In this study, photocatalytic and catalytic wet-air oxidation (CWAO) processes were used to examine removal efficiency of bisphenol A from aqueous samples over several titanate nanotube-based catalysts. Unexpected toxicity of bisphenol A (BPA) samples treated by means of the CWAO process to some tested species was determined. In addition, the CWAO effluent was recycled five- or 10-fold in order to increase the number of interactions between the liquid phase and catalyst. Consequently, the inductively coupled plasma mass spectrometry (ICP-MS) analysis indicated higher concentrations of some toxic metals like chromium, nickel, molybdenum, silver, and zinc in the recycled samples in comparison to both the single-pass sample and the photocatalytically treated solution. The highest toxicity of five- and 10-fold recycled solutions in the CWAO process was observed in water fleas, which could be correlated to high concentrations of chromium, nickel, and silver detected in tested samples. The obtained results clearly demonstrated that aqueous samples treated by means of advanced oxidation processes should always be analyzed using (i) chemical analyses to assess removal of BPA and total organic carbon from treated aqueous samples, as well as (ii) a battery of aquatic organisms from different taxonomic groups to determine possible toxicity.

INTRODUCTION

Bisphenol A (BPA), a representative of endocrine disrupting compounds (EDCs), is an industrial chemical, produced in large quantities and mainly used in the manufacture of polycarbonate plastics and epoxy resins (Chiang et al. 2004). Owing to well-known estrogen-like effects of EDCs on aquatic organisms (Sumpter 2005), there have been intensive efforts toward development of efficient technologies for the removal of EDCs from aqueous samples (Liu et al. 2009).

Among several possibilities for EDC removal from water, advanced oxidation processes (AOPs), such as heterogeneous photocatalysis, catalytic wet-air oxidation (CWAO), ozone-based technologies, and ultrasound oxidation seem to be the most promising (Liu et al. 2009; Silva et al. 2012). Numerous studies reported that photocatalytic oxidation (PCO) was found to be an efficient process for removal of recalcitrant pollutants from aqueous samples, especially pharmaceuticals, drugs and personal care products (Augugliaro et al. 2006). Catalysts used in PCO are mainly metal oxides, such as titanium dioxide, zinc oxide, zinc sulfide, ferric oxide, and tin oxide (Silva et al. 2012). To achieve more effective photocatalytic degradation of organic pollutants, TiO2-based catalysts were improved with deposition of noble metals, e.g. Ru, Au, Ag, and Pt (Chiang et al. 2004).

In the CWAO process, the other promising option for removal of recalcitrant organic compounds from aqueous samples, organic pollutants are oxidized by activated O2 species in the presence of a solid catalyst, usually at temperatures of 130–250 °C and pressures of 10–50 bar, into biodegradable intermediate products or mineralized into CO2, water and associated inorganic salts (Levec & Pintar 2007). Usually, the organic compounds are degraded over metal oxides, mixed metal oxide systems, supported noble metal catalysts, etc. (Cybulski 2007).

Experiences have shown that during chemical oxidation of pollutants, toxic intermediates and/or end-products with similar or even higher toxicity to aquatic organisms than parent chemicals could occur (Chiang et al. 2004). Nomiyama et al. (2007) reported that during the oxidative degradation of BPA using titanium dioxide, estrogenic activity increased as hydroxylated-BPA, carboxylic, phenolic, and other intermediates were produced by cleavage of a benzene ring. In our previous study, we found that oxidation during the CWAO process in the presence of TiO2 and Ru/TiO2 catalysts was more effective in regard to removal of toxicity and estrogenicity of BPA aqueous solution in comparison to photolytic/photocatalytic oxidation due to more severe reaction conditions in the CWAO process than in photolytic/photocatalytic oxidation. However, despite complete removal of BPA from some CWAO and photolytically/photocatalytically treated samples, the same or even higher toxicity or estrogenicity (or both) of some treated samples was detected in comparison to that of feed solution (Bistan et al. 2012). For this reason, samples treated by AOPs should be checked not only for their physico-chemical properties, but also for possible adverse biological effects before discharging them to the aquatic environment.

Severe experimental conditions, i.e. elevated temperatures and oxygen partial pressures during the CWAO process, provide an efficient oxidative degradation of pollutants in regard to chemical analysis, especially in the presence of titanate nanotube-based catalysts (Erjavec et al. 2013). However, unexpected toxicity of CWAO end-product solutions was obtained in toxicity tests with some tested aquatic organisms. For this reason, additional CWAO experiments with recycling of the effluent (five- and 10-fold) were performed with the aim of identifying possible reasons for toxicity of BPA aqueous samples treated by the CWAO process in a continuous-flow trickle-bed reactor. Toxicity of CWAO-treated aqueous samples was studied using bacteria Vibrio fischeri, algae Desmodesmus subspicatus, water fleas Daphnia magna, and zebrafish embryos Danio rerio. A battery of toxicity tests using test organisms from different taxonomic groups was selected because their sensitivity is highly species-specific. Efficiency of oxidative destruction of BPA was investigated by high-performance liquid chromatography (HPLC) and total organic carbon (TOC) measurements, while concentrations of metals in CWAO-treated samples were determined by means of inductively coupled plasma mass spectrometry (ICP-MS) analysis. Furthermore, oxidative destruction of BPA was investigated by means of photolytic/photocatalytic oxidation as the most common representative of low-temperature AOPs, performed in a batch glass reactor in the presence of suspended titanate nanotube-based catalysts. The obtained samples were analyzed using the same chemical and biological analyses as described for CWAO-treated samples.

MATERIALS AND METHODS

Synthesis of titanate nanotubes and catalysts' preparation and characterization are described in detail in our previous work (Erjavec et al. 2013). Briefly, an alkaline hydrothermal synthesis was applied for the synthesis of titanate nanotubes, which were then further annealed in the temperature range of 300–700 °C in steps of 100 °C. Within this temperature range a phase transition occurs from titanate to anatase structure along with simultaneous morphology and specific surface area change. The as-prepared sample (NT4) and annealed catalysts (NT4 (300), NT4 (400), NT4 (500), NT4 (600), and NT4 (700)) were examined in the reaction of BPA oxidation using two distinguished AOPs. As a reference, a commercial TiO2 (P25 Degussa) was included in the examination of catalysts, while an inert SiC was applied to determine the influence of non-catalytic contribution to BPA degradation in the CWAO system (Erjavec et al. 2013).

CWAO experiments

CWAO experiments were executed in a Microactivity-Reference (PID Eng&Tech, Spain) unit, which is a completely automated and computer-controlled continuous-flow trickle-bed reactor (Hastelloy C-276 tube) designed for catalytic tests (Pintar et al. 2008). Feed solution, which contained 10 mg/L of BPA (≥99%, Aldrich) was introduced continuously into the reactor at the flow rate of 0.5 mL/min. These standard CWAO experiments were conducted under continuous-flow operating conditions for 40 hours on stream. In a typical CWAO experiment, the reaction of BPA oxidation was conducted at 200 °C over 300 mg of a titanate nanotube-based catalyst. The properties of the catalyst bed and operating conditions can be found in our previous paper (Erjavec et al. 2013). Furthermore, a set of recycling CWAO experiments was performed in the batch-recycle trickle-bed reactor, where the treated BPA solution (300 mL, 10 mg/L at t = 0) was recycled in the above-described experimental setup. In this way, the BPA aqueous solution passed the catalyst layer either five or 10 times (50 or 100 hours on stream, respectively).

Photolytic/photocatalytic oxidation experiments

Heterogeneous photolytic/photocatalytic oxidation of BPA solution (initial concentration = 10 mg/L) was tested at atmospheric pressure in a batch slurry reactor (glass vessel, volume = 250 mL) thermostated at 20 °C (Julabo, model FP 25), magnetically stirred (360 rpm) and continuously sparged with purified air (45 L/h). The concentration of titanate nanotube-based catalyst was 0.125 g/L. After 30 minutes ‘dark’ period (for the establishing of equilibrium of the sorption process), the reactor content was illuminated by UVA high-pressure mercury lamp (150 W, with a maximum at λ = 360 nm) laid in a water-cooling jacket immersed vertically in the slurry.

Representative aqueous-phase samples were withdrawn from the reaction suspension after 75 minutes of operation and filtered through the 0.45 μm membrane filter (Sartorius) in order to remove catalyst particles.

Chemical analyses

The efficiency of the CWAO process and PCO was evaluated by determining the residual BPA content in comparison to the initial concentration, performed by means of HPLC apparatus (Spectra systemTM). The analyses were carried out in the isocratic analytical mode, using a 100 × 4.6 mm BDS Hypersil C18 2.4 μm column. The mobile phase was composed of methanol and ultrapure water (70:30 volume ratio), and was introduced into the system with a flow rate of 0.5 mL/min. The total amount of mineralized organic substances in withdrawn aqueous-phase samples was determined by measuring the TOC, using a TOC analyzer (Teledyne Tekmar, model Torch) equipped with a high-pressure non-dispersive infrared detector.

Elemental analyses were accomplished by ICP-MS. For ICP-MS measurements, a SCIEX Perkin Elmer ELAN DRC II (Perkin Elmer, Ontario, Canada) inductively coupled plasma mass spectrometer (with quadruple and single detector setup) was used. The running parameters of the instrument were checked and adjusted before every batch of measurements, using a solution with 1 μg/mL In, Ce, Th, and Mg, and 10 μg/mL Ba. The instrument calibration was carried out using a multielement Merck VI standard solution (Merck, Darmstadt, Germany) containing all elements of interest.

Toxicity tests

The luminescence of the freeze-dried marine bacterium Vibrio fischeri NRRL-B-11177, obtained from the manufacturer (Dr Lange GmbH, Düsseldorf, Germany), was measured on a LUMIStox 300 luminometer (International Organization for Standardization 2007a). The luminescent bacteria were exposed to the feed/initial solution of BPA (10 mg/L) and undiluted treated samples for 30 minutes at 15 ± 0.2 °C. The percentages of inhibition were calculated for each treated sample relative to the control.

The green, unicellular alga Desmodesmus subspicatus Chodat 1926 (SAG 86.81) was obtained from the Collection of Algal Cultures, University of Göttingen, Göttingen, Germany. A stock culture of alga was maintained in a nutrient solution according to Jaworski at a constant room temperature of 21 ± 1 °C under continuous fluorescent illumination (4000 lux) provided by four 20 W cool-white fluorescent lights (Osram). Flasks were agitated at 150 rpm for 15 minutes, alternating with 15 minutes resting, on an orbital shaker. Test flasks containing feed/initial BPA solution, treated samples, and a control (only growth medium) were constantly shaken at the same frequency as stock flasks; they were illuminated with four 40 W cool-white fluorescent lights giving an illumination of 7000 lux. After 72 hours of exposure, algal growth was determined by counting cells in a Bürker counting chamber and the inhibition of specific growth rates for each tested sample was calculated in comparison to the control (International Organization for Standardization 2004).

Daphnia magna Straus 1820 (Clone A), obtained from the ECT Oekotoxikologie, Flörsheim, Germany, were cultured at 21 ± 1 °C in 3-L aquariums covered with glass plates containing 2.5 L of Elendt M4 medium, illuminated with fluorescent bulbs (approximately 1800 lux) in a 16 hours light : 8 hours dark regime. They were fed twice a week with TetraMin® (20 mg blended in deionized water per aquarium), once a week with instant yeast (5 μg per aquarium) and four times a week with algae Desmodesmus subspicatus corresponding to 0.13 mg C/daphnia (Schweitzer et al. 2010). Neonates (<24 hours) were exposed to the feed/initial solution of BPA, undiluted treated samples, and a control. After 24 hours of exposure, immobile water fleas were counted and the percentages were calculated (International Organization for Standardization 2012).

Adult zebrafish Danio rerio Hamilton–Buchanan were bred in a temperature-controlled room in an aquarium (60 × 30 × 30 cm) containing 45 L of tap water with constant temperature (26 °C) and 12 hours light : 12 hours dark photoperiod. Fish were fed three times per day with commercially available dried fish food. One milliliter of feed/initial solution of BPA, treated samples and a control were dispensed into each hole of 24-well plates in 10 replicates and then fertilized eggs in the four- to eight-cell stages were placed into the holes (International Organization for Standardization 2007b). After 48 and 96 hours of exposure at 26 °C, lethal (egg coagulation, missing heartbeat and missing tail detachment from the yolk sac) as well as sub-lethal malformations (missing body and eye pigmentation, spine deformation, yolk sac edema, hatching) were observed and the percentages for each tested sample were calculated.

RESULTS AND DISCUSSION

Toxicity of CWAO and photocatalytically treated samples

BPA and TOC conversions of treated aqueous solutions by CWAO and photolytic/photocatalytic oxidation as well as results obtained in toxicity testing are given in Table 1 and Figure 1. Despite high removal of BPA (from 49 up to 87%) and TOC (up to 69%) from aqueous samples treated by CWAO (Table 1), high toxicity of all tested samples was determined for bacteria and algae. Furthermore, in most cases treated samples revealed even higher toxicity than the feed solution, especially in the case of sample exposure to green algae. In contrast, toxicity of CWAO-treated samples to water fleas and zebrafish embryos dropped significantly in all examined samples. Owing to the high and unexpected toxicity of CWAO-treated samples to bacteria and algae, a blank sample containing solely ultrapure water was continuously fed to the trickle-bed reactor filled with SiC (S9) (used as an inert filler to surround the thermocouple) and then subjected to toxicity tests as for the rest of the treated samples. Interestingly, even with this sample a high toxicity to bacteria (39% luminescence inhibition) and algae (57% inhibition of growth rate) was observed. However, blank sample (S9) was non-toxic to water fleas and zebrafish embryos (Figure 1(a)). Further experiments also showed that toxicity of CWAO samples treated at higher reaction temperature (230 °C) in the CWAO process significantly increased the toxicity to bacteria (to 62% inhibition of luminescence) and algae (to 98% inhibition of growth rate) in comparison to toxic effects obtained in the samples treated at 200 °C. Higher reaction temperature in the trickle-bed reactor is accompanied with higher operating pressure, both leading to more harsh reaction conditions, which, as a consequence, enhanced metal leaching from the tubular reactor.

Table 1

BPA and TOC conversion in aqueous solutions treated by means of catalytic CWAO and PCO and in recycled samples (RS)

Samples
 
    BPA conversion (%)
 
TOC removal (%)
 
CWAO PCO Catalyst Treated aqueous solution CWAO PCO CWAO PCO 
FS PIS BPA (10 mg/L) 
S1 SiC BPA 49.0 20.0 
S2 PS1 TiO2 (P25, Degussa) BPA 80.0 100.0 56.0 98.0 
S3 PS2 NT4 BPA 85.0 22.0 42.0 13.0 
S4 PS3 NT4 (300) BPA 76.0 70.0 64.0 8.0 
S5 PS4 NT4 (400) BPA 82.0 94.0 61.0 28.0 
S6 PS5 NT4 (500) BPA 78.0 100.0 58.0 78.0 
S7 PS6 NT4 (600) BPA 87.0 100.0 69.0 55.0 
S8 PS7 NT4 (700) BPA 79.0 97.0 47.0 38.0 
S9 SiC MQ water 
RS1 NT4 (600) BPA (Ra = 5) 100 64 
RS2 NT4 (600) BPA (Ra = 10) 100 81 
RS3 NT4 (600) MQ (Ra = 5) 
RS4 NT4 (600)b BPA (Ra = 10) 100 80 
Samples
 
    BPA conversion (%)
 
TOC removal (%)
 
CWAO PCO Catalyst Treated aqueous solution CWAO PCO CWAO PCO 
FS PIS BPA (10 mg/L) 
S1 SiC BPA 49.0 20.0 
S2 PS1 TiO2 (P25, Degussa) BPA 80.0 100.0 56.0 98.0 
S3 PS2 NT4 BPA 85.0 22.0 42.0 13.0 
S4 PS3 NT4 (300) BPA 76.0 70.0 64.0 8.0 
S5 PS4 NT4 (400) BPA 82.0 94.0 61.0 28.0 
S6 PS5 NT4 (500) BPA 78.0 100.0 58.0 78.0 
S7 PS6 NT4 (600) BPA 87.0 100.0 69.0 55.0 
S8 PS7 NT4 (700) BPA 79.0 97.0 47.0 38.0 
S9 SiC MQ water 
RS1 NT4 (600) BPA (Ra = 5) 100 64 
RS2 NT4 (600) BPA (Ra = 10) 100 81 
RS3 NT4 (600) MQ (Ra = 5) 
RS4 NT4 (600)b BPA (Ra = 10) 100 80 

aR – number of recycles.

bCatalyst mass = 0.5 g.

Figure 1

Toxicities of BPA feed/initial solutions (FS, PIS) and samples treated by (a) CWAO process, and (b) photocatalytic oxidation. See Table 1 for sample notation.

Figure 1

Toxicities of BPA feed/initial solutions (FS, PIS) and samples treated by (a) CWAO process, and (b) photocatalytic oxidation. See Table 1 for sample notation.

PCO was very effective for the removal of BPA from aqueous samples in the presence of TiO2 and titanate nanotube-based catalysts, as complete removal of BPA from aqueous samples was achieved in the samples PS1, PS5, and PS6. However, TOC removal was lower in comparison to most samples treated by the CWAO process (Table 1). Results obtained in toxicity tests showed that all photocatalytically treated samples were less toxic to bacteria in comparison to the initial BPA solution. Samples PS1, PS5, and PS6 were found to be non-toxic (20% of inhibition or less) to bacteria, which is in good agreement with complete BPA removal in these samples. A significant decline of acute toxicity of photocatalytically treated samples when comparing to the initial BPA solution was also observed after exposure of water fleas. Low percentages of toxic effects of initial BPA solution were observed for zebrafish embryos and especially for algae; in all photocatalytically treated samples, lethal effects on zebrafish embryos as well as algal growth inhibition remained similar to those exerted by the initial solution, or even completely disappeared (Figure 1(b)).

From the results obtained in toxicity tests with different test species, examining the CWAO-treated BPA aqueous samples (Figure 1(a)), it is evident that toxicity to bacteria and algae remained the same or was even higher compared to the untreated sample. However, this is not the case for photocatalytically treated samples (Figure 1(b)).

Recycled samples in CWAO process

Owing to the unexpected toxicity to bacteria and algae determined in the CWAO blank sample, BPA aqueous solutions were recycled five- or 10-fold to prolong residence time of the liquid phase in the batch-recycle trickle-bed reactor; results of chemical analyses and toxicity tests are listed in Tables 1 and 2.

Table 2

Toxicities of BPA initial solution (IS) and recycled CWAO samples to aquatic organisms

  V. fischeri D. magna D. rerio
 
Sample Luminescence inhibition (%) Immobility (%) Lethal effects (%) Sublethal effects (%) Hatching success (%) 
IS 70.0 55 20 100 
RS1 43.2 100 90 
RS2 46.3 100 
RS3 39.6 100 80 
RS4 47.7 100 10 10 
  V. fischeri D. magna D. rerio
 
Sample Luminescence inhibition (%) Immobility (%) Lethal effects (%) Sublethal effects (%) Hatching success (%) 
IS 70.0 55 20 100 
RS1 43.2 100 90 
RS2 46.3 100 
RS3 39.6 100 80 
RS4 47.7 100 10 10 

See Table 1 for sample notation.

In the recycled samples (RS1, RS2, RS4) complete removal of BPA, determined by HPLC analysis, was achieved; however, the level of mineralization was not as high as up to 36% of TOC still remained in these recycled samples (Table 1). High toxicity of all recycled samples was determined for V. fischeri. We found that the number of recycles and the presence or absence of BPA in recycled samples did not significantly alter the toxicity of tested samples. Moreover, sample RS3 obtained after five-fold recycle of ultrapure water induced 39.6% of luminescence inhibition. The highest toxicity of recycled samples was obtained toward water fleas, as 100% immobility was observed in all recycled samples after 24 hours; we observed that in a few hours after the start of the test all exposed water fleas were not able to swim. However, all of the recycled samples were non-toxic to zebrafish embryos after 24 and 48 hours of exposure, but a significant impact on hatching success of larvae, which is considered to be a key point in the life cycle of fish, was found, as 10-fold recycled samples (RS2 and RS4) inhibited hatching of larvae by up to 95% after 96 hours of exposure (Table 2).

From the obtained toxicity test results listed in Table 2, we assumed that some toxic metallic ions were leaching from the trickle-bed reactor construction material (i.e. Hastelloy C-276, consisting mostly of Ni (57%), Cr (16%), Mo (16%), Fe (5%), and W (4%)) during CWAO operation. Thus, ICP-MS elemental analysis was carried out to find possible reasons for the toxicity of recycled samples. For comparison, a photocatalytically reacted (PS3) and single-pass CWAO-treated sample (S7) were analyzed for potentially relevant elements; the results of ICP-MS analysis are given in Table 3.

Table 3

Concentrations of elements in single-pass sample (S7), recycled samples (RS1, RS2, RS3, RS4) obtained in CWAO experiments and photocatalytically treated end-product solution (PS3)

  Samples
 
Concentration (μg/L) S7 RS1 RS2 RS3 RS4 PS3 
Lithium 0.3 1.2 1.9 1.0 1.4 0.1 
Beryllium < 0.05 0.1 0.1 <0.05 <0.05 <0.05 
Boron 4.6 25.7 16.1 10.2 8.6 13.7 
Sodium 21.9 39.2 30.9 14.5 49.5 203.7 
Magnesium 19.8 97.9 78.5 28.4 76.4 53.8 
Aluminum < 0.05 0.1 <0.05 <0.05 <0.05 3.1 
Potassium 349.1 102.6 85.3 24.9 28.9 93.4 
Calcium < 0.05 189.4 207.2 <0.05 454.6 <0.05 
Scandium < 0.05 0.2 0.5 0.2 0.8 <0.05 
Titanium 0.1 0.6 0.5 0.3 0.6 0.6 
Vanadium 0.4 0.2 0.2 0.2 0.2 <0.05 
Chromium 153.6 997.5 1,526.1 1,013.5 2,327.1 2.5 
Manganese 2.3 0.5 0.4 0.2 0.7 0.4 
Cobalt 1.3 5.0 2.6 1.2 0.4 0.1 
Nickel 600.4 1,207.2 1,132.7 704.2 1,191.2 5.8 
Copper 0.7 0.5 0.4 0.4 0.7 0.9 
Zinc 6.9 13.4 13.0 4.4 24.6 2.7 
Germanium < 0.05 0.1 0.1 <0.05 0.1 <0.05 
Arsenic < 0.05 <0.05 0.1 0.1 <0.05 <0.05 
Rubidium < 0.05 <0.05 0.1 <0.05 <0.05 0.1 
Strontium 0.7 0.4 0.5 0.1 1.4 0.7 
Molybdenum 64.3 232.7 407.0 370.4 200.7 0.3 
Silver 0.4 0.7 10.1 10.2 9.3 0.2 
Cadmium 0.1 0.3 0.6 0.5 0.3 <0.05 
Tin < 0.05 <0.05 <0.05 <0.05 <0.05 0.1 
Barium 10.5 2.3 1.4 0.6 7.6 1.2 
Cerium < 0.05 <0.05 <0.05 0.1 <0.05 <0.05 
Tungsten 0.6 0.8 1.8 2.5 1.8 0.3 
Rhenium < 0.05 <0.05 0.1 <0.05 0.1 <0.05 
Gold < 0.05 0.1 <0.05 <0.05 <0.05 <0.05 
Mercury 0.1 0.1 0.2 0.2 0.1 0.1 
  Samples
 
Concentration (μg/L) S7 RS1 RS2 RS3 RS4 PS3 
Lithium 0.3 1.2 1.9 1.0 1.4 0.1 
Beryllium < 0.05 0.1 0.1 <0.05 <0.05 <0.05 
Boron 4.6 25.7 16.1 10.2 8.6 13.7 
Sodium 21.9 39.2 30.9 14.5 49.5 203.7 
Magnesium 19.8 97.9 78.5 28.4 76.4 53.8 
Aluminum < 0.05 0.1 <0.05 <0.05 <0.05 3.1 
Potassium 349.1 102.6 85.3 24.9 28.9 93.4 
Calcium < 0.05 189.4 207.2 <0.05 454.6 <0.05 
Scandium < 0.05 0.2 0.5 0.2 0.8 <0.05 
Titanium 0.1 0.6 0.5 0.3 0.6 0.6 
Vanadium 0.4 0.2 0.2 0.2 0.2 <0.05 
Chromium 153.6 997.5 1,526.1 1,013.5 2,327.1 2.5 
Manganese 2.3 0.5 0.4 0.2 0.7 0.4 
Cobalt 1.3 5.0 2.6 1.2 0.4 0.1 
Nickel 600.4 1,207.2 1,132.7 704.2 1,191.2 5.8 
Copper 0.7 0.5 0.4 0.4 0.7 0.9 
Zinc 6.9 13.4 13.0 4.4 24.6 2.7 
Germanium < 0.05 0.1 0.1 <0.05 0.1 <0.05 
Arsenic < 0.05 <0.05 0.1 0.1 <0.05 <0.05 
Rubidium < 0.05 <0.05 0.1 <0.05 <0.05 0.1 
Strontium 0.7 0.4 0.5 0.1 1.4 0.7 
Molybdenum 64.3 232.7 407.0 370.4 200.7 0.3 
Silver 0.4 0.7 10.1 10.2 9.3 0.2 
Cadmium 0.1 0.3 0.6 0.5 0.3 <0.05 
Tin < 0.05 <0.05 <0.05 <0.05 <0.05 0.1 
Barium 10.5 2.3 1.4 0.6 7.6 1.2 
Cerium < 0.05 <0.05 <0.05 0.1 <0.05 <0.05 
Tungsten 0.6 0.8 1.8 2.5 1.8 0.3 
Rhenium < 0.05 <0.05 0.1 <0.05 0.1 <0.05 
Gold < 0.05 0.1 <0.05 <0.05 <0.05 <0.05 
Mercury 0.1 0.1 0.2 0.2 0.1 0.1 

See Table 1 for sample notation.

The results of ICP-MS analysis showed high concentrations of chromium and nickel in the recycled samples; the concentration of chromium in 10-fold recycled sample (RS4) was more than 900 times and 15 times higher than in the photocatalytically treated sample (PS3) and single-pass sample (S7), respectively (Table 3). Slightly lower concentrations of nickel in comparison to chromium were obtained in recycled samples, but the concentration of nickel in the single-pass sample was quite high (600.4 μg/L). High concentrations of molybdenum were measured in all recycled samples; an apparently higher concentration (64.3 μg/L) of molybdenum was detected in the single-pass sample (S7) in comparison to the photocatalytically treated sample (PS3) (0.3 μg/L). Slightly increased concentrations of boron, zinc, silver, and tungsten were also detected in the recycled samples. However, measured concentrations of toxic metals such as chromium, nickel, and molybdenum in the photocatalytically treated sample were markedly lower in comparison to the CWAO recycled samples (Table 3). These findings show that, despite excellent corrosion resistance to both oxidizing and reducing media as well as excellent resistance to localized corrosion attack, Hastelloy C-276 material is not completely stable at hydrothermal conditions found in the three-phase CWAO process. The values listed in Table 3 confirm that the employed TiO2-based catalysts are chemically resistant in the employed range of operating and reaction conditions, since negligibly small concentrations of titanium in the liquid phase were measured.

Toxicity of recycled samples obtained in the CWAO process to D. magna is well correlated with the measured concentrations of chromium in the samples. Potassium dichromate has been used as a positive control in the standardized acute toxicity test with water fleas; the results of inter-laboratory testing demonstrated that the mean value of 24 hours EC50 was at 1.12 mg/L (International Organization for Standardization 2012). The results obtained in our laboratory showed that the 24 hours EC10 and 24 hours EC50 values were 0.585 and 1.28 mg/L K2Cr2O7, corresponding to 0.207 mg/L and 0.451 mg/L of chromium, respectively. The measured concentration of chromium in the single-pass sample S7 was below the 24 hours EC10; consequently, no toxic effects on water fleas were obtained. The measured values of chromium in all recycled samples (RS1–RS4) were higher than the 24 hours EC50 value, which could lead to mortality of all exposed water fleas in a few hours after the start of the toxicity test (Tables 2 and 3). According to the good correlation between toxicity of recycled samples and measured chromium concentrations, it seems that total chromium was mainly in Cr(VI) form. This is reasonable because of oxidative conditions in the CWAO reactor.

High concentrations of Ni were detected in the recycled samples, which could also contribute to extremely high toxicity of tested samples to water fleas. Reported 48 hours LC50 values were from 510 μg/L (Biesenger & Christensen 1972) up to 5.50 mg/L (Deleebeeck et al. 2008) depending on physico-chemical parameters of water, such as pH value, and hardness. Furthermore, measured concentrations of silver of the recycled samples (up to 10.2 μg/L) were high enough to cause toxic effects on Daphnia magna, as the reported 48 hours EC50 value was 1.05 μg/L (Ribeiro et al. 2014). Measured concentrations of zinc and molybdenum in the recycled samples were below the reported 24 hours (or 48 hours) LC50 values (Diamantino et al. 2000; Guilhermino et al. 2000).

In contrast to the high sensitivity of water fleas to some metals, V. fischeri is more tolerant to metals. Published data reported low toxicity of metals to luminescence bacteria; IC50 values for short-term exposure obtained in the Microtox system corresponded to 18 mg/L for Cr(VI) (15 minutes) and 43.9 mg/L for zinc sulfate (5 minutes) (Jennings et al. 2001; Hsieh et al. 2004). Toxicity of Cr(III) was reported to be similar to that of Cr(VI), as 15 minutes and 30 minutes IC50 values for Cr(III) were obtained at 15.3 mg/L and 16.0 mg/L, respectively (Lopez-Roldan et al. 2012). Accordingly to Lopez-Roldan et al. (2012), the 30 minutes EC50 value for nickel was 42.2 mg/L. Fulladosa et al. (2005) reported threshold concentration values (15 minutes EC20) for several metals tested at two pH values (6.0 and 7.0); all of the obtained threshold concentrations were significantly higher than measured concentrations in the recycled samples. From the available data about metal toxicity to Vibrio fischeri we did not find an evident reason for the inhibition of luminescence in the recycled samples. Therefore, it may be assumed that synergistic effects could play a role in the obtained toxicity to bacteria.

Literature data provide evidence that the egg chorion can protect fish embryos from different contaminants such as metals (Alsop & Wood 2011). This could be a possible reason for no lethal effects observed during the exposure of embryos to recycled samples.

CONCLUSIONS

The conversion of BPA in aqueous samples was facilitated in the presence of titanate nanotube-based catalysts with different physico-chemical properties under oxidative conditions in the CWAO process; consequently, toxicity to water fleas and zebrafish embryos significantly decreased. However, this was not the case for bacteria and algae. Additional CWAO experiments with five- and 10-fold recycling of BPA solution or ultrapure water in the batch-recycle trickle-bed reactor were performed. ICP-MS analyses of recycled samples confirmed elevated concentrations of toxic metals such as chromium, nickel, molybdenum, silver, and zinc in comparison to the photocatalytically treated sample, which was produced in the glass batch reactor. Owing to severe experimental conditions (high temperature and pressure) during the CWAO process, metals were leaching from the reactor construction material. This CWAO drawback could be avoided by applying a tubular reactor of more leaching-resistant composition and/or using novel catalysts that would reduce operating temperature and pressure. Conversely, PCO of BPA aqueous solution carried out at mild experimental conditions (i.e. atmospheric pressure and room temperature) was significantly less prone to the contamination of treated water by metals leached from the process equipment. Finally, the obtained results indisputably show that a battery of bioassays should be involved in the assessment of removal efficiency of pollutants from aqueous solutions treated by means of AOPs.

ACKNOWLEDGEMENT

The authors gratefully acknowledge the financial support of the Ministry of Education, Science and Sport of the Republic of Slovenia through Research program No. P2-0150.

REFERENCES

REFERENCES
Biesenger
K. E.
Christensen
G. M.
1972
Effects of various metals on survival, growth, reproduction and metabolism of Daphnia magna
.
J. Fish. Res. Board Can.
29
,
1691
1700
.
Deleebeeck
N. M. E.
de Schamphelaere
K. A. C.
Heijerick
D. G.
Bossuyt
B. T. A.
Janseen
C. R.
2008
The acute toxicity of nickel to Daphnia magna: predictive capacity of bioavailability models in artificial and natural waters
.
Ecotox. Environ. Safe.
70
,
67
78
.
Fulladosa
E.
Murat
C. J.
Martinez
M.
Villescusa
I.
2005
Patterns of metals and arsenic poisoning in Vibrio fischeri bacteria
.
Chemosphere
60
,
43
48
.
Guilhermino
L.
Diamantino
T.
Silva
M. C.
Soares
A. M. V. M.
2000
Acute toxicity test with Daphnia magna: an alternative to mammals in the prescreening of chemical toxicity?
Ecotox. Environ. Safe.
46
,
357
362
.
International Organization for Standardization
2004
Water Quality – Freshwater Algal Growth Inhibition Test with Unicellular Green Algae
.
ISO 8692
,
Geneva
,
Switzerland
.
International Organization for Standardization
2007a
Water Quality – Determination of the Inhibitory Effect of Water Samples on the Light Emission of Vibrio fischeri (Luminescent Bacteria Test) – Part 2: Method Using Liquid-Dried Bacteria
.
ISO 11348-2
,
Geneva
,
Switzerland
.
International Organization for Standardization
2007b
Water Quality – Determination of the Acute Toxicity of Waste Water to Zebrafish Eggs (Danio rerio)
.
ISO 15088
,
Geneva
,
Switzerland
.
International Organization for Standardization
2012
Water Quality – Determination of the Inhibition of the Mobility of Daphnia magna Straus (Cladocera, Crustacea) – Acute Toxicity Test
.
ISO 6341
,
Geneva
,
Switzerland
.
Levec
J.
Pintar
A.
2007
Catalytic wet-air oxidation processes: a review
.
Catal. Today
124
,
172
184
.
Lopez-Roldan
R.
Kazlauskaite
L.
Ribo
J.
Riva
M. C.
Gonzales
S.
Cortina
J. L.
2012
Evaluation of an automated luminescent bacteria assay for in situ aquatic toxicity determination
.
Sci. Total Environ.
440
,
307
313
.
Ribeiro
F.
Gallego-Urrea
J. A.
Jurkschat
K.
Crossley
A.
Hassellöv
M.
Taylor
C.
Soares
A. M. V. M.
Loureiro
S.
2014
Silver nanoparticles and silver nitrate induce high toxicity to Pseudokirchneriella subcapitata, Daphnia magna and Danio rerio
.
Sci. Total Environ.
466–467
,
232
241
.