Molecular motion of tethered molecules in bulk and surface-functionalized materials: A comparative study of confinement

Article abstract of DOI:10.1021/ja0556474

Achieving high degrees of molecular confinement in materials is a difficult synthetic challenge that is critical for understanding supramolecular chemistry on solid surfaces and control of host-guest complexation for selective adsorption and heterogeneous

Full text of DOI:10.1021/ja0556474

Published on Web 04/07/2006
Molecular Motion of Tethered Molecules in Bulk and
Surface-Functionalized Materials: A Comparative Study of
Confinement
Jessica L. Defreese,†,§ Son-Jong Hwang,‡ A. Nicholas G. Parra-Vasquez,†,| and
Alexander Katz*,†
Contribution from the Department of Chemical Engineering, UniVersity of California, Berkeley,
Berkeley, California 94720-1462, and the DiVision of Chemistry and Chemical Engineering,
California Institute of Technology, Pasadena, California 91125
Received August 31, 2005; E-mail: katz@cchem.berkeley.edu
Abstract: Achieving high degrees of molecular confinement in materials is a difficult synthetic challenge
that is critical for understanding supramolecular chemistry on solid surfaces and control of host-guest
complexation for selective adsorption and heterogeneous catalysis. In this Article, using 2H MAS NMR
spectroscopy of tethered carbamates as a molecular probe, we systematically investigate the degree of
steric confinement within three types of materials: two-dimensional silica surface, bulk amorphous
microporous silica, and bulk amorphous mesoporous silica. The resulting NMR spectra are described with
a simple two-site hopping model for motion and prove that the bulk silica network severely limits the molecular
mobility of the immobilized carbamate at room temperature to the same degree as surface-functionalized
materials at low-temperatures (∼210 K). Raising the temperature of the bulk materials to 413 K still
demonstrates the effect of confinement, as manifested in significantly longer characteristic times for the
immobilized carbamate relative to surface-functionalized materials at room temperature. The environment
surrounding the carbonyl functionality of the immobilized carbamate is investigated using FT-IR spectroscopy,
which shows the carbonyl stretching band to be equally shifted for all materials to lower wavenumbers
relative to its noninteracting value in carbon tetrachloride solvent. These results suggest that electrostatic
interactions between the carbonyl of the immobilized carbamate and silica surface may play an important
role in confining the immobilized carbamate and nucleating the formation of a pore wall close to the
immobilized carbamate during bulk materials synthesis.
Introduction
Confinement facilitates multiple points of contact between a
guest molecule and the chiral host active site,4 a general
condition that is desired for stereoselective control of chemical
reactivity.16 However, the high degree of molecular confinement
that is typical of biological active sites has heretofore been
difficult to reproduce in synthetic systems.5,20
Molecular confinement is a hallmark of biological catalysts
and molecular receptors.1-5 The degree of confinement in
protein active sites facilitates stereoselective molecular recogni-
tion and catalysis.4,6,7 The beneficial effect of confinement on
stereoselective transformations has also been demonstrated in
synthetic catalysts,8-14 photocatalysts,15 and binding sites.16-19
The molecular motion of an entity occluded within a host
active site can be used as a local probe of environment that
correlates inversely with degree of molecular confinement. The
motion of occluded molecules within bulk crystalline materials
† University of California, Berkeley.
‡ California Institute of Technology.
§ Current address: Process Research and Development, Bristol-Myers
Squibb Co., New Brunswick, NJ 08903-0191.
| Current address: Chemical and Biomolecular Engineering Department,
Rice University, Houston, TX 77005.
(12) Raja, R.; Thomas, J. M.; Jones, M. D.; Johnson, B. F. G.; Vaughan, D. E.
W. J. Am. Chem. Soc. 2003, 125, 14982-14983.
(1) Richards, F. M. Annu. ReV. Biophys. Bioeng. 1977, 6, 151-176.
(2) Klapper, M. H. Prog. Bioorg. Chem. 1973, 2, 55-133.
(3) Klapper, M. H. Biochim. Biophys. Acta 1971, 229, 557-566.
(4) Sundaresan, V.; Abrol, R. Protein Sci. 2002, 11, 1330-1339.
(5) Mecozzi, S.; Rebek, J. Chem.-Eur. J. 1998, 4, 1016-1022.
(6) Bergmann, M.; Fruton, J. S. J. Biol. Chem. 1937, 117, 189-202.
(7) Fersht, A. Structure and Mechanism in Protein Science: A Guide to Enzyme
Catalysis and Protein Folding; W. H. Freeman: New York, 1999.
(8) Corma, A.; Iglesias, M.; del Pino, C.; Sanchez, F. J. Chem. Soc., Chem.
Commun. 1991, 1253-1255.
(13) Hultman, H. M.; de Lang, M.; Arends, I. W. C. E.; Hanefeld, U.; Sheldon,
R. A.; Maschmeyer, T. J. Catal. 2003, 217, 275-283.
(14) 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.
(15) Sivaguru, J.; Natarajan, A.; Kaanumalle, L. S.; Shailaja, J.; Uppili, S.; Joy,
A.; Ramamurthy, V. Acc. Chem. Res. 2003, 36, 509-521.
(16) Scarso, A.; Shivanyuk, A.; Hayashida, O.; Rebek, J. J. Am. Chem. Soc.
2003, 125, 6239-6243.
(17) Fiedler, D.; Leung, D. H.; Bergman, R. G.; Raymond, K. N. J. Am. Chem.
Soc. 2004, 126, 3674-3675.
(9) Corma, A.; Iglesias, M.; del Pino, C.; Sanchez, F. J. Organomet. Chem.
1992, 431, 233-246.
(18) Leung, D. H.; Fiedler, D.; Bergman, R. G.; Raymond, K. N. Angew. Chem.,
Int. Ed. 2004, 43, 963-966.
(10) Brunet, E. Chirality 2002, 14, 135-143.
(11) Johnson, B. F. G.; Raynor, S. A.; Shepard, D. S.; Mashmeyer, T.; Thomas,
J. M.; Sankar, G.; Bromley, S.; Oldroyd, R.; Gladden, L.; Mantle, M. D.
Chem. Commun. 1999, 1167-1168.
(19) Hart, B. R.; Rush, D. J.; Shea, K. J. J. Am. Chem. Soc. 2000, 122, 460-
465.
(20) Shivanyuk, A.; Scarso, A.; Rebek, J. Chem. Commun. 2003, 1230-1231.
9
10.1021/ja0556474 CCC: $33.50 © 2006 American Chemical Society
J. AM. CHEM. SOC. 2006, 128, 5687-5694
5687

A R T I C L E S
Defreese et al.
Scheme 1. Immobilization of 1 in Three-Dimensional Microporous
Silicate Network
has been previously studied in clathrates and three-dimensional
2
zeolitic pores using H NMR spectroscopy.21,22 Most of these
molecules contain a significant amount of molecular motion,
albeit less than an isotropic liquid.23 The relatively rare exception
to this, in which the occluded molecule is rigid as in a static
situation, has been observed when there are multiple specific
interactions, such as via hydrogen bonding, of the occluded
molecule with the surrounding framework.24,25
Several previous studies of species tethered to a pre-existing
two-dimensional silica surface have utilized NMR spectroscopy
for measuring the conformational mobility of long-chain
aliphatic molecules. In studies on various surface-grafted
branched alkoxysilanes,26,27 as well as long-chain n-alkyl
silanes,28-32 several mobility trends have emerged. Carbons in
close proximity to the silica surface demonstrate more rigidity
and confinement relative to carbons further away from the silica
surface,28 and shorter alkyl chains are more disordered than
longer ones.29 Yet the degree of molecular motion of the bound
end in these systems is still relatively high in comparison to a
static situation at temperatures above 200 K.26,29,33
Our goal is to study the steric confinement of isolated
molecules that are tethered to the interior surface of bulk
microporous and mesoporous amorphous silicate material
networks. Examples of the synthesis of these bulk materials are
shown in Scheme 1 (microporous) and Scheme 2 (mesoporous).
Our study, in which the tethered species is located within a three-
dimensional network, is different from previous studies of
molecular motion of tethered species in porous amorphous
silicates, which have relied either upon tethering a molecule
onto a preexisting two-dimensional surface as illustrated in
Scheme 326,28,29,34 or upon synthesis of bulk organosilane
homopolymers.27,35 The change in pore size that results from
grafting a small molecule onto a surface is small when compared
to the pore diameter of preexisting silica supports (60 Å for
Selecto 60 and macroporous for Aeroperl 300). For this reason,
small molecules postsynthetically tethered to a silica support
exist on a two-dimensional surface rather than in any type of
three-dimensional confinement.
contact between the molecule and the network at the point of
tethering, as well as additional specific interactions with the
surface, which necessitate silica nucleation in close proximity
(distances less than 3.5 Å) to the occluded molecule. The
mesoporous silica synthesis used here has been carefully
designed to synthesize hierarchical porosity in which a mi-
croporous silicate cage is formed around the tethered molecule,
while synthesizing mesoporosity in the remainder of the
network. If the mesoporous synthesis has been successfully
accomplished, the degree of confinement in this material will
be similar to bulk microporous material, which contains no
mesoporosity.36
A critical question that we wish to address is the difference
in restricting molecular motion, or equivalently increasing degree
of confinement, when using a two-dimensional versus three-
dimensional surface for tethering. A second question is whether
it is possible to achieve the same degree of steric confinement
by synthesizing a bulk mesoporous as compared to a bulk
microporous silica network. Here, we investigate molecular
motion around the chiral carbon of immobilized molecule 1
using solid-state 2H NMR spectroscopy. The rotational mobility
about the C-O axis of molecule 1 involving the chiral carbon
is targeted because the end group dynamics are considered to
be a good probe for the geometrical confinement of the
immediate surroundings of the immobilized carbamate, and 2H
NMR spectroscopy is well suited for this purpose. For enabling
this study, we have labeled the chiral carbon with deuterium.
The nonisotopically enriched isomer of 1 has been used
previously for synthesizing bulk, microporous silica via an
The fundamental difference of the concurrent bulk synthesis
is that it allows the formation of a silicate cage around the
occluded molecule, so as to make a locally “tight fit” between
it and the amorphous material network during the process of
silica condensation. The local “tightness of fit” surrounding the
occluded molecule is driven by the requirement for covalent
(21) Shantz, D. F.; Lobo, R. F. Top. Catal. 1999, 9, 1-11.
(22) Komori, Y.; Hayashi, S. Microporous Mesoporous Mater. 2004, 68, 111-
118.
(23) Shantz, D. F.; Lobo, R. F. J. Phys. Chem. B 1998, 102, 2339-2349.
(24) Behrens, P.; van de Goor, G.; Freyhardt, C. C. Angew. Chem., Int. Ed.
Engl. 1995, 34, 2680-2682.
(25) van de Goor, G.; Freyhardt, C. C.; Behrens, P. Z. Anorg. Allg. Chem. 1995,
621, 311-322.
(26) Kelusky, E. C.; Fyfe, C. A. J. Am. Chem. Soc. 1986, 108, 1746-1749.
(27) Kang, H.-J.; Blum, F. D. J. Phys. Chem. 1991, 95, 9391-9396.
(28) Sindorf, D. W.; Maciel, G. E. J. Am. Chem. Soc. 1983, 105, 1848-1851.
(29) Neumann-Singh, S.; Villanueva-Garibay, J.; Muller, K. J. Phys. Chem. B
2004, 108, 1906-1917.
(30) Cheng, J.; Fone, M.; Ellsworth, M. W. Solid State Nucl. Magn. Reson.
1996, 7, 135-140.
(31) Pursch, M.; Brindle, R.; Ellwanger, A.; Sander, L. C.; Bell, C. M.; Handel,
H.; Albert, K. Solid State Nucl. Magn. Reson. 1997, 9, 191-201.
(32) Bayer, E.; Paulus, A.; Peters, B.; Laupp, G.; Reiners, J.; Albert, K. J.
Chromatogr. 1986, 364, 25-37.
(33) Zeigler, R. C.; Maciel, G. E. J. Am. Chem. Soc. 1991, 113, 6349-6358.
(34) Qiao, Y.; Schonhoff, M.; Findenegg, G. H. Langmuir 2003, 19, 6160-
6167.
(35) Blum, F. D.; Meesiri, W.; Kang, H.-J.; Gambogi, J. E. J. Adhes. Sci.
Technol. 1991, 5, 479-496.
(36) Ini, S.; Defreese, J. L.; Parra-Vasquez, N.; Katz, A. Mater. Res. Soc. Symp.
Proc. 2002, 723, 2.3.1-2.3.7.
9
5688 J. AM. CHEM. SOC. VOL. 128, NO. 17, 2006

Molecular Motion of Tethered Molecules
A R T I C L E S
Scheme 2. Immobilization of 1 in Three-Dimensional Mesoporous
Silicate Network
Scheme 3. Immobilization of 1 in Two-Dimensional Mesoporous
Silicate Network
CH2Si); 43.4 (CH2CH2CH2Si); 58.4 (Si(OCH2CH3)3); 126.0 (Ar-C);
127.6 (Ar-C); 128.4 (Ar-C); 142.3 (Cq); 156.0 (CdO). Mass spectrum
(FAB+): m/z 371.2121 (C18H31DNO5Si, 371.2112).
imprinting process, and the resulting imprinted material has
recently demonstrated shape-selective adsorption into the im-
printed binding pockets.37
Bulk Microporous Materials Synthesis. The bulk microporous
material was synthesized according to published procedures.37,39
Bulk Mesoporous Materials Synthesis. The mesoporous material
was synthesized by dissolving 1.80 mmol of 1 (0.668 g) and 89.6 mmol
of TEOS (20 mL) in 63 mL of absolute ethanol. The solution was
brought to reflux, and the following aliquots were added at 1 h
intervals: 0.5 mL of pH 2.0 p-toluenesulfonic acid, 0.5 mL of pH 2.0
p-toluenesulfonic acid, 3.9 mL of water, and 3.9 mL of water. The
solution was refluxed for 1 h after the last water addition. After 10
min of cooling, the solution was added to 4.4 mL of aqueous ammonium
hydroxide (pH 11.7) under vigorous stirring. After the material formed
an optically transparent gel, it was aged for 24 h at room temperature
and then dried in an oven at 40 °C for at least 10 days. The silica
monoliths were ground to <10 µm particles and Soxhlet extracted for
24 h in anhydrous acetonitrile refluxing over calcium hydride.
Surface-Grafted Mesoporous Materials Synthesis. The surface-
grafted materials were prepared by dissolving 0.1 mmol of 1 in 20 mL
of glacial acetic acid. To the solution was added 2 g of Selecto 60 or
Aeroperl 300 silica. The mixtures were stirred at 75 °C overnight. The
silica was then filtered and Soxhlet extracted in anhydrous benzene
refluxing over calcium hydride.
Experimental Section
Synthesis of 3-Triethoxysilylpropyl-carbamic Acid 1-Phenyl-1-
deutero-ethyl Ester (1). The molecule was prepared from the parent
alcohol by standard coupling procedures.38 To a solution of 1,1′-
carbonyldiimidazole (6.71 g, 41.4 mmol) and a catalytic amount of
sodium ethoxide in anhydrous THF (60 mL) was added neat 1-phe-
nylethan-1-d1-ol (5.0 g, 40.6 mmol). The solution was stirred at room
temperature for 12 h prior to the addition of 3-(aminopropyl)-
triethoxysilane (10.0 mL, 42.7 mmol). The solution was then stirred
for 3 h at room temperature. Following solvent removal, excess
imidazole was precipitated with hexane and removed by filtration. The
crude product was purified by column chromatography (Silica Gel 60,
2/1 v/v hexane/ethyl acetate) to yield a clear oil (11.53 g, yield 77%).
1H NMR 400 MHz (CDCl3) δ (ppm): 0.618-0.659 (2H, m, CH2-
CH2CH2Si(OCH2CH3)3); 1.241 (9H, t, J ) 7.0 Hz, Si(OCH2CH3)3);
1.534 (3H, s, PhCD(CH3)O); 1.595-1.670 (2H, m, CH2CH2CH2Si-
(OCH2CH3)3); 3.114-3.207 (2H, m, CH2CH2CH2Si(OCH2CH3)3); 3.832
(6H, q, J ) 7.0 Hz, Si(OCH2CH3)3); 5.145 (1H, bs, NH); 7.235-7.565
(5H, m, Ar-H). 13C{1H} NMR 400 MHz (CDCl3) δ (ppm): 7.6 (CH2-
CH2CH2Si); 18.3 (Si(OCH2CH3)3); 22.4 (PhCD(CH3)O); 23.3 (CH2CH2-
Materials Capping and Deprotection Procedures. The framework
silanol groups on each of the materials were capped with trimethylsilyl
(37) Defreese, J. L.; Katz, A. Microporous Mesoporous Mater. 2006, 89, 25-
32.
(38) Staab, H. A. Angew. Chem., Int. Ed. Engl. 1962, 1, 351-367.
(39) Defreese, J. L.; Katz, A. Chem. Mater. 2005, 17, 6503-6506.
9
J. AM. CHEM. SOC. VOL. 128, NO. 17, 2006 5689

A R T I C L E S
Defreese et al.
(TMS) groups according to published procedures.37,39,40 To confirm site-
isolation, carbamates in surface-grafted material 4 were subjected to
deprotection via published procedures,37 which synthesized primary
amines for use in catalysis experiments via cleavage of the carbamate
bond in the immobilized 1.
The mesoporous materials synthesis was carefully designed
to create a material with hierarchical porosity, possessing
microporous active sites to accomplish confinement of 1 and a
mesoporous framework for facilitating mass transport. The
synthetic strategy is based on a two-step acid-base sol-gel
hydrolysis44-46 that avoids the use of surfactants47 or block
copolymers,48 which could potentially cause local phase separa-
tion and aggregation of tethered species within hydrophobic
domains of a micellar structure. The initial acid-catalyzed step
synthesizes small oligomers of silica that condense around
isolated species of hydrolyzed 1,44,45 with low water content
contributing to maintenance of conditions favoring random
copolymerization of the silica precursor and 1.49 This acid step
that synthesizes a silicate shell around immobilized 1 is followed
by a base-catalyzed hydrolysis that synthesizes bulk mesopo-
rosity during condensation. The resulting glass is optically clear
with no visible phase separation, as opaqueness can be indicative
of aggregation processes within the bulk solid.50 The material
also displays the desired bimodal pore size distribution, contain-
ing both active site microporosity and bulk network mesopo-
rosity.36,51 A base-catalyzed probe reaction consisting of the
Knoevenagel condensation of isophthalaldehyde and malono-
nitrile52 was used to confirm the improved mass transport
characteristics of the bulk mesoporous material over its mi-
croporous analogue. The mesoporous material possessed an
observed turnover frequency 30-fold higher than that measured
for the analogous microporous material. The lack of mass
transport limitations in the mesoporous materials has enabled
studies of outer-sphere acidity effects on catalysis with com-
parisons between bulk materials and conventional surface-
functionalized materials.40
In bulk amorphous silicas utilizing an occluded carbamate
similar to 1, site isolation was demonstrated via chemisorption
of a pyrene fluorescence probe to tethered primary amines
resulting from carbamate deprotection. These sites showed
primarily monomer pyrene emission signature in comparison
to majority excimer in conventional aminopropylsilane-modified
silica.51 We have further shown that the degree of site isolation
as ascertained by pyrene fluorescence correlates to shape-
selectivity of a sequential base-catalyzed probe reaction that we
have used previously for characterizing imprinted materials.40,52
This was performed by showing that reaction of isophthalal-
dehyde and malononitrile produces two products: a monosub-
stituted product and a disubstituted product, in which both
aldehydes have reacted. Conducting the reaction using a site-
isolated catalyst inhibits the immediate formation of disubstituted
relative to monosubstituted product, whereas a clustered amine
catalyst produces both disubstituted and monosubstituted prod-
ucts immediately.36 Thus, initial reaction selectivity is a sensitive
indicator of the degree of local aggregation of the catalytic amine
active sites, and, for this reason, catalysis of this reaction was
used in the current study as a probe to verify site isolation of
FT-IR Sample Preparation. Prior to sample preparation, the
potassium bromide was dried at 300 °C under vacuum for at least 12
h. The silica materials were dried at 120 °C under vacuum for at least
12 h. Sample pellets were prepared using a hydraulic press, with 3-10
mg of silica material in approximately 200 mg of potassium bromide.
The FT-IR spectra were recorded on a N2-purged instrument, with each
spectrum the average of 128 interferograms and covering a spectral
range from 4000 to 400 cm-1. For each of the materials, both the
protected (1 intact) and the deprotected materials were examined.
Because of the low abundance of 1 within the silica material, each
deprotected material was subtracted from the protected material to
selectively observe the characteristic bands due to immobilized car-
bamate.
Results and Discussion
Materials Synthesis and Characterization. Our materials
synthesis procedures are carefully designed to ensure isolation
of the immobilized carbamate molecules. Previous studies of
molecular motions of surface-tethered complexes have shown
that high surface coverages influence the mobility of alkyl chains
on a silica surface,26-30,33,34,41 because the resulting mobility is
modified by a combination of intermolecular interactions
between tethered species and steric effects of the sur-
face.27,31,33,41,42 We are interested in isolating immobilized 1 to
minimize the effect of intermolecular interactions between
adjacent tethered molecules and ensure that we are investigating
the level of confinement that arises due to the rigid silica
network itself. In the surface-functionalized materials, isolation
of immobilized 1 is facilitated primarily through statistical
isolation by grafting to a preexisting silica surface at low surface
coverage. Although we cannot completely rule out some degree
of imprint-imprint intermolecular contact in the materials, we
have no evidence to indicate the presence of imprint-imprint
interactions in our materials. Catalysis data on materials
consisting of primary amines synthesized via deprotection of
materials used in this study demonstrate that all of our materials
lack the type of intersite aggregation observed in conventional
amine-on-silica materials (vide infra).
In the bulk amorphous materials, 1 is immobilized concur-
rently with silica synthesis, allowing the possibility of three-
dimensional steric confinement via formation of a silicate cage
around the occluded molecule. This process may be driven by
noncovalent interactions between 1 and the silica precursors in
addition to the covalent nucleation site at the terminus of the
aminopropyl tether.43 Site isolation is ensured in the microporous
material via both statistical isolation of 1 (2 mol % relative to
silica network precursors), as well as by conducting the synthesis
under acid-catalyzed conditions that promote a similar hydrolysis
and condensation rate for the organosilane and the silica network
precursors to minimize self-condensation and potential aggrega-
tion of condensed organosilane 1.44,45
(44) Brinker, C. J.; Keefer, K. D.; Schaefer, D. W.; Ashley, C. S. J. Non-Cryst.
Solids 1982, 48, 47-64.
(45) Brinker, C. J.; Keefer, K. D.; Schaefer, D. W.; Assink, R. A.; Kay, B. D.;
Ashley, C. S. J. Non-Cryst. Solids 1984, 63, 45-59.
(40) (a) Bass, J. D.; Anderson, S. L.; Katz, A. Angew. Chem., Int. Ed. 2003,
42, 5219-5222. (b) Bass, J. D.; Solovyov, A.; Pascall, A. J.; Katz, A. J.
Am. Chem. Soc. 2006, 128, 3737-3747.
(41) Zeigler, R. C.; Maciel, G. E. J. Phys. Chem. 1991, 95, 7345-7353.
(42) Pursch, M.; Sander, L. C.; Albert, K. Anal. Chem. 1996, 68, 4107-4113.
(43) Katz, A. The Synthesis and Characterization of Molecularly Imprinted
Materials. Ph.D. Thesis, California Institute of Technology, Pasadena, CA,
1999.
(46) Cao, G. Z.; Tian, H. J. Sol.-Gel Sci. Technol. 1998, 13, 305-309.
(47) Dai, S. Chem.-Eur. J. 2001, 7, 763-768.
(48) Coutinho, D.; Acevedo, A. O.; Dieckmann, G. R.; Balkus, K. J. Microporous
Mesoporous Mater. 2002, 54, 249-255.
(49) Sugahara, Y.; Inoue, T.; Kuroda, K. J. Mater. Chem. 1997, 7, 53-59.
(50) Katz, A.; Davis, M. E. Macromolecules 1999, 32, 4113-4121.
(51) Bass, J. D.; Katz, A. Chem. Mater. 2003, 15, 2757-2763.
(52) Katz, A.; Davis, M. E. Nature 2000, 403, 286-289.
9
5690 J. AM. CHEM. SOC. VOL. 128, NO. 17, 2006

Molecular Motion of Tethered Molecules
A R T I C L E S
powder pattern exhibiting recovery of the Pake doublet.53,54 The
correlation time τc in Figure 1d is larger than 10-2 s, and is
consistent with either a slower jumping motion or a static C-D
bond with librational motion, although it is not possible to tell
the difference between the two. This restriction of molecular
motion at lower temperature suggests that there is a larger
activation energy for motion involving the C-D bond in this
system than for tethered alkyl chains on silica surfaces.26,29,33
This activation energy may be increased due to interactions
between the carbamate carbonyl and the silica surface, which
may help organize the tethered molecule close to the silica
surface in its immobilized form (vida infra), resulting in greater
rigidity as observed in other systems, where long aliphatic chains
display less motion for carbons close to the silica surface and
increasing motion for those further away.26,28,33,41 The motion
of tethered molecule 1 is identical on both silica surfaces
investigated, Selecto and Aeroperl, with faster hopping motion
of the C-D bond at room temperature and reduced hopping
rate at 213 K. The rapid molecular motion indicates that 1 is
not sterically confined in the surface-grafted materials at room
temperature.
It should be easiest to achieve steric confinement of tethered
1 in the bulk microporous silica framework (2), because the
size of the micropores is commensurate with small molecule
dimensions.52 However, even for microporous material 2,
achieving confinement looks less clear-cut when compared to
previous studies of nontethered molecules in microporous
systems. In systems of cyclophosphazene inclusion compounds,
rapid rotational motion of benzene55 as well as p- and o-xylene56
is observed even at low temperatures of 150 K. In a study of
benzyltrimethylammonium cation structure directing agents
(SDAs) occluded in an as-made microporous zeolite, it was
shown that, although the SDAs experienced a reduced number
of possible configurations, they still experienced significant
motion, particularly in rapid rotation of the trimethylammonium
segment.57 The implication of these and similar studies is that
the mere presence of a micropore is insufficient for restricting
molecular motions of small molecules, particularly in untethered
form.
The effect of confinement on rotional mobility of tethered 1
by the microporous framework is shown in Figure 1e. The time
scale for the hopping motion τc is estimated to be greater than
10-2 s using the same quadrupole interaction parameters as used
in the simulation of the surface-grafted materials (Table 1). This
value is comparable to values observed in the surface-grafted
materials at 213 K. However, it is significantly larger than the
τc value observed in the surface-grafted materials at room
temperature, indicating that immobilized 1 is significantly more
confined in the bulk microporous material. Achieving the
“tightness of fit” of steric confinement in the bulk mesoporous
network of 3 is synthetically more challenging relative to the
microporous 2. Because of the bulk mesoporosity of the
framework, it could be possible for 1 to be grafted to the interior
mesoporous surface of 3, not unlike the postsynthetically grafted
1 in materials 4 and 5. This is because immobilizing 1 on a
preexisting silica surface produces anchored molecules that
sample the entire interior porosity, most of which is mesoporous
Figure 1. 2H MAS spectra of (a) surface material 5 (Aeroperl) at room
temperature, (b) surface material 4 (Selecto) at room temperature, (c) surface
material 4 (Selecto) at 413 K, (d) surface material 4 (Selecto) at 213 K, (e)
bulk microporous material 2 at room temperature, (f) bulk mesoporous
material 3 at room temperature, (g) bulk microporous material 2 at 413 K,
and (h) bulk mesoporous material 3 at 413 K.
the materials. This required that the tethered 1 illustrated in
Scheme 1-3 be deprotected with iodotrimethylsilane to liberate
primary amines for use in catalysis (see Supporting Information
for details). With confirmation of the framework porosity and
site isolation of each of the materials, we then investigated the
confinement of 1 in the bulk amorphous materials.
NMR Spectroscopy Results. To analyze the degree of
confinement and compare between different materials, we
2
utilized solid-state H MAS NMR spectroscopy to assess the
molecular motion of 1 immobilized in each of the three types
of materials: bulk microporous (2), bulk mesoporous (3), and
surface-functionalized (4 and 5). The degree of molecular
mobility of immobilized 1, particularly around the C-O bond
axis involving the chiral carbon, correlates inversely with degree
of confinement. Because of the relatively high dilution of
2
immobilized 1 in silica in each of the materials, the H MAS
NMR technique was chosen rather than the corresponding wide-
line static NMR spectroscopic methods to circumvent limitation
due to signal sensitivity.23,53 2H MAS NMR spectroscopy is
generally useful for investigating molecular motion with cor-
relation times ranging between 10-3 s and 10-8 s.53,54
As shown in Figure 1a and 1b, there is rapid molecular motion
(time scale faster than 10-7 s) in the surface-functionalized
materials 4 and 5 at room temperature. The shape of the powder
pattern can be fit to a simple two-site hopping mode of motion.54
This mobility is consistent with the results in the previously
described examples for the rotational mobility of long chain
aliphatic compounds grafted to the surface of silica.26,28,29,31,34
However, unlike the previous systems, which still display
significant molecular motion even when cooled to 200 K,26,29,33
the surface-functionalized material 4 shows a significant “freez-
ing out” of the motion at 213 K (Figure 1d), with the quadrupole
(53) Hologne, M.; Hirschinger, J. Solid State Nucl. Magn. Reson. 2004, 26,
1-10.
(54) Jelinski, L. W. High-Resolution NMR Spectroscopy of Synthetic Polymers
in Bulk. Methods in Stereochemical Analysis; VCH: Deerfield Beach, FL;
Vol. 7, pp 335-364.
(55) Meirovitch, E.; Belsky, I.; Vega, S. J. Phys. Chem. 1984, 88, 1522-1526.
(56) Meirovitch, E.; Belsky, I. J. Phys. Chem. 1984, 88, 4308-4315.
(57) Shantz, D. F.; Fild, C.; Koller, H.; Lobo, R. F. J. Phys. Chem. B 1999,
103, 10858-10865.
9
J. AM. CHEM. SOC. VOL. 128, NO. 17, 2006 5691

A R T I C L E S
Defreese et al.
Table 1. Estimated Correlation Time (τc) by Using a Two-Site
4, materials 2 (Figure 2f) and 3 (Figure 2e) experience some
broadening of individual spinning sideband line widths at higher
temperature as compared to room temperature (Figure 2d). The
origin of this line broadening is mainly caused by the slightly
increased hopping rate around the same C-O bond axis, as
manifested in the change from τc > 10-2 s to τc ≈ 10-3 s-10-4
s. Rotational motion with this new correlation time subsequently
interferes with the magic angle spinning rate to cause the
observed broadening. The distortion of the overall powder
pattern from room-temperature spectra is not well-understood,
and it may result from a number of potential causes such as an
increased contribution of jumping angle distribution, libration,
or wobble motions.53,60,61 In addition, a sharp component is
observed for all of the materials studied and is unrelated to C-D
bond dynamics. Such a phenomenon may be due to trapped
moisture and should not be interpreted as the emergence of a
sharp component due to motional averaging.
As the NMR spectroscopy results above demonstrate, im-
mobilized 1 is sterically confined within the bulk materials even
at temperatures as high as 413 K, and this is in stark contrast to
its state in the surface-grafted materials at room temperature.
This requires silica nucleation near immobilized 1 during pore
formation in 2 and 3. The one known point of nucleation is at
the aminopropyl tether, where 1 covalently condenses to the
silica network precursors. However, to create a well-formed pore
capable of true three-dimensional steric confinement, additional
interactions between the silica surface and immobilized 1 may
facilitate confinement during the process of materials synthesis
via sol-gel hydrolysis and condensation. To investigate the
potential interactions of the carbonyl group of the carbamate 1
with the silica of the pore wall, we examined each of the
materials by solid-state FT-IR spectroscopy.
Hopping Model
material scaffold
correlation time (s)a
2 (bulk microporous) at
room temperature
3 (bulk mesoporous) at
room temperature
4 (surface Selecto 60) at
room temperature
4 (surface Selecto 60) at 213 K
2 (bulk microporous) at 413 K
3 (bulk mesoporous) at 413 K
>10-2
>10-2
<10-7
>10-2
∼10-3-10-4
∼10-3-10-4
a A single correlation time, quadrupole coupling constant QCC ) ∼155
kHz, and asymmetry parameter ηQ ) 0 were used for simulating the 2H
MAS spectra.
for 3.36,51 We carefully designed the mesoporous materials
synthesis for 3 to synthesize a microporous cage around each
immobilized 1 while synthesizing mesoporosity in the remainder
of the material framework. As shown in Figure 1f, the NMR
spectroscopic data support the existence of silicate cages that
provide steric confinement for immobilized 1. At room tem-
perature, 1 occluded within the bulk mesoporous material 3
displays similar limited molecular mobility (τc > 10-2 s) as
observed in the microporous material 2, including the same
characteristic powder pattern and quadrupole parameters. Be-
cause there is essentially no observed difference in molecular
mobility between the microporous and mesoporous frameworks,
the mesoporous materials synthesis procedure discovered has
indeed been successful in creating sterically confining mi-
cropores within the bulk mesoporous framework while retaining
hierarchical porosity and associated advantages of improved
accessibility and mass transport (vide supra).
We have also examined the surface-functionalized and bulk
materials at a higher temperature to investigate how effectively
steric confinement of 1 is preserved in the presence of additional
thermal energy. We chose a temperature of 413 K, because it
is a temperature at which we have carefully verified the stability
of tethered 1 in air via thermogravimetric analysis. The surface-
functionalized material 4 at 413 K exhibits significant narrowing
of the entire powder pattern (Figure 1c) relative to the room-
temperature spectrum (Figure 1b). This is indicative of additional
degrees of freedom via an increasing combination of motional
modes so that the anisotropic character of the quadrupole
coupling is being gradually averaged out. Additionally, when
compared to the room-temperature spectrum (Figure 2a), the
line width of the individual spinning sidebands is broadened at
413 K (Figure 2b), which is consistent with a higher amount of
librational motion that is nonspecific to the C-D bond at
elevated temperature and the time scale of which begins to
interfere with the magic angle spinning rate. Returning the
surface-imprinted material 4 to room temperature (Figure 2c)
results in a return of the narrow line width, a reversibility that
confirms lack of sample degradation at elevated temperature.
FT-IR Spectroscopy Results. FT-IR spectroscopy is a
sensitive tool for exploring the effect of hydrogen bonding
and polarization of the dielectric environment surrounding the
carbonyl functionality of carbamate 1. Carbonyls not only
show a very strong characteristic stretching band in the IR
spectrum, but this band also shifts from higher to lower wave-
numbers when going from noninteracting to either hydrogen-
bonded or higher dielectric constant environments.62,63 Hydrogen
bonding between residual silanols and amide carbonyls has
been studied previously and shown to shift the carbonyl
stretching band to lower wavenumbers by 12-25 cm-1
relative to the neat amide,64,65 and a similar lowering of the
carbonyl band frequency has also been observed due to dielectric
effects upon dissolution of neat amide in a solvent such as
DMSO.63,66
Table 2 shows the results of FT-IR measurements at room
temperature. All of the silica materials clearly display a shift
(58) Laupretre, F.; Monnerie, L.; Virlet, J. Macromolecules 1984, 17, 1397-
1405.
(59) Pivcova, H.; Saudek, V.; Schmidt, P.; Hlavata, D.; Plestil, J.; Laupretre, F.
Polymer 1987, 28, 991-997.
(60) Goddard, Y. A.; Vold, R. L.; Hoatson, G. L. Macromolecules 2003, 36,
1162-1169.
Microporous material 2 (Figure 1g) and mesoporous material
3 (Figure 1h) show only slight narrowing of the powder pattern
at 413 K and retain the characteristic Pake doublet quadrupole
pattern, with only minor changes in the hopping rate relative to
the surface-grafted material at room temperature (Table 1). This
confirms that the steric confinement of the bulk mesoporous
material is the same as the bulk microporous material even at
elevated temperatures. As observed for surface-grafted material
(61) Malyarenko, D. L.; Vold, R. L.; Hoatson, G. L. Macromolecules 2001, 34,
7911-7915.
(62) Pimentel, G. C.; McClellan, A. L. The Hydrogen Bond; W. H. Freeman:
San Francisco, CA, 1960.
(63) (a) Torii, H.; Tasumi, M. J. Raman Spectrosc. 1998, 29, 537-546. (b)
Kubelka, J.; Keiderling, T. A. J. Phys. Chem. A 2001, 105, 10922-10928.
(64) Ogoshi, T.; Chujo, Y. Bull. Chem. Soc. Jpn. 2003, 76, 1865-1871.
(65) Saegusa, T. Pure Appl. Chem. 1995, 67, 1965-1970.
(66) Eaton, G.; Symons, M. C. R.; Rastogi, P. P. J. Chem. Soc., Faraday Trans.
1989, 85, 3257-3271.
9
5692 J. AM. CHEM. SOC. VOL. 128, NO. 17, 2006

Molecular Motion of Tethered Molecules
A R T I C L E S
Figure 2. 2H MAS NMR spectra displayed for the center and the first spinning sidebands of: (a) surface-grafted material 4 at room temperature, (b)
surface-grafted material 4 at 413 K, (c) surface-grafted material 4 after a return to room temperature, (d) mesoporous material 3 at room temperature, (e)
mesoporous material 3 at 413 K, and (f) microporous material 2 at 413 K. The spinning rate is 6 kHz in all spectra, except (a) where it is 5 kHz.
Table 2. FT-IR Carbonyl Stretching and Isolated Silanol
Frequencies for Silica Materials
involves no hydrogen-bonding interaction and involves instead
solvation by the silica surface, which causes polarization of the
dielectric environment surrounding the carbonyl.63,66 The fact
that all three classes of materials show the same shift of the
carbonyl band indicates an energetic driving force for 1 to
nucleate close to the silica surface in the region of the carbonyl.
During bulk silica hydrolysis and condensation, this driving
force may assist in pore-wall formation, which creates a cage
of silica closely around immobilized 1, thereby sterically
confining the molecule to the extent that rotation is significantly
limited in both the bulk microporous and the bulk mesoporous
materials. These observations of carbonyl interaction with the
silica surface even after extensive capping are supported by the
catalytic data of Corma and co-workers, who observed signifi-
cant support effects on active site reactivity even in capped silica
materials.70
C
d
(cm-
O stretch
isolated silanol band surface coverages
1
b
1 c
material scaffold
)
(cm-
)
(mmol/g)
noninteracting 1a
neat 1
2 (bulk microporous)
3 (bulk mesoporous)
4 (surface Selecto 60)
1726 ( 2
1711 ( 10
1699 ( 2
1701 ( 2
1703 ( 2
n/a
n/a
3739 ( 2
3739 ( 2
3739 ( 2
3739 ( 2
n/a
n/a
0.28
0.26
0.21
0.098
5 (surface Aeroperl 300) 1703 ( 2
a 0.01 M of 1 in CCl4. b From subtraction spectrum of protected and
deprotected materials. c From spectrum of materials deprotected with
trimethylsilyliodide.
of approximately 20 cm-1 in the carbonyl band as compared to
the noninteracting 1 at 0.01 M in CCl4, and a measurable shift
with respect to the carbonyl band for neat carbamate 1. However,
there is no discernible difference in the position of the carbonyl
band for immobilized 1 between surface-grafted and bulk
materials. Because trimethylsilyl capping in materials 2-4 still
leaves behind residual uncapped silanols that remain in the
framework of each of the materials, particularly for silanols that
may be sterically hindered,67 the observed shifts may be due to
either specific hydrogen bonds between the carbonyl and silanols
or the presence of a more polarized dielectric environment due
to the nearby presence of the silica surface. Evidence for silanols
remaining despite trimethylsilyl capping is provided by a
relatively sharp isolated silanol infrared band centered near 3740
cm-1 in all materials investigated.
The similar shift of the carbonyl stretching band in the FT-
IR spectrum of each of the materials indicates that the carbonyl
groups in both surface- and bulk-immobilized 1 are not within
a gas-phase environment and are affected by electrostatic
interactions with the silica surface. Because of the flexibility
of the aminopropyl tether in immobilized 1, we postulate that
it may be possible for the carbonyl to hydrogen bond with
residual silanols, even in the surface-grafted materials, given
the conformational flexibility in 1 and as previously described
for molecules anchored with propyl tethers on silica.68 One
possibility is a double69 hydrogen-bonding arrangement, which
is illustrated in Schemes 1-3, whereas another possibility
Conclusions
We have investigated and compared the steric confinement
of isolated tethered molecules within three types of material
networks: two-dimensional silica surface, bulk amorphous
microporous silica, and bulk amorphous mesoporous silica. The
2H MAS NMR study unequivocally demonstrates that the bulk
materials achieved thorough steric confinement of the im-
mobilized 1, severely limiting molecular mobility at both room
temperature and elevated temperature. This result is remarkable
in light of the previously explored systems, which generally
displayed significant restrictions of molecular motion only at
very low temperatures26,29 or in the presence of multiple strong
hydrogen bonds.24,25
These conclusions are supported by FT-IR spectroscopic
evidence that demonstrates that the silica surface can interact
with the carbonyl functionality of 1, causing an observable shift
of the carbonyl band in the IR spectrum. The fact that the silica
surface after TMS-capping can influence the shift of the carbonyl
peak in the IR suggests that similar interactions between the
carbonyl of the occluded molecule and silica precursors may
play an important role in nucleating the formation of a pore
wall close to 1 during materials synthesis.
These results comprise the first explicit proof of tethered
molecule confinement within bulk amorphous silica. There is
(67) Little, C. J.; Whatley, J. A.; Dale, A. D. J. Chromatogr. 1979, 171, 435-
438.
(68) Caravajal, G. S.; Leyden, D. E.; Quinting, G. R.; Maciel, G. E. Anal. Chem.
1988, 60, 1776-1786.
(69) Pihko, P. M. Angew. Chem., Int. Ed. 2004, 43, 2062-2064.
(70) Baleizao, C.; Gigante, B.; Garcia, H.; Corma, A. J. Catal. 2003, 216, 199-
207.
9
J. AM. CHEM. SOC. VOL. 128, NO. 17, 2006 5693

A R T I C L E S
Defreese et al.
broad potential for imparting significant steric confinement on
occluded molecular guests using bulk amorphous silica as a
synthetic host, which is difficult to accomplish via other methods
of host-guest encapsulation.5,20 The difficult-to-achieve con-
finement of isolated functional groups demonstrated here has
particular implications in fundamental studies and technological
applications exploiting confinement such as in selective het-
erogeneous catalysis and adsorption,71 as well as photolumi-
nescence.72,73
and Dr. Sungsool Wi at Virgina Polytechnical Institute and State
University for sharing 2H MAS NMR simulation software. We
gratefully acknowledge the National Science Foundation (CTS
0407478) for financial support, as well as a graduate research
fellowship for J.L.D. S.-J.H. acknowledges the National Science
Foundation under Grant Number 9724240 and the MRSEC
Program of the National Science Foundation under Award
Number DMR-0080065 for supporting the Caltech Solid State
NMR facility.
Acknowledgment. We thank Dr. Maggy Hologne and Dr.
Jerome Hirschinger at CNRS, Universite Louis Pasteur, France,
Supporting Information Available: Evidence for site isola-
tion. This material is available free of charge via the Internet at
http://pubs.acs.org.
(71) Davis, M. E.; Katz, A.; Ahmad, W. R. Chem. Mater. 1996, 8, 1820-1839.
(72) Ostafin, A.; Siegel, M.; Wang, Q.; Mizukami, H. Microporous Mesoporous
Mater. 2003, 57, 47-55.
(73) Makarova, O.; Ostafin, A.; Miyoshi, M.; Norris, J. J. Phys. Chem. B 1999,
103, 9080-9084.
JA0556474
9
5694 J. AM. CHEM. SOC. VOL. 128, NO. 17, 2006