The Diffusion Behavior of CO₂ Adsorption from a CO₂/N₂ Gas Mixture on Zeolite 5A in a Fixed-Bed Column

Abstract

The objective of this research was to investigate the behavior and conditions for CO₂ adsorption using a mixture of CO₂/N₂ over a fixed-bed column of zeolite 5A. The study was performed with a variation in gas composition of CO₂/N₂ as a 20/80, 50/50, and 80/20 volume %, the adsorption temperatures as 298, 333, and 373 K and the total feed flow rates as 1, 2, and 4 L/h under 100 kPa pressure. The Bohart–Adams, Yoon–Nelson, and Thomas models were used to predict the breakthrough behavior of CO₂ adsorption in a fixed column. Furthermore, the adsorption mechanism has been investigated using the kinetics adsorption of pseudo-first-order, pseudo-second-order, Boyd model, and intraparticle model. Increasing the CO₂ composition of a gas mixture resulted in a high CO₂ adsorption capacity because of the high partial pressure of CO₂. The capacity of CO₂ adsorption was decreased with increasing temperature because of physical adsorption with an exothermic reaction. The CO₂ adsorption capacity was also decreased with increasing feed flow rates with inadequate time for CO₂ adsorbates diffusion into the pores of the adsorbent before exiting the packed bed. The CO₂ adsorption by zeolite 5A confirmed that the physical adsorption with intraparticle diffusion was the rate-controlling step of the whole process.

1. Introduction

Environmental pollution is the addition of substances, such as solids, liquids, and gases, to the environment in a hazardous form. The three major types of pollution classified by the environment are air pollution, water pollution, and land pollution. Currently, disposal and air pollution are significant issues in all areas of the world because of the effect of rapid technological change, as well as emissions from industry and power plants that burn fossil fuels [1,2]. The emitted air pollution is known as greenhouse gases including carbon dioxide (CO₂), methane (CH₄), nitrous oxide (N₂O), chlorofluorocarbon (CFCs), hydrofluorocarbons (HFCs), perfluorocarbons (PFCs) and sulfur hexafluoride (SF₆). Among all of them, CO₂ is regarded as the major greenhouse gas, contributing more than 87 percent of greenhouse gas emissions, and rapidly increasing global warming and climate change [3,4,5]. There are many significant CO₂ sources with high CO₂ contributions that need to reduce CO₂ from their processes, such as landfill gas (LFG), biogas, natural gas, syngas as pre-combustion, and flue gas [6,7].

Flue gas is the gas released from a combustion plant, which usually has a composition consisting mostly of N₂, CO, and CO₂ [8,9]. To capture the CO₂ from flue gas, the composition of gases in flue gas and the characteristic for each gas in a gaseous mixture was identified to find the method for separation. In 2020, CO₂ was likely to decrease since most people have lived through lockdowns, whereby many made the abrupt shift to working from home and barely used transportation, and the production capacity was disrupted due to the COVID-19 pandemic. Nevertheless, when the situation is over the economy would be expected to be completely recovered. Despite CO₂ decreasing due to COVID-19 pandemic, the global surface temperature and climate change are likely to rise steadily. Nowadays, CO₂ removal could be achieved by applying a variety of technologies of physical and chemical processes, for instance, cryogenic distillation and membrane separation, to reduce CO₂ emissions [10,11,12]. However, the cryogenic membranes and solvent adsorption still have disadvantages such as corrosion, high costs, and high energy consumption [5]. Therefore, adsorption using solids is regarded as one of the most promising processes for CO₂ reduction. It has been widely used in CO₂ reduction technologies due to low-cost processes, low energy consumption, and being able to regenerate [13].

Porous adsorbents are the group of materials that support solid adsorption processes because of the small porosity within the structure, high surface area, and high pore volume [14]. There are many types of solid adsorbents used in the adsorption process, depending on the gas type that needs to be adsorbed or separated. Zeolite 5A is one of the solid adsorbents with a cubic lattice of sodalite form which has been used in various adsorption and separation processes. The 0.42 nm free aperture of the pores allows passage of molecules of gas adsorbates, which have a kinetic diameter less than 0.49 nm [15,16]. Zeolite 5A has a high potential to adsorb CO₂ molecular gas due to the interaction between cations positioned in zeolite 5A and the quadrupole moment of CO₂. In previous research, many adsorption and separation processes of pure and binary gas adsorption were studied on zeolite 5A. Mofarahi and Gholipour [13] used zeolite 5A to increase the efficiency of CO₂/CH₄ separation using a volumetric method with the conditions of 273–343 K and 1000 kPa. Zeolite 5A was successfully applied in CO₂/CH₄ separation. Nam et al. [17] studied the equilibrium isotherm of CH₄, C₂H₆, C₂H₄, N₂, and H₂ on zeolite 5A at 293–313 K with the pressure ranging from 0 to 2000 kPa using Pressure Swing Adsorption (PSA). Zeolite 5A could adsorb the maximum CO₂ at 293 K with a low isosteric heat of adsorption.

Furthermore, several previous studies have reported the use of zeolite 5A to investigate the conditions for CO₂/N₂ adsorption. Mendes and his co-researchers used pressure swing adsorption (PSA) to separate CO₂/N₂ on binderless zeolite 5A at temperatures ranging from 305 to 368 K and pressures up to 50 kPa at a pilot scale [18]. The binderless zeolite 5A could separate CO₂ from N₂ using the PSA process with good purity, recovery, and productivity parameters. Lui and his co-researchers investigated the conditions for CO₂/N₂ adsorption and desorption via the PSA process using zeolite 5A as an adsorbent [19]. Zeolite 5A was significantly more selective for CO₂ loading than N₂ at 303 K, with a CO₂ recovery of 91.0 percent, and a purity of 53.9 percent. Although many studies have investigated the use of zeolite 5A for CO₂/N₂ adsorption, they only evaluated the pressure and temperature factors to enhance CO₂ adsorption capacities. The effect of other factors on CO₂/N₂ adsorption, such as the effect of CO₂/N₂ composition and feed flow rate have not previously been studied. Moreover, there are still not enough reports on CO₂ adsorption behavior with a solid adsorbent to provide the efficiency of adsorbent and adsorption column. The design of adsorption is useful for high efficiency of the adsorption process in large-scale processes.

Therefore, this research aimed to study the factors that affect CO₂ adsorption from a CO₂/N₂ gas mixture including the CO₂/N₂ composition, temperature, and total feed flow rate. The breakthrough curve and kinetic models were also carried out to investigate the CO₂ adsorption behavior in a fixed-bed column.

2. Materials and Methods

2.1. Materials

In this study, a CO₂/N₂ gas mixture was obtained by combining CO₂ (99.99%) and N₂ (99.95%) in three different ratios, as 20/80, 50/50, and 80/20 %vol of a CO₂/N₂ gas mixture, as shown in Figure 1. The commercial molecular sieve zeolite 5A (CaA) powder was used in this work to investigate the efficiency of gas adsorption. The zeolite 5A was heated at 378 K for an hour to remove moisture.

2.2. Experimental Methods

The CO₂/N₂ Gas Mixture Adsorption Measurements

The amount of CO₂ inlet was measured by feeding a mixture of CO₂ and N₂ gas into the reactor at various ratios of 20/80, 50/50, and 80/20 %vol of a CO₂/N₂ gas mixture. Without adsorbent, the total flow rate was adjusted to 2 L/h at 298 K under 100 kPa of pressure, as shown in Figure 1. The renewable zeolite 5A was packed into the reactor tube at the same weight among all adsorption conditions. Following that, a CO₂/N₂ gas mixture in various ratios was fed into the reactor. The amount of CO₂ concentration was measured every 15 s for 1 h, and the amount of CO₂ adsorption was calculated using Equation (1) [20].

$$q_t=\frac{\left(C_0-C\right)}{W}\times \frac{1}{M_w}$$

(1)

where C₀ is initial concentration of CO₂ (mg/L), C is outlet concentration of CO₂ (mg/L), q_t is CO₂ adsorption capacity (mmol/g), w is amount of adsorbent (g) and M_w is molar mass of CO₂.

The ratio of CO₂/N₂ at 20/80 %vol was firstly carried out to investigate the effects of temperature and flow rate on CO₂ adsorption. The temperatures used were 298, 333, and 373 K, and the total feed flow rates were 1, 2, and 4 L/h. Each experiment condition was labeled with a symbol of CO₂ %vol as C, N₂ %vol as N, temperature as T, and total feed flow rate as F, as shown in

Table 1. Information data for CO₂ adsorption from CO₂/N₂ gas mixture on zeolite 5A.

Parameter	Value
Adsorbent	Zeolite 5A
Reactor size	0.07752 cm.id
	stainless steel tube
Weight of adsorbent	1 g
Length of bed	5 cm
Pressure	100 kPa
C20N80T298F2	CO2 20 %vol, N2 80 %vol, 298 K and 2 L/h
C50N50T298F2	CO2 50 %vol, N2 50 %vol, 298 K and 2 L/h
C80N20T298F2	CO2 80 %vol, N2 20 %vol, 298 K and 2 L/h
C20N80T333F2	CO2 20 %vol, N2 80 %vol, 333 K and 2 L/h
C20N80T373F2	CO2 20 %vol, N2 80 %vol, 373 K and 2 L/h
C20N80T298F1	CO2 20 %vol, N2 80 %vol, 298 K and 1 L/h
C20N80T298F4	CO2 20 %vol, N2 80 %vol, 298 K and 4 L/h

.

2.3. Characterization

The surface area, pore volume, pore size, and N₂ adsorption–desorption isotherms of zeolite 5A were measured before and after adsorption using the Brunauer, Emmett, and Teller (BET) method at 77 K. (BET, Micromeritics, 3Flex Surface characterization, Norcross, GA, USA). The micropore volume and external surface area were calculated using the t-plot method. The external surface area was evaluated using the BET surface area method. In addition, the micropore surface area was calculated by subtracting the external surface area from the BET surface area.

2.4. Theoretical Model

2.4.1. Fixed-Bed Experiment and Mathematical Models

Several gas flow phenomena appear in the fixed-bed adsorption column, which are based on the assumptions of axial dispersion, external resistance of film, intraparticle diffusion, and nonlinear isotherm [21]. To evaluate the performance of a fixed-bed column and the efficiency of adsorbent, it is necessary to investigate the optimization of mathematical models for designing the CO₂ adsorption process in a large-scale column. In this study, the model of Bohart–Adams, Yoon–Nelson, and Thomas were used to investigate the CO₂ breakthrough curve under the assumptions of (i) ideal gas, (ii) isothermal condition across the bed, and (iii) negligible N₂ adsorption compared to CO₂.

The Bohart–Adams model [22] is used to predict the parameters of the fixed-bed column. This model is commonly used to describe the first part of the breakthrough curve, which is based on the adsorption and depends upon the concentration of adsorbate species and the residual capacity of adsorption. The linear equation of Bohart–Adams is expressed in Equation (2).

$$ln\left(\frac{C_t}{C_0}\right)=k_{BA}{C}_{0}t-k_{BA}{N}_0\frac{Z}{F}$$

(2)

where C₀ is the influent adsorbate concentration (mg/L), C_t is the adsorbate concentration at time (mg/L), k_BA is the kinetic rate constant of the Bohart–Adams model (L/mg. min), t is time (min), N₀ is the saturate concentration (mg/L), Z is the bed height of column (cm), and F is the linear velocity calculated by dividing flow rate by the column section area (cm²/min).

The Yoon–Nelson model [23] is generally used to describe the breakthrough curve, which can predict the entire adsorption process in a fixed- bed column. This model assumes that a decreasing adsorption rate is proportional to adsorbate adsorption and breakthrough of the adsorbent. The linear equation of the Yoon–Nelson model is expressed in Equation (3).

$$ln\left(\frac{C_t}{C_0-C_t}\right)=k_{YN}t-\tau k_{YN}$$

(3)

where C₀ is the influent adsorbate concentration (mg/L), C_t is the adsorbate concentration at time (mg/L), k_YN is the kinetic rate constant of the Yoon–Nelson model (min⁻¹), t is time (min), $\tau$ is the time required for 50% adsorbate breakthrough (min).

The Thomas model [24] is the most widely used to describe the fixed-bed adsorption column. This model is based on Langmuir kinetics assumptions, which are described by a pseudo-second-order reaction, and the external resistance during the mass transfer process is negligible. The linear equation for the Thomas model is expressed in Equation (4).

$$ln\left(\frac{C_0}{C_t}-1\right)=\frac{k_{TH}{q}_{TH}m}{Q}-k_{TH}{C}_{0}t$$

(4)

where C₀ is the influent adsorbate concentration (mg/L), C_t is the adsorbate concentration at time (mg/L), k_TH is the Thomas rate constant (L/min.mg), q_TH is the equilibrium adsorbate uptake in the adsorbent (mg/g), Q is the flow rate (mL/min), t is time (min), and m is the mass of adsorbent (g).

2.4.2. Adsorption Kinetics

The adsorption kinetics in this work were determined by pseudo-first-order (PFO) and pseudo-second-order (PSO) models. The PFO model was used to describe the adsorption kinetics of physical adsorption in which the adsorbate attaches to the surface of the adsorbent through diffusion. It refers to the adsorption rate as determined by diffusion. It is expressed in Equation (5) by the Lagergren [25].

$$ln \left(q_e-q_t\right)=ln q_e-k_{1}t$$

(5)

where q_e and q_t are the amount of CO₂ adsorbed (mmol/g) at equilibrium and at any time, respectively, and k₁ is the ratio constant of the PFO model.

The chemical adsorption between adsorbate and adsorbent was described using the PSO model. It could be expressed using the equation of Ho and Mackey [26], as shown in Equation (6).

$$\frac{t}{q_t}=\frac{1}{k_2{q}_e^2}+\frac{t}{q_e}$$

(6)

where k₂ is the ratio constant of the PSO model.

2.4.3. Diffusional Mass Transfer Models

CO₂ adsorption on a solid adsorbent is typically affected by external film diffusion, intraparticle diffusion, or both, which acts as the rate-limiting step throughout the entire CO₂ adsorption process [27,28]. In general, gas adsorption on solid adsorbents involves four major steps: bulk diffusion, external film diffusion, intraparticle diffusion, and surface adsorption [28,29]. The previous studies have only considered the external film diffusion and intraparticle diffusion. Because of the rapid adsorption kinetics of bulk diffusion and surface adsorption, steps are always negligible compared to both external film diffusion and intraparticle diffusion, which are slower [26,29,30,31]. The adsorption rate adsorbed on a solid adsorbent is mainly controlled by external film diffusion and intraparticle diffusion. These two processes are described by the intraparticle diffusion model and the Boyd model as shown in Equations (7) and (8), respectively.

$$q=k_i{t}^{0.5}$$

(7)

$$-ln\left(1-F\right)=k_{f}t$$

(8)

where F is the ratio of CO₂ adsorbed at equilibrium and at time ($\frac{q_t}{q_e}$). k_i and k_f are the kinetic constant of the intraparticle diffusion model and the Boyd model, respectively.

3. Results

3.1. Effect of CO₂ Composition in a CO₂/N₂ Gas Mixture for CO₂ Adsorption

The effect of CO₂ composition on the CO₂ adsorption breakthrough curve, which was studied by changing the N₂ composition in the feed flow rate based on a constant total feed flow rate of 2 L/h at 298 K under 100 kPa of pressure. By changing the N₂ flow rate, the CO₂ composition was increased from 20 %vol to 80 %vol as shown, with gas ratios of 20/80, 50/50, and 80/20 %vol of the CO₂/N₂ gas mixture. The CO₂ adsorption breakthrough curves were shown in Figure 2a; the breakthrough times decrease and reach saturation quickly when the CO₂ composition of gas mixture was increased from 20 to 80 %vol. In addition, Figure 2b indicated that the CO₂ adsorption capacity increases from 6.42 to 7.24 mmol/g when the compositions of CO₂ in the gas mixture increased from 20 to 80 %vol.

3.2. Effect of Temperature on CO₂ Adsorption

In the flue gas industry, the partial pressure of CO₂ is typically quite low, with a composition of approximately 20 %vol. Therefore, a CO₂/N₂ gas mixture composition with 20/80 %vol was used to further investigate the effect of temperature on CO₂ adsorption.

The effect of temperatures on the breakthrough curve of CO₂ adsorption were studied at a constant CO₂/N₂ composition of 20/80 %vol onto zeolite 5A with a feed flow rate of 2 L/h. Increasing the temperature of adsorption, the breakthrough time of CO₂ adsorption significantly decreased and became steeper, as shown in Figure 3a. These results exhibited the highest CO₂ adsorption capacity of 6.43 mmol/g at the lowest temperature of 298 K, as presented in Figure 3b.

3.3. Effect of Feed Flow Rate on CO₂ Adsorption

The constant composition of the CO₂/N₂ gas mixture at 20/80%vol and constant temperature at 298 K was used to study the effect of feed flow rates as 1, 2, and 4 L/h.

Figure 4a represents the effect of feed flow rates (1, 2, and 4 L/h) on the CO₂ adsorption breakthrough curve. As the feed flow rate increases, the breakthrough time and exhaust time becomes significantly shorter and the breakthrough curve profile steeper. The highest CO₂ adsorption capacity of 7.42 mmol/g was found at 1 L/h, which was the lowest feed flow rate, as shown in Figure 4b.

Furthermore, when the total feed flow rate was increased from 1 to 4 L/h, the BET surface area and micropore surface area of zeolite 5A decreased, as listed in

Table 2. The surface properties of pure zeolite 5A and after adsorption.

Sample	Surface Area (m2 g−1)			Pore Volume (cm3 g−1)
	BET Surface Area a	External Surface Area b	Micropore Surface Area b	Micropore Volume b	Total Pore Volume c
Pure zeolite 5A	517.44	18.21	499.23	0.260	0.277
C20N80T298F1	496.51	21.80	474.71	0.254	0.274
C20N80T298F2	498.15	20.30	477.85	0.241	0.260
C20N80T298F4	502.64	19.67	482.97	0.256	0.277

^a BET method ^b t-plot method ^c N₂ adsorption isotherm at P/P₀ ≈ 0.

. This is due to the fact that the performance of CO₂ diffusion into the micropore surface area of zeolite 5A is poor at a high total feed flow rate.

3.4. Adsorption Breakthrough Curve Models Study for a Fixed-Bed Column

The breakthrough curve models of Bohart–Adams, Yoon–Nelson, and Thomas were obtained to study the fixed-bed adsorption column and predict the dynamic nature of the column. Figure 5 indicates the experimental data models of CO₂ adsorption at 20/80 %vol of CO₂/N₂ gas mixture under the different operating conditions. The breakthrough profiles of the Yoon–Nelson model fits the experimental data better than the other models, with high correlation coefficients close to 1, as shown in

Table 3. CO₂ adsorption parameters in fixed-bed column.

Model	Conditions	Parameters
		kBA (L mg−1 min−1)	N0 (mg L−1)	R2
Bohart–Adams	C20N80T298F2	2.959 × 10−6	1.419 × 1010	0.936
	C20N80T373F2	6.024 × 10−6	0.669 × 1010	0.833
	C20N80T298F4	3.730 × 10−6	1.854 × 1010	0.890
Yoon–Nelson	Conditions	kYN (min−1)	$\tau$(min)	R2
	C20N80T298F2	0.169	19.569	0.990
	C20N80T373F2	0.429	7.164	0.990
	C20N80T298F4	0.286	8.605	0.992
Thomas	Conditions	kTH (L mg−1 min−1)	qTH (mg g−1)	R2
	C20N80T298F2	4.414 × 10−6	2.387 × 107	0.987
	C20N80T373F2	4.611 × 10−6	1.249 × 107	0.811
	C20N80T298F4	4.396 × 10−6	2.829 × 107	0.905

. The poor regression coefficients of the Bohart–Adams and Thomas models indicates that neither model is adequate for studying the CO₂ adsorption process in a fixed-bed column. According to the good experimental data, fitting with the Yoon–Nelson model, the value of the Yoon–Nelson kinetic rate constant increases with increasing temperature and feed flow rate, affected by decreasing the adsorption capacity.

3.5. Adsorption Kinetics Study

The adsorption kinetics study was analyzed using PFO and PSO models of Equations (5) and (6), respectively. The CO₂ adsorption experimental data with the PFO and PSO models, with a constant CO₂/N₂ composition at 20/80 %vol at different temperatures and feed flow rates, are shown in Figure 6. The PFO model perfectly fits the experimental data of CO₂ adsorption from a CO₂/N₂ gas mixture with zeolite 5A better than PSO model with high correlation coefficient (R² > 0.97), as listed in

Table 4. Kinetic parameters of CO₂ adsorption from 20/80 %vol of a CO₂/N₂ gas mixture using zeolite 5A at different conditions.

Conditions	Experiment	Pseudo-First-Order (PFO)			Pseudo-Second-Order (PSO)
	qexp (mmol g−1)	qcal (mmol g−1)	k1 (min−1)	R2	qcal (mmol g−1)	k2 (g mmol−1min−1)	R2
C20N80T298F2	6.44	7.69	0.1034	0.993	12.77	2.45 × 10−3	0.701
C20N80T333F2	3.42	4.09	0.0834	0.986	47.85	1.92 × 10−3	0.910
C20N80T373F2	2.56	2.89	0.0996	0.996	3.35	27.20 × 10−3	0.989
C20N80T298F1	7.42	9.02	0.0994	0.996	14.97	1.90 × 10−3	0.831
C20N80T298F4	3.83	4.77	0.0620	0.988	4.84	19.40 × 10−3	0.972

. The high correlation coefficient (R² > 0.97) of the PFO model indicates that it could describe the worthy reaction mechanism of CO₂ adsorption from a CO₂/N₂ gas mixture. In all cases of adsorption conditions (Table 4), the calculated CO₂ adsorption capacities decreased when temperature and feed flow rate were increased. This is typical of exothermic adsorption which happens to the physical adsorption process [13]. Corresponding to the hypothesis of the PFO model, physical adsorption is the process in which adsorbate attaches to the surface of an adsorbent and is controlled by diffusion [25].

3.6. Adsorption Diffusion Study

In general, the adsorption kinetic models can describe only the rate retention or release of solute from an adsorbates bulk-phase to the solid adsorbent interface. This means the PFO and PSO models could not provide the contribution of the rate-controlling steps associated with adsorption kinetics on the adsorbent. Therefore, it is necessary to investigate the diffusion steps of adsorption.

According to the diffusion-controlled physical adsorption, the intraparticle diffusion model from Equation (7) was used to investigate the diffusion steps of 20/80 %vol of a CO₂/N₂ gas mixture at different temperatures and different feed flow rates, as shown in Figure 7. It depicts that the CO₂ adsorption from a CO₂/N₂ gas mixture is multilinear, influenced by more than one diffusion processes, which is represented by three multilinear sections. The three multilinear sections indicate three main diffusion steps of external film diffusion, intraparticle diffusion, and the surface adsorption at equilibrium.

Normally, external film diffusion and intraparticle diffusion can occur concurrently in the initial step of adsorption. As a result, the initial step of adsorption was examined to determine the controlling step using the Boyd model from Equation (8), as shown in Figure 8. The Boyd plots indicates that the plots are not linear at different temperatures and feed flow rates.

According to the several adsorption steps observed in the whole process of CO₂ adsorption, it is necessary to investigate a diffusion-limiting step of the whole adsorption process. The kinetics constant of intraparticle diffusion (k_i) is obtained from the slope of the multilinear plots of intraparticle diffusion, as shown in

Table 5. The diffusion kinetics constant of CO₂ adsorption from CO₂/N₂ gas mixture on zeolite 5A.

Conditions	ki,1 (mmol g−1min0.5)	ki,2 (mmol g−1min0.5)	ki,3 (mmol g−1min0.5)
C20N80T298F2	0.5212	1.8460	0.0717
C20N80T333F2	0.5711	1.0573	0.0894
C20N80T373F2	0.5589	0.2299	0.0395
C20N80T298F1	0.7363	2.2231	0.0767
C20N80T298F4	0.5148	0.9935	0.2587

. The kinetics constant (k_i,2) during the second step of adsorption is the highest value when compared to the other steps of diffusion. Therefore, there was a longer contact time between the adsorbate and adsorbent for diffusion than at the other step. Furthermore, the ratios of time required for three linear sections are listed in

Table 6. The time taken for three stages of CO₂ adsorption from CO₂/N₂ gas mixture.

Adsorption Conditions	Time Taken for Film Diffusion (min)	Time Taken for Intraparticle Diffusion (min)	Time Taken for Adsorption Equilibrium (min)	Ratio of Time For Film Diffusion to Intraparticle Diffusion
C20N80T298F2	6	18	20	0.33
C20N80T333F2	11	12	20	0.92
C20N80T373F2	12	22	15	0.50
C20N80T298F1	6	17	25	0.35
C20N80T298F4	7	13	30	0.54

. They indicate that the ratio of time taken for film diffusion to intraparticle diffusion is less than 1 at all adsorption conditions.

4. Discussion

4.1. Effect of CO₂ Composition in CO₂/N₂ Gas Mixture for CO₂ Adsorption

The partial pressure of CO₂ increased as the CO₂ composition on the gas feed increased [32]. In fact, increasing the CO₂ composition in a gas mixture results in a higher concentration gradient between the bulk phase and solid phase which affects the high mass transfer zone [33]. As a result, CO₂ adsorption was improved and reached saturation quickly at high CO₂ composition with high adsorption capacity.

Even though the CO₂/N₂ gas mixture contains N₂ gas molecules which have the same adsorption properties as CO₂ on zeolite 5A, it has no effect on CO₂ adsorption. This correlates to CO₂ having a greater quadrupole moment (4.30 × 10²⁶ ESU/cm²) and polarizability, (26.5 × 10²⁵ cm³) than N₂ (1.52 × 10²⁶ ESU/cm² quadrupole moment and 17.6 × 10²⁵ cm³ polarizability) [34]. This means that CO₂ has a stronger interaction with the surface of zeolite 5A than N₂ [19,35]. Therefore, increasing the CO₂ composition in a gas mixture contributes to a stronger interaction between CO₂ and the surface of zeolite 5A that results in a high CO₂ adsorption capacity. Corresponding to the studies of Mofarahi et al. [13] and Mulgundmath et al. [36], CO₂ could be adsorbed onto zeolite better than N₂ and CH₄ due to its greater quadrupole moment and polarizability.

4.2. Effect of Temperature on CO₂ Adsorption

At a high adsorption temperature, the CO₂ adsorption capacity decreased because of the exothermic reaction of a physical adsorption process [37,38,39]. The physical adsorption is caused by the attachment of adsorbate to adsorbent surface walls by Van der Waals forces, which is a well-known exothermic reaction with a low heat of adsorption and no reaction between adsorbate and adsorbent [13,37,40,41].

Furthermore, the molecular motion kinetics of gas adsorbates have become more dynamic, resulting in faster diffusion through the adsorbent surface at high temperatures [42]. The faster adsorbate diffusion as temperature rises has resulted in a weaker interaction between adsorbate and adsorbent, resulting in a low adsorption capacity. Corresponding to the studies of Mendes et al. [18] they compared physical and chemical adsorption of solid adsorbents to separate CO₂ from a gas mixture using temperature swing adsorption. The low temperature had a significant effect on the high CO₂ adsorption capacity of zeolite 5A.

4.3. Effect of Feed Flow Rate on CO₂ Adsorption

A higher feed flow rate acts to increase the degree of gas turbulence flow, which contributes to the faster breakthrough time caused by the lower mass transfer zone in the fixed-bed column [41,43,44,45]. In contrast, lower feed flow rates result in a slower transport of CO₂ gas adsorbate, which increases the breakthrough time and requires more adsorbate to satisfy the high adsorption capacity [42,46]. Moreover, the results of decreasing CO₂ adsorption are due to the lack of diffusion time for adsorbates into the pores of adsorbent before leaving the packed bed. Corresponding to the studies of Tobarameekul and her colleagues [47] the diffusion time of CO₂ with a flow rate of 1 L/h was longer than at 5 L/h which led to an enhanced CO₂ adsorption capacity.

4.4. Adsorption Breakthrough Curve Models Study for Fixed-Bed Column

The studies of Bohart–Adams and Thomas models on CO₂ adsorption over fixed-bed columns indicated that neither model was adequate for studying the CO₂ adsorption process in a fixed-bed column. This is because the Bohart–Adams model can only represent the initial stage of adsorption and it is only applicable to irreversible adsorption processes. This reason corresponds to the assumption of the Thomas model, which is used to describe the chemical adsorption process, whereas CO₂ adsorption on zeolite 5A is the physical adsorption that can reverse the process [48].

The value of the Yoon–Nelson kinetic rate constant increased with increasing temperature and feed flow rate was affected by decreasing the adsorption capacity. The results are consistent with the assumption of the Yoon–Nelson model, that describes the assumption monolayer, as decreasing the adsorption rate is proportional to increasing adsorbate on the adsorbent surface area [23]. Consequently, the Yoon–Nelson model is the suitable model for describing the behavior of CO₂ adsorption in a fixed-bed column.

4.5. Adsorption Kinetics and Diffusion Models Study

The PFO model perfectly demonstrated that CO₂ adsorption from a CO₂/N₂ gas mixture onto zeolite 5A was a physical adsorption process, in which the adsorbate attached to the surface of adsorbent and was controlled by diffusion [25]. The physical adsorption process is literally an exothermic reaction which has a low adsorption capacity at high temperatures and a high feed flow rate because of its weak interaction between adsorbate and adsorbent [37,38,39]. Corresponding to the studies of Gabruś et al. [49], the CO₂ adsorption capacity, calculated using the PFO model, decreased when temperature increased.

According to the intraparticle diffusion model, it indicated that CO₂ adsorption from a CO₂/N₂ gas mixture has three main steps of diffusion including the film diffusion, intraparticle diffusion, and the surface adsorption at equilibrium [50].

In part of the rate-controlling step, the kinetics constant (k_i,2) during the second step of adsorption was the highest value and it represented that the second step of diffusion has significantly affected the high CO₂ adsorption capacity. Moreover, the ratios of time required for the three linear sections indicated that intraparticle diffusion was the rate-limiting step for the overall CO₂ adsorption process on zeolite 5A, since the ratio of time taken for film diffusion to intraparticle diffusion was less than 1 [51]. This is due to the large specific surface area of the adsorbent pores resulting in a longer contact time between adsorbate and adsorbent in the second step of diffusion, compared with the other steps [52,53,54]. Therefore, the intraparticle diffusion step is the adsorption limiting step of CO₂ adsorption from a CO₂/N₂ gas mixture on 5A zeolite.

In terms of the results, the mechanism of CO₂ adsorption for all processes from a CO₂/N₂ gas mixture onto zeolite 5A are illustrated by Figure 9. This indicates that the CO₂/N₂ gas mixture is fed into a packed bed reactor using zeolite 5A as an adsorbent. Then CO₂ and N₂ gases can diffuse through the zeolite 5A sorbent layer. The CO₂ molecules could be adsorbed more than the N₂ molecules due to the greater quadrupole moment of CO₂ to interact with the cations in zeolite 5A. CO₂ molecules can diffuse from the bulk phase to film diffusion and afterword go through intraparticle diffusion to the adsorbent pore before reaching the adsorption equilibrium.

5. Conclusions

Zeolite 5A has been used to study the conditions for CO₂ adsorption from a CO₂/N₂ gas mixture and investigate the behavior of CO₂ adsorption in a fixed-bed column. The following conclusions have been drawn:

(1)

At higher CO₂ compositions, a CO₂/N₂ gas mixture resulted in an increase in CO₂ adsorption capacities. Improving the partial pressure of CO₂ and mass transfer played a significant role in the rapid adsorption of CO₂ onto zeolite 5A.

(2)

The CO₂ adsorption process onto zeolite 5A is an exothermic reaction based on the physical adsorption influenced by Van der Waals forces, because the adsorption capacities are low at high temperature, which results in greater gas kinetics.

(3)

The CO₂ adsorption capacity decreased when the feed flow rates increased because the contact time between adsorbate and adsorbent in a fixed-bed column decreased, thereby improving mass transfer, which was greatly influenced by the shallower adsorption zone.

(4)

The Yoon–Nelson model is an excellent model to describe the behavior of CO₂ adsorption in a fixed-bed column. The high temperature and high feed flow rate increase the value of the Yoon–Nelson kinetic rate constant, which results in low adsorption capacities.

(5)

The CO₂ adsorption from a CO₂/N₂ gas mixture using zeolite 5A over a fixed-bed column can be described using a PFO model because it is a physical adsorption process controlled by diffusion. In addition, the CO₂ adsorption process in a fixed-bed column involves more than one diffusion step, including film diffusion, intraparticle diffusion and adsorption on adsorbent surface. The intraparticle diffusion was observed to be the rate-limiting step controlling the CO₂ adsorption process on zeolite 5A in a fixed-bed column.