Porous hollow manganites with robust composite shells for oxidation of CO at low temperature

Abstract

Yolk- and hollow-type manganite-coated silica (SiO₂) microspheres were synthesized via the thermal hydrolysis of urea using a dissolution and deposition method. Core/shell SiO₂ microspheres were used as templates; the products were then annealed at 700 °C for 10 h under an O₂ atmosphere. Yolk microspheres (Mn₂O₃/SiO₂_1T and Mn₂O₃/SiO₂_2T) were obtained when the dissolution and deposition step was repeated once and twice, respectively, while hollow microspheres (Mn₂O₃/SiO₂_3T) were produced when the method was repeated three times. The microspheres had average diameters of 300–310 nm. The Mn₂O₃/SiO₂_3T product possessed the largest Brunauer–Emmett–Teller (BET) surface area and exhibited the best catalytic performance with regard to CO oxidation. The complete conversion of CO was achieved at approximately 200 °C. Other microspheres, including Mn₃O₄/SiO₂_3T, NiO/SiO₂_3T, and Ni/SiO₂_3T, yielded much lower activities than the Mn₂O₃/SiO₂_3T catalyst. The large surface area of the Mn₂O₃/SiO₂_3T microsphere sample, as well as the presence of Mn³⁺ ions, appears to be responsible for its superior catalytic CO oxidation activity amongst the tested catalysts.

---

Introduction

In automotive exhaust streams with catalytic converters, the catalytic oxidation of carbon monoxide (CO) to carbon dioxide (CO₂) at low temperature is an important function of emission control.¹ To date, precious metals such as Pd, Pt, Rh, and Au have been considered key elements that can effectively mitigate toxic gases.^2–5 Currently, catalytic converters, composed of Pd, Pt, and Rh nanoparticles (NPs) supported on a synthetic cordierite, are used to convert pollutants in exhaust streams into less toxic gases. Although such precious metals exhibit excellent catalytic activity, there are some challenging issues to be overcome. One issue is the high cost of the precious metals owed to their low abundance.⁶ The other regards the CO poisoning of the metal surface, which impairs the catalytic activity. These drawbacks have stimulated continuous interest in the development of low cost, highly sustainable catalysts that are able to effectively oxidize toxic gases at low temperatures.^7–10

A wide range of non-precious metal oxides have thus been considered as alternative catalysts that are suitable for CO oxidation.^11–15 A recent report by Binder et al. provides evidence that ternary oxides composed of CuO, Co₃O₄, and CeO₂ exhibit high activity for CO oxidation under simulated exhaust conditions.¹⁶ There is increasing interest in catalysts based on single oxides, which can provide high atom efficiency. Iablokov and co-workers demonstrated that Co₃O₄ NPs embedded in mesoporous silica (SiO₂) represent one of the most active metal oxides with regard to CO oxidation;¹⁷ the maximum reaction rates were achieved by Co₃O₄ NPs with sizes of 5–8 nm. Iron oxide (Fe₂O₃) NPs were also studied as catalysts for CO oxidation; this study showed that compared with commercial bulk powders, Fe₂O₃ NPs performed much more effectively as CO oxidation catalysts.¹⁸ Ren et al. prepared a range of mesoporous metal oxides using a SiO₂ template, most of which exhibited excellent activity towards CO oxidation.¹⁹ Manganese oxides, such as MnO, Mn₃O₄, Mn₂O₃, and MnO₂, have attracted much interest because they have excellent thermal stability and exhibit high levels of activity.^20–28 Among them, Mn₂O₃ is considered the most active catalyst with regard to CO oxidation at low temperature.²⁹ Most of the aforementioned studies have focused on nanometer-sized samples primarily because of their large specific surface area. However, under continuous exhaust streams, NPs are susceptible to agglomeration, especially at high temperature, which results in the formation of aggregates and eventually reduces their catalytic activity.

Here, we present highly porous microspheres with robust shells consisting of Mn₂O₃ and SiO₂, which possess large surface areas and also prevent the agglomeration of NPs. They were prepared by a dissolution and deposition method, which involves the thermal hydrolysis of urea and the use of core/shell SiO₂ microspheres as templates. Particularly, amongst the catalysts tested in the study, the greatest catalytic activity with regard to CO oxidation at low temperature was demonstrated by the hollow spherical microspheres (Mn₂O₃/SiO₂_3T). Our results render significant promise for the use of hollow composite microspheres as an inexpensive and stable metal oxide catalyst for CO oxidation.

Experimental

Chemicals

All the chemicals were obtained from commercial suppliers and used without further purification unless otherwise stated. Manganese acetyl acetonate, ethanol, and urea were purchased from Sigma-Aldrich (St. Louis, MO, USA). Tetraethyl orthosilicate (TEOS) and octadecyl trimethoxy silane (C₁₈-TMS) were purchased from Tokyo Chemical Industry Co. (Tokyo, Japan). Ammonia solution (25–28%) was obtained from Samchun Chemical Reagent Co. (Seoul, Korea).

Synthesis of core/shell SiO₂ microspheres

The core/shell SiO₂ microspheres were prepared using a slightly modified Stöber procedure.³⁰ Briefly, 75 mL of ethanol, 10 mL of deionized water, and 3 mL of ammonia solution were added to a round-bottomed flask (500 mL), and the solution was stirred for 10 min. Using a syringe, the TEOS (6 mL) was injected into the solution. Following 2 h of vigorous stirring, a mixture of TEOS (5 mL) and C₁₈-TMS (2 mL) was added, and the solution was stirred for a further 2 h. A white precipitate formed, which was separated from the solution via centrifugation. To remove all the organic residues from the precipitated powders, the powders were sintered in air at 550 °C for 6 h, yielding 2.75 g of SiO₂ powders. The obtained SiO₂ possessed a spherical shape and a core/shell structure. The core had a diameter of approximately 250 nm and the shell had a thickness of 35–45 nm, which were estimated by TEM images (Fig. S1†).

Synthesis of yolk and hollow microspheres composed of Mn₂O₃ and SiO₂ (Mn₂O₃/SiO₂_1T, Mn₂O₃/SiO₂_2T, Mn₂O₃/SiO₂_3T)

The core/shell SiO₂ microspheres (0.495 g, 8.3 × 10⁻³ mol) were dispersed in 250 mL of deionized water in a round-bottomed flask (500 mL). Manganese acetyl acetonate dihydrate (0.5 g, 2.0 × 10⁻³ mol) and urea (4.0 g, 0.067 mol) were added into the solution, which was then stirred at 80 °C for 2 h. The white microspheres gradually transformed into dark brown colloidal spheres, which were separated from the solution by centrifugation (3000 rpm) and oven-dried at 100 °C. The dried microspheres were annealed at 700 °C for 10 h under an O₂ atmosphere. The resulting product was denoted as Mn₂O₃/SiO₂_1T. Prior to annealing at 700 °C, the dried microspheres were reacted again in the presence of manganese acetyl acetonate dihydrate and urea by following the same procedure. The resulting microspheres were annealed at 700 °C for 10 h under an O₂ atmosphere. The product was denoted as Mn₂O₃/SiO₂_2T. In a similar manner, the Mn₂O₃/SiO₂_3T sample was obtained by repeating the procedure three times.

Synthesis of hollow microspheres composed of Mn₃O₄ and SiO₂ (Mn₃O₄/SiO₂_3T)

The procedure used for the synthesis of Mn₃O₄/SiO₂_3T was virtually identical to that of Mn₂O₃/SiO₂_3T, except that annealing was performed under a reducing atmosphere. The core/shell SiO₂ microspheres (0.495 g, 8.3 × 10⁻³ mol) were dispersed in 250 mL of deionized water in a round-bottomed flask (500 mL). Manganese acetyl acetonate dihydrate (0.5 g, 2.0 × 10⁻³ mol) and urea (4.0 g, 0.067 mol) were added into the solution, which was then stirred at 80 °C for 2 h. The white microspheres gradually transformed into dark brown colloidal spheres, which were separated from the solution via centrifugation (3000 rpm) and then oven-dried at 100 °C. The dried microspheres were annealed at 400 °C for 10 h under a reducing atmosphere (Ar/H₂ = 95 : 5), yielding the Mn₃O₄/SiO₂_1T microspheres. The Mn₃O₄/SiO₂_2T and Mn₃O₄/SiO₂_3T samples were prepared by repeating the procedure twice and three times, respectively.

Characterization

Powder X-ray diffraction (XRD) patterns were recorded on a Rigaku DMAX 2500 diffractometer (Japan), which employed Cu K_α radiation. High resolution scanning electron microscopy (HR-SEM) analyses were performed using a Hitachi s-5500 microscope (Hitachi, Japan). The specimens for the HR-SEM analyses were prepared by dropping powders, which were dispersed in ethanol, onto a lacey grid. Transmission electron microscopy (TEM) was performed using a JEOL JEM-2100F microscope (JEOL, Japan). The specimens for the TEM examination were prepared by dispersing samples of the finely ground powders in anhydrous ethanol, and subsequently allowing a drop of the suspension to evaporate on a 400-mesh carbon-coated grid. Energy-dispersive X-ray (EDX) spectroscopy was performed using the same microscope as that used for the TEM. The line-scan analyses were performed in scanning transmission electron microscope mode using a real-time interactive imaging system with a high-angle annular dark-field detector. The adsorption and desorption measurements were conducted at 77 K using an ASAP 2420 instrument (Micromeritics, Norcross, USA), with nitrogen as the adsorptive gas. The Brunauer–Emmett–Teller (BET) surface areas were calculated from the adsorption curve using the BET equation, with P/P₀ = 0.05–0.3. The pore-size distributions were obtained from the desorption curve using the density functional theory method. Prior to each sorption measurement, the sample was out-gassed at 300 °C for 24 h under vacuum conditions to completely remove the impurities. To investigate the elemental compositions, X-ray photoelectron spectroscopy (XPS; Theta probe AR-XPS System, Thermo Fisher Scientific, UK) analysis was conducted using a monochromated Al K_α X-ray source (hν = 1486.6 eV) at the Korea Basic Science Institute (KBSI) located in Busan. The structure of the product was simulated by the Rietveld method using the RIETAN-2000 program.

CO oxidation activity

The CO oxidation tests were performed using a fixed bed reactor. For each measurement, 60 mg of the sample was placed in the centre of the flow tube. The feed mixture, prepared using mass flow controllers (MKS Instruments, Wilmington, USA), consisted of 3.5% CO and 8.4% O₂, which was balanced with helium. The reactant mixtures were allowed to flow. The flow rate was fixed at approximately 52 mL min⁻¹. To precisely determine the conversion, the gas product was collected at each temperature over the course of the steady-state operation, which was performed for 40 min. The effluent gas was analysed using online gas chromatography (Younglin Instrument, Korea) with a thermal conductivity detector.

Results and discussion

We have developed a simple strategy to synthesize yolk and hollow microspheres with composite oxides consisting of Mn₂O₃ and SiO₂, which involves the use of core/shell SiO₂ microspheres as templates and subsequent dissolution and deposition treatments with urea. Fig. 1 schematically shows the method used to prepare the yolk and hollow microspheres composed of Mn₂O₃ and SiO₂. In brief, well-defined SiO₂ microspheres with a core/shell structure were used as templates. The SiO₂ microspheres were dispersed in a urea solution containing manganese ions. The thermal hydrolysis of urea results in the simultaneous dissolution of the core SiO₂ and deposition of the manganese species on the SiO₂ shell.³¹ Sequential dissolution and deposition stages eventually result in the formation of hollow microspheres containing manganese species and components. The subsequent annealing converts the amorphous shell into a robust shell consisting of Mn₂O₃ and SiO₂ while maintaining a hollow, spherical shape. A remarkable feature of this method is that yolk and hollow microspheres with robust oxide shells can easily be prepared by tuning the dissolution and deposition route. Moreover, yolk and hollow structures can be synthesized even when additional chemical etching is not performed.

Fig. 1 Schematic of the preparation of the yolk and hollow microspheres with composite oxides via the dissolution and deposition method. (a) Yolk–shell microsphere with large yolk and thin shell (Mn₂O₃/SiO₂_1T), (b) yolk–shell microsphere with small yolk and thick shell (Mn₂O₃/SiO₂_2T), and (c) hollow microsphere with very thick shell (Mn₂O₃/SiO₂_3T). The core and shell portions of the core/shell SiO₂ are depicted in dark- and light-colored blue shades, respectively. The shells of (a)–(c) are depicted in a purple shade, which represent a mixed composite of Mn₂O₃ and SiO₂. The terms 1T, 2T, and 3T indicate that the dissolution and deposition process has been performed once, twice, and three times, respectively. Following completion of the dissolution and deposition steps, the resulting microspheres were annealed at 700 °C for 10 h under an O₂ atmosphere.

Fig. 2 shows typical TEM and SEM images of the Mn₂O₃/SiO₂_1T, Mn₂O₃/SiO₂_2T, and Mn₂O₃/SiO₂_3T samples, which clearly illustrate the sequential changes in the microstructures due to the repeated thermal hydrolysis of the urea. Fig. 2(a) shows a TEM image of the Mn₂O₃/SiO₂_1T sample, which illustrates the white spherical line between the shell and core, indicating the slight dissolution of the core SiO₂.^32,33 The SiO₂ dissolution is more apparent in the TEM image of the second dissolution and deposition product (Mn₂O₃/SiO₂_2T), shown in Fig. 2(b), which clearly shows a yolk/shell structure formed by the dissolution of the core SiO₂ and the deposition of manganese species on the SiO₂ shell. The diameter of the Mn₂O₃/SiO₂_2T product increased to approximately 290 nm while the diameter of the core (yolk) decreased to about 70 nm. Following the third dissolution and deposition reaction, the core SiO₂ completely etched away and only the shell portion remained. The TEM image of the Mn₂O₃/SiO₂_3T product, shown in Fig. 2(c), reveals that the thickness of the shell increased to approximately 80 nm. The average diameters and shell thicknesses of the three samples are summarized in Table 1, which indicates how the microstructural changes are driven by the thermal hydrolysis of the urea. It is thus apparent that the core SiO₂ gradually dissolved while the shell steadily expanded, via the deposition of the dissolved SiO₂ and manganese species, as the period of the dissolution and deposition treatment was extended. A recent report on hollow SiO₂/TiO₂ nanospheres synthesized via a sonication-mediated etching and re-deposition method supports our conjecture on the formation of hollow microspheres with composite oxide shells.^34,35

Fig. 2 The top panels show representative TEM images of (a) Mn₂O₃/SiO₂_1T, (b) Mn₂O₃/SiO₂_2T, and (c) Mn₂O₃/SiO₂_3T. SEM images of the corresponding samples, that is, (d) Mn₂O₃/SiO₂_1T, (e) Mn₂O₃/SiO₂_2T, and (f) Mn₂O₃/SiO₂_3T, are displayed in the lower panels. Higher magnification TEM and SEM images are shown in the insets.

Table 1 Average diameters and shell thicknesses of Mn₂O₃/SiO₂_1T, Mn₂O₃/SiO₂_2T, and Mn₂O₃/SiO₂_3T. Contraction and expansion percentages are shown in parentheses

Sample	Average diameter	Core diameter (reduction%)	Shell thickness (increase%)
Core/shell SiO2	250 nm	170 nm (0%)	40 nm (0%)
Mn2O3/SiO2_1T	260 nm	155 nm (9%)	45 nm (13%)
Mn2O3/SiO2_2T	285 nm	110 nm (36%)	70 nm (75%)
Mn2O3/SiO2_3T	305 nm	0 nm (100%)	80 nm (100%)

---

Typical SEM images of the aforementioned samples are provided in the lower panel of Fig. 2. The average diameters of the microspheres are virtually identical to those estimated from the TEM images. It can be determined from the SEM images that the dissolution and deposition treatment time had no observable effect on the surface morphology, although there was a slight increase in the overall diameters. Owing to the repeated dissolution and deposition processes, numerous crystallites of Mn₂O₃ and SiO₂ appear to assemble into a spherical structure. It is notable that even following annealing at 700 °C, the hollow parent structures were essentially retained. The microspheres did not demonstrate any noticeable shrinkage, suggesting that the SiO₂ layers function as supporting materials to sustain the composite oxide shells.

EDX analysis was conducted to investigate the elemental distribution and confirm the presence of a hollow structure. As shown in Fig. 3, the presence of Si and Mn on the shell can be verified by the EDX line scan elemental profile data of the three samples. It is notable that the Mn intensity increases as the dissolution and deposition treatment period is extended while the Si intensity decreases. The Si intensity of the core of the Mn₂O₃/SiO₂_3T product appears to have been completely diminished, confirming that the sample possesses a hollow structure. These elemental profile results correlate well with the TEM and SEM images shown in Fig. 2. Considering the line profile curves obtained for the Mn₂O₃/SiO₂_3T product, it is also notable that both Mn and Si are evenly distributed on the shell. This suggests that the Mn species and dissolved SiO₂ replicas are simultaneously deposited on the porous shell.

Fig. 3 EDX elemental line scans of (a) Mn₂O₃/SiO₂_1T, (b) Mn₂O₃/SiO₂_2T, and (c) Mn₂O₃/SiO₂_3T. The top panels show TEM images that include the positions of the line scan (yellow line). The lower panel shows the EDX signal intensities for Si (blue line) and Mn (red line) across the diameter of the corresponding microsphere.

The structures of the three samples were characterized by powder XRD. All the sharp XRD peaks, illustrated in Fig. 4, correlate well with those of the α-Mn₂O₃ phase (JCPDS no. 78-0390), which has a cubic bixbyite structure. The very broad peak near 20°, particularly in the case of the Mn₂O₃/SiO₂_1T sample, is ascribed to the amorphous nature of SiO₂. When the dissolution and deposition treatment time is extended, more Mn species are deposited on the SiO₂ shell. Accordingly, the XRD peaks of the Mn₂O₃ phase are distinctive. The broad peak of the amorphous SiO₂ is spread out because of the strong peaks that originated from the crystalline Mn₂O₃ phase. Using Scherrer's method, the average grain size of the Mn₂O₃ phase was estimated from the (222) reflection of the XRD data. The grain size varies as a function of the dissolution and deposition treatment time. The average grain size of the first sample produced by the dissolution and deposition method is 47.69 nm, and the average grain size appears to increase gradually to 61.05 nm and 86.76 nm for the second and third products obtained, respectively. This indicates that the first Mn species are embedded into the mesoporous SiO₂ shell and nucleate as nano-crystallites within the shell. As a result, the average grain size of the Mn₂O₃ phase is only 8.7 nm. When the dissolution and deposition treatment time was extended, the dissolved Mn species were deposited on the external surface of the mesoporous shell. This caused the size of the Mn₂O₃ crystallites to increase.

Fig. 4 Powder X-ray diffraction patterns of (a) Mn₂O₃/SiO₂_1T (black), (b) Mn₂O₃/SiO₂_2T (red), and (c) Mn₂O₃/SiO₂_3T (blue). Theoretical XRD patterns of the cubic Mn₂O₃ phase are also displayed as vertical bars (green) for reference.

XPS was employed to evaluate the chemical environments of Mn and Si in the Mn₂O₃/SiO₂_1T, Mn₂O₃/SiO₂_2T, and Mn₂O₃/SiO₂_3T products. The detailed XPS spectra of the Mn 2p and Si 2p regions of the three samples are given in Fig. 5. Two Mn 2p signals can be observed at approximately 641.3 and 652.9 eV, which can be assigned to Mn 2p_3/2 and Mn 2p_1/2, respectively.^36–38 As summarized in Table 2, the Mn 2p binding energies of the three samples show little dependency on the dissolution and deposition treatment period, and are relatively similar to those obtained from previous XPS studies on bulk Mn₂O₃ materials.³⁹ Accordingly, the two peaks are associated with Mn₂O₃. The full width at half maximum (FWHM) values for the Mn 2p_3/2 of the three samples differ to a certain extent. However, in the case of the Mn₂O₃/SiO₂_1T product, the FWHM value is approximately 0.5 eV greater than those obtained for the Mn₂O₃/SiO₂_2T and Mn₂O₃/SiO₂_3T products. The peak broadening could be ascribed to the size of the Mn₂O₃ particles embedded in the porous SiO₂ matrix. The interfacial interaction between Mn₂O₃ and SiO₂ appears to influence the electronic properties, and the effect is enhanced as the particle size decreases. Similar behaviors were observed in the case of Mn₂O₃/SBA-15 and MnO_x/Al₂O₃ systems, where the FWHM values increased as the particle size decreased.^40,41 Fig. 5(b) shows the Si 2p XPS spectra of the three samples. The Si 2p peaks are displayed for the range of 102.9–103.2 eV. The binding energies of Si 2p correlate well with the values reported for a range of bulk SiO₂ materials. Considering the Si 2p XPS spectra, it is notable that the binding energy of the Mn₂O₃/SiO₂_1T product is greater than those of the Mn₂O₃/SiO₂_2T and Mn₂O₃/SiO₂_3T products. This indicates that the dissolution and deposition process alters the electronic environment of Si at the interfacial regions between the Mn₂O₃ and SiO₂.^34,35

Fig. 5 XPS data for Mn₂O₃/SiO₂_1T (black line), Mn₂O₃/SiO₂_2T (red line), and Mn₂O₃/SiO₂_3T (blue line) for the regions of (a) Mn 2p and (b) Si 2p.

Table 2 XPS data for Mn₂O₃/SiO₂_1T, Mn₂O₃/SiO₂_2T, and Mn₂O₃/SiO₂_3T. FWHM values are also shown. For comparison, XPS data for bulk Mn₂O₃ are also included

Catalysts	Mn				Si
	2p3/2		2p1/2		2p
	Binding energy (eV)	FWHM (eV)	Binding energy (eV)	FWHM (eV)	Binding energy (eV)	FWHM (eV)
Mn2O3/SiO2_1T	641.4	4.886	653.0	4.372	101.4	2.179
Mn2O3/SiO2_2T	641.3	4.340	652.9	3.760	101.3	2.187
Mn2O3/SiO2_3T	641.1	4.333	652.8	3.749	100.9	2.138

---

To determine the surface areas and pore sizes of the three samples, we evaluated their N₂ adsorption and desorption isotherms. As shown in Fig. 6, all three samples exhibit characteristics associated with type IV isotherm curves and show H4-type broad hysteresis loops, suggesting that mesopores are present in the shells.^42–44 The BET specific surface areas of the Mn₂O₃/SiO₂_2T and Mn₂O₃/SiO₂_3T products were calculated to be 292.52 m² g⁻¹ and 497.11 m² g⁻¹, respectively; these values are greater than that of the Mn₂O₃/SiO₂_1T product (257.35 m² g⁻¹). It is apparent that the samples obtained after the dissolution and deposition stage have large surface areas. This suggests that new porous shells composed of dissolved Mn and Si species evolve on the mesoporous SiO₂ shells without being damaged. In the case of the Mn₂O₃/SiO₂_1T, Mn₂O₃/SiO₂_2T, and Mn₂O₃/SiO₂_3T products, the pore sizes were 2.6 nm, 2.8 nm, and 3.3 nm, respectively (Fig. S2†). This slight increase in the pore size suggests that the dissolution and deposition process period does not significantly influence the pore size.

Fig. 6 Nitrogen adsorption and desorption isotherms of (a) Mn₂O₃/SiO₂_1T, (b) Mn₂O₃/SiO₂_2T, and (c) Mn₂O₃/SiO₂_3T. The pore-size distributions were calculated using the Barrett–Joyner–Halenda method.

With regard to CO oxidation, the catalytic activity of the three samples was examined. Amongst the samples, the Mn₂O₃/SiO₂_3T sample exhibited the best catalytic performance for CO oxidation over the entire temperature range, as shown in Fig. 7. This result can be explained as follows. First, in the case of the Mn₂O₃/SiO₂_3T product, the reaction temperature (T₅₀), where the CO conversion is 50%, is 110 °C; however, those of the Mn₂O₃/SiO₂_1T and Mn₂O₃/SiO₂_2T products are 165 °C and 138 °C, respectively. Secondly, the CO conversion efficiency of the Mn₂O₃/SiO₂_3T product is at least twice those of the Mn₂O₃/SiO₂_1T and Mn₂O₃/SiO₂_2T products at the same temperature. At 150 °C, the Mn₂O₃/SiO₂_3T product converts over 90% of the CO to CO₂, while the Mn₂O₃/SiO₂_1T and Mn₂O₃/SiO₂_2T products only convert approximately 37% and 66% of the CO, respectively. The low-temperature oxidation of the Mn₂O₃/SiO₂_3T product is also remarkable because even noble metals exhibit limited CO oxidation at temperatures below 200 °C. Accordingly, the Mn₂O₃/SiO₂_3T catalyst may outperform such catalysts with regard to the oxidation of CO at low temperature. The excellent activity achieved by the Mn₂O₃/SiO₂_3T product could be ascribed to its large surface area and the nanometer-sized Mn₂O₃ crystallites of the shell. To confirm the effect of the crystallite size on the CO oxidation, a commercial Mn₂O₃ powder, with an average crystallite size of over 10 μm was included in the evaluation. Under identical conditions, the T₅₀ of the Mn₂O₃ powder was determined to be approximately 231 °C and complete conversion (T₁₀₀) was achieved at 350 °C (Fig. 8). It is thus apparent that the crystallite size and surface area have an important role in enhancing the catalytic activity.

Fig. 7 CO oxidation as a function of temperature for the Mn₂O₃/SiO₂_1T (blue triangles), Mn₂O₃/SiO₂_2T (black circles), and Mn₂O₃/SiO₂_3T (red squares) products. Reaction conditions: 3.5% CO and 8.4% O₂ balanced with helium. Flow rate: 52 mL min⁻¹.

Fig. 8 CO conversion efficiencies of Mn₂O₃/SiO₂_3T (red squares) and Mn₃O₄/SiO₂_3T (red circles). Data regarding commercial bulk powders consisting of Mn₂O₃ (green inverse triangles) and Mn₃O₄ (black squares) are also included. Reaction conditions: 3.5% CO and 8.4% O₂ balanced with helium. Flow rate: 52 mL min⁻¹.

It is known that besides Mn₂O₃, other manganese oxides are also active with regard to such catalytic reactions. We initially evaluated the activity of Mn₃O₄ because it is considered more active with regard to CO oxidation compared with MnO. Hollow microspheres containing a Mn₃O₄ component, denoted as Mn₃O₄/SiO₂_3T, were simply prepared by annealing the Mn₂O₃/SiO₂_3T microspheres under a reducing atmosphere. Their catalytic performance towards CO oxidation was examined under identical conditions. The XRD data showed that only the Mn₃O₄ phase was present (Fig. S1†) and no peaks corresponding to the Mn₂O₃ phase were detected. The TEM images of the Mn₃O₄/SiO₂_3T product revealed that the surface morphology of the Mn₃O₄/SiO₂_3T product was virtually identical to that of the Mn₂O₃/SiO₂_3T product (Fig. S3†).

As illustrated in Fig. 8, the Mn₃O₄/SiO₂_3T catalyst did not show any improvement in activity compared with that of the Mn₂O₃/SiO₂_3T catalyst. Surprisingly, CO oxidation did not occur at temperatures below 200 °C. This result suggests that to promote the oxidation reaction, the Mn³⁺ species are more crucial than the Mn²⁺ ions. Presumably, the Mn³⁺ ions in the Mn₂O₃/SiO₂_3T product produce active sites for CO oxidation more effectively than the Mn²⁺ and Mn³⁺ ions in the Mn₃O₄/SiO₂_3T product. Another plausible scenario is that the spinel structure of Mn₃O₄ may be associated with the lower activity of the Mn₃O₄/SiO₂_3T catalyst. The oxide ions of Mn₃O₄ have a cubic close-packed arrangement, which may hinder the generation of oxygen vacancies. The lack of oxygen vacancies in the Mn₃O₄/SiO₂_3T product could reduce the mobility of the O²⁻ species, which may consequently reduce the activity.

For further comparison, under identical conditions, we also compared the CO oxidation properties of our products with those of other nickel-based catalysts, which are also considered active materials for CO oxidation. We synthesized hollow microspheres of NiO/SiO₂_3T and Ni/SiO₂_3T using the same dissolution and deposition method. The data obtained by the XRD and TEM analyses are given in Fig. S4 and S5,† respectively. With regard to CO conversion, we determined that the two nickel-based catalysts are much less active than the Mn₂O₃/SiO₂_3T catalyst (Fig. 8). The T₅₀ temperatures are determined to be approximately 312 °C and 325 °C for the NiO/SiO₂_3T and Ni/SiO₂_3T products, respectively, suggesting that the Ni and NiO NPs do not actively promote CO oxidation at low temperature.

As summarized in Table S1,† the T₅₀ temperature of the Mn₂O₃/SiO₂_3T product is much lower than those of other Mn-based catalysts reported thus far, demonstrating that this product exhibits excellent catalytic performance at low temperature. This value is even comparable to those achieved by Au/Mn₂O₃–Al₂O₃ (entry 7) and Ag/γ-Mn₂O₃ (entry 15) products, where expensive Au and Ag particles were combined with Mn₂O₃. It is worthy to mention that catalysts primarily on the basis of inexpensive metal oxides also show good catalytic activities under comparable conditions. Their catalytic properties are summarized in Table S2.†

Conclusions

Using the dissolution and deposition method, we have successfully developed a convenient method for the preparation of hollow microspheres composed of manganese oxides and SiO₂, without the use of etching. The resulting microspheres, Mn₂O₃/SiO₂_3T, with robust shells have large surface areas and exhibit excellent catalytic performance for CO oxidation even at temperatures below 200 °C, and can thus be used as a low cost alternative to noble metal catalysts. Future work will further evaluate the Mn₂O₃/SiO₂_3T product using simulated exhaust gases, which may achieve complete resolution for the control of CO emissions.

Acknowledgements

NHH thanks Sogang University (201610022.01) and C1 Gas Refinery Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (2016M3D3A1A01913276) for financial support. JMK also acknowledges the partial support by the Degree and Research Center (DRC) Program (2014) through the National Research Council of Science and Technology (NST) from the Ministry of Science, ICT and Future Planning.

Notes and references

1. J. Gaudet, A. Riva, E. Peterson, T. Bolin and A. Datye, ACS Catal., 2013, 3, 846.
2. M. Castegnaro, A. Killan, I. Baibich, M. Alves and J. Morais, Langmuir, 2013, 29, 7125.
3. S. Y. Hwang, E. Yurchekfrodl, C. Zhang and Z. Peng, ChemCatChem, 2016, 8, 97.
4. K. Sato, H. Tomonaga, T. Yamamoto, S. Matsumura, N. Zulkifli, T. Ishimoto, M. Koyama, K. Kusada, H. Kobayashi, H. Kitagawa and K. Nagaoka, Sci. Rep., 2016, 6, 28265.
5. W. Zhu, S. Xiao, D. Zhang, P. Liu, H. Zhou, W. Dai, F. Liu and H. Li, Langmuir, 2015, 31, 10822.
6. K. Chang, X. Hai and J. Ye, Adv. Energy Mater., 2016, 6, 1502555.
7. J. Cao, Z. Wang, R. Wang and T. Zhang, CrystEngComm, 2014, 16, 7731.
8. S. Swaminathan and M. Spiegel, Appl. Surf. Sci., 2007, 253, 4607.
9. J. Cao, Y. Zhu, K. Bao, L. Shi, S. Liu and Y. Qian, J. Phys. Chem. C, 2009, 113, 17755.
10. J. Fei, Y. Cui, X. Yan, W. Qi, Y. Yang, K. Wang, Q. He and J. Li, Adv. Mater., 2008, 20, 452.
11. X. Liu, H. Yang, L. Han, W. Liu, C. Zhang, X. Zhang, S. Wang and Y. Yang, CrystEngComm, 2013, 15, 7769.
12. L. Zhang, H. Y. Kim and G. Henkelman, J. Phys. Chem. Lett., 2013, 4, 2943.
13. J. H. Yang, J. D. Henao, M. C. Raphulu, Y. Wang, T. Caputo, A. J. Groszek, M. C. Kung, M. S. Scurrell, J. T. Miller and H. H. Kung, J. Phys. Chem. B, 2005, 109, 10319.
14. K. Zorn, S. Giorgio, E. Halwax, C. R. Henry, H. Gronbeck and G. Rupprechter, J. Phys. Chem. C, 2011, 115, 1103.
15. C. J. Karwacki, P. Ganesh, P. R. C. Kent, W. O. Gordon, G. W. Peterson, J. J. Niu and Y. Gogotsi, J. Mater. Chem. A, 2013, 1, 6051.
16. A. Binder, T. Toops, R. Unocic, J. Parks and S. Dai, Angew. Chem., Int. Ed., 2015, 54, 13263.
17. V. Iablokov, R. Barbosa, G. Pollefeyt, I. Driessche, S. Chenakin and N. Kruse, ACS Catal., 2015, 5, 5714.
18. P. Li, D. Miser, S. Rabiei, R. Yadav and M. Hajaligol, Appl. Catal., B, 2003, 43, 151.
19. Y. Ren, Z. Ma, L. Qian, S. Dai, H. He and P. Bruce, Catal. Lett., 2009, 131, 146.
20. J. E. Post, Proc. Natl. Acad. Sci. U. S. A., 1999, 96, 3447.
21. D. M. Robinson, Y. B. Go, M. Mui, G. Gardner, Z. Zhang, D. Mastrogiovanni, E. Grarfunkel, J. Li, M. Greenblatt and G. C. Dismukes, J. Am. Chem. Soc., 2013, 135, 3494.
22. S. K. Nayak and P. Jena, Phys. Rev. Lett., 1998, 81, 2970.
23. J. E. Pask, D. J. Singh, I. I. Mazin, C. S. Hellberg and J. Kortus, Phys. Rev. B: Condens. Matter Mater. Phys., 2001, 64, 024403.
24. A. E. Fischer, K. A. Pettigrew, D. R. Rolison, R. M. Stroud and J. W. Long, Nano Lett., 2007, 7, 281.
25. Y. Han, F. Chen, Z. Zhong, K. Ramesh, L. Chen and E. Widjaja, J. Phys. Chem. B, 2006, 110, 24450.
26. S. Ching, D. A. Kriz, K. M. Luthy, E. C. Njagi and S. L. Suib, Chem. Commun., 2011, 47, 8286.
27. M. Kramer, T. Schmidt, K. Stowe and W. F. Maier, Appl. Catal., A, 2006, 302, 257.
28. R. Craciun, B. Nentwick, K. Hadjiivanov and H. Knozinger, Appl. Catal., A, 2003, 243, 67.
29. K. Ramesh, L. Chen, F. Chen, Y. Liu, Z. Wang and Y. Han, Catal. Today, 2008, 131, 477.
30. W. Stöber, A. Fink and E. Bohn, J. Colloid Interface Sci., 1968, 26, 62.
31. J. B. Yoo, H. S. Kim, S. H. Kang, B. Lee and N. H. Hur, J. Mater. Chem., 2014, 2, 18929.
32. P. Burattin, M. Che and C. Louis, J. Phys. Chem. B, 1998, 102, 2722.
33. L. Hermida, A. Z. Abdullah and A. R. Mohamed, Mater. Sci. Appl., 2013, 4, 52.
34. M. Choi, C. Kim, S. O. Jeon, K. S. Yook, J. Y. Lee and J. Jang, Chem. Commun., 2011, 47, 7092.
35. J. Lee, S. H. Hwang, J. Yun and J. Jang, ACS Appl. Mater. Interfaces, 2014, 6, 15420.
36. B. J. Tan, K. J. Klabunde and P. M. A. Sherwood, J. Am. Chem. Soc., 1991, 113, 855.
37. S. Swaminathan and M. Spiegel, Appl. Surf. Sci., 2007, 253, 4607.
38. P. Pal, A. K. Giri, S. Mahanty and A. B. Panda, CrystEngComm, 2014, 16, 10560.
39. M. Zaki, M. Hasan, L. Pasupulety and K. Kumari, New J. Chem., 1998, 22, 875.
40. Y. Han, F. Chen, Z. Zhong, K. Ramesh, L. Chen and E. Widjaja, J. Phys. Chem. B, 2006, 110, 24450.
41. F. Kapteijn, A. Langeveld, J. Moulijn, A. Andreini, M. Vuurman, A. Turek, J. Jehng and I. E. Wachs, J. Catal., 1994, 150, 94.
42. H.-P. Lin, S.-T. Wong, C.-Y. Mou and C.-Y. Tang, J. Phys. Chem. B, 2000, 104, 8967.
43. J. Li, J. Lin, X. Xu, X. Zhang, Y. Xue, J. Mi, Z. Mo, Y. Fan, L. Hu, X. Yang, J. Zhang, F. Meng, S. Yuan and C. Tang, Nanotechnology, 2013, 24, 155603.
44. R. Jia, J. Chen, J. Zhao, J. Zheng, C. Song, L. Li and Z. Zhu, J. Mater. Chem., 2010, 20, 10829.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra20868a

---