Supported copper and cobalt oxides on activated carbon for simultaneous oxidation of toluene and cyclohexane in air

Abstract

Copper and cobalt oxides supported on almond shell derived activated carbon (AC) with different loadings were synthesized by sequential and co-deposition–precipitation methods leading to Cu(shell)/Co(core)/AC, Co(shell)/Cu(core)/AC and Cu–Co(mixed)/AC catalysts that were subsequently used for catalytic oxidation of gaseous mixtures of toluene and cyclohexane in air in a tubular flow reactor. The catalysts and the support were characterized by Boehm test, Brunauer–Emmett–Teller (BET) surface area, X-ray diffraction (XRD), field-emission scanning electron microscopy (FESEM), X-ray photoelectron spectroscopy (XPS), transmission electron microscopy (TEM), energy-dispersive X-ray spectroscopy (EDX), inductively coupled plasma (ICP), and thermogravimetric analysis (TGA). Catalyst efficiency for toluene and cyclohexane oxidation, both separately and in a mixture, was higher over the mixed metal oxides catalysts compared with the core–shell catalysts. An increase in the cobalt loading on the support led to a decrease in the metal oxide crystallite size and a change in the catalyst morphology. The best performance was obtained for the Cu₂–Co₆(mixed)/AC sample (Removal Efficiency >99.9%). Agglomeration of copper oxide over cobalt oxide crystallites for Cu(shell)/Co(core)/AC samples resulted in catalysts with the worst performance for complete oxidation of VOCs. Results indicated a negligible deterrence effect of toluene on cyclohexane oxidation. Furthermore, the addition of water (humid air) decreased the conversion of hydrocarbons.

---

1 Introduction

Catalytic oxidation of volatile organic compounds (VOCs) is an efficient and low operating cost technology for reduction of air pollution.¹ Emission of aromatic and aliphatic compounds to the atmosphere is a major environmental concern for many urban societies.² Typical industrial exhausts contain a mixture of VOCs and an important issue is the simultaneous oxidation of different components and the interactions between them during deep oxidation over appropriate supported catalysts.³ Indeed, it may be beneficial to determine the catalytic oxidation of each hydrocarbon both separately and in a mixture with other hydrocarbons. There are several studies reported in the open literature dealing with catalytic oxidation of different hydrocarbon mixtures. Grbic et al.⁴ investigated the complete oxidation of toluene and n-hexane in air on supported Pt over alumina catalysts and observed that the oxidation of n-hexane was inhibited by toluene in binary mixtures confirming the results presented by Gangwal et al.⁵ Gutiérrez-Ortiz et al.³ investigated the deep combustion of binary mixtures of n-hexane and chlorinated VOCs over ceria and zirconia mixed oxides observing that n-hexane inhibited the oxidation of chlorinated hydrocarbons. Blasin-Aubé et al.² studied the complete oxidation of hydrocarbon feeds containing only propane, propene, hexane, toluene, methylcyclohexane, and cyclohexane and mixtures of ethyl acetate–toluene, ethyl acetate–hexane and toluene–acetone and reported that La_(0.8)Sr_(0.2)MnO_(3+x) perovskite-type catalyst was efficient for complete oxidation of these VOCs. In addition, they concluded that toluene played an inhibitory role in the oxidation of other hydrocarbons in mixtures. A more recent investigation also indicated that the presence of toluene in a mixture inhibited the total combustion of ethanol, and in contrast, ethanol only had a mild effect on the complete oxidation of toluene.

There are extensive studies on the oxidation of different VOCs over supported noble^7–10 and transition metal^11–14 catalysts with transition metals being preferred due to their low price and availability. Complete catalytic oxidation of mixtures of VOCs can be enhanced using bimetallic nano-materials.¹⁵ Several studies have focused on using multi metallic catalysts for oxidation of VOCs in air due to their improved physical and chemical properties compared with catalysts with a single active component.¹⁶ Table 1 provides a summary of previous investigations on oxidation of VOCs using supported bimetallic catalysts. Most studies have used noble bimetallic catalysts supported on non-carbonaceous materials such as TiO₂, alumina, and silica. Activated carbon (AC) with a hydrophobic surface, high surface area, flexible surface, and low cost could be an alternative support for bimetallic catalysts for oxidation of VOCs.

Table 1 Bimetallic catalysts used for oxidation of VOCs

No.	Active site	Support	VOC	References
1	Mn–Cu	Zirconia	Benzene	17
2	Cu–Zn	Alumina	NO–CO	18
3	Cu–Co	Alumina	BTX (separately)	19
4	Pt–Rh	Alumina	Methyl cyclohexane	20
5	V–Ti, Nb–Ti	MCM-41	Styrene	21
6	Pt–Au	ZnO/Al2O3	Toluene	16 and 17
7	Au–Ag	TiO2	CO	22
8	Au–Pd	SnO2–Fe2O3	Dioxine	23
9	Co–Mn	Rh = MCM-41	Ethanol	24
10	Pd–Au	TiO2	Toluene–propene (mixture)	25
11	Au–Pd	CeO2	Toluene	26
12	Cu–Fe	SBA-15, SBA-16	Toluene	27
13	Ce–Al	Silica	Acetone	28
14	Pt–Pd	Alumina	Benzene	29
15	Pd–Ce	ZSM-5	MEK	30
16	Ag–Pt	TiO2	Toluene	31

---

Catalyst preparation method plays a significant role in the resulting structure of bimetallic nano-materials. Impregnation,^32,33 deposition–precipitation^17,23 and polyol process^18,19,34 are the general techniques used to synthesize nano-bimetallic catalysts. A recent investigation³⁵ showed that the deposition–precipitation method could result in the synthesis of well-dispersed nano-materials compared with other catalyst preparation methods. Depending on the catalyst preparation technique, an alloy or core–shell type structure was formed for the bimetallic catalysts.³⁶ The above studies used metals in the groups VIB to IIB to prepare multi component nano-catalysts. Among these metals, copper and cobalt bimetallic catalysts were prepared and characterized for oxidation of benzene, toluene, and xylene mixtures (BTX). The catalysts had a general composition of Co_yCu_xO_z depending on the preparation method and the type of support used.^19,37,38 In the present work, to extend and complete the previous studies, we synthesized new and modified nano-bimetallic catalysts for simultaneous oxidation of aliphatic and aromatic compounds. Core–shell and mixed copper–cobalt oxide catalysts on activated carbon support were prepared by the deposition–precipitation method. The activity of the prepared catalysts was determined for complete oxidation of a mixture of toluene and cyclohexane in air to water vapor and carbon dioxide and we performed a comprehensive characterization of the bimetallic catalyst samples. This study focused on investigating the effect of inlet concentration of VOCs and oxidation temperature on the performance of the prepared catalysts.

2 Experimental section

2.1 Catalyst preparation

Mesopore activated carbons derived from local Iranian agricultural solid wastes such as almond shells, walnut shells, and apricot stones were considered as catalyst supports in this study. The activated carbons were prepared by the physical activation method using water vapor as an activation agent at 800 °C. Supported copper and cobalt oxide bimetallic catalysts were prepared using a sequential and a simultaneous deposition–precipitation (DP) method to obtain a total of 8 wt% metal content with various ratios (Cu/Co = 1 : 1, 1 : 3 and 3 : 1). Nitrate solutions of copper and cobalt were prepared by addition of sufficient volume of distilled water to the metal salts. A sequential DP (sDP) was carried out for 8 hours with heating at 70–80 °C and metal ions (copper or cobalt) were precipitated on the sieved activated carbon (below mesh no. 40) by gradual addition of NaOH solution (6 wt%) in 30 minute intervals until reaching a basic solution of pH = 10–11. The prepared catalysts were washed to reach pH = 7 and dried at 110 °C for 6 hours. The dried samples were then calcined for 4 hours under N₂ flow at a calcination temperature of 500 °C. The entire process was then repeated for the other metal ion (copper or cobalt) to precipitate on the previously prepared copper or cobalt catalyst. The co-deposition–precipitation (cDP) method was performed by simultaneous addition of the nitrate solutions of copper and cobalt to the sieved activated carbon. The co-deposition–precipitation method was performed under the same conditions employed by the sequential method. Shell–core and mixed (alloy) structures²⁵ were obtained on the activated carbon by sDP and cDP methods, respectively. The catalysts were denoted by the preparation methods, Cu(x(wt%))–Co(y(wt%))/AC, Cu(x(wt%))/Co(y(wt%))/AC, or Co(y(wt%))/Cu(x(wt%))/AC, (x = 2, 4, 6 and y = 8 − x) for alloy and core–shell type, respectively. For example, the catalyst prepared by the cDP method with 4 wt% Cu and 4 wt% Co was named Cu4–Co4/AC. All the chemicals and reagents were purchased from Merck Company.

2.2 Catalyst characterization

Boehm test was used for characterization of the surface functional groups of the prepared activated carbons as support.³⁹ Nitrogen adsorption/desorption to determine the surface area and pore sizes were carried out at 77 K using a BEL (Japan) physical-adsorption apparatus. The specific surface area was calculated by Brunauer–Emmett–Teller (BET) and t-plot methods, and the average pores sizes were obtained by Barrett–Joyner–Halenda (BJH) method. The X-ray diffraction (XRD) patterns for the fresh calcined bimetallic catalysts were recorded using model D-64295 equipment from STOE Company. XRD analysis was conducted at 30 kV, 20 mA, copper Kα radiation, and scanning rate of 3° min⁻¹. Morphologies of the bimetallic catalysts were characterized using a field-emission scanning electron microscope (FESEM) at 15 kV and vacuum conditions by model S-4160 Hitachi equipment from Japan. Energy-dispersive X-ray spectroscopy (EDX) was used to determine the chemical composition of the prepared catalysts. Inductively coupled plasma (ICP) test was performed to detect the metals on the support at low concentrations. In this method, argon is ionized by a magnetic field with radio frequency of 27–40 MHz. The sample is sprayed into the argon plasma at high temperatures (10 000 K) and the atomic particles with specific emissions are identified for metal types by a PerkinElmer® Optima™ 8000 ICP-OES instrument. Transmission electron microscopy (TEM) images were used to characterize the morphology and particle size distribution of the bimetallic samples using a EM208-Philips instrument equipped with a Schottky field emission gun operated at 100 keV and magnification ranges of 180 000×. X-ray photoelectron spectroscopy (XPS) (Quantera SXM, Ulvac-Phi) was performed with non-monochromatic Al Kα (1486.6 eV) radiation. XPS peaks are related only to the electron emission from atoms near the surface due to the short length of the free path of electrons in the solid. The oxidation resistance of the support was evaluated by thermogravimetric analysis (TGA) using a TGA-25 apparatus (Mettler-Toledo, Switzerland) by heating the sample up to 1000 °C under an air flow of 55 cm³ min⁻¹ (STP) and a heating rate of 10 °C min⁻¹.

2.3 Catalyst activity

The schematic of the experimental set-up for multi component oxidation is shown in Fig. 1. A tubular fixed bed flow reactor was used to evaluate the catalytic activity of the prepared bimetallic catalysts at atmospheric pressure. The desired inlet concentrations of toluene and cyclohexane were obtained by introducing an air stream (purity 99.99%) at a constant flow-rate to isolated bubbling saturators filled with liquid toluene and cyclohexane. A digital thermocouple was located at the center of the packed reactor to record the oxidation temperature and the feed flow-rate was set by rotameters. 0.5 g of the supported catalyst was placed in the horizontal quartz tubular reactor that was 10 cm in length with an internal diameter of 12 mm. Supported cobalt and copper oxide catalysts were packed with glass beads into the middle section of the reactor in between glass wool to prevent flow channeling and pressure drop. The reactor was heated to the reaction temperature in the range of (150 to 350 °C). The reaction temperature was maintained at the desired temperature using a PID temperature controller. The initial concentration of VOCs in the feed (separately or binary composition of VOCs) was adjusted in the range of 1000 to 8000 ppmv. The reactor exit was analyzed using a Gas Chromatograph-Mass Spectroscopy (GC-Mass) (Agilent) system equipped with 5975C mass detector and a 30 m long HP-5MS stainless steel column at regular time intervals from the start of each run. No intermediate oxidation products, carbon monoxide, hydrogen, cylohexene or cyclohexanol were detected by GC-Mass, and the products only contained carbon dioxide, water, and unreacted cyclohexane and toluene. The relative humidity of the feed stream was measured by electrical thermometer and a humidity meter with ±1% accuracy and controlled automatically using a humidifier (ST-625). Catalytic oxidation experiments were also performed using feed streams containing VOCs in humid air to investigate the effect of water vapor on the performance of the catalysts. Several experiments were performed using different catalyst particle sizes and feed flow-rates to determine the operating conditions where both internal and external mass transfer resistances were negligible. All experiments were performed with particle size below Mesh number 40 using a feed flow-rate of 166 ml min⁻¹.

Fig. 1 Schematic diagram of the experimental set-up.

3 Results and discussion

3.1 Screening tests for the AC supports

To find the appropriate AC as the catalyst support, three local low cost agricultural wastes were selected as raw materials to prepare mesopore activated carbons. The properties of the collected Iranian almond shells, walnut shells, and apricot stones are presented in Table 2. Almond shells with a higher carbon content compared with apricot stones and walnut shells can be suitable for producing high surface area activated carbon. Table 3 provides the concentrations of phenolic, lactonic, carboxylic, and basic groups for each AC sample as measured by the Boehm test.³⁹ The significant amounts of acidic groups representative of oxygen groups on the surface makes the AC derived from Iranian almond shells as the most appropriate catalyst support among the prepared ACs. Iodine number (which is a good indicator of total surface area of activated carbons) determined by the DIN 53582 standard³⁹ method were 800, 430 and 750 mg g⁻¹ for almond shell, walnut shell, and apricot stone derived activated carbons, respectively. The results showed that almond shell derived activated carbon with significant amounts of oxygen groups, higher hydrophobic surface area according to Iodine number, and higher carbon content (54 wt%) would be the most suitable AC for supported metal oxide catalysts.

Table 2 Physical and chemical properties of agricultural solid wastes

Properties	Walnut shells	Apricot stones	Almond shells
pH	7.5	5.3	5.1
Moisture content (wt%)	1.97%	3.25%	4.73%
Color	Brown	Brown	Light brown
Carbon content (wt%)	40%	35.5%	54%
Hydrogen content (wt%)	4.74%	6.3%	6.26%

---

Table 3 The Boehm test results for prepared activated carbons derived from agricultural solid wastes (mequiv. g⁻¹)

Activated carbon	Basic groups	Lactonic groups	Carboxylic groups	Phenolic groups	Total acidic groups = total oxygen groups
Almond shells	0.00	0.63	0.10	0.97	1.70
Walnut shells	0.60	0.53	0.30	0.21	1.04
Apricot stones	0.80	0.30	0.20	0.41	0.91

---

One of the important properties of activated carbons as a catalyst support in the oxidation process is its resistance to oxidation. Fig. 2 illustrates the TGA results for the three activated carbon samples. There was no oxidation below 450 °C for the almond shell derived AC. The carbonization of almond shell under water vapor at 800 °C had increased the oxidation resistance. It was shown in a previous study⁷ that chemical agents such as phosphorous could modify the surface oxidation resistance of activated carbons. The TGA profiles indicated that the weight loss with increasing temperature was almost linear in the 470 to 950 °C temperature range. Some oxidation had occurred at 400 °C and 370 °C for apricot and walnut activated carbons, respectively. Almond activated carbon could therefore be used as a catalyst support for oxidation of VOCs below 400 °C.

Fig. 2 TGA for almond, apricot and walnut derived AC with 7 mg sample (heating the sample up to 1000 °C under an air flow of 55 cm³ min⁻¹ (STP)).

3.2 Characterization of bimetallic catalysts

The XRD patterns of the supported copper and cobalt oxide bimetallic catalysts on the activated carbon prepared by co-deposition–precipitation (cDP) and sequential deposition–precipitation (sDP) are shown in Fig. 3a–c. The patterns were compared with those for copper oxide, cobalt oxide, and copper cobalt oxide complex crystallites from reference patterns including: 00-002-1082, 00-021-0288, 01-074-1855, 00-001-1152, 00-001-1227, 00-002-0770, 01-076-0442, 00-001-1117, 00-002-1040, 00-003-0884 and 00-001-1142 by X'PertHighScore software version 1.0d by PANalytical BV Almelo, the Netherlands. Fig. 3a shows the XRD patterns for three bimetallic catalysts synthesized by the cDP method with three different Cu : Co ratios. There are two phases distinguished from the XRD profiles; the amorphous phase is characteristic of the carbonaceous support⁴⁰ and the peaks are observed due to metal oxide crystallites. In the XRD patterns, the metal oxide crystallites are not detected with increasing cobalt loading on the support against copper content, indicating the acceptable dispersion of copper and cobalt oxide over high surface area of mesopore almond shell derived activated carbon. Strong and narrow diffraction peaks at 2θ = 37.10°, 45.48°, 54.01°, and 66.07° are identified for Cu6–Co2/AC sample that confirms the formation of the large crystallites over the activated carbon. The peaks for Cu6–Co2/AC at (2θ = 42.38°), (2θ = 35.49°, 54.18°, and 66.07°) and (2θ = 31.23°, 37.10°, 45.48°, and 60.05°) represent CoO (00-001-1227, ICDD), CuO (00-001-1117, ICDD) and Co₂O₃·CuO (00-002-1082, ICDD), respectively. The broad peaks with low intensity in the XRD patterns of Cu4–Co4/AC at (2θ = 35.61°), (2θ = 36.52° and 42.52°) and (2θ = 38.99°) also correspond to CoO (00-001-1227, ICDD), CuO (00-001-1117, ICDD) and Co₂O₃·CuO (00-002-1082, ICDD), respectively. For Cu2–Co6/AC catalyst, with Co content increasing over 4 wt% on the activated carbon, CoCuO₂ (01-074-1855, ICDD) crystallite was formed, represented by a peak at 2θ = 36.94°, instead of Co₂O₃·CuO.

Fig. 3 XRD patterns for (a) Cux–Coy/AC , (b) Cux/Coy/AC and (c) Coy/Cux/AC samples.

Fig. 3b shows the diffraction patterns for the second group of bimetallic catalysts prepared by the sDP method with copper deposited on the calcined supported cobalt catalysts over activated carbon. The results indicated that well-formed bimetallic catalysts were generated with increasing cobalt and decreasing copper loading over the support. The metal oxides for this group of catalysts mainly consisted of CoO, CuO and CoCuO₂ while there were no detectable Co₂O₃ (00-002-0770, ICDD), Co₃O₄ (00-001-1152, ICDD), Cu₂O (00-001-1142, ICDD), Cu₂CoO₃ (00-021-0288, ICDD) and Co₂O₃·CuO (00-002-1082, ICDD) peaks in the XRD patterns. XRD patterns for Cu2/Co6/AC with high dispersion metal oxides indicated only two detectable peaks at 2θ = 36.42° and 42.32° due to the formation of CoO and CoCuO₂, respectively. Diffraction lines were detected for CoCuO₂ at 2θ = 36.42°, 73.93°, and 89.71° for Cu6/Co2/AC and for CoCuO₂ at 2θ = 37.10°, 42.38°, and 66.07° for Cu4/Co4/AC.

Fig. 3c illustrates the XRD patterns for the samples with cobalt added on the calcined copper supported on the activated carbon. Results similar to Cux/Coy/AC samples were obtained for the Coy/Cux/AC group of catalysts. It should be noted that peak intensities for Coy/Cux/AC were lower than Cux/Coy/AC and the crystallites of Co₂O₃·CuO were detected instead of CoCuO₂ at 2θ = 38.95° and 36.98° for Co2/Cu6/AC and Co4/Cu4/AC catalysts, respectively. With increasing cobalt loading, XRD patterns revealed the presence of CoCuO₂ at 2θ = 36.71° for the Co6/Cu2/AC sample. The diffraction lines in Fig. 3a–c indicated that the deposition–precipitation method had resulted in the formation of copper and cobalt oxides (CoO and CuO) over almond shell derived AC and no Cu₂O, Cu(OH)₂ and Co(OH)₂ crystallites were detected. In addition, with increasing cobalt loading to more than 4 wt%, CoCuO₂ crystallites were also formed for both cDP and sDP methods. Furthermore, small and wide peaks appeared due to high dispersion of metal oxide species on the porous support and the amorphous phase was dominant. The cDP method led to samples with lower intensity and broader peaks implying that the simultaneous deposition of copper and cobalt favors the dispersion of active sites.

The crystallite sizes were calculated by Scherrer eqn (1) assuming spherical shapes⁴¹

where D is the mean crystallite diameter (nm), K is the dimensionless shape factor (K = 0.9 for spheres), λ is the X-ray wavelength (X-ray tube: Cu, λ = 0.154 nm), β is the line broadening at half the maximum intensity, and θ is the Bragg angle. The minimum size of metal oxide crystallites was estimated for the nine catalysts prepared by cDP and sDP methods and are given in Fig. 4 indicating that there was a decrease in the crystallite size when the cobalt content increased from 2 to 6 wt% for both preparation methods. This decrease in active site (metal oxide crystallites) sizes with increasing cobalt content was more pronounced for the Cux/Coy/AC samples compared with Cux–Coy/AC and Coy/Cux/AC samples. Smaller crystallites for the cDP method compared with the sDP method were possibly due to a high interaction between copper and cobalt species during deposition–precipitation. There were no significant differences between the metal oxide size of Cux–Coy/AC and Coy/Cux/AC samples. This could be justified in that copper species can cover cobalt species due to a lower surface free energy for copper (1520 erg cm⁻²) than cobalt atoms (2420 erg cm⁻²).⁴² Fig. 4 shows that deposition–precipitation method can generally lead to well-dispersed nano-bimetallic catalysts over almond activated carbon.

Fig. 4 Particle sizes of copper–cobalt oxides bimetallic catalysts.

TEM images along with nano-metal oxide sizes of the binary transition metal catalysts collected in conventional bright field are shown in Fig. 5 for Cu2–Co6/AC, Cu6/Co2/AC, Cu2/Co6/AC, Co2/Cu6/AC and Co6/Cu2/AC catalysts prepared by cDP and sDP methods. The image for the bimetallic copper and cobalt catalyst synthesized by the cDP method (Cu2–Co6/AC) indicate the presence of homogeneous metal oxide crystallizes of uniform size below 10 nm with a narrow distribution that were not distinguished by XRD analysis. TEM images of catalysts prepared by sDP display a larger size of metal oxides compared with cDP catalysts. It is not possible to differentiate between copper and cobalt species in the TEM images due to their similar contrast.

Fig. 5 TEM images for copper–cobalt oxides bimetallic catalysts.

Large crystallites were formed by the coverage of copper over the cobalt species in Cu6/Co2/AC and Cu2/Co6/AC samples. This was also suggested by the XRD results. Fig. 5 also indicates that the clusters of Co₂/Cu₆/AC and Co₆/Cu₂/AC are smaller than clusters of the Co2/Cu6/AC and Co6/Cu2/AC samples in agreement with the metal oxide sizes calculated by Scherrer equation. The TEM images also confirm that deposition–precipitation can be an appropriate method for preparation of well-dispersed nano-catalysts supported on activated carbon. Based on statistical analyses, metal oxide crystallites size distributions were obtained by Clemex Vision software. The results are summarized in Table 4 in terms of the average diameter and the standard deviation for five catalysts.

Table 4 Mixed cobalt and copper crystallites sizes and standard deviation from TEM results

Catalysts	Average diameter (nm)	Standard deviation	Diameter range (nm)
Cu2–Co6/AC	8	1.9	6–10
Cu6/Co2/AC	49	3.7	28–70
Cu2/Co6/AC	25.5	3.8	11–40
Co2/Cu6/AC	21	3.6	12–30
Co6/Cu2/AC	19.5	3.5	9–30

---

Furthermore, ICP-OES (PerkinElmer® Optima™ 8000) was used to measure the metal content of the samples. The ICP results on the TEM images suggested a small difference between nominal metals content and ICP results indicating an acceptable adsorption of copper and cobalt over almond activated carbon during the preparation process. According to both XRD and TEM results, core–shell and alloy nano-catalysts were prepared by sDP and cDP methods, respectively.

FESEM micrographs in Fig. 6a show the morphology of the catalysts illustrating the porous texture of copper–cobalt oxide catalysts with high surface area and spherical morphology. The different morphologies of the catalysts were due to the different preparation methods (cDP and sDP). The energy dispersive X-ray tests were carried out to determine the composition of the bimetallic catalysts on the AC surface. The presence of copper, cobalt, oxygen, and carbon was detected at specified points of the catalyst as seen in Fig. 6b in agreement with ICP-OES and XRD results.

Fig. 6 (a) FESEM images for three series of copper–cobalt oxides bimetallic catalysts and (b) EDX results for Cu4–Co4/AC sample.

Adsorption isotherms of nitrogen on the almond, walnut and apricot-activated carbons are illustrated in Fig. 7 indicating a type IV isotherm for activated carbon representing a mesoporous support according to the IUPAC classification. The hydrophobic activated carbon with mesoporous texture, which is an important characteristic for adsorption of the metal ions, indicates that these ACs can be suitable supports for transition metals such as copper and cobalt.¹³ The high capacity of almond activated carbon for nitrogen adsorption demonstrates the larger surface area of this material compared with walnut and apricot based-activated carbon, in agreement with the results suggested by iodine numbers.

Fig. 7 Nitrogen adsorption/desorption for almond, apricot and walnut derived AC.

The surface area and total pore volume of the catalysts obtained by BET, BJH and t-plot analyses are summarized in Table 5. The BET surface area decreased with increasing metal content as the pores of the activated carbon were significantly blocked by the precipitating copper and cobalt ions thus reducing the surface area for adsorption of nitrogen over the support. As all the synthesized catalysts had the same metals content, they were expected to have about the same surface area. The preparation method (sDP versus cDP) and the order of metal deposition, however, were the important parameters affecting the catalyst surface area. The results indicated that Cux/Coy/AC catalysts had the highest surface area due to the aggregation of copper over cobalt in the deposition–precipitation step thereby limiting the interference during nitrogen adsorption on the catalysts. The measured surface area (according to BET and t-plot analyses) for catalysts prepared by both cDP and sDP methods decreased with increasing cobalt loading indicating the synthesis of well-dispersed catalysts in agreement with the results obtained from XRD, TEM, and FESEM characterizations. Furthermore, there is an inverse relation between surface area and dispersion of active sites over activated carbon. The calculated pore volumes form BJH tests also confirmed this inverse trend in the surface area of the prepared catalysts.

Table 5 BET, t-plot and BJH results for prepared bimetallic catalysts supported on the almond activated carbon

Sample	BET		BJH	t-plot
	Surface area (m^2 g^−1)	Total pore volume (cm^3 g^−1)	Volume pore (cm^3 g^−1)	Surface area (m^2 g^−1)
Almond AC	860.70	0.4697	0.1831	913
Walnut AC	547.12	0.3143	0.1288	580.24
Apricot AC	804.44	0.3906	0.1040	910
Cu6–Co2/AC	100	0.1904	0.1280	105.71
Cu4–Co4/AC	80	0.1870	0.1240	96.65
Cu2–Co6/AC	73.39	0.1824	0.1245	96
Co2/Cu6/AC	110	0.2015	0.1270	130.15
Co4/Cu4/AC	96	0.2012	0.1273	100
Co6/Cu2/AC	90	0.1967	0.1289	110
Cu6/Co2/AC	220	0.2539	0.1115	250.35
Cu4/Co4/AC	200.15	0.2514	0.1147	216.16
Cu2/Co6/AC	154.24	0.2218	0.1256	190.35

---

XPS measurements were also carried out to better identify the chemical state of copper and cobalt on the support and the surface properties of metal crystallites. The patterns of fresh calcined nano-bimetallic catalysts (6 wt% Co and 2 wt% Cu) prepared by three different methods are shown in Fig. 8 indicating that copper 2p spectra of selected bimetallic catalysts determined by the two main spin–orbit components Cu 2p_1/2 and Cu 2p_3/2 differed by about 19.6 eV in binding energy. The values of binding energies of Cu 2p_1/2 and Cu 2p_3/2 were measured to be 953.2 and 933.6 eV, respectively. The difference between the binding energy of copper 2p spectra represented a low copper oxidation with ionic Cu²⁺state, which is in good agreement with previous studies.⁴³ Therefore, Cu 2p_3/2 peak with binding energy of 932.7–934.6 eV can demonstrate the presence of CuO. The lower intensity of copper peaks in Cu2–Co6/AC and Co6/Cu2/AC samples is due to the random position of copper on the surface and lower concentration of copper over the support compared with the Cu2/Co6/AC sample. This can confirm the core–shell and alloy morphology of the catalysts. Calcination temperature of 500 °C was sufficient for reforming Cu(OH)₂ to CuO by deposition–precipitation method which was also reported by Huang and co-workers.⁴⁴

Fig. 8 XPS spectra for three orders of metal deposition (copper–cobalt oxides over AC).

Fig. 8 also shows the 2p spectra of cobalt characterized by two different peaks at binding energy of 796.4–796.8 eV (2p_1/2) and 780.1–781.2 eV (2p_3/2) due to the presence of Co(II) and Co(III) that was also reported in previous investigations.⁴³ A high intensity of cobalt in all samples is due to the higher cobalt loading over the activated carbon. The results confirm that CoO can be generated from Co(OH)₂ in the deposition–precipitation method and that Co(II) and Cu(II) ions represent a CoCuO₂ compound over the almond activated carbon. As was mentioned in the discussion on the XRD results, copper can gradually cover the cobalt surface due to its lower free surface energy resulting in core–shell morphology for the Cu2/Co6/AC sample.

3.3 Effect of oxidation temperature on the VOC conversion

Fig. 9a–c show the percent conversion (defined as 100 × (C₀ − C_e)/C₀ where C₀ and C_e are the inlet and exit concentrations, respectively) of toluene and cyclohexane, separately and in a binary mixture, during complete oxidation in air over the nano-bimetallic Cux–Coy/AC, Coy/Cux/AC, and Cux/Coy/AC catalysts. Experiments were carried out over 0.5 g of catalyst in a tubular fixed-reactor at different oxidation temperatures (150–350 °C) with inlet concentration of 2000 ppmv of toluene and cyclohexane in air, separately or in mixture. The steady state exit concentrations of toluene and/or cyclohexane were measured after 120 min from the start-of-run when no significant variations were observed in the outlet concentrations. As shown in Fig. 9a, conversions of toluene at different reaction temperatures were higher than cyclohexane conversions in experiments for both single component and mixtures with Cux–Coy/AC catalysts. This result can be justified by the bond strength of C–H in aromatic and aliphatic hydrocarbons. The relation between reactivity of hydrocarbons was investigated by Aube and co-workers² where the bond strength of cyclohexane and toluene were reported about 400 and 370 kJ mol⁻¹ at 177–197 °C of T_50%, respectively, suggesting a higher reactivity for toluene compared with cyclohexane in the decomposition process.

Fig. 9 Complete oxidation of toluene and cyclohexane over (a) Cu–Co/AC alloy, (b) Co(shell)/Cu(core)/AC and (c) Cu(shell)/Co(core)/AC samples at feed concentration of 2000 ppmv.

The maximum conversion of toluene and cyclohexane in single component experiments were 99.98% and 95.33%, respectively, at 350 °C over the Cu2–Co6/AC catalyst. The experimental data also indicated that the highest conversion of toluene and cyclohexane in the mixture were 95.17% and 90.1%, respectively. Similar results were reported in the study carried out by Aube and co-workers² who investigated the oxidation of cyclohexane in presence of toluene. Toluene is reported to play an inhibitor role in the catalytic oxidation of mixture of VOCs.^2,3,6 The results obtained from the present work demonstrated that toluene in mixture had no significant effect on the oxidation of cyclohexane where the difference between the conversion of cyclohexane separately and in mixture with toluene was about 5% over the Cu2–Co6/AC bimetallic catalyst.

As shown in Fig. 9b, maximum conversions for toluene were 96.34% and 89.31% separately and in mixture, respectively, while maximum conversions for cyclohexane were 93.11% and 85.74% separately and in mixture, respectively, with Co6/Cu2/AC as catalyst. There is, therefore, no significant difference in the performance of Co6/Cu2/AC and Cu2–Co6/AC for oxidation of toluene and cyclohexane as there was only minor differences between Co6/Cu2/AC and Cu2–Co6/AC in terms of metal oxide state, active site size, surface area, and dispersion. Cux/Coy/AC samples (Fig. 9c), however, show lower activity than the other series of catalysts where maximum conversions of toluene and cyclohexane were 86.67% and 77% separately, and 80% and 73% in mixture, respectively, over Cu2/Co6/AC catalyst. According to the properties of Cux/Coy/AC catalysts, the lower oxidation activity of these catalysts could be due to the larger size of metal oxide species formed as consequence of copper coating over cobalt resulting in lower dispersions.

Conversion data presented in Fig. 9a–c indicate that for the three series of catalysts, catalyst activity increased with increasing cobalt content. Furthermore, enhancement in conversion with increasing reaction temperature was more pronounced for cyclohexane compared with toluene for all three series of catalysts with both components reaching their maximum conversion at approximately 300 °C. In studies focusing on oxidation of toluene or cyclohexane, an almost linear increase in conversion with oxidation temperature has been reported.¹⁴

3.4 Effect of initial concentration on VOC conversion

To investigate the effect of inlet feed concentration on the conversion of toluene and cyclohexane, experiments were conducted using inlet feed concentrations of 1000, 2000, 4000, and 8000 ppmv at 250 °C while keeping all other operating parameters constant. Conversion data reported in Fig. 10a–c indicate that conversion of both toluene and cyclohexane decreased with increasing inlet feed concentration in agreement with results reported in other investigations¹⁴ as the oxidation products, namely water vapor and carbon dioxide, can cover the metal oxide sites on the support thereby inhibiting the oxidation of feed hydrocarbons. Fig. 10a indicates that the maximum conversion of toluene and cyclohexane were 99.9% and 95.1%, respectively, in single component experiments, and 94.11% and 90.14%, respectively, when the feed contained a mixture of both components for catalyst Cu2–Co6/AC and inlet feed concentration of 1000 ppmv for each component. Fig. 10a–c also indicate that there are no significant differences between the catalysts for oxidation of toluene and cyclohexane in mixture. Furthermore, compared with single component experiments, the presence of toluene and cyclohexane in mixture had no significant effects on the conversion of the other component. The reduction in conversion with increasing initial concentration was more pronounced for cyclohexane as compared with toluene. As shown in Fig. 10b, the conversions for toluene and cyclohexane were 97.1% and 89%, respectively, when they were present separately and 93.1% and 88.6%, respectively, when both were present in a mixture when Co6/Cu2/AC was used as the catalyst with initial concentration of 1000 ppmv for each component in the feed. These conversion levels were expected owing to the close chemical and physical properties of Cux–Coy/AC and Coy/Cux/AC catalysts. As shown in Fig. 10c, similar trends were also observed for the third group of catalysts that had lower activity for oxidation of toluene and cyclohexane.

Fig. 10 Complete oxidation of toluene and cyclohexane over (a) Cu–Co/AC alloy, (b) Co(shell)/Cu(core)/AC and (c) Cu(shell)/Co(core)/AC samples at various feed concentration and 250 °C.

For comparison, catalyst activity for oxidation of toluene in terms of μmoles of toluene per gram of metal per second,⁴⁵ calculated for the most effective catalyst in the present study and those reported by previous investigations are presented in Table 6 for oxidation at 250 °C. Blasin-Aubé et al.² reported that La_0.8Sr_0.2MnO_3+x perovskite had a very high activity for oxidation of cyclohexane and toluene, but the efficient oxidation temperature was over 450 °C. Nogueira et al.⁴⁶ also reported that complete decomposition of toluene over iron oxide supported on clay only occurred at a reaction temperature of 500 °C. In a previous investigation,¹⁹ copper and cobalt bimetallic catalysts supported on alumina were used for oxidation of a mixture of benzene, toluene, ethylbenzene and xylene at a very low concentration of about 12 mg m⁻³ and reaction temperature of 250 °C where toluene conversion was about 97.5% over Cu1.25–Co3.75/alumina prepared by the polyol method. Similar to the results presented in the present investigation, toluene removal efficiency increased with increasing cobalt content. In the present study, the removal efficiency of toluene (1000 ppmv) in a mixture with cyclohexane (1000 ppmv) was 94.11% over Cu2–Co6/AC indicating the adequate efficiency of copper and cobalt bimetallic catalysts supported on activated carbon at higher inlet concentrations. As can be seen from the data presented in Table 6, transition metals can compete with noble metals for oxidation of VOCs. Furthermore, the most active catalyst prepared in the present study (Cu2–Co6/AC) showed a superior activity in comparison with other supported transition metal catalysts at 250 °C and 2000 ppmv inlet concentration reported in the literature. At high feed concentration of toluene, the bimetallic catalysts prepared in this investigation had the best performance amongst other transition metal catalysts.

Table 6 Catalytic activity per gram (A) of metals of catalysts used for oxidation of toluene separately and in mixture of VOCs

Catalyst	Operating conditions: inlet concentration (ppmv), Oxidation temp. (°C), mixture of VOCs	A (separately) μmol g^−1 s^−1	A (mixture) μmol g^−1 s^−1	References
a Activation agent.
Pt/Al2O3	1000 ppmv–220 °C	24 817	—	47
Cu/AC(coconut)	1000 ppmv–250 °C	1687.5	—	48
Cu/AC(coconut)	150 ppmv–250 °C–(xylene, 150 ppmv)	—	1489	48
Cu/AC(coconut)	150 ppmv–250 °C–(benzene, 150 ppmv)	—	3673	48
Ni/ACfiber(Nippon-Kynol)	2000 ppmv–250 °C	1266	—	14
Co/ACfiber(Nippon-Kynol)	2000 ppmv–250 °C	1161	—	14
Cu/ACfiber(Nippon-Kynol)	2000 ppmv–250 °C	1161	—	14
Cr/ACfiber(Nippon-Kynol)	2000 ppmv–250 °C	714.72	—	14
Co/AC(rice husks) (CO2)^a	150 ppmv–250 °C–(NO, 600ppmv)		7505	13
Co/AC(rice husks) (ZnCl2)	150 ppmv–250 °C–(NO, 600ppmv)	—	2852	13
Co/AC(rice husks) (KOH)	150 ppmv–250 °C–(NO, 600ppmv)	—	3302	13
Co/AC(coconut) (ZnCl2)	150 ppmv–250 °C–(NO, 600ppmv)	—	4278	13
Co/AC(coconut) (KOH)	150 ppmv–250 °C–(NO, 600ppmv)	—	5178	13
Co/carbon nano tube	150 ppmv–250 °C–(NO, 600ppmv)	—	5404	13
Co/AC(rice husks) (CO2)	150 ppmv–250 °C–(NO, 600ppmv, CO–600 ppmv)	—	6980	13
Co/AC(rice husks) (ZnCl2)	150 ppmv–250 °C–(NO, 600ppmv–CO, 600 ppmv)	—	2477	13
Co/AC(rice husks) (KOH)	150 ppmv–250 °C–(NO, 600ppmv–CO, 600 ppmv)	—	3152	13
Co/AC(coconut) (ZnCl2)	150 ppmv–250 °C–(NO, 600 ppmv–CO, 600 ppmv)	—	3978	13
Co/AC(coconut) (KOH)	150 ppmv-250 °C–(NO, 600 ppmv–CO, 600 ppmv)	—	6830	13
Co/carbon nano tube	150 ppmv–250 °C–(NO, 600 ppmv–CO, 600 ppmv)	—	5253	13
Pt/AC(Aldrich)	2000 ppmv–160 °C	49 633	—	49
Co/Zr95Y5	1000 ppmv–250 °C	5435	—	12
Pt/Al2O3–CeO2	1000 ppmv–250 °C	10 423	—	50
LaCoO3/Ce0.9 Zr0.1 O2	1700 ppmv–250 °C	3238.6	—	51
Ag/HY	1700 ppmv–250 °C	3290.4	—	52
Pt–Au/ZnO/Al2O3	18000 ppmv–250 °C	6990.6	—	16
Pt/Al2O3	1500 ppmv–180 °C	6670.9	—	4
Pt/Al2O3	1500 ppmv–180 °C–(n-hexane, 1500 ppmv)	—	4336.1	4
Pt/TiO2	1061 ppmv–250 °C	74 450	—	6
Pd/TiO2	1061 ppmv–250 °C	74 450	—	6
Rh/TiO2	1061 ppmv–250 °C	11 168	—	6
Ir/TiO2	1061 ppmv–250 °C	5211.5	—	6
Au/TiO2	1061 ppmv–250 °C	2233.5	—	6
Pt/TiO2	1200 ppmv–250 °C–(ethanol, 1200 ppmv)	—	74 450	6
Pd/Ce0.8 Y0.2 O	1326 ppmv–250 °C	248 170	—	53
Pd/AC(lignin)	1000 ppmv–250 °C	3772.3	—	7
Au/CeO2	1000 ppmv–250 °C	30 925	—	54
Cu2–Co6/AC(almond)	1000 ppmv–250 °C–(cyclohexane, 1000 ppmv)	3089.7	2907.7	This work
Cu2–Co6/AC(almond)	2000 ppmv–250 °C–(cyclohexane, 2000 ppmv)	1780.7	1734.3	This work

---

Possible reaction routes for the deep oxidation of hydrocarbon mixtures in air depend on the type of catalyst support and active sites. Hydrocarbons, including cyclohexane and toluene, may strongly adsorb on the hydrophobic surface of activated carbon. There is a competition between toluene, cyclohexane and gaseous oxygen for adsorption on both copper oxide and cobalt oxide active sites, which could affect the reaction mechanism. Based on investigations reported in the literature, oxidation of hydrocarbon mixtures over catalysts may follow either a Langmuir–Hinshelwood¹⁵ or a Mars-van Kreleven^1,4,55 mechanism. Under the operating conditions employed in this investigation for complete oxidation of cyclohexane and toluene in air, oxygen was present in large excess. Although it seems that Mars-van Kreleven redox model is reasonable to represent the oxidation behavior of hydrocarbon mixtures, a combined mechanism that includes the equilibrium adsorption of toluene and cyclohexane as well as strong adsorption of oxygen may be more appropriate to describe the kinetic behavior. This is justified due to the presence of different active sites and a strong tendency for adsorption of reactants on both types of active sites. A simple mechanism might therefore include the following steps:

where, T, B, θ, Oθ, Tθ, Bθ, θ_T and θ_B represent toluene, cyclohexane, active sites, adsorbed oxygen, adsorbed toluene, adsorbed cyclohexane, active sites after the reaction of toluene and cyclohexane, respectively. K_ads,T, K_ads,B, k_T, k_B, k_OT and k_OB represent the equilibrium adsorption constants for toluene and cyclohexane, and reaction rate constants for oxidation of toluene and cyclohexane, and rate constants for re-oxidation of active sites after the oxidation of toluene and cyclohexane, respectively. A similar mechanism was also suggested by Grbic et al.⁴ for total oxidation of n-hexane and toluene over supported Pt over alumina.

3.5 Long term activity test

The durability of the catalyst was investigated over a 54 hours run using Cu2–Co6/AC catalyst at 250 °C using a feed containing a mixture of cyclohexane and toluene each with an inlet concentration of 1000 ppmv. A problem that could be associated with supported catalysts over activated carbon is the support burn off during reaction at high temperatures. Fig. 11 shows the high stability of Cu2–Co6/AC for the conversion of toluene and cyclohexane at the employed reaction temperature of 250 °C. TGA tests have indicated that burn off could only occur at temperatures above 450 °C.

Fig. 11 Long term performance of Cu2–Co6/AC for oxidation of toluene and cyclohexane.

3.6 Effect of water vapor on catalyst performance

To investigate the effect of water vapor on the performance of the prepared catalysts, a series of experiments were performed at 250 °C with Cu2–Co6/AC as catalyst using a feed containing a mixture of cyclohexane and toluene each with an inlet concentration of 1000 ppmv with relative humidity of 50%, 70% and 90%. The results are presented in Fig. 12 indicating the antonym effect of water vapor on the decomposition of VOCs due to the competition of water with VOC molecules for adsorption on metal oxide.⁵⁶ The conversion of both toluene and cyclohexane decreased slightly with increasing relative humidity. This decrease was more pronounced for cyclohexane. The results indicate the ability of the prepared bimetallic catalyst for total oxidation of toluene and cyclohexane in the presence of water vapor.

Fig. 12 Performance of Cu2–Co6/AC for oxidation of toluene and cyclohexane (mixture) in presence of water vapor.

4 Conclusions

Three series of copper and cobalt oxide bimetallic catalysts were synthesized using almond shell based activated carbon as support for total oxidation of toluene, cyclohexane, and their mixture in air at ambient pressure. The sequence of deposition–precipitation and the ratio of copper and cobalt loading over the activated carbon had a significant effect on the size of the metal oxide crystallites, the dispersion of active sites, the metal oxide composition, the surface area, and the structure of the prepared catalyst samples. CoO and CuO were formed while CoCuO₂ was only formed at high cobalt loading for the three series of bimetallic samples as indicated by XRD and XPS results. TEM and FESEM images revealed that the employed deposition–precipitation method for catalyst synthesis led to nano spherical structure for the bimetallic catalysts on the activated carbon support. ICP-OES analysis indicated that efficient adsorption of metal ions had occurred during the catalyst synthesis. Copper oxide crystallites covered cobalt oxide crystallites due to a lower surface energy resulting in Cux/Coy/AC catalysts with larger crystallites than Cux–Coy/AC and Coy/Cux/AC catalysts. Reactor tests indicated that all catalysts were active towards complete oxidation of toluene, cyclohexane, and their mixture. Furthermore, it was observed that toluene had negligible effect on the conversion of cyclohexane when both were present in the feed. The catalysts prepared by the co-deposition–precipitation method had slightly higher activity than those prepared by the sequential methods. The conversion of both toluene and cyclohexane increased with increasing cobalt loading. The presence of water vapor showed an antonym, albeit very small, effect on the conversion of both components in the mixture. The performance of the supported copper and cobalt oxides on activated carbon towards complete oxidation of toluene and cyclohexane were comparable to other transition metal catalysts reported in the literature.

Acknowledgements

The authors gratefully acknowledge the financial assistance of Iran National Science Foundation (INSF).

References

1. S. B. Lee, S. L. Kang, J. D. Lee and H. K. Hong, J. Ind. Eng. Chem., 1999, 5, 170–176.
2. V. Blasin-Aubé, J. Belkouch and L. Monceaux, Appl. Catal., B, 2003, 43, 175–186.
3. J. I. Gutiérrez-Ortiz, B. de Rivas, R. López-Fonseca and J. R. González-Velasco, Appl. Catal., B, 2006, 65, 191–200.
4. B. Grbic, N. Radic and A. Terlecki-Baricevic, Appl. Catal., B, 2004, 50, 161–166.
5. S. Gangwal, K. Ramanathan, P. Caffrey, M. Mullins and J. Spivey, Report no. EPA 600/7-88-017, US Environmental Protection Agency, Research Triangle Park, NC, 1988.
6. V. P. Santos, S. A. C. Carabineiro, P. B. Tavares, M. F. R. Pereira, J. J. M. Orfao and J. L. Figueiredo, Appl. Catal., B, 2010, 99, 198–205.
7. J. Bedia, J. M. Rosas, J. Rodríguez-Miraso and T. Cordero, Appl. Catal., B, 2010, 94, 8–18.
8. H. Shaoyong, Z. Changbin and H. Hong, Environ. Sci., 2009, 21, 985–990.
9. C. A. Gluhoi and B. E. Nieuwenhuys, Catal. Today, 2007, 119, 305–310.
10. T. Masui, H. Imadzu, N. Matsuyama and N. Imanaka, J. Hazard. Mater., 2010, 176, 1106–1109.
11. C. Y. Lu and M. Y. Wey, Fuel Process. Technol., 2007, 88, 557–567.
12. F. Wyrwalski, J. F. Lamonier, S. Siffert, L. Gengembre and A. Aboukaïs, Catal. Today, 2007, 119, 332–337.
13. C. Y. Lu, M. Y. Wey and K. H. Chuang, Appl. Catal., B, 2009, 90, 652–661.
14. V. Gaur, A. Sharma and N. Verma, Carbon, 2005, 43, 3041–3053.
15. K. J. Kim, S. I. Boo and H. G. Ahn, J. Ind. Eng. Chem., 2009, 15, 92–97.
16. K. J. Kim and H. G. Ahn, Appl. Catal., B, 2009, 91, 308–318.
17. C. Hongyan, L. Xiaoshuang, C. Yaoqiang, G. Maochu and W. Jianli, J. Rare Earths, 2012, 30, 871.
18. H. H. Tsenga, H. Y. Lina, Y. F. Kuoa and Y. T. Su, Chem. Eng. J., 2010, 160, 13–19.
19. C. Y. Lu, H. H. Tseng, M. Y. Weyc, L. Y. Liu, J. H. Kuo and K. H. Chuang, Fuel, 2009, 88, 340–347.
20. P. Samoila, M. Boutzeloit, I. Salem, D. Uzio, G. Mabilon, F. Epron, P. Marecot and C. Especel, Appl. Catal., A, 2012, 416, 80–88.
21. V. Parvulescu, C. Anastasescu, C. Constantin and B. L. Su, Catal. Today, 2003, 78, 477–485.
22. A. Sandoval, A. Aguilar, C. Louis, A. Traverse and R. Zanella, J. Catal., 2011, 281, 40–49.
23. M. Okumura, T. Akita, M. Haruta, X. Wang, O. Kajikawa and O. Okada, Appl. Catal., B, 2003, 41, 43–52.
24. K. Rintramee, K. Föttinger, G. Rupprechter and J. Wittayakun, Appl. Catal., B, 2012, 115–116, 225–235.
25. M. Hosseini, T. Barakat, R. Cousin, A. Aboukaïs, B. L. Su, G. De Weireld and S. Siffert, Appl. Catal., B, 2012, 111–112, 218–224.
26. D. S. Lee and Y. W. Chen, J. Taiwan Inst. Chem. Eng., 2013, 44, 40–44.
27. Á. Szegedi, M. Popova, K. Lázár, S. Klébert and E. Drotár, Microporous Mesoporous Mater., 2013, 177, 97–104.
28. L. Y. Lin and H. Bai, Appl. Catal., B, 2014, 148–149, 366–376.
29. L. Jiang, N. Yang, J. Zhu and C. Song, Catal. Today, 2013, 216, 71–75.
30. L. Yue, C. He, X. Zhang, P. Li, Z. Wang, H. Wang and Z. Hao, J. Hazard. Mater., 2013, 244–245, 613–620.
31. A. Zielińska-Jurek and A. Zaleska, Catal. Today, 2014, 230, 104–111.
32. W. W. Lonergan, D. G. Vlachos and J. G. Chen, J. Catal., 2010, 271, 239–250.
33. W. W. Lonergan, X. Xing and R. Zheng, Catal. Today, 2011, 160, 61–69.
34. H. Zhang and N. Toshima, Appl. Catal., A, 2011, 400, 9–13.
35. V. R. Choudhary and D. K. Dumbre, Top. Catal., 2009, 52, 1677–1687.
36. X. Liu, D. Wang and Y. Li, Nano Today, 2012, 7, 448–466.
37. M. De Koninck, S. C. Poirier and B. Marsan, J. Electrochem. Soc., 2006, 153, 2103–2110.
38. G. G. Volkova, T. M. Yurieva, L. M. Plyasova, M. I. Naumova and V. I. Zaikovskii, J. Mol. Catal. A: Chem., 2000, 158, 389–393.
39. M. Zabihi, A. Ahmadpour and A. HaghighiAsl, J. Hazard. Mater., 2009, 167, 230–236.
40. F. Adam and R. Thankappan, Chem. Eng. J., 2010, 160, 249–258.
41. R. J. Chimentão, G. P. Valenc, F. Medina and J. Pérez-Ramírez, Appl. Surf. Sci., 2007, 253, 5888–5893.
42. E. Machlin, An Introduction to Aspects of Thermodynamics and Kinetics Relevant to Materials Science, Elsevier Science and Technology Books, 3rd edn, 2007.
43. A. Amri, X. F. Duan, C. Y. Yin, Z. T. Jiang, M. M. Rahman and T. Pryor, Appl. Surf. Sci., 2013, 275, 127–135.
44. Z. Huang, F. Cui, J. Xue, J. Zuo, J. Chen and X. Changu, Catal. Today, 2012, 183, 45–51.
45. M. C. Alarcón, A. Maroto-Valiente, I. Rodríguez-Ramos and A. Guerrero-Ruiz, Appl. Catal., A, 2004, 275, 257–269.
46. F. G. E. Nogueira, J. H. Lopes, A. C. Silva, R. M. Lago, J. D. Fabris and L. C. A. Oliveira, Appl. Clay Sci., 2011, 51, 385–389.
47. W. G. Shim, J. W. Lee and S. C. Kim, Appl. Catal., B, 2008, 84, 133–141.
48. C. Y. Lu, M. Y. Wey and L. I. Chen, Appl. Catal., A, 2007, 325, 163–174.
49. J. C. S. Wu, Z. A. Lin, F. M. Tsai and J. W. Pan, Catal. Today, 2000, 63, 419–426.
50. Z. Abbasi, M. Haghighi, E. Fatehifar and S. Saedy, J. Hazard. Mater., 2011, 186, 1445–1454.
51. M. Alifanti, M. Florea and V. I. Pârvulescu, Appl. Catal., B, 2007, 70, 400–405.
52. S. W. Baek, J. R. Kim and S. K. Ihm, Catal. Today, 2004, 93–95, 575–581.
53. J. Lingyun, H. Mai, L. Jiqing, L. Mengfei, G. Libiao and H. Jun, J. Rare Earths, 2008, 26, 614.
54. M. Ousmanea, L. F. Liotta, G. DiCarloe, G. Pantaleod, A. M. Veneziad, G. Deganelloc, L. Retailleaua, A. Boreavea and A. Giroir-Fendler, Appl. Catal., B, 2011, 101, 629–637.
55. L. van de Beld, M. C. van der Ven and K. R. Westerterp, Chem. Eng. Process., 1995, 34, 469–478.
56. H. Y. Pan, M. Y. Xu, Z. Li, S. S. Huang and C. He, Chemosphere, 2009, 76, 721–726.

---