Synthesis of poly(diphenylacetylene) membranes by desilylation of various precursor polymers and their properties

Article abstract of DOI:10.1021/ma047632g

Diphenylacetylenes having various silyl groups (PhC≡CC ₆H₄-SiR¹R²R³, SiR ¹R²R³ = p-SiMe₃, p-SiMe ₂i-Pr, p-SiEt₃, p-SiMe₂n-C₈H ₁₇, p-SiMe₂n-C₁₈H₃₇, m-SiMe ₃, m-SiMe₂t-Bu, la-g, respectively) were polymerized with TaCl₅-n-Bu₄Sn catalyst to provide high molecular weight polymers. The formed polymers (2a-g) afforded tough free-standing membranes by casting from toluene solution. Desilylation of these Si-containing polymer membranes was carried out with trifluoroacetic acid. Para-substituted polymers (2a-e) underwent desilylation completely, while meta-subustituted polymers (2f and 2g) did not. According to thermogravimetric analysis (TGA), both Si-containing and desilylated polymers exhibited high thermal stability, and all the desilylated polymers showed practically the same onset temperature (～480°C) of weight loss. The membranes of Si-containing polymers showed lower oxygen permeability as the silyl group became bulkier. The oxygen permeability coefficients (P(O₂) 120-3300 barrers) of desilylated polymer membranes were quite different from each other irrespective of the same polymer structure. When bulkier silyl groups were removed, the oxygen permeability increased to larger extents. In ethanol/water pervaporation, both Si-containing and desilylated polymer membranes showed ethanol permselectivity, indicating that these membranes possess large excess free volumes.

Full text of DOI:10.1021/ma047632g

2704
Macromolecules 2005, 38, 2704-2709
Synthesis of Poly(diphenylacetylene) Membranes by Desilylation of
Various Precursor Polymers and Their Properties
Toshikazu Sakaguchi,† Kenichi Yumoto,† Masashi Shiotsuki,† Fumio Sanda,†
Masakazu Yoshikawa,‡ and Toshio Masuda*,†
Department of Polymer Chemistry, Graduate School of Engineering, Kyoto University,
Katsura Campus, Kyoto 615-8510, Japan, and Department of Polymer Science and Engineering,
Kyoto Institute of Technology, Kyoto 606-8585, Japan
Received November 17, 2004; Revised Manuscript Received January 12, 2005
ABSTRACT: Diphenylacetylenes having various silyl groups (PhC≡CC6H4-SiR1R2R3, SiR1R2R3 ) p-SiMe3,
p-SiMe2i-Pr, p-SiEt3, p-SiMe2n-C8H17, p-SiMe2n-C18H37, m-SiMe3, m-SiMe2t-Bu, 1a-g, respectively) were
polymerized with TaCl5-n-Bu4Sn catalyst to provide high molecular weight polymers. The formed
polymers (2a-g) afforded tough free-standing membranes by casting from toluene solution. Desilylation
of these Si-containing polymer membranes was carried out with trifluoroacetic acid. Para-substituted
polymers (2a-e) underwent desilylation completely, while meta-subustituted polymers (2f and 2g) did
not. According to thermogravimetric analysis (TGA), both Si-containing and desilylated polymers exhibited
high thermal stability, and all the desilylated polymers showed practically the same onset temperature
(∼480 °C) of weight loss. The membranes of Si-containing polymers showed lower oxygen permeability
as the silyl group became bulkier. The oxygen permeability coefficients (P(O2) 120-3300 barrers) of
desilylated polymer membranes were quite different from each other irrespective of the same polymer
structure. When bulkier silyl groups were removed, the oxygen permeability increased to larger extents.
In ethanol/water pervaporation, both Si-containing and desilylated polymer membranes showed ethanol
permselectivity, indicating that these membranes possess large excess free volumes.
Introduction
presumably affect its gas permeability. The size of
formed microvoids may be controlled by selecting the
size of silyl group of the precursor polymer. In terms of
the desilylation of membranes, however, the effects of
silyl groups on the gas permeability and the reactivity
of desilylation are not well-known yet.
In this research, diphenylacetylenes containing vari-
ous silyl groups besides trimethylsilyl group (1a-g,
Scheme 1) were polymerized, and free-standing mem-
branes of resultant polymers (2a-e) were fabricated,
which yielded the insoluble membranes of poly(diphe-
nylacetylene)s (3a-e) upon desilylation. General prop-
erties and gas permeability of the polymer membranes
(2 and 3) were examined, and the variation of gas
permeability with desilylation was also investigated.
Polymerization of substituted acetylenes with transi-
tion metal catalysts affords substituted polyacetylenes,
which have various substituents and stiff main chain
composed of alternating double bonds. Substituted
polyacetylenes exhibit interesting features such as
chromism, semiconductivity, paramagnetism, gas per-
meability, etc.1 Among substituted acetylenes, diphe-
nylacetylene polymerizes in high yield with the TaCl5-
n-Bu4Sn catalyst.2 The formed polymer is thermally
very stable [onset temperature of weight loss (T0) in air
is 500 °C], but insoluble in any solvent. On the other
hand, 1-(p-trimethylsilyl)phenyl-2-phenylacetylene (1a,
Scheme 1) provides a polymer (2a) soluble in common
organic solvents.3 2a has high molecular weight [weight-
average molecular weight (Mw) > 1 × 106] and good
thermal stability [T0 ) 420 °C], and affords a tough free-
standing membrane by the casting method. Its oxygen
permeability coefficient (P(O2)) reaches 1500 barrers,
indicating that this polymer is highly gas-permeable
among various polymers.4
Results and Discussion
Polymerization. The polymerization of several novel
monomers with silyl groups of various sizes (1a-g) was
carried out in toluene using TaCl5-n-Bu4Sn as catalyst
(Table 1). It is well-known that TaCl5-n-Bu4Sn achieves
a good yield of disubstituted acetylene polymers with
high molecular weight,1a,b which is essential for fabrica-
tion of free-standing membranes. The polymerization
of monomer 1a produced a polymer in good yield, whose
Mw was as high as 2.1 × 106. Monomers 1b-d also
polymerized in a similar way to give polymers in high
yields of 80-95%. The polymers from 1c and 1d pos-
sessed very high molecular weights over Mw > 6 × 106.
The polymerization of 1e was conducted with lowering
the monomer concentration to 0.10 M owing to the high
molecular weight of the monomer, and the formed
polymer was isolated by precipitation from acetone
because the monomer was hardly soluble in methanol.
The Mw of the resultant polymer was as high as 3.0 ×
106. Thus, it can be said that the para-substituted
diphenylacetylenes afford high molecular weight poly-
In a preliminary communication,5 we reported that
the desilylation of the 2a membrane yields insoluble
poly(diphenylacetylene) membrane which otherwise
cannot be prepared because of its insolubility. By
performing the same procedure, membranes of insoluble
poly(diarylacetylene)s having naphthyl or phenanthryl
groups can also be prepared.6 Being insoluble, these
membranes can be used for the separation of organic
mixtures as well as gas separation. Because of a solid-
state reaction, the desilylation reaction seems to gener-
ate microvoids in polymer matrix whose size will
* Corresponding author. E-mail: masuda@adv.polym.kyoto-
u.ac.jp.
† Kyoto University.
‡ Kyoto Institute of Technology.
10.1021/ma047632g CCC: $30.25 © 2005 American Chemical Society
Published on Web 03/03/2005

Macromolecules, Vol. 38, No. 7, 2005
Synthesis of Poly(diphenylacetylene) Membranes 2705
Scheme 1. Preparation of Various Poly(diphenylacetylene) Membranes via Si-Containing
Poly(diphenylacetylene)s
Table 1. Polymerization of 1a-g with TaCl5-n-Bu4Sna
Their desilylation was carried out in a hexane/trifluo-
polymerb
roacetic acid (1:1 volume ratio) mixture at room tem-
perature for 24 h. Desilylation of the membranes of
para-substituted polymers 2a-e proceeded to comple-
tion to give membranes of poly(diphenylacetylene)s 3a-
e. On the other hand, desilylation of the membranes of
2f and 2g, i.e., meta-substituted polymers, did not reach
completion under the same reaction conditions as for
the para-substituted counterparts, and hence it was
impossible to prepare the membranes of poly(dipheny-
lacetylene) from these polymers. The comparison of IR
spectra and weights of membranes before and after
desilylation indicated that 68% of the silyl groups of 2f
membrane was removed under the above-stated condi-
tions. The completely desilylated polymers 3a-e were
insoluble in any solvent (Table 2), indicating that the
incorporation of silyl groups is essential for the solubility
of poly(diphenylacetylene).
c
monomer
[M]0
yield (%)
Mw x 10-6 c
Mw/Mn
1a
1b
1c
1d
1e
1f
0.50
0.50
0.50
0.50
0.10
0.50
0.50
61
80
95
89
69d
96
21
2.1
5.3
>6.0
>6.0
3.0
5.7
0.39
3.8
1.8
3.6
2.5
14
1g
a In toluene at 80 °C for 24 h; [TaCl5] ) 20 mM, [n-Bu4Sn] )
40 mM. b Methanol-insoluble product. c Measured by GPC cali-
brated with polystyrenes as standards. d Acetone-insoluble prod-
uct.
Table 2. Solubility of the Polymersa
solvent
hexane
2a
2b
2c
2d
2e
2f
2 g
3a-e
-
+
+
+
+
-
-
-
-
-
+
+
+
+
-
-
-
-
-
+
+
+
+
-
-
-
-
+
+
+
+
+
-
-
-
-
+
+
+
+
+
-
-
-
-
-
+
+
+
+
-
-
-
-
+
+
+
+
+
-
-
-
-
-
-
-
-
-
-
-
-
-
cyclohexane
toluene
CHCl3
Thermal Stability of the Polymers. The thermal
stability of polymers 2a-g and 3a-e was examined by
TGA in air (Figures 1 and 2). The onset temperature of
THF
ethanol
methanol
DMF
DMSO
a Symbols: +, soluble; -, insoluble.
mers in high yields irrespective of bulkiness of para-
substituents. The polymerization of monomer 1f with
trimethylsilyl group at meta-position also yielded a high
molecular weight polymer. On the other hand, in the
case of monomer 1g, which has a bulkier silyl group at
meta position, the polymer yield was as low as 21% and
the Mw was no more than 3.9 × 105, which is 1 order of
magnitude lower than those of other polymers. It is
known that diphenylacetylene with a trimethylsilyl
group at the ortho position of a phenyl ring does not
undergo polymerization.7 Thus, it is not unreasonable
that diphenylacetylenes with a bulky meta substituent
such as 1g does not smoothly polymerize owing to the
steric reason.
Figure 1. TGA curves of silyl group-containing poly(diphe-
nylacetylene)s 2a-g (in air, heating rate 10 °C min-1).
Desilylation of Polymer Membranes. The solubil-
ity of polymers 2a-g is summarized in Table 2. The
present silyl group-containing poly(diphenylacetylene)s
showed good solubility in many low polarity solvents
including toluene, chloroform, tetrahydrofuran, and
cyclohexane. In addition, polymers 2d and 2e, which
have long alkyl chains, were completely soluble even in
hexane. On the other hand, none of the polymers 2a-g
dissolved in polar solvents such as N,N-dimethylforma-
mide, dimethyl sulfoxide, and lower alcohols such as
methanol and ethanol.
Figure 2. TGA curves of poly(diphenylacetylene)s 3a-e
obtained by desilylation (in air, heating rate 10 °C min-1).
Free-standing membranes could be fabricated by
casting polymers 2a-g from toluene solution. The
membranes were sufficiently tough and transparent.
weight loss (T0) for polymer 2a was 420 °C, showing its
high thermal stability. The T0 values of polymers 2b
and 2c were 350 and 300 °C, respectively, while those

2706 Sakaguchi et al.
Macromolecules, Vol. 38, No. 7, 2005
Table 3. Oxygen Permeability (P(O2)) and Oxygen/
Nitrogen Separation Factor (P(O2)/P(N2)) of the Polymer
Membranes
Table 4. P(O2) and P(O2)/P(N2) of 3a after Conditioning
with Various Solvents
solvent
P(O2) (barrer)a
P(O2)/P(N2)a
2
3
methanol
ethanol
1-propanol
2-butanol
acetone
THF
910
1100
610
760
420
420
530
510
2.2
1.6
2.2
1.6
2.6
1.8
2.1
2.0
polymer
P(O2)a
P(O2)/P(N2)
P(O2)a
P(O2)/P(N2)
a
b
c
d
e
f
1500
500
550
28
2.2
2.3
2.1
3.2
3.8
1.7
2.3
910
2100
2500
3300
120
2.2
1.6
1.5
1.5
2.8
toluene
hexane
20
1700
1000
a Measured at 25 °C after membrane 3a was immersed in each
solvent and then dried to constant weight at room temperature.
g
a At 25 °C in the units of 1 × 10-10 cm3 (STP) cm/(cm2 s cmHg)
() 1 barrer).
barrers, respectively, which are, interestingly, much
larger than those of 2b and 2c. This obvious increase
of permeability is considered to be due to the formation
of large microvoids by removal of bulky silyl groups. The
P(O2) values of membranes 2d and 2e, which have long
alkyl chains, were no more than 28 and 20 barrers,
respectively, which is compatible with previous results
that polyacetylenes with long alkyl groups show rela-
tively low gas permeability.13 In sharp contrast to 2d,
the P(O2) of membrane 3d reached up to 3300 barrers.
This seems due to the generation of large microvoids
by removal of bulky silyl group. On the other hand, the
P(O2) value of membrane 3e was 120 barrers, where the
increase of gas permeability upon desilylation is not very
large. Membrane 2e considerably swelled due to high
affinity to the desilylation solvent mixture, and conse-
quently it shrank pretty much during desilylation. This
should have more or less depressed the formation of
microvoids.
The P(O2) values of membranes 2f and 2g with silyl
groups at meta position were 1700 and 1000 barrers,
respectively, which are slightly higher than those of
para-substituted analogues (2a, 1500 barrers; 2b, 500
barrers; 2 with p-SiMe2-t-Bu is insoluble14). Unfortu-
nately, the oxygen permeability of their desilylated
membranes (3f and 3g) could not be measured because
their desilylation did not proceed completely. The
oxygen/nitrogen separation factors (P(O2)/P(N2)) of all
the polymer membranes 2 and 3 were in the range 1.5-
3.8, and the values tended to decrease as the P(O2)
increased. These results agree with the general ten-
dency of gas permeation through polymer membranes.15
Effect of Swelling Solvents on the Oxygen Per-
meability. As mentioned above, the prepared polymer
membranes were subsequently immersed in methanol,
and dried to constant weight at room temperature prior
to permeability measurements. If measured without this
operation, the permeability generally assumes lower
values, because, when immersed in methanol, polymer
membranes swell to eventually have larger excess free
volumes upon drying. Thus, the effect of various con-
ditioning agents on the oxygen permeability of mem-
brane 3a was investigated (Table 4). When 3a was
conditioned with ethanol, the P(O2) value was 1100
barrers and slightly larger than that with methanol.
When 1-propanol and 2-butanol were used, the P(O2)
values were somewhat smaller, and 610 and 760 bar-
rers, respectively. The polymer chains seem to become
mobile due to high affinity to these alcohols, and the
swollen membranes should shrink during the drying
process. The P(O2) values after treatment with acetone,
THF, toluene, and hexane which have relatively high
affinity to 3a were all 420-530 barrers and fairly small.
Thus, we can say that rather low affinity solvents such
of 2d and 2e, which have long alkyl chains, were as low
as around 260 °C. The values of 2f and 2g were 420
and 380 °C, respectively. These results imply that the
thermal degradation of these polymers occurs at the silyl
moiety at first and that the thermal stability decreases
as the alkyl chain on the silyl groups becomes longer.
It is interesting to compare the values of these polymers
with the much lower T0 (∼200 °C) of para-substituted
poly(phenylacetylenes).8 When 2a-g were heated in air
above 800 °C, the ash composed of silica remained,
whose amount agreed with the expected value in every
case.
The desilylated polymers 3a-e exhibit practically
identical TGA curves to each other, and their T0 values
are all approximately 480 °C. This temperature is
higher than those of the starting Si-containing poly-
mers, which means that the thermal stability improves
upon desilylation. This finding is consistent with the
idea that the thermal decomposition of silyl group-
containing polymers takes place at the silyl group at
first. The practically identical TGA curves of the desi-
lylated polymers show that the thermal stability de-
pends on the polymer structure obtained after desily-
lation and is independent of that before desilylation.
Oxygen Permeability of the Polymers. Polymer
membranes 2a-g and 3a-e were immersed in metha-
nol for 24 h and dried to constant weight at room
temperature, and then their permeability was measured
at 25 °C (Table 3). The P(O2) value of membrane 2a at
25 °C was 1500 barrers, while that of 3a (the product
of desilylation of membrane 2a) was 910 barrers, which
is somewhat smaller than that of 2a. Though we
reported before5 that the P(O2) of 3a reaches 6000
barrers, we were unable to reproduce this high perme-
ability thereafter. Freeman also determined the P(O2)
of 3a to be 850 barrers.9 It is, anyhow, noteworthy that
poly(diphenylacetylene) 3a exhibits high oxygen perme-
ability up to ca. 1000 barrers although it has no
spherical groups. By comparison, poly(phenylacetylene)
is much less permeable to oxygen (P(O2) ) 6.0 barrers10).
This means that microvoids are generated in 3a at the
position where silyl groups were present in 2a and that
the microvoids are retained due to the highly stiff
polymer structure.
The P(O2) values of the membranes 2b and 2c, which
have bulky silyl groups, were smaller than that of 2a,
and 500 and 550 barrers, respectively. A tendency has
been known that, as the spherical substituent of poly-
acetylenes becomes bulkier, the gas permeability de-
creases.11 This has been explained in terms of the local
mobility of substituents.12 The present finding agrees
with this tendency. The P(O2) values of the desilylated
polymer membranes 3b and 3c were 2100 and 2500

Macromolecules, Vol. 38, No. 7, 2005
Synthesis of Poly(diphenylacetylene) Membranes 2707
Table 6. Pervaporation of EtOH/H2O Mixtures through
the Polymer Membranesa
b
c
d
polymer
XEtOH
YEtOH
REtOH/H O
103R/(g m m-2 h-1
)
2
2a
2b
2c
2d
2e
2f
2g
3a
3b
3c
3d
3e
0.0985
0.0994
0.119
0.0994
0.0978
0.0941
0.0970
0.101
0.0972
0.0971
0.0982
0.0989
0.419
0.354
0.371
0.193
0.290
0.347
0.481
0.392
0.467
0.485
0.464
0.339
6.60
4.97
4.38
2.14
2.86
5.10
8.64
5.74
8.12
8.76
7.95
4.54
3.28
1.27
1.39
2.32
1.93
5.11
2.08
3.18
6.45
8.45
4.27
2.71
Figure 3. Effect of aging time on the oxygen permeability of
polymers 2a, 3a, 2f, and 3g (stored in air at 25 °C).
a At 30 °C; downstream pressure 267 Pa (2 mmHg). b Weight
fraction of ethanol in the feed. c Weight fraction of ethanol in the
d
Table 5. Gas Diffusion and Solubility Coefficients (D and
S)a
permeate. REtOH/H O ) (YEtOH/YH O)/(XEtOH/XH O).
2
2
2
and 3a for various gases. Among the P values for the
five gases examined, the P(CO2) was the largest, while
the P(N2) was the smallest in both 2a and 3a. The gas
permeation through membrane can be inspected by
dividing into the solution and diffusion terms theoreti-
cally with rubbery polymers and approximately with
glassy polymers. In general, the S value increases with
increasing critical temperature of gases, while the D
value decreases with increasing molecular diameter of
gases in any polymers. Both 2a and 3a also display this
tendency.
When the gas solubilities for 2a with 3a are com-
pared, most of the S values of 3a are larger than those
of 2a (Table 5). The gas solubility in polymer membrane
is considered to increase with increasing free volume
in polymer membrane,17 and hence the increase of gas
solubility after desilylation suggests the increase of free
volume in polymer membrane. On the other hand, the
D values of 3a were smaller than those of 2a for most
gases. This can be accounted for by the large local
mobility of trimethylsilyl group. The local mobility of
substituents plays an important role for gas diffusivity;
i.e., the gas diffusivity becomes large when the polymer
has substituents with large local mobility such as
trimethylsilyl group.12 To summarize, the desilylation
of 2a membrane results in larger gas solubility but
smaller diffusivity, and consequently, the gas perme-
ability either remains about the same or becomes
somewhat smaller.
H2
O2
N2
CO2
CH4
2a
3a
Pb
3100
9.9
1700
9.0
900
3.8
190
800
10
6900
58
2800
25
103Sc
107Dd
Pb
320
2700
8.1
240
1300
12
120
5300
87
110
2200
38
103Sc
107Dd
340
110
75
62
57
a Measured at 25 °C by the “time lag” method. b In the unit of
barrer (1 barrer ) 1 × 10-10 cm3 (STP) cm cm-2 s-1 cmHg-1). c In
the units of cm3 (STP) cm-3 cmHg-1 d In the units of cm2 s-1
. .
as methanol and ethanol are useful as agents for
conditioning substituted polyacetylene membranes to
achieve high gas permeability. In the case of high
affinity solvents, substituted polyacetylene membranes
shrink during drying, resulting in the decrease of excess
free volume and, in turn, low gas permeability.
Variation of Oxygen Permeability with Aging
Time. The variation of oxygen permeability with aging
time was examined by storing the membranes at room
temperature in air (Figure 3). The P(O2) value of
membrane 2a decreased from 1500 to 1150 barrers in
15 days and further to 1000 barrers after 30 days and
then practically leveled off. Similarly, the P(O2) values
of membranes 3a, 2f, and 2g decreased with time to
about a half of their initial values. This variation with
aging time is accounted for by the reduction of excess
free volume, in other words, the relaxation of polymer
conformation. After 90 days, these membranes were
immersed in methanol for 24 h and dried to constant
weight; consequently, their P(O2) values were restored
to the initial ones owing to the recovery of the initial
unrelaxed polymer molecular structure. For instance,
though the P(O2) of membrane 2a was 950 barrers after
90 days, it increased to ca. 1450 barrers when the
membrane was immersed in methanol for 24 h and
dried to constant weight. This variation of the perme-
ability of polymer membranes with aging time is a
general tendency observed for highly gas-permeable
disubstituted acetylene polymers.6,16 Although poly(1-
trimethylsilyl-1-propyne) [poly(TMSP)] exhibits the high-
est gas permeability among all the polymers, its P(O2)
value remarkably decreases from 6000 to around 100
barrers.16a On the other hand, the present polymers and
other poly(diarylacetylene)s6 feature rather small time
dependence of P(O2), since their permeability only
decreases to a half of their initial values.
Pervaporation of the Ethanol/Water Mixture.
The pervaporation experiments of ethanol/water mix-
ture were performed using polymer membranes 2a-g
and 3a-e, whose results are shown in Table 6. The
separation factor REtOH/H O of membrane 2a was 6.60,
2
which is similar to that of poly[1-â-naphthyl-2-(p-
trimethylsilyl)phenylacetylene]18 (REtOH/H O ) 5.30). Al-
though this value is smaller than those of2poly(TMSP)19
and poly(dimethylsiloxane)20 (17.0 and 8.65, respec-
tively), it indicates ethanol permselectivity. Other Si-
containing polymer membranes 2b-g also permeated
ethanol preferentially. These results are interesting and
important because most polymers show water permse-
lectivity in ethanol/water pervaporation. For instance,
the values of separation factor RH O/EtOH values for
2
cellulose acetate, nylon, and polyacrylonitrile having
polar groups are fairly large (8.5-70),21 and even
polyethylene and polystyrene, which are hydrophobic
Solubility and Diffusivity of Gases in the Poly-
mer Membranes. Table 5 lists the permeability,
solubility, and diffusion coefficients (P, S, and D) of 2a
polymers, display water permselectivity, whose RH O/EtOH
2
values are roughly 1.7 and 12, respectively.22,23 This is
because the ethanol molecule is bulkier than the water

2708 Sakaguchi et al.
Scheme 2. Synthesis of Monomers 1a-g
Macromolecules, Vol. 38, No. 7, 2005
ether (50 mL) was added dropwise at 0 °C, and stirring was
continued further for 5 h at room temperature. After the
completion of the reaction was confirmed by TLC, ice-water
(50 mL) was added. The reaction mixture was washed with
water, and dried over anhydrous sodium sulfate. Ether was
evaporated, and the crude product was purified by silica gel
column chromatography (eluent: hexane) to give the desired
product (12 g, 56%) as colorless liquid. Purity >99% (1H NMR).
IR (KBr, cm-1): 2859, 1598, 1560, 1508, 1248, 1100, 995, 916,
824, 807, 760. 1H NMR (CDCl3, ppm): 7.54-7.46 (m, 6H, Ar),
7.34-7.32 (m, 3H, Ar), 0.95 (d, 6H, SiCHCH3), 0.94 (m, 1H,
SiCH), 0.25 (s, 6H, SiCH3). 13C NMR (CDCl3, ppm): 139.2,
133.8, 131.6, 130.5, 128.3, 128.2, 123.4, 123.3, 89.8, 89.5, 17.5,
13.8, -5.4. Anal. Calcd for C19H22Si: C, 81.9; H, 8.0; Si, 10.1.
Found: C, 82.0; H, 7.8; Si, 10.2.
1-Phenyl-2-(p-triethylsilyl)phenylacetylene (1c). This
monomer was prepared by the same method as for 1b using
triethylchlorosilane instead of dimethylisopropylchlorosilane
to give a colorless liquid, yield 65%, purity >99% (1H NMR).
IR (KBr, cm-1): 2874, 1599, 1560, 1507, 1099, 1011, 913, 821,
755. 1H NMR (CDCl3, ppm): 7.57-7.48 (m, 6H, Ar), 7.38-
7.30 (m, 3H, Ar), 0.96 (t, 9H, SiCH2CH3), 0.80 (q, 6H, SiCH2).
13C NMR (CDCl3, ppm): 138.1, 134.1, 133.0, 131.6, 130.6,
128.3, 128.2, 123.4, 89.7, 89.6, 7.3, 3.2. Anal. Calcd for C20H24-
Si: C, 82.1; H, 8.3; Si, 9.6. Found: C, 81.7; H, 7.5; Si, 10.8.
1-Phenyl-2-(p-dimethyloctylsilyl)phenylacetylene (1d).
The identical procedure to that of 1b was applied except using
dimethyloctylchlorosilane in place of dimethylisopropylchlo-
rosilane as a colorless liquid, yield 28%, purity >99% (1H
NMR). IR (KBr, cm-1): 2854, 1654, 1560, 1508, 1248, 1101,
821, 754. 1H NMR (CDCl3, ppm): 7.54-7.46 (m, 6H, Ar), 7.36-
7.32 (m, 3H, Ar), 1.29-1.23 (m, 12H, SiCH2(CH2)6), 0.87 (t,
3H, SiCH2(CH2)6CH3), 0.74 (t, 2H, SiCH2), 0.26 (s, 6H, SiCH3).
13C NMR (CDCl3, ppm): 140.3, 133.4, 131.6, 130.6, 128.3,
128.2, 123.5, 123.3, 89.7, 89.6, 33.9, 31.9, 29.3, 23.8, 22.7, 15.6,
14.1, -3.1, -3.4. Anal. Calcd for C24H32Si: C, 82.7; H, 9.3; Si,
8.0. Found: C, 82.7; H, 9.4; Si, 7.9.
molecule and hence usually difficult to permeate through
polymer membranes. The high ethanol permselectivity
of the present polymers is attributable to the presence
of microvoids therein as well as their hydrophobicity.
In particular, the former seems to be operative in
ethanol permselectivity. The permeation rate (R) is also
considerably increased by the presence of microvoids,
and hence highly ethanol-permselective membranes
show large fluxes. It is further noted that the ethanol
permselectivity and permeation rate of polymer mem-
branes 3b-e are larger than those of the corresponding
Si-containing polymer membranes. This indicates that
a large number of microvoids are generated upon
desilylation.
Experimental Section
General Data. The molecular weights of polymers were
estimated by gel permeation chromatography (CHCl3 as elu-
ent, polystyrene calibration). IR spectra were recorded on a
Shimadzu FTIR-8100 spectrophotometer. NMR spectra were
observed on a JEOL EX-400 spectrometer. Thermogravimetric
analysis (TGA) was conducted in air with a Perkin-Elmer
TGA7 thermal analyzer.
Gas permeability coefficients of polymer membranes were
measured with a Rikaseiki K-315-N gas permeability ap-
paratus at 25 °C. The D values were determined by the time
lag method using the following equation:
1-Phenyl-2-(p-dimethyloctadecylsilyl)phenylacetyl-
ene (1e). This monomer was also prepared by the same
method as for 1b using dimethyloctadecylchlorosilane instead
of dimethylisopropylchlorosilane as a white solid, yield 78%,
mp 35.0-36.0 °C, purity >99% (1H NMR). IR (KBr, cm-1):
D ) l2/6θ
1
2850, 1599, 1560, 1508, 1255, 1101, 850, 819, 757. H NMR
Here, l is the membrane thickness and θ is the time lag, which
is given by the intercept of the asymptotic line of the time-
pressure curve to the time axis. The S values were calculated
by using equation S ) P/D.
(CDCl3, ppm): 7.56-7.50 (m, 6H, Ar), 7.36-7.34 (m, 3H, Ar),
1.31 (m, 32H, SiCH2(CH2)16), 0.90 (t, 3H, SiCH2(CH2)16CH3),
0.75 (t, 2H, SiCH2), 0.27 (s, 6H, SiCH3). 13C NMR (CDCl3,
ppm): 140.4, 133.5, 131.6, 130.6, 128.3, 128.2, 123.5, 123.3,
89.7, 89.6, 33.6, 31.9, 29.92, 29.86, 29.83, 29.82, 29.78, 29.71,
29.68, 29.67, 29.6, 29.4, 29.3, 23.8, 22.7, 15.6, 14.1, -3.0, -3.1.
Anal. Calcd for C34H52Si: C, 83.5; H, 10.7; Si, 5.8. Found: C,
83.3; H, 10.8; Si, 5.9.
The permeation experiments of ethanol/water mixture were
performed by an ordinary pervaporation technique and perme-
ates were analyzed by a Shimadzu GC-8APT gas chromato-
graph equipped with a 3.0-m long column packed with
poly(ethylene glycol) 20 M [Chromosorb W (AW-DMCS)].
Materials. TaCl5 (Strem) as main catalyst was used without
further purification, while n-Bu4Sn (Wako, Japan) as cocata-
lyst was purified by distillation. Bromoiodobenzenes (Wako,
Japan) were used without further purification. Solvents for
polymerization were purified by the usual methods. Pheny-
lacetylene (Aldrich) and various alkylchlorosilanes (Tokyo
Kasei, Japan) were commercially obtained. Monomers were
synthesized according to Scheme 2 with reference to the
literature of ethynylation24 and silylation.25 1-Phenyl-2-(p-
bromo)phenylacetylene and 1-phenyl-2-(m-bromo)phenylacety-
lene were synthesized by the reaction of phenylacetylene with
p- and m-bromoiodobenzenes, respectively.3b Monomers 1a and
1f were prepared according to the literature procedure.3b The
synthesis and analytical data of other monomers are as follows.
1-Phenyl-2-(p-dimethylisopropylsilyl)phenylacetyl-
ene (1b). A 500 mL three-necked flask was equipped with a
dropping funnel, a three-way stopcock, and a magnetic stirring
bar. After the flask was flushed with nitrogen, 1-phenyl-2-(p-
bromo)phenylacetylene (20 g, 0.078 mol) and ether (250 mL)
were added and cooled at 0 °C. At the same temperature, a
hexane solution of n-butyllithium (58 mL, 1.6 M, 93 mmol)
was added dropwise, and the mixture was stirred for 1 h. Then,
a solution of dimethylisopropylchlorosilane (13 g, 95 mmol) in
1-Phenyl-2-(m-t-butyldimethylsilyl)phenylacetylene (1
g). This monomer was prepared by the same method as for
1b using tert-butyldimethylchlorosilane and 1-phenyl-2-(m-
bromo)phenylacetylene instead of dimethylisopropylchlorosi-
lane and 1-phenyl-2-(p-bromo)phenylacetylene as a colorless
liquid, yield 64%, purity >99% (1H NMR). IR (KBr, cm-1):
1
2853, 1598, 1560, 1508, 1491, 1250, 1112, 882, 832, 757. H
NMR (CDCl3, ppm): 7.68 (s, 1H, Ar), 7.60-7.54 (m, 3H, Ar),
7.49 (s, 1H, Ar), 7.37-7.32 (m, 4H, Ar), 0.94 (s, 9H, SiCCH3),
0.31 (s, 6H, SiCH3). 13C NMR (CDCl3, ppm): 138.0, 137.4,
134.1, 131.9, 131.5, 128.2, 128.1, 127.3, 123.3, 122.5, 89.8, 89.3,
26.4, 16.8, 6.3. Anal. Calcd for C20H24Si: C, 82.1; H, 8.3; Si,
9.6. Found: C, 82.6; H, 8.1; Si, 9.3.
Polymerization. Polymerizations were carried out in a
Schlenk tube, equipped with a three-way stopcock, under dry
nitrogen. Unless otherwise specified, the reaction was carried
out at 80 °C for 24 h under the following conditions: [M]0 )
0.50 M, [TaCl5] ) 20 mM, and [n-Bu4Sn] ) 40 mM. A detailed
procedure of polymerization is as follows: The monomer
solution was prepared in a Schlenk tube by mixing monomer
1a (1.3 g) and toluene (5.0 mL). Another Schlenk tube was
charged with TaCl5 (71 mg), n-Bu4Sn (0.13 mL), and toluene
(4.9 mL); this catalyst solution was aged at 80 °C for 10 min,
and then monomer solution was added to it. Polymerization

Macromolecules, Vol. 38, No. 7, 2005
Synthesis of Poly(diphenylacetylene) Membranes 2709
can Chemical Society: Washington, DC, 2004. (b) Nunes, S.
P.; Peinemann, K. V. Membrane Technology in the Chemical
Industry; Wiley: New York, 2001. (c) Koros, W. J.; Mahajan,
R. J. Membr. Sci. 2000, 175, 181. (d) Tsujita, Y. Prog. Polym.
Sci. 2003, 28, 1377. (e) Pandey, P.; Chauhan, R. S. Prog.
Polym. Sci. 2001, 26, 853.
was run at 80 °C for 24 h, which was quenched with a small
amount of methanol. The resulting polymer was isolated by
precipitation into a large excess of methanol, and its yield was
determined gravimetrically.
Membrane Fabrication and Desilylation. Membranes
(thickness ca. 20-50 µm) of polymers (2a-g) were fabricated
by casting their toluene solution (concentrated ca. 0.50-1.0
wt %) into a Petri dish. The plate was covered with a glass
vessel to slow solvent evaporation (ca. 3-5 days). After a
membrane formed, it was immersed in methanol for 24 h and
dried to constant weight at room temperature for 24 h. As
shown in the literature,5 the desilylation of the membranes of
2a-g was carried out using trifluoroacetic acid. A detailed
procedure is as follows: The polymer membrane of 2a was
immersed in a mixture of hexane/trifluoroacetic acid (1:1
volume ratio) at room temperature for 24 h. To remove the
remaining acid in the polymer matrix, the membrane was
immersed in a mixture of hexane/triethylamine at room
temperature for 24 h. Finally, it was immersed in methanol
for 24 h to remove residual impurities, washed with methanol,
and dried to constant weight at room temperature for 24 h.
The completion of desilylation was confirmed by IR spectra of
the polymer membranes before and after the reaction; the IR
spectra of the membrane of 3a exhibited no absorptions at ca.
1250, 1120, and 860 cm-1, which were seen in 2a. After
desilylation, the weight of the 3a membrane decreased to the
values anticipated for desilylation, which also confirmed the
completion of the reaction.
(5) Teraguchi, M.; Masuda, T. Macromolecules 2002, 35, 1149.
(6) (a) Sakaguchi, T.; Shiotsuki, M.; Masuda, T. Macromolecules,
2004, 37, 4104. (b) Sakaguchi, T.; Kwak, G.; Masuda, T.
Polymer 2002, 43, 3937.
(7) Kouzai, H.; Masuda, T. Unpublished data.
(8) Fujita, Y.; Misumi, Y.; Tabata, M.; Masuda, T. J. Polym. Sci.,
Part A: Polym. Chem. 1998, 36, 3157.
(9) Freeman, B. D. Personal communication.
(10) Aoki, T.; Nakahara, H.; Hayakawa, Y.; Kokai, M.; Oikawa,
E. J. Polym. Sci., Part A: Polym. Chem. 1994, 32, 849.
(11) (a) Nagai, K.; Toy, L. G.; Freeman, B. D.; Teraguchi, M.;
Masuda, T. Polym. Mater. Sci. Eng. 1999, 81, 531. (b) Aoki,
T.; Kobayashi, Y.; Kaneko, T.; Oikawa, E.; Yamamura, Y.;
Fujita, Y.; Teraguchi, M.; Nomura, R.; Masuda, T. Macro-
molecules, 1999, 32, 79. (c) Teraguchi, M.; Masuda, T. J.
Polym. Sci., Part A: Polym. Chem. 1999, 37, 4546.
(12) (a) Kanaya, T.; Teraguchi, M.; Masuda, T.; Kaji, K. Polymer
1999, 40, 7157. (b) Kanaya, T.; Tsukushi, I.; Kaji, K.;
Sakaguchi, T.; Kwak, G.; Masuda, T. Macromolecules 2002,
35, 5559.
(13) Masuda, T.; Iguchi, Y.; Tang, B. Z.; Higashimura, T. Polymer
1988, 29, 2041.
(14) Tsuchihara, K.; Masuda, T.; Higashimura, T. J. Polym. Sci.,
Part A: Polym. Chem. 1993, 31, 547.
(15) Robeson, L. M. J. Membr. Sci. 1991, 62, 165.
(16) (a) Shimomura, H.; Nakanishi, K.; Odani, H.; Kurata, M.;
Masuda, T.; Higashimura, T. Koubunshi Ronbunshu 1986,
43, 747. (b) Nagai, K.; Higuchi, A.; Nakagawa, T. J. Polym.
Sci., Part B: Polym. Phys. 1995, 33, 289. (c) Kwak, G.;
Masuda, T. J. Polym. Sci., Part A: Polym. Chem. 2000, 38,
2964.
(17) Hu, C. C.; Chang, C. S.; Ruaan, R. C.; Lai, J. Y. J. Membr.
Sci. 2003, 226, 51.
(18) Sakaguchi, T.; Yumoto, K.; Kwak, G.; Yoshikawa, M.; Ma-
suda, T. Polym. Bull. (Berlin) 2002, 48, 271.
(19) (a) Masuda, T.; Tang, B. Z.; Higashimura, T. Polym. J. 1986,
18, 565. (b) Ishihara, K.; Nagase, Y.; Matsui, K. Makromol.
Chem., Rapid Commun. 1986, 7, 43. (c) Masuda, T.; Takat-
suka, M.; Tang, B. Z.; Higashimura, T. J. Membr. Sci. 1990,
49, 69.
(20) Miyata, T.; Obata, S.; Uragami, T. Macromolecules 1999, 32,
3712.
Acknowledgment. The authors thank Dr. Giseop
Kwak at Nara Institute of Science and Technology for
his valuable suggestions. This research was supported
by a grant-in-aid for scientific research from the Min-
istry of Education, Science, Culture, and Sports (Japan).
References and Notes
(1) For reviews of substituted polyacetylenes, see: (a) Masuda,
T.; Sanda, F. In Handbook of Metathesis; Grubbs, R. H., Ed.;
Wiley-VCH: Weinheim, Vol. 3 Chapter 3. 11, 2003. (b)
Nomura, R.; Masuda, T. In Encyclopedia of Polymer Science
and Technology; Kroshwitz, J. I., Ed.; Wiley: New York, 2003;
Vol. IA, p 1. (c) Nagai, K.; Masuda, T.; Nakagawa, T.;
Freeman, B. D.; Pinnau, I. Prog. Polym. Sci. 2001, 26, 721.
(d) Choi, S. K.; Gal, Y. S.; Jin, S. H.; Kim, H. K. Chem. Rev.
2000, 100, 1645. (e) Tabata, M.; Sone, T.; Sadahiro, Y.
Macromol. Chem. Phys. 1999, 200, 265.
(21) (a) Nagy, E.; Borlai, O.; Ujhidy, A. J. Membr. Sci. 1980, 7,
109. (b) Mulder, M. H. V.; Oude, H. J.; Hegeman, H.;
Smolders, C. A. J. Membr. Sci. 1983, 16, 269.
(22) Eustache, H.; Histi, G. J. Membr. Sci. 1981, 8, 105.
(23) Uragami, T.; Morikawa, T. Makromol. Chem. 1989, 190, 399.
(24) (a) Sonogashira, K. J. Organomet. Chem. 2002, 653, 46. (b)
Takahashi, S.; Kuroyama, Y.; Sonogashira, K.; Hagihara, N.
Synthesis 1980, 627.
(2) Niki, A.; Masuda, T.; Higashimura, T. J. Polym. Sci., Part
A: Polym. Chem. 1987, 25, 1553.
(3) (a) Tsuchihara, K.; Masuda, T.; Higashimura, T. J. Am.
Chem. Soc. 1991, 113, 8548. (b) Tsuchihara, K.; Masuda, T.;
Higashimura, T. Macromolecules 1992, 25, 5816.
(25) Habich, D.; Effenverger, F. Synthesis 1979, 841.
(4) (a) Freeman, B. D.; Pinnau, I. Advanced Materials for
Membrane Separation; ACS Symposium Series 876; Ameri-
MA047632G