Equilibrium control in enyne metathesis: Crossover studies and the kinetic reactivity of (E,Z)-1,3-disubstituted-1,3-dienes

Article abstract of DOI:10.1021/jo0482209

(Chemical Equation Presented) The stereoselectivity of diene bond formation in the ruthenium-carbene mediated intermolecular enyne metathesis was studied. Initial reaction between an alkyne and 1-hexene gave mixtures of E- and Z-isomers in the newly forme

Full text of DOI:10.1021/jo0482209

literature variation in E/Z stereoselectivity might be
explained by equilibration (e.g., the E-isomers were being
formed in certain cases due to thermodynamic control).
We hypothesized that the produced dienes might con-
tinue to react since precatalyst 1 is known to react with
internal alkenes5 and because dienes will participate in
cross alkene metathesis.6 In this report, the stereoselec-
tivity of enyne metathesis was examined using the second
generation Grubbs carbene 1. The fast metathesis pro-
duced initial mixtures of E- and Z-isomers, which was
followed by a dynamic process leading to the most stable
E-diene in certain cases (Scheme 1).
Equilibrium Control in Enyne Metathesis:
Crossover Studies and the Kinetic
Reactivity of
(E,Z)-1,3-Disubstituted-1,3-Dienes
Anthony J. Giessert and Steven T. Diver*
Department of Chemistry, University at Buffalo, the State
University of New York, Buffalo, New York 14260
diver@buffalo.edu
Received October 8, 2004
SCHEME 1. Equilibration of Stereoisomeric
Mixtures with Alkene Crossover
The stereoselectivity of diene bond formation in the ruthe-
nium-carbene mediated intermolecular enyne metathesis
was studied. Initial reaction between an alkyne and 1-hex-
ene gave mixtures of E- and Z-isomers in the newly formed
1,3-diene. However, over time the mixtures equilibrated to
form mostly the diene of the E-configuration. To evaluate
individual reactivity of the E- and Z-dienes, they were
independently synthesized. The E-diene was found to be
kinetically-stable under nominal metathesis conditions while
the Z-diene isomerized to the E-isomer. The Z-isomeric
dienes were found to react with other alkenes to produce a
new diene of E-configuration. A secondary metathesis mech-
anism involving ruthenium alkylidene intermediates was
invoked to explain the dynamic stereochemistry observed
in this study.
The reactivity of diene intermediates is a central
question in an equilibration mechanism, and yet is not
immediately answerable from prior work in enyne me-
tathesis. In the equilibration process, excess R2-CHd
CH2 will react with the dienes to eventually form mostly
E-diene product (eq 1). Equilibration might be achieved
if the reaction was under thermodynamic control since
the E-dienes are more stable than the Z-dienes. However,
equilibration may also be achieved if the Z-diene inter-
mediate of Scheme 1 was kinetically-reactive and the
E-diene was kinetically-stable. In this study, reactivity
of the dienes shown as intermediates in Scheme 1 was
directly evaluated by chemical synthesis and crossover
experiments. In the crossover paradigm, the Z-intermedi-
ate should react with R3-CHdCH2 to give crossover
product (eq 2). If the E-isomer was kinetically reactive,
it too should produce crossover product. Since equilibra-
tion requires reaction of a metal carbene, the efficiency
of this process may depend on the reactivity of the
alkylidene derived from the alkene reactant.
Enyne metathesis has proven efficient for the forma-
tion of 1,3-dienes.1 Yet as powerful as the cross enyne
metathesis is as a carbon-carbon bond-forming reaction,
its usefulness is limited by lack of selectivity (typically
about 1:1).2 Even with the more reactive second genera-
tion Grubbs’ carbene 1, mixtures of stereoisomers are
typically obtained in cross metathesis1a,3 and there are
comparatively few examples that exhibit high E-selectiv-
ity.4 Most cases exhibiting high E-selectivity make use
of the second generation Grubbs ruthenium carbene
catalyst 1 and aliphatic 1-alkenes (eq 1). Some of the
Direct kinetic monitoring of a cross metathesis showed
equilibration of a kinetic mixture of dienes. The reaction
of propargylic alkyne 2 and 1-hexene (eq 3) initially forms
a mixture of E/Z-isomers (Figure 1). The progress of the
enyne metathesis was followed by 1H NMR spectroscopy.
(1) (a) Diver, S. T.; Giessert A. J. Chem. Rev. 2004, 104, 1317-1382.
(b) Poulsen, C. S.; Madsen, R. Synthesis 2003, 1-18. (c) Mori, M. Top.
Organomet. Chem. 1998, 1, 133-154.
(4) (a) Lee, H.-Y.; Kim, B. G.; Snapper, M. L. Org. Lett. 2003, 5,
1855-1858. (b) Diver, S. T.; Giessert, A. J. Synthesis 2004, 466-471.
(c) Lee, H.-Y.; Kim, H. Y.; Tae, H.; Kim, B. G.; Lee, J. Org. Lett. 2003,
5, 3439-3442.
(2) (a) Schuster, M.; Blechert, S. Angew. Chem., Int. Ed. Engl. 1997,
36, 2037-2056. (b) Smulik, J. A.; Diver, S. T. Tetrahedron Lett. 2001,
42, 171-174.
(5) For a full account of alkene reactivity in alkene metathesis, see:
Chatterjee, A. K.; Choi, T.-L.; Sanders, D. P.; Grubbs, R. H. J. Am.
Chem. Soc. 2003, 125, 11360-11370.
(3) In particular, enol ethers give isomeric mixtures: (a) Giessert,
A. J.; Brazis, N. J.; Diver, S. T. Org. Lett. 2003, 5, 3819-3822. (b)
Giessert, A. J.; Snyder, L.; Markham, J.; Diver, S. T. Org. Lett. 2003,
5, 1793-1796.
(6) For a recent example, see: Wang, X.; Porco, J. A. J. Am. Chem.
Soc. 2003, 125, 6040-6041.
10.1021/jo0482209 CCC: $30.25 © 2005 American Chemical Society
Published on Web 01/12/2005
1046
J. Org. Chem. 2005, 70, 1046-1049

After 50 min, the ratio was found to be 55/41 E/Z, with
4% unreacted alkyne. The E/Z mixture was present until
all the alkyne was consumed, at which time the diene
mixture isomerized to form predominantly the E-diene
product E-3.7 An inspection of the data for the isomeric
dienes between 200 and 700 min shows a rate of diminu-
tion of the Z-isomer that is about equal to the growth
rate of the E-isomer. This rate was considerably slower
than the intermolecular metathesis, but occurred in the
time scale under which normal cross metatheses are
usually performed. We considered that the observed
isomerization could result from a secondary metathesis
between diene and excess 1-alkene.
Intermediate ruthenium hydrides were not observed
in the precatalysts nor were they detected in the
reaction mixtures. Isomerization of Z-3 in Figure 1
might be due to a ruthenium hydride species. It is known
that ruthenium hydrides are capable of catalytic
olefin isomerization.8 Examination of precatalyst 1 by
1H NMR showed the absence of ruthenium hydride
species, and 31P NMR showed only the resonance of 1.
Also, during the course of the reaction of eq 3, no metal
FIGURE 1. Alkyne 2 conversion to E-3.7
more effectively to metal hydrides in general, and in the
present case, isomerization would produce conjugated
styrene 7, providing a thermodynamic trap. The metath-
esis of eq 4 produced diene 6 in 43% yield as the
E-isomer.10 Significantly, the intermediate E/Z mixture
could be detected at early timepoints in the reaction, with
no â-methylstyrene detected by GC. On the basis of the
lack of alkene isomerization in the alkene reactant and
the fact that the methylene-truncated product 8 was not
found, the observed isomerization of the diene mixture
to E-6 is most likely not due to a ruthenium hydride
species.
The kinetic reactivity of the E and Z dienes 3 under
the reaction conditions was next investigated. As a test
for secondary metathesis reactivity, crossover experi-
ments were performed. Exposure of E-3 to the reaction
conditions of eq 3 with 9 equiv of 1-octene did not result
in the production of the crossover product showing that
the E-isomer is stable to the reaction conditions. The
same experiment could not be performed on pure Z-3
since the Z-isomer cannot be obtained by enyne cross
metathesis. At best, mixtures of E/Z-isomers can be
prepared. Therefore another method for the preparation
of Z-diene was necessary.
hydride species were observed by 1H NMR. Grubbs
recently showed that a ruthenium hydride complex can
form by decomposition of LnRudCH2, and isomerizes a
terminal alkene into an internal alkene.9 However,
neither the characteristic hydride peak for this complex
(δ -8.6) nor the 31P peak for CH3PPh3 (δ 34.5) were
+
detected in the reaction, or after the in situ isomerization
had occurred.
Since it is possible that hydride species were formed
below the detection limits of NMR spectroscopy, a further
experiment with an isomerization-prone reactant was
performed. In this paradigm, undetected quantities of
ruthenium hydride should result in terminal alkene
isomerization, converting the alkene reactant allyl ben-
zene 5 to â-methyl styrene 7 (eq 5). In fact, the metathesis
The desired Z-dienes were prepared by using the
stereoselective Wittig olefination. Bromoallyl acetate 9
underwent Kishi-Nozaki coupling under Taylor’s condi-
tions11 with dihydrocinnamaldehyde, which gave substi-
tuted allyl alcohol 10 (Scheme 2). Routine manipulation
furnished the desired aldehyde 11 in 53% yield over three
steps. With the desired aldehyde in hand, Wittig olefi-
nation12 was used to form the desired dienes Z-3 (>90%
1
Z-isomer by H NMR of the crude reaction) and Z-12
1
(>93-95% Z-isomer by H NMR of the crude reaction).
Access to the stereochemically pure Z-dienes permitted
investigation of their kinetic stability to the metathesis
of eq 4 is a demanding test because terminal alkenes bind
(7) The final mixture is 3:1 E/Z 3.
(8) (a) Trnka, T. M.; Morgan, J. P.; Sanford, M. S.; Wilhelm, T. E.;
Scholl, M.; Choi, T.-L.; Ding, S.; Day, M. W.; Grubbs, R. H. J. Am.
Chem. Soc. 2003, 125, 2546-2558. (b) Dinger, M. B.; Mol, J. C.
Organometallics 2003, 22, 1089-1095 (c). Sutton, A. E.; Seigal, B. A.;
Finnegan, D. F.; Snapper, M. L. J. Am. Chem. Soc. 2002, 124, 13390-
13391. (d) Ulman, M.; Grubbs, R. H. J. Org. Chem. 1999, 64, 7202-
7207.
(10) In this case, only 3 equiv of alkene were used. A minor
byproduct was the corresponding butadiene.
(11) Taylor, R. E.; Ciavarri, J. P. Org. Lett. 1999, 1, 467-469.
(12) (a) Valentine, D. H., Jr.; Hillhouse, J. H. Synthesis 2003, 317-
334. (b) Rein, T.; Pedersen, T. M. Synthesis 2002, 579-594. (c)
Nicolaou, K. C.; Harter, M. W.; Gunzner, J. L.; Nadin, A. Liebigs Ann.
Recl. 1997, 1283-1301. (d) Vedejs, E.; Peterson, M. J. Adv. Carbanion
Chem. 1996, 2, 1-85. (e) Maryanoff, B. E.; Reitz, A. B. Chem. Rev.
1989, 89, 863-927. (f) Schlosser, M. Top. Stereochem. 1970, 5, 1-30
(9) Hong, S. H.; Day, M. W.; Grubbs, R. H. J. Am. Chem. Soc. 2004,
126, 7414-7415.
J. Org. Chem, Vol. 70, No. 3, 2005 1047

TABLE 1. Crossover Results Obtained with Diene Z-12
SCHEME 2. Preparation of Z-Dienes Used in This
Studya
entry
alkenea
conversion
R
1
2
3
4
5
none
99
93
80
99
61
Me, E-12
Bu, E-3
Bu, E-3
Bu, E-3
H, 13
a Reagents and conditions: (a) CrCl2, NiCl2, DMF, PhCH2CH2-
CHO. (b) TBSCl, imidazole, CH2Cl2. (c) K2CO3, MeOH. (d) Swern
oxidation. (e) RCHPPh3Br, KHMDS, THF, -78 °C to room
temperature.
1-hexene
cis-5-decene
trans-5-decene
ethene (100 psig)d
a Conditions: 0.11-0.24 M Z-12, 5 mol % of 1, 9 equiv of alkene
(0.99-2.16 M) in C6D6, 30 °C. b 0.925 M ethene.
conditions. Exposure of Z-12 to precatalyst 1 resulted in
conversion to the E-isomer, E-12. Since evidence of
ruthenium hydrides could not be found, the isomerization
of Z-12 led us to consider a metathetical mechanism. The
two most plausible pathways are outlined in Scheme 3.
SCHEME 3
FIGURE 2. Secondary metathesis of 1-hexene with Z-12,
producing crossover isomerization product E-3.
A metal carbene (R′ ) H, alkyl, Ph) can add to the
Z-alkene A by path a, then fragment to the E-diene C,
which should also show crossover when R * R′. The
carbene entering eq 6 (LnRudCHR′) will produce a
different carbene on crossover (LnRudCHR), and the new
carbene will promote further isomerization in the re-
maining Z-diene reactant (e.g., in propagation steps).13
In path b, the opposite regiochemistry of initial metal
carbene cycloaddition (D vs B) ultimately produces the
vinyl carbene E (eq 7), which would then give productive
cross metathesis with added alkene. The vinyl carbene
E reacts via metallacyclobutane F to give crossover
product.14
The proposed metathesis-based isomerization mecha-
nisms of Scheme 3 were tested by crossover experiments
with alkenes. The crossover reaction produced the sub-
stituted E-dienes (eq 8). Diene Z-12 is effectively trans-
formed into the E-isomer via exposure to the second
generation Grubbs’ complex 1. It was also possible to
exchange the 1-methyl group of Z-12 for a butyl group
with 1-hexene and 5-decene (Table 1). The comparable
conversions obtained for internal alkenes (entries 3 and
4) compared to 1-hexene (entry 2) are notable. In fact,
the relative reaction rates for crossover/isomerization are
2.08 for 1-hexene, 2.19 for cis-5-decene, and 2.36 for
trans-decene, indicating that the alkylidenes produced
in the latter two cases react as quickly as the metal
carbene produced in the reaction with 1-hexene. This
argues convincingly that ruthenium pentylidenes are
effecting isomerization/crossover in each case, since
LnRudCH2 cannot be formed in the fast reactions of
entries 3 and 4. The effectiveness of the conversion to
E-3 is likely a function of carbene longevity; some of the
lower conversions in Table 1 are due to catalyst decom-
position. In particular, this may explain the low conver-
sion with ethylene (entry 5).15 Excess allyl trimethyl-
silane did not react with Z-12 to produce the expected
crossover product. The ineffectiveness of allyl trimeth-
ylsilane to give crossover product indicates that the
derived ruthenium carbene does not react with Z-12.
Representative crossover of Z-12 with 1-hexene is
illustrated in Figure 2. The time course of the secondary
metathesis was monitored by 1H NMR in C6D6 at 30 °C,
using a 9-fold excess of 1-hexene, identical conditions as
those in Figure 1. Most of the reaction has occurred in
the first 200 min, similar to the isomerization observed
in the 1-hexene cross metathesis of Figure 1. The data
show that the Z-diene reactant disappears at a rate about
equal to the rate of appearance of the E-crossover
product. This shows that an intermediate is not ac-
cumulated during the metathesis isomerization/cross-
over.
(13) In other words, only the initiation step will provide crossover.
The propagating carbene will bear the same R group as the diene
reactant and show apparent isomerization as the R group of the Z-diene
A is replaced with the R group to furnish the E-diene C where R′ ) R.
(14) A less likely pathway is the dipolar fragmentation of intermedi-
ate ruthenacyclobutanes which would produce open zwitterions, which
could collapse after bond rotation.
(15) For an excellent discussion of methylidene reactivity, see:
Sanford, M. S.; Love, J. A.; Grubbs, R. H. J. Am. Chem. Soc. 2001,
123, 6543-6554.
1048 J. Org. Chem., Vol. 70, No. 3, 2005

TABLE 2. Crossover Results Obtained with Diene Z-3
initial phase of the cross metathesis shown in Figure 1.
A secondary metathetical transformation to exchange
alkyl groups followed by a ruthenium hydride isomer-
ization is also unlikely. For the E-crossover product to
be produced in Figure 2, the isomerization rate must be
at least as fast as cross metathesis/crossover such that
crossover product Z-3 is never observed. If hydrides were
highly reactive, they would have shown isomerization in
the metathesis in eq 4.
entry
alkenea
conversion
R
1
2
3
4
5
none
63
54
81
75
50
Bu, E-3
Me, E-12
Me, E-12
Me, E-12
H, 13
The best explanation for crossover is the agency of
ruthenium alkylidenes not methylidenes. The use of
methylidene-free conditions17 with both cis- and trans-
5-decene (entries 3 and 4, Table 1) gave crossover, prov-
ing that alkylidenes are competent to perform the iso-
merization. This illustrates that the isomerization/cross-
over is not dependent upon ethylene nor does it require
a ruthenium methylidene species. Yet not all 1-alkenes
isomerize the Z-dienes, despite their potential to produce
LnRudCH2. Butadienes are detected as minor byproducts
(less than 10%) in metathesis isomerization/crossover
with 1-alkenes, which reveals the presence of LnRudCH2
(and possibly the formation of ethylene) in a minor
pathway. The origin of LnRudCH2 could be due to either
fragmentation of F (eq 7) or background homodimeriza-
tion of 1-alkenes. However, butadienes are less reactive
than Z-dienes used in the crossover experiments. For
instance, isolated butadiene 13 (entry 5, Table 2) gave
less than 50% conversion to cross product after 6 h under
the conditions of Figure 1. Moreover, the isolation of
butadiene as a minor byproduct in the crude reaction
shows that they are kinetically-stable under the reaction
conditions.
In summary, we have shown that certain E-selective
enyne cross metatheses produce initial E/Z mixtures that
isomerize over the course of the reaction. Crossover
between excess 1-alkenes and the stereochemically-pure
Z-isomers is consistent with an E-selective secondary
metathesis (diene-alkene cross metathesis). Internal
alkenes react at rates comparable to 1-alkenes and these
all react faster than ethylene. The data illustrate the
ruthenium alkylidenes are the probable carbene inter-
mediates in the secondary metathesis and can react with
the more reactive Z-dienes, present in the initial product
mixture. This accounts for high E-selectivity observed in
certain enyne metatheses. Further studies on the mech-
anism of enyne metathesis are in progress.
1-propene
cis-2-butene
trans-5-butene
ethene (100 psig)b
a Conditions: 0.11-0.24 M Z-3, 5 mol % of 1, 9 equiv of alkene
(0.99-2.16 M) in C6D6, 30 °C. b 0.925 M ethene.
The crossover reaction of butyl-substituted diene Z-3
with ethylene, 1-propene, and 2-butenes was similarly
explored (eq 9). Crossover of diene Z-3 was slower than
that of the lower homolog Z-12 and did not go to
completion (Table 2). This is likely due to the difference
in steric environment at the terminus of the diene. For
example, the isomerization rate by 1 alone (5 mol %) was
four times slower than the corresponding isomerization
rate of methyl-substituted diene Z-12 in Table 1.16 Use
of 1-propene (entry 2) provided similar results to cis-2-
butene (entry 3). The reaction was efficient with trans-
2-butene, providing E-12 (entry 4). Though the reactions
did not go to completion, formation of the 2-substituted
butadiene 13 was also possible by reaction with ethylene
(entry 5). The ethylene crossover proved slower than all
the others and proceeded at a rate comparable to the rate
of catalysis by 1 alone.
These data are consistent with a metathesis-based
isomerization of Z-dienes present in the initial mixture
of diene products. The isomerization by Grubbs’ complex
1 without added alkene is significant. In these cases
(entries 1 of Tables 1 and 2), the reaction of complex 1
with the Z-diene is an initiation step, which produces the
ethylidene, LnRudCHMe, on one turnover. The eth-
ylidene is the propagating carbene that effects isomer-
ization of the remaining Z-diene (R′ ) Me in path a of
Scheme 3). However, during the initiation step, mecha-
nistic path a predicts styryl transfer from complex 1, and
path b predicts the formation of â-methylstyrene. Au-
thentic samples were prepared for GC comparison.
Analysis of the 1-catalyzed isomerization of Z-12 did not
show â-methylstyrene (R′ ) Ph), but did show the styryl
transfer (product of eq 8, R ) Ph), providing support for
isomerization via eq 6. It is notable that the crossover
products are not produced as E/Z mixtures, but as the
E-isomers. This may reflect higher intrinsic selectivity
of the less reactive Z-diene (viz 1-alkenes) proceeding via
stereocontrolled formation of metallacycle B (path a,
Scheme 3) or reflect higher selectivity of a vinyl carbene
intermediate (e.g. E, path b, Scheme 3). The vinyl carbene
selectivity would be different from that of the more
reactive metal carbenes that are encountered in 1-
alkene-alkyne cross metathesis, e.g., during the fast
Acknowledgment. The authors thank Amol
Kulkarni for helpful discussions and Dr. Richard Fisher
of Materia, Inc. for providing catalyst support. We grate-
fully acknowledge the National Cancer Institute
(R01CA090603) for financial support of this work.
Supporting Information Available: Experimental pro-
cedures and full characterization data for new compounds Z-3,
E-3, E-6, 10B, 10C, 11, E-12, Z-12, and 13; conditions used
for the rate plots; and detailed procedures for the crossover
experiments. This material is available free of charge via the
Internet at http://pubs.acs.org.
JO0482209
(16) These isomerizations proceed to higher conversion at elevated
temperatures. To allow for a consistent comparison of the data in
Tables 1 and 2, we used the same reaction conditions for both
isomerization/crossover experiments.
(17) (a) Kulkarni, A. A.; Diver, S. T. Org. Lett. 2003, 5, 3463-3466.
(b) Kulkarni, A. A.; Diver, S. T. J. Am. Chem. Soc. 2004, 126, 8110-
8111.
J. Org. Chem, Vol. 70, No. 3, 2005 1049