Kinetic and spectroscopic studies of the [palladium(Ar-bian)]-catalyzed semi-hydrogenation of 4-octyne

Article abstract of DOI:10.1021/ja052729j

The kinetics of the stereoselective semi-hydrogenation of 4-octyne in THF by the highly active catalyst [Pd{(m,m′-(CF₃)₂C ₆H₃)-bian}(ma)] (2) (bian = bis(imino)acenaphthene; ma = maleic anhydride) has been investigated. The rate law under hydrogen-rich conditions is described by r = k[4-octyne]^0.65[Pd]-[H₂], showing first order in palladium and dihydrogen and a broken order in substrate. Parahydrogen studies have shown that a pairwise transfer of hydrogen atoms occurs in the rate-limiting step. In agreement with recent theoretical results, the proposed mechanism consists of the consecutive steps: alkyne coordination, heterolytic dihydrogen activation (hydrogenolysis of one Pd-N bond), subsequent hydro-palladation of the alkyne, followed by addition of N-H to palladium, reductive coupling of vinyl and hydride and, finally, substitution of the product alkene by the alkyne substrate. Under hydrogen-limiting conditions, side reactions occur, that is, formation of catalytically inactive palladacycles by oxidative alkyne coupling. Furthermore, it has been shown that (Z)-oct-4-ene is the primary reaction product, from which the minor product (E)-oct-4-ene is formed by an H₂-assisted, palladium-catalyzed isomerization reaction.

Full text of DOI:10.1021/ja052729j

Published on Web 10/13/2005
Kinetic and Spectroscopic Studies of the
[Palladium(Ar-bian)]-Catalyzed Semi-Hydrogenation of
4-Octyne
Alexander M. Kluwer,†,§ Tehila S. Koblenz,† Thorsten Jonischkeit,‡
Klaus Woelk,‡,§ and Cornelis J. Elsevier*,†
Contribution from the Van’t Hoff Institute for Molecular Sciences, UniVersiteit Van Amsterdam,
Nieuwe Achtergracht 166, 1018 WV Amsterdam, The Netherlands, and Institute of Physical and
Theoretical Chemistry, Wegelerstr. 12, UniVersity of Bonn, Bonn, 53115 Germany
Received May 20, 2005; E-mail: elsevier@science.uva.nl
Abstract: The kinetics of the stereoselective semi-hydrogenation of 4-octyne in THF by the highly active
catalyst [Pd{(m,m′-(CF3)2C6H3)-bian}(ma)] (2) (bian ) bis(imino)acenaphthene; ma ) maleic anhydride)
has been investigated. The rate law under hydrogen-rich conditions is described by r ) k[4-octyne]0.65[Pd]-
[H2], showing first order in palladium and dihydrogen and a broken order in substrate. Parahydrogen studies
have shown that a pairwise transfer of hydrogen atoms occurs in the rate-limiting step. In agreement with
recent theoretical results, the proposed mechanism consists of the consecutive steps: alkyne coordination,
heterolytic dihydrogen activation (hydrogenolysis of one Pd-N bond), subsequent hydro-palladation of the
alkyne, followed by addition of N-H to palladium, reductive coupling of vinyl and hydride and, finally,
substitution of the product alkene by the alkyne substrate. Under hydrogen-limiting conditions, side reactions
occur, that is, formation of catalytically inactive palladacycles by oxidative alkyne coupling. Furthermore, it
has been shown that (Z)-oct-4-ene is the primary reaction product, from which the minor product (E)-oct-
4-ene is formed by an H2-assisted, palladium-catalyzed isomerization reaction.
Introduction
for a wide variety of alkynes yielding the cis-alkenes, However,
the formed cis-alkene is prone to cis-trans isomerization, bond-
Homogeneous hydrogenation by transition metal complexes
has played a key role in the fundamental understanding of
catalytic reactions and has proven to be of great utility in
practical applications.1 Despite the wealth of information
available for the homogeneous hydrogenation of carbon-carbon
double bonds, remarkably few details have been reported for
alkynes, even though a large number of catalysts have been
tested and found active in this reaction.2 Stereoselective semi-
hydrogenation of carbon-carbon triple bonds is a highly desired
tool for synthetic organic chemistry, and many of the products
obtained through this reaction are useful in the synthesis of
natural products, such as biologically active compounds. For
such compounds, the stereo- and regiocontrol of the resulting
double bond is essential for its biological activity. The most
widely applied catalyst for the hydrogenation of alkynes to
afford (Z)-alkenes employs group 8 transition metals of which
the Lindlar catalyst (lead-poisoned Pd on CaCO3) is the most
prominent member.3 This catalyst shows considerable selectivity
shift isomerization, and over-reduction under the reaction
conditions. In addition, the Lindlar-catalyzed hydrogenation
reaction suffers from low chemoselectivity and a low reproduc-
ibility, leading to product mixtures that need to be separated.4
Selective homogeneous hydrogenation of alkynes to the
corresponding alkenes has been observed, for example, with
ruthenium,5 iridium,6 chromium,7,8 rhodium,9 and iron10 cata-
lysts. However, only a few of these show a good selectivity
and a reasonable activity toward a variety of alkynes containing
different functional groups; therefore, the general application
of these catalytic systems in synthetic organic chemistry remains
rather limited. Moreover, most of the reported catalytically active
systems deal with selective hydrogenation of terminal alkynes
(4) Ulan, J. G.; Maier, W. F. J. Org. Chem. 1987, 52, 3132-3142.
(5) (a) Albers, M. O.; Singleton, E.; Viney, M. M. J. Mol. Catal. 1985, 33,
77-82. (b) Albers, M. O.; Singleton, E.; Viney, M. M. J. Mol. Catal. 1985,
30, 213-217. (c) Blum, Y.; Czarkie, D.; Rahamim, Y.; Shvo, Y.
Organometallics 1985, 4, 1459-1461. (d) Frediani, P.; Giannelli, C.;
Salvini, A.; Ianelli, S. J. Organomet. Chem. 2003, 667, 197-208. (e)
Schleyer, D.; Niessen, H. G.; Bargon, J. New J. Chem. 2001, 25, 423-
426. (f) Suarez, T.; Fontal, B. J. Mol. Catal. 1988, 45, 335-344. (g)
Bianchini, C.; Bohanna, C.; Esteruelas, M. A.; Frediani, P.; Meli, A.; Oro,
L. A.; Peruzzini, M. Organometallics 1992, 11, 3837-3844.
(6) (a) Baddley, W. H.; Fraser, M. S. J. Am. Chem. Soc. 1969, 91, 3661-
3663. (b) Navarro, J.; Sagi, M.; Sola, E.; Lahoz, F. J.; Dobrinovitch, I. T.;
Katho, A.; Joo, F.; Oro, L. A. AdV. Synth. Catal. 2003, 345, 280-288. (c)
Sanchez-Delgado, R. A.; Andriollo, A.; Gonzalez, E.; Valencia, N.; Leon,
V.; Espidel, J. J. Chem. Soc., Dalton Trans. 1985, 1859-1863.
(7) Sodeoka, M.; Shibasaki, M. J. Org. Chem. 1985, 50, 1147-1149.
(8) Sodeoka, M.; Shibasaki, M. Synthesis 1993, 7, 643-658.
† Universiteit van Amsterdam.
‡ University of Bonn.
§ Current address: Department of Chemistry, University of Missouri-
Rolla, 1870 Miner Circle, Rolla, MO, USA.
(1) Chaloner, P. A.; Esteruelas, M. A.; Joo, F.; Oro, L. A. In Homogeneous
Hydrogenation; Ugo, R., James, B. R., Eds.; Kluwer Academic Publish-
ers: Dordrecht, The Netherlands, 1993.
(2) Sanchez-Delgado, R. A.; Rosales, M. Coord. Chem. ReV. 2000, 249-280.
(3) Hutchins, R. O.; Hutchins, M. G. K. In Reduction of Triple-bonded Groups,
The Chemistry of Functional Groups; Patai, S., Rappoport, Z., Eds.; John
Wiley & Sons Ltd.: New York, 1983; Vol. 1, pp 571-597.
9
15470
J. AM. CHEM. SOC. 2005, 127, 15470-15480
10.1021/ja052729j CCC: $30.25 © 2005 American Chemical Society

Kinetics of [Pd(Ar-bian)]-Catalyzed Hydrogenation
A R T I C L E S
Scheme 1. Semi-Hydrogenation of Alkynes Catalyzed by [Pd(Ar-bian)(alkene)]
as the model substrates; hence, the stereoselectivity of the
catalysts is not relevant and often not determined.
full conversion of the alkyne has been achieved (selectivity
determined at >99.5% conversion). Under these conditions, both
internal and terminal alkynes are reduced with great ease.
Besides the high stereoselectivity, the chemoselectivity is also
remarkably high as demonstrated by the presence of other
reducible functional groups in the substrate (such as carboxy
groups, nitro groups, or even alkene moieties in conjugated
enynes), which remained unaffected under the hydrogenation
conditions.
To elucidate the reaction mechanism, we report in this paper
the investigation of the overall kinetics of the highly selective
[Pd(Ar-bian)]-catalyzed semi-hydrogenation reaction, as well
as NMR studies involving deuterium labeling and parahydrogen-
induced polarization. Knowledge of the rate law was deemed
essential to assign the rate-determining step in the putative
cascade of elementary reaction steps.15,16 We have specifically
studied the semi-hydrogenation of 4-octyne employing the
zerovalent precatalyst [Pd{(m,m′-(CF3)2C6H3)-bian}(ma)] (2;
bian ) bis(imino)acenaphthene; ma ) maleic anhydride). The
presence of electron-withdrawing groups on the Ar-bian ligand
greatly enhances the rate of alkyne hydrogenation, while
maintaining the high stereo- and chemoselectivity, which have
typically been found for the [Pd(Ar-bian)] systems. A mecha-
nism for the [Pd{(m,m′-(CF3)2C6H3)-bian}(ma)]-catalyzed semi-
hydrogenation of 4-octyne is proposed in this article.
The number of homogeneous palladium systems that have
been reported as active catalysts in the semi-hydrogenation of
alkynes is still very limited. Most of these Pd catalysts are
mononuclear palladium complexes, either Pd(II)11 or Pd(0)
complexes,12 although some Pd clusters have also been found
to be highly active in this type of hydrogenation reaction.13
Considerable attention has been directed toward immobilized
palladium complexes, where the palladium is coordinated to a
ligand attached to a polymer, a clay, or other inorganic
supports.14
In the past, some authors12a have reported a highly stereo-
selective hydrogenation of internal alkynes by zerovalent
palladium catalysts bearing the rigid bidentate nitrogen ligand
bis(N-arylimino)acenaphthene (Ar-bian), and which are able to
homogeneously hydrogenate a wide variety of alkynes to form
the corresponding (Z)-alkenes (Scheme 1). The observed
selectivity toward the (Z)-alkene, using [Pd(4-MeO-C6H4-bian)-
(dmfu)] (1; dmfu ) dimethyl fumarate) as the catalyst, is very
high (>99%)11a,c for various alkynes under very mild conditions
(25 °C, 1 bar H2, see Scheme 1). Typical turnover frequencies
(TOF) of 100-200 mol mol-1 h-1 were obtained with this
catalyst. The high selectivity, that is in many cases superior to
the one obtained with the Lindlar catalyst, is maintained until
Results and Discussion
(9) (a) Blum, J.; Rosenfeld, A.; Polak, N.; Israelson, O.; Schumann, H.; Avnir,
D. J. Mol. Catal. A 1996, 107, 217-223. (b) Breuer, D.; Goen, T.; Haupt,
H. J. J. Mol. Catal. 1990, 61, 149-162. (c) Cabeza, J. A.; Fernandez-
Colinas, J. M.; Llamazares, A.; Riera, V. J. Mol. Catal. 1992, 71, L7-
L11. (d) Candlin, J. P.; Oldham, A. R. Discuss. Faraday Soc. 1968, 46,
60-71. (e) Crabtree, R. H. J. Chem. Soc., Chem. Commun. 1975, 647-
648. (f) Crabtree, R. H.; Gautier, A.; Giordano, G.; Khan, T. J. Organomet.
Chem. 1977, 141, 113-121. (g) Osborn, J. A.; Jardine, F. H.; Young, J.
F.; Wilkinson, G. J. Chem. Soc. A 1966, 1711-1732. (h) Stephan, M.;
Kohlmann, O.; Niessen, H. G.; Eichhorn, A.; Bargon, J. Magn. Res. Chem.
2002, 40, 157-160.
Preliminary Remarks. The synthesis of the precursor
catalyst [Pd{(m,m′-(CF3)2C6H3)-bian}(ma)] (2) has been achieved
by reacting the extremely versatile complex [Pd(nbd)(ma)] (3)17
with the free ligand (m,m′-(CF3)2C6H3)-bian in THF (nbd )
norbornadiene). This route has proven to be very successful for
the synthesis of a wide variety of zerovalent palladium
compounds, including the allegedly “unstable” [Pd(N)2(ma)]
complexes (where N is a simple “σ-donor-only” ligand). After
workup, catalyst (2) is an air-stable compound; however, it was
stored under nitrogen atmosphere to ensure its purity.
(10) Bianchini, C.; Meli, A.; Peruzzini, M.; Frediani, P.; Bohanna, C.; Esteruelas,
M. A.; Oro, L. A. Organometallics 1992, 11, 138-145.
(11) (a) Pelagatti, P.; Venturini, A.; Leporati, A.; Carcelli, M.; Costa, M.; Bacchi,
A.; Pelizzi, G.; Pelizzi, C. J. Chem. Soc., Dalton Trans. 1998, 2715-2721.
(b) Costa, M.; Pelagatti, P.; Pelizzi, C.; Rogolino, D. J. Mol. Catal. A 2002,
178, 21-26.
(12) (a) van Laren, M. W.; Elsevier, C. J. Angew. Chem., Int. Ed. 1999, 38,
3715-3717. (b) Sulman, E.; Deibele, C.; Bargon, J. React. Kinet. Catal.
Lett. 1999, 67, 112-117. (c) 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, 1546-1553.
In a preliminary catalytic study, we found that the [Pd(Ar-
bian)]-catalyzed hydrogenation of alkynes typically shows a
distinct induction period, after which the alkyne consumption
is linear with time. To obtain kinetic data from this reaction,
the induction behavior during the first 10% of the alkyne
conversion is disregarded, and only the alkyne consumption
between 10 and 90% conversion is determined as a function of
(13) (a) Cabeza, J. A. In Metal Clusters in Chemistry; Braunstein, P., Oro, L.
A., Raithby, P. R., Eds.; Wiley-VCH Verlag: Weinheim, Germany, 1999;
Vol. 2, pp 715-740. (b) Evrard, D.; Groison, K.; Mugnier, Y.; Harvey, P.
D. Inorg. Chem. 2004, 43, 790-796. (c) Niessen, H. G.; Eichhorn, A.;
Woelk, K.; Bargon, J. J. Mol. Catal. A 2002, 182-183, 463-470.
(14) (a) Holy, N. L.; Shelton, S. R. Tetrahedron 1981, 37, 25-26. (b) Elman,
B.; Moberg, C. J. Organomet. Chem. 1985, 294, 117-122. (c) Moberg,
C.; Rakos, L. J. Organomet. Chem. 1987, 335, 125-131. (d) Choudary,
B. M.; Sharma, G. V. M.; Bharathi, P. Angew. Chem. 1989, 101, 506-
507. (e) Ferrari, C.; Predieri, G.; Tiripicchio, A.; Costa, M. Chem. Mater.
1992, 4, 243-245. (f) Islam, M.; Bose, A.; Mal, D.; Saha, C. R. J. Chem.
Res. (S) 1998, 44-45. (g) Michalska, Z. M.; Ostaszewski, B.; Zientarska,
J.; Sobczak, J. W. J. Mol. Catal. A 1998, 129, 207-218.
(15) James, B. R. Homogeneous Hydrogenation; John Wiley and Sons: New
York, 1973; pp 198-287.
(16) Wilkinson, F. Chemical Kinetics and Reaction Mechanisms; Van Nostrand
Reinhold: New York, 1980, pp 65-130.
(17) Kluwer, A. M.; Elsevier, C. J.; Bu¨hl, M.; Lutz, M.; Spek, A. L. Angew.
Chem., Int. Ed. 2003, 42, 3501-3504.
9
J. AM. CHEM. SOC. VOL. 127, NO. 44, 2005 15471

A R T I C L E S
Kluwer et al.
Figure 1. (A) Plot of the rate of hydrogenation versus the dihydrogen pressure. (B) Plot of ln(kobs) versus ln(pH2). Conditions as given in Table 1.
Figure 2. (A) Plot of the rate of hydrogenation versus the 4-octyne concentration. (B) Plot of ln(rate) versus ln[4-octyne]. Conditions as given in Table 2.
Table 1. Kinetic Data for the Hydrogenation of 4-Octyne to
time. Furthermore, it appeared that the rate of 4-octyne
hydrogenation employing complex 2 was about 2 orders of
magnitude higher (TOF up to 18 000) than that for [Pd(Ar-bian)]
complexes that contain more electron-donating ligands (e.g.,
p-MeC6H4-bian, p-MeOC6H4-bian). Finally, the use of deuterium
has confirmed that molecular hydrogen (deuterium) is the source
of the added hydrogen atoms in the [Pd(Ar-bian)]-catalyzed
semi-hydrogenation reaction.
Dependence of the Reaction Rate on the Dihydrogen
Pressure. The dihydrogen pressure was varied between 0 and
40 bar while keeping the 4-octyne and the catalyst concentrations
constant at 0.31 M and 0.2 mM, respectively. Up to 15 bar of
dihydrogen pressure, the reaction rate depends linearly on the
dihydrogen pressure (see Figure 1A). Plots of ln(rate) versus ln
p(H2) result in a straight line for dihydrogen pressures between
1 and 15 bar, yielding a slope 1.01, showing that the hydrogena-
tion of 4-octyne is first order in dihydrogen pressure (see Figure
1B). The values of kobs are listed in Table 1.
At dihydrogen pressures higher than 25 bar, the hydrogenation
activity (kobs) remains constant (compare entries 6, 7, and 8 of
Table 1). The apparent zeroth-order dependence on the dihy-
drogen pressure suggests that the reaction rate is limited by the
mass transfer of dihydrogen gas into the THF solution. In such
a case, the dihydrogen gas concentration in solution is no longer
proportional to the applied hydrogen pressure, and consequently,
a zeroth-order dependency is observed. While the observed
4-(Z)-Octene by [Pd{(m,m′-(CF3)2C6H3)-bian}(ma)] (2)a
kcat
3
p(H2)
(bar)
10-4 [Pd]
[4-octyne]
10- kobs
(mol-2 L2
selectivity
1
1
1
1
entry
(mol L-
)
(mol L-
)
(s-
)
bar-1 s-
)
Z/E/alkaneb
1
2
3
4
5
6
7
8
0
1.0
7.0
11.3
15.3
21.5
26.0
40.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
0.31
0.31
0.31
0.31
0.31
0.31
0.31
0.31
0
0.6
4.7
0
18
22
24
23
20
18
12
94/6/0
95/5/0
93/7/0
95/5/0
95/5/0
94/6/0
93/7/0
8.3
11.0
13.4
14.4
14.3
a Conditions: reactions were carried out in a stainless steel autoclave.
Stirring speed was set at 1250 rpm, and reaction temperature was 21 °C.
b Selectivity determined between 10 and 90% alkyne conversion.
reaction rate (kobs) is constant, the kcat progressively decreases
with increasing pressure (see Table 1). The stereoselectivity of
the reaction is not affected by the hydrogen pressure, and the
ratio of (Z)-oct-4-ene/(E)-oct-4-ene/octane typically remains 95/
5/0.
Dependence of the Reaction Rate on the Substrate
Concentration. The substrate concentration was varied between
0.16 and 0.76 M, keeping a constant hydrogen pressure of 7
bar and the catalyst concentration at 0.2 mM. In Figure 2A, the
reaction rate is plotted as a function of substrate, and it clearly
shows that reaction rate reaches a maximum at a substrate
concentration of 0.32 M. At lower concentrations (entries 1-5
9
15472 J. AM. CHEM. SOC. VOL. 127, NO. 44, 2005

Kinetics of [Pd(Ar-bian)]-Catalyzed Hydrogenation
A R T I C L E S
Figure 3. (A) Plot of the rate of hydrogenation versus the catalyst concentration. (B) Plot of ln(rate) versus ln[Pd]. Conditions as given in Table 4.
Table 2. Kinetic Data for the Hydrogenation of 4-Octyne to
4-(Z)-Octene by [Pd{(m,m′-(CF3)2C6H3)-bian}(ma)] (2)a
Scheme 2. Formation of a Palladacycle
kcat
3
p(H2)
(bar)
10-4 [Pd]
[4-octyne]
10- kobs
(mol-2 L2
selectivity
Z/E/alkaneb
1
1
1
1
entry
(mol L-
)
(mol L-
)
(s-
)
bar-1 s-
)
1
2
3
4
5
6
7
8
9
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
0
0
0
41
38
35
33
27
21
11
8.6
0.16
0.21
0.25
0.31
0.37
0.44
0.67
0.76
9.2
11.3
12.1
14.3
13.8
13.0
10.3
9.1
94/6/0
95/5/0
95/5//0
95/5/0
94/6/0
95/5/0
95/5/0
95/5/0
Attempts to isolate this palladacycle failed due to the
instability of this compound, which is in accordance with reports
from other authors who were unable to prepare or isolate a
palladacycle arising from 4-octyne.20,21 The absence of hexa-
(n-propyl)benzene in the reaction mixture formed by the
trimerization of 4-octyne suggests that the palladacycle is
inactive and acts as a sink of catalyst species (catalyst deactiva-
tion product).
a Conditions: reactions were carried out in a stainless steel autoclave.
Stirring speed was set at 1250 rpm, and reaction temperature was 21 °C.
b Selectivity determined between 10 and 90% alkyne conversion by GC.
in Table 2), ln(rate) versus ln([substrate]) gives a straight line
with a slope of 0.65 (see Figure 2B). The positive broken
reaction order suggests that the substrate is involved in a pre-
equilibrium that precedes the rate-determining step.
Dependence of the Reaction Rate on the Catalyst Con-
centration. The catalyst concentration was varied between 0
and 0.25 mM, keeping the dihydrogen pressure constant at 7
bar and 4-octyne concentration at 0.31 M. The reaction rate
shows a linear dependency with respect to the palladium
concentration (see Figure 3A). Plotting ln(rate) versus ln[Pd],
as presented in Figure 3B, yields a straight line of slope 1.08,
showing that the hydrogenation of 4-octyne is first order in
palladium. The values of kobs are listed in Table 3. At concen-
trations higher than 2.0 × 10-4 mol L-1, the activity decreased;
unfortunately, we were not able to determine the order in
palladium at high catalyst concentrations due to the poor
reproducibility of the reaction in this concentration domain. This
suggests that the precatalyst is not completely activated during
the induction period. In addition, at the higher reaction rates,
the required rate of dihydrogen consumption cannot be met,
which again results in the formation of inactive palladacycles.
Derivation of the Experimental Rate Law. The results
presented here lead us to formulate the overall experimental
rate law for the stereoselective hydrogenation of 4-octyne
At concentrations higher than 0.32 M, the order in substrate
changes from +0.65 to -0.50, and 4-octyne becomes an
inhibitor for the hydrogenation reaction. The negative order of
-0.50 suggests that two substrate molecules are involved in
the inhibition (rate dependence of 1/[4-octyne]2). This change
is caused by the rate-limiting dissolution of dihydrogen gas in
combination with the high alkyne concentration. Inspecting
Tables 1 and 2 (compare entry 8 with entry 5, respectively)
shows indeed that the rates of the hydrogenation reaction
(expressed in kobs) are identical. At this rate, the mass transfer
requirements of dihydrogen into the THF solution cannot be
met, and the incipient [Pd(Ar-bian)(alkyne)] complex exists long
enough to react with a second substrate molecule to form a
palladacycle (see Scheme 2). The latter is, in turn, inactive under
these reaction conditions, thereby reducing the amount of active
hydrogenation catalyst and, hence, the overall observed catalytic
activity. It has been described by Mosely et al.18 and by Ito et
al.19 that stronger σ-donating ligands (e.g., N-donor ligands)
favor the formation of the palladacyclopentadiene instead of
the [Pd0(η2-alkyne)] complex.
(18) Moseley, K.; Maitlis, P. M. J. Chem. Soc., Chem. Commun. 1971, 1604-
(20) van Belzen, R.; Klein, R. A.; Kooijman, H.; Veldman, N.; Spek, A. L.;
Elsevier, C. J. Organometallics 1998, 17, 1812-1825.
(21) van Asselt, R.; Elsevier, C. J.; Smeets, W. J. J.; Spek, A. L. Inorg. Chem.
1994, 33, 1521-1531.
1605.
(19) Ito, T.; Hasegawa, S.; Takahashi, Y.; Ishii, Y. J. Chem. Soc., Chem.
Commun. 1972, 629-630.
9
J. AM. CHEM. SOC. VOL. 127, NO. 44, 2005 15473

A R T I C L E S
Kluwer et al.
Table 3. Kinetic Data for the Hydrogenation of 4-Octyne to
(Z)-Oct-4-ene by [Pd{(m,m′-(CF3)2C6H3)-bian}(ma)] (2)a
kcat
3
p(H2)
(bar)
10-4 [Pd]
[4-octyne]
10- kobs
(mol-2 L2
selectivity
Z/E/alkaneb
1
1
1
1
entry
(mol L-
)
(mol L-
)
(s-
)
bar-1 s-
)
1
2
3
4
5
7.0
7.0
7.0
7.0
7.0
0
0.31
0.31
0.31
0.31
0.31
0
6.0
8.8
10.6
12.1
0
28
27
28
28
1.0
1.5
1.75
2.0
95/5/0
95/5/0
94/6/0
95/5/0
a Conditions: reactions were carried out in a stainless steel autoclave.
Stirring speed was set at 1250 rpm, and reaction temperature was 21 °C.
b Selectivity determined between 10 and 90% alkyne conversion by GC.
Table 4. Kinetic Data for the Hydrogenation of 4-Octyne to
4-(Z)-Octene by [Pd{(m,m′-(CF3)2C6H3)-bian}(ma)] (2) at Different
Temperaturesa
3
10- kobs
selectivity
Z/E/alkaneb
1
entry
T (
°
C)
(s-
)
Figure 4. Eyring plot of the stereoselective hydrogenation of 4-octyne by
2.
1
2
3
4
21.0
25.4
29.0
36.5
14.0
15.3
20.5
25.9
95/5/0
95/5/0
95/5/0
94/6/0
5d series typically form stronger bonds with ligands (for example
to carbon) than their 3d and 4d counterparts.1,24 These third-
row transition metal complexes are generally less active in
catalytic reactions and, hence, can serve as stable models of
reactive intermediates proposed for a catalytic transformation.
Due to the fast kinetics of 2 in the semi-hydrogenation reaction
of 4-octyne, intermediates could neither be identified nor
detected by standard or PHIP NMR techniques (vide infra).
Therefore, recourse was taken to the platinum analogue
[Pt{(m,m′-(CF3)2C6H3)-bian}(ma)] (3) that was synthesized by
reacting [Pt(nbe)3] (nbe ) norbornene) with the m,m′-(CF3)2C6H3-
bian ligand and maleic anhydride in THF.25 Compound 3 could
be isolated in 45% yield and was characterized by NMR, mass
spectroscopy, and elemental analysis.
a Conditions: reactions were carried out in a stainless steel autoclave;
[4-octyne] ) 0.31 M, [2] ) 2.0 mM, and H2 pressure ) 7.0 bar. Stirring
speed was set at 1250 rpm. b Selectivity determined between 10 and 90%
alkyne conversion by GC.
catalyzed by 2 in THF (under conditions where dihydrogen is
nonlimiting) as follows (see Supporting Information, S1):
-d[4-octyne]
) k[4-octyne]0.65[Pd]pH2
(1)
dt
Equation 1 shows the first order in palladium catalyst and
dihydrogen pressure and the broken order of 0.65 for 4-octyne.
However, under hydrogen-limiting reaction conditions (e.g., high
substrate or catalyst concentrations), the kinetics and, hence,
the rate law change and eq 1 is no longer applicable. The
selectivity observed for the hydrogenation reaction is typically
95/5/0 ((Z)-oct-4-ene/(E)-oct-4-ene/octane) and appears to be
independent of the reaction conditions (see Tables 1-3). The
constant stereoselectivity of the 4-octyne hydrogenation under
all conditions (both dihydrogen-rich and dihydrogen-limiting
reaction conditions) suggests that a single mechanism is
operative.
Temperature Dependence of the Reaction Rate and
Activation Parameters. The effect of temperature on the
hydrogenation rate was studied in the range 20-36 °C, at a
4-octyne concentration of 0.31 M, a catalyst concentration of
2.0 mM, and a dihydrogen pressure of 7.0 bar. The Eyring plot
constructed from the obtained kinetic data is shown in Figure
4 yielding the following activation parameters; ∆Hq ) 30 ( 2
kJ mol-1; ∆Sq ) -180 ( 14 J K-1. Similar values have been
reported for ruthenium- and nickel-catalyzed hydrogenation of
alkenes.22,23 The large negative value for the entropy of
activation (∆Sq) indicates a highly ordered transition state
commensurate with the interaction of a dihydrogen molecule
with the palladium center.
Hydrogenation reactions of 4-octyne using [Pt{(m,m′-
(CF3)2C6H3)-bian}(ma)] (3) as the catalyst (ratio 4-octyne/3 )
100) in acetone-d6 at 21 °C were followed by 1H NMR. In the
1H NMR spectrum, only broad aromatic signals were detected,
and no distinct platinum complexes could be identified. When
the reaction was performed under dihydrogen-limiting conditions
(2 mmol of dihydrogen, 0.013 mmol catalyst, and a ratio
substrate/catalyst of 300 at 60 °C in THF), the color of the
solution changed from dark purple to dark red, indicating that
a new platinum species was formed. This new air- and moisture-
stable platinum complex could be isolated and characterized
by NMR, IR, and mass spectroscopy. From these spectroscopic
data, it was concluded that the obtained red solid was
[{(m,m′-(CF3)2C6H3)-bian}platina-2,3,4,5-tetra(propyl)cyclo-
pentadiene] (4), which had been formed by an oxidative
coupling (see Scheme 3). Such platinacyclopentadienes have
been reported in the literature as an intermediate in the
cyclotrimerization, isomerization, and polymerization of alkynes
and have so far only been isolated for alkynes bearing more
electron-withdrawing groups (e.g., alkoxycarbonyl or phenyl
moieties).18,26 The ease of formation of the [{(m,m′-(CF3)2C6H3)-
bian}platina-2,3,4,5-tetra(propyl)cyclopentadiene] (4) under di-
hydrogen-limiting conditions suggests that, during the hydro-
Reactions Involving the Platinum Model Catalyst [Pt-
{(m,m′-(CF3)2C6H3)-bian}(ma)] (3). Metal complexes of the
(24) Tolman, C. A.; Faller, J. W. In Homogeneous Catalysis with Metal
Phosphine Complexes; Pignolet, L. H., Ed.; Plenum Press: New York, 1983;
pp 13-109.
(25) Sprengers, J. W.; De Greef, M.; Duin, M. A.; Elsevier, C. J. Eur. J. Inorg.
Chem. 2003, 3811-3819.
(22) Huh, S.; Cho, Y.; Jun, M.-J. Polyhedron 1994, 13, 1887-1894.
(23) Angulo, I. M.; Lok, S. M.; Quiroga Norambuena, V. F.; Lutz, M.; Spek,
A. L.; Bouwman, E. J. Mol. Catal. A 2002, 187, 55-67.
9
15474 J. AM. CHEM. SOC. VOL. 127, NO. 44, 2005

Kinetics of [Pd(Ar-bian)]-Catalyzed Hydrogenation
A R T I C L E S
hydrogenation reaction proceeds by an overall pairwise addition
of the two hydrogen atoms from H2 to the substrate and may
involve a palladium(alkyne)-hydride species (such as B in
Scheme 4). Since the polarized signals are only observed in
the hydrogenation product, the exact nature of these palladium-
hydride species is not revealed within the PHIP NMR experi-
ments. Therefore, the proposed palladium-hydride species can
then arise by either homolytic (i.e., oxidative addition) or
heterolytic activation. (vide infra). The latter pathway has
recently been established by theoretical calculations of a
palladium-catalyzed semi-hydrogenation reaction as a feasible
route for cis-product formation, whereas the oxidative addition
pathway yielded only less likely, high-energy pathways.28a
It should be emphasized, that polarization signals were not
detected in the (E)-alkene, which was produced in minor
amounts (5%) during the hydrogenation. This is clear evidence
for the fact that the (E)-alkene does not arise from direct
hydrogenation of the alkyne (vide infra).
Figure 5. PHIP signals in the hydrogenation of 1-hexyne to 1-hexene with
complex (2) (d1 ) 10 s). Substrate, 1-hexyne (100 mL); solvent, acetone-
d6 (1200 mL), amount catalyst ) 2 mg, temperature ) 29 °C.
Scheme 3. Formation of the Platinacyclopentadiene (5) by
Oxidative Coupling of Two Alkyne Moieties
Attempts to detect palladium-hydride species failed due to
the short lifetime of the reactive intermediates, and hence, polar-
ization signals were only observed of the hydrogenated product
(i.e., the (Z)-alkene). Even when the substrate was omitted to
avoid depletion of palladium-hydride species by a fast hydro-
genation reaction, no polarization signals were detected, and
the catalyst rapidly decomposed to palladium metal.
In addition, attempts were made to detect signals produced
by the hydrogenation of the proposed palladacycle (E, Scheme
4) that is formed under dihydrogen-limiting conditions. To
ensure that the palladacycle was formed during the hydrogena-
tion reactions, an asymmetric substrate, the 3-phenyl propynoic
acid ethyl ester, was chosen because it is known to easily form
palladacycles.29 Furthermore, the time between parahydrogen
bursts was extended (up to 30 min) to allow the reaction to
occur between the palladium catalyst and the substrate. Although
cis-hydrogenation of the alkyne to (Z)-alkene was still observed,
neither polarization signals nor normal signals were detected
arising from the palladacycle.
genation experiments using 2, a similar palladacycle may be
formed, albeit probably in low concentrations.
PHIP NMR Studies. To ensure that the symmetry breaks
down during the catalytic hydrogenation, the PHIP experiments
presented in this paper have been carried out using the
asymmetric substrates 1-hexyne and phenyl acetylene.27 The
reactions were conducted at 29 °C in THF-d8 using [Pd{(m,m′-
(CF3)2C6H3)-bian}(ma)] (2) as catalyst. Under these conditions,
polarization signals were observed in the product (Z)-alkene
upon hydrogenating 1-hexyne as well as phenyl acetylene with
dihydrogen samples that were enriched in parahydrogen (see
Figure 5). This demonstrates that the palladium-catalyzed
It must to be noted that the observed polarization signals for
hydrogenation products catalyzed by complex 2 in nonprotic
Scheme 4. Catalytic Cycle of the Stereoselective Semi-Hydrogenation of 4-Octyne
9
J. AM. CHEM. SOC. VOL. 127, NO. 44, 2005 15475

A R T I C L E S
Kluwer et al.
solvents were only 1 order of magnitude higher than the thermal
signals produced in a normal NMR. While the enhancement of
the 1H NMR sensitivity by PHIP can theoretically reach values
of 104-105, it typically ranges about a factor of 103-104.30 The
small enhancement of the polarization in this case is explained
by the type of dihydrogen activation, that is, the heterolytic
dihydrogen activation across one of the Pd-N bonds. This
would lead to a large chemical-shift difference in the proposed
zwitterionic intermediate (B; Scheme 4), and hence, the spin
correlation is rapidly lost. In addition, the large nuclear
quadrupole moment of palladium (0.660 × 10-24 cm2) can
further decrease the signal intensity of the polarization signals.
Hence, the proposed mechanism presented in Scheme 4 is in
agreement with the findings of the parahydrogen hydrogenation
experiments as well as the results of the DFT calculations.28a
That is, the heterolytic dihydrogen cleavage coincides with the
rate-limiting step (i.e., the step with the highest activation
energy), and no stable intermediates are formed after the initial
dihydrogen activation. To readily observe the PHIP signal at
all, the reaction must be fast in an absolute sense with respect
to relaxation and loss of coherence.
PHIP NMR and catalyst 2 in the hydrogenation reaction of
1-hexyne and phenyl acetylene.
Reductive elimination of intermediate C yields the Pd(NN)-
(alkene) species D. Again, the hydrogen atoms are transferred
in such a way that the spin information in PHIP experiments is
maintained in the primary reaction product (i.e., the (Z)-alkene).
In principle, this can be achieved either by transferring the two
hydrogen atoms in a concerted pairwise manner or by transfer-
ring them very rapidly (compared to the nuclear spin coupling
constant and the spin relaxation) to the substrate in an
asynchronous way. As mentioned before, this latter route is in
accordance with the heterolytic dihydrogen splitting across one
of the Pd-N bonds.
The broken reaction order in substrate (0.65) is explained by
a pre-equilibrium between species D and A (vide infra).28b In
the case that the dihydrogen concentration becomes limiting,
or dihydrogen is no longer present during the reaction, the
palladium(alkyne) complex A reacts with a second alkyne
molecule to form the palladacycle E, which is an inactive
palladium species.
On the basis of this hydrogenation mechanism (i.e., dihy-
drogen activation is the rate-determining step), the following
rate expression can be derived (see Supporting Information): r
) kcat[octyne][Pd]catpH2 with kcat ) k2k3K1/(k3{[octene] +
K1[octyne]}. This expression is in agreement with the experi-
mentally derived rate law (eq 1).
Activation of the Precatalyst. The hydrogenation experi-
ments of 4-octyne to (Z)-oct-4-ene were carried out using the
preformed catalyst [Pd{(m,m′-(CF3)2C6H3)-bian}(ma)] (2). In
such a case, where the coordinated alkene of the precatalyst is
not identical to the substrate, the route into the catalytic cycle
needs to be considered, especially since it can influence the
performance of the catalyst.29
Mechanism for the Stereoselective Hydrogenation of
4-Octyne by [Pd{(m,m′-(CF3)2C6H3)-bian}(ma)] (2). The
kinetic and spectroscopic analysis presented above enables us
to propose a plausible catalytic cycle for palladium-catalyzed
semi-hydrogenation of 4-octyne by complex 2. The proposed
catalytic cycle, as depicted in Scheme 4, roughly resembles the
catalytic cycle proposed by van Laren et al.12a To enter the cycle,
the coordinated maleic anhydride of the precursor catalyst is
substituted (vide infra) to form the zerovalent Pd(NN)(alkyne)
species (A), which is an active intermediate entering the catalytic
cycle. It is assumed that the concentration of precursor catalyst
2 is negligible after the induction period during which species
A is formed. This is confirmed by the zero intercept of the rate-
versus-[Pd] plot and the first-order dependency of the hydro-
genation rate on the palladium concentration (see Figure 3). At
palladium concentrations below 0.2 mM, the amount of pal-
ladium in the catalytic cycle is thus equal to the total amount
of palladium in the reaction mixture.
To enter the catalytic cycle, two possibilities that determine
the faith of the coordinated maleic anhydride are envisaged:
(1) the maleic anhydride is hydrogenated to succinic anhydride,
or (2) the maleic anhydride is substituted by the large excess
of alkyne present, and free maleic anhydride is present is
solution. To investigate the latter possibility, 11 mg of [Pd{(m,m′-
(CF3)2C6H3)-bian}(ma)] (2) was dissolved in 1 mL of acetone-
d6, and 20 µL of 4-octyne was added (Pd/alkyne ratio ) 1:13).
The reaction was followed with 1H NMR for 90 min at 21 and
55 °C. Within this period of time, neither free maleic anhydride
nor any shift of the maleic anhydride protons’ resonances was
observed upon the addition of 4-octyne. This implies that there
is no significant exchange reaction between the coordinated
maleic anhydride and the dissolved 4-octyne, confirming that
electron-poor alkenes (larger π-accepting properties) coordinate
more strongly to the zerovalent palladium complexes.21
If the maleic anhydride is hydrogenated off during the hydro-
genation experiment, the reaction mixture should contain suc-
cinic anhydride. Indeed, the formation of 3,4-dideutero succinic
Under dihydrogen-rich conditions, the formation of A is
ensued by the heterolytic dihydrogen activation to obtain the
{[Pd(alkyne)(H)]-(N∩NH)+} species (B). According to high-
level DFT calculations, using an extended basis set, these species
easily collapse to [Pd(NN)(H)(alkenyl)] (C) via low-barrier
processes.28a As mentioned in the previous paragraph, attempts
to detect the palladium-hydride species (B or C) failed using
(26) (a) Mu¨ller, C.; Lachicotte, R. J.; Jones, W. D. Organometallics 2002, 21,
1118-1123. (b) Stephan, C.; Munz, C.; tom Dieck, H. J. Organomet. Chem.
1994, 468, 273-278. (c) tom Dieck, H.; Munz, C.; Mueller, C. J.
Organomet. Chem. 1990, 384, 243-255. (d) Jones, N. L.; Ibers, J. A. In
Homogeneous Catalysis with Metal Phosphine Complexes; Pignolet, L. H.,
Ed.; Plenum Press: New York, 1983; pp 111-135. (e) Moseley, K.; Maitlis,
P. M. J. Chem. Soc., Dalton Trans. 1974, 169-175.
(27) Haake, M.; Natterer, J.; Bargon, J. J. Am. Chem. Soc. 1996, 118, 8688-
8691.
2
anhydride (2.48 ppm) was detected in H NMR spectra that
(28) (a) Dedieu, A.; Humbel, S.; Elsevier, C. J.; Grauffel, C. Theor. Chem. Acc.
2004, 112, 305-312. (b) One of the reviewers suggested a mechanism in
which the alkene eliminates from a hydridopalladium(II)-vinyl complex,
which may, in the presence of excess 4-octyne, form a hydridopalladium-
(II)-vinyl(alkyne) complex reversibly. Hence, the formation of the
hydridopalladium(II) complex would clearly depend on the alkyne con-
centration, but the second step would act to inhibit the reaction.
(29) van Belzen, R.; Hoffmann, H.; Elsevier, C. J. Angew. Chem., Int. Ed. Engl.
1997, 36, 1743-1745.
were recorded when deuterium was used in the hydrogenation
reaction. In a separate hydrogenation experiment (using 1 bar
dihydrogen gas), the buildup of succinic anhydride was monitor-
ed by GC analysis, and the results are presented in Figure 6A.
The amount of succinic anhydride increases during the first
4 min, after which its concentration remains constant. Thus,
the slow formation of the active palladium catalyst reflects the
(30) Jang, M.; Duckett, S. B.; Eisenberg, R. Organometallics 1996, 15, 2863-
2865.
9
15476 J. AM. CHEM. SOC. VOL. 127, NO. 44, 2005

Kinetics of [Pd(Ar-bian)]-Catalyzed Hydrogenation
A R T I C L E S
Figure 6. (A) Formation of succinic anhydride during the hydrogenation of 4-octyne. (B) Reaction profile of the hydrogenation of 4-octyne by complex 1.
Scheme 5. Entering the Catalytic Hydrogenation Cycle
sluggish hydrogenation of maleic anhydride by [Pd(Ar-bian)-
(alkene)] systems, which has also been reported by van Asselt
et al.32 The reaction profile for this hydrogenation reaction (see
Figure 6B) displays a distinctive induction period (first 4 min)
and coincides with buildup of succinic anhydride, showing that
the coordinated maleic anhydride is gradually hydrogenated
during this period, with progressive hydrogenation of 4-octyne
by the palladium catalyst (see Scheme 5).
Stereoselectivity of the Hydrogenation and (Z)-(E) Isomer-
ization. Variation of the reaction conditions (dihydrogen pres-
sure, substrate or catalyst concentrations) does not affect the
stereoselectivity of the hydrogenation reaction (see Tables 1-3),
and a product distribution of 95/5/0 ((Z)-oct-4-ene/(E)-oct-4-
ene/octane) is typically obtained. The high stereoselectivity,
seemingly independent of the reaction conditions or conversion,
merits further investigation. The formation of the (E)-isomer
was studied using deuterium gas with both 1-octyne and phenyl
acetylene as model substrate.
2
When complex 2 is employed as the catalyst in the hydro-
genation of 1-octyne and phenyl acetylene using deuterium, there
are, besides the expected deuterium signals from the cis-
hydrogenation, also deuterium signals arising from the D3 in
the vicinal trans position relative to D1 (see Figure 7). The
deuteration of the D3 position is the result of an isomerization
reaction (vide infra). A similar deuteration pattern is obtained
when a mixture of 1-octene and 4-octyne (ratio 1:1) is
hydrogenated using deuterium (see Figure 8). In this case, the
two geminal deuterium signals of 1-octene, with roughly the
same intensity around 5.0 ppm, show that deuterium is
incorporated in both cis- and trans-vicinal positions relative to
D1 without apparent preference (compare Figure 7B and Figure
8). This shows that 1-octene can easily compete with 4-octyne
Figure 7. The 76 MHz H NMR spectra from (a) deuteration of phenyl
acetylene, and (b) deuteration of 1-octyne. Signal D3 in spectrum proves
the incorporation of the deuterium in the geminal position.
for the palladium catalyst and, while being coordinated to the
palladium, be involved in an isomerization reaction.
An isomerization mechanism that operates under dihydrogen
atmosphere has been independently proposed by Bargon et al.
and Spencer et al. and has recently been confirmed by Duckett
et al.33 This mechanism proceeds via an equilibrium between a
palladium(dihydrido)(substrate) complex and the monohydrido
alkyl complex (see Scheme 6).
To gain more insight into the formation of (E)-oct-4-ene, the
products of the hydrogenation reaction (using D2) of the mixture
(33) (a) Dunne, J. P.; Aiken, S.; Duckett, S. B.; Konya, D.; Almeida Len˜ero,
K. Q.; Drent, E. J. Am. Chem. Soc. 2004, 126, 16708-16709. (b) Yu, J.;
Spencer, J. J. Am. Chem. Soc. 1997, 119, 5257-5258. (c) Harthun, A.;
Giernoth, R.; Elsevier, C. J.; Bargon, J. Chem. Commun. 1996, 21, 2483-
2484.
(31) Drexler, H. J.; Baumann, W.; Spannenberg, A.; Fischer, C.; Heller, D. J.
Organomet. Chem. 2001, 621, 89-102.
(32) van Asselt, R.; Elsevier, C. J. J. Mol. Catal. A 1991, 65, L13-L19.
9
J. AM. CHEM. SOC. VOL. 127, NO. 44, 2005 15477

A R T I C L E S
Kluwer et al.
Figure 9. Plot of m/z ) 113/112 versus alkyne conversion. Filled squares
are the hydrogenation of 4-octyne with H2. Open dots: hydrogenation of a
mixture of 4-octyne and (Z)-oct-4-ene. Filled dots: the hydrogenation of
4-octyne with H2 in the presence of (Et3ND)+(CF3SO3)-.
hydrogen atoms after dihydrogen cleavage. In conclusion, the
primary product of the hydrogenation reaction is the (Z)-alkene,
which may be isomerized to the (E)-alkene by the palladium
catalyst assisted by dihydrogen gas according to a mechanism,
as depicted in Scheme 6. These results are in agreement with
the PHIP NMR results (vide supra).
Figure 8. The 76 MHz 2H NMR spectra from (a) deuteration of a mixture
of 4-octyne and 1-octene, and (b) deuteration of 4-octyne. Natural abundance
of deuterium in the solvent THF is marked with asterisks.
The absence of n-octane in the reaction mixture suggests that
the hydrogen-assisted isomerization reaction does not result in
over-reduction, although a palladium(hydrido)(alkyl) species is
assumed during the isomerization process. Apparently, the
â-elimination, which is, in the case of palladium, known to be
a very low-barrier process,34 is much faster than the reductive
elimination of the alkane from the incipient Pd(H)(alkyl) species.
of 4-octyne (m/z ) 110) and 4,5-(H)2 (Z)-oct-4-ene (m/z ) 112)
(ratio 5:1) was studied by GCMS. Analysis of the (E)-oct-4-
ene peak shows an increasing amount of 4,5-(H,D) (E)-oct-4-
ene (m/z ) 113). However, appropriate control experiments were
required for the validation of this interpretation since the mass
113 is “contaminated” by two contributions, namely, (1) the
isotope peak [m + 1]+ from 4,5-(H)2 (E)-oct-4-ene (with m/z
113, 9%) and (2) fragmentation [m - 1]+ of the 4,5-(D)2 (E)-
oct-4-ene (m/z ) 114). After correction of the latter contribution
(8.8% based on m/z 114), the amount of deuterium incorporation
can be expressed as a ratio of m/z 113/112 and is given in Figure
9. From this, it is concluded that a significant amount of
monodeuterated (E)-octene is formed. This compound can only
be formed by isomerization of the 5-(H)2 (Z)-oct-4-ene (m/z )
112) present at the start of the reaction and mediated by the
mechanism presented in Scheme 6.
Addition of a deuteron donor (Et3ND)+(CF3SO3)- to the
reaction mixture (using 4-octyne and H2) shows no incorporation
of deuterium, and accordingly, no 4-hydro-5-deutero (E)-oct-
4-ene (m/z ) 113) is detected in the reaction mixture. This
confirms that the reaction proceeds by incorporation of both
hydrogen atoms from one molecule of dihydrogen via a
palladium species that incorporates the substrate and these two
Conclusions
In summary, the experimentally derived rate law for the
highly stereoselective [Pd(Ar-bian)]-catalyzed hydrogenation of
4-octyne is in agreement with the proposed mechanism. The
first-order dependence in palladium concentration shows that
only mononuclear species are involved in the product formation.
The inverse dependence of the hydrogenation activity on the
substrate concentration provides evidence that an inactive
palladacycle is formed under dihydrogen-limiting conditions.
The high selectivity observed in the hydrogenation experiments
using [Pd{(m,m′-(CF3)2C6H3)-bian}(ma)] complexes is ex-
plained by the relatively strong coordination of the alkyne to
the palladium center, which only allows for the presence of small
amounts of alkene complexes. Only the latter are responsible
for the observed minor amounts of (E)-alkene. Since no n-octane
Scheme 6. Proposed Isomerization Mechanism
9
15478 J. AM. CHEM. SOC. VOL. 127, NO. 44, 2005

Kinetics of [Pd(Ar-bian)]-Catalyzed Hydrogenation
A R T I C L E S
and the Fast Atom Bombardment (FAB) mass spectrometry measure-
ments were carried out using a JEOL JMS SX/SX 102A four-sector
mass spectrometer, coupled to a JEOL MS-MP9021D/UPD system
program. In the EI-MS measurements, the samples were introduced
into the ion source via a direct insertion probe. In the FAB-MS
measurements, the samples were loaded in a matrix solution (3-
nitrobenzyl alcohol) on to a stainless steel probe and bombarded with
xenon atoms with an energy of 3 keV. During the high-resolution EI-
MS and the FAB-MS measurements, a resolving power of 10 000 (10%
valley definition) was used. Low-resolution mass spectrometry (MS)
measurements were performed in the EI mode (70 eV) on a GCMS
HP 5890/5971 apparatus equipped with a ZB-5 column (5% cross-
linked phenyl polysiloxane) with an internal diameter of 0.25 mm and
film thickness of 0.25 µm.
was detected in the reaction mixture, only a tiny amount of the
intermediate Pd(hydrido)(alkyl) species will reductively elimi-
nate to form the alkane. Thus, the combination of the stronger
coordination of the alkyne and the subsequent reaction with
dihydrogen results in the kinetic product, the (Z)-alkene.
The use of deuterium and parahydrogen gas has revealed that
both hydrogen atoms of dihydrogen are transferred to the organic
substrate, in an overall cis-pairwise manner.
Activation of the catalyst precursor 2 occurs by hydrogenation
of the coordinated maleic anhydride to succinic anhydride, which
accounts for the induction time and prevents the formation of
palladacycles before the hydrogenation of the actual substrate
starts. Substrates with more electron-withdrawing substituents,
although they can be hydrogenated cleanly to the (Z)-isomer,
readily produce palladacycles, which hamper the study of the
reaction kinetics for such substrates. Finally, the overall results
have enabled us to provide a viable reaction mechanism for
the stereoselective hydrogenation of 4-octyne in THF by
complex 2. This new insight into the mechanism of the
stereoselective semi-hydrogenation of alkynes by the complex
2 might prove to be valuable for the synthesis of organic
compounds, especially those for which a high degree of (Z)-
isomer needs to be obtained.
Hydrogenation Reactions. The kinetic studies of the catalytic semi-
hydrogenation of 4-octyne in THF with 2 as catalyst precursor have
been performed by using the initial rates method. Under the reaction
conditions used, the first 10% conversion of 4-octyne was discarded,
during which the catalyst’s activity increased (induction period). The
rate of the reaction was measured between 10 and 90% alkyne
conversion and was fitted to a straight line, yielding R2 values higher
than 0.99. For obtaining the reaction orders of the hydrogenation
reaction, the dihydrogen pressure (between 0 and 40 bar), catalyst (0
and 4 mmol L-1), and substrate concentrations (between 0.76 mol L-1
were separately changed.
)
More specifically, for a given amount of dihydrogen pressure,
catalyst, and substrate concentration (as given in Tables 1-3), a typical
hydrogenation experiment can be described as follows: the desired
amount of 4-octyne (0.5-4.5 mL) was taken from the bottle using a 1
mL (or 5 mL) plastic syringe and was then transferred to the high-
pressure autoclave. The exact amount was determined by weighing the
full syringe (syringe + 4-octyne) and weighing it again after it has
been emptied into the autoclave. The desired amount of catalyst (0-
20 µmol) was weighed in a 20 mL vial. This amount was then dissolved
in 0.5 mL of THF, which was then transferred, using a syringe, to the
autoclave. The vial was then washed three times with ca. 0.5 mL of
THF to ensure that all of the catalyst was transferred to the autoclave.
Then, 13 mL of THF was added to the autoclave (total amount of
solvent is 15 mL), after which the autoclave was closed. The electronic
stirrer was started (stirring rate was 1250 rpm), and the experiment
was started by applying the desired amount of dihydrogen pressure
(0-40 bar). During the hydrogenation reaction, small aliquots (∼1 mL)
were taken at regular intervals (0.5-2 min) under constant dihydrogen
pressure. The samples were analyzed by GC to determine the alkyne
conversion and the selectivity toward the reaction products (i.e., (Z)-
oct-4-ene, (E)-oct-4-ene, and octane). After every hydrogenation run,
the autoclave was subjected to thorough cleaning and blanc reaction
to ensure the autoclave was not active due to the presence of undesired
metallic palladium.
Experimental Section
General. Gas chromatographic analyses were run on a Varian 3300
apparatus with a DB-5 column. Dihydrogen (grade 5.0) was purchased
from Air Liquide. High-pressure hydrogenation reactions were per-
formed in a (SS 316) 50 mL stainless steel autoclave equipped with a
directly fixed transmission infrared cell of IRTRAN windows (ZnS,
transparent up to 700 cm-1, Ø 10 mm, optical path length ) 0.4 mm).
The infrared option could not be used due to the low IR sensitivity of
4-octyne. The autoclave was further equipped with a magnetic
mechanical stirrer near the bottom of the autoclave that provides a
vigorous mixing of the solution with the gas at high speeds (max )
3400 rpm). It also provides for a fast flow through the infrared cell
within several milliseconds. An electrical heater (Tmax ) 200 °C), a
temperature controller, and a pressure transducing device (Pmax ) 185
bar) were also attached to the autoclave.
All syntheses of complexes and other air- and/or water-sensitive
compounds were carried out and in dried glasswork, in dry solvents,
using standard Schlenk techniques under an atmosphere of purified
nitrogen. Diethyl ether, THF, pentane, and hexane were distilled from
sodium metal; acetone was distilled from CaSO4, and 1,4-dioxane and
dichloromethane were distilled from CaH2. Other chemicals were used
as received. The starting materials [Pt(nbe)3] and [Pd(nbd)(ma)] and
the ligand (m,m′-(CF3)2 C6H3)-bian were prepared according to literature
procedures.17,25,35
The variable temperature experiments were executed by first charging
the autoclave with 4-octyne and 13 mL of THF. The autoclave was
closed and allowed, with stirring (1250 rpm), to reach the reaction
temperature (preset temperature between 20 and 36 ( 0.2 °C). The
catalyst (5.7 µmol) was dissolved in 2 mL of THF and injected with
H2 gas (7 bar) into the autoclave. Samples were taken every 30 s, which
were analyzed by GC.
1
The H and 13C NMR spectra were recorded on a Varian Mercury
300 spectrometer (1H, 300.13 MHz; 13C, 75.48 MHz; 19F, 282.41 MHz;
2H, 76 MHz) or on a Varian Unity Inova 500 spectrometer (1H, 499.86
MHz; 13C, 125.70 MHz). Positive chemical shifts (ppm) are denoted
1
in the H and 13C NMR spectra to higher frequency from an external
tetramethylsilane reference and in the 19F NMR spectra to higher
After every experiment, the autoclave was cleaned by rinsing it with
acetone. The autoclave was then taken apart, and all of the components
that were in contact with the reaction mixture were carefully cleaned
with aqua regia. All components were washed successively with H2O
and twice with acetone. All metallic components were then cleaned
with silver polish to remove metal chloride and metal nitrate salts. The
autoclave components were washed with acetone and dried. To ensure
that all palladium metal was removed, a blanc experiment was
performed (15 mL of THF, 0.5 mL of 4-octyne, 30 bar H2 omitting
the catalyst). After a reaction time of 1 h (20 times the normal reaction
2
frequency from an external CFCl3 reference. H NMR spectra were
recorded unlocked, and the chemical shifts are denoted with respect to
an internal acetone-d6 reference.
High-resolution mass spectrometry (HRMS) measurements were
carried out at Swammerdam Institute of Life Sciences (SILS),
University of Amsterdam, The Netherlands. The Electron Impact (EI)
(34) Koga, N.; Obara, S.; Kitaura, K.; Morokuma, K. J. Am. Chem. Soc. 1985,
107, 7109-7116.
(35) Crascall, L. E.; Spencer, J. L. Inorg. Synth. 1990, 28, 126-132.
9
J. AM. CHEM. SOC. VOL. 127, NO. 44, 2005 15479

A R T I C L E S
Kluwer et al.
[Pt{(m,m′-(CF3)2C6H3)-bian}(ma)] (3). Pt(nbe)3 (70.6 mg, 0.15
mmol), (m,m′-(CF3)2C6H3)-bian (92.8 mg, 0.15 mmol), and maleic
anhydride (15.0 mg, 0.15 mmol) were dissolved in dry THF. The
reaction mixture was allowed to react for 5 days at 21 °C, during which
the color changed from yellowish brown to purple. The solvent was
evaporated, and the residue was dissolved in 2 mL of dichloromethane.
After addition of 20 mL of pentane, a purple solid deposited from
solution. The solvent was removed by decanting. The products were
washed with pentane (3 × 20 mL) and dried in vacuo. The yield was
Figure 10. Numbering scheme for Ar-bian ligands.
time), the reaction mixture was analyzed by GC. When the conversion
was less then 2%, the autoclave was considered clean.
1
55.6 mg (45%). H NMR (300.13 MHz, acetone-d6): δ 8.51 (d, 2H,
J(H,H) ) 8.7 Hz, Ar-H5), 8.50 (br, 4H, Ar-H9,13), 8.24 (br, 2H, Ar-
H11), 7.71 (d, 2H, J(H,H) ) 7.5 Hz, Ar-H4), 7.70 (pst, 2H, Ar-H3),
3.70 (br, 2H, J(Pt,H) ) 42 Hz, CdC-H). No 13C was recorded due to
the limited solubility and stability of the complex. 19F NMR (282.41
MHz, acetone-d6): δ -63.47. Anal. Calcd for: C, 42.82; H, 1.57; N,
3.12. Found: C, 42.67; H, 1.63; N, 3.04.
Hydrogenation Reactions using Deuterium. For hydrogenation
reactions using deuterium gas, a 5 mm NMR tube was filled with a 1
mL THF solution containing 1 mg of complex 2 and 30 µL of substrate.
The NMR tube was gently purged with 1.1 bar of deuterium gas for
30 s, after which a NMR spectrum was taken.
In Situ Hydrogenation Using Parahydrogen. Para-enriched dihy-
drogen containing about 50% para-H2 was prepared by passing H2
through an activated charcoal reactor at 77 K at a pressure of 3 bar, as
described in the literature.34 The hydrogenations with parahydrogen
were carried out in situ. For this purpose, a glass capillary connected
to the parahydrogen source was lowered into the NMR probe actuated
by an electromagnet (i.e., by a solenoid), which is electrically controlled
by the computer console of the NMR spectrometer. In this fashion,
parahydrogen can be bubbled through the reaction mixture for a defined
period of time followed by the detection pulse of the NMR experiment.
In most cases, a hydrogenation time of 5-10 s has turned out to be
sufficient. About 2 s after the dihydrogen addition has stopped, the
NMR detection was started. This procedure can be repeated to afford
a subsequent accumulation of several scans, which yielded a good
signal-to-noise ratio of the PHIP resonances. In addition, phase-selective
accumulation of data can be used to suppress signals of components
that are not involved in the hydrogenation or to separate PHIP signals
from standard resonances by appropriately selecting only the two-spin
order of PHIP.
[(m,m′-(CF3)2C6H3)-bian]platina-2,3,4,5-tetra(propyl)cyclopenta-
diene (4). A sealed tube equipped with a magnetic stirring bar was
charged with 0.5 mL of octyne (4.0 mmol) and 2 mL of THF. To this
solution was added Pt[(m,m′-(CF3)2C6H3)-bian](ma) (4) (12.2 mg, 0.013
mmol), and the solution was purged with dihydrogen gas for 3 min.
The tube was quickly sealed and heated to 60 °C for 60 min, during
which the purple solution gradually turned dark red. After the reaction
mixture was cooled to room temperature, the solvent was removed by
purging with a gentle stream of nitrogen. After the volume was reduced
to 0.5 mL, 50 mL of pentane was added, and the red complex
precipitated from the solution. The solvent was removed with a cannula,
and the residue was dried in vacuo. The yield was 11.5 mg (85%). 1H
NMR (300.13 MHz, CD2Cl2): δ 8.47 (d, 2H, J(H,H) ) 8.1 Hz, Ar-
H5), 8.19 (s, 2H, Ar-H11), 8.08 (s, 4H, Ar-H9,13), 7.59 (pst, 2H, Ar-H4),
7.07 (d, 2H, J(H,H) ) 7.5 Hz, Ar-H3), 3.4 (m, 4H, CH2), 2.8 (m, 4H,
CH2), 1.3 (m, 8H, CH2), 0.9 (t, 12H, CH3). 13C NMR (75.48 MHz,
2
acetone-d6): δ 148.0 (C7), 146.3 (C6), 132.7 (q, J(C,F) ) 33.3 Hz,
Five millimeter NMR tubes were charged with 100 µL of substrate,
2 mg of catalyst 2, and 1200 µL of deuterated solvent and placed into
a 200 MHz spectrometer. p-H2 enriched to 50% was prepared via
catalytic equilibration over charcoal at 77 K and injected repeatedly in
synchronization with the pulsed NMR experiment via an electrome-
chanically lowered glass capillary mechanism. Higher levels of p-H2
enrichment, namely, >97%, were achieved for some of the reactions
using a closed cycle cooler cryostat.
C10, C12), 132.5 (C5), 130.4 (C4), 126.3 (C2), 124.8 (C3), 125.5 (C9,
1
C13), 123.4 (q, J(C,F) ) 271 Hz, CF3), 123.2 (C11), (CdC of alkene
and CdN not observed). 19F NMR (282.41 MHz, acetone-d6):
δ
-63.70. Exact mass determination: found m/z 1020.2752, calcd for
C44H41N2F18Pt, 1020.2730.
N,N,N-Triethyldeuterioammonium triflate (5). A sample of 0.6
mL (6.8 mmol) of trifluoromethyl sulfonic anhydride was slowly added
to a Schlenk tube containing 1.0 mL of D2O. After the reaction mixture
was stirred for 1 h at 21 °C, 2.0 mL of triethylamine was carefully
added so that the reaction heat could be controlled. The mixture was
stirred for another hour at 21 °C, after which the solvent was evaporated.
The product could be recrystallized from a dichloromethane/hexanes
mixture at -80 °C, yielding 0.45 g. N,N,N-triethyldeuterioammonium
Synthesis. The atomic numbering for the Ar-bian ligands is given
in Figure 10.
[Pd{(m,m′-(CF3)2C6H3)-bian}(ma)] (2). An amount of 0.40 g (1.12
mmol) of (m,m′-(CF3)2C6H3)-bian was dissolved in dry THF (10 mL).
Then, Pd(nbd)(ma) (0.40 g, 1.34 mmol) was added, and the reaction
mixture was stirred for 20 min at room temperature. The reaction
mixture was filtered through Celite filter aid under N2 atmosphere to
remove traces of metallic palladium. The Celite filter aid was repeatedly
washed with dry THF (3 × 5 mL) until the washings were colorless.
The combined filtrates were evaporated to dryness in vacuo, and the
last traces of solvent were removed by stripping with dry pentane. Then,
the product was carefully washed with dry pentane and dried in vacuo.
1H NMR (499.84 MHz, acetone-d6): δ 8.40 (d, J(H,H) ) 8.5 Hz, 2H,
Ar-H5), 8.24 (br, 4H, Ar-H9,13), 8.19 (br, 2H, Ar-H11), 7.75 (pst, 2H,
Ar-H4), 7.46 (d, 2H, J(H,H) ) 7.5 Hz, Ar-H3), 4.25 (br, 2H, CdC-
H). 13C NMR (125.70 MHz, acetone-d6): δ 171.11 (CdO), 1740.40
(C1), 151.11 (C8), 146.29 (C7), 133.86 (q, 2J(C,F) ) 33.3 Hz, C10, C12),
133.11 (C5), 132.73 (C6), 129.90 (C4), 127.44 (C2), 125.86 (C3), 124.19
(q, 1J(C,F) ) 271 Hz, Ar-CF3), 123.27 (C9, C13), 122.22 (C11), (CdC
of alkene not resolved). 19F NMR (282.41 MHz, acetone-d6): δ -63.16.
IR (KBr): 1840 (CdO), 1730 (CdO). Anal. Calcd for C32H14F12N2O3-
Pd: C, 47.52; H 1.74; N 3.46. Found: C, 47.65; H, 1.81; N, 3.43.
1
triflate as a highly hygroscopic white solid. H NMR (300.13 MHz,
CD2Cl2): δ 3.15 (q, J(H,H) ) 7.0 Hz, 6H, CH2), 1.28 (t, J(H,H) ) 7.0
Hz, 9H, CH3). 13C NMR (75.48 MHz, CD2Cl2): δ 120.69 (q, J(C,F)
) 318 Hz, CF3), 47.27 (NCH2CH3), 8.67 (NCH2CH3). 19F NMR (282.41
2
MHz, CD2Cl2): δ -79.80. H NMR (76 MHz, CD2Cl2): δ 7.48 (br,
s).
Acknowledgment. This research has been supported (in part)
by the Chemistry Council of The Netherlands Foundation for
Pure Research (CW-NWO).
Supporting Information Available: Information concerning
the derivation of the rate law. This material is available free of
charge via the Internet at http://pubs.acs.org.
JA052729J
9
15480 J. AM. CHEM. SOC. VOL. 127, NO. 44, 2005