T. Sakaguchi*,
Y. Lin and
T. Hashimoto
Department of Materials Science and Engineering, Graduate School of Engineering, University of Fukui, Bunkyo, Fukui 910-8507, Japan. E-mail: sakaguchi@matse.u-fukui.ac.jp
First published on 15th June 2017
Diphenylacetylenes having various silyl groups [PhCCC6H4-R; R = p-SiMe3 (TMSDPA), p-SiEt3 (TESDPA), p-SiMe2-n-C8H17 (DMOSDPA), and p-SiPh3 (TPSDPA)] were copolymerized with diphenylacetylene having a tert-butyl group (PhCCC6H4-tertBu; TBDPA) using a TaCl5–n-Bu4Sn catalyst in various monomer feed ratios to provide high-molecular-weight copolymers in high yields. The free-standing membranes were fabricated by solution-casting, except poly(TPSDPA-co-TBDPA). Interestingly, the gas permeability of poly(TMSDPA-co-TBDPA) was higher than those of the homopolymers, poly(TMSDPA) and poly(TBDPA). The permeability of the copolymers became lower as the silyl groups became bulkier. The desilylation of membranes was carried out using a mixture of trifluoroacetic acid/hexane. When bulkier silyl groups were removed, the oxygen permeability increased to larger extents. The oxygen permeability coefficients of the copolymers and desilylated copolymers increased with increasing composition ratio of TBDPA. The gas diffusivity and gas solubility were also increased upon desilylation.
In the present study, we synthesized copolymers of diphenylacetylenes having silyl and tert-butyl groups at para position of phenyl groups (Scheme 1), and examined the desilylation of the membranes. The desilylated copolymers have still bulky tert-butyl groups, and therefore the spaces occupied by silyl groups are possibly maintained after desilylation by the steric repulsion of tert-butyl groups.
Gas permeability coefficients of polymer membranes were measured with a Tsukubarikaseiki K-315-N gas permeability apparatus equipped with an MKS Baratron detector at 25 °C. The downstream side of the membrane was evacuated at ca. 0.3 Pa, while the upstream side was filled with a gas at ca. 1 atm (105 Pa), and the increase of pressure in a downstream receiving vessel was measured. The permeability coefficient (P) expressed in barrer unit [1 barrer = 1 × 10−10 cm3 (STP) cm cm−2 s−1 cmHg−1] was calculated from the slopes of time–pressure curves in the steady state where Fick's law held. The diffusion coefficient (D) expressed in cm2 s−1 units was determined by the time lag method using the following equation:
D = l2/6θ |
S = P/D. |
M | Feed ratio | Polymerb | ||
---|---|---|---|---|
(M:TBDPA) | Yield, % | Mwc | Mw/Mnc | |
a In toluene at 80 °C for 24 h; [M]0 + [TBDPA]0 = 0.20 M, [TaCl5] = 20 mM, [n-Bu4Sn] = 80 mM.b Methanol-insoluble product.c Measured by GPC.d Acetone-insoluble product.e CHCl3-soluble part. | ||||
TMSDPA | 4:1 | 84 | 3320000 | 5.05 |
2:1 | 83 | 3230000 | 4.61 | |
1:1 | 83 | 3160000 | 4.86 | |
1:2 | 86 | 3300000 | 4.88 | |
1:4 | 93 | 2080000 | 5.43 | |
TESDPA | 4:1 | 64 | 1120000 | 2.69 |
2:1 | 66 | 1440000 | 4.21 | |
1:1 | 73 | 1800000 | 5.17 | |
1:2 | 74 | 2170000 | 4.27 | |
1:4 | 91 | 1630000 | 5.29 | |
DMOSDPA | 4:1 | 67 | 4080000 | 4.24 |
2:1 | 80 | 4560000 | 3.90 | |
1:1 | 81 | 5010000 | 4.06 | |
1:2 | 88 | 4360000 | 5.01 | |
1:4 | 87 | 4500000 | 3.57 | |
TPSDPA | 4:1 | 70d | 378000e | 2.79 |
2:1 | 76d | 376000e | 2.51 | |
1:1 | 80d | 233000e | 2.75 | |
1:2 | 74d | 335000e | 2.27 | |
1:4 | 79d | 397000e | 1.79 |
The copolymerization of TMSDPA having trimethylsilyl group with TBDPA in feed ratio of 4:1 produced poly(TMSDPA-co-TBDPA) with Mw of 3320000 in 84% yield. The copolymerizations in feed ratios of 2:1, 1:1, 1:2, and 1:4 also proceeded in the same manner and copolymers were obtained with Mw of 3230000, 3160000, 3300000, and 2080000, respectively. The copolymerizations of TESDPA having triethylsilyl group with TBDPA under the same conditions gave poly(TESDPA-co-TBDPA)s (Mw = 1120000–2170000). DMOSDPA having dimethyl-n-octylsilyl group also copolymerized with TBDPA to afford poly(DMOSDPA-co-TBDPA)s with high molecular weight (Mw = 4080000–5010000). The copolymerizations of TPSDPA having triphenylsilyl group with TBDPA provided poly(TPSDPA-co-TBDPA)s in high yields, but the obtained poly(TPSDPA-co-TBDPA)s showed poor solubility. Poly(TPSDPA-co-TBDPA)s were partially soluble in chloroform, and the Mw's of CHCl3-soluble parts were one order of magnitude lower than those of the other copolymers.
Feed ratio | Toluene | CHCl3 | THF | DMF | DMSO | Methanol |
---|---|---|---|---|---|---|
(RDPA:TBDPA) | ||||||
a Symbols: (+) soluble, (±) partly soluble, (−) insoluble. | ||||||
Poly(TMSDPA-co-TBDPA), poly(TESDPA-co-TBDPA), poly(DMOSDPA-co-TBDPA) | ||||||
4:1 | + | + | + | − | − | − |
2:1 | + | + | + | − | − | − |
1:1 | + | + | + | − | − | − |
1:2 | + | + | + | − | − | − |
1:4 | + | + | + | − | − | − |
Poly(TPSDPA-co-TBDPA) | ||||||
4:1 | ± | ± | ± | − | − | − |
2:1 | ± | ± | ± | − | − | − |
1:1 | ± | ± | ± | − | − | − |
1:2 | ± | ± | ± | − | − | − |
1:4 | ± | ± | ± | − | − | − |
DSpoly(TMSDPA-co-TBDPA) | ||||||
4:1 | − | − | − | − | − | − |
2:1 | − | − | − | − | − | − |
1:1 | − | − | − | − | − | − |
1:2 | − | ± | ± | − | − | − |
1:4 | ± | ± | ± | − | − | − |
DSpoly(TESDPA-co-TBDPA) | ||||||
4:1 | − | − | − | − | − | − |
2:1 | − | − | − | − | − | − |
1:1 | − | + | + | − | − | − |
1:2 | + | + | + | − | − | − |
1:4 | + | + | + | − | − | − |
DSpoly(DMOSDPA-co-TBDPA) | ||||||
4:1 | − | − | − | − | − | − |
2:1 | − | − | − | − | − | − |
1:1 | − | + | ± | − | − | − |
1:2 | + | + | + | − | − | − |
1:4 | + | + | + | − | − | − |
Tough free-standing membranes could be fabricated by casting polymers from their toluene solutions for poly(TMSDPA-co-TBDPA)s, poly(TESDPA-co-TBDPA)s, and poly(DMOSDPA-co-TBDPA)s. However, it was difficult to prepare a membrane of poly(TPSDPA-co-TBDPA) by solution-casting because of too poor solubility. The desilylation was carried out in a mixture of hexane/trifluoroacetic acid (1:1 volume ratio) at room temperature for 24 h to afford the desilylated membranes. It is known that the aryl–Si bond is easily cleaved by proton acids.18,19 Fig. 1 shows the IR spectra of membranes of poly(DMOSDPA-co-TBDPA) and desilylated analogue [DSpoly(DMOSDPA-co-TBDPA)] in the feed ratio of 1:4. The absorptions at 1250 cm−1 derived from the stretching of SiC–H bonds and at 1120 cm−1 derived from the vibration of Si–C completely disappeared in the spectrum after desilylation reaction. This indicates that the silyl groups were completely eliminated from the polymer membrane even the silyl groups were bulky dimethyl-n-octylsilyl groups. For poly(TMSDPA-co-TBDPA)s and poly(TESDPA-co-TBDPA)s, the completion of desilylation was confirmed by IR spectra of the polymer membranes. The color of polymer membranes changed from yellow to orange through desilylation, which would suggest that the conjugated main chains were maintained during the reaction. The results of desilylation are the same as the previous studies.10,12
Fig. 1 IR spectra of membranes of poly(DMOSDPA-co-TBDPA) and DSpoly(DMOSDPA-co-TBDPA) in the feed ratio of 1:4. |
The solubility of the desilylated copolymers, DSpoly(TMSDPA-co-TBDPA)s, DSpoly(TESDPA-co-TBDPA)s, and DSpoly(DMOSDPA-co-TBDPA)s was examined (Table 2). The desilylated copolymers showed less solubility than silyl group-containing copolymers. DSpoly(TMSDPA-co-TBDPA)s, DSpoly(TESDPA-co-TBDPA)s, and DSpoly(DMOSDPA-co-TBDPA)s in the feed ratios of 4:1 and 2:1 were practically insoluble copolymers in any solvents, indicating that the incorporation of the particular content of tert-butyl groups is essential for the solubility of poly(diphenylacetylene). The desilylated copolymers in the feed ratios of 1:1, 1:2, 1:4 indeed showed better solubility in toluene, CHCl3, and THF in general because they had tert-butyl groups in relatively high contents.
Membrane | Ratio | PO2 | PN2 | PO2/PN2 |
---|---|---|---|---|
a In the units of 1 × 10−10 cm3 (STP) cm cm−2 s−1 cmHg−1 (=1 barrer). | ||||
Poly(TMSDPA-co-TBDPA) | 1:0 | 1500 | 680 | 2.2 |
4:1 | 2200 | 1200 | 1.8 | |
2:1 | 1900 | 1000 | 1.9 | |
1:1 | 2300 | 1300 | 1.8 | |
1:2 | 2100 | 1100 | 1.9 | |
1:4 | 2700 | 1.700 | 1.6 | |
0:1 | 1100 | 690 | 1.6 | |
DSpoly(TMSDPA-co-TBDPA) | 1:0 | 900 | 410 | 2.2 |
4:1 | 1600 | 1000 | 1.6 | |
2:1 | 1900 | 1000 | 1.9 | |
1:1 | 2700 | 1600 | 1.7 | |
1:2 | 2800 | 1800 | 1.6 | |
1:4 | 3200 | 2100 | 1.5 | |
0:1 | — | — | — |
Membrane | Ratio | PO2 | PN2 | PO2/PN2 |
---|---|---|---|---|
a In the units of 1 × 10−10 cm3 (STP) cm cm−2 s−1 cmHg−1 (=1 barrer). | ||||
Poly(TESDPA-co-TBDPA) | 1:0 | 190 | 80 | 2.4 |
4:1 | 290 | 120 | 2.4 | |
2:1 | 480 | 200 | 2.4 | |
1:1 | 880 | 410 | 2.1 | |
1:2 | 1500 | 780 | 1.9 | |
1:4 | 1400 | 690 | 2.0 | |
0:1 | 1100 | 690 | 1.6 | |
DSpoly(TESDPA-co-TBDPA) | 1:0 | 760 | 440 | 1.7 |
4:1 | 1100 | 590 | 1.9 | |
2:1 | 1900 | 1100 | 1.7 | |
1:1 | 1900 | 1100 | 1.7 | |
1:2 | 2700 | 1.500 | 1.8 | |
1:4 | 2100 | 1100 | 1.8 | |
0:1 | — | — | — |
Membrane | Ratio | PO2 | PN2 | PO2/PN2 |
---|---|---|---|---|
a In the units of 1 × 10−10 cm3 (STP) cm cm−2 s−1 cmHg−1 (=1 barrer). | ||||
Poly(DMOSDPA-co-TBDPA) | 1:0 | 27 | 8.2 | 3.3 |
4:1 | 28 | 8.5 | 3.3 | |
2:1 | 31 | 9.1 | 3.4 | |
1:1 | 52 | 24 | 2.2 | |
1:2 | 230 | 150 | 1.5 | |
1:4 | 300 | 120 | 2.5 | |
0:1 | 1100 | 690 | 1.6 | |
DSpoly(DMOSDPA-co-TBDPA) | 1:0 | 650 | 300 | 2.2 |
4:1 | 670 | 310 | 2.2 | |
2:1 | 1300 | 670 | 1.9 | |
1:1 | 1800 | 1100 | 1.6 | |
1:2 | 2000 | 1200 | 1.7 | |
1:4 | 3300 | 2100 | 1.6 | |
0:1 | — | — | — |
The PO2 value of poly(TMSDPA-co-TBDPA) (4:1) obtained by the polymerization in feed ratio of 4:1 was 2200 barrer. The other poly(TMSDPA-co-TBDPA)s also exhibited high gas permeability and their PO2 values were more than 1900 barrer. Poly(TMSDPA-co-TBDPA) (1:4) exhibited the highest permeability and its PO2 value was as large as 2700 barrer. Interestingly, all the poly(TMSDPA-co-TBDPA)s exhibited high oxygen permeability compared to homopolymers. The PO2 values of homopolymers of TMSDPA and TBDPA were 1500 and 1100 barrer, respectively. The higher gas permeability of copolymers may be due to the heterogeneous structure; i.e., the copolymers has both trimethylsilyl and tert-butyl groups as substituents. It is thought that the heterogeneity of substituents generates micro-scale voids when the membrane is formed. Desilylation of homopolymer of TMSDPA decreased the PO2 value from 1500 to 900 barrer. Similarly, the desilylated poly(TMSDPA-co-TBDPA) in the low contents of TBDPA, DSpoly(TMSDPA-co-TBDPA) (4:1), showed the lower oxygen permeability than poly(TMSDPA-co-TBDPA) (4:1). This is because the membrane became dense by the elimination of bulky spherical trimethylsilyl groups. However, the desilylation of poly(TMSDPA-co-TBDPA)s with high contents of TBDPA increased the gas permeability. DSpoly(TMSDPA-co-TBDPA) (1:4) showed the highest oxygen permeability and the PO2 value was as large as 3200 barrer. It is probably because that the spaces occupied by trimethylsilyl groups were maintained as micro-scale voids due to the steric repulsion of tert-butyl groups. These findings indicate that tert-butyl groups play an important role to generate micro-scale voids through desilylation.
The PO2 values of poly(TESDPA-co-TBDPA)s were 290–1500 barrer, which are larger than that of homopolymer, poly(TESDPA). Poly(TESDPA) has flexible ethyl chains, which would occupy the free volume in the membrane.10 Therefore, poly(TESDPA) exhibits lower gas permeability than poly(TMSDPA) and poly(TBDPA). The PO2 values of poly(TESDPA-co-TBDPA)s in low TBDPA contents were smaller than that of poly(TBDPA), but the PO2 values of poly(TESDPA-co-TBDPA)s in feed ratios of 1:2 and 1:4 were larger than that of poly(TBDPA) irrespective the copolymers contained relatively low gas-permeable TESDPA component. This result also suggests that the heterogeneity of substituent enhances the gas permeability. In the previous study, the desilylation of triethylsilyl groups increased the gas permeability because the free volume in the membrane were increased due to the elimination of bulkier silyl groups.10 All the DSpoly(TESDPA-co-TBDPA)s showed higher gas permeability than poly(TESDPA-co-TBDPA)s.
Poly(DMOSDPA) has more flexible long alkyl chains and shows much lower gas permeability.10 Therefore, the PO2 values of poly(DMOSDPA-co-TBDPA)s were smaller than that of poly(TBDPA). However, the desilylation drastically enhanced the gas permeability and DSpoly(DMOSDPA-co-TBDPA) in 1:4 feed ratio showed the highest permeability. In this way, the presence of tert-butyl groups is important to achieve high gas permeability when the bulkier silyl groups were eliminated.
The oxygen/nitrogen separation factors (PO2/PN2) of all the copolymer membranes before and after desilylation were in the range 1.5–3.4, and the values tended to decrease as the PO2 increased. These results agree with the general tendency of gas permeation through polymer membranes.20
Membrane | Ratio | DO2 × 107 | DN2 × 107 | SO2 × 103 | SN2 × 103 |
---|---|---|---|---|---|
a Determined by the “time lag” method at 25 °C. In the units of cm2 s−1.b Calculated by using equation, S = P/D. In the units of cm3 (STP) cm−3 cmHg−1. | |||||
Poly(TMSDPA-co-TBDPA) | 4:1 | 55 | 46 | 40 | 26 |
2:1 | 75 | 62 | 25 | 16 | |
1:1 | 94 | 80 | 24 | 16 | |
1:2 | 69 | 65 | 30 | 17 | |
1:4 | 87 | 81 | 31 | 21 | |
DSpoly(TMSDPA-co-TBDPA) | 4:1 | 34 | 32 | 47 | 31 |
2:1 | 78 | 62 | 24 | 16 | |
1:1 | 110 | 95 | 25 | 17 | |
1:2 | 110 | 93 | 25 | 19 | |
1:4 | 120 | 100 | 27 | 21 |
Membrane | Ratio | DO2 × 107 | DN2 × 107 | SO2 × 103 | SN2 × 103 |
---|---|---|---|---|---|
a Determined by the “time lag” method at 25 °C. In the units of cm2 s−1.b Calculated by using equation, S = P/D. In the units of cm3 (STP) cm−3 cmHg−1. | |||||
Poly(TESDPA-co-TBDPA) | 4:1 | 41 | 35 | 7.1 | 3.4 |
2:1 | 62 | 53 | 7.7 | 3.8 | |
1:1 | 73 | 55 | 12 | 7.5 | |
1:2 | 100 | 65 | 15 | 12 | |
1:4 | 98 | 51 | 14 | 14 | |
DSpoly(TESDPA-co-TBDPA) | 4:1 | 83 | 70 | 13 | 8.4 |
2:1 | 100 | 88 | 19 | 13 | |
1:1 | 110 | 95 | 17 | 12 | |
1:2 | 140 | 120 | 19 | 13 | |
1:4 | 130 | 94 | 16 | 12 |
Membrane | Ratio | DO2 × 107 | DN2 × 107 | SO2 × 103 | SN2 × 103 |
---|---|---|---|---|---|
a Determined by the “time lag” method at 25 °C. In the units of cm2 s−1.b Calculated by using equation, S = P/D. In the units of cm3 (STP) cm−3 cmHg−1. | |||||
Poly(DMOSDPA-co-TBDPA) | 4:1 | 12 | 3.8 | 2.3 | 2.2 |
2:1 | 14 | 7.0 | 2.2 | 1.3 | |
1:1 | 27 | 12 | 1.9 | 2.0 | |
1:2 | 30 | 25 | 7.7 | 6.0 | |
1:4 | 31 | 15 | 9.7 | 8.0 | |
DSpoly(DMOSDPA-co-TBDPA) | 4:1 | 48 | 36 | 14 | 8.6 |
2:1 | 55 | 50 | 24 | 13 | |
1:1 | 99 | 80 | 18 | 14 | |
1:2 | 110 | 85 | 18 | 14 | |
1:4 | 180 | 150 | 18 | 14 |
The D values of DSpoly(TMSDPA-co-TBDPA) with low contents of TBDPA was lower than Si-containing membranes. However, the desilylation of poly(TMSDPA-co-TBDPA) with high contents of TBDPA increased the gas diffusivity. The increases of diffusivity would be due to the increment of free volume of membrane. In addition, DSpoly(TMSDPA-co-TBDPA)s had nearly the same gas solubility as poly(TMSDPA-co-TBDPA)s. The elimination of silyl groups affected the gas diffusivity more effectively than the gas solubility. On the other hand, the desilylation of poly(TESDPA-co-TBDPA)s increased both the D and S values. Especially in high TESDPA ratios, the large enhancement of solubility was observed through desilylation. Since both SO2 and SN2 increased in a similar fashion, the increment in solubility may be due to the increase of the free volume.6 For poly(DMOSDPA-co-TBDPA)s, the enhancement of D and S through desilylation was also quite noticeable, and the D and S values considerably increased after desilylation.
Fig. 2 Relationship between PO2 and TBDPA ratio in copolymers for poly(TMSDPA-co-TBDPA)s, poly(TESDPA-co-TBDPA)s, poly(DMOSDPA-co-TBDPA)s, and the desilylated analogues. |
The three type's desilylated copolymer showed similar oxygen permeability. This means that as the alkyl group becomes longer, an extent of increment in PO2 through desilylation became larger. In this reason, DSpoly(DMOSDPA-co-TBDPA) exhibited the highest PO2 enhancement. Similarly to the case of copolymers before desilylation, the permeability of the desilylated copolymers increased as increasing TBDPA ratio. It suggests that tert-butyl groups are important to generate micro-scale voids after desilylation as well as before desilylation.
This journal is © The Royal Society of Chemistry 2017 |