Heterogeneous Hydrogenation of Esters under Mild Conditions using Solid-Supported Phosphorus–Ruthenium Catalysts

Article abstract of DOI:10.1002/cctc.201600330

The catalytic reduction of esters by using H₂ is a sustainable alternative to the use of classic stoichiometric reagents. Nowadays, a wide range of highly efficient homogeneous catalysts are known, but these suffer from poor catalyst recoverability. Therefore, most industrial applications are based on heterogeneous catalysts, but these processes are generally operated under harsh conditions (>200 °C, >10.0 MPa). Herein, we describe the first catalytic system that combines the activity of a homogeneous catalyst with the recoverability of a heterogeneous catalyst. The presented system is capable of hydrogenating esters under very mild conditions (25 °C, 5.0 MPa) and can easily be recovered. The catalyst is based on phosphorus ligands covalently attached to a polymeric support and is readily obtained by a facile solid-phase synthetic protocol in high yields with minimal workup.

Full text of DOI:10.1002/cctc.201600330

DOI: 10.1002/cctc.201600330
Communications
Heterogeneous Hydrogenation of Esters under Mild
Conditions using Solid-Supported Phosphorus–Ruthenium
Catalysts
[
a]
[a]
[b]
[c]
Frank J. L. Heutz, Christina Erken, Mae Joanne B. Aguila, Laurent Lefort, and
Paul C. J. Kamer*
[a]
The catalytic reduction of esters by using H is a sustainable al-
tions (250–3508C at 10.0–20.0 MPa) that are compatible with
2
[
4]
ternative to the use of classic stoichiometric reagents. Nowa-
days, a wide range of highly efficient homogeneous catalysts
are known, but these suffer from poor catalyst recoverability.
Therefore, most industrial applications are based on heteroge-
neous catalysts, but these processes are generally operated
under harsh conditions (>2008C, >10.0 MPa). Herein, we de-
scribe the first catalytic system that combines the activity of
a homogeneous catalyst with the recoverability of a heteroge-
neous catalyst. The presented system is capable of hydroge-
nating esters under very mild conditions (258C, 5.0 MPa) and
can easily be recovered. The catalyst is based on phosphorus
ligands covalently attached to a polymeric support and is read-
ily obtained by a facile solid-phase synthetic protocol in high
yields with minimal workup.
only a limited set of functional groups. Arene reduction, for
example, often presents a big problem with heterogeneous
catalysts. Moreover, upon considering the toxicity of Cr com-
pounds and the high capital and operational costs associated
with high-pressure units, it is no surprise that such heteroge-
neous catalysts have not been used in the production of fine
chemicals or pharmaceuticals. Hence, there has been a strong
drive from both academia and industry in the past decades to
develop homogeneous systems capable of these transforma-
tions under milder conditions. Since pivotal work in 2006 from
[
5]
[6]
both the Milstein (complex I) and Saudan (complex II)
groups, numerous catalysts capable of ester reduction under
milder conditions have been reported (40–1508C and 1.0–
[
7]
5.0 MPa, see Figure 1).
The reduction of esters to the corresponding alcohols is a key
transformation in organic chemistry. On laboratory and fine-
chemical scale, stoichiometric hydride reagents such as LiAlH4
and NaBH are commonly employed. Whereas these reagents
4
are relatively cheap, they can be hazardous to handle and gen-
[1]
erate stoichiometric amounts of waste. A more economical
and environmentally benign approach is the catalytic conver-
sion of esters by using H which yields the desired alcohols as
2
[
2]
the sole products.
Industrially, the catalytic hydrogenation of esters is applied
for the production of long-chain alcohols, which are important
intermediates, for example, in the production of surfactants
Figure 1. Selected representative homogeneous ester hydrogenation cata-
lysts.
[
3]
and plasticizers. Typically, heterogeneous copper chromite
catalysts are employed, and this leads to high alcohol yields.
These types of catalysts, however, operate under severe condi-
However, these homogeneous systems suffer from problem-
atic recovery and recycling of the often expensive transition
[
8]
metal and ligand. Also, the bi- or multidentate P,N ligands
commonly employed are often synthesized by multistep pro-
cedures that can be low yielding in part because of trouble-
some workup procedures. The synthesis of P,N ligands by
solid-phase synthesis is an attractive alternative that can afford
[
a] F. J. L. Heutz, C. Erken, Prof. Dr. P. C. J. Kamer
EaSTCHEM School of Chemistry
University of St Andrews
North Haugh, KY16 9ST, St Andrews (United Kingdom)
E-mail: pcjk@st-andrews.ac.uk
[
b] Dr. M. J. B. Aguila
[
9]
the desired ligands in high yields with minimal workup. Addi-
tionally, ligands covalently bound to a solid support provide
immobilized homogeneous catalysts, which facilitates recy-
cling. Recently, we reported on the solid-phase synthesis of re-
Institute of Chemistry
University of the Philippines Los Banos
Laguna 4031 (The Philippines)
[
c] Dr. L. Lefort
DSM Ahead Innovative Synthesis
P.O. Box 1066, 6160 BB Geleen (The Netherlands)
[10]
cyclable supported diphosphine and phosphine–phosphite
[
11]
ligands and their application in asymmetric hydrogenation.
In this work, we demonstrate a new solid-phase synthetic
method for the preparation of supported P,N and P,N,N ligands
Supporting Information and the ORCID identification number(s) for the
author(s) of this article can be found under http://dx.doi.org/10.1002/
cctc.201600330.
ChemCatChem 2016, 8, 1896 – 1900
1896
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
and their use in the Ru-catalyzed hydrogenation of esters
under mild conditions.
the results of Kuriyama et al., who used a combination of the
[
13]
analogous dppp ligand with (S,S)-DPEN. We observed a con-
version of 80% and a selectivity of 96% towards the desired al-
cohol for methyl benzoate (S1; Table 1, entry 1). For methyl
hexanoate (S2), lower conversions were obtained (Table 1,
The application of Noyori-type ruthenium 2,2’-bis(diphenyl-
phosphino)-1,1’-binaphthyl (BINAP)/diamine catalyst systems in
ester hydrogenation has been investigated by various
[
12]
groups. Recently, Kuriyama et al. at Takasago demonstrated
that BINAP could be substituted with cheaper bidentate phos-
phorus ligands such as 1,3-bis(diphenylphosphino)propane
entry 3). Supported diphosphine L2 with a C backbone gave
4
lower activity and selectivity, which is consistent with literature
(see Table 1, entries 2 and 4). Of all the other amines tested,
only N2 showed notable conversions (Table 1, entries 5 and 6).
With the other tested Ru precursor, RuCl (PPh ) , lower conver-
[13]
(
dppp) (complex III, Figure 1). Inspired by this work, two im-
mobilized diphosphines (i.e., L1 and L2, Table 1) previously
2
3 3
sions were obtained (see the Supporting Information).
Encouraged by these positive preliminary results, we set out
to develop more active resin-supported ester hydrogenation
catalysts. The first synthetic target was an immobilized ana-
logue of the aminophosphane ligand successfully employed
by Saudan et al. in ester hydrogenation (complex II, Figure 1).
The corresponding bis(aminophosphane) ruthenium complex
showed high activity in the reduction of esters [turnover fre-
Table 1. Reduction of esters using supported ligands (L1 and L2) and
amines (N1–N6).
[a]
À1 [6c]
quency (TOF) =2200 h ]. Various procedures for synthesiz-
ing aminophosphane ligands have been reported, and an
adaptation of the method developed by Abdur-Rashid et al.
[
14]
was used to synthesize the analogous supported ligand.
The starting point of the solid-phase synthesis was support-
ed phosphine 1, which was readily obtained by treating a chlor-
omethyl-functionalized resin, here, Merrifield resin, with lithium
[
10]
phenylphosphide. Subsequently, the phosphine moiety was
lithiated by using lithium diisopropylamide (LDA), which yield-
ed the desired supported lithium phosphide (see Scheme 1).
[
b]
[c]
Entry
P^P
N^N
Substrate
Conversion
%]
Selectivity
[%]
[
1
2
3
4
5
6
L1
L2
L1
L2
L1
L2
N1
N1
N1
N1
N2
N2
S1
S1
S2
S2
S1
S1
80
37
39
15
44
20
96
75
47
21
81
55
Scheme 1. Solid-phase synthesis of supported aminophosphane ligand L3.
This was confirmed by using gel-phase NMR spectroscopy. In
addition to a small change in shift of ~d=3 ppm, lithiation led
31
to major signal broadening in the P NMR spectrum (Figure 2),
[
a] Conditions: substrate (0.5 mmol), ligand (1.0 mol%), amine (1.0 mol%),
Ru metal (0.9 mol%), KOtBu (10 mol%), THF (2.5 mL), 808C, H (5.0 MPa),
6 h. [b] Conversion of starting ester determined by GC. [c] Selectivity to-
wards the desired alcohol.
whereas a single sharp resonance at d=0.4 ppm was observed
2
7
1
by Li NMR spectroscopy.
Next, the ligand backbone was introduced by reaction of
lithiated species Li·1 with a trimethylsilyl (TMS)-protected
chloroamine. Successful incorporation of the ligand backbone
[
10]
31
published, were tested as supported ligands in ester hydro-
genation. Both diphosphines were combined with six different
was confirmed by P NMR spectroscopy and the desired prod-
uct exhibited a single resonance at d=À23.7 ppm. Removal of
the TMS groups was achieved by treatment with dilute sulfuric
acid. Upon hydrolysis, a small upfield change in chemical shift
of approximately ~d=1 ppm was observed, and this signal
amines (i.e., N1–N6) and two Ru precursors, {i.e., RuCl (PPh3)3
and [Ru(p-cymene)Cl ] }. The generated small library of 24
2
2
2
members was tested in the hydrogenation of two substrates,
that is, methyl benzoate (S1) and methyl hexanoate (S2), in
the presence of KOtBu as an activator. The most important re-
sults of this screening are depicted in Table 1 (see the Support-
ing Information for full results).
corresponded to desired supported aminophosphane L3 (d=
À24.8 ppm). Moreover, complete hydrolysis was also confirmed
by monitoring the disappearance of the distinctive TMS signal
13
by gel-phase C NMR spectroscopy and the observation of pri-
mary amine stretching frequencies by using FTIR spectroscopy.
The supported ligand was obtained in near-quantitative yield
with minimal workup, which demonstrates the power of this
The combination of L1, bearing a C backbone, with (S,S)-
3
1
,2-diphenyl-1,2-ethylenediamine [(S,S)-DPEN, N1] and [Ru(p-
cymene)Cl ] gave the best results, which is in agreement with
2
2
ChemCatChem 2016, 8, 1896 – 1900
www.chemcatchem.org
1897
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
[
15]
esters at temperatures as low as 408C. The analogous sup-
ported P,N,N ligand was readily obtained by further modifica-
tion of immobilized aminophosphane L3 through a condensa-
tion reaction with 2-pyridinecarboxaldehyde (Scheme 2). The
Scheme 2. Solid-phase synthesis of supported P,N,N ligand L4.
formed imine exhibited a resonance at d=À22.7 ppm in the
31
P NMR spectrum, and introduction of the 2-picoline moiety
Figure 2. Solid-phase synthesis of supported aminophosphane L3 monitored
by gel-phase P NMR spectroscopy.
31
was also confirmed by FTIR spectroscopy. Subsequently, the
imine was reduced by using diisobutylaluminum hydride
(
DIBAL-H). Desired supported P,N,N ligand L4 was obtained in
solid-phase synthetic approach. The phosphorus loading of the
resin was determined by using elemental analysis.
high yield (88%) with minimal workup and shows a resonance
31
at a shift of d=À23.9 ppm in the P NMR spectrum. Complete
Supported aminophosphane L3 was then applied in the hy-
drogenation of methyl benzoate (S1) and methyl hexanoate
reduction of the imine was also confirmed by using gel-phase
13
C NMR spectroscopy and the appearance of a new signal at
(
S2) by using two Ru precursors (see Table 2). Full conversion
d=55.0 ppm was observed for the NCH2 carbon atom. The
phosphorus loading was determined by using elemental analy-
sis.
[a]
Table 2. Reduction of esters using supported aminophosphane L3.
Supported P,N,N ligand L4 was screened in the hydrogena-
tion of S1 and S2. When a combination of ligand L4 and
RuCl (PPh ) was used, 100% yield towards the desired alcohol
[
b]
[c]
Entry
Ru
Substrate
Conversion
%]
Selectivity
[%]
2
3 3
[
was obtained for both substrates at 408C (Table 3, entries 1
and 2). Even at temperatures as low as 258C, almost full con-
version and selectivity were achieved for both S1 and S2
(Table 3, entries 5 and 6). To our knowledge, these conditions
are the mildest ever reported for a supported ester hydrogena-
tion catalyst. When using the [Ru(p-cymene)Cl2]2 precursor,
however, only conversions up to 60% were obtained (Table 3,
entries 3 and 4).
1
2
3
4
RuCl
2
(PPh
3
)
3
S1
S1
S2
S2
S1
S1
>99
>99
52
47
>99
95
>99
>99
64
59
>99
>99
[Ru(p-cymene)Cl
RuCl (PPh
[Ru(p-cymene)Cl
RuCl (PPh
[Ru(p-cymene)Cl
2
]
2
]
2
]
2
2
2
2
3 3
)
[
d]
d]
5
6
2
3
)
3
[
[a] Conditions: substrate (0.25 mmol), ligand (1.0 mol%), Ru metal
(0.9 mol%), KOtBu (10 mol%), THF (1 mL), 1008C, H (5.0 MPa), 16 h.
[b] Conversion of starting ester determined by GC. [c] Selectivity towards
2
the desired alcohol. [d] Performed at 808C.
[a]
Table 3. Reduction of esters using supported P,N,N ligand L4.
of S1 with over 99% selectivity to the desired alcohol was ach-
ieved with both metal precursors at 1008C and 5.0 MPa H2
[b]
[c]
Entry
Ru
Substrate
Conversion
%]
Selectivity
[%]
[
(
Table 2, entries 1 and 2). For alkyl ester S2, however, only
1
2
3
4
RuCl
RuCl
2
(PPh
(PPh
3
)
)
3
S1
S2
S1
S2
S1
S2
>99
>99
59
50
99
>99
>99
91
61
>99
99
modest conversions up to 52% were achieved (Table 2, en-
tries 3 and 4), and a significant amount of hexyl hexanoate
was formed by transesterification between the product and
the starting material. For S1, the desired product could even
be obtained in full yield at a decreased temperature of 808C
2
3
3
[Ru(p-cymene)Cl2]2
[Ru(p-cymene)Cl
2
]
2
[
d]
d]
5
6
RuCl
RuCl
2
(PPh
(PPh
3
)
)
3
[
2
3
3
97
[
a] Conditions: substrate (0.25 mmol), ligand (1.0 mol%), Ru metal
(0.9 mol%), KOtBu (10 mol%), THF (1 mL), 408C, H (5.0 MPa), 16 h.
b] Conversion of starting ester determined by GC. [c] Selectivity towards
the desired alcohol. [d] Performed at 258C.
(
Table 2, entry 5). Interestingly, complex II reported by Saudan
[
6]
2
et. al contains two P,N ligands, whereas our Ru species is only
monoligated; this demonstrates that ester hydrogenation ac-
tivity does not necessarily require two P,N ligands per Ru
center.
[
To further improve the activity of these supported ester hy-
drogenation catalysts, we turned towards supported P,N,N li-
gands. Research by Gusev et al. showed that ruthenium P,N,N
complexes (see complex IV, Figure 1) can be highly effective
ester hydrogenation catalysts for both aromatic and alkyl
The best-performing catalyst based on L4 and RuCl (PPh )
was employed to determine the substrate scope (see Figure 3).
2
3 3
Especially alkyl esters seemed to be hydrogenated with relative
ease, and almost full conversion and selectivity were achieved
for both ethyl hexanoate (S3) and hexyl hexanoate (S4).
ChemCatChem 2016, 8, 1896 – 1900
www.chemcatchem.org
1898
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
this hampered further efficient recycling in the used batch
setup. These preliminary results, however, do demonstrate that
the supported catalyst can be easily recovered and reused by
using a simple filtration step. This catalytic system would be
very suitable for application of ester hydrogenation in a contin-
uous flow system.
In summary, we prepared the first heterogeneous catalyst
able to hydrogenate esters at room temperature. For this pur-
pose, we developed an efficient solid-phase synthetic method
for the preparation of supported P,N and P,N,N ligands in high
yields and purities. After the addition of a Ru precursor, these
catalytic systems were able to hydrogenate a range of aromat-
ic and alkyl esters at temperatures as low as 258C. High con-
versions and selectivities were achieved without any observa-
ble arene reduction. Preliminary recyclability tests were per-
formed, and they demonstrated that the supported catalyst
could be easily recovered and reused. Currently, work is ongo-
ing to extend both the supported ligand library as well as the
substrate scope. Also, the possibility to use these immobilized
ester hydrogenation catalysts under flow conditions will be ex-
plored.
Figure 3. Substrate scope for ester hydrogenation using supported ligand
L4 (conversion and selectivity indicated below structures). For conditions
see Table 3.
Methyl cyclohexanoate (S5) proved to be a more challenging
substrate, and only 73% conversion and 82% selectivity to the
desired product were reached. Interestingly, no conversion was
observed for diethyl succinate (S6), even with increased reac-
tion times and at temperatures up to 1008C. Aromatic esters
ethyl benzoate (S7) and benzyl benzoate (S8) also proved
more challenging. For both substrates, moderate conversions
were reached at 408C. However, at an increased temperature
of 608C, almost full conversion and selectivity were obtained
for both substrates.
Acknowledgements
We thank the European Union (Marie Curie ITN SusPhos, Grant
Agreement No. 317404) for financial support.
Keywords: hydrogenation · immobilization · N,P ligands ·
ester reduction · solid-phase synthesis
Finally, the reusability of the solid-supported L4-based
system was investigated (Table 4). A shorter reaction time (2 h
[
1] a) J. Seyden-Penne, Reductions by the Allumino- and Borohydride in Or-
ganic Synthesis, 2nd ed., Wiley-VCH, New York, 1997; b) G. W. Gribble,
Chem. Soc. Rev. 1998, 27, 395–404; c) M. A. Fox, J. K. Whitesell in Organ-
ic Chemistry, 3rd ed., Jones and Bartlett Publishers, Sudbury, 2004,
pp. 776–580.
Table 4. Results for the recycling of ligand L4 in the hydrogenation of
[a]
S1.
[
[
2] a) L. A. Saudan in Sustainable Catalysis, Wiley, 2013, pp. 37–61; b) S.
Werkmeister, K. Junge, M. Beller, Org. Process Res. Dev. 2014, 18, 289–
[
b]
[c]
Run
Ru
Conversion
%]
Selectivity
[%]
302.
[
3] K. Weissermel, H.-J. Arpe in Industrial Organic Chemistry, Wiley-VCH,
Weinheim, 2008, pp. 193–215.
1
2
3
4
RuCl
RuCl
RuCl
RuCl
2
2
2
2
(PPh
(PPh
(PPh
(PPh
3
3
3
3
)
)
)
)
3
3
3
3
69
72
69
47
95
93
93
89
[4] a) R. Rieke, D. Thakur, B. Roberts, G. White, J. Am. Oil Chem. Soc. 1997,
74, 333–339; b) R. Rieke, D. Thakur, B. Roberts, G. White, J. Am. Oil
Chem. Soc. 1997, 74, 341–345; c) Y. Pouilloux, F. Autin, J. Barrault, Catal.
Today 2000, 63, 87–100.
[
(
a] Conditions: substrate (0.25 mmol), ligand (1.0 mol%), Ru metal
0.9 mol%), KOtBu (10 mol%), THF (1 mL), 408C, H (5.0 MPa), 2 h. [b] Con-
[5] a) J. Zhang, G. Leitus, Y. Ben-David, D. Milstein, Angew. Chem. Int. Ed.
2
2006, 45, 1113–1115; Angew. Chem. 2006, 118, 1131–1133; b) E. Fogler,
version of starting ester determined by GC. [c] Selectivity towards the de-
sired alcohol.
E. Balaraman, Y. Ben-David, G. Leitus, L. J. W. Shimon, D. Milstein, Orga-
nometallics 2011, 30, 3826–3833; c) E. Balaraman, E. Fogler, D. Milstein,
Chem. Commun. 2012, 48, 1111 –1113.
[
[
6] a) L. Saudan, P. DuPau, J. J. Riedhauser, P. Wyss, WO2006106483(A1),
instead of 16 h) was used to assess any decrease in the rate of
the reaction. Supported P,N,N ligand L4 could successfully be
used up to three times in the hydrogenation of S1 without
any loss of activity. After each run, the supernatant was re-
moved and fresh substrate and base solutions were added, all
2006; b) L. Saudan, P. Dupau, J. J. Riedhauser, P. Wyss,
WO2006106484(A1), 2006; c) L. A. Saudan, C. M. Saudan, C. Debieux, P.
Wyss, Angew. Chem. 2007, 119, 7617–7620.
7] a) P. A. Dub, T. Ikariya, ACS Catal. 2012, 2, 1718–1741; b) W. Kuriyama, T.
Matsumoto, O. Ogata, Y. Ino, K. Aoki, S. Tanaka, K. Ishida, T. Kobayashi,
N. Sayo, T. Saito, Org. Process Res. Dev. 2012, 16, 166–171; c) M. L.
Clarke, Catal. Sci. Technol. 2012, 2, 2418–2423; d) J. Pritchard, G. A. Filo-
nenko, R. van Putten, E. J. M. Hensen, E. A. Pidko, Chem. Soc. Rev. 2015,
under a gentle flow of H to ensure catalyst stability. After the
2
fourth run, some catalyst deactivation was observed, possibly
caused by introduction of trace amounts of water or oxygen
during catalyst workup. Moreover, the presence of both sub-
4
4, 3808–3833; e) S. Werkmeister, J. Neumann, K. Junge, M. Beller,
Chem. Eur. J. 2015, 21, 12226–12250.
8] a) D. J. Cole-Hamilton, Science 2003, 299, 1702–1706; b) D. J. Cole-Ham-
[
strate and H was found to be crucial for catalyst stability and
ilton, R. P. Tooze in Catalyst separation, recovery and recycling—Chemis-
2
ChemCatChem 2016, 8, 1896 – 1900
www.chemcatchem.org
1899
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
try and process design (Eds.: D. J. Cole-Hamilton, R. P. Tooze), Springer,
Dordrecht, 2006, pp. 1–8.
[12] a) L. Saudan, C. Saudan, WO2008065588, 2007; b) S. Takebayashi, S. H.
Bergens, Organometallics 2009, 28, 2349–2351.
[
9] a) D. Obrecht, J. M. Villalgordo in Solid-Supported Combinatorial and Par-
allel Synthesis of Small-Molecular-Weight Compound Libraries, Elsevier
Science ltd., Oxford, 1998, pp. 1–184; b) K. Burgess, Solid-Phase Organic
Synthesis, Wiley, New York, 2002; c) M. C. Samuels, B. H. G. Swennenhuis,
P. C. J. Kamer in Phosphorus(III) Ligands in Homogeneous Catalysis:
Design and Synthesis (Eds.: P. C. J. Kamer, P. W. N. M. v. Leeuwen), Wiley,
[13] W. Kuriyama, Y. Ino, O. Ogata, N. Sayo, T. Saito, Adv. Synth. Catal. 2010,
352, 92–96.
[14] W. Jia, X. Chen, R. Guo, C. Sui-Seng, D. Amoroso, A. J. Lough, K. Abdur-
Rashid, Dalton Trans. 2009, 8301–8307.
[15] a) D. Spasyuk, D. G. Gusev, Organometallics 2012, 31, 5239–5242; b) D.
Spasyuk, S. Smith, D. G. Gusev, Angew. Chem. Int. Ed. 2012, 51, 2772–
2
775; Angew. Chem. 2012, 124, 2826–2829.
2
012, pp. 463–479.
10] F. J. L. Heutz, M. C. Samuels, P. C. J. Kamer, Catal. Sci. Technol. 2015, 5,
296–3301.
11] F. J. L. Heutz, P. C. J. Kamer, Dalton Trans. 2016, 45, 2116–2123.
[
[
3
Received: March 21, 2016
Published online on May 24, 2016
ChemCatChem 2016, 8, 1896 – 1900
www.chemcatchem.org
1900
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim