Inorganica Chimica Acta 359 (2006) 1519–1523

www.elsevier.com/locate/ica

Synthesis, characterisation and reactivity of

oxodiperoxo-[2-(1-pyrazolyl)-6-menthylpyridine]molybdenum(VI):
The first chiral 2-(1-pyrazole)pyridineoxodiperoxomolybdenum(VI)
complex
Elisabete da Palma Carreiro, Guo Yong-En, Anthony J. Burke *

Departamento de Quı́mica and Centro de Quı́mica de Évora, Universidade de Évora, Rua Romão Romalho 59, 7000 Évora, Portugal

Received 25 September 2005; accepted 8 November 2005

Available online 4 January 2006

Abstract

We have obtained a novel chiral enantiomerically pure molybdenum complex of the type MoO(O2)2LL 0 which has been fully char-
acterised, including structural determination by X-ray crystallography and screened in a series of epoxidation reactions with simple ole-
fins giving satisfactory conversions, but low enantioselectivities. It’s mode of epoxidation has been investigated.
 2005 Elsevier B.V. All rights reserved.

Keywords: Oxodiperoxomolybdenum(VI) complex; Asymmetric catalysis; Epoxidation; Pyridine–pyrazole ligand

1. Introduction ligand and they demonstrated its effectiveness in the cata-

lytic epoxidation of a number of simple, yet unreactive ole-
The chiral epoxide functionality is a key structural unit fins. In 1998, this group reported the synthesis of an
present in a number of important natural products and bio- optically pure complex [8].
active compounds [1]. Amongst the various methods for Several years ago Kagan et al. [9,10] reported the prep-
the installment of such an important motif in a molecule, aration of a chiral molybdenum(VI) oxo–peroxo-N,N-
asymmetric catalysis represents an extremely elegant and dimethylacetamide complex giving enantioselectivities up
efficient method to achieve this goal [2]. to 35% ee. Later Park et al. [11] obtained enantioselectivi-
Despite the tremendous advances made using transition ties of 26–81% ee for the epoxidation of cis- and trans-b-
metal catalysts incorporating titanium [3] and manganese methylstyrene using the molybdenum(VI) oxo–peroxo
[4], a general highly selective catalytic epoxidation method complexes derived from (R)-piperidinylmandelamide and
remains elusive. One alternative catalytic system with (R)-piperidinylphenylacetamide, respectively. However,
potential to achieve this goal is the molybdenum based these were the only olefins examined. Unfortunately, these
Mimoun-type peroxo complex system [MoO(O2)2Ln, complexes were characterised by infra-red spectroscopy
L = ligand, n = 1, 2] [5] available from cheap molybdenum and elemental analysis only, and no information was given
trioxide and which has been demonstrated to be highly cat- regarding the structure of these interesting molybde-
alytic and efficient [5,6]. Thiel et al. [7a,7b,7c,7d,7e] success- num(VI) oxo–peroxo complexes. Schurig et al. [12] and
fully prepared a molybdenum based Mimoun-type peroxo Bartollini et al. [13] have reported enantioselective epoxida-
complex incorporating a bidentate pyridine–pyrazole tions using chiral Mimoun-type peroxo complex systems
but the ees reported were low. Very recently Gómez’s
*
Corresponding author. Tel.: +351 266745310; fax: +351 266744971.
group [14,15] reported the preparation of the very first chi-
E-mail address: [email protected] (A.J. Burke). ral molybdenum(VI) oxo–peroxo–oxazoline complex,

0020-1693/$ - see front matter  2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2005.11.021
1520 Elisabete da Palma Carreiro et al. / Inorganica Chimica Acta 359 (2006) 1519–1523

which demonstrated low activity in the epoxidation of cyc- 868.55 (O–O), 791.50, 767.79 (asymm. Mo–O), 584.48
looctene and (R)-limonene. The enantioselectivities were (symm. Mo–O) cm1. 1H NMR (400 MHz, DMSO-d6)
not measured. 8.56 (d, J = 2.6 Hz, 7-H, 1H), 7.87 (t, J = 7.9 Hz, 4-H,
1H), 7.83 (t, J = 0.8 Hz, 9-H, 1H), 7.45 (d, J = 7.7 Hz, 3-
2. Experimental H, 1H), 6.72 (d, J = 8.1, 5-H, 1H), 6.61 (t, J = 2.1 Hz, 2-
H, 1H), 5.06 (td, J = 8.1, 4.2 Hz, CHO, 1H), 2.17–2.14
2.1. General information (m, OCHCHH, 1H), 2.03–1.99 [m, CH(CH3)2, 1H], 1.75–
1.65 (m, OCHCHH, CH3CHCHH, 3H), 1.56–1.50 [m,
All reagents were obtained from Aldrich, Lancaster or CHCH(CH3)2, 1H], 1.25–1.16 [m, (CH3)2CHCHCHH,
Acros. TLC was carried out on aluminium backed Kisel- 1H], 1.1–0.96 (m, CH3CHCHH, 1H), 0.97–0.96 [m,
gel 60 F254 plates (Merck). Plates were visualised either (CH3)2CHCHCHH, 1H], 0.94 [d, J = 6.7 Hz,
by UV light or phosphomolybdic acid in ethanol. (CH3)C(CH3), 3H], 0.91 (d, J = 7.1 Hz, CH3C, 3H), 0.78
Gas chromatographic (GC) analyses of the products [d, J = 6.7 Hz, (CH3)C(CH3), 3H]; 13C NMR (75 MHz,
obtained from the epoxidation reactions were performed DMSO-d6), 162.20 (2-C), 148.69 (6-C), 142.03 (7-C),
on a Hewlett Packard (HP) 6890 series instrument 141.96 (4-C), 126.87 (9-C), 108.09 (3-C), 107.95 (5-C),
equipped with a flame ionisation detector (FID). The chro- 103.07 (2-C), 74.57 (CO), 47.09 (CHCO), 40.40
matograph was fitted with a cyclodex-B capillary column (CH2CHO), 33.88 (CH2CHCH3), 30.67 [CH(CH3)2],
(30 m, 250 lm, 0.25 lm) (Agilent 112–2532). In all cases, 25.86 (CCH3), 23.13 (CH2CHCO), 21.97 (CH3CCH3),
the olefin conversions were calculated by simply determin- 20.43 (CCH3), 16.48 (CH3CCH3); Anal. Calc. for C18H25-
ing the ratio of the peak areas for the olefin substrate and MoN3O6: C, 45.48; H, 5.30; N, 8.84. Found, C, 45.43; H,
the epoxide product. 5.55; N, 8.85%.
Melting points were recorded on a leica Galen III appa- For comparative purposes the following 1H NMR data
ratus and they are uncorrected. The 1H NMR and 13C were obtained for compound 1: 1H NMR (300 MHz,
NMR spectra were recorded on either a Bruker AMX300 DMSO-d6) 8.57 (d, J = 2.4 Hz, 7-H, 1H), 7.87 (t,
(1H: 300.13 MHz and 13C: 75 MHz) or a Bruker Advance J = 8.1 Hz, 4-H, 1H), 7.84 (s, 9-H, 1H), 7.45 (d,
(1H: 400.13 MHz and 13C: 100.61 MHz) instrument using J = 7.7 Hz, 3-H, 1H), 6.73 (d, J = 8.1, 5-H, 1H), 6.61 (s,
CDCl3 as solvent and TMS as internal standard (for mea- 2-H, 1H), 5.11 (td, J = 10.2, 4.2 Hz, CHO, 1H), 2.18–
surements made with the Bruker AMX300 instrument) and 2.14 (m, OCHCHH, 1H), 2.01–1.99 [m, CH(CH3)2, 1H],
the signal from residual CHCl3 as an internal standard (for 1.73 (m, OCHCHH, CH3CHCHH, 3H), 1.57–1.49 [m,
the measurements made with the Bruker Avance instru- CHCH(CH3)2, 2H], 1.27–1.19 [m, (CH3)2CHCHCHH,
ment). The 1H–1H COSY spectrum was obtained using 1H], 1.08–1 (m, CH3CHCHCHH, 1H), 0.97–0.96 [m,
the Bruker AMX300 instrument. 13C NMR spectra were (CH3)2CHCHH, 1H], 0.94 [d, J = 6.7 Hz, (CH3)C(CH3),
obtained with 135 DEPT editing to identify methylene 3H], 0.92 (d, J = 7.5 Hz, CH3C 3H), 0.78 [d, J = 6.7 Hz,
groups. Infra-red spectra were measured with a Perkin– (CH3)C(CH3), 3H].
Elmer Paragon 1000 model. Optical rotations were mea-
sured with a Perkin–Elmer 241 polarimeter using a 10 cm 2.3. X-ray crystallographic study
cell and the concentrations are quoted in g/100 ml. Elemen-
tal analyses were performed by the microanalysis service at C18H25MoN3O6 2, M = 475.35, yellow prisms, 0.55 ·
the CACTI centre at the University of Vigo, Spain. 0.13 · 0.12 mm, orthorhombic space group P212121,
a = 9.809(5) Å, b = 13.891(8) Å, c = 15.188(9) Å, V =
2.2. 2-(1-Pyrazolyl)-6-[(1R,2S,5R)-2-isopropyl-5-methyl- 2069(2) Å3, Z = 4, k(Mo Ka) = 0.71073 Å, m =
cyclohexyl-oxy]-pyridine oxodiperoxomolybdenum(VI) (2) 0.672 mm1, Dx = 1.525 Mg m3, 4216 measured reflec-
tions, 4216 independent reflections were obtained,
MoO3 (0.144 g, 1 mmol) was placed in a 25 ml flask with F(0 0 0) = 976, Scan range (h) = 2.478–25.8775, l =
30% H2O2 (1 ml, 8.82 mmol) and stirred at 45 C for 24 h, 0.672 mm1, Rint = 0.048.
giving a yellow solution. Ligand 1 (0.3 g, 1 mmol) was dis- The data set was collected on a Brucker Smart-CCD-
solved in methanol (1 ml) with the aid of some drops of 1000 diffractometer with 253 refined parameters and
CH2Cl2 and the reaction mixture was stirred for a further multi-scan BRUKER SADABS as the absorption correc-
24 h at room temperature. After filtration of the resulting tion, 4216 measured reflections and 4216 independent
solid, yellow crystals in needle form were obtained from reflections, T = 293(2) K, S = 1.043, Flack’s parameter
the mother liquor on standing in the refrigerator for several [16] = 0.01(4). Refinement on F2, R(F) = 0.031,
hours and they were subsequently filtered and dried under wR(F2) = 0.0618, (D/r)max = 0.001, Dqmax = 0.308 e Å3,
vacuum (0.2 g, 42%). Dqmin = 0.313 e Å3. Final R indices: R1 = 0.0310
22
M.p. 141.4–143 C (decomp.). ½aD ¼ 71:56 (c = 0.39, [I > 2r(I)], wR2 = 0.0554. R indices: R1 = 0.0484, wR2 =
CH2Cl2). IR (KBr) 3123.46, 2952.37, 2927.52, 2865.29, 0.0618 (all data).
1617.87, 1574.69 (C@C, C@N), 1519.57, 1464.28, Significant bond lengths and bond angles are listed in
1395.34, 1297.35, 1152.76, 1073.27, 955.30 (Mo@O), Table 1. The figure was drawn with ORTEP.
 Elisabete da Palma Carreiro et al. / Inorganica Chimica Acta 359 (2006) 1519–1523 1521

Table 1
Selected bond lengths (Å) and angles () for complex 2
4
Mo–O24 1.670(3) O24–Mo–N12 154.46(12) 5 3
Mo–O25 1.929(3) O24–Mo–N2 86.30(13)
Mo–O26 1.901(3) N2–Mo–N12 68.16(11) 7 O
2 O N N
Mo–O27 1.939(3) O26–Mo–O25 44.10(12) N 6 N O
Mo–O28 1.892(3) O28–Mo–O27 44.67(10)
8
N N Mo OO
Mo–N2 2.203(3) O25–Mo–N12 78.82(11) 9 O O
Mo–N12 2.423(3) O27–Mo–N12 77.87(11) 1 2
O25–O26 1.438(4) N2–N6–C7–N12 –0.2(5)
O28–O27 1.457(4) Scheme 1.

2.4. General procedures for the catalytic epoxidation of

olefins with complex (2)

Complex 2 (8.5 · 103 mmol) was placed in a 2-neck

round bottom flask fitted with a condenser and charged
with olefin (5 mmol dissolved in 3 ml of dry toluene),
TBHP (5 mmol in nonane) and dry toluene (1 ml). The
reaction mixture was heated to 100 C for 1–17 h. The reac-
tion mixture was subsequently analysed by chiral GC (see
Table 2).

3. Results and discussion

Prompted by work from the groups of Thiel [7,8] and Fig. 1. ORTEP view of complex 2 (the thermal ellipsoids are at the 50%
probability level).
Tolman [17], we set about preparing the chiral bidentate
pyrazole-pyridineoxodiperoxomolybdenum(VI) complex 2
(Scheme 1) using 2-(1-pyrazolyl)-6-menthylpyridine 1 X-ray crystal analysis. This analysis showed that the com-
(numbering in the scheme below is to aid with the hydrogen plex possessed a pentagonal–bipyramidal geometry with
assignments in the NMR spectrum) as the ligand, a ligand the sevenfold coordinated molybdenum.
which was previously screened by us in a catalytic enantio- This X-ray crystal structure shows the same structural
selective epoxidation using methyltrioxorhenium(VII) features for a similar complex reported by Thiel [7e]. It
(MTO) [18]. This chiral complex was obtained in 42% yield can be seen that the peroxo ligands are slightly tilted
from MoO3 using the procedure of Park et al. [11] and an upwards from the horizontal plane (towards the pyridine
X-ray crystal structure was obtained for this complex ring) and almost to the same degree, as can be seen from
(Fig. 1). the bond angles O27-Mo–N12 [Table 1, 77.87 (11)] and
The absolute configuration of the ligand was obtained, O25–Mo–N12 [Table 1, 78.82 (11)]. The Mo-peroxo O
and its bi-dentate coordination mode confirmed by this bonds are not all equal (Table 1) even in the case of the

Table 2
Catalytic epoxidation of simple olefins using Mo catalyst 2
H R TBHP (1.1 eq) H O R

Ar H Catalyst 2 (0.17 mol%), Ar H

toluene, 100oC
Entry Olefin Time (h) Conversion Epoxide:aldehydea Epoxide eeb(%)
1 Styrene 1 31 75:25 2
2 Styrene 17 86 54:46 1
c
3 trans-b-Methylstyrene 5 28 0
4 4-Methylstyrene 1 37 86:14 5
5 4-Methylstyrene 6 44 60:40 6
6 4-Methylstyrene 17 49 4:96 nd
a
Calculated by determining the ratio of the peak areas for the epoxide and the decomposition product.
b
Determined using GC. nd, not determined, due to the small quantity of epoxide formed.
c
Epoxide only.
1522 Elisabete da Palma Carreiro et al. / Inorganica Chimica Acta 359 (2006) 1519–1523

individual peroxo ligands. For example, in the case of the In the IR spectrum the Mo@O absorption appears at
distal peroxo ligand (containing O27 and O28) there is a 955 cm1 and the O–O absorption at 868.55 cm1. Com-
difference in the Mo–O bond of 0.047 Å and for the prox- parison of the IR spectrum of complex 2 with that for
imal peroxo ligand the difference is 0.028 Å. The O–O bond the free ligand 1 [18] shows very little changes in the
length in the case of both peroxo ligands is different, the absorption frequencies, however, the absorption frequency
distal peroxo ligand being slightly longer than the proximal for the C@C and C@N bonds in complex 2 shows an
ligand by 0.019 Å. In the case of the heterocyclic ligand sys- increase in frequency of 9 cm1 which would seem to indi-
tem the pyrazole ring is 0.22 Å closer to the metal centre cate p-back bonding between the metal and the nitrogen.
than the pyridine nitrogen, which is expected to be weaker We tested complex 2 in a series of catalytic asymmetric
on account of the trans-effect with the Mo@O bond (vide epoxidation reactions using simple olefins like styrene and
infra). The O–Mo–O bond angle for both peroxo ligands derivatives. The reactions were carried out as described
is virtually the same [O26–Mo–O25 = 44.10 (12) and below and the results are shown in Table 2.
O28–Mo–O27 = 44.67 (10)] and extremely close to those The enantioselectivities obtained for these epoxidation
in the molybdenum(VI) oxo-peroxo oxazoline complex reactions were low, a maximum ee of only 6% was obtained
reported by Gómez and coworkers [14,15]. The N2–N6– for the epoxidation of 4-methylstyrene (entry 5), all other
C7–N12 dihedral angle of 0.2(5) shows that the two het- ees being of the order 0–5%. It would appear from this
erocyclic rings are co-planar and almost certainly in conju- study that this complex has a slight preference for the
gation with one another. slightly bulkier tolyl group of 4-methylstyrene. The conver-
As was mentioned above, Thiel has reported an X-ray sions were generally moderate, but in the case of epoxida-
crystal structure for oxodiperoxo[3-(2-pyridyl)-1-pyrazol- tion of styrene over a 17-h period, a high conversion of
ylacetic acid ethyl ester]molybdenum(VI) [7e] and there 86% was obtained. In all cases epoxide decomposition
are some significant similarities between our complex products were observed, the most detectable and notable
and that of Thiel. For instance the Mo@O bond length being benzaldehyde and 4-methyl benzaldehyde, demon-
in our complex is 1.670 Å whereas for the same bond strating the lability of both the epoxides and their concom-
Thiel reported a value of 1.677 Å. The coordination of itant hydrolysis products under these reaction conditions.
the pyridine–pyrazole ligand in 2 takes place so that the This may be a result of the high lewis acidity of our Mo(VI)
pyridine nitrogen is located in an axial position opposite complex 2 which is supported by later experiments where
to the trans-directing oxo group, which is similar to what the complex was used in the absence of TBHP (vide infra)
Gómez and coworkers [14,15] observed for their oxazoline to epoxidise certain olefins. The result that was obtained
oxobis(peroxo)molybdenum complex. However, as can be with styrene after 1 h (Table 2, entry 1) is quite similar to
seen from our X-ray crystal structure the N12–Mo–O24 the result obtained by Thiel et al. [7d] after 1 h using their
bond angle of 154.46 represents a deviation of 25.54 pyrazole-pyridineoxodiperoxomolybdenum(VI) complex
from co-linearity which is a much lower value than that (0.19 mol%) for the epoxidation of styrene. The only differ-
observed for the molybdenum(VI) oxo–peroxo–oxazoline ence was that they reported obtaining tert-butylbenzoate as
complex of Gómez and coworkers [14,15] where the devi- secondary product whereas we obtained benzaldehyde.
ation was a considerable 74.7. The fact that the deviation Some time ago both Mimoun [19] and Sharpless [20] sug-
is smaller in the case of complex 2 would explain the gested disparate mechanisms for the epoxidation of olefins
longer pyridine nitrogen bond length [2.423(3) Å] com- using such complexes, however, oxygen transfer from the
pared to that in the oxo–peroxo–oxazoline complex of peroxy group to the olefin is common to the two mecha-
Gómez and coworkers [2.161(4) Å]. nisms. Later Thiel et al. [7b,7c,7d] on the basis of an earlier
NMR studies of complex 2 clearly indicated one com- publication by Sharpless and Chong [21] suggested a mech-
pound and no ligand dissociation nor the occurrence of anism where the TBHP attacks the Mo(VI) centre resulting
any Berry type rotation phenomena. What was interesting in a hydroperoxo, alkylperoxo intermediate which is pur-
was that contrary to the findings of Thiel et al. [7] with ported to be responsible for oxygen transfer to the olefin.
their complex, our complex exists as a single, enantiome- To test this hypothesis in our case, we decided to carry
rically pure diastereoisomer. The 1H NMR spectrum for out an epoxidation reaction using complex 2 and styrene
complex 2 was interpreted using a 1H–1H COSY experi- in the absence of any TBHP under the usual reaction con-
ment. On comparing the NMR spectrum for complex 2 ditions. We treated the complex with an excess of styrene
with that for the free ligand 1 [18] using DMSO-d6 as (26 equivs) and heated the mixture to 100 C in toluene
the solvent, it was found that there was very little differ- for 3 h. The reaction afforded benzaldehyde (22% yield,
ence in the 1H NMR spectra for both the non-complexed as determined by quantitative GC analysis and based on
or complexed ligand. The only difference observed was for the initial quantity of complex 2). Although no epoxide
the menthyl CHO hydrogen as it was shifted upfield by was actually observed, the implication is that epoxide
only 0.05 ppm in the Mo complex 2. The slight shielding might be formed (under these conditions where there is a
of this menthyl hydrogen is also reflected in its coupling high concentration of catalyst relative to epoxide, decom-
constants which are 2.1 and 1 Hz less than those for the position of this sensitive epoxide should be facile). An
free ligand 1[18]. alternative study was carried out using complex 2 and the
 Elisabete da Palma Carreiro et al. / Inorganica Chimica Acta 359 (2006) 1519–1523 1523

more reactive olefin cyclohexene (again in excess, 25 equiv- e-mail: [email protected] or http://www.ccdc.cam.
alents). Besides its increased reactivity cyclohexene was ac.uk).
chosen because its epoxide is less acid sensitive than styrene Supplementary data associated with this article can be
oxide. After 20 mins both cyclohexene oxide (ca. 19%) and found, in the online version, at doi:10.1016/j.ica.2005.
cyclohexenediol (ca. 16%) were observed by GC analysis, 11.021.
indicating that the Mimoum or the earlier Sharpless mech-
anism [20], with no involvement of a coordinated reactive References
alkylperoxide, might at least be a competing mechanism
in epoxidation reactions of olefins using TBHP and com- [1] P. Besse, H. Veschambre, Tetrahedron 30 (1994) 8885.
plex 2 as catalyst. [2] I.W.C.E. Arends, R.A. Sheldon, Top. Catal. 19 (2002) 133.
[3] R.A. Johnson, K.B. Sharpless, in: I. Ojima (Ed.), Catalytic Asym-
metric Synthesis, second ed., Wiley-VCH, New York, 2000, p. 231
4. Conclusions (Chapter 6A).
[4] T. Katsuki, in: I. Ojima (Ed.), Catalytic Asymmetric Synthesis,
In conclusion we have synthesised and fully character- second ed., Wiley-VCH, New York, 2000, p. 287 (Chapter 6B).
[5] D.V. Deubel, G. Frenking, P. Gisdakis, W.A. Herrmann, N. Rösch,
ised the first optically pure chiral bidentate pyrazole-pyr- J. Sundermeyer, Acc. Chem. Res. 37 (2004) 645, and references cited
idineoxodiperoxomolybdenum(VI) complex and carried therein.
out an initial series of epoxidation screening experiments. [6] O. Bortolini, F. DiFuria, G. Modena, R. Seraglia, J. Org. Chem. 50
We are currently trying to improve the enantioselectivities (1985) 2688.
using analogous systems and investigating possible expla- [7] (a) W.R. Thiel, M. Angstl, T. Priermeier, Chem. Ber. 127 (1994)
2373;
nations for the low enantioselectivites obtained with the (b) W.R. Thiel, Chem. Ber. 129 (1996) 575;
above system. (c) A. Hrock, G. Gemmecker, W.R. Thiel, Eur. J. Inorg. Chem.
(2000) 1107;
Acknowledgements (d) W.R. Thiel, M. Angstl, N. Hansen, J. Mol. Catal. A: Chem. 103
(1995) 5;
(e) W.R. Thiel, J. Mol. Catal. A: Chem. 117 (1997) 449.
We thank the Fundação para a Ciência e a Tecnologia [8] H. Glas, M. Spiegel, W.R. Thiel, Eur. J. Inorg. Chem. (1998) 275.
for generous financial support of this work in the form of [9] H.B. Kagan, H. Mimoun, C. Mark, V. Schurig, Angew. Chem. 19
a research grant (POCTI/QUI/35358/2000), including a (1979) 511.
postdoctoral grant (GY-E), which was partly funded by [10] H.B. Kagan, H. Mimoun, C. Mark, V. Schurig, Angew. Chem., Int.
Ed. Engl. 18 (1979) 485.
the European Community fund FEDER. We are grateful
[11] S.-W. Park, K.-J. Kim, S.S. Yoon, Bull. Korean Chem. Soc. 21
to Mrs. Ana Isabel Rodrigues of the Instituto Nacional (2000) 446.
de Engenharia, Tecnologia e Inovação (INETI) and the [12] V. Schurig, K. Hintzer, U. Leyrer, M. Pitchen, H.B. Kagan, J.
personnel of the NMR service at C.A.C.T.I (University Organomet. Chem. 370 (1989) 81.
of Vigo, Spain) for all 1H NMR and 13C NMR analyses. [13] O. Bartolini, F. Di Furia, G. Modena, A. Schionato, J. Mol. Catal.
35 (1986) 47.
The personnel of the microanalysis service at C.A.C.T.I
[14] M. Gómez, S. Jansat, G. Muller, G. Noguera, H. Teruel, V. Moliner,
are acknowledged for elemental analyses. The X-ray crys- E. Cerrada, M. Hursthouse, Eur. J. Inorg. Chem. (2001) 1071.
tallographic centre at the University of Santiago de Com- [15] J.A. Brito, M. Gómez, G. Muller, H. Teruel, J.-C. Clinet, E. Duñach,
postela, Spain is acknowledged for the X-ray crystal M.A. Maestro, Eur. J. Inorg. Chem. (2004) 4278.
analysis. [16] H.D. Flack, Acta. Crystallogr. A39 (1983) 876.
[17] D.L. Christenson, C.J. Tokar, W.B. Tolman, Organometallics 2148
(1995) 2148.
Appendix A. Supplementary material [18] E.P. Carreiro, G. Yong-en, A.J. Burke, J. Mol. Catal. A: Chem. 235
(2005) 285.
CCDC-284215 for complex 2 contains supplementary [19] H. Mimoun, I. Seree de Roch, L. Sajus, Tetrahedron 26 (1970)
crystallographic data for these structures. This data can 37.
[20] K.B. Sharpless, J.M. Townsend, D.R. Williams, J. Am. Chem. 94
be obtained free of charge at www.ccdc.cam.ac.uk/conts/ (1971) 295.
retrieving.html or from The Director, CCDC, 12, Union [21] A.O. Chong, K.B. Sharpless, J. Williams, J. Org. Chem. 42 (1977)
Road, Cambridge CB2 1EZ, UK (fax: +44 1223 336033; 1587.