Polymerization of epoxide with hydroxylamides as thermally latent initiators

Article abstract of DOI:10.1002/pola.28139

A new class of thermally latent initiators for the ring-opening polymerization of epoxides has been developed. The latent initiators developed herein were the hydroxylamides 1a, 1b, and 1c, which were synthesized from phthalide, 3-isochromanone, and cis-cyclohexahydrophthalide, respectively, by their ring-opening reactions with pyrrolidine. These hydroxylamides were designed so that their hydroxyl groups could attack the amide moiety intramolecularly upon heating, leading to ring closure and formation of the corresponding lactones while releasing pyrrolidine, the initiator for the anionic ring-opening polymerization of an epoxide. The temperatures at which this thermal dissociation occurred were strongly dependent on the hydroxylamide molecular structure. When using the hydroxylamides as thermally latent initiators, the polymerizations of bisphenol-A diglycidyl ether were investigated at various temperatures. This investigation clarified that the threshold temperature, that is, the temperature at which polymerization was initiated, increased in the order of 1a, 1b, and then 1c.

Full text of DOI:10.1002/pola.28139

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
Polymerization of Epoxide With Hydroxylamides as
Thermally Latent Initiators
Yanmei Wang, Mika Kimura, Atsushi Sudo, Takeshi Endo
Molecular Engineering Institute, Kinki University, 11-6 Kayanomori, Iizuka, Fukuoka 820-8555, Japan
Correspondence to: T. Endo (E-mail: tendo@moleng.fuk.kindai.ac.jp)
Received 29 February 2016; accepted 9 April 2016; published online 00 Month 2016
DOI: 10.1002/pola.28139
ABSTRACT: A new class of thermally latent initiators for the
ring-opening polymerization of epoxides has been developed.
The latent initiators developed herein were the hydroxylamides
1a, 1b, and 1c, which were synthesized from phthalide, 3-
isochromanone, and cis-cyclohexahydrophthalide, respectively,
by their ring-opening reactions with pyrrolidine. These hydrox-
ylamides were designed so that their hydroxyl groups could
attack the amide moiety intramolecularly upon heating, leading
to ring closure and formation of the corresponding lactones
while releasing pyrrolidine, the initiator for the anionic ring-
opening polymerization of an epoxide. The temperatures at
which this thermal dissociation occurred were strongly
dependent on the hydroxylamide molecular structure. When
using the hydroxylamides as thermally latent initiators, the pol-
ymerizations of bisphenol-A diglycidyl ether were investigated
at various temperatures. This investigation clarified that the
threshold temperature, that is, the temperature at which poly-
merization was initiated, increased in the order of 1a, 1b, and
C
V
then 1c.
2016 Wiley Periodicals, Inc. J. Polym. Sci., Part A:
Polym. Chem. 2016, 00, 000–000
KEYWORDS: anionic polymerization; crosslinking; initiators;
ring-opening polymerization
also an important research target because of their potentially
low toxicity and high miscibility in epoxy resins. Such non-
salt type latent initiators involve cyanoacetamide,21 amini-
mides,22 phosphonic amide esters,23 doped polyaniline,24
and sulfonates.25
INTRODUCTION Epoxy resins have been widely used in coat-
ings, paintings, and adhesive applications, because their cur-
ing reactions give materials with good mechanical strength
and chemical resistance. The curing reactions of epoxy resins
are performed by adding appropriately selected hardeners or
initiators to the epoxide mixture. When the hardeners or ini-
tiators are highly active, they are used in a so-called “two-
component system,” where they are mixed with the epoxy
resins immediately prior to the curing reactions. In contrast,
“one-component systems” where the epoxy resins and initia-
tors are premixed, and thus ready for use, are practically
more appreciated. The initiators used therein must be
“latent,” which are carefully designed so that they can be
premixed and stored stably; however, some external stimula-
tion, such as irradiation and heating, is required to trigger
their programmed dissociation into the corresponding frag-
ments that can initiate the ring-opening polymerization of
the epoxides. So far, various photo-latent1–7 and thermally
latent initiators for the ring-opening polymerization of epox-
ides have been developed, the latter of which involve diverse
onium salts such as ammonium,8,9 anilinium,10 pyridinium,11
pyradinium,12,13 imidazolium,14,15 phosphonium,13,16,17 and
sulfonium,18,19 and iodonium20 salts. In addition, the devel-
opment of nonsalt type thermally latent initiators has been
Previously, we have reported that a hydroxylamide could
also serve as a thermally latent anionic initiator for the poly-
merization of epoxides.26 As shown in Scheme 1, the species
that initiated the polymerization was piperidine, released
from the hydroxylamide by its thermally induced cyclization.
The key point in the molecular design of this hydroxylamide-
type latent initiator is the linking of the hydroxyl group and
amide group by a tether containing a rigid o-phenylene moi-
ety, so that the decrease in entropy upon reaction can be
minimized. This design also allows for the formation of a
thermodynamically stable five-membered lactone, of which
formation is favorable from the viewpoint of enthalpy. Based
on a similar concept, we have also developed phosphamide-
type latent initiators.23
Our interests currently lie in expanding the scope of the
abovementioned molecular design of hydroxylamides for
developing a variety of initiators that can initiate the ring-
opening polymerization of epoxides at various temperatures
Additional Supporting Information may be found in the online version of this article.
C
V
2016 Wiley Periodicals, Inc.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 00, 000–000
1

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
colorless liquid (1.37 g, 6.67 mmol, 67% yield). 1H NMR
(300 MHz, in CDCl3, d in ppm) 7.44–7.32 (m, 4H), 4.53 (d,
J 5 6.1 Hz, 2H), 3.98 (br t, J 5 6.1 Hz, 1H), 3.68 (t, J 5 7.0 Hz,
2H), 3.34 (t, J 5 6.5 Hz, 2H), 2.04–1.63 (4H, m); 13C NMR
(75.4 MHz, in CDCl3, d in ppm) 170.4, 139.2, 136.7, 130.0,
129.93, 127.7, 126.8, 64.15, 49.66, 46.08, 26.39, 24.72.
Synthesis of Hydroxylamide 1b
To a solution of 3-isochromanone (1.48 g, 10.0 mmol) in
methanol (10 mL), a solution of pyrrolidine (0.85 g, 12
mmol) in methanol (10 mL) was added at room tempera-
ture. The resulting solution was stirred at room temperature
for 24 h, and then concentrated under reduced pressure.
The resulting residue was dissolved in ether (50 mL) and
washed three times with 0.1 M hydrochloric acid (50 mL).
The ether solution was dried over MgSO4 and then concen-
trated under reduced pressure. The resulting residue was
fractionated using silica gel column chromatography with a
1:10 mixture of hexane and ethyl acetate as the eluent, to
obtain 1b as a colorless liquid (1.10 g, 5.02 mmol, 50%
SCHEME 1 Release of an amine-type initiator by the thermally
induced cyclization of hydroxylamide.
on demand. Herein we demonstrate the polymerization of
bisphenol A diglycidyl ether (BADGE) using a series of
hydroxylamides 1 that can be derived from pyrrolidine, a
simple 5-membered cyclic amine, as thermally latent initia-
tors (Scheme 2). Because there are a wide variety of 5-
membered cyclic amines with a broad range of functional
groups such as proline, a naturally occurring amino acid, by
proving the usefulness of 1 as latent initiators, we aimed to
broaden the range of structural diversity of hydroxyamide-
type initiators.
1
yield). H NMR (300 MHz, in CDCl3, d in ppm) 7.43–7.40 (m,
1H), 7.29–7.26 (m, 2H), 7.17–7.15 (m, 1H), 4.78 (t, J 5 6.1
Hz, 1H), 4.61 (d, J 5 6.1 Hz, 2H), 3.78 (s, 2H), 3.67 (t, J 5 6.8
Hz, 2H), 3.48 (t, J 5 6.8 Hz, 2H), 2.07–1.99 (m, 2H), 1.94–
1.87 (m, 2H); 13C NMR (75.4 MHz, in CDCl3, d in ppm)
170.2, 140.4, 139.8, 130.7, 130.4, 128.1, 127.6, 63.64, 47.32,
46.25, 38.86, 26.09, 24.28.
EXPERIMENTAL
Materials
Pyrrolidine, BADGE, glycidyl phenyl ether (GPE), and metha-
nol were purchased from Wako Pure Chemical Industries
and were used as received. Phthalide, 3-isochromanone and
cis-1,2-cyclohexanedicarboxylic anhydride were purchased
from Tokyo Chemical Industry and used as received. The
procedure for synthesizing cis-cyclohexahydrophthalide from
cis-1,2-cyclohexanedicarboxylic anhydride is described in the
Supporting Information.
Synthesis of Hydroxylamide 1c
To a solution of cis-cyclohexahydrophthalide (1.40 g, 10.0
mmmol) in methanol (10 mL), a solution of pyrrolidine
(0.85 g, 12 mmol) in methanol (10 mL) was added at room
temperature. The resulting solution was stirred at room tem-
perature for 24 h, and then concentrated under reduced
pressure. The resulting residue was dissolved in ether
(50 mL) and washed three times with 0.1 M hydrochloric
acid (50 mL). The ether solution was dried over MgSO4 and
then concentrated under reduced pressure. The resulting
residue was fractionated using silica gel column chromatog-
raphy with a 100:1 mixture of chloroform and methanol as
the eluent, to obtain 1c as a white solid (1.03 g, 4.87 mmol,
Instruments
1H NMR and 13C NMR spectra were recorded on a JEOL
AL300 NMR spectrometer with tetramethylsilane as an inter-
nal standard. Viscosity was measured with a Brookfield
viscometer (Model Cap 20001). Differential scanning calo-
rimetry (DSC) measurements were performed on a SEIKO
DSC6200. Thermogravimetry (TG) measurements were per-
formed on a SEIKO EXSTAR6000.
1
49% yield). Mp 55–57 8C; H NMR (300 MHz, in CDCl3, d in
ppm) 4.20–4.00 (m, 1H), 3.69–3.85 (m, 1H), 3.38–3.64 (m,
5H), 2.77–2.66 (m, 1H), 2.11–1.74 (m, 8H), 1.65–1.23 (m,
5H); 13C NMR (75.4 MHz, in CDCl3, d in ppm) 175.4, 63.25,
46.65, 45.88, 43.52, 38.83, 29.49, 26.11, 25.04, 24.77, 24.13,
22.02.
Synthesis of Hydroxylamide 1a
To a solution of phthalide (1.34 g, 10.0 mmol) in methanol
(10 mL), a solution of pyrrolidine (0.85 g, 12 mmol) in
methanol (10 mL) was added at room temperature. The
resulting solution was stirred at room temperature for 24 h,
and then concentrated under reduced pressure. The resulting
residue was dissolved in ether (50 mL) and washed three
times with 0.1 M hydrochloric acid (50 mL). The ether solu-
tion was dried over MgSO4 and then concentrated under
reduced pressure. The resulting residue was fractionated
using silica gel column chromatography with a 1:1 mixture
of hexane and acetone as the eluent, to obtain 1a as a
Thermal Dissociation of Hydroxylamides
Each of the hydroxylamides, 1a–1c, was heated at various
temperatures for 2 h in a glass bulb (bulb A) using a
SCHEME 2 Synthesis of hydroxylamides 1.
2
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
TABLE 1 Chemical Structure and Reaction Outcomes of Hydroxylamides, 1, from the Corresponding Lactones
Entry
1
Lactone
Hydroxylamide
Yield/%
67
Properties
Liquid
2
3
50
49
Liquid
Solid, mp 5 55–57 8C
RESULTS AND DISCUSSION
Kugelrohr distillation apparatus. Bulb A was connected to a
second bulb (bulb B) that was cooled with ice, into which
the pyrrolidine formed by the thermal dissociation of the
hydroxylamide was condensed. The residue in bulb A was
analyzed by 1H NMR spectroscopy to determine the conver-
sion of the hydroxylamide into the corresponding lactone.
Synthesis of Hydroxylamides
The hydroxylamides, 1, were readily obtained by mixing the
corresponding lactones and pyrrolidine in methanol at room
Polymerization of BADGE using Hydroxylamides
Typical procedures are given below for the polymerization
using 1a.
1. BADGE (1.36 g, 4.00 mmol) and hydroxylamide 1a
(164 mg, 0.799 mmol) were mixed and degassed at room
temperature to obtain a homogeneous mixture.
2. Three small portions of the reaction mixture (ꢀ20 mg)
were taken and put on a Brookfield viscometer and heated
at 50, 100, and 150 8C to monitor the increase in viscosity
of the mixture.
3. Several small portions of the mixture (ꢀ30 mg) were
taken and heated using a DSC instrument in both dynamic
mode (heating rate 5 10 8C/min) and isothermal mode at
several temperatures to obtain the corresponding heat
evolution profiles. These experiments were performed
under a flow of nitrogen.
4. The rest of the mixture was heated at 180 8C in an oven.
The corresponding cured materials were analyzed by TGA
and DSC in a dynamic mode (heating rate 5 10 8C/min)
under a flow of nitrogen.
MODEL POLYMERIZATIONS OF GPE AND HYDROXYLAMIDES
Typical Procedure
A mixture of GPE (150 mg, 1.00 mmol) and hydroxylamide
1a (20.5 mg, 0.10 mmol) was heated at 180 8C for 2 h. After
cooling, the mixture was dissolved in THF and analyzed by
thin layer chromatography (TLC) to check for the presence
of phthalide formed by the thermal dissociation of 1a.
FIGURE 1 1H NMR spectroscopic monitoring of the thermal
dissociation of 1a.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 00, 000–000
3

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
SCHEME 3 Polymerization of BADGE using hydroxylamide 1
as a latent initiator.
dependence of the thermal dissociation of 1a on tempera-
ture, where below 40 8C, the dissociation was negligible, at
60 8C, formation of a small amount of phthalide was
observed, and above 160 8C, more than 80% of 1a was
transformed into the phthalide.
In a similar way, using the Kugelrohr distillation apparatus
and 1H NMR spectroscopy, the thermal dissociation behav-
iors of hydroxylamides 1b and 1c were investigated. For the
calculation of the conversion of 1b into 3-isochromanone,
the corresponding signals at 5.25 and 4.56 ppm, respectively,
attributable to the methylene group located between the
aromatic group and the oxygen atom, were used. For the cal-
culation of the conversion of 1c into cis-cyclohexahydroph-
thalide, the corresponding signals at 4.21 and 4.11 ppm,
respectively, attributable to the methylene group located
between the cyclohexyl group and the oxygen atom, were
used. Figure 2 shows the dependences of the degree of dis-
sociation on temperature, where for the dissociation of 1b, a
higher temperature than that required for the dissociation of
1a was necessary. Hydroxylamide 1c was less labile than 1b,
however. It was stable below 100 8C, and its conversion into
the corresponding lactone at 180 8C was only 60%.
FIGURE 2 Temperature-dependences of the thermal dissocia-
tions of the hydroxylamides, 1.
temperature (Scheme 2). The molecular structures of the lac-
tones used, those of the resulting hydroxylamides, and the
corresponding yields are shown in Table 1. All of the reac-
tion products were isolated by column chromatography and
their structures were confirmed by NMR spectroscopy. The
1H and 13C NMR spectra of 1a are shown in Figures 1 and
S1 (in Supporting Information), respectively. The NMR spec-
tra of 1b and those of 1c are shown in Figures S2 and S3
(in Supporting Information), respectively.
Thermal Dissociation of Hydroxylamides
To investigate the thermal dissociation behaviors of the
hydroxylamides, they were heated at different temperatures
using a Kugelrohr distillation apparatus (Supporting Informa-
tion Fig. S4). For example, hydroxylamide 1a was placed in a
glass bulb (bulb A) that could be heated in the oven of the
apparatus to induce the thermal dissociation of 1a. Bulb A
was connected to another bulb (bulb B) outside of the oven,
which was cooled to condense and collect the pyrrolidine
formed by the thermal dissociation of 1a that was distilled
out from the bulb A. After heating at a certain temperature
(40–180 8C) for 2 h, the remaining residue in bulb A was
analyzed by 1H NMR spectroscopy. Figure 1 shows three
In all cases, when the hydroxylamide underwent thermal dis-
sociation to give the corresponding lactone, pyrrolidine was
distilled and condensed into bulb B. The yields of pyrrolidine
formed were much lower than those expected from the con-
versions of the hydroxylamides into the corresponding lac-
tones, presumably due to the difficulties in the complete
condensation of the formed pyrrolidine in bulb B as a result
of its high volatility.
1
spectra: (a) the H NMR spectrum of 1a, (b) that of the resi-
due after heating at 120 8C for 2 h, and (c) that of the resi-
due after heating at 160 8C for 2 h.
In these spectra, the signal attributable to the methylene
protons of 1a and the singlet signal attributable to the meth-
ylene protons of phthalide appeared at 5.32 and 4.51 ppm,
respectively, clarifying that 1a underwent the desired ther-
mal dissociation into phthalide and pyrrolidine, where the
dissociation was promoted by raising the reaction tempera-
ture. Comparison of the intensities of these singlet signals
attributable to the methylene protons allowed for calculation
of the conversion of 1a into phthalide by thermal dissocia-
tion. The conversions calculated in this way are
plotted against temperature in Figure 2, which clarified the
FIGURE 3 DSC-profiles of the hydroxylamides
1 measured
using a dynamic scan mode.
4
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
the case of 1c, this temperature was even higher. These
results implied that temperature required for the initiation
of the polymerization could be tuned by changing the struc-
ture of the hydroxylamide initiator.
Figure 4 shows the heat evolution profiles obtained by the
DSC analysis in an isothermal mode. The same reaction mix-
tures as those used for the abovementioned DSC analysis in
a dynamic scan mode were heated at 100, 150, and 180 8C.
When the reaction mixture containing 1a was heated at 100
8C, the corresponding heat evolution started after 10 min
and its peak was observed at 40 min, suggesting that 1a dis-
sociated efficiently and the resulting pyrrolidine formed initi-
ated the polymerization reaction. In contrast, no heat
evolutions were observed by heating the reaction mixtures
containing 1b and 1c at 100 8C, suggesting that their dissoci-
ations were not efficient enough to release a sufficient
amount of pyrrolidine. By elevating the temperature to 150
FIGURE 4 DSC-profiles of the hydroxylamides
1 measured
using an isothermal scan mode.
Polymerization of Epoxide with Hydroxylamides as
Thermally Latent Initiators
The performances of the three hydroxylamides as thermally
latent initiators for the polymerizations of epoxides were
evaluated by the following three methods employing BADGE
as a monomer (Scheme 3): (1) a reaction mixture comprised
of BADGE and hydroxylamide (10 mol % to epoxide moiety)
was heated in a DSC instrument using a dynamic scan mode
to investigate the temperature-dependence of reaction heat
evolution; (2) the reaction mixture was heated in the DSC
instrument using an isothermal mode to investigate the
time-dependence of reaction heat evolution; and (3) the
increase in viscosity caused by the polymerization at several
temperatures was monitored by a viscometer.
Figure 3 shows the heat evolution profiles obtained from
DSC analysis using a dynamic scan mode (25–300 8C at a
heating rate of 5 8C/min under nitrogen). As the profiles
show, the use of 1a as a latent initiator resulted in the evolu-
tion of heat at the lowest temperature. By employing 1b, the
heat evolution peak shifted to a higher temperature, and in
FIGURE 5 Time-viscosity relationships for the polymerizations
at various temperatures of BADGE using 1 as a thermal-latent
initiator.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 00, 000–000
5

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 2 Thermal Properties of the Network Polymers Obtained by the Polymerization of BADGE with Hydroxylamidea
Polymerization
Time/min
Tg
(8C)b
Td5
(8C)c
Td10
(8C)c
Entry
Hydroxylamide
1
2
3
1a
1b
1c
40
45
90
87
86
61
350
345
324
365
362
354
a
Conditions: [epoxy moieties of BADGE]0:[hydroxylamide]0 5 100:10; at
bDetermined by DSC (heating rate 10 8C/min).
cDetermined by TGA (heating rate 10 8C/min).
180 8C.
8C, a heat evolution peak appeared in the analysis of the
reaction mixture that contained 1b, and quite a broad and
low heat evolution was observed in the analysis of the mix-
ture containing 1c. Even when the temperature was elevated
to 180 8C, it was difficult to observe a clear heat evolution
from the reaction mixture containing 1c, while the polymer-
izations of the other reactions were accelerated significantly
at this temperature.
led to the significant deterioration of heat resistance, pre-
sumably due to incompletion polymerization.
Possible Mechanisms
The reaction pathways for epoxide polymerization that could
be involved in the present system using hydroxylamides as
latent initiators are shown in Scheme 4. The first process
that occurs is the thermally induced dissociation of the
hydroxylamide into the corresponding lactone and pyrroli-
dine.26 The basic chemistry of this dissociation is the alco-
holysis of the amide, which is generally disfavored due to
the low nucleophilicity of alcohols, low electrophilicity of
amides, and poor leaving ability of amines.
Figure 5 shows the time-viscosity relationships for the reac-
tion mixtures at 50, 100, and 150 8C. These relationships
implied that (1) 1a was the most labile hydroxylamide that
induced polymerization at 50 8C, (2) 1b was the second
most labile hydroxylamide that induced polymerization at
100 8C, and (3) by elevating the reaction temperature to 150
8C, the most stable hydroxylamide 1c induced the
polymerization.
However, these unfavorable factors were overcome in the
present system by the following two elements incorporated
into the molecular design of the hydroxylamide initiator: the
first being that the location of the hydroxyl group is at a
suitable position for its intramolecular nucleophilic attack on
We attempted to investigate polymerization behaviors of
BADGE using pristine pyrrolidine as an initiator by DSC anal-
ysis and viscosity measurement. However, the reaction was
not controllable: As was expected from the intrinsic reactiv-
ity of pyrrolidine with epoxide, in some cases, exotherm took
place immediately just after mixing, and the viscosity started
to increase, leading to the unreproducible DSC and viscosity
measurements.
Analyses of the Network Polymers Obtained by
Polymerization of BADGE using Hydroxylamides as
Latent Initiators
The reaction mixtures comprised of BADGE and hydroxyla-
mide 1a (10 mol % to epoxide moiety) were heated at 180
8C for 40 min. The other reaction mixtures containing 1b
and 1c were heated at 180 8C for 45 and 90 min, respec-
tively. IR analyses of the resulting network polymers
revealed that the absorption at 916 cm21, attributable to the
epoxy group of BADGE, disappeared, suggesting that the cur-
ing reactions were complete. The network polymers pre-
pared were analyzed by DSC and
TG techniques to investigate their thermal properties, and
the corresponding glass transition temperature (Tg) and tem-
peratures for 5 and 10% weight loss (Td5 and Td10) are
listed in Table 2. The use of hydroxylamides 1a and 1b as
latent initiators resulted in the formation of network poly-
mers with similar Tg and Td values. In contrast, the use of 1c
SCHEME 4 Possible reaction pathways involved in the present
polymerization system using hydroxylamide
initiator.
1 as a latent
6
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 00, 000–000

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
REFERENCES AND NOTES
the amide group, leading to the formation of a thermody-
namically favored five- or six-membered lactone; the second
point is that tethering the hydroxyl group and the amide
group together using rigid moieties, such as o-phenylene and
cyclohexyl groups, minimizes the conformational freedom of
the molecule. The reason for the more labile nature of 1a
than 1b was therefore related to the stability of the formed
lactones, that is, five-membered rings are more stable, in
general, than six-membered rings. However, the more labile
nature of 1a than 1c was related to the rigidity of the teth-
ers, that is, the o-phenylene tether is more rigid than the
cyclohexyl tether. Once pyrrolidine was formed by the ther-
mal dissociation of 1, it initiated the polymerization of the
epoxide. The propagating end in the polymerization of the
epoxide was an alkoxide, and the counter cation was a qua-
ternary ammonium.
1 K. Arimitsu, A. Oguri, M. Furutani, Mater. Lett. 2015, 140, 92–94.
2 K. Arimitsu, K. Fukuda, N. Sakai, Chem. Lett. 2014, 43, 831–
833.
3 J. V. Crivello, M. F. Aldersley, J. Polym. Sci. Part A: Polym.
Chem. 2013, 51, 801–814.
4 J. V. Crivello, J. Photopolym. Sci. Tech. 2009, 22, 575–582.
5 G. Yilmaz, B. Iskin, F. Yilmaz, Y. Yagci, ACS Macro. Lett.
2012, 1, 1212–1215.
6 M. A. Tasdelen, Y. Yagci, Aust. J. Chem. 2011, 64, 982–991.
7 Y. Yagci, S. Jockusch, N. Turro, J. Macromol. 2010, 43, 6245–
6260.
8 J. Y. Kim, N. S. Kim, P. S. Patil, K. D. Ahn, J. Kim, T. H. Kim,
J. Appl. Polym. Sci. 2009, 111, 1878–1883.
9 S. Nakano, T. Endo, Prog. Org. Coat. 1996, 28, 143–148.
10 S. J. Park, T. J. Kim, J. R. Lee, J. Polym. Sci. Part B: Polym.
Phys. 2001, 39, 2397–2406.
11 S. Nakano, T. Endo, J. Polym. Sci. Part A: Polym. Chem.
1996, 34, 475–480.
In Scheme 4, “Propagation-2” involves the ring-opening reac-
tion of the lactone by the terminal alkoxide. This reaction
gave an alkoxide, C, which could react with the epoxide into
a chain extended alkoxide, D. However, this propagation
could be competed with by a back reaction to reform the
alkoxide C and lactone. To clarify whether the lactones
released from 1 participated in the polymerization or not,
polymerization of GPE using 1 was performed at 180 8C. The
resulting mixtures were analyzed by TLC. In these cases,
phthalide and 3-isochromanone released from 1a and 1b,
respectively, were not detected, implying that they under-
went the ring-opening reaction and were incorporated into
the resulting polymers. However, cis-hexahydrophthalide
released from 1c was clearly detected, implying that this
sterically hindered lactone was less reactive in the anionic
polymerization system.
12 M. S. Kim, K. W. Lee, T. Endo, S. B. Lee, Macromolecules
2004, 37, 5830–5834.
13 S. J. Park, G. Y. Heo, D. H. Suh, J. Polym. Sci. Part A:
Polym. Chem. 2003, 41, 2393–2403.
14 E. S. Jang, S. B. Khan, J. Seo, K. Akhtar, Y. H. Nam, K. W.
Seo, H. Han, Macromol. Res. 2011, 19, 989–997.
15 M. A. M. Rahmathullah, A. Jeyarajasingam, B. Merritt, M.
Van Landingham, S. H. McKnight, G. R. Palmese, Macromole-
cules 2009, 42, 3219–3221.
16 M. K. Gupta, Macromol. Res. 2009, 17, 221–226.
17 T. Toneri, F. Sanda, T. Endo, Macromolecules 2001, 34,
1518–1521.
18 O. Shimomura, I. Tomita, T. Endo, J. Polym. Sci. Part A:
Polym. Chem. 2001, 39, 868–871.
19 O. Shimomura, I. Tomita, T. Endo, Macromol. Rapid Com-
mun. 1998, 19, 493–497.
20 J. V. Crivello, T. P. Lockhart, J. L. Lee, J. Polym. Sci. Polym.
Chem. Ed. 1983, 21, 97–109.
CONCLUSIONS
21 A. Renner, R. Moser, M. Bellus, H. Fuhrer, O. Hosang, G.
Szekely, J. Polym. Sci. Part A: Polym. Chem. 1988, 26, 1361–
1376.
We have demonstrated that three hydroxylamides, 1, served
as thermally latent initiators in the polymerization of the
epoxide, bisphenol-A diglycidyl ether. The dissociation tem-
peratures of these hydroxylamides depended on their struc-
tures, and thus the temperatures for carrying out the
22 M. Kirino, F. Sanda, T. Endo, J. Polym. Sci. Part A: Polym.
Chem. 2000, 38, 3428–3433.
23 M. Kim, F. Sanda, T. Endo, Macromolecules 2000, 33, 3499–3501.
24 T. Hino, S. Taniguchi, N. Kuramoto, Macromolecules 2006,
39, 2036–2038.
polymerizations of epoxide using
1 were tunable by
the appropriate choice of initiator. The hydroxylamides could
be synthesized easily from lactones and amines, and this
simple, and thus versatile, synthesis promises a wide range
of applications for these new types of latent initiators.
25 A. Sudo, H. Yamashita, T. Endo, J. Polym. Sci. Part A:
Polym. Chem. 2011, 49, 3631–3636.
26 F. Sanda, T. Kaizuka, A. Sudo, T. Endo, Macromolecules
2003, 36, 967–968.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2016, 00, 000–000
7