Aminimide derived from benzoylformic acid ester as photolatent base/radical initiator

Article abstract of DOI:10.1002/pola.25924

An aminimide possessing a benzoyl substituent, 1,1-dimethyl-1-(2- hydroxypropyl)amine benzoylformimide (BFI), proved to serve as an excellent photobase catalyst. BFI decomposes smoothly by the UV irradiation to give products containing tertiary amines. The effective nature of BFI as a photo/thermal dual-base catalyst was convinced by the thermal and photoinduced polymerization of epoxide/thiol system. Based on the facts that the mixture of BFI and epoxide/thiol exhibit a long pot life in dark and that it undergoes smooth polymerization by UV irradiation and heating, it was supported that BFI serves as an efficient photo/thermal latent dual-base catalysts. It was also found that BFI initiates the free radical polymerization of vinyl monomers such as 2-hydroxylethyl methacrylate (HEMA) under the UV irradiation while the mixture of BFI and HEMA also exhibit a long pot life in dark, indicating the excellent ability of BFI as a photoradical initiator.

Full text of DOI:10.1002/pola.25924

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
Aminimide Derived from Benzoylformic Acid Ester as Photolatent
Base/Radical Initiator
Manabu Kirino,1 Ikuyoshi Tomita2
1Product Development Division, Threebond Co. Ltd., 1456 Hazama-Cho, Hachioji, Tokyo 193-8533, Japan
2Department of Electronic Chemistry, Interdisciplinary Graduate School of Science and Engineering,
Tokyo Institute of Technology, Nagatsuta 4259-G1-9, Midori-ku, Yokohama 226-8502, Japan
Correspondence to: I. Tomita (E-mail: tomita@echem.titech.ac.jp)
Received 25 July 2011; accepted 7 January 2012; published online 31 January 2012
DOI: 10.1002/pola.25924
ABSTRACT: An aminimide possessing a benzoyl substituent,
1,1-dimethyl-1-(2-hydroxypropyl)amine benzoylformimide (BFI),
proved to serve as an excellent photobase catalyst. BFI decom-
poses smoothly by the UV irradiation to give products contain-
ing tertiary amines. The effective nature of BFI as a photo/
thermal dual-base catalyst was convinced by the thermal and
photoinduced polymerization of epoxide/thiol system. Based on
the facts that the mixture of BFI and epoxide/thiol exhibit a long
pot life in dark and that it undergoes smooth polymerization by
UV irradiation and heating, it was supported that BFI serves as
an efficient photo/thermal latent dual-base catalysts. It was also
found that BFI initiates the free radical polymerization of vinyl
monomers such as 2-hydroxylethyl methacrylate (HEMA) under
the UV irradiation while the mixture of BFI and HEMA also ex-
hibit a long pot life in dark, indicating the excellent ability of BFI
C
V
as
a
photoradical initiator.
2012 Wiley Periodicals, Inc.
J Polym Sci Part A: Polym Chem 50: 1556–1563, 2012
KEYWORDS: anionic polymerization; catalysts; photopolymeriza-
tion; radical polymerization
INTRODUCTION The photocuring technique is important for
many practical applications in the fields of adhesives, seal-
ants, paints, coatings, molding resins, and so on. Especially,
photoinitiated radical and cationic curing systems have been
studied widely, and they have been utilized in various indus-
trial applications.1,2 Although these photocuring systems are
effective for practical applications, curing does not occur suf-
ficiently if the materials are not transparent enough toward
the photoirradiation. To improve this issue, the postpolyme-
rization technique by heat or moisture (so-called ‘‘dual-
cure’’) is expected to assist the incomplete photocuring pro-
cess. In the radical and/or cationic systems, various initiators
have been applied to the photo/thermal dual-cure proc-
esses.3 For instance, peroxides, azo compounds, and azo-ben-
zoin initiators have been utilized as photo/thermal dual-radi-
cal initiators,4–7 and some onium salts have been also
applied as photo/thermal dual-cationic initiators.8–12 How-
ever, both the radical and the cationic systems have inevasi-
ble drawbacks: the radical system is often deactivated by ox-
ygen, and the cationic system is susceptible to weak bases
originated from moisture or impurities in monomers and/or
prepolymers.
cies would be promising methods for industrial applications.
A number of photobase catalysts have been evaluated their
ability to catalyze polymerizations of epoxide, epoxide/thiol,
epoxide/carboxylic acid, alcohol/isocyanate, and cyanoacry-
late.2,13,14 However, few photo/thermal dual-base catalyst
has been reported so far.
Aminimides are attractive candidates for photo/thermal
dual-base catalysts, because they are reported to undergo
the thermal cleavage of the NAN bonds to produce tertiary
amines,15,16 which are applicable as thermal latent base cata-
lysts for epoxides,17–21 epoxide/acid anhydride system,17 and
epoxide/thiol system,22,23 and because some of the amini-
mide derivatives such as benzimides, aminimides possessing
PhACOANꢀANþ moieties, undergo photolysis to produce
tertiary amines and amides,22–25 that can be utilized as pho-
tobase catalysts for epoxide/thiol system.22,23 Despite their
potential ability of photo/thermal dual-base catalysts, their
poor thermal activity (i.e., their relatively high activation
temperature) made it difficult to attain practical applications.
Recently, we synthesized a new class of aminimide, 1,1-di-
methyl-1-(2-hydroxypropyl)amine benzoylformimide (BFI;
Fig. 1), which was found to serve as a thermal latent base
catalyst exhibiting high activity at lower temperature. That
is, BFI generates benzoyl isocyanate, 1,1-dimethylamino-2-
In contrast, the anionic and the base-catalyzed curing system
are known to be free from these inhibition processes.
Accordingly, polymerization systems catalyzed by base spe-
Additional Supporting Information may be found in the online version of this article.
C
V
2012 Wiley Periodicals, Inc.
1556
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 1556–1563

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
fractions were directly trapped by liquid nitrogen and were
analyzed by GC–MS (conditions: sample temperature: 120
ꢁC; split flow: 40 mꢁL minꢀ1 (He gas); split ratio: 1/40; GC
ꢁ
heating profiles: 50 C for 5 min and then heated to 300 C
at a heating rate of 10 ꢁC minꢀ1; ionization voltage: 70 eV).
Differential scanning calorimetry analyses (DSC) were per-
formed on a Seiko Instruments DSC 220 under a nitrogen
atmosphere.
FIGURE 1 Structure of BFI.
propanol (1), and their further reaction products by heating
above 80 C via the Curtius-type rearrangement mechanism,
ꢁ
UV Spectra of BFI before and after UV Irradiation
A sealed crystal UV cell containing a methanol solution of
BFI (7.0 ꢂ 10ꢀ5 M) was subjected to the UV irradiation (100
mW cmꢀ2) with stirring. UV–vis spectra were measured after
designed periods.
and 1 catalyzes the polymerization of epoxide and epoxide/
thiol system.26 Because BFI has phenyl a-diketone moiety, it
is also expect that BFI also exhibits photoactivity. Thus, we
evaluated the photobase activity of BFI in detail to realize a
novel photo/thermal dual-base catalyst. The possibility of
BFI as photoradical initiator was also investigated in the
photoinitiated polymerization of an acrylate monomer.
pH Value of BFI Aqueous Solution before and after UV
Irradiation
In a sealed glass sample tube equipped with a pH electrode,
a water solution of BFI (4.0 ꢂ 10ꢀ3 M) was subjected to the
UV irradiation (100 mW cmꢀ2) with stirring, and the pH
value was evaluated after designated period.
EXPERIMENTAL
Materials
BFI was synthesized from benzoylformic acid methyl ester,
1,1-dimethylhydrazine, and propylene oxide as described in
our previous article.26 Benzoylformic acid methyl ester, 1,1-
dimethylamino-2-propanol (1), 2,2,6,6-tetramethylpiperidine-
1-oxyl (TEMPO), and benzophenone were purchased from
Tokyo Kasei Kogyo (Japan). Benzyl alcohol, pentaerythritol
tetrakis(3-mercaptopropionate), EXA-850CRP (pure grade
bisphenol A diglycidyl ether), and Irgacure 184 (1-hydroxy-
cyclohexyl phenyl ketone) were purchased from Wako Pure
Chemical Industries (Japan), Aldrich (Milwaukee), DIC
(Japan), and BASF (Germany), respectively, and they were
used as received. ACRYESTER HO (2-hydroxylethyl methacry-
late (HEMA), containing 50 ppm of 4-methoxyphenol as a
radical polymerization inhibitor) were purchased from
Mitsubishi Rayon (Japan) and used both as received and
after purification by distillation under reduced pressure.
Identification of Photodegradation Products of BFI
In a sealed quartz glass UV cell, a methanol (2.2 ml) solution
of BFI (1.0 g, 4.0 mmol) was irradiated with UV light (100
mW cmꢀ2) at 20–30 C for 50 min with stirring. During the
ꢁ
irradiation, white powder started precipitating out from the
solution after about 10 min. The resulting mixture was then
completely evaporated to dryness under reduced pressure
ꢁ
(30 C/0.23 mmHg), and the volatile portions were collected
by dry ice/acetone trap. The residual paste (0.70 g) contain-
ing the white powder was washed twice with methanol to
give 2,3-dihydroxy-2,3-diphenyl-succinic acid diamide (2)
(0.060 g, 0.20 mmol, 10% yield) as white powder in an
essentially pure form; mp 186.0–187.0 ꢁC (decomposition).
IR (solid, ATR): 3465, 3332, 3255, 3184, 1662 (s, C¼¼O),
1593, 1495, 1426, 1307 cmꢀ1
.
Methods
1H NMR (DMSO-d6): d 7.25 (s, 2H, OH), 7.28–7.36 (m, 6H,
C6H5), 7.56 (s, 2H, NH2), 7.66 (s, 2H, NH2), 7.76–7.79 (m, 4H,
C6H5). 13C NMR (DMSO-d6): d 78.7, 127.0, 127.4, 127.5,
140.9, 175.7. HRMS (FABþ) [MþH]þ: Calcd for C16H17N2O4:
m/z 301.1188. Found: m/z 301.1192, [MþNa]þ: Calcd for
The UV irradiation was performed with a Hamamatsu Pho-
tonics LC8/L8251 (200 W mercury-xenon lamp) equipped
with a direct beam uniform illumination unit (E10052-01)
without UV filter. The light intensity at 365 nm was 20 or
100 mW cmꢀ2. UV–vis spectra were measured by a Shi-
madzu UV-3100PC spectrometer. The pH values were meas-
ured by a Horiba D-54 pH meter. IR spectra were taken on a
BIO-RAD FTS-40 FTIR system using a SPECAC MK II heated
diamond ATR system. Melting points were measured on a
Mettler Toledo FP90/FP82HT thermo system and an Olym-
pus BX50 microscope at a heating rate of 2 ꢁC minꢀ1. Nu-
clear magnetic resonance (NMR) spectra were recorded with
a JEOL ECP300 spectrometer in CDCl3 (300 and 75 MHz for
1H NMR and 13C NMR, respectively). The high-resolution
mass spectrometry (HRMS) was performed with a JEOL
JMS700 system (ion mode: FABþ, matrix: nitrobenzyl alcohol
with NaI). Gas chromatographic–mass spectrometric (GC–
MS) analyses were performed on a Thermoquest GCQ GC–MS
system equipped with a Frontier Laboratories UA5 (MS/HT)-
30M-0.25F GC column and a liquid nitrogen cryo-trap. The
samples were heated at 50 ꢁC for 1 min, and the volatile
C16H16N2O4Na: m/z 323.1008, Found: m/z 323.1004. TLC
analysis of the methanol-soluble part exhibited three spots.
However, its purification by column chromatography (silica
gel and ethyl acetate/methanol) was unsuccessful, and the
unreacted BFI (0.37 g, 37%) was recovered from the mix-
ture. From the cryo-trap GC–MS analysis of the volatile por-
tions, a number of GC peaks were observed. Among them,
one GC peak was identified to be 1 from its MS spectrum in
comparison with that of the authentic sample. However, the
other GC peaks could not be identified by their MS patterns.
IR Spectra of BFI under UV Irradiation
A small portion of a benzyl alcohol solution of BFI (1.0 M)
was placed on a FTIR ATR stage (thickness of the sample:
0.004 mm, protected with a cover glass) and its real-time
FTIR spectra were measured at 25 ꢁC under the UV irradia-
tion (20 mW cmꢀ2). Conversions of the carbonyl group in
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 1556–1563
1557

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
BFI (NꢀACO) were estimated from the intensity of the peak
at 1594 cmꢀ1 with respect to that of the CAO absorption
peak in benzyl alcohol at 1014 cmꢀ1. As control experi-
ments, measurements of benzyl alcohol, a 1.0 M solution of
1 in benzyl alcohol, and a 0.5 M solution of 2 in benzyl alco-
hol were likewise performed.
Photoinduced Polymerization Studies of Epoxide/Thiol
System with BFI
To a homogenous mixture of BFI (0.0250 g, 0.100 mmol)
and benzyl alcohol (0.108 g, 1.00 mmol), EXA-850CRP
(0.688 g, 2.00 mmol) and pentaerythritol tetrakis(3-mercap-
topropionate) (0.489 g, 1.00 mmol) were added at room
temperature. Small portions of this mixture were subjected
to DSC and real-time FTIR analyses as follows.
FIGURE 2 UV–vis spectra of methanol solutions of BFI (7.0 ꢂ
10ꢀ5 and 7.0 ꢂ 10ꢀ3 M, cell length: 1.0 cm).
DSC Measurements
A portion of the mixture (50 mg) was placed in a sealed
clear borosilicate glass vial (diameter: 12 mm) and was irra-
diated with UV light (100 mW cmꢀ2) from the bottom of the
HEMA without the initiator and a mixture consisting of
HEMA (10 mmol), BFI (0.040 mmol), and TEMPO (0.020 or
0.040 mmol) were also performed to support the radical
mechanism. Mixtures of HEMA (10 mmol) and various addi-
tives (0.040 mmol) such as 1, benzophenone, benzoylformic
acid methyl ester, and 1-hydroxycyclohexyl phenyl ketone
were also subjected to the same measurements to compare
the polymerization behavior. The measurement of a mixture
consisting of HEMA (10 mmol), benzoylformic acid methyl
ester (0.040 mmol), and 1 (0.040 mmol) was also performed
to compare the photoradical activity of BFI and benzoylfor-
mic acid methyl ester in the presence of tertiary amine.
ꢁ
vial at 25 C. After the irradiation for designated period, 5.0
mg of the sample was transferred into an aluminum pan and
completely sealed. The polymerization profile was monitored
by the DSC, where the temperature of the sample was raised
from 20 to 200 ꢁC at a heating rate of 10 ꢁC minꢀ1. Subse-
quent heating cycles were monitored for determining the
glass transition temperatꢀu1re (Tg) from –30 to 100 ꢁC at a
ꢁ
heating rate of 10 C min
.
Real-Time FTIR Observations
A small aliquot of the mixture was put on a FTIR ATR stage,
whose surface was protected with a cover glass to avoid the
inhibition by oxygen. The thickness of the sample was 0.004
mm. The measurements were performed at 25 or 85 ꢁC
under the UV irradiation of 100 mW cmꢀ2 for 30 or 60 s.
The conversion was determined from the peak intensities of
RESULTS AND DISCUSSION
UV Absorption Spectra of BFI
As shown in Figure 2, the UV–vis spectra of BFI in methanol
exhibited two absorption peaks at 249 and 346 nm whose
molar extinction coefficient (e) were 1.2 ꢂ 104 L molꢀ1
cmꢀ1 and 1.0 ꢂ 102 L molꢀ1 cmꢀ1, respectively. As expected,
BFI was found to exhibit absorptions suitable for emission
lines of the high-pressure mercury lamp at 254, 313, and
both the thiol (2568 cmꢀ1) and the epoxide (914 cmꢀ1
with respect to that of the carbonyl absorption for the ester
(1736 in pentaerythritol tetrakis(3-
cmꢀ1
mercaptopropionate).
)
)
365 nm (e ¼ 1.1 ꢂ 104, 1.3 ꢂ 102, 7.8 ꢂ 10 L molꢀ1 cmꢀ1
,
Photoinitiated Radical Polymerization of Acrylic
Monomer with BFI under UV Irradiation
respectively).
Bulk Polymerization
To evaluate the photoreaction behavior of BFI, the UV–vis
spectra of BFI (7.0 ꢂ 10ꢀ5 M in methanol) were measured
after the irradiation of the high-pressure mercury lamp (Fig.
3). Consequently, the intensity of the peak at 249 nm
decreased and that of a new peak at 229 nm increased until
A homogeneous mixture of BFI (0.010 g, 0.040 mmol) and
HEMA (used both as received and after purification by distil-
lation) (1.3 g, 10 mmol) was placed between two PET films
with a thickness of 0.5 mm. The sample was irradiated with
UV light (100 mW cmꢀ2) for 50 s (5 J cmꢀ2).
the total exposure of the UV irradiation of 4.0 J cmꢀ2
.
Real-Time FTIR Observations
A clear homogeneous mixture of HEMA (as received, 1.3 g,
10 mmol) and BFI (0.010 g, 0.040 mmol) was put on a FTIR
ATR stage, and the surface of the sample was protected with
a cover glass to avoid the inhibition by oxygen. The thickness
of the sample was 0.004 mm. The real-time FTIR spectra
Variation of pH Value of Aqueous Solution of
BFI by UV Irradiation
The variation of the pH value of an aqueous solution of BFI
(4.0 ꢂ 10ꢀ3 M) by the UV irradiation was studied to evalu-
ate the basicity of photoreaction products generated from
BFI. Before the UV irradiation, the solution was almost
neutral (pH ¼ 6.2), which immediately turned to basic
(pH ¼ 9–10) after the photoirradiation (Fig. 4). This result
indicates that BFI can generate base by the UV irradiation.
were measured under the UV irradiation of 20 mW cmꢀ2
.
The conversion was estimated from the peak intensity of the
vinyl absorption in HEMA (1635 cmꢀ1) with respect to that
of the carbonyl groups (1713 cmꢀ1). The measurements of
1558
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 1556–1563

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 3 Variation of the UV–vis spectra of BFI in methanol
([BFI]0 ¼ 7.0 ꢂ 10ꢀ5 M) after the UV irradiation of 0, 0.5, 1.0,
FIGURE 5 GC profile of volatile fractions obtained by the pho-
2.0, 4.0, and 10.0 J cmꢀ2
.
toreaction of BFI in methanol.
the real-time FTIR measurements of a benzyl alcohol solu-
tion of BFI after the UV irradiation, a peak at 1594 cmꢀ1 at-
tributable to the carbonyl group in the aminimide (NꢀACO)
gradually decreased [Fig. 6(a–d)]. Based on the variation of
this peak intensity, the conversion of BFI was estimated
which indicates the smooth progress of photoinduced reac-
tions (Fig. 7). After the photolysis reaction, a peak at 1679
cmꢀ1 for the benzoyl group (Ph-CO) in BFI shifted to 1666
cmꢀ1 which is attributable to the amide carbonyl group in 2.
In our recent report,26 we described that the thermal reac-
tion of BFI accompanies the formation of urethane deriva-
tives on the basis of the similar FTIR observation, in which a
peak of aminimide decreased and a new sharp peak of the
carbonyl group in the urethane at 1771 cmꢀ1 could be
observed by the thermal treatment. As the urethane is most
probably produced by the reaction of benzoyl isocyanate and
1,1-dimethylamino-2-propanol (1), it was supported that the
thermal reaction proceeds through the Curtius-type rear-
rangement. In contrast, no sharp peak attributable to the iso-
cyanate at 2224 cmꢀ1 or to the carbonyl group in the ure-
thane at 1771 cmꢀ1 was observed in the present photolysis
Photoreaction Mechanism of BFI
To identify the photoreaction products, we carried out the
GC–MS measurement of the volatile fractions obtained by the
photoirradiation of BFI in methanol. That is, after the UV irra-
diation of a methanol solution of BFI, volatile fractions in the
resulting mixture were collected by the trap-to-trap distilla-
ꢁ
tion at 30 C/0.23 mmHg. In the GC analysis (Fig. 5), a num-
ber of peaks appeared, among which a peak at 2.81 min could
be identified as 1 from its MS spectrum in comparison with
that of the authentic sample (Supporting Information Fig. 1).
From the nonvolatile residue remained in the trap-to-trap
distillation in 70 wt %, 2,3-dihydroxy-2,3-diphenyl-succinic
acid diamide (2) was isolated in 10% yield as white solid
(Scheme 1). Besides, the unreacted BFI was recovered in
37%. These results indicate that the photolysis takes place
at least partly via the cleavage of the NAN bond in BFI.
The photolysis of BFI was studied in detail by the FTIR
measurements. Among the good solvents for BFI such as
water, alcohols, and other polar media, benzyl alcohol was
chosen as a solvent because of its high boiling point. From
FIGURE 4 Variation of the pH value of an aqueous solution of
BFI by the UV irradiation ([BFI]0 ¼ 4.0 ꢂ 10ꢀ3 M).
SCHEME 1 Products identified in photoreaction of BFI in
methanol.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 1556–1563
1559

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
FIGURE 6 FTIR spectra of a 1.0 M solution of BFI in benzyl alcohol after UV irradiation of 0, 1.2, 3.0, and 10 J cmꢀ2 (a–d), 1.0 M so-
lution of 1 in benzyl alcohol (e), 0.5 M solution of 2 in benzyl alcohol (f), and benzyl alcohol alone (g).
process. This result indicates that BFI does not generate iso-
cyanates under the photolysis conditions.
A homogeneous mixture of bisphenol A diglycidyl ether
(2.00 mmol), pentaerythritol tetrakis(3-mercaptopropionate)
(1.00 mmol), BFI (0.100 mmol), and benzyl alcohol (1.00
mmol) was irradiated with a UV lamp for designated periods,
and the resulting samples were then subjected to the DSC
It has been reported that aminimides having RACOANꢀANþ
moieties undergo the photocleavage at the NꢀANþ bonds to
produce the corresponding tertiary amines and amides. The
generation mechanism of the amides may include the hydro-
gen abstraction by the intermediate triplet nitrenes.22–25 In
the case of benzoylformic acid esters, which have analogous
structure to benzoylformic acid amide (3), it has been
reported that a dimerized products were obtained by the
photoirradiation via the ketyl radical intermediate.27–32
Although further works including the structural elucidation
of the major photolysis products from BFI should be per-
formed so as to support the proposed mechanism, we
assume that the photoreaction of BFI proceeds at least partly
by the cleavage of the NAN bonds to produce 1 and 3, and
the resulting 3 dimerizes to give 2 via the ketyl radical inter-
mediate (Scheme 2).
ꢁ
measuremꢀen1 ts in the range of 20–200 C at a heating rate of
ꢁ
10 C min (Fig. 8 and Table 1). Without the UV irradiation,
BFI serves as a highly active thermal latent base catalyst.
Thus, the polymerization of the epoxide/thiol takes place
Photoinduced Polymerization of Epoxide/Thiol System
by BFI
As described above, the base-generation from BFI by photo-
irradiation was confirmed from both pH measurements and
GC–MS studies, although the major photoproducts from BFI
have not been clarified yet. Therefore, the photoinduced
polymerization of epoxide/thiol system was carried out
with BFI to demonstrate its activity as a photobase catalyst.
FIGURE 7 Conversion of BFI estimated from the peak intensity
of the carbonyl group (NꢀACO) at 1594 cmꢀ1 by UV irradiation
in a benzyl alcohol solution (1.0 M).
1560
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 1556–1563

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 1 DSC Onset Temperature, Heat Evolved (DH), and Tg
of the Products Obtained after DSC for the Mixture of BFI and
Epoxide/Thiol System after the UV Irradiation
UV Dose
DSC Onset
Temperature (ꢁC)a
(J cmꢀ2
)
DH (J gꢀ1 a
)
Tg (ꢁC)b
0
1
3
6
9
20
a
121.1
99.3
82.7
73.2
67.9
54.8
401
400
392
392
393
362
11.8
11.1
11.1
11.3
11.0
11.8
Measured by DSC (20–200 ꢁC, at a heating rate of 10 ꢁC minꢀ1).
Measured by DSC (–30 to 100 ꢁC, at a heating rate of 10 ꢁC minꢀ1).
b
the thiol proceeds by base catalysts, the acceleration by the
prolonged photoirradiation is most probably due to the
increased amount of the photogenerated base.
Once the base was generated by the UV irradiation at the ini-
tial stage, the present reaction was found to proceed also
under dark conditions (Fig. 9). That is, the mixture of the
epoxide/thiol system and BFI was irradiated by the UV lamp
at 25 C for 1.0 min (UV dose: 6 J cmꢀ2), and the time–con-
ꢁ
version curves of both the thiol and the epoxide groups in
the resulting mixture were followed in dark at that tempera-
ture using the real-time FTIR measurements. In this case, the
polymerization proceeded smoothly and was complete
within 100 min (Table 2, run 1). When the similar experi-
SCHEME 2 One plausible photoreaction mechanism of BFI in
methanol.
ꢁ
ment was carried out at 85 C (i.e., the UV irradiation at 85
ꢁC for 1 min, run 2), the polymerization took place rapidly,
resulting in the complete conversion during the UV irradia-
tion (6 J cmꢀ2). At 85 ꢁC, the polymerization occurred
smoothly enough, even after much shorter UV irradiation
treatment (i.e., 0.5 min, run 3). Compared to the preceding
ꢁ
above 121.1 C, while the mixture exhibits long pot life (e.g.,
more than 2 weeks at room temperature), as described in
our preceding article.26 After the UV irradiation, the DSC
onset temperature decreased by increasing the total light
irradiated (i.e., UV dose). However, both the heat evolved
(DH) and Tg of the products obtained after the DSC analyses
were almost constant, indicating that the conversions of the
epoxide and the thiol, and the crosslink density of the result-
ing polymers were almost constant irrespective of the irradi-
ation conditions. As the polymerization of the epoxide and
FIGURE 9 Time–conversion curves of the thiol and the epoxide
functional groups monitored by intensities of the peaks at 2568
and 914 cmꢀ1, respectively, in the IR spectra of a mixture of
epoxide/thiol and BFI at 25 ꢁC after the UV irradiation at the ini-
tial stage for 1.0 min (UV dose: 6 J cmꢀ2).
FIGURE 8 DSC profiles of the epoxide/thiol system in the
presence of BFI after the UV irradiation of 0, 1, 3, 6, 9, and
20 J cmꢀ2
.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 1556–1563
1561

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 2 Time Required for the Complete Polymerization of the
could be excluded. These results also indicate that the initiat-
ing species are not generated from amine or HEMA. Further-
more, the consumption of HEMA took place slower if TEMPO
was added as a radical trapping agent, where the increased
amount of TEMPO remarkably suppressed the polymeriza-
tion [Fig. 10(a–c)]. Accordingly, the polymerization of HEMA
most probably proceeds in the photoinitiated radical mecha-
nism in the presence of BFI. Furthermore, as shown in Fig-
ure 10(b), it was found that the activity of BFI is high
enough compared to commercial initiators such as benzophe-
none, benzoylformic acid methyl ester, and 1-hydroxycyclo-
hexyl phenyl ketone. The high activity of BFI might be due
to the aminimide structure, because the activity of benzoyl-
formic acid methyl ester was a little lower than BFI which
Epoxide/Thiol System after UV Irradiation at the Initial Stage
Polymerization Conditions
Temperature UV Irradiation UV Dose Polymerization
Run (ꢁC)a
Time (min)
(J cmꢀ2
)
Period (min)b
1
2
3
25
85
85
1.0
6
6
3
100
1.0
1.5
1.0
0.50
a
Temperature for both UV irradiation and postpolymerization in dark.
Total time (i.e., UV irradiation and postpolymerization time) required
b
for complete conversion of thiol and epoxide, determined by the peak
intensities at 2568 and 914 cmꢀ1 in the IR spectra.
reports on the photoinduced polymerization of the epoxide/
thiol system using aminimides such as 4-nitrobenzi-
mides,22,23 the present BFI-catalyzed system exhibits remark-
ably faster polymerization in much lower temperature, indi-
cating the highly active nature of BFI as a photobase catalyst
for the epoxide/thiol system. Considering the highly active
nature of BFI as a thermal latent base catalyst,26 it was sup-
ported that BFI serves as an efficient photo/thermal latent
dual-base catalyst.
Photoinitiated Radical Polymerization of Acrylic
Monomers with BFI
It has been reported that both benzoylformic acid esters27–32
and benzoylformic acid amide33 serve as photoradical initia-
tors. Because BFI also has benzoylformic acid moiety, it is
expected that BFI exhibits photoradical activity. As men-
tioned earlier, the proposed photoreaction mechanism of BFI
also implies free radical intermediates such as products via
the hydrogen abstraction by the triplet nitrenes and/or the
ketyl radicals (Scheme 2). Thus, the ability of BFI as an ini-
tiator for the photoinitiated radical polymerization of vinyl
monomers was evaluated. In this study, 2-hydroxyethyl
methacrylate (HEMA) was chosen because of its low volatil-
ity and good ability to solubilize BFI. A homogeneous mix-
ture of HEMA (both as received and after purification, 10
mmol) and BFI (0.040 mmol) was subjected to the polymer-
ization under the UV irradiation of 5 J cmꢀ2. From both the
purified and unpurified monomers, tough transparent film
was obtained which was hardly soluble in organic solvents
such as tetrahydrofuran, chloroform, toluene, and N,N-
dimethylformamide.
The conversion of HEMA in the photopolymerization with
BFI was monitored by the FTIR measurements. For example,
a homogeneous mixture of HEMA (without purification, 10
mmol) and BFI (0.040 mmol) was subjected to the UV irradi-
ation and the conversion of HEMA was monitored by the
variation of the peak intensity of the vinyl group at 1635
cmꢀ1 in the IR spectra. It was found that the smooth conver-
sion of HEMA took place in the presence of BFI under the
UV irradiation [Fig. 10(a)]. On the other hand, the consump-
tion of HEMA was very slow in the presence of 1 [Fig. 10(e)]
or without any initiators [Fig. 10(d)]. Thus, the possibility of
the anionic mechanism in the polymerization of HEMA/BFI
FIGURE 10 Conversions of the vinyl group in the photoiniti-
ated radical polymerization of HEMA (10 mmol) in the presence
of various additives: (a) BFI (0.040 mmol), (b) BFI (0.040 mmol)
and TEMPO (0.020 mmol), (c) BFI (0.040 mmol) and TEMPO
(0.040 mmol), (d) without initiators, (e) 1 (0.040 mmol), (f) ben-
zophenone (0.040 mmol), (g) benzoylformic acid methyl ester
(0.040 mmol), (h) benzoylformic acid methyl ester (0.040
mmol) and 1 (0.040 mmol), and (i) 1-hydroxycyclohexyl phenyl
ketone (0.040 mmol).
1562
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 1556–1563

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
6 Yagci, Y.; Onen, A. J. Macromol. Sci. Chem. 1991, A28, 129–141.
was also not improved by addition of 1 [Fig. 10(h)]. The mix-
ture of HEMA and BFI exhibits a long pot life (e.g., more
than 1 year at room temperature) in dark. These results
indicate that BFI also serves as an efficient photoradical
initiator.
7 Yagci, Y.; Onen, A.; Schnabel, W. Macromolecules 1991, 24,
4620–4623.
8 Hamazu, F.; Akashi, S.; Koizumi, T.; Takata, T.; Endo, T. J.
Photopolym. Sci. Technol. 1992, 5, 247–254.
9 Reetz, I.; Bacak, V.; Yagci, Y. Polym. Int. 1997, 43, 27–32.
CONCLUSIONS
10 Toneri, T.; Sanda, F.; Endo, T. J. Photopolym. Sci. Technol.
1999, 12, 159–164.
An aminimide having a benzoyl substituent, BFI, was found
to generate both base and radical species by the photoirra-
diation. The decomposition of BFI takes place smoothly by
the irradiation of UV light, and a tertiary amine and a dia-
mide (i.e., the dimer of benzoylformic acid amide) were
detected in the degraded products. BFI exhibits high photo/
thermal dual-base activity in the polymerization of epoxide/
thiol system. The activity of BFI as a photoradical initiator
was also evaluated in the polymerization of 2-hydroxyethyl
methacrylate, in which the polymerization takes place fast
enough compared to the conventional photoradical initiators.
These results support the promising applications of BFI both
as an efficient photo/thermal dual-base catalyst and photo-
radical initiators for polymerization and curing systems. The
applications of BFI as a base/radical hybrid catalyst that can
be utilized for both the base-catalyzed and radical polymer-
ization systems (e.g., a combination of epoxides and acryl-
ates) are currently being investigated.
11 Crivello, J. V.; Kong, S. Macromolecules 2000, 33, 833–842.
12 Atmaca, L.; Kayihan, I.; Yagci, Y. Polymer 2000, 41, 6035–6041.
13 Suyama, K.; Shirai, M. Prog. Polym. Sci. 2009, 34, 194–209.
14 Crivello, J. V.; Dietliker, K. In Photoinitiators for Free Radi-
cal, Cationic and Anionic Photopolymerization, 2nd ed.; Brad-
ley, G., Ed.; Wiley: New York, 1998; pp 479–566.
15 Slagel, R. C. J. Org. Chem. 1968, 33, 1374–1378.
16 Wawzonek, S. Ind. Eng. Chem. Prod. Res. Dev. 1980, 19,
338–349.
17 Niino, H.; Noguchi, S.; Nakano, Y.; Tazuke, S. J. Appl.
Polym. Sci. 1982, 27, 2361–2368.
18 Inubishi, S.; Ikeda, T.; Tazuke, S.; Satoh, T. J. Polym. Sci.
Part A: Polym. Chem. 1987, 25, 137–150.
19 Inubishi, S.; Ikeda, T.; Tazuke, S.; Satoh, T. J. Polym. Sci.
Part A: Polym. Chem. 1987, 25, 1363–1382.
20 Lee, S. D.; Sanda, F.; Endo, T. J. Polym. Sci. Part A: Polym.
Chem. 1997, 35, 689–694.
21 Kirino, M.; Sanda, F.; Endo, T. J. Polym. Sci. Part A: Polym.
Chem. 2000, 38, 3428–3433.
The authors gratefully acknowledge Takeshi Endo of Molecular
Engineering Institute, Kinki University, who kindly directed the
authors to the research works of aminimides. The authors are
also grateful to Takao Goto of Threebond Co., for his kind assis-
tance in GC–MS measurements.
22 Katogi, S.; Yusa, M. J. Photopolym. Sci. Technol. 2002, 15,
35–40.
23 Katogi, S.; Yusa, M. J. Polym. Sci. Part A: Polym. Chem.
2002, 40, 4045–4052.
24 Robson, P.; Speakman, P. R. H. J. Chem. Soc. B 1968, 463–467.
25 Freeman, S.; Harger, M. J. P. J. Chem. Res. Synop. 1988,
192–193.
REFERENCES AND NOTES
26 Kirino, M.; Tomita, I. Macromolecules 2010, 43, 8821–8827.
1 Crivello, J. V.; Dietliker, K. In Photoinitiators for Free Radical,
Cationic and Anionic Photopolymerization, 2nd ed.; Bradley,
G., Ed.; Wiley: New York, 1998; pp 61–478.
27 Via, F. A. U.S. Patent 4,038,164 (Stauffer Chemical Co.), July
26, 1977.
28 Schafer, K. J.; Hales, J. M.; Balu, M.; Belfield, K. D.; Stry-
land, E. W. V.; Hagan, D. J. J. Photochem. Photobiol. A 2004,
162, 497–502.
2 Tsunooka, M.; Suyama, K.; Okumura, H.; Shirai, M. J. Photo-
polym. Sci. Technol. 2006, 19, 65–71.
3 Peeters, S. In Radiation Curing in Polymer Science and Tech-
nology; Fouassier, J. P.; Rabek, J. F., Eds.; Elsevier: London,
1993; Vol. III, pp 177–218.
29 Hu, S.; Neckers, D. C. J. Mater. Chem. 1997, 7, 1737–1740.
30 Hu, S.; Wu, X.; Neckers, D. C. Macromolecules 2000, 33,
4030–4033.
4 Neckers, D. C.; Jager, W. In Photoinitiation for Polymeriza-
tion: UV & EB at the Millenium, Chemistry & Technology of UV
& EB Formulations for Coatings, Inks and Paints; Wiley: New
York, 1998; Vol. VII, pp 8–12.
31 Encinas, M. V.; Lissi, E. A.; Zanocco, A.; Stewart, L. C.;
Scaiano, J. C. Can. J. Chem. 1984, 62, 386–391.
32 Hu, S.; Neckers, D. C. J. Org. Chem. 1996, 61, 6407–6415.
5 Onen, A.; Yagci, Y. J. Macromol. Sci. Chem. 1990, A27,
743–753.
33 Via, F. A. U.S. Patent 4,118,298 (Stauffer Chemical Co.), Oc-
tober 3, 1978.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE PART A: POLYMER CHEMISTRY 2012, 50, 1556–1563
1563