Organometallics
Communication
course of the 1 h electrolysis. The total charge passed is
approximately 15% less than that predicted for the 2-electron
1-proton reduction of [2][OTf], which might be attributed to
the partial decomposition of [2][OTf] under the reaction
conditions.
In summary, we report the electrohydrogenation of
benzaldehyde with a Shvo-type Mo complex (C5Ph4OH)Mo-
(CO)3(H) 3. Controlled potential electrolysis of benzaldehyde
at −0.90 V (vs Fc+/0) in ortho-difluorobenzene with the acid
[DMFH][OTf] affords a mixture of the reduction products
benzyl alcohol (BnOH) and dibenzyl ether (Bn2O) with a
Faradaic efficiency of approximately 67%. Stoichiometric
experiments reveal that the reductive protonation of either
the Mo(0) complex [(C5Ph4O)Mo(CO)3(L)] or the cationic
Mo(II) complex [(C5Ph4OH)Mo(CO)3(CH3CN)][OTf] af-
fords the hydride (C5Ph4OH)Mo(CO)3(H) 3, which was
previously shown to be competent to reduce aldehydes. These
studies provide support for an electrocatalytic mechanism
involving the rate-limiting reduction of the aldehyde by the
Mo−H 3 and subsequent reductive protonation of [(C5Ph4O)-
Mo(CO)3(L)] to regenerate the Mo−H 3. The high Faradaic
efficiencies observed with these metal−ligand bifunctional
catalysts provide new design motifs for the design of selective
electrocatalysts for electrohydrogenation relative to hydrogen
evolution.
As stoichiometric experiments had shown that (C5Ph4OH)-
Mo(CO)3(H) 3 can be generated electrochemically from
[(C5Ph4OH)Mo(CO)3(CH3CN)][OTf] ([2][OTf]), and
previous studies had shown that 3 can reduce benzaldehyde,24
the electrocatalytic reduction of aldehydes was investigated
with [2][OTf]. Initial experiments by cyclic voltammetry were
uninformative, as titration of benzaldehyde into a [2][OTf]/
[DMFH][OTf] solution, even at very low scan rates (i.e., 25
mV/s), did not show any evidence of an enhanced current at
−1000 mV. This was not unexpected, as previous studies24 had
shown that catalytic reduction of aldehydes by (C5Ph4OH)-
Mo(CO)3(H) 3 only occurs over the course of hours at 65 °C
(99% conversion after 12 h, 65 °C).
To investigate the electrocatalytic reduction of aldehydes,
controlled-potential electrolysis of an ODFB solution of 1 mM
[2][OTf], 100 mM [DMFH][OTf], and 100 mM benzalde-
hyde at −900 mV (vs Fc0/+) were carried out for 2 h. Analysis
of the resulting solution by 1H NMR (see SI) after 2 h revealed
the formation of benzyl alcohol (BnOH) and dibenzyl ether
(Bn2O), with turnover number of TON = 4.4 and a Faradaic
efficiency of approximately 67% (averaged from two runs).
The results of this experiment are summarized in the SI, Table
showed very low yield (0.3%) and Faradaic efficiency (1.7%),
confirming the catalytic role of Mo. Similarly, an experiment
with no applied potential showed no conversion (SI, Figure
S24), indicating the requirement of an electrochemical step in
the catalytic cycle.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
■
sı
General experimental methods, syntheses, additional
cyclic voltammograms, quantification procedures and
calculations, procedures for controlled potential elec-
trolysis studies, control experiments, and additional
nuclear magnetic resonance studies (PDF)
AUTHOR INFORMATION
Corresponding Author
■
The stoichiometric and catalytic experiments are consistent
with the mechanism outlined in Figure 3. In the presence of
[DMFH][OTf], [2][OTf] is converted to the Mo−H 3 by a
two-electron-one-proton reduction at the working electrode.
The slow reduction of benzaldehyde by Mo−H 3 affords
benzyl alcohol and the Mo(0) tetraphenylcyclopentadienone
complex.24 Competitive acid-catalyzed dehydration of benzyl
alcohol generates dibenzyl ether. Protonation of the Mo(0)
tetraphenylcyclopentadienone regenerates [2][OTf].
Robert M. Waymouth − Department of Chemistry, Stanford
University, Stanford, California 94305, United States;
Author
Keith C. Armstrong − Department of Chemistry, Stanford
University, Stanford, California 94305, United States;
In light of previous studies showing that the rate of aldehyde
reduction is slow at room temperatures, it is likely that the
chemical step between benzaldehyde and 3 is turnover-
limiting. The ca. 67% Faradaic efficiencies observed are
noteworthy in that they imply that the Mo−H 3, generated
electrochemically in the presence of benzaldehyde, exhibits a
selectivity for reduction of the aldehyde over protonation to
release H2 (hydrogen evolution reaction, HER). The origin of
this selectivity is not clear, but has some precedent in the
chemistry of ionic hydrogenation28 with related Mo piano
stool complexes. The related Cp’Mo(H)(CO)3 (Cp′ =
cyclopentadiene, pentamethylcyclopentadiene, pentabenzylcy-
clopentadiene29) complexes28−31 were observed to react with
triflic acid in CD2Cl2 to generate H2 and Cp’Mo(OTf)-
(CO)3,30 but when protonation of either CpMo(CO)2(PPh3)
H (CH3CN) or CpMo(CO)2(PPh3)H (CH2Cl2) was carried
out in the presence of a ketone (acetone or 3-pentanone),
competitive reduction to the alcohol and hydrogen evolution
were observed.29,31 The bifunctional nature of the hydrox-
ycyclopentadienyl ligand might also contribute to the observed
selectivity for electrohydrogenation (Figure 1).
Complete contact information is available at:
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This material is based on work supported by the National
Science Foundation (CHE-1565947, initial studies) and the
Department of Energy (DE-SC0018168). K.C.A. is grateful to
Dr. Stephen Lynch at the Stanford University NMR Facility for
valuable technical discussions and expertise. K.C.A. thanks
Daniel Marron, Dr. Katherine Walker, and the research group
of Prof. Richard N. Zare for assistance with mass spectrometry
instrumentation.
REFERENCES
■
(1) Garedew, M.; Lin, F.; Song, B.; DeWinter, T. M.; Jackson, J. E.;
D
Organometallics XXXX, XXX, XXX−XXX