Xiaohua
Ma
and
Ingo
Pinnau
*
Advanced Membranes and Porous Materials Center, Physical Sciences and Engineering Division, Chemical and Biological Engineering Program, King Abdullah University of Science and Technology, Thuwal 23955-6900, Saudi Arabia. E-mail: ingo.pinnau@kaust.edu.sa; Fax: +966 012 80821328
First published on 9th December 2015
A novel intrinsically microporous polymer was synthesized by polycondensation reaction of 1,1′,2,2′,-tetrahydroxy-tetraphenylethylene (TPE) and 2,3,5,6-tetrafluoroterephthalonitrile (TFTPN). In addition, a series of copolymers was prepared from TPE, 5,5′,6,6′-tetrahydroxy-3,3,3′,3′-tetramethylspirobisindane (TTSBI) and TFTPN. All TPE-derived polymers exhibited high molecular weight, good solubility in common organic solvents, high thermal stability and high surface area (550 to 660 m2 g−1). The CO2 permeability of a methanol-treated and 120 °C vacuum-dried TPE-TFTPN film was 862 Barrer with a moderate CO2/N2 selectivity of 26. The selectivity of the TPE-TTSBI-PIMs decreased with increasing TTSBI content coupled with a sharp increase in permeability. Molecular simulations indicated that the introduction of the tetraphenylethylene unit resulted in an increased rotational freedom of dihedral angles in the polymer main chain relative to those of the spirobisindane-based PIM-1.
Glassy ladder polymers of intrinsic microporosity were first introduced in 2004 by McKeown's and Budd's group.9,10 The state-of-the-art PIM-1 was formed by simple condensation polymerization of 5,5′,6,6′-tetrahydroxy-3,3,3′,3′-tetramethylspirobisindane (TTSBI) and 2,3,5,6-tetrafluoroterephthalonitrile (TFTPN). The resulting polymer was composed of rigid ladder-type repeat units including bulky, sterically hindered spirobisindane contortion sites. As a result, PIM-1 showed highly inefficient chain packing with a remarkable microporosity, as determined by nitrogen adsorption measurements at −196 °C (SBET ∼ 800 m2 g−1).
PIM-1 exhibited over 100-fold enhanced permeability compared to commercial low-free-volume polymers, such as Matrimid,11 coupled with a modest selectivity. The gas permeation properties of PIM-1 reached the 2008 Robeson permeability/selectivity upper bound for some gas pairs.12–15 Based on the promising results obtained for PIM-1, many other high-performance ladder PIMs,16–29 modified prototype PIM-1,30–33 and PIM polyimides (PIM-PI)34–42 have been reported with excellent gas separation properties. In fact, several ladder PIMs25–27 and PIM-PIs37,40,41 demonstrated much better performance than other known polymers, thereby redefining the previously reported permeability/selectivity upper bound for air and hydrogen separation.27 However, in the burgeoning materials class of PIMs there is still a quest for novel custom-designed building blocks to further improve the understanding of structure/property relationships for advanced gas separation membranes.
Tetraphenylethylene (TPE) is a building block widely investigated in aggregation-induced emission (AIE) materials, which originated from the restriction of the intramolecular rotation (RIR) effect in the tetraphenylethylene moiety.43,44 TPE is a bulky and sterically hindered group with great potential as a building block for the development of novel PIMs for membrane-based separation. In this study, we report for the first time the design and synthesis of tetraphenylethylene-based PIMs, as illustrated in Scheme 1. The detailed synthetic procedure and characterization are provided in the ESI.†
![]() | ||
Scheme 1 Design and synthesis of tetraphenylethylene-based homo-PIM and co-PIMs. m = 1, n = 0 (TPE-TFTPN); m = 0, n = 1 (TTSBI-TFTPN = PIM-1). |
![]() | ||
Fig. 1 Proton NMR of TPE-PIM, PIM-1 and TPE-PIM-1 copolymers; the peaks of the copolymers were normalized at 6.41 ppm. |
The resulting co-PIMs demonstrated good solubility in some common solvents such as THF, dichloromethane, chloroform, NMP, DMF, and DMAc (Table S1†). These polymers exhibited high molecular weight (Mn), ranging from 3.4 × 104 to 14.5 × 104 g mol−1, coupled with a polydispersity of 2.1 to 3.2, as determined by GPC using chloroform as the solvent. Transparent, yellow films were made by casting a 2–3% (wt/v) chloroform solution into leveled Petri dishes and evaporating the solvent under ambient conditions. The films were then soaked in methanol for 24 hours to remove any trapped casting solvent. Thereafter, the films were dried at 120 °C under vacuum for 24 h. TGA experiments confirmed that the films were solvent-free prior to the gas permeation tests.
Similar to previously reported microporous PIMs,18 all of the homo- and co-PIMs of this study exhibited completely amorphous structures as identified by wide-angle X-ray scattering shown in Fig. 2. TPE-PIM demonstrated tighter chain packing compared with PIM-1, as indicated by the much weaker peak at a low-angle around 13.6° (d-spacing = 6.5 Å), whereas higher intensity is evident at an angle of around 20° (d-spacing = 4.4 Å). As expected, the copolymers show X-ray scattering between those of the homopolymers, depending on the TPE/TTSBI ratio (Fig. 2).
![]() | ||
Fig. 2 Wide-angle X-ray scattering of TPE-PIM, PIM-1 and copolymers. TPE-25, TPE-50 and TPE-75 were normalized and shifted in the y-axial direction above PIM-1 and TPE-PIM. |
All TPE-based PIMs demonstrated high thermal stability as identified from their onset decomposition temperature over 440 °C, as illustrated in Fig. 3. It is interesting to note that TPE-PIM had a 40 °C higher decomposition temperature than PIM-1, most likely due to its higher aromatic content.
The TPE-based PIMs demonstrated high surface area (Table 1), which was calculated from their N2 adsorption isotherms at −196 °C, as illustrated in Fig. 4. All polymers exhibited type I adsorption isotherms characterized by remarkable sorption uptake at very low relative pressure (p/p0 < 0.01), indicating the presence of significant microporosity. TPE-PIM had the lowest surface area of 550 m2 g−1 of all polymers reported in this study. As the fraction of the spirobisindane repeat unit increased in the copolymers, a significant increase in the adsorption uptake at a relatively low pressure (p/p0 < 0.1) was observed, and the surface area of the TPE-based PIMs increased continuously from 550 m2 g−1 (TPE-PIM) to 770 m2 g−1 (PIM-1).
Polymers |
M
n × 10−4![]() |
PDIb |
T
d![]() |
S BET (m2 g−1) | Total pore volumed |
---|---|---|---|---|---|
a The molecular weight of the polymers was measured by GPC using chloroform as the solvent and polystyrene as the external standard. The eluent speed was 1 mL min−1. b The PDI was obtained by Mw/Mn. c The decomposition temperature was selected as the onset decomposition temperature of the TGA curve. d The total pore volume was obtained at the relative N2 pressure (p/p0) of 0.97. | |||||
TPE-PIM | 7.83 | 2.3 | 475 | 550 | 0.54 |
TPE-75 | 3.40 | 3.2 | 465 | 560 | 0.47 |
TPE-50 | 6.00 | 2.1 | 455 | 600 | 0.45 |
TPE-25 | 4.90 | 2.3 | 440 | 660 | 0.51 |
PIM-1 | 14.5 | 2.4 | 430 | 770 | 0.70 |
In order to evaluate the performance of TPE-based PIMs, the gas transport properties of the polymers were analyzed by the constant-volume/variable-pressure method. The data are summarized in Table 2. The permeability and selectivity values of PIM-1 were similar to those reported by Du et al. for the films prepared under similar conditions.18 A sharp decrease in permeability was observed for all gases when the tetraphenylethylene moiety was substituted for spirobisindane in the PIM copolymers. For example, TPE-PIM had a CO2 permeability of 862 Barrer, which is about 7-fold lower than PIM-1. However, there is a concurrent improvement in selectivity, e.g. the selectivity of O2/N2 and CO2/CH4 improved from 3.2 to 4.2 and 12 to 21, respectively. The sequence of the gas permeability of all polymers in this study was PCO2 > PH2 > PO2 > PCH4 > PN2, which is typically observed for high surface area PIMs. It is interesting to note that there is a qualitative correlation between the permeability and the surface area of the polymers, that is higher BET surface area was coupled with higher gas permeability and lower gas selectivity by increasing the spirobisindane content in the TPE-copolymers (Table 2).
Polymera | H2 | Permeability (Barrer)b | CO2 | Ideal selectivity (αX/Y) | |||||
---|---|---|---|---|---|---|---|---|---|
N2 | O2 | CH4 | H2/N2 | O2/N2 | CO2/N2 | CO2/CH4 | |||
a The membranes were cast using chloroform as the solvent and soaked in methanol for 24 hours, and thereafter dried under vacuum at 120 °C for 24 h. b 1 Barrer = 10−10 cm3 (STP) cm cm−2 s−1 cm Hg−1 = 7.5 × 10−18 m3 (STP) m m−2 s−1 Pa−1. | |||||||||
TPE-PIM | 604 | 33.4 | 138 | 41 | 862 | 18 | 4.2 | 25.8 | 20.9 |
TPE-75 | 620 | 42 | 154 | 61 | 977 | 15 | 3.7 | 23.3 | 16 |
TPE-50 | 1004 | 88 | 288 | 134 | 1869 | 11 | 3.4 | 21.3 | 14 |
TPE-25 | 2332 | 256 | 836 | 415 | 5203 | 9 | 3.3 | 20.3 | 12.5 |
PIM-1 | 3017 | 350 | 1103 | 495 | 5922 | 8.6 | 3.2 | 16.9 | 12.0 |
The permeability/selectivity trade-off plots for O2/N2, CO2/CH4, CO2/N2 and H2/N2 are illustrated in Fig. S2.† The TPE-based PIMs and co-PIMs are located below the 2008 and 2015 permeability/selectivity trade-off curves. It is suggested that the rotational freedom of the ethylene bond in the TPE-PIM repeat unit leads to more efficient polymer packing and, hence lower free volume and permeability. The calculated rotation energies of the dihedral angle of the tetraphenylethylene and spirobisindane units are highlighted in Fig. 5. When the dihedral angle was rotated from −180 to +180°, two energy minima for TPE-PIM, both at the positive dihedral angle of 130° and a negative dihedral angle around −50°, were observed. In the case of PIM-1, owing to its fused ring structure, only one energy minimum at around −50° was observed. Therefore, the conformational freedom of TPE-PIM is much larger than that of PIM-1, resulting in a more densely packed polymer structure.
![]() | ||
Fig. 5 The rotation angles selected in TPE-PIM (blue) and PIM-1 (red) and their calculated rotation energy against dihedral angle. |
Footnote |
† Electronic supplementary information (ESI) available: The synthesis details, solubility, membrane formation, testing protocol and upper bound plots of the homo- and co-PIMs for different gas pairs. See DOI: 10.1039/c5py01796c |
This journal is © The Royal Society of Chemistry 2016 |