of 126
1
Supporting Information for
Pendant Hydrogen
-
bond Donors in Cobalt Catalysts
Independently Enhance CO
2
Reduction
Alon Chapovetsky
‡1
,
Matt
hew
Welborn
‡2
,
John M. Luna
1
,
Ralf Haiges
1
,
T
homas F. Miller
III
*2
, and Smaranda C. Marinescu
*1
equal contribution
1
Department of Chemistry, University of Southern California, Los Angeles,
California
90089,
USA
2
Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena,
California 91125,
USA
* E
-
mail:
smarines@usc.edu
and
tfm@caltech.edu
.
2
Table of
Contents
General
3
Cyclic Voltammetry (CV)
3
Controlled
-
potential electrolysis (CPE)
3
-
4
TOF calculations
4
-
5
Evan’s
Method
5
-
6
Synthesis
6
-
7
Crystallographic data
7
-
1
2
Electrochemical characterization
1
3
-
20
Mechanistic studies of CO
2
reduction catalysis
20
-
3
5
Density functional calculation details
3
5
-
3
7
Computed reduction potential of 1
(II)
-
6
(II)
.
3
7
Density functional
calcluations on the binding of CO
2
3
8
-
3
9
Embedded multireference calculations on the favorability of hydrogen bonding in complex
1
(I)
-
CO
2
3
9
-
40
Structure of the CO
2
H adduct
4
0
-
41
Effect of solubility on the first protonation energy
41
-
42
Mechanism of the
second protonation: intra
-
versus intermolecular proton transfer
4
2
-
43
Comparison of the intra
-
versus intermolecular mechanism for the first protonation
4
3
Calculation of proton free energy
4
3
-
44
Acidity of the intermediates
4
4
Analysis of the
(E)
ECEC
mechanism for the second protonation step
4
5
Kinetics of the EECC mechanism
46
-
49
Geometries used in density functional studies
50
-
1
2
2
References
1
2
3
-
12
6
3
General
All manipulations of air and moisture sensitive materials were conducted under a
nitrogen atmosphere in a Vacuum Atmospheres drybox or on a dual manifold Schlenk line. The
glassware was oven
-
dried prior to use. All solvents were degassed with nitrogen and
passed
through activated alumina columns and stored over 4Å Linde
-
type molecular sieves. Deuterated
solvents were dried over 4Å Linde
-
type molecular sieves prior to use. Proton NMR spectra were
acquired at room temperature using Varian (Mercury 400 2
-
Chann
el, VNMRS
-
500 2
-
Channel,
VNMRS
-
600 3
-
Channel, and 400
-
MR 2
-
Channel) spectrometers and referenced to the residual
1 H resonances of the deuterated solvent (
1
H: CDCl
3
,
δ
7.26; C
6
D
6
,
δ
7.16; CD
2
Cl
2
,
δ
5.32;
CD
3
CN,
δ
2.94) and are reported as parts per millio
n relative to tetramethylsilane. Elemental
analyses were performed using Thermo Scientific™ FLASH 2000 CHNS/O Analyzers. All the
chemical reagents were purchased from commercial vendors and used without further
purification. The ligands
L
1
6
and complexes
1
(II)
and
6
(II)
were prepared according to the
reported literature procedures
1
,
2
.
Cyclic
Voltammetry (CV)
Electrochemistry experiments were carried out using a Pine potentiostat. The experiments
were performed in a single compartment electrochemical cell under nitrogen or CO
2
atmosphere
using a 3 mm diameter glassy carbon electrode as the work
ing electrode, a platinum wire as
auxiliary electrode and a silver wire as the reference electrode.
Ohmic drop was compensated
using the positive feedback compensation implemented in the instrument.
All experiments in this
paper were referenced relative to
ferrocene (Fc) with the Fe
3+/2+
couple at 0.0 V. Alternatively, in
cases when the redox couple of ferrocene overlapped with other redox waves of interested,
decamethylferrocene (Fc*) was as an internal standard with the Fe*
3+/2+
couple at
0.48 V.
All
electrochemical
experiments
were
performed
with
0.1
M
tetrabutylammonium
hexafluorophosphate as supporting electrolyte.
The
concentrations
of the cobalt complexes
1
(II)
6
(II)
were generally at 0.5 mM and experiments with CO
2
were performed at gas satur
ation
or
varying amounts of CO
2
in dimethylformamide (DMF)
.
Controlled
-
potential electrolysis (CPE)
CPE measurements were conducted in a two
-
chambered H cell. The first chamber held
the working and reference electrodes in 50 mL of 0.1 M
tetrabutylammonium
hexafluorophosphate and 0.5 M methanol in DMF. The second chamber held the auxiliary
electrode in 25 mL of 0.1 M tetrabutylammonium hexafluorophosphate in DMF. The two
chambers were separated by a fine porosity glass frit. The reference
electrode was placed in a
separate compartment and connected by a Vycor tip. Glassy carbon plate electrodes (6 cm
×
1 cm
×
0.3 cm; Tokai Carbon USA) were used as the working and auxiliary electrodes. Using a gas
-
tight syringe, 10 mL of gas were withdrawn f
rom the headspace of the H cell and injected into a
gas chromatography instrument (Shimadzu GC
-
2010
-
Plus) equipped with a BID detector and a
Restek ShinCarbon ST Micropacked column.
Faradaic efficiencies were determined by diving
4
the measured CO produced b
y the amount of CO expected based on the charge passed during the
bulk electrolysis experiment. For each species the
controlled
-
potential electrolysis measurements
were performed at least twice, leading to similar behavior. The reported Faradaic efficienci
es and
mmol of CO produced are average values.
TOF calculations
from cyclic voltammetry
3
Equations 1
5
were used to determine TOF from catalytic CVs. The catalytic current
(
i
cat
)
for an EECC process (E = electrochemical, C = chemical step)
is given by eq 1
, and it
corresponds to the plateau current. This equation assumes a one
-
electron diffusion current and
pseudo
-
first
-
order kinetics
(
the reaction is first order in catalyst and the concentrations of the
substrates, Q
(CO
2
)
,
is
large in comparison t
o the co
ncentration of catalyst). In eq 1, F is
Faraday’s constant (F = 96 485 C/mol),
S
is the surface area of the electrode (A = 0.07065 cm
2
for CVs),
!"#
!
is the
catalyst concentration ([cat] = 0.5 mM = 5
×
10
7
mol/cm
3
), D
cat
is the
diffusion constant of
the catalytically
-
active species (~5
×
10
6
cm
2
/s),
and
k
cat
is the rate
constant of the catalytic reaction
.
!"#
=
퐹푆
!"#
!
!"#
2
!"#
(1)
Equation 1 is simplified by standardizing with the current in the absence of
substrate
(CO
2
in this case), as described by eq 2.
In eq 2,
F is
Faraday’s constant (F = 96 485 C/mol),
S
is
the surface area of the electrode (A = 0.07065 cm
2
for CVs),
!"#
!
is the
catalyst concentration
([cat] = 0.5 mM = 5
×
10
7
mol/cm
3
), D
cat
is the diffusion constant of the catalytically
-
active
species (~5
×
10
6
cm
2
/s),
υ
is the scan rate (0.1 V/s),
R is the universal gas constant (R = 8.31 J
K
1
mol
1
),
and
T is temperature (T = 298.15 K)
.
!
=
0
.
446
×
!"#
!
!"#
!"
!"
(2
)
Dividing eq 1 by eq 2 allows for determination of
i
cat
/
i
p
and allows one to further
calculate the catalytic rate constant (
k
cat
)
without having to determine S,
!"#
!
, and
D
cat.
The ratio
of equations 1 and 2 produces equation 3 which can be rearranged to produce equation 4 in
which
k
cat
can be solved directly.
!
!"#
!
!
=
!
!
.
!!"
×
!
!
!"#
!
!"
!
=
2
.
24
×
!
!
!"#
!
!"
!
(3)
!"#
=
!
!"#
!
!
!
!
!
.
!"
!
!
!
!"
(4)
Finally, eq 4 can be simplified into eq 5, from which k
cat
can be calculated directly.
!"#
=
0
.
387
×
!
!"#
!
!
!
(5)
5
In the above calculations, the
i
p
values used correspond to the peak current obtained from
the Co
I/0
reduction. However, given that the Co
I/0
reduction is quasi
-
reversible, its homogeneity
cannot be tested. Therefore, we have also performed the
plateau current analysis
using
i
p
values
obtained from the reversible (homogeneous) Co
II/I
couple and obtained
identical results.
TOF
CPE
calculations from controlled potential electrolysis
3
Equation 6
was
used to determine TOF from CPE data, as previously reported
1
.
Th
is
equation assume
s
that electron transfer to the catalyst is fast, obeying the Nernst law. In eq 6,
i
is
the stable cur
rent transferred during CPE (
i
= charge*F.E./time, C/s), F is Faraday’s constant (F
= 96 485 C/mol), A is the surface area of the working electrode (A = 3 cm
2
for CPE),
k
cat
is the
overall rate constant of the catalytic reaction,
D is the diffusion
coefficient (~5
×
10
6
cm
2
/s),
[cat] is the concentration of the catalyst without substrate ([cat] = 0.5 mM = 5
×
10
7
mol/cm
3
), R
is the universal gas constant (R = 8.31 J K
1
mol
1
), T is temperature (T = 298.15 K),
F/RT =
38.92 V
1
.
When the electrolysi
s potential is on the plateau of the catalytic wave, the following
eq can be used to calculate TOF, as previously reported
3
.
푇푂퐹
=
!"#
=
!
!
!
!
!
!
!
!"#
!
(
6
)
Controlled potential electrolysis (
CPE
)
experiments, on the other hand, are bulk
experiments that are run for prolonged periods of time
and under constant stirring, and therefore,
replenishing of the catalyst in the diffusion layer (at the electrode
-
liquid interface).
In these
experiments
,
the
rate is limited by catalyst,
substrate
, reagents and products
(
catalyst,
CO
2
, TFE,
CO) diffusion to and away from the electrode.
These
experiments
provide information
about the
stability and selectivity of the catalyst, but do not provide any kinetic information
intrinsic to the
catalyst
3
.
Evan’s
Method
4
Evan’s method was used to determine the total spin (S) of a metal complex by
1
H NMR
spectroscopy
.
Equation
7
is used to determined the MMs (Measured Molar Susceptibility).
Δ
Hz
is the d
ifference
in hertz
between the
peaks of the
solvent
in contact with the complex and the
ones in the capillary tube and
, M is the molarity of the sample (in units of mol/L), and
Hz
NMR
is
the spectrometer frequency, in hertz (500,000,000).
푀푀푠
=
!"""
×
!"#
!
!
×
!
×
!"
!"#
(
7
)
For
1
(II)
,
Δ
Hz =
90.5
Hz
For
1
(I)
,
Δ
Hz = 131.5 Hz
Subsequently, eqs
8
and
9
are used to determine the number of unpaired electrons
!
=
푀푀푠
!
,
!
=
!"
!
×
10
!
!
(
8
)