Effects of primary and secondary surface groups in enantioselective catalysis using nanoporous materials with chiral walls

Article abstract of DOI:10.1021/ja1017706

Mesoporous materials are valuable supports for the immobilization of various molecular catalysts. Cases in which the performance of the catalyst improves after immobilization have seldom been reported, especially when it comes to enantioselective synthesis. Knowledge of how the presence of the support surface alters the properties of a bound catalyst is therefore very important. In the current article, a new periodically ordered mesoporous organosilica material (PMO) with walls exclusively made of a chiral building block is presented. The attachment of Al^III as a Lewis acid center to the chiral group furnishes the material with catalytic activity, for instance, for the asymmetric carbonyl ene reaction. Thus, the presented materials are valuable model systems for studying the effect of the chiral surface and also neighboring groups attached to the silanol groups in the network. It is reported that surface-bound Al^III exhibits significantly better performance (higher ee values) than an analogous molecular reference catalyst. Furthermore, it could be shown that the ee values increase even further when more bulky secondary groups are attached to the pore walls. Therefore, the main conclusion of the current report is that for cases in which steric conditions of a catalyst play a crucial role its immobilization inside a tailor-made mesoporous organosilica material is beneficial with respect to cooperative effects between the catalytic center and neighboring surface groups.

Full text of DOI:10.1021/ja1017706

Published on Web 04/16/2010
Effects of Primary and Secondary Surface Groups in
Enantioselective Catalysis Using Nanoporous Materials with
Chiral Walls
Andreas Kuschel and Sebastian Polarz*
Department of Chemistry, UniVersity of Konstanz, 78457 Konstanz, Germany
Received March 2, 2010; E-mail: sebastian.polarz@uni-konstanz.de
Abstract: Mesoporous materials are valuable supports for the immobilization of various molecular catalysts.
Cases in which the performance of the catalyst improves after immobilization have seldom been reported,
especially when it comes to enantioselective synthesis. Knowledge of how the presence of the support
surface alters the properties of a bound catalyst is therefore very important. In the current article, a new
periodically ordered mesoporous organosilica material (PMO) with walls exclusively made of a chiral building
block is presented. The attachment of AlIII as a Lewis acid center to the chiral group furnishes the material
with catalytic activity, for instance, for the asymmetric carbonyl ene reaction. Thus, the presented materials
are valuable model systems for studying the effect of the chiral surface and also neighboring groups attached
to the silanol groups in the network. It is reported that surface-bound AlIII exhibits significantly better
performance (higher ee values) than an analogous molecular reference catalyst. Furthermore, it could be
shown that the ee values increase even further when more bulky secondary groups are attached to the
pore walls. Therefore, the main conclusion of the current report is that for cases in which steric conditions
of a catalyst play a crucial role its immobilization inside a tailor-made mesoporous organosilica material is
beneficial with respect to cooperative effects between the catalytic center and neighboring surface groups.
15 have become important model systems in catalytic research.4,5
Chiral groups can be attached to the walls of mesoporous silica
Introduction
materials such as MCM-41 and SBA-15 via grafting,6 with the
condensation of terminal organosilanes R*Si(OR′)3 (R* ) a
chiral organic group) with surface-bound silanol groups (Si-OH)
resulting in stable Si-O-Si linkages.7 Good examples exist
for amino acids and short peptide sequences.8 Aida et al.
reported a different, very interesting method.7,9 A triblock
surfactant, the so-called lizard template containing an alkoxy-
silane headgroup covalently bound to an amino acid group
attached to a long alkyl chain together with ∼90% of a source
for unmodified SiO2, was employed for the one-pot synthesis
of a mesoporous material with ordered porosity.9 The inherent
disadvantage of the described methods is that the porous
materials contain the chiral entity only in minor amounts,
typically in the range 5-15 mol %.7 Therefore, it is a tempting
idea to attach catalytically active species to surfaces that are
chiral by themselves.5
Because of the biological preference for only one of the
alternative enantiomers (L-amino acids and D-sugars), it is
important that pharmaceutical products are associated directly
with the ability to allocate them in an enantiomerically pure
form. Enantioselective synthesis using molecular catalysts in
the homogeneous phase has already reached a profound stage
of development.1 Because of the numerous advantages of
catalysis operating in the heterogeneous mode, the immobiliza-
tion of molecular compounds on suitable supports represents
an important task. However, how an interface affects the
properties of a molecular catalyst bound to it is still a matter of
intense discussion.2,3
Because there are numerous possibilities to shape and also
to modify the surface of silica, it is one of the most important
support materials. For instance, mesoporous silica materials with
periodically ordered pore systems such as MCM-41 and SBA-
An alternative approach exists to equip mesoporous silica
materials with higher organic functionality content. Periodically
(1) Kolb, H. C.; Vannieuwenhze, M. S.; Sharpless, K. B. Chem. ReV. 1994,
94, 2483. Grubbs, R. H.; Chang, S. Tetrahedron 1998, 54, 4413.
Stephenson, G. R. E. AdVanced Asymmetric Synthesis; Chapman &
Hall: London, 1996. Procter, G. StereoselectiVity in Organic Synthesis;
Oxford University Press: Oxford, U.K., 1998. Cox, G. B. PreparatiVe
EnantioselectiVe Chromatography; Blackwell Publishing: Oxford,
U.K., 2005. Dalko, P. I.; Moisan, L. Angew. Chem., Int. Ed. 2001,
40, 3726. Dalko, P. I.; Moisan, L. Angew. Chem., Int. Ed. 2004, 43,
5138.
(4) Taguchi, A.; Schueth, F. Microporous Mesoporous Mater. 2005, 77,
1. Climent, M. J.; Corma, A.; Iborra, S.; Miquel, S.; Primo, J.; Rey,
F. J. Catal. 1999, 183, 76.
(5) Polarz, S.; Smarsly, B. J. Nanosci. Nanotechnol. 2003, 2, 581.
(6) Kresge, C. T.; Leonowicz, M.; Roth, W. J.; Vartuli, J. C.; Beck, J. S.
Nature. 1992, 359, 710. Yang, P.; Zhao, D.; Margolese, D. I.; Chmelka,
B. F.; Stucky, G. D. Nature 1998, 396, 152.
(7) Hoffmann, F.; Cornelius, M.; Morell, J.; Froeba, M. Angew. Chem.,
Int. Ed. 2006, 45, 3216.
(2) Corma, A.; Garcia, H. AdV. Synth. Catal. 2006, 348, 1391. Fraile,
J. M.; Garcia, J. I.; Herrerias, C. I.; Mayoral, J. A.; Pires, E. Chem.
Soc. ReV. 2009, 38, 695.
(8) Luechinger, M.; Kienhofer, A.; Pirngruber, G. D. Chem. Mater. 2006,
18, 1330.
(3) Li, C.; Zhang, H. D.; Jiang, D. M.; Yang, Q. H. Chem. Commun.
2007, 547.
(9) Zhang, Q. M.; Ariga, K.; Okabe, A.; Aida, T. J. Am. Chem. Soc. 2004,
126, 988.
9
6558 J. AM. CHEM. SOC. 2010, 132, 6558–6565
10.1021/ja1017706  2010 American Chemical Society

Enantioselective Catalysis Using Nanoporous Materials
A R T I C L E S
Chart 1. Asymetric Carbonyle ene Reaction24
ordered mesoporous organosilica materials (PMOs) are prepared
by pure and undiluted silsesquioxane precursors possessing a
bridging organic group (R′O)3Si-R-Si(OR′)3.10 Because the
composition of the resulting mesoporous network can be
described as RSi2O3, the interfacial accessibility of the organic
group is maximized.11 The PMO field has been described
recently in excellent review articles.7,12
There are a number of papers in which a sol-gel precursor
containing a bridging, chiral entity in addition to a pure silica
source (e.g., tetraethoxysilane) has been used for the preparation
of mesoporous materials. Those materials were also frequently
investigated in catalysis. For instance, Corma et al. have reported
in several papers materials containing a chiral bridging vanadyl
salen complex.13 Li et al. have used a similar system containing
a Mn(salen) complex and have reported several catalytic
studies.14 Garcia and Ihmels describe the stereoselective di-π-
methane rearrangement in a mesoporous material containing a
chiral 1,2-bis-(ureido)-cyclohexyl linker. Because in the latter
reports the chiral entity has been introduced via co-condensation,
it still represents the minor constituent with the majority of the
materials consisting of SiO2. Therefore, it can be envisioned
that materials made exclusively from chiral building blocks are
very interesting targets.
Examples of real PMOs (∼100% organic modification) where
chiral building block have been used are still rare.3 Our group
applied enantioselective catalytic hydroboration to precursor
(EtO)3Si-CHdCH-Si(EtO)3, resulting in a chiral PMO with
Si-C*H(OH)-CH2-Si entities embedded in the walls.15 Later,
Thomas et al. used a chiral hydroboration agent directly instead
of a stereoselective catalyst.16 The report of a PMO material
containing a chiral amino acid entity was published in 2008,
and for the first time the chirality of the surface was probed
using physisorption measurements with a chiral gas.17 Froeba
et al. prepared a PMO with chiral benzylic ether bridges from
enantioselective catalysis in 88% ee.18 An entirely different and
very interesting system was published by Crudden et al.19 They
prepared PMO materials containing a building block with axial
chirality. To the best of our knowledge, there is currently no
report on the application of a real PMO material with chiral
walls in enantioselective catalysis.20
solids with regular pore systems a variety of alternative materials
have also been used for the immobilization of molecular
catalysts. The discussion of the entire field is beyond the scope
of this article. The interested reader is referred to the extensive
review written by Glorius et al.21 In comparison to other
materials such as functional polymers, it can be expected that
chiral PMOs are favorable because of the combination of the
high density of the asymmetric groups and the possibility of
assembling these groups with geometrical precision.22
A
complementary class of materials with pores in the micropore
regime (Dp < 2 nm) should be noted as well: the so-called
homochiral, porous metal-organic framework materials.23
There are various reactions that are suitable as model reactions
for enantioselective catalysis. The ene reaction is a little-known
C-C coupling reaction. Its bandwidth ranges from cyclizations
to ring-opening reactions, or decarboxylations. Its main char-
acteristic is the addition of a CdX bond (the so-called eneophile)
to a CdC bond possessing an allylic hydrogen (Chart 1). The
ene reaction belongs to the class of pericyclic reactions and can
be catalyzed by Lewis acids activating the CdX group. AlIII or
TiIV alkoxo complexes can be used as suitable Lewis acids.
Depending on the kind of substituents, a new stereocenter is
generated, and it has been reported that the use of chiral,
molecular catalysts allows enantioselective synthesis.24 For
instance, (-)-menthol is produced industrially using the in-
tramolecular ene reaction. Inspired by this work, we attempt to
synthesize a corresponding catalytic system based on a PMO
material.
Results and Discussion
Asymmetric heterogeneous catalysis is a topic of immense
interest. Therefore, it is not surprising that besides mesoporous
Very recently, our group described a versatile PMO precursor:
1,3-bis-tri-iso-propoxysilyl-5-bromobenzene (1).25 It could be
shown that the derivatization chemistry known for halogenated
aromatic compounds can also be applied in the case of 1.26 As
a consequence, the so-called UKON-PMO materials with
organosilicate networks constructed from benzoic acid, dithio-
benzoic acid, aniline, and others have been described
recently.17,22,25,27 Here, a new mesoporous organosilica with
chiral pore walls is reported. (See Chart 2; more details are given
in the Experimental Section.) The lithiation of 1 at the 5-position
leads to a versatile nucleophilic species that can attack the
carbonyl species in acetone, resulting in the racemic (1-
(10) Asefa, T.; MacLachan, M. J.; Coombs, N.; Ozin, G. A. Nature. 1999,
402, 867. Inagaki, S.; Guan, S.; Fukushima, Y.; Ohsuna, T.; Terasaki,
O. J. Am. Chem. Soc. 1999, 121, 9611. Melde, B. J.; Holland, B. T.;
Blanford, C. F.; Stein, A. Chem. Mater. 1999, 11, 3302.
(11) MacLachlan, M. J.; Asefa, T.; Ozin, G. A. Chem.sEur. J. 2000, 6,
2507.
(12) Hatton, B.; Landskron, K.; Whitnall, W.; Perovic, D.; Ozin, G. A.
Acc. Chem. Res. 2005, 38, 305.
(13) Baleizao, C.; Gigante, B.; Das, D.; Alvaro, M.; Garcia, H.; Corma,
A. J. Catal. 2004, 223, 106. Baleizao, C.; Gigante, B.; Garcia, H.;
Corma, A. J. Catal. 2003, 215, 199. Baleizao, C.; Gigante, B.; Das,
D.; Alvaro, M.; Garcia, H.; Corma, A. Chem. Commun. 2003, 1860.
(14) Zhang, H. D.; Zhang, Y. M.; Li, C. J. Catal. 2006, 238, 369. Zhang,
H.; Li, C. Tetrahedron 2006, 62, 6640. Jiang, D. M.; Gao, J. S.; Yang,
Q. H.; Yang, J.; Li, C. Chem. Mater. 2006, 18, 6012.
(21) Heitbaum, M.; Glorius, F.; Escher, I. Angew. Chem., Int. Ed. 2006,
45, 4732.
(15) Polarz, S.; Kuschel, A. AdV. Mater. 2006, 18, 1206.
(16) Ide, A.; Voss, R.; Scholz, G.; Ozin, G. A.; Antonietti, A.; Thomas,
A. Chem. Mater. 2007, 19, 2649.
(22) Kuschel, A.; Drescher, M.; Kuschel, T.; Polarz, S. Chem. Mater. 2010,
22, 1472.
(17) Kuschel, A.; Sievers, H.; Polarz, S. Angew. Chem., Int. Ed. 2008, 47,
9513.
(23) Seo, J. S.; Whang, D.; Lee, H.; Jun, S. I.; Oh, J.; Jeon, Y. J.; Kim, K.
Nature 2000, 404, 982. Lin, W. B. J. Solid State Chem. 2005, 178,
2486. Wu, C. D.; Lin, W. B. Angew. Chem., Int. Ed. 2005, 44, 1958.
(24) Clarke, M. L.; France, M. B. Tetrahedron 2008, 64, 9003.
(25) Kuschel, A.; Polarz, S. AdV. Funct. Mater. 2008, 18, 1272.
(26) Polarz, S.; Kuschel, A. Chem.sEur. J. 2008, 14, 9816.
(27) Kuschel, A.; Luka, M.; Wessig, M.; Drescher, M.; Fonin, M.; Kiliani,
G.; Polarz, S. AdV. Funct. Mater. 2010, 20, 1133.
(18) Morell, J.; Chatterjee, S.; Klar, P. J.; Mauder, D.; Shenderovich, I.;
Hoffmann, F.; Froeba, M. Chem.sEur. J. 2008, 14, 5935.
(19) MacQuarrie, S.; Thompson, M. P.; Blanc, A.; Mosey, N. J.; Lemieux,
R. P.; Crudden, C. M. J. Am. Chem. Soc. 2008, 130, 14099.
(20) Yang, Q. H.; Han, D. F.; Yang, H. Q.; Li, C. Chem.sAsian J. 2008,
3, 1214. Li, C. Catal. ReV.sSci. Eng. 2004, 46, 419.
9
J. AM. CHEM. SOC. VOL. 132, NO. 18, 2010 6559

A R T I C L E S
Kuschel and Polarz
Chart 2. Synthesis Steps Leading to the New Mesoporous
IR (SI-2) and CD spectroscopy (Figure 1d). A mesoporous
material with a cylindrical pore-system has formed as indicated
by TEM. The pore-to-pore distance is d100 ) 11.8 nm according
to SAXS. Because only one diffraction peak is seen in SAXS,
it can be concluded that only parts of the sample are highly
ordered. The isotherm of UKON2g possesses a shape that is
typical for mesoporous solids with a capillary condensation step
at p/p0 ≈ 0.63.31 The material possesses an internal surface area
of 1045 m2/g as determined by the BET method.32 The high
amount of adsorbed gas in the pressure range below p/p0 < 0.06
and the steady increase for pressures lower than p/p0 ) 0.4
indicates the presence of large micropores. The pore-size
distribution function of UKON2g was determined from the BJH
evaluation of the adsorption branch of the isotherm.33 A pore-
size maximum at DP ) 5.3 nm is found (Figure 1b). The pore-
size determined from N2 physisorption is in good agreement
with the TEM data. In combination with the SAXS data, it can
be estimated that the pore walls of UKON2g are relatively thick
(Dthickness ≈ d100(SAXS) - Dpore(N2) ≈ 6.5 nm). The latter finding
can also be confirmed by TEM images. It is seen that the pore
walls have extensions that are similar to those of the pores
(Figures 1a and SI-2). The chemical nature of UKON2g was
investigated via solid-state MAS NMR spectroscopy. The 29Si
NMR spectrum of UKON2g (shown in Figure 2b) contains three
signals characteristic of organically modified silica materials at
δ ) -69 ppm for (HO)2RSi(OSi) = T1, -78 ppm for
(HO)RSi(OSi)2 = T2, and -87 ppm for RSi(OSi)3 = T3.34
Signals indicating the presence of pure silica (SiO2; so-called
Q signals at δ ≈ -110 ppm) are not seen. This proves that the
precursor has been used in an undiluted form and that the Si-C
bonds in UKON2g are stable. In the 13C NMR spectrum, the
signals for ChH3CgH(OH)PhSi2O3 (δg ) 72.2 ppm, δh ) 26.0
ppm) and the remaining Carom (δ ) 134.6, 142.3, and 148.4
ppm) are observed (SI-2). It was determined qualitatively via
CD spectroscopy that the stereochemical information was
maintained during the transition of the precursor (4) to the
material (5). It is important to note that CD spectroscopy applied
to solid materials is very susceptible to artifacts due to scattering
or insufficient transmission. Thus, a thin film of the UKON2g
mesoporous organosilica material has been prepared via spin
coating (Experimental Section). Still, one has to be careful with
the thickness and homogeneity of the film. Data suitable for a
qualitative comparison of the precursor (4) and the resulting
chiral mesoporous organosilica materials could be obtained for
a film of 50 nm thickness (Figure 1d). An amplitude to negative
rotational angles (λ ) 240-200 nm) and a band with a negative
sign at λ ≈ 205-207 nm is seen for both compounds, the
precursor and the resulting mesoporous organosilica material.
Organosilica Material with Chiral Pore Wallsa
a Retrosynthesis considerations are indicated by the wavy gray lines.
hydroxyethyl)benzene derivative (2). The oxidation of 2 using
pyridinium chlorochromate affords the acetophenone derivative
(3). Finally, the enantioselective hydroboration28 of the CdO
bond results in novel, chiral sol-gel precursor (-)-(S)-1,3-bis-
tri-iso-propoxysilyl-5-phenylethanol (4).
Each of the compounds (1-4) has been purified prior to use
1
in a succeeding step and was characterized by H NMR, 13C
NMR, and FT-IR spectroscopy and mass spectometry (MS).
The latter data are given in the experimental section. Because
of the chirality of 4, some additional analytical techniques
became necessary. 4 is optically active (RD ) -8.4°), and its
CD spectrum is shown in Figure 1d. In addition, the success of
the enantioselective hydroboration reaction could be probed
quantitatively. 4 was reacted with a chiral shift reagent (Sup-
porting Information SI-1) followed by 31P NMR spectroscopic
evaluation.29 The relative intensities of the two 31P NMR signals
(δ1 ) 149.25 ppm and δ1 ) 142.73 ppm) allows us to determine
the abundance of the two possible enantiomers. For the reaction
of 3 to 4, an ee value of 98% has been determined. Chemically
analogous but racemic compound 2 was taken as a reference.
Expectedly, the ee value for the formation of 2 from 1 was 0%.
New mesoporous organosilica material UKON2g was pre-
pared from 4 using amphiphilic block copolymer Pluronic P123
as a structure-directing agent at low pH () 0) under true liquid-
crystal conditions.30 The resulting material was characterized
by a variety of analytical methods including transmission
electron microscopy (TEM, Figure 1a), N2 physisorption (Figure
1b), small-angle X-ray scattering (SAXS, Figure 1c), solid-state
13C- (SI-2) and 29Si- (Figure 2b) NMR spectroscopy, and FT-
(30) Zhao, D.; Feng, J.; Huo, Q.; Melosh, N.; Fredrickson, G. H.; Chmelka,
B. F.; Stucky, G. D. Science 1998, 279, 548. Attard, G. S.; Glyde,
J. G.; Goeltner, C. G. Nature 1995, 378, 366. Attard, G. S.; Goeltner,
C. G.; Corker, J. M.; Henke, S.; Templer, R. H. Angew. Chem., Int.
Ed. Engl. 1997, 36, 1315. Polarz, S.; Antonietti, M. Chem. Commun.
2002, 2593.
(31) Gregg, S. J.; Sing, K. S. W. Adsorption, Surface Area and Porosity,
4th ed.; Academic Press: London, 1982; Vol. 2. Rouqerol, J.; Avnir,
D.; Fairbridge, C. W.; Everett, D. H.; Haynes, J. H.; Pernicone, N.;
Ramsay, J. D.; Sing, K. S. W.; Unger, K. K. Pure Appl. Chem. 1994,
66, 1739.
(32) Brunauer, S.; Emmet, P. H.; Teller, E. J. Am. Chem. Soc. 1938, 60,
309.
(28) Disimone, B.; Savoia, D.; Tagliavini, E.; Umanironchi, A. Tetrahedron:
Asymmetry 1995, 6, 301.
(33) Barret, E. P.; Joyner, L. G.; Halenda, P. H. J. Am. Chem. Soc. 1951,
73, 373.
(29) Parker, D. Chem. ReV. 1991, 91, 1441. Sullivan, G. A.; Dale, J. A.;
Mosher, H. S. J. Org. Chem. 1973, 38, 2143.
(34) Yoshina-Ishii, C.; Asefa, T.; Coombs, N.; MacLachlan, M. J.; Ozin,
G. A. Chem. Commun. 1999, 2539.
9
6560 J. AM. CHEM. SOC. VOL. 132, NO. 18, 2010

Enantioselective Catalysis Using Nanoporous Materials
A R T I C L E S
Figure 1. Selected analytical data for the new chiral, mesoporous organosilica material UKON2g. (a) TEM micrograph, (b) N2 phyisorption isotherm
(black) and BJH pore-size distribution function (gray), (c) SAXS, and (d) CD spectrum (gray) in comparison to the spectrum of the precursor (4, black) as
a reference.
Figure 2. (a) 27Al and (b) 29Si solid-state NMR spectra of the materials obtained after Me2AlCl treatment of the as-prepared UKON2g (gray curves) in
comparison to that of materials with silanol groups quenched by silylation.
The attachment of the Lewis acid AlIII to the chiral alcohol
group could be achieved via treatment of UKON2g with
Me2AlCl (Experimental Section). However, the problem was
encountered that Me2AlCl also reacts readily with surface silanol
groups as proven by the solid-state 27Al NMR spectra shown
in Figure 2a. Two signals of comparable intensity are present
at δ ) 105.4 ppm corresponding to Al-O-C* and at δ ) 97.9
ppm corresponding to the undesired Al-O-Si species. The latter
material is denoted as catalytic material (B) in the following
parts of the article (Scheme 1). Quenching of the silanol groups
in UKON2g could be achieved after its treatment with a
silylation reagent such as R3SiCl (R ) Me, Experimental
Section). It is important to note that the silylation reagent will
also initially react with the alcohol groups. However, the
resulting C-O-SiMe3 motif, which is also known as a
protecting group, can be easily reconverted to C-OH via
treatment with trifluoroacetic acid (Experimental Section). As
a result, it can be seen in the solid-state 29Si NMR spectra
(Figure 2b) that one new signal at δ ) 5.4 emerges that is typical
of Si-O-SiMe3. The treatment with trifluoroacetic acid has one
slight disadvantage. A small signal at -110 ppm has appeared
in the 29Si spectrum of the material after silylation (Figure 2b).
9
J. AM. CHEM. SOC. VOL. 132, NO. 18, 2010 6561

A R T I C L E S
Kuschel and Polarz
Scheme 1. Different Catalytic Systems Used in This Study
This signal is typical of so-called Q-type silicon atoms
((HO)4 - xSi(OSi)x) and can be explained by the minor cleavage
of the Si-C bond by traces of fluoride ions known to be
dissolved in trifluoroacetic acid. When silylated UKON2g is
treated with Me2AlCl, one obtains material C. It can be seen
clearly from the 27Al NMR spectrum that the undesired species
Al-O-Si has almost vanished. The majority of AlIII centers
are attached to the asymmetric center via an oxo bridge (Scheme
1) in material C. In addition to materials C and B, a purely
siliceous SBA-15-type mesoporous material treated analogously
with Me2AlCl (denoted as material A) and a corresponding
chiral molecular compound D were used as reference catalysts
in the asymmetric carbonyl ene reaction (R1 ) phenyl, R2 )
CCl3; see Chart 1). The reaction mixtures were investigated by
NMR spectroscopy (SI-3). The results of studies with four
catalytic systems A-D are shown in Figure 3. Depending on
reaction conditions, in addition to the signals corresponding to
the ene product, the presence of minor amounts of the starting
compounds and a byproduct originating from the oligomeriza-
tion of styrene has been observed. It is important to note that
the ene reaction may proceed according to three different
mechanisms (SI-4): a concerted mechanism involving a 6π
aromatic system, a zwitterionic mechanism, and a radical
mechanism. The latter process is obviously responsible for the
oligomerization of styrene, and indeed oligomeric R-methyl
styrene species could be identified as the byproduct by 1H NMR
spectroscopy (SI-3).
Unfortunately, there is practically no stereochemical differentia-
tion (ee values ≈ 0). However, significant differences between
the three catalytic systems containing the chiral Al species can
be observed at temperatures below 0 °C. In the case of catalyst
B, it has been found that the yield of the ene product drops
quickly with temperature and at the same time more of the
byproduct forms (Figure 3a,b). Considering the latter findings
together with the results reported for catalyst A, it can be
concluded that the AlIII species attached to silanol groups is
responsible for the formation of the undesired byproduct. The
latter assumption is supported by the fact that the molecular
reference catalyst (D) and the silanol-protected material (C)
show almost the same temperature dependency regarding the
yield of the ene product (Figure 3a; squares and spheres). The
question arises as to if there are differences in stereochemical
effects (Figure 3c). The stereochemical differentiation increases
when the temperature is lowered. Although the ee value at T )
-55 °C is 72% in the case of catalytic material C, it is only
17% for molecular reference catalyst B. The 4-fold increase in
enantioselectivity simply due to immobilization is remarkable
and requires further explanation.
The stereochemical control in enantioselective catalysis is the
result of a complex interplay of the electronic and steric
influences of the ligands and the metal center to which they
coordinate. It has been observed very frequently that the steric
demand of the chiral group at the catalyst enhances the
stereochemical control (Chart 1).35 Therefore, it can be assumed
that the presence of the pore surface increases the steric
constraints in comparison to a molecular system. Two ap-
proaches are proposed for changing such steric constraints even
further (Figure 4). It is envisioned that the mobility of guests
A is clearly the inferior catalyst in the discussed series (Figure
3; hashes). Only a relatively low amount of the ene product in
its racemic form has formed at room temperature (RT). Instead,
much of the byproduct is present. The performances of the
alternative catalysts (C and D) at RT are very similar to each
other. On the order of 80% of the ene product has formed.
(35) Noyori, R. Angew. Chem., Int. Ed. 2002, 41, 2008.
9
6562 J. AM. CHEM. SOC. VOL. 132, NO. 18, 2010

Enantioselective Catalysis Using Nanoporous Materials
A R T I C L E S
Figure 4. Dependence of the ee value on the pore size and character of
the secondary surface group.
secondary O-SiR3 groups attached to the surfaces of the
organosilica mesopores have a profound impact on the enan-
tioselectivity of the catalytic reaction. The ee values increase
in the order of R ) Me (72%) < Et (74%) < isoPr (80%). For
the explanation of this effect, one has to consider the effect of
the molecular structure of the surface on the transition state
of the ene reaction involving a six-membered ring as depicted
in Chart 1. A model of the organosilica surface derived from
molecular mechanics calculations is shown in Figure 5. It can
be seen that the presence of the secondary surface groups mainly
affects the rotational freedom of the Al-O contact formed
between the Lewis acid AlIII center and the carbonyl compound.
If the surfaces are modified by O-SiisoPr3, then this rotation is
strongly hindered and the position of the emerging stereocenter
is fixed in the vicinity of the asymmetric carbon atom of the
organosilica surface. However, for the less bulky O-SiMe3 it
is clear that the rotation is much less hindered. It is reasonable
that the stereochemical influence of the chiral center attached
to AlIII via the alkoxide C*-O-Al function is less pronounded
when the probability of the emerging stereocenter in the vicinity
decreases because of rotation (Figure 5). The discussed picture
would also explain the observed temperature dependence of the
enantioselectivity for catalytic systems B and C.
Conclusions
In the present work, the synthesis of a new, periodically
ordered mesoporous organosilica material with a chiral surface
was described. Because the surface contains chiral secondary
alcohol functions, the modification with AlIII was successful
when surface silanol groups were masked as silyl-esters. One
possible application of this material as a catalyst in the
asymmetric ene reaction was investigated. The enantioselectivity
of the described system is much higher than for an analogous
molecular reference catalyst in the homogeneous case.
The latter results points to a more general testimony of the
current article. In several reports in the past, molecular catalysts
that had already been characterized by high ee values were the
targets of an immobilization strategy. The findings reported here
indicate that better stereochemical control can be achieved even
for less-appropriate ligands when the catalyst is imbedded into
the walls of a mesoporous solid. Better stereochemical control
can also be achieved even for less-appropriate ligands when
the chiral ligand is imbedded into the walls of a mesoporous
solid. In this sense, the surface behaves as a bulky ligand. The
Figure 3. Relative yield of the ene product (a, black curves), yield of the
byproduct (b, gray curves), and enantioremic excess (c, black curves) for
four different catalytic systems: C (O), B (4), molecular catalyst D (0),
and A (hashes). See also Scheme 1.
inside the pores is restricted when the pores become smaller.
This in turn could influence the ee values. Three materials with
pore sizes of Dp ≈ 2, 5.3, and 7.8 nm were prepared
(Experimental Section and SI-5), the surface silanol groups of
all three materials were modified with the same group
(O-SiMe3), and the catalytic tests were performed at the same
temperature (T ) -55 °C). The dependence of the ee value on
pore size could be observed (Figure 4), but the effect is not
very pronounced (ee7.8 nm ) 70%, ee5.3 nm ) 72%, and ee2 nm
)
74%). Therefore, a second approach was attempted. Instead of
the modification of the free silanol groups with O-Si(Me)3,
the more bulky O-Si(Et)3 and O-Si(isoPr)3 functionalities were
attached. This time the pore size of the materials (Dp ) 5.3
nm) was kept constant. It can be seen (Figure 4) that the
9
J. AM. CHEM. SOC. VOL. 132, NO. 18, 2010 6563

A R T I C L E S
Kuschel and Polarz
Figure 5. Schematic image indicating the cooperative effect between the AlIII catalyst attached to the chiral surface center and the secondary surface groups.
A transition state such as that shown in Chart 1 has been assumed.
trimethyl borate (0.21 g, 2.0 mmol) in 2 mL of THF is prepared at
room temperature. The mixture is stirred for 30 min, and the solvent
is removed under vacuum. Two milliliters of THF are added again,
and 0.18 mL of this solution is used as a catalyst in the next step.
3 (0.38 g, 0.72 mmol) and the catalyst are dissolved in 3 mL of
THF and cooled to 0 °C. The solution is stirred for 15 min, and
BH3 ·DMS (2 M, 0.4 mL, 0.8 mmol) is added drop by drop.
Afterwards, the light-yellow solution is stirred for 48 h at RT. The
solvent is removed in vacuum, and the product (4) (0.34 g, 0.65
mmol, yield 90%) is isolated via column chromatography (silica
latter hypothesis is supported by the fact that achiral secondary
groups on the surfaces (here the silyl-ester groups) have a direct
effect on the enantioselectivity of the reaction. A cooperative
effect between the catalytically active entity and the secondary
surface groups could be proven.
Experimental Section
General Comments. Starting compounds were received from
Sigma-Aldrich and were purified and dried prior to use. All
reactions on the precursor state were performed with the exclusion
of air and moisture using the Schlenck technique. Special caution
needs to be exercised when working with alkylated lithium
compounds because of their spontaneous ignition in air.
1
gel 60, PE/EE 3/1) as a colorless liquid. H NMR (400 MHz,
3
CDCl3): δ 1.18 (d, 36H, J ) 6.1 Hz, O-CH-CH3)2), 1.45 (d,
3H, 3J ) 6.4 Hz, sec-alk CH3), 4.24 (sept, 6H, 3J ) 6.1 Hz,
3
O-CH-(CH3)2), 4.87 (q, 1H, J ) 6.4 Hz, sec-alk CH), 7.65 (s,
2H, o-arom H), 7.91 (s, 1H, p-arom H). 13C NMR (100.61 MHz,
Synthesis. 1,3-Bis-tri-iso-propoxysilyl-5-bromobenzene (1) was
prepared according to the literature.25
CDCl3):
δ 24.1 (alk CH3), 24.5 (O-CH-(CH3)2), 64.4
1-(3,5-Bis-tri-iso-propoxysilylphenyl)-ethanol (2). tertBuLi (1.5
M, 3.5 mL, 5.3 mmol) is added dropwise to a solution of 1 (1.5 g,
2.65 mmol) in 50 mL of dry Et2O at T ) -78 °C. The mixture is
stirred for 30 min, followed by the addition of acetaldehyde (0.26
g, 5.4 mmol). After the mixture is warmed to room temperature,
the solvent is removed under vacuum. Dry pentane (60 mL) is
added. The nonsoluble residue is removed via centrifugation.
Finally, the product (2) (0.84 g, 1.59 mmol, yield 60%) is isolated
by column chromatography (silica gel 60, hexane/AcOEt 10/1 f
2/1) as a colorless liquid. 1H NMR (400 MHz, CDCl3): δ 1.14 (d,
36 H, 3J ) 6.1 Hz, CH(CH3)2), 1.47 (d, 3H, 3J ) 6.4 Hz, CHCH3),
4.25 (sept, 6H, 3J ) 6.1 Hz, O-CH(CH3)2), 4.89 (q, 1H, 3J ) 6.4
(O-CH-(CH3)2), 69.4 (alk CH), 131.0, 132.7, 139.8, 142.9 (arom.
C). IR (ATR) ν: 1040 (Si-O), 1112 (Si-C), 1369-1467 (aliph +
arom C-C), 2894-2972 (aliph + arom C-H), 3421 (O-H) cm-1
.
EI-MS (70 eV) m/z: 529 (M+), 514 (-Me), 471 (-OiPr), 205
(Si(OiPr)3). Determination of ee value: 4 (0.05 g, 0.095 mmol) is
dissolved in a mixture of CDCl3 (0.2 mL) and triethylamine (0.1
mL). Afterwards, a solution of PCl@(R)-(+)-1,1′-bi-2-naphtol
(0.033 g, 0.095 mmol) in CDCl3 (0.2 mL) is added drop by drop.
The reaction mixture is stirred for 30 min, and a 31P NMR spectrum
is recorded. See also SI-1. 31P NMR (161.98 MHz, CDCl3) δ: 142.9
(+ enantiomer); 149.8 (- enantiomer).
Mesoporous Organosilica Material UKON2g (5.3 nm Pore
Size) (5). The sol-gel precursor (4) (0.45 g, 0.88 mmol) and
Pluronic P-123 (0.26 g) are dissolved in ethanol (1.2 g). Aqueous
HCl (1 M, 0.26 g) is added drop by drop while stirring. The sols
are aged in an open container for around 1 week. The resulting
monolithic pieces are dried in vacuum at 100 °C for 24 h. The
template is removed by extraction with EtOH (25 mL) and aqueous
HCl (25 mL) at T ) 45 °C. Complete removal of the block
copolymer occurs within 2-4 days. 13C MAS NMR (100.61 MHz,
cptoss, 6 kHz) δ: 25.5 (CH3-sec alkohol), 70.3 (CH-sec alkohol),
131, 138, 144 (arom C). 29Si-MAS NMR (79.5 MHz, cp, 6 kHz):
δ -78.5 (typical T-type signals). IR (ATR) ν: 1058 (Si-O +
Si-C), 2881-2980 (aliph + arom C-H), 1264-1446 (aliph +
arom C-C), 3300 (OH) cm-1. N2 physisorption: BET surface area
)1045 m2/g; BJH pore diameter ) 5.3 nm. SAXS (P ) 4 kW, Cu
4
Hz, CHCH3), 7.72 (m, 2H, o-arom H), 7.92 (t, 1H, J ) 1,2 Hz,
p-arom H). 13C NMR (100.61 MHz, CDCl3): δ 25.0 (CHCH3), 25.4
(CH(CH3)2), 65.6 (O-CH(CH3)2), 70.5 (OCH-CH3), 131.9 (C-
arom C), 133.6 (Si-arom C), 140.8 (o-arom C), 143.7 (p-arom C).
IR (ATR) ν: 1040 (Si-O), 1111 (C-O), 1369-1467 (aliph + arom
C-C), 2894-2972 (aliph + arom C-H), 3421 (O-H) cm-1. EI-
MS (70 eV): m/z ) 529 (M+), 514 (-Me), 471 (-OisoPr), 205
(Si(OisoPr)3).
1-(3,5-Bis-tri-iso-propoxysilylphenyl)-ethanone (3). Pyridinium
chlorochromate (0.24 g, 1.33 mmol) is dissolved in dry CH2Cl2
(15 mL), and 2 (0.4 g, 0.76 mmol) is added. The mixture is stirred
at room temperature for 4 h while the color changes from orange
to dark brown. After the removal of the solvent in vacuum, the
product is purified via column chromatography (silica gel 60, 1.
CH2Cl2; 2. AcOEt). The product (3) (0.322 g, 0.608 mmol, yield
80%) is isolated as a colorless liquid. 1H NMR (400 MHz, CDCl3):
1.15 (d, 36H, 3J ) 6.1 Hz, CH(CH3)2), 2.54 (s, 3H, CHCH3), 4.21
KR): q100 ) 0.63 nm-1
.
UKON2g with 2 nm Pore Size. Materials were prepared
according to the literature using ꢀ-methyl-cyclodextrine (0.26 g)
as a template.36
3
(sept, 6H, J ) 6,1 Hz, O-CH(CH3)2), 8.13 (s, 1H, p-arom H),
8.24 (s, 2H, o-arom H). 13C NMR (100.61 MHz, CDCl3): δ 25.6
(CH(CH3)2), 26.7 (CH(CH3)2), 65.7 (O-CH(CH3)2), 132.0, 135.6,
136.4, 146.1 (arom C), 198.5 (CdO). IR (ATR) ν: 1013 (Si-O),
1116 (C-O), 1354-1369 (aliph + arom C-C), 1690 (CdO),
UKON2g with 7.8 nm Pore Size. A mixture of Pluronic P-123
(0.26 g) and 1,3,5-trimethylbenzene (0.125 g) is used as a template
phase.
2890-2972 (aliph + arom C-H) cm-1
.
(-)-1-(3,5-Bis-tri-isopropoxysilyphenyl)-ethanol (4). A solution
of (1R,2S)-(+)-cis-1-amino-2-indanol (0.29 g; 2.0 mmol) and
(36) Polarz, S.; Smarsly, B.; Bronstein, L.; Antonietti, M. Angew. Chem.,
Int. Edit. 2001, 40, 4417.
9
6564 J. AM. CHEM. SOC. VOL. 132, NO. 18, 2010

Enantioselective Catalysis Using Nanoporous Materials
A R T I C L E S
Thins Films of UKON2g. The sol used for the preparation of
UKON2g was aged for 24 h. One drop was spin coated onto a
thoroughly cleaned quartz slide (2.5 × 2.5 cm2) using a spin coater
Determination of ee Values. To 0.3 mL of the reaction mixture
from the asymmertric ene reaction is added 0.1 mL of triethylamine.
Afterwards, a solution of 0.033 g of PCl@(R)-(+)-1,1′-bi-2-naphtol
(0.095 mmol) dissolved in 0.2 mL CDCl3 is added drop by drop.
The reaction mixture is stirred for 30 min, and a 31P NMR spectrum
is recorded. 31P NMR (161.97 MHz, CDCl3, determination of ee)
δ 150.95 w (+) - enantiomer; δ 153.35 w (-) - enantiomer.
Analytical Methods. NMR spectra were acquired on a Varian
Unity INOVA 400 spectrometer using dried CDCl3 as a solvent.
Solid-state NMR spectra were recorded using a Bruker DRX 400
spectrometer. The following experimental parameters were used
for the measurements. 13C: We used a cross-polarization pulse
from LOT (model SCV-10) at a rotational speed of 5000 min-1
.
The coated quartz substrates were then transferred into an atmo-
spheric chamber at T ) 40 °C and room humidity R ) 50%. After
the condensation of the silica film, the material was dried at 100
°C and then used for CD measurements. The template was not
removed, and the films were checked regarding their thickness and
homogeneity using scanning electron microscopy (SEM). Further
characterization was not done.
Quenching of Silanol Groups in UKON2g. The modification
with SiMe3 is discussed as a representative case. The modification
with alternative groups SiR3 (R ) Et, isoPr) has been performed in
an analogous way. UKON2g (0.2 g, 0.91 mmol) is dried in vacuum
at 100°. Solvent Me3Si-O-SiMe3 (7.5 g) and Me3SiCl (5 g, 46.1
mmol) are added. The mixture is heated to reflux for 15 h, and the
solid material is separated via centrifugation. To remove the
remaining Me3Si-O-SiMe3 and Me3SiCl from the pore system,
the material is stirred in 40 mL of dry THF, followed by
centrifugation and drying in vacuum at 100 °C. To remove Me3Si
groups from the alcohol function, the material is stirred in 20 mL
of EtOH and 20 mL of trifluoroacetic acid for 15 h. Afterwards,
the material is isolated via centrifugation and dried in vacuum at
100 °C.
program, a spin rate of 6 kHz, a 5 s recycle delay, a 2 ms contact
π
time, and a
/
pulse width of 6.2 µs. 29Si: We used a cross-
2
polarization pulse program, a spin rate of 6 kHz, a 40 s recycle
π
delay, a 12 ms contact time, and a /6 pulse width of 2.2 µs. The
TEM images were performed on a Zeiss Libra 120 at 120 kV
acceleration voltage. The TEM samples were prepared by quickly
dipping a carrier covered with a holey carbon foil (Plano company,
S147) in the solution, where the ground UKON materials were
dispersed in THF. FT-IR spectra were recorded by using a Perkin-
Elmer Spectrum 100 spectrometer with an ATR unit. Small-angle
X-ray scattering (SAXS) measurements were conducted with a
Bruker AXS Nanostar. N2 physisorption measurements were
recorded on a Micromeritics Tristar. CD spectra were obtained on
a Jasco J-715 instrument.
Modification of UKON-2g with AlIII. Mesoporous organosilica
material UKON2g (0.2 g, 0.91 mmol) is dried in vacuum at 100
°C. The material is cooled to -78 °C, and AlMe2Cl (1 M in hexane,
5 mL, 5 mmol) is added. The reaction proceeds for 24 h at RT.
Then, the solid material is isolated by centrifugation. To remove
unreacted residues of AlMe2Cl from the pore system, the material
is stirred three times for 24 h in 60 mL of dry pentane. Finally, the
resulting material is dried in vacuum at 80 °C and stored under
inert conditions inside a glovebox.
Catalytic Tests. CDCl3 (0.5 mL) as a solvent and chlorale (0.2
g, 1.36 mmol) are added to the AlIII-modified mesoporous orga-
nosilica material (6) (30 mg). The mixture is cooled to -36 °C,
and R-methylstryrene (0.16 g; 1.36 mmol) is added drop by drop.
The reaction mixture is stirred for 24 h at different temperatures,
and the catalyst is filtered off. The products of the reaction are
studied by NMR spectroscopy. The results of the catalytic tests
are summarized in Table 1 shown in SI-6.
Acknowledgment. We thank the Zukunftskolleg University of
Konstanz and the Deutsche Forschungs-gemeinschaft (SP 780/6-
1) for funding.
Supporting Information Available: 31P NMR investigation
of 4 in comparison to racemic 2 using a chiral NMR shift
reagent. Additional analytical data for chiral mesoporous
organosilica material UKON2g. Investigation of the catalyzed
1
carbonyl ene reaction with H NMR spectroscopy. Different
possibilities for the mechanism of the ene reaction. N2 phys-
isorption measurements of chiral mesoporous organosilica
material UKON2g with two additional pore sizes. Summarized
catalytic data. This material is available free of charge via the
Internet at http://pubs.acs.org.
JA1017706
9
J. AM. CHEM. SOC. VOL. 132, NO. 18, 2010 6565