Lowering water oxidation overpotentials using the ionisable imidazole of copper(2-(2′-pyridyl)imidazole)

Article abstract of DOI:10.1039/c6cc09208j

Rapid and low overpotential oxidation of water to dioxygen remains a key hurdle for storage of solar energy. Here, we address this issue by demonstrating that deprotonation of 2-(2′-pyridyl)-imidazole (pimH)-ligated copper complexes promotes water oxidation at low overpotential and low catalyst loading. This improves upon other work on homogeneous copper-based water oxidation catalysts, which are highly active, but limited by high overpotentials. EPR and UV-vis spectroscopic evaluation of catalyst speciation shows that at pH ≥ 12 coordinated pimH is deprotonated and a bis(hydroxide) Cu²⁺ active catalyst forms. Rapid electrochemical water oxidation (35 s^-1, 0.85 V onset potential) was observed with 150 μM catalyst. These results demonstrate that catalytic water oxidation potentials can be shifted by hundreds of mV in homogeneous metal catalysts bearing an ionisable imidazole ligand.

Full text of DOI:10.1039/c6cc09208j

ChemComm
View Article Online
View Journal
COMMUNICATION
Lowering water oxidation overpotentials using the
ionisable imidazole of copper(2-(20-pyridyl)imidazole)†
Cite this:DOI: 10.1039/c6cc09208j
Leea A. Stott,a Kathleen E. Prosser,a Ellan K. Berdichevsky,a Charles J. Walsbya and
Jeffrey J. Warren*ab
Received 18th November 2016,
Accepted 8th December 2016
DOI: 10.1039/c6cc09208j
www.rsc.org/chemcomm
Rapid and low overpotential oxidation of water to dioxygen remains coupled to energy storage in H2. For the artificial water oxidation
a key hurdle for storage of solar energy. Here, we address this issue half reaction, the most widely investigated catalysts include homo-
by demonstrating that deprotonation of 2-(20-pyridyl)-imidazole geneous Ru2 and Ir complexes;3 and heterogeneous Ir,4 Co, Ni, and
(pimH)-ligated copper complexes promotes water oxidation at low Mn5 oxides. Several recent reports investigate water oxidation using
overpotential and low catalyst loading. This improves upon other Cu complexes,6–9 Cu-solutions,10 or Cu materials.11 The key
work on homogeneous copper-based water oxidation catalysts, challenge for any of these catalysts is the ability to access high
which are highly active, but limited by high overpotentials. EPR and oxidation states implicated in proposed water oxidation mechan-
UV-vis spectroscopic evaluation of catalyst speciation shows that at isms. Here, we address this issue using homogeneous Cu com-
pH Z 12 coordinated pimH is deprotonated and a bis(hydroxide) Cu2+ plexes as examples. In proposed and calculated Cu-catalysed water
active catalyst forms. Rapid electrochemical water oxidation (35 sꢀ1
,
oxidation mechanisms,7,9 the first step is oxidation of CuII to CuIII.
0.85 V onset potential) was observed with 150 lM catalyst. These Consequently, the question is: how can the CuIII/II potential be
results demonstrate that catalytic water oxidation potentials can be modified to gain access to active Cu catalysts at lower energy?
shifted by hundreds of mV in homogeneous metal catalysts bearing an
ionisable imidazole ligand.
Photosystem II (PSII), Nature’s solution to water oxidation, uses
the 4MnIV state of the Mn4Ca oxygen evolving complex (OEC) to
generate O2.12 The deprotonation of Mn-ligating His332 of PSII may
Efficient catalysis of the four-electron oxidation of 2H2O to 4H+ affect the energetics of catalysis,13 facilitating access to MnIV.12 In
and O2, and the corresponding reduction of protons to 2H2, are another biological example, the protonation state of histidine
two key challenges for solar energy production. Perhaps the ligated to Fe in Rieske iron–sulphur clusters shifts E10(FeIII/II) by
largest technical hurdle is formation of the OQO double bond 270 mV.14 In small molecule models, it also is known that metal-
from two water molecules.1 This transformation is at the heart coordinated ligands bearing ionisable protons become more acidic
of typically high kinetic barriers (overpotentials) for water than their unligated congeners.15 Deprotonation of these ligands
oxidation. Progress is being made, but there is still a lack induces cathodic shifts in E10 (shown schematically in Fig. 1), for
of understanding of how to lower the electrochemical over- example in imidazole- and imidazoline-ligated metals (Fe, Ru,
potential for water oxidation. Herein, we demonstrate water Co) in aqueous16 and organic solvent.17 Further, in binuclear
oxidation in alkaline water using Cu2+ complexes of pimH Cu complexes, deprotonation of an imidazole results in nearly
(pimH = 2-(20-pyridyl)imidazole). Using UV-vis and EPR spectro- a 1 V shift in E10(CuII/I) in MeOH/H2O solvent.18 In general,
scopy, we determined speciation as a function of pH to show the shifts in E10 upon deprotonation range from 200–500 mV.
the role of pimH deprotonation in tuning overpotentials.
Natural photosystems collect energy from sunlight, oxidize
water, and store energy in chemical bonds (e.g., glucose). The goal
is the same in artificial photosystems, where water oxidation is
a Department of Chemistry, Simon Fraser University, 8888 University Drive,
Burnaby BC V5A 1S6, Canada. E-mail: j.warren@sfu.ca
b Canadian Institute for Applied Research, CIFAR Azrieli Global Scholar and
Bio-Inspired Solar Energy Programs, Toronto, ON M5G 1Z8, Canada
† Electronic supplementary information (ESI) available: Full experimental
Fig. 1 Metal (3+ /2+) reduction potential shift with different protonation
details, electrochemical data, UV-visible titrations, EPR data and simulation
states of ligated imidazole.
parameters. See DOI: 10.1039/c6cc09208j
This journal is ©The Royal Society of Chemistry 2016
Chem. Commun.

View Article Online
Communication
ChemComm
Table 1 Comparison of Cu–water oxidation catalysts
We take advantage of this phenomenon and demonstrate that
deprotonaton of an imidazole ligand in a Cu complex can lower
water oxidation overpotentials.
Cyclic voltammograms (CVs) of 1 : 1 mixtures of Cu(OAc)2
and pimH (150 mM) were obtained at pH values between 9 and
13 in 0.1 M NaOAc. At pH Z 11 (Fig. 2), substantial, irreversible
current response was observed, with onset at 0.85 V and peak
currents near 1.5 V. A basal plane graphite working electrode
was used, potentials are with respect to the normal hydrogen
electrode (NHE), and scan rates were 100 mV sꢀ1. At pH values
Ligand
pimH bpy bpy-OH glya L4b
pH
Electrolyte
12
12
12.4
11
11.5
e
c
c
c
d
[cat] mM
150
2.5
0.85
1.5
35
1000 1000
890 1000
3.7g 0.4
1.1 0.7
1.35 0.8
Peak current density f (mA cmꢀ2
E10 onset (V)
)
1.3
1.0
1.8
0.85
E10 peak (V)
1.35 1.2
100 0.4
TOF (calc., sꢀ1
)
33
3.6
a
b
c
Triglycylglycine. N,N0-(1,2-Phenylene)bis(N2-methyl-oxalamide) 0.1 M
d
e
f
NaOAc. 0.25 M Na phosphate. 0.1 M Na phosphate. Scan rate =
below 11, there is negligible current enhancement in compar- 100 mV sꢀ1
. .
Scan rate = 10 mV sꢀ1
g
ison to the individual reaction components or a bare electrode
(see ESI†)
potentially problematic for Cu water oxidation systems. However,
such an analysis was used for the other systems and allows
comparison on an equal footing. Differences in experimental
conditions also are problematic (e.g., electrolyte, scan rate). We
emphasize that the Cu(pimH) system has similar overpotentials
and catalytic current densities (at peak potentials), but at much
lower catalyst loadings than the other examples in Table 1.
The production of O2 in 1 : 1 CuII : pimH mixtures at constant
potentials (1.1 V, 0.1 M NaOAc, pH 12.1, 90 min) was monitored
using a fluorescence probe. Over the course of the experiment,
dissolved O2 increased from 0 to 0.5 mg mLꢀ1 using a 0.09 cm2
basal plane graphite electrode (Fig. 3). The increase in detected O2
is significantly larger than background detection of O2. We also
carried out experiments with ITO electrodes (3.3 cmꢀ2 area). The
observed currents and detected O2 scale with electrode surface
area, with B5 mg mLꢀ1 O2 detected in 30 min (see ESI†). In all
cases, gas bubbles form on the electrode surfaces, in qualitative
agreement with detected O2. The pH of these solutions remained
constant over the time course of electrolyses.
The CVs of 1 : 1 CuII : pimH mixtures exhibit lower onset poten-
tials for water oxidation at pH 12 than some reported Cu-based
catalysts (i.e., 1.0 V6 for Cu-bpy2+, 1.1 V7 for Cu(triglycylglycine)2ꢀ
,
and 0.94 V8 for Cu(bpy-OH)2+ (bpy = 2,20-bipyridyl, bpy-OH =
6,60-dihydroxy-2,20-bipyridyl)). Note that the overpotential shifts
to B0.8 V in the Cu(bpy-OH)2+ system at pH 4 13, which is
attributed to deprotonation of the ligand.
The shape of the catalytic CV waves for the Cu–pimH system
indicate that the electrocatalysis is affected by both diffusion
and the kinetics of the reaction, as observed for related Cu
systems.6–9 The limiting catalytic current increases linearly with
[Cu–pimH], between 25 and 250 mM, consistent with eqn (1).19
1/2
icat = ncatFA[Cu](kcatDCu
)
(1)
As for other Cu water oxidation systems, the Randles–Sevcik
equation is used to compare catalytic current to the reversible
diffusive current. Here, the quasireversible CuII/I couple, near 0 V,
is used as an estimate of the diffusive current, and the response
varies linearly with the square root of the scan rate (see ESI†). The
slope of a plot of icat/id versus vꢀ1/2 gives the pseudo-first order
catalytic rate constant for Cu(pimH) as 35 sꢀ1. This value is
Removal of the carbon electrode from solution and scanning in
fresh 0.1 M NaOAc alone shows no deposition of active catalysts.
During overnight electrolysis experiments, copper-coloured deposits
were observed on electrodes, suggesting that the solutions are not
stable over extended periods under controlled electrolysis.
similar to the value calculated for Cu(triglycylglycine)2ꢀ (33 sꢀ1
)
and 30% of that for Cu(bpy)2+ (100 sꢀ1). In contrast, our TOF
value is about 100-fold larger than for Cu(bpy-OH)2+ (0.4 sꢀ1) and
10-fold larger than for Cu(N,N0-(1,2-phenylene)bis(N2-methyl-
oxalamide)) (L4, 3.6 sꢀ1). A comparison of each of these catalysts
is set out in Table 1. Complex speciation (see below) makes
the applicability of the above analysis using peak currents
In contrast to reported Cu water oxidation systems, the current
response for Cu(pimH) shows several features, suggesting that
Fig. 3 Dissolved O2 concentration as a function of time during controlled
potential electrolysis experiments (1.1 V applied potential). [Cu(OAc)2] =
[pimH] = 250 mM.
Fig. 2 CVs of 150 mM Cu(OAc)2 + pimH in 0.1 M NaOAc.
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2016

View Article Online
ChemComm
Communication
Fig. 4 UV-visible spectra as a function of pH for Cu(pimH).
more than one species could be present in solution. We carried
out several spectroscopic experiments to gain greater insight
into the speciation of the Cu(pimH) system. First, equimolar
(150 mM) amounts of Cu(OAc)2 and pimH were mixed in 0.1 M
Fig. 5 Experimental (black) and simulated (red) EPR spectra of 150 mM
CuII + pimH at pH 8 and pH 12. Simulation details are given in the ESI.†
details of EPR simulation are summarized in the ESI.† At pH 9
there is approximately a 50/50 mixture of two species, suggesting a
pKa value near 9, in qualitative agreement with the UV-vis data. The
decreases in overall EPR signal intensity at intermediate pH values
also are consistent with formation of an EPR silent [Cu2(OH)2]
diamond core, as proposed for speciation of the Cu-bpy
system.6,27,28 At pH Z 12, EPR spectra are dominated by a complex
with new g-values and coupling constants, in addition to a sharp,
multi-line component. The former is assigned to the deprotonated,
hydroxide ligated complex, Cu(OH)2(pim)ꢀ, and the latter to aggre-
gated Cu species, based on comparison to spectra of Cu(OAc)2 at the
same pH. The shift in g8 is consistent with the presence of more
strongly donating ligands (i.e., pimꢀ and OHꢀ).28 The appearance of
Cu species without pim ligands is distinct from observations for
Cu(bpy)2+,6 but consistent with the weaker metal binding of pimH.29
The electrochemical, UV-vis and EPR data suggest a specia-
tion scheme that incorporates facets of the chemistry of metal-
imidazole and Cu-bpy systems. (Fig. 6). As the pH of 0.1 M
NaOAc solutions is increased, the imidazole in Cu(pimH) is
deprotonated first, with a pKa near 8.5. The anionic pimꢀ ligand
is a stronger s-donor than the neutral ligand, depressing the pKa
values of Cu-coordinated OH2 in comparison to other values
from the literature.23 As the pH is increased, EPR-silent species
develop, which are then converted to Cu(OH)2(pim)ꢀ monomers
at pH Z 12 and it appears that some Cu also is lost from the
complexes at high pH.
sodium acetate (pH 5) to give a very pale blue solution (lmax
=
690 nm, e B 250 Mꢀ1 cmꢀ1). Incrementally adjusting the pH to
B10 using 5 M NaOH causes a blue shift in lmax to 625 nm
(Fig. 4). The blue shift in the Cu d–d band is accompanied by a
red shift in the ligand p–p* bands (309 to 330 nm). These
spectroscopic changes are consistent with deprotonation of the
pimH ligand (i.e., a red shift of ligand bands20 and a corres-
ponding blue shift in Cu bands).
Further adjustment from pH 10 to pH 13 shifts the d–d band to
590 nm, with a shoulder at 530 nm. These changes are reversible
and the spectra are distinct from Cu(OAc)2 at alkaline pH. In this
case, CuII readily hydrolyses21 and we observe precipitates of Cu
after several minutes. Precipitation is not apparent in 150 mM
Cu(pimH) solutions, but at higher concentrations, solutions
become cloudy over extended periods (depending on pH and
the concentration of Cu(pimH)).
Plots of absorbance (376 nm) as a function of pH show two
inflection points at pH values of 8.5 and 11.8 (see ESI†). Similar
plots at 525 nm (Cu bands) also show the pH 8.5 inflection. As
noted above, the first transition is assigned to deprotonation of
pimH in Cu(pimH)(OH)22+ to give Cu(pim)(OH)2+. The apparent
pKa of 8.5 is larger than for the tris-chelate (pKa = 7.9),22
consistent with weaker OH2 donor ligands. The second transi-
tion likely corresponds to deprotonation of a water ligand.23
Further insights into Cu(pimH) speciation were obtained
using electron paramagnetic resonance (EPR) spectroscopy.
Cu(OAc)2 + pimH solutions (150 mM) were prepared in 0.1 M
NaOAc and flash frozen in liquid nitrogen. All spectra were
collected at 100 K. EPR spectra were simulated using EasySpin24
and all parameters are tabulated in the ESI.† At pH 8, we
observe a single species with hyperfine coupling of Cu to 14N
of pimH (Fig. 5). We attribute this spectrum to the protonated
compound Cu(pimH)2+, based on EPR spectra of 2-(2-pyridyl)-
benzimidazole Cu complexes.25 EPR spectra of CuCl2(pimH)
are reported in DMF, but hyperfine coupling was not resolved
and the solution contained a mixture of compounds.26
As the pH is raised, the intensity of the EPR signal decreases
and the spectra are best simulated by a mixture of species. Full
Incorporation of an ionisable proton in pimH offers a path-
way to a more electron rich, and therefore more easily oxidized,
Fig. 6 Proposed speciation in the Cu(pimH) system at different pH.
This journal is ©The Royal Society of Chemistry 2016
Chem. Commun.

View Article Online
Communication
ChemComm
U. Hintermair, R. H. Crabtree, G. W. Brudvig and C. A. Schmuttenmaer,
Nat. Commun., 2015, 6, 1.
Cu site. We suggest that, in analogy to other pimH transition
metal complexes, this lowers the potential at which CuII is
oxidized to CuIII to enter the putative water oxidation catalytic
cycle. Ligand non-innocence has been implicated in other
molecular Cu-based water oxidation systems,8,9 but it is not a
requirement for catalysis.7 To the best of our knowledge, pimH
does not act as a non-innocent ligand. We are now investigating
the electronic properties of these complexes in greater detail.
In sum, equimolar mixtures of CuII and 2-(20-pyridyl)imidazole
(pimH) can catalyse oxidation of water to dioxygen. Importantly
significant increase in catalytic current occurs at pH 4 12, with
B300 mV overpotential at the onset of catalytic current. We
attribute this behaviour to ionization of the ligated pimH ligand.
This phenomenon has been suggested for Co complexes30 and
for Cu(bpy-OH)2+ 8 but the result here is significant because,
deprotonation of Cu(pimH) occurs at B2 pH units lower than
for Cu(bpy-OH)2+ and shows 100-fold more rapid turnover. The
results presented here show that deprotonation of an ionisable
imidazole ligand in homogeneous water oxidation catalysts can
lower metal reduction potentials and therefore catalytic over-
potentials. We think that our detailed investigation of variable
pH speciation serves as a first-principles guide to new catalyst
designs. These principles should be applicable to any oxidation
catalyst, and we think that the next grand challenge will be
incorporating the design features that we highlight here into
heterogeneous systems.
4 J. D. Blakemore, N. D. Schley, M. N. Kushner-Lenhoff, A. M. Winter,
¨
ˆ
F. D. AoSouza, R. H. Crabtree and G. W. Brudvig, Inorg. Chem., 2012,
51, 7749.
5 M. Huynh, D. K. Bediako and D. G. Nocera, J. Am. Chem. Soc., 2014,
136, 6002.
6 S. M. Barnett, K. I. Goldberg and J. M. Mayer, Nat. Chem., 2012,
4, 498.
7 M.-T. Zhang, Z. Chen, P. Kang and T. J. Meyer, J. Am. Chem. Soc.,
2013, 135, 2048.
8 T. Zhang, C. Wang, S. Liu, J.-L. Wang and W. Lin, J. Am. Chem. Soc.,
2013, 136, 273.
9 P. Garrido-Barros, I. Funes-Ardoiz, S. Drouet, J. Benet-Buchholz,
F. Maseras and A. Llobet, J. Am. Chem. Soc., 2015, 137, 6758.
10 (a) Z. Chen and T. J. Meyer, Angew. Chem., Int. Ed., 2013, 52, 700;
(b) L. Zhu, J. Du, S. Zuo and Z. Chen, Inorg. Chem., 2016, 55, 7135.
11 S. Cui, X. Liu, Z. Sun and P. Du, ACS Sustainable Chem. Eng., 2016,
4, 2593.
12 N. Cox, M. Retegan, F. Neese, D. A. Pantazis, A. Boussac and W. Lubitz,
Science, 2014, 345, 804.
13 T. A. Stich, G. J. Yeagle, R. J. Service, R. J. Debus and R. D. Britt,
Biochemistry, 2011, 50, 7390.
14 (a) Y. Zu, J. A. Fee and J. Hirst, J. Am. Chem. Soc., 2001, 123, 9906;
(b) Y. Zu, M. M.-J. Couture, D. R. J. Kolling, A. R. Crofts, L. D. Eltis,
J. A. Fee and J. Hirst, Biochemistry, 2003, 42, 12400.
15 e.g., metal aquo complexes. See: (a) C. F. Baes and R. E. Mesmer, The
hydrolysis of cations, Wiley, New York, 1976; (b) D. W. Barnum, Inorg.
Chem., 1983, 22, 2297.
16 (a) A. R. Reddi, C. J. Reedy, S. Mui and B. R. Gibney, Biochemistry,
2007, 46, 291; (b) R. C. Rocha, F. N. Rein and H. E. Toma, J. Braz.
Chem. Soc., 2001, 12, 234; (c) S. J. Slattery, J. K. Blaho, J. Lehnes and
K. A. Goldsby, Coord. Chem. Rev., 1998, 174, 391.
17 (a) J. J. Warren and J. M. Mayer, J. Am. Chem. Soc., 2008, 130, 2774;
(b) A. Wu, J. Masland, R. D. Swartz, W. Kaminsky and J. M. Mayer,
Inorg. Chem., 2007, 46, 11190; (c) C. T. Saouma, W. Kaminsky and
J. M. Mayer, J. Am. Chem. Soc., 2012, 134, 7293.
18 A. Wilting, M. Ku¨gler and I. Siewert, Inorg. Chem., 2016, 55, 1061.
19 ncat = 4 eꢀ for water oxidation, F is the Faraday constant, A is the
electrode surface area, [Cu] is the catalyst concentration, DCu is the
diffusion coefficient, and kcat is the pseudo-first order rate constant.
20 M. Novo, M. Mosquera and F. R. Prieto, Can. J. Chem., 1992, 70, 823.
21 D. D. Perrin, J. Chem. Soc., 1960, 3189.
Work at Simon Fraser University is supported by NSERC
(to J. J. W. (RGPIN0005559) and to C. J. W. (RGPIN312575)), by
a SFU President’s Research Grant (to J. J. W.), and by the
Canadian Institute for Advanced Research (to J. J. W.). K. E. P.
acknowledges support from NSERC CGSM and Vanier scholar-
ships. L. A. S. and E. K. B. are supported by funds from the
Canada Summer Jobs Program (Project 01374504 to J. J. W.).
22 R. K. Boggess and R. B. Martin, Inorg. Chem., 1974, 13, 1525.
¨
23 (a) M. Kruppa, D. Frank, H. Leffler-Schuster and B. Konig, Inorg.
Chim. Acta, 2006, 359, 1159; (b) R. Gilson and M. C. Durrant, Dalton
Trans., 2009, 10223; (c) J. Ali-Torres, L. Rodriguez-Santiago and
M. Sodupe, Phys. Chem. Chem. Phys., 2011, 13, 7852.
24 S. Stoll and A. Schweiger, J. Magn. Reson., 2006, 178, 42.
25 K. E. Prosser, S. W. Chang, F. Saraci, P. H. Lee and C. J. Walsby,
J. Inorg. Biochem., 2017, 167, 89.
References
1 (a) J. D. Blakemore, R. H. Crabtree and G. W. Brudvig, Chem. Rev.,
2015, 115, 12974; (b) N. S. Lewis and D. G. Nocera, Proc. Natl. Acad.
Sci. U. S. A., 2006, 103, 15729; (c) D. J. Wasylenko, R. D. Palmer and
C. P. Berlinguette, Chem. Commun., 2013, 49, 218.
26 B. De Castro, C. Freire, D. Domingues and J. Gomes, Polyhedron,
1991, 10, 2541.
27 I. Fabian, Inorg. Chem., 1989, 28, 3805.
28 E. Garribba, G. Micera, D. Sanna and L. Strinna-Erre, Inorg. Chim.
Acta, 2000, 299, 253.
29 W. J. Eilbeck and F. Holmes, J. Chem. Soc. A, 1967, 1777.
30 I. Siewert and J. Gał˛ezowska, Chem. – Eur. J., 2015, 21, 2780.
2 (a) J. J. Concepcion, J. W. Jurss, M. K. Brennaman, P. G. Hoertz,
A. O. T. Patrocinio, N. Y. M. Iha, J. L. Templeton and T. J. Meyer, Acc.
Chem. Res., 1954, 2009, 42; (b) L. Duan, F. Bozoglian, S. Mandal,
B. Stewart, T. Privalov, A. Llobet and L. Sun, Nat. Chem., 2012, 4, 418.
3 (a) J. D. Blakemore, N. D. Schley, D. Balcells, J. F. Hull, G. W. Olack,
C. D. Incarvito, O. Eisenstein, G. W. Brudvig and R. H. Crabtree, J. Am.
Chem. Soc., 2010, 132, 16017; (b) S. W. Sheehan, J. M. Thomsen,
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2016