Porous potassium-titanium co-doped lithium silicate pellets capture CO2 stably over 340 cycles

Carbon capture materials often fail in the real world not because they cannot bind carbon dioxide, but because they cannot survive the journey from laboratory powder to industrial pellet. Lithium orthosilicate, or Li4SiO4, is a leading candidate for high-temperature CO2 capture, with a theoretical capacity of 36.7 wt% and abundant raw materials, yet shaping it into pellets for fluidized bed reactors tends to compress away the very pores that let CO2 reach reactive surfaces. A new study reports a way to build those pores back in without sacrificing the mechanical strength that industrial circulation demands.

Researchers fabricated porous K-Ti co-doped Li4SiO4 pellets, denoted LTSO, using a one-step extrusion-spheronization method with four sacrificial pore-forming agents, or porogens: alpha-cellulose fiber (CF), graphite (C), polyvinyl alcohol (PVA), and polyethylene (PE). Each porogen was added at 10 wt% of the total precursor mass, and the mixtures were ball-milled, plasticized with water, extruded, spheronized at 1300 rpm, and calcined at 700 °C for 8 hours in air. During calcination, the porogens burned away, leaving behind tailored pore networks in pellets measuring 1.0 to 1.5 mm in diameter. The study, published in the journal Carbon Neutrality, systematically compared how each porogen’s geometry and decomposition behavior shaped the final sorbent.

X-ray diffraction confirmed that all samples formed monoclinic Li4SiO4 as the main phase, with weak peaks from a Li2TiO3 secondary phase that stabilizes the crystal skeleton against sintering, and minor reflections from a LiKCO3 eutectic phase whose low melting temperature creates a liquid pathway for ionic and gas diffusion through the product layer during reaction. No carbon residues were detected, confirming complete oxidative removal of all four porogens. Scanning electron microscopy then revealed how each template left its signature: PVA produced worm-like ligaments, graphite created slit-like pores from the removal of laminar flakes, PE generated a foam-like structure dominated by large macropores, and cellulose fiber built a highly interconnected mesoporous network of continuous open diffusion channels penetrating the sorbent bulk.

Nitrogen adsorption measurements quantified these differences. The porogen-free LTSO pellets had a minimal specific surface area of 0.3 m2/g, a consequence of densification during granulation and calcination. In contrast, PE-templated pellets reached 4.4 m2/g and CF-templated pellets 3.6 m2/g, with enhanced porosity concentrated in mesopores of 2 to 40 nm and average pore radii near 20 nm, a regime favorable for bulk CO2 diffusion. Mechanical testing showed the trade-off the porogens must navigate: CF-LTSO retained a crush load of 15.9 N, only a marginal 3.6% reduction from the 16.5 N of dense LTSO, while PE-LTSO dropped to 12.3 N. PVA-LTSO showed the best mechanical performance, consistent with PVA acting as both binder and shaping promoter. Tests at varying CF loadings of 0, 5, 10, and 20 wt% revealed that excessive porogen deteriorates structural cohesion, making 10 wt% the optimal balance.

Kinetic analysis using a double-exponential model separated the rapid surface chemisorption stage from the slower diffusion-controlled stage. At 500 °C under 50% CO2, the surface rate constant k1 of CF-LTSO was 1.4 × 10−1 min−1, higher than the 9.0 × 10−2 min−1 of non-templated LTSO, an enhancement correlated with the increased specific surface area exposing more sites for Li2CO3 nucleation. In the diffusion stage at 600 °C under 15% CO2, CF-LTSO’s k2 reached 2.5 × 10−2 min−1 versus 1.1 × 10−2 min−1 for LTSO, indicating that interconnected mesopores mitigate the product layer resistance that otherwise blocks gas pathways in dense pellets and prematurely terminates capture.

Apparent activation energies told a similar story. The porogen-free LTSO showed the highest barrier for the surface reaction stage at 62.4 kJ mol−1, while porogen addition lowered it to a range of 51.5 to 55.7 kJ mol−1. For diffusion, CF-LTSO demonstrated 46.6 kJ mol−1, significantly lower than the 73.2 kJ mol−1 of LTSO, meaning the sorption rate in the diffusion stage is less sensitive to temperature drops and that resistance to ionic and gas transport is reduced. Temperature-programmed tests further showed that porogen-templated pellets began taking up CO2 at lower onset temperatures under 15% CO2, a simulated flue gas condition, because porosity lowers gas diffusion resistance and minimizes the kinetic hysteresis that dense pellets exhibit.

Cyclic durability testing over 100 isothermal cycles at 600 °C, with 15-minute sorption in 15% CO2/N2 and 15-minute regeneration in pure N2, revealed distinct behaviors. LTSO and PVA-LTSO pellets showed an initial decline in CO2 uptake followed by stabilization, which XRD analysis linked to a temporary phase separation of Ti species, with additional TiO2 phases appearing at the decay stage that likely disrupt Li+ diffusion pathways. PVA-LTSO rebounded rapidly to its initial level, suggesting its mesoporous network acts as a structural buffer facilitating Ti re-incorporation, whereas LTSO recovered sluggishly and incompletely. In contrast, CF-LTSO, C-LTSO, and PE-LTSO showed no obvious decay, maintaining steady capacities of approximately 0.27 gCO2/gsorbent throughout the 100 cycles. Post-cycling SEM showed CF-LTSO retained an integrated spherical morphology while LTSO exhibited significant surface coarsening and pellet fusion from severe sintering.

The decisive test came with an extended 340-cycle run. Both CF-LTSO and PE-LTSO retained stable sorption capacities exceeding 0.25 gCO2/gsorbent after an initial activation period, a longevity that most reported pelletized Li4SiO4 sorbents, typically stable for only 10 to 70 cycles, do not approach. The CF-templated pellets also sustained superior mechanical properties, with compressive strength of 15.9 N and attrition resistance showing less than 5 wt% loss after 2000 rotations, addressing the strength-kinetics trade-off that has limited industrial deployment. Benchmarking against state-of-the-art Li4SiO4-based granules in recent literature showed CF-LTSO exhibits superior or competitive capacity under similar conditions, with performance approaching that of loose Li4SiO4 powders.

The study demonstrates that combining K-Ti co-doping with alpha-cellulose fiber-templated pore engineering resolves the porosity-strength conflict in shaped sorbents. Because Li4SiO4-based sorbents operate optimally at 500 to 700 °C via chemisorption, they are suited for integration with high-temperature processes such as chemical looping and reforming, where MOF-based alternatives suffer from limited thermal stability. The granulation process is simple and scalable, preparing porous sorbents directly from raw materials through extrusion-spheronization followed by a single high-temperature calcination step, offering a rational blueprint for robust, long-life sorbents in industrial fluidized bed carbon capture systems.

Subject of Research: Pore-forming porogen templating of K-Ti co-doped Li4SiO4 pellets for high-temperature CO2 capture

Article Title: Unraveling the pore-forming mechanism and thermokinetic behaviors of porous K-Ti co-doped Li4SiO4 pellets for efficient and stable CO2 capture

Article References: Liu, Y., Jia, Q., Liu, M., Shen, Y., Zhang, R., Wan, Z., Pan, Y., & Jin, H. (2026). Unraveling the pore-forming mechanism and thermokinetic behaviors of porous K-Ti co-doped Li4SiO4 pellets for efficient and stable CO2 capture. Carbon Neutrality, 5(1), Article 11. https://doi.org/10.1007/s43979-026-00166-x

Image Credits: AI Generated

DOI: 10.1007/s43979-026-00166-x

Keywords: CO2 capture, Li4SiO4 sorbents, porogen templating, extrusion-spheronization, pellet mechanical strength, sorption kinetics, thermogravimetric analysis, cyclic stability, mesoporous structure, fluidized bed reactors, K-Ti co-doping, activation energy