High-speed living polymerization of polar vinyl monomers by self-healing silylium catalysts

Article abstract of DOI:10.1002/chem.201000961

This contribution describes the development and demonstration of the ambient-temperature, high-speed living polymerization of polar vinyl monomers (M) with a low silylium catalyst loading (≤ 0.05 mol % relative to M). The catalyst is generated in situ by protonation of a trialkylsilyl ketene acetal (^RSKA) initiator (I) with a strong Bryonsted acid. The living character of the polymerization system has been demonstrated by several key lines of evidence, including the observed linear growth of the chain length as a function of monomer conversion at a given [M]/[I] ratio, near-precise polymer number-average molecular weight (M_n, controlled by the [M]/[I] ratio) with narrow molecular weight distributions (MWD), absence of an induction period and chain-termination reactions (as revealed by kinetics), readily achievable chain extension, and the successful synthesis of well-defined block copolymers. Fundamental steps of activation, initiation, propagation, and catalyst self-repair involved in this living polymerization system have been elucidated, chiefly featuring a propagation catalysis cycle consisting of a rate-limiting C-C bond formation step and fast release of the silylium catalyst to the incoming monomer. Effects of acid activator, catalyst and monomer structure, and reaction temperature on polymerization characteristics have also been examined. Among the three strong acids incorporating a weakly coordinating borate or a chiral disulfonimide anion, the oxonium acid [H(Et₂O)₂] ⁺ [B(C ₆F₅)₄]^- is the most effective activator, which spontaneously delivers the most active R₃Si ⁺, reaching a high catalyst turn-over frequency (TOF) of 6.0 x 10³ h^-1 for methyl methacrylate polymerization by Me ₃Si⁺ or an exceptionally high TOF of 2.4 x 10⁵ h^-1 for n-butyl acrylate polymerization by iBu₃Si ⁺, in addition to its high (>90%) to quantitative efficiencies and a high degree of control over M_n, and MWD (1.07-1.12). An intriguing catalyst self-repair feature has also been demonstrated for the current living polymerization system.

Full text of DOI:10.1002/chem.201000961

DOI: 10.1002/chem.201000961
High-Speed Living Polymerization of Polar Vinyl Monomers by
Self-Healing Silylium Catalysts
Yuetao Zhang,[a] Frank Lay,[b] Pilar Garcꢀa-Garcꢀa,[b] Benjamin List,[b] and
Eugene Y.-X. Chen*[a]
Abstract: This contribution describes
the development and demonstration of
the ambient-temperature, high-speed
living polymerization of polar vinyl
monomers (M) with a low silylium cat-
alyst loading (ꢀ 0.05 mol% relative to
M). The catalyst is generated in situ by
protonation of a trialkylsilyl ketene
period and chain-termination reactions
(as revealed by kinetics), readily
AHCTUNGTRENGNaUN chievable chain extension, and the
successful synthesis of well-defined
block copolymers. Fundamental steps
of activation, initiation, propagation,
and catalyst “self-repair” involved in
this living polymerization system have
polymerization characteristics have
also been examined. Among the three
strong acids incorporating a weakly co-
ordinating borate or a chiral disulfoni-
mide anion, the oxonium acid [H-
ACHUTNGRENNUG CAHTUNGTRENNUGN
(Et2O)2]+[B(C6F5)4]À is the most effec-
tive activator, which spontaneously de-
livers the most active R3Si+, reaching a
high catalyst turn-over frequency
(TOF) of 6.0ꢀ103 hÀ1 for methyl meth-
acrylate polymerization by Me3Si+ or
an exceptionally high TOF of 2.4ꢀ
105 hÀ1 for n-butyl acrylate polymeri-
zation by iBu3Si+, in addition to its
high (>90%) to quantitative efficien-
cies and a high degree of control over
Mn and MWD (1.07–1.12). An intrigu-
ing catalyst “self-repair” feature has
also been demonstrated for the current
living polymerization system.
acetal (RSKA) initiator (I) with
a
been elucidated, chiefly featuring a
propagation “catalysis” cycle consisting
strong Brønsted acid. The living char-
acter of the polymerization system has
been demonstrated by several key lines
of evidence, including the observed
linear growth of the chain length as a
function of monomer conversion at a
given [M]/[I] ratio, near-precise poly-
mer number-average molecular weight
(Mn, controlled by the [M]/[I] ratio)
with narrow molecular weight distribu-
tions (MWD), absence of an induction
À
of a rate-limiting C C bond formation
step and fast release of the silylium cat-
alyst to the incoming monomer. Effects
of acid activator, catalyst and monomer
structure, and reaction temperature on
Keywords: anions · Brønsted acids ·
polar vinyl monomers · polymeri-
zation · silicon
Introduction
lyst-based stereochemical and architectural control, leading
to advanced or new classes of polymeric materials, often op-
erated at ambient or higher polymerization temperature
(Tp). In contrast, classical anionic polymerization, which is a
key technique used to polymerize polar vinyl monomers to
technologically important functionalized vinyl polymers,[2]
often requires low Tp or additional additives to render dis-
crete, single-site anionic active species,[3] for achieving a
good control over the polymerization characteristics, such as
the resulting polymer number average molecular weight
(Mn) and molecular weight distribution (MWD). Many strat-
egies have been developed to achieve various degrees of
control in anionic polymerization of polar vinyl monomers
at ambient temperature, and the majority of them targeted
the production of poly(methyl methacrylate) (PMMA) with
controlled structure.[4]
Discrete metal complexes incorporating specifically de-
signed organic ligands have been increasingly utilized as
highly efficient and active catalysts or initiators for coordi-
nation polymerization of polar vinyl monomers such as
(meth)acrylates and (meth)acrylamides.[1] Such coordination
polymerization is recognized as advantageous over other
polymerization methods, thanks to its precision in the cata-
[a] Dr. Y. Zhang, Prof. Dr. E. Y.-X. Chen
Department of Chemistry, Colorado State University
Fort Collins, Colorado 80523-1872 (USA)
Fax : (+1)970-491-1801
E-mail: eugene.chen@colostate.edu
[b] F. Lay, Dr. P. Garcꢁa-Garcꢁa, Prof. Dr. B. List
Max-Planck-Institut fꢂr Kohlenforschung
Of these strategies, one particularly successful, commer-
cialized living/controlled polymerization process employs a
Kaiser-Wilhelm-Platz 1, 45470 Mꢂlheim an der Ruhr (Germany)
10462
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 10462 – 10473

FULL PAPER
silyl ketene acetal (SKA) initiator and a catalyst at ambient
or higher Tp. This polymerization was developed by scien-
tists of DuPont and termed group-transfer polymerization
(GTP), based on the initially postulated associative propaga-
tion mechanism in which the silyl group that is coordinated
with a nucleophilic catalyst is intramolecularly transferred
to the incoming monomer through hypervalent anionic sili-
con species (path a, Scheme 1).[5] However, it has been sug-
lyst, we recently developed a new polymerization system,
which is catalyzed by silylium ions (R3Si+), for a highly
active, efficient, and living/controlled (meth)acrylate poly-
merization at room temperature.[12] The highly active, ambi-
philic propagating species contains both the nucleophilic
SKA moiety and the electrophilic silylium ion sites
(Scheme 2). This propagator is generated by a unique
“mono
trialkylsilyl SKA (RSKA) by a catalytic amount of [Ph3C]+
[B
(C6F5)4]À, leading to the R3Si+-activated MMA derived
ACHTUNGTRENmNUGN er-less” initiation involving oxidative activation of
AHCTUNGTRENNUNG
from vinylogous hydride abstraction of RSKA by Ph3C+
(i.e., the monomer is generated from the initiator), followed
by subsequent Michael addition of RSKA to the activated
MMA (or silylated MMA), as shown in Scheme 2. A propa-
gation “catalysis” cycle consists of a fast step of recapturing
the silylium catalyst from the ester group of the growing
polymer chain by the incoming MMA, followed by the rate-
À
determining step (r.d.s.) of the C C bond coupling by inter-
molecular Michael addition of the polymeric SKA to the si-
lylated MMA (see Scheme 2). This polymerization is propa-
Scheme 1. Proposed associative (a) and dissociative (b) pathways for
methyl methylacrylate (MMA) polymerization by using SKA.[5–7]
gested[6] that several key lines
of experimental evidence are
more consistent with a dissocia-
tive mechanism,[7] which in-
volves ester enolate anions as
the propagating species and a
rapid, reversible complexation
(termination) of small concen-
trations of enolate anions with
Scheme 2. Living/controlled (meth)acrylate polymerization catalyzed by R3Si+.[12,14]
SKA or its polymer homologue
(path b, Scheme 1). The ele-
À
mentary C C bond coupling
step in GTP is that of the Mukaiyama–Michael reaction,[8]
but activation of the inactive SKA is the critical first step in
either the associative or dissociative pathway described
above. Such activation can be viewed as reductive activation
of SKA for both pathways in GTP, namely, conversion of
the inactive, neutral Si center to the active, anionic Si
center. Regardless of the dispute over the mechanism, poly-
merization of MMA through GTP by using a SKA and a nu-
cleophilic catalyst, such as a bifluoride or oxyanion with a
low catalyst loading (0.1–1.0 mol% relative to initiator),[9]
readily produces PMMA with Mn ꢀ 20000 gmolÀ1 in a con-
trolled fashion at Tp ꢁ ambient temperatures. However, the
synthesis of PMMA with Mn ꢁ 60000 gmolÀ1 following this
procedure is considered difficult.[6] (Recently, N-heterocyclic
carbenes were utilized as nucleophilic catalysts in GTP,
which can produce polymers with higher MWs.[10]) On the
other hand, polymerization of acrylates by using GTP re-
quires the use of Lewis acidic catalysts, such as zinc halides
or alkylaluminum chlorides, and a much higher catalyst
loading (typically 10 mol% based on monomer) to achieve
a reasonable degree of polymerization control.[11]
gated in the same fashion as by using MeSKA and the exter-
nally added silylium catalyst, [Et3Si
(C6F5)4]À,[13] generated from the in situ reaction of Et3SiH
and [Ph3C]+[B(C6F5)4]À in toluene.[12] Our subsequent struc-
(toluene)]+[B-
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
ture–reactivity relationship studies have revealed a remark-
able selectivity of RSKAs (R=Me, Et, iBu, Ph, trimethylsilyl
(TMS)) on the monomer structures for the polymerization
of (meth)acrylates by RSKA/[Ph3C]+[B(C6F5)4]À.[14] Specifi-
ACTHNUTRGENNUG CAHTUNGTRENNUGN
cally, the Me3Si+ catalyst, derived from MeSKA that bears a
small silyl group, is highly active and efficient for the poly-
merization of MMA but inefficient for the polymerization
of the sterically less demanding n-butyl acrylate (nBA). In
contrast, the iBu3Si+ catalyst, derived from iBuSKA that
bears a bulky silyl group, exhibits low activity in the poly-
merization of MMA but exceptional activity, efficiency, and
control for the polymerization of nBA.[14] Overall, this novel
polymerization system can readily produce controlled low-
to-high molecular weights (Mn > 105 gmolÀ1) and well-de-
fined (Mw/Mn =1.04–1.12) homo and copolymers of (meth)a-
crylates; thereby achieving a high silylium catalyst turn-over
frequency (TOF) up to 1.5ꢀ103 hÀ1 for methacrylates and
an exceptionally high TOF up to 1.2ꢀ105 hÀ1 for acrylates,
with the reaction completed within 1 min, at 258C with a
Departing from the conventional GTP methodology and
utilizing SKA as both the initiator and the source of cata-
Chem. Eur. J. 2010, 16, 10462 – 10473
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10463

E. Y.-X. Chen et al.
low catalyst loading (typically 0.05 mol% based on mono-
mer).
Results and Discussion
In organic catalysis, Yamamoto and coworkers[15] have ex-
tensively utilized strong Brønsted acids[16] such as trifluoro-
methanesulfonimide (HNTf2) for the catalytic Mukaiyama
aldol and Michael reactions involving nucleophiles, such as
enol silyl ethers or SKAs, and acceptors, such as aldehydes,
ketones, or their a,b-unsaturated derivatives. Since our cur-
rent work, which is focused on
Generation of silylium catalysts: Protonation of MeSKA by
acid 1 in CD2Cl2 at ambient temperature is instantaneous
and produces cleanly the base-separated silylium ion pair
[Me3Si···L]+[B
Me2CHC
(C6F5)4]À (4), where
L
is the coproduct
AHCTUNGTRENN(GUN OMe)=O or Et2O (Scheme 3). The base-stabilized
the development of an alterna-
tive method for delivering the
R3Si+ catalyst by using strong
Brønsted acids, including the
oxonium acid [H
(C6F5)4]À (1),[17] [HN
[B
(C6F5)4]À (2),[18] and the
chiral acid 3 that has the chiral
disulfonimide counteranion
(Et2O)2]+[B-
ACHTUNGTRENNUNG
A
(Me2)Ph]+
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
(which recently achieved the
asymmetric Mukaiyama aldol
Scheme 3. Generation of the silylium catalyst Me3Si+ by protonation of
MeSKA with strong Brønsted acids.
reaction),[19] was in progress,
Kakuchi et al.[20] reported the
use of HNTf2 to activate SKA (thus, generating Me3SiNTf2
in situ, which is
a
superior carbonyl activator to
Me3SiOTf[21]) for living polymerization of MMA through
the same silylium-catalyzed propagation process described
above (see Scheme 2). However, the polymerization by the
SKA/HNTf2 system typically employed a reaction time of
24 h and led to PMMA with low Mn of ꢀ 17000 gmolÀ1
(but with excellent control, having Mw/Mn =1.03–1.08), thus,
giving a low catalyst TOF of typically only ꢂ80 hÀ1. Intrigu-
ingly, the previously reported, highly effective, acid-cata-
lyzed “GTP” systems by using SKA as initiator and addi-
tionally employing different combinations of a Lewis acid
Me3Si+ cation exhibits a characteristic 1H NMR signal at
d=0.68 ppm (s, 9H; SiMe3) and a 29Si NMR (HSQCAD)
signal at d=68.5 ppm, whereas the noncoordinated borate
anion is readily characterized by its 19F NMR spectrum [d=
À131.4 (brs, 8F; o-F), À161.9 (t, J=20.3 Hz, 4F; p-F),
À165.8 ppm (t, J=18.0 Hz, 8F; m-F)].[12] Interestingly, the
isolated, dried product showed a 3:1 mixture of 4a (L=
Me2CHCACHTNUTRGNENG(U OMe)=O) and 4b (L=Et2O), whereas the crystals
grown slowly over a period of one week were those of pure
4b (see the Experimental Section). Also noteworthy here is
that protonation of MeSKA with the chiral acid 3 forms
cleanly the silylated disulfonimide, Me3Si-[3-H], and the
ester coproduct is uncoordinated to the silylium ion, indica-
tive for the coordinating nature of the chiral disulfonimide
anion in acid 3.
As efforts to solve the molecular structure of the (twin)
crystals of 4b by X-ray diffraction analysis were unsuccess-
ful, we examined the same reaction by using acid 2, which
produces Me2NPh as the coproduct, a stronger base than
the ester or ether from the reaction with acid 1. Indeed, the
resulting silylium cation in 5 coordinates only to N,N-di-
methyl aniline and, gratifyingly, colorless single crystals of 5
were found suitable for X-ray diffraction analysis, yielding a
well-solved structure (Figure 1; goodness-of-fit on F2 =1.022
and final R1 =0.0381). The base-stabilized Si center adopts a
distorted tetrahedral geometry with the sum of the C-Si-C
angles being 337.598. The average C-Si-C angle of ꢂ1138 in
5 compares to 1138 in [Me3Si(py)]+IÀ (py=pyridine),[31] 1148
and
(C6F5)3,[22] Me3SiI/HgI2,[23] or Me3SiI/RAl
involve the silylium-catalyzed process as demonstrated in
the SKA/
[Ph3C]+[B(C6F5)4]À system,[12,14] thereby pointing
a
Me3Si-containing reagent, such as Me3SiOTf/B-
G
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
to potentially broad implications of such a silylium-catalyzed
process in polymerization.
The central objective of our current study was to investi-
gate the characteristics of polymerizations of polar vinyl
monomers catalyzed by silylium ions that are derived from
the reaction of SKA with the strong Brønsted acids 1, 2, and
3. Our choice of these three acids in this study, in which we
were seeking more active and controlled silylium catalyst
systems for polymerization of polar vinyl monomers, was
motivated by the following four established facts: 1) acids 1
and 2 have been widely used as potent activators in metal-
catalyzed polymerization of olefins[25] and polar vinyl mono-
ACTHNUTRGNEUNG
mers;[26] 2) the counteranion [B(C6F5)4]À in acids 1 and 2 is
among the least coordinating anions;[25,27] 3) the silylium
cation chemistry is largely dictated by the coordinating
nature of the counteranion;[28] and 4) specifically designed
chiral counteranions, such as that in 3 or its variants, can
powerfully mediate asymmetric catalysis[29] by silylium cat-
ions[19] or transition-metal cations.[30]
in [Me3Si
G
N
ACHTUNGTRENNUNG
A
R
CHTUNGTRENNUNG
iPr3Si(Br6CB11H6),[33] and 1188 in [Me3Si]+[C2H5CB11F11]À;[34]
the last structure of this series has an angle close to 1208 for
À
a nearly trigonal-planar trimethyl silyl cation. The three Si
10464
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 10462 – 10473

Living Polymerization by Self-Healing Silylium Catalysts
FULL PAPER
~
Figure 2. Comparisons of the MMA polymerization by method A (
;
slope=À0.0217, R2 =0.996) and method B
(
slope=À0.0311, R2 =
~
;
Figure 1. X-ray crystal structure of [Me3Si···NACTHUNRTGENNUG AHCTUNGTRENNGUN
(Me2)Ph]+[B(C6F5)4]À (5).
0.996) in zero-order plots of [M]t/[M]0 versus time. Conditions: solvent=
[
MeSKA]0 =5.14 mm, [1]0 =
À
À
Selected bond lengths [ꢄ] and angels [8]: Si(1) N(1) 1.9211(11), Si(1)
CH2Cl2, Tp =258C, [MMA]0 =0.935m,
MeSKA]0/[1]0 =400/2.2/0.2, [M]/[I]=200.
À
À
À
C(1) 1.8386(16), Si(1) C(2) 1.8335(16), Si(1) C(3) 1.8402(15), B(1)
0.467 mm, [MMA]0/A
CTHUNGTRENNUNG[
À
À
À
C(11) 1.6572(19), B(1) C(21) 1.6553(19), B(1) C(31) 1.6524(19), B(1)
C(41) 1.6513(19); C(1)-Si(1)-C(2) 113.03(10), C(1)-Si(1)-C(3) 112.58(8),
C(2)-Si(1)-C(3) 111.93(8), N(1)-Si(1)-C(1) 106.73(7), N(1)-Si(1)-C(2)
105.39(6), N(1)-Si(1)-C(3) 106.54(6). Hydrogen atoms are omitted for
clarity.
activation). Two observations are worth noting: first, both
procedures give the polymerization kinetics of a clearly
zero-order dependence on [M], as shown in Figure 2;
second, under otherwise identical conditions as listed in the
caption of Figure 2, the polymerization by method B (i.e.,
when the silylium catalyst is generated in the presence of
the monomer) is ꢂ1.4 times faster than that by method A
(where the silylium catalyst is preformed followed by addi-
tion of the monomer). However, it should be pointed out
that both methods achieved a quantitative monomer conver-
sion and produced PMMA with targeted MW (Mn =2.26ꢀ
104 gmolÀ1 as determined by the [M]/[I] ratio of 200) and a
narrow MWD (Mw/Mn =1.08), achieving a high initiator effi-
ciency (I*) of ꢂ90%. Considering a similarly high degree of
control and efficiency of the polymerization by both meth-
ods but faster reaction and more convenient procedures pos-
sessed by method B, we accordingly adopted this method for
the rest of the current studies.
C bonds in the cation of 5 have an essentially identical bond
À
length (1.837 ꢄ average), as do the four B C bonds
(1.655 ꢄ average) in the unassociated borate anion that has
essentially identical metric parameters to the same anion
À
found in [Me3Si
of 1.921(1) ꢄ in 5 is ꢂ0.06 ꢄ longer than that found in
(thf)]+[B
(C6F5)4]À.[12] The Si N bond length
[Me3Si(py)]+IÀ[31] (1.858(9) ꢄ), and significantly longer than
À
a typical covalent Si N bond (ꢂ1.75 ꢄ) in a tetracoordinat-
À
ed silicon center, indicative of a dative Si N bonding in 5.
Overall, the structural analysis of 5 confirms the proposed
protonation reaction of SKA with the acid, which generates
the silylium catalyst from the initiator SKA; hence, the
SKA in this polymerization system serves as both the initia-
tor and the catalyst source.
Table 1 summarizes the selected results of the MMA poly-
merization by MeSKA and acid 1 at ambient temperature.
As can be seen from Table 1, the polymerization at all inves-
tigated [M]/[I] ratios (100–800, runs 1–5) consumed all
Characteristics of homopolymerizations: In the current poly-
merization system, the added acid activator converts an
equimolar amount of the initiator SKA to the silylium cata-
lyst (Scheme 3). Hence, for a polymerization with an x[M]0/
monoACTHNUTRGENUmGN er while using a low catalyst loading of 0.05 mol%
yACHTUNGTRENNUNG[SKA]0/zACHTUNGTRENNUNG[acid]0 ratio, the real
equivalency of the initiating
SKA is yÀz, thus, giving an
actual [monomer (M)]/[initia-
Table 1. Selected results of MMA polymerization by MeSKA and acid 1 (0.05 mol% based on MMA).[a]
[b]
[c]
[c]
Run
[M]/[I]
t
Conv.
[%]
Mw/Mn
Mn
[kgmolÀ1
Mn(calcd)
[kgmolÀ1
I*[d]
[rr][b]
[mr][b]
[mm][b]
[%]
tor (I)] ratio of x/
(yÀz). Keep-
A
R
]
G
]
[%]
[%]
[%]
ing this in mind, we initially
compared the activity and effi-
ciency of the MMA polymeri-
zation by MeSKA and acid 1 by
using two different polymeri-
zation procedures detailed in
the Experimental Section:
1
2
3
4
5
100
133
200
400
800
20
35
40
75
100
100
100
100
100
1.07
1.07
1.08
1.07
1.12
11.3
14.4
22.1
42.0
74.7
10.1
13.4
20.1
40.2
80.2
90
93
91
96
107
71.7
72.2
72.2
71.5
72.4
26.6
26.0
25.9
26.5
25.6
1.7
1.8
1.9
2.0
2.0
180
[a] Conditions: solvent=CH2Cl2, Tp =258C, [MMA]0 =0.935m, [1]0 =0.467 mm, varied [MeSKA]0 to adjust the
[M]/[I] ratio to 100, 133, 200, 400, and 800. [b] Monomer conversions and PMMA methyl triad distributions
measured by 1H NMR spectroscopy. [rr]=syndiotactic racemo triad, [mr]=heterotactic triad, and [mm]=iso-
tactic meso triad. [c] Mn and MWD determined by GPC relative to PMMA standards. [d] Initiator efficiency
method A
(preactivation)
versus method B (in-reactor (I*)=Mn(calcd)/Mn(exptl), where Mn(calcd) =MWACTHNUGTRNEUNG(MMA)ꢀ[MMA]0/[I]0 ꢀconversion%+MW of chain-end groups.
Chem. Eur. J. 2010, 16, 10462 – 10473
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10465

E. Y.-X. Chen et al.
based on monomer, in relative short time periods, thus,
giving rise to a high silylium catalyst TOF of ꢂ6000 hÀ1.
This TOF value represents an approximately 4-fold rate en-
(ꢁ 90%) to near quantitative I*. All PMMAs produced by
this catalyst system at ambient temperature are syndio-rich,
having a syndiotacticity of ꢂ72% rr.
hancement over the MeSKA/
A
U
Selected plots (Figures 5 and 6) of the kinetic profile of
the MMA polymerization by MeSKA and acid 1 clearly show
a strictly linear increase of Mn with an increase in MMA
80-fold over the MeSKA/HNTf2 system (see above), high-
lighting the importance of the chosen method for delivering
the silylium catalyst, when comparing the protonation of
SKA by the acid to the vinylogous hydride abstraction of
SKA by the trityl salt; and the coordinating nature of the
counteranion, when comparing the weakly coordinating [B-
ACHTUNGTRENNUNG
(C6F5)4]À anion to the coordinating [NTf2]À anion. This poly-
merization also exhibits a high degree of control over the re-
sulting polymer MW with the [M]/[I] ratio, as shown by a
linear relationship between the polymer Mn and the [M]/[I]
ratio (Figure 3), thus, achieving nearly targeted Mnꢅs and
narrow MWDꢅs of 1.07–1.12 (Figure 4). Also noteworthy is
the high efficiency of this catalyst system, affording high
~
~
Figure 5. Plots of Mn
( ) and PDI ( ) versus MMA conversion. Condi-
tions: [M]/[I]=100, CH2Cl2, 258C (R2 =0.998).
Figure 3. Plot of Mn of PMMA versus the [M]0/[I]0 ratios. The Mn and
PDI (Mw/Mn) values for each [M]0/[I]0 ratio are included in parentheses
(Mn, PDI).
~
~
Figure 6. Plots of Mn
( ) and PDI ( ) versus MMA conversion. Condi-
tions: [M]/[I]=200, CH2Cl2, 258C (R2 =0.999).
conversion, whereas the MWD remains narrow and nearly
constant for all conversions. Overall, all results (those dis-
cussed above and further results on kinetics, chain exten-
sion, and well-defined block copolymer formation to be dis-
cussed below) demonstrate the living characters of this poly-
merization catalyzed by the silylium catalyst.
Polymerization kinetics and mechanism: Table 2 compiles
some selected kinetic results of the MMA polymerization by
MeSKA and acid 1 (0.05 mol% relative to monomer) in
Figure 4. Overlay of GPC traces of the PMMAs reported in Table 1. Mn
[kgmolÀ1] and PDI for the traces from right (low MW) to left (high MW)
are: 11.3, 1.07; 14.3, 1.07; 22.1, 1.08; 42.0, 1.07; and 74.7, 1.12.
10466
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 10462 – 10473

Living Polymerization by Self-Healing Silylium Catalysts
FULL PAPER
Table 2. Selected kinetic data for MMA polymerization by MeSKA and 1
(0.05 mol% based on MMA) in CH2Cl2 at 258C.[a]
Run [I]
[mm]
[M]/[I]0
t
A
Conv. Mn
MWD
ACHTUGNERTN(NUGN Mw/Mn) [%]
I*
A
]
1
2
3
4
5
6
7
8
9.35
100
1
3
5
17.1
28.0
39.4
49.6
71.8
86.9
97.8
10.2
19.3
25.8
34.5
51.0
67.6
87.0
98.7
21.6
35.8
66.9
95.2
1.74
2.63
3.89
5.31
7.72
9.17
1.04
1.09
1.08
1.12
1.10
1.08
1.07
1.03
1.07
1.11
1.11
1.07
1.08
1.08
1.07
1.08
1.06
1.09
1.08
1.08
1.09
1.06
1.07
1.09
1.08
1.08
1.07
1.09
1.12
1.08
1.09
1.08
1.07
1.08
1.10
1.12
104
110
104
96
95
96
96
86
109
98
99
100
94
96
92
103
92
90
92
91
128
94
95
91
93
99
96
110
93
97
93
92
93
95
101
107
7
10
13
16
1
3
5
10.3
1.69
7.01
133
9
2.44
3.63
4.76
6.87
9.70
12.2
14.3
4.30
7.96
15.0
20.9
22.1
4.60
8.24
14.9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Figure 7. Zero-order plots of [M]t/[M]0 versus time for the MMA poly-
merization by MeSKA and 1 in CH2Cl2 at 258C and varied concentrations
of acid 1 (thus, the Me3Si+ catalyst): [MMA]0 =0.935m, [MeSKA]0 =
7
10
15
20
30
5
10
20
30
40
5
10
20
30
45
60
75
5
10
20
30
45
60
90
~
~
&
^
6.08 mm ( ), 5.61 mm ( ), 5.14 mm ( ), 4.91 mm (&), 4.79 mm ( ); [1]0 =
~
~
&
^
1.402 mm ( ), 0.935 mm ( ), 0.467 mm ( ), 0.234 mm (&), 0.117 mm ( ).
4.67
2.34
200
400
100
14.5
19.2
34.9
51.7
74.7
93.9
100
5.6
8.8
15.8
23.3
36.9
48.1
72.3
91.4
22.9
32.5
38.2
4.20
4.12
7.64
1.17
800
Figure 8. Plot of lnACTHNUTRGNEUNG
(kapp) versus ln([1]0) (slope=1.13, R2 =0.972).
13.2
20.2
32.1
41.5
60.8
72.6
74.7
sulting PMMAꢅs syndiotacticity, with a constant syndiotac-
ticity of 72% rr upon variations of [Me3Si+] from 0.0125 to
0.15 mol% relative to monomer.
120
180
In another set of kinetic experiments, we kept [M]0 and
[1]0 constant but varied [MeSKA]0 (thus, the initiator concen-
tration) as [I]=[MeSKA]0À[1]0. These kinetic experiments
with [M]0/[I]0 ratio=100, 133, 200, 400, and 800, again re-
vealed that the polymerization follows the strict zero-order
kinetics with respect to [M] (Figure 9) and also the first-
order dependence on [I], given by the slope of 0.95 from a
100
[a] See footnotes of Table 1 for explanations of abbreviations and meth-
ods.
CH2Cl2 at 258C with five different [M]/[I] ratios ranging
from 100 to 800. Overall, these more detailed studies further
confirm the living character of the polymerization manifest-
ed in Table 1 and reveal that all polymerizations, from low
to high monomer conversions and in all [M]/[I] ratios inves-
tigated, are well controlled, thereby producing PMMA with
targeted Mnꢅs and narrow MWDꢅs (1.04–1.12) and achieving
high to near quantitative initiator efficiencies.
plot of lnACHTUNTRGNEUNG(kapp) versus ln[I] (Figure 10). These results indi-
In one set of kinetic experiments, we employed a constant
[M]0/[I]0 ratio of 200 by adjusting relative amounts of
[
MeSKA]0 and [1]0 (thus, the Me3Si+ catalyst concentration).
Kinetic plots of these polymerizations clearly showed no in-
duction period and a strict zero-order dependence on [M]
for all the polymerizations investigated herein (Figure 7).
However, the polymerization is first order with respect to
the catalyst concentration [Me3Si+], given by the slope of
1.13 from the double logarithm plot (Figure 8) of kapp as a
function of [1]0 (i.e., [Me3Si+]), thereby establishing an acti-
vated monomer polymerization (i.e., silylium catalysis). As
expected, the catalyst concentration does not affect the re-
Figure 9. Zero-order plot of [M]t/[M]0 versus time for the MMA polymer-
ization by MeSKA and 1 in CH2Cl2 at 258C and varied SKA concentra-
2
tions: [MMA]0 =0.935m, [1]0 =0.467 mm, [MeSKA]0 =9.82 mm ( , R =
~
2
2
2
~
&
0.993), 7.48 mm ( , R =0.995), 5.14 mm ( , R =0.999), 2.80 mm (&, R =
2
^
0.998), 1.64 mm ( , R =0.999).
Chem. Eur. J. 2010, 16, 10462 – 10473
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10467

E. Y.-X. Chen et al.
Figure 10. Plot of ln
(kapp
)
versus ln([I]), where [I]=[MeSKA]0À[1]0
(slope=0.95, R2 =0.998).
Scheme 4. Fundamental steps of activation, initiation, propagation, and
catalyst “self-repair” involved in the living polymerization catalyzed by
cate that the propagation mechanism (“catalysis” cycle) is
R3Si+–AnÀ (An=B
ACTHNUTRGNEU(NG C6F5)4).
the same as that proposed for the polymerization by SKA/ACHTUNG-
[Ph3C]+[B
R
À
mation step through intermolecular Michael addition of the
polymeric SKA to the silylium-activated (or silylated)
MMA is the rate-determining step, and the release of the si-
lylium catalyst from the ester group of the growing polymer
chain to the incoming MMA is relatively fast, thereby high-
lighting the high efficiency of the silylium catalysis in this
polymerization.
Being highly reactive and moisture-sensitive species, silyli-
um catalysts, R3Si+, can be consumed by even an extremely
low level of protic impurities.
firmed by the following chain-extension and copolymeriza-
tion experiments. In the chain-extension experiment, the
first 200 equiv of MMA were polymerized by MeSKA and
acid 1 in CH2Cl2 at 258C to give the precisely controlled
PMMA with Mn =19.9 kgmolÀ1, MWD=1.08, I*=101%,
and Tg =1288C (run 1, Table 3). Subsequently, the second
200 equiv of the MMA feed was added, and the polymeri-
zation resumed to give the precisely chain-extended PMMA
Significantly, our catalyst “self-
Table 3. Selected copolymerization results by RSKA and acid 1 at 258C.[a]
repair” study (see the Experi-
mental Section) clearly demon-
strated the self-healing ability
of the silylium catalyst. Specifi-
cally, the reaction of H2O (as a
protic impurity) with equimolar
amounts of catalyst 4, derived
from protonation of MeSKA by
acid 1, showed the clean regen-
eration of acid 1 and formation
of Me3SiOH. Upon addition of
another equivalent of MeSKA,
Run
RSKA
(R)
Solvent
M1
M2
t1, t2
[min]
Conv.
[%]
Mn(total)
[kgmolÀ1
MWD
(Mw/Mn)
I*
[%]
Tg
[8C]
G
G
]
E
1
2
3
4
5
6
7
Me
Me
Me
Me
iBu
iBu
iBu
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
toluene
toluene
toluene
MMA
MMA
none
MMA
MMA
none
none
MMA
BMA
BMA
none
nBA
35
35, 115
140
270
210
100
100
100
100
100
100
100
19.9
41.1
44.3
69.0
20.0
23.2
42.5
1.08
1.09
1.07
1.07
1.08
1.08
1.08
101
99
64
128
n.d.[b]
n.d.
70
70
n.d.
n.d.
122, À46
100
110
108
1
MMA
nBA
210, 1
[a] Conditions: [M1]/[M2]/[I]=200:200:1; [1]=0.05 mol% based on the first monomer feed, or 0.025 mol%
based on the total monomer feed. [b] n.d.=not determined.
catalyst
4
was regenerated,
which exhibited a similar polymerization activity to the orig-
inally generated catalyst before the “self-repair” process.
Overall, a combination of the detailed kinetic results with
the activation and catalyst “self-repair” studies led to the
proposed mechanism for the MMA polymerization by
RSKA and acid 1, outlined in Scheme 4. Although the prop-
agation proceeds in the same fashion as that proposed for
with Mn =41.4 kgmolÀ1, MWD=1.09, and I*=99% (run 2,
Table 3). The doubled MW after the chain extension is
shown in the overlaid GPC traces of the PMMA before and
after the chain extension depicted in Figure 11. The poly-
merization of the bulkier n-butyl methacrylate (BMA) is
slower but, nonetheless, proceeds to completion in 140 min
(run 3, Table 3), and the resulting poly(n-butyl methyl acry-
late) (PBMA) has a higher syndiotacticity (78% rr) than the
PMMA (72% rr) produced by this catalyst system. We also
showed that statistical copolymerization of MMA with
BMA (run 4, Table 3) leads to the formation of the well-de-
fined, random copolymer with a narrow MWD of 1.07 and a
Tg of 708C, which is approximately the midpoint of the Tgꢅs
for PMMA and PBMA.
the polymerization by RSKA/[Ph3C]+[B(C6F5)4]À, the cur-
ACTHNUTRGENNUG CAHTUNGTRENNUGN
rent RSKA/acid 1 system, which is faster and more effective,
proceeds through a different activation (catalyst generation)
pathway, and exhibits a unique catalyst self-healing feature,
thereby offering more attractive polymerization catalysis
features.
Chain extension and copolymerization: The living character
of the current polymerization system has been further con-
Guided by our previous structure–reactivity relationship
ACTHNGUTRENNUG CAHTUNGTRENNUNG
studies on the RSKA/[Ph3C]+[B(C6F5)4]À system,[14] we
10468
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 10462 – 10473

Living Polymerization by Self-Healing Silylium Catalysts
FULL PAPER
Figure 11. Overlay of GPC traces of the PMMA before and after the
chain-extension experiment reported in Table 3 (right: run 1, left: run 2).
Figure 12. Overlay of GPC traces of P
ACHTUNGTNER(NUNG nBA) (right: run 6, Table 3) and
PMMA-b-P(nBA) (left: run 7, Table 3).
AHCTUNGTRENNUNG
Table 4. Selected polymerization results by MeSKA and different acids.[a]
turned to the iBu3Si+ catalyst
derived from iBuSKA for the Run Acid Tp
M
[M]/[I]
t
Conv. TOF
Mn
[ꢀ103]
Mw/Mn I*
[rr]
[mr] [mm]
[8C]
[min] [%]
A
]
U
[%] [%] [%]
[%]
polymerization of nBA. Indeed,
this catalyst exhibits a lower ac-
tivity for MMA polymerization
than Me3Si+, whereas it main-
tains the excellent control and
quantitative efficiency (run 5,
Table 3), but an exceptional ac-
1
2
3
4
5
6
7
1
1
1
2
3
1
3
25 MMA
200
200
200
200
200
40
180
1980
900
1440
15
100
91.4
8.9
97.6
0
6000
1220
10
260
0
22.1
20.0
3.41 1.13
1.08
1.09
91 72.2 25.9
92 75.3 22.9
55 86.9 11.0
106 70.4 27.4
1.9
1.8
2.1
2.2
0
MMA
À40 MMA
25 MMA
25 MMA
25 DMAA 800
25 DMAA 200
18.5
1.07
100
100
64000 102
12000 30.2
1.09
1.13
78
66
20
61.5
38.5
tivity and control for polymeri- [a] Conditions: [acid]=0.05 mol% based on monomer, solvent=CH2Cl2.
zation of nBA (run 6, Table 3).
Built upon the living character
exhibited by the iBu3Si+ catalyst for both MMA and nBA
homopolymerizations, the block copolymerization by this
catalyst led to the well-defined block copolymer, PMMA-b-
acid structures (see Scheme 3). On one hand, the initiating
RSKA delivers the cation of the catalyst, R3Si+. On the
other hand, the activator acid delivers the anion of the cata-
lyst. Hence, we examined the effects of both cation and
anion on the polymerization characteristics. Consistent with
PACHTUNGTRENNUNG(nBA), with a targeted Mn and a narrow MWD of 1.08
(run 7, Table 3), as found in the GPC traces shown in
Figure 12. The copolymerization is also highly efficient with
a near quantitative efficiency and, as expected, the block co-
polymer exhibits two Tgꢅs of 1228C and À468C, correspond-
what we have previously observed for the SKA/
AHCTUNGTRENNUNG
[Ph3C]+[B-
ACHTUNGTRENNUNG
RSKA on monomer structure. Thus, the small Me3Si+ cata-
lyst derived from MeSKA and acid 1 is highly active (TOF=
6000 hÀ1), efficient (I*>90%), and living for the polymeri-
zation of MMA (run 1, Table 4), but ill-controlled for the
polymerization of the sterically less demanding nBA (e.g.,
at 45% conversion, the polymer produced gave a relatively
broad MWD of 1.53). In sharp contrast, the bulkier iBu3Si+
catalyst derived from iBuSKA and acid 1 exhibits lower activ-
ity in the polymerization of MMA (TOF=2000 hÀ1, run 5,
Table 3), but exceptional activity (TOF=2.4ꢀ105 hÀ1), effi-
ciency, and control for the polymerization of nBA (run 6,
Table 3).
ing to the hard PMMA and soft P
ly.
ACHTUNGERTN(NUNG nBA) blocks, respective-
Effects of reaction temperature and catalyst structure:
Owing to the chain-end-controlled nature of this polymeri-
zation (i.e., 4 [mm][rr]/[mr]2 ꢂ1), the Tp should have a great
influence on the resulting polymer tacticity. Accordingly, de-
creasing Tp from 258C (72.2% rr, run 1, Table 4) to 08C
(75.3% rr, run 2, Table 4) resulted in an increase in syndio-
tacticity of 3%, but at the expense of an approximately 5-
fold activity reduction. A further drop in Tp to À408C gave
a syndiotactic PMMA with 87% rr (run 3, Table 4), but the
polymerization activity was drastically diminished by more
than two orders of magnitude.
Staying with the initiator MeSKA but switching the activat-
ing acid from 1 to 2 resulted in a 23-fold reduction in activi-
ty (run 4 vs. run 1, Table 4). As both acids are paired with
The structure of the silylium catalyst in the current poly-
merization system is determined by both the SKA and the
the same anion [B
(C6F5)4]À and both react with MeSKA to
ACHTUNGTRENNUNG
instantaneously deliver the cation Me3Si+, the observed con-
Chem. Eur. J. 2010, 16, 10462 – 10473
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10469

E. Y.-X. Chen et al.
trast in activity must be attributed to the strong coordinating
N,N-dimethyl aniline, the coproduct from acid (see
(catalyst TOF=6000 hÀ1), a high efficiency (>90% to quan-
titative I* values), and a high degree of control (near-precise
Mn as determined by the [M]/[I] ratio and narrow MWDꢅs
of 1.07–1.12). Acid 2 is much less effective, due to its strong
coordinating coproduct N,N-dimethyl aniline formed upon
protonation of SKA, whereas acid 3 is ineffective for MMA
polymerization, due to its coordinating disulfonimide anion.
However, a high activity of the MeSKA/3 system has been
achieved in the polymerization of DMAA (TOF=
12000 hÀ1), attributed to the ability of this monomer to out-
compete the anion for coordination to the silylium ion.
The structure of the silylium ions also exerts a strong
effect on polymerization activity and control, as revealed by
a remarkable selectivity of RSKA on monomer structure.
Thus, the small Me3Si+ catalyst derived from MeSKA and
acid 1 is highly active, efficient, and living for the polymeri-
zation of MMA, but it is ill-controlled for the polymeri-
zation of nBA. On the other hand, the larger iBu3Si+ cata-
lyst derived from iBuSKA and acid 1 exhibits lower activity
in the polymerization of MMA, but exceptional activity
(TOF=2.4ꢀ105 hÀ1), efficiency, and control for the polymer-
ization of nBA. This system also enabled the synthesis of
well-defined methacrylate-b-acrylate (MMA-b-nBA) block
copolymers.
2
above), that significantly hampers the catalyst reactivity.
When the anion employed becomes too coordinating, as in
the case of the chiral disulfonimide anion in acid 3 (see
above), the MMA polymerization shuts down completely
(run 5, Table 4). To overcome this problem, we reasoned
that utilizing a monomer that is basic enough to outcompete
the anion for coordination to the silylium catalyst could lead
to an active polymerization system by using 3. Indeed, the
polymerization of N,N-dimethyl acrylamide (DMAA, a
stronger base than MMA) by MeSKA and acid 3 in CH2Cl2 is
highly active, reaching a high TOF of 12000 hÀ1 and produc-
ing poly(N,N-dimethyl acrylamide) (PDMAA) with
a
narrow MWD of 1.13 (run 7, Table 4). However, the result-
ing polymer is essentially atactic, with a triad distribution of
38.5% mm and 61.5% rr/mr as determined by 13C NMR
spectroscopy, and the polymerization is still about 6 times
slower than the same polymerization in the presence of acid
1 paired with the weakly coordinating anion [BACTHNUTRGNEUNG
(C6F5)4]À
(run 7 vs. run 6, Table 4). The same polymerization by
MeSKA and acid 3, but carried out in toluene, is much
slower, requiring 22 h to reach completion, but the resulting
PDMAA has a similar tacticity of 37.2% mm and 62.8% rr/
mr.
Results of kinetic studies showed that the polymerization
by MeSKA and acid 1 is zero order in monomer concentra-
tion and first order in the concentrations of initiator [SKA]
and catalyst [Me3Si+]. These results are consistent with the
propagation mechanism (“catalysis” cycle) as depicted in
Conclusion
À
In summary, this contribution presents a full account of our
investigation into the ambient-temperature, high-speed
living polymerization of polar vinyl monomers (e.g., metha-
crylates, acrylates, and acrylamides), catalyzed by silylium
ions derived from the spontaneous protonation of the SKA
initiator with the strong Brønsted acids 1, 2, and 3—which
Scheme 4, in which the C C bond formation step through
Michael addition of the polymeric SKA to the silylium-acti-
vated MMA is the rate-determining step and the release of
the silylium catalyst from the ester group of the growing
polymer chain to the incoming MMA is relatively fast,
thereby highlighting the high efficiency of the silylium catal-
ysis in this polymerization. Another interesting feature of
this polymerization is its unique catalyst “self-repair” fea-
ture, as outlined in the catalyst healing cycle of Scheme 4.
All PMMAs produced by this catalyst system at ambient
temperature are syndio-rich polymers with a syndiotacticity
of ꢂ72% rr, although the syndiotacticity of PBMA is higher
(78% rr). The syndiotacticity of PMMA can be enhanced to
75% rr and 87% rr, at the expense of substantially reduced
activity, by lowering Tp to 08C and À408C, respectively. Uti-
lization of the chiral disulfonimide anion, in attempts to reg-
ulate the MMA polymerization stereochemistry, was ham-
pered by its inactivity. Efforts to achieve a stereoselective
polymerization catalyzed by chiral silylium catalysts are cur-
rently underway.
incorporate the noncoordinating [BACTHNUTRGNEUNG
(C6F5)4]À anion or the
chiral disulfonimide anion. The living character of the cur-
rent polymerization system has been demonstrated by sever-
al key lines of evidence, including the observed linear
growth of the chain length as a function of monomer con-
version at a given [M]/[I] ratio, near-precise polymer Mn
(controlled by the [M]/[I] ratio) with narrow MWD, absence
of an induction period and chain-termination reactions (as
revealed by polymerization kinetics), readily achievable
chain extension, and the successful synthesis of the well-de-
fined block copolymers. We have also elucidated the funda-
mental steps of activation, initiation, propagation, and cata-
lyst “self-repair” involved in this living polymerization
system. Key findings of this study are summarized as fol-
lows: Among the three strong Brønsted acids investigated in
this study, the oxonium acid 1 is the most effective activator,
in that it delivers the most active silylium catalyst, R3Si+,
Experimental Section
thanks to 1) its least coordinating counteranion [BACTHNUTRGNEUNG
(C6F5)4]À
and 2) the weak basic coproduct (ether and ester) relative to
monomer. Hence, the MMA polymerization by MeSKA and
1 at ambient temperature with a low catalyst loading of
0.05 mol% (based on monomer) exhibits a high activity
Materials, reagents, and methods: All syntheses and manipulations of air-
and moisture-sensitive materials were carried out in flamed-dried
Schlenk-type glassware on a dual-manifold Schlenk line, on a high-
vacuum line (from 10À5–10À7 Torr), or in an argon-filled glove box (typi-
10470
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 10462 – 10473

Living Polymerization by Self-Healing Silylium Catalysts
FULL PAPER
cally < 1.0 ppm oxygen and moisture). NMR-scale reactions (typically in
a 0.02 mmol scale) were conducted in Teflon-valve-sealed J. Young-type
NMR tubes. High performance liquid chromatography (HPLC) grade or-
ganic solvents were first sparged extensively with nitrogen during filling
20 L solvent reservoirs and then dried by passage through activated alu-
mina (for diethyl ether (Et2O), tetrahydrofuran (THF), and CH2Cl2) fol-
lowed by passage through Q-5-supported copper catalyst (for toluene
and hexane) stainless steel columns. [D6]Benzene and [D8]toluene were
dried over sodium/potassium alloy and distilled in vacuum or filtered,
whereas C6D5Br, CD2Cl2, and CDCl3 were dried over activated Davison
4 ꢄ molecular sieves. NMR spectra were recorded on either a Varian
Inova 300 (FT 300 MHz, 1H; 75 MHz, 13C; 282 MHz, 19F) or a Varian
Inova 400 spectrometer. Chemical shifts for 1H and 13C NMR spectra
were referenced to internal solvent resonances and are reported as parts
per million (ppm) relative to SiMe4, whereas 19F NMR spectra were ref-
erenced to external CFCl3. Si NMR chemical shifts were obtained from
ACTHNGUTERNUNG CATHUNGTREN(NGNU OMe)=O or Et2O).
pair [Me3Si···L]+[B(C6F5)4]À (4, L=Me2CHC
1H NMR (CD2Cl2, 300 MHz, 238C): d=3.86 (s, 3H; OMe), 3.79 (q, J=
7.2 Hz, 8H; OCH2CH3), 2.67 (sept, J=7.2 Hz, 1H; CHMe2), 1.29 (t, J=
7.2 Hz, 12H; OCH2CH3), 1.22 (d, J=7.2 Hz, 6H; CHMe2), 0.68 ppm (s,
9H; SiMe3); 19F NMR (CD2Cl2, 282 MHz, 238C): d=À131.4 (brs, 8F; o-
F), À161.9 (t, J=20.3 Hz, 4F; p-F), À165.8 ppm (t, J=18.0 Hz, 8F; m-F).
Subsequently, the reaction mixture was dried under vacuum to remove
all volatiles. The white, solid residue was washed with hexane and then
dissolved in CD2Cl2 after being dried in vacuo for 2 h. The NMR spectra
showed that neither the ester nor the ether was completely removed and
that the product was a mixture of [Me3Si···L]+[B(C6F5)4]À in a 3:1 ratio
ACHTUNGTRENNUNG
of 4a (L=Me2CHC
AHCTUNGTRENNUNG
400 MHz, 238C): d=4.33 (s, 3H; OMe), 2.95 (brs, 1H; CHMe2), 1.40 (d,
1
J=7.2 Hz, 6H; CHMe2), 0.67 ppm (s, 9H; SiMe3); 4b: H NMR (CD2Cl2,
400 MHz, 238C): d=4.54 (q, J=7.2 Hz, 4H; OCH2CH3), 1.57 (t, J=
7.2 Hz, 6H; OCH2CH3), 0.67 ppm (s, 9H; SiMe3); 19F NMR (CD2Cl2,
282 MHz, 238C): d=À131.4 (brs, 8F; o-F), À162.0 (t, J=20.3 Hz, 4F; p-
F), À165.9 ppm (t, J=18.3 Hz, 8F; m-F).
1H/29Si HSQC or HSQCAD experiments on
a MR400-VNMRs400
1
(400 MHz, H; 79.4 MHz, 29Si) spectrometer.
It is interesting to note that the colorless crystals, obtained from layering
a solution of 1 equiv of MeSKA in hexane onto a solution of 1 in CH2Cl2
at À308C inside a freezer in the glove box over a one week, were those
of pure 4b, as shown by NMR spectroscopy. 29Si NMR (CD2Cl2,
79.4 MHz, HSQCAD): d=68.48 ppm (SiMe3). Unfortunately, X-ray dif-
fraction analysis of these single crystals did not yield a satisfactory struc-
ture due to twining.
Methyl methacrylate (MMA, 99%) and n-butyl methacrylate (BMA,
99%) were purchased from Aldrich Chemical, N,N-dimethyl acrylamide
(DMAA, 98%) from TCI America, and n-butyl acrylate (nBA) from
Acros. These monomers were first degassed and dried over CaH2 over-
night, followed by vacuum distillation. Final purification of MMA in-
volved titration with neat triACTHUNRTGNEUNG(n-octyl)aluminum (Strem Chemicals) to a
yellow end point,[35] followed by vacuum distillation. The purified mono-
mers (checked by NMR spectroscopy) were stored in brown bottles
inside a glove box freezer at À308C.
Reaction of MeSKA with acid 2: This reaction was carried out in the same
manner as the reaction of MeSKA with 1, forming cleanly [Me3Si···N-
(Me2)Ph]+[B(C6F5)4]À (5) and Me2CHC(OMe)=O. 1H NMR (CD2Cl2,
AHCTUNGTREUNNGN ACHTUNGTRENNUNG ACTHNUGTRENNGUN
Butylated hydroxytoluene (BHT-H, 2,6-di-tert-butyl-4-methylphenol,
300 MHz, 238C): d=7.59 (brs, 3H; Ph), 7.37–7.35 (m, 2H; Ph), 3.70 (s,
3H; OMe), 3.33 (brs, 6H; NMe2), 2.60 (sept, J=7.2 Hz, 1H; CHMe2),
1.16 (d, J=7.2 Hz, 6H; CHMe2), 0.52 ppm (s, 9H; SiMe3); 19F NMR
(CD2Cl2, 282 MHz, 238C): d=À131.5 (brs, 8F; o-F), À161.9 (t, J=
20.4 Hz, 4F; p-F), À165.8 ppm (t, J=18.2 Hz, 8F; m-F).
The colorless single crystals of 5 were obtained from layering solution of
1 equiv of MeSKA in hexane onto a solution of 2 in CH2Cl2 inside a freez-
er in the glove box at À308C over one week. The 1H NMR spectrum of
the isolated crystals was the same as that of 5 generated in situ from the
NMR-scale reaction (minus those peaks due to the ester, which was re-
moved). 1H NMR (CD2Cl2, 300 MHz, 238C): d=7.63–7.61 (m, 3H; Ph),
7.36–7.34 (m, 2H; Ph), 3.33 (s, 6H; NMe2), 0.52 ppm (s, 9H; SiMe3);
13C NMR (CD2Cl2, 75 MHz, 238C): d=131.2, 130.8, 120.9 (Ph), 48.17
(NMe2), À2.47 ppm (SiMe3); 19F NMR (CD2Cl2, 282 MHz, 238C): d=
À131.8 (brs, 8F; o-F), À161.9 (t, J=20.6 Hz, 4F; p-F), À165.9 ppm (t,
J=18.0 Hz, 8F; m-F); 29Si NMR (CD2Cl2, 79.4 MHz, HSQCAD): d=
55.65 ppm (SiMe3). The molecular structure of 5 was characterized by
single-crystal X-ray diffraction analysis.
99%) was purchased from Aldrich Chemical and recrystallized from
hexane prior to use. Tris(pentafluorophenyl)borane, B
kis(pentafluorophenyl)borate, [Ph3C]+[B(C6F5)4]À, and dimethylanilinium
tetrakis(pentafluorophenyl)borate, [HN
(Me2)Ph]+[B(C6F5)4]À (2), were
(C6F5)3 was further
ACHTNUGTRNE(NNUG C6F5)3, trityl tetra-
AHCTUNGTRENNUNG
A
ACHTUNGTRENNUNG
obtained as a research gift from Boulder Scientific. BACHTUNGTRENNUNG
purified by recrystallization from hexane at À308C, whereas [Ph3C]+[B-
ACHTUNGTRENNUNG
(C6F5)4]À and 2 were used as received. Trimethylsilyl dimethylketene
methyl acetal (MeSKA, 98%), diisopropylamine (ꢁ 99%), methyl isobu-
tyrate (99%), chlorotriisobutylsilane (99%), and bromopentafluoroben-
zene (99%) were purchased from Aldrich Chemical and dried over
CaH2, followed by vacuum distillation. Boron trichloride (1.0m in
hexane) and HCl (2.0m in diethyl ether) were purchased from Aldrich
Chemical and used as received. Literature procedures were employed for
the preparation of triisobutylsilyl dimethylketene methyl acetal Me2C=C-
A
ACHTUNGTRENNUNG(iBu)3 ( AHCTUTGNENRNUG ACHTUGTNRENNUG
iBuSKA),[14] [Li(Et2O)]+[B(C6F5)4]À,[36] and (R)-3,3’-bis-
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
Reaction of MeSKA with acid 3: This reaction was carried out in the same
U
ACHTUNGTRENNUNG
manner as the reaction of MeSKA with 1, forming Me3Si-[3-H] and the
(10 mL) was added to a reactor that contains a solution of HCl in diethyl
ether (1.0m, 20 mL, 20 mmol) at À308C. The mixture was stirred at this
temperature for 4 H; after which it was filtered and the filtrate was con-
centrated to ꢂ10 mL, which were then added to another solution of HCl
in diethyl ether (1.0m, 20 mL, 20 mmol) at À308C. This procedure was
repeated three times. The solvent of the filtrate was removed in vacuo,
and the solid residue was dissolved in CH2Cl2 (10 mL). The filtrate was
layered with hexane (20 mL) and diethyl ether (5 mL) and stored in a
freezer inside the glove box at À308C overnight. Filtration and drying
under vacuum of the resulting colorless crystalline solid yielded 1.30 g
free (uncoordinated) Me2CHCACHTNUTGRNEUNG
(OMe)=O. 1H NMR (CD2Cl2, 300 MHz,
238C): d=8.10–8.05 (m, 4H), 8.00–7.95 (m, 6H), 7.80–7.71 (m, 2H),
7.53–7.44 (m, 2H), 7.29–7.16 (m, 2H), 3.63 (s, 3H; OMe), 2.53 (sept, J=
6.9 Hz, 1H; CHMe2), 1.13 (d, J=6.9 Hz, 6H; CHMe2), 0.19, À0.20 ppm
(s, 9H; SiMe3); 19F NMR (CD2Cl2, 282 MHz, 238C): d=À62.85,
À62.91 ppm (CF3); 29Si NMR (CD2Cl2, HSQC): d=39.86, 24.51 ppm
(SiMe3).
Study of catalyst “self-repair”: This study was carried out in three stages.
First, to a solution of catalyst 4 (0.017 mol) in CD2Cl2 (0.5 mL) an equi-
molar amount of was added equimolar at ambient temperature. The mix-
ture was allowed to react for ꢂ15 min before NMR spectra were record-
ed, which showed the clean formation of Me3SiOH, regeneration of acid
1
(39.8%) of the pure title compound. H NMR (CD2Cl2, 300 MHz, 238C):
d=16.5 (s, 1H; H), 4.08 (q, J=7.2 Hz, 8H; OCH2CH3), 1.43 ppm (t, J=
7.2 Hz, 12H; OCH2CH3); 19F NMR (CD2Cl2, 282 MHz, 238C): d=À131.9
(brs, 8F; o-F), À162.4 (t, J=20.4 Hz, 4F; p-F), À166.4 ppm (t, J=
17.8 Hz, 8F; m-F).
Reaction of MeSKA with acid 1: In an argon-filled glove box, a Teflon-
valve-sealed J. Young-type NMR tube was charged with MeSKA (3.5 mg,
0.02 mmol) and CD2Cl2 (0.3 mL). A solution of 1 (16.6 mg, 0.02 mmol) in
CD2Cl2 (0.3 mL) was added to this tube by a pipette at ambient tempera-
ture. The mixture was allowed to react for ꢂ15 min before NMR spectra
were recorded, which showed the clean formation of base-separated ion
1, and Me2CHCACTHNUTRGNEUNG
(OMe)=O. 1H NMR (CD2Cl2, 300 MHz, 238C): d=14.8
(s, 1H; H), 3.97 (q, J=6.9 Hz, 8H; OCH2CH3), 3.70 (s, 3H; OMe), 2.58
(sept, J=6.9 Hz, 1H; CHMe2), 1.38 (t, J=6.9 Hz, 12H; OCH2CH3), 1.16
(d, J=6.9 Hz, 6H; CHMe2), 0.42 (s, 1H; SiOH), 0.07 ppm (brs, Me3SiO);
19F NMR (CD2Cl2, 282 MHz, 238C): d=À131.4 (brs, 8F; o-F), À162.0 (t,
J=20.3 Hz, 4F; p-F), À165.9 ppm (brs, 8F; m-F). Second, another equiv-
alent of MeSKA (0.017 mmol) was added to this reaction mixture. The re-
sulting NMR spectra clearly showed the regeneration of catalyst 4.
Chem. Eur. J. 2010, 16, 10462 – 10473
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10471

E. Y.-X. Chen et al.
1H NMR (CD2Cl2, 300 MHz, 238C): d=3.78 (s, 6H; OMe), 3.69 (q, J=
6.9 Hz, 8H; OCH2CH3), 2.62 (sept, J=6.9 Hz, 2H; CHMe2), 1.25 (t, J=
6.9 Hz, 12H; OCH2CH3), 1.19 (d, J=6.9 Hz, 12H; CHMe2), 0.67 (s, 9H;
SiMe3), 0.06 ppm (s, Me3SiO); 19F NMR (CD2Cl2, 282 MHz, 238C): d=
À131.4 (brs, 8F; o-F), À161.9 (t, J=20.3 Hz, 4F; p-F), À165.8 ppm (t,
J=18.2 Hz, 8F; m-F). Third, this regenerated catalyst 4 was used for the
polymerization of MMA (9.35 mmol) with MeSKA (0.047 mmol) as initia-
tor in CH2Cl2 (10 mL). A quantitative conversion of MMA was achieved
in 30 min, showing similar activity to that of the originally generated cat-
alyst.
(5 mL). The quenched mixture was precipitated into methanol (100 mL),
stirred for 1h, filtered, washed with methanol, and dried in a vacuum
oven at 508C overnight to a constant weight. The quenched aliquot was
analyzed by 1H NMR spectroscopy to show 100% monomer conversion.
The isolated and dried polymer was analyzed by gel permeation chroma-
tography (GPC), showing Mn of 2.21ꢀ104 gmolÀ1 and PDI of 1.08, rela-
tive to PMMA standards. The dried polymer was also analyzed by
1H NMR for tacticity measurement. 1H NMR (CDCl3, 300 MHz, 238C):
d=3.60 (s, OMe), 2.05 (d, J=13.6 Hz, CH2), 1.96–1.90 (m, CH2), 1.82 (s,
CH2), 1.44 (d, J=13.6 Hz, CH2), 1.21 (s, Me, [mm]=1.9%), 1.02 (s, Me,
[mr]=25.9%), 0.86 ppm (s, Me, [rr]=72.2%).
X-ray crystallographic analysis of [Me3Si···NACTHUNRTGENNUG AHCTUNGTRENNGUN
(Me2)Ph]+[B(C6F5)4]À (5):
Polymerization kinetics: Kinetic experiments were carried out in a stirred
glass reactor at ambient temperature (ꢂ258C) inside the glove box by
using stock solutions of the reagents and the procedures described previ-
ously.[3,38]
Single crystals suitable for X-ray diffraction were quickly covered with a
layer of Paratone-N oil (Exxon, dried and degassed at 1208C under
10À6 Torr for 24 h) after the mother liquor was decanted and then mount-
ed on a thin glass fiber and transferred into the cold nitrogen stream of a
Bruker SMART CCD diffractometer. The structure was solved by the
direct method and refined by using the Bruker SHELXTL program li-
brary by full-matrix least-squares on F2 for all reflections.[37] All non-hy-
drogen atoms were located by difference Fourier synthesis and refined
with anisotropic displacement parameters, whereas all hydrogen atoms
were included geometrically with Uiso tied to the Uiso of the parent atoms
and refined isotropically. CCDC-773064 contains the supplementary crys-
tallographic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
Polymer characterizations: Polymer Mn and PDI values were measured
by GPC analyses carried out at 408C and a flow rate of 1.0 mLminÀ1
,
with CHCl3 as the eluent on a Waters University 1500 GPC instrument
equipped with one PLgel 5 mm guard and three PLgel 5 mm mixed-C col-
umns (Polymer Laboratories; linear range of molecular weight=200—
2000000). The instrument was calibrated with ten PMMA standards, and
chromatograms were processed with Waters Empower software (version
2002). Glass-transition temperatures (Tg) of the polymers were measured
by differential scanning calorimetry (DSC) on a DSC 2920, TA Instru-
ment. Samples, weighted (typically in a 7–10 mg range) and sealed in her-
metic aluminum pans by using a DSC press, were first heated to 2008C at
108CminÀ1, equilibrated at this temperature for 4 min, then cooled to
08C (for methacrylate polymers) or À808C (for nBA-containing poly-
mers) at 108CminÀ1, held at this temperature for 4 min, and finally re-
heated to 2008C at 108CminÀ1. All Tg values were obtained from the
second scan after removing the thermal history. 1H NMR spectra for the
analysis of PMMA and PBMA microstructures were recorded in CDCl3
and analyzed according to the literature,[39] whereas 13C NMR spectra of
PDMAA were recorded in D2O at 808C and analyzed by using literature
procedures.[40]
X-ray crystal structural data for silylium complex 5: C35H20BF20NSi, Mr =
873.42, T=123(2) K, l=0.71073 ꢄ, crystal dimensions 0.509ꢀ0.365ꢀ
0.132 mm3, monoclinic, P21/c, a=13.6097 (3) ꢄ, b=14.1254(3) ꢄ, c=
18.6174(4) ꢄ, b=99.2180(10)8, V=3532.83(13) ꢄ3, Z=4, 1calcd
=
1.642 MgÀ3, q range for data collection=1.82–30.708, 43253 reflections
collected, 10953 unique (Rint =0.0289), goodness-of-fit on F2 =1.022, final
R1 =0.0381 and wR2 =0.1029 with I>2s(I), and residual electron density
extremes=0.396 and À0.279 eꢄÀ3
.
General polymerization procedures: Polymerizations were performed
either in 30 mL oven-dried glass reactors inside the glove box for ambi-
ent-temperature (ꢂ258C) runs or in 25 mL oven- and flame-dried
Schlenk flasks interfaced to the dual-manifold Schlenk line for runs at
nonambient temperatures. Two different polymerization procedures were
employed for comparative studies. In method A (preactivation) proce-
dures, SKA and an acid (1, 2, or 3) were premixed in CH2Cl2 or toluene
and stirred for 5 min (to cleanly generate the silylium catalyst in situ),
followed by addition of monomer to start the polymerization. In meth-
od B (in-reactor activation), SKA and monomer were premixed and the
polymerization was started by addition of the acid. In both procedures,
after the measured time interval, the polymerization was quenched by
addition of 5% HCl-acidified methanol (5 mL). For MMA and BMA
homo- and copolymerizations, the quenched mixture was precipitated
into methanol (100 mL), stirred for 1h, filtered, washed with methanol,
and dried in a vacuum oven at 508C overnight to a constant weight. For
nBA polymerizations, the polymer was isolated by evaporation of the sol-
vent and dried in a vacuum oven at room temperature to a constant
weight. For DMAA polymerizations, the polymer was precipitated into
diethyl ether (100 mL), stirred for 1h, filtered, washed with diethyl ether,
and dried in a vacuum oven at 508C overnight to a constant weight.
Acknowledgements
The work carried out at CSU was supported by the National Science
Foundation (NSF-0848845). We thank Boulder Scientific Co. for the re-
search gifts of
(C6F5)3, [Ph3C]+[B(C6F5)4]À, and [HN(Me2)Ph]+[B-
BACHTUNGTREUNNGN ACHTUNTRGENNUNG ACTHNUGTRENNGUN
AHCTUNGTRENNUNG
[1] E. Y.-X. Chen, Chem. Rev. 2009, 109, 5157–5214.
[2] H. L. Hsieh, R. P. Quirk, Anionic Polymerization. Principles and
Practical Applications, Marcel Dekker, New York, 1996.
[3] A. Rodriguez-Delgado, E. Y.-X. Chen, J. Am. Chem. Soc. 2005, 127,
961–974.
[4] Selected comprehensive reviews: a) D. Baskaran, Prog. Polym. Sci.
2003, 28, 521–5811; b) P. Vlꢆek, L. Lochmann, Prog. Polym. Sci.
1999, 24, 793–973; c) C. Zune, R. Jꢇrꢈme, Prog. Polym. Sci. 1999,
24, 631–664; d) K. Hatada, T. Kitayama, K. Ute, Prog. Polym. Sci.
1988, 13, 189–276; e) M. Van Beylen, S. Bywater, G. Smets, M.
Szwarc, D. J. Worsfold, Adv. Polym. Sci. 1988, 86, 87–143.
[5] O. W. Webster, W. R. Hertler, D. Y. Sogah, W. B. Farnham, T. V. Ra-
janBabu, J. Am. Chem. Soc. 1983, 105, 5706–5708.
[6] O. W. Webster, Adv. Polym. Sci. 2004, 167, 1–34.
[7] a) A. H. E. Mꢂller, G. Litvinenko, D. Yan, Macromolecules 1996, 29,
2346–2353; b) R. P. Quirk, J.-S. Kim, J. Phys. Org. Chem. 1995, 8,
242–248; c) R. P. Quirk, G. P. Bidinger, Polym. Bull. 1989, 22, 63–
70.
A specific polymerization example: In an argon-filled glove box, a 30 mL
oven-dried glass reactor was charged with 1 (3.9 mg, 4.7 mmol) in CH2Cl2
(5 mL). In a 20 mL glass vial, MeSKA (9.0 mg, 51.5 mmol) and MMA
(1.00 mL, 9.35 mmol) were mixed in CH2Cl2 (5 mL). The mixture in the
vial was rapidly poured into the reactor to start the polymerization at
ambient temperature. The [MMA]/ACHTUNGRNENUG[
MeSKA]/[1] ratio in this selected ex-
ample was 400:2.2:0.2, and the actual [M]/[I] ratio was 200 (as the acid
consumes 0.2 equiv of MeSKA to form the equal amount of the Me3Si+
catalyst). The reaction mixture was stirred at ambient temperature for
40 min, after which an aliquot (0.2 mL) was withdrawn from the reaction
mixture by using a syringe and quickly quenched into a 1 mL vial that
contains undried “wet” CDCl3 (0.6 mL) stabilized with of BHT-H
(250 ppm). The reactor was immediately taken out of the glove box, and
the reaction was quenched by addition of 5% HCl-acidified methanol
[8] T. Mukaiyama, Angew. Chem. 2004, 116, 5708–5733; Angew. Chem.
Int. Ed. 2004, 43, 5590–5614.
[9] D. Y. Sogah, W. R. Hertler, O. W. Webster, G. M. Cohen, Macromo-
lecules 1987, 20, 1473–1488.
10472
www.chemeurj.org
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 10462 – 10473

Living Polymerization by Self-Healing Silylium Catalysts
FULL PAPER
[10] a) J. Raynaud, Y. Gnanou, D. Taton, Macromolecules 2009, 42,
5996–6005; b) J. Raynaud, A. Ciolino, A. Baceiredo, M. Destarac,
F. Bonnette, T. Kato, Y. Gnanou, D. Taton, Angew. Chem. 2008, 120,
5470–5473; Angew. Chem. Int. Ed. 2008, 47, 5390–5393; c) M. D.
Scholten, J. L. Hedrick, R. M. Waymouth, Macromolecules 2008, 41,
7399–7404.
[11] W. R. Hertler, D. Y. Sogah, O. W. Webster, B. M. Trost, Macromole-
cules 1984, 17, 1415–1417.
[12] Y. Zhang, E. Y.-X. Chen, Macromolecules 2008, 41, 36–42.
[13] J. B. Lambert, S. Zhang, C. L. Stern, J. C. Huffman, Science 1993,
260, 1917–1918.
1188–1190; f) Y. Zhang, K. Huynh, I. Manners, C. A. Reed, Chem.
Commun. 2008, 494–496; g) S. Duttwyler, Q.-Q. Do, A. Linden,
K. K. Baldridge, J. S. Siegel, Angew. Chem. 2008, 120, 1743; Angew.
Chem. Int. Ed. 2008, 47, 1719–1722; h) K.-C. Kim, C. A. Reed,
D. W. Elliott, L. J. Mueller, F. Tham, L. Lin, J. B. Lambert, Science
2002, 297, 825–827; i) C. A. Reed, Acc. Chem. Res. 1998, 31, 325–
332.
[29] J. Lacour, V. Hebbe-Viton, Angew. Chem. 2009, 121, 4427; Angew.
Chem. Int. Ed. 2009, 48, 4363–4366.
[30] a) S. Mukherjee, B. List, J. Am. Chem. Soc. 2007, 129, 11336–11337;
b) G. L. Hamilton, E. J. Kang, M. Mba, D. Toste, Science 2007, 317,
496–499.
[31] K. Hensen, T. Zengerly, P. Pickel, G. Klebe, Angew. Chem. 1983, 95,
739–739; Angew. Chem. Int. Ed. Engl. 1983, 22, 725–726.
[32] Z. Xie, D. J. Liston, T. Jelinek, V. Mitro, R. Bau, C. A. Reed, J.
Chem. Soc. Chem. Commun. 1993, 384–386.
[14] Y. Zhang, E. Y.-X. Chen, Macromolecules 2008, 41, 6353–6360.
[15] a) H. Yamamoto, Tetrahedron 2007, 63, 8377–8412; b) M. B. Boxer,
H. Yamamoto, J. Am. Chem. Soc. 2006, 128, 48–49; c) K. Ishihara,
Y. Hiraiwa, H. Yamamoto, Synlett 2001, 1851–1854.
[16] T. Akiyama, Chem. Rev. 2007, 107, 5744–5758.
[17] P. Jutzi, C. Mꢂller, A. Stammler, H. Stammler, Organometallics
2000, 19, 1442–1444.
[18] a) E. B. Tjaden, D. C. Swenson, R. F. Jordan, Organometallics 1995,
14, 371–386; b) H. W. Turner, G. G. Hlatky, R. R. Eckman, U.S.
Patent 5,198,401, 1993.
[19] P. Garcꢁa-Garcꢁa, F. Lay, P. Garcꢁa-Garcꢁa, C. Rabalakos, B. List,
Angew. Chem. 2009, 121, 4427–4430; Angew. Chem. Int. Ed. 2009,
48, 4363–4366.
[20] R. Kakuchi, K. Chiba, K. Fuchise, R. Sakai, T. Satoh, T. Kakuchi,
Macromolecules 2009, 42, 8747–8750.
[21] B. Mathieu, L. Ghosez, Tetrahedron Lett. 1997, 38, 5497–5500.
[22] K. Ute, H. Ohnuma, T. Kitayama, Polym. J. 2000, 32, 1060–1062.
[23] R. Zhuang, A. H. X. Mꢂller, Macromolecules 1995, 28, 8035–8042;
R. Zhuang, A. H. X. Mꢂller, Macromolecules 1995, 28, 8043–8050.
[24] K. Ute, H. Ohnuma, I. Shimizu, T. Kitayama, Polym. J. 2006, 38,
999–1003.
[25] E. Y.-X. Chen, T. J. Marks, Chem. Rev. 2000, 100, 1391–1434.
[26] Y. Zhang, Y. Ning, L. Caporaso, L. Cavallo, E. Y.-X. Chen, J. Am.
Chem. Soc. 2010, 132, 2695–2709.
[33] C. A. Reed, Z. Xie, R. Bau, A. Benesi, Science 1993, 262, 402–404.
[34] T. Kꢂppers, E. Bernhardt, R. Eujen, H. Willner, C. W. Lehmann,
Angew. Chem. 2007, 119, 6462–6465; Angew. Chem. Int. Ed. 2007,
46, 6346–6349.
[35] R. D. Allen, T. E. Long, J. E. McGrath, Polym. Bull. 1986, 15, 127–
134.
[36] A. G. Massey, A. J. Park, J. Organomet. Chem. 1964, 2, 245–250.
[37] SHELXTL, Version 6.12, Bruker Analytical X-ray Solutions, Madi-
son, WI, 2001.
[38] A. Rodriguez-Delgado, E. Y.-X. Chen, Macromolecules 2005, 38,
2587–2594.
[39] a) A. D. Bolig, E. Y.-X. Chen, J. Am. Chem. Soc. 2004, 126, 4897–
4906; b) A. Rodriguez-Delgado, W. R. Mariott, E. Y.-X. Chen, Mac-
romolecules 2004, 37, 3092–3100; c) F. A. Bovey, P. A. Mirau, NMR
of Polymers, Academic Press, San Diego, 1996; d) R. Chffljꢈ, K.
Hatada, R. Kitamaru, T. Kitayama, H. Sato, Y. Tanaka, Polym. J.
1987, 19, 413–424.
[40] a) W. R. Mariott, E. Y.-X. Chen, Macromolecules 2005, 38, 6822–
6832; b) W. R. Mariott, E. Y.-X. Chen, Macromolecules 2004, 37,
4741–4743; c) M. Kobayashi, S. Okuyama, T. Ishizone, S. Nakaha-
ma, Macromolecules 1999, 32, 6466–6477; d) X. Xie, T. E. Hogen-
Esch, Macromolecules 1996, 29, 1746–1752; e) A. Bulai, M. L.
Jimeno, A.-A. de Queiroz, A. Gallardo, J. S. Roman, Macromole-
cules 1996, 29, 3240–3246.
[27] I. Krossing, I. Raabe, Angew. Chem. 2004, 116, 2116–2142; Angew.
Chem. Int. Ed. 2004, 43, 2066–2090.
[28] Selected examples: a) C. A. Reed, Acc. Chem. Res. 2010, 43, 121–
128; b) H. F. T. Klare, M. Oestreich, Dalton Trans. 2010, DOI:
10.1039/c003097; c) G. M. Miyake, Y. Zhang, E. Y.-X. Chen, Macro-
molecules 2010, 43, 4902–4908; d) H. F. T. Klare, K. Bergander, M.
Oestreich, Angew. Chem. 2009, 121, 9241; Angew. Chem. Int. Ed.
2009, 48, 9077–9079; e) C. Douvris, O. V. Ozerov, Science 2008, 321,
Received: April 15, 2010
Published online: July 21, 2010
Chem. Eur. J. 2010, 16, 10462 – 10473
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
10473