J. R. Moulins, D. J. Burnell / Tetrahedron Letters 52 (2011) 3992–3994
3993
Some metathesis reactions, especially enyne metathesis,9,10
have been improved in terms of yield and E/Z ratio when the reac-
tions were conducted under an atmosphere of ethene. The ratio-
nale10 for the improved E/Z ratio is that the products of the
primary metathesis, that is, kinetic products, undergo further
metathesis with ethene to give a terminal alkene, which in turn
undergoes metathesis again with the initial alkene substrate to
provide, after many cycles, thermodynamic products via secondary
metathesis. N-Heterocyclic carbene- and phosphine-based ruthe-
nium catalysts have been employed in enyne metathesis under
ethene. Thus, the presence of ethene was expected to encourage
equilibration of the cross-metathesis products.
When 0.1 M benzene solutions of 5a with 2.5 mol % of 1 were
maintained at 40 °C under an atmosphere of ethene, the 5b/5c
ratio was 11:1 after 2 h, and this did not change over the next
24 h. After 2 h the yield of the self-coupled products was 27%,
but after 24 h the yield was still only 32%. It appeared that the
thermodynamic E/Z ratio (ca. 11:1) had been reached, but
under the ethene atmosphere the equilibrium between 5a and its
cross-metathesis products 5b,c was favouring 5a. When a cross-
metathesis reaction of 5a was carried out with 2.5 mol % of 1 as
above, but for 3 h under N2 and then 1 h under ethene, the 5b/5c
ratio of the product was 11:1 and the yield was 45%. The E/Z ratio
was larger than the corresponding experiment in Table 1 after 4 h,
but the yield was lower. This was presumed to be due to reversion
of some of the product to 5a.
Table 1
Isolated yields and E/Z ratios from the metathesis reactions shown in Scheme 1 as a
function of time
Alkene
Yield of cross-metathesis products, E/Z ratio
5 min
30 min
4 h
24 h
3a
4a
5a
6a
16%, 4.4:1
68%, 4.4:1
23%, 1.9:1
35%, 7.7:1
27%, 4.6:1
66%, 6.4:1
41%, 3.0:1
46%, 9.7:1
34%, 5.3:1
69%, 7.2:1
66%, 6.4:1
59%, 16:1
58%, 5.7:1
62%, 7.6:1
61%, 7.9:1
51%, 18:1
The acetate 3a reacted relatively slowly (Table 1). The reaction
of 3a required 24 h for the yield of the 1,4-diacetoxy-2-butenes
3b,c to surpass 50%, and over that time the ratio of the E- and Z-
isomers increased modestly from 4.4:1 after 5 min up to 5.7:1.6
In contrast, the benzyl ether 4a underwent metathesis quickly,
and the isolated yield after 5 min was the same as after 4 h. The
yield after 24 h was a little lower, which likely reflected some
destruction of the products. In spite of little change in the com-
bined yield of the 2-butene products 4b,c, the E/Z ratio rose from
an initial ratio of 4.4:1 up to 7.6:1.
The tritylated derivative 5a and the tert-butyldimethylsilyl
derivative 6a both reacted more slowly than did 4a, and the iso-
lated yields of the 2-butene products 5b,c and 6b,c were at their
highest after 4 h. However, with these substrates the E/Z ratios
continued to increase over the next 20 h. After 24 h the ratio of
tritylated products 5b/5c had risen from 1.9:1 to 7.9:1. The idea
that a product of direct decomposition of the catalyst, which would
accumulate with time while the reaction was heated, was respon-
sible for E/Z isomerization was not consistent with the following
experiment. A solution of 1 was heated in benzene to 40 °C for
5 h. Then, 5a was added, and the mixture was heated for 30 min
under conditions that were the same as used for the results in Ta-
ble 1. The yield of 5b,c was 29%, as opposed to 41% with ‘fresh’ cat-
alyst, and so heating the catalyst had clearly degraded some of the
catalyst. However, the 5b/5c ratio was only 2.2:1, as opposed to
3.0:1 with ‘fresh’ catalyst; therefore, a decomposition product
was not enhancing the ratio of E/Z in this instance.7 Carrying out
the reaction of 5a in the presence of 2.5 mol % of 1 resulted in no
significant differences in the evolution of the E/Z ratios.8
The silylated derivatives 6b,c were obtained with a 6b/6c ratio
of 18:1 after 24 h. It seemed that the bulkier derivatives led to a
greater preponderance of the E-isomer after time. The increases
in E/Z ratios over time did not appear to be due, to a significant
extent, to a selective degradation of the Z-isomer. This is because
the increases in E/Z ratios between the 30 min and the 4 h
reactions were accompanied by increases in the yields of the
metathesis products, not decreases. The data are more consistent
with equilibration of the products via secondary metathesis from
a kinetically derived mixture with a low E/Z ratio to a thermody-
namic mixture with a higher E/Z ratio.
The yields of the cross-coupled products did not progress be-
yond 70%. Indeed, prolonging the reaction time resulted in an ero-
sion of yield, which one presumes was due to slow secondary
reactions of the products. Two experiments were conducted in
which a total of 2.5 mol % of 1 was added using 5a as the substrate
at the same initial concentration in benzene at 40 °C for 48 h. In the
first instance, the catalyst was added all at once. In the second, the
catalyst was added in four equal portions every 12 h. The yield of
the first experiment was 49% and the 5b/5c ratio was 8.4:1. The
yield of the second experiment was 67% and the 5b/5c ratio was
11:1. The yield of this second experiment was no improvement
over the yield after 4 h, but the E/Z ratio was significantly en-
hanced. In contrast with the initial experiments, these experiments
had been conducted in sealed vessels from which the byproduct of
cross-metathesis, ethene, would not have escaped.
If the E/Z ratios of cross-metathesis reactions are established by
equilibration, even with a first generation catalyst such as 1, then
the efficiency of equilibration could be assessed by comparing
the equilibration of the Z-compounds 3c–6c under the usual condi-
tions for cross-metathesis. Accordingly, benzene solutions of sam-
ples highly enriched in 3c–6c were warmed to 40 °C under
atmospheres of nitrogen gas and of ethene for 10 min in the pres-
ence of catalyst 1. The results are presented in Table 2.
The implications of the experiments under the nitrogen atmo-
sphere were that the equilibration of the Z-isomers 3c–6c was ra-
pid. Also, equilibration mediated by 1 did not require the presence
of some of the allyl compound 3a–6a. Nevertheless, the recovery of
product was not quantitative in any instance. Equilibration was
even more rapid under ethene, which implied that equilibration
via the allyl compound is more rapid. The equilibration of the
diacetate substrate 3c was anomalous in that the proportion of
the E-isomer 3b was greater under the nitrogen atmosphere than
under ethene. This is important because with catalyst 1, substrate
3c is known to homologate an alkene with higher E-selectivity than
the more reactive allyl acetate 3a.11 Our results suggest that the
reverse might be true if the acetyl protecting group on the sub-
strate were exchanged for a benzyl, trityl or TBDMS group. In every
instance with ethene, the recovery of isomerized products 3b,c–
6b,c was very poor even after just 10 min with much of the sub-
strate having reverted to the allyl compound 3a–6a. In a similar
way, ring-opening metathesis with 1 under an atmosphere of eth-
ene can provide the ring-opened product without subsequent
metathesis reactions taking place.12
Table 2
Isolated yields and E/Z ratios from the treatment of 3c–6c (0.1 M) in benzene in the
presence of 1 (2.5 mol %) at 40 °C for 10 min
Alkene
Yields and E/Z ratios
CH2CH2 atmosphereb
N2 atmospherea
3b/c 1:48
4b/c 1:23
5b/c 1:40
6b/c 1:13
90%, 1:5.3
83%, 1:2.9
86%, 1:23
96%, 1.2:1
48%, 1:10
33%, 3.1:1
21%, 1:5.3
20%, 2.6:1
a
Reaction conducted under a slow stream of dry N2.
Reaction conducted under a slow stream of ethene.
b
Moulins, Jonathan R.
