SnO2 is an n-type semiconductor with abundant surface deficient oxygen species [1, 2] and lattice oxygen that is also reducible [1, 3]. In addition,it is stable chemically with a melting point of 1630 °C [4]. Over the past decades,its properties as a gas sensing material have been widely studied [5, 6, 7]. Although it is also a potential catalytic material,especially for air pollution control reactions,studies on its catalytic properties are relatively few. During the past four years,our group has performed a series of systematic work on understanding its catalytic chemistry [3, 8, 9, 10, 11, 12, 13, 14, 15, 16]. Regular SnO2 fine powder prepared by the traditional precipitation method generally has a low surface area below 20 m2/g after calcination above 500 °C [9, 11, 12],which limits its CO and CH4 oxidation reactivity. The introduction of other metal cations such as Fe [17, 18],Cr [3, 17, 18],Mn [18],Ce [9],or Ta [8] into its lattice to form a solid solution can increase its surface area significantly and induce the formation of more active oxygen species,thus increasing its activity as well as thermal stability. On the other hand,some metal cations in the +3 valence state can react with SnO2 to form A2B2O7 pyrochlore compounds,in which Sn occupies the B site [14, 15]. These pyrochlore compounds generally possess very good thermal stability and oxygen vacancies,which makes them potential catalysts for environmental catalytic reactions. In addition,SnO2 was also reported to be a good support for noble metals as catalysts for CO and CH4 oxidation [11]. Fuller and co-workers [19] found that Pd/SnO2 showed much higher activity than Pd/SiO2 due to a synergetic effect between Pd and SnO2. Furthermore,the addition of water vapor into the reaction feed can improve the CO oxidation reactivity on Pd/SnO2,which has been confirmed by our experiments on Pd/SnO2/Al2O3 catalysts [12]. When studying Pd/SnO2 catalysts,Eguchi and co-workers [20] also found the presence of synergism between the noble metal and the support,which enhanced the activity of the catalysts significantly.
Besides these studies on regular SnO2 fine powder,our most recent results demonstrated that a SnO2 nano-rod also showed superior CO oxidation reactivity [10],although its surface area is extremely low (1 m2/g). The formation of uniform nano-rod microcrystals creates preferentially exposed (110) facets,which have been reported to be the active facets for SnO2 [10, 21]. Using density functional theory calculations,Yang and co-workers [21] also proved that the (110) planes are the active facets of SnO2. As a result,the electronic properties of the Sn cations were altered,and the SnO2 nano-rod showed the reaction behavior of noble metal catalysts. H2 temperature-programmed reduction (H2-TPR) testified that different from the regular SnO2 fine powder,the lattice oxygen of this nano-rod is non-reducible due to the formation of a rigid crystal structure. Therefore,CO oxidation on it would follow the Langmuir-Hinshelwood or Eley-Rideal mechanism,which is typical of noble metal catalysts. For comparison,using a simple co-precipitation method,uniform mesoporous Cu-SnO2 nano-sheets with a high surface area (196 m2/g) and more active surface oxygen species were successfully prepared,which also showed remarkable activity for CO oxidation [16].
Many studies have demonstrated that when prepared with a special morphology,SnO2 possesses particular characteristics that improve its gas sensing properties [5, 6, 7, 22]. However,studies on SnO2 with different morphologies as catalytic materials have been rare. To more deeply understand the catalytic chemistry of SnO2 and achieve improved catalysts for CO oxidation,in this study,SnO2 catalysts with different morphology and compositions were prepared by various methods. Mesoporous SnO2 nano-sheet with a much higher surface area than regular SnO2 fine powder can be prepared,which displayed significantly improved CO and CH4 oxidation reactivity. Using multiple characterization techniques,the reasons for the improved reaction performance were elucidated.
SnO2 nano-sheet was prepared according to the method reported in reference [23]. SnCl4·5H2O (0.7 g) was added into 50 mL distilled deionized (DDI) water (solution A). Urea (0.12 g) was added into 50 mL DDI water (solution B) to prepare a stable solution. Then solution B was added slowly into solution A with continuous stirring. After intense ultrasonic treatment,the solution mixture was transferred into a Teflon-lined stainless steel autoclave for crystallization at 160 °C for 48 h. After cooling to room temperature,the solid was obtained by filtration,washed repeatedly and thoroughly with DDI water and dried at 110 °C overnight in air atmosphere. The final catalyst was named SnO2-NS because the SEM image in Fig. 1(a,b) demonstrated that it consisted of neat and uniform nano-sheets.
SnO2 nano-rods were prepared according to the method reported in reference [24]. In a typical synthesis process,0.9 g SnCl4·5H2O was added into 100 mL ethanol/DDI water (1:1,v/v) mixture followed by the addition of 1.6 g KOH. After intense ultrasonic treatment,the solution mixture was transferred into a Teflon-lined stainless steel autoclave for crystallization at 180 °C for 24 h. After cooling to room temperature,the product was obtained by filtration,washed repeatedly and thoroughly with DDI water and dried at 110 °C overnight in an air atmosphere. The SEM image in Fig. 1(c) proved that this sample consisted of both SnO2 nano-rods and nanoparticles,therefore it was named SnO2-NR+NP.
For comparison with our previous results,the traditional precipitation method was used to prepare regular SnO2 nanopowder according to our previous publication [10]. This was named SnO2-NP because the SEM image in Fig. 1(d) demonstrated that it consisted of spherical particles.
The microstructure and morphology of the synthesized products were characterized by scanning electron microscopy (SEM). Small amounts of the dried powders were dispersed in ethanol,and a few drops were dripped onto the silicon slice support. The samples were then viewed in a Hitachi S-4800 field emission scanning electron microscope at 30 kV.
Powder X-ray diffraction (XRD) patterns were recorded on a Bruker AXS D8 Focus diffractometer instrument operating at 40 kV and 30 mA with Cu Kα radiation (λ = 0.154178 nm). Scans were taken in a 2θ range of 10°-90° and with a step of 0.02°/s. The mean crystallite size of the samples was calculated with the Scherrer equation based on the strongest peak of SnO2 with an (hkl) of (101),for which the 2θ value is 33.8°.
N2 adsorption was used to examine the porous and textural properties of the samples. The measurements were carried out at -196 °C on a Micromeritics ASAP 2020 apparatus. All the samples were pretreated in vacuum at 200 °C for 5 h before the measurement. The surface area and pore volume were obtained by the Brunauer-Emmet-Teller (BET) and Barrett- Joyner-Halenda (BJH) methods,respectively.
H2-TPR experiments were carried out on a FINESORB 3010C instrument with a 10% H2/Ar gas mixture flow. The temperature was increased from room temperature to 800 °C at a rate of 10 °C /min. A 25 mg sample was used for the test. A thermal conductivity detector (TCD) was employed to monitor the H2 consumption. For H2 consumption quantification,a calibration experiment was carried out using a high purity CuO (99.99%) sample.
X-ray photoelectron spectroscopy (XPS) tests were carried out on a Perkin-Elmer PHI1600 system using a single Mg K X-ray source operating at 300 W and 15 kV. The spectra were obtained at ambient temperature with an ultrahigh vacuum. The binding energies were calibrated using the C 1s peak of graphite at 284.6 eV as a reference.
The catalysts were evaluated for CO and CH4 oxidation in a continuous flow reactor with a gas composition of 1% CO (or 1% CH4),21% O2 and balanced by high purity N2. Typically,100 mg catalyst particles (0.2-0.3 mm) were used for each measurement with a flow rate of 30 mL/min,which corresponded to a weight hourly space velocity (WHSV) of 18000 mL h-1 g-1. The reactants and products were analyzed online on a GC9310 gas chromatograph equipped with a TDX-01 column and a TCD.
SEM was used to investigate the morphology of the SnO2 samples prepared with the different methods. The images are shown in Fig. 1. Fig. 1(a) and (b) demonstrated that the SnO2-NS sample consisted of uniform sheets with many loose pores and a thickness of 10 nm,indicating that SnO2 nano- sheets were successfully prepared. However,Fig. 1(c) shows an image containing both nano-rods and irregular spherical particles. Apparently,with the method used in this study,100% pure nano-rods were not successfully synthesized. The sample comprised both nano-rods and nanoparticles. In contrast,SnO2-NP,the sample prepared with a traditional precipitation method for comparison purpose,consisted of relatively uniform spherical particles with an average diameter of 21 nm (Fig. 1(d)),except that some particles have accumulated into larger grains. In summary,with the different methods,SnO2 with different morphologies were obtained.
The activity of the SnO2 catalysts prepared by different methods was first evaluated using CO oxidation as the probe reaction. The results are shown in Fig. 2(a). SnO2-NP,the reference sample,showed the lowest activity among the three catalysts,with complete CO oxidation achieved at 360 °C. SnO2-NS,the nano-sheet sample,showed the highest activity on which complete CO oxidation occurred at 240 °C. The parallel shift of the CO conversion curve to a 100 °C lower temperature region in comparison with SnO2-NP also testified its much improved oxidation activity. SnO2-NR+NP showed also higher CO oxidation activity than the regular SnO2-NP sample,but which was lower than that of SnO2-NS. Thus,SnO2 with special morphologies showed significantly improved CO oxidation activity over the regular SnO2 nanoparticles.
To gain a deeper understanding of the modification due to the morphology change,the reaction rate at 140 °C and the activation energy of the catalysts were calculated. The measurements were performed under differential condition with CO conversion below 20% to exclude mass and heat transfer effects. As listed in Table 1,SnO2-NS showed the highest reaction rate and lowest activation energy,and SnO2-NP showed the lowest reaction rate and highest activation energy. In comparison,both the reaction rate and activation energy over SnO2-NR+NP were between those of the above two samples.
The CH4 oxidation reactivity of the samples was also probed and compared in Fig. 2(b). The reaction rate at 440 °C and activation energy are listed also in Table 1. Since CH4 contains only four strong C-H bonds,it requires a much higher activation energy for reaction compared with CO oxidation [9]. Therefore,the reactivity difference was not as evident as in the CO oxidation case. However,both the reaction rate and activation energy basically followed the same sequence as CO oxidation. In summary,CO and CH4 oxidation results testified that with the formation of SnO2 nano-sheets and nano-rods,more active sites were introduced onto the surface of the samples,hence improving the activity.
To identify the phase composition and crystallinity of the SnO2 samples prepared by the different methods,XRD analysis was performed. The results are shown in Fig. 3. All the samples showed the diffraction features of the tetragonal rutile SnO2 phase,testifying that although they have different morphologies,they consisted of the pure rutile SnO2 phase. However,the different peak intensity of the diffraction peaks indicated that the three samples have different crystallinity. The crystallite sizes of the samples were calculated and listed in Table 2. SnO2-NP,the regular fine powder sample,has the largest crystallite size,while SnO2-NS,the nano-sheet sample,has the smallest. In addition,the crystallite size of SnO2-NR+NP was also much smaller than that of SnO2-NP but larger than that of SnO2-NS. This indicated that with the formation of special morphologies,the crystallization of the SnO2 can be impeded.
It was formerly reported that the (110) plane is the active crystal facet for SnO2 [10, 21]. A SnO2 sample with preferentially exposed (110) facets generally has improved oxidation reactivity. Therefore,to elucidate if the samples have preferentially exposed the (110) plane,the I(110)/I(101) ratios of the samples were calculated and also listed in Table 2. The intensity of the (101) peak,the strongest peak,was used as an internal standard. While SnO2-NS and SnO2-NP have a similar value for this ratio,SnO2-NR+NP has an evidently improved one,which testified that this sample has more exposed (110) facets. This was possibly due to the presence of SnO2 nano-rods in its composition. Thus,the preferentially exposed (110) facets would be one of the reasons for its improved CO and CH4 oxidation reactivity.
The textural and structural properties of the catalysts were also measured by N2 adsorption. The results are shown in Fig. 4 and Table 2. SnO2-NS possessed a surface area of 54 m2/g,which was nearly 2 times that of SnO2-NR+NP and 3 times that of SnO2-NP. The change in surface area was in line with the sequence of the crystallinity of the samples measured by XRD.
As shown in Fig. 4,all the samples have a type IV adsorption isotherm with a H3 hysteresis loop. However,compared with SnO2-NR+NP and SnO2-NP,SnO2-NS has a much larger pore volume and pore size (Table 2),indicating that this sample possessed a large amount of mesopores. The presence of a large quantity of mesopores in the solid would facilitate the diffusion of reactants and products,which would improve the activity of the SnO2-NS catalyst.
The redox properties of the SnO2 samples with different morphologies were tested by H2-TPR. The results are shown in Fig. 5. The quantification of the H2 consumption amount and O/Sn atomic ratio for the samples are also listed in Table 3. SnO2-NP showed a reduction peak at 715 °C,which was assigned to the reduction of SnO2 to metallic Sn [3, 25]. In comparison,the same reduction peaks of SnO2-NR+NP and SnO2-NS were shifted to 710 and 687 °C,respectively,indicating that the lattice oxygen species of the latter two were more reducible and active,especially SnO2-NS. The O/Sn atomic ratios of the samples were around 2.0/1,which is the stoichiometric ratio for the reduction of Sn4+ to Sn0. This further confirmed that Sn was fully oxidized in the three samples,in agreement with the XRD phase identification results. The formation of much more active oxygen species in SnO2-NS would be another reason for the higher CO and CH4 oxidation reactivity.
To further understand the surface properties of the SnO2 catalysts with different morphologies,XPS measurements were performed. The results are shown in Fig. 6 and Table 4. The binding energies were calibrated by the C 1s internal standard. The binding energy of Sn 3d was typical for Sn4+ cations,indicating that the Sn species in all the samples were fully oxidized,in line with the XRD and H2-TPR results. The ∆E between Sn 3d3/2 and Sn 3d5/2 was calculated and listed in Table 4. This generally reflects the chemical environment change of the Sn species [9, 26]. SnO2-NP has a value of 8.6 eV. In comparison,the values of SnO2-NR+NP and SnO2-NS were shifted to 8.4 eV,indicating that in these two samples,the Sn species has a similar chemical environment,but which is different from that of the regular fine powder. This was possibly due to the formation of the nano-sheet and nano-rod structure in these samples.
It was previously reported that the surface of SnO2 has a large amount of deficient oxygen species [1, 2]. The asymmetric O 1s peaks of the three samples in this study also confirmed the presence of multiple oxygen species. Therefore,the O 1s peaks of the three samples were deconvoluted and shown in Fig. 6(b). As reported previously,a deconvoluted O 1s peak with a higher binding energy was assigned to loosely bounded surface oxygen species (Oads). The peak with the lower binding energy was assigned to the surface lattice oxygen species (Olat) [9]. Based on this,the Oads/Olat ratios of the three samples were calculated and also listed in Table 4. On the surface of the SnO2-NR+NP and SnO2-NS samples,more loosely bounded oxygen species were present in comparison with SnO2-NP. For CO oxidation,the presence of this active surface oxygen species is obviously beneficial for the activity of the catalyst.
For CO oxidation in practical emission control processes,generally,1%-10% water vapor is present. The stability of a catalyst in the presence of an amount of water vapor determines its application potential. Therefore,SnO2-NS,the most active catalyst in this study,was subjected to a stability test at 240 °C in the presence of 5% water vapor. As shown in Fig. 7,with the introduction of water vapor,the CO conversion dropped from 80% in the dry feed to 60% within 2 h,but which then remained constant without further decrease during the following 25 h test with water vapor. Most importantly,after removing the water vapor,the CO conversion was completely restored,testifying that the water deactivation was not permanent. This proved that SnO2-NS was both active and also structurally stable,which makes it a potential catalyst for low temperature CO oxidation in emission control processes.
Pure SnO2 samples with different morphologies and compositions were prepared and used for CO and CH4 oxidation. SnO2-NS consisted of uniform and neat nano-sheets with a thickness of 10 nm. N2 adsorption results demonstrated that in comparison with SnO2-NP,the SnO2-NS possessed a much higher surface area and larger pore volume,and contained a large amount of mesopores,which would improve the contact of the reactants with the surface active sites and facilitate the mass transfer of reactants and products. H2-TPR and XPS results showed that more active oxygen species were formed on the surface of this nano-sheet sample. As a result,SnO2-NS displayed much improved CO and CH4 oxidation reactivity in comparison with SnO2-NP. As for SnO2-NR+NP,SEM proved that it consisted of both nano-rods and nano-particles. Compared with SnO2-NP,it has a slightly higher surface area,slightly larger pore volume,and slightly more active surface oxygen species,but possessed more active (110) facets,which were reported to be the active facet for SnO2,therefore,it showed higher activity than SnO2-NP but lower than that of SnO2-NS. Water vapor has only reversible and mild deactivation on SnO2-NS,which makes it a potential catalyst for exhaust emission control.
2015, Vol. 36 Issue (11): 2004-2010
PDF (737 KB)
2015, Vol. 36 Issue (11): 2004-2010
PDF (737 KB)