ZIF-67 Derived Cu-Co Mixed Oxides for Efficient Catalytic Oxidation of Formaldehyde at Low-Temperature

Abstract

It is still an intractable problem to exploit high-efficient Co-based catalysts for low-temperature HCHO oxidation. Herein, we synthesized a series of Cu-doped Co₃O₄ catalysts (Cu₁Co₈, Cu₁Co₄, and Cu₁Co₂ corresponded to 1/8, 1/4, and 1/2 of Cu/Co molar ratios, respectively) via in situ pyrolysis of bimetal Cu-ZIF-67 precursors and the pure Co₃O₄ sample was also prepared through directly annealing monometal ZIF-67 for comparison. Performance tests of HCHO oxidation found that Cu doping remarkably enhanced the low-temperature HCHO oxidation performance of Co₃O₄ sample, and thereinto the Cu₁Co₄ possessed the optimal HCHO oxidation activity, which achieved 90% HCHO conversion at 108 °C. The characterization results revealed that the stronger interaction between Cu and Co species (Co²⁺ + Cu²⁺ ↔ Co³⁺ + Cu⁺) of Cu₁Co₄ not only facilitates the formation of defect sites, Co³⁺ and surface adsorbed oxygen species but also improves its low-temperature reducibility, and consequently resulting in its superior HCHO oxidation performance. Furthermore, the in-situ DRIFTS results suggested that the formaldehyde oxidation over Cu₁Co₄ followed HCHO → H₂CO₂ → HCOO⁻ → CO₃²⁻ → CO₂ pathway. The present work provides a novel and facile approach to fabricating highly effective Co-based catalysts for low-temperature HCHO oxidation.

1. Introduction

Indoor formaldehyde (HCHO) is primarily generated from numerous building/upholstery materials and is one of the most harmful indoor airborne pollutants on account of its strong irritation, reproductive toxicity, and potential carcinogenesis [1,2]. In order to reduce the hazard of HCHO to residents, governments and organizations around the world have promulgated a series of recommended HCHO exposure limits. According to the World Health Organization (WHO, 2000), the HCHO guideline value in indoor air is set at 0.1 mg/m³ [3]. In Singapore, the HCHO recommendation value should not exceed 0.12 mg/m³ [4]. Recently, China has revised and issued the new “Standards for indoor air quality” (GB/T 18883-2022), in which the HCHO concentration threshold is below 0.08 mg/m³ [5]. Hence, it is imperative to explore high-efficient and low-cost HCHO elimination technologies for improving the living environment of people and meeting more and more rigorous policies. In the past decades, various removal techniques, including adsorption [6], plasma decomposition [7], photocatalysis [8], and thermal catalytic oxidation [9,10], have been studied and employed for HCHO purity. Amongst them, thermocatalysis is deemed a particularly suitable and promising approach owing to its high HCHO degradation and mineralization efficiency, excellent reusability, and no secondary pollution [11,12,13,14].

Up to now, a number of catalysts for HCHO oxidation have been designed and synthesized [15,16,17,18,19,20]. Although precious-metal-based catalysts (Pt/TiO₂, Pd/CeO₂, Au/FeO_x, Ir/Al₂O₃, etc.) manifested exceptional HCHO oxidation performance [21,22,23,24,25,26,27], the scarce terrestrial abundance and exorbitant prices of precious metals seriously restrict their large-scale application [28,29,30]. Therefore, abundant and low-cost transition-metal oxides have aroused increasing interest. Particularly, spinel-type Co₃O₄-based catalysts have been extensively investigated and utilized for HCHO catalytic oxidation profiting from their excellent reducibility and stability [31,32,33]. For instance, Lv et al. [34] reported that the cellular-like Co₃O₄ catalyst synthesized through a seeded method could totally oxidize HCHO at 165 °C. Bai et al. [35] compared the HCHO oxidation activity of nano-Co₃O₄, 2D-Co₃O₄, and 3D-Co₃O₄ catalysts and discovered that the 3D-Co₃O₄ catalyst possessed the optimal catalytic performance, mainly deriving from its unique 3D pore channel structure. However, the low-temperature O₂ activation ability of pristine Co₃O₄ catalysts is relatively poor, which in turn results in the accumulation of HCHO oxidation intermediates on active sites and deactivation of catalysts at low-temperature [31,36].

Generally, ion doping is an effective strategy to improve the O₂ activation ability of Co₃O₄ catalysts due to the strong interaction between cobalt and doping ions [37,38,39,40]. Thereinto, copper (Cu) element is considered an excellent dopant that benefits from its low valency and large ion radius, which could weaken the strength of the Co-O bond, as well as promote the generation of defects [41,42]. For example, Fan et al. [43] prepared a series of Cu_aCo_1−aO_x (a = 0.1, 0.2, 0.4, 0.6) catalysts through a solvothermal method. With an increase in the Cu doping amount, the toluene oxidation activity of the resultant catalysts was first increased and then decreased. Among them, the Cu_0.4Co_0.6O_x catalyst was the most active sample, which was ascribed to its adequate Co³⁺ and surface-adsorbed oxygen species. Moreover, Zhang et al. [44] also found that the incorporation of appropriate amounts of Cu into Co₃O₄ could facilitate O₂ activation and thereby promote the oxidation of toluene and propane. As is well known, O₂ activation is regarded as a crucial step during low-temperature HCHO oxidation, and adequate surface-adsorbed oxygen species could boost the oxidation of HCHO and intermediates [45,46]. Inspired by these, we speculate that Cu-doping might enhance the HCHO degradation performance of Co₃O₄ at low-temperature. Nevertheless, to our knowledge, the study of the Cu dopant effect on the HCHO oxidation activity of Co₃O₄ is still scarce and needs to be further investigated.

As a novel class of porous crystalline materials, zeolitic imidazolate framework-67 (ZIF-67) consisting of Co²⁺ and imidazole linkers has drawn great research interest owing to its high surface area, tunable architectures, and compositions in recent years [47]. More importantly, the ZIF-67 has been extensively applied as the precursor to synthesize various Co-based porous materials (e.g., Co₃O₄, NiCo₂O₄, CoS) by means of its unique thermal behavior and chemical reactivity [48,49,50,51]. In this work, we prepared a series of Cu-doped Co₃O₄ catalysts via in situ pyrolysis of bimetal Cu-ZIF-67 precursors. Compared with the pure Co₃O₄ catalyst, these Cu-doped Co₃O₄ samples displayed remarkably improved low-temperature HCHO oxidation performance, especially Cu₁Co₄, which exhibited the highest catalytic activity, achieving 90% of HCHO conversion at 108 °C. Comprehensive characterizations were carried out to elucidate the effect of Cu doping on the physicochemical properties of Co₃O₄, and the structure-activity relationship was discussed. Moreover, the plausible HCHO oxidation pathway over the Cu₁Co₄ catalyst was also explored by in-situ DRIFTS.

2. Results and Discussion

2.1. HCHO Oxidation Performance Tests

The HCHO oxidation ignition curves of Co₃O₄, Cu₁Co₈, Cu₁Co₄, and Cu₁Co₂ are shown in Figure 1. The HCHO oxidation activity of catalysts exhibited a volcano-shaped relationship that initially increased and subsequently decreased with the increase in the Cu/Co ratio. Thereinto, the Cu₁Co₄ displayed the optimal HCHO oxidation activity. The ignition temperatures of 50% and 90% HCHO conversion (T₅₀ and T₉₀) of catalysts were summarized in

Table 1. Ignition temperatures of Co₃O₄, Cu₁Co₈, Cu₁Co₄, and Cu₁Co₂ for HCHO oxidation.

Catalyst	T50 (°C)	T90 (°C)
Co3O4	147	157
Cu1Co8	103	122
Cu1Co4	98	108
Cu1Co2	101	112

. The T₅₀ and T₉₀ of Co₃O₄ were 147 and 157 °C, respectively. By contrast, for the Cu₁Co₈, Cu₁Co₄, and Cu₁Co₂, their T₅₀, respectively, decreased to 103, 98, and 101 °C. Meanwhile, their T₉₀ were as low as 122, 108, and 112 °C. Therefore, the HCHO degradation performance of catalysts was ranked as follows: Cu₁Co₄ > Cu₁Co₂ > Cu₁Co₈ > Co₃O₄. The above results demonstrated that moderate Cu doping could effectively improve the HCHO oxidation ability of Co₃O₄.

In view of the ineluctable existence of moisture in the actual indoor air, the effect of relative humidity on the HCHO oxidation activity of Cu₁Co₄ was assessed through a transient response experiment at 110 °C (Figure 2a). In the first 100 min, the HCHO conversion of Cu₁Co₄ was ~95% in the presence of H₂O. Next, when H₂O was removed from the feed gas, the HCHO conversion of Cu₁Co₄ gradually decreased and remained steady at ~83%. Whereas the HCHO oxidation activity of Cu₁Co₄ was completely recovered after the reintroduction of H₂O, demonstrating that H₂O had a promotional effect on Cu₁Co₄ for HCHO oxidation. It could be because H₂O accelerated the activation of O₂, which was regarded as a key step for HCHO oxidation [52]. Moreover, the stability of Cu₁Co₄ in the HCHO oxidation was evaluated by a 48-h continuous test at 110 °C. As depicted in Figure 2b, the Cu₁Co₄ maintained ~95% HCHO conversion, and no evident performance deterioration could be discerned during the whole test process, indicating the good durability of Cu₁Co₄.

2.2. Physicochemical Property Analyses

In order to determine the crystal structure and crystalline size of Co₃O₄, Cu₁Co₈, Cu₁Co₄, and Cu₁Co₂, XRD measurements were carried out (Figure 3a). The Co₃O₄ sample exhibited nine well-defined diffraction peaks at 19.1°, 31.3°, 36.9°, 38.5°, 44.9°, 55.7°, 59.4°, 65.3°, 77.4°, corresponding to (111), (220), (311), (222), (400), (422), (511), (440), (533) lattice planes of spinel-type Co₃O₄ (JCPDS 42-1467), respectively [53]. As for the Cu₁Co₈, Cu₁Co₄, and Cu₁Co₂, their XRD patterns were similar to the Co₃O₄ sample, and no metallic copper or copper oxides-related diffraction peaks were detected, possibly resulting from the high dispersion of copper species or the incorporation of Cu ions into the crystalline lattice of Co₃O₄. Of note, the (311) diffraction peaks of these Cu-Co mixed oxides shifted towards lower angles compared with pure Co₃O₄ (Figure 3b). It might be owing to the substitution of smaller Co²⁺ (0.65 Å) and Co³⁺ (0.55 Å) cations by larger Cu²⁺ (0.73 Å) cations, which would enlarge the lattice constant of Co₃O₄ [40,54]. The crystalline size of all catalysts was determined via the Scherrer formula based on the strongest (311) planes and summarized in

Table 2. Crystallite size and textural properties of Co₃O₄, Cu₁Co₈, Cu₁Co₄, and Cu₁Co₂.

Catalyst	Crystallite Size (nm)	SBET (m2/g)	Vp (cm3/g)	Average Pore Size (nm)
Co3O4	18	43	0.22	20
Cu1Co8	15	37	0.18	20
Cu1Co4	14	44	0.20	18
Cu1Co2	17	32	0.15	19

. It could be found that there was no significant crystalline size difference between the pristine Co₃O₄ and these Cu-Co mixed oxides.

The Raman spectra of samples are displayed in Figure 4a. The typical Raman-active peaks of crystalline Co₃O₄ were observed in all samples. In detail, the peaks at around 184, 465, 506, 600, and 654 cm⁻¹ were attributed to the ${\mathrm{F}}_{2\mathrm{g}}^{\left(1\right)}$, E_g, ${\mathrm{F}}_{2\mathrm{g}}^{\left(2\right)}$, ${\mathrm{F}}_{2\mathrm{g}}^{\left(3\right)}$, and A_1g vibration modes of Co₃O₄, respectively [44]. In addition, no evident characteristic bands of CuO_x were detected in the Cu₁Co₈, Cu₁Co₄, and Cu₁Co₂. As the Cu/Co ratio increased, the Raman peak positions of samples gradually shifted to higher wavenumbers (i.e., blue shift), which could be derived from the photon-confinement effect induced by the surface defects of Co₃O₄ [42,55]. Therefore, the above result demonstrated that Cu doping could effectively increase the surface defect concentration of Co₃O₄. Usually, adequate surface defect sites are advantageous to O₂ activation and generation of surface-adsorbed xygen species, which are vital for low-temperature HCHO oxidation [45,46].

The FT-IR spectra of samples are illustrated in Figure 4b. For all catalysts, the broad peak at 3700–3000 cm⁻¹ and weak peak at about 1630 cm⁻¹ were assigned to the υ(OH) and δ(OH) of surface adsorbed water [43]. Meanwhile, two strong stretching vibration peaks located at 663 and 566 cm⁻¹ were observed, which corresponded to the characteristic absorption peaks of four-coordinated Co²⁺-O and six-coordinated Co³⁺-O, respectively [42]. Likewise, no typical Cu-O characteristic absorption bands were detected over Cu₁Co₈, Cu₁Co₄, and Cu₁Co₂ catalysts.

Figure 5a–d shows the SEM images of Co₃O₄ and Cu-doped Co₃O₄ samples. For all samples, their SEM images exhibited well-defined cubic morphology, and their average size was approximately 170 nm, which suggested that Cu-doping did not obviously change the shape of the Co₃O₄ sample. In order to study the element distribution of Cu₁Co₄, EDS mapping was carried out. As depicted in Figure 5e, the Co, Cu, and O elements were evenly dispersed throughout the entire selected region. Furthermore, more detailed structure characteristics of Cu₁Co₄ were further determined by TEM. As displayed in Figure 6a, the cube-like Cu₁Co₄ consisted of numerous aggregated nanoparticles with a size of 10–15 nm, in good conformity with the estimated result of XRD (~14 nm). Simultaneously, the piled pores generated by the stacking of nanoparticles could also be discerned. The HRTEM images (Figure 6b,c) clearly showed the interplanar distance of 0.46, 0.28, and 0.24 nm, corresponding to the (111), (220), and (311) crystal planes of Co₃O₄, respectively [29,42]. In addition, no lattice fringes associated with metallic copper or copper oxides were observed. Combined with the XRD and EDS mapping results (Figure 3b and Figure 5e), it could be speculated that Cu ions were incorporated into the crystalline lattice of Co₃O₄.

To reveal the specific surface area and porous structure of Co₃O₄, Cu₁Co₈, Cu₁Co₄, and Cu₁Co₂, the N₂ adsorption-desorption isotherm tests were performed (Figure 7). It could be seen that all samples exhibited the characteristics of IV isotherms with type H3 hysteresis loops, indicating the presence of piled mesopores [27], which was in line with the TEM results (Figure 6a). Meanwhile, the BJH pore size distribution results also attested to the existence of mesopores. The textural parameters of samples are summarized in Table 2. Interestingly, compared with Cu-doped Co₃O₄ samples, the pure Co₃O₄ catalyst possessed relatively larger S_BET and V_p but displayed the worst HCHO degradation performance, demonstrating that the S_BET and V_p of catalysts might not be the major factors that determine their HCHO oxidation performance.

XPS tests were carried out to elucidate the surface element compositions and chemical states of catalysts. As illustrated in Figure 8a, the Co 2p spectra of samples displayed a spin-orbit doublet at a binding energy of 779.5–780.0 and 794.6–795.1 eV, which was identified as Co 2p_3/2 and Co 2p_1/2, respectively [53]. The ΔE_{Co 2p} (energy differences between Co 2p_3/2 peak and Co 2p_1/2 peak) of all samples were ca. 15.1 eV, suggesting the coexistence of Co²⁺ and Co³⁺ species [29]. It is interesting that the Co 2p profiles of Cu-Co samples moved towards lower binding energy (~0.5 eV) in comparison to the pure Co₃O₄ sample, revealing that there was a mutual interaction between Cu and Co species [43]. Furthermore, the surface Co³⁺/Co²⁺ molar ratios of samples were summarized in

Table 3. Surface element properties of Co₃O₄, Cu₁Co₈, Cu₁Co₄, and Cu₁Co₂.

Catalyst	Co3+/Co2+	Cu+/Cu2+	Oα/Oβ
Co3O4	0.53	-	0.33
Cu1Co8	0.59	0.16	0.51
Cu1Co4	0.78	0.29	0.55
Cu1Co2	0.69	0.21	0.53

and obeyed the sequence as follows: Cu₁Co₄ (0.78) > Cu₁Co₂ (0.69) > Cu₁Co₈ (0.59) > Co₃O₄ (0.53). Apparently, the Cu doping significantly increased the Co³⁺/Co²⁺ ratios of samples, which could be owing to the mutual interaction between Cu and Co species (Cu²⁺ + Co²⁺ ↔ Cu⁺ + Co³⁺) [42]. It is well-known that the Co³⁺ species possess strong oxidation ability and abundant Co³⁺ is beneficial to enhance the HCHO oxidation performance [31,40,41]. Hence, the Cu₁Co₄ with the highest Co³⁺/Co²⁺ ratio exhibited the best HCHO decomposition ability amongst these catalysts.

As shown in Figure 8b, the asymmetrical Cu 2p_3/2 peaks of Cu-Co samples could be fitted into two independent peaks. Specifically, the peak at 931.9 ± 0.1 eV was attributed to Cu⁺ species, while the other peak at 934.3 ± 0.1 eV belonged to Cu²⁺ species. The Cu⁺/Cu²⁺ molar ratio of Cu₁Co₄ was much higher than those of Cu₁Co₈ and Cu₁Co₂ (0.29 vs. 0.16 and 0.21), which revealed that the mutual interaction between Cu and Co species over Cu₁Co₄ was stronger compared with Cu₁Co₈ and Cu₁Co₂ (Cu²⁺ + Co²⁺ ↔ Cu⁺ + Co³⁺) [42].

The O 1s spectra of samples (Figure 8c) could be deconvoluted into two components. The peaks located at 529.7–530.1 and 531.5 ± 0.1 eV corresponded to the lattice oxygen (O_β) and surface adsorbed oxygen (O_α) species, respectively [29]. Noticeably, the binding energy of O_β centered at 530.1 eV for the pure Co₃O₄ sample shifted to 529.7 eV for Cu-doped Co₃O₄ samples, demonstrating the mutual interaction between Cu and Co species changed the coordination environment of lattice oxygen [41]. It is reported that the more electrophilic Cu²⁺ will draw electrons away from Co in the Cu-doped Co₃O₄ samples [56]. It means that the interaction between Cu²⁺–O²⁻–Co²⁺ entity makes the charge transfer between Cu and Co cations dexterously as below: Cu²⁺ + Co²⁺ ↔ Cu⁺ + Co³⁺. Therefore, the electron density of lattice oxygen in the Cu²⁺–O²⁻–Co²⁺ entity would increase because of the electronic delocalization effect, and the binding energy of lattice oxygen shifted towards a lower value due to the shielding effect [57,58,59]. As shown in Table 3, the O_α/O_β molar ratios of samples decreased in the order of Cu₁Co₄ (0.55) > Cu₁Co₂ (0.53) > Cu₁Co₈ (0.51) > Co₃O₄ (0.33). The above results indicated that Cu doping effectively enhanced the O₂ activation ability of Co₃O₄. It could be because the mutual interaction between Cu and Co species promotes the generation of surface defects, which is in line with the Raman results (Figure 4a). It has been established that the O_α species is the reactive oxygen species (ROS) in the deep oxidation of HCHO [29,45]. In this view, the Cu₁Co₄ with the highest O_α/O_β ratio should show the best HCHO degradation performance.

Normally, the low-temperature reducibility of catalysts is closely associated with their HCHO oxidation activity. H₂-TPR experiments were implemented to investigate the effect of the Cu/Co molar ratio on the redox properties of samples, and the corresponding curves are illustrated in Figure 9. For the pristine Co₃O₄, there existed two well-defined H₂ consumption peaks at 355 and 460 °C, which belonged to the sequential reduction of Co³⁺ → Co²⁺ → Co⁰ [60]. It is worth noting that the incorporation of Cu species dramatically decreased the initial reduction temperature of samples, possibly deriving from the mutual interaction between Cu and Co species [44]. For the Cu₁Co₈, it exhibited a discernible double-peak with maxima at 266 and 317 °C, in which the peak at low-temperature could be attributed to the simultaneous reduction of Cu²⁺/Cu⁺ → Cu⁰ and Co³⁺ → Co²⁺ and the latter could be assigned to the reduction of Co²⁺ → Co⁰ [29,43,61]. The TPR profiles of the Cu₁Co₄ were analogous to that of the Cu₁Co₈, but the reduction temperature of Cu₁Co₄ further moved to a lower temperature (248 and 303 °C). On the contrary, for the Cu₁Co₂, its reduction peaks slightly shifted towards a higher temperature (262 and 308 °C), suggesting that excess Cu doping would inhibit the reduction of Cu-Co oxides. Therefore, it was concluded that the incorporation of proper Cu could significantly enhance the low-temperature reducibility of Co₃O₄, consequently leading to the improvement of HCHO oxidation activity.

2.3. Reaction Pathway

In order to elucidate the mechanism of HCHO oxidation over the Cu₁Co₄ catalyst, an in-situ DRIFTS measurement was conducted (Figure 10). Several absorption vibration bands were detected during the early stage of exposure to HCHO feed gas. The peaks centered at 1420 and 1436 cm⁻¹ were assigned to the υ(OCO) in dioxymethylene (DOM) species that was generated via the attack of nucleophilic surface adsorbed oxygen species [62]. The peak located at ~1340 cm⁻¹ was ascribed to υ_s(COO) of formate species; the bands at 1550, 1567, and 1583 cm⁻¹ could be attributed to the υ_as(COO) of formate species; the peaks at 2884, 2936 and 2966 cm⁻¹ belonged to the υ(CH) of formate species [63,64,65]. As the reaction continued, the characteristic absorption peak of carbonate species (CO₃²⁻) at ~1320 cm⁻¹ could also be observed [64,66]. It demonstrated that DOM and formate species were the initial intermediates, which would be further oxidized into carbonate species. In addition, the broad band at 3700–3000 cm^–1 and the weak band at ~1635 cm^–1 were respectively assigned to υ(OH) and δ(OH) of H₂O molecules adsorbed on Cu₁Co₄ surface, which could be originated from the reactant gas or products of HCHO degradation [29].

Based on the aforementioned results, a probable HCHO degradation route over the Cu₁Co₄ catalyst is proposed. Firstly, the HCHO molecules adsorbed on the Cu₁Co₄ catalyst are promptly transformed into DOM species through the attack of nucleophilic surface adsorbed oxygen species, which were formed by the activation of O₂ over the defect sites of the Cu₁Co₄ catalyst. Next, DOM species are converted into formate species via dehydrogenation, and then, formate species further react with surface-adsorbed oxygen species to produce carbonate species. Lastly, carbonate species decompose into CO₂ and H₂O. The stronger interaction between Cu and Co species (Co²⁺ + Cu²⁺ ↔ Co³⁺ + Cu⁺) of the Cu₁Co₄ catalyst endowed it with richer defect sites, higher Co³⁺/Co²⁺ ratio, superior O₂ activation ability, and low-temperature reducibility, consequently leading to its better HCHO degradation ability.

O₂ → 2[O]_s

(1)

HCHO + [O]_s → [H₂CO₂]_s

(2)

[H₂CO₂]_s → [HCOO⁻]_s + H⁺

(3)

[HCOO⁻]_s+ [O]_s → [CO₃²⁻]_s + H⁺

(4)

[CO₃²⁻]_s + 2H⁺ → CO₂ + H₂O

(5)

3. Experimental

3.1. Materials

Co(NO₃)₂·6H₂O (99%), Cu(NO₃)₂·3H₂O (99%), and CTAB (99%) and 2-methylimidazole (98%) were purchased from Macklin Ltd. (Shanghai, China). Ethanol (99.7%) was purchased from Fuchen Ltd. (Tianjin, China). Deionized water (DI water) was obtained from UPR-Ⅳ-20L (Ulupure Ltd., Chengdu, China) and used throughout.

3.2. Preparation of Co₃O₄ and Cu-Co Oxides

Firstly, 9.08 g of 2-methylimidazole was dissolved in 80 mL DI water (marked as solution A). Then, a proper amount of Co(NO₃)₂·6H₂O and Cu(NO₃)₂·3H₂O (Co²⁺ + Cu²⁺ = 2 mmol, Cu²⁺/Co²⁺ molar ratios were 0, 1/8, 1/4 and 1/2, respectively) and 10 mg CTAB was dissolved in 20 mL DI water (marked as solution B). After that, solution B was quickly injected into solution A, and the obtained mixed solution was vigorously stirred at 25 °C for 1.5 h. These as-synthesized ZIF-67 precursors were harvested by centrifugation, washed 3 times with ethanol, and dried at 70 °C overnight. Finally, the ZIF-67 with different Cu²⁺/Co²⁺ molar ratios were annealed at 350 °C for 2 h (1 °C/min) under air atmosphere, and these acquired samples were labeled as Co₃O₄, Cu₁Co₈, Cu₁Co₄, and Cu₁Co₂, respectively.

3.3. Catalyst Characterization

PXRD measurements were conducted on a Rigaku Smartlab9 X-ray diffractometer (Rigaku, Tokyo, Japan) with Cu Kα radiation (λ = 1.5418 Å) operated at 40 kV and 40 mA. The Raman spectra were recorded on a HORIBA LabRAM HR Evolution Raman spectrometer (Horiba, Paris, France) using a 532 nm He-Ne laser. The FT-IR spectra were collected on a Nicolet iS10 FT-IR spectrometer (Thermo Scientific, Waltham, MA, USA). The morphologies and microstructures of catalysts were observed with FESEM (FEI, Waltham, MA, USA, QUANTA FEG 450) and TEM (Thermo Scientific, Hillsboro, OR, USA, Talos F200X), respectively. N₂ adsorption-desorption isotherms of catalysts were determined using an Autosorb-iQ-2 instrument (Quantachrome, Boynton Beach, FL, USA) at −196 °C. XPS spectra of catalysts were recorded on an ESCALAB 250Xi photoelectron spectroscopy (Thermo Scientific, Waltham, MA, USA) with an Al Kα X-ray source (1486.6 eV). The binding energy (B.E.) of spectra was calibrated by the C 1s peak (284.8 eV). H₂-TPR was tested in a ChemStar TPx instrument (Quantachrome, Boynton Beach, FL, USA) equipped with a TCD detector. Specifically, 30 mg of catalysts were first pretreated in Ar at 300 °C for 30 min, then cooled to 50 °C in Ar. Afterward, catalysts were reduced in a flow of 5%H₂-Ar mixture (30 mL/min) from 50 °C to 700 °C with a ramping rate of 10 °C/min.

The in-situ DRIFTS was performed on a Bruker Tensor 27 FTIR spectrometer (Bruker, Karlsruhe, Germany) equipped with CaF₂ windows. The Cu₁Co₄ catalyst was firstly pretreated in N₂ at 150 °C for 30 min to desorb impurities and then cooled to 120 °C to record the background spectrum. Afterward, feed gas containing HCHO (~80 ppm) was introduced into in situ chamber, and the spectrum was collected continuously by background subtraction.

3.4. Catalytic Evaluation

Typically, 100 mg of catalyst (0.30–0.45 mm) was packed in a quartz reactor (i.d. = 4 mm), which was placed in a furnace, and the reaction temperature was controlled by a K-type thermal couple. All gas flows were controlled by mass flow controllers. The composition of reactant gas was 80 ppm HCHO, 60% relative humidity (RH), 20% O₂, and N₂ balance. The total flow rate was 100 mL min⁻¹, corresponding to a weight hourly space velocity (WHSV) of 60,000 mL g_cat⁻¹ h⁻¹. The initial test temperature was 50 °C. The interval of adjacent temperature points is 10 °C, and the heating rate was 2 °C/min. The residence time for each temperature point was 1 h. The inlet and outlet concentrations of HCHO and CO₂ were analyzed simultaneously using an FTIR gas analyzer (Gasmet Instruments DX-4000, Vantaa, Finland). The HCHO conversion was calculated as follows:

$$\mathrm{HCHO}\mathrm{Conversion}(\%)=\frac{{{[\mathrm{CO}}_2]}_{\mathrm{out}}{-[\mathrm{CO}}_2{]}_{\mathrm{in}}}{{\left[\mathrm{HCHO}\right]}_{\mathrm{in}}}\times 100\%$$

(6)

where [CO₂]_in and [CO₂]_out are the CO₂ concentration (ppm) in the inlet and outlet gas, respectively. [HCHO]_in represents the HCHO concentration (ppm) in the inlet gas. No other carbon compounds other than CO₂ were detected during HCHO catalytic oxidation.

4. Conclusions

In conclusion, a series of Cu-doped Co₃O₄ catalysts with different Cu/Co molar ratios were constructed by in situ pyrolysis of bimetal Cu-ZIF-67 precursors and utilized for low-temperature HCHO catalytic oxidation. All Cu-doped Co₃O₄ catalysts showed remarkably improved HCHO oxidation activity in comparison to the pure Co₃O₄, amongst which the Cu₁Co₄ stood out (T₉₀ = 108 °C). The excellent HCHO oxidation performance of Cu₁Co₄ could be related to its adequate defect sites, surface adsorbed oxygen species, higher Co³⁺/Co²⁺ ratio, and superb low-temperature reducibility, which derived from its stronger interaction between Cu and Co species (Co²⁺ + Cu²⁺ ↔ Co³⁺ + Cu⁺). Moreover, in-situ DRIFTS results demonstrated that the dioxymethylene, formate, and carbonate species were primary reaction intermediates of HCHO oxidation over the Cu₁Co₄ catalyst. We anticipate that the outcomes in the present work might offer an instructive guideline for the rational design of high-performance Co-based catalysts for low-temperature HCHO oxidation.