Intrazeolite assembly of a chiral manganese salen epoxidation catalyst

Article abstract of DOI:10.1039/a607879f

Asymmetric manganese salen epoxidation catalysts are assembled and trapped in a multistep synthesis in the cages of zeolite EMT; these heterogeneous catalysts produce high enantiomeric excess in the epoxidation of aromatic alkenes with NaOCl.

Full text of DOI:10.1039/a607879f

View Article Online / Journal Homepage / Table of Contents for this issue
Intrazeolite assembly of a chiral manganese salen epoxidation catalyst
Steven B. Ogunwumi and Thomas Bein*
Department of Chemistry, Purdue University, West Lafayette, IN 47907, USA
Asymmetric manganese salen epoxidation catalysts are
assembled and trapped in a multistep synthesis in the cages
of zeolite EMT; these heterogeneous catalysts produce high
enantiomeric excess in the epoxidation of aromatic alkenes
with NaOCl.
manganese salen complexes have dimensions of ca. 1.5 (1) and
1.3 nm (2; with modifications from crystallographic data of 1,
kindly provided by E. N. Jacobsen), preventing escape of the
assembled complex from within the zeolite (crystallographic
data for EMT: see ref. 8; Fig. 1). This general approach has been
termed ‘ship-in-the-bottle synthesis’.9 No close van der Waals
contacts were observed when the complexes were positioned in
the hypercages.
The Soxhlet-extracted samples contain 0.50 ± 0.09 Mn per
unit cell of EMT in the case of 1–EMT, and 0.65 ± 0.09 Mn per
unit cell in the case of 2–EMT (XRF data). The free intrazeolite
pore volume remaining after encapsulation of the complexes
was determined using T-plot analysis from N2 sorption, after
evacuating the Soxhlet extracted samples at 120 °C. The
original void volume of calcined EMT was reduced to 0.18
ml g21 in both 1–EMT and 2–EMT, showing that a significant
fraction of the pore volume remains accessible after assembly of
the intrazeolite salen complexes in spite of some water
remaining in the zeolite after degassing.
Asymmetric epoxidation of alkenes is of great interest for the
synthesis of chiral intermediates in the pharmaceutical and
agrochemical fields.1,2 Zeolite cage structures are attractive
hosts for the design of hybrid systems.3 Such an inclusion
strategy is expected to result in easy recovery or continuous use
of the catalyst, diffusional selectivity of the reaction, and
potentially increased stability via site isolation. Very few
asymmetric reactions have been performed on zeolites.4 Here
we present an active, asymmetric epoxidation catalyst that was
assembled and trapped in the cages of crystalline zeolite EMT.
This novel heterogeneous system produces high enantiomeric
excess (ee) in the epoxidation of aromatic alkenes with NaOCl
(aq).
Chiral manganese salen epoxidation catalysts5 are partic-
ularly active for the epoxidation of cis-substituted aryl alkenes.
This family of complexes can be assembled in a stepwise
manner in the cages of the zeolite EMT. This zeolite is a
hexagonal form of the well known faujasite structure.6,7 The
hypercages of EMT are accessible through three 12-ring
windows with free dimensions of 0.69 3 0.74 nm and two 0.74
nm circular apertures, while the hypocages have only three
windows. EMT was synthesized as previously reported,7
followed by calcination at 450 °C, washing with water, and
dehydration at 300 °C and 1024 Torr before use.
Electronic absorption spectra in Fig. 2 show ligand transi-
tions of the brownish oxidized complex 1 (in solution) at 322
and 438 nm and a weak d–d transition at 500 nm [Fig. 2(a)].10
Upon zeolite encapsulation, the former bands shift to 335 and
422 nm, while the weak d–d transition is still observed at about
500 nm [Fig. 2(b)]. Strikingly, the d–d transition at 500 nm
characteristic for MnIII in these complexes is absent in the
spectrum of 3–EMT [Fig. 2(c)], with bands at 266, 333 and 438
nm. In the IR spectrum (not shown), we observe the imine band
at 1638 cm21 and a mode at 1458 cm21 for the salen ligand of
1, assembled in the zeolite and dehydrated at 110 °C (1025 Torr)
One specific intrazeolite assembly sequence was found to
result in active and enantioselective heterogeneous epoxidation
catalysts. A mixture of 1.00 g (0.16 mmol) calcined and
dehydrated zeolite EMT and 1 equiv. [0.16 mmol (18.4 mg)] of
(R,R)- or (S,S)-1,2-trans-diaminocyclohexane per EMT unit
cell were stirred and refluxed in 80 ml of dry thf for 15 h under
N2. The solvent was evaporated, and 2.00 equiv. of the required
salicylaldehyde (2-hydroxybenzaldehyde or 3-tert-butyl-5-
methylsalicylaldehyde) were added and stirred at 80 °C under
N2 for 1 h, followed by refluxing under stirring in thf for 12 h.
The zeolite host turns yellow during this process. 1 Equiv. of
Mn(O2CMe)2·4H2O (39 mg) per unit cell of EMT was added to
the slurry, which was exposed to air to oxidize the MnII.
Another 15 h refluxing under air resulted in a brownish slurry,
to which was added 1 equiv. of solid LiCl per hypercage,
followed by 24 h under reflux in thf, filtration, drying at 80 °C
under vacuum, and Soxhlet extraction in CH2Cl2 for 24 h to
remove excess complex, salen ligand, and reactants from the
zeolite exterior and interior. Intrazeolite chloro[N,NA-bis(3-
tert-butyl-5-methylsalicylidene)cyclohexanediamine]mangan-
ese(iii) (1) is denoted 1–EMT, chloro[N,NA-bis(salicylidene)-
cyclohexanediamine]manganese(iii) (2) is denoted 2–EMT, and
chloro[N,NA-bis(3-tert-butyl-5-methylsalicylidene)cyclohexane-
diamine]manganese(ii) 3 (obtained by reacting intrazeolite
salen ligand with MnII under inert conditions) is denoted
3–EMT. Complexes 1 and 2 were also synthesized in the
homogeneous phase according to ref. 5.
Before assembly of the intrazeolite guest, the micropore
volume of degassed EMT is 0.325 ml g21 6
.
The chiral
Fig. 1 View of the molecular fit of 1 in EMT (Cerius2 software)
Chem. Commun., 1997
901

View Article Online
in situ to minimize interference with the zeolite water band at
1635 cm21. In solution, the imine CNN stretch vibration of the
free ligand at 1631 cm21 shifts to 1610 cm21 upon complexa-
tion with MnIII.10b,11 A striking shift of the imine band to 1583
cm21 occurs upon complexation with MnII and oxidation to
form 1 in the EMT host which shows that a large majority of the
ligands is now coordinating the Mn ions.
activity increases significantly on addition of axial ligands such
as pyridine N-oxide, with which conversion of cis-
b-methylstyrene was 47%. Formally, 100% conversion corre-
sponds to a lower limit of 20 turnovers, but only a small fraction
of the intrazeolite salen complexes may be accessed by the
substrate during reaction. (ii) With PhIO as oxidant, 2–EMT
showed conversion of styrene and was more active than the
sterically more encumbered 1–EMT, suggesting that the access
of the bulky terminal oxidant was hindered in the zeolite–1
assembly (not in Table 1). (iii) The highest ee in epoxide, 88%
was achieved with cis-b-methylstyrene, 1–EMT plus pyridine
N-oxide as catalyst, and NaOCl as oxidant, in dichloroethane.
(iv) trans-b-Methylstyrene gives relatively high yields of trans-
epoxide but the ee is lower. (v) Smaller (non-prochiral) alkenes
such as 2,3-dimethylbut-2-ene give higher conversions than the
more bulky but more reactive aromatic alkenes, suggesting a
strong influence of the zeolite pore structure on diffusional
access to the active salen complex encapsulated in the zeolite
host. Proof for intrazeolite reactions is obtained by reacting
cholesterol with the encapsulated 1–EMT catalyst. While no
conversion was observed with the encapsulated catalyst after 18
h, in solution the epoxidation proceeded to 13% conversion
during the same time. This behaviour shows that the catalysis of
the smaller substrates truly proceeds in the zeolite pores.
Further proof that the asymmetric epoxidation reaction was
heterogeneous was obtained with a separation experiment
(oxidation of cis-b-methylstyrene with NaOCl and 1–EMT, in
acetonitrile). After removal of the catalyst, no further activity
was exhibited in the filtrate, while in the presence of catalyst the
reaction proceeds for > 30 h.
The encapsulated Mn salen complexes were studied as
asymmetric epoxidation catalysts using different substrates
(Table 1). In a representative reaction, 1 ml of dichloromethane
or acetonitrile, 1 ml NaOCl (5.25%, 0.70 mmol), 0.005 ml
Na2HPO4 (0.05 m), 0.005 ml n-decane (GC standard) and 1 drop
of 1 m NaOH were added to a solution containing 0.15 mmol of
an alkene such as cis-b-methylstyrene, along with 100 mg of
1–EMT catalyst. The slurry was stirred at 4 °C for 1 h, followed
by stirring at room temp. for !23 h. The ratio of Mn to alkene
was typically adjusted to 5 mol% (commonly used in homoge-
neous reactions). No conversion was observed when using an
MnII-exchanged, air-treated zeolite EMT without the chiral
salen ligand. The X-ray crystallinity of the zeolite host is
retained after assembly of the intrazeolite complexes, and only
slightly reduced after completion of the catalytic experiments.
The following observations can be noted (see also Table 1).
(i) 1–EMT with the bulky ligand is more active than 2–EMT
(e.g. conversion of cis-b-methylstyrene with 1–EMT was 15%
(after 24 h), and with 2–EMT it was only 1%. The catalytic
1.2
The authors gratefully acknowledge funding from the U.S.
Department of Energy for this work. We thank Professor Merritt
B. Andrus for assistance with chromatographic separations.
0.8
(c)
(b)
References
0.4
1 R. A. Sheldon and J. K. Kochi, Metal-Catalysed Oxidation of Organic
Compounds, Academic Press, New York, 1981; J. Gorzynski
Smith, Synthesis, 1984, 629.
2 Li. Deng and E. N. Jacobsen J. Org. Chem., 1992, 57, 4320; D. M.
Hodgson, P. J. Parsons and P. A. Stones, J. Chem. Soc., Chem.
Commun., 1988, 217.
3 D. E. De Vos, J. L. Meinershagen and T. Bein, Angew. Chem., Int. Ed.
Engl., 1996, 35, 2211; P. P. Knops Gerrits, D. E. De Vos, F. Thibault-
Starzyk and P. A. Jacobs, Nature, 1994, 369, 543.
(a)
0.0
250 350 450 550 650 750
λ / nm
Fig. 2 UV–VIS spectra (in diffuse reflectance mode) of oxidized 1 in
solution (a), 1–EMT (b) and 3–EMT containing MnII complexed by the
salen ligand of 1 (c)
4 (a) S. Feast, D. Bethell , P. C. B. Page, F. King, C. H. Rochester,
M. R. H. Siddiqui, D. J. Willock and G. J. Hutchings, J. Chem. Soc.,
Chem. Commun., 1995, 2409; (b) A. Corma, M. Iglesias, C. del Pino and
F. Sanchez, J. Chem. Soc., Chem. Commun., 1991, 1253; (c) A.
Carmona, A. Corma, M. Iglesias, A. San Jose and F. Sanchez,
J. Organomet. Chem., 1995, 492, 11; (d) A. Corma, M. Iglesias, V. M.
Martin and F. Sanchez, Tetrahedron Asymmetry, 1992, 3, 845; (e) U.
Bo¨hmer, K. Morgenschweis and W. Reschetilowski, Catal. Today,
1995, 24, 195; (f) A. Corma, A. Fuerte, M. Iglesias and F. Sanchez,
J. Mol. Catal. A, 1996, 107, 225.
5 W. Zhang, J. L. Loebach, S. R. Wilson and E. N. Jacobsen, J. Am.
Chem. Soc., 1990, 112, 2801; T. Katsuki, Coord. Chem. Rev., 1995, 140,
189 and references therein.
6 E. J. P. Feijen, K. De Vadder, M. H. Bosschaerts, J. L. Lievens, J. A.
Martens, P. J. Grobet and P. A. Jacobs, J. Am. Chem. Soc., 1994, 116,
2950.
Table 1 Heterogeneous and homogeneous catalysis reactions of [Mn-
(salen)Cl] in EMTa
Epoxide
Substrate selectivity
Substrate
conv.(%) (%)
ee (%)
Confign.
Heterogeneous
Me2CNCMe2
PhC(H)NCH2
(E)-PhC(H)NCH(Me)
(Z)-PhC(H)NCH(Me)d
(Z)-PhC(H)NCH(Me)e
Cholesterol
40
15
29
15
47
None
75
87
62
67
58
—
NCc
34
20
80
88
NC
S-(2)
1S,2S-(2)
1S,2R-(2)
1S,2R-(2)
—
b
—
Homogeneous
PhC(H)NCH2
(Z)-PhC(H)NCH(Me)
Cholesterolf
55
85
13
95
97
85
35
80
—
S-(2)
1S,2R-(2)
—
7 S. L. Burkett and M. E. Davis, Microporous Mater., 1993, 1, 265.
8 J. L. Lievens, J. P. Verduijn, A.-J. Bons and W. J. Mortier, Zeolites,
1992, 12, 698.
9 N. Herron, Inorg. Chem., 1986, 25, 4714.
a
Salen
=
(S,S)-N,NA-bis(3-tert-butyl-5-methylsalicylidene)-
10 (a) M. T. Rispens, A. Meetsma and B. L. Feringa, Recl. Trav. Chim.
Pays-Bas, 1994, 113, 413; (b) J. Skarzewski, A. Gupta and A. Vogt,
J. Mol. Catal. A., 1995, 103, L63.
11 J. R. Dilworth, C. A. McAuliffe and B. J. Sayle, J. Chem. Soc., Dalton
Trans., 1977, 849.
cyclohexanediamine. (Other products and distribution not given.) Reagents
and conditions: 5 mol % catalyst, reaction maintained at 0 °C for 1 h then
room temp. for 24 h; solvent, CH2Cl2; oxidant, NaOCl; Astec chiraldex
B-TA column (decane: GC standard). b Reactivity of a non-prochiral alkene.
c Not chiral. d Initial experiments addressing the recyclability of the catalyst
after Soxhlet extraction with acetonitrile showed significantly lower ee; the
origin of this reduced enantioselectivity is currently being studied. e Catalyst
f
Received in Bloomington, IN, USA, 20th November 1996; Com.
6/07879F
with pyridine N-oxide (5 mol%); solvent dichloroethane. Isolated yield
determined by 1H NMR using CH2Br2 as internal standard (after 18 h).
902
Chem. Commun., 1997
Typeset and printed by Black Bear Press Lim ited, Cam bridge, England