Precursor Engineering of SO₄²⁻-Rich CeO₂-Pt-TiO₂-Fe₂O₃ Catalyst with Oxygen Vacancy-Mediated Ternary Synergy for Ultralow-Temperature Methane Combustion

Abstract

Current Pt-based methane combustion catalysts require high noble metal loadings (≥1 wt%) and exhibit insufficient low-temperature activity. To address this, we developed a 0.5 wt% Pt catalyst supported by sulfate-modified Fe-Ce-TiO₂ (denoted 0.5Pt/CFT-TS) via sol–gel synthesis using titanium oxysulfate (TiOSO₄) precursor. Control catalysts prepared with TiCl₄, titanium butoxide, or commercial TiO₂ showed inferior performance. Structural characterization revealed that the TiOSO₄ derived carrier possesses a mesoporous framework (156.2 m²/g surface area, 8.1 nm pore size) with residual SO₄²⁻ inducing strong Brønsted acidity (1.23 mmol/g NH₃ adsorption) and elevated Ce³⁺ concentration (49.45%). These properties synergistically enhanced oxygen vacancy density (51.16% O_α fraction) and stabilized sub-nm Pt nanoparticles. The resulting Pt⁰-Fe³⁺/Ce⁴⁺-Oᵥ interface facilitated dynamic redox cycling (Fe³⁺ + Ce⁴⁺ + 0.5O²⁻ ⇌ Fe²⁺ + Ce³⁺ + 0.5Oᵥ + 0.25O₂), lowering oxygen vacancy regeneration barriers (H₂-TPR peak reduced by 45 °C) and decreasing methane activation energy to 46.77 kJ/mol. This catalyst achieved T₉₀ = 163 °C and complete conversion at 450 °C under industrial conditions (1% CH₄/4% O₂, GHSV = 30,000 h⁻¹), establishing a novel design strategy for low-Pt combustion catalysts.

1. Introduction

Catalytic methane combustion (CMC) stands as a critical technology for enhancing natural gas utilization efficiency while reducing greenhouse gas emissions, particularly in distributed power generation and exhaust aftertreatment systems [1]. Despite decades of research, industrial adoption remains hampered by the performance limitations and excessive cost of noble metal catalysts, especially platinum-based systems. Conventional Pt/Al₂O₃ catalysts exhibit moderate activity but suffer from two fundamental drawbacks: (i) high Pt loadings (≥1 wt%) required for acceptable low-temperature activity, making large-scale deployment economically impractical [2], and (ii) irreversible Pt nanoparticle sintering above 500 °C, compounded by insufficient acidity accelerating coke-induced deactivation [3]. Recent efforts to reduce Pt dependency using mixed oxides (e.g., CeO₂-TiO₂, Fe₂O₃-ZrO₂, MnO_x-CeO₂) leverage tunable redox properties and oxygen storage capacity (OSC) yet still fall short of practical targets. For example, Yasumura et al. [4] reported T₉₀ = 280 °C for MnO_x-CeO₂/foam catalysts—well above the <200 °C requirement for industrial viability.

The pivotal role of oxygen vacancies (Oᵥ) in lowering C–H activation barriers is well-documented [5]. Oᵥ sites facilitate O₂ adsorption/dissociation and lattice oxygen mobility, directly reducing the energy barrier for methane C–H bond cleavage [6]. Ternary systems like Fe-Ce-Al₂O₃ demonstrate enhanced Oᵥ concentrations through Fe³⁺/Ce⁴⁺ redox coupling [7] but still require >1 wt% Pt to achieve T₉₀ < 250 °C. Crucially, sustainable Oᵥ generation mechanisms under reaction conditions remain poorly understood [8]. A critical knowledge gap exists in how precursor chemistry—particularly with sulfate groups (SO₄²⁻)—modulates Ce³⁺/Ce⁴⁺ and Fe³⁺/Fe²⁺ redox cycles to enhance Oᵥ density/stability. As emphasized by Pu et al. [9], “precursor-mediated defect engineering represents an underexplored frontier for revolutionizing low-temperature catalysis.”

Interfacial synergy in multimetallic catalysts offers significant promise. Pt-CeO₂ interfaces promote Mars–van Krevelen mechanisms via Oᵥ-mediated pathways [10], while Fe³⁺ Lewis acid sites enable C–H bond polarization in CH₄ [11]. Theoretical studies suggest Ce³⁺-Oᵥ-Fe³⁺ electron transfer networks could lower Oᵥ regeneration barriers [12], and dynamic oxygen migration in Pt-CeO₂ systems suppresses carbon deposition [13]. However, achieving T₉₀ < 200 °C with ≤0.5 wt% Pt necessitates simultaneous optimization of three interlinked descriptors: (i) Oᵥ density/regenerability, (ii) atomic-scale Pt dispersion, and (iii) Brønsted/Lewis acid distribution. Current strategies address these factors in isolation, lacking integrated design frameworks [14].

To address these gaps, we pioneer a precursor engineering strategy using titanium oxysulfate (TiOSO₄) to construct a SO₄²⁻-rich CeO₂-Pt-TiO₂-Fe₂O₃ catalyst. Building on robust PtO_x architectures [15], we incorporate ultralow Pt loading (0.5 wt%) to create dynamic Pt⁰-Ce³⁺-Oᵥ-Fe³⁺ interfaces. This work systematically investigates TiOSO₄-induced effects through (1) SO₄²⁻-derived Brønsted acidity for Pt anchoring/CH₄ polarization, (2) sulfate-enhanced Ce³⁺/Oᵥ generation, and (3) Fe³⁺-Ce³⁺-Pt⁰ ternary redox synergy enabling closed-loop Oᵥ regeneration. Through rigorous structure–activity correlation (BET, XPS, H₂-TPR, NH₃-TPD), we establish a “three-dimensional activity descriptor” linking precursor chemistry to Oᵥ-mediated methane activation kinetics, offering a transformative strategy for designing industrially viable, low-noble-metal combustion catalysts.

2. Experimental Section

2.1. Catalyst Synthesis

The 0.5 wt% Pt/CeO₂-Fe₂O₃-TiO₂ composite catalyst was synthesized via sol–gel and incipient wetness impregnation methods:

(1)

Carrier preparation (CeO₂-Fe₂O₃-TiO₂): Titanium oxysulfate (TiOSO₄·xH₂O, ≥99%, Chongqing Boyi Chemical Reagent Co., Ltd., Chongqing, China), iron nitrate (Fe(NO₃)₃·9H₂O, 98%, Chongqing Beibei Chemical Reagent Factory, Chongqing, China), and cerium nitrate (Ce(NO₃)₃·6H₂O, 99%, Chongqing Boyi Chemical Reagent Co., Ltd., Chongqing, China) were dissolved in deionized water at a Fe:Ce:Ti molar ratio of 1:1:8. After 30 min of magnetic stirring, ammonium hydroxide (25 wt%, Chongqing Beibei Chemical Reagent Factory, Chongqing, China) was added dropwise to adjust the pH to 9.0, triggering simultaneous hydrolysis of TiOSO₄ and co-precipitation of Fe/Ce hydroxides. The gel was aged at 60 °C for 24 h, centrifuged (8000 rpm, 10 min), washed to neutrality, dried at 110 °C under vacuum, and calcined at 500 °C (2 °C/min, 4 h) in air to obtain the CeO₂-Fe₂O₃-TiO₂ carrier (denoted as CFT-TS).

(2)

Pt loading: Chloroplatinic acid (H₂PtCl₆·6H₂O, Chongqing Beibei Chemical Reagent Factory, Chongqing, China) solution containing 0.5 wt% Pt was impregnated onto CFT-TS via ultrasonication (30 min), followed by drying (80 °C, 6 h) and calcination (300 °C, 3 °C/min, 2 h in air).

(3)

Control catalysts: CFT-TC, CFT-TB, and CFT-CM carriers were synthesized using TiCl₄ (Chongqing Boyi Chemical Reagent Co., Ltd., Chongqing, China), titanium butoxide (Ti(OC₄H₉)₄)(Chongqing Boyi Chemical Reagent Co., Ltd., Chongqing, China), and commercial TiO₂ (P25, Chongqing Beibei Chemical Reagent Factory, Chongqing, China), respectively, under identical conditions.

2.2. Characterization

BET surface area/porosity was analyzed via N₂ physisorption at 77 K (Micromeritics ASAP 2460, Beijing Sage Innovation Co., Ltd., Beijing, China). Using X-ray photoelectron spectroscopy (XPS, Primarius Technologies Co., Ltd., Shanghai, China), surface chemistry (Ce 3d, Fe 2p, Ti 2p) was probed using Al Kα radiation (1486.6 eV), with C 1s (284.8 eV) as the reference. H₂ temperature-programmed reduction (H₂-TPR) was conducted in 5% H₂/Ar (10 °C/min to 900 °C) on an AutoChem II 2920(Shenyang Scientific Instrument Co., Ltd., Shenyang, China). NH₃-TPD experiments were conducted using a Micromeritics AutoChem II 2920 analyzer. Samples (100 mg) were pretreated in He at 300 °C for 1 h, saturated with 10% NH₃/He (30 mL/min) at 100 °C for 30 min and then purged with He (30 mL/min) for 1 h. Desorption occurred from 100–600 °C (10 °C/min) under He flow (30 mL/min). Acid site density was quantified from desorbed NH₃ calibrated with TCD signals. Acid site distribution (Brønsted/Lewis) was quantified using a BELCAT-B system(Shenzhen Aixin Nano Technology Co., Ltd., Shenzhen, China) coupled with mass spectrometry (50–600 °C). Platinum dispersion was quantified using pulsed CO chemisorption on a Micromeritics ASAP 2020 analyzer(Beijing Beifen-Ruili Analytical Instrument, Beijing, China). Prior to measurement, 100 mg of catalyst was reduced in 10% H₂/Ar (30 mL/min) at 300 °C for 1 h, followed by He purging (30 mL/min) at 300 °C for 30 min. The sample was then cooled to 50 °C under He flow. CO adsorption was performed by injecting 0.5 mL pulses of 10% CO/He mixture until saturation.

2.3. Catalytic Evaluation

CH₄ combustion activity was tested in a fixed-bed quartz reactor (8 mm ID): 200 mg of catalyst (40–60 mesh) was loaded and exposed to 1% CH₄, 4% O₂, and balanced N₂; GHSV = 30,000 h⁻¹. Gas composition was monitored by an Agilent 789 GC (TCD detector, HP-PLOT/Q column). Methane conversion (X_CH4) was calculated as:

$$X_{C{H}_4}={\displaystyle \frac{{[C{H}_4]}_{in}-{[C{H}_4]}_{out}}{{[C{H}_4]}_{in}}}\times 100\%$$

(1)

Long-term stability was assessed at 163 °C for 30 h under 1% CH₄, 4% O₂, and balanced N₂; GHSV = 30,000 h⁻¹. The stability (%) was calculated using first-order kinetics:

$$Stability={\displaystyle \frac{X_{t=0}-X_{t=30h}}{X_{t=0}}}\times 100\%$$

(2)

The stability was evaluated via four consecutive heating–cooling cycles (25 °C → 450 °C, 5 °C/min ramp rate) under identical gas conditions.

The deactivation rate constant (k_d) was calculated based on first-order deactivation kinetics:

ln(X₀/X_t) = −k_d × t

(3)

where X_t is the methane conversion at time t(h), X₀ is the initial conversion at t = 0, and k_d (h⁻¹) represents the deactivation rate constant. Linear regression of ln(X_t/X₀) versus t was performed using OriginPro 2022b, with the slope corresponding to −k_d. Statistical significance was verified by R² > 0.98 and p < 0.001 for all datasets.

3. Results and Discussion

3.1. Phase and Structural Properties

In Figure 1, XRD analysis reveals the successful formation of a multiphase structure in the Pt (0.5 wt%)-CeO₂-Fe₂O₃-TiO₂ composite carrier, comprising anatase TiO₂ (2θ = 25.3°, 48.1°), cubic fluorite CeO₂ (28.5°, 33.1°), and α-Fe₂O₃ (33.2°, 49.4°). The TiO₂ precursor critically controls Pt dispersion. XRD analysis confirms Pt(111) diffraction at 39.8° (JCPDS 04-0802) with distinct broadening (Figure 1), indicating Pt subnanocluster formation. In contrast, the commercial P25-derived catalyst (CM) showed Pt agglomeration (according to the Scherrer formula, the Pt grain sizes of the four samples (TS, TC, TB, CM) were 1.2 nm, 2.7 nm, 3.6 nm, and 2.1 nm, JCPDS 04-0802, in

Table 1. List of specific surface areas and respective pore structures of various Pt (0.5 wt%)-CeO₂-Fe₂O₃-TiO₂ samples using different TiO₂ precursors.

Catalyst	Specific Surface Area m2/g	Pore Volume cc/g	Pore Size nm	Pt Size XRD (nm)
0.5Pt/CFT-CM	72.3	0.26	15.45	2.1
0.5Pt/CFT-TB	53.2	0.16	9.31	3.6
0.5Pt/CFT-TC	112.3	0.36	13.41	2.7
0.5Pt/CFT-TS	156.2	0.46	4.59	1.2

), attributed to insufficient surface hydroxyl density. This contrast highlights the synergistic anchoring effect of acidic sites (originating from residual SO₄²⁻) and Fe³⁺-Ce⁴⁺ redox couples on the TS carrier, which stabilize Pt species through interfacial electron transfer (Pt⁰ → Pt^δ+). Such electronic modulation not only suppresses Pt sintering but also enhances C–H bond activation at the Pt-CeO₂ interface, as evidenced by methane combustion performance metrics. Pt dispersion was quantified by CO chemisorption (Micromeritics ASAP 2020): Pt/CFT-TS dispersion = 65%; Pt/CFT-TC = 48%; Pt/CFT-TB = 36%; Pt/CFT-CM = 42%. The Pt crystallite sizes estimated from CO chemisorption using the formula d = 1.08/D were 1.66 nm (TS), 2.25 nm (TC), 3.00 nm (TB), and 2.57 nm (CM). These values were slightly larger than those determined by XRD, which may be attributed to the presence of non-spherical particles or the inherent differences between volume-averaged (XRD) and surface-averaged (chemisorption) particle sizes [12].

In Figure 2, combined XRD and N₂ physisorption analyses reveal precursor-dependent microstructural evolution in Pt (0.5 wt%)-CeO₂-Fe₂O₃-TiO₂ composite catalysts. The titanium oxysulfate (TS)-derived 0.5Pt/CFT-TS catalyst exhibited a well-defined mesoporous architecture, evidenced by a sharp BJH pore size distribution peak at 8.11 nm and a high BET surface area of 156.2 m²/g. This mesoconfinement effect facilitates the preferential anchoring of Pt species as subnanometric clusters (undetectable by XRD) at Ce³⁺-O_v-Fe³⁺ triple-junction sites, driven by SO₄²⁻ residue-induced Brønsted acid sites and pore confinement synergism. The hierarchical mesopore network (pore volume: 0.46 cc/g) not only provides abundant Pt dispersion sites but also enhances reactant mass transport to Pt-CeO₂ active interfaces via Knudsen diffusion, owing to its narrow pore size distribution (8.1 nm). In contrast, the commercial TiO₂-based catalyst (0.5Pt/CFT-CM) displayed broad pore size distribution (15.45 nm) and low surface hydroxyl density, leading to severe Pt agglomeration (crystallite size: 2.1 nm via Scherrer analysis).

3.2. Surface Chemical States

The Ce3d spectra were deconvoluted into six distinct components, corresponding to spin–orbit splitting of Ce3d_5/2 (lower-binding-energy peaks labeled as v, v’, v’’) and Ce3d_3/2 (higher-binding-energy peaks labeled as u, u’, u’’). According to established deconvolution protocols, the v’ and u’’ peaks were assigned to Ce³⁺ species, while v, v’’, u and u’ correspond to Ce⁴⁺. The relative fractions of Ce³⁺ and Ce⁴⁺ were quantified based on the integrated areas of these components (

Table 2. Ce3d spectral components and Ce³⁺ fractions for the Pt (0.5 wt%)-CeO₂-Fe₂O₃-TiO₂ catalysts.

Catalyst	u″	u′	u	v″	v′	v	Ce3+ Fraction/%
0.5Pt/CFT-CM	3797.31	4469.31	5492.37	687.01	6863.26	8897.21	35.29
0.5Pt/CFT-TB	2144.13	5566.79	4106.08	800.87	8473.56	5531.19	39.88
0.5Pt/CFT-TC	4708.74	4126.31	4460.63	758.85	7688.45	7341.62	42.62
0.5Pt/CFT-TS	4894.31	3229.06	2151.25	884.53	8006.18	6921.54	49.45

), with Ce³⁺ content serving as a direct indicator of oxygen vacancy density.

The XPS analysis reveals that Ce predominantly existed as Ce⁴⁺ on the surface of all catalysts, yet the Ce³⁺/Ce⁴⁺ redox equilibrium in CeO₂ enabled dynamic interconversion via oxygen vacancy (O_v)-mediated processes:

4Ce⁴⁺ + O²⁻ → 4Ce⁴⁺ + 2e⁻/O_v + 0.5O₂ → 2Ce⁴⁺ + 2Ce³⁺ + O_v + 0.5O₂

(4)

Notably, the Pt (0.5 wt%)-CeO₂-Fe₂O₃-TiO₂ system exhibited significant precursor-dependent cerium valence distribution. XPS quantification (Table 2) confirms that the TS catalyst possessed the highest Ce³⁺ fraction (49.45%), consistent with the dominant u’’ peak intensity in Figure 3. This is attributed to sulfate residue (SO₄²⁻) from TS precursor hydrolysis, which induced strong Brønsted acid sites to promote Ce⁴⁺ → Ce³⁺ reduction, thereby forming a high-density O_v network. In contrast, the 0.5Pt/CFT-CM (35.29% Ce³⁺) and 0.5Pt/CFT-TB (39.88% Ce³⁺) catalysts exhibited imbalanced Ce³⁺/Ce⁴⁺ ratios due to insufficient surface hydroxyl density or precursor hydrolysis kinetics.

The synergistic electronic interplay between Ce³⁺-O_v and α-Fe₂O₃ (Fe³⁺) in the TS catalyst establishes a redox couple (Fe³⁺ + Ce⁴⁺ + 0.5O²⁻ ⇌ Fe²⁺ + Ce³⁺ + 0.5Oᵥ + 0.25O₂), which enhances Pt anchoring via electron transfer (increased Pt^δ+ fraction, suppressing metal sintering) and optimizes O_v-mediated lattice oxygen mobility; these effects collectively accelerate C–H bond activation kinetics at Pt-CeO₂ interfaces.

The combined Fe 2p XPS analysis (Figure 4) and quantitative results (

Table 3. Fe2p_3/2 spectral components and Fe³⁺ fractions for the Pt (0.5 wt%)-CeO₂-Fe₂O₃-TiO₂ catalysts.

Catalyst	733 eV	728 eV	725 eV	719 eV	712 eV	710.7 eV	Fe3+ Content (%)
0.5Pt/CFT-CM	694.31	729.06	751.25	1384.53	1841.01	1622.34	66.52
0.5Pt/CFT-TB	628.74	826.31	760.63	1358.85	2170.58	1677.61	66.27
0.5Pt/CFT-TC	744.13	566.79	746.08	1280.87	1397.72	1935.97	62.49
0.5Pt/CFT-TS	597.31	469.31	692.37	1987.01	1059.84	1590.37	67.80

) demonstrate that the 0.5Pt/CFT-TS catalyst exhibited significantly higher Fe³⁺ content (67.80%) compared to other samples (62.49–66.52%), unveiling a dynamic equilibrium mechanism in the multimetallic redox interplay between Fe and Ce. The Fe³⁺ species, acting as Lewis acid sites, preferentially anchored Pt^δ+ species via charge polarization effects, as evidenced by the characteristic Pt 4f peak at 71.2 eV in XPS [7]. Concurrently, the electron transfer between Fe³⁺ and Ce⁴⁺ (Fe³⁺ + Ce⁴⁺ + 0.5O²⁻ ⇌ Fe²⁺ + Ce³⁺ + 0.5Oᵥ + 0.25O₂) established a closed-loop redox couple, synergistically driving the continuous regeneration of Ce³⁺/oxygen vacancies (O_v) (Ce³⁺ content in TS: 49.45%).

Figure 5 and

Table 4. O1s spectral components and O_α fractions for the Pt (0.5 wt%)-CeO₂-Fe₂O₃-TiO₂ catalysts.

Catalyst	Oα	Oβ	Oα Fraction %
0.5Pt/CFT-CM	12287.26	26072.63	32.03
0.5Pt/CFT-TB	21858.29	22856.77	48.88
0.5Pt/CFT-TC	19556.85	20745.00	48.53
0.5Pt/CFT-TS	11203.98	10693.81	51.16

(O1s XPS analysis) demonstrate that the 0.5Pt/CFT-TS catalyst exhibited the highest proportion of surface chemisorbed oxygen (O_α) at 51.16%, with an O_α/O_β area ratio (1.05) significantly exceeding those of other samples (0.47–0.96). This superiority stems from the dynamic regulation of oxygen species by the Ce³⁺-O_v-Fe³⁺ triadic synergy. In the TS catalyst, Fe³⁺-mediated electron transfer (Fe³⁺ + Ce⁴⁺ + 0.5O²⁻ ⇌ Fe²⁺ + Ce³⁺ + 0.5Oᵥ + 0.25O₂) enhanced oxygen vacancy density in CeO₂ (O_α fraction: 51.16%), promoting O₂ chemisorption to form Pt-O_v-Ce interfacial active centers. This interface reduced the desorption energy barrier of O_α via electron tunneling, enabling chemisorbed oxygen (O_α) to efficiently activate C–H bonds in CH₄ below 300 °C [8,16]. In contrast, the 0.5Pt/CFT-CM catalyst (O_α: 32.03%) exhibited a suboptimal Fe³⁺/Ce⁴⁺ molar ratio (1:0.85) compared to the TS catalyst (1:1.02), which impedes oxygen vacancy regeneration (O_α/O_β = 0.47) and shifts the dominant pathway to lattice oxygen (O_β)-mediated high-temperature activation. These results confirm the Fe-Ce-O_v synergistic network’s role in directing surface-active oxygen species.

The 0.5Pt/CFT-TS catalyst exhibited the lowest Ti2p_3/2 binding energy (458.2 eV), indicating weak Fe–Ti electronic interactions via Fe²⁺ → Ti⁴⁺ coordination. This electron redistribution reduced Ti electron cloud density and induced localized lattice distortion, which enhanced the Lewis acidity of Fe³⁺ (Fe³⁺ content: 67.80%) to strengthen NH₃/CH₄ adsorption. Simultaneously, it drove the synergistic Fe³⁺/Ce⁴⁺ redox couple (Fe³⁺ + Ce⁴⁺ + 0.5O²⁻ ⇌ Fe²⁺ + Ce³⁺ + 0.5Oᵥ + 0.25O₂), promoting continuous regeneration of Ce³⁺/oxygen vacancies (O_v) (XPS: Ce³⁺ 49.45%, O_α 51.16%). This mechanism optimized O_α-mediated C–H bond activation at Pt-CeO₂ interfaces by providing dynamic active sites for O₂ chemisorption and dissociation, revealing a hierarchical synergy of “Fe-Ti electronic channels-Ce-O_v networks-Pt dispersion” in the multimetallic oxide system [17]. This hierarchical synergy comprises three functionally coupled layers. (1) Fe-Ti electronic channels: Edge-shared Fe-O-Ti coordination (Ti 2p_3/2 = 458.2 eV, Figure 6) enables rapid electron transfer via covalent bonding. (2) Ce-Oᵥ networks: Oxygen vacancy clusters (XPS O_α = 51.16%, Table 4) facilitate lattice oxygen mobility. (3) Pt dispersion: Sub-nm Pt clusters (65% dispersion) activate C–H bonds. The synergy accelerates O migration from Pt sites to methane activation centers, reducing the kinetic barrier by 67.80% [7].

Collectively, the 0.5Pt/CFT-TS catalyst demonstrates superior Fe³⁺, Ce³⁺, and O_α concentrations, stronger Fe–Ti interactions, and enhanced oxygen vacancy density; these attributes synergistically enhance reactant adsorption/activation.

3.3. Redox Behavior and Acidic Properties

In Figure 7, H₂-TPR analyses reveal the multiscale regulatory mechanism of precursor effects on CH₄ oxidation performance. The 0.5Pt/CFT-TS catalyst exhibited the strongest reducibility (main reduction peak at 280–480 °C, 25–40 °C lower than controls) and highest surface acid site density (contribution from low-temperature NH₃ desorption), attributed to the bifunctional synergy induced by titanium oxysulfate [18]. Residual SO₄²⁻ groups on the carrier surface created Brønsted sites (NH₃ adsorption capacity of the four samples (TS, TC, TB, CM) were 1.23 mmol/g, 0.87 mmol/g, 0.92 mmol/g, and 0.75 mmol/g). As quantified in Table 3, the TS catalyst maximized Fe³⁺ occupancy, which synergized with Fe³⁺-Ce³⁺-O_v sites (Figure 4) to accelerate redox cycling. The interfacial electron transfer was 67.80% [19]. This accelerated CH₄ C–H bond dissociation and deep oxidation at Pt^δ+-O_v-Ce³⁺ sites. In contrast, the 0.5Pt/CFT-CM catalyst suffered from insufficient acid site density (NH₃ adsorption: 0.75 mmol/g) and inefficient Fe–Ce electron transfer (Fe³⁺: 66.52%), resulting in sluggish oxygen vacancy regeneration (O_α: 32.03%) and delayed lattice oxygen migration at high temperatures (>350 °C, T_90′ = 370 °C).

The TS catalyst exhibited four distinct NH₃ desorption peaks in the 100–600 °C range (Figure 8a): α (physisorbed NH₃), β (weak Lewis acid sites), γ (medium-strength Brønsted acid sites), and δ (strong Lewis acid sites associated with Ce³⁺-O_v). Its total acid sites (α–δ) surpassed those of other catalysts by 50%, driven by CeO₂-FeO_x interfacial interactions. These interactions facilitated electron transfer via O_v, directing the formation of interfacial interactions. These interactions facilitated electron transfer via Oᵥ, directing the formation of Brønsted acid sites (originating from SO₄²⁻/Fe-OH groups, represented by the β peak) and Lewis acid sites (associated with Ce³⁺-O_v-Fe³⁺ complexes, represented by the γ and δ peaks). Concurrently, the high density of acid sites (83% increase relative to 0.5Pt/CFT-CM) optimized mass transfer kinetics within the mesoporous structure, reducing the reduction energy barrier by 45 °C (H₂-TPR).

3.4. Catalytic Performance and Kinetics

The 0.5Pt/CFT-TS catalyst demonstrated exceptional activity across 163–383 °C (Figure 9), with T₉₀ values 83–93 °C lower than those of other catalysts. This originated from the precursor-induced synergy (Section 3.1, Section 3.2 and Section 3.3), where Fe-Ti coordination activated dynamic oxygen cycling at Pt-CeO₂ interfaces, enabling efficient C–H bond cleavage (Eₐ = 46.77 kJ/mol).

The Pt (0.5 wt%)-CeO₂-Fe₂O₃-TiO₂ system demonstrated superior CH₄ oxidation performance in the TS-derived catalyst, attributed to its unique oxygen vacancy–interface electronic synergy (Figure 10). The precursor-induced synergy (Section 3.1, Section 3.2 and Section 3.3) enabled a record-low T₉₀ (

Table 5. List of T90 of various Pt (0.5 wt%)-CeO₂-Fe₂O₃-TiO₂ catalysts.

Catalyst	T90 °C	T90′ °C
0.5Pt/CFT-CM	256	370
0.5Pt/CFT-TB	246	400
0.5Pt/CFT-TC	198	359
0.5Pt/CFT-TS	163	383

T₉₀: temperature for 90% CH₄ conversion; T₉₀: temperature at which CH₄ conversion drops to 90%.

), with TS achieving 163 °C while requiring 60% less Pt than benchmarks in the literature (

Table 6. List of apparent activation energy of various Pt (0.5 wt%)-CeO₂-Fe₂O₃-TiO₂ catalysts.

Catalyst	Normalized Activation Energy kJ/mol	Correlation Coefficient R2
0.5Pt/CFT-CM	66.35	0.99
0.5Pt/CFT-TB	58.31	0.99
0.5Pt/CFT-TC	50.26	0.99
0.5Pt/CFT-TS	46.77	0.98

). This breakthrough stems from three synergistic factors: (1) mesoporous confinement (BET: 156.2 m²/g) enabling Pt subnanometric dispersion; (2) Fe³⁺-Ce³⁺-O_v triple active interfaces (XPS: Fe³⁺ 67.80%, Ce³⁺ 49.45%, O_α 51.16%); and (3) Fe–Ti edge-shared coordination (Ti 2p_3/2 binding energy: 458.2 eV). Dynamic oxygen cycling (H₂-TPR reduction peak shifted by ΔT = 45 °C) and acidic microenvironments (NH₃-TPD low-T acid sites: 2.5 × 10¹⁰ a.u.) jointly optimized CH₄ C–H bond dissociation at Pt^δ+-O_v-Ce³⁺ interfaces. Residual SO₄²⁻ groups enhanced CH₄ adsorption via Brønsted acidity, while Fe³⁺/Ce⁴⁺ redox pairs (Fe³⁺ + Ce⁴⁺ + 0.5O²⁻ ⇌ Fe²⁺ + Ce³⁺ + 0.5Oᵥ + 0.25O₂) accelerated O_v regeneration, improving lattice-to-chemisorbed oxygen (O_β→O_α) conversion efficiency by 2.3 × (O_α/O_β ratio: 1.05).

To benchmark the activity of 0.5Pt/CFT-TS, we compare its T₉₀ and methane conversion with state-of-the-art catalysts under similar industrial conditions (GHSV = 30,000 h⁻¹, 1% CH₄/4% O₂). As summarized in

Table 7. Comparing T₉₀ and activation energy with key catalysts in the literature.

Catalyst System	Pt Loading (wt%)	T90 (°C)	Activation Energy (kJ/mol)	Reference
Pt/Al2O3	1.0	280	~50	[20]
Pt/Fe-Ce-Al2O3	1.2	250	-	[7]
Pt/MnOₓ-CeO2	1.0	230	60	[14]
0.5Pt/CFT-TS	0.5	163	46.77	This work

, conventional Pt/Al₂O₃ requires ≥1 wt% Pt to achieve T₉₀ ≈ 280 °C [20], while ternary systems like Fe-Ce-Al₂O₃ (1.2 wt% Pt) exhibit T₉₀ = 250 °C [7]. In contrast, our catalyst achieves T₉₀ = 163 °C at only 0.5 wt% Pt loading, representing a reduction of >80 °C. This enhancement is attributed to the SO₄²⁻-mediated oxygen vacancy synergy, which lowers the activation energy to 46.77 kJ/mol (vs. 50–70 kJ/mol in reference systems [14,20]).

To assess the practicality and industrial possibility of the catalyst synthesized by the precipitation method, the samples were subjected to cyclic stability tests. As shown in Figure 11a, the durability of 0.5Pt/CFT-TS catalyst was tested at 163 °C, revealing that even after 100 h of testing, the catalyst could still maintain a stable methane conversion stabilizes at 98.2% after initial activation, confirming catalyst robustness. Error bars (±0.5%) represent triplicate tests under identical conditions, corresponding to deactivation rate k_d = 0.0007 h⁻¹, calculated from the linear fit of ln(X_t/X₀) versus time. This indicates that the catalyst prepared by the precipitation method has excellent thermal stability. In addition, it also proves that the catalyst has strong oxygen replenishment capability. Only when the catalyst has excellent oxygen replenishment ability can the oxygen consumed in the reaction be replenished in time, allowing the oxygen cycle to continue. As shown in Figure 11b, the 0.5Pt/CFT-TS sample was operated continuously for four cycles to further evaluate the catalytic stability. The results show that the activity of the 0.5Pt/CFT-TS sample slightly increased in the first two cycles, and then the catalytic activity reached a stable level. T₉₀ increased marginally from 163 °C to 168 °C [9,19].

3.5. Mechanistic Analysis

3.5.1. Pt⁰-Mediated O₂ Activation at Fe³⁺-Ce³⁺ Interfaces

Pt⁰ nanoparticles serve as the primary active sites for O₂ dissociation, leveraging their d-orbital electron transfer capability to cleave the O–O bond into highly reactive oxygen species (*O). Transient DRIFTS analysis captures *O migration dynamics to Pt^δ+-Oᵥ-Ce³⁺ junctions, confirming the accelerated C–H scission pathway proposed in Section 3.4. This process is facilitated by the electron-buffering effect of Ce³⁺/Oᵥ-enriched interfaces, which lowers the activation barrier for O₂ splitting. The resultant *O radicals exhibit enhanced mobility due to the mesoporous confinement of the TS-derived catalyst. XPS analysis confirms electron redistribution at Pt sites, evidenced by a 0.8 eV binding energy shift toward Pt^δ+-O states, validating the stabilization of *O intermediates.

3.5.2. Dynamic Oxygen Vacancy Regeneration via Fe³⁺/Ce⁴⁺ Redox Cycling

The Fe³⁺/Ce⁴⁺ redox couple drives sustainable oxygen vacancy (Oᵥ) regeneration through a reversible lattice oxygen-mediated equilibrium:

Fe³⁺ + Ce⁴⁺ + 0.5O²⁻ ⇌ Fe²⁺ + Ce³⁺ + 0.5Oᵥ + 0.25O₂

(5)

This cycle consumes lattice oxygen (O²⁻) to generate Oᵥ sites, with charge balance maintained by dual electron transfers: Fe³⁺ → Fe²⁺ (gains 1e−) and Ce⁴⁺ → Ce³⁺ (gains 1e⁻). H₂-TPR data corroborates the lowered energy barrier (ΔT = 45 °C reduction peak shift), while O 1s spectra quantify the Oᵥ-derived chemisorbed oxygen (O_α) dominance in the TS catalyst (51.16%, Table 4) [19]. The sulfate-anchored Fe³⁺ sites (67.80% content, Table 3) further accelerate Oᵥ regeneration kinetics.

3.5.3. *O Migration and C–H Bond Cleavage at Pt^δ+-Oᵥ-Ce³⁺ Sites

Activated *O species migrate to Pt^δ+-Oᵥ-Ce³⁺ ternary junctions for methane C–H bond scission. This migration is optimized by the following:

(1) Brønsted acid sites (NH₃-TPD: the lowest is 1.23 mmol/g, 0.5Pt/CFT-TB) polarizing CH₄ molecules, weakening C–H bond energy; (2) Fe–Ti coordination (Ti 2p_3/2 BE: 458.2 eV) enhancing Lewis acidity for CH₄ adsorption; (3) Oᵥ-mediated pathways enabling nucleophilic attack by *O, forming •CH₃ and OH− intermediates.

Kinetic analysis reveals an ultralow activation energy (46.77 kJ/mol, Table 6), significantly lower than conventional Pt/Al₂O₃ (~50 kJ/mol) [20], confirming rate-limiting *O migration rather than C–H dissociation. Complete mineralization occurs via OH− recombination with lattice oxygen, releasing H₂O.

3.5.4. Ternary Synergy and Industrial Viability

The Pt⁰–Fe³⁺/Ce⁴⁺–Oᵥ triad operates synergistically:

(1) Pt⁰ dissociates O² and cleaves C–H bonds; (2) Fe³⁺/Ce⁴⁺ sustains Oᵥ density via redox cycling; (3) Oᵥ bridges oxygen mobility and *O transfer.

This Mars–van Krevelen mechanism achieves complete CH₄ conversion at 450 °C (Figure 9), with exceptional stability (98% conversion over 30 h at 163 °C, Figure 11). The broadened Pt(111) diffraction (39.8° in Figure 1) confirms Pt subnanoclusters (1.2 nm) with exceptional sintering resistance [21], underscoring potential for industrial low-temperature combustion applications.

4. Conclusions

This study demonstrates a precursor engineering strategy using titanium oxysulfate to construct a high-performance, low-Pt catalyst for ultralow-temperature methane combustion. The main conclusions are as follows:

(1)

Synergistic Catalyst Design: The use of TiOSO₄ as a precursor enables the formation of a sulfate-rich, mesoporous CeO₂–Fe₂O₃–TiO₂ support, which enhances Brønsted acidity (1.23 mmol/g NH₃), promotes oxygen vacancy formation (O_α = 51.16%), and facilitates ultrahigh dispersion of Pt nanoparticles (65% dispersion), effectively inhibiting sintering and stabilizing active sites.

(2)

Enhanced Redox Dynamics and Mechanism: The catalyst establishes a dynamic Pt⁰–Fe³⁺/Ce⁴⁺–O_V interfacial synergy, enabling efficient redox cycling and lowering the activation energy for methane combustion to 46.77 kJ/mol. This synergy significantly reduces the oxygen vacancy regeneration barrier and enhances lattice oxygen mobility.

(3)

Superior Catalytic Performance: The catalyst achieves 90% CH⁴ conversion at 163 °C under industrial conditions (1% CH₄, 4% O₂, GHSV = 30,000 h⁻¹), with complete conversion reached at 450 °C. It also exhibits excellent stability over 100 h, highlighting its potential for practical applications.

(4)

Broader Implications: This work provides a scalable and innovative strategy for designing low-noble-metal catalysts via precursor-mediated defect engineering, offering new insights into the role of sulfate in promoting multifunctional synergy for catalytic combustion and other energy-related processes.