A macrocyclic chiral manganese(III) Schiff base complex as an efficient catalyst for the asymmetric epoxidation of olefins

Article abstract of DOI:10.1016/j.jcat.2005.06.021

A new chiral macrocyclic Mn(III)-salen complex has been prepared with two salicylidene moieties linked together to their 3 and 3′ positions by an aliphatic polyether bridge. This complex provides a highly enantioselective (up to 93%) catalyst for epoxidat

Full text of DOI:10.1016/j.jcat.2005.06.021

Journal of Catalysis 234 (2005) 250–255
www.elsevier.com/locate/jcat
A macrocyclic chiral manganese(III) Schiff base complex as an efficient
catalyst for the asymmetric epoxidation of olefins
Alexandre Martinez, Catherine Hemmert ∗, Bernard Meunier ∗
Laboratoire de Chimie de Coordination du CNRS, 205 route de Narbonne, 31077 Toulouse cedex 4, France
Received 4 May 2005; revised 14 June 2005; accepted 20 June 2005
Available online 9 August 2005
Abstract
ꢀ
A new chiral macrocyclic Mn(III)-salen complex has been prepared with two salicylidene moieties linked together to their 3 and 3
positions by an aliphatic polyether bridge. This complex provides a highly enantioselective (up to 93%) catalyst for epoxidation of cis-
disubstituted olefins with sodium hypochlorite as an oxygen atom donor and can be recycled up to three times without significant loss in
performance.
2005 Elsevier Inc. All rights reserved.
Keywords: Schiff base; Manganese; N, O ligands; Asymmetric catalysis; Epoxidation
1. Introduction
We recently developed another approach to creating more
robust catalysts, involving cyclization of the Jacobsen-type
ligand. Macrocyclization of the ligand is expected to en-
hance the stability of the corresponding complexes in cat-
alytic conditions in comparison with open-chain analogs
(e.g., Jacobsen’s catalyst Scheme 1), because of the macro-
cyclic effect. We have already reported preliminary results
on the synthesis and catalytic activity of a first series of
macrocyclic chiral Mn(salen) (e.g., complex 1, Scheme 1)
complexes for asymmetric epoxidation of cis-disubstituted
olefins [8]. But these first macrocyclic chiral Mn(III)-salen
complexes had two major drawbacks: (i) the enantiomeric
excess values obtained in the asymmetric epoxidation of cis-
disubstituted olefins were modest (with the best ee values
ranging from 42 to 74%), probably due to the absence of
bulky substituents in close proximity to positions 3 and 3ꢀ,
and (ii) the catalysts were not recyclable. The presence of an
additional oxygen atom at positions 3 and 3ꢀ of the ligand,
making an electron-rich aromatic ring easily oxidable, could
explain the fragility of these complexes in catalytic condi-
tions.
Chiral epoxides are important synthetic intermediates in
the development of new drugs, and there is much interest
in designing efficient catalysts for asymmetric olefin epox-
idation [1]. In 1990, Jacobsen [2] and Katsuki [3] indepen-
dently reported efficient optically active Mn(salen) catalysts
for enantioselective epoxidation. Since then, several chiral
Mn(salen) complexes have been developed, and high enan-
tioselectivities have been achieved in the epoxidation of con-
jugated cis-di-, cis-tri-, and some tetra-substituted olefins
[1,4]. But because these catalysts are associated with an oxy-
gen atom donor, the epoxidation generally proceeds with rel-
atively low turnover numbers, and most are unstable to pro-
longed oxidative conditions and thus are not recyclable [5].
Homogeneous chiral catalysts have been immobilized either
by anchoring the catalyst on a solid support or by using a
two-phase system [6]; however, both of these methods lead
to partial loss of activity and/or enantioselectivity in all but
a few examples [7].
Taking into account these first results, we report here on
the preparation and catalytic activity of a new, more robust
macrocyclic chiral Mn(III)-salen complex (e.g., complex 2,
*
Corresponding authors. Fax: +33-5-61-553003.
E-mail address: hemmert@lcc-toulouse.fr (C. Hemmert).
0021-9517/$ – see front matter
doi:10.1016/j.jcat.2005.06.021
2005 Elsevier Inc. All rights reserved.

A. Martinez et al. / Journal of Catalysis 234 (2005) 250–255
251
chromatograph equipped with a flame ionization detec-
tor and a Supelco cyclodextrin-β capillary column (β-dex
120, 30 m × 0.25 mm, 0.25 µm film) and coupled to a
Hewlett-Packard HP3395 integrator. 1,4-Dibromobenzene
or n-decane was used as an internal standard for the GC
analyses. The epoxides were identified by comparing the GC
data with data obtained from reaction of the corresponding
olefin with m-chloroperbenzoic acid.
2.2. Synthesis of complex 2
Compound 7 (80 mg, 0.1475 mmol) was dissolved in
50 ml EtOH under a nitrogen atmosphere. To the result-
ing solution was added, in succession, (1R,2R)-(−)-trans-
1,2-diaminocyclohexane (17 mg, 0.1475 mmol) and man-
ganese (II) diacetate tetrahydrate (36.4 mg, 0.1475 mmol).
After stirring overnight, air was bubbled through the so-
lution for 4 h. The reaction mixture was concentrated to
20 ml, treated with 20 ml of brine, and extracted with
2 × 50 ml of CH2Cl2. The organic layer was washed with
100 ml of H2O and dried over Na2SO4. After evapora-
tion of the solvent and drying under vacuum, 85 mg (82%)
of complex 2 was obtained as a dark-brown microcrys-
talline solid, C38H54N2O5ClMn (709): calcd. C 64.35, H
7.67, N 3.95, Mn 7.75, Cl 5.00; found C 64.66, H 7.61,
N 3.54, Mn 7.75, Cl 5.39. MS (ES): m/z = 673.55 [M–
Cl−]+. IR (KBr, cm−1): 1631(C=N). UV–vis (CH3OH):
λ(ε) = 274 nm (14,000 l mol−1 cm−1), 290 (13210), 318
Scheme 1. Macrocylic catalysts of first (1) and second (2) generations.
20
(8928), 354 (5714), 416 (4103). [α]D = −0.0231◦ (589 nm,
20 ◦C, 0.039 g/dm−3 in CH3OH, 10 cm path).
Scheme 1). Three criteria directed our choice: (i) we retained
the key features of the ligand (the chiral diimine, the bulky
substituents at the 5,5ꢀ positions, the C2 axis, and macrocy-
clization of the ligand); (ii) reinforced the steric hindrance
in the “south part” of the ligand by introducing bulky sub-
stituents; and (iii) suppressed oxygen atoms present in the
ligand of complex 1 at positions 3 and 3ꢀ. Moreover, we al-
ready showed that the length of the linker is an especially
important factor in the catalytic properties of these macro-
cyclic complexes, and so we chose a rather long and flexible
bridge to allow a nonplanar conformation, which is required
for the high-valent (salen)MnV=O species [8].
2.3. Representative catalytic experimental procedure
A typical reaction mixture contained 16 µl of 2,2ꢀ-di-
methylchromene (0.1 mmol) and internal standard (23.6 mg
of 1,4-dibromobenzene, 0.1 mmol) in 0.5 ml CH2Cl2,
5 µmol of the appropriate catalyst precursor (0.5 ml of a
10 mmol CH2Cl2 stock solution; catalyst/substrate ratio =
5%), and 4-phenylpyridine N-oxide (4.3 mg, 25 µmol). Af-
ter stirring at 0 ◦C for 10 min, 0.2 mmol NaOCl (0.4 ml of
a 0.5 mol aqueous NaOCl solution in 0.16 ml of a 0.05 mol
aqueous Na2HPO4 solution; 2 eq. of oxidant with respect to
the substrate) was added. After vigorous stirring for 2 h, the
reaction was diluted with water (2 ml) and CH2Cl2 (2 ml).
The layers were separated, and the organic phase was dried
over Na2SO4, filtered, concentrated to approximately 1 ml,
and analyzed by chiral GC.
2. Experimental
2.1. Instrumentation
Mass spectrometry analysis was performed on a Perkin-
Elmer SCIEX Api 365 (ES in MeOH) spectrophotome-
ter. The UV–vis spectrum was obtained on a Hewlett-
Packard 8452A diode array spectrophotometer. The in-
frared spectrum was recorded on a Perkin-Elmer GX 2000
spectrophotometer. Optical rotation was measured with a
Perkin-Elmer 241 polarimeter. Gas chromatography (GC)
analyses were performed on a Hewlett-Packard HP4890A
3. Results and discussion
3.1. Synthesis of catalyst 2
The synthesis of catalyst 2 is summarized in Scheme 2
[9]. The synthesis of 2,4-dibromo-4-tert-butylphenol (3) was
done as described previously [10]. The phenolic function

252
A. Martinez et al. / Journal of Catalysis 234 (2005) 250–255
◦
Scheme 2. Synthesis of complex 2. (a) Allyl bromide, CH Cl –H O, n-Bu NOH, 95%; (b) n-BuLi, Et O, (CH ) CO, −78 C, 65%;
2
2
2
4
2
3 2
◦
(c) NAH-THF, diethylene glycol ditosylate, DMF, 55%; (d) TMEDA, n-BuLi, Et O, DMF, −90 C, 40%; (e) Pd(PPh ) , MeOH, K CO , 86%;
2
3 4
2
3
(f) (1R,2R)-(−)-diaminocyclohexane, Mn(OAc) · 4H O, EtOH; (g) air, NaCl, 82%.
2
2
of 3 was protected with an allyl group to yield 4 [11]. Lithi-
ation of 4 with n-BuLi followed by quenching with acetone
produced 5 [12]. This key step corresponds to introduction
of the bulky substituents in positions 3 and 3ꢀ of the fi-
nal salen ligand. The polyether linker was introduced by
a Williamson reaction to yield 6 [13]. A diformylation re-
action followed by the deprotection of the phenol group
under mild conditions led to 7 [10,14]. The template syn-
thesis of complex 2 was achieved by mixing stoichiometric
amounts of 7, (1R,2R)-(−)-trans-1,2-diaminocyclohexane,
and Mn(OAc)2 · 4H2O, with subsequent air oxidation and
axial ligand exchange.
an acyclic olefin conjugated to an aryl group, the epoxida-
tion is nonstereospecific and affords a mixture of cis- and
trans-epoxides. A stepwise process with formation of a rad-
ical intermediate was proposed to rationalize these results
[16,17]. Using chromene as substrate, we tested the catalytic
activities of complex 2 with several oxygen atom donors
(Table 1, entries 10, 15, and 16). We obtained the best re-
sults with sodium hypochlorite in a quantitative yield with
a 100% selective formation of the corresponding epoxide
and an enantiomeric excess of 93% (Table 1, entry 10). The
ee values were 81% with PhIO and 82% with H2O2 (Ta-
ble 1, entries 15 and 16), the reaction being rather slow with
the green oxidant H2O2. In a general trend, the best ee val-
ues were obtained with NaOCl as an oxygen atom donor
for complex 2 (Table 1, entries 2 and 10), whereas com-
plex 1 gave a better stereoinduction with PhIO as oxidant
(Table 1, entries 5 and 14). These results suggest that the
nature of the active high-valent species will differ depend-
ing on the nature of the catalyst and the oxidant. In the case
of complex 1 associated with PhIO, the PhI group could be
involved in the transition state of the “Mn(salen)-oxo like”
species (PhIO–(salen)MnV=O or (salen)MnIV–OIPh), thus
inducing a better stereoinduction than with NaOCl. With cat-
alyst 2 bearing bulky substituents in positions 3 and 3ꢀ of the
ligand, such species are perhaps less involved, because of
steric hindrance, and a pure (salen)MnV=O could be the ma-
jor oxygen-transfer agent. But several other minor oxidizing
species should be involved, as proposed in the literature [17],
explaining the differences observed for the enantiomeric ex-
cesses.
3.2. Enantioselective epoxidation of cis-disubstituted
olefins
An evaluation of the catalytic activity of complex 2 was
performed with three cis-disubstituted olefins. Typical reac-
tion conditions were complex 2 (5 mol%), substrate (1 eq.),
an oxygen atom donor (NaOCl, PhIO, 2 eq. or H2O2, 3 eq.),
and the axial ligand 4-phenylpyridine N-oxide (4-PPNO,
5 eq. with respect to the catalyst). The reactions were car-
ried out at 0 ◦C for 2 h; the results are reported in Table 1.
With the three olefins used and sodium hypochlorite as ox-
idant, asymmetric induction was obviously increased for
complex 2 compared with catalyst 1 (Table 1, entries 1–2,
7–8, and 9–10). This is due mainly to the introduction of
bulky substituents in the close proximity to positions 3
and 3ꢀ of the ligand. With 1,2-dihydronaphthalene as sub-
strate, the yields of epoxide and naphthalene are in the same
range for both catalysts (entries 1–2). Naphthalene is the
main byproduct in this catalytic oxidation and results from
a dehydrogenation reaction [15]. For cis-β-methylstyrene
and 2,2ꢀ-dimethylchromene as substrates, catalyst 2 gives
slightly better conversions and epoxide yields (entries 7–8
and 9–10, respectively). In the case of cis-β-methylstyrene,
3.3. Recyclability of catalyst 2
We also tested the stability under the oxidative condi-
tions of complex 2. We have already reported that com-
plex 1 was not recyclable. The reuse of catalyst 2 was im-

A. Martinez et al. / Journal of Catalysis 234 (2005) 250–255
253
Table 1
Asymmetric epoxidation of cis-disubstituted olefins with 5% molar of catalyst and an oxygen atom donor
catalyst (5 mol%)
NaOCl (2 eq./substrate)
−−−−−−−−−−−−−−−−−−−→
4-PPNO (5 eq./catalyst)
room temperature, 2 h
a
b
Entry
Run
1
Catalyst
Substrate
Oxidant
NaOCl
Conversion (%)
90
Yield (%)
ee (%)
c
1
1
56 (19)
28
2
3
4
5
6
1
2
3
1
1
2
2
2
1
2
NaOCl
NaOCl
NaOCl
PhIO
80
70
70
93
49
55 (23)
55 (16)
55 (16)
62 (19)
35 (15)
60
60
58
42
35
c
PhIO
c
7
1
1
NaOCl
10
4 (4)
0
8
1
1
2
1
NaOCl
NaOCl
67
86
51 (11)
73
56
c
9
64
10
11
12
13
14
15
1
2
3
4
1
1
1
2
2
2
2
1
2
2
NaOCl
NaOCl
NaOCl
NaOCl
PhIO
100
100
100
100
68
100
100
95
88
51
93
91
90
80
74
81
82
c
PhIO
68
18
51
18
d
16
H O
2
2
◦
Reactions were carried out with substrate (0.1 mmol), catalyst (5 µmol) and oxidant (0.2 mmol) at 0 C in the presence of 5 eq. of 4-PPNO with respect to the
catalyst.
a
Epoxide yield (naphthalene or trans-β-methylstyrene oxide yield).
b
ee were determined by GC on a chiral capillary column (Supelco cyclodextrin-β); epoxide configurations are (1R,2S) for 1,2-dihydronaphthalene and
ꢀ
cis-β-methylstyrene and (3R,4R) for 2,2 -dimethylchromene.
c
Ref. [8].
0.3 mmol of oxidant was used.
d
proved with two substrates, 2,2ꢀ-dimethylchromene and 1,2-
dihydronaphtalene, and sodium hypochlorite as oxidant; at-
tempts to recycle complex 2 with cis-β-methylstyrene failed.
At the end of each run, the complex was recovered after
a workup by precipitation in hexane and then analyzed by
mass spectrometry. The catalyst 2 (5 mol%) can be used
three times in the epoxidation of 2,2ꢀ-dimethylchromene and
1,2-dihydronaphtalene without lost of conversion, yield, se-
lectivity, or enantioselectivity (Table 1, entries 2–4 and 10–
13). But a decrease of the yield and of the enantiomeric
excess was observed during the fourth run for the asym-
metric epoxidation of 2,2ꢀ-dimethylchromene, because cat-
alyst 2 underwent a partial oxidative decomposition (i.e,
slight decoloration of the organic phase due to leaching). We
also compared complex 2 and the commercially available
(S,S)-Jacobsen’s catalyst (Table 2). To estimate the turnover
numbers of the Jacobsen’s catalyst and complex 2, we de-
creased the amount of catalyst in the next two experiments.
First, we reduced the amount of catalyst to 0.05 mol% for
the asymmetric epoxidation of 2,2ꢀ-dimethylchromene with
NaOCl as oxidant at room temperature. The obtained con-
version, epoxide yield, and ee value were 61, 59, and 96%,
respectively, for the Jacobsen’s catalyst and 52, 50, and 86%,
respectively, for catalyst 2 (Table 2, entries 1–2). If the ee
value was better for the Jacobsen’s catalyst, then the selec-
tivities (97 to 95%) (Table 2, entries 1–2) and turnover num-
bers (1220 to 1040) (Table 2, entries 1–2) were in the same
range for both catalysts. Note that Katsuki previously re-
ported a very efficient metallosalen catalyst with a carboxy-
late group on the ethylene diimine moiety [18]. This catalyst
associated with iodosylbenzene as oxidant gave ee values
as high as 99% and a very high turnover number (9200;
8 days) for the asymmetric epoxidation of 2,2ꢀ-chromene
derivatives [18]. Because, in terms of epoxide yields and
ee values, 2,2ꢀ-dimethylchromene usually provides good re-
sults with a wide variety of Mn(salen) catalysts, we chose
1,2-dihydronaphtalene as the substrate for the second exper-
iment (Table 3, entries 1–5). The reactions were performed
with 1 mol% of catalyst and NaOCl as oxidant at room tem-
perature for 3 h. The Jacobsen’s catalyst lost activity quickly
and was quite ineffective after the first run, whereas cata-
lyst 2 could be recycled without loss in activity (Table 3, en-
tries 1–2 and 3–5, respectively). However, a significant drop
in epoxide yield was observed with catalyst 2 for the third
run (Table 3, entry 5). The corresponding turnover numbers
were 136 for the Jacobsen’s catalyst and 231 for catalyst 2.

254
A. Martinez et al. / Journal of Catalysis 234 (2005) 250–255
Table 2
ꢀ
Comparison of the Jacobsen’s catalyst and catalyst 2: asymmetric epoxidation of 2,2 -dimethylchromene with 0.05% molar of catalyst and sodium hypochlorite
catalyst (0.05 mol%)
NaOCl (2 eq./substrate)
−−−−−−−−−−−−−−−−−−−→
4-PPNO (5 eq./catalyst)
room temperature, 48 h
a
b
Entry
Catalyst
Conversion (%)
Yield (%)
Selectivity (%)
ee (%)
1
2
Jacobsen’s catalyst
Catalyst 2
61
52
59
50
97
96
96
86
ꢀ
Reactions were carried out with 2,2 -dimethylchromene (0.1 mmol), catalyst (0.05 µmol) and NaOCl (0.2 mmol) at room temperature in the presence of 5 eq.
of 4-PPNO with respect to the catalyst.
a
Epoxide yield.
b
ꢀ
ꢀ
ee were determined by chiral GC; major enantiomers: (3S,4S)-2,2 -dimethyl-3,4-epoxychromane with the Jacobsen’s catalyst and (3R,4R)-2,2 -dimethyl-
3,4-epoxychromane with catalyst 2.
Table 3
Recycling of Jacobsen’s catalyst and catalyst 2 (1% molar) in the asymmetric epoxidation of 1,2-dihydronaphthalene with sodium hypochlorite
catalyst (1 mol%)
NaOCl (2 eq./substrate)
−−−−−−−−−−−−−−−−−−−→
4-PPNO (5 eq./catalyst)
room temperature, 3 h
a
b
c
Entry
Run
Catalyst
Conversion (%)
Yield (%)
Selectivity (%)
ee (%)
1
2
1
2
Jacobsen’s catalyst
100
31
73 (27)
24 (0)
73
67
72
68
3
4
5
1
2
3
Catalyst 2
100
100
31
75 (25)
73 (26)
21 (4)
69
73
68
55
54
52
Reactions were carried out with 1,2-dihydronaphthalene (0.1 mmol), catalyst (1 µmol) and NaOCl (0.2 mmol) at room temperature in the presence of 5 eq. of
4-PPNO with respect to the catalyst.
a
Epoxide yield (naphthalene yield).
Selectivity of the epoxide.
b
c
ee were determined by chiral GC; major enantiomers: (1S,2R)-1,2-dihydronaphthalene oxide with the Jacobsen’s catalyst and (1R,2S)-1,2-
dihydronaphthalene oxide with catalyst 2.
In conclusion, we have prepared a new macrocyclic chi-
ral Schiff base (complex 2), involving a polyether bridge
with substituents to generate steric constrains and linked to
the 3 and 3ꢀ positions of the salicylidene moieties. This cat-
alyst, associated with NaOCl as oxidant, promotes highly
enantioselective catalytic epoxidation reaction (ee values
up to 93%) with cis-disubstituted olefins. In addition, cat-
alyst 2 can be used two to three times, depending on the
substrate used, without significant loss in performance. With
1,2-dihydronaphthalene as substrate, the macrocyclic com-
plex 2 displays a better robustness in oxidizing conditions
than the Jacobsen’s catalyst. These results validate the lig-
and macrocyclization strategy. Moreover, the synthetic strat-
egy developed here allows the modulation of the different
key groups, particularly the bulky substituents in positions
3 and 3ꢀ and the linker used to macrocyclize the ligand. For
example, introduction of bulkier substituents could enhance
stereoinduction, and a functionalized bridging arm could be
used to immobilize the corresponding catalyst on a solid
support. So structural variations on the catalyst (to increase
activity and enantioselectivity) could be readily available,
thus facilitating tuning of the catalytic properties. Extension
of this strategy to the design of new chiral macrocyclic MnIII
(salen) catalysts is currently underway.
Acknowledgments
This work was supported by the Centre National de la
Recherche Scientifique (CNRS).
Supplementary material
The online version of this article contains additional sup-
plementary material.
Please visit DOI: 10.1016/j.jcat.2005.06.021.
References
[1] (a) E.N. Jacobsen, in: E.W. Abel, F.G.A. Stone, G. Wilkinson (Eds.),
Comprehensive Organometallic Chemistry II, vol. 12, Pergamon, New
York, 1995, p. 1097;

A. Martinez et al. / Journal of Catalysis 234 (2005) 250–255
255
(b) E.N. Jacobsen, M.H. Wu, in: E.N. Jacobsen, A. Pfaltz, H. Ya-
mamoto (Eds.), Comprehensive Asymmetric Catalysis, vol. II,
Springer, New York, 1999, p. 649;
(b) K. Smith, C.-H. Liu, Chem. Commun. (2002) 886;
(c) M. Cavazzini, A. Manfredi, F. Montanari, S. Quici, G. Pozzi, Eur.
J. Org. Chem. (2001) 4639;
(c) T. Katsuki, in: I. Ojima (Ed.), Catalytic Asymmetric Synthesis, sec-
ond ed., Wiley–VCH, New York, 2000, p. 287;
(d) M.S. Taylor, E.N. Jacobsen, Proc. Natl. Acad. Sci. USA 101 (2004)
5368.
(d) C.E. Song, E.J. Roh, Chem. Commun. (2000) 837;
(e) H. Zhang, S. Xiang, C. Li, Chem. Commun. (2005) 1209.
[8] A. Martinez, C. Hemmert, H. Gornitzka, B. Meunier, J. Organomet.
Chem. 690 (2005) 2163.
[2] W. Zhang, J.L. Loebach, S.R. Wilson, E.N. Jacobsen, J. Am. Chem.
Soc. 112 (1990) 2801.
[9] A full account of the synthesis of compound 7 will be published in a
future full paper.
[3] R. Irie, K. Noda, Y. Yto, N. Matsumoto, T. Katsuki, Tetrahedron
Lett. 31 (1990) 7345.
[4] T. Katsuki, Synlett (2003) 281.
[10] K.S. Chan, J.X. Xu, F. Lam, J. Org. Chem. 61 (1996) 8414.
[11] C. Chu, A. Ramamurthy, A. Makriyannis, M.A. Tius, J. Org. Chem. 68
(2003) 55.
[12] J.J. Talley, I.A. Evans, J. Org. Chem. 49 (1984) 5267.
[13] S.C. Jha, N.N. Joshi, J. Org. Chem. 67 (2002) 3897.
[14] D.R. Vutukuri, P. Bharathi, Z.Yu.K. Rajasekaran, M.-H. Tran,
S. Thayumanavan, J. Org. Chem. 68 (2003) 1146.
[15] B. Meunier, S.P. de Visser, S. Shaik, Chem. Rev. 104 (2004) 3947.
[16] O. Bortolini, B. Meunier, J. Chem. Soc., Perkin Trans. II (1984)
1967.
[17] E.M. McGarrigle, D.G. Gilheany, Chem. Rev. 105 (2005) 1563, and
references cited therein.
[18] Y.N. Ito, T. Katsuki, Tetrahedron Lett. 39 (1998) 4325.
[5] (a) S.-H. Zhao, P.R. Ortiz, B.A. Keys, K.G. Davenport, Tetrahedron
Lett. 37 (1996) 2725;
(b) T.S. Reger, K.D. Janda, J. Am. Chem. Soc. 122 (2000) 6929.
[6] (a) L. Canali, D.C. Sherrington, Chem. Soc. Rev. 28 (1999) 85;
(b) S. Bräse, F. Lauterwasser, R.E. Ziegert, Adv. Synth. Catal. 345
(2003) 869;
(c) C.E. Song, S.-G. Lee, Chem. Rev. 102 (2002) 3495;
(d) P. McMorn, G.J. Hutching, Chem. Soc. Rev. 33 (2004) 108.
[7] (a) H. Sellner, J.K. Karjalainen, D. Seebach, Chem. Eur. J. 7 (2001)
2873;