Use of Periodic Mesoporous Organosilica–Benzene Adsorbent for CO2 Capture to Reduce the Greenhouse Effect

The CO2 adsorption of a phenylene-bridged ordered mesoporous organosilica (PMO–benzene) was analyzed. The maximum capture capacity was 638.2 mg·g−1 (0 °C and 34 atm). Approximately 0.43 g would be enough to reduce the amount of atmospheric CO2 in 1 m3 to pre-industrial levels. The CO2 adsorption data were analyzed using several isotherm models, including Langmuir, Freundlich, Sips, Toth, Dubinin–Radushkevich, and Temkin models. This study confirmed the capability of this material for use in reversible CO2 capture with a minimal loss of capacity (around 1%) after 10 capture cycles. Various techniques were employed to characterize this material. The findings from this study can help mitigate the greenhouse effect caused by CO2.


Introduction
Excessive increases in CO 2 concentration over a short period of time lead to an amplification of the greenhouse effect and a sharp rise in the average temperature of the planet.Before the industrial revolution, CO 2 levels were 280 ppm [1].As a result of human activities, a CO 2 level of 421 ppm was reached in 2022 [2][3][4].To address this situation, many international organizations and governments have worked to reach common agreements.One of the most important agreements was the Paris Agreement (2016) [5].The capture of CO 2 [6,7] could allow its use as a revalorized by-product, such as fuels, chemicals, and building materials, adding an economic incentive for CO 2 capture and the green economy, while also reducing the environmental footprint [8,9].Different adsorbents such as the diaminebased hybrid-slurry system [10], porous carbons [11,12], silica [13][14][15], MOFs [16][17][18], metal oxides [19][20][21][22], hydrotalcites [23,24], and periodic mesoporous organosilicas (PMOs) have been studied recently for CO 2 capture.
Selecting specific organosilica types and synthesis conditions enables precise control over the size, shape, uniformity, and periodicity of the pore structures [49].However, the shape of the adsorbent particle determines the specific surface area.
In this study, a PMO with phenylene bridges was chosen as the CO 2 adsorbent material (Figure 1).Various techniques were utilized to characterize this material, and its CO 2 capture capacity at low temperatures (0 • C, 10 • C, 20 • C, and 35 • C) and high pressure (35 atm) was evaluated.The CO 2 adsorption data were fitted by the Langmuir, Freundlich, Sips, Toth, Dubinin-Radushkevich, and Temkin models.
Selecting specific organosilica types and synthesis conditions enables precise control over the size, shape, uniformity, and periodicity of the pore structures [49].However, the shape of the adsorbent particle determines the specific surface area.
In this study, a PMO with phenylene bridges was chosen as the CO2 adsorbent material (Figure 1).Various techniques were utilized to characterize this material, and its CO2 capture capacity at low temperatures (0 °C, 10 °C, 20 °C, and 35 °C) and high pressure (35 atm) was evaluated.The CO2 adsorption data were fitted by the Langmuir, Freundlich, Sips, Toth, Dubinin-Radushkevich, and Temkin models.
The capability of this material for use in reversible CO2 capture with a minimal loss of capacity after 10 capture cycles was addressed.
The obtained results can aid in the reduction in the CO2 greenhouse effect.

PMO-Benzene
Periodic mesoporous organosilica (PMO-benzene) was synthesized via acidcatalyzed hydrolysis and the condensation of bis(triethoxysilyl)benzene following the method outlined by Burleigh [58,59].In this process, 6 g of the surfactant BrijS10-Brij76 (Polyoxyethylene (10) stearyl ether) was dissolved in a solution of 19.6 mL of HCl and 279 mL of H2O while being stirred at 50 °C for 24 h.Subsequently, 17.2 mL of 1,4bis(triethoxysilyl)benzene was added, and the mixture was continuously stirred at 50 °C for another 24 h.The solution was then kept at 90 °C under static conditions for an additional 24 h.The resulting precipitate was collected through vacuum filtration and washed with H2O.To remove the surfactant, the synthesized material was stirred in a HCl solution (1:50 ratio) at 80 °C for 12 h.Finally, the product was filtered, washed with ethanol, and dried under vacuum for 10 h.The molar ratios of Brij76, H2O, HCl, and organosilane used in the reaction were 0.11:222:3.20:0.56.The capability of this material for use in reversible CO 2 capture with a minimal loss of capacity after 10 capture cycles was addressed.

Material Characterisation
The obtained results can aid in the reduction in the CO 2 greenhouse effect.

PMO-Benzene
Periodic mesoporous organosilica (PMO-benzene) was synthesized via acid-catalyzed hydrolysis and the condensation of bis(triethoxysilyl)benzene following the method outlined by Burleigh [58,59].In this process, 6 g of the surfactant BrijS10-Brij76 (Polyoxyethylene (10) stearyl ether) was dissolved in a solution of 19.6 mL of HCl and 279 mL of H 2 O while being stirred at 50 • C for 24 h.Subsequently, 17.2 mL of 1,4-bis(triethoxysilyl)benzene was added, and the mixture was continuously stirred at 50 • C for another 24 h.The solution was then kept at 90 • C under static conditions for an additional 24 h.The resulting precipitate was collected through vacuum filtration and washed with H 2 O.To remove the surfactant, the synthesized material was stirred in a HCl solution (1:50 ratio) at 80 • C for 12 h.Finally, the product was filtered, washed with ethanol, and dried under vacuum for 10 h.The molar ratios of Brij76, H 2 O, HCl, and organosilane used in the reaction were 0.11:222:3.20:0.56.

Material Characterisation
Different techniques were used for the characterization of the PMO-benzene samples.XRD analysis was conducted using a Bruker D8 Discover A25 instrument (Karlsruhe, Germany), utilizing the ICDD 2003 database [60].N 2 adsorption isotherms were used to evaluate S BET and porosity, conducted with the Autosorb iQ2 from Quantachrome Instruments (Boynton Beach, FL, USA).Prior to this, the samples underwent degassing for 24 h at 70 • C. The data were processed with AsiQwin software (version 3.0).The surface area was calculated using the Brunauer-Emmett-Teller (BET) method.Density functional theory (DFT) was used to test the porosity.The pore volumes were calculated on the basis of the single-point method.A Mastersizer S laser diffractometer (Malvern, UK) was used to determine the particle size.Particle aggregates were separated during an ultrasonic bath (10 min).Particle structure was analyzed using a transmission electron microscope (Talos F200i TEM; Thermo Fisher Scientific) (Waltham, MA, USA).A Sievert PCTPro-2000 instrument (Setaram) (Caluire-et-Cuire, Francia) was used to obtain the CO 2 isotherms.The gases CO 2 4.5 (99.995%) and He 5.0 (99.999%) were used.
Langmuir characterizes monolayer adsorption on a uniform surface [71] with identical adsorption sites, indicating that the energy remains constant regardless of the amount adsorbed [72].Since adsorption takes place in a single layer, the surface area can be determined.In 1995, Tóth proposed [66] a correction factor χ L to enhance the accuracy of monolayer adsorption calculations.
Freundlich, on the other hand, describes adsorption on heterogeneous surfaces [73,74], enabling the assessment of the surface's heterogeneity and the adsorption intensity.
The Sips model integrates aspects of both the Langmuir and Freundlich models by depicting Freundlich-type adsorption at low pressures and resembling the Langmuir model at high pressures, thereby unifying the two approaches [75].Analogously to the Langmuir model, a correction factor, χ S , was applied.
Toth's model modifies Langmuir's equation to accurately describe adsorption isotherms that include both monolayer and multilayer formations up to the maximum relative pressure.
The D-R model is employed to describe the CO 2 adsorption based on potential energy during pore filling, which in turn determines adsorption capacity.
The Temkin model explains the heat generated in the pore filling process on the solid surface, establishing a correlation with the quantity of gas adsorbed.
MATLAB software version R2015a was used for calculation and fitting.

Characterization
High-quality PMO-benzene was characterized.As illustrated in Figure 2, the XRD pattern of the sample displays its most prominent peak (100) at d = 53.4Å.Additionally, two smaller and broader peaks were observed at approximately d = 27.3Å (110) and d = 21 Å (200).These peaks confirm that the sample is a mesoporous material with a 2D hexagonal (P6mm) structure, characteristic of the organosilica PMO-benzene, consistent with the literature [36,76].
Figure 3 presents the particle size distribution, revealing a bimodal pattern.The primary peak is narrow, ranging from 2.5 to 35 µm and centered at 13.2 µm, suggesting uniformity in particle size within the sample.A secondary peak appears between 0.3 and 2.5 µm, with a center at 1.45 µm, indicating the presence of a smaller particle group.
Figure 4 displays the N 2 adsorption-desorption isotherm, which is Type IV according to the IUPAC.The first portion of the isotherm corresponds to monolayer-multilayer adsorption, with a well-defined adsorption limit at pressures approaching p•p 0 −1 = 1 [77,78].The hysteresis in the multilayer region is indicative of capillary condensation within the mesopores.This hysteresis is classified as type H1, which is associated with a pore system that features a relatively regular arrangement and a narrow distribution of uniform mesopores [78,79], consistent with the XRD-derived crystal structure.Most pores were small mesopores, ranging from 2.4 to 4.1 nm, with a predominant pore diameter of 3.47 nm, and the total pore volume was 0.68 cm3 •g −1 (Table 1).The BET method revealed a high surface area of 928 m2 •g −1 , which explains the material's significant adsorption capacity.Additionally, several micropores were identified, contributing 139 m 2 •g −1 to the total surface area, representing 15% of the total.Figure 3 presents the particle size distribution, revealing a bimodal pattern.The primary peak is narrow, ranging from 2.5 to 35 µm and centered at 13.2 µm, suggesting uniformity in particle size within the sample.A secondary peak appears between 0.3 and 2.5 µm, with a center at 1.45 µm, indicating the presence of a smaller particle group.Figure 4 displays the N2 adsorption-desorption isotherm, which is Type IV according to the IUPAC.The first portion of the isotherm corresponds to monolayer-multilayer adsorption, with a well-defined adsorption limit at pressures approaching p•p0 −1 = 1 [77,78].The hysteresis in the multilayer region is indicative of capillary condensation within the mesopores.This hysteresis is classified as type H1, which is associated with a pore system that features a relatively regular arrangement and a narrow distribution of uniform mesopores [78,79], consistent with the XRD-derived crystal structure.Most pores were small Figure 3 presents the particle size distribution, revealing a bimodal pattern.The primary peak is narrow, ranging from 2.5 to 35 µm and centered at 13.2 µm, suggesting uniformity in particle size within the sample.A secondary peak appears between 0.3 and 2.5 µm, with a center at 1.45 µm, indicating the presence of a smaller particle group.Figure 4 displays the N2 adsorption-desorption isotherm, which is Type IV according to the IUPAC.The first portion of the isotherm corresponds to monolayer-multilayer adsorption, with a well-defined adsorption limit at pressures approaching p•p0 −1 = 1 [77,78].The hysteresis in the multilayer region is indicative of capillary condensation within the mesopores.This hysteresis is classified as type H1, which is associated with a pore system that features a relatively regular arrangement and a narrow distribution of uniform mesopores [78,79], consistent with the XRD-derived crystal structure.Most pores were small mesopores, ranging from 2.4 to 4.1 nm, with a predominant pore diameter of 3.47 nm, and the total pore volume was 0.68 cm 3 •g⁻ 1 (Table 1).The BET method revealed a high surface area of 928 m 2 •g⁻ 1 , which explains the material's significant adsorption capacity.Additionally, several micropores were identified, contributing 139 m 2 •g⁻ 1 to the total surface area, representing 15% of the total.
A TEM was used on the PMO-benzene samples.Figure 5A shows the ordered pore structure of the material from two perspectives: longitudinal and transverse to the pore network.Figure 5B shows channels with a porous diameter around 2.867 nm, i.e., a pore size which agrees with the pore-size distribution (2.4-4.1 nm).The value of 1.01 nm corresponds to the thickness of the channel wall.A TEM was used on the PMO-benzene samples.Figure 5A shows the ordered pore structure of the material from two perspectives: longitudinal and transverse to the pore network.Figure 5B shows channels with a porous diameter around 2.867 nm, i.e., a pore size which agrees with the pore-size distribution (2.4-4.1 nm).The value of 1.01 nm corresponds to the thickness of the channel wall.The CO2 adsorption capacity decreased as the temperature increased.The highest adsorption capacity was observed at 0 °C (638.2 mg•g⁻ 1 ), consistent with previous characterization results.In order to analyze the decrease in adsorption and to explore the potential for reusing the material as a CO2 adsorbent, a study was conducted involving 10 consecutive adsorption-desorption cycles at 0 °C, 10 °C, and 35 °C (Figure 7).The linear trend and maximum adsorption for each cycle are shown in Figure 8.After 10 cycles, there was a small loss in adsorption in all cases, with loss amounts between 1% and 1.5%.The working temperature did not seem to affect the adsorption capacity loss.The results indicate the material's potential for reuse without significant loss of performance.The CO 2 adsorption capacity decreased as the temperature increased.The highest adsorption capacity was observed at 0 • C (638.2 mg•g −1 ), consistent with previous characterization results.

CO2 Adsorption
In order to analyze the decrease in adsorption and to explore the potential for reusing the material as a CO 2 adsorbent, a study was conducted involving 10 consecutive adsorptiondesorption cycles at 0 • C, 10 • C, and 35 • C (Figure 7).The linear trend and maximum adsorption for each cycle are shown in Figure 8.After 10 cycles, there was a small loss in adsorption in all cases, with loss amounts between 1% and 1.5%.The working temperature did not seem to affect the adsorption capacity loss.The results indicate the material's potential for reuse without significant loss of performance.
Various models were used to fit the adsorption data.
All the isotherms fit the Freundlich model well (Figure 9).For the Langmuir model, only the isotherms at 0 and 10 • C obtained an R 2 value below 0.99 due to multilayer formation and significant adsorption in the final stage (Table 2).The samples fit the Freundlich model slightly better than the Langmuir model.The Freundlich model provided a better fit, indicating surface heterogeneity (n), with results suggesting a somewhat heterogeneous surface.

Langmuir
Freundlich PMO-benzene 0   Various models were used to fit the adsorption data.Various models were used to fit the adsorption data.only the isotherms at 0 and 10 °C obtained an R 2 value below 0.99 due to multilayer formation and significant adsorption in the final stage (Table 2).The samples fit the Freundlich model slightly better than the Langmuir model.The Freundlich model provided a better fit, indicating surface heterogeneity (n), with results suggesting a somewhat heterogeneous surface.[80] classification, the isotherms exhibited characteristics between L-type, indicative of strong intermolecular attractions and a gradual reduction in available adsorption sites, and C-type, where the number of adsorption sites remains constant.L-type curves typically align with Langmuir adsorption behavior, while C-type curves suggest linear adsorption.This shape is associated with weak intermolecular forces and a gradual occupation of adsorption sites.The monolayer filling values (qm) were generally overestimated, but using the Tóth correction yielded accurate monolayer filling predictions (qmc).This adjustment enabled the calculation of specific surface area using the Langmuir model.At 0 °C, where the adsorption curve reached a p•p0 −1 of 1, the specific surface area (SL) was 904.1 m 2 •g⁻ 1 , closely matching the BET method's value of 928 m 2 •g⁻ 1 (Table 1).The Freundlich intensity factor (nf), which measures the interaction strength between the adsorbent and adsorbate at low pressures, was above 1 but close to it, indicating favorable, albeit weak, interactions [66].This result agrees with Giles et al.'s [80] interpretation.The Langmuir equilibrium constant (KL) and Freundlich nf values (Table 2) increased with temperature, except at 35 °C, consistent with findings from other CO2 capture materials in the literature [23,24,61].The Freundlich adsorption capacity per unit According to Giles et al.'s [80] classification, the isotherms exhibited characteristics between L-type, indicative of strong intermolecular attractions and a gradual reduction in available adsorption sites, and C-type, where the number of adsorption sites remains constant.L-type curves typically align with Langmuir adsorption behavior, while C-type curves suggest linear adsorption.This shape is associated with weak intermolecular forces and a gradual occupation of adsorption sites.The monolayer filling values (qm) were generally overestimated, but using the Tóth correction yielded accurate monolayer filling predictions (qmc).This adjustment enabled the calculation of specific surface area using the Langmuir model.At 0 • C, where the adsorption curve reached a p•p 0 −1 of 1, the specific surface area (SL) was 904.1 m 2 •g −1 , closely matching the BET method's value of 928 m 2 •g −1 (Table 1).The Freundlich intensity factor (nf), which measures the interaction strength between the adsorbent and adsorbate at low pressures, was above 1 but close to it, indicating favorable, albeit weak, interactions [66].This result agrees with Giles et al.'s [80] interpretation.The Langmuir equilibrium constant (K L ) and Freundlich nf values (Table 2) increased with temperature, except at 35 • C, consistent with findings from other CO 2 capture materials in the literature [23,24,61].The Freundlich adsorption capacity per unit concentration (Kf) increased as the temperature decreased, directly correlating with higher CO 2 adsorption capacity.
These models effectively showed the adsorption trends (qmc and Kf), which rose as the temperature decreased.The Freundlich model provided a more accurate prediction of the maximum adsorption capacity (Figure 9) compared to the Langmuir model, which only accounts for monolayer adsorption.
The Sips and Toth models (three-parameter) improved the fit over the Langmuir and Freundlich models for samples between 0 and 20 • C, reflecting in higher R 2 values (Table 3).These models used relative pressures (p•p 0 −1 ) of 0.38, 0.46, and 0.54 as starting points for multilayer formation for samples at 0 • C, 10 • C, and 20 • C, respectively, obtained from the adsorption curve's derivative.For the 35 • C sample, where no multilayer formation occurred, the maximum p•p 0 −1 value was used for fitting (Figure 9), though the parameter values are not listed in Table 3.The temperature-dependent trends for qs (Sips) and qT (Toth) differed from qmc (Langmuir) due to the models' varying interpretations of the adsorption curve.The Langmuir model treated the curve as a monolayer over the entire pressure range, whereas the Sips and Toth models incorporated the pre-parameter for multilayer initiation, which varied with temperature.The Sips and Toth models predicted the CO 2 adsorption capacity equally well, as indicated by the R 2 values.The increasing qs and qt values with temperature are related to the increase in the relative pressure (p•p 0 −1 ) at which the multilayer is formed.This could be related to the increase in curvature at low p•p 0 −1 values and is in accordance with the nf value of the Freundlich model (the higher the initial curvature, the higher the nf).The heterogeneity factors from the Sips (n S ) and Toth (n T ) models agreed with the Freundlich model's heterogeneity factor (n).

Sips
Toth PMO-benzene 0 This research emphasizes the importance of using three-parameter models alongside classical two-parameter models to achieve better isotherm fitting, particularly when multilayer formation occurs (Figure 9).
For the D-R model, all isotherms had R 2 values greater than 0.9.These adsorption capacities (q D ) were slightly higher than those determined by the Langmuir model (qmc).The adsorption constant (β) was used to determine the free energy (E), which was smaller than 8 kJ•mol −1 for all samples (Table 4), suggesting the predominantly physical nature of adsorption.A slight increase in E was observed with increasing temperature.A relationship between the D-R adsorption (E) and Temkin maximum binding energy (KTk) was observed, with both parameters increasing upon increasing temperature.The same trend was observed for the Temkin constant (bTk).These findings align with the expectation of low adsorption energy [81].The physical nature of adsorption and the weak intermolecular attraction between adsorbent and adsorbate, as indicated by the fits, support the reversibility of CO 2 capture.This is further evidenced by the minimal loss of adsorption after 10 cycles (Figures 7 and 8).These properties suggest that the material is suitable for reversible multicycle CO 2 capture by changing pressure conditions.
Table 5 provides a summary of significant studies on CO 2 capture using mesoporous materials.Most of these studies focus on single temperature and pressure conditions and do not include adsorption-desorption cycling analyses.This research, however, examines the adsorption performance of PMO-benzene across various operating conditions.
The findings of this study indicate that CO 2 adsorption at 34 atm pressure and temperatures ranging from 0 • C to 35 Liu et al. [21] and Li et al. [85].The best result was 391.6 mg•g −1 (25 • C and 10 atm).In the current study, better results were obtained when working between temperatures of 0 • C and 10 • C and at a pressure of 34 atm.
In addition, the results from this study have been evaluated against various MOFtype materials as detailed in  [92] (1493 mg•g −1 ), and Furukawa et al. [93] (2400 mg•g −1 )-reported higher CO 2 capture capacities at elevated pressures and 25 • C than the highest value from our study (827.8mg•g −1 at 0 • C and 34 atm).However, these studies did not investigate adsorption-desorption cycles extensively (with Furukawa [93] conducting only a two-cycle study), which could impact the material's structure and reduce its adsorption capacity.In contrast, our study involved multiple adsorption-desorption cycles, revealing a minor capacity loss of 1% to 1.5% after 10 cycles, underscoring its potential industrial application for purifying CO 2 -rich environments.
Lastly, it is noteworthy to quantify the amount of adsorbent required to reduce current atmospheric CO 2 levels (421 ppm (mL•m −3 ) [4]) to pre-industrial levels (280 ppm (mL•m −3 ) [1]).Given CO 2 's density at 0 • C and 1 atm (1.976 mg•cm −3 ), the required amount would be 276.85mg•m −3 .Therefore, with a maximum capture capacity of 638.2 mg•g −1 at 0 • C, 0.43 g of PMO-benzene would suffice to lower the CO 2 concentration in 1 m 3 of air to pre-industrial levels.To apply this to a large volume, such as Wembley Stadium (1,139,100 m 3 ), approximately 489.8 kg of PMO-benzene would be needed to achieve this reduction.

Conclusions
In this study, the adsorbent PMO-benzene was tested at high CO 2 gas pressures (up to 35 atm) and low temperatures (0 • C, 10 • C, 20 • C, and 35 • C).

1.
The pore size was between 2.4 and 4.1 nm, while the total pore volume was 0.68 cm 3 •g −1 and S BET was 928 m 2 •g −1 .

2.
All isotherms fitted the Freundlich model well.nf > 1 was very close to 1, indicating favorable and weak interactions.

3.
The Sips and Toth models improved the results obtained by other equations at temperatures between 0 • C and 20 • C, where multilayer formation occurred.4.
The D-R and Temkin models showed a physical nature of adsorption (E < 8 kJ•mol −1 ). 5.
These results highlight that 0.43 g of PMO-benzene would be enough to reduce the CO 2 level in 1 m 3 of air to pre-industrial levels; 489.8 kg of PMO-benzene would be required to reduce the CO 2 concentration of the volume of Wembley soccer stadium to pre-industrial levels.7.
The maximum loss of adsorption capacity was 1.45% for the sample at 0 • C, after 10 adsorption-desorption cycles.Consequently, this material could be used in capture processes using changes in the pressure conditions.8.
PMO-benzene could contribute to the development of CO 2 capture and use (CCU) technology.

Figure 4 .
Figure 4. Nitrogen adsorption-desorption isotherms and pore size distribution of PMO-benzene.In the inserted figure: (Blue line) Pore volume, (Red line) Cumulative pore volume.

Figure 4 .
Figure 4. Nitrogen adsorption-desorption isotherms and pore size distribution of PMO-benzene.In the inserted figure: (Blue line) Pore volume, (Red line) Cumulative pore volume.

Table 2 .
Langmuir and Freundlich fitting results.

Table 2 .
Langmuir and Freundlich fitting results.

Table 3 .
Sips and Toth model fitting results.

Table 5 .
CO 2 capture for different materials.
• C surpasses previously reported values for PMOs in the literature.Table 5 compares these results with those for other mesoporous materials, such as those studied by Chowdhury et al. [20], Bhagiyalakshmi et al. [82], Wang et al. [83], Niu et al. [84],