Synthesis of oligo(spiroketal)s by polycondensation of silyl ethers derived from naturally occurring myo-inositol with 1,4-cyclohexanedione

Article abstract of DOI:10.1002/pola.29461

Oligo(spiroketal)s (OSKs) were synthesized from myo-inositol, a naturally occurring cyclic compound bearing six hydroxyl groups. The successful synthesis of OSKs was achieved using silyl ethers 2 derived from 1,4-di-O-alkylated myo-inositol 1 as monomers,

Full text of DOI:10.1002/pola.29461

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ORIGINAL ARTICLE
Synthesis of Oligo(spiroketal)s by Polycondensation of Silyl Ethers Derived
from Naturally Occurring myo-Inositol with 1,4-Cyclohexanedione
Atsushi Sudo
,
1,2 Tomoki Yamasaki,1 Takuro Yamashita,1 Dai Ishida1
1Graduate School of Science and Engineering Research, Kindai University, Kowakae 3-4-1, Higashi Osaka, Osaka 577-8502, Japan
2Research Institute for Science and Technology, Kindai University, Kowakae 3-4-1, Higashi-Osaka, Osaka 577-8502, Japan
Correspondence to: A. Sudo (E-mail: asudo@apch.kindai.ac.jp)
Received 1 July 2019; accepted 31 July 2019
DOI: 10.1002/pola.29461
ABSTRACT: Oligo(spiroketal)s (OSKs) were synthesized from
myo-inositol, a naturally occurring cyclic compound bearing six
hydroxyl groups. The successful synthesis of OSKs was
achieved using silyl ethers 2 derived from 1,4-di-O-alkylated
myo-inositol 1 as monomers, which underwent polycondensa-
tion with 1,4-cyclohexanedione (CHD) at 0 ꢀC in the presence of
trimethylsilyl triflate as a catalyst. Because of the irreversible
nature of the condensation reaction of silyl ethers with ketones,
the resulting OSKs 7 had higher molecular weights than previ-
ously reported OSKs that were obtained by polycondensation of
tetraols 1 with CHD, where backward hydrolysis of the ketal
functions occurred. In addition, another series of OSKs, 8, were
synthesized using silyl ethers 3 derived from 2,5-di-O-alkylated
myo-inositol 6, which are more symmetric monomers than silyl
ethers 2. Silyl ethers 3 underwent efficient polycondensation
with CHD, whereas tetraol 6 did not, demonstrating that the der-
ivation of such tetraols into the corresponding silyl ethers is a
powerful strategy to access OSKs. © 2019 Wiley Periodicals, Inc.
J. Polym. Sci., Part A: Polym. Chem. 2019
KEYWORDS: high performance polymers; myo-inositol; oligo
(spiroketal); renewable resources; silyl ether; step-growth
polymerization
from its phosphorylated derivative, phytic acid, found in rice
bran.12,13 The well-established protocol for the regioselective
dialkylation of hydroxyl groups of myo-inositol at the 1- and
4-positions enables us to synthesize a variety of tetraols 1.14,15
The substituent R can be chosen from methyl, benzyl, and allyl.
When allyl was chosen, the resulting OSK was applicable as a
precursor for functionalized OSKs based on the thiol–ene
reaction of the allyl groups. This structural diversity of OSKs
demonstrated a distinct advantage of their synthesis from
myo-inositol-derived tetraols. However, the number-average
molecular weights of OSKs ranged from 1400 to 2100, prompt-
ing us to develop a more efficient method to synthesize OSKs
with higher molecular weights.
INTRODUCTION Spiropolymers and spirooligomers have
attracted considerable attention because of their double-
stranded and thus rigid structures, permitting them to exhibit
high mechanical and thermal stability.1–6 Their rigid rod-like
structures enable the spiropolymers to undergo spontaneous
alignment into bundle-like packing, which enhances their heat
resistance and mechanical strength. For example, Yaghi and
coworkers reported the dense packing of rigid nanorods of an
oligo(spiroorthocarbonate) (OSOC) synthesized by the polycon-
densation of pentaerythritol and tetraethylorthocarbonate.7
Endo and coworkers reported the successful synthesis of a
series of OSOCs bearing alkyl or aryl side chains that enhance
their solubility.8 They demonstrated crosslinking reactions of
the OSOCs based on the ability of the spiroorthocarbonate moi-
eties in the main chain to undergo cationic ring-opening homo-
polymerization and copolymerization with epoxide.9,10
Herein, we report an improved synthesis of OSKs by polycon-
densation of silyl ethers
2
with CHD, yielding the
corresponding OSKs with higher molecular weights than the
OSKs obtained from the polycondensation of tetraol 1 with
CHD (Scheme 1). The use of silyl ether for ketalization was
developed by Noyori and coworkers.16 It allows synthesis of
cyclic ketals from silyl ethers of 1,2-diols and ketones under
mild conditions (such as 0 ꢀC) catalyzed by trimethylsilyl
trifluoromethanesulfonate (TMSOTf). The byproduct of this
reaction is TMS2O, which does not react with the product
Poly(spiroketal)s are also an attractive family of rigid
rod-like polymers, which can be synthesized easily by poly-
condensation of tetraols with diketones.1–3 Previously, we
reported the polycondensation of tetraols
1 with 1,4-
cyclohexanedione (CHD) giving oligo(spiroketal)s (OSKs)
(Scheme 1).11 The tetraols were derived from naturally
occurring myo-inositol, a cyclic hexaol that can be produced
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
ORIGINAL ARTICLE
WWW.POLYMERCHEMISTRY.ORG
1,4-Di-O-methyl-myo-inositol (1a) and 1,4-di-O-phenylmethyl-
myo-inositol (1b) were synthesized and purified according to
the reported procedures.15 1,6:3,4-Bis-[O-(2,3-dimethoxy-
butane-2,3-diyl)]-myo-inositol (4), 1,6:3,4-bis-[O-(2,3-dimetho-
xybutane-2,3-diyl)]-2,5-di-O-methyl-myo-inositol (5a), 1,6:3,4-
bis-[O-(2,3-dimethoxybutane-2,3-diyl)]-2,5-di-O-phenylmethyl-
myo-inositol (5b), 2,5-di-O-methyl-myo-inositol (6a), and
2,5-di-O-phenylmethyl-myo-inositol (6b) were synthesized
and purified according to the reported procedures.15
OH
OH
HO
OH
OH
HO
myo-inositol
Previous Work
OR
OR
O
O
O
HO
HO
OH
OH
O
O
CHD
Instruments
- H2O
O
1
n
Nuclear magnetic resonance (NMR) spectra (400 MHz for H;
100.6 MHz for 13C) were recorded on a JEOL JNM-AL400 NMR
spectrometer. Chemical shifts, δ, and coupling constants, J, are
reported in ppm and Hz, respectively. Solid-state 13C-NMR
spectra (100.6 MHz) were recorded on a Bruker Biospin NMR
AVANCE HD400WB spectrometer.
OR
1
OR
oligo(spiroketal)s (OSK)
This Work
Me3SiO
OR
OR
OSiMe3
OSiMe3
Me3SiO
Me3SiO
OSiMe3
OSiMe3
3
6
2
5
2
1
4
5
1
6
3
4
The number-average molecular weight (Mn) and weight-
average molecular weight (Mw) were estimated by size exclusion
chromatography (SEC), performed on a Tosoh HLC-8120GPC
chromatograph equipped with Tosoh TSK gel-Super HM-H
styrogel columns (6.0 mm ϕ × 15 cm). DMF containing 1 wt %
lithium bromide was used as an eluent at a flow rate of
0.6 mL min−1 after calibration with polystyrene standards.
Me3SiO
OR
2 (silyl ether of 1)
OR
3 (isomer of 2)
Polycondensation with CHD for OSK synthesis
SCHEME 1 Outline of this work.
Thermogravimetric (TG) analyses were carried out using a
SHIMADZU DTG-60 TG/DTA simultaneous measuring system
under nitrogen flow. Each sample was heated from 50 to
500 ꢀC, at a rate of 10 ꢀC min−1. Differential scanning calorim-
etry (DSC) was carried out using a SHIMADZU DSC (model
DSC-60 Plus) under nitrogen flow. Each sample was heated
from 30 ꢀC to 250–300 ꢀC depending on the thermal degrada-
tion behavior of the polymer and then cooled to 0 ꢀC at a rate
of 20 ꢀC min−1. Then, the sample was heated from 0 ꢀC at a
rate of 10 ꢀC min−1 for the acquisition of the DSC data. The
sample was cooled again to 0 ꢀC and then heated at a rate of
10 ꢀC min−1 to confirm the reproducibility.
ketals, leading to the efficient suppression of the backward
reaction from ketals to the starting materials. This method
was used by Wessig and Mollnitz for a stepwise synthesis of
well-defined rigid rod-like oligomers.4
Another highlight of this study is the use of silyl ether 3, a ste-
reoisomer of 2, as a monomer for synthesizing OSKs with dif-
ferent stereochemistry from that of OSKs obtained by the
polycondensation of silyl ether 2. Silyl ether 3 was synthe-
sized from myo-inositol via O-alkylation of two hydroxyl
groups at the 2- and 5-positions.15
Matrix-assisted laser desorption ionization-time-of-flight
(MALDI-TOF) spectra were recorded on a SHIMADZU AXIMA
Confidence Mass Spectrometer. DHB and TfONa were used as
the matrix and ionization reagents, respectively. Samples were
prepared by mixing OSKs (3 mg) with DHB (10 mg) and
TfONa (10 mg) in tetrahydrofuran (1 mL). The resulting solu-
tion was dropped on a sample plate and dried under air.
EXPERIMENTAL
Chemicals
Sodium hydride (60 wt % in mineral oil) and triethylamine
(TEA, >99.0%) were purchased from Kishida Chemical Co., Ltd.,
Osaka and used as received. (4-tert-Butyl)phenylmethyl bro-
mide (>97%), N,N-dimethylformamide (DMF; >99.5%), and 1,2-
dichloroethane (>99.7%) were purchased from Wako Pure
Chemical Industries Co., Ltd., Osaka and distilled over calcium
hydride prior to use. Chlorotrimethylsilane (TMSCl) and TMSOTf
were purchased from Shin-Etsu Chemical Osaka and used as
received. CHD (>98%) was purchased from Tokyo Chemi-
cal Industry Co., Ltd., Tokyo and was used as received.
2,5-Dihydroxybenzoic acid (DHB; >98.0%) and sodium
trifluoromethanesulfonate (TfONa; >95%) were purchased
from Wako Pure Chemical Industries Co., Ltd., and used as
received. Other reagents and solvents were purchased from
Wako Pure Chemical Industries Co., Ltd., and used as received.
Monomer Synthesis
1,4-Di-O-Methyl-2,3,5,6-Tetra-O-Trimethylsilyl-
myo-Inositol (2a)
To a solution of tetraol 1a (2.08 g, 10.00 mmol) in DMF
(100 mL), TEA (16.0 mL), and TMSCl (7.0 mL, 55 mmol) were
added successively and the resulting solution was stirred at
25 ꢀC for 24 h. To the solution, hexane (300 mL) and water
(500 mL) were added and separated. The organic layer was
dried over sodium sulfate, filtered, and concentrated under
reduced pressure. The resulting solid was recrystallized from
methanol (30 mL) to obtain 2a (4.62 g, 9.30 mmol, 93%) as a
2
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2019

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ORIGINAL ARTICLE
colorless solid: mp 79.5–80.0 ꢀC; 1H NMR (CDCl3, δ): 3.93
(br t, 1H), 3.77 (t, J = 10.7), 3.45 (s, 3H), 3.30–3.20 (m, 3H),
2.70 (dd, J = 9.3, 2.0, 1H), 0.16 (s, 9H), 0.14 (s, 9H), 0.12 (s,
9H), 0.10 (s, 9H); 13C NMR (CDCl3, δ): 83.48, 82.45, 77.18,
76.86, 74.35, 74.11, 71.18, 61.49, 57.08, 1.12, 1.05, 0.60, 0.43;
high-resolution mass spectrometry (HRMS) (electrospray ioni-
zation [ESI] m/z): [M + H]+ calcd for C20H49O6Si4, 497.2606;
found, 497.2611.
2,5-Di-O-Methyl-1,3,4,6-Tetra-O-Trimethylsilyl-myo-
Inositol (3a)
To a solution of tetraol 6a (1.04 g, 5.00 mmol) in DMF
(50 mL), trimethylamine (8.0 mL) and trimethylsilylchloride
(4.0 mL, 32 mmol) were successively added and the resulting
solution was stirred at 25 ꢀC for 24 h. To the solution, hexane
(300 mL) and water (500 mL) were added and separated. The
organic layer was dried over sodium sulfate, filtered, and con-
centrated under reduced pressure. The resulting solid was
recrystallized from methanol (15 mL) to obtain 3a (2.26 g,
4.55 mmol, 91%) as a colorless solid: mp 82.0–83.0 ꢀC; 1H
NMR (CDCl3, δ): 3.72 (t, J = 9.3, 2H), 3.55 (s, 3H), 3.47 (s, 3H),
3.37 (t, J = 2.4, 1H), 3.32 (dd, J = 9.3, 2.4 Hz), 2.71 (t, J = 9.3,
1H), 0.14 (s, 36H); 13C NMR (CDCl3, δ): 86.46, 83.80, 74.20,
73.81, 61.88, 61.40, 0.81, 0.30; HRMS (ESI, m/z): [M + H]+
calcd for C20H49O6Si4, 497.2606; found, 497.2598.
1,4-Di-O-Phenylmethyl-2,3,5,6-Tetra-O-Trimethylsilyl-myo-
Inositol (2b)
According to the procedure for the silylation of tetraol 1a into
2a, tetraol 1b (3.60 g, 10.0 mmol) was silylated. The resulting
crude 2b was distilled under vacuum using a Kugelrohr appa-
ratus to obtain purified 2b (4.88 g, 7.52 mmol, 75%) as a
white solid: mp 73.5–74.0 ꢀC; 1H NMR (CDCl3, δ): 7.54–7.33
(m, 10H), 5.01 (d, J = 12.2, 1H), 4.89 (d, J = 12.2, 1H), 4.73
(s, 2H), 4.12 (t, J = 8.8, 1H), 4.06 (t, J = 1.0, 1H), 3.78
(t, J = 9.3, 1H), 3.53 (t, J = 8.8, 1H),3.52 (dd, J = 9.3, 2.5, 1H),
3.20 (dd, J = 9.3, 2.5, 1H), 0.28 (s, 9H), 0.27 (s, 9H), 0.24 (s,
9H), 0.15 (s, 9H); 13C NMR (CDCl3, δ): 139.75, 138.68,
128.36, 127.87, 127.59, 126.97, 126.76, 81.85, 80.45, 75.02,
74.27, 74.11, 72.81, 72.24, 1.29, 1.22, 0.84, 0.32; HRMS (ESI,
m/z): [M + H]+ calcd for C32H57O6Si4, 649.3232; found,
649.3219.
2,5-Di-O-Phenylmethyl-1,3,4,6-tetra-O-Trimethylsilyl-myo-
Inositol (3b)
According to the procedure given for the silylation of tetraol
6a into 3a, tetraol 6b (1.80 g, 5.00 mmol) was silylated. The
resulting crude 3b was distilled under reduced pressure using
a
Kugelrohr apparatus to obtain purified 6b (4.84 g,
7.47 mmol, 75%) as a colorless oil, which solidified upon
cooling: mp 79.5–80.0 ꢀC; 1H NMR (CDCl3, δ): 7.50–7.10 (m,
10H), 4.80 (s, 2H), 4.77 (s, 2H), 3.94 (t, J = 9.3, 2H), 3.62 (t,
J = 2.2, 1H), 3.37 (dd, J = 9.3, 2.2, 2H), 3.04 (t, J = 9.3, 1H),
0.08 (s, 18H), 0.01 (s, 18H); 13C NMR (CDCl3, δ): 139.93,
139.65, 128.23, 127.92, 127.53, 127.31, 126.60, 126.47, 84.90,
82.59, 75.32, 74.97, 74.50, 74.00, 1.05, 0.46; HRMS (ESI, m/z):
[M + H]+ calcd for C32H57O6Si4, 649.3232; found, 649.3219.
2,5-Di-O-(4-tert-butyl)Phenylmethyl-myo-Inositol (6c)
Sodium hydride (60% dispersion in mineral oil; 2.00 g,
50.0 mmol) was placed in a flask and washed with anhydrous
hexane (10 mL) three times under argon. Then, DMF
(120 mL) and 4 (4.08 g, 10.0 mmol) were added successively.
The reaction mixture was stirred at 25 ꢀC for 1 h, and then
cooled to 0 ꢀC. To the mixture, (4-tert-butyl)phenylmethyl bro-
mide (7.50 g, 33.0 mmol) was added in drops. The mixture
was allowed to warm to 25 ꢀC and stirred for 24 h. Water
(10 mL) was added slowly, and the mixture was then diluted
with ethyl acetate (200 mL), transferred into a separation fun-
nel, and washed with water (50 mL) three times. The organic
layer was dried over sodium sulfate, filtered, and concentrated
under reduced pressure to obtain crude 1,6:3,4-Bis-[O-
(2,3-dimethoxybutane-2,3-diyl)]-2,5-di-O-(4-tert-butylphenyl)
2,5-Di-O-(4-tert-butyl)Phenylmethyl-1,3,4,6-Tetra-O-
Trimethylsilyl-myo-Inositol (3c)
According to the procedure given for the silylation of tetraol
6a into 3a, tetraol 6b (0.94 g, 2.00 mmol) was silylated. The
resulting crude 3c was recrystallized from a mixture of meth-
anol (60 mL) and ethyl acetate (10 mL) to obtain purified 3c
(0.86 g, 1.56 mmol) as a white powder: mp 136.5–137.0 ꢀC;
1H NMR (CDCl3, δ): 7.42 (br s, 4H), 7.34 (d, J = 8.4, 2H), 7.28
(d, J = 8.4, 2H), 4.83 (s, 4H), 4.01 (t, J = 9.1, 2H), 3.69 (br s,
1H), 3.43 (dd, J = 2.2, 9.1, 2H), 3.11 (t, J = 9.1, 1H), 1.37 (s,
9H), 1.34 (s, 9H), 0.16 (s, 18H), 0.09 (s, 18H); 13C NMR
(CDCl3, δ): 150.22, 149.47, 127.52, 126.64, 125.15, 124.78,
85.06, 82.21, 75.53, 74.73, 74.47, 74.03, 34.65, 34.55, 31.57,
1.13, 0.47; HRMS (ESI, m/z): [M + H]+ calcd for C40H73O6Si4,
761.4484; found, 761.4457.
1
methyl-myo-inositol (5c): mp 138.0–138.5 ꢀC; H NMR (CDCl3,
δ): 7.44 (d, J = 8.3, 2H), 7.33 (s, 4H), 7.32 (d, J = 8.3, 2H), 4.83
(s, 2H), 4.81 (s, 2H), 4.18 (dd, J = 9.3, 9.8, 2H), 3.80 (t, J = 2.4,
1H), 3.57 (dd, J = 2.4, 9.8, 2H), 3.53 (t, J = 9.3, 1H), 3.27 (s,
6H), 3.25 (s, 6H), 1.33 (s, 6H), 1.32 (s, 9H), 1.31 (s, 6H), 1.30
(s, 9H).
Crude 5c was dissolved in 1,4-dioxane (100 mL). To this solu-
tion, 35% hydrochloric acid (10 mL) was added, and the
resulting mixture was stirred at 25 ꢀC for 24 h. The volatiles
were removed under reduced pressure to obtain a solid. This
solid was recrystallized from a mixture of methanol (40 mL)
and ethyl acetate (8 mL) to obtain 6c (3.51 g, 7.44 mmol, 74%
from 4) as a white powder: mp 218.5–219.0 ꢀC; 1H NMR
(CDCl3, δ): 7.38–7.24 (br m, 8H), 4.78 (d, J = 4.9, 2H), 4.73 (d,
J = 2.9, 4H), 4.71 (d, J = 4.9, 2H), 3.71 (br t, 1H), 3.60–3.52 (m,
2H), 3.34–3.28 (br m, 2H), 3.01 (t, J = 9.0, 1H), 1.26 (s, 18H).
Polycondensations
Typical Procedure (Polycondensation of 2a with CHD)
To a solution of 2a (0.498 g, 1.00 mmol) and CHD (0.112 g,
1.00 mmol) in 1,2-dichloroethane (1.0 mL) cooled to 0 ꢀC,
TMSOTf (9.0 μL, 11 mg, 0.050 mmol) was added. The mixture
was stirred at 0 ꢀC for 24 h. To the resulting gel, a solution of
sodium hydroxide (1.0 g) in methanol (50 mL) was added and
the mixture was stirred vigorously. The resulting precipitate
was collected by filtration with suction and dried under vac-
uum to obtain 7a3 (0.264 g, 93%) as a white powder.
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2019
3

JOURNAL OF
POLYMER SCIENCE
ORIGINAL ARTICLE
WWW.POLYMERCHEMISTRY.ORG
Polycondensation of 3b with CHD
OR
OR
According to the typical procedure, polycondensation of 3b
(0.650 g, 1.00 mmol) with CHD (0.112 g, 1.00 mmol) was con-
ducted to obtain 8b (0.431 g, 99%) as a white powder.
Me3SiO
Me3SiO
OSiMe3
OSiMe3
HO
HO
OH
OH
Me3SiCl
NEt3, DMF
OR
OR
Polycondensation of 3c with CHD
According to the typical procedure, polycondensation of 3c
(0.762 g, 1.00 mmol) with CHD (0.112 g, 1.00 mmol) was con-
ducted to obtain 8c (0.427 g, 78%) as a white solid.
1a
1b
: R = Me
: R = Bn
2a
2b
: R = Me
: R = Bn
SCHEME 2 Silylation of tetraols 1.
RESULTS AND DISCUSSION
Monomer Synthesis
OH
Me
Me
The tetraols 1a and 1b were converted into 2a and 2b
(Scheme 2) by following a standard O-silylation protocol. Silyl
ethers 3, which are isomers of 2, were synthesized from diol
4 (Scheme 3).15 The two hydroxyl groups of 4 were alkylated
to obtain diethers 5, and then their 1,2-diacetal protections
were removed under acidic conditions to obtain the
corresponding tetraols 6. Finally, 6 were silylated to obtain 3.
MeO
MeO
O
O
OMe
O
O
OMe
Me
Me
OH
4
OR
Me
Me
OMe
MeO
MeO
O
O
O
1) NaH, DMF, rt
2)
RX, rt, 24 h
O
OMe
Me
Me
OR
5a
(R = Me)
Polycondensation of 2 with CHD
5b (R = PhCH2)
5c
OR
Polycondensation of 2a with CHD was performed (Scheme 4).
The conditions and the results are summarized in Table 1,
along with the results of the polycondensations of tetraols
1 with CHD and the properties of the resulting OSKs 7a1 and
7b1, which have been reported in our previous paper,11 and
shown for comparison (Entries 1 and 2). For the polyconden-
sation of 1 with CHD, dimethylsulfoxide (DMSO) was used
because tetraols 1 were soluble only in such highly polar sol-
vents. Due to the high miscibility of DMSO with water, com-
plete removal of water formed by the polycondensation from
DMSO was difficult. In addition, the high temperature also
prompted the backward reaction, that is, hydrolysis of the
ketal functions, leading to the low molecular weights of 1 rang-
ing from 1400 to 2100 (estimated by SEC).
t
HCl aq
MeOH
(R = 4- BuC6H4CH2)
HO
HO
OH
6a
(R = Me)
rt, 24 h
6b
6c
(R = PhCH2)
(R = 4- BuC6H4CH2)
OH
t
OR
OR
Me3SiO
Me3SiO
OSiMe3
OSiMe3
3a (R = Me)
3b
3c
Me3SiCl
(R = PhCH2)
(R = 4-tBuC6H4CH2)
DMF
NEt3
OR
SCHEME 3 Synthesis of silyl ethers 3.
Polycondensation of 2b with CHD
According to the typical procedure, polycondensation of 2a
(0.650 g, 1.00 mmol) with CHD (0.112 g, 1.00 mmol) was con-
ducted to obtain 7b2 (0.415 g, 95%) as a white powder.
In contrast, the polycondensation of silyl ethers 2 with CHD
occurred efficiently at 0 ꢀC. 1,2-Dichloroethane, a much less
polar solvent than DMSO could be employed because it dis-
solved both the monomers and the resulting OSKs 7. In entry
3, the polycondensation of 2a and CHD was conducted using
1 mol % of TMSOTf. The initial concentrations of 2a and CHD
were set to be 1.0 M. The resulting OSK 7a2 was isolated as a
Polycondensation of 3a with CHD
According to the typical procedure, polycondensation of 3a
(0.498 g, 1.00 mmol) with CHD (0.112 g, 1.00 mmol) was con-
ducted to obtain 8a (0.281 g, 99%) as a white powder.
OR
OR
1
cf) from tetraol
7a
7b
Me3SiO
Me3SiO
OSiMe3
OSiMe3
O
O
O
O
1 (R = Me)
1 (R = PhCH2)
O
O
n
2
:
from
OR
OR
or
2
7a
7a
2 or 3 (R = Me)
2 (R = PhCH2)
CHD (1.0 eq)
or
7b
TMSOTf
ClCH2CH2Cl, 0 oC
OR
OR
Me3SiO
Me3SiO
OSiMe3
OSiMe3
O
O
O
O
from 3:
[2 or 3]0=1.0 M
8a
8b
8c
(R = Me)
(R = PhCH2)
(R = 4- BuC6H4CH2)
n
t
OR
OR
3
SCHEME 4 Polycondensation of silyl ethers 2 and 3 with CHD.
4
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2019

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ORIGINAL ARTICLE
TABLE 1 Synthesis of OSKs and their Properties
Amount of TMSOTf
(mol %)
Time
(h)
Yield
M
n (Mw/Mn) by
M
n (Mw/Mn)
Td5
Td10
Entry Monomer
OSK (%)a
MALDI-TOF MSb
by SECc
(ꢀC)d
(ꢀC)d
1
2
3
4
5
6
7
8
1a
1b
2a
2a
2b
3a
3b
3c
–
–
1
5
5
5
5
5
48
48
24
24
24
24
24
24
7a1
7b1
7a2
7a3
7b2
8a
–
2720 (1.4)
1710 (1.5)
3630 (1.3)
3750 (1.3)
2680 (1.6)
2100 (1.9)
1400 (1.7)
328
307
341
324
–
e
e
e
36
93
95
99
99
78
–
–
–
5600 (2.9)
4500 (1.5)
347
338
277
271
302
364
350
295
293
328
f
g
–
–
f
g
8b
–
–
f
8c
–
4100 (2.3)
a Methanol-insoluble parts.
d Determined by thermogravimetric analysis.
e Not measured.
b Measured in the m/z region 1000–10,000.
c Eluent: DMF containing 1 wt
standards.
%
LiBr; calibrated with polystyrene
f Suitable conditions for efficient ionization were not found.
g Not measurable due to the poor solubility.
methanol-insoluble fraction in 36%. To evaluate the efficiency
of chain growth in the polycondensation, 7a2 was analyzed by
MALDI.
series of signals, confirming the presence of polymer chains
with different combinations of terminal structures. The signals
in Series A, B, and C were attributed to (a) OSKs bearing a diol
and ketone moiety on each of the chain ends, (b) OSKs bearing
ketone moieties on both chain ends, and (c) OSKs bearing diol
moieties on both chain ends, respectively. The signals within
TOF mass spectrometry (MS) (Fig. 1; the peak list is shown in
Table S1). Compared to 7a1 obtained by the polycondensation
of tetraol 1a with CHD, 7a2 exhibited much higher molecular
weights, confirming that the use of the silylated monomer 2a
improved the efficiency of the chain growth.
The MALDI-TOF technique allowed detailed study of the termi-
nal structures of 7a2. In Figure 2, the spectrum of 7a2 focusing
on the region from m/z 4000–6000 is shown. There are three
FIGURE
2 Expanded MALDI-TOF MA of 7a2 and peak
FIGURE 1 MALDI-TOF mass spectra of 7a1 and 7a2.
assignments.
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2019
5

JOURNAL OF
POLYMER SCIENCE
ORIGINAL ARTICLE
WWW.POLYMERCHEMISTRY.ORG
each series differed by 284 Da, which corresponded to the
molecular weight of the repeating unit of 7a2.
MALDI-TOF mass analysis of 7b2 confirmed the successful for-
mation of the OSK structures (Figs. S3 and S4, Table S2). The
spectrum showed three major series of peaks (Series A, B, and
C), confirming the presence of polymer chains with different
combinations of terminal structures.
For the 13C-NMR analysis of 7a2, solid-state spectroscopy with
cross-polarization magic angle spinning (CPMAS) was used,
because the spectra measured in solution were rather compli-
cated, presumably due to the restricted conformational
changes of the main chain. Figure 3 shows the spectrum,
where signals attributable to the quaternary carbons α and β
in the spirocyclic structures were clearly observed.
The signals within each series differ by 436 Da, which cor-
responded to the molecular weight of the repeating unit of
7b2. In addition, there was a minor series of peaks (Series C0),
whose m/z values were smaller than those of Series C by
91 Da, corresponding to the replacement of one benzyl group
by one proton in one polymer chain during the ionization pro-
cess in MALDI. Solid-state 13C-NMR (CPMAS) analysis of 7b2
confirmed its structure. Figure S5 shows the spectrum, where
signals attributable to the aromatic carbons in the benzyl
groups and quaternary carbons α and β in the spirocyclic
structures were clearly observed.
Upon confirming the successful formation of OSK by the poly-
condensation of silyl ether 2a with CHD, the polycondensation
was carried out after increasing the amount of TMSOTf to
5 mol % (Entry 4). As a result, OSK 7a3 was obtained in 94%
yield. Its MALDI-TOF mass and solid-state 13C-NMR spectra
were virtually the same as those of 7a2 (Figs. S1 and S2). SEC
analysis of 7a3 revealed that the Mn and Mw values were
larger than those of OSK 7a1 by a factor >2.
Polycondensation of 3 with CHD
We then investigated the polycondensation of tetraol 6a
with CHD under the same conditions as those for the poly-
condensation of tetraol 1a with CHD to obtain OSK 7a1,
expecting that the resulting OSK 8a would have higher ste-
reoregularity than OSK 7a. However, the target OSK 8a was
not obtained, suggesting that bisketals of 6a are less stable
than the bisketals of 1a. The former has 1,6-trans and
3,4-trans configuration, which is intrinsically less stable than
the latter having 1,2-cis and 4,5-trans configuration, leading
to greater extent of hydrolysis under the conditions. This
indicated that the use of silylether 3 would enable its poly-
condensation with CHD under much milder conditions, suit-
able for the formation of less stable bisketals in the main
chain of 8a.
Under the conditions given in Entry 4, the polycondensation
of silylether 2b with CHD was performed (Entry 5). As a
result, the corresponding OSK 7b2 was obtained in 95% yield.
SEC analysis revealed that its Mn and Mw were larger than
those of OSK 7b1 obtained by the polycondensation of tetraol
1b and CHD, confirming that the use of silylated monomer
improved the efficiency of OSK synthesis.
The polycondensation of silylether 3a with CHD was per-
formed under the same conditions used for the polycondensa-
tion of 2a (Scheme 4). The resulting OSK 8a was insoluble in
various organic solvents. Similarly, the polycondensation of
3b was carried out to obtain OSK 8b, which was also insolu-
ble in various organic solvents.
NMR analysis of 8a and 8b in solution phase was not possible
due to their insolubility. Their MALDI-TOF MS analysis was
not successful due to the difficulty in ionization of these OSKs.
Thus, these OSKs were analyzed by solid-state 13C-NMR
(CPMAS) (Figs. 3 and S5). As can be expected from the higher
symmetry of the inositol-derived component in 8a than that
in 7a, a sharp signal attributable to the two equivalent quater-
nary carbons, α, knotting the cyclic moieties in a spirocyclic
manner was observed at 112 ppm, suggesting the successful
formation of spiroketal structure in the main chain. Similarly,
the spectrum of 8b showed a sharp signal at 112 ppm, con-
firming the main chain structure of 8b.
The major reason for the insolubility of 8a and 8b is likely
their tendency to form densely packed states allowed by their
highly rigid rod-like structures with higher symmetry than 7.
Figure S6 shows three-dimensional molecular structures of
model Compounds 9 and 10 (corresponding to the repeating
FIGURE 3 Solid-state 13C-NMR spectra (CPMAS) of 7a2 and 8a.
6
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2019

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ORIGINAL ARTICLE
units of 7 and 8, respectively) that were optimized by molecu-
lar mechanics calculation. In comparison to 9 with a rather
bended conformation, 10 had more laterally extended confor-
mation, suggesting that 8 would be more rod-like than 7.
However, another possibility, that is, the formation of
crosslinked structures by intermolecular ketalization could
not be excluded completely. In the solid-state 13C-NMR spec-
tra of 8a and 8b, a weak signal was detected at 100 ppm,
which could be attributed to the quaternary carbon in the acy-
clic ketal moiety linking oligomer chains.
combined through trans and trans configurations. This consid-
eration is in good agreement with the fact that the polycon-
densation of tetraol 3 with CHD was not successful.
OSK 8c was more thermally stable than 8a and 8b. The differ-
ence in the thermal stability could be correlated with the dif-
ference in their molecular weights: OSK 8c, whose main chain
was allowed to grow efficiently in the polymerization solution,
would have higher molecular weight than 8a and 8b, which
were less soluble and thus precipitated out from the polymer-
ization solutions before sufficient propagation.
In contrast to 8a and 8b, OSK 8c obtained by the polyconden-
sation of 3c with CHD was soluble in chloroform and DMF,
presumably due to the presence of bulky and hydrophobic
tertiary butyl groups that hindered aggregation of the chains.
The solubility of 8c enabled its SEC analysis to estimate its
molecular weight, yielding Mn and Mw/Mn of 4100 and 2.3,
respectively, suggesting efficient chain growth in the polycon-
densation. It was possible to measure 1H-NMR spectrum in
CDCl3 solution (Fig. S7). Although the signals were quite
broad, the ratio of the integrated signal intensities ([aromatic
protons]: [inositol-derived cyclohexane ring + methylene of
benzyl group]: [t-butyl group]) agreed with the theoretical
value. The solid-state 13C-NMR spectrum (CPMAS) provided
more detailed information about the structure of 8c (Fig. S8).
Similar to the spectra of 8a and 8b, that of 8c indicated a sig-
nal at 112 ppm that could be attributed to the quaternary car-
bons α. Although there was a weak signal at 100 ppm to
suggest the presence of acyclic ketal, its amount would be too
small to render 8c insoluble. The ratio in signal intensity
between these two signals was virtually same as those in 8a
and 8b, suggesting that the probability of intermolecular
ketalization resulting in a spirocyclic structure in the main
chain and intermolecular ketalization leading to a crosslinking
structure was not influenced by the bulkiness of the alkyl sub-
stituent (Me or Bn or 4-t-Bu-benzyl). This observation indi-
cates that the insoluble nature of 8a and 8b is likely caused
by their packing ability.
The OSKs were also subjected to DSC analyses. OSKs 7a3, 7b2,
8a, and 8b did not exhibit glass transition, while 8c revealed
a glass transition temperature of 113 ꢀC (Fig. S10).
CONCLUSIONS
Derivation of myo-inositol into its silyl ethers and their poly-
condensation with CHD gave an efficient and versatile route
for the synthesis of OSKs with higher molecular weights than
those obtained by the conventional polycondensation of
tetraols with CHD. The use of the silyl ethers 2 and 3,
possessing different configurations, allowed us to synthesize
two series of OSKs, 7 and 8, exhibiting different properties.
OSKs 7 were soluble in organic solvents, while OSKs 8 (except
8c) were insoluble, presumably because of the efficient pack-
ing of the more symmetric and thus more rigid rod-like
chains. The difference in stereochemistry between 7 and 8 also
influenced their thermal stabilities; the OSKs 7 formed by the
ketalization of silyl ether 2 bearing 1,2-cis and 4,5-trans con-
figuration were more stable than OSKs 8 formed by the
ketalization of silyl ether 3 bearing 1,6-trans and 3,4-trans
configuration.
ACKNOWLEDGMENT
This work was supported by a Kakenhi grant (Grant-in-Aid for
Scientific Research [C] 24550262) from the Japan Society for
the Promotion of Science.
Thermal Properties of OSKs
REFERENCES AND NOTES
The obtained OSKs, 7a3, 7b2, 8a, 8b, and 8c were subjected to
TG analyses. The resulting thermograms are shown in
Figure S9. The temperatures for 5% weight loss (Td5s) and
10% weight loss (Td10s) are summarized in Table 1.
1 W. J. Bailey, A. A. Volpe, J. Polym. Sci. Part A: Polym. Chem.
1970, 8, 2109.
2
W. J. Bailey, C. F. Beam, E. D. Cappuccilli, I. Haddad,
A. A. Volpe, In Cyclopolymerization and Polymers with Chain-
Ring Structures. ACS Symposium Series; G. Butler, J. E. Kresta,
Eds., American Chemical Society: Washington, DC, 1982; Vol.
195, p. 391.
The Td5 and Td10 values of 7a3, the product of the polycon-
densation of silylether 3a with CHD, were much higher than
those of 7a1, the product of the polycondensation of tetraol
2a with CHD, confirming the enhancement of the thermal sta-
bility due to the increase of molecular weight of 7a. Similarly,
7b2, having higher molecular weight than 7b1, was more ther-
mally stable.
3 S. Makhseed, N. B. McKeown, Chem. Commun. 1999, 255.
4 P. Wessig, K. Mollnitz, J. Org. Chem. 2012, 77, 3907.
5 H. B. Sonmez, F. G. Kuloglu, K. Karadag, F. Wudl, Polym. J.
2012, 44, 217.
6 G. Akbulut, H. B. Sonmez, F. Wudl, J. Polym. Res. 2013, 20, 97.
The Td5s of OSKs 8a and 8b were lower than 300 ꢀC con-
firming that they are more thermally degradable than 7a and
7b. The higher degradability of 8 suggested that the 5–6-5
fused ring system in 8 combined through cis and trans config-
urations would be more distorted than that that in 7, which is
7
D. T. Vodak, M. Braun, L. Iordanidis, J. Ple, L. Beck,
J. C. H. Spence, M. O’Keeffe, O. M. Yaghi, J. Am. Chem. Soc.
2002, 124, 4942.
8 Y. Mori, A. Sudo, T. Endo, J. Polym. Sci. Part A: Polym. Chem.
2019, 57, 792.
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2019
7

JOURNAL OF
POLYMER SCIENCE
ORIGINAL ARTICLE
WWW.POLYMERCHEMISTRY.ORG
9 T. Hino, T. Endo, J. Polym. Sci. Part A: Polym. Chem. 2005, 43,
13 R. H. Michell, Nat. Rev. Mol. Cell Biol. 2008, 9, 151.
5323.
14 K. M. Sureshan, M. S. Shashidhar, T. Praveen, T. Das, Chem.
Rev. 2003, 103, 4477.
10 Y. Mori, A. Sudo, T. Endo, J. Polym. Sci. Part A: Polym.
Chem. 2019, 57, 1564. https://doi.org/10.1002/pola.29420.
15 A. E. Koumbis, C. D. Duarte, C. Nicolau, J. M. Lehn, Chem.
Med. Chem. 2011, 6, 169.
11 A. Sudo, T. Sano, M. Harada, D. Ishida, ACS Macro Lett.
2014, 3, 808.
16 T. Tsunoda, M. Suzuki, R. Noyori, Tetrahedron Lett. 1980, 21,
12 R. Valluru, W. Van den Ende, Plant Sci. 2011, 181, 387.
1357.
8
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2019