Investigation of microwave-assisted regeneration of zeolite 13X for efficient direct air CO₂ capture: a comparison with conventional heating method

Abstract

This study investigates the regeneration of zeolite 13X for direct air CO₂ capture by comparing microwave-assisted and conventional heating methods in a fixed-bed reactor. Zeolite 13X, a high-surface-area solid adsorbent, was tested over three adsorption/desorption cycles under ambient conditions with approximately 400 ppm of CO₂. Microwave-assisted regeneration, optimized at 300 W for 10 min (350 °C), achieved a regeneration efficiency of 95.26%, with minimal loss in adsorption capacity (9%) across cycles. Conventional heating at 350 °C for 30 min achieved a comparable efficiency of 93.90% but required significantly more time and energy. The microwave technique operates via direct dielectric heating, selectively exciting polar species such as mobile Na⁺ ions within the zeolite framework. This localized and volumetric heating enhances CO₂ desorption without requiring reactor preheating or carrier gas flow, unlike conventional methods that rely on slower conduction and convection. As a result, microwave regeneration demonstrated a tenfold reduction in energy consumption (0.06 kWh vs. 0.62 kWh for conventional heating). Statistical analysis using ANOVA identified microwave power and regeneration time as key factors, with microwave power exerting the greatest influence. The study highlights the advantages of microwave-assisted regeneration, including reduced energy demand, shorter regeneration times, and enhanced scalability. These findings emphasize its potential as a transformative approach for advancing direct air capture technologies. Compared to conventional methods, microwave-assisted regeneration offers a more energy-efficient and practical solution for CO₂ removal from ambient air.

Introduction

Carbon dioxide (CO₂) is a prominent greenhouse gas that absorbs and re-emits outgoing longwave radiation, thereby contributing to the increase in Earth’s heat content¹. Since the onset of the Industrial Revolution, atmospheric CO₂ concentrations have increased by approximately 50%, with the highest recorded level reaching 412.5 parts per million (ppm) in 2020². The combustion of fossil fuels is widely recognized as the principal source of CO₂ emissions, playing a pivotal role in supporting industrial development as a primary energy source³. Extensive research has demonstrated that, in addition to emissions from industrial processes, human activities , including industrial production, transportation, and energy consumption, also significantly contribute to the ongoing rise in atmospheric CO₂ concentrations^2–5. In response to these trends, a variety of CO₂ mitigation strategies have been actively pursued, with the ultimate goal of achieving net-zero emissions by 2050. Among these strategies, technologies for direct air capture (DAC) of CO₂ at low concentrations have been developed to address the issue of excess CO₂ emissions, as capturing CO₂ solely from industrial sources is insufficient to eliminate the overall emissions⁶.

Direct Air Capture (DAC) is an emerging technology designed to remove CO₂ directly from ambient air. In recent years, DAC has garnered significant attention due to its versatility, as it can be deployed in a variety of locations and can capture CO₂ emissions regardless of their source. The DAC process primarily involves two key stages: adsorption/absorption and regeneration (or desorption). Conventional CO₂ capture methods, commonly referred to as absorption processes, typically involve the use of aqueous amine-based solvents, such as monoethanolamine⁷. This method is considered one of the most efficient techniques for CO₂ capture. However, it has several notable drawbacks, including high energy consumption, particularly due to the energy-intensive solvent regeneration process, corrosion issues, the need for large volumes of solvents to dilute the corrosive amines, amine degradation and losses, as well as elevated operational costs. Given these limitations, the use of solid adsorbents for direct air CO₂ capture presents a promising alternative. Solid adsorbents offer several advantages over liquid amine-based systems, including lower energy requirements for regeneration, relatively fast adsorption/desorption kinetics, reasonable stability over multiple cycles, and higher CO₂ equilibrium loading^8,9. These characteristics make solid adsorbents an attractive option for enhancing the efficiency and sustainability of CO₂ capture technologies.

During the CO₂ adsorption process from ambient air, several factors can influence its efficiency, presenting challenges to air capture. These factors include the need for operation under conditions of high humidity, ambient temperature, and pressure⁶. As such, the selection of an appropriate sorbent is crucial. Commonly used solid sorbents for CO₂ adsorption include silica, zeolites, metal–organic frameworks (MOFs), activated carbon, graphene, metal oxides, alumina, and organic polymers. Among these, zeolites are often preferred due to their high surface area, optimal pore size, and excellent thermal and chemical stability at low temperatures. Their crystalline structure, composed of TO₄ tetrahedral units (where T is silicon or aluminum), enhances cation exchange sites, which attract CO₂ molecules through electrostatic interactions. Zeolite 13X, with its FAU structure and 12-membered rings, offers superior CO₂ adsorption capacity, with pore sizes ranging from 0.33 to 0.60 nm for efficient CO₂ diffusion^10,11.

For the regeneration process, adsorbent regeneration can be achieved through various methods, with the most commonly used being conventional heating. In this method, heat is transferred from an external source to the adsorbent¹². However, the relatively low thermal conductivity of zeolite 13X adsorbents results in low energy efficiency during the conventional heating process, primarily due to the limited heat transfer from the adsorbent to the captured CO₂. While increasing the regeneration temperature could accelerate heat transfer, the stability of the adsorbent imposes a limitation on the regeneration temperature. Consequently, the conventional heating method leads to limited regeneration efficiency, extended regeneration time, and higher energy demand, resulting in increased operational and investment costs⁷. Additionally, conventional heating methods are susceptible to heat loss through conduction, convection, and radiation, occurring through the walls of the device during the heating process, further diminishing overall energy efficiency¹³.

A promising and novel alternative heating method to conventional heating is microwave heating. This technique directly interacts with the material by inducing molecular vibrations or ion migration, resulting in rapid and uniform heating throughout the entire volume. Unlike the conventional heating method, which relies on surface heat transfer through conduction, radiation, or convection¹⁴, microwave regeneration offers several advantages over conventional methods, including rapid start-up and stop, non-contact heating, reduced energy consumption, and shorter regeneration times^15,16.

Based on the principles outlined above, microwave heating offers a promising and efficient method for CO₂ release. The process involves the generation of “hot spots” on the catalyst surface due to the absorption of microwave energy by materials, which convert microwave energy into heat¹⁷. This mechanism plays a crucial role in facilitating the release of CO₂. Furthermore, microwave heating triggers both thermal and non-thermal effects, including dipole rotation and ion conduction. Dipole rotation changes the microwave field direction and the movement of dipole interference and obstacles on OH⁻ groups and water, while ion conduction refers to the movement of charged species (ions). Na⁺ is loosely held in an aluminosilicate framework within zeolite 13X due to an oscillating electric field^18,19. These effects contribute to the weakening of chemical bonds, thereby enhancing the decomposition of CO₂-adsorbing materials.

Currently, there is a growing body of research investigating the use of modified microwave technology for the regeneration of solid sorbents and amine aqueous solutions to release CO₂. These studies are commonly conducted in comparison with conventional heating methods to assess the relative effectiveness of each approach, typically focusing on energy consumption, regeneration time, and CO₂ recovery. Elison et al.¹³ aimed to reduce the energy consumption associated with conventional heating methods by studying a fixed bed of zeolite 13X, which was saturated with a 150 sccm flow of 15% CO₂ at room temperature. Their results demonstrated that microwave irradiation reduced CO₂ desorption times by 50% compared to conventional regeneration techniques. This reduction in desorption time was attributed to the rapid heating rates induced by microwave radiation. Furthermore, the energy required for microwave-assisted desorption was estimated to be approximately 16–18 kJ/mol, which is less than half of the energy required for conventional thermal desorption, approximately 41.5 kJ/mol. Li et al.²⁰ investigated the microwave regeneration of TETA organic solutions for ≥ 99.99 vol% CO₂ capture and compared it to conventional conductive heating. The study found that microwave heating outperformed conductive heating in terms of heating and desorption rates, CO₂ recovery, energy consumption, and regeneration temperature. Moreover, CO₂ recovery under microwave regeneration was 2.6 to 2.7 times higher. Similarly, McGurk et al.²¹ compared microwave swing regeneration with conventional heating for CO₂ recovery from post-combustion MEA solutions. Microwave irradiation released over double the amount of CO₂ at a faster rate and reduced the energy required for the CO₂ capture reaction. Multiple cycles showed consistent stability with a 50% CO₂ recovery at 90 °C. In contrast, conventional MEA scrubbing at higher temperatures (120–140 °C) leads to energy penalties and higher costs. Microwave regeneration offers a more efficient, faster, and cost-effective alternative for post-combustion CO₂ capture. As evidenced in the study, early research on microwave-assisted regeneration predominantly focused on CO₂ capture from sources with high concentrations, with several other studies also investigating similar approaches^22–24.

However, recent research has broadened to investigate direct air capture (DAC) from low-concentration CO₂ sources, which has become critically important in addressing current climate challenges. Jang et al.²⁵ investigated microwave-based Direct Air Capture (DAC) systems and directly measured the energy consumption required by both a microwave oven and a conventional oven to regenerate the same mass of sorbent. The study compared the total energy required per unit mass of regenerated material to assess the potential benefits of microwave regeneration. The results demonstrated that microwave heating effectively regenerated MGBIG carbonate, a process that may be facilitated by water molecules co-crystallized with carbonate in the guanidine crystals. Microwave heating at 2.54 GHz with 1250 W was found to be up to 17 times faster than conventional conductive heating at 160 °C, leading to a 40% reduction in electrical energy consumption. Similarly, Nokpho et al.²⁶ investigated the regeneration of gamma-alumina sorbent modified with potassium carbonate and monoethanolamine using microwave radiation. The findings indicated that the microwave regeneration technique not only enhanced the efficiency of the process but also improved the reusability of the sorbent. Furthermore, Boylu et al.²⁷ explored a microwave-assisted Direct Air Capture (DAC) system using Zeolite 13X to capture CO₂ from ambient air. A mono-mode microwave generator was used for CO₂ desorption, with temperature and power varying from 45 to 100 °C and 5 W to 60 W. Results showed that energy consumption for CO₂ desorption ranged from 60.37 MJ (100% regeneration) to 23.97 MJ (70% regeneration). More recently, the research by Marin et al.²⁸ demonstrated a microwave heat-driven reactive CO₂ capture (RCC) process using the SrO/SrCO₃ cycle and graphitic carbon to absorb ppm-level CO₂ from humidified room temperature air. The results showed that microwaves selectively heat the mixture of graphitic carbon and SrCO₃, producing CO₂ with 85 ± 3% selectivity and stable performance over ten cycles. Rapid heating effectively prevented performance loss due to particle sintering, which is commonly observed in conventional heating systems. All these studies on microwave-assisted Direct Air Capture (DAC) further underscore the growing interest and potential of microwave technology. It highlights its increasing application in regeneration processes, demonstrating significant improvements in energy efficiency and processing time, making it a promising and intriguing technology for the future.

The motivations outlined above form the basis of this study. Microwave-assisted regeneration has demonstrated both efficiency and practicality, offering notable advantages over conventional heating methods. This research focuses on the regeneration of zeolite 13X using both unmodified microwave and conventional heating techniques for direct air capture (DAC) of CO₂ in a fixed-bed column. In contrast to previous studies that have primarily investigated modified microwave systems, this study provides a comprehensive evaluation of unmodified microwave regeneration, examining the CO₂ adsorption stability and performance of zeolite 13X across a wide temperature range and multiple adsorption/desorption cycles. A key advantage of using unmodified microwave systems is their simplicity and scalability, eliminating the need for complex modifications while still achieving efficient and uniform heating. Furthermore, unlike modified systems, the microwave regeneration process in this study does not require nitrogen purging, which simplifies the operation and reduces both energy and resource consumption, thus enhancing its practicality for real-world DAC applications. The objectives are: (1) to investigate the effects of microwave power and regeneration time of the microwave method on the performance of the adsorbent, (2) to investigate the effects of temperature in the conventional heating method on the performance of the adsorbent, (3) to compare the two methods in terms of regeneration efficiency over multiple adsorption/desorption cycles, and (4) to evaluate the energy consumption during each regeneration method. In addition, analysis of variance (ANOVA) is employed to identify the optimal operating conditions. A detailed assessment of key performance indicators, namely CO₂ adsorption capacity, CO₂ breakthrough curve, and regeneration efficiency, is conducted to provide critical insights into the effectiveness and feasibility of each regeneration method for DAC applications.

Methodology

Adsorbent and gases

The CO₂ adsorbent used in this study was commercial zeolite 13X (Si/Al 2.6–3.0), supplied by Jiangxi OIM Chemical Co., Ltd., China, in spherical pellet form. The particle size ranged from 1.6 to 2.5 mm, with a bulk density of ≥ 0.64 g/mL. Nitrogen gas (N₂) with a purity of 99.99% was supplied by Thai-Japan Company, while CO₂ at approximately 400 ppm was captured directly from ambient air.

Sorbent characterization

The specific surface area and pore structure of the zeolite 13X were analyzed using N₂ adsorption–desorption isotherms measured with a Micromeritics TriStar II Plus instrument. The surface area was calculated using the BET model, while the pore size distribution was determined using the BJH method. The adsorbent was degassed at 250 °C under vacuum for at least 6 h before testing.

Experimental setup

A fixed-bed reactor made of borosilicate glass (0.02 m diameter, 0.80 m height) was used for all experiments. Prior to adsorption, 10 g of zeolite 13X was heated at 175 °C overnight in a muffle furnace to remove impurities and moisture²⁹.

Figure 1 illustrates a system designed for direct air CO₂ capture, comprising two sections: the adsorption section and the regeneration section, as described previously. In this study, the conditions for the adsorption section are kept constant, as detailed in Section “Adsorption process”. The regeneration conditions in the regeneration section are varied to evaluate and compare the regeneration efficiency between the conventional heating method and the microwave regeneration method, as detailed in Section “Regeneration process”.

Fig. 1.

Schematic diagram of the direct air CO₂ capture process in a fixed-bed reactor: (a) regeneration using the conventional heating method, (b) regeneration using the microwave regeneration method.

In the conventional heating method, a fixed-bed reactor was placed in the conventional heating system consisting of three primary sections: gas preparation, CO₂ adsorption, and CO₂ analysis. The reactor was connected to the inlet and outlet gas pipes. The feed gas flow rate was monitored using a mass flow controller, while the outlet CO₂ concentration was measured using a CO₂ sensor (with an error limit of ± 0.01%vol CO₂), both of which were controlled by a computer. Regeneration via the conventional heating method was carried out within the same system used for the adsorption process, with the addition of a heater, which was employed to provide the necessary heating for the regeneration process, as illustrated in Fig. 1a.

For the regeneration process with the microwave regeneration method, the adsorption process was identical to that of the conventional heating system. However, for regeneration, the zeolite 13X sorbent was carefully removed from the fixed-bed column and transferred to a sealed container. The particles were evenly spread in a thin layer to achieve uniform bed distribution, ensuring even heat exposure throughout the regeneration process. Regeneration was carried out using an unmodified microwave oven (Samsung ME711K, 2.45 GHz, 100–800 W), operated independently from the adsorption setup, as shown in Fig. 1b. The oven includes a rotating turntable to help ensure uniform heating. Microwave power and exposure time were systematically varied according to the experimental design. The temperature of the adsorbent during regeneration was monitored using a non-contact infrared thermometer to maintain consistent thermal conditions throughout all cycles. After regeneration, the adsorbent was promptly returned to the fixed-bed column under sealed conditions to minimize exposure to ambient air and preserve its adsorption performance for subsequent cycles.

The electrical energy consumed during the regeneration process for both microwave-assisted and conventional heating methods was measured using a plug-in digital watt meter rated for up to 4400 W, with a measurement accuracy of ± 1%. The device was connected between the power outlet and the regeneration equipment to monitor energy use in real-time. It continuously measured voltage and current to calculate instantaneous power in watts and integrated these values over time to determine total energy consumption in kilowatt-hours. In addition, the meter provided measurements of power factor and wire temperature using a built-in thermal sensor located at the power cord, allowing for the assessment of heat buildup independently of ambient conditions. This setup enabled an accurate and consistent comparison of energy consumption between the two regeneration methods under controlled experimental conditions.

Adsorption process

In the CO₂ adsorption section, the experiment was conducted at room temperature (~ 30 °C) and a pressure of 1 bar. Initially, N₂ gas from the gas tank was introduced to the fixed-bed reactor with a flow rate of 1 L/min. This gas was purged through the column for approximately 30 min, not only to ensure the removal of all volatiles from the adsorbent, reactor, and associated pipelines, but also to cool down the zeolite 13X sorbent, which had been previously heated in the muffle furnace, as described by Erguvan et al.⁸. Subsequently, CO₂ was pumped from the ambient air using a gas pump (Airhorse AHb-20A, China) and fed into the fixed-bed column at a flow rate of 5 L/min. Adsorption continued until equilibrium was reached, as indicated by a constant CO₂ concentration at the outlet. The amount of CO₂ adsorbed per unit mass of sorbent (q_eq) was calculated using Eq. (1)^26,30:

1

where q_eq is the amount of CO₂ adsorbed per unit mass of solid sorbent at equilibrium (mgCO₂/g sorbent), m is mass of solid sorbent (g), Q is mass flow rate of gas (mg/min), C_in is mass concentration of CO₂ in inlet gas (%CO₂), C_out is mass concentration of CO₂ in outlet gas (%CO₂) and t is adsorption time (min).

Regeneration process

To evaluate the effects of microwave power and microwave regeneration time in the microwave regeneration method, and to compare the regeneration efficiency of the conventional heating method and the microwave regeneration method in terms of regeneration temperature and regeneration time, it is crucial to assess the reusability of the adsorbent through three adsorption/desorption cycles in this experiment to determine its stability and efficiency in carbon dioxide adsorption. Since changes in regeneration efficiency can be observed after only three cycles, as noted in the research of Yassin et al.¹⁵, the regeneration process employed is described in the following sections.

Microwave-assisted regeneration method

Once the zeolite 13X is saturated with CO₂, the gas flow control system is turned off to stop the introduction of CO₂ into the reactor. The zeolite 13X sorbent is then removed from the reactor and placed into a microwave, as described in Section “Experimental setup”, under the specified conditions for CO₂ desorption. After desorption, the sorbent is returned to the reactor, and N₂ gas is introduced at a flow rate of 3 L/min to cool the sorbent before initiating another cycle.

Regeneration was carried out with two factors under investigation: microwave power and regeneration time. These factors were evaluated using a 2 k factorial design, where “2” represents the two levels (high and low) of each factor, and “k” denotes the number of influencing factors being tested^26,30. To systematically assess these factors, each parameter was tested at two levels, with one center point in two replications. Microwave power was tested at low and high values of 100 W and 300 W, respectively, while regeneration time was evaluated at low and high values of 5 min and 10 min, as detailed in Table 1. These variables were converted to temperature using an infrared thermometer. The corresponding temperature range is between 90 and 350 °C. The test range for the regeneration of the zeolite 13X adsorbent is based on previous research, which suggested a broad regeneration temperature range of 90–250 °C^8,31,32. To align with the 2 k factorial design method, this study extended the temperature range to evaluate the thermal stability of the zeolite 13X adsorbent at a maximum temperature of 350 °C, as outlined in the research of Fischer et al.³³. The temperature ranges are shown in Table 2.

Table 1.

Experiment range and level of regeneration operating parameters.

Operating parameter	Symbol	Unit	Level
			Lower	Middle	Higher
Microwave power	A	watt	100	200	300
Regeneration time	B	minute	5	7.5	10

Table 2.

Effect of microwave power and regeneration time on temperature measured by an infrared thermometer.

Time (min)	Power (W)
	100	200	300
5	90 °C		120 °C
7.5		150 °C
10	300 °C		350 °C

Conventional heating regeneration method

Similar to the microwave regeneration method, once the adsorbent is saturated with CO₂, the gas flow control system is turned off to prevent the continued feeding of CO₂ into the reactor. The key difference is that the adsorbent remains in the reactor. The heater is then activated to heat the adsorbent. Once the setpoint temperature is reached, the valve is switched, and nitrogen is introduced into the reactor at a flow rate of 3 L/min. A timer is then initiated to maintain heating for a duration of 30 min to facilitate the desorption of CO₂ from the sorbent. The continuous flow of nitrogen during regeneration is another key difference between conventional heating and microwave-assisted methods. However, this requirement poses a limitation when using an unmodified microwave system, and this aspect will be further discussed in Section “Comparison between microwave and conventional heating regeneration methods”. After the desorption process, the heater is turned off; however, the nitrogen flow is not ceased and is used to cool down the sorbent before initiating the next cycle. This process is conducted at various temperatures (90 °C, 120 °C, 150 °C, 300 °C, and 350 °C) to ensure consistency and comparability with the microwave regeneration method.

To evaluate the CO₂ capture capacity and regeneration efficiency of the two regeneration methods, the regeneration efficiency was calculated using Eq. (2). Both the adsorption and desorption processes were carried out over 3 cycles.

2

where REG is the regeneration efficiency of the sorbent (%), q_avg,reg is CO₂ adsorption capacity, averaged over 3 cycles (mgCO₂/g sorbent), q_f is CO₂ adsorption capacity of 1st cycle (mgCO₂/g sorbent).

Results and discussion

Textural properties (BET) of zeolite 13X

Figure 2a depicts N₂ adsorption–desorption isotherms, specifically the relative pressure vs. quantity adsorbed determined by measuring the amount of nitrogen gas that adsorbs onto the surface of zeolite 13X and then desorbs at a constant temperature. The sorbent presented type IV isotherm typically for mesoporous material³⁴. The surface area and total pore volume at p/p₀ = 0.995 measured using the BET analysis of zeolite 13X before CO₂ adsorption test was found to be 511 m²/g and 0.363 ml/g, respectively. This value is in close agreement with the findings of Majchrzak-Kucęba et al. for zeolite 13X pellets of similar sizes (1.6–2.5 mm)³⁵.

Fig. 2.

(a) N₂ adsorption–desorption isotherms of zeolite 13X. Red and black of the scattered plots refer to adsorption and desorption measurements, respectively and (b) BJH pore size distribution of zeolite 13X.

The pore size distribution of zeolite 13X, determined using the BJH method, is illustrated in Fig. 2b. The distribution of micropores and macropores exhibits narrow peaks centered at 4.2 nm and 116.9 nm, respectively, while the mesopore distribution displays a broad peak centered at 23 nm, these findings suggest that the mesoporous structure of zeolite 13X is well-suited to accommodate CO₂ molecules, potentially enhancing physical sorption within its support framework³⁶.

Microwave-assisted regeneration

Effect of microwave power and time on CO₂ adsorption capacity

Figure 3 illustrates the CO₂ adsorption capacity of zeolite 13X across three adsorption/desorption cycles, evaluated under varying regeneration conditions using microwave heating. Specifically, Cases 1, 2, 3, 4, and 5 correspond to microwave regeneration conditions with power and duration set as follows: 100 W for 5 min (90 °C), 100 W for 10 min (120 °C), 200 W for 7.5 min (150 °C), 300 W for 5 min (300 °C), and 300 W for 10 min (350 °C). The experimental results indicate that the average CO₂ adsorption capacity of the fresh sorbent is 5.70 mgCO₂/g sorbent, as derived from the CO₂ adsorption capacity measured during the first cycle across all experimental conditions.

Fig. 3.

CO₂ adsorption capacity of microwave regeneration method. C1, C2 and C3 stand for cycle 1, cycle 2 and cycle 3, respectively.

A closer examination of the individual conditions reveals that Case 1 and Case 5 exhibit the lowest and highest average CO₂ adsorption capacities, at 1.89 mgCO₂/g sorbent and 5.39 mgCO₂/g sorbent, respectively. Notably, Case 5 demonstrated enhanced stability in CO₂ adsorption, with the adsorption capacity in the third cycle showing only a 9% reduction compared to the first cycle. In contrast, under Cases 1, 2, 3, and 4, the CO₂ adsorption capacity in the third cycle was reduced by 74%, 65%, 66%, and 43%, respectively, compared to the first cycle.

These results suggest that an increase in microwave power and regeneration time enhances CO₂ desorption efficiency, thereby improving the working capacity of zeolite 13X across successive adsorption/desorption operations. This improvement is primarily attributed to more efficient internal heating induced by microwave radiation, which interacts directly with the adsorbent through dipole rotation and ionic conduction. These mechanisms generate molecular vibrations and ion movements within the material, resulting in rapid and uniform volumetric heating. The increased energy input leads to higher internal temperatures, which facilitate the removal of CO₂ and moisture^{15,17–19,37}. These effects contribute to the preservation of accessible adsorption sites and support consistent sorbent performance.

Accordingly, experimental conditions with higher microwave power and longer regeneration times (e.g., Case 5) demonstrated higher CO₂ adsorption capacities and a smaller reduction in performance by the third cycle compared to conditions with lower power and shorter durations. Moreover, the CO₂ adsorption capacities for Cases 1, 2, and 3 followed a similar trend, with a significant increase observed between Cases 4 and 5.

CO₂ breakthrough time

Figure 4 demonstrates the CO₂ breakthrough curves for the third adsorption cycle of each case, revealing that the time required to reach equilibrium varies across the different cases. The equilibrium point corresponds to the saturation rate of the zeolite 13X sorbent. The experimental data align with the previously discussed findings, showing that the equilibrium onset time for Cases 1, 2, and 3 is approximately 500 s. In contrast, Cases 4 and 5 exhibit noticeably longer onset times, with approximately 1200 s and 1500 s, respectively.

Fig. 4.

CO₂ breakthrough curves of microwave regeneration method for the 3^rd adsorption cycle of each case.

These results are consistent with the trends in CO₂ adsorption capacity observed earlier, confirming that increased microwave power and regeneration time enhance the sorbent’s performance. The mechanisms underlying these improvements were discussed in the previous section. Notably, Case 5 exhibited the longest breakthrough time, indicating its superior CO₂ adsorption capacity during the third cycle.

Regeneration efficiency

To further investigate the effects of microwave power and regeneration time on CO₂ adsorption, the regeneration efficiency was calculated using Eq. 2 in combination with the CO₂ adsorption capacities presented in Fig. 3. The calculated regeneration efficiencies for Cases 1 through 5 were 30.04%, 36.67%, 33.91%, 66.26%, and 95.26%, respectively. These values are clearly represented in the bar graph in Fig. 10. Notably, Case 5, which employed 300 W for 10 min, exhibited the highest regeneration efficiency among all tested conditions. This enhanced performance can be attributed to the increased microwave power input and extended regeneration time, which result in more effective internal heating of the adsorbent, as discussed earlier.

Fig. 10.

Comparison of regeneration efficiency between microwave and conventional heating regeneration methods under various regeneration conditions.

From a mechanistic perspective, zeolite 13X contains a high concentration of Na⁺ cations that serve as the primary active sites for CO₂ adsorption. Under microwave irradiation, these mobile Na⁺ ions oscillate in response to the alternating electric field and generate heat through collisions with surrounding framework atoms. The abundance of Na⁺ within the supercages of the faujasite structure facilitates rapid and localized dielectric heating. This selective heating at the adsorption sites enhances CO₂ desorption while minimizing unnecessary heating of the bulk material^38,39. Consequently, higher microwave power and longer regeneration times promote more efficient energy transfer to the active sites, improving regeneration efficiency.

In contrast, Cases 1, 2, and 3 did not show significant differences in either CO₂ adsorption capacity or regeneration efficiency. This outcome can be explained by the insufficient microwave power and regeneration time in these cases, which were inadequate to elevate the temperature of the adsorbent to a level sufficient for effective CO₂ desorption. Zeolite 13X, used in pellet form, has a large surface area and a complex internal pore structure that retains CO₂. Efficient CO₂ desorption requires elevated temperatures to overcome the retention forces within the porous structure. Based on these findings, it can be inferred that a regeneration temperature of 300 °C or higher is necessary for zeolite 13X in pellet form to achieve optimal CO₂ adsorption performance. Further research within the temperature range of 150 to 300 °C is recommended to explore the regeneration process in greater detail.

Since this study employed conventional microwave systems without modifications, real-time CO₂ desorption data could not be obtained. Consequently, the experimental findings emphasize the need for further investigations using statistical analysis techniques, such as ANOVA, to comprehensively assess the influence of microwave power and regeneration time on the regeneration efficiency of the adsorbents.

Statistical analysis of microwave regeneration method

The experimental results presented herein were derived from a 2^k factorial design experiment, which was employed to examine the simultaneous effects of various process variables. This approach is particularly effective for analyzing the interactions and individual influences of multiple factors on the outcome of interest. Statistical analysis was conducted to evaluate the impact of the key factors under investigation, including microwave power and regeneration time. The Analysis of Variance (ANOVA) approach was used to evaluate whether they differed significantly from one another⁴⁰.

Table 3 presents the detailed statistical findings of this investigation. The selection of significant variables and their interactions was based on rigorous statistical screening. The analysis yielded a Model F-value of 1010.67, indicating that the model is highly statistically significant (p < 0.0001). A closer examination of the individual F-values for the main variables revealed that microwave power (Variable 1) had the most pronounced influence on the response, with an F-value of 2467.36, followed by regeneration time (Variable 2) with 387.06, and the interaction between Variables 1 and 2 (Variable 3) with 177.58. Corresponding p-values for Variables 1 and 2 were both < 0.0001, and 0.0002 for Variable 3, confirming that all factors and their interaction exerted a statistically significant effect (p < 0.05) on regeneration efficiency. Furthermore, the pure error sum of squares (SS = 7.31) was notably low, indicating minimal variability across replicate runs. This reflects a high degree of experimental precision and repeatability and enhances the reliability of the statistical conclusions drawn. Overall, the robustness of the model, combined with low experimental error, underscores the validity of the observed trends in the regeneration process.

Table 3.

ANOVA analysis table.

Source	Sum of squares	df	Mean square	F-value	p value
Model	5544.59	3	1848.20	1010.67	< 0.0001
1. Microwave power	4512.03	1	4512.03	2467.36	< 0.0001
2. Regeneration time	707.82	1	707.82	387.06	< 0.0001
3. Interaction of Microwave power-Regeneration time	324.74	1	324.74	177.58	0.0002
Pure Error	7.31	4	1.83
Cor Total	5551.90	7

Figures 5 and 6 illustrate the interaction between the dependent and independent variables in both 2D and 3D formats. The results indicate that increasing microwave power and regeneration time generally enhances regeneration efficiency, with microwave power exerting the greatest influence, as evidenced by the steepness of the graph and the color contours in both figures. This observation aligns with the findings discussed in the section on the effects of microwave power and regeneration time on CO₂ adsorption capacity and regeneration efficiency. The optimal conditions for achieving the highest regeneration efficiency were determined to be a microwave power of 300 W and a regeneration time of 10 min. However, it is important to note that there remains a gap in the tested temperature range, particularly between 200 and 250 °C. This untested region may play a crucial role in influencing moisture desorption behavior or morphological changes in the sorbent material, both of which are key factors in the efficiency of the regeneration process.

Fig. 5.

2D surface interaction between study variables and regeneration efficiency.

Fig. 6.

3D surface interaction between study variables and regeneration efficiency.

To address these limitations, further research is required to investigate the influence of moisture content and zeolite pellet morphology on the CO₂ adsorption/desorption mechanism. Existing studies suggest that the regeneration temperature range may vary based on solid sorbent shape, such as cylindrical or spherical forms, which could influence the pore structure and ease of regeneration, thereby affecting the optimal temperature range for regeneration. Additionally, zeolite 13X exhibits a significant moisture adsorption capacity, which could affect CO₂ adsorption from moist air compared to CO₂ adsorption in a controlled environment. The presence of adsorbed moisture competes for adsorption sites on zeolite, necessitating higher regeneration temperatures to evaporate the moisture before reaching the optimal temperature for efficient CO₂ desorption^13,31,32. Consequently, optimizing microwave conditioning to achieve higher temperatures may significantly influence the number of regeneration cycles.

Conventional heating regeneration method

Effect of temperature on CO₂ adsorption capacity

The temperature range for the conventional heating method used in this evaluation was determined based on the temperatures achieved during the microwave regeneration method, as observed in the experiment. The CO₂ adsorption capacity and CO₂ breakthrough curve cycle under five different temperature conditions are presented in Figs. 7 and 8. Cases 1 through 5 correspond to regeneration temperatures of 90 °C, 120 °C, 150 °C, 300 °C, and 350 °C, respectively.

Fig. 7.

CO₂ adsorption capacity of conventional heating regeneration method. C1, C2 and C3 stand for cycle 1, cycle 2 and cycle 3 respectively.

Fig. 8.

CO₂ breakthrough curves of conventional heating regeneration method for the 3^rd adsorption cycle of each case.

The experimental results exhibited a similar trend to those observed in the microwave regeneration experiment. Specifically, Cases 1 and 5, which employed regeneration temperatures of 90 °C and 350 °C, respectively, exhibited the lowest and highest average CO₂ adsorption capacities, recorded at 2.14 mgCO₂/g sorbent and 7.31 mgCO₂/g sorbent. Additionally, it was observed that the difference in CO₂ capacity between the 3rd and 1st cycles diminished as the regeneration temperature increased. The most significant reduction in capacity was seen at regeneration temperatures of 150 °C, 300 °C, and 350 °C, with decreases of 63%, 42%, and 5%, respectively. In contrast, at temperatures of 90 °C and 120 °C, the reductions in CO₂ capacity were more pronounced and similar, at 77% and 79%, respectively.

The enhanced CO₂ adsorption performance at higher regeneration temperatures can be attributed to the same fundamental mechanism observed in microwave-assisted regeneration, which is the effective removal of CO₂ and moisture, helping to restore the active adsorption sites and recover the working capacity of zeolite 13X. However, the heating mechanisms differ considerably. Conventional thermal regeneration relies on conductive and convective heat transfer, which results in slower and often non-uniform heating, particularly in materials with low thermal conductivity. Moreover, some of the thermal energy is lost to heating the reactor walls and other non-adsorbing components before reaching the sorbent, thereby reducing overall energy efficiency^41,42.

CO₂ breakthrough time

Further analysis of Fig. 8, which presents the CO₂ breakthrough curves for the third adsorption cycle under the conventional heating regeneration method, indicates that the time required to reach equilibrium increases with higher regeneration temperatures, a trend consistent with the microwave regeneration results. Specifically, Cases 1 and 2, corresponding to regeneration temperatures of 90 °C and 120 °C, respectively, reached equilibrium in approximately 500 s. Case 3, at 150 °C, required about 700 s, while Cases 4 and 5, at 300 °C and 350 °C, respectively, reached equilibrium in approximately 1500 s. The longer time to reach the saturation point suggests a higher CO₂ adsorption capacity. This is attributed to the increase in the regeneration temperature, which enhances the kinetic energy of the CO₂ molecules. Consequently, the CO₂ molecules can more readily overcome the adsorption energy barrier, facilitating their release from the zeolite structure, as elaborated in section “Effect of microwave power and time on CO₂ adsorption capacity”.

Regeneration efficiency

In evaluating the effect of temperature on regeneration performance, both CO₂ desorption profiles presented in Fig. 9 and the regeneration efficiency was calculated by substituting the CO₂ adsorption capacity values from Fig. 7 into Eq. 2.

Fig. 9.

CO₂ desorption profiles of conventional heating regeneration.

From Fig. 9, it is evident that at all temperatures, there is an initial desorption peak observed within the first 200 s. A second, broader desorption peak, however, only appears at desorption temperatures of 300 °C and 350 °C, occurring between 500 and 1000 s. This observation aligns with the findings of Guo et al.⁴³, study on the desorption characteristics of zeolite 13X, which revealed that zeolite 13X, with its FAU structure, possesses three sites (SI, SII, and SIII) involved in cation exchange with Na+. In general, SII and SIII sites participate in CO₂ adsorption, whereas the SI site does not adsorb gases. Consequently, CO₂ desorption occurs in two stages. The 1st stage involves the desorption of CO₂ from the SII site, where linear CO₂ is physisorbed to Na+ through ion–dipole interactions. The 2nd stage involves the desorption of CO₂ from the SIII site, where a stronger bond exists due to bicoordination, with CO₂ adopting a bent configuration. In this orientation, CO₂ either forms a carboxylate species (Na+–O) by coordinating with the lattice of Al or Si atoms or a carbonate structure (Na+–C) by coordinating with adjacent lattice oxygen atoms. Physisorbed CO₂ at the SII site desorbs at lower temperatures (approximately 30 °C to 120 °C), while bicoordinated CO₂ at the SIII site desorbs at higher temperatures, typically above 120 °C. Therefore, the second peak is only observed when regeneration temperatures of 300 °C and 350 °C are used. Additionally, it is observed that at a regeneration temperature of 150 °C, the first peak is higher than those at 90 °C and 120 °C, indicating the involvement of CO₂ desorption from the SIII site. Furthermore, the regeneration efficiencies further support this explanation, with significant improvements in desorption efficiency of 59.75% and 93.90% at regeneration temperatures of 300 °C and 350 °C, respectively.

Comparison between microwave and conventional heating regeneration methods

When comparing the regeneration efficiency of the microwave regeneration method with that of the conventional heating method at the same temperatures, as shown in Fig. 10, it was observed that the microwave regeneration method outperformed the conventional heating method at 90 °C, 120 °C, and 300 °C. This can be attributed to the direct and selective nature of microwave heating, which delivers energy specifically to the active adsorption sites through the vibration of mobile Na⁺ ions^38,39. In contrast, conventional heating transfers heat more slowly and unevenly through conduction and convection, leading to energy losses in non-essential components. As a result, microwave regeneration provides more efficient CO₂ desorption^15,16.

At desorption temperatures of 150 °C and 350 °C, however, the regeneration efficiency of both methods was found to be very similar. This can be explained by the fact that at 150 °C, which marks the onset of the 2nd desorption stage (related to CO₂ desorption from the SIII site), as previously discussed. Therefore, the regeneration efficiency shows minimal variation between the two methods. Further investigation into CO₂ desorption in this temperature range is warranted. At 350 °C, where CO₂ desorption is nearly complete across both stages, the regeneration efficiency of both methods is nearly identical.

However, at all temperatures, microwave regeneration demonstrates significant advantages in terms of time efficiency. Microwave regeneration requires only 5 to 10 min to achieve regeneration efficiency comparable to or better than that of conventional heating regeneration, which typically takes up to 30 min, excluding preheating time. Therefore, microwave regeneration not only offers superior time efficiency but also contributes to a reduction in energy consumption. Specifically, when compared to the conventional heating method at 350 °C, microwave regeneration consumes 0.06 kWh, whereas conventional heating consumes 0.62 kWh. This represents a reduction in energy consumption by a factor of approximately 10. Similar trends were observed at other temperatures. At 90 °C, 120 °C, 150 °C, and 300 °C, microwave regeneration required 0.01, 0.02, 0.02, and 0.03 kWh, respectively, whereas conventional heating consumed 0.08, 0.15, 0.22, and 0.46 kWh. These findings further emphasize the energy-saving potential of microwave regeneration across a range of operating temperatures, underscoring its viability as a more energy-efficient alternative to conventional heating methods.

As previously discussed, although the microwave-assisted regeneration method demonstrates superior performance in terms of mechanism, regeneration time, and energy consumption, there are two important considerations to consider. First, as shown in Figs. 3 and 7, the CO₂ adsorption capacity in the first cycle under all conventional heating conditions was higher than that observed in the first cycle using microwave-assisted regeneration. Although both sets of samples were pre-heated at 175 °C overnight in a muffle furnace before testing, the conventional heating experiments were conducted first to facilitate comparison. These tests were carried out using conditions consistent with those achievable by the microwave method to allow for a valid two-factor analysis. However, because the zeolite 13X samples designated for microwave-assisted regeneration were stored for a longer period before testing, they were more susceptible to adsorbing ambient moisture and impurities, which may not have been fully removed during regeneration at the same temperatures. Given the hydrophilic nature of zeolite 13X, this could have led to a reduction in initial CO₂ adsorption capacity in the microwave-assisted regeneration method⁴⁴. To mitigate this, future experiments should consider conducting tests within a similar time frame to minimize differences in sample storage duration. Additionally, increasing the temperature during the pre-treatment step or employing nitrogen gas for purging before the first adsorption cycle could help ensure more consistent and comparable initial conditions between the two methods. Notably, increasing the pre-treatment temperature has shown clear benefits, as demonstrated in the study by Boylu et al.²⁷, which reported a maximum CO₂ adsorption capacity of approximately 15 mgCO₂/g sorbent, compared to about 7 mgCO₂/g sorbent in our study. This significant difference highlights the need to optimize the pre-treatment temperature in future work to improve performance and comparability. Nevertheless, despite these factors, the microwave-assisted technique consistently demonstrated higher overall regeneration efficiency, underscoring its advantages in energy consumption and sorbent recovery across multiple adsorption/desorption cycles.

The second point to consider is that, unlike the conventional heating regeneration method, the microwave-assisted regeneration method does not have an N₂ purge flow during regeneration. This is a limitation of the unmodified microwave system used in this study. While N₂ as a carrier gas can enhance mass transfer, it is important to note that despite the higher energy consumption associated with the conventional method and the use of carrier gas^45,46, its CO₂ desorption performance was not as efficient as that of the microwave-assisted regeneration method, which uses lower energy and does not require carrier gas. This suggests that other factors, such as the direct dielectric heating provided by microwaves, may have a more significant impact on desorption efficiency. Therefore, although the lack of nitrogen purging in the microwave method may be viewed as a difference, it does not appear to significantly affect the overall comparison between the two methods in terms of regeneration performance.

Conclusion

This study presents a comparative analysis of the CO₂ adsorption capacity and regeneration efficiency between a microwave regeneration method and a conventional heating method over 3 adsorption/desorption cycle, applied to zeolite 13X solid adsorbent for direct CO₂ capture from air in a fixed-bed reactor. To identify the optimal conditions for maximizing regeneration efficiency, analysis of variance (ANOVA) was employed during the microwave regeneration process. Furthermore, the specific surface area and pore structure of the zeolite 13X adsorbent were characterized using N₂ adsorption–desorption isotherms, which confirmed that zeolite 13X possesses favorable properties for use as an effective adsorbent for CO₂ capture.

For the microwave-assisted regeneration method, the microwave power and regeneration time were converted into temperature values using an infrared thermometer to assess their effects on CO₂ adsorption capacity and regeneration efficiency. The results indicated that increasing microwave power and regeneration time enhanced molecular vibrations within the adsorbent, causing it to remain in the microwave field for a longer duration. This, in turn, raised the temperature of the adsorbent, which increased the kinetic energy of the CO₂ molecules. As a result, the CO₂ molecules were more easily able to overcome the adsorption energy barrier, facilitating their release from the zeolite structure. Moreover, microwave irradiation offers a distinct advantage by selectively interacting with polar molecules and active sites within the zeolite structure, leading to localized heating at adsorption sites. The optimal conditions were found to be a microwave power of 300 W and a regeneration time of 10 min, corresponding to a temperature of 350 °C, which resulted in an average CO₂ adsorption capacity of 5.39 mgCO₂/g sorbent and a regeneration efficiency of 95.26%. Additionally, under these conditions, the CO₂ adsorption capacity in the final cycle decreased by only 9% compared to the first cycle. Furthermore, ANOVA analysis demonstrated that both the microwave power and regeneration time during the experimental period significantly influenced the regeneration efficiency of the adsorbent.

For the conventional heating regeneration method, the temperature used to evaluate the effect on CO₂ adsorption capacity and regeneration efficiency was based on the temperature achieved during the microwave regeneration method. The experimental results were consistent with those observed for microwave regeneration, with the optimal condition being 350 °C, which resulted in an average CO₂ adsorption capacity of 7.31 mgCO₂/g sorbent and a regeneration efficiency of 93.90%. Additionally, under these conditions, the CO₂ adsorption capacity in the final cycle decreased by only 5% compared to the first cycle. Moreover, the increase in desorption temperature also influenced the mechanism of CO₂ adsorption at the adsorption sites of the zeolite 13X sorbent. However, one of the inherent limitations of the conventional heating method is the unavoidable heat loss through conduction, convection, and radiation, which reduces the overall thermal efficiency.

Although the CO₂ adsorption capacity and the reduction in adsorption capacity in the final cycle, compared to the first cycle, under the optimal conditions for the conventional heating regeneration method were better than those observed for the microwave regeneration method, the regeneration efficiency of the microwave regeneration method was superior. This is because the adsorption capacity is based on a random sample from the experimental batch. However, the microwave regeneration method demonstrates superior efficiency due to its ability to directly heat the adsorbent without the need to heat the reactor walls, as required by conventional heating methods. Additionally, microwave regeneration involves shorter regeneration times, leading to energy consumption that is up to 10 times lower than that of the conventional heating method.

However, it is important to note that the unmodified microwave regeneration method used in this study presents certain limitations. For example, the absence of a nitrogen purge during the regeneration process, which is commonly employed in conventional systems, may affect CO₂ desorption efficiency. While unmodified microwave systems offer significant advantages in terms of energy efficiency, regeneration speed, and operational simplicity, further investigation is needed to examine how the absence of auxiliary mechanisms—such as N₂ purging or controlled humidity—affects the long-term stability, reusability of the sorbent, and overall performance under extended operating conditions.

Nonetheless, the results of this study reinforce the promising potential of unmodified microwave regeneration as an effective and energy-efficient technique for CO₂ desorption. Despite its current limitations, microwave heating provides a direct, selective, and rapid heating mechanism that enhances regeneration efficiency while significantly reducing energy consumption. With further optimization and integration of auxiliary controls, this method could serve as a viable alternative to conventional systems for sustainable and scalable direct air capture applications.

Author contributions

B.C., P.P., X.W.—Conceptualization B.C., P.P., X.W., P.N., P.A.—Methodology B.C., P.N., P.A.—Formal analysis B.C., P.N., P.A. -Investigation B.C., P.N., P.A.—Writing—Original Draft B.C., P.P., X.W., P.N., P.A.—Writing—Review & Editing P.N., P.A.—Visualization B.C., P.P., X.W.—Supervision B.C.—Funding acquisition.

Data availability

The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.