Chiral ansa-bridged η5-cyclopentadienyl molybdenum complexes: Synthesis, structure and application in asymmetric olefin epoxidation

Article abstract of DOI:10.1016/j.jorganchem.2005.10.044

Ansa-bridged η⁵-cyclopentadienyl carbonyl molybdenum complexes were synthesized with stereogenic centers located in the side chain. An exemplary X-ray crystal structure and the catalytic activity for asymmetric olefin epoxidation are reported. In non-chiral epoxidation the compounds show a good catalytic activity, comparable to activities observed for the related non-chiral complexes of composition CpMo(CO)₃X (X = Cl, CH₃). For the asymmetric epoxidation of trans-β-methylstyrene the chiral induction is up to ca. 20%. The high ring strain of the ansa-bridged system hampers, unfortunately, its stability under oxidative condition.

Full text of DOI:10.1016/j.jorganchem.2005.10.044

Journal of Organometallic Chemistry 691 (2006) 2199–2206
www.elsevier.com/locate/jorganchem
Chiral ansa-bridged g5-cyclopentadienyl molybdenum complexes:
Synthesis, structure and application in asymmetric olefin epoxidation
*
Jin Zhao, Eberhardt Herdtweck, Fritz E. Kuhn
¨
Lehrstuhl fu¨r Anorganische Chemie der Technischen Universita¨t Mu¨nchen, Lichtenbergstrasse 4, D-85747 Garching bei Mu¨nchen, Germany
Received 30 September 2005; received in revised form 20 October 2005; accepted 20 October 2005
Available online 9 December 2005
Abstract
Ansa-bridged g5-cyclopentadienyl carbonyl molybdenum complexes were synthesized with stereogenic centers located in the side
chain. An exemplary X-ray crystal structure and the catalytic activity for asymmetric olefin epoxidation are reported. In non-chiral epox-
idation the compounds show a good catalytic activity, comparable to activities observed for the related non-chiral complexes of com-
position CpMo(CO)3X (X = Cl, CH3). For the asymmetric epoxidation of trans-b-methylstyrene the chiral induction is up to ca.
20%. The high ring strain of the ansa-bridged system hampers, unfortunately, its stability under oxidative condition.
ꢀ 2005 Elsevier B.V. All rights reserved.
Keywords: Chirality; Cyclopentadienyl; Molybdenum; Olefin epoxidation
1. Introduction
have only been achieved with a few catalytic systems,
among them chiral salene manganese (III) catalysts where
up to 98% enantiomeric excess (ee) could be obtained [6].
Yamamoto et al. developed new catalytic systems based
on chiral vanadium complexes for the asymmetric epoxi-
dation of allylic alcohols very recently. Enantiomeric
excesses of up to 97% have been reached for a variety
of substrates [7].
Several other attempts to achieve chiral epoxidation,
e.g., with Mo, W and Re based catalysts have been made,
but usually led only to moderate enantiomeric excesses [8].
The generally good catalytic activities of a variety of
molybdenum (VI)-oxo complexes, however, in oxidation
reactions make this type of complexes – in principle –
promising candidates for asymmetric catalysis by replacing
the achiral by chiral ligands [9].
Chiral epoxidations are of high interest for the synthe-
sis of chiral intermediates in chemical and pharmaceutical
processes to generate enantiomerically pure products [1,2].
Mimoun et al. [3] reported on the enantioselective epoxi-
dation of prochiral alkyl-substituted olefins in 1979 utiliz-
ing a Mo(VI) complex bearing a chiral ligand, but the
observed enantioface selectivity was not high. One year
later, Katsuki and Sharpless [4] achieved the asymmetric
epoxidation of allylic alcohols, mediated by a titanium
(IV) complex using (+)-R,R or (ꢀ)-S,S tartrate as chiral
ligand. In this case, the enantioselectivity was very high,
but the titanium complex had to be applied in stoichiom-
etric amounts. The reaction was improved later, enabeling
a reduction of the catalyst:substrate ratio to ca. 1:10–1:20
and X-ray structures of the titanium tartrate catalysts
could be determined [5]. More recently, also non-func-
tionalized prochiral cis-olefins have been successfully
applied as substrates, but high enantiometric excesses
Although the ansa-bridged g5-cyclopentadienyl com-
plexes of transition metals, in which a distal methyl group
of the substituted cyclopentadienyl ligand can undergo a
cyclometallation reaction to produce metallacyclic com-
pounds, where the metal center is coordinated to both
the g5-cyclopentadienyl and the g1-alkyl group, are rela-
tively rare, some examples with group 6 and 9 metals have
been reported [10]. Eilbracht et al. [11] have described
*
Corresponding author. Fax: +49 89 289 13473.
E-mail address: fritz.kuehn@ch.tum.de (F.E. Kuhn).
¨
0022-328X/$ - see front matter ꢀ 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2005.10.044

2200
J. Zhao et al. / Journal of Organometallic Chemistry 691 (2006) 2199–2206
several examples of this kind of reaction for a wide variety
of spiro-dienes. The resulting ligand can be considered as
chelate, where the cyclopentadienyl group and the alkyl
group are both interacting with the metal center [12].
Recently, the formation of some molybdenum and tung-
sten complexes containing a linked cyclopentadienyl-ethyl
ligand has been described and the stabilities of their
metal-alkyl bond were investigated. The crystal structure
of a tungsten tricarbonyl complex has been determined
[13]. More recently, it was found that the ansa-bridged
Mo tricarbonyl complex 1 (see Scheme 1) possesses a com-
parable catalytic activity to its alkyl- and chloro-analogues
of composition Cp0Mo(CO)3X (X = alkyl or Cl) [9d,9e]. In
fact, its epoxidation activity surpasses most other Mo-
based epoxidation catalysts (e.g., of the composition
MoX2O2L2 (X = Cl, Br, Me; L = Lewis base)) significantly
and rivals even the very active and well examined Re(VII)
epoxidation catalyst methyltrioxorhenium (MTO) [14]. The
synthetic pathway used for the preparation of ansa-bridged
compounds might allow the easy introduction of a chiral
ligand instead of a non-chiral alkyl group. The presence
of a chiral group in the immediate surrounding of the metal
center might assist in controlling the stereochemistry of
reactions taking place at the metal center. Therefore, it
could eventually increase the stereoselection in the catalytic
reactions and might lead to much higher enantiomeric
excesses as have been reached in the past with related com-
plexes, where the chirality center was usually quite far away
from the metal. Furthermore, the ansa-bridge would hold
the chiral center in place, avoiding rotation of a chiral
ligand placed on the cyclopentadienyl group not fixed in
place by an ansa-bridge.
sulfonate groups by cyclopentadiene in the presence of
excess NaNH2 affords the spiro annulated diene with inver-
sion of the configuration at the chiral carbon atom [15b].
Reaction of a THF solution of the spiro cyclopentadienes
with the tricarbonyl complex Mo(CO)3(CH3CN)3 at room
temperature takes place by a coordination followed by acti-
vation of one Cdiene–C bond and transfer of the terminal R
group to the metal center, as reported previously by Eilb-
racht et al. [11] (Schemes 2 and 3).
Complex 2 is isolated as yellow crystals in ca. 75% yield
at room temperature and complex 3 is obtained as orange
red needle shaped crystals at ꢀ30 ꢁC and as an oily solid at
room temperature in ca. 60% yield. Both compounds are
stable at room temperature, can be handled in laboratory
atmosphere and may be kept under air for some hours
without significant change.
The composition and spectroscopic data of compounds
2 and 3 were determined by elemental analysis, IR- and
NMR-spectroscopy (1H, 13C, 95Mo) as well as mass spec-
troscopy (MS). In the related ligands 4 and 5, the 1H
NMR spectra display two multiplets corresponding to
an AA0BB0 spin system for the substituted cyclopentadie-
nyl ring. The four protons of the Cp ring in complex 2
and 3, however, appear as four multiplets. This feature
is also confirmed by the 13C{1H} NMR spectra, which
show five signals for the cyclopentadienyl carbon atoms.
1
This result is different from the H NMR spectra of the
molybdenum complex containing a linked cyclopentadie-
nyl-ethyl ligand (complex 1, see Scheme 1), whose Cp ring
represents an AA0BB0 spin system [13]. The substitution
on the side chain with a phenyl or methyl group leads
to the observed magnetic inequivalency of the nuclei. In
the case of complex 3, the protons of both bridging CH
groups appear as two multiplets at ꢀ0.07 and 2.39 ppm,
respectively, the one closer to the metal center being
shifted to higher field. The two substituted methyls on
In this work, we report on the synthesis of some ansa-
bridged g5-cyclopentadienyl carbonyl molybdenum com-
plexes (see Scheme 1), in which the stereogenic centers
are located in the side chain. An exemplary X-ray crystal
structure and catalytic activity tests for asymmetric olefin
epoxidation are also described.
Ph
2. Results and discussion
Ph
MsCl, NEt3
(R)
(R)
CH2Cl2
2.1. Synthesis and spectroscopic examinations
OMs
OMs
OH
OH
The synthesis of chiral derivatives of spiro cyclopentadi-
enes 4 and 5 starts from optically active 1,2-diols according
to published procedures [15]. Displacement of the methane-
(R)-1-phenyl-1,2-ethanediol
C5H6,
NaNH2
/THF
Ph
Ph
Ph
H
H
H
(R)
Me
(S)
(R)
(S)
Mo(CO)3(CH3CN)3
THF
Mo
CO
Mo
Mo
CO
Mo
(S)
H
OC
OC
OC
CO
Me
CO
CO
OC
OC
OC
CO
2
1
2
3
4
Scheme 1.
Scheme 2.

J. Zhao et al. / Journal of Organometallic Chemistry 691 (2006) 2199–2206
2201
Me
in the same region as that of the complex 1 (ꢀ1628 ppm)
OMs
Me
Me
OH
Me
[9e].
MsCl, NEt3
CH2Cl2
(R)
(R)
(R)
(R)
MsO
HO
2.2. X-ray crystal structure of compound 2
(R,R)-butanediol
The X-ray crystal structure of racemic compound 2
has been determined (see Fig. 1). The ligands are dis-
posed in a distorted four-legged piano stool fashion sim-
ilar to that established for analogous tricarbony
cyclopentadienyl group VI metal complexes [9e,13,17].
The angles between contiguous legs range from
76.67(9)ꢁ to 81.06(8)ꢁ, being typical values for this type
of structure. The cyclopentadieny ligand is bound in a
pentahapto fashion, as inferred from the total value of
the angels at the ring (540ꢁ) and the metal ring–carbon
C5H6,
NaNH2
/THF
Me
Me
H
Me
(S)
(S)
(R)
Mo(CO)3(CH3CN)3
THF
Mo
(S)
H
OC
Me
CO
OC
˚
distances, which range from 2.287(2) to 2.338(2) A. The
3
carbonyl ligands have, as expected, a lineal arrangement,
5
with Mo–C–O angles ranging from 175.3(2)ꢁ to 177.8(2)ꢁ.
Scheme 3.
˚
The C–O bond lengths between 1.144(3) and 1.153(3) A
˚
and the averaged Mo–CCO bond length of 1.997 A (with
˚
bond lengths ranging from 1.986(2) to 2.016(2) A) are
the side chain appear as two doublets (at 1.11 and
1.46 ppm, respectively), the one closer to the metal center
being shifted to higher field. The coupling constant of
usual for terminal CO groups. The slightly elongated
Mo–Csp3 bond distance of 2.347(2) is within the
range for a methyl group bound to a molybdenum
center, e.g., as reported for Mo(g5-C5HMe4)(CO)3Me
JCH –CH ¼ 6:8 Hz is in accordance with the configuration
3
of trans-protons [11g]. These finding confirm that after
the ring-opening of the spiro ligand during the reaction
with Mo(CO)3(CH3CN)3, the product complex 3 displays
a trans-configuration as the free ligand does.
˚
(2.311(2) A) [9e], suggesting an electronically relatively
saturated metal center due to the presence of the car-
bonyl groups. The bond angles at the lateral chain are
rather different from the expected values for normal
Interestingly, the two protons of the CH2 group on the
side chain of complex 2 are diastereotopic because of the
presence of a chiral center in the neighborhood, appearing
as a quadruplet at ꢀ0.11 and at 0.50 ppm, owing to the
strong shielding effect of the neighboring metal center.
1
The H NMR shift of the CH group on the side chain,
being connected with the phenyl group, appears as
pseudo-triplet at 4.03 ppm. This result also shows that
the phenyl group is located on the bridging carbon atom
in a-position to the Cp ring, and not in a-position to the
metal center. It further indicates that the C–C bond,
which is opposite to the phenyl group has been cleaved,
which is also in accordance with the literature known
ring-opening of 1,1-dimethylspiro[2,4]hepta-4,6-diene by
a Mo carbonyl complex [11g]. This molecular disposition
was confirmed by the X-ray structure of complex 3 (see
below), which further proves that the Mo center coordi-
nates to the less hindered side of the diene system.
The 95Mo chemical shift is not an appropriate tool to
distinguish between complexes in different oxidation states,
however, it is highly sensitive to structural and electronic
variations within a series of closely related mononuclear
compounds [16], allowing insight into the electronic situa-
tion at the molybdenum center. The compounds examined
in this work exhibit highly shielded chemical shifts, which
can be in general associated with low formal oxidation
states. Compounds 2 and 3 display their 95Mo NMR signal
in CDCl3 at ꢀ1728 and ꢀ1696 ppm, respectively, which are
Fig. 1. ORTEP style plot [19e] of compound 2 in the solid state. Thermal
ellipsoids are drawn at the 50% probability level. Selected bond lengths
˚
(A) and bond angles (ꢁ): Mo–C1 2.016(2), Mo–C2 1.986(2), Mo–C3
1.988(2), Mo–C4 2.347(2), Mo–C11 2.287(2), Mo–C12 2.295(2), Mo–C13
2.324(2), Mo–C14 2.338(2), Mo–C15 2.323(2), Mo–Cg 1.974; C1–Mo–C2
81.06(9), C1–Mo–C3 80.05(9), C1–Mo–C4 143.63(9), C1–Mo–Cg 123.65,
C2–Mo–C3 104.30(10), C2–Mo–C4 78.12(9), C2–Mo–Cg 126.07, C3–Mo–
C4 76.67(9), C3–Mo–Cg 125.27, C4–Mo–Cg 92.68, Mo–C4–C5 97.1(1),
C4–C5–C11 100.4(2), C4–C5–C21 116.2(2), C11–C5–C21 115.6(2). Cg
denotes the center of gravity of the cyclopentadienyl ligand.

2202
J. Zhao et al. / Journal of Organometallic Chemistry 691 (2006) 2199–2206
tetrahedral angles showing that the Mo–C(4)–C(5)–C(11)
moiety is quite strained [13].
esting because the Jacobsen method [6] works highly effi-
cient mainly for cis-olefins. For our catalytic experiments,
we chose trans-b-methylstyrene as model substrate and
TBHP as oxidant. Like several Mo(VI)-dioxo based com-
plexes as well as g5-cyclopentadienyl carbonyl molybde-
num complexes, compounds 2 and 3 show a similar,
relatively low catalytic activity at lower reaction tempera-
ture (i.e., at room temperature and below). For example,
the yield of trans-b-methylstyrene epoxide at room temper-
ature is less than 20% after 24 h. Thus, the chiral induction
of compounds 2 and 3 for the asymmetric epoxidation was
measured at 55 ꢁC, although the lower temperature is usu-
ally regarded as being beneficial for the improvement of the
chiral catalyst selectivity [8f,8j]. For the epoxidation of
trans-b-methylstyrene, compounds 2 and 3 show good
selectivity towards the epoxide. The conversion after 4 h
reaches 66% and 50%, respectively. The obtained enantio-
meric excess for trans-b-methylstyrene is up to ca. 20%.
Compared to compound 2, compound 3 displays a better
chiral induction because of the chirality center being
located in close proximity to the Mo center (see Fig. 3).
The enantiomeric excesses are in a similar order of magni-
tude as with dioxomolybdenum (VI) complexes bearing chi-
ral ligands, such as 20-pyridinyl alcoholate ligands, cis-diol,
bis-oxazoline, etc. for the asymmetric epoxidation [8]. The
moderate enantiomeric excesses reached with the latter
dioxomolybedenum (VI) complexes has been assigned to
rather weak metal–ligand-interactions [8]. However, in the
case of complexes 2 and 3, the not satisfying ee values
obtained are likely due to the sensitivity of the compounds,
originating form the quite strong tension of the cyclic system
as already indicated by X-ray crystallography. Previously, it
has been reported that the tungsten analogue of complex 1
can be oxidized by phosphorous pentachloride and
C6H5ICl2 involving a breaking of the bond between the
metal and the attached carbon atom of the cyclic system
[13]. In order to examine the stability of complexes 2 and 3
under catalytic epoxidation conditions with TBHP as oxi-
dant, kinetic 1H NMR examinations have been carried
out. The catalyst precursor complexes are mixed with a 10-
fold excess TBHP (5.5 M in decane) at room temperature
in CDCl3. To avoid the overlap of the methyl and methane
group signals with those of TBHP, complex 1 and its tung-
sten analogue were used instead of the chiral complex 2. It
was found that the signals of the two CH2 groups in the side
chain (ꢀ0.46 and 2.92 ppm for complex 1 and ꢀ0.25 and
2.91 ppm for its tungsten analogue) and the Cp ring still
remained detectable for ca. 4 h reaction time, however, the
intensity declines gradually with the ongoing reaction. After
4 h they disappear completely and the original multiplets of
the Cp ring turn into a single peak at approximately 5 ppm.
Due to the presence of strong peaks of TBHP and decane, no
more new peaks could be observed. A similar observation
was obtained also in the case of complex 3, the CH2 and
Ph groups attached to the bridged chain as well as the multi-
plets originating from the Cp ring remain for a few hours and
then disappear, involving the appearance of new peak at ca.
2.3. Application in epoxidation catalysis
Compounds 1–3 were tested as catalysts for the epoxida-
tion of cyclooctene with TBHP. The details concerning the
catalytic reaction are given in Section 3. Blank reactions
show that no significant amount of epoxide was formed
in the absence of catalyst. A catalyst:oxidant:substrate
ratio of 1:200:100 was used in all experiments unless stated
otherwise. The catalysts were first stirred with TBHP until
a color change from orange to yellow occurred, indicative
for the oxidation of the carbonyl complexes to the corre-
sponding dioxo Mo(VI) compounds, as has been examined
and described before [9d,9e]. For cyclooctene no significant
formation of by-products (e.g., diol) was observed. Both
catalytic reactions show similar time-dependent curves, in
which the yield increases steadily in the first 2 h of the reac-
tion and then slows down (first-order kinetics). The curves
for complexes 1 and CpMo(CO)3Me were also measured
for sake of comparison (Fig. 2).
Compounds 1–3 and CpMo(CO)3Me show a similar
behavior indicating that introduction of substituents on
the side chain between the metal center (as can be also con-
cluded from the 95Mo NMR of the carbonyl precursors)
and the Cp ring does not influence much the electronic sit-
uation at the metal center and therefore also not signifi-
cantly the catalytic performance. Comparison with
CpMo(CO)3Me indicates that the replacement of a methyl
group by a ansa-bridge does not strongly influence the
overall catalytic performance, but leads to a somewhat
slower reaction. This observation may be caused by the
more pronounced steric hindrance of the ansa-bridged sys-
tems. Similar observations for more sterically crowded
related system have been reported before [9f].
We examined the compounds 2 and 3 for their catalytic
activity in the asymmetric epoxidation of unfunctionalized
trans-olefins. This class of substrates is particularly inter-
100
80
60
40
20
0
0
50
100
150
200
Time (min)
Fig. 2. Time-dependent yield of cyclooctene epoxide in the presence of
compounds 1 (closed triangles), 2 (closed squares), 3 (closed diamonds)
and CpMo(CO)3Me (closed circles) as catalysts at 55 ꢁC with 1 mol%
catalyst charge.

J. Zhao et al. / Journal of Organometallic Chemistry 691 (2006) 2199–2206
2203
25
20
15
10
5
0
0
30
60
90
120
150
180
210
240
Time (min)
Fig. 3. Time-dependent enantiomeric excess (ee) during 4 h reaction time for the asymmetric epoxidation of trans-b-methylstyrene in the presence of
compounds 2 (closed squares), 3 (closed diamonds) as catalysts at 55 ꢁC with 1 mol% catalyst charge.
5 ppm. Furthermore, the appearance of a bluish compound
can be observed, which has been identified as a mixture of
molybdenum oxide decomposition products in related reac-
tions [8,9,14]. The formation of dimeric compounds of for-
mula [Cp0MoO2]2O has been described as additional
reaction pathway for related reactions [9f,18]. Given the ring
strain in compounds 1–3, it is likely that more side reactions
appear at the same time, leading to the inconclusive appear-
ance of the NMR spectra, the reduction of the catalytic activ-
ities during the course of the reaction and the decreasing ees.
The comparatively low ee values are therefore most likely
due to the breaking of the carbon–metal under the oxidative
conditions leading to a significant decomposition of the cat-
alyst and accordingly to the disappearance of the chiral cen-
ters during the course of the catalytic reaction. A promising
way to obtain higher ee values might be the use of chiral ansa-
bridged g5-cyclopentadienyl complexes with lower ring ten-
sion (e.g., a five- or six-membered ring). The importance of
the size of the ansa-bridge will become much clearer during
the execution of these experiments. Work in this direction
is currently under way in our laboratories.
MAT 311 A and a MAT 90 spectrometer. Catalytic runs
were monitored by GC methods on a Hewlett–Packard
instrument HP 5890 Series II equipped with a FID, a Supe-
lco column Alphadex 120 and a Hewlett–Packard integra-
tion unit HP 3396 Series II. Compounds (R)-1-phenyl-1,
2-ethanediol bis(methanesulfonate), (R)-1-phenyl-spiro
[2,4]hepta-4,6-diene (4), (R,R)-2,3-butanediol bis (methane-
sulfonate) and (S,S)-1,2-dimethyl-spiro[2,4]hepta-4,6-diene
(5) were synthesised according to the literature procedures
[15].
3.1.1. (S)-[Mo(g5-C5H4CHPh-g1-CH2)(CO)3] (2)
The addition of a THF (ca. 20 mL) solution of the
ligand (R)-1-phenyl-spiro[2,4]hepta-4,6-diene (0.60 g, 3.3
mmol) to Mo(CO)3(CH3CN)3 (0.91 g, 3.0 mmol) at 0 ꢁC
produces an orange solution, which is stirred overnight at
r.t. All volatiles are removed in vacuo, and the sticky resi-
due is extracted with 15 mL hexane (three times) and fil-
tered. The obtained orange red filtrates are concentrated
and chromatographed on Florisil (60–100 mesh). The
orange yellow fraction is eluted with n-hexane and col-
lected. After cooling to ꢀ30 ꢁC yellow crystals are
obtained. Yield (1.68 g, 75%). Anal. Calc. for C16H12O3Mo
(348): C, 55.17; H, 3.45. Found: C, 55.28; H, 3.37%. IR
(KBr, m cmꢀ1): 3110 (w, m(CH) of Cp-ring), 2961, 2932,
2891 and 2859 (w, m(CH3) and m(CH2)), 2003.6, 1903.9
3. Experimental
3.1. Synthesis and characterization
All preparations and manipulations were performed
using standard Schlenk techniques under an argon
atmosphere. Solvents were dried by standard procedures
(THF, n-hexane and Et2O over Na/benzophenone; CH2Cl2
over CaH2), distilled under argon and used immediately
(vs, m(CO)); 1H NMR (CDCl3, 400 MHz, r.t.):
d
(ppm) = ꢀ0.11 (q, 1H, Mo–CH), 0.50 (q, 1H, Mo–CH),
4.03 (t, 1H, Cp–CH–Ph), 5.29 (m, 1H, Cp), 5.28 (m, 1H,
Cp), 5.21 (m, 1H, Cp) and 5.13 (m, 1H, Cp), 7.24, 7.37,
7.36 and 7,35 (m, 5H, C6H5); 13C NMR (CDCl3,
100.28 MHz, r.t.): d (ppm) = 222.4 (CO) 151.1, 144.9,
128.5, 128.4, 126.7, 126.1 (C6H5), 90.3, 88.6, 88.59, 87.9,
75.1 (C5H4), 38.57 (Cp–CH), ꢀ36.41 (Mo–CH2); 95Mo
NMR (CDCl3, 26.07 MHz, r.t.): d (ppm) = ꢀ1728; FAB-
MS (70 eV) m/z (%); M+ = 348.
˚
(THF) or kept over 4 A molecular sieves. TBHP was pur-
chased from Aldrich as 5.0–6.0 mol% solution in decane
and used after drying over molecular sieves to remove the
water (<4% when received). Microanalyses were performed
in the Mikroanalytisches Labor of the TU Munchen in
¨
Garching (Mr. M. Barth). Mid-IR spectra of isolated com-
pounds were measured on a Bio-Rad FTS 525 spectrome-
3.1.2. (R,S)-[Mo(g5-C5H4CHMe-g1-CHMe)(CO)3] (3)
The addition of a THF (ca. 20 mL) solution of the
ligand (S,S)-1,2-dimethyl-spiro[2,4]hepta-4,6-diene (0.35 g,
2.9 mmol) to Mo(CO)3(CH3CN)3 (0.80 g, 2.65 mmol) at
1
ter using KBr pellets. H, 13C, and 95Mo NMR spectra
were obtained using a 400-MHz Bruker Avance DPX-400
spectrometer. Mass spectra were obtained with a Finnigan

2204
J. Zhao et al. / Journal of Organometallic Chemistry 691 (2006) 2199–2206
0 ꢁC produces an orange solution, which is stirred over-
night at r.t. All volatiles are removed in vacuo, and the
sticky residue is extracted with 15 mL hexane (three times)
and filtered. The obtained orange red filtrates are concen-
trated and chromatographed on Florisil (60–100 mesh).
The orange yellow fraction is eluted with n-hexane and col-
lected. After removal of all solvent, compound 3 is
obtained as orange red oily solid. After cooling to
ꢀ30 ꢁC thermally unstable orange-red needles are formed.
Yield (1.68 g, 60%). Anal. Calc. for C12H12O3Mo (300): C,
48.00; H, 4.00. Found: C, 48.17; H, 4.18%. IR (KBr, m
cmꢀ1): 3118 (w, m(CH) of Cp-ring), 2969, 2938, 2897 and
2854 (w, m(CH3) and m(CH2)), 1999.5, 1902.4 (vs, m(CO));
1H NMR (CDCl3, 400 MHz, r.t.): d (ppm) = ꢀ0.07 (m,
1H, Mo–CH), 1.11 (d, J = 6.8 Hz, 3H, Mo–CH–CH3),
1.46 (d, J = 6.8 Hz, 3H, Cp–CH–CH3), 2.39 (m, 1H, Cp–
CH), 5.21 (m, 1H, Cp), 5.18 (m, 1H, Cp), 5.16 (m, 1H,
Cp) and 5.12 (m, 1H, Cp); 13C NMR (CDCl3,
100.28 MHz, r.t.): d (ppm) = 224 (CO), 89.90, 88.65,
87.74, 86.73, 74.26 (C5H4), 38.18 (Cp–CH), 24.41 (Cp–
CH–CH3), 20.79 (Mo–CH–CH3), ꢀ17.48 (Mo–CH);
95Mo NMR (CDCl3, 26.07 MHz, r.t.): d (ppm) = ꢀ1696;
FAB-MS (70 eV) m/z (%); M+ = 300.
for the X-ray diffraction study were grown from hexane.
A clear yellow plate was stored under perfluorinated
ether, transferred in a Lindemann capillary, fixed, and
sealed. Preliminary examination and data collection were
carried out on
a
kappa-CCD device (NONIUS
MACH3) with an Oxford Cryosystems cooling device
at the window of a rotating anode (NONIUS FR591)
with graphite monochromated Mo Ka radiation
˚
(k = 0.71073 A). Data collection was performed at
173 K within the H range of 2.15ꢁ < H < 25.30ꢁ. A total
of 68512 reflections were integrated, corrected for Lor-
entz, polarization, and, arising from the scaling proce-
dure, corrected for latent decay and absorption effects.
After merging (Rint = 0.058), 2497 [2210: Io > 2r(Io)]
independent reflections remained and all were used to
refine 229 parameters. The structure was solved by a
combination of direct methods and difference-Fourier
syntheses. All non-hydrogen atoms were refined with
anisotropic displacement parameters. All hydrogen
atoms were found and refined with individual isotropic
displacement parameters. Full-matrix least-squares refine-
P
2
ments were carried out by minimizing
wðF 2o ꢀ F 2cÞ
and converged with R1 = 0.0230 [Io > 2r(Io)], wR2 =
0.0497 (all data), GOF = 1.106, and shift/error < 0.001.
The final difference-Fourier map shows no striking fea-
tures [19].
3.2. Single-crystal X-ray structure determination of racemic
compound 2
Crystal data and details of the structure determina-
tion are presented in Table 1. Suitable single crystals
3.3. Catalytic reactions
The catalytic reactions were performed under an air
atmosphere, in a reaction vessel equipped with a magnetic
stirrer, immersed into a 55 ꢁC thermostated bath.
Table 1
Crystallographic data for [Mo(g5-C5H4CHPh-g1-CH2)(CO)3] (2)
Achiral catalytic epoxidation: cis-cyclooctene (800 mg,
7.3 mmol), mesitylene (1 g, internal standard), 1 mol%
(73 lmol) of compounds 2 or 3 as catalysts were added
to the reaction vessel. With the addition of TBHP
(2.65 mL, 5.5 M in n-decane) the reaction was started.
The course of the reactions was monitored by quantitative
GC analysis. Samples were taken and diluted with CH2Cl2,
and treated with a catalytic amount of MgSO4 and MnO2
to remove water and destroy the excess of peroxide, respec-
tively. The resulting slurry was filtered and the filtrate
injected into a chiral GC column. The conversion of
cyclooctene, and the formation of cyclooctene oxide were
calculated from calibration curves (r2 = 0.999) recorded
prior to the reaction course.
Compound
2
Formula
Formula weight
Color/habit
C
348.20
16H12MoO3
Yellow/plate
0.08 · 0.15 · 0.46
Orthorhombic
Pbca (No. 61)
12.7778(1)
7.5948(1)
28.3288(3)
2749.16(5)
8
173
1.683
0.957
1392
Crystal dimensions (mm)
Crystal system
Space group
˚
a (A)
˚
b (A)
˚
c (A)
3
˚
V (A )
Z
T (K)
Dcalc (g cmꢀ3
)
l (mmꢀ1
F(000)
)
H range (ꢁ)
Index ranges (h,k,l)
Number of reflections collected
Number of independent reflections (Rint
2.15–25.30
15, 9, 34
68512
2497 (0.058)
2210
2497/0/229
0.0230/0.0480
0.0294/0.0497
1.106
Chiral catalytic epoxidation: trans-b-methylstyrene
(200 mg, 1.7 mmol), mesitylene (100 mg, internal stan-
dard), and 1 mol% (17 lmol) of the compounds 2 or 3
as catalysts and 2 mL toluene as solvent were added to
the reaction vessel. With the addition of TBHP
(0.62 mL, 5.5 M in n-decane) the reaction started. The
course of the reactions was monitored by quantitative
GC analysis. The samples were processed as described
above. The enantiomeric excess was calculated with the
ratio of the peaks corresponding to both epoxides
formed.
)
Number of observed reflections (I > 2r(I))
Number of data/restraints/parameters
R1/wR2 [I > 2r(I)]a
R1/wR2 (all data)a
Goodness-of-fit (GOF)b on F2
ꢀ3
˚
Largest difference in peak and hole (e A
)
+0.34 and ꢀ0.35
P
P
P
P
2
2
1=2
a
R1 = (iFo| ꢀ |Fci)/ |Fo|; wR2 ¼ f ½wðF 2o ꢀ F c2Þ ꢁ= ½wðF o2Þ ꢁg
.
P
2
1=2
b
GOF ¼ f ½wðF 2o ꢀ F 2c Þ ꢁ=ðn ꢀ pÞg
.

J. Zhao et al. / Journal of Organometallic Chemistry 691 (2006) 2199–2206
2205
[2] R.A. Sheldon, in: B. Cormils, W.A. Herrmann (Eds.), Applied
Homogeneous Catalysis with Organometallic Compounds, second
ed., Wiley–VCH, Weinheim, 2002, pp. 412–426.
[3] H.B. Kagan, H. Mimoun, C. Marc, V. Schurig, Angew. Chem., Int.
Ed. Engl. 18 (1979) 485.
[4] T. Katsuki, K.B. Sharpless, J. Am. Chem. Soc. 102 (1980) 5974.
[5] I.D. Williams, S.F. Pederson, K.B. Sharpless, S.J. Lippard, J. Am.
Chem. Soc. 106 (1984) 6430.
[6] (a) M. Palucki, P.J. Pospisil, W. Zhang, E.N. Jacobsen, J. Am.
Chem. Soc. 116 (1994) 9333;
(b) P. Pietika¨inen, Tetrahedron 54 (1998) 4319.
[7] (a) N. Murase, Y. Hoshino, M. Oishi, H. Yamamoto, J. Org. Chem.
64 (1999) 338;
4. Conclusions
The ansa-bridged g5-cyclopentadienyl carbonyl molyb-
denum complexes 1–3 were synthesized and in the case of
complexes 2 and 3, stereogenic centers are located on the
side chain. The X-ray crystal structure of 2 shows a dis-
torted four-legged piano stool conformation similar to
that established for analogous tricarbonyl cyclopentadie-
nyl molybdenum complexes. Compounds 1–3 show a sim-
ilar catalytic behavior and similar 95Mo NMR shifts
indicating that the introduction of substitutes on the side
chain between metal center and the Cp ring does not influ-
ence much the electronic situation at the metal center and
therefore also not the catalytic performance. Comparison
to CpMo(CO)3Me, indicates that the replacement of a
methyl group by an ansa-bridge does not very strongly
influence the overall catalytic performance, but leads to
a somewhat slower reaction due to the increased steric
hindrance. For the asymmetric epoxidation of trans-b-
methylstyrene, the chiral induction is up to 20% and com-
pound 3 displays a better chiral induction, most likely
because of the closer proximity of the chiral center to
the metal atom. The ring strain leads to somewhat sensi-
tive systems that break down slowly under the oxidative
reaction conditions.
(b) Y. Hoshino, H. Yamamoto, J. Am. Chem. Soc. 122 (2000) 10452;
(c) Y. Hoshino, N. Murase, M. Oishi, H. Yamamoto, Bull. Chem.
Soc. Jpn. 73 (2000) 1653;
(d) N. Makita, Y. Hoshino, H. Yamamoto, Angew. Chem. 115
(2003) 971;
(e) W. Zhang, A. Basak, Y. Kosugi, Y. Hoshino, H. Yamamoto,
Angew. Chem. Int. Ed. 44 (2005) 4389;
(f) Z. Bourhani, A.V. Malkov, Chem. Commun. 36 (2005) 4592.
[8] (a) S. Bellemin-Laponnaz, K.S. Coleman, J.A. Osborn, Polyhedron
18 (1999) 2533;
(b) W.A. Herrmann, J.J. Haider, J. Fridgen, G.M. Lobmaier, M.
Spiegler, J. Organomet. Chem. 60 (2000) 69;
(c) F.E. Kuhn, A.M. Santos, A.D. Lopes, I.S. Gonc¸alves, J.E.
¨
´
Rodrıguez-Borges, M. Pillinger, C.C. Romao, J. Organomet. Chem.
˜
621 (2001) 207;
(d) A. Berkessel, P. Kaiser, J. Lex, Chem. Eur. J. 9 (2003) 4746;
(e) I.S. Gonc¸alves, A.M. Santos, C.C. Romao, A.D. Lopes, J.E.
˜
´
Rodrıguez-Borges, M. Pillinger, P. Ferreira, J. Rocha, F.E. Kuhn, J.
¨
Organomet. Chem. 626 (2001) 1;
Acknowledgements
(f) J. Zhao, X. Zhou, A.M. Santos, E. Herdtweck, C.C. Romao,
˜
F.E. Kuhn, Dalton Trans. (2003) 3786;
¨
(g) J.J. Haider, R.M. Kratzer, W.A. Herrmann, J. Zhao, F.E. Kuhn,
¨
J. Organomet. Chem. 689 (2004) 3735;
(h) C.E. Tucker, K.G. Davenport, Hoechst Celanese Corporation,
US Patent 5618958, 1997;
(i) M.J. Sabater, M.E. Domint, A. Corma, J. Catal. 210 (2002) 192;
(j) X. Zhou, J. Zhao, A.M. Santos, F.E. Kuhn, Z. Naturforsch. B 59
¨
(2004) 1223;
(k) R.J. Cross, P.D. Newman, R.D. Peacock, D. Stirling, J. Mol.
Catal. A 144 (1999) 273;
J.Z. thanks the Hochschul- und Wissenschaftspro-
gramm (HWP-II): Fachprogramm Chancengleichheit fur
¨
Frauen in Forschung und Lehre for a postdoctoral grant.
The Fonds der Chemischen Industrie and the Leonhard-
Lorenz-Stiftung are acknowledged for financial support.
E.H. thanks R. Ja¨ger for his help during the course of
the X-ray measurement.
(l) J. Fridgen, W.A. Herrmann, G. Eickerling, A.M. Santos, F.E.
Kuhn, J. Organomet. Chem. 689 (2004) 2752;
(m) A.A. Valente, I.S. Gonc¸alves, A.D. Lopes, J.E. Rodriguez-
Supplementary material
´
Borges, M. Pillinger, C.C. Romao, J. Rocha, X. Garcıa-Mera, New
˜
J. Chem. 25 (2001) 959.
Crystallographic data (excluding structure factors) for
the structure reported in this paper have been deposited
with the Cambridge Crystallographic Data Centre as sup-
plementary publication No. CCDC-286972 (2). Copies of
the data can be obtained free of charge on application to
CCDC, 12 Union Road, Cambridge CB2 1EZ, UK (fax:
+44 1223 336 033; e-mail: deposit@ccdc.cam.ac.uk or at
www.ccdc.cam.ac.uk/conts/retrieving.html). Supplemen-
tary data associated with this article can be found, in the
online versions, at doi:10.1016/j.jorganchem.2005.10.044.
[9] (a) F.E. Kuhn, M. Groarke, E. Bencze, E. Herdtweck, A. Prazeres,
¨
A.M. Santos, M.J. Calhorda, C.C. Romao, I.S. Gonc¸alves, A.D.
˜
Lopes, M. Pillinger, Chem. Eur. J. 8 (2002) 2370;
(b) W.R. Thiel, T. Priermeier, Angew. Chem., Int. Ed. Engl. 34
(1995) 1737;
(c) M. Abrantes, A.A. Valente, M. Pillinger, I.S. Gonc¸alves, J.
Rocha, C.C. Romao, J. Catal. 209 (2002) 237;
˜
(d) M. Abrantes, A.M. Santos, J. Mink, F.E. Kuhn, C.C. Romao,
¨
˜
Organometallics 22 (2003) 2112;
(e) J. Zhao, A.M. Santos, E. Herdtweck, F.E. Kuhn, J. Mol. Catal.
¨
A 222 (2004) 265;
(f) A.M. Martins, C.C. Romao, M. Abrantes, M.C.A. Zevedo, J.
˜
Cui, A.R. Dias, M.T. Duarte, M.A. Lomos, T. Lourenc¸o, R. Poli,
Organometallics 24 (2005) 2582.
References
[1] (a) H. Adolfsson, in: J.E. Ba¨ckvall (Ed.), Modern Oxidation
Methods, Wiley–VCH, Weinheim, 2004, pp. 21–49;
(b) Q.-H. Xia, H.-Q. Ge, C.-P. Ye, Z.-M. Liu, K.-X. Su, Chem. Rev.
105 (2005) 1603;
[10] (a) R.H. Crabtree, R.P. Dion, J. Chem. Soc., Chem. Commun. (1984)
1260;
(b) M.L.H. Green, F.W.S. Benfield, J. Chem. Soc., Dalton Trans.
(1974) 1325.
(c) F.E. Kuhn, J. Zhao, W.A. Herrmann, Tetrahedron Asym. (2005)
¨
in press.
[11] (a) P. Eilbracht, Chem. Ber. 109 (1976) 1249;
(b) P. Eilbracht, Chem. Ber. 109 (1976) 3136;

2206
J. Zhao et al. / Journal of Organometallic Chemistry 691 (2006) 2199–2206
[17] See for example: (a) T.M. Chung, Y.K. Chung, Organometallics 11
(c) P. Eilbracht, Chem. Ber. 113 (1980) 542;
(d) P. Eilbracht, Chem. Ber. 113 (1980) 1033;
(e) P. Eilbracht, Chem. Ber. 113 (1980) 1420;
(f) P. Eilbracht, Chem. Ber. 113 (1980) 2211;
(g) P. Eilbracht, W. Fassmann, W. Diehl, Chem. Ber. 118 (1985)
2314.
(1992) 2822;
(b) R. Boese, R.L. Myrabo, D.A. Newman, K.P. Vollhardt, Angew.
Chem. Int. Ed. 29 (1990) 549;
(c) K. Heck, J. Kornich, J. Organomet. Chem. 543 (1997) 153;
(d) S.H. Wang, L.H. Shiu, J. Am. Chem. Soc. 116 (1994) 5967.
[18] (a) R. Poli, Chem. Eur. J. 10 (2004) 332;
[12] J. Okuda, Comments Inorg. Chem. 16 (1994) 185.
[13] F. Amor, P. Royo, T.P. Spaniol, J. Okuda, J. Organomet. Chem. 604
(2000) 126.
(b) D. Saurenz, F. Demirhan, P. Richard, R. Poli, H. Sitzmann, Eur.
J. Inorg. Chem. 8 (2002) 1415.
[14] (a) F.E. Kuhn, A.M. Santos, P.W. Roesky, E. Herdtweck, W.
¨
[19] (a) Data Collection Software for Nonius kappa-CCD devices, Delft,
The Netherlands, 2001;
Scherer, P. Gisdakis, I.V. Yudanov, N. Ro¨sch, Chem. Eur. J. 5
(1999) 3603;
(b) Z. Otwinowski, W. Minor, Methods Enzymol. 276 (1997)
307ff;
(b) F.E. Kuhn, A.M. Santos, I.S. Gonc¸alves, A.D. Lopes, C.C.
¨
Romao, Appl. Organomet. Chem. 15 (2001) 43;
˜
(c) P. Ferreira, W.M. Xue, E. Bencze, E. Herdtweck, F.E. Ku¨hn,
Inorg. Chem. 40 (2001) 5834;
(c) A. Altomare, G. Cascarano, C. Giacovazzo, A. Guagliardi,
M.C. Burla, G. Polidori, M. Camalli, SIR-92, J. Appl. Crystallogr.
27 (1994) 435;
(d) F.E. Kuhn, W.M. Xue, A. Al-Ajlouni, S. Zhang, A.M. Santos, C.C.
Romao, G. Eickerling, E. Herdtweck, Inorg. Chem. 41 (2002) 4468.
˜
(d) A.J.C. Wilson (Ed.), International Tables for Crystallography,
vol. C, Kluwer Academic Publishers, Dordrecht, The Netherlands,
1992, Tables 6.1.1.4, 4.2.6.8, and 4.2.4.2;
(e) A.L. Spek, PLATON: A Multipurpose Crystallographic Tool,
Utrecht University, Utrecht, The Netherlands, 2001;
(f) G.M. Sheldrick, SHELXL-97, Universita¨t Go¨ttingen, Go¨ttingen,
Germany, 1998;
¨
[15] (a) S. Ciruelos, U. Englert, A. Salzer, C. Bolm, A. Maischak,
Organometallics 19 (2000) 2240;
(b) S. Ciruelos, A. Doppiu, U. Englert, A. Salzer, J. Organomet.
Chem. 663 (2002) 183;
(c) A. Doppiu, U. Englert, A. Salzer, Inorg. Chem. Acta 350 (2003)
435.
(g) L.J. Farrugia, WinGX (Version 1.70.01 January 2005), J.
Appl. Crystallogr. 32 (1999) 837.
[16] M. Minelli, J.H. Enemark, Coord. Chem. Rev. 68 (1985) 169.