Design and improvement of porous materials' properties is a frontier research area. Porous materials have versatile applications that include selective adsorption of small gas molecules for their separation or storage (Mondal and Das, 2015; Mane et al., 2017), catalysis (Huang et al., 2018), chemosensing (Gu et al., 2017), energy storage (Liao et al., 2018; Li et al., 2018), and several others (Bhanja et al., 2017; Liu and Liu, 2017; Mitra et al., 2017; Bera et al., 2018a). Metal organic frameworks (MOFs) are porous materials that utilize metal–ligand coordination bonds to stitch monomers to yield coordination polymers. In the past few decades, the scientific community has witnessed extensive research effort in the development of novel porous materials that are derived from purely organic monomers (Cong et al., 2015; Trickett et al., 2017). The resulting materials are best described as porous organic polymers (POPs) or porous organic frameworks (POFs). POPs/POFs are thus synthesized utilizing organic monomers containing lightweight, non-metallic elements (Zhang et al., 2018). The monomers are linked by strong covalent bonds, thereby imparting strength and high thermal/chemical stability to the resulting polymeric materials (Wu et al., 2019). MOFs on the other hand have relatively higher sensitivity to acid/base (lower chemical stability) since monomers (in MOFs) are linked via coordination bonds. Additionally, thermal stability of MOFs is relatively less than POPs/POFs because of the lower strength of ion–dipole interactions (ca. 200–50 kJ mol⁻¹) in the former (MOFs) than covalent bonds (ca. 200–350 kJ mol⁻¹) in the latter (COFs). For superior gas storage applications, it is expected that POPs are microporous in nature. These are also known as MOPs (microporous organic polymers) and polymers of intrinsic microporosity (PIMs) to indicate that the pore widths do not exceed 2 nm (micropores) (Sing, 1985). In current research, there has been a growth in design and construction of microporous organic polymeric materials for potential use in gas separation and storage, as sensors or as catalysts (Das et al., 2017; Dey et al., 2017). In general, polymeric materials that are obtained from rigid monomers (with limited conformational freedom) contain sufficient micropores. The resulting polymers are also highly robust and rigid. The observed microporosity is due to the inefficient packing of the rigid three-dimensional (3D) building blocks that generate voids. Triptycene, the smallest member of the iptycene family, is an example of such a rigid molecular unit with a unique paddle-wheel like arrangement of three phenyl rings fused via a bicyclo[2.2.2]octane unit. As a result, triptycene-based polymers have “internal molecular free volume (IMFV)” and these porous 3D polymeric materials have large surface area and thermal stability (Swager, 2008; Zhang et al., 2016). One of the most promising application of POPs is their potential as sorbent for selective capture of CO₂ gas. This is important because CO₂ being a greenhouse gas is recognized as major contributor to global warming and ocean acidification. One prerequisite desirable in a good CO₂ adsorbent is inclusion of sufficient CO₂-philic sites in the surface of the material. Furthermore, it is well-acknowledged that CO₂-philicity of a POP may be enhanced if the skeletal framework includes nitrogen-based functional groups. Literature reports indicated that porous and nitrogen-rich organic polymer networks have relatively better affinity and comparatively more selectivity toward carbon dioxide over other gases (Shuangzhi et al., 2016). One may attribute this to favorable acid–base interactions between acidic CO₂ molecules and basic N centers with a lone pair of electron. Various synthetic strategies have been utilized to yield N-rich POPs such as use of N-heterocycle-based monomers or amine-based monomers. However, to the best of our knowledge, porous polymers with amine linkages (–NH–) have not been explored for applications such as selective capture of CO₂ or reversible storage of H₂. Synthetic protocols for obtaining such amine (–NH–) cross-linked polymeric networks should be preferentially environmentally benign and avoid use of hazardous reagents or expensive transition metal-based catalysts (Patel et al., 2013). Polymeric materials with aromatic rings connected via amine bridges have been prepared via metal-catalyzed strategies such as the Buchwald–Hartwig (BH) coupling [Pd(II) based catalysts] or Ullmann method [catalytic Cu(I)] to form C–N bonds (Bandyopadhyay et al., 2018; Liao et al., 2018). A greener approach would be to avoid use of transition metal-based catalysts to yield such polymers in which aromatic rings are connected via amine groups.

In this present work, we report synthesis of a set of unique triptycene based POPs with amine linkers (TBPALs) without the use of metal-based catalysts. The objective was to obtain amine group containing polymeric materials as amine functional groups are known to play a significant role in improving the interaction of polymeric materials and CO₂ molecules. The synthetic protocol is facile since it involves condensation of amine (triptycene based triamine/diamine) and inexpensive phenols (such as phloroglucinol/hydroquinone) to yield a set of new nuclophilic amine functionalized polymers (TBPALs). These polymeric materials (TBPAL1–5) have been conveniently synthesized via condensation of triptycene-based triamine/diamine molecules with commercially available and inexpensive phenols such as phloroglucinol/hydroquinone. Following structural characterization, we have recorded their surface area and porous properties. Results obtained were compared with literature-reported POPs that are rich in nitrogen. The performance of TBPAL1–5 as adsorbent for CO₂ was also recorded. To the best of our knowledge, TBPAL1, 2, and 3 display significantly better CO₂/N₂ selectivity at both 273 and 298 K when compared with several nitrogen-rich networks reported to date.

Using nitrogen sorption isotherms (collected at 77 K, Figure 5), the surface area and porosity of the TBPALs were estimated. TBPALs demonstrate substantial N₂ adsorption in the low relative pressure range (P/P₀ = 0–0.01). This suggests that TBPAL1–5 have abundant micropores. In addition, moderately high increase in nitrogen uptake at medium- to high-pressure region advocates the presence of macropores in these TBPALs (Zhang et al., 2014). The surface areas (SAs) of TBPALs were measured using the Brunauer–Emmett–Teller (BET) model (Table 1 and Figure S2). The SA_BET obtained for TBPAL1, TBPAL2, TBPAL3, TBPAL4, and TBPAL5 are 775, 729, 602, 620, and 815 m² g⁻¹, respectively. The corresponding Langmuir surface areas are 1,036, 945, 942, 1,027, and 1,411 m² g⁻¹, respectively (Table 1 and Figure S3). In case of all TBPALs, the hysteresis loop in the respective sorption branches remain almost parallel over a considerable range of P/P₀. This implies that TBPALs display type H4 adsorption–desorption hysteresis.

According to IUPAC, “the type H4 loop appears to be associated with narrow slit-like pores” (Sing, 1982). A comparison with previously reported microporous materials clearly indicates that integration of triptycene units in polymeric framework with amine linkers yield TBPALs that are considerably porous. Additionally, the respective surface areas (SA_BET) are reasonably better or comparable than various previously literature reported nitrogen-rich porous materials. Representative examples of microporous materials with relatively lesser SA_BET than TBPALs are imine-POP (194 m² g⁻¹) (Yu et al., 2018), pyrene-based conjugated POPs (105–136 m² g⁻¹) (Bandyopadhyay et al., 2018), benzidine-based adsorbent material (100 m² g⁻¹) (Taskin et al., 2016), amine-based cross-linked porous polymer (305 m² g⁻¹) (Mahmoud Abdelnaby et al., 2018), g-C3N4-doped amine-rich POP (368–540 m² g⁻¹) (Ou et al., 2018) porous aromatic networks with amine linkers (178–204 m² g⁻¹) (Li et al., 2019), polyethylenimine (PEI)-linked MOPs (65–101 m² g⁻¹) (Mane et al., 2018), phthalocyanine nanoporous polymer (579 m² g⁻¹) (Neti et al., 2015) cyanate resin networks (PCN-TPC, PCN-TPPC, 662–686 m² g⁻¹) (Deng and Wang, 2017), azo-functionalized MOPs (335–706 m² g⁻¹) (Yang et al., 2015), porous polymers derived from Tröger's base (TB-MOP, 694 m² g⁻¹) (Zhu et al., 2013), as well as other MOPs such as triptycene-based polymeric nanosheet (2D-PTNS) (690 m² g⁻¹) (Chen et al., 2017), triptycene-based hexakis (metalsalphens) (356–427 m² g⁻¹) (Reinhard et al., 2018), and BODIPY-containing POPs (482–725 m² g⁻¹) (Xu et al., 2016a).

Pore dimensions of TBPALs were estimated using the N₂ sorption isotherms and by using the density functional theory (DFT) method. Pore size distribution (PSD) profile of TBPALs (Figure 5) reveals that these polymeric networks have pores that are microporous (<2 nm) as well as mesoporous (2–50 nm). Only in the case of TBPAL1 and TBPAL2 have we observed the presence of ultramicropores (<0.7 nm). Moreover, in the case of TBPAL2, there is a greater distribution of narrow ultramicropores than TBPAL1 (Figure S4). Total pore volume of TBPALs was calculated from the volume of N₂ adsorbed at a P/P₀ = 0.99 and was measured to be 0.401 cc/g for TBPAL1, 0.369 cc/g for TBPAL2, 0.446 cc/g for TBPAL3, 0.529 cc/g for TBPAL4, and 0.760 cc/g for TBPAL5 (Table 1). The data obtained also suggest that PSD is comparatively narrow for polymers (TBPAL1 and TBPAL2) that utilize both trifunctional monomers in comparison to polymers (TBPAL3–TBPAL5) having bifunctional comonomers. These results evidently hint that the PSD of TBPALs may be tuned easily by changing the number of reactive sites in the monomers. The microporosity present in TBPALs is ascribed to the incorporation of the rigid 3D triptycene units in the polymer network (Chen et al., 2016; Roy et al., 2017; Bera et al., 2018b, 2019; Tan et al., 2018).

Considering the microporous nature of TBPALs, it was our curiosity to study their ability to act as adsorbent for small gaseous molecule. Therefore, sorption isotherms of CO₂ (298 K), CH₄ (298 K), H₂ (77 K), and N₂ (298 K) were obtained with the pressures up to 1 bar. Given that hydrogen is projected as a non-polluting alternative to fossil fuel, development of materials for H₂ gas storage is an important research area. The H₂ uptake by TBPALs was found to be in the range of 11.4–16.2 mg/g (Table 2 and Figure 6). The highest value was obtained for TBPAL2 (16.2 mg/g), and this is better than various MOPs and MOFs such as azo-bridged porphyrin-based conjugated microporous polymers (8.6–11.5 mg/g) (Xu et al., 2016b), triptycene-based hexakis (metalsalphens) (6–7 mg/g) (Reinhard et al., 2018), 3D ultramicroporous triptycene-based polyimide framework (14.1 mg/g) (Ghanem et al., 2016), rigid triptycene-hexaacid H₆THA (7.8–11.8 mg/g) (Chandrasekhar et al., 2017a), porous hydrogen-bonded MOFs (1.1–1.3 mg/g) (Chandrasekhar et al., 2017b), carbazole-based POPs (11.9–12.9 mg/g) (Zhu et al., 2014), and phthalocyanine-based porous polymer (9 mg/g). (Neti et al., 2015) From the above comparison, it is clear that TBPALs reported herein show H₂ uptake abilities that are better than several microporous polymeric networks reported previously in the literature. The results also reiterate the results of theoretical studies that concluded that networks with triptycene motifs may be well suited for H₂ storage as efficient adsorbents. (Wong et al., 2009) Highest H₂ uptake at low pressure (1 bar) recorded for TBPAL2 relative to other polymers is due to the presence of narrow ultramicropores in TBPAL2, even though its corresponding BET surface area is lesser than some other TBPALs. This is related to the kinetic diameter of H₂ (0.29 nm) (Liu et al., 2015).

Capture of carbon dioxide gas at the polluting source is considered as a possible futuristic technological solution to contain the CO₂ level in the atmosphere with the objective to curtail climatic changes such as global warming due to increasing concentration of greenhouse gases. Therefore, the CO₂ adsorption properties of TBPALs were explored. Characteristic features of CO₂ isotherm of TBPALs are their fully reversible nature (absence of sorption hysteresis) as well as steep rise in the low-pressure region (Figure 7). Reversible isotherm hints at easy regeneration of adsorbent without much energy expenditure. CO₂ adsorption–desorption isotherms of TBPALs showed negligible hysteresis indicating the complete reversibility of CO₂ adsorption, suggesting physisorption process rather than chemisorption of CO₂ (Puthiaraj et al., 2017). TBPAL2 registers the highest CO₂ uptake of 163.7 and 125.0 mg g⁻¹ at 273 and 298 K, respectively, at 1 bar pressure. Presence of considerable narrow ultra micropores in TBPAL2 is also responsible for its efficient CO₂ uptake ability in the low-pressure region, and this feature is related to the kinetic diameter of CO₂ (0.33 nm) (Liu et al., 2015). Nevertheless, the extent of CO₂ uptake by these TBPALs (125–164 mg g⁻¹ at 273 K and 1 bar) is similar or superior to several previously reported POPs in general and N-rich porous polymers in particular. Representative examples of such POPs are amine-based cross-linked porous polymer (66.8 mg/g) (Mahmoud Abdelnaby et al., 2018), 3D ultramicroporous triptycene-based polyimide framework (149.6 mg/g) (Ghanem et al., 2016), triptycene-based hyper-cross-linked polymer sponge (157 mg/g) (Zhang et al., 2015), azo-functionalized MOPs (77.7–134.8 mg/g) (Yang et al., 2015), phthalocyanine nanoporous polymer (157 mg/g) (Neti et al., 2015), porous covalent triazine polymer (CTP) (6.6–29.5 mg/g) (Lee et al., 2018), and triptycene-based hexakis (metalsalphens) (50–58 mg/g) (Reinhard et al., 2018). The observed high CO₂ uptake capacity of TBPALs is due to the high affinity between amine functional groups in the polymeric framework and carbon dioxide molecules via strong dipole–dipole interactions and Lewis acid–base interactions.

We also estimated isosteric heats of CO₂ adsorption (Q_st) for TBPALs at 273 and 298 K using the Clausius–Clapeyron equation. The Q_st values for CO₂ at zero coverage were observed to be in between 30.1 and 34.9 kJ mol⁻¹ (Table 2 and Figure 8). Since the magnitude of Q_st is <40 kJ mol⁻¹, it may concluded that there is reasonable CO₂ physisorption by TBPALs (Patel et al., 2013). The reasonably high Q_st value seen at the onset of adsorption may be assigned to strong interactions between carbon dioxide molecules (Lewis acidic) and polar amine functional groups (Lewis basic). The Q_st values obtained for TBPALs are comparable with various literature reported porous materials such as porous cyanate resin networks (31.0–34.6 kJ mol⁻¹) (Deng and Wang, 2017), porous covalent triazine polymer (CTP) (22.4-34.6 kJ mol⁻¹) (Lee et al., 2018), and triptycene-based hexakis (metalsalphens) (24.4–29.1 kJ mol⁻¹) (Reinhard et al., 2018).

In addition to CO₂ and H₂, TBPALs were also considered as potential sorbent for methane gas. CH₄ gas isotherms of TBPALs were collected at 273 K and 298 K up to 1 bar pressure (Figure 9). In both cases, the isotherms are reversible. At 273 K, TBPAL2 records highest methane uptake (20.3 mg g⁻¹) while TBPAL3 registers the lowest (15.6 mg g⁻¹). We have also investigated the Q_st for methane adsorption in TBPALs and values at very low coverage are in the range of 21.2–30.6 kJ mol⁻¹ (Table 2 and Figure 8). These values suggest physisorption of methane and as good as that observed for various organic porous polymers and MOFs such as phthalocyanine nanoporous polymer (17.8 mg g⁻¹) (Neti et al., 2015), B,N-containing cross-linked polymers (PPs-BN, 15.7–18.44 mg/g) (Zhao et al., 2015), triptycene-based hexakis (metalsalphens) (7–8 mg g⁻¹) (Reinhard et al., 2018), azo-PPors (11.54–12.83 mg g⁻¹) (Jiang et al., 2015), and nanoporous triptycene-based network polyamides (4.0–8.3 mg g⁻¹) (Bera et al., 2017). Figure 10 shows comparative uptake of carbon dioxide, nitrogen, and methane by TBPALs at 273 K.

In summary, we report herein for the first time a set of five triptycene-based porous polymers (TBPALs) with hierarchical porosity in which the triptycene units are interconnected via –NH– functional groups. Such unique triptycene networks cross-linked by amine functional groups were conveniently synthesized in high yields from commercially available phloroglucinol as one of the monomers. Formation of C–N bonds did not require the use of transition metal-based catalysts, such as the ones used in Buchwald-Hartwig cross-coupling reaction or the Ullmann method. Important properties of such TBPALs are good thermal stability with improved surface area (SA_BET up to 815 m² g⁻¹) relatively to previously reported porous aromatic networks linked via amine units. For the first time, CO₂/N₂ selectivity has been measured for such amine-linked POPs, and these values are in the range of 64–92 (at 273 K). Again, for the first time, these amine-functionalized and triptycene-based POPs were explored as materials for H₂ storage. Highest H₂ uptake (16.2 mg g⁻¹) was registered by TBPAL2 and that is comparable with several hyper cross-linked polymers (HCPs) reported previously. These results suggest that amine-based polymers with robust and rigid units (such as triptycene) are a new addition to the family of nanoporous materials that display considerably high CO₂ uptake with reasonably good CO₂/N₂ selectivities for CO₂ capture applications. These results motivate us and others to develop new polyamine porous polymers with promising applications as smart materials in environment/alternative energy sectors.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation, to any qualified researcher.

ND conceived the research and supervised the experimental work related to synthesis and characterization reported herein. AA synthesized all polymeric networks reported in this manuscript. MA and AH assisted AA in the synthesis of literature reported triptycene monomers. AA characterized the polymers and was assisted by RB. All authors have contributed to interpretation of results, compiled the manuscript, and have approved the final manuscript.