Performance of CaMn_0.625Ti_0.125Fe_0.125Mg_0.125O₃ Oxygen Carrier in the Chemical Looping Combustion of Biomass

Abstract

Chemical looping combustion (CLC) has been recognized as a promising CO₂ capture technology, in which oxygen carriers (OCs) transport lattice oxygen to the fuel instead of the air. This study aims to evaluate a newly developed perovskite OC for biomass CLC and to clarify the role of staged fuel conversion in improving gas–solid redox efficiency. This is the first application of perovskite OC CaMn_0.625Ti_0.125Fe_0.125Mg_0.125O₃ in biomass CLC using a dual-stage fluidized bed. The perovskite OC was synthesized via a solid-phase synthesis method, and its performance in a dual-stage fluidized bed reactor was evaluated using pine wood chips and furfural residues as model solid fuels. The in situ conversion of volatile compounds and gasification products derived from the two biomass types was comprehensively studied. The effects of operational parameters, including temperature, OC-to-biomass ratio, and gas flow rate, on the combustion efficiency and CO₂ yield were examined. Results showed that separated gasification–combustion enhanced the combustion efficiency and CO₂ yield. At 950 °C, an OC-to-pine chip ratio of 100:1, and a gas flow rate of 0.7 L/min, the maximum combustion efficiency and CO₂ yield of 79% and 82% were obtained, respectively. Moreover, under the optimal gasification conditions (gasification rate > 99%), increasing the fuel concentration resulted in an increase in the oxygen release from 0.21 g to 0.40 g. Concurrently, the corresponding total oxygen demand increased from 4.34% to 10.56%, indicating the suitability of CaMn_0.625Ti_0.125Fe_0.125Mg_0.125O₃ in the CLC of biomass.

1. Introduction

Biomass chemical looping combustion (BCLC) is an advanced carbon-negative combustion technology that integrates the intrinsic CO₂ capture capability of chemical looping systems with the carbon-neutral properties of biomass [1]. The technology is operated via a redox-mediated process, in which metal oxide oxygen carriers (OCs) undergo cyclic redox reactions in the air and fuel reactors. During pyrolysis in the fuel reactor, biomass is thermochemically decomposed to obtain volatile species (e.g., H₂, CO, CO₂, and CH₄) and residual char (R1) [2]. Although gaseous components readily form gas–solid interactions with OCs (R2), the solid char must be gasified in situ [3] via steam/CO₂ injection (R3) to produce syngas (primarily CO and H₂) for subsequent redox reactions (R4 and R5), ultimately generating CO₂ and H₂O for sequestration [4].

Biomass → volatiles (CO, H₂, and CH₄) + Char

(R1)

volatiles + MeO_x → MeO_x−1 + CO₂ (g) + H₂O (g)

(R2)

Char + H₂O (g) → CO (g) + H₂ (g) + ash

(R3)

H₂ (g) + MeO_x → MeO_x−1 + H₂O (g)

(R4)

CO (g) + MeO_x → MeO_x−1 + CO₂ (g)

(R5)

Recent advancements have focused on chemical looping oxygen uncoupling (CLOU) materials that release gaseous oxygen at high temperatures (700–1200 °C), enhancing the solid-to-fuel conversion efficiency [5]. Current CLOU candidates include transition metal oxides (e.g., CuO [6,7], Co₃O₄ [8], and Mn₂O₃ [8,9]) and perovskite-type oxides. Among the two, perovskites (ABO₃/AB₂O₄) are particularly promising because of their nonstoichiometric oxygen capacity, structural stability, and redox durability [10]. The compositions of the A-site (typically alkaline earth metals, e.g., La [11] and Ca [12]) and B-site (transition metals, e.g., Mn [9], Fe [13,14], and Cu [15]) govern the oxygen vacancy formation and ionic mobility in the perovskite [16]. While CaMnO_3−δ systems have emerged as cost-effective candidates, their practical implementation requires elemental doping to mitigate structural instability during prolonged cycling [17]. Experimental studies have shown that B-site substitution with Ti, Mg, and Fe can considerably improve the performance of perovskites. Mattisson-Henrik et al. demonstrated that the doping of Ti and Fe in CaMnO_3−δ maintains the oxygen uncoupling capacity while stabilizing the crystalline phases. Among these, Ti and Fe effectively form a crystalline phase with Mn, whereas its Mg doping results in the formation of separated MgO domains [18]. Rydén et al. demonstrated the temperature-dependent O₂ release kinetics on CaMn_0.875Ti_0.125O₃ (Oxygen uncoupling performance improves with increasing temperature), achieving a CH₄ combustion efficiency of 99.8% via optimized oxygen partial pressure control [19]. Belviso et al. synthesized the CaMn_0.9Mg_0.1O_3−δ OC and conducted continuous testing in a 10 kWth reactor. The results confirmed a sustained operation (>350 h) with a 3% mass-specific O₂ release at 850 °C; they attributed this mechanical stability to Mg incorporation [20].

However, interactions between biomass ash and perovskite OCs pose critical challenges in CLOU, particularly due to the alkali-induced deactivation of OCs [21]. Ash components are partitioned into volatile fly ash (e.g., K, Na, S, and Cl) and residual bottom ash (e.g., Ca, Si, Mg, Al, and Fe), with SiO₂ and alkali metals having contrasting effects [22]. Research indicates that specific ash elements—notably potassium (K), calcium (Ca), and silicon (Si)—interact detrimentally with OCs, suppressing their reactivity [23]. K exhibits a propensity to migrate into the bulk structure of OC particles. This migration facilitates the formation of thermodynamically stable compounds, specifically KTi₈O_16.5, which irreversibly sequesters active material. Ca, conversely, predominantly accumulates on the external surfaces of OC particles. This surface enrichment promotes particle agglomeration, thereby disrupting the hydrodynamic behavior essential for effective fluidization within reactor systems [24]. Si reacts chemically with OC components, generating distinct, inert metal silicate phases—notably Fe₂SiO₄ (fayalite) and Al₂SiO₅ (aluminum silicate). The formation of these phases depletes the concentration of redox-active species within the OCs and significantly impairs their fundamental lattice oxygen transfer capacity [25,26,27]. Furthermore, fine particulate ash present in the reaction environment deposits onto the porous surfaces of OCs. This deposition process leads to the development of a cohesive, low-melting-point ash layer. This layer acts as a physical barrier, hindering direct gas–solid contact at the reactive sites. Consequently, carbon conversion efficiency is reduced [28], fluidization dynamics are adversely affected, and the sensitivity of the OCs to oscillating redox conditions is diminished [29]. Gu et al. observed that SiO₂-rich wheat straw ash promotes particle sintering via K-silicate formation, whereas K-rich ash promotes the reactivity of iron-based OCs [27]. Zhang’s findings corroborated this dual behavior of SiO₂-rich wheat straw ash, reporting a threshold ash content of 20%, beyond which sintering dominates the morphological evolution of the OC [30].

The speciation of alkaline metals plays a crucial role in the deactivation mechanisms of OC. KCl dominates the gas-phase alkali emissions during CLOU, with the atmosphere-dependent adsorption on OCs [31,32]. Andersson et al. performed TGA-MS analysis and reported the improved adsorption of KCl on CaMn_0.775Ti_0.125Mg_0.15O_3−δ OC under reducing conditions, with the occurrence of temperature-dependent absorption kinetics [33]. In addition to alkali metals, acidic elements (e.g., S, Cl) in solid fuels can also induce poisoning and deactivation of OCs. Sulfur compounds such as SO₂ and H₂S create additional poisoning pathways via the formation of CaSO₄ or CaS, respectively, as evidenced by a 28% decrease in the combustion efficiency of CaMn_0.9Mg_0.1O_3−δ during prolonged H₂S exposure. Moreover, the formation of a CaSO₄ crystalline phase was observed on the surface of the OC. CaMn_0.775Ti_0.125Mg_0.1O_2.9−δ in simulated syngas fuel (3000 ppm H₂S in CH₄) for 29 h of continuous operation. Compared to pure CH₄ fuel, combustion efficiency decreased by 18%, with CaSO₄ also identified in the OC. Current research on the combustion performance and morphological evolution of perovskite oxygen carriers with solid fuels predominantly focuses on mixed combustion within single-reactor systems, where fuel pyrolysis gases are typically oversimplified using CH₄ or CO for simulated combustion. This approach substantially deviates from actual pyrolysis conditions. Hence, investigating the reactivity between solid fuel pyrolysis gases and perovskite oxygen carriers in dual-reactor systems is essential.

Accordingly, this study aims to investigate the applicability of perovskite OC CaMn_0.625Ti_0.125Fe_0.125Mg_0.125O₃ for biomass chemical looping combustion in a dual-stage fluidized bed reactor, with particular emphasis on the conversion behavior of biomass pyrolysis and gasification products. Pine wood chips and pine wood char were selected as solid fuels to evaluate the CLC performance of CaMn_0.625Ti_0.125Fe_0.125Mg_0.125O₃ in two combustion modes, i.e., mixed and separated combustion modes. The optimal combustion mode (i.e., separated combustion) was further investigated by analyzing the effects of the temperature and biomass-to-OC ratio (B/O ratio) on the combustion efficiency and CO₂ yield. The microevolutionary behavior of OC particles was investigated using scanning electron microscopy–energy-dispersive X-ray spectroscopy (SEM–EDS) and X-ray diffraction (XRD).

2. Materials and Methods

2.1. Perovskite OC Synthesis

The perovskite-type OC, CaMn_0.625Ti_0.125Fe_0.125Mg_0.125O₃, was prepared via solid-state synthesis. Stoichiometric amounts of precursor materials, totaling 400 g, were accurately weighed, including 201.70 g of Ca(OH)₂ (Xin Shen Test Chemical Technology Co., Ltd., Wuhan, China, analytical grade, ≥99%), 27.15 g of TiO₂ (Xin Shen Test Chemical Technology Co., Ltd., Wuhan, China, ≥99%), 27.00 g of Fe₂O₃ (Xin Shen Test Chemical Technology Co., Ltd., Wuhan, China, ≥99%), 129.45 g of Mn₃O₄ (Xin Shen Test Chemical Technology Co., Ltd., Wuhan, China, ≥99%), and 14.67 g of MgO (Xin Shen Test Chemical Technology Co., Ltd., Wuhan, China, ≥99%). The mixture was homogenized by mechanical blending and then dried at 105 °C for 12 h to remove the adsorbed moisture. Next, the powder was milled in a planetary ball mill (Changsha Tianchuang XQM-4, Changsha, China) at 300 rpm for 3 h to reduce its particle size and increase its reactivity. The powder was granulated via a uniaxial hydraulic press (Changsha Miqi MD-30, Changsha, China) at a uniaxial pressure of 5 MPa to form cylindrical pellets, which were disaggregated using a mechanical crusher and sieved to obtain particles of 150–350 μm for subsequent redox testing.

2.2. Biomass Feedstock Preparation

Pine sawdust (PS) was selected as the experimental feedstock. Before experiments, it underwent the following standardized pretreatment procedures: (1) dehydration at 120 °C for 12 h under atmospheric pressure, (2) mechanical pulverization, and (3) particle-size classification (150–350 μm) using ASTM-standard sieves. PS-derived char (PSC) was produced via pyrolytic decomposition in a tubular reactor in an N₂ atmosphere at 850 °C for 30 min at a heating rate of 10 °C/min. Proximate and ultimate analyses were performed according to ASTM D7582 [34] and ASTM D5373 [35] protocols, respectively. The elemental and industrial analysis results of the biomass are presented in

Table 1. Elemental and proximate analyses of the biomass.

Sample	Elemental Analysis (wt.%, ad)					Industrial Analysis (wt.%, ad)
	C	H	N	S	O *	FC	V	M	A
PS	47.07	5.82	0.17	0.24	46.70	18.60	77.16	0.88	3.36
PSC	79.98	1.83	0.9	0.43	11.36	78.31	9.77	6.42	5.50

Note: FC, fixed carbon; V, volatile; M, moisture; A, ash. * The mass percentage of O is calculated as the difference between the total wt.% and the sum of the wt.% of individual elements.

.

2.3. Experimental Equipment and Procedures

Herein, the reaction performance of CaMn_0.625Ti_0.125Fe_0.125Mg_0.125O₃ was systematically examined in a dual-stage fluidized bed reactor [36] using biomass pyrolysis–gasification and redox reaction experiments. The reactor system (Figure 1) included two independently controlled stages, i.e., a primary reactor (T1, 80 mm × 30 mm) for biomass conversion and a secondary reactor (T2, 160 mm × 30 mm) for OC-mediated syngas combustion. Exhaust gases were sequentially treated by particulate filtration, tar condensation, and desiccant drying before real-time compositional analysis via nondispersive infrared (NDIR) spectroscopy, with detection limits of CO/CO₂ ± 50 ppm, CH₄/H₂ ± 100 ppm, and O₂ ± 0.1 vol.%.

The specific experimental conditions are shown in

Table 2. Experimental conditions.

Test Number	Conditions	T (°C)	Bed Material	Fuel	No. of Cycles	Atmosphere
1 (Normal CLC)	Mixed	950	30 g OC	0.3 g PS	3	0.7 L/min N2 + 0.3 L/min H2O
2 (Conversion of the gasification products of biomass)	Separated	950	30 g OC	0.3 g PS	3	0.7 L/min N2 + 0.3 L/min H2O
3 (Biomass pyrolysis)	Separated	850	30 g SiO2	0.3 g PS	3	1 L/min N2
4 (Conversion of pyrolysis products)	Separated	850	30 g OC	0.3 g PS	3	1 L/min N2
5 (Conversion of pyrolysis products)	Separated	900	30 g OC	0.3 g PS	3	1 L/min N2
6 (Combustion of pyrolysis products)	Separated	950	30 g OC	0.3 g PS	3	1 L/min N2
7 (Combustion of pyrolysis products)	Separated	950	30 g OC	0.4 g PS	3	1 L/min N2
8 (Combustion of pyrolysis products)	Separated	950	30 g OC	0.6 g PS	3	1 L/min N2
9 (In situ char gasification)	Separated	850	30 g SiO2	0.3 g PSC	3	0.7 L/min N2 + 0.3 L/min H2O
10 (Conversion of in situ char gasification products)	Separated	850	30 g OC	0.3 g PSC	3	0.7 L/min N2 + 0.3 L/min H2O
11 (Conversion of in situ char gasification products)	Separated	900	30 g OC	0.3 g PSC	3	0.7 L/min N2 + 0.3 L/min H2O
12 (Conversion of in situ char gasification products)	Separated	950	30 g OC	0.3 g PSC	3	0.7 L/min N2 + 0.3 L/min H2O
13 (Conversion of in situ char gasification products)	Separated	950	20 g OC	0.3 g PSC	3	0.7 L/min N2 + 0.3 L/min H2O
14 (Conversion of in situ char gasification products)	Separated	950	10 g OC	0.3 g PSC	3	0.7 L/min N2 + 0.3 L/min H2O

. For BCLC in the mixed mode, OCs (0.15–0.35 mm) were preoxidized in the primary reactor under an air flow (950 °C, 30 min), followed by steam injection and biomass feeding in an N₂ atmosphere. Reduction cycles were terminated upon CO₂ depletion (<0.5%), followed by oxidative regeneration with air. During separated biomass pyrolysis experiments, OCs were preoxidized in the primary reactor in air. Subsequently, biomass particles (0.15–0.35 mm) were introduced into the secondary reactor via high-pressure N₂ pulse injection in an N₂ environment. The pyrolysis gas resulting from high-temperature biomass decomposition was transferred to the primary reactor for the reaction with the preoxidized OC. Separated biomass gasification experiments were performed using a procedure similar to pyrolysis experiments, with the exception that steam was coinjected (0.1–0.5 mL/min) with biomass particles into the secondary reactor to promote gasification. Triplicate experiments were performed to ensure reproducibility (relative standard deviation < 5%), with critical parameters, including the pulse duration (100 ms) and thermal profiles (pyrolysis zone = 950 °C), rigorously controlled. This dual-mode approach enabled the comparative analysis of staged and integrated biomass conversion pathways while quantifying OC performance degradation mechanisms.

2.4. Data Evaluation

The combustion efficiency was calculated as follows:

$${\eta }_c=1-{\displaystyle \frac{{\int }_{t_0}^{t_1}\left(4F_{{CH}_4}^{\prime } +{ F}_{CO}^{\prime } +{ F}_{H_2}^{\prime } + 2F_{{CO}_2}^{″}\right)dt}{m_{fuel} \times {\mathsf{\Omega }}_{O_2}}}$$

(1)

where $F_i^{\prime }$ (i = CH₄, CO, and H₂) is the molar flow rate during the reduction phase (mol/s), $F_{{CO}_2}^{″}$ is the molar flow rate of CO₂ during the oxidation phase (mol/s), $t_0$ is the starting time of the reduction phase (s), $t_1$ is the end time of the reduction phase (s), $m_{fuel}$ is the mass of the fuel added (g), and ${\mathsf{\Omega }}_{O_2}$ is the molar amount of oxygen required for the complete combustion of the fuel (mol/g).

The carbon capture efficiency was calculated as follows:

$${\eta }_{CC}={\displaystyle \frac{{\int }_{t_0}^{t_1}(F_{{CH}_4}^{\prime } + F_{CO}^{\prime }+ F_{{CO}_2}^{\prime })dt}{{\int }_{t_0}^{t_{total}}(F_{{CH}_4} + F_{CO} + F_{{CO}_2})dt}}$$

(2)

where $F_i$ (i = CH₄, CO, and H₂) is the molar flow rate throughout the oxidation–reduction process (mol/s) and $t_{total}$ is the end time of the entire oxidation–reduction process (s).

The gas yield was calculated as follows:

$${\mathsf{\gamma }}_i={\displaystyle \frac{{\int }_{t_0}^{t_1}\left(F_i^{\prime }\right)dt}{{\int }_{t_0}^{t_1}(F_{{CH}_4}^{\prime }+ F_{CO}^{\prime }+ F_{{CO}_2}^{\prime }+ F_{H_2}^{\prime })dt}}$$

(3)

The gas yield can be interpreted as the percentage of the gas component ${\mathsf{\gamma }}_i$ (i = CO, CO₂, CH₄, and H₂) in the total gas produced.

The carbon conversion rate was calculated as follows:

$$r_i={\displaystyle \frac{F_{{CH}_4}^{\prime }+ F_{CO}^{\prime }+ F_{{CO}_2}^{\prime }}{m_{fuel \times }{n}_{c,fuel}}}$$

(4)

The carbon conversion rate reflects the rate at which the carbon in the biomass is transformed into different gaseous components during the gasification process.

The carbon conversion efficiency was calculated as follows:

$$X_c={\displaystyle \frac{{\int }_{t_0}^{t_1}(F_{{CH}_4}^{\prime }+ F_{CO}^{\prime } + F_{{CO}_2}^{\prime })dt}{m_{fuel} \times n_{c,fuel}}}$$

(5)

2.5. Characterization Methods

The morphology and surface elemental distribution of the fresh and reacted oxygen carrier particles were analyzed using scanning electron microscopy coupled with energy-dispersive X-ray spectroscopy (SEM–EDS). SEM was employed to examine the particle surface structure and morphological evolution after different chemical looping combustion modes, while EDS mapping was used to identify the spatial distribution of major elements and trace impurities on the oxygen carrier surfaces. The textural properties of the oxygen carrier, including the specific surface area, pore volume, and average pore diameter, were determined by N₂ adsorption–desorption measurements based on the Brunauer–Emmett–Teller (BET) method. Prior to analysis, the samples were degassed to remove physically adsorbed species. The crystalline phases of the oxygen carrier before and after redox reactions were identified by X-ray diffraction (XRD). XRD analysis was performed to investigate phase composition, structural stability, and possible phase evolution of the perovskite oxygen carrier during biomass chemical looping combustion.

3. Results and Discussion

3.1. Carbon Balance Analysis

As shown in

Table 3. Carbon balances.

Test Number	Carbon Balance (%)	Conditions
1	103.1 ± 1.2	Mixed
2	106.7 ± 1.4	Separated
3	107.7 ± 1.4	Separated
4	105.7 ± 0.8	Separated
9	104.5 ± 2.4	Separated
10	106.1 ± 1.5	Separated

, the carbon balance under different experimental conditions ranged from 103.1% ± 1.2% to 107.7% ± 1.4% across triplicate runs. The values exceeding 100.0% were attributed to tar deposition in the reactor and instrumental errors. Notably, in separated gasification experiments, carbon release during pyrolysis and gasification accounted for 76.4% and 23.6% of the total carbon input, respectively.

3.2. Effect of Mixed and Separated Modes of CLC

The gasification contact mode plays a crucial role in the syngas conversion efficiency of BCLC. As shown in Figure 2 and Figure 3, on PS, at 950 °C and 30% H₂O, the separated contact mode achieves superior performance, with a CO₂ concentration peak and a combustion efficiency is 24.04% (vs. 22.18% in the mixed mode) and 89% (vs. 79% in the mixed mode), respectively. This increase is attributed to the quick generation and then limited residence time of syngas in the mixed mode between the biomass and OC. In the separated mode, the syngas generated in the first reactor serves as the fluidizing gas in the second reactor filled with OC, which allows it to sufficiently interact with the OC. Additionally, the perovskite OC preferentially oxidizes CO due to its high reactivity and simple reaction pathway (CO + O → CO₂), which reduces the CO yield from 10.04% to 0.34%, while the CH₄ yield decreases from 5.65% to 2.81% because of its higher activation energy requirements. The separated mode improves syngas conversion from volatile compound–rich biomass by aligning gas generation with the redox capacity of the OC, thereby maximizing carbon utilization and combustion efficiency. This also suggests that the gas–solid contact should be considered in industrial plants.

3.3. Performance of Biomass Pyrolysis and Gasification in the Separated Mode

BCLC comprises two distinct stages, i.e., pyrolysis and gasification. The reactivity of the OC varies between these stages, necessitating the separate evaluation of its performance in each stage. Experimental results demonstrate that the separated mode yields a higher combustion efficiency and carbon conversion, more effectively reflecting the redox capability of the OC. Therefore, the reactivity of the OC toward the products of biomass pyrolysis and gasification was investigated under separated mode conditions. Furthermore, fuels employed in experiments, i.e., in situ produced biomass pyrolysis gas and gasification gas, better reflect practical conditions than synthetic gas mixtures typically used in laboratory studies.

3.3.1. Conversion of the In Situ Pyrolysis Gas from Biomass by OC

Biomass pyrolysis involves the thermochemical decomposition of organic materials under limited-oxygen or anaerobic conditions via sequential reactions, including dehydration, depolymerization, isomerization, decarboxylation, and carbonization. This process generates three primary product streams: pyrolysis oil (e.g., tars and aromatic hydrocarbons), noncondensable gases (e.g., CO, CO₂, H₂, C1–C4 hydrocarbons, and H₂S [37]), and solid residues (e.g., biochar and ash). In CLC, OCs directly interact with pyrolysis-derived volatiles via gas–solid redox reactions. The OCs facilitate the oxidation of combustible gases (e.g., CO, H₂, and CH₄) and catalytically reform tar via partial oxidation [38]. Simultaneously, alkali metals (e.g., K and Na) and contaminants (e.g., S, Cl, and heavy metals) in biomass are deposited onto OC surfaces [39], leading to the deactivation of OCs via three primary mechanisms: (1) pore blockage due to ash agglomeration, (2) active site poisoning due to sulfate/chloride formation [40], and (3) structural degradation due to redox-induced phase transformations.

PS underwent fast pyrolysis in a dual-stage fluidized bed reactor in an N₂ carrier gas to generate pyrolysis volatiles. CaMn_0.625Ti_0.125Fe_0.125Mg_0.125O₃ exhibits a significant oxygen uncoupling capacity at high temperatures, enabling the release of gaseous oxygen for syngas conversion. As shown in Figure 4a, the primary pyrolysis product at 850 °C is CO (yield = 45.39%) in the absence of OCs. After OC introduction, catalytic oxidation converts 99% of CO to CO₂, reducing the CO yield to 0.52% and increasing the CO₂ yield from 24.75% to 84.52%. Concurrently, H₂ and CH₄ yields decrease, but the reduction in the CH₄ yield is less pronounced (13.51% → 7.59%) than that in the H₂ yield because of kinetic limitations from gas-phase diffusion barriers and high activation energy (≥120 kJ/mol) [41].

Figure 4b shows the OC-mediated carbon conversion dynamics. During the initial reaction stages (carbon conversion = 0–2.5%), OCs exhibit a negligible enhancement in the conversion rates due to a high syngas flux. Progressive pyrolysis intensifies the catalytic effects of the OC, resulting in an increase in the peak carbon conversion rates from 1.38%/s (quartz sand) to 1.68%/s (OCs). Notably, introducing OCs does not alter the total carbon conversion efficiency (21.39% vs. 23.68% for inert bed materials), probably because unconverted carbon compounds, such as C₂H₄, tar, or undetectable heavy hydrocarbons, are converted to measurable CO₂ [42], thus confirming their role in accelerating kinetics without modifying thermodynamic equilibria.

3.3.2. Effect of OC on In Situ Biomass Char Gasification Gas

In situ produced biochar, which is the carbon-rich solid residue produced during biomass pyrolysis, undergoes steam-assisted gasification (C + H₂O → CO + H₂; CO + H₂O ⇌ CO₂ + H₂) to generate syngas for subsequent OC-mediated combustion. At 850 °C and 30 vol.% H₂O (N₂ balance), PSC is gasified after complete pyrolysis (200 s in an N₂ atmosphere), producing syngas primarily composed of H₂, with a peak concentration of 3.05%, and smaller amounts of CO, CO₂, and CH₄ (Figure 5a). The gasification process lasts for 1000 s, after which the gas concentrations return to baseline levels. OC introduction results in the near-complete oxidation of the syngas, as indicated by the detection of only CO₂ in the exhaust streams (Figure 5b).

Figure 6a illustrates the effect of OCs on the gas yield and carbon conversion kinetics. After the addition of the OC, the CO₂ yield considerably increases from 30.31% (inert bed) to 87.54% (OC bed), while the H₂ yield sharply decreases from 60.63% to 5.97%. This shows the selective oxidation behavior of OCs, where CO preferentially adsorbs onto active sites for rapid oxidation (CO + O* → CO₂), whereas H₂ requires dissociative adsorption (H₂ → 2H) and a higher activation energy to form H₂O [43]. The CH₄ yield remains stable (~3.5%) because of kinetic limitations in C–H bond cleavage. Transition metals, such as Fe, within the perovskite structure catalyze tar cracking, converting macromolecular carbon into CO₂ via surface-mediated oxidation, thereby improving the carbon conversion efficiency. The carbon conversion rates exhibit distinct profiles: a unimodal curve on quartz sand (indicating single-step gasification) and a bimodal trend in OC systems (Figure 6b). The secondary peak is due to CO disproportionation (2CO → C + CO₂), which temporarily suppresses conversion rates before steam reforming becomes dominant [44].

3.4. Effect of the Temperature and B/O Ratio

3.4.1. Conversion of Pyrolysis Products in the Separated Mode

The temperature critically modulates biomass pyrolysis dynamics and combustion efficiency in chemical looping systems, as demonstrated by the thermochemical evolution of PS in different thermal regimes (850–950 °C). As shown in Figure 7, CaMn_0.625Ti_0.125Fe_0.125Mg_0.125O₃ facilitates the oxidative conversion of pyrolysis volatiles at 850 °C, achieving a CO₂ yield of 87.52% via preferential CO oxidation and partial tar cracking. When the temperature increases to 900 °C, gas-phase oxidation kinetics are enhanced, increasing the CO₂ yield to 93.62% via intensified tar decomposition (92%) and steam-assisted reforming (H₂ yield = 4.2% → 6.7%, ΔT = +50 °C). After further heating to 950 °C, the CO₂ yield marginally improves (95.4%), but the carbon conversion efficiency is the same (25.1%), reflecting thermodynamic equilibrium constraints.

The oxygen-to-fuel ratio plays a critical role in determining the combustion efficiency, product distribution, and environmental impact in CLC systems. Excess oxygen promotes complete combustion, reducing the emissions of CO and unburned hydrocarbons while improving the energy conversion efficiency. However, this is accompanied by drawbacks: increased NO_x formation and higher auxiliary energy consumption for oxygen supply. By contrast, insufficient oxygen leads to incomplete fuel conversion, resulting in high concentrations of greenhouse gases (e.g., CO) and volatile organic compounds (VOCs) [45]. To address this trade-off, we adjusted the B/O ratio, which utilizes the oxygen uncoupling capability of perovskite materials. Figure 8 shows the combustion performance of CaMn_0.625Ti_0.125Fe_0.125Mg_0.125O₃ at 950 °C at different B/O ratios, which demonstrates that increasing the B/O ratio from 0.01 to 0.02 has a negligible effect on the combustion efficiency, which remains stable at ~74%. This is because of the superior oxygen release capacity of the perovskite, with 30 g of the OC providing more gaseous oxygen than is required for the complete combustion of PS. However, as the B/O ratio increases, the H₂ yield significantly decreases from 3.9% to 1.6% while the CH₄ yield remains constant at ~3.5%. This decrease in the H₂ yield is attributed to the reduced catalytic activity of the perovskite at higher concentrations of biomass-derived syngas because at high concentrations surface-active sites become saturated, hindering steam reforming reactions (C_xH_y + H₂O → CO + H₂).

3.4.2. Conversion of Gasification Products in the Separated Mode

Temperature-dependent analysis (Figure 9) shows that the combustion efficiency increases from 83.13% at 850 °C to 88.75% at 950 °C due to improved redox kinetics. At 950 °C, the initial carbon conversion rates decrease because of the dominance of CO disproportionation. However, subsequent steam reforming increases the rates (1.58%/s vs. 1.12%/s at 900 °C). This thermochemical interaction underscores the dual role of the temperature in modulating reaction pathways (disproportionation vs. oxidation) and optimizing carbon utilization.

In the separated gasification CLC experiments, 0.3 g of biomass char was used, and the B/O ratio was adjusted by varying the OC mass (10, 20, and 30 g), which is different from the separated pyrolysis CLC mode. As illustrated in Figure 10, at 950 °C, a B/O ratio of 0.03 (10 g of OC) yields 78.1% of CO₂, along with 9.46% of CO and 10.15% of CH₄. When the B/O ratio decreases to 0.01 (30 g of OC), the CO and CH₄ yields decrease to 0.34% and 2.83%, respectively, while the CO₂ yield increases to 94.66%. Correspondingly, the combustion efficiency increases from 69.57% to 88.71%. These results indicate that a higher OC loading markedly improves biomass char gasification gas conversion by supplying abundant active sites and lattice oxygen, thereby promoting the oxidation of unconverted CO and CH₄.

3.5. Characterization of OCs

3.5.1. FESEM

Field-emission SEM (FESEM) was employed to study the apparent morphology of the OC under different experimental conditions. As shown in Figure 11, the fresh OC (Figure 11(A1,A2)), synthesized via the solid-state method, exhibits a smooth surface with well-developed porous structures. After mixed combustion (Figure 11(B1,B2)), the average particle size of the OC markedly increases from 157.14 to 300.21 μm. This increase is attributed to biomass ash deposition on the OC surface, resulting in direct contact between the OC and fuel during combustion. After separated gasification (Figure 11(C1,C2)) and pyrolysis (Figure 11(D1,D2)), the OC surface becomes rough due to thermal stress and redox reactions, with increased porosity in nonsintered regions.

3.5.2. EDS

EDS mapping reveals a homogeneous distribution of Ca, Mn, Ti, Fe, and O on the surfaces of fresh and reacted OCs, confirming the structural integrity of the perovskite matrix. Mg is slightly separated on OC, probably due to its lower ionic mobility in the lattice. Notably, S is not detected in fresh OCs (Figure 12A) but is detected in all reacted OCs (Figure 12B–D), appearing as discrete hotspots on their surfaces. After CLC in the mixed mode, sulfur detected on the OC surface may have originated from the biomass ash deposited on OC particles during combustion. By contrast, after CLC in the separated mode, sulfur detected on the OC surface may have originated only from gaseous sulfur compounds (e.g., H₂S and SO₂) generated during biomass pyrolysis/gasification. This evidence indicates that the separated combustion mode not only enhances combustion efficiency but also establishes traceable corrosion pathways for acidic elements to degrade the OC.

3.5.3. BET

Table 4. Specific surface areas, pore volumes, and pore size of OCs after reactions under different conditions.

OC	Specific Surface (m2/g)	Pore Volume (cm3/g)	Pore Diameter (nm)
Fresh	0.19	0.0003	33.41
Mixed combustion	0.34	0.0005	35.69
Separated gasification	0.87	0.0004	29.99
Separated pyrolysis	0.75	0.0002	36.72

shows the Brunauer–Emmett–Teller (BET) analysis results of the OCs, which show that the specific surface area of the OC increases from 0.19 m²/g to 0.34, 0.87, and 0.75 m²/g, respectively, while the pore volume and pore size remain almost unchanged. This is attributed to (1) the capacity of CaMn_0.625Ti_0.125Fe_0.125Mg_0.125O₃ OC to release gaseous O₂ and (2) structural lattice changes during redox cycles that facilitate the formation and expansion of O₂ release pathways, which result in the better development of particle microporosity. Notably, the specific surface areas of the separated pyrolysis and gasification samples are 2.4 and 2.3 times greater than those of the mixed combustion samples, respectively. This difference can be attributed to biomass ash accumulation, which is in agreement with the SEM results (Figure 11(B1,B2)). This demonstrates that intensive gas-phase redox cycling enriches the pore structure of the OC, increasing the gas–solid contact area and thereby improving its reactivity.

3.5.4. XRD

The XRD patterns of the samples are presented in Figure 13. The primary crystalline phases of the fresh sample synthesized via ball milling and high-temperature calcination are CaTiO₃, CaFe(Ti₂O₆), and Ca_0.96Mn_0.32Si_0.64O_2.56. These phases serve as the main redox-active components. In the samples reacted under the three conditions, a new phase, CaMnO₃, is detected, which represents the decomposition product of the pristine inclusion crystals subjected to cyclic redox stress. The crystalline phases observed in the XRD patterns are consistent with previous reports on Ca–Mn–Ti–Fe perovskite systems [46,47]. Concurrently, the peak intensities of CaTiO₃ and Ca_0.96Mn_0.32Si_0.64O_2.56 decrease, indicating the partial amorphization or grain size refinement of the particle structure [43,44]. Notably, the main active phases (CaTiO₃ and Ca_0.96Mn_0.32Si_0.64O_2.56) maintain their structural integrity despite phase segregation, confirming the resistance of CaMn_0.625Ti_0.125Fe_0.125Mg_0.125O₃ to bulk degradation. The absence of Fe- or Mg-containing secondary phases indicates that the dopants are effectively stabilized within the perovskite lattice, preventing cation migration. The OCs demonstrate robust performance in BCLC systems, where ash-mediated surface reactions primarily govern long-term phase evolution without compromising redox functionality. Notably, while no sulfide crystalline phases are detected in any samples, sulfur dispersion on particle surfaces is identified via EDS analysis. This apparent discrepancy can be attributed to the limited number of experimental cycles, resulting in the insufficient sulfur-induced degradation of the OC to generate crystalline reaction products with a detectable signal intensity.

4. Conclusions

In this paper, a perovskite-type oxygen carrier, CaMn_0.625Ti_0.125Fe_0.125Mg_0.125O₃, was synthesized via a solid-state method, and its performance in biomass chemical looping combustion (BCLC) was evaluated in a dual-stage fluidized bed reactor using PS and PSC as fuels. Performance metrics (i.e., combustion efficiency and CO₂ yield) and the morphological/mineralogical evolution of the OC were systematically investigated under different combustion modes (mixed vs. separated), temperatures, and B/O ratios. The novelty of this work lies in the first evaluation of CaMn_0.625Ti_0.125Fe_0.125Mg_0.125O₃ in BCLC and in demonstrating that a two-stage configuration (decoupling biomass conversion from OC oxidation) can enhance gas–solid contact and improve overall conversion. The main conclusions of the study are as follows:

(1)

The separated combustion mode demonstrates superior performance to the mixed mode, yielding significantly higher peak CO₂ concentrations (24.04% vs. 12.81%) and combustion efficiencies (89% vs. 79%), along with substantial reductions in CO and CH₄ emissions, than the mixed combustion mode.

(2)

CaMn_0.625Ti_0.125Fe_0.125Mg_0.125O₃ markedly enhances the oxidation of gaseous products during biomass pyrolysis and char gasification. The CO₂ yields increase from 24.75% and 30.31% (without the OC) to 84.52% and 87.54% (with the OC), respectively, while CO and H₂ yields decrease significantly.

(3)

The temperature and B/O ratio significantly influence the CO₂ yield and combustion performance. High temperatures enhance oxygen release kinetics and catalytic activity, improving carbon conversion. While low B/O ratios significantly boost the combustion efficiency and suppress CO/CH₄ emissions, excessive fuel concentrations saturate active sites, inhibiting catalytic functions.

(4)

Characterization results confirm that the OC maintains overall structural robustness during BCLC, while surface evolution is mainly governed by ash/impurity deposition and redox cycling: SEM/EDS shows evident surface morphological changes and the appearance of sulfur on reacted particles; BET indicates an increased specific surface area after cycling; and XRD suggests that the main crystalline phases remain identifiable with limited phase evolution under the tested cycles.

In summary, the perovskite-type CaMn_0.625Ti_0.125Fe_0.125Mg_0.125O₃ OC demonstrates excellent oxygen release capacity and catalytic activity in BCLC, particularly under separated combustion conditions. This significantly enhances the combustion efficiency and CO₂ yield while reducing the emission of hazardous gases.