Synthesis of aliphatic polymers with high refractive index by photoinduced polyaddition of thiols to bifunctional allyl monomer containing tetrathiaspiro structure

Full text of DOI:10.1002/pola.29377

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
RAPID COMMUNICATION
Synthesis of Aliphatic Polymers with High Refractive Index
by Photoinduced Polyaddition of Thiols to Bifunctional Allyl
Monomer Containing Tetrathiaspiro Structure
Naoto Aoyagi, Takeshi Endo
Molecular Engineering Institute, Kindai University, Iizuka, Fukuoka 820-8555, Japan
Correspondence to: T. Endo (E-mail: tendo@moleng.fuk.kindai.ac.jp)
Received 28 February 2019; accepted 25 March 2019
DOI: 10.1002/pola.29377
KEYWORDS: aliphatic polymer; high refractive index; photoinduced polyaddition; tetrathiaspiro ring; thiol–ene reaction
In this article, we took note of the spiro structure in which
four sulfur atoms are bonded to a carbon atom and investi-
gated the solvent-free photocure polymerization of the bisallyl
monomer having tetrathiaspiro ring and multifunctional thiols
(T1, T2) to obtain the crosslinked polymers (P1, P2) that have
high refractive index (Scheme 1).
INTRODUCTION Polymers with high refractive index have been
utilized as the material for optical purpose in such devices as eye-
glass lenses,1 camera lenses,2 optical adhesives,3 anti-reflection
coatings,4 LED encapsulants,5 and so on. Large majority of the
reports on the synthesis of the polymers that have high refractive
index deal with the aromatic polymers having either or both
arylthioether structure and imide structure, or the aliphatic poly-
mers which have halogenated aryl group or sulfur atom on
their main chain, with most of the study dealing with linear
polymers.6,7 The aliphatic monomers which can be used as
photocurable resin and have high refractive index are rare. Inci-
dentally, the monomers to give polymers that have high refractive
index tend to be solid at room temperature and have poor com-
patibility with polymerization initiator or other monomers and
are thus unfit to the photocuring reactions. However, the develop-
ment of the photocuring reactions that use liquids and result
polymers having high refractive index is sought after with view of
the application in the nanoimprint technology such as prism
sheet8 and microlens.9
The materials to derive the bisallyl monomer having tetrathiaspiro
ring (M1), that is, episulufide (1) and trithiocanbonate (2) were
synthesized by referring to the procedure that authors previously
reported [Scheme 1(a)].15,16 First, an equivalent of 2,3-butanediol
and two equivalent of potassium thiocyanate were added to
ꢀ
allylglycidyl ether, and the mixture was stirred for 10 h at 25 C
to obtain 1 as a colorless liquid in 98% yield after purification by
silica gel column chromatography.15 Next, after the addition of
catalytic amount of 1,2-dimethyl tetrahydropyrimidine to car-
bon disulfide to generate the zwitterionic dithiocarbamate,
ꢀ
substance 1 was added and stirred for 24 h at 25 C to obtain
2 as a yellow oil in 99% yield after purification by silica gel col-
umn chromatography.16
Of various polymerization reactions, thiol–ene reaction is fit
for the photocuring reaction because it rarely suffers oxygen
inhibition and their polymerization proceeds fast to result
polymers that have uniform crosslink density with high con-
version rate and low shrinking.10–12 Recently, the study of the
UV nanoimprinting by using the photoinduced thiol–ene reac-
tion between bisacrylate having arylthioether structure and
bisthiol having diarylfluorene structure was reported.13 Unfor-
tunately, problems are cast by the fact that both monomers
are solid at room temperature due to the effect from aryl
rings requiring the use of solvent on carrying out the photo-
curing process and that the refractive index of the resulted
polymer film is lower in comparison with those of the existing
thermoplastic polymers that possess arylthioether structure.14
We next investigated the synthesis of bisallyl monomer having
tetrathiaspiro ring (M1) by way of the addition reaction
between 1 and 2 using Lewis acid as catalyst (Table 1). As a
typical experiment, a solution of 2 in dichloromethane was
stirred at 0 C, and then, 20 mol % of Lewis acid was added
to the solution all at once. A solution of 1 in dichloromethane
ꢀ
was added dropwise for 1 h to the reaction mixture and kept
ꢀ
under stirring for 12 h after raising the temperature to 25 C.
The ratio of the yield of the resulted bisallyl monomer (M1)
versus that of the byproduct poly(episulfide) (10) was derived
from 1H NMR analysis. As it turned out, M1 is obtained as
main product when boron trifluoride diethyl etherate (BF3-
Et2O) was used, but the yield was not that good because 10
Additional supporting information may be found in the online version of this article.
© 2019 Wiley Periodicals, Inc.
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2019
1

JOURNAL OF
POLYMER SCIENCE
RAPID COMMUNICATION
WWW.POLYMERCHEMISTRY.ORG
SCHEME 1 Synthesis of high refractive index crosslinked polymers (P1, P2) by photoinduced polyaddition of thiols (T1, T2) to
bifunctional allyl monomer containing tetrathiaspiro structure (M1).
was also produced as byproduct in 30% yield (entry 1). Although
epoxides are known to undergo cationic ring-opening reaction in
the presence of catalytic amount of BF3-Et2O, the episulfides have
higher reactivity than the epoxides. To give an example, the reac-
tion of dithioester with epoxide at room temperature in the pres-
ence of BF3-Et2O proceeds to give spirocyclic 1,3-oxathiolane in
70% yield, but the use of episulfide instead of epoxide makes the
cationic ring-opening reaction to proceed as the main reaction,
and the yield of spirocyclic 1,3-dithiolane becomes as low as
1%.17 We next verified that the use of copper(II) chloride (CuCl2)
as the Lewis acid resulted in a very low yield of M1 (5%),
whereas 10 was generated in the yield of 91% (entry 2). In con-
trast, the ring-opening reaction of 1 was suppressed when zinc
chloride (ZnCl2) was used as Lewis acid, and the yield of M1
remarkably increased up to 90% (entry 3). Changing the counter
anion to triflate (TfO) from Cl by using Zn(OTf)2, the reactivity
and selectivity of the addition reaction of 1 and 2 decreased
(entry 4). The reaction depicted by entry 3, the one that uses zinc
chloride, was scaled up, and the resultant product was purified by
silica gel column chromatography to obtain M1 in 87% yield. The
aliphatic bifunctional monomer containing tetrathiaspiro ring
(M1) resulted as clear colorless liquid as we expected, and the
value of refractive index (nD25) was 1.582.
The UV–Vis absorption spectrum of trithiocarbonate (2) differs
considerably from that of the bifunctional monomer (M1)
(Supporting Information, Fig. S1). An acetonitrile solution of
2 had a strong absorption band near 315 nm of π–π* transitions
and a weak band at about 442 nm of forbidden n–π* transitions,
both of which arose from the trithiocarbonyl group.18 We
assumed that the origin of the yellowish color of the substance
was caused by the weak band (442 nm). In contrast, substance
M1, in which the double bond of trithiocarbonyl group was
converted to single bond by addition reaction, no significant
absorption band was observed at the wavelength longer than
250 nm. Because the irradiation wavelength of metal halide
lamp used for the photocuring procedure is in the range of
250–450 nm, substance M1 that little absorb the light of this
wavelength range should fit as a monomer for the photocuring
reaction.
We next examined the synthesis of the crosslinked polymers
(P1, P2) by way of solvent-free photocure reaction of M1 and
2
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2019

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
RAPID COMMUNICATION
progress of curing via thiol–ene reaction was thus confirmed.
Similar to the photocuring of the M1 and T1 combination,
the photocuring occurred to the combination of M1 and
tetrafunctional thiol (T2) giving out crosslinked polymer (P2) as
pale yellow soft film after 10 min photoirradiation. The IR spec-
trum observed for P2 was compared with those of M1 and T2
in Figure 1(b). Referring to Figure 1(a) and (b), one finds that
absorption peaks at 992 and 922 cm−1, which arise from vinyl
group, weakened in P2 but yet stronger in comparison with the
peaks arising from the remnant vinyl group in P1 [Fig. 1(a)].
This situation may be explained by the fact that, all of T1 used
as starting material for P1 have primary thiol group, whereas
half of the T2 are secondary thiol group becoming less reactive
to the thiol–ene polyaddition reaction, accordingly. By similar
reason, the absorption peak (2543 cm−1) that arises from the
thiol group of P1 was barely discernible, whereas that at
2538 cm−1 of P2, coming from the thiol group of T2, was con-
firmed to be still remnant.
TABLE 1 Effect of the Lewis Acids on the Ring-Expansion Addition
of 1 with 2a
Yield (%)b
Entry
Lewis acid
M1
10
1
2
3
4
BF3-Et2O
CuCl2
67
30
91
10
22
5
ZnCl2
90 (87)c
Zn(OTf)2
29
a
We next measured the refractive index (nD25), Abbe number
(νd), 5% weight loss temperatures (Td5), and glass transition
temperature (Tg). The results are summarized in Table 2. The
values of nD25 of P1 and P2 were both around 1.63 increasing
by 0.05 from that of M1. This increase was ascribed to the
Reagents and conditions:
2 (1 mmol) + CH2Cl2 (10 mL), Lewis acid
(20 mol%), 1 (1 mmol) + CH2Cl2 (10 mL), 0 ꢀC, 1 h and 25 ꢀC, 12 h.
b
c
Determined by 1H NMR spectroscopy.
The value in parenthesis is the yield of the isolated product for 4 mmol
scale reaction.
25
rather large values of nD of T1 and T2, namely, 1.630 and
commercially available multifunctional thiols (T1, T2) [Scheme 1
(b)]. In a typical experiment, M1 and T1 or T2 were mixed in
1:1 ratio in the amount of the allyl group of M1 with respect to
that of the thiol group of T1 or T2. Then, 2 mol % of
1-hydroxycyclohexylphenyl ketone (3) was dissolved as the
photo-radical initiator. The liquid mixture was placed between
two quartz glass plates separated by the spacer of 250 μm thick-
ness and irradiated by a metal halide lamp (250–450 nm,
40 mW/cm2) at room temperature for the period of 10 min. The
photocure reaction of M1 and trifunctional thiol (T1) resulted
the formation of crosslinked polymer (P1) as pale yellow soft
film. The infrared absorption (IR) spectrum of the obtained P1
was shown in Figure 1(a) along with those observed for M1 and
T1. The intensity of the peaks arising from the vinyl group,
located at 992 cm−1 and 922 cm−1, decreased greatly in P1 from
that in M1, and the peak arising from thiol group, which is
located at 2543 cm−1, became nearly unobservable in P1. The
1.641, respectively, and to the increase of the molecular den-
sity by the crosslinking. Incidentally, this value was similar to
that of the polymer, namely 1.638, which was obtained by the
photocuring reaction of bisacrylate having arylthioether struc-
ture and the bisthiol having diarylfluorene structure.13 The
values of P1 and P2 are rather large for the values of aliphatic
polymers having neither aromatic ring nor halogens. The Abbe
number (νd) is the reciprocal of the dispersive power, which
is also an important value for optical materials. The higher νd
means a lower dispersion in the refractive index, and the νd is
defined by eq. 1.
νd = ðnd −1Þ=ðnF −nCÞ
ð1Þ
The values of νd were calculated from the values of refractive
index obtained at three wavelengths; 486 nm (nF), 588 nm
FIGURE 1 (a) IR spectra of P1, M1, and T1, (b) IR spectra of P2, M1, and T2.
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2019
3

JOURNAL OF
POLYMER SCIENCE
RAPID COMMUNICATION
WWW.POLYMERCHEMISTRY.ORG
then a solution of 2-(allyloxymethyl)thiirane (1) (4 mmol,
521 mg) in anhydrous dichloromethane (40 mL) was added
to the mixture over a period of 1 h. After the addition was
TABLE 2 Optical and Thermal Properties of P1 and P2
Film
Appearance
nD25 (−)a
νd (−)a
T
d5 (ꢀC)b
T
g (ꢀC)c
ꢀ
complete, the reaction temperature was raised to 25 C for
P1
P2
Pale yellow
Pale yellow
1.633
1.630
38.3
38.2
246
237
−24.5
−22.7
12 h, with continuous stirring. The mixture was evaporated in
vacuo, and the residue was purified by silica gel column chro-
matography (hexane/ethyl acetate = 10:1) to give M1 (1.18 g,
a
Measured by ATAGO DR-M4 Multi-Wavelength Abbe Refractome-
ter (25 ꢀC).
1
87%) as a colorless liquid. H and 13C NMR spectra of M1 are
b
c
TGA (10 ꢀC/min under N2).
DSC (5 ꢀC/min under N2).
shown in Figures S2 and S3 (Supporting Information).
1
ꢀ
H NMR (400 MHz, CDCl3, 25 C, δ): 3.33–3.58 (m, 6H, SCH2,
Allyl-OCHH), 3.69–3.81 (m, 2H, Allyl-OCHH), 3.94–4.18 (m, 6H,
CH2 CHCH2, SCH), 5.20 (d, 2H, J = 10.0 Hz, CHtrans CH), 5.28
(dd, 2H, J = 17.0, 1.4 Hz, CHcis CH), 5.83–5.94 (m, 2H,
(nd), and 656 nm (nC) (Supporting Information, Table S1). The
values for P1 and P2 were 38.3 and 38.2, respectively, and
are rather large in view of the fact that the values for conven-
tional polystyrene and polycarbonate lie around 30 when the
values of the refractive index lie around 1.59.19 The value of
Td5 as obtained by thermal gravimetric analysis (TGA) of P1
13
ꢀ
CH2 CHgem). C NMR (100 MHz, CDCl3, 25 C, δ): 43.6, 43.7,
43.9, 44.0 (SCH2), 55.6, 56.5, 57.2, 57.3 (SCH), 70.2, 70.3, 70.8,
71.1 (Allyl-OCH2), 72.1, 72.2, 72.2 (CH2 CHCH2), 84.4, 84.7,
84.9 (CS4), 117.4, 117.5 (CH2 CH), 134.2, 134.3 (CH2 CH). IR
(ATR, cm−1): 2918, 2853 (C H), 1645 (CH2 CH), 1417
(S CH2), 1085 (C O), 992, 922 (CH2 CH). EI-MS (m/z):
336 (M+). FAB-HRMS (m/z): calcd for C13H21O2S4 (M + H)+
337.0424, found 337.0423.
ꢀ
ꢀ
is 24ꢀ6 C which is lower by nearly 10 C from the value
(237 C) of P2. We infer that, as the IR analysis showed that
quantity of the unreacted polymerizable functional groups in
P2 was slightly larger than that of P1, the lowering effect to
the thermal stability showed up. The value of Tg as obtained
ꢀ
by differential scanning calorimetry (DSC)ꢀwas −24.5 C for
ꢀ
P1 which is lower than that for P2, −22.7 C, by nearly 2 C.
Polymerization Procedure of Crosslinked Polymers
(P1, P2)
We infer this difference as due to the slightly higher density
of crosslinking in P2 because of the tetrafunctionality of T2,
although the photoreactivity of M1 and T2 is slightly lower
than that of M1 and T1.
To a mixture of M1 (0.3 mmol, 101 mg) and 4-mercaptomethyl-
3,6-dithia-1,8-octanedithiol (T1) (0.2 mmol, 52 mg) or 1,2,6,7-
tetramercapto-4-thiaheptane (T2) (0.15 mmol, 37 mg) was
added 1-hydroxycyclohexylphenyl ketone (3) (6 × 10−3 mmol,
1.2 mg) at room temperature. The homogeneous liquid mixture
was placed between two quartz glass plates separated by the
spacer of 250 μm thickness and then irradiated by a metal
halide lamp (250–450 nm, 40 mW/cm2) at room temperature.
After 10 min of irradiation, crosslinked polymer (P1 or P2) was
obtained as pale yellow soft film (ca. 250 μm thick). The
obtained crosslinked polymers were analyzed by IR, TGA, DSC,
and Abbe refractometer.
To summarize, bisallyl monomer having tetrathiaspiro ring
was selectively synthesized in high yield by the addition reac-
tion of allyl-substituted episulfide and cyclic trithiocarbonate
in the presence of zinc chloride as the catalyst. In this reac-
tion, the formation of byproduct was affected by the choice of
Lewis acid; zinc chloride prevented the undesirable polymeri-
zation of episulfide, whereas copper(II) chloride gave undesir-
able poly(episulfide) as main product. The obtained bisallyl
monomer having tetrathiaspiro ring was colorless liquid, and
the values of its refractive index was large for aliphatic mono-
mers making it fit for the solventless photocuring. Photoin-
duced thiol–ene reaction of this monomer with trifunctional
or tetrafunctional thiols proceeded for 10 min in the presence
of inexpensive commercial photo-radical initiator to give
crosslinked films. Thus obtained crosslinked films had high
refractive indices for aliphatic polymers derived by photo-
curing, and the values of the Abbe number were also high.
Presently, the synthesis of tetrathiaspiro ring containing
bifunctional monomer with higher sulfur content is being con-
sidered for the purpose to obtain photocured crosslinked
polymer that has an even higher refractive index.
ACKNOWLEDGMENTS
We would like to acknowledge the help and financial support by
Mitsubishi Gas Chemical Company, Inc. involved in this work.
REFERENCES AND NOTES
1
Y. Nishimori, Y. Imagawa, K. Takemura, H. Horikoshi,
Y. Yamamoto, PCT Int. Appl., WO2018150951, 2018.
M. Shiratake, K. Ishihara, K. Hirose, S. Ikeda, N. Kato,
2
M. Kondo, S. Suzuki, K. Oshima, S. Nagayama, PCT Int. Appl.,
WO 2019044875, 2019.
3 S.-W. Lee, J.-W. Park, C.-H. Park, H.-J. Kim, Int. J. Adhes.
Adhes. 2014, 50, 93.
EXPERIMENTAL
4 K. C. Krogman, T. Druffel, M. K. Sunkara, Nanotechnology
2005, 16, S338.
Synthesis of 2,7- and 2,8-Bis(Allyloxymethyl)-
1,4,6,9-Tetrathiaspiro[4.4]Nonane (M1, Isomer Mixtures)
A mixture of 4-(allyloxymethyl)-1,3-dithiolane-2-thione (2)
(4 mmol, 825 mg) and zinc chloride (0.8 mmol, 109 ꢀmg) in
anhydrous dichloromethane (40 mL) was stirred at 0 C, and
5 X. X. Shang, S. Duan, M. Zhang, X. Y. Cao, K. Zheng,
J. N. Zhang, Y. M. Ma, R. B. Zhang, RSC Adv. 2018, 8, 9049.
6 J. G. Liu, M. Ueda, J. Mater. Chem. 2009, 19, 8907.
7 T. Higashihara, M. Ueda, Macromolecules 2015, 48, 1915.
4
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2019

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
RAPID COMMUNICATION
8 Y.-H. Ho, Y.-T. Chang, S.-W. Liu, H.-H. Lai, C.-W. Chu, C.-I. Wu,
W.-C. Tian, P.-K. Wei, Org. Electron. 2012, 13, 2667.
14 S. Seesukphronrarak, S. Kawasaki, K. Kobori, T. Takata,
J. Polym. Sci., Part A: Polym. Chem. 2007, 45, 3073.
9 D. Kang, C. Pang, S. M. Kim, H. S. Cho, H. S. Um, Y. W. Choi,
K. Y. Suh, Adv. Mater. 2012, 24, 1709.
15 N. Aoyagi, T. Endo, ChemistrySelect 2017, 2, 4465.
16 N. Aoyagi, T. Endo, Tetrahedron Lett. 2018, 59, 1702.
10 C. E. Hoyle, T. Y. Lee, T. Roper, J. Polym. Sci., Part A: Polym.
Chem. 2004, 42, 5301.
17 M. Blagoev, A. Linden, H. Heimgartner, Helv. Chim. Acta
1999, 82, 1458.
11 C. E. Hoyle, C. N. Bowman, Angew. Chem. Int. Ed. 2010, 49, 1540.
12 A. B. Lowe, Polym. Chem. 2014, 5, 4820.
18 J. Fabian, H. Viola, R. Mayer, Tetrahedron 1967, 23,
4323.
13 K. Nakabayashi, S. Sobu, Y. Kosuge, H. Mori, J. Polym. Sci.,
Part A: Polym. Chem. 2018, 56, 2175.
19 N. Sultanova, S. Kasarova, I. Nikolov, Bulg. J. Phys. 2013,
40, 258.
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2019
5