Base-Free Iridium-Catalyzed Hydrogenation of Esters and Lactones

Article abstract of DOI:10.1021/acscatal.6b00263

Half-sandwich iridium bipyridine complexes catalyze the hydrogenation of esters and lactones under base-free conditions. The reactions proceed with a variety of ester and lactone substrates. Mechanistic studies implicate a pathway involving rate-limiting hydride transfer to the substrate at high pressures of H₂ (≥50 bar).

Full text of DOI:10.1021/acscatal.6b00263

Letter
pubs.acs.org/acscatalysis
Base-Free Iridium-Catalyzed Hydrogenation of Esters and Lactones
Timothy P. Brewster,†,‡ Nomaan M. Rezayee,§ Zuzana Culakova,‡ Melanie S. Sanford,*,§
and Karen I. Goldberg*,‡
‡Department of Chemistry, University of Washington, Seattle, Washington 98195-1700, United States
§Department of Chemistry, University of Michigan, 930 North University Avenue, Ann Arbor, Michigan 48109, United States
S
* Supporting Information
ABSTRACT: Half-sandwich iridium bipyridine complexes catalyze the hydro-
genation of esters and lactones under base-free conditions. The reactions proceed
with a variety of ester and lactone substrates. Mechanistic studies implicate a
pathway involving rate-limiting hydride transfer to the substrate at high pressures of
H2 (≥50 bar).
KEYWORDS: ester, hydrogenation, acid, homogeneous catalysis, iridium
esonance-stabilized carbonyls are challenging to reduce
investigations of ester hydrogenation reactions catalyzed by 1
Rdue to the low electrophilicity of the carbonyl carbon. As are also described.17
1
Our initial studies focused on the use of catalyst 1 for the
a result, strongly basic stoichiometric reductants such as lithium
aluminum hydride are commonly employed to effect the
reduction of esters to alcohols.2 The utilization of dihydrogen
as the reductant would serve as an attractive, mild alternative.
Recent advances in this area have demonstrated the feasibility
of transition-metal-catalyzed ester hydrogenation. However, the
vast majority of these catalysts require the addition of basic
additives (most commonly alkoxide bases) to activate the
catalyst, or contain a basic site within the catalyst, which can
pose limitations with respect to functional group compati-
bility.1,3−9 To date, only a single catalyst system has been
reported to hydrogenate esters under neutral or acidic
conditions without prior activation of the catalyst.10−14 New
ester hydrogenation catalysts that can operate in the absence of
base would be a valuable complement to existing systems, as
they would allow for mild reduction in the presence of base-
sensitive functionalities.
In addition to synthetic applications, esters have been utilized
as intermediates in strategies for carbon recycling. Specifically,
formate esters serve as intermediates in a cascade hydro-
genation pathway from carbon dioxide to methanol.15 In this
system, the generation of the formate ester takes place under
acidic conditions and is thus poorly compatible with base-
driven ester hydrogenation catalysts. This provides further
motivation for the development of acid-tolerant ester hydro-
genation catalysts.
hydrogenation of ethyl acetate. Heating a 2 mM solution of 1 in
neat ethyl acetate under 60 bar H2 for 18 h in the absence of
any additives affords ethanol with a TON of 363 46 (Table
1).17 Catalysts 218 and 3,17 which are sterically similar to 1, but
a
Table 1. Catalyst Screen for Ethyl Acetate Hydrogenation
catalyst
TON
1
2
3
363 46
309 47
116
7
a
4 μmol catalyst in 2 mL of ethyl acetate (20.4 mmol), 60 bar H2, 18 h
at 120 °C. Average of 3 trials with standard deviation. Theoretical
maximum TON under these conditions = 5100. Small amounts of
diethyl ether (∼10%) are also observed.
contain different 4,4′-substituents on the bipyridine ligands
(Figure 1) were also examined. The highest turnover numbers
were obtained using 1, which contains electron-donating
methoxy substituents. An analogous trend was observed in
the hydrogenation of carboxylic acids with this series of
catalysts, suggesting that the reactions may be mechanistically
similar.17
This communication describes the use of an iridium catalyst,
[Cp*Ir(bpy-OMe)OH2][OTf]2 (1), (Cp* = pentamethylcy-
clopentadienide, bpy-OMe = 4,4′-dimethoxy-2,2′-bipyridine,
OTf = trifluoromethanesulfonate)16,17 for the hydrogenation of
esters and lactones without added base. In many cases, the
addition of a Lewis acid was found to enhance the TONs with
this catalyst. The optimization, substrate scope, and mechanistic
Received: January 26, 2016
Revised: March 21, 2016
© XXXX American Chemical Society
3113
DOI: 10.1021/acscatal.6b00263
ACS Catal. 2016, 6, 3113−3117

ACS Catalysis
Letter
saturation behavior is observed with respect to hydrogen
pressure (Figure 2B). Saturation is reached at approximately 50
bar H2, and a linear dependence on H2 pressure is observed up
to approximately 17 bar (Figure 2B).21 The dependence on
substrate concentration was determined utilizing hexyl formate
as the substrate (for ease in measurement of the concentrations
of reactant and products) in 1,2-dimethoxyethane (DME). The
reactions were run for 4 h at 100 °C under 60 bar H2.20 Under
these conditions, a linear dependence on substrate concen-
tration was observed (Figure 2C).
Figure 1. Ester hydrogenation catalysts.
Mechanistic investigations of the hydrogenation of ethyl
acetate with catalyst 1 were undertaken.19 The reactions were
conducted in neat ethyl acetate, and the % conversion and
TON were determined after 18 h of heating at 120 °C.20 The
rate of hydrogenation of ethyl acetate exhibits a linear
dependence on catalyst concentration (Figure 2A), and
These data are consistent with the ionic hydrogenation
mechanism proposed in Scheme 1.22,23 Initial addition of H2 to
Scheme 1. Proposed Mechanism for Ester Hydrogenation
the solvento complex Cp*Ir(bpy) (solv)2+ (solv = ester or
H2O) reversibly forms the iridium dihydrogen complex (I). In
neat reactions, the dihydrogen complex is then deprotonated by
substrate to generate the corresponding iridium hydride (II)
along with the protonated ester. This hydride attacks the
protonated substrate to generate a hemiacetal intermediate,
which then eliminates one equivalent of alcohol to generate a
transient aldehyde. The aldehyde is significantly more electro-
philic than the parent ester and undergoes rapid hydrogenation
to form the alcohol product. When the reactions are run in
DME, the solvent likely acts as a proton shuttle to protonate
the ester substrate prior to nucleophilic attack.
The first-order dependence on H2 pressure below 17 bar is
consistent with turnover-limiting formation of the dihydrogen
complex (I). The saturation observed at higher hydrogen
pressures suggests a change to turnover-limiting hydride
transfer. Notably, a similar mechanism (and change in turnover
limiting step) was proposed previously for the 1-catalyzed
hydrogenation of carboxylic acids.17
The scope of this transformation was next investigated. A
series of esters and lactones were evaluated using catalyst 1 and
30 bar of H2 in both neat substrate and in DME solvent.
Figure 2. Dependence of 1-catalyzed hydrogenation of esters on
catalyst concentration (A), H2 pressure (B), and substrate
concentration (C). (A) 2 mL of ethyl acetate (20.4 mmol), 30 bar
H2, 120 °C, 18 h. (B) 4 μmol of 1 in 2 mL of ethyl acetate (20.4
mmol), 120 °C, 18 h. (C) Substrate: hexyl formate; 4 μmol of 1 in
DME, 60 bar H2, 100 °C, 4 h.
1
Quantitative analysis of TON was carried out using either H
NMR spectroscopy or gas chromatography.19 Under neat
conditions, the reaction was examined in the presence and
absence of the Lewis acid Sc(OTf)3.19 On the basis of the
3114
DOI: 10.1021/acscatal.6b00263
ACS Catal. 2016, 6, 3113−3117

ACS Catalysis
Letter
a
mechanism proposed in Scheme 1, we hypothesized that a
Lewis acid should accelerate the hydrogenation reaction by
activating the ester substrate for nucleophilic attack, as seen
previously in the hydrogenation of carboxylic acids catalyzed by
1.17
Table 3. Ester Hydrogenation in DME Solvent
The hydrogenation of esters E1−E3 was first conducted
under neat conditions using 2 mM 1 and 30 bar H2 at 100 °C
for 16 h. In all cases, significant quantities of hydrogenation
products were observed, with the TON ranging from 106 to
341 (Table 2). The addition of 20 mM Sc(OTf)3 resulted in a
a
Table 2. Ester Hydrogenation under Neat Conditions
a
Average of at least 3 trials with standard deviation. Conditions: 4
μmol of 1 in 2 mL of substrate (E1, 24.8 mmol; E2, 24.4 mmol; E3,
20.4 mmol), 30 bar H2, 100 °C, 16 h. Theoretical maximum TON
b
under these conditions = E1, 6200; E2, 6300; E3, 5100. with 40 μmol
Sc(OTf)3.
marked improvement in TON. This effect is most dramatic
with ethyl formate (E1), where the TON increases from 341 to
1317 upon the addition of Sc(OTf)3.24
The hydrogenation of ester substrates E1−E11 was next
evaluated in DME (Table 3). These reactions were conducted
using 1 mmol of substrate in 1 mL of solvent under 30 bar H2
at 100 °C for 16 h using 0.5 mol % of 1. These conditions
enabled a clear comparison between different substrates
because the reactions generally proceeded to moderate
conversion. The reactivity was found to be strongly sensitive
to the size of the carbonyl substituent. For example, ethyl
formate afforded nearly quantitative conversion (TON = 173),
whereas the more sterically encumbered substrate ethyl acetate
showed much lower reactivity (TON = 27). Substrates bearing
even larger carbonyl substituents such as tert-butyl and phenyl
(E4, E5) afforded very low conversion under these conditions.
Hydride transfer is likely prohibitively slow in these systems
due to the steric demands of the substrate.
a
1 mmol substrate, 0.5 mol % 1, 1 mL of DME, 30 bar H2, 100 °C, 16
On the basis of the high reactivity of ethyl formate, the
reactions of a series of other formate esters were examined.19
Formate esters bearing alkyl ester substituents (E6−E11)
afforded methanol with TONs ranging from 89 to 115.25
Interestingly, these reactions were relatively insensitive to the
size and functionality of the alkoxy component of the ester
substrate. For example, t-butylformate (E8) and the FMOC-
protected amine E11 underwent hydrogenation with com-
parable TON and no apparent deprotection of the amine
functionality (115 and 107, respectively).
h. Yields represent an average of 3 trials
Reaction run in 2 mL DME. Average of 2 trials.
standard deviation.
b
Figure 3. Lactone substrates.
A variety of lactones (Figure 3, L1−L5) were also examined
as substrates for hydrogenation with catalyst 1. As shown in
Scheme 2, the initial product of lactone hydrogenation is a diol.
However, in the presence of catalyst 1, this diol intermediate
undergoes rapid dehydration to afford cyclic ethers as the major
product. This is noticeably different from the previously
reported Ru/Triphos catalyst system for lactone hydrogenation,
which affords the diol as the final product in the absence of
added acid.10−13
3115
DOI: 10.1021/acscatal.6b00263
ACS Catal. 2016, 6, 3113−3117

ACS Catalysis
Letter
Scheme 2. Hydrogenation of L2
AUTHOR INFORMATION
Corresponding Authors
■
*E-mail for K.I.G.: goldberg@chem.washington.edu.
*E-mail for M.S.S.: mssanfor@umich.edu.
Present Address
The results for the hydrogenation of the 5-membered
lactones γ-butyrolactone (L1) and γ-valerolactone (L2) are
shown in Table 4. In neat lactone, the addition of Sc(OTf)3 was
†Department of Chemistry, University of Memphis, Memphis,
TN 38152.
Notes
The authors declare no competing financial interest.
Table 4. Hydrogenation of L1 and L2 with 1
a
TON
ACKNOWLEDGMENTS
■
solvent
time (h)
additive
L1
291
L2
76
The authors thank Braden Zahora for contributing to the final
version of this manuscript. This work was supported by NSF
under the CCI Center for Enabling New Technologies through
Catalysis (CENTC), CHE-1205189.
b
neat
16
16
65
16
16
65
none
Sc(OTf)3
none
7
5
b
c
neat
410 53
487 16
200 19
166 54
b
neat
d
DME
none
33
23
69
4
6
4
27
19
78
5
3
6
d
e
DME
Sc(OTf)3
none
REFERENCES
d
■
DME
(1) Dub, P. A.; Ikariya, T. ACS Catal. 2012, 2, 1718−1741.
(2) Seyden-Penne, J. In Reductions by the Alumino- and Borohydrides
in Organic Synthesis, 2nd ed.; Wiley: New York, 1997; pp 1−224.
(3) Clarke, M. L. Catal. Sci. Technol. 2012, 2, 2418−2423.
(4) Spasyuk, D.; Smith, S.; Gusev, D. G. Angew. Chem., Int. Ed. 2012,
51, 2772−2775.
(5) Saudan, L. A. In Sustainable Catalysis, 1st ed.; Dunn, P. J., Hii, K.
K., Krische, M. J., Williams, M. T., Eds.; John Wiley & Sons, Inc.:
Hoboken, NJ, 2013; pp 37−61.
(6) Chakraborty, S.; Dai, H.; Bhattacharya, P.; Fairweather, N. T.;
Gibson, M. S.; Krause, J. A.; Guan, H. J. Am. Chem. Soc. 2014, 136,
7869−7972.
(7) Junge, K.; Wendt, B.; Jiao, H.; Beller, M. ChemCatChem 2014, 6,
2810−2814.
(8) Srimani, D.; Mukherjee, A.; Goldberg, A. F. G.; Leitus, G.;
Diskin-Posner, Y.; Shimon, L. J. W.; Ben David, Y.; Milstein, D. Angew.
Chem., Int. Ed. 2015, 54, 12357−12360.
(9) Zhang, J.; Leitus, G.; Ben-David, Y.; Milstein, D. Angew. Chem.,
Int. Ed. 2006, 45, 1113−1115.
(10) Geilen, F. M. A.; Engendahl, B.; Harwardt, A.; Marquardt, W.;
Klankermayer, J.; Leitner, W. Angew. Chem., Int. Ed. 2010, 49, 5510.
a
TON for cyclic ether product, average of 3 trials, standard deviation
in parentheses. 2 mL of substrate, 4 μmol 1, 30 bar H2, 100 °C. 40
μmol of Sc(OTf)3. 1 mmol substrate, 0.5 mol % 1, 1 mL DME, 30
b
c
d
e
bar H2, 100 °C. 0.5 mol % Sc(OTf)3
found to enhance reactivity.17 In contrast, in DME solvent, the
Lewis acid cocatalyst had minimal impact on TON. This may
be due to competitive coordination of the Lewis acid to DME.
The presence of the methyl group in L2 led to significantly
diminished TON in the neat reactions. This is consistent with
the observation that steric bulk around the carbonyl component
of the ester diminishes reactivity under neat conditions (E2 vs
E3, Table 2).
For the 6- and 7-membered ring lactones L3−L5, we observe
rapid conversion of starting material to a mixture of products by
NMR spectroscopy. However, only traces of the expected diol
and cyclic ether products were detected (as confirmed by
independent synthesis).19 Instead, ESI-MS revealed product
masses consistent with the formation of oligoesters (see
Supporting Information). This suggests that ring-opening
polymerization occurs more rapidly than hydrogenation for
these substrates.26,27 Analogous byproducts were not observed
for L1 or L2.
(11) Geilen, F. M. A.; Engendahl, B.; Holscher, M.; Klankermayer, J.;
̈
Leitner, W. J. Am. Chem. Soc. 2011, 133, 14349−14358.
(12) Wesselbaum, S.; vom Stein, T.; Klankermayer, J.; Leitner, W.
Angew. Chem., Int. Ed. 2012, 51, 7499−7502.
(13) vom Stein, T.; Meuresch, M.; Limper, D.; Schmitz, M.;
In conclusion, we have demonstrated that 1 catalyzes the
base-free hydrogenation of a variety of esters and lactones. This
catalyst is particularly effective for the hydrogenation of formate
esters, a substrate class that has been utilized as an intermediate
along an acid-assisted path from CO2 to methanol.15 Notably,
the presence of Lewis acid Sc(OTf)3 was not detrimental to
catalyst activity. Indeed, for reactions carried out in neat
substrate, this additive led to enhanced TONs. Mechanistic
investigations are consistent with a reaction pathway involving
turnover-limiting hydride transfer at high pressures of H2.
Consistent with this mechanism, the catalyst is very sensitive to
the size of the carbonyl substituent. Overall, this system offers a
valuable complement to previously developed, base-assisted
catalysts for the hydrogenation of esters.
Holscher, M.; Coetzee, J.; Cole-Hamilton, D. J.; Klankermayer, J.;
̈
Leitner, W. J. Am. Chem. Soc. 2014, 136, 13217−13225.
(14) Lewis acid-accelerated amide hydrogenation was found in
another case: Kita, Y.; Higuchi, T.; Mashima, K. Chem. Commun. 2014,
50, 11211−11213.
(15) Huff, C. A.; Sanford, M. S. J. Am. Chem. Soc. 2011, 133, 18122.
(16) Ogo, S.; Kabe, R.; Hayashi, H.; Harada, R.; Fukuzumi, S. Dalton
Trans. 2006, 4657−4663.
(17) Brewster, T. P.; Miller, A. J. M.; Heinekey, D. M.; Goldberg, K.
I. J. Am. Chem. Soc. 2013, 135, 16022−16025.
(18) Ogo, S.; Makihara, N.; Kaneko, Y.; Watanabe, Y. Organometallics
2001, 20, 4903−4910.
(19) For full experimental details, see Supporting Information.
(20) At this time point, reaction progress is within the linear initial
rate regime allowing for direct comparison across concentrations. See
Supporting Information for further details.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acscatal.6b00263.
(21) In contrast, the 2-catalyzed hydrogenation of glacial acetic acid
reaches saturation at approximately 25 bar H2 (ref 8).
(22) Bullock, R. M. Chem. - Eur. J. 2004, 10, 2366−2374.
(23) Eisenstein, O.; Crabtree, R. H. New J. Chem. 2013, 37, 21−27.
(24) Preliminary experiments run at lower H2 pressure (10 bar)
showed no added effect of Sc(OTf)3 on yield, suggesting that hydride
transfer is not turnover-limiting in this regime.
■
S
Experimental details and procedures, ESI-MS data, time-
course studies(PDF)
3116
DOI: 10.1021/acscatal.6b00263
ACS Catal. 2016, 6, 3113−3117

ACS Catalysis
Letter
(25) The second product of the reaction is the corresponding alcohol
(i.e., 1-hexanol from hexyl formate). Only in the cases of benzyl
formate (E9) and p-methoxybenzyl formate (E10) was an additional
product, the respective dibenzyl ether, observed in measurable
quantities. The ether byproduct is likely formed via an acid-catalyzed
condensation reaction.
(26) Lecomte, P.; Jer
́
o
̂
me, C. In Synthetic Biodegradable Polymers;
Rieger, B., Kunkel, A., Coates, W. G., Reichardt, R., Dinjus, E., Zevaco,
̈
A. T., Eds.; Springer Berlin Heidelberg: Berlin, Heidelberg, 2012; pp
173−218.
(27) Sarazin, Y.; Carpentier, J.-F. Chem. Rev. 2015, 115, 3564−3614.
3117
DOI: 10.1021/acscatal.6b00263
ACS Catal. 2016, 6, 3113−3117