Efficient disproportionation of formic acid to methanol using molecular ruthenium catalysts

Article abstract of DOI:10.1002/anie.201405457

The disproportionation of formic acid to methanol was unveiled in 2013 using iridium catalysts. Although attractive, this transformation suffers from very low yields; methanol was produced in less than 2 % yield, because the competitive dehydrogenation of formic acid (to CO₂ and H₂) is favored. We report herein the efficient and selective conversion of HCOOH to methanol in 50 % yield, utilizing ruthenium(II) phosphine complexes under mild conditions. Experimental and theoretical (DFT) results show that different convergent pathways are involved in the production of methanol, depending on the nature of the catalyst. Reaction intermediates have been isolated and fully characterized and the reaction chemistry of the resulting ruthenium complexes has been studied. The search for selectivity: Methanol is efficiently produced in >50 % yield by the disproportionation of formic acid using molecular ruthenium catalysts. Mechanistic experimental and DFT investigations have unveiled different pathways involving transient ruthenium hydride species.

Full text of DOI:10.1002/anie.201405457

.
Angewandte
Communications
DOI: 10.1002/anie.201405457
Methanol Formation
Efficient Disproportionation of Formic Acid to Methanol Using
Molecular Ruthenium Catalysts**
Solꢀne Savourey, Guillaume Lefꢀvre, Jean-Claude Berthet, Pierre Thuꢁry, Caroline Genre, and
Thibault Cantat*
Abstract: The disproportionation of formic acid to methanol
was unveiled in 2013 using iridium catalysts. Although
attractive, this transformation suffers from very low yields;
methanol was produced in less than 2% yield, because the
competitive dehydrogenation of formic acid (to CO2 and H2) is
favored. We report herein the efficient and selective conversion
of HCOOH to methanol in 50% yield, utilizing ruthenium(II)
phosphine complexes under mild conditions. Experimental
and theoretical (DFT) results show that different convergent
pathways are involved in the production of methanol, depend-
ing on the nature of the catalyst. Reaction intermediates have
been isolated and fully characterized and the reaction chemis-
try of the resulting ruthenium complexes has been studied.
reduction of CO2 to methanol [Eq. (3)]. This strategy relies on
the two-electron reduction of CO2 to FA, in an electro-
chemical cell, and this methodology is now technically and
economically available, thanks to efficient electrocatalysts.[6]
Disproportionation of FA is then required to produce
methanol. Although decomposition of FA usually proceeds
by dehydrogenation or dehydration to form CO2 and CO
[Eqs. (4) and (5)],[7] Miller, Goldberg et al. showed, for the
first time in 2013, that a molecular complex could promote the
disproportionation of FA to methanol.[8] Using [(C5Me5)Ir-
(bpy)(H2O)][OTf]2 (bpy = 2,2’-bipyridine, OTf = trifluoro-
methanesulfonate) as a catalyst, aqueous solutions of FA
could be converted to MeOH at 808C. Though promising, this
strategy currently suffers from the use of expensive iridium
catalysts and the yields of methanol do not exceed 1.9%.[8]
This effect primarily results from poor selectivity as 88% of
the reacted FA undergoes dehydrogenation instead of dis-
proportionation. Catalysts with improved activity and selec-
tivity are highly desirable to reach the potential of this
approach. Herein, we report the efficient disproportionation
of FA to methanol, with methanol yields of up to 50.2%, using
ruthenium molecular catalysts. Mechanistic insights are also
provided, based on the isolation of reactive catalytic inter-
mediates and DFT calculations.
T
he efficient conversion of carbon dioxide to methanol is
a key process in devising a methanol economy based on
a closed carbon cycle.[1] Such an achievement requires the
development of catalysts and methods to promote the six-
electron reduction of CO2 using carbon-free energy input with
a high overall Faraday efficiency. The most obvious solution
would rely on the direct electro-reduction of CO2 to methanol
[Eq. (1)]; yet current state-of-the-art catalysts proceed with
only a low faradaic yield and a low selectivity at high
overpotentials.[2] In contrast, hydrogen formation from the
(photo)electro-reduction of water has witnessed compelling
successes recently and such carbon-free sources of H2 could
serve for the reduction of CO2 to methanol [Eq. (2)].[3] The
hydrogenation of CO2 to methanol has indeed been the focus
of considerable effort and catalysts are proposed to overcome
the kinetic stability of these two nonpolar gases.[4,5] In this
respect, the conversion yield starting from H2 is still limited
because high H2 pressure is required, leading to an overall low
faradaic efficiency. An interesting alternative would consist in
ꢀ
utilizing formic acid (FA) as a C H bond shuttle in the
[*] S. Savourey, Dr. G. Lefꢀvre, Dr. J.-C. Berthet, Dr. P. Thuꢁry,
Dr. C. Genre, Dr. T. Cantat
CEA, IRAMIS, NIMBE, CNRS UMR 3299
91191 Gif-sur-Yvette (France)
E-mail: thibault.cantat@cea.fr
Homepage: http://iramis.cea.fr/Pisp/thibault.cantat/index.htm
[**] This work was supported financially by the CEA, CNRS,
CHARMMMAT Laboratory of Excellence, and the European
Research Council (ERC starting grant agreement no. 336467). T.C.
thanks the Fondation Louis D.—Institut de France for its support.
We thank CINES for computer time (Project No. c2014086494).
Supporting information for this article (including detailed descrip-
tions of experimental, spectroscopic, crystallographic, and quantum
chemical methods and results) is available on the WWW under
http://dx.doi.org/10.1002/anie.201405457.
Ruthenium complexes are well-established catalysts in
reduction chemistry and their potential was recently illus-
trated, by the groups of Milstein, Beller, and Leitner, in the
10466
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 10466 –10470

Angewandte
Chemie
hydrogenation of a variety of reluctant substrates, such as
CO2, carbonates, carbamates, and amides.[9] In addition,
ruthenium is less expensive than iridium ($75 per oz vs.
$830 per oz in 2013). The disproportionation of formic acid
was thus investigated, utilizing ruthenium(II) complexes
supported by external phosphine ligands (Scheme 1, Table 1,
and Table S1). To our delight, we observed that heating
a THF solution of FA in a sealed vessel at 1508C, in the
presence of 0.6 mol% [Ru(COD)(methylallyl)2] and
0.6 mol% CH3C(CH2PPh2)3 (triphos), resulted in the com-
plete conversion of FA to produce methanol in 5.0% yield,
after 1 h (entry 1, Table 1). This result corresponds to a 5.0%
selectivity for the formation of methanol, meaning that 5.0%
ꢀ
of the reacted C H bonds in FA are efficiently converted to
ꢀ
methanol. The remaining 95.0% of the C H bonds are
transformed to H2 via dehydrogenation [Eq. (4)] and H2
evolution was indeed confirmed by H NMR spectroscopy.
1
When H13CO2H was utilized, 13CH3OH was formed and 13CO2
was identified as the only organic by-product in this trans-
formation and no trace of carbon monoxide, formaldehyde, or
methylformate could be observed by 13C NMR spectroscopy.
The absence of CO was further confirmed by GC analysis of
the gas mixture at the end of the reaction (see the Supporting
Information (SI)). Interestingly, the reaction proceeds well in
the absence of any additive or buffer, while pH < 2 must be
maintained with HBF4 when catalyst [(C5Me5)Ir(bpy)(H2O)]-
(OTf)2 is utilized.[8]
Different supporting ligands and ruthenium precursors
were then screened so as to improve the catalytic activity and
selectivity (Table 1 and Table S1). With PPh3, 1,2-bis(diphe-
nylphosphino)ethane (dppe), P(CH2CH2PPh2)3 (PP3), and
2,6-bis(diisopropylphosphinomethyl)pyridine as ligands,
dehydrogenation of FA was observed with no formation of
methanol (entries 2 and 3 in Table 1; experiments with the
latter two ligands described in Table S1). However, replacing
the triphos ligand with the triphosphinite CH3C(CH2OPPh2)3
(OP)3 ligand led to the conversion of FA to methanol in 0.5%
yield (entry 4, Table 1).
Scheme 1. Precursor 1, deactivated complexes 2 and 3, and reaction
intermediates 4 and 5 characterized from the ruthenium-catalyzed
disproportionation of HCOOH to CH3OH.
Importantly, the reaction temperature and the initial
concentration of FA strongly influence the catalytic activity
(entries 1 and 6–12, Table 1). Decreasing the initial quantity
of FA from 2.4 to 0.8 mmol is accompanied by a decrease in
MeOH yield from 11.9 to 0.5% (entries 8, 11, and 12).
Noticeably, [Ru(COD)(methylallyl)2] is known to react with
triphos to afford complex 1 (Scheme 1), which was structur-
ally characterized (Figure S14).[5b] The catalytic activity of 1 is
similar to that of [Ru(COD)(methylallyl)2] + triphos and
MeOH was obtained in 9.7% yield (vs. 11.9% yield) with
the isolated complex 1 (entries 5 and 8). Although the
disproportionation reaction is efficient at 1508C, it also
proceeds well at 808C and is significantly slowed down only
at temperatures below 408C. For example, the transformation
of 2.4 mmol FA affords MeOH in 11.9 and 7.6% yield at 150
and 808C, respectively, while the yield drops to 1.0% at 408C
(entries 6–8, Table 1). The nature of the solvent was also
found to impact the outcome of the reaction. While the
formation of MeOH proceeds equally well in toluene and
benzene, it is considerably slower in CHCl3 where MeOH is
formed in a low 1.2% yield (entries 13–15, Table 1). In fact,
the catalytic system was found to degrade in CHCl3 at 1508C,
and yellow crystals of 2 deposited from the crude mixture
within 24 h. X-ray analysis reveals that 2 is a ruthenium(II)
hydrido chloride complex that coordinates one triphos ligand
and one molecule of CO (Figure 1). The extra chloride ligand
likely results from the abstraction of a Cl atom from the
solvent by a Ru–H species.[10]
Table 1: Catalytic disproportionation of formic acid to methanol.
Entry FA
[mmol]
L
Additive
Solv.
T
t[a] CH3OH
(1.5 mol%) (0.3 mL) [8C] [h] yield [%]
1
2
3
4
5
6
7
0.6
triphos
3 PPh3
2 dppe
(OP)3
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
THF
CHCl3
C7H8
C6H6
THF
THF
THF
THF
150
150
150
150
150
80 17 7.6
40 72 1.0
150
1
1
1
1
1
5.0
2.4
2.4
2.4
2.4
2.4
2.4
2.4
4.8
<0.1
<0.1
0.5
1[b]
9.7[b]
triphos
triphos
triphos
triphos
triphos
triphos
triphos
triphos
triphos
triphos
8
1
11.9
9[c]
40 72 1.0
80 17 26.7
10[c] 4.8
11[d] 0.8
12[e] 1.6
150
150
150
150
150
150
150
150
150
1
1
1
1
1
1
1
1
1
0.5
7.5
1.2
5.8
7.4
6.8
11.6
11.9
50.2
13
14
15
16
17
18
19
2.4
2.4
2.4
2.4
2.4
2.4
2.4
triphos NEt3
triphos H2O
triphos EtOH
triphos MSA
Another decomposition pathway of the Ru catalyst was
also identified when the disproportionation of FA was carried
out in benzene and toluene; after 24 h at 1508C, crystals of
[Ru(k3-triphos)(CO)(H)2] (3) were obtained.[11] Although 3
proves inactive in the disproportionation of FA, it catalyzes
Reaction conditions: cat. [Ru(COD)(methylallyl)2] (0.6 mol%); yields
determined by H NMR spectroscopy in deuterated solvents, using
mesitylene as an internal standard. [a] Reaction time required to achieve
100% conversion. [b] cat.: [Ru(triphos)(tmm)](0.6 mol%) (tmm=tri-
methylene methane). [c] 0.3 mol% cat. [d] 1 mol% cat. [e] 2 mol% cat.
1
Angew. Chem. Int. Ed. 2014, 53, 10466 –10470
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 10467

.
Angewandte
Communications
by the groups of Beller, Leitner, and Klankermayer to
promote the methylation of amines with CO2/H2.[9c,d] Follow-
ing this approach, the disproportionation of FA was explored
using 0.6 mol% [Ru(COD)(methylallyl)2] + triphos and
1.5 mol% MSA, at 1508C (entry 19, Table 1). Under these
conditions, FA is fully decomposed within 1 h leading to the
formation of MeOH in 50.2% yield. To the best of our
knowledge, this result represents the best selectivity obtained
for the disproportionation of FA to MeOH.
Different pathways can be proposed based on the possible
organic intermediates involved in the disproportionation of
FA (Scheme 2). FA can first undergo a catalytic reduction to
Figure 1. ORTEP views of complexes 2 (top) and 4·THF (bottom).
Solvent molecules and hydrogen atoms are omitted. Displacement
ellipsoids are drawn at the 30% probability level.
Scheme 2. Proposed pathways for the disproportionation of formic
acid to methanol.
efficiently its dehydrogenation [Eq. (4)], in agreement with
a recent report by Peruzzini, Gonsalvi et al.[7b,c] Because the
formation of CO complexes 2 and 3 was only observed 23 h
after the complete conversion of FA, it is unlikely that 3
results from the dehydration of FA [Eq. (5)]. Moreover, Cole-
Hamilton et al. have recently shown that ruthenium(II)
carbonyl complexes can form by decarbonylation of alde-
hydes or alcohols under reducing conditions, and a similar
route could account for the generation of 2 and 3 in the
presence of methanol and/or formaldehyde.[9e,12] Based on
these findings, the efficient disproportionation of 4.8 mmol
HCOOH to MeOH was achieved in 26.7% yield using
0.3 mol% [Ru(COD)(methylallyl)2] + triphos as a catalyst,
after 17 h at 808C (entry 10, Table 1). This result corresponds
to a complete conversion of FA and a selectivity of 26.7% for
MeOH production and it affords a catalytic turnover number
(TON) of 26 (TOF = 1.5 hꢀ1). In comparison, MeOH was only
obtained in 1.9% yield (as a mixture of MeOH and
methylformate), at best, with 0.002 mol% [(C5Me5)Ir(bpy)-
(H2O)][OTf]2, after 24 h at 808C.[8] Although the catalytic
activity of iridium is somewhat greater than that of the
present ruthenium system (TON = 156, TOF = 6.5 hꢀ1) it
exhibits a low selectivity for MeOH production (7%).
formaldehyde by transfer hydrogenation using a second
equivalent of FA. Since formaldehyde is highly reactive
towards reduction, it is readily reduced to MeOH either by
transfer hydrogenation (with a third equivalent of FA) or by
hydrogenation (from the competitive dehydrogenation of
FA).[14] Additionally, the competitive dehydrogenation of FA
to H2 and CO2 can also converge to methanol, as long as the
catalyst is able to promote the difficult hydrogenation of CO2
to methanol (Scheme 2). In fact, the Leitner group demon-
strated in 2013 that [1/MSA] was an efficient catalyst in the
hydrogenation of CO2 to MeOH above 1208C in the presence
of alcohols as promoters.[4b] Monitoring the product distribu-
tion for the disproportionation of FA catalyzed by [1/MSA]
reveals that the dehydrogenation of FA is strongly favored at
short reaction times and only 0.5% MeOH is formed once FA
is fully consumed. Yet, MeOH production then increases over
time via hydrogenation of CO2 assisted by the catalytic
amount of methanol formed at an earlier stage (Figure S7).[15]
Importantly, in the absence of MSA, a distinct reaction profile
was observed and the production of MeOH follows the
consumption of FA and stops once FA is fully converted. In
addition, the hydrogenation of CO2 is unlikely under the mild
conditions (< 1208C) in the absence of MSA.[5b] These
findings suggest that 1, in the absence of MSA, promotes
the disproportionation of FA via transfer hydrogenation. This
conclusion also sheds some light on the influence of the initial
concentration of FA on the selectivity of the reaction (vide
supra). Indeed, both the dehydrogenation and the dispropor-
tionation of FA produce gases and, in a closed vessel,
doubling the initial quantity of FA nearly generates a twofold
increase in pressure (Entries 8, 11, 12 in Table 1). Yet, while
the dehydrogenation of 1 mole FA produces 2 moles of gases,
In order to enhance further the productivity of MeOH, it
is necessary to increase the selectivity of the reaction and
favor the reduction of FA. In this respect, detailed exper-
imental and mechanistic studies by Klankermayer, Leitner
et al. have shown that acid promoters, such as methanesul-
fonic acid (MSA), could significantly boost the catalytic
activity of [Ru(COD)(methylallyl)2] + triphos or 1 in hydro-
genation reactions, by facilitating the formation of reactive
Ru–H species.[13] This strategy has been successfully utilized
10468 www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 10466 –10470

Angewandte
Chemie
its disproportionation to MeOH generates only 2/3 mole of
gas. According to le Chꢀtelierꢁs law, the disproportionation of
FA is thus favored over dehydrogenation, at high pressure
(i.e. at high initial concentration of FA), leading to an
increased selectivity towards MeOH production.
Using [Ru(COD)(methylallyl)2] + triphos (or 1) as a cata-
lyst, the disproportionation of FA therefore proceeds by
means of an unexplored transfer hydrogenation pathway and
we then turned our attention to the role of the metal catalyst
in this transformation. Complex 1 reacts at RT with 2 equiv
FA in THF to afford the new complex 4 [Eq. (6)]. Crystals of
implication in this reaction. Furthermore, addition of 2 equiv
FA to a THF solution of 4 resulted in the formation of
a mixture of methanol, H2, CO2, and 5 within 48 h at RT
[Eq. (7)]. This reactivity confirms that MeOH production
does not result from the hydrogenation of CO2 (which
requires high H2 pressure)[5b] but rather from the transfer
hydrogenation of FA.
Scheme 3 lays out a plausible pathway for methanol
formation supported by the available data and by DFT
computations. Competition between the decomposition of
formic acid (leading to H2 and CO2) and its disproportiona-
tion to MeOH has been investigated using DFT techniques,
with the simplified CH3C(CH2PMe2)3 ligand in place of
triphos. Both pathways rely on the formation of 5 and its
generation from 4 proceeds through the decoordination of
one k2-OCHO formate ligand and decarboxylation of the
unsaturated [Ru(triphos)(k1-OCHO)2] ((k1,k1)-4) intermedi-
ate. This step requires a relatively low activation energy of
4 were successfully grown from a THF solution and its X-ray
structure reveals the formation of a ruthenium(II) bis-
formate complex, namely [Ru(triphos)(k1-OCHO)(k2-
¼
OCHO)]. A single C H bond signal was observed for the
DG = 17.1 kcalmolꢀ1, consistent with the experimental find-
ꢀ
two formate ligands in the 1H (8.49 ppm) and 13C NMR
(172.3 ppm) spectra of 4, revealing fluxional behavior at RT.
Interestingly, 4 decarboxylates within minutes at RT to afford
complex 5. Based on 1H, 13C, and 31P NMR and ESI-MS data,
complex 5 can be formulated as the ruthenium(II) hydrido
formate complex [Ru(triphos)(H)(k2-OCHO)], characterized
ings [Eq. (7)].[17] Coordination of an incoming molecule of FA
can provide either complexes 6 or 9, which only differ by the
presence of an H-bond between the two formate ligands in 9.
Yet, these two intermediates evolve along different routes.
The presence of an acidic O–H functionality in 6 can quench
the Ru–H group to produce H2 and regenerate 4 with a very
low activation barrier (0.4 kcalmolꢀ1). In contrast, the intra-
molecular H-bond in 9 decreases the acidity of the O–H
function and, at the same time, increases the electrophilic
character of the (k2-OCHO) formate ligand, thereby facili-
2
by a typical [Ru]-H signal at d = ꢀ2.42 (dt, JPcisH = 20 Hz,
2JPtransH = 124 Hz) and a [Ru]-OCOH singlet at d = 8.24.[16]
Notably, both 4 and 5 exhibit catalytic activity similar to that
of 1 in the disproportionation of FA, thereby confirming their
Scheme 3. Computed pathways for the catalytic disproportionation of FA to methanol and FA dehydrogenation to H2 and CO2 (at the M06/6-
31+G* (H,C,O,P) + SDD (Ru) level of theory).
Angew. Chem. Int. Ed. 2014, 53, 10466 –10470
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 10469

.
Angewandte
Communications
Riduan, Y. Zhang, J. Y. Ying, Angew. Chem. Int. Ed. 2009, 48,
3322; Angew. Chem. 2009, 121, 3372; e) S. Chakraborty, J.
Zhang, J. A. Krause, H. Guan, J. Am. Chem. Soc. 2010, 132,
8872; f) C. D. N. Gomes, E. Blondiaux, P. Thuꢂry, T. Cantat,
Chem. Eur. J. 2014, 20, 7098.
tating a hydride migration to produce the acetal intermediate
10 with an activation energy of 12.2 kcalmolꢀ1. Protonolysis
of 10 with FA can then regenerate 4 with the release of
H2C(OH)2, which dehydrates to formaldehyde prior to its
easy reduction to methanol.[14] As outlined in the computed
energy surface (Scheme 3), the facial coordination of triphos
to the RuII center favors a facial arrangement of the two
formate ions in 4 and likely facilitates the transfer hydro-
genation in TS3. Overall, this mechanism highlights that the
decarboxylation of a formate ligand to produce a reactive Ru
hydride species is the rate-determining step (RDS) and it
precedes the divergent dehydrogenation/disproportionation
pathways. Experimentally, we found that the disproportiona-
tion of HCO2D, catalyzed by 1 indeed yielded a mixture of
CH3OD, CH2DOD, and CHD2OD, resulting from a hydride
scrambling enabled by the formation of H–D complex 7 (see
SI). As a result, the selectivity for the production of MeOH is
under thermodynamic control. While the dehydrogenation of
one molecule of FA is favored at low pressure (DG =
ꢀ9.9 kcalmolꢀ1 vs. ꢀ8.6 kcalmolꢀ1 for the disproportionation
route), the formation of methanol by transfer hydrogenation
is favored at high pressure, in agreement with the exper-
imental findings. Further work is underway in our laborato-
ries to translate these conclusions into the design of earth-
abundant metal catalysts with increased selectivity for the
production of methanol from FA.
[6] For a review, see H.-R. Jhong, S. Ma, P. J. A. Kenis, Curr. Opin.
Chem. Eng. 2013, 2, 191.
[7] a) A. Boddien, D. Mellmann, F. Gꢃrtner, R. Jackstell, H. Junge,
P. J. Dyson, G. Laurenczy, R. Ludwig, M. Beller, Science 2011,
333, 1733; b) I. Mellone, M. Peruzzini, L. Rosi, D. Mellmann, H.
Junge, H. Beller, L. Gonsalvi, Dalton Trans. 2013, 42, 2495; c) G.
Manca, I. Mellone, F. Bertini, M. Peruzzini, L. Rosi, D. Mellman,
H. Junge, M. Beller, A. Ienco, L. Gonsalvi, Organometallics
2013, 32, 7053.
[8] A. J. M. Miller, D. M. Heinekey, J. M. Mayer, K. I. Goldberg,
Angew. Chem. Int. Ed. 2013, 52, 3981; Angew. Chem. 2013, 125,
4073.
[9] a) E. Balaraman, C. Gunanathan, J. Zhang, L. J. W. Shimon, D.
Milstein, Nat. Chem. 2011, 3, 609; b) P. P. M. Schleker, R.
Honeker, J. Klankermayer, W. Leitner, ChemCatChem 2013, 5,
1762; c) K. Beydoun, T. vom Stein, J. Klankermayer, W. Leitner,
Angew. Chem. Int. Ed. 2013, 52, 9554; Angew. Chem. 2013, 125,
9733; d) Y. Li, I. Sorribes, T. Yan, K. Junge, M. Beller, Angew.
Chem. Int. Ed. 2013, 52, 12156; Angew. Chem. 2013, 125, 12378;
e) J. Coetzee, D. L. Dodds, J. Klankermayer, S. Brosinski, W.
Leitner, A. M. Z. Slawin, D. J. Cole-Hamilton, Chem. Eur. J.
2013, 19, 11039.
[10] J. Gomez, C. Gemel, C. Slugovc, E. Wozniak, R. Schmid, K.
Krichner, Monatsh. Chem. 2001, 132, 1137.
[11] The identity of 3 was confirmed by X-ray diffraction analysis,
see: V. I. Bakhmutov, E. V. Bakhmutov, N. V. Belkova, C.
Bianchini, L. M. Epstein, D. Masi, M. Peruzzini, E. S. Shubina,
E. V. Vorontsov, F. Zanobini, Can. J. Chem. 2001, 79, 479.
[12] N. Sieffert, R. Rꢂocreux, P. Lorusso, D. J. Cole-Hamilton, H.
Bꢄhl, Chem. Eur. J. 2014, 20, 4141.
[13] F. M. Geilen, B. Engendahl, M. Holscher, J. Klankermayer, W.
Leitner, J. Am. Chem. Soc. 2011, 133, 14349.
[14] The ability of complex 4 to promote the reduction of formal-
dehyde to methanol by transfer hydrogenation from FA was
confirmed experimentally (see SI).
[15] It is also possible that under the applied conditions the
dehydrogenation of FA to CO2 and H2 is reversible and [1/
MSA] promotes the formation of MeOH through transfer
hydrogenation from FA.
[16] The NMR data recorded for 5 are consistent with those reported
for its acetate analogue [Ru(triphos)(H)(OAc)], see: B. Chaplin,
P. J. Dyson, Inorg. Chem. 2008, 47, 47381.
[17] It is interesting to note the role played by the kinetic trans effect
related to the hydrido ligand. Two conformations can indeed be
considered for the transition state leading to the complex 4. A
mechanism involving a distorsion of the (PtransRuH) angle
(145.38) is strongly unfavored (see SI) with a computed barrier
31.2 kcalmolꢀ1 higher than that for a structure in which the
(PtransRuH) angle still close to 1808 (171.18). The labilization of
the PMe2 trans ligand in 5 is attested by the relative position of
the Ptrans signal in 31P{1H} NMR, which is at higher field than the
two Pcis peaks (see SI), and by the longer computed Ptrans–Ru
bond (2.36 ꢅ vs. 2.22 and 2.23 ꢅ for the two Pcis–Ru bonds).
Received: May 20, 2014
Published online: August 1, 2014
Keywords: density functional calculations · formic acid ·
.
homogeneous catalysis · hydrogen · methanol
[1] G. Olah, A. Goeppert, G. K. Surya Prakash, Beyond Oil and
Gas: The Methanol Economy, Wiley-VCH, Weinheim, 2009.
[2] For a review, see J.-M. Savꢂant, Chem. Rev. 2008, 108, 2348.
[3] V. Artero, M. Fontecave, Chem. Soc. Rev. 2013, 42, 2338; D. G.
Nocera, Acc. Chem. Res. 2012, 45, 767; N. Armaroli, V. Balzani,
ChemSusChem 2011, 4, 21.
[4] a) P. G. Jessop, T. Ikariya, R. Noyori, Chem. Rev. 1995, 95, 259;
b) S. Wesselbaum, T. vom Stein, J. Klankermayer, W. Leitner,
Angew. Chem. 2012, 124, 7617; Angew. Chem. Int. Ed. 2012, 51,
7499; c) R. Tanaka, M. Yamashita, K. Nozaki, J. Am. Chem. Soc.
2009, 131, 14168; d) C. Federsel, A. Boddien, R. Jackstell, R.
Jennerjahn, P. J. Dyson, R. Scopelliti, G. Laurenczy, M. Beller,
Angew. Chem. Int. Ed. 2010, 49, 9777; Angew. Chem. 2010, 122,
9971; e) C. A. Huff, M. S. Sanford, J. Am. Chem. Soc. 2011, 133,
18122.
[5] For recent results on CO2 reduction to methanol derivatives with
hydrosilanes and hydroboranes see: a) S. Bontemps, L. Vendier,
S. Sabo-Etienne, J. Am. Chem. Soc. 2014, 136, 4419; b) Y. Li, X.
Fang, K. Junge, M. Beller, Angew. Chem. Int. Ed. 2013, 52, 9568;
Angew. Chem. 2013, 125, 9747; c) O. Jacquet, X. Frogneux,
C. D. N. Gomes, T. Cantat, Chem. Sci. 2013, 4, 2127; d) S. N.
10470 www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 10466 –10470