Angewandte
Communications
Chemie
catalysts for intramolecular carbonyl olefin metathesis but not
for intermolecular reactions.[10]
A freshly open sample of anhydrous FeCl3, used in
a glovebox, did not catalyze the metathesis reaction. How-
ever, when 50 mol% of water were added, the metathesis
reaction started. Addition of a proton quencher such as di-
tert-butyl pyridine to the reaction medium (50 mol%) stops
not only the metathesis but also any other parasite reaction
(aldol, hydrolysis of the vinyl ether, etc.), for all the soluble
catalysts tested. These results indicate that in situ formed
protons are also the catalytically active species during the
metal salt-catalyzed intermolecular carbonyl–olefin metathe-
sis, and with this in mind and the positive results obtained for
soluble Brçnsted acids, a variety of commercially-available
solid acids (entries 8–15) were tested as catalysts for the
metathesis (Supporting Information, Table S4).[11] The results
in Table 1 show that all the solid Brçnsted acid catalysts tested
gave similar product yields and selectivity to the soluble acids
for alkene product 3b, including the widely available, inex-
pensive, and non-toxic pillared clay montmorillonite K10
(<10 euroskgÀ1, entries 14–15). The ester by-products 4a,b
are completely unreactive under the present reaction con-
ditions for any catalyst, according to control experiments, and
the purification of the final alkene products by column
chromatography becomes easier in the absence of additional
alkene/aldehydes/ketones in the mixture. These results support
our starting hypothesis that vinyl ethers are suitable alkene
partners for the intermolecular carbonyl–olefin metathesis.
A convenient and productive way to circumvent side
reactions during the intermolecular metathesis could be to
perform the reaction in flow over a fixed-bed reactor with
a solid catalyst, in order to rapidly separate the acid catalyst
from the reaction mixture after the metathesis reaction.
Gratifyingly, a consistent good yield of 3b (70–95%, GC)
over 8 h in-flow reaction time was obtained with fixed-bed
montmorillonite K10 (entry 15 in Table 1). Figure 1 (top)
shows that the procedure with montmorillonite K10 in
continuous mode is also suitable to prepare a variety of
trans alkenes in high yields after just 1 h on stream time
(condition A), generally in better yields than with soluble
BF3·OEt2 catalyst in batch (condition B). The main side
reaction corresponds to vinyl ether hydrolysis, which explains
the better reactivity of the in situ generated vinyl ethers from
acetals 2a–g (see also Figure S4 in the Supporting Informa-
tion).[12] Good to moderate yields are generally obtained, and
some functional groups are shown to be tolerated, including
halides (compounds 3g, 3i–l), aromatic ethers (3g), esters
(3n), amides (3t), thioethers (3v), and other alkenes (3x).
Figure 1 (bottom) shows that montmorillonite K10 is still
active after 40 h in-flow, which gives a turnover number
(TON) of 150, which is significantly higher than homogeneous
catalysts.[3] This backs up the robustness of the solid catalyst to
give access to multi-gram amounts of alkene 3b. Furthermore,
the selectivity towards the alkene product is consistently
higher than 90%, highlighting the benefits of performing the
reactions in flow with a solid catalyst (see also Table S5 in the
Supporting Information).
Figure 1. Top: Carbonyl olefin metathesis reactions of aryl aldehydes
1b,d–i with in situ formed vinyl ether 2d–g catalyzed by montmorillon-
ite K10 (A) or BF3·OEt2 (B) under the indicated reaction conditions.
Bottom: Kinetic profile for the metathesis reaction between 1b and 2d
in a fixed-bed tubular reactor with montmorillonite K10 catalyst (0.5m
toluene, 0.01 mLminÀ1 over 40 h at 1208C). Color Scheme: starting
aldehyde 1b in red, alkene product 3b in green, selectivity in blue.
Error bars account for a 5% uncertainty.
Brçnsted acids, correlate well with the pKa of the catalyst
for the homogeneous acids and also for montmorillonite K10,
and deviate towards higher values for other solid acids (see
also Figure S5, and Figure S6 in the Supporting Information
for entropy correlation).[11b] Kinetic experiments at different
stirring rates show that the reaction is controlled by diffusion
for all solid catalysts except for montmorillonite K10,[13] which
strongly supports the assumption that the reaction enthalpy is
directly related to proton strength.
Figure 3 (top) shows experimental and computational
evidences about the mechanism of the reaction (see also
Figure S7 and Table S6 in the Supporting Information). A
Hammett plot with different para-substituted acetophenones
(Supporting Information, Figure S8) gives 1 = À2.7, which
Figure 2 shows that the activation enthalpy values for the
metathesis reaction between 1b and 2b with different
2
ꢀ 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2019, 58, 1 – 5
These are not the final page numbers!