Novel synthesis of bifunctional catalysts with different microenvironments

Article abstract of DOI:10.1039/c1cc13825a

Acid-base bifunctional activity governed by -NH₂ group's microenvironment is evident from two different catalysts scrutinized by interchanging the location of -SO₃H/NH₂ groups on periodic mesoporous ethylenesilica. The hydrophobic local environment plays a significant role in one-pot deacetalization/nitroaldol condensation.

Full text of DOI:10.1039/c1cc13825a

View Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 10422–10424
www.rsc.org/chemcomm
COMMUNICATION
Novel synthesis of bifunctional catalysts with different microenvironmentsw
Manickam Sasidharan,*ab Satoru Fujita,a Masataka Ohashi,a Yasutomo Goto,a
Kenichi Nakashimab and Shinji Inagaki*a
Received 27th June 2011, Accepted 4th August 2011
DOI: 10.1039/c1cc13825a
Acid–base bifunctional activity governed by –NH2 group’s
microenvironment is evident from two different catalysts
scrutinized by interchanging the location of –SO3H/NH2 groups
on periodic mesoporous ethylenesilica. The hydrophobic local
environment plays a significant role in one-pot deacetalization/
nitroaldol condensation.
In modern catalysis, designing catalysts with spatially isolated
multiple active sites to accomplish multi-step reaction cascades
mimicking biological systems remains a major challenge. In
heterogeneous catalytic systems, the intrinsic bifunctional activity
of mesoporous catalysts often depends on the relative separation
between the acidic- and basic-sites as demonstrated by several
research groups.1–7 An elegant work describing an acidic frame-
work (–SO3H in the hydrophobic part) and –NH2 protruding the
channels (SiO2– in the hydrophilic part) of periodic mesoporous
benzenesilica was reported by Thiel et al.8 Similarly, what would
be the effect when the positions of these acid/base functionalities
were reversed on the same silica support?
To implement the idea above, periodic mesoporous ethylene-
silica9 (PMES) containing bridging –CQC– was used as a
silica support. Scheme 1 depicts catalyst preparation methods
by path A and path B, both of which involve a combination of
co-condensation and novel post-synthetic modification. Initial
co-condensation of bis(triethoxysilyl)ethylene (BTSE) with
3-aminopropyl triethoxysilane using Pluronic P123 (Mn = 5800)
leads to hybrid PMES–NH2 (path A). Post-synthetic addition
of NaHSO3 across –CQC– of PMES–NH2 results in bifunc-
tional catalyst PMES–NH2–SO3H–A. In path B, the positions
of –NH2/SO3H groups are reversed and using hybrid PMES–SH
obtained by co-condensation of BTSE and 3-mercaptopropyl
trimethoxysilane, the –NH2 group was grafted across –CQC–
through epoxidation followed by amination. The important
salient features of the present method compared to earlier
reports1–7 are that (i) attachment of –NH2/SO3H groups without
any organic spacer, (ii) no protection/deprotection steps, and (iii)
amenable for interchangeability of functional groups between
hydrophobic and hydrophilic parts.
Scheme 1 Strategies for preparation of bifunctional catalysts.
Prior to modification, the structural ordering and physical
characteristics were evaluated using X-ray diffraction (XRD),
transmission electron microscope (TEM) and N2 sorption
analyses. Both PMES–NH2 and PMES–SH after surfactant
extraction exhibit a hexagonal P6mm symmetry (Fig. S1 and S2,
ESIw). The N2 adsorption/desorption isotherms (Fig. S3, path
A, ESIw) are found to be of Type IV in nature and exhibited a
hysteresis loop characteristic of ordered mesoporous materials
with very high surface area and pore volume (Table S4, ESIw).
The addition of NaHSO3 across –CQC– of PMES–NH2 in
an O2 atmosphere10,11 under standardized reaction conditions
(Table S5, ESIw) was confirmed by 13C–CP MAS NMR
and Raman spectroscopies. The presence of a strong signal
at 147 ppm (Fig. 1, path A) corresponds to –CQC– linked to
silicon; whereas the propyl spacer shows resonance signals at
16.2, 25.5, and 35.0 ppm corresponding to different carbon
environments.9,12 The peaks marked by asterisks correspond
to spinning side bands. A band of peaks at about 55–70 ppm is
assigned to ether groups of the EO–PO–EO surfactant residue.
The appearance of a new resonance signal at 45 ppm for
the bisulfite treated PMES–NH2 confirms that –SO3H was
successfully grafted across the hydrophobic –CQC– bond.
A similar observation was also made for introduction of
–SO3H in PMES itself by our group (unpublished results).
a Toyota Central R&D Laboratories Inc., Nagakute, Aichi, 480-1191,
Japan
b Department of Chemistry, Faculty of Science and Engineering,
Saga University, 1 Honjo-machi, Saga 840-8502, Japan
w Electronic supplementary information (ESI) available: Material
synthesis and post-synthetic modification details, XRD, TEM FTIR,
N2 sorption, and catalytic data. See DOI: 10.1039/c1cc13825a
c
10422 Chem. Commun., 2011, 47, 10422–10424
This journal is The Royal Society of Chemistry 2011

View Online
The 29Si MAS NMR spectrum of PMES–NH2–SO3H–A
(Fig. 1, path A) exhibits three signals at ꢀ83.5, ꢀ74.0, and
ꢀ63.0 ppm. They can be assigned as 3T and 2T for silicon
attached with ethylene (83.5 and ꢀ74.0 ppm) and aminopropyl
(ꢀ74.0 and ꢀ63.0 ppm), respectively.12 The broad signal
centered at ꢀ74.0 ppm is the mixture of 30T and 2T for Si
connected with aminopropyl and ethylene moieties, respec-
tively. The immobilization of –SO3H was further confirmed
by the appearance of the –SO3 stretching vibrational band at
1038 cmꢀ1 in Raman spectra.13 Furthermore, the reduction in
the pore-size of PMES–NH2–SO3H–A is very small (B0.4 nm)
as seen from pore-size distribution curves (Fig. S3, path A, ESIw).
The amount of sulfonic acid estimated by potentiometric
acid–base titration was found to be 1.14 mmol gꢀ1; whereas
the elemental analysis showed 1.29 mmol gꢀ1 of amino groups.
As path B involves a new functionalization strategy, the
epoxidation has been thoroughly investigated with respect to
various oxidants, pH, and temperature (Tables S6 and S7,
ESIw). Since the RSi–CQC–SiR moiety is an electron
deficient olefin (due to conjugation with Si), the epoxidation
was performed under mild basic conditions (pH 8.5) using
anhydrous tert-butyl hydroperoxide.14,15 The formation of
epoxide was clearly seen from FTIR spectra (Fig. S8, ESIw).
The epoxy-product was subsequently treated with ammonia to get
the required bifunctional catalyst and characterized by Raman
spectroscopy. PMES–SH exhibits two strong bands at 1560
and 1300 cmꢀ1 (Fig. 2, spectrum A) which can be ascribed to
–CQC– and –C–H stretching vibrations, respectively.9 The weak
band at 646 cmꢀ1 is attributed to S–C stretching vibrations.16
After oxidation with TBHP, new bands at 1211 and 1032 cmꢀ1
corresponding to the epoxide-ring and SO3 stretching vibrations
(spectrum B), respectively, were observed.13,17 After NH3 treat-
ment, the band at 1211 cmꢀ1 vanished due to amination of
epoxide (spectrum C). Although N–H stretching vibration was
swamped in the –C–H stretching region, weak bands at 1226
and 906 cmꢀ1 originating from –N–H in plane bending and
deformation modes were clearly observed.18,19 In addition, the
minor band at 1226 cmꢀ1 is not due to the unreacted epoxide
moiety but indeed due to the N–H group as confirmed by our
parallel study (unpublished results).
The 13C NMR of PMES–SH exhibited characteristic signals
of –CQC– along with signals at 14.7, 26.9, and 28.0 ppm
originating from mercaptopropyl groups (Fig. 1, path B). After
oxidation and amination, new peaks at 38 and 50.5 ppm
attributed to –NH2 and –SO3H groups, respectively, appeared.
The 29Si MAS NMR spectrum (Fig. 1, path B) also exhibited
three signals at ꢀ82.8, ꢀ73.6, and ꢀ64.5 ppm ascribed to
3T and 2T for silicon attached with ethylene (82.8 and ꢀ73.6 ppm)
and mercaptopropyl (ꢀ73.6 and ꢀ64.5 ppm), respectively.
The amount of –SO3H and –NH2 were found to be 1.19 and
1.35 mmol gꢀ1, respectively.
The cooperative catalytic activities in a one-pot reaction
sequence comprising acid-catalyzed deacetalization of dimethyl
acetal 1 to yield the respective aldehyde 2 followed by base-
catalyzed nitroaldol (Henry) reaction to obtain 3 are shown in
Scheme 2 and Table 1. The PMES–SO3H–NH2–A (‘‘A’’ refers
to ‘‘path A’’) converts 1 (entry 1) to the corresponding aldol
product 3 in almost quantitative yields and illustrates the
efficiency of the catalysts. In contrast, the final aldol adduct
3 obtained over PMES–SO3H–NH2–B (‘‘B’’ refers to ‘‘path
B’’) was 76.9% with an appreciable amount of unreacted
benzaldehyde under similar experimental conditions. To gain
further insight, the reaction was performed over various
dimethylacetals. Entries 3–5 exhibit reactivity of various acetals
over PMES–SO3H–NH2–A and the observed selectivity for
the desired nitroolefin 3 is again always higher than 98%;
whereas PMES–SO3H–NH2–B invariably led to lower selectivity
of 3 (entries 6–8) and the sole side-product was aldehydes
formed from acid-catalyzed deprotection of dimethylacetals. It
is noteworthy to mention that, in spite of similar acid/base
contents of PMES–SO3H–NH2–A (SO3H and NH2 = 1.14
and 1.29 mmol gꢀ1, respectively) and PMES–SO3H–NH2–B
(SO3H and NH2 = 1.19 and 1.36 mmol gꢀ1, respectively), they
exhibit rather different catalytic behavior under identical
reaction conditions. Furthermore, after prolonged reaction
Fig. 1 13C MAS NMR (top two) and 29Si MAS NMR (bottom)
Fig. 2 Raman spectra of path B: (A) PMES–SH, (B) PMES–SO3Na–
spectra. The asterisk symbol indicates the spinning side bands.
epoxide, and (C) PMES–NH2–SO3H.
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 10422–10424 10423

View Online
regular ordering of catalytic sites coupled with high hydro-
phobicity. Therefore, the observed difference between these
bifunctional catalysts may be originated from the activities of
–NH2 groups situated at different microenvironments rather
than diffusion limitation. Unlike PMES–SO3H–NH2–A, the
–NH2 groups in PMES–SO3H–NH2–B are conjugated with
two electropositive silicon atoms (Si–C–C(NH2)–Si) and as a
result the basic strength could be relatively lower due to hyper-
conjugation. Furthermore, the presence of benzaldehyde in the
reaction medium and lower rate of amine-catalyzed nitroaldol
condensation of PMES–NH2 in the hydrophobic environments
corroborate our speculation that strength of the amine moiety
could play a crucial role in quantitative one-pot deacetalization
and nitroaldol condensation. Both the catalysts were recycled
four times successively (Table S10, ESIw) suggesting the high
stability of these bifunctional catalysts. The N/S contents by
elemental analysis of spent catalysts were nearly unchanged
confirming the stability of different organic tethering.
Scheme 2 One-pot deacetalization–nitroaldol condensation.
Table 1 Activity of bifunctional catalystsa
Entry
R
Catalysts
Conv. 1 [%]
2 [%]
3 [%]
1
2
3
4
5
6
7
8
H
H
A
B
A
A
A
B
B
B
100
100
100
100
100
100
100
100
100
100
Trace
100
Trace
100
1.7
23.1
1.9
1.0
2.0
21.0
19.5
21.4
10.7
3.4
Trace
100
Trace
100
98.3
76.9
98.1
99.0
98.0
79.0
80.5
78.6
89.3
96.6
Trace
Trace
Trace
Trace
NO2
OCH3
CH3
NO2
OCH3
CH3
H
H
H
H
H
9b
10c
11
12
13
14
B
B
A + PA
A + PTS
PMES–NH2
PMES–SO3H
In conclusion, we have demonstrated novel methods for
preparation of efficient and stable bifunctional catalysts without
any protection/deprotection steps. The present method enables
us to attach –SO3H/NH2 without an organic spacer. The amine
groups located in the hydrophobic environment exhibited lower
activity due to decreased basicity compared with that of the
hydrophilic SiO2 network.
H
a
Reaction conditions: aldehyde dimethyl acetal (2 mmol), CH3NO2
(5 mL), PMES–SO3H–NH2–A (22 mg, 0.025 mmol SO3H,
b
0.028 mmol), temperature 90 1C, reaction time 24 h. Reaction was
carried out for 48 h. Reaction time 65 h. PA = propylamine.
c
PTS = p-toluene sulfonic acid. The mass balance obtained by an
internal standard method (chlorobenzene standard) was in the range
of 99 ꢁ 2% for all the experiments.
Notes and references
times of 48 h and 65 h, the selectivity of 3 is increased from
76.9% to 89.3% and then to 96.6%, respectively. As expected,
no desired aldol product 3 was obtained when either of the
homogeneous analogues of sulfonic acid or amine (entries 11
and 12) were used with PMES–SO3H–NH2–A, as these func-
tionalities apparently neutralize each other. In the absence of
–SO3H (entry 13), the starting material remains unchanged
even after 24 h; whereas the presence of –SO3H (entry 14)
leads to quantitative deprotection of dimethylacetal to the
respective aldehydes.
1 R. K. Zeidan, S.-J. Hwang and M. E. Davis, Angew. Chem., Int.
Ed., 2006, 45, 6332.
2 M. J. Climent, A. Corma, V. Fornes, R. Guil-Lopez and S. Ibora,
Adv. Synth. Catal., 2002, 344, 1090.
3 J. D. Bass, S. L. Anderson and A. Katz, Angew. Chem., Int. Ed.,
2003, 42, 5219.
4 E. L. Margelefsky, A. Bendjeriou, R. K. Zeidan, V. Dufaud and
M. E. Davis, J. Am. Chem. Soc., 2008, 130, 13442.
5 K. Motokura, M. Tada and Y. Iwasawa, J. Am. Chem. Soc., 2009,
131, 7944.
6 S. Shylesh, A. Wagener, A. Seifert, S. Ernst and W. R. Thiel,
Chem.–Eur. J., 2009, 15, 7052.
The conversion of 1 to 2 is always quantitative for both
catalysts A and B; however, they differ catalytically in amine-
catalyzed transformation of 2 to 3. In order to understand the
variable cooperative catalytic behavior of these microstructurally
different PMES–SO3H–NH2–A and PMES–SO3H–NH2–B
materials, we have compared the kinetics of –NH2 (PMES–NH2)
groups in SiO2– hydrophilic (path A) and –CQC– hydrophobic
environments, where the latter was obtained by following
epoxidation/amination over PMES in a separate experiment
for comparison. The nitroaldol condensation with benzaldehyde
was performed over these two amine-functionalized materials
in the absence of –SO3H under identical reaction conditions. For
both the catalysts, the turn-over-frequency (TOF) for desired
aldol product 3 decreased with increasing reaction time due to
decreased concentration of starting materials. However, –NH2 in
the hydrophobic environment exhibited relatively lower TOF
than that located in the hydrophilic environment (Fig. S9, ESIw).
Interestingly, these –NH2 functionalized materials exhibit higher
activity than the MCM-41 based catalysts7 possibly due to
7 Y. Huang, S. Xu and V. S.-Y. Lin, Angew. Chem., Int. Ed., 2011,
50, 661.
8 S. Shylesh, A. Wagener, A. Seifert, S. Ernst and W. R. Thiel,
Angew. Chem., Int. Ed., 2010, 15, 184.
9 T. Asefa, M. J. MacLachlan, N. Coombs and G. A. Ozin, Nature,
1999, 402, 867.
10 M. S. Kharsch, E. M. May and F. R. Mayo, J. Org. Chem., 1938,
3, 175.
11 R. T. E. Schenck and I. Danishefsky, J. Org. Chem., 1951,
16, 1683.
12 Q. Yang, M. P. Kapoor and S. Inagaki, J. Am. Chem. Soc., 2002,
12, 1403.
13 J.-L. Bribes, M. Elboukari and J. Maillols, J. Raman Spectrosc.,
1991, 22, 275.
14 N. N. Schwartz and J. H. Blumbergs, J. Org. Chem., 1964,
29, 1976.
15 R. D. Temple, J. Org. Chem., 1970, 35, 1275.
16 J. Liu, Q. Yang, L. Zhang, D. Jiang, X. Shi, J. Yang, H. Zhong and
C. Li, Adv. Funct. Mater., 2007, 17, 569.
17 S. R. Davis, A. R. Brough and A. Atkinson, J. Non-Cryst. Solids,
2003, 315, 197.
18 Z.-Q. Wen, J. Pharm. Sci., 2007, 96, 2861.
19 S. Pandiarajan, M. Umadevi, V. Sasirekha, R. K. Rajaram and
V. Ramakrishnan, J. Raman Spectrosc., 2005, 36, 950.
c
10424 Chem. Commun., 2011, 47, 10422–10424
This journal is The Royal Society of Chemistry 2011