Acetylene (C
2H
2) is a crucial feedstock in the chemical industry, extensively utilized in the production of various commercial chemicals, including vinyl chloride, acrylic acid, and 1,4-butynediol [
1]. Typically, C
2H
2 is generated through the combustion of natural gas or the thermal cracking of hydrocarbons [
1,
2]. In these processes a significant amount of carbon dioxide (CO
2) also forms as a byproduct [
3]. Moreover, due to the use of water for quenching, raw C
2H
2 products also contain humidity. It is highly desired to develop efficient C
2H
2 purification techniques to meet the demand across different sectors. However, separating C
2H
2 from CO
2 presents a challenge due to their nearly identical molecular size (kinetic diameter of 3.3 Å for both) and shape, as well as similar physical properties (Table S1 in Supporting information) [
4-
9]. Traditional purification methods, including solvent extraction and cryogenic distillation [
10], are both energy-intensive and inefficient. Therefore, adsorptive separation based on porous physisorbents has aroused broad interest thanks to the merits of low energy footprint and purpose-oriented material design.
By virtue of great structure diversity and structure tunability, metal-organic frameworks (MOFs) have been developed as a promising community of porous physisorbents for adsorptive gas separation [
11-
17]. The modular nature enables MOFs to be tailored for achieving molecular sieving (pore size tuning) or enhancing separation performance (pore environment/chemistry tuning, such as incorporating specific binding sites) [
18-
21]. The introduction of functional groups through ligand design to construct hydrophobic pores in MOFs has been proven to be an effective method for improving the stability of MOFs, and meanwhile improving their separation ability of C
2H
2/CO
2 [
22-
24], C
2H
6/C
2H
4 [
25-
28], C
3H
8/C
3H
6 [
29],
n-C
4H
10/
iso-C
4H
10 [
30] and CO
2/N
2 [
31,
32] under humid conditions. In addition, ionic MOFs offer unique advantages for gas adsorption and separation due to their amenability to ion exchange [
33-
35]. The variation of cationic or anionic species within the pores allows for facile modification of the pore chemistry, which can significantly alter their capacity and selectivity in gas adsorption [
33]. Exchanging the ions in MOFs can open up new opportunities to develop advanced physisorbents potentially addressing challenges in important separations. Herein, we synthesized two new ionic MOFs, Fe-BDC-TPT-BF
4, Ni-BDC-TPT-TMA (TMA = (CH
3)
4N
+), from their parent frameworks Fe-BDC-TPT-Cl and Ni-BDC-TPT-Me
2NH
2, respectively (
Fig. 1), through ion exchange. To evaluate the effect of counter ions on gas separation performance of the resulting MOFs, static single-component gas adsorption experiments (C
2H
2 and CO
2) and dynamic breakthrough experiments (C
2H
2/CO
2 separation under dry/wet conditions) were performed on Fe-BDC-TPT-BF
4 and Ni-BDC-TPT-TMA, and compared with their parent MOFs.
The Fe/Ni-BDC-TPT-X (X = Cl, Me
2NH
2) series exhibits a good confinement effect due to the implementation of the pore space partition strategy, which effectively enhances the interaction between the framework and gas molecules, as demonstrated in the separation of C
2H
2/C
2H
4/C
2H
6 [
35], C
2H
2/CO
2 [
36-
38]. Specifically, Fe-BDC-TPT-Cl made with Fe incorporates anionic counterions, while those made with Ni incorporate cationic counterions. This difference in counterion incorporation influences the framework's interaction with specific gas molecules, making these materials highly effective for various gas separation applications. Therefore, Fe/Ni-BDC-TPT-X can modulate pore size and chemistry through ion exchange strategies, making it an excellent platform for post-synthesis modifications aimed at improving gas separation.
Fe-BDC-TPT-Cl and Ni-BDC-TPT-Me
2NH
2 were synthesized according to the literature [
35]. Subsequently, Fe-BDC-TPT-BF
4, and Ni-BDC-TPT-TMA were prepared by ion exchange with Fe-BDC-TPT-Cl and Ni-BDC-TPT-Me
2NH
2, respectively (
Fig. 1).
1H NMR confirmed that the exchange percentage for all samples exceeded 95% (Figs. S1 and S2, Tables S2-S5 in Supporting information). The phase purity of the samples was confirmed by comparing the powder X-ray diffraction (PXRD) patterns with the calculated ones, which were found to be consistent (Fig. S3 in Supporting information). The PXRD patterns of Fe-BDC-TPT-Cl, Fe-BDC-TPT-BF
4, Ni-BDC-TPT-Me
2NH
2, and Ni-BDC-TPT-TMA were subjected to indexing and refinement using the Le Bail method. Tables S6-S9 (Supporting information) presented both the initial and refined parameters, while Figs. S4-S7 (Supporting information) showed the differences between the experimental and fitted PXRD patterns. These results confirm that all materials adopt an isostructural framework.
The permanent porosity of the four materials was confirmed by N2 adsorption at 77 K. The isotherms of Fe-BDC-TPT-Cl, Fe-BDC-TPT-BF4, Ni-BDC-TPT-Me2NH2 and Ni-BDC-TPT-TMA show N2 uptakes of 586.2,448.4,379.3 and 317.5 cm3/g, respectively, at 1 atm, belonging to Type-Ⅰ adsorption isotherms. The N2 uptakes for all the frameworks decreased in the following order: Ni-BDC-TPT-Me2NH2 > Ni-BDC-TPT-TMA, Fe-BDC-TPT-Cl > Fe-BDC-TPT-BF4 (Fig. S8 in Supporting information). This trend is expected to occur due to the increasing size of exchange ions occupying the pore space, which aligns with the following order: TMA+ > Me2NH2+ > BF4− > Cl− (Table S10 in Supporting information). The Brunauer-Emmett-Teller (BET) surface area of Fe-BDC-TPT-Cl, Fe-BDC-TPT-BF4, Ni-BDC-TPT-Me2NH2 and Ni-BDC-TPT-TMA was calculated to be 1677.4, 1307.8, 1236.4, 959.2 m2/g, respectively. As the size of counter ions decreases, the pore size of Ni-BDC-TPT-TMA, Ni-BDC-TPT-Me2NH2, Fe-BDC-TPT-BF4 and Fe-BDC-TPT-Cl increase sequentially, as follows: 4.7, 6.5, 6.7, 8.6 Å (Fig. S9 in Supporting information). The chemical stability of them was tested by soaking in water for 24 h. The crystallinity of these materials remained preserved (Fig. S3). After water treatment, the N2 adsorption capacities of Fe-BDC-TPT-BF4 and Ni-BDC-TPT-TMA decreased to 269.2 and 266.5 cm3/g, respectively, showing reductions of 179.2 and 51.0 cm3/g (Fig. S10 in Supporting information). These results indicate that water has a smaller impact on Ni-BDC-TPT-TMA.
To investigate the adsorption performance of various materials for C
2H
2 and CO
2 both before and after ion exchange, as well as the separation performance of C
2H
2/CO
2, the adsorption isotherms of four materials were performed at different temperatures and 1 bar. For the cationic framework, Fe-BDC-TPT-BF
4 exhibited enhanced gas adsorption capacity after ion exchange. Specifically, Fe-BDC-TPT-BF
4 demonstrated a C
2H
2 adsorption capacity of 203.1 cm
3/g and CO
2 adsorption capacity of 93.2 cm
3/g. In contrast, Fe-BDC-TPT-Cl only had a C
2H
2 adsorption capacity of 140.8 cm
3/g and a CO
2 adsorption capacity of 70.6 cm
3/g. Two anion frameworks also exhibited similar trends (
Figs. 2a and
b). Ni-BDC-TPT-Me
2NH
2 exhibited a C
2H
2 adsorption capacity of 166.0 cm
3/g and CO
2 adsorption capacity of 95.0 cm
3/g. The larger volume of TMA
+ increased the confinement effect, resulting in higher C
2H
2 adsorption capacity: 200.1 cm
3/g for Ni-BDC-TPT-TMA (
Figs. 2c and
d). The comparison of the adsorption capacity of CO
2 and C
2H
2 by the framework before and after ion exchange is shown in
Fig. 2e. After ion exchange, the adsorption performance of the resulting material for both C
2H
2 and CO
2 changed to varying degrees. The narrow pore size and the formation of C—H···X (X = F,
π) interaction with C
2H
2 may enhance the affinity between C
2H
2 molecule and framework, thereby significantly increasing the adsorption capacity of C
2H
2.
By studying the adsorption isotherms at 298 K and 273 K, and fitting the relevant data with the Virial equation, the isosteric adsorption heats (
Qst) of four materials were calculated. As shown in Figs. S11-S18 (Supporting information), the
Qst values for C
2H
2 are 40.7 and 55.3 kJ/mol for Fe-BDC-TPT-Cl and Fe-BDC-TPT-BF
4, respectively. For CO
2, the
Qst values are 9.4 and 28.2 kJ/mol, respectively. The higher
Qst for C
2H
2 compared to CO
2 in the cationic framework system indicates a stronger affinity of the framework for C
2H
2, consistent with the trend observed in the single-component adsorption curves. In the case of Ni-BDC-TPT analogs, Ni-BDC-TPT-TMA shows a significant increase in
Qst for C
2H
2 compared to Ni-BDC-TPT-Me
2NH
2. Specifically, the
Qst values for C
2H
2 are 24.3 kJ/mol and 31.2 kJ/mol for Ni-BDC-TPT-Me
2NH
2 and Ni-BDC-TPT-TMA, respectively. For CO
2, the
Qst values are 20.1 and 22.1 kJ/mol, respectively. The higher
Qst indicates that C
2H
2 molecules exhibit a stronger affinity for the framework, resulting in greater adsorption capacity. The larger difference in
Qst between C
2H
2 and CO
2 for Ni-BDC-TPT-TMA suggests a higher selectivity for C
2H
2 over CO
2. After ion exchange, the adsorption capacity for C
2H
2 in both Fe and Ni analogs increased, however, the corresponding
Qst did not show the same behavior (
Fig. 2f).
The adsorption selectivity of the four materials for C
2H
2/CO
2 (50:50) was calculated using isothermal adsorption curve fitting parameters (Figs. S19-S24 and Tables S11-S14 in Supporting information) and ideal adsorption solution theory (IAST) at 298 K and 1 bar, respectively. The selectivity of Fe-BDC-TPT-BF
4 increases from 2.7 before ion exchange to 4.6 after exchange. Calculations show that the selectivity of the anion framework system for C
2H
2/CO
2 has also improved after ion exchange. The selectivity of Ni-BDC-TPT-Me
2NH
2 and Ni-BDC-TPT-TMA for C
2H
2/CO
2 is 2.5, 4.4, respectively. Different ion exchange percentages of Ni-BDC-TPT-TMA on the adsorption isotherms were also tested. When the exchange percentage is 51.0% and 64.1%, the C
2H
2/CO
2 selectivity is 4.1 and 4.0, respectively (Figs. S25-S29 in Supporting information). These results indicate that optimal separation efficiency for C
2H
2/CO
2 can be achieved through complete exchange with TMA
+. Among them, the selectivity of Fe-BDC-TPT-BF
4 and Ni-BDC-TPT-TMA surpasses that of many other materials for C
2H
2/CO
2 separation, such as CAU-10-H (4.0) [
39], UTSA-222 (4.0) [
40], ZJNU-100 (3.8) [
41], FJU-36a (2.8) [
42], CoV-bdc-tpt (2.6) [
43], FJU-112 (4.2) [
44], SNNU-278 (4.3) [
45], HKUST-1 (2.4) [
46] etc. (Table S6 in Supporting information).
To further investigate the separation performance of Fe-BDC-TPT-Cl, Fe-BDC-TPT-BF
4, Ni-BDC-TPT-Me
2NH
2, Ni-BDC-TPT-TMA for C
2H
2/CO
2, dynamic breakthrough column experiments were conducted at 298 K using an equimolar C
2H
2/CO
2 (50/50, v/v) dry mixture with a flow rate of 2 mL/min. As shown in
Figs. 3a and
b, Fe-BDC-TPT-BF
4 and Ni-BDC-TPT-TMA demonstrate more efficient separation. Specifically, for Fe-BDC-TPT-BF
4, compared with the parent material Fe-BDC-TPT-Cl, the signal of CO
2 is detected faster and the capture yield is slightly improved. The breakthrough interval has increased by nearly 15 min with a separation productivity of 124.6 L/kg. Similarly, Ni-BDC-TPT-TMA also exhibits improved separation performance for C
2H
2/CO
2 compared to Ni-BDC-TPT-Me
2NH
2. The retention time of Ni-BDC-TPT-TMA for C
2H
2 is about 20 min longer than Ni-BDC-TPT-Me
2NH
2, and the separation productivity has increased from 101.0 L/kg to 130.8 L/kg (
Fig. 3c). In terms of comprehensive C
2H
2 adsorption capacity and production, Ni-BDC-TPT-TMA exceeds most similar materials, such as ZJU-50a (85.2 L/kg) [
47], BUT-316 (50.3 L/kg) [
48], SIXSIF-dps-Cu (45.7 L/kg) [
49], SIXSIF-3-Ni (58.4 L/kg) (
Fig. 3d) [
5].
In industrial production, cracking streams inevitably contain small amounts of water vapor, which may affect the separation performance of MOFs. The water vapor adsorption isotherms of several materials were first tested at room temperature. As shown in Figs. S30 and S31 (Supporting information), Fe-BDC-TPT-Cl exhibits high water adsorption at low pressure, in contrast, Fe-BDC-TPT-BF
4 did not show significant water uptake until the
P/
P0 reached 0.2, indicating its lower affinity for water. The water uptake of Ni-BDC-TPT-Me
2NH
2 begins to increase sharply at lower pressures with a
P/
P0 of 0.17, whereas Ni-BDC-TPT-TMA shows a sharp increase at a
P/
P0 of 0.25. These results suggest that the materials became relatively more hydrophobic after ion exchange. To explore the separation ability of Fe-BDC-TPT-BF
4 and Ni-BDC-TPT-TMA under typical working conditions, breakthrough experiments were conducted in humid conditions. From the experimental results (
Figs. 4a-
c), it can be observed that under 35% RH, Fe-BDC-TPT-BF
4 shows that the breakthrough time for CO
2 remains nearly unchanged, while the breakthrough time for C
2H
2 significantly decreases (about 18 min). Despite showing slightly earlier breakthrough times for both CO
2 and C
2H
2 compared to dry feed gas conditions, Ni-BDC-TPT-TMA maintains its C
2H
2 purity and yield without significant decrease. The C
2H
2 adsorption capacity under humid conditions was calculated as 112.7 cm
3/g, which is 86.0% of the capacity for dry gas. And under 35% RH, Ni-BDC-TPT-TMA underwent five breakthrough cycles without a significant decline in separation performance (
Fig. 4d). Desorption of Ni-BDC-TPT-TMA with He at a flow rate of 10 mL/min yielded a C
2H
2 desorption amount of 113.8 cm
3/g, aligning with the calculated value (Fig. S32 in Supporting information). Additionally, the PXRD pattern of Ni-BDC-TPT-TMA sample remained unchanged after the cyclic experiments (Fig. S33 in Supporting information), demonstrating its high durability during the C
2H
2/CO
2 separation process.
In summary, we have varied the countering ions in Fe-BDC-TPT-Cl and Ni-BDC-TPT-Me2NH2 through ion exchange to tune their gas separation properties. Altering Me2NH2+ to TMA+ significantly increases the density of methyl groups within the pores of Ni-BDC-TPT-TMA, leading to a high C2H2 adsorption capacity (200.1 cm3/g at 298 K and 1 bar), along with a commendable C2H2/CO2 selectivity (4.4 at 298 K and 1 bar). More importantly, this modification improves the hydrophobicity of the pores in Ni-BDC-TPT-TMA, exhibiting maintained C2H2/CO2 separation performance and good recyclability in breakthrough experiments under 35% RH. This work has not only contributed new physisorbents through a facile synthesis and modification strategy, but also leveraged the material design principles for better structure functionalization and performance optimization. By tailoring the pores of MOFs, including adding, changing and removing specific entities, we can manipulate the physical and chemical properties of physisorbents to be equipped with more strengths, which are likely to help address the social and industrial challenges, even under harsh conditions.
PDF
Full
60
33
关闭全屏
No. 86 Xueyuan South Road, Haidian District, Beijing
010-62662699(Operation and Technical Support)
qkjq@cast.org.cn
