- Email: [email protected]
Oxygen permselectivities through supramolecular polymer membranes prepared by highly selective photocyclic aromatization of poly(substituted acetylene)
Accepted Manuscript Oxygen permselectivities through supramolecular polymer membranes prepared by highly selective photocyclic aromatization of poly(substituted acetylene) Guanwu Yin, Eri Ohtaka, Toshiki Aoki, Jianjun Wang, Masahiro Teraguchi, Takashi Kaneko PII:
S0032-3861(17)30861-3
DOI:
10.1016/j.polymer.2017.08.066
Reference:
JPOL 19968
To appear in:
Polymer
Received Date: 6 August 2017 Revised Date:
30 August 2017
Accepted Date: 31 August 2017
Please cite this article as: Yin G, Ohtaka E, Aoki T, Wang J, Teraguchi M, Kaneko T, Oxygen permselectivities through supramolecular polymer membranes prepared by highly selective photocyclic aromatization of poly(substituted acetylene), Polymer (2017), doi: 10.1016/j.polymer.2017.08.066. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
ACCEPTED MANUSCRIPT
RI PT
Oxygen Permselectivities through Supramolecular Polymer Membranes Prepared by Highly Selective Photocyclic Aromatization of Poly(Substituted Acetylene)
1
SC
Guanwu Yin1, Eri Ohtaka1, Toshiki Aoki1*, Jianjun Wang,2Masahiro Teraguchi1, and Takashi Kaneko1 Department of Chemistry and Chemical Engineering; Graduate School of Science
and Technology, Niigata University, Ikarashi 2-8050, Nishi-Ku, Niigata 950-2181,
M AN U
Japan. College of Chemistry and Chemical Engineering, Qiqihar University, Wenhua Street 42, Qiqihar 161006, China
2
Corresponding author: Toshiki Aoki*
TE D
Address: Department of Chemistry and Chemical Engineering, Graduate School of Science and Technology, Niigata University, Ikarashi 2-8050, Niigata, 950-2181, Japan.
AC C
EP
Tel.:/fax: +81 25 262 7280; E-mail: [email protected]
1 | 14
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
RI PT
ACCEPTED MANUSCRIPT
Graphical Abstract
SC
Oxygen Permselectivities through Supramolecular Polymer Membranes Prepared by Highly Selective Photocyclic Aromatization of Poly(Substituted Acetylene)
M AN U
Guanwu Yin1, Eri Ohtaka1, Toshiki Aoki1*, Jianjun Wang,2Masahiro Teraguchi1, and Takashi Kaneko1
TE D
: S1
OH
HO
C12H25O
OC12H25
6 5
OH
Upper bound polts f rom the literature (2016)
OH
EP
4
OH
3
OC12H25
2
S1
1
AC C
HO
in
Robeson`s upper bounds(2008) : 1
10
10
2
3
10
10
4
5
10
OH
n
OC12H25
H
OH
OH
P O2 (barrer)
nanopore
n OC12H25 H
OH
2 | 14
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
ACCEPTED MANUSCRIPT
ABSTRACT: In order to estimate oxygen permselectivities through a supramolecular polymer membrane for the first time, we prepared three kinds of composite
RI PT
membranes consisting of a supramolecular polymer membrane and a conventional organic polymer membrane. Supramolecular polymer membranes were prepared by
SC
using highly selective photocyclic aromatization of poly(substituted acetylene)s. It was found that the supramolecular polymer membrane had oxygen permselectivity (α
M AN U
= PO2/PN2) around three in spite of its low molecular weight.
Keywords: Oxygen Permselectivities; Supramolecular Polymer Membranes; Highly
Introduction
TE D
Selective Photocyclic Aromatization
EP
Gas permselective membranes are very important and valuable because necessity to remove impurities and to purify mixtures of gases, whose molecular sizes are very
AC C
similar, is enhancing for solving recent environmental problems [1-7]. As materials for gas permselective membranes, organic polymers have been usually used because they have a good self-supporting membrane forming ability, which is based on their high molecular weights (MW) and good solubility in cast solvents. However, polymeric materials having both high MW and good solubility are not many, and therefore molecular design of polymers for gas separation membranes was limited. If
3 | 14
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
ACCEPTED MANUSCRIPT
organic materials having low MW can be applied to gas permselective membranes, variation of the molecular design will be enlarged for gas separation membranes. Although supramolecular polymers were reported as new materials prepared
RI PT
from low MW compounds, [8-10] their strength were very low and therefore there have been no reports on their application to gas separation membranes which need to resist pressure differences. Therefore no data on gas permeability of supramolecular
TE D
M AN U
SC
polymers have been reported so far.
AC C
EP
Figure 1. Preparation of self-supporting supramolecular polymer membranes by using highly seletive photocyclic aromatization (SCAT).
We have previously reported highly selective photocyclic aromatization (SCAT)
of cis-cisoidal (c-c) poly(substituted acetylene)s to yield quantitatively the corresponding cyclic trimer, that is, 1,3,5-tirisubstituted benzene (Figure 1). [11-17] In the SCAT reaction by irradiation of visible light, c-c poly(substituted acetylene) membranes were quantitatively converted to the corresponding cyclic trimers. In addition, the resulting benzene derivatives maintained their self-membrane forming 4 | 14
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
ACCEPTED MANUSCRIPT
abilities in spite of their low MW (Figure 1). They were one kind of supramolecular polymers formed on the basis of hydrogen bonds and π−π stackings. However, mechanical strengths of the supramolecular membranes were very low. Although we
RI PT
tried to enhance their membrane strengths, their strengths were not enough to use separation membranes. [16]
In this communication, in order to estimate oxygen permselectivities through
SC
supramolecular polymer membranes for the first time, we prepared three kinds of
M AN U
composite membranes consisting of a supramolecular polymer membrane and a conventional organic polymer membrane (Figure 2). Supramolecular polymer
AC C
EP
TE D
membranes were prepared by SCAT of poly (substituted acetylene)s.
5 | 14
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
Figure 2. Preparation of three kinds of composite membranes consisting of supramolecular polymer membranes (S1) and organic polymer membranes (P1 or P2) by using highly seletive photocyclic aromatization (SCAT). (I) A composite membrane by partial SCAT by photoirradiation on the one side of a SCAT-active polymer(P1) membrane, (II) a composite membrane by SCAT of a membrane of a copolymer (P(1/2)) consisting of a SCAT-active sequence (P1) and a SCAT-inactive sequence (P2), (III) a composite membrane containing S1 inside nanopores of P1.
6 | 14
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
ACCEPTED MANUSCRIPT
RI PT
Results and discussion Synthesis of acetylene monomers 1 and 2 and their (co)polymers P1 and P(1/2) were carried out according to our previous reports. [18-20] Self-supporting polymer
SC
membranes of P1 and P(1/2) used for SCAT reactions were fabricated by solvent cast
M AN U
method using THF as a solvent. SCAT reactions and their estimation were carried out according to our previous report. [11-17] Oxygen and nitrogen permeability coefficients (PO2 and PN2: barrer) were measured by gas chromatographic method according to our previous report. [21-22]
TE D
As mentioned above, supramolecular polymer membranes prepared by SCAT of P1 was very week. First, we prepared blend membranes of P1 and its SCAT product S1. [21-22] Even when the S1 contents[23] were very low (less than 10 wt%), the
EP
blend membranes had lost self-membrane forming abilities. Therefore, this membrane
AC C
preparation method was not suitable for estimation of permselectivity of S1. On the other hand, when we controlled the conversion of the SCAT reaction by changing irradiation time on P1 (Method I in Figure 2)[24], the resulting composite membranes containing 0-65wt% of S1 showed self-membrane forming abilities and their permeabilities could be measured. Therefore this method (I) was suitable for estimation of permselectivity of S1 by extrapolation[25]. As a result, α (= PO2/PN2) of S1 was estimated as 2.62 (Table 1 and Figure 3, no.1). It is not high but shows the 7 | 14
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
ACCEPTED MANUSCRIPT
supramolecular membrane may have an ability to separate oxygen and nitrogen gas molecules similar to usual polymer membrane P1 (α =2.82, no.7). In general for a
AC C
EP
TE D
M AN U
SC
RI PT
membrane to show oxygen permselectivity, the membrane should not have any
8 | 14
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
ACCEPTED MANUSCRIPT
defects including molecular size pores. Therefore, the finding that S1 showed oxygen
6 Upper bound polts from the literature (2016)
5
8
Ro be Ro son` be s u so pp n` er su b pp ou er nd bo (20 un 08 d( ) 19 15 91 ) 14
764 3 5 21
2 1
1
10
100 1000 Po (barrer)
4
10000
100000
M AN U
2
SC
13 9 10 11 12
3
RI PT
4
8
II 13 III 9 10 7 11
6 5
4
TE D
3
3 2
12
1
I
2
EP
0
50
100 Po 2(barrer)
150
AC C
Figure 3. Oxygen permeation behavior through composite membranes consisting of supramolecular polymer membranes (S1) and organic polymer membranes (P1 or P2). ■ (nos.1-7, I) composite membranes by partial SCAT by photoirradiation on the one side of a SCAT-active polymer(P1) membrane; ●(nos.9-12, II) composite membranes by SCAT of a membrane of a copolymer(P(1/2)) consisting of a SCAT-active sequence (P1) and a SCAT-inactive sequence (P2); ♦ (no.13,III) a composite membrane containing S1 inside nanopores of p-P1. The data for nos. 1 and 8 were extrapolated values and the others are
permselectivity may indicate S1 membrane is dense in spite of a low MW compound. However, since the composite membranes I may be an asymmetric membrane as shown in Figure 2 (I), they must form a large area of pure S1 membranes. Therefore, 9 | 14
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
EP
it may contain some defects. To decrease such possibility to form defects in S1 membranes, we prepared composite membranes II consisting of S1 and P2 (S1/P2)
AC C
by SCAT of the copolymer of 1 and 2 (P(1/2)) as shown in Figure 2 (II). Because poly(2) is not active for SCAT reaction[26], S1/P2 could be obtained from P(1/2)[27]. Although we expected the composite membranes II would have better self-membrane forming ability than membranes I because areas of S1 must be very small, their highest content of S1 for II (6.5wt%) was unexpectedly much lower than that for I (65wt%) as shown in Table 1 and Figure 3, nos. 2 and 9). Therefore this method (II) was not suitable for estimation of permselectivity of S1 by extrapolation[25]. 10 | 14
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
ACCEPTED MANUSCRIPT
However, in the series of membranes II, when the content of S1 increased, their α (= PO2/PN2) increased, which is different from membranes I. The extrapolation value of α was 3.75 (Table 1 and Figure 3, no.8). This means the supramolecular polymer
RI PT
membrane had a higher α than the corresponding conventional polymer membrane. The value may not a precise value but this interesting behavior is possible because S1 may have a higher density than P1. In fact, when the SCAT reaction, the starting P1
SC
membranes shrank[28]. It may be caused by enhancing π−π stacking. Therefore, S1
M AN U
can be denser than P1 and enhancing their α values, although S1 is a decomposed product of P1.
Although the two α values for S1 by extrapolation were determined above mentioned, they are not direct values. Therefore it may partly contain separation
TE D
ability based on P1 or P2. To estimate more reliable α values for S1 more directly, we prepared composite membranes III as shown in Figure 2 (III) [29], where we used a nanoporous membrane of P1 (p-P1). The p-P1 membrane was prepared from a dilute
EP
chloroform solution of P1. It showed very high PO2 and very low α (Table 1 and
AC C
Figure 3, no.14). Since the α value was less than 1.0, the membrane did not have separation ability for oxygen and nitrogen. Therefore, it is suitable to estimate separation ability of S1. We prepared a composite membrane of S1 and P1 (III), that is, p-P1 membrane including S1 in its nanopores. Surprisingly the composite membrane showed much higher α value (= 3.08) than pure p-P1. [30] The selectivity can be estimated as that caused originally by S1.
11 | 14
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
ACCEPTED MANUSCRIPT
In conclusion, judging from the three series of measurements, it was found that the α values for the supramolecular polymer membrane S1 was around 3. It is quite high and similar to those for conventional organic polymers. We estimated gas
RI PT
permselectivity through supramolecular polymer membranes successfully for the first time. We are now in progress to discuss molecular design for the supramolecular
SC
polymers to enhance their gas permselective performances.
M AN U
References and Notes
[1] Ahn J, Chung W J, Pinnau I, Song J, Du N, Robertson G P, Guiver MD. J. Membr. Sci. 2010; 346: 280.
[2] Robeson L M. J. Membr. Sci. 2008; 320: 390.
TE D
[3] Aoki T. Prog. Polym. Sci. 1999; 24: 951.
[4] Aoki T, Kaneko T, Teraguchi M. Polymer 2006; 47: 4867. [5] Jung C H, Lee J E, Han S H, Park H B, Lee Y M. J. Membr. Sci. 2010; 350: 301.
EP
[6] Sen S K, Banerjee S. J. Membr. Sci. 2010; 350: 53.
AC C
[7] Thomas S, Pinnau I, Du N, Guiver M D. J. Membr. Sci. 2009; 333: 125. [8] Folmer B J B, Sijbesma R P, Versteegen R M, Rijt J A J V, Meijer E W. Adv Mater 2000; 12: 874. [9] Yoshikawa I, Sawayama J, Araki K. Angew Chem Int Ed 2008; 47: 1038. [10] Yamada N, Komatsu T, Yoshinaga H, Yoshizawa K, Edo S, Kunitake M. Angew Chem Int Ed 2003; 42: 5496. [11] Liu L, Namikoshi T, Zang Y, Aoki T, Teraguchi M, Kaneko T. J. Am. Chem. Soc. 2013; 135: 602.
12 | 14
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
ACCEPTED MANUSCRIPT
[12] Zang Y, Aoki T, Teraguchi M, Kaneko T, Ma L, Jia H. Polym. Rev. 2015; 55: 57. [13] Zang Y, Yin G, Aoki T, Teraguchi, Kaneko T. Ma L, Jia H. Chirality 2015; 27: 459. [14] Liu L, Long Q, Aoki T, Namikoshi T, Abe Y, Miyata M, Teraguchi M, Kaneko T, Wang Y,
RI PT
Zhang C. Macromol. Chem. Phys. 2015; 216: 530. [15] Miyata M, Teraguchi M, Endo H, Kaneko T, Aoki T. Chem. Lett. 2014; 43: 1476.
Kaneko T. Polym. Commun. 2013; 54: 4431.
SC
[16] Miyata M, Namikoshi T, Liu L, Zang Y, Aoki T, Abe Y, Oniyama Y, Tsutsuba T, Teraguchi M,
M AN U
[17] Yin G, Liu L, Aoki T, Namikoshi T, Teraguchi M, Kaneko T. Chem. Lett. 2016; 45: 813. [18] Aoki T, Kaneko T, Maruyama N, Sumi A, Takahashi M, Sato T, Teraguchi M. J. Am. Chem. Soc. 2003; 125: 6346.
[19] Hadano S, Kishimoto T, Hattori T, Tanioka D, Teraguchi M, Aoki T, Kaneko T, Namikoshi T,
TE D
Marwanta E. Macromol. Chem. Phys. 2009; 210: 717.
[20] Liu L, Zang Y, Jia H, Aoki T, Hadano S, Teraguchi M, Miyata M, Zhang G, Namikoshi T. Polym. Rev. 2017; 57: 89.
EP
[21] Wang J, Zang Y, Yin G, Aoki T, Urita H, Taguwa K, Liu L, Namikoshi T, Teraguchi M, Kaneko
AC C
T, Ma L, Jia H. Polymer 2014; 55: 1384. [22] Wang J, Aoki T, Liu L, Namikoshi T, Teraguchi M, Kaneko T. Chem. Lett. 2013; 42: 1090. [23] The SCAT contents (=SCAT conversion) were easily and clearly calculated from the area ratio of the GPC peaks. For the details, see Figure S1. [24] The asymmetric composite membranes were prepared just by photoirradiation on one-sides of polymer membranes. For the detail SCAT conditions, see the SI. [25] For the details of the extrapolation, see Figure S2.
13 | 14
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
ACCEPTED MANUSCRIPT
[26] Since the SCAT conversions were always less than those of the composition of monomer unit 1, it was confirmed no SCAT happened in comonomer 2 sequence whose homopolymer poly(2) is SCAT inactive.
be a relatively block type not a random type.
RI PT
[27] Since the SCAT conversion of P(1/2) could reach up to the composition of S1, the copolymer may
[28] Judging from the shrink of the membrane after SCAT reaction (see Figure S3 in the SI), the
SC
density of the SCAT product is higher than that of the original polymer.
M AN U
[29] The composite membrane III was prepared by immersing the nanoporous p-P1 membrane in methanol solution of S1. We have recently found the supramolecular polymer membranes of S1 formed from methanol solution.
[30] To confirm no change of the base polymer p-P1 itself during preparation of the composite
TE D
membrane III, a controlled experiment was carried out. Almost no change in permeability was
AC C
EP
observed (Table 2 and Figure 3, no 15).
14 | 14
ACCEPTED MANUSCRIPT Highlights for Oxygen permselectivities through supramolecular polymer membranes prepared by highly selective photocyclic aromatization of poly(substituted acetylene) Corresponding author's name: Toshiki Aoki*
RI PT
1) We developed three new preparation methods for composite membranes consisting of a supramolecular polymer membrane and a conventional organic polymer membrane, which have enough strengths for application to gas separation membranes, by highly (SCAT). (Preparationa of supramolecular
SC
selective photocyclic aromatization
polymer membranes by the SCAT reaction have been previously reported by us in
M AN U
JACS 2013 but no application was reported).
2) We found the supramolecular polymer membranes had an ability to separate oxygen and nitrogen in permeation and the permselectivity was relatively high and similar to that for usual organic polymer membranes. To the best of our knowledge, this is the first experimental result on gas separation by supramolecular polymer
AC C
EP
TE D
membranes.
S0032-3861(17)30861-3
DOI:
10.1016/j.polymer.2017.08.066
Reference:
JPOL 19968
To appear in:
Polymer
Received Date: 6 August 2017 Revised Date:
30 August 2017
Accepted Date: 31 August 2017
Please cite this article as: Yin G, Ohtaka E, Aoki T, Wang J, Teraguchi M, Kaneko T, Oxygen permselectivities through supramolecular polymer membranes prepared by highly selective photocyclic aromatization of poly(substituted acetylene), Polymer (2017), doi: 10.1016/j.polymer.2017.08.066. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
ACCEPTED MANUSCRIPT
RI PT
Oxygen Permselectivities through Supramolecular Polymer Membranes Prepared by Highly Selective Photocyclic Aromatization of Poly(Substituted Acetylene)
1
SC
Guanwu Yin1, Eri Ohtaka1, Toshiki Aoki1*, Jianjun Wang,2Masahiro Teraguchi1, and Takashi Kaneko1 Department of Chemistry and Chemical Engineering; Graduate School of Science
and Technology, Niigata University, Ikarashi 2-8050, Nishi-Ku, Niigata 950-2181,
M AN U
Japan. College of Chemistry and Chemical Engineering, Qiqihar University, Wenhua Street 42, Qiqihar 161006, China
2
Corresponding author: Toshiki Aoki*
TE D
Address: Department of Chemistry and Chemical Engineering, Graduate School of Science and Technology, Niigata University, Ikarashi 2-8050, Niigata, 950-2181, Japan.
AC C
EP
Tel.:/fax: +81 25 262 7280; E-mail: [email protected]
1 | 14
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
RI PT
ACCEPTED MANUSCRIPT
Graphical Abstract
SC
Oxygen Permselectivities through Supramolecular Polymer Membranes Prepared by Highly Selective Photocyclic Aromatization of Poly(Substituted Acetylene)
M AN U
Guanwu Yin1, Eri Ohtaka1, Toshiki Aoki1*, Jianjun Wang,2Masahiro Teraguchi1, and Takashi Kaneko1
TE D
: S1
OH
HO
C12H25O
OC12H25
6 5
OH
Upper bound polts f rom the literature (2016)
OH
EP
4
OH
3
OC12H25
2
S1
1
AC C
HO
in
Robeson`s upper bounds(2008) : 1
10
10
2
3
10
10
4
5
10
OH
n
OC12H25
H
OH
OH
P O2 (barrer)
nanopore
n OC12H25 H
OH
2 | 14
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
ACCEPTED MANUSCRIPT
ABSTRACT: In order to estimate oxygen permselectivities through a supramolecular polymer membrane for the first time, we prepared three kinds of composite
RI PT
membranes consisting of a supramolecular polymer membrane and a conventional organic polymer membrane. Supramolecular polymer membranes were prepared by
SC
using highly selective photocyclic aromatization of poly(substituted acetylene)s. It was found that the supramolecular polymer membrane had oxygen permselectivity (α
M AN U
= PO2/PN2) around three in spite of its low molecular weight.
Keywords: Oxygen Permselectivities; Supramolecular Polymer Membranes; Highly
Introduction
TE D
Selective Photocyclic Aromatization
EP
Gas permselective membranes are very important and valuable because necessity to remove impurities and to purify mixtures of gases, whose molecular sizes are very
AC C
similar, is enhancing for solving recent environmental problems [1-7]. As materials for gas permselective membranes, organic polymers have been usually used because they have a good self-supporting membrane forming ability, which is based on their high molecular weights (MW) and good solubility in cast solvents. However, polymeric materials having both high MW and good solubility are not many, and therefore molecular design of polymers for gas separation membranes was limited. If
3 | 14
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
ACCEPTED MANUSCRIPT
organic materials having low MW can be applied to gas permselective membranes, variation of the molecular design will be enlarged for gas separation membranes. Although supramolecular polymers were reported as new materials prepared
RI PT
from low MW compounds, [8-10] their strength were very low and therefore there have been no reports on their application to gas separation membranes which need to resist pressure differences. Therefore no data on gas permeability of supramolecular
TE D
M AN U
SC
polymers have been reported so far.
AC C
EP
Figure 1. Preparation of self-supporting supramolecular polymer membranes by using highly seletive photocyclic aromatization (SCAT).
We have previously reported highly selective photocyclic aromatization (SCAT)
of cis-cisoidal (c-c) poly(substituted acetylene)s to yield quantitatively the corresponding cyclic trimer, that is, 1,3,5-tirisubstituted benzene (Figure 1). [11-17] In the SCAT reaction by irradiation of visible light, c-c poly(substituted acetylene) membranes were quantitatively converted to the corresponding cyclic trimers. In addition, the resulting benzene derivatives maintained their self-membrane forming 4 | 14
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
ACCEPTED MANUSCRIPT
abilities in spite of their low MW (Figure 1). They were one kind of supramolecular polymers formed on the basis of hydrogen bonds and π−π stackings. However, mechanical strengths of the supramolecular membranes were very low. Although we
RI PT
tried to enhance their membrane strengths, their strengths were not enough to use separation membranes. [16]
In this communication, in order to estimate oxygen permselectivities through
SC
supramolecular polymer membranes for the first time, we prepared three kinds of
M AN U
composite membranes consisting of a supramolecular polymer membrane and a conventional organic polymer membrane (Figure 2). Supramolecular polymer
AC C
EP
TE D
membranes were prepared by SCAT of poly (substituted acetylene)s.
5 | 14
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
Figure 2. Preparation of three kinds of composite membranes consisting of supramolecular polymer membranes (S1) and organic polymer membranes (P1 or P2) by using highly seletive photocyclic aromatization (SCAT). (I) A composite membrane by partial SCAT by photoirradiation on the one side of a SCAT-active polymer(P1) membrane, (II) a composite membrane by SCAT of a membrane of a copolymer (P(1/2)) consisting of a SCAT-active sequence (P1) and a SCAT-inactive sequence (P2), (III) a composite membrane containing S1 inside nanopores of P1.
6 | 14
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
ACCEPTED MANUSCRIPT
RI PT
Results and discussion Synthesis of acetylene monomers 1 and 2 and their (co)polymers P1 and P(1/2) were carried out according to our previous reports. [18-20] Self-supporting polymer
SC
membranes of P1 and P(1/2) used for SCAT reactions were fabricated by solvent cast
M AN U
method using THF as a solvent. SCAT reactions and their estimation were carried out according to our previous report. [11-17] Oxygen and nitrogen permeability coefficients (PO2 and PN2: barrer) were measured by gas chromatographic method according to our previous report. [21-22]
TE D
As mentioned above, supramolecular polymer membranes prepared by SCAT of P1 was very week. First, we prepared blend membranes of P1 and its SCAT product S1. [21-22] Even when the S1 contents[23] were very low (less than 10 wt%), the
EP
blend membranes had lost self-membrane forming abilities. Therefore, this membrane
AC C
preparation method was not suitable for estimation of permselectivity of S1. On the other hand, when we controlled the conversion of the SCAT reaction by changing irradiation time on P1 (Method I in Figure 2)[24], the resulting composite membranes containing 0-65wt% of S1 showed self-membrane forming abilities and their permeabilities could be measured. Therefore this method (I) was suitable for estimation of permselectivity of S1 by extrapolation[25]. As a result, α (= PO2/PN2) of S1 was estimated as 2.62 (Table 1 and Figure 3, no.1). It is not high but shows the 7 | 14
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
ACCEPTED MANUSCRIPT
supramolecular membrane may have an ability to separate oxygen and nitrogen gas molecules similar to usual polymer membrane P1 (α =2.82, no.7). In general for a
AC C
EP
TE D
M AN U
SC
RI PT
membrane to show oxygen permselectivity, the membrane should not have any
8 | 14
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
ACCEPTED MANUSCRIPT
defects including molecular size pores. Therefore, the finding that S1 showed oxygen
6 Upper bound polts from the literature (2016)
5
8
Ro be Ro son` be s u so pp n` er su b pp ou er nd bo (20 un 08 d( ) 19 15 91 ) 14
764 3 5 21
2 1
1
10
100 1000 Po (barrer)
4
10000
100000
M AN U
2
SC
13 9 10 11 12
3
RI PT
4
8
II 13 III 9 10 7 11
6 5
4
TE D
3
3 2
12
1
I
2
EP
0
50
100 Po 2(barrer)
150
AC C
Figure 3. Oxygen permeation behavior through composite membranes consisting of supramolecular polymer membranes (S1) and organic polymer membranes (P1 or P2). ■ (nos.1-7, I) composite membranes by partial SCAT by photoirradiation on the one side of a SCAT-active polymer(P1) membrane; ●(nos.9-12, II) composite membranes by SCAT of a membrane of a copolymer(P(1/2)) consisting of a SCAT-active sequence (P1) and a SCAT-inactive sequence (P2); ♦ (no.13,III) a composite membrane containing S1 inside nanopores of p-P1. The data for nos. 1 and 8 were extrapolated values and the others are
permselectivity may indicate S1 membrane is dense in spite of a low MW compound. However, since the composite membranes I may be an asymmetric membrane as shown in Figure 2 (I), they must form a large area of pure S1 membranes. Therefore, 9 | 14
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
EP
it may contain some defects. To decrease such possibility to form defects in S1 membranes, we prepared composite membranes II consisting of S1 and P2 (S1/P2)
AC C
by SCAT of the copolymer of 1 and 2 (P(1/2)) as shown in Figure 2 (II). Because poly(2) is not active for SCAT reaction[26], S1/P2 could be obtained from P(1/2)[27]. Although we expected the composite membranes II would have better self-membrane forming ability than membranes I because areas of S1 must be very small, their highest content of S1 for II (6.5wt%) was unexpectedly much lower than that for I (65wt%) as shown in Table 1 and Figure 3, nos. 2 and 9). Therefore this method (II) was not suitable for estimation of permselectivity of S1 by extrapolation[25]. 10 | 14
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
ACCEPTED MANUSCRIPT
However, in the series of membranes II, when the content of S1 increased, their α (= PO2/PN2) increased, which is different from membranes I. The extrapolation value of α was 3.75 (Table 1 and Figure 3, no.8). This means the supramolecular polymer
RI PT
membrane had a higher α than the corresponding conventional polymer membrane. The value may not a precise value but this interesting behavior is possible because S1 may have a higher density than P1. In fact, when the SCAT reaction, the starting P1
SC
membranes shrank[28]. It may be caused by enhancing π−π stacking. Therefore, S1
M AN U
can be denser than P1 and enhancing their α values, although S1 is a decomposed product of P1.
Although the two α values for S1 by extrapolation were determined above mentioned, they are not direct values. Therefore it may partly contain separation
TE D
ability based on P1 or P2. To estimate more reliable α values for S1 more directly, we prepared composite membranes III as shown in Figure 2 (III) [29], where we used a nanoporous membrane of P1 (p-P1). The p-P1 membrane was prepared from a dilute
EP
chloroform solution of P1. It showed very high PO2 and very low α (Table 1 and
AC C
Figure 3, no.14). Since the α value was less than 1.0, the membrane did not have separation ability for oxygen and nitrogen. Therefore, it is suitable to estimate separation ability of S1. We prepared a composite membrane of S1 and P1 (III), that is, p-P1 membrane including S1 in its nanopores. Surprisingly the composite membrane showed much higher α value (= 3.08) than pure p-P1. [30] The selectivity can be estimated as that caused originally by S1.
11 | 14
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
ACCEPTED MANUSCRIPT
In conclusion, judging from the three series of measurements, it was found that the α values for the supramolecular polymer membrane S1 was around 3. It is quite high and similar to those for conventional organic polymers. We estimated gas
RI PT
permselectivity through supramolecular polymer membranes successfully for the first time. We are now in progress to discuss molecular design for the supramolecular
SC
polymers to enhance their gas permselective performances.
M AN U
References and Notes
[1] Ahn J, Chung W J, Pinnau I, Song J, Du N, Robertson G P, Guiver MD. J. Membr. Sci. 2010; 346: 280.
[2] Robeson L M. J. Membr. Sci. 2008; 320: 390.
TE D
[3] Aoki T. Prog. Polym. Sci. 1999; 24: 951.
[4] Aoki T, Kaneko T, Teraguchi M. Polymer 2006; 47: 4867. [5] Jung C H, Lee J E, Han S H, Park H B, Lee Y M. J. Membr. Sci. 2010; 350: 301.
EP
[6] Sen S K, Banerjee S. J. Membr. Sci. 2010; 350: 53.
AC C
[7] Thomas S, Pinnau I, Du N, Guiver M D. J. Membr. Sci. 2009; 333: 125. [8] Folmer B J B, Sijbesma R P, Versteegen R M, Rijt J A J V, Meijer E W. Adv Mater 2000; 12: 874. [9] Yoshikawa I, Sawayama J, Araki K. Angew Chem Int Ed 2008; 47: 1038. [10] Yamada N, Komatsu T, Yoshinaga H, Yoshizawa K, Edo S, Kunitake M. Angew Chem Int Ed 2003; 42: 5496. [11] Liu L, Namikoshi T, Zang Y, Aoki T, Teraguchi M, Kaneko T. J. Am. Chem. Soc. 2013; 135: 602.
12 | 14
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
ACCEPTED MANUSCRIPT
[12] Zang Y, Aoki T, Teraguchi M, Kaneko T, Ma L, Jia H. Polym. Rev. 2015; 55: 57. [13] Zang Y, Yin G, Aoki T, Teraguchi, Kaneko T. Ma L, Jia H. Chirality 2015; 27: 459. [14] Liu L, Long Q, Aoki T, Namikoshi T, Abe Y, Miyata M, Teraguchi M, Kaneko T, Wang Y,
RI PT
Zhang C. Macromol. Chem. Phys. 2015; 216: 530. [15] Miyata M, Teraguchi M, Endo H, Kaneko T, Aoki T. Chem. Lett. 2014; 43: 1476.
Kaneko T. Polym. Commun. 2013; 54: 4431.
SC
[16] Miyata M, Namikoshi T, Liu L, Zang Y, Aoki T, Abe Y, Oniyama Y, Tsutsuba T, Teraguchi M,
M AN U
[17] Yin G, Liu L, Aoki T, Namikoshi T, Teraguchi M, Kaneko T. Chem. Lett. 2016; 45: 813. [18] Aoki T, Kaneko T, Maruyama N, Sumi A, Takahashi M, Sato T, Teraguchi M. J. Am. Chem. Soc. 2003; 125: 6346.
[19] Hadano S, Kishimoto T, Hattori T, Tanioka D, Teraguchi M, Aoki T, Kaneko T, Namikoshi T,
TE D
Marwanta E. Macromol. Chem. Phys. 2009; 210: 717.
[20] Liu L, Zang Y, Jia H, Aoki T, Hadano S, Teraguchi M, Miyata M, Zhang G, Namikoshi T. Polym. Rev. 2017; 57: 89.
EP
[21] Wang J, Zang Y, Yin G, Aoki T, Urita H, Taguwa K, Liu L, Namikoshi T, Teraguchi M, Kaneko
AC C
T, Ma L, Jia H. Polymer 2014; 55: 1384. [22] Wang J, Aoki T, Liu L, Namikoshi T, Teraguchi M, Kaneko T. Chem. Lett. 2013; 42: 1090. [23] The SCAT contents (=SCAT conversion) were easily and clearly calculated from the area ratio of the GPC peaks. For the details, see Figure S1. [24] The asymmetric composite membranes were prepared just by photoirradiation on one-sides of polymer membranes. For the detail SCAT conditions, see the SI. [25] For the details of the extrapolation, see Figure S2.
13 | 14
A170830_1906U_SupraPerm_PolymCommun_spl_ALLincldTOC_spl__spl_REV_spl__V2.docx
ACCEPTED MANUSCRIPT
[26] Since the SCAT conversions were always less than those of the composition of monomer unit 1, it was confirmed no SCAT happened in comonomer 2 sequence whose homopolymer poly(2) is SCAT inactive.
be a relatively block type not a random type.
RI PT
[27] Since the SCAT conversion of P(1/2) could reach up to the composition of S1, the copolymer may
[28] Judging from the shrink of the membrane after SCAT reaction (see Figure S3 in the SI), the
SC
density of the SCAT product is higher than that of the original polymer.
M AN U
[29] The composite membrane III was prepared by immersing the nanoporous p-P1 membrane in methanol solution of S1. We have recently found the supramolecular polymer membranes of S1 formed from methanol solution.
[30] To confirm no change of the base polymer p-P1 itself during preparation of the composite
TE D
membrane III, a controlled experiment was carried out. Almost no change in permeability was
AC C
EP
observed (Table 2 and Figure 3, no 15).
14 | 14
ACCEPTED MANUSCRIPT Highlights for Oxygen permselectivities through supramolecular polymer membranes prepared by highly selective photocyclic aromatization of poly(substituted acetylene) Corresponding author's name: Toshiki Aoki*
RI PT
1) We developed three new preparation methods for composite membranes consisting of a supramolecular polymer membrane and a conventional organic polymer membrane, which have enough strengths for application to gas separation membranes, by highly (SCAT). (Preparationa of supramolecular
SC
selective photocyclic aromatization
polymer membranes by the SCAT reaction have been previously reported by us in
M AN U
JACS 2013 but no application was reported).
2) We found the supramolecular polymer membranes had an ability to separate oxygen and nitrogen in permeation and the permselectivity was relatively high and similar to that for usual organic polymer membranes. To the best of our knowledge, this is the first experimental result on gas separation by supramolecular polymer
AC C
EP
TE D
membranes.