Iridium-Mediated Asymmetric Hydrogenation of 2,3-Diphenylbutadiene: A Revealing Kinetic Study

Article abstract of DOI:10.1021/ja037653a

Hydrogenation of 2,3-diphenylbutadiene mediated by a chiral iridium carbene oxazoline complex was studied. The kinetics of the reaction showed that it occurred in two distinct stages. The first corresponded to slow consumption of the diene to form half-hydrogenated products, i.e., monoenes. After all the diene was consumed, however, the reaction was much faster and more stereoselective. These data were interpreted in terms of possible catalytic intermediates. Copyright

Full text of DOI:10.1021/ja037653a

Published on Web 10/30/2003
Iridium-Mediated Asymmetric Hydrogenation of 2,3-Diphenylbutadiene: A
Revealing Kinetic Study
Xiuhua Cui and Kevin Burgess*
Department of Chemistry, Texas A & M UniVersity, Box 30012, College Station, Texas 77842-3012
Received July 30, 2003; E-mail: burgess@tamu.edu
1
Chiral derivatives of Crabtree’s catalyst A, including Pfaltz’
2
3-8
phosphine oxazoline systems B, some related N,P-complexes,
and carbene oxazoline complexes like C,9,10 are unusual in that
they can be used for asymmetric hydrogenations of arylalkenes that
have no other functional groups. However, there are no reports of
these catalysts being applied to reduce prochiral dienes or polyenes.
Indeed, the literature on asymmetric hydrogenations of dienes of
any kind, functionalized or otherwise, with any catalyst, is
sparse.1
1-13
Consequently, we initiated a program to study asym-
metric hydrogenations of dienes with little or no other functionality,
beginning here with a kinetic study for 2,3-diphenylbutadiene 1.
While the data obtained in this preliminary work have no immediate
synthetic application, they are highly informative for this particular
substrate and may illustrate general facets of the mechanism of
hydrogenation by iridium catalysts A-C.
Hydrogenation of 1 to the corresponding 2,3-diphenylbutanes
could occur directly or via formation of the alkenes 2, as indicated
in Figure 1a. Authentic samples of compounds 2-4 were prepared,
the hydrogenation reaction was performed as indicated in reaction
Figure 1. (a) Hydrogenation of 2,3-diphenylbutadiene, with relative rate
constants in units of 10 mol L min from the slopes in part b. (b)
Concentrations of the reaction components as the reaction proceeds.
-
4
-1
-1
1
, and the reaction was followed via GC using a chiral column
Figure 1b).14
(
Four products formed during the initial phase of reaction 1.
Prevalent among these were the alkene 2 and its enantiomer, ent-
2. The rate ratio for formation of these two products was only 1.28;
i.e., the enantioselectivity was low at that stage. The third product
was meso-3. It formed more slowly (rate of formation of {2 +
ent-2}:meso-3 ) 8.45), while the stereoisomeric products 3 and
ent-3 were conspicuously absent during this phase of the reaction.
A series of experiments were performed to test the origin of
meso-3 in the first phase of the reaction. To facilitate this, a racemic
sample of compound 5 was prepared. Hydrogenation of an
equimolar solution of 5 and racemic 2 with catalyst C demonstrated
that the former substrate reacted slightly faster. Then, in the key
experiment, reaction 2, a mixture of 1 and racemic 5 was reduced.
Product 6 (as a mixture of stereoisomers) should prevail in reaction
2 if in reaction 1 meso-3 was formed in the initial stages of the
reaction via addition of one molecule of hydrogen, dissociation of
2 from the metal, recomplexation, and then a second reduction.
Conversely, the dominant product would be meso-3 if that same
The initial phase of the reaction (until ca. 475 min in Figure 1b)
was characterized by zero-order consumption of diene 1. Extrapola-
-
1
tion of the concentration of 1 to 1 mol L indicates there was an
induction period of ca. 4 min. Control experiments (NMR) indicate
that catalyst C in the absence of hydrogen did not exchange 1,5-
cyclooctadiene (COD) for excess 1 to any significant extent under
the conditions of this experiment. Moreover, pretreatment of C with
hydrogen in the absence of 1 gave a mixture of complexes that
were almost inactive for hydrogenation of 1 added later. Thus, the
induction period involved removal of the COD ligand in the
presence of 1 and hydrogen to give the initial catalyst species.
14212
9
J. AM. CHEM. SOC. 2003, 125, 14212-14213
10.1021/ja037653a CCC: $25.00 © 2003 American Chemical Society

C O MMU N I C A T I O N S
material in the featured reaction 1 was formed via delivery of both
molecules of hydrogen to 1 without dissociation from the iridium
center. In the event, meso-3 was the major product of reaction 2,
and Ir(V) intermediates D and E,16 i.e., via concomitant migratory
insertion of the alkene unit into an Ir-H bond, and oxidative
addition of dihydrogen. One straightforward way to explain the
experimental data described in this Communication is that, in
reaction 1, diene 1 replaces COD as a 4e donor in complex C, and
then, assuming the carbene oxazoline and diene ligands do not
dissociate and the electron count around iridium does not exceed
2
implying that direct addition of two molecules of H is the dominant
pathway. Formation of meso-3, rather than 3, is consistent with
two additions of hydrogen to the diene coordinated in a syn
conformation.
2
18, only one H entity could be accommodated around the metal.
Thus, while diene is present, the reaction is likely to be restricted
to intermediates like F and a mechanism that involves cycling
between Ir(I) and Ir(III). After the diene is consumed, however,
catalytic cycles that involve Ir(III) and Ir(V) become accessible.
In the case of diene 1, the initial phase of the reaction, possibly
involving Ir(I) and Ir(III), is less rapid and selective than the
Ir(III)/Ir(V) cycle. This could imply that asymmetric catalysis
involving Ir(I)/Ir(III) is limited, and that other metal complexes
that cannot undergo redox steps between M(III) and M(V) are
unlikely to have the same reactivity as the iridium systems A-C.
Indeed, we have prepared the rhodium analogue of complex C and
find it is relatively unreactive (see Supporting Information).
Tetrasubstituted alkene 4 (Figure 1a), the fourth product of the
initial phase of the reaction, forms in the first stage, and then its
concentration remains nearly constant. Control experiments indicate
that alkene 4 is not consumed under the conditions of this
experiment. Its cis isomer is not formed in the reaction.
After approximately 475 min, when all the diene 1 is consumed,
the reaction profile changed markedly. The fastest transformation
after that point was conversion of 2 into meso-3; this occurred 31.3
times faster than conversion of 2 into ent-3. Conversely, ent-2 was
reduced to the chiral product 3 1.94 times faster than to meso-3.
Thus, a significant stereochemical bias to the second hydrogenation
step is imparted by the chiral center of the substrates 2 and ent-2,
but the reaction is still catalyst-controlled. Even though alkene 2
was the major product after 475 min, the stereochemical outcome
was reversed at the end of the process such that 3, not ent-3,
prevailed. The overall enantioselectivity for formation of 3 was
the combined effect of the two steps: rapid conversion of 2 into
meso-3, where the substrate and catalyst stereoselectivities are
matched,15 and preferential conversion of ent-2 into 3, where the
catalyst stereoselectivity prevails over substrate control.
Acknowledgment. In memory of Arthur E. Martell; October
18, 1916 - October 15, 2003. Financial support for this work was
provided by The Robert Welch Foundation and Johnson Matthey
plc. MS spectra were recorded in the TAMU/LBMS-Applications.
Supporting Information Available: Details of the kinetic experi-
ments, syntheses performed to obtain authentic materials, control NMR
experiments, and reactions with the rhodium analogue of C (PDF).
This material is available free of charge via the Internet at http://
pubs.acs.org.
References
A series of conclusions about the hydrogenation of diene 1 can
be drawn from the data above. First, consistent with our prior
observations,10 removal of the COD ligand requires a significant
induction time and the presence of hydrogen. Once the COD group
is removed, it is rapidly replaced by 1,3-butadiene 1. The reduction
clearly occurs in two phases, corresponding to reduction of the first
diene double bond, and then the second. First, relatively slow, poorly
stereoselective hydrogenation of 1 ensues to give mostly 2 and ent-
(1) Crabtree, R. Acc. Chem. Res. 1979, 12, 331-337.
(2) Lightfoot, A.; Schnider, P.; Pfaltz, A. Angew. Chem., Int. Ed. 1998, 37,
2897-2899.
(3) Blankenstein, J.; Pfaltz, A. Angew. Chem., Int. Ed. 2001, 40, 4445-4447.
(
4) Cozzi, P. G.; Zimmermann, N.; Hilgraf, R.; Schaffner, S.; Pfaltz, A. AdV.
Synth. Catal. 2001, 343, 450-454.
(5) Menges, F.; Pfaltz, A. AdV. Synth. Catal. 2002, 344, 40-44.
(
6) Menges, F.; Neuburger, M.; Pfaltz, A. Org. Lett. 2002, 4, 4713-4716.
7) Tang, W.; Wang, W.; Zhang, X. Angew. Chem., Int. Ed. 2003, 42, 943-
946.
(
(8) Hou, D.-R.; Reibenspies, J. H.; Colacot, T. J.; Burgess, K. Chem.-Eur.
J. 2000, 7, 5391-5400.
9) Powell, M. T.; Hou, D.-R.; Perry, M. C.; Cui, X.; Burgess, K. J. Am.
Chem. Soc. 2001, 123, 8878-8879.
2, some meso-3 (mainly via a route that does not involve
(
dissociation of an alkene intermediate), and a small amount of
alkene 4. At the end of this first phase, when diene 1 is finally
consumed, the overall reaction accelerates by a factor of ap-
proximately 1.7. Thereafter, 2 is reduced mainly to meso-3, while
ent-2 is preferentially hydrogenated to 3; i.e., the enantioface
preference is reversed. Thus, the rate and stereochemical data
presented here indicate significant mechanistic differences between
the first and second stages of the reaction.
(
10) Perry, M. C.; Cui, X.; Powell, M. T.; Hou, D.-R.; Reibenspies, J. H.;
Burgess, K. J. Am. Chem. Soc. 2003, 125, 113-123.
(11) Muramatsu, H.; Kawano, H.; Ishii, Y.; Saburi, M.; Uchida, Y. J. Chem.
Soc., Chem. Commun. 1989, 769-770.
(
12) Burk, M. J.; Allen, J. G.; Kiesman, W. F. J. Am. Chem. Soc. 1998, 120,
657-663.
(
13) Beghetto, V.; Matteoli, U.; Serivanti, A. Chem. Commun. 2000, 155-
156.
(14) The reaction was repeated many times. It is relatively insensitive to small
2
H pressure changes and the reaction vessel/stirring (i.e., diffusion effects).
(
15) Masamune, S.; Choy, W.; Peterson, J. S.; Sita, L. R. Angew. Chem., Int.
Ed. Engl. 1985, 24, 1.
16) Brandt, P.; Hedberg, C.; Andersson, P. Chem.-Eur. J. 2003, 9, 339-347.
The assertions outlined above may be pertinent to other systems.
Recent theoretical studies on hydrogenation of alkenes using Pfaltz’
complexes B suggest that the mechanism proceeds via the Ir(III)
(
JA037653A
J. AM. CHEM. SOC.
9
VOL. 125, NO. 47, 2003 14213