Introduction

Pharmaceuticals (PhCs) are widespread environmental pollutants. Recent investigations document that PhC production and administration may vary both between countries and over time (Goossens et al. 2007; Kümmerer 2009; Castiglioni et al. 2006), with fluctuations that are observed not only in different seasons. For many of PhCs, only partial removal from wastewater is observed by using traditional degradation treatments, and thus, most PhCs are still found in the effluents of wastewater treatment plants (WWTPs). Moreover, the discharge into surface waters and/or in WWTPs of PhCs may give rise to ecotoxicological effects (Daughton and Ternes 1999), and in case of antibiotics to antibacterial resistance.

PhCs and their metabolites were detected worldwide at ng L−1 to μg L−1 levels in surface water bodies in different areas (Verlicchi et al. 2012; Kolpin et al. 2002; Calamari et al. 2003; Mompelat et al. 2009; Ashton et al. 2004). Safety threshold values have been defined for a limited number of PhCs. Moreover, the products of drug degradation may also represent an environmental risk, and it was found that some photoproducts are more persistent and toxic than the parent drug (Della Greca et al. 2004).

The traditional primary treatments used in municipal WWTPs are usually inefficient for the removal of most PhCs. This problem was addressed in some studies at the end of last century, and it was put into evidence that the central nervous system (CNS) drug carbamazepine is neither degraded nor adsorbed (Clara et al. 2004), and in the class of lipid regulators, the removal rate of clofibric acid (the main metabolite of clofibrate) reaches 51 %, while 83 % of bezafibrate is removed (Ternes 1998). Presently adopted treatments, such as biological filters and activated sludge, are more efficient for the removal of PhCs but still inadequate especially for some antibiotics (erythromycin, ofloxacin, sulfamethoxazole, clarithromycin, amoxicillin, tetracycline, and azithromycin), CNS drugs (fluoxetine, diazepam), analgesics/antiinflammatories (ibuprofen, mefenamic acid), and lipid regulators (fenofibric acid, fenofibrate, gemfibrozil) (Hirsch et al. 1999). The photocatalytic processes appear to be convenient and green in comparison with the traditional chemical methods. In these processes, a chemical reaction is initiated when a semiconductor photocatalyst is irradiated by light with an energy that matches or exceeds the band gap energy of the semiconductor, resulting in excited electron-hole pairs that can be applied to the production or degradation of a compound (substrate) (Hoffmann et al. 1995). Moreover, a synergistic effect was demonstrated when photocatalysis was coupled with other AOP technologies, such as ozonization, microwave, or ultrasound treatments, although in some cases, cost issues might arise (Selli et al. 2008).

Among conventional photocatalysts, the TiO2-based materials have dominated in WWTPs, although the use of energy-consuming ultraviolet (UV) lamps as radiation source is an important drawback of their use. The objective of this work is to assess an economical heterogeneous photocatalyst for the decontamination of water from micropollutants under visible light irradiation. The photocatalytic activity was tested on ciprofloxacin (CP), which is an antibiotic that belongs to the class of therapeutic drugs known as quinolones, and on ibuprofen (IBP), one the most prescribed nonsteroidal antiinflammatory drugs. Since IBP is known to be degraded photochemically under UV radiation and sunlight (Packer et al. 2003) and that the toxic compound 4-isobutylacetophenone (4-IBAP) can be formed photochemically from IBP, upon direct photolysis (Ruggeri et al. 2013), we decided to use a visible light catalyst LaFeO3. As far as we know, this is the first time that the photocatalytic efficiency of LaFeO3 was evaluated for the degradation of IBP and CP under visible light irradiation.

Lanthanum ferrite LaFeO3 was proposed as visible light photocatalyst in 2007 (Li et al. 2007), although the first detailed study on the photocatalytic degradation of organic compounds was reported by Su et al. (2010), who observed an excellent photocatalytic performance in Rhodamine B degradation. Successive studies performed by using nanostructured LaFeO3 with different morphologies (Thirumalairajan et al. 2013), or LaFeO3 nanoparticles supported on the surface of montmorillonite (Peng et al. 2016), confirmed this capability for visible light degradation of Rhodamine B. Fenton-like reaction could enhance the photocatalytic degradation of Rhodamine B through a synergistic effect (Peng et al. 2016). However, tests with dyes (or with any substrate absorbing radiation) are not appropriate in photocatalysis, because they do not distinguish between a pure photocatalytic process, a dye sensitization, or both (Barbero and Vione 2016). Little is known about the photocatalytic efficiency of LaFeO3 under visible light for the degradation of molecules other than the organic dyes. Hu et al. (2012) reported that LaFeO3 displayed photocatalytic activity for the degradation of p-chlorophenol (49.0 %) after 5 h visible light irradiation.

In this paper, the photocatalytic efficiency of LaFeO3 nanopowder in the presence of H2O2 to decontaminate water from micropollutants under visible light irradiation is reported. Moreover, aquatic toxicity tests were also performed in order to test induced toxic effects on aquatic organisms due to the presence of the degradation products of ciprofloxacin and ibuprofen.

Methods

IBP (USP specifications), CP (≥98 % HPLC grade), and H2O2 (3 wt%) were from Sigma-Aldrich and were used as purchased. Table 1 shows the structure, the main characteristics, and the occurrence in the aquatic environment of the investigated PhCs.

Table 1 Main characteristics and occurrence in the aquatic environment of the investigated PhCs
Full size table

Lanthanum ferrite LaFeO3 (LF) was prepared by the citrate auto-combustion method (Caronna et al. 2009). LF is a semiconductor with band gap energy in the range of 2.5–2.6 eV (Parrino et al. 2016; Natali Sora et al. 2013). The photocatalytic degradation reactions were carried out by using a Rayonet reactor equipped with eight lamps (8 W, GE lighting 10055-F8T5/D) emitting in the visible region. Irradiation of aqueous suspensions containing 5.0 × 10−5 M (10.0 mg L−1) of IBP, 0.003 M H2O2, 130 mg L−1 LF or 5.0 × 10−6 M (1.7 mg L−1) of CP, 0.003 M H2O2, and 130 mg L−1 LF was carried out in magnetically stirred Pyrex glass tubes. The concentrations of IBP and CP were chosen so as to have a significant spectrophotometric signal. The concentration of H2O2 corresponded to a mole amount that largely exceeded the mole amount of IBP and CP but was about 1/30 than that used in typical studies on the heterogeneous Fenton-like process. The aqueous samples taken at regular intervals of irradiation were centrifuged for 10 min (Scharlab BL-8). The solutions obtained in centrifuging suspensions after irradiation were denominated IBPt and CPt, where t was the irradiation time (0, 1, 3.5, 5, 24 h).

The absorption spectra were measured by using an ultraviolet-visible (UV-vis) spectrophotometer (Jasco V-650), with 10-mm light path. Full spectra were taken in order to monitor any spectral interference that may occur. The normalized concentration of CP and IBP was calculated by the formula C t/C 0, where C 0 and C t are the total concentrations of the pollutant solution before irradiation and after an irradiation time t, respectively.

Acute toxicity of the degradation products of CP and IBP, in terms of the dose (concentration administered) that causes a particular effect in a specified population, was investigated. The Daphnia magna (D. magna Straus, crustacean cladocera) test was performed following the guidelines UNI EN ISO 6341: 2013 (UNI 2013). Tests were performed at 25 °C, in darkness, for 24 and 48 h, after which immobility was recorded. The Pseudokirchneriella subcapitata (P. subcapitata, green alga) test was performed according to ISO 8692: 2012 (ISO 2012). Tests were performed in uniform illumination for 72 h, with determination of algal growth. The Vibrio fischeri (V. fischeri, luminescent bacterium) test was performed according to UNI EN ISO 11348–3: 2009 (UNI 2009), by using a Microtox apparatus. Toxicants influence the metabolism of the bacterium and reduce luminescence, which was measured at 15 and 30 min and compared to control. The toxicity tests were performed on (i) solutions of IBP and of CP which were not irradiated (IBP0, CP0); (ii) solutions which were irradiated for 5 h (IBP5, CP5); (iii) solutions irradiated for 24 h (IBP24), and, as reference, (iv) a solution of distilled water and H2O2. The suspensions containing the catalyst were previously centrifuged to remove LF nanopowders.

Results and discussion

The photocatalytic degradation percentage of IBP with H2O2 + LF under visible light irradiation (5 h) was 42 %. As shown in Fig. 1, the degradation after 5 h visible light exposure of only IBP was not significant, while that of IBP with only H2O2, or of IBP in the presence of LF alone, was <4 %. A different behavior was found under 24-h visible light irradiation. IBP with only H2O2 displayed a degradation percentage of 73 %, which increased to 78 % for IBP with H2O2 + LF, while the degradation percentage of IBP itself and IBP with only LF remained <4 %. In Fig. 2, UV-vis absorption spectra of aqueous solutions of IBP with H2O2 in the presence of LF irradiated for 0, 3.5, 5, and 24 h are shown. The absorbance peak observed at 220 nm was chosen for the measurement of IBP concentration (Garzón and Martìnez 2004), while the peak at 263 nm was used for the detection of 4-IBAP, which is the main degradation product of IBP (Gasco-Lopez and Izquierdo-Hornillos 1999). The concentration maximum of 4-IBAP was reached after 5 h of irradiation.

Fig. 1
figure 1

Normalized concentration of IBP in solutions of IBP alone, IBP + LF, IBP + H2O2, and IBP + LF + H2O2 after 5 h visible light irradiation. Normalized concentration of CP in solutions of CP alone, CP + LF, CP + H2O2, CP + LF + H2O2 after 5 h visible light irradiation

Full size image
Fig. 2
figure 2

UV/vis absorption spectra of solution irradiated under visible light: a IBP 5 × 10−5 M, 130 mg L−1 LF, and 0.003 M H2O2 for 0, 3.5, 5, and 24 h; b CP 5 × 10−6 M, 130 mg L−1 LF, and 0.003 M H2O2 for 0, 1, 3.5, 5, and 24 h

Full size image

The most extensive photocatalytic degradation of CP (92 %) was detected by using H2O2 + LF under 5 h visible light irradiation, while degradation under irradiation decreased down to 64 % for CP with only H2O2, 38 % for CP with only LF, and 22 % for CP alone (Fig. 1). The degradation percentage obtained for CP with only LF (38 %) was comparable with the degradation of p-chlorophenol (49.0 %) reported by Hu et al. (2012) after 5 h visible light irradiation.

The UV-vis absorption spectra of aqueous solutions of CP with H2O2 in the presence of LF irradiated for 0, 1, 3.5, 5, and 24 h showed an absorbance peak at 274 nm and a broad peak at 317 nm, which were used for CP detection (Fig. 2). No other peaks were recorded during the photocatalytic degradation experiments. For the sake of comparison, the experiments were also carried out in the dark, and both CP and IBP were not significantly transformed.

The acute toxicological effects on freshwater organisms of the degradation mixtures of CP and IBP were evaluated, and the results are shown in Table 2. Following the classification related to the chemical toxicity adopted by the Commission of the European Community in 1996, chemicals are highly toxic from 0.1 to 1 mg L−1 (expressed by EC50 values), toxic from 1 to 10 mg L−1, harmful from 10 to 100 mg L−1, and nontoxic from 100 to 1000 mg L−1. The tests performed on solution IBP0, which was not irradiated, indicate that IBP0 was harmful for P. subcapitata and V. fisheri, while it was toxic to daphnids. Solutions irradiated for 5 h (IBP5) became highly toxic (D. magna and P. subcapitata) and toxic to V. fisheri. After 24 h irradiation (IBP24), the solutions still displayed a high toxicity. Tests performed on distilled water and H2O2 without irradiation displayed no acute toxicity to P. subcapitata and to V. fisheri and a very slight toxicity to D. magna (data not reported in Table 2). Under the described experimental conditions, after 24 h irradiation, the H2O2 was fully photolyzed. Moreover, the values of the normalized concentration C 5/C 0 and C 24/C 0 of IBP were 0.58 and 0.23, respectively. These results indicate that the toxicity of photoproducts is higher than that of IBP itself and that their degradation rate is smaller. The higher toxicity is most likely due to the presence of 4-IBAP detected in the irradiated solutions by using UV/vis spectrophotometry (Fig. 2). 4-IBAP is known to be a toxic degradation product formed from IBP, when heating is applied or under oxidative conditions (Caviglioli et al. 2002).

Table 2 Dose-effect relationships for acute toxicity to IBP and CP
Full size table

The tests performed on solutions of CP without irradiation (CP0) and after 5 h irradiation (CP5) indicated that both solutions were highly toxic for D. magna, P. subcapitata, and V. fisheri. Although the normalized concentration C 5/C 0 of CP5 was only 0.08, the acute toxicity remained similar to that of CP0, which suggests that the toxicity of degradation products is comparable to that of CP itself. This result is in agreement with the fact that the antimicrobial activity of fluoroquinolones derives from the quinolone moiety (Dodd et al. 2006), which is preserved under photodegradation (Vasconcelos et al. 2009). In our study, the solutions containing 1.7 mg L−1 of CP gave a higher toxicity compared to other studies (Vasconcelos et al. 2009), but it is worth to note that in this case, H2O2 was coupled with CP, which may have a synergic toxic effect.

Conclusions

By using visible light irradiation (5 h) and in the presence of LF and aqueous H2O2, an extensive photocatalytic degradation of CP (92 %) and IBP (42 %) was observed. However, the degradation products of CP and IBP induced toxic effects on aquatic organisms, which suggests incomplete detoxification after 5 h irradiation. The main degradation product of IBP was 4-IBAP which is more toxic and showed a slower photocatalytic degradation than the parent compound. On the contrary, the toxicity of CP degradation products, although not negligible, was comparable to that of CP itself.