L-proline-grafted mesoporous silica with alternating hydrophobic and hydrophilic blocks to promote direct asymmetric aldol and Knoevenagel-Michael cascade reactions

Article abstract of DOI:10.1021/cs500385s

Promotion of heterogeneous asymmetric catalysis is of major interest in the asymmetric catalysis field. In this work, a novel strategy for the synthesis of l-proline-grafted mesoporous silica with alternating hydrophobic and hydrophilic blocks to promote the heterogeneous asymmetric catalysis was reported. The surface synergies in the neat environment and the interface acceleration in aqueous medium thereby fostered high catalytic activities and enantioselectivity in the direct aldol reaction and the Knoevenagel-Michael cascade reaction. The l-proline loading could be reduced to as low as 0.63 mol %, giving 95% ee for anti-isomers and 81% ee for syn-isomers in the catalytic asymmetric aldol reaction of nitrobenzaldehyde and cyclohexanone, which was hard to accomplish on the homogeneous counterpart. In the direct asymmetric aldol reaction of ethyl-2-oxoacetate and cyclohexanone, 82% yield in 24 h and 90% ee were achieved. More exciting, the catalysts were applied to more exigent reactions. As an example, in the Knoevenagel-Michael cascade reaction, 85% yield in 10 h and up to 91% ee was achieved.

Full text of DOI:10.1021/cs500385s

Research Article
pubs.acs.org/acscatalysis
L‑Proline-Grafted Mesoporous Silica with Alternating Hydrophobic
and Hydrophilic Blocks to Promote Direct Asymmetric Aldol and
Knoevenagel−Michael Cascade Reactions
Zhe An, Ying Guo, Liwei Zhao, Zhi Li, and Jing He*
State Key Laboratory of Chemical Resource Engineering, Beijing University of Chemical Technology, P.O. Box 98, Beijing 100029,
China
S
* Supporting Information
ABSTRACT: Promotion of heterogeneous asymmetric catal-
ysis is of major interest in the asymmetric catalysis field. In this
work, a novel strategy for the synthesis of L-proline-grafted
mesoporous silica with alternating hydrophobic and hydro-
philic blocks to promote the heterogeneous asymmetric
catalysis was reported. The surface synergies in the neat
environment and the interface acceleration in aqueous medium
thereby fostered high catalytic activities and enantioselectivity
in the direct aldol reaction and the Knoevenagel−Michael
cascade reaction. The L-proline loading could be reduced to as
low as 0.63 mol %, giving 95% ee for anti-isomers and 81% ee
for syn-isomers in the catalytic asymmetric aldol reaction of
nitrobenzaldehyde and cyclohexanone, which was hard to accomplish on the homogeneous counterpart. In the direct asymmetric
aldol reaction of ethyl-2-oxoacetate and cyclohexanone, 82% yield in 24 h and 90% ee were achieved. More exciting, the catalysts
were applied to more exigent reactions. As an example, in the Knoevenagel−Michael cascade reaction, 85% yield in 10 h and up
to 91% ee was achieved.
KEYWORDS: asymmetric catalysis, aldol reaction, cascade reaction, surface synergy, interface acceleration, mesoporous silica
work by List and Barbas and their cowokers,30−32 which applied
L-proline as a catalyst for the direct asymmetric aldol reaction
INTRODUCTION
■
Improving the catalytic activity and selectivity by virtue of
between acetone and a variety of aldehydes. Good results were
support synergies in the heterogeneous catalysis has always
been an important issue.1−5 However, it remains a challenge
achieved for α-branched aliphatic aldehydes; however, quite fair
especially when it comes to asymmetric reactions.6−8
enantioselectivities were observed for aromatic aldehydes. On
the other hand, a catalyst loading of as high as 20 to 30 mol %
Mesoporous silica, with well-ordered nanopores, narrowly
distributed pore sizes, and modifiable mesoporous surfaces,
was usually required to obtain a good isolated product yield.
Impressive progress has been made in the L-proline-catalyzed
provides potential solutions to this issue by demonstrating, for
direct asymmetric aldol reaction by derivation of proline.36−41
Using L-proline derivatives, satisfactory enantioselectivity was
observed for aromatic aldehydes.36,37 Implementing L-proline
derivative catalyzed asymmetric aldol reactions in water could
reduce the catalyst loading from 20 to 30 mol % to less than 10
mol %.38−43 To derive highly efficient and enantioselective
heterogeneous catalyst, attachment of L-proline or its structural
analogues to the surfaces of mesoporous silica solids,44−46
MCM-41, for example, or the interlayer regions of layered
solids,47 layered double hydroxides, for example, was also
attempted. Moderate enantiomer excess values were usually
achieved. Few positive cases42,48−54 have been reported
recently. But more effective strategies to produce viable
example, confinement effects.9−13 In metal-catalyzed asymmet-
ric reactions, confining catalytic sites in nanosized channels or
pores was observed to visibly improve the catalytic activity and/
or enantioselectivity.14−16 Chiral organocatalysts, which are
superior to asymmetric transition-metal catalysts because they
are metal-free, inexpensive, stable, and usually capable of
working under aerobic atmosphere with wet solvents, have
attracted much attention in the past decade.17,18 Unfortunately,
more effective routes or greater success is yet desired for highly
efficient and enantioselective heterogeneous organocataly-
sis.19,20
L-Proline was the first reported and one of the most
important organocatalysts, demonstrating efficacy in a variety of
key asymmetric carbon−carbon and carbon−heteroatom-
forming reactions,21−29 such as Diels−Alder, Michael, Mannich,
and especially direct aldol reactions.30−35 The breakthrough for
the asymmetric catalysis of L-proline came from the pioneering
Received: March 24, 2014
Revised: June 19, 2014
© XXXX American Chemical Society
2566
dx.doi.org/10.1021/cs500385s | ACS Catal. 2014, 4, 2566−2576

ACS Catalysis
Research Article
Scheme 1. Catalysts Prepared in This Work
to be added,56 and the initial molar ratio was Si/CTAB/TMAH/
NaOH/H2O = 1/0.27/0.57/0.14/100. The mixture was stirred at
room temperature for 1 h and aged at 100 °C for 24 h. The resulting
white precipitate was filtrated to yield the as-synthesized sample.
CTAB was removed by solvent extraction. As-synthesized sample (1.0
g) was stirred in 200 mL of ethanol with 3.0 g of HCl (36%) at 55 °C
for 6 h before filtration.
Modification of L-Proline. The following protocol describes the
synthesis of N-tert-butoxycarbonyl-trans-4-hydroxy-L-proline: A mix-
ture of trans-4-hydroxy-L-proline (1.5 g, 0.038 mol) in a 2:1 mixture of
THF/H2O (75 mL) was treated first with 10% aqueous NaOH (15
mL) and then with di-tert-butyldicarbonate (12 g, 0.056 mol). The
reaction mixture was stirred at room temperature overnight, and then
the THF was removed by vacuum. The residue was adjusted to pH 2−
3 by the addition of 10% aqueous NaHSO4. The acidic solution was
extracted several times with ethyl acetate. The combined organic
extracts were washed with H2O and brine, and then dried over
anhydrous Na2SO4. Removal of the desiccant and evaporation of the
solvent in vacuum gave N-tert-butoxycarbonyl-trans-4-hydroxy-L-pro-
line in a yield of 91%, which was used without further purification. 1H
NMR (600 MHz, CDCl3): δ =1.35 (s, 9H, C(CH3)3), 2.06−2.35 (m,
2H, β-CH2 Pro), 3.36−3.58 (m, 2H, δ-CH2 Pro), 4.30−4.43 (m, 2H,
α-CH Pro, γ-CH Pro). 13C NMR (400 MHz, CDCl3): δ = 177.74,
81.99, 77.33, 77.02, 76.70, 54.69, 39.01, 37.3.
The following protocol describes the synthesis of (2S,4R)-1-(tert-
butoxycarbonyl)-4-(4-vinylbenzyloxy)pyrrolidine-2-carboxylic acid:
(2S,4R)-1-(tert-Butoxycarbonyl)-4-(4-vinylbenzyloxy)pyrrolidine-2-
carboxylic acid was synthesized following the method reported by
Gruttadauria et al.42,50 with some modifications. A solution of trans-
Boc-4-hydroxy-L-proline (2.0 g, 8.65 mmol) in anhydrous THF (30
mL) was added dropwise under nitrogen atmosphere to a suspension
of NaH (60% mineral oil, 751 mg, 18.77 mmol) in anhydrous THF
(20 mL) at 0 °C. The mixture was stirred at ambient temperature for 1
h, and 18-crown-6 (228 mg, 0.86 mmol) and 4-chloromethylstyrene
(90%, 2.0 g, 11.6 mmol) were then added. The mixture was stirred for
1 h at ambient temperature and then at 50 °C overnight. After cooling
to ambient temperature, water (100 mL) was added. The aqueous
phase was extracted with cyclohexane in order to remove the
unreacted 4-chloromethylstyrene. The aqueous phase was acidified to
pH 2−3 by adding a solution of NaHSO4 (2 M) and was then
extracted with ethyl acetate. The organic phase was dried using MgSO4
and concentrated under reduced pressure to give the desired product
as pale yellow, viscous oil in 76% yield. 1H NMR (600 MHz, CDCl3):
δ = 1.41 (s, 9H, 3x −CH3), 2.10−2.50 (m, 2H, −CHCH2CH−),
3.56−3.62 (m, 2H, −CHCH2N−), 4.14−4.20 (m, 1H, −OCHH-
(CH2)2−), 4.48−4.54 (m, 3H, −CHCOOH− and −C6H4CH2O−),
5.25 and 5.75 (2H, HHCCH), 6.72 (1H, HHCCH), 7.27 and
7.40 (4H, −C6H4−). 13C NMR (400 MHz, CDCl3): δ = 177.00,
154.30, 137.20, 137.16, 136.38, 127.79, 126.25, 113.87, 80.79, 77.44,
77.32, 76.8, 60.44, 38.84, 37.81.
enantioselective heterogeneous catalyst and catalysis in exigent
reactions are much in demand.
Here we have demonstrated the achievement of highly
efficient and enantioselective catalysis by heterogeneous L-
proline catalyst in a quite low loading by use of mesoporous
silicas with alternating hydrophobic and hydrophilic blocks
(HHBs) in the pore wall. By virtue of the specific micro-
environment in the nanochannels of HHBs materials, the
resulting heterogeneous catalyst has been found to effectively
promote the asymmetric reaction through not only the surface
synergy in the neat environment but also the interface
acceleration in aqueous medium, which was hard to accomplish
with the homogeneous counterpart. In the direct asymmetric
aldol reaction of nitrobenzaldehyde and cyclohexanone, 98%
yield in 8 h and up to 96% ee were afforded with a proline
loading of 6.7 mol %. More exciting, the catalyst could be
applied to more exigent reactions with satisfactory yield and
enantioselectivity. In the direct asymmetric aldol reaction of
ethyl-(2)-oxoacetate and cyclohexanone, 82% yield in 24 h and
90% ee were achieved with a proline loading of 6.3 mol %. In
the Knoevenagel−Michael cascade reaction, 85% yield in 10 h
and up to 91% ee was achieved with a proline loading of 13.0
mol %.
EXPERIMENTAL DETAILS
■
Preparation of Mercaptopropyl-Functionalized Mesoporous
Silicas. Mercaptopropyl-functionalized mesoporous silicas were
synthesized following the method reported by Inagaki et al.55 with
some modifications. Through tailoring the molar ratio of 1,4-
bis(triethoxysilyl)benzene (BTEB), tetraethylsilicate (TEOS), and 3-
mercaptopropyl-trimethoxysilane (MPTMS), mesoporous with alter-
nating hydrophobic and hydrophilic blocks in the pore wall (HHBs-
B8T2-SH, BTEB/TEOS/ MPTMS = 72/18/10; HHBs-B5T5-SH,
BTEB/TEOS/MPTMS = 45/45/10), mesoporous silicas with hydro-
phobic surface (PMOs-B10T0-SH, BTEB/TEOS/MPTMS = 90/0/
10), and mesoporous silicas with hydrophilic surface (MCM-B0T10-
SH, BTEB/TEOS/ MPTMS = 0/90/10) were prepared. For BxTy-
SH, x/y represents the molar ratio of BTEB and TEOS, and SH refers
to the functionalization with mercaptopropyl group. Typically,
cetyltrimethylammonium bromide (CTAB) was dissolved in sodium
hydroxide aqueous solution, and then the mixture of BTEB, TEOS,
and MPTMS were added to the above solution. The initial molar ratio
was set as Si/CTAB/NaOH/H2O = 1/0.96/2.67/559 for HHBs-
B8T2-SH and PMOs-B10T0-SH, and 1/0.96/1.82/559 for HHBs-
B5T5-SH. The mixture was treated ultrasonically for 20 min, stirred at
room temperature for 12 h, and aged at 90 °C for 24 h. For MCM-
B0T10-SH, tetramethylammonium hydroxide Beilstein (TMAH) need
2567
dx.doi.org/10.1021/cs500385s | ACS Catal. 2014, 4, 2566−2576

ACS Catalysis
Research Article
Preparation of L-Proline-Grafted Mesoporous Silica. The
mercaptopropyl-functionalized mesoporous silica (1 g) was added to a
degassed solution of (2S, 4R)-1-(tert-butoxycarbonyl)-4-(4-vinyl-
benzyloxy)pyrrolidine-2-carboxylic acid (0.52 g, 1.5 mmol) and
AIBN (0.024 g, 0.145 mmol) in 50 mL of toluene. The mixture was
stirred at 110 °C for 15 h under N2 atmosphere. After cooling to
ambient temperature, the solid was filtered and washed with toluene
and CH2Cl2. A light yellow powder was obtained. Finally, the Boc
group was removed by suspending the L-proline-grafted solid in 12 mL
of CH2Cl2/TFA (v/v = 3) for 3 h. The solid was filtered and washed
with saturated NaHCO3, water, and ethanol. The obtained light yellow
powder was dried at 40 °C in vacuum for 24 h.
On the mesoporous support with hydrophobic (PMOs-B10T0-SH),
alternating hydrophobic and hydrophilic (HHBs-B8T2-SH and HHBs-
B5T5-SH), or hydrophilic (MCM-B0T10-SH) surface, trans-4-
hydroxy-L-proline was grafted via the covalent linkage using 4-
chloromethylstyrene as the linker, producing PMOs-B10T0-Pro (2),
HHBs-B8T2-Pro (3), HHBs-B5T5-Pro (4), and MCM-B0T10-Pro
(5). The preparation of the heterogeneous catalysts was elaborated in
Scheme 1. The −OH group passivated HHBs-B8T2-Pro was denoted
as HHBs-B8T2-Pro-CH3 (3′).
product as a colorless solid. The enantiomeric excess of product was
determined by chiral-phase HPLC analysis (column: Chiralpak AD-H;
flow phase: isopropanol/n-hexane (v/v = 5/95); flow rate: 1.0 mL/
min; detection wavelength: 254 nm). The absolute configuration of
aldol products was determined according to refs 39 and 58. The
product analysis by NMR and HPLC are elaborated in the Supporting
Information.
General Procedure for Knoevenagel−Michael Cascade Reaction.
Typically, 0.5 mmol of isatin, 0.5 mmol of malononitrile, 0.5 mmol of
acetone, 5 mL of solvent, and 240 mg of catalyst were oscillated at
room temperature for 10 h. The reaction progress was monitored by
TLC (hexane/ethyl acetate (v/v = 7/3)). The proline loading was
20.0% for (2S, 4R)-1-(tert-butoxycarbonyl)-4-(4-vinylbenzyloxy)
pyrrolidine-2-carboxylic acid, 15.0% for PMOs-B10T0-Pro, 13.0% for
HHBs-B8T2-Pro, 12.0% for HHBs-B5T5-Pro, and 11.2% for MCM-
B0T10-Pro. The resulting organic extracts were dried using Na2SO4.
After removal of catalyst by filtration, conversion was determined by
1H NMR analysis of the crude product. Purification by flash column
chromatography (silica gel, hexane/AcOEt) gave the product as a red
solid. The enantiomeric excess of product was determined by chiral-
phase HPLC analysis (column: Chiralpak AD-H; flow phase:
isopropanol/n-hexane (v/v = 30/70); flow rate: 1.0 mL/min;
detection wavelength: 254 nm). The absolute configuration of the
products was determined according to ref 59. Product analysis by
NMR and HPLC are elaborated in the Supporting Information.
Catalyst Recycling. The catalyst was recycled easily from the
reaction system by centrifugation. The resulting solid was thoroughly
rinsed with ethyl acetate, deionized water, and ethanol for 5 times
separately and dried in vacuum at 40 °C for the recycling catalytic
experiments.
Characterization. Powder X-ray diffraction patterns were obtained
on a Bruker D8 focus X-ray diffractometer with Cu Kα radiation
operated at 30 mA and 45 kV. TEM images were taken on a JEOL
JEM-2010 electron microscope operating at 200 kV. Nitrogen
adsorption/desorption experiments were performed at 77 K on a
Quantachrome Autosorb-1 system. The samples were degassed at 80
°C for 5 h prior to measurements. The specific surface area was
calculated by the standard BET method. The mesopore size
distribution was calculated from the desorption branch using the
Barret-Joyner-Halenda (BJH) method. The microporous surface area
and volume were calculated using the t-plot method. The external
surface area was calculated using the high-resolution αs-plot method.57
FT-IR spectra were recorded using a Nicolet 380 (Thermo)
spectrophotometer in the range of 4000−300 cm−1 with 1 cm−1
resolution with the pristine sample pellets treated in vacuum for 6 h.
Thermogravimetric analysis and differential thermal analysis (TG-
DTA) were carried out on a Pyris Diamond TG/DTA thermal analysis
system by PerkinElmer Instrument. HPLC was carried out using
Varian Prostar 210 HPLC with Prostar 325 UV−vis detector. 1H
NMR spectra of the liquid compounds were recorded on a Bruker
Avance-400 spectrometer running at 400 MHz (Bruker, Bremen,
Germany), in CDCl3 as the solvent. Chemical shifts were reported in
RESULTS AND DISCUSSION
■
Structural Properties of Heterogeneous L-Proline
Grafted Mesoporous Silica Catalyst. The powder X-ray
diffraction pattern of HHBs-B8T2-SH (Figure 1a) shows
1
the δ scale relative to residual CHCl3 (7.26 ppm) for H NMR. Data
for 1H NMR are recorded as follows: chemical shift (δ, ppm),
multiplicity (s, singlet; d, doublet; t, triplet; q, quartet; m, multiplet),
intergration, coupling constant (Hz). The solid-state NMR experi-
ments were carried out at 59.6 and 75.5 MHz for 29Si and 13C,
respectively, on a Bruker Avance-400 M solid-state spectrometer
equipped with a commercial 4 mm MAS NMR probe. The chemical
shifts were determined using δTMS = 0 ppm as a reference. The magic-
angle spinning frequencies were set to 5 kHz for all experiments. 29Si
BD/MAS NMR spectra for the silanol determination were carried out
using an Hpdec pulse sequence, which is one pulse sequence with a
high power proton decoupling. Mass spectra were recorded on a
micromass LCT spectrometer using electrspray (ES+) ionization
techniques.
Figure 1. TEM images (inset: electron diffraction patterns) and wide
and low (inset) angle powder XRD patterns of (a) HHBs-B8T2-SH
and (b) HHBs-B8T2-Pro.
General Procedure for Aldol Reactions. Typically, 83 μL of
cyclohexanone was added to 12 mg catalyst in a microreaction flask.
The proline loading was 10% for (2S, 4R)-1-(tert-butoxycarbonyl)-4-
(4-vinylbenzyloxy) pyrrolidine-2-carboxylic acid, 8.2% for PMOs-
B10T0-Pro, 6.7% for HHBs-B8T2-Pro, 5.9% for HHBs-B5T5-Pro, and
5.5% for MCM-B0T10-Pro. After sufficient wetting, 331 μL of water
was added. Stirred for 5 min, 7.6 mg (0.05 mmol) of p-
nitrobenzaldehyde was added. The reaction mixture was stirred at
25 °C for 24 h, extracted with ethyl acetate (3 × 1 mL). The resulting
organic extracts were dried using Na2SO4. After removal of the catalyst
by filtration, diastereoselectivity and conversion were determined by
1H NMR analysis of the crude aldol product. Purification by flash
column chromatography (silica gel, hexane/AcOEt) gave the aldol
typical 100, 110, and 200 reflections with a spacing of 41.0,
23.7, and 20.8 Å in the low-angle diffraction region (2θ<10°),
which can be assigned to the two-dimensional hexagonal
symmetry (p6mm) lattice. The peaks at d spacing of 7.6, 3.8,
and 2.5 Å in the region 10° < 2θ < 50° can be assigned to a
periodic structure55,60 with a spacing of 7.6 Å existing in the
pore walls. The well-ordered hexagonally symmetric structures
are fully confirmed by the TEM images. As a comparison, well-
defined hexagonally symmetric mesoporous structures can be
also observed from XRD patterns (Figure S1A) and TEM
images (Figure S2, upper images) for PMOs-B10T0-SH,
2568
dx.doi.org/10.1021/cs500385s | ACS Catal. 2014, 4, 2566−2576

ACS Catalysis
Research Article
Table 1. Textural Parameters of the Mercaptopropyl-Functionalized Samples and the L-Proline Grafted Catalysts
a
b
SBET
Smicro
Smeso
Sex
pore volume
(cm3/g)
Vmicro
Vmeso
pore size
(nm)
wall thickness
(nm)
sample
(m2/g)
(m2/g)
(m2/g)
(m2/g)
(cm3/g)
(cm3/g)
PMOs-B10T0-SH
PMOs-B10T0-Pro
PHHBs-B8T2-SH
PHHBs-B8T2-Pro
PHHBs-B5T5-SH
PHHBs-B5T5-Pro
MCM-B0T10-SH
MCM-B0T10-Pro
1330
1303
1506
1486
1239
875
0
6
0
0
3
0
0
0
1310
1280
1482
1465
1205
858
20
17
24
21
31
17
36
16
0.95
0.91
1.12
1.07
1.05
0.86
0.95
0.58
0
0.95
0.88
1.12
1.07
1.03
0.86
0.95
0.58
2.46
2.44
2.45
2.43
2.49
2.50
2.44
2.42
1.05
1.10
1.10
1.16
1.11
1.19
1.18
1.25
0.03
0
0
0.02
0
1257
817
1221
801
0
0
a
b
Pore size at the maximum distribution. Estimated as the difference between the parameter a from XRD patterns and the pore size from N2
sorption.
HHBs-B5T5-SH, and MCM-B0T10-SH. The periodic struc-
tures existing in the pore walls are also clearly observed for
PMOs-B10T0-SH and HHBs-B5T5-SH. All the mercaptoprop-
yl-functionalized samples exhibit typical type IV nitrogen
adsorption−desorption isotherms, with a uniform pore size
distribution at a maximum of 2.46−2.50 nm (Figure S3A). The
detailed textural parameters have been illustrated in Table 1.
Almost no micropores exist in the mercaptopropyl-function-
alized samples.
materials demonstrate two sets of broad signals attributed to
the resonances for organosiloxane (Tm = RSi(OSi)m(OH)3−m
m =1−3, −50 to −80 ppm) and for siloxane (Qn
Si(OSi)n(OH)4−n, n = 2−4, −90 to −120 ppm) environments.
There is no Q linkage observed for PMOs-B10T0-SH. The
presence of the Tm units gives evidence of the existence of C−
Si covalent bonds. The silanol density was quantified by curve
fitting and deconvolution of 29Si BD/MAS NMR signals
according to the reference method.61 The calculated formula is
exhibited as follows:
,
=
Solid-state 29Si and 13C NMR spectroscopies have been
demonstrated to be the most useful for providing chemical
information regarding the condensation of organosiloxane and
siloxane. As shown in the 29Si BD/MAS NMR spectra (Figure
2), HHBs-B8T2-SH, HHBs-B5T5-SH, and MCM-B0T10-SH
silanol density (μmol·g−1) =
W ·M
+
W ·M
∑
∑
Q n
Q n
Tm
Tm
wherein W and M, respectively, represent the peak area
percentage and the molar mass of Qn and Tm (n = 2−4; m = 1−
3). Silanol density was calculated as 4.23, 4.78, 5.12, and 7.19
mmol/g for PMOs-B10T0-SH, HHBs-B8T2-SH, HHBs-B5T5-
SH, and MCM-B0T10-SH, as shown in Table 2. The vacuum
FT-IR spectra (Figure S4) also clearly show the stronger Si−
OH band of MCM-B0T10-SH than that of HHBs-B8T2-SH,
confirming the more silanol quantity for MCM-B0T10-SH than
HHBs-B8T2-SH. In the 13C CP/MAS NMR spectra for PMOs-
B10T0-SH, HHBs-B8T2-SH, and HHBs-B5T5-SH (Figure
S5a, 5c, 5e), the occurrence of a large peak at about 134.7 ppm
along with sidebands (denoted with asterisks) is attributed to
the phenylene group connected to Si.62 For MCM-B0T10-SH
(Figure 3A), the resonance at 25.3 ppm could be assigned to
the 3CH2 attached to the SH moieties,63 whereas the
1
2
resonances of the CH2 and CH2 carbons appear at 4.6 and
46.2 ppm, respectively. No resonance for the 3CH2 attached to
the −SO3H moieties was detected, excluding the possibility of
the transformation from thiol to sulfonic groups by the acid
extraction.
Figure 2. 29Si BD/MAS NMR spectra of (a) PMOs-B10T0-SH, (b)
HHBs-B8T2-SH, (c) HHBs-B5T5-SH, and (d) MCM-B0T10-SH.
Table 2. Surface Properties and Chemical Composition of the Mercaptopropyl-Functionalized Samples and the L-Proline-
Grafted Catalysts
a
b
samples
S content (mmol/g)
−OH (mmol/g)
phenyl/OH (mol/mol)
1.18
CBET
N content (mmol/g)
PMOs-B10T0-SH
PMOs-B10T0-Pro
HHBs-B8T2-SH
HHBs-B8T2-Pro
HHBs-B5T5-SH
HHBs-B5T5-Pro
MCM-B0T10-SH
MCM-B0T10-Pro
0.43
0.43
0.44
0.43
0.42
0.42
0.40
0.40
4.23
4.78
5.12
7.19
0.83
0.65
1.34
1.24
2.21
2.07
5.07
4.88
0
0.34
0
1.02
0.65
0
0.28
0
0.25
0
0.23
a
b
−OH content was calculated from 29Si BD/MAS NMR spectra; CBET parameter was calculated from the vapor adsorption isotherm by the BET
method.
2569
dx.doi.org/10.1021/cs500385s | ACS Catal. 2014, 4, 2566−2576

ACS Catalysis
Research Article
Figure 3. (A) 13C CP/MAS NMR spectra of (a) MCM-B0T10-SH and (b) MCM-B0T10-Pro (* refers to sidebands). (B) TGA curves of (a)
HHBs-B8T2-SH and (b) HHBs-B8T2-Pro.
a
Table 3. Direct Asymmetric Aldol Reaction of Cyclohexanone with Aldehydes in Neat or Medium
b
ee (%)
b
entry
R =
cat.
proline loading (mol %)
time (h)
solvent
yield (%)
TON
anti/syn
anti
syn
c
c
c
c
c
c
,
,
,
,
,
,
f
1
2
3
4
5
6
7
8
9
p-nitrobenz-
1
2
3
4
5
3′
1
2
3
3
4
5
1
2
2
3
3
4
5
3
3′
2
3
3
3
3
3
3
10
72
24
24
24
48
24
72
24
24
48
24
24
72
8
<3
58
42
27
<3
41
<3
96
83
98
48
18
<3
62
>99
98
>99
31
24
54
85
27
21
28
33
93
78
f
8.2
6.7
5.9
5.5
6.7
10
7.0
6.4
3.3
76:24
79:21
76:24
73
89
80
57
f
78
60
f
d
,f
f
6.0
76:24
78
56
H2O
f
f
8.2
6.7
6.7
5.9
5.5
10
H2O
11.7
12.6
14.9
5.7
86:14
87:13
87:13
84:16
78:22
22
72
72
74
25
14
28
38
18
6
H2O
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
H2O
f
f
H2O
H2O
3.7
brine
8.2
8.2
6.7
6.7
5.9
5.5
0.63
6.7
8.0
6.7
6.7
6.7
6.7
6.7
6.3
brine
7.5
12.0
14.9
15.1
3.7
88:12
86:14
90:10
90:10
91:9
83
72
96
96
84
33
95
80
60
49
69
79
75
92
90
62
39
85
84
41
17
81
20
26
5
f
f
f
24
8
brine
brine
24
24
24
24
8
brine
brine
brine
5.0
77:23
90:10
89:21
71:29
80:20
81:19
78:22
84:16
66:34
80:20
d
f
,f
brine
81.8
11.3
2.1
brine
f
8
toluene
toluene
toluene/brine
DMSO
brine
f
8
2.1
e,f
8
2.5
7
8
4.7
28
36
15
m-nitrobenz-
o-nitrobenz-
-CO2Et
48
48
24
13.8
11.6
13.0
brine
g
brine
81
n.d.
a
b
Conditions: aldehyde (0.05 mmol), cyclohexanone (83 μL, 0.8 mmol), and solvent (331 μL). Determined by 1H NMR of the crude product and
c
d
chiral-phase HPLC analysis. The reaction in neat cyclohexanone (414 μL). Conditions: p-nitrobenzaldehyde (0.53 mmol), cyclohexanone (83 μL,
0.8 mmol), and saturated brine (331 μL). Toluene/brine = 1 (v/v). The results were repeated at least twice. Not determined.
e
f
g
As illustrated in Table 2, the phenylene content was
determined as 4.99, 4.88, and 3.33 mmol/g for PMOs-
B10T0-SH, HHBs-B8T2-SH, and HHBs-B5T5-SH. The SH
content is similar for each sample, ranging from 0.40 to 0.45
mmol/g. Combined with the 29Si MAS NMR data, the phenyl/
OH molar ratio was estimated to gradually reduce for PMOs-
B10T0-SH, HHBs-B8T2-SH, HHBs-B5T5-SH, and MCM-
B0T10-SH, confirming the gradually increased hydrophilic
surface. MCM-B0T10-SH exhibits a typical of type IV vapor
adsorption isotherm with a C value in the BET equation of
5.07, displaying distinct hydrophilic surface (Table 2). PMOs-
B10T0-SH gives a vapor adsorption isotherm of type V with a
2570
dx.doi.org/10.1021/cs500385s | ACS Catal. 2014, 4, 2566−2576

ACS Catalysis
Research Article
Figure 4. Variation of yield and ee of product catalyzed by HHBs-B8T2-Pro in the direct asymmetric aldol reaction of cyclohexanone with aldehydes
in brine with (A) reaction time and (B) recycling runs.
C value of 0.83, characteristic of a hydrophobic surface. For
HHBs-B8T2-SH, the vapor adsorption isotherm was observed
as type V, similar to the shape observed for hydrophobic
PMOs-B10T0-SH, although the C value was calculated to be
1.34. That is, HHBs-B8T2-SH is more hydrophilic than PMOs-
B10T0-SH but more hydrophobic than MCM-B0T10-SH.
Taking its pore wall structure into account, HHBs-B8T2-SH
can be considered to have an alternating hydrophobic and
hydrophilic surface. In addition, the wall thickness could be
related with the hydrophobic−hydrophilic nature of mercapto-
propyl-functionalized samples. Along with increasing hydro-
philic properties, the wall thickness presents a slight increase.
MCM-B0T10-SH exhibits a maximum wall thickness of 1.18
nm (Table 2).
After grafting trans-4-hydroxy-L-proline on the mesoporous
support with hydrophobic (PMOs-B10T0-SH), alternating
hydrophobic and hydrophilic (HHBs-B8T2-SH and HHBs-
B5T5-SH), or hydrophilic (MCM-B0T10-SH) surface, as
illustrated in Scheme 1, PMOs-B10T0-Pro (2), HHBs-B8T2-
Pro (3), HHBs-B5T5-Pro (4), and MCM-B0T10-Pro (5) were
produced. The FT-IR spectra of HHBs-B8T2-SH, HHBs-
B8T2-Pro-BOC, and HHBs-B8T2-Pro confirm the grafting
(Figure S6). The appearance of the adsorption bands at 2980
cm−1 assigned to the −CH3 of the L-proline, and the adsorption
bands at 2920 and 2850 cm−1 assigned to the −CH3 of the
BOC in the spectrum of HHBs-B8T2-Pro-BOC confirms the L-
proline grafting. The disappearance of the adsorption bands for
the −CH3 of the BOC after the acid treatment proves the
complete remove of the BOC groups. The covalent linkage of
L-proline is further supported by the emerging signal at ca.
134.7 ppm assigned to phenylene in the 13C CP/MAS NMR
spectrum of MCM-B0T10-Pro, which was absent on MCM-
B0T10-SH as the grafted precursor (Figure 3A). The N content
in the resulting heterogeneous catalyst was calculated from
element analysis as 0.34, 0.28, 0.25, and 0.23 mmol/g, which
can be taken as the content of L-proline moiety in that the N
quantity from template residue in mercaptopropyl-function-
alized supports is detected as low as negligible, as shown in
Table 2. It also indicates that S content was retained the same
value after L-proline grafting (Table 2), revealing no Si−C bond
hydrolysis occurred during the synthesis process. The grafting
effectiveness for L-proline on PMOs-B10T0-SH, HHBs-B8T2-
SH, HHBs-B5T5-SH, and MCM-B0T10-SH was calculated as
79%, 64%, 60% and 58% respectively, presenting a decrease
with the increased hydrophilic surface property. TGA curves for
HHBs-B8T2-SH and HHBs-B8T2-Pro are shown in Figure 3B.
Compared with HHBs-B8T2-SH (37.1%), the weight loss
(42.2%) from 400 to 800 °C for HHBs-B8T2-Pro not only
included the decomposition of the organic species within the
pore walls but also the decomposition of grafted L-proline.
Thus, TGA results give an N content of L-proline of 0.30
mmol/g in HHBs-B8T2-Pro, showing a good accordance with
the element analysis result of 0.28 mmol/g. In addition, after
covalent grafting, the L-proline-grafted solids well preserve the
ordered hexagonally symmetric mesoprous structures and the
periodic structure, as can be seen from the XRD patterns
(Figure 1b and Figure S1B) and TEM images (Figure S2,
bottom images). The detailed textural properties of L-proline-
grafted samples are given in Table 1. The textural structures
hardly change with typical of type IV N2 isotherms and
narrowly distributed pore sizes are well retained (Figure S2B).
The specific area and the pore volume of the L-proline-grafted
samples remarkably reduce, better evidencing the covalent
grafting (Table 1). As illustrated in Table 2, C values from the
vapor adsorption measurements were estimated as 0.65, 1.24,
2.07, and 4.88 for PMOs-B10T0-Pro, HHBs-B8T2-Pro, HHBs-
B5T5-Pro, and MCM-B0T10-Pro, confirming the gradually
increased hydrophilic surface after L-proline grafting, despite
that a slight decrease was observed for each samples compared
with the mercaptopropyl-functionalized precursors.
Catalytic Properties of Heterogeneous L-Proline-
Grafted Mesoporous Silica and Catalyst Reusability.
The catalytic direct asymmetric aldol reaction was performed
first with p-nitrobenzaldehyde and cyclohexanone as catalyzed
substrates. As shown in Table 3, in either neat or aqueous
medium (entries 1−21), all the heterogeneous catalysts
prepared in this work, except MCM-B0T10-Pro in neat, are
able to catalyze the aldol reaction of p-nitrobenzaldehyde and
cyclohexanone that hardly occurred on homogeneous vinyl-
benzene-modified 4-OH-L-proline (1). In the neat environment
(entries 1−5), PMOs-B10T0-Pro (2) gave a 58% yield (a TON
of 7.0) in 24 h, affording 73% ee for anti isomer and 57% ee for
syn isomer. HHBs-B8T2-Pro (3) gave similar TON value (6.4)
to PMOs-B10T0-Pro (2) in the same reaction time, affording
42% yield in 24 h, 89% ee for anti isomer and 78% ee for syn
isomer. HHBs-B5T5-Pro (4) gave 27% yield (a TON of 3.3) in
24 h with 80% ee for anti isomer and 60% ee for syn isomer. No
reaction observed on the distinct hydrophilic MCM-B0T10-Pro
(5). The ee observed with HHBs-B8T2-Pro (3) is comparable
2571
dx.doi.org/10.1021/cs500385s | ACS Catal. 2014, 4, 2566−2576

ACS Catalysis
Research Article
to the observation in the homogeneous catalysis of L-proline in
DMSO31 but with a much lower proline loading.
of PMOs-B10T0-Pro (2) and HHBs-B8T2-Pro (3) was
promoted more significantly in saturated brine than in water.
This is supposed to originate from the additional contribution
of the oil−water interfaces in the channels of 2 and 3. The oil−
water interface was reported previously41,64 to play a critical
role in the reaction acceleration in aqueous medium. In
saturated brine, the oil−water interface is much clearer than
that in water, able to implement the interfacial synergy more
efficiently. Owing to its hydrophobic and hydrophilic
alternating surface, HHBs-B8T2-Pro (3) benefits more from
the interface, demonstrating higher yield in brine than
hydrophobic PMOs-B10T0-Pro (2). However, a rise of the
hydrophilic blocks reduces the catalytic activity. HHBs-B5T5-
Pro (4) afforded lower yield than HHBs-B8T2-Pro (3) in each
case.
The catalytic reaction was then performed in water (entries
7−12). PMOs-B10T0-Pro (2) gave 96% yield (a TON of 11.7)
in 24 h, affording 22% ee for anti isomer and 14% ee for syn
isomer. HHBs-B8T2-Pro (3) gave 83% yield (a TON of 12.6)
in 24 h, affording 72% ee for anti isomer and 28% ee for syn
isomer. Prolonging the reaction time from 24 to 48 h, the yield
increases to 98% without expense of enantioselectivity. HHBs-
B5T5-Pro (4) gave 48% yield (a TON of 5.7) in 24 h with 74%
ee for anti isomer and 18% ee for syn isomer. On MCM-
B0T10-Pro (5), 18% yield (a TON of 3.7) in 24 h, 25% ee for
anti isomer and 6% ee for syn isomer were afforded.
In brine (entries 13−20), both yield and ee on PMOs-
B10T0-Pro (2) were improved. A 62% yield was afforded in 8 h
with 83% ee for anti isomer and 62% ee for syn isomer. When
we prolonged the reaction time to 24 h, 99% yield was
achieved, but the ee reduced to 72% ee for anti isomer and 39%
ee for syn isomer. With HHBs-B8T2-Pro (3), the catalytic
reaction was accomplished in 8 h, affording a yield of 98% and a
TON of 14.9, and 96% ee for anti isomer and 85% ee for syn
isomer (entry 16). Variation of yield and ee of product
catalyzed by HHBs-B8T2-Pro in the direct asymmetric aldol
reaction of cyclohexanone with aldehydes in brine with reaction
time has been investigated, as shown in Figure 4A. Along with
increasing reaction time, the yield exhibits a Langmuir type
increase and the ee value holds at the same value during 24 h.
This catalytic efficiency is even better than that observed for the
catalysis of trans-4-(4-tert-butylphenoxy)-L-proline41 or long-
chain diamine derived proline with trifluoroacetic acid,39 and
the ee values are comparable to that observed in aqueous
medium for the catalysis of trans-4-(4-tert-butylphenoxy)-L-
proline41 or long-chain diamine derived proline with trifluoro-
acetic acid39 or siloxyproline.38 On HHBs-B5T5-Pro (4), 31%
yield was afforded in 24 h with 84% ee for anti isomer and 41%
ee for syn isomer. MCM-B0T10-Pro (5) gave moderate yield
and ee, similar to the observation in water. A further study
demonstrated that, with HHBs-B8T2-Pro (3), the molar ratio
of cyclohexanone to p-nitrobenzaldehyde can be reduced to
1.5/1 without compromising the diastereo- and enantioselec-
tivity (entry 20). A visible direct asymmetric aldol addition has
even been achieved under a catalyst loading of as low as 0.63
mol %, giving 54% yield in 24 h and 95% ee for anti-isomers
and 81% ee for syn-isomers in saturated brine.
In either neat or aqueous medium, HHBs-B8T2-Pro (3)
provides better diastereoselectivity than other heterogeneous
catalysts investigated in this work. It is visible that the catalytic
activity and enantioselectivity rely on the surface properties of
catalyst solids and reaction medium markedly. In neat
environment, either yield or TON increases with rising surface
hydrophobicity. But in aqueous medium, HHBs-B8T2-Pro (3)
shows superior activity to PMOs-B10T0-Pro (2). In each case,
PMOs-B10T0-Pro (2) and HHBs-B8T2-Pro (3) exhibit higher
activity than MCM-B0T10-Pro (5). The superior activity of
hydrophobic or hydrophobic and hydrophilic alternating
catalyst to hydrophilic catalyst can simply be attributed to the
synergistic effects of the hydrophobic blocks in their channels.
In neat condition, the hydrophobic channels of 2 and 3
advanced the access of organic reactants to the catalytic sites,
thereby accelerating the reaction. In aqueous medium, the
mesoporous channels were able to allure from aqueous medium
and subsequently constrain organic reactants, increasing the
reactant population around the catalytic sites. The catalytic rate
In neat or aqueous environment, HHBs-B8T2-Pro (3) and
HHBs-B5T5-Pro (4), both with alternating hydrophobic and
hydrophilic blocks in pore walls, were found to afford higher
enantioselectivity than either hydrophobic catalyst (2) or
hydrophilic catalyst (5). Especially in brine, the ee values on all
heterogeneous catalysts with hydrophobic blocks were visibly
improved in comparison to in neat or in water. The oil−water
interface existing in HHBs-B8T2-Pro (3) facilitated achieving
an ee of as high as 96% in >99% yield. As reported previously,
the oil−water interface could not only accelerate the reaction
but also change the enantioselectivity in either pristine or
derived L-proline-catalyzed aldol reactions.41 When L-proline is
placed at the effective oil−water interface and the reaction
occurs around this interface, good yield and high enantiose-
lectivity could be achieved probably due to the hydrogen bonds
of the free −OH group of the interfacial water with the
reactants and transition state. In the case of hydrophilic catalyst
(5), the effective oil−water interface is difficult to form around
the active sites, leading to not only lower yield but also poor ee.
In brine, the oil/water interfaces are better defined than in
water. To further verify the above analysis of role of the oil−
water interfaces, the reaction medium was tailored as toluene
and DMSO (entries 22−25). In toluene, HHBs-B8T2-Pro (3)
is inferior to PMOs-B10T0-Pro (2) in both yield and ee. In
DMSO, yield and ee were improved but still inferior to that in
neat or in brine (entries 25). But when brine was then
introduced to fabricate the oil−water interfaces (entry 24), the
yield and ee with HHBs-B8T2-Pro (3) was both improved.
HHBs-B8T2-Pro (3) has been extensively applied to the
aldol reactions of m- or o-nitrobenzaldehydes with cyclo-
hexanone that were reported as more difficult to be catalyzed
by proline derivatives.41,48 For the aldol reactions of m-
nitrobenzaldehydes with cyclohexanone in brine (entry 26),
93% yield was afforded in 48 h with 75% ee for anti isomer. For
the aldol reactions of o-nitrobenzaldehydes with cyclohexanone
in brine (entry 27), 78% yield was afforded in 48 h with 92% ee
for anti isomer. HHBs-B8T2-Pro (3) was then applied to the
aldol reactions of ethyl-(2)-oxoacetate with cyclohexanone,
which produces a key intermediate for the synthesis of statine
and its analogues.65 An 81% yield was afforded in 24 h with
90% ee for anti isomer (entry 28).
The catalyst reusability has also been investigated. The
catalyst was easily separated from the reaction system by
centrifugation and recycled well. All of the yield and
enantiselectivity, as well as diastereoselectivity, hold at the
same level in three runs, as shown in Figure 4. Almost no
leaching was observed for HHBs-B8T2-Pro and HHBs-B5T5-
Pro (<1%) after three runs, whereas 1.5% and 1.7% were
2572
dx.doi.org/10.1021/cs500385s | ACS Catal. 2014, 4, 2566−2576

ACS Catalysis
Research Article
observed for PMOs-B10T0-Pro and MCM-B0T10-Pro (Table
S1). The recycled catalysts in three runs have effectively
preserved the long-range ordered mesoporous structures and
the periodic structure (Figure S7).
The catalysis of HHBs-B8T2-Pro (3) was then evaluated
with the Knoevenagel−Michael cascade reaction of isatin,
malononitrile, and acetone with PMOs-B10T0-Pro (2), HHBs-
B5T5-Pro (4) and MCM-B0T10-Pro (5) as control catalysts, as
illustrated in Table 4. The Knoevenagel−Michael cascade
yield and 91% ee. HHBs-B5T5-Pro (4) gave a similar variation
with HHBs-B8T2-Pro (3), despite the lower yield and ee than
HHBs-B8T2-Pro (3). In either THF/brine or toluene/brine,
MCM-B0T10-Pro (5) gave inferior yield and ee to PMOs-
B10T0-Pro (2), HHBs-B8T2-Pro (3), and HHBs-B5T5-Pro
(4). In THF/brine which has no liquid−liquid oil−water
interfaces, PMOs-B10T0-Pro (2) and HHBs-B8T2-Pro (3)
gave similar yield and ee, although in toluene/brine, which has
liquid−liquid oil−water interfaces, HHBs-B8T2-Pro (3) gave
higher yield and better ee than PMOs-B10T0-Pro (2). The
results further indicate that hydrophobic and hydrophilic
alternating surface contributes to the heterogeneous catalysis.
Variation of yield of product and intermediate and ee of
product catalyzed by HHBs-B8T2-Pro in the Knoevenagel−
Michael cascade reaction in toluene/brine with reaction time
has been investigated, as shown in Figure 5A. Along with the
reaction time, the yield of product exhibits an S-type curve, and
the ee value presents a decrease of <2% toward the reaction
time. The yield of the intermediate isatylidine malononitrile
increases in the first 4 h and then slowly decreases to 9% at 10
h. It has been reported that the reaction rate for the
Knoevenagel reaction is much faster than the following Michael
reaction,59 leading to an induction period for the whole cascade
process. The yield and enantiselectivity hold at the same level
in three runs, as shown in Figure 5B. Almost no leaching was
also observed for HHBs-B8T2-Pro.
Possible Heterogeneous Catalytic Mechanism. There
is one common observation on each heterogeneous L-proline
catalyst: a configuration inversion occurred. It was previously
reported that the aldol reaction of cyclohexanone and p-
nitrobenzaldehyde, catalyzed by either L-proline in DMSO31 or
L-proline derivatives in water or neat cyclohexanone,38−41
generally afforded (2S, 1′R) isomer as the major product. But in
our case, the (2R, 1′S) isomer was afforded as the major
product. According to the mechanism reported previ-
ously,30,31,66,67 in the proline-catalyzed homogeneous asym-
metric aldol reaction, aldehyde served to attack the enamine of
ketone. The transition states involving the re-attack on the anti-
enamine (yield 2S, 1′R and 2S, 1′S) were lower in energy than
the transition states for the si-attack on the syn-enamine (yield
2R, 1′S and 2R, 1′R), giving (2S, 1′R) isomer as the major
product. But for the L-proline grafted on the mesoporous
supports for PMOs-B10T0-Pro (2), HHBs-B8T2-Pro (3),
HHBs-B5T5-Pro (4), and MCM-B0T10-Pro (5), the meso-
Table 4. Knoevenagel−Michael Cascade Reaction of Isatin,
a
Malononitrile, and Acetone
proline
loading
b
entry
solvent
catalyst (mol %) TON yield (%) ee (%)
c
c
c
c
c
1
2
THF/brine
THF/brine
THF/brine
THF/brine
THF/brine
1
2
3
4
5
1
2
3
4
5
20.0
15.0
13.0
12.0
11.2
20.0
15.0
13.0
12.0
11.2
6.1
6.9
5.6
4.1
90
85
78
36
93
91
86
84
3
4
5
d
d
d
d
d
6
toluene/brine
toluene/brine
toluene/brine
toluene/brine
toluene/brine
7
4.3
5.4
4.4
3.2
55
64
57
26
82
91
84
63
8
9
10
a
Conditions: isatin (0.5 mmol), malonoitrile 0.5 mmol), acetone (0.5
b
1
mmol), and solvent (5 mL); reaction time: 10 h. Determined by H
NMR of the crude product and chiral-phase HPLC analysis (chiralcel
AD-H). THF/brine = 4/1 (v/v). Toluene/brine = 1 (v/v). The
results were repeated at least twice.
c
d
e
reaction of isatin, malononitrile, and acetone is an effective
approach to produce 3,3′-disubstituted oxindole which has
triggered great synthetic interest.59 To our delight, in THF/
brine, PMOs-B10T0-Pro (2) afforded 90% yield and 93% ee,
and HHBs-B8T2-Pro (3) afforded 85% yield and 91% ee
(Table 4). In toluene/brine, PMOs-B10T0-Pro (2) afforded
55% yield and 82% ee, and HHBs-B8T2-Pro (3) afforded 64%
Figure 5. Variation of yield of product and intermediate and ee of product catalyzed by HHBs-B8T2-Pro in the Knoevenagel−Michael cascade
reaction in toluene/brine with (A) reaction time and (B) recycling runs.
2573
dx.doi.org/10.1021/cs500385s | ACS Catal. 2014, 4, 2566−2576

ACS Catalysis
Research Article
Scheme 2. Mesoporous Channels Obstruct the re-attack of Aldehyde on Anti-Enamine of Ketone in the Channels of (a) PHHBs,
(b) PMOs, and (c) MCM Materials
Scheme 3. Cyclohexanone−Water Interfaces in the Channels of (a) PHHBs-B8T2-Pro (3) and (b) PMOs-B10T0-Pro (2)
porous channels could all hinder the re-attack on anti-enamine,
leading to the inverse configuration, as illustrated in Scheme 2.
In the neat environment, HHBs-B8T2-Pro (3) had more
powerful confinement than PMOs-B10T0-Pro (2) or MCM-
B0T10-Pro (5) because the carboxylic group of L-proline in
HHBs-B8T2-Pro (3) prefers to orientate close to the
hydrophilic blocks, although the benzene ring of the linker
prefers to locate close to the hydrophobic part of the pore walls
(Scheme 2).
To further confirm the synergetic effect between hydro-
phobic and hydrophilic structural counterparts, the passivation
of −OH groups of HHBs-B8T2-Pro (HHBs-B8T2-Pro-CH3)
was performed. HHBs-B8T2-Pro-CH3 was applied to the direct
asymmetric aldol reaction of p-nitrobenzaldehyde and cyclo-
hexanone. In neat, HHBs-B8T2-Pro-CH3 (3′) gave a TON
value of 6.0 (41% yield) in 24 h, and 78% ee for anti isomer and
56% ee for syn isomer (Table 3, entry 6). The passivation of
−OH groups had essentially no adverse effects on the activity
but resulted in slightly lower enantioselectivity. In brine, HHBs-
B8T2-Pro-CH3 (3′) gave a TON value of 11.3 in 24 h (85%
yield), and 80% ee or anti isomer and 20% ee for syn isomer
(Table 3, entry 21). The passivation of −OH groups caused a
decrease in both yield and ee. It supports the viewpoints that
the acceleration of the reaction depends on the hydrophobic
channels in the neat condition while relying on the oil−water
interface in brine, and the enantioselectivity promotion
depends on the oil−water interface both in neat and brine.
The enhancement of ee in saturated brine, achieved on HHBs-
B8T2-Pro (3) and PMOs-B10T0-Pro (2), is supposed to result
from the effective interfaces created in the nanochannels. The
pore wall of L-proline-grafted PMOs and HHBs materials
provided the hydrophobic environment and thus formed the
effective oil−water interface, like the hydrophobic pocket when
aldolase enzymes catalyze the reaction in water.68 As reported
previously,41,69 oil−water interface could not only accelerate
reaction but also change enantioselectivity in either pristine or
derived L-proline-catalyzed aldol reactions. The oil−water
interface is envisioned to play a crucial role in diminishing
the contacts between bulk water and the reaction transition
states. Moreover, the reaction is supposed to be facilitated by
the hydrogen bonding at the oil-in-water interface. Although
the oil−water interfaces could be formed in the channels of
both HHBs-B8T2-Pro (3) and PMOs-B10T0-Pro (2), their
interfacial shapes are different, resulting in a difference in the ee
improvement, as proposed in Scheme 3. In the channels of
HHBs-B8T2-Pro (3), the hydrophobic and hydrophilic
alternating surface leads to intermittent convex oil−water
interfaces around the catalytic sites. The convex interfaces have
no any adverse effect on the confinement. In the channels of
PMOs-B10T0-Pro (2), however, the organic reactants are
located along the hydrophobic pore wall and water at the center
regions of the channels, forming a concave continuous
interface.
As proposed in Scheme 3, when the aldol reaction occurs at
oil−water interfaces, the stability of transition states for (2R,
1′S) and (2S, 1′S) are increased due to the hydrogen bonding
2574
dx.doi.org/10.1021/cs500385s | ACS Catal. 2014, 4, 2566−2576

ACS Catalysis
Research Article
(12) Yang, Q. H.; Han, D. F.; Yang, H. Q.; Li, C. Chem.Asian J.
2008, 3, 1214−1229.
between H2O with aldehyde, thereby promoting the ee of the
isomers. However, in pure water, the oil/water interfaces are
poorer defined, producing cyclohexanone−water miscible
zones around the carboxylic group of L-proline. The miscible
zones are supposed to decrease the stability of the transition
states for (2R, 1′S) and (2S, 1′S) due to the interaction of the
phenyl ring with water.
(13) Yu, C.; He, J. Chem. Commun. 2012, 48, 4933−4940.
(14) Yang, H. Q.; Zhang, L.; Zhong, L.; Yang, Q. H.; Li, C. Angew.
Chem., Int. Ed. 2007, 46, 6861−6865.
(15) Jones, M. D.; Raja, R.; Thomas, J. M.; Johnson, B. F. G.; Lewis,
D. W.; Rouzaud, J.; Harris, K. D. M. Angew. Chem., Int. Ed. 2003, 42,
4326−4331.
(16) Raja, R.; Thomas, J. M.; Jones, M. D.; Johnson, B. F. G.;
Vaughan, D. E. W. J. Am. Chem. Soc. 2003, 125, 14982−14983.
(17) Hsiao, L. H.; Chen, S. Y.; Huang, S. J.; Liu, S. B.; Chen, P. H.;
Chan, J. C.; Cheng, S. Appl. Catal., A 2009, 359, 96−107.
(18) Bae, S. J.; Kim, S. W.; Hyeon, T.; Kim, B. M. Chem. Commun.
2000, 31−32.
CONCLUSION
■
In summary, we have developed a highly efficient heteroge-
neous asymmetric catalyst with alternating hydrophobic and
hydrophilic blocks in its pore wall, which demonstrates surface
synergies in the neat environment and interface acceleration in
aqueous medium, thereby promoting catalytic rate and
enantioselectivity impressively in the direct aldol reaction and
Knoevenagel−Michael cascade reaction. Further studies focus-
ing on the modulation of hydrophobic and hydrophilic period
of the support and the corresponding impact to the catalysis in
more exigent reaction are currently under investigation and will
be reported in due course.
(19) Dalko, P. I.; Moisan, L. Angew. Chem., Int. Ed. 2001, 40, 3726−
3748.
(20) Dondoni, A.; Massi, A. Angew. Chem., Int. Ed. 2008, 47, 4638−
4660.
(21) Eder, U.; Sauer, G.; Wiechert, R. Angew. Chem., Int. Ed. 1971,
10, 496−497.
(22) Hajos, Z. G.; Parrish, D. R. J. Org. Chem. 1974, 39, 1615−1621.
(23) List, B. Tetrahedron. 2002, 58, 5573−5590.
(24) List, B. Acc. Chem. Res. 2004, 37, 548−557.
(25) Notz, W.; Tanaka, F.; Barbas, C. F., III. Acc. Chem. Res. 2004, 37,
580−591.
(26) List, B. J. Am. Chem. Soc. 2000, 122, 9336−9337.
(27) List, B.; Pojarliev, P.; Biller, W. T.; Martin, H. J. J. Am. Chem.
Soc. 2002, 124, 827−833.
(28) List, B.; Pojarliev, P.; Martin, H. J. Org. Lett. 2001, 3, 2423−
2425.
ASSOCIATED CONTENT
* Supporting Information
XRD patterns, TEM images, N2 adsorption−desorption
isotherms, 13C CP/MAS NMR spectra, vacuum FT-IR spectra,
1H NMR data, and HPLC data of the products. This material is
available free of charge via the Internet at http://pubs.acs.org.
■
S
(29) Kumaragurubaran, N.; Juhl, K.; Zhuang, W.; Bøgevig, A.;
Jørgensen, K. A. J. Am. Chem. Soc. 2002, 124, 6254−6255.
(30) List, B.; Lerner, R. A.; Barbas, C. F., III. J. Am. Chem. Soc. 2000,
122, 2395−2396.
AUTHOR INFORMATION
Corresponding Author
*E-mail: hejing@mail.buct.edu.cn. Fax: +86-10-6442 5385 .
Tel.: +86-10-6442 5280.
■
(31) Sakthivel, K.; Notz, W.; Bui, T.; Barbas, C. F., III. J. Am. Chem.
Soc. 2001, 123, 5260−5267.
(32) Notz, W.; List, B. J. Am. Chem. Soc. 2000, 122, 7386−7387.
(33) Northrup, A. B.; MacMillan, D. W. C. J. Am. Chem. Soc. 2002,
124, 6798−6799.
Notes
The authors declare no competing financial interest.
(34) Northrup, A. B.; Mangion, I. K.; Hettche, F.; MacMillan, D. W.
C. Angew. Chem., Int. Ed. 2004, 43, 2152−2154.
(35) Pidathala, C.; Hoang, L.; Vignola, N.; List, B. Angew. Chem., Int.
Ed. 2003, 42, 2785−2788.
ACKNOWLEDGMENTS
■
This work is supported by NSFC, PCSIRT (IRT1205), and
973 Project (2011CBA00504). J.H. particularly appreciates the
financial aid of China National Funds for Distinguished Young
Scientists from the NSFC.
(36) Tang, Z.; Yang, Z. H.; Chen, X. H.; Cun, L. F.; Mi, A. Q.; Jiang,
Y. Z.; Gong, L. Z. J. Am. Chem. Soc. 2005, 127, 9285−9289.
(37) Tang, Z.; Jiang, F.; Yu, L. T.; Cui, X.; Gong, L. Z.; Mi, A. Q.;
Jiang, Y. Z.; Wu, Y. D. J. Am. Chem. Soc. 2003, 125, 5262−5263.
(38) Hayashi, Y.; Sumiya, T.; Takahashi, J.; Gotoh, H.; Urushima, T.;
Shoji, M. Angew. Chem., Int. Ed. 2006, 45, 958−961.
(39) Mase, N.; Nakai, Y.; Ohara, N.; Yoda, H.; Takabe, K.; Tanaka,
F.; Barbas, C. F., III. J. Am. Chem. Soc. 2006, 128, 734−735.
(40) Hayashi, Y.; Aratake, S.; Okano, T.; Takahashi, J.; Sumiya, T.;
Shoji, M. Angew. Chem., Int. Ed. 2006, 45, 5527−5529.
(41) Huang, J.; Zhang, X.; Armstrong, D. W. Angew. Chem. 2007,
119, 9231−9235.
REFERENCES
■
(1) Tanev, P. T.; Chibwe, M.; Pinnavaia, T. J. Nature 1994, 368,
321−323.
(2) Margelefsky, E. L.; Zeidanb, R. K.; Davis, M. E. Chem. Soc. Rev.
2008, 37, 1118−1126.
(3) Motokura, K.; Tada, M.; Iwasawa, Y. J. Am. Chem. Soc. 2007, 129,
9540−9541.
(4) Motokura, K.; Tada, M.; Iwasawa, Y. Angew. Chem., Int. Ed. 2008,
47, 9230−9235.
(42) Gruttadauria, M.; Giacalone, F.; Marculescu, A. M.; Meo, P. L.;
Riela, S.; Noto, R. Eur. J. Org. Chem. 2007, 28, 4688−4698.
(43) Gruttadauria, M.; Giacalone, F.; Noto, R. Adv. Synth. Catal.
2009, 351, 33−57.
(5) Motokura, K.; Tada, M.; Iwasawa, Y. J. Am. Chem. Soc. 2009, 131,
7944−7945.
(6) Trindade, A. F.; Gois, P. M. P; Afonso, C. A. M. Chem. Rev. 2009,
109, 418−514.
́ ́ ́
(44) Doyaguez, E. G.; Calderon, F.; Sanchez, F.; Fernandez-
̈
(7) Xiang, S.; Zhang, Y. L.; Xin, Q.; Li, C. Angew. Chem., Int. Ed.
2002, 41, 821−824.
Mayoralas, A. J. Org. Chem. 2007, 72, 9353−9356.
(45) Gao, J. S.; Liu, J.; Jiang, D. M.; Xiao, B.; Yang, Q. H. J. Mol.
Catal. A: Chem. 2009, 313, 79−87.
(8) Yang, H. Q.; Li, J.; Yang, J.; Liu, Z. M.; Yang, Q. H.; Li, C. Chem.
Commun. 2007, 1086−1088.
(46) Zamboulis, A.; Rahier, N. J.; Gehringer, M.; Cattoen, X.; Niel,
̈
(9) Wan, Y.; Zhao, D. Y. Chem. Rev. 2007, 107, 2821−2860.
G.; Bied, C.; Moreau, J. J. E.; Man, M. W. C. Tetrahedron: Asymmetry
2009, 20, 2880−2885.
(10) Hoffmann, F.; Cornelius, M.; Morell, J.; Froba, M. Angew.
̈
Chem., Int. Ed. 2006, 45, 3216−3251.
(11) Li, C.; Zhang, H. D.; Jiang, D. M.; Yang, Q. H. Chem. Commun.
2007, 547−558.
(47) An, Z.; Zhang, W. H.; Shi, H. M.; He, J. J. Catal. 2006, 241,
319−327.
2575
dx.doi.org/10.1021/cs500385s | ACS Catal. 2014, 4, 2566−2576

ACS Catalysis
Research Article
(48) Gao, J. S.; Liu, J.; Tang, J. T.; Jiang, D. M.; Li, B.; Yang, Q. H.
Chem.Eur. J. 2010, 16, 7852−7858.
(49) Jebors, S.; Enjalbal, C.; Amblard, M.; Mehdi, A.; Subra, G.;
Martinez, J. J. Mater. Chem. 2013, 1, 2921−2925.
(50) Giacalone, F.; Gruttadauria, M.; Marculescu, A. M.; Noto, R.
Tetrahedron Lett. 2007, 48, 255−259.
(51) Gruttadauria, M.; Giacalone, F.; Marculescu, A. M.; Salvo, A. M.
P.; Noto, R. ARKIVOC 2009, 8, 5−15.
(52) Gruttadauria, M.; Giacalone, F.; Marculescu, A. M.; Noto, R.
Adv. Synth. Catal. 2008, 350, 1397−1405.
(53) Gruttadauria, M.; Salvo, A. M. P.; Giacalone, F.; Agrigento, P.;
Noto, R. Eur. J. Org. Chem. 2009, 31, 5437−5444.
(54) Giacalone, F.; Gruttadauria, M.; Agrigento, P.; Campisciano, V.;
Noto, R. Catal. Commun. 2011, 16, 75−80.
(55) Yang, Q. H.; Kapoor, M. P.; Inagaki, S. J. Am. Chem. Soc. 2002,
124, 9694−9695.
(56) Zhao, X. S.; Audsley, F.; Lu, G. Q. J. Phys. Chem. B 1998, 102,
4143−4146.
(57) Kruk, M.; Jaroniec, M.; Sayari, A. Langmuir 1997, 13, 6267−
6273.
(58) Pousse, G.; Cavelier, F. L.; Humphreys, L.; Rouden, J.; Blanchet.
J. Org. Lett. 2010, 12, 3582−3585.
(59) Liu, L.; Wu, D.; Li, X.; Wang, S.; Li, H.; Li, J.; Wang, W. Chem.
Commun. 2012, 48, 1692−1694.
(60) Inagaki, S.; Guan, S. Y.; Ohsuna, T.; Terasaki, O. Nature 2002,
416, 304−307.
́
(61) García, N.; Benito, E.; Guzman, J.; Tiemblo, P.; Morales, V.;
García, R. A. Microporous Mesoporous Mater. 2007, 106, 129−139.
(62) Kapoor, M. P.; Setoyama, N.; Yang, Q.; Ohashi, M.; Inagaki, S.
Langmuir 2005, 21, 443−449.
(63) Yang, Q.; Kapoor, M. P.; Shirokura, N.; Ohashi, M.; Inagaki, S.;
Kondoc, J. N.; Domen, K. J. Mater. Chem. 2005, 15, 666−673.
(64) Jung, Y.; Marcus, R. A. J. Am. Chem. Soc. 2007, 129, 5492−5502.
(65) Tao, J.; Hoffman, R. V. J. Org. Chem. 1997, 62, 6240−6244.
(66) Bahmanyar, S.; Houk, K. N.; Martin, H. J.; List, B. J. Am. Chem.
Soc. 2003, 125, 2475−2479.
(67) Hoang, L.; Bahmanyar, S.; Houk, K. N.; List, B. J. Am. Chem.
Soc. 2003, 125, 16−17.
(68) Berkssel, A.; Groger, H. Asymmetric Organocatalysis; Wiley-
VCH: Weinheim, 2005.
(69) Delaney, J. P.; Brozinski, H. L.; Henderson, L. C. Org. Biomol.
Chem. 2013, 11, 2951−2960.
2576
dx.doi.org/10.1021/cs500385s | ACS Catal. 2014, 4, 2566−2576