Document Type : Research Paper
Author
Department of Metallurgy and Materials Engineering, University of Kashan, Kashan, Iran
Abstract
Keywords
INTRODUCTION
Synthetic dyes as one of the organic compounds released from various chemical industries such as textile, pigment, food, and cosmetics, are hazardous organic pollutants that have caused severe threats to both human health and the environment [1]. Even low dye concentrations contaminated in effluents may cause mutagenesis and carcinogenesis, and block the ecosystem [2]. Anionic reactive dyes, comprising reactive red 141 (RR141) dye are complex, very stable and resistant to decolorization due to their aromatic anthraquinone structure with high molecular weight [3]. Hence, developing an efficient and cost-effective technology to promptly eliminate these harmful dyes in wastewater is essential. So far, several methods, including biosorption [4], active carbon adsorption [5], Fenton-like oxidation [6], photocatalytic [7], and electrocoagulation [8], have been studied to remove dye pollution from industrial wastewater. Most of the mentioned methods generally have drawbacks for dye elimination, which limit their applications. Recently, the photocatalytic process as a new approach has been adopted and gained extensive attention. It offers the advantages of economy, high efficiency in decolorizing the dyes without producing harmful products, and environmental friendliness [9]. Numerous semiconducting photocatalysts are utilized to decolorize dyes such as SnO2 [10], TiO2 [11], ZnO [12], Ag3PO4 [13], and BiVO4 [14]. Among them, ZnO is generally an appropriate candidate owing to its non-toxic nature, chemical stability, photostability, abundance, cost effectiveness, eco-friendliness, suitable band gap (around 3.0 eV), strong oxidation capability, and potential photocatalytic efficiency [15-17], which acts as a superior photocatalyst material for the degradation of dye pollutants. Unfortunately, because of the wide bandgap, ZnO can only be excited under UV light, which is a small fraction of sunlight (3-5%) [18]. Besides, the high recombination rate of photoinduced electron-hole pairs greatly limited the photocatalytic efficiency of ZnO [19]. Some approaches have been developed to solve these deficiencies, such as noble metal deposition [20], element doping [21], and coupling with narrow band gap semiconductors to form binary or trinary heterojunctions [22, 23]. Among the narrow band gap materials, Co3O4, a p-type semiconductor has been widely studied for the photocatalytic degradation of dyes because of its absorption capability of visible light and valence band potential being close to the conduction band potential of ZnO. Hence, it creates a suitable heterojunction with ZnO for the maximum photocatalytic application [24]. The rapid recombination of charge carriers has reduced the photocatalytic efficiency of Co3O4, which can be partially modified by introducing doping elements to the Co3O4 lattice [25]. Recently, Abo Zeid et al. [26] reported that Ni-, Cu, Ni- and Cd, Ni-doped Co3O4 nanoparticles show much higher photocatalytic efficiency for the degradation of methyl orange dye than that of pure Co3O4. Keerthana et al. [27] synthesized 1 M Sn-doped Co3O4 nanoparticles for the photocatalytic degradation of methylene blue dye under visible light. Fida et al. [28] successfully doped Co3O4 with Mn, a composite of carbon nanotubes via co-precipitation. Herein, binary ZnO/Ni-Co3O4 nanocomposites have been successfully fabricated and the synergistic effects of Ni-doped Co3O4 coupled with ZnO nanoparticles was investigated to photocatalytic decolorization of reactive red 141 dye. The prepared heterojunction exhibited enhanced photocatalytic activity compared with single ZnO and Co3O4 nanoparticles. The probable photocatalytic mechanism of ZnO/Ni-Co3O4 nanocomposite under tungsten lamp irradiation was also discussed.
MATERIALS AND METHODS
Synthesis of ZnO/Ni-doped Co3O4 nanocomposite
In the present work, a sol-gel process operating at air atmospheric was employed to synthesize ZnO/Ni-doped Co3O4 nanocomposite. All chemical reagents were purchased from Sigma-Aldrich and used without further purification. To prepare a 0.25 M solution of pure Co3O4, 1.8 g hexahydrate nitrate cobalt (Co(NO3)2·6H2O; 99%) and 3.4 g urea (CO(NH2)2; 97%) were dissolved in 30 mL of water and stirred magnetically for 30 min. To obtain Ni-doping, hexahydrate nickel nitrate (Ni(NO3)3·6H2O; 97%) was added to the precursor solution with different amounts (2, 5, and 8 wt%) for all doped samples. The crystalline Fe-doped Co3O4 nanoparticles were obtained after drying at room temperature and then calcination at 400 °C for 3 h. To synthesize ZnO nanoparticles, 2.2 g of zinc nitrate hexahydrate (Zn(NO3)2·6H2O; 98%) was dissolved in 40 mL of DI water and stirred for around 30 min (solution 1). In another beaker, 8 g sodium hydroxide (NaOH; ≥ 98%) was added to 40 mL DI water (solution 2). In the next step, 10 mL of solution 1 was added dropwise to solution 2 and stirred for 2 h to achieve a milky white solution. Afterward, the treated solution was dried and calcined at 450 °C for 1 h. Finally, the prepared ZnO and Ni-doped Co3O4 nanoparticles with a weight ratio of 80:20 were mixed manually to obtain the homogeneous ZnO/Ni-doped Co3O4 composites.
Sample Characterization
A field emission scanning electron microscope (FE-SEM, MIRA3, TESCAN) equipped with energy dispersive X-ray spectroscopy (FESEM-EDS) were utilized to characterize the surface morphology and constituent composition of the as-prepared products. A PANalytical X’Pert Pro diffractometer with CuKα radiation (λ=1.542 oA) at 40 kV and 30 mA was used to obtain the crystalline phase, as well as the phase composition of the as-fabricated samples. The functional groups of the prepared samples were measured using the Fourier transform infrared spectroscopy (FTIR, Nicolet Magna 550 IR) in the wavenumber range from 400 to 4000 cm-1. UV-visible diffuse reflectance spectra (UV-vis DRS) were recorded on a UV-vis V900 spectrometer in the wavelength range of 300-800 nm.
Photocatalytic decolorization process
Reactive red (RR141) dye was used to investigate the photocatalytic activity of the synthesized individual ingredients and its coupled nanostructures. The photocatalytic decolorization process was performed in a simple photoreactor (250 mL capacity) with a 400 W tungsten lamp as the light source. In a particular procedure, 50 mg of the prepared photocatalyst was added to 50 mL (10 mg/L) aqueous solution of RR141 dye. Then, the catalyst-containing solution was continuously stirred for 20 min in the dark to achieve adsorption-desorption equilibrium between the dye molecules and catalyst. At selected time intervals, 5 mL of aliquot of the mixed solution was withdrawn from the reactor and filtered for analysis through Millipore syringe filter membrane to separate the residual catalyst particulates and solution. The change in the concentration of RR141 was monitored at the corresponding wavelength with the UV-vis V900 spectrometer during the photocatalytic process. The photocatalytic decolorization efficiency (η) of the RR141 was determined according to the following formula:
η=(A0−At)/A0)100%
where A0 and At are the absorbances of the RR141 solution at the initial and each time interval of light irradiation, respectively.
RESULTS AND DISCUSSION
Materials Characterizations
SEM image and EDX pattern of ZnO/5Ni-doped Co3O4 nanocomposite prepared by sol-gel method are represented in Fig. 1. As displayed in Fig. 1a, nanocomposites are composed of both ZnO and Co3O4 nanoparticles that have a spherical morphology with a uniform size distribution and slight agglomeration. The average particle size of the nanoparticles was in the range of about 20-40 nm. EDS analysis was performed to determine the elemental composition of ZnO/5Ni-doped Co3O4 nanocomposite, as shown in Fig. 1b. The EDS spectrum demonstrated the existence of Zn, Co, and O and a small amount of Ni with elemental weight percentage of 46.9 % for Zn, 30.2 % for Co, 22.4 % for O, and 0.5 % for Ni. Assuming that all of the Zn atoms were bound with O as ZnO, the weight ratio of Co to O was 30:11, which was consistent with the expected formation of Co3O4. Thus, EDS data confirm the formation of the ZnO/5Ni-doped Co3O4 nanocomposite without any impurities.
Fig. 2 shows the XRD patterns of ZnO nanoparticles, Co3O4 nanoparticles, and ZnO/Ni-doped Co3O4 heterostructured nanocomposites with different amounts of Ni dopant. XRD analysis of ZnO nanoparticles reveals the main diffraction peaks at 2θ values of 31.7o, 34.4o, 36.1o, 47.5o, 56.5o, 62.9o, and 67.9o, which are attributed to the (100), (002), (101), (102), (110), (103), and (112) planes and matched with the JCPDS card No. 36-1451 of the hexagonal wurtzite structure of ZnO [29]. The apparent peaks of Co3O4 are of the cubic spinel type phase located at 2θ of 19.0o, 31.3°, 36.9°, 38.6°, 44.8°, 55.7o, 59.4°, 65.2°, and 77.3° corresponding to crystal planes of (111), (220), (311), (222), (400), (422), (511), (400), and (533), respectively (JCPDS card No. 043-1003) [30]. The XRD pattern of ZnO/Co3O4 heterojunction contains the characteristic diffraction peaks of both ZnO and Co3O4 materials in the nanocomposite. No impurity peak or no remarkable shift from the diffraction peaks of ZnO and Co3O4 was observed in the ZnO/Co3O4 sample, confirming the coexistence of ZnO and Co3O4 without the formation of a solid solution. In the case of ZnO/Ni-doped Co3O4 samples, no extra peaks associated to Ni clusters, nickel oxides or other oxide phases can be observed, indicating the high purity of synthesized samples. It is also showed that the peaks of Co3O4 are slightly shifted towards lower diffraction angles with increasing the Ni content This suggests that the Ni ions were well incorporated into the Co sites of Co3O4 lattice. Further, the crystallite size and lattice parameters of Ni-doped Co3O4 were calculated from the XRD analysis and listed in Table 1. The data analysis indicated that there is a decrease in the values of the average crystallite size and an increase in lattice constant by increasing the Ni contents. The values of average crystallite size obtained by using the classical Debye-Scherrer formula [31] were found be 28.2 nm for Co3O4 nanoparticles and it decreased to 19.4 nm by increasing the Ni content steadily up to 8 wt%. The lattice parameter values for the cubic structure were determined using the following equation:
A slight increase from 8.0882 to 8.1267 oA was observed in lattice parameter of Co3O4 structure with doping by Ni from 0 to 8 wt%. This can be explained by the substitution of smaller Co2+ (0.65 oA) or Co3+ (0.54 oA) ions by larger Ni2+ ions (0.69 oA) in coordination sites [32].
FTIR data were conducted to confirm the functional groups present in Co3O4 and Ni-doped Co3O4 nanoparticles and the results are illustrated in Fig. 3. All samples exhibited the broad absorption bands around 3522 cm-1 and 1658 cm-1 are corresponds to the stretching and bending vibrations of hydroxyl groups (O-H) generated by the adsorbed water molecules. As can be clearly seen in FTIR spectra, two absorption bands were found at 559 cm-1 and 657 cm-1, which were assigned to the stretching vibration of metal-oxygen bonding interactions (Co-O or Ni-O) within the spinel Co3O4 structure [33]. The first band at 559 cm-1 is attributed to the vibrations of Co3+ ions occupying the octahedral sites in the spinel structure of Co3O4 [34]. Similarly, the second band at 657 cm-1 is associated with the Co2+ ions in the tetrahedral sites [35]. These peaks for Ni-doped Co3O4 exhibited a slight shift compared to Co3O4 due to Ni dopant.
UV-vis DRS analysis was employed to determine the optical properties of synthesized photocatalysts. Fig. 4 shows the UV-vis spectra of ZnO nanoparticles, Co3O4 nanoparticles, and ZnO/Ni-doped Co3O4 heterostructured nanocomposites. Pristine ZnO displayed the absorbance band around 300-380 nm in the UV light region. For pristine Co3O4 nanoparticles was two pronounced absorption bands at 380-540 nm and 730-800 nm, which are ascribed to the ligand-to-metal charge transfer process (O2−→Co2+ and O2−→Co3+) in Co3O4 structure [30]. The absorption curves of ZnO/2Ni-Co3O4, ZnO/5Ni-Co3O4, and ZnO/8Ni-Co3O4 nanocomposites indicated a slightly red shift compared with pristine ZnO. It confirms that the coupling ZnO with Co3O4 and Ni dopant amount effects on the extended light absorption toward the visible region, suggesting enhanced visible-light-driven photocatalytic activity.
Meanwhile, the optical band gap energies (Eg) are determined according to Tauc relation using Eq. (2) [36]:
αhν=A(hν-Eg)n
where, α is the absorption coefficient, hν is photon energy, A is constant, Eg is the bandgap energy, and n=2 for a indirectly allowed transition and n=1/2 for a directly allowed transition. Here, the best fitting gives n=1/2, indicating that the prepared photocatalysts have directly allowed electronic transitions. The band gap energies of samples were estimated by extrapolating the linear part of plots of (αhν)2 axis to (hν) axis at (𝛼hυ)2=0, as shown in the inset of Figs. 4a-f. Eg estimation from the Tauc plots yielded value of 3.25 eV for ZnO nanoparticles. For the Co3O4 nanoparticles, two band gap energies of 2.54 eV and 1.73 eV assigned to the charge transfer processes. It can be also found that the Eg values for ZnO/Co3O4, ZnO/2Ni-Co3O4, ZnO/5Ni-Co3O4, and ZnO/8Ni-Co3O4 nanocomposites was 2.9, 3.78, 2.71, and 2.56 eV, respectively. The optical band gap of ZnO/Co3O4 sample is about 0.35 eV lower than those of pristine ZnO nanoparticles, which this change may be due to the fact that the addition of Co3O4 adjusts the fermi energy levels across the ZnO/Co3O4 heterojunction [37]. Meanwhile, the incorporation of Ni into the Co3O4 structure gradually reduces the Eg of ZnO/Co3O4 nanocomposites. Ni ions contribute to generate the defects, oxygen vacancies as well as increase the number of charge carriers that leads to more conductivity of Co3O4 nanoparticles.
The photocatalytic activity of the produced photocatalyst materials was evaluated by the decolorization of RR141 solution under tungsten lamp irradiation. In order to study the optimum photocatalyst dosage on the photocatalytic decolorization process, different amount of ZnO/5Ni-Co3O4 nanocomposite were examined, while the initial dye concentration was kept constant at 10 mg/L. Fig. 5 depicts decolorization of RR141 dye with the variation in the amount of ZnO/5Ni-Co3O4 nanocomposite from 0.5 g/L to 1.2 g/L in the presence of tungsten lamp irradiation. It is observed that the photocatalytic decolorization efficiency and rate initially increased from 83.7% (0.0177 min-1) for 0.5 g/L to 97.4% (0.0373 min-1) for 1 g/L and thereafter decreases to 94.5% (0.0300 min-1) for 1.2 g/L. Considering these results, 1 g/L dosage is the optimum amount of photocatalyst for the photocatalytic degradation of the dye. At catalyst concentration below this optimal value, the effective surface of catalyst is fewer available to the dye molecules and absorption of light. Increase in catalytic dosage increases the number of active sites on the surface, which results in more generation of •O2– and •OH radicals [38]. Furthermore, when the catalyst dosage exceeds above 1 g/L, a negative effect was observed in the photocatalytic efficiency of the system. This phenomenon could be explained by the light scattering and the screening effect on the surface of the photocatalyst [39].
The effect of initial dye concentration on the decolorization of RR141 by ZnO/5Ni-Co3O4 nanocomposite was studied with variation of dye concentrations (5,10, 20, and 30 mg/L) at constant catalyst dosage of 1 g/L. As can be seen in Fig. 6, the decolorization efficiency of RR141 was decreased from 98 to 81.8% with increasing the initial concentration of dye from 5 to 30 mg/L after 100 min. The large amount of adsorbed dye molecules on the catalyst surface leads to lack of direct contact between them, reduction of penetrated light and decreased active sites to the dye-loaded surface, causing an inhibitive role on the reaction of dye molecules with photogenerated electron/holes or reactive radicals [40].
Figs. 7a and b illustrate the plots of C/C0 and -Ln (C/C0) against the reaction time, respectively. It can be observed that the decolorization of RR141 is resistant in the absence of irradiation. Also, the self-decolorization test confirmed that RR141 is almost no decolorization without a photocatalyst, demonstrating that the photolysis effect may be negligible. Thus, catalyst as well as light irradiation are required for efficient photocatalytic decolorization of RR141 dye. From the photocatalytic experiments, it can be found that among the synthesized photocatalysts, ZnO/5Ni-Co3O4 nanocomposite exhibited the best photocatalytic efficiency where 97.4% of RR141 was decolorized within 100 min of irradiation with a rate constant of 0.0373 min-1. The good linearity relationship approves that the decolorization of RR141 obeys the pseudo-first-order reaction kinetics. This was then followed by ZnO/8Ni-Co3O4 nanocomposite where it indicated a decolorization of 93.1% with a rate constant of 0.0252 min-1. The photocatalytic activity declined to 79.4% with a rate constant of 0.0144 min-1 when using ZnO/2Ni-Co3O4 nanocomposite as the photocatalyst. This values for ZnO/Co3O4 nanocomposite were 67.1% and 0.0107 min-1, respectively. However, pristine ZnO and Co3O4 nanoparticles decreased 34.8% and 39.2% of RR141 within 100 min of irradiation with the rate constants of 0.0042 min-1 and 0.0048 min-1, respectively. The high activity of nanocomposites is possibly due to the suppressed recombination of photogenerated carriers as well as the improved charge transport, which derived from the synergistic effects of Ni-doping in the Co3O4 lattice and its coupling with ZnO nanoparticles. The efficiency of the synthesized photocatalyst for decolorization of RR 141 dye in comparison to some recently reported ZnO based photocatalysts was also summarized in Table 2.
To examine the photocatalytic stability and reusability, the repeated experiments of RR141 decolorization the optimized ZnO/5Ni-Co3O4 nanocomposite were implemented for six subsequent runs in a total of 10 h. After each test, the photocatalyst was recovered through washing by distiller water and then was reused in the next run using the same conditions. As depicted in Fig. 8a, the removal efficiency decreased from 97.4% in the first cycle to 93.8% in the third run, suggesting that the prepared composite has well stability for the practical applications. Fig. 8b shows the XRD patterns of ZnO/5Ni-Co3O4 nanocomposite before use and after six cycling test, demonstrating tha the positions of the feature peaks before and after use are the same, and the intensities of their peaks only have a negligible decrease. It well exhibits an excellent stability of the ZnO/5Ni-Co3O4 photocatalyst after photocatalytic activity.
To better understand the roles of active species participated during the photocatalytic reactions, different quenching agents were added into the ZnO/5Ni-Co3O4 system. Benzoquinone (BQ; as a quencher of O2−), Isopropanol (IPA; as a quencher of OH−) and Triethanolamine (TEA; as a quencher of h+) were respectively used under the same conditions with photocatalytic decolorization experiments. Fig. 9 presents the photocatalytic efficiency of RR141 over ZnO/5Ni-Co3O4 heterostructure was 79, 33.9, and 55.4% with TEA, IPA and BQ, respectively. These results suggest that hydroxyl radicals play more important role in RR141 decolorization process, compared to the photo-generated superoxide (•O2–) radicals and holes.
Based on the above-mentioned results, a probable mechanism of the decolorization via ZnO/Ni-doped Co3O4 heterojunction was proposed and illustrated in Fig. 10. ZnO is an n-type semiconductor and Co3O4 is known as a p-type semiconductor. The Fermi level of the n-type ZnO is located near the conduction band (CB), while that of p-type Co3O4 is located near the valence band (VB). The CB and VB potentials of semiconductor materials can be estimated using following relations [47]:
ECB= χ-E0-0.5Eg
EVB= ECB+Eg
where ECB and EVB are the potentials of the conduction and valence band edge, χ is defined as the absolute electronegativity of constituent atoms, E0 is the energy of free electrons on the hydrogen scale (about 4.5 eV), which the χ values for ZnO and Co3O4 are 5.79 eV [9] and 5.90 eV [48], respectively, and Eg is band gap of the photocatalyst. Ni doping into Co3O4 lattice can induce additional mid-gap energy levels just below the CB edge, which creates photogenerated electron-trapping sites [49]. After contact, the Fermi levels of the two components reaches a thermodynamic equilibrium which leads to create an electric field in the Interface of n-p ZnO/Co3O4 heterojunction and shift the band positions in the two components. Thus, the band positions of the Co3O4 is up shifted beyond the band positions of ZnO according to type-II band gap configuration. Therefore, it can be concluded that the Ni doping in the Co3O4 structure along with its coupling with ZnO serve to separate the charge carriers, resulting in the enhanced photocatalytic activity. Under light irradiation, both ZnO and Co3O4 can be excited, and photo-generated electrons move from the CB of Co3O4 to that of ZnO. This electrons on the CB of ZnO were trapped by molecular oxygen to form •O2– radicals, while the photo-generated holes migrated from the VB of ZnO to that of Co3O4 react with hydroxyl ions (OH–) to form •OH radicals. These free radicals are strong oxidizing agents to decompose dye molecules.
ZnO/Ni-Co3O4 + hν → ZnO/Ni-Co3O4 (e–CB+h+VB)
Ni-Co3O4 (e-CB) + ZnO → ZnO (e-CB)
ZnO (h+VB) + Ni-Co3O4→ Ni-Co3O4 (h+VB)
O2 + ZnO (e-CB) → •O2–
Ni-Co3O4 (h+VB) + H2O (ads) → •OH + H+
Ni-Co3O4 (h+VB) + OH- (ads) → •OH
•O2– + H+ → •HO2
•HO2 + •HO2 → H2O2 + O2
H2O2 + e- → •OH + OH-
•O2– + •OH + h+ + RR141 dye molecules → decolorized products
CONCLUSION
In summary, a type-II band alignment heterojunction of ZnO/Ni-doped Co3O4 nanocomposites with different amounts of Ni dopant has successfully fabricated via sol-gel method, and the resultant photocatalysts demonstrated a significant enhancement and stability in photocatalytic decolorization efficiency of RR141 dye. The photocatalytic results revealed a maximum decolorization performance of 97.4% with a rate constant of 0.0373 min-1 after 100 min light irradiation. The prepared nanocomposites exhibited outstanding stability even after six cycles, which confirms their potential for reuse and long-term practical applications in the environmental field.
CONFLICT OF INTEREST
The authors declare that there is no conflict of interests regarding the publication of this manuscript.