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
/
Chemistry surrounding tin: from a new electrocatalyst for CO₂ reduction to syngas to a novel CF₂H transfer reagent and related computational studies
(USC Thesis Other)
Chemistry surrounding tin: from a new electrocatalyst for CO₂ reduction to syngas to a novel CF₂H transfer reagent and related computational studies
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
Chemistry Surrounding Tin: From a New Electrocatalyst for CO
2
Reduction to Syngas to a Novel CF
2
H Transfer Reagent and
Related Computational Studies
by
John-Paul Jones
__________________________________
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)
August 2015
Copyright 2015 John-Paul Jones
ii
DEDICATION
To
Calvin, Rory, and Oliver
iii
ACKNOWLEDGEMENTS
Without the help of the people listed here this thesis would not have been
possible. Dr. George Olah first sparked my interest in chemistry when I was in high
school, and has continued to inspire me throughout my time as a Ph.D. student. Dr. Surya
Prakash has provided support in innumerable ways since I first set foot in his lab as a
summer intern in 2004. Dr. Matt Moran got me up to speed when I joined the lab as a
first year graduate student. As I continued my studies, I took up far too much of Dr. Alain
Goeppert’s time with questions about the design of my high pressure reactor among other
things. When my focus shifted to electrochemistry, Drs. Sri Narayan, Souradip
Malkhandi, Aswin Karthik, and Charlie Krause all provided me with the education
necessary to achieve my goals. There have been many Olah/Prakash group members
during my tenure with the group, and I am grateful to all of them for their help and
support.
The staff at Loker have also been very helpful, and I would like to thank Jessie,
Carole, David and Ralph for all that they have done for me. Don, Mike, and Ramon at the
USC Machine Shop helped immensely by not only machining the high pressure
electrolyzer to my specifications, but helping design solutions to problems that arose with
the various designs. Ross Lewis also helped greatly with the design and fabrication of the
electronics used to control the high pressure electrolyzer.
I owe my wife, Sarah, a huge debt of gratitude for supporting and encouraging me
throughout my time as a graduate student. I would definitely not be here without her and
iv
I hope she can forgive me for taking a little longer than initially planned to finish. My
family has also been very supportive in helping me achieve my educational goals
throughout my life leading up to this point, and for that I am extremely thankful.
v
TABLE OF CONTENTS
DEDICATION ........................................................................................................................ ii
ACKNOWLEDGEMENTS ........................................................................................................ iii
LIST OF TABLES ................................................................................................................ viii
LIST OF FIGURES ...................................................................................................................x
LIST OF SCHEMES .............................................................................................................. xiii
ABBREVIATIONS ............................................................................................................... xiv
ABSTRACT ........................................................................................................................ xvi
Chapter 1: Electrochemical CO
2
Reduction....................................................................1
1.1 Introduction .........................................................................................................1
1.2 The mechanism of electrochemical CO
2
reduction on metal electrodes .............3
1.2.1 CO
2
reduction efficiency.....................................................................7
1.3 Methods for CO
2
reduction ...............................................................................10
1.3.1 CO
2
reduction to formate on metal electrodes in aqueous systems ..10
1.3.2 Carbon monoxide from CO
2 .........................................................................................
12
1.3.3 Higher order products from CO
2
reduction ......................................15
1.3.4 Electrochemical reduction of CO
2
in flow cells ...............................18
1.3.5 High pressure electrochemical CO
2
reduction ..................................21
1.3.6 Molecular catalyst assisted CO
2
reduction .......................................24
1.3.7 Non-aqueous solvent systems for CO
2
reduction .............................28
1.3.8 Solid oxide approaches to CO
2
reduction .........................................31
1.3.9 Direct photochemical reduction of CO
2
to methanol ........................36
1.4 Electrochemical water splitting .........................................................................37
1.5 Summary ...........................................................................................................41
1.6 References .........................................................................................................42
vi
Chapter 2: Reduction of CO
2
and water to syngas using gold or tin nanoparticles
in a high pressure flow electrolyzer .....................................................................54
2.1 Introduction .......................................................................................................54
2.2 Electrolyzer and system development ...............................................................55
2.2.1 High pressure electrolyzer design .....................................................56
2.2.2 High pressure flow system design ....................................................61
2.3 Results and discussion .......................................................................................66
2.3.1 Screening conditions .........................................................................66
2.3.2 Effect of pressure and flow rate ........................................................71
2.3.3 CO
2
reduction on gold ......................................................................73
2.3.4 Tin as a cathode for syngas production.............................................76
2.3.5 The effect of the cation in CO
2
reduction .........................................81
2.3.6 Increasing current density .................................................................83
2.3.7 Tokuyama anion conducting membrane ...........................................85
2.4 Technoeconomic analysis ..................................................................................87
2.4.1 Assumptions ......................................................................................88
2.4.2 Conditions for formic acid production ..............................................89
2.4.3 Conditions for CO production ..........................................................93
2.4.4 Comparison to commercial producers ..............................................96
2.5 Conclusion .........................................................................................................99
2.6 Experimental ...................................................................................................102
2.7 References .......................................................................................................103
Chapter 3: Difluoromethylation of Aryl (Heteroaryl) Iodides and β-Styrenyl
Halides using Copper Mediated Tributyl(difluoromethyl)stannane ..............106
3.1 Introduction .....................................................................................................106
3.2 Synthesis of TBTCF
2
H ....................................................................................109
3.3 Copper mediated difluoromethylation of iodoarenes with TBTCF
2
H ............112
3.4 Copper mediated difluoromethylation of β–styrenes with TBTCF
2
H ............115
3.5 Mechanistic studies .........................................................................................116
3.5.1 Part 1: transmetallation ...................................................................116
3.5.2 Part 2: ipso-substitution ..................................................................120
3.5.3 Overall mechanism .........................................................................123
3.6 Conclusion .......................................................................................................124
3.7 Experimental ...................................................................................................125
3.7.1 Materials and Instrumentation .......................................................125
3.7.2 Synthesis of (n-Bu)
3
SnCF
2
H (2a) ..................................................126
3.7.3 Synthesis of Me
3
SnH .....................................................................127
vii
3.7.4 Synthesis of Me
3
SnCF
2
H (2b) .......................................................127
3.7.5 Synthesis of Ph
3
SnCF
2
H (2c) ........................................................128
3.7.6 Synthesis of Cy
3
SnCF
2
H (2d) .......................................................129
3.7.7 Synthesis of (2-halovinyl)arenes ...................................................130
3.7.8 Synthesis of difluoromethylated arenes and styrenes ....................130
3.7.9 VT NMR studies ...........................................................................131
3.7.10 Representative Spectra ................................................................132
3.7.11 Computational details ..................................................................142
3.7.11.1 Example computational structures .................................143
3.8 References .......................................................................................................146
Chapter 4: Group Electronegativities: A Simple Computational Method ...............151
4.1 Introduction .....................................................................................................151
4.2 Experimental Methodology .............................................................................154
4.3 Results and Discussion ....................................................................................160
4.4 Comparison to Literature Values ....................................................................166
4.5 Conclusion .......................................................................................................171
4.6 References .......................................................................................................172
BIBLIOGRAPHY .................................................................................................................175
viii
LIST OF TABLES
Table 1.1: Standard reduction potentials for CO
2
reduction ................................................2
Table 1.2: Full cell electrochemical reactions derived from thermodynamic data ..............8
Table 1.3: Formate production on metal electrodes in aqueous media ..............................11
Table 1.4: Electrochemical reduction of CO
2
to CO on metals under ambient
conditions ...............................................................................................................13
Table 1.5: Higher Products from CO
2
Reduction ..............................................................16
Table 1.6: Electrochemical flow reactors for CO
2
reduction .............................................20
Table 1.7: High pressure CO
2
reduction ............................................................................22
Table 1.8: CO
2
reduction using molecular catalysts ..........................................................26
Table 1.9: CO
2
reduction in non-aqueous solvents ............................................................29
Table 1.10: CO
2
reduction using SOECs ...........................................................................34
Table 2.1: Screening CO
2
reduction conditions .................................................................67
Table 2.2: Formic Acid Baseline Conditions .....................................................................89
Table 2.3: CO Baseline Conditions ...................................................................................94
Table 2.4: Syngas Hypothetical Conditions.......................................................................95
Table 3.1: CaI
2
Initiated Optimization of (n-Bu)
3
SnCF
2
H Synthesis ..............................110
Table 3.2: Conversion of R
3
SnH to R
3
SnCF
2
H ...............................................................111
Table 3.3: Difluoromethylation of Iodoarenes using 2a ..................................................113
Table 3.4: Difluoromethylation of β-styrenyl halides using 2a .......................................115
Table 3.5: Comparison of Copper Mediated Difluoromethylation ..................................122
Table 4.1: Group ENs (scaled to Pauling values) determined from the natural charge
on the R–group of R–X (R = proton, methyl, phenyl; B3LYP/aug-cc-pVTZ)....158
ix
Table 4.2: Group electronegativities (scaled to Pauling values) determined from the
natural charge on the R group of R–X (R = vinyl; B3LYP/aVTZ) .....................161
x
LIST OF FIGURES
Figure 1.1. Schematic of an electrochemical flow cell with buffer layer for CO
2
reduction ................................................................................................................19
Figure 1.2. Typical high pressure reactor modified for CO
2
electrolysis ..........................23
Figure 1.3. The dependence of the faradaic efficiencies of reduction products on
the CO
2
pressure in the electrochemical reduction of CO
2
on a Ag-GDE at
300 mA cm
-2
in 0.5 mol dm
-3
KHCO
3
...................................................................24
Figure 1.4. Molecular catalysts for electrochemical CO
2
reduction ..................................27
Figure 1.5. Diagram of solid-oxide three electrode cell ....................................................31
Figure 1.6. I-V curves from an SOEC with a LSCM/GDC cathode operated at
900 °C at various atmospheres ...............................................................................32
Figure 1.7. Water electrolysis in acid and base .................................................................37
Figure 1.8. Alkaline electrolysis cell .................................................................................38
Figure 1.9. Capital Cost Breakdown of PEM Electrolyzer ................................................40
Figure 2.1. High pressure electrolyzer version 1 ...............................................................56
Figure 2.2. High pressure electrolyzer version 2 ...............................................................58
Figure 2.3. High pressure electrolyzer with floating flow field (version 3) ......................59
Figure 2.4. Cutaway view of high pressure flow electrolyzer ...........................................60
Figure 2.5. Initial high pressure flow system design .........................................................62
Figure 2.6. Automated high pressure flow system ............................................................63
Figure 2.7. Voltage to current signal using an op-amp ......................................................65
Figure 2.8. Effect of flow rate and pressure on CO FE on a gold cathode ........................72
Figure 2.9. CO
2
reduction on gold with a 0.1 M NaOH anolyte solution .........................74
Figure 2.10. CO
2
reduction on gold with a 0.1 M Li2CO3 anolyte solution .....................75
xi
Figure 2.11. CO
2
reduction at various current densities on a tin electrode .......................77
Figure 2.12. CO
2
reduction on tin with 0.1 M NaOH anolyte solution .............................78
Figure 2.13. CO
2
reduction on tin with a 0.1 M Li
2
CO
3
anolyte solution .........................80
Figure 2.14. Total CO
2
reduction FE on tin and gold ........................................................81
Figure 2.15. CO
2
reduction on tin with a 0.1 M Li
2
CO
3
electrolyte at 60 mA/cm
2
..........84
Figure 2.16. CO
2
reduction on silver using the Tokuyama membrane ..............................86
Figure 2.17. Cost of Formic Acid as a Function of Cell Voltage ......................................90
Figure 2.18. Cost of Formic Acid as a Function of FE ......................................................91
Figure 2.19. Cost of Formic Acid as a Function of Electricity Cost .................................92
Figure 2.20. Production of Syngas in a 2:1 H
2
:CO Ratio by Various Methods ................95
Figure 2.21. Costs Compared.............................................................................................98
Figure 3.1. Low temperature (-30 °C)
19
F NMR spectrum of TBTCF
2
H (2a) with
TMAF ..................................................................................................................119
Figure 3.2. Ambient temperature
19
F NMR spectrum of TBTCF
2
H (2a) with TMAF
after two days .......................................................................................................120
Figure 3.3. Sn(CH
3
)
3
FCF
2
H + DMF-Cu-I in DMF (transition state, Scheme 3.2,
Structure D) (aug-cc-pVDZ) with hydrogen atoms on methyl groups
removed for clarity ...............................................................................................143
Figure 3.4. Iodobenzene + DMF-Cu-CF
2
H square planar complex in DMF (Scheme
3.2, Structure H) ...................................................................................................145
Figure 4.1. Linear correlations between various group electronegativity scales .............153
Figure 4.2. Sum of the charges on the vinyl group for H
2
C=CHX (X = F, Cl, Br, I)
graphed against Pauling electronegativities .........................................................157
Figure 4.3. Graphical representation of group electronegativities via NBO charge
distribution ...........................................................................................................160
Figure 4.4. Correlation between calculated (this work; vinyl group) and Inamoto
and Masuda’s group electronegativities...............................................................167
xii
Figure 4.5. Absolute residuals from regression analysis .................................................167
Figure 4.6. Correlation with Dailey and Shoolery group electronegativities ..................168
Figure 4.7. Correlation with Wells group electronegativities ..........................................169
Figure 4.8. Correlation with Boyd and Boyd’s group electronegativities .......................170
Figure 4.9. Correlation with Suresh and Koga’s group electronegativities .....................171
xiii
LIST OF SCHEMES
Scheme 1.1: Mechanism for electrochemical CO2 reduction on metal surfaces
in water.....................................................................................................................4
Scheme 1.2: Equilibrium reactions of CO
2
with water ........................................................6
Scheme 1.3: Possible interaction between CO
2
and BF
4
-
..................................................30
Scheme 1.4: Water-gas shift reaction ................................................................................33
Scheme 3.1: Methods for Difluoromethylation ...............................................................108
Scheme 3.2: DFT calculations for part 1: transmetallation .............................................117
Scheme 3.3: DFT calculation for part 2: ipso-substitution ..............................................121
xiv
ABBREVIATIONS
B3LYP Becke, 3-parameter, Lee-Yang-Parr
DAQ Data acquisition
DCM Dichloromethane
DFT Density functional theory
DMA Dimethylacetamide
DMF Dimethylformamide
DMPU 1,3-Dimethyl-3,4,5,6-tetrahydro-2(1H)-pyrimidinone
DMSO Dimethylsulfoxide
E
0
Standard reduction potential referenced to the standard hydrogen electrode
EN Electronegativity
FE Faradaic efficiency
FT Fischer-Tropsch
gEN Group electronegativity
HER Hydrogen evolution reaction
HMPA Hexamethylphosphoramide
HPLC High performance liquid chromatography
LSM Lanthanum strontium manganite
MEA Membrane electrode assembly
MFC Mass flow controller
NBO Natural bond orbital
NHE Normal hydrogen electrode
xv
NMP n-Methyl-2-pyrrolidone
OER Oxygen evolution reaction
PEM Polymer electrolyte membrane
PID Proportional-integral-derivative
SCE Saturated calomel electrode
SHE Standard hydrogen electrode
SOEC Solid oxide electrolysis cell
TBAF Tetra-n-butylammonium fluoride
TBAT Tetra-n-butylammonium difluorotriphenylsilicate
TBTCF
2
H Tributyl(difluoromethyl)stannane
TCD Thermal conductivity detector
THF Tetrahydrofuran
TMAF Tetramethylammonium fluoride
TMSCF
3
(Trifluoromethyl)trimethylsilane
TOF Turn over frequency
YSZ Yttria stabilized zirconia
xvi
ABSTRACT
Electrochemical CO
2
reduction is discussed in detail in Chapter 1, from its
beginnings on mercury electrodes in aqueous media to systems producing primarily
carbon monoxide, higher order products, the use of flow cells, the use of high pressure,
developments in molecular catalysis, non-aqueous solvent systems, solid oxide
electrolysis approaches, and direct photochemical reduction to methanol. These widely
varying methods are compared with an eye towards selecting processes that would be
suitable for scale up in the conversion of CO
2
to methanol. Finally, a brief survey of
available electrochemical water splitting technologies is included due to its relevance for
producing methanol from CO
2
.
The design and testing of a novel, high pressure flow electrolysis cell is covered
in Chapter 2. Design of the cell plates themselves is covered in detail, as several
iterations were tested before arriving at a workable solution. Schematic depictions of the
system used for controlling the reaction and connections of all the supporting equipment
are also included. Results obtained from CO
2
reduction to CO and formate at elevated
pressure under flow conditions are then presented, starting with screening conditions,
followed by an in-depth study on the effect of the cation (Na
+
or Li
+
) on CO
2
reduction.
A techno-economic analysis follows, where the approximate cost of syngas and methanol
produced using the above methods are calculated by making various assumptions and
extrapolations.
xvii
A new reagent, tributyl(difluoromethyl)stannane, is synthesized and used for
difluoromethylation of aryl and heteroaryl iodides and β-styrenyl halides with a copper
catalyst in Chapter 3. A novel synthetic procedure for the preparation of difluoromethyl
stannanes from their respective tin hydride precursors using the Rupert-Prakash reagent is
described. The use of these novel compounds for difluoromethylation of aryl and
heteroaryl iodides as well as β-styrenyl halides using copper iodide as a transfer reagent
is then described. Extensive computational modeling of the reaction pathway is given as
well, with a mechanism proposed.
Electronegativity is a particularly important concept in chemistry, but a universal
method to calculate group electronegativity has yet to be accepted. In Chapter 4 Group
electronegativities are derived using simple NBO calculations on structures optimized at
ab initio and density functional theory methods. These group electronegativities are then
compared to those published by others using both experimental and computational
approaches.
1
1 Electrochemical CO
2
Reduction
1.1 Introduction
Electrochemical CO
2
reduction could potentially complete the anthropogenic
carbon cycle by recycling CO
2
from various sources into feedstock materials for fuels
and chemical manufacturing. Today, civilization relies mainly on coal and oil reserves for
fuels and feedstock chemicals, but this cannot continue as the world depletes these fossil
fuel reserves.
1
An anthropogenic carbon cycle, where we are able to make our own fuels
and feedstock materials from CO
2
, has been a long standing challenge and one that
continues to this day.
2-3
The current capacity for industrial CO
2
utilization is exceedingly
small. One reason is that we have relied on abundant, relatively cheap fossil fuel oil
reserves for over 100 years and another reason is that technology to replace oil
economically does not exist yet. Although the Sabatier reaction was developed in 1902
4
to convert CO
2
and H
2
into methane and schemes were developed before the Second
World War for partial synthesis of liquid fuels from alternative sources,
5
we have not
seen a fully sustainable system where all of the CO
2
produced by a given system is
recycled. Of course, the energy for this recycling must come from a source that does not
generate its own CO
2
, therefore we must use either renewable energy (solar, wind,
geothermal, hydro, etc.) or nuclear energy to generate the power to accomplish this
recycling. Our power grid itself still relies heavily on CO
2
-producing sources such as
coal, so the work presented herein only makes sense if its implementation refrains from
using fossil fuel based power sources.
2
For a thorough review of this topic from one of the major researchers in this area,
Hori’s excellent review published in 2008 is available.
6
The field is still very much alive
and well despite the initial work that took place many decades ago.
7
Since much of the
defining work in electrochemical CO
2
reduction was performed quite a while ago, we
will attempt to mix old work with new work to give the reader an idea of where the field
has come and where it is heading. We chose to take this approach in part because many
of the highest performing systems were reported a long time ago,
8
therefore a review of
only current electrochemical CO
2
reduction might miss these important milestones. This
review has been divided into various types of electrochemical CO
2
reduction, with a
section devoted to the mechanism as well. By including both historical and current data
in our sections, we hope to provide the reader with a feel for how varied the field of
electrochemical CO
2
reduction is and the diversity of chemistry that has been applied in
this area of seemingly simple transformation.
Table 1.1: Standard reduction potentials for CO
2
reduction from ref.
9
Half-Cell Reaction E°
CO
2
+ 2H
+
+ 2e
-
HCOOH -0.11
CO
2
+ 2H
+
+ 2e
-
CO + H
2
O -0.10
CO
2
+ 4H
+
+ 4e
-
CH
2
O + H
2
O -0.028
CO
2
+ 6H
+
+ 6e
-
CH
3
OH + H
2
O +0.031
CO
2
+ 8H
+
+ 8e
-
CH
4
+ 2H
2
O +0.17
Depending on the conditions and electrocatalyst, a variety of reduced products
can be generated electrochemically from CO
2
, some of which are presented in Table 1.1.
None of these reactions are well-separated by potential, so the control between various
products must come from choice of catalyst and conditions rather than potential.
3
Although all of the standard potentials shown in Table 1.1 are close to the standard
potential for hydrogen evolution, generation of the first intermediate (carbon dioxide
radical anion, CO
2
•-
) has been estimated to occur only at -2.21 V vs. SCE.
10
Although it
can be stabilized, this high energy intermediate must be produced for CO
2
reduction to
occur on a traditional metal electrode, and the result is that CO
2
reduction generally
requires significant over-potentials to occur at a reasonable rate. Water is typically
present as a proton source (and electrolyte) and thus the competing hydrogen evolution
reaction (HER) must be taken into account while considering CO
2
reduction. For this
reason, many metals which are reported for CO
2
reduction also have relatively high HER
overpotentials. A balancing act must be conducted to find the optimal CO
2
reduction
electrode that will reduce CO
2
selectively at high rates and low overpotentials without
reducing water simultaneously.
We have made as much effort as possible (i.e. reported potentials were converted
to a common reference electrode) to present the data in the following tables in a way that
is useful to the reader, allowing one to compare results from different research groups
based on published results from various sources to generate all the relevant data. The
quality of the published data, however, varied quite a bit and we had to make judicious
choices in selecting relevant and reliable data.
1.2 The mechanism of electrochemical CO
2
reduction on metal electrodes
The mechanism for electrochemical CO
2
reduction has been studied for many
years, with the typical aim of the research being directed towards understanding why
4
different metals give different products.
6-7,11-13
Many groups continue to report interesting
and novel research based on understanding CO
2
reduction on metallic electrodes.
14-20
Scheme 1.1 depicts the various paths that CO
2
reduction can follow based on the
electrode used. The first step, generation of CO
2
•-
, is critical because it is the rate limiting
step and the coordination of this intermediate determines if the 2e- reduction product will
be either CO or formate.
18
The CO
2
•-
intermediate is very high in energy (-2.21 V vs.
SCE)
10
and readily reacts with either water (to form formate or CO) or whatever else is
present in solution, including another CO
2
molecule. Subsequent reduction steps take
place almost instantaneously compared to the first step. Stabilization of this high energy
intermediate, therefore, is a key to achieving a high rate and energy efficient CO
2
reduction process.
Scheme 1.1: Mechanism for electrochemical CO
2
reduction on metal surfaces in water
One can generally divide metal electrodes for CO
2
reduction into three categories
based on their tendency to bind the CO
2
•-
intermediate and whether they are able to
reduce CO. Group 1 consists of those that do not bind the CO
2
•-
intermediate and cannot
5
reduce CO. Group 2 metals bind the CO
2
•-
intermediate, but cannot reduce CO. Group 3
(copper) binds the CO
2
•-
intermediate and can reduce CO. There is also another group of
metals that bind hydrogen strongly, thereby excluding CO
2
reduction in aqueous media.
Group 1 includes metals such as Pb, Hg, In, Sn, Cd, and Tl, whose tendency to bind the
CO
2
•-
intermediate is so low that CO
2
reduction is thought to take place through an outer-
sphere mechanism, generally giving formate (or formic acid) as a product. Group 2 is
composed of metals such as Au, Ag, Zn, and Ga, which bind the CO
2
•-
intermediate to
varying degrees, but cannot reduce CO,
21
so they typically yield CO as the major product
of CO
2
reduction. Copper is the only commonly studied (for CO
2
reduction) metal that
falls into group 3, binding the CO
2
•-
intermediate, and reducing CO
22
to higher reduction
products such as alcohols and hydrocarbons.
23
Examples of metals that do not reduce
CO
2
readily are Ni, Fe, Pt and Ti, which bind hydrogen strongly enough to exclude CO
2
reduction for the most part in aqueous media. Metals from this last group have been
studied for CO
2
reduction under high-concentration conditions,
24-31
but the results are
generally not too promising because CO adsorbs irreversibly on most of them.
6
Although
these classes of metals can provide general guidelines, conditions can significantly alter
the product distribution on a given electrode.
12
This variability indicates that metals have
the ability to reduce CO
2
by multiple mechanisms, probably determined by surface
oxidation
32-35
and crystal planes
36-44
as well as the reaction conditions.
Since CO
2
reduction is typically conducted in aqueous media, one must also
consider HER, which competes with CO
2
reduction and is more facile for the majority of
metals. In fact, many CO
2
reduction catalysts are often chosen not for their ability to
6
catalyze CO
2
reduction itself, but rather because of their high HER overpotentials. Some
of the most commonly used electrochemical CO
2
reduction catalysts for formate are: Sn,
Pb, and Bi, all of which also happen to have significantly high HER overpotentials.
6
One of the many challenges in electrochemical CO
2
reduction has to do with the
interaction between CO
2
and water. When CO
2
is introduced to an aqueous system, a
complex series of reversible reactions are set up, as described in Scheme 1.2. The relative
Scheme 1.2: Equilibrium reactions of CO
2
with water
ratios of these compounds can be adjusted via pH, an increase of which will lead to an
increase in the amount of carbonate species dissolved in solution. While this seems like a
quick and easy way to achieve high CO
2
concentration in aqueous solutions, it has
generally been accepted that dissolved CO
2
(CO
2(aq)
in Scheme 1.2) is the only species
that can be electrochemically reduced on metal electrodes
6-7,11
and not the carbonate
species. A recent spectroscopic investigation by Innocent et al.
14
has suggested that
hydrogen carbonate can also be reduced under certain conditions on lead electrodes. This
is in stark contrast with the mechanism proposed in Scheme 1.1, and is supported by
infrared analysis showing the absence of dissolved CO
2
during formate production
electrochemically. It is unclear, however, if this particular result represents an exception
rather than a rule, since there is so much experimental evidence from years past
6-7,11
that
suggests to the contrary.
7
Electrochemical reduction of CO
2
is therefore a balancing act between finding an
electrode material that is active enough to reduce CO
2
efficiently and at a high rate to a
specific product, while being inactive for competing reactions, primarily HER. The pH of
the solution is also a key concern if one is conducting CO
2
reduction in an aqueous
environment because the pH directly impacts the HER as well as the solubility of CO
2
. A
fundamental understanding of the mechanism for CO
2
reduction on metallic surfaces is
important, because it gives the reader a baseline that can be used to judge the divergent
approaches, which will be presented in the forthcoming sections.
1.2.1 CO
2
reduction efficiency
Most publications related to electrochemical CO
2
reduction make some mention
of the efficiency of the process, but few explain how these values are generated and what
they mean with regard to the viability of electrochemical CO
2
reduction. To begin with,
there are two distinct types of efficiency, faradaic efficiency (FE) and energy efficiency.
Faradaic efficiency is quite simple to calculate and understand: it is the percentage of
electrons that end up in the desired product. If this value was discussed in the context of
synthetic chemistry, it would be called “selectivity” instead of “efficiency”, which may
be why some readers who are not electrochemists by training may misunderstand this
term. Faradaic efficiency is typically calculated by counting the number of electrons that
go into the system during a reaction (easily accomplished using a modern potentiostat)
and comparing this to the amount of desired product produced.
8
Table 1.2: Full cell electrochemical reactions derived from thermodynamic data
45
Reduction Product Full Cell Reaction E (V)
Formic Acid CO
2(aq)
+ H
2
O
(l)
HCOOH
(aq)
+ ½O
2(g)
1.496
Formate CO
2(aq)
+ OH
-
(aq)
HCOO
-
(aq)
+ ½O
2(g)
1.207
Carbon Monoxide CO
2(aq)
CO
(g)
+ ½O
2(g)
1.567
The second type of efficiency is the energy efficiency, which is the amount of
energy in the products divided by the amount of electrical energy put into the system.
Energy efficiency is determined by calculating the enthalpy change for the full cell
reaction from the standard heats of formation and comparing it with the electrical energy
put into the full cell. The enthalpy change divided by the number of Faraday equivalents
yields the thermoneutral potential. Table 1.2 lists three examples of full cell CO
2
reduction reactions, calculated at thermoneutral conditions. Energy efficiency is best
estimated from full cell experiments where CO
2
reduction occurs on one electrode of the
cell, and oxygen evolution occurs on the other electrode of the cell. However, with half-
cell studies, measured electrode potentials are reported vs. a reference electrode.
Therefore, calculations of energy efficiency must rely on estimates of the full cell
potential. Alternatively, the measured half-cell potentials reported vs. the normal
hydrogen electrode can be compared with the thermoneutral potential for the CO
2
reduction half-cell reaction (CO
2
+ H
2
product + H
2
O). In this review, we have
converted the reported electrode potentials to potentials against the standard hydrogen
electrode (SHE) for comparison where possible. While this conversion does not allow for
efficiency calculations, it does allow for comparisons between processes.
9
Although some research groups have assembled full cells and are thus able to
measure the efficiency of their system,
46
these reports are rare and so we have not
attempted to calculate energy efficiency for the majority of reports, which include
information only on the potential of the working electrode. Generally speaking, the lower
the cell voltage, the more energy efficient the process becomes. Delacourt et al.
47
calculated the energy efficiency of their full CO
2
reduction cells, which ranged from 35-
50% depending on the current density. This is in line with the calculated values in Table
1.2 given their cell voltage of approximately 3 to 4.5 V depending on current density.
Since they also report individual electrode potentials, it can be seen that a cathode
potential of approximately -1.2 V vs. SHE gave a full cell energy efficiency of almost
50%, while a cathode potential of -1.8 V vs. SHE gave a full cell energy efficiency of
30%. While these values cannot be translated directly to other systems where only the
potential of the working electrode is given, they can provide a rough guide.
Other practical aspects interfere with calculating energy efficiency. In the case of
photoelectrochemical cells, a calculation of energy efficiency becomes even more
convoluted because some of the energy utilized to undergo the transformation comes
from a light source. A similar situation presents itself when considering solid oxide
electrolysis cells (SOECs, discussed in section 3.8), where some of the energy for the
transformation comes from the heat applied to the system, and some of the energy comes
from the potential applied. Measured potentials are often not corrected for ohmic losses
in the electrolyte, a correction that should ideally be made if the actual electrical
efficiency is desired.
10
1.3 Methods for CO
2
reduction
1.3.1 CO
2
reduction to formate on metal electrodes in aqueous systems
One of the first products identified from the electrochemical reduction of CO
2
on
metal electrodes was formate
7
as it can be observed readily on a mercury electrode.
Throughout the years, this has been the primary product observed on metals which have
very high HER overpotentials such as Pb, Hg, In, Sn, Cd, and Tl.
6
Table 1.3 provides a
selection of results from various research groups
15,48-51
for formate production in aqueous
media on metal electrodes. Of particular note is entry 1 in Table 1.3, which includes one
of the few references describing CO
2
reduction in acidic media. This is surprising
because almost exclusively H
2
evolution has been reported under similar conditions,
6
yet
the authors report very high (97%) faradaic efficiency towards formate coupled with one
of the highest current densities reported for a system at ambient pressure.
11
Table 1.3: Formate production on metal electrodes in aqueous media
Entry Electrode E vs. SHE (V) HCOO
-
FE (%)
Current Density
(mA/cm
2
)
Ref. Electrolyte
1 Pb -2.76 97 115
48
0.35 M Na
2
SO
4
, H
2
SO
4
, pH 2.0
2 Pb -1.63 97.4 5.0
49
0.1 M KHCO
3
3 Hg -1.51 99.5 0.5
49
0.1 M KHCO
3
4 In -1.55 94.9 5.0
49
0.1 M KHCO
3
5 Sn -1.48 88.4 5.0
49
0.1 M KHCO
3
6 Cd -1.63 78.4 5.0
49
0.1 M KHCO
3
7 Tl -1.60 95.1 5.0
49
0.1 M KHCO
3
8 Pb -1.56 39 0.5
50
0.2 M K
2
CO
3
Fixed Bed
9 Pb -1.59 90 2.5
15
0.5 M NaOH
10 Pb -1.64 49 10
15
0.5 M NaOH
11 Sn -1.60 70 27
51
0.5 M NaHCO
3
12 Sn -1.76 63.5 26.7
52
0.5 M KHCO
3
So far, we have considered only bulk metal surfaces for electrochemical CO
2
reduction, but recent studies have provided evidence that the surface oxides play an
important role. Chen and Kanan
19
carefully removed all traces of SnO
x
species from their
electrodes prior to electrolysis under standard (CO
2
saturated solution of NaHCO
3
)
conditions and discovered extraordinarily low CO
2
reduction efficiency of 0 to ~30%
depending on the potential. When they subsequently electrodeposited a mixture of
Sn/SnO
x
, the characteristic faradaic efficiency of tin was restored. The obvious
conclusion from this work is that tin oxides play an important role in CO
2
reduction, and
analogous oxide layers may be important for other metals, especially those which favor
formate production. The effects of the cation and anion species in the electrolyte have
also been studied recently
52
for CO
2
reduction on tin electrodes. Such new findings
highlight just how little we know about the supposedly simple proton coupled two
electron reduction of CO
2
.
12
1.3.2 Carbon monoxide from CO
2
Production of carbon monoxide from CO
2
is a promising pathway given the
robust options for downstream processing of syngas, a mixture of CO and H
2
, using
Fischer-Tropsch (FT) chemistry. Typically, silver and gold are the electrodes of choice
for CO
2
reduction to CO due to their binding of CO
2
and inability to reduce CO.
6
These
metals have higher HER exchange current densities than typical formate forming metals
such as lead and tin.
53
Researchers in this field have generally pointed out that the most
efficient way to utilize a process like this may be to simultaneously produce both CO and
H
2
, which would be fed directly into a downstream processing reactor to generate more
reduced products via traditional FT heterogeneous catalysis.
47,54
For example, if methanol
is the desired end product, a ratio of 2:1 for H
2
:CO (or 33% FE for CO), also known as
metgas, would be desirable since this is the ratio of syngas needed to generate methanol.
Altering this ratio, the catalyst, temperature, etc. for the downstream processing unit
would give access to a wide range of FT hydrocarbon products that are typically derived
from fossil fuels.
13
Table 1.4: Electrochemical reduction of CO
2
to CO on metals under ambient conditions
Entry Electrode E vs. SHE (V) CO FE (%)
Current Density
(mA/cm
2
)
Ref. Electrolyte
1 Au -1.5 50 20
13
0.5 M KHCO
3
2 Au -1.10 80 7
55
0.1 M KHCO
3
3 Ag -1.3 92.3 20
56
0.2 M K
2
SO
4
4 Ag -1.46 64.6 50
56
0.2 M K
2
SO
4
5 Ag -2.96 52.7 100
56
0.2 M K
2
SO
4
6 Ag -1.46 60 50
47
0.5 M KHCO
3
7 Ag -1.56 30 50
17
0.5 M KHCO
3
8 Au -1.91 33 100
17
0.5 M KHCO
3
9 Au/C -2.22 64 200
17
0.5 M KHCO
3
10 Au NP -0.4 98 6
57
0.5 M KHCO
3
11 Ag NP -0.6 92 18
38
0.5 M KHCO
3
12 ACF/Ni
[a]
-1.56 30.3 47
58
0.5 M KHCO
3
13 40 wt% Ag/TiO
2
-1.6 90 100
59
1 M KOH
[a] ACF/Ni = 1300 m
2
g
-1
Ni particles supported on activated carbon fibers
Table 1.4 gives some examples of CO
2
reduction to CO on metal catalysts at
atmospheric pressure. Many of these results are not very impressive from the perspective
of achieving very high faradaic efficiency, but the authors are often not targeting the
highest possible FE, but rather a particular ratio of H
2
:CO that would be useful for
downstream processing. Given that gold and silver electrodes are expected to bind CO
2
to
some degree, one might expect lower overpotentials (and thus greater energy efficiency)
for equivalent CO
2
reduction rates compared to formate-producing electrodes such as
lead and tin. Indeed, under identical conditions, gold electrodes tend to require lower over
potentials than most other metal for a given current density according to Hori et al.
49
They tested a wide variety of metal electrodes and determined that the gold electrode
required only -1.14 V vs. NHE to achieve a partial current density to CO of 4.36 mA/cm
2
,
compared to tin, for example, at -1.48 V vs. NHE to achieve a partial current density to
14
formate of 4.42 mA/cm
2
. While these data suggest that gold has some ability to lower the
barrier for CO
2
reduction, silver does not appear to bind CO
2
as well, with a potential of -
1.37 V vs. NHE to achieve a partial current density towards CO of 4.08 mA/cm
2
.
Recently, several research groups have investigated nanoparticles (both gold
57
and
silver
38
) for CO
2
reduction to CO using both computational
60
and experimental
20,57
methods. As evidenced by Entry 10 in Table 1.4, the faradaic efficiencies and low
overpotentials achievable when using nanoparticles are quite impressive compared to
bulk gold electrodes. High geometric current density is generally challenging to achieve
using nanoparticles as they typically rely on a small absolute quantity of catalyst. One of
the most promising aspects of nanoparticles could be their resistance to poisoning as the
report from Chen and co-workers
57
showed significant improvement in longevity while
using gold nanoparticles as a catalyst compared to bulk gold. Studies of these systems
indicate that there may be a change in the mechanism
57
associated with going from bulk
gold to nanoparticles, although it may have more to do with the ratio of edges to planes
20
than some fundamental size effect. Investigations of single crystals of silver also indicate
that there is a strong dependence on crystal face for reduction efficiency.
36
A major hurdle to commercialization of CO
2
reduction to CO on silver and gold
electrodes comes from the deactivation that they experience in a relatively short period of
time.
47
The cause of this deactivation or poisoning is not entirely understood, however
one of the initial hypotheses is that a small portion of the CO
2
is being reduced all the
way to graphitic carbon and this is being deposited on the catalyst surface.
61-62
This
hypothesis appears to have originated from studies on copper,
63
where black deposits on
15
a copper electrode after electrolysis were confirmed to be graphitic in nature by XPS
analysis. Graphitic deposits have subsequently been observed on gold electrodes.
64
Others
65
have suggested that organic molecules may adhere to the catalyst surface and
thereby alter the product composition. Regardless of the root cause of the poisoning
observed on silver and gold electrodes, it is an issue that must be addressed if these
electrode materials are to be used for extended periods in practical applications.
Although gold and silver electrodes may not be the most efficient for generating
H
2
, combining two processes, CO synthesis and H
2
synthesis, in one reactor may be a
valuable advantage industrially. If the overall goal of CO
2
reduction is to recycle CO
2
into useful hydrocarbon fuels, the CO pathway may be the most promising method
currently available due to the combination of relatively high energy efficiency coupled
with relatively high current density.
1.3.3 Higher order products from CO
2
reduction
Although carbon monoxide and formate can be potentially used as feedstock
chemicals for further processing,
1
there is significant interest in conditions which lead to
further reduction of CO
2
in the research community.
42,66-74
One of the simplest
approaches to produce highly reduced products from CO
2
is to use copper as a cathode,
which is described in a recent review.
75
Table 1.5 highlights a few of the contributions
from various groups, dating back to 1983,
42
when it was discovered that semiconductors
had the potential to convert CO
2
into methanol, albeit at very low current densities and
faradaic efficiencies. Many of these reports also involve irradiation of some sort,
32,68
16
which provides a way to reduce the applied cathode potential, but complicates
comparison to other processes. We have not attempted to account for the solar energy
used in Table 1.5, so the results presented provide the reported potentials applied to the
cells. We have chosen to focus on reports of relatively selective CO
2
reduction to higher
products since it seems that generation of a wide variety of reduced CO
2
products would
further complicate downstream processing rather than streamlining it.
Table 1.5: Higher Products from CO
2
Reduction
Entry Electrode
E vs. NHE
(V)
Major
Product
FE (%) J (mA/cm
2
) Ref. Electrolyte
1 n-GaAs -1.06 MeOH 1 0.16
42
Sat. Na
2
SO
4
2 p-GaAs -1.06 MeOH 0.52 0.08
42
Sat. Na
2
SO
4
3 p-InP -1.06 MeOH 0.8 0.06
42
Sat. Na
2
SO
4
4 Mo -0.56 MeOH 84 0.12
67
0.2 M Na
2
SO
4
5 CuO -1.3 MeOH 28 6.9
72
0.5 M KHCO
3
6 CuO -1.60 C
2
H
4
30 n/a
a
69
0.1 M KHCO
3
7 CuO -2.00 CH
4
55 n/a
a
69
0.1 M Na
2
SO
4
7 Pt-Ru/C -0.06 MeOH 7.5 0.4
76
Flow Cell
8 Pd/Pyridine -0.51 MeOH 30 0.04
66
0.5 M NaClO
4
+
Pyridinium
9
n-GaP/
Pyridine
-0.06 MeOH 90 0.27
68
10mM Pyridinium pH
5.2
10 Polyaniline -0.16
CH
3
COO
H
57 10.7
77
0.1 M LiClO
4
in
MeOH
b
11 Cu -2.80 CH
4
20 15
78
0.5 M LiCl in MeOH
c
12 La
1.8
Sr
0.2
CuO
4
-1.96 EtOH 16.4 180
79
0.5 M KOH
13 Cu -1.61 CH
4
19 20
80
0.5 M KHCO
3
Copper (and its oxides)
35
can electrochemically reduce CO to higher order
products such as methane and even form C-C bonds, but a wide range of products are
typically formed on copper electrodes.
81-82
Schouten and co-workers
37
have recently
focused on understanding the mechanisms by which various products are formed on
17
copper by electrochemically reducing CO instead of CO
2
. They have determined that,
perhaps not surprisingly, the crystal face of the copper cathode has a significant effect on
the nature of the product, which is a conclusion supported by other researchers.
39-41
The
effect of the crystal face has even been studied by computational chemistry.
44
A model
has also been developed
83
to calculate the concentration of CO
2
and H
+
at the copper
surface given the stirring rate, buffer capacity, and current density. The effect of the
electrolyte on CO
2
reduction using a copper electrode has also been examined,
84
with the
conclusion being that KHCO
3
generally leads to high currents compared to NaHCO
3
.
Mixing copper with another metal
85
or a metal oxide
43
has also been demonstrated to
potentially boost production of alcohols from CO
2
, with studies on the effects of copper
oxides supporting this conclusion.
33
Conversely, earlier studies pointed to copper oxides
being a cause for enhanced HER and thus negatively impacting CO
2
reduction.
86
Such
contradictory results suggest that the processes are complex and poorly understood.
Another approach that has received increased attention in recent years
87-91
has
been the use of pyridinium as a co-catalyst for CO
2
reduction, primarily to methanol. This
approach is represented by entries 8 and 9 in Table 1.5 and represents an exciting new
avenue for electrochemical reduction of CO
2
to higher products. Entry 9 is a
photoelectrochemical process, so the applied potential is even lower than the standard
potential for the half-cell reaction. These reports may be a better fit for the molecular
catalysis section (vide infra), but pyridinium is such a simple molecule compared to the
catalysts typically considered in that section that it seemed better to include the reports
18
here. Also, there generally seems to be a fairly strong link between the electrode and the
catalyst.
One of the major hurdles still to be overcome has to do with the poisoning effects
often observed on copper electrodes.
63,92-95
The cause of this poisoning was originally
thought to be deposition of graphitic carbon on the electrode surface,
63,92
however more
recent studies indicate that it may be a result of unwanted metal ions (such as iron)
present in solution.
93
Either way, poisoning is a serious concern if the aim for this
technology is commercialization. Although carbon monoxide and formate can be used as
chemical feedstock materials to generate other materials, the ability to go further than the
2 electron reduction stage in one step efficiently would be a major breakthrough, possibly
allowing electrochemical CO
2
reduction to be competitive with other technologies in the
marketplace.
1.3.4 Electrochemical reduction of CO
2
in flow cells
For electrochemical CO
2
reduction to become an industrially viable process, it
must move out of the lab setting where batch electrochemical cells are used to conduct
experiments and into flow cells that can be scaled up to match the other industrial
processes.
96
Several research groups have attempted this transformation,
15,46-47,76,97-103
typically using experimental setups similar to the one pictured in Figure 1.1.
97
Although
almost all of the research groups use different setups, many of them use a buffer layer to
augment a membrane for their electrolyte because of the strong effect of the membrane
on the efficiency and selectivity of the reaction.
104
Delacourt et al.
47
compared the
19
effectiveness of several configurations, some with a buffer and some without, and came
to the conclusion that a buffer is almost essential to achieve decent product selectivity in
a flow cell.
Figure 1.1. Schematic of an electrochemical flow cell with buffer layer for CO
2
reduction
Given the drastically different cell design compared to traditional electrochemical
cells, one cannot expect the results to be directly comparable to those presented in Tables
1.3 and 1.4. Some of the more recent results utilizing this approach are presented in Table
1.6 and give good current density and faradaic efficiency results compared to results at
ambient pressure in batch electrolysis cells, although the batch results
17-18
from Delacourt
20
et al. were better than their flow results.
47
It should be noted that most of the results in
Table 1.6 report cell potential, rather than electrode potential vs. NHE, so the voltage
values are generally higher. If improved performance has to do with improved mass
transport or other phenomenon related to the flow cell design is beyond the scope of this
review and, to the best of our knowledge, this has not been extensively investigated by
any group to date.
Table 1.6: Electrochemical flow reactors for CO
2
reduction
Entry Electrode
Cell Potential
(V)
Current Density
(mA/cm
2
)
Major
Product
FE (%) Ref. Electrolyte
1 Sn -3.75 160 HCOO- 80
46
0.5 M KCl/HCl
2 Pb -1.59
a
2.5 HCOO- 90
15
0.5 M NaOH
3 Pb -1.64
a
10 HCOO- 49
15
0.5 M NaOH
4 Sn -1.7 10 HCOO- 90
97
0.1 M KHCO3
5 Pt-Ru/C -0.06
a
0.4 MeOH 7.5
76
Nafion MEA
6 Pb -2.2 10 HCOO- 25
b
98
1.0 M NaHCO
3
7 Ag -3.5 80 CO 33
47
0.5 M KHCO3
8 Sn -2.7 60 HCOO- 91
101
0.45 M KHCO
3
+ 2 M KCl
9 Sn -4.1 290 HCOO- 64
101
0.45 M KHCO
3
+ 2 M KCl
[a] Cathode potential vs. NHE. [b] Current not stable
Flow cell CO
2
electrolysis deserves a significant amount of attention from the
general research community because it is likely the most economical method for
electrolysis, and the electrochemical environment is typically quite different from
traditional electrolysis cells. Whether these designs can be scaled up to industrially
relevant sizes has been studied extensively by Oloman and Li,
96,99-101
and there does
appear to be a much greater chance compared to scaling up the typical glass
electrochemical cells that are used for so many of the investigations reported in this
21
review. Another issue that flow electrolyzers like the ones described here have to deal
with is the oxygen evolution reaction (OER), which must occur in tandem with CO
2
reduction. Because the cathode and anode are matched in a flow cell, the energy
efficiency of the whole system can often be easily obtained. This information is
extremely useful to determine the feasibility or cost of the process as a whole because
approximations related to the anode do not have to be made.
1.3.5 High pressure electrochemical CO
2
reduction
One of the major issues often described for electrochemical CO
2
reduction in
aqueous media has to do with the lack of solubility of CO
2
in water at ambient pressure
and temperature. CO
2
saturated water only contains 0.034 M CO
2
at standard temperature
and pressure.
105
A simple solution to resolving this problem is to increase the pressure of
CO
2
in the system to raise the concentration of CO
2
according to Henry’s law while
maintaining an aqueous solvent environment. Significant effort was made towards
optimizing this approach for electrochemical CO
2
reduction in the 90s. Despite the
relatively vintage age of these reports, they contain some of the highest current densities
8
and CO
2
reduction efficiencies ever reported in the literature, thus they have been
included in Table 1.7 as a benchmark for some of the newer approaches to CO
2
reduction. Despite their propensity to evolve H
2
, even platinum, palladium, and rhodium
can be made to yield CO
2
reduced products at high pressure, albeit at moderate
efficiencies compared to metals like lead, tin, and indium.
24
22
Table 1.7: High pressure CO
2
reduction
Entry Electrode
Pressure
(atm)
E vs. SHE
(V)
Current Density
(mA/cm
2
)
Major
Product
FE (%) Ref.
1 Pt 30 -1.28 163 HCOO- 50.4
24
2 Cu 30 -1.44 163 HCOO- 53.7
24
3 Hg 20 n/a 200 HCOO- 100.9
a
106
4 In 60 n/a 200 HCOO- 107.6
a
106
5 Sn 30 -1.19 163 HCOO- 92.3
24
6 Pb 30 -1.37 163 HCOO- 95.5
24
7 Pb 60 n/a 200 HCOO- 100.4
a
106
8 Pb 50 -1.56 0.7 HCOO- 88
50
9 Bi 30 -1.22 82.7 HCOO- 82.7
24
10 Ag 20 -1.02 300 CO 86.0
8
11 Pd 20 -1.13 300 CO 57.5
8
12 Rh 30 -1.21 163 CO 61.0
24
[a] The origin of FE values >100% is not explained in the papers referenced
Typically, a high pressure autoclave, like the one depicted in Figure 1.2 is used
for high pressure studies of CO
2
electrolysis. Since these high pressure devices are almost
always fabricated out of metal, several considerations must be made to ensure proper
operation for electrolysis experiments. Insulated wires must be fed through the autoclave
and attached to the electrodes housed inside using gas-tight fittings rated for high
pressure. A glass or Teflon liner must also be used to insulate the electrolyte from the
metal autoclave. Magnetic stirring is also critical since these experiments are intended to
overcome some of the mass transport limitations inherent to CO
2
electrolysis and the
experiments are generally carried out at static pressure, i.e. without bubbling gas to
provide some degree of mixing.
23
Figure 1.2. Typical high pressure reactor modified for CO
2
electrolysis
One generally expects to see similar behavior to that presented in Figure 1.3 when
increasing the pressure of CO
2
in a system. That is, with increasing CO
2
concentration,
CO
2
reduction increases as well. This is not surprising, however it is interesting to note
that the product distribution between various CO
2
-reduced products can change as well.
In fact, some very interesting relationships
107
have been noted on copper electrodes with
respect to CO
2
pressure and current density.
24
Figure 1.3. The dependence of the faradaic efficiencies of reduction products on the CO
2
pressure in the electrochemical reduction of CO
2
on a Ag-GDE at 300 mA cm
-2
in 0.5
mol dm
-3
KHCO
3
: ( ) CO, ( ) HCOO
-
, ( ) H
2
. Reproduced from ref.
8
High pressure CO
2
reduction stands out as being one of the most feasible methods
for achieving a commercial electrochemical process given the high current densities and
efficiencies observed on a variety of electrodes. Despite the relatively simple methods
and use of normal metal electrodes, researchers have been able to demonstrate multiple
amperes of current per square centimeter,
8
something which has only recently been
replicated with high temperature solid oxide electrolyzers (discussed in section 3.8).
These methods deserve attention because of their ability to reduce large quantities of
CO
2
, albeit in a lab scale apparatus.
1.3.6 Molecular catalyst assisted CO
2
reduction
The use of molecular catalysts to aid in the reduction of CO
2
is a relatively old
concept
108-109
which has become quite popular lately.
110-123
Dubois et al.
124
reviewed this
25
topic in much more detail than what has been covered here. We have looked at molecular
electrocatalysts for CO
2
reduction from a broader perspective and in comparison to other
methods of electrochemical CO
2
reduction. The promise of molecular catalysis is that one
can achieve high selectivity and low potential relative to the reduction on a metal surface.
Typically, the molecular catalyst is expected to undergo a fairly facile reduction at an
otherwise inactive (towards CO
2
reduction) electrode, and then chemically reduce CO
2
on
its own. Several groups have set about determining the mechanism of their homogeneous
molecular CO
2
reduction electrocatalysts,
109,111,113,115-117,124-128
but such discussions are
beyond the scope of this section. There are also reports in the literature of covalent
attachments
112
made between electrodes and molecular catalysts to achieve a stronger
interaction between the catalyst and the electrode. Others have chosen to use insoluble
catalysts, which are then physically connected to an electrode surface.
129-130
A small but
diverse selection of results from these widely varying reports using such molecular
catalysts have been collected and represented in Table 1.8.
26
Table 1.8: CO
2
reduction using molecular catalysts
[a] E vs. carbon electrode.
Entry Catalyst E vs. SHE (V)
Current Density
(mA/cm
2
)
TOF (s
-1
)
Major
Product
FE (%) Ref. Notes
1 Mn(bipyridyl)CO
3
Br -1.7
a
0.6 CO 85
110
0.1 M TBAP in 5% H
2
O in MeCN
2 Ir-pincer -1.41 0.6 HCOO- 93
117
1% MeCN in 0.1 M NaHCO
3
3 Co-TPP -0.90 100 CO 74.8
129
0.5 M KHCO
3
4 Co-TPP -0.88 100 CO 97.3
129
20 atm CO
2
, 0.5 M KHCO
3
5 Fe-TPP -0.84 100 CO 42.4
129
0.5 M KHCO
3
6 Fe-TPP -0.92 100 CO 80.9
129
20 atm CO
2
, 0.5 M KHCO
3
7 Co-Tetraaza -1.26 2.5 e-3 CO 45
108
0.1 M KNO
3
in H
2
O:MeCN 2:1
8 Ni-Tetraaza -1.36 1.7 e-3 CO 65
108
0.1 M KNO
3
in H
2
O:MeCN 2:1
9 Polypyridyl Ru -1.52 1.3
CO/
HCOO-
n/a
115
0.1 M nBu
4
NPF
6
in MeCN
10 AgDAT/C -1.4 34 CO 70
131
Flow Cell
11 FeTDHPP -1.16 0.31 1000 CO 94
118
2 M H
2
O in DMF
12 Ni(cyclam) -1.21 1.8 90 CO 90
132
0.8 M tetrabutylammonium PF
6
27
A great deal of diversity
108,110,115,117-118,129,131,133
is present in molecular catalysts
for electrochemical CO
2
reduction, only a few of which are presented in Figure 1.4.
Caution should be exercised while comparing values from Table 1.8 to each other and to
other types of electrochemical CO
2
reduction because they are fundamentally different in
that they operate indirectly on CO
2
rather than directly as metal electrodes do. Still, the
output of a given systems is definitely comparable if the goal is to produce reduction
products from CO
2
. Typically, systems involving homogeneous molecular catalysts
struggle to achieve high current densities, where they cannot compete with metallic
catalysts for the most part. Some research groups
129,134
have immobilized molecular
catalysts on solid supports, entries 3-6 for example, which gave high current densities.
Figure 1.4. Molecular catalysts for electrochemical CO
2
reduction
28
There is a great deal of attention being paid to molecular catalysts for CO
2
reduction with good reason. They hold the promise of improving two of the major
stumbling blocks present for CO
2
reduction in general: to decrease the overpotential
required for CO
2
reduction and to increase the selectivity towards the desired product. It
seems that if the selectivity and catalytic abilities of these molecules could be combined
with a solid metallic catalyst, we could solve the major problems now facing
electrochemical CO
2
reduction. Many of these catalysts systems can be expensive,
however.
1.3.7 Non-aqueous solvent systems for CO
2
reduction
Like most approaches to electrochemical CO
2
reduction, non-aqueous solvents
have been explored extensively. Typically, one expects to utilize the non-aqueous system
to increase the solubility of CO
2
compared to water. At least one research group has even
gone so far as to use supercritical CO
2
as a medium for CO
2
reduction,
135
which would
seem to alleviate any potential concentration concerns. Methanol, for example, dissolves
>4 times as much CO
2
at ambient conditions as water,
105
therefore one can think of a
non-aqueous solvent as an alternative to using high pressure. Saeki and co workers have
tested a variety of metal electrodes for their CO
2
reduction properties in methanol.
136
They have reported extremely high current densities on a copper electrode in methanol,
81
however they have not included the electrode potential of their process. The use of
methanol as a solvent for electrochemical CO
2
reduction has been studied extensively by
29
Kaneco and co workers.
34,78,137-143
Low temperature,
137,139-140,144-145
various cathode
materials,
34,142-143
and the effect of the cation
137,141
have all been investigated.
Alternatively, use of non-aqueous solvent systems for CO
2
reduction can be
thought of as similar to molecular catalysis in some respects because the interaction
between the solvent and the CO
2
molecule can be thought of as a type of catalysis to
lower the activation energy for CO
2
reduction.
146
Generally, however, water must still be
present in the system or else the products will differ drastically from the aqueous case.
12
Table 1.9 provides a summary of some results from the literature over the years with
various non-aqueous solvents.
12,77-78,146-149
Table 1.9: CO
2
reduction in non-aqueous solvents
Entry Electrode
E vs.
SHE (V)
Current Density
(mA/cm
2
)
Solvent
Major
Product
FE
(%)
Ref. Notes
1 Pb -2.4 n/a
Propylene
Carbonate
H
2
C
2
O
4
73.3
12
0.1 M TEAP
2 In -2.4 n/a
Propylene
Carbonate
CO 85.3
12
0.1 M TEAP
3 p-InP -1.9 50 MeOH CO 60
147
Photo
4 p-InP -2.7 100 MeOH CO 93
147
Photo, 40 atm
5 p-GaAs -2.7 100 MeOH CO 82
147
Photo, 40 atm
6 p-Si -2.3 50 MeOH CO 75
147
Photo, 40atm
7 Cu -2.8 15 MeOH CH
4
20
78
0.5 M LiCl, 10
atm
8 Polyaniline -0.16 10.7 MeOH CH
3
COOH 57
77
0.1 M LiClO
4
,
20 atm
9 Cu -3.5
a
70 MeOH CO 84
137
0.5 M CsOH, 10
atm
10 Cu -1.4
b
20 Bmim-PF
6
CO 50
148
104 atm
11 Ag -2.0
c
0.5
18% Emim-
BF
4
in H
2
O
CO 96
146
Electrochemica
l surface area
12 Bi-CMEC -1.71 5.8
Bmim-BF
4
in MeCN
CO 95
149
[a] Ag wire quasi-reference electrode [b] Pt wire quasi-reference electrode [c] Cell
voltage
30
One observation that stands out in Table 1.9 is that the favored product in non-
aqueous systems appears to be CO, regardless of the electrode material. This is partially a
function of the lack of water in most of these systems, which precludes the possibility of
producing formate efficiently on electrodes that would otherwise favor formate
production (such as lead or indium in entries 1 and 2). Entries 9-11 utilize ionic liquids,
which are generally nitrogen-containing salts and bend the CO
2
molecules in solution
according to IR data,
150
thereby facilitating electrochemical CO
2
reduction at decreased
electrode potentials. In fact, one might question our inclusion of Entry 11 in this table
since the primary solvent is water, so the ionic liquid used can be thought of as a catalyst
or electrolyte. Scheme 1.3 gives a representation of an interaction between a popular
ionic liquid anion, BF
4
-
and CO
2
, where the CO
2
molecule interacts with the slightly basic
BF
4
-
to generate a slightly bent intermediate complex that would then undergo
electrolysis. Snuffin et al. have tried CO
2
reduction with the BF
3
Cl
-
anion in order to test
the hypothesis that Cl
-
could leave, giving the Lewis acid BF
3
, which would then form
the BF
3
-CO
2
Lewis acid-base adduct.
151
Scheme 1.3: Possible interaction between CO
2
and BF
4
-
One of the major hurdles to adoption of non-aqueous systems for CO
2
reduction is
cost, since the vast majority of non-aqueous solvents are much more expensive than
water itself, especially if one uses ionic liquids. Of course, since these solvents are not
31
consumed in the reaction, they can be treated as catalysts, which would be recycled
numerous times or used continuously in a flow system.
1.3.8 Solid oxide approaches to CO
2
reduction
Figure 1.5. Diagram of solid-oxide three electrode cell
Using a solid oxide electrolysis cell (SOEC) to reduce CO
2
to CO is
fundamentally different from any of the previously discussed methods because the
temperatures used (typically ~800-900 °C) present the possibility for both chemical
reduction of CO
2
as well as electrochemical reduction of CO
2
. SOECs
152-153
are
composed of three parts: a high temperature ion conducting electrolyte (typically yttria
stabilized zirconia, or YSZ), a cathode (traditionally Ni-cermets, or ceramic-metallic
materials), and an anode. Figure 1.5 represents a typical setup for a single, three electrode
32
solid oxide test cell. The process involves dissociation of CO
2
into CO and oxygen atoms,
which migrate through the electrolyte to the anode, where they are combined to form O
2
.
The attraction towards CO
2
reduction in SOECs has to do with the energy efficiency of
the process
154
as well as the very high current densities that SOECs can support (see
Figure 1.6).
Figure 1.6. I-V curves from an SOEC with a LSCM/GDC cathode operated at 900 °C at
various atmospheres, reproduced from ref.
155
One of the first reports comes in the form of a regenerative fuel cell which was
proposed for a Mars exploration mission,
156
where CO was oxidized to CO
2
to generate
power at night, and CO
2
was reduced to CO to store power during the day. This process
was originally envisioned due to the high concentration of CO
2
in the Martian
atmosphere. One of the first in-depth reports on co-electrolysis of steam and CO
2
came in
2006 from Idaho National Lab.
157
Several research groups familiar with solid oxide
systems have now experimented with CO
2
reduction,
154-155,157-160
and their results have
been quite encouraging. Most of the work in this area has focused on attempting to co-
33
electrolyze CO
2
and steam to produce both CO and H
2
simultaneously. This approach is
ideal from the standpoint of downstream processing, where the product could be directly
fed into a Fischer-Tropsch reactor
158
to produce higher hydrocarbon products or
methanol.
Scheme 1.4: Water-gas shift reaction
The chemical components present in a solid oxide electrolysis cell are: CO
2
, H
2
O,
H
2
, and CO, which sets up the possibility for both the water gas shift reaction and the
reverse water gas shift reaction, shown in Scheme 1.4. As a result of this equilibrium, it
becomes difficult to calculate Faradaic efficiency, which we generally associate with
traditional electrochemical CO
2
reduction, because the CO which has been produced
could be a result of the direct electrochemical reduction of CO
2
, or from the chemical
reaction of CO
2
with H
2
. This problem has not presented itself previously in our
discussions of CO
2
reduction in the presence of H
2
because the temperatures that have
been previously reported are close to room temperature. Therefore, the water gas shift
equilibrium reaction does not take place to any significant degree. Models have been
created to help understand the performance of these complex systems,
154,161-162
and they
generally appear to agree that a significant portion of CO
2
reduction is due to the reverse
water gas shift reaction.
CO
2
reduction in an SOEC is drastically different from CO
2
reduction in a
conventional electrochemical cell. Therefore, we attempted to convert the data presented
in some of the papers cited above into a form which could be used for comparison with
34
traditional electrolysis systems. This endeavor is not straightforward, however, since
most of the reports of CO
2
reduction in SOECs add some combination of H
2
, CO
2
, H
2
O,
and sometimes even CO to the inlet of their reactors. Thus, calculating meaningful FEs
required subtracting out the inlet composition to generate a “net” FE value. Since there
are multiple reactions occurring on these surfaces, these values should also be taken as
“faradaic efficiency equivalents” because a percentage of the CO produced may come
from the reverse water-gas shift reaction, rather than direct electrochemical reduction of
CO
2
. The net result is the same, however, so we hope that Table 1.10 will be useful for
the reader.
Table 1.10: CO
2
reduction using SOECs
Entry Cathode T (°C)
Cell
Potential
(V)
Feed
(H
2
:CO
2
:H
2
O:CO:Inert)
CO FE
Equivalent
Current
Density
(mA/cm
2
)
Ref. Notes
1 Ni/ScSZ 800 1.15 7:14:19:0:61 50 230
154
Button Cell
2 Ni/Sc/SZ 800 1.4 12:12.7:11.4:0:63.3 62 188
154
Stack
3 Ni/YSZ 800 1.3 1:1:2:0:0 n/a 1000
160
4 LSCM/YSZ 900 2.25
a
0:9:0:1:0 n/a 540
155
Three
Electrode
5 LSCM/GDC 900 2.25
a
0:9:0:1:0 n/a 610
155
Three
Electrode
6 Ni/YSZ 900 2.25
a
0:9:0:1:0 n/a 800
155
Three
Electrode
7 Ni/YSZ 850 1.25 0:1:1:1:1 n/a 900
159
8 BCZYZ 614 2.65 0:1:0:0:0 29.5 1500
163
Proton
Conductor
[a] E vs. LSM/Scandium stabilized zirconia quasi-reference electrode
Faradaic efficiencies could not be calculated for Entries 4-7 because the product
compositions were not given in the original papers. It is important to check the
composition of gas at the outlet of the reactor, because over-reduction of CO
2
to soot
35
(carbon black) could take place at elevated temperatures and potentials.
154
In addition, we
could not calculate a faradaic efficiency value for Entry 3 because the amount of H
2
consumed during the course of the reaction was more than that produced for all their data
points, so the resulting calculation would produce a negative faradaic efficiency value.
This means that one must supply hydrogen gas from some other source, so it’s unclear
how this could be accounted for in terms of faradaic efficiency. Entry 8
163
is the only
example that we have come across in the literature of using a high temperature proton
conducting membrane for CO
2
electrolysis.
SOECs represent an exciting new path to CO
2
reduction by combining aspects of
chemical CO
2
reduction with electrochemical CO
2
reduction. Determining how
competitive SOEC technology is with traditional electrochemical and chemical methods
would require a far more in-depth analysis than what we have provided here, however.
This technology is being actively studied by industry, particularly Haldor Topsoe,
164
which claims that SOECs for CO
2
reduction can proceed far more efficiently than in
PEM type electrolyzers. They have presented data which suggests that, at 950 °C, current
densities of ~1.5 A/cm
2
can be achieved at ~1.3 V, although full details of their
experiments are not available. SOECs appear to also have inspired others to experiment
with “intermediate” temperature (~300 °C) materials for CO
2
electrolysis,
165
where the
possibility to form hydrocarbons exists. One drawback to this technology is that the cells
can generally only sustain a limited number of stop-start cycles.
166
While this may not
prove to be a significant limitation compared to other systems since all chemical reactors
will show some degradation with many start-stop cycles, it could complicate practical
36
applications where energy is coming from intermittent sources such as solar and wind
power.
1.3.9 Direct photochemical reduction of CO
2
to methanol
Direct photochemical reduction of CO
2
to methanol shares some similarities with
electrochemical CO
2
reduction, particularly when it comes to molecular catalysts used in
both cases. One recent report in particular cites the pyridinium approach to
photoelectrochemical CO
2
reduction, but instead uses ruthenium(II) trisphenanthroline as
a chromophore to generate the electrons necessary for CO
2
reduction instead of a
semiconductor.
167
Although the selectivity for methanol is quite low, the conversion of
CO
2
directly to methanol using photochemistry has been demonstrated. Typically,
photochemical (or photocatalytic) CO
2
reduction gives either formate or CO as a major
product, an area which was reviewed in 2010.
168
One of the major limitations to
photochemical CO
2
reduction as it is generally practiced is that a sacrificial hydride
source (generally an amine, ascorbic acid, or 1-benzyl-1,4-dihydronicotinamide) must be
added to the solution to substitute for the anode which would generally be used in
electrochemical CO
2
reduction. While several groups are actively working on
photochemical regeneration of hydride donor,
169
it remains a major challenge. Turnover
frequency and turnover number for photochemical catalysts for CO
2
reduction are still
limited (TOFs are generally around 20 hr
-1
), and will require significant improvement if
these processes are to be considered for industrial application. In addition to the
molecular catalyst approach, several research groups have experimented with
37
semiconductors and metal oxides such as silicon carbide,
170
WO
3
,
171
InVO
4
,
172
TiO
2
,
173-
176
InTaO
4
,
177
NiO,
176
and ZnO
176
either by themselves or in combination with various
other heterogeneous catalysts to achieve the same goal. Contamination of semiconductors
can cause spurious results,
178
however, so meticulous cleaning of the surface is essential
to obtaining reproducible results. A challenge to producing methanol on semiconductors
irradiated with light is that the reaction can be reversible,
170,176
so strategies to mitigate
methanol oxidation are essential to realizing an industrial process.
1.4 Electrochemical water splitting
Acid
Cathode
Anode
Overall
Base
Cathode
Anode
Overall
Figure 1.7. Water electrolysis in acid and base
Electrochemical water splitting is a critical technology that impacts most of the
anthropogenic carbon cycles in some way. Water electrolysis takes place according to the
equations in Figure 1.7 depending on if the media is acidic or basic. The overall reaction
is the same regardless of the pH of the solution, however. Ursua et al. recently reviewed
this field comprehensively,
179
so we will give only an overview here. There are three
types of water electrolysis based on the type of electrolyte: alkaline, PEM, and SOEC.
Alkaline electrolyzers (Figure 1.8) are by far the most mature of the three, with the first
demonstration in 1800,
180
and virtually all of the large scale electrolysis facilities utilize
38
this technology. The main advantages that alkaline electrolysis cells have over competing
technologies are maturity, low installation cost, large scalability, and long lifetime. Since
alkaline electrolysis is the oldest technology, everything else must demonstrate
improvements relative to it.
Figure 1.8. Alkaline electrolysis cell
The alkaline electrolyzers on the market today are mostly considered “advanced
alkaline electrolyzers”,
181
which feature several design improvements including
minimized distance between the electrodes,
182
new separator materials,
183-184
higher
temperature of operation, and new electrocatalysts. Very high hydrogen production rates
can be achieved using alkaline electrolyzers, with power demands reaching into the
OH
-
OH
-
OH
-
OH
-
+
H
2
O
H
2
O
H
2
H
2
H
2
H
2
O
H
2
O
O
2
O
2
anode
cathode
diaphragm
oxygen
evolution
hydrogen
evolution
alkaline
solution
alkaline
solution
39
megawatt range. A concern with alkaline electrolyzers is that they cannot be quickly
switched on and off, which is problematic if they are to be coupled with fluctuating
energy generation from wind, solar, or even in an environment when they must only be
used during low energy demand.
PEM electrolyzers are available commercially,
179
often from the same companies
that produce alkaline electrolyzers. However, they struggle with high initial cost and
short lifetime compared to alkaline electrolyzers. Their main advantages include the
production of higher purity gases (limiting the need for purification), high current
density, high voltage efficiency, and rapid response to changes in load.
185
These
advantages are offset by their high cost, shorter lifetime, and smaller size compared to
alkaline electrolyzers. The cost of a PEM stack is broken down in Figure 1.9 with the
largest single portion going to the flow fields and separators and the next largest one to
the MEA itself. These values are high due to the specialized materials required for
operation within the acidic environment of the PEM electrolyzer. Conversely, most of the
advantages of PEM electrolyzers (high current density, high voltage efficiency, and rapid
response) are related to their acidic nature.
40
Figure 1.9. Capital Cost Breakdown of PEM Electrolyzer.
186
SOECs operate by utilizing materials capable of transporting oxygen atoms
through solid ceramic materials at very high temperatures (see Section 1.3.8 vide supra).
These designs offer very high efficiencies and current densities and are potentially much
cheaper than PEM cells due to the use of non-noble metal catalysts, but they are still in
the process of being commercialized. The very high temperatures they require constitutes
both a strength and a weakness, as it increases the electrical efficiency to values much
higher than those achievable with either alkaline electrolyzers or PEM cells, but induces
many problems in the form of material durability and compatibility at such high
temperatures. Another disadvantage of SOECs is their inability to tolerate many start/stop
cycles due to the strain to internal components caused by varying degrees of thermal
expansion at high temperature.
166
Also, their high temperature of operation indicates that
they likely would only operate efficiently within a narrow load range, much like alkaline
electrolyzers. The ability to use waste heat makes this technology very attractive to
nuclear power, however, where the excess heat generated by the reactor could be used to
Flow
Fields and
Separators
48%
MEA
24%
Balance of
Stack
23%
Balance of
Cell
5%
41
heat the SOEC and thereby significantly increase the overall efficiency of a nuclear
power plant.
1.5 Summary
The understanding of the processes within electrochemical reduction of CO
2
has
come a long way since its early days,
7
and we now have processes which may one day be
commercialized. For such a simple molecule, seemingly innumerable conditions have
been applied to realize its reduction to useful products. While we now know much more
about the factors affecting electrochemical reduction than we did decades ago, there is
still uncertainty in the pathways for reduction, especially for those reductions which
proceed past the 2 electron point. Despite all of the effort put into electrochemical CO
2
reduction, there are still many problems with the current catalysts and conditions which
must be addressed for these technologies to achieve widespread use:
Heterogeneous electrode catalysts which significantly decrease the energy barrier
for CO
2
reduction must be identified
The mechanisms that give rise to higher reduction products must be fully
understood and applied to heterogeneous electrode catalysts
Poisoning must be further studied and suppressed to the lowest levels possible
Solid oxide CO
2
electrolyzers must be studied further for their durability, the
active mechanisms, and the overall energy input compared to traditional
electrochemical cells
42
While there are still challenges with regard to cathode development for CO
2
reduction, the anode also requires attention. Although clearly not the focus of this review,
a large portion of the voltage drop in a CO
2
electrolyzer occurs due to the oxygen
evolution reaction, which has a standard potential of 1.229 V.
45
Also, the oxygen
produced cannot be readily utilized. This problem is being tackled by many groups
185
because of its application to water electrolysis, but a clear solution has been elusive so
far.
Given the world’s increasing consumption of fossil fuels and the additional
anthropogenic CO
2
that they create in our atmosphere, the need for efficient CO
2
electrolysis is imminent. Compared to other technologies for CO
2
mitigation, such as
sequestration for example, electrolysis provides us with the possibility to close the loop
and generate useful products from CO
2
. Assuming that these products can be turned into
fuels, a carbon neutral cycle created entirely by humans can even be envisioned.
1.6 References
(1) Olah, G. A.; Goeppert, A.; Prakash, G. K. S. Beyond Oil and Gas: The Methanol
Economy; John Wiley & Sons, Inc., 2009.
(2) Olah, G. A.; Goeppert, A.; Prakash, G. K. S. J. Org. Chem. 2009, 74, 487.
(3) Olah, G. A.; Prakash, G. K. S.; Goeppert, A. J. Am. Chem. Soc. 2011, 133, 12881.
(4) Sabatier, P.; Senderens, J. B. J. Chem. Soc., Faraday Trans. 1902, 82, 333.
(5) Fischer, F.; Tropsch, H. Brennst. Chem. 1923, 4, 276.
(6) Hori, Y. In Mod. Aspect. Electroc.; Vayenas, C., White, R., Gamboa-Aldeco, M.,
Eds.; Springer New York: 2008; Vol. 42, p 89.
43
(7) Teeter, T. E.; Van Rysselberghe, P. J. Chem. Phys. 1954, 22, 759.
(8) Hara, K.; Sakata, T. Bull. Chem. Soc. Jpn. 1997, 70, 571.
(9) Carbon Dioxide Chemistry: Environmental Issues; 1st ed.; Pradier, J. P.; Pradier,
C. M., Eds.; Woodhead Publishing, 1994.
(10) Lamy, E.; Nadjo, L.; Saveant, J. M. J. Electroanal. Chem. 1977, 78, 403.
(11) Paik, W.; Andersen, T. N.; Eyring, H. Electrochim. Acta 1969, 14, 1217.
(12) Ikeda, S.; Takagi, T.; Ito, K. Bull. Chem. Soc. Jpn. 1987, 60, 2517.
(13) Hori, Y.; Murata, A.; Kikuchi, K.; Suzuki, S. J. Chem. Soc., Chem. Commun.
1987, 728.
(14) Innocent, B.; Pasquier, D.; Ropital, F.; Hahn, F.; Léger, J. M.; Kokoh, K. B. Appl.
Catal., B 2010, 94, 219.
(15) Innocent, B.; Liaigre, D.; Pasquier, D.; Ropital, F.; Léger, J. M.; Kokoh, K. B. J.
Appl. Electrochem. 2009, 39, 227.
(16) Eneau-Innocent, B.; Pasquier, D.; Ropital, F.; Léger, J. M.; Kokoh, K. B. Appl.
Catal., B 2010, 98, 65.
(17) Delacourt, C.; Newman, J. J. Electrochem. Soc. 2010, 157, B1911.
(18) Delacourt, C.; Ridgway, P. L.; Newman, J. J. Electrochem. Soc. 2010, 157,
B1902.
(19) Chen, Y.; Kanan, M. W. J. Am. Chem. Soc. 2012, 134, 1986.
(20) Zhu, W.; Michalsky, R.; Metin, Ö.; Lv, H.; Guo, S.; Wright, C. J.; Sun, X.;
Peterson, A. A.; Sun, S. J. Am. Chem. Soc. 2013, 135, 16833.
(21) Hori, Y.; Murata, A.; Takahashi, R.; Suzuki, S. Chem. Lett. 1987, 16, 1665.
(22) Hori, Y.; Koga, O.; Yamazaki, H.; Matsuo, T. Electrochim. Acta 1995, 40, 2617.
(23) Hori, Y.; Takahashi, R.; Yoshinami, Y.; Murata, A. J. Phys. Chem. B 1997, 101,
7075.
(24) Hara, K.; Kudo, A.; Sakata, T. J. Electroanal. Chem. 1995, 391, 141.
44
(25) Hara, K.; Kudo, A.; Sakata, T.; Watanabe, M. J. Electrochem. Soc. 1995, 142,
L57.
(26) Koga, O.; Matsuo, T.; Yamazaki, H.; Hori, Y. Bull. Chem. Soc. Jpn. 1998, 71,
315.
(27) Ougitani, Y.; Aizawa, T.; Sonoyama, N.; Sakata, T. Bull. Chem. Soc. Jpn. 2001,
74, 2119.
(28) Christensen, P. A.; Higgins, S. J. J. Electroanal. Chem. 1995, 387, 127.
(29) Yoshitake, H.; Takahashi, K.; Ota, K.-i. J. Chem. Soc., Faraday Trans. 1994, 90,
155.
(30) Giner, J. Electrochim. Acta 1963, 8, 857.
(31) Lu, G.; Wang, H.; Bian, Z.; Liu, X. ScientificWorldJournal 2013, 2013, 8.
(32) Ghadimkhani, G.; de Tacconi, N. R.; Chanmanee, W.; Janaky, C.; Rajeshwar, K.
Chem. Commun. 2013, 49, 1297.
(33) Momose, Y.; Sato, K.; Ohno, O. Surf. Interface Anal. 2002, 34, 615.
(34) Ohya, S.; Kaneco, S.; Katsumata, H.; Suzuki, T.; Ohta, K. Catal. Today 2009,
148, 329.
(35) Chang, T.-Y.; Liang, R.-M.; Wu, P.-W.; Chen, J.-Y.; Hsieh, Y.-C. Mater. Lett.
2009, 63, 1001.
(36) Hoshi, N.; Kato, M.; Hori, Y. J. Electroanal. Chem. 1997, 440, 283.
(37) Schouten, K. J. P.; Pérez Gallent, E.; Koper, M. T. M. ACS Catal. 2013, 3, 1292.
(38) Lu, Q.; Rosen, J.; Zhou, Y.; Hutchings, G. S.; Kimmel, Y. C.; Chen, J. G.; Jiao, F.
Nat. Commun. 2014, 5.
(39) Takahashi, I.; Koga, O.; Hoshi, N.; Hori, Y. J. Electroanal. Chem. 2002, 533,
135.
(40) Hori, Y.; Takahashi, I.; Koga, O.; Hoshi, N. J. Phys. Chem. B 2001, 106, 15.
(41) Hori, Y.; Wakebe, H.; Tsukamoto, T.; Koga, O. Surf. Sci. 1995, 335, 258.
(42) Canfield, D.; Frese, K. W. J. Electrochem. Soc. 1983, 130, 1772.
45
(43) Andrews, E.; Ren, M.; Wang, F.; Zhang, Z.; Sprunger, P.; Kurtz, R.; Flake, J. J.
Electrochem. Soc. 2013, 160, H841.
(44) Durand, W. J.; Peterson, A. A.; Studt, F.; Abild-Pedersen, F.; Nørskov, J. K. Surf.
Sci. 2011, 605, 1354.
(45) Standard Potentials in Aqueous Solution; Bard, A. J.; Parsons, R.; Jordan, J.,
Eds.; IUPAC, 1985.
(46) Whipple, D. T.; Finke, E. C.; Kenis, P. J. A. Electrochem. Solid-State Lett. 2010,
13, B109.
(47) Delacourt, C.; Ridgway, P. L.; Kerr, J. B.; Newman, J. J. Electrochem. Soc. 2008,
155, B42.
(48) Mahmood, M. N.; Masheder, D.; Harty, C. J. J. Appl. Electrochem. 1987, 17,
1159.
(49) Hori, Y.; Wakebe, H.; Tsukamoto, T.; Koga, O. Electrochim. Acta 1994, 39,
1833.
(50) Köleli, F.; Balun, D. Appl. Catal., A 2004, 274, 237.
(51) Prakash, G. K. S.; Viva, F. A.; Olah, G. A. J. Power Sources 2013, 223, 68.
(52) Wu, J.; Risalvato, F.; Zhou, X.-D. ECS Trans. 2012, 41, 49.
(53) Kita, H. J. Electrochem. Soc. 1966, 113, 1095.
(54) Olah, G. A.; Prakash, G. K. S. Electrolysis of carbon dioxide in aqueous media to
carbon monoxide and hydrogen for production of methanol US 12/754,952 March
20, 2012
(55) Ohmori, T.; Nakayama, A.; Mametsuka, H.; Suzuki, E. J. Electroanal. Chem.
2001, 514, 51.
(56) Hori, Y.; Ito, H.; Okano, K.; Nagasu, K.; Sato, S. Electrochim. Acta 2003, 48,
2651.
(57) Chen, Y.; Li, C. W.; Kanan, M. W. J. Am. Chem. Soc. 2012, 134, 19969.
(58) Yamamoto, T.; Hirota, K.; Tryk, D. A.; Hashimoto, K.; Fujishima, A.; Okawa, M.
Chem. Lett. 1998, 27, 825.
46
(59) Ma , S.; Lan , Y.; Perez, G. M. J.; Moniri, S.; Kenis , P. J. A. ChemSusChem 2014,
866.
(60) Kauffman, D. R.; Alfonso, D.; Matranga, C.; Qian, H.; Jin, R. J. Am. Chem. Soc.
2012, 134, 10237.
(61) Shiratsuchi, R.; Nogami, G. J. Electrochem. Soc. 1996, 143, 582.
(62) Yano, H.; Shirai, F.; Nakayama, M.; Ogura, K. J. Electroanal. Chem. 2002, 533,
113.
(63) DeWulf, D. W.; Jin, T.; Bard, A. J. J. Electrochem. Soc. 1989, 136, 1686.
(64) Ishihara, A.; Fujimori, T.; Motohira, N.; Ota, K.-i.; Kamiya, N. Bull. Chem. Soc.
Jpn. 2001, 74, 1517.
(65) Kostecki, R.; Augustynski, J. Ber. Bunsen-Ges. Phys. Chem. 1994, 98, 1510.
(66) Seshadri, G.; Lin, C.; Bocarsly, A. B. J. Electroanal. Chem. 1994, 372, 145.
(67) Frese, K. W., Jr.; Leach, S. C.; Summers, D. P. Electrochemical reduction of
aqueous carbon dioxide to methanol US 4,609,441A 1986
(68) Barton, E. E.; Rampulla, D. M.; Bocarsly, A. B. J. Am. Chem. Soc. 2008, 130,
6342.
(69) Yano, J.; Yamasaki, S. J. Appl. Electrochem. 2008, 38, 1721.
(70) Centi, G.; Perathoner, S.; Wine, G.; Gangeri, M. Green Chem. 2007, 9, 671.
(71) Gangeri, M.; Perathoner, S.; Caudo, S.; Centi, G.; Amadou, J.; Bégin, D.; Pham-
Huu, C.; Ledoux, M. J.; Tessonnier, J. P.; Su, D. S.; Schlögl, R. Catal. Today
2009, 143, 57.
(72) Le, M.; Ren, M.; Zhang, Z.; Sprunger, P. T.; Kurtz, R. L.; Flake, J. C. J.
Electrochem. Soc. 2011, 158, E45.
(73) Qu, J.; Zhang, X.; Wang, Y.; Xie, C. Electrochim. Acta 2005, 50, 3576.
(74) Schrebler, R.; Cury, P.; Suárez, C.; Muñoz, E.; Gómez, H.; Córdova, R. J.
Electroanal. Chem. 2002, 533, 167.
(75) Gattrell, M.; Gupta, N.; Co, A. J. Electroanal. Chem. 2006, 594, 1.
47
(76) Shironita, S.; Karasuda, K.; Sato, K.; Umeda, M. J. Power Sources 2013, 240,
404.
(77) Aydin, R.; Köleli, F. J. Electroanal. Chem. 2002, 535, 107.
(78) Kaneco, S.; Iiba, K.; Katsumata, H.; Suzuki, T.; Ohta, K. Electrochim. Acta 2006,
51, 4880.
(79) Schwartz, M.; Cook, R. L.; Kehoe, V. M.; MacDuff, R. C.; Patel, J.; Sammells, A.
F. J. Electrochem. Soc. 1993, 140, 614.
(80) Zhai, Y.; Chiachiarelli, L.; Sridhar, N. ECS Trans. 2009, 19, 1.
(81) Saeki, T.; Hashimoto, K.; Fujishima, A.; Kimura, N.; Omata, K. J. Phys. Chem.
1995, 99, 8440.
(82) Komatsu, S.; Tanaka, M.; Okumura, A.; Kungi, A. Electrochim. Acta 1995, 40,
745.
(83) Gupta, N.; Gattrell, M.; MacDougall, B. J. Appl. Electrochem. 2006, 36, 161.
(84) Salazar-Villalpando, M. D. ECS Trans. 2011, 33, 77.
(85) Jia, F.; Yu, X.; Zhang, L. J. Power Sources 2014, 252, 85.
(86) Terunuma, Y.; Saitoh, A.; Momose, Y. J. Electroanal. Chem. 1997, 434, 69.
(87) Keith, J. A.; Carter, E. A. J. Am. Chem. Soc. 2012, 134, 7580.
(88) Lim, C.-H.; Holder, A. M.; Musgrave, C. B. J. Am. Chem. Soc. 2012, 135, 142.
(89) Morris, A. J.; McGibbon, R. T.; Bocarsly, A. B. ChemSusChem 2011, 4, 191.
(90) Barton Cole, E.; Lakkaraju, P. S.; Rampulla, D. M.; Morris, A. J.; Abelev, E.;
Bocarsly, A. B. J. Am. Chem. Soc. 2010, 132, 11539.
(91) Kamrath, M. Z.; Relph, R. A.; Johnson, M. A. J. Am. Chem. Soc. 2010, 132,
15508.
(92) Shiratsuchi, R.; Aikoh, Y.; Nogami, G. J. Electrochem. Soc. 1993, 140, 3479.
(93) Hori, Y.; Konishi, H.; Futamura, T.; Murata, A.; Koga, O.; Sakurai, H.; Oguma,
K. Electrochim. Acta 2005, 50, 5354.
48
(94) Yano, H.; Tanaka, T.; Nakayama, M.; Ogura, K. J. Electroanal. Chem. 2004, 565,
287.
(95) Smith, B. D.; Irish, D. E.; Kedzierzawski, P.; Augustynski, J. J. Electrochem. Soc.
1997, 144, 4288.
(96) Oloman, C.; Li, H. ChemSusChem 2008, 1, 385.
(97) Wu, J.; Risalvato, F. G.; Sharma, P. P.; Pellechia, P. J.; Ke, F.-S.; Zhou, X.-D. J.
Electrochem. Soc. 2013, 160, F953.
(98) Narayanan, S. R.; Haines, B.; Soler, J.; Valdez, T. I. J. Electrochem. Soc. 2011,
158, A167.
(99) Li, H.; Oloman, C. J. Appl. Electrochem. 2005, 35, 955.
(100) Li, H.; Oloman, C. J. Appl. Electrochem. 2006, 36, 1105.
(101) Li, H.; Oloman, C. J. Appl. Electrochem. 2007, 37, 1107.
(102) Köleli, F.; Atilan, T.; Palamut, N.; Gizir, A. M.; Aydin, R.; Hamann, C. H. J.
Appl. Electrochem. 2003, 33, 447.
(103) Stevens, G. B.; Reda, T.; Raguse, B. J. Electroanal. Chem. 2002, 526, 125.
(104) Aeshala, L. M.; Rahman, S. U.; Verma, A. Sep. Purif. Technol. 2012, 94, 131.
(105) Aschenbrenner, O.; Styring, P. Energy Environ. Sci. 2010, 3, 1106.
(106) Todoroki, M.; Hara, K.; Kudo, A.; Sakata, T. J. Electroanal. Chem. 1995, 394,
199.
(107) Hara, K.; Tsuneto, A.; Kudo, A.; Sakata, T. J. Electrochem. Soc. 1994, 141, 2097.
(108) Fisher, B. J.; Eisenberg, R. J. Am. Chem. Soc. 1980, 102, 7361.
(109) Bernatis, P. R.; Miedaner, A.; Haltiwanger, R. C.; DuBois, D. L. Organometallics
1994, 13, 4835.
(110) Bourrez, M.; Molton, F.; Chardon-Noblat, S.; Deronzier, A. Angew. Chem., Int.
Ed. 2011, 50, 9903.
49
(111) Leung, K.; Nielsen, I. M. B.; Sai, N.; Medforth, C.; Shelnutt, J. A. J. Phys. Chem.
A 2010, 114, 10174.
(112) Yao, S. A.; Ruther, R. E.; Zhang, L.; Franking, R. A.; Hamers, R. J.; Berry, J. F.
J. Am. Chem. Soc. 2012, 134, 15632.
(113) Schneider, J.; Jia, H.; Kobiro, K.; Cabelli, D. E.; Muckerman, J. T.; Fujita, E.
Energy Environ. Sci. 2012, 5, 9502.
(114) Saravanakumar, D.; Song, J.; Jung, N.; Jirimali, H.; Shin, W. ChemSusChem
2012, 5, 634.
(115) Chen, Z.; Chen, C.; Weinberg, D. R.; Kang, P.; Concepcion, J. J.; Harrison, D. P.;
Brookhart, M. S.; Meyer, T. J. Chem. Commun. 2011, 47, 12607.
(116) Kang, P.; Cheng, C.; Chen, Z.; Schauer, C. K.; Meyer, T. J.; Brookhart, M. J. Am.
Chem. Soc. 2012, 134, 5500.
(117) Kang, P.; Meyer, T. J.; Brookhart, M. Chemical Science 2013, 4, 3497.
(118) Costentin, C.; Drouet, S.; Robert, M.; Savéant, J.-M. Science 2012, 338, 90.
(119) Isaacs, M.; Armijo, F.; Ramírez, G.; Trollund, E.; Biaggio, S. R.; Costamagna, J.;
Aguirre, M. J. J. Mol. Catal. A: Chem. 2005, 229, 249.
(120) Breedlove, B. K.; Takayama, D.; Ito, T. Chem. Lett. 2004, 33, 1624.
(121) Tanaka, K.; Mizukawa, T. Appl. Organomet. Chem. 2000, 14, 863.
(122) Mizukawa, T.; Tsuge, K.; Nakajima, H.; Tanaka, K. Angew. Chem., Int. Ed. 1999,
38, 362.
(123) Meser Ali, M.; Sato, H.; Mizukawa, T.; Tsuge, K.; Haga, M.-a.; Tanaka, K.
Chem. Commun. 1998, 249.
(124) Rakowski Dubois, M.; Dubois, D. L. Acc. Chem. Res. 2009, 42, 1974.
(125) Collomb-Dunand-Sauthier, M.-N.; Deronzier, A.; Ziessel, R. Inorg. Chem. 1994,
33, 2961.
(126) Grodkowski, J.; Dhanasekaran, T.; Neta, P.; Hambright, P.; Brunschwig, B. S.;
Shinozaki, K.; Fujita, E. J. Phys. Chem. A 2000, 104, 11332.
50
(127) Costentin, C.; Drouet, S.; Passard, G.; Robert, M.; Savéant, J.-M. J. Am. Chem.
Soc. 2013, 135, 9023.
(128) Steffey, B. D.; Curtis, C. J.; DuBois, D. L. Organometallics 1995, 14, 4937.
(129) Sonoyama, N.; Kirii, M.; Sakata, T. Electrochem. Commun. 1999, 1, 213.
(130) Cheung, K.-C.; Guo, P.; So, M.-H.; Lee, L. Y. S.; Ho, K.-P.; Wong, W.-L.; Lee,
K.-H.; Wong, W.-T.; Zhou, Z.-Y.; Wong, K.-Y. J. Organomet. Chem. 2009, 694,
2842.
(131) Tornow, C. E.; Thorson, M. R.; Ma, S.; Gewirth, A. A.; Kenis, P. J. A. J. Am.
Chem. Soc. 2012, 134, 19520.
(132) Froehlich, J. D.; Kubiak, C. P. Inorg. Chem. 2012, 51, 3932.
(133) Raebiger, J. W.; Turner, J. W.; Noll, B. C.; Curtis, C. J.; Miedaner, A.; Cox, B.;
DuBois, D. L. Organometallics 2006, 25, 3345.
(134) Magdesieva, T. V.; Yamamoto, T.; Tryk, D. A.; Fujishima, A. J. Electrochem.
Soc. 2002, 149, D89.
(135) Abbott, A. P.; Eardley, C. A. J. Phys. Chem. B 2000, 104, 775.
(136) Saeki, T.; Hashimoto, K.; Kimura, N.; Omata, K.; Fujishima, A. J. Electroanal.
Chem. 1996, 404, 299.
(137) Kaneco, S.; Iiba, K.; Katsumata, H.; Suzuki, T.; Ohta, K. Chem. Eng. J. 2007,
128, 47.
(138) Kaneco, S.; Iiba, K.; Katsumata, H.; Suzuki, T.; Ohta, K. J. Solid State
Electrochem. 2007, 11, 490.
(139) Kaneco, S.; Iiba, K.; Ohta, K.; Mizuno, T.; Saji, A. Electrochim. Acta 1998, 44,
573.
(140) Kaneco, S.; Iiba, K.; Ohta, K.; Mizuno, T.; Saji, A. J. Electroanal. Chem. 1998,
441, 215.
(141) Kaneco, S.; Katsumata, H.; Suzuki, T.; Ohta, K. Energy Fuels 2005, 20, 409.
(142) Kaneco, S.; Ueno, Y.; Katsumata, H.; Suzuki, T.; Ohta, K. Chem. Eng. J. 2006,
119, 107.
51
(143) Kaneco, S.; Yamauchi, H.; Katsumata, H.; Suzuki, T.; Ohta, K. In Stud. Surf. Sci.
Catal.; Sang-Eon Park, J.-S. C., Kyu-Wan, L., Eds.; Elsevier: 2004; Vol. Volume
153, p 277.
(144) Mizuno, T.; Naitoh, A.; Ohta, K. J. Electroanal. Chem. 1995, 391, 199.
(145) Mizuno, T.; Ohta, K.; Kawamoto, M.; Saji, A. Energy Sources 1997, 19, 249.
(146) Rosen, B. A.; Salehi-Khojin, A.; Thorson, M. R.; Zhu, W.; Whipple, D. T.; Kenis,
P. J. A.; Masel, R. I. Science 2011, 334, 643.
(147) Hirota, K.; Tryk, D. A.; Yamamoto, T.; Hashimoto, K.; Okawa, M.; Fujishima, A.
J. Phys. Chem. B 1998, 102, 9834.
(148) Zhao, G.; Jiang, T.; Han, B.; Li, Z.; Zhang, J.; Liu, Z.; He, J.; Wu, W. J.
Supercrit. Fluids 2004, 32, 287.
(149) DiMeglio, J. L.; Rosenthal, J. J. Am. Chem. Soc. 2013, 135, 8798.
(150) Kazarian, S. G.; Briscoe, B. J.; Welton, T. Chem. Commun. 2000, 2047.
(151) Snuffin, L. L.; Whaley, L. W.; Yu, L. J. Electrochem. Soc. 2011, 158, F155.
(152) Nicholas, J. D. Interface 2013, 22, 49.
(153) Minh, N.; Mogensen, M. Interface 2013, 22, 55.
(154) Stoots, C. M.; Hartvigsen, J. J.; O’Brien, J. E.; Herring, J. S. J. Fuel Cell Sci.
Technol. 2008, 6, 011014.
(155) Yue, X.; Irvine, J. T. S. J. Electrochem. Soc. 2012, 159, F442.
(156) Sridhar, K. R.; Foerstner, R. J. Propul. Power 2000, 16, 1105.
(157) Stoots, C. M.; O'Brien, J. E.; Hartvigsen, J. J.; American Nuclear Society: 2007, p
72.
(158) Hartvigsen, J.; Elangovan, S.; Frost, L.; Nickens, A.; Stoots, C. M.; O'Brien, J. E.;
Herring, J. S. ECS Trans. 2008, 12, 625.
(159) Ebbesen, S. D.; Knibbe, R.; Mogensen, M. J. Electrochem. Soc. 2012, 159, F482.
(160) Zhan, Z.; Kobsiriphat, W.; Wilson, J. R.; Pillai, M.; Kim, I.; Barnett, S. A. Energy
Fuels 2009, 23, 3089.
52
(161) Ni, M. J. Power Sources 2012, 202, 209.
(162) Shi, Y.; Luo, Y.; Cai, N.; Qian, J.; Wang, S.; Li, W.; Wang, H. Electrochim. Acta
2013, 88, 644.
(163) Xie, K.; Zhang, Y.; Meng, G.; Irvine, J. T. S. J. Mater. Chem. 2011, 21, 195.
(164) Hansen, J. B. In Sustainable Fuels from Renewable Energies IASS Potsdam,
Germany, 2013.
(165) Hallinder, J.; Holtappels, P.; Mogensen, M. ECS Trans. 2012, 41, 61.
(166) Ormerod, R. M. Chem. Soc. Rev. 2003, 32, 17.
(167) Boston, D. J.; Xu, C.; Armstrong, D. W.; MacDonnell, F. M. J. Am. Chem. Soc.
2013, 135, 16252.
(168) Doherty, M. D.; Grills, D. C.; Muckerman, J. T.; Polyansky, D. E.; Fujita, E.
Coord. Chem. Rev. 2010, 254, 2472.
(169) Dibenedetto, A.; Stufano, P.; Macyk, W.; Baran, T.; Fragale, C.; Costa, M.;
Aresta, M. ChemSusChem 2012, 5, 373.
(170) Gondal, M.; Ali, M.; Dastageer, M.; Chang, X. Catal. Lett. 2013, 143, 108.
(171) Chen, X.; Zhou, Y.; Liu, Q.; Li, Z.; Liu, J.; Zou, Z. ACS Appl. Mater. Inter. 2012,
4, 3372.
(172) Jia, Z. F.; Wang, F. M.; Bai, Y.; Liu, N. Adv. Mat. Res. 2011, 2033.
(173) Li, X.; Liu, H.; Luo, D.; Li, J.; Huang, Y.; Li, H.; Fang, Y.; Xu, Y.; Zhu, L.
Chem. Eng. J. 2012, 180, 151.
(174) Luo, D.; Chen, C.; Zhang, N.; Hong, S.; Wu, H.; Liu, Z. In Z. Phys. Chem. 2009;
Vol. 223, p 1465.
(175) Wu, J. C. S.; Lin, H.-M.; Lai, C.-L. Appl. Catal., A 2005, 296, 194.
(176) Yahaya, A. H.; Gondal, M. A.; Hameed, A. Chem. Phys. Lett. 2004, 400, 206.
(177) Tsai, C.-W.; Chen, H. M.; Liu, R.-S.; Asakura, K.; Chan, T.-S. J. Phys. Chem. C
2011, 115, 10180.
53
(178) Yui, T.; Kan, A.; Saitoh, C.; Koike, K.; Ibusuki, T.; Ishitani, O. ACS Appl. Mater.
Inter. 2011, 3, 2594.
(179) Ursua, A.; Gandia, L. M.; Sanchis, P. Proc. IEEE 2012, 100, 410.
(180) Kreuter, W.; Hofmann, H. Int. J. Hydrogen Energy 1998, 23, 661.
(181) Wendt, H.; Imarisio, G. J. Appl. Electrochem. 1988, 18, 1.
(182) Fischer, J.; Hofmann, H.; Luft, G.; Wendt, H. AIChE J. 1980, 26, 794.
(183) Vandenborre, H.; Leysen, R.; Baetslé, L. H. Int. J. Hydrogen Energy 1980, 5,
165.
(184) Vermeiren, P.; Adriansens, W.; Moreels, J. P.; Leysen, R. Int. J. Hydrogen
Energy 1998, 23, 321.
(185) Carmo, M.; Fritz, D. L.; Mergel, J.; Stolten, D. Int. J. Hydrogen Energy 2013, 38,
4901.
(186) Ayers, K. E.; Anderson, E. B.; Capuano, C.; Carter, B.; Dalton, L.; Hanlon, G.;
Manco, J.; Niedzwiecki, M. ECS Trans. 2010, 33, 3.
54
2 Reduction of CO
2
and water to syngas using gold or tin in a high
pressure flow electrolyzer
2.1 Introduction
The concentration of CO
2
in the atmosphere has been steadily rising since the
industrial revolution when mankind began burning fossil fuels (coal, oil, natural gas).
1
Currently, an industrially viable process which can convert CO
2
back into the fuels which
we use every day does not exist. The goal of electrochemical CO
2
reduction is to develop
a method which could efficiently turn CO
2
into a feedstock (preferably syngas, a
combination of CO and H
2
) for chemical plants and refineries to produce fuels. We are
particularly interested in a ratio of H
2
:CO of 2:1, also known as “metgas”,
2
which would
be ideal for producing methanol in a separate reactor. Methanol produced this way could
be used in a number of applications, both transportation related as well as stationary, and
possibly help to alleviate our reliance on foreign oil. This concept only makes sense if the
power for electrochemical CO
2
reduction comes from renewable sources (solar, wind,
etc.) or nuclear power. The CO
2
could be procured from any source, even the air,
3
however large emitters of high concentration CO
2
streams, such as coal burning power
plants, would probably benefit most from these technologies in the short term.
Concurrent electrolysis of CO
2
and water to form CO and H
2
is an old concept,
with research dating back many decades.
4-5
A comprehensive review of CO
2
reduction on
metal electrodes was published by Hori several years ago.
6
The innovative approach
described herein combines a polymer electrolyte membrane (PEM) flow cell with high
55
pressure CO
2
, a combination which, to the best of our knowledge, has not been described
in the literature.
Selective electrochemical CO
2
reduction to CO has been widely reported
7-10
and
is generally accepted to be the preferred pathway on metal electrodes such as gold and
silver. We will demonstrate here that, under the right conditions, tin can be used as well
as an efficient CO
2
to CO electrocatalyst. Based on the work of Delacourt et al.,
11
we
assume that Equations 2.1 and 2.2 describe the evolution of CO on silver and gold
electrodes in basic media.
Equation 2.1
Equation 2.2
These metals also evolve a significant amount of H
2
at the potentials required for
CO
2
reduction if water is present (as required by Equation 2.2 for CO
2
reduction to take
place), so some current will go towards the hydrogen evolution reaction (HER). HER is
not necessarily detrimental, however, since this process is intended to be used directly in
a syngas reactor which will take CO and H
2
as its feedstock.
2.2 Electrolyzer and system development
Since the goal of this project was to reduce CO
2
under pressure in a flow
electrolyzer, experiments could not be conducted until a suitable electrolyzer could be
acquired. To the best of our knowledge, a commercial source does not exist for a high
pressure flow electrolyzer or fuel cell, so we were forced to design and fabricate an
appropriate electrolyzer in house. In addition to designing the electrolyzer itself,
56
considerable development work was devoted to designing and testing the various
components to support the flow system.
2.2.1 High pressure electrolyzer design
The first version of the high pressure electrolyzer is illustrated in Figure 2.1, and
is very similar to a standard graphite fuel cell, however it was made of 316 stainless steel
and had an o-ring groove. A desired pressure rating can be achieved easily by controlling
the dimensions of the o-ring groove and o-ring hardness. The purpose of the o-ring
groove was to obtain the high pressure rating that was desired, since a simple Teflon
gasket is unlikely to remain air tight at high pressure. This simple design was made by
the USC machine shop and then gold plated for chemical resistance.
Figure 2.1. High pressure electrolyzer version 1
57
Although the cell described in Figure 2.1 held pressure without issues, problems
continually arose. Electrical contact between the column flow pattern and the electrodes
proved troublesome, carbon electrodes regularly disintegrated entirely inside the cell, and
sealing between compartments was generally non-existent. Precision was also required
when tightening the bolts holding the cell together to obtain even spacing between the
metal plates. To address the issues of crushing electrodes and electrical contact, the cell
was modified by shaving down the columns to make space for a gold-plated nickel mesh,
the hypothesis being that a nickel mesh would impart less stress on the fragile graphite
electrodes than the steel pillars did. The nickel mesh did not perform as predicted, and no
improvement in MEA stability, electrical contact, or sealing was observed. Additionally,
the gold deposited on the nickel mesh likely contained numerous impurities which
interfered with CO
2
reduction.
The next attempted solution involved machining additional grooves into the
stainless steel endplates to accept two additional, smaller o-rings, as shown in Figure 2.2.
The concept behind version 2 of the high pressure electrolyzer was that the “low
pressure” o-rings would solve the issue of crossover from the anode to the cathode while
simultaneously holding the MEA in a more well-defined location. The nickel mesh was
kept and used as a buffer between the pillar design and the graphite electrodes. The
Teflon gaskets from version 1 were removed in an effort to simplify the cell and
hopefully limit the possibility for misalignment of the various components to become an
issue. Testing various polymeric coatings for the stainless steel plates revealed
polyethylene to be the best material in this application. Polyethylene was applied by
58
cutting a 0.003 inch thick sheet of polyethylene to roughly the correct size and hot-
pressing it onto the stainless steel plate. Excess polyethylene was then removed with an
X-Acto knife so that the steel pillars were exposed for electrical contact.
Figure 2.2. High pressure electrolyzer version 2
Version 2 of the high pressure electrolyzer finally solved the crossover issue, but
retained many of the earlier problems, particularly the crushing of MEAs. Upon further
analyzing the crushed MEAs, it became apparent that part of the problem had to do with
the fact that Nafion swells when it is hydrated. The thickness of the MEAs increased
drastically after hydration compared to when they were first placed in the cell. This
59
swelling was on the order of a few thousandths of an inch, but was apparently enough to
crush the graphite electrodes when housed in the cell.
Given the various problems with the cell, version 3 was developed as shown in
Figure 2.3. The major change in version 3 is the floating flow field on the anode side,
which is a block of graphite that sits in a 10 mm deep hole where the steel flow field once
was. Underneath the graphite flow field is a stainless steel wave disc spring which allows
the flow field to move up and down ~2 mm while providing electrical contact. This
amount of travel is more than enough to account for the Nafion swelling, thereby
alleviating the crushing problem.
Figure 2.3. High pressure electrolyzer with floating flow field (version 3)
60
The cutaway view of the high pressure floating flow field electrolyzer in Figure
2.4 shows the direction of flow of water and CO
2
through the cell, as well as the O
2
and
CO that is produced at the anode and cathode, respectively. The holes in the graphite
flow field match up to holes in the cathode stainless steel endplate so that the flow is
directed through the flow field and not around it. There is no direct seal between the
stainless steel endplate and the graphite flow field, but the tolerance between the two is
quite small, so it is unlikely that a significant amount of water bypasses the MEA.
Figure 2.4. Cutaway view of high pressure flow electrolyzer
The floating flow field was machined from isostatically pressed graphite instead
of stainless steel because stainless steel can become oxidized when positively charged in
this environment and leach ions like iron, which could clog the membrane. The low
61
pressure o-rings were also removed because they were determined to no longer be
necessary. The o-ring grooves in the plates remained, serving as sealing devices
themselves even without o-rings. During the hydration process, the Nafion membrane
flows into the small o-ring grooves, providing a good low-pressure seal. Originally, tin
was deposited electrochemically on the cathode plate to guard against HER, however it
did not make any difference in our testing (as should be the case given that electrolyte is
not technically in contact with the stainless steel endplate), so the cathode end plate was
simply kept clean and any buildup was removed abrasively. This design proved to be
sufficient to ensure that reliable results could be achieved with the high pressure
electrolyzer, and no further development work was done to improve the design beyond
this point.
2.2.2 High pressure flow system design
During the development of the high pressure flow cell, the system around the cell
was also being designed and refined. The initial design is illustrated in Figure 2.5. CO
2
came from a standard high pressure cylinder, through a bubbler to hydrate it, then a
needle valve to control the flow, then the cathode plate, followed by a gas-liquid
separator to remove water and ionic compounds, a manually actuated back pressure
regulator, and finally a GC to measure gaseous products. On the anode side, nitrogen
pressure was applied to a water reservoir which caused the water to flow through a needle
valve to control flow, followed by the anode plate, then a manually actuated back
pressure regulator, and finally to waste. Two 5 watt heating pads and a thermocouple
62
were connected to a Cole-Parmer Digi-Sense temperature controller to maintain constant
temperature, generally 40 °C.
Figure 2.5. Initial high pressure flow system design
Although the arrangement illustrated in Figure 2.5 allowed pressure to be built-up
inside the cell, there were several issues which needed to be addressed. The first is that
pressure fluctuated significantly even with the back pressure regulator set to a constant
position. The second is that flow on either side was not consistent. The pressure in the
anode and cathode compartments of the cell has to be similar because a pressure
differential across the cell can cause the membrane or electrodes to break, or can force
water or CO
2
through or around the MEA by breaking the low pressure seal or simply
63
diffusing through the membrane. Maintaining the desired flow and pressure proved to be
a challenging task, with adjustments constantly needed to keep the system in a relatively
steady state. Since the purpose of a flow system is to achieve steady state operation, it
was clear that changes needed to be made, the culmination of which are presented in
Figure 2.6.
Figure 2.6. Automated high pressure flow system
a
[a] DAQ: Data Acquisition, PS: Power Supply, LP: Low Pressure, HP: High Pressure,
GC: Gas Chromatograph, MFC: Mass Flow Controller, I/P: Current to Pressure, HPLC:
High-Performance Liquid Chromatography
64
The first change was to the anode side, where an HPLC pump was installed. This
allowed the flow across the anode to be set precisely, with the necessary pressure to
achieve that flow generated by the HPLC pump. On the cathode side, a high pressure
mass flow controller (MFC - Brooks SLA5850 series) was installed to achieve constant
flow. Pressure in the system was controlled via air-actuated back pressure regulators
(Tescom 26-1700 series). These back pressure regulators take air pressure as an input and
scale it to pressure applied to the regulator’s actuator. This system is controlled by two
separate I/P converters (Omega IP610 series), which take a pressurized air supply and
lower the pressure based on a current signal. The current signal was generated by two
National Instruments DAQs (NI USB-6008), which were connected via USB to a
computer running the LabVIEW program. This arrangement allowed the back pressure
regulators to be controlled automatically within the LabVIEW program, however in order
to do this successfully a pair of pressure transducers (Omega PX309 series) were also
connected to the system with their outputs fed into the DAQs as well.
The NI DAQ used (USB-6008) does not have the capability to produce current
signals natively, so circuitry had to be implemented to generate a current signal using a
voltage input. A relatively simple op-amp circuit was implemented as shown in Figure
2.7. The Texas Instruments OP117 was used in this application. Three output signals
were required (one for each I/P converter and one for the MFC), so a small project box
was assembled with three OP117’s, a common power supply, and the two National
Instrument DAQs.
65
Figure 2.7. Voltage to current signal using an op-amp
With these parts in place, a LabVIEW program was written which adjusted the
pressure applied to the back pressure regulators based on a proportional-integral-
derivative (PID) controller to achieve a desired pressure set point. Although the original
plan for this system had been to use PID controllers for both anode and cathode
compartments, the behavior of the PID controller on the anode side caused too much
fluctuation, so pressure on the anode was set using a simple pressure set point. Drift in
this set point did occur, but was generally 10-20 psi over 8 hours or so, which was
considered acceptable. When fully operational, this system maintained steady pressure
(within 10 psi) and flow (within 0.1 mL/min) indefinitely.
66
2.3 Results and discussion
2.3.1 Screening conditions
Table 2.1 lists 31 experiments where catalyst, catalyst particle size, membrane
composition, catholyte composition, anolyte composition, cell potential, pressure, and
flow rate were modified to determine reasonable operating conditions for the CO
2
reduction cell. Experiments were conducted in constant potential mode for six hours with
a platinum black on carbon paper anode and a Nafion 117 membrane. Exceptional
experiments have been highlighted in red for either high current density, high formate
faradaic efficiency (FE), high CO FE, or high hydrogen evolution reaction (HER) FE.
Due to the drastically different methods used to measure effluent composition (GC and
NMR) and the fact that GC measurements were taken every 15 minutes, while NMR
could only be run once at the end of the experiment, the total FE does not always add up
to 100%.
The electrocatalyst has quite a significant effect on the reaction, with tin generally
giving higher current density at equivalent conditions, while lead generally gave higher
formate efficiency with lower current density. Smaller particle size (going from 100 mesh
tin particles to <150 nm tin particles) gave higher current density, but at the expense of
formate efficiency. While higher current density was expected given the higher absolute
surface area, lower formate efficiency was not expected. A report in the literature noticed
a similar effect for gold nanoparticles,
12
however, and these findings are in line with
theirs. We also briefly experimented with tin plating a lead powder electrode in Entries 9
67
Table 2.1: Screening CO
2
reduction conditions
Entry Electrode Membrane Catholyte Anolyte
Cell
Potential
(V)
Cathode
Pressure
(psig)
Current
Density
(mA/cm
2
)
Formate
(%)
CO
(%)
H
2
(%)
Total
(%)
1 100 mesh Sn Nafion-H CO
2
0.1 M LiOH 3.3 300 23.7 27% 36% 29% 92%
2 100 mesh Sn Nafion-H CO
2
0.1 M LiOH 3.3 300 17.8 34% 54% 10% 98%
3 100 mesh Sn Nafion-H CO
2
0.1 M LiOH 3.3 300 13.8 29% 57% 11% 97%
4 100 mesh Sn Nafion-H CO
2
0.1 M LiOH 3.6 300 24.2 39% 51% 10% 100%
5 300 mesh Pb Nafion-H CO
2
0.1 M LiOH 3.3 300 13.4 48% 0% 19% 68%
6 300 mesh Pb Nafion-H CO
2
0.1 M LiOH 3.3 300 7.9 29% 0% 23% 51%
7 300 mesh Pb Nafion-H CO
2
0.1 M LiOH 3.3 300 13.3 37% 26% 23% 86%
8 300 mesh Pb Nafion-H CO
2
0.1 M LiOH 3.6 300 14.6 39% 29% 21% 89%
9
Sn plated on
Pb
Nafion-H CO
2
0.1 M LiOH 3.3 300 12.3 28% 34% 27% 89%
10
Sn plated on
Pb
Nafion-H CO
2
0.1 M LiOH 3.3 300 11.0 28% 46% 19% 94%
11
Sn plated on
Pb
Nafion-H CO
2
0.1 M LiOH 3.3 300 11.2 26% 0% 0% 26%
12 100 mesh Sn Nafion-H CO
2
0.1 M LiOH 3.3 300 14.2 32% 49% 14% 95%
13 100 mesh Sn Nafion-H CO
2
+ H
2
O 0.1 M LiOH 3.3 300 12.3 35% 48% 9% 92%
14 100 mesh Sn Nafion-H CO
2
+ H
2
O 0.1 M LiOH 3.3 300 19.9 30% 43% 15% 88%
15 100 mesh Sn Nafion-H CO
2
+ H
2
O 0.1 M LiOH 3.3 300 10.7 31% 51% 12% 95%
16 100 mesh Sn Nafion-H CO
2
+ H
2
O 0.1 M LiOH 3.3 300 9.4 28% 88% 21% 137%
17 100 mesh Sn Nafion-Li CO
2
0.1 M LiOH 3.3 300 76.8 7% 9% 84% 100%
18 100 mesh Sn Nafion-Li CO
2
0.1 M LiOH 3.3 300 67.0 7% 13% 69% 90%
19 100 mesh Sn Nafion-Li CO
2
0.1 M LiOH 3.3 300 30.6 25% 50% 13% 88%
20 100 mesh Sn Nafion-Na CO
2
0.1 M
NaOH
3.3 300 79.9 13% 16% 68% 97%
68
Table 2.1 (continued): Screening CO
2
reduction conditions
Entry Electrode Membrane Catholyte Anolyte
Cell
Potential
(V)
Cathode
Pressure
(psig)
Current
Density
(mA/cm
2
)
Formate
(%)
CO
(%)
H
2
(%)
Total
(%)
21 100 mesh Sn Nafion-Na CO
2
0.1 M
NaOH
3.3 300 53.8 19% 27% 46% 91%
22 100 mesh Sn Nafion-Na CO
2
0.1 M
NaOH
3.6 300 132.4 12% 11% 78% 101%
23 100 mesh Sn Nafion-Na CO
2
0.1 M
NaOH
3.3 450 79.2 15% 18% 49% 82%
24 100 mesh Sn Nafion-Na CO
2
+ H
2
O
0.1 M
NaOH
3.3 300 53.2 18% 27% 50% 95%
25 100 mesh Sn Nafion-Na CO
2
H
2
O 3.3 300 55.5 1% 7% 85% 92%
26 <150 nm Sn Nafion-H CO
2
0.1 M LiOH 3.3 300 97.2 10% 13% 72% 95%
27 <150 nm Sn Nafion-H CO
2
0.1 M LiOH 3.3 300 71.2 8% 11% 73% 92%
28 <150 nm Sn Nafion-H CO
2
0.1 M LiOH 3.3 450 78.3 11% 12% 54% 77%
29 <150 nm Sn Nafion-H CO
2
0.1 M LiOH 3.6 300 121.3 6% 5% 86% 96%
30 <150 nm Sn Nafion-H CO
2
0.1 M LiOH 3.0 300 26.2 17% 28% 31% 77%
31 <150 nm Sn Nafion-H CO
2
0.1 M LiOH 3.6 450 68.0 11% 13% 77% 101%
69
through 11. There does not appear to be a synergistic effect from the combination of tin
and lead, at least in the way that we combined them through first painting lead on toray
carbon paper, followed by electroplating tin.
Membrane composition was a factor that produced interesting results. Nafion is
typically sold in the sodium ion form, and then treated in a sequence of sulfuric acid and
hydrogen peroxide baths to produce the proton form. This process was reversed in several
experiments by boiling the membrane in 0.1 M NaOH or LiOH before pressing the MEA.
Surprisingly, current density increased significantly when either Nafion-Li or Nafion-Na
pre-treated membranes were used. Unfortunately, this was also coupled with significantly
higher HER, so CO
2
reduction FE plunged. Despite this, the amount of formate being
produced actually increased compared to other experiments with much higher formate
efficiencies. For example, in Entry 4, 39% formate efficiency and 24.2 mA/cm
2
current
density were observed, for 9.4 mA/cm
2
formate current density. In Entry 22, 12%
formate efficiency and 132.4 mA/cm
2
were observed, for 15.9 mA/cm
2
formate current
density. Both of these experiments used the same electrocatalyst with otherwise almost
identical conditions, except for the pre-treatment of the membrane with boiling 0.1 M
NaOH in Entry 22. Similar experiments for LiOH produced similar results, however the
current densities were higher with NaOH-treated membranes.
One example of pure water on the anode is given, Entry 25, where 85% hydrogen
was detected and only 1% formate. This further reinforced earlier findings that acidic
Nafion does not facilitate efficient CO
2
reduction. Since the composition of the anolyte
produced such profound effects on CO
2
electrolysis, several other electrolytes and
70
concentrations were tried. Most combinations other than 0.1 M LiOH, 0.1 M NaOH, and
0.1 M Li
2
CO
3
failed, however. Sodium salts other than hydroxide resulted in a clogged
cathode flow field, as did potassium based salts. Higher concentration also resulted in
cathode flow field clogging, so 0.1 M was retained as the concentration of choice. Li
2
CO
3
was chosen for future experiments instead of LiOH because LiOH tends to be harder to
work with, and likely forms Li
2
CO
3
in solution when it reacts with CO
2
in the atmosphere
anyway.
The amount of water present on the cathode side was a concern because the CO
2
used was dry. While water is the source of HER and undesirable for that reason, it is also
essential to CO
2
reduction. Two ways to mitigate this potential issue were developed, the
first was to use an overpressure on the anode side of 5-15 psi to “push” water through the
membrane. This approach yielded mixed results, nevertheless 5 psi overpressure was
used as the standard condition for the reactions described in Table 2.1. The next idea was
to use a second HPLC pump to introduce water with the pressurized CO
2
stream on the
cathode side at a very low flow rate. Flow rates of 0.05-0.15 mL/min water were tested
on the cathode, but the results did not indicate a significant improvement. Entries 12-16
relate to adding extra water to the cathode, where lower current densities were observed
after adding water, as well as variable amounts of CO and H
2
. Interestingly, virtually no
change whatsoever was observed for the formate efficiency.
Temperature was also briefly investigated, but results proved inconclusive and are
not presented here. In general, increased temperature led to increased HER and a faster
rate of MEA degradation. Since low temperature operation was not an option, the
71
intermediate temperature of 40 °C was settled on simply to maintain a constant
temperature for future experiments without causing undue harm to the cell or increasing
HER.
Increasing pressure much above 300 psig did not significantly affect the results,
although in some cases we did observe modest increases in either formate efficiency or
current density, but not both. The explanation for this is likely that CO
2
concentration is
not limiting at the cathode at 300 psi, and that there is some other factor that is limiting
the CO
2
reduction current density. Increasing flow rate, the other way to increase CO
2
availability, gave mixed results and is not shown in Table 2.1, but was explored in detail
in section 2.3.2.
2.3.2 Effect of pressure and flow rate
A series of experiments (each using a newly fabricated MEA) was conducted to
determine the exact effect of pressure by holding all other factors constant and increasing
pressure in 150 psi steps with the results presented in Figure 2.8. As might be expected,
increased CO
2
pressure (and thus CO
2
concentration) increases the FE of CO. However,
the increase does not continue for pressures higher than 300 psig. Since the flow reactor
has a fixed pillar design, the amount of mixing in the cell is proportional to the
volumetric flow rate inside the cell. Flow rates are determined on the cathode in mL/min
CO
2
, but this is measured at atmospheric pressure, not at high pressure. Therefore, during
the cell operation at 300 psig (21 atm), the volumetric flow inside the cell can be
calculated using the ideal gas law as being 1.2 mL/min as opposed to 25 mL/min at
72
atmospheric pressure. At 600 psig, there is only 0.59 mL/min volumetric flow, which is
less than half of the volumetric flow on the anode side. The CO FE increased
dramatically from 38% to 56% when the flow increased from 25 mL/min to 75 mL/min,
which is the highest FE observed on a gold electrode under similar conditions.
Figure 2.8. Effect of flow rate and pressure on CO FE on a gold cathode
a
[a] All experiments carried out galvanostatically at 30 mA/cm
2
for 1 hour, 0.5-0.8 m Au
6.3 mg/cm
2
on 10% Teflonized Toray paper cathode, Pt black 4.5 mg/cm
2
on non-
Teflonized Toray paper anode, Nafion 117 membrane, indicated cell pressure, 25 mL/min
CO
2
flow rate across the cathode except where indicated, 2 mL/min 0.1 M NaOH flow
rate across the anode, 40 °C.
A flow of 75 mL/min under atmospheric pressure is only 1.8 mL/min at 600 psi.
These results indicate that, at the lower flow rate, a region where mass transport is the
limiting factor has been reached. Although more experiments would be needed to explain
this effect more precisely, volumetric flow appears to be responsible for degraded
0
10
20
30
40
50
60
0 150 300 450 600
CO Faradaic Efficiency (%)
Pressure (psig)
75 mL/min
25 mL/min
73
performance at high pressures. An improved flow field could potentially alleviate some
of these issues, as could a higher flow rate, but neither was attainable in the present
system. An improved flow field would likely be the best solution since, at 25 mL/min
(the standard flow rate), only around 1% of the CO
2
is converted into products. An inert
gas could be used to increase volumetric flow (and thus mixing) at higher pressures,
however this would defeat the purpose of using high pressure in the first place, which is
to increase the concentration of CO
2
at the electrode interface.
2.3.3 CO
2
reduction on gold
Carbon monoxide is the most commonly observed product when using a gold
working electrode for CO
2
reduction.
6
Most reports in the literature do not give real-time
data related to CO production using gold electrodes. However, many reports
8,13-15
mention a poisoning effect which has been observed on gold electrodes in a variety of
conditions. Reports in the literature generally collect FE data quite early,
16
specifically to
avoid this poisoning problem. Initial experiments conducted using the pressurized flow
reactor indicated complex behavior on gold electrodes, so a relatively long 8 hour
experiment was conducted with analysis of the product gasses every 15 minutes, the
results of which are shown in Figure 2.9. Experiments at 2, 4, 8, and 16 hours utilizing
the same experimental conditions showed that 8 hours was sufficient time for product
gasses to stabilize.
74
Figure 2.9. CO
2
reduction on gold with a 0.1 M NaOH anolyte solution
a
[a] Carried out galvanostatically at 30 mA/cm
2
for 8 hours, 0.5-0.8 m Au at 10.42
mg/cm
2
on 6% Teflonized Toray paper cathode, Pt black 4.44 mg/cm
2
on non-Teflonized
Toray paper anode, Nafion 117 membrane, 300 psig cell pressure, 25 mL/min CO
2
flow
rate across the cathode, 1 mL/min 0.1 M NaOH flow rate across the anode, 40 °C. Total
integrated faradaic efficiencies were H
2
: 66%, CO: 38%, HCOO
-
: 0% (104% total FE).
Average cell potential: -4.18 V.
Despite a relatively constant potential required to achieve 30 mA/cm
2
over 8
hours, the product distribution in Figure 2.9 shows significant time-dependent effects.
Surprisingly, at 6 hours the trend reverses and CO production increases along with a
decrease in H
2
evolution. CO production initially peaks at ~2 hours and then begins a
gradual decline. CO FE starts out at 83% at 30 minutes, falling to 50% at 2 hours, then
29% at 5.5 hours before increasing up to 37% at 8 hours.
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0 2 4 6 8 10
Product Flow (mL/min)
Time (hours)
Applied current ends
H2 (mL/min)
CO (mL/min)
75
Changing the anolyte composition from 0.1 M NaOH to 0.1 M Li
2
CO
3
changed
the product distribution significantly, as shown in Figure 2.10. Product composition was
steady throughout the 8 hour run, with no discernable oscillation observed, although the
FE for CO was significantly lower (26% overall) than observed when 0.1 M NaOH was
used as the anolyte (38% overall). The difference between the experiment with sodium
and the experiment with lithium is quite striking, and leads us to the hypothesis that the
interactions between the electrode and the cations are critical to electrochemical
reduction of CO
2
, discussed further in section 2.3.5.
Figure 2.10. CO
2
reduction on gold with a 0.1 M Li
2
CO
3
anolyte solution
a
[a] Carried out galvanostatically at 30 mA/cm
2
for 8 hours, 0.5-0.8 m Au at 10.35
mg/cm
2
on 6% Teflonized Toray paper cathode, Pt black 3.92 mg/cm
2
on non-Teflonized
Toray paper anode, Nafion 117 membrane, 300 psig cell pressure, 25 mL/min CO
2
flow
rate across the cathode, 1 mL/min 0.1 M Li
2
CO
3
flow rate across the anode, 40 °C. Total
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0 2 4 6 8 10
Product Flow (mL/min)
Time (hours)
H2 (mL/min)
CO (mL/min)
76
integrated faradaic efficiencies were H
2
: 82%, CO: 26%, HCOO
-
: 0% (108% total FE).
Average cell potential: -3.02 V.
Results from the experiments on gold electrodes demonstrate that, while gold
shows almost perfect selectivity for CO as its CO
2
reduction product, it also produces a
significant amount of H
2
under the conditions tested here when experiments are
continued until they reach a steady state. The FE for CO was as high as 83% at the
beginning of the NaOH experiments (Figure 2.9), indicating that data taken from the
beginning of an experiment is not necessarily indicative of the steady state. A lack of
steady state data in the literature could be the reason that FE’s reported here are lower
than many reports in the literature. The choice of salt for the anolyte solution turned out
to be surprisingly important. While 0.1 M NaOH gave overall a higher CO
2
reduction
current, 0.1 M Li
2
CO
3
produced a much more consistent product composition. While it
definitely appears that there is some form of poisoning on gold electrodes during CO
2
reduction, it appears to be quite weak, and therefore there are likely many options to
reduce or eliminate this possible problem.
2.3.4 Tin as a cathode for syngas production
Tin is not a precious metal (like gold), and is not toxic like some of the other
metals which tend to reduce CO
2
efficiently like lead.
6
Therefore, its use in almost any
application would be preferred over a more expensive or harmful metal. Although tin is
generally reported to produce primarily formate during CO
2
electrolysis,
6
conditions have
previously been identified which result in the production of primarily CO.
17
In an attempt
77
to better understand CO
2
reduction on a tin electrode in a high pressure flow system, a
broad range of current densities was investigated, from 3 mA/cm
2
to 500 mA/cm
2
. The
experiments were run in a random order on the same MEA to ensure that temporal effects
would not present themselves. Figure 2.11 plots the results from this series of
experiments.
Figure 2.11. CO
2
reduction at various current densities on a tin electrode
a
[a] All experiments carried out galvanostatically for 2 hours, <150 nm Sn 7.61 mg/cm
2
on 10% Teflonized Toray paper cathode, Pt black 4.81 mg/cm
2
on non-Teflonized Toray
paper anode, Nafion 117 membrane, 450 psig cell pressure, 50 mL/min CO
2
flow rate
across the cathode, 2 mL/min 0.1 M NaOH flow rate across the anode, 40 °C.
As expected, the average voltage increases nearly linearly with logarithmic
current density increase. FE for formate peaks at 10 mA/cm
2
, while FE for CO peaks at
30 mA/cm
2
. Interestingly, CO is always the favored CO
2
reduction product, except at the
lowest current density, where the results are not particularly reliable due to detection
0%
5%
10%
15%
20%
25%
30%
35%
40%
1 10 100 1000
Faradaic Efficiency (%)
Current Density (mA/cm
2
)
CO
HCOO-
78
limits. The fact that this system generally favors CO production on tin suggests that the
environment is almost non-aqueous in nature given the similarity in product distribution
to prior work on tin in non-aqueous conditions.
17
This is surprising given the fact that
Nafion requires a significant amount of water to be present to function as a cation
exchange membrane. The shape of the CO and formate curves are similar, however CO
appears to be favored over a broader range of current densities than formate. Another
interesting point is the 100 mA/cm
2
data point, where 30% FE for CO at 4.0 V was
observed. This particular experiment produces almost exactly the H
2
:CO ratio of 2:1 that
would be ideal for methanol synthesis. Current densities this high on gold electrodes in
this system yield higher H
2
formation.
Figure 2.12. CO
2
reduction on tin with 0.1 M NaOH anolyte solution
a
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 2 4 6 8 10
Product Flow (mL/min)
Time (hours)
H2 (mL/min)
CO (mL/min)
79
[a] Carried out galvanostatically at 30 mA/cm
2
for 8 hours, <150 nm Sn at 6.50 mg/cm
2
on 6% Teflonized Toray paper cathode, Pt black 3.05 mg/cm
2
on Toray paper anode,
Nafion 117 membrane, 300 psig cell pressure, 25 mL/min CO
2
flow rate across the
cathode, 1 mL/min 0.1 M NaOH flow rate across the anode, 40 °C. Total integrated
faradaic efficiencies were H
2
: 67%, CO: 21%, HCOO
-
: 10% (98% total FE). Average cell
potential: -3.48 V.
With preliminary information about the behavior of tin as a cathode material in
the high pressure flow system in hand, focus turned to comparison with the gold system.
An experiment was carried out on a tin electrode using 0.1 M NaOH as the anolyte
solution (Figure 2.12) demonstrated a slow decline of the FE for CO, while the FE for H
2
slowly increased over 8 hours of operation. Unlike what was observed in the case of the
gold electrode, no FE oscillation was observed, although the CO FE in Figure 2.12 is not
entirely stable, and may hide an oscillation which is taking place. Like the gold
experiment, the production rate of CO is initially greater than that of H
2
, but drops
significantly as the experiment continues, possibly indicating some sort of weak
poisoning effect. Had the experiment ended at 2 hours, the conclusion would have been
that CO production was approximately equal to H
2
production, even slightly higher for
the beginning of the run. This initial spike in activity quickly subsided and a slow
degradation in CO FE was observed from ~2 hours onward. Interestingly, the cell
potential remained relatively constant throughout the experiment, indicating that the
change of the product distribution does not coincide with an increase in resistance at the
electrode.
80
Figure 2.13. CO
2
reduction on tin with a 0.1 M Li
2
CO
3
anolyte solution
a
[a] Carried out galvanostatically at 30 mA/cm
2
for 8 hours, <150 nm Sn at 6.05 mg/cm
2
on Toray paper cathode, Pt black 4.98 mg/cm
2
on 6% Teflonized Toray paper anode,
Nafion 117 membrane, 300 psig cell pressure, 25 mL/min CO
2
flow rate across the
cathode, 1 mL/min 0.1 M Li
2
CO
3
flow rate across the anode, 40 °C. Total integrated
faradaic efficiencies: H
2
: 21%, CO: 46%, HCOO
-
: 27% (94% total FE). Average cell
potential: -3.57 V.
Given that Li
2
CO
3
had produced interesting results when used in the anolyte in
the gold system (Figure 2.10), the same conditions were applied to a tin electrode, with
the results presented in Figure 2.13. Surprisingly, much higher CO
2
reduction FE was
observed compared to the NaOH based anolyte with the tin electrode (Figure 2.12). The
FE of CO is quite consistent throughout the 8 hour experiment, however there does
appear to be a significant increase in H
2
FE approximately 5 hours into the experiment.
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
0 2 4 6 8 10
Product Flow (mL/min)
Time (hours)
H2 (mL/min)
CO (mL/min)
81
2.3.5 The effect of the cation in CO
2
reduction
In sections 2.3.3 and 2.3.4, clear differences were observed when either NaOH or
Li
2
CO
3
was used on the anode, summarized in Figure 2.14. CO
2
reduction on gold was
~1:3 in favor of H
2
evolution when Li
2
CO
3
was used (Figure 2.10), compared to ~1:1.7
when NaOH was used (Figure 2.9). Total CO
2
reduction (CO + formate) FE on tin was
roughly 3.5:1 compared to the FE of H
2
evolution when Li
2
CO
3
was used (Figure 2.13),
compared to roughly 1:2 in favor of H
2
evolution when NaOH was used (Figure 2.12). In
both cases on tin, selectivity for CO compared to formate was roughly 2:1 in favor of CO.
The drastic differences in CO
2
reduction FE can only be attributed to the difference of
anolyte, as all other factors were kept as consistent as possible, and several experiments
were run with similar sets of conditions to verify reproducibility.
Figure 2.14. Total CO
2
reduction FE on tin and gold
a
[a] integrated results from experiments represented in Figures 2.9, 2.10, 2.12, and 2.13
0
10
20
30
40
50
60
70
80
Tin Gold
Total CO
2
Reduction FE (%)
Electrode
Sodium
Lithium
82
The difference between NaOH and Li
2
CO
3
is most likely related to the cation
itself since that is the only species which can pass through the Nafion membrane to
interact with the working electrode. It is highly unlikely that the difference between the
anions, which can only interact with the platinum counter electrode, is responsible for
such a dramatic shift in selectivity on the working electrode. The differences in FE can be
attributed to one of two causes: 1. The cation interacts with CO
2
or a reaction
intermediate 2. The cation interacts with the surface of the electrode. Both processes are
certainly taking place, and it has been shown that CO
2
interacts more strongly with Li
+
than Na
+
.
18
However, the drastic changes in reactivity cannot be attributed to simple
interactions between the cation and CO
2
because Li
+
favors CO
2
reduction on a tin
electrode and disfavors CO
2
reduction on a gold electrode (Figure 2.14).
Cationic species are known to alter the adsorption behavior of CO
2
and CO on the
surface of gold.
19
We therefore hypothesize that the cationic species present at the
cathode alters the adsorption behavior of CO
2
, CO, or both and can lead to a reversible
poisoning or oscillation effect which was observed in the case of a sodium cation on a
gold electrode. Oscillation of CO oxidation on platinum has been observed,
20
where co-
adsorption of anions and other species led to nontrivial interactions which lead to system
instability or oscillation. Therefore, in the absence of another reasonable explanation,
there is likely a specific interaction between the cation and the surface of the working
electrode which affects the binding of reactants and products since lithium favored CO
2
reduction on tin, while sodium favored CO
2
reduction on gold. To the best of our
83
knowledge, this is the first demonstration of such an effect for electrochemical CO
2
reduction, and highlights the importance of electrolyte choice for this particular reaction.
2.3.6 Increasing current density
One of the major concerns for any electrochemical process is the current density,
since this directly affects the cost of the system.
21
Reports of high energy efficiency
abound in the literature,
8,22-24
but reports of high energy density have become relatively
scarce recently. While short duration experiments at high current densities work without
issue in our system (see Figure 2.11), steady state operation is much harder to achieve
under such conditions. Typically, the cathode flow field becomes clogged after a certain
amount of time. It is unclear why this occurs in our system. Figure 2.15 plots the product
distribution from an 8 hour experiment at 60 mA/cm
2
, which is the highest current
density that we have been able to operate the cell at while still achieving relatively stable
results.
84
Figure 2.15. CO
2
reduction on tin with a 0.1 M Li
2
CO
3
electrolyte at 60 mA/cm
2 a
[a] Carried out galvanostatically at 60 mA/cm
2
for 8 hours, <150 nm Sn at 6.38 mg/cm
2
on Toray paper cathode, Pt black 3.48 mg/cm
2
on 6% Teflonized Toray paper anode,
Nafion 117 membrane, 300 psig cell pressure, 50 mL/min CO
2
flow rate across the
cathode, 2 mL/min 0.1 M Li
2
CO
3
flow rate across the anode, 40 °C. Total integrated
faradaic efficiencies: H
2
: 61%, CO: 23%, HCOO
-
: 20% (104% total FE). Average cell
potential: -3.83 V.
The rate of CO production at 60 mA/cm
2
is almost identical to the rate of CO
production at 30 mA/cm
2
(Figure 2.13), so the additional current density is almost
exclusively going towards H
2
production. This result suggests that mass transport has
become limiting compared to the experiment at 30 mA/cm
2
, however the CO
2
flow rate
was doubled (50 mL/min compared to the standard 25 mL/min) which should have
significantly increased CO
2
mass transport in the system.
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
0 2 4 6 8 10
Product Flow (mL/min)
Time (hours)
H2 (mL/min)
CO (mL/min)
85
2.3.7 Tokuyama anion conducting membrane
Given that CO
2
reduction is almost exclusively conducted in basic media, it
seemed logical to test the reaction using a hydroxide conducing membrane like the
Tokuyama membrane instead of a cation conducting membrane like Nafion. Since
negative ions flow though the membrane instead of positive ions, one issue with this
setup is that the flow of ions is opposite to the usual case with Nafion. Hydroxide ions
will be formed on the cathode as a by-product of CO
2
reduction, and these ions must then
flow through the membrane to be oxidized to O
2
at the anode. To provide the necessary
water molecules to the cathode of the reactor, a second HPLC pump was connected to the
system which delivered 0.15 mL/min 0.1 M NaOH to the cathode in addition to 30
mL/min CO
2
.
Although current flow was observed when a tin electrode on a Tokuyama MEA
was tested, only traces of formate were detected in the gas-liquid separator. Despite this,
<50% of the current was accounted for when analyzing the effluent gasses (for H
2
and
CO) by GC. Thinking about this further, it was surmised that the formate ions were most
likely adsorbing into the Tokuyama membrane instead of making their way to the gas-
liquid separator. In the case of a Nafion membrane, anionic species like formate are
rejected from the membrane because they carry the same charge as the sulfonic acid
groups on the surface. With the Tokuyama membrane, the positively charged alkyl
ammonium groups attract the formate ions, which are produced on a tin electrode during
CO
2
reduction.
86
Figure 2.16. CO
2
reduction on silver using the Tokuyama membrane
[a]
[a] Carried out potentiostatically at 3.3 V for 2 hours, 0.5-1.0 m Ag on 10% Teflonized
Toray paper cathode, Pt on non-Teflonized Toray paper anode, 300 psig cell pressure, 30
mL/min CO
2
, 0.15 mL/min 0.1 M NaOH flow rate across the cathode, 2 mL/min H
2
O
flow rate across the anode, 50 °C.
Since formate could not be observed when reducing CO
2
in a Tokuyama
membrane, CO
2
reduction on silver was explored with the goal of generating CO instead
of formate. Presumably, CO would not be trapped by the membrane as was the case with
formate. Figure 2.16 gives the results of one such experiment, showing the current
density, H
2
and CO flow in the effluent. An average of 23.0 mA/cm
2
was observed over 2
hours, with 47.5% CO FE and 44.9% H
2
FE. These figures fall fairly well in line with
other experiments previously performed under similar conditions but using Nafion
instead of the Tokuyama membrane. Unfortunately, it appears that CO production starts
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0
5
10
15
20
25
30
35
40
45
50
0 50 100 150 200
Gas in Effluent (mL/min)
Current Density (mA/cm
2
)
Time (min)
Current Density
H2 in Effluent
CO in Effluent
87
off as the major product of the reaction, but drops of significantly at 45 minutes and falls
gradually until the end of the reaction. The current density also falls considerably from
the initial state, where it appeared to start close to 45 mA/cm
2
. This decreasing
performance could be partially due to catalyst deactivation, as has been suggested for
gold electrodes evolving CO.
Although additional experimentation may have yielded interesting insights into
CO
2
reduction using an anionic membrane, the current system was not well-equipped for
such studies. The main issue revolves around the gas-liquid separator, which has very
limited capacity of ~120 mL. This small capacity makes it challenging to pass a large
volume of liquid over the cathode since an overflow will end up in the GC after going
through the back pressure regulator. A significant redesign would be required for the
system to be capable of sustaining reasonable liquid flow on the cathode for any length of
time.
2.4 Technoeconomic analysis
Based on extensive electrochemical CO
2
reduction presented vide supra, a series
of cost estimates have been made to determine the rough price of chemicals obtained
using these processes. Given that the goal of CO
2
reduction is to produce fuels and
commodity chemicals, cost is of the utmost importance. While the following exercise is
not exhaustive, ballpark cost estimates are provided and the factors influencing cost
examined.
88
2.4.1 Assumptions
The only industrial scale process that resembles CO
2
electrolysis is water
electrolysis. With alkaline water electrolysis, it has been estimated that over 80% of the
cost is due to electricity prices.
25
Although there are significant differences between
water electrolysis and CO
2
electrolysis, we will assume that the cost of electricity will
remain the major associated cost with the products produced. Other costs, such as the cost
to build and maintain the reactor, finance, etc. have not been considered. Since CO
2
is
likely to either be free or even have a negative cost, i.e. that there will be a monetary
incentive to use it, we will not consider it in our analysis.
A potential roadblock to CO
2
reduction scale up is that less than 1% of the CO
2
that flows across the cathode is consumed. Perhaps this would not be a significant issue if
pumps could easily re-circulate CO
2
, or perhaps conditions could be found to convert a
much larger percentage of the incoming gas. At the very least, it seems that a significant
investment in CO
2
recycling equipment would be necessary for large scale
electrochemical CO
2
reduction. Methods for increasing CO
2
conversion have not been
investigated because it was never a parameter that has been focused on, but it may prove
to be a significant barrier to industrial application.
For the purposes of this analysis, electricity generation will be assumed to take
place on site or very close so that the cost of electricity can be brought down to what it
costs to generate, rather than what it costs to buy from the grid. Rather than selecting a
particular strategy for electricity generation, a cost of 4 ¢/kWh has been selected, which
falls in the range of electricity produced from coal, natural gas, wind, and is more
89
expensive than nuclear.
26
While it would not be advantageous to the environment to use
coal or natural gas to power a CO
2
electrolyzer, wind and nuclear power would be good
choices given that they do not emit CO
2
. The cost of solar power is currently too high to
be considered for electrochemical CO
2
reduction.
2.4.2 Conditions for formic acid production
Through a great deal of testing various catalysts and conditions, a conservative
estimate for the operating conditions and expected product distribution has been
generated, which could be expected for a large scale production facility. To begin with, a
single experiment has been selected as a baseline for CO
2
reduction to formate using the
conditions specified in Table 2.2.
Table 2.2: Formic Acid Baseline Conditions
Controlled
Variables
Working Electrode 100 mesh Tin powder
Membrane Nafion-H
Counter Electrode Platinum black
Catholyte Pure CO
2
Anolyte 0.1 M LiOH
Cell Potential 3.6 V
Cathode Pressure 300 psig
Anode Pressure 305 psig
Cathode Flow 30 mL/min
Anode Flow 2 mL/min
Temperature 50 °C
Measured
Values
Average Current Density 24.2 mA/cm
2
Formate FE 39.3 %
CO FE 51.0 %
H
2
FE 10.2 %
Assumed Cost of Electricity 4 ¢/kWh
Cost of Electrochemical Formic Acid $427/tonne
90
Since a large portion of the current in Table 2.2 is going towards CO and H
2
(syngas), these products could potentially be sold as an additional revenue stream. This
additional revenue stream has not been considered in this analysis, however. Using these
conditions as a baseline, the only other necessary piece of information was the price of
electricity. The cost to heat up the electrolyzer to 50 °C has not been considered because
the energy efficiency of the electrolysis is low enough that it would almost certainly heat
itself up to and beyond 50 °C and may require cooling rather than heating. One additional
step in the process which has also not been considered is the acidification of the resulting
lithium formate to formic acid. This may well be a significant cost, but there is no readily
available information regarding this step, so it has not been taken into account. One could
instead determine the production cost per metric ton of lithium formate, but since this
would almost certainly be a very small volume product, it seems unlikely to be the end
point for our process.
Figure 2.17. Cost of Formic Acid as a Function of Cell Voltage
$150
$200
$250
$300
$350
$400
$450
$500
1.0 2.0 3.0 4.0
Cost ($/tonne)
Cell Potential (V)
91
With these input values, an initial conservative estimated value for production of
formic acid was calculated to be $427 per metric ton. Figures 2.17-2.19 illustrate the
dependence of the cost on various factors including operating voltage (Figure 2.17),
faradaic efficiency (Figure 2.18), and cost of electricity (Figure 2.19), with the baseline
conditions indicated with a red square. FE is the only factor which does not follow a
linear trend. One point to make about the FE, however, is that the other two products (CO
and H
2
) could also be sold and a certain amount of the cost could be recovered in that
way. There is no silver lining for the cost of cell voltage, however, and a more energy
efficient system would probably have the biggest impact on the cost of producing formic
acid from CO
2
using electrolysis. The cost of electricity also plays a major role in the cost
of formic acid produced through electrolysis.
Figure 2.18. Cost of Formic Acid as a Function of FE
$0
$100
$200
$300
$400
$500
$600
$700
$800
$900
0% 20% 40% 60% 80% 100%
Cost ($/tonne)
Faradaic Efficiency
92
Figure 2.19. Cost of Formic Acid as a Function of Electricity Cost
Although lithium formate isn’t an item that is typically sold in large volumes,
sodium and potassium formate are, and electrochemically produced sodium formate can
be directly compared to commercially available sodium formate since there is no
acidification step is involved. Using the same “baseline” conditions of Table 2.2, a price
of $553 per metric ton of sodium formate produced electrochemically from CO
2
can be
calculated. This includes the cost of sodium hydroxide, which was estimated to be $450
per metric ton. The baseline conditions call for LiOH as the anolyte, which would
produce lithium formate rather than sodium formate, but these are relatively conservative
estimates, so it seems plausible that conditions could be found to generate similar values
while using NaOH as an anolyte instead.
Based on reports in the literature, it seems that faradiac efficiency could be
improved significantly, potentially up to 90%, while current density could also increase
$150
$250
$350
$450
$550
$650
$750
$850
$950
0 2 4 6 8 10
Cost ($/tonne)
Electricity Cost (¢/kWh)
93
to above 100 mA/cm
2
. Cell potential should also drop with improved design and cell
components, as much of our cell potential is likely being lost to resistance since we
typically measure ~1 Ω across the 4 cm
2
cell. Reducing the cell potential to 3.0 V seems
like a reasonably achievable goal without resorting to exotic catalysts or methods. Given
90% FE and 3.0 V instead of 39% FE and 3.6 V gives a cost of formic acid of $155 per
metric ton, which is a huge improvement from $427 per metric ton that was estimated as
a conservative value. Again, it should be emphasized that this does not take into account
acidification, but the price/tonne will be quite similar for production of sodium formate,
so one can use that as a basis for comparison.
2.4.3 Conditions for CO production
Table 2.3 gives a summary of the conditions used and the measured values for a
conservative electrolysis experiment using a gold electrode to generate CO and H
2
. Since
sodium ions move from the anode to the cathode and regenerate sodium hydroxide, the
cost of sodium hydroxide was not included since it is not consumed, and thus could be
recovered and reused. Although catalyst poisoning has been reported to be an issue for
CO
2
reduction on gold electrodes, it does not appear to be a significant problem in the
system tested here, and likely is reversible in nature. Tin can also be used to generate
~90% syngas in ~2:1 ratio of H
2
:CO at 4.0 V and 100 mA/cm
2
. Rather than use tin as a
baseline in this analysis, however, gold will be considered since it is the best known
catalyst for the reduction of CO
2
to CO.
94
Table 2.3: CO Baseline Conditions
Controlled
Variables
Working Electrode Gold powder
Membrane Nafion-Na
Counter Electrode Platinum black
Catholyte Pure CO
2
Anolyte 0.1 M NaOH
Current Density 30 mA/cm
2
Cathode Pressure 600 psig
Anode Pressure 605 psig
Cathode Flow 90 mL/min
Anode Flow 2 mL/min
Temperature 40 °C
Measured
Values
Average Cell Potential 2.92 V
CO FE 56.0 %
H
2
FE 43.8 %
Assumed Cost of Electricity 4 ¢/kWh
Cost of Electrochemical CO $399/tonne
For Table 2.3, a set of experimental conditions were chosen that gave the highest
percentage of CO measured, rather than the lowest cell potential at a given ratio of
H
2
:CO. The gold catalyst has the characteristic of producing both H
2
and CO in varying
quantities depending on the conditions, so one can fine tune the ratio to the desired
composition for downstream processing. For methanol synthesis, a ratio of 2:1 H
2
:CO
would be desirable, so these conditions might not be ideal for that application. An
alternative would be to use a commercial electrolyzer or other source of H
2
to make up
the H
2
deficiency because it may prove cheaper overall; however it would increase the
complexity of the system by adding another step. The values given in Table 2.4 and
represented graphically in Figure 2.20 assume that conditions can either be found for
simultaneous production of H
2
and CO in a ratio of 2:1 (very close to our observations),
or a system capable of sustaining 95% faradaic efficiency for CO.
95
Table 2.4: Syngas Hypothetical Conditions
Simultaneous
Production
Separate H
2
O
Electrolysis
Separate H
2
from Methane
Hypothetical
Values
Cell Potential 3.0 V 3.0 V 3.0 V
CO FE 32 % 95 % 95 %
H
2
FE 63 % 0 % 0 %
Assumed
Cost of H
2
n/a $2700/tonne $750/tonne
Cost of Electricity 4 ¢/kWh 4 ¢/kWh 4 ¢/kWh
Cost of Syngas (H
2
:CO 2:1) $634/tonne $551/tonne $306/tonne
Figure 2.20. Production of Syngas in a 2:1 H
2
:CO Ratio by Various Methods
Commercial water electrolyzers produce exclusively hydrogen at a price of
$2560-2970 per metric ton, while hydrogen from natural gas can be produced for a much
lower price of $750 per metric ton. Taking these values into account, it does not appear
that the price per metric ton of syngas is significantly reduced by using H
2
from
$0
$100
$200
$300
$400
$500
$600
$700
Simultaneous
Production
Additional H2
Electrolyzer
Additional H2 from
Methane
Cost ($/tonne)
Extra H2
CO2 Electrolyzer
96
electrolysis. On the other hand, H
2
from methane lowers the cost by approximately 50%
compared to simultaneous production of CO and H
2
. Using H
2
from methane would not
solve any environmental problems, however, since the methane used to generate H
2
would release CO
2
back into the environment.
The trends for the cost per tonne of CO are the same as for formate production, so
Figures 2.16-2.18 will not be replicated for CO production here. Suffice to say that cell
potential and cost of electricity have the same linear influence on the cost of CO, while
the FE follows an exponential trajectory. If it is assumed that concurrent production of
CO and H
2
is the preferred route, then optimization of CO FE would likely not be critical.
Therefore, the most important performance metric to improve in this case is the cell
potential. Since gold is a slightly better catalyst for CO
2
reduction than tin, it is already
feasible to operate the cell at ~3.0 V and achieve reasonable current densities of ~30
mA/cm
2
. Getting values higher than this and closer to 100 mA/cm
2
may require higher
cell potentials, but it is not unreasonable to assume that improved system design (less
resistance, etc.) could achieve 100 mA/cm
2
with ≤ 3.0 cell voltage.
2.4.4 Comparison to commercial producers
Figure 2.21 compares the cost of formic acid and syngas produced from CO
2
electrochemically to the price of the same products produced commercially with existing
technologies. Although access to accurate and current prices for bulk chemicals was
elusive, rough prices for formic acid and methanol have been determined, which will be
used as a basis for comparison. On a large scale, it has been reported that formic acid
97
ranges from $890 per metric ton in Western Europe to $1,250 per metric ton in the
United States.
27
Both of these values compare favorably to the conservative estimate of
$426 per metric ton, although formate is the product rather than formic acid and
acidification has not been accounted for, among other costs. Assuming that the other
costs and the acidification step would double the cost, a revised estimate of $852 per
metric ton would still be competitive with Western Europe and significantly cheaper than
the US price. If improvements such as lower resistance and improved efficiency could be
realized upon scale up, this price would drop further to be more competitive on the world
market.
Whether or not there is a large enough market for formic acid to make this
feasible at the moment is not something that has been considered because simply
producing formic acid to sell on the current formic acid market was never the end goal of
this project. Instead, formic acid would ideally be converted into methanol for use as a
fuel, or used directly in formic acid fuel cells, or indirectly in hydrogen fuel cells using
formic acid decomposition catalysts. Given the size of the energy market, it seems that
the challenge would be to produce enough to make even a small dent in the energy
landscape.
98
Figure 2.21. Costs Compared
Syngas is not sold commercially to the best of our knowledge, so the price of
methanol was used as a substitute for it. Although this may seem a bit strange given that
we are proposing production of methanol from syngas, ~80%
28
of the cost of a methanol
production plant is taken up by the process to produce syngas. Thus, if we take 80% of
the cost of methanol, we have a rough guide for how much syngas costs industrially. If
we assume that methanol is ~$1.20 per US gallon, we get a price of $400 per metric ton
of methanol and, assuming 80% of that accounts for syngas, $320 per metric ton of
syngas. Since this syngas is clearly intended for methanol production, the ratio is likely
close to 2:1 H
2
:CO, which compares to our calculated value of $634 per metric ton
above. Using water electrolysis instead of simultaneous generation of CO and H
2
would
cut our cost to $551 per metric ton, or using hydrogen from methane would cut our cost
to $306 per metric ton.
$0
$200
$400
$600
$800
$1,000
$1,200
$1,400
Syngas Formic Acid
Price (USD/tonne)
Commercial
CO2 Electrolysis
99
While there would not be much benefit in producing CO electrochemically only
to combine it with H
2
produced from methane, it is interesting how close the cost of this
last scenario is to the estimated cost for syngas directly from methane. The cost for either
of the non-methane routes is significantly higher and would naturally lead to methanol
with a much higher price than what is available now, but with the benefit of being
completely fossil fuel free if we assume that electricity would be from non-CO
2
producing sources.
If we assume that syngas generation would still account for 80% of the total cost
of methanol, then methanol produced from electrochemically generated syngas would
cost $793 per metric ton, or $2.38 per US gallon, or roughly double the current price of
methanol. This translates into $4.85 per US gallon for the equivalent amount of energy
contained in one gallon of gasoline. Although this is higher than the cost of gasoline at
the pump in the US today, it is lower than the cost of gasoline at the pump in Europe. It
should also be emphasized that this price would be for methanol produced exclusively
from CO
2
, rather than methanol produced from natural gas or methane. Of course this
price does not include the taxes and various other fees, which are levied on gasoline
prices at the pump, so the comparison is not entirely valid.
2.5 Conclusion
A high pressure flow electrochemical reactor system for the conversion of CO
2
to
CO has been designed and demonstrated. Increased CO
2
pressure increases CO
2
reduction FE, as long as the system is not operating under mass transport limitations. The
100
continuous flow design lends itself to easy scale up, however it presents some challenges
particularly with high pressure CO
2
reduction, many of which were addressed here.
Although CO
2
reduction has been performed under high pressure
29
and flow conditions
before,
13
it has not, to our knowledge, been performed in a high pressure flow
configuration. This configuration significantly changed the product distribution of the
reaction, enabling the use of tin as a cathode material for the selective production of CO
from CO
2
. The effect of the electrolyte was surprisingly significant in our system, with
NaOH producing better results on gold, and Li
2
CO
3
producing better results on tin. It was
also shown that the loss of activity on a gold electrode is reversible, rather than a
permanent poison, suggesting that a change in conditions can lead to a non-trivial change
in the observed performance.
The results presented here highlight some of the possibilities and challenges that
face CO
2
reduction in a pressurized flow reactor. While a reactor system of this type
seems ideal for scale up since it operates under flow conditions (essential for a large scale
process) and under pressure (reducing its size and making downstream handling much
easier), there are several issues that make CO
2
reduction challenging under such
conditions. Typically, aqueous CO
2
reduction is done in neutral to slightly basic
conditions using electrolyte salts like NaHCO
3
.
6
In a PEM, the electrolyte is typically the
membrane itself, and (as we have reinforced here) the electrolyte can have a profound
effect on the FE distribution of various products.
16
While others have dealt with this issue
in a flow reactor by introducing a buffer layer between the electrode and the membrane,
16
that design would have been challenging to implement in a high pressure reactor. As a
101
result, the membrane itself must serve as the only electrolyte, leaving the other process
variables to control the electrolysis. This is likely why CO
2
reduction to CO on gold in
this system proceeds with only 10% FE (see Figure 2.8) at ambient pressure when others
report much higher FEs
4
under otherwise similar conditions.
High pressure solves some of the issues related to FE, however it did not generate
the high current densities and FEs that have been reported for static systems under
pressure.
29
Within the first hour of the experiment, FE was generally quite high for both
tin and gold, however this FE dropped off significantly as HER took over as the
predominant product in most cases. Using Li
2
CO
3
as the anolyte salt with a tin cathode
mitigated this problem, yielding a system which consistently favored CO
2
reduction
instead of HER over the course of the entire 8 hour experiment. A significant change in
product composition compared to what has been reported in the literature was also
observed when using a tin electrode at elevated pressure in a flow system. While tin has
generally been reported to produce primarily formate as a CO
2
reduction product, even
under pressure, only a small fraction of the CO
2
reduced under pressure in our flow
system ended up as formate. The majority of the CO
2
reduction current went towards CO,
allowing us to consider using this system for syngas production instead of formate
production. Tin is much cheaper than gold, and has a lower HER exchange current
density than gold, two distinct advantages which may be important in the scale up of this
process. That being said, the ability to run a continuous system under pressure should
surely be an advantage upon scale up and integration into existing chemical processing
plants, particularly Fischer-Tropsch plants, which typically run at elevated pressure.
102
2.6 Experimental
Membrane electrode assemblies (MEAs) were fabricated by first mixing slurries
of the desired metal catalyst (gold, tin, or platinum) with deionized water (18.2 MΩ-cm
Millipore filtered) and 5% Nafion binder. 99.87% pure tin nanopowder was used as
obtained from Sigma-Aldrich, with a particle size of <100 nm by TEM. 99.98% pure
spherical 0.6 micron gold powder from Alfa Aesar was used as obtained. High surface
area (45-52 m
2
/g) platinum black was used as obtained from Premetek. These slurries
were then applied to either 6% teflonized carbon paper or bare carbon paper cut to 2x2
cm squares using a detail nylon paint brush. After drying for at least 2 hours in a vacuum
oven at 60 °C, the electrodes were weighed to determine the mass loading. The electrodes
were then pressed onto a sheet of Nafion 117 using a hot press at 2000 lbs and 120 °C for
5 minutes. Upon removal from the press, the MEAs were cut to the correct dimension
using a 1-5/8” circular die. The MEAs were then placed between the anode and cathode
endplates and the entire enclosure was sealed by tightening the six bolts around the
perimeter of the cell in a star pattern to 30 in-lbs using a torque wrench. The cell was then
connected to the rest of the system and slowly pressurized to the desired pressure as
evenly as possible. A high flow rate was used initially to build pressure before the desired
flow rate was set. The desired conditions (pressure, temperature, and flow rate on each
side) were maintained for at least one hour before electrochemical experiments were
started. Anolyte solutions were prepared by first mixing the desired salt (NaOH,
NaHCO
3
, LiOH, Li
2
CO
3
) with the desired amount of deionized water and then pre-
103
electrolyzed by applying 1.05 V for at least 12 hours to two platinum wires immersed in
the solution while stirring.
Electrochemical measurements were carried out using a Solartron 1287
potentiostat controlled with a Windows PC running CorrWare. Analysis of the effluent
gasses was carried out every 15 minutes (the minimum amount of time required to fully
elute the CO
2
peak) using a Varian GC with a TCD detector and a Carboxen 1010 plot
column at 120 °C using argon as the carrier gas. Integration of the H
2
and CO peaks was
compared to reference standards of 1% CO or H
2
in argon. Analysis of the liquid
products in the gas-liquid separator was carried out on a Varian 400 MHz NMR
spectrometer by analyzing the liquid for any organic products. A known quantity of
dimethylformamide (DMF) was added to the gas-liquid separator and a small sample of
this mixture was taken, diluted with D
2
O, and analyzed by
1
H NMR. Integration of the
formate peak relative to DMF was used to determine FE.
2.7 References
(1) Olah, G. A.; Goeppert, A.; Prakash, G. K. S. Beyond Oil and Gas: The Methanol
Economy; John Wiley & Sons, Inc., 2009.
(2) Goeppert, A.; Czaun, M.; Jones, J.-P.; Surya Prakash, G. K.; Olah, G. A. Chem.
Soc. Rev. 2014.
(3) Goeppert, A.; Zhang, H.; Czaun, M.; May, R. B.; Prakash, G. K. S.; Olah, G. A.;
Narayanan, S. R. ChemSusChem 2014, 7, 1386.
(4) Hori, Y.; Murata, A.; Kikuchi, K.; Suzuki, S. J. Chem. Soc., Chem. Commun.
1987, 728.
(5) Hori, Y.; Wakebe, H.; Tsukamoto, T.; Koga, O. Electrochim. Acta 1994, 39,
1833.
104
(6) Hori, Y. In Mod. Aspect. Electroc.; Vayenas, C., White, R., Gamboa-Aldeco, M.,
Eds.; Springer New York: 2008; Vol. 42, p 89.
(7) Hori, Y.; Ito, H.; Okano, K.; Nagasu, K.; Sato, S. Electrochim. Acta 2003, 48,
2651.
(8) Chen, Y.; Li, C. W.; Kanan, M. W. J. Am. Chem. Soc. 2012, 134, 19969.
(9) Lu, Q.; Rosen, J.; Zhou, Y.; Hutchings, G. S.; Kimmel, Y. C.; Chen, J. G.; Jiao, F.
Nat. Commun. 2014, 5.
(10) Ma , S.; Lan , Y.; Perez, G. M. J.; Moniri, S.; Kenis , P. J. A. ChemSusChem 2014,
866.
(11) Delacourt, C.; Ridgway, P. L.; Newman, J. J. Electrochem. Soc. 2010, 157,
B1902.
(12) Mistry, H.; Reske, R.; Zeng, Z.; Zhao, Z.-J.; Greeley, J.; Strasser, P.; Cuenya, B.
R. J. Am. Chem. Soc. 2014.
(13) Delacourt, C.; Ridgway, P. L.; Kerr, J. B.; Newman, J. J. Electrochem. Soc. 2008,
155, B42.
(14) Shiratsuchi, R.; Nogami, G. J. Electrochem. Soc. 1996, 143, 582.
(15) Yano, H.; Shirai, F.; Nakayama, M.; Ogura, K. J. Electroanal. Chem. 2002, 533,
113.
(16) Delacourt, C.; Newman, J. J. Electrochem. Soc. 2010, 157, B1911.
(17) Ikeda, S.; Takagi, T.; Ito, K. Bull. Chem. Soc. Jpn. 1987, 60, 2517.
(18) Bonelli, B.; Civalleri, B.; Fubini, B.; Ugliengo, P.; Areán, C. O.; Garrone, E. J.
Phys. Chem. B 2000, 104, 10978.
(19) Farkas, A. P.; Solymosi, F. J. Phys. Chem. C 2009, 113, 19930.
(20) Malkhandi, S.; Bauer, P. R.; Bonnefont, A.; Krischer, K. Catal. Today 2013, 202,
144.
(21) Ursua, A.; Gandia, L. M.; Sanchis, P. Proc. IEEE 2012, 100, 410.
105
(22) Innocent, B.; Liaigre, D.; Pasquier, D.; Ropital, F.; Léger, J. M.; Kokoh, K. B. J.
Appl. Electrochem. 2009, 39, 227.
(23) Barton, E. E.; Rampulla, D. M.; Bocarsly, A. B. J. Am. Chem. Soc. 2008, 130,
6342.
(24) Rosen, B. A.; Salehi-Khojin, A.; Thorson, M. R.; Zhu, W.; Whipple, D. T.; Kenis,
P. J. A.; Masel, R. I. Science 2011, 334, 643.
(25) Simbeck, D.; Chang, E. Hydrogen Supply: Cost Estimate for Hydrogen Pathways-
Scoping Analysis, National Renewable Energy Laboratory, Golden, CO, 2002.
(26) OECD/NEA Projected Costs of Generating Electricity; OECD Publishing, 2005.
(27) Bizzari, S. N.; Blagoev, M. CEH Marketing Research Report: Formic Acid, SRI
consulting, 2010.
(28) Pavone, A. Mega Methanol Plants, SRI Consulting, 2003.
(29) Hara, K.; Kudo, A.; Sakata, T. J. Electroanal. Chem. 1995, 391, 141.
106
3 Difluoromethylation of Aryl (Heteroaryl) Iodides and β-Styrenyl
Halides using Copper Mediated Tributyl(difluoromethyl)stannane
3.1 Introduction
Fluorine is the 13
th
most abundant element in the earth’s crust but less than 15
naturally occurring organofluorine compounds (all of them being monofluorinated) are
known.
1-3
In other words, the entire field of organofluorine chemistry is synthetic or
manmade. Organofluorine compounds have found applications in a wide variety of fields
such as pharmaceuticals, agrochemicals, fuel cell membranes, refrigerants, materials,
surfactants, solvents, catalysts, etc.
4-8
Trifluoromethylation chemistry is an established
area of research and there exist a number of reagents such as
(trifluoromethyl)trimethylsilane (TMSCF
3
or the Rupert-Prakash reagent),
9-10
(trifluoromethyl)trimethoxyborate for nucleophilic trifluoromethylations,
11-16
Langlois
reagent for radical trifluoromethylations
17-18
or Umemoto’s and Togni’s reagents for
electrophilic trifluoromethylations.
19-24
However, the field of difluoromethylation is
relatively unexplored. Selective introduction of the difluoromethyl group (CF
2
H) into
organic molecules is of interest due to its special biological properties such as
enhancement of membrane permeability, binding affinity and bioavailability.
2,25-26
The
CF
2
H functionality is isosteric and isopolar to the hydroxyl (OH) group and is reasonably
hydrophobic.
27
At the same time the CF
2
H group is weakly acidic and capable of
participating in weak hydrogen bonding interactions. Because of these properties, the
CF
2
H group has been a part of various biologically active compounds like enzyme
inhibitors, sugars,
28-29
and agrochemicals such as pesticides and herbicides.
30
Therefore,
107
methods for introduction of the CF
2
H group have great potential in the areas of
pharmaceuticals, agrochemicals, and material science.
7,31-32
Several methods have been developed for the preparation of CF
2
H-containing
compounds, including the deoxofluorination of aldehydes using SF
4
, DAST and its
derivatives.
33-34
Magnesium metal mediated reductive difluoromethylation of
chlorosilanes using difluoromethyl sulfides, sulfoxides, sulfones
35
and nucleophilic
introduction of a CF
2
H group into a carbonyl compound has been reported using
difluoromethyl phenyl sulfone,
36
(difluoromethyl)dimethylphenylsilane and
(chlorodifluoromethyl)trimethylsilane.
27,37
Direct transfer of a difluoromethyl group to a
heteroarene using zinc difluoromethanesulfinate (DFMS), believed to proceed via a
radical pathway, was reported by Baran’s group.
38
Direct access to regiospecific
difluoromethylated arenes, however, has been a challenge until recently.
As shown in Scheme 3.1, Amii and coworkers reported a CuI catalyzed three-step
approach for the synthesis of difluoromethyl aromatics and hetero aromatics by a C-C
coupling reaction between aryl iodides and α-silyldifluoroacetates followed by hydrolysis
and decarboxylation.
39-40
Fier and Hartwig reported a one-step copper mediated (CuI)
nucleophilic difluoromethylation of iodoarenes using TMSCF
2
H.
41
Although the products
were obtained in excellent yields, the methodology requires significant excess of
TMSCF
2
H, is limited to electron-rich and electron-neutral iodoarenes, and the method is
not tolerant to aldehydes and ketones due to the competitive nucleophilic addition
reaction at the carbonyl center.
42-43
108
Scheme 3.1: Methods for Difluoromethylation
R
I
TMSCF
2
COOEt
R
CF
2
COOEt
1.hydrolysis
2. decarboxylation
R
CF
2
H
R
I
R
I
CuI, CsF
TMSCF
2
H, NMP
120 °C, 24h
R
CF
2
H
R
CF
2
H
or or
- Does not tolerate R= CHO,
COR or EWG
- Requires 5 eq. TMSCF
2
H
X
R
I
R
X'
CuI, KF
(n-Bu)
3
SnCF
2
H, DMA
100-120 °C, 24h
X
R
CF
2
H
R
CF
2
H
or or
- Tolerates R= CHO, COR,
COOEt
- Requires 2 eq.
(n-Bu)
3
SnCF
2
H
- X= CH, N
Amii et al. Org. Lett. 2011
Hartwig et al. JACS 2012
This work
X = CH, N
X' = I, Br
We were interested in developing a novel reagent for difluoromethylation of
iodoarenes tolerant towards electron-withdrawing as well as electron-donating
functionalities and particularly towards carbonyl groups. Given tin’s relatively low
affinity for oxygen compared to silicon, we hypothesized that a tin-based CF
2
H reagent
may be useful for ipso-difluoromethylation of iodoarenes preferentially in the presence of
carbonyl groups. Herein, the facile synthesis of tributyl(difluoromethyl)stannane
(TBTCF
2
H) and its application in the directed difluoromethylation of iodoarenes and β-
styrenyl halides is reported. This methodology has been extended to activated,
deactivated, and heterocyclic iodoarenes including carbonyl substituents in moderate to
good yields and β-styrenyl halides in good to excellent yields.
109
3.2 Synthesis of TBTCF
2
H
Equation 1: Decomposition of Me
3
SiCF
3
to CF
2
Carbene
Me
3
SiCF
3
Me
3
SiX+ CF
3
-
:CF
2
+ F
-
X
-
The Prakash group has demonstrated that singlet CF
2
carbene can be generated
using TMSCF
3
(Equation 1) in the presence of tetra-n-butylammonium
difluorotriphenylsilicate (TBAT), and NaI.
44
This carbene can then add across an alkene
or alkyne by a [2+1] cycloaddition reaction. We wanted to use this methodology to insert
a CF
2
carbene into a metal-hydrogen bond. While screening various compounds
containing metal-hydrogen bonds, it was observed that the CF
2
carbene could be easily
inserted into the Sn-H bond of (n-Bu)
3
Sn-H to afford (n-Bu)
3
SnCF
2
H under mild
conditions in isolated yields >80%. Cullen et al. had previously reported insertion of CF
2
carbene generated from Me
3
SnCF
3
into the Sn-H bond of timethyltin hydride under harsh
conditions (150
o
C, 24h) reporting 63% yield of the insertion product.
45
Results from
optimization of this reaction with respect to solvent and equivalents of TMSCF
3
are
shown in Table 3.1. Although this reaction takes place at room temperature, an induciton
period exists which makes it advisable to run the reaction at slightly elevated
temperature. Temperature increases dramatically after the induction period, so high
temperatures are not necessary, however.
110
Table 3.1: CaI
2
Initiated Optimization of (n-Bu)
3
SnCF
2
H Synthesis
[a]
Entry Me
3
SiCF
3
(equiv.)
Solvent (4 mL) %Yield (2a)
b
1 2.25 DMA 85
2 2.25 DMF 72
3 2.25 NMP 86
4 2.25 THF 0
5 2.25 THF:HMPA (1:1) 0
6 2.25 DMSO 0
7 1.1 NMP 63
8 1.4 NMP 79
9 1.7 NMP 82
[a] Reaction conditions: (n-Bu)
3
SnH: 8.68 mmol, CaI
2
: 0.34 mmol [b] Determined by
19
F
NMR
Compared to sodium iodide and several fluoride initiators, we found calcium
iodide to be an ideal initiator to generate CF
2
carbene from TMSCF
3
at 45
o
C, giving
TBTCF
2
H (2a) as the major product in less than an hour. The reaction proceeds in a
variety of polar aprotic solvents (Table 3.1) such as dimethylformamide (DMF),
dimethylacetamide (DMA) and n-methylpyrrolidone (NMP). Although a slight (10 %)
excess of TMSCF
3
still gives 63 % yield of 2a (Table 3.1, Entry 7), increasing the
amount to 2.25 equivalents increased the yield to 86 % (Table 3.1, Entry 3) in NMP. The
only significant side product is (n-Bu)
3
Sn-CF
3
, which can be easily removed by adding a
1 M solution of TBAF in tetrahydrofuran (THF). TBAF selectively reacts with (n-
Bu)
3
Sn-CF
3
at room temperature and converts it into (n-Bu)
3
Sn-F within an hour. During
(n-Bu)
3
SnH
+
Me
3
SiCF
3
(n-Bu)
3
SnCF
2
H
+
Me
3
SiF
Solvent, CaI
2
45 °C, 60 min
2a 1a
111
this period any unreacted TMSCF
3
also reacts with fluoride to form the volatile products
TMSF and CF
3
H. After extraction with dichloromethane (DCM), 2a can be obtained by
eluting with hexanes over a plug of silica. Conversion of (n-Bu)
3
Sn-CF
3
to (n-Bu)
3
Sn-F
was necessary as 2a and (n-Bu)
3
Sn-CF
3
elute together and could not be easily separated
by column chromatography.
Table 3.2: Conversion of R
3
SnH to R
3
SnCF
2
H
[a]
[a] Reaction conditions: All reactions were carried out in 5 gram scale of starting R
3
SnH,
2.2 equiv. of TMSCF
3
, 0.07 equiv. CaI
2
, 45-50 °C for 1 hour
The purification method reveals the lower reactivity of 2a compared to TMSCF
3
and (n-Bu)
3
SnCF
3
. This procedure works generally for R
3
SnH to R
3
SnCF
2
H (Table 3.2),
however we focused on 2a due to its easy synthesis, non-volatile nature and stability in
air. The starting tributyltin hydride is also one of the few tin hydrides which is
inexpensive, commercially available, has relatively low toxicity (because of low
volatility) and is stable compared to other analogues such as Ph
3
SnH, (C
6
H
11
)
3
SnH and
Me
3
SnH. To the best of our knowledge, this is the first reported synthesis of TBTCF
2
H
(2a), triphenyl(difluoromethyl)stannane (2c), and tricyclohexyl(difluoromethyl)stannane
(2d). This procedure can be carried out in multi-gram scale to yield pure 2a, 2c, and 2d
R
3
SnH
+
Me
3
SiCF
3
R
3
SnCF
2
H
+
Me
3
SiF
CaI
2
, DMA,
1a-d 2a-d
product # isolated yield (
19
F NMR yield)
2a 80% (86%) 2b (54%) 2c 54% (65%) 2d 57% (67%)
Sn CF
2
H
n-Bu
n-Bu n-Bu
Sn CF
2
H
C H
3
C H
3
CH
3
Sn CF
2
H
Ph
Ph Ph
Sn CF
2
H
Cy
Cy Cy
112
using simple protocols under an inert environment. The isolation of
trimethyl(difluoromethyl)stannane (2b) proved to be a challenge (due to its volatility),
and was not pursued further. Compound 2a is easily purified and can be stored for long
periods of time (weeks to months at least) in air without decomposition.
3.3 Copper mediated difluoromethylation of iodoarenes with TBTCF
2
H
After extensive screening work to determine conditions suitable for the copper
mediated difluoromethylation of iodoarenes, the desired product (Scheme 3.1) was
observed using CuI in DMA with KF as an initiator. Reaction time, temperature,
concentration, equivalents of reagent, amounts of CuI, and KF, and the type of aryl
substituent were all parameters which were systematically optimized. The reaction was
found to be sensitive to the amounts of CuI, KF, 2a, and the ratio of KF to CuI. Statistical
analysis identified two sets of optimized conditions; method A which consists of using
1.3 equiv. CuI, 3 equiv. KF, 2 equiv. 2a, 24 h and 100
o
C and works for β-styrenyl
halides, iodonaphthalenes and iodo substituted aldehydes/ketones. Method B works for
Br, Cl, Ph, and CF
3
substituted iodoarenes and calls for 1.3 equiv. CuI, 3 equiv. KF, 3
equiv. 2a, 24h, and 120
o
C. The use of CsF as the initiator gives the desired product in
lower yield. Only traces of the desired product were observed when other Cu(I) halides
such as CuBr and CuCl were used. Less than 5% of product was observed by
19
F NMR in
the control reactions when no KF was added.
113
Table 3.3: Difluoromethylation of Iodoarenes using 2a
[a]
[a] Reaction conditions: Iodoarene: 0.5 mmol, 2a: 1-1.5 mmol, CuI: 0.65 mmol, KF: 1.5
mmol, 4 mL DMA, under N
2
, 20 mL microwave vial at 100-120 °C, 24 h
X
R
I
X
R
CF
2
H
3a-r
CF
2
H
O
Br
CF
2
H
CF
2
H
3f 71% (78%)
3a 61% (69%)
3e 74% (80%)
3q (59%)
3c 61% (64%)
3g 66% (71%)
2a: 2-3 eq.
CuI: 1.3 eq.
KF: 3 eq.
100-120 °C
DMA, 24 h
F
3
C CF
2
H
3p (78%)
OHC
CF
2
H
CF
2
H
O
F
3
C
CF
2
H
3o 51% (55%)
CF
2
H
Ph
3d 51% (53%)
CF
2
H
CF
2
H
Ph
O
3n 44% (48%)
CF
2
H
CHO
3b 70% (82%)
CF
2
H
H
3
CO
O
3h 75% (78%)
product # isolated yield (
19
F NMR yield)
Method A: (n-Bu)
3
SnCF
2
H: 2 eq., 100 °C
Method B: (n-Bu)
3
SnCF
2
H: 3 eq., 120 °C
CF
2
H
CN
3i 65% (68%)
N CF
2
H
3j 60% (67%)
N
Br
CF
2
H
3l (75%)
N CF
2
H
Br
3k 49% (64%)
3m (32%)
CF
2
H MeO
Cl
CF
2
H
3r (50%)
X = CH, N
114
Table 3.3 lists the results for the conversion of a series of iodoarenes and
iodoheteroarenes to their respective CF
2
H substituted arenes or heteroarenes using DMA
as a solvent, which gives slightly increased yields over analogous solvents such as DMF,
1,3-dimethyl-3,4,5,6-tetrahydro-2(1H)-pyrimidinone (DMPU) or NMP. Although yields
vary, the method is extremely selective, and amenable to several types of functionalities
including aldehydes, ketones, esters and other halogens. Iodo substituted heteroarenes
(3j, 3k, and 3l) also gave the desired products. Reactions with bromoarenes as starting
materials gave only low yields, rarely >5%. In products 3k, 3l, and 3o the bromo
substitutents remain intact. This selectivity could be exploited by using bromides such as
3k, 3l, and 3o as starting materials in coupling reactions
46
to generate more complex
difluoromethylated molecules. Compounds 3l, 3m and 3p-3r were not isolated due to
their volatility or relatively low yield.
115
3.4 Copper mediated difluoromethylation of β–styrenes with TBTCF
2
H
Table 3.4: Difluoromethylation of β-styrenyl halides using 2a
[a]
[a] Reaction conditions: β-styrenyl halide: 0.5 mmol, 2a: 1 mmol, CuI: 0.65 mmol, KF:
1.5 mmol, 4 mL DMA, under N
2
, 20 mL microwave vial at 100 °C, 24 h
Based on the success of the copper-catalyzed ipso-substitution reaction detailed
above, we assumed that β-styrenyl substituents should also undergo similar reactions.
Our findings supported this premise, giving yields 20-30% higher than those for similar
iodoarenes. The β-styrenyl bromides gave the desired products, albeit at slightly lower
yields than their iodo counterparts. No isomerization of the double bond was observed
during the reaction with the β-styrenyl halides, indicating that the reaction proceeds with
retention of configuration. Results of experiments with substituted styrenyl iodides (cis,
R
X
R
CF
2
H
4a-h
4a 85% (94%) 4b 77% 4c 72%
product # isolated yield (
19
F NMR yield)
4d 76%
2a: 2 eq.
CuI: 1.3 eq.
KF: 3 eq.
DMA, 100 °C, 24 h
CF
2
H
Me
CF
2
H
CF
2
H
CF
2
H
Br
CF
2
H
Br
CF
2
H
F
Ph
Ph CF
2
H
Ph
4e 68% 4f 60%
4h 32%
CF
2
H
4g (48%)
X = I
X = Br
116
trans, electron-rich, and electron poor) are presented in Table 3.4 and showed similar
reactivity, giving the desired products under method A in good to excellent yields (60-
85%).
3.5 Mechanistic studies
3.5.1 Part 1: transmetallation
To better understand the mechanism of the reaction and related intermediates, an
extensive computational study of the hypothesized copper-CF
2
H species
47
was initiated,
analogous to the reported copper-CF
3
species which are well investigated.
16,48-51
Computational studies were based on DMF because it is easier to model than DMA and
the solvents produced similar results experimentally. The substituents on TBTCF
2
H (2a)
were also simplified from n-butyl to methyl. All calculations were conducted using the
Gaussian 09
52
computational package, using the B3LYP/cc-pVTZ level of theory for
neutral reactions and B3LYP/aug-cc-pVDZ for reactions involving anions.
Since an unstable CF
2
H anion is not likely to exist independently in solution
(especially since nucleophilic addition to carbonyl groups was not observed), we then
began to look for a reasonable mechanism for transferring CF
2
H from TBTCF
2
H (2a) to a
Cu-CF
2
H species similar to what has been reported in the literature.
53-54
Interestingly, no
transition state could be found by simply combining TBTCF
2
H (Scheme 3.2, Structure A)
and CuI (or CuCF
2
H) in a plausible way. Only when the pentacoordinate anion B was
first formed was a transition state successfully optimized involving CuI (Scheme 3.2,
117
Structure D). Similarly, a suitable transition state involving CuI when starting from the
octahedral dianion C was not found. If no productive transition states exist when starting
from A or C, then the necessary amounts of CuI and KF should be closely tied. Indeed,
extensive experimental screening work showed a high degree of correlation between the
amounts of KF and CuI. High concentrations of KF relative to CuI left TBTCF
2
H
unreacted, while high concentrations of CuI relative to KF produced mostly by-products
such as CF
2
H
2
. This observation is taken to be evidence that the reaction is highly
dependent on the amount of "active" reagent which, based on computational modelling, is
believed to be [(n-Bu)
3
SnCF
2
HF]
-
(Scheme 3.2, Structure B).
Scheme 3.2: DFT calculations for part 1: transmetallation
a
[a] Calculations performed at B3LYP/aug-cc-pVDZ, free energies (ΔG) reported in
kcal/mol
Low temperature
19
F NMR studies at -30 °C were undertaken to slow the fast
exchange of F
-
ions that was observed at higher temperatures. TBTCF
2
H (2a) was
118
disolved in DMF-d
7
with a sub-stoichometric amount of tetramethylammonium fluoride
(TMAF). TMAF was used instead of KF due to to the increased solubility of TMAF
compared to KF, particuarly at low temperature. NMR analysis (Figure 3.1) showed two
distinct peaks at -136.7 ppm and -148.1 ppm which were assigned to Sn-F species given
the large
1
J coupling constant (1780 Hz and 2024 Hz, respectively)
55
of the
119
Sn
satellites present in the spectrum. TMAF was also observed (-79.7 ppm) and HF
2
-
species
(-145.5, J
H-F
= 120 Hz) as reported in literature.
56-57
Additional peaks were also observed
overlapping in the region between -125.5 and -126.2 ppm where 2a is generally detected,
thus indicating that an activated CF
2
H species had potentially formed (Scheme 3.2,
Structure B). Unfortunately, the region around -125.5 to -126.2 ppm became too crowded
to deconvolute and quantify the various species present in the spectrum. These
observations occured before observation of CF
2
H
2
(which was observed after heating to
room temperature), indicating that 2a had not yet decomposed to tri-n-butyltin fluoride.
On warming to room temperature (Figure 3.2) and leaving the solution to stand for two
days, the “Sn-F” species disappeared, the additional signals around the TBTCF
2
H (2a)
peak disappeared, and CF
2
H
2
(-144.0 ppm, J
H-F
= 50.2 Hz) was the only product
observed.
119
Figure 3.1. Low temperature (-30 °C)
19
F NMR spectrum of TBTCF
2
H (2a) with TMAF
120
Figure 3.2. Ambient temperature
19
F NMR spectrum of TBTCF
2
H (2a) with TMAF after
two days
3.5.2 Part 2: ipso-substitution
The second part in the overall mechanism is the ipso-substitution where the
hypothesized copper-CF
2
H species substitutes the iodide on an arene, as illustrated in
Scheme 3.3. This transformation is more straightforward than the first part, with
coordination the likely rate determining step. It is possible that there is a higher energy
barrier to oxidative addition since we were unable to find a transition state, however it
seems unlikely given the small energy differences between the coordinated species and
the square planar products.
121
Scheme 3.3: DFT calculation for part 2: ipso-substitution
a
[a] Calculations performed at B3LYP/cc-pVTZ, free energies (ΔG) reported in kcal/mol
Adding solvent considerations (using IEFPCM) into the calculation resulted in a
decrease in free energy of 8.0 kcal/mol of the hypothesized CuCF
2
H intermediate,
however a significant stabilization beyond that of 14.1 kcal/mol was achieved by
explicitly including one molecule of DMF with the oxygen atom directly interacting with
the copper. Additional solvent molecules did not yield stable structures, so a single,
oxygen interacting DMF molecule was used in all calculations involving a copper (I)
species with only one formal ligand. In comparing the Cu-CF
3
species to the Cu-CF
2
H
species, it was found that adding a molecule of DMF to the Cu-CF
2
H anion increased
stabilization by 8.5 kcal/mol more than adding a molecule of DMF to the Cu-CF
3
anion.
122
Table 3.5: Comparison of Copper Mediated Difluoromethylation
[a]
Substrate
ΔG for
Coordination
(kcal/mol)
ΔG for Oxidative
Addition
(kcal/mol)
Iodobenzene 12.7 -0.7
trans- β-iodostyrene 8.8 2.8
cis- β-iodostyrene 8.8 -2.0
Bromobenzene N/A N/A
trans- β-bromostyrene 9.8 1.2
Chlorobenzene N/A N/A
trans- β-chlorostyrene 9.7 2.4
[a] Calculations done using the B3LYP/cc-pVTZ level of theory for optimizations and
energies using the Gaussian 09 computational chemistry package
A similar mechanism was modeled for the -styrenyl equivalent (both cis and
trans isomers), with results presented in Table 3.5. The results of this computational
study correlate relatively well with our experimental results, i.e. that -styrenyl halides
should be easier to difluoromethylate than arenes, and that isomerization around the
double bond is not expected. The difference between cis- and trans- -iodostyrene is
within error for coordination, the hypothesized rate determining step. Although
iodostyrenes should react more readily than bromostyrenes, the difference was not found
to be profound experimentally. Coordination complexes with bromobenzene and
chlorobenzene could not be found computationally. Interaction complexes could be
formed between the Cu-CF
2
H and the haloarenes, but not with the Cu-CF
2
H positioned
close to the halogen. This may explain why we have been unable to find conditions to
efficiently convert bromo and chloroarenes to their respective CF
2
H products. It may be
123
that a different catalyst or ligand system is required to facilitate the formation of the
productive η
2
complex.
3.5.3 Overall mechanism
Part 1 (the transmetalation step) involves the transfer of CF
2
H from tin to copper,
and likely is rate determining based on computational studies, while part 2 (ipso-
substitution) involves the reaction between the proposed CuCF
2
H species and the starting
material. Direct observation of CuCF
2
H was unsuccessful, likely due to the short half-life
of the species at higher temperatures (>-30 °C) according to previous reports.
54
The
19
F
NMR peak assigned to Cu(CF
2
H)
2
by Hartwig
41
for difluoromethylation using TMSCF
2
H
was not observed when 2a was used. Therefore, we propose that difluoromethylation
using 2a is mechanistically different from difluoromethylation using TMSCF
2
H. This
mechanistic difference is highlighted by the preference of TMSCF
2
H for iodobenzenes
with electron donating groups as opposed to the preference of 2a for iodobenzenes with
electron withdrawing groups. The highest energy transition state for part 2 (the ipso-
substitution step) could be either the coordination of CuCF
2
H to the starting material or
the oxidative addition. There are two competing pathways for the reaction of CuCF
2
H,
namely decomposition
54
and coordination to the starting material. This competition is the
likely cause of varying yields of the difluoromethylated products. The product of
oxidative addition is a copper (III) species formally, which seems unusual at first, but has
been postulated before as a reactive intermediate for copper mediated coupling
reactions.
58
124
Although CuI should be catalytic (as we start and end with it in our mechanism),
we did not find good experimental evidence to suggest that the reaction is truly catalytic
in CuI. At best, one could describe the reaction as sub-stoichiometric in CuI under certain
reaction conditions. This indicates that there may be other pathways for decomposition of
the copper species to insoluble (aggregates) or otherwise unreactive products. A suitable
ligand for copper may yield a system that is truly catalytic, however development of this
possibility was not pursued. A few readily available ligands were added to various
reaction mixtures, but no positive benefit was observed. An improved catalyst could
definitely benefit this particular reaction since yield of some products is less than stellar
and copper iodide is intimately involved with both TBTCF
2
H (2a) and the aryl starting
material.
3.6 Conclusion
In summary, an efficient procedure for the preparation of difluoromethylated tin
compounds from TMSCF
3
has been developed. The primary difluoromethylated tin
compound studied was TBTCF
2
H (2a) due to its long term stability and ease of handling,
which allowed it to be stored in air for weeks to months without decomposition. We
believe that this compound could be a useful difluoromethylating reagent given its
stability and ease of handling. A single-step transformation of iodoarenes and
iodoheteroarenes to their ipso-CF
2
H substituted counterparts using TBTCF
2
H (2a)
mediated by copper has also been demonstrated. Compared to difluoromethylation using
TMSCF
2
H, difluoromethylation using TBTCF
2
H (2a) is extremely selective, allowing
125
iodoarenes with carbonyl functional groups to be used directly. The difluoromethylation
protocol gives even higher yields with β-styrenyl iodides, where the reactivity is such that
β-styrenyl bromides can be used to give reasonable yields. A mechanism for the transfer
of CF
2
H from tin to a copper intermediate and subsequent reaction with iodoarenes was
proposed based on extensive computational and low temperature NMR work. Choice of
solvent is critical in this reaction because the solvent molecules must directly stabilize the
Cu-CF
2
H species. Given the solvent’s direct role in the reaction, development of the
copper catalyst would likely be the most effective means to improve product yield and
increase the scope of the reaction to include starting materials that are not feasible using
this approach.
3.7 Experimental
3.7.1 Materials and Instrumentation
1
H,
13
C,
19
F and
119
Sn NMR spectra were recorded on Varian 500 MHz or 400
MHz NMR spectrometers.
1
H NMR chemical shifts were determined relative to the
signal of a residual protonated solvent, CDCl
3
(δ 7.26 ppm).
13
C NMR chemical shifts
were determined relative to the
13
C signal of solvent, CDCl
3
(δ 77.23 ppm).
19
F NMR
chemical shifts were determined relative to CFCl
3
as an internal standard (δ 0.0 ppm).
119
Sn NMR chemical shifts were determined relative to Me
4
Sn as an internal standard (δ
0.0 ppm). Mass spectral data were recorded on a Bruker 300-MS TQ Mass Spectrometer
at 70 eV for EI and 20 eV for CI (CH
4
was used as the reagent gas) or an Agilent 6120
126
MS. HRMS data was obtained from University of Arizona Mass Spectrometry Facility.
Unless otherwise mentioned all the reactants, reagents and solvents were purchased from
commercial sources.
3.7.2 Synthesis of (n-Bu)
3
SnCF
2
H (2a)
In a glove box, CaI
2
(3.4 mmol, 1.0 g), (n-Bu)
3
SnH (86.8 mmol, 25.3g), TMSCF
3
(130.2 mmol, 18.5 g), were added to a 185 mL pressure tube (rated to 150 psi) that was
equipped with a stirring bar. This was followed by addition of 40 mL of NMP (1-Methyl-
2-pyrrolidone) and then the pressure tube was sealed and taken out of the glove box.
After a brief induction period at 45 °C, initiation of the reaction is visible due to the
dissolution of the solids and darkening of the solution. This is also accompanied by
increase in pressure up to 4 atm (60 psi) and temperature rising up to 90
o
C. After 1h, the
solution was allowed to cool to room temperature, and 20 mL of TBAF solution (1M in
THF) was added at room temperature and stirred for 45 min. The resulting mixture was
extracted with dichloromethane (4 X 100 mL), and the combined organic phase was
washed with water three times and then dried over magnesium sulfate. The solvent was
removed by rotary evaporation and purified by column chromatography on silica gel with
hexanes to give analytically pure TBTCF
2
H (2a) product in >99% purity and 80%
isolated yield.
Characterization:
1
H NMR (500 MHz, CDCl
3
) δ 6.41 (t, J = 44.9 Hz, 1H), 1.58 –
1.50 (m, 6H), 1.31 (dq, J = 14.7, 7.3 Hz, 6H), 1.07 – 1.02 (m, 6H), 0.89 (t, J = 7.3 Hz,
9H).
13
C NMR (126 MHz, CDCl
3
) δ 130.14 (t, J = 280.3 Hz), 28.95 (s), 27.45 (s), 13.83
127
(s), 9.33 (t, J = 2.5 Hz).
19
F NMR (470 MHz, CDCl
3
) δ -125.11 (d, J = 45.0 Hz).
119
Sn
NMR (186 MHz, CDCl
3
) δ -47.98 (t, J = 217.5 Hz). HRMS (EI, m/z) for (C
13
H
28
F
2
Sn) :
calculated 291.1135 [(M-CF
2
H)+]; found 291.1155 [(M-CF
2
H)+].
3.7.3 Synthesis of Me
3
SnH
Me
3
SnH was synthesized according to the procedure described by Lipshutz and
Reuter.
59
3.7.4 Synthesis of Me
3
SnCF
2
H (2b)
In a glove box, CaI
2
(1.8 mmol, 0.530 g), Me
3
SnH (35.4 mmol, 5.87g), TMSCF
3
(70.7 mmol, 10.1 g), were added to a 185 mL pressure tube (rated to 150 psi) that was
equipped with a stirring bar. This was followed by addition of 30 mL of DMA
(dimethylacetamide) and then the pressure tube was sealed and taken out of the glove
box. After a brief induction period at 45 °C, initiation of the reaction is visible due to the
dissolution of the solids and darkening of the solution. This is also accompanied by
increase in pressure up to 6 atm (90 psi) and temperature rising up to 90
o
C. After 1h, the
solution was allowed to cool to room temperature, and 20 mL of TBAF solution (1M in
THF) was added at room temperature and stirred for 45 min. The resulting mixture was
extracted with dichloromethane (4 X 25 mL), and the combined organic phase was
washed with water three times and then dried over magnesium sulfate.
Characterization:
19
F NMR (470 MHz, CDCl
3
) δ -127.37 (d, J = 45.2 Hz). The
1
H
NMR and
13
C NMR were found to be consistent with previously reported spectra.
45
128
3.7.5 Synthesis of Ph
3
SnCF
2
H (2c)
In a glove box, CaI
2
(1.02 mmol, 0.300g), Ph
3
SnH (14.2 mmol, 5 g), TMSCF
3
(28.5 mmol, 4.19 g), were added to a 20 mL microwave vial that was equipped with a
stirring bar. This was followed by addition of 10 mL of DMA and then the vial was
sealed and taken out of the glove box. After a brief induction period at 50 °C, initiation of
the reaction is visible due to the dissolution of the solids and darkening of the solution.
After 1h, the solution was allowed to cool to room temperature, and 10 mL of TBAF
solution (1M in THF) was added at room temperature and stirred for 45 min. The
resulting mixture was extracted with dichloromethane (4 X 10 mL), and the combined
organic phase was washed with water three times and then dried over magnesium sulfate.
The solvent was removed by rotary evaporation and purified by column chromatography
on silica gel with hexanes/ ethyl acetate to give analytically pure
triphenyl(difluoromethyl)stannane product in >99% purity and 54 % isolated yield.
Characterization:
1
H NMR (500 MHz, CDCl
3
) δ 7.63 – 7.58 (m, 6H), 7.46 – 7.41
(m, 9H), 6.84 (t, J = 44.7 Hz, 1H).
13
C NMR (126 MHz, CDCl
3
) δ 137.39 (s), 135.36 (t, J
= 2.7 Hz), 130.04 (t, J = 6.1 Hz), 129.90 (s), 129.84 (t, J = 281.4 Hz), 129.18 (s).
19
F
NMR (470 MHz, CDCl
3
) δ -122.62 (d, J = 44.7 Hz).
119
Sn NMR (149 MHz, CDCl
3
) δ -
176.30 (t, J = 278.2 Hz). HRMS (EI, m/z) for (C
19
H
16
F
2
Sn): calculated 351.0196 [(M-
CF
2
H)+]; found 351.0208 [(M-CF
2
H)+].
129
3.7.6 Synthesis of Cy
3
SnCF
2
H (2d)
In a glove box, CaI
2
(1.02 mmol, 0.150 g), Cy
3
SnH (13.5 mmol, 5g), TMSCF
3
(27
mmol, 3.8 g), were added to a 20 mL microwave vial that was equipped with a stirring
bar. This was followed by addition of 10 mL of DMA and then the vial was sealed and
taken out of the glove box. After a brief induction period at 50 °C, initiation of the
reaction is visible due to the dissolution of the solids and darkening of the solution. After
1h, the solution was allowed to cool to room temperature, and 10 mL of TBAF solution
(1M in THF) was added at room temperature and stirred for 45 min. The resulting
mixture was extracted with dichloromethane (4 10 mL), and the combined organic
phase was washed with water three times and then dried over magnesium sulfate. The
solvent was removed by rotary evaporation and purified by column chromatography on
silica gel with hexanes to give analytically pure tricyclohexyl(difluoromethyl)stannane
product in >99% purity and 57% isolated yield.
Characterization:
1
H NMR (399 MHz, CDCl
3
) δ 6.47 (t, J = 44.6 Hz, 1H), 1.97 –
1.83 (m, 6H), 1.73 – 1.56 (m, 18H), 1.40 – 1.20 (m, 9H).
13
C NMR (100 MHz, CDCl
3
) δ
132.26 (t, J = 281.2 Hz), 32.34 (s), 29.27 (s), 28.04 (t, J = 2.1 Hz), 27.18 (t, 3.2 Hz).
19
F
NMR (470 MHz, CDCl
3
) δ -123.04 (d, J = 44.6 Hz).
119
Sn NMR (149 MHz, CDCl
3
) δ -
119.38 (t, J = 189.4 Hz). HRMS (EI, m/z) for (C
19
H
34
F
2
Sn): calculated 369.1604 [(M-
CF
2
H)+]; found 369.1589 [(M-CF
2
H)+].
130
3.7.7 Synthesis of (2-halovinyl)arenes
All the E/Z (2-halovinyl)arenes were synthesized according to literature
procedures with minor changes wherever required. 4a, 4g and 4h were synthesized
according to a procedure reported by Petasis,
60
4b-4d were synthesized from the
corresponding benzylbromides
61
and 4e and 4f were synthesized according to a procedure
reported by Thadani et al.
62
3.7.8 Synthesis of difluoromethylated arenes and styrenes
Method A (3a-m, 4a-h): To a 20 mL microwave vial equipped with a magnetic
stir bar was added anhydrous KF (0.087 g, 1.5 mmols), CuI (0.124 g, 0.65 mmol) aryl
iodide (0.5 mmol) and TBTCF
2
H (2a, 0.341 g, 1 mmol) in that order under nitrogen
atmosphere in a glove box. To the above reaction mixture 4 mL DMA (N,N-
dimethylacetamide) was added and heated to 100
o
C in an oil bath protected from light
for 24 h. The reaction mixture was filtered and washed with 20 mL dichloromethane. The
resulting filtrate was extracted with dichloromethane, and the combined organic phase
was washed with water (5 x 10 mL) and then dried over magnesium sulfate. The solvent
was removed by rotary evaporation. Pure product was isolated after silica-gel column
chromatography with either hexanes/ethylacetate or pentane/diethylether to afford the
difluoromethylated arene (heteroarene), β-difluromethylated styrene product.
Method B (3n-r): To a 20 mL microwave vial equipped with a magnetic stir bar
was added anhydrous KF (0.087 g, 1.5 mmols), CuI (0.124 g, 0.65 mmol), aryl iodide
(0.5 mmol) and TBTCF
2
H (2a, 0.512 g, 1.5 mmol) in that order under nitrogen
131
atmosphere in a glove box. To the above reaction mixture 4 mL DMA (N,N-
dimethylacetamide) was added and heated to 120
o
C in an oil bath protected from light
for 24 h. The reaction mixture was filtered and washed with 20 mL dichloromethane. The
resulting filtrate was extracted with dichloromethane, and the combined organic phase
was washed with water (5 x 10 mL) and then dried over magnesium sulfate. The solvent
was removed by rotary evaporation. Pure product was isolated after silica-gel column
chromatography with either hexanes/ethylacetate or pentane/diethylether to afford the
difluoromethylated arene (heteroarene), β-difluromethylated styrene product.
3.7.9 VT NMR studies
Tetramethylammonium fluoride (5 mg, 0.05 mmol) and 0.7 mL DMF-d
7
were
added to a J. Young valve NMR tube in the glove box. The solution was taken out of the
glove box and cooled to -50
o
C at which point TBTCF
2
H (2a, 25 mg, 0.015mmol) and
C
6
F
6
(1 L) were added. The reaction was monitored by VT NMR (-30 °C). C
6
F
6
was
used as internal reference standard (-164.9 ppm).
132
3.7.10 Representative Spectra:
TBTCF
2
H (2a)
1
H NMR (CDCl
3
)
133
13
C NMR (CDCl
3
)
134
19
F NMR (CDCl
3
)
135
119
Sn NMR (CDCl
3
)
136
4-(difluoromethyl)benzaldehyde (3a):
1
H NMR (CDCl
3
)
137
13
C NMR (CDCl
3
)
138
19
F NMR (CDCl
3
)
139
(E)-(3,3-difluoroprop-1-en-1-yl)benzene (4a):
1
H NMR (CDCl
3
)
140
13
C NMR (CDCl
3
)
141
19
F NMR (CDCl
3
)
142
3.7.11 Computational details
All calculations were performed using the Gaussian 09
52
Rev. A.02 computational
chemistry program installed on a computer with an Intel
®
Core i7 920 processor (4 cores,
8 threads, 2.66 GHz) with 12 GB of RAM and using Red Hat Enterprise Linux version
5.3. Four processors were allocated to each job along with 6 GB of RAM. Geometries
and input files were generated by the GaussView 5.0.8 program and then submitted to a
simple queue system which allowed for two jobs to run simultaneously. The hybrid DFT
B3LYP method was used in all cases, along with the cc-pVTZ basis set for neutral
species and the aug-cc-pVDZ basis set for reactions involving anionic species. For
calculations involving iodine or tin, cc-pVTZ-PP or aug-cc-pVDZ-PP (downloaded from
the EMSL basis set exchange) was used for the heavy atom (or atoms) along with the
GENECP keyword. The cc-pVTZ or aug-cc-pVDZ basis set was specified for all other
atoms in such calculations. Solvent corrections were made using the polarizable
continuum model (PCM) using DMF as a solvent, except where the effect of such a
solvent correction was being investigated.
Frequencies were calculated at the same level of theory and used to verify that the
minimum found by the optimization algorithm was indeed a true minimum by the
absence of negative frequencies, except for the transition state, which has one negative
frequency. The frequency calculation also includes the thermal correction to the energy
calculation, so free energies were taken from this file and used as a basis of comparison.
143
3.7.11.1 Example computational structures
Figure 3.3. Sn(CH
3
)
3
FCF
2
H + DMF-Cu-I in DMF (transition state, Scheme 3.2,
Structure D) (aug-cc-pVDZ) with hydrogen atoms on methyl groups removed for clarity
Optimized Cartesian coordinates:
C -4.0316160 -1.7788980 -1.2419330
H -3.8411690 -2.6182530 -1.9254810
H -4.7720390 -2.0969640 -0.4942630
H -4.4615820 -0.9474680 -1.8216290
C -0.9120120 -0.8096190 -2.0176260
H 0.1418050 -0.6157140 -1.7868640
H -0.9828370 -1.7227310 -2.6247640
H -1.3214720 0.0301190 -2.5946650
C -2.9721060 0.3217390 1.1659300
H -2.2047580 1.0209160 1.5117130
H -3.7968750 0.8702790 0.6925400
H -3.3696880 -0.2448830 2.0193940
C -0.1700340 -1.2272390 1.5376780
I -0.0109190 3.1051470 -0.0614800
F 0.5676360 -0.2222660 2.3281600
F 0.8180970 -2.2394990 1.3625740
Cu 0.7904310 0.7071710 0.1271990
C 3.2702310 -0.9244510 0.1255190
H 3.0893460 -1.1463410 1.1851870
2 2. .0 03 33 33 35 5 Å Å
2 2. .5 57 79 95 51 1 Å Å
2 2. .6 68 83 32 20 0 Å Å
3 3. .4 49 93 32 25 5 Å Å
144
O 2.4655360 -0.2332930 -0.5391900
H -0.8304930 -1.6529860 2.3126030
C 5.2824340 -2.2523920 0.4616360
H 5.3756040 -3.2645490 0.0449930
H 6.2791110 -1.7920910 0.4989660
H 4.8797240 -2.3188460 1.4771110
C 4.7814650 -1.2576290 -1.7640890
H 5.7660980 -0.7730080 -1.8090510
H 4.8435220 -2.2307400 -2.2701020
H 4.0392800 -0.6306270 -2.2621680
N 4.3924600 -1.4495420 -0.3705250
Sn -2.1622220 -1.1196720 -0.2565330
F -1.9443910 -3.0248090 0.4811890
Imaginary Frequencies: 1
Energies from frequency calculation:
Zero-point correction= 0.231141 (Hartree/Particle)
Thermal correction to Energy= 0.257358
Thermal correction to Enthalpy= 0.258302
Thermal correction to Gibbs Free Energy= 0.167451
Sum of electronic and zero-point Energies= -2857.344236
Sum of electronic and thermal Energies= -2857.318019
Sum of electronic and thermal Enthalpies= -2857.317075
Sum of electronic and thermal Free Energies= -2857.407926
145
Figure 3.4. Iodobenzene + DMF-Cu-CF
2
H square planar complex in DMF (Scheme 3.2,
Structure H)
Optimized Cartesian coordinates:
Cu 0.0724190 0.4578820 -0.2495310
C 1.9470700 0.3752230 0.1550010
C 2.8514780 0.3429590 -0.8987160
C 2.3826700 0.2940590 1.4717330
C 4.2105480 0.1841690 -0.6271290
H 2.5205240 0.4284650 -1.9248620
C 3.7429530 0.1350590 1.7314970
H 1.6844250 0.3393780 2.2962770
C 4.6575240 0.0786120 0.6849480
H 4.9147920 0.1473000 -1.4480790
H 4.0820750 0.0577380 2.7563700
H 5.7125250 -0.0400970 0.8915410
I -0.2468130 -2.1503720 -0.2327660
C 0.4643200 2.4152060 -0.1651830
F 0.4645280 2.8485600 -1.4570060
F -0.6117870 2.9816490 0.4594820
H 1.3631800 2.7668890 0.3285030
C -2.9379780 0.6642220 -0.7120120
H -3.6433730 0.6744440 -1.5451840
1 1. .9 96 68 84 47 7 Å Å
1 1. .9 91 19 95 58 8 Å Å
2 2. .6 62 27 77 77 7 Å Å
1 1. .9 99 97 79 95 5 Å Å
146
N -3.5045430 0.5018340 0.4721010
C -4.9430170 0.2685930 0.5747220
H -5.4036100 1.0504610 1.1775630
H -5.3897570 0.2690040 -0.4154320
H -5.1281820 -0.6964110 1.0461370
C -2.7620400 0.5121870 1.7278930
H -2.6055300 -0.5034590 2.0900930
H -1.8045580 1.0027120 1.5943100
H -3.3366440 1.0686900 2.4662380
O -1.7261240 0.8149470 -0.9654990
Imaginary Frequencies: none
Zero-point correction= 0.216534 (Hartree/Particle)
Thermal correction to Energy= 0.236088
Thermal correction to Enthalpy= 0.237032
Thermal correction to Gibbs Free Energy= 0.162822
Sum of electronic and zero-point Energies= -2655.061901
Sum of electronic and thermal Energies= -2655.042347
Sum of electronic and thermal Enthalpies= -2655.041403
Sum of electronic and thermal Free Energies= -2655.115613
3.8 References
(1) Gribble, G. W. Chem. Soc. Rev. 1999, 28, 335.
(2) Kirk, K. L. Org. Process Res. Dev. 2008, 12, 305.
(3) O'Hagan, D.; Schaffrath, C.; Cobb, S. L.; Hamilton, J. T. G.; Murphy, C. D.
Nature 2002, 416, 279.
(4) Uneyama, K. Organofluorine Chemistry; Blackwell: Oxford, 2006.
(5) Müller, K.; Faeh, C.; Diederich, F. Science 2007, 317, 1881.
(6) Banks, R. E.; Smart, B. E.; Tatlow, J. C. Organofluorine Chemistry: Principles
and Commercial Applications; Plenum: New York, 2000.
(7) Filler, R.; Kobayashi, Y.; Yagupolskii, L. M. Organofluorine Compounds in
Medicinal Chemistry and Biomedical Applications; Elsevier: Amsterdam, 1993.
147
(8) Hiyama, T.; Kanie, K.; Kusumoto, T.; Morizawa, Y.; Shimizu, M.
Organofluorine Compounds: Chemistry and Application; Springer: Berlin, 2000.
(9) Beckers, H.; Bürger, H.; Bursch, P.; Ruppert, I. J. Organomet. Chem. 1986, 316,
41.
(10) Krishnamurti, R.; Bellew, D. R.; Prakash, G. K. S. J. Org. Chem. 1991, 56, 984.
(11) Prakash, G. K. S.; Yudin, A. K. Chem. Rev. (Washington, D. C.) 1997, 97, 757.
(12) Prakash, G. K. S.; Hu, J. ACS Symp. Ser. 2005, 911, 16.
(13) Prakash, G. K. S.; Mandal, M. J. Fluorine Chem. 2001, 112, 123.
(14) Singh, R. P.; Shreeve, J. M. Tetrahedron 2000, 56, 7613.
(15) Knauber, T.; Arikan, F.; Röschenthaler, G.-V.; Gooßen, L. J. Chem. –A Eur. J.
2011, 17, 2689.
(16) Morimoto, H.; Tsubogo, T.; Litvinas, N. D.; Hartwig, J. F. Angew. Chem., Int. Ed.
2011, 50, 3793.
(17) Langlois, B. R.; Laurent, E.; Roidot, N. Tetrahedron Lett. 1991, 32, 7525.
(18) Ji, Y.; Brueckl, T.; Baxter, R. D.; Fujiwara, Y.; Seiple, I. B.; Su, S.; Blackmond,
D. G.; Baran, P. S. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 14411.
(19) Umemoto, T.; Ishihara, S. J. Am. Chem. Soc. 1993, 115, 2156.
(20) Umemoto, T.; Adachi, K.; Ishihara, S. J. Org. Chem. 2007, 72, 6905.
(21) Umemoto, T.; Ishihara, S. Tetrahedron Lett. 1990, 31, 3579.
(22) Umemoto, T. Chem. Rev. 1996, 96, 1757.
(23) Kieltsch, I.; Eisenberger, P.; Togni, A. Angew. Chem., Int. Ed. 2007, 46, 754.
(24) Liu, T.; Shen, Q. Org. Lett. 2011, 13, 2342.
(25) Erickson, J. A.; McLoughlin, J. I. J. Org. Chem. 1995, 60, 1626.
(26) Prakash, G. K. S.; Chacko, S. Curr. Opin. Drug Discovery Dev. 2008, 11, 793.
148
(27) Yudin, A. K.; Prakash, G. K. S.; Deffieux, D.; Bradley, M.; Bau, R.; Olah, G. A.
J. Am. Chem. Soc. 1997, 119, 1572.
(28) Houlton, J. S.; Motherwell, W. B.; Ross, B. C.; Tozer, M. J.; Williams, D. J.;
Slawin, A. M. Z. Tetrahedron 1993, 49, 8087.
(29) Kaneko, S.; Yamazaki, T.; Kitazume, T. J. Org. Chem. 1993, 58, 2302.
(30) Goure, W. F.; Leschinsky, K. L.; Wratten, S. J.; Chupp, J. P. J. Agr. Food Chem.
1991, 39, 981.
(31) Smart, B. E. Chem. Rev. 1996, 96, 1555.
(32) Furuya, T.; Kuttruff, C.; Ritter, T. Curr. Opin. Drug Disc. Dev. 2008, 11, 803.
(33) Singh, R. P.; Shreeve, J. M. Synthesis 2002, 2561.
(34) Dolbier, W. R.; Xie, P.; Zhang, R.; Xu, W.; Chang, Y.; Abboud, K. A. J. Org.
Chem. 2008, 73, 2469.
(35) Prakash, G. K. S.; Hu, J.; Olah, G. A. J. Org. Chem. 2003, 68, 4457.
(36) Prakash, G. K. S.; Hu, J.; Wang, Y.; Olah, G. A. Eur. J. Org. Chem. 2005, 2218.
(37) Prakash, G. K. S.; Hu, J. Acc. Chem. Res. 2007, 40, 921.
(38) Fujiwara, Y.; Dixon, J. A.; Rodriguez, R. A.; Baxter, R. D.; Dixon, D. D.; Collins,
M. R.; Blackmond, D. G.; Baran, P. S. J. Am. Chem. Soc. 2012, 134, 1494.
(39) Fujikawa, K.; Fujioka, Y.; Kobayashi, A.; Amii, H. Org. Lett. 2011, 13, 5560.
(40) Fujikawa, K.; Kobayashi, A.; Amii, H. Synthesis 2012, in press.
(41) Fier, P. S.; Hartwig, J. F. J. Am. Chem. Soc. 2012, 134, 5524.
(42) Zhao, Y.; Huang, W.; Zheng, J.; Hu, J. Org. Lett. 2011, 13, 5342.
(43) Hagiwara, T.; Fuchikami, T. Synlett 1995, 717.
(44) Wang, F.; Luo, T.; Hu, J.; Wang, Y.; Krishnan, H. S.; Jog, P. V.; Ganesh, S. K.;
Prakash, G. K. S.; Olah, G. A. Angew. Chem., Int. Ed. 2011, 50, 7153.
(45) Cullen, W. R.; Sams, J. R.; Waldman, M. C. Inorg. Chem. 1970, 9, 1682.
149
(46) Mee, S. P. H.; Lee, V.; Baldwin, J. E. Angew. Chem., Int. Ed. 2004, 43, 1132.
(47) Fujikawa, K.; Fujioka, Y.; Kobayashi, A.; Amii, H. Org. Lett. 2011, 13, 5560.
(48) Litvinas, N. D.; Fier, P. S.; Hartwig, J. F. Angew. Chem., Int. Ed. 2012, 51, 536.
(49) Dubinina, G. G.; Furutachi, H.; Vicic, D. A. J. Am. Chem. Soc. 2008, 130, 8600.
(50) Dubinina, G. G.; Ogikubo, J.; Vicic, D. A. Organometallics 2008, 27, 6233.
(51) Kieltsch, I.; Dubinina, G. G.; Hamacher, C.; Kaiser, A.; Torres-Nieto, J.;
Hutchison, J. M.; Klein, A.; Budnikova, Y.; Vicic, D. A. Organometallics 2010,
29, 1451.
(52) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.;
Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.;
Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.;
Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven,
T.; J. A. Montgomery, J.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.;
Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.;
Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.;
Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo,
C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;
Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.;
Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.;
Daniels, A. D.; Farkas, Ö.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D.
J.; Gaussian, Inc.: Wallingford CT, 2009.
(53) Eujen, R.; Hoge, B.; Brauer, D. J. J. Organomet. Chem. 1996, 519, 7.
(54) Burton, D. J.; Hartgraves, G. A. J. Fluorine Chem. 2007, 128, 1198.
(55) Gingras, M.; Chabre, Y. M.; Raimundo, J. M. Synthesis 2006, 2006, 182.
(56) Sun, H.; DiMagno, S. G. J. Am. Chem. Soc. 2005, 127, 2050.
(57) Christe, K. O.; Wilson, W. W. J. Fluorine Chem. 1990, 47, 117.
(58) Hu, H.; Snyder, J. P. J. Am. Chem. Soc. 2007, 129, 7210.
(59) Lipshutz, B. H.; Reuter, D. C. Tetrahedron Lett. 1989, 30, 4617.
(60) Petasis, N. A.; Zavialov, I. A. Tetrahedron Lett. 1996, 37, 567.
150
(61) Bull, J. A.; Mousseau, J. J.; Charette, A. B. Org. Lett. 2008, 10, 5485.
(62) Beshai, M.; Dhudshia, B.; Mills, R.; Thadani, A. N. Tetrahedron Lett. 2008, 49,
6794.
151
4 Group Electronegativities: A Simple Computational Method
4.1 Introduction
The concept of electronegativity (EN) is one of the cornerstones of chemistry, and
adds to the essential understanding, rationalization, and modeling for systems in organic,
inorganic and physical chemistry. Key concepts such as acidity, bond polarities, dipole
moments, and inductive effects are conceived of, and taught, in terms of EN. However,
despite its usefulness, electronegativity is not a well-defined concept. In 1932 Pauling put
forward a definition for electronegativity as “the power of an atom in a molecule to
attract electrons to it,” while at the same time suggesting a scale for its measurement.
1
While many other scales have since been constructed for atomic electronegativity,
2-6
including those derived from the effective nuclear charge of an atom,
5,7-8
it is accepted
that EN does not have a quantum mechanical operator, and therefore cannot be directly
measured or calculated.
Another concept that is fundamentally related to EN is group EN (gEN): the
ability of a bonded group of atoms to attract electron density to itself (by extension of
Pauling’s definition). Group ENs are of more use to chemists when studying concepts
such as acidity, chemical reactivity, activating groups, and protecting groups, as there are
many more polyatomic than monatomic substituents. Many attempts have been made to
measure, and define a scale for, gENs.
9-13
A detailed review of “Sigma- and Pi-Electron
Delocalization” was written by Krygowski,
9
where the concept of the substituent effects
is explored from a theoretical viewpoint. The title of the review reveals the progression of
the field, whereby the majority of researchers seek to separate and quantify σ and π
152
effects independently. Although this likely provides a more accurate physical picture of
the substituents being investigated, the simplicity and universal applicability of Pauling’s
EN scale is lost. In this work, we seek to retain the simplicity of a single value for group
electronegativity while increasing accuracy and accessibility through the use of modern,
reliable computational methods. Previous attempts to generate a single value for gEN
have not produces comparable values. Figure 4.1 plots the correlation of gEN values for
overlapping substituent groups from Inamoto et al.,
14
Wells,
15
and Dailey et al.
16
As is
evident from the linear regression between the various scales, a good correlation is not
observed between any two scales, with the highest R
2
value being 0.8081 between
Inamoto et al. and Dailey et al. This comparison highlights the differences between
various approaches and the need for more work in this field to provide an easily
measured and accurate scale that can be used on a wide variety of substituents.
153
Figure 4.1. Linear correlations between various group electronegativity scales
R² = 0.7204
R² = 0.8081
R² = 0.6473
2
2.5
3
3.5
4
2 2.5 3 3.5 4
Electronegativity B
Electronegativity A
Inamoto to Wells
Inamoto to D&S
Wells to D&S
The purpose of the present work is to detail a simple method for calculating group
electronegativities that considers the sum of both and effects. Natural charges,
determined by Natural Bond Orbital (NBO) analysis, on a group, R (R = H, CH
3
,
CH=CH
2
, C
6
H
5
), bound to the substituent of interest, X, in a series of R–X molecules, are
calculated and compared with previously determined values (scaled to Allred-corrected
Pauling electronegativities for a more direct comparison).
17
The rationale for using the
vinyl and phenyl groups as “test” groups is that they should be better suited to detecting π
effects given their double bond structures. Using this method, the gEN of any substituent
154
can be determined from a straight-forward NBO analysis. As a proof of concept, the
group electronegativities for 60 substituents of synthetic interest were determined.
Computed NMR parameters (either chemical shifts or coupling constants) may
yield more accurate electronegativities than natural charges. For example, Dailey et al.
determined their group electronegativities by correlating the difference between the
1
H
chemical shifts, δ(CH
2
) – δ(CH
3
), of numerous CH
3
CH
2
X species and creating a
correlation table (similar to what we have done in this work with natural charges and
Pauling electronegativities) with the electronegativities of the halogens as determined by
Huggins.
18
However, while Gaussian 09
19
can calculate both chemical shifts and coupling
constants, the program does not include all of the parameters, particularly spin-orbit
coupling, necessary to make these computed values reliable for heavier atoms. The effect
of spin-orbit coupling on chemical shifts for the series of CX
3
+
cations (X = F, Cl, Br, I)
has been demonstrated by Kaupp.
20
As such, these values were not used for the present
study, though it is noted that such values, if computed accurately, may provide more
accurate group electronegativities than can natural charges determined by NBO
calculations.
4.2 Experimental Methodology
All molecules computed for this study were optimized (unconstrained, C1
symmetry) using Gaussian 09 (Revision A.02).
19
Density functional theory (DFT,
B3LYP) and Møller-Plesset second-order perturbation (MP2) methods with the aug-cc-
pVTZ basis set, except for the elements Sn, Br, Hg, I, and Xe, which used aug-cc-pVTZ-
155
PP, were used for optimization and frequency calculations. Basis sets were obtained from
the EMSL Basis Set Exchange.
21
The basis sets used here are relatively large, but
probably do not affect the overall gEN calculation significantly, so smaller basis sets
could be used in the interest of time, as long as correlation with the halogen series is first
established. Since the resulting electronegativities were similar using DFT and MP2
methods (Table 4.1), DFT methods were used, as they were computationally less
expensive. All species were deemed to be in their ground-state configurations by the
absence of negative frequencies in the vibrational output. Natural bond orbital analyses
were performed on the densities with the NBO program (version 3.1).
22-24
All natural
charges were converted to Allred-corrected Pauling electronegativities
17
using a least-
squares method that fit the charges on F, Cl, Br and I, similar to the methodology
employed by Dailey and Shoolery.
16
For HX molecules, electronegativities were calculated from the natural charges on
the hydrogen atom. The total charge on the methyl, vinyl and phenyl groups for the series
of CH
3
X, H
2
C=CHX, and C
6
H
5
X molecules were taken as the sum of the natural charges
on all of the atoms of these groups. For the methyl group, the natural charges on the three
hydrogen atoms were not uniform, which is a consequence of the asymmetry of the
electron density distribution, as observed previously for a series of A(XY)
4
molecules.
25
However, the difference in charges was equal to or lesser than 0.02, and, assuming
uniform charge, only influenced the final group electronegativity by no more than 0.05.
The elements F, Cl, Br, and I were chosen as reference points to Pauling EN
because it was hypothesized that the Pauling electronegativities (redetermined by
156
Allred)
17
determined for these elements are the most reliable. The alkali metals were
excluded because it was expected (and confirmed) that the gas-phase determination of the
charge of R–X (X = Li, Na) would not correspond to the values determined by Pauling on
strongly ionic compounds of lithium and sodium. Similarly, attempts to determine the
electronegativity of other elements (N, O, P, S, etc.) using this method failed to give
results consistent with known values, presumably because of complex interactions with
the resulting complex species (NR
3
, OR
2
, etc.). Despite these limitations, it was
determined that the halogen series were sufficient to obtain trend lines that would be
adequate for predicting the electronegativities of various substituents.
To convert charges to Pauling electronegativities the charges on, for example, the
vinyl group of H
2
C=CHX (X = F (0.34586), Cl (0.02096), Br (–0.03836), I (–0.13182))
were plotted against the Pauling electronegativities (as redetermined by Allred)
17
for
those elements (F, 3.98; Cl, 3.16; Br, 2.96; I, 2.66). The resulting graph is given in Figure
4.2.
157
Figure 4.2. Sum of the charges on the vinyl group for H
2
C=CHX (X = F, Cl, Br, I)
graphed against Pauling electronegativities.
y = 2.7181x + 3.0564
R² = 0.9959
2.5
2.7
2.9
3.1
3.3
3.5
3.7
3.9
4.1
-0.2 -0.1 0 0.1 0.2 0.3 0.4
Pauling Electronegativities
Computed charge on vinyl group
The resulting Equations (4.1-4.4) were determined as follows:
Equation 4.1 HX: y = 2.8399x + 2.4145 (X = F, Cl, Br, I; R
2
= 1.0000)
Equation 4.2 CH
3
X: y = 3.0296x + 2.8582 (X = H, F, Cl, Br, I; R
2
= 0.9974)
Equation 4.3 H
2
C=CHX: y = 2.7181x + 3.0564 (X = F, Cl, Br, I; R
2
= 0.9959)
Equation 4.4 C
6
H
5
X: y = 2.6561x + 3.0955 (X = F, Cl, Br, I, R
2
= 0.9960)
The charges obtained (x) were used to determine the electronegativities of X (y;
see Table 4.1 and Table 4.2).
158
Table 4.1: Group ENs (scaled to Pauling values) determined from the natural charge on the R–group of R–X (R = proton,
methyl, phenyl; B3LYP/aug-cc-pVTZ)
group (X) charge on H
a
EN (χ) charge on CH
3
a
EN (χ) charge on C
6
H
5
EN (χ)
BeH
–0.43777
1.17
–0.60819
1.02
–0.61389
1.46
BH
2
–0.10658 2.11 –0.28440 2.00 –0.24328 2.45
B(OH)
2
–0.13448 2.03 –0.28490 (–0.28932) 2.00 (2.14) –0.27087 2.38
BF
2
–0.15674 1.97 –0.29195 1.97 –0.28115 2.35
CH
3
0.20443 (0.19878) 3.00 (3.00) 0.00000 (0.00000) 2.86 (2.94) –0.03113 3.01
Et 0.19267 2.96 0.00528 2.87 –0.02715 3.02
n-Pr 0.19179 2.96 0.00804 2.88 –0.02473 3.03
C(CH
3
)
3
0.17719 2.92 0.01792 2.91 –0.01653 3.05
Bn
b
0.20934 3.01 0.02559 2.94 –0.00490 3.08
CH
2
F 0.15790 2.86 0.00208 2.86 –0.00861 3.07
CHF
2
0.12540 2.77 0.00779 2.88 –0.01557 3.05
CF
3
0.10512 (0.09928) 2.71 (2.71) 0.01060 2.89 –0.00116 3.09
CH=CH
2
0.18443 2.94 0.00184 2.86 –0.02314 3.03
C≡CH 0.22631 3.06 0.02883 2.95 –0.00269 3.09
C
6
H
5
0.20305 2.99 0.03111 2.95 0.00000 3.10
C
6
F
5
0.24391 3.11 0.07375 3.08 0.05063 3.23
C(O)H 0.10064 (0.09585) 2.70 (2.70) –0.01506 (–0.02447) 2.81 (2.87) 0.00424 3.11
C(O)CH
3
0.08850 2.67 –0.02108 2.79 –0.00462 3.08
C(O)NH
2
0.09250 2.68 –0.02020 2.80 –0.01677 3.05
C(O)OH 0.08419 (0.08127) 2.65 (2.66) –0.00323 (–0.00613) 2.85 (2.92) 0.01076 3.12
C(O)OCH
3
0.08269 2.65 –0.00245 2.85 0.00494 3.11
C(O)Cl 0.14397 2.82 0.02714 2.94 0.04270 3.21
Bz
c
0.09916 2.70 –0.01036 2.83 0.00168 3.10
Cbz
d
0.11304 2.74 –0.02837 2.77 0.00996 3.12
CN 0.22012 3.04 0.04402 (0.03818) 2.99 (3.04) 0.02915 3.17
NH
2
0.35097 3.41 0.14626 (0.15029) 3.30 (3.35) 0.04414 3.21
N(CH
3
)
2
0.33779 (0.33560) 3.37 (3.38) 0.16934 3.37 0.04595 3.22
NHAc
e
0.38998 3.52 0.20230 3.21 0.14245 3.47
NCO 0.41222 3.59 0.22434 3.54 0.16806 3.54
NO 0.20742 3.00 0.14877 3.31 0.19740 3.62
NO
2
0.30903 (0.31377) 3.29 (3.32) 0.22486 (0.22244) 3.54 (3.55) 0.24510 3.75
OH 0.46122 (0.46476) 3.72 (3.75) 0.27020 3.68 0.19561 3.62
OCH
3
0.45665 (0.45234) 3.71 (3.71) 0.28324 3.72 0.20171 3.63
OAc
e
0.48762 3.80 0.33106 3.86 0.28000 3.84
ONO
2
0.48440 (0.47570) 3.79 (3.78) 0.34281 (0.33243) 3.90 (3.85) 0.31796 3.94
159
Table 4.1: Continued
group (X) charge on H
a
EN (χ) charge on CH
3
a
EN (χ) charge on C
6
H
5
EN (χ)
OSO
2
CH
3
0.50562 (0.50232) 3.85 (3.86) 0.34740 3.91 0.31015 3.92
OSO
2
C
6
H
5
0.50229 3.84 0.34155 3.89 0.29944 3.89
OSO
2
CF
3
0.51117 (0.50818) 3.87 (3.87) 0.37063 3.98 0.32668 3.96
OSO
2
F 0.51066 3.86 0.36448 3.96 0.33320 3.98
OTeF
5
0.50641 (0.50788) 3.85 (3.87) 0.35899 3.95
k
0.36062 4.05
OF 0.45184 3.70 0.32285 3.84 0.37739 4.10
Al(CH
3
)
2
–0.41228 1.24 –0.57938 1.10 –0.59209 1.52
SiH
3
–0.16078 1.96 –0.37598 1.72 –0.41053 2.01
Si(CH
3
)
3
–0.21486 1.80 –0.41672 1.60 –0.45867 1.88
SiF
3
–0.27835 1.62 –0.43136 1.55 –0.45562 1.89
PH
2
–0.01112 2.38 –0.22596 2.17 –0.27906 2.36
P(CH
3
)
2
–0.05802 2.25 –0.25539 2.08 –0.30014 2.30
P(O)(CH
3
)
2
–0.12122 2.07 –0.26467 2.06 –0.29751 2.31
PF
2
–0.14657 2.00 –0.28970 1.98 –0.03113 2.25
SH 0.13800 (0.13372) 2.81 (2.81) –0.06440 (–0.07519) 2.66 (2.73) –0.17276 2.64
SCH
3
0.11495 (0.11137) 2.74 (2.75) –0.08436 2.60 –0.20176 2.56
S(O)CH
3
0.00999 2.44 –0.11295 2.52 –0.16317 2.66
SO
2
CH
3
–0.03275 (–0.03840) 2.32 (2.32) –0.11143 (–0.12556) 2.52 (2.59) –0.13130 2.75
SO
2
C
6
H
5
–0.02286 2.35 –0.10511 2.54 –0.12725 2.76
SO
2
CF
3
0.00609 (–0.00292) 2.43 (2.42) –0.07750 2.62 –0.08835 2.86
SF
5
0.03215 (0.01231) 2.51 (2.47) –0.01135 2.82 –0.04323 2.98
HgCl –0.21526 1.80 –0.33598 1.84 –0.39014 2.06
HgOAc
e
–0.22256 1.78 –0.33622 1.84 –0.37332 2.10
Sn(CH
3
)
3
–0.23295 1.75 –0.37676 1.72 –0.43068 1.95
XeF –0.02640 2.34 –0.11204 2.52 –0.17202 2.64
a
Values in parentheses determined from MP2/aug-cc-pVTZ electron densities.
b
Bn = CH
2
C
6
H
5
.
c
Bz = C(O)C
6
H
5
.
d
Cbz = C(O)OCH
2
C
6
H
5
.
e
Ac = C(O)CH
3
.
160
4.3 Results and Discussion
Figure 4.3. Graphical representation of group electronegativities via NBO charge
distribution
The electronegativities determined in this work are provided in Table 4.2 (R =
vinyl) and Table 4.1 (R = proton, methyl, and phenyl), and compared with selected
experimental and computational values.
14-16,26-28
A graphical representation of a series of
first row substituents (-BH
2
, -CH
3
, -NH
2
, and –OH) is presented in Figure 4.3. The
validity of the current method hinges on two assumptions: 1) that there is a linear
relationship between the charge on the R–group and the electronegativity of the
substituent, X, and 2) that the substituent-induced changes in the charge on the R–group
are accurately predicted by natural charges. The first assumption was validated through
the least-squared method as described above, using the halogen series as the benchmark
(Figure 4.2). The second assumption was validated by comparison with previous
experimental and computational values, as well as by comparison with expected trends.
161
Table 4.2: Group electronegativities (scaled to Pauling values) determined from the natural charge on the R group of R–X (R
= vinyl; B3LYP/aVTZ).
group (X)
charge
electronegativity (χ)
a
this work ref.
14
ref.
15
ref.
16
ref.
26
ref.
27
ref.
28
BeH –0.58700 1.46 1.90 1.88
BH
2
–0.21435 2.47 2.31 2.25
B(OH)
2
–0.23956 2.41 2.56
BF
2
–0.25378 2.37
CH
3
–0.00186 3.05 2.73 2.51 2.77 2.77 2.79
Et 0.00359 3.07 2.74 2.74 2.77 2.76
n-Pr 0.00731 3.08 2.74 2.77
C(CH
3
)
3
0.01267 3.09 2.75 2.74 2.77
Bn
b
0.02334 3.12 2.76
CH
2
F 0.01205 3.09 2.87 2.74 2.82 3.02
CHF
2
0.01379 3.09 3.01 2.86 3.21
CF
3
0.02092 3.11 3.16 3.45 2.77 2.91 3.37
CH=CH
2
0.00000 3.06 3.00 3.13 2.77 2.80 3.04
C≡CH 0.01931 3.11 3.24 3.40 3.14 2.86
C
6
H
5
0.02311 3.12 2.94 3.13 2.68 2.80
C
6
F
5
0.06829 3.24
C(O)H 0.20623 3.13 3.06 2.59 2.67 2.82 3.19
C(O)CH
3
0.01666 3.10 3.06 2.67 2.81 3.07
C(O)NH
2
0.00178 3.06 2.95 2.67 2.82 3.08
C(O)OH 0.02985 3.14 3.03 3.00 2.55 2.80 2.84 3.27
C(O)OCH
3
0.02814 3.13 3.04 2.84 2.85 3.21
C(O)Cl 0.06491 3.23 2.95 2.86
Bz
c
0.03001 3.14 2.91
Cbz
d
0.00497 3.07
CN 0.04924 3.19 3.35 3.40 2.49 3.24 2.89
NH
2
0.06084 3.22 3.17 3.45 2.99 3.31 3.25 3.28
N(CH
3
)
2
0.06751 3.24 3.19 3.13 3.34 3.25 3.31
NHAc
e
0.13585 3.43 3.21
NCO 0.17921 3.54 3.65 3.61 3.29 3.88
NO 0.19265 3.58 3.66 3.25
NO
2
0.25571 3.75 3.53 3.49 3.54 3.33 3.64
OH 0.20795 3.62 3.60 3.76 3.53 3.64 3.60 3.65
OCH
3
0.21822 3.65 3.64 3.76 3.68 3.58 3.70
OAc
e
0.29013 3.85 3.61
162
Table 4.2. Contuned
group (X)
charge
electronegativity (χ)
a
this work ref.
14
ref.
15
ref.
16
ref.
26
ref.
27
ref.
28
ONO
2
0.29456 3.86 3.95
OSO
2
CH
3
0.29338 3.85 3.58
OSO
2
C
6
H
5
0.29220 3.85 3.58
OSO
2
CF
3
0.32045 3.93
OSO
2
F 0.33015 3.95
OTeF
5
0.34662 4.00 3.87
f
OF 0.32584 3.94 3.64
Al(CH
3
)
2
–0.56499 1.52 1.99 1.96
SiH
3
–0.37839 2.03 2.42 1.76 2.24
Si(CH
3
)
3
–0.42524 1.90 2.26 2.23
SiF
3
–0.42814 1.89 2.52 2.28 2.78
PH
2
–0.23963 2.41 2.43 2.51 2.46 2.54
P(CH
3
)
2
–0.27247 2.32 2.48 2.51 2.47
P(O)(CH
3
)
2
–0.26898 2.33 2.66
PF
2
–0.28901 2.27 2.70 2.78
SH –0.15391 2.64 2.78 2.96 2.42 2.60 2.86 2.84
SCH
3
–0.17018 2.59 2.76 2.96 2.53 2.86 2.80
S(O)CH
3
–0.14349 2.67 2.96 2.87
SO
2
CH
3
–0.11362 2.75 3.09 2.90
SO
2
C
6
H
5
–0.11069 2.76 3.10
SO
2
CF
3
–0.07140 2.86 3.13 3.13
SF
5
–0.02686 2.98 3.15 2.84 3.22
HgCl –0.36853 2.05 1.70
HgOAc
e
–0.35612 2.09 1.74
Sn(CH
3
)
3
–0.40005 1.97 2.26
XeF –0.15187 2.64
a
Literature values scaled to Pauling electronegativities.
b
Bn = CH
2
C
6
H
5
.
c
Bz = C(O) C
6
H
5
.
d
Cbz = C(O)OCH
2
C
6
H
5
.
e
Ac =
C(O)CH
3
.
f
From ref.
29
.
163
Based on our electronegativities derived from the charge on H, CH
3
, CH=CH
2
and
C
6
H
5
in a series of R–X species (Table 4.1 and Table 4.2), it was concluded that, using
natural charges, the charge on the vinyl or phenyl group is an overall better predictor of
gEN than H or CH
3
. Although it may be tempting to use R
2
as the sole measure of
effectiveness of the test group, restraint must be exercised because the R
2
values are all
very high and the test group with the highest R
2
value, H, produces the least plausible
results (Table 4.1). For example, the series -CH
3
, -CH
2
F, -CHF
2
, -CF
3
produces the
opposite trend that one would expect, i.e. that CH
3
is the most electronegative group and
CF
3
is the least electronegative. This observation indicates that using H as a test group
does not sufficiently capture the effects of electronegativity as we have been trained to
understand it. CH
3
as a test group produces similar, albeit less pronounced, results.
Therefore, the highest possible correlation with the halides does not necessarily indicate
that a test group will be the best for determining group electronegativity. The success of
the phenyl and vinyl groups could be due to their π-bound structure, as we had
hypothesized at the beginning of this project. The results using vinyl (Table 4.2) and
phenyl (Table 4.1) are in close agreement; therefore, only vinyl is described in the
ensuing discussion, as these species are computationally less expensive than their phenyl
analogs.
Many of the a priori trends in electronegativity are accurately reproduced in the
present work; for example:
164
Across the second period:
BeH < BH
2
< CH
3
< NH
2
< OH
1.46 2.47 3.05 3.22 3.62
Across the third period:
Al(CH
3
)
2
< Si(CH
3
)
3
< P(CH
3
)
2
< SCH
3
1.52 1.90 2.32 2.59
With increasing carbon substitution on carbon:
CH
3
< Et < n-Pr < C(CH
3
)
3
< Bn
3.05 3.07 3.08 3.09 3.12
With increasing fluorine substitution on carbon:
CH
3
< CH
2
F ≈ CHF
2
< CF
3
3.05 3.09 3.09 3.11
With more electron-withdrawing substituents on N:
NH
2
< N(CH
3
)
2
< NHAc < NCO < NO < NO
2
3.22 3.24 3.43 3.54 3.58 3.75
With more electron-withdrawing substituents on O:
OH < OCH
3
< OAc ≈ OSO
2
CH
3
< ONO
2
< OSO
2
CF
3
< OTeF
5
3.62 3.65 3.85 3.85 3.86 3.93 4.00
With more electron-withdrawing substituents on S:
SCH
3
< S(O)CH
3
< SO
2
CH
3
< SO
2
CF
3
< SF
5
2.59 2.67 2.75 2.86 2.98
165
Apart from these trends, gENs increase as the oxidation state of the central atom
of the substituent increases (e.g., SCH
3
(2.59), S(O)CH
3
(2.67), SO
2
CH
3
(2.75)). As well,
groups bonded through period 2 elements have higher electronegativities than their
period 3 counterparts. As predicted, the influence on the gEN by atoms not directly
bound to R is greatly reduced. However, some discrepancies exist:
1) The gEN determined for CH
3
(3.05) is lower than CF
3
(3.11) as predicted, though the
difference was expected to be greater based on previously determined values. While
this result may be due to the strongly ionic nature of the C–F bond (see point 3
below), it also fits with the general trend that the atom directly bonded to R dominates
the gEN value
2) The groups bonded through period 3 elements differ markedly from Inamoto and
Masuda’s values.
14
However, the previous work is based on the covalent boundary
method of Gordy,
3
and used different scaling factors for the different periods.
3) The group ENs of BF
2
(2.37), SiF
3
(1.89), and PF
2
(2.27) are computed to be lower
than BH
2
(2.47), SiH
3
(2.03), and PH
2
(2.41), respectively, contrary to a priori
expectations of perfluorinated analogs. While it is tempting to attribute this result to
strong π-back donation by fluorine, prior work has demonstrated that fluorine is a
very weak π-donor,
29
and the above results may be attributable to the strong ionic
nature of the E–F bond (E = B, Si, P).
166
4.4 Comparison to Literature Values
No matter how sound the theoretical reasoning for a computational method,
comparison to experimental data is generally the best way to ensure that a computational
model is valid. Unfortunately, there appears to be no universally accepted method for the
determination of group electronegativities in the literature, although methods from
Inamoto,
14
Wells,
15
Dailey and Shoolery,
16
Marriott and coworkers,
26
Boyd and Boyd,
27
and Suresh and Koga
28
have all been proposed. None of these methods agree with each
other based on our research (see Figure 4.1), so it is understandable why there is no
accepted method for determining group electronegativity. Of the three experimental
methods, Inamoto and Masuda tested the highest number of functional groups, so this
report provides the largest data set for comparison. Figure 4.4 shows the correlation
between an overlapping selection of our calculated values (using charges for the vinyl
group) and the corresponding Inamoto and Masuda values (scaled to Pauling
electronegativities). It is evident from this figure that a significant correlation exists (R
2
=
0.87), and the residuals from regression analysis between the two data sets (Figure 4.5)
reveals only methyl and benzyl to be significant outliers. There do not appear to be any
significant groups of outliers, suggesting that the method should be tolerant to diverse
substituent groups. The heavy atom containing group SnMe
3
correlated very well with
Inamoto, indicating that our method is applicable even for heavy atoms.
167
Figure 4.4. Correlation between calculated (this work; vinyl group) and Inamoto and
Masuda’s group electronegativities.
14
1.5
2
2.5
3
3.5
4
4.5
1.5 2 2.5 3 3.5
Calculated Vinyl Electronegativities
Inamoto Electronegativities
y = 1.6438 x - 0.9037
R
2
= 0.8680
Figure 4.5. Absolute residuals from regression analysis
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
F
Cl
Br
I
B(OH)2
Me
CF3
CHO
COOH
COOMe
C6H5
CN
Bn
NH2
NMe2
NO2
OH
OMe
OAc
OSO2Me
OSO2Ph
SiMe3
PMe2
PF2
POMe2
SH
SMe
SOMe
SO2Me
SO2CF3
SO2Ph
SF5
SnMe3
Absolute Residual
R-Group
168
Given the relatively good correlation between our values and Inamoto’s, similar
correlation studies were done with some of the other group electronegativity scales that
have been reported. Correlation with Dailey and Shoolery
16
(Figure 4.6) demonstrated
some level of correlation, but a relatively low R
2
of 0.77 indicates that there are likely
significant differences between the two scales. Dailey and Shoolery’s group
electronegativity scale is based on the difference between the
1
H NMR shifts for the CH
3
and CH
2
peaks for a series of CH
3
CH
2
-X molecules.
Figure 4.6. Correlation with Dailey and Shoolery group electronegativities
16
y = 0.69x + 1.1529
R² = 0.768
2
2.5
3
3.5
4
4.5
2 2.5 3 3.5 4 4.5
Calculated Vinyl Electronegativities
Dailey Electronegativities
Correlation with Wells
15
(Figure 4.7) was no better than with Dailey and
Shoolery, indicating that significant differences exist between our method and these older
experimental methods.
169
Figure 4.7. Correlation with Wells group electronegativities
15
y = 0.8464x + 0.5253
R² = 0.7537
1.5
2
2.5
3
3.5
4
4.5
1.5 2 2.5 3 3.5 4 4.5
Calculated Vinyl Electronegativities
Wells Electronegativities
Boyd and Boyd’s electronegativities
27
are based on the atoms in molecules theory,
and their results correlate relatively well with ours, as plotted in Figure 4.8. Although
their method is also computational in nature, it differs significantly from the method
described here. Their method utilizes the “bond critical point” model which calculates the
position of “critical points” between atoms. By relating the critical point between a group
of interest and hydrogen to the Pauling electronegativities of lithium and fluorine (at 1.0
and 4.0, respectively), they are able to generate electronegativities for any group. Despite
using a proton as their “test” group (which we found to be inaccurate), their results do not
appear to be significantly different from those described here.
170
Figure 4.8. Correlation with Boyd and Boyd’s group electronegativities
27
y = 0.9609x + 0.4344
R² = 0.8332
1
1.5
2
2.5
3
3.5
4
4.5
1 2 3 4 5
Calculated Vinyl Electronegativities
Boyd Electronegativities
Suresh and Koga
28
defined electronegativity in reference to the molecular
electrostatic potential, a quantity which can be calculated for any given group as part of a
molecule. Their approach was quite similar to ours in that they used a substituent group
(methyl) as their test group rather than an atom which is how Boyd and Boyd
27
calculated
their electronegativity values. They chose the methyl group specifically to avoid π-effects
which is in stark contrast to the choice made in this work, where a test group which
would include π-effects was desired. The correlation between our values and Suresh and
Koga’s values is plotted in Figure 4.9. The correlation is relatively good, with the
exception of the R group SiF
3
, which they estimate to have much higher electronegativity
(2.549) than our method predicts (1.89).
171
Figure 4.9. Correlation with Suresh and Koga’s group electronegativities
28
y = 0.996x + 0.0737
R² = 0.8424
1
1.5
2
2.5
3
3.5
4
4.5
1 1.5 2 2.5 3 3.5 4 4.5
Calculated Vinyl Electronegativities
Suresh Electronegativities
4.5 Conclusion
A valid and convenient method for the determination of group electronegativities
by a systematic study of the natural charge on the various R–groups (R = proton, methyl,
vinyl and phenyl) of R–X is described. The vinyl and phenyl groups were found to be
excellent predictors of gENs for a diverse group of 60 synthetically interesting
substituents, giving results that were found to qualitatively agree with previously reported
experimental and computed values. This methodology offers a relatively straightforward
procedure to compute the gEN for virtually any substituent via a simple Gaussian
calculation.
172
The major failing of the method presented here is that it still does not provide
adequate separation between various substituent groups containing the same connecting
atom. In this way, the method overestimates the contribution from the bonding atom. It is
unclear exactly why that is the case given the relatively simple NBO approach that we
have used should provide an accurate description of the partial charges on each atom
including all effects from neighboring atoms. It is possible that the partial charges
calculated by the NBO program are not accurate enough to tease out the relatively subtle
differences that atoms more than one bond away impart. The atoms in molecules
approach from Boyd and Boyd
27
and the molecular electrostatic potential approach used
by Suresh and Koga
28
perhaps provide a more rigorous definition of the electronegativity
term, however they are also more challenging calculations to carry out compared to a
simple calculation of partial charge.
4.6 References
(1) Pauling, L. J. Am. Chem. Soc. 1932, 54, 3570.
(2) Mulliken, R. S. J. Chem. Phys. 1934, 2, 782.
(3) Gordy, W. Phys. Rev. 1946, 69, 604.
(4) Allred, A. L.; Rochow, E. G. J. Inorg. Nucl. Chem. 1958, 5, 264.
(5) Sanderson, R. T. J. Am. Chem. Soc. 1983, 105, 2259.
(6) Allen, L. C. J. Am. Chem. Soc. 1989, 111, 9003.
(7) Lackner, K. S.; Zweig, G. Phys. Rev. D 1983, 28, 1671.
(8) Bratsch, S. G. J. Chem. Educ. 1984, 61, 588.
173
(9) Krygowski, T. M.; Stepień, B. T. Chem. Rev. 2005, 105, 3482.
(10) Ozimiński, W. P.; Dobrowolski, J. C. J. Phys. Org. Chem. 2009, 22, 769.
(11) Greever, J. C. J. Chem. Educ. 1978, 55, 538.
(12) Bratsch, S. G. J. Chem. Educ. 1985, 62, 101.
(13) Huai, L.; Qingwei, W.; Lixin, L. J. Chem. Educ. 1992, 69, 783.
(14) Inamoto, N.; Masuda, S. Chem. Lett. 1982, 11, 1007.
(15) Wells, P. R. Prog. Phys. Org. Chem. 1968, 6, 111.
(16) Dailey, B. P.; Shoolery, J. N. J. Am. Chem. Soc. 1955, 77, 3977.
(17) Allred, A. L. J. Inorg. Nucl. Chem. 1961, 17, 215.
(18) Huggins, M. L. J. Am. Chem. Soc. 1953, 75, 4123.
(19) Frisch, M. J.; Trucks, G. W.; Shchlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.;
Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.;
Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.;
Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven,
T.; Montgomery, J. A. J.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.;
Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.;
Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. s.; tomasi, J.; Cossi, M.;
Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo,
C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;
Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.;
Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.;
Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D.
J.; Gaussian, Inc.: Wallingford, CT, 2009.
(20) Kaupp, M.; Malkina, O. L.; Malkin, V. G. Chem. Phys. Lett. 1997, 265, 55.
(21) Le, M.; Ren, M.; Zhang, Z.; Sprunger, P. T.; Kurtz, R. L.; Flake, J. C. J.
Electrochem. Soc. 2011, 158, E45.
(22) Glendening, E. D.; Reed, A. E.; Carpenter, J. E.; Weinhold, F.; Version 3.1 ed.;
Theoretical Chemistry Institute of Wisconsin: Madison, WI, 1990, p NBO.
(23) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988, 88, 899.
174
(24) Reed, A. E.; Weinstock, R. B.; Weinhold, F. J. Chem. Phys. 1985, 83, 735.
(25) Heard, G. L.; Gillespie, R. J.; Rankin, D. W. H. J. Mol. Struct. 2000, 520, 237.
(26) Marriott, S.; Reynolds, W. F.; Taft, R. W.; Topsom, R. D. J. Org. Chem. 1984,
49, 959.
(27) Boyd, R. J.; Boyd, S. L. J. Am. Chem. Soc. 1992, 114, 1652.
(28) Suresh, C. H.; Koga, N. J. Am. Chem. Soc. 2002, 124, 1790.
(29) Mercier, H. P. A.; Moran, M. D.; Schrobilgen, G. J.; Suontamo, R. J. J. Fluorine
Chem. 2004, 125, 1563.
175
BIBLIOGRAPHY
Abbott, A. P.; Eardley, C. A. J. Phys. Chem. B 2000, 104, 775.
Aeshala, L. M.; Rahman, S. U.; Verma, A. Sep. Purif. Technol. 2012, 94, 131.
Allen, L. C. J. Am. Chem. Soc. 1989, 111, 9003.
Allred, A. L. J. Inorg. Nucl. Chem. 1961, 17, 215.
Allred, A. L.; Rochow, E. G. J. Inorg. Nucl. Chem. 1958, 5, 264.
Andrews, E.; Ren, M.; Wang, F.; Zhang, Z.; Sprunger, P.; Kurtz, R.; Flake, J. J.
Electrochem. Soc. 2013, 160, H841.
Aschenbrenner, O.; Styring, P. Energy Environ. Sci. 2010, 3, 1106.
Aydin, R.; Köleli, F. J. Electroanal. Chem. 2002, 535, 107.
Ayers, K. E.; Anderson, E. B.; Capuano, C.; Carter, B.; Dalton, L.; Hanlon, G.; Manco,
J.; Niedzwiecki, M. ECS Trans. 2010, 33, 3.
Banks, R. E.; Smart, B. E.; Tatlow, J. C. Organofluorine Chemistry: Principles and
Commercial Applications; Plenum: New York, 2000.
Standard Potentials in Aqueous Solution; Bard, A. J.; Parsons, R.; Jordan, J., Eds.;
IUPAC, 1985.
Barton Cole, E.; Lakkaraju, P. S.; Rampulla, D. M.; Morris, A. J.; Abelev, E.; Bocarsly,
A. B. J. Am. Chem. Soc. 2010, 132, 11539.
Barton, E. E.; Rampulla, D. M.; Bocarsly, A. B. J. Am. Chem. Soc. 2008, 130, 6342.
Beckers, H.; Bürger, H.; Bursch, P.; Ruppert, I. J. Organomet. Chem. 1986, 316, 41.
Bernatis, P. R.; Miedaner, A.; Haltiwanger, R. C.; DuBois, D. L. Organometallics 1994,
13, 4835.
Beshai, M.; Dhudshia, B.; Mills, R.; Thadani, A. N. Tetrahedron Lett. 2008, 49, 6794.
176
Bizzari, S. N.; Blagoev, M. CEH Marketing Research Report: Formic Acid, SRI
consulting, 2010.
Bonelli, B.; Civalleri, B.; Fubini, B.; Ugliengo, P.; Areán, C. O.; Garrone, E. J. Phys.
Chem. B 2000, 104, 10978.
Boston, D. J.; Xu, C.; Armstrong, D. W.; MacDonnell, F. M. J. Am. Chem. Soc. 2013,
135, 16252.
Bourrez, M.; Molton, F.; Chardon-Noblat, S.; Deronzier, A. Angew. Chem., Int. Ed. 2011,
50, 9903.
Boyd, R. J.; Boyd, S. L. J. Am. Chem. Soc. 1992, 114, 1652.
Bratsch, S. G. J. Chem. Educ. 1984, 61, 588.
Bratsch, S. G. J. Chem. Educ. 1985, 62, 101.
Breedlove, B. K.; Takayama, D.; Ito, T. Chem. Lett. 2004, 33, 1624.
Bull, J. A.; Mousseau, J. J.; Charette, A. B. Org. Lett. 2008, 10, 5485.
Burton, D. J.; Hartgraves, G. A. J. Fluorine Chem. 2007, 128, 1198.
Canfield, D.; Frese, K. W. J. Electrochem. Soc. 1983, 130, 1772.
Carmo, M.; Fritz, D. L.; Mergel, J.; Stolten, D. Int. J. Hydrogen Energy 2013, 38, 4901.
Centi, G.; Perathoner, S.; Wine, G.; Gangeri, M. Green Chem. 2007, 9, 671.
Chang, T.-Y.; Liang, R.-M.; Wu, P.-W.; Chen, J.-Y.; Hsieh, Y.-C. Mater. Lett. 2009, 63,
1001.
Chen, X.; Zhou, Y.; Liu, Q.; Li, Z.; Liu, J.; Zou, Z. ACS Appl. Mater. Inter. 2012, 4,
3372.
Chen, Y.; Kanan, M. W. J. Am. Chem. Soc. 2012, 134, 1986.
Chen, Y.; Li, C. W.; Kanan, M. W. J. Am. Chem. Soc. 2012, 134, 19969.
Chen, Z.; Chen, C.; Weinberg, D. R.; Kang, P.; Concepcion, J. J.; Harrison, D. P.;
Brookhart, M. S.; Meyer, T. J. Chem. Commun. 2011, 47, 12607.
177
Cheung, K.-C.; Guo, P.; So, M.-H.; Lee, L. Y. S.; Ho, K.-P.; Wong, W.-L.; Lee, K.-H.;
Wong, W.-T.; Zhou, Z.-Y.; Wong, K.-Y. J. Organomet. Chem. 2009, 694, 2842.
Christe, K. O.; Wilson, W. W. J. Fluorine Chem. 1990, 47, 117.
Christensen, P. A.; Higgins, S. J. J. Electroanal. Chem. 1995, 387, 127.
Collomb-Dunand-Sauthier, M.-N.; Deronzier, A.; Ziessel, R. Inorg. Chem. 1994, 33,
2961.
Costentin, C.; Drouet, S.; Passard, G.; Robert, M.; Savéant, J.-M. J. Am. Chem. Soc.
2013, 135, 9023.
Costentin, C.; Drouet, S.; Robert, M.; Savéant, J.-M. Science 2012, 338, 90.
Cullen, W. R.; Sams, J. R.; Waldman, M. C. Inorg. Chem. 1970, 9, 1682.
Dailey, B. P.; Shoolery, J. N. J. Am. Chem. Soc. 1955, 77, 3977.
Delacourt, C.; Newman, J. J. Electrochem. Soc. 2010, 157, B1911.
Delacourt, C.; Ridgway, P. L.; Kerr, J. B.; Newman, J. J. Electrochem. Soc. 2008, 155,
B42.
Delacourt, C.; Ridgway, P. L.; Newman, J. J. Electrochem. Soc. 2010, 157, B1902.
DeWulf, D. W.; Jin, T.; Bard, A. J. J. Electrochem. Soc. 1989, 136, 1686.
Dibenedetto, A.; Stufano, P.; Macyk, W.; Baran, T.; Fragale, C.; Costa, M.; Aresta, M.
ChemSusChem 2012, 5, 373.
DiMeglio, J. L.; Rosenthal, J. J. Am. Chem. Soc. 2013, 135, 8798.
Doherty, M. D.; Grills, D. C.; Muckerman, J. T.; Polyansky, D. E.; Fujita, E. Coord.
Chem. Rev. 2010, 254, 2472.
Dolbier, W. R.; Xie, P.; Zhang, R.; Xu, W.; Chang, Y.; Abboud, K. A. J. Org. Chem.
2008, 73, 2469.
Dubinina, G. G.; Furutachi, H.; Vicic, D. A. J. Am. Chem. Soc. 2008, 130, 8600.
Dubinina, G. G.; Ogikubo, J.; Vicic, D. A. Organometallics 2008, 27, 6233.
178
Durand, W. J.; Peterson, A. A.; Studt, F.; Abild-Pedersen, F.; Nørskov, J. K. Surf. Sci.
2011, 605, 1354.
Ebbesen, S. D.; Knibbe, R.; Mogensen, M. J. Electrochem. Soc. 2012, 159, F482.
Eneau-Innocent, B.; Pasquier, D.; Ropital, F.; Léger, J. M.; Kokoh, K. B. Appl. Catal., B
2010, 98, 65.
Erickson, J. A.; McLoughlin, J. I. J. Org. Chem. 1995, 60, 1626.
Eujen, R.; Hoge, B.; Brauer, D. J. J. Organomet. Chem. 1996, 519, 7.
Farkas, A. P.; Solymosi, F. J. Phys. Chem. C 2009, 113, 19930.
Fier, P. S.; Hartwig, J. F. J. Am. Chem. Soc. 2012, 134, 5524.
Filler, R.; Kobayashi, Y.; Yagupolskii, L. M. Organofluorine Compounds in Medicinal
Chemistry and Biomedical Applications; Elsevier: Amsterdam, 1993.
Fischer, F.; Tropsch, H. Brennst. Chem. 1923, 4, 276.
Fischer, J.; Hofmann, H.; Luft, G.; Wendt, H. AIChE J. 1980, 26, 794.
Fisher, B. J.; Eisenberg, R. J. Am. Chem. Soc. 1980, 102, 7361.
Frese, K. W., Jr.; Leach, S. C.; Summers, D. P. Electrochemical reduction of aqueous
carbon dioxide to methanol US 4,609,441A 1986
Frisch, M. J.; Trucks, G. W.; Shchlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman,
J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.;
Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.;
Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.;
Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.;
Montgomery, J. A. J.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.;
Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.;
Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. s.; tomasi, J.; Cossi, M.;
Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo,
C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;
Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.;
Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.;
Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D.
J.; Gaussian, Inc.: Wallingford, CT, 2009.
Froehlich, J. D.; Kubiak, C. P. Inorg. Chem. 2012, 51, 3932.
179
Fujikawa, K.; Fujioka, Y.; Kobayashi, A.; Amii, H. Org. Lett. 2011, 13, 5560.
Fujikawa, K.; Kobayashi, A.; Amii, H. Synthesis 2012, in press.
Fujiwara, Y.; Dixon, J. A.; Rodriguez, R. A.; Baxter, R. D.; Dixon, D. D.; Collins, M. R.;
Blackmond, D. G.; Baran, P. S. J. Am. Chem. Soc. 2012, 134, 1494.
Furuya, T.; Kuttruff, C.; Ritter, T. Curr. Opin. Drug Disc. Dev. 2008, 11, 803.
Gangeri, M.; Perathoner, S.; Caudo, S.; Centi, G.; Amadou, J.; Bégin, D.; Pham-Huu, C.;
Ledoux, M. J.; Tessonnier, J. P.; Su, D. S.; Schlögl, R. Catal. Today 2009, 143,
57.
Gattrell, M.; Gupta, N.; Co, A. J. Electroanal. Chem. 2006, 594, 1.
Ghadimkhani, G.; de Tacconi, N. R.; Chanmanee, W.; Janaky, C.; Rajeshwar, K. Chem.
Commun. 2013, 49, 1297.
Giner, J. Electrochim. Acta 1963, 8, 857.
Gingras, M.; Chabre, Y. M.; Raimundo, J. M. Synthesis 2006, 2006, 182.
Glendening, E. D.; Reed, A. E.; Carpenter, J. E.; Weinhold, F.; Version 3.1 ed.;
Theoretical Chemistry Institute of Wisconsin: Madison, WI, 1990, p NBO.
Goeppert, A.; Czaun, M.; Jones, J.-P.; Surya Prakash, G. K.; Olah, G. A. Chem. Soc. Rev.
2014.
Goeppert, A.; Zhang, H.; Czaun, M.; May, R. B.; Prakash, G. K. S.; Olah, G. A.;
Narayanan, S. R. ChemSusChem 2014, 7, 1386.
Gondal, M.; Ali, M.; Dastageer, M.; Chang, X. Catal Lett 2013, 143, 108.
Gordy, W. Phys. Rev. 1946, 69, 604.
Goure, W. F.; Leschinsky, K. L.; Wratten, S. J.; Chupp, J. P. J. Agr. Food Chem. 1991,
39, 981.
Greever, J. C. J. Chem. Educ. 1978, 55, 538.
Gribble, G. W. Chem. Soc. Rev. 1999, 28, 335.
180
Grodkowski, J.; Dhanasekaran, T.; Neta, P.; Hambright, P.; Brunschwig, B. S.;
Shinozaki, K.; Fujita, E. J. Phys. Chem. A 2000, 104, 11332.
Gupta, N.; Gattrell, M.; MacDougall, B. J Appl Electrochem 2006, 36, 161.
Hagiwara, T.; Fuchikami, T. Synlett 1995, 717.
Hallinder, J.; Holtappels, P.; Mogensen, M. ECS Trans. 2012, 41, 61.
Hansen, J. B. In Sustainable Fuels from Renewable Energies IASS Potsdam, Germany,
2013.
Hara, K.; Kudo, A.; Sakata, T. J. Electroanal. Chem. 1995, 391, 141.
Hara, K.; Kudo, A.; Sakata, T.; Watanabe, M. J. Electrochem. Soc. 1995, 142, L57.
Hara, K.; Sakata, T. Bull. Chem. Soc. Jpn. 1997, 70, 571.
Hara, K.; Tsuneto, A.; Kudo, A.; Sakata, T. J. Electrochem. Soc. 1994, 141, 2097.
Hartvigsen, J.; Elangovan, S.; Frost, L.; Nickens, A.; Stoots, C. M.; O'Brien, J. E.;
Herring, J. S. ECS Trans. 2008, 12, 625.
Heard, G. L.; Gillespie, R. J.; Rankin, D. W. H. J. Mol. Struct. 2000, 520, 237.
Hirota, K.; Tryk, D. A.; Yamamoto, T.; Hashimoto, K.; Okawa, M.; Fujishima, A. J.
Phys. Chem. B 1998, 102, 9834.
Hiyama, T.; Kanie, K.; Kusumoto, T.; Morizawa, Y.; Shimizu, M. Organofluorine
Compounds: Chemistry and Application; Springer: Berlin, 2000.
Hori, Y. In Mod. Aspect. Electroc.; Vayenas, C., White, R., Gamboa-Aldeco, M., Eds.;
Springer New York: 2008; Vol. 42, p 89.
Hori, Y.; Ito, H.; Okano, K.; Nagasu, K.; Sato, S. Electrochim. Acta 2003, 48, 2651.
Hori, Y.; Koga, O.; Yamazaki, H.; Matsuo, T. Electrochim. Acta 1995, 40, 2617.
Hori, Y.; Konishi, H.; Futamura, T.; Murata, A.; Koga, O.; Sakurai, H.; Oguma, K.
Electrochim. Acta 2005, 50, 5354.
Hori, Y.; Murata, A.; Kikuchi, K.; Suzuki, S. J. Chem. Soc., Chem. Commun. 1987, 728.
Hori, Y.; Murata, A.; Takahashi, R.; Suzuki, S. Chem. Lett. 1987, 16, 1665.
181
Hori, Y.; Takahashi, I.; Koga, O.; Hoshi, N. J. Phys. Chem. B 2001, 106, 15.
Hori, Y.; Takahashi, R.; Yoshinami, Y.; Murata, A. J. Phys. Chem. B 1997, 101, 7075.
Hori, Y.; Wakebe, H.; Tsukamoto, T.; Koga, O. Electrochim. Acta 1994, 39, 1833.
Hori, Y.; Wakebe, H.; Tsukamoto, T.; Koga, O. Surf. Sci. 1995, 335, 258.
Hoshi, N.; Kato, M.; Hori, Y. J. Electroanal. Chem. 1997, 440, 283.
Houlton, J. S.; Motherwell, W. B.; Ross, B. C.; Tozer, M. J.; Williams, D. J.; Slawin, A.
M. Z. Tetrahedron 1993, 49, 8087.
Hu, H.; Snyder, J. P. J. Am. Chem. Soc. 2007, 129, 7210.
Huai, L.; Qingwei, W.; Lixin, L. J. Chem. Educ. 1992, 69, 783.
Huggins, M. L. J. Am. Chem. Soc. 1953, 75, 4123.
Ikeda, S.; Takagi, T.; Ito, K. Bull. Chem. Soc. Jpn. 1987, 60, 2517.
Inamoto, N.; Masuda, S. Chem. Lett. 1982, 11, 1007.
Innocent, B.; Liaigre, D.; Pasquier, D.; Ropital, F.; Léger, J. M.; Kokoh, K. B. J Appl
Electrochem 2009, 39, 227.
Innocent, B.; Pasquier, D.; Ropital, F.; Hahn, F.; Léger, J. M.; Kokoh, K. B. Appl. Catal.,
B 2010, 94, 219.
Isaacs, M.; Armijo, F.; Ramírez, G.; Trollund, E.; Biaggio, S. R.; Costamagna, J.;
Aguirre, M. J. J. Mol. Catal. A: Chem. 2005, 229, 249.
Ishihara, A.; Fujimori, T.; Motohira, N.; Ota, K.-i.; Kamiya, N. Bull. Chem. Soc. Jpn.
2001, 74, 1517.
Ji, Y.; Brueckl, T.; Baxter, R. D.; Fujiwara, Y.; Seiple, I. B.; Su, S.; Blackmond, D. G.;
Baran, P. S. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 14411.
Jia, F.; Yu, X.; Zhang, L. J. Power Sources 2014, 252, 85.
Jia, Z. F.; Wang, F. M.; Bai, Y.; Liu, N. Adv. Mat. Res. 2011, 2033.
Kamrath, M. Z.; Relph, R. A.; Johnson, M. A. J. Am. Chem. Soc. 2010, 132, 15508.
182
Kaneco, S.; Iiba, K.; Katsumata, H.; Suzuki, T.; Ohta, K. Electrochim. Acta 2006, 51,
4880.
Kaneco, S.; Iiba, K.; Katsumata, H.; Suzuki, T.; Ohta, K. J Solid State Electrochem 2007,
11, 490.
Kaneco, S.; Iiba, K.; Katsumata, H.; Suzuki, T.; Ohta, K. Chem. Eng. J. 2007, 128, 47.
Kaneco, S.; Iiba, K.; Ohta, K.; Mizuno, T.; Saji, A. Electrochim. Acta 1998, 44, 573.
Kaneco, S.; Iiba, K.; Ohta, K.; Mizuno, T.; Saji, A. J. Electroanal. Chem. 1998, 441, 215.
Kaneco, S.; Katsumata, H.; Suzuki, T.; Ohta, K. Energy Fuels 2005, 20, 409.
Kaneco, S.; Ueno, Y.; Katsumata, H.; Suzuki, T.; Ohta, K. Chem. Eng. J. 2006, 119, 107.
Kaneco, S.; Yamauchi, H.; Katsumata, H.; Suzuki, T.; Ohta, K. In Stud. Surf. Sci. Catal.;
Sang-Eon Park, J.-S. C., Kyu-Wan, L., Eds.; Elsevier: 2004; Vol. Volume 153, p
277.
Kaneko, S.; Yamazaki, T.; Kitazume, T. J. Org. Chem. 1993, 58, 2302.
Kang, P.; Cheng, C.; Chen, Z.; Schauer, C. K.; Meyer, T. J.; Brookhart, M. J. Am. Chem.
Soc. 2012, 134, 5500.
Kang, P.; Meyer, T. J.; Brookhart, M. Chemical Science 2013, 4, 3497.
Kauffman, D. R.; Alfonso, D.; Matranga, C.; Qian, H.; Jin, R. J. Am. Chem. Soc. 2012,
134, 10237.
Kaupp, M.; Malkina, O. L.; Malkin, V. G. Chem. Phys. Lett. 1997, 265, 55.
Kazarian, S. G.; Briscoe, B. J.; Welton, T. Chem. Commun. 2000, 2047.
Keith, J. A.; Carter, E. A. J. Am. Chem. Soc. 2012, 134, 7580.
Kieltsch, I.; Dubinina, G. G.; Hamacher, C.; Kaiser, A.; Torres-Nieto, J.; Hutchison, J.
M.; Klein, A.; Budnikova, Y.; Vicic, D. A. Organometallics 2010, 29, 1451.
Kieltsch, I.; Eisenberger, P.; Togni, A. Angew. Chem., Int. Ed. 2007, 46, 754.
Kirk, K. L. Org. Process Res. Dev. 2008, 12, 305.
183
Kita, H. J. Electrochem. Soc. 1966, 113, 1095.
Knauber, T.; Arikan, F.; Röschenthaler, G.-V.; Gooßen, L. J. Chem. –A Eur. J. 2011, 17,
2689.
Koga, O.; Matsuo, T.; Yamazaki, H.; Hori, Y. Bull. Chem. Soc. Jpn. 1998, 71, 315.
Köleli, F.; Atilan, T.; Palamut, N.; Gizir, A. M.; Aydin, R.; Hamann, C. H. J Appl
Electrochem 2003, 33, 447.
Köleli, F.; Balun, D. Appl. Catal., A 2004, 274, 237.
Komatsu, S.; Tanaka, M.; Okumura, A.; Kungi, A. Electrochim. Acta 1995, 40, 745.
Kostecki, R.; Augustynski, J. Ber. Bunsen-Ges. Phys. Chem. 1994, 98, 1510.
Kreuter, W.; Hofmann, H. Int. J. Hydrogen Energy 1998, 23, 661.
Krishnamurti, R.; Bellew, D. R.; Prakash, G. K. S. J. Org. Chem. 1991, 56, 984.
Krygowski, T. M.; Stepień, B. T. Chem. Rev. 2005, 105, 3482.
Lackner, K. S.; Zweig, G. Phys. Rev. D 1983, 28, 1671.
Lamy, E.; Nadjo, L.; Saveant, J. M. J. Electroanal. Chem. 1977, 78, 403.
Langlois, B. R.; Laurent, E.; Roidot, N. Tetrahedron Lett. 1991, 32, 7525.
Le, M.; Ren, M.; Zhang, Z.; Sprunger, P. T.; Kurtz, R. L.; Flake, J. C. J. Electrochem.
Soc. 2011, 158, E45.
Leung, K.; Nielsen, I. M. B.; Sai, N.; Medforth, C.; Shelnutt, J. A. J. Phys. Chem. A
2010, 114, 10174.
Li, H.; Oloman, C. J Appl Electrochem 2005, 35, 955.
Li, H.; Oloman, C. J Appl Electrochem 2006, 36, 1105.
Li, H.; Oloman, C. J Appl Electrochem 2007, 37, 1107.
Li, X.; Liu, H.; Luo, D.; Li, J.; Huang, Y.; Li, H.; Fang, Y.; Xu, Y.; Zhu, L. Chem. Eng.
J. 2012, 180, 151.
Lim, C.-H.; Holder, A. M.; Musgrave, C. B. J. Am. Chem. Soc. 2012, 135, 142.
184
Lipshutz, B. H.; Reuter, D. C. Tetrahedron Lett. 1989, 30, 4617.
Litvinas, N. D.; Fier, P. S.; Hartwig, J. F. Angew. Chem., Int. Ed. 2012, 51, 536.
Liu, T.; Shen, Q. Org. Lett. 2011, 13, 2342.
Lu, G.; Wang, H.; Bian, Z.; Liu, X. ScientificWorldJournal 2013, 2013, 8.
Lu, Q.; Rosen, J.; Zhou, Y.; Hutchings, G. S.; Kimmel, Y. C.; Chen, J. G.; Jiao, F. Nat.
Commun. 2014, 5.
Luo, D.; Chen, C.; Zhang, N.; Hong, S.; Wu, H.; Liu, Z. In Z. Phys. Chem. 2009; Vol.
223, p 1465.
Ma , S.; Lan , Y.; Perez, G. M. J.; Moniri, S.; Kenis , P. J. A. ChemSusChem 2014, 866.
Magdesieva, T. V.; Yamamoto, T.; Tryk, D. A.; Fujishima, A. J. Electrochem. Soc. 2002,
149, D89.
Mahmood, M. N.; Masheder, D.; Harty, C. J. J Appl Electrochem 1987, 17, 1159.
Malkhandi, S.; Bauer, P. R.; Bonnefont, A.; Krischer, K. Catal. Today 2013, 202, 144.
Marriott, S.; Reynolds, W. F.; Taft, R. W.; Topsom, R. D. J. Org. Chem. 1984, 49, 959.
Mee, S. P. H.; Lee, V.; Baldwin, J. E. Angew. Chem., Int. Ed. 2004, 43, 1132.
Mercier, H. P. A.; Moran, M. D.; Schrobilgen, G. J.; Suontamo, R. J. J. Fluorine Chem.
2004, 125, 1563.
Meser Ali, M.; Sato, H.; Mizukawa, T.; Tsuge, K.; Haga, M.-a.; Tanaka, K. Chem.
Commun. 1998, 249.
Minh, N.; Mogensen, M. Interface 2013, 22, 55.
Mistry, H.; Reske, R.; Zeng, Z.; Zhao, Z.-J.; Greeley, J.; Strasser, P.; Cuenya, B. R. J.
Am. Chem. Soc. 2014.
Mizukawa, T.; Tsuge, K.; Nakajima, H.; Tanaka, K. Angew. Chem., Int. Ed. 1999, 38,
362.
Mizuno, T.; Naitoh, A.; Ohta, K. J. Electroanal. Chem. 1995, 391, 199.
185
Mizuno, T.; Ohta, K.; Kawamoto, M.; Saji, A. Energy Sources 1997, 19, 249.
Momose, Y.; Sato, K.; Ohno, O. Surf. Interface Anal. 2002, 34, 615.
Morimoto, H.; Tsubogo, T.; Litvinas, N. D.; Hartwig, J. F. Angew. Chem., Int. Ed. 2011,
50, 3793.
Morris, A. J.; McGibbon, R. T.; Bocarsly, A. B. ChemSusChem 2011, 4, 191.
Müller, K.; Faeh, C.; Diederich, F. Science 2007, 317, 1881.
Mulliken, R. S. J. Chem. Phys. 1934, 2, 782.
Narayanan, S. R.; Haines, B.; Soler, J.; Valdez, T. I. J. Electrochem. Soc. 2011, 158,
A167.
Ni, M. J. Power Sources 2012, 202, 209.
Nicholas, J. D. Interface 2013, 22, 49.
O'Hagan, D.; Schaffrath, C.; Cobb, S. L.; Hamilton, J. T. G.; Murphy, C. D. Nature 2002,
416, 279.
OECD/NEA Projected Costs of Generating Electricity; OECD Publishing, 2005.
Ohmori, T.; Nakayama, A.; Mametsuka, H.; Suzuki, E. J. Electroanal. Chem. 2001, 514,
51.
Ohya, S.; Kaneco, S.; Katsumata, H.; Suzuki, T.; Ohta, K. Catal. Today 2009, 148, 329.
Olah, G. A.; Goeppert, A.; Prakash, G. K. S. Beyond Oil and Gas: The Methanol
Economy; John Wiley & Sons, Inc., 2009.
Olah, G. A.; Goeppert, A.; Prakash, G. K. S. J. Org. Chem. 2009, 74, 487.
Olah, G. A.; Prakash, G. K. S. Electrolysis of carbon dioxide in aqueous media to carbon
monoxide and hydrogen for production of methanol US 12/754,952 March 20,
2012
Olah, G. A.; Prakash, G. K. S.; Goeppert, A. J. Am. Chem. Soc. 2011, 133, 12881.
Oloman, C.; Li, H. ChemSusChem 2008, 1, 385.
Ormerod, R. M. Chem. Soc. Rev. 2003, 32, 17.
186
Ougitani, Y.; Aizawa, T.; Sonoyama, N.; Sakata, T. Bull. Chem. Soc. Jpn. 2001, 74,
2119.
Ozimiński, W. P.; Dobrowolski, J. C. J. Phys. Org. Chem. 2009, 22, 769.
Paik, W.; Andersen, T. N.; Eyring, H. Electrochim. Acta 1969, 14, 1217.
Pauling, L. J. Am. Chem. Soc. 1932, 54, 3570.
Pavone, A. Mega Methanol Plants, SRI Consulting, 2003.
Petasis, N. A.; Zavialov, I. A. Tetrahedron Lett. 1996, 37, 567.
Carbon Dioxide Chemistry: Environmental Issues; 1st ed.; Pradier, J. P.; Pradier, C. M.,
Eds.; Woodhead Publishing, 1994.
Prakash, G. K. S.; Chacko, S. Curr. Opin. Drug Discovery Dev. 2008, 11, 793.
Prakash, G. K. S.; Hu, J. ACS Symp. Ser. 2005, 911, 16.
Prakash, G. K. S.; Hu, J. Acc. Chem. Res. 2007, 40, 921.
Prakash, G. K. S.; Hu, J.; Olah, G. A. J. Org. Chem. 2003, 68, 4457.
Prakash, G. K. S.; Hu, J.; Wang, Y.; Olah, G. A. Eur. J. Org. Chem. 2005, 2218.
Prakash, G. K. S.; Mandal, M. J. Fluorine Chem. 2001, 112, 123.
Prakash, G. K. S.; Viva, F. A.; Olah, G. A. J. Power Sources 2013, 223, 68.
Prakash, G. K. S.; Yudin, A. K. Chem. Rev. (Washington, D. C.) 1997, 97, 757.
Qu, J.; Zhang, X.; Wang, Y.; Xie, C. Electrochim. Acta 2005, 50, 3576.
Raebiger, J. W.; Turner, J. W.; Noll, B. C.; Curtis, C. J.; Miedaner, A.; Cox, B.; DuBois,
D. L. Organometallics 2006, 25, 3345.
Rakowski Dubois, M.; Dubois, D. L. Acc. Chem. Res. 2009, 42, 1974.
Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988, 88, 899.
Reed, A. E.; Weinstock, R. B.; Weinhold, F. J. Chem. Phys. 1985, 83, 735.
187
Rosen, B. A.; Salehi-Khojin, A.; Thorson, M. R.; Zhu, W.; Whipple, D. T.; Kenis, P. J.
A.; Masel, R. I. Science 2011, 334, 643.
Sabatier, P.; Senderens, J. B. J. Chem. Soc., Faraday Trans. 1902, 82, 333.
Saeki, T.; Hashimoto, K.; Fujishima, A.; Kimura, N.; Omata, K. J. Phys. Chem. 1995, 99,
8440.
Saeki, T.; Hashimoto, K.; Kimura, N.; Omata, K.; Fujishima, A. J. Electroanal. Chem.
1996, 404, 299.
Salazar-Villalpando, M. D. ECS Trans. 2011, 33, 77.
Sanderson, R. T. J. Am. Chem. Soc. 1983, 105, 2259.
Saravanakumar, D.; Song, J.; Jung, N.; Jirimali, H.; Shin, W. ChemSusChem 2012, 5,
634.
Schneider, J.; Jia, H.; Kobiro, K.; Cabelli, D. E.; Muckerman, J. T.; Fujita, E. Energy
Environ. Sci. 2012, 5, 9502.
Schouten, K. J. P.; Pérez Gallent, E.; Koper, M. T. M. ACS Catal. 2013, 3, 1292.
Schrebler, R.; Cury, P.; Suárez, C.; Muñoz, E.; Gómez, H.; Córdova, R. J. Electroanal.
Chem. 2002, 533, 167.
Schwartz, M.; Cook, R. L.; Kehoe, V. M.; MacDuff, R. C.; Patel, J.; Sammells, A. F. J.
Electrochem. Soc. 1993, 140, 614.
Seshadri, G.; Lin, C.; Bocarsly, A. B. J. Electroanal. Chem. 1994, 372, 145.
Shi, Y.; Luo, Y.; Cai, N.; Qian, J.; Wang, S.; Li, W.; Wang, H. Electrochim. Acta 2013,
88, 644.
Shiratsuchi, R.; Aikoh, Y.; Nogami, G. J. Electrochem. Soc. 1993, 140, 3479.
Shiratsuchi, R.; Nogami, G. J. Electrochem. Soc. 1996, 143, 582.
Shironita, S.; Karasuda, K.; Sato, K.; Umeda, M. J. Power Sources 2013, 240, 404.
Simbeck, D.; Chang, E. Hydrogen Supply: Cost Estimate for Hydrogen Pathways-
Scoping Analysis, National Renewable Energy Laboratory, Golden, CO, 2002.
Singh, R. P.; Shreeve, J. M. Tetrahedron 2000, 56, 7613.
188
Singh, R. P.; Shreeve, J. M. Synthesis 2002, 2561.
Smart, B. E. Chem. Rev. 1996, 96, 1555.
Smith, B. D.; Irish, D. E.; Kedzierzawski, P.; Augustynski, J. J. Electrochem. Soc. 1997,
144, 4288.
Snuffin, L. L.; Whaley, L. W.; Yu, L. J. Electrochem. Soc. 2011, 158, F155.
Sonoyama, N.; Kirii, M.; Sakata, T. Electrochem. Commun. 1999, 1, 213.
Sridhar, K. R.; Foerstner, R. J. Propul. Power 2000, 16, 1105.
Steffey, B. D.; Curtis, C. J.; DuBois, D. L. Organometallics 1995, 14, 4937.
Stevens, G. B.; Reda, T.; Raguse, B. J. Electroanal. Chem. 2002, 526, 125.
Stoots, C. M.; Hartvigsen, J. J.; O’Brien, J. E.; Herring, J. S. J. Fuel Cell Sci. Technol.
2008, 6, 011014.
Stoots, C. M.; O'Brien, J. E.; Hartvigsen, J. J.; American Nuclear Society: 2007, p 72.
Sun, H.; DiMagno, S. G. J. Am. Chem. Soc. 2005, 127, 2050.
Suresh, C. H.; Koga, N. J. Am. Chem. Soc. 2002, 124, 1790.
Takahashi, I.; Koga, O.; Hoshi, N.; Hori, Y. J. Electroanal. Chem. 2002, 533, 135.
Tanaka, K.; Mizukawa, T. Appl. Organomet. Chem. 2000, 14, 863.
Teeter, T. E.; Van Rysselberghe, P. J. Chem. Phys. 1954, 22, 759.
Terunuma, Y.; Saitoh, A.; Momose, Y. J. Electroanal. Chem. 1997, 434, 69.
Todoroki, M.; Hara, K.; Kudo, A.; Sakata, T. J. Electroanal. Chem. 1995, 394, 199.
Tornow, C. E.; Thorson, M. R.; Ma, S.; Gewirth, A. A.; Kenis, P. J. A. J. Am. Chem. Soc.
2012, 134, 19520.
Tsai, C.-W.; Chen, H. M.; Liu, R.-S.; Asakura, K.; Chan, T.-S. J. Phys. Chem. C 2011,
115, 10180.
Umemoto, T. Chem. Rev. 1996, 96, 1757.
189
Umemoto, T.; Adachi, K.; Ishihara, S. J. Org. Chem. 2007, 72, 6905.
Umemoto, T.; Ishihara, S. Tetrahedron Lett. 1990, 31, 3579.
Umemoto, T.; Ishihara, S. J. Am. Chem. Soc. 1993, 115, 2156.
Uneyama, K. Organofluorine Chemistry; Blackwell: Oxford, 2006.
Ursua, A.; Gandia, L. M.; Sanchis, P. Proc. IEEE 2012, 100, 410.
Vandenborre, H.; Leysen, R.; Baetslé, L. H. Int. J. Hydrogen Energy 1980, 5, 165.
Vermeiren, P.; Adriansens, W.; Moreels, J. P.; Leysen, R. Int. J. Hydrogen Energy 1998,
23, 321.
Wang, F.; Luo, T.; Hu, J.; Wang, Y.; Krishnan, H. S.; Jog, P. V.; Ganesh, S. K.; Prakash,
G. K. S.; Olah, G. A. Angew. Chem., Int. Ed. 2011, 50, 7153.
Wells, P. R. Prog. Phys. Org. Chem. 1968, 6, 111.
Wendt, H.; Imarisio, G. J Appl Electrochem 1988, 18, 1.
Whipple, D. T.; Finke, E. C.; Kenis, P. J. A. Electrochem. Solid-State Lett. 2010, 13,
B109.
Wu, J.; Risalvato, F.; Zhou, X.-D. ECS Trans. 2012, 41, 49.
Wu, J.; Risalvato, F. G.; Sharma, P. P.; Pellechia, P. J.; Ke, F.-S.; Zhou, X.-D. J.
Electrochem. Soc. 2013, 160, F953.
Wu, J. C. S.; Lin, H.-M.; Lai, C.-L. Appl. Catal., A 2005, 296, 194.
Xie, K.; Zhang, Y.; Meng, G.; Irvine, J. T. S. J. Mater. Chem. 2011, 21, 195.
Yahaya, A. H.; Gondal, M. A.; Hameed, A. Chem. Phys. Lett. 2004, 400, 206.
Yamamoto, T.; Hirota, K.; Tryk, D. A.; Hashimoto, K.; Fujishima, A.; Okawa, M. Chem.
Lett. 1998, 27, 825.
Yano, H.; Shirai, F.; Nakayama, M.; Ogura, K. J. Electroanal. Chem. 2002, 533, 113.
Yano, H.; Tanaka, T.; Nakayama, M.; Ogura, K. J. Electroanal. Chem. 2004, 565, 287.
190
Yano, J.; Yamasaki, S. J Appl Electrochem 2008, 38, 1721.
Yao, S. A.; Ruther, R. E.; Zhang, L.; Franking, R. A.; Hamers, R. J.; Berry, J. F. J. Am.
Chem. Soc. 2012, 134, 15632.
Yoshitake, H.; Takahashi, K.; Ota, K.-i. J. Chem. Soc., Faraday Trans. 1994, 90, 155.
Yudin, A. K.; Prakash, G. K. S.; Deffieux, D.; Bradley, M.; Bau, R.; Olah, G. A. J. Am.
Chem. Soc. 1997, 119, 1572.
Yue, X.; Irvine, J. T. S. J. Electrochem. Soc. 2012, 159, F442.
Yui, T.; Kan, A.; Saitoh, C.; Koike, K.; Ibusuki, T.; Ishitani, O. ACS Appl. Mater. Inter.
2011, 3, 2594.
Zhai, Y.; Chiachiarelli, L.; Sridhar, N. ECS Trans. 2009, 19, 1.
Zhan, Z.; Kobsiriphat, W.; Wilson, J. R.; Pillai, M.; Kim, I.; Barnett, S. A. Energy Fuels
2009, 23, 3089.
Zhao, G.; Jiang, T.; Han, B.; Li, Z.; Zhang, J.; Liu, Z.; He, J.; Wu, W. J. Supercrit. Fluids
2004, 32, 287.
Zhao, Y.; Huang, W.; Zheng, J.; Hu, J. Org. Lett. 2011, 13, 5342.
Zhu, W.; Michalsky, R.; Metin, Ö.; Lv, H.; Guo, S.; Wright, C. J.; Sun, X.; Peterson, A.
A.; Sun, S. J. Am. Chem. Soc. 2013, 135, 16833.
Abstract (if available)
Abstract
Electrochemical CO₂ reduction is discussed in detail in Chapter 1, from its beginnings on mercury electrodes in aqueous media to systems producing primarily carbon monoxide, higher order products, the use of flow cells, the use of high pressure, developments in molecular catalysis, non‐aqueous solvent systems, solid oxide electrolysis approaches, and direct photochemical reduction to methanol. These widely varying methods are compared with an eye towards selecting processes that would be suitable for scale up in the conversion of CO₂ to methanol. Finally, a brief survey of available electrochemical water splitting technologies is included due to its relevance for producing methanol from CO₂. ❧ The design and testing of a novel, high pressure flow electrolysis cell is covered in Chapter 2. Design of the cell plates themselves is covered in detail, as several iterations were tested before arriving at a workable solution. Schematic depictions of the system used for controlling the reaction and connections of all the supporting equipment are also included. Results obtained from CO₂ reduction to CO and formate at elevated pressure under flow conditions are then presented, starting with screening conditions, followed by an in‐depth study on the effect of the cation (Na⁺ or Li⁺) on CO₂ reduction. A techno‐economic analysis follows, where the approximate cost of syngas and methanol produced using the above methods are calculated by making various assumptions and extrapolations. ❧ A new reagent, tributyl(difluoromethyl)stannane, is synthesized and used for difluoromethylation of aryl and heteroaryl iodides and β-styrenyl halides with a copper catalyst in Chapter 3. A novel synthetic procedure for the preparation of difluoromethyl stannanes from their respective tin hydride precursors using the Rupert‐Prakash reagent is described. The use of these novel compounds for difluoromethylation of aryl and heteroaryl iodides as well as β-styrenyl halides using copper iodide as a transfer reagent is then described. Extensive computational modeling of the reaction pathway is given as well, with a mechanism proposed. ❧ Electronegativity is a particularly important concept in chemistry, but a universal method to calculate group electronegativity has yet to be accepted. In Chapter 4 Group electronegativities are derived using simple NBO calculations on structures optimized at ab initio and density functional theory methods. These group electronegativities are then compared to those published by others using both experimental and computational approaches.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Selective fluoroalkylation methods and synthesis of water-soluble organic molecules for organic redox flow batteries
PDF
Novel fluoroalkylation reactions and microwave-assisted methodologies
PDF
Electrocatalysts for direct liquid-feed fuel cells and advanced electrolytes for lithium-ion batteries
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
Electrochemical pathways for sustainable energy storage and energy conversion
PDF
Integrated capture and conversion of carbon dioxide from air into methanol and other C1 products
PDF
Stereoselective nucleophilic fluoromethylations: from methodology to mechanistic studies
PDF
Design and modification of electrocatalysts for use in fuel cells and CO₂ reduction
PDF
The modification of catalysts and their supports for use in various fuel cells
PDF
Direct C−H arylation for the synthesis of conjugated polymers
Asset Metadata
Creator
Jones, John-Paul
(author)
Core Title
Chemistry surrounding tin: from a new electrocatalyst for CO₂ reduction to syngas to a novel CF₂H transfer reagent and related computational studies
School
College of Letters, Arts and Sciences
Degree
Doctor of Philosophy
Degree Program
Chemistry
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
C-C coupling,density functional calculations,difluoromethylation,electrochemical CO₂ reduction,electrochemistry,flow electrolysis,group electronegativity,high pressure CO₂ reduction,OAI-PMH Harvest,stannanes
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Prakash, G. K. Surya (
committee chair
), Olah, George A. (
committee member
), Shing, Katherine (
committee member
)
Creator Email
johnpauj@usc.edu,jpjones9700@gmail.com
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c3-589308
Unique identifier
UC11300381
Identifier
etd-JonesJohnP-3543.pdf (filename),usctheses-c3-589308 (legacy record id)
Legacy Identifier
etd-JonesJohnP-3543.pdf
Dmrecord
589308
Document Type
Dissertation
Format
application/pdf (imt)
Rights
Jones, John-Paul
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
C-C coupling
density functional calculations
difluoromethylation
electrochemical CO₂ reduction
electrochemistry
flow electrolysis
group electronegativity
high pressure CO₂ reduction
stannanes