Catalytic Ester Metathesis Reaction and Its Application to Transfer Hydrogenation of Esters

Article abstract of DOI:10.1021/acscatal.6b00827

We report a Ru-complex-catalyzed ester metathesis reaction where an unsymmetrical ester such as ethyl hexanoate can be transformed to a mixture of starting material, hexyl ethanoate, ethyl acetate, and hexyl hexanoate in equal proportions, as expected from a classical metathesis reaction with 0.2 mol % catalyst. A 20× excess of low boiling alcohol, such as ethanol, allows for the transfer of an acyl moiety to the sacrificial low boiling ethyl acetate product, while significantly increasing the functional group tolerance and substrate scope; yields of alcohols can reach 90%, which represents an attractive alternative to current high H₂ pressure hydrogenation protocols for Ru-based ester reduction chemistry. Both reactions have not been reported previously in the field of Ru-catalyzed transformations of the ester functionality.

Full text of DOI:10.1021/acscatal.6b00827

Letter
pubs.acs.org/acscatalysis
Catalytic Ester Metathesis Reaction and Its Application to Transfer
Hydrogenation of Esters
Abhishek Dubey and Eugene Khaskin*
Okinawa Institute of Science and Technology Graduate University, 1919-1 Tancha, Onna-son, Kunigami-gun, Okinawa, 904-0495
Japan
S
* Supporting Information
ABSTRACT: We report a Ru-complex-catalyzed ester meta-
thesis reaction where an unsymmetrical ester such as ethyl
hexanoate can be transformed to a mixture of starting material,
hexyl ethanoate, ethyl acetate, and hexyl hexanoate in equal
proportions, as expected from a classical metathesis reaction
with 0.2 mol % catalyst. A 20× excess of low boiling alcohol,
such as ethanol, allows for the transfer of an acyl moiety to the
sacrificial low boiling ethyl acetate product, while significantly
increasing the functional group tolerance and substrate scope;
yields of alcohols can reach 90%, which represents an attractive
alternative to current high H2 pressure hydrogenation protocols
for Ru-based ester reduction chemistry. Both reactions have not
been reported previously in the field of Ru-catalyzed trans-
formations of the ester functionality.
KEYWORDS: metathesis, esters, pincer complex, transfer hydrogenation, homogeneous catalysis, ruthenium
he efficient, Ru-pincer-complex-catalyzed alcohol coupling
their low catalyst loadings. However, the full reactivity scope for
both the forward and reverse reaction is still absent as ester
hydrogenation under H2 pressure is reported in preference to
the alcohol coupling reaction; this may be because the latter
reaction is unselective unless a primary and secondary (or
benzylic/aliphatic) alcohol are used or the homocoupling of
one alcohol is the desired outcome.
Since most of the catalysts are capable of promoting both
ester hydrogenation and alcohol coupling, we decided to see if
we could observe the simultaneous reduction and oxidation of
different parts of the ester molecule in one pot. According to
Scheme 1, the overall reaction should be accomplished with
hydrogen gas as a cocatalyst. However, only high pressures of
H2 had been reported so far for ester hydrogenation, and under
these conditions alcohols should be obtained as the exclusive
product.
Gratifyingly, when exploring the reactivity of an unsym-
metrical ester, ethyl hexanoate 10, with commercially available
catalysts 2−4, we found that upon the addition of activating
base, the catalyst was capable of reacting and rearranging the
ester without the need of any addition of hydrogen cocatalyst.
After 16 h of heating at 80 °C in the presence of 1 mol % of
catalyst 2 (Table 1), 1HNMR in C6D6 (Supporting Information
(SI), Figures S1 and S2) confirmed complete scrambling of the
starting materials to a statistical equilibrium of products
T
reaction to form esters with the concurrent release of H2
gas was first reported in 2005;1 the reverse reaction, or
hydrogenation of esters to form alcohols, was published soon
afterward (Scheme 1).2 Today, both reactions are well-
Scheme 1. Alcohol Coupling Reaction To Form Esters and
Ester Hydrogenation under H2 Pressure
established processes of increasing industrial relevance.3,4
Oxidative alcohol coupling is attractive as it represents an
atom-economical, green-chemistry alternative to previous
synthetic routes, with H2 being the only byproduct that leaves
as gas during the reaction. Likewise, reduction of esters to
alcohols normally requires stoichiometric reagents such as
lithium aluminum hydride; however, reduction with H2, despite
the high pressures (often 50 bar) and specialized equipment
required, offers an atom economical way of obtaining the
desired alcohol. Some of the applications of this attractive
chemistry include reducing triglycerides into important
component alcohols5 as well as decomposing polyesters to
alcohol components.6
We summarize some of the complexes relevant to ester
reactivity, along with the date and application in Figure 1.1,7−14
It can be seen that last-generation Ru catalysts are notable for
Received: March 22, 2016
Revised: April 25, 2016
© XXXX American Chemical Society
3998
DOI: 10.1021/acscatal.6b00827
ACS Catal. 2016, 6, 3998−4002

ACS Catalysis
Letter
a
Table 1. Optimization of Ester Metathesis
entry
catalyst (X mol %)
efficiency
1
2
3
4
5
6
7
8
2 (1 mol %)
94%
74%
89%
99%
99%
99%
91%
trace
2 (0.5 mol %)
2 (0.2 mol %)
3 (1.0 mol %)
3 (0.5 mol %)
3 (0.2 mol %)
3 (0.1 mol %)
4 (1.0 mol %)
a
Amounts based on 1 mg of catalyst in 3 mL of toluene in an 11 mL
closed container.
a
Table 2. Esters Tested in Metathesis and Products
Figure 1. A brief overview of pincer catalysts and their application.
(Scheme 2), meaning that true metathesis had occurred.
Organometallic mediated metathesis has previously been
observed with alkenes,15 alkynes,16 and alkanes,17,18 but not
with esters.
Scheme 2. Metathesis Reaction with Ethyl Hexanoate 10
Although all the products are obtained as a statistical mixture,
the boiling point differences do make isolation of a particular
desired product a possibility. The reaction may also find
applicability in the flavoring industry where changing an ester
composition results in different scent, and a mixture of simple
esters can now react not only via transesterification but also via
metathesis scrambling. To the best of our knowledge, the
transformation is unique; thus, the full scope of its practicality
may become more apparent in the future.
Screening of the catalysts 2−4 for the ester metathesis
identified 3 as the most effective (Table 1, entry 6). Catalyst 2
was also effective but gave slightly poorer yields (Table 1,
entries, 1−3), whereas catalyst 4 failed to give a metathesis
product. A short screening showed that toluene/benzene was
the only solvent appropriate for the reaction, as THF and
CH3CN gave no conversion and reactions in hexane performed
poorly, likely due to solubility of the catalyst. Other activating
bases, such as KHMDS, had no effect on outcomes.
With the optimal conditions in hand (Table 1, entry 6), we
decided to investigate the applicability of the reaction to various
esters that contained benzylic, olefin, halogen, and ether
moieties. Currently, functional groups or compounds that can
form a conjugated, stabilized double bond with the acyl moiety
are not tolerated with the catalysts we tested. Importantly,
however, alkyl (Table 1), aryl (Table 2, entry 5), and mixed
a
Amounts based on 1 mg of catalyst in 3 mL of toluene in an 11 mL
closed container.
alkyl-aryl (Table 2, entries 3−4) esters were all scrambled
equally well during the reaction at 0.2 mol % catalyst loadings.
Control reactions without either the activating base, or catalyst,
gave no conversion. Metathesis efficiency was determined by
the difference between unsymmetrical products (i.e., 10 and 11
for Table 1; see GC traces in the SI for each entry).
Methyl ester 14 is not active in the reactions (Table 2, entry
1−2), which may be due to catalyst inactivation by carbonyls
formed in situ from the methanol. After long reaction times,
benzyl acetate 19 (entry 4) is selective for transformation to
3999
DOI: 10.1021/acscatal.6b00827
ACS Catal. 2016, 6, 3998−4002

ACS Catalysis
Letter
ethyl benzoate 20, reflecting the greater thermodynamic
stability of an ester with an acyl group in the benzylic position.
From NMR experiments, it was found that the reaction will
proceed at temperatures as low as 50 °C with 19; however, a
temperature of 80 °C and a 16 h heating period were used as
general conditions. We determined efficiency to be at ∼99%
when the starting material was at ≤25% of the initial amount,
and the product was correspondingly ≥25%. The amounts may
not exactly end up at 25% if the product is more
thermodynamically stable than the starting material.
The reaction with isopropanol as the sacrificial H2 donor
resulted in 43% yield of hexanol (40TON; Table 3, entry 9),
which is less than our maximum yields obtained with ethanol.
In the case of a secondary alcohol, ester metathesis with
isopropanol cannot take place, and the cost of making acetone
may be greater than the energy gained by making primary
alcohols from an ester.
Interestingly, other substrates that were not active in ester
metathesis are also converted to alcohols, often in good yields
(Table 4, entry 1, 5, 11, 14−16).
In light of the excellent results obtained with the metathesis
reaction, we decided to see if it can be applied to make a single
product selectively. There are a number of reports on transfer
hydrogenation of ketones, aldehydes, alkynes, olefins, and
imines.19 We are only aware of one report from the Nikonov
group in 2015, where transfer hydrogenation of esters is
reported with a Ru half-sandwich complex and isopropanol as
the sacrificial hydrogen donor.20 Good activity (∼20 TON) is
obtained only with secondary trifluoroacetate esters.
Transfer hydrogenation should be possible if an excess of
primary alcohol, such as ethanol is introduced together with the
ester. The advantages of ethanol and the ethyl acetate
byproduct include low boiling points and cost due to them
being common solvents and biofuels. Formally, the reaction
would be ester metathesis, but technically transfer dehydrogen-
ation (TH) of an ester would have occurred because products
can be isolated after removing solvent (toluene and ethanol)
and byproducts (ethyl acetate) under vacuum. By avoiding high
pressures of H2 gas, TH of esters has the advantages of safety
and ease of operation.
Our protocol was also applicable in the hydrogenation of
triglycirides, hydrogenations of which with H2 pressure have
recently been reported.5 The long-chain alcohols obtained from
these natural products are of commercial interest and are
normally obtained by stoichiometric reduction; milder
alternatives involving H2 or transfer hydrogenation are highly
desired. Along with the alcohol product (∼57% for tripalmitin),
a significant amount of the intermediate ethyl ester of palmitate
is also observed. The yield of the alcohol here is limited by the
solubility of the starting triglyceride in the toluene solvent.
Attempts to use other solvents or perform a neat reaction gave
worse outcomes.
We show our working mechanistic hypotheses for this
reaction in Figure 2. The metathesis reaction of benzyl acetate
1
19 was followed by HNMR, confirming the formation of the
three products. We observed that the rate of appearance of
symmetrical metathesis products, which form at an equal rate to
each other, is more rapid than the formation of the more
thermodynamically stable, unsymmetrical ethyl benzoate 20
(SI, Figures S7−S16). According to pathway I, symmetrical
products are formed initially, and the unsymmetrical product
can be generated when the catalyst subsequently reacts with
these symmetrical products, or by the trans-esterification (TE)
reaction active in the presence of base (TE is slower than
metathesis and is discussed in more detail in the SI). An acyl Ru
species that is suggested by the mechanism also helps explain
the rapid deactivation of the catalyst when methyl esters are
involved; it is unlikely that this type of organometallic species
(RCO-Ru) would be tolerant of other functional groups as well.
Acyl-Ru species have been isolated and characterized for
monometallic Ru complexes.21
As we had expected, using ethanol is a viable strategy for the
hydrogenation of esters via our procedure. However, the
catalyst loading has to be increased to 1 mol % in order to get
acceptable yields as reactivity with ethanol lowers the rate of
useful reactions with the ester substrate (Table 3). At 20 equiv
a
Table 3. Optimization of Transfer Hydrogenation of Esters
Pathway II may be active in TH with ethanol, where free H2
from excess alcohol in the closed reaction system opens up an
opportunity for Ru dihydride complex formation in situ. The
activity of some substrates in hydrogenation that proved to be
inactive in metathesis, as well the dramatic improvement in the
activity of catalyst 4 (Table 4, entry 5), suggests that both
pathways I and II can be active in transfer hydrogenation of
esters, and it is likely that only pathway I is active in one-
component ester metathesis. In-situ transfer hydrogenation in a
reaction between a primary ester and a secondary alcohol has
been hinted at by an earlier Milstein report where trace ketone
byproducts were obtained; however, the authors focused on
generating secondary alcohol esters, and because only sym-
metrical primary esters were used as reactants, metathesis could
also not be observed.22 Presumably, this reaction proceeded via
pathway II (Figure 2), but acyl group reduction could not be
obtained due to the absence of sacrificial primary alcohols.
The current work shows that it is possible to think of the
ester functionality as an easily modifiable moiety where
metathesis can occur under the right catalysis conditions.
Until now, this was thought of as a two-pot reaction that
required different conditions and instrumental setups for each
entry
catalyst (X mol %)
H2 source (equiv )
yield (%)
1
2
3
4
5
6
7
8
9
2 (1 mol %)
2 (1 mol %)
2 (1 mol %)
3 (0.2 mol %)
3 (1 mol %)
4 (1 mol %)
3 (1 mol %)
3 (1 mol %)
3 (1 mol %)
ethanol (5 equiv)
57
65
71
60
89
70
20
18
43
ethanol (10 equiv)
ethanol (20 equiv)
ethanol (20 equiv)
ethanol (20 equiv)
ethanol (20 equiv)
benzyl alcohol (10 equiv)
benzyl alcohol (20 equiv)
iso-propanol (20 equiv)
a
Amounts based on 1 mg of catalyst in 3 mL of toluene in an 11 mL
closed container.
of ethanol to substrate, ethyl hexanoate is converted to hexanol
at 89% yield (Table 3, entry 5) after 16 h at 80 °C. Fewer
equivalents of ethanol led to lower conversions after 16 h of
reaction time. Interestingly, catalyst 4, which was very poorly
active in metathesis, gave alcohol product, albeit to a lesser
extent than 3. The lower reactivity in entries 7−8 is likely the
result of product inhibition as aromatic benzylic alcohol and
esters outcompete the aliphatic ester for binding to the catalyst.
4000
DOI: 10.1021/acscatal.6b00827
ACS Catal. 2016, 6, 3998−4002

ACS Catalysis
Letter
Table 4. Substrates Tested for Transfer Hydrogenation and
e
Yields of Alcohol
Figure 2. Working mechanistic hypotheses.
sacrificial substrate. We hope that this report spurs further
interest and research in the reduction of esters via Ru-catalyzed
ester scrambling as well as investigation of new catalysts23 that
display a wider functional group tolerance.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acscatal.6b00827.
■
S
Stoichiometric/kinetic NMR experiments, mechanistic
discussion, GC/FID traces, and DFT calculations (PDF)
AUTHOR INFORMATION
Corresponding Author
*E-mail: eugene.khaskin@oist.jp.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
The authors wish to acknowledge the Okinawa Institute of
Science and Technology for full funding of the project.
■
REFERENCES
■
(1) Zhang, J.; Leitus, G.; Ben-David, Y.; Milstein, D. J. Am. Chem. Soc.
2005, 127, 10840−10841.
(2) Zhang, J.; Leitus, G.; Ben-David, Y.; Milstein, D. Angew. Chem.,
Int. Ed. 2006, 45, 1113−1135.
(3) Trincado, M.; Banerjee, D.; Gruetzmacher, H. Energy Environ. Sci.
2014, 7, 2464−2503.
(4) Werkmeister, S.; Junge, K.; Beller, M. Org. Process Res. Dev. 2014,
18, 289−302.
(5) Acosta-Ramirez, A.; Bertoli, M.; Gusev, D. G.; Schlaf, M. Green
Chem. 2012, 14, 1178−1188.
(6) Krall, E. M.; Klein, T. W.; Andersen, R. J.; Nett, A. J.; Glasgow, R.
W.; Reader, D. S.; Dauphinais, B. C.; Mc Ilrath, S. P.; Fischer, A. A.;
Carney, M. J.; Hudson, D. J.; Robertson, N. J. Chem. Commun. 2014,
50, 4884−4887.
(7) Kuriyama, W.; Matsumoto, T.; Ogata, O.; Ino, Y.; Aoki, K.;
Tanaka, S.; Ishida, K.; Kobayashi, T.; Sayo, N.; Saito, T. Org. Process
Res. Dev. 2012, 16, 166−171.
(8) Spasyuk, D.; Smith, S.; Gusev, D. G. Angew. Chem., Int. Ed. 2013,
52, 2538−2542.
(9) Spasyuk, D.; Gusev, D. G. Organometallics 2012, 31, 5239−5242.
(10) Chakraborty, S.; Dai, H.; Bhattacharya, P.; Fairweather, N. T.;
Gibson, M. S.; Krause, J. A.; Guan, H. J. Am. Chem. Soc. 2014, 136,
7869−7872.
(11) Junge, K.; Wendt, B.; Jiao, H.; Beller, M. ChemCatChem 2014, 6,
2810−2814.
a
Only substrate given with alcohol part of ester as product; presence
b
of metathesis isomers confirmed as side-products. Catalyst 4 was
used; conversion factor used to calculate yield based on hexanol*2;
products and remaining material were a mix of compounds where the
c
double bond isomerizes to the 2,3, and 4 positions of the chain. Yield
d
calculated based on alcohol obtained/2. Mixture of olefin isomers.
e
Amounts based on 1 mg of catalyst in 3 mL of toluene in an 11 mL
closed container; 20 equiv of ethanol except for entries 9 and 10 (30
equiv of ethanol).
step. The ability to carry out this reaction rapidly and efficiently
leads to some interesting applications such as an efficient ester
transfer hydrogenation reaction with ethanol acting as the
4001
DOI: 10.1021/acscatal.6b00827
ACS Catal. 2016, 6, 3998−4002

ACS Catalysis
Letter
(12) Sun, Y.; Koehler, C.; Tan, R.; Annibale, V. T.; Song, D. Chem.
Commun. 2011, 47, 8349−8351.
(13) Fogler, E.; Garg, J. A.; Hu, P.; Leitus, G.; Shimon, L. J. W.;
Milstein, D. Chem. - Eur. J. 2014, 20, 15727−15731.
(14) Filonenko, G. A.; Aguila, M. J. B.; Schulpen, E. N.; van Putten,
R.; Wiecko, J.; Mueller, C.; Lefort, L.; Hensen, E. J. M.; Pidko, E. A. J.
Am. Chem. Soc. 2015, 137, 7620−7623.
(15) Astruc, D. New J. Chem. 2005, 29, 42−56.
(16) Fuerstner, A.; Davies, P. W. Chem. Commun. 2005, 18, 2307−
2320.
(17) Vidal, V.; Theolier, A.; Thivolle-Cazat, J.; Basset, J.-M. Science
1997, 276, 99−102.
(18) Goldman, A. S.; Roy, A. H.; Huang, Z.; Ahuja, R.; Schinski, W.;
Brookhart, M. Science 2006, 312, 257−261.
(19) Wang, D.; Astruc, D. Chem. Rev. 2015, 115, 6621−6686.
(20) Lee, S.-H.; Nikonov, G. I. ChemCatChem 2015, 7, 107−113.
(21) Hirano, M.; Sato, H.; Kurata, N.; Komine, N.; Komiya, S.
Organometallics 2007, 26, 2005−2016.
(22) Gnanaprakasam, B.; Ben-David, Y.; Milstein, D. Adv. Synth.
Catal. 2010, 352, 3169−3173.
(23) A recent report on new Os-based catalysts that efficiently
catalyse hydrogenation of esters suggests that Ru may not be unique to
this transformation. See also refs 10 and 11. Spasyuk, D.; Vicent, C.;
Gusev, D. G. J. Am. Chem. Soc. 2015, 137, 3743−3746.
4002
DOI: 10.1021/acscatal.6b00827
ACS Catal. 2016, 6, 3998−4002