Efficient aerobic epoxidation of alkenes in perfluorinated solvents catalysed by chiral (salen) Mn complexes

Article abstract of DOI:10.1039/a800558c

Chiral complexes selectively soluble in perfluorocarbons have been synthesized for the first time and tested as catalysts for the epoxidation of alkenes under fluorous biphasic conditions.

Full text of DOI:10.1039/a800558c

View Article Online / Journal Homepage / Table of Contents for this issue
Efficient aerobic epoxidation of alkenes in perfluorinated solvents catalysed by
chiral (salen) Mn complexes
Gianluca Pozzi,*† Flavio Cinato, Fernando Montanari and Silvio Quici
Centro CNR and Dipartimento di Chimica Organica e Industriale dell’Universita`, Via Golgi 19, I-20133 Milano, Italy
Chiral complexes selectively soluble in perfluorocarbons
have been synthesized for the first time and tested as
catalysts for the epoxidation of alkenes under fluorous
biphasic conditions.
proper chiral 1,2-diamine (Scheme 1).‡ The synthesis of the key
compound 3 was easily achieved following the pathway
reported in the same scheme. The free OH groups of
3,5-diiodosalicyclic acid were protected by methylation with
Me2SO4 (quantitative yield) before the coupling reaction with 2
equiv. of C8F17I mediated by Cu powder (65% yield).9 The ester
4 was converted into 3 by reduction to benzylic alcohol (70%
yield) followed by oxidation (85% yield) and demethylation
(88% yield). The use of trifluoromethylbenzene as solvent
greatly improved the yield of the oxidative step,10 which was
conveniently carried out under aqueous-organic two-phase
conditions according to a procedure already reported by us.11
The free salen ligands were soluble in cold perfluorocarbons,§
CCl2FCF2Cl and Et2O, soluble in hot EtOH and only sparingly
soluble in cold halogenated solvents. The corresponding
manganese complexes (Mn–1 and Mn–2), obtained in quantita-
tive yield by refluxing an ethanolic solution of ligand with an
excess of Mn(OAc)2·4H2O, were soluble in cold perfluor-
ocarbons and CCl2FCF2Cl, but completely insoluble in com-
mon organic solvents.
Mukaiyama and co-workers have shown that optically active
(salen)Mn complexes catalyse the enantioselective epoxidation
of alkenes in the presence of molecular oxygen and a sacrificial
aldehyde.12 Fluorocarbons dissolve large quantities of molec-
ular oxygen and this property has been exploited in oxidation
reactions, including the epoxidation of alkenes under FBS
conditions in the presence of aliphatic aldehydes.7a,8b Com-
plexes Mn–1 and Mn–2 were thus first tested adapting
Mukaiyama conditions to a typical FBS procedure (see Table 1
for results and conditions). Reactions were carried out in the
dark at 20 °C under atmospheric pressure of O2, with Mn–1 and
Mn–2 immobilized in the fluorinated phase. In the absence of
the catalyst, conversions were negligible. As already reported in
The use of perfluorocarbons as reaction media in preparative
organic chemistry and homogeneous catalysis is a topic of
growing interest.1 In particular, the value of a new phase-
separation and immobilization technique known as FBS
(Fluorous Biphase Systems), developed by Horva´th and Ra´bai,2
is now becoming evident to both academic and industrial
researchers. Principles and current achievements of this tech-
nique have been recently reviewed.3 The introduction of FBS
raises a number of questions, among them the possible
application to enantioselective reactions. Besides helping
recovery and reuse of precious chiral reagents or catalysts, it has
been repeatedly proposed that the unique solvation environment
provided by perfluorocarbons might have unforeseen beneficial
effects on the selectivity of the reactions.4 However, neither
examples of enantioselective FBS reactions nor feasibility
studies have been reported so far. Metal complexes of salen
ligands are able to catalyse the epoxidation of unfunctionalized
alkenes.5 The synthetic route to salen ligands is straightforward
and can be adapted to the preparation of many different
compounds, including chiral ones.6 This offers the opportunity
to increase the very limited number of known catalysts for FBS
oxidations,7,8 and also to begin assessing the true potential of
FBS in enantioselective catalysis. For these reasons we have
synthesized two optically active (salen)MnIII complexes (Ja-
cobsen–Katsuki catalysts) selectively soluble in fluorocarbons
(Scheme 1). They were tested as catalysts in the epoxidation of
alkenes under FBS conditions, in the presence of various
oxygen donors.
Both C2 symmetric salen ligands 1 and 2 were obtained in
good yields (75 and 82%, respectively) by the condensation of
2 equiv. of the perfluoroalkylated salicylaldehyde 3 with the
Table 1 Expoxidation of alkenes with O2–pivalaldehyde catalysed by chiral
(salen)Mn complexes under FBS conditionsa
Conversion Yield Ee
t/h (%)b
(%)c
(%)
CO2H
OH C8F17
CO2Me
OMe C8F17
C8F17
CHO
OH
C8F17
Catalyst Substrate
i, ii
iii–v
Mn–1d
Mn–1d
Mn–1d
Mn–1f
Mn–1d
Mn–1d
Mn–1d
Mn–2d
Mn–2d
Mn–2d
Indene
1,2-Dihydronaphthalene
Styrene
3-Nitrostyrene
trans-b-Methylstyrene
trans-Stilbene
cis-Stilbene
Indene
1,2-Dihydronaphthalene
Styrene
2
8
5
12
5
12
12
3
100
85
100
70
100
80
88
100
95
100
83
70
86
36
75
78
85h
77
73
81
92e
I
10e
n.d.e
n.d.e
n.d.g
n.d.g
—
I
4
3
R
N
R
N
H
H
90e
R
R
vi
8
5
13e
H
H
3 +
n.d.e
H2N
NH2
C8F17
C8F17
OH HO
C8F17
a
Conditions: [catalyst] = 0.005 m in D-100; [substrate] = 0.33 m in
CH2Cl2; [pivalaldehyde] = 1 m in CH2Cl2; [N-hexylimidazole] = 0.033 m
in CH2Cl2; volume of D-100 = 3 ml; volume of CH2Cl2 = 3 ml. T = 20 °C.
Stirring rate = 1300 rpm. b Determined by capillary GC integration against
an external standard (dichlorobenzene). c Isolated epoxide. d A second run
with the recovered perfluorocarbon layer gave almost the same results.
e Determined by capillary GC using a Cyclodex-B chiral column. n.d. = not
detected. f The catalyst bleached in the first run. g Determined by 1H NMR
spectroscopy in the presence of Eu(hfc)3. h cis:trans epoxide = 3:1.
C8F17
1 R–R = –(CH2)4–
2 R = Ph
Scheme 1 Reagents and conditions: i, Me2SO4, K2CO3, acetone, reflux;
ii, C8F17I, Cu, DMF, 125 °C; iii, LAH, Et2O, 0 °C; iv, aq. NaOCl, KBr (10
mol%), TEMPO, PhCF3, 5 °C; v, BBr3, CH2Cl2 278 °C, then toom temp.;
vi, EtOH, reflux
Chem. Commun., 1998
877

View Article Online
Table 2 Epoxidation of 1,2-dihydronaphthalene catalysed by Mn–1 under
chains on the Mn–oxo intermediate responsible for the oxygen
transfer.16,17 Further synthetic efforts are required in order to
improve enantiofacial discrimination, still maintaining the
peculiar solubility of the catalysts in fluorocarbons.
We thank Ausimont S.p.A. (Milano) for a kind gift of
Galden®-D100 and the Progetto Strategico per la Difesa dai
Rischi Chimico-Industriali ed Ecologici, CNR (Rome) for
financial support.
FBS conditionsa
Conver- Yield Ee
sion
Oxidant
Solvent
t/h
0.5
24
10
4
6
6
4
T/°C
(%)b
(%)b
(%)c
d
30% H2O2
PhIO
Bu4NHSO5
MCPBA–NMOe
MCPBA–NMOe
MCPBA–NMOe
MCPBA–NMOe,f
MeCN
MeCN
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
20
20
20
278
250
220
250
—
85
95
10
70
98
90
—
65
89
9
56
86
85
—
6
8
5
7
Notes and References
† E-mail: lupo@iumchz.chimorg.unimi.it
5
71
‡ Selected data for 1: mp 108–109 °C; [a]2D0 2105.2 (c 0.5 in Et2O); dH(300
MHz, CDCl3) 13.45 (br s, 2 H), 8.29 (s, 2 H), 7.65 (d, J 2.1, 2 H), 7.47 (d,
J 2.1, 2 H), 3.48–3.37 (m, 2 H), 2.16–1.68 (m, 8 H); dF(282 MHz, CDCl3)
281.3 (t, J 10, 3 F), 2106.9 (t, J 15, 2 F), 2120.2 (br s, 2 F), 2122.1 (m,
6 F), 2123.2 (br s, 2 F), 2126.6 (br s, 2 F). For Mn–1: [a]2D0 2780 (c 0.01
in CCl2FCF2Cl); lmax(CCl2FCF2Cl)/nm 465, 385, 335. For 2: mp
48–50 °C; [a]2D0 28.8 (c 0.5 in Et2O); dH(CDCl3) 13.24 (br s, 2 H), 8.44 (s,
2 H), 7.65 (d, J 2.1, 2 H), 7.52 (d, J 2.1, 2 H), 7.27–7.05 (m, 10 H), 4.82 (s,
2 H); dF(282 MHz, CDCl3) 281.1 (t, J 10, 3 F), 2106.8 (t, J 14, 2 F),
2120.3 (br s, 2 F), 2121.9 (m, 6 F), 2123.8 (br s, 2 F), 2126.6 (br s, 2 F).
For Mn–2: [a]2D0 +90 (c 0.01 in CCl2FCF2Cl); lmax(CCl2FCF2Cl)/nm 480,
385, 330.
a Conditions: [catalyst] = 0.005 m in D-100; [substrate] = 0.1 m in CH2Cl2;
volume of D-100 = 1 ml; volume of CH2Cl2 = 1 ml; oxidant: 0.2 mmol.
Stirring rate = 1300 rpm. b Determined by capillary GC integration against
an internal standard (dichlorobenzene). c Determined by capillary GC using
d
a Cyclodex-B chiral column.
The catalyst bleached in 30 min.
e
NMO = N-methylmorpholine N-oxide (0.5 mmol). Addition of the
oxidant was carried out according to the procedure described in ref. 13.
f Substrate = indene.
§ The following commercially available perfluorocarbons were tested as
solvents: FC-72 (3M, mainly perfluorohexanes, bp 56 °C), FC-75 (3M,
mainly perfluoro-n-butyltetrahydrofuran, bp 102 °C), D-100 (Ausimont,
mainly n-perfluorooctane, bp 100 °C).
the case of the cobalt complex of a perfluoroalkylated
tetraarylporphyrin, the presence of two distinct phases did not
preclude the oxidation of the substrate dissolved in the organic
solvent.7a Both complexes were found to be active catalysts,
affording epoxides in isolated yields up to 85%. Moreover we
were able to use the catalysts in substantially lower amounts
than those required under homogenous conditions
(catalyst = 1.5% with respect to the alkene, instead of 12%).12
The brown perfluorocarbon layer recovered by simple decanta-
tion could be generally recycled in a second run without
appreciable decrease of activity. Rather surprisingly, only
indene was epoxidized with high enantioselectivity ( > 90%).
All the other alkenes that we tested gave low (15%) or even 0%
ee. The same trend was observed when reactions were carried
out in the presence of other oxygen donors more commonly
used in combination with chiral (salen)Mn. Results are
exemplified in Table 2 for the epoxidation of 1,2-dihydro-
naphthalene and indene catalysed by Mn–1. Epoxidations with
the couple m-chloroperbenzoic acid–N-methylmorpholine
N-oxide (MCPBA–NMO) were very slow at 278 °C, probably
because of hindered mass-transfer between the two liquid
phases. Although still lower than in the case of homogeneous
systems,13 reaction rates in the FBS became reasonable at
250 °C. Catalyst Mn–2 gave lower ee (58%) and conversion
(50%) with respect to Mn–1 in the MCPBA–NMO epoxidation
of indene. In the case of 1,2-dihydronaphthalene ee was
improved (15 vs. 7%), but conversion of the substrate was not
satisfactory (45% after 5 h).
Despite the low enantioselectivities generally observed with
our prototype catalysts, the FBS approach described here offers
distinct advantages over other reported methods of immobiliza-
tion of salen complexes. Note that in just one case the reported
activities, chemoselectivities and ees of heterogenized
Jacobsen–Katsuki catalysts were as high as those obtained with
one of the best homogeneous chiral (salen)Mn complexes.14
Stability of the catalyst immobilized in the fluorocarbon layer
toward bleaching is increased and its activity remains high. The
easy separation of the products from the catalyst which can be
readily reused is another considerable benefit that was not
given, for instance, by embedding salen complexes into the
pores of zeolites.15 The present results show that an improve-
ment in enantioselectivity does not necessarily follow from the
use of FBS versions of chiral catalysts. In this context, the
behaviour of Mn–1 and Mn–2 is coherent with a strong
electron-withdrawing effect exerted by the four perfluoroalkyl
1 Early work in this field; M. Vogt, Ph.D. Thesis, Rheinisch-
Westfa¨lischen Technischen Hochschule, Aachen, Germany, 1991;
D.-W. Zhu, Synthesis, 1993, 953; I. T. Horva´th and J. Ra´bai, (Exxon
Comp.), US Pat. 5 463 082, 1995.
2 I. T. Horva´th and J. Ra´bai, Science, 1994, 266, 72.
3 D. P. Curran, Chemtracts: Org. Chem., 1996, 9, 75; B. Cornils, Angew.
Chem., Int. Ed. Engl., 1997, 36, 2057.
4 A. Gladysz, Science, 1994, 266, 55; J. J. J. Juliette, I. T. Horva´th and
J. A. Gladysz, Angew. Chem., Int. Ed. Engl., 1997, 36, 1610.
5 E. G. Samsel, K. Srinivasan and J. K. Kochi, J. Am. Chem. Soc., 1985,
107, 7606.
6 E. N. Jacobsen, Catalytic Asymmetric Synthesis, ed. I. Ojima, VCH,
Weinheim, 1993, p. 159; T. Katsuki, J. Mol. Cat., 1996, 113, 87 and
references cited therein.
7 (a) G. Pozzi, F. Montanari and S. Quici, Chem. Commun., 1997, 69; (b)
G. Pozzi, M. Cavazzini, S. Quici and S. Fontana, Tetrahedron Lett.,
1997, 38, 7605.
8 (a) S. G. DiMagno, P. H. Dussault and J. A. Schultz, J. Am. Chem. Soc.,
1996, 118, 5312; (b) I. Klement, H. Lu¨tjens and P. Knochel, Angew.
Chem., Int. Ed. Engl., 1997, 36, 1454; (c) J.-M. Vincent, A. Rabion,
V. K. Yachandra and R. H. Fish, Angew. Chem., Int. Ed. Engl., 1997, 36,
2346.
9 V. C. R. McLoughlin and J. Thrower, Tetrahedron, 1969, 25, 5921;
G. Pozzi, I. Colombani, M. Miglioli, F. Montanari and S. Quici,
Tetrahedron, 1997, 53, 6145.
10 A. Ogawa and D. P. Curran, J. Org. Chem., 1997, 62, 450.
11 P. L. Anelli, F. Montanari and S. Quici, Org. Synth., 1993, Coll. Vol.
VIII, 367.
12 T. Yamada, K. Imagawa, T. Nagata and T. Mukaiyama, Chem. Lett.,
1992, 2231; K. Imagawa, T. Nagata, T. Yamada and T. Mukaiyama,
Chem. Lett., 1994, 527.
13 M. Palucki, G. J. McCormick and E. N. Jacobsen, Tetrahedron Lett.,
1995, 36, 5457.
14 I. F. J. Vankelecom, D. Tas, R. F. Parton, V. Van de Vyver and
P. A. Jacobs, Angew. Chem., Int. Ed. Engl., 1996, 35, 1346.
15 S. B. Ogunwumi and T. Bein, Chem. Commun., 1997, 901; M. J.
Sabater, A. Corma, A. Domenech, V. Forne´s and H. Garc`ıa, Chem.
Commun., 1997, 1285; L. Frunza, H. Kosslick, H. Landmesser, E. Ho¨ft
and R. Friecke, J. Mol. Catal., 1997, 123, 179.
16 A. Robert, A. Tsapara and B. Meunier, J. Mol. Catal., 1993, 85, 13;
P. J. Pospisil, D. H. Carsten and E. N. Jacobsen, Chem. Eur. J., 1996, 2,
974.
17 M.-A. Guillevic, A. M. Arif, I. T. Horva´th and J. A. Gladysz, Angew.
Chem., Int. Ed. Engl., 1997, 36, 1612.
Received in Liverpool, UK, 20th January 1998; 8/00558C
878
Chem. Commun., 1998