1 Introduction

The discharge of tetracycline hydrochloride (TC-HCl) and other antibiotics into the aquatic environment has been widely recognized as a severe environmental problem [1,2,3]. The presence of antibiotics in water body could lead to long-term adverse impacts on ecosystems and human health [2, 4, 5]. Much attention has been paid to develop efficient methods such as adsorption [6, 7], electrolysis [8], advanced oxidation processes [9,10,11] and microbial degradation [12] to degrade the recalcitrant antibiotics. Advanced oxidation processes (AOPs) have been recognized as promising techniques for completely removing tetracycline hydrochloride from polluted water comparing to other wastewater treatments [13, 14].

Among different AOPs, photocatalysis has received much attention as a kind of energy saving and green technology [8, 15, 16]. Bismuth based semiconductors such as Bi2O3 [17, 18], Bi2MoO6 [19], Bi2WO6 [20], BiVO4 [21], Bi2GeO5 [22] and BiOX (X = Cl, Br, I) [23,24,25] are widely used in photocatalysis for removing toxic organic compounds. As a representative Bi-based photocatalyst, Bi2MoO6 could be excited by visible light because of its suitable band gap (2.4–2.8 eV) [26]. Moreover, its unique crystal structure is composed of altering (Bi2O2)2+ and (MoO6)2− layers, which was proved to be beneficial to charge transfer [27]. However, it is still hard for Bi2MoO6 to meet the standard of practical applications due to the fierce recombination rate of photo-excited electron–hole pairs. Up to now, several methods such as controllable synthesis [28], ion doping [29], self-doping [30] and constructing heterojunction [31] have been developed to promote the photocatalytic efficiency of Bi2MoO6.

In recent years, coupling of photocatalysis and electron acceptors has been proved to be an effective method to produce a rate-enhancing effect on degrading pollutants [32]. Compared to H2O2, persulfate (PS) is easier to be transported, applied and stored due to its solid state at room temperature [33, 34]. After trapping the electrons on the conduction band, PS is excited with the generation of highly efficient sulfate radical (SO ·−4 ) and hydroxyl radical (·OH). Compared to nonselective ·OH, SO ·−4 is more selective toward organic pollutants. Moreover, SO ·−4 has stronger oxidize activity than ·OH since the redox potential of SO ·−4 (2.5–3.1 V) is higher than that of ·OH (1.89–2.72 V) [35]. More recently, integrating photocatalyst, visible light and PS to degrade contaminants has attracted more and more interest from researchers [36,37,38]. However, to the best of our knowledge, the report focused on investigating the photocatalytic process of Bi2MoO6 with the assistant of PS is still limited.

Based on the previous studies, the object of our article was to investigate the potential application of Vis/Bi2MoO6/PS system in the degradation of TC-HCl. The effects of several important parameters on the removal process of TC-HCl were studied. A possible photocatalytic mechanism was proposed according to free radical scavenging experiments. This study provided a promising way to improve the photocatalytic performance of Bi2MoO6 by adding PS.

2 Experimental

2.1 Chemicals

All the reagents were used as-received directly without further purification. Tetracycline Hydrochloride (TC-HCl) was bought from Macklin Biochemical Technology Co., Ltd. (Shanghai, China). All other reagents (bismuth nitrate pentahydrate (Bi(NO3)3·5H2O), sodium molybdate dihydrate (Na2MoO4·2H2O), sodium persulfate (Na2S2O8), sodium hydroxide (NaOH), nitric acid (HNO3), absolute ethanol (C2H5OH), sodium chloride (NaCl), sodium sulfate (Na2SO4), potassium nitrate (KNO3), tertbutanol (TBA), phenol, ethylenediamine tetraacetic acid disodium salt (EDTA-2Na)) were obtained from Kelong Chemical Reagent Co., Ltd. (Chengdu, China) at analytical grade. Ultrapure water (18.25 MΩ cm at 25 °C) obtained from a Milli-Q system was used throughout experiments.

2.2 Synthesis of Bi2MoO6

Bi2MoO6 was prepared by a simple hydrothermal method according to the previous articles [39, 40]. At first, 2 mmol (0.97 g) Bi(NO3)3·5H2O and 1 mmol (0.24 g) Na2MoO4·2H2O were added into 80 mL ultrapure water. Subsequently, the mixture was vigorously stirred at room temperature for 30 min to form homogenous suspension. Then the resultant suspension was transferred to a 100 mL autoclave Teflon vessel and hydrothermally treated at 120 °C for 24 h. After the autoclave cooling down to room temperature naturally, the as-prepared powders were washed with ultrapure water and absolute ethanol for three times in turn and dried at 60 °C for 12 h.

2.3 Characterization of Bi2MoO6

The crystal characteristics of the as-prepared Bi2MoO6 were measured by powder X-ray diffraction (XRD, RU-200B/D/MAX-RB, Japan) with Cu Kα radiation over the two-theta region of 10°–60°. The surface element compositions were investigated using the X-ray photoelectron spectroscopy (XPS, ESCALAB 250Xi, USA) where C 1s located at 284.6 eV was used as reference standard. The morphological features of Bi2MoO6 were recorded on a field emission scanning electron microscopy (FESEM, Regulus8100, Japan). The UV–Vis absorption spectra and diffuse reflectance spectra (DRS) were analyzed by UV–Vis spectrophotometer (Lambda 750S, USA) using BaSO4 as reference standard.

2.4 Degradation experiments and analyses

The photocatalytic activities of Bi2MoO6 were evaluated by degrading TC-HCl under visible light irradiation. A 300 W Xe lamp (CEL-HXF300/CEL-HXUV-300) equipped with a 420 nm cut-off filter was used as a light source. The solution pH was not adjusted unless specially illustrated. A desired amount of Bi2MoO6 was introduced to 100 mL TC-HCl solution at first. The obtained mixture was vigorously stirred in the dark for 30 min to establish the adsorption/desorption equilibrium. After stirring, a certain amount of PS and other additives if necessary were added. Subsequently, the suspension was exposed to visible light to initiate the reaction. At designated time interval, 3 mL sample was extracted and centrifuged immediately to remove the catalyst. The residual concentration of TC-HCl was determined by measuring the absorbance at the maximum wavelength of 357 nm on a UV–Vis spectrophotometer (UVmini-1280, Shimadzu).

The degradation ratio (DR) was calculated via the following equation:

$${\text{DR}}(\% ) = (1 - {\text{C}}/{\text{C}}_{0} ) \times 100\%$$
(1)

where C0 is the initial concentration of TC-HCl at time 0, C represents the concentration of TC-HCl at a given period of time t.

The pseudo-first-order kinetic model was used to investigate the degradation performance of TC-HCl under different systems. The value of apparent kinetic constant (Kapp) was calculated according to the following formula [41]:

$$- \ln ({\text{C}}/{\text{C}}_{0} ) = {\text{K}}_{\text{app}} {\text{t}}$$
(2)

3 Results and discussion

3.1 Characterization of Bi2MoO6

The X-ray diffraction pattern of the as-prepared Bi2MoO6 was shown in Fig. 1. Well-defined peaks at 10.86°, 23.52°, 28.26°, 32.54°, 33.06°, 36.04°, 46.78°, 47.16°, 55.56°, 56.22°, 58.56° were respectively responding to the (020), (111), (131), (200), (060), (151), (202), (260), (331), (280), (262) planes of orthorhombic Bi2MoO6 phase (JCPDS card no. 21-0102) [40]. No other obvious peaks could be found, indicating the high purity of Bi2MoO6.

Fig. 1
figure 1

XRD pattern of the as-prepared Bi2MoO6 photocatalyst

Full size image

XPS analysis was carried out to investigate the chemical compositions and states of surface elements. As shown in Fig. 2a, signals of Bi, Mo, O and C were found in the survey spectrum of Bi2MoO6, where C came from the XPS instrument. The Bi 4f core-level spectra demonstrated a binding energy at 164.3 eV for Bi 4f5/2 and at 159.0 eV for Bi 4f7/2, revealing that Bi was in trivalent oxidation state [42]. For the spectrum in the Mo 3d region, two sharp peaks located at 235.4 eV and 232.3 eV were respectively ascribed to the Mo 3d3/2 and Mo 3d5/2, which corresponded to the + 6 oxidation state of Mo [43]. The detailed O 1 s spectra was shown in Fig. 2d. The peak at 529.9 eV could be assigned to the lattice oxygen [44].

Fig. 2
figure 2

XPS of Bi2MoO6: a full spectra of sample; b Bi 4f peaks; c Mo 3d peaks; d O1s peaks

Full size image

The microstructure and morphology of Bi2MoO6 were characterized by FESEM technique.  Figure 3 presented the representative FESEM images of Bi2MoO6. It was clearly that Bi2MoO6 was composed of many irregular nanosheets with smooth surface. With respect to the high-resolution image (Fig. 3b), the nanosheets were quite thin and its width ranged from 500 nm to 2 μm. Moreover, the nanosheets were stacked closely.

Fig. 3
figure 3

FESEM images of Bi2MoO6

Full size image

The UV–Vis DRS in the range of 200–800 nm was utilized to determine the optical properties of Bi2MoO6. As shown in Fig. 4a, it was obvious that the photocatalyst had a strong absorption capacity in visible light region. The corresponding band gap energy (Eg) of Bi2MoO6 could be calculated using the following Kubelka–Munk equation [45]:

$$\alpha {\text{h}}\nu = {\text{A}}({\text{h}}\nu - {\text{E}}_{\text{g}} )^{{{\text{n}}/2}}$$
(3)

where \(\alpha ,{\text{h}},\nu ,\) A are the absorption coefficient, Planck constant, light frequency and the proportionality constant, respectively. The value of n depending on transition nature is 1 due to the direct band-gap characteristic of Bi2MoO6 [45]. Therefore, the equation can be expressed as another form as follows:

$$(\alpha {\text{h}}\nu /{\text{A}})^{2} = h\nu - {\text{E}}_{\text{g}}$$
(4)

As illustrated in Fig. 4b, the band gap energy of Bi2MoO6 estimated from the plot of (αhν)2 versus photo energy (hν) was 2.58 eV. The potentials of conduction band (CB) and valence band (VB) are generally determined according to the following empirical equations [46]:

$${\text{E}}_{\text{CB}} = {\text{X}} - {\text{E}}_{0} - \frac{1}{2}{\text{E}}_{\text{g}}$$
(5)
$${\text{E}}_{\text{VB}} = {\text{E}}_{\text{CB}} + {\text{E}}_{\text{g}}$$
(6)

where ECB is the edge potential of CB, the value of X which represents the electronegativity of the semiconductor is 5.5 eV for Bi2MoO6 [47], E0 is the energy of free electrons on the hydrogen scale (about 4.5 eV), Eg is the band gap energy of the semiconductor (2.58 eV, from Eq. (4)), EVB is the VB edge potential.

Fig. 4
figure 4

a UV–Vis DRS spectra of Bi2MoO6, b (αhv)2 versus photon energy (hv) of Bi2MoO6

Full size image

According to the above formulas, the EVB and ECB values of Bi2MoO6 were calculated to be 2.29 eV and − 0.29 eV, respectively.

3.2 Degradation of TC-HCl under different systems

Preliminary experiments were conducted to evaluate the degradation efficiency of TC-HCl under various systems and the results were shown in Fig. 5. No obvious degradation of TC-HCl was observed in the case of visible light irradiation, revealing that TC-HCl was quite stable under visible light. Only 4% of TC-HCl was degraded by 4 g/L PS, indicating that it was difficult for PS to directly degrade TC-HCl. Interestingly, the degradation efficiency of TC-HCl increased to 26.9% when combing visible light with 4 g/L PS. Adsorption capacity was investigated using bare Bi2MoO6 as catalyst. Approximately 20% of TC-HCl was adsorbed within 5 min but no more TC-HCl was further adsorbed. This phenomenon could be attributed to the achievement of adsorption/desorption equilibrium. Compared to PS alone, only 13% improvement of TC-HCl decay was observed in the Bi2MoO6/PS system, which suggested that PS was difficult to be excited by Bi2MoO6 in the dark. Resulting from the efficient photocatalytic activity of Bi2MoO6, about 48.2% of TC-HCl decay was achieved in Vis/Bi2MoO6 system. Compared to other systems, it was obvious that the Vis/Bi2MoO6/PS process exhibited the optimum performance toward TC-HCl degradation and almost 83% of TC-HCl was degraded within 60 min.

Fig. 5
figure 5

Time-dependent TC-HCl concentration under different systems (conditions: Bi2MoO6 loading = 0.5 g/L; initial TC-HCl concentration = 20 ppm; initial PS concentration = 4 g/L; unbuffered initial pH ~ 4.4)

Full size image

As shown in Table 1, TC-HCl degradation obeyed the two-stage pseudo first-order kinetics. The degradation process could be divided into the first (sharp) stage (t < 10 min) and the second (gradual) stage (t > 10 min). According to the previous literatures, the rate constant calculated in the first stage was commonly used to evaluate whether the combination of Vis/oxidant and Vis/catalyst was synergistic [32, 48, 49]. Table 2 presented the relative kinetic parameters for TC-HCl degradation in the first stage under various systems. A synergy index (ratio of the degradation rate obtained with Vis/Bi2MoO6/PS system and the sum of degradation rates attained with Vis/PS and Vis/Bi2MoO6 systems) of 5.13 (≫ 1) was obtained, indicating the combination of Vis/PS and Vis/Bi2MoO6 was synergistic [49].

Table 1 The apparent rate constant (kapp) and correlation coefficient (R2) for the degradation of TC at different reaction time by the Vis/Bi2MoO6/PS system
Full size table
Table 2 The apparent rate constant (kapp) and correlation coefficient (R2) for the degradation of TC in the first stage under different systems
Full size table

3.3 Effects of parameters on TC-HCl degradation in Vis/Bi2MoO6/PS system

To investigate the effect of catalyst loading on the photocatalytic degradation of TC-HCl by the Vis/Bi2MoO6/PS system, experiments were carried out by varying the catalyst dosages (0 to 0.8 g/L). As shown in Fig. 6, removal efficiency was markedly accelerated with an increase of catalyst loading (0 to 0.5 g/L). This phenomenon might be attributed to the increase of available active sites for PS decomposition. Nevertheless, a slight decrease of TC-HCl decay was observed with increasing the Bi2MoO6 loading from 0.5 to 0.8 g/L. Previous studies have reported that the existence of excessive catalyst can lead to the low rate of visible light penetration because of turbid solution caused by the formation of aggregates and consequent self-bonding of catalyst particles [32]. The catalyst dosage of 0.5 g/L with maximum removal efficiency of TC-HCl was applied in the following experiments.

Fig. 6
figure 6

Effect of Bi2MoO6 loading on the removal efficiency of TC-HCl with the Vis/Bi2MoO6/PS system (conditions: initial TC-HCl concentration = 20 ppm; initial PS concentration = 4 g/L; unbuffered initial pH ~ 4.4)

Full size image

The degradation performance of Vis/Bi2MoO6/PS system at various initial TC-HCl concentrations was investigated and the results were presented in Fig. 7. It can be seen that the removal efficiency of TC-HCl decreased significantly with increasing the TC-HCl concentration from 20 to 120 ppm. At an initial TC-HCl concentration of 20 ppm, approximately 83% removal was achieved within 60 min. Nevertheless, the degradation rate decreased to 66.8% for 40 ppm TC-HCl, 51.1% for 80 ppm TC-HCl and 38.6% for 120 ppm TC-HCl. At high substrate concentrations, the interaction of catalyst surface and visible light was restricted because more photos would be absorbed by TC-HCl molecules or the formed by-products. Another reason might be attributed to the fact that a large number of by-products would compete with TC-HCl molecules to react with generated radicals [49].

Fig. 7
figure 7

Effect of TC-HCl concentration on the removal efficiency of TC-HCl with the Vis/Bi2MoO6/PS system (conditions: Bi2MoO6 loading = 0.5 g/L; initial PS concentration = 4 g/L; unbuffered initial pH ~ 4.4)

Full size image

To evaluate the influence of PS concentration on the degradation rate of TC-HCl, a series of experiments were carried out at different concentrations of PS (0 to 10 g/L). The obtained results were shown in Fig. 8. In the absence of PS, 48.2% extent of TC-HCl was removed after 60 min. The degradation rate of TC-HCl were 63.4%, 70.7%, 83% for the concentration of PS at 0.2 g/L, 2 g/L, 4 g/L, respectively. Obviously, the degradation efficiency of TC-HCl increased sharply with increasing the concentration of PS from 0 to 4 g/L. This was because more reactive radicals would generate on the surface of Bi2MoO6 with increasing the amount of PS which worked as the donor of sulfate and hydroxyl radicals. On the contrary, there was no promotion of degradation rate when the PS concentration further increased to 10 g/L. Excess amounts of PS could not be activated efficiently because limited active sites could be provided for PS activation at fixed Bi2MoO6 loading and illumination intensity. Additionally, the radical–radical recombination parasite reactions were suspected to be occurred at high concentration of radicals. Both SO ·−4 and ·OH were consumed leading to the generation of less active radicals including S2O ·−8 and HSO ·−4 (Eqs. (7) – (10)) [50, 51]. Compared to other concentration of PS, 4 g/L which performed the quite excellent degradation efficiency was applied in the following study.

$${\text{SO}}_{ 4}^{ \cdot - } + {\text{ S}}_{ 2} {\text{O}}_{ 8}^{ 2- } \to {\text{SO}}_{ 4}^{ 2- } + {\text{ S}}_{ 2} {\text{O}}_{ 8}^{ \cdot - }$$
(7)
$${\text{SO}}_{4}^{ \cdot - } + {\text{SO}}_{4}^{ \cdot - } \to {\text{S}}_{2} {\text{O}}_{8}^{2 - }$$
(8)
$${\text{S}}_{2} {\text{O}}_{8}^{2 - } + \cdot {\text{OH}} \to {\text{OH}}^{ - } + {\text{S}}_{2} {\text{O}}_{8}^{ \cdot - }$$
(9)
$${\text{SO}}_{4}^{ \cdot - } + \cdot {\text{OH}} \to {\text{HSO}}_{4}^{ - } + 1/2{\text{O}}_{2}$$
(10)

It was necessary to elucidate how initial solution pH influence the oxidation process of TC-HCl since solution pH played an important role in PS activation [34]. The solution pH was adjusted to five different pH values of 2.1, 3.4, 4.0, 6.1, 10.2 by 1 M NaOH and 1 M HNO3. The initial solution pH without adjustment was 4.4. The results were presented in Fig. 9. The degradation of TC-HCl at three different pH values from 3.4 to 6.1 had no obvious difference from that of pH 4.4 at basic conditions. However, when the degradation process was conducted at strange acid medium (pH 2.1), the removal efficiency of TC-HCl declined rapidly with a final removal of 69.5%. This phenomenon might be due to the reaction between H+ and active radicals (Eqs. (11) and (12)) [52]. At a pH of 10.2, the TC-HCl degradation slightly decreased to 78.8%. This decrease was attributed to the capture activity of OH− according to Eq. (13) [53]. In conclusion, the Vis/Bi2MoO6/PS system could be applied at a wide range of pH to degrade TC-HCl and the pH without adjustment is also feasible in the treatment of TC-HCl.

$$\cdot {\text{OH}} + {\text{H}}^{ + } + {\text{e}}^{ - } \to {\text{H}}_{2} {\text{O}}$$
(11)
$${\text{SO}}_{4}^{ \cdot - } + {\text{H}}^{ + } + {\text{e}}^{ - } \to {\text{HSO}}_{4}^{ \bullet - }$$
(12)
$${\text{SO}}_{4}^{ \cdot - } + {\text{OH}}^{ - } \to \cdot {\text{OH}} + {\text{SO}}_{4}^{2 - }$$
(13)
Fig. 8
figure 8

Effect of PS concentration on the removal efficiency of TC-HCl with the Vis/Bi2MoO6/PS system (conditions: Bi2MoO6 loading = 0.5 g/L; initial TC-HCl concentration = 20 ppm; unbuffered initial pH ~ 4.4)

Full size image
Fig. 9
figure 9

Effect of solution pH on the removal efficiency of TC-HCl with the Vis/Bi2MoO6/PS system (conditions: Bi2MoO6 loading = 0.5 g/L; initial TC-HCl concentration = 20 ppm; initial PS concentration = 4 g/L)

Full size image

In the actual water matrix, the occurrence of inorganic anions will influence the wastewater conditions even the fate of pollutants through electron exchange [53]. Figure 10 illustrated the effects of three inorganic ions including Cl−, SO42− and NO3− at a fixed concentration of 20 mM on TC-HCl degradation. Near 83% extent of TC-HCl was degraded within 60 min in the absence of anions. Nevertheless, the degradation efficiency of TC-HCl decreased to 73% in the case of 20 mM Cl− addition. The negative effect of Cl− resulted from the phenomenon that the existence of Cl− will transform SO ·−4 and ·OH to selective radicals which were not favorable for TC degradation (Eqs. (14)– (16)) [54]. The presence of 20 mM SO42− decreased the degradation efficiency of TC-HCl from 83% to 72%. The possible reactions involved in the transformation process of radicals with the addition of SO42− were listed in Eqs. (17) and (18) [55]. A negligible influence of 20 mM NO3− on the degradation efficiency of TC-HCl was observed.

$${\text{SO}}_{4}^{ \cdot - } + {\text{Cl}}^{ - } \to {\text{SO}}_{4}^{2 - } + {\text{Cl}} \cdot$$
(14)
$${\text{Cl}}^{ \cdot } + {\text{Cl}}^{ - } \to {\text{Cl}}_{2}^{ \cdot - }$$
(15)
$${\text{Cl}}^{ - } + \cdot {\text{OH}} \to {\text{ClOH}}^{ \cdot - }$$
(16)
$${^{ \cdot } {\text{OH}}} + {\text{SO}}_{4}^{{2 - }} \to {\text{SO}}_{4}^{{ \cdot - }} + {\text{OH}}^{ - }$$
(17)
$${\text{SO}}_{4}^{ \cdot - } + {\text{SO}}_{4}^{2 - } \to {\text{S}}_{2} {\text{O}}_{8}^{2 - } + {\text{e}}^{ - }$$
(18)
Fig. 10
figure 10

Effect of coexisting anions on the removal efficiency of TC-HCl with the Vis/Bi2MoO6/PS system (conditions: Bi2MoO6 loading = 0.5 g/L; initial TC-HCl concentration = 20 ppm; initial PS concentration = 4 g/L; unbuffered initial pH ~ 4.4)

Full size image

3.4 Reusability and stability of Bi2MoO6 in Vis/Bi2MoO6/PS system

Recycle experiments were carried out to investigate the reusability and stability of Bi2MoO6 under optimum conditions (0.5 g/L Bi2MoO6; 4 g/L PS; initial pH 4.4 and 20 ppm TC, from 3.3 section). After each degradation cycle, the used samples were separated immediately from TC-HCl solution by centrifugation. Subsequently, the catalyst was re-dispersed into fresh TC-HCl solution and the photocatalytic reactions were re-initialed under the same conditions compared to the first run. Figure 11 compared the degradation results of repeated experiments. About 83% of TC-HCl was removed within 60 min in the first run while 74.5% TC-HCl removal efficiency was achieved in the third run, which indicated that the photocatalytic efficiency of TC-HCl showed acceptable decrease even after three cycles. The recycle experiments implied that Bi2MoO6 was an excellent and stable catalyst in the Vis/Bi2MoO6/PS system.

Fig. 11
figure 11

Stability of Bi2MoO6 in Vis/Bi2MoO6/PS system (conditions: Bi2MoO6 loading = 0.5 g/L; initial TC-HCl concentration = 20 ppm; initial PS concentration = 4 g/L; unbuffered initial pH ~ 4.4)

Full size image

3.5 Mechanism for TC degradation in Vis/Bi2MoO6/PS system

In order to detect the main reactive radicals involved in the degradation system, several quenching tests were carried out. TBA was used as the scavenger of ·OH due to its much lower rate constant with SO ·−4 (4.0 × 105 M−1 S−1) compared to ·OH (6.0 × 108 M−1 S−1) [56]. Phenol which reacts with SO ·−4 and ·OH at high and comparable rate constants (6.6 × 109 M−1 S−1 and 8.8 × 109 M−1 S−1 for SO ·−4 and ·OH, respectively) was employed to quench SO ·−4 and ·OH [57]. The addition of ethylenediamine tetraacetic acid disodium salt (EDTA-2Na) and Nitrogen (N2) were applied to trap h+ and O ·−2 , respectively [58].

As illustrated in Fig. 12, when 1 mM TBA was added into the system, the degradation rate of TC-HCl decreased from 83 to 72.8%. In the presence of 1 mM phenol, the photocatalytic activity was inhibited significantly and only 35% of TC-HCl was removed after 60 min. The different inhibition effect of TBA and phenol suggested that SO ·−4 played a much more indispensable role than ·OH in the degradation process of TC-HCl [54]. The degradation efficiency of TC-HCl rapidly decreased to 11.6% with the addition of 0.5 mM EDTA-2Na, which resulted from the strong oxidization capacity of h+ [58]. There was about 4% loss in removal efficiency of TC under N2 saturation, indicating that O ·−2 also participated in the degradation process but played a negligible role.

Fig. 12
figure 12

Effect of radical scavengers on the removal efficiency of TC-HCl with the Vis/Bi2MoO6/PS system (conditions: Bi2MoO6 loading = 0.5 g/L; initial TC-HCl concentration = 20 ppm; initial PS concentration = 4 g/L; unbuffered initial pH ~ 4.4)

Full size image

According to these trapping experiments, a possible mechanism was proposed to explain the enhanced photocatalytic performance toward the degradation of TC-HCl by Vis/Bi2MoO6/PS system. As shown in Fig. 13, under visible light illumination, the catalyst was excited with the generation of electron–hole pairs (Eq. (19)). PS activation was achieved by the reaction between free electrons on the conduction band and PS anions (Eq. (20)) [34]. As electrons were partly consumed by PS, the recombination rate of photo-induced electron/hole pairs was decelerated. Furthermore, SO ·−4 would react with H2O/OH− resulting in the production of ·OH (Eqs. (21)– (22)) [59]. ·OH cannot be generated through sole photocatalysis since the potential of h+ on the valence band of Bi2MoO6 (2.29 eV) was more negative than the potential of H2O/·OH (2.4 eV) [26]. The dissolved oxygen (O2) was difficult to be reduced by the photo-induced e− to O ·−2 because the reduction potential of conduction band for Bi2MoO6 (− 0.29 eV) was a little lower than the potential of O2/O ·−2 (− 0.33 eV) [60]. h+ on the conduction band could directly react with organic compounds [26].

$${\text{Bi}}_{2} {\text{MoO}}_{6} + {\text{h}}\nu \to {\text{h}}^{ + } + {\text{e}}^{ - }$$
(19)
$${\text{S}}_{2} {\text{O}}_{8}^{2 - } + {\text{e}}^{ - } \to {\text{SO}}_{4}^{ \cdot - } + {\text{SO}}_{4}^{2 - }$$
(20)
$${\text{SO}}_{4}^{ \cdot - } + {\text{H}}_{2} {\text{O}} \to \cdot {\text{OH}} + {\text{SO}}_{4}^{2 - } + {\text{H}}^{ + }$$
(21)
$${\text{SO}}_{4}^{ \cdot - } + {\text{OH}}^{ - } \to \cdot {\text{OH}} + {\text{SO}}_{4}^{2 - }$$
(22)
$${\text{SO}}_{4}^{{ \cdot - }} {\text{/}}^{ \cdot } {\text{OH/h}}^{ + } + {\text{TC - HCl}} \to {\text{products}}$$
(23)

4 Conclusions

In our study, the ternary Vis/Bi2MoO6/PS system was proved to be more efficient than unary or binary systems toward TC-HCl removal in aqueous phase. Increasing Bi2MoO6 dosage or PS concentration in a range would rapidly accelerate the degradation process. Higher TC-HCl concentrations were adverse to the process of TC-HCl removal. The novel system demonstrated ideal removal efficiency at a wide range of initial solution pH. The existence of inorganic anions such as Cl− and SO42− had a negative effect on the TC-HCl removal efficiency while NO3− did not affect it. The catalyst exhibited excellent reusability and stability in multiple runs. The results of trapping experiments indicated that SO ·−4 together with h+ were mainly responsible for the degradation process. A suitable mechanism was proposed to explain the rate-enhancing effect of Vis/Bi2MoO6/PS system. After the catalyst was activated by visible light, PS would react with e− which resulted in the generation of SO ·−4 and ·OH. Meanwhile, the recombination rate of electron/hole pairs was decelerated since part of e− was consumed by PS. In summary, degrading organic pollutants using Vis/Bi2MoO6/PS system is a promising technology.

Fig. 13
figure 13

The schematic diagram of Vis/Bi2MoO6/PS phtotocatalytic system

Full size image