Please do not adjust margins
ChemComm
Page 3 of 4
DOI: 10.1039/C5CC07255G
Journal Name
COMMUNICATION
decreased and the anti/syn ratio of the product was inversed (entry for the high anti/syn ratio of 85/15 and enantioselectivity of 92%
2’). Moreover, the reaction became very sluggish, affording only obtained when water was used as additive (Table 1, entry 8). The
trace products if no acids were added. Similar result was observed importance of the nitro-group on p-nitrobenzaldehyde in hydrogen-
when water was added as the additive. That indicates that water is bonding was examined by using benzaldehyde to take the place of
too weak as an acid to assist the direct aldol reaction. The p-nitrobenzaldehyde as the reactant, and it was found that both the
transition-state model of the protonated catalytic center was aldehyde conversion and enantioselectivity decreased markedly
proposed by Luo et al.,7 showing that E-isomer of the enamine and (Table S2, ESI).
steric hindrance between the cyclohexene and benzene rings were
the main factors for formation of anti-isomer with (2R,1’S) enantio- DDAC-SH-SBA-15 was refluxed in a toluene solution containing
selectivity in this reaction (Scheme S1). trimethylsilyl chloride for 24 h to convert the surface silanol to
In order to provide evidence of the roles of the silanol groups,
When immobilized DDAC-SH-SBA-15 was used as the catalyst trimethylsilyl ether. The catalytic performance of resultant Me3Si-
(10 mol%), the aldol reaction is markedly accelerated even without DDAC-SH-SBA-15 in the aldol reaction of CH and NBA is poor and
any additive (entry 3), in contrast to the trace conversion over only trace amount of aldehyde conversion is obtained (Table 1,
pristine organocatalyst. With TFA as the additive, the reaction rate is entry 9). This result confirms the importance of the hydrogen-
increased but is slightly slower than that of the pristine bonding provided by the silanol groups on the silica wall in the
organocatalyst (entry 4). However, it is surprising to find that the catalytic reaction.
anti/syn ratio of the products is reversed from 76/24 obtained in
homogeneous system to 27/73 after the chiral diamines DDAC were
immobilized on SBA-15. Products of similar reversed anti/syn ratio
are observed with 2,2,3,3-tetrafluoro succinic acid as the additive
(entry 5). Nevertheless, the enantioselectivities of syn-isomers are
rather low with these strong acid additives.
If weak acid additives were introduced, the diastereo-
selectivities over DDAC-SH-SBA-15 were found to be also reversed
from that observed in homogeneous system (entries 6-7).
Moreover, (2R,1’S) isomer is the main product, similar to that
observed in the homogeneous system with strong acid additives. To
our delight, an extremely high anti/syn ratio of 85/15 and (2R,1’S)
enantioselectivity of 92% were obtained when water was used as
the additive (entry 8). This finding is unique since water is not an
effective additive in the homogeneous system.
Changes in stereoselectivity with supported catalysts were
previously reported in enantioselective reactions promoted by
immobilized metallic complexes.12 A complete reversion in anti/syn
Scheme 3 Proposed transition states for syn- and anti-aldoladducts in the
confined space of DDAC-SH-SBA-15.
ratio and an increase in the enantioselectivity of the major product
were described in the Mukaiyama-aldol reaction of 1-phenyl-1-
trimethylsilyloxyethene with α-ketoesters catalyzed by clay
supported bis(oxazoline)–copper complexes. It is attributed to a
change in the symmetry of the complex when it is placed in the
vicinity of the clay surface. That reason however is not applicable to
our case with the organic catalyst.
The opposite catalytic performance and different product
selectivities observed on SBA-15 immobilized catalyst is obviously
attributed to the confined space in the mesoporous silica. With
strong acid additives, the gegen ion may adhere to protonated
catalytic center in the confined space (Scheme 3A). It repels the
aldehyde substrate from approaching the catalytic center and the
unfavorable syn isomers are formed as the main products. The low
enantioselectivities of syn products confirm this proposal. More
interestingly, high enantioselectivities of anti-aldol (2R,1’S) product
were obtained when less acidic additives or even water were added
(Table 1, entries 6-8). These results were totally contrary to that
observed under homogeneous condition.
The recyclability of the catalyst was examined by separating the
spent catalysts from the liquid products by simple filtration, and the
recovered catalyst was used in the next run. Entries 1-3 in Table S3
(ESI) show the results under the same reaction condition as that of
Entry 8, Table 1 but using 0.2 mL water as the solvent without acid
additive. The recycled catalyst shows only slight decrease in
conversions, diastereoselectivities and enantioselectivities. Another
series of experiments were carried out to see if the changes in
stereoselectivity over the supported DDAC catalyst was permanent.
The same sample of solid catalyst using water additive was
recovered by filtration, washing with acetylacetone and methanol,
and used with TFA additive for second run, and finally recovered
and reused again with water. The results were shown in Entries 2-4
in Table S2 (ESI). The recyclability of the catalyst and the
dependence of anti/syn ratios and enantioselectivities on the
additives were clearly demonstrated.
In conclusion, SBA-15 immobilized primary-tertiary diamine
catalyst developed in this work demonstrated unprecedented
alternation of the diastereoselectivity of the products in asymmetric
direct aldol reactions, attributing to the confined space of
mesopores and the interaction of the surface silanol groups with
the intermediate species in the transition state. High anti-
diastereoselectivities of the products were obtained when water
was added in direct aldol reaction of cyclohexanone with p-
nitrobenzaldehyde, providing a green route for organic synthesis.
To achieve the anti-aldol product, some forces should be
present to keep the aldehyde and cyclohexene in E-geometry under
the catalytic transition-state. One of the possible forces is the
hydrogen-bonding between the nitro-group at the para-position of
the benzaldehyde and surface silanol group, as shown in Scheme
3B. This kind of interaction is probably more drastic if the less acidic
additive or water is present. In addition, water may also contribute
the cage effect and trap the reactants and the organic moieties
around catalytic centers. These two roles of water probably account
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins