Enhancement of TiO2 activity under visible light by N,S codoping for Pb(II) removal from water

This paper deals with a systematic study on the co-doping N,S on TiO2 photocatalyst to improve its activity under visible light on the removal of Pb(II) from the aqueous media. The co-doping TiO2 by N,S atoms was conducted in an autoclave by one-step hydrothermal of TiO2 mixed with nitric and sulfuric acids as the sources of N and S, respectively. The mole ratio of TiO2:nitric acid:sulfuric acid was varied as 1:1:0.5, 1:1:1, and 1:1:1.5 to find the best ratio toward the activity. The co-doped photocatalysts obtained were characterized by specular reflectance UV/Vis (SRUV), X-ray diffraction (XRD), and fourier transform infrared (FTIR) instruments. A batch experiment was carried out for oxidation of Pb(II), driven by a combination of visible light and TiO2-N,S photocatalyst. The research results attribute that co-doping N,S into TiO2 has remarkably narrowed the gap in the TiO2 structure, emerging in the visible region. It was also proven that the co-doped in TiO2 can considerably enhance its activity in the removal of Pb(II) under visible light, and the highest activity was owned by TiO2-N,S (1:1:1). Furthermore, the most effective removal of Pb(II) 10 mg/L (98%) could be reached by employing 500 mg L-1 of the TiO2-N,S (1:1:1) dose, 45 min of the time, and the solution pH at 7. The Pb(II) removed is due to the photo-oxidation induced by OH radicals to form the handleable PbO2.

Compared to metal elemental dopants, the non-metals are more interesting due to their smaller size allowing them to be inserted into the TiO 2 crystal lattice facilely [2,4,[9][10][11]. Furthermore, among the non-metal dopants, nitrogen has attracted extensive interest as it can be easily introduced into the TiO 2 structure, due to its atomic size, which is comparable with that of oxygen, its low ionization energy, and high stability [2][3][4]15]. In accordance, N-doped TiO 2 demonstrates significant photocatalytic activity under visible light irradiation [2][3][4]. It was also reported that increasing the dopant loaded has improved the photodegradation of organic pollutants, but the further increase was found to show the opposite effect [2,5,[8][9][10][11]15]. This unexpected trend is generated by the turning role of the dopant from preventing becomes servicing recombination [5,6,11,15]. Consequently, the fast recombination may proceed that declines the effectiveness of the photocatalytic organic degradation.
Considering the advantages of using N dopant, as presented above, the double elemental dopants of N combined with C [7,12], with P [11], as well as with S dopants [13][14][15][16] have been intensively studied. Due to the pronoun effect shown by double dopants of N and S in the improving photodegradation under visible light exposure, co-doping with N and S is growing of interest [13][14][15][16]. Several studies have investigated the codoping TiO 2 with N-S atoms for accelerating the degradation of residual antiinflammatory drug [13], antibiotic residual [14], p-chlorophenol [15], and phenol [16] in the presence of the visible light, and high results have been obtained.
To the best of our knowledge, the co-doped TiO 2 -N,S has not been examined for oxidation of hazardous heavy metals such as Pb (II). The heavy metal become an environmental concern due to its characters including wider disposal sources, rapid accumulative in biotic tissues, and high hazard for human health [17][18][19][20][21][22][23]. To prevent the dangerous effect of Pb(II) ion on the ecosystem and humans, an effective detoxification method is urgently required to treat the corresponding wastewater before reaching the environment. Removal of Pb(II) ions from water has been frequently performed by adsorption techniques [17][18][19]. Unfortunately, at the end of the process, the adsorbent saturated with Pb(II) can generate undesired hazardous solid waste, which creates a new problem in the environment. The most suitable method is believed to be oxidation of Pb(II) resulting from the less toxic and handleable PbO 2 [20][21][22][23]. The effective Pb(II) oxidation can be obtained through the photo-Fenton process [20] and photocatalysis over TiO 2 photocatalyst under UV illumination [20][21][22][23].
However, detoxification of Pb(II) by using N,S-coped TiO 2 in the presence of visible light so far is untraceable in the literature. The application of the N,S co-doped TiO 2 for photo-oxidation of Pb(II) in the solution contributes plausibly a novelty in Pb(II) remediation as well as the application of TiO 2 -N,S photocatalyst. Under the circumstance, in the present research, photooxidation of Pb(II) over TiO 2 -N,S photocatalyst under visible light irradiation is addressed.
Concerning the co-doping N,S into TiO 2 , urea, and thiourea are the most employed as the sources of the N and S dopants [2,3,9,10,13,14,16] and other organic compounds [2,11,12]. However, using organic amine for N and S dopant sources can inevitably lead to organic residues on the photocatalyst surface [4,5] that can decrease the activity of the photocatalyst. To avoid such weaknesses, in this paper, nitric and sulfuric acids are proposed as the simple inorganic N and S dopant sources respectively, into the TiO 2 structure. The co-doping is performed in one step suggesting a fast and practice process. Furthermore, to reach the maximum Pb(II) photo-oxidation result, the influences of some important parameters controlling the effectiveness of Pb(II) oxidation such as dopant amount in the photocatalyst, photocatalyst mass, irradiation time, and solution pH are also evaluated.

Co-doping of TiO 2 by N and S atoms
The co-doping TiO 2 with N and S elements was employed by a pressured hydrothermal technique in the autoclave [24]. Powder of TiO 2 about 0.8 g was suspended in 20 mL of a mixture of distilled water (10 mL) and ethanol (10 mL), followed by stirring for 30 min to make a good suspension. Into the suspension, 20 mL of HNO 3 5 M and 20 mL of H 2 SO 4 5 M were added, accompanied by stirring for 30 min to get a homogenous mixture. The mixture was then put in the autoclave and heated at a constant temperature of 150°C for 6 h. After that, the co-doped TiO 2 photocatalyst was collected and washed with deionized water three times to remove the weakly adsorbed species. Then, the photocatalyst was dried at 110°C and continued with calcination at 400°C for 2 h. The quantities of TiO 2 , HNO 3, and H 2 SO 4 introduced would give a mole ratio of Ti:N:S = 1:1:1. This as-prepared photocatalyst was coded as TiO 2 -N,S(1:1:1).
The same procedure was copied for processes by addition of HNO 3 and H 2 SO 4 with the same concentration (5M) but different volume, giving mole ratio of Ti:N:S = 1:1:1.5 and 1:1:0.5, as well as by using single HNO 3

Characterization
The effect of co-doping on the crystallinity of TiO 2 was followed by using an X-ray diffractometer of 6000X Shimadzu with Cu K radiation. The XRD patterns were recorded from 4-40 o of the 2 theta angles. The absorption edge and band gap energy of the samples were measured using a UV/visible spectrophotometer equipped with a Specular reflectance accessory of the UV-1800 series. The SRUV spectra were taken from 300-800 nm of the wavelength. Infrared spectra (IR) with the wavenumber of 4000-400 cm −1 were recorded on a Shimadzu Prestige 21 Infrared spectrophotometer that was used to evaluate the success of the N,S doping. The determination of Pb(II) concentration in the solution from the photooxidation process was carried out by using 3110 Perkin-Elmer flame Atomic Absorption Spectrophotometry elemental analysis.

Photocatalytic removal of Pb(II) from water under visible irradiation
In this typical process, 25 mg of the co-doped photocatalyst mixed with 50 mL of a solution containing Pb(II) 10 mg/L was stirred to get a homogenous mixture. The mixture was placed in a container glass, and the container was put in the photo-process apparatus equipped with 4 visible lamps (TL-D Intensity @20 W, 2000 lm/m 2 ). Then, the visible lamps were turned on, and the mixture was magnetically stirred for 30 min. After the desired time, the solution in the container was filtered under 41 Whatman filter paper to collect the solution containing the Pb(II) residue. The concentration of Pb(II) ions left in the solution was analyzed by using AAS based on the respective standard curve. The removed Pb(II) (in %) from the solution was calculated by the following equation: where Co is the initial amount of Pb(II) (mg) and C represents the amount of Pb(II) left in the solution (mg).
The same procedure proceeded with various mole ratios of Ti:N:S, photocatalyst masses, irradiation time, and solution pH, as well as by using mono-S-doped and Ndoped photocatalyst as a comparison.

Result and discussion
Characterization of The N,S Co-doped TiO 2

SRUV spectra
The SRUV spectra of the co-doped TiO 2 along with un-doped photocatalysts are exhibited in Fig. 1. By taking the intersection of the absorbance at the respective wavelength, the maximal absorbance can be obtained. From the maximal absorbance, the band gap energies of the photocatalyst can be calculated, and the calculation results are presented in Table 1.
Data in the Table 1 indicates that by co-doping, the absorption of the photocatalyst shifts significantly into a longer wavelength, entering the visible region. The shifts are due to the considerable decrease of the band gap energy, resulting from narrowing the gap, which is synergically created by the two dopant (N and S) atoms. Further, the decreasing band gap energy is observable in more effective when the amount of dopant S is enlarged giving larger narrowing the gap. This data provides evidence of the success . Similar patterns of the co-doped photocatalyst samples to the pattern of the undoped TiO 2 are also clearly observed. The difference that appeared in the intensities of the doped photocatalyst is lower than the intensities of the undoped one. With the increase of the amount of S in the co-doped photocatalyst, the intensities are seen to be considerably decreased. The lowering intensities imply the crystallinity deforms partially, due to the insertion of the atom dopants in the crystal lattice [2,10]. The more amount of S dopant produces, more S inserted into the lattice leading to higher deformation of the crystal. This alteration of the intensities is good proof of the success of the N,S codoping in TiO 2 . Other studies have also obtained the same trend [2,5,8,11].

FTIR data
Several characteristic peaks belonging to TiO 2 are notable in the FTIR spectra as seen in Fig. 3. It is assigned that with the small sulfuric acid concentration, the S doping is undetectable, and by introducing higher concentration doping N and S have been successfully proceeded [9,10,15].

Influence of co-doping
It is observable in Fig. 4 that co-doping promotes higher removal of Pb(II) under visible light compared to the undoped. Moreover, increasing the amount of S doped in co-doped TiO 2 has improved the removal, but the opposite effect is seen when the amount of S dopant is further enlarged. The improvement of the photo-oxidation in the presence of the co-doped under visible light illumination is promoted by narrowing the gap or reducing the Eg, allowing TiO 2 to be responsive by visible light. In addition, the dopant atoms can also service separation of the photogenerated electrons and hole by capturing the electrons [5,11,12,14], which detains the recombination. In contrast, when the S dopant is in excessive amount, the role of the dopant switches into the recombination center [5,6,11,15], which accelerates the recombination, and thus diminishes the Pb(II) removal. The Pb(II) ion removal from the aqueous media under visible light and in the presence of TiO 2 -N.S can be stimulated by 3 possible reactions. The first reaction is precipitation of Pb 2+ with -OH anionic into Pb(OH) 2 , written as reaction (4), that can only occur at pH higher than 7. The photocatalytic reaction was conducted at pH 5, suggesting that no precipitation of Pb(OH) 2 occurs.
The second possibility is the reduction of Pb 2+ induced by electrons released by TiO 2 during light irradiation into Pb 0 following reaction (5) [22]. Since the standard reduction potential is a negative value suggesting that the reduction is impossible thermodynamically [20,21].
The last possible reaction is oxidation by hole (h + ) with oxidation potential as much as 3.5 V, exhibiting a strong oxidizing agent. In addition to the hole, OH radical also acts as a strong oxidizing agent with 2.8 V of the oxidation potential. The oxidation of Pb 2+ with OH radicals to be Pb(IV)O 2 is represented by reactions (6) and (7).
It is known that the standard reduction potential of Pb(IV)/Pb(II) is −0.67 V [21], implying that the reduction is unlikely to proceed. Hence, it is obvious that photooxidation of Pb(II) is more favorable. The more possible oxidation of Pb(II) is in agreement with the study reported [23] that Pb(II) could be oxidized by chlorine and was accelerated by Mn(VII).

Influence of photocatalyst mass
The photocatalytic removal of Pb(II) is significantly controlled by photocatalyst mass, as displayed in Fig. 5. Notably, the effectiveness of the Pb(II) removal enhances sharply with the enlargement of the photocatalyst mass, but the oxidation is detrimental when the mass is in excess. With the higher photocatalyst mass, more OH radicals can be provided. Also, enhancing the photocatalyst mass can enlarge the active surface of the photocatalyst that improve the Pb(II) adsorption to be oxidized by OH radicals. The more number of OH radicals and Pb(II) adsorbed is beneficial to accelerate the photooxidation that reaches the maximum result.
The photocatalyst mass exceeding the optimum level (35 mg) can escalate significantly the turbidity, which hinders the penetration of the light. Consequently, the number of OH radicals formed may be declined and thus diminish photooxidation. A study also obtained similar results [10,15].
Moreover, the figure also demonstrates that co-doping results in higher oxidation compared to the mono-doped. It is evidence that co-doping can noticeably increase the visible responsive of TiO 2 and effectively prevent the recombination, due to the synergic effect of the double dopants [8,[11][12][13][14]. The higher photooxidation of Pb(II) shown by TiO 2 -N over that of TiO 2 -S implies that the effect of doping N is more effective than S dopant. This trend was also reported as well [6]. Influence of the irradiation time Figure 6 demonstrates a sharp incremental of the Pb(II) photooxidation as the extending time up to 30 min, but there is a negative effect with the longer time than 30 min. Prolonging time of the irradiation facilitates TiO 2 to provide more OH radicals and also allows OH radicals to contact with Pb 2+ effectively. With the longer than 30 min, the solid of PbO 2 resulting from the oxidation may be formed in a larger amount [9,10,15,20], which can cover the surface of the photocatalyst. In accordance, the formation of OH radicals is fewer, and thus, the photooxidation of Pb(II) declines. Some studies have also found a similar trend [9,10,15,20].

Influence of pH solution
The photooxidation effectiveness with the alteration pH can also be seen in Fig.  6. Increasing pH up to 7 exhibits an elevation of the effectiveness, but the contrary effect is seen at pH higher than 7. At lower pH, the surface of TiO 2 is protonated to be positive charge [8,10,11,21], presented as reaction (7), that prevents providing OH radicals, and Pb(II) exists prominently as Pb 2+ [19][20][21]. The same charges of the photocatalyst surface and Pb(II) ions restrict the adsorption of Pb 2+ on the surface of TiO 2 . The fewer OH radicals and low Pb 2+ adsorption explain the low photooxidation of Pb 2+ by OH radicals on the surface of the photocatalyst. At higher pH, the protonation should decrease or quiet, allowing TiO 2 to form OH radicals maximally and to adsorb Pb 2+ effectively, which is beneficial in the high photooxidation. When the pH is climbed up to basic condition (pH 9), the most surface of TiO 2 is in the negative forms, as seen in reaction 6 [11,21], that is to form OH radicals. In such pH Pb(II) precipitates as Pb(OH) 2 [19,21]. These conditions are certainly adverse for Pb 2+ photooxidation by OH radicals.

Conclusions
It is evidence that N and S atoms from HNO 3 and H 2 SO 4 , respectively, have been successfully co-doped into TiO 2 . The co-doping N,S into TiO 2 has significantly decreased the Eg from 3.2 eV into 2.53-2.95 eV depending on the dopant loaded and thereby noticeably improves the photocatalytic removal of Pb(II) under visible light. Furthermore, the effectiveness of Pb(II) photooxidation is found to be considerably directed by the fraction of dopants, and the optimum mole fraction of the dopants are Ti:N:S = 1:1:1. Additionally, by applying 0.5 g/L of the TiO 2 -N,S (1:1: 1) photocatalyst dose in 45 min of time and pH 7, the highest removal of 10 mg/L Pb(II) could be achieved that is approximately 98%. The removal of Pb(II) from the aqueous is due to the photocatalytic oxidation induced by OH radicals to form handleable PbO 2 .