Abstract
The objective of this study was to systematically evaluate the performance, kinetic behavior, and operational robustness of a Moving Bed Biofilm Reactor (MBBR) for the removal of Bisphenol-A (BPA) from synthetic wastewater under increasing pollutant loads. Specifically, the study aimed to (i) assess organic matter and ammoniacal nitrogen removal efficiencies, (ii) determine the kinetic parameters governing BPA biodegradation, and (iii) investigate the impact of high BPA concentrations on nitrification and biofilm resilience. The reactor maintained high soluble chemical oxygen demand (sCOD) removal (>80%) at BPA concentrations up to 50 mg/L, declining to 47% at 160 mg/L. Despite this, BPA removal consistently reached 99% after acclimation, with maximum volumetric removal rates rising from 0.04 to 0.97 kgBPA/m³·d, demonstrating robust biofilm adaptation. Apparent ammoniacal nitrogen (NH4+-N) removal remained stable, averaging 96%, but the maximum removal rate declined by 87% as BPA increased, indicating a strong inhibitory effect on nitrifying bacteria. The findings provide insight into the adaptability of biofilm-based systems for the treatment of recalcitrant organic contaminants and their potential scalability for complex wastewater streams.
Key words
Biofilm Adaptation; Bisphenol-A (BPA); Inhibition Kinetics; Moving Bed Biofilm Reactor (MBBR); Nitrification Inhibition; Nitrogen Removal
INTRODUCTION
Endocrine disruptors chemicals (EDCs) are regarded by regulatory organizations such as the World Health Organization (WHO), the United States Environmental Protection Agency (USEPA), and the European Environment Agency (EEA) as substances of great concern among chemical polluting substances (Amin et al. 2018). EDCs are compounds capable of interfering with the endocrine system of animals and humans, and can be of different classes, such as alkylphenols, pesticides, phthalates, natural estrogens, phytoestrogens, among other pharmaceutical substances, synthetic, and natural compounds (Bila & Dezotti 2007). Adverse effects caused by exposure to EDCs include diseases such as obesity, infertility, learning difficulties, memory loss, diabetes, cardiovascular diseases, as well as suppression of the immune system, increasing susceptibility to infections (Bergman et al. 2013).
Among the many chemical compounds with endocrine-disrupting characteristics, Bisphenol-A (BPA) stands out, as this molecule can bind to various hormone receptors, such as estrogenic and androgenic receptors, glucocorticoid receptors, and thyroid receptors (Demeneix & Slama 2019). With an annual production of about 3 million tons, Bisphenol-A is used in the production of polycarbonate plastics and epoxy resins, among other components, many of which come into contact with food, being the main route of human contamination with BPA (Costa et al. 2014). Calafat et al. (2005), in the United States, and Bushnik et al. (2010), in Canada, found that BPA is present in 95% and 70% of urine samples collected, respectively, evidencing the widespread exposure to this substance across the population.
BPA is introduced into the environment through various routes: contamination from production, handling and transportation in the form of dust; leachate leakage in landfills; and, mainly, through wastewater from industries that produce BPA or use it in the composition of other products. In many cases, the existing wastewater treatment technologies are not efficient in degrading or removing BPA, leading to concentrations of up to 150 µg/L in industrial effluents (Juan et al. 2021). In 2006, the total emissions of Bisphenol-A to the environment were estimated as 100 tons (Keykavoos et al. 2013). In surface waters it has been found in concentrations ranging from 0.05 µg/L to 3.61 mg/L (Mutseyekwa et al. 2017).
Bisphenol-A is associated with various metabolic diseases. Studies by Lang et al. (2008), Melzer et al. (2010), and Shankar & Teppala (2011) provide evidence of this correlation, showing that higher concentrations of BPA in urine samples are related to diagnoses of type-2 diabetes and cardiovascular diseases in the population of the United States, both for obese and normal-weight individuals. BPA is also associated with thyroid function dysregulation, altering the production of thyroid hormones thyrotropin (TSH), triiodothyronine (T3), and thyroxine (T4). However, the effect of BPA is not the same for different thyroid hormones, with an increase in the concentration of T3 observed with increasing BPA in urine samples (Meeker et al. 2010, Wang et al. 2012), but an inverse relationship with respect to T4 and TSH (Meeker & Ferguson 2011).
To reduce human and environmental contamination with Bisphenol-A, particularly in aquatic environments, various treatment methods, including biological and physicochemical technologies, have been studied and applied for the removal of BPA from industrial and domestic effluents. Diverse processes were studied, such as: membrane separation through microfiltration (Chen et al. 2018, Zielińska et al. 2016), nanofiltration (Escalona et al. 2014, Koloti et al. 2017), reverse osmosis (Fujioka et al. 2015), ultrafiltration (Zielińska et al. 2017), and membrane distillation (Naidu et al. 2017, Tufail et al. 2021); adsorption processes (de Lima et al. 2021, Zhang et al. 2021); as well as advanced oxidative processes, like Fenton/photo-Fenton (Xiao et al. 2020, Zhu et al. 2018), photocatalysis (Hunge et al. 2021, Wang et al. 2018), ozonation (Mu et al. 2021, Wang et al. 2021); and biological processes (Ouada et al. 2018, Bilal et al. 2019).
Wastewater treatment plants (WWTPs) commonly use biological treatments to remove organic compounds due to its cost–benefit and flexibility. Unlike conventional suspended growth biological processes, widely used in WWTPs, attached growth technologies may present more resilience to fluctuations of organic load and pH, as well as to toxic and recalcitrant compounds (van Haandel & van der Lubbe 2012).
One excellent alternative among attached growth processes is the Moving Bed Biofilm Reactor (MBBR), which works by providing protected surface for biofilm growth in plastic supports kept fluidized inside a reactional volume through aeration or mechanical stirring. It combines the advantages of other biofilm technologies and of activated sludge reactors – such as using the entire reactor volume for microorganism growth, longer sludge ages without the need to recycle sludge, better microorganisms acclimation, and smaller installation areas – and eliminates many of their drawbacks – such as lower head loss and absence of blockages when compared to fixed-bed processes, and lower solids generation than conventional activated sludge. Additionally, MBBRs present ease of operation and maintenance, adaptability to changes in operational conditions and feed loads, as well as versatility for aerobic, anoxic, and anaerobic processes (Bassin & Dezotti 2008, Ødegaard 2006).
Although the MBBR is a well-developed technology, applied for various wastewater treatment processes, few studies have applied the MBBR for the removal of BPA in wastewater. Despite being a robust process, it has been assessed mainly as part of a treatment train, used as pre- or post-treatment for membranes or oxidative processes (Wang et al. 2022, Buhari et al. 2023). Consequently, there is a significant lack of long-term studies that evaluate the continuous adaptation of nitrifying biofilms to progressively higher toxic loads of BPA. Furthermore, the ability of an MBBR to maintain stable nitrogen removal while the heterotrophic community adapts to BPA as the sole carbon source remains poorly understood. To address this gap, the present study investigates the removal of BPA in a bench-scale MBBR over a 770-day period, evaluating the functional resilience of the biofilm and the impacts on organic matter and ammoniacal nitrogen removal under concentrations ranging from 5 to 160 mgBPA/L.
MATERIALS AND METHODS
MBBR Configuration
The experimental unit used in this study was composed of a cylindrical glass reactor with an internal diameter of 6.7 cm, with a semi-spherical base for better hydrodynamic conditions. The reactor was fed in upward flow mode, promoting better homogenization of the mixture, using a peristaltic pump to maintain the flow rate at 1.2 L/d, which provided a 6h hydraulic retention time (HRT), based on previous studies of Gaioto et al. (2023) and Matheus et al. (2020). The MBBR had a working volume of 310 mL, filled with AnoxKaldnes® K1 carriers (500 m²/m³) made of high-density polyethylene (HDPE) at a filling fraction of 50%, resulting in 250 m²/m³ effective specific area. Aeration was promoted through an air diffuser porous stone on the bottom of the reactor and a rotameter to control air flow, maintaining appropriate hydrodynamic and oxygenation conditions. The schematic representation is shown in Fig. 1. As this study utilized synthetic wastewater and bench-scale reactors without involving human subjects or live animals, approval from an ethics committee was not required.
Schematic representation of the bench-scale MBBR experimental setup: (1) Influent synthetic wastewater; (2) Peristaltic pump; (3) Cylindrical glass reactor with AnoxKaldnes® K1 carriers; (4) Air diffuser stone; (5) Rotameter for aeration control; (6) Effluent; (7) AnoxKaldnes® K1 carrier.
Continuous MBBR Operation
The reactor operated for 770 days, of which the first 120 days corresponded to the phase of inoculation and biofilm development. Inoculation of the MBBR was made using 3 K1 carriers with established biofilm from 3 different bench-scale MBBRs fed with similar synthetic sanitary effluent used in this work, composed of 0.330 g glucose/L (CODeq ≈ 350 mg/L), 0.33 g NaHCO3/L, 0.1146 g NH4Cl/L (NH4+-N = 30mg/L), 0.025 g K2HPO4/L, 0.020 g KH2PO4/L, and 0.5 mL of trace metal solution/L, to provide biofilm development. The trace metal solution is described elsewhere (Vishniac & Santer 1957). Following the inoculation and biofilm development period, the next 650 days of continuous operation can be divided into seven different phases, as summarized in Table I. These phases correspond to increasing BPA’s concentration in the influent, ranging from 0 to 160 mgBPA/L. Each operational phase was maintained until steady-state conditions were achieved, defined as a period in which effluent concentrations of sCOD, NH4+-N, and BPA remained stable with a coefficient of variation below 10% across at least four consecutive sampling events (conducted twice a week). Given the HRT of 6 h , the durations listed in Table I (36–154 days) were sufficient to ensure both hydraulic and biological equilibrium.
BPA concentration for each phase was first calculated, and the corresponding mass of BPA was determined based on the volume of synthetic effluent to be prepared. The calculated mass of BPA was initially dissolved in 2 L of synthetic effluent to ensure proper mixing and solubility. This solution was then diluted to the appropriate final volume of synthetic effluent. Table I also shows the theoretical equivalent COD (BPA-CODeq) for each BPA concentration used.
Throughout the 650 days of operation, the dissolved oxygen concentration ranged from 7,5 to 8.2 mgO2/L, the pH was kept at near neutral conditions (ranging from 6.4 to 7.8), and the temperature ranged from 24°C to 31°C.
To evaluate the effects of the bisphenol-A presence on the removal of organic matter and ammoniacal nitrogen, related parameters were monitored between P1 and P7, as described in Table II (section Analytical Methods). Samples (50 mL) from the influent and the effluent streams were collected regularly and filtered immediately using cellulose nitrate membranes with 0,45 µm average pore diameter. Usually, the parameters were analyzed on the same day of sampling; however, in some cases, the analyses were conducted on later days, with the samples being preserved at -4°C in 50 mL Falcon tubes.
Kincannon-Stover Kinetic Model
Data obtained from continuous operation was fitted to the adapted Kincannon-Stover model (Hosseiny & Borghei 2002) in order to obtain a maximum substrate utilization rate and biodegradability. The maximum substrate removal rate (Umax) and the saturation constant (KB) can be calculated from the slope and interception of the line obtained from the plot of 1/VRR (the inverse of the volumetric removal rate, VRR) against the 1/VLR (the inverse of the volumetric loading rate, VLR). This model can be applied to datasets where there are variations in some of the conditions throughout the time, as was observed in BPA concentration in this work. In contrast, sCOD removal and nitrification processes exhibited low variability throughout the operation, which results in a lack of a wide range of loading rates necessary for a statistically significant kinetic fit. To ensure the robustness of the BPA kinetic model, validation was performed using a multi-indicator approach, including the coefficient of determination (R²), Root Mean Square Error (RMSE), Residual Sum of Squares (RSS), and 95% confidence intervals for the model parameters.
Kinetic Batch Trials
Batch trials were conducted to obtain the maximum removal rates of COD, ammoniacal nitrogen, and bisphenol-A at similar operating conditions as those of continuous operation. To this end, tests were conducted at the final stages of each operating phase, as detailed in Table III to Table IV, to ensure that maximum removal rates were representative of a fully acclimated and stabilized biofilm. The batch trials were carried out according to the following protocol: with the reactor outlet and inlet closed, all liquid content was replaced with a new batch of synthetic effluent. Samples were collected (8 mL in P1 and 10 mL in P2-P7) at intervals of 15 min during the first 2 h (0 to 120 min) of the batch trials, 20 min from 120 to 240 min duration, and 30 min from 240 to 360 min. The samples were immediately filtered through 0,45 µm cellulose nitrate membranes, to halt microbial activity, and refrigerated at 4°C.
Considering that the substrate is in excess at the beginning of the batch trials, zero-order removal kinetics can be assumed (Henze et al. 2008). Thus, the maximum volumetric removal rate (r, in kg/m³·d) and maximum superficial removal rate (s, in kg/m²·d) were determined through linear regression of the first points of concentration against time. Outliers were removed from the linear regression using the statistical concept of Cook’s Distance. It was calculated for each data point of the batch trials and those with Cook’s Distance higher than three times the average Cook’s Distance were considered outliers and removed from the linear regression (Barnett et al. 1975).
Analytical Methods
The list of analyzed properties and their frequency is seen in Table II, together with their respective analytical method.
Bisphenol-A
Bisphenol-A concentration analysis followed the 4-aminoantipyrine colorimetric method adapted from the procedure applied by Ettinger et al. (1951) for the determination of phenolic compounds. This method consists of the reaction between phenolic compounds and 4-aminoantipyrine in the presence of potassium ferricyanide. This reaction produces a colored complex, which can be quantified using a visible light spectrophotometer (460 nm). The intensity of the color is proportional to the concentration of phenolics in the sample, allowing for their determination in environmental or industrial water samples. As this method determines the presence of phenolic compounds in the samples, the results obtained do not necessarily reflect the presence of BPA only, since it may have been decomposed into other substances that still contain the phenolic ring, being detected by the method applied. Figure S1 (Supplementary Material) shows the calibration curve for BPA concentration with the method with a coefficient of determination (R²) of 0.997.
Statistical Analysis
The statistical analyses carried out when processing the data in this study used the one-way ANOVA statistical test from the Statistics Kingdom (2022) online calculator. The one-way ANOVA test checks the null assumption that the averages of two or more datasets are equal. The online calculator also runs the Turkey HSD, that compares each pair of datasets separately, providing more information if the difference between the sample averages reflects a real difference between the groups, or is due to the random noise inside each group. The normality of each group is evaluated by the Shapiro Wilk test; and the Levene’s test is used to evaluate the equality of variances. Data referring to the concentration or percentage removal were analyzed, with a significance level of α = 0.05 The associated statistics, F-value and p-value were recorded and are reported. A lower F-value and a higher p-value suggest greater confidence in accepting the null hypothesis of ANOVA (i.e., the averages of the groups are equal).
RESULTS
Chemical Oxygen Demand
The degradation of organic matter was monitored by measuring the chemical oxygen demand (COD) in both the influent and effluent streams of the reactor throughout the 650 days of continuous operation. In the influent, total chemical oxygen demand (tCOD) originated exclusively from soluble organic matter introduced to the reactor from glucose and bisphenol-A (phases 2 to 6), or only from bisphenol-A (phase 7). In the effluent, over 97% of the total COD originated from soluble organic matter. Thus, no report on suspended COD is given at this time.
Fig. 2 shows the average sCOD values in the influent and the effluent (sCODin and sCODef, respectively) obtained for each phase and the respective average sCOD removal. The corresponding contributions of bisphenol-A to the influent and effluent average sCOD are also plotted (BPA-sCODin and BPA-sCODef). The higher influent chemical oxygen demand observed in P6 and P7 can be attributed to higher BPA concentration. In fact, the contribution of BPA to the inlet sCOD ranged from 4% during P2 to almost 50% in P6, after which BPA completely substituted any other organic matter source during P7.
Average sCOD in the influent and effluent and removal percentage during the continuous operation. The fraction of BPA-related sCOD is shown in each column.
The overall average influent sCOD (P1 to P7) was 300 ± 75 mg/L, while the effluent was 40 ± 34 mg/L. However, during phases P6 and P7, the average influent sCOD was 450 mg/L, whereas the average effluent sCOD was 154 mg/L. A statistically significant difference is observed for the removal of sCOD among phases P1 to P7 (ANOVA, p-value = 4.12287x10-7, F = 8.781990), mainly due to the decrease from an average of over 80% (P1 to P5) to 70% (P6) then 47% (P7). This reduction coincides with the maximum removal rates gradually converging to the apparent removal rate over the phases (as seen in Table III), as the capacity decreases and gets closer to the volumetric loading rate. It reflects the transition of the organic substrate from glucose to BPA, a compound with lower biodegradability due to its molecular structure. At high concentrations, BPA can induce toxic shock in microbial communities, slowing sCOD degradation and necessitating longer retention times for complete removal. However, findings by Ferro Orozco et al. (2013) indicate that the gradual increase in BPA concentration allows for microbial acclimation, enhancing degradation efficiency over time.
Furthermore, the elevated sCOD levels observed in the effluent during P6 and P7 suggest the accumulation of BPA biodegradation by-products, a phenomenon also reported by Ferrer-Polonio et al. (2021). As the 4-aminoantipyrine method specifically targets the phenolic ring, a high BPA removal efficiency coupled with a lower overall sCOD removal indicates that the original molecule was transformed into intermediate metabolites that lack the phenolic structure but contribute to the residual organic matter, leading to the observed rise in effluent sCOD values.
Additionally, when sCOD apparent removal rates were analyzed (Table III), it can be observed that they follow a similar trend with the average sCOD levels in the influent. It can be observed that there is a slight reduction in the apparent removal rates following the introduction of BPA in the influent (ANOVA between P1 and P2, p-value = 0.000820259, F = 4.389136). From phases P3 to P6 there is a gradual increase in sCOD apparent removal rates (ANOVA, p-value = 0.00285712, F = 5.631232) following each increase in BPA concentration in the feed. In turn, during the beginning of P7 there is another decrease in the apparent sCOD removal rates (P7.1 in Table III) as a result of the substitution of glucose as main organic matter source to BPA as the only source. On the later stages of the phase, the apparent sCOD increased again (P7.2 in Table III), likely as a consequence of the microbiota adaptation. Overall, the results show little variation throughout the operation (apart from P6.1 and P7.1) for volumetric and superficial sCOD apparent removal rates as the organic matter source was shifted from an easily biodegradable source (glucose) to organic materials harder to metabolize, such as BPA and, possibly, its biodegradation by-products.
The batch trials conducted to assess the kinetics of organic matter removal (as detailed in section Kinetic Batch Trials) resulted in an overall trend of decreasing maximum removal rates from phase 2 to 7. Fig. 3 presents the plots of sCOD reduction during the batch trials conducted at each phase, corresponding to each applied BPA concentration (0, 5, 15, 30, 50, 100 and 160 mBPA/L). For the batch trial of P1 (0 mgBPA/L), a removal percentage above 85% was achieved after 140 minutes, whereas for the batch trials of P2 and P3 (5 and 10 mgBPA/L, respectively), removal efficiencies above 99% were achieved within the same 140 min. For P4 (30 mgBPA/L), a maximum removal of 97% was attained after 180 minutes; and for P5, it took 210 minutes to reach a 90% removal.
A more substantial decrease of the sCOD maximum removal was noticed in the batch trial P6.1 (100 mgBPA/L), when 270 minutes have passed before reaching only 75% removal. It further decreased to 50% during P6.2 within the same reaction duration, remaining the same at P7.1. It rose again in test P7.2 to 75% removal after 330 minutes. The results indicate that the presence of 100 and 160 mgBPA/L negatively impacted the reactor. The observed decrease in sCOD removal efficiency can be attributed to the toxicity of BPA at high concentrations and its higher contribution to sCOD, as well as the formation of by-products resistant to biodegradation processes. Furthermore, the toxic effects of BPA on organic matter removal take longer to manifest compared to its effects on BPA and nitrogen removals. Although high removal percentages were achieved at intermediary BPA concentrations, it can be observed that longer times were necessary at each step to achieve high removal efficiencies, whereas for higher concentrations (P6 and P7), low removal percentages were achieved even for the entire test duration. This slowdown emphasizes the inhibitory effects of high BPA concentrations on microbial metabolism and aligns with findings by Zielińska et al. (2014) on toxic impacts of BPA on nitrifying and COD-removing bacteria.
Differently from what was observed for apparent removal rates – which, throughout the continuous operation, showed little variations – volumetric and superficial maximum removal rates increased from P1 (before BPA introduction to the system) to P3 (15 mgBPA/L), decreasing significantly from phases P3 to P7.
Additionally, it can be observed that there was a marked decline in maximum removal rates during Phase P7, where BPA became the sole carbon source, reflecting the limitations of the biofilm in handling high concentrations of recalcitrant organic matter. BPA’s known inhibitory effects on microbial metabolism (Dinçer 2021, Ferrer-Polonio et al. 2021) likely contributed to this decrease, as high levels of BPA can disrupt microbial enzymatic processes, reducing the biofilm’s overall biodegradation capacity. The drop in maximum volumetric removal rates from an average of 2.56 ± 0.5 kg/(m³·d) in earlier phases (P1 to P6.1) to 1.15 kg/(m³·d) and 1.36 kg/(m³·d) (P7.1 and P7.2, respectively), and maximum superficial removal rates from 10.22 ± 1.95 g/(m²·d) (P1 to P6.1) to 4.60 and 5.43 g/(m²·d) (P7.1 and P7.2, respectively), reinforces the importance of co-metabolites (such as glucose) in sustaining high organic matter biodegradation rates for complex wastewaters.
These findings suggest that the overall sCOD removal efficiency in the MBBR system is significantly enhanced when a readily biodegradable carbon source is present alongside recalcitrant compounds, such as BPA. This observation highlights the reactor’s dependance on balanced organic loads to maintain optimal performance in terms of total carbon removal.
Bisphenol-A
Bisphenol-A was first fed into the reactor at the start of P2, on day 202. BPA influent concentrations (BPAin) ranged from 5 to 160 mgBPA/L, corresponding to the different phases shown in Fig. 4, which also shows the effluent BPA content (BPAef) and respective percentage removal. Throughout the 154 days of P2 (day 202 to 356), BPA inlet concentrations had an average of 4.38 ± 0.67 mg/L, with an average removal of 73 ± 20%. The degradation profile during P2 (Figure S2) shows a distinct acclimation period, where BPA degradation increased from an average of 52% removal during the first 57 days to 93% removal after day 325. The first 93 days of P2 served as an adaptation phase of the biomass to the BPA presence, with a steady increase of the BPA degradation until day 295, when 100% degradation was achieved and an average removal rate of 93 ± 4% was observed in the following 30 days.
Average bisphenol-A concentration in the influent and the effluent, and removal percentages for each phase during continuous operation of the MBBR.
After the 154 days of P2, the BPA influent concentration was raised to 10 mg/L during P3 (127 days), 30 mg/L during P4 (62 days), and 50 mg/L during P5 (73 days). The respective average BPA removal efficiencies were 98 ± 3%, 97 ± 5%, and 99 ± 0.5%. Subsequently, the inlet BPA concentration was increased to 100 mg/L in P6 (116 days) with an 89 ± 11% average removal, followed by the increase to 160 mgBPA/L in P7 (36 days), being the only organic matter source and achieving an average removal of 89 ± 10%.
Although the average removals for P6 and P7 are lower than those of the previous phases, this is due to an acclimation period when these phases started (P6.1, 98 days, and P7.1, 8 days), with mean BPA removals of 87% and 79%, respectively. After the acclimation in P6 and P7, removals as high as those of the previous phases were reached: 95% and 99% for P6.2 and P7.2, respectively. This temporary initial decrease in removal efficiency - which can be assumed due to toxic effects of increase in BPA concentration - followed by an increase in removal efficiency after an acclimation period indicated resilience and robustness by the biofilm.
The successful recovery of removal efficiency after initial declines in phases P6 and P7 reinforces the importance of gradually increasing BPA concentration to allow enzyme induction and microbial selection. In addition to co-metabolism, the implementation of feeding strategies, in which the pollutant load of an industrial effluent is increased in stages, in full-scale MBBRs can minimize transient inhibitory effects on heterotrophic and nitrifying ecosystems. This operational adjustment can ensure more stable performance under fluctuating industrial loads, guaranteeing good performance even with “shocks” of high concentrations in the reactor feed.
The greater extent of the initial decline in removal efficiency in P7 compared to P6 underscores the role of co-metabolism in sustaining high removal rates. While the removal of BPA reached 99% in P7.2, once adapted to the absence of a co-substrate, there was a significant drop in sCOD removal efficiency (from >80% to 47%) and the reduction in maximum volumetric removal rates, which suggests that cometabolism played a major role for high-rate performance in phases P2 to P6, when glucose was available as a co-substrate. The presence of a readily biodegradable substrate (glucose) likely supported higher biomass density and the induction of non-specific enzymes, such as oxygenases, which facilitated faster BPA degradation kinetics. Thus, while direct metabolism is possible after long-term acclimation, cometabolism remains the most efficient pathway for rapid pollutant degradation in mixed-stream scenarios.
Bisphenol-A is known to have toxic effects on microorganisms involved in biological processes depending on BPA concentration, sludge age, and biomass adaptation to the presence of BPA. Adapted biomass shows higher BPA removal efficiencies than those without any kind of acclimation, which also results in fewer negative effects on overall organic matter removal and nitrification (Seyhi et al. 2013, Eio et al. 2015, 2014, Ferrer-Polonio et al. 2021, Zielińska et al. 2014). Ferro Orozco et al. (2013) used activated sludge batch reactors to study the acclimation of microorganisms to BPA’s presence with two procedures, the first maintaining BPA concentration constant at 40 mg/L and the second doubling BPA concentration at each acclimation period from 40 mg/L to 360 mg/L. They found that increasing the concentration of BPA in steps favored the removal efficiency of BPA with less impact on biological processes, similarly to what is observed in this study, where increasing BPA concentration in steps results in high BPA removal.
The kinetics of BPA removal were also studied through the batch trials. Fig. 5 displays the results of the batch trials conducted from phases P2 to P7, corresponding to 5, 15, 30, 50, 100 and 160 mgBPA/L, respectively. During the batch trials of P2 and P3, above 95% removal was achieved after 180 min. In P4 (30 mgBPA/L), it took 210 min to achieve 97% removal; whereas for P5, 240 min had passed to reach 98%. However, for P6.1 (100 mgBPA/L), it took 360 min to attain a concentration reduction of only 74%, which increased to 95% during P6.2. Similar behavior was noticed in P7: only 79% in P7.1 and an increase to 99% in P7.2. The lower terminal BPA removal percentage for P6.1 and P7.1 are in line with the results observed during the acclimation period in continuous operation, as discussed above. This is possibly due to a toxic shock caused by the high BPA concentrations (Han et al. 2023, Tian et al. 2023), which becomes more likely during the batch trials since the biofilm is exposed to higher BPA concentrations than in continuous operation (where the actual BPA content within the reactor is the effluent concentration).
During the latest phase, BPA was the only organic substrate, showing a high degree of adaptation of the microorganisms to the BPA presence, still reaching high removal percentages. Despite that, it can be observed that at each phase, longer times were necessary to achieve high removal efficiencies, almost reaching a contact time equal to the MBBR HRT (6h). That is probably due to the complex nature of BPA and the degradation by-products that results in a cumulative effect slowing the removal rates at higher BPA concentrations. Additionally, the reduced availability of easily degradable organic matter could further affect the removal efficiency, resulting in progressively lower removal rates.
The maximum volumetric and superficial BPA removal rates calculated from the batch trials are listed along the apparent removal rates in Table IV. As the BPA feed concentration was raised, the maximum volumetric removal rates increased steadily from 0.04 kg/(m³·d) in P2 to 0.97 kg/(m³·d) during P7.2. The exceptions are the initial parts of phases P6 and P7 (P6.1 and P7.1), which had a reduction of the rate in comparison to those of the end of the previous phases. This is in line with the forementioned hypothesis of BPA toxic shock followed by acclimation in phases P6 and P7. Naturally, superficial BPA maximum removal rates followed a similar profile, increasing with the increase in BPA concentration, from 0.16 g/(m²·d) at 5 mgBPA/L to 3.9 g/(m²·d) at 160 mgBPA/L.
The results show that with an increase in BPA concentration, an increase in maximum removal rates also occurred. Consequently, the apparent removal rates rose as well since both the BPA removal capacity (maximum removal rate) and the load were increasing. The difference between the maximum and apparent removal rates decreased with the increasing BPA concentration, evidencing that the BPA load was rising in a greater proportion than the capacity.
Although direct microbial characterization through molecular techniques (e.g., 16S rRNA sequencing) was not conducted, the evolution of the kinetic parameters provides strong evidence of selective enrichment of the biofilm. The increase in the maximum BPA removal rates from phase P2 (0.04 kgBPA/(m³·d)) to P7 (0.97 kgBPA/(m³·d)) suggests that the biofilm community successfully adapted to high pollutant concentration by increasing the density of specialized BPA-degraders microorganisms. This functional adaptation is further supported by the observed shift in solids composition (section Solids), where Volatile Adhered Solids (VAS) increased in the final phases, indicating a regeneration of active biomass specifically acclimated to BPA as the sole carbon source.
Moreover, the BPA concentration data from the continuous operation were fitted to the Kincannon-Stover kinetic model, as detailed in the section Kincannon-Stover Kinetic Model, to determine the maximum substrate removal rate and biodegradability. As shown in Fig. 6 data from the initial 93 days of Phase P2 (represented by ○) were excluded from the linear regression because they represented a biological adaptation period characterized by unstable degradation rates that do not reflect steady-state kinetics.
Plot of BPA data from continuous operation according to the linearized Kincannon-Stover model. The first 93 days of P2 (○) were not considered for the determination of Kincannon-Stover parameters in the linear regression.
After disregarding this acclimation period, the model demonstrated a statistically robust fit with a correlation coefficient (R²) was higher than 0.99, with a root mean square error (RMSE) of 1.967 and a residual sum of squares (RSS) of 235.92 for n = 61 observations. These indicators, coupled with tight 95% confidence intervals for the slope (1.1 0066 ± 0.0136) and intercept (-1.04611 ± 0.4346), indicating that the model accurately captures the substrate utilization kinetics. The low RMSE relative to the magnitude of the experimental values (1/VRR) demonstrates the model’s high predictive power and the stability of the steady-state data collected throughout the operational phases.
The maximum substrate removal rate (Umax) and the saturation constant (KB) were determined to be 0.96 (CODeq = 2.42) and 1.05 (CODeq = 2.65) kgBPA/(m³·d), respectively. The estimated biodegradability () obtained was 91%, which is consistent with that observed during batch trials and continuous operation, as degradation percentages were consistently over 90% throughout the operation of the reactor, disregarding the initial acclimation periods in P2, P6 and P7. It is important to note that the 4-aminoantipyrine method employed targets the phenolic moieties of the molecule. Therefore, the removal efficiency reported reflects the degradation of the parent compound’s primary structure, representing the removal of the endocrine-disrupting core and possible phenolic byproducts, rather than complete mineralization to CO2 and H2O.
Furthermore, the convergence between the model-derived Umax (0.96 kgBPA/(m³·d)) and the experimental maximum rate measured in P7.2 (0.97 kgBPA/(m³·d)) validates that the reactor reached its maximum BPA removal capacity by the end of the study, with 160 mgBPA/L substituting the source of organic matter in the reactor.
Nitrogen
Throughout the 650 days of continuous operation, nitrogen conversion was monitored by analyzing both the influent and effluent streams of the reactor for ammonium, nitrate, and nitrite. The concentration results obtained are shown in Fig. 7 together with the ammonium percentage removal.
Average NH4+-N, NO2--N and NO3--N concentrations in the influent and the effluent, and removal percentages of NH4+-N for each phase during continuous operation of the MBBR.
The average removal of ammoniacal nitrogen (NH4+-N) remained stable at 96% throughout most of the operational phases. However, a significant temporary drop to 87% was observed during the initial stage of phase P7 (P7.1, days 734 to 741), coinciding with the total replacement of glucose by BPA as the sole carbon source. Following this critical acclimation period, the system demonstrated robust recovery, with NH4+-N removal efficiency returning to an average of 96% in the final stage (P7.2, days 741 to 770). This recovery is further evidenced by the kinetic data in Table V, showing the biofilm’s capacity to stabilize even under high chemical stress.
The average NH4+-N influent concentration throughout the continuous operation was 25 ± 3.8 mgN/L, while in the effluent it was 1.0 ± 1.5 mgN/L. The highest mean values for effluent ammonium concentrations were observed in P6 and P7, with 3.4 ± 1.6 and 3.9 ± 1.3 mgN/L, respectively. This reduction in NH4+-N removal efficiency can be explained by the toxic effects of high BPA concentrations on microbiological activity, such as inhibition of ammonia-oxidizing bacteria (AOB), as reported by Zielińska et al. (2014) and Dinçer (2021).
In the effluent, the NO2--N concentration throughout the continuous operation was lower than 0.6 mgN/L, averaging 0.3 ± 0.3 mgN/L, except for P6, for which nitrite nitrogen concentration achieved a maximum of 3.3 mgN/L, and 1.2 ± 1.6 mgN/L on average. Moreover, NO3--N had an overall average effluent concentration of 21 ± 7 mgN/L, apart from phase P6, which presented an average of 13 ± 1.1 mgN/L, with a maximum of 15 mgN/L. These results also show that higher BPA concentrations affect nitrification processes, slowing the complete oxidation to nitrate, accumulating nitrite and reducing the formation of nitrate.
Differently from what is observed from sCOD and BPA apparent removal rates, ammoniacal nitrogen apparent removal rates have insignificant variations between each of the operation phases (Table V), not having statistically significant difference (ANOVA, p-value = 0.000532, F = 4.610) in NH4+-N removal percentages between the phases without BPA (P1) and with BPA (P2 – P7) fed to the reactor.
Also, nitrate apparent formation rates (Table VI) decrease considerably with the increase in BPA concentrations, between phases P2 and P6, however, it increases during P7 to values similar to the ones observed in P2. The rapid consumption of nitrite generating nitrate during the P7 phase significantly increases the formation of nitrate.
Notwithstanding, the observed stability in NH4+-N apparent removal rates throughout different phases, despite the increase in BPA concentrations and the effects observed in sCOD removal, suggests a significant resilience of the nitrifying community. This consistency in removal efficiency, with an average of 96% throughout the operational phases, highlights the biofilm’s capacity to shield AOBs and other nitrifying microorganisms from the inhibitory effects of BPA (Sabater et al. 2002). This resilience in nitrogen removal indicates that nitrifiers, potentially located within the deeper layers of the biofilm, are less exposed to external pollutant stress, thus maintaining stable nitrification even under elevated BPA loads.
Interestingly, the stability of NH4+-N removal during the co-metabolic phases (P2-P6) further demonstrates that the addition of a co-substrate did not compromise nitrification performance. The fact that the lowest nitrogen removal efficiency (87%) occurred at the beginning of P7, when BPA was the sole substrate and its toxicity was less ‘diluted’ by the metabolic activity of the heterotrophic community, indicates that a co-substrate may exert a protective effect on nitrifiers by maintaining biofilm integrity and reducing direct toxic shock (Torresi et al. 2016, Piculell et al. 2016). This ability to support nitrifying bacteria even under toxic conditions is promising for the removal of ammoniacal nitrogen in wastewaters contaminated with complex recalcitrant compounds.
Maximum NH4+-N removal rates decrease significantly from phase P1 to phase P7. The maximum volumetric and superficial removal rates decreased over 87% from 0.55 kgN/m³·d and 2.2 gN/m²·d, respectively, during P1, to 0.07 kgN/m³·d and 0.28 gN/m²·d, during P7.2. Additionally, maximum superficial and volumetric NO3--N formation rates decreased initially by over 74% between P2 and P7. These results show a very clear impact on ammoniacal nitrogen removal rates, as well as the impacts on nitratation, due to very high concentrations of BPA in the influent. These results are to be expected, since BPA has been proved to harshly affect the biological nitrification process (Dinçer 2021, Zielińska et al. 2014).
The great difference between the maximum and apparent ammoniacal nitrogen removal rates reveal an overestimation of the HRT (6 h) up to phase P5. Although the maximum rates decreased progressively from phase to phase, suggesting some level of inhibitory effect on AOBs, the overestimated HRT ensured that the applied load remained low enough to prevent a noticeable loss in removal efficiency. Naturally, as the maximum rates got lower phase after phase, longer times were needed to achieve high removal percentages during the batch trials, as can be seen in Fig. 8a. While for P1 120 min were needed to achieve 100% removal, for P2 it took 160 min. Complete NH4+-N removal was reached in 160 min during P3 and P4, and in 180 min in P5. However, for P6 and P7, it was not observed complete NH4+-N removal in the time equivalent to the HRT (240 min), needing 270 min to achieve over 90% removal during P6.1, P6.2 and P7.1 batch trials, whereas for P7.2 300 min was needed.
Results of nitrogen transformation obtained from batch trials conducted at the end of each phase. a) NH4+-N; b) NO2--N; c) NO3--N.
As can be seen from Fig. 8b and Fig. 8c, during the batch trials of phases P6 and P7, the highest concentration of BPA resulted in a slower nitratation step, accumulating nitrite, whereas nitrate output was the lowest. The NO3--N output in the batch trials significantly decreased over the phases, from 21 mgN/L during P1 to 10 mgN/L during P7, following the increase in BPA concentration. This is not observed, however, throughout the continuous operation, where the output was constant with an average of 21 mgN/L throughout most of the phases, with the exception of phase P6, which yielded an average of 13 mgN/L, while in P7 it was 25 mgN/L. It may be theorized that the toxic effect of the BPA was amplified in the batch trials, as the biofilm was exposed to higher BPA bulk concentrations than during continuous operation (when the actual BPA concentration in the MBBR is the effluent concentration).
The observed fluctuations in nitratation rates further clarify the nature of BPA’s impact on nitrifying bacteria. As shown in Table VI, the maximum nitrate formation rate dropped in Phase P6 but nearly tripled by the end of Phase P7.2 (from 0.0563 to 0.1402 kgN/m³·d). This readjustment indicates that the inhibitory effect of high BPA concentrations was reversible and likely linked to metabolic competition or non-lethal enzymatic interference rather than direct cellular death. The biofilm’s ability to restore nitratation capacity even under a 160 mg/L BPA load suggests that the nitrifying community remained viable and protected within the biofilm matrix, successfully adapting to the chemical stress over time.
Despite these robust results, the use of synthetic wastewater lacks the complexity of real industrial matrices (e.g., heavy metals, salinity, and competing organic fractions), which may influence biofilm mass transfer and species competition in full-scale applications.
Solids
Solids concentrations were also monitored and analyzed throughout the operation of the reactor, following the methods and frequency described in Table II. The analysis comprised adhered solids, as volatile adhered solids (VAS) and fixed adhered solids (FAS), resulting in total adhered solids (TAS); and suspended solids, as volatile suspended solids (VSS) and fixed suspended solids (FSS), comprising the total suspended solids (TSS). The results are shown in Fig. 9, as the average concentrations of solids in each phase, along the VAS/TAS ratio and the percentage of adhered solids in relation to the total solids in the reactor (TAS/(TAS+TSS)).
TAS concentrations remained relatively stable during the operation with an average of 8.8 ± 1.3 g/m², with lowest concentration found during phase P3 (7.0 ± 2.0 g/m²). Furthermore, the ratio of VAS to TAS decreased between phases P2 and P3, reaching its lowest value, 0.65 during phase P3, showing a reduction in biofilm activity, likely due to the accumulation of inorganic materials and BPA degradation by-products, reflected by the increase in FAS during the same period. This was also observed by Zielińska et al. (2014) and Ferrer-Polonio et al. (2021), whose works show that BPA degradation can result in the formation of intermediate by-products that contribute to fixed solids. However, as the microbial community adapted to BPA in later phases, the concentration of FAS decreased, while VAS concentration increased, suggesting an enhanced capacity for BPA degradation and biofilm regeneration.
While TAS concentrations were high, TSS concentrations were, comparatively, very low throughout the operation of the MBBR, remaining stable with an average of 17.2 ± 2.1 g/m³. Furthermore, the TAS/(TAS+TSS) ratio remained constant and near 1 across all phases, showing a high capacity to retain biomass within the biofilm, ensuring that the majority of biological activity is confined within the biofilm and minimizing the release of suspended solids into the effluent.
The increase in FAS during certain phases, particularly under higher BPA concentrations, could potentially lead to mass transfer limitations if biofilm thickening is not properly managed. However, the system’s ability to stabilize both TAS and VAS after the initial phase of BPA exposure indicates that the MBBR can sustain high levels of BPA degradation without significant impacts on biofilm integrity.
CONCLUSIONS
This study investigated the efficacy of a Moving Bed Biofilm Reactor (MBBR) system in removing BPA from synthetic wastewater and assessed its impact on ammoniacal nitrogen and organic matter degradation. Across the phases, the reactor demonstrated a high capacity for organic matter removal, achieving average sCOD removal efficiencies of over 80% during phases P1 through P5. However, as BPA concentrations increased significantly in phases P6 and P7, with BPA comprising up to 100% of the organic matter in P7, sCOD removal efficiencies declined to 70% and 47%, respectively. This reduction is attributed to the recalcitrant nature of BPA and its by-products, which are known to resist biodegradation.
NH4+-N removal remained consistent, with an average efficiency of 96% across most phases. While a temporary decrease to 87% occurred during the initial acclimation to high BPA concentrations as the sole substrate (phase P7.1), the system demonstrated significant resilience, with removal efficiencies recovering to 96% in phase P7.2. This performance, despite the 87% reduction in maximum removal capacity, was sustained due to the conservative HRT employed, which provided a sufficient safety margin for nitrifying bacteria protected within the biofilm matrix.
The study also revealed that as BPA concentrations increased, both maximum and apparent BPA removal rates steadily increased together, with the maximum volumetric removal rates rising from 0.04 kg/m³·d in phase P2 to 0.97 kg/m³·d in phase P7.2. The latter agreed with the maximum removal rate derived from the Kincannon-Stover continuous operation kinetic model, which also revealed 91% biodegradability of the BPA. Despite the adverse effects on nitrification and organic matter removal at high BPA concentrations, the biofilm displayed resilience and an ability to adapt, suggesting the potential for co-metabolism strategies in full-scale applications to optimize biodegradation rates.
The study demonstrates that the MBBR system is robust for BPA removal and could be further optimized for complex wastewater streams. Key operational recommendations include the maintenance of co-metabolic conditions, the implementation of phased feeding strategies to prevent toxic shocks, and the selection of conservative HRT to safeguard nitrification stability.
Acknowledgements
This work was supported by Brazilian agencies CAPES (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior), CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico), FAPERJ (Fundação de Amparo à Pesquisa do Estado do Rio de Janeiro), and the Norwegian Research Council. The authors also acknowledge the infrastructure and support provided by the Water Pollution Control Laboratory (LabPol) at the Federal University of Rio de Janeiro (UFRJ). V.F. de Mello acknowledges the support of a CAPES scholarship. This work was supported by Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) – Finance Code 001, Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Fundação de Amparo à Pesquisa do Estado do Rio de Janeiro (FAPERJ), and the Norwegian Research Council, Grant 322529: Protecting Aquatic Ecosystem and Human Health From Micropollutants (PATCHER).
Data availability
References
- AMERICAN PUBLIC HEALTH ASSOCIATION (APHA), AMERICAN WATER WORKS ASSOCIATION (AWWA) & WATER ENVIRONMENT FEDERATION (WEF). 2023. 5220 D. Chemical Oxygen Demand (COD) – Closed Reflux, Colorimetric Method. In: Lipps WC et al. (Eds), Standard Methods for the Examination of Water and Wastewater, Washington, D.C.: American Public Health Association (APHA), p. 547-548.
- AMIN MM, BINA B, EBRAHIMI A, YAVARI Z, MOHAMMADI F & RAHIMI S. 2018. The occurrence, fate, and distribution of natural and synthetic hormones in different types of wastewater treatment plants in Iran. Chinese J Chem Eng 26: 1132-1139.
-
ASTM INTERNATIONAL. 2008. ASTM D1426-08: Standard Test Methods for Ammonia Nitrogen in Water. West Conshohocken, PA: ASTM International. https://doi.org/10.1520/D1426-08
» https://doi.org/10.1520/D1426-08 - BARNETT V, NETER J & WASSERMAN W. 1975. Applied Linear Statistical Models., 5th ed., USA: McGraw-Hill, 258 p.
- BASSIN JP & DEZOTTI M. 2008. Processos e técnicas para o controle ambiental de efluentes líquidos: Volume 5 da Série Escola Piloto de Engenharia Química, 1st ed., Rio de Janeiro: E-papers.
- BERGMAN Å, HEINDEL JJ, JOBLING S, KIDD KA & ZOELLER RT. 2013. State of the Science of Endocrine Disrupting Chemicals – 2012: Summary for Decision-Makers. United Nations Environ Program World Heal Organ.
- BILA DM & DEZOTTI M. 2007. Desreguladores endócrinos no meio ambiente: efeitos e conseqüências. Quim Nova 30: 651-666.
- BILAL M, JING Z, ZHAO Y & IQBAL HMN. 2019. Immobilization of fungal laccase on glutaraldehyde cross-linked chitosan beads and its bio-catalytic potential to degrade bisphenol A. Biocatal Agric Biotechnol 19: 101174.
- BUHARI J, HASAN HA, KURNIAWAN SB, ABDULLAH SRS & OTHMAN AR. 2023. Future and challenges of co-biofilm treatment on ammonia and Bisphenol A removal from wastewater. J Water Process Eng 54: 103969.
- BUSHNIK T, HAINES D, LEVALLOIS P, LEVESQUE J, VAN OOSTDAM J & VIAU C. 2010. Lead and bisphenol A concentrations in the Canadian population. Heal Reports 21: 7-18.
- CALAFAT AM, KUKLENYIK Z, REIDY JA, CAUDILL SP, EKONG J & NEEDHAM LL. 2005. Urinary concentrations of bisphenol A and 4-nonylphenol in a human reference population. Environ Health Perspect 113: 391-395.
- CHEN Y, GAO P, SUMME MJ, PHILLIP WA & WEI N. 2018. Biocatalytic membranes prepared by inkjet printing functionalized yeast cells onto microfiltration substrates. J Memb Sci 550: 91-100.
- COSTA EMF, SPRITZER PM, HOHL A & BACHEGA TASS. 2014. Effects of endocrine disruptors in the development of the female reproductive tract. Arq Bras Endocrinol Metabol 58: 153-161.
- DE LIMA HHC, LLOP MEG, MANIEZZO R dos S, MOISÉS MP, JANEIRO V, ARROYO PA, GUILHERME MR & RINALDI AW. 2021. Enhanced removal of bisphenol A using pine-fruit shell-derived hydrochars: Adsorption mechanisms and reusability. J Hazard Mater 416: 126167.
-
DEMENEIX B & SLAMA R. 2019. Endocrine Disruptors: from Scientific Evidence to Human Health Protection. Brussels: European Parliament, Policy Department for Citizens’ Rights and Constitutional Affairs. Available at: https://www.europarl.europa.eu/RegData/etudes/STUD/2019/608866/IPOL_STU(2019)608866_EN.pdf
» https://www.europarl.europa.eu/RegData/etudes/STUD/2019/608866/IPOL_STU(2019)608866_EN.pdf - DINÇER AR. 2021. Kinetic Modelling and Determination of Octyl Phenol Ethoxylate (OPE) and Bisphenol A (BPA) (used as plastic additives) Inhibition Constants for Nitrogen Conversion. Glob NEST J 23: 162-168.
- EIO EJ, KAWAI M, NIWA C, ITO M, YAMAMOTO S & TODA T. 2015. Biodegradation of bisphenol A by an algal-bacterial system. Environ Sci Pollut Res 22: 15145-15153.
- EIO EJ, KAWAI M, TSUCHIYA K, YAMAMOTO S & TODA T. 2014. Biodegradation of bisphenol A by bacterial consortia. Int Biodeterior Biodegrad 96: 166-173.
- ESCALONA I, FORTUNY A, STÜBER F, BENGOA C, FABREGAT A & FONT J. 2014. Fenton coupled with nanofiltration for elimination of Bisphenol A. Desalination 345: 77-84.
- ETTINGER MB, RUCHHOFT CC & LISHKA RJ. 1951. Sensitive 4-Aminoantipyrine Method for Phenolic Compounds. Anal Chem 23: 1783-1788.
- FERRER-POLONIO E, ALVIM CB, FERNÁNDEZ-NAVARRO J, MOMPÓ-CURELL R, MENDOZA-ROCA JA, BES-PIÁ A, ALONSO-MOLINA JL & AMORÓS-MUÑOZ I. 2021. Influence of bisphenol A occurrence in wastewaters on biomass characteristics and activated sludge process performance. Sci Total Environ 778: 146355.
- FERRO OROZCO AM, LOBO CC, CONTRERAS EM & ZARITZKY NE. 2013. Biodegradation of bisphenol-A (BPA) in activated sludge batch reactors: Analysis of the acclimation process. Int Biodeterior Biodegradation 85: 392-399.
- FUJIOKA T, KHAN SJ, MCDONALD JA & NGHIEM LD. 2015. Rejection of trace organic chemicals by a hollow fibre cellulose triacetate reverse osmosis membrane. Desalination 368: 69-75.
-
GAIOTO FC, MATHEUS MC, DE SOUZA-CHAVES BM, CAMPOS JC, BARRA TA, DE ALMEIDA AZEVEDO D, BASSIN JP & DEZOTTI M. 2023. Treatment of a pesticide-containing wastewater by biological and physicochemical processes: seeking the best conditions towards reuse. Brazilian J Chem Eng 41: 1109-1125. https://doi.org/10.1007/s43153-023-00394-z.
» https://doi.org/10.1007/s43153-023-00394-z -
HAN Y, DAI H, RONG X, JIANG H & XUE Y. 2023. Research Progress of Methods for Degradation of Bisphenol A. Molecules 28: 8028. https://doi.org/10.3390/molecules28248028
» https://doi.org/10.3390/molecules28248028 - HENZE M, LOOSDRECHT M van, EKAMA G & BRDJANOVIC D. 2008. Biological Wastewater Treatment: Principles, Modelling and Design, London: IWA Publishing.
- HOSSEINY SH & BORGHEI SM. 2002. Modeling of Organic Removal in a Moving Bed Biofilm Reactor (MBBR). Sci Iran 9: 53-58.
- HUNGE YM, YADAV AA, KHAN S, TAKAGI K, SUZUKI N, TESHIMA K, TERASHIMA C & FUJISHIMA A. 2021. Photocatalytic degradation of bisphenol A using titanium dioxide@nanodiamond composites under UV light illumination. J Colloid Interface Sci 582: 1058-1066.
- JUAN DA, HASAN HA, MUHAMAD MH, ABDULLAH SRS, BAKAR SNHA & BUHARI J. 2021. Physico-chemical and Biological Techniques of Bisphenol A Removal in an Aqueous Solution. J Ecol Eng 22: 136-148.
- KEYKAVOOS R, MANKIDY R, MA H, JONES P & SOLTAN J. 2013. Mineralization of bisphenol A by catalytic ozonation over alumina. Sep Purif Technol 107: 310-317.
- KOLOTI LE, GULE NP, AROTIBA OA & MALINGA SP. 2017. Laccase immobilized dendritic nanofibrous membranes as a novel approach towards the removal of bisphenol A. Environ Technol 39: 392-404.
- LANG IA, GALLOWAY TS, SCARLETT A, HENLEY WE, DEPLEDGE M, WALLACE RB & MELZER D. 2008. Association of urinary bisphenol A concentration with medical disorders and laboratory abnormalities in adults. J Am Med Assoc 300: 1303-1309.
- LIPPS WC, BRAUN-HOWLAND EB & BAXTER TE. 1999. Standard Methods for the Examination of Water and Wastewater, 20th ed., Washington, DC: APHA, AWWA, WEF, Method 2540 D: Total Suspended Solids Method, 2540 p.
- MATHEUS MC, LOURENÇO GR, SOLANO BA, DEZOTTI MWC & BASSIN JP. 2020. Assessing the impact of hydraulic conditions and absence of pretreatment on the treatability of pesticide formulation plant wastewater in a moving bed biofilm reactor. J Water Process Eng 36: 101243.
- MEEKER JD, CALAFAT AM & HAUSER R. 2010. Urinary bisphenol A concentrations in relation to serum thyroid and reproductive hormone levels in men from an infertility clinic. Environ Sci Technol 44: 1458-1463.
- MEEKER JD & FERGUSON KK. 2011. Relationship between Urinary Phthalate and Bisphenol A Concentrations and Serum Thyroid Measures in U.S. Adults and Adolescents from the National Health and Nutrition Examination Survey (NHANES) 2007–2008. Environ Health Perspect 119: 1396-1402.
- MELZER D, RICE NE, LEWIS C, HENLEY WE & GALLOWAY TS. 2010. Association of urinary bisphenol A concentration with heart disease: Evidence from NHANES 2003/06. PLoS ONE 5: 1-9.
- MU J, LI S, WANG J, LI X, CHEN W, TONG X, TANG Y & LI L. 2021. Efficient catalytic ozonation of bisphenol A by three-dimensional mesoporous CeOx-loaded SBA-16. Chemosphere 278: 130412.
- MUTSEYEKWA ME, DOĞAN Ş & PIRGALIOĞLU S. 2017. Ozonation for the removal of bisphenol A. Water Sci Technol 76: 2764-2775.
- NAIDU G, JEONG S, CHOI Y & VIGNESWARAN S. 2017. Membrane distillation for wastewater reverse osmosis concentrate treatment with water reuse potential. J Memb Sci 524: 565-575.
- ØDEGAARD H. 2006. Innovations in wastewater treatment: The moving bed biofilm process. Water Sci Technol 53: 17-33.
-
OUADA SB, ALI RB, LEBOULANGER C, ZAGHDEN H, CHOURA S, OUADA H Ben & SAYADI S. 2018. Effect and removal of bisphenol A by two extremophilic microalgal strains (Chlorophyta). J Appl Phycol 30: 1765-1776. https://doi.org/10.1007/s10811-017-1386-x.
» https://doi.org/10.1007/s10811-017-1386-x - PICULELL M, WELANDER P, JÖNSSON K & WELANDER T. 2016. Evaluating the effect of biofilm thickness on nitrification in moving bed biofilm reactors. Environ Technol 37: 732-743.
- SABATER S, GUASCH H, ROMANÍ A & MUÑOZ I. 2002. The effect of biological factors on the efficiency of river biofilms in improving water quality. Hydrobiologia 469: 149-156.
- SEYHI B, DROGUI P, BUELNA G & BLAIS JF. 2013. Biodegradation of Bisphenol-A in aerobic membrane bioreactor sludge. Water Sci Technol 68: 1926-1931.
- SHANKAR A & TEPPALA S. 2011. Relationship between Urinary Bisphenol A Levels and Diabetes Mellitus. J Clin Endocrinol Metab 96: 3822-3826.
-
STATISTICS KINGDOM. 2022. One-way ANOVA test calculator. Available at: https://www.statskingdom.com/180Anova1way.html Accessed on: January 8, 2025.
» https://www.statskingdom.com/180Anova1way.html - TIAN K, YU Y, QIU Q, SUN X, MENG F, BI Y, GU J, WANG Y, ZHANG F & HUO H. 2023. Mechanisms of BPA Degradation and Toxicity Resistance in Rhodococcus equi. Microorganisms 11: 67.
- TORRESI E, FOWLER JS, POLESEL F, BESTER K, ANDERSEN HR, SMETS BF, PLOSZ BG & CHRISTENSSON M. 2016. Biofilm thickness influences biodiversity in nitrifying MBBRs – Implications on micropollutant removal. Environ Sci Technol 50: 9279-9288.
- TUFAIL A, ALHARBI S, ALRIFAI J, ANSARI A, PRICE WE & HAI FI. 2021. Combining enzymatic membrane bioreactor and ultraviolet photolysis for enhanced removal of trace organic contaminants: Degradation efficiency and by-products formation. Process Saf Environ Prot 145: 110-119.
- US EPA - UNITED STATES ENVIRONMENTAL PROTECTION AGENCY. 1971a. Method 354.1: Nitrogen, Nitrite (Colorimetric, NED) by Spectrophotometer. Methods for Chemical Analysis of Water and Wastes. Cincinnati, OH: US EPA.
- US EPA - UNITED STATES ENVIRONMENTAL PROTECTION AGENCY. 1971b. Method 352.1: Nitrogen, Nitrate (Colorimetric, Brucine) by Spectrophotometer. Methods for Chemical Analysis of Water and Wastes. Cincinnati, OH: US EPA.
- VAN HAANDEL AC & VAN DER LUBBE JGM. 2012. Handbook of Biological Wastewater Treatment: Design and Optimisation of Activated Sludge Systems, 2nd ed., London: IWA Publishing.
- VISHNIAC W & SANTER M. 1957. The Thiobacilli. Bacteriol Rev 21: 195-213.
- WANG C-Y, ZHANG Y-J, WANG W-K, PEI D-N, HUANG G-X, CHEN J-J, ZHANG X & YU H-Q. 2018. Enhanced photocatalytic degradation of bisphenol A by Co-doped BiOCl nanosheets under visible light irradiation. Appl Catal B Environ 221: 320-328.
- WANG F, HUA J, CHEN M, XIA Y, ZHANG Q, ZHAO R, ZHOU W, ZHANG Z & WANG B. 2012. High urinary bisphenol A concentrations in workers and possible laboratory abnormalities. Occup Environ Med 69: 679-684.
- WANG G, HAMBLY AC, WANG G, TANG K & ANDERSEN HR. 2022. Engineered manganese redox cycling in anaerobic–aerobic MBBRs for utilisation of biogenic manganese oxides to efficiently remove micropollutants. Chem Eng J 446: 136998.
- WANG L, YUN J, ZHANG H, SI J, FANG X & SHAO L. 2021. Degradation of Bisphenol A by ozonation in rotating packed bed: Effects of operational parameters and co-existing chemicals. Chemosphere 274: 129769.
- XIAO S, ZHOU C, YE X, LIAN Z, ZHANG N, YANG J, CHEN W & LI H. 2020. Solid-Phase Microwave Reduction of WO₃ by GO for Enhanced Synergistic Photo-Fenton Catalytic Degradation of Bisphenol A. ACS Appl Mater Interfaces 12: 32604-32614.
- ZHANG W, SUN P, LIU D, ZHAO Q, ZOU B, ZHOU L & YE Z. 2021. Method to fabricate porous multifunction β-cyclodextrin modified resin for ultrafast and efficient removal of Cu(II) and bisphenol A. J Taiwan Inst Chem Eng 119: 286-297.
- ZHU Y, ZHU R, XI Y, XU T, YAN L, ZHU J, ZHU G & HE H. 2018. Heterogeneous photo-Fenton degradation of bisphenol A over Ag/AgCl/ferrihydrite catalysts under visible light. Chem Eng J 346: 567-577.
- ZIELIŃSKA M, BUŁKOWSKA K, CYDZIK-KWIATKOWSKA A, BERNAT K & WOJNOWSKA-BARYŁA I. 2016. Removal of bisphenol A (BPA) from biologically treated wastewater by microfiltration and nanofiltration. Int J Environ Sci Technol 13: 2239-2248.
- ZIELIŃSKA M, CYDZIK-KWIATKOWSKA A, BERNAT K, BUŁKOWSKA K & WOJNOWSKA-BARYŁA I. 2014. Removal of bisphenol A (BPA) in a nitrifying system with immobilized biomass. Bioresour Technol 171: 305-313.
- ZIELIŃSKA M, CYDZIK-KWIATKOWSKA A, BUŁKOWSKA K, BERNAT K & WOJNOWSKA-BARYŁA I. 2017. Treatment of Bisphenol A-Containing Effluents from Aerobic Granular Sludge Reactors with the Use of Microfiltration and Ultrafiltration Ceramic Membranes. Water Air Soil Pollut 228: 282.


















