Enhanced Catalytic Ozonation of Formaldehyde over MOFs- Derived MnO_x Catalysts with Diverse Morphologies: The Role of Oxygen Vacancies

Abstract

Metal–organic frameworks (MOFs) have become a hot topic in various research fields nowadays. And MOF-derived metal oxides prepared by the sacrificial template method have been widely applied as catalysts for pollutant removal. Accordingly, we prepared a series of MOF-derived MnO_x catalysts with diverse morphologies (rod-like, flower-like, slab-like) via the pyrolysis of MOF precursors, and the as-prepared MnO_x catalysts demonstrated superior performance compared to the one prepared using the co-precipitation method. MnO_x-II, with a flower-like structure, exhibited excellent activity for formaldehyde (HCHO) catalytic ozonation at room temperature, reaching complete HCHO conversion at O₃/HCHO of 1.5 and more than 90% CO₂ selectivity at an O₃/HCHO ratio of 2.5. On the basis of various characterization methods, it was clarified that the enhanced catalytic performance of MnO_x-II benefited from its larger BET surface area, abundant oxygen vacancies, better redox ability at lower temperature, and more Lewis acid sites. The H₂O resistance and stability tests were also conducted. Furthermore, DFT calculations substantiated the enhanced adsorption of HCHO and O₃ on oxygen vacancies, while in–situ DRIFTS measurements elucidated the degradation pathway of HCHO during catalytic ozonation through detected intermediates.

1. Introduction

Driven by the accelerating urbanization and industrialization trends, the emission of gaseous pollutants has increased significantly, causing great damage to air quality worldwide. Typically, volatile organic compounds (VOCs), as crucial atmospheric pollutants, consistently induce adverse health effects and air quality issues, like dust haze and photochemical smog [1]. Common in indoor environments, formaldehyde (HCHO) usually leads to respiratory problems and lung damage, and worse yet, long exposure to HCHO even causes leukemia and cancers [2]. To mitigate the adverse effect of HCHO on the atmosphere and public health, developing efficient control means has become imperative.

Catalytic ozonation is a desirable method for pollution treatment, which can mineralize VOCs into harmless products (H₂O and CO₂) without secondary pollution. With the promotion of ozone (O₃), the reaction temperature is further lowered due to its strong oxidizability and produced free radicals (·O^2–, ·O^2–, ·OH) [3]. Moreover, the selection of catalysts with superior catalytic activity is important to impel the proceeding of the catalytic ozonation reaction. To date, numerous catalysts have been employed for HCHO degradation, involving precious metals (Au, Pt, Pd, etc.) [4,5,6,7] and non-precious metals (Mn, Co, etc.) [8,9]. However, the exorbitant price and deactivation tendency of precious metals restrict their practical application. Hence, efficient and economical catalysts are urgently required for HCHO removal.

Manganese oxides (MnO_x) have been widely applied to pollutant treatment as promising materials because of their low cost, multiple valence states, and diverse shapes [10]. Rational design strategies for better catalytic activity of MnO_x have been extensively investigated, such as regulating oxygen vacancies [11], adjusting morphologies [12], and altering preparation methods [13]. Zhao et al. [14] reported that the catalytic ozonation on MnO_x (O₃/HCHO=3:1) promoted the HCHO conversion (~100%) at ambient temperature, confirming the unique ability of ozone in promoting the performance of MnO_x under low temperature. Saputra et al. [15] systematically designed the morphologies of α-Mn₂O₃ (cubic, octahedral, truncated-octahedral), and the catalyst with a cubic shape exhibited higher catalytic activity for its bigger specific surface area and abundant active surface facets. Wang et al. [12] prepared several MnO_x catalysts with distinct morphologies (rod-, wire-, tubular-, and flower-like), and rod-like α-MnO₂ achieved complete toluene oxidation at lower temperatures, relating to its plentiful oxygen vacancies and reliable redox ability. Accordingly, the morphological control of MnO_x catalysts might prove to be critical to optimize catalytic performance.

Recently, MOFs have been intensively investigated in the catalysis field for some attractive advantages, including large surface area, controllable morphology, and high porosity [16,17,18,19,20]. With plenty of electron-deficient open metal sites, MOFs have strong interaction with VOCs and can easily adsorb and degrade pollution gases, which brings a broader prospect for VOC removal [21]. Furthermore, MOFs can be pyrolyzed as sacrificial precursors at high temperature in an aerobic atmosphere to fabricate diverse porous metal oxides, which can inherit the porosity and well-defined morphology of pristine MOFs [22]. By selecting appropriate precursors and control strategies, diverse morphologies of MOF-derived nano-structure catalysts can be rationally designed. Zhang et al. [23] researched the activity of Mn₂O₃ obtained from Mn-MOFs with different precursors (MIL-100, MOF-74, BTC). The catalyst derived from the cubic Mn-MIL-100 precursor performed well, associating with its higher Mn³⁺/Mn⁴⁺ ratio and plentiful surface oxygen species. Li et al. [24] fabricated MOF-derived MnO_x-m catalysts with a nanorod-self-assembly hollow microsphere structure. Compared with other derivatives, the MnO_x-m catalysts presented superior catalytic performance due to their abundant exposed active sites, desirable redox ability, and plentiful active oxygen species. These studies illustrate that the catalyst performance is highly associated with morphological structures. However, there have been few studies on the catalytic ozonation of HCHO over MOF-derived MnO_x catalysts with different morphologies so far.

Herein, a series of MOFs-derived MnO_x catalysts with diverse structures were synthesized in this work via the pyrolysis of different MOFs precursors. The catalytic activity was evaluated through the measurement of HCHO conversion and CO₂ selectivity during the catalytic ozonation reaction at ambient temperature. Meanwhile, all kinds of characterization methods were adopted to illustrate the structure-activity relationship. Moreover, the intermediates generated during HCHO ozonation were also detected through in-situ DRIFTS analysis. Integrated with DFT calculation, the reaction mechanism was further clarified.

2. Results and Discussion

2.1. Catalytic Ozonation of HCHO

Figure 1a presents the catalytic performance of MnO_x catalysts during HCHO ozonation at room temperature by varying O₃/HCHO ratios (0.5~3.0). As shown in Equation (1), the complete mineralization of HCHO with ozone required a theoretical O₃/HCHO ratio of 2.0. Clearly, the increase in the O₃/HCHO ratio significantly improved the HCHO conversion of the catalysts. The O₃/HCHO ratio corresponding to 90% HCHO conversion (O₉₀) was the following sequence: MnO_x-II (1.1) < MnO_x-I (1.3) < MnO_x-III (1.4) < MnO_x-IV (1.8). Compared to the MOF-derived MnO_x catalysts, MnO_x-IV prepared by the co-precipitation method had undesirable catalytic activity, getting a 100% conversion at O₃/HCHO of 3.0. Meanwhile, MnO_x-II with a flower-like structure exhibited remarkably higher HCHO conversion, achieving 100% HCHO conversion when the O₃/HCHO ratio was 1.5. It was worth mentioning that almost no O₃ residual was observed during the catalytic tests.

HCHO + (2 − x)O₃ → (1 − x)CO₂ + xCO + (2 − x)O₂ + H₂O

(1)

To reduce hazardous intermediates and promote green environmental protection, environmentally friendly CO₂ and H₂O are regarded as ideal final products. Consequently, the CO₂ selectivity results of MnO_x catalysts are shown in Figure 1b. Obviously, for all the catalysts, the CO₂ selectivity had a nearly linear increase with the O₃/HCHO ratio. And on the whole range of O₃ input, MnO_x-II maintained the highest CO₂ selectivity, which attained above 90% CO₂ selectivity at an O₃/HCHO ratio of 2.5, and the order for the CO₂ selectivity was as follows: MnO_x-II > MnO_x-III > MnO_x-I > MnO_x-IV.

2.2. Catalyst Characterization

2.2.1. Crystalline Structures

The crystalline phases of the MnO_x catalysts were identified through XRD analysis (Figure 2). And the diffraction peaks at 23.1°, 32.9°, 38.2°, 45.1°, 49.3°, 55.1°, and 65.7° were ascribed to the (2 1 1), (2 2 2), (4 0 0), (3 3 2), (4 3 1), (4 4 0), and (6 2 2) planes of α-Mn₂O₃ (PDF#71-0636), while the weak peaks at 18.1°, 28.8°, and 36.6° were indexed to the (2 0 0), (3 1 0), and (4 0 0) plane of MnO₂ (PDF#44-0141) [25]. Compared with other MnO_x catalysts, MnO_x-I and MnO_x-II displayed a relatively weaker diffraction intensity, implying their smaller particle sizes, which introduced sufficient lattice defects and enhanced catalytic performance [26]. Moreover, the crystallite size was calculated by the Scherrer formula, and the crystallite sizes of MnO_x-I, MnO_x-II, MnO_x-III, and MnO_x-IV were 35.6 nm, 32.3 nm, 36.7 nm, and 37.2 nm, respectively. MnO_x-II showed the smallest crystallite size.

To explore the corresponding morphologies of MOF precursors and MOF-derived MnO_x catalysts, the images of SEM characterization are depicted in Figure 3. Clearly, MnO_x-I presented a rod-like structure composed of multiple stacked nanorods, while MnO_x-II exhibited a flower-like structure with ordered assembly nanosheets, and MnO_x-III consisted of many stacked slabs. Prepared by the co-precipitation method, MnO_x-IV showed an amorphous structure with agglomerated nanoparticles, which might result in a reduction in surface area. After calcination, the surface of MOF-derived MnO_x catalysts became rougher, and a lot of voids appeared, attributable to the thermolysis of organic ligands under thermal treatment. Fortunately, the morphologies of MOF precursors were basically retained on MOF-derived MnO_x catalysts, and the generated surface defects on porous structures accelerated both active site formation and VOCs–catalyst contact efficiency [27].

2.2.2. Textual Properties

To explore the textual properties of MnO_x catalysts, their N₂ adsorption–desorption isotherms (Type IV, H3 hysteresis loop) and BJH pore size distributions are depicted (Figure S1). The BET surface area and pore volume are arranged in the sequence of MnO_x-II > MnO_x-I > MnO_x-III > MnO_x-IV (

Table 1. Textual properties of prepared catalysts.

Catalyst	BET Surface Area /m2·g−1	Pore Volume a /cm3·g−1	Average Pore Diameter b/nm
MnOx-I	19.6	0.077	10.3
MnOx-II	20.0	0.079	10.7
MnOx-III	18.7	0.075	10.9
MnOx-IV	15.8	0.044	16.0

^a BJH desorption cumulative volume of pores. ^b BJH desorption average pore diameter.

), in accord with the HCHO conversion sequence. Higher surface area usually promotes the generation of surface active sites, the adsorption of VOCs, and the activation of ozone [28,29]. With the optimal catalytic activity, MnO_x-II possessed a higher surface area and pore volume, 20.0 m²·g⁻¹ and 0.079 cm³·g⁻¹, respectively. Moreover, MnO_x-IV exhibited the least favorable catalytic activity in HCHO degradation, with lower surface area and pore volume.

2.2.3. Surface Properties

To research the valence states of elements on the catalyst surface, the XPS curves are depicted in Figure 4, and the content of elements after deconvolution is presented in

Table 2. XPS results of synthesized catalysts.

Catalysts	Mn 2 p3/2		O 1 s		Mn3+/Mn4+	Oad/Ola
	Mn4+ (%)	Mn3+ (%)	Ola (%)	Oad (%)
MnOx-I	36.8	63.2	47.7	52.3	1.72	1.10
MnOx-II	35.1	64.9	44.9	55.1	1.85	1.23
MnOx-III	40.4	59.6	48.2	51.8	1.48	1.07
MnOx-IV	40.9	59.1	49.3	50.7	1.44	1.03

. As displayed in Figure 4a, the binding energies at 641.4 and 643.3 eV were ascribed to Mn³⁺ and Mn⁴⁺ species, respectively [30,31]. The Mn³⁺/Mn⁴⁺ ratio declined as the sequence MnO_x-II (1.85) > MnO_x-I (1.72) > MnO_x-III (1.48) > MnO_x-IV (1.44), fitting with the order of HCHO conversion. The existence of Mn³⁺ ions could promote the formation of oxygen vacancies through a charge balance mechanism [32]. Thus, a higher Mn³⁺/Mn⁴⁺ ratio usually suggests more oxygen vacancies on the catalyst surface, which can promote the decomposition of ozone and the catalytic ozonation of VOCs [33].

The O 1s spectra identified lattice oxygen species (O_la) at 529.7 eV and adsorbed oxygen species (O_ad) at 530.7 eV, respectively (Figure 4b). Moreover, the ratio of O_ad/O_la ranked in order of MnO_x-II (1.23) > MnO_x-I (1.10) > MnO_x-III (1.07) > MnO_x-IV (1.03). Compared to O_la species, O_ad species showed higher oxygen mobility, which could accelerate oxygen vacancy generation and promote the reactant activation process [34,35].

To sum up, MnO_x-II had the highest Mn³⁺ and O_ad contents, which was conducive to increased oxygen vacancies and accelerated oxygen mobility. Different from the MOF-derived MnO_x catalysts, MnO_x-IV presented a lower Mn³⁺ and O_ad proportion, fitting with its undesirable catalytic efficiency. Moreover, the results of XPS analysis aligned with the HCHO conversion order, indicating the critical role of oxygen vacancies in affecting catalytic performance.

2.2.4. Temperature Programmed Studies

A Raman spectrum study was performed to explore the oxygen vacancies in MnO_x catalysts (Figure 5a). The characteristic peaks at 630~650 cm⁻¹ corresponded to the v2(Mn−O) stretching vibration of [MnO₆] chains, while Mn−O bond strength usually correlates to oxygen vacancies formation [36]. Based on Hooke’s law, the strength of the Mn−O bond was determined by the following equation (Equation (2)):

$$\omega ={\displaystyle \frac{1}{2\pi c}}\sqrt{{\displaystyle \frac{k}{\mu }}}$$

(2)

where ω represents the stretching vibration shift, k refers to Mn−O bond strength, μ refers to effective mass and c represents the light velocity, respectively.

The ω value of catalysts decreased in the order: MnO_x-IV (650.0 cm⁻¹) > MnO_x-III (647.9 cm⁻¹) > MnO_x-I (645.8 cm⁻¹) > MnO_x-II (633.5 cm⁻¹). Performing well in catalytic activity, MnO_x-II showed the lowest ω value, which corresponded to the lowest strength. Weaker Mn−O bond strength indicates an increase in oxygen vacancy defects as well as higher oxygen mobility because of the easily stripped oxygen species, which can facilitate HCHO deep oxidation [37,38]. Hence, increasing oxygen vacancy concentration usually enhances catalytic performance, consistent with the XPS analysis results.

H₂-TPR measurement was implemented to assess the reducibility of synthesized MnO_x catalysts, with relevant profiles presented in Figure 5b. Clearly, the curves of catalysts presented two obvious reduction peaks, while the peak at 300~400 °C belonged to Mn⁴⁺ to Mn³⁺ transformation (MnO₂ → Mn₂O₃) and another one (400~500 °C) was assigned to Mn³⁺ to Mn²⁺ reduction (Mn₂O₃ → MnO), respectively, corresponding to the main existence of Mn⁴⁺ and Mn³⁺ from XPS analysis [39,40]. For the peaks at lower temperature (300~400 °C), the temperature of reduction peaks ranked as MnO_x-II < MnO_x-III < MnO_x-I < MnO_x-IV. Lower reduction temperature usually means better reducibility, which can promote oxygen species migration and mobility [41,42]. The reduction temperature of MnO_x-IV was relatively higher, corresponding to its weak reducibility and poor catalytic activity. With better redox ability at lower temperatures, the low-temperature activity of MnO_x-II would be greatly promoted.

To analyze the acidity of MnO_x catalysts, the NH₃-TPD profiles are presented in Figure 5c. With desorption temperature increasing, the two peaks were ascribed to Lewis acid sites (physisorbed NH₃ and NH₄⁺) and Brönsted acid sites (desorption of NH₄⁺), respectively [43]. Generally, surface acidity is of great importance in the removal procedure of VOCs, covering reactant adsorption, ozone decomposition, and resultant desorption [44]. Notably, Lewis acid sites are closely correlated to reactive oxygen species formation and C–C bond breaking, which facilitate the further reaction of HCHO [45,46]. With desirable catalytic activity, MnO_x-II showed a lower desorption temperature (86 °C) and more Lewis acid sites, signifying its better acidity at lower temperatures, which would facilitate the degradation of HCHO indoors.

2.3. Effect of H₂O

As an inevitable component in indoor air, water vapor (H₂O) is known to be a possible influence factor for catalytic performance. In the long-time H₂O resistance test, two different concentrations of H₂O (RH = 50% and RH = 100%) were added to reaction gas (Figure 6). Meanwhile, the HCHO conversion and outlet CO₂ selectivity over MnO_x-II catalyst at 30 °C were monitored during the test, with an O₃/HCHO ratio of 2.0. With the addition of a low concentration of H₂O (RH = 50%), the HCHO conversion was basically unchanged after 2 h. Subsequently, introducing a higher concentration of H₂O (RH = 100%), the HCHO conversion slightly decreased to a stable platform at ~97% and then gradually recovered to ~99% after stopping H₂O. Evidently, H₂O of high concentration resulted in catalyst poisoning to a certain extent, and the slight deactivation was fortunately reversible. In addition, the outlet CO₂ selectivity obviously increased when introducing H₂O, as the generated OH active radicals may be conducive to the complete oxidation of HCHO [47]. Moreover, the stability test was conducted, and the HCHO conversion was maintained at ~100% for 7 h, demonstrating the desirable stability of MnO_x-II.

2.4. In-Situ DRIFTS Measurement

To study the HCHO ozonation reaction mechanism on MnO_x-II catalysts, in situ DRIFTS spectra were collected to record the evolution of intermediate products. As shown in Figure 7a, the dynamic changes in the HCHO adsorption process (30 ppm HCHO+N₂) at ambient temperature were conducted with the change in reaction time. Typically, the peaks located between 800 and 1200 cm⁻¹ were identified as DOM species [48]. While the bands at 1579 cm⁻¹ and 1376 cm⁻¹ were ascribed to the asymmetric and symmetric (COO) stretch vibration of formates, respectively. Also, the band at 2858 cm⁻¹ was associated with various hydrocarbon vibrations [49]. Additionally, the peaks at 3690 and 1350 cm⁻¹ belonged to hydroxyl groups and carbonate species, respectively [50]. The weak inverse peaks of hydroxyl groups may reflect the utilization of OH groups and the continuous regeneration, possibly facilitating the further conversion of DOM species [51].

When the adsorption of HCHO was saturated, 90 ppm O₃ was added to the feed gas. Clearly, the intensity of DOM bands (1071 and 958 cm⁻¹) gradually disappeared, as presented in Figure 7b. Simultaneously, the peaks of formate species (1376, 1579, and 2858 cm⁻¹) decreased with time, and the peaks corresponding to different modes of carbonate species, such as hydrocarbonate species (1606 cm⁻¹), bicarbonate species (1420 cm⁻¹) and monodentate carbonate species (1531 cm⁻¹) emerged [52]. Moreover, the band that appeared at 2349 cm⁻¹ corresponded to CO₂ formation as the ideal final product. And the band of hydroxyl groups (3690 cm⁻¹) was basically unchanged, which may be replenished by the generated H₂O during the HCHO catalytic ozonation.

2.5. Theoretical Calculation and Reaction Mechanism

For heterogeneous catalysis, the adsorption of reactants on MnO_x catalysts with α-Mn₂O₃ structure was an important process [53]. To find out the role of oxygen vacancies in the reactant adsorption procedure, the adsorption energies and bond lengths of O₃ and HCHO on the α-Mn₂O₃ (1 1 1) surface were evaluated (Figure 8). With an oxygen vacancy (V₀), the calculated O₃ adsorption energy (−0.46 eV) and Mn-O distance (1.935 Å) were lower than the original ones (E_ads = −0.34 eV, L_Mn-O = 1.951 Å). As for the adsorption of HCHO, the adsorption energy was −0.36 eV, higher than the configuration with V₀ (E_ads = −0.52 eV), and the Mn–O bond length was in the same situation (L_Mn-O = 2.44 Å, L_Mn-O = 2.34 Å with V₀). Typically, more negative adsorption energies and shorter bond lengths always indicate an enhanced interaction between reactants and catalytic surfaces [54]. Thus, the existence of oxygen vacancies may promote reactant adsorption and consequently accelerate catalytic reaction.

In summary, coupling in situ DRIFTS with DFT analysis, a feasible reaction route of HCHO catalytic ozonation is depicted in Figure 9. Firstly, with the promotion of oxygen vacancies, the adsorption process of HCHO and O₃ on surface reactive sites of MnO_x catalysts occurred. And then the active oxygen species immediately converted HCHO into DOM species, while the surface reactive hydroxyl groups were related to the formation of formate species. Then, on the surface oxygen vacancies, O₃ was quickly decomposed to O₂ molecules and reactive oxygen radicals (O*). Finally, formate species were further oxidized into carbonic acid and decomposed to form CO₂ and H₂O, producing harmless degradation products.

3. Materials and Methods

3.1. Catalyst Preparation

For the synthesis of MnO_x-I, 5 mmol Mn(CH₃COO)₂·4H₂O (Aladdin, Shanghai, China) and 15 mmol H₃BTC (Aladdin, China) were placed in deionized (DI) water and ethanol (Sinopharm, China), respectively. Next, they would be vigorously blended and kept under constant stirring (25 °C, 15 h).

To obtain MnO_x-II, 5 mmol Mn(CH₃COO)₂·4H₂O and a certain amount of PVP (Aladdin, China) were dispersed in a mixture solution (vol ratio of C₂H₅OH and H₂O was 2:1) (A solution). Next, 10 mmol H₃BTC was dissolved in the same C₂H₅OH/H₂O (2:1) solution (B solution). Subsequently, after the thorough mixing of A and B, the admixture was left to stand for 12 h at ambient temperature.

During the typical synthesis of MnO_x-III, 0.1 mol/L Mn(CH₃COO)₂·4H₂O (5 mmol) solution and 0.1 mol/L 8-hydroxyquinoline (10 mmol) (Aladdin, Shanghai, China) solution were mixed vigorously and maintained for 5 h at normal temperature.

Typically, the MnO_x-IV was synthesized by the co-precipitation method for comparison via the intensive mixing of Mn(CH₃COO)₂·4H₂O solution and NaOH (Sinopharm, Beijing, China) solution.

All the samples were separated through centrifugation and washed with DI water and ethanol. After being totally dried and then annealed at 500 °C for 3 h, the obtained catalysts were ground to 40~60 mesh. In summary, the synthesis scheme of MOF-derived MnO_x catalysts is presented in Figure 10.

3.2. Catalytic Activity

The catalytic ozonation of HCHO over synthesized catalysts was performed on an experimental setup, as depicted in Figure S2. Meanwhile, the detailed procedures were summarized in Supporting Information (Text S1).

The catalytic activity was evaluated by conversion efficiencies of HCHO and O₃, and the selectivity of CO and CO₂, which were calculated with the following equations:

$$[\mathrm{HCHO}]\mathrm{conv}.={\displaystyle \frac{[\mathrm{CO}]\mathrm{outlet}+[{\mathrm{CO}}_2]\mathrm{outlet}}{[\mathrm{HCHO}]\mathrm{initial}}}\times 100\%$$

(3)

$$[{\mathrm{O}}_3]\mathrm{conv}.={\displaystyle \frac{[{\mathrm{O}}_3]\mathrm{initial}-[{\mathrm{O}}_3]\mathrm{outlet}}{[{\mathrm{O}}_3]\mathrm{initial}}}\times 100\%$$

(4)

$$[{\mathrm{CO}}_2]\mathrm{selec}.={\displaystyle \frac{[{\mathrm{CO}}_2]\mathrm{outlet}}{[\mathrm{CO}]\mathrm{outlet}+[{\mathrm{CO}}_2]\mathrm{outlet}}}\times 100\%$$

(5)

where [HCHO]_initial and [O₃]_initial are the initial concentration of HCHO and O₃, unit ppm, respectively. [CO]_outlet, [CO₂]_outlet and [O₃]_outlet are the outlet concentration of CO, CO₂ and O₃, unit ppm, respectively.

3.3. Catalyst Characterization

The samples were evaluated by XRD, SEM, XPS, BET, NH₃-TPD, H₂-TPR, and in situ DRIFTS characterizations. And the materials in detail are presented in Supporting Information (Text S2).

3.4. DFT Calculation

The DFT calculations were conducted with the CASTEP package using Material Studio software 2019 [55]. And the exchange–correlation interaction was processed with the generalized gradient approximation (GGA) using the Perdew–Burke–Enzerhof (PBE) function. A 650 eV cutoff energy was adopted for all DFT calculations. From XRD analysis, a an α-Mn₂O₃ phase of MnO_x catalysts was confirmed, and the Mn₂O₃ (1 1 1) surface was cleaved, as shown in Figure S3 [56,57]. And a 16 Å vacuum layer was introduced in the z-direction to eliminate the next layer’s influence. The adsorption energy (Eads) of adsorbates on the catalyst surface was determined by the equation below:

E_ads = E_{(adsorbate-catalyst)} − (E_catalyst + E_adsorbate)

(6)

where E_{(adsorbate-catalyst)} is the total energy of catalyst-adsorbate system, E_catalyst is the total energy of catalyst, and E_adsorbate is the total energy of the gas-phase molecule.

4. Conclusions

To completely remove HCHO under room temperature, the catalytic ozonation method with higher efficiency and fewer byproducts is an efficient way. Herein, a series of efficient MOF-derived MnO_x catalysts with distinct shapes were fabricated via a facile sacrificial template strategy for HCHO catalytic ozonation. The flower-like MnO_x-II catalyst showed excellent catalytic activity (100% HCHO conversion, O₃/HCHO = 1.5) and higher CO₂ selectivity (above 90%, O₃/HCHO = 2.5), while the MnO_x-IV prepared by the co-precipitation method exhibited poor catalytic performance. The unique flower-like structure of MnO_x-II endows it with expanded surface area and abundant surface active sites, facilitating the catalytic ozonation process of HCHO. Moreover, its higher Mn³⁺ and O_ad content promoted the formation of oxygen vacancies, as verified by Raman spectra. Therefore, the presence of oxygen vacancies appears crucial for catalysis, accelerating reactant adsorption, ozone decomposition, and complete oxidation of HCHO. In addition, the better acidity and reducibility, sufficient active oxygen species, and improved oxygen mobility also contributed to the excellent activity of MnO_x-II. In stability and H₂O tolerance experiments, MnO_x-II maintained ~100% HCHO conversion without H₂O and ~97% HCHO conversion when RH = 100%, respectively. Based on DFT calculation, the crucial role of oxygen vacancies in promoting HCHO and O₃ adsorption process was elucidated. Moreover, through the detection of intermediate products, a feasible catalytic mechanism was raised (HCHO, DOM, formates, H₂CO₃, H₂O+CO₂, successively).