Nanoscale organization of thiol and arylsulfonic acid on silica leads to a highly active and selective bifunctional, heterogeneous catalyst

Article abstract of DOI:10.1021/ja804082m

Ordered mesoporous silicas functionalized with alkylsulfonic acid and thiol group pairs have been shown to catalyze the synthesis of bisphenols from the condensation of phenol and various ketones, with activity and selectivity highly dependent on the distance between the acid and thiol. Here, a new route to thiol/sulfonic acid paired catalysts is reported. A bis-silane precursor molecule containing both a disulfide and a sulfonate ester bond is grafted onto the surface of ordered mesoporous silica, SBA-15, followed by simultaneous disulfide reduction and sulfonate ester hydrolysis. The resulting catalyst, containing organized pairs of arylsulfonic acid and thiol groups, is significantly more active than the alkylsulfonic acid/thiol paired catalyst in the synthesis of bisphenol A and Z, and this increase in activity does not lead to a loss of regioselectivity. The paired catalyst has activity similar to that of a randomly bifunctionalized arylsulfonic acid/thiol catalyst in the bisphenol A reaction but exhibits greater activity and selectivity than the randomly bifunctionalized catalyst in the bisphenol Z reaction.

Full text of DOI:10.1021/ja804082m

Published on Web 09/13/2008
Nanoscale Organization of Thiol and Arylsulfonic Acid on
Silica Leads to a Highly Active and Selective Bifunctional,
Heterogeneous Catalyst
Eric L. Margelefsky,† Anissa Bendje´riou,‡ Ryan K. Zeidan,† Ve´ronique Dufaud,*,‡
and Mark E. Davis*,†
DiVision of Chemistry and Chemical Engineering, California Institute of Technology,
Pasadena, California 91125, and Laboratoire de Chimie, Ecole Normale Supe´rieure de Lyon,
46 allee d’Italie, 69364 Lyon cedex 07, France
Received May 30, 2008; E-mail: mdavis@cheme.caltech.edu; vdufaud@ens-lyon.fr
Abstract: Ordered mesoporous silicas functionalized with alkylsulfonic acid and thiol group pairs have
been shown to catalyze the synthesis of bisphenols from the condensation of phenol and various ketones,
with activity and selectivity highly dependent on the distance between the acid and thiol. Here, a new route
to thiol/sulfonic acid paired catalysts is reported. A bis-silane precursor molecule containing both a disulfide
and a sulfonate ester bond is grafted onto the surface of ordered mesoporous silica, SBA-15, followed by
simultaneous disulfide reduction and sulfonate ester hydrolysis. The resulting catalyst, containing organized
pairs of arylsulfonic acid and thiol groups, is significantly more active than the alkylsulfonic acid/thiol paired
catalyst in the synthesis of bisphenol A and Z, and this increase in activity does not lead to a loss of
regioselectivity. The paired catalyst has activity similar to that of a randomly bifunctionalized arylsulfonic
acid/thiol catalyst in the bisphenol A reaction but exhibits greater activity and selectivity than the randomly
bifunctionalized catalyst in the bisphenol Z reaction.
Introduction
the synthesis of the solid. When the spacer is removed by
deprotection, one is left with the positioned functional group
pair.2g
Organic-inorganic hybrid materials represent an attractive
class of solids that allows for molecular-level fine-tuning of
catalytic materials.1 Of current interest in the design and
synthesis of solid catalysts is the creation of materials that
contain multiple types of active centers.2 These functionalities
may be used to perform several steps in a reaction sequence or
work in a cooperative manner to alter the characteristics (rates,
selectivities) of a single reaction. Logically, cooperative behavior
between functional groups should depend on the control of the
distance between the reactive groups in order to optimize the
catalysis for a particular reaction. We have been developing
positioned monofunctional and bifunctional catalytic materials
where the functional groups are present in a highly ordered
fashion at predefined distances. The synthetic route to these
materials involves the integration of incipient functional groups,
in a protected form, positioned around a central spacer during
One reaction where cooperative, heterogeneous catalysis has
been established is the synthesis of bisphenols. Bisphenols, such
as bisphenol A (BPA) and bisphenol Z (BPZ), are important
industrial feedstocks, especially as monomers in polycarbonate
materials and epoxy resins. They are synthesized by the acid-
catalyzed condensation between a ketone and phenol, yielding
the desired p,p′ isomer and a byproduct, the o,p’ isomer (Scheme
1). The addition of thiols as a cocatalyst is known to improve
both the rate of reaction and the selectivity to the desired
isomer.3,4
Mineral acids can be used to catalyze the bisphenol conden-
sation reaction, but solid acid catalysts such as polymeric ion-
exchange resins are typically used for commercial bisphenol A
production due to their recyclability. Thiols have been either
added as a homogeneous feed additive4 or bound to the resin
surface by ion-pairing.5 Several solid catalysts bearing both acid
and thiol groups covalently attached to a solid support have
been reported. Thiols have been covalently tethered to polymeric
resins containing sulfonic acid groups,6,7 and polysiloxane
catalysts containing randomly organized alkylsulfonic acid and
alkylthiol groups have also been reported to have good catalytic
† California Institute of Technology.
‡ Ecole Normale Supe´rieure de Lyon.
(1) Wight, A. P.; Davis, M. E Chem. ReV. 2002, 102, 3589–3613, and
references therein.
(2) (a) Macquarrie, D. J.; Jackson, D. B.; Tailland, S.; Wilson, K.; Clark,
J. H. Stud. Surf. Sci. Catal. 2000, 129, 275–282. (b) Asefa, T.; Kruk,
M.; MacLachan, M. J.; Coombs, N.; Grondey, H.; Jaroniec, M.; Ozin,
G. A. J. Am. Chem. Soc. 2001, 123, 8520–8530. (c) Margolese, D.;
Melero, J. A.; Christiansen, S. C.; Chmelka, B. F.; Stucky, G. D. Chem.
Mater. 2000, 12, 2448–2459. (d) Hall, S. R.; Fowler, C. E.; Lebeau,
B.; Mann, S. Chem. Commun. 1999, 2, 201–202. (e) Bass, J. D.; Katz,
A. Chem. Mater. 2006, 18, 1611–1620. (f) Bhaumik, A.; Tatsumi, T.
J. Catal. 2000, 189, 31–39. (g) Margelefsky, E. L.; Zeidan, R. K.;
Davis, M. E. Chem. Soc. ReV. 2008, 37, 1118–1126.
(3) Jerabek, K.; Hun, L. G.; Setinek, K. Collect. Czech. Chem. Commun.
1989, 54, 321–325.
(4) Jansen, J. E. U.S. Patent 2,468,982, 1949.
(5) McNutt, B. W.; Gammill, B. B. U.S. Patent 3,394,089, 1968.
(6) Apel, F. N.; Conte, L. B.; Bender, H. L. U.S. Patent 3,153,001, 1964.
(7) Wagner, R. B. U.S. Patent 3,172,916, 1960.
9
13442 J. AM. CHEM. SOC. 2008, 130, 13442–13449
10.1021/ja804082m CCC: $40.75
2008 American Chemical Society

Spatially Organized Arylsulfonic Acid/Thiol Catalysts
A R T I C L E S
Scheme 1. Synthesis of Bisphenol A (Top) and Bisphenol Z (Bottom) from Phenol and Either Acetone or Cyclohexanone
Scheme 2. Synthesis of Previously Reported11 Catalysts SBA-a,
Containing Alkylsulfonic Acid Sites, and SBA-at-p, Containing
Alkylsulfonic Acids Paired with Alkyl Thiols
activity,8 but the thermal stability of polymeric resins is a key
limitation of these materials.
Catalytic materials bearing multiple sulfonic acid functional
groups positioned at varying distances from one another on the
surface of mesoporous solids have been prepared to explore
the effects that the spatial arrangement of active sites have on
catalytic activity and selectivity.9 A series of organosilane
precursors containing either disulfide or sulfonate ester func-
tionalities (synthons of the eventual sulfonic acid groups) were
synthesized. From these molecular precursors, a variety of
organic-inorganic hybrid, mesostructured SBA-15 silica ma-
terials were prepared using a postsynthetic grafting procedure
that led to disulfide- and sulfonate ester-modified silicas, which
were subsequently converted to sulfonic acid-modified materials.
The sulfonic acid-containing SBA-15 silica materials were tested
as catalysts for the condensation reaction of phenol and acetone
to bisphenol A, and positive cooperative effects were demon-
strated, although the origin of the effects was not determined.
We have investigated the addition of thiol to the sulfonic acid-
catalyzed reactions by using homogeneous and SBA-15-
immobilized sulfonic acids.10 Inclusion of homogeneous thiol
catalysts along with either type of sulfonic acid catalyst
accelerated the rate of reaction and shifted the regioselectivity
to favor the formation of the p,p′-bisphenol isomer (bisphenol
A) over the unwanted o,p′-bisphenol isomer. By immobilizing
the thiol on the same solid as the sulfonic acid, reaction rates
and regioselectivities were enhanced even further from those
obtained from using mixtures of these two functional groups
(when they are both homogeneous, or a combination of
homogeneous and heterogeneous). These observations led us
to design SBA-15 catalysts functionalized with discrete alkyl-
sulfonic acid/thiol pairs formed from a sultone intermediate (see
Scheme 2).11 The activity and selectivity of these catalysts were
found to increase markedly as the acid/thiol distance decreased.
Furthermore, the catalyst in which the acid and thiol groups
were separated by three carbon atoms was 3 times more active
than a material containing randomly distributed acid and thiol
groups in the synthesis of bisphenol A and 14 times more active
in the synthesis of bisphenol Z.11 These results highlight how
the nanoscale organization of catalytic surfaces can be tuned to
provide a level of reactivity control unachievable through
traditional random functionalization approaches.
Here, we report a new route to acid/thiol paired catalysts
involving the design of an organosilane precursor which is
grafted to a surface at two points and then cleaved to deprotect
the catalytic site. The precursor, bis-silane 3 containing a
disulfide group and an aryl sulfonate ester separated by two
carbon atoms, was prepared according to Scheme 3. After this
molecule was grafted onto the silica surface, reduction of the
disulfide bond and hydrolysis of the sulfonate ester afforded a
thiol and arylsulfonic acid group, respectively, in close proximity
on the silica surface (Scheme 4). This material was used to
catalyze the condensation between phenol and either acetone
or cyclohexanone, and its performance was compared to those
of the randomly bifunctionalized analogue and the alkylsulfonic
acid/thiol paired catalyst obtained by the sultone route.11
Typically, increasing the strength of an acid catalyst increases
catalytic activity at the expense of selectivity. It was hoped that,
(8) Wieland, S.; Panster, P. Stud. Surf. Sci. Catal. 1997, 108, 67–74.
(9) Dufaud, V.; Davis, M. E. J. Am. Chem. Soc. 2003, 125, 9403–9413.
(10) Zeidan, R. K.; Dufaud, V.; Davis, M. E. J. Catal. 2006, 239, 299–
306.
(11) Margelefsky, E. L.; Zeidan, R. K.; Dufaud, V.; Davis, M. E. J. Am.
Chem. Soc. 2007, 129, 13691–13697.
9
J. AM. CHEM. SOC. VOL. 130, NO. 40, 2008 13443

A R T I C L E S
Margelefsky et al.
Scheme 3. Synthesis of Disulfide Silane 2 and Bis-silane 3
Scheme 4. Synthesis of Paired Acid/Thiol Catalyst SBA-AT-p by Deprotection from Grafted Disulfide/Sulfonate Ester Intermediate SBA-g3
through careful positioning of the acid and thiol groups, higher
activity can be achieved without sacrificing selectivity.
carbon adjacent to the disulfide and the carbon adjacent to the
arylsulfonate ester group.
Synthesis of Organic-Inorganic Hybrid Materials. SBA-15
was functionalized with bis-silane 3 by grafting in refluxing
toluene. The resulting material is denoted SBA-g3. The loading
was deliberately kept very low so that the effect of pairing could
be seen: the loading used was ∼0.2 mmol/g (the same loading
used for the previously reported alkylsulfonic acid/thiol paired
catalysts).11 SBA-1513 was chosen as a support because of its
high surface area, large pore diameter (∼6 nm), and ease of
functionalization. The rigidity of the silica matrix ensures also
that the bound functional groups do not change their positioning
during surface chemical transformations or catalysis.
SBA-g3 was characterized by X-ray diffraction (XRD),
nitrogen adsorption/desorption porosimetry, 13C{1H} and
29Si{1H} cross-polarization/magic-angle spinning (CP/MAS)
NMR, and thermogravimetric analysis (TGA).
The X-ray diffraction pattern of SBA-g3 (see Supporting
Information) exhibits three clear peaks in the 2θ range of
0.6-3°, characteristic of hexagonally ordered mesophases with
a d(100) spacing of 96.4 Å. The structure of the mesoporous
material is thus preserved throughout the grafting process.
Results and Discussion
Synthesis of Molecular Precursor. The synthesis of a precur-
sor containing both a disulfide linkage and a sulfonate group
bearing two terminal trialkoxysilane groups was accomplished
in three steps (Scheme 3). The unsymmetrical disulfide 2-(py-
ridin-2-yldisulfanyl)ethanol, 1, was obtained in 68% yield from
2-mercaptoethanol, HO(CH2)2SH, with 2,2′-dithiodipyridine via
a thiol-disulfide interchange reaction.12 The pyridyl group,
which is a well-known reversible blocking reagent for thiol
functionalities, was then easily removed by reaction with (3-
mercaptopropyl)triethoxysilane, leading to the formation of
2-((3-(triethoxysilyl)propyl)disulfanyl)ethanol, 2, in 60% yield.
The last step involves the esterification of the commercially
available 2-(4-chlorosulfonylphenyl)ethyltrimethoxysilane with
2 in the presence of an excess of triethylamine. The reaction
leads to bis-silane 3 in a relatively low yield (40%), probably
due to the impurities present in the commercial sulfonyl chloride
1
reagent (mainly free sulfonic acid). 3 was characterized by H
and 13C NMR. Note, in particular, in the 13C NMR spectrum
(Figure 1, top) the peaks characteristic of the mercaptoethanol
spacer at 36.6 and 68 ppm that correspond respectively to the
(13) (a) Zhao, D.; Feng, J.; Huo, Q.; Melosh, N.; Fredrickson, G. H.;
Chmelka, B. F.; Stucky, G. D. Science 1998, 279, 548–552. (b) Zhao,
D.; Huo, Q.; Feng, J.; Chmelka, B. F.; Stucky, G. D. J. Am. Chem.
Soc. 1998, 120, 6024–6036. (c) Zhao, D.; Sun, J.; Li, Q.; Stucky, G. D.
Chem. Mater. 2000, 12, 275–279. (d) Kruk, M.; Jaroniec, M. Chem.
Mater. 2000, 12, 1961–1968.
(12) (a) Guo, W.; Pleasants, J.; Rabenstein, D. L. J. Org. Chem. 1990, 55,
373–376. (b) Houk, J.; Whitesides, G. M. J. Am. Chem. Soc. 1987,
109, 6825–6836.
9
13444 J. AM. CHEM. SOC. VOL. 130, NO. 40, 2008

Spatially Organized Arylsulfonic Acid/Thiol Catalysts
A R T I C L E S
Figure 1. 13C NMR spectrum of bis-silane 3 (top) and 13C{1H} CP/MAS spectrum of SBA-g3 (bottom). / indicates solvent peak (DMSO-d6).
Table 1. Catalyst Characterization Data
indicating that most of the grafted silanes are covalently bonded
a
to the surface at two points, as shown in Scheme 4. Because 3
contains two trialkoxysilane groups, it is possible that one end
could graft to the silica surface, leaving the other end free. This
would be evidenced by a T0 peak in the 29Si spectrum at -41
ppm, but since no such peak is seen, it appears that nearly every
precursor molecule is grafted to the surface at both ends.
Quantitative determination of the organic content of SBA-
g3 was performed by thermogravimetric analysis in air. The
TGA profile (see Supporting Information) showed three weight
loss regions. At temperatures up to about 200 °C, weight loss
was observed, accompanied by an endothermic DSC peak,
presumably due to the desorption of water. A second region of
weight loss followed at temperatures ranging from 200 to 600
°C, presumably arising from the decomposition and desorption
of volatile organics. This weight loss was taken as an estimate
of the total amount of the organic. Thus, the functional group
loading for SBA-g3 was found to be ∼0.4-0.5 mmol/g,
depending on how many residual methoxy/ethoxy groups are
left on the surface. Note also that the grafted complex is
thermally stable up to 220 °C. Further weight loss occurred at
temperatures above 600 °C, likely due to the condensation of
silanols of the silica framework.
Generation of Catalytic Solids and Characterization of
Active Sites. Since the two types of functional groups are
introduced in a protected form, they must be deprotected in order
to remove the spacer and generate the active catalytic material.
This sequence requires that the deprotection of one functional
group does not degrade the other. Fortuitously in this case, the
deprotection of the disulfide and sulfonate ester groups of SBA-
g3 could be accomplished in one step using aqueous tris(2-
carboxyethyl)phosphine hydrochloride (TCEP ·HCl). The phos-
phine reduces the disulfide to a thiol,14 and the weakly acidic
aqueous solution hydrolyzes the sulfonate ester to a sulfonic
entry
material
SBET (m2/g) Dpb (nm) H+ c (mmol/g) SHd (mmol/g) SH/H+
1
2
3
4
5
6
SBA-15
SBA-g3
SBA-A
SBA-AT-p
SBA-AT-r
SBA-T
860
230
6.3
5.8
0.20
0.21
0.13
430
6.0
0.19
0.12
0.32
0.90
0.92
a Specific surface area, calculated using the BET method. b Average
pore diameter, calculated from adsorption isotherm using the BJH
method. c Acid loading, measured by ion exchange/titration. d Thiol
loading, measured by reaction with Ellman’s reagent.
Nitrogen adsorption-desorption results (Table 1, Supporting
Information) show a type IV isotherm, typical of mesoporous
solids. The unfunctionalized SBA-15 has a surface area of 860
m2/g. After grafting with the large bis-silane 3, the surface area
decreases to 230 m2/g, and the pore size is reduced from 6.3 to
5.8 nm. While the surface area of SBA-15 would be expected
to decrease after grafting with a large organosilane such as 3,
the nearly 4-fold decrease observed is too large to be attributed
only to the presence of organic surface groups. It is also likely
that some of the pore mouths are blocked by organic species,
rendering those pores inaccessible to adsorbent.
The integrity of the organic fragment after grafting was
confirmed by 13C CP/MAS NMR spectroscopy. The 13C CP/
MAS spectrum of the grafted molecule corresponds to the
solution-phase spectrum of 3, confirming that the molecule
remains intact upon grafting (see Figure 1). In particular,
the presence of the central spacer that determines the distance
between the masked functionalities is observed at 68 ppm as a
resolvable peak in the solid-state NMR spectrum.
The tertiary silicon peaks in the 29Si CP/MAS spectrum
confirm the presence of covalently bonded organic groups
(Figure 2, top). T1, T2, and T3 peaks can be seen, corresponding
to organosilanes grafted to the surface at one, two, and three
points, respectively. Of these three peaks, the largest is T2,
(14) Burns, J. A.; Butler, J. C.; Moran, J.; Whitesides, G. M. J. Org. Chem.
1991, 56, 2648–2650.
9
J. AM. CHEM. SOC. VOL. 130, NO. 40, 2008 13445

A R T I C L E S
Margelefsky et al.
Figure 2. 29Si{1H} CP/MAS spectrum of SBA-g3 (top), showing T1, T2, and T3 peaks corresponding to silanes covalently bonded to the surface at one,
two, and three points, respectively. Following the cleavage of the mercaptoalcohol linker to form SBA-AT-p (bottom), the T3 peak is larger, the T2 peak is
shifted slightly downfield, and the T1 peak is nearly absent.
Figure 3. 13C{1H} CP/MAS NMR spectra of (a) SBA-g3, (b) SBA-AT-p, and (c) SBA-AT-r. For peak assignments, see Supporting Information, Figures
S6 and S8.
acid. The mercaptoalcohol linker is then free to diffuse away
from the acid/thiol site. The resulting deprotected material is
denoted SBA-AT-p (for acid/thiol paired).
(5.8-6.0 nm). The powder XRD pattern of SBA-AT-p (Sup-
porting Information) indicates that the hexagonal ordering is
maintained.
The 13C CP/MAS spectrum of SBA-AT-p is shown in Figure
3b. The peaks corresponding to the mercaptoethanol linker (37
and 68 ppm) are absent, confirming that the deprotection step
is complete. The residual alkoxy peaks are also absent, as these
groups are also hydrolyzed during the aqueous reaction. The
29Si CP/MAS spectrum (Figure 2, bottom) shows an increase
in the T3 signal and a decrease in the T1 signal; thus, the aqueous
deprotection reaction appears to further condense the alkoxysi-
lane moieties with the surface.15
The acid content of SBA-AT-p was measured by ion-
exchange with NaCl, followed by titration of the HCl with
NaOH. Thiol content was measured by reaction with Ellman’s
reagent,16 followed by spectrophotometric quantification of the
liberated 2-nitro-5-mercaptobenzoate. The catalyst characteriza-
tion data are summarized in Table 1. The acid/thiol ratio was
nearly unity, as expected (0.21 mmol/g acid, 0.19 mmol/g thiol).
Catalysts containing only thiols, only arylsulfonic acids, and
randomly distributed arylsulfonic acid and thiol groups were
also synthesized (denoted SBA-T, SBA-A, and SBA-AT-r,
respectively) according to Scheme 5. The acid groups were
generated by sulfonate ester hydrolysis, using the same aqueous
TCEP·HCl treatment employed for SBA-AT-p. The disulfide
silane 2 was used as the source of thiol groups for SBA-AT-r
because a free thiol could attack the sulfonate ester during the
The surface area measurements show that the BET surface
area increased from 230 (SBA-g3) to 430 m2/g (SBA-AT-p)
after the deprotection step (Table 1, entries 2, 4). This is due to
the removal of the mercaptoethanol linker and also likely to
the unblocking of some of the blocked pore mouths. The median
pore diameter did not change significantly during deprotection
(16) Badyal, J. P.; Cameron, A. M.; Cameron, N. R.; Coe, D. M.; Cox, R.;
Davis, B. G.; Oates, L. J.; Oye, G.; Steel, P. G. Tetrahedron Lett.
2001, 42, 8531–8533.
(15) The same phenomenon was noted by Dufaud and Davis.9 for a similar
aqueous TCEP treatment of organic functionalized SBA-15.
9
13446 J. AM. CHEM. SOC. VOL. 130, NO. 40, 2008

Spatially Organized Arylsulfonic Acid/Thiol Catalysts
A R T I C L E S
Scheme 5. Synthesis of SBA-T, SBA-A, and SBA-AT-ra
a (Top) Grafting 3-mercaptopropyltrimethoxysilane (MPTMS) onto SBA-15 generates SBA-T, containing only thiol sites. (Middle) Grafting sulfonate
ester precursor 4 generates intermediate SBA-g4, which is hydrolyzed to SBA-A, containing only acid sites. (Bottom) Grafting both disulfide 2 and sulfonate
ester 4 generates intermediate SBA-g2,4 which is deprotected to form SBA-AT-r, containing randomly distributed acid and thiol sites.
Table 2. Catalytic Data for Bisphenol A Synthesisa
grafting process. 13C CP/MAS NMR spectroscopy indicates that,
in all cases, the cleavage of the protected groups is quantitative
per-site yield
to yield clean thiol- and/or sulfonic acid-functionalized hybrid
entry
catalyst(s)
p,p′
o,p′
total
isomer ratio
materials (see Supporting Information). As expected, the 13C
NMR spectrum of SBA-AT-r is essentially that of SBA-AT-p
(Figure 3c) because the same functional groups are present on
the surface of both materials; the only difference is the spatial
positioning of these groups. It is also worth noting that the thiol/
acid ratio was fixed at unity in this randomly bifunctionalized
material to provide comparison with the organized thiol/acid
solid.
1
2
3
4
5
6
7
8
9
SBA-Tb
SBA-A
SBA-A + SBA-T
pTSA
<0.1
23
28
21
35
106
108
2
<0.1
11
11
13
12
7
6
1.1
5.6
<0.2
34
39
34
47
113
114
3.1
83
2.1
2.6
1.6
pTSA + SBA-T
SBA-AT-r
SBA-AT-p
SBA-ac
2.9
15.2
19.3
1.8
SBA-at-pc
78
14.0
Catalytic Reactions. The above materials were used to
catalyze the synthesis of bisphenols A and Z. The catalytic tests
were carried out under the same reaction conditions as those
reported in ref 11.
The catalytic data for the bisphenol A reaction are sum-
marized in Table 2. As expected, SBA-T was almost completely
inactive due to its lack of acid sites (entry 1). SBA-A exhibited
some catalytic activity, but very low selectivity due to its lack
of thiol sites (entry 2). When a physical mixture of these two
catalysts was employed (entry 3), the activity and selectivity
improved only slightly beyond that for SBA-A alone, because
the acid and thiol groups are spatially isolated and unable to
a Reaction conditions: 0.02 mmol of H+, 6 mmol of acetone, 24
mmol of phenol, 90 °C, 24 h. Per-site yield calculated as mmol of
product/mmol of H+. b 0.02 mmol of SH was used; per-site yield
calculated on the basis of thiol sites instead of acid sites. c Data taken
from ref 11.
interact except at the outer surfaces of the particles. When a
homogeneous acid, p-toluenesulfonic acid (pTSA), was used,
the activity was very close to that of SBA-A, albeit with an
even lower selectivity (entry 4). pTSA and SBA-T used together
(entry 5) gave somewhat improved activity and selectivity, as
the acid was able to enter the silica pores and interact with the
surface-bound thiols. The catalysts containing both acid and thiol
9
J. AM. CHEM. SOC. VOL. 130, NO. 40, 2008 13447

A R T I C L E S
Margelefsky et al.
Table 3. Catalytic Data for Bisphenol Z Synthesisa
further improvement in activity and a dramatic increase in
selectivity that is not seen in the bisphenol A reaction.
per-site yield
entry
catalyst(s)
p,p′
o,p′
total
isomer ratio
Conclusions
10
11
12
13
14
15
16
17
18
19
SBA-Tb
SBA-A
SBA-A + SBA-T
SBA-A + PrSH
pTSA
0
0
0
An ordered mesoporous silica catalyst containing arylsulfonic
acid and alkylthiol groups organized in pairs was synthesized
by grafting a designed bis-silane precursor onto SBA-15,
followed by linker cleavage. In the bisphenol A reaction, the
random and paired arylsulfonic acid/thiol catalysts exhibit
similar catalytic performance, whereas in the slower bisphenol
Z reaction, the paired catalyst outperforms the randomly
distributed catalyst. Compared to catalysts containing weaker
alkylsulfonic acid sites, the arylsulfonic acid catalysts are more
active while maintaining similar selectivities. Thus, it appears
that the common tradeoff between activity and selectivity can
be circumvented by appropriate catalyst design.
8.1
7.9
7.3
3.2
4.5
17
27
0.3
13
3.8
3.6
4.2
2.6
3.0
4.0
2.0
0d
11.9
11.4
11.5
5.8
7.5
21
29
0.3
14
2.1
2.2
1.8
1.2
1.5
4.3
13.6
N/Ad
14.3
pTSA + SBA-T
SBA-AT-r
SBA-AT-p
SBA-ac
SBA-at-pc
0.9
a Reaction conditions: 0.02 mmol of H+, 6 mmol of cyclohexanone,
24 mmol of phenol, 90 °C, 24 h. Per-site yield calculated as mmol of
product/mmol of H+. b 0.02 mmol of SH was used; per-site yield
calculated on the basis of thiol sites instead of acid sites. c Data taken
from ref 11. d o,p′ isomer below detection limit.
The acid/thiol distance in SBA-AT-p was set by the mer-
captoethanol spacer used in the precursor. Other mercaptoal-
cohol linkers could be used, allowing the acid/thiol distance to
be tuned. Replacing the alkyl spacer (which is somewhat
flexible) with a more rigid one might also lead to a greater
uniformity in acid/thiol distance. This fine-tuning of acid/thiol
distance would be expected to be especially important for even
slower bisphenol condensations and could potentially catalyze
the condensation of phenol with less-reactive ketones without
requiring very high reaction temperatures.
sites in close proximity on the same silica support exhibited
very high activity and good selectivity to the desired p,p′ isomer.
Whether they are organized or randomly distributed, the 1:1
thiol/acid materials (entries 6, 7) exhibited very similar activity
(total yield ∼113-114 per acid site), despite the difference in
the spatial positioning of the acid and thiol groups. The paired
catalyst was, however, the more selective (isomer ratio of 19.3
versus 15.2). Compared to the alkylsulfonic acid-containing
catalysts SBA-a and SBA-at-p obtained via the sultone route11
(entries 8, 9), it is clear that the greater acid strength of the
arylsulfonic acid groups leads to greater overall activity without
sacrificing selectivity.
The heterogeneous acid/thiol catalysts were also tested in the
condensation of cyclohexanone with phenol to form bisphenol
Z. This reaction is slower and has been previously shown to be
more highly sensitive to acid/thiol positioning than the bisphenol
A reaction.11 The catalytic results are shown in Table 3.
The yield of bisphenol Z is 3-6 times lower than that of
bisphenol A at the same reaction conditions due to the lower
reactivity of the cyclohexanone reactant. SBA-A gives a per-
site yield of only 11.9 (entry 11), and no improvement is seen
when a physical mixture of SBA-A and SBA-T is used (entry
12). Notably, no cooperativity is seen when a homogeneous
thiol is used in conjunction with SBA-A (entry 13), and only a
minor improvement is seen with the homogeneous acid catalyst
when heterogeneous thiol is added (entries 14, 15). Only when
the acid and thiol groups are both immobilized on the silica
surface does cooperativity become evident. SBA-AT-r (entry
16) has a yield (21) and selectivity (4.3) approximately twice
those of SBA-A. Organizing the acid and thiol sites into pairs
(entry 17) further increases the yield to 29 and imparts a very
high selectivity (13.6). Compared to the alkylsulfonic acid-
containing catalyst SBA-at-p (entry 19), the arylsulfonic acid/
thiol paired catalyst SBA-AT-p is twice as active while retaining
the same high selectivity.
Experimental Section
Materials. Dichloromethane, pentane, and toluene were dried
by distillation over CaH2, CaSO4, and Na, respectively, and stored
over activated 4 Å molecular sieves. All other solvents were
analytical grade and used as received. Tris(2-carboxyethyl)phos-
phine hydrochloride was purchased from Alfa Aesar. Tetraethox-
ysilane (TEOS), 2-mercaptoethanol, and 2-(4-chlorosulfonylphe-
nyl)ethyltrimethoxysilane (50% in dichloromethane) were obtained
from Acros Organics. (3-Mercaptopropyl)triethoxysilane was pur-
chased from Gelest. All other chemicals were purchased from
Aldrich and used as received. All reactions were performed under
an argon atmosphere.
2-(Pyridin-2-yldisulfanyl)ethanol, 1. A solution of 2-mercap-
toethanol (0.87 g, 11.1 mmol) in dry CH2Cl2 (10 mL) was added
dropwise at room temperature to a solution of 2,2′-dithiopyridine
(4.89 g, 22.2 mmol) in CH2Cl2 (20 mL). After complete addition
of the mercaptoethanol, the solution was stirred for 3 h at room
temperature. The initially colorless solution turned yellow due to
the production of pyridine-2-thione. After the solvent was removed
in Vacuo, chromatography on silica gel (8:2 pentane/ethyl acetate)
afforded 2-(pyridin-2-yldisulfanyl)ethanol (1.4 g, 68% yield) as a
yellow oil.
1H NMR, CDCl3, δ 2.71 ppm (t, J ) 5.5 Hz, 2H, C5H4NS2-
CH2CH2OH), 3.59 ppm (q, J ) 5.5 Hz, 2H, SSCH2CH2OH), 5.5
ppm (t, J ) 6.4 Hz, 1H, SSCH2CH2OH), 6.89, 7.33, 8.2 ppm (m,
4H, pyridyl); 13C NMR, CDCl3, δ 42.1 ppm (SSCH2CH2OH), 58.9
ppm (SSCH2CH2OH), 120.9, 121.3, 137.1, 149.2, and 159.2 ppm
(pyridyl carbons).
2-((3-(Triethoxysilyl)propyl)disulfanyl)ethanol (Organosilane
2). A solution of (3-mercaptopropyl)triethoxysilane (1.61 g, 6.74
mmol) in dry CH2Cl2 (10 mL) was added slowly via syringe to a
solution of 1 (1.40 g, 7.49 mmol) in CH2Cl2 (20 mL). The yellow
solution was stirred at room temperature overnight. The crude
product was purified by chromatography on silica gel (6:4 cyclo-
hexane/ethyl acetate) to afford 2 (1.27 g, 60% yield).
These data demonstrate that catalytic cooperativity in the
synthesis of bisphenol Z is much more dependent on acid/thiol
distance than for bisphenol A. Combining a homogeneous acid
with a heterogeneous thiol (or vice versa) does not provide the
necessary proximity for good catalytic activity because the
homogeneous component is dispersed throughout the liquid
phase. Immobilizing both the acid and thiol onto silica fixes
the groups on the two-dimensional surface, leading to greater
proximity and catalytic performance. Finally, organizing the acid
and thiol groups into closely spaced surface pairs leads to a
1H NMR, CDCl3, δ 0.6 ppm (m, 2H, SiCH2CH2), 1.1 ppm (t, J
) 7 Hz, 9H, SiOCH2CH3), 1.6 ppm (m, 2H, SiCH2CH2), 2.6 ppm
(t, J ) 7 Hz, 2H, SSCH2CH2CH2), 2.7 ppm (t, J ) 6 Hz, 2H,
HOCH2CH2SS), 3.4 ppm (br, 1H, OH), 3.7 ppm (m, 8H,
9
13448 J. AM. CHEM. SOC. VOL. 130, NO. 40, 2008

Spatially Organized Arylsulfonic Acid/Thiol Catalysts
A R T I C L E S
SiOCH2CH3 and HOCH2CH2SS); 13C NMR, CDCl3, δ 9.2 ppm
(SiCH2CH2), 17.7 ppm (SiOCH2CH3), 22.4 ppm (SSCH2CH2CH2),
40.9 ppm (SSCH2CH2CH2), 41.5 ppm (HOCH2CH2SS), 58.3 ppm
(SiOCH2CH3), 60.3 ppm (HOCH2CH2SS); HRMS (ESI+) m/z (M
+ Na+) (C11H27SiS2O4) calcd 337.0939, found 337.0942.
2-((3-(Triethoxysilyl)propyl)disulfanyl)ethyl4-(2-(Trimethoxy-
silyl)ethyl)benzenesulfonate (Bis-silane 3). A solution of 2 (1.23
g, 3.93 mmol) and dry triethylamine (1.02 g, 10.1 mmol) in dry
CH2Cl2 (20 mL) was cooled to 0 °C, and 2-(4-chlorosulfonylphe-
nyl)ethyltrimethoxysilane (50% in CH2Cl2, 1.4 mL, 2.8 mmol) was
added dropwise via syringe. The mixture was allowed to warm
slowly to room temperature and stirred overnight. The solvent was
removed in Vacuo, and the crude product was purified by chroma-
tography on tetramethylorthosilicate-passivated silica gel (6:4
CH2Cl2/ethyl acetate) to afford 3 (0.67 g, 40% yield).
When two organosilanes were grafted onto the same SBA-15
material, two separate solutions of organosilane in toluene were
prepared and were added dropwise simultaneously to the SBA-15
suspension.
Surface Disulfide Reduction and Sulfonate Ester Hydro-
lysis. Functionalized silica (0.5 g) containing disulfide and/or
sulfonate ester surface groups was suspended in an aqueous solution
of TCEP·HCl (16 mM, 50 mL) and stirred at 50 °C for 48 h. After
filtration, the solids were washed with water (5 × 100 mL) and
then suspended in 0.5 N HCl for ∼1 h to acidify, followed by
washing with water (5 × 100 mL). The material was then dried
under high vacuum.
Quantification of Acid Sites. Acid-containing silica (40 mg)
and 2 N aqueous NaCl (4 mL) were stirred at room temperature
for 24 h. The solids were filtered off using positive air pressure
filtration and were washed with water (4 × 2 mL). The combined
filtrate was titrated with 0.01 N NaOH using phenol red as indicator.
Quantification of Thiol Sites. The procedure developed by
Badyal et al.16 was followed for quantification of resin-bound thiols
in organic solvents. Thiol-containing silica (5 mg) was suspended
in methanol (4 mL), and 1 mL of a solution containing Ellman’s
reagent (4 mg/mL) and diisopropyl ethylamine (0.05 mL/mL) was
added. After stirring at room temperature for 4 h, the solids were
removed by syringe filtration, and the absorbance of the filtrate
was measured at 412 nm, using an experimentally determined
1H NMR, CD2Cl2, δ 0.7 ppm (m, 2H, SiCH2CH2CH2), 1.0 ppm
(m, 2H, SiCH2CH2), 1.2 ppm (t, J ) 7 Hz, 9H, SiOCH2CH3), 1.8
ppm (m, 2H, SiCH2CH2CH2), 2.7 ppm (t, J ) 7 Hz, 2H,
SiCH2CH2), 2.8 ppm (m, 4H, SO2OCH2CH2SS and SiCH2CH2CH2),
3.6 ppm (s, 9H, SiOCH3), 3.8 ppm (q, J ) 7 Hz, 6H, SiOCH2CH3),
4.3 ppm (t, J ) 7 Hz, 2H, SO2OCH2CH2SS), 7.4 (d, J ) 8.5 Hz,
2H, phenyl proton ꢀ to ethyl chain) and 7.8 ppm (d, J ) 8.5 Hz,
2H, phenyl proton ꢀ to sulfonate ester group); 13C NMR, CD2Cl2,
δ 9.2 ppm (SiCH2CH2CH2), 10.7 ppm (SiCH2CH2), 17.7 ppm
(SiOCH2CH3), 22.5 ppm (SiCH2CH2CH2), 28.8 ppm (SiCH2CH2),
36.6 ppm (SO2OCH2CH2SS), 41.6 ppm (SiCH2CH2CH2), 50.3 ppm
(SiOCH3), 58.2 ppm (SiOCH2CH3), 68.0 ppm (SO2OCH2CH2SS),
127.8 and 128.8 ppm (aromatic carbons), 133.0 ppm (aromatic
carbon attached to ethyl group), 151.4 ppm (aromatic carbon
attached to sulfonate ester); HRMS (ESI+) m/z (M + Na+)
(C22H42Si2S3O9) calcd 625.1427, found 625.1427.
Ethyl 4-(2-(Trimethoxysilyl)ethyl)benzenesulfonate (Orga-
nosilane 4). 2-(4-Chlorosulfonylphenyl)ethyltrimethoxysilane (50%
in dichloromethane, 1.9 mL, 3.8 mmol) was added dropwise through
a septum to a solution of dry ethanol (0.88 g, 19 mmol) and
triethylamine (1.56 g, 15.4 mmol) at 0 °C. After the addition, the
mixture was allowed to warm slowly to room temperature. The
reaction mixture was stirred at 25 °C overnight. After the solvent
was removed in Vacuo, filtration on silica gel (ethyl acetate) afforded
4 (0.5 g, 40% yield).
extinction coefficient of 11 mM-1
.
Catalytic Reaction: Condensation of Phenol and Ketone. An
amount of catalyst corresponding to 0.02 mmol of H+ (∼100-200
mg) was added to a vial and dried under high vacuum at 80 °C
overnight. Phenol (2.2 g, 24 mmol) and ketone (6 mmol) were
added, the vial was sealed under argon, and the contents were stirred
at 90 °C for 24 h. The catalyst was removed by filtration and washed
with acetonitrile to a total filtrate volume of 25 mL, and the products
were quantified by HPLC analysis. Per-site yield was calculated
on the basis of the number of acid sites present, and selectivity
was defined as the ratio of bisphenol isomers (p,p′/o,p′).
When homogeneous catalysts were used, the homogeneous
catalyst was dissolved in ketone and then added to the phenol and
heterogeneous catalyst.
Acknowledgment. This work was supported in part by the
National Science Foundation under Grant No. OISE-0436985 and
the Department of Energy (M.E.D.). E.L.M. also acknowledges
support from a National Science Foundation Graduate Research
Fellowship, and A.B. would like to thank the CNRS for a
postdoctoral grant. The International Relation Department of the
CNRS is also acknowledged by V.D. for specific funding for this
project (PICS 4016).
1H NMR, CDCl3, δ 0.9 ppm (m, 2H, SiCH2CH2), 1.21 ppm (t,
J ) 7 Hz, 3H, CH3CH2O), 2.74 ppm (m, 2H, C6H4CH2), 3.5 ppm
(s, 9H, SiOCH3), 4.0 ppm (q, J ) 7 Hz, 2H, CH3CH2O), 7.33 ppm
(d, J ) 8.2 Hz, 2H, phenyl protons ꢀ to ethyl chain), 7.7 ppm (d,
J ) 8.2 Hz, 2H, phenyl protons ꢀ to sulfonate ester group); 13C
{1H} NMR, CDCl3, δ 10.8 (SiCH2CH2) ppm, 14.6 ppm
(CH3CH2O), 28.7 ppm (C6H4CH2), 50.4 ppm (SiOCH3), 66.8
ppm (CH3CH2O), 127.8 and 128.8 (aromatic carbons), 133.0 ppm
(aromatic carbon attached to ethyl group), 150.8 ppm (aromatic
carbon attached to sulfonate group).
Organic-Functionalized SBA-15 Materials. SBA-15 (1.0 g,
synthesized according to the literature procedure15) was suspended
in dry toluene (40 mL). A solution of organosilane in toluene (10
mL) was added dropwise via syringe. The suspension was stirred
for 2 h at room temperature and then refluxed overnight. After
cooling to room temperature, the solids were filtered, washed several
times with toluene, and dried under high vacuum.
Supporting Information Available: 13C{1H} CP/MAS NMR
spectra of SBA-g4, SBA-A, SBA-g2,4, SBA-AT-r, and SBA-
T; N2 adsorption/desorption isotherms of SBA-15, SBA-g3, and
SBA-AT-p; X-ray diffraction data of SBA-15, SBA-g3, and
SBA-AT-p; TGA data of SBA-g3 and SBA-AT-p. This material
is available free of charge via the Internet at http://pubs.acs.org.
JA804082M
9
J. AM. CHEM. SOC. VOL. 130, NO. 40, 2008 13449