Angewandte
Chemie
different from the results observed in the living cationic
chain end, i.e., an acetal proton originating from methanol
polymerization using HX/MX , in which the protonic acid
quenching the polymerization. As indicated by the filled
circles in Figure 3b, the acetal-chain-end content from all w-
ends [ÀOCH /(ÀSC(S)Z) + (ÀOCH )] nearly agrees with the
n
determines the molecular weight through the formation of
[
5–8,29]
stable dormant CÀX species.
Additionally, a low con-
3
3
centration of TfOH ([TfOH] = 0.05 mm or ca. 8 ppm in the
TfOH content ([TfOH] /([3] + [TfOH] )). This result indi-
0
0
0
0
À4
À3
solution; [TfOH] /[M] = 1 ꢀ 10
;
[TfOH] /[2] = 5 ꢀ 10 )
cates that the growing carbocationic species exists at the same
concentration as that of the initial TfOH and is terminated by
methanol, whereas the dormant thioester terminal is intact
upon quenching, and that the cationic polymerization pro-
ceeds not through a Lewis acid-catalyzed mechanism, which
0
0
0
0
efficiently induces the metal-free “living” or controlled
[
3,29,30]
cationic polymerization.
To further clarify the polymerization mechanism, the
effects of TfOH were investigated, because it was already
reported that the CÀS bonds in thiocarbonylthio compounds
has already been reported for the thiocarbonylthio com-
[27]
[27,28]
can also be activated by Lewis acids. The TfOH concen-
tration was thus changed between 0.05 and 5.0 mm, while the
concentrations of the monomer and RAFT agent were held
constant ([M] /[3] = 500/10 mm). As the [TfOH] increased,
pounds,
but through a RAFT mechanism. Thus, the
growing carbocationic species undergoes propagation without
significant irreversible side reactions and reversibly inter-
changes with the dormant thioester terminal, resulting in
a control of the molecular weight as in radical RAFT
0
0
0
the polymerization became faster (e.g., [TfOH] , time, con-
0
[31]
version = 0.05 mm, 90 min, 96%; 5.0 mm, 2 s, > 99%) (Fig-
ure S5), and Mn gradually decreased, whereas the MWDs
remained narrow (M /M ꢀ 1.1) (Figure 3a). Figure 3b shows
polymerizations.
The cationic RAFT mechanism was further confirmed by
H NMR spectroscopy of the model reaction, in which an
1
w
n
equimolar mixture of two RAFT agents (RÀSC(S)Z) with
different R and Z moieties (3: R = CH CH(OiBu), Z = NEt ,
3
2
8
: R = CH CH(OEt), Z = OEt) were treated with a small
3
amount of TfOH ([3] /[8] /[TfOH] = 30/30/0.15 mm) in the
0
0
0
absence of monomer at À408C (Figure S6). Even after 24 h,
no exchange reaction occurred between the two RAFTagents
(Figure S6C), indicating that TfOH does not work as a Lewis
acid catalyst for the thiocarbonylthio compounds. In contrast,
when the mixture was treated with ZnCl , a typical Lewis acid
2
catalyst for the living cationic polymerization, the exchange
reaction occurred to result in the exchanged products at least
[32]
within 30 min (Figure S6D). These results clearly demon-
strate that the cationic polymerization with RÀSC(S)Z/TfOH
proceeds through the RAFT mechanism.
The choice of protonic acid is also important for inducing
the controlled cationic polymerization. A series of protonic
acids with different acidity or different nucleophilicity of their
anions were used in conjunction with 2 for IBVE (Figure S7).
A stronger acid, triflylimide (Tf NH), induced a faster
2
polymerization to give polymers with controlled molecular
weights. However, when using a weaker acid, such as
TfOH·Py, CH SO H, or CF CO H, no polymers were
3
3
3
2
obtained, most likely due to the formation of stable adducts
of IBVE cation and a more nucleophilic counteranion.
The versatility of the RAFT cationic polymerization was
then examined for other vinyl ethers (EVE: ethyl vinyl ether,
CEVE: 2-chloroethyl vinyl ether) and p-methoxystyrene
(pMOS). Although the conditions were optimized depending
on the reactivity of the monomers, all of these monomers
were successfully polymerized in a controlled fashion to give
polymers with controlled molecular weights (Figure S8). In
addition, p-hydroxystyrene (pHS) or p-vinylphenol, which
usually cannot be polymerized with a Lewis acid due to the
phenolic functional groups, were also successfully polymer-
ized without protection of the hydroxy group, to give
Figure 3. Effects of [TfOH] on a) SEC curves of the polymers obtained
0
at conversion >95% and b) M observed by SEC and terminal acetal
n
1
groups by H NMR spectroscopy in the polymerization of IBVE at
[
1
(
M] /[3] /[TfOH] =500/10/0.05–5.0 mm in n-hexane/CH Cl /Et O (80/
0
0
0
2
2
2
1
0/10 vol%) at À408C; c) typical H NMR spectrum of the polymers
M (SEC)=2000, M (NMR)=1800, M /M =1.07) obtained at [M] /
n n w n 0
[
3] /[TfOH] =500/10/0.10 mm (conversion=34%).
0 0
the ratio of the molecular weight measured using SEC
M (obs)] to the calculated one [M (calcd)] assuming the
[
n
n
[
33,34]
formation of one polymer chain per molecule of 3. As
polymers with controlled molecular weights (Figure S9).
[
TfOH] increased, the ratio decreased to below 1, indicating
Another notable feature of the cationic RAFT polymer-
ization is its good compatibility with the radical RAFT
polymerization, which would enable a more direct synthesis
of block copolymers between cationically and radically
0
1
that more chains were formed. Furthermore, in the H NMR
spectrum of the obtained polymer (Figure 3c), a new peak (h)
appeared at 4.6 ppm, which can be attributed to another w-
Angew. Chem. Int. Ed. 2014, 53, 1 – 6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3
These are not the final page numbers!
Uchiyama, Mineto
