Effect of the substrate and catalyst chirality on the diastereoselective epoxidation of limonene using Jacobsen-type catalysts

Article abstract of DOI:10.1002/chir.20755

Chiral and achiral Jacobsen's catalysts in their homogeneous form or immobilized on Al-MCM-41 exhibit similar catalytic activity during diastereoselective epoxidation of limonene when in situ generated dimethyldioxirane is used as oxidizing agent. Experim

Full text of DOI:10.1002/chir.20755

CHIRALITY 22:403–410 (2010)
Effect of the Substrate and Catalyst Chirality on the
Diastereoselective Epoxidation of Limonene
Using Jacobsen-Type Catalysts
´
´
*
JAIRO CUBILLOS, MARIA VARGAS, JULIANA REYES, AIDA VILLA, AND CONSUELO MONTES DE CORREA
Chemical Engineering Department, Universidad de Antioquia, Medellı´n, Colombia
ABSTRACT
Chiral and achiral Jacobsen’s catalysts in their homogeneous form or
immobilized on Al-MCM-41 exhibit similar catalytic activity during diastereoselective
epoxidation of limonene when in situ generated dimethyldioxirane is used as oxidizing
agent. Experimental observations suggest that not only the catalyst chiral center but
also the substrate chiral center participates in the preferential formation of most diaster-
eomers. Remarkable turnover numbers (TON), up to 288, was achieved over the hetero-
geneous catalysts in comparison to their homogeneous counterparts (TON up to 46).
Catalyst leaching rather than catalyst oxidative degradation was identified as the main
source of catalyst deactivation during reutilization tests. Chirality 22:403–410,
C
VV
2010.
2009 Wiley-Liss, Inc.
KEY WORDS: chirality; manganese salen complexes; achiral catalyst; heterogeneous
catalysis; catalyst degradation
INTRODUCTION
substrate is an enantiomerically pure compound and the
catalyst is either chiral or achiral.
Enantiomerically pure epoxides are synthetically useful
for synthesizing many chiral nonracemic compounds.1 R-
(1)-1, 2-Limonene oxide is commercially available and
relatively inexpensive but, usually marketed as a 1:1 mix-
ture of the cis- and trans-epoxides (Aldrich, 97% pure). In
general, oxidation of R-(1)- and S-(2)- limonene yield a va-
riety of products. Epoxides are formed if oxidation occurs
at olefinic positions due to the reactivity of the more sub-
stituted double bonds.2 These substrates have two olefinic
bonds (1, 2 and 8, 9 in Fig. 1) and the oxidation can take
place at either or both of these bonds. Furthermore, two
types of diastereomers (cis and trans) are expected for
each of the epoxide products (see Fig. 2).3 Typically in
these reactions, a 1:1 ratio of the cis/trans diastereomers
are formed as the chiral center in the molecule does not
seem to affect the orientation of the incoming oxygen
atom.2,3
Chiral and achiral salen transition metal complexes have
become a matter of current interest because of their wide
range of applications as catalysts for asymmetric epoxida-
tion reactions.4 Figure 3 illustrates the general structure of
metal salen complexes. When substituents R1 and R3 are
equal to R2 and R4, respectively, the complex corresponds
to an achiral catalyst. However, in the chiral catalyst either
R1 is different from R2 and R3 different from R4.
Epoxidation of unfunctionalized olefins catalyzed by chi-
ral manganese (III) salen complexes, initially developed
by Jacobsen and Katsuki, have emerged as practical meth-
ods for the synthesis of optically active epoxides.6 Among
them, the most important catalyst for enantioselective
epoxidation of unfunctionalized olefins is the Jacobsen’s
catalyst (with both optical configuration R,R and S,S), a
chiral manganese salen complex which displays high activ-
ity and enantioselectivity for asymmetric epoxidation of
conjugated cis-disubstituted and trisubstituted prochiral
olefins in homogeneous phase. However, the separation
and recycling of this catalyst are still problematic issues.7
In addition, salen transition metal complexes are very
expensive materials, mainly in their chiral forms.8 The
conventional way to solve these problems is to immobilize
chiral manganese (III) salen complexes onto solid sup-
ports. Therefore, many efforts have been devoted to heter-
ogenize these catalysts using a solid matrix as support.9
Various approaches for immobilization of chiral manga-
nese (III) salen complexes have been described in an
excellent review, which includes grafting the catalyst on a
solid inorganic support such as silica or MCM-41, encap-
sulation into the pores of zeolites, physical entrapment in a
It is known that any asymmetric reaction requires the
presence of a chiral component in the reaction medium.5
In this way, two types of asymmetric catalytic transforma-
tions can be distinguished. Enantioselective reaction,
which can take place when the substrate is an achiral or
prochiral compound and the catalyst is a chiral compound,
Contract grant sponsor: CENIVAM-Colciencias; Contract grant number:
432-2004
Contract grant sponsor: Universidad de Antioquia
*Correspondence to: Jairo Antonio Cubillos Lobo, Calle 67 No. 53-108,
Medell´ın, Colombia. E-mail: jacubil@gmail.com
Received for publication 16 March 2008; Accepted 22 May 2009
DOI: 10.1002/chir.20755
Published online 29 June 2009 in Wiley InterScience
or diastereoselective reaction which can occur when the (www.interscience.wiley.com).
C
VV
2009 Wiley-Liss, Inc.

404
CUBILLOS ET AL.
Fig. 1. Chemical structure of limonene enantiomers: (a) R-(1)-Limo-
nene, (b) S-(2)-Limonene.
Fig. 3. Chemical structures of metal salen complexes.
polydimethylsiloxane membrane (polymer support), clays,
and activated carbon.9 However, heterogenization of
homogeneous catalysts generally leads to lower catalytic
activity than that reached by their homogeneous counter-
parts.9 In addition, the instability of the catalyst itself
under reaction conditions avoids its useful reutilization.9
Therefore, to carry out successful heterogeneous asym-
metric epoxidation, not only the design of supported
catalysts but also the selection of appropriate oxidation
conditions is very important.10 In this sense, Cubillos et al.
reported that the use of in situ generated dimethyldioxir-
ane (DMD) as oxidizing agent improved the stability of
the Jacobsen’s catalyst during the enantioselective epoxi-
dation of three proquiral olefins in comparison to other
common oxygen sources such as sodium hypochlorite and
m-chloroperbenzoic acid (m-CPBA).11
Recently, two research groups independently reported
the heterogeneous diastereoselective epoxidation of R-
(1)-limonene using chiral salen manganese complexes as
catalysts in the system oxygen/sacrificial aldehyde,
according to Mukaiyama’s conditions.12,13 However, the
catalytic behavior of the corresponding achiral catalyst
was not explored in these works. Recently, Ratnasamy and
coworkers14 used a heterogeneous achiral Jacobsen’s cata-
lyst (without the terbutyl groups) on limonene epoxidation
but low catalytic activities were obtained. It is widely
known that terbutyl groups favor the approaching of
incoming olefin to the catalytically active center.15
immobilized on Al-MCM-41 by ionic bond using in situ
generated DMD as the oxygen source. The presence of
framework aluminum in Al-MCM-41 generates a negative
charge which offers the possibility of immobilizing manga-
nese salen complexes of cationic nature.16 Results suggest
that the chiral center of the substrate also plays an impor-
tant role in the formation of the predominant diaster-
eomer. Additionally, reaction conditions used in this work
markedly improved catalyst stability during oxidative deg-
radation, so the utilization of a stable heterogeneous
catalyst could be accomplished.
MATERIALS AND METHODS
Characterization
FTIR spectra of the solid samples in the 4000–400 cm21
range were obtained with a Nicolet Avatar 330 FTIR spec-
trometer, using KBr as reference. Atomic absorption anal-
yses of manganese were performed in UNICAM-929
equipment. Thermal gravimetric analysis (TGA) was per-
formed in a TGA 2950 apparatus at a heating rate of 2
K/min. Surface area measurements were performed by N2
adsorption at 21968C in a Micromeritics ASAP 2010 appa-
ratus. Powder X-ray diffraction (XRD) patterns were
recorded on a RIGAKU D-Max/IIIB diffractometer using
Cu Ka radiation. UV–vis spectra was recorded in the
range 200–800 nm with a Lamba 4 Perkin Elmer spectrom-
eter equipped with a diffuse reflectance attachment, using
BaSO4 as reference. Reaction samples were analyzed by
GC-FID with a Varian Star 3400 gas chromatograph using
helium as carrier gas and two capillary columns, DB-1 (50
m long, 0.32 mm id and 1.20 lm film thickness) and
Lipodex-G (50 m long and 0.25 mm id).
In this article, we report on the diastereoselective epoxi-
dation of limonene (optical configuration R and S) using
the Jacobsen’s catalyst in its homogeneous chiral (optical
configuration R,R and S,S) and achiral forms, as well as,
Synthesis of Homogeneous Catalysts
The Jacobsen’s catalyst in its two optically active forms
denoted as R,R-Jacobsen and S,S-Jacobsen and a simple
achiral version of the Jacobsen’s catalyst denoted as achi-
ral Jacobsen (see Fig. 4) were used in this work. These
catalysts were prepared according to reported proce-
dures17 by reacting 8.50 mmol of 3,5-di-tert-butyl-2-hydrox-
ybenzaldehyde with the corresponding diamine
Fig. 2. Chemical structure of the possible limonene oxides originating
from R-(1)-limonene epoxidation: (a) cis-(1)-1,2-limonene oxide, (b)
trans-(1)-1,2-limonene oxide, (c) cis-(1)-8,9-limonene oxide, (d) trans-(1)-
8,9-limonene oxide, (e) diepoxide.
Chirality DOI 10.1002/chir

EPOXIDATION OF LIMONENE USING JACOBSEN-TYPE CATALYSTS
405
Fig. 4. Manganese (III) salen complexes: (a) R,R-Jacobsen, (b) S,S-Jacobsen, (c) achiral catalyst.
Heterogenization of the Jacobsen’s Catalyst
(4.20 mmol) and subsequent Mn(III) incorporation by met-
alation of the salen ligand. As diamine component, we
used (1R, 2R)-(1)-1,2-diaminocyclohexane L-tartrate for the
R,R-Jacobsen, (1S, 2S)-(1)-1,2-diaminocyclohexane L-tar-
trate for the S,S-Jacobsen, and 1,2-diamino ethane for the
achiral Jacobsen. In all cases, 8.50 mmol manganese (II)
acetate (Mn(C2H3O2)2ꢀ4H2O) was used as the Mn source.
Also, the optically pure salen ligands ((R,R)-(2)-N,N0-
bis(3,5-di-tert-butylsalicylidene)cyclohexane-1,2-diamine),
((S,S)-(2)-N,N0-bis(3,5-di-tert-butylsalicylidene)cyclohex-
ane-1,2-diamine), and achiral salen ligand (bis(3,5-di-tert-
butylsalicylidene)ethano-1,2-diamine) were prepared. The
purity and identity of these soluble materials were con-
firmed by FTIR, TGA, and UV–vis.
Synthesis of Al-MCM-41. Mesoporous Al-MCM-41
was prepared following the procedure reported by van
Hooff18 with slight modifications. In a typical synthesis,
48 g tetraethylammonium hydroxide (TEAOH, 35 wt %,
aqueous), 0.84
g sodium aluminate (NaAlO2), and
116.4 g H2O were mixed together and stirred at room
temperature for 1 h. Then, 40 g of tetradecyltrimethylam-
monium bromide (TDTMABr) was added and the result-
ing mixture stirred during
4 h. Finally, Ludox-HS
40 (60.92 g) was added drop wise over a period of 1 h
and stirring continued at ambient temperature for
additional 4 h. The gel composition was:
Fig. 5. Immobilization of the homogeneous catalyst by ionic bond in Al-MCM-41.
Chirality DOI 10.1002/chir

406
CUBILLOS ET AL.
Fig. 6. Diastereoselective epoxidation of R-(1)-limonene in the presence of chiral and achiral manganese salen complexes and in situ generated
DMD as oxidizing agent.
with 0.82 g of either optically pure salen ligand or 0.74 g
achiral salen ligand in 20 ml CH2Cl2. In all cases, an
excess of the amount of salen ligand with respect to Mn
was used (Mn/salen 5 0.5 mmol/mmol) to guarantee the
full complexation of the Mn atoms. After that, the inert
atmosphere was replaced by flowing air and stirring con-
tinued for 6 h. The mixture was cooled to room tempera-
ture and the solvent removed by reduced pressure at
room temperature. The resulting solid was Soxhlet
extracted with CH2Cl2 at 608C and toluene at 1408C until
the washing solvent remained colorless. The immobilized
catalysts on Al-MCM-41 were coded R,R-Al-MCM-41, S,S-
Al-MCM-41 and achiral-Al-MCM-41. Figure 5 depicts the
immobilization method used in this work. The heterogene-
ous process was confirmed by XRD, N2-sorption, Mn
atomic absorption TGA, and FTIR.
1:0 NaAlO2: 40 SiO2: 11:6 TDTMABr:
28 TEAOH: 800 H2O
The gel was heated under static conditions at 1108C for
7 days. After the second, fourth, and sixth day, pH was
adjusted to 10.5 using CH3COOH (10 wt %, aqueous). After
crystallization, the solid phase was recovered by filtration
and washed with abundant deionized water. The white
material was dried at 608C overnight, followed by calcina-
tion in flowing air at 5408C for 6 h (heating rate 18C/min).
Synthesis of Mn-Al-MCM-41. Calcined Al-MCM-41
(3 g) was added to a manganese (II) acetate solution in
water (100 ml, 0.2 M) and stirred for 24 h. The material
was then filtered, washed, dried, and stirred again in a
fresh manganese (II) acetate solution for additional 24 h.
This procedure was repeated twice. Finally, the obtained
solid was calcined in flowing air at 5508C for 8 h before
use.
Catalytic Measurements
Catalysts were tested on the diastereoselective epoxida-
tion of R-(1)- and S-(2)-limonene at room temperature
(see Fig. 6). In a typical reaction, 0.702 g of R-(1)- or
Immobilization by Ion Exchange in Mn-Al-MCM-41
A method similar to that reported by Hutchings and S-(2)-limonene, 1 g of sodium bicarbonate, and 0.03 g of
coworkers19 was used to immobilize all homogeneous cat- homogeneous catalyst (0.047 mmol of optically pure
alysts. In a typical procedure, dry Mn-Al-MCM-41 (3 g) catalysts, 0.052 mmol of achiral catalyst) or 0.1 g of hetero-
was treated, under reflux in argon atmosphere for 24 h, geneous catalyst (0.008 mmol enantiomerically pure
Fig. 7. Formation of DMD from Oxone1 (KHSO5 as active agent) and acetone.
Chirality DOI 10.1002/chir

EPOXIDATION OF LIMONENE USING JACOBSEN-TYPE CATALYSTS
407
Fig. 9. Nitrogen adsorption-desorption isotherms: (a) Al-MCM-41, (b)
Mn-Al-MCM-41, (c) R,R-Al-MCM-41.
Fig. 8. XRD patterns: (a) Al-MCM-41, (b) Mn-Al-MCM-41, (c) R,R-Al-
MCM-41.
sities did remarkably decrease, suggesting that the incor-
poration of the manganese salen complex destroyed the
mesopores of Al-MCM-41.
catalysts, 0.009 mmol achiral catalyst) were dissolved in 30
ml of acetone. The pH of the resulting mixture, denoted A,
was adjusted between 8.0 and 8.5 using NaHCO3 (5 wt %
aqueous). On the other hand, a solution, denoted B, was
prepared by dissolving 1.23 g of Oxone1 (2KHSO5ꢁKH-
SO4ꢁK2SO4) in 25 ml of H2O. Solution B was slowly added
to mixture A while stirring and maintaining the pH in the
range 8.0–8.5 with 5 wt % aqueous NaHCO3. The proce-
dure to separate the homogeneous catalyst was as follows:
inorganic salts were isolated by centrifugation, the result-
ing liquid mixture extracted with 30 ml of dichlorome-
thane, the aqueous phase discarded, and the organic layer
treated by vacuum distillation (1608C and 0.08 MPa). On
the other hand, for heterogeneous reactions, the catalyst
was easily separated by filtration and both solid and liquid
mixtures were retained for further use. Solid samples
were thoroughly washed with water to remove inorganic
salts originated from the oxygen source and buffer solu-
tion. In both cases, homogeneous and heterogeneous
reactions, the catalyst free liquid mixture was concen-
trated under vacuum and aliquots analyzed by GC-FID
using a DB-1 column to determine conversion of R-(1)-
limonene and selectivity to 1,2-(1)-limonene oxide and a
Lipodex-G column to determine percent diasteromeric
excess (% de) between cis-(1)-1,2-limonene oxide and
trans-(1)-1,2-limonene oxide.
N2 adsorption isotherms and textural properties of
Al-MCM-41, Mn-Al-MCM-41, and R,R-Al-MCM-41 are pre-
sented in Figure 9 and Table 1. These isotherms are
characteristic of mesoporous materials. During the immo-
bilization process, BET surface area and total pore volume
decreased from 1113 m2/g for Al-MCM-41 to 999 m2/g for
Mn-Al-MCM-41. Mesopore volume decreased from 0.915
for Al-MCM-41 to 0.336 cm3/g for Mn-Al-MCM-41. BET
surface area and mesopore volume of Mn-Al-MCM-41
decreased when transformed to R,R-Al-MCM-41. There-
fore, it appears that the complex was deposited inside the
pore system of the support.22
Elemental analysis and TGA. Table 2 shows Mn(III)
catalyst loading of R,R-Al-MCM-41 determined by atomic
absorption and TGA. Neither the Mn atom nor the salen
ligand in Al-MCM-41 was detected. After ion exchange,
Mn-Al-MCM-41 contained 0.548 wt % Mn. After treatment
with the salen ligand, Mn loading decreased to 0.446 wt %
in chiral-Al-MCM-41. From TGA a salen ligand loading of
9.105 wt % in chiral-Al-MCM-41 was determined. The R,R-
Al-MCM-41 was obtained with an excess of ligand (Mn/
salen 5 0.488 mmol/mmol) which corresponds to that
used in the preparation method (Mn/salen 5 0.5 mmol/
mmol). The fact that salen ligand is in excess after Soxhlet
extraction suggests that uncomplexed salen ligand is
mainly located in inaccessible sites of the support. As the
metal loading is the limiting reagent, the catalyst loading
in chiral-Al-MCM-41 corresponds to 5 wt %.
DMD was obtained in situ from reaction between
KHSO5 (active component of Oxone1) and acetone in a
slightly basic reaction medium (pH 5 8.0–8.5). Figure 7
depicts the formation of DMD.20
TABLE 1. Surface area and pore volume of
synthesized catalysts
RESULTS AND DISCUSSION
Catalyst Characterization
Structural and textural properties. The XRD pat-
terns of Al-MCM-41 exhibit a very intense peak at 2y 5
2.5 and two additional peaks with low intensities at 2y 5
4.3 and 2y 5 4.9 (Fig. 8a), which can be indexed to a hex-
agonal lattice.21 It is observed that after ionic exchange
and catalyst immobilization (Figs. 8b and 8c); peak inten-
BET surface
area (m2/g)
Pore volume
BJH (cm3/g)
Material
Al-MCM-41
Mn-Al-MCM-41
R,R-Al-MCM-41
1113
999
712
0.915
0.336
0.163
Chirality DOI 10.1002/chir

408
CUBILLOS ET AL.
TABLE 2. Catalyst loading
TABLE 3. Control reactionsa
Salen
Selectivity
Mn
ligand
Catalyst
loading
(wt %)
Conversion to 1,2-limonene Diasteromeric
loading loading
Mn/salen
Material
(%)
epoxide (%)
excess (de %)b
Material
(wt %)a
(wt %)b (mmol/mmol)
None
Al-MCM-41
Mn-Al-MCM-41
53
55
74
34
61
70
23
24
29
Al-MCM-41
Mn-Al-MCM-41
R,R-Al-MCM-41
0
0
0
–
–
0
0
0.548
0.446
9.105
0.488
5c
aReaction conditions: 5 mmol R-(1)-limonene or S-(2)-limonene; 4 mmol
KHSO5; 0.03 g homogeneous catalyst; 0.1 g (5 wt %) heterogeneous cata-
lyst; 1 g NaHCO3; 30 ml acetone; 35 min reaction time; 258C.
aDetermined by Mn atomic absorption.
bDetermined by TGA.
bReferred to cis-1,2-limonene oxide (predominant epoxide).
cDetermined from Mn loading.
In general, low asymmetric inductions were reached (de
up to 56%) compared to those reported for cis-disubstituted
olefins (ee up to 99%),4 showing a favorable steric interac-
tion between one of the alkene substituents and the active
manganese(III) salen complex plane, which can be
explained from the side-on approach model.25 A commer-
cial sample of (cis/trans)-1,2-limonene oxide (Aldrich, 97%)
was used to identify cis and trans diastereomers. 10% de of
trans-(1)-1,2-limonene oxide was quantified in this sample
by using a Lipodex-G chromatographic column.
The results of three control experiments are listed in
Table 3. In the absence of catalyst, moderate R-(1)-limo-
nene conversion (53%), low selectivity to (cis/trans) 1,2-
limonene oxide (34%), and low de (23%) was obtained. Sim-
ilar results were obtained with S-(2)-limonene. A remark-
able selectivity increase was observed over Al-MCM-41
(61%) and Mn-Al-MCM-41 (70%). These results indicate
that the support and Mn atoms have a positive effect
on the stability of (cis/trans) 1,2-limonene oxide. The fact
that the initial pH increased from 7.0 to 8.0 in the presence
of these materials (Al-MCM-41 and Mn-Al-MCM-41)
allows us to infer that the epoxide ring opening reaction
rate could be reduced compared with the oxygen transfer
reaction.26 Only a slight increase in conversion was
obtained in the presence of Mn-Al-MCM-41. On the other
hand, neither Al-MCM-41 nor Mn-Al-MCM-41 produced
remarkable effect on de.
Spectral properties. The FTIR spectra of chiral-Al-
MCM-41, Figure 10d, show characteristic bands of the or-
ganic structure of manganese (III) salen complexes (Fig.
10a).23 Although these bands are weak and not sufficient
enough to elucidate a structure, the typical band at 1540
cm21 is distinguished. This signal is ascribed to the N-Mn
and O-Mn stretching vibrations and was absent in the
FTIR spectra of Al-MCM-41 (Fig. 10b) and Mn-Al-MCM-41
(Fig. 10c), thus confirming the presence of the Mn salen
complex inside the mesoporous support. Additionally, the
OH stretching band centered at 3500 cm21 is characteris-
tic of terminal silanol groups from the support.24
Catalytic tests
Oxidation of limonene. Oxidation of R-(1)- and S-
(2)-limonene can yield a variety of products, such as (cis/
trans) 1,2-limonene oxide, (cis/trans) 8,9-limonene oxides,
(cis/trans) 1,2 and 8,9 limonene oxides (epoxides), car-
veol, and carvone. However, under the reaction conditions
used in this study, 1,2-limonene oxide was the main prod-
uct with minor amounts of diepoxide. It is known that the
endocyclic double bond (Fig. 1, bond 1,2) is more reactive
than the exocyclic one (Fig. 1, bond 8,9).2
Tables 4 and 5 show activity results using the homoge-
neous catalysts, their corresponding fresh heterogeneous
catalysts, as well as, reused heterogeneous catalysts for
oxidation of R-(1)-limonene and S-(2)-limonene, respec-
tively. In general, selectivity improved in the presence of
the homogeneous catalysts compared with the uncata-
lyzed reaction. Additionally, it is worth to note that the
product stereochemistry is strongly dependent on the
absolute configuration of both the catalyst and the limo-
nene. Thus, the combination of R-(1)-limonene with R,R-
Jacobsen or (S)-(2)-limonene with S,S-Jacobsen forms a
matched pair (see row 1 in Tables 4 and 5), giving rise to
higher de, 56% and 45%, respectively. Whereas R-(1)-limo-
nene with S,S-Jacobsen or S-(2)-limonene with R,R-Jacob-
sen generates a mismatched pair (see row 2 in Tables 4
and 5) and the corresponding de are reduced to 5–6%.
This phenomenon has been described as double asymmet-
ric induction process,27 and can be rationalized assuming
the qualitative transition-state model proposed for trisubsti-
Fig. 10. FTIR spectra: (a) chiral catalyst, (b) Al-MCM-41, (c) Mn-Al-
MCM-41, (d) R,R-Al-MCM-41.
Chirality DOI 10.1002/chir

EPOXIDATION OF LIMONENE USING JACOBSEN-TYPE CATALYSTS
409
TABLE 4. Results of catalytic activity using R-(1)-limonene as substratea
Selectivity
TONc
Conversion
(%)
to 1,2-limonene
epoxide (%)
Diastereomeric
excess (de %)b
(mmol cis-(1)-1,2-limonene
Catalyst
oxide/mmol catalyst)
R,R-Jacobsen
S,S-Jacobsen
55
60
100
92
56
5
46
7
Achiral Jacobsen
R,R-Al-MCM-41
S,S-Al-MCM-41
Achiral-Al-MCM-41
R,R-Al-MCM-41d
Achiral-Al-MCM-41d
R,R-Al-MCM-41e
Achiral-Al-MCM-41e
53
100
78
81
60
70
60
71
100
68
80
81
49
62
58
64
50
35
23
36
25
24
22
22
38
288
58
260
117
157
134
160
aReaction conditions: 5 mmol R-(1)-limonene; 4 mmol KHSO5; 0.03 g homogeneous catalyst; 0.1 g (5 wt %) heterogeneous catalyst; 1 g NaHCO3; 30 ml
acetone; 35 min reaction time; 258C.
bReferred to cis-1,2-limonene oxide (predominant epoxide).
cTurnover number (TON) 5 mmol cis(1)-1,2-limonene epoxide/mmol catalyst.
dFirst reuse.
eSecond reuse.
tuted olefins in which the limonene interacts via a skewed activity experiments using commercial limonene oxide
side-on approach with the metal-oxo intermediate com- instead of R-(1)-limonene or S-(2)-limonene proved that
plex. Additionally, for each substrate, similar de and turn- our heterogeneous catalysts did not promote the kinetic
over numbers (TON) were obtained with the best optically separation among cis/trans isomers. In contrast, it was
pure catalyst and the achiral catalyst (compare rows 1 and confirmed that limonene oxide undergoes further epoxida-
3 either in Table 4 or 5). In particular, the similarity tion leading to diepoxide (conversion ca. 50%). This may
between de values can be due to both substrate and cata- be the reason why the selectivity to (cis/trans) 1,2-limo-
lyst chiral center. In other words, not only the catalyst chi- nene oxide over the heterogeneous catalyst was not
ral center but also the substrate chiral center participates complete.
in the preferential formation of cis-(1)-1,2-limonene oxide.
Finally, the reutilization issue was addressed. Heteroge-
With the exception of S,S-Al-MCM-41 for the oxidation neous catalysts underwent a progressive loss of their
of S-(1)-limonene (compare rows 1 and 4, Table 5), higher initial catalytic activity through two reuses. To identify the
conversions and TON were reached over the heterogene- origin of the catalytic activity loss, the chemical structure
ous catalysts because lower amount of the manganese of the fresh homogeneous catalyst and reused twice were
salen complex was used in the heterogeneously catalyzed investigated by FTIR. Figure 11 shows that the catalyst
reaction. The TON increase can also be associated with did not suffer oxidative degradation, because the main sig-
catalyst isolation on the support surface, preventing in this nal assigned to the catalyst (1540 cm21) is retained after
way, deactivation routes by oligomerization.8 Catalytic two consecutive runs. Therefore, the decrease in catalytic
TABLE 5. Results of catalytic activity using S-(2)-limonene as substratea
Selectivity to
1,2-limonene
epoxide (%)
TONc
Conversion
(%)
Diastereomeric
excess (de %)b
(mmol cis-(1)-1,2-limonene
oxide/mmol catalyst)
Catalyst
S,S-Jacobsen
R,R-Jacobsen
80
60
63
68
72
76
65
70
66
68
75
77
87
67
73
73
74
75
70
72
45
6
11
6
Achiral Jacobsen
S,S-Al-MCM-41
R,R-Al-MCM-41
Achiral-Al-MCM-41
S,S-Al-MCM-41d
Achiral-Al-MCM-41d
S,S-Al-MCM-41e
Achiral-Al-MCM-41e
46
23
20
20
18
22
20
23
10
43
48
50
43
49
46
46
aReaction conditions: 5 mmol S-(2)-limonene; 4 mmol KHSO5; 0.03 g homogeneous catalyst; 0.1 g (5 wt %) heterogeneous catalyst; 1 g NaHCO3; 30 ml
acetone; 35 min reaction time; 258C.
bReferred to cis-1,2-limonene oxide (predominant epoxide).
cTurnover number (TON) 5 mmol cis(1)-1,2-limonene epoxide/mmol catalyst.
dFirst reuse.
eSecond reuse.
Chirality DOI 10.1002/chir

410
CUBILLOS ET AL.
8. Xia Q-H, Ge H-Q, Ye C-P, Liu Z-M, Su K-X. Advances in homogeneous
and heterogeneous catalytic asymmetric epoxidation. Chem Rev
2005;105:1603–1662.
9. Baleiz˜ao C, Garc´ıa H. Advances in homogeneous and heterogeneous
catalytic asymmetric epoxidation. Chem Rev 2006;106:3987–4043.
10. Song CE, Roh EJ, Yu BM, Chi DY, Kim SC, Lee K-J. Heterogeneous
asymmetric epoxidation of alkenes catalysed by a polymer-bound (pyr-
rolidine salen)manganese(III) complex. Chem Commun 2000;615–
616.
11. Cubillos J, Hoelderich WF. Jacobsen’s catalyst anchored on Al-MCM-
41 and NH2 group modified Si-MCM-41 as heterogeneous enantiose-
lective epoxidation catalyst using in situ generated dimethyldioxirane
as oxidant. Rev Fac Ing Univ Antioquia 2007;41:31–47.
12. Schuster C, Mo¨llman E, Tompos A, Ho¨lderich W. Highly stereoselec-
tive epoxidation of (2)-a-pinene over chiral transition metal (salen)
complexes occluded in zeolitic hosts. Catal Lett 2001;74:69–75.
Fig. 11. FTIR spectra of R,R-Jacobsen: (a) fresh, (b) reused twice.
13. Bhattacharjee S, Anderson JA. Synthesis and characterization of novel
chiral sulfonato-salen-manganese(III) complex in a zinc-aluminium
LDH host. Chem Commun 2004;554–555.
activity of the heterogeneous catalyst is caused by leach-
ing of the catalyst rather than its oxidative degradation.
14. Saikia L, Srinivas D, Ratnasamy P. Chemo-, regio- and stereo-selective
aerial oxidation of limonene to the endo-1,2-epoxide over Mn(Salen)-
sulfonated SBA-15. Appl Catal A 2006;309:144–154.
15. Gaquere A, Liang S, Hsu F-L, Bu XR. The readily available tert-pentyl
group as a most effective simple directing group for asymmetric syn-
thesis: a case study on Salen-Mn(III)-catalyzed epoxidation. Tetrahe-
dron: Asymmetry 2002;13:2089–2093.
CONCLUSIONS
Optically active and achiral Jacobsen’s catalysts either ho-
mogeneous or immobilized on Al-MCM-41 exhibited similar
catalytic activity during diastereoselective epoxidation of R-
(1)-limonene using in situ generated DMD as oxidizing
agent. In particular, the similarity among de values obtained
using both the homogeneous and the heterogeneous cata-
lysts, suggests that the R-(1)-limonene chiral center, takes
part in the preferential formation of cis-(1)-1,2-limonene
oxide. This result suggests that the chirality of the susb-
strate plays an important role in the stereochemical forma-
tion of new chiral centers and the process occurs through
classical double asymmetric induction.
Catalyst immobilization of the oxo-manganese active
species on the support surface favored the stability toward
oligomerization leading to improved catalytic productivity.
Catalyst reutilization was not successfully accomplished
likely due to catalyst leaching rather than oxidative
degradation.
16. Piaggio P, McMorn P, Langham C, Bethell D, Bulman-Page PC,
Hancock FE. Asymmetric epoxidation of stilbene by manganese(III)
chiral salen complex immobilized in Al-MCM-41.
1998;22:1167–1169.
N J Chem
17. Larrow JF, Jacobsen EN. A practical method for the large-scale prepa-
ration of [N,N0-bis(3,5-di-tertbutylsalicylidene)-1,2-cyclohexanediami-
nato(2-)]manganese(III) chloride, a highly enantioselective epoxida-
tion catalyst. J Org Chem 1994;59:1939–1942.
18. Busio M, Jaenchen J, van Hooff JHC. Aluminum incorporation in
MCM-41 mesoporous molecular sieves. Microporous Mater 1995;5:
211–218.
19. Piaggio P, Morn P, Murphy D, Bethell D, Page PCB, Hancock FE, Sly
C, Kerton OJ, Hutchings GJ. Enantioselective epoxidation of (Z)-stilbene
using a chiral Mn(III)-salen complex: effect of immobilisation on MCM-
41 on product selectivity. Chem Soc Perkin Trans 2000;2:2008–2015.
20. Webb KS, Ruszkay SJ. Oxidation of aldehydes with Oxone1 in aque-
ous acetone. Tetrahedron 1998;54:401–410.
21. De Clercq B, Lefebvre F, Verpoort F. Immobilization of multifunc-
tional Schiff base containing ruthenium complexes on MCM-41. Appl
Catal A: Gen 2003;247:345–364.
LITERATURE CITED
22. Chaube VD, Shylesh S, Singh AP. Synthesis, characterization and
catalytic activity of Mn(III)- and Co(II)-salen complexes immobilized
mesoporous alumina. J Mol Catal A: Chem 2005;241:79–87.
1. Surburg H, Panten J. Common fragrance and flavor materials. Wein-
heim: Wiley-VCH; 2006. p 3.
2. Royals EA, Leffingwe JC. Reaction of the limonene 1,2-oxides. I. The
stereospecific reaction of the (1)-cis- and (1)-trans-limonene 1,2-
oxides. J Org Chem 1966;31:1937–1944.
23. Deshpande S, Srinivas D, Ratnasamy P. EPR and catalytic investiga-
tion of Cu(Salen) complexes encapsulated in zeolites.
1999;188:261–269.
J Catal
3. Hutter R, Mallat T, Baiker A. Titania-silica mixed oxides. III. Epoxida-
tion of a-isophorone with hydroperoxides. J Catal 1995;153:177–189.
24. Kureshy RI, Ahmad I, Khan NH, Abdi SHR, Phatak K, Jasra RV.
Chiral Mn(III) salen complexes covalently bonded on modified MCM-
41 and SBA-15 as efficient catalysts for enantioselective epoxidation of
nonfunctionalized alkenes. J Catal 2006;238:134–141.
4. Larrow JF, Jacobsen EN. Asymmetric processes catalyzed by chiral
(salen)metal complexes. Topics Organomet Chem 2004;6:123–152.
25. Dalton CT, Ryan KM, Wall VM, Bousquet C, Gilheany DG. Recent
progress towards the understanding of metal-salen catalysed asym-
metric alkene epoxidation. Top Catal 1998;5:75–91.
5. Sheldon RA. Chirotechnology. New York: Marcel Dekker, Inc.; 1993.
p 19.
6. Wang D, Wang M, Zhang R, Wang X, Gao A, Ma J, Sun L. Asymmet-
ric epoxidation of styrene and chromenes catalysed by dimeric chiral
(pyrrolidine salen)Mn(III) complexes. Appl Catal A: Gen 2006;315:
120–127.
26. Ready JM, Jacobsen EN. A practical oligomeric [(salen)Co] catalyst
for asymmetric epoxide ring-opening reactions. Angew Chem Int Ed
2002;41:1374–1377.
7. McMorn P, Hutching G. Heterogeneous enantioselective catalysts:
strategies for the immobilisation of homogeneous catalysts. Chem Soc
Rev 2004;33:108–122.
27. Masamune S, Shoy W, Petersen JS, Sita LR. Double asymmetric syn-
thesis and a new strategy for stereochemical control in organic syn-
thesis. Angew Chem Int Ed Engl 1985;24:1–30.
Chirality DOI 10.1002/chir