CO₂ Adsorption of Aminopropyltrimethoxysilane-and-Tetraethylenepentamine-Co-Modified Mesoporous Silica Gel

Abstract

3-Aminopropyltrimethoxysilane (APTS) and tetraethylenepentamine (TEPA)-co-functionalized mesoporous silica gel (MSG) composites were prepared for CO₂ adsorption. The surface functional groups, thermal stability, and pore structures of the composites were tested using FT-IR, TGA, and N₂ adsorption–desorption techniques, respectively. The effects of the amine loading, adsorption temperature, and influent flow rate on CO₂ adsorption were experimentally investigated. The results indicate that the synergistic effects of APTS, TEPA, and Si-OH on MSG enhanced CO₂ adsorption performance. In addition, the amine-co-modified MSG exhibited good cyclic regenerability and rapid adsorption kinetics.

1. Introduction

Excessive consumption of fossil fuels has triggered significant growth in anthropogenic CO₂ emissions [1,2,3]. According to the National Oceanic and Atmospheric Administration (NOAA), the level of carbon in the air rose from 280 ppm in 1750 to 420 ppm in 2023 [4], triggering severe environmental effects, for example, the greenhouse effect, global warming, and melting glaciers. Considering that fossil fuels will remain our primary energy sources, developing friendly CO₂ capture technology is urgently required.

Various CO₂ capture methods have been proposed for CO₂ separation, for example, solvent absorption [5], adsorption [6,7], membrane sequestration [8], etc. Among these technologies, CO₂ adsorption using amine-functionalized porous solid sorbents has attracted significant attention due to its higher adsorption–desorption rates, lower energy consumption, and the almost insignificant degree of corrosion it triggers. Amine-functionalized porous solid sorbents can be prepared via grafting [9], impregnation [10], and a combination of grafting and impregnation [11,12]. Our group [11] prepared an APTS-and-TEPA-co-functionalized MCM-41 composite through a step combining grafting and impregnation. The sorbent showed good adsorption performance because of the synergistic effect of APTS and TEPA in the MCM-41. Sanz et al. [12] verified that amino groups successively incorporated via a combined method interacted more easily with CO₂ molecules in comparison with a single-amine-modification method.

Currently, the supports used to prepare amine-functionalized porous solid composites are primarily mesopores, such as MCM-41 [11,13], multi-walled carbon nanotubes [14], metal–organic frameworks (MOFs) [15], nano-silica [16], etc. The costs of these supporting materials are very high, and these materials have been estimated to account for the vast majority of all the sorbents [16]. To address this situation, several relatively inexpensive materials have been explored, including carbon-based materials [17,18] and Al₂O₃ [19]. Yan et al. [18] estimated the cost for 30 wt.% PEI-modified clover leaf-shaped Al₂O₃ to be USD 30/kg, which is significantly below that for an SBA-15-based composite, namely, USD 760/kg. However, the maximum adsorption capacity of CA/30 is only 1.13 mmol g⁻¹, which has limited its practical application in large-scale CO₂ capture. Therefore, the development of new, inexpensive amine-functionalized porous materials with high performance is vital for CO₂ capture.

MSG is a commercialized mesoporous sieve with a highly developed surface area, mesopores and surface hydroxyl groups, and good mechanical stability, making it suitable for loading organic amines for CO₂ adsorption. In addition, compared to mesoporous sieves such as MCM-41 and SBA-15, the price of MSG is relatively cheap. In this study, MSG was selected as a support, APTS and TEPA were used as an active site provider, and a step combining grafting and impregnation was adopted to prepare APTS-and-TEPA-co-modified MSG sorbents. The adsorption and regeneration processes were evaluated in a fixed-bed reactor. In addition, the CO₂ adsorption kinetics were also analyzed.

2. Materials and Methods

2.1. Materials

MSG was purchased from Qingdao Bangkai Separation Materials Co., Ltd. (Qingdao, China). APTS with a purity of 95% and TEPA with a purity of 90% were sourced from Tianjin BASF Chemical Co., Ltd. (Tianjin, China). Anhydrous ethanol (purity ≥ 99.5%) was supplied by Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). N₂ (99.999%) and the simulated flue gas (85 vol.% N₂ + 15 vol.% CO₂) were obtained from Weiyang Gas Co., Ltd. (Weifang, China). The high-purity nitrogen gas (99.999% N₂) and the simulated flue gas mixture (85 vol.% N₂ balanced with 15 vol.% CO₂) used in this study were commercially procured from Weiyang Gas Co., Ltd.

2.2. Sorbent Preparation

The APTS-and-TEPA-co-modified MSG was prepared using a combination of grafting and impregnation. The details of this process are as follows.

Firstly, APTS was grafted onto MSG. Dried MSG (2 g) was added to a three-necked flask containing 200 mL of anhydrous ethanol, and the suspension was continuously stirred to make sure the MSG was sufficiently dispersed. Following the addition of 0.4 mL of deionized water, the reaction mixture was further stirred for 30 min to ensure complete homogenization. Subsequently, the temperature was increased to 70 °C, and a specific volume of APTS was introduced dropwise into the solution, which was continuously refluxed for 10 h. The synthesized powder was subjected to three successive anhydrous ethanol washes and then oven-dried at 80 °C (12 h). The dried composite sample was designated as APTSx-MSG, where x indicates the weight percentage of APTS in the APTSx-MSG.

Secondly, TEPA was impregnated onto the APTSx-MSG. A specific volume of TEPA was added to 20 mL of anhydrous ethanol and stirred for 30 min. Then, 2 g of APTSx-MSG was introduced, and the slurry was stirred for another 3 h. Subsequently, the resulting white powder was oven-dried at 80 °C (12 h). The obtained composites were named TEPAy-APTSx-MSG, where y is the weight percentage of TEPA in the TEPAy-APTSx-MSG.

2.3. Characterization

The FT-IR spectra were acquired using a FT-IR Bruker Tensor-27 spectrometer (Bruker, Karlsruhe, Germany). The spectral parameters were systematically collected at a spectral resolution of 4 cm⁻¹ in a range from 4000 cm⁻¹ to 400 cm⁻¹.

The thermal stability of the sorbents was measured using a thermogravimetric analysis instrument (TA, Newcastle, DE, USA). The analysis was conducted under a N₂ atmosphere (100 mL min⁻¹) with a linear heating regime increasing from 30 to 600 °C at 10 °C min⁻¹. The N₂ adsorption–desorption isotherms were obtained at 77 K using an ASAP 2460 (Micromeritics, Norcross, GA, USA). Prior to gas adsorption analysis, the sorbents were degassed at 80 °C under high-vacuum conditions for 12 h. Specific surface area determinations were performed by applying the Brunauer–Emmett–Teller (BET) theory, and the pore size distribution was derived from the desorption isotherm using the Barrett–Joyner–Halenda (BJH) method.

2.4. Adsorption and Regeneration

The CO₂ adsorption and regeneration processes were evaluated in a self-assembled fixed-bed reactor (200 mm × 10 mm, i.d.), as shown in Figure 1. Before adsorption, 2 g of the sample was activated at 100 °C for 1 h under a nitrogen stream. Subsequently, the fixed bed was cooled to a predetermined adsorption temperature, and the simulated flue gas, at a flow rate of 30 mL min⁻¹, was introduced to displace the N₂ carrier gas; simultaneously, the outlet CO₂ concentration was recorded using a gas chromatograph. When the outlet CO₂ concentration had reached 5% of its initial inlet concentration, the adsorption process finished with a breakthrough stage, where the time and adsorption capacity were denoted as breakthrough time and breakthrough adsorption capacity, respectively. When the outlet CO₂ concentration was identical to its initial inlet concentration, the adsorption reached an equilibrium stage, and the adsorption capacity was the saturated adsorption capacity. Soon afterwards, the sample was heated at 100 °C in N₂ gas, and ten CO₂ adsorption–desorption cycles were induced.

The CO₂ adsorption capacity was determined using the following Equation (1),

$$q_t={\displaystyle \frac{Q\times {\displaystyle {\int }_0^t(C_0-C)dt}}{m}}$$

(1)

where q_t is the CO₂ adsorption capacity at time t, given in mmol g⁻¹; Q is the influent flow rate, in mL min⁻¹; C₀ and C are the influent and effluent CO₂ concentrations, respectively, in mmol L⁻¹; m is the weight of the sorbent, in g; and t is the adsorption time, in min.

3. Results and Discussion

3.1. Characterization

The FT-IR spectra for the prepared samples used to investigate the functional group changes of the MSG before and after amine co-modification are shown in Figure 2. The obvious band at 3450 cm⁻¹ for all the investigated samples can be attributed to the stretching vibration mode of the Si-OH group and physically adsorbed water molecules, and the bands at 1093, 800, and 473 cm⁻¹ were assigned to the asymmetric and symmetric stretching vibrations and bending vibrations of Si-O-Si, respectively [20], suggesting that the amine modification did not change the framework structure of MSG. After APTS grafting, two new bands at 1569 and 1475 cm⁻¹ appeared, which correspond to the asymmetric and symmetric stretching vibrations of -NH₂ from APTS, respectively [21]. For the TEPA-modified APTS-MSG, the two bands appearing at 2930 and 2850 cm⁻¹ were attributed to the symmetric and asymmetric stretching vibrations of -CH- in TEPA [22], suggesting that TEPA was successfully impregnated into APTS-MSG. Moreover, the bands at 1475 and 1569 cm⁻¹ intensified with the increase in the TEPA loading, suggesting that more TEPA was introduced.

Figure 3 displays the weight-loss profiles of MSG before and after amine modification. There was no remarkable mass loss for MSG before 600 °C, demonstrating the superior thermal resistance of the MSG support. Two obvious mass losses were observed for the amine-functionalized MSG. The first mass loss, occurring below 100 °C, was due to the desorption of the adsorbed CO₂ and moisture in the air, and the second mass loss, occurring over 100 °C, can primarily be ascribed to the thermal decomposition of TEPA and APTS. The starting decomposition temperatures for TEPA40-MSG and APTS30-MSG were 130 and 280 °C, respectively, while this temperature for TEPA30-APTS30-MSG was 150 °C, with the value ranging between that for MSG and APTS30-MSG. The alkanolamine of APTS was chemically loaded onto the pore surface of the MSG. After physical impregnation with TEPA, the amino functional groups from TEPA engaged in hydrogen bonding interactions with the hydroxyl groups from APTS and the Si-OH in MSG; a new reticular structure formed, which helped to disperse TEPA and diffuse CO₂; and more CO₂ molecules were captured [11]. Thus, the thermal stability of the APTS-and-TEPA-co-modified MSG was superior to that of the TEPA-modified MSG.

Figure 4 shows the N₂ adsorption–desorption isotherms of the MSG-based composite sorbents. According to the IUPAC classification, all the samples characterized demonstrated a typical IV adsorption–desorption isotherm, and the state of the MSG before and after modification suggests mesoporous structure [10,11]. Corresponding textural properties are presented in

Table 1. Textural properties of the MSG-based composite sorbents.

Sorbent	SBET [m2 g−1]	Pore Volume [cm3 g−1]	Pore Diameter [nm]
MSG	334.12	0.93	8.07
APTS30-MSG	237.26	0.67	7.51
TEPA20-APTS30-MSG	123.33	0.41	7.88
TEPA30-APTS30-MSG	75.21	0.28	8.77
TEPA40-APTS30-MSG	38.08	0.18	9.84
TEPA40-MSG	93.41	0.32	8.25

. The specific surface area and total pore volume of MSG were 334 m² g⁻¹ and 0.93 cm³ g⁻¹, respectively, while these values were 237 m² g⁻¹ and 0.67 m³ g⁻¹ for APTS30-MSG, 75 m² g⁻¹ and 0.28 m³ g⁻¹ for TEPA30-APTS30-MSG, and 93 m² g⁻¹ and 0.32 m³ g⁻¹ for TEPA40-MSG, respectively, with the corresponding values being lower than those for the pure MSG. In addition, the specific surface area and pore volume all decreased with the decrease in the amine loading, which revealed that more active components were immobilized in the MSG and that the pores in the MSG had been filled [23]. Thus, we successfully prepared the composite sorbents by combining grafting and impregnation.

3.2. CO₂ Adsorption Performance

3.2.1. Effect of the Amine Loading

To determine the optimal APTS and TEPA loadings, different loadings of APTS-and-TEPA modified MSG were prepared to investigate the gel’s CO₂ adsorption performance. The adsorption temperature and influent velocity were 70 °C and 30 mL min⁻¹, respectively, and the adsorption capacity data are provided in

Table 2. The CO₂ adsorption capacity of MSG before and after modification.

Sample	Breakthrough Time [min]	Breakthrough Adsorption Capacity [mmol g−1]	Saturated Adsorption Capacity [mmol g−1]
MSG	2	0.20	0.35
TEPA10-MSG	6	0.60	0.76
TEPA20-MSG	10	1.01	1.22
TEPA30-MSG	16	1.61	1.83
TEPA40-MSG	18	1.81	2.21
TEPA50-MSG	14	1.41	1.96
APTS20-MSG	8	0.80	1.09
TEPA20-APTS20-MSG	12	1.21	1.68
TEPA30-APTS20-MSG	22	2.21	2.38
TEPA40-APTS20-MSG	26	2.61	2.83
TEPA50-APTS20-MSG	16	1.61	1.93
APTS30-MSG	10	1.01	1.11
TEPA10-APTS30-MSG	14	1.41	1.73
TEPA20-APTS30-MSG	20	2.01	2.19
TEPA30-APTS30-MSG	28	2.81	3.04
TEPA40-APTS30-MSG	22	2.21	2.40

. As shown in Table 2, the CO₂ adsorption capacity of the modified MSG, especially the APTS-and-TEPA-co-modified MSG, was much higher than that of the pure MSG, showing that amine modification remarkably enhanced the adsorption performance of the MSG. These findings are indicative of an acid–base reaction in which the primary and secondary amines reacted with CO₂ via a zwitterion mechanism to form carbamates, which allowed more CO₂ molecules to be captured [24,25].

For the TEPA-modified MSG, both the breakthrough and saturated adsorption capacity first increased and then decreased as the TEPA loading increased from 10 to 50%, and TEPA40-MSG showed a maximum saturated adsorption capacity of 2.21 mmol g⁻¹. However, when the TEPA loading was further increased to 50 wt.%, the CO₂ adsorption capacity exhibited a notable decline, reaching 1.96 mmol g⁻¹. This trend agrees well with previously reported results [26,27]. The amines’ active sites allowed more CO₂ to be captured, but an excessive amount of amine can also block the pores or agglomerate on the inner surface of MSG, which, in this case, impaired both the uniform distribution of the amine active sites and the effective diffusion of CO₂ molecules [27]. For the APTS-modified MSG, the maximum saturated adsorption capacity was 1.11 mmol g⁻¹.

For TEPA40-APTS20-MSG and TEPA30-APTS30-MSG, the saturated adsorption capacities were 2.83 and 3.04 mmol g⁻¹, respectively, which are remarkably greater than those of the TEPA or APTS co-functionalized MSG. This performance enhancement can primarily be attributed to the synergistic interaction between APTS, TEPA, and Si-OH on the pore surface of MSG [11,28]. APTS reacted with the abundant hydroxyl groups within the pores of MSG, forming Si-O-Si networks, which improved TEPA dispersion and exposed more high-affinity CO₂-binding sites.

3.2.2. Effect of the Adsorption Temperature

Adsorption temperature serves as a critical determinant in CO₂ capture processes, and the CO₂ adsorption performance of TEPA30-APTS30-MSG was systematically evaluated at 40, 55, 70, and 85 °C. The (a) CO₂ breakthrough curves and (b) adsorption capacity of TEPA30-APTS30-MSG are shown in Figure 5. As evidenced by the data in Figure 5, both the breakthrough time and adsorption capacity initially increased as the temperature increased from 40 to 70 °C and then decreased as the temperature further increased to 85 °C, and the maximum breakthrough and saturated adsorption capacities were 2.81 and 3.04 mmol g⁻¹ at 70 °C, respectively.

According to previous research, the CO₂ adsorption process is exothermic, and a relatively low temperature is helpful for CO₂ adsorption in thermodynamics [11]. However, a low temperature not only hinders the dispersion and exposure of the actives sites of TEPA but also limits the movement of CO₂ molecules [29,30]; with the increase in the adsorption temperature, the active sites became more flexible and evenly dispersed, and the reduction in CO₂ diffusion in the pores of MSG was reduced, indicating that a relatively high temperature is positively conducive in terms of dynamics. Therefore, both thermodynamics and dynamics play important roles in the CO₂ adsorption process, and 70 °C was identified as the most favorable adsorption temperature in the subsequent study.

3.2.3. Effect of Influent Flow Rate

The CO₂ adsorption performance of TEPA30-APTS30-MSG at different influent flow rates was examined, and the (a) CO₂ breakthrough adsorption curves and (b) adsorption capacity of TEPA30-APTS30-MSG are shown Figure 6. When the influent flow rate was 30 mL min⁻¹, the breakthrough time was 26 min. As the influent flow rate increased, the breakthrough time decreased to 12 min at 40 mL min⁻¹, 8 min at 50 mL min⁻¹, and 4 min at 60 mL min⁻¹, and the corresponding breakthrough and saturated adsorption capacity significantly decreased. A high flow rate is favorable for reaching adsorption equilibrium due to the corresponding low mass transfer resistance [31]. However, as the influent flow rate increased, the optimal contact duration of CO₂ within the sorbent was significantly shortened, and CO₂ molecules could not sufficiently make contact with the amino active sites. So, a 30 mL min⁻¹ influent flow rate is optimal in the studied range.

The CO₂ adsorption capacity of TEPA30-APTS30-MSG was compared with the reported values in

Table 3. The optimal CO₂ adsorption capacity for the MSG-based composite sorbents.

Support	Amine	Loading [%]	Temperature [°C]	CO2 Partial Pressure [kPa]	Adsorption Capacity [mmol g−1]	Ref.
MSG	PEI	33	25	—	1.07	[32]
MSG	PEI	30	75	15	2.12	[22]
MSG	Poly- AAM	39	60	17	0.64	[33]
MSG	AEATPMS	28	70	60	0.58	[34]
MSG	TEPA	40	70	15	2.21	Present work
MSG	TEPA/ APTS	30/30	70	15	3.04	Present work

: the former showed significantly superior performance, indicating that a combination of grafting and impregnation is suitable for preparing MSG-supported composite sorbents.

3.3. Regeneration

A robust regenerative capacity serves as a critical indicator for assessing an adsorbent in industrial applications. In this study, TEPA30-APTS30-MSG was subjected to ten adsorption–desorption cycles, and the adsorption and desorption temperatures were 70 and 100° C, respectively. The saturated adsorption capacity for TEPA30-APTS30-MSG after every regeneration is indicated in Figure 7. As shown in the figure, the adsorption capacity of TEPA30-APTS30-MSG showed only a slight change, with the value decreasing from 3.04 mmol g⁻¹ for the fresh sorbent to 2.95 mmol g⁻¹ after ten regenerations, and reduced by only 2.96%, potentially because of TEPA’s volatilization. Therefore, our combined-technique-prepared MSG-supported composite sorbent showed good regeneration performance.

3.4. CO₂ Adsorption Kinetics

To gain deeper insights into the adsorption processes of the MSG-supported composite sorbent, the pseudo-first-order, pseudo-second-order, and Avrami models were selected to analyze the experimental data. In general, the pseudo-first-order and pseudo-second-order models are based on the hypothesis that the adsorption processes pertaining to the sorbents are physisorption and chemisorption, respectively, and the Avrami model is grounded in the fundamental assumption that the adsorption process is the combination of physisorption and chemisorption [35,36,37]. The models’ equations are presented below.

$$\mathrm{Pseudo}\text{-}\mathrm{first}\text{-}\mathrm{order} \mathrm{model}\text{:} q_t=q_e(1-e^{-k_{f}t})$$

(2)

$$\mathrm{Pseudo}\text{-}\mathrm{second}\text{-}\mathrm{order} \mathrm{model}\text{:} q_t={\displaystyle \frac{k_s{q_e}^2}{1+k_s{q}_{e}t}}t$$

(3)

$$\mathrm{Avrami}\text{:} q_t=q_e[1-e^{-(k_{a}t)n_a}]$$

(4)

q_t (mmol g⁻¹) and q_e (mmol g⁻¹) are the adsorption capacity at time t and equilibrium time, respectively; k_f (min⁻¹), k_s (g mmol⁻¹ min⁻¹) and k_a (min⁻¹) are the rate constant; n_a is the Avrami exponent.

The fitting results at (a) 40 °C, (b) 55 °C, and (c) 70 °C are shown in Figure 8 and

Table 4. The fitting parameters for TEPA30-APTS30-MSG yielded by three kinetic models at different temperatures.

Kinetic Model	Parameter	40 °C	55 °C	70 °C
Pseudo-first-order model	qe,cal (qe,exp)	2.39 (2.18)	2.67 (2.37)	4.04 (3.04)
	kf	0.0351	0.0602	0.0689
	R2	0.9606	0.9615	0.9714
Pseudo-second-order model	qe,cal (qe,exp)	3.19 (2.18)	3.68 (2.37)	6.26 (3.04)
	ks	0.0038	0.0134	0.0188
	R2	0.9401	0.9450	0.9656
Avrami	qe,cal (qe,exp)	2.21 (2.18)	2.41 (2.37)	3.20 (3.04)
	ka	0.0523	0.0703	0.0768
	na	1.6706	1.6645	1.5888
	R2	0.9926	0.9920	0.9912

. As can be seen from the figures, the adsorption data for TEPA30-APTS30-MSG suggest less deviation from the Avrami model compared to that for the pseudo-first-order and pseudo-second-order models, and the values of R² for Avrami fitting range from 0.9912 to 0.9926, which are significantly higher than those for the other two models; therefore, the Avrami model demonstrated superior fitting performance with respect to the experimental data. The adsorption process for the APTS-and-TEPA-co-functionalized MSG was not a singular process of either physisorption or chemisorption but a combined and complex process [11].

In Table 4, the kinetic constant of k_a increased from 0.0523 to 0.0768 min⁻¹ as the temperature increased from 40 to 70 °C, suggesting that in the studied temperature range, dynamic factors play a leading role in CO₂ adsorption process, and these results are consistent with those in Section 3.2.2.

4. Conclusions

APTS and TEPA co-modification was confirmed to be an effective method of preparing the MSG-supported composite sorbents for CO₂ capture. The CO₂ adsorption capacity of TEPA30-APTS30-MSG reached 3.04 mmol g⁻¹ at 70 °C, and the influent flow rate was 30 mL min⁻¹, which is higher than that of the TEPA- or APTS-modified MSG. In addition, the adsorption capacity reduced by 2.96% after ten regenerations. The kinetic fitting of CO₂ adsorption on TEPA30-APTS30-MSG indicated that the adsorption process involved both physisorption and chemisorption mechanisms and that dynamic factors play a leading role in the studied temperature range.