Facile synthesis of an amphiphilic 1,3,5-trisubstituted benzene as a novel surface modifier by selective photocyclic aromatization and efficient improvement of oxygen permselectivity by the addition of the surface modifier

Article abstract of DOI:10.1246/cl.130372

An amphiphilic 1,3,5-trisubstituted benzene as a new 2D surface modifier (T-EO) which has six hydroxy groups and three oligoethylene groups was successfully and easily synthesized by selective photocyclic aromatization in high conversion and selectivity f

Full text of DOI:10.1246/cl.130372

doi:10.1246/cl.130372
1090
Published on the web June 4, 2013
Facile Synthesis of an Amphiphilic 1,3,5-Trisubstituted Benzene as a Novel Surface Modifier
by Selective Photocyclic Aromatization and Efficient Improvement of Oxygen Permselectivity
by the Addition of the Surface Modifier
Jianjun Wang,1 Toshiki Aoki,*1 Lijia Liu,2 Takeshi Namikoshi,3 Masahiro Teraguchi,1 and Takashi Kaneko1
1Graduate School of Science and Technology, Niigata University, 2-8050 Ikarashi, Nishi-ku, Niigata 950-2181
2College of Materials Science and Chemical Engineering, Harbin Engineering University, Harbin 150001, P. R. China
3Material Science and Engineering, Kitami Institute of Technology, 165 Koen-cho, Kitami, Hokkaido 090-8507
(Received April 21, 2013; CL-130372; E-mail: toshaoki@eng.niigata-u.ac.jp)
An amphiphilic 1,3,5-trisubstituted benzene as a new 2D
surface modifier (T-EO) which has six hydroxy groups and three
oligoethylene groups was successfully and easily synthesized
by selective photocyclic aromatization in high conversion and
selectivity for the first time. We prepared four kinds of blend
membranes based on PVA or poly(trimethylsilylphenylacety-
lene) (poly(SPA)) containing the new amphiphilic cyclic trimer
(T-EO) or the corresponding polymer (P-EO), that is, T-EO/
PVA, P-EO/PVA, and T-EO/poly(SPA) membranes prepared
by conventional solvent casting method (Method I) and T-EO/
poly(SPA) membranes prepared by a new solvent casting
method (Method II). T-EO/PVA membranes by Method I and
T-EO/poly(SPA) membranes by Method II showed better
performance in oxygen permeation than P-EO/PVA membranes
by Method I and T-EO/poly(SPA) membranes by Method I,
respectively. A PVA-based membrane containing 1.0 wt % T-EO
showed the best performance, that is, the ¡ became twice
that of the original PVA membrane without any decrease in
hv
SCAT
H
n
P-EO
T-EO
Scheme 1. Selective photocyclic aromatization (SCAT) of
P-EO.
In this study, to improve effectiveness of the surface
modifiers, a new surface active compound, T-EO (Scheme 1)
was designed. T-EO is a 1,3,5-trisubstituted benzene derivative
which has triphenylbenzene as a hydrophobic planer (2D) core
and three ortho-substituted oligo(ethylene oxide) groups as
hydrophilic regions and in addition it has six hydroxy groups
which can form hydrogen bonds. If T-EO was used as a surface
modifier of hydrophilic polymer membranes, the hydrophilic
surface exposed by hydrophobic air can be covered by the
hydrophobic 2D part of T-EO, because the triphenylben-
zene portion tends to accumulate on the surface and the
oligo(ethylene oxide) groups can work as an anchor segment.
In addition, if the six hydroxy groups form hydrogen bonds
between the molecules, a planar i.e., 2D supramolecular surface
can be formed. However, since the molecular structure of T-EO
is complicated and has three kinds of functional groups, it is
difficult to obtain it in a high yield by a simple synthetic route.
For example, the direct cyclization from EO (Scheme S1)3 to
T-EO is difficult because the catalyst for the direct cyclization
reaction has low functional group tolerance.
PO . The ideal improvement may be caused by effective surface
2
modification by T-EO.
Permselective membranes separating gases such as oxygen
and nitrogen are very useful for many practical solutions to
environmental and energy problems. Hence, many studies have
been reported on such membranes from not only organic
polymers but also inorganic compounds.1 The requirements for
oxygen permselective membranes are (1) high permeability
coefficient (PO : cm3(STP) cm cm¹2 s¹1 cmHg¹1), (2) high sep-
2
aration factor (¡ ¼ PO =PN ), and (3) good membrane forming
2
2
ability (high mechanical strength). However, these three
characteristics have been unobtainable at the same time in
almost all membrane materials reported. For example, mem-
branes having higher PO had lower ¡ and membranes having
2
higher ¡ had lower PO , that is, PO and ¡ showed a tradeoff
2
2
relationship.1a Also membranes having higher PO tended to be
2
too flexible and membranes with higher ¡ tended to be too
brittle.
We found and reported recently a novel polymer reaction
called SCAT (highly selective photocyclic aromatization)
reaction.4 The SCAT reaction has many advantages.4a Among
the advantages, high conversion, high selectivity, and simple
procedure (only light irradiation) are very useful for synthesis of
T-EO and therefore we selected SCAT as the synthetic route. In
this communication, a facile synthesis of T-EO by the SCAT
reaction and enhancement of ¡ by surface modification by using
T-EO are reported.
To overcome the above problems, that is, to obtain materials
satisfying the three requirements simultaneously, we reported
surface modifications of conventional polymer membranes with
sufficient mechanical strength by solvent casting a mixture of
small amounts of surface active polymers and conventional base
polymers.2 It was an effective method to enhance ¡ with a small
decrease in PO without change to the good membrane forming
2
abilities of the base polymers. The enhancement of ¡, however,
was not sufficient.
T-EO was obtained quantitatively and easily by light
irradiation (SCAT) on P-EO without any purification. For the
Chem. Lett. 2013, 42, 1090-1092
© 2013 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/

1091
A
B
Table 1. Oxygen permeation and contact angles of four kinds
of blend membranes
5
4
3
2
1
3.4
3.0
T-EOa
Or
P-EO
T-EO/PVA
(Method I)
P-EO/PVA
(Method I)
5.0
10.0
10.0
No.
1.0
5.0
5.0
ªd
PO
2
¡
ªd
b
c
b
c
PO
¡
2
1.0
1
2
3
4
5
6
0
13.7
1.71
1.80
2.68
3.88
3.73
4.11
39.1
40.4
40.1
40.4
51.5
53.1
13.7 1.71
39.1
®
®
42.2
30.2
34.2
α
α
0.50
0.060 13.5
0.50
1.0
5.0
10
®
®
®
®
13.7
12.7
8.8
2.6
2.2
5.0
1.0
0.50
0.50
1.0
12.9 2.15
8.8 2.25
4.9 3.82
0.060
0
0
5.1
10
12
14
16
18
4
8
12
16
T-EO/poly(SPA)
T-EO/poly(SPA)
PO2 × 10-3/barrer
PO2 × 10/barrer
No. T-EOa
(Method I)
(Method II)
PO
¡
ªd
PO
¡
ªd
e
c
e
c
Figure 1. Relationship between the PO and ¡ of the blend
2
2
2
membranes (the numbers in the figure are the content of
7
8
9
0
171
150
136
120
2.36
2.45
2.55
2.63
106
105
107
104
171
142
128
111
2.36
2.58
2.88
3.22
¹1
106
100
95
87
c¡ ¼
additives (wt %)). A: : T-EO/PVA blend membranes;
:
0.50
1.0
5.0
P-EO/PVA blend membranes. B: : T-EO/poly(SPA) blend
membranes prepared by Method I; : T-EO/poly(SPA) blend
¹10
10
membranes prepared by Method II (1 barrer = 10
cm3(STP) cm cm¹2 s¹1 cmHg¹1).
aIn wt %. bIn 10¹13 cm3(STP) cm cm¹2 s¹1 cmHg
.
PO =PN
.
dContact angles of water droplets on the air
2
2
surface of the blend membranes (°). eIn 10
¹10
from 1.71 to 3.88 without any drop in PO when only 1.0 wt % of
2
cm3(STP) cm cm¹2 s¹1 cmHg
.
¹1
the cyclic trimer (T-EO) was added. On the other hand, by
adding the same amount (1.0 wt %) of the corresponding
polymer (P-EO) into the PVA membrane ( ), the permselectivity
showed almost no change. By increasing the content of T-EO
from 1.0 to 10 wt % in T-EO/PVA blend membranes, the
permselectivity had almost no change but the permeability
decreased markedly (Figure 1A and Table 1). In summary, the
PVA-based blend membrane containing 1.0 wt % of T-EO had
the best performance and the small addition of T-EO to PVA was
effective for enhancing the permselectivity of the PVA mem-
experimental details, procedures, and results of synthesis of the
monomer (EO), the polymer (PO), and the cyclic trimer (T-EO),
see the Supporting Information.3 T-EO was used as a 2D surface
modifier and three kinds of blend membranes containing T-EO
were prepared. The blend membranes were prepared by the
following two methods: Method I (conventional method): A
toluene solution of the polymer and 0.060-5.0 wt % of the
additives was cast on
a poly(tetrafluoroethylene) sheet.
Method II (new method): A toluene solution of poly(trimethyl-
silylphenylacetylene) (poly(SPA)) and a methanol solution
of T-EO was first blended together, and then the solution
branes without any drop in PO .
2
The T-EO (5.0 wt %)/poly(SPA) membranes ( ) prepared
by Method II we developed showed a higher permselectivity
compared with the ones ( ) having the same content of T-EO by
Method I (Figure 1B and Table 1). The permselectivities of the
membranes prepared by Method II increased more than those by
Method I although the decreases in permeability were almost the
same (Figure 1B and the lower lines of Table 1). In summary,
we measured oxygen permeation through the two kinds of
T-EO/poly(SPA) blend membranes prepared by the two
methods I and II. As a result, the membranes prepared by
Method II showed a better performance on enhancing the
permselectivity compared with the ones prepared by Method I.
In conclusion, by comparing these four kinds of membranes
above-mentioned, we conclude that the PVA-based membrane
containing 1.0 wt % T-EO on the surface was the best judging
from the performance and the degree of improvement.
was cast on
a poly(tetrafluoroethylene) sheet. Oxygen
and nitrogen permeability coefficients (PO
,
and PN :
2
2
cm3(STP) cm cm¹2 s¹1 cmHg¹1) were measured by gas chroma-
tography at 25 °C using YANACO GTR-10.
A 1,3,5-trisubstituted benzene (T-EO: Scheme 1) which
has six hydroxy groups and three oligoethylene groups was
synthesized successfully for the first time by a selective
photocyclic aromatization (SCAT) method we developed and
reported recently.4a In addition, the conversion and selectivity
were very high, and the procedure was pretty simple, that is,
only light irradiation to the corresponding polymer membrane.
Four kinds of blend membranes containing T-EO were prepared
by using the two methods. One was conventional method
(Method I) using one solvent which can solve the base polymer
and the additives, and the other was newly developed method
(Method II) using two solvents which were miscible but have
different solubilities for the base polymer and the additive (T-
EO) and different boiling points. Oxygen permeation behavior
(PO and ¡ ¼ PO =PN ) through the resulting four kinds of
To determine the reason for the effective improvement of
the ¡ values without large drops in PO described above, we
2
investigated the surfaces of the blend membranes by measuring
contact angles (ª) of water droplets. The results are shown in
Figure 2 and Table 1.
Judging from the ª values in Figure 2A and the upper part
of Table 1, Nos. 1-6, the surface of T-EO/PVA blend mem-
branes ( ) became more hydrophobic than the original pure
polymer membrane (PVA) but the surface of P-EO/PVA blend
2
2
2
blend membranes were measured. The results were shown in
Table 1 and Figure 1.
As shown in Table 1, Nos. 1 and 4 and Figure 1A, the ¡
values of T-EO/PVA blend membranes ( ) increased steeply
Chem. Lett. 2013, 42, 1090-1092
© 2013 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/

1092
A
B
hydrophilic than poly(SPA), T-EO could be accumulated at
the surface. As we described above, T-EO/poly(SPA) blend
membranes prepared by Method II showed better performances
of oxygen permselectivity than those by Method I (Figure 1B).
Therefore, the effective improvement and good performance in
oxygen permeation of T-EO/poly(SPA) blend membranes
prepared by Method II are thought to be caused by the effective
surface-modification.
80
100
120
80
60
40
20
0
3
6
0
4
8
12
T-EO or P-EO in membrane/wt%
T-EO in membrane/wt%
In conclusion, we reported the four kinds of blend
membranes based on PVA or poly(SPA) containing the new
amphiphilic cyclic trimer (T-EO), which was simply prepared in
a high yield by SCAT reaction, developed previously by us, of
the corresponding polymer (P-EO), that is, T-EO/PVA, P-EO/
PVA, and T-EO/poly(SPA) membranes prepared by conven-
tional solvent casting (Method I) and T-EO/poly(SPA) mem-
branes prepared by a new solvent casting method we developed
(Method II). T-EO/PVA membranes prepared by Method I and
T-EO/poly(SPA) membranes prepared by Method II showed
better performance in oxygen permeation than P-EO/PVA
membranes prepared by Method I and T-EO/poly(SPA) mem-
branes prepared by Method I, respectively. Especially the
1.0 wt % of T-EO containing PVA-based membranes showed
the best result, because the ¡ became twice that of the original
Figure 2. Plots of contact angles vs. the content of additives in
the blend membranes. A: : T-EO/PVA blend membranes;
P-EO/PVA blend membranes. B: : T-EO/poly(SPA) blend
membranes prepared by Method I; : T-EO/poly(SPA) blend
membranes prepared by Method II.
:
membranes ( ) did not become more hydrophobic. Therefore
the surface of T-EO/PVA blend membranes were more
effectively modified by the newly synthesized additives
(T-EO) than those of P-EO/PVA blend membranes. In addition,
as described above, T-EO/PVA blend membranes showed much
better oxygen permselectivity than P-EO/PVA blend membranes
(Figure 1A). Therefore, the effective improvement and good
performance of T-EO/PVA blend membranes are thought to be
caused by the effective surface-modification by T-EO.
PVA membrane without any decrease in PO . The ideal
2
improvement may be caused by effective surface modification
by T-EO. Since T-EO has a planer region in its molecule, it can
arrange in 2D manner on the surface. The 2D supramolecular
surface may improve the performance. Further investigation
about surface structures are now in progress.
In the case of addition of 5.0 wt % of T-EO or P-EO to PVA,
they showed almost the same PO but the former membrane
2
showed about twofold higher ¡ than the latter as described
above (Figure 1A) and simultaneously the ª value of the former
membrane was higher than that of the latter (Table 1, No. 5, and
Figure 2A). Therefore, the modified surface can be one of the
reasons for the effective enhancement of ¡. The difference
between T-EO and P-EO may be caused by the difference of
their molecular assembly structures on the surface. It may mean
T-EO is a good 2D surface modifier. Especially in the case of
addition of 1.0 wt % of T-EO to PVA, the blend membrane
showed about twice higher ¡ than the original pure PVA
This study was partly supported by a Grant-in-Aid for
Scientific Research (B) (No. 19350054) from the Japan Society
for the Promotion of Science, the National Science Foundation
of China (No. 21204014), and the Fundamental Research Funds
for the Central Universities, China (Nos. HEUCFR1227 and
HEUCFT1009).
membrane with almost no drop in PO (Figure 1A). It may also
2
References and Notes
be caused by effective surface modification although no clear
change in ª value was observed. It is thought to be that the
surface was modified very effectively and no alternation of the
inside or bulk structure of the blend membrane and as a result
1
a) J. Ahn, W.-J. Chung, I. Pinnau, J. Song, N. Du, G. P. Robertson,
M. D. Guiver, J. Membr. Sci. 2010, 346, 280. b) L. M. Robeson,
J. Membr. Sci. 2008, 320, 390. c) T. Aoki, Prog. Polym. Sci. 1999,
24, 951. d) T. Aoki, T. Kaneko, M. Teraguchi, Polymer 2006, 47,
4867. e) C. H. Jung, J. E. Lee, S. H. Han, H. B. Park, Y. M. Lee,
J. Membr. Sci. 2010, 350, 301. f) S. K. Sen, S. Banerjee, J. Membr.
Sci. 2010, 350, 53. g) S. Thomas, I. Pinnau, N. Du, M. D. Guiver,
J. Membr. Sci. 2009, 333, 125. h) J. Li, J. Wang, Y. Zang, T. Aoki, T.
Kaneko, M. Teraguchi, Chem. Lett. 2012, 41, 1462. i) T. Sato, N.
Yoshida, A. Ishida, M. Teraguchi, T. Aoki, T. Kaneko, Polymer
2013, 54, 2231. j) T. Kaneko, K. Yamamoto, M. Asano, M.
Teraguchi, T. Aoki, J. Membr. Sci. 2006, 278, 365.
only ¡ was increased and no drop in PO was observed.
2
Judging from the ª values in Figure 2B and the lower part
of Table 1, Nos. 7-10, the surface of T-EO/poly(SPA) blend
membranes prepared by Method II which was developed in this
study became more hydrophilic than the original pure polymer
membrane (poly(SPA)) but the surface of T-EO/poly(SPA)
blend membranes prepared by Method I (conventional method)
did not become more hydrophilic. Therefore, the surfaces of
T-EO/poly(SPA) blend membranes prepared by Method II were
more effectively modified by the newly synthesized additives
(T-EO) than those by Method I. It was unusual that the surfaces
were modified by more hydrophilic additives by Method II. We
have achieved such unusual modification by Method II. The
reason for the success is not clear at present but we speculate the
reason as follows: The point is the difference of the boiling
points (bp’s) of the two solvents used. Since the more hydro-
philic solvent (methanol) having a lower bp, which is evaporat-
ing first, has higher solubility for T-EO, which is more
2
a) Y. Kawakami, T. Aoki, Y. Yamashita, Polym. Bull. 1987, 17, 293.
b) Y. Kawakami, H. Karasawa, H. Kamiya, T. Aoki, Y. Yamashita,
Polym. J. 1986, 18, 237. c) Y. Kawakami, T. Aoki, Y. Yamashita, M.
Hirose, A. Ishitani, Macromolecules 1985, 18, 580.
3
4
Supporting Information is available electronically on the CSJ-
Journal Web site, http://www.csj.jp/journals/chem-lett/index.html.
a) L. Liu, T. Namikoshi, Y. Zang, T. Aoki, S. Hadano, Y. Abe, I.
Wasuzu, T. Tsutsuba, M. Teraguchi, T. Kaneko, J. Am. Chem. Soc.
2013, 135, 602. b) Y. Zang, K. Nakao, H. Yotsuyanagi, T. Aoki, T.
Namikoshi, T. Tsutsuba, M. Teraguchi, T. Kaneko, Polymer 2013,
54, 1729. c) L. Liu, Y. Zang, S. Hadano, T. Aoki, M. Teraguchi, T.
Kaneko, T. Namikoshi, Macromolecules 2010, 43, 9268. d) T. Aoki,
T. Kaneko, N. Maruyama, A. Sumi, M. Takahashi, T. Sato, M.
Teraguchi, J. Am. Chem. Soc. 2003, 125, 6346.
Chem. Lett. 2013, 42, 1090-1092
© 2013 The Chemical Society of Japan
www.csj.jp/journals/chem-lett/