35-4.38).

Figure 4.34 a) Cyclic voltammetry of electrodes modified with varying numbers of grafting scans
in acetonitrile solutions with 0.1 M TBAPF6 electrolyte (no diazonium salt in solution) (ν = 100
mV/s). b) Catalyst loading as a function of the number of grafting scans applied (Pi = -0.6 V, Ps =
-1.6 V, ν = 1 V/s) as determined by cyclic voltammetry (red) and ICP-OES (blue). c) Double-layer
capacitance (Cdl) measurement for a modified electrode (n = 5), plots indicate CV experiments at
the open-circuit potential (0.120 V) with varying scan rates (ν = 1, 5, 10, and 20 mV/s). d) Cdl of
deposited films as a function of the number of grafting scans, illustrating the linear increase in
electrochemically-active surface coverage.

151

Figure 4.35 Cyclic voltammetry of a modified glassy carbon electrode (n = 5) in acetonitrile with
0.1 M TBAPF6 supporting electrolyte.

Figure 4.36 Cyclic voltammetry of a modified glassy carbon electrode (n = 1) in acetonitrile with
0.1 M TBAPF6 supporting electrolyte.

152

Figure 4.37 Cyclic voltammetry of a modified glassy carbon electrode (n = 10) in acetonitrile with
0.1 M TBAPF6 supporting electrolyte.

Figure 4.38 Cyclic voltammetry of a modified glassy carbon electrode (n = 20) in acetonitrile with
0.1 M TBAPF6 supporting electrolyte.

After the initial cathodic scan, subsequent scans exhibit lower current densities due to
inefficient reoxidation of the reduced film. If an anodic scan (up to Ps = +0.4 V) is performed
between cathodic scanning, the cathodic current density remains stable for repeated experiments
(Figure 4.39). Similarly, if the modified electrode is allowed to oxidize in air overnight (12 hours),
the redox feature at -1.95 V returns to the current density measured for the first scan after grafting.
This behavior coincides with a dramatic color change in these films. After cathodic sweeping (P s
= -2.6 V), a color change from orange to blue is observed for deposited films (Figure 4.40). After
resting in air for 5 minutes, the blue film returns to its original orange color. This color change is
153

not observed if the cathodic scan is followed by an anodic sweep (Ps = +0.4 V). This behavior is
consistent with the reported behavior for poly([2,2'-bipyridine]-5,5'-diyl), which undergoes
electrochemical n-doping.
24,49
As such, this behavior points to the successful generation of
conjugated polymers with the desired poly(Re(CO)3Cl[2,2'-bipyridine]-5,5'-diyl) structure.

Figure 4.39 Three sequential cyclic voltammetry scans of a modified FTO electrode (n = 5) from
-0.5 V to -2.25 V. The fourth scan begins with an anodic sweep (Pi = -0.6 V, Ps = +0.6 V, Pf = -
2.25 V). All scans in acetonitrile with 0.1 M TBAPF6 (ν = 100 mV/s).

154

Figure 4.40 a) Modified glassy carbon stick electrodes (n = 1, 5, 10, and 20) prepared with varying
catalyst loadings. b) Unmodified glassy carbon electrode (n = 0) and modified glassy carbon
electrodes (n = 5) immediately after grafting with CV scans to Ps = -2.60 V (blue coloration) and
an identically-modified electrode after 5 minutes of air exposure (orange coloration). c) A
modified Au electrode (n = 10) with a portion of the electrode not modified for visual comparison
(n = 0).

Table 4.2 Catalyst loading for FTO electrodes as determined by cyclic voltammetry (CV) and
ICP-OES.

CV Loading ICP Bulk Loading
n C nmol ppm nmol
1 0.00009 0.7 0.018 1.0
5 0.00037 2.7 0.026 1.4
10 0.00036 3.7 0.036 2.0
20 0.00065 5.4 0.040 2.2

155

Figure 4.41 Double-layer charging current density (ΔJ = Janodic - Jcathodic) at the open-circuit
potential for a modified glassy carbon electrode (n = 1) as a function of scan rate.

Figure 4.42 Double-layer charging current density (ΔJ = Janodic - Jcathodic) at the open-circuit
potential for a modified glassy carbon electrode (n = 5) as a function of scan rate.

Figure 4.43 Double-layer charging current density (ΔJ = Janodic - Jcathodic) at the open-circuit
potential for a modified glassy carbon electrode (n = 10) as a function of scan rate.

156

Figure 4.44 Double-layer charging current density (ΔJ = Janodic - Jcathodic) at the open-circuit
potential for a modified glassy carbon electrode (n = 20) as a function of scan rate.

To assess the degree of control over catalyst loading provided by electrochemical
deposition, we performed a series of studies to determine the electrochemically-active surface
coverage of the rhenium bipyridine species. After washing the working electrode, CV scans in
clean electrolyte solution (no diazonium salt in solution) were used to quantify the electroactive
rhenium bipyridine concentration. The current-time plots were integrated for the range of the 2-
electron reduction feature at -1.95 V (as a convenient electrochemical signature of electroactive
rhenium sites). The charge passed in this region was then converted to [Re] using Equation 4.1,
though this provides only an estimate of coverage due to the peak broadness of these reduction
features and possible side phenomena.
50
To corroborate these estimates, bulk catalyst loadings
were determined by ICP-OES measurements for films digested in nitric acid. The results of these
quantification studies are plotted in Figure 4.34b (details tabulated in Table 4.2). The linear
correlation between electroactive rhenium coverage and the number of grafting scans performed
demonstrates that nanomolar control over the surface coverage is feasible through this
electropolymerization technique. Further, the double layer capacitance, Cdl, of each film was
measured to provide a proxy for electroactive surface coverage (Figure 4.34c).
51
Cyclic
voltammetry scans were performed with variable scan rates in a 100 mV window around the open-
157

circuit potential for each electrode (Figures 4.41-4.44). A linear correlation was observed between
the Cdl value and the number of grafting scans applied, shown in Figure 4.34d, which agrees with
the linear trend observed from our estimation of surface coverage by Equation 4.1. As a surface
coverage of ~0.1 nM/cm
2
corresponds to a dense monolayer for analogous metal complexes, the
catalyst loadings measured in this study suggest that multilayer film growth is achieved during the
initial grafting scan (~10 equivalent layers for n = 1).
47

4.3.4 Catalytic Studies
4.3.4.1 Cyclic Voltammetry
Voltammograms with modified glassy carbon electrodes under a saturated CO2 atmosphere
(shown in Figure 4.45a) exhibit an increase in current and loss of reversibility relative to
experiments under nitrogen atmosphere. While CVs under CO2 show increasing current densities
with increasing surface coverage of catalyst, experiments performed after the addition of a proton
source (0.5 M trifluoroethanol, Figure 4.45b) result in identical behavior for an electrode prepared
with n = 10 and one prepared with n = 20, verifying that an optimal film thickness is reached before
n = 20. This saturation in current density for films with higher catalyst loading in the presence of
an external proton source indicates hindered mass transport as a result of the increased film
thickness.

158

Figure 4.45 a) Cyclic voltammograms of modified electrodes (n = 1, 5, 10, and 20) in acetonitrile
solution with 0.1 M TBAPF6 electrolyte under inert atmosphere (dashed) and 1 atmosphere of CO2
(solid). b) Cyclic voltammograms of modified electrodes (n = 1, 5, 10, and 20) in acetonitrile
solution with 0.1 M TBAPF6 electrolyte under 1 atmosphere of CO2 (dashed) and with 0.5 M
trifluoroethanol added (solid).

Figure 4.46 Foot-of-the-wave analysis for a) modified graphite rod electrode (n = 10) and b)
modified Nafion-MWCNT electrode (n = 10).

As glassy carbon and FTO represent flat substrates, surface coverage of the catalytic film
is limited in these cases. In an effort to improve the electrochemically-active surface coverage of
this material, alternative carbon-based electrodes were explored for CO2 electrolysis studies. We
chose to explore graphite rod substrates and a composite electrode material composed of multi-
walled carbon nanotubes (MWCNT) with a proton-conducting binder (Nafion® 117) on FTO as
high-surface-area electrodes. For the MWCNT electrodes, a MWCNT-Nafion ink was prepared
and drop-cast onto FTO substrates for subsequent electropolymerization of 2. The results of cyclic
159

voltammetry experiments conducted on modified electrodes are presented in Figure 4.47. Upon
switching from a nitrogen atmosphere to a CO2-saturated atmosphere, both electrodes show a large
increase in current density with a corresponding loss in reversibility and a change in waveform to
that of the kinetic zone, indicating fast catalysis with minimal substrate depletion near the electrode
surface. As plateau currents were not reached for these systems, catalytic rate constants (kobs) were
determined by foot-of-the-wave analysis (Figure 4.46, details described in experimental methods).
For a modified graphite rod electrode (n = 10), kobs was calculated to be 29.8 s
-1
. For the modified
MWCNT-Nafion electrode, a rate constant of 19.7 s
-1
was calculated. Due to the increased passive
charging currents measured and slower kobs calculated for the Nafion-MWCNT electrode, graphite
rod electrodes were selected as conductive substrates for controlled potential electrolysis studies.

Figure 4.47 Cyclic voltammograms in acetonitrile solutions with 0.1 M TBAPF6 supporting
electrolyte under either N2 (blue) or CO2 (red) atmosphere with a) modified graphite rod electrode
(n = 10) and b) modified Nafion-MWCNT (n = 10).

4.3.4.2 Controlled Potential Electrolysis
Controlled potential electrolysis (CPE) experiments were performed with modified
graphite rod electrodes to determine the current efficiencies, product distributions, and lifetimes
of these catalysts. Electrodes were prepared with various catalyst loadings (Figure 4.48) and
assessed under electrolytic conditions for two hours at -2.25 V under CO2 atmosphere. This
160

potential was selected for CPE studies, as it corresponds to the potential at which the modified
electrode reaches current densities of 10 mA/cm
2
under saturated CO2 atmosphere by cyclic
voltammetry. The results of these studies are summarized below in Figure 4.49, with details
tabulated in Table 4.3. In all cases, CO was the only product determined by GC and no formic
acid was detected by
13
C NMR, illustrating high selectivity for CO production. For the lowest
catalyst coverage studied (n = 1), 3.3 μmol CO was produced during a 2 hour electrolysis
experiment with FE = 48±5%. Increasing the catalyst loading (n = 5) led to an improvement in
activity, producing 18.8 μmol CO with a FE= 99±7%. This corresponds to a TON of 3583, based
on the bulk catalyst loading by ICP-OES, and an overall TOF of 0.50 s
-1
. For n = 10, a modified
electrode produced 27.2 μmol CO with FE = 88±8%, corresponding to a TON of 3606 and TOF
of 0.50 s
-1
. For the highest coverage studied (n = 20), 13.8 μmol CO was produced with a FE =
64±6% (TON = 623, TOF = 0.06 s
-1
).

Figure 4.48 Catalyst loading for modified graphite rod electrodes (n = 1, 5, 10, and 20) as
determined by cyclic voltammetry (red) and ICP (blue).

161

Figure 4.49 Results of controlled potential electrolysis experiments with modified graphite rod
electrodes (n = 1, 5, 10, and 20). All experiments were performed at -2.25 V for 2 hours in
acetonitrile solutions with 0.1 M TBAPF6 supporting electrolyte. Left axis corresponds to the
Faradaic efficiency (red) and the right axis corresponds to TON based on ICP-OES bulk loading
(blue).

Table 4.3 Summary of controlled potential electrolysis studies with TON and TOF calculated
based on cyclic voltammetry (CV) estimated loading and ICP estimates. All experiments
performed in acetonitrile with 0.1 M TBAPF6 supporting electrolyte for 2 hours at -2.25 V.
Modified graphite rod electrodes served as the working electrodes, with a Ag wire reference and
graphite rod counter electrode.

n

CV

ICP

C CO
(μmol)
FE
(%)
Loading
(nmol)
TON

TOF
(s
-1
)
Loading
(nmol)
TON

TOF
(s
-1
)
1 1.327 3.32 48±5 4.1 806±80 0.112 2.9 1149±110 0.319
5 3.638 18.82 99±7 11.7 1606±160 0.223 5.3 3583±360 0.498
10 5.987 27.20 88±8 18.0 1508±150 0.210 7.5 3606±360 0.501
20 4.174 13.81 64±6 34.7 398±40 0.055 22.2 623±60 0.086

Based on these results, it is apparent that both FE and TON increase up to an optimal value
for catalyst loading, beyond which both parameters decrease. The low FE and TON for low catalyst
loadings (n = 1) are consistent with the behavior reported for a chemically-grafted diazonium-
substituted rhenium bipyridine species on graphene substrates.
52
Chemical grafting led to much
lower catalyst coverages (0.1 nmol/cm
2
) and lower FE for CO production (≤ 44%, comparable to
the n = 1 electrode). As catalyst loading in modified graphite rod electrodes is increased, side
162

processes are eliminated by passivating the electrode surface, and the FE for CO production
increase. Based on the trend observed in Figure 4.49, the optimal catalyst loading is reached before
n = 20 and appears to lie between n = 5 and n = 10. Using an average of the loadings measured
through ICP-OES for n = 5 and n = 10, an optimal coverage of ~6 nmol is estimated. As previously
discussed, increased catalyst loading corresponds to both reduced electron transfer and mass
transport kinetics, which are expected to hinder catalytic activity for thick films. These results
demonstrate that high catalytic performance can be reached with relatively low concentrations of
catalyst.

4.3.4.3 Photocatalysis
The photocatalytic activities of modified TiO2 substrates were measured to assess the
applicability of this material towards device fabrication for artificial photosynthesis. As catalyst
immobilization has previously been shown to improve excited state lifetimes, the selection of
supporting substrate, anchoring method, and device architecture has a profound impact on the
photocatalytic activity.
29,53,54
The electrode architecture explored in this study is depicted below in
Figure 8. Mesoporous TiO2 (m-TiO2) was selected as a wide-bandgap semiconducting support
with a rough morphology for increased surface area. The semiconducting nature of TiO2 enables
light absorption and subsequent exciton generation, albeit with poor spectral overlap with the solar
spectrum (Eg = 3.2 eV), while the roughness of these substrates allows for increased catalyst
coverage relative to the flat FTO and glassy carbon electrodes. As the flatband potential of TiO 2
in acetonitrile solution was reported to be about -2.4 V, a strong driving force exists for electron
transfer from the TiO2 conduction band to the conjugated film with a catalytic onset potential of -
1.9 V.
55
This was expected to prevent back-transfer of electrons from the rhenium sites to the TiO2
substrate, though midgap states have been shown to facilitate electron transfer processes for
163

surface-immobilized redox species.
56
Triethanolamine (TEOA) was selected as a sacrificial
electron donor and all photocatalytic experiments were performed under sacrificial conditions in
DMF (5:1 DMF:TEOA) to study the reductive half-reaction.
57,58
Electrodes were prepared with
various catalyst loadings, which was estimated both by cyclic voltammetry and ICP-OES (Figure
4.50).

Figure 4.50 Catalyst loading for modified TiO2 electrodes (n = 1, 5, 10, and 20) as determined by
cyclic voltammetry (red) and ICP (blue).

Scheme 4.2 Device architecture and energy diagram for photocatalytic CO2 reduction.

As TiO2 has been shown to perform unassisted CO2RR, albeit with typically low selectivity
for CO production, the activity of the bare substrates under photocatalytic conditions was
164

measured to benchmark this background activity.
55
The bare TiO2 substrate generate 0.11 μmol
CO during 5 hours of irradiation. By introducing a 399 nm cut-on filter between the light source
and substrate, excitation of TiO2 was inhibited and no CO was detected after 5 hours of irradiation.
Following these control studies, the photocatalytic activities of modified devices were tested with
various catalyst loadings. The results of these studies are summarized in Figure 4.51, with details
tabulated in Table 4.4. For the lowest coverage studied (n = 1) a TON of 70 and an overall TOF
of 14.0 hr
-1
was measured after 5 hours of irradiation. For n = 5, a TON of 28 and an overall TOF
of 5.6 hr
-1
was determined. After introducing a 399 nm cut-on filter, a similar activity was
measured for a film modified under the same conditions (n = 5, TON = 31, TOF = 6.1 hr
-1
). This
confirms that the CO generated for modified electrodes is predominantly generated by the catalyst
moieties rather than the mesoporous TiO2 electrode. This is consistent with dense surface
modification, as seen by the disappearance of the Sn 3d XPS signal on FTO electrodes (Figure
4.21). For n = 10, the photocatalytic activity decreases to a TON of 26 and TOF of 5.2 hr
-1
. The
highest coverage studied (n = 20) displayed the lowest catalytic activity, with a TON of 22 and
TOF of 4.4 hr
-1
. After 5 hours of irradiation, the solvent mixture had changed from colorless to
pale orange and the color intensity of the post-catalysis devices had diminished.

165

Figure 4.51 TON as a function of catalyst loading for photocatalytic studies with modified TiO 2
electrodes.

Table 4.4 Summary of photocatalytic studies with TON and TOF calculated based on cyclic
voltammetry (CV) estimated loading and ICP estimates. All studies performed in 10 mL 5:1
DMF:TEOA under illumination for 5 hours.
*
399 nm cut-on filter was introduced for this
measurement.

n

CV

ICP

t (s) CO
(μmol)
Loading
(nmol)
TON

TOF
(hr
-1
)
Loading
(nmol)
TON

TOF
(hr
-1
)
1 18000 0.41 4.0 103±10 20.6 5.8 70±7 14.0
5 18000 0.31 7.9 39±4 7.8 10.9 28±3 5.6
5
*
18000 0.40 8.3 48±5 9.6 13.0 31±3 6.1
10 18000 0.39 15.8 25±2 4.9 15.1 26±3 5.2
20 18000 0.40 31.6 13±1 2.5 18.4 22±2 4.4

These results suggest promising unbiased photocatalytic activity for thin films of
poly(Re(CO)3Cl[2,2'-bipyridine]-5,5'-diyl). Prior systems incorporating rhenium bipyridine
complexes immobilized on TiO2 were reported to exhibit photocatalytic TON up to 62 during 24
hours for a phosphonated complex, and a photoelectrocatalytic TON of 70 for a carboxylate-
substituted complex on Cu2O-TiO2 photocathodes (during 1.5 hours at -2.05 V).
59,14
The measured
TON of 70 for the n = 1 electrode is comparable to these previous studies. The high activity for
low catalyst loadings demonstrate that only a small quantity of the deposited material remains
active through the 5 hour irradiation experiment. Beyond this loading, photocatalytic activity drops
166

to TON ≤ 28. The homogeneous, unsubstituted rhenium bipyridine species was reported to
generate CO with a TON of 27 during 4 hours of irradiation in a mixture of 5:1 DMF:TEOA.
Furthermore, the color change of both the electrodes and DMF solution reveal that catalyst
desorption occurs for TiO2 substrates and that these polymers are partially soluble in DMF
following surface desorption. As rhenium bipyridine complexes are known to exhibit absorption
features with large molar extinction coefficients, it is likely that the activity of films with higher
catalyst loadings are dominated by the desorbed material, which absorbs incident photons before
they are able to reach the electrode-catalyst assembly. For photons that do reach the device, a thin
layer of photocatalytic moieties at the surface are likely excited and prevent deeper-lying
chromophores from accessing
3
MLCT excited states. As reductive quenching is also dependent on
the diffusion kinetics of TEOA through the film, film thickness may inhibit the sacrificial reductant
from quenching interior photoexcited states.

4.3.5 Post-Catalysis Electrode Characterization
Modified electrodes were analyzed after catalytic studies to better understand the lifecycle
and possible deactivation pathways of the catalyst. In all cases, the cyclic voltammograms of
modified electrodes after sustained bulk electrolysis or irradiation studies (>2 hours) exhibit
substantially decreased current densities relative to the as-deposited films (>90% decrease). XPS
on the post-catalysis FTO electrodes indicates a decrease in intensity for the high-resolution
rhenium 4f peaks, implying a loss of material from the electrode surface (Figure 4.52). No features
corresponding to Re
0
or oxidized rhenium species were detected, confirming that the formation of
nanoparticles did not occur during catalytic studies. A similar decrease in intensity was observed
for the carbonyl stretching modes of the post-catalysis films in IRRAS studies (Figure 4.53).
These peaks were also 10 cm
-1
red-shifted relative to the pre-catalysis film, suggesting a more
167

nucleophilic rhenium center, which is consistent with the reduced rhenium bipyridine molecular
complex. The appearance of two new features at 1604 and 1660 cm
-1
is observed, and have
previously been attributed to the presence of Re-CO2
-
species—an intermediate in the proposed
catalytic cycle for the homogeneous complex.
60,61
UV-Vis studies on the post-catalysis FTO
electrodes show a similar decrease in intensity for the electronic transitions of the film (Figure
4.54). No shifts in these features were observed, nor were any new features present after catalytic
studies. This provides a strong indication that dimers are not generated under catalytic conditions,
as the characteristic electronic transitions of the dimer species remain absent.

Figure 4.52 High-resolution XPS of the Re 4f region for a modified FTO substrate (n = 20) before
(blue) and after (red) a 1 hour controlled potential electrolysis experiment at -2.25 V in acetonitrile
solution with 0.1 M TBAPF6 supporting electrolyte under saturated CO2 atmosphere.

168

Figure 4.53 IRRAS studies of a modified FTO substrate (n = 20) before (black) and after (red) a
1 hour controlled potential electrolysis experiment at -2.25 V in acetonitrile solution with 0.1 M
TBAPF6 supporting electrolyte under saturated CO2 atmosphere.

Figure 4.54 UV-Vis studies of a modified FTO substrate (n = 20) before (blue) and after (red) a 1
hour controlled potential electrolysis experiment at -2.25 V in acetonitrile solution with 0.1 M
TBAPF6 supporting electrolyte under saturated CO2 atmosphere.

4.4 CONCLUSIONS
We report here the directed electropolymerization of molecular films for photocatalytic
and electrocatalytic CO2 reduction. A diamine-functionalized rhenium bipyridine system was
converted to a bis(diazonium) species, complex 2, which was immobilized to a broad range of
substrates. The catalyst loading was controlled by cyclic voltammetry, and the resulting films
exhibit a high degree of long-range order, supported by polarized IRRAS studies. The electronic
absorption spectra reveal new transitions in the visible range upon polymerization of 2, consistent
169

with the generation of conjugated polymers with intraligand π-π
*
and metal-πpolymer transitions.
Cyclic voltammetry of these films shows a large, irreversible increase in current under CO2
atmosphere at a potential that coincides with the rhenium bipyridine reduction features. CPE
studies performed on a modified graphite rod electrode confirm the catalytic reduction of CO2 to
CO with turnover numbers up to 3606 in 2 hours, TOF of 0.50 s
-1
, and FE of 99% for electrolysis
at -2.25 V vs Fc/Fc
+
. Photocatalytic studies on modified TiO2 electrodes confirm promising
activity with TONs up to 70 after 5 hours of irradiation, corresponding to an overall TOF of 14 hr
-
1
. This work demonstrates that (1) rigid, conjugated, and highly-oriented polymers can be produced
on a broad range of substrates by the electropolymerization of aryl p-bis(diazonium) salts, and that
(2) conjugated polymers of rhenium bipyridine moieties produced in this manner exhibit promising
electrocatalytic and photocatalytic performance.

4.5 EXPERIMENTAL METHODS
4.5.1 Materials and Synthesis
All manipulations of air- and moisture-sensitive materials were conducted under nitrogen
atmosphere in a Vacuum Atmospheres glovebox or on a dual manifold Schlenk line with oven-
dried glassware. Water was deionized with the Millipore Synergy system (18.2 MΩ·cm). All other
solvents used were degassed with nitrogen, passed through activated alumina columns, and stored
over 4Å Linde-type molecular sieves. Proton NMR spectra were acquired using a Varian 400-MR
2-Channel spectrometer at room temperature and referenced to the residual
1
H resonances of the
deuterated solvent (
1
H: CD3CN, δ 1.95 ppm). The [2,2'-bipyridine]-5,5'-diamine ligand was
synthesized according to reported literature procedures.
1
Complex 1 was synthesized using similar
170

methods with the ones reported for analogous complexes.
2
All other chemical reagents were
purchased from commercial vendors and used without further purification.

4.5.2 Synthesis of 2
Complex 1 (45 mg) was suspended in acetonitrile (1.9 mL) and briefly sonicated for 5
minutes. Separately, a solution of nitrosonium tetrafluoroborate (24 mg) was dissolved in a
minimal amount of acetonitrile (0.9 mL). Both solutions were cooled to -40 °C. Once cooled, the
suspension of 1 was added dropwise to the NOBF4 solution, leading to an immediate color change
from pale-yellow to dark blue. Addition of diethyl ether (6 mL) resulted in the formation of a dark
blue precipitate, which was collected by filtration, and stored in the dark at -27 °C (
1
H in CD3CN:
δ 10.10, 9.29 and 9.02 ppm).

4.5.3 Electrochemistry
Electrochemistry experiments were carried out in acetonitrile solution with 0.1 M TBAPF6
electrolyte using a Pine potentiostat. The pseudo-reference electrode used was a Ag wire purchased
from VWR. The platinum wire used as a counter electrode was purchased from Alfa Aesar. Ohmic
drop was compensated using the positive feedback compensation implemented in the instrument.
All experiments in this paper were referenced relative to ferrocene (Fc) with the Fe3
+/2+
couple at
0.0 V. Electrochemical experiments were carried out in a three electrode configuration
electrochemical cell under a nitrogen or CO2 atmosphere using glassy carbon, graphite rod, carbon
nanotubes, FTO, TiO2, or gold as the working electrode. The reference and counter electrodes were
isolated in glass capillaries with Vycor frits.

171

CPE measurements were conducted in a two-chambered H cell. The first chamber held the
working and reference electrodes in 40 mL of 0.1 M tetrabutylammonium hexafluorophosphate
(TBAPF6) in acetonitrile. The second chamber held the auxiliary electrode in 20 mL of 0.1 M
TBAPF6 in acetonitrile. The two chambers were separated by a fine porosity glass frit and the
reference electrode was placed in a separate compartment connected by a Vycor tip. Graphite rod
electrodes (Tokai Carbon USA) were used as the working and auxiliary electrodes. Using a
gastight syringe, 2 mL of gas were withdrawn from the headspace of the H cell and injected into a
gas chromatography instrument (Shimadzu GC-2010-Plus) equipped with a BID detector and a
Restek ShinCarbon ST Micropacked column. Faradaic efficiencies were determined by dividing
the measured CO produced by the amount of CO expected based on the charge passed during the
bulk electrolysis experiment. For each species the controlled-potential electrolysis measurements
were performed at least twice, leading to similar behavior. The reported Faradaic efficiencies and
µmol of CO produced are average values.

4.5.4 Electrochemical Analysis
The electrochemically-active catalyst loading was estimated by cyclic voltammetry. A
cathodic scan sweeping from Pi = -0.6 V to Pi = -2.25 V was performed for the modified electrode
under a nitrogen atmosphere. The resulting current-time plot was integrated for the film redox
feature at -1.95 V, which was used to determine and estimated catalyst loading based on Equation
4.1 below. Q represents the total charge passed at the electrode for the cathodic wave (C), F
represents the Faraday constant (96,485 C mol
-1
), n represents the number of electrons for the
reduction event (2), and A represents the area of the electrode (cm
2
).

172

[𝑅𝑒 ] =
𝑄 𝐹𝑛𝐴 Eq. 4.1

The catalytic rate constant (kobs) for modified graphite rod and carbon nanotube electrodes
was approximated by cyclic voltammetry. Plateau currents are not reached for these materials as a
result of electrochemical side-processes, so a foot-of-the-wave approach was employed. The
current behavior at the foot of the catalytic wave was analyzed and kobs was extracted under the
assumption that side processes are absent in this region (𝐸 <<𝐸 cat
0
). The electrochemical behavior
of redox-active films has been modelled by a series of current-potential relationships, with
pertinent relations shown below in Equations 4.2–4.6.
63-65
The catalytic current at the foot-of-the-
wave (i) for an EECC or ECCE process (E = electrochemical, C = chemical step) can be modelled
by Equation 4.2. In Equation 4.2, F is Faraday’s constant (F = 96,485 C mol
-1
), 𝐶 𝑝 0
is the catalyst
concentration in the film, De is the effective diffusion constant of the electron, k1 is the rate constant
of the catalytic reaction, κA is the partition coefficient for the substrate (CO2) in the film, and 𝐶 𝐴 0

is the bulk concentration of substrate (CO2). Equation 4.4 describes the peak current response in
the absence of substrate (i
p
0
) . The effects of 𝐶 𝑝 0
and De are negated by substituting Equation 4.3
into Equation 4.2, and dividing by Equation 4.4. This generates the simplified Equation 4.5.
Plotting
𝑖 i
p
0
as a function of 1 + exp[
F
RT
(E − E 1
2
0
)] therefore provides a linear correlation with
the slope given by Equation 4.6, where kobs is equal to k
1
𝜅 𝐴 𝐶 𝐴 0
and represents the observed rate
constant in the hypothetical absence of side processes.

i =
i
c
1 +exp[
F
RT
(E −E
1
2
0
)]
Eq. 4.2
173

i
c
= 2F C
p
0
√D
𝑒 √k
1
𝜅 𝐴 𝐶 𝐴 0
Eq. 4.3
i
p
0
= 0.446F C
p
0
√
Fv
𝑅𝑇
𝐷 𝑒 Eq. 4.4
i
i
p
0
=
4.48√
RT
Fv
k
1
𝜅 𝐴 𝐶 𝐴 0
1 +exp[
F
RT
(E −E
1
2
0
)]
Eq. 4.5
𝑆𝑙𝑜𝑝𝑒 = 4.48√
𝑅𝑇
𝐹 v
𝑘 𝑜𝑏𝑠 Eq. 4.6

4.5.5 Gas Chromatography
Gaseous products were quantified using a Shimadzu GC-2010-Plus instrument equipped
with a BID detector and a Restek ShinCarbon ST Micropacked column. In a typical experiment, 2
mL of gas were withdrawn from the headspace of the electrochemical cell with a gas-tight syringe
and injected into the instrument. Calibration plots were prepared with multiple injections of syngas
standards purchased from Praxair, Inc. Faradaic efficiencies were determined by dividing the
amount of gaseous product produced as measured by gas chromatography by the amount of gas
expected based on the total charge measured during controlled potential electrolysis. Multiple runs
were performed for each condition studied, leading to similar behavior. The reported μmol of gas
produced (and subsequently FE and TON) are averaged values and error bars are determined from
multiple experiments.

4.5.6 Product Detection
Detection of formate was performed according to literature precedent.
2
Following CPE studies, a
5 mL aliquot of each electrolysis solution was collected and extracted with 2 mL D2O. The aqueous
174

portion was acidified with one drop HCl and tested for formate by
1
H NMR spectroscopy. Formate
was not detected in any of the experiments performed. Detection of methanol, aldehydes, and other
alcohols was carried out through NMR spectroscopy studies, and detection of methane, carbon
monoxide, and hydrogen was carried out through gas chromatography.

4.5.7 UV-Vis Spectroscopy
UV-Vis spectra were collected using a UV-1800 Shimadzu UV spectrophotometer. FTO
samples were studied in transmittance mode and the spectrum measured for an unmodified FTO
substrate was subtracted as background.

4.5.8 SEM
Scanning electron microscopy (SEM) was performed on a JEOL JSM 7001F scanning
electron microscope using an accelerating voltage of 15 kV.

4.5.9 AFM
Atomic force microscopy (AFM) topography images were collected in tapping mode using
an Agilent 5420 SPM instrument S3. The probe tips were Tap300-G Silicon AFM probes (resonant
frequency 300 kHz, force constant 40 N/m) purchased from Budgetsensors.com and aligned prior
to use. Images were collected with a scan rate of 0.25 lines per second and over an area of 20 µm
2
.
All samples were imaged under one atmosphere of air at room temperature.

175

4.5.10 ICP-OES
Inductively coupled plasma optical emission spectroscopy (ICP-OES) measurements were
performed using a Thermo Scientific iCAP 7000 ICP-OES.

4.5.11 Photocatalytic Studies
Mesoporous TiO2 (m-TiO2) electrodes were prepared by spin-coating a suspension of 20
mg TiO2 nanoparticles (anatase, ~20 nm) in 10 mL ethanol onto fluorine-doped tin oxide (FTO)
electrodes, which were then annealed at 450 °C for 30 minutes. All photocatalysis experiments
were conducted using a ThorLabs HPLS-30-03 solid state light source with a wavelength range of
350 to 700 nm. For studies conducted with a filter, a 399 nm cutoff filter (purchased from Schott
Glass) was introduced. Using a gastight syringe, 2 mL of gas were withdrawn from the headspace
of the photocatalysis cell and injected into a gas chromatography instrument (Shimadzu GC-2010-
Plus) equipped with a BID detector and a Restek ShinCarbon ST Micropacked column. TONs
were determined by dividing the measured CO produced by the catalyst loading determined by
ICP-OES or CV analysis. All studies were conducted in 5:1 mixtures of DMF:TEOA (10 mL).

4.5.12 Computational Methods
All calculations were run using the Q-CHEM program package.
66
Geometry optimizations
were run with unrestricted DFT calculations at the M06 level of theory with a composite basis set.
8

The Pople 6-31G* basis set was used for H, C, N, and O atoms and the Hay–Wadt VDZ (n+1)
effective core potentials and basis sets (LANL2DZ) were used for Cl and Re atoms.
67-71
All
optimized geometries were verified as stable minima with frequency calculations at the same level
of theory. The M06 functional was used throughout this study, as it provides reduced Hartree-Fock
176

exchange contributions and includes empirical fitting for accuracy in organometallic systems.
Single point energy calculations were run with a larger 6-311G** basis for H, C, N, and O atoms.
Kohn-Sham orbital images are presented with isovalues of 0.05 for clarity.

177

4.6 REFERENCES
1. Energy International Administration (EIA). International Energy Outlook Executive
Summary. 2018, 1–14.
2. Gray, H. B. Powering the Planet with Solar Fuel. Nat. Chem. 2009, 1, 7.
3. Lewis, N. S. Research Opportunities to Advance Solar Energy Utilization. Science 2016,
351 (6271), aad1920.
4. House, R. L.; Iha, N. Y. M.; Coppo, R. L.; Alibabaei, L.; Sherman, B. D.; Kang, P.;
Brennaman, M. K.; Hoertz, P. G.; Meyer, T. J. Artificial Photosynthesis: Where Are We
Now? Where Can We Go? J. Photochem. Photobiol. C Photochem. Rev. 2015, 25, 32–45.
5. Kim, D.; Sakimoto, K. K.; Hong, D.; Yang, P. Artificial Photosynthesis for Sustainable
Fuel and Chemical Production. Angew. Chemie Int. Ed. 2015, 54 (11), 3259–3266.
6. Qiao, J.; Liu, Y.; Hong, F.; Zhang, J. A Review of Catalysts for the Electroreduction of
Carbon Dioxide to Produce Low-Carbon Fuels. Chem. Soc. Rev. 2014, (43), 631-675.
7. Benson, E. E.; Kubiak, C. P.; Sathrum, A. J.; Smieja, J. M. Electrocatalytic and
Homogeneous Approaches to Conversion of CO2 to Liquid Fuels. Chem. Soc. Rev. 2009,
38 (1), 89–99.
8. Morris, A. J.; Meyer, G. J.; Fujita, E. Molecular Approaches to the Photocatalytic
Reduction of Carbon Dioxide for Solar Fuels. Acc. Chem. Res. 2009, 42 (12), 1983–1994.
9. Kortlever, R.; Shen, J.; Schouten, K. J. P.; Calle-Vallejo, F.; Koper, M. T. M. Catalysts and
Reaction Pathways for the Electrochemical Reduction of Carbon Dioxide. J. Phys. Chem.
Lett. 2015, 6 (20), 4073–4082.
10. Bullock, R. M.; Das, A. K.; Appel, A. M. Surface Immobilization of Molecular
Electrocatalysts for Energy Conversion. Chem. - A Eur. J. 2017, 23 (32), 7626–7641.
11. Hawecker, J.; Lehn, J.-M.; Ziessel, R. Electrocatalytic Reduction of Carbon Dioxide
Mediated by Re(Bipy)(CO)3Cl (Bipy = 2,2′-Bipyridine). J. Chem. Soc., Chem. Commun.
1984, 984 (6), 328–330.
12. Hawecker, J.; Lehn, J.-M.; Ziessel, R. Efficient Photochemical Reduction of CO2 to CO by
Visible Light Irradiation of Systems Containing Re(Bipy)(CO)3X or Ru(Bipy)3
2+
-Co
2+

Combinations as Homogeneous Catalysts. J. Chem. Soc., Chem. Commun. 1983, 9, 536–
538.
13. Blakemore, J. D.; Gupta, A.; Warren, J. J.; Brunschwig, B. S.; Gray, H. B. Noncovalent
Immobilization of Electrocatalysts on Carbon Electrodes for Fuel Production. J. Am. Chem.
Soc. 2013, 135 (49), 18288–18291.
178

14. Schreier, M.; Luo, J.; Gao, P.; Moehl, T.; Mayer, M. T.; Grätzel, M. Covalent
Immobilization of a Molecular Catalyst on Cu2O Photocathodes for CO2 Reduction. J. Am.
Chem. Soc. 2016, 138 (6), 1938–1946.
15. Oh, S.; Gallagher, J. R.; Miller, J. T.; Surendranath, Y. Graphite-Conjugated Rhenium
Catalysts for Carbon Dioxide Reduction. J. Am. Chem. Soc. 2016, 138 (6), 1820–1823.
16. Friebe, C.; Hager, M. D.; Winter, A.; Schubert, U. S. Metal-Containing Polymers via
Electropolymerization. Adv. Mater. 2012, 24 (3), 332–345.
17. Windle, C. D.; Perutz, R. N. Advances in Molecular Photocatalytic and Electrocatalytic
CO2 reduction. Coord. Chem. Rev. 2012, 256 (21–22), 2562–2570.
18. O’Toole, T. R.; Margerum, L. D.; Westmoreland, T. D.; Vining, W. J.; Murray, R. W.;
Meyer, T. J. Electrocatalytic Reduction of CO2 at a Chemically Modified Electrode. J.
Chem. Soc. Chem. Commun. 1985, 20, 1416.
19. Cabrera, C.; Abruña, H. D. Electrocatalysis of CO2 Reduction at Surface Modified Metallic
and Semiconducting Electrodes. J. Electroanal. Chem. 1986, 209, 101–107.
20. O’Toole, T. R.; Sullivan, B. P.; Bruce, M. R.-M.; Margerum, L. D.; Murray, R. W.; Meyer,
T. J. Electrocatalytic Reduction of CO2 by a Complex of Rhenium in Thin Polymeric Films.
J. Electroanal. Chem. Interfacial Electrochem. 1989, 259 (1–2), 217–239.
21. Kamata, R.; Kumagai, H.; Yamazaki, Y.; Sahara, G.; Ishitani, O. Photoelectrochemical
CO2 Reduction Using a Ru(II)-Re(I) Supramolecular Photocatalyst Connected to a Vinyl
Polymer on a NiO Electrode. ACS Applied Materials & Interfaces 2018, Article ASAP,
DOI: 10.1021/acsami.8b05495.
22. Chan, W. K. Metal Containing Polymers with Heterocyclic Rigid Main Chains. Coord.
Chem. Rev. 2007, 251 (17–20), 2104–2118.
23. Maruyama, T.; Yamamoto, T. System Based on Chelating π-Conjugated Poly (2,2′-
Bipyridine-5,5′-Diyl) and Platinum for Photoevolution of H2 from Aqueous Media and
Spectroscopic Analysis of the Catalyst. J. Phys. Chem. B 1997, 101 (19), 3806–3810.
24. Yamamoto, T.; Maruyama, T.; Zhou, Z. H.; Ito, T.; Fukuda, T.; Yoneda, Y.; Begum, F.;
Ikeda, T.; Sasaki, S.; Takezoe, H.; Fukuda, A.; Kubota, K. π-Conjugated Poly(Pyridine-
2,5-Diyl), Poly(2,2′-Bipyridine-5,5′-Diyl), and Their Alkyl Derivatives. Preparation,
Linear Structure, Function as a Ligand to Form Their Transition Metal Complexes,
Catalytic Reactions, n-Type Electrically Conducting Properties. J. Am. Chem. Soc. 1994,
116 (11), 4832–4845.

179

25. Hayashida, N.; Yamamoto, T. Synthesis of Ru- and Os-Complexes of π-Conjugated
Oligomers of 2,2′-Bipyridine and 5,5′-Bipyrimidine. Optical Properties and Catalytic
Activity for Photoevolution of H2 from Aqueous Media. Bulletin of the Chemical Society
of Japan. 1999, 72 (5), 1153–1162.
26. Ley, K. D.; Whittle, C. E.; Bartberger, M. D.; Schanze, K. S. Photophysics of π-Conjugated
Polymers That Incorporate Metal to Ligand Charge Transfer Chromophores. J. Am. Chem.
Soc. 1997, 119 (14), 3423–3424.
27. Portenkirchner, E.; Gasiorowski, J.; Oppelt, K.; Schlager, S.; Schwarzinger, C.;
Neugebauer, H.; Knör, G.; Sariciftci, N. S. Electrocatalytic Reduction of Carbon Dioxide
to Carbon Monoxide by a Polymerized Film of an Alkynyl-Substituted Rhenium(I)
Complex. ChemCatChem 2013, 5 (7), 1790–1796.
28. Mahouche-Chergui, S.; Gam-Derouich, S.; Mangeney, C.; Chehimi, M. M. Aryl
Diazonium Salts: A New Class of Coupling Agents for Bonding Polymers,
Biomacromolecules and Nanoparticles to Surfaces. Chem. Soc. Rev. 2011, 40 (7), 4143–
4166.
29. Bangle, R.; Sampaio, R. N.; Troian-Gautier, L.; Meyer, G. J. Surface Grafting of Ru(II)
Diazonium-Based Sensitizers on Metal Oxides Enhances Alkaline Stability for Solar
Energy Conversion. ACS Appl. Mater. Interfaces 2018, 10 (3), 3121–3132.
30. Marshall, N.; Locklin, J. Reductive Electrografting of Benzene (p-Bisdiazonium
Hexafluorophosphate): A Simple and Effective Protocol for Creating Diazonium-
Functionalized Thin Films. Langmuir 2011, 27 (21), 13367–13373.
31. Smieja, J. M.; Kubiak, C. P. Re(Bipy-tBu)(CO)3Cl-Improved Catalytic Activity for
Reduction of Carbon Dioxide: IR-Spectroelectrochemical and Mechanistic Studies. Inorg.
Chem. 2010, 49 (20), 9283–9289.
32. Zhao, Y.; Truhlar, D. G. The M06 Suite of Density Functionals for Main Group
Thermochemistry, Thermochemical Kinetics, Noncovalent Interactions, Excited States,
and Transition Elements: Two New Functionals and Systematic Testing of Four M06-Class
Functionals and 12 Other Function. Theor. Chem. Acc. 2008, 120 (1–3), 215–241.
33. Wadt, W. R.; Hay, P. J. Ab Initio Effective Core Potentials for Molecular Calculations.
Potentials for Main Group Elements Na to Bi. J. Chem. Phys. 1985, 82 (1), 284–298.
34. Shao, Y.; Molnar, L.; Jung, Y.; Kussmann, J.; Ochsenfeld, C.; Brown, S.; Gilbert, A.;
Slipchenko, L.; Levchenko, S.; O'Neill, D.; DiStasio, R.; Lochan, R.; Wang, T.; Beran, G.;
Besley, N.; Herbert, J.; Lin, C.; Van Voorhis, T.; Chien, S.; Sodt, A.; Steele, R.; Rassolov,
V.; Maslen, P.; Korambath, P.; Adamson, R.; Austin, B.; Baker, J.; Byrd, E.; Dachsel, H.;
Doerksen, R.; Dreuw, A.; Dunietz, B.; Dutoi, A.; Furlani, T.; Gwaltney, S.; Heyden, A.;
Hirata, S.; Hsu, C.; Kedziora, G.; Khalliulin, R.; Klunzinger, P.; Lee, A.; Lee, M.; Liang,
W.; Lotan, I.; Nair, N.; Peters, B.; Proynov, E.; Pieniazek, P.; Rhee, Y.; Ritchie, J.; Rosta,
180

E.; Sherrill, C.; Simmonett, A.; Subotnik, J.; Woodcock, H.; Zhang, W.; Bell, A.;
Chakraborty, A.; Chipman, D.; Keil, F.; Warshel, A.; Hehre, W.; Schaefer, H.; Kong, J.;
Krylov, A.; Gill, P.; Head-Gordon, M. Advances in Methods and Algorithms in a Modern
Quantum Chemistry Program Package. Phys. Chem. Chem. Phys. 2006, 8 (27), 3172-3191.
35. Kariuki, J. K.; McDermott, M. T. Formation of Multilayers on Glassy Carbon Electrodes
via the Reduction of Diazonium Salts. Langmuir 2001, 17 (19), 5947–5951.
36. Smieja, J. M.; Benson, E. E.; Kumar, B.; Grice, K. A.; Seu, C. S.; Miller, A. J. M.; Mayer,
J. M.; Kubiak, C. P. Kinetic and Structural Studies, Origins of Selectivity, and Interfacial
Charge Transfer in the Artificial Photosynthesis of CO. Proc. Natl. Acad. Sci. U.S.A. 2012,
109 (39), 15646–15650.
37. Jousselme, B.; Bidan, G.; Billon, M.; Goyer, C.; Kervella, Y.; Guillerez, S.; Hamad, E. A.;
Goze-Bac, C.; Mevellec, J. Y.; Lefrant, S. One-Step Electrochemical Modification of
Carbon Nanotubes by Ruthenium Complexes via New Diazonium Salts. J. Electroanal.
Chem. 2008, 621 (2), 277–285.
38. Andrieux, C. P.; Dumas-Bouchiat, J. M.; Saveant, J. M. Catalysis of Electrochemical
Reactions at Redox Polymer Electrodes. Effect of the Film Thickness. J. Electroanal.
Chem. 1980, 114 (1), 159–163.
39. Ceccato, M.; Bousquet, A.; Hinge, M.; Pedersen, S. U.; Daasbjerg, K. Using a Mediating
Effect in the Electroreduction of Aryldiazonium Salts to Prepare Conducting Organic Films
of High Thickness. Chem. Mater. 2011, 23 (6), 1551–1557.
40. Machan, C. W.; Sampson, M. D.; Chabolla, S. A.; Dang, T.; Kubiak, C. P. Developing a
Mechanistic Understanding of Molecular Electrocatalysts for CO2 Reduction Using
Infrared Spectroelectrochemistry. Organometallics 2014, 33 (18), 4550–4559.
41. Permatasari, F. A.; Aimon, A. H.; Iskandar, F.; Ogi, T.; Okuyama, K. Role of C-N
Configurations in the Photoluminescence of Graphene Quantum Dots Synthesized by a
Hydrothermal Route. Sci. Rep. 2016, 6, 1–8.
42. Ricks, A. M.; Anfuso, C. L.; Rodríguez-Córdoba, W.; Lian, T. Vibrational Relaxation
Dynamics of Catalysts on TiO2 Rutile (110) Single Crystal Surfaces and Anatase
Nanoporous Thin Films. Chem. Phys. 2013, 422, 264–271.
43. Kurtz, D. A.; Brereton, K. R.; Ruo, K. P.; Tang, H. M.; Felton, G. A. N.; Miller, A. J. M.;
Dempsey, J. L. Bathochromic Shifts in Rhenium Carbonyl Dyes Induced through
Destabilization of Occupied Orbitals. Inorg. Chem. 2018, 57 (9), 5389–5399.
44. Walters, K. A.; Premvardhan, L. L.; Liu, Y.; Peteanu, L. A.; Schanze, K. S. Metal-to-
Ligand Charge Transfer Absorption in a Rhenium (I) Complex That Contains a π-
Conjugated Bipyridine Acceptor Ligand. Chem. Phys. Lett. 2001, 339, 255–262.
181

45. Hossain, M. M.; Ivanov, M. V.; Wang, D.; Reid, S. A.; Rathore, R. Spreading Electron
Density Thin: Increasing the Chromophore Size in Polyaromatic Wires Decreases
Interchromophoric Electronic Coupling. J. Phys. Chem. C 2018, 112 (31), 17688–17675.
46. Savéant, J. M. Electron Hopping between Localized Sites. Coupling with Electroinactive
Counterion Transport. J. Phys. Chem. 1988, 92 (5), 1011–1013.
47. Denisevich, P.; Abruña, H. D.; Leidner, C. R.; Meyer, T. J.; Murray, R. W.
Electropolymerization of Vinylpyridine and Vinylbipyridine Complexes of Iron and
Ruthenium: Homopolymers, Copolymers, Reactive Polymers. Inorg. Chem. 1982, 21 (6),
2153–2161.
48. Costentin, C.; Saveant, J.-M. Cyclic Voltammetry Analysis of Electrocatalytic Films. J.
Phys. Chem. C 2015, 119 (22), 12174-12182.
49. Miyamae, T.; Yoshimura, D.; Ishii, H.; Ouchi, Y.; Seki, K.; Miyazaki, T.; Koike, T.;
Yamamoto, T. Ultraviolet Photoelectron Spectroscopy of Poly(Pyridine-2,5-Diyl),
Poly(2,2′-Bipyridine-5,5′-Diyl), and Their K-Doped States. J. Chem. Phys. 1995, 103 (7),
2738–2744.
50. Kramer, W. W.; McCrory, C. C. L. Polymer Coordination Promotes Selective CO 2
Reduction by Cobalt Phthalocyanine. Chem. Sci. 2016, 7 (4), 2506–2515.
51. Pognon, G.; Brousse, T.; Bélanger, D. Effect of Molecular Grafting on the Pore Size
Distribution and the Double Layer Capacitance of Activated Carbon for Electrochemical
Double Layer Capacitors. Carbon 2011, 49 (4), 1340–1348.
52. Zhou, X.; Micheroni, D.; Lin, Z.; Poon, C; Li, Z.; Lin, W. Graphene-Immobilized fac-
Re(bipy)(CO)3Cl for Syngas Generation from Carbon Dioxide. ACS Appl. Mater.
Interfaces 2016, 8, 4192-4198.
53. Gatty, M. G.; Pullen, S.; Sheibani, E.; Tian, H.; Ott, S.; Hammarström, L. Direct Evidence
of Catalyst Reduction on Dye and Catalyst Co-Sensitized NiO Photocathodes by Mid-
Infrared Transient Absorption Spectroscopy. Chem. Sci. 2018, 9 (22), 4983–4991.
54. Pekarek, R. T.; Kearney, K.; Simon, B. M.; Ertekin, E.; Rockett, A. A.; Rose, M. J.
Identifying Charge Transfer Mechanisms across Semiconductor Heterostructures via
Surface Dipole Modulation and Multiscale Modeling. J. Am. Chem. Soc. 2018. Article
ASAP, DOI: 10.1021/jacs.8b05057.
55. Habisreutinger, S. N.; Schmidt-Mende, L.; Stolarczyk, J. K. Photocatalytic Reduction of
CO2 on TiO2 and Other Semiconductors. Angew. Chemie - Int. Ed. 2013, 52 (29), 7372–
7408.
182

56. Staniszewski, A.; Morris, A. J.; Ito, T.; Meyer, G. J. Conduction Band Mediated Electron
Transfer across Nanocrystalline TiO2 Surfaces. J. Phys. Chem. B 2007, 111 (24), 6822–
6828.
57. Morimoto, T.; Nakajima, T.; Sawa, S.; Nakanishi, R.; Imori, D.; Ishitani, O. CO2 Capture
by a Rhenium(I) Complex with the Aid of Triethanolamine. J. Am. Chem. Soc. 2013, 135
(45), 16825–16828.
58. Abdellah, M.; El-Zohry, A. M.; Antila, L. J.; Windle, C. D.; Reisner, E.; Hammarström, L.
Time-Resolved IR Spectroscopy Reveals a Mechanism with TiO2 as a Reversible Electron
Acceptor in a TiO2–Re Catalyst System for CO2 Photoreduction. J. Am. Chem. Soc. 2017,
139 (3), 1226–1232.
59. Windle, C. D.; Pastor, E.; Reynal, A.; Whitwood, A. C.; Vaynzof, Y.; Durrant, J. R.; Perutz,
R. N.; Reisner, E. Improving the Photocatalytic Reduction of CO2 to CO through
Immobilisation of a Molecular Re Catalyst on TiO2. Chem. - A Eur. J. 2015, 21 (9), 3746–
3754.
60. Takeda, H.; Koike, K.; Inoue, H.; Ishitani, O. Development of an Efficient Photocatalytic
System for CO2 Reduction Using Rhenium(I) Complexes Based on Mechanistic Studies.
J. Am. Chem. Soc. 2008, 130 (6), 2023–2031.
61. Kou, Y.; Nabetani, Y.; Masui, D.; Shimada, T.; Takagi, S.; Tachibana, H.; Inoue, H. Direct
Detection of Key Reaction Intermediates in Photochemical. J. Am. Chem. Soc. 2014, 136
(136), 6021–6030.
62. Shinde, D. B.; Aiyappa, H. B.; Bhadra, M.; Biswal, B. P.; Wadge, P.; Kandambeth, S.;
Garai, B.; Kundu, T.; Kurungot, S.; Banerjee, R. A Mechanochemically Synthesized
Covalent Organic Framework as a Proton-Conducting Solid Electrolyte. J. Mater. Chem.
A 2016, 4 (7), 2682–2690.
63. Savéant, J.-M. Elements of Molecular and Biomolecular Electrochemistry; Wiley-
Interscience: New York, NY, 2006.
64. Costentin, C.; Savéant, J.-M. Multielectron, Multistep Molecular Catalysis of
Electrochemical Reactions: Benchmarking of Homogeneous Catalysts. ChemElectroChem
2014, 1 (7), 1226–1236.
65. Rountree, E. S.; McCarthy, B. D.; Eisenhart, T. T.; Dempsey, J. L. Evaluation of
Homogeneous Electrocatalysts by Cyclic Voltammetry. Inorg. Chem. 2014, 53 (19), 9983–
10002.
66. Shao, Y.; Molnar, L.; Jung, Y.; Kussmann, J.; Ochsenfeld, C.; Brown, S.; Gilbert, A.;
Slipchenko, L.; Levchenko, S.; O'Neill, D.; DiStasio, R.; Lochan, R.; Wang, T.; Beran, G.;
Besley, N.; Herbert, J.; Lin, C.; Van Voorhis, T.; Chien, S.; Sodt, A.; Steele, R.; Rassolov,
V.; Maslen, P.; Korambath, P.; Adamson, R.; Austin, B.; Baker, J.; Byrd, E.; Dachsel, H.;
183

Doerksen, R.; Dreuw, A.; Dunietz, B.; Dutoi, A.; Furlani, T.; Gwaltney, S.; Heyden, A.;
Hirata, S.; Hsu, C.; Kedziora, G.; Khalliulin, R.; Klunzinger, P.; Lee, A.; Lee, M.; Liang,
W.; Lotan, I.; Nair, N.; Peters, B.; Proynov, E.; Pieniazek, P.; Rhee, Y.; Ritchie, J.; Rosta,
E.; Sherrill, C.; Simmonett, A.; Subotnik, J.; Woodcock, H.; Zhang, W.; Bell, A.;
Chakraborty, A.; Chipman, D.; Keil, F.; Warshel, A.; Hehre, W.; Schaefer, H.; Kong, J.;
Krylov, A.; Gill, P.; Head-Gordon, M. Advances in Methods and Algorithms in a Modern
Quantum Chemistry Program Package. Phys. Chem. Chem. Phys. 2006, 8 (27), 3172-3191.
67. Zhao, Y.; Truhlar, D. G. The M06 Suite of Density Functionals for Main Group
Thermochemistry, Thermochemical Kinetics, Noncovalent Interactions, Excited States,
and Transition Elements: Two New Functionals and Systematic Testing of Four M06-Class
Functionals and 12 Other Function. Theor. Chem. Acc. 2008, 120 (1–3), 215–241.
68. Ditchfield, R.; Hehre, W. J.; Pople, J. A. Self-Consistent Molecular-Orbital Methods. IX.
An Extended Gaussian-Type Basis for Molecular-Orbital Studies of Organic Molecules. J.
Chem. Phys. 1971, 54 (2), 724–728.
69. Hehre, W. J.; Ditchfield, K.; Pople, J. A. Self-Consistent Molecular Orbital Methods. XII.
Further Extensions of Gaussian-Type Basis Sets for Use in Molecular Orbital Studies of
Organic Molecules. J. Chem. Phys. 1972, 56 (5), 2257–2261.
70. Hay, P. J.; Wadt, W. R. Ab Initio Effective Core Potentials for Molecular Calculations.
Potentials for the Transition Metal Atoms Sc to Hg. J. Chem. Phys. 1985, 82 (1), 270–283.
71. Wadt, W. R.; Hay, P. J. Ab Initio Effective Core Potentials for Molecular Calculations.
Potentials for Main Group Elements Na to Bi. J. Chem. Phys. 1985, 82 (1), 284–298.
184

CHAPTER 5

Immobilized Molecular Wires on Carbon Cloth Electrodes Facilitate CO2 Electrolysis

A portion of this chapter has appeared in print:

Orchanian, N. M.; Hong, L. E.; Marinescu, S. C.* “Immobilized Molecular Wires on Carbon Cloth
Electrodes Facilitate CO2 Electrolysis”, ACS Catal., 2019, 9, 10, 9393-9397.
185

5.1 ABSTRACT
Conjugated molecular wires of rhenium bipyridine complexes were grown on flexible,
lightweight, carbon cloth electrodes through reductive diazonium electropolymerization. CO2
electrolysis studies reveal rapid (kcat ~ 40 s
-1
) and selective (Faradaic efficiency >99%) conversion
to CO with turnover numbers (TON) per rhenium site reaching ~290,000 and catalytic currents
(icat) >10 mA/cm
2
. This represents over a 80-fold increase in activity relative to our prior graphite
systems, and a ~25-fold increase relative to the highest-performing immobilized rhenium
bipyridine catalyst to date. The high activity of these electrodes is explained by a mechanism
initiated via electrochemical charging of the π-conjugated backbone followed by anion
dissociation, CO2 coordination, and protonation. As numerous metal-bipyridine complexes are
known for a broad scope of electrocatalytic transformations, these integrated carbon cloth devices
are anticipated to serve as a platform for future studies.

Keywords: Surface Modification, CO2 Reduction, Electrocatalysis, Molecular Wires,
Electropolymerization, Metallopolymer, Solar Energy Conversion

5.2 INTRODUCTION
The emergence of low-cost, high-efficiency photovoltaic cells has led to an increasing
availability of affordable solar electricity.
1
While solar energy has penetrated grid-scale
deployment, the spatial and temporal variability of sunshine has fueled a growing need for
electricity storage.
2
In particular, new technologies are required to consume excess current during
times of low-energy-demand and release this energy during peak hours.
3
One attractive approach
to this problem is the production of solar fuels: driving the transformation of abundant small-
molecules to value-added chemical products through electrolysis.
4
As the atmospheric
concentration of CO2 is steadily increasing with fossil fuel consumption, recycling CO2 into
186

industrial chemicals both provides a channel for energy storage and sequesters this atmospheric
pollutant.
5
The CO generated through such a process could then be used for methanol production
or for Fischer-Tropsch chemistry to synthesize liquid fuels.
6,7,8
As the byproduct of this reaction
is water, CO2 electrolyzers are also desirable for CO2-to-O2 converters in atmospheric control
systems for manned spaceflight.
9,10
For these applications to come to fruition, advanced methods
to integrate electrocatalysts with a supporting electrode are critical.
11

Our group has previously reported the application of a directed electropolymerization
methodology for the generation of surface-immobilized 1D metallopolymers.
12
This methodology
is based on the electrochemical reduction of bis(diazonium) salts which leads to surface attachment
and subsequent 5,5'-directed polymerization. These molecular wires containing rhenium
bipyridine tricarbonyl moieties offer robust integration with a broad range of supporting
electrodes. An optimal catalyst loading on graphite rod electrodes was determined resulting in
turnover numbers (TON) per rhenium site for CO2-to-CO conversion to reach ~3600 during two
hours of electrolysis, with increased catalyst loadings displaying electron- and mass-transport
limitations. Prior studies related to the conjugation of molecular units to graphite electrodes have
shown that conjugation imparts unique functionality to catalytic sites, with a conjugated rhenium
phenanthroline systems reaching TON per Re site of ~12,000 for CO2 reduction to CO.
13,14
While
the electropolymerization methodology imparts a break in conjugation between the electrode and
catalyst, related studies have shown that the rhenium bipyridine units in these wires exhibit
conjugation with respect to each other. This extended conjugation, as well as the broad substrate
scope, tunable catalyst loading, and unique 1D growth mechanism of our methodology warranted
further exploration. In particular, carbon cloth substrates have recently been demonstrated as
flexible, lightweight electrodes for CO2 electrolysis with immobilized molecular complexes.
15,16

187

We proposed to transfer our electropolymerization studies to these high-surface-area carbon cloth
electrodes (CCEs) to mitigate the previously-observed electron- and mass-transport
limitations.
16,17
Herein, we report the electrochemical modification of CCEs with 1D
metallopolymers for rapid and selective CO2-to-CO conversion.

5.3 RESULTS AND DISCUSSION

Figure 5.1 High-resolution scanning electron microscopy (HR-SEM) imaging of the carbon cloth
electrode and modification procedure via directed electropolymerization.

Electrodes were modified under the optimized conditions reported previously by our group,
illustrated schematically in Figure 5.1. An unmodified CCE was immersed in a 1 mM solution of
the bis(diazonium) monomer, [Re(CO)3(Cl)(5,5'-bis(diazonium)-2,2'-bipyridine)], and
functionalized via cyclic-voltammetry-initiated electropolymerization (see SI for details). CCEs
were prepared with a sequence of five cyclic voltammetry (CV) scans from -0.6 V to -1.6 V at a
scan rate of 1 V/s (all potentials referenced vs Fc/Fc
+
), then washed thoroughly to remove any
unbound material.

The electrochemical behavior of Re-CCE was probed by CV studies in the absence of
monomer (Figure 5.2) to confirm the electroactive nature of the resulting molecular film.
18
A Re-
CCE device was transferred to a solution of 0.1 M [nBu4N][PF6] in acetonitrile and variable scan
188

rate experiments were performed under inert and CO2 atmosphere. Under inert atmosphere (N2),
the two-electron reduction of the rhenium bipyridine moiety is observed as a broad two-electron
reduction event at -1.85 V with near-zero peak separation at low scan rates (12.1 mV separation at
50 mV/s). This broad feature exhibits linearly increasing peak current densities as a function of
the scan rate, as expected for surface-confined redox systems (see Figure 5.3). The measured peak
separation increases with scan rate, as observed previously for these systems, and has been
attributed to diffusion-like transport of electrons through the π-conjugated wires.
19
A pre-wave is
also observed at -1.40 V, which has previously been assigned to ion-coupled population of trap
states associated with the π-backbone.
20,21

Figure 5.2 Cyclic voltammetry of Re-CCE in acetonitrile solvent with 0.1 M [nBu4N][PF6]
electrolyte under N2 atmosphere (left) and CO2 atmosphere (right).

189

Figure 5.3 Randles-Sevcik analysis of Re-CCE, indicating a linear dependency of J (mA/cm
2
) on
the scan rate. Studies were performed in acetonitrile solution with 0.1 M [nBu4N][PF6] supporting
electrolyte.

Under CO2 atmosphere, the redox feature assigned to the electroactive polymer displays a
loss of reversibility and takes on a sharp onset current, suggesting catalytic reactivity. As the scan
rate is increased, the waveshape changes to reach a pseudo-plateau shape at 100 mV/s. This change
in waveshape indicates diffusion-controlled current, as is consistent with mass-transport-limited
CO2 coordination to the active site.
22
At sufficiently high scan rates (500 mV/s), this pseudo-
plateau shape is lost and the catalyst reenters a kinetic regime with higher charging currents than
observed at slow scan rates.

To probe the stability and selectivity of Re-CCE under CO2 electrolysis conditions,
controlled-potential electrolysis (CPE) studies were performed. In a typical experiment, the Re-
CCE device served as the working electrode and a silver wire separated from solution by a Vycore
frit was employed as a pseudo-reference electrode in acetonitrile solvent with 0.1 M [nBu4N][PF6]
supporting electrolyte. The working cell was separated by a glass frit from a counter cell containing
an unmodified CCE as the counter electrode. As shown in Figure 5.4, an average current density
(Javg) of ~11 mA/cm
2
was maintained for two hours of electrolysis at -2.70 V. The decrease in
190

current after 100 minutes corresponded with the evolution of a large quantity of bubbles at the
working electrode, which had begun to cover the Re-CCE surface and hinder catalysis. Samples
of the headspace were collected at 60 minute intervals and analyzed by gas chromatography to
quantify the gaseous products. In all cases, >99% FE towards CO production (FECO) were
measured with no concomitant H2 production. It was determined that 622.7 μmol CO was produced
over the course of two hours of electrolysis, indicating rapid and selective catalysis. Analogous
studies in the absence of CO2 or catalyst (unmodified CCE) indicate no CO production, confirming
that the molecular units are responsible for the observed catalytic behavior (see Table 5.1).

Figure 5.4 Controlled potential electrolysis study with Re-CCE under H-cell conditions in
acetonitrile solvent with 0.1 M [nBu4N][PF6] supporting electrolyte.

191

Table 5.1 Summary of results from controlled-potential electrolysis (CPE) studies. All studies
were performed in acetonitrile solution with 0.1 M [nBu4N][PF6] supporting electrolyte.
Sample Loading
(nmol)
N2/CO2
(1 atm)
Potential
(V vs
Fc/Fc
+
)
Duration
(s)
Charge
Passed
(C)
μmol
CO

μmol
H2
FE
(%
CO)
FE
(%
H2)
CCE 0 CO2 -2.25 3600 0.88 0 0 N/A N/A
CCE 0 CO2 -2.70 3600 0.94 0 0 N/A N/A
Re-CCE-
1
3.2(3) N2 -2.25 3600 1.02 0 0 N/A N/A
Re-CCE-
1
3.2(3) CO2 -1.85 3600 1.04 0.30(3) 0 1 N/A
Re-CCE-
1
3.2(3) CO2 -2.25 3600 1.19 6.0(6) 0 98(6) N/A
Re-CCE-
1
3.2(3) CO2 -2.70 3600 111.40 581(60) 0 99(6) N/A
Re-CCE-
1
3.2(3) CO2 -2.90 1200 15.62 84(8) 0 99(6) N/A
Re-CCE-
2
2.1(2) N2 -2.70 3600 1.33 0 0 N/A N/A
Re-CCE-
2
2.1(2) CO2 -2.70 7200 122.36 622(60) 0 99(6) N/A

Table 5.2 Calculation of catalytic parameters (TON, TOF) based on ICP loading. All studies
performed under 1 atmosphere CO2.
Sample Loading
(nmol)
Potential
(V vs
Fc/Fc
+
)
Duration
(s)
μmol CO

TONtot (CO) TOFtot
(CO)
Re-CCE-1 3.2(3) -1.85 3600 0.30(3) 103(10) 0.03(1)
Re-CCE-1 3.2(3) -2.25 3600 6.0(6) 1,847(200) 0.5(1)
Re-CCE-1 3.2(3) -2.70 3600 581(60) 177,909(18000) 49(5)
Re-CCE-1 3.2(3) -2.90 1200 84(8) 25,033(2500) 20(2)
Re-CCE-2 2.1(2) -2.70 7200 622(60) 287,474(29000) 39(4)

Based on a measured loading of 4.0(4) nmol/cm
2
for Re-CCE, calculated catalytic
parameters of TONtot = 287,474 and TOFtot = 39.9 s
-1
were determined (see Table 5.2). The TONtot
measured for Re-CCE represents over a 80-fold increase relative to our prior systems and an ~25-
fold increase relative to the highest-performing immobilized rhenium bipyridine system to date
(TON ~12,000). The high TOFtot determined for this system represents a dramatic increase relative
to that measured on graphite rod electrodes (0.50 s
-1
), suggesting that the improved mass transport
192

kinetics provided by carbon cloth substrates facilitate rapid CO2 electrolysis. A single electrode
was able to be reused up to four times for a total of ~4.5 hours of electrolysis with no catalyst
leaching observed (Tables 5.1-5.2).

Figure 5.5 High-resolution Re 4f (left) and Cl 2p (right) X-ray photoelectron spectroscopy of Re-
CCE before (top) and after (bottom) two hours of electrolysis.

To characterize the elemental composition of the Re-CEE system, high-resolution X-ray
photoelectron spectroscopy (XPS) studies were explored. As shown in Figure 5.5, the as-prepared
Re-CEE exhibits XPS signatures in both the Re 4f and Cl 2p regions. Analysis of the Re 4f region
reveals two features with binding energies of 43.3 eV and 41.0 eV, corresponding to the 4f5/2 and
4f7/2 levels, respectively. These features exhibit full-width-half-maximum (FWHM) values of 1.1
eV, indicating a single rhenium environment. The binding energies observed in this region are
consistent with our previously-characterized materials (43.4 eV and 41.0 eV), which suggests that
polymerization proceeds similarly on these CCE systems.
12
Analysis of the Cl 2p region reveals
193

two features with binding energies of 198.7 eV and 197.1 eV, which correspond to the 2p 3/2 and
2p1/2 levels, respectively. Only one chlorine environment is observed, as indicated by a FWHM of
1.2 eV, and the binding energies measured are analogous to our prior immobilized systems. The
bulk loading of rhenium on Re-CCE was determined by inductively-coupled plasma mass
spectrometry (ICP-MS) for the calculation of per-Re-site total turnover number (TONtot) and total
turnover frequency (TOFtot). Samples were digested overnight in concentrated nitric acid for ICP-
MS studies, and no rhenium was detected for the unmodified carbon cloth electrode (see SI for
details).

Following electrolysis studies, the Re-CCE devices were analyzed again by XPS to assess
whether structural changes could be observed in the post-catalysis films. The Re 4f and Cl 2p
regions of Re-CCE are shown in Figure 5.5 after two hours of electrolysis at -2.70 V. The Re 4f
region retains two features associated with the 4f5/2 and 4f7/2 levels, with a minor shift (0.2 eV)
towards lower binding energies (43.1 eV and 40.8 eV, respectively). This minor shift in binding
energy is not consistent with the formation of rhenium nanoparticles, but appears to be consistent
with the previously-reported n-doping observed in 1D rhenium bipyridine metallopolymers.
17
It
was observed that π-conjugated rhenium bipyridine wires exhibit a color change from yellow to
blue upon doping, and our prior studies on glassy carbon electrodes similarly indicated a color
change from yellow to blue in the as-deposited films after electroreduction, suggesting
electrochemical n-doping occurs upon polarization.
12
The post-electrolysis Cl 2p region exhibits
no features, which suggests that the chloride ligands are quantitatively dissociated upon reduction
of the π-backbone. This is consistent with the reduction-initiated chloride dissociation observed in
our prior studies related to these 1D polymers as well as analogous rhenium bipyridine complexes.
The absence of chloride following electrolysis provides strong evidence that all of the rhenium
194

sites along the molecular wires are electroactive. No rhenium was detected in the post-electrolysis
solution by ICP-MS, suggesting the metallopolymers are stable to sustained catalytic studies.

Figure 5.6 Kinetic isotope effect (KIE) study of Re-CCE. Cyclic voltammetry was carried out in
the presence of 0.5 M H2O (blue trace) and 0.5 M D2O (red trace) to determine a ratio of current
densities for KIE estimation. KIE was estimate at -2.25 V (dashed line).

Figure 5.7 Catalytic Tafel plots collected in under 1 atmosphere CO2 in acetonitrile solvent with
0.1 M [nBu4N][PF6] supporting electrolyte.

195

Table 5.3 Comparison to selected literature examples.
Sample Potential Duration (s) FECO (%) TONCO TOFCO (s
-1
) Ref.
[Re(bpy)] -2.11 V vs
Fc/Fc

50400 98 26 5  10
-3
4
[Re(apbpy)]
polymer
-2.10 V vs
Fc/Fc
+

1740 100 402.4 0.23 5
[Re(vbpy)]
polymer
-1.55 V vs
SCE
4800 90 512 0.1 6
GCC-Re -2.00 V vs
Fc/Fc
+

5040 96(3) 12000 2.5 7
[Re(bdbpy)]
polymer
-2.25 V vs
Fc/Fc
+

3600 99(7) 3606 0.50(5) 3
Re-CCE-2 -2.70 V vs
Fc/Fc
+

7200 99(6) 287,474(29000) 39(4) this
work

To probe the mechanism of CO2 reduction with Re-CCE, additional CV studies were
performed. CV scans under CO2 in the presence of 0.5 M H2O and 0.5 M D2O were explored to
probe the kinetic-isotope effect (KIE, see SI).
23
In the presence of either additive, the catalytic
current increased relative to CO2 alone, confirming a proton-coupled reduction mechanism (see
SI). Based on the ratio of the currents generated under H2O and D2O, a KIE of 1.2 was calculated
at -2.25 V, corresponding to the potential applied during electrolysis studies (see Figure 5.6). A
measured KIE >1 indicates that a proton is involved in the rate-determining step of catalysis, as
has been shown to be the case for the unsubstituted Re(bpy)(CO)3Cl complex. To further
understand the mechanism of activity, Tafel plots were prepared and Tafel slopes were determined
in both the low-overpotential regime and high-overpotential regime (see Figure 5.7).
24
A large
Tafel slope of ~240 mV/dec was determined at the potential applied during electrolysis (-2.25 V),
which has previously been associated with rate-limiting electron transfer. This Tafel slope
continues to increase linearly at more cathodic potentials (~690 mV/dec at -2.55 V), indicating
that mass transport of CO2 becomes limiting under these conditions.
25
Together, these data suggest
that a PCET step for the conversion of the metal-carboxylic acid to the neutral tetracarbonyl
196

species may be rate-limiting under electrolysis conditions, which closely parallels that of the
homogeneous species.

Based on the studies described above, a catalytic mechanism is proposed in Scheme 5.1
which parallels that of the molecular species.
26,27
When the carbon cloth electrode potential (EF,CC)
is below the Fermi level of the electroactive molecular wires (measured as the redox potential,
EF,polymer = -1.85 V), only Ohmic current responses are observed. Raising the electrode Fermi level
by increasing the applied potential switches “on” the device, entering the regime where E F,CC >
EF,poly. In this regime, electrons are able to transfer to the π-extended LUMO of the polymer
backbone, facilitating chloride dissociation as is the case for the unmodified rhenium bipyridine
molecular species. This generates the active form of the catalyst, which subsequently undergoes
mass-transport-limited coordination of CO2 followed by protonation to generate the metal-
carboxylic acid. An additional one-electron reduction with protonation facilitates the loss of H2O
and generation of the tetracarbonyl rhenium species. This species then undergoes another one-
electron reduction to release CO and regenerate the active form of the catalyst. When the electrode
potential is switched “off”, electron transfer to the molecular wires is inhibited. Acetonitrile
solvent is then bound to fill the open coordination site at rhenium, generating the post-electrolysis
film with no detected chloride.

197

Scheme 5.1 Proposed mechanism of catalysis.

5..4 CONCLUSION
In conclusion, we have applied a directed electropolymerization methodology to generate
electrode-catalyst devices with remarkable activity towards CO2 reduction. By translating this
broadly-applicable immobilization technique to high-surface-area carbon cloth electrodes (CCEs),
a TONtot per Re site of ~290,000 was measured. This represents an increase of over 80-fold relative
to our prior optimized system and ~25-fold relative to the highest-performing immobilized
rhenium bipyridine systems under analogous electrolysis conditions. Stable current densities of
~11 mA/cm
2
were generated, leading to the evolution of a large quantity of gas bubbles, which
passivated the device in an H-cell configuration. Faradaic efficiencies were determined to be >99%
for CO production, with no competitive H2 evolution detected by gas chromatography. As
numerous metal-bipyridine catalysts are known for a broad scope of electrocatalytic
transformations, we posit that this methodology will serve as a platform for future studies.
198

5.5 EXPERIMENTAL METHODS
5.5.1 Materials and Synthesis
All manipulations of air- and moisture-sensitive materials were conducted under nitrogen
atmosphere in a Vacuum Atmospheres glovebox or on a dual manifold Schlenk line with oven-
dried glassware. Water was deionized with the Millipore Synergy system (18.2 MΩ·cm).
Acetonitrile was degassed with nitrogen, passed through activated alumina columns, and stored
over 4Å Linde-type molecular sieves. The [2,2'-bipyridine]-5,5'-diamine ligand and complex were
synthesized according to reported literature procedures.
28,29
The bis(diazonium) monomer was
synthesized according to our previous report.
12
All other chemical reagents were purchased from
commercial vendors and used without further purification.

5.5.2 Synthesis of diamino bipyridine complex, [Re(CO)3(Cl)([2,2'-bipyridine]-5,5'-diamine)]
The diamino bipyridine complex, [Re(CO)3(Cl)([2,2'-bipyridine]-5,5'-diamine)], prepared
according to precedent.
29
Briefly, a ligand suspension was prepared with 56 mg [2,2'-bipyridine]-
5,5'-diamine in 1.8 mL methanol with sonication for 5 min. In a separate flask, 110 mg Re(CO)5Cl
was brought to reflux in 40 mL toluene. Once at reflux, the ligand suspension was added dropwise
to the rhenium solution and an immediate color change to pale yellow was observed. This was
allowed to heat at reflux for one hour before cooling to room temperature. The reaction flask was
transferred to a freezer at -27 °C overnight, and was subsequently filtered to isolate the pale yellow
product in 85% yield. This was recrystallized by slow diffusion of ether into a concentrated DMF
solution (
1
H in CD3CN: δ 8.25, 8.01, 7.28, and 6.22 ppm).

199

5.5.3 Synthesis of bis(diazonium) monomer, [Re(CO)3(Cl)( [2,2'-bipyridine]-5,5'-
bis(diazonium))]
12
Recrystallized diamino complex (45 mg) was suspended in acetonitrile (1.9 mL) and
briefly sonicated for 5 minutes to generate a uniform suspension. A solution of nitrosonium
tetrafluoroborate (24 mg) was prepared in an inert glovebox and dissolved in acetonitrile (0.9 mL).
Both solutions were cooled to -40 °C in dry ice-acetonitrile baths. Once cooled, the suspension of
diamino complex was added dropwise to the NOBF4 solution, and an immediate color change
from pale-yellow to dark blue was observed. Diethyl ether (6 mL) was added and a dark blue
precipitate was formed, which was collected by filtration, and stored in the dark at -27 °C (
1
H in
CD3CN: δ 10.14, 9.26 and 9.08 ppm).

5.5.4 Modification of CCE Samples

Carbon cloth electrodes (CCEs) were modified based on our previously-optimized
procedure. Modifications to this procedure were made to facilitate parallel-modification of several
electrodes. A stock solution (40 mL) of 0.1 mM bis(diazonium) monomer was prepared in dry
acetonitrile solution with 0.1 M [nBu4N][PF6] supporting electrolyte. An oven-dried, nitrogen
purged 20 mL scintillation vial was capped with a Teflon adapter holding the working, pseudo-
reference, and counter electrodes. A 10 mL aliquot of the grafting stock solution was injected into
the vial and a series of five cyclic voltammetry (CV) scans were run between -0.6 V and -1.6 V at
a scan rate of 1 V/s. Up to three carbon cloth electrodes could be connected in parallel to modify
several samples simultaneously with ICP-determined bulk catalyst loadings within 20% for each
sample (though electrode modification was not optimized for parallel grafting). A clean vial with
a fresh 10 mL aliquot of grafting solution was used for each set of electrodes produced. All Re-
200

CCE samples in this study were prepared individually in a clean vial and fresh 10 mL aliquot of
grafting solution.

5.5.5 Electrochemistry
Electrochemistry experiments were carried out in acetonitrile solution with 0.1 M [nBu-
4N][PF6] electrolyte using a Pine potentiostat. A silver wire purchased from VWR was used as a
pseudo-reference electrode. An unmodified carbon cloth was used as a counter electrode was
purchased from Alfa Aesar. Ohmic drop was compensated using the positive feedback
compensation implemented in the instrument. All reported potentials are referenced relative to
ferrocene (Fc) at 0.0 V. Electrochemical experiments were carried out in a three electrode
configuration electrochemical cell under a nitrogen or CO2 atmosphere using unmodified CCE or
Re-CCE working electrodes. The reference and counter electrodes were isolated in glass capillaries
with Vycor frits and electrochemical studies were performed in a single-cell configuration.

CPE measurements were conducted in a two-chambered H cell. In the first chamber, the
working and reference electrodes were immersed in 40 mL of 0.1 M tetrabutylammonium
hexafluorophosphate (TBAPF6) in acetonitrile. The counter electrode (graphite rod) was placed in
the second chamber in 20 mL of 0.1 M TBAPF6 in acetonitrile. The two chambers were separated
by a fine porosity glass frit and the reference electrode was placed in a separate compartment
connected by a Vycor tip. Carbon cloth electrodes (Fuel Cell Earth, Inc.) were used as the working
and auxiliary electrodes. For gas chromatography experiments, 2 mL of gas were withdrawn from
the headspace of the H cell with a gas-tight syringe. This was injected into a gas chromatography
instrument (Shimadzu GC-2010-Plus) equipped with a BID detector and a Restek ShinCarbon ST
201

Micropacked column. Faradaic efficiencies were determined by dividing the amount of CO
produced as measured by gas chromatography by the amount of CO expected based on the total
charge measured during controlled potential electrolysis. For each experiment, the controlled-
potential electrolysis measurements were performed at least twice (with two samples prepared
under identical conditions), leading to similar behavior. The reported Faradaic efficiencies, TON,
TOF, and µmol of CO produced are average values. TON was calculated by dividing the total mols
of CO generated (as determined by gas chromatography) by the total mols of rhenium in the film
(as determined by ICP). TOF was calculated by dividing TON by the duration of the electrolysis
experiment.

5.5.6 Electrochemical Analysis
Calculation of TOFCPE (Eq. 5.1), and TONCPE (Eq. 5.2) were performed according to
established protocol.
28
This method assumes Nernstian electron transfer to the catalyst. As this
method does not directly consider the quantity of gaseous products generated, the resulting TOFCPE
and TONCPE represent the overall activity towards two-electron-reduced products (both CO and
H2). In Eq. 5.1, i represents the average current during CPE (i = Charge*F.E./time, C/s), F
represents Faraday’s constant (F = 96 485 C/mol), A represents the surface area of the working
electrode (2.5 cm
2
), D represents the diffusion coefficient (determined to be 1 × 10
-5
cm
2
/s based
on Randles-Sevcik analysis of the oxidation feature at -0.21 V), and [cat] represents the bulk
catalyst concentration ([cat] = 1 mM = 1 × 10
-6
mol/cm
3
). In Eq. 5.2, t represents the duration of
the electrolysis experiment (s
-1
).

𝑇 𝑂 𝐹 𝐶𝑃𝐸 =
𝑖 2
𝐹 2
𝐴 2
𝐷 [ 𝑐𝑎 𝑡 ]
2
Eq. 5.1
202

𝑇 𝑂 𝑁 𝐶𝑃𝐸 = 𝑇 𝑂 𝐹 𝐶𝑃𝐸 × 𝑡 Eq. 5.2

Wash tests were performed by removing the catalyst solution from the H-cell via syringe
under a positive pressure of CO2, and rinsing the working cell three times with acetonitrile. The
cell was maintained under 1 atm of CO2, and the electrode was not removed from the cell during
these washings to prevent O2-exposure. As such, a small amount of catalyst remains in the cell
despite these washings, due to the presence of small amounts of catalyst trapped in the pores of
the glass frit that separates the counter and working compartments of the electrolysis cell. The
additional current measured during the wash tests relative to the blank CPE study is representative
of this small amount of solubilized catalyst. No visible material is present on the electrode
following CPE, and much lower current densities are observed in the wash test relative to any of
the preparative CPE studies. As such, these studies strongly suggest that the observed reactivity is
due to solubilized catalyst, as no gaseous products are observed following three wash cycles.

5.5.7 XPS
XPS data were collected using a Kratos AXIS Ultra instrument. The monochromatic X-ray
source was the Al K α line at 1486.6 eV. Low-resolution survey spectra were acquired between
binding energies of 1–1200 eV with 160 eV pass energy. Higher-resolution detailed scans, with a
resolution of ~0.1 eV, were collected on individual XPS lines of interest with 20 eV pass energy.
The sample chamber was maintained at < 2 × 10
–9
Torr during analysis. The XPS data were
analyzed and fit using the CasaXPS software. Re 4f features were fit with spin-orbit component Δ
= 2.43 eV, and Cl 2p features were fit with Δ = 1.60 eV.

203

5.5.8 High-Resolution Scanning Electron Microscopy, HR-SEM
High-resolution scanning electron microscopy (HR-SEM) was performed on a FEI Nova
NanoSEM 450 scanning electron microscope using an accelerating voltage of 15 kV. Images
acquired for carbon cloth electrodes (CCE) illustrate the high surface area of these substrates.

5.5.9 ICP-OES
Inductively coupled plasma optical emission spectroscopy (ICP-OES) measurements were
performed using a Thermo Scientific iCAP 7000 ICP-OES. Samples were digested in 2mL nitric
acid overnight with sonication. These nitric acid samples were subsequently diluted to 25 mL with
DI H2O. Calibration plots were prepared using a series of five rhenium standard solutions with
[Re] = 0.2 mM, 0.8 mM, 1.0 mM, 2.0 mM, and 4.0 mM. Reported loadings represent an average
of three measurements. No rhenium was detected for the unmodified carbon cloth electrode, nor
was rhenium detected in the post-electrolysis solution following catalytic studies.

204

5.6 REFERENCES
1. Fu, R.; Chung, D.; Lowder, T.; Feldman, D.; Ardani, K.; Fu, R.; Chung, D.; Lowder, T.;
Feldman, D.; Ardani, K. U.S. Solar Photovoltaic System Cost Benchmark : Q1 2017 U.S.
NREL 2017, No. September, 1–66.
2. Schmidt, O.; Hawkes, A.; Gambhir, A.; Staffell, I. The Future Cost of Electrical Energy
Storage Based on Experience Rates. Nat. Energy 2017, 2, 1–8.
3. Gielen, D.; Boshell, F.; Saygin, D. Climate and Energy Challenges for Materials Science.
Nat. Mater. 2016, 15, 117–120.
4. Nocera, D. G. Solar Fuels and Solar Chemicals Industry. Acc. Chem. Res. 2017, 50, 616–
619.
5. Francke, R.; Schille, B.; Roemelt, M. Homogeneously Catalyzed Electroreduction of
Carbon Dioxide - Methods, Mechanisms, and Catalysts. Chem. Rev. 2018, 118, 4631–
4701.
6. Qiao, J.; Liu, Y.; Hong, F.; Zhang, J. A Review of Catalysts for the Electroreduction of
Carbon Dioxide to Produce Low-Carbon Fuels. Chem. Soc. Rev., 2014, 43, 631-675.
7. Olah, G. A.; Prakash, G. K. S.; Goeppert, A. Anthropogenic Chemical Carbon Cycle for a
Sustainable Future. J. Am. Chem. Soc. 2011, 133, 12881–12898.
8. Olah, G. A. Beyond Oil and Gas: The Methanol Economy. Angew. Chemie - Int. Ed. 2005,
44, 2636–2639.
9. Schreckenghost, D.; Thronesbery, C.; Bonasso, P.; Kortenkamp, D.; Martin, C. Intelligent
Control of Life Support for Space Missions. IEEE Intell. Syst. 2002, 17, 24–31.
10. Breedlove, B. K.; Ferrence, G. M.; Washington, J.; Kubiak, C. P. A Photoelectrochemical
Approach to Splitting Carbon Dioxide for a Manned Mission to Mars. Matt. Design, 2001,
22, 577-584.
11. Bullock, R. M.; Das, A. K.; Appel, A. M. Surface Immobilization of Molecular
Electrocatalysts for Energy Conversion. Chem. - A Eur. J. 2017, 23, 7626–7641.
12. Orchanian, N. M.; Hong, L. E.; Skrainka, J. A.; Esterhuizen, J. A.; Popov, D. A.;
Marinescu, S. C. Surface-Immobilized Conjugated Polymers Incorporating Rhenium
Bipyridine Motifs for Electrocatalytic and Photocatalytic CO2 Reduction. ACS Appl.
Energy Mater. 2018, 2, 110–123.
13. Jackson, M. N.; Oh, S.; Kaminsky, C. J.; Chu, S. B.; Zhang, G.; Miller, J. T.; Surendranath,
Y. Strong Electronic Coupling of Molecular Sites to Graphitic Electrodes via Pyrazine
Conjugation. J. Am. Chem. Soc. 2018, 140, 1004–1010.
205

14. Oh, S.; Gallagher, J. R.; Miller, J. T.; Surendranath, Y. Graphite-Conjugated Rhenium
Catalysts for Carbon Dioxide Reduction. J. Am. Chem. Soc. 2016, 138, 1820–1823.
15. Rotundo, L.; Filippi, J.; Gobetto, R.; Miller, H. A.; Rocca, R.; Nervi, C.; Vizza, F.
Electrochemical CO2 Reduction in Water at Carbon Cloth Electrodes Functionalized with
a Fac-Mn(apbpy)(CO)3 Br Complex. Chem. Commun. 2019, 55, 775–777.
16. Cheung, P. L.; Lee, S. K.; Kubiak, C. P. Facile Solvent-Free Synthesis of Thin Iron
Porphyrin COFs on Carbon Cloth Electrodes for CO2 Reduction. Chem. Mater. 2019, 31,
1908–1919.
17. Ley, K. D.; Whittle, C. E.; Bartberger, M. D.; Schanze, K. S. Photophysics of π-Conjugated
Polymers That Incorporate Metal to Ligand Charge Transfer Chromophores. J. Am. Chem.
Soc. 1997, 119, 3423–3424.
18. Savéant, J.-M. Molecular Catalysis of Electrochemical Reactions. Mechanistic Aspects.
Chem. Rev. 2008, 108, 2348–2378.
19. Andrieux, C. P.; Dumas-Bouchiat, J. M.; Savéant, J. M. Kinetics of Electrochemical
Reactions Mediated by Redox Polymer Films. J. Electroanal. Chem. 1984, 169, 9–21.
20. Jousselme, B.; Bidan, G.; Billon, M.; Goyer, C.; Kervella, Y.; Guillerez, S.; Hamad, E. A.;
Goze-Bac, C.; Mevellec, J. Y.; Lefrant, S. One-Step Electrochemical Modification of
Carbon Nanotubes by Ruthenium Complexes via New Diazonium Salts. J. Electroanal.
Chem. 2008, 621, 277–285.
21. Blauch, D. N.; Savéant, J. M. Dynamics of Electron Hopping in Assemblies of Redox
Centers. Percolation and Diffusion. J. Am. Chem. Soc. 1992, 114, 3323–3332.
22. Costentin, C.; Savéant, J.-M. Multielectron, Multistep Molecular Catalysis of
Electrochemical Reactions: Benchmarking of Homogeneous Catalysts. ChemElectroChem
2014, 1, 1226–1236.
23. Chapovetsky, A.; Welborn, M.; Luna, J. M.; Haiges, R.; Miller, T. F.; Marinescu, S. C.
Pendant Hydrogen-Bond Donors in Cobalt Catalysts Independently Enhance CO2
Reduction. ACS Cent. Sci. 2018, 4, 397–404.
24. Dunwell, M.; Luc, W.; Yan, Y.; Jiao, F.; Xu, B. Understanding Surface-Mediated
Electrochemical Reactions: CO2 Reduction and Beyond. ACS Catal. 2018, 8, 8121–8129.
25. Rosen, J.; Hutchings, G. S.; Lu, Q.; Rivera, S.; Zhou, Y.; Vlachos, D. G.; Jiao, F.
Mechanistic Insights into the Electrochemical Reduction of CO2 to CO on Nanostructured
Ag Surfaces. ACS Catal. 2015, 5, 4293–4299.
26. Keith, J. A.; Grice, K. A.; Kubiak, P.; Carter, E. A. Elucidation of the Selectivity of Proton-
Dependent Electrocatalytic. J. Am. Chem. Soc. 2013, 135, 15823–15829.
206

27. Riplinger, C.; Sampson, M. D.; Ritzmann, A. M.; Kubiak, C. P.; Carter, E. A. Mechanistic
Contrasts between Manganese and Rhenium Bipyridine Electrocatalysts for the Reduction
of Carbon Dioxide. J. Am. Chem. Soc. 2014, 136, 16285–16298.
28. Albrecht, M.; Janser, I.; Lützen, A.; Hapke, M.; Fröhlich, R.; Weis, P. 5,5'-Diamino-2,2'-
Bipyridine: A Versatile Building Block for the Synthesis of Bipyridine/Catechol Ligands
That Form Homo- and Heteronuclear Helicates. Chem. - A Eur. J. 2005, 11, 5742–5748.
29. Popov, D. A.; Luna, J. M.; Orchanian, N. M.; Haiges, R.; Downes, C. A.; Marinescu, S. C.
A 2,2'-Bipyridine-Containing Covalent Organic Framework Bearing Rhenium(i)
Tricarbonyl Moieties for CO2 Reduction. Dalt. Trans. 2018, 47, 17450–17460.
207

BIBLIOGRAPHY
1) U.S. Energy Information Administration, Internation Energy Outlook. 2019.
2) IEA. 2020 World Energy Outlook: Executive Summary. Oecd/Iea 2020, 11.
3) IEA. 2019 World Energy Outlook: Executive Summary. Oecd/Iea 2019, 11.
4) U.S. Energy Information Administration, Annual Energy Outlook. 2020.
5) U.S. Energy Information Administration, Short-Term Energy Outlook. 2020.
6) Keeling, Ralph F.; Keeling, Charles D. Scripps CO2 Program Data. 2017 UC San Diego
Library Digital Collections. Accessed on May 6, 2020.
7) Joos et al. Atmospheric Chemistry and Physics. 2013, 13(5), 2793-2825.
8) Turner, J. A. Science 2004, 305, 972-974.
9) Chu, S.; Majumdar, A. Nature 2012, 488, 294-303.
10) Alanne, K.; Saari, A. Renew. Sustain. Energy Rev. 2006, 10 (6), 539–558.
11) Thiam, D.-R. Renew. Energy 2010, 35 (8), 1615–1623.
12) Lacey-Barnacle, M.; Robison, R.; Foulds, C. Energy Sustain. Dev. 2020, 55, 122–138.
13) Nocera, D. G. Acc. Chem. Res. 2017, 50 (3), 616–619.
14) Graves, C.; Ebbesen, S. D.; Mogensen, M.; Lackner, K. S. Renew. Sustain. Energy Rev.
2011, 15 (1), 1–23.
15) Detz, R. J.; Reek, J. N. H.; Van Der Zwaan, B. C. C. Energy Environ. Sci. 2018, 11 (7),
1653–1669.
16) Olah, G. A.; Prakash, G. K. S.; Goeppert, A. J. Am. Chem. Soc. 2011, 133 (33), 12881–
12898.
17) Seh, Z. W.; Kibsgaard, J.; Dickens, C. F.; Chorkendorff, I.; Nørskov, J. K.; Jaramillo, T.
F. Science 2017, 355, eaad4998.
18) House, R. L.; Iha, N. Y. M.; Coppo, R. L.; Alibabaei, L.; Sherman, B. D.; Kang, P.;
Brennaman, M. K.; Hoertz, P. G.; Meyer, T. J. J. Photochem. Photobiol. C Photochem.
Rev. 2015, 25, 32–45.
208

19) Kim, D.; Sakimoto, K. K.; Hong, D.; Yang, P. Angew. Chemie Int. Ed. 2015, 54 (11), 3259–
3266.
20) Walter, M. G.; Warren, E. L.; McKone, J. R.; Boettcher, S. W.; Mi, Q.; Santori, E. A.;
Lewis, N. S. Chem. Rev. 2010, 110 (11), 6446–6473.
21) Züttel, A.; Remhof, A.; Borgschulte, A.; Friedrichs, O. Philos. Trans. R. Soc. A Math. Phys.
Eng. Sci. 2010, 368 (1923), 3329–3342.
22) Rowsell, J. L. C.; Millward, A. R.; Park, K. S.; Yaghi, O. M. J. Am. Chem. Soc. 2004, 126
(18), 5666–5667.
23) Rostrup-Nielsen, J. R. Catal. Today 1993, 18 (4), 305–324.
24) Rostrup-Nielsen, J. R. Catal. Today 2002, 71, 243–247.
25) Meyer, K.; Ranocchiari, M.; van Bokhoven, J. A. Energy Environ. Sci. 2015, 8 (7), 1923–
1937.
26) Luna, P. De; Hahn, C.; Higgins, D.; Jaffer, S. A.; Jaramillo, T. F.; Sargent, E. H. Science
2019, 364, 6438.
27) Zeng, M.; Li, Y.; Houlachi, G. J. Mater. Chem. A 2015, 3 (29), 14942–14962.
28) Fujishima, K. Honda, Nature 1972, 238, 37–38.
29) US Department of Energy's Office of Energy Efficiency & Renewable Energy.
30) Demirdoven, N.; Deutch, J.; Golden, D. M. Science. 2004, 305 (5686), 974–976.
31) Verma, S.; Kim, B.; Jhong, H.-R. M.; Ma, S.; Kenis, P. J. A. ChemSusChem 2016, 9 (15),
1972–1979.
32) Qiao, J.; Liu, Y.; Hong, F.; Zhang, J.; Koper, M. T. M. M. Chem. Soc. Rev. 2014, 43 (2),
631–675.
33) Berggren, M. 19th IMPCA Asian Methanol Conference, Singapore, 1st–3rd November,
2016
34) Olah, G. A. Angew. Chemie - Int. Ed. 2005, 44 (18), 2636–2639.
35) Olah, G.; Goeppert, A.; K. Surya Prakash, G. J. Org. Chem. 2008, 74 (2), 487–498.
36) Surampudi, S.; Narayanan, S. R. R.; Vamos, E.; Frank, H.; Halpert, G.; LaConti, A.; Kosek,
J.; Prakash, G. K. S.; Olah, G. A. A. J. Power Sources 1994, 47 (3), 377–385.
209

37) Olah, G. A. Catalysis Letters, 2004 93(1-2), 1-2.
38) Alberico, E.; Nielsen, M. Chem. Commun. 2015, 51 (31), 6714–6725.
39) Takeda, H.; Ishitani, O. Coord. Chem. Rev. 2010, 254 (3–4), 346–354.
40) Barlow, J. M.; Yang, J. Y. ACS Cent. Sci. 2019, 5 (4), 580–588.
41) Artero, V.; Saveant, J. M. Energy Environ. Sci. 2014, 7 (11), 3808–3814.
42) Costentin, C.; Savéant, J.-M. ChemElectroChem 2014, 1 (7), 1226–1236.
43) Costentin, C.; Passard, G.; Savéant, J. M. J. Am. Chem. Soc. 2015, 137 (16), 5461–5467.
44) Costentin, C.; Savéant, J.-M. J. Am. Chem. Soc. 2017, 139 (24), 8245–8250.
45) Bligaard, T.; Morris Bullock, R.; T. Campbell, C.; G. Chen, J.; C. Gates, B.; J. Gorte, R.;
W. Jones, C.; D. Jones, W.; R. Kitchin, J.; L. Scott, S. ACS Catal. 2016, 6 (4), 2590–2602.
46) U.S. Department of the Interior, U.S. Geological Survey—USGS Fact Sheet. 2002.
47) Morales-Guio, C. G.; Stern, L.-A.; Hu, X. Chem. Soc. Rev. 2014, 43, 6555-6569.
48) Kortlever, R.; Shen, J.; Schouten, K. J. P.; Calle-Vallejo, F.; Koper, M. T. M. J. Phys.
Chem. Lett. 2015, 6 (20), 4073–4082.
49) Su, B.; Cao, Z.-C.; Shi, Z.-J. Acc. Chem. Res. 2015, 48 (3), 886–896.
50) Bairagya, M. Das; Bujol, R. J.; Elgrishi, N. Chem. - A Eur. J. 2020, 1–11.
51) Artero, V. Nat. Energy 2017, 2, 17131.
52) Wang, H. Chem. Soc. Rev. 2013, 42, 8360–8375.
53) Holm, R. H.; Kennepohl, P.; Solomon, E. I. Chem. Rev. 1996, 96, 2239−2314.
54) Appel, A. M.; Bercaw, J. E.; Bocarsly, A. B.; Dobbek, H.; Dubois, D. L.; Dupuis, M.;
Ferry, J. G.; Fujita, E.; Hille, R.; Kenis, P. J. A.; Kerfeld, C. A.; Morris, R. H.; Peden, C.
H. F.; Portis, A. R.; Ragsdale, S. W.; Rauchfuss, T. B.; Reek, J. N. H.; Seefeldt, L. C.;
Thauer, R. K.; Waldrop, G. L. Chem. Rev. 2013, 113 (8), 6621–6658.
55) Beinert, H. JBIC 2000, 5, 2-15.
56) Ogata, H.; Nishikawa, K.; Lubitz, W. Nature 2015, 520 (7548), 571–574.
57) Armor, J. N. A Catal. Today 2011, 163 (1), 3–9.
210

58) Zhou, Z. Y.; Tian, N.; Li, J. T.; Broadwell, I.; Sun, S. G. Chem. Soc. Rev. 2011, 40 (7),
4167–4185.
59) Mizuno, N.; Misono, M. Chem. Rev. 1998, 98 (1), 199–217.
60) Thoi, V. S.; Sun, Y.; Long, J. R.; Chang, C. J. Chem. Soc. Rev. 2013, 42, 2388-2400.
61) Morris, A. J.; Meyer, G. J.; Fujita, E. Acc. Chem. Res. 2009, 42 (12), 1983–1994.
62) Costentin, C.; Saveant, J.-M. J. Phys. Chem. C 2015, 119, 22, 12174-12182.
63) Bullock, R. M.; Das, A. K.; Appel, A. M. Chem. - A Eur. J. 2017, 23 (32), 7626–7641.
64) Louis, M. E.; Fenton, T. G.; Rondeau, J.; Jin, T.; Li, G. Comments Inorg. Chem. 2016, 36
(1), 38–60.
65) Love, C. J.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Chem. Rev.
2005, 105, 1103−1169
66) Ulman, A. Chem. Rev. 1996, 96, 1533–1554.
67) Meyer, T. J.; Sheridan, M. V.; Sherman, B. D. Chem. Soc. Rev. 2017, 46 (20), 6148–6169.
68) Materna, K. L.; Crabtree, R. H.; Brudvig, G. W. Chem. Soc. Rev. 2017, 46 (20), 6099–
6110.
69) Mahouche-Chergui, S.; Gam-Derouich, S.; Mangeney, C.; Chehimi, M. M. Chem. Soc.
Rev. 2011, 40 (7), 4143–4166.
70) Blakemore, J. D.; Gupta, A.; Warren, J. J.; Brunschwig, B. S.; Gray, H. B. J. Am. Chem.
Soc. 2013, 135 (49), 18288–18291.
71) Bowers, C. M.; Rappoport, D.; Baghbanzadeh, M.; Simeone, F. C.; Liao, K. C.; Semenov,
S. N.; Zaba, T.; Cyganik, P.; Aspuru-Guzik, A.; Whitesides, G. M. J. Phys. Chem. C 2016,
120 (21), 11331–11337.
72) Chen, F.; Li, X.; Hihath, J.; Huang, Z.; Tao, N. J. Am. Chem. Soc. 2006, 128 (49), 15874–
15881.
73) Furukawa, H.; Cordova, K. E.; O’Keeffe, M.; Yaghi, O. M. Science 2013, 341 (6149),
1230444–1230444.
74) Downes, C. A.; Marinescu, S. C. ChemSusChem, 2017, 10, 4374–4392
75) O’Toole, T. R.; Margerum, L. D.; Westmoreland, T. D.; Vining, W. J.; Murray, R. W.;
Meyer, T. J. J. Chem. Soc. Chem. Commun. 1985, 20, 1416.
211

76) Andrieux, C. P.; Dumas-Bouchiat, J. M.; Saveant, J. M. J. Electroanal. Chem. 1980, 114
(1), 159–163.
77) Wang, J. L.; Wang, C.; Lin, W. ACS Catal. 2012, 2 (12), 2630–2640.
78) Wang, S.; Wang, X. Small 2015, 11 (26), 3097–3112.
79) Drake, T.; Ji, P.; Lin, W. Acc. Chem. Res. 2018, 51 (9), 2129–2138.
80) Wang, M.; Chen, L.; Sun, L. Energy Environ. Sci. 2012, 5 (5), 6763–6778.
81) George, G. N.; Costa, C.; Moura, J. J. G.; Moura, I. J. Am. Chem. Soc. 1999, 121 (11),
2625–2626.
82) Luca, O. R.; Crabtree, R. H. Chem. Soc. Rev. 2013, 42 (4), 1440–1459.
83) Arévalo, R.; Chirik, P. J. J. Am. Chem. Soc. 2019, 141, 23, 9106-9123.
84) Fogeron, T.; Todorova, T. K.; Porcher, J. P.; Gomez-Mingot, M.; Chamoreau, L. M.;
Mellot-Draznieks, C.; Li, Y.; Fontecave, M. A ACS Catal. 2018, 8 (3), 2030–2038.
85) Letko, C. S.; Panetier, J. A.; Head-Gordon, M.; Tilley, T. D. J. Am. Chem. Soc. 2014, 136
(26), 9364–9376.
86) Lee, K. J.; McCarthy, B. D.; Rountree, E. S.; Dempsey, J. L Inorg. Chem. 2017, 56 (4),
1988–1998.
87) Eisenberg, R.; Gray, H. B. Inorg. Chem. 2011, 50 (20), 9741–9751.
88) McNamara, W. R.; Han, Z.; Yin, C.-J.; Brennessel, W. W.; Holland, P. L.; Eisenberg, R.
Proc. Natl. Acad. Sci. 2012, 109 (39), 15594–15599.
89) Solis, B. H.; Hammes-Schiffer, S. J. Am. Chem. Soc. 2012, 134 (37), 15253–15256.
90) Smieja, J. M.; Kubiak, C. P. Inorg. Chem. 2010, 49 (20), 9283–9289.
91) Kumar, B.; Smieja, J. M.; Kubiak, C. P. J. Phys. Chem. C 2010, 114 (33), 14220–14223.
92) Clark, M. L.; Cheung, P. L.; Lessio, M.; Carter, E. A.; Kubiak, C. P. ACS Catal. 2018, 8
(3), 2021–2029.
93) Hawecker, J.; Lehn, J.-M.; Ziessel, R. J. Chem. Soc., Chem. Commun. 1984, 984 (6), 328–
330.
94) Hawecker, J.; Lehn, J.-M.; Ziessel, R. J. Chem. Soc., Chem. Commun. 1983, No. 9, 536–
538.
212

95) Smieja, J. M.; Benson, E. E.; Kumar, B.; Grice, K. A.; Seu, C. S.; Miller, A. J. M.; Mayer,
J. M.; Kubiak, C. P Proc. Natl. Acad. Sci. U.S.A. 2012, 109 (39), 15646–15650.
96) Keith, J. A.; Grice, K. A.; Kubiak, P.; Carter, E. A. J. Am. Chem. Soc. 2013, 135, 15823–
15829.
97) IEA. 2018 World Energy Outlook: Executive Summary. Oecd/Iea 2018, 11.
98) U.S. Energy Information Administration – EIA. “New Methanol Plants Expected to
Increase Industrial Natural Gas Use through 2020” 2019, https://www.eia.gov/.
99) Collin, J. P.; Sauvage, J. P. Coord. Chem. Rev. 1989, 93, 245–268.
100) Windle, C. D.; Perutz, R. N. Coord. Chem. Rev. 2012, 256, 2562–2570.
101) Francke, R.; Schille, B.; Roemelt, M. Chem. Rev. 2018, 118, 4631–4701.
102) Zhou, X.; Micheroni, D.; Lin, Z.; Poon, C.; Li, Z.; Lin, W. ACS Appl. Mater. Interfaces
2016, 8, 4192–4198.
103) Kumar, B.; Smieja, J. M.; Sasayama, A. F.; Kubiak, C. P. Chem. Commun. 2012, 48, 272–
274.
104) Wang, J. W.; Huang, H. H.; Sun, J. K.; Zhong, D. C.; Lu, T. B. ACS Catal. 2018, 8, 7612–
7620.
105) Elgrishi, N.; Chambers, M. B.; Artero, V.; Fontecave, M. Phys. Chem. Chem. Phys. 2014,
16, 13635–13644.
106) Froehlich, J. D.; Kubiak, C. P. Inorg. Chem. 2012, 51, 3932–3934.
107) Chen, Z.; Concepcion, J. J.; Brennaman, M. K.; Kang, P.; Norris, M. R.; Hoertz, P. G.;
Meyer, T. J. Proc. Natl. Acad. Sci. 2012, 109, 15606–15611.
108) Kang, P.; Chen, Z.; Nayak, A.; Zhang, S.; Meyer, T. J. Energy Environ. Sci. 2014, 7, 4007–
4012.
109) Chen, Z.; Kang, P.; Zhang, M. T.; Meyer, T. J. Chem. Commun. 2014, 50, 335–337.
110) Chapovetsky, A.; Do, T. H.; Haiges, R.; Takase, M. K.; Marinescu, S. C. J. Am. Chem.
Soc. 2016, 138, 5765–5768.
111) Nichols, E. M.; Derrick, J. S.; Nistanaki, S. K.; Smith, P. T.; Chang, C. J. Chem. Sci. 2018,
9, 2952–2960.
213

112) Downes, C. A.; Yoo, J. W.; Orchanian, N. M.; Haiges, R.; Marinescu, S. C. H2 Chem.
Commun. 2017, 53, 7306–7309.
113) Jeletic, M. S.; Helm, M. L.; Hulley, E. B.; Mock, M. T.; Appel, A. M.; Linehan, J. C. ACS
Catal. 2014, 4, 3755–3762.
114) Dey, S.; Ahmed, M. E.; Dey, A. Inorg. Chem. 2018, 57, 5939–5947.
115) Jain, R.; Mashuta, M. S.; Buchanan, R. M.; Grapperhaus, C. A. Eur. J. Inorg. Chem. 2017,
2017, 3714–3719.
116) Haddad, A. Z.; Kumar, D.; Ouch Sampson, K.; Matzner, A. M.; Mashuta, M. S.;
Grapperhaus, C. A. J. Am. Chem. Soc. 2015, 137, 9238–9241.
117) Barry, B. M.; Stein, B. W.; Larsen, C. A.; Wirtz, M. N.; Geiger, W. E.; Waterman, R.;
Kemp, R. A. Inorg. Chem. 2013, 52, 9875–9884.
118) Chapovetsky, A.; Welborn, M.; Luna, J. M.; Haiges, R.; Miller, T. F.; Marinescu, S. C.
ACS Cent. Sci. 2018, 4, 397–404.
119) Russo, T. V.; Martin, R. L.; Hay, P. J. J. Chem. Phys. 1995, 102, 8023–8028.
120) Li, Z.; Liu, W. J. Chem. Theory Comput. 2016, 12, 238–260.
121) Costentin, C.; Robert, M.; Savéant, J. M. Chem. Soc. Rev. 2013, 42, 2423–2436.
122) McCarthy, B. D.; Donley, C. L.; Dempsey, J. L. Chem. Sci. 2015, 6, 2827–2834.
123) Lam, Y. C.; Nielsen, R. J.; Gray, H. B.; Goddard III, W. A. ACS Catal. 2015, 5, 2521-
2528.
124) Kütt, A.; Leito, I.; Kaljurand, I.; Sooväli, L.; Vlasov, V. M.; Yagupolskii, L. M.; Koppel,
I. A. J. Org. Chem. 2006, 71, 2829-2838.
125) Pegis, M. L.; Roberts, J. A.; Wasylenko, D. J.; Mader, E. A.; Appel, A. M.; Mayer, J. M.
Inorg. Chem. 2015, 54, 11883−11888.

214

126) Shao, Y.; Molnar, L.; Jung, Y.; Kussmann, J.; Ochsenfeld, C.; Brown, S.; Gilbert, A.;
Slipchenko, L.; Levchenko, S.; O'Neill, D.; DiStasio, R.; Lochan, R.; Wang, T.; Beran, G.;
Besley, N.; Herbert, J.; Lin, C.; Van Voorhis, T.; Chien, S.; Sodt, A.; Steele, R.; Rassolov,
V.; Maslen, P.; Korambath, P.; Adamson, R.; Austin, B.; Baker, J.; Byrd, E.; Dachsel, H.;
Doerksen, R.; Dreuw, A.; Dunietz, B.; Dutoi, A.; Furlani, T.; Gwaltney, S.; Heyden, A.;
Hirata, S.; Hsu, C.; Kedziora, G.; Khalliulin, R.; Klunzinger, P.; Lee, A.; Lee, M.; Liang,
W.; Lotan, I.; Nair, N.; Peters, B.; Proynov, E.; Pieniazek, P.; Rhee, Y.; Ritchie, J.; Rosta,
E.; Sherrill, C.; Simmonett, A.; Subotnik, J.; Woodcock, H.; Zhang, W.; Bell, A.;
Chakraborty, A.; Chipman, D.; Keil, F.; Warshel, A.; Hehre, W.; Schaefer, H.; Kong, J.;
Krylov, A.; Gill, P.; Head-Gordon, M. Phys. Chem. Chem. Phys. 2006, 8, 3172-3191.
127) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996, 77, 3865–3868.
128) Ditchfield, R.; Hehre, W. J.; Pople, J. A. J. Chem. Phys. 1971, 54, 724–728.
129) Hehre, W. J.; Ditchfield, K.; Pople, J. A. J. Chem. Phys. 1972, 56, 2257–2261.
130) Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005, 7, 3297.
131) Feller, D. Comput. Chem. 1996, 17, 1571-1586.
132) Schuchardt, K. L.; Didier, B. T.; Elsethagen, T; Sun, L.; Gurumoorthi, V.; Chase, J.; Li, J.;
Windus, T. L. J. Chem. Inf. Model. 2007, 47, 1045-1052.
133) Sheldrick, G. M. Acta Cryst. 1990, A46, 467.
134) Sheldrick, G. M. Acta Cryst. 2008, A64, 112.
135) Müller, P. Cryst. Rev. 2009, 15, 57.
136) Benson, E. E.; Kubiak, C. P.; Sathrum, A. J.; Smieja, J. M. Chem. Soc. Rev. 2009, 38 (1),
89–99.
137) Costentin, C.; Robert, M.; Savéant, J. M. Acc. Chem. Res. 2015, 48 (12), 2996–3006.
138) Artz, J.; Müller, T. E.; Thenert, K.; Kleinekorte, J.; Meys, R.; Sternberg, A.; Bardow, A.;
Leitner, W. Chem. Rev. 2018, 118 (2), 434–504.
139) Jeoung, J.-H.; Dobbek, H. 2010, 330, 1461–1465.
140) Chen, L.; Guo, Z.; Wei, X. G.; Gallenkamp, C.; Bonin, J.; Anxolabéhère-Mallart, E.; Lau,
K. C.; Lau, T. C.; Robert, M. J. Am. Chem. Soc. 2015, 137 (34), 10918–10921.
141) Guo, Z.; Cheng, S.; Cometto, C.; Anxolabéhère-Mallart, E.; Ng, S.-M.; Ko, C.-C.; Liu, G.;
Chen, L.; Robert, M.; Lau, T.-C. J. Am. Chem. Soc. 2016, 138 (30), 9413–9416.
215

142) Cometto, C.; Chen, L.; Lo, P. K.; Guo, Z.; Lau, K. C.; Anxolabéhère-Mallart, E.; Fave, C.;
Lau, T. C.; Robert, M. ACS Catal. 2018, 8 (4), 3411–3417.
143) Zhang, E. X.; Wang, D. X.; Huang, Z. T.; Wang, M. X. J. Org. Chem. 2009, 74 (22), 8595–
8603.
144) Pun, S.; Chung, W.; Lam, K.; Guo, P.; Chan, P.; Wong, K.; Che, C.; Chen, T.; Peng, S. J.
Chem. Soc., Dalton Trans. 2002, 575–583.
145) Piguet, C. J. Chem. Ed. 1997, 74 (7), 815–816.
146) Mealli, C.; Hoffman, R.; Stockis, A. Inorg. Chem. 1984, 23, 65-74.
147) Saveant, J.-M. Chem. Rev. 2008, 108, 2348–2378.
148) Grosvenor, A. P.; Kobe, B. A.; Biesinger, M. C.; McIntyre, N. S. Surf. Interface Anal.
2004, 36, 1564-1574.
149) Grosvenor, A. P.; Biesinger, M. C.; Smart, R. St.C.; McIntyre, N. S. Surface Science 2006,
600, 1771–1779.
150) Gray, H. B. Nat. Chem. 2009, 1, 7.
151) Lewis, N. S. Science 2016, 351 (6271), aad1920.
152) Schreier, M.; Luo, J.; Gao, P.; Moehl, T.; Mayer, M. T.; Grätzel, M. J. Am. Chem. Soc.
2016, 138 (6), 1938–1946.
153) Oh, S.; Gallagher, J. R.; Miller, J. T.; Surendranath, Y. J. Am. Chem. Soc. 2016, 138 (6),
1820–1823.
154) Friebe, C.; Hager, M. D.; Winter, A.; Schubert, U. S. Adv. Mater. 2012, 24 (3), 332–345.
155) Cabrera, C.; Abruña, H. D. J. Electroanal. Chem. 1986, 209, 101–107.
156) O’Toole, T. R.; Sullivan, B. P.; Bruce, M. R.-M.; Margerum, L. D.; Murray, R. W.; Meyer,
T. J. J. Electroanal. Chem. Interfacial Electrochem. 1989, 259 (1–2), 217–239.
157) Kamata, R.; Kumagai, H.; Yamazaki, Y.; Sahara, G.; Ishitani, O. ACS Applied Materials
& Interfaces 2018, Article ASAP, DOI: 10.1021/acsami.8b05495.
158) Chan, W. K. Chem. Rev. 2007, 251 (17–20), 2104–2118.
159) Maruyama, T.; Yamamoto, T. J. Phys. Chem. B 1997, 101 (19), 3806–3810.
216

160) Yamamoto, T.; Maruyama, T.; Zhou, Z. H.; Ito, T.; Fukuda, T.; Yoneda, Y.; Begum, F.;
Ikeda, T.; Sasaki, S.; Takezoe, H.; Fukuda, A.; Kubota, K. J. Am. Chem. Soc. 1994, 116
(11), 4832–4845.
161) Hayashida, N.; Yamamoto, T. Bulletin of the Chemical Society of Japan. 1999, 72 (5),
1153–1162.
162) Ley, K. D.; Whittle, C. E.; Bartberger, M. D.; Schanze, K. S. J. Am. Chem. Soc. 1997, 119
(14), 3423–3424.
163) Portenkirchner, E.; Gasiorowski, J.; Oppelt, K.; Schlager, S.; Schwarzinger, C.;
Neugebauer, H.; Knör, G.; Sariciftci, N. S. ChemCatChem 2013, 5 (7), 1790–1796.
164) Bangle, R.; Sampaio, R. N.; Troian-Gautier, L.; Meyer, G. J. ACS Appl. Mater. Interfaces
2018, 10 (3), 3121–3132.
165) Marshall, N.; Locklin, J. Langmuir 2011, 27 (21), 13367–13373.
166) Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120 (1–3), 215–241.
167) Wadt, W. R.; Hay, P. J. J. Chem. Phys. 1985, 82 (1), 284–298.
168) Kariuki, J. K.; McDermott, M. T. Langmuir 2001, 17 (19), 5947–5951.
169) Jousselme, B.; Bidan, G.; Billon, M.; Goyer, C.; Kervella, Y.; Guillerez, S.; Hamad, E. A.;
Goze-Bac, C.; Mevellec, J. Y.; Lefrant, S. J. Electroanal. Chem. 2008, 621 (2), 277–285.
170) Ceccato, M.; Bousquet, A.; Hinge, M.; Pedersen, S. U.; Daasbjerg, K. Chem. Mater. 2011,
23 (6), 1551–1557.
171) Machan, C. W.; Sampson, M. D.; Chabolla, S. A.; Dang, T.; Kubiak, C. P. Organometallics
2014, 33 (18), 4550–4559.
172) Permatasari, F. A.; Aimon, A. H.; Iskandar, F.; Ogi, T.; Okuyama, K. Role of C-N Sci.
Rep. 2016, 6, 1–8.
173) Ricks, A. M.; Anfuso, C. L.; Rodríguez-Córdoba, W.; Lian, T. Chem. Phys. 2013, 422,
264–271.
174) Kurtz, D. A.; Brereton, K. R.; Ruo, K. P.; Tang, H. M.; Felton, G. A. N.; Miller, A. J. M.;
Dempsey, J. L. Inorg. Chem. 2018, 57 (9), 5389–5399.
175) Walters, K. A.; Premvardhan, L. L.; Liu, Y.; Peteanu, L. A.; Schanze, K. S. Chem. Phys.
Lett. 2001, 339, 255–262.
217

176) Hossain, M. M.; Ivanov, M. V.; Wang, D.; Reid, S. A.; Rathore, R. J. Phys. Chem. C 2018,
112 (31), 17688–17675.
177) Savéant, J. M. J. Phys. Chem. 1988, 92 (5), 1011–1013.
178) Denisevich, P.; Abruña, H. D.; Leidner, C. R.; Meyer, T. J.; Murray, R. W. Inorg. Chem.
1982, 21 (6), 2153–2161.
179) Miyamae, T.; Yoshimura, D.; Ishii, H.; Ouchi, Y.; Seki, K.; Miyazaki, T.; Koike, T.;
Yamamoto, T. J. Chem. Phys. 1995, 103 (7), 2738–2744.
180) Kramer, W. W.; McCrory, C. C. L. Chem. Sci. 2016, 7 (4), 2506–2515.
181) Pognon, G.; Brousse, T.; Bélanger, D. Carbon 2011, 49 (4), 1340–1348.
182) Gatty, M. G.; Pullen, S.; Sheibani, E.; Tian, H.; Ott, S.; Hammarström, L. Chem. Sci. 2018,
9 (22), 4983–4991.
183) Pekarek, R. T.; Kearney, K.; Simon, B. M.; Ertekin, E.; Rockett, A. A.; Rose, M. J. J. Am.
Chem. Soc. 2018. Article ASAP, DOI: 10.1021/jacs.8b05057.
184) Habisreutinger, S. N.; Schmidt-Mende, L.; Stolarczyk, J. K. Angew. Chemie - Int. Ed. 2013,
52 (29), 7372–7408.
185) Staniszewski, A.; Morris, A. J.; Ito, T.; Meyer, G. J. J. Phys. Chem. B 2007, 111 (24),
6822–6828.
186) Morimoto, T.; Nakajima, T.; Sawa, S.; Nakanishi, R.; Imori, D.; Ishitani, O. J. Am. Chem.
Soc. 2013, 135 (45), 16825–16828.
187) Abdellah, M.; El-Zohry, A. M.; Antila, L. J.; Windle, C. D.; Reisner, E.; Hammarström, L.
J. Am. Chem. Soc. 2017, 139 (3), 1226–1232.
188) Windle, C. D.; Pastor, E.; Reynal, A.; Whitwood, A. C.; Vaynzof, Y.; Durrant, J. R.; Perutz,
R. N.; Reisner, E. Chem. - A Eur. J. 2015, 21 (9), 3746–3754.
189) Takeda, H.; Koike, K.; Inoue, H.; Ishitani, O. J. Am. Chem. Soc. 2008, 130 (6), 2023–2031.
190) Kou, Y.; Nabetani, Y.; Masui, D.; Shimada, T.; Takagi, S.; Tachibana, H.; Inoue, H. J. Am.
Chem. Soc. 2014, 136 (136), 6021–6030.
191) Shinde, D. B.; Aiyappa, H. B.; Bhadra, M.; Biswal, B. P.; Wadge, P.; Kandambeth, S.;
Garai, B.; Kundu, T.; Kurungot, S.; Banerjee, R. J. Mater. Chem. A 2016, 4 (7), 2682–
2690.
218

192) Savéant, J.-M. Elements of Molecular and Biomolecular Electrochemistry; Wiley-
Interscience: New York, NY, 2006.
193) Rountree, E. S.; McCarthy, B. D.; Eisenhart, T. T.; Dempsey, J. L. Inorg. Chem. 2014, 53
(19), 9983–10002.
194) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82 (1), 270–283.
195) Fu, R.; Chung, D.; Lowder, T.; Feldman, D.; Ardani, K.; Fu, R.; Chung, D.; Lowder, T.;
Feldman, D.; Ardani, K. “U.S. Solar Photovoltaic System Cost Benchmark: Q1 2017”
NREL 2017, No. September, 1–66.
196) Schmidt, O.; Hawkes, A.; Gambhir, A.; Staffell, I. Nat. Energy 2017, 2, 1–8.
197) Gielen, D.; Boshell, F.; Saygin, D. Nat. Mater. 2016, 15, 117–120.
198) Schreckenghost, D.; Thronesbery, C.; Bonasso, P.; Kortenkamp, D.; Martin, C. IEEE
Intell. Syst. 2002, 17, 24–31.
199) Breedlove, B. K.; Ferrence, G. M.; Washington, J.; Kubiak, C. P. Matt. Design, 2001, 22,
577-584.
200) Orchanian, N. M.; Hong, L. E.; Skrainka, J. A.; Esterhuizen, J. A.; Popov, D. A.;
Marinescu, S. C. ACS Appl. Energy Mater. 2018, 2, 110–123.
201) Jackson, M. N.; Oh, S.; Kaminsky, C. J.; Chu, S. B.; Zhang, G.; Miller, J. T.; Surendranath,
Y. J. Am. Chem. Soc. 2018, 140, 1004–1010.
202) Rotundo, L.; Filippi, J.; Gobetto, R.; Miller, H. A.; Rocca, R.; Nervi, C.; Vizza, F. Chem.
Commun. 2019, 55, 775–777.
203) Cheung, P. L.; Lee, S. K.; Kubiak, C. P. Chem. Mater. 2019, 31, 1908–1919.
204) Savéant, J.-M. Chem. Rev. 2008, 108, 2348–2378.
205) Andrieux, C. P.; Dumas-Bouchiat, J. M.; Savéant, J. M. J. Electroanal. Chem. 1984, 169,
9–21.
206) Blauch, D. N.; Savéant, J. M. J. Am. Chem. Soc. 1992, 114, 3323–3332.
207) Dunwell, M.; Luc, W.; Yan, Y.; Jiao, F.; Xu, B. ACS Catal. 2018, 8, 8121–8129.
208) Rosen, J.; Hutchings, G. S.; Lu, Q.; Rivera, S.; Zhou, Y.; Vlachos, D. G.; Jiao, F. ACS
Catal. 2015, 5, 4293–4299.
219

209) Riplinger, C.; Sampson, M. D.; Ritzmann, A. M.; Kubiak, C. P.; Carter, E. A. J. Am. Chem.
Soc. 2014, 136, 16285–16298.
210) Albrecht, M.; Janser, I.; Lützen, A.; Hapke, M.; Fröhlich, R.; Weis, P. Chem. - A Eur. J.
2005, 11, 5742–5748.
211) Popov, D. A.; Luna, J. M.; Orchanian, N. M.; Haiges, R.; Downes, C. A.; Marinescu, S. C.
Dalt. Trans. 2018, 47, 17450–17460.
212) Orchanian, N. M.; Hong, L. E.; Marinescu, S. C. ACS Catal., 2019, 9, 10, 9393-9397.
213) USC Graduate School. Thesis and Dissertation Guidelines. 2020 University of Southern
California Library Digital Collections. Accessed on May 6, 2020.
https://graduateschool.usc.edu/current-students/thesis-dissertation-submission/guidelines-
for-format-and-presentation/

Abstract (if available)

Abstract The utilization of carbon dioxide as a feedstock chemical has the potential to revolutionize the anthropogenic carbon cycle and enable a carbon-neutral chemical infrastructure. Just as biological systems have developed photosynthetic cycles capable of producing chemical products from atmospheric carbon dioxide, synthetic molecular systems have been developed which can upgrade carbon dioxide to industrial chemical feedstocks. Particularly effective systems for the electrocatalytic conversion of carbon dioxide are coordination complexes which incorporate transition metal ions, as these systems can selectively bind carbon dioxide over protons thus inhibiting competitive hydrogen evolution. While these molecular systems exhibit unique reactivity, practical realization of grid-scale electrolysis requires the development of heterogeneous systems for reduced catalyst separations cost and improved catalyst recyclability. I therefore explored systematic studies to investigate new molecular catalyst targets and subsequently developed methodologies for the immobilization of molecular systems to electrode surfaces, ultimately resulting in a novel electropolymerization route for the surface-assembly of one-dimensional molecular wires.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Electrocatalytic thiolate- and selenolate-based coordination polymers for solar energy conversion

PDF

Design of catalysts for the transformation of abundant small molecules into solar fuels

PDF

Design, synthesis, and study of polypyridine based molecular and heterogenized molecular electrocatalysts for CO₂ reduction

PDF

Two-dimensional metal dithiolene metal-organic frameworks as conductive materials for solar energy conversion

PDF

Cobalt phosphino-thiolate complexes: applications towards electrocatalytic small molecule conversion to synthetic fuels

PDF

Plasmonic enhancement of catalysis and solar energy conversion

PDF

Rhenium bipyridine catalysts with pendant amines: substituent and positional effects on the electrocatalytic reduction of CO₂ to CO

PDF

Ultrafast dynamics of excited state intra- and intermolecular proton transfer in nitrogen containing photobases

PDF

Two-coordinate coinage metal complexes for solar fuels and organic LED chemistry

PDF

Adiabatic and non-adiabatic molecular dynamics in nanoscale systems: theory and applications

PDF

Design, synthesis, and investigation of a bio inspired CO₂ reduction catalyst

PDF

Exploring new frontiers in catalysis: correlating crystal chemistry and activity in layered silicates

PDF

Vibrational spectroscopy and molecular orientation at interfaces: correlating the structure of a material's surface to its properties

PDF

Conversion of carbon dioxide via electrochemical method using a trimetallic cobalt complex

PDF

Studying iron and nickel analogues of an efficient carbon dioxide reduction electrocatalyst

PDF

Probing charge transfer and electric fields at the bulk and interface using vibrational spectroscopy

PDF

Synthesis and characterization of novel donor-acceptor systems for solar water splitting

PDF

Dithiolate-based metal-organic frameworks for electrocatalytic hydrogen evolution

PDF

Molecular and polymeric donors for bulk heterojunction organic solar cells

PDF

The adsorption and selective deposition of molecular and nanocluster ions on carbon based devices

Asset Metadata

Creator Orchanian, Nicholas Melkon (author)

Core Title Catalysis for solar fuels production: molecular and materials approaches

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Chemistry

Publication Date 05/07/2020

Defense Date 05/06/2020

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag catalysis,coordination chemistry,coordination polymers,functional materials,green chemistry,inorganic chemistry,OAI-PMH Harvest,renewable energy,solar energy conversion,solar fuels,sustainability

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Marinescu, Smaranda C. (committee chair), Dawlaty, Jahan M. (committee member), El-Naggar, Moh Y. (committee member)

Creator Email norchanian@gmail.com,orchania@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-300574

Unique identifier UC11664325

Identifier etd-OrchanianN-8453.pdf (filename),usctheses-c89-300574 (legacy record id)

Legacy Identifier etd-OrchanianN-8453.pdf

Dmrecord 300574

Document Type Dissertation

Rights Orchanian, Nicholas Melkon

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA