An efficient method for the preparation of oxo molybdenum salalen complexes and their unusual use as hydrosilylation catalysts

Article abstract of DOI:10.1021/ic901794h

A number of molybdenum(VI) dioxo salalen complexes were prepared from the reaction of Mo(CO)₆ and salen ligands containing bulky substituants, providing a novel and facile entry to Mo-salalen compounds. Two of the complexes were characterized b

Full text of DOI:10.1021/ic901794h

11290 Inorg. Chem. 2009, 48, 11290–11296
DOI: 10.1021/ic901794h
An Efficient Method for the Preparation of Oxo Molybdenum Salalen Complexes
and Their Unusual Use as Hydrosilylation Catalysts
Jeanette E. Ziegler, Guodong Du,† Philip E. Fanwick, and Mahdi M. Abu-Omar*
Brown Laboratory, Department of Chemistry, Purdue University, West Lafayette, Indiana 47907.
†Current address: Department of Chemistry, University of North Dakota, Grand Forks, ND 58202
Received September 8, 2009
A number of molybdenum(VI) dioxo salalen complexes were prepared from the reaction of Mo(CO)6 and salen ligands
containing bulky substituents, providing a novel and facile entry to Mo-salalen compounds. Two of the complexes
were characterized by single-crystal X-ray diffraction. Reduction with organic phosphines or silanes afforded the
monooxo molybdenum(IV) complexes, along with dinuclear molybdenum(V) species featuring a bridged oxo ligand
(μ-O). One of the dinuclear complexes as well as a molybdenum(VI) dioxo salan complex was characterized
structurally. All of the molybdenum compounds except the monooxo molybdenum(IV) were fully characterized by
NMR, mass spectrometry, and elemental analyses. Investigations of acetophenone and 4-Ph-2-butanone reduction
with PhSiH3 showed that all of these molybdenum oxo complexes could serve as catalysts at reasonably low loading
(1 mol % Mo) and ∼110 °C. The time profiles and efficacy of catalysis varied depending on the precursor form of the
catalyst, MoVI(O)2 vs (O)MoV-O-MoV(O) vs MoIV(O). Solvent effects, radical scavenger probes, and other
mechanistic considerations reveal that the monooxo molybdenum(IV) is the most likely active form of the catalyst.
Introduction
and co-workers on the use of cis-Re(O)2(I)(PPh3)2 as a
catalyst for the hydrosilylation of aldehydes and ketones.5
Severaloxo and imido Re6,7 and Mo8,9 compounds have been
evaluated for the catalytic hydrosilylation of organic carbo-
nyl compounds. Several mechanistic schemes were put forth
on the basis of experimental and computational investiga-
tions.10 Among the catalysts screened so far, cationic mono-
oxo rhenium(V) oxazoline and salen complexes exhibit the
Use of the salen ligand framework in transition-metal
chemistry is well-documented, especially in the area of
asymmetric catalysis.1 Less common is the related salan
(reduced salen) complexes, which often display different
structural conformation and thus lead to different reactivity
and selectivity.2 In recent years, partially reduced salen
complexes (Figure 1), salalen for short, have emerged as a
new platform in asymmetric catalysis.3,4
Application of high oxidation state transition metals,
particularly rhenium and molybdenum, in catalytic reduc-
tions has been studied actively since the initial report by Toste
(6) (a) Du, G.; Fanwick, P. E.; Abu-Omar, M. M. Inorg. Chim. Acta 2008,
361, 3184–3192. (b) Du, G.; Fanwick, P. E.; Abu-Omar, M. M. J. Am. Chem.
Soc. 2007, 129, 5180–87. (c) Du, G.; Abu-Omar, M. M. Organometallics 2006,
25, 4920–23. (d) Ison, E. A.; Cessarich, J. E.; Du, G.; Fanwick, P. E.; Abu-Omar,
M. M. Inorg. Chem. 2006, 45, 2385–87. (e) Ison, E. A.; Trivedi, E. R.; Corbin,
R. A.; Abu-Omar, M. M. J. Am. Chem. Soc. 2005, 127, 15374–5.
(7) (a) Royo, B.; Romao, C. C. J. Mol. Catal. A: Chem. 2005, 236, 107–
112. (b) Reis, P. M.; Royo, B. Catal. Commun. 2007, 8, 1057–1059. (c) Reis,
P. M.; Costa, P. J.; Romao, C. C.; Fernandes, J. A.; Calhorda, M. J.; Royo, B.
Dalton Trans. 2008, 1727–1733. (d) Nolin, K. A.; Ahn, R. W.; Toste, F. D. J. Am.
Chem. Soc. 2005, 127, 12462–12463.
(8) (a) Peterson, E.; Khalimon, A. Y.; Simionescu, R.; Kuzmina, L. G.;
Howard, J. A. K.; Nikonov, G. I. J. Am. Chem. Soc. 2009, 131, 908–909.
(b) Pontes da Costa, A.; Reis, P. M.; Gamelas, C.; Romao, C. C.; Royo, B. Inorg.
Chim. Acta 2008, 361, 1915–1921. (c) Fernandes, A. C.; Romao, C. C.
Tetrahedron Lett. 2005, 46, 8881–8883. (d) Fernandes, A. C.; Romao, C. C.
J.Mol. Cat. A: Chem. 2006, 253, 96–98. (e) Fernandes, A. C.; Romao, C. C.
J. Mol. Cat. A: Chem. 2007, 272, 60–63. (f) Fernandes, A. C.; Fernandes, R.;
Romao, C. C.; Royo, B. Chem. Commun. 2005, 213–4. (g) Reis, P. M.; Romao,
C. C.; Royo, B. Dalton Trans. 2006, 1842–1846.
*To whom correspondence should be addressed. E-mail: mabuomar@
purdue.edu.
(1) (a) Katsuki, T. Chem. Soc. Rev. 2004, 33, 437–444. (b) Katsuki, T.
Coord. Chem. Rev. 1995, 140, 189–214.
(2) For use of salan in polymerization catalysts, see: (a) Sergeeva, E.;
Kopilov, J.; Goldberg, I.; Kol, M. Chem. Commun. 2009, 3053–55. (b) Yeori,
A.; Goldberg, I.; Shuster, M.; Kol, M. J. Am. Chem. Soc. 2006, 128, 13062–63.
(3) Matsumoto, K.; Saito, B.; Katsuki, T. Chem. Commun. 2007, 3619–
3627.
(4) (a) Fujita, H.; Uchida, T.; Irie, R.; Katsuki, T. Chem. Lett. 2007, 36,
1092–1093. (b) Yeori, A.; Gendler, S.; Groysman, S.; Goldberg, I.; Kol, M. Inorg.
Chem. Commun. 2004, 7, 280–2. (c) Saito, B.; Katsuki, T. Angew. Chem., Int.
Ed. 2005, 44, 4600–2. (d) Yamaguchi, T.; Matsumoto, K.; Saito, B.; Katsuki, T.
Angew. Chem., Int. Ed. 2007, 46, 4729–31. (e) Saito, B.; Egami, H.; Katsuki, T.
J. Am. Chem. Soc. 2007, 129, 1978–86.
(5) (a) Kennedy-Smith, J. J.; Nolin, K. A.; Gunterman, H. P.; Toste, F. D.
J. Am. Chem. Soc. 2003, 125, 4056–4057. (b) Nolin, K. A.; Krumper, J. R.;
Pluth, M. D.; Bergman, R. G.; Toste, F. D. J. Am. Chem. Soc. 2007, 129, 14684–
96.
(9) (a) Fernandes, A. C.; Romao, C. C. Tetrahedron Lett. 2007, 48, 9176–
9. (b) Fernandes, A. C.; Fernandes, J. A.; Paz, F.A. A.; Romao, C. C. Dalton
Trans. 2008, 6686–6688.
(10) Du, G.; Abu-Omar, M. M. Curr. Org. Chem. 2008, 12, 1185–1198.
r
pubs.acs.org/IC
Published on Web 10/21/2009
2009 American Chemical Society

Article
Inorganic Chemistry, Vol. 48, No. 23, 2009 11291
Figure 1. Salen, salan, and salalen ligands (R=alkyl, halide, etc.; R0=
Me or H).
highest activities. However, the chiral version gives low
enantioselectivity,6a except for a mono-oxo oxazoline com-
plex that has been shown to catalyze the hydrosilylation of
imines with high enantioselectivity.7d In contrast, the Mo
catalysts examined thus far have been primarily simple dioxo
Mo(VI) such as Mo(O)2Cl2.
Our initial thought was to examine oxomolybdenum(IV)
salen analogues on the basis of diagonal relationships of the
periodic table and their isoelectronic configuration to
oxorhenium(V). We were also interested in understanding
the effect of charge on catalytic activity. Oxorhenium(V)
oxazoline and salen are cationic while their molybdenum
analogues would be neutral. In this report, we describe a
novel and simple entry to molybdenum salalen complexes,
attempts to obtain isolable Mo(IV) monooxo complexes, and
their application in catalytic hydrosilylation.
Figure 2. ORTEP diagram (drawn at 50% probability ellipsoids)
of Mo(O)2(salalen-tBu2) compound 1. The hydrogen atoms except the
˚
one on nitrogen are omitted for clarity. Selected bond distances (A) and
bond angles (deg): Mo1-O13, 1.697(3); Mo1-O12, 1.734(3); Mo1-O11,
1.917(3); Mo1-O118, 2.012(3); Mo1-N111, 2.216(4); Mo1-N18,
2.281(4); N18-C17, 1.285(5); N111-C110, 1.492(6); N111-H111, 0.75(4);
O13-Mo1-N18, 171.67(13); O12-Mo1-O118, 159.08(13); O11-
Mo1-N111, 155.52(14); O13-Mo1-O12, 101.58(15); O13-Mo1-O11,
105.47(13).
stereochemistry. The presence of a medium strength
broadband at 3120 cm-1 is indicative of NH group. On
the basisof theseobservations, the product wasformulated
as a salalen cis-dioxo complex, MoVI(O)2(salalen-3,5-tBu2)
(1). X-ray diffraction analysis on single crystals confirmed
this assignment (see below).
We extended this synthetic approach further to a
number of substituted salen ligands incorporating a chiral
backbone (eq 1). Two complexes were isolated and fully
Results
1
characterized. The H spectra show a pattern similar to
Synthesis and Characterization of Dioxomolybdenum-
(VI) Salalen Complexes. In an effort to prepare MoIV(O)-
(salen) complexes and test their applicability in catalytic
hydrosilylation of organic carbonyl compounds, we first
followed a literature procedure11 that calls for refluxing
Mo(CO)6 with H2salen (the parent ligand;no substitu-
ents on the phenyl rings and an ethyl bridged diimine)
under air. As reported, a dark brown powder was
obtained, which is insoluble in common organic solvents.
We reasoned that t-butyl substituents on the phenyl ring
would facilitate dissolution of the Mo-containing pro-
duct. Indeed, when H2(3,5-tBu2-salen) was employed
instead of the parent salen in the reaction, the isolated
orange product was soluble in common organic solvents,
including THF, CH2Cl2, acetone, methanol, benzene,
and hexanes. Unlike the presumed MoIV(O)(salen),11 this
that observed for 1 above, which is consistent with the
reduction of one of the CdN linkages on the ligand to an
amine HC;NH. It is noteworthy that two diastereomers
with equal intensity were observed for 2 and 3 due to the
chirality at the amine nitrogen.
1
compound is diamagnetic and gives a clean H NMR.
However, the spectrum cannot be accounted for by
either a symmetrically or asymmetrically coordinated
salen-tBu2 ligand. In the aromatic region, one singlet
(8.09 ppm), assignable to one imine proton (HC=N),
was accompanied by four doublets (at 7.61, 7.21, 7.18,
and 6.19 ppm, with J=2.4 Hz), assignable to four ArH
protons, suggesting one of the two imine CdN double
bonds was hydrogenated during the course of the reaction
(see eq 1). Accordingly, two doublets (J=12.9 Hz) were
observed at 4.62 and 4.01 ppm. Furthermore, the FT-IR
showed multiple strong bands in the ModO stretch-
ing region, 910-840 cm-1, characteristic of cis dioxo
1 and 2 were characterized structurally by single crystal
X-ray diffraction analysis. ORTEP drawings along with
selected bond lengths and angles are shown in Figures 2
and 3 (where saladach refers, by analogy, to the half-
reduced parent salalen ligand with a cyclohexane diamine
(dach) bridge). The structures reveal that the metal center
adopts a distorted octahedral geometry. The two oxo
ligands are located mutually cis with bond angels of
101.58 (1) and 101.59° (2). Trans to the oxo ligands are
an imine nitrogen and a phenoxy oxygen from the salan
portion. Consistent with the stronger trans effect of oxo
(11) Sabry, D. Y.; Youssef, T. A.; El-Medani, S. M.; Ramadan, R. M.
J. Coord. Chem. 2003, 56, 1375–1381.

11292 Inorganic Chemistry, Vol. 48, No. 23, 2009
Ziegler et al.
Figure 3. ORTEP diagram (drawn at 50% probability ellipsoids) of
Mo(O)2(saladach-tBu) compound 2. The hydrogen atoms except the one
˚
on nitrogen are omitted for clarity. Selected bond distances (A) and bond
Figure 4. ORTEP diagram (drawn at 50% probability ellipsoids) of
Mo(O)2(salan-3,5-tBu2) compound 4. The hydrogen atoms except the one
angels (deg): Mo1-O12, 1.702(3); Mo1-O11, 1.712(2); Mo1-O118,
1.924(3); Mo1-O111, 2.016(2); Mo1-N18, 2.200(3); Mo1-N111,
2.297(3); N111-C112, 1.277(5); N18-C17, 1.501(4); N18-H18, 0.99(4);
N28-H28, 0.71(4); O12-Mo1-O111, 160.20(12); O118-Mo1-N18,
155.79(12); O11-Mo1-N111, 169.51(12); O12-Mo1-O11, 101.59(12);
O12-Mo1-O118, 100.87(13).
˚
on nitrogen are omitted for clarity. Selected bond distances (A) and bond
angles (deg): Mo1-O2, 1.688(4); Mo1-O1, 1.709(4); Mo1-O12,
1.924(4); Mo1-O62, 1.941(4); Mo1-N2, 2.347(5); Mo1-N5, 2.351(5);
O2-Mo1-O1, 107.6(2); O2-Mo1-O12, 101.9(2); O1-Mo1-O12,
93.82(19); O2-Mo1-O62, 94.3(2); O1-Mo1-O62, 96.43(19); O12-
Mo1-O62, 157.24(17); O2-Mo1-N2, 86.55(19); O1-Mo1-N2,
165.3(2); O12-Mo1-N2, 78.82(17); O62-Mo1-N2, 86.38(18);
O2-Mo1-N5, 156.93(19); O1-Mo1-N5, 94.51(19); O12-Mo1-N5,
82.70(18); O62-Mo1-N5, 76.29(17); N2-Mo1-N5, 72.03(16).
ꢀ
˚
groups, the Mo-O(phenoxy) bond (2.012 in 1, 2.016 A in 2)
trans to the oxo is longer than the Mo-O(phenoxy) bond
(1.917 in 1, 1.924 A in 2) trans to the amine nitrogen.
˚
1
results to PPh3 (by UV-vis and H NMR). However,
The salalen ligands bind to the molybdenum center in a
fac-mer fashion, with the salan portion fac and the salen
portion mer, which are the typical binding modes for the
parent salan and salen ligands, respectively.
careful quantitative integration of the NMR signals
showed that only about half of the starting material
converted to MoIV(O)(salalen-tBu2) (eq 2). Upon recrys-
tallization, a yellow crystalline material was obtained,
which was found to be NMR silent. X-ray crystal diffrac-
tion analysis revealed a μ-oxo-bridged MoV dinuclear
complex (Figure 5), presumably resulting from a facile
reaction between MoVI and MoIV. The individual ar-
rangement around the metal center is similar to the parent
mononuclear Mo(O)2(salalen-tBu2) (1). Two terminal
oxygens, trans to the imine nitrogens, adopt an anti
configuration.
Synthesis and Structure of cis-Mo(O)2(salan-tBu2) (4).
In the context of asymmetric catalysis, we prepared cis-
Mo(O)2(salan-tBu2) with a chiral backbone from Mo(O)2-
(acac)2 and H2salan. The simplicity of the product’s
1H-NMR spectrum suggests cis-R conformation. This
stereochemistry is confirmed by X-ray structural analysis,
which revealed a symmetric arrangement of the salan
ligand with two amine nitrogens trans to the two multiply
bonded oxo ligands (Figure 4).
Reduction of Dioxo Molybdenum with Phosphines and
Silanes. Because our initial targets are Mo(IV) complexes,
we attempted to prepare oxo MoIV salalen by reduction of
dioxo MoVI complexes with organic phosphines and
silanes. The latter is relevant to hydrosilylation. Most
reactions were carried out with Mo(O)2(salalen-tBu2) (1).
Despite several attempts of varying concentrations and
identity of the phosphine, 1 proved to be resistant to
reduction with phosphine under ambient conditions.
However, at elevated temperature (ca. 100 °C) a color
change from orange to dark red was observed and
accompanied by formation of OPPh3. The Mo product
was diamagnetic and displayed shifted 1H NMR signals
in comparison to the starting dioxo complex. We con-
clude that the product is a five-coordinate monooxo
molybdenum(IV) because the byproduct OPPh3 and
excess PPh3 are not coordinated based on their 31P
chemical shifts. The resulting MoIV complex is air sensi-
tive. It turns yellow upon exposure to air. Use of polymer-
In the context of hydrosilylation, we also employed
organic silanes as reductants. No reactions were observed
with Et3SiH and Ph2SiH2 even at elevated temperature.
When two equivalents of PhSiH3 was employed, reduc-
tion was observed at ∼ 110 °C to afford MoIV(O)-
(salalen-tBu2) (5) as one of the major products based on
1H NMR. Prolonged heating, however, led to the disap-
pearance of 5. It is worth noting that during the course of
reaction, there is another intermediate species observed by
1H NMR, alongside the starting MoVI(O)2(salalen-tBu2)
1
bound PPh3 afforded the same product, based on H
NMR, allowing for easy removal of the OPPh3 byproduct
via filtration.
Trimethyl phosphine (PMe3), a stronger reducing
agent, still required elevated temperature and gave similar

Article
Inorganic Chemistry, Vol. 48, No. 23, 2009 11293
Table 1. Hydrosilylation of Acetophenone Catalyzed by Mo(O)2(salalen-tBu2)
(1)a
entry
silane
solvent
time/ h
% conversion
1
2
Et3SiH
Cl3SiH
Ph2SiH2
PhSiH3
PhSiH3
PhSiH3
PhSiH3
PhSiH3
PhSiH3
PhSiH3
benzene
benzene
benzene
benzene
benzene
toluene
1,2-dichlorobenzene
chloroform
dichloromethane
acetonitrile
24
24
24
24
7
N.R.
N.R.
20
N.R.
>98
>98
>98
42
3
4b
5
6
7
8
9
10
4
23
28
24
28
8
38
a Reaction conditions: 1.0 mmol acetophenone, 1.2 mmol silane, and
5 mol % 1 in a sealed NMR tube heated to ca. 110 °C. b Room
temperature.
Table 2. Molybdenum-Catalyzed Hydrosilylation of Acetophenone (A) and
a
4-Phenylbutanone (B) with PhSiH3
Figure 5. ORTEP diagram (drawn at 50% probability ellipsoids) of
μ-oxo-(Mo(O)(salalen-tBu2))2 compound 6. The hydrogen atoms, except
˚
the ones on nitrogen, are omitted for clarity. Selected bond distances (A)
entry
substrate
catalyst
time/ h
% conv.
% yieldb
and bond angels (deg): Mo1-O2, 1.686(2); Mo1-O1, 1.8884(3);
Mo1-O4, 1.992(2); Mo1-O3, 2.013(2); Mo1-N1, 2.204(3); Mo1-N2,
2.302(3); N1-H1, 0.83(4); O2-Mo1-O1, 98.74(9); O1-Mo1-O4,
93.38(7); O1-Mo1-O3, 162.10(7); O4-Mo1-N1, 153.42(11); O2-
Mo1-N2, 169.10(11); Mo1-O1-Mo1, 180.000(12).
1
2
3
4
5
A
B
A
B
A
A
A
A
1
1c
3c
3c
2
7
19
16
19
7
7
3
8
>98
>98
>98
>98
97
>98
93
>98
80
76
80
83
__
58
__
__
6
4c
5
and the reduction product 5. Due to the overlap, only
signals in the 5.5-2.5 ppm region can be clearly identi-
fied. For example, peaks at 5.07 and 3.30 ppm can be
assigned to the benzylic CH2 adjacent to the amine
nitrogen. This intermediate is tentatively formulated as
MoIV(OH)(OSiPhH2)(salalen-tBu2), resulting from a 3 þ
2 addition of the Si-H bond across the Mo(O)2 core.12
For cis-dioxo molybdenum or rhenium compounds such
as Mo(O)2Cl2 and ReI(O)2(PPh3)2, 2 þ 2 cycloaddition of
Si-H has been reported.5,7 In our case, a 2 þ 2 pathway
would result in a seven-coordinate molybdenum, and
thus appears less likely. Supposedly, further elimination
of PhH2SiOH from MoIV(OH)(OSiPhH2)(salalen-tBu2)
would lead to the MoIVO(salalen-tBu2) product. How-
ever, we were unable to confirm the identity of this
intermediate species by ESI-MS.
Catalytic Hydrosilylation of Acetophenone and 4-Phe-
nylbutanone. Mo(O)2(salalen-tBu2), 1, was investigated
initially as a typical representative of the obtained, solu-
ble, and pure molybdenum dioxo complexes. Consistent
with the observed reactivity with organic silanes, no
catalysis was detected with acetophenone (a standard
ketone) at room temperature. As for tertiary silanes such
as Et3SiH, no hydrosilylation product was detected even
when the reaction mixture was heated in a sealed NMR
tube to 110 °C. With a more active dihydrosilane,
Ph2SiH2, a small amount of hydrosilylation product
could be detected. When a primary silane, PhSiH3, was
employed at 110 °C in a sealed NMR tube, quantitative
conversion of acetophenone was observed (Table 1). A
control reaction under identical conditions (aceto-
phenone and PhSiH3 at 110 °C in a sealed NMR tube)
in the absence of catalyst resulted in no conversion and
the starting compounds were stable under these condi-
tions. A number of solvents were examined to discern
7
8d
6
a Reaction conditions: 1.0 mmol ketone, 1.2 mmol PhSiH3, 5 mol %
catalyst in benzene, sealed NMR tube at ∼110 °C. b Isolated alcohol
yield after acid aqueous workup. c 1 mol % catalyst. d In 1,2-dichlor-
obenzene.
medium effects on the efficiency of the reaction. Benzene,
toluene, and 1,2-dichlorobenzene performed best. There
seems to be no discernible dependencies on solvent polarity
or coordinating ability. Hence, benzene or toluene was
the solvent of choice from this point forward.
In addition to acetophenone, we investigated 4-phenyl-
butanone as a representative aliphatic ketone. It was
found to be comparable in reactivity to the aryl ketone
acetophenone (Table 2). In the case of 4-phenylbutanone,
a mixture of hydrosilylation products were observed
1
(eq 3). These products were characterized by H NMR
and GC-MS. Interestingly, in 1,2-dichlorobenzene only
one product was observed, which is the expected initial
hydrosilylation product (H-C-OSiH2Ph; see eq 3). Despite
the complication of several silylated products, acid workup
afforded the alcohol in high to modest isolated yields
(Table 2). Different catalysts were studied to probe the
dependence on the oxidation state and ligand identity of
the catalyst precursor. The dioxomolybdenum(VI) catalyst
containing a chiral ligand (2) behaved comparably to 1.
So did the fully reduced salan complex 4.
Reaction progress for all of the catalysts was followed
by 1H NMR. All dioxomolybdenum(VI) complexes
(1-4) as well as the dinuclear μ-oxo molybdenum(V) (6)
showed an induction period. In comparison, the monooxo
(12) For [3 þ 2] addition of tertiary silane to OsO4 :Valliant-Saunders, K.;
Gunn, E.; Shelton, G. R.; Hrovat, D. A.; Borden, W. T.; Mayer, J. M. Inorg.
Chem. 2007, 46, 5212–19.

11294 Inorganic Chemistry, Vol. 48, No. 23, 2009
Ziegler et al.
an intramolecular Meerwein-Ponndorf-Verley (MPV)
reaction.15 Remarkably the reaction stops at half way
without going all the way to the salan complex. Although
the preparation is usually simple and straightforward,
these two routes are often metal and ligand specific.
Our study herein represents the first examples for
preparation of Mo(VI) salalen complexes from the salen
ligands. Despite the resemblance with the above-men-
tioned intramolecular MPV reduction, the reaction is
more complex since it also involves the oxidation of
Mo(0) to Mo(VI). Notably the reaction is quite general,
as a number of Mo complexes are obtained in a similar
fashion, as long as t-butyl groups are present in the 3,30-
positions. In retrospect, the partially reduced dioxo
Mo(VI) species was probably present in the original
preparation intended for MoIV(O)(salen);11 ESI-MS of
the sample using the parent salen ligand without substit-
uents in our hands showed a pattern consistent with
Mo(O)2(salalen). However, multinuclear species can also
be seen. We attribute these complications to the lack of
steric bulk in unsubstituted salen. Furthermore, the poor
solubility of the product with the parent salen ligand and
apparent paramagnetism prevented further characteriza-
tion. By comparison to Mo, reaction of W(CO)6 with
salen ligands only led to recovery of the free ligand under
the same conditions.
Figure 6. Reaction progress for the hydrosilylation of acetophenone
with PhSiH3 at 110 °C catalyzed by dioxomolybdenum(VI) 2 (open
circles), dinuclear molybdenum(V) 6 (open squares), and oxomolybde-
num(IV) 5 (hatched squares). % Conversion was determined by 1H
NMR. The lines are a smoothing function and not a kinetic fit.
molybdenum(IV) (5) did not exhibit an induction period
(Figure 6). After the induction period, precursors 1-4
and 6 appear to follow comparable conversion rates to the
oxomolybdenum(IV) catalyst 5. Even though 5 did not
show an induction period, it should be noted that 5 did
not catalyze hydrosilylation under ambient temperature.
Temperatures of around 110 °C were still necessary in the
case of oxomolybdenum(IV) catalyst. The use of a chiral
ligand in catalysts 2, 3, and 4 deserves a comment.
Although these catalysts gave high conversions and
excellent yields of the alcohol after aqueous workup,
none of them showed significant asymmetric induction
(<8 ee%) with either substrate, acetophenone or 4-phe-
nylbutanone. This could be attributed to the high reaction
temperature, which was necessary.
Attempts to prepare MoVI(O)2(3,5-tBu2-salen) from
Mo(O)2(acac)2 and H2(3,5-tBu2-salen) were unsuccessful
despite the successful synthesis of MoVI(O)2(3,5-(OMe)2-
salen) by this methodology;16 multiple products formed
and were not further identified. This is not surprising
because the success of Mo- and Re-salen preparations is
quite sensitive to substituents on the phenyl rings as well
as the diamine bridge, and hard to predict.17 Conse-
quently, a systematic variation of substituents in these
systems is often difficult to achieve. In contrast, cis-dioxo
Mo-salan compounds with sterically bulky substituents
have been prepared easily, though higher nucleation
could pose some problems for ligands with less bulky
substituents.18 This suggests that the steric bulk is one of
the driving forces for the partial reduction of the salen
ligands, so that the unfavorable steric interaction could be
relieved when one of the CdN double bonds is replaced
by a saturated C-N single bond.
Mechanistic Considerations of Mo-Catalyzed Hydrosi-
lylation. The dioxo functionality was suggested to be a
critical feature for cis-Re(O)2I(PPh3)2.19 This idea has
been supported by the reactivity of Mo(O)2Cl2 and by
computational studies of these systems.20 We and others,
however, have shown that the dioxo functionality is not a
prerequisite for catalysis, but the presence of an open
coordination site or a labile ligand is. In contrast, the
Discussion
Synthesis of Oxo Molybdenum Salalen Complexes.
There are several approaches available in the literature
for the generation of salalen complexes. The most com-
mon approach is coordination of metal precursors with
salalen ligands.13 This method offers opportunities for
easy tuning of steric and electronic properties and has
been utilized in the synthesis of a number of catalysts. The
drawback is that the ligand synthesis has to involve the
stepwise construction of imine and amine linkages. The
salalen complexes can also be generated by two other less
generally applicable routes. One is the oxidative dehy-
drogenation of the C-N bonds in metal-salan com-
plexes;14 the other is from in situ reduction of salen
upon coordination with titanium(IV), presumably via
(13) (a) Berkessel, A.; Brandenburg, M.; Leitterstorf, E.; Frey, J.; Lex, J.;
Schaefer, M. Adv. Synth. Catal. 2007, 349, 2385–2391. (b) Berkessel, A.;
Frauenkron, M.; Schwenkreis, T.; Steinmetz, A. J. Mol. Catal. A: Chem. 1997,
117, 339–346. (c) Berkessel, A.; Bats, J. W.; Schwarz, C. Angew. Chem., Int. Ed.
1990, 29, 106–108. (d) Schwenkreis, T.; Berkessel, A. Tetrahedron Lett. 1993,
34, 4785.
(14) (a) Bottcher, A.; Elias, H.; Jager, E.-G.; Langfelderova, H.; Mazur,
M.; Muller, L.; Paulus, H.; Pelikan, P.; Rudolph, M.; Valko, M. Inorg.
Chem. 1993, 32, 4131–38. (b) Bottcher, A.; Elias, H.; Muller, L.; Paulus, H.
Angew. Chem., Int. Ed. 1992, 31, 623–5.
(16) Yamanouchi, K.; Yamada, S. Inorg. Chim. Acta 1974, 9, 161–164.
(17) Re: (a) Gerber, T. I. A.; Luzipo, D.; Mayer, P. J. Coord. Chem. 2005,
58, 1505–1512. (b) Herrmann, W. A.; Rauch, M. U.; Artus, G. R. J. Inorg. Chem.
1996, 35, 1988–1991. (c) El-Medani, S. M. J. Coord. Chem. 2004, 57, 497–507.
(18) (a) Whiteoak, C. J.; Britovsek, G. J.; Gibson, V. C.; White, A. J.
Dalton Trans. 2009, 2337–44. (b) Yang, H.; Wang, H.; Zhu, C. J. Org. Chem.
2007, 72, 10029–34.
(19) Thiel, W. R. Angew. Chem., Int. Ed. 2003, 42, 5390–5392.
(20) (a) Chung, L. W.; Lee, H. G.; Lin, Z.; Wu, Y.-D. J. Org. Chem. 2006,
71, 6000–6009. (b) Costa, P. J.; Romao, C. C.; Fernandes, A. C.; Royo, B.; Reis,
P. M.; Calhorda, M. J. Chem.;Eur. J. 2007, 13, 3934–41. (c) Drees, M.;
Strassner, T. Inorg. Chem. 2007, 46, 10850–59.
(15) Matsumoto, K.; Sawada, Y.; Saito, B.; Sakai, K.; Katsuki, T. Angew.
Chem., Int. Ed. 2005, 44, 4935–39.

Article
Inorganic Chemistry, Vol. 48, No. 23, 2009 11295
Experimental Section
dioxo MoVI and dinuclear oxo MoV compounds investi-
gated in this study are coordinatively saturated and do
not contain a labile ligand. This premise is in agreement
with the observation of an induction period for both
dioxo MoVI and μ-oxo dinuclear MoV catalysts. We
hypothesize on the basis of experimental observations
that during this induction period dioxo MoVI and μ-oxo
MoV are reduced by the organic silane. This proposal
is also consistent with a ReVII system in which the
ReVII precursor was first reduced to ReV by dihydrogen
prior to catalysis.7c We observe an intermediate species
during the reduction of Mo(O)2(salalen-tBu2) (1), tenta-
tively formulated as MoIV(OH)(OSiPhH2)(salalen-tBu2).
However, this intermediate is not likely to be the active
catalyst because it is detected mostly during the induction
period when comparing the reaction profile with and
without a ketone substrate. In the presence of excess
PhSiH3, monooxo MoIV can be reduced further to an
unidentified paramagnetic species. However, when this
over-reduced species was applied as a catalyst by adding
ketone to the reaction mixture, no hydrosilylation pro-
duct was observed at 110 °C. These observations collec-
tively point to MoIV(O)-salalen as the active species, and
indeed, MoIV(O)(salalen-tBu2) compound 5 exhibits the
highest activity (Table 2, entry 7) and lacks an induction
period (Figure 6).
We also considered a radical mechanism because it has
been shown in several systems that organic silanes can
undergo radical reactions with transition metals.8g,20c In
the presence of the radical scavenger CBr4, the conver-
sion of acetophenone occurred with a much shorter
induction period and reactions were complete within 1
h, regardless of the catalyst (1, 5, or 6). However,
GC-MS analysis of the reaction products showed a
small amount of the hydrosilylation product, and in-
stead the major species was PhCHBrCH3, which is the
result of deoxygenative halogenation.21 This finding
may suggest that there is at least some radical charac-
ter/involvement in the hydrosilylation reaction. In the
presence of CBr4, the hydrosilylation pathway is totally
inhibited and another reaction prevails, yielding a to-
tally different product.
General. Solvents were purified with a solvent purification
system (Anhydrous Engineering Inc.) prior to use. Organosi-
lanes and ketones were obtained from commercial sources and
used as received without further purification. The saldach ligand
and its derivatives were prepared by refluxing an ethanol solution
of (R, R)-1,2-diaminocyclohexane (dach) with two equivalents
of appropriate salicyaldehydes. Mo(CO)6 and Mo(O)2(acac)2
were obtained from commercial sources.
NMR spectra were recorded on Varian Inova 300 instruments
at 20 °C. Proton resonances were referenced internally to residual
solvent peaks (CHCl3, 7.26; CHDCl2, 5.32; CD2HCN, 1.94).
Infrared spectra were recorded as KBr pellets on a Perkin-Elmer
2000 FT-IR spectrometer. UV-vis data were recorded on a
Shimadzu 2501 spectrophotometer and reported as λmax in nm
(log ε). Optical rotation was measured on a Rudolph Research
Autopol III polarimeter. Mass spectrometry was performed by the
Purdue University Campus-Wide Mass Spectrometry Center on a
Hewlett-Packard Engine mass spectrometer (GC/MS) or a Finni-
ganMAT LCQ mass spectrometer system (ESI). Enantiomeric
excesses of the resulting silylethers were determined using an
Agilent 6890 gas chromatograph equipped with a chiral CyclodexB
column (30 m ꢀ 0.25 mm ID, 0.25 μm thick) on the corresponding
alcohols after desilylation. Elemental microanalyses were done by
the Purdue University Microanalytical Laboratory.
Mo(O)2(Salalen-tBu2) (1). A mixture of N,N’-ethylene-bis-
(3,5-ditbutyl-salicylideneimine), H2(3,5-tBu2-salen), (241 mg,
0.489 mmol), Mo(CO)6 (125 mg, 0.473 mmol) in THF (35 mL)
was refluxed under air for 22 h. After cooling to ambient
temperature, the resulting reddish solution was filtered and
the filtrate was dried in vacuo. The residue was taken up by
CH2Cl2 and layered with hexanes and stored at -20 °C. After
several days the product was collected by filtration. Yield: 137 mg,
47%. X-ray quality crystals were obtained by slow evaporation
of a concentrated CH2Cl2 solution of the product. 1HNMR
(CD2Cl2, 300 MHz): δ 8.09 (s, 1H, CH=N), 7.61 (d, J=2.4 Hz,
1H), 7.21 (d, J=2.4 Hz, 1H), 7.18 (d, J=2.4 Hz, 1H), 6.91 (d, J=
2.4 Hz, 1H), 5.61 (br s, 1H, NH, varies depending on the
condition), 4.62 (d, J = 12.9 Hz, 1H, benzylic), 4.01 (d, J =
12.9 Hz, 1H, benzylic), 3.72 (m, 1H, ethylene), 3.44 (m, 1H,
ethylene), 3.21 (m, 2H, ethylene), 1.49 (s, 9H, tbutyl), 1.32 (s, 9H,
t
t
tbutyl), 1.26 (s, 9H, butyl), 1.11 (s, 9H, butyl). FT-IR (KBr):
3120 (NH), 1645 (CdN), 907, 875, 842 (ModO) cm-1. MS
(ESIþ): m/z 623 (98MoO2L þ Hþ), with correct isotopic pattern
for Mo. Anal. Calcd (found) for C32H48N2O4Mo: C, 61.92
(61.68); H, 7.79 (7.71); N, 4.51 (4.59).
Mo(O)2(3-tBu-Saladach) (2). A mixture of (R,R)-1,2-cyclo-
hexanediamino-N,N0-bis(3-t-butyl-salicylidene), H2(3-tBu-saldach),
(333.5 mg, 0.767 mmol), Mo(CO)6 (186.7 mg, 0.707 mmol) in
THF (50 mL) was refluxed under air for 19 h in which time
the color changed from yellow to green-brown and then
red-brown. After cooling to ambient temperature, the mixture
was filtered and the filtrate was dried in vacuo. The residue was
treated with a small amount of hexanes and filtered again. The
dark yellow solid obtained was dissolved in benzene and layered
with hexanes. The resulting yellow crystals, suitable for X-ray
diffraction crystallography, and powders were collected by
filtration. Yield: 134 mg, 34%.
In conclusion, a facile synthetic pathway to oxo molyb-
denum salalen complexes has been described. These
compounds serve as catalysts for the reduction of ke-
tones to alcohols with PhSiH3 at 110 °C. Dioxo mo-
lybdenum(VI) and dinuclear μ-oxo molybdenum(V)
complexes display a long induction period. In contrast,
monooxo molybdenum(IV) is the most reactive catalyst
and lacks an induction period. At elevated temperatures
(ca. 110 °C) dioxo molybdenum(VI) salalen complexes
are reduced by organic phosphines (PPh3 or PMe3) and
silanes (Ph2SiH2 or PhSiH3) to mononuclear molyb-
denum(IV) and dinuclear μ-oxo molybdenum(V). Mo
catalysts containing chiral salalen ligands do not give a
significant asymmetric induction (less than 8% ee).
The results in hand are most consistent with the proposal
that monooxo molybdenum(IV) is the active form of
the catalyst; and involvement of radical pathways cannot
be excluded.
1HNMR (CD2Cl2, 300 MHz): δ 8.08 (s, 1H, CH=N), 8.04
(s, 1H, CH=N), 7.57 (m, 2H), 7.30 (m, 3H), 7.14 (d, 1H), 7.00
(m, 3H), 6.79 (m, 1H), 6.59 (m, 2H), 6.02 (br s, 1H, NH), 5.51
(br, 1H, NH), 4.61 (d, J=12.3 Hz, 1H, benzylic), 4.20-4.40
(m, 3H, benzylic), 3.16 (m, 1H, C6H10), 2.84 (m, 1H, C6H10),
2.65(m, 1H, C6H10), 2.32 (m, 3H, C6H10), 1.88 (m, 2H, C6H10),
1.50 (s, 9H, tbutyl), 1.43 (s, 9H, tbutyl), 1.11 (s, 9H, tbutyl), 0.96
t
(s, 9H, butyl) and other parts of C6H10. MS (ESIþ): m/z 564
(98MoO2L þ Hþ), with correct isotopic pattern for Mo. Anal.
Calcd (found) for C28H38N2O4Mo: C, 59.78 (59.55); H, 6.81
(6.91); N, 4.98 (4.91).
(21) Yoshiyuki, O.; Daigo, O.; Makoto, Y.; Akio, B. J. Am. Chem. Soc.
2002, 124, 13690–13691.

11296 Inorganic Chemistry, Vol. 48, No. 23, 2009
Ziegler et al.
Mo(O)2(3,5-tBu2-Saladach) (3). A mixture of (R,R)-1,2-cy-
clohexanediamino-N,N0-bis(3,5-di-t-butyl-salicylidene), H2-
(3,5-tBu2-saldach), (553 mg, 1.01 mmol), Mo(CO)6 (238 mg,
0.902 mmol) in THF (70 mL) was refluxed under air for 19 h in
which time the color changed from yellow to green-brown and
then dark purple. After cooling to ambient temperature, the
resulting dark brown solution was filtered and the filtrate was
dried in vacuo. The residue was treated with a small amount of
hexanes, filtered, and washed with cold hexanes. The product
was obtained as an orange powder. Yield: 232 mg, 38%.
1HNMR (CD2Cl2, 300 MHz): δ 8.06 (s, 1H, CH=N), 8.02
(s, 1H, CH=N), 7.61 (d, J=2.1 Hz, 1H), 7.59 (d, J=2.4 Hz, 1H),
7.28 (d, J=2.4 Hz, 1H), 7.24 (d, J=2.4 Hz, 1H), 7.16 (d, J=
2.4 Hz, 1H), 7.09 (d, J=2.4 Hz, 1H), 7.01 (d, J=2.4 Hz, 1H), 6.74
(d, J=2.4 Hz, 1H), 6.15 (br s, 1H, NH), 5.62 (br, 1H, NH), 4.60
(d, J=12.9 Hz, 1H, benzylic), 4.18-4.40 (m, 3H, benzylic), 3.15
(m, 1H, C6H10), 2.75 (m, 1H, C6H10), 2.62 (m, 1H, C6H10), 2.40
(m, 2H, C6H10), 2.23 (m, 2H, C6H10), 1.60-1.95 (m, 6H, C6H10),
1.51 (s, 9H, tBu), 1.43 (s, 9H, tBu), 1.35 (s, 9H, tBu), 1.33 (s, 9H,
tBu), 1.28 (s, 9H, tBu), 1.22 (s, 9H, tBu), 1.13 (s, 9H, tBu), 0.94
(s, 9H, tBu). MS (ESIþ): m/z 675 (98MoO2L þ Hþ), with correct
isotopic pattern for Mo. Anal. Calcd (found) for C36H54N2O4-
Mo: C, 64.08 (64.25); H, 8.07 (7.99); N, 4.15 (3.96).
Mo(O)2(3,5-tBu2-Salan) (4). The salan ligand was prepared in
86% yield by reduction of the corresponding salen ligand with
NaBH4 in THF-H2O, following a literature protocol.22 The
dioxo Mo complex was synthesized by refluxing a mixture of
Mo(O)2(acac)2 and H2Salan (slight excess) in MeOH, according
to a previous report.23 Yield: 80%. 1H NMR (C6D6, 300 MHz):
δ 7.49 (d, J=2.4 Hz, 2H), 6.86 (d, J=2.4 Hz, 2H), 5.23 (d, J=
14.7 Hz, 2H, benzylic), 3.64 (d, J=14.7 Hz, 2H, benzylic), 2.08
(m, 4H), 1.66 (s, 18H, tbutyl), 1.32 (s, 18H, tbutyl), 0.90 (m, 2H),
0.099 (m, 4H). NH signals were not located.
X-ray Data Collection and Structure Solution. A summary of
structure determination is provided here and further details can
be found in the Supporting Information. Crystallographic data
for 1: C32H47MoN2O4, 0.465(CH2Cl2), F.W.=659.18, 0.44 ꢀ 0.40 ꢀ
0.13 mm crystal dimension, triclinic, space group P1(#2),
˚
a=14.569(3), b=15.011(6), c=18.323(5) A; R=72.878(13),
3
˚
β =88.824(7), γ=72.504(11)°, V=3642.3(19) A . Z=4, Fcalcd
=
1.202 g/cm3, u=0.451 mm-1, 44292 reflections collected, 12726
unique (Rint=0.064), R1=0.057, R2=0.144, for 8629 reflections
with Fo2 > 2σ(Fo2).
Crystallographic data for 2: C28H38MoN2O4, F.W.=562.57,
0.38 ꢀ 0.31 ꢀ 0.25 mm crystal dimension, monoclinic, space
group P 1 21 1(#4). a=8.50330(10), b=22.8862(4), c=14.9349(3)
˚
A; β=106.2526(6)°, V=2790.30(8) A . Z=4, Fcalcd=1.339 g/cm3,
3
˚
u=0.490 mm-1, 30847 reflections collected, 12688 unique (Rint
=
2
0.039), R1=0.037, R2=0.075, for 9942 reflections with Fo2 >2σ(Fo ).
Crystallographic data for 4: C36H56MoN2O4,H2O,0.65-
(CH4O), F.W.=715.64, 0.38 ꢀ 0.15 ꢀ 0.13 mm crystal dimen-
sions. Hexagonal, space group P31 (#144), a=17.0026(13), c=
13.6160(12) A, V=3408.9(5) A . Z=3, Fcalcd=1.046 g/cm3, u=
0.315 mm-1, 9437 reflections collected, 6931 unique (Rint=0.079),
R1=0.055, R2=0.136, for 5432 reflections with Fo2 > 2σ(Fo2).
Crystallographic data for 6: C64H96Mo2N4O7,2(C7H8),
F.W. = 1409.66, 0.35 ꢀ 0.20 ꢀ 0.06 mm crystal dimension,
3
˚
˚
monoclinic, C12/c1(#15), a = 30.4913(17), b = 13.6081(8) c =
3
˚
˚
17.9982(6) A; β=93.323(3)°, V=7455.4(7) A . Z=4, Fcalcd
=
1.256 g/cm3, u=0.379 mm-1, 42133 reflections collected, 6562
unique (Rint=0.094), R1=0.045, R2=0.111, for 4763 reflections
with Fo2 > 2σ(Fo2).
Catalytic Hydrosilylation of Ketones: General Procedure. To a
J. Young NMR tube was added ketone (1.0 mmol), silane
(1.2 mmol), Mo catalyst (1 or 5 mol %), and solvent (0.6 mL).
The resulting solution was capped and heated in a sand bath at
1
MoIV(O)(tBu2-Salalen) (5). In a nitrogen-filled glovebox a
three-neck round-bottom flask with a magnetic stir bar was
charged with Mo(O)2(Salalen-tBu2), 1, (0.200 g, 0.322 mmol)
dissolved in toluene. The reaction flask was capped and taken
out of the glovebox. Under nitrogen atmosphere a condenser
was attached to the round-bottom flask. To this solution was
added PMe3 1.0 M solution in toluene (1.61 mL, 1.61 mmol)
through one of the side arms. The resulting solution was heated
in a sand bath at ∼120 °C for 24 h. The reaction mixture was
filtered under inert atmosphere and the filtrate was dried in
vacuo. Yield 0.049 g, 25%.
∼110 °C. The reaction progress was monitored by H NMR.
After the reaction was complete, as indicated by the complete
disappearance of ketone, the reaction mixture was hydrolyzed in
acidic ether for a few hours. After extraction with diethylether,
the organic phase was dried, concentrated, and subjected to
flash chromatography on a silica gel column, and eluted with
hexanes-EtOAc. The isolated alcohols were identified by com-
paring the 1H NMR spectra with literature reports.
Reaction of MoVI(O)2(Salalen-tBu2) with PhSiH3. A mixture
of Mo(O)2(salalen-tBu2) (22 mg, 0.035 mmol), PhSiH3 (8.6 μL,
0.070 mmol) in C6D6 (0.62 mL) was heated in a sand bath to
∼100 °C and the reaction progress was monitored by 1H NMR
spectroscopy. New peaks in the 5.5-1.9 ppm region were
observed alongside the starting complex MoVI(O)2-
(salalen-tBu2). Further heating to 110 °C for several hours
resulted in increasing the intensity of these peaks (∼60% of
total detectable Mo), accompanied by nearly complete disap-
pearance of starting MoVI(O)2(salalen-tBu2) and the appearance
of MoIV(O)(salalen-tBu2), 5. This intermediate species was
tentatively assigned as MoIV(OH)(OSiPhH2)(salalen-tBu2).
1HNMR (C6D6, 300 MHz): δ 5.07 (d, J=14.1 Hz, 1H, benzylic),
3.30 (d, J=13.8 Hz, 1H, benzylic), 2.42 (br, 2H, OH and NH).
Other peaks were difficult to assign because of their close
proximity and overlap to peaks from compounds 1 and 5.
1HNMR (CD2Cl2, 300 MHz): δ 8.15 (s, 1H, CH=N), 7.65 (m,
1H, aromatic), 7.35 (m, 1H, aromatic), 7.17 (s, 1H, aromatic),
6.76 (s, 1H, aromatic), 4.19 (d, J=8.6 Hz, 1H, benzylic), 3.52
(m, 2H, NH þ benzylic), 3.33 (m, 1H, ethyl), 2.90 (m, 2H, ethyl),
2.59 (m, 1H, ethyl), 1.48 (s, 9H, tbutyl), 1.35 (s, 9H, tbutyl), 1.23
(s, 9H, tbutyl), 1.13 (s, 9H, tbutyl). MS (ESIþ): m/z 606
(98MoOL) with the correct isotope pattern for Mo.
μ-Oxo-(Mo(O)(salalen-tBu2))2 (6). In a nitrogen-filled glove-
box a three-neck round-bottom flask with a magnetic stir bar
was charged with Mo(O)2(Salalen-tBu2), 1, (0.200 g, 0.322 mmol
dissolved in toluene. The reaction flask was capped, taken out of
the glovebox, and under nitrogen atmosphere a condenser was
attached to the flask. To this solution was added PMe3 1.0 M
solution in toluene (644 μL, 0.644 mmol) through one of the side
arms. The resulting solution was heated in a sand bath at ∼120 °C
for 24 h. The mixture was filtered and the remaining dark red solid
was rinsed with toluene and dried in vacuo. Yield 0.0912 g, 46%.
Anal. Calcd (found) for C67H96Mo2N4O7: C, 62.73 (62.73); H,
7.90 (7.94); N, 4.57 (4.70).
Acknowledgment. This research was supported by the Che-
mical Sciences Division, Office of Basic Energy Sciences, U.S.
Department of Energy (Grant DE-FG02-06ER15794).
Supporting Information Available: Crystallographic tables for
atomic coordinates and equivalent isotropic parameters, bond
lengths, and angles for complexes 1, 2, 4, and 6 (CIF). Experi-
mental details for X-ray studies (PDF). This material is available
free of charge via the Internet at http://pubs.acs.org.
(22) Zeynizadeh, B.; Behyar, T. Bull. Chem. Soc. Jpn. 2005, 78, 307–315.
(23) Subramanian, P.; Spence, J. T.; Ortega, R.; Enemark, J. H. Inorg.
Chem. 1984, 23, 2564–72.