Synthesis and MMA polymerization of chiral ansa-zirconocene ester enolate complexes with C2- and Cs-ligation

Article abstract of DOI:10.1016/j.jorganchem.2006.04.036

A series of chiral ansa-zirconocene ester enolate complexes incorporating C₂- or C_s-symmetric ligands, including neutral rac-(EBI)ZrCl[OC(OⁱPr){double bond, long}CMe₂] (1), rac-(EBI)Zr(OTf)[OC(OⁱPr){double bond, long}CMe₂] (2), rac-(EBI)Zr(OTf)[OC(OMe){double bond, long}C(Me)CH₂C(Me₂)C(OⁱPr){double bond, long}O] (3), [Me₂C(Cp)(Flu)]ZrMe[OC(OⁱPr){double bond, long}CMe₂] (4), and cationic [Me₂C(Cp)(Flu)]Zr⁺(THF)[OC(OⁱPr){double bond, long}CMe₂][MeB(C₆F₅)₃]^- (5), have been synthesized. Within the neutral C₂-ligated zirconocene ester enolate series, the chloride derivative 1 is inactive toward any methyl methacrylate (MMA) additions, the methyl derivative rac-(EBI)ZrMe[OC(OⁱPr){double bond, long}CMe₂] adds cleanly only 1 equiv. of MMA, and the triflate derivative 2 can add either 1 equiv. of MMA to form the single-MMA-addition product 3 or multiple equivalents of MMA to form P(MMA). Unlike the C_s-ligated methyl cation [Me₂C(Cp)(Flu)ZrMe]⁺, which is inactive for MMA polymerization under various conditions, the C_s-ligated ester enolate cation 5 is moderately active for polymerization of MMA and N,N-dimethylacrylamide at ambient temperature; the resulting P(MMA) has a high molecular weight of M_n = 388 000 Da but a low syndiotacticity of [rr] = 64%, and the polymerization conforms to a chain-end control mechanism.

Full text of DOI:10.1016/j.jorganchem.2006.04.036

Journal of Organometallic Chemistry 691 (2006) 3490–3497
www.elsevier.com/locate/jorganchem
Synthesis and MMA polymerization of chiral ansa-zirconocene ester
enolate complexes with C2- and Cs-ligation
*
Antonio Rodriguez-Delgado, Wesley R. Mariott, Eugene Y.-X. Chen
Department of Chemistry, Colorado State University, Fort Collins, CO 80523-1872, USA
Received 4 February 2006; received in revised form 29 April 2006; accepted 29 April 2006
Available online 5 May 2006
Abstract
A series of chiral ansa-zirconocene ester enolate complexes incorporating C2- or Cs-symmetric ligands, including neutral rac-(EBI)Zr-
Cl[OC(OiPr)@CMe2] (1), rac-(EBI)Zr(OTf)[OC(OiPr)@CMe2] (2), rac-(EBI)Zr(OTf)[OC(OMe)@C(Me)CH2C(Me2)C(OiPr)@O] (3),
[Me2C(Cp)(Flu)]ZrMe[OC(OiPr)@CMe2] (4), and cationic [Me2C(Cp)(Flu)]Zr+(THF)[OC(OiPr)@CMe2][MeB(C6F5)3]ꢀ (5), have been
synthesized. Within the neutral C2-ligated zirconocene ester enolate series, the chloride derivative 1 is inactive toward any methyl meth-
acrylate (MMA) additions, the methyl derivative rac-(EBI)ZrMe[OC(OiPr)@CMe2] adds cleanly only 1 equiv. of MMA, and the triflate
derivative 2 can add either 1 equiv. of MMA to form the single-MMA-addition product 3 or multiple equivalents of MMA to form
P(MMA). Unlike the Cs-ligated methyl cation [Me2C(Cp)(Flu)ZrMe]+, which is inactive for MMA polymerization under various con-
ditions, the Cs-ligated ester enolate cation 5 is moderately active for polymerization of MMA and N,N-dimethylacrylamide at ambient
temperature; the resulting P(MMA) has a high molecular weight of Mn = 388000 Da but a low syndiotacticity of [rr] = 64%, and the
polymerization conforms to a chain-end control mechanism.
ꢀ 2006 Elsevier B.V. All rights reserved.
Keywords: Metallocene catalysts; Chiral zirconocenes; MMA polymerization; Ester enolates; Chain-end control; Site control
1. Introduction
[Me2C(Cp)(Flu)] (Cp = g5-cyclopentadienyl; Flu = g5- or
g3-fluorenyl). In attempts to produce syndiotactic poly
(methyl methacrylate), P(MMA), at least five research
groups [5] have investigated the MMA polymerization
using the Cs-symmetric zirconocene dimethyl complex
[Me2C(Cp)(Flu)]ZrMe2 [6], in combination with borane
or borate-based activators; however, they all found no
polymerization activity. We reasoned that the inactivity
of the derived methyl cation, [Me2C(Cp)(Flu)ZrMe]+,
could simply be due to a problem with the initiation step
that involves presumably an ineffective Zr–Me transfer to
the coordinated monomer; hence, the active cationic ester
enolate propagating species cannot be generated (Eq. (1)).
In this contribution, we report our efforts to solve this
long-standing problem by first, the synthesis of the
preformed neutral zirconocene ester enolate complex
[Me2C(Cp)(Flu)]ZrMe[OC(OiPr)@CMe2] and second,
the investigation into the MMA polymerization
There has been increasing interest in controlled polymer-
ization of methyl methacrylate (MMA) and N,N-dimethyl-
acrylamide (DMAA) using group 4 metallocene and
related ester or amide enolate catalysts incorporating metal-
locene-type achiral C2v [1] and chiral C2 [2] or C1 [3] as well
as constrained-geometry-type prochiral Cs [4] ligand sym-
metries. Some of these polymerization systems are living
and stereospecific, allowing for a high degree of control
over polymer molecular weight, molecular weight distribu-
tion, and tacticity as well as for the synthesis of well-
defined block and stereoblock copolymers.
Obviously missing from the above catalyst list is a cat-
alyst system based on the metallocene-type Cs-ligand:
*
Tel.: +1 970 491 5609; fax: +1 970 491 1801.
E-mail address: eychen@lamar.colostate.edu (E.Y.-X. Chen).
0022-328X/$ - see front matter ꢀ 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2006.04.036

A. Rodriguez-Delgado et al. / Journal of Organometallic Chemistry 691 (2006) 3490–3497
3491
characteristics using the corresponding cationic complex
[Me2C(Cp)(Flu)]Zr+(THF)[OC(OiPr)@CMe2][MeB(C6-
F5)3]ꢀ, which models the proposed active propagating
species.
scale) were conducted in Teflon-valve-sealed J. Young-type
NMR tubes. HPLC grade organic solvents were sparged
extensively with nitrogen during filling of the solvent reser-
voir and then dried by passage through activated alumina
(for THF, Et2O, and CH2Cl2) followed by passage through
Q-5-supported copper catalyst (for toluene and hexanes)
stainless steel columns. Benzene-d6 and toluene-d8 were
degassed, dried over sodium/potassium alloy, and filtered
before use, whereas CD2Cl2 was degassed and dried over
OMe
OMe
O
O
MMA
Zr
Zr
Me
MMA
(initiation)?
˚
activated Davison 4 A molecular sieves. NMR spectra were
1
anticipated active species
recorded on either a Varian Inova 300 (FT 300 MHz, H;
75 MHz, 13C; 282 MHz, 19F) or a Varian Inova 400 spec-
ð1Þ
1
trometer. Chemical shifts for H and 13C spectra were ref-
A second, closely related issue that we are addressing
in the current contribution pertains to our continued
efforts seeking for effective MMA polymerization cata-
lysts based on neutral chiral zirconocene ester enolate
complexes. Such neutral catalysts are of interest for the
development of an effective catalyst system devoid of
any cocatalysts. We previously found that the neutral
methyl zirconocene ester enolate complex, rac-(EBI)Zr-
Me[OC(OiPr)@CMe2] [EBI = C2H4(Ind)2] [2c], reacts
cleanly with 1 equiv. of MMA in non-coordinating polar
solvents such as CH2Cl2 at ambient temperature to form
the single-MMA-addition product, rac-(EBI)ZrMe-
[OC(OMe)@C(Me)CH2C(Me2)C(OiPr)@O] [2b]; no fur-
ther MMA additions took place, however, in the
presence of excess MMA with extended reaction times
(24 h). We hypothesized that the size (Cl vs. Me) and
lability (OTf vs. Me and Cl) of the X ligand in a neutral
zirconocene ester enolate complex series, rac-(EBI)Zr-
X[OC(OiPr)@CMe2] (X = Me, Cl, OTf), should affect
the reactivity of these complexes towards possible further
MMA additions (Eq. (2)). To seek for appropriate neu-
tral zirconocene ester enolate complexes that can effect
further MMA additions to yield P(MMA), we synthe-
sized the other two neutral zirconocene ester enolate
derivatives (X = Cl, OTf) to complete the series and sub-
sequently examined their reactivity towards MMA
additions.
erenced to internal solvent resonances and are reported as
parts per million relative to tetramethylsilane, whereas 19F
NMR spectra were referenced to external CFCl3. Elemen-
tal analyses were performed by Desert Analytics, Tucson,
Arizona.
n-Butyllithium (1.6 M in hexanes), chlorotrimethylsi-
lane, 2-cyclopentadienyl-2-fluorenylpropane, 1,2-dibromo-
ethane, diisopropylamine, 2,6-di-tert-butyl-4-methylphenol
(butylated hydroxytoluene, BHT-H), indene, isobutyriyl
chloride, isopropanol, lithium dimethylamide, triflic acid,
trimethyl amine, and zirconium tetrachloride were pur-
chased from Aldrich Chemical Co., trimethylaluminum
(neat) from Strem Chemical Co., methyllithium (1.6 M
in diethyl ether) from Acros. and trimethylsilyl trifluo-
romethanesulfonate (Me3SiOTf) from Alfa Aesar. These
reagents were used as received, except for the following
reagents that were further treated or purified: the amines
were degassed using three freeze–pump–thaw cycles and
BHT-H was recrystallized from hexanes.
Methyl methacrylate (MMA) and N,N-dimethylacryla-
mide (DMAA) were purchased from Aldrich Chemical
Co. and TCI America, respectively; they were purified by
first degassing and drying over CaH2 overnight, followed
by vacuum distillation; final purification of MMA involved
titration with neat tri(n-octyl)aluminum to a yellow end
point [7] followed by a second vacuum distillation. The
purified monomers were stored in brown bottles over acti-
˚
vated Davison 4-A molecular sieves (for DMAA) in a
X
X
OPri
n
MMA
MMA
ꢀ30 ꢁC freezer inside the glovebox.
P(MMA)
Zr
Zr
X = ?
Tris(pentafluorophenyl)borane, B(C6F5)3, was obtained
as a research gift from Boulder Scientific Co. and further
purified by recrystallization from hexanes at ꢀ30 ꢁC. The
(C6F5)3B Æ THF adduct was prepared by addition of THF
to a toluene solution of the borane followed by removal
of the volatiles and drying in vacuo. Alkyl (isopropyl and
tert-butyl) isobutyrates and their corresponding lithium
salts were prepared according to modified literature proce-
dures [8]; the isolated lithium ester enolates were stored in a
freezer at ꢀ30 ꢁC inside the glovebox. Literature proce-
dures were employed for the preparation of the following
zirconium complex precursors: rac-(EBI)ZrX2 (X =
NMe2, Cl, Me) [9], [Me2C(Cp)(Flu)]ZrMe2 [6], rac-(EBI)-
ZrMe[OC(OiPr)@CMe2] [2c], and rac-(EBI)Zr[OC(OiPr)@
CMe2]2 [2c].
O
O
O
OPri
OMe
ð2Þ
2. Experimental
2.1. Materials and methods
All syntheses and manipulations of air- and moisture-
sensitive materials were carried out in flamed Schlenk-type
glassware on a dual-manifold Schlenk line, a high-vacuum
line (typically 10ꢀ5 to 10ꢀ7 Torr), or in an argon or nitro-
gen-filled glovebox (typically <1.0 ppm oxygen and mois-
ture). NMR-scale reactions (typically in a 0.02 mmol

3492
A. Rodriguez-Delgado et al. / Journal of Organometallic Chemistry 691 (2006) 3490–3497
2.2. Synthesis of rac-(EBI)ZrCl[OC(OiPr)@CMe2] (1)
23 ꢁC): d 154.20 (ZrOC), 129.15, 128.96, 126.75, 126.81,
126.20, 125.85, 124.76, 124.59, 123.90, 121.99, 120.71,
120.48, 116.52, 116.09, 114.98, 110.06, 106.87, 106.11 (18
resonances for indenyl-ring carbons), 87.26 (@CMe2),
68.81 (OCHMe2), 29.01 and 28.58 (CH2CH2), 21.82,
(OCHMe2), 21.57 (OCHMe2), 18.91 (CF3) 17.68 (@CMe2),
17.38 (@CMe2). 19F NMR (CD2Cl2, 23 ꢁC): d ꢀ74.95.
In an argon-filled glovebox, a 30 mL glass reactor was
charged with rac-(EBI)ZrCl2 (0.126 g, 0.301 mmol) and
10 mL toluene. To this stirred reactor was added a tolu-
ene (5 mL) solution of LiOC(OiPr)@CMe2 (0.049 g,
0.361 mmol). The resulting turbid pale yellow mixture
was stirred for 24 h at ambient temperature, after which
all volatiles were removed in vacuo, giving a yellow oily
residue. The residue was extracted with hexanes, and the
resulting suspension was filtered twice through a pad of
Celite; the filtrate was cooled to ꢀ30 ꢁC inside the freezer
of the glovebox for about 10 h to afford complex 1 as a
bright yellow microcrystalline solid. The combined yield
from several crops of recrystallization, after filtration
and drying, was 0.12 g (78%). Anal. Calc. for
C27H29O2ClZr: C, 63.33; H, 5.71. Found: C, 63.31; H,
5.71%.
2.4. Isolation of rac-(EBI)Zr(OTf)
[OC(OMe)@C(Me)CH2C(Me2)C(OiPr)@O] (3): the
single MMA addition product of 2
In an argon-filled glovebox, a 20 mL glass vial was
charged with 2 (0.080 g, 0.127 mmol) and 10 mL CH2Cl2.
To this pre-cooled (ꢀ30 ꢁC) and stirred yellow solution
was added MMA (0.013 g, 0.127 mmol). The resulting
solution was stirred for 1 h at ambient temperature, after
which an aliquot was taken and dried in vacuo for 1H
and 19F NMR analysis which showed the formation of sin-
gle MMA addition product 3. All volatiles of the remaining
solution were removed in vacuo, giving a yellow oil. The
residue was extracted with hexanes and filtered through a
pad of Celite; the filtrate was cooled to ꢀ30 ꢁC inside the
freezer of the glovebox for about 48 h to afford the product
as a bright orange microcrystalline solid. The combined
yield from several crops of recrystallization, after filtration
and drying, was 0.055 g (60%). Significantly, the isolated
complex 3 contains (as shown by NMR) a small amount
of oligomers (i.e., higher MMA addition products), indi-
cating that multiple MMA additions can occur and thus
a possibility to produce high polymer in the presence of
an excess of MMA using neutral zirconocenes either 2 or 3.
1H NMR (CD2Cl2, 23 ꢁC) for 3: d 7.81 (d, J = 8.4 Hz,
2H), 7.44–7.17 (m, 6H), 6.76 (s, br, 1H), 6.42 (s, br, 1H),
6.36 (s, br, 1H), 6.27 (s, br, 1H), 4.93 (sept, J = 6.3 Hz,
1H, OCHMe2), 3.87 (m, 4H, C2H4), 3.12 (s, 3H, OMe),
2.18 (qAB, 2H, CH2), 1.24 (d, J = 6.3 Hz, 3H, OCHMe2),
1.21 (d, J = 6.3 Hz, 3H, OCHMe2), 1.12 (s, 3H, @CMe),
1.11 and 1.09 (s, 6H, ˜CMe2). 19F NMR (CD2Cl2,
23 ꢁC): d ꢀ75.05.
1H NMR (C6D6, 23 ꢁC) for 1: d 7.32–7.08 (m, 6H), 6.95–
6.88 (m, 2H), 6.58 (d, J = 3.3 Hz, 1H), 6.28 (d, J = 3.3 Hz,
1H), 6.01 (d, J = 3.3 Hz, 1H), 5.83 (d, J = 3.3 Hz, 1H), 3.99
(sept, J = 6.3 Hz, 1H, OCHMe2), 3.14–3.04 (m, 4H, C2H4),
1.79 (s, 3H, @CMe2), 1.47 (s, 3H, @CMe2), 1.16 (d,
J = 6.3 Hz, 3H, OCHMe2), 1.13 (d, J = 6.3 Hz, 3H,
OCHMe2). 13C{1H} NMR (C6D6, 23 ꢁC): d 154.01
(ZrOC), 130.81, 129.41, 129.30, 126.18, 125.91, 125.46,
125.36, 123.99, 123.08, 123.07, 121.93, 121.24, 120.52,
119.29, 114.33, 114.05, 108.86, 104.93 (18 resonances for
indenyl-ring carbons), 85.45 (@CMe2), 68.40 (OCHMe2),
28.89 and 28.57 (CH2CH2), 22.88, (OCHMe2), 22.01,
(OCHMe2), 17.77 (@CMe2), 17.35 (@CMe2).
2.3. Synthesis of rac-(EBI)Zr(OTf)[OC(OiPr)@CMe2]
(2)
In an argon-filled glovebox, a 30 mL glass reactor was
charged with rac-(EBI)Zr[OC(OiPr)@CMe2]2 (0.100 g,
0.165 mmol) and 10 mL toluene. To this pre-cooled
(ꢀ30 ꢁC) and stirred yellow solution was added Me3SiOTf
(0.046 g, 0.265 mmol) via pipette. The resulting dark
orange solution was stirred for 80 h at ambient tempera-
ture, after which all volatiles were removed in vacuo,
affording an orange oily residue. The residue was extracted
with hexanes, and the resulting suspension was filtered
twice through a pad of Celite; the filtrate was cooled to
ꢀ30 ꢁC inside the freezer of the glovebox for about 16 h
to afford 0.092 g (89%) complex 2 as a bright orange micro-
crystalline solid after filtration and drying. Anal. Calc. for
C28H29O5F3SZr: C, 53.74; H, 4.67. Found: C, 54.37; H,
4.89%.
2.5. Synthesis of [Me2C(Cp)(Flu)]ZrMe-
[OC(OiPr)@CMe2] (4)
In an argon-filled glovebox, a 30 mL glass reactor was
charged with [Me2C(Cp)(Flu)]ZrMe2 (0.195 g, 0.498 mmol)
and 10 mL toluene. To this pre-cooled (ꢀ30 ꢁC) and stirred
solution was added Me3SiOTf (0.143 g, 0.647 mmol) via
pipette. The resulting dark orange solution was stirred
for 24 h at ambient temperature, after which all volatiles
were removed in vacuo, affording a red oily solid. The
crude product was extracted with hexanes, and the result-
ing suspension was filtered twice through a pad of Celite;
the red filtrate was cooled to ꢀ30 ꢁC inside the freezer of
the glovebox to afford the spectroscopically pure triflate
derivative [Me2C(Cp)(Flu)]Zr(OTf)Me as a red microcrys-
talline solid. The combined yield from several crops of
1H NMR (CD2Cl2, 23 ꢁC) for 2: d 7.87–7.82 (m, 2H),
7.82–7.17 (m, 6H), 6.76 (d, J = 3.3 Hz, 1H), 6.45 (d,
J = 3.3 Hz, 1H), 6.42 (d, J = 3.3 Hz, 1H), 6.28 (d,
J = 3.3 Hz, 1H), 3.91–3.88 (m, 4H, C2H4), 3.63 (sept,
J = 6.3 Hz, 1H, OCHMe2), 1.50 (s, 3H, @CMe2), 1.29 (s,
3H, @CMe2), 1.13 (d, J = 6.3 Hz, 3H, OCHMe2), 1.06
(d, J = 6.3 Hz, 3H, OCHMe2). 13C{1H} NMR (C6D6,

A. Rodriguez-Delgado et al. / Journal of Organometallic Chemistry 691 (2006) 3490–3497
3493
recrystallization, after filtration and drying, was 0.196 g
(75%).
67.74 (OCHMe2), 40.89 (˜CMe2), 29.11 and 28.90
(˜CMe2), 26.22 (Zr–Me), 21.96 and 21.58 (OCHMe2),
16.88 and 16.59 (@CMe2).
1H NMR (C6D6, 23 ꢁC) for [Me2C(Cp)(Flu)]Zr-
(OTf)Me: d 7.81 (d, 1H, Flu), 7.58 (d, 1H, Cp), 7.31–6.80
(m, 6H, Flu), 6.18 (q, 1H, Cp), 6.02 (q, 1H, Cp), 5.42 (q,
1H, Cp), 4.74 (q, 1H, Cp), 1.65 (s, 3H, ˜CMe2), 1.56 (s,
3H, ˜CMe2), ꢀ0.77 (Zr–Me) 19F NMR (C6D6, 23 ꢁC): d
ꢀ77.21.
2.6. In situ generation of ion pair
[Me2C(Cp)(Flu)]Zr+(THF)
[OC(OiPr)@CMe2][MeB(C6F5)3]ꢀ (5)
In an argon-filled glovebox, a 30 mL glass reactor was
charged with [Me2C(Cp)(Flu)]Zr(OTf)Me (0.100 g,
0.190 mmol) and 10 mL toluene. To this pre-cooled
(ꢀ30 ꢁC) and stirred red solution was added a toluene
(2 mL)solutionofLiOC(OiPr)@CMe2 (0.033 g,0.247 mmol).
The resulting solution changed from clear red to a turbid
orange suspension upon warming to ambient temperature
over 20 min with a moderate exotherm being detected.
All volatiles were immediately removed in vacuo, the resi-
due extracted with hexanes, and the resulting suspension
filtered twice through a pad of Celite. The filtrate was
cooled to ꢀ30 ꢁC inside the freezer of the glovebox for
about 48 h to afford complex 4 as an orange solid. The
combined yield from several crops of recrystallization, after
filtration and drying, was 0.063 g (65%). The isolated prod-
The cationic ester enolate species 5 can be readily gener-
ated by in situ mixing of equimolar amounts of neutral
complex 4 and (C6F5)3B Æ THF in CD2Cl2, as shown by
1
NMR. H NMR (CD2Cl2, 23 ꢁC) for 5: d 8.33 (d, 1H,
Flu), 8.24 (d, 1H, Flu), 8.05 (t, 2H, Flu), 7.53–7.39 (m,
4H, Flu), 6.60 (q, 1H, Cp), 6.01 (m, 2H, Cp), 5.87 (q,
1H, Cp), 3.56 (sept, J = 6.3 Hz, 1H, OCHMe2), 3.58–3.45
(m, 4H, a-CH2, THF), 2.51 (s, 3H, ˜CMe2), 2.47 (s, 3H,
˜CMe2), 1.87 (m, 2H, b-CH2, THF), 1.76 (m, 2H,
b-CH2, THF), 1.38 (s, 3H, @CMe2), 1.34 (s, 3H, @CMe2),
1.12 (d, J = 6.3 Hz, 3H, OCHMe2), 1.02 (d, J = 6.3 Hz,
3H, OCHMe2), 0.48 (s, br, B–Me). 19F NMR (CD2Cl2,
3
23 ꢁC): d ꢀ131.52 (d, JF–F = 20.9 Hz, 6F, o-F), ꢀ163.60
3
(t, JF–F = 20.9 Hz, 3F, p-F), ꢀ166.20 (m, 6F, m-F).
1
uct is spectroscopically pure, as shown by its H NMR
2.7. General polymerization procedures
spectrum (see Fig. 1); however, owing to its thermal insta-
bility, this ester enolate complex is not suitable for subject
to elemental analysis.
Polymerizations were performed in 20-mL glass reactors
inside the glovebox at ambient temperature (ꢁ23 ꢁC). In a
typical procedure for the polymerization using the in situ
1H NMR (CD2Cl2, 23 ꢁC) for 4: d 8.11 (d, 1H, Flu), 8.02
(d, 1H, Flu), 7.97 (d, 1H, Flu), 7.44 (t, 1H, Flu), 7.29–7.11
(m, 4H, Flu), 6.18 (q, 1H, Cp), 6.07 (q, 1H, Cp), 5.68 (q,
1H, Cp), 5.60 (q, 1H, Cp), 3.51 (sept, J = 6.3 Hz, 1H,
OCHMe2), 2.30 (s, 3H, ˜CMe2), 2.19 (s, 3H, ˜CMe2),
1.32 (s, 3H, @CMe2), 1.17 (s, 3H, @CMe2), 1.04 (d,
J = 6.0 Hz, 3H, OCHMe2), 0.90 (d, J = 6.3 Hz, 3H,
OCHMe2), ꢀ1.28 (Zr–Me). 13C{1H} NMR (CD2Cl2,
23 ꢁC): d 152.49 (ZrOC), 127.95, 126.81, 124.49, 124.08,
123.87, 123.22, 122.42, 122.27, 121.72, 121.06, 120.12,
119.36, 119.11, 115.94, 111.58, 102.18, 100.38, 84.15 (18
resonances for Flu and Cp carbons), 84.56 (@CMe2),
generated cationic complex 5, complex
4
(10 mg,
19.8 lmol) and (C6F5)3B Æ THF (11.5 mg, 19.8 lmol) were
dissolved in 1.25 mL CH2Cl2 and allowed to stir for
10 min at ambient temperature to cleanly generate the cat-
alyst 5. To this reactor was quickly added 200 equiv. of
MMA (3.96 mmol) via pipette, and the sealed flask was
kept with vigorous stirring. After the desired time interval,
a 0.2 mL aliquot was taken from the reaction mixture via
syringe and quickly quenched into a 4 mL vial containing
0.6 mL of undried ‘‘wet’’ CDCl3 stabilized by 250 ppm of
BHT-H. The quenched aliquots were analyzed by 1H
NMR to obtain monomer conversion data. The polymeri-
zation was immediately quenched after the removal of the
aliquot by the addition of 5 mL 5% HCl-acidified metha-
nol. For MMA polymerization, the quenched mixture
was precipitated into 100 mL of methanol, stirred for 1 h,
filtered, washed with methanol, and dried in a vacuum
oven at 50 ꢁC overnight to a constant weight. For DMAA
polymerization, the quenched mixture was precipitated
into 100 mL of diethyl ether, stirred for 30 min, and the
solvent was decanted off. An additional 75 mL of diethyl
ether was used to wash the polymer and then decanted;
the P(DMAA) product was dried in a vacuum oven at
50 ꢁC to a constant weight.
Me2C=
Me2C<
Zr-Me
Me2CH-
Flu
Me2CH
Cp
CHDCl2
2.8. Polymer characterizations
8.0
7.0
6.0
5.0
4.0
3.0
2.0
1.0
0.0
-1.0
ppm
Fig. 1. 1H NMR (CD2Cl2) spectrum of [Me2C(Cp)(Flu)]ZrMe-
[OC(OiPr)@CMe2] (4).
Gel permeation chromatography (GPC) analyses of the
polymers were carried out at 40 ꢁC and a flow rate of

3494
A. Rodriguez-Delgado et al. / Journal of Organometallic Chemistry 691 (2006) 3490–3497
1.0 mL/min, with CHCl3 as the eluent, on a Waters Uni-
versity 1500 GPC instrument equipped with four 5 lm
PL gel columns (Polymer Laboratories) and calibrated
using 10 PMMA standards. Chromatograms were pro-
cessed with Waters Empower software (2002); number-
average molecular weight and polydispersity of polymers
are given relative to PMMA standards. NMR spectra of
the polymers were recorded in CDCl3 and analyzed accord-
ing to literature procedures [10].
C(Me)CH2C(Me2)C(OiPr)@O] (3, Scheme 1). However,
unlike the methyl derivative, which cleanly affords the sin-
gle-MMA-addition product in the presence of 1 or more
equiv. of MMA [2b], the addition product 3 derived from
the reaction of complex 2 with 1 equiv. of MMA is accom-
panied by the formation of a small amount of higher
MMA addition products, thereby suggesting a possibility
to produce polymer in the presence of an excess of MMA
1
(vide infra). Nevertheless, both H and 19F NMR spectra
of 3 feature characteristic signals for a triflate zirconocene
ester enolate structure (which differs from that of the
precursor 2) and additionally an isopropyl ester group
3. Results and discussion
1
3.1. Neutral zirconocene ester enolate complexes with
C2-ligation
[–C(OiPr)@O]. Specifically, the H NMR (CD2Cl2) signal
for the methine proton in –CHMe2 (sept, 3.6 ppm)
attached to the enolate ligand in 2 is substantially down-
field shifted by 1.3 ppm to 4.9 ppm (sept) for the methine
proton in –CHMe2 now attached to the ester group in 3.
In agreement with these spectroscopic changes on going
from 2 to 3, the added MMA now becomes a methyl ester
enolate ligand [i.e., –OC(OMe)@C(Me)CH2–] bound to Zr
in 3, which can be readily characterized by 1H NMR [d 3.12
(OMe), 2.18 (AB quartet for CH2), 1.12 (@CMe)]. The tri-
flate ligand in 3 is still covalently bound to Zr, as the 19F
NMR for the OTf signal (ꢀ75.0 ppm in CD2Cl2) remained
the same on going from 2 to 3, whereas the identical chem-
ical shift for the –CHMe2 proton of the isopropyl ester
group in 3 to that in free isopropyl isobutyrate (4.9 ppm
in CD2Cl2) suggests that the terminal ester group in 3 is
presumably not coordinated to the Zr center. Most signif-
icantly, unlike the single-MMA-addition product derived
from the methyl derivative, the triflate derivative 3 can
undergo further MMA additions to form polymer (vide
infra).
The synthesis of the chloride derivative rac-(EBI)Zr-
Cl[OC(OiPr)@CMe2] (1) was accomplished in a straightfor-
ward fashion by the reaction of the dichloride zirconocene
precursor rac-(EBI)ZrCl2 with 1.2 equiv. of LiOC(OiPr)@
CMe2 in toluene (Scheme 1). The NMR spectroscopic
characteristics of 1 are nearly identical to those of the
known methyl derivative rac-(EBI)ZrMe[OC(OiPr)@
CMe2] [2c], except for noticeable down-field shifts of the
1H NMR signals for the methine proton in –CHMe2 (by
0.3 ppm) and those for the indenyl C5-ring protons as com-
pared with the methyl derivative. The triflate derivative
rac-(EBI)Zr(OTf)[OC(OiPr)@CMe2] (2) was conveniently
synthesized by the reaction of rac-(EBI)Zr[OC(OiPr)@
CMe2]2 [2c] with Me3SiOTf, where the reaction coproduct
silyl ketene acetal was removed by extensive vacuum
(Scheme 1).
Unlike the methyl derivative, the chloride analog 1 does
not react with MMA in CH2Cl2 at ambient temperature.
On the other hand, the triflate derivate 2 reacts with
1 equiv. of MMA to form the corresponding single-
MMA-addition product rac-(EBI)Zr(OTf)[OC(OMe)@
3.2. Neutral and cationic zirconocene ester enolate
complexes with Cs-ligation
Scheme 2 outlines the procedures for the synthesis of the
neutral zirconocene ester enolate complex [Me2C(Cp)
(Flu)]ZrMe[OC(OiPr)@CMe2] (4) and the generation of
Cl
LiO
Cl
Cl
toluene
LiCl
Zr
Zr
Zr
Zr
+
PriO
O
OPri
1
2
Me
Me
OTf
Me
toluene
Me4Si
PriO
O
Zr
Zr
+ Me3SiOTf
OTf
O
Me3SiOTf, toluene
Me3SiO
O
LiO
toluene
LiOTf
OPri
PriO
PriO
PriO
[MeB(C6F5)3]
OPri
OPri
O
X
O
O
X
(C6F5)3B THF
CH2Cl2
OPri
OMe
Zr
Zr
Zr
Zr
Me
THF
O
O
O
CH2Cl2
OPri
X = Cl, Me, OTf
OMe
X = Me; OTf (3)
5
4
Scheme 1.
Scheme 2.

A. Rodriguez-Delgado et al. / Journal of Organometallic Chemistry 691 (2006) 3490–3497
3495
the corresponding cationic complex [Me2C(Cp)(Flu)]-
3.3. Polymerization of MMA and DMAA
Zr+(THF)[OC(OiPr)@CMe2][MeB(C6F5)3]ꢀ (5). Thus,
treatment of [Me2C(Cp)(Flu)] ZrMe2 with Me3SiOTf read-
ily affords the methyl triflate intermediate [Me2C(Cp)(Flu)]-
Zr(OTf)Me, which subsequently reacts with LiOC(OiPr)@
CMe2 to form the methyl zirconocene ester enolate com-
plex 4. The 1H NMR spectrum (Fig. 1) shows C1-symmetry
for 4 in a CD2Cl2 solution, featuring an AA0BB0 pattern
for the four Cp protons as well as two doublets and two
singlets for the two diastereotopic isopropoxy methyl and
two bridging isopropylidene methyl groups, respectively.
Clean, instantaneous, and quantitative generation of the
THF-stabilized cationic ester enolate complex 5 was
accomplished by treatment of the methyl zirconocene ester
enolate 4 with (C6F5)3B Æ THF in CD2Cl2 at ambient tem-
perature (Fig. 2). Most characteristically upon formation
Among the neutral C2-ligated chiral zirconocene ester
enolate complex series, rac-(EBI)ZrX[OC(OiPr)@CMe2]
(X = Me, Cl, OTf), the investigation into their reactivity
toward MMA in
a non-coordinating polar solvent
(CH2Cl2) at ambient temperature showed that the chloride
derivative 1 is inactive toward any MMA additions, the
methyl derivative adds only 1 equiv. of MMA, and the tri-
flate derivative 2 can add either 1 equiv. of MMA to form
the single-MMA-addition product 3 or multiple equiva-
lents of MMA to form a polymer product. Monitoring of
the reaction between 2 or 3 and 200 equiv. of MMA by
1H NMR in CD2Cl2 clearly showed gradual conversion
of MMA to P(MMA), whereas the OTf signal monitored
by 19F NMR gradually shifted from ꢀ75.0 ppm as a sharp
singlet at the beginning of the polymerization to
ꢀ76.1 ppm as a broad singlet during the course of the poly-
merization, suggesting that the OTf ligand becomes more
partially dissociated from the Zr center as the chain grows.
The preparative scale polymerization in a [MMA]0/[3]0
ratio of 200 for 24 h at ambient temperature indeed yielded
P(MMA) with Mn = 10300 and Mw/Mn = 1.24, upon
which the monomer conversion reached only 22% (run 1,
Table 1), giving an initiator efficiency (I*) = 49% [I* =
Mn(calcd.)/Mn(exptl.), where Mn(calcd.) = MW(MMA) ·
[MMA]0/[3]0 · conversion%]. The low activity (low conver-
sion, long reaction time) of this neutral species is in drastic
contrast to that of the cationic zirconocenium ester enolate
species, which typically effects quantitative monomer con-
versions within just a few minutes of reaction time under
similar conditions and also produces highly isotactic
1
of the cation, the sharp singlet at ꢀ1.28 ppm in H NMR
for the Zr–Me protons in 4 disappeared and a broad singlet
appeared at 0.48 ppm for the B–Me protons of the
[MeB(C6F5)3]ꢀ anion in 5. On going from neutral 4 to cat-
ion 5, protons for the rest of the moieties all down-field
shifted, consistent with the formation of the more-electron
deficient, cationic species 5. Conversely, on going from
(C6F5)3B Æ THF to [MeB(C6F5)3]ꢀ, para- and meta-fluorines
in 19F NMR upfield shifted by 8.7 and 4.1 ppm (see Fig. 2),
respectively, consistent with the borate anion formation.
These 1H and 19F NMR observations, coupled with a small
chemical shift difference of <3 ppm [D (m,p-F) = 2.6 ppm
in 5] between the para- and meta-fluorines in the 19F
NMR spectrum for the anion, clearly show the formation
of the cation
5 paired with the non-coordinating
[MeB(C6F5)3]ꢀ anion [11].
Fig. 2. 1H (bottom) and 19F NMR (top) spectra of 5 generated in situ by mixing 4 and (C6F5)3B Æ THF in CD2Cl2; peaks marked with correspond to
*
(C6F5)3B Æ THF added in excess for comparison.

3496
A. Rodriguez-Delgado et al. / Journal of Organometallic Chemistry 691 (2006) 3490–3497
Table 1
MMA and DMAA polymerization results by zirconocene ester enolate complexesa
104 Mn (g/mol)
PDIb (Mw/Mn)
[mm]c (%)
[mr]c (%)
[rr]c (%)
b
Run no.
Monomer
Complex
Time (h)
Conv.(%)
1
2
3
4
5
a
MMA
MMA
MMA
DMAA
DMAA
3
24
4
4
1.5
4
25
65
55
51
23
1.03
38.8
1.95
>500
>200
1.24
1.57
1.28
Bimodald
Bimodald
46.5
4.0
7.0
n.d.
n.d.
26.7
32.0
31.8
n.d.
n.d.
26.8
64.0
61.2
n.d.
n.d.
5
4 + 5
5
5e
Carried out in CH2Cl2 at ambient temperature inside a glovebox; [monomer]0/[Zr]0 = 200.
Number average molecular weight (Mn) and polydispersity index (PDI) determined by GPC relative to PMMA standards.
Methyl triad distribution (tacticity) determined by 1H NMR spectroscopy.
b
c
d
The reaction mixture gelled at 1.5 h; there exists also a low molecular weight (Mn = 5.4 kg/mol for run 4 and 0.8 kg/mol for run 5) fraction on GPC
traces.
e
The solvent volume is increased by 4-fold.
polymer via an enantiomorphic-site control mechanism [2].
Surprisingly, despite its chiral C2-ligation in neutral 3, the
resulting polymer is not highly isotactic as is the polymer
by the cationic species; rather, the polymer produced by
neutral 3 shows an unusual tacticity with a methyl triad
distribution of [mm] = 46.5%, [mr] = 26.7%, [rr] = 26.8%.
This tacticity is presumably a result of double stereo-differ-
entiation, namely the interplay between the enantiomor-
phic-site control (by the chiral metal center) and chain-end
control (by the stereogenic center of the last enchained
monomer) mechanisms. To the best of our knowledge,
the only other C2-ligated chiral neutral group 4 metallo-
cene complex that has been found active for MMA poly-
merization is a zirconocene complex incorporating the
chelating isopropylidene-bridged Cp o-carboranyl ligand:
rac-Zr(g5:g1-CpCMe2CB10H10C)2 [12]; this complex, how-
ever, polymerizes MMA in polar, coordinating THF
(which typically retards or shuts down the coordinative
anionic polymerization of MMA by metallocene catalysts)
and yields unexpectedly syndio-rich ([rr] = 65.7%) P(MMA)
despite its C2 symmetry. Furthermore, the initiator effi-
ciency I* based on the polymerization results for 24 h is cal-
culated to be ꢁ510%, indicating considerable chain transfer
processes. Hence, the polymerization characteristics of the
neutral carboranyl complex are inconsistent with a coordi-
native anionic mechanism and are also in sharp contrast to
those of the neutral ester enolate complex 3.
As pointed out earlier, the methyl cation [Me2C(Cp)
(Flu)ZrMe]+ is inactive for MMA polymerization under
various conditions. Satisfyingly, the ester enolate cation 5
is moderately active for MMA polymerization. Thus, the
in situ generated cationic ester enolate 5 polymerizes
MMA in CH2Cl2 ([MMA]0/[5]0 = 200) for 4 h at ambient
temperature to high molecular weight P(MMA) with
Mn = 388000 and Mw/Mn = 1.57, upon which the mono-
mer conversion reached 65% (run 2, Table 1); this gives a
low initiator efficiency (I*) <4%, implying that the active
propagating species is unstable at ambient temperature
and that the majority of the active species has been deacti-
vated due to decomposition. The resulting polymer has a
low syndiotacticity, with a methyl triad distribution of
[rr] = 64%, [mr] = 32%, [mm] = 4%. A triad test using
4[mm][rr]/[mr]2 gave 1.0, conforming to a chain-end control
mechanism; this result compares well to that observed for
the related cationic titanium ester enolate complex bearing
the prochiral Cs-constrained geometry ligand, [Me2Si(g5-
Me4C5)(tBuN)]Ti+(THF)[OC(OiPr)@CMe2][MeB(C6F5)3]ꢀ
[4a], but it is in sharp contrast to that observed for the iso-
structural constrained geometry zirconium complex which
produces highly isotactic P(MMA) via a site control mech-
anism at low temperatures [4b]. These results further show
the apparent high sensitivity of relative rates of racemiza-
tion (at the chiral metal center by the process of ligand
exchange) and propagation (at the C–C bond formation
step by intramolecular Michael addition) to polymerization
activity and stereospecificity in the MMA polymerization
by the Cs-ligated metallocene system; variations in the type
of metal and Cs-ligation can drastically change the poly-
merization activity and stereocontrol mechanism.
As the unimetallic propagation by the cationic ester eno-
late 5 is hampered by its low initiator efficiency, we also
investigated the MMA polymerization by the bimetallic
propagation pathway [1e] with a mixture containing both
neutral ester enolate 4 (as initiator) and cationic ester eno-
late 5 (as catalyst), which is generated in situ by the reac-
tion of 2 equiv. of 4 with 1 equiv. of (C6F5)3B Æ THF. The
MMA polymerization by this two-component system pro-
duces P(MMA) with significantly lower molecular weight
(Mn = 19500, Mw/Mn = 1.28), upon which the monomer
conversion reached 55% (run 3, Table 1); this gives a sub-
stantially higher initiator efficiency (I* = 56%), as com-
pared to the polymerization by 5 alone.
Complex 5 is also active for DMAA polymerization (run
4), producing essentially atactic P(DMAA). This polymer-
ization is complicated by gel formation, and the resulting
polymer has a bimodal molecular weight distribution con-
sisting of a predominant high molecular weight fraction
(Mn > 5 · 106) and a low molecular weight fraction
(Mn = 5400); this behavior can be explained by the sub-
stantial catalyst deactivation present in the current catalyst
system (vide supra) to form presumably neutral zircono-
cene species that can also readily initiate the DMAA poly-
merization but via a radical process [13]. Attempts to avoid
the gel formation by increasing the solvent volume by

A. Rodriguez-Delgado et al. / Journal of Organometallic Chemistry 691 (2006) 3490–3497
3497
4-fold (run 5) resulted in similar polymerization character-
References
istics but with much lower activity.
[1] (a) G. Stojcevic, H. Kim, N.J. Taylor, T.B. Marder, S. Collins,
Angew. Chem. Int. Ed. 43 (2004) 5523–5526;
(b) M. Ferenz, F. Bandermann, R. Sustmann, W. Sicking, Macromol.
Chem. Phys. 205 (2004) 1196–1205;
4. Conclusions
In efforts to address the inactivity of MMA polymeriza-
tion by the zirconocene methyl cation with Cs-ligation,
[Me2C(Cp)(Flu)ZrMe]+, and the neutral methyl zircono-
cene ester enolate with C2-ligation, rac-(EBI)ZrMe[O-
C(OiPr)@CMe2], we have synthesized a series of neutral
and cationic chiral ansa-zirconocene ester enolate
complexes 1–5 with C2- or Cs-ligation. Within the neutral
zirconocene ester enolate complex series, rac-(EBI)ZrX-
[OC(OiPr)@CMe2] (X = Cl, Me, OTf), the chloride
derivative 1 is inactive toward any MMA additions, the
methyl derivative adds cleanly only 1 equiv. of MMA,
and the triflate derivative 2 can add either 1 equiv. of
MMA to form the single-MMA-addition product 3 or mul-
tiple equivalents of MMA to form a polymer product. The
relatively high reactivity of the triflate derivative in this ser-
ies toward MMA additions can be attributed to the lability
of the OTf ligand which can gradually and partially disso-
ciate from the metal center to accommodate monomer
enchainment as the polymer chain grows.
(c) A.D. Bolig, E.Y.-X. Chen, J. Am. Chem. Soc. 124 (2002) 5612–
5613;
(d) F. Bandermann, M. Ferenz, R. Sustmann, W. Sicking, Macromol.
Symp. 174 (2001) 247–253;
(e) Y. Li, D.G. Ward, S.S. Reddy, S. Collins, Macromolecules 30
(1997) 1875–1883.
[2] (a) W.R. Mariott, E.Y.-X. Chen, Macromolecules 38 (2005) 6822–
6832;
(b) A. Rodriguez-Delgado, E.Y.-X. Chen, Macromolecules 38 (2005)
2587–2594;
(c) A.D. Bolig, E.Y.-X. Chen, J. Am. Chem. Soc. 126 (2004) 4897–
4906.
[3] J. Jin, W.R. Mariott, E.Y.-X. Chen, J. Polym. Chem. A 41 (2003)
3132–3142.
[4] (a) A. Rodriguez-Delgado, W.R. Mariott, E.Y.-X. Chen, Macromol-
ecules 37 (2004) 3092–3100;
(b) H. Nguyen, A.P. Jarvis, M.J.G. Lesley, W.M. Kelly, S.S.
Reddy, N.J. Taylor, S. Collins, Macromolecules 33 (2000) 1508–
1510.
[5] (a) A.D. Bolig, E.Y.-X. Chen, J. Am. Chem. Soc. 123 (2001) 7943–
7944;
(b) H. Frauenrath, H. Keul, H. Ho¨cker, Macromolecules 34 (2001)
14–19;
Unlike the methyl cation [Me2C(Cp)(Flu)ZrMe]+,
which is inactive for MMA polymerization under various
conditions, the ester enolate cation 5 having the same liga-
tion, which is derived from the methyl zirconocene ester
enolate 4, is active for polymerization of MMA and
DMAA under ambient conditions. The resulting
P(MMA) has high molecular weight but low syndiotactic-
ity, and the polymerization conforms to a chain-end con-
trol mechanism. These results, coupled with the findings
from the titanium and zirconium ester enolate complexes
bearing the related prochiral Cs-constrained geometry
ligand [4], indicate that current efforts to produce highly
syndiotactic P(MMA) using Cs-ligated metallocene cata-
lysts via a site-control mechanism are complicated by the
subtle interplay between the relative rates of racemization
at the chiral metal center and propagation at the C–C bond
formation step. Appropriate site-controlled syndiospecific
metallocene catalysts are those that can regulate the rela-
tive rates of those events, presumably achievable by varia-
tions of the ligand sterics; research toward this direction is
currently underway.
(c) P.A. Cameron, V. Gibson, A.J. Graham, Macromolecules 33
(2000) 4329–4335;
(d) Y.-X. Chen, M.V. Metz, L. Li, C.L. Stern, T.J. Marks, J. Am.
Chem. Soc. 120 (1998) 6287–6305;
(e) K. Soga, H. Deng, T. Yano, T. Shiono, Macromolecules 27
(1994) 7938–7940.
[6] (a) A. Razavi, U. Thewalt, J. Organomet. Chem. 445 (1993) 111–
114;
(b) J.A. Ewen, R.L. Jones, A. Razavi, J.D. Ferrara, J. Am. Chem.
Soc. 110 (1988) 6255–6256.
[7] R.D. Allen, T.E. Long, J.E. McGrath, Polym. Bull. 15 (1986) 127–
134.
[8] Y.-J. Kim, M.P. Bernstein, A.S. Galiano Roth, F.E. Romesberg,
P.G. Williard, D.J. Fuller, A.T. Harrison, D.B. Collum, J. Org.
Chem. 56 (1991) 4435–4439.
[9] G.M. Diamond, R.F. Jordan, J.L. Petersen, J. Am. Chem. Soc. 118
(1996) 8024–8033.
[10] (a) A. Bulai, M.L. Jimeno, A.-A.A. de Queiroz, A. Gallardo, J.S.
Roman, Macromolecules 29 (1996) 3240–3246;
(b) R.C. Ferguson, D.W. Ovenall, Macromolecules 20 (1987) 1245–
1248;
(c) R. Subramanian, R.D. Allen, J.E. McGrath, T.C. Ward, Polym.
Prepr. 26 (1985) 238–240.
[11] (a) J. Klosin, G.R. Roof, E.Y.-X. Chen, K.A. Abboud, Organomet-
allics 19 (2000) 4684–4686;
(b) A.D. Horton, J. de With, A.J. van der Linden, H. van de Weg,
Organometallics 15 (1996) 2672–2674.
Acknowledgments
[12] E. Hong, Y. Kim, Y. Do, Organometallics 17 (1998) 2933–
2935.
[13] W.R. Mariott, E.Y.-X. Chen, Macromolecules 37 (2004) 4741–
4743.
Funding for this work was provided by the National Sci-
ence Foundation and Colorado State University. We thank
Boulder Scientific Co. for the gift of B(C6F5)3.