Through-Space Charge Interaction Substituent Effects in Molecular Catalysis Leading to the Design of the Most Efficient Catalyst of CO2-to-CO Electrochemical Conversion

Article abstract of DOI:10.1021/jacs.6b07014

The starting point of this study of through-space substituent effects on the catalysis of the electrochemical CO₂-to-CO conversion by iron(0) tetraphenylporphyrins is the linear free energy correlation between through-structure electronic effec

Full text of DOI:10.1021/jacs.6b07014

Subscriber access provided by University of Idaho Library
Article
A Study of Through-Space Charge Interaction Substituent
Effects in Molecular Catalysis Leading to the Design of the Most
Efficient Catalyst of CO2-to-CO Electrochemical Conversion
Iban Azcarate, Cyrille Costentin, Marc Robert, and Jean-Michel Savéant
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.6b07014 • Publication Date (Web): 11 Nov 2016
Downloaded from http://pubs.acs.org on November 11, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 15
Journal of the American Chemical Society
A Study of Through-Space Charge Interaction Substituent Effects in
Molecular Catalysis Leading to the Design of the Most Efficient Catalyst
of CO2-to-CO Electrochemical Conversion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Iban Azcarate, Cyrille Costentin*, Marc Robert* and Jean-Michel Savéant*
Université Paris Diderot, Sorbonne Paris Cité, Laboratoire d'Electrochimie Moléculaire, Unité Mixte de
Recherche Université - CNRS N° 7591, Bâtiment Lavoisier, 15 rue Jean de Baïf, 75205 Paris Cedex 13, France.
Corresponding author's email addresses: cyrille.costentin@univ-paris-diderot.fr; robert@univ-paris-diderot.fr;
saveant@univ-paris-diderot.fr .
ABSTRACT.
The starting point of this study of through space-substituent effects on the catalysis of the electrochemical
CO2-to-CO conversion by iron(0)-tetraphenyl porphyrins is the linear free energy correlation between through-
structure electronic effects and the iron(I/0) standard potential established in reference Error! Bookmark not
defined.. The introduction of four positively charged trimethylanilinium groups in para-position of the TPP
phenyls results in an important positive deviation from the correlation and a parallel improvement of the
catalytic Tafel plot. The assignment of this catalysis boosting effect to the coulombic interaction of these
positive charges with the negative charge borne by the initial Fe0–CO2 adduct is confirmed by the negative
deviation observed when the four positive charges are replaced by four negative charges borne by sulfonate
groups also installed in the para-position of the TPP phenyls. The climax of this strategy of catalysis boosting
by means of coulombic stabilization of the initial Fe0–CO2 adduct is reached when four positively charged
trimethylanilinium groups are introduced in ortho-position of the TPP phenyls. The addition of a large
concentration of a weak acid − phenol − helps cleaving one of the C-O bonds of CO2. The efficiency of the
resulting catalyst is unprecedented as can be judged by catalytic Tafel plot benchmarking with all presently
available catalysts of the electrochemical CO2-to-CO conversion. The maximal turnover frequency is as high as
106 s-1 and is reached at an overpotential of only 220 mV; the extrapolated turnover frequency at zero
overpotential is larger than 300 s-1. This catalyst leads to a highly selective formation of CO (practically 100%)
in spite of the presence of a high concentration of phenol which could have favor H2 evolution. It is also very
stable showing no significant alteration after more than eighty hours electrolysis.
Introduction
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 15
1
We have analyzed in details elsewhere how substituents exerting a through-structure electronic effect
influence the thermodynamics and kinetics of the catalysis of the electrochemical CO2-to-CO-conversion,
taking as example a series of phenyl-substituted FeI/0tetraphenyl porphyrins. Catalytic Tafel plots (Figure 1)
1
2
3
4
5
6
7
8
9
relating the electrochemical turnover frequency (TOF) to the overpotential,
η
= ± E0 - E ; with
(
)
t,r
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0
t,r
0
E
= E
= -0.74 V vs. SHE (see SI) were used to gauge the substituent effects.
CO2 /CO,DMF
E = Ec0at
l
logTOF
ogTOFmax
Good
Catalysts
Bad
Catalysts
η
kcat
TOF =
TOF =
,
η
= Et0r − E
F



1+ exp
1+ exp
E − E0
(
)



cat
RT
TOFmax
F
F






E0 − E0
×exp −
η
(
)






tr
cat
RT
RT
E : electrode potential
Et0r: standard potential of the target reaction
Ec0at: catalyst standard potential
Cc0at: catalyst concentration
Fig. 1. Catalytic Tafel plot.
They also allow the rational benchmarking of the catalysts, independently of the cell characteristics for the
Error! Bookmark not defined.
2
electrochemical CO2-to-CO-conversion,
but also for other reactions. Two conflicting
trends appear for these through-structure effects: positive shifts of the catalyst standard potential, entailing a
favorable decrease of the overpotential η, are accompanied by unfavorable decreases of TOFmax (and vice
versa). These two factors obey a linear free-energy relationship in which the standard potential of the FeI/0
0
couple (
) may serve as a thermodynamic index measuring the standard free energy of each substituent. This
Ecat
2
ACS Paragon Plus Environment

Page 3 of 15
Journal of the American Chemical Society
is the main finding of Reference 1 that will be used in the following discussion rather than the otherwise
important detailed analysis of the substituent effects on each step of the four-step process it also provided. The
relevant results for FeTPP and the three perfluorophenyl porphyrins (Chart 1) are recalled in Figure 2 under the
1
2
3
4
5
6
7
8
Ec0at
and
RT ln10 / F logk
form of catalytic Tafel plots (Figure 2a) and of the linear correlation between
(
)
cat
9
(Figure 2b).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Chart 1.
a
b
5
4
0.3
logTOF (s−1
)
RT ln10 / F log kcat (s-1)
(
)
FeTPP
0.25
0.2
3
FeF5TPP
FeF10TPP
2
1
0.15
0.1
0
FeF20TPP
Ec0at (V vs. SHE)
η
(V)
-1
0.2
0.4
0.6
0.8
1
-1
-1.1
-1.2
-1.3
-1.4
-1.5
-1.6
Fig. 2. a: Catalytic Tafel plots of the porphyrins of Charts 1 for [PhOH] = 3M. b: correlation between
0
TOFmax = kcat
and
, recalling the through-structure substituents effect. The code of colors is the same as in
Ecat
Chart 1.
In seeking to solubilize in water a catalyst of the same family we were led to synthetize and test the porphyrin
designed by Fe-p-TMA (for Iron tetra-(p-N,N,N-TriMethylAnilinium)porphyrin, Chart 2), which bears four
positively charged trimethylammonio substituents in para positions of the phenyl rings. We have thus obtained a
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 15
catalyst that is not only soluble in water but which has proved very effective to the point of being integrated,
after deposition on the cathode surface, into an operational electrolyzer. 3
1
2
3
4
5
6
7
8
9
The question that then naturally arises is to know whether these charged substituents may or may not exert the
same through-structure inductive effect as the Chart 1 porphyrins, which characteristics are recalled in Figure 2.
The existence of a specific effect of the positive charges borne by the four trimethylanilinium groups was then
Chart 2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
checked by comparison with a porphyrin bearing four negative charges (Fe-p-PSULF in Chart 2), which shows
RT ln10 / F logk
an opposite deviation from the
and
(
)
Ec0at
correlation.
cat
These charged substituents in para-position of the phenyl rings are quite distant from the iron catalytic center.
Thus, the specific coulombic effect, even if significant, is not expected to be very strong. This is the reason that
we prepared and tested another tetra-trimethylanilinium porphyrin, which bears the same four
trimethylammonio groups but in the ortho-positions of the phenyl groups (Fe-o-TMA). Compared to Fe-p-TMA
the coulombic interaction effects are expected to be much stronger because of the greater proximity of the
charges to the reacting center. This is indeed what happens as detailed in the next sections, making this
molecule a considerably better catalyst of the electrochemical CO2-to-CO-conversion than all molecules that
have been tested so far. Particular care was exerted to prepare this molecule, to analyze its cyclic voltammetric
RT ln10 / F logk
responses so as to obtain its catalytic Tafel plot and its representative point in the
vs.
(
)
cat
Ec0at
stability over long-time preparative-scale electrolyzes.
correlation. This catalyst is also remarkably performant in terms of CO/H2 selectivity and in terms of
4
ACS Paragon Plus Environment

Page 5 of 15
Journal of the American Chemical Society
Results and Discussion
1
2
3
4
5
6
7
8
Figure 3 shows the FeI/0 cyclic voltammetric responses obtained with the eight porphyrins of Charts 1 and 2
under 1 atm CO2 in the presence of 3 M phenol. It immediately appears, before any treatment of the raw data,
that the Fe-o-TMA is a far much better catalyst than any other molecules in the series both in terms of current
and potential. We will be back to this point after the raw data have been treated to see how the substituent
effects in the Chart 2 porphyrins fall in line or not with the through-structure electronic effects detected and
rationalized with the Chart 1- substituents. Error! Bookmark not defined.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fig. 3. Cyclic voltammetry of Fe-o-TMA (light blue), FeF20TPP (magenta), FeF10TPP (red), Fe-p-TMA
(purple), FeF5TPP (green), FeTPP (black) and Fe-p-PSULF (orange) in the potential domain of the catalytic CO2
reduction wave in DMF + 0.1 M n-Bu4NPF6 + 0.1 M H2O + 3 M PhOH, at 0.1 V/s under 1 atm. CO2 (catalyst
concentration is 1mM). The current, i, is normalized against the peak current of the reversible one-electron
FeII/FeI reversible wave,
obtained at the same scan rate (0.1 V/s).
0
ip
As seen in Figure 4, the cyclic voltammetry response of the three charged porphyrins of Chart 2 is similar to
the one of the neutral porphyrins previously studied in reference 1. As already observed,4 the electro-
withdrawing inductive effect of the four trimethylanilinium groups in para-position of Fe-p-TMA are
0
Ec0at
Ecat
responsible for quite a significant shift of
toward positive potential (
= -1.263 V vs. SHE, 165 mV
more positive than with FeTPP). This effect is even stronger when the trimethylanilinium groups are placed in
ortho-position. (Fe-o-TMA) = -0.944 V vs. SHE (484 mV more positive than with FeTPP) which is the
Ec0at
most positive potential ever reported for an iron porphyrin CO2-reduction catalyst. On the other hand, the
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 15
0
sulfonate groups of Fe-p-PSULF exert almost no effect on the standard potential of the catalyst (
= -1.428
Ecat
1
2
3
V vs. SHE).
4
5
6
7
8
9
The formation of CO is clearly attested by the observation of a cathodic shift of the FeI/FeII reoxidation wave
on the reverse scan in the cyclic voltammetric catalytic responses of the three porphyrins of Chart 2. As with the
other porphyrins, it matches what can be observed in the cyclic voltammetry of the FeII/FeI couple in the
presence of CO. We will confirm the quasi-exclusive formation of CO during long-term preparative scale
electrolyzes.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
It is noticed (Figures 3 and 4) that the current-potential responses of the three porphyrins of interest show peaks
Error! Bookmark not
instead of the plateaus expected for fast catalytic processes. As analyzed in detail elsewhere,
defined.c,4
this is due to the interference of secondary phenomena such as substrate or co-substrate consumption,
inhibition by product and possibly other phenomena that all increase with the charge passed. One way of
fighting the interference of such phenomena is to raise the scan rate, and thereby, decrease the charge passed, so
as to get back to a S-shaped current potential responses and derive the rate constant from the ensuing plateau
current.
Fe-p-TMA
Fe-p-PSULF
Fe-o-TMA
0
-0.5
-1
-1.5
0
-0.5
-1
-1.5
0
-0,5
-1
-1,5
-2
200
180
160
140
120
100
80
60
40
20
0
220
200
180
160
140
120
100
80
i
i0p
i
i0p
i
i0p
60
40
20
0
E (V vs. SHE)
E (V vs. SHE)
E (V vs. SHE)
-20
2
-20
i
i0p
i
i0p
i
i0p
2
1
1
0
0
-1
-1
-2
-3
-2
E (V vs. SHE)
E (V vs. SHE)
E (V vs. SHE)
-3
0
-0.5
-1
-1.5
0
-0.5
-1
-1.5
0
-0,5
-1
-1,5
-2
6
ACS Paragon Plus Environment

Page 7 of 15
Journal of the American Chemical Society
Fig. 4. Cyclic voltammetry of the substituted iron porphyrin of Chart 2 (conc.: 1 mM) in DMF + 0.1 M n-
Bu4NPF6 + 0.1 M H2O at 0.1 V/s under argon (black) and under 1 atm. CO2 in the presence of 3 M PhOH. The
dotted black line in the top right-hand figure is the CV response observed under 1 atm. Ar in the presence of 3 M
PhOH. The bottom figures are a blow-up of the upper figures showing the formation of CO on the reverse scan
(see text).
1
2
3
4
5
6
7
This is the treatment that has been applied to the raw cyclic voltammetric data as shown in Figure 5. It is
8
-0.9 -1.0 -1.1 -1.2 -1.3 -1.4 -1.5 -1.6 -0.9
-1.1
-1.3
-1.5
-1.7
-1.9
-2.1
9
300
250
200
150
100
50
140
120
100
80
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
i
i0p
i
i0p
Fe-p-PSULF
Fe-p-TMA
60
40
20
0
0
E (V vs. SHE)
E (V vs. SHE)
-50
-20
i
i0p
1400
1200
Fe-o-TMA
1000
800
600
400
200
0
E (V vs. SHE)
-200
-0.7
-0.9
-1.1
-1.3
-1.5
Fig. 5. Elimination of the secondary phenomena by raising the scan rate (see text). Cyclic voltammetry of the
substituted iron porphyrins of Chart 2 (conc.: 1 mM) in DMF + 0.1 M NBu4PF6 + 0.1 M H2O under 1 atm. CO2
in the presence of 3 M PhOH at scan rates (V/s), from bottom to top: Fe-p-TMA : 0.1, 1, 5, 10; Fe-p-PSULF :
0.1, 1, 2; Fe-o-TMA : 0.1, 0.5, 1, 2, 6, 10, 20, 30, 48, 96, 115. The current, i, is normalized against the peak
i 0
II
I
current of the reversible one-electron Fe /Fe reversible wave,
obtained at 0.1 V/s.
p
worth noting, en passant, that the scan rates required to reach an S-shaped CV responses are, as expected, the
larger the stronger catalysis (in increasing order: Fe-p-PSULF, Fe-p-TMA, Fe-o-TMA).
RTln10/ F logk
Ec0at
for these
The values of kcat required to establish the catalytic Tafel plots and the
vs.
(
)
cat
three porphyrins may then be derived from the plateau current values, using the following equations. For the
i = FSC0 Dcat 2kcat
catalytic plateau current:
, and for the one-electron diffusion current at 0.1 V/s:
pl
cat
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 15
Fv = 0.1V/s
ipl
i0p
(
)
. Using the ratio
2kcat
RT
i0p = FS ×0.446×Cc0at Dcat
= 2.24
avoids determining S
1
2
3
RT
0.1 F
4
5
6
0
Dcat
(concentration and diffusion coefficient of the catalyst, respectively).
(electrode surface area),
and
Ccat
7
8
9


2

ipl
0.1
F
Thus: kcat
=
0
2.242
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60


2RT
ip


The values of kcat thus obtained were then used to locate the catalytic Tafel plots of the three Chart 2 porphyrins
(
)
Ec0at
correlation
RTln10/ F logk
(Figure 6a) and to introduce the corresponding data points in the
diagram (Figure 6b).
vs.
cat
A through-space effect clearly appears for Fe-p-TMA in so far its representative point stands above the
through-structure correlation line (purple dot in Figure 6b). That this effect results from the positive charges
borne by the substituents is confirmed by the observation that the introduction of negative charges in similar
positions produces the reverse effect (Figure 6b). Although clearly present, these effects are necessarily small
since the charges are rather distant from the reacting center, viz., the initial adduct between CO2 and the iron(0)
complex.
a
b
7
6
0.4
RT ln10 / F log kcat (s-1
)
logTOF (s−1
)
(
)
0.35
0.3
5
4
0.25
0.2
3
2
0.15
0.1
1
0
0.05
0
cat
η
(V)
E
(V vs. SHE)
-1
0
0
0.2
0.4
0.6
0.8
1
-0.8 -0.9 -1 -1.1 -1.2 -1.3 -1.4 -1.5 -1.6
TOFmax = kcat
Fig. 6. a: Catalytic Tafel plots of the porphyrins of Charts 1 and 2. b: correlation between
and
0
, recalling the through-structure substituents effect and showing the coulombic interaction effects of
Ecat
positively and negatively charged substituents, including the huge gain in reactivity and overpotential brought
about by positively charged substituents when located in ortho position of the phenyl ring. The code of colors is :
Fe-o-TMA (light blue), FeF20TPP (magenta), FeF10TPP (red), Fe-p-TMA (purple), FeF5TPP (green), FeTPP
(black) and Fe-p-PSULF (orange).
8
ACS Paragon Plus Environment

Page 9 of 15
Journal of the American Chemical Society
These observations encouraged us to introduce positive charged substituents closer to the reacting center in
spite of the expected synthetic difficulties related to steric congestion of this type of substituted iron-porphyrin.
The result − Fe-o-TMA − proved to be the best catalyst of the whole iron porphyrin series with considerable
gains in terms of overpotential and of turnover frequency as can be seen in Figure 6. The maximal turnover
frequency is as high as 106 s-1 and the extrapolated turnover frequency at zero overpotential larger than 300 s-1.
The reason for this leap forward is most likely the stabilization of the initial iron(0)–CO2 adduct by the
interaction of the negative charge borne by the oxygens of CO2 in this adduct with the nearby positive charges
borne by the trimethylanilinium substituents on the porphyrin phenyls. The presence of phenol in large
concentration then helps the proton-assisted reductive cleavage of one of the carbon-oxygen bond of CO2 (see
Scheme 1 in reference Error! Bookmark not defined.). This is a particularly striking example of the power of
close-distance through–space interactions in boosting of catalysis. Another type of close-distance through–space
interactions as already been implemented to boost CO2 reduction catalysis, albeit with a lesser efficiency,
namely the introduction of ortho phenol groups in the
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Chart 3
porphyrin phenyls (see Chart 3),5 in which stabilization of the initial Fe0 – CO2 adduct is achieved by means of
H-bonding rather than by means of coulombic interactions as in the present case. Comparison between the
coulombic-interaction and H-bond-interaction series is shown in Figure 7, as well as with the best examples of
molecular catalysts that can be found in the literature. It appears that Fe-o-TMA is to a large extent the most
efficient of the whole series in terms of catalytic Tafel plots.
The next step was to look after the selectivity and durability of the catalyst at the preparative scale. Figure 8
shows the Faradaic yields in CO and H2 for Fe-o-TMA and Fe-p-TMA over 7 hours electrolysis time. In both
cases the catalysts appear stable and the CO-vs.-H2 selectivity is excellent. Actually, a slight improvement in
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 15
the selectivity is observed for Fe-o-TMA which presents a CO selectivity of 100% compared to a CO selectivity
of 93% for Fe-p-TMA. The durability of Fe-o-TMA was evaluated by performing a series of 4 consecutive
electrolyzes with the same catalytic solution at an overpotential of 220 mV (at which the TOF is 106 s-1), over a
total of 84 h. The first and fourth electrolysis were done in a glassy carbon crucible, while the
1
2
3
4
5
6
7
8
9
8
a, b
logTOF (s−1
)
Fe-o-TMA
Fe-p-TMA
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
6
4
a, b
c, 6
FCAT
c, 6
2
CAT
b, 1
0
FeTPP
7
N2=Mn(CO)3
-2
-4
-6
-8
-10
-12
8
Re(bpy)(CO)3(py)
9
m-(triphos)2Pd2
10
Mn(bpytBu)(CO)3
11
RuII(tpy-)(bpy-)
11
RuII(tpy-)(Mebim-py-)
η
(V)
0
0.2
0.4
0.6
0.8
1
1.2
a: this work, b: in the presence of 3M PhOH, c :see Chart 3.
py=pyridine, tpy = 2,2':6',2"-terpyridine, bpy =2,2'-bipyridine,
Mebimpy = 2,6-bis(1-methyl benzimidazol-2-yl)pyridine
Fig. 7. Benchmarking of the main molecular catalysts of the CO2-to-CO electrochemical conversion in DMF or
acetonitrile 5,6,7,8,9,10 by means of their catalytic Tafel plots. The vertical arrows indicate the overpotential values
at which the electrolysis were carried out (see text and Figure 8).
Fe-p-TMA
Faradaic yield (%)
Fe-o-TMA
120
100
80
60
40
20
0
120
100
80
Faradaic yield (%)
92%
7%
93%
0%
60
40
20
0
Time (hours)
Time (hours)
-20
-20
0
1
2
3
4
5
6
7
8
0
1
2
3
4
5
6
7
8
9
Fig. 8. a: Preparative-scale electrochemical CO2-to-CO conversion catalyzed by Fe-p-TMA (at an overpotential
of 450 mV, purple vertical arrow in Figure 6) and Fe-o-TMA (at an overpotential of 220 mV, light blue vertical
arrow in Figure 6) in DMF + 0.1 M NBu4PF6 + 0.1 M H2O under 1 atm. CO2 in the presence of 3M PhOH
(catalyst conc.: 0.5 mM). CO (red dots) and H2 (blue dots) Faradaic yields.
second and third one were performed at a Hg pool. From the almost linear evolution of the charge vs. time
(Figure 9a), it was concluded that CO2 was continuously reduced with no noticeable sign of catalyst
10
ACS Paragon Plus Environment

Page 11 of 15
Journal of the American Chemical Society
degradation. Note that some PhOH was added to the catalytic solution after each electrolysis to keep a sufficient
concentration and maintain the activity. The selectivity toward CO remained equal to 100% during the whole
experiment, regardless of the nature of the working electrode. This good selectivity falls in line with the CV
responses shown in top-right-hand Figure 4, where it is seen that the current observed under Ar in the presence
of 3M phenol is considerably less that the catalytic current observed in the presence of CO2.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
a
b
180
160
140
120
100
80
140
120
100
80
Charge (C)
Faradaic yield (%)
CO
60
40
60
20
40
H2
0
20
Time (hours)
Time (hours)
0
-20
0
12 24 36 48 60 72 84
0
12 24 36 48 60 72
Fig. 9. Long run (84 hours) preparative-scale electrochemical CO2-to-CO conversion catalyzed by Fe-o-TMA (at
an overpotential of 220 mV, light blue vertical arrow in Figure 6) in DMF + 0.1 M nBu4NPF6 + 0.1 M H2O
under 1 atm. CO2 in the presence of 3 M PhOH (catalyst conc.: 0.5 mM). a. Charge evolution and b. Faradaic
yields (CO: full circles, H2: open circles) during 24 h on a carbon electrode (blue), then 18 h on a mercury
electrode (green), then during a second 18 h period again on a mercury electrode (red) and finally during another
24 h period on a carbon electrode (yellow).
A control electrolysis performed with a fresh solution containing no catalyst, with a used glassy carbon crucible
demonstrate that no iron catalyst was deposited on the electrode (see SI). The stability of Fe-o-TMA has also
been monitored by means of UV-vis spectroscopy and cyclic voltammetry (see SI, Figure S1). While some
shifts in the Soret band and in the FeIII/II/I CV waves resulting from products’ formation are observed, the
catalytic FeI/0 catalytic CV wave remains remarkably stable.
Conclusions
Through space-substituent effects on the catalysis of the electrochemical CO2-to-CO conversion by iron(0)-
tetraphenyl porphyrins has been first investigated and evidenced by the introduction of four positively charged
trimethylanilinium groups in para-position of the TPP phenyls (Fe-p-TMA). The assignment of this catalysis
boosting effect to the coulombic interaction of these positive charges with the negative charges borne by the
11
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 15
initial Fe0–CO2 adduct has been further confirmed by the negative catalytic effect observed when the four
1
2
3
positive charges in Fe- -TMA are replaced by four negative charges provided by four sulfonate groups installed
p
4
5
in the para-position of the TPP phenyls (Fe-
p-PSULF). Optimization of the catalysis by means of coulombic
6
7
stabilization of the initial Fe0–CO2 adduct was reached when four positively charged trimethylanilinium groups
8
9
are introduced in the ortho position of the TPP phenyls (Fe-o-TMA). The exceptional efficiency of the resulting
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
catalyst is unprecedented with maximal turnover frequency as high as 106 s-1 and is reached at a low
overpotential of 220 mV. The selectivity for CO production is close to 100% while the catalyst appears
extremely stable upon long term electrolysis, with no significant alteration for more than three and a half days.
Experimental
An original synthetic pathway has been developed to synthesize Fe-
porphyrins in Chart 2 is described in the Supporting Information (SI), as well as information concerning the
other chemicals and the instrumentation and procedure for cyclic voltammetry. Fe- -TMA was prepared
according to the flow diagram in Chart 4 from the reaction of 2-nitrobenzaldehyde and pyrrole in 5 steps with
o-TMA. The synthesis of the other Fe
o
9% overall yield. First, tetra-
o
-nitrophenylporphyrine (
1
) was prepared by condensation of 2-nitrobenzaldehyde
with pyrrole in refluxing acetic acid. 11 Porphyrin
1
was then reduced by SnCl2 in concentrated HCl to give the
tetra-o-aminophenylporphyrine (2) as a statistic mixture of four atropisomers (see inset in Chart 4 for a
schematic representation of the atropisoemers). Column chromatography led to the isolation of pure α,β,α,β
isomer. The three remaining isomers can be converted to a new statistic mixture of the four atropisomers after
refluxing in toluene for additional 6 hours. An almost complete conversion of
can be achieved by repeating this chromatography/heating cycle. Treatment of
reductive amination with formaldehyde and sodium cyanoborohydride yield the metallated dimethylamino
intermediate . The mild conditions used in these two successive steps allowed the isolation of in a pure
α,β,α,β configuration.
Finally, porphyrin
desired Fe- -TMA as a triflate salt. Several milder quaternization methods were tried with no decisive success.
2
into the desired α,β,α,β isomer
2
with FeBr2, followed by a
4
4
4
was quaternized by reaction with a large excess of methyl triflate in hot DMF to yield the
o
It may therefore be that some isomerization occurs leading to the formation small amounts of the
12
ACS Paragon Plus Environment

Page 13 of 15
Journal of the American Chemical Society
α,α,β,β
,
α,α,α,β and α,α,α,α isomers. Their separation was not attempted since they present similar
1
2
3
4
5
6
7
8
capabilities to stabilize the initial Fe0-CO2 adduct through coulombic interaction as the α,β,α,β isomer. The
catalyst was fully characterized by IR, UV-vis absorbance, elemental analysis and mass spectrometry (see SI for
details).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Chart 4
13
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 15
0
Supporting Information. Experimental details. Determination of
according to the solvent and to the
ECO /CO
2
1
2
3
acids present. This material is available free of charge via the Internet at http://pubs.acs.org.
4
5
6
Acknowledgments. Partial financial support from the SATT IDF Innov (project 098) is gratefully
7
8
acknowledged.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
References and Notes
1. Azcarate, I. ; Costentin, C. ; Robert ; Savéant, J. M., submitted.
2. Artero, V.; Savéant, J.-M. Energy Environ. Sci. 2014, 7, 3808.
3. (a) Costentin, C.; Robert, M.; Savéant, J.-M.; Tatin, A. Proc. Natl. Acad. Sci. U.S.A. 2015
Tatin, A.; Comminges, C.; Kokoh, B.; Costentin, C.; Robert, M.; Savéant, J.-M. Proc. Natl. Acad. Sci. U.S.A.
2016 113, 5526.
, 112, 6882. (b)
,
4. (a) Costentin, C.; Drouet, S.; Robert, M.; Savéant, J.-M. J. Am. Chem. Soc. 2012
19949. (b) Costentin, C.; Drouet, S.; Passard, G.; Robert, M.; Savéant, J.-M. J. Am. Chem. Soc. 2013
9023. (c) Costentin, C.; Passard, G.; Savéant, J.-M. J. Am. Chem. Soc. 2015 137, 5461.
,
134, 11235; 2012
,
134
135
,
,
,
,
5. Costentin, C.; Passard, G.; Robert, M.; Savéant, J.-M. Proc. Natl. Acad. Sci. U.S.A. 2014, 111, 14990.
6. Sampson, M. D.; Nguyen, A. D.; Grice, K. A.; Moore, C. E.; Rheingold, A. L.; Kubiak, C. P. J. Am. Chem.
Soc. 2014 136, 5460.
,
7. Wong, K.-Y.; Chung, W.-H.; Lau, C.-P. J. Electroanal. Chem. 1998, 453, 161.
8. Raebiger, J. W.; Turner, J. W.; Noll, B. C.; Curtis, C. J.; Miedaner, A.; Cox, B.; DuBois, D. L.
Organometallics 2006 25, 3345.
,
14
ACS Paragon Plus Environment

Page 15 of 15
Journal of the American Chemical Society
1
2
3
4
9. Smieja, J. M.; Sampson, M. D.; Grice, K. A.; Benson, E. E.; Froehlich, J. D.; Kubiak, C. P. Inorg. Chem.
2013 52, 2484.
,
5
6
7
8
9
10. Chen, Z.; Chen, C.; Weinberg, D. R.; Kang, P.; Concepcion, J. J.; Harrison, D. P.; Brookhart, M. S.;
Meyer, T. J. Chem.Comm. 2011 47, 12607.
,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
11. Collman J. P.; Gagne, R. R.; Reed, C.; Halbert, T. R.; Lang, G.; Robinson, W. T.J. Am. Chem. Soc. 1975,
97, 1427.
TOC
15
ACS Paragon Plus Environment