The Effect of H₂O₂ Pretreatment on TiO₂-Supported Ruthenium Catalysts for the Gas Phase Catalytic Combustion of Dichloromethane (CH₂Cl₂)

Abstract

In the study presented herein, a series of TiO₂-supported Ru catalysts were prepared through over-impregnation with different amounts of H₂O₂ and applied to the catalytic combustion of dichloromethane (DCM). The experimental results show that the optimal (1 H₂O₂)-Ru@TiO₂ catalyst sample yields 90% DCM conversion at 301 °C, which is 40 °C lower than the temperature required by the (0 H₂O₂)-Ru@TiO₂ catalyst. This good activity is related to its high number of acid sites (especially Brønsted acid sites), an appropriate amount of uniformly dispersed RuO₂ particles, and strong interaction between RuO₂ and the TiO₂ support. In addition, excessive H₂O₂ leads to the growth and phase segregation of RuO₂, which weakens the interaction between RuO₂ and the TiO₂ support and decreases catalytic activity. Moreover, H₂O₂ addition also contributes to the stability of catalysts, which possibly results from the re-dispersion of RuO₂ on the TiO₂ support during the reaction.

1. Introduction

Chlorinated volatile organic compounds (CVOCs) are widely used in industry as solvents and degreasing and cleaning agents; however, they pose considerable harm to human health and the environment [1,2,3]. Catalytic oxidation is considered an advantageous technique to eliminate CVOCs due to its high conversion rate, low ignition temperature, low operating energy consumption, and limited number of by-products [4,5].

The catalysts used for catalytic oxidation mainly consist of noble metal catalysts (e.g., Pt, Pd, and Ru) [6,7,8] and transition metal oxides (e.g., CeO₂-ZrO₂, Cr₂O₃, and MnO₂) [9,10,11,12]. Supported Pt or Pd catalysts demonstrate relatively higher catalytic activity than transition metal oxides; however, they are expensive and sensitive to high temperatures and chlorine poisoning. Transition metal oxide catalysts are relatively inexpensive and have good resistance to chlorine poisoning; however, their low catalytic activity and poor selectivity limit their wider application [13]. In contrast, supported Ru catalysts have been demonstrated to be highly active and stable components for CVOCs oxidation due to their excellent catalytic dechlorination activity via the Deacon pathway [14].

TiO₂, ZnO [15], CeO₂ [16,17], WO₃ [18], and LaMnO₃ [19] are often employed as catalyst supports. Among these, TiO₂ is one of the classical acid catalysts, which is considered an important factor for the degradation of short-chain CVOCs (e.g., DCM) due to its high surface area, large number of pores for the dispersal of active particles, and remarkable interaction between group 8 noble metals [6,20]. At present, researchers mainly focus their attention on the effect of catalyst support behavior (e.g., the hierarchical structure of the support [13] and chemical modification of the support [21]) on the performance of catalysts. However, the behavior of active species [22], such as the valence and dispersion as reported in several works [23,24,25], also plays an important role in catalytic activity [26]. Ru species are considered as the active sites in the catalytic combustion of CVOCs [27], and H₂O₂ is a green and moderate oxidant used to change the valence of Ru, which is also beneficial to the catalyst’s surface dispersion [28]. In the following paper, we propose a simple and feasible method to oxidize Ru species through the addition of different amounts of H₂O₂, aiming to prepare catalysts with varying behavior of Ru species. Moreover, the structural and chemical properties of these catalysts were characterized using XRD, BET, TEM, XPS, H₂-TPR, NH₃-TPD, and pyridine-IR methods to determine the effects of the morphology, acidity, and valence of Ru species and interaction between RuO₂ species and TiO₂ on catalytic activities.

2. Results and Discussion

2.1. Catalyst Analysis

Figure 1 presents the XRD diffraction patterns of the (0, 1, and 3 H₂O₂)-Ru@TiO₂ catalysts, where 2θ = 25.3°, 37.8°, 48.1°, 53.9°, 55.1°, and 62.7° represent the diffraction peak of the typical anatase phases (JCPDS No. 21-1272) and 2θ = 27.9° and 35.1° are assigned to RuO₂ (JCPDS No. 40-1290). All samples show almost identical XRD diffraction patterns, indicating that all catalysts maintain the initial support TiO₂ crystal structure (the raw material P25, mostly anatase TiO₂) after Ru impregnation and H₂O₂ treatment. The Ru content of (0, 1, and 3 H₂O₂)-Ru@TiO₂ catalysts measured by ICP-OES are 0.28 wt%, 0.26 wt%, and 0.26 wt% (see

Table 1. Different properties of P25 and catalysts.

Samples	Ru Content/wt%	RuO2 Particle Size/nm	Specific Surface Area /m2·g−1
Initial support TiO2 (P25)	—	—	58.5
(0 H2O2)-Ru@TiO2 catalyst	0.28	6.4	49.1
(1 H2O2)-Ru@TiO2 catalyst	0.26	7.0	49.7
(3 H2O2)-Ru@TiO2 catalyst	0.26	8.5	48.8

), respectively, which are close to the theoretical value. The specific surface area of P25 is 58.5 m²·g⁻¹, and all three catalysts exhibit a slight decrease. Regardless of whether H₂O₂ is added, all samples have extremely poor XRD peaks for RuO₂, owing to the low loading amount of Ru. A further analysis revealed that the RuO₂ crystallite sizes of the (0, 1, and 3 H₂O₂)-Ru@TiO₂ catalysts are 6.4, 7.0, and 8.5 nm (see Table 1), respectively, as calculated using Scherrer’s equation (see Table S1). Using the above results, it was possible to prove that H₂O₂ treatment promotes the growth of RuO₂ crystallite, especially the (3 H₂O₂)-Ru@TiO₂ catalyst when the amount of H₂O₂ present is excessive.

The HRTEM images of different samples are provided in Figure 2. The morphologies of the (0 and 1 H₂O₂)-Ru@TiO₂ catalysts are almost identical, as shown in Figure 2a–d, where their mainly exposed lattice fringe at around 0.35 nm corresponds to anatase (1 0 1) facets and at 0.24 nm corresponds to anatase (0 0 4) facets [29]. Moreover, the EDS mapping images illustrate that the Ru species of the (1 H₂O₂)-Ru@TiO₂ catalyst disperse more uniformly than those of the (0 H₂O₂)-Ru@TiO₂ catalyst (Figure S1a,b), which might result from the formation of RuO₂ nanoparticles during oxidation with H₂O₂.

With the increased addition of excess H₂O₂, a new structure of RuO₂ nanorods emerges in the (3 H₂O₂)-Ru@TiO₂ catalyst, where their mainly exposed lattice fringe at around 0.32 nm corresponds to RuO₂ (1 1 0) facets [30] (Figure 2e,f). The EDS mapping images (Figure S1c) indicate that the nanorods were enriched in Ru and O species but not Ti species, which also proves that the nanorods comprised RuO₂. Furthermore, the XRD analysis results also proved the growth of RuO₂ nanorods, and this growth was found to eventually cause the phase segregation of RuO₂ at higher H₂O₂ concentrations.

The XPS results for Ti 2p provided in Figure 3a show two main peaks centered at 458.5–458.8 eV and 464.6–465.0 eV, which could be ascribed to the typical Ti⁴⁺ oxidation state in a tetragonal structure [21]. It can be clearly seen that the binding energy shifts to a lower energy level after H₂O₂ treatment for the (1 and 3 H₂O₂)-Ru@TiO₂ samples compared with the (0 H₂O₂)-Ru@TiO₂ catalyst. This binding energy shift indicates the chemical attachment of the Ru species to the TiO₂ surface, probably resulting in a changed electron atmosphere and stronger interaction with the support following H₂O₂ treatment [31,32,33].

Figure 3b shows the Ru 3d5/2 XPS profiles of the studied catalysts. The peaks centered at 279.8 and 279.9 eV can be attributed to Ru⁰ [21]. The second peak at around 208.3 eV is assigned to RuO_x species. The third peak centered at 281.0 and 281.2 eV is ascribed to RuO₂ [34]. All species of Ru exist in each catalyst; however, the amount of RuO₂ in the (0 H₂O₂)-Ru@TiO₂ catalyst is significantly lower than that in the (1 and 3 H₂O₂)-Ru@TiO₂ catalysts (see

Table 2. The content of different Ru species from fresh and spent samples analyzed by XPS.

Samples	Fresh Catalysts			Spent Catalysts
	Ru0	RuOx	RuO2	Ru0	RuOx	RuO2
(0 H2O2)-Ru@TiO2 catalyst	41.72%	30.79%	27.49%	32.60%	47.76%	19.64%
(1 H2O2)-Ru@TiO2 catalyst	28.42%	34.99%	36.59%	29.03%	38.66%	32.31%
(3 H2O2)-Ru@TiO2 catalyst	29.82%	29.65%	40.53%	27.46%	33.85%	38.69%

), which indicates that H₂O₂ treatment could oxidize Ru³⁺ species to a higher valence state of RuO₂ and prevent the generation of excess Ru⁰ following heating. Furthermore, a greater amount of RuO₂ was observed in the (3 H₂O₂)-Ru@TiO₂ catalyst compared to the (1 H₂O₂)-Ru@TiO₂ catalyst, which also proves the growth and phase segregation of RuO₂ in accordance with the XRD and TEM results.

As the O1s XPS results displayed in Figure 3c show, the peaks centered at around 529.6 eV can be assigned to oxygen ions in the crystal lattice (O²⁻), and the peaks presented at 531.2 and 531.7 eV can be ascribed to adsorbed oxygen species (O⁻/O₂²⁻) [35]. The amount of adsorbed oxygen species for the (1 and 3 H₂O₂)-Ru@TiO₂ catalysts is higher than that of the (0 H₂O₂)-Ru@TiO₂ catalyst (see Table S2), suggesting that the H₂O₂ treatment of RuO₂ increases the adsorption of hydroxyl groups on the surface of TiO₂, which might create abundant active oxygen species to improve catalytic activity during the progression of the oxidation reaction [36,37].

The H₂-TPR profiles of the catalysts are shown in Figure 3d. H₂-TPR profiles are mainly used to obtain the interaction information between metal oxides with carriers and to analyze the content of different species of metal oxides. The reduction peak between 150–250 °C and 300–450 °C can be attributed to the reduction in RuO₂ species interacting with the TiO₂ support weakly and strongly, respectively [38]. The reduction temperature increased for the (1 H₂O₂)-Ru@TiO₂ catalyst compared to the (0 H₂O₂)-Ru@TiO₂ sample, implying that the interaction between RuO₂ species and the TiO₂ support increased, which corresponds to the XPS analysis results for Ti 2p. However, the reduction temperature decreased slightly for the (3 H₂O₂)-Ru@TiO₂ catalyst compared to the (1 H₂O₂)-Ru@TiO₂ sample, indicating that interaction between RuO₂ species and the TiO₂ support decreased. It can be inferred that the phase segregation of RuO₂ nanorods, as shown in the TEM images, results in weakened interaction.

The total acid intensity of the catalysts was evaluated using NH₃-TPD (see Figure 4a). The peaks at 120–300 °C and 300–400 °C are attributed to NH₃ desorption from weak and moderate acidic sites, respectively. The NH₃ desorption peaks of the (0 H₂O₂)-Ru@TiO₂ catalyst are observed at 188 °C and 351 °C, which are assigned to weak and moderate acidic sites, respectively. With the increased addition of H₂O₂, the weak acidic site disappears and the moderate acidic site shifts to a higher temperature. The peaks of the (1 H₂O₂)-Ru@TiO₂ and (3 H₂O₂)-Ru@TiO₂ catalysts were found to shift to 366 °C and 375 °C, respectively, suggesting that the catalysts have an increased number of moderate acidic sites.

The pyridine-IR spectra of catalysts with different H₂O₂ additions are used to distinguish Brønsted and Lewis acids (see Figure 4b). The peaks at 1606 cm⁻¹ are ascribed to the physical adsorption of pyridine. The bands at 1445 and 1540 cm⁻¹ are assigned to Lewis and Brønsted acid sites [21], respectively, and the information on the percentage of different acid sites on the catalysts is also presented in Figure 4b. It can be seen that all catalysts contain mostly Lewis acid sites and generate more Brønsted acid sites with H₂O₂ treatment.

2.2. Catalytic Activity and Discussion

The activities of the different samples are presented in Figure 5. The activities of pure TiO₂ and pure RuO₂ samples were carried out separately, using an identical amount of TiO₂ or RuO₂ as they are present in the catalyst. Pure TiO₂ showed extremely poor activity due to its small active sites. The activity of the pure RuO₂ sample increased due to its high catalytic dechlorination activity via the Deacon pathway, reaching 90% conversion (T₉₀) at 413 °C. When TiO₂ and RuO₂ were physically mixed, the activity further improved with a T₉₀ value of 369 °C. However, the activity of this mixed sample is lower than that of the (0, 1, and 3 H₂O₂)-Ru@TiO₂ catalysts, although they have identical amounts of TiO₂ and RuO₂. Comparing the activities of the prepared catalysts, the (0 H₂O₂)-Ru@TiO₂ catalyst showed a lower activity with a T₉₀ value of 344 °C than that of the (1 and 3 H₂O₂)-Ru@TiO₂ catalysts which had T₉₀ values of 301 °C and 317 °C, respectively.

It is widely accepted that there exist two important steps during the CVOCs catalytic combustion process [39,40]: (a) the adsorption, dissociation, and oxidation of the C−Cl/C-C bond at acidic sites and (b) the desorption of products such as CO₂, H₂O, HCl, and Cl₂. The chlorine species that are dissociatively adsorbed on the active sites can be removed in the form of Cl₂ via the Deacon pathway catalyzed by RuO₂ species rather than Ru⁰ species [38,41]. The H₂O₂ treatment catalysts, the (1 and 3 H₂O₂)-Ru@TiO₂ catalysts, contain more RuO₂ species than the (0 H₂O₂)-Ru@TiO₂ catalyst, as shown in the XPS results, which contribute to the increasing the catalytic activity.

The EDS mapping images illustrate that the Ru species of the (1 H₂O₂)-Ru@TiO₂ catalyst were dispersed more uniformly than those of the (0 H₂O₂)-Ru@TiO₂ catalyst and strengthened the interaction between RuO₂ species and the TiO₂ support analyzed using their H₂-TPR profiles. It can be inferred that the increased interaction can also promote the catalytic combustion activity of DCM.

The NH₃-TPD results prove that the increased acidity of (1 H₂O₂)-Ru@TiO₂ catalyst can facilitate the adsorption and dissociation of CVOCs whilst simultaneously increasing catalytic activity, as reported [42,43]. Furthermore, the pyridine-IR characterization results illustrate that three catalysts contain a similar percentage of Lewis acid sites, where it is generally believed that the C−Cl bonds cleave and the intermediates are eventually oxidized to CO₂ [44]. However, the (1 H₂O₂)-Ru@TiO₂ catalyst generates a higher percentage of Brønsted acid sites than the (0 H₂O₂)-Ru@TiO₂ catalyst, with this process providing sufficient H protons to react with Cl, forming HCl at low temperatures and facilitating Cl desorption [44,45]. The increased number of Brønsted acid sites is conducive to accelerating the catalytic cycle whilst simultaneously increasing catalytic activity.

Moreover, the activity of the (3 H₂O₂)-Ru@TiO₂ catalyst with greater H₂O₂ addition was lower than that of the (1 H₂O₂)-Ru@TiO₂ catalyst. It can be inferred that the phase segregation of RuO₂ nanorods under excess H₂O₂ conditions might weaken the interaction between RuO₂ and the TiO₂ support and decrease the catalytic combustion activity of DCM.

2.3. Stability Tests

The stability tests of the (0, 1, and 3 H₂O₂)-Ru@TiO₂ catalysts were carried out and the results are presented in Figure 6. The catalysts were evaluated at different temperatures due to the difference in catalytic activities, which makes certain that each catalyst degrades roughly the same amount of DCM. It can be found that the DCM conversion over (1 and 3 H₂O₂)-Ru@TiO₂ catalysts only decreased by 0.2–0.3% and 0.1–0.2% after a 150 h reaction at 320 °C and 330 °C, respectively. By contrast, the (0 H₂O₂)-Ru@TiO₂ catalyst decreased 0.6–0.7% after a 150 h reaction at 360 °C, showing more obvious deactivation. These results implied that the addition of H₂O₂ contributes to the catalytic stability, which may be ascribed to the uniformly dispersed RuO₂ and strong interaction between RuO₂ and the TiO₂ support.

The spent catalysts (after 150 h reaction) are analyzed by XRD, as shown in Figure S2. The XRD patterns of spent catalysts are similar to the fresh catalysts, implying that the anatase TiO₂ cannot transfer to the rutile phase during the stability test because the reaction temperature is lower than its transition temperature.

The spent catalysts are characterized by XPS, as shown in Figure 7. The binding energies of Ti 2p change by only a little, as shown in Figure 7a, which indicates that the TiO₂ support is stable in its reaction. Figure 7b shows the O1s XPS profiles of the spent catalysts; the changes in peaks centered at 529.6 eV (assigned to oxygen ions in the crystal lattice) are negligible. The peak assigned to adsorbed oxygen species for the (0 H₂O₂)-Ru@ TiO₂ catalyst shifts to the lower binding energy form 531.7 eV to 531.1 eV, which might be related to its greater decrease during the stability test. Furthermore, the amount of adsorbed oxygen species (generally considered as surface active oxygen species) for all spent catalysts decreased compared to fresh catalysts (see Table S2), which is probably associated with the decrease in DCM conversion.

As the Ru 3d XPS results displayed in Figure 7c show, the binding energies of all species of Ru almost remain unchanged. The amount of Ru⁰ and RuO₂ in the 0 H₂O₂-Ru@TiO₂ catalyst shows a greater increase and decrease, respectively. However, this change is not obvious for the (1 and 3 H₂O₂)-Ru@TiO₂ catalysts (see Table 2). It is inferred that a part of RuO₂ is reduced to Ru⁰ during the stability test, which might cause a decrease in DCM conversion. Notably, the decrease in the amount of RuO₂ as well as the DCM conversion is at a minimum in the (3 H₂O₂)-Ru@TiO₂ catalyst during stability tests, which possibly results from the re-dispersion of RuO₂ on the TiO₂ support in the reaction [38].

3. Experiment

3.1. Catalyst Preparation

The catalysts were prepared using the over-impregnation method. A total of 0.0762 g of RuCl₃·H₂O (37.5–41.0 wt% Ru, Aladdin, Shanghai, China) was added to 80 mL deionized water, whereby the concentration of Ru was 0.0037 mol·min⁻¹. Subsequently, 10 g of TiO₂ (P25, Macklin, Shanghai, China) and different amounts of 30 wt% H₂O₂ solution (0 g, 0.0335 g, and 0.1005 g) were added to the solution separately, maintaining the molar ratio of Ru: H₂O₂ at 1:0, 1:1, and 1:3, respectively. The solution was stirred for 2 h at 95 °C and then cooled to 80 °C. Subsequently, the solutions were stirred and evaporated at 80 °C until they were completely evaporated. Finally, the obtained samples were dried in an oven at 80 °C for 12 h and calcined in a muffle furnace at 450 °C for 4 h. The resulting catalysts were denoted as (0 H₂O₂)-Ru@TiO₂, (1 H₂O₂)-Ru@TiO₂, and (3 H₂O₂)-Ru@TiO₂, respectively, and each catalyst contained a theoretical 0.30 wt% loading amount of Ru.

Moreover, three samples (“Pure RuO₂”, “Pure TiO₂”, and mixed “RuO₂ and TiO₂”) were synthetized in a similar process to catalyst preparation. For “Pure RuO₂” samples, H₂O₂ solution and 10 g of TiO₂ were not added and the other steps were exactly the same as the catalyst preparation method. Similarly, 0.0762 g of RuCl₃·H₂O was not added in the preparation of “Pure TiO₂” samples. A total of 0.015 g of pure RuO₂ and 4.985g of pure TiO₂ were ground mechanically together to obtain the sample of “mixed RuO₂ and TiO₂”.

3.2. Catalyst Characterization

The crystal phases of the samples were analyzed using an X-ray powder diffractometer (Bruker-AXS D8 Advance, Karlsruhe, Germany). The operating speed is 10 degree·min⁻¹ in the 2θ range of 5–90°. A high-resolution transmission electron microscope (HRTEM) was performed on JEM-F200 (Tokyo, Japan) to examine the morphologies of different catalysts. Energy-dispersive X-ray spectroscopy (EDS, JED-2300T, Tokyo, Japan) was applied to determine the element types on the observed region of the catalysts. The valence band of the catalysts was obtained though X-ray photoelectron spectroscopy (XPS) using a Thermal Scientific ESCALAB 250Xi instrument with an Al K_α X-ray. The Ru content of each catalyst was determined by an inductively coupled plasma optical emission spectrometer (ICP-OES, Spectro Arcos FHX22, Kleve City, Germany). The specific surface area was analyzed by N₂ adsorption on Quadrasorb-SI instrument (Quantachrome, Boynton Beach, FL, USA).

The H₂-TPR and NH₃-TPD profiles of the catalysts were measured on Auto Chem II 2920 (Micromeritics, Atlanta, GA, USA). The samples were pretreated with pure He at 300 °C for 30 min and then cooled to 30 °C. Thereafter, 10% H₂/He was applied through the reactor for the H₂-TPR experiments, and then the temperature was increased at a rate of 10 °C·min⁻¹ from 30 to 800 °C at a heating rate of 10 °C·min⁻¹. In the NH₃-TPD experiments, the sample was flushed with 5% NH₃/He for 30 min, and then treated with pure He at 100 °C for 30 min to remove the physically adsorbed NH₃. Thereafter, chemically adsorbed NH₃ was detected with the same temperature programs of H₂-TPR. The pyridine-IR spectra of the catalysts were performed on a Thermo Fisher Nicolet iS50 model Fourier transform infrared spectrometer (Waltham, MA, USA) with pyridine/2,4,6-trimethylpyridine adsorption between 1350 and 1750 cm⁻¹.

3.3. Activity and Stability Tests

Catalytic activities during DCM oxidation were carried out in a microreactor. The inner diameter and length of the reactor are 6 mm and 70 cm, respectively. A quantity of 0.2 g (0.0008 g for the sample of RuO₂ only) catalyst (40–60 mesh) mixed with quartz (0.3 g, 20–40 mesh) was loaded in the microreactor. The flow rates of DCM, N₂, and O₂ were 0.05 mL·min⁻¹, 90 mL·min⁻¹, and 10 mL·min⁻¹, respectively, where the concentration of DCM was 500 mg·L⁻¹ with GHSV 30,000 mL·g⁻¹·h⁻¹. The product was sampled online at 120 °C through a six-way valve and analyzed using a gas chromatograph (GC, 9790IIT-2, FULI, Zhejiang, China) equipped with an FID detector and a capillary column (624, 30 m × 0.32 mm ×1.0 μm, Lanzhou Donglilong Information Technology Co. Ltd., Lanzhou, China).

The conversion of DCM is calculated as follows:

$$\mathrm{DCM} \mathrm{conversion}(\%)={\displaystyle \frac{{\left[\mathrm{DCM}\right]}_{\mathrm{in}}-{\left[\mathrm{DCM}\right]}_{\mathrm{out}}}{{\left[\mathrm{DCM}\right]}_{\mathrm{in}}}}$$

where [DCM]_in and [DCM]_out are the inlet DCM concentration and the outlet DCM concentration, respectively.

4. Conclusions

In this study, we investigated TiO₂-supported Ru catalysts through treatment involving various H₂O₂ concentrations and examined the catalytic combustion activity of DCM. The activity of the (0 H₂O₂)-Ru@TiO₂ catalyst with a T₉₀ value of 344 °C was lower than the (1 and 3 H₂O₂)-Ru@TiO₂ catalysts with T₉₀ values of 301 °C and 317 °C, respectively. The effect of H₂O₂ treatment on the morphology, dispersion, and valence of Ru species was verified using our TEM and XPS results; in addition, the acidity and interaction between RuO₂ species and the TiO₂ support were analyzed using NH₃-TPD, pyridine-IR spectra, and H₂-TPR characterization. With the addition of H₂O₂, the (1 H₂O₂)-Ru@TiO₂ catalyst was found to comprise a higher number of acid sites (especially Brønsted acid sites), an appropriate amount of uniformly dispersed RuO₂, and strong interaction between RuO₂ and the TiO₂ support, which are characteristics conducive to improving the catalytic combustion activity and stability of DCM. However, excess H₂O₂ addition on the (3 H₂O₂)-Ru@TiO₂ catalyst led to phase segregation forming RuO₂ nanorods, which weakened the interaction between RuO₂ and the TiO₂ support and decreased catalytic activity.