CO2 reduction catalyzed by mercaptopteridine on glassy carbon

Article abstract of DOI:10.1021/ja5081103

The catalytic reduction of CO₂ is of great current interest because of its role in climate change and the energy cycle. We report a pterin electrocatalyst, 6,7-dimethyl-4-hydroxy-2-mercaptopteridine (PTE), that catalyzes the reduction of CO₂ and formic acid on a glassy carbon electrode. Pterins are natural cofactors for a wide range of enzymes, functioning as redox mediators and C1 carriers, but they have not been exploited as electrocatalysts. Bulk electrolysis of a saturated CO₂ solution in the presence of the PTE catalyst produces methanol, as confirmed by gas chromatography and ¹³C NMR spectroscopy, with a Faradaic efficiency of 10-23%. FTIR spectroelectrochemistry detected a progression of two-electron reduction products during bulk electrolysis, including formate, aqueous formaldehyde, and methanol. A transient intermediate was also detected by FTIR and tentatively assigned as a PTE carbamate. The results demonstrate that PTE catalyzes the reduction of CO₂ at low overpotential and without the involvement of any metal.

Full text of DOI:10.1021/ja5081103

Communication
pubs.acs.org/JACS
CO2 Reduction Catalyzed by Mercaptopteridine on Glassy Carbon
Dongmei Xiang, Donny Magana, and R. Brian Dyer*
Department of Chemistry, Emory University, Atlanta, Georgia 30322, United States
S
* Supporting Information
idine (PTE) (Figure 1a) as an electrocatalyst for CO2
ABSTRACT: The catalytic reduction of CO2 is of great
current interest because of its role in climate change and
the energy cycle. We report a pterin electrocatalyst, 6,7-
dimethyl-4-hydroxy-2-mercaptopteridine (PTE), that cata-
lyzes the reduction of CO2 and formic acid on a glassy
carbon electrode. Pterins are natural cofactors for a wide
range of enzymes, functioning as redox mediators and C1
carriers, but they have not been exploited as electro-
catalysts. Bulk electrolysis of a saturated CO2 solution in
the presence of the PTE catalyst produces methanol, as
confirmed by gas chromatography and 13C NMR spec-
troscopy, with a Faradaic efficiency of 10−23%. FTIR
spectroelectrochemistry detected a progression of two-
electron reduction products during bulk electrolysis,
including formate, aqueous formaldehyde, and methanol.
A transient intermediate was also detected by FTIR and
tentatively assigned as a PTE carbamate. The results
demonstrate that PTE catalyzes the reduction of CO2 at
low overpotential and without the involvement of any
metal.
reduction. In water, PTE is predominately in the keto form
(Figure 1a, right structure) as indicated by its FTIR spectrum.
Pterins are natural cofactors for a wide range of enzymes,
functioning as redox mediators and C1 carriers. In its
biologically active form, the pyrazine ring is fully reduced
he catalytic reduction of CO2 is of great current interest.
T
CO2 is a potent greenhouse gas, and its anthropogenic
emission has accelerated as global energy demand has
increased. The current atmospheric CO2 level of over 400
ppm has been implicated as the primary source of climate
forcing, leading to calls for the development of a carbon-neutral
energy cycle.1,2 A sustainable approach to recycling CO2 back
to fuels will require the development of an efficient approach
for its reduction using renewable energy sources such as solar
energy. CO2 is a very stable molecule, however, and its direct
reduction is slow and requires a high overpotential.3,4 The
single-electron reduction of CO2 encounters a large barrier due
to the reorganization energy required to bend the stable linear
molecule to produce the radical anion.5,6
This problem has spurred significant research to develop
electrochemical7−10 or photochemical11−16 catalysts for effi-
cient CO2 reduction. Catalyst development has primarily
focused on metals and metal complexes. Various metals
including Pt, Au, Zn, and Fe have been reported to catalyze
CO2 reduction, primarily to the two-electron reduction
products CO and formate.7 Molecular catalysts based on
polypyridyl complexes of Ru and Re have been developed, with
labile ligand sites to coordinate a CO2 molecule.12,17 All of
these approaches suffer from the need for high overpotentials
to achieve reduction of CO2.
Figure 1. PTE electrocatalysis. (a) Equilibrium structures of PTE. (b)
Cyclic voltammograms of saturated CO2 solutions with (red) and
without (black) PTE catalyst on a glassy carbon electrode. PTE alone
under Ar is shown for comparison (blue). Conditions: 10 mM PTE,
100 mM KCl, 100 mM phosphate buffer (pH 6.3), scan rate of 1 mV/
s. (c) Cyclic voltammograms of 33 mM formic acid with (red) and
without (black) PTE on a glassy carbon electrode in 100 mM KCl
solution at pH 4.6 and 4.2, respectively.
Inspired by the role of pterin in biological catalysis, we have
explored the use of 6,7-dimethyl-4-hydroxy-2-mercaptopter-
Received: August 7, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja5081103 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
(H4-PTE), and C1 substituents are carried at N5 of the
pyrazine ring. Pterin generally acts as a two-electron reductant,
transferring the electrons as hydride.18 Methanogenesis in
archaeal anaerobes depends on pterin cofactors such as
methanopterin (MPT). C1 is loaded onto N5 of the reduced
MPT as a formyl group and then converted to a cyclic
methylene intermediate and finally a methyl group. Critical
steps along this pathway have been shown to occur with the
MPT cofactor alone and no enzyme present, although the
initial steps usually involve a separate C1 carrier.19,20 Thus,
MPT has been shown to condense with formaldehyde to
produce the N5-formyl derivative, which in turn cyclizes to
form the CH2-N5MPT intermediate on the route to forming
methanol or methane. Furthermore, the active (reduced) form
of pterin can be produced electrochemically.21 These character-
istics make pterin a promising candidate as an electrocatalyst
for CO2 reduction.
We observed reduction of CO2 catalyzed by PTE at low
overpotential on a glassy carbon electrode without the
involvement of any metals. Electrochemical measurements
were performed using a Reference 3000 potentiostat (Gamry
Instruments). The working electrode was polished consec-
utively with 1 μm and then 0.05 μm alumina and sonicated
before use. The electrochemistry of PTE on glassy carbon in
buffer solution purged with Ar gas is quasi-reversible with a
two-electron redox couple centered at −0.68 V vs Ag/AgCl, as
shown in Figure 1b (blue curve). In contrast, when the solution
is saturated with CO2, a catalytic wave indicating CO2
reduction is observed (red curve), characterized by significantly
greater current flow and irreversible reduction (Figure 1b).
CO2 reduction does not occur in the absence of PTE (black
curve). A catalytic reduction wave is also observed for formic
acid in the presence of PTE, as shown in Figure 1c. No
reduction of formic acid is observed without the PTE catalyst
(Figure 1c, black trace). These results suggest that PTE can act
as a single, multifunctional catalyst for successive two-electron
reduction steps of CO2, i.e., from CO2 to formic acid and then
to formaldehyde or further reduced products.
The products of CO2 reduction were determined in bulk
electrolysis using a reticulated vitreous carbon (RVC) electrode
as the working electrode. Following bulk electrolysis of a
saturated CO2 solution containing the PTE catalyst in 100 mM
KCl at pH 6.3 (100 mM phosphate buffer) for ∼2 h at −0.65 V
(vs Ag/AgCl), the products were detected by gas chromatog-
raphy and 13C NMR spectroscopy. Both methods demon-
strated that methanol is formed in bulk electrolysis of a
saturated CO2 solution, whereas it is not detected when the
solution is purged with Ar gas or in the absence of the PTE
catalyst (Figures S1 and S2 in the Supporting Information). For
bulk electrolysis of a 13CO2 saturated solution, the 13C NMR
spectrum (Figure S2) shows peaks at 48 and 171 ppm due to
13C-enriched methanol and formate, respectively, demonstrat-
ing that the origin of these products is from CO2 reduction.
The concentration of methanol produced was determined by
GC analysis (Figure S3), which indicated that the Faradaic
efficiency of methanol formation is between 10−23%. Formate
and formaldehyde are the other major reduction products
detected by IR and 13C NMR spectroscopy, but these could not
be quantified by the GC method.
Figure 2. FTIR analysis of bulk electrolysis reactions (scan rate of 1
mV/s, RVC working electrode). (a, b) Difference FTIR spectra
(spectra at indicated potentials minus the spectrum at no applied
potential) of PTE under (a) Ar and (b) saturated CO2 solutions
scanned at 1 mV/s. (c) Simulated spectrum from reference spectra in
D2O. Spectra were acquired on a Varian 3100 FTIR spectrometer (50
μm path length CaF2 cell, 2 cm−1 resolution) using LabVIEW to
synchronize bulk electrolysis and FTIR data collection.
FTIR difference spectra (spectra at indicated potentials minus
the spectrum at no applied potential) of PTE under (a) Ar or
(b) CO2 atmosphere and (c) the simulated spectrum using
reference spectra of the products. Reduction of PTE under Ar
(Figure 2a) leads to a progressive bleach of the ground-state
bands (the unreduced spectrum is shown for comparison) as
the potential is scanned more negative. A concomitant buildup
of positive features is also observed, all shifted to lower
frequency, characteristic of pterin reduction.22 It should be
noted that the ketone absorbance at 1690 cm−1 is only slightly
shifted upon reduction (to 1679 cm−1), whereas the pyrazine
CN mode at 1512 cm−1 disappears. This result is consistent
with a two-electron reduction of the pyrazine ring, as expected
from previous electrochemical studies.21 When the same
experiment is repeated with a saturated CO2 solution (Figure
2b), as the potential becomes more negative a progression of
new peaks due to the reduction of CO2 is observed. At −0.40
V, the bleach features due to reduction of PTE are present, but
the corresponding positive peaks are missing, indicating that
the reduced PTE reacts rapidly with CO2 and is not detected
on the 1−2 min time scale of the FTIR experiment. We
simulated the IR difference spectra using a linear combination
of reference spectra of the products in D2O and subtracting the
initial PTE spectrum, as shown in Figure 2c. The simulated
spectrum at −0.75 V gives the best fit with a formate:formal-
dehyde:methanol product ratio of 2:1.5:1. All of the major
The progress of the electrocatalytic reaction was followed
using FTIR spectroelectrochemistry by flowing the sample
solution from the electrochemical cell into an infrared
transmission cell during bulk electrolysis. Figure 2 shows the
B
dx.doi.org/10.1021/ja5081103 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
absorbances of formate (1350 and 1370 cm−1), formaldehyde
(1442 cm−1), and methanol (1466 cm−1) are well-fit by this
procedure. The simulation does not account for several peaks,
the most prominent at 1625 cm−1, that are present throughout
the reaction. We postulate that these features are due to a PTE-
carbamate intermediate formed by addition of CO2 to one of
the nitrogens of the reduced pyrazine ring, by analogy to the IR
spectra of known carbamates.23 Buildup of the reduced PTE is
also observed at the lowest potential (−0.9 V), as evidenced by
the appearance of the positive peak at 1679 cm−1 (the ketone
absorbance of two-electron-reduced PTE). Reduced PTE may
accumulate at this potential as a result of faster catalysis and
depletion of CO2 in the solution.
Comparison of the present results with previous work using
pyridinium in water to catalyze the reduction of CO2 to
methanol provides additional mechanistic insight.24−26 The
Bocarsly group observed CO2 reduction to methanol on a Pt
electrode with low overpotential.25 They proposed a multistep
mechanism in which pyridinium works as a one-electron
shuttle, with the initial reduction of pyridinium to form the
pyridinyl radical anion followed by reaction with CO2 to form
the carbamate adduct. This mechanism was questioned by
another study that observed no methanol production, only
competitive proton reduction.26 Theoretical work also noted
several problems, including a pKa of ∼27 for the deprotonation
of the pyridinyl radical in water, presenting a high energy
barrier to the formation of the carbamate intermediate,27,28 and
a discrepancy between the calculated and observed redox
potentials of the pyridinium/pyridinyl couple.28−30 Alternative
mechanisms have been proposed that involve surface
interactions of pyridinium with the Pt electrode29 or a
surface-bound dihydropyridine species.31,32 Finally, the Mac-
Donnell group reported that pyridinium catalyzes CO2
reduction in concert with a Ru or Re photosensitizer without
the presence of a metal electrode.16,24 Therefore, the role of the
metal electrode in these reactions remains unclear.
where competitive reduction of protons is not an issue. The
observed mechanism is consistent with the known ability of
reduced pterins to catalyze two-electron reductions via hydride
transfer to C1 substrates.
In summary, PTE acts as a molecular electrocatalyst for CO2
reduction on a glassy carbon electrode. Gas chromatography
and NMR and FTIR spectroscopy provide evidence of two-
electron reduction of the PTE catalyst and reaction with CO2
to form an intermediate species that is likely a PTE-carbamate,
followed by successive two-electron reductions to formate,
formaldehyde, and methanol. The efficiency of methanol
production is modest, but it is worth noting that the PTE
catalyst lacks the extended substituent at position 6 of the
pyrazine ring normally present in methanogenic cofactors such
as MPT.19 Consequently, PTE cannot stabilize a cyclic
methylene intermediate, which may limit the efficiency of
further reduction steps beyond the first two-electron reduction
to formate. We are currently exploring pterin derivatives that
better mimic MPT as a means to improve the overall efficiency.
ASSOCIATED CONTENT
■
S
* Supporting Information
GC and 13C NMR product analysis (Figures S1−S3), FTIR
reference spectra (Figure S4), and cyclic voltammogram
comparisons (Figure S5). This material is available free of
charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
briandyer@emory.edu
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the NSF through Grant DMR-
1409851 (R.B.D.).
We conclude that the mechanism of PTE catalysis is
fundamentally different from that of pyridinium. Clearly no
metal is involved in the reduction reactions on glassy carbon.
The detection of two-electron-reduced PTE and a transient
intermediate formed with CO2 (likely a carbamate) indicates
that the catalyst can act as a two-electron reductant and C1
carrier. The only products detected are formate, formaldehyde,
and methanol, consistent with successive two-electron reduc-
tion steps. The difference in mechanism is due in part to
different electrochemical behavior of the catalysts, which is
evident from a comparison of the cyclic voltammograms of
pyridinium and PTE on platinum, gold, and glassy carbon
electrodes (Figure S5). Pyridinium shows a quasi-reversible
redox reaction on a Pt electrode under Ar, but it does not show
an oxidation wave on gold or any redox reactions on a glassy
carbon electrode. Pt has a significantly lower hydrogen
overpotential than the other electrodes, and therefore, the
reduction of protons is competitive with the reduction of the
pyridinium catalyst. This results in ample hydrogen atoms
adsorbed on the electrode surface that can react with surface
species, which may play a role in pyridinium catalysis on Pt.
Glassy carbon has a high overpotential for proton reduction,
and no reaction is observed with pyridinium. In contrast, the
PTE catalyst shows quasi-reversible redox chemistry under Ar
on all three of these electrodes, indicating that its redox
chemistry is not necessarily coupled to proton reduction.
Furthermore, it reduces CO2 on glassy carbon under conditions
REFERENCES
■
(1) Hansen, J.; Sato, M.; Ruedy, R.; Nazarenko, L.; Lacis, A.;
Schmidt, G. A.; Russell, G.; Aleinov, I.; Bauer, M.; Bauer, S.; Bell, N.;
Cairns, B.; Canuto, V.; Chandler, M.; Cheng, Y.; Del Genio, A.;
Faluvegi, G.; Fleming, E.; Friend, A.; Hall, T.; Jackman, C.; Kelley, M.;
Kiang, N.; Koch, D.; Lean, J.; Lerner, J.; Lo, K.; Menon, S.; Miller, R.;
Minnis, P.; Novakov, T.; Oinas, V.; Perlwitz, J.; Perlwitz, J.; Rind, D.;
Romanou, A.; Shindell, D.; Stone, P.; Sun, S.; Tausnev, N.; Thresher,
D.; Wielicki, B.; Wong, T.; Yao, M.; Zhang, S. J. Geophys. Res.: Atmos.
2005, 110, No. D18104.
(2) Hansen, J.; Sato, M.; Russell, G.; Kharecha, P. Philos. Trans. R.
Soc., A 2013, 371, No. 20120294.
(3) Jitaru, M.; Lowy, D. A.; Toma, M.; Toma, B. C.; Oniciu, L. J.
Appl. Electrochem. 1997, 27, 875−889.
(4) Kumar, B.; Llorente, M.; Froehlich, J.; Dang, T.; Sathrum, A.;
Kubiak, C. P. Annu. Rev. Phys. Chem. 2012, 63, 541−569.
(5) Kafafi, Z. H.; Hauge, R. H.; Billups, W. E.; Margrave, J. L. Inorg.
Chem. 1984, 23, 177−183.
(6) Yoshioka, Y.; Jordan, K. D. Chem. Phys. Lett. 1981, 84, 370−374.
(7) Hori, Y. Mod. Aspects Electrochem. 2008, 42, 89−189.
(8) Kondratenko, E. V.; Mul, G.; Baltrusaitis, J.; Larrazabal, G. O.;
Perez-Ramirez, J. Energy Environ. Sci. 2013, 6, 3112−3135.
(9) Costentin, C.; Robert, M.; Savea
42, 2423−2436.
́
nt, J.-M. Chem. Soc. Rev. 2013,
(10) Oh, Y.; Hu, X. Chem. Soc. Rev. 2013, 42, 2253−2261.
(11) Morris, A. J.; Meyer, G. J.; Fujita, E. Acc. Chem. Res. 2009, 42,
1983−1994.
C
dx.doi.org/10.1021/ja5081103 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
(12) Benson, E. E.; Kubiak, C. P.; Sathrum, A. J.; Smieja, J. M. Chem.
Soc. Rev. 2009, 38, 89−99.
(13) Ikeue, K.; Yamashita, H.; Anpo, M.; Takewaki, T. J. Phys. Chem.
B 2001, 105, 8350−8355.
(14) Dhakshinamoorthy, A.; Navalon, S.; Corma, A.; Garcia, H.
Energy Environ. Sci. 2012, 5, 9217−9233.
(15) Mao, J.; Li, K.; Peng, T. Y. Catal. Sci. Technol. 2013, 3, 2481−
2498.
́
(16) Boston, D. J.; Pachon, Y. M. F.; Lezna, R. O.; de Tacconi, N. R.;
MacDonnell, F. M. Inorg. Chem. 2014, 53, 6544−6553.
(17) Dubois, M. R.; Dubois, D. L. Acc. Chem. Res. 2009, 42, 1974−
1982.
(18) Keltjens, J. T.; Vanderdrift, C. FEMS Microbiol. Lett. 1986, 39,
259−303.
(19) Dimarco, A. A.; Bobik, T. A.; Wolfe, R. S. Annu. Rev. Biochem.
1990, 59, 355−394.
(20) Maden, B. E. H. Biochem. J. 2000, 350, 609−629.
(21) Dryhurst, G.; Raghavan, R.; Egeserpkenci, D.; Karber, L. G. Adv.
Chem. Ser. 1982, 201, 457−487.
(22) Moore, J.; Wood, J. M.; Schallreuter, K. U. Biochemistry 1999,
38, 15317−15324.
(23) Hisatsune, C. Can. J. Chem. 1984, 62, 945−948.
(24) Boston, D. J.; Xu, C. D.; Armstrong, D. W.; MacDonnell, F. M.
J. Am. Chem. Soc. 2013, 135, 16252−16255.
(25) Cole, E. B.; Lakkaraju, P. S.; Rampulla, D. M.; Morris, A. J.;
Abelev, E.; Bocarsly, A. B. J. Am. Chem. Soc. 2010, 132, 11539−11551.
́
(26) Costentin, C.; Canales, J. C.; Haddou, B.; Saveant, J.-M. J. Am.
Chem. Soc. 2013, 135, 17671−17674.
(27) Keith, J. A.; Carter, E. A. J. Am. Chem. Soc. 2012, 134, 7580−
7583.
(28) Lim, C. H.; Holder, A. M.; Musgrave, C. B. J. Am. Chem. Soc.
2013, 135, 142−154.
(29) Ertem, M. Z.; Konezny, S. J.; Araujo, C. M.; Batista, V. S. J. Phys.
Chem. Lett. 2013, 4, 745−748.
(30) Yan, Y.; Gu, J.; Bocarsly, A. B. Aerosol Air Qual. Res. 2014, 14,
515−521.
(31) Keith, J. A.; Carter, E. A. Chem. Sci. 2013, 4, 1490−1496.
(32) Keith, J. A.; Carter, E. A. J. Phys. Chem. Lett. 2013, 4, 4058−
4063.
D
dx.doi.org/10.1021/ja5081103 | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX