Exploring the reactivity of tungsten bis(imido) dimethyl complexes with methyl aluminium reagents: Implications for ethylene dimerization

Article abstract of DOI:10.1039/c0dt00110d

The reaction of [WCl₂(NAr)₂(DME)] (1) with excess Me₃Al affords the dimethyl complex [WMe₂(N{Ar}AlMe ₂{μ-Cl})(NAr)] (2), which on treatment with THF or MeAlCl ₂ yields [WMe₂(NAr)₂(THF)] (3) and [WMe ₂(N{Ar}AlMe(Cl){μ-Cl})(NAr)] (5), respectively. Complex 3 is unstable in solution dissociating to form [WMe₂(NAr)₂] (4) that may be isolated as an adduct with PMe₃, [WMe ₂(NAr)₂(PMe₃)] (6). While complex 2 is inert towards ethylene, complex 3 reacts rapidly to afford a mixture of methane and but-1-ene (1:4). Neither complex 2 nor 3 react with propylene. Reaction of 3 with a C₂H₄/C₂D₄ (1:1) affords a mixture of isotopomers that is consistent with complete isotopic scrambling. The structures of complexes 1, 2, and 3 have been determined by X-ray diffraction. The Royal Society of Chemistry 2010.

Full text of DOI:10.1039/c0dt00110d

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Exploring the reactivity of tungsten bis(imido) dimethyl complexes with
methyl aluminium reagents: implications for ethylene dimerization†
William R. H. Wright,a Andrei S. Batsanov,a Judith A. K. Howard,a Robert P. Tooze,b Martin J. Hanton*b and
Philip W. Dyer*a
Received 10th March 2010, Accepted 21st May 2010
First published as an Advance Article on the web 21st June 2010
DOI: 10.1039/c0dt00110d
The reaction of [WCl2(NAr)2(DME)] (1) with excess Me3Al affords the dimethyl complex
[WMe2(N{Ar}AlMe2{m-Cl})(NAr)] (2), which on treatment with THF or MeAlCl2 yields
[WMe2(NAr)2(THF)] (3) and [WMe2(N{Ar}AlMe(Cl){m-Cl})(NAr)] (5), respectively. Complex 3 is
unstable in solution dissociating to form [WMe2(NAr)2] (4) that may be isolated as an adduct with
PMe3, [WMe2(NAr)2(PMe3)] (6). While complex 2 is inert towards ethylene, complex 3 reacts rapidly to
afford a mixture of methane and but-1-ene (1 : 4). Neither complex 2 nor 3 react with propylene.
Reaction of 3 with a C2H4/C2D4 (1 : 1) affords a mixture of isotopomers that is consistent with complete
isotopic scrambling. The structures of complexes 1, 2, and 3 have been determined by X-ray diffraction.
initiators [WCl2(NAr)2] exhibit ethylene dimerization behaviour
Introduction
(W : Al 1 : 15) at elevated pressures (40 bar).11–13
Since the pioneering work of Ziegler in the mid-1950s metal-
Recently, the nature of the interplay between Me2AlCl and
catalysed alkene oligomerisation has attracted an overwhelming
research effort from both academia and industry.1–3 An area of
various bis(imido) tungsten species of relevance to ethylene
dimerization via a concerted metallacyclic coupling cycle, has been
explored computationally.14 In particular, this study probed the
key degradation manifold of the metallacyclopentane (A) via a
combined b-hydride/reductive elimination pathway (Scheme 1).
This revealed an unexpectedly high barrier to reductive elimination
from B (72.4 kcal mol-1), something accounted for by the presence
of the two strongly donating imido ligands.
particular importance is the catalytic dimerization of alkenes,
which provides a straightforward and economical route for the
manufacture of higher olefinic feedstocks for industry.4 Here,
amongst the most desirable products are the linear, short-chain
materials (C4–C20) that are finding increasing use as comonomers
in polyethylene manufacture, as plasticizers, and as synthetic
lubricant base oils.5 For these applications, the design of catalysts
that engender high levels of control over the selectivity and extent
of chain branching remains a crucial goal.
Although a number of metal-based dimerization initiator sys-
tems have been reported,4,6 of particular interest are those derived
from WCl6, which display a desirable selectivity for the head-to-
tail and tail-to-tail products resulting from a-olefin coupling.7 In
general, these tungsten-based initiators are generated in situ by
sequential reaction of WCl6 with an aryl amine at elevated tem-
peratures, followed by addition of a co-initiator, typically EtAlCl2
(EADC) or Et3Al2Cl3 (W : amine : Al 1 : 2 : 10).8,9 While the exact
nature of the resulting active initiator species is unknown, it is
well documented that efficient removal of HCl is one key feature
in ensuring high catalytic activity.10,11 This observation strongly
suggests that either amido and/or imido tungsten complexes
are generated during pro-initiator preparation. Such a notion is
supported by the reports that, in combination with EADC, discrete
imido complexes of the type [WCl4(NAr)] (W : Al 1 : 11) are active
for propylene dimerization, while related tungsten bis(imido) pro-
Scheme 1 Proposed metallacyclic b-hydride elimination and reductive
elimination/olefin dissociation steps of bis(imido) tungsten-initiated ethy-
lene dimerization.
In order to attenuate the electron density at the tungsten centre
and hence facilitate reductive elimination, coordination of the
Lewis acidic alkyl aluminium to the nitrogen lone pairs of the
imido fragments was investigated. The lowest energy arrangements
were computed to be those in which there is both imido-N→Al and
Cl→W coordination (Fig. 1). As anticipated the double Lewis acid
adduct A-Al2 achieves the most significant lowering of the energy
barrier to reductive elimination, by 24.3 kcal mol-1 relative to that
determined for A. Clearly the overall free energy of activation for
metallacycle degradation remains rather high, something that is at
odds with the observed catalytic performance of such systems.11
Together, these contradictory observations inspired us to in-
vestigate model reactions between discrete tungsten bis(imido)
aCentre for Sustainable Chemical Processes, Department of Chemistry,
Durham University, South Road, Durham, UK DH1 3LE. E-mail: p.w.
dyer@durham.ac.uk; Fax: +44(0)191 334 2150; Tel: +44(0)191 384 4737
bSasol Technology (UK) Ltd, Purdie Building, North Haugh, St Andrews,
Fife, UK KY16 9ST
† CCDC reference numbers 769100–769102. For crystallographic data in
CIF or other electronic format see DOI: 10.1039/c0dt00110d
7038 | Dalton Trans., 2010, 39, 7038–7045
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Fig. 1 Computed modes of Me2AlCl coordination to the metallacyclic
bis(imido) tungsten complex A.14
◦
˚
Table 1 Selected bond distances (A) and angles ( ) in 1
Molecule
A
B
W–N(1)
W–N(2)
W–Cl(1)
W–Cl(2)
W–O(1)
W–O(2)
N(1)–C
N(2)–C
1.7598(17)
1.7599(17)
2.3841(7)
2.3878(9)
2.3494(15)
2.3446(15)
1.393(3)
1.7571(18)
1.7579(17)
2.3852(8)
2.3873(7)
2.3307(16)
2.3539(16)
1.393(3)
Fig. 2 Molecular structure of [WCl2(NAr)2(DME)] (1), showing inde-
pendent molecule A only. Henceforth H atoms are omitted for clarity and
the thermal ellipsoids set at the 50% level.
binding as a result of the limited p-bonding capacity of the
tungsten centre.19–22
1.395(3)
1.393(3)
N(1)–W–N(2)
Cl(1)–W–Cl(2)
O(1)–W–O(2)
N(1)–W–Cl(1)
N(1)–W–Cl(2)
N(2)–W–Cl(1)
N(2)–W–Cl(2)
W–N(1)–C
W–N(2)–C
103.89(8)
154.28(2)
70.06(6)
99.73(6)
97.11(6)
97.45(6)
97.19(6)
163.06(15)
161.47(15)
104.91(8)
154.89(2)
70.33(6)
98.64(6)
97.24(6)
97.38(6)
97.18(6)
164.93(15)
168.14(15)
Reactions of [WCl2(NAr)2(DME)] (1) with MexAlCl3-x
It has long been established that the nature of aluminium- and
boron-based co-initiators has a pronounced impact upon the activ-
ity and selectivity of homogeneous early transition metal-catalysed
alkene polymerisation.23 As would be anticipated similar effects
have recently also been exemplified for the tungsten-based dimeri-
sation systems of interest here.9 Consequently, an examination of
the reaction of the dichloride complex 1 with variously-substituted
alkyl aluminium reagents was carried out. Initially, the reaction
with Me3Al (W : Al 1 : 5) was undertaken and cleanly afforded
a single tungsten-containing product [WMe2(N{Ar}AlMe2{m-
Cl})(NAr)] (2), which was obtained following complete conversion
of 1 (Scheme 2). Complex 2 presents two inequivalent aryl imido
environments together with distinct resonances attributable to
complexes and MexAlCl3-x reagents in vitro in order to gain a
greater understanding of the role of Lewis acidic activators in
ethylene dimerization.
Results and discussion
Preparation and structural characterisation of
[WCl2(NAr)2(DME)] (1)
the Al- and W-bound methyl groups according to H and 13C
1
1
{ H} NMR spectroscopy. The observed asymmetry results from
The starting point for this investigation was the previously
reported complex [WCl2(NAr)2(DME)] (1), which was prepared
according to the known procedure.15 To provide a comparison with
the structures of the various tungsten alkyl complexes described
below, the molecular structure of complex 1 was determined and
found to possess an asymmetric unit comprising two molecules (A
and B) of similar conformation (Fig. 2, Table 1).
coordination of an Me2AlCl fragment (generated in situ during
transmetallation) to one of the imido ligands. Excess Me3Al was
found to be essential for clean generation of 2, since in this reaction
stoichiometric amounts of the alkyl aluminium are sequestered by
the DME liberated from 1.24
The tungsten atom adopts a strongly distorted octahedral coor-
dination with the two chloride and two imido ligands occupying
mutually trans and cis positions, respectively, as found for the
related complex [WCl2(NPh)2(bipy)] reported by Bradley et al.16
As is the case for the majority of cis-octahedral d0 bis(imido)
complexes, the p-donor chlorides of 1 are tilted away from the
two multiply bonded ligands due to electronic factors.17,18 The
metric parameters associated with the two imido nitrogen atoms of
either of the independent molecules A or B are identical to within
experimental error, with a mean W–N–C angle of 164.4◦ and an
˚
average W–N distance of 1.759(2) A, data that are comparable
with the parameters determined for [WCl2(NPh)2(bipy)] (165.8◦
˚
and 1.78(1) A, respectively). In line with established precedent, the
W–N bond distances of 1 are best regarded as being intermediate
Scheme
2 Summary of the synthesis and reactivity of
between those for X2 (2-electron) and LX2 (4-electron) imido
[WMe2(N{Ar}AlMe2{m-Cl})(NAr)] (2).
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 7038–7045 | 7039
©

View Article Online
An X-ray diffraction study was used to confirm the connectivity
of 2 (Fig. 3, Table 2). The complex has an approximate non-
crystallographic Cs symmetry, with the mirror plane passing
through the W, Al, Cl, and both N atoms. While the aryl
substituent at N(1) lies in this plane, the ring at N(2) is perpen-
dicular to it (with an angle between the ring planes of 85.4◦).
The coordination geometry of tungsten is intermediate between
trigonal bipyramidal (with N(1) and Cl axial) and square-based
pyramidal (with N(2) apical), but much closer to the former, as
indicated by Reedijk’s parameter t = 0.77 (1 for an ideal TBP
and 0 for SBP).25 There is an additional ‘umbrella’ distortion,26
whereby all three equatorial ligands are bent away from the axial
terminal imido ligand, N(1)Ar, which forms the highest-order
bond with the central atom. The equatorial sites are occupied by
two methyl groups and an unsymmetrically-bridging imido ligand,
N(2)Ar, which is a constituent of a four-membered –W–N–Al–
Cl- metallacycle, with a dative (Al)Cl→W interaction completing
the tungsten’s coordination sphere. Consequently, the Al atom
acquires a near-tetrahedral geometry, with N(2) being trigonal
ꢀ
planar (
N(2) ~ 360◦). Such a bonding scheme is supported
ang
by the inequivalence of the two W–N bond distances, which differ
by 0.12 A. The shorter W–N(1) bond has a degree of pseudo-
˚
triple bond character arising from lone pair donation to tungsten
(LX2 binding) and, as expected, is significantly shorter than the
˚
W–N bonds of the parent complex 1 (average 1.759(2) A); the
slight bending at N(1) results from steric repulsion between the
aryl substituents.22 Clearly, coordination of the Lewis acid to one
of the two imido ligands of [WMe2(NAr)2] alleviates competition
for the limited p-bonding capacity of the tungsten centre.27 The
Me2AlCl-bound imido ligand has a W–N(2)-C(21) bond angle
of 133.24(12)◦ and a relatively long W–N(2) bond distance of
˚
1.8613(15) A. Notably, the structure predicted computationally
(A-Al) very reliably reproduces that determined experimentally
for 2.14
Fig. 3 Molecular structure of [WMe2(N{Ar}AlMe2{m-Cl})(NAr)] (2).
Structurally related –M–N–Al–Cl– metallacycles have been
observed previously in the complexes [Cp2ZrC(Ph)=C(Ph)-
C(Me)=NAl(Cl)Me2]28 and [Cp2ZrCH(Ph)N(Ph)Al(Cl)Me2].29
◦
˚
Table 2 Selected bond distances (A) and angles ( ) for [WMe2(N-
{Ar}AlMe2{m-Cl})(NAr)] (2) and [WMe2(NAr)2(THF)] (3)
˚
The Al–N bonds of 2 (1.958(2) A) and of the two Zr com-
˚
plexes (1.888(2) and 1.952(7) A, respectively), are shorter than
2
3, A
3, B
3, C
those of traditional donor–acceptor adducts, e.g. Me3N·AlClMe2
30
31
˚
˚
(2.010(1) A) or 5,6-benzoquinoline·AlClMe2 (2.016(3) A),
W–N(1)
W–N(2)
W–C(1)
1.7402(15)
1.8613(15)
2.129(2)
2.139(2)
2.6440(6)
1.399(2)
1.439(2)
169.63(5)
108.27(7)
94.95(8)
96.81(9)
82.00(5)
79.20(7)
79.68(7)
113.16(8)
114.7(1)
123.6(1)
165.9(1)
133.2(1)
2.2757(8)
1.958(2)
1.947(2)
1.951(2
1.764(4)
1.784(4)
2.157(5)
2.159(5)
2.322(3)
1.391(6)
1.387(6)
156.0(2)
111.1(2)
97.2(2)
94.0(2)
88.8(2)
76.9(2)
76.3(2)
1.746(4)
1.771(4)
2.173(5)
2.151(5)
2.305(3)
1.396(6)
1.395(6)
159.7(2)
110.8(2)
96.5(2)
94.0(2)
89.4(2)
77.3(2)
76.5(2)
1.757(4)
1.773(4)
2.151(5)
2.159(5)
2.311(4)
1.397(6)
1.386(6)
158.7(2)
111.4(2)
96.8(2)
94.1(2)
89.8(2)
76.8(2)
76.0(2)
while the Al–Cl bond of 2 is longer than the (terminal) single
32
˚
bonds in NAl(Cl)Me2 moieties (2.14–2.20 A). The Cl→W bond
˚
W–C(2)
(2.6440(6) A) is marginally longer than previously reported aryl-
W–Xa
˚
Cl→W dative bonds, which lie in the range 2.57–2.61 A, as a
N(1)–C
N(2)–C
consequence of the high trans influence of the imido ligand.33 An
analogous –M–N–B–Cl– metallacycle has also been reported in
the complex [Os(Cl){HB(pz)3}{N(Ph)-B(Cl)Ph2}].34
N(1)–W–Xa
N(1)–W–N(2)
N(1)–W–C(1)
N(1)–W–C(2)
X–W–N(2)a
X–W–C(1)a
X-W–C(2)a
N(2)–W–C(1)
N(2)–W–C(2)
C(1)–W–C(2)
W–N(1)–C
W–N(2)–C
Al–Cl
Reactivity of [WMe2(N{Ar}AlMe2{l-Cl})(NAr)] (2)
Although complex 2 is prepared in the presence of excess AlMe3
(W : Al 1 : 5) the formation of products comparable to the double
Lewis adduct A-Al2 proposed computationally are not observed.14
Indeed, a previously isolated sample of complex 2 does not
react with AlMe3 (W : Al 1 : 5) even at elevated temperatures
(60 ◦C). Nevertheless, 2 reacts rapidly with MeAlCl2 (W : Al 1 : 3)
at RT, resulting in complete displacement of the coordinated
Me2AlCl and the clean formation of [WMe2(N{Ar}Al(Cl)Me{m-
Cl})(NAr)] (5) (Scheme 2); again, no evidence of the formation of
the double Lewis adduct was obtained. Clearly the creation and
retention of –W–N–Al–Cl– metallacycles is a key requirement
for Lewis acid coordination to the imido residues of these
types of tungsten complex. Once formed, the resulting Lewis
acid–Lewis base adducts are quite stable as exemplified by a
variable temperature 1H NMR spectroscopic study (+25 to 60 ◦C;
C6D5Cl) of 2, which gave no indication of either decomposition or
dissociation of Lewis acid.
111.0(2)
109.6(2)
130.2(2)
176.9(4)
154.7(3)
111.8(2)
108.9(2)
130.7(2)
178.1(4)
150.1(3)
110.3(2)
109.6(2)
130.9(2)
178.2(4)
153.0(3)
Al–N(2)
Al–C(3)
Al–C(4)
Cl–Al–N(2)
Cl–Al–C(3)
Cl–Al–C(4)
N(2)-Al–C(3)
N(2)-Al–C(4)
C(3)-Al–C(4)
W–Cl–Al
90.47(5)
107.94(8)
106.32(8)
112.55(9)
114.93(9)
119.9(1)
78.07(2)
109.10(8)
117.6(1)
W–N(2)-Al
Al–N(2)-C
a X = Cl (2); X = O(1) (3).
7040 | Dalton Trans., 2010, 39, 7038–7045
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
It has recently been shown that dimerization initiators generated
in situ from WCl6/ArNH2/RxAlCl3-x (60 ◦C) are only active
when the aluminium co-initiator bears both an alkyl and a
halide group such as for MeAlCl2, EtAlCl2, and EtAlCl2; use of
AlEt3, AlMe3, or MAO (methyl aluminoxane) all fail to promote
dimerization.9 Furthermore, it was shown that the activity of
the initiator system decreases as a function of the aluminium
reagent in the order EtAlCl2>Et3Al2Cl3>Et2AlCl. Collectively,
these observations closely parallel our findings for the well-defined
imido complexes reported here: not only is N→Al and (Al)Cl→W
coordination facile, but also MeAlCl2 displaces Me2AlCl (2→5,
Scheme 2). Hence, as inferred computationally, formation of –
W–N–Al–Cl– metallacycles is likely to be a key feature of these
tungsten-based dimerisation initiators.14
In order to further probe the strength and nature of the Me2AlCl
coordination in 2, reactions with Lewis bases were undertaken.
Treatment of 2 with excess THF affords [WMe2(NAr)2(THF)]
(3), accompanied by elimination of [Me2AlCl(THF)], as apparent
from 1H NMR analysis of the reaction mixture (Scheme 2).
Following recrystallisation from diethyl ether, single crystals of
3 suitable for X-ray diffraction analysis were obtained. The
asymmetric unit comprises three molecules A, B, and C of similar
geometry (Fig. 4, Table 2).
ligands of 3 adopt a bonding arrangement that is intermediate
between that of an LX2 and X2 donor.22
In C6D6 solution the THF ligand of 3 is labile at RT, immediately
dissociating to give what appears to be the base-free complex
[WMe2(NAr)2] (4) and apparently uncoordinated THF as indi-
1
cated by H NMR spectroscopic analysis (Scheme 2).‡ Identical
behaviour is observed for 3 in C6D5Cl solution, with no 1H NMR
spectral changes being apparent over the temperature window
-40 ◦C to RT. Complex 4 may also be prepared quantitatively
through reaction of 3 with one equivalent of (C6F5)3B in C6D6,
which liberates an equimolar quantity of the known adduct
[(C6F5)3B(THF)].37§ However, although complex 4 is stable in
solution for days, all attempts at its isolation resulted in its
decomposition. This is consistent with the observation that
exposure of a crystalline sample of 3 to vacuum (30 min) results in
the removal of THF and subsequent degradation. This contrasts
with the analogous molybdenum complex [MoMe2(NAr)2(THF)],
which is stable under reduced pressure.35 In contrast, reaction
of THF complex 3 with trimethyl phosphine cleanly affords
[WMe2(NAr)2(PMe3)] (6), which is stable in the solid state even
on prolonged exposure to vacuum.
It is important to note that although complex 3 may be prepared
in high yield by sequential reaction of 1 with AlMe3 followed
by THF, all attempts to prepare 3 through reaction of 1 with a
slight excess of MeMgCl in THF, resulted in the formation of an
intractable mixture of products, none of which could be assigned
to 3. This contrasts with the analogous molybdenum system
where [MoMe2(NAr)2(THF)] may be prepared in good yield
from reaction of [MoCl2(NAr)2(DME)] with methyl magnesium
chloride in THF solution.22
Reactions of complexes 2 and 3 with C2H4 and C3H6
An evaluation of the role of Me2AlCl coordination upon the
reactivity of tungsten bis(imido) species of relevance to alkene
dimerization has been undertaken by comparing the behaviour of
complexes 2 and 3/4 with ethylene (5 eq.) (Scheme 3). Notably, no
reaction occurred between the tungsten-alkyl aluminium adduct 2
even on prolonged heating (48 h, 60 ◦C), irrespective of the solvent
(C6D6 or C6D5Cl). Similarly, complex 2 does not initiate propylene
dimerization, with no reaction taking place. This lack of reactivity
Fig. 4 Molecular structure of [WMe2(NAr)2(THF)] (3), showing inde-
pendent molecule A only.
The tungsten’s coordination geometry in 3 is broadly similar to
that in 2 (with the THF occupying the position of the chloride
ligand) although slightly more distorted towards SBP (mean t =
0.46) with the N(2)Ar ligand in the pseudo-apical position and
the W, O and both N atoms lying in one plane. The shift to
SBP geometry is even more pronounced in the related dimethyl
molybdenum complex [MoMe2(NAr)2(PMe2Ph)]35 (t = 0.26),
probably due to bigger steric demands of PMe2Ph compared with
THF.18b The arene ring bonded to N(1) of 3 is inclined to this
plane by 16.7◦ (molecule A), 7.8◦ (B) and 3.7◦ (C), while the ring
at N(2) is nearly perpendicular to it (dihedral angles 84.9, 83.8 and
Scheme 3 Behaviour of complexes 2 and 3/4 towards ethylene.
70.9◦, respectively). The W–C bond distances of 3 at 2.157(5) and
‡ Upon dissolution in C6D6 the THF moiety of complex 3 gives rise to
resonances at 3.47 and 1.40 ppm by 1H NMR spectroscopy, which are
identical to those found for an authentic sample of free THF.
§ The formation of [(F5C6)3B(THF)] was confirmed spectroscopically by
comparison with the data from an authentic sample.
36
˚
2.159(5) A are unremarkable, and are comparable to those of 2.
Despite the differences in W–N–C bond angle, the two W–N bond
˚
distances of 1.764(4) and 1.784(4) A again indicate that the imido
This journal is The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 7038–7045 | 7041
©

View Article Online
Table 3 Isotopomer distributions predicted by theory and observed
following reaction of complex 3 with C2H4/C2D4 (1 : 1)
must result from insertion of an alkene into a W–H bond that
would be inherent in such a process. This is supported by recent
observations made for the in situ-generated systems, which gave
rise to identical isotopmer scrambling together with the formation
of linear decenes that resulted from dimerisation of 1-pentene.9 As
such we feel that a chain growth mechanism must be given the same
theoretical considerations granted to the metallacycle (oxidative
coupling) pathway for tungsten-mediated alkene dimerization.
Predicted
Isotopomer Chain-growth Metallacycle Fully Scrambled Observed a
C4H8
1
1
0
1
2
1
0
1
1
1
0
0
0
2
0
0
0
1
1
8
1.08
1.71
2.70
3.80
4.29
3.68
1.13
1.00
>0.2
C4H7D
C4H6D2
C4H5D3
C4H4D4
C4H3D5
C4H2D6
C4HD7
C4D8
28
56
70
56
28
8
Conclusions
Discrete tungsten imido complexes in the presence of aluminium-
based Lewis acids are now firmly established as pro-initiators
in alkene dimerization, yet the nature of the actual active
species remains unclear. A recent computational study pro-
posed that the formation of bimetallic complexes of the type
[WMe2(N{Ar}AlMe(X){m-Cl})(NAr)] could be implicated in the
catalytic process since coordination of the Lewis acid in this
manner significantly lowered the activation barrier for one of the
key catalytic steps, namely tungstenacyclopentane degradation.14
However, it has been demonstrated here that although transmetal-
lation reactions between methyl aluminium reagents and tungsten
bis(imido) dihalide complexes such as 1 occur readily to afford
bimetallic complexes of the type [WMe2(N{Ar}AlMe(X){m-
Cl})(NAr)], where X = Me (2), Cl (5), such compounds do not
react with either ethylene or propylene.
In contrast, the labile dialkyl derivative [W(NAr)2Me2(THF)]
(3), generated through displacement of the Lewis acid moiety
of 2 by THF, reacts rapidly with ethylene to generate a mixture
of, as yet, unidentified species that initiate catalytic dimerization
of ethylene affording methane and but-1-ene in a 1 : 4 ratio,
respectively. The methane evolved during the initial stages of the
reaction is suggestive of initial ethylene insertion into a tungsten–
methyl bond, followed by b-hydride elimination and reductive
elimination sequence delivering the catalytically active species,
rather than oxidative coupling.
1
a GC-MS.
of complex 2 towards alkenes has been attributed to the strong
coordination of the Me2AlCl fragment to the imido moiety and
tungsten, which prevent initial alkene association (the LUMO of
the 14e- W(NAr)2Me2 fragment being occupied by (Al)Cl→W
donation).38 Clearly these observations are at odds with the
high dimerization activity demonstrated by the RxAlCl3-x/WCl6-
based systems derived in situ. One possible explanation for this
behaviour is that bis(imido) species are not in fact generated in the
reaction of WCl6 and ArNH2 as proposed,12,14 rather that mixed
mono(imido)/amido systems are formed as suggested previously.11
In contrast to the behaviour of 2, addition of ethylene (5 eq.) to
solutions of complex 4 (generated upon dissolution of 3) results in
an immediate reaction at RT. Analysis of the reaction mixture
after 5 min by 1H NMR spectroscopy confirms the complete
disappearance of 3 and 4 and the appearance of a number
of unattributable new resonances. After 48 h at RT ethylene
conversion is complete, yielding but-1-ene, but-2-ene (trace), and
methane (alkene : alkane 4 : 1 by integration).¶ Further quantities
of ethylene (2 ¥ 10 eq.) added to the reaction mixture are slowly
converted (days) to but-1-ene. Although reproducible, the origin
of the methane evolved during the reaction of 4 with C2H4 remains
unaccounted for. No C6 components are observed (which would
arise from co-dimerisation of C2 and C4 alkenes) in agreement with
the failure of these discrete W-species 3/4 to initiate dimerisation
of propene.
As stated, it has been widely proposed that W-imido-
promoted selective alkene dimerization occurs via a metallacycle
mechanism,12,39 something that is consistent with the mode of
action of discrete chromium-based alkene trimerisation catalysts.40
Hence, it was decided to probe this assumption via the now well-
established C2H4/C2D4 co-oligomerisation experiment.41–45 To this
end, a C6D6 solution of complex 3 was treated with a 1 : 1 mixture
of C2H4/C2D4 and the resulting products analysed. As shown in
Table 3, the observed distribution of isotopomers is characteristic
of a system in which significant scrambling has occurred.46
This result implies that rather than the expected metallacycle
mechanism a chain-growth process (Cossee-type mechanism) may
be in operation. While we have not conclusively proven here that
a chain-growth process is in operation, the observed scrambling
Together with the findings of Tobisch’s recent computational
study,14 the observations made here strongly suggest that simple
discrete tungsten bis(imido) complexes, of the type outlined
in Fig. 1 are unlikely to be the true active components in
homogeneous catalytic alkene dimerization. With this in mind,
further detailed studies of the reactions of complexes 2 and 3 with
Lewis acids and alkenes are being actively pursued.
Experimental
All operations were conducted under an atmosphere of dry
nitrogen using standard Schlenk and cannula techniques, or
in a nitrogen-filled Saffron Scientific glove box. NMR-scale
reactions were conducted using NMR tubes fitted with Young’s
tap valves. Bulk solvents were purified using an Innovative
Technologies SPS facility and degassed prior to use. NMR
solvents (C6D6 and C6D5Cl) were dried over P2O5, distilled and
degassed before use. DME (Aldrich) was distilled under nitrogen
from sodium/benzophenone. When appropriate, liquid reagents
were dried, distilled and deoxygenated. Nitrogen gas was passed
through a drying column (silica/CaCO3/P2O5) and ethylene and
propylene were dried using a column of P2O5 and a c.H2SO4 gas
washing bottle. The known compound [WCl2(NAr)2(DME)] (1)
¶ Comparison was made against an authentic sample of methane dissolved
in C6D6 in a sealed NMR tube, which gives a resonance at 0.30 ppm.
The formation of but-1-ene was verified by GC analysis of the volatile
components from the reaction of 3 with C2H4 after 48 h at RT.
7042 | Dalton Trans., 2010, 39, 7038–7045
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
was prepared using standard literature procedures,15 while neat
Me3Al was purchased from Albermarle R & D centre. MeAlCl2
was obtained by slow evaporation of hexane from MeAlCl2
solutions obtained from Aldrich.
Dissolution of 3 in either C6D6 or C6D5Cl leads to spontaneous
dissociation of THF affording [WMe2(NAr)2] (4), which is only
stable in solution and was hence characterised by NMR spec-
troscopy in situ. 1H NMR (C6D6, 400 MHz): d 7.07 (6H, b, Hmeta
and Hpara), 3.62 (4H, septet, CH3CH, 3JHH = 7.2 Hz), 1.25 (6H, s,
W(CH3)2, 2JWH = 7.6 Hz), 1.17 (24H, d, CH3CH, 3JHH = 7.2 Hz)
GC analyses were performed on an Agilent Technologies 6890
N GC system equipped with PONA (50 m ¥ 0.020 mm ¥ 0.50 mm)
and MDN12 (60 m ¥ 0.025 mm ¥ 0.25 mm) columns. GC-MS
analyses were performed on an Agilent Technologies 6890 N GC
system equipped with a MDN12 (60 m ¥ 0.025 mm ¥ 0.25 mm)
column, coupled to an Agilent Technologies 5973 N MSD Mass
Spectrometric instrument. Elemental analyses were performed by
the Analytical Services Department of the Chemistry Department,
Durham University. Routine NMR spectra were collected on a
Varian Unity 200, Varian Mercury 400 or an Avance 400 MHz
Bruker spectrometer. Chemical shifts were referenced to residual
protio impurities in the deuterated solvent (1H) or the 13C shift
of the solvent (13C). Solvent proton shifts (ppm): C6D6, 7.15;
C6D5Cl, 7.13, 6.98, 6.95. Solvent carbon shifts (ppm): C6D6,
128.3. 13C NMR spectra were assigned with the aid of DEPT 135
and gHMQC 1H/13C correlation experiments. Chemical shifts are
reported in ppm and coupling constants in Hz.
1
ppm. 13C{ H} NMR (C6D6, 175.9 MHz): d 152.8 (Cipso), 143.1
(Cortho), 126.0 (Cmeta), 123.0 (Cpara), 48.6 (W(CH3)2), 29.3 (CH3CH),
23.9 (CH3CH) ppm.
Preparation of [WMe2(NAr)2(PMe3)] (7)
To a C6D6 solution of [WMe2(NAr)2(THF)] (40 mg, 0.06 mmol)
was added PMe3 (0.19 mmol) by vacuum transfer. After 20 min,
the volatile components were removed in vacuo, giving 6 as a light
1
brown solid. H NMR (C6D6, 700 MHz): d 7.12 (4H, d, Hmeta
,
3JHH = 7.7 Hz), 7.02 (2H, t, Hpara, 3JHH = 7.7 Hz), 3.88 (4H, septet,
CH3CH, 3JHH = 7.7 Hz), 1.25 (24H, d, CH3CH, 3JHH = 7.7 Hz),
2
1.08 (6H, s, W(CH3)2, JWH = 5.2 Hz), 0.93 (9H, d, P(CH3)3,
2JPH = 7.7 Hz) ppm. 13C{ H} NMR (C6D6, 175.9 MHz): d 154.1
1
(Cipso), 144.2 (Cortho), 125.1 (Cmeta), 123.0 (Cpara), 28.8 (CH3CH),
1
24.4 (CH3CH), 21.9 (W(CH3)2), 13.4 (P(CH3)3, JPC = 18 Hz)
ppm. 31P{ H} NMR (C6D6, 283.2 MHz): d -25.0 ppm.
1
Preparation of [W(N{Ar}AlMe2{l-Cl})(NAr)Me2] (2)
Preparation of [(C6F5)3B(THF)]
To a solution of [W(NAr)2Cl2(DME)] (1) (50 mg, 0.07 mmol) in
C6D6 (0.8 mL) was added Me3Al (25 mg, 0.34 mmol). Removal of
volatile components from the C6D6 solution in vacuo gave a small
quantity of a tan brown solid. This material was recrystallized
from the minimal amount of hexane (-5 ◦C) to give single crystals
of W(N{Ar}AlMe2{m-Cl})(NAr)Me2 (2) of sufficient quality for
X-ray diffraction analysis. 1H NMR (C6D6, 200 MHz): d 6.96 (6H,
b, Hmeta and Hpara), 3.50 (2H, septet, CH3CH, 3JHH = 6.8 Hz), 2.65
(2H, septet, CH3CH, 3JHH = 6.6 Hz), 1.56 (6H, s, W(CH3)2), 1.17
A sample of B(C6F5)3 (200 mg, 0.39 mmol) was dissolved in THF (5
mL) and allowed to react for 20 min before all volatile components
were removed in vacuo to give [(C6F5)3B(THF)] as a white solid
in quantitative yield. 1H NMR (C6D6, 400 MHz): d 3.19 (4H, br,
1
n 1 = 9.4 Hz, OCH2), 0.92 (4H, br, n 1 = 11.0 Hz, CH2); 19F{ H}
2
2
NMR (C6D6, 376 MHz): d -132.9 (2F, d, 3JFF = 21 Hz, o-C6F5),
-154.8 (1F, 3JFF = 21 Hz, p-C6F5), -162.7 (2F, br t, 3JFF = 19 Hz,
m-C6F5). Anal. Calcd for C22H8BF15O: C, 45.24; H, 1.38. Found:
C; 45.00, H; 1.21.
3
3
(6H, d, CH3CH, JHH = 6.8 Hz), 1.03 (6H, d, CH3CH, JHH
=
3
6.8 Hz), 0.89 (12H, d, CH3CH, JHH = 6.6 Hz), 0.15 (6H, s,
1
Al(CH3)2Cl) ppm. 13C{ H} NMR (C6D6, 125.6 MHz): d 151.3
Reaction of [W(N{Ar}AlMe2{l-Cl})(NAr)Me2] (2) with C2H4
(Cipso), 149.2 (Cortho), 140.8 (Cmeta), 124.8 (Cpara), 71.1 (CH3OCH2),
61.3 (CH3OCH2), 50.9 (W(CH3)2), 29.4 (CH3CH), 28.6 (CH3CH),
26.3 (CH3CH), 24.2 (CH3CH), -6.80 (Al(CH3)3) ppm. Despite
repeated attempts, satisfactory elemental analyses for 2 could not
be obtained.
Complex 1 (50 mg, 0.07 mmol) was dissolved in C6D6 or C6D5Cl
(0.8 mL), followed by the addition of Me3Al (25 mg, 0.34 mmol)
to both solutions. 1H NMR spectroscopic analysis confirmed the
formation of 2 in situ in each case. The samples were then allowed
to contact ethylene (5 eq., 0.15 mmol) prior to heating at 60 ◦C for
1 h. Irrespective of the solvent, no reaction between ethylene and
1
2 was found to occur (verified using H NMR spectroscopy). A
Preparation of [WMe2(NAr)2(THF)] (3) and [WMe2(NAr)2] (4)
similar lack of reaction was found when samples of 2 were heated
at 60 ◦C for 1 h under ethylene (1 atm).
Complex 1 (0.80 g, 1.15 mmol) was dissolved in CH2Cl2 (30 mL)
and Me3Al (414 mg, 5.75 mmol) also dissolved in CH2Cl2 (10
mL) was added. The combined solution was allowed to stir for
approximately five minutes before the volatile components were
then removed in vacuo to give a dark brown solid, which was
then re-dissolved in hexane (40 mL). To this hexane solution
was added THF (0.41 mL, 5.76 mmol) and, after 5 min, the
mixture was concentrated in vacuo. This resulted in the immediate
formation of a yellow precipitate that was collected via filtration
and recrystallized from diethyl ether (-15 ◦C), giving light brown
crystals (which were dried under a flow of N2) of sufficient quality
for X-ray diffraction analysis; yield 200 mg (27%). Anal. calcd. for
C30H48N2OW: C 56.60, H 7.60, N 4.40; found: C 56.80, H 7.52 N
4.70.
Reaction of [W(N{Ar}AlMe2{l-Cl})(NAr)Me2] (2) with MeAlCl2
Complex 1 (30 mg, 0.04 mmol) was reacted with Me3Al (18 mg,
0.25 mmol) in C6D6. This solution was then treated with MeAlCl2
(0.14 mmol, 1 M in hexanes, 0.14 mL) and, after a 30 min
reaction period, the resulting mixture was dried in vacuo. This
material was analyzed using 1H NMR spectroscopy. The resulting
spectrum presented signals assignable to complex 5, together
with resonances attributable to an unknown [MexAlCl3-x(DME)]
adduct, a by-product of the reaction of [W(NAr)2Cl2(DME)] (2)
and Me3Al. 1H NMR (C6D6, 400 MHz): d 6.95 (6H, b, Hmeta and
Hpara), 3.58 (2H, CH3CH, septet 3JHH = 6.8 Hz), 2.59 (2H, CH3CH,
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 7038–7045 | 7043
©

View Article Online
Table 4 Crystal data and X-ray experimental details
1
2
3
CCDC no.
Formula
Formula weight
T/K
769100
C28H44Cl2N2O2W
695.40
769101
C28H46AlClN2W
656.95
769102
C30H48N2OW
636.55
120
120
120
Crystal system
Space group (no.)
Triclinic
Monoclinic
P21/n (no. 14b)
11.691(2)
17.983(3)
14.342(2)
90
92.64(1)
90
3012.1(8)
4, 1.449
3.97
0.11 ¥ 0.26 ¥ 0.34
0.4874–0.6692
35373, 8025
58
0.144, 0.028
0.018, 0.038
Monoclinic
P21/c (no. 14)
19.506(2)
16.994(2)
26.404(3)
90
95.20(2)
90
8716(2)
12, 1.455
4.00
0.15 ¥ 0.23 ¥ 0.26
0.1700–0.4323
121256, 25037
60
0.103, 0.046
0.045, 0.119
¯
P1 (no. 2)
˚
a/A
10.5913(14)
16.009(2)
18.714(3)
80.84(3)
87.75(3)
76.79(2)
3049.5(7)
4, 1.515
3.99
0.20 ¥ 0.30 ¥ 0.35
0.2419–0.4314
43557, 17064
60
0.098, 0.024
0.020, 0.046
˚
b/A
˚
c/A
a (◦)
b (◦)
g (◦)
3
˚
V/A
Z and rc/g cm-3
m(Mo-Ka)/mm-1
Crystal size/mm
Transmission factor, min-max
Reflections collected, unique
2q max. (◦)
Rint before and after correction
R1 [I > 2s(I)] and wR2 (all data)a
ꢀ
ꢀ
ꢀ
ꢀ
2
a R1 = ꢀFo| - |Fcꢀ/ |Fo|, wR2 = [ w(Fo - Fc2)2/ w(Fo2)2]1/2
.
b Non-standard space group.
septet 3JHH = 6.8 Hz), 1.60 (6H, s, W(CH3)2), 1.26 (6H, d, CH3CH,
3JHH = 6.4 Hz), 1.01 (6H, d, CH3CH, 3JHH = 6.8 Hz), 0.87 (12H,
d, CH3CH, 3JHH = 6.8 Hz), 0.14 and 0.13 (3H, s, Al(CH3)2Cl).
Acknowledgements
Durham University and Sasol Technology UK Ltd. (studentship
WRHW) are warmly acknowledged for financial support of this
work. We thank Dr A. M. Kenwright, Mrs. C. Heffernan, and Mr
I. McKeag for assistance with NMR studies.
Reaction of [WMe2(NAr)2(THF)] (3) with C2H4
[WMe2(NAr)2(THF)] (3) (20 mg, 0.03 mmol) was dissolved
in C6D6 (0.8 mL) and the resulting solution then allowed to
contact ethylene (5 eq., 0.17 mmol) and analyzed using 1H
NMR spectroscopy. After a 48 h reaction period but-1-ene was
observable in the H NMR spectrum. The presence of but-1-ene
was confirmed using GC analysis.
Notes and references
1 K. Ziegler, E. Holzkamp, H. Breil and H. Martin, Angew. Chem., 1955,
67, 426.
2 P. Kuhn, D. Se´meril, D. Matt, M. Chetcuti and P. Lutz, Dalton Trans.,
2007, 515.
1
3 C. Janiak, Coord. Chem. Rev., 2006, 250, 66.
4 S. Muthukumaru Pillai, M. Ravindranathan and S. Sivaram, Chem.
Rev., 1986, 86, 353.
5 D. Vogt, in Applied Homogeneous Catalysis with Organometallic
Compounds, B. Cornils, W. A. Herrmann, ed.; VCH, New York, 2nd
Edtn., 2002, pp. 240-253.
Reaction of [WMe2(NAr)2(THF)] (3) with C2H4/C2D4
In an NMR tube fitted with
a
J. Young’s valve
6 (a) For reviews see: Y. Chauvin, H. Olivier, in Applied Homogeneous
Catalysis with Organometallic Compounds, B. Cornils, W. A. Herrmann,
ed.; VCH, New York, 1996, pp. 258-268; (b) H. Olivier-Bourbigou,
L. Saussine, in Applied Homogeneous Catalysis with Organometallic
Compounds, B. Cornils, W. A. Herrmann, ed.; VCH, New York, 2nd
Edtn., 2002, pp. 253-265; (c) J. Skupin´ska, Chem. Rev., 1991, 91, 613;
(d) F. Speiser, P. Braunstein and L. Saussine, Acc. Chem. Res., 2005, 38,
784.
[WMe2(NAr)2(THF)] (3) (20 mg, 0.03 mmol) was dissolved
in C6D6 (0.8 mL) and then frozen (-78 ◦C). The resulting frozen
solution was allowed to contact C2H4 (0.17 mmol) and C2D4
(0.17 mmol). After 48 h at RT, the mixture was analyzed using 1H
NMR spectroscopy and by GC and GC-MS.
7 (a) H. R. Menapace, G. S. Benner and N. A. Maly, (Goodyear) U.S. Pat.
3,784, 629-3,784,631, 1972; (b) G. W. Lawson, (Goodyear) U.S. Pat.
3,813,453, 1974; (c) N. A. Maly, H. R. Menapace and G. S. Benner,
(Goodyear) U.S. Pat. 3,903,193, 1975.
8 H. R. Menapace, N. A. Maly, J. L. Wang and L. G. Wideman, J. Org.
Chem., 1975, 40, 2983.
9 M. J. Hanton, L. Daubney, T. Lebl, S. Polas, D. M. Smith and A.
Willemse, Dalton Trans., 2010, 10.1039/c0dt00106f.
10 D. E. Hendriksen, (Exxon) U.S. Pat. 5,059,739, 1991.
11 M. J. Hanton, R. P. Tooze, (Sasol Technology UK) WO2005089940,
2005.
X-Ray crystallography
Diffraction experiments (Table 4) were carried out on a Siemens
3-circle diffractometer with a SMART 1000 CCD area detector,
˚
using graphite-monochromated Mo-Ka radiation (l = 0.71073 A)
and Cryostream (Oxford Cryosystems) open-flow N2 cryostats.
Reflection intensities were corrected for absorption by semi-
empirical method based on Laue equivalents;47 the structures
were solved by Patterson method and refined by full-matrix least
squares against F2 of all data, using SHELXTL software.48
12 H. Olivier and P. Laurent-Ge´rot, J. Mol. Catal. A: Chem., 1999, 148,
43.
7044 | Dalton Trans., 2010, 39, 7038–7045
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
13 (a) M. J. Hanton, P. W. Dyer, R. P. Tooze, and W. R. H. Wright,
unpublished results.; (b) W. R. H. Wright, Ph.D. Thesis, Durham
University, 2009.
30 T. Gelbrich, J. Sieler and U. Dumichen, Z. Kristallogr., 2000, 215, 127.
31 H.-S. Sun, X.-M. Wang, X.-Z. You and X.-Y. Huang, Polyhedron, 1995,
14, 2159.
32 F. H. Allen and R. Taylor, Chem. Soc. Rev., 2004, 33, 463.
33 (a) T. Kreickmann, S. Arndt, R. R. Schrock and P. Mu¨ller,
Organometallics, 2007, 26, 5702; (b) S. Arndt, R. R. Schrock and P.
Mu¨ller, Organometallics, 2007, 26, 1279.
34 T. J. Crevier and J. M. Mayer, Angew. Chem., Int. Ed., 1998, 37, 1891.
35 V. C. Gibson, C. Redshaw, G. L. P. Walker, J. A. K. Howard, V. J. Hoy,
J. M. Cole, L. G. Kuzmina and D. S. De Silva, J. Chem. Soc., Dalton
Trans., 1999, 161.
14 S. Tobisch, Dalton Trans., 2008, 2120.
15 R. R. Schrock, R. T. DePue, J. Feldman, K. B. Yap, D. C. Yang, W. M.
Davis, L. Park, M. DiMare, M. Schofield, J. T. Anhaus, E. Walborsky,
E. Evitt, C. Kruger and P. Betz, Organometallics, 1990, 9, 2262.
16 D. C. Bradley, R. J. Errington, M. B. Hursthouse, R. L. Short, B. R.
Ashcroft, G. R. Clark, A. J. Nielson and C. E. F. Rickard, J. Chem.
Soc., Dalton Trans., 1987, 2067.
17 (a) See for example: D. Wang, R. Kro¨ll, M. Mayr, K. Wurst and M. R.
Buchmeiser, Adv. Synth. Catal., 2006, 348, 1567; (b) A. J. Nielson,
M. W. Glenny, C. E. F. Rickard and J. M. Waters, J. Chem. Soc., Dalton
Trans., 2000, 4569.
18 (a) D. E. Wigley, Prog. Inorg. Chem., 1994, 42, 239; (b) D. del R´ıo, F.
Montilla, A. Pastor, A. Galindo, A. Monge and E. Gutie´rrez-Puebla,
J. Chem. Soc., Dalton Trans., 2000, 2433.
36 A search of the May 2009 release of the Cambridge Structural Database
(F. H. Allen, and R. Taylor, Chem. Soc. Rev., 2004, 33, 463) for all
species containing W–CH3 moieties indicated that for such complexes
˚
W–C bond distances of between 2.05 and 2.36 A are typical.
37 (a) C. Lorber, R. Choukroun and L. Vendier, Organometallics, 2008,
27, 5017; (b) L. H. Doerrer, A. J. Graham, D. Haussinger and M. L. H.
Green, J. Chem. Soc., Dalton Trans., 2000, 813.
38 D. S. Williams, J. T. Anhaus, M. H. Schofield, R. R. Schrock and W. M.
Davis, J. Am. Chem. Soc., 1991, 113, 5480.
39 S. Tobisch, Organometallics, 2007, 26, 6529.
40 T. Agapie, J. A. Labinger and J. E. Bercaw, J. Am. Chem. Soc., 2007,
129, 14281; T. Agapie, S. J. Schofer, J. A. Labinger and J. E. Bercaw,
J. Am. Chem. Soc., 2004, 126, 1304; D. S. McGuinness, Organometallics,
2009, 28, 244; M. J. Overett, K. Blann, A. Bollmann, J. T. Dixon, D.
Haasbroek, E. Killian, H. Maumela, D. S. McGuinness and D. H.
Morgan, J. Am. Chem. Soc., 2005, 127, 10723.
41 K. Maruyama, T. Kuroki, T. Mizoroki and A. Ozaki, Bull. Chem. Soc.
Jpn., 1971, 44, 2002.
42 J. Soto, M. L. Steigerwald and R. H. Grubbs, J. Am. Chem. Soc., 1982,
104, 4479.
43 T. Agapie, J. A. Labinger and J. E. Bercaw, J. Am. Chem. Soc., 2007,
129, 14281.
44 T. Agapie, S. J. Schofer, J. A. Labinger and J. E. Bercaw, J. Am. Chem.
Soc., 2004, 126, 1304.
45 A. K. Tomov, V. C. Gibson, G. J. P. Britovsek, R. J. Long, M. van
Meurs, D. J. Jones, K. P. Tellmann and J. J. Chiros, Organometallics,
2009, 28, 7033.
19 D. S. Williams, M. H. Schofield, J. T. Anhaus and R. R. Schrock, J. Am.
Chem. Soc., 1990, 112, 6728.
20 P. W. Dyer, V. C. Gibson, J. A. K. Howard, B. Whittle and C. Wilson,
J. Chem. Soc., Chem. Commun., 1992, 1666.
21 W. A. Nugent, and J. M. Mayer, Metal–Ligand Multiple Bonds, Wiley-
Inter Science, New York, 1988.
22 V. C. Gibson, E. L. Marshall, C. Redshaw, W. Clegg and M. R. J.
Elsegood, J. Chem. Soc., Dalton Trans., 1996, 4197.
23 (a) F. Song, R. D. Cannon, S. J. Lancaster and M. Bochmann, J. Mol.
Catal. A: Chem., 2004, 218, 21; (b) W. Kaminsky, M. Arndt-Rosenau,
in Applied Homogeneous Catalysis with Organometallic Compounds,
B. Cornils, W. A. Herrmann, ed.; VCH, New York, 2nd Edtn., 2002,
pp. 213-229.
24 Although adducts of MexAlCl(3-x) (x = 0 - 3) with DME have been
reported, no spectroscopic or structural details were provided: N. C.
Means, C. M. Means, S. G. Bott and J. L. Atwood, Inorg. Chem., 1987,
26, 1466.
25 A. A. Addison, T. N. Rao, J. Reedijk, J. van Rijn and G. C. Verschoor,
J. Chem. Soc., Dalton Trans., 1984, 1349.
26 S. Alvarez, P. Alemany, D. Casanova, J. Cirera, M. Llunell and D.
Avnir, Coord. Chem. Rev., 2005, 249, 1693.
27 W. A. Nugent, R. J. McKinney, R. V. Kasowski and F. A. Van-Catledge,
Inorg. Chim. Acta, 1982, 65, L91.
28 P. Binger, F. Sandmeyer and C. Kruger, Organometallics, 1995, 14,
2969.
46 Extractive Ion Trace GC-MS revealed complete scrambling of both the
unoligomerised ethene and the C4-fraction.
47 G. M. Sheldrick, SADABS, version 2.10, Bruker AXS, Madison,
Wisconsin, USA, 2003.
29 C. J. Harlan, B. M. Bridgewater, T. Hascall and J. R. Norton,
Organometallics, 1999, 18, 3827.
48 G. M. Sheldrick, SHELXTL, version 6.4, Bruker AXS, Madison,
Wisconsin, USA, 2003.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 7038–7045 | 7045
©