Superior performance of a chiral catalyst confined within mesoporous silica

Article abstract of DOI:10.1039/a902441g

A chiral catalyst from the ligand 1,1'-bis(diphenylphosphino)ferrocene (dppf) anchored to the inner walls of the mesoporous support MCM-41 and co-ordinated to Pd(II) has been shown to exhibit a degree of regioselectivity and enantiomeric excess, in the allylic amination of cinnamyl acetate, which is far superior to that of its homogeneous counterpart or that of a surface-bound analogue attached to a non-porous silica.

Full text of DOI:10.1039/a902441g

View Article Online / Journal Homepage / Table of Contents for this issue
Superior performance of a chiral catalyst confined within mesoporous silica
Brian F. G. Johnson,*a Stuart A. Raynor,a Douglas S. Shephard,a Thomas Mashmeyer,ab John Meurig
Thomas,*b Gopinthar Sankar,b Stefen Bromley,b Richard Oldroyd,b Lynn Gladdenc and Mike D. Mantlec
a Department of Chemistry, The University of Cambridge, Lensfield Road, Cambridge, UK CB2 1EW.
E-mail: bfgj1@cam.ac.uk
b Davy-Faraday Research Laboratories, The Royal Institution of Great Britain, 21 Albemarle Street, London, UK
W1X 4BS
c Department of Chemical Engineering, The University of Cambridge, New Museum Site, Cambridge, UK CB2 3QZ
Received (in Cambridge, UK) 26th March 1999, Accepted 12th May 1999
A chiral catalyst from the ligand 1,1A-bis(diphenylphosphi-
no)ferrocene (dppf) anchored to the inner walls of the
mesoporous support MCM-41 and co-ordinated to PdII has
been shown to exhibit a degree of regioselectivity and
enantiomeric excess, in the allylic amination of cinnamyl
acetate, which is far superior to that of its homogeneous
counterpart or that of a surface-bound analogue attached to
a non-porous silica.
a degree of catalytic regioselectivity as well as an enantiomeric
excess that is far superior to either the homogeneous counterpart
or the cabosil-bound catalyst (Cabosil is a non-porous, high-
area silica).
Our conceptual methodology was to take a homogeneous
system of known catalytic behaviour and, by performing
suitable modifications, to tether this catalyst within the
mesopore.11 Care is taken to ensure that all activity is confined
to the internal surface of the MCM-41 channels. This is
achieved by selectively deactivating the external surface of the
support. We have shown previously by electron microscopy, on
functionalised mesopores stained with Ru6-based clusters, that
this is a reliable and satisfactory method.12
Our approach to the preparation of the catalytic system is
shown in Scheme 1. The mesoporous framework was first
treated with Ph2SiCl2 under non-diffusive conditions to deacti-
vate the exterior walls of the material. The interior walls of this
same material were then derivatised with 3-bromopropyltri-
chlorosilane to give the activated MCM-41 1.
The ferrocenyl-based ligand, (S)-1-[(R)-1A,2-bis(diphenyl-
phosphino)ferrocenyl]ethyl-N,NA-dimethylethylenediamine 2,
was prepared from ferrocene by literature methods.13,14 On
Stereo- and regio-selective catalysis lies at the heart of current
developments in pharmaceutical, agrochemical and cognate
industries.1–3 The recognition that a small amount of a chiral
catalyst may yield large quantites of a targeted chiral product
has already led to the development of several ingenious
methods for the design of such catalysts. These include
modified metals,4 metal oxides5 and zeolites,6,7 and the so-
called ‘ship-in-bottle’ or ‘tea-bag’ systems.8–10 In this work we
have been able to demonstrate that a chiral ligand derived from
1,1A-bis(diphenylphosphino)ferrocene (dppf) bonded to an ac-
tive metal centre (in this case PdII)† and tethered, via a
molecular link of appropriate length, to the inner walls of a
mesoporous silica support (MCM-41, ca. 30 Å diameter) yields
Scheme 1 The synthetic routes used in the preparation of the homogeneous catalyst 3, the mesopore-supported 6 and the carbosil-bound 7. Reagents and
conditions: formation of 1(a) Ph2SiCl2–THF, 25 °C, (b) Cl3SiCH2CH2CH2Br–THF, 278 °C; i, THF, 25 °C; ii, THF, 25 °C; iii PdCl2–MeCN–THF, 25 °C;
iv PdCl2–MeCN–THF, 25 °C.
Chem. Commun., 1999, 1167–1168
1167

View Article Online
Table 1 Catalytic results
Catalysta Conversionb (%) Straight chainc (%) Branched (%) eed (%)
3 (S)
7 (S)
6 (S)
6 (R)
76
98
99+
99+
99+
98
49
—
2
51
50
—
43
99+
93
50
b
a Symbols in parentheses denote chirality of the directing group.
%
Conversion is stated relative to the use of benzylamine. c Regio- and
enantio-selectivity determined by gas chromatography on a Chiraldex G-
DA column (Alltech) with g-cyclodextrin as the active phase (20 m).
Conditions: He pressure 12.5 psi; temperature ramped 50–180 °C at 10 °C
min21 and then held for the duration of the run. d Major stereoisomer
possesses the same chirality as the catalyst. Retention time ca. 38 min.
(ee). Three catalysts were examined: the homogeneous {(S)-
1-[(R)-1A,2-bis(diphenylphosphino)ferrocenyl]ethyl-N,NA-di-
methylethylenediamine} palladium dichloride 3, the cabosil-
supported one 7 and that attached within the mesoporous
material 6. Preliminary results of the studies are listed in Table
1. With the homogeneous catalyst 3 the reaction is directed
solely towards the straight chain product whilst catalyst 7 shows
some of the desired regioselectivity by producing 2% of the
branched product. Unfortunately the enantioselectivity of the
reaction is relatively low with an ee of 43%. The use of the
mesopore-confined catalyst 6 promotes a dramatic change in the
regioselectivity of the reaction, producing 51% of the branched
product. The enantioselectivity of the catalysis is also greatly
improved relative to the cabosil-bound catalyst with the ee
approaching 100%.
These results indicate that the control exercised by the MCM-
41 on the activity of the ferrocenyl catalyst is considerable. The
profound changes in the regio- and enantio-selectivity are
clearly apparent from the data listed in Table 1. Other, related
systems are currently under investigation.
This work was supported by the EPSRC, ICI (Wilton) and the
Newton Trust (S. A. R.), a Royal Society Fellowship and
Peterhouse (D. S. S).
Fig. 1 Computer model of the catalytic centre inside MCM-41.
treatment of the activated MCM-41 1 with an excess of 2 yields
the chiral catalytic precursor 4 which, on reaction with PdCl2–
MeCN, gives the required catalyst 6. In a separate experiment
the closely related cabosil-supported catalyst 7 was prepared in
a similar manner from surface-activated cabosil and 2 followed
by addition of the required palladium(ii) salt.
The catalyst 6 was fully characterised by a range of
techniques. First, the structural integrity of the MCM-41
catalytic precursor 4 was established by MAS NMR and
EXAFS spectroscopy. The presence of the tethered ferrocenyl
catalysts was confirmed by comparison of the 13C MAS NMR
spectrum of 4 with that of the unattached precursor 2. Apart
from the additional signals arising from the propyl tethering unit
the spectra were essentially the same. Examination of the 31P
MAS NMR of 2 and 4 showed identical chemical shift values.
On incorporation of the PdII ion to yield the active catalyst 6,
significant changes in the aliphatic region of the 13C MAS NMR
and in the 31P MAS NMR spectra were noted. Two 31P
resonances at d 15.9 and 34.4 were recorded, clearly indicating
two different phosphorous environments, these have been
assigned as one being trans to a nitrogen and the other being
trans to a chloride.15 The change in the aliphatic region of the
13C MAS NMR corresponds to changes in the methyl
resonances of the amines, further corroborating the assignment
of the 31P spectrum. This entire arrangement was borne out by
a detailed EXAFS analysis of 6. The coordination environment
of the cationic species within the mesopore is depicted in
Fig. 1.
Notes and references
† Other metal ions, e.g. Rh(i) may also be employed.
1 R. Noyori, Asymmetric Catalysis in Organic Synthesis, Wiley, New
York, 1994.
2 J. M. Brown and S. G. Davies, Nature, 1994, 370, 418.
3 Advances in Catalytic Processes: Asymmetric Chemical Transforma-
tions, ed. M. Boyle, JAI, Greenwich, CT, vol. 1.
4 A. Pfalz and T. Heinz, Top. Catal., 1997, 4, 229.
5 A. Baiker, Curr. Opin. Solid State Mater. Sci., 1998, 3, 86.
6 Top Catal., ed. D. G. Blackmond and W. Leitner, 1997, 4, Part 1.
7 A. Corma, M. Iglesias, C. del Pino and F. Sanchez, J. Chem. Soc., Chem.
Commun., 1991, 1253.
8 J. M. Thomas, Philos. Trans. R. Soc. London, Ser. A, 1990, 333, 173.
9 C. L. Hill, Activation and Functionalisation of Alkanes, Wiley, New
York, 1989.
10 M. J. Sabater, A. Corma, A. Domenech, V. Fornes and H. Garcia, Chem.
Commun., 1997, 1285.
11 J. M. Thomas, T. Maschmeyer, B. F. G. Johnson and D. S. Shephard,
J. Mol. Catal., 1999, 141, 139.
12 D. S. Shephard, W. Z. Zhou, T. Mashmeyer, J. M. Matters, C. L. Roper,
S. Parsons, B. F. G. Johnson and M. J. Duer, Angew. Chem., Int. Ed.,
1998, 37, 2719.
In testing our catalysts we decided to use an allylic amination
reaction between cinnamyl acetate and benzylamine (Scheme
2). This reaction has two possible products: a straight chained
product (which is favoured due to the retention of the
delocalised p system) and a chiral branched product. The aim of
the reaction is to produce the greatest possible yield of the
branched product with the highest possible enantiomeric excess
13 G. W. Gokel and I. K. Ugi, J. Chem. Educ., 1972, 49, 294.
14 T. Hayashi, T. Mise, M. Fukushima, M. Kagotani, N. Nagashima, Y.
Hamada, A. Matsumoto, S. Kawakami, M. Konishi, K. Yamamoto and
M. Kumada, Bull. Chem. Soc. Jpn., 1980, 53, 1138.
15 T. Suzuki, M. Kita, K. Kashiwabara and F. Fujita, Bull. Chem. Soc. Jpn.,
1990, 63, 3434; P. J. Stang, D. H. Cao, S. Siato and A. M. Arif, J. Am.
Chem. Soc., 1995, 117, 6273.
Ph
OAc
Ph
NHCH2Ph
THF, 40 °C
+
+
*
PhCH2NH2
Ph
NHCH2Ph
Scheme 2 The catalytic reaction between cinnamyl acetate and benzyla-
mine.
Communication 9/02441G
1168
Chem. Commun., 1999, 1167–1168