Controlled depolymerization of poly(5-ethyl-5-phenyl-1,3-dioxan-2-one): Selective liberation of cyclic carbonate monomer from polymer chain end

Article abstract of DOI:10.1021/ma035049e

Poly(EP6CC) was studied using several techniques. The compound was found to undergo controlled depolymerization with releasing the monomer EP6CC selectively from the chain end in the presence of the HCl-OEt₂ complex, regardless of whether it is in a solution state or it is grafted on an insoluble cross-linked polymer.

Full text of DOI:10.1021/ma035049e

Macromolecules 2004, 37, 251-253
251
Sch em e 1
Con tr olled Dep olym er iza tion of
P oly(5-eth yl-5-p h en yl-1,3-d ioxa n -2-on e):
Selective Liber a tion of Cyclic Ca r bon a te
Mon om er fr om P olym er Ch a in En d
Atsu sh i Su d o,† Kyok o Ka ta ok a ,‡ F u m io Sa n d a ,§ a n d
Ta k esh i En d o*,
Scheme 2 illustrates the present polymerization-
depolymerization system. For the polymerization, potas-
sium tert-butoxide was employed as an anionic initiator,
as previously reported.8 The polymerization was carried
out in tetrahydrofuran (THF) at -40 °C. By monitoring
the polymerization reaction by 1H NMR analysis, it was
revealed that the conversion of the monomer was
determined to be 32% at 1 h after the initiation, and
then the conversion reached 77% at 6 h, which would
be the equilibrium point under the conditions. The
number-average molecular weight (Mn) and weight-
average molecular weight (Mw) of the resulting polymer,
which were estimated by size exclusion chromatography
(SEC) of the crude mixture before precipitation, were
9000 and 13 300, respectively. These values did not
change after precipitation. The formed polymer 1a was
isolated in 62% yield as methanol-insoluble material.
The presence of hydroxyl group at the polymer terminal
Molecular Engineering Institute, Kinki University,
11-6 Kayanomori, Iizuka, Fukuoka 820-8555, J apan;
Chemical Resources Laboratory, Tokyo Institute of
Technology, Nagatsuta-cho 4259, Midori-ku,
Yokohama 226-8503, J apan; Department of Polymer
Chemistry, Graduate School of Engineering,
Kyoto University, Yoshida-honmachi, Sakyo-ku,
Kyoto 606-8501, J apan; and Department of Polymer Science
and Engineering, Faculty of Engineering, Yamagata
University, J onan, Yonezawa, Yamagata 992-8510, J apan
Received J uly 22, 2003
Revised Manuscript Received November 30, 2003
To date, there have been various studies on depolym-
erization of polymers. A small difference in enthalpy
between monomer and polymer, which is often found
in the case of polymerization of disubstituted olefins,
leads to the equilibrium nature of the polymerization.
On the basis of this equilibrium nature, the polymer can
be depolymerized into its original monomer under
appropriate conditions. In most cases, depolymerization
proceeds with random scission of the polymer main
chain to give a mixture of monomer and residual
oligomers, which prevent the selective regeneration of
the monomer. On the other hand, in some cases,
depolymerization initiates selectively from the chain end
of the polymer to release the monomer.1-4 Such a so-
called “unzipping” depolymerization is potentially as a
highly effective method for chemical recycle of monomer.
Furthermore, if the degree of unzipping can be well-
controlled, it can be a powerful chisel for nanoscale
control of figures and patterns of polymer materials.
We have reported equilibrium polymerization behav-
iors of cyclic monomers.5-9 For example, six-membered
cyclic carbonates (6CC) undergo anionic polymerization,
and on the basis of equilibrium nature, the formed
polymer (poly6CC) undergoes anionic depolymerization
(Scheme 1).8 This equilibrium nature depends on the
bulkiness of the substituents at the 5-position of the
monomer: The more bulky the substituents are, the
more easily depolymerization undergoes. In this anionic
depolymerization system, a significant amount of oli-
gomers are formed besides the monomer in the early
stage of the depolymerization, suggesting that the main
reaction involved in the system is a random attack of
anionic species to the carbonate group in the main
chain. During our efforts to control the depolymerization
of poly(6CC), we found that a cationic system is capable
of achieving a controlled unzipping depolymerization,
which is the main subject of this Communication.10
was confirmed by its IR absorption at 3100 cm-1
.
Sch em e 2
Thus, we examined the acid-promoted depolymeriza-
tion of the obtained polycarbonate 1a using HCl-OEt2
as a promoter.11 The time dependence of Mn and Mw/Mn
of the polymer and that of yield of EP6CC for the
depolymerization are shown in Table 1. As can be seen
from these data, the yield of EP6CC increased with the
elapse of time, and accordingly, Mn of the resulting
polymer decreased by depolymerization. The time de-
pendence of yield of EP6CC was not an exactly linear
one but showed the tendency that the rate of depolym-
erization slightly declined upon elongation of time,
suggesting the equilibrium nature of the polymerization.
The previously reported monomer concentration at
equilibrium in the anionic polymerization of EP6CC is
0.4 M (at 20 °C, the initial monomer concentration )
0.6 M, initiator ) t-BuOK, in THF), which means that
† Kinki University.
‡ Tokyo Institute of Technology.
§ Kyoto University.
Yamagata University.
* To whom correspondence should be addressed: Tel & Fax
+81-238-26-3090; e-mail tendo@yz.yamagata-u.ac.jp.
2
approximately /3 of the repeating unit in the polymer
can be depolymerized into EP6CC.8 Thus, the monomer
10.1021/ma035049e CCC: $27.50 © 2004 American Chemical Society
Published on Web 12/23/2003

252 Communications to the Editor
Macromolecules, Vol. 37, No. 2, 2004
Ta ble 1. Acid -P r om oted Dep olym er iza tion of
P oly(EP 6CC)
Sch em e 3
depolymerization of 1a a
depolymerization of 3a b
yield of
yield of
recovered
c
depolym
time/h
Mn
recovered
(Mw/Mn)c
monomer/%d
monomer/%e
0
3
6
24
48
72
9000 (1.27)
nd f
9000 (1.30)
8300 (1.31)
7800 (1.31)
7100 (1.31)
(0)
nd f
4
9
16
20
(0)
3
4
9
nd f
14
a
Conditions: [polymer-OH]0 ) 0.25 M; [HCl-OEt2]0 ) 0.125
b
M. Conditions: [polymer-OH]0 ) 0.17 M; [HCl-OEt2] ) 0.51
M. c Estimated by SEC analysis (eluent: THF, calibrated with
polystyrene standards) of the crude mixture. Determined by 1H
d
NMR. e Determined by GC. f Not determined.
yield in the present depolymerization seems to be still
much lower than the expected one from the potential
depolymerizability of poly(EP6CC). This low monomer
yield in the present depolymerization is presumably due
to the following two reasons: (1) The acid-promoted
depolymerization is much slower than that promoted by
basic reagent, and thus the depolymerization did not
reach its equilibrium yet. (2) Solvent should have strong
effects on the equilibrium nature: Since highly electron
donative THF was used for the previously reported
anionic depolymerization while electron deficient dichlo-
romethane was used in the present acidic depolymeri-
zation, the equilibrium monomer concentration should
be different.
It is noteworthy that no formation of the correspond-
ing oligomers was found by SEC analysis of the obtained
mixture, suggesting that the mechanism of the depo-
lymerization does not involve a random scission of the
main chain but is a selective intramolecular attack of
the terminal hydroxyl group to the carbonyl group of
the main chain, which is activated by protonation. The
polymer 1b12 having an acetylated terminal did not
undergo depolymerization under the same conditions,
indicating that the terminal hydroxyl group is indis-
pensable for initiation of the depolymerization. During
the depolymerization of 1a , no significant change in
Mw/Mn was observed, while ideal unzipping depolym-
erization of well-defined polymer obtained by living
polymerization should have an intrinsic tendency of
decrease in polydispersity. The change of polydispersity
during the depolymerization is considered to signifi-
cantly have reflected the change during the polymer
growth in the polymerization: The polymer 1a , formed
by initiation with t-BuOK, should have a large polydis-
persity in the early stage of the polymerization because
the initiation reaction is slower than the propagation.
Thus, it can be considered that the polydispersity should
increase with retracing the propagation to compensate
the intrinsic tendency of decrease in polydispersity for
an ideal well-controlled depolymerization process, find-
ing no significant change in it.
carbonate, i.e., poly(EP6CC) grafted on a cross-linked
polystyrene (Scheme 3). Graft polymerization of EP6CC
was carried out using a solid-supported anionic initiator
2, which was prepared from cross-linked hydroxy-
methylated polystyrene beads by treatment with potas-
sium hexamethyldisilazide (KN(SiMe3)2) in THF.13 The
polymerization came to equilibrium with the corre-
sponding depolymerization at 3 h after the initiation,
and the monomer conversion at this point was deter-
mined to be 76% by GC analysis of the mixture. The
resulting beads 3a were filtered and washed by THF to
remove soluble parts. From the weight increase of the
obtained beads, conversion of EP6CC to the grafted
poly(EP6CC) was calculated to be 35%. IR analysis of
the beads showed strong absorptions at 1753 and at
3120 cm-1, which were attributable to acyclic carbonate
group in the main chain of the grafted poly(EP6CC) and
hydroxyl group at the chain end, respectively. The
soluble parts consisted of the monomer (24%) and oligo-
and poly(EP6CC) (total 41%), which would be formed
by backbiting reaction of the propagating polymer and
by intermolecular nucleophilic attack of the propagating
polymer to the carbonate group of the other polymer
chain. The beads 3a were exposed to the conditions for
acetylation to obtain the beads 3b,12 on which the
immobilized poly(EP6CC) had a masked hydroxyl ter-
minal by esterification. Complete masking of the ter-
minal hydroxyl group was confirmed by disappearance
of its IR absorption at 3120 cm-1
.
Depolymerization of the grafted poly(EP6CC) was
examined under similar conditions to those adopted for
the linear polymer. Figure 1 illustrates the apparatus
used, which was designed so that GC analysis of the
solution phase can be easily performed with the aid of
temporary removal of the beads by filtration under a
nitrogen atmosphere. In this depolymerization system
for the grafted poly(EP6CC), the amount of HCl-OEt2
complex was increased to 3 equiv to the terminal
hydroxyl group because 0.5 equiv of the complex did not
sufficiently promote the depolymerization, presumably
due to some steric effects arisen by cross-linked poly-
styrene. As shown in Table 1, the monomer was gradu-
ally released from the beads by the depolymerization.
SEC analysis of the solution phase revealed that elution
Addition of molecular sieve (MS4A), which traps
water, is essential to achieve well-controlled depolym-
erization: After the depolymerization for 72 h in the
absence of MS4A, Mn of the resulting polymer decreased
to 7500 with increase of Mw/Mn to 1.53, suggesting that
random scission of the main chain would undergo by
hydrolysis of the protonated carbonate group.
The present polymerization-depolymerization system
was successfully applicable to a solid-supported poly-

Macromolecules, Vol. 37, No. 2, 2004
Communications to the Editor 253
Kirshenbaum, G. S., Eds.; Elsevier: New York, 1986; pp
105-114.
(4) Vogl, O.; Miller, H. C.; Sharkey, W. H. Macromolecules 1972,
5, 658.
(5) Endo, T.; Sanda, F. React. Funct. Polym. 1997, 33, 241.
(6) Sanda, F.; Hitomi, M.; Endo, T. J . Polym. Sci., Part A:
Polym. Chem. 2001, 39, 3159.
(7) Nishida, H.; Yamashita, M.; Endo, T.; Tokiwa, Y. Macro-
molecules 2000, 33, 6982.
(8) Matsuo, J .; Aoki, K.; Sanda, F.; Endo, T. Macromolecules
1998, 31, 4432.
(9) For the recent work by the other authors, see: Kanoh, S.;
Nishimura, T.; Senda, H.; Ogawa, H.; Motoi, M.; Tanaka,
T.; Kano, K. Macromolecules 1999, 32, 2438.
(10) Enzymatic degradation of polycarbonate is also of recent
great interest. For example, see: Tsutsumi, C.; Nakagawa,
K.; Shirahama, H.; Yasuda, H. Macromol. Biosci. 2002, 2,
223.
(11) HCl-OEt2 has been found to be an effective promoter in
polymerizations of cyclic lactones and cyclic carbonates
using alcohol as an initiator; see: Shibasaki, Y.; Sanada,
H.; Yokoi, M.; Sanda, F.; Endo, T. Macromolecules 2000, 33,
4316.
(12) 1b was obtained by acetylation of hydroxyl group at the
terminal of 1a : 1a (200 mg) was treated with acetic
anhydride (2 mL) and pyridine (4 mL) at room temperature
for 24 h. The mixture was poured into methanol (30 mL) to
obtain precipitates, which were redissolved in THF and were
reprecipitated with methanol to obtain 1b (191 mg, 95%),
of which SEC-estimated Mn and Mw/Mn were identical with
those of 1a . For acetylation of the solid-supported polymer
3a, dichloromethane and 4-(dimethylamino)pyridine (DMAP)
were additionally used as a solvent and a catalyst, respec-
tively.
(13) The procedure: To a THF dispersion (5 mL) of hydroxy-
methylated cross-linked polystyrene beads (purchased from
Novabiochem Co., Ltd.; 0.87 mmol of -OH per 1 g; 115 mg,
[-OH] ) 0.10 mmol), a THF solution of KN(TMS)2 (0.5 M;
0.20 mL, 0.10 mmol) was added at 0 °C under N2. The
mixture was shaken at 0 °C for 30 min. The beads were
filtered and washed with THF (5 mL) twice under N2 and
swelled in THF (3 mL). To this THF dispersion of the beads,
a THF solution (2 mL) of EP6CC (206 mg, 1.0 mmol) was
added. The mixture was shaken at -40 °C for 3 h. The
reaction was quenched by addition of saturated aqueous
solution of ammonium chloride/methanol (1/9 (v/v), 0.1 mL).
The beads were filtered, washed by THF and methanol, and
dried under vacuum to give 3a (187 mg).
F igu r e 1. Apparatus for depolymerization of the solid-
supported poly(EP6CC): (a) for reaction; (b) for sampling for
GC analysis.
of polymeric and oligomeric products was negligible,
indicating that selective depolymerization successfully
proceeded without random scission of the main chain
of the beads-supported poly(EP6CC), similarly to the
depolymerization of the free poly(EP6CC). In addition,
the terminal hydroxyl group of the grafted poly(EP6CC)
was found to be indispensable for successful depolym-
erization: When 3b, of which chain end was capped by
acetylation, was exposed to the depolymerization condi-
tions, only a negligible amount of EP6CC (<2%) was
liberated from the beads even after 72 h.
In summary, poly(EP6CC) undergoes controlled de-
polymerization with releasing the monomer EP6CC
selectively from the chain end in the presence of the
HCl-OEt2 complex, regardless of whether it is in a
solution state or it is grafted on an insoluble cross-linked
polymer. More detailed investigations on equilibrium
behaviors of the polymerizations of EP6CC and other
cyclic carbonates on various material surfaces are
currently underway.
Refer en ces a n d Notes
(1) Lehman, F. A.; Braner, G. M. Anal. Chem. 1961, 33, 673.
(2) Barker, S. J .; Bruce, M. B. Polyacetal; Elsevier: New York,
1970.
(3) Persak, K. J .; Fleming, R. A. In High Performance Poly-
mers: Their Origin and Development; Seymour, R. B.,
MA035049E