New Porous Silicon-Containing Organic Polymers: Synthesis and Carbon Dioxide Uptake

: The design and synthesis of new multifunctional organic porous polymers has attracted signiﬁcant attention over the years due to their favorable properties, which make them suitable for carbon dioxide storage. In this study, 2-, 3-, and 4-hydroxybenzaldehyde reacted with phenyltrichlorosilane in the presence of a base, a ﬀ ording the corresponding organosilicons 1–3 , which further reacted with benzidine in the presence of glacial acetic acid, yielding the organic polymers 4–6 . The synthesized polymers exhibited microporous structures with a surface area of 8.174–18.012 m 2 g − 1 , while their pore volume and total average pore diameter ranged from 0.015–0.035 cm 3 g − 1 and 1.947–1.952 nm, respectively. In addition, among the synthesized organic polymers, the one with the meta -arrangement structure 5 showed the highest carbon dioxide adsorption capacity at 323 K and 40 bar due to its relatively high surface area and pore volume.


Introduction
Fossil fuels are under pressure due to the constantly increasing energy demand in many industrial applications. However, their combustion leads to high levels of greenhouse gas (e.g., carbon dioxide (CO 2 )) emissions [1], which in turn cause several environmental problems including weather changes and global warming [2]. Therefore, the capture of CO 2 plays a vital role in reducing the negative effects of global warming [3][4][5][6]. The efficient capture and storage of CO 2 requires its separation from other gases, such as hydrogen (H 2 ), methane (CH 4 ), and nitrogen (N 2 ), which can be achieved under high-pressure conditions during pre-combustion or at atmospheric pressure and temperature during post-combustion [7]. The storage process involves the adsorption of CO 2 followed by desorption of the gas in a pure form.
To date, the commercial scale processes for the separation of CO 2 from gas flue have mainly focused on the use of amines, aqueous ammonia, and potassium carbonate as absorbents [8,9]. Although the use of chilled ammonia is simple and easy and generates ammonium carbonate as a solid at a relatively low temperature (<20 • C) [9], the process requires significantly high energy levels. Alternatively, a mixture of polyethylene glycol and dimethyl ether has been used as a CO 2 absorbent [10], as it requires low energy for the generation step. However, this method is suitable only for streams with high pressure. Ionic liquids also require low energy for the solvent regeneration and can absorb CO 2 through physisorption [11], but have high viscosity, which limits the absorption rate of CO 2 [12]. In contrast, solid CO 2 adsorbents are more beneficial in terms of energy efficiency compared to absorbents, as they can form van der Waals (physisorption) or covalent (chemisorption) bonds with CO 2 [13].
Porous materials, such as activated carbons, zeolites, molecular sieves, and metal oxides, are common CO 2 physical adsorbents [14] and their properties can be easily modified by incorporating several functional groups [15]. In particular, porous materials should have low density and high CO 2 adsorption capacity, surface area, and physical and chemical stability under practical conditions [16]. Porous organic polymers (POPs) have high CO 2 adsorption efficiency and stability in harsh environments, well-defined structure of porous networks, good lifetime, and recyclability, which make them suitable materials for gas storage [17]. Hence, the development of new POPs as gas storage media, heterogeneous catalysts, molecular sensors, light harvesters, and other applications has recently attracted considerable attention [18][19][20][21][22][23][24][25]. POPs can be synthesized either through simple, convenient, and high yielding synthetic procedures or by using suitable monomers [18], which may bear multifunctional groups that lead to high-porosity materials with different geometries [26][27][28][29][30]. For example, MCM-41, produced from pulverized coal fly ash, has a large pore volume and acts as an excellent CO 2 storage medium in the presence of amine [31]. The implementation of both strategies allows full control of both the functionality and the porosity of the synthesized POPs. Furthermore, it has been reported that porous polymeric materials with tetrahedral geometry have a high surface area, porosity, rigidity, and gas adsorption capacity [32][33][34][35][36][37].
In this regard, our group has recently designed, synthesized, and examined the application of various materials as gas storage media [38][39][40][41][42][43]. Herein, we synthesized a series of new polymeric materials that contain silicon and a Schiff base moiety to investigate their properties and applicability as CO 2 storage media. Specifically, three silicon-containing polymers were successfully prepared with a porous amorphous surface and different particle diameters. Further characterizations revealed that all polymers were thermally stable at temperatures of up to 400 • C and could reversibly adsorb CO 2 within their pores. Finally, the estimation of the CO 2 uptake by the three polymers revealed that their pore size distribution, surface area, and geometry could significantly affect their CO 2 storage capacity.

Instrumentation
The FTIR and 1 H NMR spectra (500 MHz) were recorded on a Shimadzu 8400 spectrophotometer (Shimadzu, Tokyo, Japan) and a Bruker DRX500 NMR spectrometer (Bruker, Zürich, Switzerland), respectively. Energy dispersive X-ray spectroscopy was analyzed on a Bruker XFlash 610. The field emission scanning electron microscopy (FESEM) images were obtained using a SIGMA 500 VP microscope (ZEISS Microscopy, Jena, Germany), while the X-ray powder diffraction (XRD) spectra were measured on an ADX-2500 X-ray diffraction instrument (Angstrom Advanced, Inc., Stoughton, MA, USA). The thermogravimetric analysis (TGA) was performed on a TGA4000 thermogravimetric analyzer (PerkinElmer, Waltham, MA, USA) and the differential scanning calorimetry (DSC) analysis was carried out on an ASTM E1356 calorimeter (Intertek, Wilton UK). A Quantchrome analyzer was used to record the nitrogen (N 2 ) adsorption-desorption isotherms at 77 K, while the surface area and pore sizes were calculated using the Brunauer-Emmett-Teller (BET) method. The pore volumes were determined at a relative pressure (p/p • ) of 0.98. The CO 2 uptake was estimated on an H-sorb 2600 high-pressure volumetric adsorption analyzer (Gold APP Instrument Corporation, Beijing, China) at 323 K and 40 bar. Before the measurements, the polymers were dried at 150 • C under inert atmosphere (N 2 ) for 5 h and degassed at 50 • C under dynamic vacuum for 1 h. A standard procedure was used for the preparation of sample for adsorption experiments and the samples were sputter coated by gold (ca. 15 nm).

Synthesis of Organosilicons 1-3
To a stirred solution of 2-, 3-, or 4-hydroxybenzaldehyde (3.66 g, 30 mmol) in dry pyridine (20 mL) at 0 • C under nitrogen atmosphere, a solution of phenyltrichlorosilane (2.11 g, 10 mmol) in dry tetrahydrofuran (THF; 10 mL) was added dropwise over 30 min through a dropping funnel. The mixture was stirred at 40-45 • C for 6 h and then was allowed to cool to room temperature. The generated solid was removed by filtration and washed with THF (3 × 25 mL). The filtrate was then concentrated under reduced pressure, affording a yellow oil, which was washed with cold distilled water (2 × 10 mL) and dried under reduced pressure at 60 • C for 2 h to give the corresponding organosilicon 1, 2, or 3.

Synthesis of Polymers 4-6
A mixture of 1, 2, or 3 (4.65 g, 10 mmol) and benzidine (2.76 g, 15 mmol) in THF (25 mL) containing glacial acetic acid (AcOH; 0.5 mL) was stirred under reflux for 5 h under nitrogen atmosphere. The mixture was allowed to cool to room temperature, generating a solid, which was afterwards filtered, washed with THF (2 × 10 mL) and hexane (3 × 10 mL), and dried under reduced pressure for 2 h to give the corresponding polymer 4, 5, or 6 as an orange powder.

Synthesis of Organosilicons 1-3
To a stirred solution of 2-, 3-, or 4-hydroxybenzaldehyde (3.66 g, 30 mmol) in dry pyridine (20 mL) at 0 °C under nitrogen atmosphere, a solution of phenyltrichlorosilane (2.11 g, 10 mmol) in dry tetrahydrofuran (THF; 10 mL) was added dropwise over 30 min through a dropping funnel. The mixture was stirred at 40-45 °C for 6 h and then was allowed to cool to room temperature. The generated solid was removed by filtration and washed with THF (3 × 25 mL). The filtrate was then concentrated under reduced pressure, affording a yellow oil, which was washed with cold distilled water (2 × 10 mL) and dried under reduced pressure at 60 °C for 2 h to give the corresponding organosilicon 1, 2, or 3.

Synthesis of Polymers 4-6
A mixture of 1, 2, or 3 (4.65 g, 10 mmol) and benzidine (2.76 g, 15 mmol) in THF (25 mL) containing glacial acetic acid (AcOH; 0.5 mL) was stirred under reflux for 5 h under nitrogen atmosphere. The mixture was allowed to cool to room temperature, generating a solid, which was afterwards filtered, washed with THF (2 × 10 mL) and hexane (3 × 10 mL), and dried under reduced pressure for 2 h to give the corresponding polymer 4, 5, or 6 as an orange powder.

Scheme 1. Synthesis of organosilicons 1-3.
A strong absorption band was observed at 1686-1713 cm −1 in the FTIR spectra of 1-3 (Figures S1-S3), which was attributed to the aldehyde carbonyl group, while no absorption band corresponding to the hydroxyl group of the hydroxybenzaldehydes was detected, implying the effective reaction of the starting aldehydes. The formation of the desired organosilicons was confirmed by the appearance of the characteristic peak of the Si-O group at 1161-1174 cm −1 . The most common FTIR absorption bands detected for 1-3 are reported in Table 1.
Moreover, a singlet peak was observed at 9.93-9.91 μmol/mol in each of the 1 H NMR spectra of 1-3 ( Figures S4-S6), which corresponded to the protons of the aldehyde groups, while the protons of the aromatic moieties were also detected ( Table 2). In particular, the structure of 2 was further confirmed by the 13 C NMR spectroscopic data ( Figure S7), where a very strong peak was observed at A strong absorption band was observed at 1686-1713 cm −1 in the FTIR spectra of 1-3 (Figures S1-S3), which was attributed to the aldehyde carbonyl group, while no absorption band corresponding to the hydroxyl group of the hydroxybenzaldehydes was detected, implying the effective reaction of the starting aldehydes. The formation of the desired organosilicons was confirmed by the appearance of the characteristic peak of the Si-O group at 1161-1174 cm −1 . The most common FTIR absorption bands detected for 1-3 are reported in Table 1. Moreover, a singlet peak was observed at 9.93-9.91 µmol/mol in each of the 1 H NMR spectra of 1-3 ( Figures S4-S6), which corresponded to the protons of the aldehyde groups, while the protons of the aromatic moieties were also detected ( Table 2). In particular, the structure of 2 was further confirmed by the 13 C NMR spectroscopic data ( Figure S7), where a very strong peak was observed at 191.3 µmol/mol, corresponding to the carbonyl group of aldehyde, while all the expected aromatic carbons were identified within the 153.9-116.8 µmol/mol region.

Synthesis of Polymers 4-6
The reaction of 1-3 and benzidine in boiling THF in the presence of glacial acetic acid for 5 h gave the corresponding polymers 4-6 in 79-85% yields (Scheme 2). The FTIR spectra of 4-6 (Figures S8-S10) indicated the absence of the carbonyl group, clearly implying the polymerization of 1-3. Additional absorption bands were also detected with the 1605-1647 cm −1 region, which were attributed to the azomethane bond (C=N). The most common FTIR absorption bands of 4-6 and their melting points, yields, and color are shown in Table 3.

Synthesis of Polymers 4-6
The reaction of 1-3 and benzidine in boiling THF in the presence of glacial acetic acid for 5 h gave the corresponding polymers 4-6 in 79-85% yields (Scheme 2). The FTIR spectra of 4-6 (Figures S8-S10) indicated the absence of the carbonyl group, clearly implying the polymerization of 1-3. Additional absorption bands were also detected with the 1605-1647 cm −1 region, which were attributed to the azomethane bond (C=N). The most common FTIR absorption bands of 4-6 and their melting points, yields, and color are shown in Table 3.  To further confirm the formation of the desired polymers, the elemental composition of 4-6 was determined by energy-dispersive X-ray spectroscopic measurements ( Figures S11-S13), indicating the presence of carbon, nitrogen, oxygen, and silicon in 84.4-85.6, 6.5-7.1, 6.5-7.0, and 1.0-1.5%, respectively.

Surface Morphology of Polymers 4-6
The FESEM technique provides not only less distorted, but also high-resolution pictures for the surface of the examined material [44]. Thus, it was used in this study to explore the surface morphology of the synthesized polymers 4-6 ( Figure 1 and Figures S14-S16). To further confirm the formation of the desired polymers, the elemental composition of 4-6 was determined by energy-dispersive X-ray spectroscopic measurements ( Figures S11-S13), indicating the presence of carbon, nitrogen, oxygen, and silicon in 84.4-85.6, 6.5-7.1, 6.5-7.0, and 1.0-1.5%, respectively.

Surface Morphology of Polymers 4-6
The FESEM technique provides not only less distorted, but also high-resolution pictures for the surface of the examined material [44]. Thus, it was used in this study to explore the surface morphology of the synthesized polymers 4-6 ( Figure 1 and Figures S14-S16).  In particular, the FESEM images shown in Figure 1 indicated that 4-6 were porous materials and irregular in shape, uniform, and contained cracks within smoother surfaces. However, despite their similar surface morphology, the particles within the surface had different shape, size, and diameter. Specifically, the particle diameter of polymers 4, 5, and 6 was 35.3-109.8, 40.2-178.4, and 45.2-208.2 nm, respectively.
The X-ray powder diffraction was also used to examine the surface of the synthesized polymers 4-6, as it can provide useful information about the crystallinity, defects, average grain size, and strain of the examined material [45]. As shown in Figure 2, polymers 4-6 had amorphous and very similar structures, which contained an aura that was described as a broad hump. There was no degree of any crystalline form within the polymeric structure. The amorphous structure could result from the sliding of the layers, which destructed the crystalline forms. It should also be noted that there were no peaks indicating a specific recurring shape within the structures. In particular, the FESEM images shown in Figure 1 indicated that 4-6 were porous materials and irregular in shape, uniform, and contained cracks within smoother surfaces. However, despite their similar surface morphology, the particles within the surface had different shape, size, and diameter. Specifically, the particle diameter of polymers 4, 5, and 6 was 35.3-109. 8 The X-ray powder diffraction was also used to examine the surface of the synthesized polymers 4-6, as it can provide useful information about the crystallinity, defects, average grain size, and strain of the examined material [45]. As shown in Figure 2, polymers 4-6 had amorphous and very similar structures, which contained an aura that was described as a broad hump. There was no degree of any crystalline form within the polymeric structure. The amorphous structure could result from the sliding of the layers, which destructed the crystalline forms. It should also be noted that there were no peaks indicating a specific recurring shape within the structures.

TGA and DSC of Polymers 4-6
TGA is mainly used to explore the thermal stability and composition of polymers [46]. When the temperature rises, polymers may lose weight, and volatile components, such as low molecular weight oligomers, moisture, or residual solvents, can be abstracted. It is also known that the degradation of

TGA and DSC of Polymers 4-6
TGA is mainly used to explore the thermal stability and composition of polymers [46]. When the temperature rises, polymers may lose weight, and volatile components, such as low molecular weight oligomers, moisture, or residual solvents, can be abstracted. It is also known that the degradation of polymers at high temperatures leads to the formation of volatile components due to chain scission, elimination of side groups, and unzipping to monomers [47,48]. Thus, the weight loss of the synthesized polymeric materials could be estimated by TGA. Figure 3 shows a significant weight loss was observed between 285 • C and 400 • C ( Table 4), suggesting that the polymers 4-6 consisted of only one phase due to the lack of secondary products. DSC is also used to measure various thermal parameters of materials [49]. The glass transition (T g ) and melting (T m ) temperatures for polymers 4-6 were determined from the DSC measurements (Table 4, Figure 3). According to these results, we concluded that polymers 4-6 are amorphous and thermally stable materials at a temperature up to 290 • C and have endothermic peaks. elimination of side groups, and unzipping to monomers [47,48]. Thus, the weight loss of the synthesized polymeric materials could be estimated by TGA. Figure 3 shows a significant weight loss was observed between 285 °C and 400 °C ( Table 4), suggesting that the polymers 4-6 consisted of only one phase due to the lack of secondary products. DSC is also used to measure various thermal parameters of materials [49]. The glass transition (Tg) and melting (Tm) temperatures for polymers 4-6 were determined from the DSC measurements (Table 4, Figure 3). According to these results, we concluded that polymers 4-6 are amorphous and thermally stable materials at a temperature up to 290 °C and have endothermic peaks.

Surface Area and Porosity of Polymers 4-6
The BET method is generally used to estimate the surface area of the examined adsorbents [50,51]. The collected data are displayed in the form of a N 2 adsorption isotherm, which is obtained by plotting the quantity of the adsorbed gas against the relative pressure. The N 2 adsorption-desorption measurements of polymers 4-6 at 77 K were of type III with no single-layer adsorption and their average pore diameter and total volume were less than 2 nm (i.e., microporous materials) and 0.015-0.034 cm 3 g −1 , respectively ( Table 5). The N 2 adsorption-desorption and pore size distribution isotherms of polymers 4-6 are shown in Figures 4 and 5. The BET surface area (S BET ) of polymers 4, 5, and 6 was relatively low (Table 5). Clearly, both the surface area and pore volume of polymers 4-6 are lower compared with those for activated carbons [52].

Surface Area and Porosity of Polymers 4-6
The BET method is generally used to estimate the surface area of the examined adsorbents [50,51]. The collected data are displayed in the form of a N2 adsorption isotherm, which is obtained by plotting the quantity of the adsorbed gas against the relative pressure. The N2 adsorptiondesorption measurements of polymers 4-6 at 77 K were of type III with no single-layer adsorption and their average pore diameter and total volume were less than 2 nm (i.e., microporous materials) and 0.015-0.034 cm 3 g −1 , respectively ( Table 5). The N2 adsorption-desorption and pore size distribution isotherms of polymers 4-6 are shown in Figures 4 and 5. The BET surface area (SBET) of polymers 4, 5, and 6 was relatively low (Table 5). Clearly, both the surface area and pore volume of polymers 4-6 are lower compared with those for activated carbons [52]. Table 5. Textural properties of polymers 4-6.
Polymer SBET (m 2 g −1 ) a Total Volume (cm 3 g −1 ) b Pore Size (nm) c,d 4 12     Porous materials are often described based on the size of the pores. However, the reported average pore size cannot be used to describe the porosity of the synthesized polymers. The synthesized POPs have mainly microporous structures, along with mesoporous, macroporous, and other materials.

CO2 Adsorption of Polymers 4-6
The CO2 adsorption of polymers 4-6 was investigated at 323 K and 40 bar. High temperature and pressure were used in order to maximize the CO2 uptake by polymers based on our previous work [40][41][42][43]. The CO2 adsorption isotherms for the adsorbed CO2 within the polymer pores under the applied temperature (323 K) and pressure (40 bar) conditions are shown in Figure 6. Polymers 4-6 showed a reasonable CO2 uptake in the range of 11.295-30.581 cm 3 g −1 (Table 6), while polymer 5 with the meta-arrangement exhibited the highest CO2 uptake (6.0 wt%), probably due to its relatively Porous materials are often described based on the size of the pores. However, the reported average pore size cannot be used to describe the porosity of the synthesized polymers. The synthesized POPs have mainly microporous structures, along with mesoporous, macroporous, and other materials.

CO 2 Adsorption of Polymers 4-6
The CO 2 adsorption of polymers 4-6 was investigated at 323 K and 40 bar. High temperature and pressure were used in order to maximize the CO 2 uptake by polymers based on our previous work [40][41][42][43]. The CO 2 adsorption isotherms for the adsorbed CO 2 within the polymer pores under the applied temperature (323 K) and pressure (40 bar) conditions are shown in Figure 6. Polymers 4-6 showed a reasonable CO 2 uptake in the range of 11.295-30.581 cm 3 g −1 (Table 6), while polymer 5 with the meta-arrangement exhibited the highest CO 2 uptake (6.0 wt%), probably due to its relatively high surface area (18.012 m 2 g −1 ) and total pore volume (0.034 cm 3 g −1 ) compared to the other two polymers with orthoand para-arrangements. Similar findings have been reported for polyphosphates containing 1,4-diaminobenzene and benzidine [41,42]. It was thus clearly indicated that the pore size distribution and surface area can affect the CO 2 storage capacity of polymers 4-6 under the applied conditions. Since these parameters are probably mainly controlled by the geometry of the polymer building blocks, the higher CO 2 adsorption capacity of 5 could also be explained. conditions. Since these parameters are probably mainly controlled by the geometry of the polymer building blocks, the higher CO2 adsorption capacity of 5 could also be explained. The adsorbent-adsorbate interactions were relatively weak and the gas was clustered around the -CH=N group within the polymer surface [53]. The surface area of polymeric materials is not the only contributing factor to the adsorption of CO2. However, polar groups such as -CH=N, -OH, NH2, and O=C-NH within the skeleton of the POPs can lead to a high CO2 uptake by tuning the isoelectric head of adsorption [54][55][56][57][58].  The storage capacity of POPs 4-6 was also comparable to that reported for metal complexes containing valsartan [39], and telmisartan [40], and for polyphosphates containing 1,4diamainobenzene [41]. Instead, polyphosphates consisting of benzidine [42] and melamine Schiff bases [43] showed better CO2 uptake than 4-6, possibly due to the tunable pore size of their particles and the presence of functional groups. Nevertheless, in this case, it should be considered that a direct comparison between phosphate-and silicate-containing polymers is difficult. The newly synthesized POPs are easy to produce, have high thermal stability, and their surface area and pore structures can be tuned to maximize CO2 uptake. Therefore, these materials have potential to be used in gas capture applications.

Conclusions
Three silicon-containing POPs with ortho-, meta-, and para-arrangements were synthesized and their structures and properties were established. The synthesized porous polymers had different properties and adsorption capacity toward CO2 due to their different surface area and pore size distribution. Among the synthesized POPs, the meta-arranged polymer (5) exhibited the highest CO2 uptake, as it had a relatively high surface area and pore total volume compared to polymers 4 and 6.  The adsorbent-adsorbate interactions were relatively weak and the gas was clustered around the -CH=N group within the polymer surface [53]. The surface area of polymeric materials is not the only contributing factor to the adsorption of CO 2 . However, polar groups such as -CH=N, -OH, NH 2 , and O=C-NH within the skeleton of the POPs can lead to a high CO 2 uptake by tuning the isoelectric head of adsorption [54][55][56][57][58].
The storage capacity of POPs 4-6 was also comparable to that reported for metal complexes containing valsartan [39], and telmisartan [40], and for polyphosphates containing 1,4-diamainobenzene [41]. Instead, polyphosphates consisting of benzidine [42] and melamine Schiff bases [43] showed better CO 2 uptake than 4-6, possibly due to the tunable pore size of their particles and the presence of functional groups. Nevertheless, in this case, it should be considered that a direct comparison between phosphate-and silicate-containing polymers is difficult. The newly synthesized POPs are easy to produce, have high thermal stability, and their surface area and pore structures can be tuned to maximize CO 2 uptake. Therefore, these materials have potential to be used in gas capture applications.

Conclusions
Three silicon-containing POPs with ortho-, meta-, and para-arrangements were synthesized and their structures and properties were established. The synthesized porous polymers had different properties and adsorption capacity toward CO 2 due to their different surface area and pore size distribution. Among the synthesized POPs, the meta-arranged polymer (5) exhibited the highest CO 2 uptake, as it had a relatively high surface area and pore total volume compared to polymers 4 and 6.