Synthesis, Characterization, and Gas Adsorption Performance of Amine-Functionalized Styrene-Based Porous Polymers

In recent years, porous materials have been extensively studied by the scientific community owing to their excellent properties and potential use in many different areas, such as gas separation and adsorption. Hyper-crosslinked porous polymers (HCLPs) have gained attention because of their high surface area and porosity, low density, high chemical and thermal stability, and excellent adsorption capabilities in comparison to other porous materials. Herein, we report the synthesis, characterization, and gas (particularly CO2) adsorption performance of a series of novel styrene-based HCLPs. The materials were prepared in two steps. The first step involved radical copolymerization of divinylbenzene (DVB) and 4-vinylbenzyl chloride (VBC), a non-porous gel-type polymer, which was then modified by hyper-crosslinking, generating micropores with a high surface area of more than 700 m2 g−1. In the following step, the polymer was impregnated with various polyamines that reacted with residual alkyl chloride groups on the pore walls. This impregnation substantially improved the CO2/N2 and CO2/CH4 adsorption selectivity.


Introduction
Anthropogenic carbon dioxide is considered among the main cause of the greenhouse effect responsible for global warming. More than 65% of anthropogenic CO 2 is generated from burning fossil fuels during the energy production. To limit the increase in the Earth's temperature, reducing greenhouse gas emissions is necessary. Therefore, development of new technologies for producing clean energy with minimal CO 2 emissions is necessary. CO 2 can be reduced using carbon capture and storage (CCS) technologies based on the capture, transport, and storage of CO 2 [1,2].
CCS technologies are considered potentially promising in reducing CO 2 emissions, and thereby relieving climate change [3,4]. Adsorption is considered an effective alternative for CO 2 capture, offering possible energy savings compared to other established separation technologies [5,6].
In recent years, the research interest in solid adsorbent materials has increased. Compared to traditional separation procedures, they are easy to operate and offer relatively low energy requirements [7].
Considering the direction of society towards a circular economy of materials, the use of plastic waste as a source of functional materials for various applications has become an important challenge for the future [12][13][14].

Synthesis of Porous Polymer
A hyper-crosslinked VBC-DVB microporous polymer was prepared and functionalized in two steps, as shown in the schema in Figure 1a. The first step involved the suspension polymerization of DVB and 4-vinylbenzyl chloride (VBC). The resulting polymer was then hyper-crosslinked under the conditions shown in Figure 1a. During this operation, certain chloromethyl groups within the polymer network were transformed into crosslinked methylene bridges. This additional crosslinking (hyper-crosslinking) induces phase separation within the polymer gel, resulting in the creation of pores between the hyper-crosslinked nodules. However, the conversion of a chloromethyl group into a methylene bridge can proceed only if a neighboring polymer chain is available in the vicinity of the chloromethyl group. The probability of finding such a reaction partner in the surface layer of the nodules, which is on the walls of the newly created pores, is low; hence, unreacted chloromethyl groups are available for further modification on the pore walls [24]. In the second step, various polyamine species were applied to react with the alkyl chloride groups and functionalize the pore surface to improve the selectivity of the material for CO 2 (see Figure 1b).

Preparation of Hyper-Crosslinked Porous Polymer Sample (HCLPP)
The polymerization was performed as follows: 20.2 g (130 mmol) of VBC was mixed with 1 g of technical grade (80%) DVB (corresponding to 5% crosslinking of the polymer matrix) and 0.6 g of AIBN (initiator) was added. An aqueous solution (1000 mL) containing 10 g of polyvinyl alcohol (Mowiol 4-88, MW = 31000, Sigma Aldrich, Prague, Czech  The polymerization was performed as follows: 20.2 g (130 mmol) of VBC was mixed with 1 g of technical grade (80%) DVB (corresponding to 5% crosslinking of the polymer matrix) and 0.6 g of AIBN (initiator) was added. An aqueous solution (1000 mL) containing 10 g of polyvinyl alcohol (Mowiol 4-88, MW = 31000, Sigma Aldrich, Prague, Czech Republic) and 50 g of sodium chloride was introduced into the reactor. The polymerization mixture was poured, the reactor was closed, and the mixture was stirred at 400 rpm. After one hour of stirring at room temperature, the temperature was raised to 80 • C, and the mixture was allowed to react for 24 h. Then, the reactor was opened, and the polymer was separated in a sieve, washed with ethanol, and dried. The resulting partially bonded balls were broken in a grinding mill.
Hyper-crosslinking: 15 g of polymer was suspended in 100 mL of dichloroethane. After 30 min, 20 mL of stannic chloride was added and the mixture was stirred for 5 h and then left for another 24 h. The polymer was then filtered and washed thoroughly with ethanol.

Functionalization of HCLPP
The prepared hyper crosslinked polymer 0.4 g was suspended in 20 mL of acetonitrile. Amine group-containing reagents (ethylenediamine EDA, diethylenetriamine DETA) or dendrimers (see Figure 1b, left) 0.2 g were added dropwise (0.1 mg in 2 mL of acetonitrile). Reaction mixture was then stirred at 50 • C overnight (12 h). After cooling to room temperature, the reaction mixture was centrifuged, washed twice with acetonitrile, and then dried under reduced pressure. Subsequently, the solid was suspended in 5 mL of 10% K 2 CO 3 . After centrifugation, the samples were washed twice with water and dried under reduced pressure.

Characterization 2.3.1. Properties of the HCLPPs
All forms of the prepared hyper-crosslinked materials were characterized using the following methods: Textural characteristics, including the specific surface area, pore volume, and pore diameter distribution, were determined based on nitrogen adsorption-desorption isotherms measured at 77 K using a commercial automated volumetric gas adsorption analyzer ASAP 2050 (Micromeritics, Norcross, GA, USA). Before the measurements, the samples were degassed under deep vacuum at 373 K for 4 h. The morphology and elemental composition were evaluated using a scanning electron microscope (SEM) equipped with an energy-dispersive X-ray (EDX) spectrometer (Quantax, Bruker, Bremen, Germany) in the range of 5-30 kV. Using this technique, the films were analyzed at an accelerating voltage of 15 kV. Thermogravimetric analysis (TGA) of samples was performed using a TG 209 F1 Libra equipment (NETZSCH, Selb, Germany) in the temperature range from 25 • C to 1000 • C in an inert argon atmosphere with a flow rate of 50 mL·min −1 . The samples were heated from ambient temperature to 1000 • C at a heating rate of 4 • C min −1 , the weight of each sample was approximately 7 mg.

Characterization of Dendrimers
NMR spectra were measured using a Bruker Avance 400, Bruker, Bremen, Germany ( 1 H at 400.1 MHz; 13 C at 100.6 MHz; 29 Si { 1 H} (INEPT technique) at 79.5 MHz) at 25 • C. 1 H and 13 C NMR signals of the prepared compounds were assigned to the corresponding atoms using gHSQC, gCOSY, and gHMBC 2D NMR correlation spectra. 1 H and 13 C chemical shifts (δ/ppm) are given relative to residual solvent signals (δ H /δ C : DMSO-d6 2.50/39.52); 29 Si spectra were referenced to an external standard hexamethyldisilane (−19.87 ppm). HRMS spectra were measured using a MicroTOF-QIII instrument (Bruker Daltonics, Bruker, Bremen, Germany) with an ESI or APCI ionization source in the positive mode.

Gas Adsorption Measurement
The adsorption isotherms of the tested gases were measured using the volumetric method. The adsorption of pure gas in the prepared materials was measured using a simple apparatus shown in Figure 2 constructed in our laboratory.

Properties of the HCLPPs
All forms of the prepared hyper-crosslinked materials were characterized using the following methods: Textural characteristics, including the specific surface area, pore volume, and pore diameter distribution, were determined based on nitrogen adsorption-desorption isotherms measured at 77 K using a commercial automated volumetric gas adsorption analyzer ASAP 2050 (Micromeritics, Norcross, GA, USA). Before the measurements, the samples were degassed under deep vacuum at 373 K for 4 h. The morphology and elemental composition were evaluated using a scanning electron microscope (SEM) equipped with an energy-dispersive X-ray (EDX) spectrometer (Quantax, Bruker, Bremen, Germany) in the range of 5-30 kV. Using this technique, the films were analyzed at an accelerating voltage of 15 kV. Thermogravimetric analysis (TGA) of samples was performed using a TG 209 F1 Libra equipment (NETZSCH, Selb, Germany) in the temperature range from 25 °C to 1000 °C in an inert argon atmosphere with a flow rate of 50 mL⋅min −1 . The samples were heated from ambient temperature to 1000 °C at a heating rate of 4 °C min −1 , the weight of each sample was approximately 7 mg.  13 C NMR signals of the prepared compounds were assigned to the corresponding atoms using gHSQC, gCOSY, and gHMBC 2D NMR correlation spectra. 1 H and 13 C chemical shifts (δ/ppm) are given relative to residual solvent signals (δH/δC: DMSO-d6 2.50/39.52); 29 Si spectra were referenced to an external standard hexamethyldisilane (−19.87 ppm). HRMS spectra were measured using a MicroTOF-QIII instrument (Bruker Daltonics, Bruker, Bremen, Germany) with an ESI or APCI ionization source in the positive mode.

Gas Adsorption Measurement
The adsorption isotherms of the tested gases were measured using the volumetric method. The adsorption of pure gas in the prepared materials was measured using a simple apparatus shown in Figure 2 constructed in our laboratory.  The thermostated part of the apparatus consisted of two chambers with calibrated volumes, V 1 and V 2 , separated by a valve. The pressure in the apparatus was measured Polymers 2023, 15, 13 6 of 13 using a pressure transducer. A heated box maintains a constant temperature of the entire system during the absorption experiment.
Before the experiment, a weighed amount of the sample was placed in the adsorption chamber, V 2 . First, to remove dissolved gases and impurities, the apparatus chambers were evacuated at least 5 h at temperature 80 • C. Subsequently, the system temperature was set to the thermostat at the required experimental value. Chamber V 2 was closed, while volume V 1 was filled with the studied gas at the desired pressure p 1 (from zero to approximately 400 kPa). At the start of the sorption measurement, volumes V 1 and V 2 were connected, and the pressure inside the apparatus was monitored by a computer. The pressure initially dropped immediately to a certain value because of the interconnection of the chambers. Subsequently, the pressure decreased further, albeit more slowly, owing to the adsorption of gas in the sample material. The time required to attain equilibrium was approximately 3 h.
The amount of gas adsorbed in the material was calculated from the balance as the difference between the initial and equilibrium amounts of gas in the system: The ideal gas state Equation (3) is used for the moles in the gas-phase calculation (N gas mole number): The adsorbed amount is related to the material weight m to determine the adsorbed phase concentration q:

Properties of HCLPPs
The textural properties of these materials are listed in Table 1. The results revealed that the pristine hyper-crosslinked porous polymer had the highest Brunauer-Emmett-Teller surface area (S BET = 757 m 2 g −1 ) and its structure was partially microporous and mesoporous. a S BET the specific surface area calculated by the BET method from the N 2 adsorption isotherm at T = 77 K; b S meso is the specific surface area of mesopores (t-plot method using the standard isotherm according to Harkins and Jura [39]); c V tot the specific total volume of pores determined from the N 2 adsorption isotherm at p/p 0 = 0.99; d V micro the specific volume of micropores (t-plot method determined by the t-plot method using the standard isotherm according to Harkins and Jura).
The reaction of the prepared porous material with polyamine species (amines and amine groups containing dendrimers) decreased the specific surface area and total volume of pores, while in the case of modification by dendrimers, the decrease was smaller than that for amines. The reduction in specific surface area due to amine functionalization was also reported by Lee et al. [32], see Table 6.
Similarly, the total and micropore volumes decreased significantly after amine modification. However, the modification by dendrimers reduced the total and micropore volumes only slightly, which indicated difficulties in filling the micropores with voluminous dendrimer molecules affecting only the region of mesopores and not the micropores.
To investigate the morphology and elemental composition of the prepared HCLPP, a SEM equipped with an EDX spectroscopy detector was used. From the photographs in Figure 3, the morphologies of all prepared polymers were similar, and the resulting materials had different particle sizes of irregular shape with no observable changes after functionalization by agents containing amino groups.
The reaction of the prepared porous material with polyamine species (amines and amine groups containing dendrimers) decreased the specific surface area and total volume of pores, while in the case of modification by dendrimers, the decrease was smaller than that for amines. The reduction in specific surface area due to amine functionalization was also reported by Lee et al. [32], see Table 6.
Similarly, the total and micropore volumes decreased significantly after amine modification. However, the modification by dendrimers reduced the total and micropore volumes only slightly, which indicated difficulties in filling the micropores with voluminous dendrimer molecules affecting only the region of mesopores and not the micropores.
To investigate the morphology and elemental composition of the prepared HCLPP, a SEM equipped with an EDX spectroscopy detector was used. From the photographs in Figure 3, the morphologies of all prepared polymers were similar, and the resulting materials had different particle sizes of irregular shape with no observable changes after functionalization by agents containing amino groups. Elemental analyses of the prepared polymers, performed using EDX, are given in Table 2. After functionalization, the nitrogen content in the materials considerably increased. The modification by amines increased the nitrogen content to more than 9%, while with the dendrimer, the value of approximately 5% was achieved. Simultaneously, Elemental analyses of the prepared polymers, performed using EDX, are given in Table 2. After functionalization, the nitrogen content in the materials considerably increased. The modification by amines increased the nitrogen content to more than 9%, while with the dendrimer, the value of approximately 5% was achieved. Simultaneously, the modification lowered the chlorine content, indicating a reaction with the chloride units. The results show a more effective reaction and exchange of chloride groups for amine groups and a significantly higher density of amine groups on the surface of the porous material for amine functionalization compared to that of modification with a dendrimer. These findings were in agreement with the morphological data shown in Table 1.
Elemental analysis of the dendrimer-modified sample also revealed the presence of sulfur and silicon in the modifying agent. A small amount of silicon was also detected in other porous polymers, which was probably caused by contamination during the grinding of the porous samples in the ceramic bowl before the analyses. Thermal stability of polymers was evaluated by TGA in the temperature range of 25 to 1000 • C under argon atmosphere with heating rate of 4 • C min −1 . The TGA results are shown in Figure 4. The profiles of all the samples were very similar. A weight loss of approximately 6% at a lower temperature up to 250 • C was observed, probably caused by the evaporation of water, residual solvent, and trapped gases in the porous structure of the material. Significant weight loss followed as the temperature increased to 445 • C, corresponding to the decomposition of polystyrene-based polymers [28,40].
dendrimer. These findings were in agreement with the morphological data shown in Table 1. Elemental analysis of the dendrimer-modified sample also revealed the presence of sulfur and silicon in the modifying agent. A small amount of silicon was also detected in other porous polymers, which was probably caused by contamination during the grinding of the porous samples in the ceramic bowl before the analyses.
Thermal stability of polymers was evaluated by TGA in the temperature range of 25 to 1000 °C under argon atmosphere with heating rate of 4 °C min −1 . The TGA results are shown in Figure 4. The profiles of all the samples were very similar. A weight loss of approximately 6% at a lower temperature up to 250 °C was observed, probably caused by the evaporation of water, residual solvent, and trapped gases in the porous structure of the material. Significant weight loss followed as the temperature increased to 445 °C, corresponding to the decomposition of polystyrene-based polymers [28,40]. Compared to waste expanded PS undergoing complete weight loss in the range of 400-440 °C [27,40], the prepared porous polymers were more thermally stable, having a total weight loss at 1000 °C of only approximately 58% to 65%. This high thermal stability Compared to waste expanded PS undergoing complete weight loss in the range of 400-440 • C [27,40], the prepared porous polymers were more thermally stable, having a total weight loss at 1000 • C of only approximately 58% to 65%. This high thermal stability is because of the rigid cross-linked porous structure formed during the Friedel-Crafts reaction [27,41].

Gas Adsorption and Selectivity
The adsorption isotherms of gases (CO 2 , CH 4 , N 2 , H 2 , and O 2 ) were measured using a simple volumetric sorption apparatus designed in our laboratory (the scheme is given in Figure 2). All sorption measurements were performed at a constant, actively controlled temperature of 25 • C. Table 3 presents a comparison of the adsorption capacities of the unmodified and modified polymers for pure gases, determined from the measured isotherms at a pressure of 100 kPa. The results demonstrated that polymer modification by amines led to a decrease in the amount of adsorbed gas, with the exception of oxygen. This could be attributed to the reduced surface area of the amine-functionalized forms of the polymer. However, the decrease in the adsorption capacity was not as significant as the decrease in the specific surface area or pore volume of the modified materials. Lee et al. [32] reported similar trends in CO 2 and N 2 capacities at 25 • C for the porous polymer network synthesized by solvothermal polymerization of DVB and VBC, see Table 6.
In contrast, in the case of the material modified with a dendrimer containing amino groups, an increase in the adsorption performance of CO 2 and oxygen was observed, despite the fact that, similar to other modified forms of the polymer, the specific surface area also decreased, but to a lesser extent.
However, if the adsorption capacities of the materials were normalized to the specific surface area, the efficiency of CO 2 capture was significantly improved, thus confirming the positive effect of functionalization (Table 4). While the gas-uptake ability of the porous material decreased after functionalization, the selectivity increased in all cases. In particular, the CO 2 /N 2 and O 2 /N 2 separation efficiencies of the HCLPPs after amination were improved. Table 5 shows the selectivity values for different gas pairs obtained from the adsorption isotherms of pure gases at 100 kPa. To verify the precision of the measurements using the simple apparatus constructed in our laboratory, we also performed control measurements of CO 2 adsorption using the automated commercial apparatus ASAP 2050 (Micromeritics, Norcross, GA, USA). As shown in Figure 5, the agreement of the results was excellent; therefore, the apparatus provided the correct results.
To verify the precision of the measurements using the simple apparatus constructed in our laboratory, we also performed control measurements of CO2 adsorption using the automated commercial apparatus ASAP 2050 (Micromeritics, Norcross, GA, USA). As shown in Figure 5, the agreement of the results was excellent; therefore, the apparatus provided the correct results. The obtained CO2 adsorption data of the prepared materials were compared with other hyper-crosslinked porous polystyrene-based polymers studied for CO2 capture application reported in literature (Table 6), for examples HCLPs prepared via the Friedel-Craft reaction between waste expanded polystyrene and 1,2-dichloroethane as both the solvent and the crosslinking agent [27], similar polymers produced using four cross-linkers in order to enhance the porosity of the polymeric adsorbent [28], or amine-functionalized porous polymer network prepared by conventional radical polymerization. In all cases, the material prepared in our study showed a comparable or slightly lower CO2 adsorption capacity, but in most cases the higher selectivity. The obtained CO 2 adsorption data of the prepared materials were compared with other hyper-crosslinked porous polystyrene-based polymers studied for CO 2 capture application reported in literature (Table 6), for examples HCLPs prepared via the Friedel-Craft reaction between waste expanded polystyrene and 1,2-dichloroethane as both the solvent and the crosslinking agent [27], similar polymers produced using four cross-linkers in order to enhance the porosity of the polymeric adsorbent [28], or amine-functionalized porous polymer network prepared by conventional radical polymerization. In all cases, the material prepared in our study showed a comparable or slightly lower CO 2 adsorption capacity, but in most cases the higher selectivity.

Conclusions
In summary, we presented a method of preparation of a new porous polymeric adsorbent by the suspension polymerization of divinylbenzene and 4-vinylbenzyl chloride, followed by modification with amines and branched dendrimers containing amino groups in order to enhance its CO 2 affinity. Synthesized hyper-crosslinked porous material showed a high apparent surface area and selective CO 2 adsorption over CH 4 , N 2 and H 2 , which significantly increased after amines functionalization. Therefore, produced porous polymers represent promising organic porous materials for CO 2 uptake from gas mixtures. The results indicated that the reaction of the prepared polymer with the dendrimer was not as effective as its reaction with amines due to the partially microporous structure being too tight for the possible interaction with the branched dendrimer molecules. Based on this finding, we will focus in the future on the development of polymers with mesoporous structure that may be more suitable for dendrimer loading. The prepared porous polymer structure and its amine-functionalized forms are considered for testing as fillers into mixed-matrix membranes with improved CO 2 /CH 4 and CO 2 /N 2 separation performance as well.