First Evidence That the Mechanism of Catalytic Hydrogenation with Homogeneous Palladium and Rhodium Catalysts Is Strongly Influenced by Substrate Polarity

Full text of DOI:10.1021/ja964179i

J. Am. Chem. Soc. 1997, 119, 5257-5258
5257
Scheme 1
First Evidence That the Mechanism of Catalytic
Hydrogenation with Homogeneous Palladium and
Rhodium Catalysts Is Strongly Influenced by
Substrate Polarity
Scheme 2
Jinquan Yu and Jonathan B. Spencer*
UniVersity Chemical Laboratory
UniVersity of Cambridge, Lensfield Road
Cambridge CB2 1EW, U.K.
ReceiVed December 4, 1996
Homogeneous palladium and rhodium catalysts have assumed
an important position in synthetic chemistry because of their
utility in the semireduction of alkynes1-2 and the asymmetric
reduction of prochiral alkenes.3-6 Although there have been
many studies conducted, the electronic mode in which the
transition metal hydrogen species adds to the double bond of
the alkene has never been determined. It has been reported that
the geminal hydrogens of terminal alkenes can be exchanged
by deuterium with homogeneous catalysts.7-9 We have recently
found that during the reduction of cis-alkenes, such as cis-
crotonic acid, with deuterium some trans-alkenes are formed
that contain one deuterium atom (Scheme 1). This identifies
that deuterium is directly involved in the isomerization of the
cis-alkenes and has allowed us to investigate for the first time
the electronic properties of the hydrogen bonded to the metal,
by determining how electron-donating or -withdrawing groups
conjugated to the double bond influence the location of the
hydrogen addition. If the metal hydrogen bond is polarized in
the catalyst, the hydrogen should selectively add to the most
electronically favored end of the double bond.
We report in this paper that by using this methodology we
have been able to demonstrate that the palladium hydrogen bond
can be polarized in either mode (a) Pdδ+-Hδ- or (b) Pdδ-
-
Hδ+), whereas the rhodium catalyst studied is dominated by
mode a (Rhδ+-Hδ-). This provides strong evidence that the
mechanism of catalytic hydrogenation with homogeneous
catalysts is a two electron process that can be dramatically
affected by the Coulombic nature of the substrate.
ated compounds, such as crotonic acid, are polarized so that
the carbon remote to carbonyl group is electron deficient
(compared to the carbon next to the functional group) and readily
react at this position with nucleophilic species. The deuterium
distribution in 1 and 2 would therefore be consistent with the
palladium-hydrogen bond being polarized so that the hydrogen
was electronegative and the palladium electropositive (mode a
Pdδ+-Hδ-) just prior to their addition to the double bond. The
intermediate formed after addition can then receive another
hydrogen from the palladium to form the saturated alkane or
rotate with the subsequent elimination of the hydrogen attached
to the remote carbon to give the deuterated trans-alkene (Scheme
2).
To investigate how a functional group with the opposite
characteristics to a carbonyl group might influence the mode
of addition of the palladium hydrogen species to the double
bond, compounds containing a phenyl group conjugated to the
double bond were studied using unlabeled 3 and the cis-isomer
of 4 (Table 1). The aromatic group, particularly with a
p-methoxy substituent, will polarize the double bond so that
the remote carbon is electron rich compared to the one next to
the functional group. The deuterium distribution in compounds
3 and 4 (formed using the palladium catalyst) shows that the
label is located solely in the position remote to the aromatic
group. This result is completely opposite to what would be
predicted if the palladium hydrogen bond is polarized in mode
a (Pdδ+-Hδ-) as was proposed for crotonic acid (1) and ethyl
crotonate (2). More consistent with this regioselectivity of the
deuteration would be for the palladium-hydrogen bond to be
polarized in the opposite way (mode b Pdδ--Hδ+) so that the
palladium was electronegative and the hydrogen electropositive
(Scheme 2).
Pure cis-alkenes with a variety of different functional groups
attached (to the double bond) were individually isomerized using
deuterium gas and either the palladium or rhodium catalyst to
give a mixture of the labeled trans-alkene (10-45%) and alkane.
NMR analysis of the trans-alkenes clearly showed that the
deuterium is not evenly distributed across the double bond
(Table 1). Deuterium is found only to be present in crotonic
acid (1) and ethyl crotonate (2) at the carbon remote to the
carbonyl group when the isomerization is carried out with bis-
[1,2-bis(diphenylphosphino)ethane]palladium(0). R,â-Unsatur-
(1) (a) Schrock, R. R.; Osborn, J. A. J. Am. Chem. Soc. 1971, 93, 3089
-3091. (b) Schrock, R. R.; Osborn, J. A. J. Am. Chem. Soc. 1976, 98,
2143 -2147.
(2) (a) Yoshida, T.; Youngs, W. J; Sakaeda, T.; Ueda, T.; Otsuka, S.;
Ibers, J. A. J. Am. Chem. Soc. 1983, 105, 6273-6278. (b) Burch, R. R.;
Shusterman, A. J.; Muetterties, E. L.; Teller, R. G.; Williams, J. M. J. Am.
Chem. Soc. 1983, 105, 3546-3556.
(3) Miyashita, A.; Yasuda, A.; Takaya, H.; Toriumi, K.; Ito, T.; Souchi,
T.; Noyori, R. J. Am. Chem. Soc. 1980, 102, 7932-7934.
(4) (a) Noyori, R. Science 1990, 248, 1194-1199. (b) Noryori, R.;
Takaya, H. Acc. Chem. Res. 1990, 23, 345-350.
(5) (a) Ashby, M. T.; Halpern, J. J. Am. Chem. Soc. 1991, 113, 589-
594. (b) Ashby, M. T.; Khan, M. A.; Halpern, J. J. Organometallics 1991,
10, 2011-2015.
(6) Brown, J. M. Angew. Chem., Int. Ed. Engl. 1987, 26, 190-203. (b)
Brunner, H. Synthesis 1988, 645-654.
(7) Gungor, M.; Jardine, F. H.; Wheatley, J. D. J. Chem. Soc., Dalton.
Trans. 1988, 1653-1656.
(8) Hardick, D. J.; Blagbrough, I. S.; Potter, B. V. L. J. Am. Chem. Soc.
1996, 118, 5897-5903.
(9) Harthun, A.; Selke, R.; Bargon, J. Angew. Chem., Int. Ed. Engl. 1996,
35, 2505-2507.
S0002-7863(96)04179-0 CCC: $14.00 © 1997 American Chemical Society

5258 J. Am. Chem. Soc., Vol. 119, No. 22, 1997
Communications to the Editor
Table 1
deuterium
distribution (%)
relative ratio of
deuterium (%)
isomerization
product
catalysta
I
II
I
II
b
1
Figure 1. Part of the H NMR spectrum (250 MHz, CDCl3) (double
bond region) for the trans-enol ether 8 formed by the isomerization of
the cis-isomer using D2 and (a) bis[1,2-bis(diphenylphosphino)ethane]-
palladium(0) or (b) chlorotris(triphenylphosphine)rhodium(I). The
triplets observed inside the doublets show deuterium hydrogen coupling
that occurs in the labeled compound.
lysts is most likely a two-electron process that, for palladium,
is strongly influenced by the Coulombic properties of the
substrate.
Previous work has established that the formation of trans-
alkenes by â-elimination occurs with cis-geometry11 (Scheme
2) and would predict 100% incorporation of deuterium into
alkenes 1-8. Intriguingly, the level of deuterium incorporated
is substantially less (Table 1), suggesting either a different
mechanism is taking place or there are competing mechanisms.
It is possible that the addition of metal and hydrogen to the
double bond is not concerted. If the metal adds first to the
double bond of the cis-alkene, to give a charged intermediate
whose lifetime was sufficient for rotation to take place to the
trans-conformation, then isomerization could occur without
incorporation of deuterium. Alternatively, hydrogen could add
first to the double bond to give a charged intermediate that again
would allow rotation to take place, with the double bond being
reformed by elimination of a hydrogen or a deuterium. This
last step might be subject to a kinetic isotope effect and should
give at least a 50% incorporation of deuterium into the trans-
alkene. Having functional groups conjugated to the double bond
that are capable of stabilizing such charged intermediates might
favor these stepwise pathways and explain why such molecules
isomerize readily during hydrogenation compared to simple cis-
alkenes.
a Pd: bis[1,2-bis(diphenylphosphino)ethane]palladium(0). Rh: chlo-
rotris(triphenylphosphine)rhodium(I) (Wilkinson’s catalyst). b Cis/trans
refers to the deuterium being cis/trans to the aromatic group. Some
compound was identified with 2 deuteriums in the terminal position of
the double bond. The proportion of this compound increased with a
longer reaction time as did that of the monodeuterated species.
To further investigate which mode of addition is preferred,
cis-alkenes with functional groups conjugated to both ends of
the double bond were isomerized with deuterium and the
palladium catalyst. The results show for cis-isomer of cinnamic
acid (5) that either position can be deuterated; 86% of the
deuterium is remote to the carbonyl group with only 14% remote
to the phenyl group. The double bond of 5 is strongly polarized
and can therefore promote both modes of addition, however,
the deuterium distribution shows that mode a is favored.
Compounds 6 and 7 differ from 5 by having an o- or p-methoxy
group in the aromatic ring and result in a significant increase
in the proportion of deuterium remote to this aromatic group
compared to the same position in 5. The aromatic ring
substituted with an o- or p-methoxy group will be more electron
releasing than the phenyl ring of cinnamic acid and hence may
aid attack of an electrophilic hydrogen on the carbon remote to
this group.
The results clearly show that steric effects have only a minor
role on the regioselectivity of the metal hydrogen addition to
the double bond of these substrates since compound 7 containing
a p-methoxy group is sterically no more demanding than
cinnamic acid (5).
When the same approach was used to investigate the
electronic properties of a rhodium catalyst [chlorotris(tri-
phenylphosphine)rhodium(I)], the results with compounds 3, 5,
and 8 (Table 1) reveal that compared with the palladium catalyst
rhodium-hydrogen addition occurs predominantly by mode a
(Rhδ+-Hδ-). This difference in electronic properties as
highlighted for enol ether 8 (Figure 1) cannot be explained by
the metal’s influence on a one-electron process.10 This strongly
suggests that catalytic hydrogenation with homogeneous cata-
The identification that the mode of addition is strongly
governed by the polarity of the substrate should greatly aid the
understanding of asymmetric reduction of alkenes as it affords
an insight into the likely structure of the intermediate formed
during their hydrogenation.12 The approach reported here should
also prove useful for the rational design of new catalysts, as
their electronic properties can be readily determined using this
methodology.
Acknowledgment. The Royal Society for the award of a Royal
Society University Research Fellowship (J.B.S.) and British Council
and the Chinese Government for the award of Sino-British Friendship
Scholarship (J. Yu).
Supporting Information Available: Experimental details and
spectroscopic and analytical data for compounds 1-8 (6 pages). See
any current masthead page for ordering and Internet access instructions.
JA964179I
(11) (a) Evans, J.; Schwartz, J.; Urquhart, P. W. J. Organomet. Chem.
1974, 81, C37-C39. (b) McCarthy, T. J.; Nuzzo, R. G.; Whitesides, G. M.
J. Am. Chem. Soc. 1981, 103, 3396-3403.
(12) (a) Halpern, J. In Asymmetric Synthesis; Morrison, J. D., Ed.;
Acedemic Press, Inc., Ltd.: London, 1985; Vol. 5, pp 41-68. (b) Brown,
J. M. Chem. Soc. ReV. 1993, 25, 25-41.
(10) (a) Kochi, J. K. In Free Radicals; Olah, G. A., Ed.; John Willey &
Sons: New York, 1982; Vol. 2, pp 283-288 . (b) Leory, C.; Peeters, D.;
Wilante, C. J. Mol. Struct. 1982, 5, 217-223.