Chiral molybdenum(VI) and tungsten(VI) 2′-pyridinyl alcoholate complexes. Synthesis, structure and catalytic properties in asymmetric olefin epoxidation

Article abstract of DOI:10.1016/S0022-328X(00)00112-1

A new class of chiral molybdenum(VI) and tungsten(VI) complexes of the type MO₂L*₂ (M = Mo, W; L* = chiral 2′-pyridinyl alcoholate), available through several synthetic pathways, and their catalytic behavior in the asymmetric epoxidation of unfunctionalized olefins are reported. MO₂Cl₂, MO₂(acac)₂, and Na₂[MO₄] (M = Mo, W) served as starting materials for the synthesis of the chiral molybdenum(VI) or tungsten(VI) complexes, respectively. The new oxo complexes were fully characterized including X-ray crystallographic analyses. The chiral 2′-pyridinyl alcoholate ligands were derived from either (-)-menthone, (-)-fenchone, (-)-camphor or (+)-camphor. For catalytic runs in the enantioselective epoxidation, trans-methylstyrene was used as model substrate and tert-butylhydroperoxide as the oxidant. The molybdenum complexes exhibit good catalytic activity and substantial optical induction. By way of contrast, the analogous tungsten complexes have low activities at comparable optical yields.

Full text of DOI:10.1016/S0022-328X(00)00112-1

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 603 (2000) 69–79
Chiral molybdenum(VI) and tungsten(VI) 2%-pyridinyl alcoholate
complexes. Synthesis, structure and catalytic properties in
asymmetric olefin epoxidation
Wolfgang A. Herrmann *, Joachim J. Haider 1, Jo¨rg Fridgen, Gerhard M. Lobmaier 2,
Michael Spiegler
Anorganisch-chemisches Institut der Technischen Uni6ersita¨t Mu¨nchen, Lichtenbergstr. 4, D-85747 Garching, Germany
Received 15 November 1999
Abstract
A new class of chiral molybdenum(VI) and tungsten(VI) complexes of the type MO2L* (M=Mo, W; L*=chiral 2%-pyridinyl
2
alcoholate), available through several synthetic pathways, and their catalytic behavior in the asymmetric epoxidation of
unfunctionalized olefins are reported. MO2Cl2, MO2(acac)2, and Na2[MO4] (M=Mo, W) served as starting materials for the
synthesis of the chiral molybdenum(VI) or tungsten(VI) complexes, respectively. The new oxo complexes were fully characterized
including X-ray crystallographic analyses. The chiral 2%-pyridinyl alcoholate ligands were derived from either (−)-menthone,
(−)-fenchone, (−)-camphor or (+)-camphor. For catalytic runs in the enantioselective epoxidation, trans-methylstyrene was
used as model substrate and tert-butylhydroperoxide as the oxidant. The molybdenum complexes exhibit good catalytic activity
and substantial optical induction. By way of contrast, the analogous tungsten complexes have low activities at comparable optical
yields. © 2000 Elsevier Science S.A. All rights reserved.
Keywords: Molybdenum; Tungsten; Asymmetric catalysis; Epoxidation; 2%-Pyridinyl alcoholate
1. Introduction
symmetric ketones. This has already been successfully
demonstrated for various 2%-pyridinyl alcoholates bear-
ing aryl or alkyl moieties, respectively (Fig. 1) [5].
The important feature of these ligands is their strong
resistance to ligand degradation which is crucial for the
application in oxidation catalysis [6]. In our case this
Recently, we described the synthesis of new dioxo-
molybdenum(VI) complexes of the type Mo2O2L2 (L=
2%-pyridinyl alcoholate) and their application as
catalysts for the selective oxidation of terminal n-alke-
nes with molecular oxygen and hydroperoxides [1,2].
We also reported the analogous dioxotungsten(VI)
complexes [3]. Other ligand systems [4] are known for
olefin epoxidation but the use of 2%-pyridinyl alcoholate
ligands is attractive because they can be easily prepared
in a broad range by the reaction of 2-lithiopyridine with
* Corresponding author. Tel.: +49-89-32093080; fax: +49-89-
28913473.
E-mail address: lit@arthur.anorg.chemie.tu-muenchen.de (W.A.
Herrmann)
1 Present address: Bayer AG, KA-F+E/KCH, F41, D-4153g Dor-
magen, Germany.
2 Present address: Gru¨nau-Illertissen GmbH, Postfach 1063, D-
89251 Illertissen, Germany.
Fig. 1. Structure of achiral molybdenum(VI) and tungsten(VI) 2%-
pyridinyl alcoholate complexes.
0022-328X/00/$ - see front matter © 2000 Elsevier Science S.A. All rights reserved.
PII: S0022-328X(00)00112-1

70
W.A. Herrmann et al. / Journal of Organometallic Chemistry 603 (2000) 69–79
chiral reducing agents such as chlorodiisopinocam-
pheylborane [10]. We decided to investigate the reaction
of 2-lithiopyridine with appropriate prochiral ketones
from the chiral pool [7] which directly leads to
diastereomerically pure 2%-pyridinyl alcohols as conve-
nient alternative.
Scheme 1. Diastereoselective synthesis of chiral 2%-pyridinyl alcohols
As it can be clearly seen from Scheme 1, the nucle-
ophilic attack of 2-lithiopyridine at carbonyl groups is
always preferred from the sterically less hindered side
which corresponds to the predicted attack according to
the model of Felkin and Ahn [11].
Due to the constraint of the molecular geometry in
the norbornane backbone or bulky side chains such as
an isopropyl group, (+)-camphor (1), (−)-fenchone
(2) and (−)-menthone (3) are high-rated candidates for
this synthetic strategy (Fig. 2).
The carbonyl groups in camphor and fenchone are
accessible from different stereosides. Camphor is obvi-
ously attacked from the endo-side due to the sterically
demanding methyl group at the C1-bridge of the nor-
bornane backbone whereas the nucleophilic addition to
fenchone is exclusively performed from the exo-side.
Since in fenchone two geminal methyl groups are lo-
cated in a-position to the carbonyl group, the addition
of 2-lithiopyridine is guided by the higher steric demand
of the C2 bridge compared to the C1-bridge.
The reaction products, i.e. the desired ligands,
(1R,2R,4R) - 1,7,7 - trimethyl - 2 - (2% - pyridinyl)bicyclo-
[2.2.1]heptan-2-ol (4) ((+)-campy), (1S,2S,4S)-1,7,7-
trimethyl-2-(2%-pyridinyl)bicyclo-[2.2.1]heptan-2-ol (5)
((−)-campy), (1R,2R,4S)-1,3,3-trimethyl-2-(2%-pyri-
dinyl)bicyclo[2.2.1]heptan-2-ol (6) ((−)-fenpy) (1S,2S,
5R)-5-methyl-2-isopropyl-1-(2%-pyridinyl)cyclohexan-
1-ol (7) ((−)-menpy), shown in Fig. 2, were prepared
and were fully characterized. The 1H-NMR signals
were completely assigned by 2D-NMR techniques and
X-ray crystallography was used to confirm the absolute
configuration of the ligand (cf. (−)-fenpy (6) in Fig. 3)
(Tables 1 and 2).
using ketones from the chiral pool.
has been achieved by the use of tertiary alcohol and
pyridine as oxidation-stable functional groups.
If unsymmetrical ketones are used as starting mate-
rial for ligand synthesis chiral 2%-pyridinyl alcoholates
are obtained [7]. For this reason we also became inter-
ested in the evaluation of our ligand concept for cata-
lytic asymmetric applications. In the field of
molybdenum chemistry several approaches for catalytic
asymmetric epoxidation have been reported but none
have yielded high enantiomeric excess [4]. The catalyti-
cally active species is normally prepared in situ from
MoO2(acac)2, tert-butylhydroperoxide and an excess of
chiral N/O or O/O-ligands, e.g. N-alkyl ephedrine [4a],
N-methylprolinol [4b] or diisopropyl tartrate [4d]. In
the case of N-methylprolinol the secondary alcohol at
the single chiral center is presumably subject to oxida-
tion which erases the chirality in the ligand [4b]. This
detrimental effect can be eliminated in the case of chiral
2%-pyridinyl alcohols. For this reason we have expanded
our studies of this synthetic strategy to chiral 2%-
pyridinyl alcohols by using a straightforward approach
that leads directly to stereoisomerically pure materials
in one step. Early results proved our strategy both in
ligand and complex synthesis and in asymmetric epoxi-
dation catalysis [8]. We now report the synthesis and
characterization of chiral complexes of the type MO2L*
2
(M=Mo, W; L*=chiral 2%-pyridinyl alcoholate) and
describe the evaluation of their catalytic potential in the
asymmetric epoxidation of unfunctionalized olefins.
2. Results and discussion
2.2. Synthesis of chiral dioxomolybdenum(VI) and
dioxotungsten(VI) complexes
2.1. Diastereoselecti6e synthesis of chiral 2%-pyridinyl
alcohols
For the development of a straightforward synthetic
route to dioxomolybdenum(VI) and dioxotungsten(VI)
2%-pyridinyl alcoholate complexes 13–20, three different
metal precursors bearing the cis-dioxo metal fragment
were tested (Scheme 2).
Starting with the dioxodichlorides MoO2Cl2 (7) or
WO2Cl2 (8) we found that, despite its poor solubility in
organic solvents, WO2Cl2 (8) reacts with 2%-pyridinyl
alcohols in refluxing THF where a THF-adduct is
presumably formed as intermediate. Both chloro
ligands are replaced by the pyridinyl alcohols to form
complexes of the type WO2L2 (L=2%-pyridinyl alcoho-
late) (17–20). The products are obtained as microcrys-
talline colorless air- and moisture-resistant solids. In the
There are two synthetic pathways for the synthesis of
enantiomerically pure 2%-pyridinyl alcohols. One strat-
egy involves the formation of a racemic mixture of
2%-pyridinyl alcohols from the addition of 2-lithiopy-
ridine to unsymmetrical ketones followed by kinetic
resolution of the enantiomers. This can be achieved by
acylation of the alcohol, enzymatic cleavage of one
enantiomer and subsequent chromatographic separa-
tion [9]. The other method uses the synthesis of a-ke-
topyridines from 2-lithiopyridine and appropriate
nitriles followed by reductive hydrogenation to a single
enantiomer of the chiral 2%-pyridinyl alcohol employing

W.A. Herrmann et al. / Journal of Organometallic Chemistry 603 (2000) 69–79
71
Fig. 2. Examples for chiral 2%-pyridinyl alcohols as ligands.
case of MoO2Cl2 (7) the molybdenum–chlorine bond is
not cleaved by our ligands. If, however, the 2%-pyridinyl
alcohol is deprotonated with n-butyllithium, the lithium
alcoholate reacts with MoO2Cl2 (7) and the desired
MoO2L2 complexes 13–16 (L=2%-pyridinyl alcoholate)
are formed.
In an alternative approach we also demonstrated the
suitability of dioxobis(acetylacetonato) compounds
MoO2(acac)2 (9) and WO2(acac)2 (10), respectively, as
versatile precursors for ligand substitution. They are
much easier to handle than the dioxodichlorides be-
cause of their high resistance to air and moisture. The
dioxobis(acetylacetonato) compounds react with two
equivalents of 2%-pyridinyl alcohol in dry methanol to
form the pyridinyl alcoholate complexes (13)–(20) of
analytical purity in high yields. Depending upon the
substituents at the quaternary a-carbon of the 2%-
pyridinyl alcohol, the complexes either precipitate im-
mediately or after partial removal of the solvent. In our
case the chiral ligands bear bulky hydrophobic hydro-
carbon substituents. Therefore the complexes precipi-
tate immediately after addition of the ligand to a
methanol solution of the respective dioxobis(acetylacet-
onato) compound and can be filtered off.
Fig. 3. PLATON-representation of (1R,2S,4S)-1,3,3-trimethyl-2-(2%-
pyridinyl)bicyclo[2.2.1]-heptan-2-ol (−)-fenpy (6). Thermal ellipsoids
represent 50% probability levels.
Table 1
Selected bond lengths (pm) and bond angles (°) of (−)-fenpy (6)
Bond lengths
Bond angles
C11–N1
C15–N1
C11–C1
O1–H
134.0(3)
133.3(3)
154.0(3)
91.0(3)
C11–N1–C15
N1–C11–C1
C11–C1–O1
O1–C1–C2
O1–C1–C6
N1–C11–C12
119.2(19)
113.7(17)
106.8(15)
112.4(16)
107.6(15)
120.8(19)
Finally, we found a way to use the metal salts
Na2[MoO4] (11) and Na2[WO4] (12) as the most inex-
pensive and easy-to-handle starting materials to prepare
the aforesaid type of 2%-pyridinyl alcoholate complexes.
A solution of the corresponding ligand in acetic acid is
C1–O1
142.7(2)

72
W.A. Herrmann et al. / Journal of Organometallic Chemistry 603 (2000) 69–79
Table 2
denum(VI) and tungsten(VI) 2%-pyridinyl alcoholate
complexes towards moisture.
Crystallographic data and measuring parameters for (−)-fenpy (6)
All molydenum and tungsten complexes of the type
MO2L2* (M=Mo, W) derived from the four chiral
ligands were prepared and fully characterized. In the
complexes 13–20, an approximately 15 ppm downfield
shift of the quaternary alcoholate carbon atom signal in
the 13C-NMR spectrum indicates the coordination to
the metal center. Another NMR probe is represented
by the C2% carbon atom in ortho position to the nitro-
gen atom at the pyridine moiety (Table 3).
Compound
(−)-fenpy (6)
Mr
231.34 (C15H21NO)
Monoclinic
2547.95(19)
753.29(3)
1368.79(10)
100.212(4)
2585.6(3)
C2 (5)
Crystal system
a (pm)
b (pm)
c (pm)
i (°)
V (106 pm3)
Space group (no.)
Dcalc
1.189
The infrared spectra show the typical symmetric and
asymmetric MꢀO stretch vibrations in the region of 896
and 934 cm−1 indicating the presence of the cisoid
dioxometal moiety in the complexes. The data obtained
from elemental analysis and mass spectroscopy confirm
the predicted structure type. At this level of characteri-
zation, however, the exact arrangement of the ligands
at the metal center is not fully determined. For this
reason we recrystallized the products from CH3OH/
CHCl3 yielding colorless crystals and X-ray crystallo-
graphic analyses were performed. ORTEP-style plots of
13 and 19 are depicted in Figs. 4 and 5 as examples for
a molybdenum and tungsten complex (Tables 4–6).
In both complexes two anionic N,O-chelating ligands
and two oxo ligands are bound to the metal adopting a
distorted octahedral coordination geometry around the
molybdenum or tungsten center, respectively. The dis-
tortion from the idealized octahedral geometry arising
from the acute bite angle of the bidentate N,O-ligands
varies from 70.8° for WO2((−)-fenpy)2 (19) to 71.5° for
MoO2((+)-campy)2 (13). As in MoO2(acac)2 (7), and
WO2(acac)2 (8), dioxomolybdenum(VI) or dioxotung-
sten(VI) 2%-pyridinyl alcoholate complexes, two oxygen
ligands and the metal form a oxometal fragment with a
cis-arrangement. With 104.93° (13) and 105.40° (19) the
bond angles of the cis-dioxo fragment remain un-
changed. The neutral N-donor ligand atoms of the
pyridine ring are in a trans-position to the cis-dioxo-
l (pm)
Z
154.178 (Cu–Ka)
8
1008
F (000)
v (cm−1
)
5.7
hkl-Range
−30/30, −9/9,
−16/0
3.8–67.8
193
4739
4535
4535
475
0.24/−0.29
−0.2(2)
0.0535
q Range (°)
T (K)
No. of reflections measured
No. of unique reflections
No. of used reflections (I/|(I)\0.001)
No. of parameters
Residual electron density (e A
−3
,
)
Flack parameter
aR1
awR2
GoF a
0.1135
1.085
a R1=ꢀ(ꢁꢁFoꢁ−ꢁFcꢁꢁ)/ꢀꢁFoꢁ, wR2=[ꢀw(ꢁFoꢁ−ꢁFcꢁ)2/ꢀwF2o]1/2
.
Table 3
Comparison of selected 13C-NMR signals as indicator for coordina-
tion of the chelating ligands at the molybdenum or tungsten center,
respectively a
Scheme 2. Synthetic pathways to complexes of the type MO2L*
2
(M=Mo, W) using different oxometal precursors.
Compound
l (COH) (ppm)
l (C2%ortho) (ppm)
MoO2((+)-campy)2 (13)
MoO2((−)-campy)2 (14)
MoO2((−)-fenpy)2 (15)
MoO2((−)-menpy)2 (16)
WO2((+)-campy)2 (17)
WO2((−)-campy)2 (18)
WO2((−)-fenpy)2 (19)
WO2((−)-menpy)2 (20)
96.41 (82.64)
96.14 (82.65)
99.62 (83.62)
92.82 (77.17)
94.81 (82.64)
95.83 (82.65)
99.42 (83.62)
92.62 (77.17)
166.07 (162.30)
166.44 (163.54)
164.94 (163.25)
169.65 (165.25)
166.06 (162.30)
167.06 (163.54)
165.57 (163.25)
170.18 (165.35)
added to an excess of the respective metal salt in
aqueous solution. The solution turns turbid and after
stirring overnight the product can be filtered off. Com-
pared to the previous approaches this method is favor-
able both for economic and ecological reasons, since
inexpensive starting materials and environmentally be-
nign solvent, water, are employed. It also successfully
demonstrates the excellent stability of the dioxomolyb-
a NMR signals of the free ligand are given in brackets.

W.A. Herrmann et al. / Journal of Organometallic Chemistry 603 (2000) 69–79
73
2.3. Catalytic results
We examined the complexes described above for their
activity in the asymmetric epoxidation of unfunctional-
ized olefins. This class of substrates is particularly
interesting because the Jacobsen method works highly
efficient only for cis-substituted olefins [14].
In the field of molybdenum-catalyzed asymmetric
epoxidation, functionalized olefins, e.g. allylic alcohols
or amides, are often chosen as substrates and yield
enantiomeric excesses up to 53% [4b] whereas for un-
functionalized olefins only one procedure is known and
that yields only a 14% e.e. [15] or only stoichiometric
procedures are known [4c]. To the best of our knowl-
edge, no asymmetric epoxidation procedure is known
for tungsten-based systems [16].
For our catalytic experiments we chose trans-methyl-
styrene as model substrate and tert-butylhydroperoxide
as oxidant (Scheme 3). In a typical catalytic run one
equivalent of substrate was mixed with 1 mol% catalyst
Table 4
Fig. 4. PLATON-representation of MoO2((+)-campy)2 (13). Thermal
Selected bond lengths (pm) and bond angles (°) of MoO2((+)-
campy)2 (13)
ellipsoids represent 50% probability levels.
Bond lengths
Bond angles
Mo–O1(sb)
Mo–O2
Mo–O3(db)
Mo–O4
Mo–N1
Mo–N2
O1–C1
C1–C11
C11–N1
O2–C21
C21–C31
C31–N2
194.8(2)
194.4(2)
170.6(2)
169.9(2)
233.9(2)
232.1(2)
144.0(4)
152.6(5)
133.9(4)
142.9(4)
153.0(4)
134.7(4)
O3–Mo–O4
O2–Mo–O4
O1–Mo–N1
O4–Mo–N2
N1–Mo–N2
O1–Mo–O2
Mo–N1–C11
N1–C11–C1
N2–C31–C21
C11–C1–O1
C31–C21–O2
C1–O1–Mo
C21–O2–Mo
105.4(11)
105.6(10)
71.6(9)
71.4(10)
79.2(8)
147.5(12)
114.5(18)
114.2(3)
113.3(3)
106.1(3)
106.9(2)
129.7(3)
130.0(18)
Table 5
Selected bond lengths (pm) and bond angles (°) of WO2((−)-fenpy)2
(19)
Fig. 5. PLATON-representation of WO2((−)-fenpy)2 (19). Thermal
ellipsoids represent 50% probability levels.
Bond lengths
Bond angles
molybdenum fragment resulting from the trans-direct-
ing effect of the oxo ligands. The chelating ligands can
bind in two different ways to the metal center forming
a D- or L-isomer. Since we employed stereoisomerically
pure ligands both isomers can be distinguished as
diastereomers. In all cases we found only one set of
signals in the 13C-NMR spectrum. This means only one
isomer is actually formed. A comparison of the crystal
structures from a set of six complexes [12,13] demon-
strates that molybdenum and tungsten complexes
derived from (+)-campy and (−)-menpy prefer the
L-form whereas the (–)-campy and (−)-fenpy-derived
complexes favor the D-form.
W–O1 (sb)
W–O2
W–O3 (db)
W–O4
W–N1
W–N2
193.1(19)
193.4(18)
173.4(2)
173.9(2)
232.5(2)
232.1(2)
142.2(3)
153.7(4)
134.0(3)
142.2(3)
153.1(4)
134.2(3)
O1–W–O2
O1–W–O3
O3–W–N1
O4–W–N2
N1–W–N2
O3–W–O4
W–N1–C11
N1–C11–C1
N2–C31–C21
C11–C1–O1
C31–C21–O2
C1–O1–W
C21–O2–W
146.7(9)
93.4(9)
70.8(8)
71.4(10)
86.9(8)
104.9(10)
115.3(18)
113.7(2)
113.9(2)
106.0(2)
106.0(2)
128.8(15)
128.5(15)
O1–C1
C1–C11
C11–N1
O2–C21
C21–C31
C31–N2

74
W.A. Herrmann et al. / Journal of Organometallic Chemistry 603 (2000) 69–79
Table 6
Table 7
Crystallographic data and measuring parameters for MoO2((+)-
campy)2 (13) and WO2((−)-fenpy)2 (19)
Comparison of the results for the molybdenum and tungsten-cata-
lyzed asymmetric epoxidation of trans-b-methylstyrene a
Compound
MoO2((+)-
campy)2 (13)
WO2((−)fenpy)2·
CH3OH (19)
Compound
Conversion (%)
e.e. (%)
MoO2((+)-campy)2 (13)
MoO2((−)-campy)2 (14)
MoO2((−)-fenpy)2 (15)
MoO2((−)-menpy)2 (16)
WO2((+)-campy)2 (17)
WO2((−)-campy)2 (18)
WO2((−)-fenpy)2 (19)
WO2((−)-menpy)2 (20)
76
73
71
81
33
34
31
36
26
25 b
15
4
24
24 b
12
4
Empirical formula
Mr
Crystal system
a (pm)
b (pm)
c (pm)
C30H40N2O4Mo C31H44N2O5W
588.60
740.58
Monoclinic
1355.38(7)
831.79(2)
1381.79(7)
116.880(5)
1389.5(1)
P21 (4)
Orthorhombic
1389.37(2)
1481.01(2)
1529.73(2)
90
3147.7(7)
P212121 (19)
1.563
i (°)
V (106 pm3)
Space group (no.)
Dcalc
a 1 mol% catalyst, two equivalents tert-butylhydroperoxide, 50°C,
1.407
,
6 h, N2, molecular sieve (4 A).
l (pm)
Z
71.073 (Mo–Ka) 71.073 (Mo–Ka)
b By use of the enantiomeric (−)-campy ligand the (R,R)-epoxide
2
4
was obtained.
F (000)
616
5.1
1504
–
v (cm−1
)
campy-substituted complexes. When the (−)-campy
substituted complexes were employed we observed the
inversion of the optical induction with formation of the
(R,R)-epoxide, as expected. This clearly shows the de-
pendence of stereoselectivity on the enantiomeric nature
of the ligand. When the chiral complexes are compared
by terms of the influence of steric demand on the
stereoselectivity, it can also be clearly seen that the
bulkier norbornane-type ligands gave the highest opti-
cal inductions. These results are encouraging but also
leave the question why the optical inductions for
molybdenum and tungsten systems are poor compared
to other systems [14]. Perhaps, this can be explained by
detrimental effects from tert-butanol which is inevitably
formed during the reaction from tert-butylhydroperox-
ide [17]. For example, the polarity of the solvent
(toluene or methylene chloride) is increased by the
formation of tert-butanol. Since for unfunctionalized
olefins only non-covalent interactions between the cata-
lytically-active species and the substrate can contribute
to the optical induction, it is quite clear that polar
solvent molecular would impair these interactions lead-
ing to lower enantiomeric excess. In a series of control
experiments we employed several aliphatic alcohols,
such as methanol, ethanol, isopropanol and tert-bu-
tanol, as the solvent. In all cases, no optical induction
was observed although the system was still catalytically
active. Another possible side effect of alcohols is their
potential to function as additional monodentate ligands
thus competing with the chelating ligand for the metal
center. Since we prepared our system in a 100:200:1
ratio (substrate:oxidant:catalyst), it is also clear that a
high excess of competitive achiral ligand is produced
during the reaction.
hkl-Range
−15/15, −9/9, −17/17, −14/14,
−16/16
2.8–24.6
193
−19/19
4.1–26.4
293
q Range (°)
T (K)
No. of reflections measured 17 015
No. of unique reflections
No. of used reflections
(I/|(I)\0.001)
5854
4520
4520
5854
5831
No. of parameters
362
530
Residual electron density
0.35/−0.46
0.84/−0.79
−3
,
(e A
)
Flack parameter
−0.04(3)
0.0286
0.0477
0.916
–
aR1
0.0157
0.0416
1.06
awR2
GoF a
a R1=ꢀ(ꢁꢁFoꢁ−ꢁFcꢁꢁ)/ꢀꢁFoꢁ, wR2=[ꢀw(ꢁFoꢁ−ꢁFcꢁ)2/SwF2o]1/2
.
Scheme 3. Epoxidation of trans-methylstyrene (M=Mo, W).
and two equivalents of tert-butylhydroperoxide (S/C
ratio 100) and the reaction was performed over 6 h at
50°C under exclusion of air and moisture. Aliquots
were taken every 10 min and were quenched by addi-
tion of manganese dioxide and magnesium sulfate,
filtered and analyzed by chiral gas chromatography.
The results are depicted in Table 7.
A comparison of the catalytic activity reveals good
conversions for all molybdenum complexes which are in
the range of 70%. The tungsten complexes, however,
were only half as active. The enantioselectivity, on the
other hand, depends more strongly on the ligand than
on the metal, e.g. complexes bearing the same ligand
type gave similar results. The highest optical induction
was 26% for the (S,S)-epoxide by using the (+)-
This becomes more apparent when the reaction
mechanism is considered. In a control experiment we
prepared a stoichiometric mixture of a chiral molybde-
num complex with trans-b-methylstyrene in the absence

W.A. Herrmann et al. / Journal of Organometallic Chemistry 603 (2000) 69–79
75
4. Experimental
of tert-butylhydroperoxide under nitrogen atmosphere.
No reaction was observed which means that the
oxo ligands of the cis-dioxomolybdenum fragment
do not transfer oxygen to the olefin. An oxometal
transfer mechanism under change of oxidation state
at the molybdenum center, as found for the Jacobsen-
type manganese salene complexes [14], can be clearly
ruled out. Therefore, it is plausible to assume the
existence of a peroxometal reaction pathway where the
d0 central metal acts as Lewis-acid to activate the
hydroperoxide [18]. In our case since the molybdenum
and tungsten centers are coordinatively saturated by an
octahedral coordination sphere and the oxo ligands are
‘chemically inert’, the reaction mechanism therefore
must involve a dissociative step with respect to the
chelating ligand.
For this reason we prepared a stoichiometric mixture
of an achiral dimethylsubstituted dioxomolybde-
num(VI)-2%-pyridinyl alcoholate complex with tert-
butylhydroperoxide and monitored the signals of the
methyl groups in a temperature row by NMR. When
the chelating ligand is bound to the metal center, the
signal of the two methyl groups was split due to their
different stereochemical environments. Only in the pres-
ence of an alkylhydroperoxide did the separate singlet
bands collapse at about 60 °C indicating the existence
of a dissociative step. Although the cleavage of a
Mo–O or Mo–N bond cannot be distinguished in this
experiment, it is plausible to assume that the Mo–O
bond is cleaved with the subsequent addition of the
hydroperoxide to the Lewis-acidic metal center. An
‘oxenoid’ oxygen is generated which is subsequently
transferred to the olefin. The stereodifferentiation is
then produced by the chiral centers through weak p–p-
interactions between the pyridine ring of the ligand and
the phenyl ring of the aromatic olefin. The catalytic
tests in protic solvents yielded no optical induction at
all to confirm this hypothesis.
4.1. General remarks
All reactions were carried out under nitrogen with
use of standard Schlenk techniques. Only freshly dis-
tilled, dry and oxygen-free solvents were used. The H-
1
and 13C-NMR spectra were recorded at 399.65 and
100.53 MHz on a FT Jeol GX 400 instrument. IR
spectra were recorded on a Perkin–Elmer 1600 FT-IR
spectrometer. Elemental analyses were performed in the
Mikroanalytische Labor of the Technical University
Munich (M. Barth). Mass spectra were recorded on a
Finnigan MAT 90-spectrometer. Catalytic runs were
monitored by chiral GC methods on a Hewlett–Pack-
ard (HP 5970 B) instrument equipped with a Chiraldex
g-TA column (Alltech),
(HP5970 B) and integration unit (HP 3394).
a mass-selective detector
WO2(acac)2 was prepared according to the procedure
from the literature [3]. MoO2(acac)2, (+)-camphor,
(−)-camphor, (−)-fenchone, (−)-menthone, 2-bro-
mopyridine, tert-butylhydroperoxide, trans-b-methyl
styrene, were purchased from Aldrich or Fluka and
used without further purification.
4.2. X-ray crystallography
Suitable single crystals for the X-ray diffraction stud-
ies were grown by standard techniques from saturated
solutions in CH3OH/CHCl3 at room temperature (r.t.).
All structures were solved by a combination of direct
methods, difference-Fourier syntheses and least-squares
methods. Neutral atom scattering factors for all atoms
and anomalous dispersion corrections for the non-hy-
drogen atoms were taken from the International Tables
for X-Ray Crystallography. All calculations were per-
formed on a DEC 3000 AXP workstation with the
STRUX-V system, including the programs PLATON-92
SIR-92 AND SHELXL-93 [19].
,
4.3. Data collection, structure solution and refinement
for the complexes 6, 13 and 19
3. Conclusion
A summary of the collection and refinement data are
reported in Table 2. Preliminary examination and data
collection were carried out in the case of 13 and 19 on
an imaging plate diffraction system (IPDS; Stoe & Cie)
equipped with a rotating anode (Nonius FR591; 50 kV;
60 mA; 3.0 kW; graphite monochromated Mo–Ka
radiation) and in the case of 6 on a four cycle diffrac-
tometer (CAD4; Nonius) equipped with a fine focus
sealed tube (50 kV; 24 mA; 1.2 kW; graphite
monochromated Cu–Ka radiation). The data collection
was performed at 193 K within the q-range of 3.8°B
qB67.8° (6), 2.8°BqB24.6° (13) and 4.1°BqB26.4°
(19). A total number of 4739 (17 015 and 5854) reflec-
Chiral 2%-pyridinyl alcoholates are suitable ligands
for the application in catalytical asymmetric olefin
epoxidation due to their high resistance to oxidative
degradation. The ligands and their molybdenum(VI)
and tungsten(VI) complexes can be obtained in good to
excellent yields by a straightforward synthesis. The
catalytic results show not only the suitability of this
new system for asymmetric epoxidation but also some
of the limitations when using tert-butylhydroperoxide
as the oxidant. Efforts to identify more effective oxi-
dants and catalysts for these reactions are being investi-
gated in our laboratories.

76
W.A. Herrmann et al. / Journal of Organometallic Chemistry 603 (2000) 69–79
tions were collected. After merging a sum of 4535, 4520
and 5854 independent reflections remained and were
used for all calculations. Data were corrected for
Lorentz and polarization effects. All ‘heavy atoms’ of
the asymmetric unit were anisotropically refined. The
hydrogen atoms were calculated in ideal positions (rid-
ing model) for 13 and 19, for 6 all hydrogen atoms were
located in difference Fourier maps and refined isotropi-
cally. Full matrix least-squares refinements were carried
out by minimizing Sw(F2o–Fc2)2 with the SHELXL
weighting scheme and stopped at shift/errB0.001. The
final refinement (on F2o) of 475 (362, 530) parameters
converged at R1=0.0535 (0.0286, 0.0157), wR2=
0.1135 (0.0477, 0.0416) and GoF=1.085 (0.916, 1.06).
ppm): l=162.3 (C2%, s), 147.37 (C6%, d), 135.53 (C4%, d),
121.57 (C3%, d), 120.58 (C5%, d), 82.64 (C2, s), 53.47 (C1,
s), 50.51 (C7, s), 45.34 (C4, d), 44.24 (C3, t), 30.70 (C6,
t), 26.98 (C5, t), 21.32 (C8, q), 21.17 (C9, q), 9.94 (C10,
q).
4.4.2. (1S,2S,4S)-1,7,7-Trimethyl-2-(2%-pyridinyl)-
bicyclo[2.2.1]heptan-2-ol (5)
Yield: 10.0 g, 36%.
EI-MS: 231 (M+), 213 (M+–H2O). IR (KBr, cm−1):
w(OH)=3370. 1H-NMR (CDCl3, 400 MHz, 298 K,
ppm): l=8.52 (H6%, d, 1H), 7.63 (H4%, dd, 1H), 7.41
(H3%, d, 1H), 7.14 (H5%, t, 1H), 5.25 (OH, s, 1H), 2.30
(H3eq, m, 1H), 2.09 (H3ax, d, 1H), 1.88 (H4, t, 1H), 1.78
(H5eq, m, 1H), 1.31 (H5ax, m, 1H), 1.28 (H6eq, m, 1H),
1.24 (H10, s, 3H), 0.88 (H8, s, 3H), 0.81 (H6ax, m, 1H),
0.78 (H9, s, 3H). 13C-NMR (CDCl3, 100 MHz, 298 K,
ppm): l=163.54 (C2%, s), 147.38 (C6%, d), 135.55 (C4%, d),
121.59 (C3%, d), 120.59 (C5%, d), 82.65 (C2, s), 53.48 (C1,
s), 50.51 (C7, s), 45.36 (C4, d), 44.24 (C3, t), 30.71 (C6,
t), 26.99 (C5, t), 21.33 (C8, q), 21.17 (C9, q), 9.94 (C10,
q).
4.4. General procedure for the synthesis of the ligands
4–7
To a solution of 200 ml dry diethylether under
nitrogen was added 80 ml of 1.6 M n-butyllithium/hex-
ane at −78°C (dry ice/isopropanol). After 10 min, 11.5
ml (0.123 mmol) of 2-brompyridine dissolved in 10 ml
of diethylether was added and the clear solution turned
to dark red. The solution was stirred for additional 30
min. After the addition of 130 mmol of the appropriate
ketone dissolved in 25 ml of diethylether the reaction
solution was stirred for 2 h and the temperature was
not allowed to raise above −40°C. The solution was
allowed to warm up to r.t. and carefully hydrolyzed by
addition of 5 ml of saturated aqueous ammonium
chloride solution. For the isolation of the reaction
product the organic phase was extracted with 5×100
ml of HCl (10%) solution until the ether phase became
almost clear. The aqueous phase was then neutralized
with a NaOH (10%) solution until an intense white
precipitate occurs followed by extraction with ether.
Finally, the organic phase was dried over sodium sul-
fate, filtered and the solvent removed under high vac-
uum resulting in a brown residue. The crude product
can be further purified by dissolving the product in a
small amount of ether, filtration over celite followed by
recrystallisation. In all four cases the product was ob-
tained as a colorless crystalline material of analytical
purity.
4.4.3. (1R,2S,4R)-1,3,3-trimethyl-2-(2%-pyridinyl)-
bicyclo[2.2.1]heptan-2-ol (6)
Yield:15.9 g, 56%.
EI-MS (m/z): 231 (M+), 213 (M+-H2O). IR (KBr,
cm−1): w(OH)=3364. 1H-NMR (CDCl3, 400 MHz,
298 K, ppm): l=8.47 (H6%, d, 1H), 7.62 (H4%, dd, 1H),
7.51 (H3%, d, 1H), 7.12 (H5%, t, 1H), 5.82 (OH, s, 1H),
3
2.32 (H6en, m, 1H), 2.22 (H7, m, J(H7, H7)=11 Hz,
5J(H7, H5en)=2 Hz, 5J(H7, H6en)=1 Hz), 1.84 (H5en, d,
1H), 1.77 (H4, d, 1H), 1.47 (H5ex, m, 1H), 1.32 (H7, d,
1H), 1.12 (H6ex, m, 1H), 0.97 (H8/9, s, 3H), 0.95 (H10, s,
3H), 0.41 (H8/9, s, 3H). 13C-NMR (CDCl3, 100 MHz,
298 K, ppm): l=163.25 (C2%, s), 146.69 (C6%, d), 134.99
(C4%, d), 123.11 (C3%, d), 121.35 (C5%, d), 83.62 (C2, s),
51.86 (C1, s), 48.87 (C4, d), 45.97 (C3, s), 42.00 (C7, t),
32.52 (C6, t), 29.20 (C9/8, q), 24.37 (C5, t), 22.24 (C9/8
q), 17.12 (C10, q).
,
4.4.4. (1S,2S,5R)-5-Methyl-2-isopropyl-1-
(2%-pyridinyl)cyclohexan-1-ol (7)
Yield: 20.6 g, 72%.
4.4.1. (1R,2R,4R)-1,7,7-Trimethyl-2-(2%-pyridinyl)-
bicyclo[2.2.1]heptan-2-ol (4)
Yield: 10.2 g, 36%.
EI-MS (m/z) 233 (M+); 215 (M+–H2O). IR (KBr,
cm−1): w(OH)=3323. 1H-NMR (CDCl3, 400 MHz,
298 K, ppm): l=8.48 (H6%, d, 1H), 7.66 (H4%, dd, 1H),
7.31 (H3, d, 1H), 7.14 (H5%, t, 1H), 5.20 (OH, s, 1H),
1.95 (H5, m, 1H), 1.85 (H4eq, m, 1H), 1.70 (H3ax, d, 1H),
1.65 (H2, m, 1H), 1.58 (H3eq, d, 1H), 1.55 (H6eq, m, 1H),
1.32 (H6ax, dd, 1H), 1.15 (H7, m, 1H), 1.02 (H4ax, 1H),
0.89 (H10, d, 3H), 0.83 (H8, d, 3H), 0.67 (H9, d, 3H).
13C-NMR (CDCl3, 100 MHz, 298 K, ppm): l=165.35
(C2%, s), 146.97 (C6%, d), 136.80 (C4%, d), 121.53 (C3%, d),
EI-MS: 231 (M+), 213 (M+–H2O). IR (KBr, cm−1):
w(OH)=3371. 1H-NMR (CDCl3, 400 MHz, 298 K,
ppm): l=8.52 (H6%, d, 1H), 7.63 (H4%, dd, 1H), 7.41
(H3%, d, 1H), 7.13 (H5%, t, 1H), 5.25 (OH, s, 1H), 2.29
(H3eq, m, 1H), 2.08 (H3ax, d, 1H), 1.88 (H4, t, 1H), 1.78
(H5eq, m, 1H), 1.31 (H5ax, m, 1H), 1.28 (H6eq, m, 1H),
1.25 (H10, s, 3H), 0.81 (H8, s, 3H), 0.74 (H6ax, m, 1H),
0.71 (H9, s, 3H). 13C-NMR (CDCl3, 100 MHz, 298 K,

W.A. Herrmann et al. / Journal of Organometallic Chemistry 603 (2000) 69–79
77
cm−1): w(MoꢀO)=902, 912. 1H-NMR (CDCl3, 400
119.25 (C5%, d), 77.17 (C1, s), 50.68 (C6, t), 50.01 (C2, d),
35.24 (C4, t), 28.47 (C5, d), 27.40 (C10, q), 23.59 (C5, t),
22.37 (C8, q), 21.93 (C3, t), 18.49 (C9, q).
3
MHz, 298 K, ppm): l=8.57 (H6%, d, J(H6%, H5%)=5.2
3
3
Hz, 1H), 7.41 (H4%, dd, J(H4%, H5%)=7.5 Hz, J(H4%,
H3%)=8.2 Hz, 1H), 7.15 (H3¦, d, J(H3%, H4%)=8.2 Hz,
3
1H), 6.99 (H5%, dd, J(H5%, H6%)=6.0 Hz, J(H5%, H4%)=
6.7 Hz, 1H), 2.78 (H3eq, m, 1H), 1.96 (H4, m, 1H), 1.88
(H3ax, m, 1H), 1.84 (H5eq, m, 1H), 1.47 (H10, s, 3H),
1.32 (H5ax, m, 1H), 1.29 (H6eq, m, 1H), 1.23 (H6ax, m,
1H), 0.93 (H8, s, 3H), 0.91 (H9, s, 3H). 13C-NMR
(CDCl3, 100 Hz, 298 K, ppm): l=166.44 (C2%), 147.90
(C6%), 136.90 (C4%), 122.31 (C3%), 121.87 (C5%), 96.14 (C2),
51.36 (C7), 50.55 (C3), 45.90 (C4), 31.19 (C6), 27.38 (C5),
21.62 (C8), 21.06 (C9), 11.78 (C10).
3
3
4.5. General procedures for the synthesis of the
dioxomolybdenum(VI) and dioxotungsten(VI) complexes
13–20
(a) To a solution of 4.3 mmol (2.2 equivalents) of
ligands 4–7 in 10 ml of dry methanol under nitrogen
atmosphere at r.t. were added 654 mg (2 mmol) (one
equivalent) dioxomolybdenum(VI)bis(acetylacetonate)
and the resulting suspension was stirred for 30 min.
After the reaction has been completed, the volume of
the solution was reduced and the reaction products
13–20 precipitate as white microcrystalline solids. The
supernatant was filtered off with a Whatman filter-
wrapped cannula and the obtained solid washed with
cold dry methanol. Finally, the product was dried
under high vacuum to give a white powder of analytic
purity (1H-NMR). No attempt was made to optimize
the product yield. For X-ray diffraction studies the
compounds 13 and 19 were recrystallized from
methanol/chloroform to give colorless crystals.
(b) An acetic acid solution of 3.6 mmol of ligands
4–7 is added to 75 ml of an aqueous solution of an
excess of metal salt, Na2[MoO4] or Na2[WO4] (10
mmol). The solution is adjusted to pH 4 by adding
hydrochloric acid or ammonia. After stirring overnight
at r.t. the product is filtered off, washed with hot
distilled water and dried in vacuo [20].
4.5.3. Bis[(1R,2S,4R)-1,3,3-trimethyl-2-(2%-pyridinyl)-
bicyclo[2.2.1]heptan-2-olato]dioxomolybdenum(VI) (15)
Yield: 929 mg, 80%.
Anal. Calc. for C30H40N2O4Mo: C, 59.68; H, 6.78; N,
4.48. Found: C, 60.74; H, 6.85; N, 4.63%. IR (KBr,
cm−1): w(MoꢀO)=921, 896. 1H-NMR (CDCl3, 400
MHz, 298 K, ppm): l=8.82 (H6%, d, 1H), 7.69 (H4%, dd,
1H), 7.61 (H3%, d, 1H), 7.17 (H5%, t, 1H), 3.40 (H3eq, m,
1H), 1.83 (H3ax, d, 1H), 1.92 (H4, dd, 1H), 1.80 (H5eq
,
m, 1H), 1.41 (H10, s, 3H), 1.31 (H5ax, m, 1H), 1.28
(H6eq, m, 1H), 1.20 (H6ax, m, 1H), 0.79 (H8, s, 3H), 0.69
(H9, s, 3H). 13C-NMR (CDCl3, 100 MHz, 298 K, ppm):
l=164.94 (C2%), 147.15 (C6%), 136.66 (C4%), 124.36 (C3%),
121.91 (C5%), 99.62 (C2), 54.74 (C1), 50.65 (C4), 49.08
(C3) 42.78 (C7), 31.81 (C6), 30.12 (C8), 24.94 (C5), 22.21
(C9), 18.88 (C10).
4.5.4. Bis[(1S,2S,5R)-5-methyl-2-isopropyl-1-
(2%-pyridinyl)cyclohexan-1-olato]diooxomolybdenum(VI)
(16)
4.5.1. Bis[(1R,2R,4R)-1,7,7-trimethyl-2-(2%-pyridinyl)bi-
cyclo[2.2.1]heptan-2-olato]dioxomolybdenum(VI) (13)
Yield: 918 mg, 79%.
Yield: 974 mg, 84%.
Anal. Calc. for C30H40N2O4Mo: C, 59.68; H, 6.78;
N, 4.48. Found: C, 59.78; H, 7.13; N, 4.50%. IR
Anal. Calc. for C30H44N2O4Mo: C, 59.80; H, 7.48; N,
4.73. Found: C, 59.78; H, 7.98; N, 4.27%. IR (KBr,
1
(KBr, cm−1): w(MoꢀO)=902, 912. H-NMR (CDCl3,
cm−1): w(MoꢀO) 916, 902 cm−1 1H-NMR (CDCl3,
.
400 MHz, 298 K, ppm): l=8.49 (H6%, d, 1H), 7.50
(H4%, dd, 1H), 7.11 (H3%, d, 1H), 6.97 (H5%, t, 1H), 2.70
(H3eq, m, 1H), 1.92 (H4, dd, 1H), 1.83 (H3ax, d, 1H),
400 MHz, 298 K, ppm): l=8.79 (H6%, d, 1H), 7.77 (H4%,
dd, 1H), 7.29 (H3%, d, 1H), 7.24 (H5%, t, 1H), 2.28 (H5,
m, 1H), 1.95 (H3,4, m, 2H), 1.68 (H2,4,6, m), 1.65 (H2, m,
1H), 1.50 (H7, m, 1H), 1.32 (H6, d, 1H), 1.04 (H3, m,
1H), 0.99 (H10, d, 3H), 0.83 (H8,9, d, 3H), 0.57 (H8,9, d,
3H). 13C-NMR (CDCl3, 100 MHz, 298 K, ppm): l=
169.65 (C2%, s), 147.70 (C6%, d), 136.74 (C4%, d), 122.31
(C3%, d), 120.35 (C5%, d), 92.81 (C1, s), 54.09 (C2, t), 50.34
(C6, t), 35.24 (C3, t), 28.54 (C5, d), 27.06 (C7, d), 24.48
(C10, q), 22.48 (C4, t), 21.98 (C8/9, q), 19.83 (C9/8, q).
1.80 (H5eq, m, 1H), 1.41 (H10, s, 3H), 1.31 (H5ax
,
m, 1H), 1.28 (H6eq, m, 1H), 1.22 (H6ax, m, 1H), 0.90
(H8, s, 3H), 0.83 (H9, s, 3H). 13C-NMR (CDCl3, 100
MHz, 298 K, ppm): l=166.07 (C2%, s), 147.74 (C6%, d),
137.09 (C4%, d), 122.29 (C3%, d), 122.20 (C5%, d), 96.41
(C2, s), 51.18 (C7, s), 50.27 (C3, t), 45.74 (C4, d), 31.03
(C6, t), 27.15 (C5, t), 21.37 (C8, q), 20.89 (C9, q), 11.61
(C10, q).
4.5.5. Bis[(1R,2R,4R)-1,7,7-trimethyl-2-(2%-pyridinyl)-
bicyclo[2.2.1]heptan-2-olato]dioxotungsten(VI) (17)
Yield: 408 g, 83%.
4.5.2. Bis[(1S,2S,4S)-1,7,7-trimethyl-2-(2%-pyridinyl)bi-
cyclo[2.2.1]heptan-2-olato]dioxomolybdenum(VI) (14)
Yield: 953 mg, 82%.
Anal. Calc. for C30H40N2O4W·H2O: C, 51.88; H,
6.10; N, 4.03. Found: C, 51.97; H, 6.12; N, 3.82%. MS
Anal. Calc. for C30H40N2O4Mo: C, 59.68; H, 6.78; N,
4.48. Found: C, 59.74; H, 7.0214; N, 4.50%. IR (KBr,

78
W.A. Herrmann et al. / Journal of Organometallic Chemistry 603 (2000) 69–79
(CI, 70 eV) m/z (%)=678.5 (0.77) [M++2], 676.5
s, 3H) 1.32 (H7ex, m, 1H), 1.22 (H6ex, m, 1H), 0.80 (H8,
s, 3H), 0.72 (H9, s, 3H). 13C-NMR (CDCl3, 100 MHz,
298 K, ppm): l=165.57 (C2%), 147.43 (C6%), 137.12 (C4%),
124.67 (C3%), 122.18 (C5%), 99.41(C2), 55.18 (C1), 50.20
(C4), 49.07 (C3), 42.97 (C7), 31.73 (C6), 29.98 (C8), 25.00
(C5), 22.40 (C9), 18.93 (C10).
(1.03) [M+], 247.3 (2.89), 231.2 (22.11), 121.0 (100). IR
1
(KBr, cm−1) w=929s [was(WꢀO)], 898s [ws(WꢀO)]. H-
NMR (CDCl3, 400 MHz, 298 K, ppm): l=8.64 (H6%,
d, 3J(H6%, H5%)=4.5 Hz, 1H), 7.59 (H4%, dd, 3J(H4%,
H3%)=9 Hz, 3J(H4%, H5%)=7 Hz, 1H), 7.20 (H3%, d,
3
3J(H3%, H4%)=6.5 Hz, 1H), 7.05 (H5%, t, J(H5%, H6%)=8
3
3
Hz, J(H5%, H4%)=7 Hz, 1H), 2.82 (H3eq, dt, J(H3ex
,
4.5.8. Bis[(1S,2S,5R)-5-methyl-2-isopropyl-1-
(2%-pyridinyl)cyclohexan-1-olato]dioxotungsten(VI) (20)
Yield: 450 g, 91%.
H3en)=13 Hz, 1H), 1.99 (H4, dd, J(H4, H3ex)=4 Hz,
3
1H), 1.88 (H5ex, m, 1H), 1.84 (H3ax, d, J(H3en, H3ex)=
3
13 Hz, 1H), 1.51 (H10, s, 3H), 1.38 (H5en, m, 1H), 1.22
(H6ex, m, 1H), 1.20 (H6en, m, 1H), 0.99 (H8, s, 3H), 0.94
(H9, s, 3H). 13C-NMR (CDCl3, 100 MHz, 298 K, ppm):
l=166.06 (C2%), 146.92 (C6%), 136.37 (C4%), 121.74 (C3%),
121,20 (C5%), 94.81 (C2), 59.42 (C1), 50.48 (C7), 49.82
(C3), 44.95 (C4), 30.07 (C6), 26.38 (C5), 20.55 (C8), 20.07
(C9), 10.85 (C10).
Anal. Calc. for C30H44N2O4W·2H2O: C, 50.29; H,
6.75; N, 3.91. Found: C, 50.19; H, 6.92; N, 3.57%. MS
(CI, 70 eV) m/z (%)=679.8 (43.61) [M+], 661.8 (35.89),
594.8 (8.54), 568.7 (2.23), 463.9 (10.93), 354.9 (2.54),
233.0 (10.89), 216.0 (100.00), 148.0 (27.42). IR (KBr,
cm−1) w=934 [was(WꢀO)], 896 [ws(WꢀO)]. 1H-NMR
(CDCl3, 400 MHz, 298 K, ppm): l=8.82 (H6%, d,
3
3J(H6%, H5%)=5.1 Hz, 1H), 7.80 (H4%, dd, J(H4%, H5%)=
3
4.5.6. Bis[(1S,2S,4S)-1,7,7-trimethyl-2-(2%-pyridinyl)-
bicyclo[2.2.1]heptan-2-olato]dioxotungsten(VI) (18)
Yield: 399 g, 81%.
6 Hz, 1H), 7.33 (H3%, d, J(H3%, H4%)=6 Hz, 1H), 7.31
(H5%, t, 3J(H5%, H4%)=6 Hz, 1H), 3.45 (H2, d, 3J(H2,
H7)=4 Hz, 1H), 2.29 (H5, m, 1H), 1.95 (H3, d, 1H),
1.71 (H6, m, 1H), 1.68 (H2, m, 1H), 1.65 (H6, m, 1H),
Anal. Calc. for C30H40N2O4W·H2O: C, 51.88; H,
6.10; N, 4.03. Found: C, 51.93; H, 6.14; N, 3.91%. MS
(CI, 70 eV) m/z (%)=678.5 (0.77) [M++2], 676.5
(1.03) [M+], 247.3 (2.89), 231.2 (22.11), 121.0 (100). IR
3
1.33 (H6, dd, J(H6ax, H6eq)=8 Hz, 1H), 1.10 (H7, m,
3
1H), 1.03 (H10, d, J(H10, H5)=6.9 Hz, 3H), 0.94 (H8,
3
3
d, J(H8, H7)=6.9 Hz, 3H), 0.59 (H9, d, J(H9, H7)=
6.6 Hz, 3H). 13C-NMR (CDCl3, 100 MHz, 298 K,
ppm): l=170.18 (C2%), 147.99 (C6%), 139.14 (C4%), 122.60
(C3%), 120.69 (C5%), 92.52 (C1), 54.06 (C6), 50.61 (C2),
35.29 (C4), 28.53 (C5), 26.91 (C10), 24.63 (C5), 22.53
(C8), 21.96 (C3), 20.00 (C9).
1
(KBr, cm−1) w=929s [was(WꢀO)], 898s [ws(WꢀO)]. H-
NMR (CDCl3, 400 MHz, 298 K, ppm): l=8.64 (H6%,
d, 3J(H6%, H5%)=5.2 Hz, 1H), 7.59 (H4%, dd, 3J(H4%,
3
H3%)=7.5 Hz, J(H4%, H5%)=7.5 Hz, 1H), 7.19 (H3%, d,
3
3J(H3%, H4%)=7.5 Hz, 1H), 7.05 (H5%, t, J(H5%, H6%)=
6.8 Hz, 3J(H5%, H4%)=6.0 Hz, 1H), 2.82 (H3eq, dt,
3J(H3ex, H3en)=13.5 Hz, 1H), 1.99 (H4, m, 1H), 1.86
(H5ex, m, 1H), 1.83 (H3ax, m, 1H), 1.56 (H10, s, 3H),
1.38 (H5en, m, 1H), 1.26 (H6ex, m, 1H), 1.23 (H6en, m,
1H), 0.99 (H8, s, 3H), 0.93 (H9, s, 3H). 13C-NMR
(CDCl3, 100 MHz, 298 K, ppm): l=167.06 (C2%),
147.92 (C6%), 137.36 (C4%), 122.74 (C3%), 122,21 (C5%),
95.83 (C2), 60.44 (C1), 51.48 (C7), 50.82 (C3), 45.91 (C4),
31.07 (C6), 27.38 (C5), 21.55 (C8), 21.06 (C9), 11.86
(C10).
4.6. General procedure for the epoxidation of
trans-i-methylstyrene with molybdenum and tungsten
compounds 13–20
A total of 200 mg (1.7 mmol) of trans-b-methyl-
styrene and 10 mg (1.0 mol%) catalyst was dissolved in
2 ml of chloroform. After the addition of 615 ml (5.5 M)
of tert-butyl hydroperoxide solution, the reaction mix-
ture was stirred for up to 16 h at 50°C.
For GC-analysis aliquots of 10 ml were taken,
quenched with manganese dioxide at 0°C on an ice
bath, diluted with 1 ml of chloroform and dried over
magnesium sulfate. The enantiomeric excess and con-
version was determined on a chiral GC column. The
products were identified by GC/MS and co-injection of
reference substances.
4.5.7. Bis[(1R,2S,4R)-1,3,3-trimethyl-2-(2%-pyridinyl)-
bicyclo[2.2.1]heptan-2-olato]dioxotungsten(VI) (19)
Yield: 434 g, 88%.
Anal. Calc. for C30H40N2O4W·H2O: C, 51.88; H,
6.10; N, 4.03. Found: C, 51.77; H, 6.47; N, 3.86%. MS
(CI, 70 eV) m/z (%)=678.8 (95.84) [M++2], 676.8
(100.0) [M+], 659.8 (16.37), 594.9 (4.41), 230.0 (4.71),
214.0 (70.16), 143.8 (4.14). IR (KBr, cm−1) w=932
[was(WꢀO)], 896 [ws(WꢀO)]. 1H-NMR (CDCl3, 400
5. Supplementary material
3
MHz, 298 K, ppm): l=8.87 (H6%, d, J(H6%, H5%)=5
3
3
Hz, 1H), 7.72 (H4%, t, J(H4%, H3%)=8 Hz, J(H4%, H5%)=
Crystallographic data (excluding structure factors)
for the structural analysis have been deposited with the
Cambridge Crystallographic Data Centre, CCDC nos.
138893 for 6, 138894 for 13 and 138895 for 19. Copies
of the data can be obtained free of charge from The
6 Hz, 1H), 7.65 (H3%, d, J(H3%, H4%)=8 Hz, 1H), 7.23
3
(H5%, t, J(H5%, H4%)=6 Hz, 1H), 2.68 (H6en, m, 1H),
3
2.24 (H7en, d, J(H7en, H7en)=10.5 Hz, 1H), 2.04 (H5en
,
2
m, 1H), 1.77 (H4, m, 1H), 1.55 (H5ex, m, 1H), 1.45 (H10,

W.A. Herrmann et al. / Journal of Organometallic Chemistry 603 (2000) 69–79
79
lytic Oxidation (ADHOC-99), York, UK, 19–23 July, 1999. (c)
W.A. Herrmann, J. Fridgen, M. Spiegler, Conference on
Organometallic Chemistry 1999 (XIIIth FECHEM), Lisbon,
Portugal, August 29th–September 3rd, 1999.
Director, CCDC, 12 Union Rd., Cambridge CB2 1EX,
UK (Fax: +44-1223-336033; e-mail: deposit@ccdc.
com.ac.uk or www: http://www.ccdc.cam.ac.uk).
[9] J.I. Uenishi, T. Hiraoka, S. Hata, K. Nishiwaki, O. Yoemitsu, J.
Org. Chem. 63 (1998) 2481.
[10] C. Bolm, M. Ewald, M. Felder, G. Schlingloff, Chem. Ber. 125
(1992) 1169.
Acknowledgements
[11] (a) N.T. Ahn, Top. Curr. Chem. 88 (1980) 145. (b) J. Mulzer,
Nachr. Chem. Tech. Lab. 32 (1984) 16. (c) M. Cherest, H.
Felkin, N. Prudent, Tetrahedron Lett. 1 (1968) 4199.
[12] (a) J.J. Haider, Ph.D. Thesis, Technische Universita¨t Mu¨nchen,
Munich, 1998. (b) W.A. Herrmann, J.J. Haider, G.M. Lobmaier,
R.W. Fischer, DE 19827046 AI.
This work was financially supported by the Deutsche
Forschungsgemeinschaft (SPP Sauerstofftransfer/Per-
oxidchemie) and the Bayerische Forschungsverbund fu¨r
Katalyse (FORKAT).
[13] Unpublished results except for complex 13 and 19.
[14] (a) E.N. Jacobsen, in: I. Ojima (Ed.) Catalytic Asymmetric
Syntheses, VCH, Weinheim, 1993, p. 159. (b) E.N. Jacobsen, W.
Zhang, A.R. Muci, D.R. Ecker, L. Deng, J. Am. Chem. Soc. 113
(1991) 7063. (c) E.N. Jacobsen, W. Zhang, L. Deng (Research
Corporation Technologies, Inc.), WO 93/03838, 1991 [Chem.
Abstr. 120 (1994) 7876v].
References
[1] W.A. Herrmann, G.M. Lobmaier, T. Priermeier, M.R. Mattner,
B. Scharbert, J. Mol. Catal. 117 (1997) 455.
[2] (a) W.A. Herrmann, G.M. Lobmaier, U. Dingerdissen, R.
Schulz (Hoechst AG), DE-A-197 29 837, 1997. (b) W.A. Herr-
mann, G.M. Lobmaier, U. Dingerdissen, B. Scharbert, R. Schulz
(Hoechst AG), DE-A-197 29 838, 1997.
[15] K. Tani, M. Hanafusa, S. Otsuka, Tetrahedron Lett. (1979)
3017.
[16] (a) C. Venturello, Chim. Ind. 75 (1993) 283. (b) K.A. Jorgensen,
Chem. Rev. 89 (1989) 431. (c) M. Quenard, V. Bonmarin, G.
Gelbard, L. Krumenacker, New. J. Chem. 13 (1989) 183. (d) O.
Bortolini, F.D. Furia, G. Modena, R. Seraglia, J. Org. Chem. 50
(1985) 2688. (e) J. Prandi, H.B. Kagan, H. Mimoun, Tetrahe-
dron Lett. 27 (1986) 2617. (f) S. Sakaguchi, Y. Nishiyama, Y.
Ishii, J. Org. Chem. 61 (1996) 5307. (g) C. Venturello, E. Alneri,
M. Ricci, J. Org. Chem. 48 (1983) 2831. (h) H.C. Stevens, A.J.
Kaman, J. Am. Chem. Soc. 87 (1965) 734. (i) L. Salles, F.
Robert, V. Semmer, Y. Jeannin, J.-M. Bre´geault, Bull. Soc.
Chim. Fr. 133 (1996) 319.
[3] W.A. Herrmann, J. Fridgen, G.M. Lobmaier, M. Spiegler, New
J. Chem. (1999) 5.
[4] (a) S.-I. Yamada, T. Mashiko, S. Terashima, J. Am. Chem. Soc.
99 (1977) 1988. (b) S. Coleman Kammula, E.T. Duim-Kollstra,
J. Organomet. Chem. 246 (1983) 53. (c) V. Schurig, K. Hintzer,
U. Leyrer, C. Mark, P. Pitchen, H.B. Kagan, J. Organomet.
Chem. 370 (1989) 81. (d) H. Brunner, H. Zintl, J. Organomet.
Chem. 411 (1991) 375. (e) K. Kaneda, S. Haruna, T. Imanaka,
M. Hamamoto, Y. Nishiyama, Y. Ishii, Tetrahedron Lett. 33
(1992) 6827.
[5] (a) B.E. Schultz, S.F. Gheller, M.C. Muetterties, M.J. Scott,
R.H. Holm, J. Am. Chem. Soc. 115 (1993) 2714. (b) W.A.
Herrmann, G.M. Lobmaier, T. Priermeier, M.R. Mattner, B.
Scharbert, J. Mol. Catal. 117 (1997) 455. (c) G.M. Lobmaier,
Ph.D. Thesis, Technische Universita¨t Mu¨nchen, Munich, 1997.
(d) I. Kim, Y. Nishihara, R.F. Jordan, R.D. Rogers, A.D.
Rheingold, G.P.A. Yap, Organometallics 16 (1997) 314. (e) J.
Fridgen, Diploma Thesis, Technische Universita¨t Mu¨nchen, Mu-
nich, 1998.
[17] R. Sheldon, in: W.A. Herrmann, B. Cornils (Eds.), Applied
Homogeneous Catalysis with Organometallic Compounds, vol.
2, VCH-Wiley, Weinheim, 1996, p. 411.
[18] R.A. Sheldon, Top. Curr. Chem. 164 (1993) 21.
[19] (a) IPDS Operating System Version 2.8, Stoe & Cie, GmbH,
Darmstadt, Germany, 1997. (b) A. Altomare, G. Cascarano, C.
Giacovazzo, A. Guagliardi, M.C. Burla. (c) G. Polidori, M.
Camalli, SIR-92, University Bari, Italy, 1992. (d) A.J.C. Wilson
(Ed.) International Tables for Crystallography, vol. C, tables
6.1.1.4 (pp. 500–502, 219–222) and 4.2.4.2 (pp. 193–199),
Kluwer Academic Publishers, Dordrecht, The Netherlands, 1992.
(e) G.M. Sheldrick, SHELXL-93, in: G.M. Sheldrick, C. Kru¨ger,
R. Goddard (Eds.), Crystallographic Computing 3, Oxford Uni-
versity Press, UK, 1993, pp. 175–189. (f) A.L. Spek, PLATON-92,
An Integrated Tool for the Analysis of the Results of a Single
Crystal Structure Determination, Acta Crystallogr. Sect. A 46
(1990) C34.
[6] T.J. Collins, Acc. Chem. Res. 27 (1994) 279.
[7] (a) G. Chelucci, F. Soccolini, Tetrahedron Asymmetry 3 (1992)
1235. (b) M. Genov, K. Kostova, V. Dimitrov, Tetrahedron
Asymmetry 8 (1997) 1869.
[8] (a) W.A. Herrmann, J.J. Haider, G.M. Lobmaier, The XVIIIth
International Conference on Organometallic Chemistry
(ICOMC), Munich, Germany, 16–21 August, 1998. (b) W.A.
Herrmann, J.J. Haider, J. Fridgen, M. Spiegler, 7th International
Symposium on Dioxygen Activation and Homogeneous Cata-
[20] T. Honjo, S. Kawasaki, Bull. Chem. Soc. Jpn. 61 (1988) 2247.
.
.