Synthesis of new mixed ring zirconocenes and evidence of ring loss

Article abstract of DOI:10.1016/S0022-328X(97)00786-9

The synthesis of the substituted silyl cyclopentadiene ligand [C₅H₅-Si(CH₃)₂(p-C₆H ₄Br)] (Cp^SiH) (1) is described. Deprotonation of (1) gives (Cp^SiLi) (2) which reacts with 1/2 ZrCl₄ to produce the bent, sandwich complex Zr(Cp^Si)₂Cl₂ (3). Compound (3) can be alkylated to give Zr(Cp^Si)₂(CH₃)₂ (4). Reaction of Cp*ZrCl₃·(Et₂O)_1/2 with (2) in ether produces (3) and Zr(Cp^Si)(C₅Me₅)Cl₂ (5) as the minor and major products, respectively, whereas reaction in toluene produces only 5. Et₂O and THF are found to catalyze the observed ring loss during the synthesis of 5. Alkylation of 5 produces Zr(Cp^Si)(C₅Me₅)(CH₃)₂ (6).

Full text of DOI:10.1016/S0022-328X(97)00786-9

Journal of Organometallic Chemistry 555 (1998) 1–4
Synthesis of new mixed ring zirconocenes and evidence of ring loss
Faisal A. Shafiq, David E. Richardson, James M. Boncella *
Department of Chemistry and Center for Catalysis, Uni6ersity of Florida, Gaines6ile, FL 32611, USA
Received 14 July 1997; received in revised form 27 October 1997
Abstract
The synthesis of the substituted silyl cyclopentadiene ligand [C5H5–Si(CH3)2(p-C6H4Br)] (CpSiH) (1) is described. Deprotona-
tion of (1) gives (CpSiLi) (2) which reacts with 1/2 ZrCl4 to produce the bent, sandwich complex Zr(CpSi)2Cl2 (3). Compound (3)
can be alkylated to give Zr(CpSi)2(CH3)2 (4). Reaction of Cp*ZrCl3 ·(Et2O)1/2 with (2) in ether produces (3) and Zr(-
CpSi)(C5Me5)Cl2 (5) as the minor and major products, respectively, whereas reaction in toluene produces only 5. Et2O and THF
are found to catalyze the observed ring loss during the synthesis of 5. Alkylation of 5 produces Zr(CpSi)(C5Me5)(CH3)2 (6). © 1998
Elsevier Science S.A. All rights reserved.
Keywords: Zirconium; Metallocenes; Ring loss
1. Introduction
2. Results and discussion
The area of homogeneous metallocene-catalyzed
olefin polymerization has recently received much atten-
tion, in part because ligand modifications can result in
desirable changes to the properties of the resulting
polymers [1]. Our most recent efforts in this area have
been directed toward introducing functionalities to the
cyclopentadienyl (Cp) ligand so that subsequent modifi-
cations can be performed either before or after attach-
ment to the metal center of interest [2]. Herein we
report some unusual features pertaining to the synthesis
of new metallocenes bearing a bromophenyl ring bound
to the Cp group via a silicon bridge. We found during
the course of our syntheses that Cp* rings can be lost
when attempting to prepare mixed ring metallocenes
containing this group if ethereal solvents are used in
even trace quantities.
2.1. Synthesis of (C5H5)Si(CH3)2(C6H4Br)(1) (H–CpSi)
Halogen–lithium exchange of n-BuLi with 1,4 dibro-
mobenzene [3] in Et2O followed by addition to
(CH3)2SiCl2 gives (CH3)2SiCl(C6H4Br) as a colorless oil
in near quantitative yields (Scheme 1). Its isolation is
possible, but not necessary since in situ addition of
either NaCp or LiCp in THF to this species produces a
yellow solution containing one major product and a
* Corresponding author. Tel.: +1 352 3929512; fax: +1 352
3923255.
Scheme 1.
0022-328X/98/$19.00 © 1998 Elsevier Science S.A. All rights reserved.
PII S0022-328X(97)00786-9

2
F.A. Shafiq et al. / Journal of Organometallic Chemistry 555 (1998) 1–4
minor, colorless, lower boiling fraction which can be
removed under vacuum at 100°C. The remaining yellow
oil is predominantly the desired substituted cyclopenta-
diene (C5H5)Si(CH3)2(C6H4Br) (1) (H–CpSi) as identi-
2.3. The synthesis of (Cp*)(CpSi)Zr(Cl)2 (5)
Reaction of Cp*ZrCl3 ·(Et2O)1/2 with 2 produces
(Cp*)(CpSi)Zr(Cl)2 (5) and (CpSi)2ZrCl2 (3) in as large
as a 3:1 ratio (isolated yields) when the reactions are
performed in Et2O (3 and 5, Eq. 4). Relative ratios vary
between batches but significant amounts of 3 are always
produced.
1
fed by H NMR spectroscopy and M/S (Scheme 1).
1
The H NMR of (1) displays two broadened Cp ring
resonances and we attribute this to rapid sigmatropic
shifts of either the methine proton, the silyl group or
both [4]. The spectrum is qualitatively very similar to
that of C5H5–Si(CH3)3 [5]. Sigmatropic rearrangements
are well known in related silyl substituted cyclopentadi-
enyl systems [4], and we have not investigated the
features of this fluxionality in greater detail. Compound
(1) is readily deprotonated by n-BuLi in pentane to give
the base-free lithium salt Li(C5H4)Si(CH3)2(C6H4Br) (2)
(Li–CpSi) in good
1
When the reaction is followed by H NMR in THF-d8,
3 and 5 are observed to form concurrently. Similar
results are obtained when the reaction is run at various
temperatures and reaction times. Efforts to separate 5
and 3 by chromatography or fractional crystallization
were unsuccessful. Both complexes are soluble in most
organic solvents and have similar retention times on
alumina and silica. We initially thought that 3 was
produced because varying amounts of ZrCl4 may be left
over from the synthesis of Cp*ZrCl3, but this hypothe-
sis proved to be false. In contrast, we find that reaction
of Cp*ZrCl3 · (Et2O)1/2 with 2 in refluxing toluene over
yield as an air sensitive, white solid that is insoluble in
aliphatic and aromatic hydrocarbons and soluble in
ethers (Eq. 1). Exposure of 2 to moisture produces 1 as
1
observed by H NMR.
2.2. The synthesis of (CpSi)2ZrCl2 (3) and
(CpSi)2Zr(CH3)2 (4)
The reaction of two eq of 2 with one eq of ZrCl4 in
Et2O produces (CpSi)2ZrCl2 (3) as an air-stable white
solid that is soluble in most common organic solvents
(3, Eq. 2). Alkylation of (3) with CH3MgBr produces
the dimethyl complex (CpSi)2Zr(CH3)2 (4) as a colorless
solid (4, Eq. 3). Compound 4 is air sensitive and has
solubility properties similar to 3, thus limiting isolated
5 days produces 5 exclusively (5, Eq. 5).
Toluene
Cp*ZrCl3 · (Et2O)2+2
“
5
(5)
5 day reflux
Although we have no mechanistic information re-
garding this process, we speculate that loss of LiCp* is
promoted by ethereal solvents. This hypothesis is fur-
ther supported by the observation that if even a small
amount (~4 ml) of THF is added to the toluene (~50
ml) reaction mixture, a small amount of 3 is produced,
while the reaction is complete within 3 h. Although ring
loss is not usually reported, it is certainly occurring in
the present case.
1
yields. H NMR analysis of the crude product indicates
near quantitative conversion to 4. It is noteworthy that
use of CH3MgCl does not afford complete alkylation to
the dimethyl 4 but yields the tentatively characterized
monomethyl species (CpSi)2Zr(CH3)Cl as the major
product. Use of excess CH3MgCl to drive the reaction
is ineffective and leads to decomposition.
2.4. The synthesis of (Cp*)(CpSi)Zr(CH3)2 (6)
Reaction of 5 with MeMgBr in ether produces a
white solid after workup which has been identified as
the dimethyl species (Cp*)(CpSi)Zr(CH3)2 (6) in moder-
ate isolated yield due to its high solubility (6, Eq. 6).
The spectroscopic properties of this compound are
unremarkable, and once again the use of MeMgBr as
the alkylating agent gives the best yield and most
complete reaction.

F.A. Shafiq et al. / Journal of Organometallic Chemistry 555 (1998) 1–4
3
Our interest in the synthesis of these metallocenes has
been to functionalize the C–Br bond on the phenyl
group. We anticipate that reaction of the dimethyl
species 4 and 6 with n-BuLi at low temperature in
ethereal solvents will result in halogen lithium exchange
at the C–Br position. We have attempted to perform
the halo-Li exchange reaction on compounds 4 and 6
with little success using n-BuLi as the Li source. When
the exchange reaction is carried out at low temperature,
trapping the resultant Li reagent with various elec-
trophiles gave numerous products with a high degree of
variability from sample to sample. The number of
products observed in our halogen–lithium exchange
reactions suggested that this pathway is not feasible.
Negishi et al. have recently reported on their studies of
reaction of metallocenes with alkyl lithiums in which
half-sandwich species are formed via loss of LiCp [6].
Given the ring loss that we observed during the synthe-
sis of 5, it is possible that similar reactions have
thwarted our attempts to lithiate the phenyl ring of the
CpSi ligand.
served, to which 13.77 ml (2.5M, 34.4 mmol) of n-BuLi
was added slowly to generate p-Li–C6H4Br. This mix-
ture was left at −50°C (~15 min), warmed to ambient
temperature (10 min), recooled to −50°C then added
dropwise to (CH3)2SiCl2 (5.12 ml) in Et2O (25 ml) and
stirred for an hour at room temperature. The solution
was then cooled to −78°C and a −78°C solution of a
stoichiometric amount (34.4 mmol) of NaCp or LiCp in
THF (80 ml) was added via cannula and stirred at
room temperature for 5–6 h to give a yellow solution.
The reaction was quenched by pouring it into ice/cold
water. The product was extracted from the aqueous
layer into pentane then dried with CaSO4. After filtra-
tion, the yellow solution was concentrated to leave a
yellow oil. The oil was then heated to ~95°C under
reduced pressure which removed a colorless uniden-
tified fraction and left behind a thick, yellow, oil. Yield
1
78%. H NMR (C6D6, 22°C): l -0.03 (s, 6H, Si(CH3)2);
3.34 (br s, 1H, methine); 6.38 (br s, 2H, Cp ring), 6.60
(br s, 2H, Cp ring), 7.03 (d, JH–H=8.4 Hz, 2H,
phenyl), 7.34 (d, JH–H=8.4 Hz, 2H, phenyl).
In conclusion, our results for the synthesis of the
mixed sandwich dichloride complex 5 demonstrate that
the Cp* ring can easily be displaced when ethereal
solvents are present even in small quantities. Although
reaction times are longer when toluene is used as the
solvent, loss of the Cp* ring is not observed. We also
find that functionalization of CpSi via bromo-lithium
exchange once it is on the metal does not appear to be
a viable approach for substitution at the para position
in zirconocene compounds.
3.2. [LiC5H4–Si(CH3)2(p-C6H4Br)] (CpSiLi) (2)
Compound 1 (2.00 g, 7.16 mmol) was dissolved in
~40 ml pentane and cooled to −50°C. 2.87 ml n-BuLi
(2.5 M, 7.17 mmol) was injected slowly whereupon
precipitation of a white solid was observed. The reac-
tion was allowed to warm to room temperature and left
stirring for ~5 h. The solid was cannula filtered, washed
2X with pentane, and dried under reduced pressure
leaving a flocculent white solid. Yield 70%.
3.3. Zr(CpSi)2Cl2 (3)
3. Experimental details
A flask was charged with 2 (0.900 g, 3.16 mmol) and
freshly sublimed ZrCl4 (0.366 g, 1.44 mmol) in a dry
box and suspended in cold (−50°C) Et2O. Ca. 10 ml of
THF was then added to this mixture and the suspen-
sion immediately dissolved to form a peach colored
solution that was left to stir overnight. Solvent was
removed under reduced pressure and the remaining oily
solid was redissolved in 25 ml CH2Cl2/6 ml 4 M HCl.
The product was extracted twice with CH2Cl2 and the
combined fractions were washed with ca. 10 ml H2O.
The solution was dried over CaSO4 and filtered. After
the solvent was removed under reduced pressure, the
remaining oily solid was triturated with pentane
overnight to leave a fine white solid. Analysis of the
crude reaction mixture revealed near quantitative con-
version to (3). Due to high solubility, the isolated yield
is ca. 50%. 1H NMR (C6D6, 22°C): l 0.55 (s, 12H,
Si(CH3)2); 5.78 (t, JH–H=2.6 Hz, 4H, Cp ring), 6.27 (t,
All syntheses were performed under a dry argon
atmosphere using standard Schlenk techniques.
Cp*ZrCl3 ·(Et2O)1/2 was prepared according to a pub-
lished route [7]. Tetrahydrofuran (THF), diethyl ether
(Et2O), toluene and pentane were distilled from potas-
1
sium or sodium benzophenone ketyl. H and 13C NMR
spectra were recorded on a Varian VXR-300 (300
MHz) or a General Electric QE-300 (300 MHz) spec-
trometer. Elemental analyses were performed by the
University of Florida Department of Chemistry Analyt-
ical Services or by Atlantic Microlabs, Norcross, GA,
USA. Despite repeated attempts, elemental analyses of
compounds (4–6) were consistently 1–2% low in car-
bon even though they were of high purity as judged by
1H NMR spectroscopy.
3.1. [C5H5–Si(CH3)2(p-C6H4Br)] (CpSiH) (1)
J
H–H=2.6 Hz, 4H, Cp ring), 7.09 (d, JH–H=8.4 Hz,
1,4 dibromobenzene (8.12 g, 34.4 mmol) was dis-
solved in ~50 ml of Et2O and cooled to −50°C. Once
the solution had cooled, a white suspension was ob-
4H, phenyl), 7.33 (d, JH–H=8.4 Hz, 4H, phenyl). Ele-
mental Analysis Calc: C, 43.51; H, 4.23. Found: C,
43.53; H, 4.05.

4
F.A. Shafiq et al. / Journal of Organometallic Chemistry 555 (1998) 1–4
3.4. Zr(CpSi)2(CH3)2 (4)
1.56 mmol) of a solution of CH3MgBr was syringed
into the reaction mixture which was stirred at −50°C
for 1 h and then warmed to room temperature and left
overnight. Cannula filtration to remove the salts fol-
lowed by removal of the solvent left a sticky white
solid. Recrystallization from benzene/pentane at −20
gives 6 as a fine white solid. Analysis of the crude
reaction mixture revealed near quantitative conversion
to 6. Owing to high solubility, the isolated yield is ca.
36%. 1H NMR (C6D6, 22°C): l −0.35 (s, 6H, Zr–
CH3); 0.47 (s, 6H, Si(CH3)2); 1.67 (s, 15H, Cp*); 5.60 (t,
Compound 3 (0.206 g, 0.286 mmol) was dissolved in
ca. 15 ml Et2O and cooled −78°C. Methyl magnesium
bromide, 191 ml (3.0 M soln in Et2O, 0.573 mmol) was
added dropwise and stirred 1 h at −78°C and then
warmed slowly to room temperature overnight. The
product was cannula filtered and the salts were washed
2X with Et2O. The solvent was removed from the
combined fractions to leave a colorless oil which was
washed 1X with pentane and then pumped dry leaving
a white powder. Isolated yield ca. 80%. 1H NMR
(C6D6, 22°C): l −0.13 (s, 6H, Zr–CH3), 0.31 (s, 12H,
Si(CH3)2); 5.83 (t, JH–H=2.3 Hz, 4H, Cp ring), 5.95 (t,
J
H–H=2.4 Hz, 2H, Cp ring), 6.05 (t, JH–H=2.4 Hz,
2H, Cp ring), 7.16 (d, JH–H=8.4 Hz, 2H, phenyl), 7.34
(d, JH–H=8.1 Hz, 2H, phenyl).
JH–H=2.3 Hz, 4H, Cp ring), 7.08 (d, JH–H=8.4 Hz,
4H, phenyl), 7.34 (d, JH–H=8.4 Hz, 4H, phenyl).
Acknowledgements
3.5. Zr(CpSi)(C5Me5)Cl2 (5)
We acknowledge the Mobil Chemical Company for
the support of this research.
Cp*ZrCl3. (Et2O)1/2 (1.00 g, 2.70 mmol) and 2 (0.84
g, 2.94 mmol) were mixed in a Schlenk tube and
suspended in ca. 30 ml of toluene. The vessel was fitted
with a condenser and refluxed for 5 days. Solvent was
stripped and the remaining oily solid was redissolved in
25 ml CH2Cl2/ 6 ml 4 M HCl. The product was
extracted twice with CH2Cl2 and the combined frac-
tions were washed with ca. 10 ml of H2O. The solution
was dried over CaSO4 and filtered. After the solvent
was removed under reduced pressure, the remaining
oily solid was triturated with pentane overnight to leave
a fine white solid. Yield ca. 52% 1H NMR (C6D6,
22°C): l 0.70 (s, 6H, Si(CH3)2); 1.69 (s, 15H, Cp*), 5.67
(t, JH–H=2.6 Hz, 2H, Cp ring), 6.40 (t, JH–H=2.6 Hz,
2H, Cp ring), 7.18 (d, JH–H=8.1 Hz, 2H, phenyl), 7.33
(d, JH–H=8.4 Hz, 2H, phenyl).
References
[1] (a) P.C. Mohring, N.J. Coville, J. Organomet. Chem. 479 (1994)
1. (b) J.A. Ewen, M.J. Elder, R.L. Jones, L. Haspeslagh, J.L.
Atwood, S.G. Bott, K. Robinson, Makromol. Chem. Macromol.
Symp. 48/49 (1991) 253. (c) R.F. Jordan, Adv. Organomet.
Chem. 32 (1991) 325.
[2] (a) H.J. Miller, B.S. Strickler, K.A. Abboud, J.M. Boncella, D.E.
Richardson, Organometallics 16 (1997) 1628. (b) H.J. Miller,
K.A. Abboud, J.M. Boncella, D.E. Richardson, in preparation.
[3] (a) H. Gilman, W. Longhorn, F. Moore, J. Am. Chem. Soc. 62
(1940) 2327. (b) H. Gilman, R. Gorsich, J. Am. Chem. Soc. 77
(1955) 3919. (c) H. Gilman, R. Gorsich. J. Am. Chem. Soc. 78
(1956) 2217. (d) H. Gilman, R. Gorsich. J. Am. Chem. Soc. 79
(1957) 2625.
[4] P. Jutzi, Chem. Rev. 86 (1986) 983.
3.6. Zr(CpSi)(C5Me5)(CH3)2 (6)
[5] C.S. Kraihanzel, M.L. Losee, J. Am. Chem. Soc. 90 (1968) 4701.
[6] D. Kondakov, E. Negishi, J. Chem. Soc. Chem. Comm. (1996)
963.
Compound 5 (0.448 g, 0.778 mmol) was dissolved in
ca. 20 ml Et2O and cooled to −50°C. 519 ml (3.0 M,
[7] P.T. Wolczanski, J.E. Bercaw, Organometallics 1 (1982) 793.
.