Elucidating the special role of strong metal–support interactions in Pt/MnO₂ catalysts for total toluene oxidation

Abstract

With the aim to elucidate the role of strong metal–support interactions (SMSIs) in total toluene deep oxidation, herein, we prepared a series of Pt/MnO₂ catalysts with different MnO₂ crystal structures (α-, β-, γ- and hollow-MnO₂); it was found that the structures and catalytic behaviors of the MnO₂ crystals decorated by Pt nanoparticles (NPs) changed dramatically, especially the mobility of surface adsorbed oxygen and lattice oxygen. Among all these catalysts, the as-synthesized Pt/α-MnO₂ exhibited best catalytic activity for total toluene oxidation (T₉₀ = 170 °C). TGA, high-temperature XRD (HT-XRD) and quasi in situ XPS were applied to estimate the reconfiguration of gaseous oxygen on the surface of MnO₂-based catalysts. A series of characterization results indicate that the SMSIs between the Pt NPs and MnO₂ result in improved low-temperature reducibility, optimized surface chemical states, and the weakening of surface lattice oxygen species (Mn–O bands). The in situ DRIFTS study showed that the lattice oxygen could be involved in the activation of toluene without oxygen, whereas the presence of gas-phase oxygen was essential for the deep oxidation of toluene into CO₂ and H₂O on the surface of MnO₂-based catalysts.

---

New concepts

The utilization of strong metal–support interactions (SMSIs) between precious metals and metal oxide supports is a promising approach to improve the catalytic property and durability of catalysts, and the formation of SMSIs can significantly alter the natural properties of the catalysts. However, the special role of SMSIs in catalytic VOC deep oxidation remains unclear. If the SMSIs of the functional catalysts could be optimized by tuning the crystal structures of the transition-metal oxide supports, novel structures with SMSIs would be endowed with prominent performances in catalytic reactions. Based on this new concept, the present study provides scientific insights into the significant effect of SMSIs in the Pt/MnO₂ catalysts on the surface oxygen species. The biggest contribution of Pt/MnO₂ SMSIs is the weakening of surface Mn–O bonds and the improved mobility of the surface lattice oxygen. We speculate that the toluene oxidation simultaneously follows the Langmuir–Hinshelwood (L–H) mechanism and the Mars–van Krevelen (MvK) mechanism on the surface of the MnO₂-based catalysts. The results provide a deep understanding of the positive effects of SMSI on the natural structures of the catalysts.

Introduction

In recent years, the concentration of volatile organic compounds (VOCs), including ethyl acetate, formaldehyde, benzene and toluene, has been steadily increasing in the environment due to the continuous emissions of these VOCs; this is harmful to air quality and the human health.^1,2 To date, catalytic oxidation has been considered as a highly efficient technology to eliminate the VOCs.³ Supported platinum (Pt) catalysts have been widely studied in the field of heterogeneous catalysis due to their outstanding activity for VOC degradation.^4–6 It has been revealed that control over the particle size of Pt nanoparticles (NPs) can influence the structure–property relationships of supported Pt catalysts. Consequently, a significant number of studies have been devoted towards the development of uniform ultra-small Pt NPs or single-atom catalysts;^6–9 on the other hand, it has been disclosed that the surface interface, that is, the so-called strong metal–support interaction (SMSI) in supported metal catalysts, would significantly alter the activity and stability of the catalysts.^10–13 In 1978, SMSI was first observed by Tauster et al.,¹⁴ and since then, the SMSIs have attracted extensive attentions because of their positive effects. It has been established that the generation of SMSIs usually requires the metal NPs and supporting materials to lead to partial encapsulation during thermal treatment;¹³ the SMSIs can hinder the growth of small metal nanoparticle crystals, promoting the bonding structures and electron transfers between the metals and the supports.¹³

To date, various kinds of promising approaches have been introduced for the utilization of the SMSIs of supported metal catalysts; however, the structural characteristics of supporting materials are occasionally overlooked. Moreover, only a limited number of studies have been reported on the tailoring of the SMSIs of supported metal catalysts via modification of the crystal structures of transition-metal oxide supports.^15,16 Generally, transition-metal oxide supports can crucially influence the overall oxygen utilization of catalysts as well as the electronic and chemical states of precious metal NPs. Thus, tailoring of the natural structure/morphology of oxide supports would provide an efficient strategy for the manipulation of SMSIs. Among transition-metal oxide catalysts, MnO₂ catalysts with five phase structures (α-, β-, γ-, δ-, and ε-MnO₂) have been applied in different catalytic applications,^17–21 such as in the VOC oxidation, CO oxidation, ozone decomposition and NO_x SCR, mainly owing to their special physico-chemical properties. To achieve optimal performance of MnO₂, various novel structures including nanotubes, nanorods, nanoflowers, nanospheres and hollow microspheres have been consecutively reported.^22–24 To the best of our knowledge, only a few studies have been reported on Pt/MnO₂ systems, especially on the elucidation of the natural structures of Pt/MnO₂; moreover, the promotional roles of their SMSIs in the catalytic deep oxidation of VOCs have been rarely investigated, which may have an important significance in the development of novel catalysts with remarkable performance.

In this study, we successfully reported the construction of Pt/MnO₂ SMSIs by tuning the crystal structures of different MnO₂ catalysts (α-, β-, γ- and hollow-MnO₂) with the aim to systemically investigate the special role of SMSIs in toluene oxidation. Moreover, the relationship between the Pt/MnO₂ SMSIs and the catalytic performances for toluene oxidation was identified by the ex situ characterization techniques, including XRD, SEM, TEM, H₂-TPR, O₂-TPD, TGA, XPS, and high-temperature XRD (HT-XRD), as well as quasi in situ XPS. The in situ DRIFTS results provided evidences for the elucidation of the reaction mechanism of toluene deep oxidation.

Results and discussions

The size of the as-prepared Pt NPs was examined by transmission electron microscopy (TEM), as shown in Fig. 1a–d. It can be clearly observed that the as-prepared Pt NPs are monodisperse, and the average diameter is mainly distributed in the range of around 3.5 nm, as shown in Fig. 1d. In addition, the lattice fringe of the Pt NPs is 0.23 nm, corresponding to the (111) plane of the Pt NPs. The SEM images of the as-prepared carbon spheres with a smooth surface are shown in Fig. 1e and f, revealing that a monodisperse spherical shape with a diameter of around 1.0 μm has been successfully achieved by the hydrothermal treatment of the glucose solution. The carbon spheres were mainly used as templates to prepare the hollow γ-MnO₂ structure.

Fig. 1 (a) Low-, (b) medium-, and (c) high-magnification TEM images, and (d) particle-size distribution images of Pt nanoparticles; (e and f) SEM images of the carbon spheres.

X-ray diffraction (XRD) was used to investigate the crystal structures of the as-obtained MnO₂ products, as shown in Fig. S1 (ESI†). All the diffraction peaks could be perfectly ascribed to α-MnO₂ (JCPDS 44-0141), β-MnO₂ (JCPDS 24-0735) and γ-MnO₂ (JCPDS 42-1316). The XRD pattern of hollow MnO₂ (H-MnO₂) is in good agreement with that of γ-MnO₂ (JCPDS 42-1316). In addition, α-MnO₂ and β-MnO₂ possess high crystallinities according to their high-intensity and narrow characteristic peaks; in contrast, γ-MnO₂ and H-MnO₂ display low-intensity XRD patterns. Compared with that of the γ-MnO₂ sample, the XRD pattern of the H-MnO₂ sample indicates that its crystallinity is relatively lower, and more exposed planes can be detected. Lower crystallinity will be beneficial to the formation of more lattice defects that would enhance the surface oxygen vacancies and catalytic activities of the catalysts; this has been further confirmed by the following characterizations.

Fig. 2 shows the SEM images, schematics and crystal structures of different MnO₂ samples. As shown in Fig. 2e and f, the γ-MnO₂ product possesses a uniform sphere-like morphology with the average diameter of about 1.5 μm. The sphere-like morphology was composed by the aggregation of cubic-shaped MnO₂; the hollow γ-MnO₂ (H-MnO₂) sample was further synthesized by the NH₄HCO₃ co-precipitation method using a carbon sphere as the template. For the H-MnO₂ sample, the SEM images (as shown in Fig. 2a and b) reveal that there are similar sphere-like nanostructures assembled from cubes; however, a hollow spherical structure of MnO₂ is formed. α-MnO₂ shows uniform nanorods, and typically, an individual nanorod is about 600–800 nm long and about 40 nm wide, as shown in Fig. 2i and g. It can be clearly observed that β-MnO₂ possesses a pompon-like morphology, which is composed of tetragonal nanorods with the width of 20 nm (Fig. 2m and n).

Fig. 2 SEM images of different MnO₂ samples: (a and b) H-MnO₂, (e and f) γ-MnO₂, (i and j) α-MnO₂, and (m and n) β-MnO₂; (c, g, k and o) schematics and (d, h, l and p) crystal structures of H-MnO₂, γ-MnO₂, α-MnO₂ and β-MnO₂, respectively.

Fig. 3 shows the TEM images of different MnO₂ samples after the loading of Pt NPs. It can be observed that numerous Pt NPs are anchored on the surface of different MnO₂ samples, and the interplanar spacing of 0.230 nm is in accordance with the (111) plane of the Pt NPs in these samples. The as-fabricated Pt/H-MnO₂ sample presents hollow spheres with the diameter of about 1.5 μm and the wall thickness of 0.25 μm; the rough surface of the cubic-shaped MnO₂ consists of many small crystalline particles, and embedded pores can be observed, as shown in Fig. 3a and b. Fig. 3c reveals that the lattice fringes of about 0.479 nm and 0.24 nm can be well indexed to the (200) and (400) planes of γ-MnO₂, respectively. In Fig. 3d and e, it can be observed that Pt/γ-MnO₂ also has a uniform sphere-like morphology, which is formed by the interconnected cubic structures, and the average diameter is about 1.6 μm. The intraplanar spacing (as shown in Fig. 3f) is approximately 0.240 nm, which is attributed to the (400) plane of γ-MnO₂. In addition, Fig. 3g shows that Pt/α-MnO₂ displays nanorod structures with the length of about 600–900 nm, which is identical to the morphology of the α-MnO₂ support. The interplanar spacing was measured to be about 0.24 nm, assigned to the (211) crystal plane of the α-MnO₂ phase, as shown in Fig. 3i. In Fig. 3j, it can be obviously observed that Pt/β-MnO₂ exhibits intersected tetragonal nanorods, and the average width of the nanorods is about 20 nm. The lattice fringe with the intraplanar distance of 0.310 nm can be ascribed to the (110) plane of the β-MnO₂ crystal structure, as shown in Fig. 3l. Moreover, it can be observed that the upper surfaces of the Pt NPs in Pt/α-MnO₂ and Pt/β-MnO₂ form partial encapsulation due to the migration of the MnO₂ layer; this confirms the generation of forceful SMSIs (as shown in Fig. 3i and l, respectively).

Fig. 3 TEM images of different Pt/MnO₂ samples: (a–c) Pt/H-MnO₂, (d–f) Pt/γ-MnO₂, (g–i) Pt/α-MnO₂ and (j–l) Pt/β-MnO₂.

The complete oxidation of toluene over different MnO₂ and Pt/MnO₂ catalysts was estimated at the continuous flow rate of 100 mL min⁻¹ and the weight hourly space velocity (WHSV) of 60 000 mL g⁻¹ h⁻¹, as shown in Fig. 4a and b. T₁₀, T₅₀ and T₉₀ corresponding to the temperatures needed for 10%, 50% and 90% toluene conversion are summarized in Table 1 and Fig. S2 (ESI†). As shown in Fig. 4a, the toluene conversion increases with an increase in temperature, and the H-MnO₂ catalyst has highest activity for toluene oxidation among all the MnO₂ structures, achieving T₁₀, T₅₀ and T₉₀ at 205, 223, and 230 °C, respectively. The β-MnO₂ catalyst presents a relatively poor activity for toluene oxidation, and T₅₀ and T₉₀ obtained using it are 27 and 45 °C, higher than those found in the case of the H-MnO₂ catalyst, respectively. The order of T₉₀ for toluene conversion with these MnO₂ catalysts is H-MnO₂ > γ-MnO₂ > α-MnO₂ > β-MnO₂.

Fig. 4 Toluene conversion over different (a) MnO₂ and (b) Pt/MnO₂ samples, (c) Pt/α-MnO₂ with different Pt loadings, and (d) Arrhenius plots for toluene oxidation over different MnO₂ and Pt/MnO₂ samples. Toluene concentration was 1000 ppm balanced air, and all the reactions were kept at WHSV = 60 000 mL g⁻¹ h⁻¹.

Table 1 The BET and XPS data and catalytic performances of different MnO₂-based samples

Sample	Pt loading (wt%)^ICP	S BET (m^2 g^−1)	V pore (cm^3 g^−1)	Pore diameter (nm)	Mn^4+/Mn^3+	Oads/Ototal	Pt^0/Pttotal	T 10 (°C)	T 50 (°C)	T 90 (°C)
H-MnO2	—	75.4	0.25	11.5	5.87	0.349	—	205	223	230
γ-MnO2	—	88.0	0.20	7.8	5.81	0.291	—	210	227	238
α-MnO2	—	97.6	0.35	33.8	5.62	0.264	—	220	245	263
β-MnO2	—	21.1	0.11	25.5	4.04	0.243	—	225	250	275
Pt/H-MnO2	0.38	54.0	0.23	14.1	5.74	0.346	0.765	145	165	180
Pt/γ-MnO2	0.39	68.6	0.23	10.5	5.99	0.312	0.739	150	175	190
Pt/α-MnO2	0.39	71.0	0.37	39.8	7.64	0.302	0.788	145	156	170
Pt/β-MnO2	0.40	18.4	0.10	26.7	5.12	0.256	0.745	150	170	185

---

Particularly, all the Pt-anchored MnO₂ catalysts exhibited significantly improved catalytic activities for toluene oxidation when compared with single MnO₂ crystal structures, as shown in Fig. 4b. However, the order trend of T₉₀ for toluene conversion over these Pt/MnO₂ catalysts obviously changed, which was Pt/α-MnO₂ > Pt/H-MnO₂ > Pt/β-MnO₂ > Pt/γ-MnO₂. Apparently, Pt/α-MnO₂ exhibited a more outstanding performance with the T₁₀, T₅₀, and T₉₀ values of 145, 156, and 170 °C, respectively, whereas Pt/γ-MnO₂ exhibited a lower catalytic performance with the reaction temperatures for toluene conversions being 150, 175 and 190 °C, respectively. Furthermore, Pt/α-MnO₂ has a lower activation energy (E_a = 46.73 kJ mol⁻¹) that has been calculated via the Arrhenius plots shown in Fig. 4d. In addition, with an increase in the content (from 0.4 wt% to 0.8 wt%) of the Pt NPs on the surface of α-MnO₂, the catalytic activity for toluene oxidation increased, as displayed in Fig. 4c. When 0.8 wt% Pt NPs was loaded on α-MnO₂, the T₁₀, T₅₀ and T₉₀ values of this catalyst were 140, 153, and 158 °C, respectively. From the water vapor and cycling results shown in Fig. S3 (ESI†), it was found that the Pt/α-MnO₂ catalyst exhibited high catalytic stability and moisture resistance during toluene oxidation. As shown in Fig. S4 (ESI†), after the long-term testing in the presence of moisture, the conversion of toluene over the Pt/α-MnO₂ catalyst could still be maintained. The prominent catalytic performance of the Pt/α-MnO₂ catalyst can be attributed to its increased oxygen release ability resulting from the highlighted strong metal–support interaction between the Pt NPs and the α-MnO₂ support.

The surface area and pore size distributions of the MnO₂ and Pt/MnO₂ catalysts were further analyzed, as summarized in Fig. S5 (ESI†) and Table 1. Obviously, the as-synthesized MnO₂ and Pt/MnO₂ samples displayed classical type-IV adsorption–desorption isotherms with type H3 hysteresis loops in the relative pressure range of 0.6–1.0; this indicated the existence of slit-like mesoporous structures in the bulk samples. Among all the MnO₂ samples, the β-MnO₂ crystal structure had the smallest specific BET surface area and pore volume of 21.1 m² g⁻¹ and 0.109 cm³ g⁻¹, respectively, whereas α-MnO₂ showed the largest specific surface area of 97.6 m² g⁻¹ and pore volume of 0.346 cm³ g⁻¹. Note that the average pore size distributions of the H-MnO₂ and γ-MnO₂ samples are narrower than those of the α-MnO₂ and β-MnO₂ samples; this indicates that both the H-MnO₂ and the γ-MnO₂ samples have small pores and poor crystallinity; this has been verified by the XRD patterns. Moreover, the average pore size of H-MnO₂ is obviously at about 11.5 nm, which is slightly different from that of γ-MnO₂ (7.8 nm). Note that the specific surface areas of the as-synthesized Pt/MnO₂ catalysts increased when the Pt NPs were embedded on different MnO₂ crystal structures when compared with the case of the single MnO₂ samples. Moreover, no significant changes were observed in their pore volumes and pore sizes. Pt/α-MnO₂ also exhibited largest specific surface area and pore volume among all the Pt-supported catalysts, in which α-MnO₂ with higher specific surface area could be more conducive to the immobilization of Pt NPs.

According to the previous studies, the reduction of MnO₂ follows a continuous process: MnO₂ → Mn₂O₃ → Mn₃O₄ → MnO (Mn⁴⁺ → Mn³⁺ → Mn²⁺).^25–27 In the H₂-TPR profiles of these MnO₂ samples, there are two or three reduction peaks, as shown in Fig. S6a (ESI†). Obviously, β-MnO₂ exhibits a maximum H₂ consumption peak at 292 °C with a broad peak at 393 °C due to the contribution of H₂ in the reduction of MnO₂ to Mn₃O₄ and then to MnO; for the α-MnO₂ sample, there is a broad overlapping reduction peak in the range from 200 to 400 °C, whereas the high-temperature reduction peak at 605 °C can be considered to be due to the reduction of some of the surface MnO phase. The main reduction peaks of γ-MnO₂ are at about 250 and 377 °C; its first H₂ consumption peak due to the reduction of higher metal species (Mn⁴⁺) is lower than those of α-MnO₂ and β-MnO₂. Moreover, a reduction peak occurring at 100–200 °C can be obviously observed, attributed to the reduction of surface oxygen species. In addition, it is clear that the reduction peak of H-MnO₂ exhibits a certain similarity to that of γ-MnO₂, and the intensity of the reduction peak of the surface oxygen species at 172 °C is increased; this suggests that the gas-phase oxygen can be adsorbed/excited to rich active oxygen, which would contribute to the catalytic reaction of VOC oxidation over H-MnO₂.

When different MnO₂ crystal structures supported the Pt NPs, the reduction peaks of the Pt/MnO₂ products obviously shifted to lower temperature regions, as shown in Fig. S6b (ESI†); the reduction peaks of these Pt/MnO₂ samples could be divided into two temperature regions (marked as I and II). The region I, occurring at about 60–150 °C, is attributed to the reduction of a low amount of PtO_x and the formation of a strong Mn–O–Pt bond due to a strong metal–support interaction (SMSI) between the surface lattice oxygen of MnO₂ and the adjacent Pt NPs.¹⁸ The region II located at about 150–350 °C could be attributed to the stepwise reduction of MnO₂. In comparison, Pt/α-MnO₂ exhibits best low-temperature reducibility, whereas both Pt/γ-MnO₂ and Pt/H-MnO₂ possess similar reducibility. Moreover, the order of reducibility for the Pt/MnO₂ samples is inconsistent with that of the MnO₂ structures. The abovementioned results suggest that the difference in the interfacial interactions between the surface unsaturated Pt species and the different MnO₂ crystal structures leads to changes in the low-temperature reducibility; moreover, the surface Mn–O bonds are significantly weakened by the Pt NPs due to the formation of SMSIs. The weakening of the surface lattice oxygen would give rise to oxygen mobility and the formation of more active oxygen species.

To further confirm the effects of the formation of strong metal–support interactions on different oxygen species, the O₂-TPD and thermogravimetric analyses were carried out, as shown in Fig. S7 (ESI†). Some distinguished desorption peaks are clearly observed: one at below 150 °C, one at about 250–450 °C (I region), and another in the range of 500–650 °C (II region). The first desorption peak at about 107 °C is attributed to the release of surface active oxygen species (O_ads, O₂⁻ or O₂²⁻) that would be weakly bound to the surface of the catalysts; the region I desorption peak is attributed to surface lattice oxygen species (surface O_latt and O²⁻), and the oxygen desorption peak occurring in the temperature region II generally originates from lattice oxygen species (O_latt and O²⁻).^18,28,29 Hence, according to the first desorption peak, it is noticeable that Pt/α-MnO₂ exhibits largest surface oxygen vacancies with highest intensity of the surface-active oxygen species; this is well consistent with the H₂-TPR results. On the other hand, the position of the desorption peak I is shifted to a lower temperature (at about 365 °C); this demonstrates enhanced O_latt mobility due to an extremely strong metal–support interaction between α-MnO₂ and the Pt NPs when compared with the case of other α-MnO₂ and MnO₂-supported Pt samples.

Both the α-MnO₂ and the Pt/α-MnO₂ samples were further characterized by nitrogen/air thermogravimetric analysis (N₂-TGA and Air-TGA) to verify the surface active oxygen species that are beneficial for toluene oxidation, as shown in Fig. S8 (ESI†). The first weight losses at about 172 °C for the α-MnO₂ and Pt/α-MnO₂ catalysts are 1.47 wt% and 2.82 wt% under a N₂ atmosphere, respectively, which can be attributed to the desorption of the surface adsorbed oxygen or OH species.^30,31 At higher temperatures (200–500 °C), a small weight loss is observed, attributed to the loss of surface lattice oxygen. Moreover, the maximum weight loss for Pt/α-MnO₂ due to the bulk lattice oxygen release taking place at above 550 °C was approximately 6.56 wt%, whereas a minor change with a 5.82 wt% weight loss was observed in α-MnO₂ since it was relatively stable during the thermal treatment. Importantly, the weight loss for Pt/α-MnO₂ was higher than that for α-MnO₂ during the heating process; this indicated that the surface oxygen could be much more easily abstracted from Pt/α-MnO₂ by nitrogen than from α-MnO₂,³¹ and the presence of Pt NPs in α-MnO₂ improves the surface oxygen vacancies of the catalyst. In addition, the temperature values of the weight loss are in accordance with the O₂-TPD investigation results. In air, the weight losses for the α-MnO₂ and Pt/α-MnO₂ catalysts below 180 °C were approximately 1.76 wt% and 1.22 wt%, respectively, and their weight loss trend in the range of 180–500 °C in an air environment was similar to that under a N₂ atmosphere. However, compared with the case of the N₂ atmosphere, the weight losses of the Pt/α-MnO₂ catalyst in the air environment were decreased; this could explain the role of activated oxygen in the reaction process. Since the surface-adsorbed oxygen species coordinate with the oxygen vacancies or the interface of SMSIs are usually considered as the active species in VOC oxidation, this weight loss stage implies that the activated surface-adsorbed oxygen is gradually liberated from the surface of the catalysts. The result is in good agreement with the abovementioned O₂-TPD measurement results; this confirms that the Pt NPs anchored on MnO₂ can significantly enhance the activation of the surface oxygen species for toluene oxidation.

Fig. S9 (ESI†) shows the Mn 2p and O 1s spectra of different MnO₂ and Pt/MnO₂ catalysts. As revealed in Fig. S9a (ESI†), the binding energies (BE) at 642.7 and 641.7 eV are ascribed to the surface Mn⁴⁺ and Mn³⁺ species, respectively.^32,33 The surface Mn⁴⁺/Mn³⁺ molar ratios calculated by the Mn 2p quantitative analysis are shown in Table 1. Obviously, it can be observed that the Mn⁴⁺/Mn³⁺ molar ratios of these catalysts are distinguishing. H-MnO₂ possesses highest surface Mn⁴⁺/Mn³⁺ molar ratio (5.87), whereas β-MnO₂ shows lowest molar ratio of surface Mn⁴⁺/Mn³⁺ (4.04). The sequence of the surface Mn⁴⁺/Mn³⁺ ratios for these MnO₂ catalysts is as follows: H-MnO₂ > γ-MnO₂ > α-MnO₂ > β-MnO₂. Based on a literature study,²⁵ a catalyst with a higher manganese valence (Mn⁴⁺) would easily generate abundant surface adsorbed oxygen (O_ads) species. As shown in Fig. S9c (ESI†), the binding energies (BE) at around 529.7 and 531.5 eV are ascribed to the surface lattice oxygen (O_latt) and adsorbed oxygen (O_ads) species, respectively.^20,34,35 The amount of surface O_ads species follows the order H-MnO₂ (0.349) > γ-MnO₂ (0.291) > α-MnO₂ (0.264) > β-MnO₂ (0.243), which is consistent with the tendency of the surface Mn⁴⁺ species.

The supported-Pt samples were further investigated via the XPS measurement; the curves of the surface elements (Mn 2p, O 1s and Pt 4f) are shown in Fig. S9 and S10 (ESI†). The concentrations of the surface Mn⁴⁺ and O_ads species on Pt/MnO₂ are generally increased when compared with those of the single MnO₂ supports. Certainly, among these Pt-based samples, the concentrations of the surface Mn⁴⁺ and O_ads species on Pt/α-MnO₂ increased even more dramatically due to the contribution of a remarkable SMSI that leads to the weakening of the surface lattice oxygen species. The Pt 4f spectrum (as shown in Fig. S10, ESI†) reveals that Pt 4f_7/2 can be divided into two peaks at 71.3 and 72.4 eV, which are ascribed to the Pt⁰ and Pt²⁺ species, respectively.^6,36,37 It can be observed that Pt in these Pt/MnO₂ samples mainly exist as Pt⁰in the surface composition, and the percentages of Pt⁰ are similar for all the samples. In addition, the major peak of Pt 4f_7/2 for Pt/α-MnO₂ shifts towards a lower binding energy by around 0.3 eV; this can be due to the strong metal–support interaction between the Pt NPs and the α-MnO₂ support. It has been reported that the O_ads species would play significant roles in the total VOC oxidation.^24,25,38,39 This result manifests that the presence of Pt NPs can cause a series of significant changes in the surface structure of the catalysts.

To further analyze the effect of Pt NPs on the crystal structure of MnO₂, high-temperature XRD (HT-XRD) tests were carried out at different temperatures. Fig. S11 (ESI†) shows the HT-XRD patterns for the α-MnO₂ and Pt/α-MnO₂ samples. For the α-MnO₂ sample, all the diffraction peaks are ascribed to the characteristic peaks of the α-MnO₂ (JCPDS 44-0141) phase, and the intensity of the (211), (321), and (710) reflections of the MnO₂ phase gradually increases and slightly shifts to lower angles with an increase in temperature. In addition, other diffraction peaks undergo no significant changes when the heating temperature is increased from 25 °C to 250 °C. In contrast, the (400) and (521) diffraction peaks in the Pt/α-MnO₂ samples become weak and obviously shift to lower angles due to the thermal effects of lattice distortion. However, the variation trends of other major diffraction peaks in the Pt/α-MnO₂ sample are similar to those in the α-MnO₂ sample. Therefore, the overall crystal structure of the MnO₂ phase was maintained; however, the local surface-exposed planes that could interact with the Pt NPs were changed. According to the O₂-TPD, XPS and TGA results, it has been confirmed that the Pt NPs decorated on the surface of α-MnO₂ would greatly improve the surface oxygen vacancies and the mobility of the surface lattice oxygen. In this characterization, the larger shifted value of the angles for the Pt/α-MnO₂ catalyst could be attributed to the formation of a strong metal–support interaction. The reduced intensity of the partial diffraction peaks with the increasing temperature could be due to the effect of heating on the surface lattice oxygen (Mn–O or Mn–O–Pt) release dissociation. Hence, α-MnO₂ is relatively stable during the heating treatment, whereas the surface chemical elemental states of Pt/α-MnO₂ would be easily activated at the active sites at higher temperatures.

Next, the quasi in situ XPS measurement was carried out to characterize the subtle structural rearrangement of the α-MnO₂ and Pt/α-MnO₂ samples after the N₂ and air pre-treatment. The quasi in situ XPS results reveal that there are significant changes in the surface concentration of various elements. When the temperature is increased from 150 °C to 250 °C, the surface Mn⁴⁺/Mn³⁺ molar ratios on the α-MnO₂ catalyst decrease under the same atmosphere (Fig. 5 and Fig. S12, ESI†). In addition, its Mn⁴⁺/Mn³⁺ molar ratios after air pre-treatment are higher than those after the N₂ pre-treatment. The amount of surface O_ads species also decreases with the increasing temperature (Fig. 5 and Fig. S12, Table S1, ESI†); this indicates that the surface-adsorbed oxygen may be gradually liberated at higher temperatures. Significantly, when the N₂ pretreatment was switched to the air pretreatment, the amount of surface O_ads species similarly indicated a slight decrease, suggesting that the gaseous oxygen species were rearranged into surface lattice oxygen on the catalyst surface during heating. For the Pt/α-MnO₂ catalyst, the changes in the surface elements are similar to those for the α-MnO₂ catalyst. However, the surface O_ads species in the Pt/α-MnO₂ catalyst exhibit a sharp decrease due to the presence of gaseous oxygen; this indicates that the gaseous oxygen is more easily activated on the surface of the Pt/α-MnO₂ catalyst. The average oxidation state (AOS) of the manganese atom was further calculated according to the binding energy of the Mn 3s patterns (in Fig. S13, ESI†), as shown in Table S1 (ESI†). The quasi in situ Pt 4f XPS (Fig. S14, ESI†) shows that the valence state of the Pt species has not significantly changed.

Fig. 5 Quasi in situ (a and b) Mn 2p and (c and d) O 1s XPS spectra of α-MnO₂ and Pt/α-MnO₂ samples at different temperatures (150 and 250 °C) and atmospheres (N₂ and air).

To study the formed intermediates and the used oxygen species during the toluene oxidation process, the in situ DRIFTS was used to detect the transient species on the catalyst surface, as shown in Fig. 6 and Fig. S14 (ESI†). Before the adsorption and oxidation process of toluene, the α-MnO₂ and Pt/α-MnO₂ catalysts were pre-treated by N₂ for 1.0 h at 300 °C. The in situ DRIFT spectra of Pt/α-MnO₂ exposed to 500 ppm toluene/N₂ and toluene/air at 50 °C for different times are shown in Fig. 6a and b. The bands that occur at around 1490 and 1596 cm⁻¹ can be ascribed to ν(C=C) of the skeletal vibrations of the aromatic rings, where the methyl (–CH₃) group, the C–H stretching vibration and C–H in-plane bending vibration of the benzene rings have been immediately observed at 3070, 3027 and 1025 cm⁻¹, respectively, suggesting that toluene is adsorbed on the surface of the Pt/α-MnO₂ catalyst.^40–42 The 1176 cm⁻¹ absorption band is related to the antisymmetric C–C stretching vibration of the benzene ring. The bands at 2950, 2879, 1369, and 1303 cm⁻¹ are ascribed to ν_as(C–H), ν_s(C–H) bending vibrations of methylene (–CH₂) groups, and the CH₂ deformation mode, respectively, suggesting that the CH₃ group of toluene has been cleaved into methylene to form benzyl (C₆H₅–CH₂) on the surface of the Pt/α-MnO₂ catalyst.^28,43,44 The bands at around 1078, 1146, 1167, 1180 and 1213 cm⁻¹ are ascribed to the C–O stretching vibrations. The bands at 1167, 1146 and 1078 cm⁻¹ are the characteristic bands of the typical alkoxide species (C₆H₅–CH₂–O), and the band at 1213 cm⁻¹ belongs to the C–O vibration of the surface phenolate (C₆H₅–OH).^41,45 In addition, the bands at 1412 and 1667 cm⁻¹ are ascribed to the C–O–H and ν(C=O) stretching vibrations, respectively, suggesting the formation of benzaldehyde species (C₆H₅–CHO).⁴¹ The band located at 1820 cm⁻¹ corresponds to the ν(C=O) stretching vibration of acid anhydride as a by-product. The band centered at 1546 cm⁻¹ is related to the stretching vibration of the formate species (–COO), and the distinct bands located at 3300–3600 and 1627 cm⁻¹ are ascribed to the stretching vibrations of the OH group due to the contribution of toluene derivatives or water, indicating the formation of benzoate species (C₆H₅–COO) as the key intermediate during toluene oxidation.⁴² It can be found that the key intermediates on the surface of Pt/α-MnO₂ exposed to toluene/N₂ and toluene/air are similar (Fig. 6a and b). Various key intermediates can be observed during the adsorption process of toluene, and the band intensities of the key products increase with time; this suggests that the lattice oxygen may be involved in transformation of the adsorbed toluene to other species with or without gas-phase oxygen at lower temperatures.

Fig. 6 In situ DRIFT spectra of Pt/α-MnO₂ exposed to 500 ppm toluene/N₂ and toluene/air at 50 °C for (a and b) different times, and at (c and d) different temperatures (50, 100, 160, 200 and 240 °C), respectively; and (e) the schematic of various intermediates during the degradation of toluene.

When the Pt/α-MnO₂ catalyst was introduced into toluene/N₂ and toluene/air at different temperatures, obvious changes occurred in various species with an increase in temperature, as shown in Fig. 6c and d. The intensity of various bands increases to maximum at 160 °C in toluene/N₂ and toluene/air. Subsequently, when the temperature is increased to above 160 °C, the intensity of the bands (toluene and key intermediates) shows no significant change without oxygen (Fig. 6c); on the contrary, the concentration of the key intermediates (benzoate, benzaldehyde and alkoxide species) gradually decreases with an increase in temperature in the presence of gas-phase oxygen. Moreover, the in situ DRIFTS of α-MnO₂ under the same conditions (Fig. S15, ESI†) are similar to those of Pt/α-MnO₂, in which the major difference is the tendency of the intermediate species to decrease with temperature. The abovementioned results imply that both the gaseous oxygen and the lattice oxygen in the Pt/α-MnO₂ and α-MnO₂ catalysts can be involved in the initial activation of toluene; however, only gaseous oxygen is available for the deep oxidation of toluene into CO₂ and H₂O because the gaseous oxygen species can be reconstituted into the surface lattice oxygen on the catalyst surface. It has been confirmed that the weakening of the surface Mn–O bonds and the improved mobility of the lattice oxygen because of the contribution of the Pt NPs play an important role in the catalytic oxidation reaction. According to the in situ DRIFTS characterization, we can predict that the Langmuir–Hinshelwood (L–H) mechanism and the Mars-van Krevelen (MvK) mechanism simultaneously act during the toluene catalytic oxidation on the surface of the MnO₂-based catalysts; this has also been reported in the case of CO oxidation.^18,46 The reaction pathway could be mainly followed as (Fig. 6e)^28,41–45 benzyl radical → benzaldehyde → benzoate → formate species, and finally the complete oxidation into CO₂ and H₂O.

Conclusions

In summary, we showed that the SMSIs of Pt/MnO₂ catalysts significantly depended on different MnO₂ crystal structures (α-, β-, γ- and H-MnO₂), and different MnO₂-based catalysts were applied to the toluene deep oxidation reaction. The shapes of the MnO₂ supports not only led to different surface Mn⁴⁺ and O_ads amounts, low-temperature reducibility, and catalytic behaviors for total toluene oxidation, but also affected the correlation between them and the Pt NPs. Experimental results confirmed that the relationship between the catalyst structure and the catalytic activity was closely associated with the difference in the Pt/MnO₂ SMSIs. Among the MnO₂ catalysts, the as-synthesized H-MnO₂ catalyst with abundant surface O_ads species and a well-defined hollow structure exhibited best catalytic activity for total toluene oxidation (T₉₀ = 230 °C). More interestingly, when different MnO₂ crystal structures were used as supports, the Pt/α-MnO₂ catalyst showed highly superior activity (T₉₀ = 170 °C) for toluene oxidation, whereas Pt/γ-MnO₂ exhibited lowest catalytic activity. The improved activity was ascribed to the introduction of Pt NPs to form SMSIs. Due to the special role of the SMSIs in the Pt/MnO₂ catalysts, the SMSIs would greatly improve the surface oxygen vacancies and the mobility of the surface lattice oxygen; moreover, they would affect the surface-exposed Pt NPs. The in situ DRIFTS results demonstrated that the lattice oxygen on the catalyst surface could be involved in the initial activation of the toluene molecules to transform them into other species; however, the gaseous oxygen replenished by the oxygen vacancies played an essential role in the deep oxidation of the toluene molecules to CO₂ and H₂O.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This research was financially supported by the National Natural Science Foundation of China (No. 51878292, 51108187, 51672273, B5151050), the Science and Technology Planning Project of Guangdong Province China (No. 2015B0202236002), and the Guangdong Natural Science Foundation (No. 2016A030311003).

Notes and references

1. L. F. Liotta, H. Wu, G. Pantaleo and A. M. Venezia, Catal. Sci. Technol., 2013, 3, 3085.
2. W. Tang, W. Xiao, S. Wang, Z. Ren, J. Ding and P.-X. Gao, Appl. Catal., B, 2018, 226, 585–595.
3. H. Huang, Y. Xu, Q. Feng and D. Y. C. Leung, Catal. Sci. Technol., 2015, 5, 2649–2669.
4. X. Chen, Z. Zhao, Y. Zhou, Q. Zhu, Z. Pan and H. Lu, Appl. Catal., A, 2018, 566, 190–199.
5. E. Yang, E. J. Jang, J. G. Lee, S. Yoon, J. Lee, N. Musselwhite, G. A. Somorjai, J. H. Kwak and K. An, Catal. Sci. Technol., 2018, 8, 3295–3303.
6. R. Peng, S. Li, X. Sun, Q. Ren, L. Chen, M. Fu, J. Wu and D. Ye, Appl. Catal., B, 2018, 220, 462–470.
7. G. Liu, Y. Tian, B. Zhang, L. Wang and X. Zhang, J. Hazard. Mater., 2019, 367, 568–576.
8. J. Liu, ACS Catal., 2016, 7, 34–59.
9. B. Qiao, A. Wang, X. Yang, L. Allard, Z. Jiang, Y. Cui, J. Liu, J. Li and T. Zhang, Nat. Chem., 2011, 3, 634–641.
10. S. Li, Y. Xu, Y. Chen, W. Li, L. Lin, M. Li, Y. Deng, X. Wang, B. Ge, C. Yang, S. Yao, J. Xie, Y. Li, X. Liu and D. Ma, Angew. Chem., Int. Ed., 2017, 56, 10761–10765.
11. Y. Wang, D. Widmann, M. Heenemann, T. Diemant, J. Biskupek, R. Schlögl and R. J. Behm, J. Catal., 2017, 354, 46–60.
12. K. Murata, Y. Mahara, J. Ohyama, Y. Yamamoto, S. Arai and A. Satsuma, Angew. Chem., Int. Ed., 2017, 56, 15993–15997.
13. K. Fujiwara, K. Okuyama and S. E. Pratsinis, Environ. Sci.: Nano, 2017, 4, 2076–2092.
14. S. Tauster, S. Fung and R. Garten, J. Am. Chem. Soc., 1978, 100, 170–175.
15. R. Peng, X. Sun, S. Li, L. Chen, M. Fu, J. Wu and D. Ye, Chem. Eng. J., 2016, 306, 1234–1246.
16. S. Zhang, Z. Xia, T. Ni, Z. Zhang, Y. Ma and Y. Qu, J. Catal., 2018, 359, 101–111.
17. H. Xu, N. Yan, Z. Qu, W. Liu, J. Mei, W. Huang and S. Zhao, Environ. Sci. Technol., 2017, 51, 8879–8892.
18. N. Zhang, L. Li, R. Wu, L. Song, L. Zheng, G. Zhang and H. He, Catal. Sci. Technol., 2019, 9, 347–354.
19. B. Zhang, G. Cheng, W. Ye, X. Zheng, H. Liu, M. Sun, L. Yu, Y. Zheng and X. Cheng, Dalton Trans., 2016, 45, 18851–18858.
20. J. Jia, P. Zhang and L. Chen, Appl. Catal., B, 2016, 189, 210–218.
21. D. Jampaiah, V. K. Velisoju, P. Venkataswamy, V. E. Coyle, A. Nafady, B. M. Reddy and S. K. Bhargava, ACS Appl. Mater. Interfaces, 2017, 9, 32652–32666.
22. F. Shi, F. Wang, H. Dai, J. Dai, J. Deng, Y. Liu, G. Bai, K. Ji and C. T. Au, Appl. Catal., A, 2012, 433–434, 206–213.
23. B. Bai, J. Li and J. Hao, Appl. Catal., B, 2015, 164, 241–250.
24. F. Wang, H. Dai, J. Deng, G. Bai, K. Ji and Y. Liu, Environ. Sci. Technol., 2012, 46, 4034–4041.
25. W. Tang, X. Wu, D. Li, Z. Wang, G. Liu, H. Liu and Y. Chen, J. Mater. Chem. A, 2014, 2, 2544–2554.
26. Z. Sihaib, F. Puleo, J. M. Garcia-Vargas, L. Retailleau, C. Descorme, L. F. Liotta, J. L. Valverde, S. Gil and A. Giroir-Fendler, Appl. Catal., B, 2017, 209, 689–700.
27. M. Luo, Y. Cheng, X. Peng and W. Pan, Chem. Eng. J., 2019, 369, 758–765.
28. Y. Wang, H. Arandiyan, Y. Liu, Y. Liang, Y. Peng, S. Bartlett, H. Dai, S. Rostamnia and J. Li, ChemCatChem, 2018, 10, 3429–3434.
29. B. Li, Q. Yang, Y. Peng, J. Chen, L. Deng, D. Wang, X. Hong and J. Li, Chem. Eng. J., 2019, 366, 92–99.
30. L. Ma, C. Y. Seo, X. Chen, K. Sun and J. W. Schwank, Appl. Catal., B, 2018, 222, 44–58.
31. P. Li, X. Chen, L. Ma, A. Bhat, Y. Li and J. W. Schwank, Catal. Sci. Technol., 2019, 9, 1165–1177.
32. H. Pan, Y. Jian, C. Chen, C. He, Z. Hao, Z. Shen and H. Liu, Environ. Sci. Technol., 2017, 51, 6288–6297.
33. Y. Wei, J. Liu, W. Su, Y. Sun and Y. Zhao, Catal. Sci. Technol., 2017, 7, 1565–1572.
34. Q. Zhang, S. Mo, B. Chen, W. Zhang, C. Huang and D. Ye, Mol. Catal., 2018, 454, 12–20.
35. N. Huang, Z. Qu, C. Dong, Y. Qin and X. Duan, Appl. Catal., A, 2018, 560, 195–205.
36. B. Wang, B. Chen, Y. Sun, H. Xiao, X. Xu, M. Fu, J. Wu, L. Chen and D. Ye, Appl. Catal., B, 2018, 238, 328–338.
37. X. Weng, Q. Liu, J. J. Feng, J. Yuan and A. J. Wang, J. Colloid Interface Sci., 2017, 504, 680–687.
38. J. Zhang, Y. Li, L. Wang, C. Zhang and H. He, Catal. Sci. Technol., 2015, 5, 2305–2313.
39. S. Mo, S. Li, W. Li, J. Li, J. Chen and Y. Chen, J. Mater. Chem. A, 2016, 4, 8113–8122.
40. H. Sun, Z. Liu, S. Chen and X. Quan, Chem. Eng. J., 2015, 270, 58–65.
41. X. Yang, X. Yu, M. Lin, X. Ma and M. Ge, Catal. Today, 2019, 327, 254–261.
42. S. Zhao, F. Hu and J. Li, ACS Catal., 2016, 6, 3433–3441.
43. Y. T. Lai, T. C. Chen, Y. K. Lan, B. S. Chen, J. H. You, C. M. Yang, N. C. Lai, J. H. Wu and C. S. Chen, ACS Catal., 2014, 4, 3824–3836.
44. Y. Wang, W. Deng, Y. Wang, L. Guo and T. Ishihara, Mol. Catal., 2018, 459, 61–70.
45. X. Chen, X. Chen, E. Yu, S. Cai, H. Jia, J. Chen and P. Liang, Chem. Eng. J., 2018, 344, 469–479.
46. Y. G. Wang, D. C. Cantu, M. S. Lee, J. Li, V. A. Glezakou and R. Rousseau, J. Am. Chem. Soc., 2016, 138, 10467–10476.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental section, Table S1 and Fig. S1–S15. See DOI: 10.1039/c9nh00303g

---