Tuning cooperativity by controlling the linker length of silica-supported amines in catalysis and CO2 capture

Article abstract of DOI:10.1021/ja305601g

Cooperative interactions between aminoalkylsilanes and silanols on a silica surface can be controlled by varying the length of the alkyl linker attaching the amine to the silica surface from C1 (methyl) to C5 (pentyl). The linker length strongly affects the catalytic cooperativity of amines and silanols in aldol condensations as well as the adsorptive cooperativity for CO₂ capture. The catalytic cooperativity increases with the linker length up to propyl (C3), with longer, more flexible linkers (up to C5) providing no additional benefit or hindrance. Short linkers (C1 and C2) limit the beneficial amine-silanol cooperativity in aldol condensations, resulting in lower catalytic rates than with the C3+ linkers. For the same materials, the adsorptive cooperativity exhibits similar trends for CO₂ capture efficiency.

Full text of DOI:10.1021/ja305601g

Communication
pubs.acs.org/JACS
Tuning Cooperativity by Controlling the Linker Length of Silica-
Supported Amines in Catalysis and CO2 Capture
Nicholas A. Brunelli, Stephanie A. Didas, Krishnan Venkatasubbaiah,† and Christopher W. Jones*
School of Chemical & Biomolecular Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332-0100
S
* Supporting Information
capping of the silanols eliminated the beneficial acid
ABSTRACT: Cooperative interactions between amino-
alkylsilanes and silanols on a silica surface can be
controlled by varying the length of the alkyl linker
attaching the amine to the silica surface from C1 (methyl)
to C5 (pentyl). The linker length strongly affects the
catalytic cooperativity of amines and silanols in aldol
condensations as well as the adsorptive cooperativity for
CO2 capture. The catalytic cooperativity increases with the
linker length up to propyl (C3), with longer, more flexible
linkers (up to C5) providing no additional benefit or
hindrance. Short linkers (C1 and C2) limit the beneficial
amine−silanol cooperativity in aldol condensations,
resulting in lower catalytic rates than with the C3+ linkers.
For the same materials, the adsorptive cooperativity
exhibits similar trends for CO2 capture efficiency.
component and decreased the observed rate of reaction,
confirming the importance of the cooperative interactions, but
significantly modifying the local environment. Subsequent
studies using primary amines have focused on tuning the
strength of the acid site.17,24,25 A consensus is forming that
weaker acid sites result in the greatest enhancement of the
reaction rate in prototypically base-catalyzed coupling reactions
such as the aldol condensation, with many examples utilizing
surface silanols as the weak acid component.21,23,26,27 While the
importance of tuning the acid strength has been demonstrated,
no studies have investigated the impact of the length and
flexibility of otherwise electronically similar amines in their
cooperative interactions with the silanols.
Amine−silanol interactions have also recently been suggested
to be important in the adsorption of CO2.28 Surfaces grafted
sparsely with propylamines so they could easily interact with
silanols effectively adsorbed carbon dioxide. Capping the
silanols decreased the carbon dioxide adsorption. While this
trend was unambiguous, it is unclear whether the cause was
associated with prevention of cooperative amine−silanol
interactions or simply isolation of the amine sites as a result
of the introduction of the bulky trimethylsilyl groups. The latter
issue, isolation, is important because two amines are typically
thought to be required to adsorb CO2 under dry conditions,
resulting in a maximum amine efficiency29,30 of 50% (i.e., 0.5
mol of CO2 adsorbed/mol of amine). With the cooperative
amine−silanol mechanism, an amine efficiency of 100% is
theoretically possible provided that the amines have sufficient
flexibility to interact with adjacent silanols. Since the organo-
silane is relatively expensive, a higher amine efficiency would be
very desirable from a practical perspective.
uning cooperative interactions between two identical or
Tdifferent sites has important implications for many
systems, including ligand−receptor binding,1 catalysis,2−5 and
CO2 adsorption.6−8 The ultimate examples of tuned cooper-
ative interactions involve protease enzymes that typically
employ a trio of catalytic sites (the catalytic triad9), including
acidic, basic, and hydrogen-bonding residues. These highly
evolved sites have complex architectures that facilitate the
precise organization of acid and base groups to activate both
the reacting electrophile and nucleophile in catalytic reactions.
The complex enzyme architecture spatially organizes the
functional groups and maintains a degree of flexibility that
allows the catalytic sites and substrates to adopt a favorable
conformation to promote the cooperative interaction. Slight
modifications of these parameters can have a significant impact
on the catalytic rates and even the mechanistic pathways.2,10,11
Some of these enzyme features can be achieved in simpler
catalysts through the synthesis of bifunctional molecules11−15
and/or via immobilization of the acid and base components on
a support in such a way that quenching is not possible because
of the restricted mobility or weak interactions of the sites.16−19
Past works have colocalized the acid and base sites on organic
molecules such as an amino acids14 and ionic liquids20 or on
polymeric18 or mesoporous silica supports.21 Of particular
importance is the acid−base cooperative catalytic interaction of
amines and silanols that was first demonstrated for a soluble
amine mixed with mesoporous silica.22 The benefit of acid−
base colocalization on solid supports was subsequently
demonstrated through immobilization of primary amines on
mesoporous silica at moderate densities.21,23 Trimethylsilyl
In this work, the effect of the spatial separation of the amines
and silanols was investigated by systematically varying the alkyl
chain length of the organosilane grafted onto a well-defined
SBA-15 mesoporous silica support. Using an array of materials
functionalized with alkyl linkers varying from methyl to pentyl
(Scheme 1), we determined the amine linker length depend-
ence of the reaction rate for the acid−base cooperative aldol
condensation of 4-nitrobenzaldehyde and acetone. The
importance of amine−silanol interactions during CO2 adsorp-
tion was also demonstrated, and structure−function trends for
the two applications have been elucidated.
A description of the synthetic procedures for organosilane
synthesis, surface grafting, material characterization, catalytic
Received: June 11, 2012
Published: August 14, 2012
© 2012 American Chemical Society
13950
dx.doi.org/10.1021/ja305601g | J. Am. Chem. Soc. 2012, 134, 13950−13953

Journal of the American Chemical Society
Communication
Scheme 1. Aminosilanes Grafted onto Mesoporous SBA-15
Silica with Controlled Linker Lengths, Including C1
(Methyl), C2 (Ethyl), C3 (Propyl), C4 (Butyl), and C5
(Pentyl)
testing, and CO2 adsorption testing can be found in the
Supporting Information (SI), but a brief summary is provided
here. Only two of the five organosilanes used were available
commercially. The other three aminosilanes were synthesized
from the appropriate halogenated alkylsilanes using anhydrous
ammonia in a Parr reactor.31 The pentyl-based organosilane
required an initial hydrosilylation of 5-bromo-1-pentene with
triethoxysilane to yield the alkylsilane.32 NMR characterization
confirmed the syntheses of the three relatively uncommon
organosilanes.
The different organosilanes were separately grafted onto
calcined SBA-15 silica under mild conditions. Extensive
characterization using standard techniques, including 13C
cross-polarization magic-angle-spinning (CP-MAS) NMR spec-
troscopy, elemental analysis, and nitrogen physisorption,
confirmed the functionalization (Figures S1−S10 in the SI).
All of the materials were functionalized with low molar loadings
that were approximately equal (∼0.32 mmol g−1) based on
nitrogen content from elemental analysis. Still, slight differences
in the grafting density due to differences in the hydro-
phobicities, reactivities, and polarities of the organosilanes
could be expected. The low loadings allowed efficient acid−
base (silanol−amine) catalytic cooperativity and permitted
direct investigation of the effect of the linker length on the
chemical reactivity.
The materials were tested for pseudo-equilibrium CO2
adsorption capacity using 400 ppm CO2 in He to limit the
amount of physisorbed CO2, thereby probing the behavior of
the amine sites. The moisture-free conditions used to measure
CO2 capacity meant that CO2 was adsorbed in one of two
manners: (1) adsorption utilizing two primary amine groups or
(2) adsorption between an amine and silanols on the surface, as
observed by Weitz and co-workers using differential FTIR
measurements with 2% CO2.28 The relative importance of these
two mechanisms depends on the amine density. In view of the
low amine loadings used, the latter mechanism was surmised to
be more important.
Figure 1. Computed amine efficiencies for adsorption of CO2 (400
ppm in He) on materials functionalized with amines having different
linker lengths.
efficiency to 13% was observed. Adding a third methylene
group to the chain provided the necessary flexibility to produce
a significant increase in the amine efficiency to ∼20%, which is
similar to those for highly functionalized materials reported
previously in ultradilute CO2 capture, where amine−silanol
cooperativity would not be observed.33 No further increase in
the amine efficiency was observed with the additional flexibility
imparted with the butyl or pentyl linkers. If cooperation of
amines with silanols were the dominant CO2 capture
mechanism, an increase in amine efficiency might be expected
with an increase in linker length, as more amines would be able
to interact with silanols. The low amine efficiency suggests that
the dominant mechanism of CO2 adsorption is cooperative
amine−amine adsorption, but additional detailed spectroscopic
studies are needed to confirm or refute this hypothesis.
The silanols were removed by treating the materials with
hexamethyldisilazane (HMDS) to create a second set of
materials that were identical to the first except that the vast
majority of the surface silanols were removed. Without the
silanols, the CO2 adsorption capacity decreased significantly for
materials with the longer linkers (C3 to C5) because the
capping molecule disrupted the cooperative interaction
necessary for efficient CO2 adsorption, resulting in amine
efficiencies of ∼10%.
The materials were also tested for their catalytic efficiency in
the aldol condensation of 4-nitrobenzaldehyde with acetone
(Scheme 2), a standard test reaction used to assess
Scheme 2. Aldol Condensation of 4-Nitrobenzaldehyde and
Acetone
The amine efficiencies for all of the materials are shown in
Figure 1 (numerical values are given in Table S1 in the SI). The
material with the shortest linker (C1) adsorbed a very small
amount of CO2 (0.024 mmol g−1), resulting in an amine
adsorption efficiency of only 8%. The low amount of
adsorption was less than that by the bare support material
(i.e., 0.032 mmol g−1), indicating that the organosilane does not
significantly contribute to the adsorption of CO2 and
additionally blocks potential CO2 physisorption sites. The
low amine efficiency indicates that the low flexibility of the
amine site limits the interaction of the amine with the surface
and that the amine sites are relatively isolated from one
another. Further flexibility was imparted to the amine group by
using the ethyl linker (C2), yet only a slight increase in amine
cooperativity in aminosilica materials.14,19,34−36 The linker
length had a significant impact on the observed reaction rates,
as shown in Figure 2. The observed rate increased with the
number of carbons in the carbon chain up to the propyl chain.
Increasing the linker length beyond propyl resulted in identical
catalytic rates. This is in contrast to the results of previous work
with carboxylic acids and primary amines, where the additional
flexibility associated with longer chains between the functional
groups resulted in decreased activity.34,37 This observation
13951
dx.doi.org/10.1021/ja305601g | J. Am. Chem. Soc. 2012, 134, 13950−13953

Journal of the American Chemical Society
Communication
(acidic activation of the ketone) may not occur to an
appreciable extent, whereas the acidic activation of the aldehyde
may proceed more readily, resulting in a rate that is faster than
that found for the methyl linker but slower than the catalytic
rate for the propyl (and longer) linkers. Distinguishing between
these two (and other possible) explanations will require
additional experimentation.
The acid−base catalytic cooperativity was further confirmed
with the silanol-capped materials. The catalytic rates decreased
considerably when the acidic silanols were removed, as shown
in Figure 3. After silanol removal, the materials with propyl
Figure 2. Conversion of 4-nitrobenzaldehyde in the aldol
condensation with acetone for the catalysts with different alkyl linkers
at 50 °C using a catalyst loading of 10 mol % amine based on CHN
analysis.
suggests that the amines and silanols interact only weakly (most
likely because of the weak acidity of the silanol), whereas
amines and carboxylic acids interact more strongly unless the
relative interaction of the two functional groups is constrained,
as is the case with proline.11
The material with the shortest chain (C1) had the lowest
observed reaction rate. The observed reaction rate with this
catalyst showed an interesting behavior. An initial high apparent
rate of conversion of 4-nitrobenzaldehyde was observed over
the first hour of reaction, after which slow yet gradual
conversion occurred over the remaining reaction time. The
initial 10% conversion of 4-nitrobenzaldehyde corresponds to
the amine loading in the reaction, suggesting a relatively rapid
reaction of the aldehyde with the amine sites on the silica
surface (Scheme S1 in the SI). We surmise that the adsorbed
reactant was not readily converted to the aldol product because
the reacted benzaldehyde did not contain a hydrogen α to the
carbonyl, preventing the aldehyde from forming the enamine
necessary for the reaction to proceed efficiently.38 This
demonstrates that the single methylene linker does not permit
sufficient flexibility of the amine site to allow the reaction to
proceed through a silanol-activated pathway. Instead, the
reaction proceeds through a much slower base-catalyzed
mechanism at a rate consistent with catalysis by a
homogeneous, unsupported primary amine (Figure S11).
The material with a slightly longer chain length (C2) was
found to catalyze the aldol condensation with an efficiency
between those of methyl (slow) and propyl (fast) function-
alized materials. The two most plausible explanations for this
observation are that (1) only limited numbers of surface
silanols are accessible to the short C2 linkers or (2) the
mechanism is modified. The intermediate value of the rate
could reflect an irregular placement of the silanols on the silica
surface, with some amine sites being within the optimal
distance of silanols to enhance the activity and others not. An
alternative explanation is that the ethyl linker causes the
reaction to proceed through a modified mechanism (see
Schemes S2 and S3). Examination of the proposed cooperative
mechanism for a proline catalyst showed that the reaction is
accelerated at two instances as a result of acidic protonation of a
carbonyl, one for the ketone and the other for the aldehyde.14,38
The close proximity of the amine and carboxylic acid in proline
allows both steps to be cooperatively catalyzed. With the
constrained flexibility due to the ethyl linker, the first activation
Figure 3. Conversion of 4-nitrobenzaldehyde in the aldol
condensation with acetone for the catalysts with different alkyl linkers
after HMDS treatment at 50 °C using a catalyst loading of 10 mol %
amine based on CHN analysis. The data for C1 and C3 without
HMDS treatment are included for comparison.
through pentyl linkers had activities comparable to that of the
unmodified methyl-functionalized material, confirming the
important role of acid−base cooperativity for the C3−C5
linkers. This further demonstrates that the unmodified
aminomethyl group was spatially constrained and unable to
catalyze the reaction in a cooperative manner. The methyl- and
ethyl-functionalized materials after HMDS treatment gave even
lower activities than the propyl material treated with HMDS.
The bulky trimethylsilyl groups presumably add steric
constraints that further slow the reaction.
The importance of the length of the aminosilane linker in
two cooperative chemical reactions, catalysis of the aldol
condensation and CO2 adsorption, has been demonstrated
using a systematic array of amine-functionalized silica materials
with different linker lengths. Short linkers limited the
interaction and correspondingly the amine−silanol coopera-
tivity in catalytic and CO2 adsorption applications. For the
longer linkers (C3−C5), the catalytic and adsorptive
cooperativities remained relatively constant, suggesting that
amine−silanol quenching reactions do not occur and that a
propyl linker is sufficiently long to allow for effective
cooperative interactions. These materials should prove highly
valuable in further fundamental studies probing the cooperative
catalytic mechanisms of other acid−base bifunctional reactions
as well as in helping to create more efficient catalysts and
adsorbents. In particular, the methyl linker can be used to
distinguish between cooperative effects with the surface and
outer-sphere effects, as the short linker does not require
modification of the surface to limit cooperative surface
interactions.21
13952
dx.doi.org/10.1021/ja305601g | J. Am. Chem. Soc. 2012, 134, 13950−13953

Journal of the American Chemical Society
Communication
(22) Kubota, Y.; Goto, K.; Miyata, S.; Goto, Y.; Fukushima, Y.; Sugi,
Y. Chem. Lett. 2003, 32, 234.
(23) Kubota, Y.; Yamaguchi, H.; Yamada, T.; Inagaki, S.; Sugi, Y.;
ASSOCIATED CONTENT
■
S
* Supporting Information
Syntheses of organosilanes and mesoporous materials;
mechanistic proposals for catalytic reactions; and materials
characterization, including 13C CP MAS NMR spectra, nitrogen
physisorption isotherms, thermogravimetric analysis curves, and
X-ray diffraction patterns. This material is available free of
charge via the Internet at http://pubs.acs.org.
Tatsumi, T. Top. Catal. 2010, 53, 492.
(24) Brunelli, N. A.; Venkatasubbaiah, K.; Jones, C. W. Chem. Mater.
2012, 24, 2433.
(25) Ji, C.; Peng, Y.; Huang, C.; Wang, N.; Jiang, Y. Synlett 2005, 986.
(26) Xie, Y.; Sharma, K. K.; Anan, A.; Wang, G.; Biradar, A. V.; Asefa,
T. J. Catal. 2009, 265, 131.
(27) Sharma, K. K.; Asefa, T. Angew. Chem., Int. Ed. 2007, 46, 2879.
(28) Danon, A.; Stair, P. C.; Weitz, E. J. Phys. Chem. C 2011, 115,
11540.
(29) Choi, S.; Drese, J. H.; Jones, C. W. ChemSusChem 2009, 2, 796.
(30) Bollini, P.; Didas, S. A.; Jones, C. W. J. Mater. Chem. 2011, 21,
15100.
(31) Schafer, O.; Bauer, A.; Kneissl, A. Aminomethylene functional
siloxanes. U.S. Pat. Appl. 2006/0122413 A1, 2006.
(32) Pichon, B. P.; Man, M. W. C.; Bied, C.; Moreau, J. J. E. J.
Organomet. Chem. 2006, 691, 1126.
(33) Didas, S. A.; Kulkarni, A. R.; Sholl, D. S.; Jones, C. W.
ChemSusChem 2012, DOI: 10.1002/cssc.201200196.
(34) Yu, X.; Yu, X.; Wu, S.; Liu, B.; Liu, H.; Guan, J.; Kan, Q. J. Solid
State Chem. 2011, 184, 289.
AUTHOR INFORMATION
■
Corresponding Author
cjones@chbe.gatech.edu
Present Address
†School of Chemical Sciences, National Institute of Science
Education and Research (NISER), Institute of Physics Campus,
Bhubaneswar-751005, Orissa, India.
Notes
The authors declare no competing financial interest.
(35) Iborra, S.; Climent, M. J.; Corma, A.; Forne, V. Adv. Synth.
Catal. 2002, 344, 1090.
ACKNOWLEDGMENTS
■
The authors acknowledge financial support from the U.S.
Department of Energy, Office of Basic Energy Sciences,
through Catalysis Science Contract DE-FG02-03ER15459.
N.A.B. acknowledges E. W. Ping, W. Xu, and J. Copeland for
meaningful discussions.
(36) Zeidan, R. K.; Hwang, S. J.; Davis, M. E. Angew. Chem., Int. Ed.
2006, 45, 6332.
(37) Margelefsky, E. L. Cooperative Catalysis by Bifunctionalized
Mesoporous Silica. Ph.D. Dissertation, California Institute of
Technology, Pasadena, CA, 2008.
(38) List, B.; Hoang, L.; Martin, H. J. Proc. Natl. Acad. Sci. U.S.A.
2004, 101, 5839.
REFERENCES
■
(1) Wurtz, J.-M.; Bourguet, W.; Renaud, J.-P.; Vivat, V.; Chambon,
P.; Moras, D.; Gronemeyer, H. Nat. Struct. Biol. 1996, 3, 87.
(2) Konsler, R. G.; Karl, J.; Jacobsen, E. N. J. Am. Chem. Soc. 1998,
120, 10780.
(3) Venkatasubbaiah, K.; Zhu, X.; Kays, E.; Hardcastle, K. I.; Jones,
C. W. ACS Catal. 2011, 1, 489.
(4) Lyaskovskyy, V.; Bruin, B. D. ACS Catal. 2012, 2, 270.
(5) Tan, K. L. ACS Catal. 2011, 1, 877.
(6) Caplow, M. J. Am. Chem. Soc. 1968, 90, 6795.
(7) Donaldson, T. L.; Nguyen, Y. N. Ind. Eng. Chem. Fundam. 1980,
19, 260.
(8) Danckwerts, P. V. Chem. Eng. Sci. 1978, 34, 443.
(9) Brady, L.; Brzozowski, A. M.; Derewenda, Z. S.; Dodson, E.;
Dodson, G.; Tolley, S.; Turkenburg, J. P.; Christiansen, L.; Huge-
Jensen, B.; Norskov, L.; Thim, L.; Menge, U. Nature 1990, 343, 767.
(10) Nakayama, K.; Maruoka, K. J. Am. Chem. Soc. 2008, 130, 17666.
(11) Sakthivel, K.; Notz, W.; Bui, T.; Barbas, C. F., III. J. Am. Chem.
Soc. 2001, 123, 5260.
(12) Okino, T.; Hoashi, Y.; Furukawa, T.; Xu, X.; Takemoto, Y. J.
Am. Chem. Soc. 2005, 127, 119.
(13) Okino, T.; Hoashi, Y.; Takemoto, Y. J. Am. Chem. Soc. 2003,
125, 12672.
(14) List, B.; Lerner, R. A.; Barbas, C. F., III. J. Am. Chem. Soc. 2000,
122, 2395.
(15) Mase, N.; Nakai, Y.; Ohara, N.; Yoda, H.; Takabe, K.; Tanaka,
F.; Barbas, C. F., III. J. Am. Chem. Soc. 2006, 128, 734.
(16) Shao, Y.; Guan, J.; Wu, S.; Liu, H.; Liu, B.; Kan, Q. Microporous
Mesoporous Mater. 2010, 128, 120.
(17) Zeidan, R. K.; Davis, M. E. J. Catal. 2007, 247, 379.
(18) Kwong, C. K.-W.; Huang, R.; Zhang, M.; Shi, M.; Toy, P. H.
Chem.Eur. J. 2007, 13, 2369.
(19) Solin, N.; Han, L.; Che, S.; Terasaki, O. Catal. Commun. 2009,
10, 1386.
́
(20) Boronat, M.; Climent, M. J.; Corma, A.; Iborra, S.; Monton, R.;
Sabater, M. Chem.Eur. J. 2010, 16, 1221.
(21) Bass, J. D.; Solovyov, A.; Pascall, A. J.; Katz, A. J. Am. Chem. Soc.
2006, 128, 3737.
13953
dx.doi.org/10.1021/ja305601g | J. Am. Chem. Soc. 2012, 134, 13950−13953