Synthesis and application of layered double hydroxide-hosted catalysts for stereoselective epoxidation using molecular oxygen or air

Article abstract of DOI:10.1016/j.jcat.2004.04.008

A novel chiral sulfonato-salen manganese(III) complex has been intercalated into a Zn(II)-Al(III) layered double hydroxide (LDH) host to produce a stable, active, and selective heterogeneous epoxidation catalyst. Powder X-ray diffraction, TGA, and IR and

Full text of DOI:10.1016/j.jcat.2004.04.008

Journal of Catalysis 225 (2004) 398–407
www.elsevier.com/locate/jcat
Synthesis and application of layered double hydroxide-hosted catalysts
for stereoselective epoxidation using molecular oxygen or air
∗
Samiran Bhattacharjee, Trevor J. Dines, and James A. Anderson
Surface Chemistry and Catalysis Group, Division of Physical and Inorganic Chemistry, University of Dundee, Dundee, DD1 4HN, Scotland, UK
Received 10 March 2004; revised 6 April 2004; accepted 7 April 2004
Available online 25 May 2004
Abstract
A novel chiral sulfonato-salen manganese(III) complex has been intercalated into a Zn(II)–Al(III) layered double hydroxide (LDH) host
to produce a stable, active, and selective heterogeneous epoxidation catalyst. Powder X-ray diffraction, TGA, and IR and UV–visible spec-
troscopies confirm the successful intercalation of the Mn complex within the LDH structure. Molecular modeling calculations predict the
orientation of the complex within the layered structure. The intercalated sulfonato-salen–manganese(III) complex was found to be effective
in the stereoselective epoxidation of R-(+)-limonene and (−)-α-pinene at room temperature. At close to 100% conversion, R-(+)-limonene
was converted to the corresponding epoxide with good selectivity and de (diastereomeric excess), whereas (−)-α-pinene was converted with
91.7% selectivity and 98.1% de. A significant advantage of this catalyst over the equivalent homogeneous counterpart was the ability to
activate oxygen or air at atmospheric pressures in place of NaOCl. The catalyst could be recycled without loss of efficiency.
2004 Elsevier Inc. All rights reserved.

Keywords: Chiral salen–manganese(III); Layered double hydroxide; LDH host catalyst; R-(+)-Limonene; (−)-α-Pinene; Heterogeneous air epoxidation
1
. Introduction
pure epoxides [1]. Among these, epoxidation of the ap-
propriate alkene catalyzed under homogeneous conditions
by chiral salen–Mn(III) complexes are an efficient synthe-
sis method [2]. Collman et al. [3] showed that cytochrome
P450 analogs (manganese porphyrin complexes) catalysed
the epoxidation of olefins while chiral iron–porphyrin com-
plexes show high enantioselectivities in the epoxidation of
terminal alkenes [4]. Unfortunately, these modified por-
phyrins are rather difficult to synthesize. Jacobsen et al. [5]
reported catalytically active compounds prepared from more
readily synthesized Mn(III) salen complexes that gave high
enantioselectivities for the epoxidation of unfunctionalized
alkenes under homogeneous conditions and using sodium
hypochlorite as oxidant. Despite high activity, selectivity,
and chiral induction, this system has several disadvantages
including the use of sodium hypochlorite and the difficulty
in recovering the catalyst after use. In an attempt to over-
come the problem of catalyst recovery, one approach has
been to attach the active system to zeolites [6], to silica via
chloropropyl spacers, metallated with manganese [7] or us-
ing organo-functionalized mesoporous materials [8]. These
anchored systems have shown reasonable properties in the
epoxidation of alkenes. Kureshy et al. [9] recently reported
Much attention has been paid over recent years to the
establishment of ecologically acceptable processes in the
chemicals industries. Approaches involve reducing the ex-
tent of by-product formation by using heterogeneous cat-
alysts in place of stoichiometric reagents and employing
reagents which maximise atom efficiency. One example of
the former would be the use of solid acid catalysts to re-
place corrosive, liquid mineral acids for reactions such as
isomerisation and alkylation. Maximising atom efficiency in,
for example, an oxidation reaction might be achieved by the
use of oxygen or air as an oxidising agent instead of the use
of NaOCl, H2O2 or peroxy acids.
One important oxidation reaction employed in industry
as a route to the formation of useful intermediates is epox-
idation. Additionally, some oxygenated monoterpenes are
currently used in the flavors and perfumes industries. In
recent years, much attention has been given to the devel-
opment of routes for the preparation of enantiomerically
*
Corresponding author. Fax: +44 1382 345517.
E-mail address: j.a.anderson@dundee.ac.uk (J.A. Anderson).
0021-9517/$ – see front matter  2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2004.04.008

S. Bhattacharjee et al. / Journal of Catalysis 225 (2004) 398–407
399
immobilization of dicationic chiral Mn (III) complexes be-
tween the layers of a montmorillonite clay. This system was
tested in the epoxidation of nonfunctionalized alkenes using
NaOCl as oxidant and gave a higher ee for certain alkenes
compared to that found under homogeneous conditions.
Among the different layered structures which might be
considered as potential hosts, hydrotalcite-based materials
have received attention over recent years in applications such
as anion exchangers and catalysts [10–16]. Studies on the
intercalation of transition-metal complexes within hydro-
talcite layers are less abundant, although intercalation of a
dioxomolybdenum(VI) anion into a Zn(II)–Al(III) layered
double hydroxide host was reported to show useful catalytic
properties in the oxidation of thiols [17] and reduction of
nitrobenzene [18]. Recent advances in the pillaring of hy-
drotalcites by polyoxometalate anions have demonstrated
that the gallery height may be increased sufficiently to allow
oxidation of larger dimension organic compounds [19,20].
We have recently reported [21] that chiral sulfonato-salen–
sure of molecular oxygen. While details of the catalytic ac-
tivities with other alkenes using molecular oxygen and other
oxidants are currently being pursued [22], the present paper
deals with the synthesis and characterisation of the LDH,
[Zn2.15Al0.86(OH)6.02][Mn(Cl)(L)]0.19[C H COO]
0.48
6
5
·
2H O as well as its action as a heterogeneous catalyst in the
2
epoxidation of (R)-limonene and (−)-α-pinene by molecu-
lar oxygen as well as in air at atmospheric pressure and using
a range of benign solvents. To our knowledge, this represents
the first report of the successful use of a heterogenised sys-
tem containing a chiral salen–manganese(III) complex for
the epoxidation of (R)-limonene and (−)-α-pinene at room
temperature using air at atmospheric pressure.
2
. Experimental methods
2
.1. Preparation of metal complex catalyst
II
III
manganese(III) complex intercalated into a Zn –Al lay-
ered double hydroxide showed high conversion, selectivity,
and de in the oxidation of (R)-limonene using elevated pres-
A scheme of the preparation procedure is shown in
Scheme 1.
Scheme 1. (i) Ref. [24]; (ii) Ref. [23]; (iii) 5.54 g of 1 and 2.97 g of 2; (iv) 1.72 g Mn(O CMe) · 4H O and 1.75 g of 3, water; (v) saturated NaCl solution;
2
2
2
(
vi) 5.0 g LDH-[C H COO] and 1.71 g of 5, water, and stirred.
6 5

400
S. Bhattacharjee et al. / Journal of Catalysis 225 (2004) 398–407
2
.1.1. Preparation of sulfonato-salen ligand, H2L (3)
2.1.4. Preparation of LDH-[Mn(Cl)(L)] (6)
The LDH-[Mn(Cl)(L)] was obtained by the partial substi-
tution of intercalated C6H5COO ions by the [Mn(Cl)(L]
(
R,R)-1,2-Diammoniumcyclohexane mono-(+)-tartrate
2
−
(
(
2) from trans-diaminocyclohexane (99%, Aldrich) and
2R,3R)-(+)-tartaric acid (99.5%, Aldrich) and sodium
ions. The LDH-[C6H5COO] was prepared by mixing a solu-
tion of zinc(II) nitrate tetrahydrate (29.75 g) (98%, Aldrich)
and aluminum(III) nitrate (12.50 g) (99.997%, Aldrich) in
decarbonated water, together with a further separate so-
lution prepared by dissolving benzoic acid (21.96 g) and
NaOH (15.60 g) in decarbonated water under a nitrogen
atmosphere. The gel-like mixture was digested at 348 K
for 62 h. Upon cooling, the product was isolated by filtra-
tion, washed with water and ethanol, and dried overnight at
salicylaldehyde-5-sulfonate (1) from salicylaldehyde (98%,
Aldrich) and aniline (99.5%, Aldrich) were prepared using
published methods [23,24]. A mixture of 2.97 g of (R,R)-
1
,2-diammonium cyclohexane mono-(+)-tartrate and 3.12 g
of potassium carbonate were combined with 20 ml water–
ethanol (1:4) into a two-necked round-bottomed flask with
reflux condenser and an addition funnel. The mixture was
heated and stirred with a magnetic stirrer. A solution of
sodium salicylaldehyde-5-sulfonate (5.54 g) in 20 ml of wa-
ter was added dropwise to the above solution through an
addition funnel with constant stirring and gentle heating.
The resulting mixture was refluxed for 1 h with stirring and
cooled to room temperature. The volume was reduced by
3
33 K. Yield: 90%.
Na2[Mn(Cl)(L)] (1.71 g) was dissolved in decarbonated
water and LDH-[C6H5COO] (5.0 g) was added to the solu-
tion and stirred for 10 h at room temperature under a nitro-
gen atmosphere. The pale green product was filtered off and
washed with water and dried at overnight 333 K. Yield: 90%.
5
0% by rotary evaporation until a yellow solid was sepa-
rated, which was filtered off, and washed with ethanol. The
yellow solid was then recrystallised from water–diethyl ether
mixture and dried over silica gel. Yield: 98%.
2
.1.5. Catalyst characterisation
IR spectra of samples were recorded using KBr disks
on a Perkin-Elmer 1720X FTIR spectrometer and electronic
spectra on a Perkin-Elmer UV/VIS Spectrometer Lambda
16. The X-ray powder diffraction patterns were recorded on
a Philips PW 1010 X-ray generator with Cu-Kα (1.5402 Å)
radiation at 1 min . Thermogravimetric analyses, in the
range 298–971 K, were carried out on a Mettler Toledo Star
2
.1.2. Preparation of Na2[Mn(OAc)(L)] · 2H2O (4)
To an aqueous solution (30 ml) of Mn(O2CMe)2 · 4H2O
o
−1
(
1.72 g) (99+%, Aldrich), an aqueous solution (20 ml) of the
e
ligand (1.75 g) was added dropwise while stirring. Stirring
was continued for 1 h and the mixture left standing for 2 h.
The green solid was separated and filtered and washed with
cold water and ethanol and dried over silica gel. Yield: 95%.
System under a flow of air and a heating rate of 5 K min 1.
Manganese and aluminum concentrations were estimated by
literature methods [25] using UV–vis and fluorescence spec-
troscopies, respectively, while zinc was determined using a
UNICAM 939/959 atomic absorption spectrometer. Charac-
terisation details are given in Table 1.
−
2
.1.3. Preparation of Na2[Mn(Cl)(L)] · 2H2O (5)
A saturated aqueous solution of sodium chloride (3 ml)
was added to the mixture after the complete addition of lig-
and [as describe for Na2[Mn(OAc)(L)] · 2H2O] and stirring
was continued for 1 h and the mixture allowed to stand for
2.2. Catalytic reaction
Epoxidation of (R)-limonene and (−)-α-pinene over the
LDH-[Mn(L)(Cl)] catalyst with molecular oxygen or air was
carried out in a twin-necked round-bottom flask equipped
with condenser. In a typical run, 3.70 mmol of (R)-limonene,
2
h. The green solid was separated and filtered and washed
with cold water and ethanol and dried over silica gel. Yield:
7%.
9
Table 1
Characterisation of catalyst precursors, the LDH-host precursor, and hosted Mn(III) complex
a
b
Compound
Analytical data (%)
[Mn(Cl)(L)]
λ
max
2
(
%)
(nm)
C
H
N
Mn
–
Zn
–
Al
–
H L (3)
45.72
3.74
5.36
–
–
2
(
(
(
45.61)
38.93
39.17)
35.25
35.16)
(3.84)
3.76
(3.74)
3.27
(5.32)
4.10
(4.15)
3.98
Na [Mn(OAc)(L)] · 2H O (4)
8.23
(8.15)
8.20
–
–
–
–
800, 574, 465,
363, 295, 250
798, 576, 466,
365, 295, 250
277, 234
2
2
Na [Mn(Cl)(L)] · 2H O (5)
2
2
(3.22)
(4.11)
(8.05)
LDH-[C H COO]
34.61
5.58
6
5
(
(
34.75)
29.03
29.14)
(5.67)
4.69
(4.81)
LDH-[Mn(Cl)(L)] (6)
24.88
(25.06)
799, 552, 446,
350, 270, 226
a
Calculated values are shown in parentheses.
In Nujol mull.
b

S. Bhattacharjee et al. / Journal of Catalysis 225 (2004) 398–407
401
9
.2 mmol pivalaldehyde, 1.78 mmol N-methylimidazole,
cipitation of a solution of zinc and aluminum nitrates with an
aqueous solution of NaOH and benzoic acid. [Mn(Cl)(L)]
2
−
and 0.100 g of catalyst (0.85% Mn) were stirred at room tem-
perature (298 K) while bubbling molecular oxygen or air at
atmospheric pressure. After completion of the reaction, the
catalyst was filtered off and the selectivity and conversion
were measured using a Perkin-Elmer GC Autosystem XL
fitted with 10% SP 2330(100/120)Supelcoport capillary col-
umn and a FID. The reaction products were analysed using
a Hewlett Packard GC/MS (with FID also) after separation
over a CYDEX-B fused silica chiral column.
compound was intercalated into the zinc(II)–aluminum(III)
layered double hydroxide at room temperature from aque-
ous medium by the exchange of the benzoate ion. Although
complete elimination of the benzoate ion was the objective,
only partial substitution by the manganese compound was
achieved.
III
2−
3
.3. Characterisation of [Mn -(L)(X)] and
LDH-[Mn(Cl)(L)] catalysts
2
.3. Computational details
The free Schiff-base ligand exhibits strong bands at 1110
The structure of the chiral (sulfonato–salen)–manganese
III) complex was calculated using the Gaussian 98 pro-
−1
and 1035 cm in the IR spectrum (Fig. 1A) due to the an-
(
−
tisymmetric and symmetric stretching modes of the SO3
gram [26], with the hybrid SCF-DFT method B3-LYP.
This incorporates Becke’s three-parameter Hybrid Func-
tional [27] and the Lee, Yang, and Parr correlation func-
tional [28], and generally gives superior results to HF-SCF
calculations, which neglect the effects of electron correla-
tion. There is no published crystal structure of this complex
moiety [32] and these are slightly blue-shifted in the Mn(III)
for comparison, so the structure of a related complex, chloro-
ꢀ
(
1R,2R)-(−)-(1,2-cyclohexanediamino-N,N -bis(3,5-di-t-
butyl-salicylidene))–manganese(III) [29] was used to pro-
vide an initial estimate of the geometry prior to optimisation.
Geometry optimisation was performed with the LanL2DZ
basis set, which employs Dunning–Huzinaga double zeta
(
DZ) basis functions [30] on C, H, N, and O and Los Alamos
effective core potential with DZ functions on Mn, Cl, and
S [31]. This effective core potential basis set was judged to
be a reasonable compromise in view of the large size of the
complex.
(A)
3
. Results and discussion
III
2−
3
.1. Synthetic aspect of [Mn –(salfonato-salen)(X)]
The synthetic route of chiral (sulfonato-salen)–manga-
nese(III) and its LDH compound is shown in Scheme 1. The
chiral Schiff base ligand (3) was readily prepared as a beau-
tiful crystalline yellow solid in 98% yield by refluxing two
equivalents of sodium salicylaldehyde-5-sulfonate (1) with
(
R,R)-1,2-diammoniumcylohexane mono-(+)-tartrate (2)
in a water–ethanol medium. In aqueous solution, the chiral
ligand instantly reacted with manganese(II) acetate tetrahy-
drate to produce dianionic manganese(III) (4) compound in
95% yield. The acetyl ligand in [Mn(OAc)(L)]2 [L
−
2−
=
∼
chiral salen (sulfonato)] was readily replaced by chloride at
room temperature to give [Mn(Cl)(L)]2 (5) with good yield
−
(
97%). The manganese compounds are water soluble.
(B)
3
.2. Synthesis of LDH-[Mn–Salen]
Fig. 1. (A) IR spectra of (a) free sulfonato-salen, (b) Na [Mn(Cl)(L)], and
2
The host hydrotalcite-like material, ZnAl–[C6H5COO]
was prepared (see experimental section for details) by copre-
(c) Na [Mn(OAc)(L)] and (B) (a) Na [Mn(Cl)(L)], (b) LDH-[Mn(Cl)(L)],
2
2
and (c) LDH-[C H COO].
6 5

402
S. Bhattacharjee et al. / Journal of Catalysis 225 (2004) 398–407
complexes (Fig. 1A). The NNOO (donors of ligand function-
alities) mode of metal coordination, i.e., through the pheno-
lic oxygen and azomethine nitrogen of the salen ligand was
in the gallery height may be indicative of the intercala-
2
−
tion of the [Mn(L)(Cl)] ion. The calculated dimension of
ꢀꢀ
chloro[N,N -bis(salicylidene)-cyclohexanediamine]-man-
ganese-(III) is 13.0 Å [35]. Elemental analysis (Table 1) for
the catalyst is consistent with the formula unit [Zn2.15Al0.86-
(OH)6.02][Mn(Cl)(L)]0.19[C6H5COO]0.48 · 2H2O. The ratios
[Al/(Zn + Al)] of catalyst 6 and LDH-[C6H5COO] are al-
−1
apparent from the blue shift (ca. 23 cm ) of the ν(C–O)
−
1
and the red shift (ca. 12 cm ) of the ν(C=N) vibrations
−1
of the free ligand (1522 and 1632 cm respectively) [33].
Further support for this coordination mode is provided by
bands at ca. 574 and ca. 425 cm due to the ν(Mn–O)
and ν(Mn–N), respectively [33]. The IR spectrum (Fig. 1B)
of catalyst 6 shows bands at 1116 and 1032 cm due to
the presence of the sulfonato group and at 573 cm for
ν(Mn–O), indicating the presence of the [Mn(Cl)(L)]2 ion
−1
II
III
most identical, indicating that no loss of either Zn or Al
occurred during the exchange procedure.
−1
The thermogravimetric analysis and first derivatives
curves of LDH-[Mn(L)(Cl)], LDH-[C6H5COO], and the
free manganese(III) complex are shown in Fig. 3. The mass
loss behaviour of the latter showed that decomposition of
the complex occurred between 515 and 583 K (Fig. 3A).
A similar mass loss was found for catalyst 6 (Fig. 3C), but
no weight loss was observed for the LDH-[C6H5COO] be-
tween 503 and 583 K (Fig. 3B). This mass loss results from
decomposition of the manganese complex and is consistent
with the presence of the Mn(III) complex 5 within the LDH
host.
−
1
−
within the layered double hydroxide. Bands in the region
−1
1
650–1500 cm (Fig. 1B) are not readily assignable due
to the coexistence of carboxylato and phenyl ring vibration
−1
in LDH-[C6H5COO]. A broad band around 3430 cm in
the spectra of both LDH-[C6H5COO] and LDH-[Mn(L)(Cl)]
is attributed to the stretching mode of hydrogen-bonded hy-
droxyl groups and water molecules.
The X-ray powder diffraction patterns (Fig. 2) of LDH-
5
[
C6H5COO] and LDH-[Mn(Cl)(L)] showed that the basal
The field-free D term of manganese (III) in weak Oh
5
2
spacing of the LDH-[C6H5COO] was increased from 15.22
to 18.78 Å following the exchange process. The gallery
height of the catalyst is 14.1 Å when the thickness of
the brucite layers (4.7 Å) was subtracted. This increase
field is split into a Eg ground and a Tg excited state. Since
the ground state is an orbital doublet, Jahn–Teller distor-
tion becomes pronounced giving rise to a five-coordinate
square pyramidal situation. To a reasonable approximation,
Fig. 2. X-ray powder patterns of (a) LDH-[C H COO] and (b) LDH-[Mn(Cl)(L)].
6
5

S. Bhattacharjee et al. / Journal of Catalysis 225 (2004) 398–407
403
(A)
(B)
(C)
Fig. 3. TGA (upper) and first derivative curves (lower) of (A) Na [Mn(Cl)(L)], (B) LDH-[C H COO], and (C) LDH-[Mn(Cl)(L)].
2
6 5
ꢀ
ꢀ
the energy level diagram is as in Fig. 4, which dictates the oc-
chloro[N, N -bis(salicylidene)-cyclohexanediamine]-man-
ganese-(III) trapped in the cages of the synthetic zeo-
lite EMT is claimed to be 13.0 Å [34]. This structure is
significantly distorted at the Mn(III) centre, and presum-
ably results from a computer-generated fit of the complex
into the zeolite host. By contrast, the published crystal
structure of chloro-(1R,2R)-(−)-(1,2-cyclohexanediamino-
5
5
currence of three d–d transitions assignable to B2 ← B1,
5
5
5
5
A1 ← B1 and E ← B1 [33]. The electronic spectra
Fig. 5) of the free manganese(III) complex (4) (in Nujol
mull) displayed bands at 800, 574, 465, 363, 295 and 250 nm
Table 1). Of these, bands at 800 and 574 nm are due to tran-
(
(
5
5
5
5
sitions B2 ← B1 and A1 ← B1 [33,35]. The UV–visible
spectrum of catalyst 6 (in Nujol mull) showed similar fea-
tures (Table 1) to the free complex, indicating that following
intercalation, no change to the manganese (III) coordination
centre took place.
ꢀ
N,N -bis(3,5-di-t-butyl-salicylidene))-manganese(III) [29]
has dimensions which are comparable with our calculated
structure.
The experimental observation from X-ray powder diffrac-
tion of the LDH-[Mn(Cl)(L)] catalyst showed d = 18.78 Å
in height. The other phases can also be confirmed by the
−
3
.4. Structure calculation of [Mn(L)(Cl)]2 and
◦
LDH-[Mn(Cl)(L)]
medium intensity peaks at 8.6, 13.8, and 19.7 (2θ/ ) for the
product. In the case of the polyoxometalate pillared layered
double hydroxides [36], the interlayer distances (observed
from X-ray powder diffraction) were smaller than the numer-
ical sum of the Keggin in diameter and the ideal thickness of
the hydroxide layer due to strong interaction between the ba-
A projection view of the [Mn(L)(Cl)]2 ion from the B3-
LYP/LanL2DZ calculation is shown in Fig. 6. This calcula-
tion revealed that the dimension of the free manganese(III)
complex is ca. 17–18 Å. The calculated dimension of
−

404
S. Bhattacharjee et al. / Journal of Catalysis 225 (2004) 398–407
Fig. 4. Probable energy-level diagram of Mn(III) in square-pyramidal coordination.
Fig. 5. UV–vis spectra of (a) LDH-[Mn(Cl)(L)], (b) free chiral sulfonato–manganese(III) complex (5) and (c) LDH-[C H COO] in Nujol mull.
6
5
sic layered double hydroxides and the acidic Keggin ions. In
the present case, the gallery height of the LDH-[Mn(Cl)(L)]
catalyst is 14.1 Å when the thickness of the brucite layers
psi) and at 298 K. The reactions are expressed by Eqs. (1)
and (2):
(
4.7 Å) was subtracted. In a similar manner to Keggin ions
in an LDH host [36], the present catalyst showed smaller
d spacings than required to locate a perpendicularly orien-
tated Mn complex, although it is clearly possible for the Mn
complex to adopt such an orientation by penetration of the
terminal sulfonato groups into the layers. Alternatively the
Mn complex might be orientated either with the long axis
parallel to the hydroxide layers or diagonally at an angle less
LDH-[Mn-salen]/RT
−−−−−−−−−−−−−−−→
−
,
(1)
R-(+)-Limonene
+-cis-Limonene-1,2-epoxide
◦
than 90 to the layers.
LDH-[Mn-salen]/RT
−−−−−−−−−−−−−−−→
−
.
(2)
3
.5. Catalytic activity of the oxidation of alkenes
(
−)-α-Pinene
(−)-α-Pinene oxide
The LDH-[Mn(Cl)(L)] catalyst was tested in the epoxi-
dation of (R)-limonene and (−)-α-pinene using molecular
The epoxidation of (R)-limonene was tested at atmospheric
pressure of molecular oxygen using various solvents [Eq. (1)].
oxygen and also using air at atmospheric pressure (14.5

S. Bhattacharjee et al. / Journal of Catalysis 225 (2004) 398–407
405
Fig. 6. Calculated structure of the sulfanato–Mn(III) salen ion (5)
Table 2
Table 4
a
Epoxidation of (R)-limonene and (−)-α-pinene using molecular oxygen
Substrate
Performance of reused catalyst in the epoxidation of (R)-limonene and (−)-
α-pinene
Additive
Conversion (%)
Selectivity (%)
de (%)
Run
Reagent
Oxidant
Conversion
(%)
Selectivity
(%)
de
(%)
(
(
(
(
R)-Limonene
R)-Limonene
−)-α-Pinene
−)-α-Pinene
–
96.6
94.7
100.0
100.0
88.5
93.3
90.0
93.3
33.4
42.8
98.1
98.3
b
N-MeI
–
N-MeI
a
a
a
1
2
3
1
2
3
(R)-Limonene
(R)-Limonene
(R)-Limonene
Air
Air
Air
Oxygen
Oxygen
Oxygen
97
97
96
100
100
100
87
89
86
90.0
93.3
89.7
31
32
31
98.1
98.1
98.1
b
a
Reaction conditions: 3.7 mmol substrate, 9.2 mmol pivalaldehyde,
8.5 ml toluene, 0.100 g catalyst, 14.5 psi molecular oxygen, temperature
98 K and 6 h.
b
b
(−)-α-Pinene
(−)-α-Pinene
1
2
(−)-α-Pineneb
b
N-Methylimidazole, 1.78 mmol.
a
Reaction conditions: 3.7 mmol (R)-limonene, 9.2 mmol pivalaldehyde,
18.5 ml toluene, 0.100 g catalyst and air (14.5 psi), temperature 298 K, and
Table 3
reaction time 8 h.
b
Influence of solvent and choice of oxidant on the epoxidation of (R)–
Reaction conditions: 3.7 mmol (−)-α-pinene, 9.2 mmol pivalaldehyde,
limonene and (−)-α-pinenea
18.5 ml toluene, 0.100 g catalyst, and molecular oxygen (14.5 psi), temper-
ature 298 K, and reaction time 6 h.
Substrate
Solvent Oxidant
Conversion Selectivity de
(%)
(%)
(%)
b
b
1,2-epoxide with identical catalytic activity as when using
molecular oxygen. Replacing the toluene by acetone and us-
ing air as oxidant, the reaction proceeded to give similar
conversion, selectivity, and de (Table 3). The above obser-
vation showed that the modified Jacobsen’s catalyst within
the LDH host could be employed just as effectively using
air or oxygen as oxidant and showed little if any dependence
on the type of solvent used. The comparative data displayed
in Fig. 7 show that epoxidation of (R)-limonene using the
present LDH-based Mn–salen catalyst gave much higher
selectivity using air as oxidant and acetone as solvent or
using dioxygen and toluene, than those previously reported
for zeolite entrapped Mn–salen [38] using fluorobenzene as
solvent and high pressures of molecular oxygen, while still
achieving equivalent activity and de.
(
(
(
(
(
R)-Limonene Toluene Dioxygen
R)-Limonene Toluene Air
R)-Limonene Acetone Air
96.6
97.0
98.3
100.0
100.0
88.5
87.0
88.5
90.0
91.7
33.4
31.0
32.4
98.1
98.2
c
c
−)-α-Pinene
−)-α-Pinene
a
Toluene Dioxygen
Acetone Air
b
Reaction conditions: 3.7 mmol substrate, 9.2 mmol pivalaldehyde,
8.5 ml solvent, 0.100 g catalyst and temperature 298 K.
1
b
Reaction time 6 h.
Reaction time 8 h.
c
The results are summarised in Tables 2–4. In toluene, us-
ing pivalaldehyde at 298 K, (R)-limonene was converted
into (+)-cis-limonene-1,2-epoxide with 88.5% selectivity
and 33.4% de (Table 3). Addition of catalytic amounts of
N-methylimidazole improved the de to 42.8% while the se-
lectivity was increased to 93.3%. Yamada et al. [37] found
that N-methylimidazole could be used to improve the ee in
the epoxidation of 1,2-dihydronaphthalenes, suggesting that
this axial base is capable of directing the manner in which
the substrate molecule interacts with the catalyst active site.
Under the above identical reaction conditions, using air as
oxidant at atmospheric pressure and with toluene as solvent
at 298 K, (R)-limonene was converted to (+)-cis-limonene-
The epoxidation of (−)-α-pinene at 298 K was also in-
vestigated using molecular oxygen or air as oxidant and
using a variety of different solvents [Eq. (2)]. Under at-
mospheric pressure of O , (−)-α-pinene was converted to
2
(−)-α-pinene oxide with high conversion, selectivity, and de
(Table 2). In this case, selectivity and de were not affected
by the presence of the donor ligand, N-methylimidazole.
The (−)-α-pinene also produced epoxide in good conver-

406
S. Bhattacharjee et al. / Journal of Catalysis 225 (2004) 398–407
Fig. 7. Comparison of conversion, selectivity, and de (%) of (R)-limonene
over (a) LDH hosted Mn–salen complex and (c) zeolite-immobilized
Mn(salen-2) [38], and for (−)-α-pinene over (b) LDH hosted Mn–salen
complex, (d) zeolite-immobilized Mn(salen-2) [38], and (e) zeolite-
immobilized Co(salen) [38].
sion, selectivity, and de in the presence of acetone as solvent
and using air as oxidant (Table 3), similar catalytic prop-
erties observed in (R)-limonene epoxidation in toluene or
acetone as solvent using molecular oxygen and also in air.
Using air or molecular oxygen, the conversion, selectivity,
and de % were not affected by changing the solvent from
toluene to acetone (Table 3). Using atmospheric pressure
of either oxidant, the conversion, selectivity, and de were
higher than those reported for zeolite-entrapped Mn–salen
Jacobsen’s catalyst [38]. An immobilized cobalt–salen com-
plex [38] was found to produce the highest catalytic activity
in a series of zeolite entrapped metal–salen systems. A com-
parison of data obtained for (−)-α-pinene epoxidation is
shown in Fig. 7. The modified Jacobsen’s catalyst shows
higher conversion, selectivity, and de, when hosted in the
layered double hydroxide catalyst than a zeolite-entrapped
Mn–salen [38] in heterogeneous systems. A higher de but
slightly lower selectivity was also found for our catalyst
when compared with recently reported results for a zeolite-
entrapped Co–salen catalyst [38]. The present results are at
odds with the claim [38] that Jacobsen’s catalyst is not par-
ticularly suited for the (−)-α-pinene epoxidation reaction,
indicating that the use of the LDH as host provides some ad-
ditional features to the catalytic process.
The present results suggest that de and selectivity are
not only dependent on the ligand substituents or the cen-
tral metal ion, but possibly also on the presence of positive
charge interaction of the layered hydroxides host and the
gallery height. The origin of the high catalytic activity may
also be associated with a sufficiently large gallery spacing
of the LDH-[Mn(Cl)(L)] catalyst which allowed greater ac-
cessibility of the substrate to interact with the active centre
of the Mn(III) salen complex between the double hydroxide
layers.
Fig. 8. IR spectra of (a) fresh Na [Mn(Cl)(L)], (b) fresh LDH-[Mn(Cl)(L)]
2
and LDH-[Mn(Cl)(L)] catalyst after (c) first run, (d) second run, and (e)
third run.
tive of the number of cycles. No evidence for leaching of Mn
or decomposition of the catalyst complex was observed dur-
ing the catalysis reaction and no Mn could be detected by
AA spectroscopic measurement of the liquid reaction mix-
ture after catalytic reaction. The IR spectrum (Fig. 8) of the
solid catalyst after reuse was also identical to the fresh cata-
lyst.
The advantages of the present catalyst for epoxidation re-
actions with respect to the homogeneous Jacobsen’s catalyst
or anchored Mn salen systems are:
(
a) Use of molecular oxygen or air instead of sodium
hypochloride as oxidant as used in Jacobsen’s sys-
tem [5]. NaOCl requires careful handling and storage
considerations and, in addition to being toxic and cor-
rosive, has the potential to release chlorine gas. Use of
air or oxygen improves the atom efficiency of the over-
all process with respect to the use of NaOCl and avoids
the formation of NaCl.
(
b) The use of air or oxygen at atmospheric pressures over-
comes the need for complex pressure vessels and addi-
tional safety features which are essential for a system
operating at elevated pressures.
(
c) The catalyst is active at room temperature using acetone
or toluene as solvent instead of fluorobenzene required
by others [38] to achieve good activity. The latter is an
eye, skin, and respiratory irritant with potential toxico-
logical effects.
The stability of the catalyst was studied by performing
repeated epoxidation reactions using the same reaction con-
ditions as described above. At the end of each reaction cy-
cle, the catalyst was recovered by filtration and washed with
toluene, dried, and reused. The results are shown in Table 4
for catalyst reused up to three times. The conversion (%),
selectivity (%), and de (%) were almost identical, irrespec-
Mechanistic studies as well as comparative studies with
the use of the other transition metal–salen complexes into
the layered double hydroxide in the epoxidation of different
olefins as well as the use of other oxidants are in progress to
understand the role of the layered double hydroxide host on
the enantio- and diastereoselectivities.

S. Bhattacharjee et al. / Journal of Catalysis 225 (2004) 398–407
407
4
. Conclusions
[13] B.M. Choudary, M.L. Kantam, A. Rahaman, C.V. Reddy, K.K. Rao,
Angew. Chem. Int. Ed. 40 (2001) 763.
[14] B.M. Choudary, N.S. Chowdari, K. Jyothi, M.L. Kantam, J. Am.
Chem. Soc. 124 (2002) 5341.
−
The chiral SO3 salen–Mn(III) complex intercalated into
Zn/Al-LDH has been prepared and tested for its epoxida-
tion activity using (R)-limonene and (−)-α-pinene as model
compounds using air and molecular oxygen as oxidant at at-
mospheric pressure and at 298 K under a very mild reaction
condition using acetone as a solvent. The chiral salen ligand
and metal complexes synthesised in this study from aque-
ous medium were obtained in high yield and high purity.
The LDH-[Mn(Cl)(L)] catalyst showed higher conversion,
selectivity, and de % for (R)-limonene and (−)-α-pinene
epoxidation when compared with other published data for
other Mn–salen catalysts.
[
15] O. Clause, B. Rebours, E. Merlen, F. Trifiro, A. Vaccario, J. Catal. 133
(1992) 231.
[
[
16] H. Schaper, J.J. Berg-Slot, W.H. Stork, J. Appl. Catal. 54 (1989) 79.
17] A. Cervilla, A. Corma, V. Fornes, E. Llopis, P. Palanca, F. Rey, A.
Ribera, J. Am. Chem. Soc. 116 (1994) 1595.
[18] A. Cervilla, A. Corma, V. Fornes, E. Llopis, F. Perez, F. Rey, A. Rib-
era, J. Am. Chem. Soc. 117 (1995) 6781.
[
[
19] F. Cavani, F. Trifiro, A. Vaccari, Catal. Today 11 (1991) 173.
20] T. Tatsumi, K. Yamamoto, H. Tajuma, H. Tominaga, Chem. Lett.
(1992) 815.
[21] S. Bhattacharjee, J.A. Anderson, Chem. Commun. 5 (2004) 554.
[22] S. Bhattacharjee, J.A. Anderson, unpublished work.
[
[
23] J.F. Larrow, E.N. Jacobsen, Organ. Synth. 75 (1998) 1.
24] M. Botsivali, D.F. Evans, P.H. Missen, M.W. Upton, J. Chem. Soc.,
Dalton Trans. (1985) 1147;
Acknowledgments
K.J. Berry, F. Moya, K.S. Murray, A.B. van der Bergen, B.O. West,
J. Chem. Soc., Dalton Trans. (1982) 109.
We thank the Royal Society (London) for a DWSV (to
S.B.), the EPSRC (GR/S67371/01) for a Research Fellow-
ship, Dr. G. Berry (University of Dundee) for performing the
TGA experiments, and Drs. M. Paterson and I.M. Eggleston
for helpful discussions.
[25] J.H. Kennedy, Analytical Chemistry: Practice, Saunders, New York,
1990;
G.D. Christian, Analytical Chemistry, second edition, Wiley, New
York, 1977.
[
26] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb,
J.R. Cheeseman, V.G. Zakrzewski, J.A. Montgomery Jr., R.E. Strat-
mann, J.C. Burant, S. Dapprich, J.M. Millam, A.D. Daniels, K.N.
Kudin, M.C. Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi, R.
Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford, J. Ochterski,
G.A. Petersson, P.Y. Ayala, Q. Cui, K. Morokuma, D.K. Malick, A.D.
Rabuck, K. Raghavachari, J.B. Foresman, J. Cioslowski, J.V. Ortiz,
B.B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. Gom-
perts, R.L. Martin, D.J. Fox, T. Keith, M.A. Al_Laham, C.Y. Peng, A.
Nanayakkara, C. Gonzalez, M. Challacombe, P.M.W. Gill, B. Johnson,
W. Chen, M.W. Wong, J.L. Andres, C. Gonzalez, M. Head-Gordon,
E.S. Replogle, J.A. Pople, Gaussian 98, Revision A. 5, Gaussian Inc.
Pittsburgh, PA, 1998.
References
[
[
1] R.A. Sheldon, Chirotechnology, Dekker, New York, 1993, Chap. 9.
2] T. Katsuki, in: E.N. Jacobsen, A. Apafaltz, H. Yamamoto (Eds.),
Comprehensive Asymmetric Catalysis, vol. II, Springer, Berlin, 1999,
p. 621;
E.N. Jacobsen, M.H. Wu, in: E.N. Jacobsen, A. Apafaltz, H. Yama-
moto (Eds.), Comprehensive Asymmetric Catalysis, vol. II, Springer,
Berlin, 1999, p. 649;
V.K. Aggarwal, in: E.N. Jacobsen, A. Apafaltz, H. Yamamoto (Eds.),
Comprehensive Asymmetric Catalysis, vol. II, Springer, Berlin, 1999,
p. 679.
[
[
[
27] A.D. Becke, J. Chem. Phys. 98 (1993) 5648.
28] C. Lee, W. Yang, R.G. Parr, Phys. Rev. B 37 (1988) 785.
29] J.W. Yoon, T.-S. Yoon, S.W. Lee, W. Shin, Acta Crystallogr. C 55
[3] J.P. Collman, T. Kodadek, S.A. Raybuck, B. Meunies, Proc. Natl.
Acad. Sci. USA 80 (1983) 7039.
(1999) 1766.
[
30] T.H. Dunning Jr., P.J. Hay, in: H.F. Schaefer III (Ed.), Modern Theo-
retical Chemistry, vol. 3, Plenum, New York, 1976, p. 1.
31] P.J. Hay, W.R. Wadt, J. Chem. Phys. 82 (1985) 270, 284 and 299.
32] D. Lin-Vien, N.B. Colhup, W.G. Fateley, J.C. Grasselli, The Hand
Book Infrared and Raman Characteristic Frequencies of Organic
Molecules, Academic Press, San Diego, 1991.
[4] J.P. Collman, Z. Wang, A. Straumanis, M. Quelquejeu, J. Am. Chem.
Soc. 121 (1999) 460.
[
[
[
[
5] E.N. Jacobsen, W. Zhang, J. Org. Chem. 56 (1991) 437.
6] M.J. Sabater, A. Corma, A. Domenech, V. Fornes, H. Garcia, Chem.
Commun. (1997) 1285.
[
[
[
7] B.M. Choudary, N.S. Chowdari, M.L. Kantam, P.L. Santhi, Catal.
Lett. 76 (2001) 213.
8] D.-W. Park, S.-D. Choi, S.-J. Choi, C.-Y. Lee, G.-J. Kim, Catal.
Lett. 78 (2002) 145.
9] R.I. Kureshy, N.H. Khan, S.H.R. Abdi, I. Ahmad, S. Singh, R.V. Jasra,
J. Catal. 221 (2004) 234.
[33] R. Mukhopadhyay, S. Bhattacharjee, R. Bhattacharyya, J. Chem. Soc.,
Dalton Trans. (1994) 2799.
[
[
34] S.B. Ogunwumi, T. Bein, Chem. Commun. (1997) 901.
35] J. Skarzewski, A. Gupta, A. Vogt, J. Mol. Catal. A: Chem. 103 (1995)
L63.
[36] J. Wang, Y. Tian, R. Wang, J.L. Colon, A. Clearfield, Mater. Res. Soc.
Symp. Proc. 233 (1991) 63.
[10] S. Miyata, Clays Clay Miner. 23 (1975) 369.
[
11] F. Trifiro, A. Vaccario, in: Comprehensive Supramolecular Chemistry,
vol. 7, Pergamon–Elsevier Science, Oxford, 1996, p. 251.
12] B. Sels, V.D. De, M. Buntinx, F. Pierard, A. Kirsch-De Mesmaeken,
P.A. Jacobs, Nature 400 (1999) 855.
[37] T. Yamada, K. Imagawa, T. Nagata, T. Mukaiyama, Chem. Lett. (1992)
2231.
[
[38] C. Scuster, W.F. Holderich, Catal. Today 60 (2000) 193.