Sustainable PH₃ Purification over MOF-Derived Ce-Doped CuO Materials: Enhanced Performance and Closed-Loop Resource Recovery

Abstract

To address the limitations of low CuO loading and poor dispersion in conventional supported adsorbents, in this study, MOF (metal–organic framework)-derived CuO with Ce doping (Cu_xCe_yO) was synthesized and used for the adsorption–oxidation of PH₃ under low-temperature and low-oxygen conditions. The results demonstrated that Ce doping increased the PH₃ capacity of the adsorbent from 75.54 mg·g⁻¹ (MOF-derived CuO) to 226.87 mg·g⁻¹ (Cu₁Ce_0.2O). The characterization results indicated that Ce doping significantly altered the physicochemical properties of CuO. Specifically, Cu₁Ce_0.2O exhibited optimal CuO dispersion, the highest adsorbed oxygen concentration, superior redox performance, an increased number of basic sites, and a larger specific surface area and pore volume, all contributing to its improved performance. Analysis of the exhausted adsorbent revealed the formation of Cu₃P and phosphoric acid. And the deactivation of the adsorbent can be attributed to the consumption of CuO and the blockage of pore structure. Surprisingly, the exhausted adsorbent demonstrated considerable photocatalytic performance due to the formation of Cu₃P, enabling the resource utilization of the waste adsorbent, making it a promising material for the adsorption–oxidation of PH₃. This waste-to-resource conversion reduces hazardous solid waste while creating value-added photocatalysts, establishing a sustainable lifecycle from pollutant removal to functional material regeneration.

1. Introduction

Phosphine (PH₃) is a gaseous pollutant in the environment [1,2,3], with a wide range of sources. Natural sources of PH₃ include wetlands, marshes, volcanic activities, and biological decomposition, while anthropogenic sources primarily arise from chemical production activities [4]. PH₃ is a highly toxic gas with an irritating odor. Even at low concentrations, it can cause harm to the health of animals and humans, potentially leading to death [5]. In industrial production, the exhaust gas emitted by the phosphorus chemical industry contains a certain concentration of PH₃. If directly discharged into the air, it will cause damage to the environment and also affect the recovery and utilization of valuable resource gases such as CO [6]. Therefore, waste gas containing PH₃ must be treated, and secondary pollution should be minimized as much as possible.

At present, the common methods for removing PH₃ are mainly divided into dry and wet methods. The wet method removes PH₃ through liquid-phase absorption or liquid-phase catalysis, offering the advantages of high efficiency and absorption capacity [7]. High-concentration PH₃ is often removed using absorption towers. However, the wet method also presents challenges, such as complex absorption equipment, high water consumption, and the risk of secondary pollution. The dry method achieves the adsorption or decomposition and removal of PH₃ through adsorbents or catalysts. This approach features a simple process, high efficiency, and low cost, making it a method with great potential. Common dry methods include gas–solid adsorption [8], catalytic decomposition [4], and adsorption–oxidation [9,10]. The gas–solid adsorption method can efficiently adsorb low-concentration PH₃, but it has a limited adsorption capacity. Catalytic decomposition can decompose PH₃ into P and H₂, but it requires a high temperature, making it energy-intensive and subject to strict operational requirements. The adsorption–oxidation method can efficiently remove PH₃ at low temperatures and convert PH₃ to metal phosphide and phosphoric acid, this method is pollution-free and has the advantages of recycling, high adsorption capacity, and stability. The above characteristics made it be considered a highly promising approach.

The adsorbents used in the adsorption–oxidation method are mostly supported materials, typically prepared by loading strong oxidizing active components such as metals or metal oxides, onto common carriers such as carbon materials [11,12], molecular sieves [13,14], TiO₂ [15], and γ-Al₂O₃ [16]. Previous studies have demonstrated that, among various metal oxides, CuO as the active component exhibits the best activity in PH₃ removal [10,17]. However, many commonly used supported materials face issues such as CuO agglomeration and low loading amount. To further improve the adsorption capacity of PH₃, increasing the loading and dispersion of CuO is crucial. In addition to the conventional loading methods, some studies have also explored obtaining high-loading CuO materials by calcining Cu-containing materials with a special structure as precursors. In recent years, metal–organic framework (MOF) materials have attracted considerable attention due to their advantages [18,19,20,21], such as large specific surface area, tunable pore structure and versatile surface chemistry. Some studies have used MOFs as precursors to synthesize metal oxide nanoparticles, including CuO materials using Cu-based MOFs. MOF-199 is a typical copper-based MOF material. Previous studies have shown that using MOF-199 as a precursor can obtain a kind of CuO material with the characteristics of high CuO loading, a well-developed porous structure, and highly dispersed CuO particles. Duan et al. [22] obtained a material with well-developed porous structure and highly dispersed CuO particles by carbonizing MOF-199 composite material. This material exhibited an SO₂ adsorption capacity of 233.11 mg/g, surpassing previously reported adsorbents [22]. Yang et al. identified the best preparation conditions for CO oxidation by calcining Cu-BTC under various atmospheres, thereby altering the structure of the material and the composition of CuO/Cu₂O [23]. This method enables the production of CuO materials with high loading and dispersion, while also offering a larger specific surface area compared to pure CuO, which enhances their adsorption or catalytic performance.

Additionally, research has explored the addition of bimetal [24,25,26,27], carbon nanotubes [28], SiO₂ [29] and other additives into MOF-derived CuO materials to increase their specific surface area and enhance interaction between active components, thereby further improving their performance. Among these, Ce is frequently used to modify CuO materials due to its beneficial characteristics and strong interaction with Cu [30]. Zhang et al. used MOFs as the precursor to prepare a CuO/CeO₂ catalyst with well-dispersed active sites and excellent oxidation performance for CO [31]. Zhang and Wu et al. quantitatively analyzed the oxygen vacancy concentration in Cu-Ce catalysts and found that the unique role of the Cu-Ce interface contributed to the material’s excellent stability and catalytic performance [32]. These studies indicate that Ce doping enhances the dispersion of CuO particles, while CeO₂, after calcinating, increases the specific surface area and generates more oxygen vacancies [33]. Moreover, the interface between CuO and CeO₂ facilitates unique interactions, including the presence of active sites with enhanced redox properties, which further improve the performance of the adsorbent [34,35].

In this study, a composite adsorbent of CuO/CeO₂ was synthesized using MOFs as the precursor, which is mixed with Ce and then co-calcined. This adsorbent features a facile synthesis approach, with high CuO-loading while maintaining high dispersion, and abundant oxygen vacancies, leading to a considerable adsorption–oxidation performance for PH₃ at low temperature. The superior performance of the adsorbent minimizes both the replacement frequency and the generation of industrial solid waste. Furthermore, its low-temperature operational efficiency significantly curtails energy consumption and carbon emissions. Collectively, these advancements position the solution as a paradigm of energy conservation, emission reduction, and sustainable development principles. The structure–activity relationship of the adsorbent and the adsorption–oxidation process were also analyzed using various characterization techniques. It is noteworthy that the deactivated adsorbent contains Cu₃P, which can be utilized in photocatalytic reactions and holds potential for secondary reuse, thereby helping to prevent resource waste and reduce environmental pollution, embodying the core concept of “waste recycling”.

2. Materials and Methods

2.1. Preparation of Adsorbents

The copper nitrate (Cu(NO₃)₂·3H₂O), trimesic acid (C₉H₆O₆), Ce(NO₃)₃·6H₂O, DMF (C₃H₇NO), anhydrous ethanol (C₂H₆O) and RhB (C₂₈H₃₁ClN₂O₃) were purchased from Macklin Biochemical Tech Co., Ltd. (Shanghai, China). All chemicals were analytical reagents (AR) and used without further purification. PH₃ gas was purchased from Dalian Special Gases Co., Ltd. (Dalian, China). O₂ and N₂ were purchased from Pingxiang Jinchang Gas Co., Ltd. (Pingxiang, China).

Firstly, 1 g of trimesic acid was dissolved in 17 mL of DMF and 17 mL of anhydrous ethanol, while 2 g of Cu(NO₃)₂·3H₂O was dissolved in 17 mL of deionized water. The above two solutions were then mixed and subjected to ultrasound for 30 min. The mixture was subsequently transferred to a 100 mL reactor for hydrothermal reaction at 85 °C for 24 h. Following the completion of the reaction, the reactor was cooled to room temperature. The obtained material was washed 3 times with a mixture of deionized water and anhydrous ethanol, then placed in a vacuum oven for drying at 160 °C for 5 h. After cooling to room temperature, sky-blue powder identified as MOF-199 was obtained. The obtained MOF-199 material was placed in a muffle furnace and heated at a rate of 5 °C/min from room temperature to 450 °C. After 2 h of calcination, a black solid material, designated as MOF-derived CuO was obtained.

A certain amount of MOF-199 material was placed in a beaker, while a certain amount of Ce(NO₃)₃·6H₂O was dissolved in 20 mL of deionized water. The above two precursors were mixed in a magnetic mixer for 3 h, then dried in an oven at 100 °C. The obtained material was placed in a muffle furnace and calcined at a rate of 5 °C/min from room temperature to 450 °C for 2 h. Finally, a black solid material was obtained. Based on the mass ratio of MOF to Ce(NO₃)₃·6H₂O, the resulting materials were denoted as Cu₁Ce_0.14O, Cu₁Ce_0.2O, Cu₁Ce_0.25O, Cu₁Ce_0.33O, Cu₁Ce_0.5O, and Cu₁Ce₁O, respectively. Meanwhile, Cu(NO₃)₂·3H₂O and Ce(NO₃)₃·6H₂O were calcined under the same conditions to obtain CuO and CeO₂ for comparation.

2.2. Activity Tests

PH₃ adsorption–oxidation experiments were carried out in a quartz tube reactor. First, 0.3 g of the adsorbent was loaded into a quartz tube, which was then placed in the reactor and heated to 90 °C. The simulated gas mixture containing 1000 ppm PH₃ and 1% O₂ was prepared by controlling the flow rate. The total volume flow rate was set to 100 mL·min⁻¹, corresponding to a weight hourly space velocity (WSHV) of 20,000 mL·h⁻¹·g⁻¹. The mixed gas was introduced into the quartz tube, with the internal temperature of the reaction system maintained at 90 °C. The PH₃ concentration in the outlet gas was measured by gas chromatography (FULI-9790 II), with samples collected every 30 min for analysis. All experiments were conducted until the removal efficiency dropped below 90%, containing an outlet gas PH₃ concentration exceeding 100 ppm. Meanwhile, the breakthrough ability of PH₃ was defined in the equation, and the performance of the adsorbent was evaluated more objectively. Each test was repeated at least twice for each adsorbent to ensure the reliability of the data collected.

Calculation formula of PH₃ removal efficiency (Re):

$$\mathrm{RE} (\%)={\displaystyle \frac{{({\mathrm{P}\mathrm{H}}_3)}_{\mathrm{i}\mathrm{n}}-{\left({\mathrm{P}\mathrm{H}}_3\right)}_{\mathrm{o}\mathrm{u}\mathrm{t}}}{{\left({\mathrm{P}\mathrm{H}}_3\right)}_{\mathrm{i}\mathrm{n}}}}\times 100\%$$

(1)

Calculation formula of PH₃ breakthrough capacity (BC):

$$\mathrm{BC} (\mathrm{mg}·{\mathrm{g}}^{-1})={\displaystyle \frac{\mathrm{Q}{\int }_0^{\mathrm{t}}[{\left({\mathrm{P}\mathrm{H}}_3\right)}_{\mathrm{i}\mathrm{n}}-{\left({\mathrm{P}\mathrm{H}}_3\right)}_{\mathrm{o}\mathrm{u}\mathrm{t}}]\mathrm{d}\mathrm{t}}{\mathrm{m}}}$$

(2)

where (PH₃)_in/(PH₃)_out is the phosphine concentration (ppm) in the inlet/outlet gas flow through the reaction system, m is the mass of the adsorbent involved in each experiment (g), t is the reaction time (min) at the breakthrough point of phosphine (1 ppm), and Q is the total flow rate (mL·min⁻¹).

2.3. Characterization Methods

XRD (X-ray diffraction analysis), SEM (scanning electron microscopy), EDS (energy dispersive spectroscopy), TEM (transmission electron microscopy), N₂ adsorption-desorption isotherms, TG (thermogravimetry), XPS (X-ray photoelectron energy spectrum), H₂-TPR (hydrogen temperature-programmed reduction), CO₂-TPD (CO₂ temperature-programmed desorption), and EPR (electron paramagnetic resonance) were used to reveal the structural characteristics of the adsorbent. Specific details of the characterization methods are shown in the Supplementary Materials.

3. Results and Discussion

3.1. Characterization of Adsorbents

To investigate the effect of high-temperature calcination on the morphology of adsorbents, SEM was employed for characterization. Figure 1 presents the SEM and EDS mapping images of MOF-199 and the CuO derived from it. As illustrated in Figure 1a, the MOF-199 exhibits a typical octahedral structure with a smooth and regular surface [36]. The EDS images indicate that the C, Cu, and O elements in the adsorbent are evenly distributed across the surface. MOF-199 is formed by coordinating a Cu dimer with four trimeric acid ligands, which are interconnected to create a “cage-channel” structure [36,37,38]. This unique arrangement facilitates the orderly and uniform distribution of elements within the material. In contrast, as shown in Figure 1b, the CuO derived from the MOF has significantly changed compared to that of MOF-199. After high-temperature calcination, the octahedral structure transforms into spherical particles with smaller sizes and rough surfaces, likely due to the collapse of the MOF material’s skeleton, resulting in the formation of aggregated CuO and an increase in porosity. The EDS images reveal a significant decrease in C content following high-temperature calcination, which may be attributed to the reaction between C and O₂ in the air, leading to the conversion of C into gaseous products that detach from the material.

To further investigate the effect of Ce doping on the morphology of the adsorbent, Cu₁Ce_0.14O, Cu₁Ce_0.2O, Cu₁Ce₁O, and CeO₂ were characterized using SEM. As shown in Figure 1c, CeO₂ exhibits an irregular blocky morphology with a rough and porous surface. In Figure 1d, the morphology of Cu₁Ce_0.2O has significantly changed compared to the MOF-derived CuO. Unlike the previous spherical particles, the Cu₁Ce_0.2O displays an irregular bulk morphology after Ce doping, in which the surface is porous and the particle size is larger than that of MOF-derived CuO. This indicates that Ce doping effectively alters the morphology of the adsorbent, imparting some morphological characteristics of CeO₂. EDS-mapping images reveal that, compared to MOF-derived CuO, the Cu and Ce elements are uniformly dispersed across the surface of Cu₁Ce_0.2O, indicating successful Ce doping and improved dispersion of CuO. As shown in Figure 1e,f, the SEM images of Cu₁Ce_0.14O and Cu₁Ce₁O also appear as irregular blocks. However, EDS-mapping images indicate that, compared to Cu₁Ce_0.2O, there is significant agglomeration of Cu on the surfaces of Cu₁Ce_0.14O and Cu₁Ce₁O, suggesting that the amount of Ce doping affects the dispersion of CuO, which may influence the performance of the adsorbent.

Since the morphology of the adsorbent changed significantly after Ce doping, the physical properties were likely affected. Hence, N₂ adsorption–desorption isotherms were employed to determine the pore structure characteristics of adsorbents, with the results presented in

Table 1. Textural properties of the prepared adsorbents.

Samples	SBET (m2·g−1)	Pore Volume (mL·g−1)	Average Pore Size (nm)
MOF-199	1109.41	0.5980	1.078
MOF-derived CuO	6.51	0.0105	3.234
Cu1Ce0.14O	20.37	0.0374	3.672
Cu1Ce0.2O	28.01	0.0579	4.136
Cu1Ce1O	70.20	0.0980	2.791
exhausted Cu1Ce0.2O	13.82	0.0232	3.360
CeO2	77.48	0.2195	5.667

. MOF-199 exhibits a very large specific surface area and pore volume due to its unique structure, which is consistent with the typical characteristics of MOF materials. A large specific surface area typically indicates better adsorption properties. However, excellent redox properties are also crucial for adsorbents to effectively remove PH₃. After high-temperature calcining, the specific surface area and pore volume of MOF-derived CuO decreased significantly, indicating the collapse of the MOF structure. In contrast, the specific surface area and pore volume of Cu₁Ce_0.14O, Cu₁Ce_0.2O, and Cu₁Ce₁O are all larger than those of MOF-derived CuO, suggesting that Ce doping can effectively enhance the specific surface area and pore volume of the adsorbent. Moreover, both the specific surface area and pore volume increase with the amount of Ce doping. Figure S1 presents the N₂ adsorption isotherms and pore size distribution curves for several adsorbents. The adsorption isotherm of MOF-199 is classified as Type I, indicating that it is a microporous material, which is corroborated by the corresponding pore size distribution curve. In contrast, the adsorption isotherms of MOF-derived CuO and Cu₁Ce_0.2O are classified as Type II, with pore size distributions predominantly in the mesoporous range. This suggests that after high-temperature calcining, the adsorbents transformed into mesoporous materials due to the collapse of the MOF structure.

To investigate the composition and crystalline phases of all samples, the prepared sorbent was analyzed using XRD. As shown in Figure 2a, the characteristic diffraction peaks of MOF-199 at 2θ = 6.7°, 9.4°, 11.6°, 13.4°, 17.4°, 18.9° are consistent with those reported in the literature [36], indicating that MOF-199 was successfully synthesized. After high-temperature calcination, the characteristic peaks of MOF-199 disappeared, and a series of characteristic diffraction peaks belonging to CuO were observed in the XRD pattern of MOF-derived CuO [23,39,40,41] (2θ = 32.5°, 35.5°, 38.7°, 48.7°, 61.5°). This result indicates that high-temperature calcination led to the collapse of the MOF structure and the formation of CuO, consistent with the SEM results. However, compared to MOF-derived CuO, the XRD pattern of Cu₁Ce_0.2O did not change significantly, indicating that Ce doping did not substantially affect the formation of CuO crystalline phases. Meanwhile, characteristic peaks corresponding to CeO₂ were observed, confirming the successful Ce doping. Additionally, the characteristic peaks of CuO were significantly weakened, suggesting that Ce doping enhances the dispersion of CuO, which may favor the adsorption–oxidation of PH₃. Figure S2 shows the XRD patterns of adsorbents with varying Ce doping amounts (Cu₁Ce_0.14O, Cu₁Ce_0.2O, Cu₁Ce₁O). It can be observed that as the Ce doping amount increases, the intensity of the CuO characteristic peak gradually decreases. The decreased intensity of the CuO characteristic peak may indicate higher dispersion of CuO, which could be beneficial for the adsorption–oxidation of PH₃. However, the intensity of the CuO characteristic peak is significantly weakened in Cu₁Ce₁O, likely due to an overproduction of CeO₂, which may obscure CuO and affect the adsorption–oxidation performance of the adsorbent. To explore the changes occurring during calcination in the preparation of the adsorbent, TG analysis of MOF-derived CuO and Cu₁Ce_0.2O is shown in Figure 2b. Two distinct weight loss stages can be observed in the thermal weight curve during the heating process. The first phase (30–200 °C) can be attributed to the desorption and volatilization of adsorbed water, while the second phase (210–400 °C) corresponds to the decomposition of the organic framework and the formation of the active component CuO, resulting in a significant loss of carbon, consistent with previous EDS mapping results. Additionally, it can be observed that Cu₁Ce_0.2O exhibits significant weight loss at 200 °C compared to MOF-derived CuO, which may be attributed to the decomposition of Ce(NO₃)₃·6H₂O, resulting in the formation of CeO₂. The samples continued to lose mass gradually as the temperature increased, likely due to further carbon loss.

To obtain the chemical composition and valence state information of the adsorbent, XPS was employed for analysis. Figure 2c presents the O 1s spectra of MOF-derived CuO, Cu₁Ce_0.14O, Cu₁Ce_0.2O, and Cu₁Ce₁O. The peaks in the O 1s spectra at 530.0 eV, 531.8 eV, and 532.9 eV correspond to lattice oxygen [42,43] (O_latt), adsorbed oxygen (O_ads), and hydroxyl oxygen (O_-OH), respectively.

Table 2. The relative content of lattice oxygen (O_latt), adsorbed oxygen (O_ads), and hydroxy oxygen (O_-OH) of MOF-derived CuO, Cu₁Ce_0.14O, Cu₁Ce_0.2O, and Cu₁Ce₁O.

Samples	Oads (%)	Olatt (%)	O-OH (%)
MOF-derived CuO	18.87	60.34	20.79
Cu1Ce0.14O	21.05	65.45	13.50
Cu1Ce0.2O	25.71	57.99	16.30
Cu1Ce1O	19.83	64.16	16.01

displays the relative content of O_latt, O_ads, and O_-OH for each adsorbent. Compared to MOF-derived CuO, Cu₁Ce_0.14O, Cu₁Ce_0.2O, and Cu₁Ce₁O exhibit higher adsorbed oxygen content, indicating that Ce doping enhances the formation of adsorbed oxygen. The presence of adsorbed oxygen, which is more active in redox reactions [43,44], may facilitate the adsorption–oxidation of PH₃. Furthermore, an increased amount of adsorbed oxygen typically signifies a greater number of oxygen vacancies on the material’s surface. This suggests that Ce doping may lead to an increase in oxygen vacancies, thereby enhancing the adsorbed oxygen content. Additionally, Figure 2c indicates that Ce doping shifts the O peak to a lower binding energy (529.8 eV), likely due to the interaction between Cu and Ce [32,45,46]. To further demonstrate that Ce doping resulted in increased oxygen vacancies, EPR was employed to analyze the oxygen vacancy of MOF-derived CuO and Cu₁Ce_0.2O. As shown in Figure 2d, no signal was detected in MOF-derived CuO, while a distinct signal corresponding to oxygen vacancies was observed in Cu₁Ce_0.2O at a g value of approximately 2.07 [47]. This finding indicates that Ce doping significantly increased the number of oxygen vacancies on the adsorbent surface. Previous studies have established that the unique fluorite structure and excellent oxygen storage and release capacity of CeO₂ facilitate the formation of oxygen vacancies [46,47].

Additionally, PH₃ is an acidic gas that is more readily captured by adsorbents with stronger basicity. To investigate the effect of Ce doping on the intensity and quantity of basic sites, CO₂-TPD was utilized to analyze MOF-derived CuO, Cu₁Ce_0.2O, and CeO₂. As illustrated in Figure 2e, Ce doping had a substantial impact on the basic sites. Generally, the desorption peak area and temperature in the CO₂-TPD profile correspond to the number and intensity of basic sites, respectively. No basic sites were observed in the MOF-derived CuO. In contrast, the Cu₁Ce_0.2O exhibited a distinct CO₂ desorption peak at 122 °C, indicating the presence of weak basic sites [48]. Furthermore, the CO₂-TPD curves of CeO₂ revealed two distinct CO₂ release peaks at 130 °C and 460 °C, corresponding to weak and strong basic sites, respectively [49]. This suggests that Ce doping enhances the generation of basic sites and increases their strength in Cu₁Ce_0.2O compared to MOF-derived CuO. These basic sites not only increase the basicity of the adsorbent but also serve as active sites, potentially enhancing the adsorption and activity towards PH₃. In the adsorption–oxidation of PH₃, oxidation reaction may occur, making the redox properties of the adsorbent crucial. To explore the redox properties of the adsorbents, MOF-derived CuO and Cu₁Ce_0.2O were analyzed using H₂-TPR. The results, shown in Figure 2f, indicate that a reduction peak is present in the H₂-TPR curves of both MOF-derived CuO and Cu₁Ce_0.2O, attributed to the reduction of CuO, with peak positions at 235 °C and 179 °C, respectively. The lower peak position suggests that Cu₁Ce_0.2O exhibits superior redox properties, demonstrating stronger oxidation capabilities and a more effective performance in removing the reducing gas PH₃ [50,51].

3.2. Adsorbents Performance

All prepared adsorbents were subjected to PH₃ adsorption–oxidation experiments at 90 °C, with the test results presented in Figure 3. The results indicate that MOF-199 can maintain complete removal of PH₃ for approximately 30 min, corresponding to a breakthrough capacity of 17.69 mg·g⁻¹. This finding demonstrates that specific surface area is not the most critical factor for PH₃ removal. In contrast, the activity of MOF-derived CuO was significantly enhanced compared to MOF-199, with a breakthrough capacity increasing to 75.54 mg·g⁻¹. This suggests that utilizing MOF material as a precursor to obtain CuO particles is a viable strategy. After Ce doping, the activity of the adsorbent was significantly enhanced, but varying amounts of Ce doping can lead to substantial differences in activity. The breakthrough time of Cu₁Ce_0.14O reached 270 min, corresponding to a breakthrough capacity of 136.53 mg·g⁻¹. The performance of the adsorbent increased with the amount of Ce doping until Cu₁Ce_0.2O exhibited the highest performance, with a breakthrough time of 420 min and a breakthrough capacity of 226.87 mg·g⁻¹. However, the continuous increase in Ce doping amount led to a decline in adsorbent performance, as evidenced by the breakthrough capacity of Cu₁Ce₁O, which reached only 121.31 mg·g⁻¹. As evidenced by EDS mapping (Figure 1f) and quantitative analysis in Table 2, excessive Ce doping significantly reduced both the dispersion degree of CuO nanoparticles and the relative concentration of adsorbed oxygen species, thereby leading to a marked decline in PH₃ adsorption–oxidation capacity. Previous characterization results indicate that Cu₁Ce_0.2O possesses optimal CuO dispersion, the highest adsorbed oxygen concentration, superior redox properties, an increased number of basic sites, and a larger specific surface area and pore volume. These characteristics are speculated to contribute to its enhanced performance. Additionally, it is evident that CeO₂ exhibits negligible adsorption activity towards PH₃, indicating that the primary active component of the adsorbent is CuO. Figure S3 shows the PH₃ adsorption–oxidation performance of Cu₁Ce_0.2O prepared at different calcination temperatures. As the calcination temperature increases, the adsorbent’s performance initially improves and then declines. As mentioned previously, Figure 2b shows that the samples progressively lose mass with increasing temperature, likely due to additional carbon loss. Therefore, it is speculated that the residual carbon may provide a supporting role. As the calcination temperature increases, excessive carbon loss leads to a pronounced collapse of the adsorbent’s structure, which consequently diminishes its performance. To investigate the effect of PH₃ concentrations, humidity, and oxygen levels on the adsorption–oxidation performance of Cu₁Ce_0.2O for PH₃, a series experiments were carried out. Figures S4–S6 show the adsorption–oxidation performance of Cu₁Ce_0.2O for PH₃ at varying PH₃ concentrations, humidity, and oxygen levels, respectively. It can be seen from Figure S4 that as the PH₃ concentration decreased, the adsorbent’s performance improved, attributed to the active sites on the adsorbent being able to fully contact and completely react with the PH₃ molecules. As shown in Figure S5 (RH corresponds to the relative humidity), higher humidity can enhance the performance of the adsorbent, this may be because the appropriate humidity leads to the enhancement of surface hydroxyl activity and promotes the production of reactive oxygen species. According to Figure S6, the presence of trace oxygen facilitates the reaction process, while excessive oxygen leads to a decline in the performance of the adsorbent; the cause may be that excessive oxygen changes the selectivity of the reaction, leading to the formation of excessive H₃PO₄ and covering the active site of the adsorbent.

3.3. Mechanism of PH₃ Adsorption–Oxidation over Cu₁Ce_0.2O

Ce doping significantly enhanced the adsorption–oxidation performance of MOF-derived CuO. However, continuous dephosphorization leads to the accumulation of oxidation products, which deactivates the adsorbent. A series of characterizations were conducted to investigate the reaction products and mechanisms of PH₃ adsorption–oxidation over Cu₁Ce_0.2O. As shown in Figure 4a, XRD patterning compares the fresh and exhausted Cu₁Ce_0.2O adsorbent. The peak intensity corresponding to CuO in the exhausted adsorbent is significantly decreased, indicating that CuO is the main active component in the adsorption–oxidation of PH₃ and that a substantial amount of CuO was consumed during the reaction. Additionally, the characteristic diffraction peaks of Cu₃P (2θ = 41.6°, 45.1°, 46.2°) were observed, confirming that a redox reaction occurred between CuO and PH₃ and Cu₃P was formed [52]. Figure S3 presents a TEM image of the exhausted adsorbent, revealing a lattice spacing of approximately 0.20 nm, which corresponds to crystalline Cu₃P and is consistent with the XRD results [53].

The chemical composition and elemental valence states of fresh and exhausted adsorbents were confirmed using XPS. Figure 4b presents the Cu 2p spectra for the fresh and exhausted adsorbents. For the fresh adsorbent, distinct signal peaks were detected at 933.8 eV and 953.8 eV, corresponding to the binding energy of Cu²⁺ (CuO), while the peak at 942.7 eV represents the strong satellite peak of Cu²⁺ [54]. In contrast, the exhausted adsorbent exhibited new signal peaks at 933.1 eV and 952.9 eV, which correspond to the binding energy of Cu⁺ [55], presumed to be associated with Cu₃P [56]. This observation is consistent with the XRD and TEM results, further confirming that CuO participated in the reaction and transformed into Cu₃P. Figure 4c displays the P 2p spectra for the exhausted adsorbent, revealing two signal peaks at 133.7 eV and 129.6 eV. The peak at 133.7 eV corresponds to P⁵⁺ (P₂O₅ or H₃PO₄) [50], while the peak at 129.6 eV corresponds to P³⁻ (Cu₃P), further confirming the presence of Cu₃P [56]. The detection of P₂O₅ or H₃PO₄ in the products indicates that, in addition to chemisorption, oxidation of PH₃ also occurred. As shown in Figure 4d, the O 1s spectra for fresh and exhausted adsorbents were analyzed. For the exhausted adsorbent, the signal peak corresponding to lattice oxygen (O_latt) was nearly absent, indicating significant consumption of CuO. Figure 4e presents the Ce 3d spectra for both fresh and exhausted adsorbents, revealing seven signal peaks in the fresh adsorbent. The peaks at 882.9 eV, 889.4 eV, 898.7 eV, and 907.9 eV were attributed to Ce³⁺, while those at 886.1 eV, 901.3 eV, and 917.2 eV corresponded to Ce⁴⁺ [57]. After the reaction, the relative contents of Ce³⁺ and Ce⁴⁺ changed significantly, with an increase in the relative content of Ce³⁺, suggesting that some Ce⁴⁺ was reduced to Ce³⁺ during the reaction. This transformation can be attributed to the interaction between Cu and Ce. Previous studies have shown that Ce⁴⁺ and Cu⁺ at the interface of CuO and CeO₂ can be converted into Ce³⁺ and Cu²⁺, leading to the generation of more oxygen vacancies and increased adsorbed oxygen [32], which facilitates the reaction. To further demonstrate the presence of the CuO and CeO₂ interface, the TEM image of the fresh adsorbent is shown in Figure 4f, where lattice stripe spacings of approximately 0.23 nm and 0.31 nm correspond to CuO and CeO₂ [39,57], respectively, confirming the existence of the CuO and CeO₂ interface.

Based on the experiments and characterization results presented above, a possible mechanism of the PH₃ adsorption–oxidation process over the Cu₁Ce_0.2O was proposed and illustrated in Figure 5. A series of reactions occurred between PH₃ and the Cu₁Ce_0.2O during the adsorption–oxidation process, resulting in the formation of Cu₃P and H₃PO₄ or P₂O₅. Therefore, the chemical equation for the adsorption–oxidation process is inferred as follows:

PH₃ + 3CuO + 3e⁻ → Cu₃P + 3H⁺ + 3O²⁻

(3)

$$2{\mathrm{PH}}_3+3{\mathrm{O}}_2 \stackrel{\mathrm{C}\mathrm{u}\mathrm{O}}{\to }{\mathrm{P}}_2{\mathrm{O}}_5+6{\mathrm{H}}^{+}{\mathrm{O}}^{2-}+4{\mathrm{e}}^{-}$$

(4)

P₂O₅ + 3H₂O → 2H₃PO₄

(5)

It is hypothesized that the adsorption–oxidation of PH₃ by the adsorbents involves both catalytic oxidation and chemisorption. Additionally, as indicated by the results in Table 1, the specific surface area and pore volume of Cu₁Ce_0.2O decreased after the adsorption–oxidation of PH₃. This decrease is attributed to the accumulation of products and the blockage of pore diameters. Once the active component CuO is fully consumed, or if the products cover the surface and obstruct the pore structure, preventing PH₃ from contacting the active component, the adsorbent becomes deactivated.

3.4. Recycling of the Exhausted Adsorbents

The reaction product is highly crystalline Cu₃P, a typical P-type semiconductor that may be utilized for the resource conversion of exhausted adsorbents. According to the literature, Cu₃P is frequently employed in photocatalytic reactions, such as the degradation of organic pollutants and the decomposition of aquatic hydrogen [58,59]. Therefore, exhausted Cu₁Ce_0.2O may be reused in photocatalysis. Rhodamine B (RhB) is a common organic pollutant in the printing and dyeing industry, posing significant threats to animals, plants, and human health. It is challenging to degrade naturally, necessitating treatment [60,61]. Various methods are available for its remediation, among which photocatalytic degradation of RhB is a promising approach. Consequently, an experiment to test the photocatalytic degradation activity of the deactivated adsorbents was designed. Initially, 30 mg of catalyst was placed in 50 mL of an aqueous Rhodamine B solution (50 mg/L), achieving adsorption equilibrium after 30 min in a dark environment. The photocatalytic degradation reaction was then conducted under visible light, with the concentration of RhB in the solution measured every 15 min. The experimental results are presented in Figure 6. After 30 min of dark adsorption, all materials adsorbed approximately 5% to 10% of RhB. Following visible light irradiation, the exhausted adsorbent exhibited significantly enhanced photocatalytic degradation of RhB, with degradation reaching about 80% after 180 min, while the fresh adsorbent achieved only about 15% degradation. This indicates that the fresh adsorbent lacks photocatalytic degradation capability, whereas Cu₃P in the exhausted adsorbent effectively facilitates photocatalytic degradation. Additionally, a series of experiments were conducted on the catalytic degradation of RhB under dark conditions using the exhausted adsorbent. The results, shown in Figure S4, indicated that the degradation rate of RhB was 40% lower than under visible light conditions, confirming the photocatalytic activity of this material. The photocatalytic degradation performance of different catalysts toward RhB are summarized in

Table 3. The photocatalytic degradation performance of different catalysts toward RhB.

Samples	Experimental Condition	Time (min)	Removal (%)	Ref.
Fe3O4/PPE-2	ultraviolet light, 50 mg catalyst, RhB 50 mg·L−1	160	98.2	[62]
PTA@MIL-101 (Cr)	visible light, 30 mg catalyst, RhB 30 mg·L−1	180	97.81	[63]
Iodine-doped TiO2 nanoparticles	visible light, 100 mg catalyst, RhB 50 mg·L−1	140	82.36	[64]
Sn/SnO2	visible light, 50 mg catalyst, RhB 10 mg·L−1	300	90	[65]
Exhausted Cu1Ce0.2O	visible light, 30 mg catalyst, RhB 50 mg·L−1	180	80	this work

. Considering the experimental conditions, the exhausted Cu₁Ce_0.2O exhibited commendable photocatalytic degradation activity. To investigate the photocatalytic stability of exhausted adsorbent, Figure S9 shows the XRD pattern of exhausted adsorbent after RhB catalytic degradation (designated as exhausted adsorbent-AR). Compared with exhausted adsorbent, the composition of the exhausted adsorbent-AR remained unchanged. Figure S10 illustrates the cyclic performance of the exhausted adsorbent in the catalytic degradation of RhB under visible light conditions. After one cycle, the exhausted adsorbent maintained its performance, indicating excellent stability. However, after two and three cycles, some of the adsorbent was lost, which resulted in a reduction of the photocatalytic degradation performance to 60%. These experimental findings demonstrate that exhausted adsorbents possess commendable photocatalytic activity and can be reused effectively. In a word, the application of exhausted adsorbent in photocatalytic degradation not only achieves resource recovery from waste but also aligns with the principles of a circular economy.

4. Conclusions

In conclusion, Cu₁Ce_0.2O adsorbent was synthesized using MOF-199 as the precursor and applied to the adsorption–oxidation of PH₃ under low-temperature and low-oxygen conditions. Due to its optimal CuO dispersion, high adsorbed oxygen concentration, superior redox performance, abundant basic sites, and well-developed pore structure, Cu₁Ce_0.2O exhibited a PH₃ capacity of up to 226.87 mg·g⁻¹. Furthermore, experimental results and a series of characterizations indicated that adsorption and oxidation reactions occurred simultaneously. CuO chemically adsorbed PH₃, resulting in the formation of high-value p-type semiconductor Cu₃P. Concurrently, O₂ was catalyzed by CuO to undergo a redox reaction with PH₃, producing P₂O₅ and H₃PO₄. Additionally, the exhausted adsorbent demonstrates excellent photocatalytic performance and significant potential for recycling.