Redox-active cyclopentadienyl Ni complexes with quinoid N-heterocyclic carbene ligands for the electrocatalytic hydrogen release from chemical fuels

Article abstract of DOI:10.1039/c3nj00276d

We now report the electrocatalytic dehydrogenation of tetrahydroquinaldine by an electron-rich CpNi N-heterocyclic carbene (NHC) with quinoid ligand motifs and explore the effects of quinone additives on CpNi compounds without quinoid NHC ligands. Our CpNi(NHC) catalyst exhibits dehydrogenative electrocatalytic activity and demonstrates that a molecular catalyst precursor can be viable in the electrode-driven (H⁺ + e^-) release step of "virtual hydrogen storage".

Full text of DOI:10.1039/c3nj00276d

NJC
View Article Online
View Journal
LETTER
Redox-active cyclopentadienyl Ni complexes with
quinoid N-heterocyclic carbene ligands for the
Cite this: DOI: 10.1039/c3nj00276d
electrocatalytic hydrogen release from chemical fuels†
Received (in Porto Alegre, Brazil)
15th March 2013,
Oana R. Luca, Daria L. Huang, Michael K. Takase and Robert H. Crabtree*
Accepted 8th April 2013
DOI: 10.1039/c3nj00276d
www.rsc.org/njc
We now report the electrocatalytic dehydrogenation of tetra- capable of 1eꢀ chemical steps, incorporation of redox-active
hydroquinaldine by an electron-rich CpNi N-heterocyclic carbene ligands has been proposed.7
(NHC) with quinoid ligand motifs and explore the effects of
quinone additives on CpNi compounds without quinoid NHC
ligands. Our CpNi(NHC) catalyst exhibits dehydrogenative electro-
(1)
catalytic activity and demonstrates that a molecular catalyst pre-
cursor can be viable in the electrode-driven (H+ + eꢀ) release step of
‘‘virtual hydrogen storage’’.
Quinones have been of interest in electrochemical dehydro-
genation chemistry as they readily accept (H+ + eꢀ) equivalents
With the rising economic and environmental costs of natural
and can be regenerated at an electrode.4 They are also effective
gas and oil, much effort has been devoted to developing
terminal H2 acceptors for Pd catalyzed dehydrogenation
reactions.8,9 We have previously shown that high potential
quinones such as 2,3-dichloro-5,6-dicyanoquinone (DDQ) are
electrocatalysts for desaturation of C–N bonds, although having
a limited substrate scope.4 We believed that tethering quinones
as (H+ + eꢀ) acceptor motifs to electron-rich catalyst frame-
works might impart reactivity towards dehydrogenation chemistry
(Fig. 1).
The known reaction of nickelocene with azolium salts has
been previously used as a versatile synthetic methodology
leading to electron-rich CpNi(NHC) compounds.10–13 Bielawski
and coworkers14 have previously reported the synthesis of
hydrogen fuel storage as an alternative, low-CO2 fuel source.
However, storage and transportation methods of free H2 face
many practical problems such as diffusion through metals and
high flammability.1 Easily storable liquid hydrogen carriers
that can release H2 equivalents as protons and electrons,
therefore offer an attractive solution. The fuel can be thought
of as ‘‘virtual hydrogen storage’’1 material that can be
recycled by reversible electrohydrogenation–dehydrogenation.2
The current infrastructure developed for the handling of
liquid petroleum products may also be amenable to this
technology.2,3
Only one electrode-driven electrochemical process of desatu-
ration has been reported with a molecular catalyst precursor to
date.4 Even non-electrochemical hydrogenation–dehydrogenation
catalytic systems for heterocycles are rare and require costly
late transition metals like ruthenium5 or iridium.6 Research
efforts have thus been focused on synthesizing catalysts with
abundant first-row transition metals. In order to impart multi-
electron reactivity to such base metal compounds, normally only
Department of Chemistry, Yale University 225 Prospect St., New Haven,
CT 06520-8107, USA. E-mail: robert.crabtree@yale.edu
† Electronic supplementary information (ESI) available: Additional CVs, controls,
experimental details and X-ray information. CCDC 915452. For ESI and crystallo-
graphic data in CIF or other electronic format see DOI: 10.1039/c3nj00276d
Fig. 1 ORTEP representation of the X-ray structure of compound 1 (40%
probability ellipsoids). H atoms were omitted for clarity.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem.

View Article Online
Letter
NJC
quinone-annulated NHCs and the corresponding Rh and Ag
complexes. We have now synthesized the quinone NHC
compound 1 by the direct reaction of nickelocene with the
imidazolium-annulated quinoid salt (QIm in eqn (1)). The new
compound 1 was also successfully crystallized by slow evapora-
tion from dichloromethane and the X-ray structure determined
(see Section S2 (ESI†) for more details). The crystal structure
shows very similar characteristics when compared to related
CpNi NHC compounds13 with similar ‘‘piano-stool’’ geometry.
The diene distortion (the presence of two equally shortened
non-adjacent C–C bonds) was also observed in the structure of
compound 1.
The cyclic voltammogram of 1 reveals two quasi-reversible
oxidations the first of which at 0.78 V vs. NHE, we tentatively
assign to a metal-centered oxidation (Fig. 2). The second signal
at 1.35 V vs. NHE may be oxidation of Clꢀ, although a definite
assignment requires further investigation. A quasi-reversible
quinone-centered reduction response is observed at ꢀ0.31 V vs.
NHE, followed by another reduction at ꢀ1.1 V vs. NHE (Fig. 2)
which we believe is associated with the metal center.
Fig. 3 Cyclic voltammetry of 1-catalytic run: 2 mM 1 in a 2 mL solution of
0.1 M NBu4BF4 in acetonitrile at a Pt electrode with increments of added
tetrahydroquinaldine.
Redox-active 1 has now been shown to act as an operation- N,N-dimethylbenzimidazolium chloride (2, Fig. S3-2 (ESI†) for
ally homogeneous dehydrogenation catalyst for the dehydro- CVs). Compound 2 was prepared as previously reported via the
genation of the saturated N-heterocyclic substrate: 1,2,3,4- method depicted in eqn (1).13 The results of our electrocatalytic
tetrahydroquinaldine to give the corresponding arene. Cyclic studies with 2 are summarized in Table 1, entries b–d.
voltammograms of compound 1 in the presence of incremental
The utility of quinones as ligand components in metal
additions of this substrate show a sharp increase in current catalysts has been limited by difficulty of synthesis.16 Moreover,
response, characteristic of a catalytic process (Fig. 3). As recent studies have explored the nature of metal–quinone
derivatives of anilines often show polyaniline-forming side- interactions in catalytic processes.17 We therefore looked at the
reactions,15 the current responses had to be correlated with possibility of adding a free quinone to a CpNiNHC catalyst that
isolated yields of the desired fully desaturated product (eqn (2)) lacked a tethered quinone ligand. The results shown in entries c
to show efficient and selective charge consumption (details in and d of Table 1 indicate that the free quinone is indeed capable
the ESI†). The reaction was shown to occur in 70–90% faradaic of activating the catalysts for the desaturation process without
and chemical yield (Table 1) at a Pt surface, with the largest needing to be incorporated in the ligand. The result with 2 and
chemical yield of 78% obtained with 1. We also investigated no added quinone (entry b) suggests the quinone is necessary for
the role of a free quinone additive on an otherwise electro- the dehydrogenation process to occur in quantifiable yields, but
catalytically unreactive non-quinoid CpNi NHC derived from it does not necessarily have to be attached to the metal catalyst.
Moreover, the addition of 10 equivalents of benzoquinone
(related to the 0.5 mol% catalyst loading) raises the product
yield by 19% from the catalytic run with only one equivalent
(Table 1, entry c) of benzoquinone (Table 1, entry d).
Electrode materials other than Pt also proved suitable under
the conditions, albeit with lower yields: Reticulated Vitreous
Carbon (RVC)—55% and Stainless Steel 60 (SS60)—30%. Plots
of charge vs. time after 4 h of electrolysis at 1 V vs. NHE with the
three electrode materials are shown in Fig. 4. and the product
yields in Table 2.
Ni/Pt has previously been shown to be a viable catalyst
option for a fuel cell operating with an organic fuel.18 In our
own work, controls with suspended NiO and metallic Ni
powders suspended in solution have shown significant current
consumption; however modest product (40 and 30% respec-
tively) and faradaic yields (53 and 34% respectively) were
observed (Table S2, ESI†). This is indicative of parasitic non-
productive processes occurring under these conditions, likely
associated with substrate side-reactivity due to polyaniline
formation, as has been previously observed.15
Fig. 2 Cyclic voltammetry of 1: 2 mM 1 in a 0.1 M NBu4BF4 solution at a Pt
working electrode at 100 mV sꢀ1
.
c
New J. Chem.
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013

View Article Online
NJC
Letter
Table 1 Chemical and faradaic yields for tetrahydroquinaldine dehydrogena-
tion experiments at a Pt electrode 0.5 mol% catalysts 1 and 2
Table 2 Chemical and faradaic yields for tetrahydroquinaldine dehydrogena-
tion at a Pt electrode (entry 1), SS60 (2) and RVC (3) by 1 (0.5%). Plots of current
vs. time shown in Fig. 4
Electrode
material
Chemical
yield (%)
Faradaic
efficiency (%)
Entry
1
2
3
Pt
SS60
RVC
78
30
55
80
86
95
Chemical
yielda (%)
Faradaic
Entry
a
Catalyst
yieldb (%)
Conclusions
78
80
Dehydrogenation of 1,2,3,4-tetrahydroquinaldine is not only
possible with quinone catalyst 1, but also with free quinone
added to the non-quinoid catalyst 2 (entries c and d in Table 1)
that in the absence of the quinone is inactive. Our CpNi(NHC)
catalyst is one of the first examples of first-row transition metal
complexes exhibiting dehydrogenative electrocatalytic activity
at room temperature and shows that a molecular catalyst
precursor can be viable in the electrode-driven H2 release step
of ‘‘virtual hydrogen storage’’.1
b
N/A
N/A
c
51
68
71
90
Experimental
Synthesis and characterization of compound 1
d
To a 20 mL solution of 1.1 equivalents of nickelocene (41 mg,
0.22 mmol) in anhydrous THF, the yellow quinone-annulated
dimesitylimidazolium chloride ImQ (0.2 mmol) was added as a
solid as shown in eqn (1). The resulting suspension was then
refluxed for 4 hours. 1 was isolated as a red solid in 69% yield
by column chromatography in 4 : 1 hexanes–ethylacetate as a
brown-red solid. FT-ICR MS analysis was performed at the Yale
Keck Proteomics facility on a 9.4 T Bruker Qe FT-ICR MS.
Elemental Analysis was performed by Robertson Microlit.
1H NMR (500 MHz, CD2Cl2) d 8.01 (s, 2H, ArCH), 7.74 (s, 2H,
ArCH), 7.20 (s, 4H, MesCH), 4.58 (s, 5H, CpCH), 2.50 (s, 6H,
e
f
None
0
0
0
0
5 mol% benzoquinone
a
b
Isolated yields. Determined from the ratio of current used for
current used for efficient product formation vs. total passed current.
Mesp-CH3), 2.08 (s, 12H, Meso,o -CH3). 13C{1H} (126 MHz, CD2Cl2)
0
d 141.19, 135.85, 135.35, 133.00, 130.66, 128.25, 94.34, 22.46,
19.66. FT ICR MS: [M-Cl] calculated 557.1734, found 557.1721.
Elemental analysis. Expected: C: 68.77%, H: 5.26%, N: 4.72%
Found: C: 68.59%, H: 5.23%; N: 4.44%.
Acknowledgements
We thank Daryl Smith for suggestions for our electrochemical
cell design. The work was supported as part of the Center for
Electrocatalysis, Transport Phenomena, and Materials (CETM)
for Innovative Energy Storage, an Energy Frontier Research
Center funded by the U.S. Department of Energy, Office of
Basic Energy Sciences under Award Number DE-SC00001055.
We thank Ms. Julie Thomsen for useful discussions.
Fig. 4 Plots of charge passed vs. time during 4 h of electrolysis at 1 V vs. NHE
with the three different electrode materials. Background runs with just the model
fuel (200 mL 1,2,3,4-tetrahydroquinaldine) in the absence of catalyst are shown in
short dash-dot—SS60, dash-dot—RVC, dot-dot—Pt. Catalytic runs were per-
formed at 0.5 mol% 1. The top dashed gray line indicates the level of charge
expected to pass for 100% efficient current conversion for the reaction depicted
in eqn (2). Background current consumptions observed in the absence of catalyst
1 (at the bare electrode) in the presence of substrate does not correlate with
formation of desired product and are depicted by dashed lines. Benzoquinone
was unable to mediate the heterocycle desaturation in the absence of 2 at
5 mol% (entry f in Table 1).
Notes and references
1 R. H. Crabtree, Energy Environ. Sci., 2008, 1, 134.
2 R. Crabtree and O. Luca, WO Pat. 2,012,112,758, 2012.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem.

View Article Online
Letter
NJC
3 P. Driscoll, E. Deunf, L. Rubin, O. Luca, R. Crabtree, 11 V. Ritleng, E. Brenner and M. J. Chetcuti, J. Chem. Educ.,
C. Chidsey, J. Arnold and J. Kerr, ECS Trans., 2011, 35, 3. 2008, 85, 1646.
4 O. R. Luca, T. Wang, S. J. Konezny, V. S. Batista and 12 V. Ritleng, A. M. Oertel and M. J. Chetcuti, Dalton Trans.,
R. H. Crabtree, New J. Chem., 2011, 35, 998. 2010, 39, 8153.
5 G. P. Pez, A. R. Scott, A. C. Cooper, H. Cheng, F. C. Wilhelm 13 O. R. Luca, B. A. Thompson, M. K. Takase and R. H. Crabtree,
and A. H. Abdourazak, US Pat., 7351395, 2008. J. Organomet. Chem., DOI: 10.1016/j.jorganchem.2012.10.038.
6 Z. Wang, J. Belli and C. M. Jensen, Faraday Discuss., 2011, 14 M. D. Sanderson, J. W. Kamplain and C. W. Bielawski, J. Am.
151, 297. Chem. Soc., 2006, 128, 16514.
7 O. R. Luca and R. H. Crabtree, Chem. Soc. Rev., 2013, 15 S. Mu and J. Kan, Electrochim. Acta, 1996, 41, 1593.
˜
42, 1440.
16 L. A. Berben, D. C. Craig, C. Gimbert-Surinach, A. Robinson,
8 B. V. Popp, J. L. Thorman and S. S. Stahl, J. Mol. Catal. A:
Chem., 2006, 251, 2.
K. H. Sugiyarto and S. B. Colbran, Inorg. Chim. Acta, 2011,
370, 374.
9 B. V. Popp and S. S. Stahl, in Organometallic Oxidation, 17 Z. Liu, R. J. Deeth, J. S. Butler, A. Habtemariam,
catalysis, ed. F. Meyer and C. Limberg, Springer, New York,
2007.
M. E. Newton and P. J. Sadler, Angew. Chem., Int. Ed.,
2013, 52, 1.
´
´
10 R. A. Kelly III, N. M. Scott, S. Dıez-Gonzalez, E. D. Stevens 18 G. L. L. Soloveichik, J. P. Zhao and I. Ji-Cheng, US Pat.,
and S. P. Nolan, Organometallics, 2005, 24, 3442. 20080248339, 2008.
c
New J. Chem.
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013