In situ 1H-PHIP-NMR studies of the stereoselective hydrogenation of alkynes to (E)-alkenes catalyzed by a homogeneous [Cp*Ru]+ catalyst

Article abstract of DOI:10.1039/b007201j

The hydrogenation of internal alkynes using a [Cp*Ru(alkene)]⁺ complex leads to the formation of (E)-alkenes. This ruthenium complex represents one of the few homogeneous catalysts that trans-hydrogenate internal alkynes directly and stereoselectively. We have studied its stereoselectivity by in situ PHIP-NMR spectroscopy (PHIP = para-hydrogen induced polarization). With this method the initially formed products can be identified and characterized even at very low concentrations and low conversions. Furthermore, their subsequent fate can be evaluated with high sensitivity and with time resolution. Different alkyne substrates were used to demonstrate the universal applicability of this catalyst. The catalyst is not active in combination with terminal alkynes, however, possibly due to the formation of a rather stable vinylidene complex. A mechanism proceeding via a binuclear complex is proposed to explain the formation of the (E)-alkenes.

Full text of DOI:10.1039/b007201j

View Article Online / Journal Homepage / Table of Contents for this issue
In situ 1H-PHIP-NMR studies of the stereoselective hydrogenation
of alkynes to (E)-alkenes catalyzed by a homogeneous [Cp*Ru]‘
catalyst
Dana Schleyer, Heiko G. Niessen and Joachim Bargon*
Institute of Physical and T heoretical Chemistry, University of Bonn, W egelerstrasse 12,
D-53115 Bonn, Germany. E-mail: bargon=uni-bonn.de; Fax: ]49 228 73 9424
Received (in Freiburg, Germany) 4th September 2000, Accepted 19th December 2000
First published as an Advance Article on the web 9th February 2001
The hydrogenation of internal alkynes using a [Cp*Ru(alkene)]` complex leads to the formation of (E)-alkenes.
This ruthenium complex represents one of the few homogeneous catalysts that trans-hydrogenate internal alkynes
directly and stereoselectively. We have studied its stereoselectivity by in situ PHIP-NMR spectroscopy
(PHIP \ para-hydrogen induced polarization). With this method the initially formed products can be identiÐed
and characterized even at very low concentrations and low conversions. Furthermore, their subsequent fate can be
evaluated with high sensitivity and with time resolution. Di†erent alkyne substrates were used to demonstrate the
universal applicability of this catalyst. The catalyst is not active in combination with terminal alkynes, however,
possibly due to the formation of a rather stable vinylidene complex. A mechanism proceeding via a binuclear
complex is proposed to explain the formation of the (E)-alkenes.
The cationic ruthenium complex [Cp*Ru(g4-CH CH2
(E)-alkenes. A reaction mechanism derived from our results,
which is consistent with claims published by others pre-
viously,4 is proposed.
3
CHCH2CHCOOH)][CF SO ], 1, has been shown to selec-
3
3
tively hydrogenate sorbic acid (trans,trans-2,4-hexadienoic
acid) to (Z)-3-hexenoic acid.1 Mechanistic studies on that
hydrogenation have been undertaken successfully before in
our group.2 Thereby, it has been found that this simple cata-
lyst 1 is an attractive choice for the selective homogeneous
(E)-hydrogenation of alkynes. Normally, homogeneous tran-
sition metal complexes hydrogenate alkynes mainly to (Z)-
alkenes.3 Traditional methods for trans-hydrogenation of
alkynesÈeither homogeneous or heterogeneousÈof which
some yield high amounts of (E)-alkenes, however, frequently
do not tolerate functional groups. Heterogeneous catalysis
typically permits easier removal of the catalyst; by contrast
homogeneous catalysis is much more product speciÐc and
stereoselective, which is desirable for easier and low-cost
workup of the products. For example, totally hydrogenated
and re-arranged products can usually be suppressed or limited
to a certain degree through variation of the reaction condi-
tions and through catalyst design.
Experimental
In order to investigate the initial hydrogenation of alkynes in
situ at low conversion rates, PHIP-NMR spectroscopy
(PHIP \ para-hydrogen induced polarization) was used in
this study to obtain insight into details of the reaction mecha-
nism. The PHIP e†ect gives rise to a characteristically di†er-
ent pattern for every alternative hydrogen transfer possibility
to the substrate, consisting of characteristic emission and
absorption signals in the in situ-recorded NMR spectrum of
the hydrogenation product.5 Both in the intermediates and in
the Ðnal hydrogenation products, not only chemically, but
also magnetically di†erent protons will lead to speciÐc and
distinguishable polarization patterns via which alternate reac-
tion schemes can be discriminated. Since the characteristic
PHIP patterns are only observed if the two para-hydrogen
atoms are transferred pair-wise by the catalyst, that is while
maintaining the pair correlation between the two formerly
para-hydrogen protons via magnetic interaction, one can
obtain detailed and hence subtle information about the hydro-
genation mechanism. Even if the primarily formed reaction
product rearranges subsequently, its original identity mani-
fests itself in the polarization pattern. Whether the rearranged,
that is the eventually formed more thermodynamically stable
product, also appears in polarization depends on the reaction
conditions and the kinetics. As an additional advantage, the
PHIP e†ect leads to a 104-fold increase of the NMR sensi-
tivity, which enables us to detect small amounts of interme-
diates in the catalytic cycle or by-products of the reaction.
Our investigations were performed under a hydrogen pres-
sure of 1 bar at a temperature of 298 K using an NMR
spectrometer operating at a nominal proton resonance fre-
quency of 200 MHz. For all investigations 5 mg of the catalyst
were weighed into a standard 5 mm NMR tube and dissolved
In principle, the thermodynamically more stable (E)-
alkenesÈas formed here via homogeneous hydrogenationÈ
could be the consequence of subsequent isomerizations of ini-
tially formed (Z)-alkenes, whereby the isomerizations do not
occur until a high degree of conversion of the substrate is
reached. The direct homogeneous hydrogenation of an alkyne
to an (E)-alkene by a mononuclear catalyst is difficult to
envision, because the hydrogen transfer would have to occur
from two di†erent sides of the substrate. Therefore, it may not
proceed in one step. More likely is a pathway via a catalyst
that contains two metal centers, since this can lead to the
direct formation of an (E)-alkene. In this contribution, we
report on the stereoselective trans-hydrogenation of internal
alkynes to (E)-alkenes using 1. The key features of this reac-
tion are: mild reaction conditions (hydrogen pressures of only
1 bar and temperatures around 300 K), a high tolerance of
other functional groups within the substrate molecule (e.g.
hydroxyl, carbonyl, acetal), and the selective production of
DOI: 10.1039/b007201j
New J. Chem., 2001, 25, 423È426
423
This journal is ( The Royal Society of Chemistry and the Centre National de la Recherche ScientiÐque 2001

View Article Online
Table 1 Experimental 1H-NMR data of the hydrogenation products
at 200 MHz (in methanol-d , 293 K)
4
R
R
d
d
3J(H H )/Hz
1
2
A
B
A B
CH
CH
CH
CH
CH
5.42
5.46
5.98
6.24
5.44
5.43
6.26
5.57
6.36
5.56
5.43
5.40
5.41
6.07
5.40
5.42
5.36
5.42
6.40
5.44
5.43
6.38
5.68
6.61
5.41
5.64
5.67
5.54
6.99
5.83
15.0
15.0
15.7
15.8
15.0
15.0
16.0
15.6
16.0
15.5
15.4
15.2
15.6
16.1
15.8
3
3
3
3
3
C H
2
5
C3 CCH
3
C H
6
C H
2
5
5
5
5
7
5
5
C H
2
5
5
5
C D
C D
2
2
C H
2
C H
6
CH OH
C H
Fig. 1 OleÐnic region of the 200 MHz 1H-PHIP-NMR spectrum
recorded during the hydrogenation of 3-hexyne-1-ol using the Cp*Ru
catalyst.
2
3
CH OH
CH CH OH
CH(OH)CH CH
CH(OH)CH CH
CH CH(OH)CH
C(O)CH
CH(OC H )
C H
6
2
C H
2
2
2
CH
2
3
3
3
3
C H
2
2
5
5
5
7
C H
2
in 600 ll methanol-d ; 20 ll of the substrate were added sub-
2
C H
2
4
3
sequently. The whole system was purged with argon. This pro-
C H
2
5 2
3
cedure is well established in our group and has been found to
produce good results.
characteristic PHIP polarization pattern. These polarization
signals prove the pair-wise transfer of the para-hydrogen to
the substrate. The observed antiphase coupling constant of
15.5 Hz is a typical value for an oleÐnic trans coupling con-
stant and clearly identiÐes the formation of the corresponding
Results
As a typical example for the stereoselective character of this
type of hydrogenation, we show the 1H-PHIP-NMR spectrum
of 3-hexyne-1-ol using 1 in methanol-d (Fig. 1), exhibiting the
4
Fig. 2 OleÐnic region of the 200 MHz 1H-PHIP-NMR spectrum recorded during the hydrogenation of 1-phenyl-1-propyne at various tem-
peratures using the Cp*Ru catalyst.
424
New J. Chem., 2001, 25, 423È426

View Article Online
Fig. 3 Proposed dinuclear mechanism for the hydrogenation of internal alkynes with the Cp*Ru catalyst 1.
(E)-alkene by trans-hydrogenation of the substrate. Additional
1H-PHIP-NMR data of other (E)-alkenes obtained through
hydrogenation of alkynes with 1 are presented in Table 1.
Whereas for most of the investigated substrates studied here
the stereoselectivity of the formation of the (E)-product was
found to be temperature independent, phenyl-substituted
alkynes represent an exception. Accordingly, PHIP-NMR
experiments revealed a change of selectivity during the hydro-
genation of 1-phenyl-1-propyne depending on the tem-
perature. At 283 K the (E)-alkene is the main product,
whereas the formation of the (Z)-isomer is favored with
increasing temperature. Fig. 2 shows the 1H-PHIP-NMR
spectra recorded during the hydrogenation of 1-phenyl-1-
propyne at di†erent temperatures.
From our PHIP-NMR studies we conclude that any
pathway proceeding via (Z/E)-isomerization for the formation
of the (E)-alkene product can be excluded, because no initial
formation of the necessary (Z)-product can be detected even
at low temperatures. In addition, no (Z/E)-isomerization
occurred when (Z)-3-hexene-1-ol was exposed to catalyst 1
under the usual reaction conditions for the hydrogenation.
This experiment supports the proposed reaction mechanism
(Fig. 3) for the direct trans-hydrogenation.
Control experiments using a homogeneous rhodium cata-
lyst, the Wilkinson catalyst, have shown that in this case the
hydrogenation leads directly to the formation of the (Z)-
alkenes with characteristically di†erent polarization patterns.
Furthermore, when using either the Wilkinson or any other of
the investigated rhodium catalysts, no formation of the (E)-
product could be observed by PHIP-NMR spectroscopy.
Discussion
We have studied the hydrogenation of various terminal or
internal aliphatic alkynes, including some containing various
functional groups. We found that the hydrogenation rates for
these di†erent types of alkynes are not much inÑuenced by
functional groups, but vary signiÐcantly with the position of
the triple bond within the molecule. In particular, alkynes
with internal triple bonds are found to be much more reactive
than those with terminal ones, to the degree that no polarized
products have ever been observed for the latter. Since the
hydrogenation of terminal alkynes proceeds in principle
readily with other transition metal catalysts, such as rhodium,
the lack of activity of this ruthenium catalyst could be due to
the formation of a stable catalystÈsubstrate complex. In con-
trast, using the catalyst 1 the hydrogenation of internal
alkynes proceeds readily and leads to the corresponding (E)-
alkenes with high stereoselectivity. Even if the reaction is
carried out below 280 K, no signals due to the (Z)-alkene are
detected in the in situ PHIP-NMR spectra. The inability of 1
to yield any PHIP polarization when hydrogenating terminal
alkynes might alternatively indicate that it proceeds according
to another mechanism, not ““pair-wiseÏÏ.
Based on steric reasoning a catalyst containing only one
metal center should most likely lead to the formation of
Fig. 4 Formation of
alkynes.
a stable vinylidene complex with teminal
New J. Chem., 2001, 25, 423È426
425

View Article Online
(Z)-alkenes as the main hydrogenation product of internal
alkynes. Therefore, to explain the observed trans-
hydrogenation, a di†erent mechanism has to be considered.
Binuclear rhodium complexes have been reported pre-
viously to e†ect the homogeneous hydrogenation of alkynes
to (E)-alkenes.4 The intermediate is assumed to be a bridged
alkyneÈdihydrogen complex, which enables a di†erent stereo-
chemical pathway from that typical for mononuclear rhodium
complexes. In the case of ruthenium, the formation of dimeric
complexes with alkynes has also been reported before.6 Since
the stereoselectivity observed hereÈas has been pointed out
beforeÈcan only be explained if the reaction takes place
involving two metal centers, we propose the mechanism out-
lined in Fig. 3.
According to this concept, the hydrogenation has to take
place rapidly and without total separation of the two para-
hydrogen atoms. Otherwise the pair correlation of the two
proton spins would be lost and no PHIP-NMR spectrum
should be observed. The observation of a PHIP-NMR spec-
trum and steric considerations clearly favor the existence of a
dimeric complex as the catalytically active species. Further-
more, our results prove that the two hydrogen atoms trans-
ferred to the alkyne stem from the same hydrogen molecule.
In addition, if the (Z)-alkene was formed Ðrst, followed by
subsequent isomerization to the (E)-form, this fact would
manifest itself in the polarization spectrum, where signals
characteristic for the (Z)-alkene should appear. In the event
the (Z)-product would hypothetically exist for a very short
time (a necessary requirement to be considered a product), this
would manifest itself in the form of polarization signals. Fur-
thermore, kinetic considerations would only favor the non-
observability of the (E)-alkene if the corresponding (Z)-alkene
was formed Ðrst but not the other way around; fast relaxation
would more likely eliminate the signals for the (E)-alkene.
Since PHIP-NMR spectroscopy provides the opportunity to
investigate the reaction under initial kinetic conditions
without probable isomerization at high conversions, it cannot
be excluded that subsequent reactions lead to the formation of
some (Z)-isomer. Therefore, by just looking at thermal
product spectra at nearly complete conversion, no distinction
between a trans- or cis-speciÐc hydrogenation would be pos-
sible if both products were Ðnally present.
internal alkynes. We assume that the relatively small demand
for space and the open structure of the resulting activated
complex are important features that characterize and enable
this binuclear mechanism. Sterically hindered substrates may
obstruct the double-sided coordination of the catalyst. This
may apply to phenyl-substituted alkynes as well, but it is also
possible that some other form of speciÐc coordination occurs
between the phenyl group and the active center of the catalyst.
With increasing temperature, the reaction pathway changes
to a mononuclear mechanism that gives rise to the formation
of (Z)-alkenes. Further investigations of phenyl-substituted
alkynes are warranted. As shown in previous studies, the rate
of addition of the dihydrogen to a metal center, not that of the
transfer of the hydrogen to the substrate molecule, is the rate-
determining step at room temperature and 1 bar of hydro-
gen.2
As mentioned before, 1 does not exhibit any hydrogenation
activity towards 1-alkynes. Therefore, we propose that a ter-
minal alkyne complex isomerizes into a vinylidene complex
via a 1,2-hydrogen shift. The corresponding mechanism is out-
lined in Fig. 4. Such an acetyleneÈvinylidene rearrangement is
well known and has been described in the literature pre-
viously.7
Acknowledgements
We thank B. Drieen-Holscher, W. Keim, and S. Steines of
the Institute of Technical and Petrol Chemistry, RWTH
Aachen, for samples of the catalyst, for helpful discussions,
and continuous support during the course of these investiga-
tions. We also thank the German Federal Ministry for Science
and Technology (BMBF), the Deutsche Forschungsgemein-
schaft (DFG), and the Fonds der Chemischen Industrie,
Frankfurt, for Ðnancial support. H. G. N. thanks the
Studienstiftung des deutschen Volkes for a graduate research
fellowship.
References
1
2
3
S. Steines, U. Englert and B. Drieen-Holscher, Chem. Commun.,
2000, 217.
H. G. Niessen, D. Schleyer, S. Wiemann, J. Bargon, S. Steines and
B. Drieen-Holscher, Magn. Reson. Chem., 2000, 38, 747.
(a) R. R. Schrock and J. A. Osborn, J. Am. Chem. Soc., 1976, 98,
2143; (b) M. W. van Laren and C. J. Elsevier, Angew. Chem.,
1999, 111, 3926; M. W. van Laren and C. J. Elsevier, Angew.
Chem., Int. Ed., 1999, 38, 3715.
The Ðnding that no intermediates as outlined in the scheme
depicted in Fig. 3 have been observed in our PHIP studies is
attributed to the fact that in all proposed intermediates the
two transferred hydrogen atoms are chemically and magnetically
equivalent, which makes the detection of these complexes
impossible. We are aware that the reaction mechanism is not
yet fully established and is partly based on chemical reason-
ing; however, we Ðnd the mechanism consistent with the liter-
ature and our results. Therefore, we consider it at least a good
working hypothesis and a starting point for further investiga-
tions and discussions. At this point, kinetic investigations to
prove and substantiate the mechanism further are under the
way.
4
5
(a) R. R. Burch, E. L. Muetterties, R. G. Teller and J. M. Wil-
liams, J. Am. Chem. Soc., 1982, 104, 4257; (b) R. R. Burch, A. J.
Shusterman, E. L. Muetterties, R. G. Teller and J. M. Williams, J.
Am. Chem. Soc., 1983, 105, 3546.
(a) T. C. Eisenschmidt, R. U. Kirss, P. P. Deutsch, S. I. Hommel-
toft, R. Eisenberg, J. Bargon, R. G. Lawler and A. L. Balch, J.
Am. Chem. Soc., 1987, 109, 8091; (b) J. Bargon, in Applied Homo-
geneous Catalysis with Organometallic Compounds, ed. B. Cornils
and W. A. Herrmann, VCH, Weinheim, 1996, vol. II, p. 672.
(a) H. Suzuki, H. Omori, D. H. Lee, Y. Yoshida, M. Fukushima,
M. Tanaka and Y. Moro-oka, Organometallics, 1994, 13, 1129; (b)
H. Omori, H. Suzuki, T. Kakigano and Y. Moro-oka,
Organometallics, 1992, 11, 989.
6
7
Accordingly, 1 represents the Ðrst homogeneous ruthenium
catalyst that is capable of a direct trans-hydrogenation of
J. Silvestre and R. Ho†mann, Helv. Chim. Acta, 1985, 68, 1461.
426
New J. Chem., 2001, 25, 423È426