Dinuclear niobium(III) and tantalum(III) complexes with thioether and selenoether ligands [{MIIIX2(L)}2(μ-X) 2(μ-L)] (M = Nb, Ta; X = Cl, Br; L = R2S, R 2Se): Syntheses, structures, and the optimal conditions and the mechanism of the catalysis for regioselective cyclotrimerization of alkynes

Article abstract of DOI:10.1016/j.jorganchem.2013.07.035

A series of dinuclear niobium(III) and tantalum(III) halide complexes with chalcogen donors [{M^IIIX₂(L)}₂(μ-X) ₂(μ-L)] (M = Nb; X = Cl, Br; L = R₂S, R₂Se) (M = Ta; X = Cl, Br; L = R₂S) have been prepared by the reaction of dimetal(V) decahalide (M₂X₁₀) with magnesium and L. The structures of five of those new complexes were determined by X-ray crystallography to have a M-M double bond. It is evidenced that the solvent and the temperature play important roles in achieving high yield and regioselectivity for cyclotrimerization of alkynes. The reaction of the niobium(III) chloride complexes (L = dimethyl sulfide (Me₂S), tetrahydrothiophene (C₄H₈S, THT, thiolane)) with phenylacetylene at room temperature in toluene gave both head-to-tail cycloadded trimers of alkyne, 1,3,5-(Ph)₃-2,4,6-(H)₃-benzene and head-to-head cycloadded 1,2,4-(Ph)₃-3,5,6-(H)₃-benzene, in high yields. The smaller the thioether ligands, the higher the catalytic activity. The niobium(III) chloride complexes with selenoether (L = dimethyl selenide (Me₂Se), tetrahydroselenophene (C₄H ₈Se, THSe, selenolane) have higher rates for the reaction with alkynes, but low activity for the cyclotrimerization. Tantalum(III) chloride complexes (L = Me₂S, THT, tetrahydrothiopyran (C₅H ₁₀S, THTP, thiane)) have lower catalytic activities than the niobium(III) ones, because the Ta-S(μ-L) bond lengths are shorter than those of niobium analogs. The stability of the precursor complexes toward the first oxidative addition depends on the M-S(μ-L) bond strength, and controls the concentration of catalytic active species. The niobium(III) bromide complexes (L = Me₂S, THT) and the tantalum(III) bromide one (L = Me₂S) react with alkynes to give head-to-tail compounds regioselectively, but are less catalytically active than those of chloride complexes.

Full text of DOI:10.1016/j.jorganchem.2013.07.035

Journal of Organometallic Chemistry 745-746 (2013) 288e298
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Dinuclear niobium(III) and tantalum(III) complexes with thioether and
III
selenoether ligands [{M X (L)} (
m
-X)2(
m
-L)] (M ¼ Nb, Ta; X ¼ Cl, Br;
2
2
L ¼ R S, R Se): Syntheses, structures, and the optimal conditions and
2
2
the mechanism of the catalysis for regioselective cyclotrimerization
of alkynes
Masatoshi Matsuura ,*, Takashi Fujihara , Masaki Kakeya , Tomoaki Sugaya ,
Akira Nagasawa
a
a
b
,
a
a
*
a
Department of Chemistry, Graduate School of Science and Engineering, Saitama University, Shimo-Okubo 255, Sakura-ku, Saitama 338-8570, Japan
Comprehensive Analysis Center for Science, Saitama University, Shimo-Okubo 255, Sakura-ku, Saitama 338-8570, Japan
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 19 February 2013
Received in revised form
A series of dinuclear niobium(III) and tantalum(III) halide complexes with chalcogen donors
III
[
{M X
2
(L)}
2
(
m
-X)
2
(
m
-L)] (M ¼ Nb; X ¼ Cl, Br; L ¼ R
2
S, R
2
Se) (M ¼ Ta; X ¼ Cl, Br; L ¼ R
2
S) have been
prepared by the reaction of dimetal(V) decahalide (M
2
X
10) with magnesium and L. The structures of five
6
July 2013
of those new complexes were determined by X-ray crystallography to have a MeM double bond. It is
evidenced that the solvent and the temperature play important roles in achieving high yield and
regioselectivity for cyclotrimerization of alkynes. The reaction of the niobium(III) chloride complexes
Accepted 11 July 2013
Keywords:
Niobium
Tantalum
Chalcogen
Cyclotrimerization
Alkyne
Regioselectivity
(
L ¼ dimethyl sulfide (Me
temperature in toluene gave both head-to-tail cycloadded trimers of alkyne, 1,3,5-(Ph)
benzene and head-to-head cycloadded 1,2,4-(Ph) -3,5,6-(H) -benzene, in high yields. The smaller the
thioether ligands, the higher the catalytic activity. The niobium(III) chloride complexes with selenoether
(L ¼ dimethyl selenide (Me Se), tetrahydroselenophene (C Se, THSe, selenolane) have higher rates for
the reaction with alkynes, but low activity for the cyclotrimerization. Tantalum(III) chloride complexes
L ¼ Me S, THT, tetrahydrothiopyran (C 10S, THTP, thiane)) have lower catalytic activities than the
niobium(III) ones, because the TaeS( -L) bond lengths are shorter than those of niobium analogs. The
stability of the precursor complexes toward the first oxidative addition depends on the MeS( -L) bond
strength, and controls the concentration of catalytic active species. The niobium(III) bromide complexes
S) react with alkynes to give head-to-tail
2
S), tetrahydrothiophene (C
4
H
8
S, THT, thiolane)) with phenylacetylene at room
3
-2,4,6-(H)
3
-
3
3
2
4 8
H
(
2
5
H
m
m
(
L ¼ Me
2
S, THT) and the tantalum(III) bromide one (L ¼ Me
2
compounds regioselectively, but are less catalytically active than those of chloride complexes.
Ó 2013 Elsevier B.V. All rights reserved.
1. Introduction
complexes as catalysts has been focused on those of Ti [2], Zr [3],
Mo [4], etc., while the catalytic activity of the lower oxidation
During the last six decades, numerous studies have revealed
that many transition metal complexes catalyze the cyclo-
trimerization of alkynes, but even now it is a major challenge for
synthetic chemistry to develop methods and conditions with high
regio- and chemoselectivity in the cyclotrimerization of a wide
variety of alkynes [1]. The development of early transition metal
states of Nb and Ta in discrete complexes remains relatively un-
explored. It has been reported that the mononuclear Nb and
Ta complexes, [NbCl
3
(DME)] and [TaCl
3
(DME)] (DME ¼ 1,2-
dimethoxyethane), respectively, have catalytic activity to
let alkynes cyclotrimerize [5,6]. These catalytic reactions always
gave a mixture of 1,3,5- and 1,2,4-cyclotrimers, which are formed
through head-to-tail and head-to-head additions, respectively,
and the yields generally do not depend on the properties of the
substituent on the alkynes.
*
Corresponding authors.
On the other hand, the dinuclear Nb and Ta complexes
E-mail addresses: masatoshi_m@chem.saitama-u.ac.jp (M. Matsuura), fuji@
chem.saitama-u.ac.jp (T. Fujihara).
III
[{M X
2
(L)}
2
(
m
-X)
2
(
m
-L)] ([Nb
2
Cl
6
(Me
2
S)
3
] (1a) (Me
2
S ¼ dimethyl
0
022-328X/$ e see front matter Ó 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jorganchem.2013.07.035

M. Matsuura et al. / Journal of Organometallic Chemistry 745-746 (2013) 288e298
289
sulfide), [Nb
thiolane)), [Ta
2
Cl
6
(THT)
(Me
3
] (1b) (THT ¼ tetrahydrothiophene (C
4
H
8
S,
and high yields to prepare [Nb
(1b) rather than using Na/Hg or Na/K [17]. We therefore employed
Mg for synthesizing other complexes 1ce1h, [Ta Cl (Me S) ] (2a),
[Ta Cl (THT) ] (2b), [Ta Cl (THTP) ] (2d), and [Ta Br (Me S) ] (2g).
All these complexes were provided by the reaction of dimetal(V)
decahalide (M 10) with Mg and L in a mixture of CH Cl or toluene
with diethyl ether (Et O) for several days as shown in Scheme 3. All
the prepared complexes are air-unstable solids. Single crystals
2 6 2 3 2 6 3
Cl (Me S) ] (1a) and [Nb Cl (THT) ]
2
Cl
6
2
S) ] (2a), and [Ta Cl (THT) ] (2b)) are well
3
2
6
3
known to act as a starting materials for synthesizing other
dinuclear and multinuclear complexes (Scheme 1) [7e9]. Espe-
cially, reactions of [Nb
with alkynes give various products including polymers, cyclo-
2
6
2
3
2
6
3
2
6
3
2
6
2
3
2 6 3 2 6 3
Cl (THT) ] (1b) and [Ta Cl (THT) ] (2b)
2
X
2
2
2
trimers, and complexes with
of alkyne [10e14]. Recently, we have found that the reaction of
Nb Cl (Me S) ] (1a) and [Nb Cl (THT) ] (1b) with various al-
kynes (R C^CR ) gave cycloadded trimer, 1,3,5-(R -2,4,6-(R
benzene regioselectively in high yields, even at room temperature
in CH Cl . We have also suggested a plausible mechanism for the
h
-alkene depending upon the sort
2
[
2
6
2
3
2
6
3
were obtained for the new complexes [Nb
[Nb Cl (THTP) ] (1d), [Nb Br (Me S) ] (1g), [Ta
and [Ta Br (Me S)
2
Cl
Cl
6
(Et
(THTP)
2
S)
3
]
(1c),
] (2d)
1
2
1
)
3
2
)
3
-
2
6
3
2
6
2
3
2
6
3
1
13
2
6
2
3
] (2g). The H and C NMR spectra in CDCl
3
for
2
2
all the complexes suggest that the molecular structures in the solid
state (see below) remain unchanged in solution. The diamagnetism
of these complexes is rationalized by the formation of a MeM
cyclotrimerization (Scheme 2) [15]. The reaction mechanism
consists of three consecutive processes. In the first step, after
leaving of the bridging sulfur in the precursor complex, the
oxidative addition of the alkynes to the metal center takes place
2
double bond, although M in the oxidation number þ3 has d
electron configuration.
V
2
2 1 2 2 2
to give [{M X (L)(h -R C]CR )} (m-X) ]. The addition of the
second alkyne to form a five-membered metallacycle is the sec-
ond step, and it is the source of the regioselectivity. The third step
forms metallaeheptacyclic complex. This step is also regiose-
lective. On reaction of the complex with a new alkyne, the orig-
inal catalytically-active complex is reproduced.
We still do not know the determining factors for such high
regioselectivity and high yields. It will be necessary to examine the
influences of the sort of central metal, the electronic state and steric
bulkiness of the sulfur and selenium donor and the halide, and
environments such as temperature and solvent, to verify the
optimal conditions for the catalytic reaction.
2.2. Molecular structure of the catalyst [{MIIIX
2 2 2
(L)} (m-X) (m-L)]
Molecular structures were determined by the single crystal X-
ray crystallographic analyses for the new complexes [Nb
(1c), [Nb Cl (THTP) ] (1d), [Nb Br (Me S) ] (1g), [Ta Cl
(2d) and [Ta Br (Me S)
2
Cl
6
(Et
2
S)
3
]
]
2
6
3
2
6
2
3
2
6
(THTP)
3
2
6
2
3
] (2g). The ORTEP drawing and selected
bond lengths and angles of 1c are shown in Fig. 1 and Table 1,
respectively, and those of others are collected in Fig. S1, and
Tables S3 and S4, respectively. We have reanalyzed the molecular
structures of two known complexes [Nb
2 6 3
Br (THT) ] (1h) and
[Ta Cl (Me S) ] (2a) [18,19] (shown in Fig. S1), and used those re-
2
6
2
3
As one step towards this goal we report here on the
syntheses and structures of the new complexes of the type
sults for the discussion about the influences of the MeM double
bond and coordination bonds. The above mentioned five new
complexes have very similar structures with those of known ones
[15,18e20].
III
[
{M X
2
(L)}
Cl
thiane)), [Nb
Nb Cl (THSe)
selenolane)), [Nb
2
(
m
-X)
(THTP)
Cl (Me
] (1f) (THSe ¼ tetrahydroselenophene (C
Br (Me S) (1g), [Ta Cl (THTP) (2d), and
] (2g), and on the reaction of all these complexes
1ae1h, 2a, 2b, 2d, and 2g) with phenylacetylene under various
2
(
m
-L)]: [Nb
] (1d) (THTP ¼ tetrahydrothiopyran (C
Se) ] (1e) (Me
Se ¼ dimethyl selenide),
Se,
2
Cl
6
(Et
2
S)
3
] (1c) (Et
2
S ¼ diethyl sul-
fide), [Nb
2
6
3
5
H10S,
2
6
2
3
2
[
2
6
3
4
H
8
2.2.1. Presence of MeM double bond
2
6
2
3
]
2
6
3
]
Each metal center has a pseudo-octahedral environment with a
shared trigonal face consisting of one sulfur donor ligand and two
halides, the coordination of which is completed by terminal two
halides and one sulfur donor. Formation of the metalemetal
bonding is obvious from the MeM distance, e.g., NbeNb
[
(
2 6 2 3
Ta Br (Me S)
conditions to reveal optimal conditions.
ꢀ
ꢀ
2
[
.6888(4) A in [Nb
Ta Cl (THTP) ] (2d), and the former is longer than the latter. This
trend is clearly seen for other pairs of Nb and Ta complexes, i.e.
2 6 3
Cl (THTP) ] (1d); TaeTa 2.6730(5) A in
2
. Results and discussion
2
6
3
.1. Syntheses of [{MIIIX
2 2 2
(L)} (m-X) (m-L)]
2
[
(
Nb
1b) vs. [Ta
2
Cl
6
(Me
2
S)
Cl
3
] (1a) vs. [Ta
2 6 2 3 2 6 3
Cl (Me S) ] (2a), and [Nb Cl (THT) ]
2
6
(THT) ] (2b) [15,18,20]. Those distances lie within
3
We have previously adopted Tsunoda’s method, which uses
III
III
III
III
III
the range of usual atomic distances of Nb eNb and Ta eTa
double bonds, 2.61e2.94 A and 2.51e2.84 A, respectively [7,21].
2 2 2
magnesium as a reductant, to prepare [{Nb Cl (L)} (m-Cl) (m-
ꢀ
ꢀ
L)] [15,16]. It is highly useful due to the convenience of treatment
A further indication of the strong metalemetal attraction is the
displacement of the metal centers toward one another away from
the center of the idealized octahedral coordination sphere. Aver-
X
L
X
X
L
X
ages of deviations of the metal center from the Cl
Cl(2)eCl(3)eCl(4) or Cl(3)eCl(4)eCl(5)eCl(6) in each coordination
plane Cl(1)e
4
M
M
ꢀ
ꢀ
sphere are 0.167 A and 0.181 A in [Nb
[Ta Cl (THTP) ] (2d), respectively. Averages of deviations of the
metal center from the Br plane Br(1)eBr(2)eBr(3)eBr(4) or Br(3)e
Br(4)eBr(5)eBr(6) are 0.187 A and 0.204 A in [Nb
and [Ta Br (Me S) ] (2g), respectively. Deviations in the complexes
2 6 3
Cl (THTP) ] (1d) and
X
2
6
3
X
L
4
ꢀ
ꢀ
2
6 2 3
Br (Me S) ] (1g)
1
1
1
1
1
1
1
1
a (M = Nb, X= Cl, L = Me S)
2a (M = Ta, X= Cl, L = Me S)
2
2
2
6
2
3
b (M = Nb, X= Cl, L = THT)
2b (M = Ta, X= Cl, L = THT)
2d (M = Ta, X= Cl, L = THTP)
with Br are similar with those in the corresponding complexes with
Cl, and the trend is common for Nb and Ta complexes: deviations
are larger in Ta complexes than that in Nb ones, and in complexes
with Br than in complexes with Cl. The tendency may be due to the
more enhanced covalency of MeM and Mem-S bonds in Ta com-
plexes than that of Nb complexes.
The influence of the thioether ligand on the NbeNb and TaeTa
bonds is little, since the complexes of the same metal center and the
same halide have very similar MeM distances despite the thioether
c (M = Nb, X= Cl, L = Et S)
2
d (M = Nb, X= Cl, L = THTP)
2g (M = Ta, X= Br, L = Me S)
2
e (M = Nb, X= Cl, L = Me Se)
2
f (M = Nb, X= Cl, L = THSe)
g (M = Nb, X= Br, L = Me S)
2
h (M = Nb, X= Br, L = THT)
Scheme 1. Dinuclear Nb and Ta complexes [{MIII
X (L)} (m-X) (m-L)] (1ae2g).
2 2 2

290
M. Matsuura et al. / Journal of Organometallic Chemistry 745-746 (2013) 288e298
R1
Cl
Cl
L
L
Cl
NbIII
L
Cl
Cl
2
R
R
, − L
Cl
1
2
NbIII
R2
V
Nb
V
Nb
R2
L, − 2 R1
R2
Cl
Cl
Cl
Cl
Cl
L
Cl
L
1
st step
R1
Precursor
Active species
2
R1
R2
2
nd step
R2
R1
Cl
2
R1
R2
L
R1
Cl
Cl
R2
V
V
Nb
L
Nb
R2
Cl
Cl
R1
Cl
R1 R2
2
R1
R2
3
rd step
R1
R2
R2
R2
R1
R2
R1
R1
Cl
L
Cl
Cl
Cl
R2
R1
R1
V
Nb
V
R1
R2
Nb
R2
R2
Cl
L
Cl
R1
R1
R2
Scheme 2. Plausible mechanism for the cyclotrimerization.
ꢀ
ligands, such as for Nb and Cl, [Nb
2
Cl
6
(Me
2
S)
Cl
] (2b) and [Ta
3
] (1a), [Nb
(THTP)
Cl
2
Cl
(1d), and
(THTP) ] (2d),
6
(THT)
3
]
2.6785(8) A, with those having bromide bridges, [Nb
2
Br
6
(Me
2 3
S) ]
ꢀ
ꢀ
(
[
1b), [Nb
Ta Cl (Me
2
Cl
S)
6
(Et
2
S)
3
]
(1c) and [Nb
Cl (THT)
2
6
3
]
(1g) 2.7253(4) A and [Ta
trend, and which holds for the complexes with THT [19]. The ionic
radius of the bromide ion, 1.96 A, is larger than that of chloride ion
2
6
Br (Me
2
S)
3
] (2g) 2.7072(8) A, reveals the
2
6
2
3
] (2a), [Ta
2
6
3
2
6
3
ꢀ
respectively.
On the other hand, influence of the halide is obvious. The MeM
bonding is shorter in the complexes with Cl than in ones with Br,
and the trend is common for Nb and Ta complexes. Comparison of
C(4)#1
C(4)
2
the structures of complexes with Me S having chloride bridges,
ꢀ
[
Nb
2
Cl
6
(Me
2
S)
3
]
(1a) 2.6880(3)
A
and [Ta
2
Cl
6
(Me
2
S)
3
]
(2a)
C(3)
C(3)#1
Cl(3)#1
S(1)
S(1)#1
X
Cl(3)
L
X
X
L
X
C(2)
C(2)#1
L / Mg
M X
2
M
M
C(1)
C(1)#1
10
Et O-CH Cl or Et O-toluene
2 2 2 2
Nb(1)
Nb(1)#1
X
X
r.t.
L
1
1
1
1
1
1
1
1
a (88%)
b (64%)
c (88%)
d (84%)
e (92%)
f (63%)
g (51%)
h (69%)
2a (60%)
2b (56%)
2d (53%)
2g (75%)
Cl(1)
Cl(2)
C(5)
Cl(1)#1
Cl(2)#1
S(2)
C(5)#1
C(6)#1
C(6)
Fig. 1. The ORTEP drawings of 1c with the numbering scheme. Ellipsoids are drawn at
their 50% probability level; hydrogen atoms were omitted for clarity.
Scheme 3. SynthesesofdinuclearNbandTacomplexes, [{MIII
2 2 2
X (L)} (m-X) (m-L)](1ae2g).

M. Matsuura et al. / Journal of Organometallic Chemistry 745-746 (2013) 288e298
291
Table 1
Selected bond lengths (A) and angles (deg) for [Nb
ꢀ
2 6 2 3
Cl (Et S) ] (1c).
[
Nb Cl (Et S)
] (1c)a
2
6
2
3
Nb(1)eNb(1)#1
Nb(1)eCl(1)
Nb(1)eCl(2)
Nb(1)eCl(3)
Nb(1)eCl(3)#1
Nb(1)eS(1)
Nb(1)eS(2)
Nb1eCl(1)eCl(2)eCl(3)eCl(3)#1
S(1)eS(1)#1
2.6903(18)
Cl(1)eCl(2)
Cl(3)eCl(3)#1
3.6613 (41)
3.0976 (53)
2.381(2)
2.369(3)
2.513(2)
2.468(2)
2.628(3)
2.412(3)
0.1610 (19)
6.1659 (56)
Nb(1)eCl(3)eNb(1)#1
Nb(1)eS(2)eNb(1)#1
S(1)eNb(1)eS(2)
S(2)eNb(1)eNb(1)#1
C(1)eS(1)eC(3)
65.39(7)
67.78(10)
170.42(8)
56.11(5)
103.8(8)
103.3(9)
C(5)eS(2)eC(5)#1
a
Symmetry transformations used to generate equivalent atoms:#1 y, x, ꢁz.
ꢀ
1
.81 A [22]. Bromide having bigger ionic radius compared with
We have already discussed the difference in the bond nature
between NbeS and NbeCl from the incorporation of the 3d orbital
[15], but the treatment seems incorrect. In this study, we did the
chloride may lead to the elongation of the metalehalide bond
ꢀ
ꢀ
(
2.63 A for Br, 2.50 A for Cl), and the decreased angle of metale
ꢀ
ꢀ
halideemetal (62 for Br, 65 for Cl). We do not know whether
these changes in geometry may cause or result of the elongation of
MeM bonding.
The results of the X-ray crystallography explain that MeM
bonding is shorter in the chloride complexes than in the bromide
ones. Quantum chemical calculations, Natural Bond Orbital (NBO)
Analysis, were carried out for assessing the electronic effect on the
MeM bonding. The occupancy of electrons on the NbeNb bonding
NBO calculations and revealed that the
s
*(Nbe
hybrid of
extent of the stabilization of the complex resulting from the exis-
m
-S) consists of the
3
.24 4.06
2.23
sp
d
hybrid of Nb and the sp
m-S (Table S8). The
ꢁ
1
tence of one bridging sulfur donor ligand Me
2
S (1042 kJ mol ) is
larger than that from the existence of two bridging chlorides
ꢁ1
(297 kJ mol ). Although there is an obvious difference in in-
teractions of -S with NbeNb bonding from those of -Cl with Nbe
-S in those
m
m
Nb bonding, no incorporation of d orbitals of
interactions.
m
2 6 2 3 2 6 2 3
in [Nb Cl (Me S) ] (1a) is larger than that in [Nb Br (Me S) ] (1g).
The larger occupancy on the NbeNb bonding leads to the shorter
bond lengths. Therefore, the elongation of MeM bonding in
bromide-bridged complexes are understandable from the view-
point of electronic factors (Tables S7 and S9).
2.3. Active species
Our previous paper suggests that the catalytic active interme-
diate retains the dinuclear structure throughout the reaction, and
that the structure is essential for regioselectivity [15]. The latter is
rationalized from the catalytic abilities of analogous catalysts
2.2.2. Coordination bonds to sulfur atoms and halides
The influence of the sort of the thioether ligand on the NbeS and
TaeS bonds is little, since Nb complexes with chloride, such as
Nb Cl (Me S) ] (1a), [Nb Cl (THT) ] (1b), [Nb Cl (Et S) ] (1c)
and [Nb Cl (THTP) (1d), and Ta complexes with chloride
Ta Cl (Me S) ] (2a), [Ta Cl (THT) ] (2b) and [Ta Cl (THTP) ] (2d),
3 2 6 3
[NbCl (DME)] and [Mo Cl (THT) ] [4a,5]. Quantum chemical cal-
culations may also help the understanding on the difference in the
stabilities of dinuclear and mononuclear complex.
[
2
6
2
3
2
6
3
2
6
2
3
2
6
3
]
[
2
6
2
3
2
6
3
2
6
3
It has been reported that the mononuclear Nb complex,
have very similar MeS lengths with each other, respectively.
3
[NbCl (DME)] has catalytic activity to let alkynes cyclotrimerize
However the difference in bond length is seen among metal-to-
bridging thioether (m-S) bonding: the bond length of TaeS in
[5]. This catalytic reaction always gave a mixture of 1,3,5- and
1,2,4-cyclotrimers, which are formed through head-to-tail and
head-to-head additions, respectively. The yields of both cyclo-
trimers generally do not depend on the properties of the sub-
ꢀ
[
[
Ta
Nb
2
Cl
Cl
6
(THTP)
3
] (2d) (2.389 A) is shorter than that of NbeS in
ꢀ
2
6
(THTP) ] (1d) (2.423 A), while the length of Ta to the ter-
3
ꢀ
minal thioether (
h
h
-S) bonding (2.624 A) is almost the same as Nbe
-S (2.639 A). The similar trend is seen for other pairs of Nb and Ta
complexes, i.e. [Nb Cl (Me S) ] (1a) vs. [Ta Cl (Me S) ] (2a), and
Nb Cl (THT) ] (1b) vs. [Ta Cl (THT) ] (2b) [15,18,20].
If the metalesulfur bond were purely ionic, the distance should
stituent on the alkynes. The reaction of [NbCl
C^CR (R Si. R ¼ H, Ph.) gave complexes with
¼ Ph, Et, Me
metallacyclopropene [NbCl C]CR )] [5]. For the analogous
Ta complexes, the reaction of [TaCl
etc), as well as [TaCl (DME)], with alkyne gave cyclotrimerized
products [6]. Therefore, in the Nb case it is reasonable that an
3
(DME)] with
ꢀ
R
1
2
1
3
2
-R
2
2
6
2
3
2
6
2
3
3
(
h
1
2
-R
2
[
2
6
3
2
6
3
3
(
h
1
C]CR
2
)] (R
1
, R
2
¼ Et,
3
ꢀ
coincide with the sum of ionic radii of metal (0.72 A for six-
coordinate Nb and Ta ) and sulfide (1.84 A) to be 2.56 A. The
III
III
ꢀ
ꢀ
2
intermediately generated [NbCl
active species. The dinuclear Mo complex [Mo
same skeletal structure as [Nb Cl (THT) ] (1b). However, for the
3
(
h
-R
1
C]CR
2
)] has a catalytically
distance to the bridging ligand is fairly shorter, and that to the
2
Cl (THT) ] has the
6
3
terminal ligand (
the interaction of
of -S. The difference is due to the influence of the covalency of the
h
-S) is longer, than this value. This suggests that
2
6
3
m-S with metal center is much stronger than that
former Mo complex, the feature of the catalytic activity is entirely
different, i.e. exhibiting linear oligomerization, cyclotrimerization,
h
metalesulfur interactions.
and polymerization, from those of the latter, and a mononuclear
2
The trend in MeS bond length is common for Nb and Ta
complexes, in spite of the chloride and bromide complexes. We
reveal that the size of the triangular shape which consists of two
terminal and one bridging sulfurs (
center remains similar even if the halide alternates. The distance
catalytically active species [MoCl
Et. R
example, the reaction of [Mo
(R
¼ Me, Et, H. R
2
(THT)
2
(
h
-R
1
C]CR
2
)] (R
1
¼ Me,
2
¼ Ph) is generated from the reaction with alkynes [4a]. For
2
Cl (THT) ] with alkynes R C^CR
6
3
1
2
h
-Se
m
-Se
h
-S) on each metal
1
2
¼ Ph, n-Bu, n-Pr) gave a mixture of 1,3,5- and
1,2,4-cyclotrimers. Although the authors claimed the steric bulk-
iness of the alkyne plays a dominant role in giving a regiose-
lectivity for the cyclotrimers. We tend to think the difference in
ꢀ
of
h
-Se
h
-S in the chloride complex [Nb
2
Cl
6
(Me
2
S)
Br
-S and Nbe
] (1a) are similar with those of [Nb Br (Me
1g), respectively, the sort of halide influences little on the
3
] (1a): 6.233 A
(Me S) ] (1g):
-S in
S)
is similar with that in the bromide complex [Nb
2
6
2
3
ꢀ
6
.197 A. Since the lengths of both Nbe
m
h
catalytic activity between [Nb
comes from the structure of active species. Cotton and coworkers
reported that the [M Cl (THT)
] (M ¼ Nb (1b), Ta (2b)) with al-
kynes R C^CR (R ¼ Ph; R ¼ t-Bu. R ¼ Me) gave
¼ Ph, R
2 6 3 2 6 3
Cl (THT) ] and [Mo Cl (THT) ]
[Nb Cl (Me S)
(
2
6
2
3
2
6
2
3
]
2
6
3
triangular shape.
1
2
1
2
1
2

292
M. Matsuura et al. / Journal of Organometallic Chemistry 745-746 (2013) 288e298
[
{MCl
2
(THT)( 2
h
-R
1
C]CR
2
)}
2
(
m
-Cl)
2
m
] which retains the dinuclear
-Cl bonds. We also confirmed
All of the complexes dealt in the present study are easily
decomposed in highly polar solvents, i.e. CH CN, EtOH etc. In the
presence of excess amount of electron pair donor, such as THF,
terminal sulfur donor ligands are substituted, although the bridging
sulfur-based ligand did not dissociate from the metal center [7r,s].
Thus we examined the catalytic reaction of the complexes in non-
structure without cleaving of Me
the generation of this dinuclear complex with
reaction of 2-hexyne with 1/6 equiv of [Nb
CDCl
3
2
h
-alkene from the
2
Cl (THT) ] (1b) in
6
3
1
3
(0.4 mL) by monitoring with H NMR at room temperature
[15]. In conclusion, it was thought that the intermediate is in the
dinuclear structure throughout the reaction, and the structure is
essential for regioselectivity.
2 6 2 3
polar solvents. [Nb Cl (Me S) ] (1a) reacted with phenylacetylene
to give cycloadded trimers in various solvents shown in Table 2. As
seen in Table 2, the solvent plays an important role for determining
the regioselectivity and the yield of the cyclotrimers. The catalytic
The DFT calculations suggest that the “catalytically active spe-
cies” has a dinuclear structure. We have carried out calculations on
2
[
{NbCl
2
(Me
2
S)(
(Me
h
-CH
h
3
C]CCH
3
)} -Cl)
2
(
m
2
], and mononuclear com-
activity was decreased in CHCl
3 4
, CCl , toluene and benzene. In these
2
plex [NbCl
3
2
S)(
-CH
3
C]CCH )] generated from the cleavage
3
solvents oxidation and/or the dissociation of ligands of the catalyst
1
13
of the bond to bridging Cl (Scheme S1). We first characterized
optimized geometries as potential minima on the potential energy
surface. The mononuclear complex has a distorted trigonal bipyr-
amidal structure (Fig. S4). We obtained the total energy of each
complex, and calculate the total energy difference between dinu-
clear and mononuclear complexes (see Supporting information). One
molecule of dinuclear complex is more stable than two molecules
precursor complex was not observed. The H and C NMR spectra
were monitored [Nb Cl (Me S) ] (1a) in CCl at room temperature
2
6
2
3
4
for 4 h. The signal intensities of free ligands did not increase.
Therefore, the low activity these solvents may be due to the local
elevation of the temperature despite enough stirring and sufficient
cooling, since the system seems to produce some heat during the
reaction in these solvents. The regioselective production of 3
ꢁ
1
of mononuclear complexes by 148 kJ mol
.
catalyzed by [Nb
2 6 2 3 2 2
Cl (Me S) ] (1a) took place in CH Cl , THF, and
6 5
C H
Cl. The system is kept under mild conditions because the re-
2
.4. Optimal conditions for Nb complexes
action is less exothermic in these solvents, and gives 1,3,5-isomer
regioselectively. The improvement of the selectivity in haloarenes
is a general tendency (entry 6 and 12), while haloalkanes except
The reaction of [Nb
with phenylacetylene has been reported to give head-to-tail
cycloadded trimers of alkyne, 1,3,5-(Ph) -2,4,6-(H) -benzene (3)
regioselectively in CH Cl at room temperature, and these com-
plexes act as catalysts to give cycloadded products regioselectively
15]. It is important to understand the mechanism and limiting
2 6 2 3 2 6 3
Cl (Me S) ] (1a) and [Nb Cl (THT) ] (1b)
2 2
CH Cl are not suitable for solvents. Eventually, the optimal con-
3
3
ditions in the solvent and temperature were disclosed to be CH
and below 40 C, respectively.
2
Cl
2
ꢀ
2
2
[
2.5. Influence of the metal center on the catalytic activity
factors for this regioselective cyclotrimerization in high yields.
Therefore, we investigated the reaction under various conditions,
particularly of the temperature and the solvent, in order to disclose
the influence on the catalytic activity and the regioselectivity. In our
We investigated the influence of the metal center on the cata-
lytic ability. Data shown in Table 3 suggest that Ta complexes have
lower activity towards phenylacetylene than Nb ones. It is
reasonably supposed that the higher the electron density at the Me
M double bond which is back-donated to the bridging sulfur-donor
ligand, the lower the concentration of catalytically active species in
experiments, depending upon the conditions,
through the head-to-head addition 1,2,4-(Ph) -3,5,6-(H)
4) was found in addition to 3 in product mixtures. These products
a
cyclotrimer
3
3
-benzene
(
1
were identified by H NMR spectra.
the first step. Since the Me(m-S)eM interaction is stronger in the Ta
To confirm the effects of temperature on the product, we
complexes than in the Nb ones, the concentration of the catalyti-
cally active species is lower in the Ta system than the Nb ones to
decrease the catalytic activity in the first step.
checked the reaction [Nb
2
Cl
6
(Me
2
S)
3
] (1a) in CHCl
3
or toluene at
ꢀ
lower and higher temperatures than room temperature (10e18 C).
The system generated 3 highly regioselectively in both the solvents
It is reported that Nb and Ta complexes [Nb
Cl (THT) ] (2b) gave different products from each other in the
reaction with 1,2-bis(diphenylphosphino)ethane (dppe). In those
systems [Nb Cl (THT) ] (1b) allows the bridging sulfur donor to
2 6 3
Cl (THT) ] (1b) and
ꢀ
ꢀ
at 0 C, while the regioselectivity disappears in CH
2
Cl
2
at 40 C. The
[Ta
2
6
3
higher temperature in toluene, the regioselectivity becomes
inverted, i.e. the predominant product was switched from 1,3,5-
isomer to 1,2,4-isomer. The same tendency was observed for
2
6
3
leave, and the original face-sharing bioctahedral structure changes
into an edge-sharing ones with two bridging chlorides
[
2 6 3
Nb Cl (THT) ] (1b). Thus, temperature is a more important factor
than the solvent for emerging the regioselectivity for 1,3,5-isomer.
[{NbCl
ctahedral structure remains in [Ta
inertness or stability of the bridging sulfur donor, even though two
[Ta Cl (THT) ] (2b) dimerize to give a tetranuclear complex [{Ta
Cl) -THT)Cl -dppe) ] [7p]. In the present work, we checked
the bond lengths of the complexes closely, and reveal that the Ta
complexes have a slightly shorter coordination bonds than Nb
complexes. This feature leads to the less “labile” Ta complexes than
the Nb ones.
2
(dppe)}
2
(
m
-Cl)
2
]. On the other hand, the face-sharing bio-
We carried out the experiments for checking the stability of
2
Cl (THT) ] (2b) due to the
6
3
ꢀ
[Nb
2
Cl
6
(Me
2
S)
3
] (1a) both at 40 C and room temperature in CD
2
Cl
2
,
and obtained results that the complex is decomposed faster at
higher temperature (Table S6). This lower stability at high tem-
2
6
3
2
(m-
2
(
m
4
}
2
(
m
2
ꢀ
perature may be responsible for the lower yield at 40 C than at
room temperature.
The second step, addition of the second alkyne to form a h2
-
butadiene moiety, is the rate-determining and the key step for the
regioselectivity. The carbonecarbon bond should be formed be-
tween incoming and coordinated alkynes in the head-to-tail fashion
to give 1,3,5-isomerat low temperature. On the otherhand, at higher
temperature, the insertion of the second alkyne into the NbeC bond
becomes possible, since the thermal motion provides space which
accommodates the incoming ligand, and leads to the binding to the
metal center. The reason for the inverted regioselectivity is still
uncertain, but the differences of thermal stabilities of isomeric form
in the product of the second step, dinuclear five-membered metal-
lacycles, may be responsible (Scheme 2, [15]).
We also tested reactions of [Ta
acetylene relevance to the solvent and temperature (Table 3). We
have found that [Ta Cl (Me S) ] (2a) maintains regioselectivity
2 6 2 3
Cl (Me S) ] (2a) with phenyl-
2
6
2
3
even if the solvent is altered or the temperature is elevated. For Nb
complexes the solvent and temperature affect the activity and
ꢀ
regioselectivity remarkably (see above). The conditions under 40 C
2 2 2 6 2 3
in CH Cl are necessary for Nb complex [Nb Cl (Me S) ] (1a) to give
1,3,5-benzene derivatives regioselectively, but these conditions are
not necessary for Ta complex 2a. This may arise from the robustness
2 6 2 3
of the precursor complex [Ta Cl (Me S) ] (2a), and the catalyst

M. Matsuura et al. / Journal of Organometallic Chemistry 745-746 (2013) 288e298
293
Table 2
Cyclotrimerization of phenylacetylene catalyzed by [Nb
2
6
Cl (Me
2
S)
3
] (1a) and [Nb
2
Cl
6
(THT)
3
] (1b) under various conditions.
Ph
Ph
1a and 1b
Ph
Ph
+
Ph
Ph
(200 equiv)
Ph
3
4
ꢀ
Isomer ratioa
Entry
Catalyst
Duration (h)
Solvent
T ( C)
Yield (%)
3
4
1b
2
3
4
5
6
7
8
9
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1b
1b
1b
1b
1a
1a
4
4
4
4
4
4
4
4
4
4
4
1
1
1
1
4
4
CH
Toluene
CHCl
CCl
Benzene
Cl
Cl
r.t.
r.t.
r.t.
r.t.
r.t.
r.t.
r.t.
40
40
60
70
r.t.
40
40
50
0
>99:1
48
51
60
44
87
86
51
36
34
33
>99:1
72
55
42
96
88
96
67
78
30
64
56
15
40
78
73
54
5
2
2
52
49
40
56
13
14
49
64
66
67
3
4
C
6
H
5
THF
CH Cl
2
2
Toluene
Toluene
Toluene
10
11
12
13
14
15
16
17
C
CH
6
H
4
Cl
Cl
2
2
2
28
45
58
4
14
31
42
6
Toluene
Toluene
Toluene
CHCl
3
0
12
33
a
From the intensity ratio of 1H NMR.
Ref. [15].
b
activating temperature range of Ta complexes becomes higher than
that of Nb ones.
The reaction of [Ta
fold equivalent in toluene-d
2 6 3
Cl (THTP) ] (2d) with phenylacetylene in 200-
1
13
8
was followed by H and C NMR
[
Ta
Cl
mixture, and 2d itself is gradually decomposed with the residual O
and H O in solution. In this study, we refer to the influences of the
metal and thioether ligands on the catalytic activity. Ta complexes
have lower catalytic activity than Nb ones, and [Nb Cl (THTP)
1d) has lowest catalytic activity among the Nb complexes and
scarcely gave cyclotrimerized products (see below). Therefore, it
seems reasonable that [Ta Cl (THTP) ] (2d) has no catalytic activity
in CH Cl (Table 3). Meanwhile, the catalytic activity of
Ta Cl (Me S) ] (2a) is more enhanced in toluene than in CH Cl
2
Cl
6
(THTP)
3
] (2d) is so inert toward phenylacetylene in
spectra for 4 h. Cyclotrimerized product 3 scarcely generated in this
reaction (3:phenylacetylene ¼ 1:200).
CH
2
2
that phenylacetylene is recovered from the reaction
2
The above mentioned experimental results may suggest that
2
the contributions of the m-S ligand and the MeM bonding are
more significant in Ta complexes than in Nb complexes. Eventu-
2
6
3
]
ally, the influence of the difference in metal center in
III
(
[{M Cl
2
(L)}
2
(
m
-Cl)
2
(
m
-L)] on the catalytic activity may be due to
the concentration of the catalytically active species produced in
the first step, which depends on the strength of the back-
donation from metalemetal bond to the bridging sulfur donor
in the precursor complexes.
2
6
3
2
2
[
2
6
2
3
2
2
.
Table 3
Cyclotrimerization of phenylacetylene catalyzed by [Ta
2
Cl
6
(Me
2
S)
3
] (2a), [Ta
Cl
2 6
(THT)
3
] (2b), and [Ta
2
6
Cl (THTP)
3
] (2d) under various conditions.
Ph
Ph
Ph
2a, 2b and 2d
+
Ph
Ph
Ph
(
200 equiv)
Ph
4
3
ꢀ
Isomer ratioa
Entry
Catalyst
Duration (h)
Solvent
T ( C)
Yield (%)
3
4
1
2
3
4
5
6
7
2a
2a
2a
2a
2a
2b
2d
4
4
4
4
4
CH
CH
Toluene
Toluene
Toluene
2
Cl
Cl
2
2
r.t.
40
r.t.
40
60
r.t.
r.t.
>99:1
76
95
96
92
8
77
42
44
50
15
0
2
24
5
4
8
1
46
CH
CH
2
Cl
Cl
2
2
>99:1
e
2
e
a
From the intensity ratio of 1H NMR peaks.

294
M. Matsuura et al. / Journal of Organometallic Chemistry 745-746 (2013) 288e298
2.6. Influence of the chalcogen donor and halide on the catalytic
ethylene sulfide, which does not form complexes but leads to
activity
decomposition.
In addition, we tried to synthesize the Nb complex with thio-
At first we discuss about the catalytic activities of Nb complexes
with sulfur donors. As shown in Table 4, [Nb Cl (Et S) ] (1c) acts as
a catalyst to give 3 regioselectively with lower activity than
phene having conjugate p-systems, but have not obtained any
proof of generation of complexes. This may be due to the low
donating ability of cyclic unsaturated thioether ligands.
2
6
2
3
[
Nb
2
Cl
6
(Me
2
S)
3
] (1a) and [Nb
2
Cl
6
(THT)
3
] (1b). The complex with
The trend which the catalytic activity depends on the size of thi-
oether ligands is applicable to the case of the complexes with sele-
THTP [Nb
2
Cl
6
(THTP)
3
] (1d) scarcely gave cyclotrimerized products,
although the regioselectivity is sustained (79:21). The low yields
0.5%) suggest that [Nb Cl (THTP) ] (1d) reacts with alkynes as a
reactant, not as a catalyst. Judging from the yield of the product,
noether ligands [Nb
have already determined the structure of the [Nb
analog [{NbCl (Me Se)} -Cl) -Me S)] which has dimethyl sele-
nide atthe terminalpositions[23]. The bond lengthsofthecomplex in
terminal positions are longer than [Nb Cl (Me S) ] (1a) probably due
to the difference in the ionic radius of Se (1.98 A) from that of S
2
Cl
6
(Me
2
Se)
3
] (1e) and [Nb
2
Cl
6
(THSe)
Cl (Me
6
3
] (1f). We
(
2
6
3
2
2
Se) ] (1e)
3
2
2
2
(m
2
(
m
2
[
[
Nb
Nb
2
Cl
Cl
6
(Me
2 3
S) ] (1a) has a highest catalytic activity among various
2
6
L
3
] for the cyclotrimerization of alkynes, and a sequence of
2
6
2
3
ꢀ
2
ꢁ
2ꢁ
,
the decreasing activity is,
Nb Cl (Me S) ] (1a) > [Nb
1c) > [Nb Cl (THTP) ] (1d). The smaller the thioether ligands, the
2
ꢁ
2ꢁ
[
2
6
2
3
2
Cl
6
(THT)
3
] (1b) > [Nb
2
Cl
6
(Et
2
S)
3
]
while the donating ability of Se is larger than that of S . We dis-
closed that the Nb complexes with selenoethers, [Nb Cl (Me Se)
(1e) and [Nb Cl (THSe) ] (1f), exhibit higher rates incatalytic reaction
than those with sulfur donors, but have the lower activities, for which
the higher sensitivity to air (e.g.; O) compared with
[Nb Cl (Me S) ] (1a) and [Nb Cl (THT) ] (1b) is responsible. There-
fore, [Nb Cl (Me S) ] (1a) with dimethyl sulfide having eventually
(
2
6
3
2
6
2
3
]
higher the catalytic activity.
2
6
3
For the complexes with acyclic saturated thioethers, the reac-
tivity is probably associated with the stability of the active species,
which is mainly governed by the bulkiness of the hydrocarbon
2 2
O , H
2
6
2
3
2
6
3
moiety. [Nb
compared with [Nb
2
Cl
6
(Et
2
S)
3
] (1c) with ethyl group is extremely unstable
(Me S) ] (1a) with methyl group. Other Nb
2
6
2
3
2
Cl
6
2
3
smallest size in our series of sulfur donors has highest activity.
For the effect of the halide on the catalytic reaction, it is sug-
complexes with di-t-butyl sulfide and with t-butyl methyl sulfide
have not been synthesized, after many trials to prepare those
complexes. On the other hand for the complexes with cyclic
saturated thioethers, the inertness of the complex to give active
species may be responsible for the low catalytic activity. Although
gested that the bromide complexes, [Nb
[Nb Br (THT) ](1h) and [Ta Br (Me S) (2g) give
lectively. However the catalytic activities of the complexes are
lower than the chloride ones, i.e. [Nb Cl (Me S) (1a),
[Nb Cl (THT) ] (1b) and [Ta Cl (Me S) ] (2a), respectively (Table 4).
It is quite reasonable that the deviation of the metal center from the
4
X plane is more significant for complexes with Br than ones with Cl
2
Br
6
(Me
2
S)
3
]
(1g),
2
6
3
2
6
2
3
]
3 regiose-
2
6
2
3
]
[
Nb
2
Cl
6
(THT)
3
] (1b) with a less bulky thioether has a higher cat-
Cl (THTP) ] (1d) with bulky ones, the
2
6
3
2
6
2
3
alytic activity than [Nb
2
6
3
former is not so extremely stable as the latter. The steric bulkines is
therefore not the determining factor but the electronic properties.
(see above), since the steric bulkiness of the Br is higher than Cl. The
enhanced effect of bromide on preventing alkynes from
approaching to the coordination vicinity compared with chloride
allows to the catalytic activities decrease in Nb complexes. There-
fore, steric factor plays a significant role in controlling the catalytic
behavior.
The donor ability of [Nb
than [Nb Cl (THTP) ] (1d), which leads to the dissociation of the
bridging thioether ligand bound to the complex with -back
2 6 3
Cl (THT) ] (1b) may be slightly higher
2
6
3
p
donation-based bonds. The exception is trimethylene sulfide. We
have tried to synthesize the complex having trimethylene sulfide
by the reaction of NbCl
solution indicating the generation of [Nb
5
with trimethylene sulfide, and a purple
Cl ] was obtained.
Furthermore, NBO analysis revealed the lower electron density
2
6
L
3
around the metal center in [Nb
2 6 2 3
Br (Me S) ] (1g) than in
However the complex is extremely unstable to decompose at
[Nb Cl (Me S) ] (1a) (see above). Thus the lability of precursor to
give catalytically active species decreases in complexes with Br, and
consequently the complex has a lower catalytic activity. The trend
2
6
2
3
ꢀ
higher temperature than ꢁ18 C even in solution. The reason for
the unstableness is speculated to be reactivity of trimethylene
sulfide with Nb center inside the complex, just as is seen for
is common for Nb and Ta complexes, probably because the Mem-S
Table 4
Cyclotrimerization of phenylacetylene catalyzed by [Nb
2
Cl
6
(Et
2
S)
3
]
2
(1c), [Nb Cl
6
(THTP)
3
]
(1d), [Nb
Cl
2 6
(Me
2
Se)
3
]
(1e), [Nb
2
Cl
6
3 2 6 2 3
(THSe) ] (1f), [Nb Br (Me S) ] (1g),
[
Nb Br (THT) ] (1h), and [Ta Br (Me S) ] (2g) in CH Cl at room temperature.
2
6
3
2
6
2
3
2
2
Ph
Ph
Ph
1c-1h and 2g
+
Ph
200 equiv)
Ph
Ph
CH Cl r.t.
2
2
(
Ph
4
3
Entry
Complex
Duration (h)
Isomer ratioa
Yield (%)
3
4
1
2
3
4
5
6
7
1c
1d
1e
1f
1g
1h
2g
1
1
1
0.17
4
>99:1
79
>99:1
>99:1
91
10
0.5
20
15
16
11
2
21
9
10
3
1
4
90
97
a
Determined by 1H NMR.

M. Matsuura et al. / Journal of Organometallic Chemistry 745-746 (2013) 288e298
295
bond lengths for Nb are very similar with those of Ta for each of
complexes with Cl or Br.
4. Experimental
.1. General
4
2
.7. Efficiency as a catalyst
All the reactions were carried out under a dry argon atmosphere
by using standard Schlenk tube techniques. All the solvents were
dehydrated and purified by distillation: Et O and toluene were
refluxed over sodium benzophenone ketyl and distilled, CH Cl
CHCl and n-hexane were refluxed over CaH and distilled. These
We have calculated overall turnover frequencies for overall
2
cyclotrimerization products (TOF) and turnover numbers (TON),
which are collected in Tables S15 and S16, respectively. The ef-
2
2
,
3
2
ficiency as
Nb Cl (Me
a
catalyst is discussed based on those data.
] (1a) in CH Cl at room temperature shows the
purified solvents were stored under a dry argon atmosphere. Other
reagents employed were in the reagent grade and used as received
without further purification.
[
2
6
2
S)
3
2
2
most highest TON. The most effective factor on TON is the alkyl
chain in thioether ligands than other factors, such as metal,
halide and chalcogen. On the other hand, five-membered cyclic
sulfur donor ligands have an advantage for high TOF in
both Nb and Ta complexes. It seems that TOF mainly depends
both on the formation constant of catalytically active species in
the first step, which is closely related to the stability of the
precursor complex, and the rate of conversion into the five-
membered metallacycle species. Furthermore, the decrease in
the stability of the active species makes TON lower due to the
retardation in the yield of recovery of active species after the
catalytic cycle.
1
13
H and C NMR spectra were recorded on a Bruker DPX-400,
1
13
DRX-400, AVANCE500 NMR spectrometer. For H and C NMR
measurements, Me Si was used an internal standard (
Elemental analyses were carried out using FISONS EA 1112
at the Comprehensive Analysis Center for Science, Saitama
4
d
¼ 0).
University. Crystals of [Nb
Nb Cl (THTP) ] (1d) (CCDC 923585), [Nb
23588), [Nb Br (THT) ] (1h) (CCDC 923587), [Ta
CCDC 923582), [Ta Cl (THTP) (2d) (CCDC 923583), and
2
Cl
6
(Et
2
S)
3
]
(1c) (CCDC 923586),
Br (Me S) ] (1g) (CCDC
Cl (Me S) ] (2a)
[
2
6
3
2
6
2
3
9
2
6
3
2
6
2
3
(
[
2
6
3
]
Ta
2
Br
6
(Me
2
S)
3
] (2g) (CCDC 923584) were grown in CH
14) at room temperature or at ꢁ18 C and submitted for X-
2
Cl
2
ehex-
ꢀ
ane (C
6
H
Among Nb complexes, TON is higher for that with dimethyl
chalcogenides than for those with five-membered cyclic chalcogen
donors, while the TOF exhibits a reverse relationship. These phe-
nomena can be understood by the stabilities of active species and
ray crystallography measurements. The X-ray diffraction data for
all the compounds were collected on Bruker SMART APEX
diffractometer and BRUKER APEX II Ultra diffractometer equipped
with CCD area detector. Crystallographic data and structure
precursor, respectively. For example, [Nb
chalcogen donor Et S shows 1/3 of TOF and 1/12 of TON compared
with those of [Nb Cl (Me S) ] (1a) with Me S. This means that
Nb Cl (Et S) ] (1c) is more stable by three times in precursor and
less stable by 36 times in the active species than those of
Nb Cl (Me S) ] (1a). On the other hand, [Nb Cl (THT) ] (1b) with
THT shows 9 times of TOF and 9 times of TON compared with those
of [Nb Cl (Et S) ] (1c) with Et S. Cyclic chalcogen donor makes
both TOF and TON higher.
2 6 2 3
Cl (Et S) ] (1c) with
refinement for [Nb
of others are shown in Tables S1 and S2, respectively. The frame
data were acquired with the SMART-W2K/NT software using Mo K
2 6 2 3
Cl (Et S) ] (1c) are shown in Table 5, and those
2
2
6
2
3
2
a
[
2
6
2
3
ꢀ
radiation (
l
¼ 0.71073 A) from a fine-focus tube. The observed
frames were processed using the SAINT-WK2/NT software to give
the hkl data set. The crystals used for the diffraction measure-
ments show no decomposition during the data collection. Ab-
sorption correction was performed using the SADABS program
[
2
6
2
3
2
6
3
2
6
2
3
2
[
24]. The structures are solved by the direct method [25] or Pat-
2
terson method [26] and refined by least-squares method on F
using SHELEx-97 incorporated in SHELxTL-NT program [27]. All
non-hydrogen atoms for all the complexes were refined aniso-
tropically. Hydrogen atoms were included and refined using a
riding model for all the complexes.
3
. Conclusion
The structures of Nb and Ta complexes [Nb
Nb Cl (THTP) ] (1d), [Nb Br (Me S) ] (1g), [Ta Cl
and [Ta Br (Me S) ] (2g) were determined by X-ray crystallog-
2
Cl
6
(Et
2
S)
3
] (1c),
[
2
6
3
2
6
2
3
2
6
(THTP)
3
] (2d),
2
6
2
3
III
III
raphy. A dinuclear structure with a M eM double bond is com-
mon for all the complexes.
Table 5
In the reaction with phenylacetylene to give the 1,3,5-
Crystallographic data of [Nb
2 6 2 3
Cl (Et S) ] (1c).
cyclotrimerized product, [Nb
2 6 2 3
Cl (Me S) ] (1a) exhibits the
highest performance both in the catalytic activity and in the
Compound
1c
regioselectivity. The optimal conditions are: [Nb
2
Cl
6
(Me
2
S)
3
] (1a)
Empirical formula
Formula weight
Temperature (K)
Crystal system
Space group
12 6 2 3
C H30Cl Nb S
669.06
150(2)
ꢀ
in CH
2
Cl
2
at room temperature (<40 C). The dinuclear frame-
work of the complex with chalcogen donors and halides seems
essential to give rise to the regioselectivity for the 1,3,5-trimer,
while the activity varies with each of the central metal and
chalcogen donors and halides. On the other hand, both activity
and regioselectivity depends strongly upon the external condi-
tions such as the temperature and the solvent. This behavior is
seen in the following observations. The difference in the central
metal gives little influence on the regioselectivity, but significant
influence on the activity: the regioselective catalysis is observed
Tetragonal
P 4
1 1
2 2
ꢀ
a (A)
8.3794(6)
8.3794(6)
35.899(4)
90
90
90
2520.6(4)
4
1.763
b ( ꢀA )
ꢀ
c (A)
ꢀ
a
b
g
( )
ꢀ
( )
ꢀ
( )
ꢀ
3
V (A )
Z
Dcalcd. (Mg/m3)
Crystal size (mm)
ꢀ
ꢀ
below 40 C for Nb complexes, while even over 40 C for Ta
complexes. Furthermore, the difference in chalcogen donor and
halide affects strongly on the activity, but weakly on the
regioselectivity.
0.12 ꢂ 0.06 ꢂ 0.06
ꢁ
10 ꢃ h ꢃ 9
Index ranges
ꢁ10 ꢃ k ꢃ 10
ꢁ45 ꢃ l ꢃ 40
1.036
Goodness-of-fit on F2
Final R indices
Trials for isolation of the reaction intermediate in each step, and
the catalytic reaction with other alkynes are in progress for the
purpose of clarifying the mechanism.
R
wR
1
¼ 0.0632
[I > 2
s(I)]
2
¼ 0.1185

296
M. Matsuura et al. / Journal of Organometallic Chemistry 745-746 (2013) 288e298
4
4
.2. Syntheses of [{MX
2
(L)}
2
(
m
-X)
2
(
m
-L)] (1ae2g)
terminal-C
bridge-C
10S). Anal. Found (calcd. for C15
5
H
H
10S), 26.9 (4C, C
b
, terminal-C
10S), 23.4 (1C, C
Nb ): C, 25.27 (25.55);
5
H
10S), 26.6 (2C, C
b
,
5
10S), 25.2 (2C, C , terminal-C
g
5
H
g
, bridge-
.2.1. Preparation of [{NbCl
The complex was prepared by the method proposed in the
literature [15]. Diniobium decachloride Nb Cl10 (2.3 g, 4.3 mmol),
Mg (0.57 g, 23.4 mmol) were suspended in CH Cl (76 mL), then
dimethyl sulfide (C S, Me S, 1.7 mL, 23 mmol) was add to the
mixture, and finally Et O (12 mL) at room temperature. The mixture
2
(Me S)}
2
2
(m
-Cl) -Me
2
(
m
2
S)] (1a)
C
5
H
H
30Cl
6
2 3
S
H, 4.24 (4.29).
2
2
2
4.2.5. Synthesis of [{NbCl
2
(Me
2
Se)}
Cl10 (0.5 g, 0.93 mmol), Mg (0.11 g,
4.6 mmol) were suspended in CH Cl (75 mL), then dimethyl
selenide (C Se, Me Se, 0.35 mL, 4.6 mmol) was add to the
mixture, and finally Et O (11.5 mL) at room temperature. The
2 2 2
(m-Cl) (m-Me Se)] (1e)
H
2 6
2
Diniobium decachloride Nb
2
2
2
2
was stirred for 4 days at room temperature. The suspended solid
formed the precipitation, which was removed by filtration, and the
resultant filtrate was concentrated to dryness leaving purple
powder. The crude product was washed with hexane and dried
under reduced pressure (2.2 g, yield 88%). H NMR (CDCl
2
H
6
2
2
mixture was stirred for 4 h at room temperature. The suspended
solid caused the precipitation, which was removed by filtration,
and the resultant filtrate was concentrated to dryness leaving
purple powder. The crude product was washed with hexane and
dried under reduced pressure (0.6 g, yield 92%). H NMR (CDCl
1
3
, 300 K):
13
d
3.33 (6H, bridge-Me
300 K, CDCl ): 30.0 (2C, bridge-Me
Anal. Found (calcd. for C Nb
2
S), 2.63 (12H, terminal-Me
S), 22.7 (4C, terminal-Me
): C, 12.33 (12.32); H, 3.00
2
S). C NMR
1
(
3
d
2
2
S).
3
,
C
13
6
H
18Cl
6
2
S
3
300 K):
NMR (300 K, CDCl
Me Se).
d
2.76 (6H, bridge-Me
2
Se), 1.83 (12H, terminal-Me
2
Se).
(
3.10).
3
): 31.6 (4C, terminal-Me
d
2
Se), 22.7 (2C, bridge-
2
4
.2.2. Preparation of [{NbCl
Diniobium decachloride Nb
.6 mmol) were suspended in CH
S, THT, thiolane, 0.75 mL, 8.5 mmol) was add to
the mixture, and finally Et O (6 mL) at room temperature. The
2
(THT)}
Cl10 (1.2 g, 2.2 mmol), Mg (0.16 g,
Cl (37 mL), then tetrahy-
2 2
(m-Cl) (m-THT)] (1b)
2
4.2.6. Synthesis of [{NbCl
Diniobium decachloride Nb
13.3 mmol) were suspended in CH
droselenophene (C Se, THSe, selenolane, 0.72 g, 1.3 mmol) was
add to the mixture, and finally Et O (11.5 mL) at room temperature.
2
(THSe)}
Cl10 (0.7 g, 1.3 mmol), Mg (0.32 g,
Cl (75 mL), then tetrahy-
2 2
(m-Cl) (m-THSe)] (1f)
6
2
2
2
drothiophene (C
H
4 8
2
2
2
4 8
H
mixture was stirred for 4 days at room temperature. The suspended
solid formed the precipitation, which was removed by filtration,
and the resultant filtrate was concentrated to dryness leaving
purple powder. The crude product was washed with hexane and
2
The mixture was stirred for 4 h at room temperature. The sus-
pended solid caused the precipitation, which was removed by
filtration, and the resultant filtrate was concentrated to dryness
leaving purple powder. The crude product was washed with hexane
1
dried under reduced pressure (0.9 g, yield 64%). H NMR (300 K,
1
CDCl
.31 (4H, H
300 K, CDCl
S), 30.4 (4C, C
Anal. Found (calcd. for C12
3.65).
3
):
d
4.03 (4H, H
, bridge-C
): 38.0 (4C, C
, terminal-C
a
, bridge-C
S), 2.15 (8H, H
, bridge-C
S), 27.9 (2C, C
Nb ): C, 21.75 (21.74); H, 3.55
4
H
8
S), 3.41 (8H, H
a
, terminal-C
4
H
8
S),
S). C NMR
, terminal-
, bridge-C S).
and dried under reduced pressure (0.7 g, yield 63%). H NMR (300 K,
13
2
(
C
b
4
H
8
b
, terminal-C
4
H
8
CDCl
3
):
d
3.45 (4H, H
a
, bridge-C
, bridge-C
Se). C NMR (300 K, CDCl ):
, terminal-C Se), 31.9 (2C, C
, terminal-C Se).
4
H
8
Se), 2.87 (8H, H
Se), 1.25 (8H, H
44.6 (2C, C , bridge-C
, bridge-C Se), 31.3
a
, terminal-
, terminal-
Se),
3
d
a
4
H
8
S), 35.7 (2C, C
a
C
C
4
H
H
8
Se), 2.03 (4H, H
b
4
H
8
b
1
3
H
4 8
b
4
H
8
b
4
H
8
4
8
3
d
a
4 8
H
H24Cl
6
S
2 3
32.8 (4C, C
(4C, C
a
4
H
8
b
4 8
H
(
b
4 8
H
4
.2.3. Synthesis of [{NbCl
Diniobium decachloride Nb
.6 mmol) were suspended in CH
10S, Et S, 0.5 mL, 4.6 mmol) was add to the mixture, and finally
O (11.5 mL) at room temperature. The mixture was stirred for 4 h
2
(Et
2
S)}
Cl10 (0.5 g, 0.93 mmol), Mg (0.11 g,
Cl (75 mL), then diethyl sulfide
2
(
m
-Cl)
2
(
m
-Et
2
S)] (1c)
4.2.7. Synthesis of [{NbBr
Diniobium decabromide Nb
7.40 mmol) were suspended in CH
10 mmol) was add to the mixture, and finally Et
2
temperature. The mixture was stirred for 3 days at room temper-
ature. The suspended solid caused the precipitation, which was
removed by filtration, and the resultant filtrate was concentrated to
dryness leaving violet powder. The crude product was washed with
2
(Me
2
S)}
Br10 (2.4 g, 2.4 mmol), Mg (0.18 g,
Cl (72 mL), then Me S (0.75 mL,
O (12 mL) at room
2 2 2
(m-Br) (m-Me S)] (1g)
2
2
4
2
2
2
2
2
(
C
4
H
2
Et
2
at room temperature. The suspended solid formed the precipita-
tion, which was removed by filtration, and the resultant filtrate was
concentrated to dryness leaving purple powder. The crude product
was washed with hexane and dried under reduced pressure (0.6 g,
yield 88%).
1
hexane and dried under reduced pressure (1.0 g, yield 51%). H NMR
(CDCl
3
, 300 K):
C NMR (300 K, CDCl
nal-Me S). Anal. Found (calcd. for C
d
3.60 (6H, bridge-Me
): 35.9 (2C, bridge-Me
2 3
Nb S
2
S), 2.77 (12H, terminal-Me
S), 24.8 (4C, termi-
): C, 8.07 (8.46); H,
2
S).
1H NMR (300 K, CDCl
terminal-Et S), 1.72 (6H, H
S). C NMR (300 K, CDCl
, terminal-Et S), 13.8 (4C, C
S).
):
d
3.80 (4H, H
, bridge-Et S), 1.45 (12H, H
): 36.4 (2C, C , bridge-Et S), 29.9 (4C,
, terminal-Et S), 11.8 (2C, C , bridge-
, bridge-Et
S), 3.10 (8H, H
,
13
d
3
a
2
a
3
2
2
b
2
b
, terminal-
2
6
H
18Br
6
13
Et
C
Et
2
3
d
a
2
2.16 (2.13).
a
2
b
2
b
2
4.2.8. Preparation of [{NbBr
2
(THT)}
2
(
m
-Br)
Br10 (2.4 g, 2.4 mmol), Mg
(0.18 g, 7.3 mmol) were suspended in CH Cl (72 mL), then THT
(0.7 mL, 7.9 mmol) was add to the mixture, and finally Et
2
(m-THT)] (1h)
Diniobium decabromide Nb
2
4
.2.4. Synthesis of [{NbCl
Diniobium decachloride Nb
2
(THTP)}
Cl10 (2.3 g, 4.3 mmol), Mg (0.52 g,
Cl (75 mL), then tetrahy-
10S, THTP, thiane, 1.4 mL, 14 mmol) was add to
O (12 mL) at room temperature. The
2
(
m
-Cl)
2
(
m
-THTP)] (1d)
2
2
2
2
O
21 mmol) were suspended in CH
2
2
(12 mL) at room temperature. The mixture was stirred for 2 days
at room temperature. The suspended solid formed the precipi-
tation, which was removed by filtration, and the resultant
filtrate was concentrated to dryness leaving violet powder. The
5
drothiopyran (C H
the mixture, and finally Et
mixture was stirred for 2 days at room temperature. The suspended
solid formed the precipitation, which was removed by filtration,
and the resultant filtrate was concentrated to dryness leaving
purple powder. The crude product was washed with hexane and
2
crude product was washed with hexane and dried under
1
reduced pressure (1.6 g, yield 69%). H NMR (300 K, CDCl
3
):
d
4.29 (4H, H
(4H, H , bridge-C
(300 K, CDCl ):
minal-C S), 29.5 (4C, C
S). Anal. Found (calcd. for C12
H, 2.83 (2.60).
a
, bridge-C
S), 2.13 (8H, H
39.5 (2C, C , bridge-C
, terminal-C
4
H
8
S), 3.47 (8H, H
, terminal-C
S), 39.1 (4C, C
S), 26.8 (2C, C , bridge-
Nb ): C, 15.90 (15.50);
a
, terminal-C
S). C NMR
, ter-
4 8
H S), 2.34
1
13
dried under reduced pressure (2.5 g, yield 84%). H NMR (300 K,
CDCl
b
4
d
H
8
b
4 8
H
3
):
10S), 2.48 (4H, H
10S),1.69 (4H, H , terminal-C
): 44.1 (2C, C
d
3.79 (4H, H
a
, bridge-C
5
H
H
10S), 3.14 (8H, H
10S), 2.00 (8H, H
10S),1.51 (2H, H
, bridge-C
a
, terminal-
, terminal-
3
a
4
H
8
a
C
C
5
H
H
b
, bridge-C
5
b
4
H
8
b
4
H
8
b
5
g
5
H
g
, bridge-C
5
H
10S).
C
4
H
8
H
24Br
6
2 3
S
1
3
C NMR (300 K, CDCl
3
d
a
5
H
10S), 34.8 (4C, C ,
a

M. Matsuura et al. / Journal of Organometallic Chemistry 745-746 (2013) 288e298
297
4
.2.9. Preparation of [{TaCl
Ditantalum decachloride Ta
2 mmol) were suspended in toluene (75 mL), then Me
4 mmol) was add to the mixture, and finally Et O (10 mL) at room
2
(Me
2
S)}
Cl10 (2.4 g, 3.4 mmol), Mg (0.28 g,
S (1.0 mL,
2
(
m
-Cl)
2
(
m
-Me
2
S)] (2a)
of a precipitate. We calculate the completion of the reaction
determined by the comparison of literature data [15]. The solution
was removed by filtration and the resultant filtrate was concen-
trated to dryness. The crude product was washed with hexane and
dried under reduced pressure. The results are collected in Tables 2e
4, respectively.
2
1
1
2
2
temperature. The mixture was stirred for 2 days at room temper-
ature. The suspended solid formed the precipitation, which was
removed by filtration, and the resultant filtrate was concentrated to
dryness leaving yellow brown powder. The crude product was
washed with hexane and dried under reduced pressure. (1.6 g, yield
6
terminal-Me
2
9
Triphenylbenzene, 3 and 4: Identification and regioisomeric
ratio of 1,3,5- and 1,2,4-triphenylbezene were carefully determined
1
by comparison of literature data [15] as well as by checking the H
1
0%). H NMR (CDCl
3
, 300 K):
S). C NMR (300 K, CDCl
S). Anal. Found (calcd. for C
d
3.40 (6H, bridge-Me
): 41.0 (2C, bridge-Me
Ta ): C,
2
S), 2.66 (12H,
NMR using the signal at
d
¼ 7.93.
13
2
3
d
2
S),
3.1 (4C, terminal-Me
2
6
H
18Cl
6
2 3
S
4.4. Analysis data. Compounds 3 and 4
.47 (9.47); H, 2.38 (2.28).
1
Compound 3: H NMR (Acetone-d
6
)
d
d
7.93 (s, 3H), 7.88 (d, 6H),
4
.2.10. Preparation of [{TaCl
2
(THT)}
Cl10 (2.4 g, 3.4 mmol), Mg (0.28 g,
2 mmol) were suspended in toluene (75 mL), then THT (1.2 mL,
4 mmol) was add to the mixture, and finally Et O (10 mL) at room
2 2
(m-Cl) (m-THT)] (2b)
7
.52 (t, 6H), 7.42 (tt, 3H).
1
Ditantalum decachloride Ta
2
Compound 4: H NMR (Acetone-d
6
)
7.18e7.78 (m, 18H).
1
1
2
Acknowledgments
temperature. The mixture was stirred for 2 days at room temper-
ature. The suspended solid formed the precipitation, which was
removed by filtration, and the resultant filtrate was concentrated to
dryness leaving reddish brown powder. The crude product was
washed with hexane and dried under reduced pressure (1.6 g, yield
5
This work has been supported by the programs of the Grants-in-
Aid for Scientific Research (to TF, No. 23510115) from the Japan
Society for the Promotion of Science.
1
6%). H NMR (CDCl
3
, 300 K):
S), 2.51 (4H, H
S). C NMR (300 K, CDCl
S), 38.4 (4C, C , terminal-C S), 30.5 (4C, C
7.8 (2C, C , bridge-C S). Anal. Found (calcd. for C12
d
3.77 (4H, H
, bridge-C
): 42.0 (2C, C
4 8
, terminal-C H
a
, bridge-C
S), 2.09 (8H, H
, bridge-
S),
):
4 8
H S), 3.44
Appendix A. Supplementary data
(
8H, H
a
, terminal-C
4
H
8
b
4
H
8
b
,
1
3
terminal-C
C
2
4
H
8
3
d
a
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.jorganchem.2013.07.035.
H
4 8
a
4
H
8
b
b
4
H
8
H24Cl
6
Ta
2
S
3
C, 16.74 (17.18); H, 2.73 (2.88).
References
4
.2.11. Synthesis of [{TaCl
Ditantalum decachloride Ta
2 mmol) were suspended in toluene (75 mL), then THTP (1.4 mL,
4 mmol) was add to the mixture, and finally Et O (10 mL) at room
2
(THTP)}
2
(
m
-Cl)
2
(
m
-THTP)] (2d)
[
1] For recent reviews, see: (a) S. Li, L. Zhou, K.-I. Kanno, T. Takahashi,
2
Cl10 (2.4 g, 3.4 mmol), Mg (0.28 g,
J. Heterocycl. Chem. 48 (2011) 517e528;
(b) G. Dominguez, J. Perez-Castells, Chem. Soc. Rev. 40 (2011) 3430e3444;
1
1
(c) R. Hua, M.V.A. Abrenica, P. Wang, Curr. Org. Chem. 15 (2011) 712e729;
(d) T. Takahashi, S. Li, K. Kanno, Heterocycles 80 (2010) 725e738;
(e) K. Tanaka, Chem. Asian J. 4 (2009) 508e518.
2
temperature. The mixture was stirred for 4 days at room temperature.
The suspended solid formed the precipitation, which wasremoved by
filtration, and the resultant filtrate was concentrated to dryness
leaving reddish brown powder. The crude product was washed with
[2] (a) N. Morohashi, K. Yokomakura, T. Hattori, S. Miyano, Tetrahedron Lett. 47
(
(
2006) 1157e1161;
b) F.T. Ladipo, V. Sarveswaran, J.V. Kingston, R.A. Huyck, S.Y. Bylikin, S.D. Carr,
R. Watts, S. Parkin, J. Organomet. Chem. 689 (2004) 502e514;
(c) O.V. Ozerov, B.O. Patrick, F.T. Ladipo, J. Am. Chem. Soc. 122 (2000)
1
hexane and dried under reduced pressure (1.6 g, yield 53%). H NMR
6423e6431;
(
C
1
d
300 K, CDCl
10S), 2.76 (4H, H
.62 (6H, H , bridge- and terminal-C
55.3 (2C, C , bridge-C 10S), 33.8 (4C, C
, terminal-C 10S), 25.7 (2C, C , bridge-C
10S), 23.2 (1C, C , bridge-C 10S). Anal. Found (calcd. for
Ta ): C, 20.50 (20.45); H, 3.32 (3.43).
3
):
d
3.74 (4H, H
a
, bridge-C
5
H
10S), 3.12 (8H, H
, terminal-C
10S). C NMR (300 K, CDCl
, terminal-C 10S), 27.1 (4C,
10S), 24.2 (2C, C , ter-
a
, terminal-
10S),
):
(
(
d) O.V. Ozerov, F.T. Ladipo, B.O. Patrick, R. Ph, H. Me, J. Am. Chem. Soc. 121
1999) 7941e7942;
5
H
b
, bridge-C
5
H10S), 1.91 (8H, H
b
5
H
1
3
g
5
H
3
(e) T. Takahashi, C.-F. Hsi, Jpn. Kokai Tokkyo Koho (2002), 2002-265393.
[3] (a) T. Takahashi, F.-Y. Tsai, Y. Li, K. Nakajima, M. Kotora, J. Am. Chem. Soc.
a
5
H
a
5
H
1
21 (1999) 11093e11100;
C
b
5
H
b
5
H
g
(
(
b) T. Takahashi, Jpn. Kokai Tokkyo Koho (2003), 2003-261469;
c) F. Calderazzo, G. Pampaloni, P. Pallavicini, J. Straehle, K. Wurst, Organ-
minal-C
C
5
H
g
5
H
15
H
30Cl
6
S
2 3
ometallics 10 (1991) 896e901;
d) der L. van Arjan, C.J. Schaverien, N. Meijboom, C. Ganter, A.G. Orpen,
J. Am. Chem. Soc. 117 (1995) 3008e3021;
e) Y. Wielstra, S. Gambarotta, A. Meetsma, B. De Jan L., M.Y. Chiang, Or-
(
4
.2.12. Synthesis of [{TaBr
Ditantalum decabromide Ta
.2 mmol) were suspended in toluene (75 mL), then Me
.9 mmol) was add to the mixture, and finally Et O (5 mL) at room
2
(Me
2
S)}
Br10 (2.0 g, 1.7 mmol), Mg (0.12 g,
S (0.5 mL,
2 2 2
(m-Br) (m-Me S)] (2g)
(
2
ganometallics 8 (1989) 2696e2702.
[
4] (a) P.M. Boorman, M. Wang, M. Parvez, J. Chem. Soc. Dalton Trans. Inorg.
Chem. (1996) 4533e4542;
5
6
2
2
b) I. Czelu ꢁs niak, P. Koci e˛ cka, T. Szyma nꢁ ska-Buzar, J. Organomet. Chem. 716
(2012) 70e78;
(
temperature. The mixture was stirred for 2 days at room temperature.
The suspended solid formed the precipitation, which wasremoved by
filtration, and the resultant filtrate was concentrated to dryness
leaving red powder. The crude product was washed with hexane and
dried under reduced pressure (1.3 g, yield 75%). H NMR (CDCl
(
(
(
c) K. Sekiya, K. Hakkaku, Jpn. Kokai Tokkyo Koho (2005), 2005-112777;
d) R. Matusiak, A. Keller, J. Mol. Catal. A Chem. 195 (2003) 29e35;
e) G.A. Ardizzoia, S. Brenna, G. LaMonica, A. Maspero, N. Masciocchi,
J. Organomet. Chem. 649 (2002) 173e180;
(f) R. Hara, Q. Guo, T. Takahashi, Chem. Lett. (2000) 140e141;
(g) R.D. Adams, J.E. Babin, M. Tasi, J.G. Wang, Organometallics 7 (1988) 755e
1
3
,
13
3
00 K):
300 K, CDCl
Anal. Found (calcd. for C
d
3.75 (6H, bridge-Me
): 46.9 (2C, bridge-Me
Ta ): C, 6.99 (7.01); H, 1.77 (1.77).
2
S), 2.88 (12H, terminal-Me
2
S). C NMR
764.
(
3
d
2
S), 25.1 (4C, terminal-Me
2
S).
[5] J.B. Hartung Jr., S.F. Pedersen, Organometallics 9 (1990) 1414e1417.
[6] T. Oshiki, K. Takai, Jpn. Kokai Tokkyo Koho (2002), 2002-60354.
[
6
H
18Br
6
S
2 3
7] [Nb and Ta complexes with double bond] (a) F.A. Cotton, W.J. Roth, Inorg.
Chem. 22 (1983) 3654e3656;
4
.3. Catalytic cyclotrimerization of phenylacetylene with complexes
(
(
(
(
(
b) J.A.M. Canich, F.A. Cotton, Inorg. Chem. 26 (1987) 4236e4240;
c) F.A. Cotton, M.P. Diebold, M. Matusz, W.J. Roth, Inorg. Chim. Acta 112
1986) 147e152;
d) F.A. Cotton, M.P. Diebold, W.J. Roth, Inorg. Chem. 26 (1987) 3323e3327;
e) F.A. Cotton, X. Feng, P. Guetlich, T. Kohlhaas, J. Lu, M. Shang, Inorg. Chem.
Phenylacetylene in 200-fold equivalent was added to each
complexes 1ae2g (50 mg) in any solvent (20 mL), and the mixture
was stirred at any temperature. The solution caused the formation
33 (1994) 3055e3063;

298
M. Matsuura et al. / Journal of Organometallic Chemistry 745-746 (2013) 288e298
(
(
(
4
(
4
(
8
f) F.A. Cotton, M.P. Diebold, W.J. Roth, Inorg. Chem. 24 (1985) 3509e3510;
g) F.A. Cotton, J.H. Matonic, C.A. Murillo, J. Cluster Sci. 7 (1996) 655e662;
h) F.A. Cotton, S.A. Duraj, L.L. Falvello, W.J. Roth, Inorg. Chem. 24 (1985)
389e4393;
i) E. Babaian-Kibala, F.A. Cotton, P.A. Kibala, Inorg. Chem. 29 (1990) 4002e
005;
j) F.A. Cotton, L.M. Daniels, C.A. Murillo, X. Wang, Inorg. Chem. 36 (1997)
96e901;
3531;
(e) F.A. Cotton, M.P. Diebold, W.J. Roth, Inorg. Chem. 27 (1988) 2347e2352;
(f) J.L. Seela, J.C. Huffman, G. Christou, J. Am. Chem. Soc. (1987) 1258e1260.
[10] F.A. Cotton, W.J. Roth, Inorg. Chim. Acta 85 (1984) 17e21.
[11] F.A. Cotton, W.T. Hall, Inorg. Chem. 20 (1981) 1285e1287.
[12] F.A. Cotton, W.T. Hall, K.J. Cann, F.J. Karol, Macromolecules 14 (1981)
233e236.
[13] F.A. Cotton, W.T. Hall, Inorg. Chem. 19 (1980) 2354e2356.
[14] F.A. Cotton, W.T. Hall, Inorg. Chem. 19 (1980) 2352e2354.
[15] M. Kakeya, T. Fujihara, T. Kasaya, A. Nagasawa, Organometallics 2 (2006)
4131e4137.
[16] (a) M. Tsunoda, L.G. Hubert-Pfalzgraf, Inorg. Synth. 21 (1982) 16e18;
(b) J.L. Templeton, R.E. McCarley, Inorg. Chem. 17 (1978) 2293e2299;
(c) L.G. Hubert-Pfalzgraf, M. Tsunoda, J.G. Riess, Inorg. Chim. Acta 41 (1980)
283e286.
[17] (a) E.T. Maas Jr., R.E. McCarley, Inorg. Chem. 12 (1973) 1096e1101;
(b) A.D. Allen, S. Naito, Can. J. Chem. 54 (1976) 2948e2952.
[18] F.A. Cotton, R.C. Najjar, Inorg. Chem. 20 (1981) 2716e2719.
[19] J.L. Templeton, W.C. Dorman, J.C. Clardy, R.E. McCarley, Inorg. Chem. 17
(1978) 1263e1267.
(
(
(
(
k) F.A. Cotton, M.P. Diebold, W.J. Roth, Inorg. Chim. Acta 149 (1988) 105e110;
l) F.A. Cotton, L.R. Falvello, R.C. Najjar, Inorg. Chem. 22 (1983) 375e377;
m) F.A. Cotton, W.J. Roth, Inorg. Chem. 22 (1983) 868e870;
n) F.A. Cotton, M.P. Diebold, S.A. Duraj, W.J. Roth, Polyhedron 4 (1985) 1479e
1
484;
(o) F.A. Cotton, L.R. Falvello, R.C. Najjar, Inorg. Chim. Acta 63 (1982) 107e111;
(p) P.F. Gilletti, V.G. Young, T.M. Brown, Inorg. Chem. 28 (1989) 4034e4036;
(q)M. Etienne, J.-C. Hierso, P.J. Daff, B. Donnadieu, F. Dahan, Polyhedron23 (2004)
3
(
1
(
79e383;
r) E. Babaian-Kibala, F.A. Cotton, M. Shang, Acta Crystallogr. C47 (1991) 1617e
621;
s) F.A. Cotton, S.A. Duraj, W.J. Roth, Acta Crystallogr. C41 (1985) 878e881.
[
8] [Dinuclear Nb and Ta complexes without double bond] (a) L.G. Hubert-
[20] M. Kakeya, T. Fujihara, A. Nagasawa, Acta Crystallogr. E62 (2006) m553e
m554.
Pfalzgraf, G. Aharonian, Inorg. Chim. Acta 100 (1985) L21eL22;
(
b) A.R. Chakravarty, F.A. Cotton, M.P. Diebold, D.B. Lewis, W.J. Roth, J. Am.
Chem. Soc. 108 (1986) 971e976;
c) F.A. Cotton, L.M. Daniels, J.H. Matonic, X. Wang, C.A. Murillo, Polyhedron
6 (1997) 1177e1191;
d) F.A. Cotton, S.A. Duraj, W.J. Roth, J. Am. Chem. Soc. 106 (1984) 6987e
993;
e) J.A.M. Canich, F.A. Cotton, Inorg. Chem. 26 (1987) 3473e3478;
f) F.A. Cotton, W.J. Roth, J. Am. Chem. Soc. 105 (1983) 3734e3735;
g) J.A.M. Canich, F.A. Cotton, L.R. Falvello, S.A. Duraj, Inorg. Chim. Acta 143
1988) 185e192;
[21] (a) M. Kakeya, T. Fujihara, A. Nagasawa, Acta Crystallogr. E60 (2004) m893e
m894;
(
1
(
6
(
(
(
(
(
(b) F.A. Cotton, M. Shang, Inorg. Chem. 29 (1990) 508e514;
(c) P.N. Riley, J.R. Clark, P.E. Fanwick, I.P. Rothwell, Inorg. Chim. Acta 288
(1999) 35e39;
(d) H. Kawaguchi, T. Matsuo, Angew. Chem. Int. Ed. 41 (2002) 2792e2794;
(e) la M. de Javier, M. Gomez, P. Gomez-Sal, P. Royo, Inorg. Chem. 32 (1993)
5454e5457;
(f) A. Caselli, E. Solari, R. Scopelliti, C. Floriani, N. Re, C. Rizzoli, A. Chiesi-
Villa, J. Am. Chem. Soc. 122 (2000) 3652e3670;
(g) H. Brunner, G. Gehart, W. Meier, J. Wachter, B. Wrackmeyer, B. Nuber,
M.L. Ziegler, J. Organomet. Chem. 436 (1992) 313e324;
(h) R.B. Wilson Jr., A.P. Sattelberger, J.C. Huffman, J. Am. Chem. Soc. 104
(1982) 858e860;
h) G.C. Campbell, J.A.M. Canich, F.A. Cotton, S.A. Duraj, J.F. Haw, Inorg.
Chem. 25 (1986) 287e290;
i) F.A. Cotton, M.P. Diebold, W.J. Roth, J. Am. Chem. Soc. 109 (1987) 5506e
514;
j) F.A. Cotton, S.A. Duraj, W.J. Roth, J. Am. Chem. Soc. 106 (1984) 4749e
751;
(
5
(
4
(i) A.J. Scioly, M.L. Luetkens Jr., R.B. Wilson Jr., J.C. Huffman, A.P. Sattelberger,
Polyhedron 6 (1987) 741e757;
(
(
(
(
k) J.A.M. Canich, F.A. Cotton, Inorg. Chim. Acta 159 (1989) 163e168;
l) J.A.M. Canich, F.A. Cotton, S.A. Duraj, W.J. Roth, Polyhedron 5 (1986) 895e898;
m)F.A. Cotton, M.P. Diebold, W.J. Roth, J. Am. Chem. Soc. 108 (1986) 3538e3539;
n) L.G. Hubert-Pfalzgraf, M. Tsunoda, D. Katoch, Inorg. Chim. Acta 51 (1981)
(j) C. Ting, L. Messerle, J. Am. Chem. Soc. 111 (1989) 3449e3450;
(k) A.P. Sattelberger, R.B. Wilson Jr., J.C. Huffman, Inorg. Chem. 21 (1982)
2392e2396;
(l) A.P. Sattelberger, R.B. Wilson s Jr., J.C. Huffman, Inorg. Chem. 21 (1982)
4179e4184;
(m) E. Babaian-Kibala, F.A. Cotton, Inorg. Chim. Acta 171 (1990) 71e74;
(n) F.A. Cotton, L.M. Daniels, C.A. Murillo, X. Wang, J. Am. Chem. Soc. 118
(1996) 4830e4833.
8
(
4
1e84;
o) P.F. Gilletti, D.A. Femec, F.I. Keen, T.M. Brown, Inorg. Chem. 31 (1992)
008e4012;
(
(
p) M.D. Fryzuk, D.H. McConville, Inorg. Chem. 28 (1989) 1613e1614;
q) S. Gupta, S. Sharma, O.P. Vermani, A.K. Narula, Indian J. Chem. Sect. A
[22] R.D. Shannon, Acta Crystallogr. Sect. A Cryst. Phys. Diffr. Theor. Gen. Crys-
tallogr. A32 (1976) 751e767.
Inorg. Bio-inorg. Phys. Theor. Anal. Chem. 36A (1997) 232e234;
(
(
r) M.E. Clay, T.M. Brown, Inorg. Chim. Acta 72 (1983) 75e80;
s) F.A. Cotton, W.T. Hall, J. Am. Chem. Soc. 101 (1979) 5094e5095.
[23] M. Matsuura, T. Fujihara, A. Nagasawa, S.W. Ng, Acta Crystallogr. E68 (2012)
m1166.
[
9] [Multinuclear Nb and Ta complexes] (a) F.A. Cotton, W.J. Roth, Inorg. Chim.
[24] G.M. Sheldrick, SADABS, University of Göttingen, Germany, 1996.
[25] A. Altomare, G. Cascarno, C. Giacovazzo, A. Gualiardi, M.C. Burla, G. Polidori,
M. Camalli, J. Appl. Crystallogr. 27 (1994) 435e436.
[26] P.T. Beurskens, et al., Technical Report of Crystallography Laboratory, Uni-
versity of Nijmegen, Nijmegen, The Netherlands, 1994.
[27] G.M. Sheldrick, SHELEX-97, University of Gözttingen, Germany, 1997.
Acta 126 (1987) 161e166;
(
(
b) J.A. Canich, F.A. Cotton, S.A. Duraj, Inorg. Chim. Acta 156 (1989) 41e46;
c) F.A. Cotton, M.P. Diebold, R. Llusar, W.J. Roth, J. Am. Chem. Soc. (1986)
1276e1278;
(
d) F.A. Cotton, S.A. Duraj, W.J. Roth, J. Am. Chem. Soc. 106 (1984) 3527e