The isolation of [Pd{OC(O)H}(H)(NHC)(PR3)] (NHC = N-heterocyclic carbene) and its role in alkene and alkyne reductions using formic acid

Article abstract of DOI:10.1021/ja311087c

The [Pd(SIPr)(PCy₃)] complex efficiently promotes a tandem process involving dehydrogenation of formic acid and hydrogenation of C-C multiple bonds using H₂ formed in situ. The isolation of a key catalytic hydridoformatopalladium species, [Pd{OC(O)H}(H)(IPr)(PCy ₃)], is reported. The complex plays a key role in the Pd(0)-mediated formation of hydrogen from formic acid. Mechanistic and computational studies delineate the operational role of the palladium complex in this efficient tandem sequence.

Full text of DOI:10.1021/ja311087c

Communication
pubs.acs.org/JACS
The Isolation of [Pd{OC(O)H}(H)(NHC)(PR3)] (NHC = N‑Heterocyclic
Carbene) and Its Role in Alkene and Alkyne Reductions Using Formic
Acid
Julie Broggi,† Vaclav Jurcík,† Olivier Songis,† Albert Poater,‡ Luigi Cavallo,§ Alexandra M. Z. Slawin,†
́
̌
and Catherine S. J. Cazin*,†
†EaStCHEM School of Chemistry, University of St Andrews, St Andrews KY16 9ST, U.K.
‡Institut de Química Computacional, Departament de Química, Universitat de Girona, Campus de Montilivi, E-17071 Girona, Spain
§KAUST Catalyst Research Center, Physical Sciences and Engineering Division, King Abdullah University of Science and
Technology, Thuwal 23955-6900, Kingdom of Saudi Arabia
S
* Supporting Information
Scheme 1. Oxidative Addition of H2 to a Pd(0) Center10
ABSTRACT: The [Pd(SIPr)(PCy3)] complex efficiently
promotes a tandem process involving dehydrogenation of
formic acid and hydrogenation of C−C multiple bonds
using H2 formed in situ. The isolation of a key catalytic
hydridoformatopalladium species, [Pd{OC(O)H}(H)-
(IPr)(PCy3)], is reported. The complex plays a key role
in the Pd(0)-mediated formation of hydrogen from formic
acid. Mechanistic and computational studies delineate the
operational role of the palladium complex in this efficient
tandem sequence.
and its saturated-NHC-bearing relative [Pd(SIPr)(PCy3)] (3)
[SIPr = N,N′-bis(2,6-diisopropylphenyl)imidazolidin-2-yli-
dene] as promising catalysts enabling the hydrogenation of
C−C multiple bonds using formic acid as the hydrogen source.
To ascertain the exact role of formic acid in the initial steps
of the reported “transfer hydrogenation” process,8 1 was first
reacted with formic acid in a J-Young NMR tube at low
temperature (Figure 1).11 At −70 °C, conversion of 1 into a
hydride-bearing species, 4 [δP = −43 ppm, δH = −18.88 ppm
(2JH−P = 4.5 Hz)], was observed. Increasing the temperature led
he catalytic hydrogenation of C−C multiple bonds is one
T
of the most important transformations in organic
chemistry.1 Such reactions are usually carried out using
molecular hydrogen, with two major drawbacks: H2 is highly
flammable and readily forms explosive mixtures with air,2 and
the stoichiometry of H2 is difficult to control, oftentimes
leading (when this is possible) to substrate over-reduction.
Catalytic transfer hydrogenation represents a viable alternative
to the more classical reduction method using molecular
hydrogen or metal hydrides.3 Among various possible hydrogen
sources, alcohols,3d,4 water,5 formic acid,6 and alkylammonium
formates7 have found the broadest applications. A Pd(OAc)2/
PtBu3 system has been reported to catalyze the reduction of
alkenes using formic acid.8 The transfer hydrogenation was
presumed to occur through the formation of a undefined
hydridoformatopalladium species followed by migratory
insertion of the substrate into the Pd−H bond. No trace of
hydrogen formation was reported.
We recently described the remarkable activity of [Pd(NHC)-
(PCy3)] complexes (NHC = N-heterocyclic carbene) in alkene
hydrogenation under a hydrogen atmosphere.9 Further
investigation using a congener permitted the first isolation of
monomeric palladium dihydride species, demonstrating the
feasibility of direct activation of H2 by a Pd(0) complex
(Scheme 1).10
Figure 1. Variable-temperature 31P{1H} NMR experiments on
[Pd(IPr)(PCy3)] (1) in the presence of HCOOH (1.05 equiv) in
THF-d8. Peaks labeled with * are due to a minor impurity (see the SI).
The catalytic transfer hydrogenation,8 hydrogen activation,10
and catalytic hydrogenation properties9 displayed by Pd(0)−
NHC complexes encouraged us to examine [Pd(IPr)(PCy3)]
(1) [IPr = N,N′-bis(2,6-diisopropylphenyl)imidazol-2-ylidene]
Received: November 11, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja311087c | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
to the disappearance of signals associated with formic acid and
the formation of the new complex 4, which gradually reverted
back to 1. Surprisingly, release of H2 was observed.12 At 60 °C,
the reaction concluded with complete conversion of the formic
acid and quantitative recovery of complex 1 [other minor
compounds observed during the reaction were attributed to
side products generated by traces of water; see the Supporting
Information (SI)]. Interestingly, under these conditions the
formation of the trans-dihydride complex 2 was never observed
spectroscopically.10 Labeling studies13 with selectively deuter-
ated formic acid (DCOOH and HCOOD) allowed the origin
of the Pd−H proton to be assigned as the acidic H of formic
acid, suggesting the composition [Pd{OC(O)H}(H)(IPr)-
(PCy3)] for complex 4.
ability of a homogeneous complex such as 1 to promote the
dehydrogenation of formic acid at low temperature represents,
to the best of our knowledge, a true first.21
Complex 4 was the only Pd species detected during the
mechanistic study and thus is presumed to be a key
intermediate or the catalyst resting state involved in this
hydrogen formation. The propensity of 4 to convert back into 1
with the generation of H2 appears to support this hypothesis
(Scheme 2). To shed light on the unexpected hydrogen
Scheme 2. Pd(0)-Mediated Dehydrogenation of Formic Acid
To establish the atom connectivity in 4 unequivocally (and in
spite of its thermal instability), single crystals were grown at low
temperature,14 and the diffraction data confirmed the structure
assigned on the basis of labeling and NMR studies (Figure 2).
production property displayed by 4, a computational
investigation of the individual reaction steps using density
functional theory (DFT) was undertaken. The results revealed
that dehydrogenation occurs in a stepwise manner without
large energy barriers (Figure 4). HCOOH coordination with
Figure 2. Molecular structure of 4. Selected bond distances (Å) and
angles (deg): Pd−C, 2.064(7); Pd−O, 2.140(5); Pd−P, 2.322(2);
Pd−H, 1.76(2); C−Pd−O, 85.9(2); C−Pd−P, 169.91(18); O−Pd−P,
102.89(16); C−Pd−H, 92(3); O−Pd−H, 174(3); P−Pd−H, 80(3).14
Figure 4. Free energy profile for the dehydrogenation of formic acid.
The isolation of such a palladium complex is, to the best of our
knowledge, unprecedented. Darensbourg did report the
isolation of a greyish [Pd{OC(O)H}(H)(PCy3)2] complex,
but the reported spectroscopic data suggest the formation of a
mixture of isomers under the conditions employed.15 Achieving
such high stability in this architecture may very well be linked
to the presence of the NHC, a strategy having precedent in
synthetic organometallic chemistry.16
spontaneous dissociation of the acid proton to form 4 requires
6.2 kcal/mol. Consistent with the crystallographic structure, the
DFT-optimized geometry of 4 has the formato H atom
pointing away from the metal (see Figures 2 and 4). The first
step of the C−H activation involves a conformational
rearrangement of 4 to a slightly less stable geometry with the
formato H atom pointing toward the metal (structure 4′ in
Figure 4). The energy barrier separating 4 from 4′ through
transition state TS1 is only 1.8 kcal/mol. At this point,
activation of the C−H bond of the formato ligand by the Pd
atom generates 2 through transition state TS2 with a barrier of
12.2 kcal/mol with liberation of a CO2 molecule. TS2 is a very
late transition state in which the CO2 molecule is practically
formed and dissociated from the metal center. According to the
calculations, concerted isomerization of 2 to its cis-H2 analogue
proved to have a very high energy barrier (>30 kcal/mol).
Thus, a mechanism involving dissociation of PCy3 to give
intermediate A, followed by coordination of PCy3 to A to give
Pd dihydride intermediate B, and finally formation of
intermediate C containing a Pd-bound H2 molecule seems
more likely. Dissociation of H2 from C regenerates 1. Since the
total process was computed to release 10.6 kcal/mol, liberation
of both CO2 and H2 in the medium probably drives the
Because of safety and environmental concerns, the use of
formic acid as a hydrogen storage material has received much
attention over the past decade (Figure 3).17 In spite of the
efficient reduction of CO2 to formic acid catalyzed by
homogeneous Ru catalysts, which has been extensively studied
by the Jessop and Joo
́
groups,18 the catalytic dehydrogenation
of formic acid remains an area of current interest.19,20 The
Figure 3. Formic acid as a hydrogen storage material.
B
dx.doi.org/10.1021/ja311087c | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
reaction toward complete consumption of formic acid. The
difference in the stabilities of complexes 1, 4, and 2 easily
explains why no trace of the dihydride species 2 was observed
during the experimental study.
During preliminary labeling studies with complex 1 and
deuterated formic acid (DCOOD, DCOOH, or HCOOD), HD
1
and H2 were observed by H NMR spectroscopy. While the
formation of HD using DCOOD could be easily explained by
exchange between the acidic D and any labile H, the formation
of H2 observed with DCOOH and HCOOD was more
surprising. Those data could be easily related to the isotope
exchange reaction observed when complex 1 was reacted with
HD.10 To gain a better understanding of the process, a series of
experiments involving hydrogenation of 5 catalyzed by 3 were
performed using 1.05 equiv of DCOOD under various
pressures of H2 (see Table S1 in the SI). In addition to the
expected incorporation of two or zero deuterium atoms into
the final product, the monodeuterated compound was formed
in high concentration (Scheme 4). This can be explained by
We previously reported the excellent catalytic performance of
1 in the hydrogenation of C−C bonds under H2.9 The unique
behavior of 1 in the dehydrogenation of formic acid could
permit the hydrogenation of C−C bonds using a tandem
dehydrogenation/hydrogenation sequence. The ability of
Pd(0) species to catalyze such a process was recently reported
using NH3BH3 as the hydrogen source.22 The release of 1 equiv
of H2 in the reaction medium permitted more selective
hydrogenation than classical methodology. To test the
efficiency of the tandem process, kinetic profiling of the
hydrogenation of trans-stilbene (5) using HCOOH as the
hydrogen source was performed. HCOOH (1.05 equiv) was
added to an NMR tube charged with 5 (1 equiv) and 1 (5 mol
%) in THF-d8 at −20 °C, and the progress of the reaction was
Scheme 4. Pd(0)-Mediated Dehydrogenation of Formic Acid
1
monitored by H NMR spectroscopy at various time intervals
at 45 °C (Figure 5). The rapid disappearance of formic acid was
scrambling of H2 and D2 to form HD mediated by 3. The high
yield of the monodeuterated species formed suggests that under
the reaction conditions, the H−D scrambling would be as easily
accomplished as the delivery of H to the alkene. Computational
analysis of this H2/D2 scrambling process mediated by 1 clearly
shows it to be energetically facile (see the SI).
A short survey of the substrate scope was performed to
highlight the versatility of the tandem reaction. Reduction of a
variety of substrates including cyclic and acyclic olefins, α,β-
unsaturated aldehydes, α,β-unsaturated ketones, and other α,β-
unsaturated carbonyl compounds was investigated. The
presence of functional groups such as acids, amides, esters,
ketones, or nitriles did not affect the hydrogenation, and
excellent isolated yields were generally obtained (Figure 6).
Figure 5. Hydrogenation of trans-stilbene using HCOOH as the
hydrogen source at 45 °C.
accompanied by the formation of H2 and fast conversion of 5 to
1,2-diphenylethane. After 30 min at this temperature, no formic
acid or hydrogen could be detected by 1H NMR spectroscopy.
Complex 1 was the only organopalladium species observed at
the end of the tandem transformation. This suggests initial
rapid consumption of formic acid to give formato complex 4
followed by fast conversion of 4 into the catalytically relevant
species 2, which is most likely in equilibrium with precatalyst 1
(Scheme 3).9 Computational analysis of this proposed reaction
profile indicated that initial displacement of PCy3 by the alkene
and H2 occurs, followed by conversion of the alkene to an alkyl
moiety and then reductive elimination in two low-energy steps.
The overall energy profile is provided in the SI.
Scheme 3. Formation of H2 Starting from Formic Acid
Involving Formato Complex 4 and Dihydride Complex 2
Figure 6. Hydrogenation of alkenes using 3 and formic acid.
Since the reduction proceeded efficiently without the need to
use excess formic acid, we next challenged the economy of this
process in the selective semihydrogenation of alkynes (Figure
7). Gratifyingly, using 1.05 equiv of formic acid gave 94% cis-
stilbene. When the scope was expanded to other substrates,
good to excellent Z selectivity was achieved with both aromatic
C
dx.doi.org/10.1021/ja311087c | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Backvall, J.-E.; Andersson, P. G.; Brandt, P. Chem. Soc. Rev. 2006, 35,
̈
237.
(4) Miecznikowski, J. R.; Crabtree, R. H. Polyhedron 2004, 23, 2857.
(5) Campana, A. G.; Estev
́
ez, R. E.; Fuentes, N.; Robles, R.; Cuerva,
̃
J. M.; Bunuel, E.; Cardenas, D.; Oltra, J. E. Org. Lett. 2007, 9, 2195.
́
̃
(6) Koike, T.; Ikariya, T. Adv. Synth. Catal. 2004, 346, 37.
(7) (a) Tani, K.; Ono, N.; Okamoto, S.; Sato, F. J. Chem. Soc., Chem.
Commun. 1993, 386. (b) Hauwert, P.; Maestri, G.; Sprengers, J. W.;
Catellani, M.; Elsevier, C. J. Angew. Chem., Int. Ed. 2008, 47, 3223.
(c) Hauwert, P.; Boerleider, R.; Warsink, S.; Weigand, J. J.; Elsevier, C.
J. J. Am. Chem. Soc. 2010, 132, 16900.
Figure 7. Semireduction of alkynes using 3 and formic acid.
and aliphatic alkynes, and only minor over-reduction to the
corresponding alkanes was observed.23
(8) (a) Brunel, J. M. Tetrahedron 2007, 63, 3899. (b) Brunel, J. M.
Synlett 2007, 330.
In summary, the use of a mixed NHC/PR3 environment
about a palladium center has made possible the isolation of a
unique hydridoformatopalladium complex, 4, involved in the
dehydrogenation of formic acid. Efficient catalytic hydro-
genation of alkenes and alkynes using formic acid proceed via
a tandem sequence using 3. The [Pd(NHC)(PCy3)] complexes
examined play a dual role in catalysis: they facilitate the
dehydrogenation of formic acid to generate hydrogen and
subsequently make use of the H2 formed in situ to reduce C−C
multiple bonds. DFT calculations support the experimental
observations and mechanistic conclusions. The reduction of a
variety of cyclic and acyclic olefins bearing diverse functional
groups was achieved in very good isolated yields. The system
proved to be stereo- and chemoselective for the semi-
hydrogenation of aromatic and aliphatic internal alkynes to
the corresponding Z alkenes under mild conditions. Studies
aimed at capitalizing on the fascinating reactivity of these Pd(0)
complexes are presently ongoing in our laboratories.
(9) Jurcí
2509.
(10) Fantasia, S.; Egbert, J. D.; Jurcí
̌
k, V.; Nolan, S. P.; Cazin, C. S. J. Chem.Eur. J. 2009, 15,
̌
k, V.; Cazin, C. S. J.; Jacobsen,
H.; Cavallo, L.; Heinekey, D. M.; Nolan, S. P. Angew. Chem., Int. Ed.
2009, 48, 5182.
1
(11) For H and 31P{1H} NMR spectra of reactions between 1 and
formic acid in THF-d8 recorded at different temperatures, see the SI.
(12) Formation of H2 was clearly observed by 1H NMR spectroscopy
(see the SI).
(13) Spectra from the labeling studies are available in the SI.
(14) CCDC-873769 contains the supplementary crystallographic
data for 4. These data can be obtained free of charge from the
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
(15) Darensbourg, D. J.; Wiegreffe, P.; Riordan, C. G. J. Am. Chem.
Soc. 1990, 112, 5759.
(16) (a) N-Heterocyclic Carbenes in Transition Metal Catalysis and
Organocatalysis; Cazin, C. S. J., Ed.; Springer: London, 2011.
(b) Nolan, S. P. Acc. Chem. Res. 2011, 44, 91. (c) Scott, N. M.;
Nolan, S. P. Eur. J. Inorg. Chem. 2005, 1815.
(17) Enthaler, S. ChemSusChem 2008, 1, 801.
(18) (a) Johnson, T. C.; Morris, D. J.; Wills, M. Chem. Soc. Rev. 2010,
ASSOCIATED CONTENT
* Supporting Information
Experimental procedures and details, NMR spectra of
stoichiometric reactions and catalytic products, a CIF for 4,
Cartesian coordinates and energies for all species discussed in
the text, and computational details. This material is available
free of charge via the Internet at http://pubs.acs.org.
■
S
39, 81. (b) Joo,
́
F. ChemSusChem 2008, 1, 805.
(19) (a) Boddien, A.; Mellmann, D.; Gartner, F.; Jackstell, R.; Junge,
H.; Dyson, P. J.; Laurenczy, G.; Ludwig, R.; Beller, M. Science 2011,
̈
333, 1733. (b) Loges, B.; Boddien, A.; Gartner, F.; Junge, H.; Beller,
̈
M. Top. Catal. 2010, 53, 902. (c) Fellay, C.; Yan, N.; Dyson, P. J.;
Laurenczy, G. Chem.Eur. J. 2009, 15, 3752. (d) Fellay, C.; Dyson, P.
J.; Laurenczy, G. Angew. Chem., Int. Ed. 2008, 47, 3966. (e) Fukuzumi,
S.; Kobayashi, T.; Suenobu, T. ChemSusChem 2008, 1, 827.
(20) For a recent contribution examining the formation of formato
complexes using CO2 insertion into M−H bonds of Ni and Pd pincer
complexes, see: Suh, H.-W.; Schmeier, T. J.; Hazari, N.; Kemp, R. A.;
Takase, M. K. Organometallics 2012, 31, 8225.
AUTHOR INFORMATION
Corresponding Author
cc111@st-andrews.ac.uk
Notes
■
(21) Hydrogen production experiments were carried out using 0.5
mol% [Pd(IPr)(PCy3)] (see the SI).
The authors declare no competing financial interest.
(22) Hartmann, C. E.; Jurcí
Commun. 2013, 49, 1005.
(23) The addition of a drop of mercury resulted in a noticeable
decrease in conversion (see the SI). For the limitations of the mercury
drop test, see: Crabtree, R. H. Chem. Rev. 2012, 112, 1536.
̌
k, V.; Songis, O.; Cazin, C. S. J. Chem.
ACKNOWLEDGMENTS
The EPSRC and the Royal Society (University Research
Fellowship to C.S.J.C.), the Spanish MICINN (Ramon
Contract RYC-2009-05226 to A.P.), the EC (Career Integra-
tion Grant CIG09-GA-2011-293900 to A.P.), and Generalitat
de Catalunya (BE Fellowship 2011BE100793 to A.P.) are
acknowledged for financial support. Umicore AG is thanked for
a generous gift of materials.
■
́
y Cajal
REFERENCES
■
(1) de Vries, J. G.; Elsevier, C. J. Handbook of Homogeneous
Hydrogenation; Wiley-VCH: Weinheim, Germany, 2007.
(2) Molkov, V. In Fire and Explosion Hazards: Proceedings of the 5th
International Seminar; Bradley, D., Drysdale, D., Molkov, V., Carvel, R.,
Eds.; University of Edinburgh: Edinburgh, U.K., 2007.
(3) For reviews of transfer hydrogenation reactions, see: (a) Brieger,
G.; Nestrick, T. J. Chem. Rev. 1974, 74, 567. (b) Zassinovich, G.;
Mestroni, G.; Gladiali, S. Chem. Rev. 1992, 92, 1051. (c) Gladiali, S.;
Alberico, E. Chem. Soc. Rev. 2006, 35, 226. (d) Samec, J. S. M.;
D
dx.doi.org/10.1021/ja311087c | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX