Synthesis, Characterization and Application of Amine-Functionalized Hierarchically Micro-Mesoporous Silicon Composites for CO2 Capture in Flue Gas

An efficient CO2 adsorbent with a hierarchically micro-mesoporous structure and a large number of amine groups was fabricated by a two-step synthesis technique. Its structural properties, surface groups, thermal stability and CO2 adsorption performance were fully investigated. The analysis results show that the prepared CO2 adsorbent has a specific hierarchically micro-mesoporous structure and highly uniformly dispersed amine groups that are favorable for the adsorption of CO2. At the same time, the CO2 adsorption capacity of the prepared adsorbent can reach a maximum of 3.32 mmol-CO2/g-adsorbent in the actual flue gas temperature range of 303–343 K. In addition, the kinetic analysis results indicate that both the adsorption process and the desorption process have rapid adsorption/desorption rates. Finally, the fitting of the CO2 adsorption/desorption experimental data by Avrami’s fractional kinetic model shows that the CO2 adsorption rate is mainly controlled by the intra-particle diffusion rate, and the temperature has little effect on the adsorption rate.


Introduction
For climate change, the ultimate global goal is to avoid dangerous disturbances to the climate system. To achieve this goal, UNFCCC member states have shown strong support for measurable guardrail targets, such as "Avoid 1.5 • C" or "Avoid 450 ppm CO 2 ". However, the CO 2 concentration in the atmosphere has increased sharply from 340 ppm in 1980 to 408 ppm in 2019 [1], and if existing human activities do not sufficiently and quickly change, the global temperature increase will exceed 1.5 • C. Therefore, a variety of methods should be adopted to reduce CO 2 levels, such as planting more trees, replacing fossil energy with renewable energy, improving the efficiency of coal-fired plants, and using CO 2 capture and storage (CCS) to reduce CO 2 content [2,3].
CCS technology is an attractive method because it helps reduce the release of CO 2 into the environment and allows the continued use of coal to meet the world's energy needs [2,4]. According to different coal-fired plant configurations, three main approaches can be applied to reduce CO 2 emissions from the flue gas of these plants, including precombustion capture, oxyfuel combustion, and post-combustion capture [5]. Among these CCS technologies, post-combustion capture technology (which uses wet/dry adsorbents to absorb CO 2 from the flue gas that is produced by the combustion of fossil fuels) is the most popular industrial method [6]. Post-combustion capture technologies generally include absorption [7], porous materials adsorption [8], membrane separation [9], cryogenics and hydration etc. [5]. At present, some experiments and computational studies have been conducted on the above-mentioned technologies [10][11][12].

Sample Preparation 2.2.1. Synthesis of Support Material
The preparation of the hierarchically micro-mesoporous silicon material was as follows: First, 4.54 g of TPAOH was dissolved in 16.87 g of deionized water at room temperature and stirred for 30 min, and then 0.17 g of aluminum isopropoxide and 0.1 g of NaOH were added. After stirring for 30 min, the mixture was heated to 303 K, and 8.5 g of TEOS was dropped into the mixture. After 16 h of vigorous stirring, the final solid-liquid mixture was transferred to a Teflon kettle and subjected to hydrothermal treatment at 423 K for 24 h. The microcrystalline emulsion was then cooled to room temperature and the final resultant was labeled as microcrystalline emulsion A. Second, 4.0 g of P123 (5800) was dissolved in a mixture of 144 g of H 2 O and 7.9 g of hydrochloric acid solution (35 wt.%) at 313 K under stirring. After the P123 was completely dissolved, 4.0 g of n-Butanol was added. After stirring continuously for 1 h, the above microcrystalline emulsion A was added and sonicated for 30 min. After vigorously stirring at 313 K for 24 h the resultant was transferred to a Teflon kettle and subjected to hydrothermal treatment at 373 K for 24 h. The solid-liquid mixture was filtered to obtain a solid product and the solid product was dried at 373 K for 10 h. After that, the obtained solid sample was calcined at 823 K for 6 h under a temperature programmed system (5 K/min) to remove the template. The resulting material is herein denoted as micro-mesoporous silicon (MMS).

Preparation of Amine-Functionalized Hierarchically Micro-Mesoporous Adsorbent
The amine-loaded hierarchically micro-mesoporous adsorbent was produced by a wet impregnation method. The ethylenediamine (EDA); diethylenetriamine (DETA); tetraethylenepentamine (TEPA); pentaethylenehexamine (PEHA); and polyetherimide (PEI) were used as amine-modified materials. In a typical experiment, 1 g of amine-modified material was dissolved in 50 mL of ethanol at room temperature, and after stirring for 30 min, 1 g of MMS was added, and then the mixture was stirred for 2 h. After that, the mixture was evaporated at 353 K and then dried in air at 373 K for 1 h. Finally, a white composite was obtained. The obtained composites are denoted as AMMS-E, AMMS-D, AMMS-T, AMMS-P and AMMS-PEI, respectively. Here, the AMMS represents amine-functionalized micro-mesoporous silicon, and the E, D, T, P, and PEI stand for EDA, DETA, TEPA, PEHA, and PEI, respectively.

Characterization
The surface crystallinity of the AMMS was analyzed by the X-ray diffraction (XRD) measurement using a Rigaku powder diffractometer (D8 ADVANCE, Bruker, Bremen, Germany) with Cu Kα radiation (λ = 0.15406 nm). The tube voltage was 45 kV and the current was 40 mA. The XRD diffraction patterns were taken in the 2θ range of 0.5-10 • and 5-60 • at a scan speed of 2 • /min. The surface morphology was determined with a scanning electron microscope (S4800, Hitachi, Tokyo, Japan).
The surface area analyzer (Autosorb iQ, The Quantachrome Instruments U.S., Boynton Beach, FL, USA) was used for the nitrogen adsorption/desorption test. Before the measurement, the sample was degassed at 573 K under nitrogen flow for 3 h. The surface area of the powder was obtained using the Brunauer-Emmett-Teller (BET) method. The pore size distributions were calculated by the Barrett-Joyner-Halenda (BJH) equation. The total pore volume was determined from the amount of adsorbed N 2 at P/P 0 = 0.99.
A TG (TG 209F3, Netzsch, Selb, Germany) instrument was used to analyze the thermal stability of AMMS, and the thermogravimetric analysis test was performed at a temperature of 303 K to 973 K with a heating rate of 10 K/min under a dynamic N 2 atmosphere.
The surface chemical groups were analyzed by Fourier Transform Infrared (FTIR, Nicolet IS50, Thermo Fisher Scientific, Waltham, MA, USA) spectroscopies.

Adsorption/Desorption of CO 2
A TGA device (TG209F3, Netzsch) was used for the CO 2 adsorption/desorption test. A total of 10 mg of AMMS samples were loaded into a 0.05 cm 3 platinum sample pan. After stabilization, each sample was heated to 423 K in a nitrogen stream (120 cm 3 min −1 ), then cooled to the adsorption temperature, and a CO 2 (99.99%)/N 2 (99.99%) mixture (10 Vol.% CO 2 ) was introduced until the mass of the sample no longer increased (about 50 min). The adsorption capacity was calculated based on the AMMS sample mass increase using Equation (1).
where q a is the adsorption capacity of CO 2 , mmol g −1 ; M 0 is the mass of pure adsorbent, and g; M 1 is the mass of the sample after adsorbing CO 2 g. Once the adsorption process reached equilibrium, the temperature programmed desorption (TPD) test was performed on the same experimental system. The adsorbent was heated to the required desorption temperature (383 K) in a nitrogen stream (120 cm 3 min −1 ). The total amount of CO 2 was calculated based on the loss of sample mass using Equation (1).

Adsorption/Desorption Kinetic Models
Avrami's fractional kinetic model (Equation (2)) is used to study the kinetics of CO 2 adsorption on AMMS materials. Usually, Equation (2) can be transformed into Equation (3) in the form of desorption component y. Thus, Equation (3) can be employed to further understand the desorption kinetics of the CO 2 desorption process [27].
where q e and q t (mmol·g −1 ) denote the equilibrium capacity and the adsorption capacity at any time t (s), respectively, k a is the kinetic constant of the Avrami model, the n represents as the Avrami exponent, which is often in fractional form, and reflects possible mechanism changes in the adsorption process [28].

Results and Discussions
3.1. Characterization of the AMMS 3.1.1. XRD Analysis Figure 1 provides the small-angle XRD pattern of the MMS and AMMS powder. The spectrum for the MMS and AMMS powder reveals a strong peak at about 2θ = 0.55 • that corresponds to the (100) crystal face, indicating that the composites materials MMS and AMMS may have a p6 mm hexagonal-symmetry mesoporous structure that is similar to SBA-15 [29]. However, we cannot find the (110) and (200) peaks that are normally observed in SBA-15 from the spectra of the MMS and AMMS powders, which may indicate that the mesoporous system of MMS and AMMS has become disordered [30]. This disorder may be attributed to the formation of microporous structures that are similar to ZSM-5 materials in MMS and AMMS materials. where e q and t q (mmol·g −1 ) denote the equilibrium capacity and the adsorption capacity at any time t (s), respectively, a k is the kinetic constant of the Avrami model, the n represents as the Avrami exponent, which is often in fractional form, and reflects possible mechanism changes in the adsorption process [28].

Characterization of the AMMS
3.1.1. XRD Analysis Figure 1 provides the small-angle XRD pattern of the MMS and AMMS powder. The spectrum for the MMS and AMMS powder reveals a strong peak at about 2θ = 0.55° that corresponds to the (100) crystal face, indicating that the composites materials MMS and AMMS may have a p6 mm hexagonal-symmetry mesoporous structure that is similar to SBA-15 [29]. However, we cannot find the (110) and (200) peaks that are normally observed in SBA-15 from the spectra of the MMS and AMMS powders, which may indicate that the mesoporous system of MMS and AMMS has become disordered [30]. This disorder may be attributed to the formation of microporous structures that are similar to ZSM-5 materials in MMS and AMMS materials.  The locations of the characteristic Bragg diffraction peaks of the MMS and AMMS samples are almost the same, but the peak intensity (XRD patterns in wide-angle) of the The locations of the characteristic Bragg diffraction peaks of the MMS and AMMS samples are almost the same, but the peak intensity (XRD patterns in wide-angle) of the AMMS is reduced, indicating that the amine-modified material is loaded into the micropores of MMS. At the same time, the micro-mesoporous structure of MMS is retained.

SEM Analysis
The surface morphology of MMS and AMMS-T is presented in Figure 2, and the SEM images of other AMMS are shown in Figure S1. The SEM micrograph of MMS shows that this solid is formed by the aggregation of small amorphous material units (like the ZSM-5 material [31], and a certain number of pores are formed between the units. In contrast to the surface morphology of the MMS material, the SEM micrograph of AMMS-T shows that the surface of AMMS-T is distributed with highly dispersed and uniform particles. This result indicates that the amine-modified material was successfully impregnated into the pores of the MMS material, and the loading of the amine-modified material resulted in a decrease in the surface area and pore volume of the structure, which was further confirmed in the BET surface area analysis. At the same time, the highly uniformly dispersed amine on the solid surface may be beneficial to the adsorption of CO 2 [32]. The subsequent CO 2 adsorption analysis also proved this possibility. that the surface of AMMS-T is distributed with highly dispersed and uniform particles. This result indicates that the amine-modified material was successfully impregnated into the pores of the MMS material, and the loading of the amine-modified material resulted in a decrease in the surface area and pore volume of the structure, which was further confirmed in the BET surface area analysis. At the same time, the highly uniformly dispersed amine on the solid surface may be beneficial to the adsorption of CO2 [32]. The subsequent CO2 adsorption analysis also proved this possibility.

BET Analysis
Through nitrogen adsorption experiments at 77 K, the structural characteristics of MMS and AMMS, such as surface area, average pore size and pore volume were analyzed. The nitrogen-adsorption isotherms of MMS and AMMS-T are presented in Figure 3, and the nitrogen adsorption isotherms of other AMMS are displayed in Figure S2. All adsorption isotherms represent the typical type IV features of mesoporous silica. The pore size distribution of MMS and AMMS-T are also displayed in Figure 3. The pore size distribution curve of MMS has two obvious peaks at 1.6 nm and 3.5 nm, indicating that its pore structure is mainly composed of micro-pores and meso-pores. However, the pore size distribution of AMMS-T is relatively random and irregular. This phenomenon may be caused by the disordered accumulation of amine-modified materials on the surface of the pores during the impregnation process.

BET Analysis
Through nitrogen adsorption experiments at 77 K, the structural characteristics of MMS and AMMS, such as surface area, average pore size and pore volume were analyzed. The nitrogen-adsorption isotherms of MMS and AMMS-T are presented in Figure 3, and the nitrogen adsorption isotherms of other AMMS are displayed in Figure S2. All adsorption isotherms represent the typical type IV features of mesoporous silica. The pore size distribution of MMS and AMMS-T are also displayed in Figure 3. The pore size distribution curve of MMS has two obvious peaks at 1.6 nm and 3.5 nm, indicating that its pore structure is mainly composed of micro-pores and meso-pores. However, the pore size distribution of AMMS-T is relatively random and irregular. This phenomenon may be caused by the disordered accumulation of amine-modified materials on the surface of the pores during the impregnation process.   Table 2 summarizes the surface area, average pore size and pore volume of all the materials. It was observed that after amine modification, the surface area and pore volume of AMMS were significantly reduced. This can be explained by amine impregnation, where amine material accumulates on the surface of the MMS porous channels, resulting in a decrease in surface area and pore volume. These results are consistent with the XRD measurement results.   Table 2 summarizes the surface area, average pore size and pore volume of all the materials. It was observed that after amine modification, the surface area and pore volume of AMMS were significantly reduced. This can be explained by amine impregnation, where amine material accumulates on the surface of the MMS porous channels, resulting in a decrease in surface area and pore volume. These results are consistent with the XRD measurement results.

TG Analysis
The thermal behavior of AMMS was studied by TG analysis. Figure 4 shows the mass loss curve of AMMS. Two stages of mass loss are observed in the TG curve of AMMS. Due to the removal of adsorbed water and CO 2 , the mass loss of 3-8 wt.% in the first stage is approximately between 303 K and 373 K. We noticed that when the temperature is higher than 373 K, the mass loss curves display a relatively straight line within a certain temperature range, indicating that there is basically no mass loss in the sample within this temperature range. For AMMS-T, the temperature range is 373 K to 423 K, AMMS-P is 373 K to 483 K, and AMMS-PEI is 373 K to 543 K. As the temperature continues to rise, the samples begin to undergo the second stage of the mass loss process. The results suggest that AMMS-T, AMMS-E, AMMS-D, AMMS-P and AMMS-PEI have a thermal stability of 423 K, 483 K, 403 K, 483 K, and 543 K, respectively. Finally, when the sample mass remains constant, the total mass loss of AMMS-T, AMMS-P and AMMS-PEI is about 55%, 61% and 55%, respectively. The reason for the different total mass loss of the samples may be attributed to the volatilization and decomposition of different amine-modified materials.   The FTIR spectrums of MMS and AMMS in the domain of 400-4000 cm −1 are demonstrated in Figure 5. For MMS, the adsorption peak around 462 cm −1 , 804 cm −1 and 1092 cm −1 could be attributed to the Si-O-Si bending vibrations [33], the symmetric stretching vibrations of Si-O-Si bonds [33], and the asymmetric stretching vibrations of Si-O-Si bonds [34]. Moreover, the adsorption band at 552 cm −1 represents the typical vibration band of five-or six-membered rings of X-O-X, where X can be Al or Si [35]. In addition, the adsorption band at 972 cm −1 is ascribed to the defective Si-OH group [36]. Furthermore, the adsorption bands at 1632 cm −1 and 3448 cm −1 are due to the H-O-H bending vibration and the typical of O-H stretching vibration, respectively [37]. The FTIR spectrum of AMMS displays some new adsorption peaks at 1314 cm cm −1 , 1589 cm −1 , 1673 cm −1 , 2839 cm −1 , and 2958 cm −1 . The peak at 1314 cm −1 is attrib C-N tensile vibration. The adsorption peaks at 1475 cm −1 and 1589 cm −1 represe stretching vibrations which are associated with asymmetric and symmetric bendin primary amines (-NH2). In addition, the adsorption peak at 1673 cm −1 is related bending of secondary amines (-N(R)H) in amine-modified materials. Moreover, sorption peaks at 2839 cm −1 and 2958 cm −1 are due to the CH2 asymmetric and sym stretching modes of the amine chain in the AMMS. Compared with the MMS [ cannot observe the peak at 1632 cm −1 from the FTIR spectrum of AMMS. At the sam the intensity of the peak at 3448 cm −1 decreases with the impregnation of amine. T son for this may be due to the interaction of amine and MMS. Overall, the abov analysis indicates that the amine-modified material is effectively loaded into the p the MMS. The FTIR spectrum of AMMS displays some new adsorption peaks at 1314 cm −1 , 1475 cm −1 , 1589 cm −1 , 1673 cm −1 , 2839 cm −1 , and 2958 cm −1 . The peak at 1314 cm −1 is attributed to C-N tensile vibration. The adsorption peaks at 1475 cm −1 and 1589 cm −1 represent N-H stretching vibrations which are associated with asymmetric and symmetric bending of the primary amines (-NH 2 ). In addition, the adsorption peak at 1673 cm −1 is related to the bending of secondary amines (-N(R)H) in amine-modified materials. Moreover, the adsorption peaks at 2839 cm −1 and 2958 cm −1 are due to the CH 2 asymmetric and symmetric stretching modes of the amine chain in the AMMS. Compared with the MMS [38], we cannot observe the peak at 1632 cm −1 from the FTIR spectrum of AMMS. At the same time, the intensity of the peak at 3448 cm −1 decreases with the impregnation of amine. The reason for this may be due to the interaction of amine and MMS. Overall, the above FTIR analysis indicates that the amine-modified material is effectively loaded into the pores of the MMS.

Adsorption Properties Analysis
CO 2 adsorption experiments were carried out at the required temperature with C 0 = 10 vol.% of CO 2 (inlet CO 2 concentration). The CO 2 adsorption capacities (q a ) of MMS and AMMS are presented in Table 3. In addition, the adsorption curves of MMS and AMMS-T at three temperatures (303 K, 323 K, and 343 K) are shown in Figure 6, and the adsorption curves of the other AMMS are displayed in Figure S3.  It can be seen from Table 3 that the qa of MMS (at different temperatures) is very limited. At the same time, as the temperature increases from 303 K to 343 K, the qa decreases significantly. This may be due to the exothermic behavior of the adsorption phenomenon. However, the qa of AMMS-T increases with increasing temperature. The reason for this may be that the loaded amine-modified material has higher molecular activity and mobility at higher temperatures, so the qa of AMMS shows an increasing trend when the temperature increases from 303 K to 343 K.
From Table 3, we observed that under similar conditions AMMS-T has the best adsorption capacity (qa = 3.32 mmol-CO2/g-adsorbent at 343 K) among all the AMMS. This result is consistent with the results of the SEM analysis. The previous SEM analysis results indicated that among all the surface morphologies of AMMS samples, the surface particle dispersion of AMMS-T was the most consistent and uniform. It is precisely because of the highly uniform dispersion of the amine-modified material particle that agglomeration of the particles is avoided, so that more amine groups of the amine-modified material can contact and react with CO2 molecules, thereby improving the ability to adsorb CO2.
The adsorption capacity of AMMS-T was compared with the values that were obtained under similar conditions for the CO2 adsorbents proposed in the literature (Table  4). We noticed that AMMS-T exhibited a relatively good adsorption capacity under similar conditions. It can be seen from Table 3 that the q a of MMS (at different temperatures) is very limited. At the same time, as the temperature increases from 303 K to 343 K, the q a decreases significantly. This may be due to the exothermic behavior of the adsorption phenomenon. However, the q a of AMMS-T increases with increasing temperature. The reason for this may be that the loaded amine-modified material has higher molecular activity and mobility at higher temperatures, so the q a of AMMS shows an increasing trend when the temperature increases from 303 K to 343 K.
From Table 3, we observed that under similar conditions AMMS-T has the best adsorption capacity (q a = 3.32 mmol-CO 2 /g-adsorbent at 343 K) among all the AMMS. This result is consistent with the results of the SEM analysis. The previous SEM analysis results indicated that among all the surface morphologies of AMMS samples, the surface particle dispersion of AMMS-T was the most consistent and uniform. It is precisely because of the highly uniform dispersion of the amine-modified material particle that agglomeration of the particles is avoided, so that more amine groups of the amine-modified material can contact and react with CO 2 molecules, thereby improving the ability to adsorb CO 2 .
The adsorption capacity of AMMS-T was compared with the values that were obtained under similar conditions for the CO 2 adsorbents proposed in the literature (Table 4). We noticed that AMMS-T exhibited a relatively good adsorption capacity under similar conditions. The abscissa in Figure 6 denotes the time and the ordinate represents the q a . The overall adsorption curve profiles for 303 K, 323 K, and 343 K display similar adsorptive behavior. All the adsorption processes are completed in a short time, and the adsorption rate is approximately the same. This means that although a higher temperature can increase the CO 2 adsorption capacity, the adsorption rate is not controlled by the adsorption temperature. The adsorption rate is further analyzed in the subsequent kinetic analysis of the adsorption process.

Adsorption/Desorption Kinetics Analysis
Previous adsorption capacity studies have shown that AMMS-T exhibits the best adsorption capacity among all AMMS (q a = 3.32 mmol-CO 2 /g-adsorbent, at 343 K). Therefore, AMMS-T was used as a CO 2 adsorbent to analyze the characteristics of the adsorption/desorption kinetics of the CO 2 adsorption/desorption process in this study. Multiple adsorption kinetics models have been employed to quantitatively analyze the adsorption and to explore the adsorption mechanism; for example, Lagergren's pseudo-first-order model [48]; Ho's pseudo-second-order model [49]; the classical intracrystalline diffusion model [50]; and Avrami's fractional-order kinetic model [51]. Among these models, since the fractional order of the Avrami model can be used to characterize the complexity of the reaction mechanism or the simultaneous occurrence of multiple reaction paths [52,53], the Avrami fractional-order kinetic model is well suited to analyzing the CO 2 adsorption kinetics and adsorption mechanisms [54]. For the Avrami model, if n = 1, it means that the adsorption process has uniform adsorption [55,56]. If n = 2, then the adsorption process may show a perfect one-dimensional growth at the adsorption point after uniform adsorption [57].
The kinetics character of CO 2 adsorption on AMMS-T were investigated by isothermal adsorption at several T (303, 323 and 343 K) and at various C 0 (10, 15, 20, and 40 vol.%). The experimental data were simulated using the Avrami model and the results can be found in Figure 7 and Table 5.
As described in Table 5, the adsorption rate constant k a increases with the increase in C 0 at the same T. However, the value of k a only slightly changes with the increase in T under the same C 0 . The high partial pressure of CO 2 can promote the faster diffusion of CO 2 to the active sites on the surface of the solid pores. Therefore, it can be deduced from the simulation results that k a is mainly controlled by the intra-particle diffusion process, and temperature has little effect on the k a . Molecules 2022, 27, x FOR PEER REVIEW 12 of 18 Figure 7. The results of the Avrami kinetics models and the experimental CO2 uptakes of AMMS-T.
As described in Table 5, the adsorption rate constant ka increases with the increase in C0 at the same T. However, the value of ka only slightly changes with the increase in T under the same C0. The high partial pressure of CO2 can promote the faster diffusion of CO2 to the active sites on the surface of the solid pores. Therefore, it can be deduced from the simulation results that ka is mainly controlled by the intra-particle diffusion process, and temperature has little effect on the ka.
It can be seen from Table 5 that most Avrami exponent n are greater than 1 and less than 2, indicating that additional adsorption may be preferentially provided on the existing adsorption sites and resulting in one-dimensional growth at these adsorption points after uniform adsorption. The value of n is less than 2, because as the adsorption progresses, not all the adsorption sites that have adsorbed CO2 start a new uniform one-dimensional adsorption, but only some sites continue to adsorb CO2; the rest do not continue to absorb CO2.
The ka can be demonstrated by the Arrhenius equation: where A denotes the Arrhenius pre-exponential factor, Ea is the activation energy, and R is the universal ideal gas constant. A plot of lnka versus 1/T is shown in Figure 8. Table 6 presents the A and Ea that were obtained by linear regression of the experimental data.  It can be seen from Table 5 that most Avrami exponent n are greater than 1 and less than 2, indicating that additional adsorption may be preferentially provided on the existing adsorption sites and resulting in one-dimensional growth at these adsorption points after uniform adsorption. The value of n is less than 2, because as the adsorption progresses, not all the adsorption sites that have adsorbed CO 2 start a new uniform onedimensional adsorption, but only some sites continue to adsorb CO 2 ; the rest do not continue to absorb CO 2 .
The k a can be demonstrated by the Arrhenius equation: where A denotes the Arrhenius pre-exponential factor, E a is the activation energy, and R is the universal ideal gas constant. A plot of lnk a versus 1/T is shown in Figure 8. Table 6 presents the A and E a that were obtained by linear regression of the experimental data.   As shown in Table 6, the absolute value of Ea decreases with the increase in C0, indicating that a high CO2 partial pressure promotes the adsorption of CO2. The result is consistent with the previous analysis results, that is, the CO2 adsorption rate is mainly controlled by the intra-particle diffusion process. The Ea of AMMS-T is less than that of liquid ammonia absorption (40−50 kJ·mol −1 [58,59]), which indicates that AMMS-T may be a promising and effective CO2 adsorbent. Figure 9 displays the TPD curve, the desorption curve at 383 K and the fitting curve of the Avrami fractional kinetic model. The TPD curve is basically a horizontal straightline segment between 343 K and 353 K, indicating that the adsorbent that was saturated with CO2 had almost no mass loss, which means that when T ≤ 353 K, no desorption process occurs. Then, the TPD curve is a downwardly sloping line segment between 353 K and 383 K, and mass loss begins to appear, indicating that CO2 is gradually desorbed from the surface of the adsorbent. Finally, when the temperature is maintained at 383 K, the TPD curve presents a vertical line segment, indicating that CO2 is quickly and completely desorbed from the surface of the adsorbent. Moreover, according to the desorption curve, the CO2 is almost completely desorbed within 6 min, which means that the desorption rate is very fast. All these results suggest that the optimal desorption temperature for AMMS-T is about 383 K.  As shown in Table 6, the absolute value of E a decreases with the increase in C 0 , indicating that a high CO 2 partial pressure promotes the adsorption of CO 2 . The result is consistent with the previous analysis results, that is, the CO 2 adsorption rate is mainly controlled by the intra-particle diffusion process. The E a of AMMS-T is less than that of liquid ammonia absorption (40−50 kJ·mol −1 [58,59]), which indicates that AMMS-T may be a promising and effective CO 2 adsorbent. Figure 9 displays the TPD curve, the desorption curve at 383 K and the fitting curve of the Avrami fractional kinetic model. The TPD curve is basically a horizontal straight-line segment between 343 K and 353 K, indicating that the adsorbent that was saturated with CO 2 had almost no mass loss, which means that when T ≤ 353 K, no desorption process occurs. Then, the TPD curve is a downwardly sloping line segment between 353 K and 383 K, and mass loss begins to appear, indicating that CO 2 is gradually desorbed from the surface of the adsorbent. Finally, when the temperature is maintained at 383 K, the TPD curve presents a vertical line segment, indicating that CO 2 is quickly and completely desorbed from the surface of the adsorbent. Moreover, according to the desorption curve, the CO 2 is almost completely desorbed within 6 min, which means that the desorption rate is very fast. All these results suggest that the optimal desorption temperature for AMMS-T is about 383 K. From Figure 9, we also noticed that the R 2 values for the fitted Avrami fractionalorder model are 0.99725, which verifies that the Avrami model can describe the experimental data well.

Cyclic CO2 Adsorption/Desorption Behavior of the AMMS-T
The persistent cyclic adsorption/desorption behavior of the adsorbent is essential for long-term operation. Figure 10 describes the adsorption capacity of the AMMS-T in repeated cycles of CO2 adsorption at 343 K and desorption at 383 K. The cycle data indicate that the adsorption performance of the AMMS-T is quite stable, and the adsorption capacity is still about 3.05 mmol g −1 after five adsorption/regeneration cycles.

Conclusions
This study presents an effective amine-functionalized micro-mesoporous silicon adsorbent for CO2 capture. The adsorbent was prepared by the impregnation method using From Figure 9, we also noticed that the R 2 values for the fitted Avrami fractional-order model are 0.99725, which verifies that the Avrami model can describe the experimental data well.

Cyclic CO 2 Adsorption/Desorption Behavior of the AMMS-T
The persistent cyclic adsorption/desorption behavior of the adsorbent is essential for long-term operation. Figure 10 describes the adsorption capacity of the AMMS-T in repeated cycles of CO 2 adsorption at 343 K and desorption at 383 K. The cycle data indicate that the adsorption performance of the AMMS-T is quite stable, and the adsorption capacity is still about 3.05 mmol g −1 after five adsorption/regeneration cycles. From Figure 9, we also noticed that the R 2 values for the fitted Avrami fra order model are 0.99725, which verifies that the Avrami model can describe the mental data well.

Cyclic CO2 Adsorption/Desorption Behavior of the AMMS-T
The persistent cyclic adsorption/desorption behavior of the adsorbent is esse long-term operation. Figure 10 describes the adsorption capacity of the AMMSpeated cycles of CO2 adsorption at 343 K and desorption at 383 K. The cycle data that the adsorption performance of the AMMS-T is quite stable, and the adsorption ity is still about 3.05 mmol g −1 after five adsorption/regeneration cycles.

Conclusions
This study presents an effective amine-functionalized micro-mesoporous sil sorbent for CO2 capture. The adsorbent was prepared by the impregnation metho

Conclusions
This study presents an effective amine-functionalized micro-mesoporous silicon adsorbent for CO 2 capture. The adsorbent was prepared by the impregnation method using micro-mesoporous silicon as a support and TEPA as a modifier. The resulting sample was characterized with the XRD, SEM, TG, nitrogen adsorption and FTIR analysis. The characterization results show that the prepared AMMS-T maintains a micro-mesoporous structure after amine loading, and there are highly uniformly dispersed amine groups distributed on the surface of the pore which is beneficial to the adsorption of CO 2 .
The CO 2 adsorption capacity of AMMS-T displays an upward trend with the increase in temperature in the range of 303-343 K, and the maximum adsorption capacity is 3.32 mmol-CO 2 g −1 -adsorbent. The results of the adsorption kinetic analysis illustrate that Avrami's fractional kinetic model can fit the CO 2 adsorption experimental data well. In addition, the simulation results reveal that the CO 2 adsorption rate of AMMS-T is mainly affected by the intra-particle diffusion rate, and temperature has little effect on it. The optimal regeneration temperature of AMMS-T is about 383 K and the CO 2 desorption process can be well simulated by the Avrami model.
In summary, the prepared AMMS-T shows excellent CO 2 adsorption/desorption performance. The good agreement between the results of the Avrami model and the experimental data shows that the kinetic constants that were obtained by the simulation are highly effective for designing the actual CO 2 adsorption process.

Data Availability Statement:
The data that support the findings of this study are available from the corresponding author, [Liu Y.], upon reasonable request.