Introduction

With the increasing awareness of people’s environmental protection and increasing attention to health, drug residues in the environment have received increasing attention from researchers as well as the public. With the increasing use of antibiotics, high antibiotic concentrations have been detected in wastewater, as well as on the surface waters, drinking water, soil, and other environments now (Huang et al. 2019; Sanganyado and Gwenzi 2019; Cerqueira et al. 2019; Tong et al. 2019). In addition to ecotoxicity, they may cause increased bacterial resistance (Zhang et al. 2015; Hu et al. 2007). It has been reported that the proportion of drug-resistant bacteria in various poultry and livestock has also increased greatly in recent years (Van Boeckel et al. 2019).

Fluoroquinolones (FQs) have been widely used to treat a broad spectrum of bacterial infections over the past few decades (Zhang et al. 2015; Liu et al. 2016; Zhao et al. 2016). Ofloxacin (OFX), a popular FQ antibiotic, is a third-generation quinolone, which is mainly used in the treatment of acute and chronic infections of the respiratory tract, throat, and tonsils caused by gram-negative bacteria (Watanabe et al. 2001). OFX is discharged in non-metabolic form mainly through wastewater produced by the industries such as pharmaceuticals and waste generated by humans and livestock, enter into municipal wastewater treatment plants through sewer networks, and finally enter into the natural environment (Zhang et al. 2015; Jara et al. 2020; Radjenovic et al. 2007). However, existing sewage treatment facilities have low processing efficiency with regard to antibiotics (Jia et al. 2012). According to reports, various antibiotic residues have been detected in sewage treatment plants around the world, such as the average content of OFX in the effluent of sewage treatment plants has reached an astonishing 97.0 ng/L (Wang et al. 2020a). Hence, finding an effective antibiotics degradation method is essential.

Various technologies have been developed for the removal of various antibiotics from the water now, such as adsorption, advanced oxidation processes (AOPs), biodegradation/photodegradation/sonic degradation, and membrane separation (Peng et al. 2012; Jiang et al. 2016; Gorito et al. 2018; Elmolla and Chaudhuri 2010; Wang and Wang 2019a; 2018a). However, they have not been widely applied due to their low removal efficiency and high operational cost. By contrast, advanced oxidation technology mainly generates highly active oxidation species through a series of chemical oxidation processes for the degradation of antibiotics or converts them to small molecule substances, which could enhance their biodegradability and the removal rate (Zhou et al. 2021; Sharma et al. 2016; Wilde et al. 2013; Wang and Wang 2019b; 2018b). AOPs use strong oxidation agents to degrade organic pollutants. According to the different ways used to produce oxidation agents, AOPs can be classified into many different types (Wang and Zhuan 2020; Wang and Chen 2020). Among them, Fe(VI) advanced oxidation is a promising technology for the degradation of organic pollutants in wastewater (Pi et al. 2021; Shin et al. 2018; Wang et al. 2020b). It is characterized by rapid degradation of pollutants and high pollutant removal efficiency; hence, it can be regarded as green technology and has been widely used in wastewater treatment (Gong et al. 2020; Li et al. 2005).

In this work, K2FeO4 (Fe(VI)) was selected as the oxidant and OFX was selected as the target contaminant. The degradation efficiency and kinetics of OFX by Fe(VI) under different reaction conditions were investigated. In addition, on the basis of the intermediate products identified, a possible degradation pathway of OFX by Fe(VI) was proposed. This research provides theoretical support for the treatment of OFX-containing wastewater using an Fe(VI) advanced oxidation process.

Materials and methods

Reagents

All standards of the pharmaceuticals used (sodium hydroxide (NaOH), hydrochloric acid (HCl), sodium thiosulfate (Na2S2O3), ammonium acetate (CH3COONH4), phosphoric acid (H3PO4), tert-butanol (TBA), isopropanol (IPA), sodium bicarbonate (NaHCO3), sodium nitrate (NaNO3), sodium sulfate (Na2SO4), sodium chloride (NaCl), ammonium chloride (NH4Cl), ferric chloride (FeCl3), potassium chloride (KCl), and humic acid (HA)) were of analytical purity grade and purchased from Sinopharm Chemical Reagents (Shanghai, China). The solvent (methanol) was HPLC grade and was purchased from Merck (Darmstadt, Germany). OFX was obtained from Aladdin (Shanghai, China). The water was produced using a UPH purification system (ULUP-I, Ulpure, China, resistivity ≥ 18.2 MΩcm).

Experimental procedures

Fe(VI) was prepared using a hypochlorite oxidation method (Li et al. 2005). The reaction was carried out in a 250 mL glass conical flask under a constant stirring rate of 400 rpm at 25℃ in a water bath using a thermostatic magnetic stirrer. The reaction between OFX (8.0 μM) and Fe(VI) (40.0–160.0 μM) was initiated by mixing them in equal volumes of 100.0 mL; the pH of the reaction mixtures was adjusted using 1.0 mM HCl or NaOH buffer. The reaction was quenched completely at certain reaction times (i.e., 0–5 min) using 200.0 μL of 1.0 mM Na2S2O3 solution. Samples were filtered using 0.45 μm polytetrafluoroethylene syringe filters (Fisherbrand, Fisher Scientific) and transferred into 2.0 mL high-performance liquid chromatography (HPLC) vials for analysis. All experiments took an average of three parallel samples.

Analytical methods

The concentrations of OFX in the samples were measured by HPLC (LC-2030PLUS, Shimadzu, Japan) equipped with a SinChrom ODS-B column (5 μm, 4.6 mm × 250 mm), and the detection was performed using a G1365MWD UV detector at 293 nm. The mobile phase was composed of 60% ammonium acetate (1% phosphoric acid adjusted to pH 2.7) and 40% methanol. Samples were analyzed at a flow rate of 1.0 mL/min, the injection volume was set to 20 μL, and the column temperature was set at 25℃.

The intermediates of OFX were measured using a quadrupole time-of-flight (Q-TOF) liquid chromatography/mass spectrometry (LC/MS) (6545Q-TOF, Agilent, USA). Isocratic elution was performed at a flow rate of 0.3 mL/min with 0.1% (v/v) ammonium acetate (A) and methanol (B). The mass spectra with electrospray ionization (ESI) source were recorded across the range of 50–500 m/z in positive scan mode.

Results and discussion

Influence of initial oxidant concentration and degradation kinetics

The effect of the initial Fe(VI) concentration on OFX degradation is shown in Fig. 1a. It can be observed that the pseudo-first-order rate constants for OFX degradation (Kobs) increased from 0.0031 to 0.0125 s−1 when the Fe(VI) concentration increased from 40 to 160 μM. Then, Ln[Fe(VI)]0 and LnKobs were fitted by linear regression, and the results are shown in Fig. 1b. The correlation coefficient R2 was greater than 0.99, which means that the degradation of OFX by Fe(VI) follows the second-order reaction kinetics equation. The rate equations under different initial concentrations of Fe(VI) are shown in Table 1.

Fig. 1
figure 1

Changes in OFX as a function of initial Fe(VI) concentrations. (a) Degradation of OFX under different initial concentrations of Fe(VI). (b) Linear fitting of Ln[Fe(VI)]0 and LnKobs at different Fe(VI) concentrations. Reaction conditions: [OFX]0 = 8 μM, [Fe(VI)]0 = 40–160 μM, T = 25℃, pH = 7.0

Full size image
Table 1 Kinetic equation of the degradation of OFX by Fe(VI)
Full size table

As shown in Table 1, the average value of the secondary reaction rate constant (Kapp) for the degradation of OFX by Fe(VI) was 78.28 M−1 s−1. As the initial concentration of Fe(VI) increased from 40 to 160 μM, the half-life t1/2 decreased from 224 to 55 s. The results showed that with an increase in the initial concentration of [Fe(VI)], the degradation rate of OFX gradually increased; similar results have been reported by Wang et al. (2019) and Liu et al. (2019).

Influence of pH and the corresponding kinetics

The degradation of OFX by Fe(VI) under different pH conditions was measured, and the results are shown in Fig. 2a. When the pH increased from 5 to 10, the Kobs decreased from 0.0168 to 0.0039 s, implying that the degradation rate of OFX by Fe(VI) increased with a decrease in the pH value; similar results have been reported by Zheng et al. (2020) and Han et al. (2018).

Fig. 2
figure 2

Changes in OFX as a function of pH. (a) Degradation of OFX at different pH. (b) Relationship between pH value and apparent second-order reaction kinetic constant (Kapp). Reaction conditions: [OFX]0 = 8 μM, [Fe(VI)]0 = 120 μM, T = 25℃, pH = 5–10

Full size image

Because OFX and Fe(VI) have various ionisable groups, their final ionization morphologies can be determined based on different pH values (Huang et al. 2017; Peterson et al. 2015). Therefore, the kinetic equations of the pH and Kapp values were simulated using Eq. (1)

$${K}_{app}\left[Fe\left(VI\right)\right]\left[OFX\right]={\sum }_{i=\mathrm{1,2},3,j=\mathrm{1,2}}{K}_{ij}{a}_{i}{\beta }_{j}\left[Fe\left(VI\right)\right]\left[OFX\right]$$
(1)

where αi and βj represent the distribution coefficients of OFX and Fe(VI), respectively. When the OFX dissociation constants pKa1 and pKa2 are 6.11 and 8.18, respectively, α1, α2, and α3 refer to the fractions of OFX+, OFX0, and OFX−, respectively (Peterson et al. 2015). For Fe(VI) (pKa = 7.23, pH ≥ 5 in the experiment), β1 and β2 refer to the fractions of HFeO4− and FeO42−, respectively (Han et al. 2018; Zajicek et al. 2015). Under alkaline conditions, the reaction between FeO42− and OFX− was very slow (Zheng et al. 2020; Han et al. 2018; Huang et al. 2017); therefore, the reaction could be ignored in the model calculation. In addition, because it was difficult for HFeO4− and OFX− (or FeO42− and OFX+) to exist at the same time under certain pH conditions, their reaction could also be ignored in the model calculation; therefore, the effect of pH on the Kapp value of Fe(VI) can be further simplified as Eq. (2):

$${K}_{app}={K}_{11}{a}_{1}{\beta }_{1}+{K}_{21}{a}_{2}{\beta }_{1}+{K}_{22}{a}_{2}{\beta }_{2}$$
(2)

The experimental values of Kapp (Meas Kapp) under different pH conditions were fitted with Eq. (2), and the results are shown in Fig. 2b; it can be seen that they exhibited a significant curve correlation (R2 ≥ 0.99).

The results showed that K11, K21, and K22 were 150.23 M−1S−1, 94.7 M−1S−1, and 45.1 M−1S−1, respectively. Among them, K11 or K21 was > K22, indicating that HFeO4− was more oxidizing than FeO42−, and with the gradual decrease in pH (when PH ≥ 5), HFeO4− would become more dominant in the system. Moreover, K11 > K21 or K22 implied that HFeO4− showed higher reactivity toward OFX+ than OFX0 or OFX−. In conclusion, the degradation rate of OFX gradually increased with a continuous decrease in pH.

Influence of temperature and the corresponding kinetics

The degradation of OFX was measured as the temperature increased from 10 to 30℃; Kobs correspondingly increased from 0.0055 to 0.0110 s (Fig. 3a). This illustrates that the degradation rate of OFX gradually increases with an increase in temperature (Han et al. 2018). Figure 3b shows the results of fitting LnKapp and 1000/T, and a good linear relationship (R2 ≥ 0.99) can be observed. This shows that the degradation of OFX with temperature change satisfies the Arrhenius equation as follows (Luo et al. 2015; Han et al. 2018):

Fig. 3
figure 3

Changes in OFX as a function of temperature. (a) Degradation of OFX at different temperatures. (b) Linear fitting of LnKapp and 1000/T. (c) Linear fitting of Ln(Kapp/T) and 1/T. Reaction conditions: [OFX]0 = 8 μM, [Fe(VI)]0 = 120 μM, pH = 7.0; T = 10–30℃

Full size image
$${LnK}_{app}=-\left(Ea/R\right)\times \left(1/T\right)+LnA$$
(3)

Here, the molar gas constant R is 8.314 J·mol−1·k−1, and hence, the reaction activation energy Ea can be calculated to be 24.2 kJ·mol−1. The results suggest that the reaction between Fe(VI) and OFX could occur even when the activation energy was relatively low.

In Fig. 3c, the results of fitting Ln(Kapp/T) and 1/T are shown; a good linear relationship (R2 ≥ 0.99) is observed. This implies that the degradation of OFX with temperature change satisfies the Eying equation as follows (Luo et al. 2015):

$$\mathrm{Ln}\left({K}_{app}/T\right)=-\left(\Delta H/R\right)\times \left(1/T\right)+Ln\left({K}_{B}/h\right)+\Delta S/R$$
(4)

where the Boltzmann constant kB is 1.38 × 10−23 J·k−1 and the Planck constant h is 6.626 × 10−34 J·s; therefore, ΔH and ΔS can be calculated as 21.319 kJ·mol−1 and − 135.62 J·mol−1·k−1, respectively. The results show that the degradation of OFX by Fe(VI) was an endothermic reaction. With an increase in temperature, the number of effective collisions between the polymers increased, resulting in an acceleration of the reaction rate.

Influence of free radicals in reaction system

Studies have shown that the degradation of organic pollutants by Fe(VI) includes the direct oxidation by Fe(VI) and the indirect oxidation by generated free radicals (Zhang et al. 2012). And Fe(VI) and its intermediates (Fe(V) and Fe(IV)) can generate highly reactive hydroxyl radicals (OH•) during their self-decomposition process (Zhang et al. 2012; Noorhasan et al. 2010). In traditional advanced oxidation process experiments, OH• plays a significant role (Wang et al. 2018; Chen et al. 2020, 2019; Wang and Wang 2020). Therefore, in order to verify the contribution of ROS, TBA and IPA were used as free radical scavengers of OH•. As shown in Fig. 4, the OFX degradation rate decreased by 4.57% and 5.82% with the addition of TBA and IPA to the reaction system, respectively. Thus, it can be concluded that Fe(VI) does exist OH• in the experiments of degrading OFX, and OH• contributed little to the degradation of OFX. Therefore, it can be considered that the degradation of OFX is mainly due to the oxidation by high-valent iron-based. Similar results have been reported by Shao et al. (2019) and Jin et al. (2021).

Fig. 4
figure 4

Influence of OH• on OFX degradation. Reaction conditions: [OFX]0 = 8 μM, [Fe(VI)]0 = 120 μM, T = 25℃, pH = 7.0

Full size image

Influence of anions, cations, and organic matter

The common anions, cations, and organics in water were selected to investigate their influence on the degradation of OFX by Fe(IV). K+, Na+, Cl−, SO42−, and NO3− had almost no effect; Fe3+ and NH4+ greatly promoted OFX degradation; while HCO3− only had a small inhibitory effect on OFX degradation in the reaction system (Fig. 5). In addition, organic matter, represented by HA, significantly inhibited the degradation of OFX by Fe(IV) (Fig. 5).

Fig. 5
figure 5

OFX degradation as a function of water components. Reaction conditions: [OFX]0 = 8 μM, [Fe(IV)]0 = 120 μM, [K+]0 = [Na+]0 = [NH4+]0 = [Fe3+]0 = [HCO3−]0 = [NO3−]0 = [SO42−]0 = [Cl−]0 = [humic acid (HA)]0 = 50 μM, T = 25℃, pH = 7.0

Full size image

In water, HCO3− can slightly inhibit the degradation of OFX by Fe(VI). On the one hand, the presence of bicarbonate can increase the solution pH, which in turn affects the stability of the oxidant. As shown in Fig. 2, the degradation rate of OFX by Fe(VI) decreases with increasing pH. The reason is that Fe(VI) is more present in the form of FeO42− under alkaline conditions, and the reaction between FeO42− and OFX− is very slow (Zheng et al. 2020; Han et al. 2018; Huang et al. 2017). On the other hand, the HCO3− will react with OH•, thereby depleting free radicals in solution and reducing the reaction rate (Wang and Wang 2021).

However, the pH of the solution raised after bicarbonate hydrolysis is limited, and Fig. 4 also proves that the contribution of OH• is small. Therefore, HCO3− has only a small inhibitory effect on OFX degradation in the reaction system.

Fe3+ enhanced the OFX removal by Fe(VI) mostly via self-catalysis of Fe(VI) to generate more Fe(IV) or Fe(V) (Jiang et al. 2016; Ma et al. 2016; Zhao et al. 2018a). Moreover, NH4+ was reductive, which could promote the generation of Fe(IV) or Fe(V) from Fe(VI), thus improving the removal rate of OFX (Zhao et al. 2018b). Since HA and OFX compete with Fe(VI), HA has a significant inhibitory effect on the Fe(VI) degradation of OFX (Sharma et al. 2016; Horst et al. 2013).

Intermediate products and degradation pathways

Through a study of relevant references and Q-TOF LC/MS analysis of intermediate products of OFX degradation (Michael et al. 2013; Bi et al. 2019; Xue et al. 2017; Meng et al. 2021; Jin et al. 2021), and possible degradation pathways were proposed (Fig. 6). There are two main ways in which Fe(IV) degrades OFX. One of the degradation pathways is as follows: Firstly, the methyl group on the benzoxazine ring of OFX was carboxylated to form P1 (m/z = 391.1). P1 was then converted to P3 (m/z = 303.1) through decarboxylation. After the opening of the piperazine and oxazine rings, P3 was transformed into P5 (m/z = 178.1), and finally, P7 (m/z = 110.0) was formed through pyridine ring opening, defluorination, deamination, and hydroxylation. The other degradation pathway is as follows: OFX was converted into P2 (m/z = 337.1) through pyridine ring opening, and then P4 (m/z = 218.1) was formed by the opening of the piperazine and oxazine rings. P4 was finally transformed into P6 (m/z = 122.0) by deamination, defluorination, and dehydroxylation. P6 and P7 were further oxidized to form small molecular intermediates, which were finally decomposed into small molecular acids, CO32−, H2O, NO3−, and minerals (Zhou et al. 2021).

Fig. 6
figure 6

Pathways of OFX degradation by Fe(VI)

Full size image

Conclusion

This study showed that Fe(VI) is an effective oxidant for the degradation of OFX. The degradation rate of OFX by Fe(VI) was in accordance with the second-order kinetic equation. OFX degradation by Fe(VI) was directly proportional to the initial concentration of Fe(VI) and temperature and inversely proportional to the pH. Under the conditions of [Fe(VI)]:[OFX] = 15:1, T = 25℃, and pH = 7.0, the removal efficiency of 8 μM OFX reached more than 90% in 4 min. Seven intermediates were identified by Q-TOF LC/MS, and the opening loop, decarboxylation, breaking the C-N bond, and deamination were the main reaction processes identified in their formation. In conclusion, Fe(VI) advanced oxidation technology has obvious advantages in the treatment of wastewater containing OFX.