Rising CO2 levels in the atmosphere are a critical concern for global climate change and environmental risks. By 2100, it could increase to 26 billion tons per year. This forecast has consequently emphasized the importance of prioritizing the capture and adsorption of CO2 from an environmental perspective1. The ongoing release of CO2 into the atmosphere has resulted in long-term changes in the global climate, including rising temperatures, sea levels, and more frequent occurrences of extreme weather events. Four primary methods for CO2 separation have emerged: absorption, adsorption, cryogenic, and membrane technologies2,3. The choice of the suitable CO2 capture technique relies on several factors, encompassing the origin of CO2, the magnitude of the capture process, the desired level of purity for the captured CO2, and the envisioned application of the captured CO24. At present, absorption and adsorption represent the prevailing methods employed for CO2 capture, whereas cryogenic and membrane technologies are still in their nascent stages of development5. Researchers are actively engaged in investigating diverse methodologies to adsorb CO2 as a means of mitigating its emissions6. Porous materials such as zeolite7, silica8, MOF9, carbon10, and polymer11 have been used to adsorb CO2, each of which has its advantages and disadvantages.

Zeolite is a material with a crystalline structure that can be either naturally occurring or synthesized12. It contains aluminosilicate minerals and exhibits a distinctive three-dimensional framework with well-organized pores and channels. Zeolites have crystal structures with a rigid framework that includes pores and channels formed as TO4, where T can be silica and aluminum. Aluminum atoms attract the oxygens and produce an excellent site for cation transfer13. Cation in the structure of zeolites plays a crucial role in CO2 capture because it can attract CO2 into the zeolite6. Zeolites are promising CO2 adsorbents with high surface area, suitable pore size, and excellent thermal and chemical stability14. Multiple types of zeolites have undergone thorough investigation to assess their potential in adsorbing CO2 gas generated from industrial processes. Zeolite 4A15, zeolite 13X16, ZK-517, ZSM-518, β-zeolite19, and Na-X20 are among the zeolite types that have demonstrated potential in applications related to CO2 capture. These zeolites possess distinctive pore structures, substantial surface areas, and excellent thermal stability, rendering them highly desirable options for CO2 adsorption. Zeolite 13X has demonstrated exceptional selectivity for CO221. ZK-5 possesses a distinctive cage-like structure that can be modified to improve its adsorption properties for CO222. Similarly, ZSM-523 and β-zeolite24 have exhibited significant CO2 adsorption capacity in the studies. Additionally, Na-X has shown good stability and regeneration properties25. Zeolite 4A is characterized by a substantial concentration of adsorption sites attributable to the presence of aluminum atoms within its framework. These sites exhibit a robust affinity towards CO2 molecules, facilitating effective capture and retention of the gas.

There are several methods for synthesizing zeolites, such as hydrothermal synthesis, sol–gel synthesis, microwave-assisted synthesis, and organic template synthesis. The hydrothermal Method is the most commonly used technique for synthesizing zeolite when working with kaolin26. Kaolin-based zeolites have high adsorption capacity due to their combination of mesoporous and microporous structures. Its natural abundance makes it a cost-effective option for large-scale applications, and its environmentally friendly properties make it a sustainable choice for CO2 capture solutions. The unique characteristics of kaolin-based zeolite allow for customization and tailored modifications, resulting in enhanced CO2 capture performance27,28.

In recent literature, the modification of zeolites with amine functional groups was studied, and it was demonstrated to enhance their CO2 capture capabilities27,29,30,31,32,33,34. Other modifications, such as carbon modification35, silica modification7, MOF modification36, acid treatment37 and ion exchange38 have also been investigated, showcasing their potential in improving the CO2 adsorption performance of zeolites. Structural modifications achieved through various techniques, as demonstrated in relevant research studies39,40, can improve the adsorption performance of Zeolite 4A. presents a comprehensive analysis of the advantages and disadvantages of these modifications.

Table 1 presents a comprehensive analysis of the advantages and disadvantages of these modifications.

Table 1 Comparison of different methods of zeolite modification.

The incorporation of amines onto the surface of zeolites can significantly enhance their CO2 adsorption capabilities. The interaction between amine-modified zeolites and CO2 occurs through chemisorption, wherein a chemical bond forms between the CO2 molecule and the amine group on the zeolite surface43. As mentioned in presents a comprehensive analysis of the advantages and disadvantages of these modifications.

Table 1, amine-modified zeolites offer multiple advantages, including increased CO2 adsorption capacity, selective CO2 capture, High regeneration, and low energy consumption44. Amine functionalization of zeolites can be achieved through two main methods: grafting and impregnation. The grafting method is a well-established approach that involves attaching amine-containing molecules to the zeolite surface via covalent bonds45. This method typically results in higher degrees of functionalization, increased stability of the amine groups, and improved selectivity. However, there may be a reduction in CO2 adsorption capacity and difficulties in regeneration due to high pressure46. Compared to the grafting method, impregnation is less complex and easier to implement. It allows for a straightforward introduction of functional groups onto the zeolite surface, making it a practical choice for modifying the material45. Impregnation method involves immersing the zeolite in a solution containing the desired amine compound to deposit amine-containing molecules onto its surface. After impregnation, the zeolite undergoes washing and drying. Impregnation is a straightforward and versatile approach, but it may result in lower levels of functionalization and less stable amine groups compared to grafting. Amine impregnation involves incorporating amines, such as MEA, DEA, and TEPA, into the pores of a zeolite material47. Fashi et al. utilized 2% piperazine to modify zeolite 13X and improve its CO2 adsorption capabilities48. Babaei et al. examined Na-Y zeolite with a silicon-to-aluminum ratio of 2.5, utilizing varying quantities of amine. When comparing NaY-2-MAE to NaY-2-DEA, they found that the steric barrier is reduced in NaY-2-MAE, leading to increased adsorption. Moreover, five functional groups in TEPA resulted in higher adsorption levels49. Ahmad et al. studied the modification of zeolite β by incorporating melamine to enhance its CO2 adsorption performance. The modified zeolite demonstrated a significant CO2 adsorption capacity of 162.36 mg/g at 298 K and 1 bar, attributable to the increased number of active sites and the improved hydrophobicity of the zeolite surface resulting from the modification50. Panda et al. worked on modifying zeolite 4A with different amines such as propylene amine, butyl amine, pentyl amine, isopropyl amine, isobutyl amine, and isopentyl amine. The optimum result of modified-zeolite by butylamine and iso-butylamine was 108.68 and 112.64 mg/g at 298 K and 1 bar51. Garshasbi et al. prepared 13 × zeolite and acid modification of Iranian kaolin, which showed an adsorption capacity of 352 mg/g21. Thakkar et al. synthesized ZSM-5, Y, and SAPO-34 zeolites using kaolin and modified them using TEPA amine to increase CO2 adsorption27. Murge et al. synthesized and modified zeolite Y by amine TEPA, the best adsorption performance related to Z-Y-3 at 303 K and 1 bar was 114 mg/g52.

Table 2 presents an overview of the studies and experiments on various amines for modifying zeolite structures. The modifications involved alterations in the chemical structure and properties of the zeolites, such as pore size, surface area, and functionality, leading to changes in the adsorption capacity and selectivity. The tabulated results offer insights into the potential of amines as modifying agents for enhancing the adsorption performance of zeolites, as well as the conditions required for achieving optimal results Table 2.

Table 2 Review of studies on different modified zeolites with amine for CO2 adsorption.

Researchers frequently utilize the response surface methodology (RSM) as a statistical modeling technique to better understand the behavior of chemical systems and improve their performance. RSM is considered a valuable tool for optimizing chemical processes56. Pashaei et al. demonstrated the effectiveness of RSM in optimizing the CO2 absorption process into piperazine solutions5. Gill et al. used the RSM to evaluate the effect of parameters such as activation temperature and burning degree on CO2 absorption capacity57. Karimi et al. used the RSM method to model the CO2 adsorption capacity by modifying a commercially activated carbon58. Khajeh et al. used RSM to optimize the operational conditions, reactor temperature and pressure, and acid concentration for activating the surface and wt% NaOH to raise the adsorption capacity performance59. The influence of amine loading on the adsorbent structures derived from kaolin-modified zeolite for CO2 capture has received limited attention in previous studies, especially using RSM. This article aims to fill this research gap by investigating the effects of amine loading on these structures and evaluating their capability for CO2 adsorption.

In this study, we aim to enhance the CO2 adsorption capacity of zeolite synthesized from kaolin. Our focus was on using two specific amines, Tetraethylenepentamine (TEPA) and Diethanolamine (DEA), as modifiers for the zeolite. RSM was utilized to optimize the experiments and operating conditions for the modified zeolites and evaluate their CO2 adsorption capability. Furthermore, we analyzed the kinetic and thermodynamic characteristics of CO2 capture of adsorbents.

Material and method


Iranian Kaolin was employed to synthesize 4A-zeolite. Sodium hydroxide (NaOH) and methanol were procured from Merck. Tetraethylenepentamine (TEPA) and diethanolamine (DEA), both of analytical grades, were used as amines during the adsorbent synthesis and obtained from Sigma Aldrich.

4A-zeolite synthesis

Zeolite 4A was synthesized using a prescriptive method described in reference27. The synthesis process involved the calcination of Iranian Kaolin at a temperature of 600 °C for 2 h, with a heating rate of 5 °C/min. Subsequently, 5 g of metakaolin were immersed in 100 mL of 2M NaOH solution in a round-bottom flask, and treated by stirring under reflux for 48 h at a temperature of 100 °C. The resulting mixture was then washed with deionized water until the pH reached 7, after which it was dried at 100 °C for 12 h. The resulting product was a highly porous 4A-zeolite. The synthesis process is illustrated in Fig. 1.

Figure 1
figure 1

Schematic of 4A-zeolite synthesis.

Impregnation of 4A-zeolite by DEA and TEPA

The optimal conditions for CO2 adsorption experiments, including temperature, pressure, and amine loading, were determined using Design Expert software to create an experimental design. Modified zeolites were then prepared based on the data generated by the software, with an amine loading of 5–25 wt%, to meet the specific requirements of the experimental design. For preparing amine-modified zeolite, we used the wet impregnation route. In this method, 4A-zeolite was modified with five different loadings of 5, 10, 15, 20, and 25 wt% tetraethylenepentamine (TEPA) and DEA (diethanolamine). In the preparation of 4A-25%TEPA, a solution comprising 0.33 g of TEPA and 100 mL of methanol was mixed and subjected to stirring for 20 min at 60 °C. One gram of prepared 4A-zeolite was added to the mixed solution. The solution was continuously stirred for 4 h at 500 rpm in a 100 mL beaker. The resulting mixture is dried at 100 °C for 12 h and placed in an oven to obtain a soft white powder. We repeated the process for DEA loadings of 5, 10, 15, 20, and 25 wt%. The schematic of this process is shown in Fig. 2. Prepared samples were denoted 4A-5%TEPA, 4A-10%TEPA, 4A-15%TEPA, 4A-20%TEPA, and 4A-25%TEPA. These steps are repeated for DEA.

Figure 2
figure 2

Modification of 4A-Zeolite from kaolin with TEPA and DEA.


The N2 adsorption/desorption method is a scientific method used to determine the surface area of solid materials. This test is used for measuring the adsorption of gas molecules onto the zeolite's surface at varying pressures, which is then used to calculate the specific surface area using the BET equation60. This test was achieved by the ASAP 2020 model at 77 K. The FTIR (Fourier Transform Infrared Spectroscopy) test is a popular analytical technique used in various fields to analyze the chemical composition of a zeolite. The test involves passing infrared radiation through a sample and measuring its absorption or transmission at different wavelengths to determine the chemical bonds present in the sample. The infrared spectrum generated from the analysis presents a distinctive identifying characteristic of the sample's composition, thus enabling the identification of unfamiliar compounds, assessment of purity, and tracking of chemical reactions. X-ray diffraction (XRD) analysis is a technique that is used for the investigation of crystal structure in materials across a broad spectrum is facilitated. This test determines the composition, purity, crystallinity, and phase identification of kaolin, zeolite, and modified zeolite60,61. The test device for this analysis worked at 40 mA and 40 kV. Scanning Electron Microscopy (SEM) is an electron microscopy technique that enables the acquisition of high-resolution images of the surface of a zeolite. The resulting image provides detailed information about the zeolite's morphology, topography, and composition with sub-nanometer resolution.

Gas adsorption setup

This study evaluated the quantity of CO2 adsorbed on zeolites through a CO2 adsorption test pilot, as depicted in Fig. 3. Initially, a 0.5 g sample was loaded into the reactor, and a vacuum was produced using a vacuum pump. Subsequently, high-purity N2 was introduced into the chamber for 30 min before introducing CO2 gas, which flowed over the adsorbent for 3600 s. The experiments were performed at various pressures and temperatures. During the investigation, the temperature of the CO2 gas was regulated using an electric heater, and changes in temperature and pressure were continuously recorded by a computer. Upon achieving equilibrium, which took approximately one hour, the device recorded the internal pressure (Pf). Subsequently, the adsorption parameters were determined using the recorded data62. The experiments were repeated three times, and the data average was reported to minimize experimental error.

Figure 3
figure 3

Adsorption pilot to measure CO2 capture.

The CO2 adsorption percentage and adsorption capacity of adsorbents are calculated using Eqs. (1) and (2), respectively.

$$\% \, \mathrm{ Adsorption}=\frac{{\mathrm{P}}_{\mathrm{i}}-{\mathrm{P}}_{\mathrm{f}}}{{\mathrm{P}}_{\mathrm{i}}}\times 100$$
$$\mathrm{q}=\frac{{\mathrm{m}}_{\mathrm{i}}-{\mathrm{m}}_{\mathrm{f}}}{{\mathrm{m}}_{\mathrm{i}}} =\frac{{(\mathrm{P}}_{\mathrm{i}}-{\mathrm{P}}_{\mathrm{f}}){\mathrm{VM}}_{\mathrm{CO}2}}{\mathrm{RTmZ}}$$

The equation involves Pi as the initial pressure, Pf as the equilibrium pressure, V is the reactor volume, \({M}_{CO2}\) is the molecular weight of CO2, R is a gas constant, m is the mass adsorbent, T is the temperature, and Z is the compressibility coefficient. The compressibility factor is obtained from virial equations [Eqs. (3)–(6)]. Table 3 displays the specific properties and calibration details of the developed unit.

Table 3 Specific properties and calibration details of the developed unit.
$$\frac{{\mathrm{BP}}_{\mathrm{c}}}{{\mathrm{RT}}_{\mathrm{c}}}={\mathrm{F}}^{(0)}\left({\mathrm{T}}_{\mathrm{R}}\right)+\upomega {\mathrm{F}}^{(1)}\left({\mathrm{T}}_{\mathrm{R}}\right)$$

Response surface methodology (RSM)

RSM is the statistical technique to model and optimize complex relationships between multiple input variables and output responses56. The investigation conducted in this research involved an examination of the effects of various factors, such as temperature, pressure, and loading percentage of two different amines on 4A-zeolite, to enhance the performance of CO2 capacity. RSM was implemented to optimize these factors, utilizing a central composite design (CCD) based on a four-factor approach, which included temperature (A), pressure (B), amine wt% (C), and amine type (D), as detailed in Table 4.

Table 4 The process factors in the RSM modeling for 4A-zeolite with TEPA and DEA.

Total of 52 tests (Table S1 in the supplementary) were conducted under varying operating conditions, and the resulting experimental data was used to establish the relationship between the X variable and the Y response through a design model [Eq. (7)]

$$y={a}_{0}+{\sum }_{i=1}^{n}{a}_{i}{x}_{i}{\sum }_{i=1}^{n}{a}_{ii}{x}_{i}^{2}+{\sum }_{i}^{n}{\sum }_{j}^{n}{a}_{ij}{x}_{i}{x}_{j}+\varepsilon$$

In this formula [Eq. (7)] y is the dependent variable, a0 is the intercept, ai is the regression coefficient for the n predictor variable xi, and aii is the coefficient for the squared term of each predictor, aij is the coefficient for the interaction terms, and ε represents the residual error term. The formula suggests that the dependent variable y is a function of the importance of the predictor variables, the regression coefficients, and the error term. In linear regression, the aim is to determine the values of the regression coefficients that minimize the sum of the squared differences between the predicted and actual values of y. RSM has been used to predict experimental data by fitting a mathematical model to the observed data, which can then be used to make predictions for new combinations of input variables. By employing this approach, we can ascertain the optimal conditions necessary to achieve a desired response and assess the responsiveness of the response to variations in the input variables.

Results and discussion

Characterization results

N2 adsorption–desorption results of 4A-zeolite and modified zeolites indicate in Table 5 and Fig. 4b. The pore size distribution of zeolite 4A, 4A-TEPA, and 4A-DEA was analyzed using the Barrett–Joyner–Halenda equation (BJH)63, and the results are present in Fig. 4a. The BJH pore diameters for 4A, 4A-TEPA, and 4A-DEA were determined to be 11.1 nm, 10.1 nm, and 6 nm, respectively. Table 5 provides information on the micropore volume, BET surface area, and micropore surface area of the zeolite 4A and modified zeolites.

Table 5 BET Characterization of zeolite 4A and its modified structures.
Figure 4
figure 4

(a) The pore-size distribution by BJH method for adsorbent samples, (b) N2 adsorption/desorption at 77K.

The FT-IR spectrum of Fig. 5 shows the comparison of three samples: 4A-zeolite, 4A-15%TEPA, and 4A-10%DEA. In the spectrum of the commercial 4A-zeolite there are several prominent peaks and troughs that can be identified and analyzed. The peak at around 3429 cm−1 corresponds to the O–H stretching vibration of adsorbed water molecules, indicating the presence of water in the zeolite. In the spectrum of the modified zeolite with DEA and TEPA, some several additional peaks and troughs can be identified and analyzed. The peak at around 3417 cm−1 corresponds to the O–H stretching vibration of adsorbed water molecules, similar to the commercial zeolite 4A. The peak at around 1544 cm−1 corresponds to the N–H bending vibration of the amine groups, further confirming the presence of the amines64. Comparing the spectra of the modified zeolites with the unmodified 4A-zeolite reveals changes in peak intensity and position. For the FT-IR structure of 4A-zeolite, the vibration bands at 1001 cm−1 and 570 cm−1 could be assigned to the stretching vibration of Si–O or Al–O units and the vibration of Si–O–Al units in the 4A-zeolite structure, respectively63.

Figure 5
figure 5

FT-IR of zeolite 4A, Zeolite 4A-TEPA, and Zeolite 4A-DEA.

Figure 6 shows the XRD pattern of samples. For zeolite 4A and amine-modified Zeolites, the positions of characteristic peaks are consistent, the same as based 4A-zeolite. The most intense peaks occur at 17.2°, 26.17°, 35.1°, 40.2°, and 60° 2θ, which can be indexed to an amorphous structure consistent with zeolite 4A because it was synthesized from clay materials65. All zeolites exhibit additional peaks when compared to the untreated 4A-zeolite. The samples used for 4A, 4A-TEPA, and 4A-DEA were free of impurities, as confirmed by the similarities observed in their XRD patterns.

Figure 6
figure 6

XRD profiles of Zeolite 4A, Zeolite 4A-TEPA, and Zeolite 4A-DEA.

In Fig. 7a, the bright and dark areas represent variations in the topography and composition of the zeolite 4A surface. The bright regions correspond to high points on the surface, while the dark areas correspond to low points or pores within the zeolite structure. In Fig. 7b and c, the presence of pores within the zeolite structure emerges as a critical characteristic of this material, as it offers sites for the adsorption of CO2. The effect of amine in 4A-zeolite determines in Table 5.

Figure 7
figure 7

SEM images of (a) Zeolite 4A, (b) Zeolite 4A-10%DEA, and (c) Zeolite 4A-15%TEPA.

RSM results

This research study utilized response surface methodology (RSM) based on central composite design (CCD) to explore and optimize CO2 adsorption by two modified zeolite 4A adsorbents with TEPA and DEA. The investigation employed a factorial design incorporating four factors temperature, pressure, the percentage of amine used for surface modification, and the type of amine, resulting in a total of 52 tests. The study findings include the values of the independent factors and the CO2 adsorption capacity. The amount of change shown by the response variables in the assimilation process indicates the magnitude and direction of their influence, as determined by the corresponding signs. The values about the TEPA and DEA factors were obtained using Eqs. (8, 9), respectively, providing a practical means to explain the resultant predictive relationship.

$${q}_{C{O}_{2}}(\mathrm{TEPA})=139.42197-5.89412\times T\text{+ 43.07921}\times P\text{+ 0.478457}\times C-\text{0.487504}\times T\times P -0.091647\times T\times C+0.824998\times P\times C+0.108529\times \left({T}^{2}\right)+1.05560\times \left({P}^{2}\right)+0.042884\times ({C}^{2})$$
$${q}_{C{O}_{2}}(\mathrm{DEA})=+274.43678-9.03449\times T\text{+ 48.61391}\times P-1.41403\times C-\text{0.487504}\times T\times P -0.091647\times T\times C+0.824998\times P\times C+0.108529\times \left({T}^{2}\right)+1.05560\times \left({P}^{2}\right)+0.042884\times ({C}^{2})$$

It is imperative to assess the significance of the model, its independent parameters, and any interactions and second-order terms that may exist within it for a desirable response.

The ANOVA analysis Table 6. ANOVA model of adsorption capacity Presents statistical data concerning the parameters of temperature, pressure, amine loading, CO2 adsorption capacity, and efficiency. Statistical analysis conducted on the model yielded a Model F-value of 34.19, which suggests that the model is significant. The likelihood of an F-value of this magnitude occurring due to random error is only 0.01%, indicating a high degree of confidence in the model's validity. In line with established conventions, model terms with p-values below 0.0500 consider statistically significant36, while those with p-values greater than 0.1000 deem insignificant. Accordingly, in the present analysis, the terms A, B, AD, and A2 consider significant contributors to the model.

Table 6 ANOVA model of adsorption capacity.

In contrast, those with p-values greater than 0.1000 are deemed to have no significant impact. The model terms of temperature and pressure have F-values of 10.46 and 403.20, respectively. In other words, the high F-values provide evidence that the model is meaningful, as the model terms have a substantial influence on the response variable. The correlation coefficient (0.9212) obtained for the CO2 uptake capacity indicates a satisfactory agreement between the correlation coefficients and experimental data. The difference between the predicted R2 of 0.7300 and the adjusted R2 of 0.8943 is within 0.2, indicating a good match. Additionally, Adeq Precision, which gauges the signal-to-noise ratio, should exceed 4, But in this case, it measures 28.981, indicating an adequate signal. Hence, this model can be applied to explore the design space. Ratio more significant than 4 is desirable; a ratio of 28.981 indicates an adequate signal. This model can be used to navigate the design space. The performance of a proposed model for CO2 adsorption evaluates using residual plots and a comparison of actual and predicted values. Figure 8a displays the residual field, which depicts the deviation between actual and predicted values of the model's responses. The degree of appropriateness and normality displayed by the distribution of data points surrounding the linear regression line in this plot indicates whether the errors distribute adequately. Figure 8b shows a story of the actual and predicted values, revealing a good agreement between the two. This result suggests that the proposed model can accurately predict the amount of CO2 adsorption under various operating conditions.

Figure 8
figure 8

The CCD forecasted the CO2 capture capacity estimated value in comparison to (a) the normal probability and (b) the actual capture capacity.

In order to further assess the reliability of the final model, a residual plot was generated by plotting the predicted response values against the residuals, as shown in Fig. 9a. The resulting plot displayed a scattered distribution of points along the x-axis, ranging from + 3.58751 to − 3.58751, with no discernible trends. This observation suggests that the models were adequate and reliable, with a consistent variance observed across the range of responses. Moreover, this approach served as an additional tool for evaluating the suitability of the final model. We utilized the method of tracing predicted responses toward the model residuals, as illustrated in Fig. 9a and b, to examine whether there were any notable levels of constant variance.

Figure 9
figure 9

The predicted CO2 uptake capacity in relation: (a) Externally studentized residual, and (b) Run number.

Researchers in the RSM often use three-dimensional response surfaces to study and determine the best conditions. These surfaces can analyze the connections between factor variables and responses. Researchers can examine the impact of variables on a system in great detail by focusing on the response functions of two parameters while keeping all other parameters constant. Figure 10 presents a three-dimensional diagram, obtained using response surface methodology, for two zeolite modifications with DEA (Fig. 10a) and TEPA (Fig. 10b), depicting the impact of the interaction between pressure and temperature variables on the CO2 adsorption capacity of modified-zeolites. We evaluated both zeolite modifications at different temperatures and pressures, ranging from 25 to 65 °C and 1 to 9 bar, respectively. Afterward, we analyzed the obtained data using Design Expert software. The analysis revealed that an increase in pressure caused increasing in CO2 adsorption for both modifications, while higher temperatures resulted in a reduction in adsorption capacity. We also examined an amine concentration of 15% for both adsorbents.

Figure 10
figure 10

Interactions between P*T and their impact on CO2 adsorption capacity (a) Zeolite 4A-DEA and (b) Zeolite 4A-TEPA.

Deviation plots

The deviation plot illustrates the comprehensive impact of all process parameters on the response function, with the central point (0) serving as the midpoint of the operating range. This outcome offers valuable insights into the overall behavior of the studied system. Figure 11 presents a perturbation plot highlighting the effect of all four operating parameters, namely temperature, pressure, and wt% amine, at the reference points. The results reveal that an increase in temperature (A) and wt% amine (C) leads to a decrease in the CO2 capacity of DEA, whereas an increase in pressure (B) enhances CO2 capture. Furthermore, Fig. 11a and b demonstrates that TEPA and DEA exhibit similar behavior under the experimental conditions tested.

Figure 11
figure 11

Deviation curves for responses of two kinds of modified zeolite with coded factors for (a) 4A-DEA, and (b) 4A-TEPA.

Optimum adsorption operating parameters

This study aimed to identify the optimal combination of independent variables, namely pressure, temperature, and wt% of amine, to achieve maximum adsorption performance. The Response Surface Methodology (RSM) optimization technique is proposed by conducting a series of tests. Input parameters were given ranged values to achieve the maximum response of CO2 adsorption capacity. Tables 7 and 8 present the limiting conditions for zeolites modified with DEA and TEPA amines, respectively.

Table 7 Optimization of condition for 4A-DEA.
Table 8 Optimization of condition for 4A-TEPA.

Optimization of condition for 4A-DEA and 4A-TEPA

Determining the optimal operating and structural conditions for modified zeolites to achieve the maximum CO2 adsorption capacity is the main objective of this study. Optimizing variable components is one of the methods to enhance CO2 adsorption efficiency in modified zeolites. Table 9 presents the optimal values and specified ranges for the CO2 adsorption process when TEPA and DEA utilize to modify 4A-zeolite.

Table 9 The optimal values of uptake optimization results by zeolite-modified adsorbents.

After obtaining the optimal conditions, we plan to perform isotherm, kinetic, and thermodynamic modeling on the modified zeolites.

Influence of amine loading quantity on CO2 adsorption capacity

Amines are commonly used as functional groups on adsorbent materials due to their ability to interact with CO2 molecules through chemisorption, resulting in enhanced CO2 capture performance. By incorporating amine functional groups, more sites for CO2 adsorption are introduced that improve the adsorption capacity through increased surface interactions. Amine loading plays a direct role in CO2 adsorption efficacy as amines are the primary active sites for CO2 adsorption in solid adsorbents that are amine-based and functionalized. Very high loading of amine functional groups can lead to steric hindrance, limiting access to the amine sites and reducing CO2 adsorption efficiency66. Therefore, careful optimization of TEPA or DEA loading is essential to balance the benefits of increased loading without negatively affecting adsorption performance. The CO2 adsorption capacity of the adsorbents at different DEA and TEPA loadings is shown in Fig. 12. According to Fig. 12, the ideal DEA and TEPA loading for 4A-zeolite were 10 and 15 wt%, respectively. Due to the low DEA load and the decreased tendency of pore-clogging in 4A-10%DEA, the adsorption capacity is the highest67. The broadest pore size distribution, which enhances TEPA dispersion and CO2 molecule diffusion, is 4A-15%TEPA (Fig. 4). The best CO2 adsorption capacity is found in 4A-10%DEA, which has more micropores (Table 5) hence lowers the resistance to mass transfer and boosts the capacity for CO2 adsorption68.

Figure 12
figure 12

CO2 adsorption capacity of the adsorbents at different DEA and TEPA loadings.

Isotherm modeling

Utilizing isotherm modeling is crucial in investigating CO2 adsorption in amine-modified zeolites, as it provides a quantitative description of the adsorption behaviours of CO2 on the modified zeolite surface. The Langmuir Eq. (10), Freundlich Eq. (11), and Dubinin-Radushkevich (D–R) Eq. (12) models were used in this study to describe the behaviour, as they are among the various isotherm models available.

$${q}_{e}={q}_{m}{e}^{-\lambda {\omega }^{2}}$$

The parameters qe and qm represent the equilibrium and maximum adsorption capacities of CO2, respectively, and measure in units of mg/g. The Langmuir isotherm model is characterized by the parameter KL, which represents the Langmuir equilibrium constant (bar−1). The Freundlich isotherm model is characterized by the parameters kF (mg g−1 bar−1/n), Pe (bar), and n (Freundlich isotherm constant). The D–R isotherm model characterizes by two parameters, namely the constant of the model (λ) in mol2/J2, and the Polanyi potential (ω) in KJ/mol units.

CO2 adsorption isotherms using these models plot at 298 K and pressures ranging from 1 to 9 bar, as depicted in Fig. 13a–c. The results indicated that an increase in uptake pressure led to a rise in the rate of CO2 adsorption. Table 10 presents the experimental findings and the R2 correlation coefficients for all coefficients of isotherm parameter models. Based on the nonlinear regression technique and the R2 values, the theoretical isotherms rank in order of effectiveness for explaining and predicting adsorption for the behavior of modified zeolite as Freundlich > Langmuir > D–R. The Freundlich isotherm model's ability to fit well with the adsorption data indicates that the modified zeolite surface is not uniform and has a wide range of adsorption energies. This behavior explains the heterogeneous surface with broad adsorption energy distribution through the Freundlich constant and exponent parameters. A high Freundlich constant shows that the modified zeolite has a high adsorption capacity, while a low exponent means a more linear adsorption isotherm. In conclusion, the Freundlich isotherm model provides valuable information about CO2 adsorption on amine-modified zeolites and can help optimize their design and performance for CO2 capture applications.

Figure 13
figure 13

Comparison of isothermal models and experimental values of CO2 adsorption at the temperature of 298 K and pressure of 5 bar by (a) 4A, (b) 4A-15%TEPA, and (c) 4A-10%DEA.

Table 10 Theoretical isotherm models of CO2 adsorption at 5 bar and 298 K.

Kinetics modeling

The analysis of the adsorption rate, through kinetics, is crucial in determining the required residence time for evaluating the adsorption reaction. In the study of adsorption data, two primary categories of mathematical models are commonly utilized: adsorption reaction models and adsorption diffusion models. Although both models describe the kinetic process of adsorption, they represent different aspects of the kinetic analysis69. In the case of reaction models, experimental data fit with differential equations such as pseudo-first order, pseudo-second order, etc. (Table 11), which help to determine the reaction order and rate constants46. In contrast, adsorption diffusion models base on three consecutive steps: external diffusion or film diffusion (i.e., diffusion across the gas film surrounding the adsorbent particle), internal diffusion or intraparticle diffusion (i.e., diffusion of gas in the pores and, or along the pore walls), and mass action (i.e., adsorption and desorption between gas molecules and active sites)69. The variables qt, k1, k2, and kA use represent the adsorption capacity and rate constants of the first-order, second-order, and fractional-order models. In addition to n, a, and b use employing mean the kinetic model parameters, that important in characterizing the adsorption process's kinetic behavior and understanding the underlying mechanisms.

Table 11 Fitting parameters of kinetic models.

The first-order model assumes that the rate of solute uptake changes proportionally to the difference in saturation concentration and the amount of solid uptake over time, indicating a physical adsorption process. If the R2 value of the latter model decreases, as shown in Table 11, it suggests that chemical adsorption plays an increasingly important role in the adsorption processes70. The Rate Controlling Model has been commonly used to analyze mass transfer mechanisms and has established intraparticle diffusion as the sole determining factor in regulating the process rate71. Based on the data presented in Table 11 and the correlation coefficient (R2) values of the kinetic models, it is evident that the fractional-order adsorption kinetic model is the best-suited approach for describing the CO2 adsorption capacity and reaction time. This model provides a more thorough and accurate description of adsorption phenomena that deviate from integer order kinetics (Fig. 14). It considers various factors, such as surface heterogeneity, multilayer adsorption, and the interactions between adsorbate molecules, which are all crucial in the intricate nature of the adsorption process72. Correlation coefficient values (R2) ranging from 0.97470 to 0.99337 at 5 bar (298, 308 K) suggest that the fractional-order adsorption kinetic model provides the best fit. Table 11 displays the corresponding kinetic parameters.

Figure 14
figure 14

Comparison of kinetic models and experimental values of CO2 adsorption at the temperature of 298 K and pressure of 5 bar by (a) 4A, (b) 4A-15%TEPA, and (c) 4A-10%DEA.

Thermodynamic modeling

The thermodynamic parameters, including Gibbs free energy (ΔG°), Enthalpy (ΔH°), and Entropy (ΔS°), are essential for understanding the adsorption process21. Equations (13) provide a way to determine the values of ΔH° and ΔS° by plotting ln (KL) against the inverse of temperature (1/T) and can be used to calculate ΔG°. The universal gas constant (R) and the absolute temperature (T) are represented by 8.314 J/mol K and K, respectively.

$$\Delta {G}^{0}=-RT ln{K}_{d}$$
$$\Delta {G}^{0}=\Delta {H}^{0}-T\Delta {S}^{0}$$
$$ln{K}_{d}=\frac{\Delta {S}^{0}}{R}-\frac{\Delta {H}^{0}}{RT}$$

Table 12 shows the thermodynamic parameters, with negative values of ΔG° that the spontaneity of the adsorption mechanism23. For 4A-zeolite, the ΔG° values shift inversely with temperature, indicating decreased adsorption feasibility at higher temperatures. The ΔG° values for 4A-zeolite are between − 9.219 and − 9.648 kJ/mol, suggesting physical adsorption. However, for DEA-4A and TEPA-4A, the ΔG° values increase with temperature, indicating both physical and chemical adsorption. By using (16, the values of ΔH° and ΔS° were obtained from the slope and intercept of ln (K) as a function of (1/T) (Fig. 15). ΔH° value for 4A-zeolite − 4.9, indicating an exothermic adsorption mechanism. Moreover, the ΔS° value suggests an associative mechanism for the adsorption process. Calculated thermodynamic data indicate the CO2 adsorption process on zeolite was physisorption, exothermic, and spontaneous51.

Table 12 Thermodynamic parameters for 4A-zeolite, 4A-15%TEPA and 4A-10%DEA.
Figure 15
figure 15

Arrhenius plot for the heat of adsorption.

Adsorption capacity comparison with recent adsorbents

This section involved a comparative analysis between the current study and other relevant investigations focused on CO2 adsorption, employing various recent adsorbents. The outcomes of various similar studies were consolidated and presented in Table 13. Both 4A-10%DEA and the 4A-15%TEPA samples demonstrated considerable adsorption capacities, measuring 413.69 mg/g and 404.89 mg/g, respectively. A comparison between this study and the other research endeavors revealed the superior performance and a notable adsorption capability of the resulting amine-functionalized samples for CO2 capture applications.

Table 13 Some studies on CO2 capture by most recent adsorbents.

Adsorption mechanism

The adsorption mechanism of CO2 on amine-modified 4A-zeolite involves the chemisorption of CO2 molecules onto the amine groups present on the surface of the zeolite. The TEPA molecule comprises both primary amine (R1NH2) and secondary amine (R1R2NH) functional groups, both of which can participate in the reaction with CO2 and yield a carbamate ion, as shown in Fig. 16. On the other hand, the DEA molecule contains the R1R2NH functional group, which is secondary amine and is responsible for the chemical reaction between the amines and CO2. The presence of amine groups enhances the adsorption capacity of the material for CO2, while the increased hydrophilicity of the modified zeolite surface promotes the physisorption of CO2 molecules. The van der Waals forces between the zeolite and CO2 molecules, influenced by the Si and Al atoms in the zeolite structure, also play a role in the adsorption mechanism by attracting and holding CO2 molecules on the surface49. The strength of the interaction between CO2 molecules and the amine groups on the zeolite surface influenced factors such as the type of amine used for modification, the amine loading, and the pore size of the zeolite. Efficient capture of CO2 by amine-modified zeolites attributed to a combination of chemisorption and physisorption mechanisms.

Figure 16
figure 16

Schematic of CO2 adsorption on modified zeolites.

Cyclic performance

The ability to reuse the adsorbent is essential for industrial applications. Amine-based adsorbents may degrade or leach amines under harsh conditions, potentially affecting their long-term performance and reusability. In a series of ten adsorption cycles at 298 K and 5 bar, both types showed a slight decrease in adsorption potential after recycling at 410 K for 8 h. The 4A-15%TEPA adsorption potential was reduced by 3%, and the 4A-10%DEA adsorption potential decreased by approximately 2% (Fig. 17). These results suggest potential effectiveness in high-value industrial applications.

Figure 17
figure 17

Adsorption–desorption of CO2 and reusability of amine-modified zeolites.


We successfully synthesized and modified 4A-zeolite from kaolin to improve its performance in capturing CO2 in this study. To characterize the synthesized samples, we utilized several analytical techniques, including scanning electron microscopy (SEM), X-ray diffraction (XRD), Fourier-transform infrared spectroscopy (FT-IR), and Brunauer–Emmett–Teller (BET) testing. Our study assessed the effectiveness of modifications in enhancing the CO2 adsorption capacity of 4A-zeolite and tested the adsorption capacity of the modified zeolites at different temperatures and pressures. Utilizing Response Surface Methodology (RSM), we evaluated the CO2 adsorption performance of the modified zeolites by optimizing the operating conditions. CO2 adsorption experiments were performed at varying temperatures, pressures, and amine concentrations. The optimal adsorption capacity of the 4A-TEPA adsorbent is 477.342 mg/g, achieved at a temperature of 25.05 °C, pressure of 8.991 bar, and amine concentration of 15.275 wt%. Similarly, the 4A-DEA adsorbent exhibits an optimal adsorption capacity of 579.468 mg/g, with optimal operational variables of 25.270 °C, 8.870 bar, and 11.112 wt% amine concentration. After subjecting the adsorbents to recycling in an oven at 410 K for 8 h, the 4A-15%TEPA adsorption potential experienced a reduction of 3%, while the 4A-10%DEA adsorption potential showed a decrease of approximately 2%. The high R2 value of 0.9212 confirmed the excellent agreement between the experimental data and the model employed in this study. Furthermore, the kinetic and thermodynamic analyses have shown that the adsorption process of the modified zeolites is affected by both physisorption and chemisorption mechanisms. After analyzing various kinetic models, it was determined that the fractional-order adsorption model was the most appropriate. Overall, the results of this research highlight the promising potential of amine-functionalized 4A-zeolite as an effective adsorbent for CO2 capture. The material demonstrates notable advantages such as cost-effectiveness, high CO2 adsorption capacity, and a lack of reagent requirements. The method proposed in this study has the potential to facilitate the production of high-performance zeolites for various industrial applications.