Document Type : Research Paper
Authors
1 Assistant Professor, Nanotechnology Research Center, Research Institute of the Petroleum Industry, Tehran, Iran
2 M.S. Student, Environmental and Biotechnology Department, Research Institute of the Petroleum Industry, Tehran, Iran
Abstract
Cobalt oxide catalysts supported on oxidized multi-walled carbon nanotubes (MWCNT) for the low-temperature catalytic oxidation of carbon monoxide were prepared by an impregnation-ultrasound method. These catalysts were characterized by N2 adsorption/desorption, TEM, XRD, Raman, and H2-TPR methods. The XRD and Raman results indicated that the phase of the synthesized cobalt oxide was in the Co3O4 form. The effects of cobalt oxide loading and reaction temperature were studied on the catalytic oxidation conversion of carbon monoxide. The TEM image of the best catalyst (14 wt.% metal oxide loading) revealed a good dispersion of Co3O4 over the surface of the support with an average particle size of 11-16 nm. Under the reaction conditions of T= 200-250 °C, P=1 bar, CO = 600 ppm, O2 = 5 vol.%, GHSV = 30,000 hr.−1, and Co3O4= 14 wt.%, CO conversion was 91%.
Keywords
Main Subjects
CO from the exhaust gases of automobiles and combustion engines remains one of the most serious pollutants (Chai et al., 2017; Lv et al., 2016). A well-proven method for CO abatement is the catalytic oxidation of CO to carbon dioxide due to its economic advantages and simplicity (Lee et al., 2016; Dong et al., 2017). Although noble metal catalysts with/without different supports, including Au-CeO2 (Bansmann et al., 2017), gold nanoparticles (Qiao et al., 2016), Pd/TiO2 (Bratan et al., 2017), Pd/SnO2 (Jin et al., 2012) exhibit a high activity in this reaction, the high cost and easily sintering limit their extensive applications (Chai et al., 2017). Among the non-precious metal catalysts, transition metal oxide catalysts such as Co3O4 (Wang et al., 2017), CuO (Wang et al., 2017), and Fe2O3 (Cui et al., 2017) have attracted much attention in the catalytic oxidation reaction of CO due to their high activity and thermal stability (Bao et al., 2011). Some research findings have shown that the catalytic activity and redox property of Co3O4 are strongly influenced by the formation of oxygen vacancies (Zhang et al., 2017; Chen et al. 2017), controlling the crystalline size (Kant Sharma et al., 2017) and its oxidation state (Chen et al., 2016). Therefore, the performance of cobalt oxide, which is one of the most promising catalysts for the complete oxidation of CO, appears to be critically dependent on the preparation method (Yuan et al., 2017; Liu et al., 2017) and the nature of the support (Wang, 2016).
Carbon nanotubes have proved attractive and competitive as a catalyst support in heterogeneous catalytic reactions due to their distinctive morphology, high thermal resistance, and conductivity (Payan et al., 2018; Fattahi et al., 2015; Liu et al., 2018; Zhou et al., 2017). CNTs, with the unique one-dimensional tubular and mesoporous structure, simplify the transmission of reactants and products in chemical reactions (Huang et al., 2007; Santillan-Jimenez et al., 2011). CNTs have a hydrophobic surface with low solubility and dispersivity in polar solvents (Mazov et al., 2012). The introduction of oxygenated groups such as (OH and COOH) to CNT surface has been reported to increase the wetting characteristics of CNTs and to increase the dispersion of metal particles on their surface (Kundu et al., 2008). These oxygenated groups can bond metal oxides onto the CNT surface (Chuang et al., 2011). Gao and coworkers (Gao et al., 2015) have reported that CuO-CeO2 catalyst supported on OMWNTs, with the nominal ratio of Cu to Ce equal to 5/5 and a total metal oxide loading of 20 wt.%, had a high activity and selectivity for PROX of CO in comparison with CuO-CeO2 catalysts supported on other materials, including activated carbon, γ-Al2O3, and SiO2 in an H2-rich stream. In this work, cobalt oxide catalysts supported on OMWNTs were prepared by an ultrasound-assisted impregnation method, and their performance was tested for the catalytic oxidation of CO at low temperatures. The effects of CoOx loading and reaction temperature on conversion and selectivity were also studied. Phase transformation, redox, and structural properties were characterized by N2 adsorption/desorption, XRD, Raman, TEM, and H2-TPR techniques.
2. Experimental
2.1. Catalyst preparation and characterization
a. Oxidation of the support
The employed MWNTs (95% purity, 10–30 nm in OD) were prepared by a chemical vapor deposition (CVD) method over a Co–Mo/MgO catalyst at Research Institute of the Petroleum Industry (RIPI), Iran (Rashidi et al., 2007). Following the purification of the raw MWNTs, they were oxidized in a 1:3 (v/v) solution of HNO3 (65%) and H2SO4 (98%) in an ultrasonic bath (Fisher Scientific, 130 W and 40 kHz) maintained at 60 °C for 2 hrs. The solution was then filtered, and washed using de-ionized water to remove excess acid; it was then dried in the oven at 110 °C in air for 24 hrs (Xing et al., 2005).
b. Catalyst preparation
Supported cobalt oxide catalysts at different loadings (2–18 wt.%) were prepared by the impregnation technique aided by ultrasonic treatment. The OMWNTs were impregnated using certain amounts of an aqueous solution of Co(NO3)2 6H2O as the precursor for cobalt oxide. The solutions were subjected to an ultrasonic treatment for 1 hr. Afterwards, the samples were first dried in a rotary evaporator at 78 °C and then in an oven at 100 °C for 24 hrs. Loading of cobalt oxide on the OMWNTs was controlled by adjusting the ratio between the OMWNTs and cobalt nitrate. The catalysts were calcined in a tubular furnace at 400 °C for 3 hrs under a blanket of argon as a carrier gas to obtain CoOx/OMWNTs catalysts at different loadings.
c. Catalyst characterization
The specific surface areas of the support and catalysts were measured by nitrogen adsorption at -196 °C with the BET method on a Belsorp II (BEL Co., Japan) instrument. The samples were degassed at 250 °C for 3 hrs under vacuum. Total pore volume and mean pore diameter of the samples were calculated from adsorption and desorption isotherms at a relative pressure (P/P° of 0.98). The crystalline phase of the support and cobalt oxides over the OMWNTs was characterized by XRD. The measurements were obtained using an X’Pert MPD-PHILIPS X-ray diffractometer with Cu Ka (k = 1.5056 Å) radiation. The voltage and current of the anode were 40 kV and 30 mA respectively. The scanning range of 2θ was from 5 to 70 ° with a step size of 0.04 °/s. The diffraction patterns were manually analyzed with the Joint Committee of Powder Diffraction Standard (JCPDS) card. Raman spectra were employed at ambient temperature on an Almega Thermo Nicolet Dispersive Raman Spectrometer with an argon-ion laser at an excitation wavelength of 514 nm. The TEM observations were made with a Zeiss EM900 microscope operated at 120 keV to observe the dispersion of active sites on the support surface. The samples were prepared by the ultrasonic dispersion of the catalysts in ethanol, and the suspensions were then dropped onto a carbon-coated copper grid. The temperature programmed reduction (TPR) experiments were performed on a Micromeritics-2900 system equipped with a thermal conductivity detector (TCD). 50 mg of the catalyst was reduced in a flow of 5% H2 under the blanket of argon at a flow rate of 40 cc min−1 and at a linear heating rate of 10 °C min−1 from 25 to 800 °C.
2.2. Catalytic activity tests
The CO oxidation experiments were carried out in a fixed bed quartz reactor at atmospheric pressure between 100 and 250 °C. The gas composition was 600 ppm CO and 5 vol.% O2 balanced with He as the carrier gas. Before adding the reactant gas into the reactor, the gases were mixed in a glass chamber. In all the tests, the total flow rate was fixed at 500 cc.min-1, which corresponds to a GHSV of 30,000 hr.-1. CO and O2 concentrations in the inlet and outlet gases were measured using an on-line flue gas analyzer (Testo model 340, Germany) equipped with CO and O2 sensors (Figure 1). The reaction temperature was controlled by a K-type thermocouple inserted directly into the catalyst bed. The catalytic activity was calculated by the following equation:
(1) |
Figure 1
A schematic of the activity test system.
3. Results and discussion
3.1. Effect of reaction temperature
The activity of the support and the 14 wt.% CoOx/OMWNTs in the CO oxidation at reaction temperatures between 100 and 250 °C is detailed in Figure 2. In this temperature range, CO conversion on the OMWNTs was very low (5–6%, Figure 2a), which could be attributed to the CO adsorption by carbon nanotubes (Azizi et al., 2011). As the temperature increased, the CO conversion increased from 55% at 100 °C to 83% at 150 °C; it finally reached a maximum conversion of 91% at temperatures more than 200 °C (Figure 2b); thus, the optimum low reaction temperature over CoOx/OMWNTs for CO oxidation was 200 °C.
Figure 2
Effect of reaction temperature on CO conversion of a) OMWNTs and b) 14 wt.% CoOx/OMWNTs.
3.2 Effect of CoOx loading
The effect of different cobalt oxide loadings (2–18 wt.%) on CO conversion is displayed in Figure 3. Over all the reaction temperatures, changing the CoOx loadings from 2 to 14 wt.% resulted in an increased conversion due to an increased number of active sites. By increasing CoOx loading over the support to 18 wt.%, the agglomeration of the active sites occurred, and the catalyst activity declined.
Figure 3
Effect of CoOx loading at different reaction temperatures on CO conversion.
The surface area, total pore volume, and mean pore diameter of the OMWNTs and 10 wt.%, 14 wt.%, and 18 wt.% CoOx/OMWNTs are tabulated in Table 1. After impregnation, the surface area of the catalysts increased, but the pore volume and pore diameter dropped by the formation of cobalt oxide nanoparticles on the support. On increasing the cobalt loading from 10 to 14 wt.%, the changes in total pore volume and mean pore diameter were negligible, whereas the catalyst had the maximum surface area (208 m2gr.-1) (see Table 1).
Table 1
Surface area, total pore volume, and mean pore diameter of OMWNTs and CoOx/OMWNTs.
Samples |
SBET (m2gr.-1) |
Total pore volume (cm3gr.-1) |
Mean pore diameter (nm) |
OMWNTs |
185.8 |
0.86 |
18.56 |
10 wt.% cobalt oxide |
199.2 |
0.69 |
13.38 |
14wt.% cobalt oxide |
208 |
0.65 |
13.39 |
18 wt.% cobalt oxide |
191.3 |
0.53 |
10.95 |
A TEM image of cobalt oxide nanoparticles (14 wt.%) over OMWNTs in Figure 4a shows a good dispersion of the active phase on the surface of the support with an average particle size of 11-16 nm. However, the surface area, pore volume, and pore diameter of the 18 wt.% CoOx catalyst were all decreased due to the agglomeration of CoOx. A TEM image of the 18 wt.% CoOx catalyst is illustrated in Figure 4b. Figure 5 depicts the XRD patterns of the pure cobalt oxide (prepared by the ultrasonic-assisted impregnation method), OMWNTs, and CoOx/OMWNTs (10 wt.%, 14 wt.%, and 18 wt.%).
Figure 4[E1]
TEM image of a) 14 wt.% CoOx/OMWNTs and b) 18 wt.% CoOx/OMWNTs.
Over the pure cobalt oxide (Figure 5a), diffraction peaks observed at 17.0, 30.3, 36.8, 39.0, 46.2, 59.4, and 64.2 ° can be indexed as (1 1 1), (2 2 0), (3 1 1), (2 2 2), (4 0 0), (5 1 1), and (4 4 0) lattice planes of spinel Co3O4 (JCPDS no ≠ 76-1802) respectively. No diffraction peaks corresponding to CoO or Co phases are observed. Figure 5b displays the peaks at 2θ of 25.7, 41.9, 43.16, and 50.9° for the (002), (100), (101), and (004) diffractions of graphite (JCPDS # 08-0415), indicating that the structure of the MWNTs support was not destroyed through the ultrasonic treatment (Belin and Epron, 2005).
Figure 5
XRD patterns of a) cobalt oxide, b) OMWNTs, and c) CoOx/OMWNTs catalysts 10 wt.% loadings, d) CoOx/OMWNTs catalysts 14 wt.% loadings, and e) CoOx/OMWNTs catalysts 18 wt.% loadings.
Over CoOx/ OMWNTs, the intensity of Co3O4 peaks increased slightly at higher cobalt loadings. The Raman spectra of MWNTs, OMWNTs, and 14 wt.% cobalt oxide are drawn in Figure. 6. It can be seen that MWNTs have two main peaks at about 1345 and 1567 cm−1 (Figure 6a). The peak around 1345 cm−1 (D-band) is associated with the vibrations of carbon atoms in the disordered graphite structure, i.e., the defects. The G-band is due to the first order scattering of the E2g mode of graphite. Therefore, the ratio of D-band intensity to G-band (ID/IG) is related to the defect density in carbon nanotubes and provides direct information on the oxidation degree of carbon nanotubes (Zou et al., 2008; Chaunchaiyakul et al., 2016). As shown in Figure 6, the ID/IG ratio of the OMWNTs is higher than that of MWNTs, confirming the formation of defect sites by the addition of the oxygen groups. The Raman scattering reflections in Figure 6c at 482 cm-1 is ascribed to the Eg mode of Co3O4, and the reflections at about 520 and 625 cm-1 are attributed to F2g; the reflection at 695 cm-1 stands for the A1g mode of Co3O4 (Pu et al., 2017). These results confirm the existence of the crystalline structure of cobalt oxide in the Co3O4 form, which is in agreement with the XRD results. Decreasing the ID/IG ratio of the OMWNTs from 1.21 to 1.17 in the 14 wt.% catalyst can be attributed to the anchoring of cobalt oxide nanoparticles onto the defect sites of OMWNTs.
Figure 6
Raman spectra of a) MWNTs, b) OMWNTs, and c) 14 wt.% cobalt oxide/OMWNTs.
Figure 7 displays the H2-TPR analysis of the pure cobalt oxide, OMWNTs, and CoOx/OMWNTs (10 wt.%, 14 wt.%, and 18 wt.%) to investigate their reduction behavior. The TPR profile of the cobalt oxide with two typical reduction peaks is plotted in Figure 7a, which indicates a two-step reduction process of Co3O4 to CoO at about 250–350 °C and CoO to metallic cobalt in the region of 350–560 °C according to Equations 2 and 3 (Tang, Wang and Chien, 2008):
(2) |
|
(3) |
Xia et al. (Xia et al., 2010) found out that the reduction profile of Co3O4 includes a low-temperature peak in the range of 200–350 °C and a high-temperature peak at approximately 500 °C. With respect to the H2-TPR profile of the catalysts (10 wt.%, 14 wt.%, and 18 wt.%) in Figures 7c, 7d, and 7e, the existence of these two main peaks proves the presence of cobalt oxide in the form of Co3O4,which is in agreement with the results reported by XRD (Figure 5) and Raman (Figure 6) studies. Comparing the profile of the support (Figure 7b) with that of the catalysts (Figures 7c, 7d, and 7e), the increase in the temperature range of the second peak to 750 °C can be attributed to the carbon methanation of the support in the presence of hydrogen at high temperatures (> 500 °C) (Xiong et al., 2013).
Figure 7
H2-TPR profile of a) cobalt oxide, b) OMWNTs, c) 10 wt.% cobalt oxide catalyst, d)14 wt.% cobalt oxide catalyst, and e) 18 wt.% cobalt oxide catalyst.
4. Conclusions
A series of CoOx over oxidized multi-walled carbon nanotube catalysts have been prepared by an impregnation-ultrasound method at different loadings (2-18 wt.%). The activity of the prepared catalysts was examined for the catalytic oxidation of carbon monoxide. The effects of CoOx loading and reaction temperature on CO conversion were also discussed. Increasing CoOx loading from 2 to 14 wt.% increased conversion, while higher loadings to 18 wt.% led to the agglomeration of cobalt oxide particles and increased catalyst crystallinity. The XRD and Raman profiles of the catalysts confirmed that the cobalt oxide was in the form of Co3O4, which is very active in redox reactions. The catalyst having 14 wt.% metal loading had the highest surface area and a good dispersion of Co3O4 in the range of 11-16 nm; in addition, it exhibited the maximum catalytic performance toward the oxidation of CO into CO2 at about 91% at temperatures more than 200 °C. Under the objective to evaluate the capability of the prepared catalyst for CO removal from exhaust gases of stationary sources, in reality, simultaneously investigating the catalyst efficiency in the presence of water vapor, SO2, and NOX emissions are suggested. Finally, in order to increase the thermal stability of the catalyst at high temperatures, our studies will focus on usage of carbon-metal oxide hybrid supports in the future.
Nomenclature
BET |
: Brunauer–Emmett–Teller |
CNTs |
: Carbon nanotubes |
CO |
: Carbon monoxide |
COOH |
: Carboxyl |
CoOx |
: Cobalt oxide |
GHSV |
: Gas hourly space velocity |
H2-TPR |
: H2-temperature-programmed reduction |
JCPDS |
: Joint Committee on Powder Diffraction Standards |
OD |
: Outside diameter |
OH |
: Hydroxyl |
OMWNTs |
: Oxidized multi-walled carbon nanotubes |
PPROX |
: Preferential oxidation |
Redox |
: Reduction-oxidation |
TEM |
: Transmission electron microscopy |
XCO |
: Conversion of CO |
XRD |
: X-ray powder diffraction |
[E1]A b