Ethylene polymerization using tetramethyl(2-methylthioethyl)cyclopentadienyl complexes of cobalt

Article abstract of DOI:10.1021/om030414b

Tetramethyl(2-methylthioethyl)cyclopentadiene complexes of cobalt have been synthesized and tested in ethylene polymerization. In combination with MMAO-3A, tetramethyl(2-methylthiothyl) cyclopentadienylcobalt diiodide (4) was shown to polymerize ethylene

Full text of DOI:10.1021/om030414b

Organometallics 2003, 22, 4699-4704
4699
Eth ylen e P olym er iza tion Usin g
Tetr a m eth yl(2-m eth ylth ioeth yl)cyclop en ta d ien yl
Com p lexes of Coba lt
Olafs Daugulis, Maurice Brookhart,* and Peter S. White
Department of Chemistry, University of North Carolina at Chapel Hill,
Chapel Hill, North Carolina 27599-3290
Received J une 3, 2003
Tetramethyl(2-methylthioethyl)cyclopentadiene complexes of cobalt have been synthesized
and tested in ethylene polymerization. In combination with MMAO-3A, tetramethyl(2-
methylthioethyl)cyclopentadienylcobalt diiodide (4) was shown to polymerize ethylene to
linear polyethylene with Mn’s in the range of 5000-68 000. The activity of the catalyst is
improved compared to the previously described catalysts containing the Cp*Co+(PR3) unit.
Cationic (tetramethyl(2-methylthioethyl)cyclopentadienyl)methylcobalt nitrile complexes 8
were also shown to be competent ethylene polymerization catalysts, producing polyethylene
with similar properties as observed with 4/MMAO-3A.
In tr od u ction
ylphosphine derivative 1c is not active as a polymeri-
zation catalyst, while trimethylphosphine and phosphite
complexes 1a ,b do polymerize ethylene.3g While the less
bulky cyclopentadienyl derivatives are intrinsically
more active than their pentamethylcyclopentadienyl
analogues 1a ,b, they are substantially less stable and
exhibit short lifetimes.3g
During the past decade olefin polymerizations using
late transition metal catalysts have attracted consider-
able attention.1 Relative to early metal d0 catalysts,
different enchainment mechanisms are frequently ob-
served for late metal systems leading to different
polymer microstructures. In addition, reduced oxophi-
licity of these systems allows the copolymerization of
ethylene and certain polar vinyl monomers.2 In the mid
1980s we and others reported cobalt(III) complexes of
the type 1, which are well-defined ethylene polymeri-
zation catalysts.3 These complexes polymerize ethylene
in a living fashion, producing polyethylenes with narrow
polydispersities and moderate molecular weights. The
catalyst resting state was shown to be a â-agostic
complex, but alkyl ethylene complexes such as 2 were
determined to be intermediates in the catalytic cycle.
Unfortunately, the activity of these complexes is
relatively low, generally 1-2 turnovers/min at 1 atm
ethylene.3c,f,g Activity was shown to be dependent on the
steric bulk of the complex; for example, the trieth-
(1) For recent reviews see: (a) Gibson, V. C.; Spitzmesser, S. K.
Chem. Rev. 2003, 103, 283. (b) Ittel, S. D.; J ohnson, L. K.; Brookhart,
M. Chem. Rev. 2000, 100, 1169. (c) Mecking, S. Coord. Chem. Rev.
2000, 203, 325. Pioneering studies: (d) Wilke, G. Angew. Chem., Int.
Ed. Engl. 1988, 27, 185. (e) Keim, W.; Kowaldt, F. H.; Goddard, R.;
Kru¨ger, C. Angew. Chem., Int. Ed. Engl. 1978, 17, 466. (f) Klabunde,
U.; Ittel, S. D. J . Mol. Catal. 1987, 41, 123.
(2) (a) J ohnson, L. K.; Mecking, S.; Brookhart, M. J . Am. Chem. Soc.
1996, 118, 267. (b) Mecking, S.; J ohnson, L. K.; Wang, L.; Brookhart,
M. J . Am. Chem. Soc. 1998, 120, 888. (c) J ohnson, L. K.; McLaine S.
J .; Sweetman, K. J .; Wang, Y.; Bennett, A. M. A.; Wang, L.; McCord,
E. F.; Ionkin, A.; Ittel, S. D.; Radzewich, C. E.; Schiffino, R. S. WO
200344066, 2003. Review: (d) Boffa, L. S.; Novak, B. M. Chem. Rev.
2000, 100, 1479.
We report here an investigation of Co(III) ethylene
polymerization catalysts derived from the tetramethyl-
(2-methylthioethyl)cyclopentadienyl ligand (general struc-
ture 3). The -SCH3 pendant ligand is much less bulky
than the PMe3 or P(OMe)3 ligands of 1a -c. Further-
more, the π-donor properties of the sulfide ligand could
conceivably lower the barrier to ethylene insertion
through stabilization of the electron-deficient transition
state for migratory insertion.
(3) (a) Schmidt, G. F.; Brookhart, M. J . Am. Chem. Soc. 1985, 107,
1443. (b) Brookhart, M.; Volpe, A. F., J r.; Lincoln, D. M. J . Am. Chem.
Soc. 1990, 112, 5634. (c) Brookhart, M.; Volpe, A. F., J r.; DeSimone,
J . M. Polym. Prepr. 1991, 32, 462. (d) Brookhart, M.; DeSimone, J .
M.; Grant, B. E.; Tanner, M. J . Macromolecules 1995, 28, 5378. (e)
Cracknell, R. B.; Orpen, A. G.; Spencer, J . L. J . Chem. Soc., Chem.
Commun. 1984, 326. (f) Enders, M.; Ludwig, G.; Pritzkow, H. Orga-
nometallics 2001, 20, 827. (g) Volpe, A. F., J r. Thesis, University of
North Carolina at Chapel Hill, 1991.
Resu lts a n d Discu ssion
Syn th esis of th e Coba lt Com p lexes. Potential
precursors to the catalytically active cationic cobalt(III)
alkyl complexes include the cobalt(III) dihalides (acti-
vated with aluminum alkyls), the cobalt(I) ethylene
10.1021/om030414b CCC: $25.00 © 2003 American Chemical Society
Publication on Web 10/15/2003

4700 Organometallics, Vol. 22, No. 23, 2003
Daugulis et al.
Sch em e 1
F igu r e 1. ORTEP view of 6. Selected interatomic dis-
tances (Å) and angles (deg): Co(1)-S(1) ) 2.1750(9), Co-
(1)-C(1) ) 1.980(4), Co(1)-C(2) ) 1.996(3), C(1)-C(2) )
1.400(6), Co(1)-S(1)-C(4) ) 99.29(13), Co(1)-S(1)-C(3) )
111.49(16), C(3)-S(1)-C(4) ) 99.65(21).
complex (activated through protonation), and cobalt(III)
dialkyl complexes activated through protonolysis. Syn-
theses of these precatalysts are described below.
Sch em e 2
Treatment of dicobalt octacarbonyl with tetramethyl-
(2-methylthioethyl)cyclopentadiene4a,5 yields the inter-
mediate dicarbonyl complex, which, without isolation,
was oxidized with iodine to give tetramethyl(2-meth-
ylthioethyl)cyclopentadienylcobalt diiodide (4) in an
excellent yield (Scheme 1). No νCO band is observed in
the IR spectrum, indicating that the sulfide arm has
displaced CO from the cobalt center. A halogen-bridged
dimer structure with a decoordinated sulfide arm can
be excluded on the basis of the observation of diaste-
reotopic ring methyl groups by NMR spectroscopy (see
Experimental Section). The bis-ethylene complex 5 was
obtained by reduction of the diiodide with potassium in
the presence of ethylene.6 Unfortunately, the ethylene
complex is a very air-sensitive oil, and attempts to
purify it were not successful. No conditions were found
where the bis-ethylene complex could be converted to
the monoethylene complex 6. However clean samples
of 6 could be obtained through treatment of 4 with
ethylmagnesium bromide, initially anticipated to yield
the cobalt diethyl complex. Recrystallization of the
product from pentanes yielded a red-brown, air- and
temperature-sensitive solid whose 1H and 13C NMR
spectra showed that the cobalt ethylene complex 6 was
formed. Presumably decomplexation of the sulfide arm
from a diethyl complex, followed by â-hydride elimina-
tion, reductive elimination of ethane, and recoordination
of sulfide, leads to the observed complex 6.
of 6 is shown in Figure 1. The carbon-carbon bond
length for the bound ethylene is somewhat elongated
compared to free ethylene (1.400(6) vs 1.339 Å).7 The
cobalt-sulfur bond length is 2.1750(9) Å, which is
within the range of the reported values for Cp-Co
complexes.8 As expected, the geometry around sulfur is
approximately tetrahedral, with the nonbonding pair of
electrons on S occupying the fourth site.
Due to the high sensitivity of the ethylene complex
6, a more convenient entry into cationic cobalt(III)
compounds was sought. The reaction of the diiodide with
methyllithium afforded the corresponding dimethyl
complex 7 as a red-violet, crystalline substance (Scheme
2).9 This compound is stable at RT in crystalline form;
however, in chlorinated hydrocarbons slow decomposi-
tion is observed. Protonation of 7 using HB(ArF)4‚2Et2O
(ArF ) 3,5-(CF3)2C6H3) in CH2Cl2 followed by complex-
ation with either acetonitrile or 3,5-(CF3)2C6H3CN af-
forded the cationic methyl nitrile complexes 8a ,b as
violet-brown stable solids.
The structure of the complex was verified by single-
crystal X-ray diffraction analysis. The ORTEP diagram
(4) (a) Krut’ko, D. P.; Borzov, M. V.; Petrosyan, V. S.; Kuz’mina, L.
G.; Churakov, A. V. Izv. Akad. Nauk SSSR, Ser. Khim. 1996, 984 (in
Russian). (b) Krut’ko, D. P.; Borzov, M. V.; Kuz’mina, L. G.; Churakov,
A. V.; Lemenovskii, D. A.; Reutov, O. A. Inorg. Chim. Acta 1998, 280,
257. A review about cyclopentadienylmetal complexes bearing pendant
ligands: (c) Butenscho¨n, H. Chem. Rev. 2000, 100, 1527. For cyclo-
pentadienylchromium complexes with pendant sulfide groups, see: (d)
Do¨hring, A.; Go¨hre, J .; J olly, P. W.; Kryger, B.; Rust, J .; Verhovnik,
G. P. J . Organometallics 2000, 19, 388.
NMR Stu d ies of th e Com p lexes. Due to a signifi-
cant barrier to inversion at the chiral sulfur atom, the
(7) Tables of Interatomic Distances and Configuration in Molecules
and Ions; Sutton, L. E., Ed.; The Chemical Society: London, 1958.
(8) (a) Sheldrick, W. S.; Hauck, E.; Korn, S. J . Organomet. Chem.
1994, 467, 283. (b) Harada, T.; Takayama, C.; Kajitani, M.; Sugiyama,
T.; Akiyama, T.; Sugimori, A. Bull. Chem. Soc. J pn. 1998, 71, 2645.
(9) (a) Preparation of CpCoMe2PPh3: Yamazaki, H.; Hagihara, N.
Bull. Chem. Soc. J pn. 1965, 38, 2212. (b) Protonation of CpCoMe2-
PMe3: Hofmann, W.; Werner, H. Chem. Ber. 1982, 115, 119.
(5) Frith, S. A.; Spencer, J . L. Inorg. Synth. 1990, 28, 273.
(6) (a) Green, M. L. H.; Pardy, R. B. A. J . Chem. Soc., Dalton Trans.
1979, 355. (b) J utzi, P.; Kristen, M. O.; Dahlhaus, J .; Neumann, B.;
Stammler, H.-G. Organometallics 1993, 12, 2980. (c) Holle, S.; J olly,
P. W. J . Organomet. Chem. 2000, 605, 157. Reference 6c describes
Cp-Co ethylene complexes with pendant amino groups.

Ethylene Polymerization Using Co Complexes
Organometallics, Vol. 22, No. 23, 2003 4701
Ta ble 1. P olym er iza tion of Eth ylen e by 4/MMAO or P MAO-IP (en tr ies 1-13) a n d 8a ,b (en tr ies 14-17) in
Ch lor oben zen ea
no.
pressure, psi
time, h
2.83
3
1
1
1
1
1
temp, °C
g PE
TOF,b h-1
Mn
Br/1000Cc
MWD
1d
2
3
14
400
100
400
400
400
400
400
400
800
800
400
400
400
400
400
400
26
5-6
5.0
3.4
0.7
3.5
3.7
3.9
3.8
6.7
3.6
2.8
4.2
2.2
2.9
1.1
0.3
2.6
1.4
630
4000
2500
11 800
68 000
23 500
28 000
23 800
28 000
29,000
30 900
21 100
22 900
31 100
7900
5200
24 700
ND
25 200
31 600
9
1
3
3
2
2
2
1
2
2
2
4
2.4
2.0
2.2
2.1
2.4
2.2
2.0
2.1
3.0
2.1
1.9
2.3
3.7
1.9
ND
2.3
2.0
26-28
28-31
28-31
28-31
28-31
26-31
28-31
28-32
28-32
59-62
59-62
26-29
26-29
26-31
26-30
4
12 500
13 200
13 900
13 600
8,000
12 900
20 000
15 000
15 700
6900
5e
6
7e
8
3
9f
0.25+0.75
0.5
1
0.5
1.5
1
1
1
1
10
11
12
13
14g
15h
16i
17j
3
3900
1100
6500
5000
10
ND
6
4
a
0.01 mmol precatalyst in 100 mL of chlorobenzene; entries 1-5, 8-13 MMAO-3A activator; entries 6-7 PMAO-IP activator, entries
2-4, 6, 8-13 Al/Co ratio 300. TOF calculated assuming all Co is active. c Branching/1000C determined by 1H NMR spectroscopy. Al/
b
d
Co ratio 220, 0.1 mmol precatalyst. e Al/Co ratio 1000. f Pressurize to 400 psi, polymerize for 15 min, then vent reactor, wait 3 h, then
g
h
i
repressurize to 400 psi, polymerize for 45 min. Catalyst 8a , no activator. Catalyst 8a , no activator, 2 equiv of CH3CN added. Catalyst
8b (0.014 mmol), no activator. Catalyst 8b, no activator.
j
excess of CD3CN. Since the first-order rate constant
obtained by using 15 equiv of CD3CN (kobs ) (1.9 ( 0.1)
× 10-4 s-1, ∆Gq ) 20.5 ( 0.1 kcal/mol) is nearly identical
to the rate constant obtained using 47 equiv (kobs
)
(1.7 ( 0.1) × 10-4 s-1, ∆Gq ) 20.6 ( 0.1 kcal/mol), the
exchange most likely proceeds by the rate-determining
dissociation of acetonitrile (kobs ) kdissoc) followed by
trapping of the coordinatively unsaturated intermediate
with free acetonitrile. However, rate-limiting dissocia-
tion of the sulfide ligand cannot be rigorously excluded.
P olym er iza tion of Eth ylen e. Initially the polym-
erization of ethylene was studied using the combination
of diiodide 4 with MMAO-3A in chlorobenzene (Table
1). Under 1 atm of ethylene at room temperature after
2.8 h polyethylene with Mn ) 11 800 (MWD ) 2.4) was
obtained with a turnover frequency (TOF) 630 TO/h. For
comparison, catalyst 10/MAO in toluene under 1 atm
NMR spectra of complexes 4 and 7 exhibit signals for
diastereotopic methyl groups on the cyclopentadienyl
ring below room temperature. For complex 8a , in which
both the sulfur and cobalt centers are stereogenic, two
diastereomers are observed at low temperature. The
inversion process for 7 is shown above. To determine
the energetics of inversion at sulfur, variable-temper-
ature NMR studies of complexes 4, 7, and 8a were
carried out.10 For complex 4, the barrier to inversion at
sulfur was determined to be 15.8 ( 0.1 kcal/mol (tetra-
chloroethane-d2) and for 7 13.9 ( 0.1 kcal/mol (meth-
ylcyclohexane-d14). For 8a in THF-d8, in which coales-
cence studies were carried out, a 1/0.64 ratio of
diastereomers is observed. The barriers of inversion at
sulfur were measured to be 15.3 ( 0.1 (minor to major
diastereomer) and 15.6 ( 0.1 kcal/mol (major to minor
diastereomer). For comparison, in complex 9 the barrier
of inversion at sulfur is 10.5-11.2 kcal/mol depending
on the solvent.4a
ethylene produces PE with Mn ) 9900 (MWD ) 4.5) and
42 TO/h (TOF corresponds to 3 h run).3g At 5 °C, 400
psi ethylene pressure polyethylene with Mn ) 68 000
was obtained (entry 2, TOF ) 4000 h-1). As expected,
at higher temperatures increased TOF (ca. 13 000 at
RT) and decreased Mn (ca. 28,000) is observed (entry
4). The decrease in the calculated TOF in the 3 h run
at 28-31 °C, 400 psi (entry 8) is probably due to the
precipitation of the polymer. Polymerization for 15 min,
followed by venting the reactor, waiting for 3 h, then
repressurizing with ethylene, and polymerizing for an
additional 45 min afforded almost the same amount of
polyethylene as the 1 h uninterrupted run (compare
entry 9 to entry 4).11 These results suggest that catalyst
Acetonitrile self-exchange in 8a was studied in CD2-
Cl2 solution at -2 °C by treatment of 8a with a large
(10) Inversion of sulfides coordinated to transition metals has been
widely studied. For a review, see: (a) Abel, E. W.; Bhargawa, S. K.;
Orrell, K. G. Prog. Inorg. Chem. 1984, 32, 1. Recent examples: (b)
Albe´niz, A. C.; Espinet, P.; Lin, Y.-S. Organometallics 1996, 15, 5010.
(c) Dupont, J .; Gruber, A. S.; Fonseca, G. S.; Monteiro, A. L.; Ebeling,
G. Organometallics 2001, 20, 171. (d) Canovese, L.; Lucchini, V.; Santo,
C.; Visentin, F.; Zambon, A. J . Organomet. Chem. 2002, 642, 58. (e)
Shan, X.; Espenson, J . H. Organometallics 2003, 22, 1250.
(11) A referee has suggested that catalyst lifetime could be shortened
under actual polymerization conditions. We cannot rule this out, but
since the resting state is an agostic species, we believe this experiment
accurately reflects catalyst stability under polymerization conditions.

4702 Organometallics, Vol. 22, No. 23, 2003
Daugulis et al.
Sch em e 3
is stable for hours at room temperature. At 60 °C
catalyst is unstable and reduced lifetimes and Mn’s
(5000-8000) are observed (see entries 12 and 13).
The TOF is nearly linear with respect to ethylene
pressure. A comparison of entries 3, 4, and 10 shows
that as the pressure increases from 100 to 400 psi to
800 psi the corresponding increase in TOF is 2500/h to
12 800/h to 20 000/h. These results suggest that the
major catalyst resting state under these conditions is a
cobalt alkyl complex and not a cobalt alkyl ethylene
complex, consistent with the observations in the case
of the earlier P(OMe)3-based catalyst.3 As established
in the case of the P(OMe)3 complex 1a , these kinetics
are consistent with a preequilibrium between a cobalt
agostic alkyl complex and a cobalt alkyl ethylene
complex which favors the cobalt alkyl complex.
acetonitrile binding.13 The complex 8b, containing the
more weakly binding bis-3,5-trifluoromethylbenzoni-
trile, was also examined. As expected, this complex is
more reactive than the acetonitrile complex 8a , exhibit-
ing a TOF of 5000-6500/h at 400 psi ethylene (entries
16, 17, Table 1). Using 8b, chain growth could be
observed by NMR spectroscopy at 0 °C in CD2Cl2.
Treatment of a 0.017 M solution of 8b in CD2Cl2 with 9
equiv of ethylene results in consumption of C2H4 (ca. 3
equiv in 30 min) and growth of an alkyl chain. No vinylic
resonances are evident. Only signals for coordinated
nitrile are observed, indicating that the catalyst resting
state is the cationic cobalt alkyl nitrile complex.
Branching densities in the polyethylene are quite low,
1-10 branches/1000 carbons, and are in the range of
high-density polyethylene. Tm’s fall in the narrow range
between 132 and 133 °C. Only vinylic end groups were
detected in low molecular weight samples. The detection
of vinylic end groups, the similarity of Mn values from
MMAO activation of diiodide relative to the activator-
free nitrile complexes 8, and the insensitivity of molec-
ular weight to Co/Al ratios all point to â-hydride
elimination as the dominant chain transfer mechanism
for both catalyst systems. No significant chain transfer
to Al appears to occur in the MMAO-3A or PMAO-IP
activated systems.
No significant difference was observed between using
an Al/Co ratio of 300 and 1000 (entry 5 vs entry 4).
PMAO-IP could also be used as activator with no
difference between Al/Co ratios of 300 and 1000 (entries
6 and 7).
Monomodal molecular weight distributions of ca. 2 for
polymer prepared from MMAO activation of diiodide
indicate a single site catalyst. The use of large quantities
of MMAO precludes NMR studies for identification of
the catalyst resting state. On the basis of earlier studies,
the likely resting state is a cationic cobalt alkyl complex,
but since MMAO can serve as a reducing agent,12
verification of this assumption is desired. While pro-
tonation of Cp*(P(OMe)3)Co(C2H4) leads to an agostic
ethyl complex3a (a model for the catalyst resting state),
similar attempts to generate a spectroscopically char-
acterizable cationic agostic ethyl complex from low-
temperature protonation of 6 have not been successful.
However, as noted above, protonation of dimethyl
complex 7 and trapping with CH3CN yields the cationic
methyl acetonitrile complex 8a . This cationic methyl
complex serves as an “activator-free” complex for eth-
ylene polymerization (entry 14). At 400 psi ethylene, RT,
polyethylene is produced with a very similar molecular
weight, MWD, and branching level compared to PE
produced under similar conditions with 4/MMAO (entry
4). These results support a cationic Co(III) alkyl inter-
mediate in the case of MMAO and PMAO activation.
The lower TOF in the case of acetonitrile complex 8a is
proposed to be due to competitive binding of ethylene
and CH3CN at the Co(III) center, as shown in Scheme
3. Performing the polymerization under the conditions
of entry 14, but with addition of 2 equiv of acetonitrile,
further lowers the TOF to 1100 (entry 15), supporting
the notion that the lower TOF is due to competitive
Su m m a r y
Polymerization of ethylene by cobalt complexes con-
taining a tetramethyl(2-methylthioethyl)cyclopentadi-
enyl unit has been investigated. In combination with
MMAO-3A, tetramethyl(2-methylthioethyl)cyclopenta-
dienylcobalt diiodide (4) was shown to polymerize eth-
ylene, producing polyethylene with Mn up to 68 000. The
activity of the catalyst is substantially improved com-
pared to the previously described catalysts containing
the Cp*Co+(PR3) unit. Cationic alkyl complexes [tet-
ramethyl(2-methylthioethyl)cyclopentadienyl]Co(CH3)-
L+ (L ) CH3CN, 8a ; (CF3)2C6H3CN, 8b) were synthe-
sized as activator-free initiators. Although ethylene
polymerization activity was reduced relative to 4/MMAO
due to competitive nitrile binding, the properties of the
polyethylene produced were identical to those of poly-
ethylene produced from MMAO activation. These re-
sults suggest that both systems involve a cationic Co(III)
alkyl ethylene complex as the key intermediate and that
(13) From exchange studies, the estimated rate of acetonitrile
dissociation from 8a at RT is ca. 0.04 s-1. The measured TOF (ca. 1
s-1 at 400 psi ethylene) exceeds this value. This suggests that at high
ethylene pressure the trapping of the agostic alkyl complex [Me4CpCH2-
CH2SMe]Co(CH2Pol)+ by ethylene and insertion is significantly faster
than retrapping by acetonitrile to return to species 10a (Scheme 3).
Such a mechanism, in which a significant fraction of the catalyst rests
as the acetonitrile complex, is fully consistent with suppression of the
reactivity by added CH3CN.
(12) (a) Kooistra, T. M.; Knijnenburg, Q.; Smits, J . M. M.; Horton,
A. D.; Budzelaar, P. M. H.; Gal, A. W. Angew. Chem., Int. Ed. 2001,
40, 4719. (b) Gibson, V. C.; Humphries, M. J .; Tellmann, K. P.; Wass,
D. F.; White, A. J . P.; Williams, D. J . Chem. Commun. 2001, 2252.

Ethylene Polymerization Using Co Complexes
Organometallics, Vol. 22, No. 23, 2003 4703
rated and the distillate checked for oligomers by NMR. The
following polymer Tm values were determined (entry number
in Table 1: Tm, °C): 1, 132.2; 2, 132.5; 6, 133.0.
chain transfer occurs via â-hydride elimination and not
through chain transfer to aluminum in the case of
MMAO activation.
[Me4Cp CH2CH2SMe]CoI2 (4). This compound was made
using a precedented procedure.5 Co2(CO)8 (3.28 g, 9.6 mmol,
Strem) was mixed with 1,3-cyclohexadiene (1.3 mL, 13.6 mmol,
Aldrich) and CH2Cl2 (10 mL). To the magnetically stirred
brown-red solution was added dropwise a solution of Me4-
CpHCH2CH2SMe (3.41 g, 17.4 mmol) in CH2Cl2 (10 mL) with
venting into the atmosphere through a silicon oil bubbler
(evolution of CO). The mixture was refluxed for 1.5 h. After
evaporation under vacuum CH2Cl2 (10 mL) was added to the
residue, followed by a dropwise addition of a solution of I2 (4.85
g, 19.1 mmol, Fisher Scientific) in CH2Cl2 (80 mL) with venting
into the atmosphere through a silicon oil bubbler (CO evolu-
tion). The solution was stirred for 1 h at RT, and the color
changed from brown to dark green. After evaporation of about
half of the solvent the residue was purified by flash chroma-
tography on silica gel (17 × 6.4 cm) in 1/1 CH2Cl2/hexanes
under air. The first 500 mL were blank. The eluent was
changed to CH2Cl2 and the elution continued. The next 750
mL contained impurities followed by 1500 mL containing the
desired product. Solvent was changed to 4/1 CH2Cl2/EtOAc,
and the next 1500 mL contained additional product. Fractions
containing the product were evaporated to give a dark green
solid (7.92 g, 89.6%). 1H NMR (tetrachloroethane-d2, -10 °C):
δ 4.11-3.97 (m, 1H); 3.91-3.77 (m, 1H); 2.43-2.22 (m, 2H);
2.28 (s, 3H); 2.11 (s, 3H); 2.08 (s, 3H); 2.00 (s, 3H); 1.97 (s,
3H). 1H NMR (tetrachloroethane-d2, 105 °C): δ 4.00 (t, 2H;
J ) 6.8 Hz); 2.33 (t, 2H; J ) 6.8 Hz); 2.32 (s, 3H); 2.17 (s, 6H);
2.15 (s, 6H). 13C{1H} (CD2Cl2, 0 °C): δ 108.7; 97.4; 94.0; 90.3;
86.6; 54.3; 22.8; 21.9; 13.6; 13.0; 11.6; 10.0. Anal. Calcd for
Exp er im en ta l Section
Gen er a l Con sid er a tion s. All the operations related to
catalysts were carried out under an argon atmosphere using
standard Schlenk techniques. The 1H and 13C spectra were
recorded using Bruker 300, 400, or 500 MHz spectrometers
and referenced against residual solvent peaks (1H, 13C). Flash
chromatography was performed using 60 Å silica gel (SAI).
GPC was performed at DuPont Analytical in Wilmington, DE,
using a Waters 150-C AL/GPC equipped with a Polymer
Standards Service “polyolefin” high-speed GPC column. The
mobile phase was 1,2,4-trichlorobenzene at 135 °C, flow rate
of 5 mL/min, RI detector. Elemental analyses were performed
by Atlantic Microlab Inc. of Norcross, GA.
Ma ter ia ls. Anhydrous solvents were used in the reactions.
Solvents were distilled from drying agents or passed through
alumina columns under an argon or nitrogen atmosphere.
NMR solvents were vacuum transferred from P2O5 and de-
gassed by repeated freeze-pump-thaw cycles. The following
starting materials were made using literature procedures: HB-
(ArF)4‚2Et2O14 and tetramethyl(2-methylthioethyl)cyclopenta-
diene.4a NaB(ArF)4 was purchased from Boulder Scientific;
MMAO (modified methylaluminoxane) and PMAO-IP were
from Akzo Nobel.
Analysis of polymer branching by 1H NMR spectroscopy was
carried out using a published method.15 Since the end group
signals (vinylic) were too small to be integrated correctly, the
end group and allylic integrals were adjusted to match the
molecular weight of polymer observed by GPC. In some
polymers, trace amounts of residual MAO were observed
upfield of the CH3 group signals. The NMR was measured in
bromobenzene-d5 at 120 °C. Errors in coalescence and ∆Gq
measurements were calculated assuming temperature varia-
tions of (1 K. Errors in rate constants were obtained assuming
5% error in measurements.
C
12H19CoSI2: C 28.37, H 3.77. Found: C 28.71, H 3.79.
Coalescence temperature of peaks at 2.11 and 2.00 ppm
(∆δ ) 33.2 Hz) as well as 2.08 and 1.97 ppm (∆δ ) 33.9 Hz)
was determined to be +41 °C (tetrachloroethane-d2).
[Me4Cp CH2CH2SMe]Co(eth ylen e)2 (5). Complex 4 (0.508
g, 1.0 mmol) was mixed with K (0.086 g, 2.2 mmol). The
mixture was cooled to 0 °C and the Schlenk flask flushed with
ethylene followed by the addition of THF (10 mL). The mixture
was stirred at 0 °C with continuous bubbling of ethylene until
the color changed from green to brownish red (about 40 min)
and then an additional 10 min. The solution was evaporated,
and the residue was extracted with pentane and filtered
through a pad of Celite. After evaporation a dark brown, very
air-sensitive oil was obtained. Attempts to purify the complex
were not successful due to noncrystallinity and instability. 1H
NMR (toluene-d8, 25 °C): δ 2.48-2.31 (m, 4H); 1.82 (s, 3H);
1.70-1.59 (m, 4H); 1.47 (s, 6H); 1.29 (s, 6H); 0.95-0.86
(m, 4H).
Sp ectr a l Da ta for th e B(Ar F )4 Cou n ter ion . The following
1H and 13C spectroscopic assignments of the B(ArF)4 counterion
were invariant for different temperatures and are not reported
in the spectroscopic data for each of the cationic complexes.
B[3,5-C6H3(CF3)2]4- (B(ArF)4): 1H NMR (CD2Cl2): δ 7.74 (br s,
8H), 7.57 (s, 4H). 13C{1H} NMR (CD2Cl2): δ 162.2 (q, J C-B
37.4 Hz), 135.2, 129.3 (q, J C-F ) 31.3 Hz), 125.0 (q, J C-F
272.5 Hz), 117.9.
)
)
Gen er a l P olym er iza tion P r oced u r e. Polymerizations
were carried out in a mechanically stirred 300 mL Parr reactor
equipped with an electric heating mantle controlled by a
thermocouple in the reaction mixture. The reactor was charged
with chlorobenzene (100 mL) and heated for 1 h at 150 °C.
After cooling to RT the solvent was poured out and the reactor
heated under vacuum at 150 °C for 1 h. The reactor was filled
with Ar, cooled to RT, pressurized to 200 psi ethylene, and
vented three times. A solution of the catalyst in 100 mL of
chlorobenzene was added to the reactor via cannula, the
appropriate activator added via syringe, the reactor pressur-
ized with ethylene to the required pressure, and the pressure
maintained at this value during the polymerization. After that
the reaction mixture was stirred for the appropriate time. The
reactor was occasionally cooled with an external ice bath to
keep the temperature in the desired range. After venting, the
reaction mixture was poured into methanol and the polyeth-
ylene was washed with acidic methanol and then dried under
vacuum at 80 °C. The solvent from precipitation was evapo-
[Me4Cp CH2CH2SMe]Co(eth ylen e) (6). To a stirred sus-
pension of 4 (0.508 g, 1.0 mmol) in diethyl ether (8 mL) was
added EtMgBr (0.83 mL of a 3 M solution in diethyl ether, 2.5
mmol, Aldrich) at -78 °C. No reaction was observed. The
cooling bath was removed and the solution concentrated under
vacuum to half of the initial volume, followed by the addition
of THF (5 mL). The solution was stirred at RT for 10 min.
The color changed from green to black to red. The reaction
mixture was filtered through a pad of Florisil cooled to
-78 °C eluting with diethyl ether, the solvent was evaporated,
and the residue was crystallized from a small amount of pen-
tane at -30 °C. The product was isolated as red-brown crystals
(0.13 g, 46.1%) that decompose at RT in the solid state and in
solution. The structure was verified by X-ray crystallography.
1H NMR (CD2Cl2, -60 °C): δ 3.59-3.45 (m, 1H); 3.21-3.10
(m, 1H); 2.50-2.37 (m, 1H); 2.32 (s, 3H); 2.28 (s, 3H); 2.20-
2.09 (m, 1H); 1.85 (s, 3H); 1.28-0.99 (m, 4H); 1.00 (s, 6H).
13C{1H} (CD2Cl2, -30 °C): δ 106.6; 91.9; 88.9; 84.1; 80.8; 53.8;
31.3; 29.5; 22.1; 20.7; 11.9; 11.7; 9.0. Two Me peaks on the Cp
overlap at 9.0 ppm.
(14) Brookhart, M.; Grant, B.; Volpe, A. F. Organometallics 1992,
11, 3920.
(15) Daugulis, O.; Brookhart, M.; White, P. S. Organometallics 2002,
21, 5935.

4704 Organometallics, Vol. 22, No. 23, 2003
Daugulis et al.
Ta ble 2. Cr ysta llogr a p h ic Da ta Collection
P a r a m eter s for 6
[Me4Cp CH2CH2SMe]CoMe2 (7). To the solution of 4 (1.52
g, 3.0 mmol) in THF (30 mL) was added MeLi (4.2 mL of a
1.6 M solution in diethyl ether, 6.6 mmol, Aldrich) at -78 °C.
The color of the solution changed from dark green to red. The
solution was stirred for 1 h at -78 °C, the cooling bath was
removed, and two-thirds of the solvent was evaporated under
reduced pressure. The residue was filtered through a pad of
Florisil at -78 °C eluting with diethyl ether. The eluent was
evaporated, the residue was dissolved in pentane (40 mL) and
filtered through Celite into a Schlenk flask precooled to -78
°C, and Celite was washed with additional pentane (10 mL).
After 4 h the solvent was transferred away from the crystals
by cannula, and the crystals were washed with a minimal
amount of cold pentane and dried under vacuum. The product
was obtained as red-violet crystals (0.41 g, 48.3%) and is stable
at RT under inert atmosphere for at least days. 1H NMR
(methylcyclohexane-d14, -10 °C): δ 3.65-3.43 (m, 1H); 3.27-
3.09 (m, 1H); 2.43-2.24 (m, 1H); 2.21-2.02 (m, 1H); 1.85 (s,
formula
C14H23CoS
mol wt
cryst syst
space group
282.33
monoclinic
P21/c
a, Å
b, Å
c, Å
9.17860(20)
9.44180(20)
15.8065(3)
97.631(1)
1357.70(5)
4
â, deg
V, Å3
Z
dens calcd, Mg/m3
F(000)
1.381
593.98
0.30 × 0.20 × 0.20
-100
1.5418
140.0
11.09
8669
2413
2069
cryst dimens, mm
temp, °C
radiation (λ, Å)
2θ max, deg
µ, mm-1
total no. of rflns
total no. of unique rflns
no. of obsd data (I>2.5σ(I))
no. of refined params
RF, %
1
6H); 1.70 (s, 3H); 1.10 (s, 3H); 1.07 (s, 3H); -0.05 (s, 6H). H
NMR (methylcyclohexane-d14, 60 °C): δ 3.34 (t, 1H; J ) 6.8
Hz); 2.18 (t, 1H; J ) 6.8 Hz); 1.81 (s, 6H); 1.64 (s, 3H); 1.05 (s,
6H); -0.05 (s, 6H). 13C{1H} (CD2Cl2, -10 °C): δ 113.5; 89.5;
89.1; 86.4; 85.9; 55.0; 21.7; 16.5; 9.3; 9.0; 8.3; -6.0; -6.8. Two
238
0.041
Rw, %
0.048
GOF
1.7705
Me peaks on the Cp overlap at 8.3 ppm. Anal. Calcd for C14
-
H
25CoS: C 59.14, H 8.86. Found: C 58.65, H 8.75. Coalescence
8a , only after protonation 3,5-(CF3)2C6H3CN (56 µL, 0.33 mmol,
Aldrich) was added instead of acetonitrile. After drying a
violet-brown foam was obtained, 0.38 g (92.4%). 1H NMR (CD2-
Cl2, -10 °C; 1.3/1 mixture of diastereomers): δ 8.26 (s, 1H);
8.23 (s, 1H); 8.1 (s, 2H; ArF-CN peaks overlap for both
diastereomers); 4.26-4.11 (m, 1H); 3.88-3.74 (m, 2H); 3.44-
3.29 (m, 1H); 2.69-2.48 (m, 1H); 2.47-2.35 (m, 2H); 2.34-
2.16 (m, 1H); 2.24 (s, 3H); 2.17 (s, 3H); 2.10 (s, 3H); 1.95
(s, 3H); 1.60 (s, 3H); 1.53 (s, 3H); 1.30 (s, 3H); 1.29 (s, 3H);
1.21 (s, 3H); 1.20 (s, 6H; overlap of two Me groups); 1.17
(s, 3H). Multiplets at 4.26, 3.44, 2.47 ppm are mutually
coupled. Signals at 3.88, 2.69, 2.34 ppm are mutually coupled.
Anal. Calcd for C54H37CoBF30SN: C 47.28, H 2.72, N 1.02.
Found: C 47.02, H 2.60, N 1.04.
Aceton itr ile Self-Exch a n ge Kin etics for 8a . To the
solution of 8a (10 mg, 0.0085 mmol) in CD2Cl2 (0.6 mL) cooled
to -78 °C was added CD3CN (6.7 µL, 0.128 mmol). The NMR
tube was placed in a probe cooled to -2 °C and the loss of
bound acetonitrile peaks at 2.36 and 2.29 ppm monitored
versus time, giving kobs ) 1.9 × 10-4 s-1, ∆Gq ) 20.5 ( 0.1
kcal/mol. The second experiment was carried out as above,
except using 21 µL (0.40 mmol) of CD3CN. The following data
were obtained: kobs ) 1.7 × 10-4 s-1, ∆Gq ) 20.6 ( 0.1 kcal/
mol.
temperature of multiplets centered at 3.5 and 3.2 ppm (∆δ )
107.4 Hz) was determined to be 17 °C, and that of multiplets
at 2.3 and 2.1 ppm (∆δ ) 65.7 Hz) 11 °C (methylcyclohexane-
d
14).
[Me4CpCH2CH2SMe]Co(Me)(NCCH3)+B(Ar F)4- (8a). Solid
HB(ArF)4‚2Et2O (0.313 g, 0.31 mmol) was mixed with 7 (0.085
g, 0.30 mmol). After cooling to -78 °C CH2Cl2 (2 mL) was
added and the dark red-brown reaction mixture stirred for 15
min at that temperature followed by the addition of CH3CN
(78 µL, 1.5 mmol). The color changed to brown-red. After
stirring for 15 min at -78 °C the solution was warmed to RT,
evaporated, coevaporated with hexanes, and dried under
vacuum to give 0.35 g (99.4%) of product as a violet-brown
solid. 1H NMR (CD2Cl2, -10 °C; 1/1 mixture of diastereo-
mers): δ 4.19-4.06 (m, 1H); 3.81-3.71 (m, 2H); 3.36-3.25 (m,
1H); 2.59-2.45 (m, 1H); 2.38-2.24 (m, 2H); 2.36 (s, 3H;
complexed CH3-CN); 2.29 (s, 3H; complexed CH3-CN); 2.24-
2.12 (m, 1H); 2.19 (s, 3H); 2.10 (s, 3H); 1.88 (s, 3H); 1.55
(s, 3H); 1.47 (s, 3H); 1.22 (s, 3H); 1.19 (s, 3H); 1.10 (s, 3H);
1.09 (s, 3H); 1.08 (s, 3H); 1.07 (s, 3H). Multiplets at 4.19, 3.36,
2.38 ppm are mutually coupled. Signals at 3.81, 2.59, 2.24 ppm
are mutually coupled. 13C{1H} (CD2Cl2, -40 °C; mixture of 2
diastereomers): δ 116.5; 115.6; 99.5; 99.2; 97.9; 96.3; 93.7; 93.1;
80.6; 77.2; 54.4; 52.3; 22.2; 21.2; 18.2; 15.8; 9.5; 9.4; 9.3; 9.2;
8.5; 8.4; 8.3; 7.8; 4.8; 4.7; 1.7; 0.3. Signals for two Cp ring
X-r a y Cr ysta l Str u ctu r e (6). Diffraction data were col-
lected on a Bruker SMART 1K diffractometer using the ω-scan
mode. Refinement was carried out with the full-matrix least-
carbons are missing. Anal. Calcd for C47H37CoBF24SN:
C
48.10; H 3.18; N 1.19. Found: C 47.78, H 3.04, N 1.26. Rates
of inversion at sulfur in 8a (THF-d8 solution) were determined
using line shape analysis of peaks at 2.52, 2.45 (set 1) and
2.20, 2.00 (set 2), which broaden and coalesce. The analysis
was carried out at +20, +25 °C(set 1) and +30, +35, +40 °C
(set 2).16 Using set 1 at +20 °C, kminorfmajor ) 29 s-1, kmajorfminor
squares method based on
F (NCRVAX) with anisotropic
thermal parameters for all non-hydrogen atoms. Hydrogen
atoms were inserted in calculated positions and refined riding
with the corresponding atom. Complete details of X-ray data
collection are given in Table 2.
) 19 s-1; at +25 °C, kminorfmajor ) 43 s-1, kmajorfminor ) 27 s-1
.
)
Using set 2 at +30 °C, kminorfmajor ) 66 s-1, kmajorfminor
Ack n ow led gm en t. We are grateful to DuPont and
the National Science Foundation (CHE-0107810) for
support of this research and to DuPont Analytical for
GPC analyses.
42 s-1; at +35 °C, kminorfmajor ) 104 s-1, kmajorfminor ) 66 s-1; at
+40 °C, kminorfmajor ) 172 s-1, kmajorfminor ) 110 s-1. Equilibrium
ratio of the two diastereomers in THF-d8 at 0 °C is 1/0.637.
1H spectra of 8a in THF-d8 are given in the Supporting
Information.
Su p p or tin g In for m a tion Ava ila ble: X-ray crystallogra-
phy data for 6, NMR data for 8a in THF-d8, and graphs for
the determination of acetonitrile self-exchange rate. This
material is available free of charge via the Internet at
http://pubs.acs.org.
[Me 4C p C H 2C H 2S Me ]C o (Me )[3,5-(C F 3)2C 6H 3C N ]+B -
-
(Ar F )4 (8b). The reaction was carried out as in the case of
(16) The DNMR simulations were performed with an updated
version of WINDNMR (http://www.chem.wisc.edu/areas/reich/plt/
windnmr.htm). Reich, H. J . WinDNMR Dynamic NMR Spectra for
Windows. J . Chem. Ed. Software, Ser. D 1996, 3D, No. 2.
OM030414B