Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Dithiolate-based metal-organic frameworks for electrocatalytic hydrogen evolution
(USC Thesis Other)
Dithiolate-based metal-organic frameworks for electrocatalytic hydrogen evolution
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
DITHIOLATE-BASED METAL-ORGANIC FRAMEWORKS FOR
ELECTROCATALYTIC HYDROGEN EVOLUTION
by
Keying Chen
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)
December 2021
Copyright 2021 Keying Chen
ii
Dedication
For Mom, Dad, and Grandma
iii
Acknowledgements
It was the late night of August 5
th
, 2016, when my 14-hour flight from Hong Kong to Los
Angeles was finally about to descend. While amazed at the vast City of Angels with her beautiful
night lights, I felt my heart racing: What exactly did I sign myself up for? Now, almost exact five
years later, standing at the end of my graduate career, I am so glad that I made the decision to
pursue my Ph.D. at USC, where I achieved so many things that I am truly proud of. Of course,
none of these would have been possible without all the amazing people that surround and support
me continuously throughout this journey.
I want to first thank my advisor, Prof. Smaranda Marinescu, for her guidance and mentorship,
as well as her support both in and outside of the lab over the past five years. Without her, I would
not be where I am right now. I also want to thank her for being a role model and a constant source
of inspiration, as an established female scientist with an international background and a proud
mother of three. I also want to acknowledge the rest of my committee, Prof. Richard Brutchey,
Prof. Jayakanth Ravichandran, Prof. Brent Melot, and Prof. Sri Narayan. Thank you for being part
of my Ph.D. journey and for all the support and advice you provided along the way.
Next, I want to say a big thank you to all the lab mates I have had in the Marinescu group. I
am so lucky to have you all by my side to navigate through graduate school together. Courtney,
thank you for being the best graduate mentor. From electrochemistry, to manuscript writing, and
to job searching, literally every step I took in graduate school, you were there to lend me a hand. I
cannot imagine where I would be without you, and I do wish someday I could return the favor. To
Andrew, thank you for laying the groundwork for my research project and teaching me all the
characterization techniques using your Cemma superuser privilege. It was fun to watch hockey
and eat Domino’s with you during those 18-hour reaction days. Alon, thank you for being a
iv
constant source of laughter and delicious Twix. Damir, thank you for your kindness and all your
help when I first joined the lab. It was a lot of fun to explore Koreatown with you and Geo after a
long day of work. Eric, thank you for being a responsible lab dad and always looking out for every
of us, even after you graduated. Nick, thank you for all the Valentine’s Day candies, all the lab
coordination, and being my life guard during those tert-BuLi reactions. Ashley, I am so glad we
got much closer during COVID (while we were not supposed to). It was nice to have someone to
vent to and to share gossip with, thank you for being there for me. Jeremy, I officially acknowledge
you as my first Jewish friend, and I want you to know that I appreciate all the cultural insights you
shared with me. Come to me for Chinese food next Christmas. Jeff, I constantly got blown away
by how kind you are and how much you know. Thank you for all the pro chemistry tips, career
advice, and Starbucks trips. I am looking forward to seeing you soon again in Boston. David,
Adam, and Jeremiah, thank you for choosing to be part of the Marinescrew and for your kind and
lively spirits. I am so glad to be here with you in your early graduate journey, and I believe in you
to have a successful graduate career.
I also want to acknowledge the loving community of the USC Chemistry Department. I want
to thank Prof. Surya Prakash, the chair of the department, for being so inspirational and supportive,
especially during the difficult pandemic times. Special thank you to Michele, for answering my
emails almost every other day over the past five years. From purchasing order, to paycheck, and
to fellowship, I certainly would not be able to get anything done without your help. Magnolia,
thank you for sorting out all the international student issues for me and withstanding me sending
you work messages on Instagram. I also want to give a big thank you to Dr. Jessica Parr for
mentoring me as a Chemical Education Fellow, and letting me participate in her general chemistry
v
lecture. Your passion for teaching is so inspiring and I truly appreciate all the career advice you
shared with me.
Over the past five years, I have made precious bonds with so many friends, who constantly
provide me with relentless support and immense joy. Geo, the first friend I ever made in the U.S.,
I still remember the days when we would take silly selfies in lab and during seminars together, or
stay late in LJS to solve an insane amount of NMR and organometallic homework together. Thanks
for appreciating me and my cultural background and introducing me to a world filled with novelty,
from language, to music, to game, to food, and to so much more. Carlos, you are one of the kindest
people I have ever met in my life, and I am so lucky to have you as a close friend. Thanks for all
the company (and rides) to those countless adventures we have been through together, and for
always being there for me when I needed emotional support. To Aaron, you are so smart, creative,
enthusiastic, and genuine; being friends with you makes me feel proud of myself. I am really glad
that we were able to get on some adventures together at the end of our Ph.D.; those memories are
truly precious. Thanks for all the movie nights and delicious ube bread, egg tarts, and Filipino food.
Mami, you are the funniest person that I know who does not think they are being funny. You are
so brave, passionate, and ambitious; I really admire your fighter’s spirit and I am so grateful to
have you in my life. Victor, the chillest person I have ever met. Thanks for your company in all
those trips to Asian grocery stores, for giving Randy so much love, and for watching the zombie
show Kingdom with me. I am excited to visit you in Thailand someday. Nick and Sanket, it was
great to have you guys to go through the whole graduate school together. I miss the good old times
we had in Hillview, and then at the Adams house. Thanks for always being around and supporting
me. To all the other amazing and adorable people that I was fortunate enough to make a bond with,
Shuyang, Bailey, Wei, Ben, Van, Billy, Christina, Lanja, James, Zelin, Robert, Abegail, Aneesh,
vi
Seda, and Ana, you all made my graduate school so much more colorful and joyful. Thank you for
all the laughter, support, advice, gossip, food, and boba.
I also want to thank the group of Chinese friends I met outside of the Chemistry Department,
Cheng, Jikun, Yuntao, Ruixue, Qingman, Wanxin, Xueqing, Liangyu, Hanfan. Thank you all for
being here with me in this foreign land and providing me with a little harbor to go to when I craved
Chinese food or when I felt homesick during traditional Chinese festivals. To all my friends back
home, although we are thousands of miles apart, our friendship is never diminished by time and
distance. Thank you for loving me as who I am and being so supportive all the time.
Words cannot express how grateful I am to my family. To my parents, thank you for the
unconditional support and love you give me throughout my life. Although you never fully
understood my intention to pursue a Ph.D. overseas, you still chose to stand with me and provided
me with everything I could possibly ask for. Grandma, thanks for taking good care of me when I
was little and showing me the good virtues of being a kind, compassionate, generous, and
conscientious person. You shaped who I am, I love you so much.
To Bryce, thank you for your love, patience, support, and everything. I am so fortunate to have
you by my side throughout this journey, where you provided me with endless courage and bad
jokes. Thanks for appreciating and respecting me and my culture, and introducing me to all the
delicious American cuisine and Jurassic Park, Star Wars, Pirates of the Caribbean movies. Being
with you makes me love myself more and motivates me to achieve higher every day. Can’t wait
to make the world a better place with you.
vii
Table of Contents
Dedication ....................................................................................................................................... ii
Acknowledgements ....................................................................................................................... iii
Table of Contents .......................................................................................................................... vii
List of Tables ................................................................................................................................... x
List of Figures ............................................................................................................................... xii
Abstract ...................................................................................................................................... xviii
Chapter 1: General Introduction ...................................................................................................... 1
1.1 Global Climate Crisis and Energy Outlook ........................................................................... 2
1.2 Storage of Renewables through Electrocatalysis .................................................................. 4
1.3 Hydrogen Evolution Reaction (HER) ................................................................................... 5
1.4 Homogeneous and Heterogeneous Electrocatalysts .............................................................. 8
1.5 Metal-Organic Frameworks (MOFs) for Electrocatalysis ................................................... 10
1.6 Dithiolate-Based Metal-Organic Frameworks for Hydrogen Evolution ............................. 12
1.7 Outline of This Work ........................................................................................................... 13
1.8 References ........................................................................................................................... 16
Chapter 2: Improving and Understanding the Hydrogen Evolving Activity of a Cobalt Dithiolene
Metal-Organic Framework ............................................................................................................ 19
2.1 Introduction ......................................................................................................................... 20
2.2 Results and Discussion ........................................................................................................ 26
2.2.1 Optimization of the Electrochemical HER Activity of CoTHT .................................. 26
2.2.2 Electrochemical Characterization of 1 ......................................................................... 33
2.2.3 Electrocatalytic Stability of 1 ....................................................................................... 37
2.2.4 Determination of the Structural Model for CoTHT ..................................................... 41
2.2.5 DFT Studies of the HER Mechanism Using PBEsol+D3 ............................................ 44
2.2.6 DFT Studies of the HER Mechanism Using HSE06+D3 ............................................. 46
2.2.7 Discussion of the HER Mechanism .............................................................................. 48
2.3 Conclusions ......................................................................................................................... 50
2.4 Supplementary Experimental Information .......................................................................... 51
2.4.1 General Considerations ................................................................................................. 51
2.4.2 Synthesis of CoTHT .................................................................................................... 51
2.4.3 Deposition of CoTHT for Electrochemical Study ....................................................... 52
2.4.4 Electrochemical Methods ............................................................................................. 52
2.4.5 Physical Characterization Methods .............................................................................. 54
2.4.6 General DFT Methods .................................................................................................. 55
2.4.7 Comparing Different DFT Models ............................................................................... 58
2.4.8 Comparing Different Configurations of CoTHT ......................................................... 62
2.4.9 Molecular Modeling ..................................................................................................... 63
2.4.10 The HER Mechanism of CoTHT in the AB Configuration ....................................... 63
2.4.11 The HER Mechanism of CoTHT in the AA Configuration ...................................... 68
2.4.12 The HER Mechanism in the AB Configuration Using Hubbard Correction (+U) ..... 68
2.5 References ........................................................................................................................... 70
viii
Chapter 3: Hydrogen Evolving Activity of Dithiolene-Based Metal-Organic Frameworks with
Mixed Cobalt and Iron Centers ..................................................................................................... 78
3.1 Introduction ......................................................................................................................... 79
3.2 Results and Discussion ........................................................................................................ 81
3.2.1 Synthesis and Characterization ..................................................................................... 81
3.2.2 Electrocatalytic H2 Evolution Studies. ......................................................................... 86
3.2.3 Discussion ..................................................................................................................... 96
3.3 Conclusions ....................................................................................................................... 100
3.4 Supplementary Experimental Information ........................................................................ 102
3.4.1 General Experimental Methods .................................................................................. 102
3.4.2 Synthesis of FeTHT ................................................................................................... 102
3.4.3 Synthesis of Mixed Fe/Co THT .................................................................................. 103
3.4.4 Deposition of MOFs for Electrochemistry Study ....................................................... 103
3.4.5 Electrochemical Methods ........................................................................................... 103
3.4.6 Physical Characterization Methods ............................................................................ 105
3.4.7 Computational Methods for FTIR Spectrum Simulation ........................................... 106
3.5 References ......................................................................................................................... 107
Chapter 4: Cu[Ni(2,3-pyrazinedithiolae)2] Metal-Organic Framework for Electrocatalytic
Hydrogen Evolution .................................................................................................................... 113
4.1 Introduction ....................................................................................................................... 114
4.2 Results and Discussion ...................................................................................................... 116
4.2.1 Electrochemical Studies ............................................................................................. 116
4.2.2 Physical Characterizations .......................................................................................... 123
4.2.3 Density Functional Theory Calculations .................................................................... 127
4.2.4 Discussion ................................................................................................................... 134
4.3 Conclusions ....................................................................................................................... 138
4.4 Supplementary Experimental Methods ............................................................................. 138
4.4.1 Synthesis of Cu[Ni(2,3-pyrazinedithiolate)2] (1) ....................................................... 138
4.4.2 Deposition of 1 for Electrochemistry Study ............................................................... 139
4.4.3 Electrochemical Methods ........................................................................................... 139
4.4.4 Physical Characterization Methods ............................................................................ 141
4.5 References ......................................................................................................................... 142
Chapter 5: Synthesis and Investigation of a Cobalt Triphenylenehexaselenolate Metal-Organic
Framework for Electrocatalytic Hydrogen Evolution ................................................................. 148
5.1 Introduction ....................................................................................................................... 149
5.2 Results and Discussion ...................................................................................................... 151
5.2.1 Synthesis and Physical Characterization .................................................................... 151
5.2.2 Electrochemical Characterizations of CoTHSe(AP) ................................................. 157
5.2.3 Electrochemical Characterizations of CoTHSe(OX) ................................................. 162
5.3 Conclusions and Future Direction ..................................................................................... 163
5.4 Experimental Details ......................................................................................................... 165
5.4.1 Synthesis of THSe
t
Bu ................................................................................................. 165
5.4.2 Synthesis of CoTHSe via Method 1 ........................................................................... 169
5.4.3 Synthesis of CoTHSe via Method 2 ........................................................................... 172
5.4.4 Physical Characterizations .......................................................................................... 173
ix
5.4.5 Electrochemistry Methods .......................................................................................... 174
5.5 References ......................................................................................................................... 175
Bibliography ................................................................................................................................ 178
x
List of Tables
Table 1.1 Elementary steps for the HER and the corresponding theoretical Tafel slope .............. 15
Table 2.1. Summary of the HER activity of dithiolene-based MOFs ........................................... 24
Table 2.2. Characteristic values of ink mixtures with variable CB wt% ...................................... 31
Table 2.3. Characteristic values of electrodes deposited with variable amounts of 1. .................. 32
Table 2.4. Values extracted from fitting the EIS data of 1 to the 2TS equivalent circuit ............. 35
Table 2.5. Characteristic values for 1 before and after 9 h CCE. .................................................. 39
Table 2.6. Characteristic values for 1 before and after 100 CV cycles. ........................................ 39
Table 2.7. Free energy change for the addition of one proton/electron pair to the cobalt atom on a
bare CoTHT MOF in the AB configuration with increasing energy cutoff and K-point values using
PBEsol+D3. ................................................................................................................................... 58
Table 2.8. Free energy difference between the AA and AB configurations calculated using
different methods. .......................................................................................................................... 62
Table 2.9. Free energies for the binding of H and H2 to the A-site cobalt atom in CoTHT calculated
using different methods. ................................................................................................................ 67
Table 3.1 Metal content in mixed metal THT-based MOFs determined by ICP-OES ................. 82
Table 3.2 Rct values of FeTHT derived from fitting EIS data to the 2TS model ......................... 88
Table 3.3 Characteristic values of FeTHT extracted from the EIS responses shown in Figure 3.9b.
....................................................................................................................................................... 89
Table 3.4 Electrocatalytic properties of THT-based MOFs .......................................................... 91
Table 3.5 Rct values of the materials derived from fitting EIS data to the 2TS model .................. 91
Table 3.6 Rct and Cdl values of 2 before and after CPE, extracted from EIS at η = 168 mV ........ 93
Table 3.7 Turnover frequency (TOF) calculations for 1, 2, and 3 based on the CPE experiment
shown in Figure 3.12. .................................................................................................................... 93
Table 4.1 HER activity of 1 before and after CPE (-0.59 V vs. RHE) ........................................ 122
Table 4.2 EDX measurement results of 1 before and after electrolysis (at -0.59 V vs. RHE) .... 127
Table 4.3 ICP-OES measurement results of 1 before and after electrolysis (at -0.59 V vs. RHE)
..................................................................................................................................................... 127
Table 4.4 Relative energies (eV) for the addition of two hydrogen atoms (2H) per unit cell of
Cu[Ni(pdt)2] computed using PBE-D3 functional ....................................................................... 130
Table 4.5 Cu-N and Ni-S bond lengths (Å) of the pristine MOF and H atom added structures (on
N center) after geometry optimization. ........................................................................................ 131
Table 4.6 Magnetic moments (μB) on Cu and Ni centers before and after the addition of H atoms
on the N center ............................................................................................................................. 131
xi
Table 4.7 CM5(Bader) Charges on Cu and Ni centers before and after the addition of H atoms on
the N center .................................................................................................................................. 131
Table 4.8 Absolute and relative energies (eV) of various structures shown in Figure 4.19 ........ 133
Table 5.1 Binding energy (eV) of XPS peaks for CoTHSe in comparison with analogous literature
precedents .................................................................................................................................... 154
Table 5.2 Binding energy (eV) of XPS peaks for CoTHSe synthesized via Method 2 .............. 155
Table 5.3 Tafel slopes of electrodes prepared with different catalyst loadings .......................... 161
xii
List of Figures
Figure 1.1. Renewable share of total final energy consumption in 2017 ........................................ 3
Figure 1.2. Schematic of a sustainable energy landscape based on electrocatalysis ....................... 5
Figure 1.3. An electrochemical water splitting device .................................................................... 7
Figure 1.4 Examples of conducive 2D MOFs ............................................................................... 11
Figure 1.5 Cobalt dithiolene-based MOFs (CoBHT and CoTHT) formed from the reaction of Co
2+
and the trinucleating ligands, benzenehexathiolate (BHT) or triphenylene-2,3,6,7,10,11-
hexathiolate (THT). ....................................................................................................................... 13
Figure 2.1. Structural schematics of the dithiolene-based 2D MOFs containing trinucleating
ligands, such as triphenylene-2,3,6,7,10,11-hexathiolate (THT) and benzenehexathiolate (BHT).
....................................................................................................................................................... 22
Figure 2.2. (a) Cyclic voltammograms (CVs) of three individual glassy carbon electrodes (GCEs)
modified with the same batch of CoTHT film, scan rate: 100 mV/s. (b) Nyquist plots (markers)
with respective fits (solid lines) measured at -0.47 V vs. RHE. All measurement were performed
in N2-satuarted pH 1.3 aqueous solutions. ..................................................................................... 27
Figure 2.3. (a) CVs of GCE_1 (red) and GCE_2 (blue), prepared by drop casting 10 uL of the
CoTHT/Nafion mixture onto the GCE surface, scan rate: 100 mV/s; (b) Nyquist plots (markers)
with respective fits (solid lines) measured at -0.468 V vs. RHE. All measurements were performed
in N2-saturated pH 1.3 aqueous solutions. ..................................................................................... 28
Figure 2.4. Top down SEM images of 1 ........................................................................................ 29
Figure 2.5. Synchrotron PXRD pattern of pristine CoTHT (red) and 1 (blue), wavelength:
0.412750Å. .................................................................................................................................... 29
Figure 2.6. (a) CVs of one GCE deposited with only CB and Nafion and three GCEs prepared by
drop casting 10 uL of ink mixture 1 onto the electrode surface; (b) CVs of GCE_1 subjected to 10
repetitive CV cycles; (c) Nyquist plot (markers) of GCE_1 with fit (solid line) measured at -0.168
V vs. RHE, Rct = 34.6 𝛺. Scan rate for CVs: 100 mV/s. All measurements were performed in N2-
saturated pH 1.3 aqueous solutions. .............................................................................................. 30
Figure 2.7. (a) Polarization curves of a bare GCE and GCEs deposited with ink mixtures with
variable carbon black wt%, scan rate: 5 mV/s. (b) Nyquist plots (markers) of each electrode with
respective fits (solid lines) measured at -0.168 V vs. RHE. All measurements were performed in
N2-saturated pH 1.3 H2SO4 solutions at room temperature. .......................................................... 31
Figure 2.8. Tafel plots of (a) CoTHT/Nafion mixture and (b) ink composites with variable CB
wt%, extracted from polarization curves presented in Figure 2.7. ................................................ 31
Figure 2.9. Polarization curves of electrodes deposited with variable amount of 1 in N2-saturated
pH 1.3 aqueous solutions, scan rate: 5 mV/s. (b) Nyquist plots (markers) of electrodes deposited
with variable amount of 1 with respective fits (solid lines) measured at -0.168 V vs. RHE (c)
Comparison of current density (mA/cm
2
) vs. loading (μL) under different overpotentials. (d)
Polarization curves of 1 (red) and 20% Pt/C (black) in N2-saturated pH 1.3 aqueous solutions, the
inset shows the corresponding Tafel plot. Scan rate: 5 mV/s. ....................................................... 33
xiii
Figure 2.10. (a) Nyquist and (b) Bode plots of 1 (markers) with respective fits (solid lines)
measured at variable overpotentials. All measurements were performed in N2-saturated pH 1.3
aqueous solutions. .......................................................................................................................... 34
Figure 2.11. Tafel plot obtained from EIS spectra (red) and polarization curve (blue). ............... 35
Figure 2.12. (a) Polarization curves illustrating the pH-dependent HER activity of 1, scan rate: 5
mV/s; (b) Catalytic onset potential as a function of pH, slope: 57.5 mV/dec. .............................. 36
Figure 2.13. (a) Scan rate (mV/s) dependence experiments of 1 in a N2-saturated pH 10 aqueous
solution. (b) Plot of log |jp| versus log(𝜐). ...................................................................................... 36
Figure 2.14. (a) 2 h controlled potential electrolysis (CPE) of 1 (red, 9.73×10
-7
molCo) and carbon
black (blue) at -0.19 V vs. RHE, Faradaic efficiency = 98%. (b) 8 h CPE of 1 (1.00×10
-6
molCo)
at -0.31 V vs. RHE. (c) 24 h CPE of 1 at -0.31 V vs. RHE. All measurements were performed in
N2-saturated pH 1.3 aqueous solutions. ......................................................................................... 38
Figure 2.15. (a) Controlled current electrolysis (CCE) of 1 at 10 mA/cm
2
in a N2-saturated pH 1.3
H2SO4 solution; (b) Polarization curves before and after electrolysis, scan rate: 5 mV/s; (c) Nyquist
plots (markers) with respective fits (solid lines) before and after electrolysis, measured at -0.168
V vs. RHE. ..................................................................................................................................... 38
Figure 2.16. (a) CVs (cathodic scans) of 1 (1.16×10
-6
molCo/cm
2
) as a function of electrochemical
cycling in N2-saturated pH 1.3 H2SO4 solution, scan rate: 20 mV/s; (b) Polarization curves before
and after cycling, scan rate: 5 mV/s; (c) Nyquist plots (markers) with respective fits (solid lines)
before and after cycling, measured at -0.168 V vs. RHE. ............................................................. 39
Figure 2.17. Top down SEM images of 1 on RDE after electrolysis. ........................................... 40
Figure 2.18. XPS spectra of 1 before (red) and after (blue) electrolysis: (a) Survey spectrum, (b)
High-resolution Co 2p region, (c) High-resolution S 2s region. ................................................... 40
Figure 2.19. XRD of 1 after 2 h CPE. ........................................................................................... 41
Figure 2.20. A comparison of the geometries of the AA (a) and the AB configuration (b and c).
The spheres representing cobalt, sulfur, carbon, and hydrogen are colored blue, yellow, brown,
and white, respectively. The AB configuration has two distinct cobalt sites, labeled as A and B
site. Panel (b) shows that there are 4 A sites and 2 B sites in each hexagonal pore. ..................... 42
Figure 2.21. A comparison of the experimental (blue), and simulated PXRD patterns of CoTHT
in the AB (red) and AA (green) configurations. The simulated PXRD patterns are displayed using
the Materials Studio (version 8.0) suite of programs by Accelrys. Left panel: the overall pattern;
right panel: magnified at the [100] and [200] reflection peaks for easier comparison. ................. 44
Figure 2.22. The calculated free energy changes for the HER pathways in CoTHT that adopts (a)
AB stacking configuration or (b) AA stacking configuration. The DFT calculations were
performed using the PBEsol+D3 method. ..................................................................................... 45
Figure 2.23. The calculated free energy changes for the HER pathways in CoTHT that adopts (a)
AB stacking configuration or (b) AA stacking configuration. The DFT calculations were
performed using the HSE06+D3 method. ..................................................................................... 47
Figure 2.24. The lattice parameters a (panel a), c (panel b), and magnetic moment (µB) per formula
unit of Co3(THT)2 (panel c) of the AB configuration as a function of U using PBEsol+D3. The
xiv
experimental lattice parameters are 𝑎 = 22.52 Å 𝑎𝑛𝑑 𝑐 = 6.6 Å; and the experimental magnetic
moment is 1.55 µB. ......................................................................................................................... 60
Figure 2.25. The labeling scheme used to describe where the proton/electron pairs bind to the
active sites in CoTHT for the (a) A site of the AB configuration, (b) B site of the AB configuration,
and (c) AA configuration. Panel (d) shows a proton/electron pair bound to the S(1) atom in the AB
configuration. Panel (a) shows the periodic boundary conditions by showing Co(2) on both the top
and the bottom as these are identical and both are shown for visualization purposes. ................. 64
Figure 2.26. A comparison of the relative free energies between intermediates at the A site of the
AB configuration of CoTHT. ....................................................................................................... 65
Figure 2.27. A comparison of the relative free energies between intermediates at the B site of the
AB configuration of CoTHT. ....................................................................................................... 66
Figure 2.28. A comparison of the relative free energies between intermediates of the AA
configuration of CoTHT. .............................................................................................................. 68
Figure 2.29. A comparison of the relative free energies between intermediates (a) bare and CoH
and (b) CoH and CoH2 of the AB configuration of CoTHT as a function of U. Both reactions occur
at the 1-cobalt site of the AB configuration. ................................................................................. 69
Figure 3.1 (a) Synchrotron and simulated PXRD patterns of THT-based MOFs, X-ray wavelength:
0.4127 Å. (b) DFT simulated FTIR spectrum of the cobalt dithiolene molecular model (black trace)
and the experimental FTIR spectra of the MOF series. ................................................................. 83
Figure 3.2 High resolution XPS spectra of 1, 2, and 3: (a) Co 2p region; (b) Fe 2p region; (c) C 1s
region; (d) S 2s region. .................................................................................................................. 85
Figure 3.3 High-resolution XPS spectra of 2: (b) Co 2p core level; (c) Fe 2p core level. ............ 85
Figure 3.4 Top-down SEM images of (a) 1, (b) 2, and (c) 3. ........................................................ 85
Figure 3.5 (a) Top-down SEM image of 2. EDX spectra of 2 revealing the presence and
homogeneous distribution of (b) S, (c) Co, and (d) Fe. ................................................................. 86
Figure 3.6 (a) Polarization curves of FeTHT (red), bare glassy carbon electrode (black), and
components of the ink (grey). (b) Tafel plot obtained from polarization curve. Scan rate: 5 mV/s,
rotation rate of RDE: 2000 rpm, loading of catalyst: 2.85 × 10
-7
molFe/cm
2
. All measurements
were performed in N2-saturated pH 1.3 aqueous solutions. .......................................................... 88
Figure 3.7 (a) The two-time constant serial model (2TS) employed to fit the EIS response of
FeTHT. (b) Nyquist plots (markers) of FeTHT at various overpotentials with respective fits (solid
lines). (c) and (d): Bode plots (markers) of FeTHT at various overpotentials with respective fits
(solid lines). All measurements were performed in N2-saturated pH 1.3 aqueous solutions. ....... 88
Figure 3.8 Consecutive CV scans of FeTHT in a N2-saturated pH 1.3 aqueous solution, scan rate:
50 mV/s. ......................................................................................................................................... 89
Figure 3.9 (a) 1 h CPE experiment of FeTHT at -0.518 V vs. RHE, FE: 72.2 ± 1.4%. (b) Nyquist
plots (marks) fitted to 2TS model (lines) showing EIS responses at different overpotentials before
and after CPE. ................................................................................................................................ 89
Figure 3.10 Polarization curves of FeTHT, mixed-metal MOF 1-3, and CoTHT with current
normalized by (a) geometric surface area and (b) bulk loading of the Co metal centers. ............. 90
xv
Figure 3.11 Nyquist (a) and Bode plots (b and c) showing EIS responses of the materials at various
overpotentials. For FeTHT and 1, η = 368 mV; for 2 and 3, η = 168 mV ................................... 91
Figure 3.12 Controlled potential electrolysis (CPE) of 1 (red), 2 (blue), and 3 (green) at -0.418 V
versus RHE. ................................................................................................................................... 93
Figure 3.13 Nyquist (a) and Bode (b) plots showing the EIS responses of 2 at η = 168 mV before
and after CPE. ................................................................................................................................ 93
Figure 3.14 Consecutive CV scans of (a) FeTHT, (b) 1, (c) 2, (d) 3 in pH 10 electrolyte solutions,
scan rate: 50 mV/s. ........................................................................................................................ 94
Figure 3.15 XRD patterns of (a) MOF 2 and (b) FeTHT before and after 1 h CPE experiment . 95
Figure 3.16 High resolution XPS of FeTHT before and after 1 h CPE: (a) Fe 2p; (b) S 2s ......... 96
Figure 3.17 High resolution XPS of MOF 2 before and after 1 h CPE: (a) Co 2p; (b) Fe 2p; (c) S
2s .................................................................................................................................................... 96
Figure 3.18 The geometry of the cobalt dithiolene molecular model ......................................... 107
Figure 4.1 Structure of Cu[Ni(pdt)2] (1). Color coding for atoms: green (Ni), cyan (Cu), yellow
(S), dark blue (N), and gray (C). H atoms are omitted for clarity. .............................................. 116
Figure 4.2 X-ray diffraction patterns of 1: (1) as-prepared (red); (2) after 20 minutes sonication in
the presence of Nafion, ethanol, and water (blue); (3) after soaking in an electrolyte solution of pH
1.3 for 18 h (purple). Inset is an image of the dark purple-colored sonicated catalyst mixture. . 117
Figure 4.3 Polarization curve of 1 obtained from linear sweep voltammetry (LSV) in a N2-saturated
pH 1.3 aqueous solution; scan rate: 5 mV/s, RDE rotation rate: 1600 rpm. ............................... 118
Figure 4.4 (a) Tafel plot of 1 derived from the polarization curve shown above. (b) Nyquist plot of
1 (markers) and the respective fit (line) at -0.52 V vs. RHE; Inset shows the 2TS equivalent circuit
used to fit the EIS response. ........................................................................................................ 118
Figure 4.5 (a) Polarization curve of 1 magnified at the catalytic onset region, scan rate: 5 mV/s. (b)
CV scan of 1 with the inset focused on the higher potential region, scan rate: 50 mV/s. ........... 119
Figure 4.6 (a) 1 h CPE experiment of 1 at -0.59 V vs. RHE; (b) Nyquist plots (markers) and
respective fits (lines) before and after electrolysis; EIS potential: -0.49 V vs. RHE. ................. 120
Figure 4.7 Controlled potential electrolysis performed at -0.50 V vs RHE in a N2-saturated pH 1.3
solution. ....................................................................................................................................... 121
Figure 4.8 (a) 4 h electrolysis of 1 at -0.59 V vs RHE. (b) LSVs of 1 acquired before, after 2 h, and
after 4 h CPE, in comparison with the Pt/C catalyst; scan rate: 5 mV/s. .................................... 121
Figure 4.9 CPE and subsequent wash test conducted at -0.59 V vs RHE in a N2-saturated pH 1.3
aqueous solution. ......................................................................................................................... 122
Figure 4.10 (a) Polarization curves of 1 obtained in various-pH electrolytes, scan rate: 5 mV/s. (b)
Plot of onset potential vs. corresponding pH values. .................................................................. 123
Figure 4.11 1 h CPE performed at -0.59 V vs RHE in N2-saturated pH 4.6 aqueous solution. .. 123
Figure 4.12 SEM images of (a) as-prepared 1; (b) composite of 1 after sonication with water,
ethanol, and Nafion; (c) 1 after 1 h electrolysis; (d) 1 after 4 h electrolysis. .............................. 124
xvi
Figure 4.13 PXRD patterns of 1 before and after 1 h CPE at -0.59 V vs RHE in a N2-saturated pH
1.3 solution. ................................................................................................................................. 124
Figure 4.14 High-resolution XPS spectra of the as-prepared 1 (purple) and the post-CPE 1 sampled
at two different spots (blue and red). (a) Ni 2p region; (b) Cu 2p region; (c) S 2s region. ......... 126
Figure 4.15 The primitive cell used for DFT calculations .......................................................... 128
Figure 4.16 A pictorial representation of a 2×2×2 supercell of Cu[Ni(pdt)2] MOF upon addition
of two H atoms to different sites per unit cell of the MOF (prior structural optimization). (a) H
atoms are added to N centers attached to two different Cu centers (2H-diff-Cu-N); (b) H atoms
are attached to N centers, which are cis with respect to the same Cu center (2H-same-Cu-cis-N);
(c) H atoms are attached to N centers, which are trans with respect to the same Cu center (2H-
same-Cu-trans-N); (d) H atoms are attached to S centers, which are coordinated to two different
Ni centers (2H-diff-Ni-S); (e) H atoms are attached to S centers, which are cis with respect to the
same Ni center (2H-same-Ni-cis-S-1); (f) H atoms are attached to S centers, which belong to the
same pdt ligand (2H-same-Ni-cis-S-2), (g) H atoms are attached to S centers, which are trans with
respect to the same Ni center (2H-same-Ni-trans-S). Color coding for atoms: green (Ni), cyan
(Cu), yellow (S), dark blue (N), white (H), and gray (C). ........................................................... 129
Figure 4.17 Optimized structure of 2H-diff-Cu-N using the PBE-D3 functional. ..................... 131
Figure 4.18 (a) Top view and (b) side view of a 2×2×2 supercell (shown for visualization
purpose only) of Cu[Ni(pdt)2] MOF (after structural optimization) upon addition of four hydrogen
atoms (4 H) per unit cell of the MOF. ......................................................................................... 133
Figure 4.19 Preliminary calculations to determine the active site of the catalyst, where one H atom
is added to: (a) the Ni site; (b) the S site closer to N; (c) the S site closer to Cu; (d) the Cu site. (e):
Optimized structure of (d), where the H atom transferred to the N atom after structural optimization.
..................................................................................................................................................... 133
Figure 5.1 Structures of the active sites of [NiFe] and [NiFeSe] hydrogenase ........................... 149
Figure 5.2 Synthetic routes of CoTHSe MOFs ........................................................................... 152
Figure 5.3 PXRD patterns of CoTHSe synthesized via (a) Method 1 with NaOAc as base; (b)
Method 1 with KOH as base; (c) Method 2. ................................................................................ 153
Figure 5.4 The XPS spectra of CoTHSe(AP) synthesized by Method 1 (KOH as base). (a) Survey
spectrum; (b) Co 2p region; (c) O 1s region; (d) Se 3d region .................................................... 154
Figure 5.5 The XPS spectra of CoTHSe(OX) synthesized by Method 1 (KOH as base). (a) Survey
spectrum; (b) Co 2p region; (c) O 1s region; (d) Se 3d region. ................................................... 155
Figure 5.6 The XPS spectra of CoTHSe(AP) synthesized by Method 2. (a) Survey spectrum; (b)
Co 2p region; (c) O 1s region; (d) Se 3d region. ......................................................................... 156
Figure 5.7 The XPS spectra of CoTHSe(OX) synthesized by Method 2. (a) Survey spectrum; (b)
Co 2p region; (c) O 1s region; (d) Se 3d region. ......................................................................... 156
Figure 5.8 Polarization curve and Tafel plot of CoTHSe(AP) synthesized by Method 2, loading:
15 μL. ........................................................................................................................................... 157
Figure 5.9 EIS response of CoTHSe(AP) synthesized by method 2, loading: 15 uL. ................ 159
xvii
Figure 5.10 (a) Polarization curves of CoTHSe(AP) collected before and after EIS. (b) CVs with
variable scan rates, only cathodic sweeps are shown for simplicity. .......................................... 159
Figure 5.11 Cdl measurements of electrodes deposited with variable amounts of CoTHSe(AP) ink.
..................................................................................................................................................... 161
Figure 5.12 Polarization curves of CoTHSe(AP) with variable loadings .................................. 161
Figure 5.13 Polarization curve of CoTHSe(OX) in comparison with CoTHSe(AP) and the
corresponding Tafel plot .............................................................................................................. 162
Figure 5.14 Polarization curves of CoTHSe(OX) and CoTHSe(AP) with current normalized by
Cdl ................................................................................................................................................ 163
Figure 5.15 Nyquist plots of CoTHSe(OX) and CoTHSe(AP) collected at -0.48 V vs RHE ... 163
Figure 5.16 The synthetic route of THSe
t
Bu .............................................................................. 166
Figure 5.17
1
H NMR spectrum of LiSe
t
Bu(dioxane) (500 MHz, CD3CN) ................................. 167
Figure 5.18
1
H NMR spectrum of HBT (500 MHz, CDCl3), HBT is poorly soluble in common
NMR solvents. ............................................................................................................................. 167
Figure 5.19
1
H NMR spectrum of THSe
t
Bu (500 MHz, CDCl3) ............................................... 168
Figure 5.20
77
Se NMR spectrum of THSe
t
Bu (500 MHz, CDCl3) ............................................. 168
Figure 5.21
13
C NMR spectrum of THSe
t
Bu (500 MHz, CDCl3) .............................................. 169
Figure 5.22 The synthetic route of CoTHSe via Method 1 ......................................................... 171
Figure 5.23
1
H NMR spectrum of THSeAc (500 MHz, CDCl3) ................................................ 172
xviii
Abstract
In the face of a looming climate crisis, immediate action needs to be taken to decarbonize the
current fossil-fuel-dominated energy economy by adopting a more sustainable alternative. Thanks
to the alliance between scientific breakthroughs and policy support, recent years have witnessed
the rapid growth in renewable electricity generation using solar and wind energy. However, the
integration of renewables into other major energy sectors, such as transportation and thermal
production, is still very limited. To address this challenge, electrocatalysis provides a feasible
solution, where abundant small molecules are converted into value-added products via the input
of renewable electricity to realize the storage of renewables in chemical bonds. The products can
then be consumed as fuels for transportation or thermal production, or as chemical feedstocks in
industrial processes. The key to realize this electrocatalysis-based sustainable energy future is to
develop highly efficient electrocatalysts that are composed of abundant elements and can facilitate
chemical catalysis under mild conditions.
In this dissertation, several electrocatalysts that are based on metal-organic frameworks (MOFs)
are developed for green hydrogen production from water. The electrocatalytic H2 production is of
particular interest because H2 is a very important chemical feedstock in industrial productions and
a promising carbon-free energy carrier. In recent years, MOFs have emerged as an extensive class
of highly functional materials with unique properties such as high porosities, large surface areas,
and extraordinary structural and compositional variabilities. The application of MOFs in clean
energy is an emerging field of research and is of great significance in the context of current climate
crisis. A brief outline of this dissertation is provided below:
Chapter 1 presents a general introduction of the dissertation, including a brief discussion on
the current global energy status, the fundamentals of electrocatalytic hydrogen production, general
xix
design principles of electrocatalysts, as well as the frontiers in MOF-based electrocatalysis. In
Chapter 2, the HER performance of a known dithiolene-based MOF, the cobalt triphenylene-
2,3,6,7,10,11-hexathiolate (THT) MOF, is optimized by unraveling the reaction mechanism and
identifying the key factors that dictate the overall catalytic performance. The optimization results
in the most active MOF-based electrocatalyst for hydrogen production that comprises only earth
abundant elements. Chapter 3 discusses the role of metal centers in the HER activity using the
examples of a series of iron and cobalt/iron mixed-metal dithiolate MOFs. In Chapter 4, a
conductive three-dimensional dithiolene-based MOF, the Cu[Ni(2,3-pyrazinedithiolate)2] MOF, is
investigated as an HER electrocatalyst for the first time. Lastly, Chapter 5 presents the synthesis
and HER characterization of a diselenolate-based MOF, an analogous derivative of the dithiolate
MOFs. This study is to highlight the role of the chalcogen within the ligand, which is inspired by
nature where the selenium-containing [NiFe] hydrogenase displays much higher activity than its
sulfur-only analogue for hydrogen production.
1
Chapter 1
General Introduction
2
1.1 Global Climate Crisis and Energy Outlook
It is an undeniable fact that the human activities have drastically changed the course of planet
Earth. The deadly heatwaves and floods in Europe, the devastating wild fires in Australia, and the
long-lasting droughts in western United States, these are just a few cautionary tales told by the
changing climate. It is estimated that the global average temperature has increased by
approximately 1 degree Celsius since the pre-industrial period, which is directly linked to the
anthropogenic greenhouse gas emissions.
1
Carbon dioxide (CO2), produced mainly from the
combustion of fossil fuels, accounts for two-thirds of the total greenhouse emission, followed by
methane and nitrous oxides, generated mostly from agricultural activities.
2
In the face of the
climate crisis, many local and national governments as well as intergovernmental organizations
have established targets to shift onto a path of decarbonization and to reach carbon-neutral in the
foreseeable future. The Intergovernmental Panel on Climate Change (IPCC) has set the goal to
limit global warming to well below 2, preferably 1.5 degree Celsius, compared to pre-industrial
levels, to prevent further irreversible damages to the planet. This requires the CO2 emissions to
decline by at least 25% by 2030 relative to the 2010 levels, and to reach net zero by around 2070.
3
To achieve such goals, the key is to decrease the reliance on traditional fossil fuels by shifting the
current energy system to a renewable-based one.
Thanks to the alliance between scientific breakthroughs and policy support, renewable energy
has been growing rapidly in the past decade. The main growths are seen in the power (electricity)
sector, driven by the cost reductions in wind and solar photovoltaic (PV) technologies and
sustained policy support. In 2020, renewables produced 38% of the European Union’s electricity,
marking the first time in history that the renewables generated more electricity than fossil fuels.
4
Globally, solar PV and onshore wind are among the most cost effective ways to increase power
3
capacity in many countries today, and the total installed capacity of them combined are on course
to surpass natural gas in 2023 and coal in 2024.
5
Despite the significant progress in the power
sectors, shares of renewables in the total final energy consumption (TFEC) is growing very slowly.
As of 2018, modern renewable energy only accounted for an estimate of 11% of TFEC, which is
largely due to the limited shares of renewables in other major energy sectors such as heating,
cooling, and transport. As shown in Figure 1.1, in 2017, the highest contributions of renewables
(26.4%) is made to the power sector, which only accounted for 17% of TFEC. The end uses in the
transport sector made up for 32% of TFEC, yet only 3.3% was supplied by renewable energy.
Although positive progress has occurred in the transport sector, such as the advancement in both
biofuels and electric vehicles, it is largely offset by the rapidly rising energy demand. Energy use
in the thermal sector, typically for heating and cooling in buildings and industrial processes,
accounted for more than half of TFEC, of which 10.1% was fulfilled by renewables. Out of the
10.1%, modern bioenergy represented the largest heat contributor (7.2%), followed by renewable
electricity (1.9%) and solar and geothermal heat (1.0%). The profound lag of the renewable uptake
in the thermal and transport sectors is mainly ascribed to insufficient policy support as well as the
slow development in new technologies. It is severely hindering the progress towards the global
climate and development goals, especially in view of the ever-growing global energy demand.
Figure 1.1. Renewable share of total final energy consumption in 2017
6
4
1.2 Storage of Renewables through Electrocatalysis
To accelerate the integration of renewables into the thermal and transport sectors, one of the
key strategies is to improve the flexibility of the renewable electrical energy by storing it off the
grid through electrochemical technologies. Given the diffuse and intermittent nature of solar and
wind energy, this strategy is highly desirable as it allows for a broader geographical distribution
of renewables. One of the central electrochemical technologies is the development and
commercialization of Lithium-ion batteries, which has directly advanced the booming industry of
electric vehicles (EVs). As of 2019, around 7.2 million electric cars were on the world’s road,
including battery EVs and plug-in hybrid EVs, gaining momentum for the electrification of the
transport sector.
6
Another promising approach to store the renewables is realized through electrocatalysis, where
abundant small molecules (e.g. water, carbon dioxide, and nitrogen), are converted into higher
value products (e.g. hydrogen, hydrocarbons, and ammonia) with the input of renewable electricity.
As illustrated in Figure 1.2, electrocatalysis presents a propitious opportunity to advance the
penetration of renewables in all energy sectors. For instance, the renewable hydrogen produced
via water splitting can be used as an energy carrier to produce carbon-free electricity in fuel cells.
The hydrocarbons generated by CO2 reduction can serve as fuels or chemical feedstocks, and the
electroreduction of nitrogen to ammonia holds the potential to produce fertilizers in a sustainable
and decentralized manner. Additionally, the raw materials used in these chemical transformations
are easily accessible and the deployment of them poses minimum environmental consequences.
The electrocatalytic conversion of CO2 possesses the bonus effect of sequestering the atmospheric
CO2, giving a truly carbon-neutral fuel production and consumption cycle. However, despite the
promises held by electrocatalysis, improvement in practical efficiencies are required for many of
5
the aforementioned processes to be viable for large-scale applications. To realize the envisioned
electrocatalysis-supported energy landscape, it is crucial to develop efficient electrocatalytic
systems that are cost-effective and can function under benign conditions.
Figure 1.2. Schematic of a sustainable energy landscape based on electrocatalysis
7
1.3 Hydrogen Evolution Reaction (HER)
Hydrogen (H2) is one of the most important chemical feedstocks in the contemporary industrial
world. It is primarily used in the production of ammonia (mainly for fertilizers), the refinery
processing of petrochemicals, and as the hydrogenating agent for food and drug productions.
Currently, ten million metric tons of H2 are produced in the United States every year, mostly
through a centralized steam-methane reforming process.
8
Although using low-cost natural gas to
generate H2 is economically favorable, its heavy reliance on fossil fuels leaves a huge carbon
footprint and necessitates the development of a carbon-neutral pathway for H2 generation to
comply with the current climate targets. In addition to its use as a chemical feedstock, H2 has also
emerged as a promising clean energy carrier with a gravimetric energy density around three times
higher than gasoline. The most desirable way of using H2 as a fuel is through the fuel cell
6
technology, a power generation technology that converts chemical energy to electrical energy with
much higher energy efficiencies than conventional combustion engines. When H2 is supplied to a
fuel cell (along with oxygen), the only products are electricity, water, and heat, resulting in
virtually zero green-house gas emissions. The scalability of fuel cells also allows for a wide range
of applications from laptop computer to utility power station. To date, hydrogen fuel cell
technology has been commonly applied in light-duty vehicles and are expected to be able to power
heavy duty transportations such as trains, boats, and airplanes in the near future. The wide
deployment of such clean electricity generation technology will drastically push forward the
transition toward a low-carbon energy economy.
In view of the prospects of H2 as a clean energy carrier and an important chemical feedstock,
it is paramount to shift away from steam-methane reforming and produce green H2 through a both
economical and sustainable pathway. Scientists have explored a wide portfolio of technologies to
transition toward a truly carbon-free hydrogen economy, among which the electrocatalytic water
splitting is of particular interest for the following reasons. First, this process has the advantage of
producing extremely pure H2 (>99.9%), obviating further procedures for product separation and
purification.
9
Second, water electrolysis offers opportunities for synergy with renewable electricity
generation, where renewable energy is essentially stored in H2. The storage of renewable electricity
off the grid helps to mitigate the intermittent nature of renewables and increase their shares in the
overall energy structure. Additionally, the recent drop in the price of solar and wind electricity
generation makes H2 production by electrolysis even more cost-effective and economically viable.
Lastly, water electrolyzers can be performed in a small-scale, decentralized manner, allowing for
the production of H2 in remote areas and alleviating the pressure of product distribution.
7
The electrocatalytic water splitting is composed of the following two half reactions (under
acidic conditions):
2H
+
+ 2e
-
⟶ H2 (1)
2H2O ⟶ 4H
+
+ O2 + 4e
-
(2)
Equation (1) represents the generation of H2 through the combination of two electrons and two
protons (H
+
), and is commonly referred to as the hydrogen evolution reaction (HER). Equation (2)
depicts the oxidation of water to generate oxygen gas (O2), which is commonly referred to as the
oxygen evolution reaction (OER). In an electrolysis setup, these two reactions occur on two
different electrodes, with the former happening on a cathode and the latter on an anode. Figure 1.3
is showing a schematic illustration of a polymer electrolyte membrane (PEM) electrolyzer, where
the two half reactions are separated by a solid plastic membrane that allows for the H
+
generated
in the anodic side to diffuse to the cathodic side while separating the products from the two
compartments.
Figure 1.3. An electrochemical water splitting device
10
The HER is a two-proton, two-electron transfer reaction, which requires the presence of a
catalyst to minimize the kinetic barrier to achieve efficient H2 production. The current industrial
8
standard catalysts for PEM water electrolysis are based on platinum (Pt). Though effective, Pt is
an extremely rare metal, with an average global production of around 200,000 kg/year, which is
reflected in its high market prbigice.
11
Therefore, for practical application and global deployment
of green H2 production through electrolysis, the development of efficient electrocatalysts that
employ only earth-abundant elements is crucial.
Broadly speaking, there are generally two strategies to improve the activity of an
electrocatalytic system. The first is to improve the intrinsic activity of each catalytic site, which
will require the understanding of the mechanism of the catalytic pathway and the capability to fine-
tune the active sites synthetically. The second approach is to improve the number of accessible
active sites on a given electrode to improve the overall activity of the system. This strategy focuses
more on the physical processes, such as charge and mass transport, that occur within the
electrode/catalyst architecture, and is more related to the optimization of a given electrocatalytic
system for practical applications. Both of these design principles will be applied and discussed in
the following chapters in greater detail.
1.4 Homogeneous and Heterogeneous Electrocatalysts
In general terms, electrocatalysts are broadly divided into two categories, the homogeneous
catalyst and the heterogeneous catalyst. The former is also commonly referred to as the molecular
catalyst, as the catalyst itself is a well-defined molecule, and in most cases an organometallic
complex. Thanks to its molecular nature, the mechanism of homogeneous catalysis could be
elucidated by capturing the reaction intermediates of the catalyst or characterizing them in situ via
spectroelectrochemical techniques (SEC). The understanding of the reaction mechanisms could
offer insights into catalyst modification, which can be achieved by fine-tuning the chemical
environment of the active sites through novel synthetic design. While the high synthetic tunability
9
of the homogeneous catalyst offers tremendous possibilities for the discovery of electrocatalysts
with high efficiency and selectivity, one flaw of this type of catalysts is their diffusion-limited
activity. When in use, the molecular catalyst is dissolved in an electrolyte solution but it is only
active when in the diffusion layer near the electrode surface. The diffusion kinetics limits the
concentration of the catalyst on the electrode surface, and hence the overall reaction rate of the
catalytic process. Therefore, in any given moment, only a small portion of the total catalyst is
actually performing catalysis, which is not economically ideal. The freely diffusing catalyst can
crossover between the anodic and cathodic compartments, preventing its practical application in
energy conversion devices. Additionally, the dissolution of molecular catalysts often requires the
need for organic solvents due to solubility limitations, introducing environmental and safety
concerns. The solution-phase species could be reactive towards alternative processes, such as
bimolecular decomposition, leading to compromised catalyst stability and reusability. On the other
hand, the heterogeneous catalyst is directly mounted onto the working electrode surface, and is
more robust and stable compared to the molecular systems. The insolubility nature of these
catalysts allows for the use of aqueous electrolyte media and facile separation and recycle of the
catalysts. However, heterogenous catalysts typically lack well-defined active sites, making it
difficult to investigate the reaction mechanisms and rationally improve the catalytic system.
The heterogenization of molecular catalysts has become an emerging topic of research in recent
years, as it provides a pathway to combine the advantages of homogeneous and heterogeneous
catalysts while in the meantime suppressing the drawbacks they suffer from. For example, by
incorporating a molecular catalyst into a heterogeneous structure, the activity of the catalyst is no
longer diffusion-limited, meaning the entirety of the catalyst could in theory participate in catalysis.
This makes the catalytic system more cost effective, as less catalyst is needed to achieve the target
10
activity. Additionally, site isolation prevents the bimolecular deactivation or decomposition,
improving the overall stability of the catalyst. After heterogenization, solubility is no longer a
limiting factor, so environmentally benign electrolyte solvents could be used. Lastly, the well-
defined molecular nature of the active sites allows for a bottom-up understanding the reaction
mechanism and in turn provides opportunities to improve the catalytic performance through
synthetic modifications. Common approaches to heterogenize molecular catalysts include covalent
attachment, adsorption (noncovalent) attachment, and surface polymerization.
12
1.5 Metal-Organic Frameworks (MOFs) for Electrocatalysis
Recent years have witnessed a blossoming interest in the employment of metal-organic
frameworks (MOFs) as electrocatalysts for clean energy conversion. MOFs are typically
constructed by connecting metal/metal clusters with organic linkers to afford an extended two- or
three- dimensional structure. Such ordered structural motif provides a perfect platform for the
heterogenization of molecular catalysts, where well-defined catalytic sites can be introduced as
building blocks to construct the framework scaffold. Metal site isolation can improve the stability
of the catalyst by limiting bimolecular decomposition, while preserving the well-defined molecular
nature of the active sites. Additionally, many unique properties of MOFs are also very attractive
for catalytic applications. For instance, the large surface area and permanent porosity can promote
high catalyst loading and active site accessibility. The hybrid nature of MOFs provides abundant
opportunities for catalyst design and modification. However, conventional MOFs exhibit very low
electrical conductivity (below 10
-10
S cm
-1
), because they are typically constructed by hard metal
ions and redox-inactive ligands, which provides no low-energy pathway for charge transport nor
any free charge carriers.
13–15
This characteristic limits their application in electrocatalysis, which
requires the efficient electron transfer toward the active sites.
11
Many novel design strategies have been developed in recent years to overcome the low
conductivity of MOFs while maintaining their porosity and high surface area. Generally, these
strategies focus on offering low-energy charge transport pathways and/or increasing the number
of free charge carriers, and have resulted in MOFs with room-temperature conductivity that is up
to thirteen orders of magnitude higher than conventional MOFs.
15
Among all the conductive MOFs
reported so far, a special sub-class of two-dimensional (2D) π-conjugated MOFs exhibit some of
the highest conductivities.
15,16
This family of 2D MOFs are typically constructed by benzene- or
triphenylene-derived ligands with ortho-disubstituted S, O, or N atoms that form a square-planar
coordination geometry with a single transition metal node. Their remarkably high conductivity
originates from the extended 2D π conjugation that offers low energy pathway for through-bond
conductivity within the 2D sheet, as well as the π-π stacking interaction that allows for through-
space conductivity between the 2D sheets.
Figure 1.4 Examples of conducive 2D MOFs. (a) ref 17 (b) ref 18 (c) ref 19 (d) ref 18
While conductivity is an important factor for achieving high electrocatalytic performance,
recent studies have also suggest that other factors, such as the kinetics of the redox center, as well
as the diffusion of reactants/products and counter ions can also control the overall rate of catalysis,
as many MOF-based electrocatalysts rely on a redox hopping charge transfer mechanism.
20
Therefore, to realize a MOF catalyst with industrial relevance, efforts should be devoted to not
only optimizing the chemical environment of the individual catalytic sites to engineer high
12
turnover frequencies, but also to improve the bulk properties of the entire framework to allow for
efficient mass and charge transportation.
1.6 Dithiolate-Based Metal-Organic Frameworks for Hydrogen Evolution
Molecular metal dithiolene complexes have been extensively explored for efficient
electrocatalytic and photoelectrocatalytic HER.
21–24
The high HER activity of these complexes
originates from the non-innocent, redox active nature of the dithiolene ligands. Specifically,
dithiolene complexes featuring cobalt as the metal center are suggested to have significant metal-
ligand mixing in the frontier orbital, allowing for multielectron transfer processes through metal-
ligand cooperativity.
25
Eisenberg and coworkers have performed extensive studies on
[Co(bdt)2][TBA] (where bdt = benzene-1,2-dithiolate, TBA = tetrabutylammonium) as an active
electrocatalyst and photocatalyst in mixed aqueous/organic media.
21
The catalyst is capable of
performing electrocatalytic HER with near unity faradaic efficiency and high turnover numbers.
When combined with a photosensitizer such as Ru(bpy)3
2+
or CdSe quantum dots, efficient
photoelectrocatalytic generation of H2 can also be achieved.
21,22
In view of the outstanding HER activity of molecular dithiolene complexes, recent efforts have
been devoted to immobilizing these molecular complexes to realize the transition from
homogeneous to heterogeneous catalysis. In the Marinescu group, we have successfully integrated
the cobalt dithiolene catalytic units into a one-dimensional (1D) coordination polymer by using
the benzenetetrathiolate ligand (BTT),
26
and into extended 2D framework structures using
trinucleating ligand scaffolds, such as benzenehexathiolate (BHT) or triphenylene-2,3,6,7,10,11-
hexathiolate (THT) (Figure 1.5).
27
The as-synthesized cobalt dithiolene MOFs exhibit high HER
activity under fully aqueous media and near unity faradaic efficiency, representing the first
example of a MOF displaying high intrinsic electrocatalytic activity.
84
Subsequent work by Feng
13
and coworkers investigated the HER performance of a nickel dithiolene 2D MOF as well as a
series of metal dithiolene-diamine 2D MOFs, all of which display promising activities toward the
HER.
28,29
While these preliminary studies have demonstrated the great potential of metal
dithiolate-based 2D MOFs for electrocatalytic application, many challenges still remain in this
field of research. For example, there is minimal knowledge about the catalytic mechanism on these
catalysts, prohibiting the rational improvement of the catalytic performance. Additionally, due to
the top-down catalyst deposition method, these catalytic systems typically lack a robust
electrode/catalyst architecture, which leads to a compromised apparent activity. Lastly, taking
advantage of MOF’s high synthetic tunability, the library of the dithiolene MOF could to be further
expanded by altering the metal and the ligand identities through novel synthetic modifications.
Figure 1.5 Cobalt dithiolene-based MOFs (CoBHT and CoTHT) formed from the reaction of Co
2+
and the
trinucleating ligands, benzenehexathiolate (BHT) or triphenylene-2,3,6,7,10,11-hexathiolate (THT).
1.7 Outline of This Work
The broad goal of this work is to develop efficient HER electrocatalysts based on dithiolate
MOFs, and this goal is approached from two different angles. The first is to optimize the HER
performance of a known dithiolene-based MOF, the cobalt triphenylene-2,3,6,7,10,11-
hexathiolate (THT) MOF, by unraveling the reaction mechanism and identifying the key factors
that dictate the overall catalytic performance, which will be discussed in detail in Chapter 2. The
second approach is to explore the HER performance of new dithiolene-based MOFs by altering
the metal and the ligand identities. Chapter 3 investigates the role of metal centers in the HER
M = Co, Ni, Fe
M
S
S
S
M
S
S
M
S
MTHT
or
M
S
S
MBHT
S
M
S
S
M
S
2H
+
+ 2e
-
H
2
14
activity, where a series of iron and cobalt/iron mixed metal dithiolene MOFs were synthesized and
electrochemically characterized. In Chapter 4, a conductive 3D dithiolene-based MOF, the
Cu[Ni(2,3-pyrazinedithiolate)2] MOF, is investigated as an HER electrocatalyst for the first time.
Lastly, Chapter 5 will present the synthesis and HER characterization of a diselenolate-based MOF,
an analogous derivative of the dithiolate MOFs. This is to investigate the role of the chalcogen
moiety of the ligand, which is inspired by previous studies that suggest the superior HER activity
of the cobalt diselenolate complex and polymer compared to the dithiolate analogues.
30,31
Several figures of merit will be used to benchmark the HER activity of the catalyst of interest,
and are introduced herein. First is the onset of catalysis, which represents the potential where a
drastic increase in current density is observed and marks as the onset of the catalytic reaction.
Second is the overpotential to reach a current density of 10 mA/cm
2
, which is the approximate
current density expected for a 10% efficient solar-to-fuels conversion device under 1 sun
illumination.
32
Overpotential (η) is the difference between the thermodynamic standard reduction
potential and the actual operating potential, which represents the additional driving forces needed
to overcome the kinetic hindrances experienced in the electrochemical process.
Tafel slope, a measurement of the current response to the applied potential, provides
information associated with the rate-determining steps and is also used as a kinetic descriptor of
the catalyst.
33
It is typically obtained through a Tafel plot, which is generated by replotting the
polarization curve as a plot of log(j) vs. η with the slope of the linear portion defined as the Tafel
slope and expressed as follows:
𝑑 𝑙𝑜𝑔(𝑗)
𝑑 𝜂
= 2.303𝑅𝑇/𝛼𝑛𝐹
While R (ideal gas constant), T (temperature), F (Faraday constant), and n (the number of
transferred electrons) are all constants, α, which represents the charge transfer coefficient, is
15
characteristic to a given catalyst. As the Tafel slope is inversely proportional to α, a small Tafel
slope should indicate a high charge transfer ability and is hence preferable. Tafel slope also
provides important implications for reaction mechanism. For HER under acidic conditions, the
reaction proceeds with two possible mechanisms. The first step, for either of the mechanisms, is
always the Volmer discharging reaction, where a surface adsorbed hydrogen (Hads) is formed.
Depending on the surface coverage of Hads, the next step could either be the electrochemical
desorption of H2, known as the Heyrovsky reaction, or the chemical desorption of H2, also referred
to as the Tafel reaction. For the Heyrovsky reaction, the Hads reacts with a proton and an electron
from solution to evolve H2, whereas for the Tafel step, two adjacent Hads join together to evolve
H2. The theoretical Tafel slopes of the above-mentioned elementary steps are calculated to be 120
mV/dec, 40 mV/dec, and 30 mV/dec, for Volmer, Heyrovsky, and Tafel reaction, respectively.
33
Comparison between the experimentally determined Tafel slope and the theoretical ones can help
to elucidate the HER mechanism as well as the rate-determining step.
Table 1.1 Elementary steps for the HER and the corresponding theoretical Tafel slope
Step Reaction Theoretical Tafel Slope (mV/dec)
Volmer reaction 𝐻
!
(𝑎𝑞)+𝑒
"
⟶ 𝐻
#$%
120
Heyrovsky reaction 𝐻
#$%
+𝑒
"
+𝐻
!
⟶ 𝐻
&
40
Tafel reaction 𝐻
#$%
+𝐻
#$%
⟶ 𝐻
&
30
Finally, the efficiency of hydrogen production is described by the faradaic efficiency (FE),
which is calculated by comparing the H2 produced (determined by gas chromatography) with the
charge supplied during an electrolysis experiment. The current response over a long period of
controlled potential electrolysis is used to evaluate the stability of the catalyst under operating
conditions, which is also an important parameter when analyzing the catalytic performance.
16
1.8 References
(1) NASA. A Degree of Concern: Why Global Temperatures Matter – Climate Change: Vital
Signs of the Planet. NASA’s Global Climate Change Website. 2019.
(2) US Energy Protection Agency. Global Greenhouse Gas Emissions Data | Greenhouse Gas
(GHG) Emissions | US EPA. United States Environmental Protection Agency. 2014.
(3) IPCC, I. G. P. for C. C. Global Warming of 1.5°C. An IPCC Special Report on the Impacts
of Global Warming of 1.5°C above Pre-Industrial Levels and Related Global Greenhouse Gas
Emission Pathways, in the Context of Strengthening the Global Response to the Threat of Climate
Change,; 2018.
(4) Agora Energiewende and Ember. The European Power Sector in 2020: Up-to-Date
Analysis on the Electricity Transistion; 2021.
(5) IEA (International Energy Agency). Renewables 2020; 2020.
(6) REN21. Renewables 2020 Global Status Report; 2020.
(7) She, Z. W.; Kibsgaard, J.; Dickens, C. F.; Chorkendorff, I.; Nørskov, J. K.; Jaramillo, T. F.
Combining Theory and Experiment in Electrocatalysis: Insights into Materials Design. Science
2017, 355 (6321), eaad4998.
(8) LeValley, T. L.; Richard, A. R.; Fan, M. The Progress in Water Gas Shift and Steam
Reforming Hydrogen Production Technologies – A Review. Int. J. Hydrogen Energy 2014, 39
(30), 16983–17000.
(9) Santos, D. M. F.; Sequeira, C. A. C.; Figueiredo, J. L. Hydrogen Production by Alkaline
Water Electrolysis. Quim. Nova 2013, 36 (8), 1176–1193.
(10) U.S. Department of Energy. Hydrogen Production: Electrolysis
https://www.energy.gov/eere/fuelcells/hydrogen-production-electrolysis.
(11) McKone, J. R.; Marinescu, S. C.; Brunschwig, B. S.; Winkler, J. R.; Gray, H. B. Earth-
Abundant Hydrogen Evolution Electrocatalysts. Chem. Sci. 2014, 5 (3), 865–878.
(12) Bullock, R. M.; Das, A. K.; Appel, A. M. Surface Immobilization of Molecular
Electrocatalysts for Energy Conversion. Chem. Eur. J. 2017, 23, 7626–7641.
(13) Stassen, I.; Burtch, N.; Talin, A.; Falcaro, P.; Allendorf, M.; Ameloot, R. An Updated
Roadmap for the Integration of Metal–Organic Frameworks with Electronic Devices and Chemical
Sensors. Chem. Soc. Rev. 2017, 46 (11), 3185–3241.
(14) Morozan, A.; Jaouen, F. Metal Organic Frameworks for Electrochemical Applications.
Energy Environ. Sci. 2012, 5 (11), 9269.
(15) Sun, L.; Campbell, M. G.; Dincă, M. Electrically Conductive Porous Metal-Organic
17
Frameworks. Angew. Chem., Int. Ed. 2016, 55 (11), 3566–3579.
(16) Kusamoto, T.; Nishihara, H. Zero-, One- and Two-Dimensional Bis(Dithiolato)Metal
Complexes with Unique Physical and Chemical Properties. Coord. Chem. Rev. 2019, 380, 419–
439.
(17) Sheberla, D.; Sun, L.; Blood-Forsythe, M. A.; Er, S.; Wade, C. R.; Brozek, C. K.; Aspuru-
Guzik, A.; Dincă, M. High Electrical Conductivity in Ni3(2,3,6,7,10,11-Hexaiminotriphenylene)2,
a Semiconducting Metal–Organic Graphene Analogue. J. Am. Chem. Soc. 2014, 136 (25), 8859–
8862.
(18) Aubrey, M. L.; Kapelewski, M. T.; Melville, J. F.; Oktawiec, J.; Presti, D.; Gagliardi, L.;
Long, J. R. Chemiresistive Detection of Gaseous Hydrocarbons and Interrogation of Charge
Transport in Cu[Ni(2,3-Pyrazinedithiolate)2] by Gas Adsorption. J. Am. Chem. Soc. 2019, 141 (12),
5005–5013.
(19) Clough, A. J.; Skelton, J. M.; Downes, C. A.; de la Rosa, A. A.; Yoo, J. W.; Walsh, A.;
Melot, B. C.; Marinescu, S. C. Metallic Conductivity in a Two-Dimensional Cobalt Dithiolene
Metal–Organic Framework. J. Am. Chem. Soc. 2017, 139 (31), 10863–10867.
(20) Lin, S.; Usov, P. M.; Morris, A. J. The Role of Redox Hopping in Metal–Organic
Framework Electrocatalysis. Chem. Commun. 2018, 54 (51), 6965–6974.
(21) Das, A.; Han, Z.; Haghighi, M. G.; Eisenberg, R. Photogeneration of Hydrogen from Water
Using CdSe Nanocrystals Demonstrating the Importance of Surface Exchange. Proc. Natl. Acad.
Sci. 2013, 110 (42), 16716–16723.
(22) McNamara, W. R.; Han, Z.; Alperin, P. J.; Brennessel, W. W.; Holland, P. L.; Eisenberg,
R. A Cobalt–Dithiolene Complex for the Photocatalytic and Electrocatalytic Reduction of Protons.
J. Am. Chem. Soc. 2011, 133 (39), 15368–15371.
(23) Lv, H.; Ruberu, T. P. A.; Fleischauer, V. E.; Brennessel, W. W.; Neidig, M. L.; Eisenberg,
R. Catalytic Light-Driven Generation of Hydrogen from Water by Iron Dithiolene Complexes. J.
Am. Chem. Soc. 2016, 138 (36), 11654–11663.
(24) Zarkadoulas, A.; Field, M. J.; Artero, V.; Mitsopoulou, C. A. Proton-Reduction Reaction
Catalyzed by Homoleptic Nickel-Bis-1,2-Dithiolate Complexes: Experimental and Theoretical
Mechanistic Investigations. ChemCatChem 2017, 9 (12), 2308–2317.
(25) Ray, K.; Begum, A.; Weyhermüller, T.; Piligkos, S.; van Slageren, J.; Neese, F.; Wieghardt,
K. The Electronic Structure of the Isoelectronic, Square-Planar Complexes [Fe
II
(L)2]
2-
and
[Co
III
(LBu)2]
-
(L
2-
and (LBu)
2-
= Benzene-1,2-Dithiolates): An Experimental and Density
Functional Theoretical Study. J. Am. Chem. Soc. 2005, 127 (12), 4403–4415.
(26) Downes, C. A.; Marinescu, S. C. Efficient Electrochemical and Photoelectrochemical H2
Production from Water by a Cobalt Dithiolene One-Dimensional Metal–Organic Surface. J. Am.
Chem. Soc. 2015, 137 (43), 13740–13743.
18
(27) Clough, A. J.; Yoo, J. W.; Mecklenburg, M. H.; Marinescu, S. C. Two-Dimensional Metal–
Organic Surfaces for Efficient Hydrogen Evolution from Water. J. Am. Chem. Soc. 2015, 137 (1),
118–121.
(28) Dong, R.; Pfeffermann, M.; Liang, H.; Zheng, Z.; Zhu, X.; Zhang, J.; Feng, X. Large-Area,
Free-Standing, Two-Dimensional Supramolecular Polymer Single-Layer Sheets for Highly
Efficient Electrocatalytic Hydrogen Evolution. Angew. Chem., Int. Ed. 2015, 54 (41), 12058–
12063.
(29) Dong, R.; Zheng, Z.; Tranca, D. C.; Zhang, J.; Chandrasekhar, N.; Liu, S.; Zhuang, X.;
Seifert, G.; Feng, X. Immobilizing Molecular Metal Dithiolene-Diamine Complexes on 2D Metal-
Organic Frameworks for Electrocatalytic H2 Production. Chem. Eur. J. 2017, 23 (10), 2255–2260.
(30) Downes, C. A.; Marinescu, S. C. Bioinspired Metal Selenolate Polymers with Tunable
Mechanistic Pathways for Efficient H2 Evolution. ACS Catal. 2017, 7 (1), 848–854.
(31) 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 (53), 7306–7309.
(32) McCrory, C. C. L.; Jung, S.; Ferrer, I. M.; Chatman, S. M.; Peters, J. C.; Jaramillo, T. F.
Benchmarking Hydrogen Evolving Reaction and Oxygen Evolving Reaction Electrocatalysts for
Solar Water Splitting Devices. J. Am. Chem. Soc. 2015, 137 (13), 4347–4357.
(33) Shinagawa, T.; Garcia-Esparza, A. T.; Takanabe, K. Insight on Tafel Slopes from a
Microkinetic Analysis of Aqueous Electrocatalysis for Energy Conversion. Sci. Rep. 2015, 5 (1),
13801.
19
Chapter 2
Improving and Understanding the Hydrogen Evolving Activity of a Cobalt
Dithiolene Metal-Organic Framework
A portion of this chapter has appeared in print:
Chen, K.; Downes, C. A.; Schneider, E.; Goodpaster, J. D.; Marinescu, S. C. “Improving and
Understanding the Hydrogen Evolving Activity of a Cobalt Dithiolene Metal-Organic
Framework”. ACS Appl. Mater. Interfaces 2021, 13 (14), 16384-16395.
Reprinted with permission from ACS Appl. Mater. Interfaces 2021, 13 (14), 16384-16395. Copyright 2021 American
Chemical Society.
20
2.1 Introduction
In response to the increasing global energy demand and the threat of climate change caused by
the consumption of fossil fuels, scientists have extensively explored clean energy alternatives to
fossil fuels.
1
One of the most promising candidates is hydrogen (H2) because of its high gravimetric
energy density and carbon-free combustion product.
2,3
Currently, the majority of H2 is industrially
produced from steam-methane reforming, a process that is extremely energy-intensive and releases
a large amount of greenhouse gases.
4
Electrocatalytic water splitting is a more sustainable route as
it simply requires the combination of water and electricity to produce H2, where the electricity can
be generated from renewable energy sources such as solar energy.
5
However, this route remains
thwarted by the kinetic bottleneck that arises from the multi-electron nature of both the reductive
and oxidative processes involved in the water splitting reaction, i.e. the hydrogen evolution
reaction (HER) and the oxygen evolution reaction (OER).
6
Therefore, the development of efficient
electrocatalysts is required to overcome the significant activation barriers associated with water
splitting. To date, the best-performing catalysts for the HER are based on platinum, which is too
scarce and expensive for practical scalability needs. Hence, to achieve global deployment of these
clean energy technologies, the design and implementation of catalysts that can function under
benign conditions and comprise only inexpensive, earth-abundant elements is paramount.
7,8
Metal-organic frameworks (MOFs) have recently emerged as a promising class of materials
for electrocatalytic applications,
9–14
such as the HER,
15–17
OER,
18–21
oxygen reduction reaction
(ORR),
22–27
and CO2 reduction reaction (CO2RR).
28–34
The hybrid nature of MOFs offers great
tunability for catalyst design and the high surface areas and permanent porosity of these materials
allow for substrate diffusion to the catalytic active sites. Some of the earliest examples of
electrocatalytic MOFs are two-dimensional (2D) dithiolene-based, where well-defined metal-
21
dithiolene catalytic units are integrated into extended 2D frameworks through conjugated
trinucleating ligands such as benzenehexathiol (BHT) and triphenylene-2,3,6,7,10,11-hexathiol
(THT) (Figure 2.1). These 2D dithiolene-based MOFs, which employ non-precious metal centers
such as cobalt, nickel, or copper, have successfully demonstrated electrocatalytic HER activity in
acidic aqueous media (Table 2.1).
15–17,35–37
Additional reports of one-dimensional (1D) thiolate-
based coordination polymers (CPs) have also indicated promising electrocatalytic HER activity.
38–
43
In an early example, we have explored the HER performance of the 2D cobalt hexathiolate
systems, CoBHT and CoTHT, which display overpotentials of 340 mV and 530 mV, respectively,
to achieve a current density of 10 mA/cm
2
in pH 1.3 aqueous solutions.
15
This robust performance
has prompted ensuing studies and improvement of the materials. For the CoBHT system, the
overpotential is identified as dependent on the thickness of the deposited catalyst film with the
optimal film thickness generating overpotentials as low as 185 mV.
35
For the CoTHT system,
work by Dong and coworkers has illustrated that an ink composite of CoTHT (labeled as CoTHT
(powder) in Table 2.1) displays an overpotential of 496 mV to reach 10 mA/cm
2
.
17
The same
authors have also shown that the fabrication of a monolayer film further reduces the overpotential
to 323 mV. The reduction in overpotential for the thin film in comparison to the bulk powder is
most likely attributed to enhanced charge transport properties as 2D MOFs commonly exhibit
improved conductivities when transitioning from powders to thin films because of a reduction in
the number of grain boundaries.
44,45
However, the monolayer film of CoTHT still operates at a
large overpotential and the necessity of a monolayer to see minimal improvement in performance
is not ideal for practical and scalable energy-converting devices. Therefore, it is necessary to
understand and further improve the activity of the bulk CoTHT material through alternative
methods.
22
Figure 2.1. Structural schematics of the dithiolene-based 2D MOFs containing trinucleating ligands, such as
triphenylene-2,3,6,7,10,11-hexathiolate (THT) and benzenehexathiolate (BHT).
One of the bottlenecks of the extensive assessment of the intrinsic catalytic activity of the 2D
dithiolene-based MOFs resides in the immobilization of the material. Typically, the material is
synthesized through an interfacial reaction followed by a “top-down” deposition method, by
directly immersing the electrode into the reaction mixture or drop casting the film onto the
electrode substrate. This “top-down” deposition method is facile but yields irreproducible results,
which was extensively discussed in the study of a cobalt anthracenetetrathiolate coordination
polymer, where a large variance in catalytic responses was observed for the same batch of
material.
41
This irreproducibility is explained by the differences in bulk loading, the number of
electrochemically accessible active sites, the electron transfer process, and the electrical
integration between the catalyst and the support, which are all introduced by the poor properties
of the catalyst/electrode architecture constructed through the “top-down” deposition method.
Study with the CoBHT system also suggests that the as-deposited films invariably display
significant cracking, which appears more severe for thicker films.
35
Therefore, the “top-down”
deposition method hinders the expression and accurate measurement of the intrinsic activity of the
material.
In addition to the poor deposition techniques for dithiolene-based MOFs that limit the
determination of their intrinsic activities, the lack of detailed information regarding their
M = Co, Ni, Cu
M
S
S
S
M
S
S
M
S
MTHT
or
M
S
S
MBHT
S
M
S
S
M
S
23
mechanisms for the HER also hinders our understanding of these materials. The mechanisms of
analogous cobalt bisdithiolene molecular complexes, however, have been explored
computationally. The work presented by Hammes-Schiffer and coworkers suggests that, during
the catalytic cycle, [Co(bdt)2]
-
(bdt = 1,2-benzenedithiolate) undergoes one-electron reduction,
followed by two protonation events occurring on two different sulfur atoms.
46
Upon the second
one-electron reduction, an intramolecular proton shift takes place, leading to the formation of a
Co
III
-hydride intermediate, which can couple with the adjacent SH moiety to produce H2 and
regenerate [Co(bdt)2]
-
.
46
The presence of different substitutes on the aryl moiety leads to a different
sequence of reduction and protonation events.
46
More recently, computational studies performed
on cobalt-bis(diaryldithiolene) species suggest that the first proton transfer event occurs on the
cobalt center, rather than the ligand, to generate a Co
III
-hydride directly. Upon the second reduction
and protonation event, the formation of H2 proceeds via a homocoupling mechanism, where a
cobalt-H2 intermediate is involved.
47,48
These rigorous theoretical investigations to deduce the mechanism of H2 evolution by cobalt
bisdithiolene complexes are insightful but should not be used solely to elucidate the mechanism
of the HER with dithiolene-based MOFs. Mechanisms that are unavailable for molecular
dithiolene complexes might be accessible to dithiolene-based MOFs because of their unique
extended structures. Although extensive computational studies have not been undertaken to
evaluate possible mechanistic pathways for the HER with dithiolene-based MOFs, DFT studies
were recently reported assessing and identifying the most favorable site for H adsorption. For
CoTHT, DFT calculations reported by Dong et al. suggest that the 1H adsorption on the Co site
is slightly thermodynamically favored over the S site (-2.25 eV vs. -2.10 eV, respectively).
17
Additionally, DFT studies performed by Huang et al. on CuBHT indicate that the lowest Gibbs
24
free energy for hydrogen adsorption is found when the H atom adsorbs on the “Cu-edge site” on
the (100) surface of CuBHT nanoparticles.
36
For one-dimensional cobalt dithiolene coordination
polymers, the Gibbs free energy of the adsorption of hydrogen atoms identified the S atoms as the
preferred catalytic site for the HER.
49
Additionally, the activity of the S atoms can be improved by
interacting with the alkali metal cations from the electrolytes.
Table 2.1. Summary of the HER activity of dithiolene-based MOFs
Catalyst
Onset η
(mV)
η (mV)
@10mA/cm
2
Tafel Slope
(mV/dec)
Condition Ref.
CoTHT (ink) 61 143
71
pH 1.3
This
work
CoTHT (multi-layer) 403 530
161
a
pH 1.3 15
CoTHT (single-layer) 110 323
82
0.5 M H
2
SO
4
17
CoTHT (powder) 139 496 157
0.5 M H
2
SO
4
CoTHT/graphene 99 426 137
0.5 M H
2
SO
4
CoTHTA
b
(single-layer, 0.8 nm)
92 283 71
0.5 M H
2
SO
4
NiTHTA (single-layer) 107 315 76
0.5 M H
2
SO
4
NiTHT
(single-layer, 0.7 nm)
110 333 80.5
0.5 M H
2
SO
4
16
CoBHT (film, 244 nm) - 185 88 pH 1.3 35
CuBHT (nanoparticle) 200 450 95 pH 0.0 36
NiAT
c
150 370 128 pH 1.3 37
[a] Tafel slope measured in pH 2.6; [b] THTA stands for a mixture of THT ligand and THA (triphenylenehexamine)
ligand; [c] NiAT stands for bis(aminothiolato)nickel.
Despite these promising previous reports on the electrocatalytic HER activity of cobalt, nickel,
and copper hexathiolate frameworks, structure-activity studies cannot be accurately performed due
to the poor reproducibility of the HER performance of these materials, caused by the poor
properties of the MOF-modified electrode architectures. As mentioned above, there are simply too
many variables at play – differences in bulk loading, the number of electrochemically accessible
active sites, the electron transfer process, and the electrical integration between the catalyst and
25
the electrode substrate. Thus, the development of a method to more accurately measure the HER
activity of these frameworks, by overcoming the previously highlighted limitations, is paramount.
To address the aforementioned problems, we adopt a deposition method that employs an ink
composite (1) comprised of CoTHT, carbon black, and Nafion. This deposition method has often
been used for the preparation of electrodes with MOF-based electrocatalysts.
17,24,43
Carbon black,
an inexpensive conductive additive, is necessary to improve the electrical conductivity of the
composite and the proton conductive polymeric binder Nafion increases the integration between
the composite and the electrode. Such simple modifications to the bulk material result in a drastic
enhancement of the HER performance compared to all previous reports on the electrocatalytic
activity of metal dithiolene-based MOFs (Table 2.1). This result highlights the high intrinsic
activity of CoTHT, which was previously overshadowed by the poor bulk properties of the
catalyst/electrode architecture.
Additionally, a series of extensive density functional theory (DFT) calculations were
performed to explore the HER mechanism for CoTHT for the first time. To achieve more accurate
modeling of the extended 2D framework, the computational hydrogen electrode (CHE) is applied,
which has been used with great success in modeling HER for extended heterogeneous systems.
50,51
The calculation results suggest that the CoTHT sheets can adopt two different stacking geometries,
referred to as the AA and the AB stacking modes, with the latter displaying lower free energy. Due
to the lack of the experimental local structural evidence and the possibility of the co-existence of
the two stacking geometries, the HER mechanistic pathways were calculated on both stacking
configurations. The results suggest that, regardless of the stacking geometry or the applied
functional, the HER in CoTHT MOF resembles more closely to that of the cobalt-
bis(diaryldithiolene) species, where the proton/electron transfers occur more favorably at the
26
cobalt centers. The calculated onset potential of catalysis is in good agreement with the
experimentally observed value, supporting the reliability of the calculated mechanistic pathways.
2.2 Results and Discussion
2.2.1 Optimization of the Electrochemical HER Activity of CoTHT
Previously reported electrochemical studies of CoTHT utilize the direct transfer of the as-
synthesized 2D film by immersing the electrode into the reaction mixture.
15
In an attempt to
improve the crystallinity of the material, we modified the synthetic method from a gas-liquid
interfacial reaction to a liquid-liquid interfacial reaction, yet the film grown this way inhibits
deposition of the material on to the electrode using the previous method. It is feasible to deposit
the material through an alternative “top-down” method by directly drop casting the as-synthesized
films onto the electrode substrate. However, the electrodes fabricated from this immobilization
method yield a wide variance in catalytic and electrochemical impedance responses. As shown in
Figure 2.2, three electrodes modified with the same batch of material display an onset potential of
catalysis ranging from -0.12 V (GCE2) to -0.45 V (GCE3) with a charge transfer resistance (Rct)
at -0.47 V (vs. RHE) ranging from 1867 Ω (GCE1) to 4424 Ω (GCE3). This irreproducibility is
mainly ascribed to the following problems associated with the deposition method: poor control
over bulk loading, poor electrical integration between the electrode and the material, and
inhomogeneous distribution of the material on the electrode surface. The weak physical contact
also leads to delamination of the material from the electrode during electrochemical testing,
rendering a low apparent long-term stability. The poor properties of the catalyst-modified
electrodes introduced by the “top-down” deposition method prevent an adequate understanding of
the catalytic activity of the material. Therefore, a better deposition method is necessary in order to
adequately evaluate the intrinsic activity of CoTHT.
27
Figure 2.2. (a) Cyclic voltammograms (CVs) of three individual glassy carbon electrodes (GCEs) modified with the
same batch of CoTHT film, scan rate: 100 mV/s. (b) Nyquist plots (markers) with respective fits (solid lines) measured
at -0.47 V vs. RHE. All measurement were performed in N2-satuarted pH 1.3 aqueous solutions.
To this end, a variety of ink compositions were explored starting with a CoTHT/Nafion
mixture. Nafion is chosen as the polymeric binder due to its high proton conductivity. The mixture
was fabricated as a suspension in water and ethanol through sonication. The bulk loading of the
catalyst can be readily controlled by drop casting a specific volume of the suspension using a
micro-syringe. The electrodes prepared this way yield more reproducible results, accompanied by
a slight decrease in Rct, as the material is distributed more homogeneously on the electrode surface
(Figure 2.3). However, the absolute value of Rct is still relatively high (~900 Ω at -0.47 V vs. RHE),
partially owing to the poor electrical conductivity of CoTHT. The reported room-temperature
electrical conductivity of the as-prepared CoTHT film is approximately 3.2 × 10
-2
S·cm
-1
, which
is further reduced to 1.4 × 10
-3
S·cm
-1
when the material is pressed into a pellet due to the
introduction of a large number of grain boundaries.
45
Therefore, to ensure an adequate conductivity
for an efficient HER, and to investigate the intrinsic activity of CoTHT, the catalyst composite
was further modified by the addition of carbon black (Vulcan XC-72, CB) to make a conductive
ink composite 1. Dong and coworkers previously reported an analogous ink material (labeled as
CoTHT (powder) in Table 2.1); however, the material was synthesized through a different method
(a hydrothermal method where a suspension of Co(NO3)2, THT, and aqueous ammonia (NH4OH)
28
was heated to 120 °C in an autoclave) and displayed a modest overpotential to reach 10 mA/cm
2
(496 mV).
17
Figure 2.3. (a) CVs of GCE_1 (red) and GCE_2 (blue), prepared by drop casting 10 uL of the CoTHT/Nafion mixture
onto the GCE surface, scan rate: 100 mV/s; (b) Nyquist plots (markers) with respective fits (solid lines) measured at
-0.468 V vs. RHE. All measurements were performed in N2-saturated pH 1.3 aqueous solutions.
Scanning electron microscopy (SEM) images of 1 reveal the bulk morphology of the composite
(Figure 2.4a), with sheet-like morphology of the CoTHT film identified at higher magnification
(Figure 2.4b). The integrity of the structure is confirmed by powder X-ray diffraction (PXRD),
where similar peaks are observed compared to the pristine CoTHT (Figure 2.5), though the
intensity is diminished. Preliminary static cyclic voltammogram (CV) studies of 1 (Figure 2.6a)
show a drastic increase in the achievable current density (~ 90 mA/cm
2
at -0.23 V vs. RHE), which
is associated with the catalytic reaction induced by the active species, CoTHT, as the control
experiment with only CB and Nafion does not show any appreciable activity within the
experimental potential window. The same electrode is also subjected to 10 successive CV cycles
(Figure 2.6b) and no visual loss of material or apparent drop in activity is observed within 10
cycles. The dramatic improvement in activity is likely resulted from the electronic communication
between CoTHT and CB that facilitates the electron transport process, which is an integral
component of the resultant HER activity. Indeed, the electrochemical impedance spectroscopy
(EIS) measurement reveals that the Rct is significantly reduced to 34 Ω at -0.17 V vs. RHE (Figure
29
2.6c). However, a large capacitive charging of the material is observed due to the participation of
CB, and as the catalyst is effectively evolving H2, the data collected from a static CV setup is
extremely noisy. Therefore, in order to accurately measure the catalytic response of 1, a glassy
carbon rotating disk electrode (RDE) was used as the substrate for subsequent electrochemical
tests. Linear sweep voltammograms (LSVs) with a scan rate of 5 mV/s under a rotation speed of
1600 rpm were used to characterize the electrocatalytic activity of the material. The onset potential
of catalysis, overpotential (η) to reach a current density of 10 mA/cm
2
, and Tafel slope extracted
from LSV are the three metrics used to evaluate the performance. EIS measurements were also
conducted to understand the charge transfer processes during catalysis. All measurements were
performed in N2-saturated pH 1.3 H2SO4 aqueous solutions at room temperature, unless otherwise
stated.
Figure 2.4. Top down SEM images of 1
Figure 2.5. Synchrotron PXRD pattern of pristine CoTHT (red) and 1 (blue), wavelength: 0.412750Å.
30
Figure 2.6. (a) CVs of one GCE deposited with only CB and Nafion and three GCEs prepared by drop casting 10 uL
of ink mixture 1 onto the electrode surface; (b) CVs of GCE_1 subjected to 10 repetitive CV cycles; (c) Nyquist plot
(markers) of GCE_1 with fit (solid line) measured at -0.168 V vs. RHE, Rct = 34.6 𝛺. Scan rate for CVs: 100 mV/s.
All measurements were performed in N2-saturated pH 1.3 aqueous solutions.
To establish a mixture that can yield the optimal activity, ink composites with variable CB
contents were prepared and tested in the RDE setup. As shown in Figure 2.7a, the CoTHT/Nafion
mixture (0 wt% CB) displays nominal activity with a catalytic onset potential of -474 mV (vs.
RHE). The activity of the catalyst is substantially enhanced by adding as little as 10 wt% of CB,
highlighted by a much earlier onset potential of -91 mV and an overpotential of 185 mV to achieve
10 mA/cm
2
. The Rct is drastically decreased from 2948 Ω to 55.1 Ω at -168 mV (Figure 2.7b),
suggesting a much faster charge transfer process. Upon increasing the amount of CB, a small
fluctuation in activity is observed, which reaches an apex at around 24 wt% (characteristic values
summarized in
Table 2.2). Although higher CB contents can further facilitate electron transport throughout
the material, it can also adversely dilute the concentration of the active species, CoTHT, resulting
31
in a higher Rct and diminished activity of the overall material. Tafel analysis reveals a substantial
decrease in the Tafel slope upon incorporation of CB (Figure 2.8), indicating a much favored HER
kinetics and a possible change in the mechanistic pathways. For CoTHT/Nafion mixture, a Tafel
slope of 150 mV/dec suggests the Volmer reaction could be the rate-limiting step, whereas for ink
composites with Tafel slopes in the range of 73.6 to 92.8 mV/dec, Volmer or Heyrovsky reaction
is likely the rate-limiting step.
52,53
Figure 2.7. (a) Polarization curves of a bare GCE and GCEs deposited with ink mixtures with variable carbon black
wt%, scan rate: 5 mV/s. (b) Nyquist plots (markers) of each electrode with respective fits (solid lines) measured
at -0.168 V vs. RHE. All measurements were performed in N2-saturated pH 1.3 H2SO4 solutions at room temperature.
Table 2.2. Characteristic values of ink mixtures with variable CB wt%
wt% of carbon black 0 10 24 32 50
Onset (mV vs. RHE) -474 -91.0 -73.0
-88.5
-83.8
η (mV) @ 10 mA/cm
2
> 500 185 164
178
189
Tafel slope (mV/dec) 150 83.2 73.6
78.0
92.8
Rct (Ω) @ -168 mV 2948 55.1 27.9 42.9 48.1
Figure 2.8. Tafel plots of (a) CoTHT/Nafion mixture and (b) ink composites with variable CB wt%, extracted from
polarization curves presented in Figure 2.7.
32
Understanding the effect of catalyst loading on electrocatalytic activity is crucial for
determining the scalability and practicality of the catalyst for energy conversion applications.
Previously, we have observed a thickness-dependent HER activity for an analogous material,
CoBHT, where film thickness influences the apparent activity of the catalyst by affecting the
charge transfer and proton permeation processes.
35
Herein, we explore similar effects by tuning
the bulk loading of 1. As shown in Figure 2.9, an increase in catalyst loading from 3 µL to 20 µL
results in an enhancement in the overall activity: the overpotential decreases from >400 mV to as
low as 143 mV and the Tafel slope decreases from 101 mV/dec to 70.6 mV/dec. However, when
the loading is further increased to 25 μL, a slightly diminished performance is observed, suggesting
that the proton permeation through the thick catalyst layer becomes a limiting factor (Table 2.3).
In all, the optimized HER activity of 1 is described by a Tafel slope of 70.6 mV/dec, an onset
potential of -60.9 mV, and an η of 143 mV, which is only 100 mV more compared to Pt/C, the
industrial standard for the HER (Figure 2.9d).
Table 2.3. Characteristic values of electrodes deposited with variable amounts of 1.
Loading (µL) 3 5 10 15 20 25
Loading (10
-6
molCo/cm
2
)
a
0.32 0.48 1.15 1.32 1.91 1.96
Onset (mV vs. RHE) -162 -148 -130
-77.8
-60.9 -64.5
η (mV) @ 10 mA/cm
2
> 400 322 241
160
143 171
Tafel slope (mV/dec) 101 92.5 80.1 73.3 70.6 90.2
Rct (Ω) @ -168 mV 280 276 98.6 28.7 24.3 39.7
[a] Cobalt loading was determined by ICP-OES measurements.
33
Figure 2.9. Polarization curves of electrodes deposited with variable amount of 1 in N2-saturated pH 1.3 aqueous
solutions, scan rate: 5 mV/s. (b) Nyquist plots (markers) of electrodes deposited with variable amount of 1 with
respective fits (solid lines) measured at -0.168 V vs. RHE (c) Comparison of current density (mA/cm
2
) vs. loading
(μL) under different overpotentials. (d) Polarization curves of 1 (red) and 20% Pt/C (black) in N2-saturated pH 1.3
aqueous solutions, the inset shows the corresponding Tafel plot. Scan rate: 5 mV/s.
2.2.2 Electrochemical Characterization of 1
To obtain a physical picture of the electrode/electrolyte interface and the processes occurring
at the electrode surface, EIS measurements were performed over a range of overpotentials (68 mV
- 138 mV), and the respective Nyquist and Bode plots are presented in Figure 2.10. The Nyquist
plot for all overpotentials is dominated by a single semicircle at low frequency range (Figure 2.10a).
This low-frequency feature is diminished and shifted to higher frequencies at larger overpotentials,
as shown in the Bode phase plot (Figure 2.10b), suggesting that this feature is overpotential-
dependent and is related to the catalytic kinetics. There is also a small feature at very high
frequencies, which appears overpotential-independent in the Bode phase plot. Based on these
qualitative evaluations of the spectra, the EIS data was subsequently fitted with a two-time constant
34
serial model (2TS, Figure 2.10a inset), with the first time constant (Rct-CPE1, t1) corresponding to
the charge transfer kinetics and the second (R2-CPE2, t2) related to non-Faradaic origins such as
the porosity of the electrode surface or the interfacial resistance between the electrode and
catalyst.
41,54–56
The values obtained from this fit are summarized in Table 2.4. As expected, the
solution resistance (Rs ≈ 37 Ω) is not affected by varying overpotentials, whereas the Rct values
drastically decrease from 1230 Ω at η = 68 mV to 103 Ω at η = 138 mV. Moreover, the values of
Cdl (double-layer capacitance, extracted from EIS) remain relatively unchanged over the course of
the measurement, indicating the integrity of the electrochemically active surface area. Ideally, the
second time constant (t2) should not be affected by the change of overpotential, which contradicts
the fitted results. Admittedly, this high-frequency feature is not fitted well using the simple 2TS
model, indicating the occurrence of possible complex physical processes at the interface during
catalysis, such as morphological change. The charge-transfer Tafel slope is derived from the linear
fit of the plot of log(1/Rct) against potential.
37,55
As shown in Figure 2.11, the Tafel slope obtained
this way (64.4 mV/dec) is similar, yet smaller, than the one obtained from the polarization curve
(70.9 mV/dec), as the former excludes the contributions from the kinetic processes described by
t2.
Figure 2.10. (a) Nyquist and (b) Bode plots of 1 (markers) with respective fits (solid lines) measured at variable
overpotentials. All measurements were performed in N2-saturated pH 1.3 aqueous solutions.
35
Table 2.4. Values extracted from fitting the EIS data of 1 to the 2TS equivalent circuit
η (mV) 68 78 88 98 108 118 128 138
Rs (Ω) 36.6 36.6 36.6
36.6
36.7 36.8 36.9 36.2
CPE1 (mF∙s
n-1
) 13.2 13.3 13.5
13.6
13.9 14.0 13.9 13.7
n1 1.00 1.00 1.00
1.00
1.00 1.00 1.00 1.00
Rct (Ω) 1230 795 525 345 245 176 130 103
Cdl (mF)
a
13.2 13.3 13.5 13.6 13.9 14.0 13.9 13.7
CPE2 (mF∙s
n-1
) 29.2 29.8 13.5 32.5 38.0 35.1 34.6 32.4
n2 0.416 0.404 0.376 0.369 0.331 0.333 0.322 0.347
R2 (Ω) 32.7 35.3 46.5 49.5 170 138 150 77.6
[a] Cdl is calculated using the following equation:
57
𝐶
!"
= [
#$%
('
!
"#
( '
$%
"#
)
(#"')
]
*/,
Figure 2.11. Tafel plot obtained from EIS spectra (red) and polarization curve (blue).
The pH dependence of the catalytic performance of 1 was also tested. As the pH of the
electrolyte solution increases from 0.4 to 4.6, the HER activity progressively diminishes (Figure
2.12a), characterized by a delayed catalytic onset and an increase in the overpotential to reach 10
mA/cm
2
. The onset potential of the catalytic event shifts in a nearly Nernstian fashion by 57.5 mV
per pH unit (Figure 2.12b), suggesting the electrochemical reaction involves an equal number of
protons and electrons, as expected for the HER (2H
+
+ 2e
-
à H2). In pH 10 aqueous buffer
solutions (Figure 2.13a), a quasi-reversible feature is observed at E1/2 = -0.14 V vs. NHE (0.45 V
vs. RHE) with a relatively large peak-to-peak separation (∆Ep) of 56 mV at a scan rate of 5 mV/s,
36
indicating a sluggish charge transfer process at the electrode/electrolyte interface. The correlation
between the peak current (ip) and the scan rate (υ) can provide information about the nature of the
electron transfer event. For a perfectly surface-confined electron transfer process, ip is proportional
to υ, in contrast to the υ
1/2
dependence for diffusing species in the case of molecular catalysis.
58
The log |jp| - log υ plot derived from Figure 2.13a shows a slope between 0.5 and 1 (Figure 2.13b),
suggesting the redox feature observed here involves both a surface-confined and a diffusion-
limited process, which is likely related to the charge hopping within the catalyst layer.
59
Figure 2.12. (a) Polarization curves illustrating the pH-dependent HER activity of 1, scan rate: 5 mV/s; (b) Catalytic
onset potential as a function of pH, slope: 57.5 mV/dec.
Figure 2.13. (a) Scan rate (mV/s) dependence experiments of 1 in a N2-saturated pH 10 aqueous solution. (b) Plot of
log |jp| versus log(𝜐).
37
2.2.3 Electrocatalytic Stability of 1
Controlled potential electrolysis (CPE) studies of 1 on a glassy carbon plate electrode were
performed in a pH 1.3 H2SO4 solution at -0.19 V vs. RHE for 2 hours (Figure 2.14). Analysis of
the gas mixture in the headspace of the working compartment by gas chromatography confirms
the production of H2 with a Faradaic efficiency of 98%. However, a gradual decline in the
generated current is observed over the course of 2 hours (from 9.2 mA to 6.1 mA). Longer duration
CPE experiments were also performed (Figure 2.14) and the gradual decrease in activity was again
present. To further assess the long-term durability of 1, controlled current electrolysis (CCE,
Figure 2.15) and CV cycling experiments (Figure 2.16) were performed using the RDE setup. In
both cases, a diminished activity is observed (Table 2.5, Table 2.6). For example, after 9 hours of
CCE, the overpotential required to reach 10 mA/cm
2
increases from 172 mV to 247 mV,
accompanied by an increase in Tafel slope from 77.8 mV/dec to 101 mV/dec, an increase in Rct (η
= 168 mV) from 28.2 Ω to 70.7 Ω, and a decrease in Cdl from 15.8 mF to 13.3 mF. The reduced
Cdl value indicates a possible loss in the number of catalytic active sites.
SEM images (Figure 2.17) reveal an apparent change in morphology after long-term
electrolysis, as the surface of the electrode appears more porous and cracked, which could
potentially reduce the electrical conductivity of the catalyst and weaken the contact between the
electrode substrate and the catalyst, causing more active units to become inaccessible to the
electrons. X-ray photoelectron spectroscopy (XPS) measurements show no obvious change in the
Co 2p and S 2s regions after electrolysis (Figure 2.18), indicating the material does not decompose
during electrolysis. PXRD pattern of 1 after electrolysis shows broadening of the peaks, but the
peak position is similar to the pristine material (Figure 2.19), indicating the overall structure of the
38
material is partially maintained during catalysis. ICP-OES analysis of the post-electrolysis
electrolyte shows negligible amount of cobalt present in the solution.
Figure 2.14. (a) 2 h controlled potential electrolysis (CPE) of 1 (red, 9.73×10
-7
molCo) and carbon black (blue) at -0.19
V vs. RHE, Faradaic efficiency = 98%. (b) 8 h CPE of 1 (1.00×10
-6
molCo) at -0.31 V vs. RHE. (c) 24 h CPE of 1 at
-0.31 V vs. RHE. All measurements were performed in N2-saturated pH 1.3 aqueous solutions.
Figure 2.15. (a) Controlled current electrolysis (CCE) of 1 at 10 mA/cm
2
in a N2-saturated pH 1.3 H2SO4 solution; (b)
Polarization curves before and after electrolysis, scan rate: 5 mV/s; (c) Nyquist plots (markers) with respective fits
(solid lines) before and after electrolysis, measured at -0.168 V vs. RHE.
39
Table 2.5. Characteristic values for 1 before and after 9 h CCE.
Tafel slope (mV/dec) Rs (ohm) Rct (ohm) Cdl (mF)
Before 77.8 42.7 28.2
15.8
After 101 40.3 70.7
13.3
%change +29.8% -5.62% +151% -15.8%
Figure 2.16. (a) CVs (cathodic scans) of 1 (1.16×10
-6
molCo/cm
2
) as a function of electrochemical cycling in N2-
saturated pH 1.3 H2SO4 solution, scan rate: 20 mV/s; (b) Polarization curves before and after cycling, scan rate: 5
mV/s; (c) Nyquist plots (markers) with respective fits (solid lines) before and after cycling, measured at -0.168 V vs.
RHE.
Table 2.6. Characteristic values for 1 before and after 100 CV cycles.
Tafel slope (mV/dec) Rs (ohm) Rct (ohm) Cdl (mF)
Before 66.7 35.8 23.0
18.3
After 84.6 34.0 79.9
14.9
%change +26.8% -5.03% +247% -18.6%
40
Figure 2.17. Top down SEM images of 1 on RDE after electrolysis.
Figure 2.18. XPS spectra of 1 before (red) and after (blue) electrolysis: (a) Survey spectrum, (b) High-resolution Co
2p region, (c) High-resolution S 2s region.
41
Figure 2.19. XRD of 1 after 2 h CPE.
2.2.4 Determination of the Structural Model for CoTHT
To calculate the HER mechanism for CoTHT, the structural model of the material needs to be
determined first. Periodic DFT was used to capture the extended MOF structure using two layers
of CoTHT in the unit cell with implicit solvent and electrolyte included.
60,61
The revised Perdew-
Burke-Ernzerhof exchange correlation function for solids (PBEsol) with Van der Waals (D3)
correction was applied to determine the structure,
62
which has shown high accuracy in modeling
CoTHT.
45
The D3 correction is chosen instead of the Hubbard correction (U) because of the
following reasons: (1) It provides a more reasonable estimation of the experimentally observed
magnetic moment; while the addition of U leads to an unphysical increase in the calculated
magnetic moments; (2) The D3 correction shows an accurate calculation of the unit cell parameters
compared to the experimental results; (3) The inclusion of Van der Waals correction takes into
account the dispersion forces between 2D CoTHT layers. Further discussions of the DFT methods
including the choice of modeling functionals are presented in section 2.4.7. Herein, the discussion
will solely focus on the geometry optimization calculated by PBEsol+D3.
The catalytically active form of CoTHT could adopt either a fully eclipsed AA stacking mode
or an AB slipped-parallel stacking mode, where one layer is slipped relative to the neighboring
42
one along the a or b vectors. Previous work performed on an analogous MOF, Ni3(HITP)2, found
the AB stacking mode as the most energetically favored structure where one layer was slipped by
1.8 Å.
44
In contrast, prior work on CoTHT suggested the fully eclipsed AA stacking structure as
the most energetically favored stacking mode, although local minima were present at offsets of
~1.75 Å along the a or b axes.
45
However, this previous study on CoTHT did not allow for
geometry distortions of the sheets. Here, full geometry relaxations were performed, which allow
for the non-planar distortion of the CoTHT sheet and for the unit cell to change shape using the
PBEsol+D3 functional.
Figure 2.20. A comparison of the geometries of the AA (a) and the AB configuration (b and c). The spheres
representing cobalt, sulfur, carbon, and hydrogen are colored blue, yellow, brown, and white, respectively. The AB
configuration has two distinct cobalt sites, labeled as A and B site. Panel (b) shows that there are 4 A sites and 2 B
sites in each hexagonal pore.
Figure 2.20 shows the two calculated geometric configurations of CoTHT with corresponding
bond lengths. For AA stacking (Figure 2.20a), each layer is perfectly eclipsed and the Co–Co
distance between the neighboring sheets is 3.1 Å. For AB stacking, it is found that a 1.6 Å
43
displacement along the a and b vectors leads to the minimum energy configuration (Figure 2.20b,
c). The sliding of the layers creates two different bonding configurations of cobalt, which are
referred to as the A and B site. A site represents the 5-coordinated cobalt site, where the cobalt is
raised out of the plane of the four basal sulfur atoms, and forms an axial interaction with a fifth
sulfur atom in the neighboring sheet (Co–Saxial bond length of 2.6 Å). The Co–Co distance is 3.1
Å, which is identical to the one observed in the AA configuration. The square pyramidal
coordination geometry observed in the AB configuration closely resembles the dimeric structures
found in the molecular bisdithiolene complexes.
64–66
Figure 2.20c shows the orientation of the
neighboring layers within the AB configuration, which adopts a “zig-zag” pattern. The B site
cobalt atoms are puckered out of the square planes and have an alternating 2.4 Å and 3.8 Å spacing
between each other.
Calculations using the PBSsol+D3 method suggest the AB stacking mode is 45.1 kcal/mol
lower in energy compared to the AA configuration. This value was further compared with the
results obtained from other exchange-correlation functionals and in all cases the AB configuration
was found to be lower in energy (details see section 2.4.8). The calculated magnetic moment of
the AA and the AB configuration using PBEsol+D3 is 1.8 µB and 1.4 µB per formula unit,
respectively. The experimental magnetic moment was reported as 1.55 µB,
45
which lies in-between
the calculated values for AA and AB configuration. The XRD patterns for both the AA and AB
configurations were also calculated; however, the experimentally measured peaks are located
between the corresponding peaks of the calculated AA and AB structure (Figure 2.21). Admittedly,
the experimental peaks are relatively broad, so it is difficult to distinguish which stacking geometry
is dominating the overall structure of CoTHT. Due to the lack of the experimental evidence of the
44
local structure of CoTHT, the following text will discuss the calculated mechanistic pathways of
both the AB and AA stacking modes in order to present a more comprehensive picture.
Figure 2.21. A comparison of the experimental (blue), and simulated PXRD patterns of CoTHT in the AB (red) and
AA (green) configurations. The simulated PXRD patterns are displayed using the Materials Studio (version 8.0) suite
of programs by Accelrys. Left panel: the overall pattern; right panel: magnified at the [100] and [200] reflection peaks
for easier comparison.
2.2.5 DFT Studies of the HER Mechanism Using PBEsol+D3
The free energy changes caused by the addition of proton and electron pairs can be calculated
using the CHE model.
50,51
For instance, the free energy change caused by the addition of one proton
and one electron to the bare MOF can be calculated as:
∆𝐺 = 𝜇(CoH)−𝜇(Bare)−J
'
&
𝜇(H
2
)−𝑒𝑈L (1)
where 𝜇 is the chemical potential, 𝑒 is charge, and 𝑈 is the applied potential. Figure 2.22
shows the lowest free energy pathway for the HER in CoTHT starting at the bare MOF for both
AB and AA stacking modes. The black traces represent the pathway under zero applied potential
(U = 0 V). To obtain a theoretical onset potential, a non-zero U was applied (via equation 1) to the
catalytic cycle to determine when the free energy between reactants and products are similar for
the rate-determining step (∆𝐺
()*
≅ 0), which is shown as the red traces in Figure 8. Although this
method does not explicitly consider reaction barriers, the condition of isoenergetic reactant and
product for the rate-determining step (RDS) has been applied successfully to explain experimental
45
observations regarding hydrogen evolution.
67,68
More details regarding the DFT methods for the
mechanistic studies are listed in section 2.4.10 and 2.4.11.
Figure 2.22. The calculated free energy changes for the HER pathways in CoTHT that adopts (a) AB stacking
configuration or (b) AA stacking configuration. The DFT calculations were performed using the PBEsol+D3 method.
As mentioned previously, when the material adopts the AB stacking configuration, the cobalt
atoms have two different coordination environments, which are referred to as the A and B sites.
Detailed calculations have been performed on both sites considering different proton/electron pair
configurations, and the results suggest that the HER requires less energy when occurs at the A site.
Therefore, the following discussion will only focus on the lowest free energy pathway on the A
site, which is believed to be the dominant catalytic cycle. As shown in Figure 2.22a, at zero applied
potential, the first proton/electron pair is added to a cobalt atom (CoH), which is uphill in energy
by 2.2 kcal/mol. Next, a second proton/electron pair is added to form an H2 molecule adsorbed to
a cobalt atom (CoH2), which is predicted to be uphill in energy by 1.3 kcal/mol. Finally, this H2 is
released to regenerate the bare form of the catalyst. Upon setting ∆𝐺
()*
= 0, the theoretical onset
potential was calculated to be -0.098 V vs. RHE (Figure 2.22a, red trace). The applied potential
does not affect hydrogen dissociation since there is no electron transfer involved; however, this
step is downhill and should happen rapidly. Therefore, DFT results suggest a relatively simple
46
HER mechanism for CoTHT, which involves only three different configurations at the cobalt
center, without the participation of any sulfur moieties. Additional steps involving proton and
electron transfer to the sulfur atoms were also considered; however, all of these intermediates were
shown to be less energetically favorable (see 2.4.10 for supplementary details).
For AA stacking configuration, various possible mechanistic pathways were also calculated
(section 2.4.11), but again only the lowest free energy pathway is shown here (Figure 2.22b). This
catalytic cycle is very similar to the one for the AB stacking mode yet with different free energy
requirements. Both electron/proton transfer steps occur at the cobalt site and are uphill in energy,
with an energy increase of 1.1 kcal/mol and 3.0 kcal/mol, respectively. The theoretical onset
potential is -0.13 V vs. RHE for the AA stacking configuration.
Our model assumes that the electron transfer is relatively fast compared to proton transfer.
Therefore, it models 1 more closely than the pristine CoTHT or CoTHT/Nafion, as the carbon
black provides sufficient electrical conductivity for electron transfer. Thus, the corresponding
difference between the predicted onset potential for the AB configuration (-0.098 V vs. RHE) and
the earliest experimental onset potential of 1 (-0.061 V vs. RHE, Table 2.3) is only 0.8 kcal/mol,
which is in good agreement with the accuracy of DFT calculations for extended systems.
69
On the
other hand, the calculated onset potential for AA structure is -0.13 V vs. RHE, corresponding to a
1.6 kcal/mol deviation from the experimental onset. Therefore, based on the onset potential, the
catalytic system is more closely modeled by the AB stacking configuration.
2.2.6 DFT Studies of the HER Mechanism Using HSE06+D3
Although the PBEsol+D3 provided an accurate estimation of the experimental results, it is still
possible that the model using PBEsol+D3 is over delocalized compared to the true system. Thus,
additional calculations were performed using a hybrid functional, the Heyd-Scuseria-Ernzerhof
47
(HSE06) functional with the D3 correction. Given the high computational cost of HSE06+D3,
these additional calculations were solely focused on the study of the HER mechanism based on
the geometries optimized by the PBEsol+D3. Using HSE06+D3, the magnetic moment for the AA
and the AB configuration were calculated to be 2.4 µB and 1.8 µB per formula unit. The calculated
magnetic moment for the AB configuration is in better agreement with the experimentally
observed value of 1.55 µB, as is observed for the PBEsol+D3 calculation.
Figure 2.23. The calculated free energy changes for the HER pathways in CoTHT that adopts (a) AB stacking
configuration or (b) AA stacking configuration. The DFT calculations were performed using the HSE06+D3 method.
The lowest free energy pathways for the HER on both the AB and AA stacking mode were
calculated by the HSE06+D3 functional, with the results shown in Figure 2.23. For the AB
configuration, at zero applied potential, the first proton/electron pair is added to a cobalt atom
(CoH in Figure 2.23a), which is downhill in energy by 4.3 kcal/mol. Next, a second proton/electron
pair is added to the same cobalt atom to form a H2 molecule adduct (CoH2 in Figure 2.23a), which
is predicted to be uphill in energy by 2.0 kcal/mol. For the AA configuration (Figure 2.23b), at
zero applied bias the first electron/proton transfer step is uphill by 4.9 kcal/mol, while the second
electron/proton transfer step is largely downhill by 30.5 kcal/mol. This large energy decrease
suggests a strong binding between cobalt and H2, hence a further electron transfer is likely required
48
to dissociate H2 and complete the catalytic cycle. Using HSE06+D3, the theoretical onset potential
is determined as -0.073 V and -0.180 V vs. RHE for the AB and AA stacking configuration,
respectively. Comparing these results to those provided by the PBEsol+D3, a relatively large
difference in the free energy changes can be observed. However, these two functionals provided
qualitatively similar results with the reaction energies being lower for the AB configuration, along
with a relatively close prediction of the onset potential (-0.098 V for PBEsol+D3 and -0.073 V for
HSE06+D3).
2.2.7 Discussion of the HER Mechanism
The combined computational and experimental analyses of hydrogen evolution in CoTHT
provide significant insight into the mechanism. While the DFT results showed a heavy dependency
on the exchange-correlation functional, a few general conclusions can still be drawn from the
calculations. First, the DFT results suggest that the AB stacking mode is energetically favorable
and is an important configuration for efficient catalysis. Second, it is found that regardless of the
stacking mode of the layers, the HER is governed by a Volmer-Heyrovsky mechanism, where the
initial formation of the adsorbed-hydrogen is followed by the reduction of a second proton and the
concomitant formation of an H-H bond. Only cobalt is found to serve as the catalytic active center,
in contrast to the [Co(bdt)2]
-
complex where the formation of S–H moieties is involved in the
catalytic cycle.
46
Therefore, the HER mechanisms for CoTHT resembles more closely to that of
the cobalt-bis(diaryldithiolene) species, where the proton/electron transfers occur more favorably
at the cobalt centers.
47,48
Last, for both the AA and AB stacking geometries, depending on the level
of theory, either the Volmer step (first proton-electron transfer) or the Heyrovsky (second
proton/electron transfer) could be rate determining, and the ∆𝐺
()*
ranges between 2 to 3 kcal/mol.
Previous calculations performed using a CHE-like model on the Pt(111) surface and the MoS2
49
surface gave a ∆𝐺
()*
of 2.5 and 2.7 kcal/mol,
67,68
respectively, which are of similar magnitude
compared to the ∆𝐺
()*
calculated for CoTHT. While the Pt and MoS2 are known for their
outstanding HER performance, our computational results suggest that the CoTHT should also be
intrinsically active for the HER, as is observed in the experimental results.
Experimentally, Tafel analyses can help to determine the mechanism of the HER catalysts.
Under ideal scenario, when the Volmer step is rate-determining, the theoretical Tafel slope is ~120
mV/dec, whereas when the Heyrovsky reaction is rate-determining, the theoretical Tafel slope is
~40 mV/dec.
70–72
Deviations from theoretical Tafel slopes should be expected as transfer
coefficients (α) and surface adsorption could affect the rate. Indeed, from Figure 2.13 the relatively
large peak-to-peak separation (∆Ep) suggests deviation from the ideal theoretical case. Regardless,
Tafel slope analyses are useful in mechanistic analysis to determine the rate-determining processes.
Therefore, since CoTHT/Nafion has a Tafel slope closer to 120 mV/dec (experimentally measured
as 150 mV/dec, Table 2.2), the Volmer reaction is likely the RDS. However, when CB is
incorporated to form 1, the electron transfer rate is significantly increased, leading to the shift of
rate-determining step from the Volmer reaction to the Heyrovsky reaction. This can be seen
experimentally as the Tafel slope for 1 is as low as 70.6 mV/dec. This intermediate Tafel slope
between 40 ~ 120 mV/dec can be partially contributed by a complex mechanism where either
Heyrovsky or Volmer step serves as the RDS.
In summary, the DFT and experimental results both suggest a potentially convoluted
mechanism pathway, where both Heyrovsky and Volmer steps can play the role of RDS. While
the details of the computational model lead to differences in the energetics of the catalytic pathway,
the estimated free energy changes for the RDS are similar to what was calculated for Pt and MoS2,
and hence should warrant efficient hydrogen production on the cobalt active sites. However, the
50
experimental results show that the high intrinsic activity of CoTHT can only be expressed by
modifying the catalyst with the addition of Nafion and carbon black. These together highlight the
importance of a robust electrode/catalyst architecture that can support an efficient proton/electron
transfer to the active sites. Further calculations including reaction barriers, explicit electrolyte, as
well as the modelling of proton/electron transfer pathways are required to gain a more
comprehensive picture of the overall mechanism.
2.3 Conclusions
In conclusion, the HER activity of CoTHT is significantly improved upon the addition of
Nafion and carbon black to form an ink composite 1, due to enhanced electrode-catalyst integration
and a much faster electron transport process. Composite 1 exhibits an optimal overpotential of 143
mV to reach 10 mA/cm
2
and a Tafel slope of 70.6 mV/dec in pH 1.3 aqueous solutions. With this
activity, 1 outperforms all the other analogous dithiolene-based MOF catalysts that have been
reported in the literature (Table 2.1). To better understand the HER activity of CoTHT, the
catalytic pathway was investigated through DFT calculations. The calculations suggest that
CoTHT can adopt two different stacking configurations, the AA and the AB configurations, with
the latter displaying lower free energy. The DFT calculations were performed on both
configurations to provide a more comprehensive mechanistic picture. It is found that, regardless
of the stacking mode, the catalytic cycle involves three steps: (1) a Volmer discharge reaction to
form a cobalt hydride, (2) a Heyrovsky step where a proton/electron pair leads to the formation of
a cobalt-H2 intermediate, and (3) H2 evolution. The calculated onset of catalysis is in good
agreement with the experimentally observed value, supporting the reliability of the calculated
mechanism. Overall, the work presented herein highlights the great intrinsic activity of CoTHT,
reveals the significance of the properties of the catalyst/electrode architecture, and provides
51
insightful information about the mechanistic pathways of the material. Future studies will focus
on synthesizing the MOF via a solvothermal approach, which will allow for the direct growth of
the MOF thin film on the electrode substrate with improved charge transport properties.
73,74
It will
also potentially give rise to more crystalline materials, which can allow for detailed
characterization of the local structure to identify the AA or AB stacking mode. Additionally, more
detailed theoretical studies are also underway to further understand the HER mechanism as well
as proton and electron transfer processes.
2.4 Supplementary Experimental Information
2.4.1 General Considerations
All manipulations of air and moisture sensitive materials were conducted under a nitrogen
atmosphere in a Vacuum Atmospheres glovebox or on a dual manifold Schlenk line. The glassware
was oven-dried prior to use. Water was deionized with the Millipore Synergy system (18.2 MW·cm
resistivity). All the solvents used were degassed under vacuum and refilled with nitrogen (10 ×).
Triphenylene-2,3,6,7,10,11-hexathiol ligand (THT) was synthesized according to literature
procedure.
75
All other chemical reagents were purchased from commercial vendors and used
without further purification.
2.4.2 Synthesis of CoTHT
The synthesis of CoTHT was previously reported by our laboratory.
15
A 120 mL jar was
charged with a solution of CoCl2∙6H2O (40.0 mg, 0.168 mmol) in water (40 mL). Separately, a
suspension of THT (2.5 mg, 0.006 mmol) in N-methyl-2-pyrrolidone (NMP) (0.1 mL) was diluted
with ethyl acetate until the total volume of the suspension reached 5 mL, sealed, and briefly
sonicated to form a uniform suspension. Ethyl acetate (35 mL) was gently layered on top of the
aqueous solution to create a liquid-liquid interface; the suspension of THT was then gently added
52
to the ethyl acetate layer and the jar was sealed and allowed to stand still. A black film appeared
at the liquid-liquid interface over 5 days, which was then collected as powder, solvent exchanged
with methanol (3 × 20 mL), and dried under vacuum.
2.4.3 Deposition of CoTHT for Electrochemical Study
Preparation of ink composite (1): 2 mg of CoTHT was mixed with a desired amount of carbon
black (Vulcan XC-72R), 20 µL of Nafion solution (0.5 wt%), 45 µL of water, and 135 µL of
ethanol, followed by sonication for 1 hour to form a uniformly dispersed suspension. A desired
amount of suspension was then drop casted onto a glassy carbon electrode (GCE) or a rotating disk
electrode (RDE, glassy carbon insert) using a microsyringe, and dried in nitrogen atmosphere at
room temperature.
Direct deposition of CoTHT: Deposition was carried out by drop casting the film formed at
the liquid-liquid interface to the electrode substrate using a glass pipette. Following deposition,
the electrode was washed with water and methanol.
Preparation of CoTHT/Nafion mixture: 2 mg of CoTHT was mixed with 20 µL of Nafion
solution (0.5 wt%), 45 µL of water, and 135 µL of ethanol, followed by sonication for 1 hour to
form a uniformly dispersed suspension. 10 µL of suspension was then drop casted onto a GCE and
dried in nitrogen atmosphere at room temperature.
2.4.4 Electrochemical Methods
Electrochemical experiments were carried out using a VersaSTAT 3 potentiostat in a three
electrode configuration electrochemical cell under an inert atmosphere. A glassy carbon electrode
(GCE, 0.07065 cm
2
surface area) or a rotating disk electrode (RDE, glassy carbon insert, 0.196
cm
2
surface area) was used as the working electrode. GCE and RDE were polished with 0.05 µm
Al2O3 polish powder and sonicated in water prior to use. A graphite rod, purchased from Graphite
53
Machining, Inc. (Grade NAC-500 Purified, < 10 ppm ash level), was used as the counter electrode.
The reference electrode, placed in a separate compartment and connected by a Vycor tip, was
based on an aqueous Ag/AgCl/saturated 3.5 M KCl electrode. The reference electrode in aqueous
media was calibrated externally relative to ferrocenecarboxylic acid (Fc-COOH) at pH 7.0, with
the Fe
3+/2+
couple at 0.28 V vs. Ag/AgCl. All potentials reported in this paper were converted to
the reversible hydrogen electrode (RHE) by adding a value of (0.205 + 0.059 × pH) V, or to the
normal hydrogen electrode (NHE) by adding a value of 0.205 V (for variable pH studies).
The aqueous solutions used in the electrochemical experiments were prepared as follows. For
the pH 0.4 solution, 5.61 mL of 18.7 M H2SO4 was dissolved in 200 mL water. For the pH 1.3
solution, 0.534 mL of 18.7 M H2SO4 was added to 200 mL 0.1 M NaClO4. For the pH 2.6 solution,
citric acid (3.750 g) and Na2HPO4 (0.620 g) were dissolved in 200 mL 0.1 M NaClO4. For the pH
4.6 solution, NaOAc (1.605 g) and acetic acid (1.2 mL) was added to 200 mL 0.1 M NaClO4. For
the pH 10.0 solution, NaHCO3 (0.678 g) and Na2CO3 (1.264 g) were dissolved in 200 mL 0.1 M
NaClO4. The pH of the solutions was measured with a benchtop Mettler Toledo pH meter. Prior
to electrochemical testing, the solution was purged with nitrogen thoroughly.
Controlled potential electrolysis (CPE) measurements to determine Faradaic efficiency were
conducted in a sealed two-chambered H cell where the first chamber held the working and
reference electrodes in 40 mL of pH 1.3 aqueous solution and the second chamber held the counter
electrode in 20 mL of pH 1.3 aqueous solution. The two chambers were both under N2 and
separated by a fine porosity glass frit. CPE experiments were performed with a glassy carbon plate
electrode (6 cm × 1 cm × 0.3 cm; Tokai Carbon USA) as the working electrode and a graphite rod
as the counter electrode. The reference electrode was a Ag/AgCl/saturated 3.5 M KCl (aq)
electrode separated from the solution by a Vycor tip. Using a gas-tight syringe, 2 mL of gas was
54
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. To determine the Faradaic efficiency, the theoretical H2 amount based on
total charge flowed is compared with the GC-detected H2 produced from controlled-potential
electrolysis.
Electrochemical impedance spectroscopy (EIS) measurements were carried out at different
overpotentials in the frequency range of 100 kHz – 0.1 Hz with 10 mV sinusoidal perturbations.
Experimental EIS data were analyzed and fitted with the ZSimpWin software.
The obtained polarization curves were corrected for iR loss according to the following equation:
Ecorr = Emea – iRs
Where Ecorr is the iR-corrected potential, Emea is the experimentally measured potential, and Rs
is the solution resistance extracted from the fitted EIS data.
2.4.5 Physical Characterization Methods
X-ray photoelectron spectroscopy (XPS) data were collected using a Kratos AXIS Ultra
instrument. The monochromatic X-ray 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
"+
Torr. The XPS data were analyzed using the CasaXPS software.
High resolution synchrotron powder X-ray diffraction data was collected using the 11-BM
beamline mail-in program at the Advanced Photon Source (APS), Argonne National Laboratory,
with an average wavelength of 0.412750 Å. Discrete detectors covering an angular range from 0.5
55
to 30
o
2θ are scanned over a 34
o
2θ range, with data points collected every 0.001
o
2θ and scan
speed of 0.01
o
/s.
Inductively coupled plasma optical emission spectroscopy (ICP-OES) measurements were
performed using a Thermo Scientific iCAP 7000 ICP-OES. After electrochemical tests, the
composite was removed from the electrode surface, digested in concentrated nitric acid, and tested
by ICP-OES to determine the concentration of the cobalt ions.
Scanning electron microscopy (SEM) was performed on a JEOL JSM 7001F scanning electron
microscope using an accelerating voltage of 30 kV.
2.4.6 General DFT Methods
Electronic structure calculations were performed using density functional theory (DFT) and
the Perdew-Burke-Ernzerhof (PBE) or revised Perdew-Burke-Ernzerhof exchange correlation
function for solids (PBEsol), or the Heyd-Scuseria-Ernzerhof (HSE06) functional with the Vienna
ab initio simulation program (VASP).
62,76–81
The Van der Waals correction (D3) and the U
correction were used in some calculations. The plane-wave cut off for all calculations was set to
500 eV and k-point sampling was done with the Monkhorst-Pack method with (2,2,10), unless
otherwise noted.
82
All calculations were performed spin-polarized, and the calculations were
initialized with a magnetic moment of +1 on each cobalt atom (through use of the MAGMOM tag)
of CoTHT, unless otherwise noted. The VASPsol program was used to describe solvent with
additional modifications for electrolyte.
60,61,83
The relative permittivity was set to 78.4, and the
Debye length for the electrolyte was set to 3.0 Å. The total energy is:
𝐸[𝜌
DFT
] = 𝐸
KS-DFT
[𝜌
),-
]+𝐸
Cav
[𝜌
DFT
]
+UV𝜙(𝑟)Y𝑁
nuc
(𝑟)−𝜌
DFT
(𝑟)[\𝑑
.
(𝑟)+
𝜖(𝑟)
8𝜋
|𝜙(𝑟)|
&
−
𝜖(𝑟)𝜅(𝑟)
&
8𝜋
|𝜙(𝑟)
&
|
56
where 𝜙(𝑟) is the electron potential, 𝜌
DFT
is the electron density from DFT, 𝑁
nuc
(𝑟) is the
atom nuclear charge density, and 𝐸
KS-DFT
[𝜌
DFT
] is the KS-DFT energy. The dielectric is smoothly
varying and solved with:
𝜖Y𝜌
DFT
(𝑟)[ = 1+(𝜖
/
−1)𝑆(𝜌
DFT
(𝑟))
where 𝜖
/
is the dielectric of the bulk and the inverse Debye screening length is solved with:
𝜅(𝑟)
&
= 𝜅(𝑟)
&
𝑆(𝜌
DFT
(𝑟))
where 𝜅(𝑟) is the inverse Deybe screening length of the bulk and:
𝑆Y𝜌
DFT
(𝑟)[ =
1
2
𝑒𝑟𝑓𝑐(
logg
𝜌
DFT
(𝑟)
𝜌
0
(𝑟)
h
𝜎√2
)
where 𝜌
0
(𝑟) defines at what electron density a cavity starts to form and 𝜎 is a parameter for
the width of the cavity. In this case 𝜌
0
= 4.73∗10
".
Å
.
and 𝜎 = 0.6. The contribution to the
cavitation energy is:
𝐸
Cav
[𝜌
DFT
] = 𝜏Uo∇𝑆Y𝜌
DFT
(𝑟)[o𝑑
.
(𝑟)
where 𝜏 is the effect surface tension which in this case 𝜏 = 0.525 meV/Å
&
. Two layers of the
CoTHT MOF were modeled. Calculations where initialized with the lattice parameters: 𝑎 =
23.0 Å,𝑏 = 23.0 Å,𝑐 = 6.6 Å,𝛼 = 90.0°,𝛽 = 90.0°,𝛾 = 120.0° and the geometries were
allowed to relax including the unit cell. Free energies for each system were calculated by adding
zero-point energy (ZPE), entropies, and heat capacities to the electronic energy obtained from
VASP.
84
Frequencies were calculated by treating the degrees of freedom of added hydrogen atoms
and the atoms they were coordinated to as vibrational and that there were no significant vibrational
changes to other parts of the system.
85
Several geometries had frequencies lower than 50 cm
-1
that
were then set to 50 cm
-1
.
57
Table 2.7 shows the energy for the addition of a proton/electron pair to the cobalt atom in the
AB configuration with increasing Ecut and increasing K-point sampling using PBEsol+D3. The
energy does not vary by more than 0.1 kcal/mol for calculations at 500 ~ 1000 eV energy cutoff,
and does not vary at all when the energy cutoff is greater than 700 eV; therefore, for computational
efficiency reasons a value of 500 eV was chosen. A minor change of 0.6 kcal/mol was observed
when the K-point was increased from (2,2,5) to (2,2,10). Therefore, all final energy calculations
were performed using (2,2,5) K-points; however, all geometry optimization calculations were
performed using (2,2,5) K-points due to computational cost considerations. Additionally, (2,2,5)
K-points were used for all frequency calculations for the thermodynamic corrections. Finally,
calculations based on a 4-layer CoTHT model were also performed, and it was found that the
energy change is only different by 1.1 kcal/mol in comparison to the 2-layer model (last entry in
Table 2.7). Therefore, the 2-layer model was used throughout the rest of this study.
As previously reported,
45
5 unique magnetic configurations can be isolated in CoTHT, with
the lowest energy configuration corresponding to a magnetic moment of 1.8 µB and 1.4 µB per
formula unit for the AA and the AB configuration, respectively (obtained using PBEsol+D3). The
1.8 µB magnetic moment is in agreement with the previous theoretical calculations on the AA
configuration of this MOF. These numbers are also in agreement with experimental magnetic
moment of 1.55 µB.
45
In the AB configuration, this corresponds to a local moment of ~0.47 µB per
cobalt with negligible spin density on the ligands. Again, as previously discussed, this is consistent
with two-thirds of the Co having a trivalent state (S = 0) and one exhibiting a formal divalent state
(S = ½).
45
This is significant since mixed oxidation states are associated with charge delocalization.
Due to the cost of exact exchange DFT functionals, we only performed electronic energy
calculations at the HSE06+D3 level of the theory. In the HSE06 free energies reported here, we
58
use the geometries obtained using PBEsol+D3 to calculate the electronic energies. We then added
the thermodynamic corrections obtained using PBEsol+D3 to the HSE06 electronic energies.
Table 2.7. Free energy change for the addition of one proton/electron pair to the cobalt atom on a bare CoTHT MOF
in the AB configuration with increasing energy cutoff and K-point values using PBEsol+D3.
Layers of
CoTHT
K-points Energy Cutoff (eV) Co-H addition energy (kcal/mol)
2 2´2´5 500 1.6
2 2´2´5 600 1.7
2 2´2´5 700 1.7
2 2´2´5 800 1.7
2 2´2´5 900 1.7
2 2´2´5 1000 1.7
2 2´2´10 500 2.3
2 2´2´10 600 2.3
4 2´2´5 500 2.8
2.4.7 Comparing Different DFT Models
CoTHT is a highly conductive MOF compared to conventional MOFs. The appropriate DFT
model is challenging as one might expect that DFT+U models to perform better due to localization
of the 3d electrons; however, DFT+U typically does not perform well on traditional conductors
such as metals. Nevertheless, to determine the applicability of DFT+U to the system, we performed
a series of calculations with variable U values using PBE and PBEsol to calculate the lattice
parameters and magnetic moments.
Figure 2.24 shows the lattice parameters a and c, as well as the magnetic moment as a function
of the U parameter in PBEsol+U+D3 calculations. Using U = 0.0 eV gives lattice parameters of
𝑎 = 23.02 Å,𝑏 = 23.02 Å,𝑐 = 6.45 Å,𝛼 = 91.6°,𝛽 = 87.3°,𝛾 = 119.8° , which is in good
agreement with the experimentally observed lattice parameters of 𝑎 = 22.52 Å,𝑏 = 22.52 Å,𝑐 =
6.6 Å,𝛼 = 90.0°,𝛽 = 90.0°,𝛾 = 120.0°. Increasing the value of U only leads to a slight deviation
of the lattice parameters from the experiment. For example, when U = 8.0 eV, the lattice parameters
59
are 𝑎 = 23.23 Å,𝑏 = 23.20 Å,𝑐 = 6.43 Å,𝛼 = 89.6°,𝛽 = 87.2°,𝛾 = 119.8° . As shown in
Figure 2.24, the inclusion of U leads to less than 1% change in the lattice parameters. Therefore,
any choice of U in the range of 0.0 ~ 8.0 eV can yield a reasonable lattice parameter that is
comparable with the experimental geometry. Hence, we cannot determine the applicability of U
solely base on lattice parameters.
Next, we explored the magnetic moments as a function of the U parameter. As stated above
and previously reported,
45
significant care must be taken in these calculations to ensure the proper
spin and magnetic states are being modeled correctly. Through initialization using +1 magnetic
moment on each cobalt atom we converge to a state that has a smaller spin density on one cobalt
atom, while the other two cobalt atoms have an equally larger spin density. Our calculations are
suggesting that one cobalt atom has more trivalent state (B site, Figure 2.20) and the other two
atoms (A site, Figure 2.20) have a mixed valance state between the trivalent and divalent states.
Therefore, PBEsol+D3 without a U does give an experimentally consistent spin and magnetic
states and a total magnetic moment of 1.4 µB per formula unit with experimental magnetic moment
of 1.55 µB.
45
Beginning from this state, we now increase the U value from 0.0 to 8.0 eV and plot the
magnetic moment in Figure S19c. It was observed that, at U = 1.0 eV, the magnetic moment has
increased to 1.7 µB per formula unit, equal distance from the experimental magnetic moment
compared to U = 0.0 eV. Further increasing of U to 2.0 - 4.0 eV leads to an increase of the magnetic
moment where the spin density is delocalized across all three cobalt atoms, such that each cobalt
atom has a similar local magnetic moment, but still with a mixed valance state. Further increasing
beyond 4.0 eV leads to all cobalt atoms having unpaired electrons. At U = 5.0 eV, all mixed valance
character is lost. Then even higher spin states are seen for U = 6.0 to 8.0 eV. Therefore, application
60
of U increases the spin state of the cobalt atoms in CoTHT, leading to magnetic moments
incomparable to the experiment. We also tried initializing these calculations using +0.5 and 0.0
magnetic moment on each cobalt atom, which all lead to either higher energies or higher magnetic
moments. Initializing using the U=0.0 density led to similar results as initializing using +1
magnetic moment.
This data suggest that a large U value is certainly not appropriate to model CoTHT, while one
could consider a small value (~1 eV) could be considered appropriate. To further explore if U is
appropriate we applied linear response theory to calculate the value of U.
86
However, this
calculation gave a value of U = 5.04 eV, which based on the above calculations, would not be
appropriate as it results in the qualitatively incorrect spin and magnetic state.
Figure 2.24. The lattice parameters a (panel a), c (panel b), and magnetic moment (µB) per formula unit of Co3(THT)2
(panel c) of the AB configuration as a function of U using PBEsol+D3. The experimental lattice parameters are 𝑎 =
22.52 Å 𝑎𝑛𝑑 𝑐 = 6.6 Å; and the experimental magnetic moment is 1.55 µB.
U (eV)
Magnet Moment
(per formula unit)
0
1
2
3
4
5
0 1 2 3 4 5 6 7 8
6.38
6.4
6.42
6.44
6.46
6.48
22.9
23.0
23.1
23.2
23.3
c (Å) a (Å)
(a)
(b)
(c)
61
Therefore, the aforementioned results suggest that DFT+U is not appropriate for this system.
Since the experimentally observed state suggests that it is a mixed oxidation state with charge
delocalization, it is perhaps not surprising that the addition of U, which increases charge
localization,
87
does not appropriately model the system. Additionally, not adding U is consistent
with not applying U in the case of conductors as this is a relatively conductive MOF, and its
conductivity properties were previously calculated using PBEsol+D3.
45
As we have now determined that U is not appropriate, an alternative approach frequently used
to study the mixed-valence states in molecular complexes is to utilize a hybrid functional which
incorporates exact exchange.
88
Therefore, we considered the use of the Heyd-Scuseria-Ernzerhof
(HSE06) functional with the D3 correction. Given the computational cost of performing these
calculations, we were unable to perform geometry optimizations and used the PBEsol+D3
geometries. Additionally, we could only perform calculations using k-point sampling with the
Monkhorst-Pack method of (2,2,5).
Again, we must be careful about the spin state convergence. Here, application of a +1 magnetic
moment for each cobalt atom leads to a 3.0 µB per formula unit for both the AA and AB
configurations, with one unpaired electron on each cobalt atom. As is frequently seen, the
application of exact exchange increases the stability of the higher spin configuration in transition
metals.
89
However, when the PBEsol+D3 electron densities were used for the initial configurations,
we could converge to a higher energy spin state with the correct qualitative spin and magnetic
description. Using this protocol, we calculated magnetic moment of the AA and the AB
configuration to be 2.4 µB and 1.8 µB per formula unit, respectively. Therefore, the magnetic
moment of the AB configuration is slightly higher but still in-line with the experimental value of
1.55 µB.
62
Taken together, our results from DFT+U and HSE06 modeling suggest that both are over
stabilizing the high-spin state of the cobalt atoms in contrast to the more experimentally observed
mixed-valence state of CoTHT. As reasonable values for the magnetic moment were obtained
using HSE06+D3 and PBEsol+D3, we have discussed those results in the above text. The
mechanism of HER calculated using PBEsol+U are presented below for completeness.
2.4.8 Comparing Different Configurations of CoTHT
Table 2.8. Free energy difference between the AA and AB configurations calculated using different methods.
Method EAB -EAA (kcal/mol)
PBE -33.7
PBEsol -56.4
PBE + D3 -39.2
PBEsol + D3 -45.1
PBE + U -16.9
PBEsol + U -5.9
PBE + U + D3 -10.6
PBEsol + U + D3 -10.1
HSE06 -119.5
HSE06 + D3 -108.4
Table 2.8 shows the energy difference between AB and AA configurations (EAB - EAA). These
calculations were performed including 2 layers of CoTHT in a charge-neutral supercell. The
negative values correspond to the lower energy of the AB configuration. Again, full geometry
relaxations were performed, which allowed for the unit cell to change shape and the CoTHT sheet
to become non-planar, with the exception of the HSE06 and HSE06+D3 calculations that used the
PBEsol+D3 geometry. Here, not only PBEsol+D3 was tested; the PBE and the effects of the D3
correction and the U correction with U = 4.0 eV were also studied. Table 2.8 shows that, for all
combinations of PBE, PBEsol, with or without the D3 and U correction, the AB configuration is
always lower in energy. However, the various methods of the calculations lead to significantly
63
different values – PBEsol + U gives an energy difference of only -5.9 kcal/mol and PBEsol gives
an energy difference of -56.4 kcal/mol. The HSE06 and HSE06+D3 calculations give significantly
lower energies for the AB configuration, which could be due in part to being unable to afford
geometry optimization using the HSE06 functional.
2.4.9 Molecular Modeling
Molecular modeling of CoTHT was carried out using the Materials Studio (version 8.0) suite
of programs by Accelrys. The MS Reflex module was used to calculate the simulated PXRD
patterns of CoTHT in the AA and AB configurations. Line broadening for crystallite size was not
calculated. Comparison of the experimental and simulated PXRD patterns in the AA and AB
configurations is shown in Figure 2.21. The difference between the PXRD patterns of the two
stacking modes is very small, and both configurations compare similarly with the experimental
pattern. Due to the broadness of the experimental PXRD pattern, it is difficult to identify which
exact stacking mode is dominating the overall structure of CoTHT.
2.4.10 The HER Mechanism of CoTHT in the AB Configuration
As discussed in the main text, the AB configuration has two different cobalt sites, the A site
and the B site, which have different coordination environments. Therefore, calculations were
performed to explore the HER at both sites. Figure 2.25 illustrates the active site for HER in
CoTHT for both the A and B sites in the AB configuration and the AA configuration. When the
proton/electron pair binds to the cobalt atom in the AB configuration, it is more energetically
favorable to be placed in the larger void space, which is between Co(1) and Co(2) in the above
layer in Figure 2.25a and b. It is noted that for the first hydrogen addition to a sulfur atom in the
AA configuration, all 8 sulfurs are equivalent (Figure 2.25c); however, this is not the case in the
AB configuration. Figure 2.25d shows a proton/electron pair bound to the S(1) atom in the A site
64
of the AB configuration. The calculations were initialized with the hydrogen atom being placed in
the plane of CoTHT; however, the geometries always optimize with the hydrogen atom being
located between the layers of CoTHT. As a result, the geometries were optimized on both sides
of the sulfur atom and the lower energy configuration is listed below (Figure 2.25d). The detailed
mechanistic plots for individual active sites displayed below are based on the atom-labeling
scheme shown in Figure 2.25.
Figure 2.25. The labeling scheme used to describe where the proton/electron pairs bind to the active sites in CoTHT
for the (a) A site of the AB configuration, (b) B site of the AB configuration, and (c) AA configuration. Panel (d)
shows a proton/electron pair bound to the S(1) atom in the AB configuration. Panel (a) shows the periodic boundary
conditions by showing Co(2) on both the top and the bottom as these are identical and both are shown for visualization
purposes.
Figure 2.26a illustrates the HER pathways where the first step is the proton/electron addition
to the Co atom at the A site. The lowest free energy pathway was found to involve a simple three-
step mechanism. First, an initial hydrogen and electron transfer occurs to form Co(1)H (Figure
2.22a). Second, an initial hydrogen and electron transfer occurs to form Co(1)H2. This can be
followed by hydrogen gas evolution to return to the bare state. Additional steps involving proton
and electron transfer to the sulfur atoms were also considered; however, all of these intermediates
were shown to be higher in energy (Figure 2.26a). Therefore, although these intermediates are low
enough in energy to be formed, our results suggest the dominant pathway should be simple: bare,
to Co(1)H, and then Co(1)H2.
65
Figure 2.26b demonstrates the HER mechanism where the first step is a proton/electron
transfer to a sulfur atom at the A site. Here, all of the pathways are higher in energy than the ones
presented in Figure 2.26a, where the initial proton/electron transfer happens at a cobalt site.
However, it is noteworthy that the energy difference between Co(1)H2/S(1)H/S(3)H and
S(1)H/S(3)H is only 2.4 kcal/mol, lower than the 3.5 kcal/mol energy difference between the bare
and Co(1)H2 in Figure 2.26a. This suggests that if the intermediate S(1)H/S(3)H is formed, the
HER mechanism will require less energy input to proceed.
Figure 2.26. A comparison of the relative free energies between intermediates at the A site of the AB configuration of
CoTHT.
Next, the mechanism at the B site of the AB configuration is considered. Figure 2.27a and 27b
show the intermediates for the first proton/electron transfer at the cobalt and sulfur atoms,
respectively. Generally, the HER pathways at the B site are considerably higher in energy
compared to those at the A site. The free energy to form H2 on the cobalt site is 12.9 kcal/mol.
This suggests that, while feasible under electrochemical conditions, it is likely playing a smaller
role in the overall rate, especially when the overpotential input is low. However, one interesting
observation regarding the B site is that the protonation of the S atom is thermodynamically
downhill, suggesting that it should happen rapidly.
66
In summary, the DFT results obtained here suggest the A site is the more likely active site for
catalysis in the AB configuration of CoTHT. However, it is likely that both the A site and B site
are catalytically active under higher overpotentials. Future work is to improve the model to include
explicit solvation and calculate transition states.
Figure 2.27. A comparison of the relative free energies between intermediates at the B site of the AB configuration of
CoTHT.
As large differences were seen in the energy of the AA and AB configurations depending on
the choice of functional and whether to include the D3 or U correction (Table 2.8), the simple
mechanism at the A site of the AB configuration was also calculated using different calculation
methods, with the results summarized in Table 2.9. Again, a large variation in the free energy
differences is seen here. When no D3 of U correction is applied, the PBE and PBEsol functional
give relatively similar results with both steps being uphill in energy and each step being within a
few kcal/mol of each other. When the D3 correction is applied, lower energies for the intermediates
are obtained for PBEsol, while higher energies are obtained for PBE.
We used a U correction of U = 4.0 eV which is in the range of suggested U values for metal-
organic frameworks.
90
In general, when the U correction is applied, high energies for the
intermediates are observed, with PBEsol with D3 correction being the exception. Such high-energy,
less accessible intermediates contradicts with the high activity of CoTHT observed experimentally;
67
hence the U correction is not considered reliable. Additionally, it has been suggested that the U
parameter for cobalt is oxidation state specific.
90
Since the oxidation state changes during HER, it
suggests that the U correction could be introducing additional error.
Table 2.9. Free energies for the binding of H and H2 to the A-site cobalt atom in CoTHT calculated using different
methods.
Method CoH (kcal/mol) CoH2 (kcal/mol)
PBE 3.6 9.8
PBEsol 5.8 6.1
PBE + D3 9.4 15.3
PBEsol + D3 2.2 3.5
PBE + U 17.2 19.2
PBEsol + U 9.3 5.1
PBE + U + D3 16.1 25.0
PBEsol + U + D3 2.2 -2.0
PBEsol is optimized for solids, so it is believed that this functional is more accurate for
modeling CoTHT compared to PBE. Additionally, as the interactions between layers of CoTHT
should have significant Van der Waals interactions, the PBEsol + D3 is the considered best choice
for this system. Hence, the majority of the results presented in this paper were obtained using this
functional combination. The relatively good agreement between the experimental and the
theoretical onset potential calculated from PBEsol + D3 also provides support for its validity.
However, given the large spread of energies varied with the choice of functional and corrections,
it is believed that our quantitative results are less conclusive compared to our qualitative
mechanistic study. Therefore, the results presented herein should be interpreted as possible
pathways that are playing a role in the overall HER mechanism of CoTHT. Further calculations
of reaction barriers and explicit electrolyte are required to gain a more comprehensive picture of
the overall mechanism, which are planned in the future work.
68
2.4.11 The HER Mechanism of CoTHT in the AA Configuration
As the current experimental evidence cannot accurately identify the local structure of CoTHT,
it is possible that the AA configuration still plays a role in its overall catalytic activity. Therefore,
the mechanism in the AA configuration was also explored. Similar to previous discussion, the
mechanism pathways were studied separately based on where the initial proton/electron pair is
added, which are shown in Figure 2.28.
These results show a very similar mechanism compared to the AB configuration. The lowest
energy pathway begins with a proton/electron transfer to cobalt to form Co(1)H. An additional
proton/electron transfer follows this step to form Co(1)H2. This is then followed by H2 evolution
and returning to the starting state. All other pathways are higher in energy; therefore, these results
suggest that in the AA configuration the formation of SH is not relevant to the catalytic pathway.
Figure 2.28. A comparison of the relative free energies between intermediates of the AA configuration of CoTHT.
2.4.12 The HER Mechanism in the AB Configuration Using Hubbard Correction (+U)
As discussed above, due to the U correction stabilizing the higher spin states of CoTHT, the
most accurate model of CoTHT is determined to be PBEsol+D3. Here, we also calculated the
mechanism presented in the main text as a function of U to briefly illustrate how application of U
would affect the free energy of the intermediates for the AB configuration. Again, significant care
69
is required when performing these calculations to ensure the best spin state is achieved throughout
the reaction profile. Therefore, we initialize all of these calculations by using the electron density
obtained using PBEsol+D3 without U.
Figure 2.29a shows the free energy change between bare CoTHT and CoH as a function of U.
We see that this step is surprisingly robust as a function of U with small (~2 kcal/mol) increase
until U = 4.0 eV. Above this value, we see large changes depending on U, which is due to the
changes in the spin state of cobalt at higher U values. Figure 2.29b shows the free energy change
between CoH and CoH2 as a function of U. Here, we see larger variation as a function of U with
the free energy lowering by around 6 to 8 kcal/mol for U = 3.0 to 5.0 eV. Again, above this value
we see large changes relative to U = 0.0 eV, as the spin states of cobalt has changed.
Figure 2.29. A comparison of the relative free energies between intermediates (a) bare and CoH and (b) CoH and
CoH2 of the AB configuration of CoTHT as a function of U. Both reactions occur at the 1-cobalt site of the AB
configuration.
U (eV)
(a)
(b)
ΔG (kcal/mol)
CoH à CoH
2
-30
-20
-10
0
10
20
-30
-20
-10
0
10
20
30
40
0 1 2 3 4 5 6 7 8
ΔG (kcal/mol)
bare à CoH
-8
-6
-4
-2
0
2
0 1 2 3 4 5
U (eV)
ΔG (kcal/mol)
0
2
4
6
8
10
0 1 2 3 4 5
ΔG (kcal/mol)
U (eV)
70
Taken together, these results suggest the main conclusions regarding the HER mechanism
would not change if a modest value of U (3.0 ~ 5.0 eV) were used. Therefore, CoTHT is a highly
intrinsically active material for the HER, whose apparent activity is mainly limited by electron
conductivity, which is why significantly higher activities are obtained when carbon black is
introduced into the catalyst mixture.
2.5 References
(1) Chu, S.; Majumdar, A. Opportunities and Challenges for a Sustainable Energy Future.
Nature 2012, 488 (7411), 294–303.
(2) Kim, D.; Sakimoto, K. K.; Hong, D.; Yang, P. Artificial Photosynthesis for Sustainable
Fuel and Chemical Production. Angew. Chem., Int. Ed. 2015, 54 (11), 3259–3266.
(3) She, Z. W.; Kibsgaard, J.; Dickens, C. F.; Chorkendorff, I.; Nørskov, J. K.; Jaramillo, T. F.
Combining Theory and Experiment in Electrocatalysis: Insights into Materials Design. Science
2017, 355 (6321), eaad4998.
(4) LeValley, T. L.; Richard, A. R.; Fan, M. The Progress in Water Gas Shift and Steam
Reforming Hydrogen Production Technologies – A Review. Int. J. Hydrogen Energy 2014, 39
(30), 16983–17000.
(5) Roger, I.; Shipman, M. A.; Symes, M. D. Earth-Abundant Catalysts for Electrochemical
and Photoelectrochemical Water Splitting. Nat. Rev. Chem. 2017, 1 (1), 0003.
(6) Queyriaux, N.; Kaeffer, N.; Morozan, A.; Chavarot-Kerlidou, M.; Artero, V. Molecular
Cathode and Photocathode Materials for Hydrogen Evolution in Photoelectrochemical Devices. J.
Photochem. Photobiol. C Photochem. Rev. 2015, 25, 90–105.
(7) McKone, J. R.; Marinescu, S. C.; Brunschwig, B. S.; Winkler, J. R.; Gray, H. B. Earth-
Abundant Hydrogen Evolution Electrocatalysts. Chem. Sci. 2014, 5 (3), 865–878.
(8) Thoi, V. S.; Sun, Y.; Long, J. R.; Chang, C. J. Complexes of Earth-Abundant Metals for
Catalytic Electrochemical Hydrogen Generation under Aqueous Conditions. Chem. Soc. Rev. 2013,
42 (6), 2388–2400.
(9) Downes, C. A.; Marinescu, S. C. Electrocatalytic Metal–Organic Frameworks for Energy
Applications. ChemSusChem 2017, 10 (22), 4374–4392.
(10) Solomon, M. B.; Church, T. L.; D’Alessandro, D. M. Perspectives on Metal–Organic
Frameworks with Intrinsic Electrocatalytic Activity. CrystEngComm 2017, 19 (29), 4049–4065.
(11) Maeda, H.; Sakamoto, R.; Nishihara, H. Coordination Programming of Two-Dimensional
Metal Complex Frameworks. Langmuir 2016, 32 (11), 2527–2538.
71
(12) Sakamoto, R.; Takada, K.; Pal, T.; Maeda, H.; Kambe, T.; Nishihara, H. Coordination
Nanosheets (CONASHs): Strategies, Structures and Functions. Chem. Commun. 2017, 53 (43),
5781–5801.
(13) Ko, M.; Mendecki, L.; Mirica, K. A. Conductive Two-Dimensional Metal–Organic
Frameworks as Multifunctional Materials. Chem. Commun. 2018, 54 (57), 7873–7891.
(14) Kusamoto, T.; Nishihara, H. Zero-, One- and Two-Dimensional Bis(Dithiolato)Metal
Complexes with Unique Physical and Chemical Properties. Coord. Chem. Rev. 2019, 380, 419–
439.
(15) Clough, A. J.; Yoo, J. W.; Mecklenburg, M. H.; Marinescu, S. C. Two-Dimensional Metal–
Organic Surfaces for Efficient Hydrogen Evolution from Water. J. Am. Chem. Soc. 2015, 137 (1),
118–121.
(16) Dong, R.; Pfeffermann, M.; Liang, H.; Zheng, Z.; Zhu, X.; Zhang, J.; Feng, X. Large-Area,
Free-Standing, Two-Dimensional Supramolecular Polymer Single-Layer Sheets for Highly
Efficient Electrocatalytic Hydrogen Evolution. Angew. Chem., Int. Ed. 2015, 54 (41), 12058–
12063.
(17) Dong, R.; Zheng, Z.; Tranca, D. C.; Zhang, J.; Chandrasekhar, N.; Liu, S.; Zhuang, X.;
Seifert, G.; Feng, X. Immobilizing Molecular Metal Dithiolene-Diamine Complexes on 2D Metal-
Organic Frameworks for Electrocatalytic H2 Production. Chem. Eur. J. 2017, 23 (10), 2255–2260.
(18) Lu, X. F.; Liao, P. Q.; Wang, J. W.; Wu, J. X.; Chen, X. W.; He, C. T.; Zhang, J. P.; Li, G.
R.; Chen, X. M. An Alkaline-Stable, Metal Hydroxide Mimicking Metal-Organic Framework for
Efficient Electrocatalytic Oxygen Evolution. J. Am. Chem. Soc. 2016, 138 (27), 8336–8339.
(19) Wang, L.; Wu, Y.; Cao, R.; Ren, L.; Chen, M.; Feng, X.; Zhou, J.; Wang, B. Fe/Ni Metal-
Organic Frameworks and Their Binder-Free Thin Films for Efficient Oxygen Evolution with Low
Overpotential. ACS Appl. Mater. Interfaces 2016, 8 (26), 16736–16743.
(20) Zhao, S.; Wang, Y.; Dong, J.; He, C. T.; Yin, H.; An, P.; Zhao, K.; Zhang, X.; Gao, C.;
Zhang, L.; et al. Ultrathin Metal-Organic Framework Nanosheets for Electrocatalytic Oxygen
Evolution. Nat. Energy 2016, 1 (12), 16184.
(21) Duan, J.; Chen, S.; Zhao, C. Ultrathin Metal-Organic Framework Array for Efficient
Electrocatalytic Water Splitting. Nat. Commun. 2017, 8, 15341.
(22) Miner, E. M.; Fukushima, T.; Sheberla, D.; Sun, L.; Surendranath, Y.; Dincă, M.
Electrochemical Oxygen Reduction Catalysed by Ni3(Hexaiminotriphenylene)2. Nat. Commun.
2016, 7 (1), 10942.
(23) Miner, E. M.; Gul, S.; Ricke, N. D.; Pastor, E.; Yano, J.; Yachandra, V. K.; Van Voorhis,
T.; Dincă, M. Mechanistic Evidence for Ligand-Centered Electrocatalytic Oxygen Reduction with
the Conductive MOF Ni3(Hexaiminotriphenylene)2. ACS Catal. 2017, 7 (11), 7726–7731.
(24) Miner, E. M.; Wang, L.; Dincă, M. Modular O2 Electroreduction Activity in Triphenylene-
72
Based Metal–Organic Frameworks. Chem. Sci. 2018, 9 (29), 6286–6291.
(25) Usov, P. M.; Huffman, B.; Epley, C. C.; Kessinger, M. C.; Zhu, J.; Maza, W. A.; Morris,
A. J. Study of Electrocatalytic Properties of Metal–Organic Framework PCN-223 for the Oxygen
Reduction Reaction. ACS Appl. Mater. Interfaces 2017, 9 (39), 33539–33543.
(26) Liu, X. H.; Hu, W. L.; Jiang, W. J.; Yang, Y. W.; Niu, S.; Sun, B.; Wu, J.; Hu, J. S. Well-
Defined Metal-O6 in Metal-Catecholates as a Novel Active Site for Oxygen Electroreduction. ACS
Appl. Mater. Interfaces 2017, 9 (34), 28473–28477.
(27) Lions, M.; Tommasino, J.-B.; Chattot, R.; Abeykoon, B.; Guillou, N.; Devic, T.;
Demessence, A.; Cardenas, L.; Maillard, F.; Fateeva, A. Insights into the Mechanism of
Electrocatalysis of the Oxygen Reduction Reaction by a Porphyrinic Metal Organic Framework.
Chem. Commun. 2017, 53 (48), 6496–6499.
(28) Lin, S.; Diercks, C. S.; Zhang, Y.-B.; Kornienko, N.; Nichols, E. M.; Zhao, Y.; Paris, A.
R.; Kim, D.; Yang, P.; Yaghi, O. M.; et al. Covalent Organic Frameworks Comprising Cobalt
Porphyrins for Catalytic CO2 Reduction in Water. Science 2015, 349 (6253), 1208–1213.
(29) Kornienko, N.; Zhao, Y.; Kley, C. S.; Zhu, C.; Kim, D.; Lin, S.; Chang, C. J.; Yaghi, O.
M.; Yang, P. Metal–Organic Frameworks for Electrocatalytic Reduction of Carbon Dioxide. J. Am.
Chem. Soc. 2015, 137 (44), 14129–14135.
(30) Diercks, C. S.; Lin, S.; Kornienko, N.; Kapustin, E. A.; Nichols, E. M.; Zhu, C.; Zhao, Y.;
Chang, C. J.; Yaghi, O. M. Reticular Electronic Tuning of Porphyrin Active Sites in Covalent
Organic Frameworks for Electrocatalytic Carbon Dioxide Reduction. J. Am. Chem. Soc. 2018, 140
(3), 1116–1122.
(31) Matheu, R.; Gutierrez-Puebla, E.; Monge, M. Á.; Diercks, C. S.; Kang, J.; Prévot, M. S.;
Pei, X.; Hanikel, N.; Zhang, B.; Yang, P.; et al. Three-Dimensional Phthalocyanine Metal-
Catecholates for High Electrochemical Carbon Dioxide Reduction. J. Am. Chem. Soc. 2019, 141
(43), 17081–17085.
(32) Hod, I.; Sampson, M. D.; Deria, P.; Kubiak, C. P.; Farha, O. K.; Hupp, J. T. Fe-Porphyrin-
Based Metal–Organic Framework Films as High-Surface Concentration, Heterogeneous Catalysts
for Electrochemical Reduction of CO2. ACS Catal. 2015, 5 (11), 6302–6309.
(33) Johnson, E. M.; Haiges, R.; Marinescu, S. C. Covalent-Organic Frameworks Composed of
Rhenium Bipyridine and Metal Porphyrins: Designing Heterobimetallic Frameworks with Two
Distinct Metal Sites. ACS Appl. Mater. Interfaces 2018, 10 (44), 37919–37927.
(34) Meng, Z.; Luo, J.; Li, W.; Mirica, K. A. Hierarchical Tuning of the Performance of
Electrochemical Carbon Dioxide Reduction Using Conductive Two-Dimensional
Metallophthalocyanine Based Metal-Organic Frameworks. J. Am. Chem. Soc. 2020, 142 (52),
21656–21669.
(35) Downes, C. A.; Clough, A. J.; Chen, K.; Yoo, J. W.; Marinescu, S. C. Evaluation of the H2
Evolving Activity of Benzenehexathiolate Coordination Frameworks and the Effect of Film
73
Thickness on H2 Production. ACS Appl. Mater. Interfaces 2018, 10 (2), 1719–1727.
(36) Huang, X.; Yao, H.; Cui, Y.; Hao, W.; Zhu, J.; Xu, W.; Zhu, D. Conductive Copper
Benzenehexathiol Coordination Polymer as a Hydrogen Evolution Catalyst. ACS Appl. Mater.
Interfaces 2017, 9 (46), 40752–40759.
(37) Sun, X.; Wu, K.-H.; Sakamoto, R.; Kusamoto, T.; Maeda, H.; Ni, X.; Jiang, W.; Liu, F.;
Sasaki, S.; Masunaga, H.; et al. Bis(Aminothiolato)Nickel Nanosheet as a Redox Switch for
Conductivity and an Electrocatalyst for the Hydrogen Evolution Reaction. Chem. Sci. 2017, 8 (12),
8078–8085.
(38) Downes, C. A.; Marinescu, S. C. Efficient Electrochemical and Photoelectrochemical H2
Production from Water by a Cobalt Dithiolene One-Dimensional Metal–Organic Surface. J. Am.
Chem. Soc. 2015, 137 (43), 13740–13743.
(39) Downes, C. A.; Marinescu, S. C. One Dimensional Metal Dithiolene (M = Ni, Fe, Zn)
Coordination Polymers for the Hydrogen Evolution Reaction. Dalt. Trans. 2016, 45 (48), 19311–
19321.
(40) Downes, C. A.; Marinescu, S. C. Bioinspired Metal Selenolate Polymers with Tunable
Mechanistic Pathways for Efficient H 2 Evolution. ACS Catal. 2017, 7 (1), 848–854.
(41) Downes, C. A.; Marinescu, S. C. Understanding Variability in the Hydrogen Evolution
Activity of a Cobalt Anthracenetetrathiolate Coordination Polymer. ACS Catal. 2017, 7 (12),
8605–8612.
(42) Wang, L.; Tranca, D. C.; Zhang, J.; Qi, Y.; Sfaelou, S.; Zhang, T.; Dong, R.; Zhuang, X.;
Zheng, Z.; Seifert, G. Toward Activity Origin of Electrocatalytic Hydrogen Evolution Reaction on
Carbon-Rich Crystalline Coordination Polymers. Small 2017, 13 (37), 1700783.
(43) Ji, Z.; Trickett, C.; Pei, X.; Yaghi, O. M. Linking Molybdenum-Sulfur Clusters for
Electrocatalytic Hydrogen Evolution. J. Am. Chem. Soc. 2018, 140 (42), 13618–13622.
(44) Sheberla, D.; Sun, L.; Blood-Forsythe, M. A.; Er, S.; Wade, C. R.; Brozek, C. K.; Aspuru-
Guzik, A.; Dincă, M. High Electrical Conductivity in Ni3(2,3,6,7,10,11-Hexaiminotriphenylene)2,
a Semiconducting Metal–Organic Graphene Analogue. J. Am. Chem. Soc. 2014, 136 (25), 8859–
8862.
(45) Clough, A. J.; Skelton, J. M.; Downes, C. A.; de la Rosa, A. A.; Yoo, J. W.; Walsh, A.;
Melot, B. C.; Marinescu, S. C. Metallic Conductivity in a Two-Dimensional Cobalt Dithiolene
Metal–Organic Framework. J. Am. Chem. Soc. 2017, 139 (31), 10863–10867.
(46) Solis, B. H.; Hammes-Schiffer, S. Computational Study of Anomalous Reduction
Potentials for Hydrogen Evolution Catalyzed by Cobalt Dithiolene Complexes. J. Am. Chem. Soc.
2012, 134 (37), 15253–15256.
(47) Panetier, J. A.; Letko, C. S.; Tilley, T. D.; Head-Gordon, M. Computational
Characterization of Redox Non-Innocence in Cobalt-Bis(Diaryldithiolene)-Catalyzed Proton
74
Reduction. J. Chem. Theory Comput. 2016, 12 (1), 223–230.
(48) 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 (26), 9364–9376.
(49) Wang, Y.; Liu, X.; Liu, J.; Al-Mamun, M.; Wee-Chung Liew, A.; Yin, H.; Wen, W.; Zhong,
Y. L.; Liu, P.; Zhao, H. Electrolyte Effect on Electrocatalytic Hydrogen Evolution Performance of
One-Dimensional Cobalt-Dithiolene Metal-Organic Frameworks: A Theoretical Perspective. ACS
Appl. Energy Mater. 2018, 1 (4), 1688-1694.
(50) Nørskov, J. K.; Rossmeisl, J.; Logadottir, A.; Lindqvist, L.; Kitchin, J. R.; Bligaard, T.;
Jónsson, H. Origin of the Overpotential for Oxygen Reduction at a Fuel-Cell Cathode. J. Phys.
Chem. B 2004, 108 (46), 17886–17892.
(51) Skúlason, E.; Karlberg, G. S.; Rossmeisl, J.; Bligaard, T.; Greeley, J.; Jónsson, H.; Nørskov,
J. K. Density Functional Theory Calculations for the Hydrogen Evolution Reaction in an
Electrochemical Double Layer on the Pt(111) Electrode. Phys. Chem. Chem. Phys. 2007, 9 (25),
3241–3250.
(52) Li, Y.; Wang, H.; Xie, L.; Liang, Y.; Hong, G.; Dai, H. MoS2 Nanoparticles Grown on
Graphene: An Advanced Catalyst for the Hydrogen Evolution Reaction. J. Am. Chem. Soc. 2011,
133 (19), 7296–7299.
(53) Shinagawa, T.; Garcia-Esparza, A. T.; Takanabe, K. Insight on Tafel Slopes from a
Microkinetic Analysis of Aqueous Electrocatalysis for Energy Conversion. Sci. Rep. 2015, 5 (1),
13801.
(54) Navarro-Flores, E.; Chong, Z.; Omanovic, S. Characterization of Ni, NiMo, NiW and NiFe
Electroactive Coatings as Electrocatalysts for Hydrogen Evolution in an Acidic Medium. J. Mol.
Catal. A Chem. 2005, 226 (2), 179–197.
(55) Vrubel, H.; Moehl, T.; Grätzel, M.; Hu, X. Revealing and Accelerating Slow Electron
Transport in Amorphous Molybdenum Sulphide Particles for Hydrogen Evolution Reaction. Chem.
Commun. 2013, 49 (79), 8985.
(56) Pham, K.-C.; Chang, Y.-H.; McPhail, D. S.; Mattevi, C.; Wee, A. T. S.; Chua, D. H. C.
Amorphous Molybdenum Sulfide on Graphene–Carbon Nanotube Hybrids as Highly Active
Hydrogen Evolution Reaction Catalysts. ACS Appl. Mater. Interfaces 2016, 8 (9), 5961–5971.
(57) Merki, D.; Vrubel, H.; Rovelli, L.; Fierro, S.; Hu, X. Fe, Co, and Ni Ions Promote the
Catalytic Activity of Amorphous Molybdenum Sulfide Films for Hydrogen Evolution. Chem. Sci.
2012, 3 (8), 2515–2525.
(58) Staniland, S. S.; Rawlings, A.; Bramble, J.; Tolosa, J.; Wilson, O.; García-Martínez, J. C.;
Binns, C. Novel Methods for the Synthesis of Magnetic Nanoparticles. Molecular Biology, 2014,
8, 85–128.
75
(59) Lin, S.; Usov, P. M.; Morris, A. J. The Role of Redox Hopping in Metal–Organic
Framework Electrocatalysis. Chem. Commun. 2018, 54 (51), 6965–6974.
(60) Fishman, M.; Zhuang, H. L.; Mathew, K.; Dirschka, W.; Hennig, R. G. Accuracy of
Exchange-Correlation Functionals and Effect of Solvation on the Surface Energy of Copper. Phys.
Rev. B 2013, 87 (24), 245402.
(61) Mathew, K.; Sundararaman, R.; Letchworth-Weaver, K.; Arias, T. A.; Hennig, R. G.
Implicit Solvation Model for Density-Functional Study of Nanocrystal Surfaces and Reaction
Pathways. J. Chem. Phys. 2014, 140 (8), 084106.
(62) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient Approximation Made Simple.
Phys. Rev. Lett. 1996, 77 (18), 3865–3868.
(63) Buendía, F.; Beltrán, M. R. Theoretical Study of Hydrogen Adsorption on Co Clusters.
Comput. Theor. Chem. 2013, 1021, 183–190.
(64) Baker-Hawkes, M. J.; Dori, Z.; Eisenberg, R.; Gray, H. B. The Crystal and Molecular
Structure of the Tetra-n-Butylammonium Salt of the Dianionic Dimer of Bis(1,2,3,4-
Tetrachlorobenzene-5,6-Dithiolato)Cobaltate. J. Am. Chem. Soc. 1968, 90 (16), 4253–4259.
(65) Alvarez, S.; Vicente, R.; Hoffmann, R. Dimerization and Stacking in Transition-Metal
Bisdithiolenes and Tetrathiolates. J. Am. Chem. Soc. 1985, 107 (22), 6253–6277.
(66) Eisenberg, R.; Gray, H. B. Noninnocence in Metal Complexes: A Dithiolene Dawn. Inorg.
Chem. 2011, 50 (20), 9741–9751.
(67) Nørskov, J. K.; Bligaard, T.; Logadottir, A.; Kitchin, J. R.; Chen, J. G.; Pandelov, S.;
Stimming, U. Trends in the Exchange Current for Hydrogen Evolution. J. Electrochem. Soc. 2005,
152 (3), J23–J26.
(68) Hinnemann, B.; Moses, P. G.; Bonde, J.; Jørgensen, K. P.; Nielsen, J. H.; Horch, S.;
Chorkendorff, I.; Nørskov, J. K. Biomimetic Hydrogen Evolution: MoS2 Nanoparticles as Catalyst
for Hydrogen Evolution. J. Am. Chem. Soc. 2005, 127 (15), 5308–5309.
(69) Gautier, S.; Steinmann, S. N.; Michel, C.; Fleurat-Lessard, P.; Sautet, P. Molecular
Adsorption at Pt(111). How Accurate Are DFT Functionals? Phys. Chem. Chem. Phys. 2015, 17
(43), 28921–28930.
(70) Thomas, J. G. N. Kinetics of Electrolytic Hydrogen Evolution and the Adsorption of
Hydrogen by Metals. Trans. Faraday Soc. 1961, 57 (0), 1603–1611.
(71) Chen, Z.; Cummins, D.; Reinecke, B. N.; Clark, E.; Sunkara, M. K.; Jaramillo, T. F. Core–
Shell MoO3–MoS2 Nanowires for Hydrogen Evolution: A Functional Design for Electrocatalytic
Materials. Nano Lett. 2011, 11 (10), 4168–4175.
(72) Zhang, W.; Haddad, A. Z.; Garabato, B. D.; Kozlowski, P. M.; Buchanan, R. M.;
Grapperhaus, C. A. Translation of Ligand-Centered Hydrogen Evolution Reaction Activity and
76
Mechanism of a Rhenium-Thiolate from Solution to Modified Electrodes: A Combined
Experimental and Density Functional Theory Study. Inorg. Chem. 2017, 56 (4), 2177–2187.
(73) Deblase, C. R.; Silberstein, K. E.; Truong, T.-T.; Abruñ, H. D.; Dichtel, W. R. β-
Ketoenamine-Linked Covalent Organic Frameworks Capable of Pseudocapacitive Energy Storage.
J. Am. Chem. Soc 2013, 135, 53.
(74) Deblase, C. R.; Hernández-Burgos, K.; Silberstein, K. E.; Rodríguez-Calero, G. G.; Bisbey,
R. P.; Abruña, H. D.; Dichtel, W. R. Rapid and Efficient Redox Processes within 2D Covalent
Organic Framework Thin Films. ACS Nano 2015, 9 (3), 3178–3183.
(75) Sakamoto, R.; Kambe, T.; Tsukada, S.; Takada, K.; Hoshiko, K.; Kitagawa, Y.; Okumura,
M.; Nishihara, H. Π-Conjugated Trinuclear Group-9 Metalladithiolenes With a Triphenylene
Backbone. Inorg. Chem. 2013, 52 (13), 7411–7416.
(76) Slater, J. C. A Simplification of the Hartree-Fock Method. Phys. Rev. 1951, 81 (3), 385–
390.
(77) Perdew, J. P.; Wang, Y. Accurate and Simple Analytic Representation of the Electron-Gas
Correlation Energy. Phys. Rev. B 1992, 45 (23), 13244–13249.
(78) Kresse, G.; Furthmüller, J. Efficiency of Ab-Initio Total Energy Calculations for Metals
and Semiconductors Using a Plane-Wave Basis Set. Comput. Mater. Sci. 1996, 6 (1), 15–50.
(79) Kresse, G.; Furthmüller, J. Efficient Iterative Schemes for Ab Initio Total-Energy
Calculations Using a Plane-Wave Basis Set. Phys. Rev. B 1996, 54 (16), 11169–11186.
(80) Kresse, G.; Hafner, J. Ab Initio Molecular Dynamics for Liquid Metals. Phys. Rev. B 1993,
47 (1), 558–561.
(81) Kresse, G.; Hafner, J. Ab Initio Molecular-Dynamics Simulation of the Liquid-Metal--
Amorphous-Semiconductor Transition in Germanium. Phys. Rev. B 1994, 49 (20), 14251–14269.
(82) Monkhorst, H. J.; Pack, J. D. Special Points for Brillouin-Zone Integrations. Phys. Rev. B
1976, 13 (12), 5188–5192.
(83) Letchworth-Weaver, K.; Arias, T. A. Joint Density Functional Theory of the Electrode-
Electrolyte Interface: Application to Fixed Electrode Potentials, Interfacial Capacitances, and
Potentials of Zero Charge. Phys. Rev. B 2012, 86 (7), 75140.
(84) Jinnouchi, R.; Anderson, A. B. Electronic Structure Calculations of Liquid-Solid Interfaces:
Combination of Density Functional Theory and Modified Poisson-Boltzmann Theory. Phys. Rev.
B 2008, 77 (24), 245417.
(85) Jones, G.; Jakobsen, J.; Shim, S.; Kleis, J.; Andersson, M.; Rossmeisl, J.; Abildpedersen,
F.; Bligaard, T.; Helveg, S.; Hinnemann, B. First Principles Calculations and Experimental Insight
into Methane Steam Reforming over Transition Metal Catalysts. J. Catal. 2008, 259 (1), 147–160.
77
(86) Cococcioni, M.; de Gironcoli, S. Linear Response Approach to the Calculation of the
Effective Interaction Parameters in the LDA+U Method. Phys. Rev. B 2005, 71 (3), 035105.
(87) Kulik, H. J. Perspective: Treating Electron over-Delocalization with the DFT+U Method.
J. Chem. Phys. 2015, 142 (24), 240901.
(88) Ghosh, S.; Singh, S. K.; Tewary, S.; Rajaraman, G. Enhancing the Double Exchange
Interaction in a Mixed Valence {VIII–VII} Pair: A Theoretical Perspective. Dalt. Trans. 2013, 42
(47), 16490.
(89) Radoń, M. Revisiting the Role of Exact Exchange in DFT Spin-State Energetics of
Transition Metal Complexes. Phys. Chem. Chem. Phys. 2014, 16 (28), 14479–14488.
(90) Mann, G. W.; Lee, K.; Cococcioni, M.; Smit, B.; Neaton, J. B. First-Principles Hubbard U
Approach for Small Molecule Binding in Metal-Organic Frameworks. J. Chem. Phys. 2016, 144
(17), 174104.
78
Chapter 3
Hydrogen Evolving Activity of Dithiolene-Based Metal-Organic Frameworks
with Mixed Cobalt and Iron Centers
A portion of this chapter has appeared in print:
Chen, K.; Downes, C. A.; Goodpaster, J. D.; Marinescu, S. C. “Hydrogen Evolving Activity of
Dithiolene-Based Metal-Organic Frameworks with Mixed Cobalt and Iron Centers”. Inorg. Chem.
2021, accepted.
79
3.1 Introduction
Metal-organic frameworks (MOFs) have attracted growing attention as a new class of
functional materials over the last two decades. Their unique properties, such as large surface area,
permanent porosity, structural regularity, and high synthetic tunability, endow MOFs with
potential for a wide spectrum of applications.
1
In the context of increased global energy
consumption and environmental issues, the development of MOF-based electrocatalysts has
become an emerging research topic in recent years.
2
The rigid structure and hybrid nature of MOFs
make them an excellent platform for the heterogenization of molecular catalysts, a novel strategy
to construct catalytic materials with improved stability and reusability.
2,3
More importantly, this
strategy allows for the understanding and optimization of the activity of a bulk material at the
molecular level. To date, many MOF-based electrocatalytic systems have been reported for various
reactions, including the hydrogen evolution reaction (HER),
4–14
oxygen evolution reaction
(OER),
15–20
oxygen reduction reaction (ORR),
21–26
and CO2 reduction (CO2RR).
27–33
Dithiolene-based MOFs are among the most active MOFs for electrocatalytic HER.
4–9,14,34
This
class of 2D MOFs are typically constructed by linking single-metal centers with trinucleating
ligands such as benzenehexathiolate (BHT) and triphenylene-2,3,6,7,10,11-hexathiolate (THT).
MOFs incorporating cobalt or nickel as active metal centers have been extensively studied for the
HER,
5,7,9
with CoTHT displaying the lowest overpotential (143 mV) to reach the benchmarking
metric of 10 mA/cm
2
, a Tafel slope of only 71 mV/dec, and with near unity Faradaic efficiency
(FE) for H2 evolution.
4
However, the electrocatalytic activity of the iron dithiolene-based MOFs
remains relatively unexplored. The molecular iron dithiolene complexes have been shown to
exhibit moderate activity towards visible-light-driven photocatalytic HER with high turnover
numbers in conjunction with CdSe quantum dots as a photosensitizer and ascorbic acid as the
80
electron donor.
35
Additionally, prior report has suggested that the iron dithiolene coordination
polymer based on the BHT ligand (FeBHT) was capable of performing HER with decent activity,
characterized by a Tafel slope of 119 mV/dec and an overpotential of 473 mV to reach 10 mA/cm
2
in fully aqueous mdia.
8
Although the electrocatalytic activity of FeTHT has not been reported,
recent studies regarding its physical properties have suggested that FeTHT films exhibit similar
room-temperature electrical conductivity compared to that of the CoTHT films, with values
ranging between 0.02 and 0.3 S×cm
-1
.
36,37
The room-temperature mobility of FeTHT was reported
to be ~220 cm
2
V
-1
s
-1
, a record high value for MOFs.
38
Based on these previous studies and the fact
that iron is inexpensive and more abundant than cobalt, the investigation of the HER performance
of FeTHT is of great interest.
The hybrid nature of MOFs allows for a high degree of variability and multivariate synthetic
tunability. Such tunability has been explicitly studied by Yaghi and coworkers where MOFs were
shown to incorporate up to 8 unique ligands and 10 different metals without disturbing the
crystalline structure of the overall framework.
39,40
In the context of electrocatalysis, introducing
multiple metal centers into the same structure is a common strategy to improve the catalytic
performance.
41–43
For example, it is well-known that the incorporation of Fe into Co or Ni
(oxy)hydroxides can dramatically improve the OER performance of the catalysts.
44,45
Chang and
coworkers also reported the synthesis of the multivariate Co/Cu COF-367 for CO2 reduction. The
incorporation of Cu allowed for dilution of the active Co sites and substantially enhanced activity
on a per-cobalt basis.
29
Hence, we sought to utilize the high synthetic tunability of MOFs to design
a series of mixed Co/Fe MOF electrocatalysts and investigate the influence of different metal
moieties on the HER activity.
81
Herein, a series of dithiolene-based MOFs incorporating Co and Fe metal centers with variable
ratios were successfully synthesized and characterized. The presence of both metals was analyzed
by various spectroscopic techniques. The electrocatalytic HER activity of the materials was studied
in pH 1.3 aqueous solutions. It is found that, unlike its cobalt analogue, FeTHT exhibits minimal
activity towards the HER. More interestingly, the incorporation of the Fe centers also drastically
reduces the activity and stability of the mixed-metal frameworks. It is proposed that the FeTHT
undergoes alternative Faradaic processes under catalytic conditions, which alter its local structure
and electrochemical behavior, resulting in a material with diminished activity for the HER.
3.2 Results and Discussion
3.2.1 Synthesis and Characterization
Prior reports have suggested that both CoTHT and FeTHT can be synthesized through the
liquid-liquid interfacial synthesis, which was proposed to lead to materials with high structural
similarity.
5,36,37
Therefore, it was expected that similar synthetic methodology would be applicable
to the mixed-metal MOFs. To tune the stoichiometries of the metals in the resulting MOFs, metal
precursors were prepared as a series of solutions of CoCl2 and FeCl2 with different molar ratios
and underwent similar liquid-liquid interfacial synthetic procedure (see section 3.4 for
experimental details). The resulting mixed-metal frameworks were collected as powder, solvent
exchanged with water (3´) and methanol (3´), then dried under vacuum. We confirmed the
incorporation and determined the concentration of each metal within the frameworks by
inductively coupled plasma optical emission spectroscopy (ICP-OES) measurements with results
summarized in Table 3.1. Analysis by ICP-OES reveals that both Co and Fe are successfully
incorporated into the framework through this synthetic method. Additionally, altering the feeding
molar ratios of the precursors results in MOFs with varied metal compositions, and such variability
82
is relatively reproducible by comparing the composition of the materials prepared from three
individual trials (Table 3.1). It is also found that the feeding Fe:Co molar ratio in the precursor
solutions is typically much higher than that of the resulting MOFs. For example, when the
precursor contains the same amount of CoCl2 and FeCl2, the resulting product has an Fe content
as low as 5.29% (MOF 3 in Table 3.1). Even when FeCl2 was 1000 times more than CoCl2 in the
precursor, Fe still makes up only 76.1% of the total metal centers in the product (MOF 1 in Table
3.1). When the precursor solution contains FeCl2 and CoCl2 in the 100:1 ratio, the resulting product
has an Fe content of 46.1% (MOF 2 in Table 3.1). These results suggest that the incorporation rate
of Fe into the framework is much slower compared to that of Co. During synthesis, the slower
growth of the FeTHT framework was visually observed.
Table 3.1 Metal content in mixed metal THT-based MOFs determined by ICP-OES
Material
Feeding
Fe:Co
molar ratio
Product Fe:Co Molar Ratio
Average Fe:Co
Ratio
Average
Fe%
Trial 1 Trial 2 Trial 3
1 1000 3.21 3.14 3.22 3.19 ± 0.04 76.1%
2 100.0 9.52×10
-1
7.58×10
-1
8.55×10
-1
(8.54 ± 0.97) × 10
-1
46.1%
3 1.00 5.24×10
-2
6.13×10
-2
5.46×10
-2
(5.61 ± 0.47) × 10
-2
5.29%
Powder X-ray diffraction (PXRD) measurements confirm the crystalline structure of the as-
synthesized series of THT-based MOFs (Figure 3.1a). MOFs 1, 2, and 3 exhibit very similar PXRD
patterns, with prominent peaks at 1.2
o
, 2.4
o
, 3.2
o
and 4.2
o
, indicating a similar crystalline structure
regardless of the metal compositions. The PXRD patterns of the mixed-metal MOFs (1, 2, 3) also
match with those of the Co and Fe-only MOFs, as well as with their simulated PXRD patterns. To
confirm the structural similarity of the MOF series, FTIR spectrum was acquired for each MOF
and compared to the simulated spectrum of a cobalt dithiolene molecular model (see section 3.4.7
for modelling details). As shown in Figure 3.1b, each MOF displays a very similar FTIR pattern
83
and also shows relatively good agreement with the calculated spectrum. The only notable
difference between the experimental and computational result is for the C–H stretching modes,
which are experimentally observed at 2970–2870 cm
-1
, but are calculated to be at 3150–3020 cm
-
1
. This is likely due to the discrepancy between the simulation model and the actual extended
structure, which is discussed in detail in section 3.4.7. While the molecular model also influences
the exact peak positions of the remainder of the spectrum, the number of peaks and relative
positions allows for assignment of the spectrum as follows. The peaks at 1600–1300 cm
-1
correspond to C–C bond stretches, the peaks at 1170–970 cm
-1
correspond to C–H in-plane
bending, the peaks at 925–800 cm
-1
correspond to C–H out-of-plane bending, and the peaks at
680–525 cm
-1
correspond to C–C out-of-plane bending. Finally, the S–H bond stretches at around
2520 cm
-1
observed in the experimental spectra is likely caused by the residual unreacted THT
ligand in the sample, and the absorption at ~3300 cm
-1
is attributed to residual water that was
introduced during the synthesis.
Figure 3.1 (a) Synchrotron and simulated PXRD patterns of THT-based MOFs, X-ray wavelength: 0.4127 Å. (b) DFT
simulated FTIR spectrum of the cobalt dithiolene molecular model (black trace) and the experimental FTIR spectra
of the MOF series.
The chemical composition of the materials was further probed by X-ray photoelectron
spectroscopy (XPS), which indicates the presence of Fe, Co, S, and C. The peak characteristics of
each element appear similar for materials across the series, yet with variable peak intensities
84
(Figure 3.2). Hence, 2 will be used herein as a representative example for discussions of the XPS
features (Figure 3.3). Two sets of peaks are observed in the Co 2p region, with binding energies
of ~779 eV and ~794 eV. Deconvolution of these signals generates 5 peaks. The peaks at 779.8
and 794.8 eV are assigned to Co
III
, the peaks at 778.6 and 793.6 eV are assigned to Co
II
, and the
broad peak at 782.2 eV is assigned to the Fe LMM Auger peak generated by the Al Kα source. For
the Fe 2p regions, deconvolution of the signals at ~708 eV and ~721 eV also gives rise to a mixed
valency of Fe
II
/Fe
III
, and the interference of the Co Auger peak (~713.2 eV) is again present.
Excluding the Auger peaks, the characteristics of the Co and Fe regions are similar to those
reported for materials with solely Co or Fe as the metal center, suggesting similar coordination
environments.
5,36
The Co 2p region of CoTHT displays a set of broad peaks at around 779 and
794 eV, corresponding to the mixed-oxidation states of Co
II/III
induced by the non-innocent nature
of the dithiolene ligands.
5
On the other hand, the Fe 2p region of FeTHT also shows a broad set
of peaks around 708 eV and 721 eV, again fitted to the mixed-valency of Fe
II/III
.
36
Therefore, the
similarities between the XPS results of the mixed-metal MOFs and the Co and Fe only MOFs
further confirm that both metals are incorporated into the framework structure with the expected
coordination environment.
The morphology of the materials was evaluated by scanning electron-microscopy (SEM). As
shown in Figure 3.4, materials with different metal ratios all display a sheet-like morphology,
consistent with the previous reports on CoTHT and FeTHT.
5,37
The energy-dispersive X-ray
spectroscopy (EDX) of 2 reveals the uniform distribution of Co, Fe, and S throughout the material
(Figure 3.5), further confirming the chemical composition.
85
Figure 3.2 High resolution XPS spectra of 1, 2, and 3: (a) Co 2p region; (b) Fe 2p region; (c) C 1s region; (d) S 2s
region.
Figure 3.3 High-resolution XPS spectra of 2: (b) Co 2p core level; (c) Fe 2p core level.
Figure 3.4 Top-down SEM images of (a) 1, (b) 2, and (c) 3.
86
Figure 3.5 (a) Top-down SEM image of 2. EDX spectra of 2 revealing the presence and homogeneous distribution of
(b) S, (c) Co, and (d) Fe.
3.2.2 Electrocatalytic H2 Evolution Studies.
The electrocatalytic HER performance of FeTHT was first investigated by linear sweep
voltammetry (LSV) in N2-saturated pH 1.3 H2SO4 aqueous solutions using a three-electrode
configuration (see section 3.4.5 for details). In contrast to its cobalt counterpart, FeTHT displays
poor electrocatalytic HER activity within the investigated electrochemical window (Figure 3.6).
A catalytic onset is seen at -0.440 V versus RHE and at large overpotentials (h > 0.70 V) the
maximum current density achieved is only 3.5 mA·cm
-2
, far less than the benchmarking metric of
10 mA·cm
-2
for the HER. This poor performance is related to the large charge transfer resistance
(Rct) obtained by fitting the electrochemical impedance spectroscopy (EIS) data with the two-time
constant serial (2TS) model (Figure 3.7). Although Rct decreases at larger overpotentials (Table
3.2) as expected for an enhanced HER rate with increased catalytic driving force, the large Rct
87
at -0.518 V versus RHE (Rct = 923 Ω) is still too prohibitive to allow for efficient catalysis.
Sluggish charge transfer kinetics in the electrochemical window investigated thus precludes the
ability to achieve high electrocatalytic activity. Tafel analysis reveals a Tafel slope of 210 mV/dec
for FeTHT (Figure 3.6), suggesting a large energy barrier for the rate-limiting Volmer-like
discharge reaction.
46
In addition to the low initial electrocatalytic activity, successive cyclic voltammetry (CV)
experiments of FeTHT in a pH 1.3 solution revealed an unstable current response (Figure 3.8). A
predominant irreversible redox wave is observed around 0 V versus RHE in the first cathodic
sweep of the CV. During sequential scans, this feature gradually diminishes, shifting to more
anodic potentials, and eventually disappearing. This behavior suggests a possible change in the
electronic properties of the material under electrochemical polarization. Controlled potential
electrolysis (CPE) experiments were performed to further evaluate the stability and H2 production
efficiency of FeTHT. As shown in Figure 3.9, the current response is relatively stable yet
extremely low (~0.5 mA) even under a large overpotential input (η = 0.518 V). This behavior
further testifies the low activity of the catalyst as observed in the LSV studies (Figure 3.6). The
Faradaic efficiency (FE) for H2 production is quantified to be 72.2 ± 1.4%, suggesting more than
a quarter of the electrons were consumed in other alternative Faradaic processes, likely related to
the changing redox features observed in the CV studies (Figure 3.8). The EIS measurements reveal
a drastic increase in the Rct after CPE (Figure 3.9, Table 3.3), indicating a much hindered HER
kinetics after polarization. In view of the poor electrocatalytic activity and cyclability of FeTHT,
attention was then turned to the electrochemical study of the mixed-metal MOFs, with the
hypothesis that the inactive Fe sites could serve as the structural building blocks to dilute the Co
sites and enhance the activity on a per-Co basis, similar to the report by Chang and coworkers.
29
88
Figure 3.6 (a) Polarization curves of FeTHT (red), bare glassy carbon electrode (black), and components of the ink
(grey). (b) Tafel plot obtained from polarization curve. Scan rate: 5 mV/s, rotation rate of RDE: 2000 rpm, loading of
catalyst: 2.85 × 10
-7
molFe/cm
2
. All measurements were performed in N2-saturated pH 1.3 aqueous solutions.
Figure 3.7 (a) The two-time constant serial model (2TS) employed to fit the EIS response of FeTHT. (b) Nyquist
plots (markers) of FeTHT at various overpotentials with respective fits (solid lines). (c) and (d): Bode plots (markers)
of FeTHT at various overpotentials with respective fits (solid lines). All measurements were performed in N2-
saturated pH 1.3 aqueous solutions.
Table 3.2 Rct values of FeTHT derived from fitting EIS data to the 2TS model
η (mV) 368 418 468 518
Rct (Ω) 5677 3581 1605 922.7
89
Figure 3.8 Consecutive CV scans of FeTHT in a N2-saturated pH 1.3 aqueous solution, scan rate: 50 mV/s.
Figure 3.9 (a) 1 h CPE experiment of FeTHT at -0.518 V vs. RHE, FE: 72.2 ± 1.4%. (b) Nyquist plots (marks) fitted
to 2TS model (lines) showing EIS responses at different overpotentials before and after CPE.
Table 3.3 Characteristic values of FeTHT extracted from the EIS responses shown in Figure 3.9b.
Potential
(V vs. RHE)
Before CPE After CPE
RS (Ω) Rct (Ω) Cdl (mF) RS (Ω) Rct (Ω) Cdl (mF)
-0.418 5.48 210 1.65 5.84 940 1.64
-0.518 5.15 35.0 1.58 5.78 151 1.64
The HER activity of the mixed-metal MOFs 1-3 was first evaluated in pH 1.3 aqueous solutions
using LSV. As shown in Figure 3.10, upon increasing the Co content within the framework, the
overall activity is substantially enhanced, with 3 exhibiting the highest activity for the mixed-
metals MOFs. The catalytic onset is reduced from -0.440 V for FeTHT to -0.091 V for 3 and the
Tafel slope is lowered from 210 mV/dec for FeTHT to 80.2 mV/dec for 3 (characteristic values
90
summarized in Table 3.4). Such enhancement in activity correlates positively with the diminished
Rct value and enhanced double layer capacitance (Cdl) extracted from the EIS measurements
(Figure 3.11). For 1, the Rct is as high as 4190 W even at h = 368 mV, whereas for 3 the Rct is only
46.8 W at h = 168 mV. The Cdl values are used as an indicator of the electrocatalytically active
surface area, and are extrapolated from the EIS spectra instead of cyclic voltammograms to obtain
a more accurate description of the catalyst under the HER conditions.
47
The much larger Cdl value
(15.9 mF) for 3 compared to 1.12 mF for 1 indicates the increase in Co content gives rise to a
larger electrochemically active surface area (Table 3.5), suggesting Co metal center to be the actual
catalytically active site. However, upon normalizing the current by the bulk loading of Co, 2 and
3 display similar activity that is much higher than the activity of 1, and on par with that of the
CoTHT (Figure 3.10). This result was unanticipated, as one would expect that by diluting the
electroactive Co sites with catalytically inert Fe sites, the activity would be boosted on a per-cobalt
basis, as was reported for the multivariate Co/Cu COF-367 catalysts mentioned previously.
29
The
experimental results observed herein suggest that, instead of being an inert site and serving merely
as a diluting factor, the Fe site is playing a detrimental role and adversely influencing the overall
activity at high concentrations.
Figure 3.10 Polarization curves of FeTHT, mixed-metal MOF 1-3, and CoTHT with current normalized by (a)
geometric surface area and (b) bulk loading of the Co metal centers.
91
Table 3.4 Electrocatalytic properties of THT-based MOFs
Material FeTHT 1 2 3 CoTHT
Onset (V) vs. RHE -0.440 -0.400 -0.117 -0.091 -0.079
Tafel Slope (mV/dec)
b
210 101 95.2 80.2 81.6
Figure 3.11 Nyquist (a) and Bode plots (b and c) showing EIS responses of the materials at various overpotentials.
For FeTHT and 1, η = 368 mV; for 2 and 3, η = 168 mV
Table 3.5 Rct values of the materials derived from fitting EIS data to the 2TS model
Material FeTHT 1 2 3
η (mV) 368 368 168 168
Rct (Ω) 5677 4190 215 46.8
Cdl (mF)
a
0.261 1.12 2.67 15.9
[a] Cdl is calculated using the following equation: 𝐶
!"
= [
#$%
('
!
"#
( '
$%
"#
)
(#"')
]
*/,
To further investigate the stability and HER performance of the mixed-metal MOFs, CPE
experiments were performed. Figure 3.12 shows the current responses of all three materials during
a one hour electrolysis experiment at an overpotential of 0.418 V. MOF 1 exhibits a relatively
92
stable, yet low current response (~ 0.1 mA), whereas 3 is able to maintain a relatively high current
(~ 8 mA) over the period of one hour. The most interesting behavior is observed for 2, where the
current starts off high (~ 5 mA), but gradually declines over time (~ 1.6 mA at 60 min). The EIS
measurements of 2 before and after electrolysis reveal a drastic increase in Rct from 98.5 Ω to 940
Ω and a substantial decrease in Cdl from 19.3 mF to 4.25 mF (measured at η = 168 mV, Figure
3.13, Table 3.6). The changes in Rct and Cdl suggest that the decline in activity is a result of the
impeded charge transfer processes along with the diminished electrocatalytically active surface
area. The ICP-OES analysis of the post-CPE electrolyte solutions revealed that the amount of Co
and Fe in the solution is negligible, suggesting the diminished activity was not caused by the
dissolution of the catalyst. Additionally, the Faradaic efficiency (FE) for H2 generation for 1, 2,
and 3 are 70.3%, 72.8%, and 86.6%, respectively, whereas CoTHT has a near unity FE for H2
production as reported previously.
4,5
The extend of reduction in the FE correlates with the amount
of Fe present in the framework, indicating that the FeTHT domains within the framework are
going through alternative faradaic processes during electrolysis. The turnover frequency (TOF) of
the catalysts was estimated by dividing the amount of H2 produced (in moles) during the 1 hour
CPE experiment by the catalyst loading (in moles of metals) determined by ICP-OES (Table 3.7).
It is noteworthy that the as-calculated TOF is a lower-bound estimation of the actual TOF, as not
all metal sites are electrochemically accessible and therefore, active, during electrolysis. If we
assume only Co is responsible for catalysis, the corresponding TOFCo is 11.8 h
-1
, 85.7 h
-1
, and 132
h
-1
for 1, 2, and 3, respectively (Table 3.7). These results demonstrate that the TOFCo decreases
drastically when more Fe is present in the catalyst, further suggesting that Fe is inhibiting the
catalytic activity of the Co sites.
93
Figure 3.12 Controlled potential electrolysis (CPE) of 1 (red), 2 (blue), and 3 (green) at -0.418 V versus RHE.
Figure 3.13 Nyquist (a) and Bode (b) plots showing the EIS responses of 2 at η = 168 mV before and after CPE.
Table 3.6 Rct and Cdl values of 2 before and after CPE, extracted from EIS at η = 168 mV
Trial Rct (ohm) Cdl (mF)
Before CPE 98.5 19.3
After CPE 940 4.25
Table 3.7 Turnover frequency (TOF) calculations for 1, 2, and 3 based on the CPE experiment shown in Figure 3.12.
MOF 3 2 1
Charge Passed (C) 27.39 10.01 0.17
H2 Produced (mol) 1.23×10
"1
3.78×10
"2
6.25×10
"3
Fe (mol) 5.42×10
"4
3.64×10
"3
1.62×10
"3
Co (mol) 9.31×10
"3
4.41×10
"3
5.31×10
"4
Total metal (mol) 9.85×10
"3
8.04×10
"3
2.15×10
"3
TOFCo (h
-1
) 132 85.7 11.8
TOFtotal (h
-1
) 125 47.0 2.91
94
The poor stability of the mixed-metal MOFs is also revealed by the CV studies in pH 10
electrolyte solutions. As shown in Figure 3.14, when 1 is subjected to consecutive CV scans, the
intensity of the redox feature centered around -0.3 V versus RHE diminishes gradually. The
instability of the redox wave is also observed for 2 and 3, albeit less drastic within the CV timescale
in comparison to 1. The CoTHT framework displays redox feature at similar potential, yet with
better cyclability.
4
The correlation between the extent of reduction in the redox feature and the
content of Fe within the framework again suggests that the Fe domain is responsible for the
instability of the material under electrochemical conditions.
Figure 3.14 Consecutive CV scans of (a) FeTHT, (b) 1, (c) 2, (d) 3 in pH 10 electrolyte solutions, scan rate: 50 mV/s.
Physical characterizations were performed on FeTHT and MOF 2 after electrolysis to provide
insights into the electrochemical deactivation of the material. For 2, the PXRD pattern taken after
1 hour CPE experiment shows a drastically diminished peak intensity at 4.5° and 9.1°,
95
corresponding to the [100] and [200] reflection, respectively (Figure 3.15). In the case of FeTHT,
the intensity of these two peaks further diminishes to essentially an undetectable level (Figure
3.15). The change in PXRD patterns suggests that the electrolysis leads to a deteriorated coherence
within the 2D sheets, which is likely induced by the FeTHT domains, as the Co-only analogue is
capable of maintaining its structural integrity based on our previous study.
4
To detect the origin of
the structural changes, XPS measurements were applied to elucidate the chemical environment
after electrolysis. As shown in Figure 3.16, the Fe 2p region of the post-CPE FeTHT displays a
very broad peak ranging from 702 eV to 727 eV, suggesting the formation of new bonding
environments of Fe centers. Additionally, a new satellite peak appears at around 233 eV in the S
2s region (Figure 3.16), which could be contributed by the sulfonic acid group in the Nafion,
though the possibility of a modified S environment in the dithiolene linker cannot be excluded.
For post-CPE analysis of MOF 2, the Fe 2p and S 2s regions display similar behaviors compared
to those of FeTHT (Figure 3.17). The Co 2p peaks are present but hard to distinguish, because of
the sloped baseline of the region, which is likely resulted from the interference of the Fe Auger
series. Nevertheless, previous work with the Co-only analogue has suggested that electrolysis does
not alter the Co 2p XPS features.
4
In short, the XPS characterizations indicate the transformation
of the bonding environment for Fe, and possibly S for FeTHT and MOF 2.
Figure 3.15 XRD patterns of (a) MOF 2 and (b) FeTHT before and after 1 h CPE experiment
96
Figure 3.16 High resolution XPS of FeTHT before and after 1 h CPE: (a) Fe 2p; (b) S 2s
Figure 3.17 High resolution XPS of MOF 2 before and after 1 h CPE: (a) Co 2p; (b) Fe 2p; (c) S 2s
3.2.3 Discussion
Based on previous reports on molecular metal dithiolene complexes, FeTHT is expected to be
an inferior electrocatalyst for the HER compared to CoTHT. Computational studies focused on
the electronic structure of [Fe(bdt)2] and [Co(bdt)2] (bdt = benzene-1,2-dithiolate) molecular
97
complexes have suggested that there are significant differences in the composition of the frontier
orbitals. For [Co(bdt)2], the frontier orbitals possess mixed metal-ligand character (non-innocent),
yet for [Fe(bdt)2], these orbitals are predominantly metal-centered.
48
Therefore, the dithiolate
ligands in the [Fe(bdt)2] complex are considered innocent. Eisenberg and coworkers have
illustrated that [Fe(bdt)2] has a more negative reduction potential (-0.90 V vs. SCE) compared to
[Co(bdt)2] (-0.64 V vs. SCE), and exhibits lower activity when incorporated into a photocatalytic
HER system.
35,49
The diminished HER activity of iron dithiolene complexes is expected to be
retained upon incorporation into the extended THT-based framework. However, the extent of the
reduction in the HER activity for FeTHT in comparison to CoTHT is much greater than
anticipated. For the molecular [Fe(bdt)2] complex, although its activity is inferior compared to the
Co analogue, it is still capable of performing photocatalytic HER for up to 80 hours with high
turnover numbers.
35
Additionally, our previous study regarding an analogous polymer, FeBHT,
where iron dithiolene active units are incorporated into an extended framework using BHT as
linkers, has shown moderate activity towards the HER with a Tafel slope of 119 mV/dec and an
overpotential of 473 mV to reach 10 mA/cm
2
.
8
However, in this study, the electrocatalytic testing
of FeTHT shows a minimum activity and an inability to achieve the HER benchmarking metric
of 10 mA·cm
-2
within the investigated potential window, while CoTHT is reported to be among
the most active MOF-based HER catalysts.
4
Such large performance gap between CoTHT and
FeTHT is also surprising from a physical property point of view, as a recent report demonstrates
that the room-temperature electrical conductivity of the as-prepared FeTHT is comparable to that
of CoTHT,
36,37
suggesting that FeTHT should be sufficiently conductive to promote charge
transport in the pristine oxidation state.
98
Electrochemical characterizations suggest that under catalytic conditions, the FeTHT domains
undergo alternative Faradaic processes that alter its electrochemical behavior. First, the diminished
redox feature observed in the consecutive CV scans of FeTHT (Figure 3.8) suggests that the
material is dynamically changing under operating conditions and the electrons seem to be trapped
within the framework. Second, the EIS response of FeTHT reveals a much larger Rct after 1 hour
CPE experiment (Figure 3.9), indicating a more hindered HER kinetics on the post-CPE catalyst,
or in other words, the material becomes less active during electrolysis. The electrochemical
instability of FeTHT persists within the mixed-metal frameworks. For instance, the CV studies of
the mixed-metal MOFs in pH 10 electrolyte solutions highlight the limited cyclability of the Fe-
containing MOFs (Figure 3.14). Within the same timescale, the change in the redox feature is more
drastic for 1 relative to 2 and 3, indicating that when more Fe is present in the framework, the
material is more prone to have altered electrochemical behaviors. Furthermore, the compromised
FEs for Fe-containing MOFs demonstrate that electrons undergo alternative Faradaic processes
other than the HER. It is observed that the loss of FE correlates positively with the increase in Fe
content, indicating that the FeTHT domains are responsible for the loss of FE. Post-electrolysis
physical characterizations of the catalysts suggest that the altered electrochemical behavior could
be attributed to the changes in the local coordination geometry of the Fe center. The PXRD
measurements of MOF 2 and FeTHT after electrochemical treatment display peaks with
diminished intensity, indicating a deteriorated coherence within the 2D sheets (Figure 3.15).
Additionally, the XPS features of the Fe 2p and S 2s regions are very different between the pristine
and post-electrolysis catalysts, suggesting a change in the bonding environment of the Fe sites
(Figure 3.16, Figure 3.17).
99
In an attempt to lower the catalyst cost and enhance the activity on a per-site basis, a series of
mixed Co/Fe MOFs were tested for the hydrogen evolution reaction. However, the polarization
curves normalized on the cobalt loading (Figure 3.10) suggest that the presence of Fe does not
boost the activity of the Co sites. Furthermore, the CPE experiments (Figure 3.12) show that the
presence of Fe can in fact suppress the performance of the Co sites. For 2, where the metal
composition is roughly 50% Co and 50% Fe, the current response drastically decreases over the
course of the electrolysis, indicating that the high intrinsic activity of the Co sites are masked in
the presence of Fe. The calculated average TOF of the MOFs (Table 3.7) also illustrates that when
more Fe is present in the framework, the Co sites appear to behave less active for the HER.
To explain the adverse impact of Fe sites on the HER activity of the mixed-metal frameworks,
it is proposed that the electrochemical treatment of FeTHT yields a material with attenuated charge
transfer ability, rendering the originally active Co sites inaccessible by the electrons. This
conductivity-switching phenomenon has been observed by Boettcher and coworkers while
studying transition metal oxyhydroxides for electrocatalytic OER.
44,45,50
The electrical
conductivity of the oxyhydroxides is found to be higher under anodic potentials and lower under
cathodic potentials. Although similar behavior has not been reported for any THT-based materials,
it has been seen in some analogous BHT-based materials. For example, the NiBHT and AgBHT
frameworks are found to exhibit a lower electrical conductivity upon chemical reduction.
51,52
These prior studies, together with the electrochemical behavior of FeTHT observed in this work,
suggest the possibility of a declined electron transfer ability of FeTHT under electrocatalytic
conditions, which could be caused by the change in the coordination geometry of the Fe sites.
For example, it is possible that the diminished redox feature observed for FeTHT is due to the
low ability of the reduced FeTHT to undergo electron transfer, which essentially forms an
100
insulating layer at the electrode/catalyst interface that prevents further electronic communication.
Additionally, the large Tafel slope of FeTHT (210 mV/dec) indicates a large kinetic barrier for
the first one-electron one-proton reduction (Volmer step), which could arise from the insulating
nature of the reduced FeTHT. Upon incorporation of Co, which is known to be active under
cathodic potentials, the energy barrier for electron transfer is lowered, leading to a decrease in
Tafel slope (Table 3.4). Following the same line, the deactivation mechanism of the mixed-metal
MOFs can also be rationalized. During electrolysis, the FeTHT domains would be reduced and
deactivated gradually. As these FeTHT domains are distributed homogeneously throughout the
framework, as suggested by the EDX mapping, their deactivation can impede electron transfer
within the entire structure, rendering the active Co sites catalytically inert. This aligns well with
the changes in the Rct and Cdl values of MOF 2 observed after 1 hour of electrolysis (Table 3.6);
the almost 10 fold increase in Rct suggests a severely hindered charge transfer kinetics, whereas
the drastically decreased Cdl reveals a diminished number of the catalytically active sites. When
more Fe is present in the framework, the deactivation process is accelerated, which is observed in
the more pronounced deactivating behavior of 2 in comparison to 3. As for 1, due to the
significantly higher Fe concentration within the structure, the material deactivates rapidly and
exhibits electrocatalytic activity analogous to that of FeTHT.
3.3 Conclusions
In summary, taking advantage of the synthetic tunability of MOFs, a series of THT-based
MOFs with varying stoichiometries of cobalt and iron were synthesized and characterized. The
composition of the materials can be readily altered by tuning the ratios of the iron and cobalt
precursors, as revealed by ICP-OES and confirmed by XPS and EDX measurements. The materials
within the series possess similar crystalline structures regardless of the chemical composition, as
101
suggested by PXRD and FTIR measurements. The electrocatalytic HER activities of the materials
were characterized in pH 1.3 aqueous solutions. It is found that, unlike its Co analogue, FeTHT
exhibits minimal activity towards the HER, with a catalytic onset of -0.440 V versus RHE and a
Tafel slope of 210 mV/dec. Upon incorporation of more Co into the framework, the overall
electrocatalytic HER activity of the framework increases drastically, indicating Co being the
catalytically active center. However, further investigation suggests that Fe centers within the
framework are not merely inert spectator species serving to dilute the catalytically active Co
centers, but rather active inhibitors of efficient and steady H2 evolution. The long-term stability of
the material as well as the FE for H2 production drastically decline when more Fe is present in the
structure. Post-electrolysis characterizations of the catalysts suggest a diminished crystallinity and
a drastically altered chemical environment for the Fe sites. It is proposed that the poor activity of
FeTHT and the adverse effect of the Fe centers in the mixed-metal frameworks originate from the
ability of iron-containing materials to participate in alternative Faradaic processes, leading to a
material that is less electrochemically active for the HER. This hypothesis is supported by EIS
characterizations and CV studies where an irreversible redox peak is observed to diminish over
consecutive CV scans. Future experimental studies such as in-situ conductivity measurements and
X-ray absorption spectroscopy measurements, as well as density functional theory calculations are
necessary to fully elucidate the deactivation mechanism of the Fe and mixed-metal MOFs. Our
work suggests that, while incorporating tunable molecular units into extended frameworks serves
as an attractive strategy to build heterogeneous electrocatalysts, precautions should be taken when
carrying out this strategy, as the physical properties of the bulk material, such as the ability to
move charge, could be altered under catalytic conditions and adversely impact the overall activity
of the material. Furthermore, although diluting expensive, highly active metals with cheaper, more
102
abundant, and catalytically inert metals is a method to increase the per-site activity and decrease
the overall cost of the catalyst, our observation of the decrease in activity of the Co active sites
with increasing Fe content demonstrates that the identity of the diluting metal should be chosen
carefully.
3.4 Supplementary Experimental Information
3.4.1 General Experimental Methods
All manipulations of air and moisture sensitive materials were conducted under a nitrogen
atmosphere in a Vacuum Atmospheres Glovebox or on a dual manifold Schlenk line. The
glassware was oven-dried prior to use. Water was deionized with the Millipore Synergy system
(18.2 MW·cm resistivity). All the solvents used were degassed under vacuum and refilled with
nitrogen (10 ×). Triphenylene-2,3,6,7,10,11-hexathiol ligand (THT) was synthesized according to
literature procedures.
53
All other chemical reagents were purchased from commercial vendors and
used without further purification.
3.4.2 Synthesis of FeTHT
The synthesis of FeTHT was previously reported by our laboratory.
54
A 120 mL jar was
charged with a solution of FeCl2∙4H2O (40 mg, 0.20 mmol) in water (40 mL). Separately, a
suspension of THT (2.5 mg, 0.006 mmol) in N-methyl-2-pyrrolidone (NMP) (0.1 mL) was diluted
with ethyl acetate until the total volume of the suspension reached 5 mL, sealed, and briefly
sonicated to form a uniform suspension. Ethyl acetate (35 mL) was gently layered on top of the
aqueous solution to create a liquid-liquid interface; the suspension of THT was then gently added
to the ethyl acetate layer and the jar was sealed and allowed to stand still. A black film appeared
at the liquid-liquid interface over 5 days, which was then collected as powder, solvent exchanged
with water (3 × 20 mL), methanol (3 × 20 mL), and dried under vacuum.
103
3.4.3 Synthesis of Mixed Fe/Co THT
A series of aqueous solutions with different molar ratios of FeCl2∙4H2O and CoCl2∙6H2O were
prepared and added to a 120 mL jar. The total metal amount was controlled to be 0.20 mmol for
each jar. Separately, a suspension of THT (2.5 mg, 0.006 mmol) in N-methyl-2-pyrrolidone (NMP)
(0.1 mL) was diluted with ethyl acetate until the total volume of the suspension reached 5 mL,
sealed, and briefly sonicated to form a uniform suspension. Ethyl acetate (35 mL) was gently
layered on top of the aqueous solution to create a liquid-liquid interface; the suspension of THT
was then gently added to the ethyl acetate layer and the jar was sealed and allowed to stand still.
A black film appeared at the liquid-liquid interface over 5 days, which was then collected as
powder, solvent exchanged with water (3 × 20 mL), methanol (3 × 20 mL), and dried under
vacuum.
3.4.4 Deposition of MOFs for Electrochemistry Study
The THT-based MOFs were deposited as an ink mixture. To generate the ink mixture, 2 mg of
as-prepared MOF powder was mixed with 0.5 mg of carbon black (Vulcan XC-72R), 20 µL of
Nafion solution (0.5 wt%), 45 µL of water, and 135 µL of ethanol, followed by sonication for 30
mins to form a uniformly dispersed suspension. A desired amount of this suspension was then drop
casted onto a freshly polished glassy carbon electrode (GCE) or a rotating disk electrode (RDE,
glassy carbon insert) using a micro syringe, and dried in a nitrogen atmosphere at room temperature.
3.4.5 Electrochemical Methods
Electrochemistry experiments were carried out using a VersaSTAT 3 potentiostat in a three-
electrode configuration electrochemical cell under an inert atmosphere. A glassy carbon electrode
(GCE, 0.07065 cm
2
surface area, used for CV studies) or a rotating disk electrode (RDE, glassy
carbon insert, 0.196 cm
2
surface area, used for obtaining the steady-state polarization curves) was
104
used as the working electrode. GCE and RDE were polished with 0.05 µm Al2O3 polish powder
and sonicated in water prior to use. A graphite rod, purchased from Graphite Machining, Inc.
(Grade NAC-500 Purified, < 10 ppm ash level), was used as the counter electrode. The reference
electrode, placed in a separate compartment and connected by a Vycor tip, was based on an
aqueous Ag/AgCl/saturated 3.5 M KCl electrode. The reference electrode in aqueous media was
calibrated externally relative to ferrocenecarboxylic acid (Fc-COOH) at pH 7.0, with the Fe
3+/2+
couple at 0.28 V vs. Ag/AgCl. All potentials reported in this paper were converted to the reversible
hydrogen electrode (RHE) by adding a value of (0.205 + 0.059 × pH) V, or to the normal hydrogen
electrode (NHE) by adding a value of 0.205 V (for pH 10 CV studies). The aqueous solutions used
in the electrochemical experiments were prepared as follows. For the pH 1.3 solution, 0.534 mL
of 18.7 M H2SO4 was added to 200 mL 0.1 M NaClO4. For the pH 10.0 solution, NaHCO
3
(0.678
g) and Na
2
CO
3
(1.264 g) were dissolved in 200 mL 0.1 M NaClO4. The pH of the solutions was
measured with a benchtop Mettler Toledo pH meter. Prior to each electrochemical experiment, the
electrolyte solution was purged with nitrogen thoroughly.
Controlled potential electrolysis (CPE) measurements to determine Faradaic efficiency were
conducted in a sealed two-chambered H-cell where the first chamber held the working and
reference electrodes in 40 mL of pH 1.3 aqueous solution and the second chamber held the counter
electrode in 20 mL of pH 1.3 aqueous solution. The two chambers were both under N2 and
separated by a fine porosity glass frit. CPE experiments were performed with a glassy carbon plate
electrode (6 cm × 1 cm × 0.3 cm; Tokai Carbon USA) as the working electrode and a graphite rod
as the counter electrode. The reference electrode was a Ag/AgCl/saturated 3.5 M KCl (aq.)
electrode separated from the solution by a Vycor tip. Using a gas-tight syringe, 2 mL of gas was
withdrawn from the headspace of the H-cell and injected into a gas chromatography instrument
105
(Shimadzu GC-2010-Plus) equipped with a BID detector and a Restek ShinCarbon ST
Micropacked column. To determine the Faradaic efficiency, the theoretical H2 amount based on
total charge flowed is compared with the GC-detected H2 produced from controlled-potential
electrolysis.
Electrochemical impedance spectroscopy (EIS) measurements were carried out at different
overpotentials in the frequency range of 100 kHz – 0.1 Hz with 10 mV sinusoidal perturbations.
Experimental EIS data were analyzed and fitted with the ZSimpWin software.
The obtained polarization curves were corrected for iR loss according to the following equation:
Ecorr = Emea – iRs
Where Ecorr is the iR-corrected potential, Emea is the experimentally measured potential, and Rs
is the solution resistance extracted from the fitted EIS data.
3.4.6 Physical Characterization Methods
X-ray photoelectron spectroscopy (XPS) data were collected using a Kratos AXIS Ultra
instrument. The monochromatic X-ray source was the Al Kα 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
"+
Torr. The
XPS data were analyzed using the CasaXPS software.
High resolution synchrotron powder X-ray diffraction data was collected using the 11-BM
beamline mail-in program at the Advanced Photon Source (APS), Argonne National Laboratory,
with an average wavelength of 0.412750 Å. Discrete detectors covering an angular range from 0.5
to 30
o
2θ are scanned over a 34
o
2θ range, with data points collected every 0.001
o
2θ and scan
speed of 0.01
o
/s.
106
In-house powder X-ray diffraction (PXRD) studies were performed on a Rigaku Ultima IV X-
Ray diffractometer in reflectance parallel beam/parallel slit alignment geometry. The measurement
employed Cu Kα line focused radiation at 1760 W (40 kV, 44 mA) power and a Ge crystal detector
fitted with a 2 mm radiation entrance slit. Samples were mounted on zero-background sample
holders and were observed using a 0.08° 2θ step scan from 2.0 – 40.0° with an scan rate of 1° per
minute.
Inductively coupled plasma optical emission spectroscopy (ICP-OES) measurements were
performed using a Thermo Scientific iCAP 7000 ICP-OES. After electrochemical tests, the
composite was removed from the electrode surface, digested in concentrated nitric acid, and tested
by ICP-OES to determine the concentration of the Co and Fe in the catalyst. The post-CPE
electrolyte solution samples were prepared by adding certain amount of nitric acid to ensure the
consistency of the background.
Scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDX) were
performed on an FEI Nova NanoSEM 450 scanning electron microscope using an accelerating
voltage of 15 kV.
3.4.7 Computational Methods for FTIR Spectrum Simulation
The electronic structure calculations were performed using Gaussian 16, Revision C.01
program.
9
All geometry optimizations and frequency calculations were performed using the
B3LYP functional
10
and 6-311G(d,p) basis set.
11
All geometries were characterized by frequency
analysis calculations to be local minima (without any imaginary frequency). As B3LYP tends to
overestimate frequencies, a frequency scaling factor of 0.967 was used.
12
For the IR spectra, the
vibrational frequencies were broaden using a IR peak half-width at half height of 20 cm
-1
.
107
Due to the large unit cell associated with CoTHT, the calculation of IR spectrum using periodic
DFT is prohibitively computationally expensive. Therefore, we created a molecular model of
CoTHT with a cobalt atom coordinated by two THT ligands. This geometry was cut from a
periodic model of CoTHT that was studied in our previous work.
3
The sulfur atoms were capped
with hydrogen to saturate the sulfur bonds and mimic the periodic system. While the periodic
system shows a mixed oxide state of cobalt, for the molecular model, we used a doublet spin state
as this gave the lowest energy. The geometry of the model is provided below.
Figure 3.18 The geometry of the cobalt dithiolene molecular model
3.5 References
(1) Zhou, H. C.; Long, J. R.; Yaghi, O. M. Introduction to Metal-Organic Frameworks. Chem.
Rev. 2012, 112 (2), 673–674.
(2) Downes, C. A.; Marinescu, S. C. Electrocatalytic Metal–Organic Frameworks for Energy
Applications. ChemSusChem 2017, 10 (22), 4374–4392.
(3) Drake, T.; Ji, P.; Lin, W. Site Isolation in Metal-Organic Frameworks Enables Novel
Transition Metal Catalysis. Acc. Chem. Res. 2018, 51 (9), 2129–2138.
(4) Chen, K.; Downes, C. A.; Schneider, E.; Goodpaster, J. D.; Marinescu, S. C. Improving
and Understanding the Hydrogen Evolving Activity of a Cobalt Dithiolene Metal–Organic
Framework. ACS Appl. Mater. Interfaces 2021, 13 (14), 16384–16395.
(5) Clough, A. J.; Yoo, J. W.; Mecklenburg, M. H.; Marinescu, S. C. Two-Dimensional Metal–
Organic Surfaces for Efficient Hydrogen Evolution from Water. J. Am. Chem. Soc. 2015, 137 (1),
118–121.
(6) Sun, X.; Wu, K.-H.; Sakamoto, R.; Kusamoto, T.; Maeda, H.; Ni, X.; Jiang, W.; Liu, F.;
Sasaki, S.; Masunaga, H.; et al. Bis(Aminothiolato)Nickel Nanosheet as a Redox Switch for
108
Conductivity and an Electrocatalyst for the Hydrogen Evolution Reaction. Chem. Sci. 2017, 8 (12),
8078–8085.
(7) Dong, R.; Zheng, Z.; Tranca, D. C.; Zhang, J.; Chandrasekhar, N.; Liu, S.; Zhuang, X.;
Seifert, G.; Feng, X. Immobilizing Molecular Metal Dithiolene-Diamine Complexes on 2D Metal-
Organic Frameworks for Electrocatalytic H2 Production. Chem. Eur. J. 2017, 23 (10), 2255–2260.
(8) Downes, C. A.; Clough, A. J.; Chen, K.; Yoo, J. W.; Marinescu, S. C. Evaluation of the H
2 Evolving Activity of Benzenehexathiolate Coordination Frameworks and the Effect of Film
Thickness on H2 Production. ACS Appl. Mater. Interfaces 2018, 10 (2), 1719–1727.
(9) Dong, R.; Pfeffermann, M.; Liang, H.; Zheng, Z.; Zhu, X.; Zhang, J.; Feng, X. Large-Area,
Free-Standing, Two-Dimensional Supramolecular Polymer Single-Layer Sheets for Highly
Efficient Electrocatalytic Hydrogen Evolution. Angew. Chemie Int. Ed. 2015, 54 (41), 12058–
12063.
(10) Qin, J.-S.; Du, D.-Y.; Guan, W.; Bo, X.-J.; Li, Y.-F.; Guo, L.-P.; Su, Z.-M.; Wang, Y.-Y.;
Lan, Y.-Q.; Zhou, H.-C. Ultrastable Polymolybdate-Based Metal–Organic Frameworks as Highly
Active Electrocatalysts for Hydrogen Generation from Water. J. Am. Chem. Soc. 2015, 137 (22),
7169–7177.
(11) Hod, I.; Deria, P.; Bury, W.; Mondloch, J. E.; Kung, C.-W.; So, M.; Sampson, M. D.; Peters,
A. W.; Kubiak, C. P.; Farha, O. K.; et al. A Porous Proton-Relaying Metal-Organic Framework
Material That Accelerates Electrochemical Hydrogen Evolution. Nat. Commun. 2015, 6 (1), 8304.
(12) Wu, Y.-P.; Zhou, W.; Zhao, J.; Dong, W.-W.; Lan, Y.-Q.; Li, D.-S.; Sun, C.; Bu, X.
Surfactant-Assisted Phase-Selective Synthesis of New Cobalt MOFs and Their Efficient
Electrocatalytic Hydrogen Evolution Reaction. Angew. Chem., Int. Ed. 2017, 56 (42), 13001–
13005.
(13) Micheroni, D.; Lan, G.; Lin, W. Efficient Electrocatalytic Proton Reduction with Carbon
Nanotube-Supported Metal–Organic Frameworks. J. Am. Chem. Soc. 2018, 140 (46), 15591–
15595.
(14) Huang, X.; Yao, H.; Cui, Y.; Hao, W.; Zhu, J.; Xu, W.; Zhu, D. Conductive Copper
Benzenehexathiol Coordination Polymer as a Hydrogen Evolution Catalyst. ACS Appl. Mater.
Interfaces 2017, 9 (46), 40752–40759.
(15) Gong, Y.; Shi, H.-F.; Hao, Z.; Sun, J.-L.; Lin, J.-H. Two Novel Co(II) Coordination
Polymers Based on 1,4-Bis(3-Pyridylaminomethyl)Benzene as Electrocatalysts for Oxygen
Evolution from Water. Dalt. Trans. 2013, 42 (34), 12252.
(16) Kung, C.-W.; Mondloch, J. E.; Wang, T. C.; Bury, W.; Hoffeditz, W.; Klahr, B. M.; Klet,
R. C.; Pellin, M. J.; Farha, O. K.; Hupp, J. T. Metal–Organic Framework Thin Films as Platforms
for Atomic Layer Deposition of Cobalt Ions To Enable Electrocatalytic Water Oxidation. ACS
Appl. Mater. Interfaces 2015, 7 (51), 28223–28230.
(17) Lu, X. F.; Liao, P. Q.; Wang, J. W.; Wu, J. X.; Chen, X. W.; He, C. T.; Zhang, J. P.; Li, G.
109
R.; Chen, X. M. An Alkaline-Stable, Metal Hydroxide Mimicking Metal-Organic Framework for
Efficient Electrocatalytic Oxygen Evolution. J. Am. Chem. Soc. 2016, 138 (27), 8336–8339.
(18) Johnson, B. A.; Bhunia, A.; Ott, S. Electrocatalytic Water Oxidation by a Molecular
Catalyst Incorporated into a Metal–Organic Framework Thin Film. Dalt. Trans. 2017, 46 (5),
1382–1388.
(19) Lin, S.; Pineda-Galvan, Y.; Maza, W. A.; Epley, C. C.; Zhu, J.; Kessinger, M. C.; Pushkar,
Y.; Morris, A. J. Electrochemical Water Oxidation by a Catalyst-Modified Metal-Organic
Framework Thin Film. ChemSusChem 2017, 10 (3), 514–522.
(20) Shen, J.-Q.; Liao, P.-Q.; Zhou, D.-D.; He, C.-T.; Wu, J.-X.; Zhang, W.-X.; Zhang, J.-P.;
Chen, X.-M. Modular and Stepwise Synthesis of a Hybrid Metal–Organic Framework for Efficient
Electrocatalytic Oxygen Evolution. J. Am. Chem. Soc. 2017, 139 (5), 1778–1781.
(21) Miner, E. M.; Fukushima, T.; Sheberla, D.; Sun, L.; Surendranath, Y.; Dincă, M.
Electrochemical Oxygen Reduction Catalysed by Ni3(Hexaiminotriphenylene)2. Nat. Commun.
2016, 7 (1), 10942.
(22) Miner, E. M.; Gul, S.; Ricke, N. D.; Pastor, E.; Yano, J.; Yachandra, V. K.; Van Voorhis,
T.; Dincă, M. Mechanistic Evidence for Ligand-Centered Electrocatalytic Oxygen Reduction with
the Conductive MOF Ni3(Hexaiminotriphenylene)2. ACS Catal. 2017, 7 (11), 7726–7731.
(23) Miner, E. M.; Wang, L.; Dincă, M. Modular O2 Electroreduction Activity in Triphenylene-
Based Metal–Organic Frameworks. Chem. Sci. 2018, 9 (29), 6286–6291.
(24) Usov, P. M.; Huffman, B.; Epley, C. C.; Kessinger, M. C.; Zhu, J.; Maza, W. A.; Morris,
A. J. Study of Electrocatalytic Properties of Metal–Organic Framework PCN-223 for the Oxygen
Reduction Reaction. ACS Appl. Mater. Interfaces 2017, 9 (39), 33539–33543.
(25) Lions, M.; Tommasino, J.-B.; Chattot, R.; Abeykoon, B.; Guillou, N.; Devic, T.;
Demessence, A.; Cardenas, L.; Maillard, F.; Fateeva, A. Insights into the Mechanism of
Electrocatalysis of the Oxygen Reduction Reaction by a Porphyrinic Metal Organic Framework.
Chem. Commun. 2017, 53 (48), 6496–6499.
(26) Liu, X. H.; Hu, W. L.; Jiang, W. J.; Yang, Y. W.; Niu, S.; Sun, B.; Wu, J.; Hu, J. S. Well-
Defined Metal-O6 in Metal-Catecholates as a Novel Active Site for Oxygen Electroreduction. ACS
Appl. Mater. Interfaces 2017, 9 (34), 28473–28477.
(27) Kornienko, N.; Zhao, Y.; Kley, C. S.; Zhu, C.; Kim, D.; Lin, S.; Chang, C. J.; Yaghi, O.
M.; Yang, P. Metal–Organic Frameworks for Electrocatalytic Reduction of Carbon Dioxide. J. Am.
Chem. Soc. 2015, 137 (44), 14129–14135.
(28) Hod, I.; Sampson, M. D.; Deria, P.; Kubiak, C. P.; Farha, O. K.; Hupp, J. T. Fe-Porphyrin-
Based Metal–Organic Framework Films as High-Surface Concentration, Heterogeneous Catalysts
for Electrochemical Reduction of CO2. ACS Catal. 2015, 5 (11), 6302–6309.
(29) Lin, S.; Diercks, C. S.; Zhang, Y.-B.; Kornienko, N.; Nichols, E. M.; Zhao, Y.; Paris, A.
110
R.; Kim, D.; Yang, P.; Yaghi, O. M.; et al. Covalent Organic Frameworks Comprising Cobalt
Porphyrins for Catalytic CO2 Reduction in Water. Science 2015, 349 (6253), 1208.
(30) Johnson, E. M.; Haiges, R.; Marinescu, S. C. Covalent-Organic Frameworks Composed of
Rhenium Bipyridine and Metal Porphyrins: Designing Heterobimetallic Frameworks with Two
Distinct Metal Sites. ACS Appl. Mater. Interfaces 2018, 10 (44), 37919–37927.
(31) Meng, Z.; Luo, J.; Li, W.; Mirica, K. A. Hierarchical Tuning of the Performance of
Electrochemical Carbon Dioxide Reduction Using Conductive Two-Dimensional
Metallophthalocyanine Based Metal-Organic Frameworks. J. Am. Chem. Soc. 2020, 142 (52),
21656–21669.
(32) Diercks, C. S.; Lin, S.; Kornienko, N.; Kapustin, E. A.; Nichols, E. M.; Zhu, C.; Zhao, Y.;
Chang, C. J.; Yaghi, O. M. Reticular Electronic Tuning of Porphyrin Active Sites in Covalent
Organic Frameworks for Electrocatalytic Carbon Dioxide Reduction. J. Am. Chem. Soc. 2018, 140
(3), 1116–1122.
(33) Matheu, R.; Gutierrez-Puebla, E.; Monge, M. Á.; Diercks, C. S.; Kang, J.; Prévot, M. S.;
Pei, X.; Hanikel, N.; Zhang, B.; Yang, P.; et al. Three-Dimensional Phthalocyanine Metal-
Catecholates for High Electrochemical Carbon Dioxide Reduction. J. Am. Chem. Soc. 2019, 141
(43), 17081–17085.
(34) Kusamoto, T.; Nishihara, H. Zero-, One- and Two-Dimensional Bis(Dithiolato)Metal
Complexes with Unique Physical and Chemical Properties. Coord. Chem. Rev. 2019, 380, 419–
439.
(35) Lv, H.; Ruberu, T. P. A.; Fleischauer, V. E.; Brennessel, W. W.; Neidig, M. L.; Eisenberg,
R. Catalytic Light-Driven Generation of Hydrogen from Water by Iron Dithiolene Complexes. J.
Am. Chem. Soc. 2016, 138 (36), 11654–11663.
(36) Clough, A. J.; Orchanian, N. M.; Skelton, J. M.; Neer, A. J.; Howard, S. A.; Downes, C.
A.; Piper, L. F. J.; Walsh, A.; Melot, B. C.; Marinescu, S. C. Room Temperature Metallic
Conductivity in a Metal–Organic Framework Induced by Oxidation. J. Am. Chem. Soc. 2019, 141
(41), 16323-16330.
(37) Clough, A. J.; Skelton, J. M.; Downes, C. A.; de la Rosa, A. A.; Yoo, J. W.; Walsh, A.;
Melot, B. C.; Marinescu, S. C. Metallic Conductivity in a Two-Dimensional Cobalt Dithiolene
Metal–Organic Framework. J. Am. Chem. Soc. 2017, 139 (31), 10863–10867.
(38) Dong, R.; Han, P.; Arora, H.; Ballabio, M.; Karakus, M.; Zhang, Z.; Shekhar, C.; Adler,
P.; Petkov, P. St.; Erbe, A.; et al. High-Mobility Band-like Charge Transport in a Semiconducting
Two-Dimensional Metal–Organic Framework. Nat. Mater. 2018, 17 (11), 1027–1032.
(39) Deng, H.; Doonan, C. J.; Furukawa, H.; Ferreira, R. B.; Towne, J.; Knobler, C. B.; Wang,
B.; Yaghi, O. M. Multiple Functional Groups of Varying Ratios in Metal-Organic Frameworks.
Science 2010, 327, 846–850.
(40) Wang, L. J.; Deng, H.; Furukawa, H.; Gándara, F.; Cordova, K. E.; Peri, D.; Yaghi, O. M.
111
Synthesis and Characterization of Metal-Organic Framework-74 Containing 2, 4, 6, 8, and 10
Different Metals. Inorg. Chem. 2014, 53 (12), 5881–5883.
(41) Zhang, B.; Zheng, Y.; Ma, T.; Yang, C.; Peng, Y.; Zhou, Z.; Zhou, M.; Li, S.; Wang, Y.;
Cheng, C. Designing MOF Nanoarchitectures for Electrochemical Water Splitting. Adv. Mater.
2021, 33 (17), 2006042.
(42) Zhao, Q.; Lin, X.; Zhou, J.; Zhao, C.; Zheng, D.; Song, S.; Jing, C.; Zhang, L.; Wang, J. A
Tunable Amorphous Heteronuclear Iron and Cobalt Imidazolate Framework Analogue for
Efficient Oxygen Evolution Reactions. Eur. J. Inorg. Chem. 2021, 2021 (8), 702–707.
(43) Dang, Y.; Han, P.; Li, Y.; Zhang, Y.; Zhou, Y. Low-Crystalline Mixed Fe-Co-MOFs for
Efficient Oxygen Evolution Electrocatalysis. J. Mater. Sci. 2020, 55 (28), 13951–13963.
(44) Burke, M. S.; Kast, M. G.; Trotochaud, L.; Smith, A. M.; Boettcher, S. W. Cobalt-Iron
(Oxy)Hydroxide Oxygen Evolution Electrocatalysts: The Role of Structure and Composition on
Activity, Stability, and Mechanism. J. Am. Chem. Soc. 2015, 137 (10), 3638–3648.
(45) Batchellor, A. S.; Boettcher, S. W. Pulse-Electrodeposited Ni-Fe (Oxy)Hydroxide Oxygen
Evolution Electrocatalysts with High Geometric and Intrinsic Activities at Large Mass Loadings.
ACS Catal. 2015, 5 (11), 6680–6689.
(46) Shinagawa, T.; Garcia-Esparza, A. T.; Takanabe, K. Insight on Tafel Slopes from a
Microkinetic Analysis of Aqueous Electrocatalysis for Energy Conversion. Sci. Rep. 2015, 5 (1),
13801.
(47) Downes, C. A.; Marinescu, S. C. Understanding Variability in the Hydrogen Evolution
Activity of a Cobalt Anthracenetetrathiolate Coordination Polymer. ACS Catal. 2017, 7 (12),
8605–8612.
(48) Ray, K.; Begum, A.; Weyhermüller, T.; Piligkos, S.; van Slageren, J.; Neese, F.; Wieghardt,
K. The Electronic Structure of the Isoelectronic, Square-Planar Complexes [Fe
II
(L)2]
2-
and
[Co
III
(LBu)2]
-
(L
2-
and (LBu)
2-
= Benzene-1,2-Dithiolates): An Experimental and Density Functional
Theoretical Study. J. Am. Chem. Soc. 2005, 127 (12), 4403–4415.
(49) Das, A.; Han, Z.; Haghighi, M. G.; Eisenberg, R. Photogeneration of Hydrogen from Water
Using CdSe Nanocrystals Demonstrating the Importance of Surface Exchange. Proc. Natl. Acad.
Sci. 2013, 110 (42), 16716–16723.
(50) Stevens, M. B.; Enman, L. J.; Batchellor, A. S.; Cosby, M. R.; Vise, A. E.; Trang, C. D.
M.; Boettcher, S. W. Measurement Techniques for the Study of Thin Film Heterogeneous Water
Oxidation Electrocatalysts. Chem. Mater. 2017, 29 (1), 120–140.
(51) Kambe, T.; Sakamoto, R.; Hoshiko, K.; Takada, K.; Miyachi, M.; Ryu, J. H.; Sasaki, S.;
Kim, J.; Nakazato, K.; Takata, M.; et al. π-Conjugated Nickel Bis(Dithiolene) Complex Nanosheet.
J. Am. Chem. Soc. 2013, 135 (7), 2462–2465.
(52) Liu, L.; Tu, Z.; Xu, W.; Chen, J.; Zou, Y.; Yi, Y.; Wu, X.; Li, H.; Liang, Y.; Huang, X.; et
112
al. Highly Conducting Neutral Coordination Polymer with Infinite Two-Dimensional Silver–
Sulfur Networks. J. Am. Chem. Soc. 2018, 140 (45), 15153–15156.
(53) Sakamoto, R.; Kambe, T.; Tsukada, S.; Takada, K.; Hoshiko, K.; Kitagawa, Y.; Okumura,
M.; Nishihara, H. Π-Conjugated Trinuclear Group-9 Metalladithiolenes With a Triphenylene
Backbone. Inorg. Chem. 2013, 52 (13), 7411–7416.
(54) Clough, A. J.; Yoo, J. W.; Mecklenburg, M. H.; Marinescu, S. C. Two-Dimensional Metal–
Organic Surfaces for Efficient Hydrogen Evolution from Water. J. Am. Chem. Soc. 2015, 137 (1),
118–121.
113
Chapter 4
Cu[Ni(2,3-pyrazinedithiolate)2] Metal-Organic Framework for Electrocatalytic
Hydrogen Evolution
A portion of this chapter has appeared in print:
Chen, K.; Ray, D.; Ziebel, M. E.; Gaggioli, C. A.; Gagliardi, L.; Marinescu, S. C. “Cu[Ni(2,3-
pyrazinedithiolate)2] Metal-Organic Framework for Electrocatalytic Hydrogen Evolution ”. ACS
Appl. Mater. Interfaces 2021, 13 (29), 34419.
Reprinted with permission from ACS Appl. Mater. Interfaces 2021, 13 (29), 34419. Copyright 2021 American
Chemical Society.
114
4.1 Introduction
In response to the threat of climate change, it is paramount to displace the current fossil-fuel-
dominated energy economy with a more sustainable alternative. Recent years have witnessed
increased shares of renewable energy in the power sector of the overall energy structure.
1
For
example, renewables were estimated to contribute to 26% of the global electricity generation in
2018.
1
However, the implementation of renewable energy in other major energy sectors, such as
transportation, is still very limited, partially due to the intermittent nature of the renewable sources
and the lack of renewable energy storage technologies.
1
To address this problem, electrocatalysis
provides a feasible solution, where abundant small molecules are converted into value-added
products with the input of renewable electricity, allowing for the indirect conversion/storage of
renewable energy into chemical bonds, which can then be used as fuels and chemical feedstocks
in transportation and industrial processes.
2
This strategy can greatly increase the flexibility of
renewable energy, facilitate the decarbonization of the overall energy structure, and the
distribution of renewable sources in a broader geographic scope. Among all the electrocatalytic
processes, the generation of H2 is of particular interest, driven by the booming of the H2 fuel-cell
industry. However, one of the most pressing challenges to implementing a commercially viable
electrocatalytic system is to develop electrocatalysts that comprise of only abundant elements and
can function effectively under mild conditions.
Metal-organic frameworks (MOFs) have recently attracted great attention in electrocatalysis
due to their unique properties, such as large surface areas, high stabilities, ordered crystalline
structures, and well-defined chemical motifs. The high synthetic tunability of MOFs offers
immense opportunities for the design and optimization of efficient electrocatalysts. Additionally,
MOFs provide a great platform for the heterogenization of homogeneous catalysts, which can be
115
achieved by employing well-understood molecular catalytic units as the building blocks to
construct an extended framework structure. This allows for the design of heterogeneous materials
with well-defined active sites that are viable for real-world applications.
3
Currently, MOF-based
electrocatalysts have demonstrated capability in various reactions including the hydrogen
evolution reaction (HER),
4–11
oxygen evolution reaction (OER),
12–20
oxygen reduction reaction
(ORR),
21–26
and CO2 reduction reaction (CO2RR).
27–34
For the HER, dithiolene-based MOFs are
among the most active and well understood.
4–6,11
This family of MOFs is typically constructed by
linking single-metal nodes with trinucleating dithiolene ligands to form a fully-conjugated two-
dimensional (2D) sheet. For example, MOFs composed of 2,3,6,7,10,11-triphenylenehexathiolate
(THT) linkers and cobalt or nickel metal centers have demonstrated excellent HER activity in
fully aqueous media with a near-unity faradaic efficiency (FE).
4–6,11
Herein, a three-dimensional (3D) dithiolene-based MOF, Cu[Ni(pdt)2], where pdt = pyrazine-
2,3-dithiolate, is studied as an electrocatalyst for the HER for the first time. Cu[Ni(pdt)2] is
constructed by nickel bis(pyrazine-2,3-dithiolate) complexes that are linked by 4-coordinated
square planar Cu centers through the N atoms of the pyrazine moiety (Figure 4.1). Unlike
previously studied dithiolene-MOFs that typically possess a 2D geometry, Cu[Ni(pdt)2] forms a
3D structure with one-dimensional (1D) square channels along the c direction.
35
An interesting
conductivity-switch behavior was previously reported for Cu[Ni(pdt)2], where a 10
4
-fold increase
in conductivity was observed upon oxidation with I2.
35
Although Cu[Ni(pdt)2] has been studied for
applications in gas separation,
36,37
its electrocatalytic applications remain underdeveloped. In
contrast, the electrocatalytic HER performance of the molecular analogues of this material has
been extensively explored.
38–41
In one of the prior reports, a [Ni(dcpdt)2]
2-
(dcpdt = 5,6-
dicyanopyrazine-2,3-dithiolate) complex was shown to perform HER with low overpotentials
116
(330–400 mV) and high faradaic efficiencies (92–100%) in pH 4–6 aqueous solutions.
38
The
excellent HER activity of these molecular complexes prompted us to explore the competency of
Cu[Ni(pdt)2], where the Ni-dithiolene active units are incorporated into a 3D framework, allowing
for the transitioning from homogeneous to heterogeneous catalysis. Additionally, the 3D
architecture could be beneficial for proton diffusion toward the active sites, an important factor for
preferable HER kinetics as suggested by previous studies with other dithiolene-based
electrocatalysts.
11,42
Figure 4.1 Structure of Cu[Ni(pdt)2] (1). Color coding for atoms: green (Ni), cyan (Cu), yellow (S), dark blue (N), and
gray (C). H atoms are omitted for clarity.
4.2 Results and Discussion
4.2.1 Electrochemical Studies
To investigate the electrocatalytic hydrogen-evolving activity of Cu[Ni(pdt)2] (1), detailed
electrochemical measurements were performed in a three-electrode configuration. To improve the
catalyst adhesion to the electrode substrate, 1 was mixed with Nafion, sonicated in ethanol and
water to form a homogeneous suspension, and deposited as a composite. The powder X-ray
diffraction (PXRD) of the composite displays peaks similar to that of the as-prepared material
(Figure 4.2), suggesting that the crystallinity of the material is maintained after sonication in the
presence of Nafion, ethanol, and water. The composite was drop-casted onto a freshly polished
117
glassy carbon electrode (GCE) using a microsyringe and dried under ambient conditions prior to
use. A previous report of 1 illustrated its excellent water stability over a wide pH range (pH 1–
12).
36
Therefore, the electrocatalytic activity of 1 was first measured in pH 1.3 H2SO4 aqueous
solutions. The PXRD pattern of 1 was acquired after soaking the material in a solution of pH 1.3
for 18 h, to further confirm the stability of the material in the electrolyte solution (Figure 4.2).
Figure 4.2 X-ray diffraction patterns of 1: (1) as-prepared (red); (2) after 20 minutes sonication in the presence of
Nafion, ethanol, and water (blue); (3) after soaking in an electrolyte solution of pH 1.3 for 18 h (purple). Inset is an
image of the dark purple-colored sonicated catalyst mixture.
The polarization curve of 1 obtained from linear sweep voltammetry (LSV) is presented in
Figure 4.3. 1 exhibits a catalytic onset at -0.43 V vs RHE and an overpotential (𝜂) of 0.53 V to
reach a current density of 10 mA/cm
2
. The Tafel plot derived from the polarization curve revealed
a Tafel slope of 69.0 mV/dec and an exchange current density (j0) of 10
-6.7
mA/cm
2
(Figure 4.4).
The charge-transfer resistance (Rct) of 1 is 77 Ω at 𝜂 = 0.52 V, obtained by fitting the
electrochemical impedance spectroscopy (EIS) response with a two-time constant serial (2TS)
model (Figure 4.4). Characterized by the aforementioned metrics, the HER activity of 1 is
comparable to that of other dithiolene-based metal-organic frameworks or coordination polymers.
It should be noted that the measurements conducted herein did not employ any sophisticated thin-
film-growth techniques or utilize any conductive matrix, such as carbon black or carbon nanotubes,
which are common strategies used in MOF-based electrocatalysis to facilitate the charge transport
118
through active sites and enhance the overall catalyst performance.
6,11,43
The absence of such
electrode modification techniques in this work highlights the sufficient intrinsic electrical
conductivity of 1 to support efficient catalytic reactions at the active sites.
Figure 4.3 Polarization curve of 1 obtained from linear sweep voltammetry (LSV) in a N2-saturated pH 1.3 aqueous
solution; scan rate: 5 mV/s, RDE rotation rate: 1600 rpm.
Figure 4.4 (a) Tafel plot of 1 derived from the polarization curve shown above. (b) Nyquist plot of 1 (markers) and
the respective fit (line) at -0.52 V vs. RHE; Inset shows the 2TS equivalent circuit used to fit the EIS response.
It is noteworthy that 1 exhibits a precatalytic redox feature at around -0.26 V vs RHE (Figure
4.5), a feature that has seldom been observed for other heterogenized dithiolene-based catalysts.
However, a pre-catalytic wave has been observed for the molecular dithiolene complexes, where
it was assigned to either hydrogenation of the ligand or decomposition of the molecular
complex.
38,44
Cyclic voltammetry (CV) studies of 1 reveal a complex redox behavior at more
positive potential regions, highlighted by a dominant return oxidation peak at 0.34 V vs RHE
119
(Figure 4.5). Additionally, the crossing of the CV trace in the catalytic potential window, where
the return sweep shows a larger current than the initial cathodic sweep (Figure 4.5), indicates that
the activity of the material increases as a function of time.
45
These observations are suggestive that
the material is undergoing certain irreversible dynamic transitions under the applied potentials that
could result in a more active catalyst.
Figure 4.5 (a) Polarization curve of 1 magnified at the catalytic onset region, scan rate: 5 mV/s. (b) CV scan of 1 with
the inset focused on the higher potential region, scan rate: 50 mV/s.
To further analyze the stability of 1 and its faradaic efficiency (FE) for hydrogen production,
controlled potential electrolysis (CPE) was performed in a pH 1.3 aqueous solution over the course
of 1 h (Figure 4.6). The magnitude of the current drastically increased (from 5 mA to 14 mA)
during the first 20 min of the electrolysis and then plateaued for the rest of the experiment. This
enhancement in current was accompanied by a decrease in Rct (from 46 Ω to 13 Ω) and an increase
in the double-layer capacitance (Cdl, from 44 μF to 55 μF), as revealed by the EIS spectra measured
at -0.49 V vs. RHE (Figure 4.6). The decrease in Rct indicates enhanced charge-transfer kinetics
and the increase in Cdl suggests a larger electrocatalytically active surface area, both of which are
beneficial for promoting a more efficient HER. This activation behavior of 1 was reproducible
when identical experiments were repeated using different batches of the material. The average FE
for H2 production after 1 h of electrolysis was determined to be 77.0 ± 5.7%. The moderate FE
120
along with the changes in current, Rct, and Cdl suggest that alternative faradaic processes are
occurring concurrently with the HER, which lead to the dynamic changes of 1 and result in a more
active form of the catalyst.
Figure 4.6 (a) 1 h CPE experiment of 1 at -0.59 V vs. RHE; (b) Nyquist plots (markers) and respective fits (lines)
before and after electrolysis; EIS potential: -0.49 V vs. RHE.
To further probe the electrochemical activation process of 1, longer-duration CPE experiments
were performed. During one of the experiments, the headspace of the electrochemical cell was
sampled three times throughout the 2 h electrolysis (Figure 4.7). It was found that after 15 min of
electrolysis, the FE for the HER was only 42.7%, which increased to 82.9% during the remainder
of the first hour and to 93.4% during the second hour. This suggests that the main loss of electrons
occurred at the beginning of the electrolysis, where a drastic current increase was observed. In
other words, the reconstruction of 1 occurs rapidly under the catalytic conditions, producing a
more active and stable material with higher FE for H2 production. In another experiment, 1 was
subjected to electrolysis for 4 h, and the LSV was acquired after 2 and 4 h. As shown in Figure
4.8, the activities of the catalyst were very similar after 2 and 4 h of electrolysis, with an onset of
catalysis at -0.23 V vs RHE, an average 𝜂
'567/06
- of 0.35 V, and an average Tafel slope of 75
mV/dec (Table 4.1). These results demonstrate that the activated catalyst has a much higher HER
activity compared to the pristine catalyst, evidenced by a 200 mV reduction in catalytic onset and
121
𝜂
'567/06
-. The Tafel slope of the activated catalyst appears to be slightly higher than that of the
pristine catalyst (73 mV/dec vs. 69 mV/dec). This is likely due to the more favorable
electrochemical kinetics of the activation processes compared to the HER. To confirm the
heterogeneous nature of the activated catalyst, a wash test was performed, where the working
electrode was replaced by a bare glassy carbon electrode after CPE, without replenishing the
electrolyte solution. As shown in Figure 4.9, the wash test did not display any discernible activity,
suggesting that the species in solution did not contribute to the high activity observed during the
electrolysis.
Figure 4.7 Controlled potential electrolysis performed at -0.50 V vs RHE in a N2-saturated pH 1.3 solution.
Figure 4.8 (a) 4 h electrolysis of 1 at -0.59 V vs RHE. (b) LSVs of 1 acquired before, after 2 h, and after 4 h CPE, in
comparison with the Pt/C catalyst; scan rate: 5 mV/s.
122
Table 4.1 HER activity of 1 before and after CPE (-0.59 V vs. RHE)
Trial Onset (V) vs. RHE η @ 10 mA/cm
2
(V) Tafel slope (mV/dec)
Before -0.43 0.53 69.0
After 2 hour -0.23 0.36 76.1
After 4 hour -0.23 0.33 72.7
Figure 4.9 CPE and subsequent wash test conducted at -0.59 V vs RHE in a N2-saturated pH 1.3 aqueous solution.
To understand the role of protons in the activation process, the electrochemical behavior of 1
was evaluated in electrolyte solutions with higher pH values. The polarization curves of 1 in pH
1.3, pH 4.8 and pH 7.3 electrolytes were obtained, and the corresponding catalytic onset potential
was plotted against the pH value (Figure 4.10). The slope of this plot is 27.6 mV/dec. According
to the Nernst equation, if the number of protons and electrons are equal in an electrochemical
reaction, the slope of the potential–pH plot should be 59 mV/dec, which is expected for the HER.
However, the obtained slope for 1 is 27.6 mV/dec, nearly half of the expected value, indicating
that the ratio of the protons and electrons involved in the electrochemical reaction at the onset
potential is around 1:2. In other words, more electrons are consumed than that needed for the HER.
This result is consistent with the moderate FE observed for 1, suggesting the consumption of
electrons by alternative faradaic processes that occur concurrently with the HER. The CPE
experiment was also carried out in a pH 4.6 solution under similar conditions (Figure 4.11). The
activation process still persisted but is much slower, as an increase in current was observed during
123
the entire course of the electrolysis. The FE for hydrogen production was 86.3 ± 0.6% as compared
to that of 77.0 ± 5.7% in pH 1.3, indicating that the HER is more favored under higher pH
conditions and the side process that leads to the loss of the FE is likely proton-dependent.
Figure 4.10 (a) Polarization curves of 1 obtained in various-pH electrolytes, scan rate: 5 mV/s. (b) Plot of onset
potential vs. corresponding pH values.
Figure 4.11 1 h CPE performed at -0.59 V vs RHE in N2-saturated pH 4.6 aqueous solution.
4.2.2 Physical Characterizations
The electrochemical behavior of 1 suggests that the material is dynamically changing at
negative potentials in acidic conditions. To investigate this transition process, a series of physical
characterizations were performed after electrolysis. First, the material was analyzed under
scanning electron microscopy (SEM) to reveal the morphological changes during catalysis. As
shown in Figure 4.12, the as-prepared MOF displays a sphere-like particle morphology, which is
maintained after sonication with Nafion, water, and ethanol during the catalyst deposition process.
124
However, after electrochemical treatments, the catalyst transitions into a plate-like morphology
where the boundaries between the particles are less defined. This transition in morphology was
accompanied by a structural change as revealed by the PXRD measurements of the post-echem
MOF (Figure 4.13). The crystallinity of 1 was completely lost after 1 h of electrolysis, suggesting
that structural changes induced under catalytic conditions.
Figure 4.12 SEM images of (a) as-prepared 1; (b) composite of 1 after sonication with water, ethanol, and Nafion; (c)
1 after 1 h electrolysis; (d) 1 after 4 h electrolysis.
Figure 4.13 PXRD patterns of 1 before and after 1 h CPE at -0.59 V vs RHE in a N2-saturated pH 1.3 solution.
125
Next, the material was analyzed by X-ray photoelectron spectroscopy (XPS) to reveal its
chemical composition after catalysis. Figure 4.14 shows the high-resolution XPS spectra of 1
before and after 1 h CPE at -0.59 V vs RHE. Two sampling spots were collected for the post-CPE
sample to capture a better representation of the catalyst surface. For the Ni 2p region, both before-
and after-CPE spectra display similar signals, with a major set of peaks at 853.8 and 871.1 eV,
suggestive of Ni(II) ions, as expected for 1 (Figure 4.14). The absence of the Ni(0) signal, which
typically appears at 852.6 eV, precludes the formation of Ni(0) nanoparticles on the electrode
surface, which has been reported in other analogous nickel dithiolate systems.
45,46
The Cu 2p
spectra of the as-prepared sample show a set of Cu 2p peaks at 932.3 and 952.2 eV, accompanied
by a set of satellite peaks at 934.4 and 954.3 eV (Figure 4.14). For the postelectrolysis sample, one
sampling spot showed an attenuated Cu signal, with one set of Cu 2p peaks at 932.3 and 952.2 eV,
whereas the other spot showed no presence of Cu. Although these two sampling sites showed
different Cu features, which are likely due to the inhomogeneous distribution of the catalyst on the
electrode surface, they both suggest that the surface of the catalyst layer has a lower Cu
concentration after electrocatalysis, as compared to that of the pristine catalyst. The S 2s region of
the as-prepared sample exhibits a peak around 226.2 eV, which is maintained after electrolysis,
suggesting that the integrity of the Ni(pdt)2 moiety is likely preserved during catalysis (Figure
4.14). The broad satellite peak around 233 eV is attributed to the S moieties in Nafion, which was
used to bind the catalyst onto the electrode.
To further evaluate the chemical composition changes of the catalyst on a bulk scale, the
materials were analyzed by the energy-dispersive X-ray spectroscopy (EDX), with results
summarized in Table 4.2. The relative ratio of Ni and Cu was found to be 1.00:0.88 in the pristine
catalyst, with a slight deficiency of Cu likely due to Cu vacancies in the framework. After 1 h of
126
electrolysis at -0.59 V vs RHE, the relative ratio of Ni and Cu increased to 1.00:1.18, which further
increased to 1.00:1.54 when the electrolysis was extended to 4 hours. These results suggest that
the amount of Cu increases relative to Ni during catalysis, meaning more Ni was leaching out of
the framework than Cu. However, these EDX results seem to conflict with the XPS observations,
which suggest a decrease in the relative Cu content after catalysis. Nonetheless, additional
evidence provided by the inductively coupled plasma-optical emission spectroscopy (ICP-OES)
of the digested catalyst is in support of the EDX results, where an increase in the relative Cu
content was observed in the postcatalytic material (Table 4.3). Additionally, the postelectrolysis
electrolyte solution was found to contain both Cu and Ni, with Ni having a much higher
concentration (Ni/Cu = 1.00:0.30). This further supports that more Ni was leaching out of the
framework during catalysis.
Figure 4.14 High-resolution XPS spectra of the as-prepared 1 (purple) and the post-CPE 1 sampled at two different
spots (blue and red). (a) Ni 2p region; (b) Cu 2p region; (c) S 2s region.
127
Table 4.2 EDX measurement results of 1 before and after electrolysis (at -0.59 V vs. RHE)
Atomic Ratio Ni Cu
Before electrolysis 1.00 0.88
After 1 hour electrolysis 1.00 1.18
After 4 hour electrolysis 1.00 1.54
Table 4.3 ICP-OES measurement results of 1 before and after electrolysis (at -0.59 V vs. RHE)
Molar ratio Ni Cu
Before electrolysis - catalyst 1.00 0.80
After 1 hour electrolysis - catalyst 1.00 1.29
After 1 hour electrolysis - electrolyte 1.00 0.30
After 4 hour electrolysis - catalyst 1.00 1.78
After 4 hour electrolysis - electrolyte 1.00 0.31
The aforementioned characterizations of the MOF suggest that the catalyst is indeed
dynamically changing during catalysis, evidenced by a change in morphology, crystallinity, and
chemical composition. However, the process appears to be rather complicated, especially in view
of the conflicting evidence provided by XPS, EDX, and ICP-OES. Therefore, computational
studies were performed to shed light on the activation mechanism of the catalyst.
4.2.3 Density Functional Theory Calculations
Density functional theory calculations were performed on Cu[Ni(pdt)2] MOF using the VASP
package.
47–50
We started from the experimental structure of 1
37
and then added H atoms to model
the effect of H atom addition on its structural properties. The primitive cell of the MOF contains
2 formula units of Cu[Ni(pdt)2] (Figure 4.15). Next, two H atoms were added per primitive cell to
examine the preferred site of hydrogen addition. H atoms are used in the modeling instead of
protons (H
+
) to better mimic the electrocatalytic conditions, a combination of protons and electrons.
The H atoms were placed in positions close to the N atom within the pyrazine ring or the S atom
within the thiolate group of the pyrazinedithiolate linker. All geometries were optimized using the
PBE functional
51,52
with Grimme dispersion correction
53
and Becke–Johnson damping.
54
A
128
planewave energy cutoff of 520 eV, energy convergence criteria of 10
-5
eV, and force convergence
criteria of 0.02 eV/Å were used for all calculations. The Brillouin zone was sampled using a Γ-
centered 3×3×2 k-point grid, and projected-augmented wave function (PAW) pseudopotentials
were used for all of the structural optimizations.
55,56
Figure 4.15 The primitive cell used for DFT calculations
In the pristine MOF, the average Cu–N bond length is 2.05 Å and the average Ni–S bond length
is 2.16 Å, in good agreement with the previously reported experimental and computational bond
lengths.
37
Both Cu and Ni ions are in the formal +2 oxidation state and adopt a square planar
coordination geometry. Using PBE-D3 functional, the magnetic moment of the Cu center was
calculated to be -0.43 μB, consistent with one unpaired electron in the d
9
Cu(II) center in the square
planar crystal field. The calculated magnetic moment for Ni, using the PBE-D3 functional, is near
0.00 μB, consistent with zero unpaired electrons in the d
8
Ni(II) square planar configuration. It was
also found that the N atoms display a calculated magnetic moment of -0.06 μB, which could be
caused by covalent σ-bonding with the paramagnetic Cu ions or by a slight radical character on
the organic ligand.
To determine the preferred binding sites, various configurations of the H atoms were
considered as shown in Figure 4.16, with their respective electronic energy compared (Table 4.4).
The binding of the H atoms to the N sites is more favorable than that to the S sites by at least 1.43
eV per unit cell. Furthermore, for the N-binding configurations, the addition of one H atom per Cu
129
center (2H-diff-Cu-N) is favored over the addition of two H atoms to a single Cu center by at least
0.51 eV (2H-same-Cu-N).
Figure 4.16 A pictorial representation of a 2×2×2 supercell of Cu[Ni(pdt)2] MOF upon addition of two H atoms to
different sites per unit cell of the MOF (prior structural optimization). (a) H atoms are added to N centers attached to
two different Cu centers (2H-diff-Cu-N); (b) H atoms are attached to N centers, which are cis with respect to the same
Cu center (2H-same-Cu-cis-N); (c) H atoms are attached to N centers, which are trans with respect to the same Cu
center (2H-same-Cu-trans-N); (d) H atoms are attached to S centers, which are coordinated to two different Ni centers
(2H-diff-Ni-S); (e) H atoms are attached to S centers, which are cis with respect to the same Ni center (2H-same-Ni-
cis-S-1); (f) H atoms are attached to S centers, which belong to the same pdt ligand (2H-same-Ni-cis-S-2), (g) H
atoms are attached to S centers, which are trans with respect to the same Ni center (2H-same-Ni-trans-S). Color
coding for atoms: green (Ni), cyan (Cu), yellow (S), dark blue (N), white (H), and gray (C).
130
Table 4.4 Relative energies (eV) for the addition of two hydrogen atoms (2H) per unit cell of Cu[Ni(pdt)2] computed
using PBE-D3 functional
Binding Sites Relative Energy of H Atom Addition (eV)
2H-diff-Cu-N 0.00
2H-same-Cu-trans-N 0.51
2H-same-Cu-cis-N 0.83
2H-diff-Ni-S 1.43
2H-same-Ni-cis-S-1 2.27
2H-same-Ni-cis-S-2 2.27
2H-same-Ni-trans-S 2.43
Since the addition of H atoms on the N centers is favored, we decided to look into the structural
and electronic properties of the configurations where H atoms are added to the N centers in detail.
It is found that, for 2H-diff-Cu-N, two of the Cu–N bonds are cleaved and the originally 4-
coordinated Cu centers become 2-coordinated (Figure 4.17). This Cu–N bond cleavage also results
in a distortion of the coordination environment of the Cu centers. Specifically, the Cu–N bond
length is shortened to 1.86 Å as compared to 2.05 Å in the pristine MOF, and the distance between
the Cu center and the uncoordinated N atom is 3.40 Å. Similar distortion in the structure was also
observed for 2H-same-Cu-cis-N and 2H-same-Cu-trans-N binding structures (Table 4.5). The
Ni(pdt)2 linker stays intact for all considered binding motifs, as suggested by the similar Ni–S bond
length before and after H atom addition. However, part of the Ni(pdt)2 linkers is detached from the
framework due to the cleavage of the Cu–N bonds. To investigate the electronic property of the
MOF after H atom addition, the magnetic moment of the Cu center was calculated for all three N-
binding geometries (Table 4.6). Upon Cu–N bond cleavage, the calculated magnetic moment of
the Cu center decreases from 0.43 μB to 0.00 μB, suggesting a formal reduction from Cu(II) to
Cu(I). The relative stability of the closed shell d
10
Cu(I) system compared to that of the d
9
Cu(II)
system is likely to make the cleavage of Cu–N bond favorable. Further CM5 (Bader) charge
analysis
57–62
also revealed that upon H atom addition, the charge on the Cu centers decreases from
131
0.70(0.98) e
-
to 0.53(0.70) e
-
, consistent with the reduction of Cu(II) to Cu(I). On the other hand,
the CM5 and Bader charges on the Ni centers remain unchanged (Table 4.7) for all of the structures,
suggesting no change in the formal oxidation states of the Ni centers.
Figure 4.17 Optimized structure of 2H-diff-Cu-N using the PBE-D3 functional.
Table 4.5 Cu-N and Ni-S bond lengths (Å) of the pristine MOF and H atom added structures (on N center) after
geometry optimization.
Bond
Lengths
Pristine
MOF
2H-diff-Cu-N
2H-same-Cu-
cis-N
2H-same-Cu-
trans-N
4H-2-Cu-
trans-N
Cu1-N1 2.05 1.86 1.91 2.03 3.17
Cu1-N2 2.05 3.41 2.03 1.92 1.95
Cu1-N3 2.05 1.86 1.98 3.36 3.43
Cu1-N4 2.05 3.62 3.65 2.02 1.92
Cu2-N5 2.05 3.61 3.99 1.91 1.91
Cu2-N6 2.05 1.87 3.56 3.98 3.15
Cu2-N7 2.05 3.40 1.91 1.91 1.95
Cu2-N8 2.05 1.86 1.91 4.22 3.40
Ni1-S1 2.16 2.17 2.19 2.16 2.18
Ni1-S2 2.16 2.17 2.19 2.17 2.18
Ni1-S3 2.16 2.16 2.19 2.17 2.19
Ni1-S4 2.16 2.16 2.13 2.17 2.19
Ni2-S5 2.16 2.16 2.16 2.16 2.18
Ni2-S6 2.16 2.16 2.15 2.17 2.18
Ni2-S7 2.16 2.17 2.18 2.17 2.17
Ni2-S8 2.16 2.17 2.16 2.18 2.17
Table 4.6 Magnetic moments (μB) on Cu and Ni centers before and after the addition of H atoms on the N center
Magnetic
Moments(μB)
Pristine
MOF
2H-diff-Cu-
N
2H-same-
Cu-cis-N
2H-same-
Cu-trans-N
4H-2-Cu-
trans-N
Cu1 -0.43 0.00 0.00 0.00 -0.02
Cu2 -0.43 0.00 0.00 0.00 0.02
Ni1 -0.01 0.00 0.00 0.00 0.00
Ni2 0.00 0.00 0.00 0.00 0.00
Table 4.7 CM5(Bader) Charges on Cu and Ni centers before and after the addition of H atoms on the N center
132
CM5(Bader)
Charge
Pristine
MOF
2H-diff-Cu-
N
2H-same-Cu-
cis-N
2H-same-Cu-
trans-N
4H-2-Cu-
trans-N
Cu1 0.70(0.98) 0.53(0.71) 0.58(0.80) 0.57(0.80) 0.51(0.70)
Cu2 0.70(0.98) 0.53(0.70) 0.52(0.73) 0.52(0.73) 0.51(0.70)
Ni1 0.29(0.69) 0.28(0.64) 0.27(0.63) 0.29(0.66) 0.28(0.66)
Ni2 0.29(0.69) 0.27(0.64) 0.28(0.63) 0.26(0.63) 0.27(0.65)
Intrigued by the aforementioned results where the addition of two H atoms breaks Cu–N bonds
and partially detaches the Ni(pdt)2 molecular complex from the Cu centers, we explored the
addition of four H atoms to the same structures to see if it is possible to fully dissociate the Ni(pdt)2
molecular complex. As shown in Table 4.4, the 2H-diff-Cu-N and 2H-same-Cu-trans-N are the
two most stable structures for 2 H atom addition. Therefore, two H atoms were added per Cu center
in a trans fashion to generate the 4 H atom addition structure (4H-2Cu-trans-N). After structural
optimization, it is observed that more Cu–N bonds are cleaved compared to the two H addition
calculation, leaving some of the Ni(pdt)2 molecular complexes fully detached from the MOF
framework. As shown in Figure 4.18, the 2×2×2 supercell formed by stacking the primitive
cells displays a layered structure, where free Ni(pdt)2 complexes float in between the one-
dimensional [Cu2Ni(pdt)2] chains. Each Cu is coordinated to two different pyrazine rings via N
atoms with an average Cu–N bond length of 1.93 Å (Table 4.5). However, the cleavage of the
additional Cu-N bonds for the 4 H atom addition structure does not lead to further reduction of the
Cu(I) centers to Cu(0) while the Ni(II) centers again stay unchanged (Tables 4.6 and 4.7). It is
speculated that the pyrazine ligands are reduced during the process, as previous studies with the
[Ni(dcpdt)2]
2-
molecular complex suggest a ligand-based PCET process in the HER mechanism.
40
133
Figure 4.18 (a) Top view and (b) side view of a 2×2×2 supercell (shown for visualization purpose only) of
Cu[Ni(pdt)2] MOF (after structural optimization) upon addition of four hydrogen atoms (4 H) per unit cell of the MOF.
Figure 4.19 Preliminary calculations to determine the active site of the catalyst, where one H atom is added to: (a) the
Ni site; (b) the S site closer to N; (c) the S site closer to Cu; (d) the Cu site. (e): Optimized structure of (d), where the
H atom transferred to the N atom after structural optimization.
Table 4.8 Absolute and relative energies (eV) of various structures shown in Figure 4.19
Structure Absolute Energy (eV) Relative Energy (eV)
H on Ni -306.600662 1.19
H on S near N -306.914625 0.87
H on S near Cu -306.997072 0.79
H on N
-307.787581 0.00
134
Furthermore, some additional calculations were performed to determine the active sites of the
catalysts for the HER. Starting with the optimized 2H-diff-Cu-N structure (Figure 4.17), another
H atom was added to the Ni, S, or Cu site (Figure 4.19). While the addition of H atom to the Ni or
S sites leads to a relatively stable structure, the H atom added to the Cu site transfers immediately
to the neighboring N atom of the pyrazine ring upon structural optimization (Figure 4.19e). Such
optimized structure has a lower energy compared to the S or Ni H-binding structures (Table 4.8).
These computational results suggest that the Cu center has low H affinity and likely does not
participate in the HER mechanism. In contrast, the high H affinity of the pyrazine linker suggests
that the HER mechanism is very likely ligand-based, similar to that observed for the molecular
analogue.
40
4.2.4 Discussion
Based on the experimental and computational results, it is proposed that under catalytic
conditions, 1 undergoes a decomposition process, which involves the cleavage of Cu–N bonds
caused by the protonation of the N site. The DFT calculations suggest that the preferred protonation
sites are the N atoms within the pyrazine rings, instead of the S atoms as seen for many other
dithiolene-based complexes. It is further shown that the addition of the H atom to the N atom
induces the cleavage of Cu–N bonds, which subsequently results in the reduction of Cu(II) to Cu(I)
and a distortion of the coordination geometry of the Cu centers, confirmed by a change in the Cu–
N bond length. It is predicted that such changes in the local coordination environment can alter the
extended framework structure, leading to loss of crystallinity and a change from the pristine
morphology, which is consistent with the results obtained from the PXRD and SEM studies. The
DFT calculations also help to explain the complex redox features observed during the CV
experiments (Figure 4.5), where the precatalytic wave could be partially attributed to the cleavage
135
of the Cu–N bonds and the reduction of the Cu centers, and the return oxidation peak could be
related to the oxidation of the Cu sites.
The 4 H atom addition calculations suggest that, in the presence of abundant electrons and
protons, it is possible to fully dissociate a portion of the Ni(pdt)2 complexes from the framework.
It is proposed that the detached Ni(pdt)2 moieties can subsequently diffuse and migrate toward the
catalyst/electrolyte interface and eventually leach into the electrolyte solution. This explains the
contradictory results of XPS, EDX, and ICP. As XPS is a surface analysis technique, with a typical
sampling depth of 5 nm, it is only capable of capturing the characteristics of the surface of the
catalyst layer, which is dominated by the free Ni moieties. On the other hand, EDX and ICP are
bulk analysis techniques that reveal the chemical composition of the bulk catalyst, in which case
the Cu will be the dominant metal species. However, it is noteworthy that the dissolution of the
Ni(pdt)2 complex is not complete and the activated catalyst still contains a significant amount of
Ni as suggested by the EDX and ICP measurements (Table 4.2, Table 4.3). This is consistent with
the DFT results, where the four H atom addition leads to the formation of free Ni(pdt)2 complexes
sandwiched in between 1D [Cu2Ni(pdt)2] chains.
Notably, the DFT-predicted mechanism for the transformation of 1 is unprecedented compared
to other analogues dithiolene-based electrocatalysts. The most well-understood CoTHT 2D MOF,
constructed by 2,3,6,7,10,11-triphenylenehexathiolate linkers and cobalt ion nodes, displays good
stability under catalytic conditions for up to 24 hours of electrolysis, with near unity FE for the
HER.
11
On the other hand, the NiBHT coordination polymer, constructed by benzenehexathiolate
linkers and nickel ion nodes, exhibits a compromised stability and an FE between 65% to 75%,
42
which are attributed to the instability of the NiS4 core, as reported by many studies on Ni
dithiolene-based catalysts.
45,63,64
Unlike these previous cases, the electrochemically induced
136
transformation of 1 is caused by the high proton affinity of the N atoms and the weak Cu–N bond.
While the molecular Ni(pdt)2 complexes have illustrated excellent stability,
38–40
it is crucial to
engineer more rigid linker-node connectivity in future studies to incorporate the well-defined
Ni(pdt)2 active units into an extended MOF structure with improved stability.
Regardless of the mechanism of the transformation process, experimental evidence clearly
suggests that the electrochemically treated species 1 is more stable and more active compared to
the pristine catalyst. As shown by the CPE experiments (Figure 4.6, Figure 4.7), 1 transforms
quickly under catalytic conditions, with the most drastic current enhancement occurring during the
first 15 min of the electrolysis. The low FE observed during this period is explained by the DFT
results, where the cleavage of Cu–N bonds involves the reduction of Cu centers and hence the
consumption of electrons. The stabilized current response and enhanced FE during later periods of
electrolysis suggest that the transformed 1 is more stable and produces H2 more selectively (Figure
4.7). The higher HER activity of the post-electrolysis 1 is characterized by the LSV measurements,
where a 200 mV reduction in the catalytic onset and 𝜂
'567/06
- (Figure 4.8) is observed.
Additionally, EIS also suggests a nearly 4-fold reduction in the Rct, indicating a much favored
HER kinetics at the new catalyst interface (Figure 4.6b). The wash test following the CPE
experiment (Figure 4.9) shows that the activated catalyst is still heterogeneous in nature, meaning
the observed high activity is not contributed by the in-solution Cu or Ni-containing species.
Although the real identity of the active center is not yet clear, DFT calculations and physical
characterization techniques all suggest that the electrochemically treated 1 could have a
significantly different structure compared to the pristine MOF.
Previous studies with analogous nickel pyrazinedithiolate molecular complexes have
illustrated that the protonation of the N atom within the pyrazine ring is an essential step in the
137
HER mechanism.
38–40
In our case, the DFT studies reveal that the cleavage of the Cu–N bond leads
to the formation of protonated Ni(pdt)2 moieties, as shown in the 2 H atom addition studies, which
could serve as the activated form of the catalyst. Further DFT calculations also suggest that the Cu
sites are likely not involved in the HER mechanism, because when a H atom is added to Cu, it
immediately transfers to the neighboring N atom of the pyrazine ring (Figure 4.19e). Furthermore,
the calculated energy of the H-binding structure is the lowest when this H is on the N, compared
to S and Ni. In view of such DFT results, along with the previously reported ligand-based PCET
mechanism for the molecular [Ni(dcpdt)2]
2-
molecular complex, we have strong reason to believe
that the HER occurs preferably on the Ni(pdt)2 moieties. However, due to the lack of structural
information of the catalyst, detailed mechanistic studies and computational modeling cannot be
performed at the current stage.
The DFT calculations reported here only model what could happen to the MOF during catalysis,
but not how likely, or under what conditions this transformation will occur. Therefore, they cannot
offer a definitive answer about what is the ultimate structure of the MOF under the experimental
conditions. However, the current experimental and computational results suggest that the dynamic
changes of the MOF catalyst during electrolysis are induced by the cleavage of the Cu–N bonds,
resulting in a more active HER catalyst, with activity on par with that of the analogous dithiolate-
based MOFs. Considering the inhomogeneity of the system, it is expected that the activated form
of the catalyst has a complex local structure and an amorphous long-range order (as suggested by
PXRD), which necessitates more complex structural characterizations to probe the identity of the
catalytically active sites experimentally.
138
4.3 Conclusions
In summary, a 3D dithiolene-based MOF, Cu[Ni(2,3-pyrazinedithiolate)2] (1), is investigated
as a possible electrocatalyst for the hydrogen evolution reaction in fully aqueous media. It is found
that the pristine MOF catalyst exhibits a moderate activity with a catalytic onset at -0.43 V vs.
RHE, an overpotential (𝜂) of 0.53 V to reach a current density of 10 mA/cm
2
, a Tafel slope of 69.0
mV/dec, and an exchange current density (j0) of 10
-6.7
mA/cm
2
. It is further observed that during
controlled potential electrolysis, 1 undergoes a rapid activation process that results in a more active
material with a 200 mV reduction in the catalytic onset and 𝜂
'567/06
- . To understand the
activation process, physical characterizations were performed on the postelectrolysis 1. PXRD
measurement revealed a complete loss of the crystallinity of the material after electrochemical
treatment, while the SEM images showed a drastic change in morphology. Elemental analysis by
EDX and ICP-OES suggested a relative increase in the Cu content in the bulk catalyst, while the
XPS indicated a higher Ni concentration on the surface level. DFT calculations were performed to
further provide insights into the process. It is found that the addition of H atoms to the MOF can
cause the Cu–N bond to break and leave the Ni(pdt)2 moieties partially or fully detached from the
MOF structure. Future studies are required to pinpoint the true identity of the activated
electrocatalyst, the catalyst transformation pathway, as well as the HER mechanism. The
understanding of these processes will provide insights into the modification of the current MOF
catalyst to achieve better stability and HER performance.
4.4 Supplementary Experimental Methods
4.4.1 Synthesis of Cu[Ni(2,3-pyrazinedithiolate)2] (1)
The synthesis of Cu[Ni(pdt)2] (1) was adapted from a previously reported procedure.
37
A
Schlenk flask was charged with dry, degassed CH3CN (150 mL) under a dry, Ar atmosphere. Solid
139
Na[Ni(pdt)2] (0.350 g, 0.983 mmol) was added to the flask, and the suspension was stirred until
all solid had dissolved. In a separate Schlenk flask, CuI (0.187 g, 0.982 mmol) was dissolved in
dry, degassed CH3CN (40 mL). The CuI solution was then added dropwise to the Na[Ni(pdt)2]
solution with constant stirring, resulting in the rapid precipitation of Cu[Ni(pdt)2] as a dark red
solid. The resulting suspension was stirred for 2 hours, and the solid was collected by filtration in
air, washed with CH3CN (3 × 50 mL), and dried under vacuum at 90 °C for 24 hours. The solid
was then further activated on a Micromeritics ASAP 2420 instrument equipped with a turbo pump,
yielding 0.243g (61%) of the product as a dark red microcrystalline solid. The identity of the
product was confirmed via powder X-ray diffraction.
4.4.2 Deposition of 1 for Electrochemistry Study
1 was deposited as a suspension composed of 2 mg of the MOF, 20 µL of Nafion solution (0.5
wt%, purchased from Sigma-Aldrich
â
), 45 µL of water, and 135 µL of ethanol. The mixture was
sonicated for 20 mins to form a uniformly dispersed suspension. 10 µL of this suspension was then
drop casted onto a freshly polished glassy carbon electrode using a micro syringe, and dried at
room temperature prior to use.
4.4.3 Electrochemical Methods
Electrochemistry experiments were carried out using a VersaSTAT 3 potentiostat in a three-
electrode configuration electrochemical cell under an inert (N2) atmosphere. A rotating disk
electrode (RDE, with glassy carbon insert, 0.196 cm
2
surface area) was used as the working
electrode. The rotation rate of the RDE is set to 1600 rpm, if not otherwise stated. The glassy
carbon electrode was polished with 0.05 µm Al2O3 polish powder and sonicated in Millipore water
for 10 minutes prior to use. A graphite rod, purchased from Graphite Machining, Inc. (Grade NAC-
500 Purified, < 10 ppm ash level), was used as the counter electrode. The reference electrode,
140
placed in a separate compartment and connected by a Vycor tip, was based on an aqueous
Ag/AgCl/saturated 3.5 M KCl electrode. The reference electrode in aqueous media was calibrated
externally relative to ferrocenecarboxylic acid (Fc-COOH) at pH 7.0, with the Fe
3+/2+
couple at
0.28 V vs. Ag/AgCl. All potentials reported in this paper were converted to the reversible hydrogen
electrode (RHE) by adding a value of (0.205 + 0.059 × pH) V, or to the normal hydrogen electrode
(NHE) by adding a value of 0.205 V (for variable pH studies).
The aqueous electrolyte solutions used in the electrochemical experiments were prepared as
follows. For the pH 1.3 solution, 0.534 mL of 18.7 M H2SO4 was added to 200 mL 0.1 M NaClO4.
For the pH 4.6 solution, NaOAc (1.605 g) and acetic acid (1.2 mL) was added to 200 mL 0.1 M
NaClO4. For the pH 7 solution, NaH
2
PO
4
(0.936 g) and Na
2
HPO
4
(3.273 g) were dissolved in 200
mL 0.1 M NaClO4. The pH of the solutions was measured with a benchtop Mettler Toledo pH
meter. Prior to each electrochemical experiment, the electrolyte solution was purged with nitrogen
thoroughly to avoid the interference of oxygen reduction reaction.
Electrochemical impedance spectroscopy (EIS) measurements were carried out at different
overpotentials in the frequency range of 100 kHz – 0.1 Hz with 10 mV sinusoidal perturbations.
Experimental EIS data were analyzed and fitted with the ZSimpWin software.
The obtained polarization curves were corrected for iR loss according to the following equation:
Ecorr = Emea – iRs
Where Ecorr is the iR-corrected potential, Emea is the experimentally measured potential, and Rs
is the solution resistance extracted from the fitted EIS data.
Controlled potential electrolysis (CPE) measurements were conducted in a sealed two-
chambered H-cell where the first chamber held the working and reference electrodes in 40 mL of
electrolyte solution and the second chamber held the counter electrode in 20 mL of electrolyte
141
solution. The two chambers were both under N2 and separated by a fine porosity glass frit. CPE
experiments were performed with a glassy carbon plate electrode (6 cm × 1 cm × 0.3 cm; Tokai
Carbon USA) as the working electrode and a graphite rod as the counter electrode. The reference
electrode was a Ag/AgCl/saturated 3.5 M KCl (aq) electrode separated from the solution by a
Vycor tip. Using a gas-tight syringe, 2 mL of gas was 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. To determine the Faradaic
efficiency, the theoretical H2 amount based on total charge flowed was compared with the GC-
detected H2 produced from controlled-potential electrolysis.
4.4.4 Physical Characterization Methods
X-ray photoelectron spectroscopy (XPS) data were collected using a Kratos AXIS Ultra
instrument. The monochromatic X-ray 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
"+
Torr. The XPS data were analyzed using the CasaXPS software.
Powder X-ray diffraction (PXRD) was performed on a Rigaku Ultima IV X-Ray diffractometer
in reflectance parallel beam/parallel slit alignment geometry. The measurement employed Cu Kα
line focused radiation at 1760 W (40 kV, 44 mA) power and a Ge crystal detector fitted with a 0.6
mm radiation entrance slit. Samples were mounted on zero-background sample holders and were
observed using a 0.01° 2θ step scan from 5.0 – 50.0° with a scan rate of 1°/min.
Inductively coupled plasma optical emission spectroscopy (ICP-OES) measurements were
performed using a Thermo Scientific iCAP 7000 ICP-OES. After electrochemical tests, the
142
catalyst was collected from the electrode surface, digested in 2 mL of concentrated nitric acid,
diluted by Millipore water to 25 mL, and tested by ICP-OES to determine the concentration of the
Co and Cu moieties.
Scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDX) were
performed on an FEI Nova NanoSEM 450 scanning electron microscope using an accelerating
voltage of 15 kV.
4.5 References
(1) REN21. Renewables 2019 Global Status Report. https://www.ren21.net/gsr-2019/.
(2) Seh, Z. W.; Kibsgaard, J.; Dickens, C. F.; Chorkendorff, I.; Nørskov, J. K.; Jaramillo, T. F.
Combining Theory and Experiment in Electrocatalysis: Insights into Materials Design. Science
2017, 355 (6321), eaad4998.
(3) Downes, C. A.; Marinescu, S. C. Electrocatalytic Metal–Organic Frameworks for Energy
Applications. ChemSusChem 2017, 10 (22), 4374–4392.
(4) Clough, A. J.; Yoo, J. W.; Mecklenburg, M. H.; Marinescu, S. C. Two-Dimensional Metal–
Organic Surfaces for Efficient Hydrogen Evolution from Water. J. Am. Chem. Soc. 2015, 137 (1),
118–121.
(5) Dong, R.; Zheng, Z.; Tranca, D. C.; Zhang, J.; Chandrasekhar, N.; Liu, S.; Zhuang, X.;
Seifert, G.; Feng, X. Immobilizing Molecular Metal Dithiolene-Diamine Complexes on 2D Metal-
Organic Frameworks for Electrocatalytic H2 Production. Chem. Eur. J. 2017, 23 (10), 2255–2260.
(6) Dong, R.; Pfeffermann, M.; Liang, H.; Zheng, Z.; Zhu, X.; Zhang, J.; Feng, X. Large-Area,
Free-Standing, Two-Dimensional Supramolecular Polymer Single-Layer Sheets for Highly
Efficient Electrocatalytic Hydrogen Evolution. Angew. Chem., Int. Ed. 2015, 54 (41), 12058–
12063.
(7) Qin, J.-S.; Du, D.-Y.; Guan, W.; Bo, X.-J.; Li, Y.-F.; Guo, L.-P.; Su, Z.-M.; Wang, Y.-Y.;
Lan, Y.-Q.; Zhou, H.-C. Ultrastable Polymolybdate-Based Metal–Organic Frameworks as Highly
Active Electrocatalysts for Hydrogen Generation from Water. J. Am. Chem. Soc. 2015, 137 (22),
7169–7177.
(8) Hod, I.; Deria, P.; Bury, W.; Mondloch, J. E.; Kung, C.-W.; So, M.; Sampson, M. D.; Peters,
A. W.; Kubiak, C. P.; Farha, O. K.; et al. A Porous Proton-Relaying Metal-Organic Framework
Material That Accelerates Electrochemical Hydrogen Evolution. Nat. Commun. 2015, 6 (1), 8304.
(9) Micheroni, D.; Lan, G.; Lin, W. Efficient Electrocatalytic Proton Reduction with Carbon
Nanotube-Supported Metal–Organic Frameworks. J. Am. Chem. Soc. 2018, 140 (46), 15591–
15595.
143
(10) Wu, Y.-P.; Zhou, W.; Zhao, J.; Dong, W.-W.; Lan, Y.-Q.; Li, D.-S.; Sun, C.; Bu, X.
Surfactant-Assisted Phase-Selective Synthesis of New Cobalt MOFs and Their Efficient
Electrocatalytic Hydrogen Evolution Reaction. Angew. Chem., Int. Ed. 2017, 56 (42), 13001–
13005.
(11) Chen, K.; Downes, C. A.; Schneider, E.; Goodpaster, J. D.; Marinescu, S. C. Improving
and Understanding the Hydrogen Evolving Activity of a Cobalt Dithiolene Metal–Organic
Framework. ACS Appl. Mater. Interfaces 2021, 13 (14), 16384–16395.
(12) Kung, C.-W.; Mondloch, J. E.; Wang, T. C.; Bury, W.; Hoffeditz, W.; Klahr, B. M.; Klet,
R. C.; Pellin, M. J.; Farha, O. K.; Hupp, J. T. Metal–Organic Framework Thin Films as Platforms
for Atomic Layer Deposition of Cobalt Ions To Enable Electrocatalytic Water Oxidation. ACS
Appl. Mater. Interfaces 2015, 7 (51), 28223–28230.
(13) Gong, Y.; Shi, H.-F.; Hao, Z.; Sun, J.-L.; Lin, J.-H. Two Novel Co(II) Coordination
Polymers Based on 1,4-Bis(3-Pyridylaminomethyl)Benzene as Electrocatalysts for Oxygen
Evolution from Water. Dalt. Trans. 2013, 42 (34), 12252.
(14) Lu, X.-F.; Liao, P.-Q.; Wang, J.-W.; Wu, J.-X.; Chen, X.-W.; He, C.-T.; Zhang, J.-P.; Li,
G.-R.; Chen, X.-M. An Alkaline-Stable, Metal Hydroxide Mimicking Metal–Organic Framework
for Efficient Electrocatalytic Oxygen Evolution. J. Am. Chem. Soc. 2016, 138 (27), 8336–8339.
(15) Johnson, B. A.; Bhunia, A.; Ott, S. Electrocatalytic Water Oxidation by a Molecular
Catalyst Incorporated into a Metal–Organic Framework Thin Film. Dalt. Trans. 2017, 46 (5),
1382–1388.
(16) Lin, S.; Pineda-Galvan, Y.; Maza, W. A.; Epley, C. C.; Zhu, J.; Kessinger, M. C.; Pushkar,
Y.; Morris, A. J. Electrochemical Water Oxidation by a Catalyst-Modified Metal-Organic
Framework Thin Film. ChemSusChem 2017, 10 (3), 469–469.
(17) Shen, J.-Q.; Liao, P.-Q.; Zhou, D.-D.; He, C.-T.; Wu, J.-X.; Zhang, W.-X.; Zhang, J.-P.;
Chen, X.-M. Modular and Stepwise Synthesis of a Hybrid Metal–Organic Framework for Efficient
Electrocatalytic Oxygen Evolution. J. Am. Chem. Soc. 2017, 139 (5), 1778–1781.
(18) Wang, L.; Wu, Y.; Cao, R.; Ren, L.; Chen, M.; Feng, X.; Zhou, J.; Wang, B. Fe/Ni Metal-
Organic Frameworks and Their Binder-Free Thin Films for Efficient Oxygen Evolution with Low
Overpotential. ACS Appl. Mater. Interfaces 2016, 8 (26), 16736–16743.
(19) Zhao, S.; Wang, Y.; Dong, J.; He, C. T.; Yin, H.; An, P.; Zhao, K.; Zhang, X.; Gao, C.;
Zhang, L.; et al. Ultrathin Metal-Organic Framework Nanosheets for Electrocatalytic Oxygen
Evolution. Nat. Energy 2016, 1 (12), 16184.
(20) Duan, J.; Chen, S.; Zhao, C. Ultrathin Metal-Organic Framework Array for Efficient
Electrocatalytic Water Splitting. Nat. Commun. 2017, 8, 15341.
(21) Miner, E. M.; Fukushima, T.; Sheberla, D.; Sun, L.; Surendranath, Y.; Dincă, M.
Electrochemical Oxygen Reduction Catalysed by Ni3(Hexaiminotriphenylene)2. Nat. Commun.
2016, 7 (1), 10942.
144
(22) Miner, E. M.; Gul, S.; Ricke, N. D.; Pastor, E.; Yano, J.; Yachandra, V. K.; Van Voorhis,
T.; Dincă, M. Mechanistic Evidence for Ligand-Centered Electrocatalytic Oxygen Reduction with
the Conductive MOF Ni3(Hexaiminotriphenylene)2. ACS Catal. 2017, 7 (11), 7726–7731.
(23) Miner, E. M.; Wang, L.; Dincă, M. Modular O2 Electroreduction Activity in Triphenylene-
Based Metal–Organic Frameworks. Chem. Sci. 2018, 9 (29), 6286–6291.
(24) Lions, M.; Tommasino, J.-B.; Chattot, R.; Abeykoon, B.; Guillou, N.; Devic, T.;
Demessence, A.; Cardenas, L.; Maillard, F.; Fateeva, A. Insights into the Mechanism of
Electrocatalysis of the Oxygen Reduction Reaction by a Porphyrinic Metal Organic Framework.
Chem. Commun. 2017, 53 (48), 6496–6499.
(25) Usov, P. M.; Huffman, B.; Epley, C. C.; Kessinger, M. C.; Zhu, J.; Maza, W. A.; Morris,
A. J. Study of Electrocatalytic Properties of Metal–Organic Framework PCN-223 for the Oxygen
Reduction Reaction. ACS Appl. Mater. Interfaces 2017, 9 (39), 33539–33543.
(26) Liu, X. H.; Hu, W. L.; Jiang, W. J.; Yang, Y. W.; Niu, S.; Sun, B.; Wu, J.; Hu, J. S. Well-
Defined Metal-O6 in Metal-Catecholates as a Novel Active Site for Oxygen Electroreduction. ACS
Appl. Mater. Interfaces 2017, 9 (34), 28473–28477.
(27) Kornienko, N.; Zhao, Y.; Kley, C. S.; Zhu, C.; Kim, D.; Lin, S.; Chang, C. J.; Yaghi, O.
M.; Yang, P. Metal-Organic Frameworks for Electrocatalytic Reduction of Carbon Dioxide. J. Am.
Chem. Soc. 2015, 137 (44), 14129–14135.
(28) Hod, I.; Sampson, M. D.; Deria, P.; Kubiak, C. P.; Farha, O. K.; Hupp, J. T. Fe-Porphyrin-
Based Metal–Organic Framework Films as High-Surface Concentration, Heterogeneous Catalysts
for Electrochemical Reduction of CO2. ACS Catal. 2015, 5 (11), 6302–6309.
(29) Kung, C. W.; Audu, C. O.; Peters, A. W.; Noh, H.; Farha, O. K.; Hupp, J. T. Copper
Nanoparticles Installed in Metal-Organic Framework Thin Films Are Electrocatalytically
Competent for CO2 Reduction. ACS Energy Lett. 2017, 2 (10), 2394–2401.
(30) Johnson, E. M.; Haiges, R.; Marinescu, S. C. Covalent-Organic Frameworks Composed of
Rhenium Bipyridine and Metal Porphyrins: Designing Heterobimetallic Frameworks with Two
Distinct Metal Sites. ACS Appl. Mater. Interfaces 2018, 10 (44), 37919–37927.
(31) Lin, S.; Diercks, C. S.; Zhang, Y.-B.; Kornienko, N.; Nichols, E. M.; Zhao, Y.; Paris, A.
R.; Kim, D.; Yang, P.; Yaghi, O. M.; et al. Covalent Organic Frameworks Comprising Cobalt
Porphyrins for Catalytic CO2 Reduction in Water. Science 2015, 349 (6253), 1208–1213.
(32) Diercks, C. S.; Lin, S.; Kornienko, N.; Kapustin, E. A.; Nichols, E. M.; Zhu, C.; Zhao, Y.;
Chang, C. J.; Yaghi, O. M. Reticular Electronic Tuning of Porphyrin Active Sites in Covalent
Organic Frameworks for Electrocatalytic Carbon Dioxide Reduction. J. Am. Chem. Soc. 2018, 140
(3), 1116–1122.
(33) Matheu, R.; Gutierrez-Puebla, E.; Monge, M. Á.; Diercks, C. S.; Kang, J.; Prévot, M. S.;
Pei, X.; Hanikel, N.; Zhang, B.; Yang, P.; et al. Three-Dimensional Phthalocyanine Metal-
Catecholates for High Electrochemical Carbon Dioxide Reduction. J. Am. Chem. Soc. 2019, 141
145
(43), 17081–17085.
(34) Meng, Z.; Luo, J.; Li, W.; Mirica, K. A. Hierarchical Tuning of the Performance of
Electrochemical Carbon Dioxide Reduction Using Conductive Two-Dimensional
Metallophthalocyanine Based Metal-Organic Frameworks. J. Am. Chem. Soc. 2020, 142 (52),
21656–21669.
(35) Kobayashi, Y.; Jacobs, B.; Allendorf, M. D.; Long, J. R. Conductivity, Doping, and Redox
Chemistry of a Microporous Dithiolene-Based Metal-Organic Framework. Chem. Mater. 2010, 22
(14), 4120–4122.
(36) Peng, Y.-L.; Pham, T.; Li, P.; Wang, T.; Chen, Y.; Chen, K.-J.; Forrest, K. A.; Space, B.;
Cheng, P.; Zaworotko, M. J.; et al. Robust Ultramicroporous Metal-Organic Frameworks with
Benchmark Affinity for Acetylene. Angew. Chem., Int. Ed. 2018, 57 (34), 10971–10975.
(37) Aubrey, M. L.; Kapelewski, M. T.; Melville, J. F.; Oktawiec, J.; Presti, D.; Gagliardi, L.;
Long, J. R. Chemiresistive Detection of Gaseous Hydrocarbons and Interrogation of Charge
Transport in Cu[Ni(2,3-Pyrazinedithiolate)2] by Gas Adsorption. J. Am. Chem. Soc. 2019, 141 (12),
5005–5013.
(38) Koshiba, K.; Yamauchi, K.; Sakai, K. A Nickel Dithiolate Water Reduction Catalyst
Providing Ligand-Based Proton-Coupled Electron-Transfer Pathways. Angew. Chem., Int. Ed.
2017, 56 (15), 4247–4251.
(39) Aimoto, Y.; Koshiba, K.; Yamauchi, K.; Sakai, K. A Family of Molecular Nickel
Hydrogen Evolution Catalysts Providing Tunable Overpotentials Using Ligand-Centered Proton-
Coupled Electron Transfer Paths. Chem. Commun. 2018, 54 (91), 12820–12823.
(40) Koshiba, K.; Yamauchi, K.; Sakai, K. Consecutive Ligand-Based PCET Processes
Affording a Doubly Reduced Nickel Pyrazinedithiolate Which Transforms into a Metal Hydride
Required to Evolve H2. Dalt. Trans. 2019, 48 (2), 635–640.
(41) Hayashi, M.; Takahashi, Y.; Yoshida, Y.; Sugimoto, K.; Kitagawa, H. Role of d -Elements
in a Proton–Electron Coupling of d –π Hybridized Electron Systems. J. Am. Chem. Soc. 2019, 141
(29), 11686–11693.
(42) Downes, C. A.; Clough, A. J.; Chen, K.; Yoo, J. W.; Marinescu, S. C. Evaluation of the H2
Evolving Activity of Benzenehexathiolate Coordination Frameworks and the Effect of Film
Thickness on H2 Production. ACS Appl. Mater. Interfaces 2018, 10 (2), 1719–1727.
(43) Micheroni, D.; Lan, G.; Lin, W. Efficient Electrocatalytic Proton Reduction with Carbon
Nanotube-Supported Metal–Organic Frameworks. J. Am. Chem. Soc. 2018, 140 (46), 15591–
15595.
(44) 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 (5), 2827–2834.
146
(45) Fang, M.; Engelhard, M. H.; Zhu, Z.; Helm, M. L.; Roberts, J. A. S. Electrodeposition from
Acidic Solutions of Nickel Bis(Benzenedithiolate) Produces a Hydrogen-Evolving Ni–S Film on
Glassy Carbon. ACS Catal. 2014, 4 (1), 90–98.
(46) Hu, C.; Ma, Q.; Hung, S.-F.; Chen, Z.-N.; Ou, D.; Ren, B.; Chen, H. M.; Fu, G.; Zheng, N.
In Situ Electrochemical Production of Ultrathin Nickel Nanosheets for Hydrogen Evolution
Electrocatalysis. Chem 2017, 3 (1), 122–133.
(47) Kresse, G.; Furthmüller, J. Efficiency of Ab-Initio Total Energy Calculations for Metals
and Semiconductors Using a Plane-Wave Basis Set. Comput. Mater. Sci. 1996, 6 (1), 15–50.
(48) Kresse, G.; Furthmüller, J. Efficient Iterative Schemes for Ab Initio Total-Energy
Calculations Using a Plane-Wave Basis Set. Phys. Rev. B 1996, 54 (16), 11169–11186.
(49) Kresse, G.; Hafner, J. Ab Initio Molecular-Dynamics Simulation of the Liquid-Metal--
Amorphous-Semiconductor Transition in Germanium. Phys. Rev. B 1994, 49 (20), 14251–14269.
(50) Kresse, G.; Hafner, J. Ab Initio Molecular Dynamics for Liquid Metals. Phys. Rev. B 1993,
47 (1), 558–561.
(51) Perdew, J. P.; Wang, Y. Accurate and Simple Analytic Representation of the Electron-Gas
Correlation Energy. Phys. Rev. B 1992, 45 (23), 13244–13249.
(52) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient Approximation Made Simple.
Phys. Rev. Lett. 1996, 77 (18), 3865–3868.
(53) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. A Consistent and Accurate Ab Initio
Parametrization of Density Functional Dispersion Correction (DFT-D) for the 94 Elements H-Pu.
J. Chem. Phys. 2010, 132 (15), 154104.
(54) Becke, A. D.; Johnson, E. R. A Density-Functional Model of the Dispersion Interaction. J.
Chem. Phys. 2005, 123 (15), 154101.
(55) Kresse, G.; Joubert, D. From Ultrasoft Pseudopotentials to the Projector Augmented-Wave
Method. Phys. Rev. B 1999, 59 (3), 1758–1775.
(56) Blöchl, P. E. Projector Augmented-Wave Method. Phys. Rev. B 1994, 50 (24), 17953–
17979.
(57) Bibila Mayaya Bisseyou, Y.; Bouhmaida, N.; Guillot, B.; Lecomte, C.; Lugan, N.;
Ghermani, N.; Jelsch, C. Experimental and Database-Transferred Electron-Density Analysis and
Evaluation of Electrostatic Forces in Coumarin-102 Dye. Acta Crystallogr. Sect. B Struct. Sci.
2012, 68 (6), 646–660.
(58) Sanville, E.; Kenny, S. D.; Smith, R.; Henkelman, G. Improved Grid-Based Algorithm for
Bader Charge Allocation. J. Comput. Chem. 2007, 28 (5), 899–908.
(59) Yu, M.; Trinkle, D. R. Accurate and Efficient Algorithm for Bader Charge Integration. J.
147
Chem. Phys. 2011, 134 (6), 064111.
(60) Tang, W.; Sanville, E.; Henkelman, G. A Grid-Based Bader Analysis Algorithm without
Lattice Bias. J. Phys. Condens. Matter 2009, 21 (8), 084204.
(61) Marenich, A. V.; Jerome, S. V.; Cramer, C. J.; Truhlar, D. G. Charge Model 5: An
Extension of Hirshfeld Population Analysis for the Accurate Description of Molecular Interactions
in Gaseous and Condensed Phases. J. Chem. Theory Comput. 2012, 8 (2), 527–541.
(62) Wang, B.; Li, S. L.; Truhlar, D. G. Modeling the Partial Atomic Charges in
Inorganometallic Molecules and Solids and Charge Redistribution in Lithium-Ion Cathodes. J.
Chem. Theory Comput. 2014, 10 (12), 5640–5650.
(63) Downes, C. A.; Marinescu, S. C. One Dimensional Metal Dithiolene (M = Ni, Fe, Zn)
Coordination Polymers for the Hydrogen Evolution Reaction. Dalt. Trans. 2016, 45 (48), 19311–
19321.
(64) Zarkadoulas, A.; Field, M. J.; Artero, V.; Mitsopoulou, C. A. Proton-Reduction Reaction
Catalyzed by Homoleptic Nickel-Bis-1,2-Dithiolate Complexes: Experimental and Theoretical
Mechanistic Investigations. ChemCatChem 2017, 9 (12), 2308–2317.
148
Chapter 5
Synthesis and Investigation of a Cobalt Triphenylenehexaselenolate Metal-
Organic Framework for Electrocatalytic Hydrogen Evolution
149
5.1 Introduction
In nature, hydrogenase enzymes are capable of catalyzing the hydrogen evolution reaction
(HER) with high efficiencies at minimal overpotential.
1
[NiFe] hydrogenase is one of the most
well-known hydrogenase, the understanding of which has inspired the design of numerous
synthetic HER catalysts.
2
However, more recent studies have suggested that the [NiFe]
hydrogenase actually have a strong bias toward H2 oxidation, whereas the hydrogen production
process is inhibited by the product (H2) itself.
3
In contrast, [NiFeSe] hydrogenase, a selenium-
containing subclass of the [NiFe] hydrogenase, has recently gained much attention as it exhibits a
higher H2 production rate compared to that of the standard [NiFe] hydrogenase with high tolerance
to H2 and O2. The selenium is present in the selenocysteine amino acid residue within the first
coordination sphere of the active site, in replacement of the sulfur-containing cysteine residue in
the original [NiFe] hydrogenase (Figure 5.1). The HER activity of [NiFeSe] hydrogenase was
reported to be up to 40 times higher than that of the corresponding [NiFe] hydrogenase,
4
and the
increased activity is proposed to result from the higher nucleophilicity of the selenocysteine amino
acid residue relative to that of the cysteine, and the lower pKa of a selenocysteine selenol compared
to that of a cysteine thiol.
3
The enhanced O2-tolerance of [NiFeSe] hydrogenase in comparison to
the conventional [NiFe] hydrogenase is attributed to the ease with which selenocysteine can be
oxidized and reduced back, along with the large size of the Se atom that all together help to protect
the active nickel center from the attack of O2.
3
Figure 5.1 Structures of the active sites of [NiFe] and [NiFeSe] hydrogenase
Fe
S
Ni
S
S
S
OC
NC
NC
Cys
[NiFe]-hydrogenase
Cys
Fe
S
Ni
Se
S
S
OC
NC
NC
Cys
[NiFeSe]-hydrogenase
Cys
Cys
Cys
Sec
Cys
Figure 1. Structures of [NiFe]- and [NiFeSe]-Hase active sites.
150
The discovery and understanding of the [NiFeSe] hydrogenase has inspired the development
of many selenium-containing synthetic electrocatalysts that are derived from their sulfur-
containing analogues.
5–10
For example, Cui and coworkers have reported that the CoSe2
nanoparticles display higher HER activity compared to that of the CoS2 nanoparticles.
5
It has been
proposed that the weaker Se–H bond (276 kJ/mol) compared to the S–H bond (363 kJ/mol) can
facilitate the hydrogen dissociation from the active sites, resulting in a faster active site turnover,
similar to the case of [NiFeSe] hydrogenase as discussed above.
11
A previous work from our group
has shown that the cobalt benzenetetraselenolate (CoBTSe) one-dimensional (1D) polymer
displays substantially higher HER activity compared to its sulfur-only analogue (CoBTT), with a
217 mV improvement in the overpotential to reach a 10 mA/cm
2
current density.
10
This
enhancement in catalytic performance is explained by an alternative mechanism pathway that is
not available for the CoBTT analogue, which could involve the recombination of two protonated
selenium moieties following a second electron reduction to produce H2.
Herein, a cobalt triphenylene-2,3,6,7,10,11-hexaselenolate (CoTHSe) metal-organic
framework (MOF) is synthesized and evaluated as an electrocatalyst for the HER. The CoTHSe
MOF is a derivative of the cobalt triphenylene-2,3,6,7,10,11-hexathiolate (CoTHT) MOF, which
serves as one of the most active MOF-based electrocatalysts for the HER in the literature.
12
Hence,
we set to explore here the HER activity change as a result of the replacement of the sulfur by
selenium in the ligand scaffold. In the case of [NiFeSe] hydrogenase, the introduction of Se
improves the O2 tolerance of the enzymes. Therefore, it would be interesting to explore if such
functionality of Se is preserved within the MOFs.
3
Although the application of CoTHSe for
electrocatalytic HER has not been explored, its synthesis and physical characterizations have been
previously reported by Zhu and coworkers.
13
Herein, the attention will be focused on the
151
electrochemical characterization of the material, with an emphasis on the evaluation of how
oxidation state influences the electrochemical behavior and the HER activity.
5.2 Results and Discussion
5.2.1 Synthesis and Physical Characterization
The synthesis of the cobalt triphenylene-2,3,6,7,10,11-hexaselenolate (CoTHSe) was
attempted through two different routes (Figure 5.2, see section 5.4 for more experimental details).
The first method is based on the synthesis of an analogous triphenylene-2,3,6,7,10,11-hexathiolate
(THT) ligand, where the sulfur-alkyl bond was cleaved by free electrons generated in Na and
liquid NH3 through a Birch reduction reaction.
14
Using similar reaction conditions, treatment of
2,3,6,7,10,11-hexakis(tert-butylselenyl)triphenylene (THSe
t
Bu) with excess Na in liquid NH3
leads to the in-situ generation of the hexaselenolate ligand, which was subsequently treated with
acetic anhydride to introduce an acetyl protecting group that allows for product isolation and
enhanced air stability.
15
The acetyl protecting group was subsequently removed in the presence of
a base (KOH or NaOAc) to allow for the in-situ formation of the hexaselenolate ligand, which was
then treated with cobalt(II) acetate to generate the corresponding CoTHSe MOF. The second
synthetic approach is adopted directly from the work by Zhu and coworkers,
13
where the Se-alkyl
bond is cleaved in the presence of BBr3 to generate THSeBBr, which was immediately metalated
with cobalt(II) acetate through a solvothermal reaction due to the limited stability of the Se–B
bond. The as-prepared CoTHSe MOF was either worked up and stored under an inert atmosphere
(CoTHSe(AP)), or exposed to air to generate the oxidized MOF (CoTHSe(OX)).
The crystallinity of the MOFs was evaluated by powder X-ray diffraction (PXRD) studies. As
shown in Figure 5.3, the as-prepared CoTHSe by method 2 displays prominent diffraction peaks
at 2θ = 4.3
o
, 8.8
o
, 11.7
o
, and 15.3
o
, along with a broad peak at around 28
o
. The oxidized CoTHSe
152
synthesized by the same method displays significantly diminished peak intensities, with only peaks
at 2θ = 4.3
o
, 8.8
o
, and 27
o
remained observable. MOFs synthesized by method 1 show a poorer
crystallinity, but the peaks at 2θ = 4.3
o
, 8.8
o
, corresponding to the [100] and [200] reflections, are
still prominent, though very broad. Again, the oxidized MOFs are less crystalline compared to the
as-prepared ones. These XRD results are consistent with the those reported previously by Zhu and
coworkers in terms of the overall XRD pattern and peak positions. However, they only reported
on the high crystallinity of the oxidized MOF, without discussing the crystallinity of the as-
prepared MOF. In contrast, our result seems to suggest that oxidation leads to diminished
crystallinity. Future efforts are required to reproduce the synthesis with higher purity precursors
to further improve the crystallinity of the acquired MOFs.
Figure 5.2 Synthetic routes of CoTHSe MOFs
Br Br
Br
Br
Br
Br
LiSe
t
Bu(dioxane)
+
DMI
(1) 70
o
C, 17 hr
(2) CH
3
I
R R
R
R
R
R
R = Se
(1) Na/NH
3
, refluxing
(2) MeOH
(3) THF, Ac
2
O
Method 1
Method 2
BBr
3
, PhF, 70
o
C
Se Se
Se
Se
Se
Se
B
Br
B
Br
B
Br
Se Se
Se
Se
Se
Se
O O
O
O
O
O
(1) KOH or NaOAc, EtOH
(2) Co(OAc)
2
, reflux
Se
Co Se
Se Co
Se
Se
Co
Se
Se
Se
Se
Co
Se
Se
Se
Se Co
Se
Se
Se
Se
Co Se
Se
Se
Se Co
Se
Se
Se
Se
Co Se
Se
Se Se
Se
Se
Co
Se
Se
Se
Se
Co Se
Se
Se
Se
Se
Se Co
Se
Se
Se
Se
Co
Se
Se
Se
CoTHSe
Air
CoTHSe(AP)
CoTHSe(OX)
HBT
THSeBBr
THSe
t
Bu
THSeOAc
153
Figure 5.3 PXRD patterns of CoTHSe synthesized via (a) Method 1 with NaOAc as base; (b) Method 1 with KOH as
base; (c) Method 2.
Preliminary X-ray photoelectron spectroscopy (XPS) measurements were performed on
CoTHSe(AP) and CoTHSe(OX) (synthesized via Method 1 using KOH) to probe the elemental
composition of the materials as well as the chemical environment of each element. The XPS survey
spectra reveal the presence of Co, O, C, and Se (Figure 5.4a and Figure 5.5a), and the high
resolution spectra were collected under each elemental region. For CoTHSe(AP), deconvolution
of the rather complex Co 2p region gives rise to three different sets of peaks (Figure 5.4b), with
binding energies of the 2p 3/2 peaks at 779.5, 781.0, and 787.0 eV. The O 1s region shows a single
peak at 534.1 eV (Figure 5.4c). The presence of oxygen could be contributed by the partial surface
oxidation during sample preparation or the oxygen present in the carbon tape, which serves as the
substrate for XPS sample preparation. The Se 3d region shows a dominant peak at around 55.8 eV
along with a satellite peak at 60.9 eV (Figure 5.4d). Deconvolution of these peaks results in two
sets of peaks due to spin-orbit splitting, with binding energies of the 3d 5/2 peaks at 55.4 and 60.7
eV. The Co 2p spectrum of CoTHSe(OX) is somewhat similar to that of the as-prepared one,
although additional features and/or intensities are observed(Figure 5.5b). Three different sets of
peaks are again present, yet with a slightly lower binding energy of 779.3, 780.8, and 784.8 eV.
Interestingly, the O 1s region for the oxidized MOF is very different in comparison to that of the
as-prepared MOF (Figure 5.5c), where two oxygen environments with binding energies of 530.9
and 532.7 eV are observed. For Se 3d, the main peak is again accompanied by a satellite peak
154
(Figure 5.5d), but the relative intensity of the satellite peak is greater compared to that observed in
the case of CoTHSe(AP). Deconvolution of these peaks suggest a binding energy of 56.0 eV and
58.7 eV for 3d 5/2 peaks. All of the binding energies mentioned above are tableted in Table 5.1,
along with the XPS results reported for the cobalt benzenetetraselenolate (CoBTSe) polymer and
the CoTHSe reported by Zhu and coworkers.
10, 13
Table 5.1 Binding energy (eV) of XPS peaks for CoTHSe in comparison with analogous literature precedents
Peaks AP OX CoBTSe
10
CoTHSe
13
Co
2p 3/2
Peak 1 779.5 779.3 778.5 778.5
Peak 2 781.0 780.8 ~781 (weak) /
Peak 3 787.0 784.8 /
~786.0
(strong)
Se
3d 5/2
Main Peak 55.4 56.0 54.7 ~55.0
Satellite 60.7 58.7 / ~60.0
Figure 5.4 The XPS spectra of CoTHSe(AP) synthesized by Method 1 (KOH as base). (a) Survey spectrum; (b) Co
2p region; (c) O 1s region; (d) Se 3d region
155
Figure 5.5 The XPS spectra of CoTHSe(OX) synthesized by Method 1 (KOH as base). (a) Survey spectrum; (b) Co
2p region; (c) O 1s region; (d) Se 3d region.
The XPS spectra of the MOFs synthesized using Method 2 were also collected. As shown in
Figure 5.6 and Figure 5.7, the general pattern of these spectra are very similar to the XPS results
discussed above, where the Co 2p regions are deconvoluted into three sets of peaks for both the
AP and OX samples, and the Se 4d region shows a drastic change upon oxidation. Better fitting
and analysis of the XPS are required to reveal key differences between the AP and OX samples.
Table 5.2 Binding energy (eV) of XPS peaks for CoTHSe synthesized via Method 2
Peaks AP OX
Co
2p 3/2
Peak 1 779.8 780.0
Peak 2 781.0 783.3
Peak 3 785.6 787.7
Se
3d 5/2
Main Peak 55.3 56.3
Satellite 60.0 58.7
156
Figure 5.6 The XPS spectra of CoTHSe(AP) synthesized by Method 2. (a) Survey spectrum; (b) Co 2p region; (c) O
1s region; (d) Se 3d region.
Figure 5.7 The XPS spectra of CoTHSe(OX) synthesized by Method 2. (a) Survey spectrum; (b) Co 2p region; (c) O
1s region; (d) Se 3d region.
157
5.2.2 Electrochemical Characterizations of CoTHSe(AP)
Electrochemical characterizations of CoTHSe were performed in N2-satuarted pH 1.3 aqueous
electrolytes using a three-electrode cell setup (see section 5.4.5 for electrochemical details). The
polarization curve of CoTHSe(AP) synthesized by Method 2 was acquired by linear sweep
voltammetry (LSV, Figure 5.8), where two electrochemical regions are present, marked by an
onset potential of -0.13 and -0.35 V vs RHE, respectively. The corresponding Tafel slope of these
two regions is 107 mV/dec and 102 mV/dec, extrapolated from the Tafel plot shown in Figure
5.8b. This interesting two-onset behavior has been previously observed for the CoBTSe polymer,
explained by a shift in catalytic mechanism under different applied potentials – at more positive
potentials, protonation followed by reduction to produce H2 is favored (ECCE), whereas at more
negative potentials, reduction followed by protonation is favored (ECEC).
10
In terms of the HER
activity, an overpotential (η) of 430 mV is needed to achieve a current density of 10 mA/cm
2
,
which is around 290 mV larger than the analogous thiolate MOF and 80 mV larger than the
CoBTSe polymer. However, it is to note that the activity presented herein for the CoTHSe(AP) is
not fully optimized.
Figure 5.8 Polarization curve and Tafel plot of CoTHSe(AP) synthesized by Method 2, loading: 15 μL.
To understand the two-onset behavior observed herein, electrochemical impedance
spectroscopy (EIS) was performed at -0.18, -0.28, and -0.38 V (vs RHE), correlating to the two
158
onset regions as well as the plateau region in between them. The crude EIS data is presented in
Figure 5.9. The Nyquist plots under all three potentials display the shape of a depressed semi-
circle, with the one at -0.28 V having a larger radius than the ones at -0.18 V and -0.38 V (Figure
5.9a). Typically for HER catalysts, the radius of Nyquist plot decreases under more negative
potentials, because they provide a larger driving force that leads to a more favored charge-transfer
kinetics. The EIS response observed herein suggests that the electrochemical reactions occurring
at -0.28 V are likely kinetically hindered by a lack of chemical reactants not a lack of electrons.
This also explains the plateau-shaped LSV curve around -0.28 V, which is usually indicative of a
diffusion-limited catalytic process.
16
However, it is noteworthy that at the current stage, the
identity of the electrochemical reaction occurring at the given potential is unclear, so it is more
appropriate to address it as a chemical-limited process rather than a diffusion-limited process. The
EIS data is not properly fitted with an equivalent electrical circuit, as the post-EIS LSV
measurement revealed a drastic change in the polarization curve (Figure 5.10a). This suggests that
the material is dynamically changing during the course of the EIS measurements, so quantitative
analysis of these EIS results is insubstantial.
17
The post-EIS polarization curve has a diminished
current enhancement under the first onset potential, indicating that the corresponding
electrochemical reaction is less favored, which could be due to a depletion of the chemical
reactants as mentioned above. The Tafel slope around the first and second onset also changed to
127 mV/dec and 84.4 mV/dec, compared to the previously measured 107 mV/dec and 102 mV/dec.
The increase in Tafel slope for the first onset aligns well with the diminished current response,
suggesting a larger kinetic hindrance under this potential range. On the other hand, the decreased
Tafel slope for the second onset suggests an enhanced kinetics for the electrochemical process
159
occurring at this potential window, which could be a result of the activation of the electrocatalyst
and requires further investigations.
Figure 5.9 EIS response of CoTHSe(AP) synthesized by method 2, loading: 15 uL.
Figure 5.10 (a) Polarization curves of CoTHSe(AP) collected before and after EIS. (b) CVs with variable scan rates,
only cathodic sweeps are shown for simplicity.
Cyclic voltammograms (CVs) were collected at variable scan rates to further investigate the
two-onset phenomenon. As shown in Figure 5.10b, the first onset shifts to more negative potentials
under higher scan rates, whereas the second onset stays largely unchanged. This further suggests
that the first onset potentially involves a rate-limiting chemical step, whereas the second onset
does not. Future studies can focus on performing similar variable scan rate studies in higher pH
electrolyte solutions to gain further evidence.
The loading of the catalyst was also varied to evaluate its impact on the electrochemical profile.
Catalyst loading can be easily controlled by varying the volume of the catalyst ink drop casted
during electrode preparation (see section 5.4.5 for electrochemical details). Here, electrodes with
160
three different loadings, 5 μL, 10 μL, and 15 μL, were prepared and analyzed. Double-layer
capacitance (Cdl) measurements were first performed with results summarized in Figure 5.11. The
Cdl value is used to represent the electrocatalytically active surface area (ECSA), which is indictive
of the actual loading of the catalyst.
18
It was found that higher volume loading of the catalyst
indeed results in a higher ECSA. The Cdl value of the 5 μL, 10 μL, and 15 μL electrode are 0.85
mF, 1.43 mF, and 2.33 mF, respectively. Polarization curve of each electrode was then acquired
by LSV (Figure 5.12), where a diminished current enhancement under the first onset was observed
for electrodes with a lower catalyst loading. The Tafel slopes derived from the polarization curves
are summarized in Table 5.3. The 0.85 mF loading electrode has the largest Tafel slope around the
first onset region, which is consistent with its minimal current enhancement under this potential
window. This further suggests that the electrochemical processes occurring around the first onset
potential involves the participation of chemical moieties within the MOF structure. It is also
interesting that the Tafel slopes around the second onset potential range between 70 to 90 mV/dec,
which is quite low for MOF-based electrocatalysts. However, further analyses are necessary to
attribute the observed electrochemical processes to hydrogen production.
161
Figure 5.11 Cdl measurements of electrodes deposited with variable amounts of CoTHSe(AP) ink.
Figure 5.12 Polarization curves of CoTHSe(AP) with variable loadings
Table 5.3 Tafel slopes of electrodes prepared with different catalyst loadings
Loading (mF)
Tafel slope near first onset
(mV/dec)
Tafel slope near second onset
(mV/dec)
0.85 217 73.9
1.43 139 92.9
2.33 127 84.4
162
5.2.3 Electrochemical Characterizations of CoTHSe(OX)
Preliminary electrochemical characterizations were also performed for CoTHSe(OX)
synthesized via Method 1 using KOH as the base for the deprotection reaction. As shown in Figure
5.13, CoTHSe(OX) displays only one catalytic onset at -0.35 V vs RHE, which is very close to
the second onset displayed by CoTHSe(AP). The 𝜂
'5 67/06
- for the two electrodes are both
around 440 mV, but it is important to note that the Cdl value determined for the CoTHSe(OX) is
only 0.25 mF, much lower than the 1.45 mF for CoTHSe(AP). If we normalize the current by Cdl
(Figure 5.14), it can be seen that the oxidized MOF displays a much higher current density than
the as-prepared MOF, suggesting the greater intrinsic activity of each catalytically active site.
Additionally, Tafel slope for the CoTHSe(OX) is only 48.5 mV/dec, which is significantly lower
than that of the CoTHSe(AP) as discussed above. Such low Tafel slope suggests a favorable
charger-transfer kinetics of the electrochemical process, in line with the high intrinsic activity
revealed by the polarization curve. The EIS spectra of CoTHSe(OX) also shows a much smaller
radius for the semicircle, indicating a smaller charge-transfer resistance (Figure 5.15). The
drastically different electrochemical profiles of the as-prepared and oxidized MOFs suggest that
the oxidation state of the material can indeed influence the electrochemical properties and possibly
the HER mechanism of the catalyst, which will require future investigations.
Figure 5.13 Polarization curve of CoTHSe(OX) in comparison with CoTHSe(AP) and the corresponding Tafel plot
163
Figure 5.14 Polarization curves of CoTHSe(OX) and CoTHSe(AP) with current normalized by Cdl
Figure 5.15 Nyquist plots of CoTHSe(OX) and CoTHSe(AP) collected at -0.48 V vs RHE
5.3 Conclusions and Future Direction
In conclusion, preliminary work presented herein illustrates the successful synthesis of the
CoTHSe MOF and its capability of performing electrocatalytic HER under fully aqueous
conditions. Future work should focus on the following aspects:
First, further synthetic optimizations should be performed to acquire materials with higher
crystallinity. Such optimization could likely be achieved by purifying reaction intermediates,
modifying the synthetic conditions (scales, temperatures, solvents) of the synthesis, especially
those of the final metalation step. Following the synthesis, comprehensive physical
characterizations such as XRD (synchrotron), XPS, SEM, EDX, and BET should be performed to
fully analyze the materials, and compare the differences between the as-prepared and oxidized
164
MOF. Detailed analysis of XPS might be crucial to elucidate the difference in oxidation states for
these two materials, which might require collaboration with researchers that have such expertise.
Next, several key electrochemical experiments need to be performed to fully understand the
electrochemical behavior of the materials. First, CV studies under more basic conditions (e.g. pH
4, 7, 10) can be performed to evaluate the redox features of CoTHSe(AP) and CoTHSe(OX), the
comparison of which can provide insights into how oxidation states influences the electrochemical
profile of these catalysts. Controlled potential electrolysis (CPE) should be done under different
potentials for CoTHSe(AP), to evaluate the stability of the catalyst and the products generated at
the two different electrochemical regions. If the first onset is responsible for hydrogen production,
future studies can explore the mechanism of the corresponding electrochemical process and
develop strategies to promote such process due to its low overpotential requirements. Detailed
physical characterizations should also be conducted following the CPE to understand the
transitions of the catalysts during catalysis. Some other electrochemical experiments can also be
performed, such as the extraction of kinetic Tafel slope through EIS studies,
19
variable pH studies,
variable rotation rate studies, kinetic isotope studies, and CV cycling studies to further evaluate
the stability of the catalyst.
The preliminary electrochemical characterizations of CoTHSe suggest that the activity of this
MOF is inferior to the analogous thiolate MOF (CoTHT) as it requires a larger overpotential to
reach 10 mA/cm
2
current density. This is not consistent with what was observed for the analogous
1D polymer, where CoBTSe is more active than CoBTT.
10
Future work could exploit
computational tools to understand the origin of the different activity observed herein. However,
the Tafel slopes of CoTHSe is actually lower than that of the CoTHT in some cases, suggesting
possible difference in mechanistic pathways between the two MOFs. Therefore, it would also be
165
interesting to explore the mechanisms for the selenium-containing MOFs using both experimental
and computational tools. Previous work with the 1D CoBTSe polymer has suggested the
involvement of the Se–H species in the catalytic cycle, whereas for the CoTHT MOF,
computational studies suggest only cobalt center being responsible for catalysis. As Se is known
to be more readily protonated, it would not be surprising if CoTHSe adopts a totally different HER
mechanism than the original sulfur analogue.
5.4 Experimental Details
5.4.1 Synthesis of THSe
t
Bu
The synthesis of THSe
t
Bu is adopted from the reported procedures by Tuner and Vaid, with
the reaction route presented in Figure 5.16.
15
The reaction precursors, hexabromotriphenylene
(HBT) and LiSe
t
Bu(dioxane), were synthesized according to prior literature reports and their
respective
1
H-NMR spectrum is shown in Figure 5.17 and Figure 5.18.
15,20
The synthesis of
LiSe
t
Bu(dioxane) calls for the use of
t
BuLi, which is a highly flammable pyrophoric substance.
The researcher MUST be properly trained in order to handle the chemical safely.
Under a N2 atmosphere, HBT (0.74 g, 1 mmol, 1.0 eq.), LiSe
t
Bu(dioxane) (2.2 g, 9.5 mmol,
9.5 eq.), and 1,3-dimethyl-2-imidazolidinone (DMI, 60 mL) were packed in a Schlenk flask, and
the reaction mixture was stirred overnight under room temperature(~ 18 h). During the course of
the reaction, a drastic color change was observed, where the original orange color became much
brighter. The reaction mixture was a suspension during the whole period of the reaction, with more
white precipitate formed as the reaction proceeded, hence vigorous stirring must be ensured to
avoid congelation. The reaction mixture was then cooled to 0
o
C to help the precipitation of the
product. The solid product was subsequently filtered off and washed with a copious amount of
water, and a small amount of cold MeOH and Et2O. The final product was a white powder (crude
166
yield: 70%), which was analyzed by
1
H NMR spectroscopy:
1
H NMR (500 MHz, CDCl3, Figure
5.19), δ 8.94 (s, 1H), δ 1.55 (s, 9H);
77
Se NMR (500 MHz, CDCl3, Figure 5.20), δ 528.05;
13
C
NMR (500 MHz, CDCl3, Figure 5.21), δ 137.49, 132.13, 128.73, 45.68, 32.22.
Notes: (1) This reaction works better in a smaller scale (100 mg HBT, 300 mg LiSe
t
Bu, and 8 mL
DMI); (2) Heating the reaction leads to side products, room temperature works the best; (3) DMI
is chosen based on past experience with the synthesis of THT, the use of DMF could be attempted
as well in the future. It is difficult to fully remove DMI from the product, make sure to wash with
a large amount of water (DMI
1
H-NMR peaks: δ 3.27, 2.78); (4) THSe
t
Bu is somewhat air stable,
so the workup was performed in air, but the product should be stored under an inert atmosphere
right after the workup is done. Recrystallization with alcohols has been attempted, but did not
result in purer products, which could be caused by the limited air stability of the product.
Figure 5.16 The synthetic route of THSe
t
Bu
Se
1)
t
BuLi/THF
2) Dioxane
LiSe
t
Bu(dioxane)
HBT/DMI
70
o
C, 110
o
C
R R
R
R
R
R
R = Se
167
Figure 5.17
1
H NMR spectrum of LiSe
t
Bu(dioxane) (500 MHz, CD3CN)
Figure 5.18
1
H NMR spectrum of HBT (500 MHz, CDCl3), HBT is poorly soluble in common NMR solvents.
168
Figure 5.19
1
H NMR spectrum of THSe
t
Bu (500 MHz, CDCl3)
Figure 5.20
77
Se NMR spectrum of THSe
t
Bu (500 MHz, CDCl3)
169
Figure 5.21
13
C NMR spectrum of THSe
t
Bu (500 MHz, CDCl3)
5.4.2 Synthesis of CoTHSe via Method 1
Draw from our previous experience on the synthesis of CoTHT, the synthesis of CoTHSe was
first attempted using a Birch reduction reaction (see reaction scheme in Figure 5.22). Under a N2
atmosphere, around 25 mL of liquid NH3 was condensed in a 3-neck flask under -78
o
C (dry
ice/IPA bath). Next, Na metal chunks (0.38 g) were added into liquid NH3 to form a dark blue
solution, followed by the addition of THSe
t
Bu (0.43 g). Dry ice/IPA bath was then added to the
cold finger condenser, and the cold bath around the 3-neck flask was subsequently removed to
enable the reflux of liquid NH3. After 4 h, degassed MeOH (5 mL) was added to the reaction vessel
to quench the unreacted excess Na, leaving behind a green/yellow colored suspension. The reaction
mixture was first allowed to warm up to room temperature until most of the NH3 was boiled off,
then the residue MeOH was removed under vacuum, resulting in a grey colored solid residue.
170
The second part of this synthesis is to install an acetyl protecting group to improve the stability
of the hexaselenolate ligand.
15
10 mL THF was added to the above reaction vessel and the mixture
was stirred for 15 mins, after which all the volatiles were removed under vacuum. Another 10 mL
of fresh THF was added to suspend the solid, followed by the addition of 2 mL acetic anhydride.
The reaction was allowed to stir for 1 h, during which the suspended solid was partially dissolved,
and the solvent was then removed under vacuum. Water (25 mL) was added, and the product was
extracted with dichloromethane (4 × 25 mL). The combined organic layers (yellow in color) were
washed with water, dried over Na2SO4, and filtered. The solvent was removed under vacuum
leaving an off-white solid (THSeAc). The crude product was characterized using
1
H NMR
spectroscopy (Figure 5.23): δ 8.90 (s, 1H), δ 2.54 (s, 3H). While some prominent solvent and
impurity peaks are present (e.g. the DMI
1
H-NMR peaks at δ 3.27, 2.78), several diagnostic peaks
suggest the completion of the intended reaction: the shift of the aromatic proton peak from δ 8.94
to δ 8.90, the disappearance of the alkyl proton peak at δ 1.55, and the appearance of the acetyl
proton peak at δ 2.54. However, future effort should be devoted to further purifying the acquired
product.
To generate the CoTHSe MOF, THSeAc was suspended in degassed EtOH. Under vigorous
stirring, an excess amount of base (KOH or NaOAc) was added to the suspension, and allowed to
stir for 10 min, during which the solid fully dissolved to give a yellow clear solution. Next,
Co(OAc)2·4H2O was added to the reaction vessel, upon which solid precipitated out and the color
of the mixture changed from yellow to greenish yellow and eventually dark green. The reaction
was heated to reflux for 24 h, and then cooled to room temperature. Half of the product suspension
was transferred to a N2-filled Schlenk flask using a syringe, and the volatiles were then removed
under vacuum and the flask was brought into the glove box for air-free workup. The second half
171
of the reaction was exposed to air and stirred for 12 h to ensure complete oxidation of the MOF.
Both of the oxidized and as-prepared CoTHSe were washed with water, methanol,
dichloromethane, and diethyl ether thoroughly, and then dried under vacuum to yield a very dark
green colored solid.
Notes: (1) Refluxing liquid NH3 is required to yield the target product. (2) Due to the harsh
condition of Birch reduction, a glass stir bar should be used instead of a regular Teflon-coated stir
bar, to avoid the disintegration of the Teflon coat. (3) Future attempts could use Li instead of Na,
Ar instead of N2, to match the literature reported procedures. If Li is used, make sure to switch to
Ar atmosphere, as Li reacts with N2. If Ar is used, make sure to switch to a N2 atmosphere when
removing volatiles under vacuum, Ar should NOT be condensed!
14
Figure 5.22 The synthetic route of CoTHSe via Method 1
R R
R
R
R
R
R = Se
(1) Na/NH
3
, refluxing
(2) MeOH
(3) THF, Ac
2
O
’R R’
’R
’R
R’
R’
R’ =
O
Se
(1) Base, EtOH
(2) Co(OAc)
2
, reflux
CoTHSe MOF
172
Figure 5.23
1
H NMR spectrum of THSeAc (500 MHz, CDCl3)
5.4.3 Synthesis of CoTHSe via Method 2
The second synthetic route of CoTHSe is directly adopted from the previous work by Zhu and
coworkers.
13
Under an Ar atmosphere, THSe
t
Bu (0.21 g, 1 eq.) was dissolved in fluorobenzene
(12 ml) in a 3-neck flask equipped with a condenser. Then BBr3 (0.31 g, 0.12 mL, 6 eq.) was added
via a syringe. The reaction mixture was heated at 70 ℃ for 12 h under vigorous stirring. During
which, a lightly yellow precipitate formed, and the reaction mixture turned into darker orange
overnight. Then the reaction was cooled to room temperature, and the stirring was turned off to
allow the product to settle. The dark orange reaction solvent was then cannula transferred out of
the reaction vessel, leaving behind the light yellow solid product (THSeBBr), which was washed
with air-free pentane several times via cannula transfer. Due to the limited stability of the
THSeBBr, it was not further isolated and purified and was directly used for the MOF synthesis.
173
25 mL of degassed EtOH was added to the above reaction vessel to form a suspension. Solid
KOH (0.17 g, 15 eq.) was then added under vigorous stirring, upon which the solids initially
dissolved partially giving a greenish colored suspension, but a new precipitate formed later on
leaving a light yellow colored suspension. After 10 mins of stirring, Co(OAc)2·4H2O was added
and the reaction was heated to reflux for 24 h, during which the color of the reaction turned brown.
The workup of the resultant MOF was the same as described for method 1, where the MOF was
split into two halves, with one kept under air-free condition, and the other one oxidized. Both as-
prepared and oxidized MOFs have a dark green color, but the oxidized one is a little darker.
Notes: (1) BBr3 is a fuming liquid compound, which promptly reacts with water to release HBr.
Make sure to order a sure-seal bottle of BBr3, and fully quench the needle/syringe used for
delivering the compound with water after use.
5.4.4 Physical Characterizations
X-ray photoelectron spectroscopy (XPS) data were collected using a Kratos AXIS Ultra
instrument. The monochromatic X-ray 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
"+
Torr. The XPS data were analyzed using the CasaXPS software.
Powder X-ray diffraction (PXRD) was performed on a Rigaku Ultima IV X-Ray diffractometer
in reflectance parallel beam/parallel slit alignment geometry. The measurement employed Cu Kα
line focused radiation at 1760 W (40 kV, 44 mA) power and a Ge crystal detector fitted with a 0.6
mm radiation entrance slit. Samples were mounted on zero-background sample holders and were
observed using a 0.01° 2θ step scan from 2.0 – 40.0° with a scan rate of 1°/min.
174
5.4.5 Electrochemistry Methods
Electrochemistry experiments were carried out using a VersaSTAT 3 potentiostat in a three-
electrode configuration electrochemical cell under an inert (N2) atmosphere. A rotating disk
electrode (RDE, with glassy carbon insert, 0.196 cm
2
surface area) was used as the working
electrode. The rotation rate of the RDE is set to 1600 rpm, if not otherwise stated. The glassy
carbon electrode was polished with 0.05 µm Al2O3 polish powder and sonicated in Millipore water
for 10 minutes prior to use. A graphite rod, purchased from Graphite Machining, Inc. (Grade NAC-
500 Purified, < 10 ppm ash level), was used as the counter electrode. The reference electrode,
placed in a separate compartment and connected by a porous Teflon tip, was based on an aqueous
Ag/AgCl/1 M KCl electrode, directly purchased from CH Instruments, Inc. All potentials reported
in this paper were converted to the reversible hydrogen electrode (RHE) by adding a value of
(0.235 + 0.059 × pH) V.
The catalyst was deposited as a suspension composed of 2 mg of the MOF, 20 µL of Nafion
solution (0.5 wt%, purchased from Sigma-Aldrich
â
), 45 µL of water, and 135 µL of ethanol. The
mixture was sonicated for 20 mins to form a uniformly dispersed suspension A desired amount of
this suspension was then drop casted onto a freshly polished glassy carbon electrode using a micro
syringe, and dried at room temperature prior to use. The pH 1.3 aqueous electrolyte solutions was
prepared by adding 0.534 mL of 18.7 M H2SO4 to 200 mL 0.1 M NaClO4. The pH of the solutions
was measured with a benchtop Mettler Toledo pH meter. Prior to each electrochemical experiment,
the electrolyte solution was purged with nitrogen thoroughly to avoid the interference of oxygen
reduction reaction.
175
Electrochemical impedance spectroscopy (EIS) measurements were carried out at different
overpotentials in the frequency range of 100 kHz – 0.1 Hz with 10 mV sinusoidal perturbations.
Experimental EIS data were analyzed and fitted with the ZSimpWin software.
The obtained polarization curves were corrected for iR loss according to the following equation:
Ecorr = Emea – iRs
Where Ecorr is the iR-corrected potential, Emea is the experimentally measured potential, and Rs
is the solution resistance extracted from the fitted EIS data.
Controlled potential electrolysis (CPE) measurements were conducted in a sealed two-
chambered H-cell where the first chamber held the working and reference electrodes in 40 mL of
electrolyte solution and the second chamber held the counter electrode in 20 mL of electrolyte
solution. The two chambers were both under N2 and separated by a fine porosity glass frit. CPE
experiments were performed with a glassy carbon plate electrode (6 cm × 1 cm × 0.3 cm; Tokai
Carbon USA) as the working electrode and a graphite rod as the counter electrode. The reference
electrode was a Ag/AgCl/saturated 1 M KCl (aq) electrode separated from the solution by a Vycor
tip. Using a gas-tight syringe, 2 mL of gas was 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. To determine the Faradaic efficiency,
the theoretical H2 amount based on total charge flowed was compared with the GC-detected H2
produced from controlled-potential electrolysis.
5.5 References
(1) Lubitz, W.; Ogata, H.; Rüdiger, O.; Reijerse, E. Hydrogenases. Chem. Rev. 2014, 114 (8),
4081–4148.
(2) Schilter, D.; Camara, J. M.; Huynh, M. T.; Hammes-Schiffer, S.; Rauchfuss, T. B.
Hydrogenase Enzymes and Their Synthetic Models: The Role of Metal Hydrides. Chem. Rev. 2016,
116 (15), 8693–8749.
176
(3) Wombwell, C.; Caputo, C. A.; Reisner, E. [NiFeSe]-Hydrogenase Chemistry. Acc. Chem.
Res. 2015, 48 (11), 2858–2865.
(4) Valente, F. M. A.; Oliveira, A. S. F.; Gnadt, N.; Pacheco, I.; Coelho, A. V.; Xavier, A. V.;
Teixeira, M.; Soares, C. M.; Pereira, I. A. C. Hydrogenases in Desulfovibrio Vulgaris
Hildenborough: Structural and Physiologic Characterisation of the Membrane-Bound [NiFeSe]
Hydrogenase. J. Biol. Inorg. Chem. 2005, 10 (6), 667–682.
(5) Kong, D.; Cha, J. J.; Wang, H.; Lee, H. R.; Cui, Y. First-Row Transition Metal
Dichalcogenide Catalysts for Hydrogen Evolution Reaction. Energy Environ. Sci. 2013, 6 (12),
3553–3558.
(6) Kong, D.; Wang, H.; Lu, Z.; Cui, Y. CoSe2 Nanoparticles Grown on Carbon Fiber Paper:
An Efficient and Stable Electrocatalyst for Hydrogen Evolution Reaction. J. Am. Chem. Soc. 2014,
136 (13), 4897–4900.
(7) Wang, H.; Kong, D.; Johanes, P.; Cha, J. J.; Zheng, G.; Yan, K.; Liu, N.; Cui, Y. MoSe2
and WSe2 Nanofilms with Vertically Aligned Molecular Layers on Curved and Rough Surfaces.
Nano Lett. 2013, 13 (7), 3426–3433.
(8) Henckel, D. A.; Lenz, O. M.; Krishnan, K. M.; Cossairt, B. M. Improved HER Catalysis
through Facile, Aqueous Electrochemical Activation of Nanoscale WSe2. Nano Lett. 2018, 18 (4),
2329–2335.
(9) 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 (53), 7306–7309.
(10) Downes, C. A.; Marinescu, S. C. Bioinspired Metal Selenolate Polymers with Tunable
Mechanistic Pathways for Efficient H2 Evolution. ACS Catal. 2017, 7 (1), 848–854.
(11) Anantharaj, S.; Ede, S. R.; Sakthikumar, K.; Karthick, K.; Mishra, S.; Kundu, S. Recent
Trends and Perspectives in Electrochemical Water Splitting with an Emphasis on Sulfide, Selenide,
and Phosphide Catalysts of Fe, Co, and Ni: A Review. ACS Catal. 2016, 6 (12), 8069–8097.
(12) Chen, K.; Downes, C. A.; Schneider, E.; Goodpaster, J. D.; Marinescu, S. C. Improving
and Understanding the Hydrogen Evolving Activity of a Cobalt Dithiolene Metal–Organic
Framework. ACS Appl. Mater. Interfaces 2021, 13 (14), 16384–16395.
(13) Cui, Y.; Yan, J.; Chen, Z.; Xing, W.; Ye, C.; Li, X.; Zou, Y.; Sun, Y.; Liu, C.; Xu, W.; et
al. Synthetic Route to a Triphenylenehexaselenol-Based Metal Organic Framework with Semi-
Conductive and Glassy Magnetic Properties. iScience 2020, 23 (1), 100812.
(14) Sakamoto, R.; Kambe, T.; Tsukada, S.; Takada, K.; Hoshiko, K.; Kitagawa, Y.; Okumura,
M.; Nishihara, H. Π-Conjugated Trinuclear Group-9 Metalladithiolenes With a Triphenylene
Backbone. Inorg. Chem. 2013, 52 (13), 7411–7416.
(15) Turner, D. L.; Vaid, T. P. Synthesis of Protected Benzenepolyselenols. J. Org. Chem. 2012,
177
77 (20), 9397–9400.
(16) Costentin, C.; Saveant, J.-M. Cyclic Voltammetry Analysis of Electrocatalytic Films. J.
Phys. Chem. C 2015, 119 (22), 12174–12182.
(17) Anantharaj, S.; Noda, S. Appropriate Use of Electrochemical Impedance Spectroscopy in
Water Splitting Electrocatalysis. ChemElectroChem 2020, 7 (10), 2297–2308.
(18) Yoon, Y.; Yan, B.; Surendranath, Y. Suppressing Ion Transfer Enables Versatile
Measurements of Electrochemical Surface Area for Intrinsic Activity Comparisons. J. Am. Chem.
Soc. 2018, 140 (7), 2397–2400.
(19) Downes, C. A.; Clough, A. J.; Chen, K.; Yoo, J. W.; Marinescu, S. C. Evaluation of the H2
Evolving Activity of Benzenehexathiolate Coordination Frameworks and the Effect of Film
Thickness on H2 Production. ACS Appl. Mater. Interfaces 2018, 10 (2), 1719–1727.
(20) Breslow, R.; Jaun, B.; Kluttz, R. Q.; Xia, C. Ground State Pi-Electron Triplet Molecules
of Potential Use in the Synthesis of Organic Ferromagnets. Tetrahedron 1982, 38 (6), 863–867.
178
Bibliography
(1) NASA. A Degree of Concern: Why Global Temperatures Matter – Climate Change: Vital
Signs of the Planet. NASA’s Global Climate Change Website. 2019.
(2) US Energy Protection Agency. Global Greenhouse Gas Emissions Data | Greenhouse Gas
(GHG) Emissions | US EPA. United States Environmental Protection Agency. 2014.
(3) IPCC, I. G. P. for C. C. Global Warming of 1.5°C. An IPCC Special Report on the Impacts
of Global Warming of 1.5°C above Pre-Industrial Levels and Related Global Greenhouse Gas
Emission Pathways, in the Context of Strengthening the Global Response to the Threat of Climate
Change; 2018.
(4) Agora Energiewende and Ember. The European Power Sector in 2020: Up-to-Date
Analysis on the Electricity Transistion; 2021.
(5) IEA (International Energy Agency). Renewables 2020; 2020.
(6) REN21. Renewables 2020 Global Status Report; 2020.
(7) She, Z. W.; Kibsgaard, J.; Dickens, C. F.; Chorkendorff, I.; Nørskov, J. K.; Jaramillo, T. F.
Combining Theory and Experiment in Electrocatalysis: Insights into Materials Design. Science
2017, 355 (6321), eaad4998.
(8) LeValley, T. L.; Richard, A. R.; Fan, M. The Progress in Water Gas Shift and Steam
Reforming Hydrogen Production Technologies – A Review. Int. J. Hydrogen Energy 2014, 39
(30), 16983–17000.
(9) Santos, D. M. F.; Sequeira, C. A. C.; Figueiredo, J. L. Hydrogen Production by Alkaline
Water Electrolysis. Quim. Nova 2013, 36 (8), 1176–1193.
(10) U.S. Department of Energy. Hydrogen Production: Electrolysis
https://www.energy.gov/eere/fuelcells/hydrogen-production-electrolysis.
(11) McKone, J. R.; Marinescu, S. C.; Brunschwig, B. S.; Winkler, J. R.; Gray, H. B. Earth-
Abundant Hydrogen Evolution Electrocatalysts. Chem. Sci. 2014, 5 (3), 865–878.
(12) Bullock, R. M.; Das, A. K.; Appel, A. M. Surface Immobilization of Molecular
Electrocatalysts for Energy Conversion. Chem. Eur. J. 2017, 23, 7626–7641.
(13) Stassen, I.; Burtch, N.; Talin, A.; Falcaro, P.; Allendorf, M.; Ameloot, R. An Updated
Roadmap for the Integration of Metal–Organic Frameworks with Electronic Devices and Chemical
Sensors. Chem. Soc. Rev. 2017, 46 (11), 3185–3241.
(14) Morozan, A.; Jaouen, F. Metal Organic Frameworks for Electrochemical Applications.
Energy Environ. Sci. 2012, 5 (11), 9269.
(15) Sun, L.; Campbell, M. G.; Dincă, M. Electrically Conductive Porous Metal-Organic
179
Frameworks. Angew. Chemie Int. Ed. 2016, 55 (11), 3566–3579.
(16) Kusamoto, T.; Nishihara, H. Zero-, One- and Two-Dimensional Bis(Dithiolato)Metal
Complexes with Unique Physical and Chemical Properties. Coord. Chem. Rev. 2019, 380, 419–
439.
(17) Sheberla, D.; Sun, L.; Blood-Forsythe, M. A.; Er, S.; Wade, C. R.; Brozek, C. K.; Aspuru-
Guzik, A.; Dincă, M. High Electrical Conductivity in Ni3(2,3,6,7,10,11-Hexaiminotriphenylene)2,
a Semiconducting Metal–Organic Graphene Analogue. J. Am. Chem. Soc. 2014, 136 (25), 8859–
8862.
(18) Aubrey, M. L.; Kapelewski, M. T.; Melville, J. F.; Oktawiec, J.; Presti, D.; Gagliardi, L.;
Long, J. R. Chemiresistive Detection of Gaseous Hydrocarbons and Interrogation of Charge
Transport in Cu[Ni(2,3-Pyrazinedithiolate)2] by Gas Adsorption. J. Am. Chem. Soc. 2019, 141 (12),
5005–5013.
(19) Clough, A. J.; Skelton, J. M.; Downes, C. A.; de la Rosa, A. A.; Yoo, J. W.; Walsh, A.;
Melot, B. C.; Marinescu, S. C. Metallic Conductivity in a Two-Dimensional Cobalt Dithiolene
Metal–Organic Framework. J. Am. Chem. Soc. 2017, 139 (31), 10863–10867.
(20) Lin, S.; Usov, P. M.; Morris, A. J. The Role of Redox Hopping in Metal–Organic
Framework Electrocatalysis. Chem. Commun. 2018, 54 (51), 6965–6974.
(21) Das, A.; Han, Z.; Haghighi, M. G.; Eisenberg, R. Photogeneration of Hydrogen from Water
Using CdSe Nanocrystals Demonstrating the Importance of Surface Exchange. Proc. Natl. Acad.
Sci. 2013, 110 (42), 16716–16723.
(22) McNamara, W. R.; Han, Z.; Alperin, P. J.; Brennessel, W. W.; Holland, P. L.; Eisenberg,
R. A Cobalt–Dithiolene Complex for the Photocatalytic and Electrocatalytic Reduction of Protons.
J. Am. Chem. Soc. 2011, 133 (39), 15368–15371.
(23) Lv, H.; Ruberu, T. P. A.; Fleischauer, V. E.; Brennessel, W. W.; Neidig, M. L.; Eisenberg,
R. Catalytic Light-Driven Generation of Hydrogen from Water by Iron Dithiolene Complexes. J.
Am. Chem. Soc. 2016, 138 (36), 11654–11663.
(24) Zarkadoulas, A.; Field, M. J.; Artero, V.; Mitsopoulou, C. A. Proton-Reduction Reaction
Catalyzed by Homoleptic Nickel-Bis-1,2-Dithiolate Complexes: Experimental and Theoretical
Mechanistic Investigations. ChemCatChem 2017, 9 (12), 2308–2317.
(25) Ray, K.; Begum, A.; Weyhermüller, T.; Piligkos, S.; van Slageren, J.; Neese, F.; Wieghardt,
K. The Electronic Structure of the Isoelectronic, Square-Planar Complexes [Fe
II
(L)2]
2-
and
[Co
III
(LBu)2]
-
(L
2-
and (LBu)
2-
= Benzene-1,2-Dithiolates): An Experimental and Density
Functional Theoretical Study. J. Am. Chem. Soc. 2005, 127 (12), 4403–4415.
(26) Downes, C. A.; Marinescu, S. C. Efficient Electrochemical and Photoelectrochemical H2
Production from Water by a Cobalt Dithiolene One-Dimensional Metal–Organic Surface. J. Am.
Chem. Soc. 2015, 137 (43), 13740–13743.
180
(27) Clough, A. J.; Yoo, J. W.; Mecklenburg, M. H.; Marinescu, S. C. Two-Dimensional Metal–
Organic Surfaces for Efficient Hydrogen Evolution from Water. J. Am. Chem. Soc. 2015, 137 (1),
118–121.
(28) Dong, R.; Pfeffermann, M.; Liang, H.; Zheng, Z.; Zhu, X.; Zhang, J.; Feng, X. Large-Area,
Free-Standing, Two-Dimensional Supramolecular Polymer Single-Layer Sheets for Highly
Efficient Electrocatalytic Hydrogen Evolution. Angew. Chem., Int. Ed. 2015, 54 (41), 12058–
12063.
(29) Dong, R.; Zheng, Z.; Tranca, D. C.; Zhang, J.; Chandrasekhar, N.; Liu, S.; Zhuang, X.;
Seifert, G.; Feng, X. Immobilizing Molecular Metal Dithiolene-Diamine Complexes on 2D Metal-
Organic Frameworks for Electrocatalytic H2 Production. Chem. Eur. J. 2017, 23 (10), 2255–2260.
(30) Downes, C. A.; Marinescu, S. C. Bioinspired Metal Selenolate Polymers with Tunable
Mechanistic Pathways for Efficient H 2 Evolution. ACS Catal. 2017, 7 (1), 848–854.
(31) 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 (53), 7306–7309.
(32) McCrory, C. C. L.; Jung, S.; Ferrer, I. M.; Chatman, S. M.; Peters, J. C.; Jaramillo, T. F.
Benchmarking Hydrogen Evolving Reaction and Oxygen Evolving Reaction Electrocatalysts for
Solar Water Splitting Devices. J. Am. Chem. Soc. 2015, 137 (13), 4347–4357.
(33) Shinagawa, T.; Garcia-Esparza, A. T.; Takanabe, K. Insight on Tafel Slopes from a
Microkinetic Analysis of Aqueous Electrocatalysis for Energy Conversion. Sci. Rep. 2015, 5 (1),
13801.
(34) Chu, S.; Majumdar, A. Opportunities and Challenges for a Sustainable Energy Future.
Nature 2012, 488 (7411), 294–303.
(35) 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.
(36) Roger, I.; Shipman, M. A.; Symes, M. D. Earth-Abundant Catalysts for Electrochemical
and Photoelectrochemical Water Splitting. Nat. Rev. Chem. 2017, 1 (1), 0003.
(37) Queyriaux, N.; Kaeffer, N.; Morozan, A.; Chavarot-Kerlidou, M.; Artero, V. Molecular
Cathode and Photocathode Materials for Hydrogen Evolution in Photoelectrochemical Devices. J.
Photochem. Photobiol. C Photochem. Rev. 2015, 25, 90–105.
(38) Thoi, V. S.; Sun, Y.; Long, J. R.; Chang, C. J. Complexes of Earth-Abundant Metals for
Catalytic Electrochemical Hydrogen Generation under Aqueous Conditions. Chem. Soc. Rev. 2013,
42 (6), 2388–2400.
(39) Downes, C. A.; Marinescu, S. C. Electrocatalytic Metal–Organic Frameworks for Energy
Applications. ChemSusChem 2017, 10 (22), 4374–4392.
181
(40) Solomon, M. B.; Church, T. L.; D’Alessandro, D. M. Perspectives on Metal–Organic
Frameworks with Intrinsic Electrocatalytic Activity. CrystEngComm 2017, 19 (29), 4049–4065.
(41) Maeda, H.; Sakamoto, R.; Nishihara, H. Coordination Programming of Two-Dimensional
Metal Complex Frameworks. Langmuir 2016, 32 (11), 2527–2538.
(42) Sakamoto, R.; Takada, K.; Pal, T.; Maeda, H.; Kambe, T.; Nishihara, H. Coordination
Nanosheets (CONASHs): Strategies, Structures and Functions. Chem. Commun. 2017, 53 (43),
5781–5801.
(43) Ko, M.; Mendecki, L.; Mirica, K. A. Conductive Two-Dimensional Metal–Organic
Frameworks as Multifunctional Materials. Chem. Commun. 2018, 54 (57), 7873–7891.
(44) Clough, A. J.; Yoo, J. W.; Mecklenburg, M. H.; Marinescu, S. C. Two-Dimensional Metal–
Organic Surfaces for Efficient Hydrogen Evolution from Water. J. Am. Chem. Soc. 2015, 137 (1),
118–121.
(45) Lu, X. F.; Liao, P. Q.; Wang, J. W.; Wu, J. X.; Chen, X. W.; He, C. T.; Zhang, J. P.; Li, G.
R.; Chen, X. M. An Alkaline-Stable, Metal Hydroxide Mimicking Metal-Organic Framework for
Efficient Electrocatalytic Oxygen Evolution. J. Am. Chem. Soc. 2016, 138 (27), 8336–8339.
(46) Wang, L.; Wu, Y.; Cao, R.; Ren, L.; Chen, M.; Feng, X.; Zhou, J.; Wang, B. Fe/Ni Metal-
Organic Frameworks and Their Binder-Free Thin Films for Efficient Oxygen Evolution with Low
Overpotential. ACS Appl. Mater. Interfaces 2016, 8 (26), 16736–16743.
(47) Zhao, S.; Wang, Y.; Dong, J.; He, C. T.; Yin, H.; An, P.; Zhao, K.; Zhang, X.; Gao, C.;
Zhang, L.; et al. Ultrathin Metal-Organic Framework Nanosheets for Electrocatalytic Oxygen
Evolution. Nat. Energy 2016, 1 (12), 16184.
(48) Duan, J.; Chen, S.; Zhao, C. Ultrathin Metal-Organic Framework Array for Efficient
Electrocatalytic Water Splitting. Nat. Commun. 2017, 8, 15341.
(49) Miner, E. M.; Fukushima, T.; Sheberla, D.; Sun, L.; Surendranath, Y.; Dincă, M.
Electrochemical Oxygen Reduction Catalysed by Ni3(Hexaiminotriphenylene)2. Nat. Commun.
2016, 7 (1), 10942.
(50) Miner, E. M.; Gul, S.; Ricke, N. D.; Pastor, E.; Yano, J.; Yachandra, V. K.; Van Voorhis,
T.; Dincă, M. Mechanistic Evidence for Ligand-Centered Electrocatalytic Oxygen Reduction with
the Conductive MOF Ni3(Hexaiminotriphenylene)2. ACS Catal. 2017, 7 (11), 7726–7731.
(51) Miner, E. M.; Wang, L.; Dincă, M. Modular O2 Electroreduction Activity in Triphenylene-
Based Metal–Organic Frameworks. Chem. Sci. 2018, 9 (29), 6286–6291.
(52) Usov, P. M.; Huffman, B.; Epley, C. C.; Kessinger, M. C.; Zhu, J.; Maza, W. A.; Morris,
A. J. Study of Electrocatalytic Properties of Metal–Organic Framework PCN-223 for the Oxygen
Reduction Reaction. ACS Appl. Mater. Interfaces 2017, 9 (39), 33539–33543.
(53) Liu, X. H.; Hu, W. L.; Jiang, W. J.; Yang, Y. W.; Niu, S.; Sun, B.; Wu, J.; Hu, J. S. Well-
182
Defined Metal-O6 in Metal-Catecholates as a Novel Active Site for Oxygen Electroreduction. ACS
Appl. Mater. Interfaces 2017, 9 (34), 28473–28477.
(54) Lions, M.; Tommasino, J.-B.; Chattot, R.; Abeykoon, B.; Guillou, N.; Devic, T.;
Demessence, A.; Cardenas, L.; Maillard, F.; Fateeva, A. Insights into the Mechanism of
Electrocatalysis of the Oxygen Reduction Reaction by a Porphyrinic Metal Organic Framework.
Chem. Commun. 2017, 53 (48), 6496–6499.
(55) Lin, S.; Diercks, C. S.; Zhang, Y.-B.; Kornienko, N.; Nichols, E. M.; Zhao, Y.; Paris, A.
R.; Kim, D.; Yang, P.; Yaghi, O. M.; et al. Covalent Organic Frameworks Comprising Cobalt
Porphyrins for Catalytic CO2 Reduction in Water. Science 2015, 349 (6253), 1208–1213.
(56) Kornienko, N.; Zhao, Y.; Kley, C. S.; Zhu, C.; Kim, D.; Lin, S.; Chang, C. J.; Yaghi, O.
M.; Yang, P. Metal–Organic Frameworks for Electrocatalytic Reduction of Carbon Dioxide. J. Am.
Chem. Soc. 2015, 137 (44), 14129–14135.
(57) Diercks, C. S.; Lin, S.; Kornienko, N.; Kapustin, E. A.; Nichols, E. M.; Zhu, C.; Zhao, Y.;
Chang, C. J.; Yaghi, O. M. Reticular Electronic Tuning of Porphyrin Active Sites in Covalent
Organic Frameworks for Electrocatalytic Carbon Dioxide Reduction. J. Am. Chem. Soc. 2018, 140
(3), 1116–1122.
(58) Matheu, R.; Gutierrez-Puebla, E.; Monge, M. Á.; Diercks, C. S.; Kang, J.; Prévot, M. S.;
Pei, X.; Hanikel, N.; Zhang, B.; Yang, P.; et al. Three-Dimensional Phthalocyanine Metal-
Catecholates for High Electrochemical Carbon Dioxide Reduction. J. Am. Chem. Soc. 2019, 141
(43), 17081–17085.
(59) Hod, I.; Sampson, M. D.; Deria, P.; Kubiak, C. P.; Farha, O. K.; Hupp, J. T. Fe-Porphyrin-
Based Metal–Organic Framework Films as High-Surface Concentration, Heterogeneous Catalysts
for Electrochemical Reduction of CO2. ACS Catal. 2015, 5 (11), 6302–6309.
(60) Johnson, E. M.; Haiges, R.; Marinescu, S. C. Covalent-Organic Frameworks Composed of
Rhenium Bipyridine and Metal Porphyrins: Designing Heterobimetallic Frameworks with Two
Distinct Metal Sites. ACS Appl. Mater. Interfaces 2018, 10 (44), 37919–37927.
(61) Meng, Z.; Luo, J.; Li, W.; Mirica, K. A. Hierarchical Tuning of the Performance of
Electrochemical Carbon Dioxide Reduction Using Conductive Two-Dimensional
Metallophthalocyanine Based Metal-Organic Frameworks. J. Am. Chem. Soc. 2020, 142 (52),
21656–21669.
(62) Downes, C. A.; Clough, A. J.; Chen, K.; Yoo, J. W.; Marinescu, S. C. Evaluation of the H
2 Evolving Activity of Benzenehexathiolate Coordination Frameworks and the Effect of Film
Thickness on H2 Production. ACS Appl. Mater. Interfaces 2018, 10 (2), 1719–1727.
(63) Huang, X.; Yao, H.; Cui, Y.; Hao, W.; Zhu, J.; Xu, W.; Zhu, D. Conductive Copper
Benzenehexathiol Coordination Polymer as a Hydrogen Evolution Catalyst. ACS Appl. Mater.
Interfaces 2017, 9 (46), 40752–40759.
(64) Sun, X.; Wu, K.-H.; Sakamoto, R.; Kusamoto, T.; Maeda, H.; Ni, X.; Jiang, W.; Liu, F.;
183
Sasaki, S.; Masunaga, H.; et al. Bis(Aminothiolato)Nickel Nanosheet as a Redox Switch for
Conductivity and an Electrocatalyst for the Hydrogen Evolution Reaction. Chem. Sci. 2017, 8 (12),
8078–8085.
(65) Downes, C. A.; Marinescu, S. C. Efficient Electrochemical and Photoelectrochemical H2
Production from Water by a Cobalt Dithiolene One-Dimensional Metal–Organic Surface. J. Am.
Chem. Soc. 2015, 137 (43), 13740–13743.
(66) Downes, C. A.; Marinescu, S. C. One Dimensional Metal Dithiolene (M = Ni, Fe, Zn)
Coordination Polymers for the Hydrogen Evolution Reaction. Dalt. Trans. 2016, 45 (48), 19311–
19321.
(67) Downes, C. A.; Marinescu, S. C. Understanding Variability in the Hydrogen Evolution
Activity of a Cobalt Anthracenetetrathiolate Coordination Polymer. ACS Catal. 2017, 7 (12),
8605–8612.
(68) Wang, L.; Tranca, D. C.; Zhang, J.; Qi, Y.; Sfaelou, S.; Zhang, T.; Dong, R.; Zhuang, X.;
Zheng, Z.; Seifert, G. Toward Activity Origin of Electrocatalytic Hydrogen Evolution Reaction on
Carbon-Rich Crystalline Coordination Polymers. Small 2017, 13 (37), 1700783.
(69) Ji, Z.; Trickett, C.; Pei, X.; Yaghi, O. M. Linking Molybdenum-Sulfur Clusters for
Electrocatalytic Hydrogen Evolution. J. Am. Chem. Soc. 2018, 140 (42), 13618–13622.
(70) Solis, B. H.; Hammes-Schiffer, S. Computational Study of Anomalous Reduction
Potentials for Hydrogen Evolution Catalyzed by Cobalt Dithiolene Complexes. J. Am. Chem. Soc.
2012, 134 (37), 15253–15256.
(71) Panetier, J. A.; Letko, C. S.; Tilley, T. D.; Head-Gordon, M. Computational
Characterization of Redox Non-Innocence in Cobalt-Bis(Diaryldithiolene)-Catalyzed Proton
Reduction. J. Chem. Theory Comput. 2016, 12 (1), 223–230.
(72) 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 (26), 9364–9376.
(73) Wang, Y.; Liu, X.; Liu, J.; Al-Mamun, M.; Wee-Chung Liew, A.; Yin, H.; Wen, W.; Zhong,
Y. L.; Liu, P.; Zhao, H. Electrolyte Effect on Electrocatalytic Hydrogen Evolution Performance of
One-Dimensional Cobalt-Dithiolene Metal-Organic Frameworks: A Theoretical Perspective. ACS
Appl. Energy Mater. 2018, 1 (4), 1688–1694.
(74) Nørskov, J. K.; Rossmeisl, J.; Logadottir, A.; Lindqvist, L.; Kitchin, J. R.; Bligaard, T.;
Jónsson, H. Origin of the Overpotential for Oxygen Reduction at a Fuel-Cell Cathode. J. Phys.
Chem. B 2004, 108 (46), 17886–17892.
(75) Skúlason, E.; Karlberg, G. S.; Rossmeisl, J.; Bligaard, T.; Greeley, J.; Jónsson, H.; Nørskov,
J. K. Density Functional Theory Calculations for the Hydrogen Evolution Reaction in an
Electrochemical Double Layer on the Pt(111) Electrode. Phys. Chem. Chem. Phys. 2007, 9 (25),
3241–3250.
184
(76) Li, Y.; Wang, H.; Xie, L.; Liang, Y.; Hong, G.; Dai, H. MoS2 Nanoparticles Grown on
Graphene: An Advanced Catalyst for the Hydrogen Evolution Reaction. J. Am. Chem. Soc. 2011,
133 (19), 7296–7299.
(77) Navarro-Flores, E.; Chong, Z.; Omanovic, S. Characterization of Ni, NiMo, NiW and NiFe
Electroactive Coatings as Electrocatalysts for Hydrogen Evolution in an Acidic Medium. J. Mol.
Catal. A Chem. 2005, 226 (2), 179–197.
(78) Vrubel, H.; Moehl, T.; Grätzel, M.; Hu, X. Revealing and Accelerating Slow Electron
Transport in Amorphous Molybdenum Sulphide Particles for Hydrogen Evolution Reaction. Chem.
Commun. 2013, 49 (79), 8985.
(79) Pham, K.-C.; Chang, Y.-H.; McPhail, D. S.; Mattevi, C.; Wee, A. T. S.; Chua, D. H. C.
Amorphous Molybdenum Sulfide on Graphene–Carbon Nanotube Hybrids as Highly Active
Hydrogen Evolution Reaction Catalysts. ACS Appl. Mater. Interfaces 2016, 8 (9), 5961–5971.
(80) Merki, D.; Vrubel, H.; Rovelli, L.; Fierro, S.; Hu, X. Fe, Co, and Ni Ions Promote the
Catalytic Activity of Amorphous Molybdenum Sulfide Films for Hydrogen Evolution. Chem. Sci.
2012, 3 (8), 2515–2525.
(81) Staniland, S. S.; Rawlings, A.; Bramble, J.; Tolosa, J.; Wilson, O.; García-Martínez, J. C.;
Binns, C. Novel Methods for the Synthesis of Magnetic Nanoparticles. In Molecular Biology; 2014;
Vol. 8, pp 85–128.
(82) Fishman, M.; Zhuang, H. L.; Mathew, K.; Dirschka, W.; Hennig, R. G. Accuracy of
Exchange-Correlation Functionals and Effect of Solvation on the Surface Energy of Copper. Phys.
Rev. B 2013, 87 (24), 245402.
(83) Mathew, K.; Sundararaman, R.; Letchworth-Weaver, K.; Arias, T. A.; Hennig, R. G.
Implicit Solvation Model for Density-Functional Study of Nanocrystal Surfaces and Reaction
Pathways. J. Chem. Phys. 2014, 140 (8), 084106.
(84) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient Approximation Made Simple.
Phys. Rev. Lett. 1996, 77 (18), 3865–3868.
(85) Buendía, F.; Beltrán, M. R. Theoretical Study of Hydrogen Adsorption on Co Clusters.
Comput. Theor. Chem. 2013, 1021, 183–190.
(86) Baker-Hawkes, M. J.; Dori, Z.; Eisenberg, R.; Gray, H. B. The Crystal and Molecular
Structure of the Tetra-n-Butylammonium Salt of the Dianionic Dimer of Bis(1,2,3,4-
Tetrachlorobenzene-5,6-Dithiolato)Cobaltate. J. Am. Chem. Soc. 1968, 90 (16), 4253–4259.
(87) Alvarez, S.; Vicente, R.; Hoffmann, R. Dimerization and Stacking in Transition-Metal
Bisdithiolenes and Tetrathiolates. J. Am. Chem. Soc. 1985, 107 (22), 6253–6277.
(88) Eisenberg, R.; Gray, H. B. Noninnocence in Metal Complexes: A Dithiolene Dawn. Inorg.
Chem. 2011, 50 (20), 9741–9751.
185
(89) Nørskov, J. K.; Bligaard, T.; Logadottir, A.; Kitchin, J. R.; Chen, J. G.; Pandelov, S.;
Stimming, U. Trends in the Exchange Current for Hydrogen Evolution. J. Electrochem. Soc. 2005,
152 (3), J23–J26.
(90) Hinnemann, B.; Moses, P. G.; Bonde, J.; Jørgensen, K. P.; Nielsen, J. H.; Horch, S.;
Chorkendorff, I.; Nørskov, J. K. Biomimetic Hydrogen Evolution: MoS2 Nanoparticles as Catalyst
for Hydrogen Evolution. J. Am. Chem. Soc. 2005, 127 (15), 5308–5309.
(91) Gautier, S.; Steinmann, S. N.; Michel, C.; Fleurat-Lessard, P.; Sautet, P. Molecular
Adsorption at Pt(111). How Accurate Are DFT Functionals? Phys. Chem. Chem. Phys. 2015, 17
(43), 28921–28930.
(92) Thomas, J. G. N. Kinetics of Electrolytic Hydrogen Evolution and the Adsorption of
Hydrogen by Metals. Trans. Faraday Soc. 1961, 57 (0), 1603–1611.
(93) Chen, Z.; Cummins, D.; Reinecke, B. N.; Clark, E.; Sunkara, M. K.; Jaramillo, T. F. Core–
Shell MoO3–MoS2 Nanowires for Hydrogen Evolution: A Functional Design for Electrocatalytic
Materials. Nano Lett. 2011, 11 (10), 4168–4175.
(94) Zhang, W.; Haddad, A. Z.; Garabato, B. D.; Kozlowski, P. M.; Buchanan, R. M.;
Grapperhaus, C. A. Translation of Ligand-Centered Hydrogen Evolution Reaction Activity and
Mechanism of a Rhenium-Thiolate from Solution to Modified Electrodes: A Combined
Experimental and Density Functional Theory Study. Inorg. Chem. 2017, 56 (4), 2177–2187.
(95) Deblase, C. R.; Silberstein, K. E.; Truong, T.-T.; Abruñ, H. D.; Dichtel, W. R. β-
Ketoenamine-Linked Covalent Organic Frameworks Capable of Pseudocapacitive Energy Storage.
J. Am. Chem. Soc 2013, 135, 53.
(96) Deblase, C. R.; Hernández-Burgos, K.; Silberstein, K. E.; Rodríguez-Calero, G. G.; Bisbey,
R. P.; Abruña, H. D.; Dichtel, W. R. Rapid and Efficient Redox Processes within 2D Covalent
Organic Framework Thin Films. ACS Nano 2015, 9 (3), 3178–3183.
(97) Sakamoto, R.; Kambe, T.; Tsukada, S.; Takada, K.; Hoshiko, K.; Kitagawa, Y.; Okumura,
M.; Nishihara, H. Π-Conjugated Trinuclear Group-9 Metalladithiolenes With a Triphenylene
Backbone. Inorg. Chem. 2013, 52 (13), 7411–7416.
(98) Slater, J. C. A Simplification of the Hartree-Fock Method. Phys. Rev. 1951, 81 (3), 385–
390.
(99) Perdew, J. P.; Wang, Y. Accurate and Simple Analytic Representation of the Electron-Gas
Correlation Energy. Phys. Rev. B 1992, 45 (23), 13244–13249.
(100) Kresse, G.; Furthmüller, J. Efficiency of Ab-Initio Total Energy Calculations for Metals
and Semiconductors Using a Plane-Wave Basis Set. Comput. Mater. Sci. 1996, 6 (1), 15–50.
(101) Kresse, G.; Furthmüller, J. Efficient Iterative Schemes for Ab Initio Total-Energy
Calculations Using a Plane-Wave Basis Set. Phys. Rev. B 1996, 54 (16), 11169–11186.
186
(102) Kresse, G.; Hafner, J. Ab Initio Molecular Dynamics for Liquid Metals. Phys. Rev. B 1993,
47 (1), 558–561.
(103) Kresse, G.; Hafner, J. Ab Initio Molecular-Dynamics Simulation of the Liquid-Metal--
Amorphous-Semiconductor Transition in Germanium. Phys. Rev. B 1994, 49 (20), 14251–14269.
(104) Monkhorst, H. J.; Pack, J. D. Special Points for Brillouin-Zone Integrations. Phys. Rev. B
1976, 13 (12), 5188–5192.
(105) Letchworth-Weaver, K.; Arias, T. A. Joint Density Functional Theory of the Electrode-
Electrolyte Interface: Application to Fixed Electrode Potentials, Interfacial Capacitances, and
Potentials of Zero Charge. Phys. Rev. B 2012, 86 (7), 75140.
(106) Jinnouchi, R.; Anderson, A. B. Electronic Structure Calculations of Liquid-Solid Interfaces:
Combination of Density Functional Theory and Modified Poisson-Boltzmann Theory. Phys. Rev.
B 2008, 77 (24), 245417.
(107) Jones, G.; Jakobsen, J.; Shim, S.; Kleis, J.; Andersson, M.; Rossmeisl, J.; Abildpedersen,
F.; Bligaard, T.; Helveg, S.; Hinnemann, B. First Principles Calculations and Experimental Insight
into Methane Steam Reforming over Transition Metal Catalysts. J. Catal. 2008, 259 (1), 147–160.
(108) Cococcioni, M.; de Gironcoli, S. Linear Response Approach to the Calculation of the
Effective Interaction Parameters in the LDA+U Method. Phys. Rev. B 2005, 71 (3), 035105.
(109) Kulik, H. J. Perspective: Treating Electron over-Delocalization with the DFT+U Method.
J. Chem. Phys. 2015, 142 (24), 240901.
(110) Ghosh, S.; Singh, S. K.; Tewary, S.; Rajaraman, G. Enhancing the Double Exchange
Interaction in a Mixed Valence {VIII–VII} Pair: A Theoretical Perspective. Dalt. Trans. 2013, 42
(47), 16490.
(111) Radoń, M. Revisiting the Role of Exact Exchange in DFT Spin-State Energetics of
Transition Metal Complexes. Phys. Chem. Chem. Phys. 2014, 16 (28), 14479–14488.
(112) Mann, G. W.; Lee, K.; Cococcioni, M.; Smit, B.; Neaton, J. B. First-Principles Hubbard U
Approach for Small Molecule Binding in Metal-Organic Frameworks. J. Chem. Phys. 2016, 144
(17), 174104.
(113) Zhou, H. C.; Long, J. R.; Yaghi, O. M. Introduction to Metal-Organic Frameworks. Chem.
Rev. 2012, 112 (2), 673–674.
(114) Downes, C. A.; Marinescu, S. C. Electrocatalytic Metal–Organic Frameworks for Energy
Applications. ChemSusChem 2017, 10 (22), 4374–4392.
(115) Drake, T.; Ji, P.; Lin, W. Site Isolation in Metal-Organic Frameworks Enables Novel
Transition Metal Catalysis. Acc. Chem. Res. 2018, 51 (9), 2129–2138.
(116) Sun, X.; Wu, K.-H.; Sakamoto, R.; Kusamoto, T.; Maeda, H.; Ni, X.; Jiang, W.; Liu, F.;
187
Sasaki, S.; Masunaga, H.; et al. Bis(Aminothiolato)Nickel Nanosheet as a Redox Switch for
Conductivity and an Electrocatalyst for the Hydrogen Evolution Reaction. Chem. Sci. 2017, 8 (12),
8078–8085.
(117) Chen, K.; Downes, C. A.; Schneider, E.; Goodpaster, J. D.; Marinescu, S. C. Improving
and Understanding the Hydrogen Evolving Activity of a Cobalt Dithiolene Metal–Organic
Framework. ACS Appl. Mater. Interfaces 2021, 13 (14), 16384–16395.
(118) Clough, A. J.; Yoo, J. W.; Mecklenburg, M. H.; Marinescu, S. C. Two-Dimensional Metal–
Organic Surfaces for Efficient Hydrogen Evolution from Water. J. Am. Chem. Soc. 2015, 137 (1),
118–121.
(119) Dong, R.; Pfeffermann, M.; Liang, H.; Zheng, Z.; Zhu, X.; Zhang, J.; Feng, X. Large-Area,
Free-Standing, Two-Dimensional Supramolecular Polymer Single-Layer Sheets for Highly
Efficient Electrocatalytic Hydrogen Evolution. Angew. Chem., Int. Ed. 2015, 54 (41), 12058–
12063.
(120) Qin, J.-S.; Du, D.-Y.; Guan, W.; Bo, X.-J.; Li, Y.-F.; Guo, L.-P.; Su, Z.-M.; Wang, Y.-Y.;
Lan, Y.-Q.; Zhou, H.-C. Ultrastable Polymolybdate-Based Metal–Organic Frameworks as Highly
Active Electrocatalysts for Hydrogen Generation from Water. J. Am. Chem. Soc. 2015, 137 (22),
7169–7177.
(121) Hod, I.; Deria, P.; Bury, W.; Mondloch, J. E.; Kung, C.-W.; So, M.; Sampson, M. D.; Peters,
A. W.; Kubiak, C. P.; Farha, O. K.; et al. A Porous Proton-Relaying Metal-Organic Framework
Material That Accelerates Electrochemical Hydrogen Evolution. Nat. Commun. 2015, 6 (1), 8304.
(122) Wu, Y.-P.; Zhou, W.; Zhao, J.; Dong, W.-W.; Lan, Y.-Q.; Li, D.-S.; Sun, C.; Bu, X.
Surfactant-Assisted Phase-Selective Synthesis of New Cobalt MOFs and Their Efficient
Electrocatalytic Hydrogen Evolution Reaction. Angew. Chemie Int. Ed. 2017, 56 (42), 13001–
13005.
(123) Micheroni, D.; Lan, G.; Lin, W. Efficient Electrocatalytic Proton Reduction with Carbon
Nanotube-Supported Metal–Organic Frameworks. J. Am. Chem. Soc. 2018, 140 (46), 15591–
15595.
(124) Gong, Y.; Shi, H.-F.; Hao, Z.; Sun, J.-L.; Lin, J.-H. Two Novel Co(Ii) Coordination
Polymers Based on 1,4-Bis(3-Pyridylaminomethyl)Benzene as Electrocatalysts for Oxygen
Evolution from Water. Dalt. Trans. 2013, 42 (34), 12252.
(125) Kung, C.-W.; Mondloch, J. E.; Wang, T. C.; Bury, W.; Hoffeditz, W.; Klahr, B. M.; Klet,
R. C.; Pellin, M. J.; Farha, O. K.; Hupp, J. T. Metal–Organic Framework Thin Films as Platforms
for Atomic Layer Deposition of Cobalt Ions To Enable Electrocatalytic Water Oxidation. ACS
Appl. Mater. Interfaces 2015, 7 (51), 28223–28230.
(126) Johnson, B. A.; Bhunia, A.; Ott, S. Electrocatalytic Water Oxidation by a Molecular
Catalyst Incorporated into a Metal–Organic Framework Thin Film. Dalt. Trans. 2017, 46 (5),
1382–1388.
188
(127) Lin, S.; Pineda-Galvan, Y.; Maza, W. A.; Epley, C. C.; Zhu, J.; Kessinger, M. C.; Pushkar,
Y.; Morris, A. J. Electrochemical Water Oxidation by a Catalyst-Modified Metal-Organic
Framework Thin Film. ChemSusChem 2017, 10 (3), 514–522.
(128) Shen, J.-Q.; Liao, P.-Q.; Zhou, D.-D.; He, C.-T.; Wu, J.-X.; Zhang, W.-X.; Zhang, J.-P.;
Chen, X.-M. Modular and Stepwise Synthesis of a Hybrid Metal–Organic Framework for Efficient
Electrocatalytic Oxygen Evolution. J. Am. Chem. Soc. 2017, 139 (5), 1778–1781.
(129) Lin, S.; Diercks, C. S.; Zhang, Y.-B.; Kornienko, N.; Nichols, E. M.; Zhao, Y.; Paris, A.
R.; Kim, D.; Yang, P.; Yaghi, O. M.; et al. Covalent Organic Frameworks Comprising Cobalt
Porphyrins for Catalytic CO2 Reduction in Water. Science 2015, 349 (6253), 1208.
(130) Clough, A. J.; Orchanian, N. M.; Skelton, J. M.; Neer, A. J.; Howard, S. A.; Downes, C.
A.; Piper, L. F. J.; Walsh, A.; Melot, B. C.; Marinescu, S. C. Room Temperature Metallic
Conductivity in a Metal–Organic Framework Induced by Oxidation. J. Am. Chem. Soc. 2019,
jacs.9b06898.
(131) Dong, R.; Han, P.; Arora, H.; Ballabio, M.; Karakus, M.; Zhang, Z.; Shekhar, C.; Adler,
P.; Petkov, P. St.; Erbe, A.; et al. High-Mobility Band-like Charge Transport in a Semiconducting
Two-Dimensional Metal–Organic Framework. Nat. Mater. 2018, 17 (11), 1027–1032.
(132) Deng, H.; Doonan, C. J.; Furukawa, H.; Ferreira, R. B.; Towne, J.; Knobler, C. B.; Wang,
B.; Yaghi, O. M. Multiple Functional Groups of Varying Ratios in Metal-Organic Frameworks.
Science 2010, 327, 846–850.
(133) Wang, L. J.; Deng, H.; Furukawa, H.; Gándara, F.; Cordova, K. E.; Peri, D.; Yaghi, O. M.
Synthesis and Characterization of Metal-Organic Framework-74 Containing 2, 4, 6, 8, and 10
Different Metals. Inorg. Chem. 2014, 53 (12), 5881–5883.
(134) Zhang, B.; Zheng, Y.; Ma, T.; Yang, C.; Peng, Y.; Zhou, Z.; Zhou, M.; Li, S.; Wang, Y.;
Cheng, C. Designing MOF Nanoarchitectures for Electrochemical Water Splitting. Adv. Mater.
2021, 33 (17), 2006042.
(135) Zhao, Q.; Lin, X.; Zhou, J.; Zhao, C.; Zheng, D.; Song, S.; Jing, C.; Zhang, L.; Wang, J. A
Tunable Amorphous Heteronuclear Iron and Cobalt Imidazolate Framework Analogue for
Efficient Oxygen Evolution Reactions. Eur. J. Inorg. Chem. 2021, 2021 (8), 702–707.
(136) Dang, Y.; Han, P.; Li, Y.; Zhang, Y.; Zhou, Y. Low-Crystalline Mixed Fe-Co-MOFs for
Efficient Oxygen Evolution Electrocatalysis. J. Mater. Sci. 2020, 55 (28), 13951–13963.
(137) Burke, M. S.; Kast, M. G.; Trotochaud, L.; Smith, A. M.; Boettcher, S. W. Cobalt-Iron
(Oxy)Hydroxide Oxygen Evolution Electrocatalysts: The Role of Structure and Composition on
Activity, Stability, and Mechanism. J. Am. Chem. Soc. 2015, 137 (10), 3638–3648.
(138) Batchellor, A. S.; Boettcher, S. W. Pulse-Electrodeposited Ni-Fe (Oxy)Hydroxide Oxygen
Evolution Electrocatalysts with High Geometric and Intrinsic Activities at Large Mass Loadings.
ACS Catal. 2015, 5 (11), 6680–6689.
189
(139) Downes, C. A.; Marinescu, S. C. Understanding Variability in the Hydrogen Evolution
Activity of a Cobalt Anthracenetetrathiolate Coordination Polymer. ACS Catal. 2017, 7 (12),
8605–8612.
(140) Stevens, M. B.; Enman, L. J.; Batchellor, A. S.; Cosby, M. R.; Vise, A. E.; Trang, C. D.
M.; Boettcher, S. W. Measurement Techniques for the Study of Thin Film Heterogeneous Water
Oxidation Electrocatalysts. Chem. Mater. 2017, 29 (1), 120–140.
(141) Kambe, T.; Sakamoto, R.; Hoshiko, K.; Takada, K.; Miyachi, M.; Ryu, J. H.; Sasaki, S.;
Kim, J.; Nakazato, K.; Takata, M.; et al. π-Conjugated Nickel Bis(Dithiolene) Complex Nanosheet.
J. Am. Chem. Soc. 2013, 135 (7), 2462–2465.
(142) Liu, L.; Tu, Z.; Xu, W.; Chen, J.; Zou, Y.; Yi, Y.; Wu, X.; Li, H.; Liang, Y.; Huang, X.; et
al. Highly Conducting Neutral Coordination Polymer with Infinite Two-Dimensional Silver–
Sulfur Networks. J. Am. Chem. Soc. 2018, 140 (45), 15153–15156.
(143) REN21. Renewables 2019 Global Status Report; https://www.ren21.net/gsr-2019/.
(144) Seh, Z. W.; Kibsgaard, J.; Dickens, C. F.; Chorkendorff, I.; Nørskov, J. K.; Jaramillo, T. F.
Combining Theory and Experiment in Electrocatalysis: Insights into Materials Design. Science
2017, 355 (6321), eaad4998.
(145) Qin, J.-S.; Du, D.-Y.; Guan, W.; Bo, X.-J.; Li, Y.-F.; Guo, L.-P.; Su, Z.-M.; Wang, Y.-Y.;
Lan, Y.-Q.; Zhou, H.-C. Ultrastable Polymolybdate-Based Metal–Organic Frameworks as Highly
Active Electrocatalysts for Hydrogen Generation from Water. J. Am. Chem. Soc. 2015, 137 (22),
7169–7177.
(146) Wu, Y.-P.; Zhou, W.; Zhao, J.; Dong, W.-W.; Lan, Y.-Q.; Li, D.-S.; Sun, C.; Bu, X.
Surfactant-Assisted Phase-Selective Synthesis of New Cobalt MOFs and Their Efficient
Electrocatalytic Hydrogen Evolution Reaction. Angew. Chem., Int. Ed. 2017, 56 (42), 13001–
13005.
(147) Lu, X.-F.; Liao, P.-Q.; Wang, J.-W.; Wu, J.-X.; Chen, X.-W.; He, C.-T.; Zhang, J.-P.; Li,
G.-R.; Chen, X.-M. An Alkaline-Stable, Metal Hydroxide Mimicking Metal–Organic Framework
for Efficient Electrocatalytic Oxygen Evolution. J. Am. Chem. Soc. 2016, 138 (27), 8336–8339.
(148) Lin, S.; Pineda-Galvan, Y.; Maza, W. A.; Epley, C. C.; Zhu, J.; Kessinger, M. C.; Pushkar,
Y.; Morris, A. J. Electrochemical Water Oxidation by a Catalyst-Modified Metal-Organic
Framework Thin Film. ChemSusChem 2017, 10 (3), 469–469.
(149) Shen, J.-Q.; Liao, P.-Q.; Zhou, D.-D.; He, C.-T.; Wu, J.-X.; Zhang, W.-X.; Zhang, J.-P.;
Chen, X.-M. Modular and Stepwise Synthesis of a Hybrid Metal–Organic Framework for Efficient
Electrocatalytic Oxygen Evolution. J. Am. Chem. Soc. 2017, 139 (5), 1778–1781.
(150) Wang, L.; Wu, Y.; Cao, R.; Ren, L.; Chen, M.; Feng, X.; Zhou, J.; Wang, B. Fe/Ni Metal-
Organic Frameworks and Their Binder-Free Thin Films for Efficient Oxygen Evolution with Low
Overpotential. ACS Appl. Mater. Interfaces 2016, 8 (26), 16736–16743.
190
(151) Miner, E. M.; Wang, L.; Dincă, M. Modular O2 Electroreduction Activity in Triphenylene-
Based Metal–Organic Frameworks. Chem. Sci. 2018, 9 (29), 6286–6291.
(152) Kornienko, N.; Zhao, Y.; Kley, C. S.; Zhu, C.; Kim, D.; Lin, S.; Chang, C. J.; Yaghi, O.
M.; Yang, P. Metal-Organic Frameworks for Electrocatalytic Reduction of Carbon Dioxide. J. Am.
Chem. Soc. 2015, 137 (44), 14129–14135.
(153) Kung, C. W.; Audu, C. O.; Peters, A. W.; Noh, H.; Farha, O. K.; Hupp, J. T. Copper
Nanoparticles Installed in Metal-Organic Framework Thin Films Are Electrocatalytically
Competent for CO2 Reduction. ACS Energy Lett. 2017, 2 (10), 2394–2401.
(154) Kobayashi, Y.; Jacobs, B.; Allendorf, M. D.; Long, J. R. Conductivity, Doping, and Redox
Chemistry of a Microporous Dithiolene-Based Metal-Organic Framework. Chem. Mater. 2010, 22
(14), 4120–4122.
(155) Peng, Y.-L.; Pham, T.; Li, P.; Wang, T.; Chen, Y.; Chen, K.-J.; Forrest, K. A.; Space, B.;
Cheng, P.; Zaworotko, M. J.; et al. Robust Ultramicroporous Metal-Organic Frameworks with
Benchmark Affinity for Acetylene. Angew. Chem., Int. Ed. 2018, 57 (34), 10971–10975.
(156) Koshiba, K.; Yamauchi, K.; Sakai, K. A Nickel Dithiolate Water Reduction Catalyst
Providing Ligand-Based Proton-Coupled Electron-Transfer Pathways. Angew. Chemie Int. Ed.
2017, 56 (15), 4247–4251.
(157) Aimoto, Y.; Koshiba, K.; Yamauchi, K.; Sakai, K. A Family of Molecular Nickel
Hydrogen Evolution Catalysts Providing Tunable Overpotentials Using Ligand-Centered Proton-
Coupled Electron Transfer Paths. Chem. Commun. 2018, 54 (91), 12820–12823.
(158) Koshiba, K.; Yamauchi, K.; Sakai, K. Consecutive Ligand-Based PCET Processes
Affording a Doubly Reduced Nickel Pyrazinedithiolate Which Transforms into a Metal Hydride
Required to Evolve H2. Dalt. Trans. 2019, 48 (2), 635–640.
(159) Hayashi, M.; Takahashi, Y.; Yoshida, Y.; Sugimoto, K.; Kitagawa, H. Role of d -Elements
in a Proton–Electron Coupling of d –π Hybridized Electron Systems. J. Am. Chem. Soc. 2019, 141
(29), 11686–11693.
(160) Micheroni, D.; Lan, G.; Lin, W. Efficient Electrocatalytic Proton Reduction with Carbon
Nanotube-Supported Metal–Organic Frameworks. J. Am. Chem. Soc. 2018, 140 (46), 15591–
15595.
(161) 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 (5), 2827–2834.
(162) Fang, M.; Engelhard, M. H.; Zhu, Z.; Helm, M. L.; Roberts, J. A. S. Electrodeposition from
Acidic Solutions of Nickel Bis(Benzenedithiolate) Produces a Hydrogen-Evolving Ni–S Film on
Glassy Carbon. ACS Catal. 2014, 4 (1), 90–98.
(163) Hu, C.; Ma, Q.; Hung, S.-F.; Chen, Z.-N.; Ou, D.; Ren, B.; Chen, H. M.; Fu, G.; Zheng, N.
191
In Situ Electrochemical Production of Ultrathin Nickel Nanosheets for Hydrogen Evolution
Electrocatalysis. Chem 2017, 3 (1), 122–133.
(164) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. A Consistent and Accurate Ab Initio
Parametrization of Density Functional Dispersion Correction (DFT-D) for the 94 Elements H-Pu.
J. Chem. Phys. 2010, 132 (15), 154104.
(165) Becke, A. D.; Johnson, E. R. A Density-Functional Model of the Dispersion Interaction. J.
Chem. Phys. 2005, 123 (15), 154101.
(166) Kresse, G.; Joubert, D. From Ultrasoft Pseudopotentials to the Projector Augmented-Wave
Method. Phys. Rev. B 1999, 59 (3), 1758–1775.
(167) Blöchl, P. E. Projector Augmented-Wave Method. Phys. Rev. B 1994, 50 (24), 17953–
17979.
(168) Bibila Mayaya Bisseyou, Y.; Bouhmaida, N.; Guillot, B.; Lecomte, C.; Lugan, N.;
Ghermani, N.; Jelsch, C. Experimental and Database-Transferred Electron-Density Analysis and
Evaluation of Electrostatic Forces in Coumarin-102 Dye. Acta Crystallogr. Sect. B Struct. Sci.
2012, 68 (6), 646–660.
(169) Sanville, E.; Kenny, S. D.; Smith, R.; Henkelman, G. Improved Grid-Based Algorithm for
Bader Charge Allocation. J. Comput. Chem. 2007, 28 (5), 899–908.
(170) Yu, M.; Trinkle, D. R. Accurate and Efficient Algorithm for Bader Charge Integration. J.
Chem. Phys. 2011, 134 (6), 064111.
(171) Tang, W.; Sanville, E.; Henkelman, G. A Grid-Based Bader Analysis Algorithm without
Lattice Bias. J. Phys. Condens. Matter 2009, 21 (8), 084204.
(172) Marenich, A. V.; Jerome, S. V.; Cramer, C. J.; Truhlar, D. G. Charge Model 5: An
Extension of Hirshfeld Population Analysis for the Accurate Description of Molecular Interactions
in Gaseous and Condensed Phases. J. Chem. Theory Comput. 2012, 8 (2), 527–541.
(173) Wang, B.; Li, S. L.; Truhlar, D. G. Modeling the Partial Atomic Charges in
Inorganometallic Molecules and Solids and Charge Redistribution in Lithium-Ion Cathodes. J.
Chem. Theory Comput. 2014, 10 (12), 5640–5650.
(174) Lubitz, W.; Ogata, H.; Rüdiger, O.; Reijerse, E. Hydrogenases. Chem. Rev. 2014, 114 (8),
4081–4148.
(175) Schilter, D.; Camara, J. M.; Huynh, M. T.; Hammes-Schiffer, S.; Rauchfuss, T. B.
Hydrogenase Enzymes and Their Synthetic Models: The Role of Metal Hydrides. Chem. Rev. 2016,
116 (15), 8693–8749.
(176) Wombwell, C.; Caputo, C. A.; Reisner, E. [NiFeSe]-Hydrogenase Chemistry. Acc. Chem.
Res. 2015, 48 (11), 2858–2865.
192
(177) Valente, F. M. A.; Oliveira, A. S. F.; Gnadt, N.; Pacheco, I.; Coelho, A. V.; Xavier, A. V.;
Teixeira, M.; Soares, C. M.; Pereira, I. A. C. Hydrogenases in Desulfovibrio Vulgaris
Hildenborough: Structural and Physiologic Characterisation of the Membrane-Bound [NiFeSe]
Hydrogenase. JBIC J. Biol. Inorg. Chem. 2005, 10 (6), 667–682.
(178) Kong, D.; Cha, J. J.; Wang, H.; Lee, H. R.; Cui, Y. First-Row Transition Metal
Dichalcogenide Catalysts for Hydrogen Evolution Reaction. Energy Environ. Sci. 2013, 6 (12),
3553–3558.
(179) Kong, D.; Wang, H.; Lu, Z.; Cui, Y. CoSe2 Nanoparticles Grown on Carbon Fiber Paper:
An Efficient and Stable Electrocatalyst for Hydrogen Evolution Reaction. J. Am. Chem. Soc. 2014,
136 (13), 4897–4900.
(180) Wang, H.; Kong, D.; Johanes, P.; Cha, J. J.; Zheng, G.; Yan, K.; Liu, N.; Cui, Y. MoSe2
and WSe2 Nanofilms with Vertically Aligned Molecular Layers on Curved and Rough Surfaces.
Nano Lett. 2013, 13 (7), 3426–3433.
(181) Henckel, D. A.; Lenz, O. M.; Krishnan, K. M.; Cossairt, B. M. Improved HER Catalysis
through Facile, Aqueous Electrochemical Activation of Nanoscale WSe2. Nano Lett. 2018, 18 (4),
2329–2335.
(182) Anantharaj, S.; Ede, S. R.; Sakthikumar, K.; Karthick, K.; Mishra, S.; Kundu, S. Recent
Trends and Perspectives in Electrochemical Water Splitting with an Emphasis on Sulfide, Selenide,
and Phosphide Catalysts of Fe, Co, and Ni: A Review. ACS Catal. 2016, 6 (12), 8069–8097.
(183) Cui, Y.; Yan, J.; Chen, Z.; Xing, W.; Ye, C.; Li, X.; Zou, Y.; Sun, Y.; Liu, C.; Xu, W.; et
al. Synthetic Route to a Triphenylenehexaselenol-Based Metal Organic Framework with Semi-
Conductive and Glassy Magnetic Properties. iScience 2020, 23 (1), 100812.
(184) Turner, D. L.; Vaid, T. P. Synthesis of Protected Benzenepolyselenols. J. Org. Chem. 2012,
77 (20), 9397–9400.
(185) Costentin, C.; Saveant, J.-M. Cyclic Voltammetry Analysis of Electrocatalytic Films. J.
Phys. Chem. C 2015, 119 (22), 12174–12182.
(186) Anantharaj, S.; Noda, S. Appropriate Use of Electrochemical Impedance Spectroscopy in
Water Splitting Electrocatalysis. ChemElectroChem 2020, 7 (10), 2297–2308.
(187) Yoon, Y.; Yan, B.; Surendranath, Y. Suppressing Ion Transfer Enables Versatile
Measurements of Electrochemical Surface Area for Intrinsic Activity Comparisons. J. Am. Chem.
Soc. 2018, 140 (7), 2397–2400.
(188) Breslow, R.; Jaun, B.; Kluttz, R. Q.; Xia, C. Ground State Pi-Electron Triplet Molecules
of Potential Use in the Synthesis of Organic Ferromagnets. Tetrahedron 1982, 38 (6), 863–867.
Abstract (if available)
Abstract
In the face of a looming climate crisis, immediate action needs to be taken to decarbonize the current fossil-fuel-dominated energy economy by adopting a more sustainable alternative. Thanks to the alliance between scientific breakthroughs and policy support, recent years have witnessed the rapid growth in renewable electricity generation using solar and wind energy. However, the integration of renewables into other major energy sectors, such as transportation and thermal production, is still very limited. To address this challenge, electrocatalysis provides a feasible solution, where abundant small molecules are converted into value-added products via the input of renewable electricity to realize the storage of renewables in chemical bonds. The products can then be consumed as fuels for transportation or thermal production, or as chemical feedstocks in industrial processes. The key to realizing this electrocatalysis-based sustainable energy future is to develop highly efficient electrocatalysts that are composed of abundant elements and can facilitate chemical catalysis under mild conditions. ❧ In this dissertation, several electrocatalysts that are based on metal-organic frameworks (MOFs) are developed for green hydrogen production from water. Electrocatalytic hydrogen production is of particular interest because hydrogen is a very important chemical feedstock in industrial productions and a promising carbon-free energy carrier. In recent years, MOFs have emerged as an extensive class of highly functional materials with unique properties such as high porosities, large surface areas, and extraordinary structural and compositional variabilities. The application of MOFs in clean energy is an emerging field of research and is of great significance in the context of the current climate crisis. A brief outline of this dissertation is provided below: ❧ Chapter 1 presents a general introduction of the dissertation, including a brief discussion on the current global energy status, the fundamentals of electrocatalytic hydrogen production, general design principles of electrocatalysts, as well as the frontiers in MOF-based electrocatalysis. In Chapter 2, the HER performance of a known dithiolene-based MOF, the cobalt triphenylene-2,3,6,7,10,11-hexathiolate (THT) MOF, is optimized by unraveling the reaction mechanism and identifying the key factors that dictate the overall catalytic performance. The optimization results in the most active MOF-based electrocatalyst for hydrogen production that comprises only earth-abundant elements. Chapter 3 discusses the role of metal centers in the HER activity using the examples of a series of iron and cobalt/iron mixed-metal dithiolate MOFs. In Chapter 4, a conductive three-dimensional dithiolene-based MOF, the Cu[Ni(2,3-pyrazinedithiolate)2] MOF, is investigated as an HER electrocatalyst for the first time. Lastly, Chapter 5 presents the synthesis and HER characterization of a diselenolate-based MOF, an analogous derivative of the dithiolate MOFs. This study is to highlight the role of the chalcogen within the ligand, which is inspired by nature where the selenium-containing [NiFe] hydrogenase displays much higher activity than its sulfur-only analog for hydrogen production.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
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
Electrocatalytic thiolate- and selenolate-based coordination polymers for solar energy conversion
PDF
Rhenium bipyridine catalysts with pendant amines: substituent and positional effects on the electrocatalytic reduction of CO₂ to CO
PDF
Hot electron driven photocatalysis and plasmonic sensing using metallic gratings
PDF
Understanding the mechanism of oxygen reduction and oxygen evolution on transition metal oxide electrocatalysts and applications in iron-air rechargeable battery
PDF
Hydrogen energy system production and storage via iridium-based catalysts
PDF
Catalysis for solar fuels production: molecular and materials approaches
PDF
Photophysical properties of luminescent iridium and coinage metal complexes
PDF
Novel methods for functional group interconversions in organic synthesis and structural characterization of new transition metal heterogeneous catalysts for potential carbon neutral hydrogen storage
PDF
Thermoelectric and transport studies of low dimensional materials & hot electron-driven photocatalysis on plasmon-resonant grating nanostructures
PDF
Organic photovoltaics: from materials development to device application
PDF
Design of catalysts for the transformation of abundant small molecules into solar fuels
PDF
Sustainable continuous flow syntheses of colloidal inorganic nanoparticle catalysts
PDF
Two-coordinate coinage metal complexes for solar fuels and organic LED chemistry
PDF
Solution phase synthesis routes to functional nanomaterials for energy storage
PDF
Design, synthesis, and study of polypyridine based molecular and heterogenized molecular electrocatalysts for CO₂ reduction
PDF
Electrochemical pathways for sustainable energy storage and energy conversion
PDF
Porphyrin based near infrared‐absorbing materials for organic photovoltaics
PDF
Development of new bifunctional iridium complexes for hydrogenation and dehydrogenation reactions
Asset Metadata
Creator
Chen, Keying
(author)
Core Title
Dithiolate-based metal-organic frameworks for electrocatalytic hydrogen evolution
School
College of Letters, Arts and Sciences
Degree
Doctor of Philosophy
Degree Program
Chemistry
Degree Conferral Date
2021-12
Publication Date
09/13/2021
Defense Date
08/16/2021
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
dithiolene,electrocatalysis,hydrogen evolution,metal-organic frameworks,OAI-PMH Harvest,renewable energy storage,two-dimensional materials
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Marinescu, Smaranda C. (
committee chair
), Brutchey, Richard L. (
committee member
), Ravichandran, Jayakanth (
committee member
)
Creator Email
keying.chen@hotmail.com,keyingch@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-oUC15909836
Unique identifier
UC15909836
Legacy Identifier
etd-ChenKeying-10058
Document Type
Dissertation
Format
application/pdf (imt)
Rights
Chen, Keying
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 author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.
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
Repository Email
cisadmin@lib.usc.edu
Tags
dithiolene
electrocatalysis
hydrogen evolution
metal-organic frameworks
renewable energy storage
two-dimensional materials