Oxygen Reduction Catalysis at a Dicobalt Center: The Relationship of Faradaic Efficiency to Overpotential

Article abstract of DOI:10.1021/jacs.5b12828

The selective four electron, four proton, electrochemical reduction of O₂ to H₂O in the presence of a strong acid (TFA) is catalyzed at a dicobalt center. The faradaic efficiency of the oxygen reduction reaction (ORR) is furnished from a systematic electrochemical study by using rotating ring disk electrode (RRDE) methods over a wide potential range. We derive a thermodynamic cycle that gives access to the standard potential of O₂ reduction to H₂O in organic solvents, taking into account the presence of an exogenous proton donor. The difference in ORR selectivity for H₂O vs H₂O₂ depends on the thermodynamic standard potential as dictated by the pK_a of the proton donor. The model is general and rationalizes the faradaic efficiencies reported for many ORR catalytic systems.

Full text of DOI:10.1021/jacs.5b12828

Subscriber access provided by The Chinese University of Hong Kong
Communication
Oxygen Reduction Catalysis at a Dicobalt Center:
The Relation of Faradaic Efficiency to Overpotential
Guillaume Passard, Andrew M. Ullman, Casey N Brodsky, and Daniel G. Nocera
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.5b12828 • Publication Date (Web): 13 Feb 2016
Downloaded from http://pubs.acs.org on February 14, 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 5
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Oxygen Reduction Catalysis at a Dicobalt Center: The Relation of
Faradaic Efficiency to Overpotential
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
Guillaume Passard, Andrew M. Ullman, Casey N. Brodsky, Daniel G. Nocera*
Department of Chemistry and Chemical Biology, Harvard University, 12 Oxford Street, Cambridge, MA 02138, United States
Supporting Information Placeholder
ABSTRACT: The selective four electron, four proton, electrochem-
ical reduction of O2 to H2O in the presence of a strong acid (TFA)
is catalyzed at a dicobalt center. The faradaic efficiency of the oxy-
gen reduction reaction (ORR) is furnished from a systematic elec-
trochemical study by using rotating ring disk electrode (RRDE)
methods over a wide potential range. We derive a thermodynamic
Scheme 1. Dicobalt complexes composed a Co(III)2(OH)2 diamond
cycle that gives access to the standard potential of O2 reduction to
core stabilized by a DPEN ligand with anionic acetamidate (1) and
H2O in organic solvents, taking into account the presence of an
exogenous proton donor. The difference in ORR selectivity for
acetate (2) bridging ligands.
H2O vs. H2O2 depends on the thermodynamic standard potential
as dictated by the pKa of the proton donor. The model is general
and rationalizes the faradaic efficiencies reported for many ORR
catalytic systems.
This dicobalt motif is useful for ORR because it is soluble in water
and nonaqueous solutions and affords access to a wide range of
overpotentials and faradaic efficiencies for ORR. We now develop a
general framework to shed light on (a) how solvent and acid
strength affect the overpotential of ORR catalysis and (b) the cor-
relation between ORR overpotential and faradaic efficiency.
Complex 1 was prepared by the addition of the DPEN ligand to
Renewable energy resources have the potential to impact climate
change by mitigating carbon emissions attendant to a fossil based
Co(NO3)2 in a 1:1 water:acetone mixture. Air oxidation to furnish
the two Co(III) centers is slow, thus H2O2 was used to drive the
oxidation. The bridging acetamidate ligand was furnished by heat-
ing MeCN solutions of the PF6 salt of the dicobalt complex. Com-
plex 2 was prepared in a similar fashion, except using a Co(OAc)2
precursor salt to furnish the bridging acetate ligand. The com-
pounds were characterized by NMR, mass spectrometry, and ele-
mental analysis (see SI for details).
Figure 1 shows CVs of 1 and 2 in MeCN in the absence of O2
(red trace). The reversible CV wave corresponds to the
Co2(III,III)/Co2(III,II) couple; redox processes of the DPEN lig-
and occur at a very negative potential (Figure S1). Consistent with
the poorer electron-donating property of the acetate bridge, the
Co2(III,III)/Co2(III,II) couple for 2 is shifted anodically by 250
mV as compared to that of 1. The presence of O2 (8.1 mM) causes
a loss of reversibility for the Co2(III,III)/Co2(III,II) wave (blue
trace, Figure 1) and concomitant slight increase in current. This
loss of reversibility is a clear indication that O2 binds to complexes
1 and 2 following the one-electron reduction of the dicobalt core.
Whereas 1 exhibits a single reduction wave, 2 exhibits a small sec-
ond wave that becomes more pronounced upon O2 binding. In
contrast to these small current-voltage perturbations of the com-
plexes in the presence of O2, addition of acetic acid (AcOH) to 1
and 2 in the presence of O2 results in a single large reduction wave.
Such a significant increase in current is consistent with a catalytic
ORR process. The need for both AcOH and O2 to engender the
fuel infrastructure.1,2 Solar energy may be stored in the form of
chemical bonds and subsequently recovered on demand in the
form of electricity by using fuel cells.3,4 The cathodic 4e–, 4H+ oxy-
gen reduction reaction (ORR) of hydrogen-based fuel cells is kinet-
ically challenging and overall energy conversion efficiencies depend
on the selective production of H2O. Platinum can meet the de-
manding criteria of efficient ORR3,5 but the metal is critical, spur-
ring efforts to develop catalysts based on earth abundant transition
metals such as cobalt6–10 and iron.11–18 ORR in such systems is often
performed in nonaqueous solution using a strong acid (e.g., tri-
fluoroacetic acid (TFA), protonated N,N-dimethylformamide
(DMF) or perchloric acid) as the proton donor.14,15,19,20 In assessing
activity, we and others often employ the faradaic efficiency for H2O
production as a performance benchmark.14,
However, faradaic
19,20
efficiency is difficult to interpret in the absence of a thermodynamic
potential that correctly accounts for the activity of the proton in
nonaqueous solution.
Dicobalt complexes in acetonitrile (MeCN) and in the presence
of O2 and strong acid demonstrate excellent faradaic efficiencies for
H2O production.21 We have prepared the pair of dicobalt complex-
es shown in Scheme 1 wherein a diamond Co(III)2(OH)2 core is
stabilized by the six coordinate ligand, dipyridylethane naphthy-
ridine (DPEN). The complex is similar to the first row metal com-
plexes of DPEN (and its fluorinated analog, DPFN) prepared by
Tilley and coworkers22– excepting the anionic bridging ligand.
24
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 5
1
2
3
4
5
6
7
8
9
Figure 1. Cyclic voltammetry of (a) 1 and (b) 2 in MeCN (0.1 M n-
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
Figure 2. (a) 1 (0.5 mM) in MeCN (0.1 M n-Bu4PF6) and AcOH (50
Bu4NPF6) under Ar (▬) and O2 (▬), and in the presence of 50 mM
mM) under O2 at different rotation rates: 100 (▬), 250 (▬), 500
(▬), 750 (▬) and 1000 rpm (▬). v = 0.02 V s–1. Ring potential: 1.17
V. Disk current is positive and ring current is negative. (b) Faradaic
yield for H2O2 production obtained by application of eq 1 at different
rotation rates.
AcOH (▬); v = 0.1 V s–1.
large current indicates that protons are necessary for catalytic ORR
turnover. Moreover, the current increase exhibits a dependence on
the amount of AcOH (Figure S2), which confirms the implication
of protons in the mechanism of ORR. At the highest acid concen-
trations, the foot of the catalytic wave shifts towards less negative
potentials, demonstrating that the kinetics of the catalytic reaction
improve with increasing acid concentration. With [AcOH] concen-
tration in excess of 50 mM (Figure S2), the current decreases due
to deactivation of the catalyst via acidolysis.
The products of ORR are typically water and hydrogen peroxide
with the latter being the undesirable product owing to its lower cell
voltage. Consequently, the design of ORR catalysts typically em-
phasizes the selectivity of H2O vs H2O2 production. The commonly
used metric for this selectivity is faradaic efficiency, as measured by
RRDE for which the theory is very well defined.25 The faradaic yield
is often determined arbitrarily at the potential for which the disk
current is highest, though it is more representative to take into
account the average yield throughout the entire catalytic region.
The faradaic efficiency for H2O2 production as a function of poten-
tial may be determined by using:
Figure 3. (a) Faradaic efficiency of 1 towards H2O production as a
function of acid pKa of different acids listed in Table 1 in MeCN (•)
and DMF (•). (b) Faradaic efficiency of 1 towards H2O production as a
function of solution pH adjusted by aqueous phosphate buffer.
crease of the ring current would be observed with an increase in
rotation rate. At a given potential, the ring and disk currents at vari-
ous rotation rates show a variance of ~10%, however with no trend
in rotation rate (Figure 2b). We attribute this variance ( 10%) in
the collection efficiency. The average faradaic efficiency of H2O
production of 44% in Figure 2b was determined by taking 1 – [far-
adaic efficiency H2O2], which was determined as the average value
across the entire trace for all rotation rates.
This RRDE study of ORR by 1 was expanded to include proton
donors of different strength in MeCN (Figures S5–S8) and DMF
(Figures S9 and S10) at different acid concentrations. The pKa of
the acids used in these studies are given in Table 1 in DMF and
MeCN.26–29 For a given acid, the current increases and the foot of
the wave shifts slightly to more positive potentials with acid con-
centration (e.g., Figure S2). These observations indicate that the
effect of acid concentration is rooted in the kinetics of ORR. Most
significantly, the H2O yield in organic solvents increases with in-
creasing acid strength. Figure 3a plots the average faradaic efficien-
cy of H2O for the different acids using eq 1. We note that the fara-
daic efficiency changed minimally with acid concentration (Table
S1) over a range where acidolysis of the compound was minimal (≤
50 mM acid concentration). As shown in Figure 3a, the faradaic
efficiency varies substantially with pKa (e.g., 1 with TFA in MeCN
gives an average faradaic efficiency for H2O production of 96% as
compared to 33% for the more weakly acidic phenol).
2ꢂꢃ(ꢁ)/ꢄ
( )
ꢂꢅ ꢁ + ꢂꢃ(ꢁ)/ꢄ
( )
%HꢀOꢀ ꢁ =
(1)
where ir(E) and id(E) are the ring and the disk current, respectively,
at potential E and N is the collection efficiency of the rotating ring
disk electrode,25 which was experimentally determined to be 0.26.
The potential at the disk was scanned through the appropriate cata-
lytic region while the potential at the ring was held at 1.17 V (all
potentials are reported vs. NHE) to ensure complete oxidization of
H2O2. Figure 2a presents a representative experiment showing the
ring and disk currents resulting from ORR of 1 in MeCN acidified
with AcOH. Figure 2b shows the corresponding faradaic yield for
H2O2 production across the potential range of ORR catalysis, ob-
tained by application of eq 1 at different rotation rates. In applying
eq 1, the average was taken over a potential range that gave a relia-
ble measure of the current at the disk (id > 0.05 mA). Control ex-
periments in the absence of catalyst show that there is no current
on the ring regardless of the acid and its concentration within the
limits of our potential window. Current from direct reduction of O2
at the disk in the absence of catalyst occurs at more negative poten-
tials (Edisk < –0.3 V vs NHE with strong acid to Edisk < –0.64 V vs
NHE in the absence of acid). The faradaic efficiency should be the
same at different rotation rates, unless hydrogen peroxide is an
intermediate in the reaction leading to water. In this case, a de-
The dependence of ORR faradaic yield on the different acids can
be explained by considering the thermodynamic standard potential
of the reduction of O2 to H2O as a function of acid pKa (Table 1).
We have developed a thermodynamic cycle akin to that developed
ACS Paragon Plus Environment

Page 3 of 5
Journal of the American Chemical Society
similar, ORR may be driven at much lower effective overpotential
Table 1. pKa of different acids and thermodynamic standard
potential of O2 reduction to H2O in MeCN and DMF.
in DMF owing to the significantly reduced value of ꢁꢆꢋ
1
2
3
4
5
6
7
8
⁄
ꢀꢈꢇꢆ,ꢈꢉ
ꢇ
(arising from a difference in ꢏꢈꢋ ,ꢊ→ꢍꢎ, Table S2). Indeed, Figure
ꢐ
MeCN
pKa HA E0/V vs. NHE
DMF
Acid, HAa
S3 shows the typical catalytic CVs of 1 in MeCN and DMF. A shift
of 650 mV is observed at the foot of the catalytic wave caused by
the difference of 779 mV between the two standard potentials (see
Table 1). Nonetheless this gain in terms of thermodynamics is
somewhat offset in terms of kinetics, as the catalytic current at the
in DMF is significantly reduced as compared to MeCN.
Experiments performed on 1 in water at different pHs lead to the
results (Figure S11) that are similar to that observed for ORR in
nonaqueous solutions. The apparent standard potential decreases
with pH according to eq 5, and accordingly ΔEORR decreases with
increasing pH. Concomitant with his lowering in the effective
overpotential, the faradaic efficiency for H2O production decreases
(Figure 3b).
Changing the bridging anion from the acetamidate in 1 to ace-
tate in 2 yields similar RRDE results in MeCN (Figures S12 and
S13) and water (Figure S14). The trend of the average faradaic
efficiency with pKa in MeCN (Figure S15) and pH in water (Figure
S16) is similar to that observed for ORR catalyst 1. A positive shift
of the catalytic wave is observed for 2 resulting in a decrease in
ΔEORR and an attendant decrease in the effective ORR overpoten-
tial (Figure S4). This gain in terms of thermodynamics is offset in
terms of kinetics, as the current at the maximum of the catalytic
wave is smaller in 2 than 1 (Figure S14a).
pKa HA
18.8
13.3
10.0
4.8
E0/V vs. NHE
–0.301
0.020
PhOH
AcOH
ClAcOH
TFA
27.2
22.3
15.3
12.6
0.430
0.720
1.133
1.293
0.213
0.518
9
a
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
phenol (PhOH), acetic acid (AcOH), chloroacetic acid (ClAcOH) and
trifluoroacetic acid (TFA).
by Costentin et al. for CO2 reduction to CO (Scheme S1).30 Using
this cycle, the standard potential is given by:
ꢁꢋ
= ꢁꢌ,ꢊ + ꢁꢋ
⁄
⁄
ꢆꢇ ꢀꢈꢇꢆ,ꢈꢉ,ꢊ
ꢆꢇ ꢀꢈꢇꢆ,ꢍꢎ
ꢋ
(2)
4∆ꢏꢈꢋꢐ
−
− 2∆ꢏ
ꢈꢇꢆ,ꢊ→ꢍꢎ
ꢒꢓ ln10
ꢑ
,ꢊ→ꢍꢎ
−
pꢔꢍ ꢈꢉ,ꢊ
2ꢑ
where eq 2 accounts for the standard potential of ORR to furnish
H2O in solvent S and in the presence of HA (ꢁꢋ ꢀꢈ ꢆ,ꢈꢉ,ꢊ) to that
⁄
ꢆꢇ
ꢇ
in aqueous solution ꢁꢆꢋ ꢀꢈ ꢆ,ꢍꢎ, corrected for the interliquid junc-
⁄
ꢇ
ꢇ
tion potential, ꢁꢌ,ꢊ, between water and solvent S. The terms in eq 2
are defined within Scheme S1. Substituting the constants in eq 2 for
MeCN and DMF, respectively, yields:
Our studies establish that an increase in the effective overpoten-
tial of ORR (i.e., increased ΔEORR) is accompanied by an increase in
faradic efficiency. Our model is general and it applies to previously
published ORR catalysts. Figure 4 shows a plot of reported faradaic
efficiencies versus effective overpotential for 1 and 2 together with
selected ORR catalysts operating in organic solvents.14,15,21 As con-
veyed by eq 6, the effective overpotential, ΔEORR, is obtained by
subtracting the standard reduction potential of each catalyst from
the standard potential of O2 reduction in a given solvent as deter-
ꢒꢓ ln10
ꢁꢆꢋ
ꢁꢆꢋ
= 2.038 −
= 0.799 −
pꢔꢍ,ꢈꢉ,ꢕꢖꢗꢘ
(3)
(4)
⁄
⁄
ꢀꢈꢇꢆ,ꢕꢖꢗꢘ
ꢇ
ꢇ
ꢑ
ꢒꢓ ln10
ꢑ
pꢔꢍ,ꢈꢉ,ꢙꢕꢚ
ꢀꢈꢇꢆ,ꢙꢕꢚ
We note that these equations contain no term for acid concentra-
tion, which is reflected in the lack of a noticeable concentration
effect on selectivity (Table S1). For the case of water, the standard
potential evolves with pH in the standard fashion,
mined from eq 2. We note the value of ∆ꢏꢈꢋ
used in eq 2 has
ꢐ
,ꢊ→ꢍꢎ
recently been re-examined;31 whereas the various values of
∆ꢏꢈꢋ
(Table S3) alter the absolute value of ΔEORR, the trend
ꢒꢓ ln10
ꢑ
ꢒꢓ ln10
ꢑ
ꢁꢆꢋ´
= ꢁꢆꢋ
−
pH = 1.23 −
ꢐ
(5)
pH
,ꢊ→ꢍꢎ
⁄
⁄
ꢀꢈ ꢆ,ꢍꢎ
ꢀꢈ ꢆ,ꢍꢎ
ꢇ
ꢇ
ꢇ
ꢇ
in Figure 4 is maintained. We have confined our analysis to ORR
catalysts operating in DMF and MeCN, as the thermodynamic
parameters needed for other solvent systems are unknown or poor-
ly defined; thus literature examples such as those in acetone20 and
benzonitrile19 are not included in our analysis. Additionally, the
standard potential for solid state catalysts is unknown; hence Fig-
ure 4 does not include solid state catalyst in the comparison.
The result of Figure 4 is striking inasmuch as very different cata-
lysts with respect to metal and ligand type are undistinguished with
regard to ORR faradaic efficiency. High faradaic efficiencies are
obtained only at high effective overpotentials. Per the model em-
bodied by eq 2, most studies achieve these high overpotentials by
employing strong acids in nonaqueous solutions. In this regard,
compound 1 appears to exhibit better performance in DMF but
this is not a result of better intrinsic catalyst activity but due to a
lowering of the effective overpotential (eq 4 as compared to 3) as a
result of greater activity of the proton in DMF (Table S3). It is
interesting to note that the reported faradaic efficiencies for most
catalysts to date are indifferent to the kinetics of the ORR reaction.
For instance, in compounds 4 and 5, a proton relay group is posi-
From eqs 3 and 4, we obtain the E0 values presented in Table 1.
These data reveal that the improved selectivity for H2O production
with increasing acid strength (Figure 3a) scales with the increasing
standard potential of the ORR reaction for the different acids. In-
asmuch as the applied potential to drive the ORR is defined by the
reduction potential of the catalyst,
Δꢁꢆꢛꢛ = ꢁꢋ ꢀꢈ ꢆ,ꢈꢉ,ꢊ – ꢁꢗꢋꢜ (ꢝꢝꢝ,ꢝꢝꢝ)/ꢗꢜ (ꢝꢝ,ꢝꢝꢝ)
(6)
⁄
ꢆꢇ
ꢇ
ꢇ
ꢇ
the faradaic efficiency scales with ΔEORR, which is the effective
overpotential. Thus the ORR pathway leading to the O–O bond
cleavage needed to produce H2O becomes favored in strong acid
where there is a high effective overpotential. Conversely, with weak
acids, the effective overpotential for ORR to H2O is small, and the
pathway leading to H2O production is disfavored relative to H2O2.
The same trend of improved H2O selectivity with stronger acid is
observed in DMF solution (Figure 3a, navy circles). Whereas the
faradaic efficiency for H2O production in DMF and MeCN are
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 5
(2) Lewis, N. S.; Nocera, D. G. The Bridge 2015, 46, 41.
1
2
3
4
5
6
7
8
(3) Borup, R.; Meyers, J.; Pivovar, B.; Kim, Y. S.; Mukundan, R.;
Garland, N.; Myers, D.; Wilson, M.; Garzon, F.; Wood, D.; Zelenay, P.;
More, K.; Stroh, K.; Zawodzinski, T.; Boncella, J.; McGrath, J. E.; Inaba,
M.; Miyatake, K.; Hori, M.; Ota, K.; Ogumi, Z.; Miyata, S.; Nishikata, A.;
Siroma, Z.; Uchimoto, Y.; Yasuda, K.; Kimijima, K.; Iwashita, N. Chem. Rev.
2007, 107, 3904.
(4) Cook, T. R.; Dogutan, D. K.; Reece, S. Y.; Surendranath, Y.; Teets,
T. S.; Nocera, D. G. Chem. Rev. 2010, 110, 6474.
(5) Steele, B. C. H.; Heinzel, A. Nature 2001, 414, 345.
(6) Anson, F. C.; Shi, C.; Steiger, B. Acc. Chem. Res. 1997, 30, 437.
(7) Kadish, K. M.; Shen, J.; Frémond, L.; Chen, P.; Ojaimi, M. E.;
Chkounda, M.; Gros, C. P.; Barbe, J.-M.; Ohkubo, K.; Fukuzumi, S.; Gui-
lard, R. Inorg. Chem. 2008, 47, 6726.
(8) Dogutan, D. K.; Stoian, S. A.; McGuire, R.; Schwalbe, M.; Teets, T.
S.; Nocera, D. G. J. Am. Chem. Soc. 2011, 133, 131.
(9) Geiger, T.; Anson, F. C. J. Am. Chem. Soc. 1981, 103, 7489.
(10) McGuire Jr, R.; Dogutan, D. K.; Teets, T. S.; Suntivich, J.; Shao-
Horn, Y.; Nocera, D. G. Chem. Sci. 2010, 1, 411.
(11) Collman, J. P.; Devaraj, N. K.; Decréau, R. A.; Yang, Y.; Yan, Y.-L.;
Ebina, W.; Eberspacher, T. A.; Chidsey, C. E. D. Science 2007, 315, 1565.
(12) Collman, J. P.; Decréau, R. A.; Lin, H.; Hosseini, A.; Yang, Y.; Dey,
A.; Eberspacher, T. A. Proc. Natl. Acad. Sci. U. S. A. 2009, 106, 7320.
(13) Collman, J. P.; Ghosh, S.; Dey, A.; Decréau, R. A.; Yang, Y. J. Am.
Chem. Soc. 2009, 131, 5034.
(14) Carver, C. T.; Matson, B. D.; Mayer, J. M. J. Am. Chem. Soc. 2012,
134, 5444.
(15) Wasylenko, D. J.; Rodríguez, C.; Pegis, M. L.; Mayer, J. M. J. Am.
Chem. Soc. 2014, 136, 12544.
(16) Chatterjee, S.; Sengupta, K.; Samanta, S.; Das, P. K.; Dey, A. Inorg.
Chem. 2015, 54, 2383.
(17) Samanta, S.; Mittra, K.; Sengupta, K.; Chatterjee, S.; Dey, A. Inorg.
Chem. 2013, 52, 1443.
(18) Samanta, S.; Das, P. K.; Chatterjee, S.; Sengupta, K.; Mondal, B.;
Dey, A. Inorg. Chem. 2013, 52, 12963.
(19) Rosenthal, J.; Nocera, D. G. Acc. Chem. Res. 2007, 40, 543.
(20) Kakuda, S.; Peterson, R. L.; Ohkubo, K.; Karlin, K. D.; Fukuzumi,
S. J. Am. Chem. Soc. 2013, 135, 6513.
(21) Fukuzumi, S.; Mandal, S.; Mase, K.; Ohkubo, K.; Park, H.; Benet-
Buchholz, J.; Nam, W.; Llobet, A. J. Am. Chem. Soc. 2012, 134, 9906.
(22) Davenport, T. C.; Tilley, T. D. Angew. Chem. Int. Ed. Engl. 2011,
50, 12205.
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
Figure 4. Faradaic efficiency for H2O production vs. Δꢁꢆꢛꢛ(=
ꢁꢆꢋ ꢀꢈ ꢆ,ꢈꢉ,ꢊ – ꢁꢞꢋꢍꢟꢍꢠꢡꢢꢟ) for: 1 in MeCN (●) with (1a) PhOH, (1b)
⁄
ꢇ
MeCNꢇ, (1c) ClAcOH, (1d) TFA; 1 in DMF with (●), (1e) AcOH,
(1f) TFA; 2 in MeCN (●) with (2a) PhOH, (2b) AcOH; (3) Fe tetra-
phenylporphyrin in DMF with HClO4 (●), ref 15; (4) Fe meso-tetra(2-
carboxyphenyl)porphine in MeCN with (HDMF)+ (●), ref 14; (5) Fe
meso-tetra(4-carboxyphenyl)porphine in MeCN with (HDMF)+ (●),
ref 14; (6) CoIII2(trpy)2(μ-bpp)(μ-1,2-O2)]3+ (bpp = bis(pyridyl)-
pyrazolate, trpy = terpyridine) in MeCN with TFA (●), ref 21.
tioned toward and away from a Fe porphyrin ring, respectively.
Whereas the kinetics of the ORR are affected by the involvement of
the proton relay,14 the faradaic efficiency for the two compounds is
similar as consequence of similar effective overpotentials.
In summary, we have developed a model that shows that ORR
selectivity of catalysts is largely dictated by the effective overpoten-
tial. Our model reveals that in most systems reported to date, high
ORR selectivities for H2O is a result of large effective overpoten-
tials for the reaction, achieved by the use of strong acids. The chal-
lenge to developing better ORR catalysts will be to maintain high
catalytic efficiencies under conditions where the overpotential for
ORR is greatly reduced.
ASSOCIATED CONTENT
(23) Davenport, T. C.; Tilley, T. D. Dalton Trans. 2015, 44, 12244.
(24) Davenport, T. C.; Ahn, H. S.; Ziegler, M. S.; Tilley, T. D. Chem.
Commun. 2014, 50, 6326.
(25) Allen J. Bard, L. R. F. Electrochemical Methods: Fundamentals and
Applications, 2nd Edition; Wiley: New York, 2001.
(26) Andrieux, C. P.; Gamby, J.; Hapiot, P.; Savéant, J.-M. J. Am. Chem.
Soc. 2003, 125, 10119.
(27) Costentin, C.; Evans, D. H.; Robert, M.; Savéant, J.-M.; Singh, P. S.
J. Am. Chem. Soc. 2005, 127, 12490.
(28) Kütt, A.; Rodima, T.; Saame, J.; Raamat, E.; Mäemets, V.;
Kaljurand, I.; Koppel, I. A.; Garlyauskayte, R. Y.; Yagupolskii, Y. L.; Ya-
gupolskii, L. M.; Bernhardt, E.; Willner, H.; Leito, I. J. Org. Chem. 2011, 76,
391.
(29) Izutsu, K. Acid-Base Dissociation Constants in Dipolar Aprotic Sol-
vents; Blackwell: Boston, 1990.
Supporting Information. Full experimental details, additional elec-
trochemical data. This material is available free of charge via the Inter-
net at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
dnocera@fas.harvard.edu
ACKNOWLEDGMENT
This work was supported by the U.S. Department of Energy Office of
Science under Award Number DE-SC0009758. We also thank the
TomKat Charitable Trust for support of this work. C.N.B. acknowl-
edges the NSF’s Graduate Research Fellowship Program. We are grate-
ful to Prof. Cyrille Costentin for helpful discussions.
(30) Costentin, C.; Drouet, S.; Robert, M.; Savéant, J.-M. Science 2012,
338, 90.
(31) Pegis, M. L.; Roberts, J. A. S.; Wasylenko, D. J.; Mader, E. A.;
Appel, A. M.; Mayer, J. M. Inorg. Chem. 2015, 54, 11883.
REFERENCES
(1) Lewis, N. S.; Nocera, D. G. Proc. Natl. Acad. Sci. U. S. A. 2006, 103,
15729.
ACS Paragon Plus Environment

Page 5 of 5
Journal of the American Chemical Society
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
5
ACS Paragon Plus Environment