Homogeneous Palladium-Catalyzed Transfer Hydrogenolysis of Benzylic Alcohols Using Formic Acid as Reductant

Article abstract of DOI:10.1002/chem.201801466

We report the first homogeneous palladium-based transfer hydrogenolysis of benzylic alcohols using an in situ formed palladium-phosphine complex and formic acid as reducing agent. The reaction requires a catalyst loading as low as only 1 mol % of palladium and just a slight excess of reductant to obtain the deoxygenated alkylarenes in good to excellent yields. Besides demonstrating the broad applicability for primary, secondary and tertiary benzylic alcohols, a reaction intermediate could be identified. Additionally, it could be shown that partial oxidation of the applied phosphine ligand was beneficial for the course of the reaction, presumably by stabilizing the active catalyst. Reaction profiles and catalyst poisoning experiments were used to characterize the catalyst, the results of which indicate a homogeneous metal complex as the active species.

Full text of DOI:10.1002/chem.201801466

A Journal of
Accepted Article
Title: Homogeneous Palladium-Catalyzed Transfer Hydrogenolysis of
Benzylic Alcohols Using Formic Acid as Reductant
Authors: Benjamin Ciszek and Ivana Fleischer
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201801466
Link to VoR: http://dx.doi.org/10.1002/chem.201801466
Supported by

Chemistry - A European Journal
10.1002/chem.201801466
COMMUNICATION
Homogeneous Palladium-Catalyzed Transfer Hydrogenolysis of
Benzylic Alcohols Using Formic Acid as Reductant
[a]
[a]
Benjamin Ciszek and Ivana Fleischer*
Abstract: Herein we report the first homogeneous palladium-based
transfer hydrogenolysis of benzylic alcohols using an in situ formed
palladium-phosphine complex and formic acid as reducing agent. The
reaction requires a catalyst loading as low as only 1 mol% of
palladium and just a slight excess of reductant to obtain the
deoxygenated alkylarenes in good to excellent yields. Besides
demonstrating the broad applicability for primary, secondary and
tertiary benzylic alcohols, a reaction intermediate could be identified.
Additionally, it could be shown that partial oxidation of the applied
phosphine ligand was beneficial for the course of the reaction,
presumably by stabilizing the active catalyst. Reaction profiles and
catalyst poisoning experiments were used to characterize the catalyst,
the results indicate a homogeneous metal complex as active species.
Intrigued by the lack of a homogeneous palladium-based
transfer hydrogenation of benzylic alcohols, we were determined
to develop such a method using formic acid as reductant
(Scheme 1). Compared to conventionally used hydride sources,
formic acid bears several advantages concerning its transport and
2 2
and H
.[11]
handling and can be easily produced from CO
Furthermore, it exhibits a higher chemoselectivity than hydrogen
gas as overreduction of aromatic systems is prevented.[12] In the
transfer hydrogenolysis of alcohols with formic acid, only water
2
and CO are obtained as by-products. Re-using the latter in formic
acid synthesis would make up for a highly atom economic and
sustainable process.
Benzylic Hydrogenolysis: Literature
The reduction of C-O to C-H bonds constitutes an essential
transformation in organic chemistry, e.g. in the deoxygenation of
alcohols or carbonyl compounds.[1] The large number of methods
reported in the last century account for this, with the most popular
one being the Barton-McCombie-deoxygenation.[2] A severe
drawback of these methods is the need for harsh reaction
conditions and high amounts of reducing agents, leading to
decrased selectivity and toxic waste. Therefore, catalytic
approaches such as hydrogenolysis are of considerable interest.
The term hydrogenolysis refers to the cleavage of C-C or C-
heteroatom bonds by using hydrogen. Since Padoa and Ponti
OH
H
het. Pd
+
H2 (or R SiH, HCOOH)
3
Ar
Ar
• high catalyst loading
overreduction of aromatics possible
•
Our Work
OH
Ar
H
Pd(II), Ligand
HA
+
HCOOH
Ar
HOMOGENEOUS CATALYST
HCOOH AS REDUCTANT
•
•
milder conditions
mechanistic studies
• no overreduction
• H O, CO waste
described the nickel-catalyzed hydrogenolysis of furfuryl alcohol
in 1908,[3] numerous protocols for this transformation were
2
2
developed. More recent works, especially those concerning the
transformation of benzylic alcohols and ketones into the
corresponding alkylarenes, apply transition metal catalysts like
nickel,[4] cobalt,[4] palladium[5] or ruthenium.[6] Molecular hydrogen
is often chosen as reductant because of its low price, the high
atom economy of the reduction and for water being the only by-
product formed.[7] Nonetheless, hydrogen gas is highly flammable
and requires special pressure equipment. For these reasons, so-
called transfer hydrogenolysis reactions are applied, using
alternative hydride donors like silanes,[5a-d, 8] alcohols,[4]
cyclohexadiene[9] or formic acid.[10] When applying formic acid as
reducing agent, palladium is most frequently used as catalyst;
however, all reported methods for the deoxygenation of benzylic
alcohols with formic acid so far rely on heterogeneous palladium
on solid support, requiring high catalyst loadings.[10] Thus, the
development of a more selective and active homogeneous
catalyst system appears highly attractive. Additionally,
homogeneous catalysis usually allows for easier catalyst
modification and mechanistic investigations.
Scheme 1. Heterogeneous versus homogeneous palladium-catalyzed transfer
hydrogenolysis.
We chose 1-phenylethanol (1a) as a model substrate for the
optimization experiments (Scheme 2). Based on our previous
studies using Pd-catalysts,[13] the applied catalyst system
consisted of a palladium precatalyst, bidentate phosphine ligand
L1 and a Brønsted acid co-catalyst. The initial optimization (for
II
extended details see the SI) showed that use of Pd as well as
soluble Pd0 precursors in the reaction resulted in excellent to
2
quantitative product yields with Pd(acac) being superior. With
heterogeneous Pd/C, no significant amount of product was
detected. Without metal, no transfer hydrogenolysis occured at all.
The testing of different acid co-catalysts made apparent that only
sulfonic acids like methane- or para-toluenesulfonic acid led to
high product yields. When weaker Brønsted acids were applied,
a
possible intermediate of the reaction was identified:
Esterification of formic acid with 1-phenylethanol gave
-phenylethyl formate (3), the role of this species will be
1
discussed below. It was found that the transfer hydrogenolysis
proceeds smoothly also at lower temperatures, although a
prolongation of the reaction time was required in some cases to
obtain a good product yield. The reaction was then further
[a]
B. Ciszek, Prof. Dr. I. Fleischer
Institute of Organic Chemistry
Eberhard Karls University of Tuebingen
Auf der Morgenstelle 18, 72076 Tuebingen
E-mail: ivana.fleischer@uni-tuebingen.de
optimized in terms of loading in Pd(acac)
2
,
L1 and
Supporting information for this article is given via a link at the end of
the document.
methanesulfonic acid as well as for the amount of formic acid.
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201801466
COMMUNICATION
partially oxidized (two signals at d = 62.07 and 24.15 ppm;
[
Pd] source (1 mol%)
Intermediate:
L1 (4 mol%)
integration resulted in a ratio of ligand L1 to ligand oxide L1-(O)
2
O
acid co-catalyst (16 mol%)
of 3:1, see SI for details), while batch B contained pure L1. To
investigate this phenomenon, we conducted a series of transfer
hydrogenolysis experiments in which the applied ligand was
OH
HCOO
H (2 equiv)
H
O
H
Ph
1,2-dichloroethane, 18 h
Ph
Ph
2
varied in presence of L1-(O) as well as the phosphine oxide was
1
a
2a
3
varied in presence of L1 (for details see the Supporting
Information). In both series of experiments, PdII and Pd0
Ligand:
[Pd] source:
Acid:
precursors were used. As expected, none of the other applied
✓
✓
✓ Pd(acac)2
✓✓ MeSO H, pTolSO H
t
3
3
P( Bu)2
PdCl2, Pd(OAc)2,
✗ TFA, PhCOOH,
phosphine/L1-(O)
product. When applying L1 with different phosphine oxides no
effect on the reaction outcome was observed with Pd(dba) as
precursor, however, starting from Pd(acac) product yields
2
mixtures yielded significant amounts of
t
P( Bu)2
Pd(dba)2
(PhO) P(O)OH
2
✗
Pd/C, no metal
✗ no acid
L1
2
2
Scheme 2. Optimization of reaction conditions for the palladium-catalyzed
transfer hydrogenolysis of 1-phenylethanol.
differed depending on the introduced phosphine oxide. It was
apparent that the presence of a phosphine oxide with certain
structural properties was beneficial for the transfer hydrogenation
II
if a Pd precursor was used, possibly through a stabilizing effect
in the formation of the active catalyst.[15]
Next, the applied phosphine ligand was varied (Table 1). To
our surprise, only one of the tested ligands, namely L1, led to a
significant amount of product. The use of other bidentate
phosphines either gave no or only minor amounts of product 2a.
Monodentate phosphines like L7 or L8 also showed only marginal
activity. Exclusion of any ligand led to a breakdown of the desired
reactivity. When searching for reasons for the observed exclusive
ligand effect, we came across reports about L1 being able to form
a relatively stable palladium-phosphine-hydride complex under
With the optimized reaction conditions at hand and having
gained an insight in the role of the phosphine oxide, a thorough
substrate screening was performed (Table 2). The reactivity of
various substituted aryl ethanols was investigated, showing a
strong influence of the electronic and steric properties of the
introduced substituents. Methyl-substituted 1-phenylethanols
provided the respective products 2b and 2m in good to very good
yields of 47% and 82% (entries 2, 11). The transformation of para-
and ortho-methoxy-substituted 1-phenylethanols led only to
traces of the desired products, due to an acid-induced elimination
and subsequent dimerization of these electron-rich substrates
similar conditions which is able to isomerize C-C double bonds as
well as to perform the carbonylation of alkenes.[14] We assume
this to be an initial hint towards the structure of the catalytically
active species in the transfer hydrogenolysis reported in this work.
(
entries 3, 12). However, the meta-substituted isomer was
Table 1. Ligand effects in the palladium-catalyzed transfer hydrogenolysis of
deoxygenated in a good yield of 65% (entry 14). 1,2,3,4-
Tetrahydro-1-naphthol as a more rigid derivative of the model
substrate 1a also was converted to the corresponding alkane in
quantitative yield (entry 17). 1-Phenylethanols decorated with F,
Cl and Br were tolerated well in the reaction, no dehalogenation
occured despite using Pd as catalyst. The corresponding para-
halogenated ethylbenzenes were obtained in good to excellent
yields of 62% to 94% (entries 4-6). 1-(para-Iodophenyl)ethanol
was deoxygenated only in trace amounts (entry 7), the
intermediary formate was observed exclusively. Introduction of an
ester group hindered the transfer hydrogenolysis, the
corresponding alkane was obtained in 35% yield, accompanied
by a great amount of non-reacted formate intermediate (entry 8).
Strongly deactivated substrates were not converted, only the
intermediate formates were observed (see the SI for details).
Biphenyl or naphthyl groups in the substrate resulted in excellent
isolated product yields (entries 9, 15, 16).
1
-phenylethanol (1a).[a]
Pd(acac)2 (1 mol%)
ligand (4 mol%)
MeSO H (16 mol%)
3
OH
Ph
HCOOH (2 equiv)
H
1,2-dichloroethane,
00 °C, 18 h
Ph
1
1
a
2a
PR2
PR2
t
P( Bu)
2
2
O
Ph P
2
PPh2
t
n
P( Bu)
L1: >99% yield[b]
L2 n = 1: 0%
L3 n = 2: 0%
L4 R = Ph: 0%
L5 R = Bu: 0%
PPh2
t
R
R
Fe
P
R
(tBu) P
2
L6: 3%
L7 R = Ph: 1%[c]
L8 R = Me: 4%[c]
We were delighted to see that primary benzylic alcohols were
also transformed to the corresponding toluenes in nearly
[
(
a] General reaction conditions: 1a (1.0 mmol), Pd(acac)
4 mol% if not stated otherwise), MeSO
,2-dichloroethane (3 mL), stirred for 18 h at 100 °C in a Schlenk pressure tube.
b] Yields determined via quantitative GC-FID with n-pentadecane as internal
2
(1 mol%), ligand
3
H (16 mol%), formic acid (2.0 mmol) in
quantitative yields, excluding
hydrogenation pathway through
entries 10, 11). Diphenylmethanol, which is unable to form an
a
possible elimination-
1
[
styrene derivatives
standard. [c] Amount of applied ligand: 8 mol%.
(
alkene via dehydration, was also transformed in a very good yield
of 87% (entry 19). 4-Hydroxy-4-phenylbutan-2-one as a simple
aldol product was reduced only at the benzylic hydroxy function
in a good yield of 59% (entry 18). Tertiary benzylic alcohols were
deoxygenated in good to quantitative yields (entries 20 - 22).
In the further course of our studies on the homogeneous
transfer hydrogenolysis, the use of a new batch (B) of L1 from a
different supplier led to significantly lower product yields for
several substrates than the previously used batch (A). 31P-NMR
analysis of both samples showed batch A was delivered already
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201801466
COMMUNICATION
Acetophenone was deoxygenated in only 21% (not shown),
demonstrating the clear selectivity of the catalyst for benzylic
alcohols over benzylic ketones.
elimination to the corresponding styrene derivative is not required.
In several cases, a formate intermediate was obtained as side-
product and it was also detected by GC in the transformation of
successful substrates. In order to confirm its formation as reaction
Table 2. Substrate scope of the palladium-catalyzed transfer hydrogenolysis of
intermediate, formate
3
was subjected to the transfer
benzylic alcohols.
Pd(acac)2 (1 mol%)
hydrogenolysis conditions (Table 3). Notably, the reaction took
place in the absence of formic acid (entry 1). An additional
equivalent of HCOOH led to a higher yield (entry 2). Furthermore,
we followed the reaction of the acetate analogue 4 as substrate
under standard conditions (entry 3). Here, a transesterification to
the formate was observed on GC-FID before any product
formation occurred. From these results we concluded that the
formate ester is in fact the reaction intermediate.
L1 (4 mol%)
L1-(O)2 (1.3 mol%)
MeSO H (16 mol%)
3
OH
HCOOH (2 equiv)
H
R2
R1
R2
1
,2-dichloroethane,
00 °C, 18 h
Ar
Ar
R1
1
1
2
Yield
%][b]
Entry[a]
product
Ar
R1
R2
[
Table 3. Transformation of the possible reaction intermediate and a respective
analogue in the transfer hydrogenolysis.
1
2
3
4
5
6
7
8
9
2a
2b
2c
2d
2e
2f
Ph
Me
Me
Me
Me
Me
Me
Me
Me
Me
H
H
H
>99
47
Pd(acac)2 (1 mol%)
4-Me-C
4-MeO-C
4-F-C
4-Cl-C
4-Br-C
4-I-C
4-CO Me-C
6
H
4
L1 (4 mol%)
L1-(O)2 (1.3 mol%)
R
traces[c]
70
6
H
4
H
MeSO H (16 mol%)
3
O
O
H
HCOOH (x equiv.)
6
H
4
H
1
,2-dichoroethane
00 °C , 18 h
Ph
Ph
1
6
H
4
H
94
3: R = H
4: R = CH
3
2a
6
H
4
H
62
Entry[a]
R
H
H
formic acid (x equiv.)
Yield (2a) [%][b]
traces[d]
2g
2h
2i
6
H
4
H
35[d]
1
2
3
0
1
2
65
91
88
2
6
H
4
H
4-Ph-C
Ph
6
H
4
H
>99 (96)
98
1
1
1
1
1
1
1
1
1
1
2
2
2
0
1
2
3
4
5
6
7
8
9
0
1
2
2j
H
CH
3
2k
2l
4-Me-C
2-MeO-C
2-Me-C
3-MeO-C
6
H
4
H
H
93
[
(
a] General reaction conditions: Ester substrate (1.0 mmol), Pd(acac)
2
1.0 mol%), L1 (4.0 mol%), L1-(O)
2
(1.3 mol%) (partially oxidized L1 was
H
4
4
Me
Me
Me
Me
Me
H
traces[c]
82
6
applied), MeSO
(
3
H (16 mol%), formic acid (0-2.0 mmol) in 1,2-dichloroethane
3.0 mL), stirred for 18 h at 100 °C in a Schlenk pressure tube. [b] Yields
2m
2n
2o
2p
2q
2r
6
H
4
H
determined via quantitative GC-FID with n-pentadecane as internal standard.
6
H
H
65
In order to establish the homogeneous nature of the catalyst
system, a series of experiments was conducted.[16] The reaction
was monitored over time by withdrawing samples from the
reaction mixture (for details see the SI). The obtained reaction
profile exhibited an initial linear shape without significant induction
period typical for heterogeneous catalysis. Furthermore,
qualitative catalyst poisoning experiments were conducted using
dibenzo[a,e]cyclo-octatetraene (dct)[17] and mercury[18] (for details
2-naphthyl
H
(94)
(95)
>99
59
1-naphthyl
H
Ph
Ph
Ph
Ph
Ph
Ph
-(CH
2
)
3
-
H
CH C(O)CH
2
3
H
2s
2t
Ph
Me
Ph
Ph
H
87
see the SI). Findings from these experiments propose
homogeneous catalyst as both additives inhibited the reaction.
a
Me
Me
Ph
75
As Pd nanoparticles (NPs) should be excluded as the active
catalyst, commercially available Pd(0) EnCat® 30NP was tested
as metal source (Table 4). While under the optimized reaction
conditions a yield of 2a of 35% was obtained (entry 1), no product
was formed without ligand (entry 2). Interestingly, doubling the
amount of L1 led to a higher yield of 61% (entry 3). From this we
deduce that Pd NPs are not the active catalyst species, but may
serve as a reservoir from which metal can leach into solution to
form the active homogeneous catalyst. Such cases of so-called
"cocktail catalysis" were already described in the literature.[19]
2u
2v
66
(>99)
[
a] General reaction conditions: Alcohol (1.0 mmol), Pd(acac)
2
(1.0 mol%), L1
(1.3 mol%) (partially oxidized L1 was applied), MeSO
16 mol%), formic acid (2.0 mmol) in 1,2-dichloroethane (3.0 mL), stirred for
8 h at 100 °C in a Schlenk pressure tube. [b] Yields determined via quantitative
(
(
4.0 mol%), L1-(O)
2
3
H
1
GC-FID with n-pentadecane as internal standard. Isolated yields in brackets. [c]
Acid-induced dehydrative elimination/dimerization of the aryl ethanol observed.
d] Large amounts of unreacted 1-arylethyl formate were detected after the
[
reaction.
The overall reaction might proceed via different intermediates.
The results obtained from the substrate screening indicate that
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201801466
COMMUNICATION
Table 4. Nanoparticles as Pd source in the transfer hydrogenolysis.
In summary, a new palladium-based homogeneous catalyst
for the transfer hydrogenolysis of benzylic alcohols was
developed. Formic acid was used as a convenient and potentially
sustainable reductant. A series of primary, secondary and tertiary
alcohols was successfully converted to the corresponding alkyl
arenes. The influence of aryl substituents on the reactivity was
evaluated. We found that formic acid not only acts as reductant,
but it also activates the substrate via esterification. An essential
finding of our investigations was the positive effect of partially-
oxidized ligand on the activity of PdII catalyst precursors. In
addition, the homogeneous nature of the active catalyst was
proposed.
[
Pd] NPs (1 mol%)
L1 (x mol%)
L1-(O) (0.3x mol%)
MeSO H (y mol%)
HCOOH (2 equiv)
2
3
OH
H
1
1
,2-dichloroethane,
00 °C, 18 h
Ph
Ph
1
a
2a
Entry[a]
x
y
Yield (2a) [%][b]
1
2
3
4
4.0
0
16
16
16
32
35
3
8.0
8.0
61
66
Acknowledgements
[a] General reaction conditions: 1a (1.0 mmol), [Pd] NPs (1 mol%), L1
Financial support from Fonds der Chemischen Industrie (Liebig
fellowship IF), the Deutsche Bundesstiftung Umwelt (PhD
fellowship BC) and the University of Tübingen (Institutional
(
x mol%), L1-(O) (0.3x mol%) (partially oxidized L1 was applied), acid co-
2
catalyst (y mol%), formic acid (2.0 mmol) in 1,2-dichloroethane (3 mL), stirred
for 18 h at 100 °C in a closed schlenk pressure tube. [b] Yields determined via
quantitative GC-FID with n-pentadecane as internal standard.
Strategy
of the
University
of Tübingen: Deutsche
Forschungsgemeinschaft ZUK 63) is gratefully acknowledged.
Based on our findings in the reaction optimization, the
transformation of the formate intermediate and the poisoning
experiments, we propose the following mechanism for the transfer
hydrogenolysis reaction: The formation of the cationic palladium-
hydride complex A is initiated either by the acid co-catalyst or
formic acid while the substrate 1a is transformed into the formate
intermediate 3 via esterification (Scheme 3). We do not exclude
an equilibrium with styrene for substrates that allow it. The
reduction occurs via a six-membered distorted chair-like transition
state B in which palladium coordinates the carbonyl group of the
formate. The hydride is transferred from the metal center to the
substrate, releasing the desired deoxygenated product 2a and a
formate anion which ligates to the palladium center. Palladium-
formato complex C regenerates the active catalyst by release of
Keywords: Hydrogenolysis • Palladium • Formic acid •
Homogeneous catalysis • Alcohols
[
1]
2]
J. M. Herrmann, B. König, Eur. J. Org. Chem. 2013, 2013, 7017-7027.
D. H. R. Barton, S. W. McCombie, J. Chem. Soc., Perkin Trans. 1 1975,
[
1574-1585.
[3]
[4]
Padoa, Ponti, Gazz. Chim. Ital. 1907, 37, 105.
B. H. Gross, R. C. Mebane, D. L. Armstrong, Appl. Catal., A. 2001, 219,
281-289.
[5]
a) R. J. Rahaim, R. E. Maleczka, Org. Lett. 2011, 13, 584-587; b) A.
Volkov, K. P. J. Gustafson, C.-W. Tai, O. Verho, J.-E. Bäckvall, H.
Adolfsson, Angew. Chem. 2015, 127, 5211-5215; Angew. Chem. Int. Ed.
2015, 54, 5122-5126; c) H. Wang, L. Li, X.-F. Bai, J.-Y. Shang, K.-F.
Yang, L.-W. Xu, Adv. Synth. Catal. 2013, 355, 341-347; d) N. C.
Mamillapalli, G. Sekar, Adv. Synth. Catal. 2015, 357, 3273-3283; e) N.
C. P. Araujo, A. F. Brigas, M. L. S. Cristiano, L. M. T. Frija, E. M. O.
Guimaraes, R. M. S. Loureiro, J. Mol. Catal. A: Chem. 2004, 215, 113-
120; f) M. Tamaki, G. X. Han, V. J. Hruby, J. Org. Chem. 2001, 66, 3593-
2
CO , closing the catalytic cycle.
OH
[
Pd(II)]
[Pd(0)]
O
Ph
HA
1a
3596; g) C.-B. Yu, Y.-G. Zhou, Angew. Chem. 2013, 125, 13607-13610;
H
OH
O
Angew. Chem. Int. Ed. 2013, 52, 13365-13368.
A
[
6]
7]
N. Kalutharage, C. S. Yi, J. Am. Chem. Soc. 2015, 137, 11105-11114.
N. Thakar, N. F. Polder, K. Djanashvili, H. van Bekkum, F. Kapteijn, J. A.
Moulijn, J. Catal. 2007, 246, 344-350.
O
H
CO , H
2
[
P
P
Ph
Pd
H
3
OCO
Ph
[8]
a) S. C. A. Sousa, T. A. Fernandes, A. C. Fernandes, Eur. J. Org. Chem.
2016, 2016, 3109-3112; b) M. Yasuda, Y. Onishi, M. Ueba, T. Miyai, A.
A
Baba, J. Org. Chem.2001, 66, 7741-7744.
[9]
J. S. Bajwa, J. Slade, O. Repič, Tetrahedron Lett. 2000, 41, 6025-6028.
[
10] a) X. Liu, G. Lu, Y. Guo, Y. Guo, Y. Wang, X. Wang, J. Mol. Catal. A:
Chem. 2006, 252, 176-180; b) J. Feng, C. Yang, D. Zhang, J. Wang, H.
Fu, H. Chen, X. Li, Appl. Catal., A. 2009, 354, 38-43; c) S. Sawadjoon,
A. Lundstedt, J. S. M. Samec, ACS Catal. 2013, 3, 635-642.
P
P
Ph
Pd
O
H
H
O
O
O
C
Pd
P
[
11] a) A. K. Singh, S. Singh, A. Kumar, Catal. Sci. Technol. 2016, 6, 12-40;
b) A. Álvarez, A. Bansode, A. Urakawa, A. V. Bavykina, T. A. Wezendonk,
M. Makkee, J. Gascon, F. Kapteijn, Chem. Rev. 2017, 117, 9804-9838;
c) D. Mellmann, P. Sponholz, H. Junge, M. Beller, Chem. Soc. Rev. 2016,
P
H
B
H
45, 3954-3988.
2
a
[
12] T. Maegawa, A. Akashi, H. Sajiki, Synlett 2006, 2006, 1440-1442.
13] a) P. H. Gehrtz, V. Hirschbeck, I. Fleischer, Chem. Commun. 2015, 51,
Ph
[
1
2574-12577; b) V. Hirschbeck, P. H. Gehrtz, I. Fleischer, J. Am. Chem.
Scheme 3. Proposed reaction mechanism for the palladium-catalyzed transfer
Soc. 2016, 138, 16794-16799.
hydrogenolysis with formic acid as reductant.
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201801466
COMMUNICATION
[
14] a) G. R. Eastham, R. P. Tooze, B. T. Heaton, J. A. Iggo, R. Whyman, S.
Zacchini, Chem. Commun. 2000, 609-610; b) G. R. Eastham, R. P.
Tooze, M. Kilner, D. F. Foster, D. J. Cole-Hamilton, J. Chem. Soc.,
Dalton Trans. 2002, 1613-1617; c) V. Goldbach, L. Falivene, L. Caporaso,
L. Cavallo, S. Mecking, ACS Catal. 2016, 6, 8229-8238.
[16] a) R. H. Crabtree, Chem. Rev. 2012, 112, 1536-1554; b) J. A. Widegren,
M. A. Bennett, R. G. Finke, J. Am. Chem. Soc. 2003, 125, 10301-10310.
[17] D. R. Anton, R. H. Crabtree, Organometallics 1983, 2, 855-859.
[18] N. T. S. Phan, M. Van Der Sluys, C. W. Jones, Adv. Synth. Catal. 2006,
348, 609-679.
[15] S. E. Denmark, R. C. Smith, S. A. Tymonko, Tetrahedron 2007, 63, 5730-
[19] V. P. Ananikov, I. P. Beletskaya, Organometallics 2012, 31, 1595-1604.
5738.
This article is protected by copyright. All rights reserved.

Chemistry - A European Journal
10.1002/chem.201801466
COMMUNICATION
Entry for the Table of Contents
COMMUNICATION
Hydrogenolytic cleavage of benzylic
C-OH bonds was achieved using a
homogeneous Pd-based catalyst and
formic acid as both reducing and
activating agent.
Benjamin Ciszek, Ivana Fleischer*
H
O
Page No. – Page No.
H
OH
Homogeneous Palladium-Catalyzed
Transfer Hydrogenolysis of Benzylic
Alcohols Using Formic Acid as
Reductant
P
P
Pd
H
O
H
OH
C
O
O
H
This article is protected by copyright. All rights reserved.