The Pd4(dppm)4(H)22+ Cluster: A Precatalyst for the Homogeneous Hydrogenation of Alkynes

Article abstract of DOI:10.1021/ic0347992

The catalytic properties of the title cluster toward the homogeneous hydrogenation of phenylacetylene, diphenylethyne and phenyl-1-propyne have been investigated as a function of temperature, pressure, solvents, substrate and cluster concentrations, and counterions. The title cluster is a precatalyst that exhibits a good catalytic activity under mild conditions (1 atm of H ₂ at 20 °C) for the hydrogenation of alkynes and alkenes. For the alkyne substrates, the turnover frequencies (tof's) range between 200 and 500 h^-1, and the product distribution varies as: cis-products, 75-90%; trans-products; 0-8% after 3 h of reaction. Based on the graphs -d[substrate]/dt vs [Pd₄]^1/2, the mechanism indicates a cluster dissociation into two dimers (presumably of the type Pd ₂(dppm)₂(H)(solvent)⁺). The variations of tof (or -d[substrate]/dt) as a function of [substrate] and pressure of H ₂ are linear. At 1600 psi of H₂, the tof can reach 2500-3000 h^-1 (in THF). The tof also increases with temperature reaching a maximum at ～35 °C (tof: 1000-1300 h^-1), but at higher temperatures cluster decomposition begins to occur, leading to a rapid decrease in rates of catalysis. At 50 °C, no catalysis is observed. The hydrogenation reaction can be stopped at the corresponding cis-alkenes with ～95% yields, depending on the substrate and experimental conditions used. The tof's also vary with the solvent, where stronger coordinating solvent molecules give higher tof's. In addition, the tof's do not change with the nature of the counterion, which acts as spectator in the catalysis.

Full text of DOI:10.1021/ic0347992

Inorg. Chem. 2004, 43, 790−796
The Pd4(dppm)4(H)22+ Cluster: A Precatalyst for the Homogeneous
Hydrogenation of Alkynes
,1b
David Evrard,1a,b Katherine Groison,1b Yves Mugnier,1a and Pierre D. Harvey*
Contribution from the Laboratoire de Synthe`se et d’EÄ lectrosynthe`se Organome´talliques,
(LSEO-CNRS UMR 2595), Faculte´ des Sciences Gabriel, UniVersite´ de Bourgogne,
6 BouleVard Gabriel, 21000 Dijon, France, and the De´partement de Chimie, UniVersite´ de
Sherbrooke, Sherbrooke, Que´bec, Canada J1K 2R1
Received July 10, 2003
The catalytic properties of the title cluster toward the homogeneous hydrogenation of phenylacetylene, diphenylethyne
and phenyl-1-propyne have been investigated as a function of temperature, pressure, solvents, substrate and
cluster concentrations, and counterions. The title cluster is a precatalyst that exhibits a good catalytic activity under
mild conditions (1 atm of H at 20 °C) for the hydrogenation of alkynes and alkenes. For the alkyne substrates, the
2
turnover frequencies (tof’s) range between 200 and 500 h-1, and the product distribution varies as: cis-products,
75−90%; trans-products; 0−8% after 3hof reaction. Based on the graphs −d[substrate]/dt vs [Pd4]1/2, the mechanism
indicates a cluster dissociation into two dimers (presumably of the type Pd2(dppm)2(H)(solvent)+). The variations of
tof (or −d[substrate]/dt) as a function of [substrate] and pressure of H are linear. At 1600 psi of H , the tof can
2
2
reach 2500−3000 h-1 (in THF). The tof also increases with temperature reaching a maximum at ∼35 °C (tof:
1000−1300 h-1), but at higher temperatures cluster decomposition begins to occur, leading to a rapid decrease in
rates of catalysis. At 50 °C, no catalysis is observed. The hydrogenation reaction can be stopped at the corresponding
cis-alkenes with ∼95% yields, depending on the substrate and experimental conditions used. The tof’s also vary
with the solvent, where stronger coordinating solvent molecules give higher tof’s. In addition, the tof’s do not
change with the nature of the counterion, which acts as “spectator” in the catalysis.
Introduction
plane, resembling picket fences.4 This cyclic structure has
been confirmed from X-ray crystallography for a close
derivative issued from the redox reaction between the title
cluster and the reducing tetraphenylborate anion to form the
diamagnetic [Pd4(dppm)4(H)]BPh4 species.5 The title cluster
is also a precursor of the electrocatalytic reduction of H+
into H2 and oxidative formate decomposition.6 Homogeneous
catalyzed hydrogenations are clearly important processes,7
but for Pd-catalyzed reactions, the number of published
works appears to be relatively smaller than for other active
transition metal complexes.8 A close look at the known
The synthesis of a purple compound active for the catalytic
hydrogenation of alkynes and alkenes was reported by Kirss
and Eisenberg over 10 years ago2 but was of an unknown
composition for several years until it was confidently
identified as the title cluster as characterized by our teams
using various spectroscopic tools.3 The structure consists of
a cyclic Pd4 frame supported by four bridging dppm ligands
(dppm ) bis(diphenylphosphino)methane), in which the two
hydrides are bonded to the Pd centers; in solution the
complex exhibits a rapid fluxional process at room temper-
ature, keeping a C4-symmetry of the molecule. A closer look
at the structure suggests that two cavities described by the
16 dppm-phenyl groups are formed above and below the Pd4
(4) Meilleur, D.; Harvey, P. D. Can. J. Chem. 2001, 79, 552.
(5) Evrard, D.; Meilleur, D.; Drouin, M.; Mugnier, Y.; Harvey, P. D. Z.
Anorg. Allg. Chem. 2002, 628, 2286.
(6) Meilleur, D.; Rivard, D.; Harvey, P. D.; Gauthron, I.; Lucas, D.;
Mugnier, Y. Inorg. Chem. 2000, 39, 2909.
* To whom correspondence should be addressed: Telephone: (819) 821-
2005. Fax: (819) 821-8017. E-mail: pharvey@courrier.usherb.ca.
(1) (a) Universite´ de Bourgogne. (b) Universite´ de Sherbrooke.
(2) Kirss, R. U.; Eisenberg, R. Inorg. Chem. 1989, 28, 3372.
(3) Gauthron, I.; Gagnon, J.; Zhang, T.; Rivard, D.; Lucas, D.; Mugnier,
Y.; Harvey, P. D. Inorg. Chem. 1998, 37, 1112.
(7) (a) James, B. R. Homogeneous Hydrogenation; Wiley: New York,
1973. (b) James, B. R. In ComprehensiVe Organometallic Chemistry;
Wilkinson, G., Stone, F. G. A., Abel, E. W., Eds.; Pergamon: Oxford,
1982; Vol. 8, Chapter 51. (c) Chaloner, P. A.; Esteruelas, M. A.; Jo´o,
F.; Oro, L. A. Homogeneous Hydrogenation; Kluwer Academic:
Dordrecht, Netherlands, 1994.
790 Inorganic Chemistry, Vol. 43, No. 2, 2004
10.1021/ic0347992 CCC: $27.50 © 2004 American Chemical Society
Published on Web 12/25/2003

2+
Pd4(dppm)4(H)2 Cluster
divided by the total number of moles of cluster. These values were
extracted at the very beginning of the reaction where the graphs of
tof vs time were fairly linear. The samples for GC/MS analyses
were collected with an Ar-purged syringe, degassed with Ar-
bubbling, and stored in the dark in the freezer (-5 °C) until analysis.
High-Pressure Hydrogenations. The hydrogenation reactions
under high pressure were performed in a 200-mL Parr Instrument
autoclave (4001 model) equipped with a gas inlet-outlet valve and
a liquid sampling valve. Mechanical stirring provided agitation. In
this contribution, no heating was provided when high-pressure
experiments were performed. A 100-mL solution of THF (or other
desired solvents) containing 4 × 10-6 mol of [Pd4(dppm)4H2](BF4)2
(or other salts depending on the experiments), 0.004 mol of
substrate, and 0.004 mol of methylcyclohexane was placed inside
the reactor. The reactor was purged three times with H2 gas at the
desired pressure, while the stirring was activated. All the runs were
24 h.
homogeneous Pd catalysts indicates that the active species
all appear to be mononuclear,8 except for one material, a
species formulated as {Pd5(PPh)2}n. In this case, no structural
characterization is available, but this catalyst turns out to be
very efficient.8b To our knowledge, besides this material, the
title cluster and Pd2(dppm)3,9 there is no other polynuclear
palladium species that has been investigated for its homo-
geneous catalyzing hydrogenation properties. One wonders
what the effects of the presence of more than one metal atom
in the active catalysts are, with respect to cooperativity.
We now report the catalytic properties of the title cluster
toward the homogeneous hydrogenation of phenylacetylene,
diphenylethyne, and phenyl-1-propyne as a function of
temperature, pressure, solvents, cluster and substrate con-
centrations, and counterions. The mechanism involves the
cluster dissociation into active d9-d9 dimers.
Variable Concentration Hydrogenations. The hydrogenation
reactions using various concentrations of cluster (4 × 10-6, 10-5
,
Experimental Section
2 × 10-5, 6 × 10-5, 8 × 10-5, 2 × 10-4, and 10-4 M) and substrates
-
Materials. The title cluster as BF4-, Cl-, and BPh4 salts has
(4 × 10-3, 10-2, 2 × 10-2, 6 × 10-2, 8 × 10-2, 10-1, 2 × 10-1
,
been prepared according to literature procedures.3,6 The phenyl-
acetylene, diphenylethyne, 1-phenyl-1-propyne, cyclohexene, and
methylcyclohexane starting substrates and the corresponding hy-
drogenated products styrene, cis- and trans-stilbene, cis- and trans-
phenyl-1-propene, ethylbenzene, 1,2-diphenylethane, and propyl-
benzene used to calibrate the GC/MS instrument, were purchased
from Aldrich, and were used as received. The solvents THF
(Anachemia), acetonitrile (Fisher), acetone (Fisher), CH2Cl2 (ACP),
pyridine (Aldrich), and DMF (Aldrich) were purified according to
literature procedures.10 The cluster was kept under inert atmosphere
at all times, even in the solid state, for best results and reproduc-
ibility.
and 4 × 10-1 M) were performed at 20 °C and 600 psi of H2. The
procedure was the same as described above.
GC/MS Analyses. The GC/MS analyses were performed with
a Hewlett-Packard 5890 series II spectrometer coupled with a
selective mass detector HP 5971 series. The hydrogenated products
were analyzed by temperature-programmed gas chromatography
(10 °C/min) from 60 to 250 °C with a DB-5MS column. A 0.9
mL/min helium flow has been used. The instrument was calibrated
using authentic samples of the substrates and their hydrogenation
products.
NMR Experiments. Variable-temperature NMR spectra were
recorded on a Bruker WM 300 spectrometer (1H NMR: 300.15
MHz, 31P NMR: 121.497 MHz). The reference was the residual
nondeuterated solvent. The chemical shifts are reported with respect
to TMS (1H NMR) and H3PO4 (31P NMR).
UV-Visible Experiments. UV-visible spectra were recorded
on a Hewlett-Packard (HP 8452A) diode array spectrophotometer.
Space-Filling Model and Simulations. The space-filling model
shown in Figure 3 was generated using the commercially available
PC-model 7.0 (Serena software) which uses the MMX empirical
parameters, and using ORTEP 32 and POV-Ray 3.5. The former
software is used to compute the skeleton for Pd4, while ORTEP
and POV-Ray generate the space-filling drawing. For the PC-model
computations, no particular constraint on bond lengths or angles
was applied, but comparisons with known structures containing the
same Pd2(dppm)2 skeleton from the literature were made, to ensure
the validity of the results. Simulated curves of Figures 2 and 5
were obtained from the regression function of the commercially
available Sigma Plot 2001 (Version 7.0) software.
Variable-Temperature Hydrogenation. The hydrogenation
reactions at 1 atm of H2 as a function of temperature were performed
using two 250-mL round flasks connected to each other with a
cannula. The first one contains 100 mL of THF, and H2 bubbling
was allowed to pass through this first flask to minimize the effect
of evaporation with time. In these experiments, the level of solvent
in the reaction flask never changed. The second flask contained 50
mL of THF, 0.002 mol of substrates ([substrate] ) 0.04 M), and
0.002 mol of methylcyclohexane as internal standard ([methylcy-
clohexane] ) 0.04 M). The whole system was allowed to reach
the desired temperature and purged for 1 h prior to addition of 2 ×
10-6 mol of Pd4(dppm)4H2(BF4)2 ([Pd4] ) 4 × 10-5 M). The
turnover frequency (tof) is defined as the number of moles of
consumed starting alkyne (or alkene when it applies) per hour,
(8) See, for examples: (a) Tsuji, J. Palladium Reagents and Catalysts;
Wiley: New York, 1995. (b) Moiseev, I. I.; Vargaftic, M. N. New J.
Chem. 1998, 1217. (c) van Asselt, R.; Elsevier, C. J. J. Mol. Catal.
1991, 65, L13. (d) Tani, K.; Ono, N.; Okamoto, S.; Sato, F. J. Chem.
Soc., Chem. Commun. 1993, 386. (e) Tour, J. M.; Pendalwar, S. L.
Tetrahedron. Lett. 1990, 31, 4719. (f) Suzuki, N.; Tsukanaka, T.;
Nomoto, T.; Ayaguchi, Y.; Izawa, Y. J. Chem. Soc., Chem. Commun.
1983, 515. (g) Borriello, C.; Ferrara, M. L.; Orabona, I.; Panunzi, A.;
Ruffo, F. J. Chem. Soc., Dalton Trans. 2000, 2545. (h) Pelagatti, P.;
Bacchi, A.; Carelli, M.; Costa, M.; Fochi, A.; Ghidini, P.; Leporati,
E.; Masi, M.; Pelizzi, C.; Pelizzi, G. J. Organomet. Chem. 1999, 583,
94.
(9) Stern and Maples examined the catalytic activity of this complex with
respect to homogeneous hydrogenation. Because the complex was
handled in air, the true nature of the catalyst may be unknown. Stern,
E.; Maples, P. K. J. Catal. 1972, 27, 120.
(10) (a) Perrin, D. D.; Armarego, W. L. F.; Perrin, D. R. Purifications of
Laboratory Chemicals; Pergamon Press: Oxford, 1966. (b) Gordon,
A. J.; Ford, R. A. The Chemist’s Companion, a Handbook of Practical
Data, Techniques and References; Wiley: New York, 1972; p 436.
Results and Discussion
Hydrogenation under Mild Conditions. The [Pd4(dppm)4-
(H)2](BF4)2 cluster catalyses the homogeneous hydrogenation
of phenylacetylene, diphenylethyne, and phenyl-1-propyne
at 20 °C and 1 atm of H2 in THF with tof’s 500, 200, and
500 h-1, respectively, to form the semi- (cis- and trans- when
it applies) and fully hydrogenated corresponding products
(the definition for tof is provided in the Experimental
Section). These values are good when compared with
literature data for other catalysts (Table 1).11-13 However,
the values still fall short when compared with the best known
Inorganic Chemistry, Vol. 43, No. 2, 2004 791

Evrard et al.
Table 1. Literature tof Comparison for Homogeneous Hydrogenation
of Phenylacetylenea
Figure 1. Typical example of product evolution during a homogeneous
hydrogenation. Here the reaction is for diphenylethyne in THF at 20 °C
and 1 atm of H2 using the [Pd4(dppm)4(H)2](BF4)2 cluster as catalyst (b )
alkyne; 9 ) alkane; 0 ) cis-alkene; 4 ) trans-alkene).
Table 2. Product Distribution in THF after 24 h at 20 °C/1 atm H2
alkyne
(%)
cis-alkene
trans-alkene
alkane
(%)
substrate
(%)
(%)
PhCtCH
PhCtCMe
PhCtCPh
0
0
0
35.2
56.4
6.4
s
16.4
1.4
64.8
27.2
92.2
a Other works reporting the homogeneous hydrogenation of phenylace-
tylene under mild conditions exist (see, for a recent example: Bacchi, A.;
Carcelli, M.; Gabba, L.; Ianelli, S.; Pelagatti, P.; Pelizzi, G.; Rogolino, D.
Inorg. Chim. Acta 2003, 342, 229), but the tof values are not reported.
b The tof value at 1 atm of H2 is not reported.
Pd catalyst, a species formulated as {Pd5(PPh)2}n.8b,12 Other
reports on efficient catalytic conversion of phenylacetylene
in ethylbenzene at 20 °C/1 atm of H2 by palladium species
(where complete reaction occurs within 3 h, for example)
also exist,13 but the lack of experimental data for tof does
not allow making any meaningful comparison.
The evolution of the products as a function of time (see
Figure 1 as an example for phenylacetylene) shows that
styrene and the kinetic products, cis-stilbene and cis-1-
phenyl-1-propene, are the first products to be observed,
reaching a maximum concentration within ∼3 h. This
maximum value (or yield at a given time) varies from 75 to
90% and is somewhat, at first glance, similar or better than
other mononuclear M(0) catalysts14 but falls shorter than that
found for some other mononuclear Pd(0) complexes of
bidentate nitrogen-containing donors, and thiosemicarba-
zone- and thiobenzoylhydrazone-containing Pd(II) catalysts
(g90%).15
Figure 2. Evolution of the UV-vis spectra of [Pd4(dppm)4(H)2](BF4)2 in
THF at 20 °C, monitoring its stability over a period of 24 h (curve a ) 0,
progressing by intervals of 1 h until 8 h, curve i ) 24 h).
where the addition of H2 onto PhCtCD led to the cis-
addition PhCHdCHD product.2 However, the thermody-
namic trans-alkene products are also observed in this work,
but always in a smaller quantity. At room temperature, the
maximum trans-/cis- ratio is about 1/4 after 24 h (Table 2),
but is generally smaller for most other experiments. The
quantity of alkanes increases with reaction time, while the
amount of alkenes decreases, consistent with a stepwise
mechanism. No reactivity is observed between the cluster
The presence of observable alkenes is consistent with the
greater reactivity of the alkynes,16 and the favored cis-
stereochemistry agrees with Kirss and Eisenberg’s findings
(11) See, for examples: (a) Castiglioni, M.; Deabate, S.; Giordano, R.;
King, P. J.; Knox, S. A. R.; Sappa, E. J. Organomet. Chem. 1998,
571, 251. (b) Adams, R. D.; Barnard, T. S.; Li, Z.; Wu, W.; Yamamoto,
J. H. J. Am. Chem. Soc. 1994, 116, 9103. (c) Adams, R. D.; Li, Z.;
Swepson, P.; Wu, W.; Yamamoto, J. J. Am. Chem. Soc. 1992, 114,
10657.
(12) Berenblyum, A. S.; Aseeva, A. P.; Lakham, L. I.; Petrovsky, V. V.;
Moiseev, I. I. IzV. Akad. Nauk SSSR, Ser. Khim. 1980, 1227 (in
Russian).
(13) (a) Bianchini, C.; Meli, A.; Peruzzini, M.; Vizza, F.; Zanobini, F.;
Frediani, P. Organometallics 1989, 8, 2080. (b) Frediani, P.; Giannelli,
C.; Salvini, A.; Ianelli, S. J. Organomet. Chem. 2003, 667, 197.
(14) See, for exampl: (a) van Laren, M. W.; Duin, M. A.; Klerk, C.; Naglia,
M.; Rogolino, D.; Pelagatti, P.; Bacchi, A.; Pelizzi, C.; Elsevier, C. J.
Organometallics 2002, 21, 1546. (b) Sodeoka, M.; Shibasaki, M. J.
Org. Chem. 1985, 50, 1147.
1
and the substrate under Ar as monitored by H NMR.
The Pd4 Thermal Stability. The graph of tof as a function
of temperature (see Figure 2 as an example) shows an
increase between 20 and 35 °C and then a decrease down to
zero at 50 °C. Above 40 °C, a rapid cluster decomposition
(15) (a) Pelagatti, P.; Venturini, A.; Leporati, A.; Carcelli, M.; Costa, M.;
Bacchi, A.; Pelizzi, G.; Pelizzi, C. J. Chem. Soc., Dalton Trans. 1998,
2715. (b) van Laren, M. W.; Elsevier, C. J. Angew. Chem., Int. Ed.
1999, 38, 3715.
(16) Trost, B. M.; Braslau, R. Tetrahedron. Lett. 1989, 30, 4657.
792 Inorganic Chemistry, Vol. 43, No. 2, 2004

2+
Pd4(dppm)4(H)2 Cluster
is observed in the presence or absence of any substrate. This
high-temperature decomposition is accompanied by the
deposition of a black Pd-containing precipitate; a material
that does not further hydrogenate alkynes or alkenes in our
experimental conditions. No solid deposit is observed in
experiments where the temperature does not exceed 40 °C
within 24 h, but a slow decomposition occurs over a period
of a few days at 20 °C. Again this solid is not active for the
hydrogenation reactions investigated in this work. The bell-
shape curve (Supporting Information) is approximately the
same for all three substrates.
The monitoring of this decomposition process by UV-
vis spectroscopy (Figure 2) allows the extraction of the rates
of decomposition. The graph of log(A - A ) vs time (Figure
2, see inset) indicates a first-order reaction. The slopes
extracted from the graphs log(A - A ) vs time are -1.0 ×
10-3, -1.9 × 10-3, -5.8 × 10-3, -1.6 × 10-2, and
-4.6 × 10-2 h-1 for MeCN, DMF, THF, Py, and CH2Cl2,
respectively. MeCN, DMF, and THF exhibit slower rates of
decomposition, while for Py and CH2Cl2, reactivity with Pd4
is noticed, and the observed slopes are 1 order of magnitude
greater. CH2Cl2 reacts with the title cluster to form Pd2(dppm)2-
Cl2 (minor) and Pd(dppm)Cl2 (major product) agreeing with
Kirss and Eisenberg’s previous findings.2 The pyridine
solvent (Py) also reacts with the cluster to form a new, but
uncharacterized species, which is discussed further below.
Figure 3. Space-filling model for the Pd4(dppm)4(H)22+ cluster showing
the cavity giving access to the Pd atoms. This is one possible conformation
as an example, due to the multiple relative conformations of the dppm
ligands and orientations of the phenyl groups.
In the presence of H2 (instead of Ar), cluster decomposition
occurs more rapidly (based on 1H NMR and UV-vis:
Supporting Information), which suggests reactivity between
H2 and the title cluster, presumably producing unstable Pd-H
intermediates. From 1H and 31P NMR, no intermediate could
be detected under all experimental conditions used in this
work at 1 atm of H2.17
Figure 4. Graph of -d[phenylacetylene]/dt vs [Pd4]1/2 showing a linear
dependence. Experimental conditions: THF at 20 °C and 600 psi of H2
using the [Pd4(dppm)4(H)2](BF4)2 cluster as catalyst.
Mechanism. The space-filling model for Pd4(dppm)4(H)22+
(Figure 3) shows a cavity that is clearly too small to allow
any efficient interactions between the metal atoms and the
substrates in a side-on geometry, as verified from modeling.
However, the relatively impressive tof value of 500 h-1 at
room temperature for both PhCtCH and PhCtCPh strongly
The monitoring by GC/MS of the products’ distribution
with time at high temperatures (i.e., ∼45 °C) shows a rapid
substrate consumption (<2-3 h), but the complete produc-
tion of the corresponding alkanes is never observed, even
after 24 h. Instead, the typical bluish color of the solutions
changes to yellow (with a small black Pd-containing suspen-
sion), witnessing the catalyst’s decomposition. The maximum
amount of intermediate alkenes is observed after 2 h in all
cases. For styrene, this peak amount fluctuates between 80
and 90%, while for cis- and trans-stilbenes, their corre-
sponding values are 80 ( 5 and about 10%, respectively.
Homogeneous System. The presence of cluster decom-
position at high temperatures calls into questions the nature
of the catalysis (homogeneous vs heterogeneous). Attempts
to hydrogenate cyclohexene were made, knowing that this
substrate is not or only very slowly hydrogenated by
homogeneous Pd species but is readily hydrogenated by
colloidal or heterogeneous Pd.15b Indeed, cyclohexene is not
at all hydrogenated by Pd4 in THF (600 psi of H2 at 20 °C)
after several hours, demonstrating the likely homogeneous
character of this catalyst.18
2+
suggests that Pd4(dppm)4(H)2 is not the active catalyst.
This observation leads one to investigate the dependence
of the cluster and substrate concentrations, and H2 pressure
on tof (or -d[alkyne]/dt), and to propose a mechanism in
which the first step involves the dissociation of the title
cluster 1 into less sterically demanding d9-d9 dimers of the
type Pd2(µ-dppm)2(H)(S)+ (2), where S ) solvent (Scheme
1: mechanism adapted for alkynes).19 The graphs -d[alkyne]/
dt versus [Pd4]1/2 are indeed linear (see Figure 4 for PhCt
(18) (a) The mercury test18b was not suitable in this case, since the title
cluster reacts rapidly with Hg. In fact, reactivity between a Pd species
and Hg has been observed before.18c (b) Fusi, A.; Ugo, R.; Psaro, R.;
Braunstein, P.; Dehand, J. J. Mol. Catal. 1982, 16, 217. (c) Harvey,
P. D.; Aye, K. T.; Hierso, K.; Isabel, E.; Lognot, I.; Mugnier, Y.;
Rochon, F. D. Inorg. Chem. 1994, 33, 5981.
(19) d9-d9 dimeric species axially functionalized by good coordinating
solvent molecules have already been reported in the past. See, for
examples: (a) Maekawa, M.; Munakata, M.; Kudora-Sowa, T.;
Suenaga, Y. Polyhedron 1998, 17, 3657. (b) Maekawa, M.; Munakata,
M.; Kuroda-Sowa, T.; Suenaga, Y. Inorg. Chim. Acta 1998, 281, 116.
(c) Hill, R. H.; De Mayo, P.; Puddephatt, R. J. Inorg. Chem. 1982,
21, 3642.
(17) High pressure 1H NMR experiments (pression ) 1000 bar of H2 using
acetone-d6 at 20 °C) performed at the University of Lausanne, also
did reveal any intermediates.
Inorganic Chemistry, Vol. 43, No. 2, 2004 793

Evrard et al.
Scheme 1
Figure 5. Graph of the tof as a function of pressure of H2 for phenyl-1-
propyne in THF at 20 °C using [Pd4(dppm)4(H)2](BF4)2 as catalyst.
of phenyl-1-propyne (as an example) where an increase of
about 10-fold is observed (from 200 to 2500 h-1) between
atmospheric pressure and 1600 psi at 20 °C. This figure
exhibits a first-order dependence on the pressure of H2, which
is consistent with the oxidative addition step of H2 on 4 to
give 5.24 As a result of high pressures, the corresponding
alkane products are generated far more rapidly, and all the
reactions reach completion (i.e., 100% of alkanes) within 2
or 3 h instead of 24 h for 1 atm of H2. The tof vs H2 pressure
graphs exhibit slopes of 1.06, 1.43, and 1.57 h-1/psi for
phenylacetylene, 1-phenyl-1-propyne, and diphenylethyne,
respectively. Intermediate 5 reductively eliminates the al-
1
kenes, and regenerates the active dimer 2. By H and 31P
NMR, no new species has been detected under catalytic
conditions, indicating that the concentration of the intermedi-
ates must be very small, which indicates a small displacement
of the first equilibrium from 1 to 2 in the mechanism of
Scheme 1.
CH as an example), while the graphs -d[alkyne]/dt versus
[Pd4] are not (see one example in the Supporting Informa-
tion), hence demonstrating the dissociation of the cluster into
active species.
The graph of -d[phenylacetylene]/dt vs [phenylacetylene]
is also linear (Supporting Information), illustrating the first-
order dependence. This result is consistent with the second
step of the mechanism where one molecule of substrate
inserts onto the Pd-Pd bond of the active catalyst to give
species 3.20,21 The related [Pt2H(S)(dppm)2(µ-RCCR)]+ com-
plexes (S ) MeCN, MeC(O)Me; R ) CF3) have already
been reported in the literature by Puddephatt and Thomp-
son.22
The kinetic analysis allows us to formulate the following
rate law:
r ) k[alkyne] [Pd4]1/2 P(H2)
(1)
The addition of a 10-fold excess (/[Pd4]) of PPh3 to the
solutions strongly slows down the catalysis. A long induction
period of 15 min where no hydrogenation occurs is observed,
and after 1.5 h the distribution of products is phenylace-
tylene: 95% and styrene: 5%. This result is consistent with
PPh3 competing for the coordination site otherwise occupied
by a solvent molecule on the intermediates 2-4, where the
Then, the coordinated alkyne migration-insertion onto the
Pd-H bond of 3 produces another d9-d9 intermediate, 4,
which undergoes an oxidative addition with H2 to form a
suspected species Pd2(µ-dppm)2(H)(η1-CRdCHR′)(µ-H)+
5.23 Figure 5 shows tof vs H2 pressure for the hydrogenation
(23) The lack of literature precedent for this complex 5, or related
compounds, is likely due to the well-known fragility of Pd-H
species.24a However, the presence of stable species such as Pd2(µ-
dppm)2(aryl)2(µ-H)+ (aryl ) various substituted benzenes),23a Pd2-
(20) Addition of alkynes onto Pd-Pd bonds is well documented. See, for
examples: (a) Davies, J. A.; Pinkerton, A. A.; Syed, R.; Vilmer, M.
J. Chem. Soc., Chem. Commun. 1988, 47. (b) Davies, J. A.;
Kirschbaum, K.; Kluwe, C. Organometallics 1994, 13, 3664. (c)
Kluwe, C.; Muller, J.; Davies, J. A. J. Organomet. Chem. 1996, 526,
385. (d) Kluwe, C.; Davies, J. A. Organometallics 1995, 14, 4257.
(e) Balch, A. L.; Lee, C.-L.; Lindsay, C. H.; Olmstead, M. M. J.
Organomet. Chem. 1979, 177, C22. (f) Lee, C.-L.; Hunt, C. T.; Balch,
A. L. Inorg. Chem. 1981, 20, 2498. (g) Higgins, S. J.; Shaw, B. L. J.
Chem. Soc., Dalton Trans. 1988, 457. (h) Higgins, S. J.; Shaw, B. J.
J. Chem. Soc., Chem. Commun. 1986, 1629.
(dppm)2(aryl)(Et)(µ-H)+ 23b,c and Pt2(dppm)2(H)2(µ-H)+,23d,e suggests
,
that such a formulation is reasonable. (a) Grushin, V. V. Chem. ReV.
1996, 96, 2011. (b) Stockland, R. A., Jr.; Anderson, G. K.; Rath, N.
P. J. Am. Chem. Soc. 1999, 121, 7945. (c) Stockland, R. A., Jr.;
Anderson, G. K.; Rath, N. P. Inorg. Chim. Acta 2000, 300, 395. (d)
Brown, M. P.; Puddephatt, R. J.; Rashidi, M.; Seddon, K. R. Inorg.
Chim. Acta 1977, 23, L33. (e) Brown, M. P.; Puddephatt, R. J.; Rashidi,
M.; Seddon, K. R. J. Chem. Soc., Dalton Trans. 1978, 516.
(24) Attempts to synthesize intermediates 2-5 stubbornly failed. For
instance, addition of one equivalent of NaBH4 to Pd2(dppm)2Cl2 in
the presence of a 10-fold excess of phenylacetylene resulted in a 1:1
mixture of starting material and [Pd4(dppm)4H2](Cl)2. The alkyne
substrate acted as a spectator and 2 was never observed.
(21) No 1,2-shift reaction20b,c was observed either by 1H NMR or the
appearance of other peaks in the GC/MS.
(22) Puddephatt, R. J.; Thomson, M. A. Inorg. Chem. 1982, 21, 725.
794 Inorganic Chemistry, Vol. 43, No. 2, 2004

2+
Pd4(dppm)4(H)2 Cluster
Pd-PPh3 bond is stronger. This PPh3 molecule greatly slows
down the oxidative addition of H2 onto intermediate 4 to
give 5.25
(or alkynes) may be possible. For alkene substrates, such a
process renders cis-trans isomerization possible. All in all,
this proposal is highly speculative, and the presence of the
minor transient trans-product must be explained by another
mechanism which constitutes a separate study from this work.
Both the cis- and trans-stilbene in THF 20 °C/600 psi of
H2 are converted into 1,2-diphenylethane with nearly com-
parable rates, considering the uncertainties (tof ) 1000 and
800 h-1; for cis- and trans-, respectively), indicating that
Pd4 is able to hydrogenate trans-alkenes as well. A similar
mechanism as shown in Scheme 1 may apply for the
hydrogenation of olefins (Supporting Information).
Properties at High Pressure. The products’ evolution
with time for the hydrogenation reaction of phenylacetylene
at high pressures (200-1600 psi of H2) is similar to that of
Figure 1. Styrene appears and disappears faster with higher
pressures (i.e., within 2 h), but the relative amount of styrene
at its maximum concentration, generally observed after 1 h,
remains fairly constant at around 65 ( 10% for all experi-
ments. This result indicates that the relative rates of
hydrogenation of phenylacetylene and styrene remain very
similar to each other, for all investigated pressures.
For diphenylethyne, similar observations are made. The
maximum amount observed for cis-stilbene fluctuates be-
tween 50 and 65% for all pressures. Similarly, the maximum
quantity of the intermediate trans-stilbene product also stays
the same at about 4%. This observation reinforces the
hypothesis that the apparent regioselectivity is not lost at
high pressures.
The same observations are drawn for the phenyl-1-propyne
substrate, except that the trans-product is never observed at
more than 8%. These experiments above indicate that the
complete production of the corresponding alkanes is best
carried out at higher pressures. Pressure at 200 psi, which is
the lowest investigated pressure in this work, is fully adequate
for this purpose. For the generation of cis-alkenes, best results
are obtained after 2 h using THF as the solvent at 30-40
°C and at 1 atm of H2 (> 75%).
Solvent Dependence. The tof’s for the hydrogenation of
phenyl-1-propyne in various solvents at 20 °C and 600 psi
of H2 have been measured. These values can be divided into
three groups: (1) large for DMF (1800), (2) medium for
THF (1240), MeC(O)Me (1130), and MeCN (1060), and (3)
small for Py (660 h-1). Indeed, complete conversion of
phenyl-1-propyne in propylbenzene (as an example) is
performed within 4 h in DMF instead of, say, 12 h in THF.
This result is not unprecedented for us, as a previous
investigation on the electrocatalytic properties of the title
cluster also indicated that better results are obtained for this
solvent for the reduction of H+ and oxidative decomposition
- 6
of HCO2 . The medium-sized tof values for THF, MeC-
(O)Me, MeCN (which cannot be separated on the basis of
the uncertainties) reflect a lower coordinating ability of
the solvent molecules with respect to DMF (i.e. DMF >
THF ≈ MeC(O)Me) on the intermediates 2-4 shown in
Scheme 1.27
cis- vs trans-Products. Occasional amounts of minor
trans-alkene products were observed as stated. Pd4 does not
induce the cis-trans isomerization, or at least not efficiently
Py exhibits a significantly lower tof (660 h-1). The results
are obscured by the presence of a reactivity with this solvent
where a yellow coloration appears with time.28 With this
solvent, the hydrogenation of phenylacetylene at 600 psi of
H2 at 20 °C leads to 75% of styrene and 25% of ethylbenzene
after 45 min. Except for the relatively rapid appearance of
styrene, these proportions do not differ greatly from THF at
the same pressure. However for 1,2-diphenyl-acetylene, cis-
and trans-stilbene and diphenylethane were detected with a
proportion of 80, 2, and 17%, respectively, after 24 h.
Overall, these observations are consistent with catalyst
poisoning; however, the cis-/trans- ratio has greatly im-
proved. For 1-phenyl-1-propyne, 97% of cis-1-phenyl-1-
propene, and only 3% of propylbenzene are obtained after
4 h of catalysis. These results indicate that the hydrogenation
can be stopped at the semihydrogenated products.29
1
(in the absence of H2 as monitored by H NMR and GC/
MS) using both cis- and trans-alkenes as starting materials.
This result is interesting since it implies that a second
mechanism, involving different intermediates, must operate
to explain the presence of the trans-alkene products. Specu-
latively, 1 and 2 in Scheme 1 have not been considered to
2+
add H2 to form the corresponding Pd4(dppm)4(H)4 (1-
+
H2) and Pd2(dppm)2(H)3 (2-H2) potential intermediates.
The former (1-H2) has already been suspected in the past
to be one of the active intermediates in the H2-evolution
reaction catalyzed by 1.6 The second potential intermediates
(2-H2) are not known, but the corresponding platinum
analogue Pt2(dppm)2(H)3 is.26 For such reactive H-rich
+
species, atom or hydride transfer to uncoordinated alkenes
-
-
The Counterions. The counterions BF4 , BPh4 , and Cl-
exhibit no important effect on tof within the experimental
uncertainties. For example, the tof’s for phenyl-1-propyne
in THF at 20 °C at 600 psi of H2 are 1000, 1333, and 1000
(25) (a) An additional experiment where two competitive substrates (i.e.
[diphenylethyne] ) [cis-stylbene] ) 0.02 M, at 20 °C, and 600 psi of
H2) were inserted with the cluster was performed to demonstrate that
the alkyne substrate is preferentially hydrogenated over the alkene.
The alkyne substrate is hydrogenated first, resulting in an increase of
the total [cis-alkene]. Then, [cis-alkene] starts decreasing after 20 min.
2+
(b) Addition of 7 equiv of PPh3 to Pd4(dppm)4(H)2 in (CD3)2CO
(27) Cotton, F. A.; Wilkinson, G.; Gaus, P. Basic Inorganic Chemistry,
3rd ed.; Wiley: Toronto, 1987; p 208.
leads to a new product that exhibits a peak at +8.3 ppm in the 31P
NMR spectra. No attempt was made to identify the species. A
structurally related compound, [Pd2(dppm)2(PPh3)(C6Cl5)](BPh4), has
been reported in the literature.25c (c) Espinet, P.; Fornies, J.; Fortunato,
C.; Hidalgo, G.; Martinez, F.; Thomas, M.; Welch, A. J. J. Organomet.
Chem. 1986, 317, 105.
(28) (a) Reactivity of pyridine with d9-d9 and d8-d8 Pd2(dppm)2 species
is precedented.27b-d (b) Balch, A. L.; Hunt, C. T.; Lee, C.-L.; Olmstead,
M. M.; Farr, J. P. J. Am. Chem. Soc. 1981, 103, 3764. (c) Uson, R.;
Fornies, J.; Espinet, P.; Martinez, F.; Fortuno, C.; Menjon, B. J.
Organomet. Chem. 1983, 256, 365. (d) Chin, C. C. H.; Yeo, J. S. L.;
Loh, Z. H.; Vittal, J. J.; Henderson, W.; Hor, T. S. A. J. Chem. Soc.,
Dalton Trans. 1998, 3777.
(26) Brown, M. P.; Fisher, J. R.; Hill, R. H.; Puddephatt, R. J.; Seddon, K.
R. Inorg. Chem. 1981, 20, 3516.
Inorganic Chemistry, Vol. 43, No. 2, 2004 795

Evrard et al.
-
-
h-1 for BF4 , BPh4 , and Cl-, respectively. This result is
consistent with Scheme 1 where the counterion is a “specta-
tor”.
Acknowledgment. P.D.H. thanks NSERC (Natural Sci-
ences and Engineering Research Council). Y.M. is grateful
to CNRS (Centre National de la Recherche Scientifique) and
Conseil Re´gional de Bourgogne for funds. Dr. Lothar Helm
and Professor Andre´ E. Merbach (EÄ cole Polytechnique
Fe´de´rale de Lausanne, Switzerland) are acknowledged for
their help with the
pressure. Professor Jean Lessard is acknowledged for fruitful
discussion.
Conclusions
The title cluster is one of the very good Pd-catalyst for
the homogeneous hydrogenation of alkynes, despite the fact
that decomposition is observed at temperatures exceeding
45 °C. The mechanism proceeds via a cluster dissociation,
and depending on the experimental conditions, selectivity
for cis-trans isomers and semi- and fully hydrogenated
products is possible.
1H NMR measurements under high
Supporting Information Available: Graph of -d[phenylace-
tylene]/dt vs [Pd4]. Experimental conditions: THF at 20 °C and
600 psi of H2 using the [Pd4(dppm)4(H)2](BF4)2 cluster as catalyst,
graph of -d[phenylacetylene]/dt) vs [phenylacetylene] at 600 psi
of H2 at 20 °C, graphs of log A vs time for [Pd4(dppm)4(H)2](BF4)2
in THF at 20 °C in the presence of 1 atm of H2 and diphenylethyne,
graph of the tof as a function of temperature for phenyl-1-propyne
in THF at 1 atm H2 using [Pd4(dppm)4(H)2](BF4)2 as catalyst, and
possible mechanism for the homogeneous hydrogenation of alkenes
by [Pd4(dppm)4(H)2](BF4)2. This material is available free of charge
via the Internet at http://pubs.acs.org.
(29) (a) Another example where the hydrogenation is stopped at the alkene
using Py as the solvent exists: Kerr, J. M.; Suckling, C. J.; Bamfield,
P. J. Chem. Soc., Perkin Trans. 1 1990, 4, 887. (b) One possible
explanation for this result is the relative lability of the donors. PPh3
is not as labile as THF for instance (the commonly used solvent in
this work), and its presence in the solution leads to a saturation of the
active sites. This coordination strongly inhibits the catalyses. On the
other hand, Py is less labile than THF but more than PPh3. As alkynes
are more reactive than alkenes based on this work, it is possible that
hydrogenation reactions for both alkynes and alkenes are slowed to
the point that the hydrogenation of alkenes are too slow to be observed
in the conditions used.
IC0347992
796 Inorganic Chemistry, Vol. 43, No. 2, 2004