Comparative Analysis on Carbon Mitigation by High-Temperature Lithium Adsorption Systems

Abstract

High-temperature adsorption is a promising technology for carbon mitigation, and it can be applied in direct carbon capture and the integration with utilization. Lithium-based adsorbents, known for their high CO₂ uptake and rapid kinetics, have garnered significant interest. However, adsorption performance, cycling stability, and degradation behavior of this type of adsorbent are rarely reported and compared under comparable conditions. In this work, nine lithium-based adsorbents were synthesized and characterized for their physicochemical properties. Dynamic and isothermal thermogravimetric analysis were conducted to determine adsorption/desorption equilibrium temperatures, evaluate CO₂ adsorption characteristics under varying thermal conditions, and assess cycling stability over 20 adsorption–desorption cycles. The results reveal exceptional initial CO₂ capacities for α-Li₅AlO₄, Li₅GaO₄, Li₅FeO₄, and Li₆ZnO₄; however, these values decline to 30.2 wt.%, 24.3 wt.%, 41.6 wt.%, and 44.2 wt.% after cycling. In contrast, Li₂CuO₂ and Li₄SiO₄ exhibit lower initial capacities but possess superior cycling stability with final values of 21 wt.% and 21.6 wt.%. Phase composition and microstructural analysis identify lithium carbonate and metal oxides as primary products, and microstructural sintering was observed during cycling. This study could provide insights into the trade-offs between the initial capacity and cycling stability of lithium-based adsorbents, offering guidelines for adsorbent optimization through doping or pore engineering to advance high-temperature CO₂ capture technologies.

1. Introduction

The massive CO₂ emissions from fossil fuel utilization have been identified as the primary driver of global climate change. According to the Intergovernmental Panel on Climate Change (IPCC), global CO₂ emissions must decrease by approximately 45% from 2010 levels by 2030 to limit the rise of the global average temperature to within 1.5 °C above pre-industrial levels [1]. Carbon capture, utilization, and storage (CCUS) is globally recognized as one of the most effective strategies for CO₂ emissions mitigation, with the capture process constituting the most critical step of CCUS [2,3]. Consequently, the development of novel and efficient carbon capture technologies holds significant importance for achieving sustainable development across energy, environmental, and societal domains [4,5].

High-temperature carbon adsorption is recognized as one of the most promising technologies for efficient CO₂ capture due to notable advantages, such as high adsorption capacity and rapid reaction kinetics [6]. The cornerstone of CO₂ adsorption technology lies in developing adsorbents with high capacity and good stability. Common high-temperature CO₂ adsorbents include calcium oxide (CaO) [7,8] and lithium orthosilicate (Li₄SiO₄) [7,9,10]. Although CaO exhibits a high theoretical CO₂ adsorption capacity of 0.786 g/g, its practical capacity undergoes rapid degradation with increasing adsorption–desorption cycles [7]. For instance, modified CaO derived from natural limestone usually experiences a rapid decline in adsorption capacity to approximately 0.1 g/g. Synthetic CaO-based adsorbent pellets typically retain only 0.2–0.3 g/g after cyclic adsorption and desorption, accompanied by limited mechanical strength of 1–10 MPa [11,12]. In comparison, doped Li₄SiO₄ particles demonstrate superior cyclic stability, maintaining a practical adsorption capacity of approximately 0.3 g/g during cyclic reactions, along with enhanced compressive strength of 10–30 MPa [10,13]. Furthermore, Li₄SiO₄ requires a regeneration temperature approximately 200 °C lower than CaO for CO₂ desorption and reduces energy consumption during regeneration to about 60% of that required for CaO adsorbents, which confers significant operational energy efficiency advantages for Li₄SiO₄ in industrial-scale applications.

Lithium-based adsorbents represented by Li₄SiO₄ exhibit significant research prospects due to their good adsorption capacity and low regeneration energy consumption [9,14,15,16]. Existing studies have demonstrated that doping with carbonates or inert metal oxides can enhance adsorption performance and improve long-term cycling stability [10,17,18]. However, the improvement in the CO₂ adsorption capacity of Li₄SiO₄ is constrained by its maximum theoretical adsorption value of 36.7 wt.% and cannot be further advanced [19,20,21,22]. In contrast, lithium-based materials synthesized via changing the Li/Si molar ratio or replacing silicates in Li₄SiO₄ have gradually gained attention for their promising CO₂ adsorption potential [4]. Durán-Muñoz et al. [23] synthesized Li₈SiO₆ via a solid-state reaction and reported exceptional CO₂ adsorption performance, ranging from 650 to 700 °C, demonstrating a maximum CO₂ uptake of 52 wt.% at 650 °C under a CO₂ atmosphere. Ávalos-Rendón et al. [24] evaluated the adsorbent of Li₅GaO₄ through dynamic thermogram analysis in a CO₂ flow. It was reported that Li₅GaO₄ was able to adsorb CO₂ above 112 °C, and a maximum adsorption capacity of 39 wt.% was observed from 568 to 709 °C. They also investigated two types of lithium aluminates (LiAlO₂ and Li₅AlO₄), revealing that LiAlO₂ did not capture CO₂, whereas Li₅AlO₄ could adsorb CO₂ in the temperature range of 200–700 °C. It was seen that CO₂ uptake reached 35.7 wt.% at 700 °C in pure CO₂ in the first 3 min. Then, the adsorption rate decreased, and a maximum CO₂ adsorption capacity of 47.8 wt.% was demonstrated after 30 min [25]. Togashi et al. [26] first reported the high-temperature CO₂ adsorption features of Li₄TiO₄ and found that single-phase Li₄TiO₄ can capture CO₂ within the temperature range of 300-856 °C, reaching a maximum adsorption capacity of 42 wt.% at 856 °C. However, the regeneration of Li₄TiO₄ appeared to be very slow, as CO₂ desorption remained incomplete even at 1000 °C. Zhou et al. [27] evaluated the CO₂ adsorption properties of Li₆ZnO₄ in the temperature range of 500–750 °C. The maximum capacity of 76 wt.% was achieved at 700 °C in a mixed gas containing 20 vol.% CO₂, and 60% of the capacity was retained after 10 cycles. Additionally, CO₂ adsorption potentials of other types of lithium-based adsorbents, such as Li₆CoO₄ [28], Li₅FeO₄ [29], and Li₅SbO₅ [30], were explored, but significantly different behaviors in capturing CO₂ were demonstrated.

Through reviewing the literature, it is seen that some novel lithium-based adsorbents exhibit promising CO₂ adsorption potential, with even higher adsorption capacities than the typical Li₄SiO₄ adsorbents. However, current research on novel lithium-based adsorbents remains scarce, and pure CO₂ is mainly employed in adsorption testing, which is far from the requirement in capturing low-concentration CO₂ under flue gas conditions. Additionally, existing studies have primarily focused on evaluating the reaction characteristics of these novel lithium-based adsorbents in single or very limited cycles, with insufficient investigation into their stability under prolonged adsorption–desorption cycling. Therefore, it is imperative to conduct experimental tests on potential novel lithium-based adsorbents under low-concentration CO₂ conditions and systematically compare their adsorption performance and long-term cycling stability. Such efforts could further advance the development and industrial application of high-temperature CO₂ capture technologies.

In summary, the objective of this work is to screen lithium-based adsorbents with high adsorption capacity and cycling stability while elucidating their CO₂ capture characteristics. First, nine adsorbents of α-Li₅AlO₄, β-Li₅AlO₄, Li₆CoO₄, Li₅GaO₄, Li₂CuO₂, Li₅FeO₄, Li₆ZnO₄, Li₈SiO₆, and Li₄SiO₄ with promising CO₂ capacity and reasonable reaction conditions were synthesized using a solid-state reaction method to preliminarily investigate their fundamental physicochemical properties. Subsequently, dynamic CO₂ adsorption–desorption tests and isothermal adsorption experiments were conducted to evaluate the basic reaction characteristics of these materials for CO₂ adsorption. Finally, cyclic adsorption–desorption tests and post-cycling physicochemical features of the adsorbents were analyzed to clarify their reaction behaviors and the evolution of morphology and pore structure during cycling. This work systematically identifies novel lithium-based adsorbents with promising potential, clarifies their fundamental reaction mechanisms and CO₂ capture behaviors, and thereby establishes a foundation for developing next-generation high-temperature CO₂ adsorption technologies.

2. Materials and Methods

2.1. Materials and Adsorbents Synthesis

In this work, all precursors used are analytical reagents (>98% purity), purchased from Shanghai Aladdin Biochemical Technology Co., Shanghai, China, Lithium-based CO₂ adsorbents were synthesized via the solid-state method. Specifically, stoichiometric amounts of Li₂O(Li₂CO₃) and other oxides were precisely weighed in order to synthesize adsorbents with the molecular formula as shown in

Table 1. Precursors and preparation conditions of adsorbents.

Adsorbent	Precursors	Temperature	Atmosphere	Time
α-Li5AlO4	Li2O, γ-Al2O3	500 °C	Air	24 h
β-Li5AlO4	Li2O, γ-Al2O3	900 °C	Air	24 h
Li6CoO4	Li2O, CoO	800 °C	N2	12 h
Li5GaO4	Li2O, Ga2O3	500 °C	Air	24 h
Li2CuO2	Li2O, CuO	685 °C	Air	24 h
Li5FeO4	Li2O, Fe2O3	850 °C	Air	20 h
Li6ZnO4	Li2O, ZnO	800 °C	N2	14 h
Li8SiO6	Li2O, SiO2	800 °C	Air	8 h
Li4SiO4	Li2CO3, SiO2	900 °C	Air	24 h

. The raw materials were intensively mixed using a planetary ball mill at 400 rpm for 65 min. Subsequently, the mechanically mixed powder was compressed into pellets under 5-ton uniaxial pressure and calcined in a muffle furnace under controlled atmosphere conditions with predetermined temperatures. The calcined products were then ground using an agate mortar and sieved to obtain final lithium-based adsorbents.

2.2. Adsorption and Desorption Test

CO₂ adsorption and desorption characteristics of the synthetic adsorbents were evaluated using a thermogravimetric analyzer (NETZSCH, TG209 F3, Selb, Germany). The instrument features a microbalance with a precision of 0.0001 mg, a temperature range of room temperature to 1000 °C, and a heating rate range of 0–50 °C/min. In this work, three types of experiments were conducted: dynamic adsorption–desorption test, isothermal CO₂ adsorption test, and cyclic adsorption–desorption test. Prior to formal testing, blank experiments with empty crucibles were performed under identical temperature, gas flow, and atmosphere conditions to eliminate systematic errors and ensure data accuracy.

Dynamic adsorption–desorption test: This test aims to identify the optimal adsorption temperature window and equilibrium conditions by analyzing continuous CO₂ adsorption and desorption profiles under linearly increasing temperatures. After baseline correction, 5 mg of adsorbent was loaded into the crucible, which was heated from room temperature to 925 °C at a constant rate of 10 °C/min under a gas mixture of 100 mL/min containing 15 vol.% CO₂ balanced by N₂. Real-time mass changes were recorded to generate dynamic adsorption–desorption curves.

Isothermal CO₂ adsorption test: Isothermal tests were performed to determine the temperature-dependent adsorption characteristics. A total of 5 mg of adsorbent was heated to predetermined temperatures at a rate of 10 °C/min under a pure N₂ atmosphere. Upon reaching target temperatures, the gas composition was instantaneously switched to 15 vol.% CO₂ balanced by N₂ while maintaining a total flow rate of 100 mL/min. The adsorption process was monitored isothermally for 1 h to obtain CO₂ adsorption curves.

Cyclic adsorption–desorption test: Cyclic capacity and stability were assessed through 20 consecutive CO₂ adsorption–desorption cycles. Specifically, a 5 mg sample in the crucible was heated from room temperature to target adsorption temperature at a rate of 10 °C/min in N₂, followed by adsorption in 15 vol.% CO₂ balanced by N₂ for 30 min. After the completion of adsorption, the sample was heated to target temperatures at a rate of 20 °C/min in N₂, and desorption was conducted for 20 min. Next, it was cooled to the adsorption temperature at a rate of −20 °C/min, and a complete adsorption and desorption cycle was achieved, which was repeated 20 times.

2.3. Characterization of Adsorbents

Phase composition of the adsorbents synthesized was tested using an X-ray diffractometer (BRUKER, D8 ADVANCE, Karlsruhe, Germany) with a Co target irradiated by an electron beam at 50 kV and 55 mA. Wide-angle X-ray diffraction (XRD) patterns were acquired over a 2θ range of 10–90° at a scanning rate of 6 °/min. Then, phase composition of the adsorbent powder was determined by comparing the obtained diffraction patterns with standard reference patterns. Surface morphology of the synthesized adsorbent was characterized using a scanning electron microscope (ZEISS, GeminiSEM 300, Oberkochen, Germany). Prior to the test, the adsorbent was uniformly dispersed on conductive adhesive and sputter-coated with gold for 60 s to mitigate charging effects caused by its low electrical conductivity. Microporous structural properties of the adsorbent were analyzed using a fully automatic specific surface area and pore size analyzer (MICROMERITICS, ASAP 2460, Norcross, GA, USA). Samples were first degassed under vacuum at 300 °C for 6 h to remove surface impurities. Subsequently, N₂ adsorption–desorption isotherms were measured at −196 °C. The Brunauer–Emmett–Teller (BET) method and Barrett–Joyner–Halenda (BJH) model were applied to calculate the specific surface area, total pore volume, average pore diameter, and pore size distribution, respectively.

3. Results and Discussion

3.1. Microstructural Properties of Fresh Adsorbents

Surface morphologies of nine synthetic adsorbents captured via environmental scanning electron microscopy (ESEM) are shown in Figure 1. A comparison of the SEM images of α-Li₅AlO₄ and β-Li₅AlO₄ reveals distinct structural differences. The surface of the former one is primarily composed of small spherical particles, whereas the latter exhibits a denser arrangement with uniformly distributed sharp flake-like crystals. In contrast to the irregular particle shapes and polydisperse sizes of Li₆CoO₄ and Li₆ZnO₄, the adsorbents of Li₅GaO₄, Li₂CuO₂, Li₅FeO₄, Li₈SiO₆, and Li₄SiO₄ demonstrate much looser surfaces with better pore structures. Notably, unique striped channels are observed on the Li₅FeO₄ surface, which may facilitate gaseous component transport during the CO₂ adsorption process.

Specific surface area, pore volume, and average pore size of the nine lithium-based adsorbents prepared are summarized in Figure 2. A positive correlation is observed between the specific surface area and pore volume. A larger specific surface area provides more active surfaces for CO₂ adsorption and reaction, while a higher pore volume enhances the capacity of the material to accommodate gaseous components, both of which correlate with more developed internal pore structures. Specifically, α-Li₅AlO₄, Li₅FeO₄, and Li₄SiO₄ exhibit relatively higher specific surface areas (2.9605 m²/g, 2.5128 m²/g, and 1.3672 m²/g, respectively) and pore volumes (0.0185 cm³/g, 0.0102 cm³/g, and 0.0068 cm³/g, respectively), indicating their more developed internal pore structure compared to other adsorbents. This structural advantage likely facilitates CO₂ diffusion and reaction by offering abundant transport channels and active sites. In contrast, Li₆CoO₄ and Li₈SiO₆ show the lowest specific surface areas (0.5274 m²/g and 0.5993 m²/g) and pore volumes (0.0030 cm³/g and 0.0040 cm³/g), reflecting their denser internal pore structures. Through analyzing the average pore size, it reveals that Li₂CuO₂, Li₅FeO₄, and Li₄SiO₄ possess smaller average pore diameters of 22.04 nm, 16.30 nm, and 19.75 nm, respectively, while Li₅GaO₄ exhibits the largest average pore size of 30.77 nm. Overall, the data on specific surface area, pore volume, and average pore size suggest that α-Li₅AlO₄, Li₅FeO₄, and Li₄SiO₄ possess more developed pore structures than the other synthetic lithium-based adsorbents.

Further examining pore size distribution curves, it is observed that α-Li₅AlO₄ displays a unimodal distribution centered at around 25 nm, with minimal pores < 10 nm. β-Li₅AlO₄ exhibits a bimodal distribution with peaks at 3.5 nm and 25 nm. Li₆CoO₄ contains limited pores in the 2–4 nm range. Li₅GaO₄ features pores predominantly within 20–60 nm. Li₂CuO₂ and Li₅FeO₄ show concentrated pore diameters at 2–6 nm. Li₆ZnO₄ has a narrow peak at 10 nm, indicating uniform pore size distribution. Li₈SiO₆ and Li₄SiO₄ display pores primarily in 10–40 nm and 3–5 nm ranges, respectively. Notably, Li₆CoO₄, Li₂CuO₂, Li₅FeO₄, and Li₄SiO₄ are dominated by pores < 5 nm, with Li₅FeO₄ exhibiting significantly higher peak intensity, suggesting it has relatively dense small pores and superior meso-pore structure compared to other adsorbents.

3.2. CO₂ Adsorption Characteristics of Nine Adsorbents

To determine adsorption/desorption equilibrium temperatures of the adsorbents and identify the temperature ranges corresponding to fast CO₂ adsorption and desorption rates, dynamic CO₂ adsorption–desorption curves were measured under a 15 vol.% CO₂ atmosphere balanced with N₂ during a linear temperature ramp to 925 °C, as shown in Figure 3. Notably, all adsorbents except Li₄SiO₄ initiate CO₂ adsorption reactions only above 600 °C. The high onset adsorption temperatures consequently result in elevated adsorption/desorption equilibrium temperatures. Due to the temperature limitation of the thermogravimetric analyzer (TGA) used in this study (maximum 925 °C in practice), β-Li₅AlO₄, Li₅GaO₄, Li₆ZnO₄, and Li₈SiO₆ fail to reach equilibrium adsorption/desorption states during dynamic testing. Consequently, their desorption temperatures were set to 925 °C in the subsequent cycling experiments. Furthermore, adsorption/desorption equilibrium temperatures for α-Li₅AlO₄, Li₆CoO₄, Li₂CuO₂, Li₅FeO₄, and Li₄SiO₄ were determined as 847.9 °C, 838.4 °C, 763.5 °C, 783.5 °C, and 644.4 °C, respectively, under 15 vol.% CO₂. Accordingly, their desorption temperatures in cycling tests were set to 900 °C, 900 °C, 850 °C, 900 °C, and 700 °C. Among all adsorbents, α-Li₅AlO₄, β-Li₅AlO₄, Li₅GaO₄, Li₅FeO₄, Li₆ZnO₄, and Li₈SiO₆ exhibit CO₂ adsorption capacities exceeding 40 wt.%, with α-Li₅AlO₄ and Li₆ZnO₄ achieving the highest capacities of 49.97 wt.% and 57.34 wt.%, respectively.

Based on the adsorption temperature ranges derived from dynamic adsorption/desorption curves, isothermal adsorption tests were conducted on lithium-based adsorbents under a 15 vol.% CO₂ atmosphere at three distinct temperatures, with results illustrated in Figure 4. It is observed that most adsorbents approach CO₂ adsorption equilibrium within 10 min, demonstrating rapid adsorption kinetics. Furthermore, owing to the extended reaction time under isothermal conditions, CO₂ adsorption capacities obtained in these tests significantly exceed those from dynamic adsorption/desorption experiments. Adsorbents, including α-Li₅AlO₄, Li₅GaO₄, Li₅FeO₄, Li₆ZnO₄, and Li₈SiO₆, all demonstrate maximum CO₂ adsorption capacities above 56 wt.%. Notably, Li₆ZnO₄ and Li₅FeO₄ achieve maximum adsorption capacities of 84.1 wt.% at 750 °C and 79.9 wt.% at 725 °C, respectively, far surpassing the observed CO₂ capacity of Li₄SiO₄ (16.7 wt.%). The isothermal adsorption results confirm the superior CO₂ adsorption potential of these materials, while even the poor adsorbents of β-Li₅AlO₄ and Li₂CuO₂ still demonstrate capacities of 50.3 wt.% and 34.7 wt.%, respectively. It is noteworthy that all adsorbents in this study were synthesized via a basic solid-state method. The optimization of preparation methods and parameters could potentially further enhance their CO₂ adsorption performance.

To elucidate the post-adsorption products and derive reaction mechanism in CO₂ adsorption, X-ray diffraction (XRD) analysis was performed to characterize the phase composition of the adsorbents before and after CO₂ adsorption. Here, to obtain sufficient sample quantities required for XRD testing, the adsorbents were first subjected to CO₂ adsorption in a tubular furnace under a 15 vol.% CO₂ atmosphere balanced by N₂ at their optimal adsorption temperatures for 1 h, followed by cooling to room temperature prior to XRD measurements. XRD patterns of the adsorbents before and after adsorption are presented in Figure 5. The results indicate that both α-Li₅AlO₄ and β-Li₅AlO₄ react with CO₂ to form Li₂CO₃ and LiAlO₂, with the adsorption reaction described by Equation (1). For Li₆CoO₄, the post-adsorption phases comprise Li₂CO₃ and CoO, corresponding to the reaction in Equation (2). Li₂CuO₂ and Li₆ZnO₄, upon CO₂ adsorption, generate Li₂CO₃ alongside their respective metal oxides (CuO and ZnO), as represented by Equations (3) and (6). In addition to Li₂CO₃, LiGaO₂ and LiFeO₂ are predominantly formed after CO₂ adsorption by Li₅GaO₄ and Li₅FeO₄, respectively, with their reactions governed by Equations (4) and (5). Furthermore, the reaction of Li₈SiO₆ and Li₄SiO₄ with CO₂ both produce Li₂CO₃ and Li₂SiO₃, and the reactions are detailed in Equations (7) and (8). However, it should be noted that the equations presented are the dominant reaction paths, and some secondary reactions may occur during the adsorption and desorption processes. For example, a small amount of phases like Li₁₀Zn₄O₉, LiCoO₂, and Li₂O were observed in the literature, which, however, are not stable and do not easily form the main phases by complex reactions [27,28,31].

2Li₅AlO₄ + 4CO₂ → 4Li₂CO₃ + 2LiAlO₂

(1)

Li₆CoO₄ + 3CO₂ → 3Li₂CO₃ + CoO

(2)

Li₂CuO₂ + CO₂ → Li₂CO₃ + CuO

(3)

2Li₅GaO₄ + 4CO₂ → 4Li₂CO₃ + 2LiGaO₂

(4)

2Li₅FeO₄ + 4CO₂ → 4Li₂CO₃ + 2LiFeO₂

(5)

Li₆ZnO₄ + 3CO₂ → 3Li₂CO₃ + ZnO

(6)

Li₈SiO₆+ 3CO₂ → 3Li₂CO₃ + Li₂SiO₃

(7)

Li₄SiO₄ + CO₂ → Li₂CO₃ + Li₂SiO₃

(8)

3.3. Cyclic Performance and Evolution of Adsorbents

To evaluate the cycling stability of nine lithium-based adsorbents under repeated adsorption/desorption cycles, 20-cycle tests were conducted under their optimal adsorption and desorption temperatures determined from prior experiments. The adsorption involves exposure to a 15 vol.% CO₂ atmosphere balanced by N₂ for 30 min, while desorption was performed under pure N₂ for 20 min. As shown in Figure 6, α-Li₅AlO₄ exhibits an adsorption capacity of 62.1 wt.% in the first cycle, which sharply declines to 45.4 wt.% after the second cycle. Despite some fluctuation, it gradually decreases to 30.2 wt.% at the end of 20 cycles. Similarly, Li₅GaO₄ shows an initial capacity of 48.7 wt.%, dropping to 37.4 wt.% after the second cycle and further declining to 24.3 wt.% after 20 cycles. Li₅FeO₄ demonstrates a deactivation trend analogous to α-Li₅AlO₄, with an initial capacity of 59.2 wt.% in the first cycle. Notably, the total sample mass decreases from 100% to 85.4% after the first desorption cycle, likely due to evaporation, yet its adsorption capacity rebounds to 90.5 wt.% in the fourth cycle and declines gradually to 41.6 wt.% after 20 cycles. Li₆ZnO₄ achieves a remarkable capacity of 80.8 wt.% in the first cycle, which drops to 67.5 wt.% in the second cycle and exhibits a gradual deactivation, thereafter, retaining a CO₂ capacity of 44.2 wt.% at the end of the test. Generally, the four adsorbents of α-Li₅AlO₄, Li₅GaO₄, Li₅FeO₄, and Li₆ZnO₄ display high initial capacities and maintain significant adsorption capabilities even after prolonged cycling. Therefore, further improvements in cycling stability through strategies such as doping are expected to develop novel, efficient, and durable CO₂ adsorbents.

In contrast, β-Li₅AlO₄, Li₆CoO₄, and Li₈SiO₆ exhibit a rapid capacity degradation despite good initial adsorption values of 39.9 wt.%, 48.5 wt.%, and 58.3 wt.%, respectively. After 20 cycles, their capacities decline to 13.3 wt.%, 11.3 wt.%, and 3.2 wt.%, reflecting poor cycling stability. Conversely, Li₂CuO₂ and Li₄SiO₄, though with lower initial capacities (28 wt.% and 19.3 wt.%, respectively), demonstrate exceptional cycling stability. Li₂CuO₂ retains a CO₂ capacity of 21 wt.% after 20 cycles, while Li₄SiO₄ even shows a slight capacity increase to 21.6 wt.%. For these two materials, structural modifications to enhance initial adsorption capacity should be prioritized while preserving their inherent cycling stability.

To investigate morphological changes of the adsorbents induced by cyclic reactions, the samples subjected to 20 cycles of CO₂ adsorption and desorption were imaged using environmental scanning electron microscopy (ESEM), and the images are shown in Figure 7. Compared to their initial morphologies, all adsorbents exhibit varying degrees of sintering after cycling, with the observation of agglomerated larger particles. It is seen that the sharp flake-like particles on the surface of fresh β-Li₅AlO₄ adsorbent almost disappear after cyclic reactions. Notably, Li₆CoO₄, Li₅GaO₄, Li₆ZnO₄, and Li₄SiO₄ display the most severe sintering, with post-cycling surfaces appearing smooth, dense, and non-porous.

Furthermore, N₂ adsorption–desorption isotherms were obtained on the adsorbents after cyclic reaction to analyze their specific surface areas, pore volumes, average pore diameters, and pore size distributions, and the results are summarized in Figure 8. Comparative analysis reveals substantial reductions in specific surface areas and pore volumes after 20 cycles. Among nine adsorbents, Li₆CoO₄, Li₅GaO₄, Li₆ZnO₄, and Li₄SiO₄ exhibit the smallest values in specific surface area and pore volume, consistent with their pronounced sintering observed in Figure 7. These results indicate that repeated cycling not only induces particle aggregation with the observation of transition from point to surface contact, but also damages the internal pore structure. The disappearance of micro- and mesopores due to particle coalescence further diminishes the specific surface area. A comparative analysis of pore sizes before and after cycling demonstrates a general decrease in average pore diameters during cycling. Pore size distribution curves of the four densely sintered adsorbents almost approach a straight line, with minimal pores < 5 nm, suggesting pore evolution toward larger sizes. Additionally, reduced intensity in pore size distribution curves after cycling reflects a significant decline in pore quantity across all size ranges. In summary, the cycling of CO₂ adsorption and desorption triggers surface and internal sintering in adsorbents, characterized by the coalescence of small particles, pore isolation, and depletion of micro- and mesopores. The loss of pores drastically reduces the internal surface area, limiting gas diffusion pathways and CO₂-accessible active sites, ultimately leading to progressive degradation of adsorption performance with cycling.

4. Conclusions

In this work, nine lithium-based adsorbents were synthesized, and their CO₂ adsorption characteristics, cycling stability under adsorption/desorption conditions, and the microstructural evolution during cycling were systematically investigated and compared. Among the nine adsorbents, α-Li₅AlO₄, Li₅FeO₄, and Li₄SiO₄ exhibit relatively developed porous structures and higher specific surface areas. Turning to CO₂ adsorption, α-Li₅AlO₄, Li₅FeO₄, and Li₆ZnO₄ display exceptionally high initial CO₂ adsorption capacities of 62.1 wt.%, 59.2 wt.%, and 80.8 wt.%, respectively. However, their performance gradually degrades during cycling, declining to 30.2 wt.%, 41.6 wt.%, and 44.2 wt.% after 20 cycles. In contrast, β-Li₅AlO₄ and Li₆CoO₄ show moderate initial capacities of 39.9 wt.% and 48.5 wt.%, respectively, but suffer severe cycling instability, with capacities dropping to 13.3 wt.% and 11.3 wt.% after 20 cycles. Li₂CuO₂ and Li₄SiO₄, despite their lower initial capacities of 28 wt.% and 19.3 wt.%, respectively, demonstrate remarkable cycling stability with minimal capacity loss. Moreover, reaction equations for CO₂ adsorption were established through phase composition analysis before and after adsorption. All reactions involve the formation of lithium carbonate and corresponding salts or oxides via interactions between lithium-based adsorbents and CO₂. Morphological and pore structure characterization reveal that sintering during cycling, coupled with a reduction in micro- and mesopores, hinders gas-phase diffusion and surface reactions, ultimately leading to performance degradation.