SR

---

Article pubs.acs.org/IECR

Degradation of Phenol by Vis/Co-TiO2/KHSO5 Hybrid Co/SR− Photoprocess at Neutral pH Qingkong Chen,1 Fangying Ji,1,* Qian Guo,2 Wei Guan,1 Peng Yan,1 Ling Pei,1 and Xuan Xu1 1

Key Laboratory of Three Gorges Reservoir Region’s Eco-Environment, Ministry of Education, Chongqing University, Chongqing 400045, People’s Republic of China 2 T. Y. Lin International Engineering Consulting (China) Co., Ltd., Chongqing 401121, People’s Republic of China ABSTRACT: Co−TiO2 catalysts have been synthesized by the sol−gel method, and they were characterized by X-ray fluorescence, Brunauer−Emmett−Teller, scanning electron microscopy, high-resolution transmission electron microscopy, X-ray photoelectron spectroscopy, and ultraviolet diffuse reflectance spectroscopy technologies. To evaluate the activities of the catalysts, a series of tests have been conducted to decontaminate the phenol simulated wastewater at neutral pH. Because Co− TiO2 represented ability to decompose KHSO5 to generate sulfate radicals (Co/SR) and visible light photocatalytic activities (photo), we defined them as a type of “bifunctional catalyst”. Moreover, the obvious combined effect of Co/SR and photo was also observed. Results indicated that the cobalt species takes a function to integrate sulfate radical based advanced oxidation reaction and the TiO2 photocatalytic process in one heterogeneous system. Under optimum conditions of DCo2+ = 1.0%, TC = 773 K, MKHSO5/Mphenol = 10:1, and Ccata = 1.0 g/L, about 76.2% total organic content removal and 100% degradation of 50.0 mg/ L phenol could be achieved after 2 h of hybrid Co/SR−photo treatment. Finally, the possible catalytic mechanism of the synergistic system has been proposed.

1. INTRODUCTION Industrial wastewaters usually contain refractory organic compounds, and they are generally resistant to conventional physical, chemical, and biological wastewater treatment. Phenol and phenolic compounds are the most commonly found refractory organic pollutants in effluents from petrochemical, coal processing, and pharmaceutical industries.1 The advanced oxidation technologies (AOTs) have the greatest potential to eliminate these pollutants due to their ability to generate highly oxidizing transient species such as hydroxyl radical (•OH) in situ through different possible ways. But the specific treatment requirements were needed depending on the targeted effluent quality,2 and that may imply high operation cost because of the high doses of reagents required for mineralization. For example, Fenton and TiO2 photocatalysis technologies are recognized as more environmental friendly among various AOTs. The advantages of the Fenton reaction (FeII/H2O2) include the simple handling of this process, no external energy needed, fast reaction of all AOTs, and good performance under mild conditions. The TiO2 can be excited by endless solar light, and it is nontoxic, highly active, photostable, chemically stable, biologically inert, and low in cost.3 However, these technologies also face some disadvantages that hinder practical application for organic wastewater. The application of conventional Fenton was confined to the narrow operating pH range 3−4, and the Fenton reagent is invalidated by the generation of ferric hydroxide (Fe(OH)3) quickly. The major drawback of TiO2 photocatalysis was its high efficiency occurred just under ultraviolet irradiation (λ ≤ 385 nm), using only 3−4% of the solar energy that reaches the earth. Moreover, electron−hole (ecb−/hvb+) recombination in the absence of proper electron acceptors in TiO2 photocatalysis is extremely inefficient; thus, the oxidation reaction rate was slower, and most TiO2 © 2013 American Chemical Society

photocatalytic oxidation to degrade industrial wastewater was taken in the laboratory and only at a pilot scale.4 Therefore, to enhance their practical applications of industrial interest, some efforts have been made to develop simultaneous hybrid reactions to improve the removal of refractory compounds, reducing the final operation cost. The electro- and ultrasonic-Fenton or TiO2 photocatalysis with electron acceptors (e.g., UV/TiO2/H2O2,5 UV/TiO2/O3 6) and even Fenton−photo hybrid processes (e.g., UV/TiO2/Fe2+/ H2O2 7 and UV/Fe2O3−TiO2/H2O2 8) have attracted many researchers’ eyeballs. When Fenton is combined with TiO2 photocatalytic reaction, the efficacy of both systems for the production of radical are enhanced.9 The conduction band electrons (ecb−) and photoenergy (UV) accelerate the cycle of FeII/FeIII; at the same time, FeIII and H2O2 inhibit the electron−hole (ecb−/hvb+) recombination by scavenging electrons (eqs 1−7). Furthermore, electrons from the conduct band of TiO2 decomposes more H2O2 to •OH, which promotes efficiency of the Fenton reaction drastically. But, lower pH and UV light are still necessary for these hybrid processes; because the FeII/H2O2 Fenton only showed high catalytic activity at acidic condition, FeIII just can be decomposed to FeII by UV radiation effectively,10 and TiO2 cannot be activated efficaciously under visible light. TiO2 + UV → hvb+ + ecb−

(1)

Fe II + H 2O2 → Fe III + •OH + OH−

(2)

Received: Revised: Accepted: Published: 12540

April 19, 2013 July 18, 2013 August 5, 2013 August 5, 2013 dx.doi.org/10.1021/ie4012547 | Ind. Eng. Chem. Res. 2013, 52, 12540−12549

Industrial & Engineering Chemistry Research

Article

Fe III + H 2O2 → Fe II + HO2• +H+

(3)

Fe III + ecb− → Fe II

(4)

Fe II + hvb+ → Fe III

(5)

Fe III + UV → Fe II

(6)

H 2O2 + ecb− → •OH + OH−

2. EXPERIMETAL SECTION 2.1. Chemicals. The cobalt nitrate (Co(NO3)2·6H2O) and phenol (C6H6O) were purchased from Chendu Kelong Chemical Reagent Co., Ltd. (Sichuan, China). Tetra-n-butyl titanate (C16H36O4Ti) was obtained from Shanghai Qiangshun Co., Ltd. (Shanghai, China). Potassium peroxymonosulfate (KHSO5) was available from Shaoguan Sunuo Chemical Reagent Co., Ltd. (Guangdong, China). All chemicals used in this study were analytical grade reagents and were used as received without further purification. Deionized water was used in the present work. 2.2. Synthesis of Co−TiO2 Powders. Co−TiO2 catalysts were synthesized by sol−gel method. In a typical process, a stoichiometric amount of Co(NO3)2 (dissolved in deionized water fist) and tetrabutyl titanate (molar percentage of Co = 0.5, 1.0, and 1.5%) was separately dissolved in ethanol to form two solutions (marked as solution A and solution B, respectively). Then solution A was added dropwise to solution B with continuous stirring. After gelation, the solids were aged for 48 h. Furthermore, the gels were dried in an oven at a temperature of 383 K to remove water. Finally, in order to control their crystal phase and enhance their mechanical stability, the catalysts were calcined in a furnace at different temperatures (673, 773, and 873 K) for 6 h. The resulting Co− TiO2 catalyst was ground thoroughly and labeled as XCo− YTiO2, where X stands for the molar ratio of Co to Ti and Y stands for the calcinations temperature. Co−TiO2/KHSO5 sulfate radical based advanced oxidation process, Co−TiO2 visible light photocatalysis, and Co−TiO2 based hybrid reaction are denoted as Co−TiO2/KHSO5 (Co/SR), vis/Co−TiO2 (photo), and vis/Co−TiO2/KHSO5 (Co/SR−photo), respectively. 2.3. Co/SR, Photo, and Co/SR−Photo Hybrid Processes. The catalytic activities of Co−TiO2/KHSO5, vis/Co− TiO2, and vis/Co−TiO2/KHSO5 processes were evaluated in terms of phenol degradation. All of the experiments were performed in glass cylindrical reactors containing 1.0 L of 50.0 mg/L phenol (0.53 mM) solution with adjusted pH of 7.0 using 0.5 M phosphate buffer17,18 that were placed on a magnetic stirrer plate. But, in the test of the effect of initial phenol concentration on the activity of catalyst, the concentration of the phenol was set as 50 and 100 mg/L. The aim of doubling the concentration of phenol was to further confirm the “combined effect”. If the degradation efficiency of hybrid reaction was still higher than the sum of the individual process, the combined effect exists. 2.3.1. Co−TiO2/KHSO5. After the addition of the amount of catalyst (0.5, 1.0, or 1.5 g/L), the solution was allowed to reach adsorption equilibrium between the catalyst and phenol for 45 min (there was no considerable phenol adsorption on the surface of the samples; the maximum value < 5.0%). Then, KHSO5 was added into the solution at KHSO5 to phenol molar ratio of 5:1, 10:1, and 20:1. For measuring the concentration of phenol during 2 h of reaction, 5.0 mL samples were withdrawn at specified time intervals and quenched with 5.0 mL of methanol (in total organic content (TOC) tests, using 0.1 M NaNO2 solution instead) to prevent further reaction. The samples were centrifugated at 4500 rpm for 15 min and analyzed using a standard 4-aminoantipyrene photometric method (HJ 503-2009)19 by UV−vis spectrophotometer (Hitachi UV3010, Tokyo, Japan).

(7) II

II

Recently, in comparison with Fe /H2O2, Co /KHSO5 has become a promising alternative. Its proposed mechanism is similar to that of Fenton (eqs 8 and 9), and now it is named as a type of sulfate radical based advanced oxidation process (SRAOP). CoII + HSO5− → CoIII + SO4•− + OH−

(8)

CoIII + HSO5− → CoII + SO5•− + H+

(9)

II

The advantages of Co /KHSO5 include the following: (1) CoII is more stable than FeII in a wide pH range, even in alkaline condition, and it is hardly oxidized by dissolved oxygen to CoIII in aqueous solution;11 (2) sulfate radical (SO4•−), generated by decomposition of KHSO5,12 exhibited much higher degradation efficiency than •OH in a wider pH range from 2 to 9.13 Besides the utilization in sulfate radical based advanced process, cobalt (Co) also has been used to modify some photocatalysts such as TiO2. Doping Co2+ into TiO2 lattice has been reported as an effective way to shift the photoresponse of TiO2 into the visible light region14 or enhance its photocatalytic activity15 because Co2+ ions in lattice form impurity energy levels in the bandgap of TiO2 and promote separation of photoinduced electrons and holes (ecb−/hvb+). Consequently, we could make an assumption that it is possible for cobalt and TiO2 to form a kind of bifunctional catalyst that can couple sulfate radical based advanced oxidation process with TiO2-photocatalysis in one heterogeneous catalytic reaction system effectively. In this paper, Co−TiO2 powders were synthesized via sol−gel method. Phenol was chosen as the model organic pollutant due to it being a typical recalcitrant contaminant without sensitizing as a dye.16 The main novelties of our work include the following: (1) The Co−TiO2 catalyst used in this study possesses the ability to decompose KHSO5 to generate sulfate radicals; meanwhile, it also represents photocatalytic activity under visible light. So, the sulfate radical advanced oxidation reaction and visible light photocatalytic process can be integrated in one heterogeneous advanced oxidation system. (2) The Co−TiO2 based photocatalysis and sulfate radical advanced oxidation reaction can take place under visible light at neutral pH condition. So, to some extent, it overcomes the drawbacks of traditional TiO2photocatalysis and iron based Fenton. (3) When the hybrid reaction was composed by visible light, KHSO5, and Co−TiO2, the combined effect of sulfate radical based advanced oxidation process and photocatalytic reaction causes drastic kinetic enhancement in phenol oxidation and TOC removal. At last, considering the practical application in organic wastewater disposal, this bifunctional catalyst can be used in both light and dark environments. Furthermore, it is more cost effective than traditional Fenton and TiO2 catalysts since it can adapt to ambient condition and utilize visible light more effectively. 12541

dx.doi.org/10.1021/ie4012547 | Ind. Eng. Chem. Res. 2013, 52, 12540−12549

Industrial & Engineering Chemistry Research

Article

Figure 1. (a) SEM image and (b) HRTEM image of 1.0Co−773TiO2.

phase contents of samples were measured by “relative intensity ratio (RIR) method”.21 The morphology of the catalysts was measured by a high-resolution transmission electron microscope (HRTEM, JEM-2100, JEOL). Scanning electron microscopy (SEM) images were collected on an environmental field emission scanning electron microscope (FEI, Hillsboro, OR, USA). An ASAP-2010 (Micromeritics, Norcross, GA, USA) surface area analyzer was used to determine the Brunauer−Emmett−Teller (BET) surface areas of the catalysts. The X-ray photoelectron spectroscope (Shimadzu Amicus) with Mg Kα X-rays was used to determine the chemical states of surface cobalt, titanium, and oxygen in the catalyst. Atomic concentrations were obtained from the X-ray photoelectron spectra based on peak areas and sensitivity factors. UV−vis diffuse reflectance spectroscopy (UV-DRS) were obtained under atmospheric conditions using a UV−vis spectrophotometer (UV3010, Hitachi) mounted with an integrating sphere attachment for diffuse reflectance measurement. The band gap energy (Eg) of samples is calculated using the Kubelka−Munk function22

2.3.2. Vis/Co−TiO2. The photocatalytic experiments were conducted under atmospheric conditions using a 1000 W Xe lamp as the light source. All of the UV lights with wavelength shorter than 420 nm were removed by a cutoff filter (JB420). The temperature of the phenol solution in the reactors was kept at about 298 K by cooling water outside the reactors. In order to ensure adsorption equilibrium, the solution was stirred for about 45 min in the dark, prior to irradiation. For measuring the concentration of phenol during 2 h of reaction, 5.0 mL samples were withdrawn at specified time intervals. Then, the samples were centrifugated at 4500 rpm for 15 min and analyzed using a 4-aminoantipyrene photometric method (HJ 503-2009) by UV−vis spectrophotometer. 2.3.3. Vis/Co−TiO2/KHSO5. The experimental condition and test method are the same as Co−TiO2/KHSO5 process, and 1000 W Xe lamp with light filter (λ ≥ 420 nm) as the light source. TOC present in the samples was determined using a multi N/C 2100 analyzer (Analytik Jena AG, Jena, Germany). An atomic absorption spectrometer (Hitachi 180-80) was used to analyze the cobalt leaching from the catalyst. All measurements were repeated two times, and the results were reproducible within the experiments errors (±3%). 2.3.4. Reusability and Stability of Co−TiO2 in Multiple Runs. Five recycling runs of the Co−TiO2 were conducted, and the catalyst was recycled under the same reaction conditions to evaluate the reusability of the Co−TiO2 catalyst. After every run, the suspensions were centrifugated and the catalyst was collected, washed thoroughly, and dried in an oven at 353 K before the next round. Leaching Co2+ concentrations were tested by atomic absorption spectroscopy (AAS) after every run by using supernatant of centrifugated catalysts. The comparative tests of the catalysts’ stability include the following: (1) comparing the variations of leaching Co2+ ion concentrations in the vis/Co−TiO2/KHSO5 process at neutral pH with time; (2) monitoring the leaching Co2+ ion concentrations in five recycling runs. 2.4. Characterization and Analysis. The real cobalt mass fractions were tested by X-ray fluorescence technology (PPM100%, Baird). Crystallographic structure, crystalline degree, and chemical composition of catalysts were investigated with an Xray diffractometer (XRD-6000, Shimadzu, Kyoto, Japan). The crystallite size was calculated by “Scherrer equation”,20 and the

(αhν)1/2 = A(hν − Eg )

(10)

Eg is the optical band gap of catalyst, and A is a constant.

3. RESULTS AND DISCUSSION 3.1. Catalyst Characterization. Figure 1a shows a typical SEM image of Co−TiO2. The powder is shaped like irregular blocks, and some of them are agglomerated. The sizes of the powders are around 20 nm. The results of BET surface areas (SBET) are summarized in Table 1. In comparison with pure TiO2 (10.5 m2/g), doping Co2+ ion expands the surface area of TiO2; the higher the Co2+ doping concentration was, the larger the surface area obtained (TC = 773 K). But, the SBET becomes smaller at higher calcination temperature (873 K) with a corresponding increase in crystallite size. From XRD patterns (Figure 2), it is possible to identify the mixtures of anatase and rutile phases in both pure TiO2 and Co−TiO2 samples, and no peaks associated with cobalt oxide phases were detected. Thus, this may either indicate that the concentration of cobalt oxide formed in the samples is too low to be measured or the Co2+ ions have been doped into the TiO2 lattice, since the ionic 12542

dx.doi.org/10.1021/ie4012547 | Ind. Eng. Chem. Res. 2013, 52, 12540−12549

Industrial & Engineering Chemistry Research

Article

region and the absorption fringe is between 760 and 800 nm. The absorption bands at 200−400 nm are due to the charge transfer from the valence band to the conduction band of TiO2. The absorption band in the visible light region of 520−700 nm is accredited to the 2E(G), 4T1(P), 2A1(G)→ground state 4 A2(F) transitions for high-spin Co2+ (3d7) in tetrahedral coordination.24,25 The absorptions at 420 and 700 nm over the Co−TiO2 are assigned to the 1A1g→1T2g and 1A1g→1T1g transitions of Co3+ ions in the octahedral symmetry.26 Therefore, the CoII and CoIII may both exist in the lattice of Co−TiO2. The value of Eg is 1.56 eV for 1.0Co−773TiO2 and 2.9 eV for TiO2. On the basis of the above results, it is predicted that, in contrast to the TiO2, the doped samples may possess photocatalytic activity in the visible light region. Moreover, the difference in absorbance of TiO2 and Co_TiO2 in the ultraregion is not significant, so Co−TiO2 and pure TiO2 should have similar capability to be excited by UV light. So, it can be deduced that the potocatalytic activity of Co− TiO2 should be higher than pure TiO2. XPS tests were used to investigate the oxidation states of cobalt on the surface of Co− TiO2 catalysts. Figure 4 summarizes the XPS spectra of the 1.0Co−773TiO2 sample. There are two major peaks with binding energies at 780.2 and 796.2 eV in Co2p XPS spectra, corresponding to Co2p3/2 and Co2p1/2, respectively, which is characteristic of a Co3O4 phase,27−29 and this result coincides with the HRTEM image (Figure 1b). The two strong peaks centered at 458.1 and 463.8 eV symmetrically of the Ti spectrum are in agreement with the binding energies of Ti2p3/2 and Ti2p1/2, respectively.15,30 This confirms that Ti is present as Ti4+. The broad and asymmetric O1s peak is coherently fitted by two nearly Gaussian components, centered at 529.7 and 531.7 eV.23,27 The first peak on the low binding energy (BE) side of the O1s spectrum (529.7 eV) was due to the O2− ions in the lattice oxygen species from TiO2 and Co3O4; a shoulder at higher binding energy of 531.7 eV is identified as surface hydroxyl groups (i.e., Ti−OH and Co−OH) that are ubiquitous in air-exposed cobalt oxide materials.28 From the HRTEM photo (Figure 1b), the distances between two set planes in TiO2 are about 0.35 and 0.32 nm, respectively, which are in good agreement with the d-spacing of the (101) plane of anatase TiO2 and the (110) plane of rutile TiO2. The result is well-consistent with that of the XRD pattern in Figure 2, where the peaks of the (101) plane and the (110) plane are much stronger. The lattice fringes with interplanar spacing of 0.24 nm are also observed; it corresponds to the (311) plane of Co3O4. This result supports the conclusion that Co3O4 was formed. 3.2. Catalytic Activities Experiment for Co−TiO2. 3.2.1. Comparative Tests. Comparative experiments of phenol oxidation were carried out to test catalytic activities of the resulting 1.0Co−773TiO 2 powder at neutral pH. The degradation of 50 mg/L phenol as a function of reaction time is shown in Figure 5. Without KHSO5 addition and visible light, but only with Co−TiO2 as catalyst, the removal efficiency was less than 0.5%, so the absorption of phenol by catalyst is not obvious. It was also seen that approximately over 10.0% of phenol was degraded by dark/KHSO5 or KHSO5/vis processes; and almost no degradation was observed in the presence of visible light alone. Thus the visible light cannot cause the direct photolysis of phenol obviously or enhance the removal rate with KHSO5. The order is vis/Co−TiO2/KHSO5 > Co−TiO2/ KHSO5 > vis/Co−TiO2 > vis/KHSO5 ≈ dark/KHSO5 > vis/ TiO2 ≈ dark/Co−TiO2 > vis only. The combined effect of sulfate radical based advanced oxidation reaction and photo-

Table 1. Structural Property of Catalysts crystal sizeb (nm) sample TiO2, 773 K 1.0Co− 673TiO2 0.5Co− 773TiO2 1.0Co− 773TiO2 1.5Co− 773TiO2 1.0Co− 873TiO2

phase content (%)

SBETa (m2/g)

crystallinity (%)

anatase

rutile

anatase

rutile

10.5 65.0

67.4 17.8

20.5 13.0

21.9

10.4 100

89.6

22.1

75.4

20.4

20.3

35.1

64.9

28.4

69.3

18.7

18.8

70.6

29.4

33.8

66.2

16.5

15.2

66.3

33.7

2.6

65.1

26.3

100

a

SBET: BET specific surface area. b(101) for anatase TiO2, (110) for rutile TiO2.

Figure 2. XRD patterns of Co−TiO2 with different calcination temperatures and different Co2+ doping concentrations.

radius Co2+ (0.65 Å) is smaller than that of Ti4+ (0.69 Å).23 Furthermore, with the increasing of Co2+ doping concentration (from 0 to 1.5%) in Co−TiO2, the percentage of anatase phase shifts to higher value obviously, and the value of the rutile phase declines (Table 1); it may be concluded that Co2+ doping promotes the TiO2 rutile-to-anatase phase transition. In addition, since the color of the Co−TiO2 powders is green and becomes darker and darker by enhancing the Co2+ concentration (0.5−1.5%), the cobalt oxide species may be located on the surface of Co−TiO2.16 From the UV-DRS spectra (Figure 3), there are new absorption bands present between 400 and 800 nm of all Co−TiO2 samples; that means band gap absorption onset was extended into the visible light 12543

dx.doi.org/10.1021/ie4012547 | Ind. Eng. Chem. Res. 2013, 52, 12540−12549

Industrial & Engineering Chemistry Research

Article

Figure 3. UV-DRS spectra of samples: (a) different calcination temperatures; (b) different Co2+ doping concentrations (insert: Eg of 1.0Co− 773TiO2 and TiO2, 773 K).

the sum of vis/Co−TiO2 (25.7%) and Co−TiO2/KHSO5 (39.8%). Therefore, we can say that Co−TiO2/KHSO5 or vis/Co−TiO2 has limited effect, but their hybrid process exhibits an unexpected oxidation activity, suggesting a combined effect between Co−TiO2/KHSO5 and vis/Co− TiO2 is indispensable to the high catalytic activity of the hybrid reaction. At last, to rule out the possibility that the observed catalytic activity was caused by the leaching of Co2+, the leaching Co2+ ion concentrations of the catalysts were monitored as a function of reaction time (vis/Co−TiO2/ KHSO5 process) by AAS, and the curves were shown in Figure 6. We can find the concentration of leaching Co2+ ions increased initially and then maintained a fixed value of the 1.0Co−673TiO2 sample (Figure 7). So, an experiment in homogeneous process (Co2+/KHSO5 and vis/Co2+/KHSO5) was carried out using the same concentration of Co2+ ions that were leached out from the catalyst (maximum value). Under the same operating conditions, the degradation efficiency of homogeneous Co2+/KHSO5 process after 120 min was 17.5%, which was much less than the removal of 65.3% in the heterogeneous Co−TiO2/KHSO5 system. Moreover, the visible light cannot elevate the degradation efficiency obviously. Therefore, the catalytic activity was majorly attributed to the 1.0Co−773TiO2 catalyst but not the dissolved cobalt ions. 3.2.2. Effect of DCo2+ on Catalytic Activities of Co−TiO2. The effect of the Co2+ doping concentration (DCo2+) on phenol removal is illustrated in Figure 8. A great enhancement in

Figure 4. XPS survey spectra of 1.0Co−773TiO2.

catalytic process enhanced the removal of phenol significantly. The removal rate reaches 65.3% within 120 min using the Co− TiO2/KHSO5 and 42.1% in the vis/Co-TiO2 process were both higher than the KHSO5 alone; moreover, the phenol was removed completely with the simultaneous presence of Co− TiO2, KHSO5, and visible light. In Figure.6, the initial phenol concentration was doubled (100 mg/L); the combined effect was also observed. The degradation rate of vis/Co-TiO2/ KHSO5 hybrid Co/SR−Photo process (74.9%) was higher than

Figure 5. Removal of phenol under different conditions: Ccata = 1.0 g/L, MKHSO5/Mphenol = 10:1, and pH = 7.0. 12544

dx.doi.org/10.1021/ie4012547 | Ind. Eng. Chem. Res. 2013, 52, 12540−12549

Industrial & Engineering Chemistry Research

Article

Figure 6. Removal of phenol by vis/Co−TiO2, Co−TiO2/KHSO5, and vis/Co−TiO2/KHSO5 processes: (a) Cphenol = 50 mg/L, Ccata = 1.0 g/L, MKHSO5/Mphenol = 10:1, and pH = 7.0; (b) Cphenol = 100 mg/L, Ccata = 1.0 g/L, MKHSO5/Mphenol = 5:1, and pH = 7.0.

Table 2. Real Cobalt Mass Fraction of Fresh Catalysts, Concentration of Leached Co2+, and Removal Rate of Phenol by Homogenous Co/SR sample 1.0Co− 673TiO2 0.5Co773TiO2 1.0Co773TiO2 1.5Co773TiO2 1.0Co873TiO2

percentage of Co in fresh catalystsa (%)

Co2+ concn (mg/L)

leaching Co/fresh Cob

removal rate (%)

0.653

0.344

5.11

57.7

0.344

0.042

1.22

9.6

0.688

0.096

1.40

17.5

1.022

0.284

2.77

36.2

0.713

0.009

0.13

7.5

Figure 7. Cobalt leaching curve in vis/Co−TiO2/KHSO5 hybrid Co/ SR−photo processes: Ccata = 1.0 g/L, MKHSO5/Mphenol = 10:1, and pH = 7.0.

a

phenol degradation of the Co/SR process was observed by increasing the DCo2+ of samples from 0.5 to 1.5%, but the photocatalytic activity of 1.0Co−773TiO2 was the highest. The higher DCo2+ will generate more Co3O4 on the surface of catalysts and more leached Co2+ ions (Figure 7 and Table 2), the active sites for the Co/SR reaction, so the effect of the Co/ SR reaction will be more powerful. A reasonable Co2+ ion concentration doped in the TiO2 lattice will enhance the photocatalytic activity of TiO2 as it can promote the separation of light induced electron−hole pairs effectively, as well as narrow the band gap of TiO2. If the DCo2+ is too high, the Co2+ ion may also serve as a recombination center that may decrease activity of Co−TiO2. From Figure 3b, we can see when the DCo2+ was 1.0%, the catalyst represented the best light

absorbance ability, and this also supports the result of catalytic activities tests. Moreover, a higher percentage of anatase phase (Table 1) may be better for the photocatalytic activity of 1.0Co−773TiO2. The combined effect was obviou in the vis/ Co−TiO2/KHSO5 hybrid process with three samples. For example, in contrast to the individual process, the phenol removal of 0.5Co−773TiO2 in the hybrid reaction has increased by 15.7% for Co/SR and 21.3% for photo, respectively. But we also can see both 1.5Co−773TiO2 and 1.0Co−773TiO2 removed phenol completely, though photocatalytic activity of 1.0Co−773TiO2 is better. This may be

Percentage of Co in fresh catalysts: Tested by XRF and calculated with mass fraction. bLeaching Co/fresh Co: Calculated on the basis of 1.0 g of catalyst.

Figure 8. Effect of catalyst Co2+ doping concentrations on the removal rate of phenol. Reaction conditions: Cphenol = 50 mg/L; Ccata = 1.0 g/L; MKHSO5/Mphenol = 10:1; pH = 7.0; 298 K. 12545

dx.doi.org/10.1021/ie4012547 | Ind. Eng. Chem. Res. 2013, 52, 12540−12549

Industrial & Engineering Chemistry Research

Article

Figure 9. Effect of catalyst calcination temperatures on the degradation efficiency of phenol. Reaction conditions: Cphenol = 50 mg/L; Ccata = 1.0 g/L; MKHSO5/Mphenol = 10:1; pH = 7.0; 298 K.

Figure 10. Effect of (a) MKHSO5/Mphenol (Cphenol = 50 mg/L; Ccata = 1.0g/L; pH = 7.0; 298 K) and (b) Ccata (Cphenol = 50 mg/L; MKHSO5/Mphenol = 10:1; pH = 7.0; 298 K) on the removal rate of phenol.

using the same concentration of cobalt ions that were leached out from the catalysts. Under the same operating conditions, the removal of the homogeneous process after 2 h was 57.7%, 17.5%, and 7.5%. The degradation efficiency of cobalt ions released from 1.0Co−773TiO2 and 1.0−873TiO2 was much less than the removal of 65.3% and 57.4% in the heterogeneous Co−TiO2/KHSO5 process, so their catalytic activity was majorly due to the heterogeneous catalyst. However, it seems that the homogeneous Co/SR reaction has a remarkable contribution to the hybrid Co/SR−photo process of 1.0Co− 673TiO2. Furthermore, we also can find the catalyst calcined in 673 K was not stable as its Co2+ leaching was continuous with time (Figure 6). So, considering the stability, catalytic performance, and the health concerns about the adverse effect of dissolved cobalt in water, the 1.0Co−773TiO2 may be the best choice. 3.2.4. Effect of MKHSO5/Mphenol on Catalytic Activities of Co−TiO2. The results of the effect of MKHSO5/Mphenol on the removal of phenol are depicted in Figure 10a. The removal rate increased with increasing molar ratio from 5 to 10. However, when MKHSO5/Mphenol reached 20, the final removal rate decreased. This phenomenon can be explained by the oxidant and electron acceptor not being enough when MKHSO5/Mphenol was 5:1; moreover, if the KHSO5 concentration was too high, such as in MKHSO5/Mphenol = 20:1, the competitive adsorption of KHSO5 and phenol may occur on the catalyst surface. It is harmful for the heterogeneous oxidation process, so the combined effect decreases. This test indicates that the molar ratio 10:1 is best. 3.2.5. Effect of Ccata on Catalytic Activities of Co−TiO2. From Figure 10b we can see the effect of catalyst dosage (Ccata)

explained by the stronger Co/SR effect compensating for its poor photocatalytic performance. On the basis of the observed results, 1.0% was chosen as the DCo2+ in the hybrid Co/SR− photo process. 3.2.3. Effect of TC on Catalytic Activities of Co−TiO2. Figure 9 shows a comparative performance of the catalysts with different calcination temperatures (TC). As is shown, Co/SR, photo, and Co/SR−photo hybrid processes were evidently affected by the TC. When TC was 673 K, the catalyst obtained the best Co/SR activity; it removed 100% phenol in Co/SR and hybrid processes, but its photocatalytic performance was poor. This may ascribe to its larger SBET and lower crystallinity (Table 1). The larger SBET can locate more Co3O4 on the catalyst surface, but lower crystallinity may cause the decreasing of photocatalytic activity. The 1.0Co−773TiO2 showed better photocatalytic activities because a reasonable calcination temperature is a benefit for crystallization of catalyst and can enhances the anatase phase percentage of the sample (Table 1). However, if the TC was too high (873 K), the TiO2 would transform into the rutile phase completely; the crystal size became bigger and the SBET decreased severely (Figure 2 and Table 1). All of these are harmful for the photocatalytic activity of the catalyst. In addition, the TC usually affects the ion leaching of the catalyst, and the contribution of the homogeneous process (Co2+/KHSO5) increased. But, when TC was higher, the better crystallinity is obtained. Better crystallinity means less ion leaching. To compare the contributions of the homogeneous Co/SR, the concentrations of leached cobalt ion of 1.0Co−673TiO2, 1.0Co−773TiO2, and 1.0Co−873TiO2 found by using AAS were 0.344 mg L−1, 0.096 mg L−1, and 0.009 mg L−1, respectively (Figure 6 and Table 2). An experiment in homogeneous process was implemented 12546

dx.doi.org/10.1021/ie4012547 | Ind. Eng. Chem. Res. 2013, 52, 12540−12549

Industrial & Engineering Chemistry Research

Article

Figure 11. TOC removal of phenol: (a) vis/Co−TiO2, Co−TiO2/KHSO5, and vis/Co−TiO2/KHSO5 processes. Reaction conditions: Cphenol = 50 mg/L; Ccata = 1.0 g/L; MKHSO5/Mphenol = 10:1; pH = 7.0; 298 K. (b) Reaction conditions: Cphenol = 50 mg/L; MKHSO5/Mphenol = 10:1; pH = 7.0; 298 K. (c) Reaction conditions: Cphenol = 50 mg/L; Ccata = 1.0 g/L; pH = 7.0; 298 K.

on reactivity. With the amount of catalyst increasing, the active sites for Co/SR and photocatalytic reaction increased. But superfluous catalyst (such as 1.5 g/L in the test) also leads to the consumption of KHSO5 quickly. A great amount of SO4•− was generated in a short time and might not react sufficiently with phenol molecularly since a great part of them were annihilated in mutual collision or with reactor. Moreover, the penetration of light was diminished when the suspension catalyst powder in solution was too much; it would also decrease the effect of photocatalysis. This may be the reason why the degradation rate of phenol was improved with the dosage increasing from 0.5 to 1.0 g/L but decreasing from 1.0 to 1.5 g/L. 3.2.6. TOC Removal of Phenol. TOC removal efficiency is a significant standard from which to evaluate a new advanced oxidation technology, because our ultimate aim is to oxidize the organic pollutant to CO2 and H2O completely. In this research, the hybrid process not only enhances the removal rate of phenol but also improves its mineralization (Figure 11a). In contrast with 39.8% of photo and 44.5% of Co/SR, the hybrid process reached a 76.2% TOC removal rate. Furthermore, the operation parameters (MKHSO5/Mphenol and Ccata) affected the TOC removal. The TOC−time evolution curves (Figure 11b,c) showed a tendency similar to phenol removal. It proves the favorable concentration and light irradiation are significant to mineralization of phenol. 3.3. Reusability of the Co−TiO2 Catalyst. As shown in Figure 12, the regenerated catalyst exhibited good performance. The activity of the catalyst (reaction rate) dropped slightly with the used time. The activity reduction may be due to conglomeration of the catalyst during the centrifugal process and effect of contaminant adsorption on reactive sites on the catalyst surface, as observed for other heterogeneous catalysts. The concentration of the dissolved cobalt ions from Co−TiO2 in five runs did not fluctuate obviously (0.096, 0.094, 0.094, 0.094, and 0.094 mg/L). Therefore, the Co−TiO2 has good catalytic performance for reuse. 3.4. Proposed Catalytic Mechanism of Co−TiO2. In this study, Co2+ ions were doped into TiO2 by sol−gel method. A part of them is incorporated into the TiO2 lattice and those remaining formed Co3O4 immobilized on the surface of catalyst that is proved by UV-DRS and XPS investigation. Cobalt appears in two redox states in Co3O4 as CoII in CoO and CoIII in Co2O3; that is why Co−TiO2 has activity to active KHSO5 to generate SO4•− (eqs 8 and 9).11,13,28,29 Moreover, some cobalt ions leached from catalysts perform homogeneous Co/SR reaction. The contribution of this homogeneous effect may change with DCo2+ and TC (see Table 2). On the other hand,

Figure 12. Degradation of phenol in consecutive runs using the recycled 1.0Co−773TiO2 in the vis/Co−TiO2/KHSO5 process. Reaction conditions: Cphenol = 50 mg/L; MKHSO5/Mphenol = 10:1; pH = 7.0; 298 K.

the cobalt ions in TiO2 lattice enhance the photocatalytic activity by reduction of photoinduced electron−hole (ecb−/ hvb+) recombination (eqs 11 and 12), and they also formed impurity energy levels in the TiO2 bandgap that extend the photoresponse of Co−TiO2 into the visible light region.14,15,31 CoIII + ecb− → CoII

(11)

CoII + hvb+ → CoIII

(12)

In the cis/Co−TiO2/KHSO5 hybrid process, the conduct band electrons from catalyst may promote the catalytic cycle of CoII/ CoIII 31 (eqs 11 and 12), like some photocatalysis and FeII/ FeIII−H2O2 mixed processes;7,9 thus, they enhance the effect of the Co/SR process. Simultaneously, KHSO5 takes a function of electron acceptor to inhibit ecb−/hvb+ recombination (eqs 13 and 14).31 The separation of ecb−/hvb+ makes valence band holes more available for oxidation of adsorbed organic compounds, and for oxidation of surface hydroxyl groups and adsorbed water molecules into •OH. The SO4•− and •OH combined to a hybrid radical system, and its oxidizability may be more powerful than a single one (eqs 15−17).

12547

HSO5− + ecb− → SO4•− + OH−

(13)

HSO5− + ecb− → •OH + SO4 2 −

(14)

SO4•− + H 2O → SO4 2 − + •OH + H+

(15)

dx.doi.org/10.1021/ie4012547 | Ind. Eng. Chem. Res. 2013, 52, 12540−12549

Industrial & Engineering Chemistry Research

■

Article

hvb+ + H 2O → •OH + H+

(16)

hvb+ + OH− → •OH

(17)

(11) Cao, X. Z.; Song, T. Y.; Wang, X. Q. Inorganic Chemistry, 3rd ed.; , High Education Press: Beijing, 2000; pp 1026−1035. (12) Anipsitakis, G. P.; Stathatos, E.; Dionysiou, D. D. Heterogeneous Activation of Oxone Using Co3O4. J. Phys. Chem. B. 2005, 109, 13052−13055. (13) Chan, K. H.; Chu, W. Degradation of Atrazine by CobaltMediated Activation of Peroxymonosulfate: Different Cobalt Counteranions in Homogenous Process and Cobalt Oxide Catalysts in Photolytic Heterogeneous Process. Water Res. 2009, 43, 2513−2521. (14) Le, T. T.; Akhtar, M. S.; Park, D. M.; Lee, J. C.; Yang, O. B. Water Splitting on Rhodamine-B Dye Sensitized Co-Doped TiO2 Catalyst under Visible Light. Appl. Catal., B 2012, 111−112, 397−401. (15) Ganesh, I.; Gupta, A. K.; Kumar, P. P.; Chandra, Sekhar, P. S.; Radha, K.; Padmanabham, G.; Sundararajan, G. Preparation and Characterization of Co-Doped TiO2 Materials for Solar Light Induced Current and Photocatalytic Applications. Mater. Chem. Phys. 2012, 135, 220−234. (16) Long, M. C.; Cai, W. M.; Cai, J.; Zhou, B. X.; Chai, X. Y.; Wu, Y. H. Efficient Photocatalytic Degradation of Phenol over Co3O4/BiVO4 Composite under Visible Light Irradiation. J. Phys. Chem. B 2006, 110, 20211−20216. (17) Huang, K. C.; Couttenye, R. A.; Hoag, G. E. Kinetics of HeatAssisted Persulfate Oxidation of Methyl tert-Butyl Ether (MTBE). Chemosphere 2002, 49, 413−420. (18) Liang, C. J.; Wang, Z. S.; Bruell, C. J. Influence of pH on Persulfate Oxidation of TCE at Ambient Temperatures. Chemosphere 2007, 66, 106−113. (19) Water QualityDetermination of Volatile Phenolic Compounds 4-AAP Spectrophotometric Method, National Environmental Protection Standard HJ 503-2009, Ministry of Environmental Protection, The People’s Republic of China, Beijing, PRC, 2009. (20) Chitrakar, R.; Makita, Y.; Hirotsu, T.; Sonoda, A. Selective Uptake by Akaganeite (β-FeOOH) of Phosphite from Hypophosphite and Phosphite Solutions. Ind. Eng. Chem. Res. 2012, 51, 972−979. (21) Al-Jaroudi, S. S.; Ul-Hamid, A.; Mohammed Abdul-Rashid, I.; Saner, S. Use of X-ray Powder Diffraction for Quantitative Analysis of Carbonate Rock Reservoir Samples. Powder Technol. 2007, 175, 115− 121. (22) Cao, C. L.; Hu, C. G.; Shen, W. D.; Wang, S. X.; Wang, J. L.; Tian, Y. S. Fabrication of a Novel Heterostructure of Co3O4-Modified TiO2 Nanorod Arrays and Its Enhanced Photoelectrochemical Property. J. Alloys Compd. 2013, 550, 137−143. (23) Liu, X. H.; He, X. B.; Fu, Y. B. Effects of Doping Cobalt on the Structures and Performances of TiO2 Photocatalyst. Acta Chimca Sinica 2008, 66, 1725−1730. (24) Jayakumar, O. D.; Sudakar, C.; Persson, C.; Sudarsan, V.; Sakuntala, T.; Naik, R.; Tyagi, A. K. 1D Morphology Stabilization and Enhanced Magnetic Properties of Co:ZnO Nanostructures on Co Doping with Li: A Template-Free Synthesis. Cryst. Growth Des. 2009, 9, 4450−4455. (25) Kim, Y. D.; Cooper, S. L.; Klein, M. V.; Jonker, B. T. Spectroscopic Ellipsometry Study of the Diluted Magnetic Semiconductor System Zn (Mn, Fe, Co) Se. Phys. Rev. B 1994, 49, 1732− 1742. (26) Zayat, M.; Levy, D. Blue CoAl2O4 Particles Prepared by the Sol−Gel and Citrate−Gel Methods. Chem. Mater. 2000, 12, 2763− 2769. (27) Lu, Y. C.; Lin, Y. H.; Wang, D. J.; Wang, L. L.; Xie, T. F.; Jiang, T. F. A High Performance Cobalt-Doped ZnO Visible Light Photocatalyst and Its Photogenerated Charge Transfer Properties. Nano Res. 2011, 4, 1144−1152. (28) Yang, Q. J.; Choi, H.; Dionysiou, D. D. Nanocrystalline Cobalt Oxide Immobilized on Titanium Dioxide Nanoparticles for the Heterogeneous Activation of Peroxymonosulfate. Appl. Catal., B 2007, 74, 170−178. (29) Shi, P. H.; Su, R. J.; Wan, F. Z.; Zhu, M. C.; Li, D. X.; Xu, S. H. Co3O4 Nanocrystals on Graphene Oxide as a Synergistic Catalyst for Degradation of Orange II in Water by Advanced Oxidation

AUTHOR INFORMATION

Corresponding Author

*Tel./Fax: + 86 (23) 65127537. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank the Fundamental Research Funds for the Central Universities (Grant No. CDJXS12210002), the Major Project Foundation of Science and Technology Innovation in Minister of Education (Grant No. 708071), the National Natural Science Foundation of China (Grant No. 51108483), and the Natural Science Foundation Project of CQ CSTC (Grant No. cstcjjA20002) for financial support of this research.

■

ABBREVIATIONS BE = binding energy (eV) Ccata = dosage of catalyst (g/L) Cphenol = initial phenol concentration (mg/L) DCo2+ = Co2+ ion doping concentration (%) MKHSO5/Mphenol = molar ratio of KHSO5 to phenol photo = photocatalysis/photocatalytic process TC = calcinations temperature (K)

■

REFERENCES

(1) Santos, A.; Yustos, P.; Rodriguez, S.; Simon, E.; Romero, A. Fenton Pretreatment in the Catalytic Wet Oxidation of Phenol. Ind. Eng. Chem. Res. 2010, 49, 5583−5587. (2) Rey, A.; Carbajo, J.; Adán, C.; Faraldos, M.; Bahamonde, A.; Casas, J. A.; Rodriguez, J. J. Improved Mineralization by Combined Advanced Oxidation Processes. Chem. Eng. J. 2011, 174, 134−142. (3) Sarmah, S.; Kumar, A. Photocatalytic Activity of Polyaniline− TiO2Nanocomposites. Indian J. Phys. 2011, 85, 713−726. (4) Liu, L. J.; Zhang, G. L.; Wang, L.; Huang, T.; Qin, L. Highly Active S-Modified ZnFe2O4 Heterogeneous Catalyst and Its PhotoFenton Behavior under UV−Visible Irradiation. Ind. Eng. Chem. Res. 2011, 50, 7219−7227. (5) Riga, A.; Soutsas, K.; Ntampegliotis, K.; Karayannis, V.; Papapolymerou, G. Effect of System Parameters and of Inorganic Salts on the Decolorization and Degradation of Procion Hexl Dyes. Comparison of H2O2/UV, Fenton, UV/Fenton, TiO2/UV, and TiO2/ UV/H2O2 Processes. Desalination 2007, 211, 72−86. (6) Wu, D. H.; You, H.; Zhang, R.; Chen, C.; Lee, D. J. Ballast Waters Treatment Using UV/Ag−TiO2 + O3 Advanced Oxidation Process with Escherichia coli and Vibrio alginolyticus as Indicator Microorganisms. Chem. Eng. J. 2011, 174, 714−718. (7) Selvam, K.; Muruganandham, M.; Sobana, N.; Swaminathan, M. Enhancement of UV-Assisted Photo-Fenton Degradation of Reactive Orange 4 Using TiO2-P25 Nanoparticles. Sep. Purif. Technol. 2007, 54, 241−247. (8) Cong, Y. Q.; Li, Z.; Zhang, Y.; Wang, Q.; Xu, Q. Synthesis of αFe2O3/TiO2 Nanotube Arrays for Photoelectro-Fenton Degradation of Phenol. Chem. Eng. J. 2012, 191, 356−363. (9) Kim, H. E.; Lee, J. S.; Lee, H. S.; Lee, C. H. Combined Effects of TiO2 Photocatalysis in Combination with Fenton-like Reactions on Oxidation of Organic Compounds at Circumneutral pH. Appl. Catal., B 2012, 115−116, 219−224. (10) Chen, F.; Xie, Y. D.; He, J. J.; Zhao, J. C. Photo-Fenton Degradation of Dye in Methanolic Solution under Both UV and Visible Irradiation. J. Photochem. Photobiol., A 2001, 138, 139−146. 12548

dx.doi.org/10.1021/ie4012547 | Ind. Eng. Chem. Res. 2013, 52, 12540−12549

Industrial & Engineering Chemistry Research

Article

Technology Based on Sulfate Radicals. Appl. Catal., B 2012, 123−124, 265−272. (30) Huang, W.; Zuo, Z. J.; Han, P. D.; Li, Z. H.; Zhao, T. D. XPS and XRD Investigation of Co/Pd/TiO2 Catalysts by Different Preparation Methods. J. Electron Spectrosc. Relat. Phenom. 2009, 173, 88−95. (31) Ram, M. K.; Andreescu, S.; Ding, H. M. Nanotechnology for Environmental Decontamination; McGraw-Hill Professional: New York, 2011; pp 173−175.

12549

dx.doi.org/10.1021/ie4012547 | Ind. Eng. Chem. Res. 2013, 52, 12540−12549