Selective Hydrogenation of Functionalized Alkynes to (E)-Alkenes, Using Ordered Alloys as Catalysts

Article abstract of DOI:10.1021/acscatal.5b02953

Intermetallic Pd₃Pb acts as a highly selective alkyne semihydrogenation catalyst that is greatly superior to the conventional Lindlar catalyst. Density functional theory (DFT) calculations demonstrate an ideal adsorption property of Pd₃Pb, where the surface holds alkynes while releasing alkenes. A tandem catalytic system that is comprised of Pd₃Pb/SiO₂ for alkyne semihydrogenation and RhSb/SiO₂ for alkene isomerization allows one-pot (E)-alkene synthesis from a functionalized alkyne, which is the first success using heterogeneous catalysts. A variety of functionalized alkynes with aldehyde, ketone, carboxylic acid, and ester moieties are hydrogenated into the corresponding (E)-alkene in good to excellent yields under 1 atm H₂ at room temperature.

Full text of DOI:10.1021/acscatal.5b02953

Letter
pubs.acs.org/acscatalysis
Selective Hydrogenation of Functionalized Alkynes to (E)‑Alkenes,
Using Ordered Alloys as Catalysts
Shinya Furukawa* and Takayuki Komatsu*
Department of Chemistry, Graduate School of Science and Enginnering, Tokyo Institute of Technology, Ookayama, Meguro-ku,
Tokyo 152-8550, Japan
S
* Supporting Information
ABSTRACT: Intermetallic Pd3Pb acts as a highly selective
alkyne semihydrogenation catalyst that is greatly superior to the
conventional Lindlar catalyst. Density functional theory (DFT)
calculations demonstrate an ideal adsorption property of Pd3Pb,
where the surface holds alkynes while releasing alkenes. A tandem
catalytic system that is comprised of Pd3Pb/SiO2 for alkyne
semihydrogenation and RhSb/SiO2 for alkene isomerization
allows one-pot (E)-alkene synthesis from a functionalized alkyne,
which is the first success using heterogeneous catalysts. A variety of functionalized alkynes with aldehyde, ketone, carboxylic acid,
and ester moieties are hydrogenated into the corresponding (E)-alkene in good to excellent yields under 1 atm H2 at room
temperature.
KEYWORDS: alkyne hydrogenation, alkene isomerization, (E)-alkene, ordered alloy, intermetallic compound, Pd3Pb, RhSb
he semihydrogenation of alkynes to alkenes is a very
hydrogenation of CC bonds is favored kinetically and
Timportant chemical transformation in many industrial and thermodynamically compared with heteroatom-containing
synthetic applications.1 To date, several catalytic systems
unsaturated bonds such as CO, CN, CN, and NO.
This hydrogenation is also expected to be compatible with acid-
labile substrates, such as amines. Moreover, the use of hydrogen
would allow for a one-pot protocol for semihydrogenation and
isomerization under a single reaction condition. However,
hydrogen-mediated alkene isomerization is always associated
with irreversible overhydrogenation to alkane, which signifi-
cantly lowers the yield of the desired (E)-alkene at a high
conversion. This drawback has been a significant hindrance for
the successful development of (E)-alkene synthesis reactions.
Recently, we reported on the discovery of an unprecedented
catalysis that overcomes this problem, where a RhSb-ordered
alloy (intermetallic compound) selectively catalyzes the isomer-
ization of (Z)-stilbene to (E)-stilbene under 1 atm H2 with a
negligible byproduction of the overhydrogenated diphenyl-
ethane.8 A detailed study combined with several character-
izations and theoretical calculations revealed that the surface of
RhSb is dominated by one-dimensionally aligned Rh rows
separated by Sb atoms (1D-planes), which allows surface
hydrogen diffusion in one direction (Figure 1).
applicable to alkyne semihydrogenation, represented by Lindlar
reduction, have been reported.2 The catalytic hydrogenation of
inner alkynes, being either homogeneous or heterogeneous,
shows an intrinsic stereoselectivity to (Z)-alkenes, because of
their syn addition style. In contrast, the catalytic hydrogenation
of alkynes to (E)-alkenes, in principle, barely occurs. Although a
Birch-type reduction can be applied to (E)-alkene synthesis, the
reaction is stoichiometric and the strongly reductive condition
with alkali metals in liquid ammonia is essentially incompatible
to functionalized alkynes with base-labile or reducible moieties.3
In this context, the development of an efficient catalytic system
for (E)-alkene synthesis that is both mild and functional-group-
tolerant is highly attractive and challenging.
A simple and promising synthetic root involves a two-step
conversion via catalytic hydrogenation to (Z)-alkene, followed
by isomerization to the thermodynamically stable (E) counter-
part. Alkene isomerization generally requires the rotation of an
alkenyl CC bond. The conventional protocols to rotate the
CC bond employ photoexcitation to a triplet state,4 where a
π-system is twisted, or acid-catalyzed formation of carbocation
intermediates.5 However, these protocols are not very
compatible with functionalized alkynes, because the functional
groups typically undergo undesired photodecomposition (e.g.,
Norrish reactions of carbonyl6) or acid-catalyzed side reactions
(e.g., aldol condensation or hydrolysis of esters). A promising
candidate may be hydrogen-mediated alkene isomerization. In
this process, the half-hydrogenation of a (Z)-alkene to an alkyl
intermediate allows for C−C rotation, and the subsequent β−H
elimination can produce the corresponding (E)-alkene.7 The
The geometric constraints of the 1D-planes and steric
hindrance from one alkyl group of (Z)-alkene limit hydrogen
access toward the alkenyl carbon to one direction, enabling
one-atom hydrogenation for isomerization but inhibiting two-
atom hydrogenation for overhydrogenation (Figure 1).8 In the
present study, the unprecedented surface stereochemistry
Received: December 26, 2015
Revised: February 15, 2016
© XXXX American Chemical Society
2121
DOI: 10.1021/acscatal.5b02953
ACS Catal. 2016, 6, 2121−2125

ACS Catalysis
Letter
in Table 1). Similar results were obtained when PdCu and
PdZn were used as catalysts (entries 2 and 3, respectively, in
Table 1). These catalysts produced relatively high (E)-alkene
contents, which was likely due to the occurrence of
isomerization, as well as overhydrogenation. The catalysts
Pd20Sn13, PdIn, PdFe, and Pd2Ga exhibited low or moderate
conversions (5%−58%), even after 90 min. Both PdAg (an
industrial catalyst for acetylene semihydrogenation) and Pd−
Pb/CaCO3 resulted in relatively good selectivities (entries 8
and 9 in Table 1). However, they were below 90%, because of
overhydrogenation. The selectivity was improved in the
presence of quinoline modifier (entry 10 in Table 1), as
typically applied in organic synthesis.2a,b Conversely, Pd3Bi and
Pd3Pb exhibited excellent selectivities (entries 11 and 12 in
Table 1), even at the 99% conversion of alkyne. The highest
selectivity was obtained using Pd3Pb, resulting in an almost
stoichiometric conversion into the corresponding alkene.
Figure 2 shows the time course of the product yield in the
hydrogenation of 4-phenyl-3-butyn-2-one with the Pd−Pb/
CaCO3 and Pd3Pb/SiO2 catalysts.
Figure 1. Equilibrium crystal shape of RhSb and atomic arrangement
on the surface (left). Schematic illustration of limited hydrogen access
to the alkenyl carbon on the 1D-plane ((020) plane, right).8
governed by RhSb/SiO2 was employed as an effective alkene
isomerization methodology for (E)-alkene synthesis. Herein,
we report the first example of a heterogeneous catalytic system
for (E)-alkene synthesis from various functionalized alkynes.
First, we optimized the alkyne semihydrogenation catalysts
using Pd-based bimetallic materials, including a commercially
available Lindlar catalyst (Pd−Pb/CaCO3, Pd 5 wt %, TCI). A
series of Pd-based solid solution alloys and intermetallic
compounds supported on silica (PdxMy/SiO2, where M = Ag,
Bi, Cu, Fe, Ga, In, Pb, Sn, and Zn) were prepared by
conventional impregnation, followed by H2 reduction at 400
°C−800 °C (see the Supporting Information for the
experimental details). The formation of the desired alloy or
intermetallic phase was confirmed by X-ray diffraction (XRD)
(see Figure S1 in the Supporting Information). The catalytic
performances of these palladium-based catalysts were compared
in the hydrogenation of 4-phenyl-3-butyn-2-one under 1 atm
H2 at 25 °C (see Table 1).
For all catalysts, an overhydrogenated alkane (4-phenyl-
butan-2-one) was formed as the main byproduct. As commonly
known, monometallic Pd/SiO2 shows low alkene selectivity,
because of the significant overhydrogenation to alkane (entry 1
Table 1. Semihydrogenation of 4-Phenyl-3-butyn-2-one
Using Various Palladium-Based Catalysts
Figure 2. Time course of the product yield in the hydrogenation of 4-
phenyl-3-butyn-2-one over Pd−Pb/CaCO3 (left) and Pd3Pb/SiO2
(right).
a
For Pd−Pb/CaCO3, overhydrogenation to alkane proceeded
significantly after the complete hydrogenation of alkyne,
resulting in a lower alkene yield. Thus, the commercial Lindlar
catalyst could not prevent the undesired overhydrogenation
and was not suitable for (E)-alkene synthesis. Conversely, when
Pd3Pb/SiO2 was used, the overhydrogenation was successfully
inhibited even after the complete conversion of alkyne. Note
that the alkene selectivity was maintained at 93% at twice the
reaction time for the complete alkyne conversion, which was
higher than those for Pd−Pb/CaCO3 (54%) and the quinoline-
modified Pd−Pb/CaCO3 (86%, Figure S2 in the Supporting
Information). According to a report in the literature, the Pb
species of the Pd−Pb/CaCO3 catalyst is deposited on the
surface of monometallic palladium,9 which is different from the
Pd3Pb/SiO2 catalyst, which comprised the intermetallic phase.
Based on these results, we concluded that Pd3Pb/SiO2 was the
catalyst most suitable for the (E)-alkene synthesis. We also
checked the catalytic performance of RhSb/SiO2 for the alkyne
hydrogenation, which showed a very low conversion (Table 1,
entry 13). This may be explained by the geometric effect of
RhSb 1D-plane, as shown in Figure 1: for alkyne, hydrogen
alkene
time
conversion
(%)
selectivity
b
entry
catalyst
Pd/SiO2
(min)
(%)
Z:E
1
2
3
4
5
6
7
8
16
20
25
90
90
90
90
30
16
16
60
95
90
98
98
98
5
65
31
78
95
96
95
91
84
88
97
94
98
86
73:27
74:26
87:13
88:12
90:10
92:8
PdCu/SiO2
PdZn/SiO2
Pd20Sn13/SiO2
PdIn/SiO2
7
PdFe/SiO2
23
58
99
99
99
99
99
4
Pd2Ga/SiO2
PdAg/SiO2
Pd−Pb/CaCO3
93:7
85:15
91:9
c
9
cd
,
10
11
12
13
94:6
Pd3Bi/SiO2
Pd3Pb/SiO2
RhSb/SiO2
92:8
86:14
78:22
a
Reaction conditions: alkyne, 0.5 mmol; catalyst, 50 mg (Pd or Rh: 3
wt %); solvent, 5 mL (THF); atmosphere, 1 atm H2; temperature, 25
°C. Selectivity to (Z)- and (E)-alkenes. Pd 5 wt %, 20 mg.
b
c
d
Quinoline (2 mg) was added in the reaction mixture.
2122
DOI: 10.1021/acscatal.5b02953
ACS Catal. 2016, 6, 2121−2125

ACS Catalysis
Letter
access to the alkynyl carbons is completely blocked by the
connecting functional groups.
The monometallic Pd(111) surface showed largely (−107 kJ
mol−1) and moderately (−49 kJ mol−1) negative Ead values for
acetylene and ethylene adsorption, respectively. The latter
indicates that alkene desorption from palladium is relatively
slow, which causes overhydrogenation to alkane. For other
bimetallic surfaces, the Ead values for ethylene were higher than
those of palladium in the following order:
The observed difference in alkene selectivity among the
tested catalysts has often been explained by the adsorption
energy of alkene.10 The adsorption energy of alkene is typically
reduced in the presence of the secondary metal atoms adjacent
to Pd atoms, which accelerates alkene desorption, thereby,
inhibiting overhydrogenation. However, alkyne adsorption is
also weakened, which lowers the hydrogenation rate of alkyne.
Based on this concept, we performed density functional theory
(DFT) calculations to compare the adsorption energies of
alkene and alkyne on the surface of the palladium-based
materials used in this study (Pd, PdAg, Pd−Pb, Pd3Bi, Pd3Pb,
and PdIn). As models of adsorbed alkene and alkyne, ethylene
and acetylene with di-σ adsorption to Pd atoms were
considered, respectively. Figure 3 shows the optimized
Pd < PdAg < Pd−Pb < Pd3Bi < Pd3Pb ≪ PdIn
This order is consistent with those of alkene selectivity at the
complete conversion of alkyne (Pd < PdAg < Pd−Pb < Pd3Bi <
Pd3Pb), demonstrating the aforementioned rationale for alkene
selectivity. Although PdIn showed the highest Ead value for
ethylene, that for acetylene was also quite positive. This agrees
with the very low catalytic activity of PdIn (Table 1, entry 5).
Thus, appropriate modification in Ead is needed to develop an
efficient catalyst for alkene semihydrogenation. In this context,
Pd3Pb provides an ideal adsorption property, i.e., slightly
positive Ead for ethylene and moderately negative Ead for
acetylene. Note that this property cannot be obtained with the
Pb-deposited Pd surface (Pd−Pb). The modifier second metal
atoms should be incorporated into the lattice, as observed for
intermetallic Pd3Pb.
We subsequently performed a one-pot synthesis of (E)-
alkene in the presence of Pd3Pb/SiO2 and RhSb/SiO2. Figure 5
shows the time course of the product yield in the hydro-
genation of 3-phenyl-2-propyn-1-one using these catalysts.
Figure 3. Optimized structures of di-σ bonded ethylene on Pb-
deposited-Pd(111) (left) and Pd3Pb(111) (right) surfaces. For the
former, Pb atoms were deposited on fcc hollow sites of Pd(111) with a
(3 × 3) structure.
structures of ethylene adsorbed on Pb-deposited-Pd(111) and
Pd3Pb(111) surfaces, the former of which represents a
simplified model surface of the Lindlar catalyst (structures for
other surfaces and acetylene are shown in Figure S3 in the
Supporting Information).
These models clearly distinguish whether the Pb atoms are
incorporated into the surface lattice or not, which is the crucial
difference between the Lindlar catalyst and intermetallic Pd3Pb.
The calculated adsorption energy (Ead) of acetylene and
ethylene on the surfaces of various Pd-based materials are
summarized in Figure 4.
Figure 5. Time course of the product yield in the hydrogenation of 3-
phenyl-2-propyn-1-one in the presence of Pd3Pb/SiO2 and RhSb/
SiO2. Reaction conditions: alkyne, 0.5 mmol; catalyst, Pd3Pb/SiO2
(Pd: 3 wt %) 100 mg, RhSb/SiO2 (Rh: 3 wt %) 100 mg; solvent
(THF), 5 mL; atmosphere, 1 atm H2; temperature, 25 °C.
At the early stage of the reaction, the semihydrogenation of
the alkyne to the (Z)-alkene mainly proceeded. In the latter
half, the isomerization of the (Z)-alkene to the (E) counterpart
took place almost exclusively. Thus, the tandem catalytic
reaction system allowed for the successive reactions of
semihydrogenation and isomerization. During the reaction,
overhydrogenation to the alkane was successfully inhibited,
Figure 4. Calculated adsorption energies of di-σ bonded acetylene and
ethylene on the surfaces of various Pd-based surfaces: Pd(111),
PdAg(110), Pd(111)-(3 × 3)-2Pb (Pd−Pb), Pd3Bi(100), Pd3Pb(111),
and PdIn(110). For each material, the most densely packed surface
was employed as a most stable surface.
2123
DOI: 10.1021/acscatal.5b02953
ACS Catal. 2016, 6, 2121−2125

ACS Catalysis
Letter
resulting in a 94% yield of the desired (E)-alkene and the
production of only a small amount of the overhydrogenated
alkane. Alcoholic species originating from carbonyl hydro-
genation were not detected.
In a similar fashion, the hydrogenation of various function-
alized alkynes was performed using the tandem catalytic system.
Table 2 shows the resulting alkyne conversion, alkene
selectivity and E:Z ratios, which reflects the (Z)-alkene
conversion in the second step.
convenient handling of the synthesis. This catalytic system
worked well also in a gram-scale condition, which afforded a
good isolated yield of the desired (E)-alkene (76%, entry 5).
Recycling tests revealed that the catalysts could be used at least
4 times without reducing the selectivity (see Table S1 in the
Supporting Information). Although the conversion rate
decreased in the reuse, high product yields comparable to
that for the initial run could be obtained at longer reaction
times.
In this catalytic system, inhibiting the overhydrogenation to
alkane is one of the critical points to obtain a high (E)-alkene
yield. The conventional Lindlar catalyst, in this context, is not
suitable for this reaction, because the unavoidable over-
hydrogenation during the second isomerization step would
significantly decrease the alkene yield. The other critical point is
that the RhSb/SiO2 catalyst and its hydrogen-mediated
isomerization methodology are highly compatible to various
functional groups. However, our attempt to hydrogenation of
nitro-containing alkyne (1-nitro-4-phenylethynyl-benzene) to
the corresponding (E)-alkene resulted in hydrogenation of both
alkyne and nitro moieties (data not shown). A control
experiment using RhSb/SiO2 alone revealed that RhSb/SiO2
is active for nitro-hydrogenation.
Table 2. Hydrogenation of Various Functionalized Alkynes
to the Corresponding (E)-Alkenes Using Pd3Pb/SiO2 and
a
RhSb/SiO2 Catalysts
In conclusion, we have discovered that a Pd3Pb ordered alloy,
as a highly selective alkyne semihydrogenation catalyst, is
greatly superior to the conventional Lindlar catalyst. DFT
calculations have demonstrated an ideal adsorption property of
Pd3Pb, where the surface holds alkynes while releasing alkenes.
The combination of the two original catalysts, i.e., Pd3Pb/SiO2
for alkyne semihydrogenation and RhSb/SiO2 for alkene
isomerization, allowed for one-pot (E)-alkene synthesis from
a functionalized alkyne, which is the first success using
heterogeneous catalysts. A variety of functionalized alkynes
with aldehyde, ketone, carboxylic acid, and ester moieties were
hydrogenated into the corresponding (E)-alkene in good to
excellent yields under 1 atm H2 at room temperature.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acscatal.5b02953.
■
S
Experimental and computational details, XRD patterns,
recycling test, and optimized structures (PDF)
AUTHOR INFORMATION
Corresponding Authors
*E-mail: furukawa.s.af@m.titech.ac.jp (S. Furukawa).
*E-mail: komatsu.t.ad@m.titech.ac.jp (T. Komatsu).
■
a
Reaction condition is identical to that described in Figure 5.
b
c
Selectivity to (E) and (Z) alkenes. Alkyne, 6.0 mmol; Pd3Pb/SiO2,
Notes
d
e
200 mg, RhSb/SiO2, 200 mg. Isolated yield. Alkyne, 0.1 mmol;
The authors declare no competing financial interest.
Pd3Pb/SiO2, 25 mg, RhSb/SiO2, 75 mg.
REFERENCES
■
(1) (a) Blaser, H.-U.; Schnyder, A.; Steiner, H.; Rossler, F.;
̈
Various functionalized alkynes with aldehyde (entries 1 and
9), ketone (entries 2 and 7), carboxylic acid (entry 3), ester
(entries 4 and 10), and alcohol (entry 6) moieties were
converted to the corresponding (E)-alkene in good to excellent
yields (typically 81%−94%). In most cases, isomerization
successfully proceeded to provide high E:Z ratios (typically
>99:1). Thus, the tandem catalytic system exhibited a high
functional group tolerance in the (E)-alkene synthesis reaction.
Note that this catalytic system worked well under 1 atm H2 at
room temperature, which allowed for the simple and
Baumeister, P. In Handbook of Heterogeneous Catalysis; Wiley−VCH
Verlag GmbH & Co. KGaA: Weinheim, Germany, 2008; pp 3284−
3308. (b) Heller, D.; De Vries, A. H.; De Vries, J. G. In Handbook of
Homogeneous Hydrogenation; de Vries, J. G., Elsevier, C. J., Eds.;
Wiley−VCH: Weinheim, Germany, 2007; Vol. 3, pp 1483−1516.
(c) Molnar, A.; Sarkany, A.; Varga, M. J. Mol. Catal. A: Chem. 2001,
173, 185−221.
(2) (a) Lindlar, H. Helv. Chim. Acta 1952, 35, 446−456. (b) Lindlar,
H.; Dubuis, R. Organic Syntheses 2003, 46, 89−92. (c) Rajaram, J.;
Narula, A. P. S.; Chawla, H. P. S.; Dev, S. Tetrahedron 1983, 39, 2315−
2124
DOI: 10.1021/acscatal.5b02953
ACS Catal. 2016, 6, 2121−2125

ACS Catalysis
Letter
2322. (d) Brunet, J. J.; Caubere, P. J. Org. Chem. 1984, 49, 4058−4060.
(e) Sajiki, H.; Mori, S.; Ohkubo, T.; Ikawa, T.; Kume, A.; Maegawa,
T.; Monguchi, Y. Chem.Eur. J. 2008, 14, 5109−5111. (f) Lee, Y.;
Motoyama, Y.; Tsuji, K.; Yoon, S. H.; Mochida, I.; Nagashima, H.
ChemCatChem 2012, 4, 778−781. (g) Mitsudome, T.; Yamamoto, M.;
Maeno, Z.; Mizugaki, T.; Jitsukawa, K.; Kaneda, K. J. Am. Chem. Soc.
2015, 137, 13452−13455.
(3) Radkowski, K.; Sundararaju, B.; Furstner, A. Angew. Chem., Int.
Ed. 2013, 52, 355−360.
(4) (a) Inoue, Y.; Takamuku, S.; Sakurai, H. J. Phys. Chem. 1977, 81,
7−11. (b) Weiss, R.; Warshel, A. J. Am. Chem. Soc. 1979, 101, 6131−
6133. (c) Saltiel, J.; Neuberger, K. R.; Wrighton, M. J. Am. Chem. Soc.
1969, 91, 3658−3659.
(5) Pitchumani, K.; Joy, A.; Prevost, N.; Ramamurthy, V. Chem.
Commun. 1997, 127−128.
(6) Wang, Z. In Comprehensive Organic Name Reactions and Reagents;
John Wiley & Sons: New York, 2010; pp 2062−2066.
(7) Horiuti, J.; Polanyi, M. Trans. Faraday Soc. 1934, 30, 1164−1172.
(8) Furukawa, S.; Ochi, K.; Luo, H.; Miyazaki, M.; Komatsu, T.
ChemCatChem 2015, 7, 3472−3479.
(9) Palczewska, W.; Jablonski, A.; Kaszkur, Z.; Zuba, G.; Wernisch, J.
J. Mol. Catal. 1984, 25, 307−316.
(10) Studt, F.; Abild-Pedersen, F.; Bligaard, T.; Sorensen, R. Z.;
Christensen, C. H.; Norskov, J. K. Science 2008, 320, 1320−1322.
2125
DOI: 10.1021/acscatal.5b02953
ACS Catal. 2016, 6, 2121−2125