Synthesis and structure of boron-bridged constrained geometry complexes of titanium

Article abstract of DOI:10.1039/b315708c

The boron-bridged constrained geometry titanium complexes [Ti{η⁵:η¹-(C₅H₄)B(NR ₂)NPh}(NMe₂)₂] [R = iPr (3), SiMe₃ (4)] and [Ti{η⁵:η¹-(C₉

Full text of DOI:10.1039/b315708c

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis and structure of boron-bridged constrained geometry
complexes of titanium
Holger Braunschweig,*a Frank M. Breitling,a,b Carsten von Koblinski,c Andrew J. P. White†b
and David J. Williamsb
a Institut für Anorganische Chemie, Julius-Maximilians-Universität Würzburg, Am Hubland,
D-97074 Würzburg, Germany
b Imperial College London, Exhibition Road, London, UK SW7 2AZ
c Institut für Anorganische Chemie, RWTH Aachen, Templergraben 55, D-52056 Aachen,
Germany
Received 3rd December 2003, Accepted 5th February 2004
First published as an Advance Article on the web 13th February 2004
The boron-bridged constrained geometry titanium complexes [Ti{η5:η1-(C5H4)B(NR2)NPh}(NMe2)2] [R = iPr (3),
SiMe3 (4)] and [Ti{η5:η1-(C9H6)B(NiPr2)NPh}(NMe2)2] (12) have been prepared in good yields by amine elimination
reaction from [Ti(NMe2)4]. Subsequent deamination–chlorination with excess Me3SiCl yielded the corresponding
dichloro-complexes (5, 6, 13). Reaction of the analogous ligand precursors (C5H5)B(NiPr2)N(H)R (R = Cy, tBu)
with [Ti(NMe2)4] did not result in the expected bridged compounds, but rather in the half-sandwich complexes
[Ti{(η5-C5H4)B(NiPr2)N(H)R}(NMe2)3] [R = Cy (9), tBu (10)]. All compounds were fully characterised by means
of multinuclear NMR spectroscopy. Thorough investigation of substituent effects was achieved by comparative
X-ray diffraction studies on complexes 3, 5, 6 and 12.
Introduction
Group IV metallocenes have been first studied in polymeris-
ation research because they can be used to model the active sites
in the classically heterogeneous Ziegler–Natta type polymeris-
ation of olefins in a homogeneous environment. With the
discovery of methylaluminoxane (MAO) as a highly efficient
activator, the interest in Group IV metallocenes was shifted
from mechanistic studies towards utilisation of these com-
pounds as catalysts on an industrial scale. Due to uncertainties
about the structure of MAO, and consequently its interaction
with the metallocenes, a clear understanding of the poly-
merisation process remained elusive.1 In 1990, Bercaw pre-
sented the well-defined scandium complex I that is isoelectronic
with the proposed cationic active species in Group IV mediated
polymerisations of olefins and is itself an active polymerisation
catalyst even in the absence of a co-catalyst.2
Cp(centroid)–M–N angle and is hence, a crucial parameter in
the polymerisation reaction. Extending our synthetic method
for the preparation of boron-bridged ligand precursors5 and
corresponding metallocenes6,7 that are stabilised by an amino
substituent such as IV and V, we recently communicated the
first boron-bridged CGC of titanium.8,9 The very short and
rigid bridging moiety in these compounds and its poten-
tially electron withdrawing effect may result in advantageous
catalytic properties.10
In due course, groups both in industry and academia
employed the same general ligand framework (II) to synthesise
corresponding Group IV constrained geometry complexes
(CGC) such as III. Complexation was usually achieved by reac-
tion of the ligand precursor with a suitable metal compound via
a dilithiation-salt elimination sequence or amine/alkane elimin-
ation. CGC of Group IV metals are known to be active olefin
polymerisation catalysts when activated with MAO or other
co-catalysts.3
In the present paper we report the synthesis of a series of
novel boron-bridged CGC of titanium and comparative struc-
tural studies of these compounds by means of X-ray analysis.
The industrial interest in these new systems can be generally
attributed to the unique properties of the polymers obtained.
The combination of
a decreased tendency to undergo
β-hydride elimination and an ability to readily incorporate
higher α-olefins results in linear polymers with long sidechains.
The linear low density polyethylene (LLDPE) produced com-
bines the high processability of low density polyethylene
(LDPE), obtained from the ICI radical process, with the
strength and toughness of metallocene produced high density
polyethylene (HDPE).3a,4
Beside efforts to tune polymerisation characteristics by
modification of the cyclopentadienyl (Cp) and amido frag-
ments of the ligand framework, a variety of different bridging
moieties were reported.3b,c The bridging moiety affects the
Results and discussion
Synthesis and characterisation of new boron-bridged
cyclopentadienyl CGC
We have previously reported the synthesis of boron-bridged
CGC type ligands.5b–d The incorporating zirconocene,
[Zr{η5:η1-(C9H6)B(NiPr2)NPh}2], has been prepared via a
dilithiation-salt elimination sequence which demonstrated the
stability of the boron bridge against nucleophilic attack by
lithium alkyl reagents.5b Nevertheless, dilithiation of the
ligand precursors (C5H5)B(NiPr2)N(H)Ph (1) and (C9H7)-
B(NiPr2)N(H)Ph (11) with 2 equiv. of LinBu, reaction with
[TiCl3(thf )3] and subsequent oxidation by 0.5 equiv. of PbCl2
† X-Ray crystallography.
D a l t o n T r a n s . , 2 0 0 4 , 9 3 8 – 9 4 3
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4
938

View Article Online
always resulted in product mixtures that could not be
separated.11
An amine elimination route, employing [Ti(NMe2)4], proved
to be more fruitful. The reaction of ligand precursors (C5H5)-
B(NiPr2)N(H)Ph (1) and (C5H5)B{N(SiMe3)2}N(H)Ph (2) with
[Ti(NMe2)4] resulted in the formation of the boron-bridged
CGC 38 and 4 (Scheme 1). When [Ti(NMe2)4] was added to a
hexane solution of the respective ligand precursor at Ϫ78 ЊC,
no reaction in terms of colour change from yellow to dark
orange and release of HNMe2 was observed below Ϫ10 ЊC.
Yields after recrystallisation were good. Subsequent reaction
with Me3SiCl gave the corresponding dichloro-complexes 5 and
6 in good yields (Scheme 1). Compounds 3, 5 and 6 were
obtained as orange or red crystals, whereas 4 was obtained as a
dark red oil. The compounds proved to be air- and moisture-
sensitive, very soluble in CH2Cl2 and moderately soluble in
toluene and hexane.
Scheme 2
might be both the result of increased steric bulk of the substi-
tuents R compared to a phenyl group and a significantly lower
acidity of the alkyl-NH in comparison to the phenyl-NH group.
The formulations of 9 and 10 are evident from integration
1
ratios in the H NMR spectra as well as from the IR spectra
that display characteristic NH stretching bands at 3441 and
3456 cmϪ1, respectively, for 9 and 10 (in comparison to 3435
and 3454 cmϪ1, respectively, for the ligand precursors 7 and
813).
The same synthetic route as outlined above was extended to
the preparation of the indenyl CGC 12 and 13 (Scheme 3). In
the first step, elevated temperatures and extended reaction times
had to be applied to obtain sufficient conversion. While the Cp
complexes 3 and 4 were formed very smoothly, the synthesis of
the indenyl derivative 12 suffered from the formation of signifi-
cant amounts of undesired products of the type [LTi(NMe2)3],
where only the indenyl or the amido moiety of the newly intro-
duced ligand is coordinated to the metal centre. The yield was
accordingly modest, though after two subsequent recrystallis-
ation steps the analytical pure target compound 12 could
be obtained in the form of orange crystals. Deamination–
chlorination of 12 with Me3SiCl gave the dichloro complex 13
in high yield as a red microcrystalline material. 11B NMR chem-
ical shifts of 12 and 13 at δ 27.7 and 28.0, respectively, again
resemble the value observed for the ligand precursor.5b 1H and
13C NMR spectra of 12 and 13 were unremarkable and showed
Scheme 1
Cp complexes 3–6 were characterised by multinuclear NMR
techniques. The 11B NMR chemical shifts of δ 27.8 and 28.4,
respectively, for the iPr2N substituted complexes 3 and 5, and
at δ 33.0 and 33.1, respectively, for the (Me3Si)2N substituted
complexes 4 and 6, closely resemble those of the respective lig-
and precursors.5d The relative down field shift for the (Me3Si)2N
substituted compounds 4 and 6, with respect to those com-
pounds with the iPr2N substituent on boron, can be attributed
to the deshielding effect of the Me3Si groups. 1H and 13C NMR
spectra for 3–5 display two signals each for the methine groups
of the Cp moiety, indicating approximate Cs symmetry in solu-
1
tion. In contrast, the H and 13C NMR spectra of 6 show four
sets of signals for the methine groups and the 29Si NMR spec-
trum displays two signals indicating C1 symmetry in solution.
Such a geometry would appear to arise from a torsion of the
N(SiMe3)2 fragment (vide infra) and hindered rotation about
the B–N(SiMe3)2 axis. The signals for the ipso-C of the Cp
ring could not be detected in the room temperature 13C NMR
spectra of 3–6 due to quadrupolar 13C–11B coupling.12
In the case of ligand precursors (C5H5)B(NiPr2)N(H)R [R =
Cy (7), tBu (8)] performing the reaction in toluene under reflux
conditions resulted only in the formation of the non-bridged
half-sandwich complexes 9 and 10 (Scheme 2). This observation
Scheme 3
D a l t o n T r a n s . , 2 0 0 4 , 9 3 8 – 9 4 3
939

View Article Online
all of the expected signals apart from the signal for the ipso-C
of the indenyl moiety.
X-Ray crystallographic analysis of complexes 3, 5, 6 and 12
The molecular structures of 5, 6 and 12 as determined by single
crystal X-ray analysis are shown in Figs. 1–3, respectively. Com-
plex 5 crystallised with two independent molecules (A and B) in
the asymmetric unit each having essentially identical geometries
– the rms deviation of the best fit between the two molecules is
only 0.058 Å. The gross structures of the previously reported
species 3,8 and complexes 5, 6 and 12 are very similar with each
exhibiting a characteristic angling of the B–C(1) bond to the
[C5] ring plane (Table 1). There are, however, a few notable
differences between the four structures.
Fig. 3 The molecular structure of 12. Selected bond lengths (Å)
and angles (Њ): Ti–N(3) 1.915(2), Ti–N(4) 1.918(3); C5–Ti–N(3) 115.7,
C5–Ti–N(4) 123.8.
an angle (ꢀ) of ca. 3–5Њ to the normal to the ring plane. The
Ti–C5 ring centroid and Ti–N(1) bond lengths in 3 and 12 are
significantly longer than those observed in 5 and 6 reflecting the
very different electronic environments of titanium centres bear-
ing two NMe2 ligands compared to those with two chlorides. In
all four structures both the B and N(2) centres have trigonal
planar geometries, but whereas in 3, 5 and 12 these two trigonal
planes are nearly coplanar, in 6 the steric bulk of the SiMe3
substituents forces a twist of ca. 38Њ between them. This twist,
however, is not accompanied by any significant lengthening
of the B–N(2) bond, though there is an increased bend of the
B–C(1) vector out of the [C5] ring plane, and a noticeably flatter
geometry around N(1).
In the structure of 5 adjacent independent molecules are
linked by C–H ؒ ؒ ؒ π interactions to form an extended chain
that propagates along the crystallographic b axis (Fig. 4). No
significant packing interactions are observed between adjacent
molecules in the crystals of 6, the closest intermolecular contact
being the approach of the C(4)–H proton in one molecule to
the Cl(1) centre in another [C ؒ ؒ ؒ Cl 3.74 Å, H ؒ ؒ ؒ Cl 2.84 Å,
C–H ؒ ؒ ؒ Cl 158Њ]. There are no noteworthy intermolecular
interactions in the crystals of 12.
Fig.
1
The structure of one of the two crystallographically
independent molecules (A) present in the crystals of 5. Selected bond
lengths (Å) and angles (Њ) (values for the second independent molecule
in square brackets): Ti–Cl(1) 2.260(2) [2.2588(14)], Ti–Cl(2) 2.2677(15)
[2.2707(15)]; C5–Ti–Cl(1) 117.1 [118.5], C5–Ti–Cl(2) 118.2 [118.6].
Fig. 4 Part of one of the extended C–H ؒ ؒ ؒ π linked chains of
alternating A and B type molecules present in the crystals of 5. The
hydrogen bonding geometries [H ؒ ؒ ؒ π] (Å) and [C–H ؒ ؒ ؒ π] (Њ) are
(a) 2.81, 155 and (b) 2.84, 158.
Fig. 2 The molecular structure of 6. Selected bond lengths (Å) and
angles (Њ): Ti–Cl(1) 2.2385(12), Ti–Cl(2) 2.2553(13); C5–Ti–Cl(1) 120.3,
C5–Ti–Cl(2) 117.1.
If one considers the C5 aromatic rings in all four complexes
as η1-moieties, the geometry at titanium can be viewed as dis-
torted tetrahedral with angles in the range 99.2(1)–123.82(11)Њ,
the most acute angle in each case being associated with the bite
of the chelating ligand. In each structure the metal centre is
slightly displaced (δ) from a position directly underneath the C5
ring centroid such that the Ti–C5 ring centroid vector subtends
Conclusions
A series of new boron-bridged constrained geometry complexes
of titanium, [Ti{η5:η1-(CxHy)B(NR2)NPh}RЈ2], has been pre-
pared and some have been structurally characterised. It has
been demonstrated that the chelating ligand framework can
readily be altered by modification of the η5-coordinated frag-
D a l t o n T r a n s . , 2 0 0 4 , 9 3 8 – 9 4 3
940

View Article Online
Table 1 Comparative selected geometric parameters for complexes 3,8 5, 6 and 12
3
5 (mol. A)
5 (mol. B)
6
12
[R1 = C(12),
R2 = C(15)]
2.027(3)
4.4
[R1 = C(12),
R2 = C(15)]
1.989(4)
3.1
[R1 = C(12Ј),
R2 = C(15Ј)]
1.995(4)
3.5
[R1 = Si(12),
R2 = Si(16)]
1.992(4)
3.5
[R1 = C(19),
R2 = C(16)]
2.054(3)
4.9
Ti–C5 a/Å
ꢀb/Њ
δc/Å
0.16
0.11
0.12
0.12
0.18
Ti–N(1)/Å
B–N(1)/Å
B–N(2)/Å
B–C(1)/Å
2.020(2)
1.428(3)
1.409(3)
1.603(3)
99.2(1)
131.0(2)
103.57(18)
125.5(2)
360.0
1.937(3)
1.464(6)
1.382(6)
1.602(6)
99.7(2)
132.9(4)
99.4(3)
127.6(4)
359.9
1.930(4)
1.448(6)
1.386(6)
1.605(6)
99.7(2)
131.4(4)
100.1(3)
128.5(4)
360.0
1.941(3)
1.436(5)
1.419(5)
1.595(5)
99.73(9)
130.0(3)
101.8(3)
127.5(3)
359.3
2.020(2)
1.428(4)
1.414(4)
1.610(4)
99.86(10)
130.6(3)
104.0(2)
125.4(3)
360.0
N(1)–Ti–C5 a/Њ
N(1)–B–N(2)/Њ
N(1)–B–C(1)/Њ
N(2)–B–C(1)/Њ
Σ[X–B–Y]d/Њ
∆Be/Å
0.006
0.016
0.021
0.064
0.004
B–N(2)–R1/Њ
B–N(2)–R2/Њ
R1–N(2)–R2/Њ
Σ[X–N(2)–Y]d/Њ
∆N(2)e/Å
123.2(2)
122.0(2)
114.8(2)
360.0
0.009
2.1
102.4(2)
124.2(2)
131.7(2)
358.3
0.115
78.2
122.8(4)
121.8(3)
115.3(3)
359.9
0.001
1.7
105.8(3)
119.5(2)
133.4(3)
358.7
0.102
72.0
123.1(4)
121.6(4)
115.2(3)
359.9
0.020
1.9
105.8(3)
119.3(2)
133.6(3)
358.7
0.103
73.8
122.7(2)
115.5(2)
121.6(2)
359.8
0.043
38.0
102.5(2)
127.7(2)
129.8(3)
360.0
0.015
41.2
122.1(2)
123.5(2)
114.4(2)
360.0
0.012
1.3
102.2(2)
124.7(2)
131.8(2)
358.7
0.103
77.2
τ[B–N(2)]f/Њ
Ti–N(1)–B/Њ
Ti–N(1)–Ph/Њ
B–N(1)–Ph/Њ
Σ[X–N(1)–Y]d/Њ
∆N(1)e/Å
τ[N(2)–Ph]f/Њ
B–C(1) ؒ ؒ ؒ [C5]g/Њ
∆B[C5]h/Å
148.6
0.848
148.2
0.853
148.9
0.860
145.3
0.919
148.2
0.878
a “C5” refers to the centroid of the five-membered aromatic ring. b ꢀ Is the angle between the normal to the [C5] ring plane and the Ti ؒ ؒ ؒ C5 ring
centroid vector. c δ Is the offset of the Ti centre away from a position perpendicularly beneath the C5 ring centroid. d Sum of the angles around the
central atom. e ∆X is the deviation of atom X from the plane of its substituents. f Torsion angle about linkage. g Angle of B–C(1) bond to [C5] ring
plane. h Deviation of B from [C5] plane.
ment (cyclopentadienyl vs. indenyl) or the boron bridge [BNiPr2
vs. BN(SiMe3)2], in order to tune the electronic and steric char-
acteristics of the complexes. Increasing the steric demand of the
amido fragment in the chelating ligand by substituting phenyl
for cyclohexyl or tert-butyl was not feasible, as the reaction of
the respective ligand precursors with [Ti(NMe2)4] resulted only
in the formation of non-bridged species. In preliminary experi-
ments, the reported dichloro complexes could be activated with
MAO for the polymerisation of ethylene and styrene. Further
experiments to establish the influence of variations in the
described ligand framework on the polymerisation character-
istics are in progress.
spectrometer. Elemental analyses were recorded on a Carlo-
Erba elemental analyzer, model 1106.
Preparations
[Ti{ꢀ5:ꢀ1-(C5H4)B(NiPr2)NPh}(NMe2)2] (3). (C5H5)B(NiPr2)-
N(H)Ph (0.83 g, 3.09 mmol) was dissolved in 15 mL toluene
and [Ti(NMe2)4] (0.69 g, 3.09 mmol) in 5 mL hexane was added
dropwise at Ϫ78 ЊC. The yellow solution was stirred for 20 min
at Ϫ78 ЊC, before warming it slowly to ambient temperature.
At Ϫ10 ЊC, the colour changed from yellow to orange and
formation of HNMe2 was observed. After 30 min at ambient
temperature, the mixture was heated to 35 ЊC for 15 min. The
solvent was removed in vacuo, the red residue dissolved in
30 mL hexane and stored at Ϫ30 ЊC to yield pure 3 as orange
crystals (1.00 g, 2.49 mmol, 80%).
Experimental
General considerations
Analytical data for 3 have been previously reported.8
All manipulations were carried out under a dry nitrogen
atmosphere using common Schlenk techniques. Solvents were
dried by standard procedures, distilled, and stored under nitro-
gen over molecular sieves. [Ti(NMe2)4] and Me3SiCl were used
as supplied without further purification. Ligand precursors
were synthesised as previously reported.5b–d
[Ti{ꢀ5:ꢀ1-(C5H4)B(N(SiMe3)2)NPh}(NMe2)2] (4). The same
procedure was employed as for 3. After removing all volatile
components, 4 was obtained as a dark red oil (2.10 g, 4.55
mmol, 99%).
1H NMR (CDCl3): δ 0.11 (s, 18H, SiMe3), 3.09 (s, 12H,
NMe2), 5.97 (m, 2H, CHCp), 6.46 (m, 2H, CHCp), 6.8–7.3
(m, 5H, CHPh). 13C NMR (toluene-d8): δ 3.93 (SiMe3), 47.44
(NMe2), 115.56, 120.05 (CHCp), 121.65, 124.03, 128.71 (CHPh),
153.24 (ipso-CPh). 11B NMR (toluene-d8): δ 33.0. 29Si NMR
(toluene-d8): δ 0.27. MS (CI); m/z (%): 329 (8) [MϩϪ Ti(NMe2)2
ϩ 3H] and ligand fragments (Found: C, 54.20; H, 8.31; N,
12.01. C21H39BN4Si2Ti requires: C, 54.54; H, 8.50; N 12.12%).
NMR experiments were performed on either a Varian Unity
500 spectrometer, a Bruker DPX-400 spectrometer, a JEOL-
EX270 Delta Upgrade spectrometer or a Bruker Mercury
1
200 spectrometer. Chemical shifts for H and 13C-{1H} NMR
spectra were referenced to internal solvent resonances and
are reported relative to tetramethylsilane. Chemical shifts for
11B-{1H} and 29Si-{1H} NMR spectra were referenced to BF3ؒ
OEt2 and tetramethylsilane, respectively, as external standards.
Routine coupling constants for 1H NMR spectra are not
reported. Mass spectra were recorded either on a Finnigan
MAT 95, a Thermo Finnigan Trio 1000, or a Micromass
Platform II spectrometer. IR spectroscopy was conducted on
CH2Cl2 solutions and performed on a Bruker Vector 22 FT-IR
[Ti{ꢀ5:ꢀ1-(C5H4)B(NiPr2)NPh}Cl2] (5). A solution of 3
(0.17 g, 0.42 mmol) in 10 mL hexane was reacted with Me3SiCl
(0.50 g, 4.60 mmol) at 0 ЊC. The mixture was allowed to come to
ambient temperature and stirred for 16 h. The yellow–orange
precipitate was filtered off, washed with 10 mL hexane and
D a l t o n T r a n s . , 2 0 0 4 , 9 3 8 – 9 4 3
941

View Article Online
dried in vacuo. Recrystallisation from CH2Cl2 at 4 ЊC yields 5
(0.16 g, 0.42 mmol, 98%) as orange crystals.
6.3–7.7 (m, 11H, CHInd/Ph). 13C NMR (CDCl3): δ 21.35, 21.66,
27.20, 27.71 (Me), 44.61, 47.03 (CHiPr), 116.38, 121.78, 125.14,
126.16, 127.53, 127.95, 128.54, 129.18, 129.39 (CHInd/Ph),
133.45, 134.80 (quaternary CInd), 151.82 (ipso-CPh). 11B NMR
(benzene-d6): δ 28.0. MS (EI); m/z (%): 314 (100) [Mϩ Ϫ TiCl2 Ϫ
2H] (Found: C, 58.05; H, 5.65; N, 6.18. C21H25BCl2N2Ti
requires: C, 57.98; H, 5.79; N 6.44%).
Analytical data for 5 have been previously reported.8
[Ti{ꢀ5:ꢀ1-(C5H4)B(N(SiMe3)2)NPh}Cl2] (6). The same pro-
cedure was employed as for 5. Recrystallisation from hexane
at ambient temperature yields pure 6 as orange prismatic and
needle shaped crystals (0.78 g, 1.75 mmol, 68%).
1H NMR (benzene-d6): δ 0.10 (s, 9H, SiMe3), 0.20 (s, 9H,
SiMe3), 5.96 (m, 1H, CHCp), 6.07 (m, 1H, CHCp), 6.31 (m, 1H,
CHCp), 6.61 (m, 1H, CHCp), 6.75–7.45 (m, 5H, CHPh). 13C
NMR (benzene-d6): δ 3.74, 3.87 (SiMe3), 119.01, 121.23,
122.72, 123.08, 123.25, 126.23, 126.45 (CHCp/Ph), 152.65
(ipso-CPh). 11B NMR (benzene-d6): δ 33.1. 29Si NMR (benzene-
d6): δ 1.7, 5.6 (s, SiMe3). MS (CI); m/z (%): 445 (54) [Mϩ], 329
(29) [MϩϪ TiCl2 ϩ 3H], 236 (20) [(Me3Si)2NBCpHϩ], 162 (60)
X-Ray crystallography
Crystal data for 5. C17H23BCl2N2Ti, M = 385.0, monoclinic,
P21/c (no. 14), a = 12.575(2), b = 18.852(2), c = 16.8556(13) Å,
β = 109.247(11)Њ, V = 3772.5(8) Å3, Z = 8 (two independent
molecules), Dc = 1.356 g cmϪ3, µ(Mo-Kα) = 0.74 mmϪ1, T =
183 K, orange platy needles; 6624 independent measured reflec-
tions, F 2 refinement, R1 = 0.050, wR2 = 0.097, 3940 independ-
ent observed reflections [|Fo| > 4σ(|Fo|), 2θmax = 50Њ], 391
parameters.
ϩ
[(Me3Si)2NH2 ] (Found: C, 45.44; H, 5.98; N, 6.30. C17H27-
BCl2N2Si2Ti requires: C, 45.86; H, 6.11; N 6.29%).
[Ti{(ꢀ5-C5H4)B(NiPr2)N(H)Cy}(NMe2)3] (9). (C5H5)B(Ni-
Pr2)N(H)Cy (0.75 g, 2.73 mmol) dissolved in 20 mL of toluene
was treated at 0 ЊC with [Ti(NMe2)4] (0.61 g, 2.73 mmol). The
reaction mixture was refluxed for 3 h while the colour changed
from yellow to dark red. The solvent was removed in vacuo to
give 9 as a dark red oil (1.23 g, 2.70 mmol, 99%) that is virtually
free from impurities.
Crystal data for 6. C17H27BCl2N2Si2Ti, M = 445.2, mono-
clinic, P21/c (no. 14), a = 7.1231(11), b = 19.0371(10), c =
17.219(2) Å, β = 97.423(14)Њ, V = 2315.4(5) Å3, Z = 4, Dc = 1.277
g cmϪ3, µ(Cu-Kα) = 6.26 mmϪ1, T = 203 K, orange blocky
needles; 3423 independent measured reflections, F 2 refinement,
R1 = 0.049, wR2 = 0.126, 2855 independent observed absorp-
tion corrected reflections [|Fo| > 4σ(|Fo|), 2θmax = 120Њ], 215
parameters.
1H NMR (toluene-d8): δ 1.19 (d, 12H, MeiPr), 0.90–2.00 (m,
10H, CH2), 3.18 (s, 18H, NMe2), 3.55–3.75 (m, 2H, CHiPr), 5.93
(m, 2H, CHCp), 6.36 (m, 2H, CHCp). 13C NMR (toluene-d8):
δ 23.74 (MeiPr), 25.75, 26.28, 38.70 (CH2), 46.45 (CHCy), 50.30
(NMe2), 51.20 (CHiPr), 111.27, 119.71 (CHCp). 11B NMR (tolu-
ene-d8): δ 28.3. MS (CI); m/z (%): 454 (5) [Mϩ ϩ H], 409 (71)
[Mϩ Ϫ HNMe2], 275 (20) [Mϩ Ϫ Ti(NMe2)3 ϩ 3H]. IR/cmϪ1
3441w ν(NH) (Found: C, 60.32; H, 10.27; N, 15.23. C23H48BN5-
Ti requires: C, 60.93; H, 10.67; N 15.45%).
¯
Crystal data for 12. C25H37BN4Ti, M = 452.3, triclinic, P1
(no. 2), a = 9.6645(10), b = 9.8449(13), c = 13.8532(14) Å,
α = 88.386(10), β = 80.406(8), γ = 75.145(9)Њ, V = 1256.1(2) Å3,
Z = 2, Dc = 1.196 g cmϪ3, µ(Mo-Kα) = 0.36 mmϪ1, T = 203 K,
orange blocks; 4427 independent measured reflections, F 2
refinement, R1 = 0.052, wR2 = 0.132, 3445 independent observed
reflections [|Fo| > 4σ(|Fo|), 2θmax = 50Њ], 272 parameters.
[Ti{(ꢀ5-C5H4)B(NiPr2)N(H)tBu}(NMe2)3] (10). The same
procedure was employed as for 9. Recrystallisation from hexane
at Ϫ30 ЊC gave pure 10 as an orange microcrystalline material
(1.08 g, 2.87 mmol, 99%).
CCDC reference numbers 225922 (5), 225923 (6) and 225924
(12).
See http://www.rsc.org/suppdata/dt/b3/b315708c/ for crystal-
lographic data in CIF or other electronic format.
1H NMR (benzene-d6): δ 1.16 (d, 12H, MeiPr), 1.19 (s, 9H,
MetBu), 3.18 (s, 18H, NMe2), 3.51 (m, 2H, CHiPr), 5.91 (m, 2H,
CHCp), 6.46 (m, 2H, CHCp). 13C NMR (benzene-d6): δ 24.00
(MetBu), 34.05 (MeiPr), 46.4 (br, CHiPr), 49.66 (ipso-CtBu),
50.45 (NMe2), 110.98, 120.87 (CHCp). 11B NMR (benzene-d6):
δ 29.1. MS (CI); m/z (%): 383 (100) [Mϩ Ϫ NMe2], 339 (80)
[Mϩ Ϫ 2NMe2], 183 (47) [Mϩ Ϫ Ti(NMe2)3 Ϫ C5H4]. IR/cmϪ1
3456w ν(NH) (Found: C, 58.75; H, 10.57; N, 16.19. C21H46-
BN5Ti requires: C, 59.02; H, 10.85; N 16.39%).
Acknowledgements
This work was supported by BASF AG Ludwigshafen, BMBF,
DFG, EPSRC, R. Soc. F. M. B. thanks the Fonds der
Chemischen Industrie for a doctoral scholarship.
References
1 (a) H. H. Brintzinger, D. Fischer, R. Mülhaupt, B. Rieger and
R. M. Waymouth, Angew. Chem., Int. Ed. Engl., 1995, 34, 1143, and
references therein; (b) W. Kaminsky, J. Chem. Soc., Dalton Trans.,
1998, 1413, and references therein.
2 P. J. Shapiro, E. Bunel, W. Schaefer and J. E. Bercaw,
Organometallics, 1990, 9, 867.
3 (a) A. L. McKnight and R. M. Waymouth, Chem. Rev., 1998, 98,
2587, and references therein; (b) G. J. P. Britovsek, V. C. Gibson and
D. F. Wass, Angew. Chem., Int. Ed. Engl., 1999, 38, 428, and
references therein; (c) V. C. Gibson and S. K. Spitzmesser,
Chem. Rev., 2003, 103, 283, and references therein.
4 S. Chum, W. J. Kruper and M. J. Guest, Adv. Mater., 2000, 12, 1759.
5 (a) H. Braunschweig, C. v. Koblinski, M. Neugebauer, U. Englert
and X. Zheng, J. Organomet. Chem., 2001, 619, 305;
(b) H. Braunschweig, C. v. Koblinski, F. M. Breitling, K. Radacki,
C. Hu, L. Wesemann, T. Marx and I. Pantenburg, Inorg. Chim. Acta,
2003, 350, 467; (c) H. Braunschweig, M. Kraft, M. Homberger,
F. M. Breitling, A. J. P. White, U. Englert and D. J. Williams,
Appl. Organomet. Chem., 2003, 17, 421; (d ) H. Braunschweig,
F. M. Breitling, M. Homberger, C. v. Koblinski, A. J. P. White and
D. J. Williams, Z. Anorg. Allg. Chem., 2003, 629, 2244.
[Ti{ꢀ5:ꢀ1-(C9H6)B(NiPr2)NPh}(NMe2)2] (12). The same
procedure was employed as for 3. Recrystallisation twice from
hexane at Ϫ30 ЊC and subsequently 4 ЊC afforded 12 (1.25 g,
2.77 mmol, 55%) as orange crystals.
1H NMR (benzene-d6): δ 0.91, 1.03 (br d, 3H, MeiPr), 1.64 (br
d, 6H, MeiPr), 2.29 (s, 6H, NMe2), 3.08 (s, 6H, NMe2), 3.28 (m,
1H, CHiPr), 3.73 (m, 1H, CHiPr), 6.2–7.8 (m, 11H, CHInd/Ph). 13
C
NMR (benzene-d6): δ 22.81, 22.03, 26.98, 27.67 (MeiPr), 44.53
(CHiPr), 46.40 (NMe2), 46.99 (CHiPr), 49.48 (NMe2), 107.38,
120.44, 121.41, 123.23, 123.41, 123.61, 124.35, 124.97, 126.81
(CHInd/Ph), 128.78, 132.00 (quaternary CInd), 155.37 (ipso-CPh).
11B NMR (benzene-d6): δ 27.7. MS (CI); m/z (%): 319 (40)
[Mϩ Ϫ Ti(NMe2)2 ϩ 3H] (Found: C, 65.89; H, 8.03; N, 12.22.
C25H37BN4Ti requires: C, 66.39; H, 8.25; N 12.39%).
[Ti{ꢀ5:ꢀ1-(C9H6)B(NiPr2)NPh}Cl2] (13). The same procedure
was employed as for 5. Recrystallisation from toluene at Ϫ30 ЊC
yields pure 13 as red microcrystals (0.17 g, 0.39 mmol, 87%).
1H NMR (benzene-d6): δ 0.74 (d, 3H, Me), 0.79 (d, 3H, Me),
1.44 (br d, 6H, Me), 3.13 (m, 1H, CHiPr), 3.39 (m, 1H, CHiPr),
6 (a) H. Braunschweig, C. v. Koblinski and M. O. Kristen, Ger. Pat.
Appl., DE 19858016A1, 1998; (b) H. Braunschweig, C. v. Koblinski
and R. Wang, Eur. J. Inorg. Chem., 1999, 69; (c) H. Braunschweig,
D a l t o n T r a n s . , 2 0 0 4 , 9 3 8 – 9 4 3
942

View Article Online
C. v. Koblinski, M. Mamuti, U. Englert and R. Wang, Eur. J. Inorg.
Chem., 1999, 1899.
7 A. J. Ashe, III, X. Fang and J. W. Kampf, Organometallics, 1999, 18,
2288.
8 H. Braunschweig, C. v. Koblinski and U. Englert, Chem. Commun.,
2000, 1049.
9 For reviews on boron-bridged Group IV metallocenes and related
compounds, see: (a) P. J. Shapiro, Eur. J. Inorg. Chem., 2001, 321;
(b) H. Braunschweig, F. M. Breitling, E. Gullo and M. Kraft,
J. Organomet. Chem., 2003, 680, 31; (c) S. Aldridge and C. Bresner,
Coord. Chem. Rev., 2003, 244, 71.
10 J. A. Klang and D. B. Collum, Organometallics, 1988, 7,
1532.
11 C. v. Koblinski, PhD Thesis, RWTH Aachen, 2000.
12 B. Wrackmeyer, Prog. NMR Spectrosc., 1979, 12, 227.
13 H. Braunschweig and F. M. Breitling, unpublished results.
D a l t o n T r a n s . , 2 0 0 4 , 9 3 8 – 9 4 3
943