Kinetic Effects of Ciprofloxacin, Carbamazepine, and Bisphenol on Biomass in Membrane Bioreactor System at Low Temperatures to Treat Urban Wastewater

This study analysed the kinetic results in the presence and absence of micropollutants (bisphenol A, carbamazepine, ciprofloxacin, and the mixture of the three compounds) obtained with respirometric tests with mixed liquor and heterotrophic biomass in a membrane bioreactor (MBR) working for two different hydraulic retention times (12–18 h) and under low-temperature conditions (5–8 °C). Independently of the temperature, the organic substrate was biodegraded faster over a longer hydraulic retention time (HRT) with similar doping, which was probably due to the longer contact time between the substrate and microorganisms within the bioreactor. However, low values of temperature negatively affected the net heterotrophic biomass growth rate, with reductions from 35.03 to 43.66% in phase 1 (12 h HRT) and from 37.18 to 42.77% in phase 2 (18 h HRT). The combined effect of the pharmaceuticals did not worsen the biomass yield compared with the effects caused individually.


Introduction
Today, society has a strong environmental conscience, and attempts are being made to minimise the impact of human activities on the environment [1]. For a resource as important to the population and human activities as water, the correct management and treatment of wastewater is fundamental. Inadequate treatment can release contaminants into the natural environment. Over the last few decades, there has been a rise in the number of emerging contaminants detected in water around the world [2]. These pollutants are pharmaceutically active compounds and endocrine-disrupting chemicals [3]. Pharmaceuticals may induce a variety of physiological changes, which can be reversible or not, in non-target aquatic organisms [4]. They can reach the aquatic environment via a high number of routes, such as discharge of both raw and wastewater treated by municipal [5], hospital [6], and industrial wastewater treatment plants (WWTPs); sewer leakage [7]; landfill leachate [8]; or treated wastewater used for irrigating [9]. These compounds can be found in both raw influents and treated effluents of WWTPs, so they are of particular interest [10]. However, existing treatment plants are not designed to remove such contaminants, leading to discharge into aquatic environments [11].
Within this group of contaminants are carbamazepine, bisphenol, and ciprofloxacin. These pharmaceuticals have been chosen for the study because of their widespread use worldwide, their detection in wastewater plants, and their presence on the European Union list of substances for monitoring purposes at the European level. whose pore size is 0.04 µm and whose area is 0.93 m 2 per membrane. Municipal wastewater from the primary settling tank of the Los Vados WWTP is the influent of the bioreactor. The entire system is schematically represented in Figure 1. The plant operates with combined cycles of filtration and backwashing with periods of 9.67 min and 0.33 min, respectively. Filtration is carried out from the outside to the inside by means of sucking. A recirculation pump from the membrane tank to the mixing tank allows a uniform MLSS concentration to be maintained. The membrane bioreactor is supplied with a constant air flux, and waste sludge is purged from the system.
The study was carried out during the months of December and January. Under the plant operating conditions of 12 HRT (flow rate of 7.02 L/h) and 18 HRT (flow rate of 4.67 L/h) and at low operating temperatures due to the climatic conditions of the region, no increase in the fouling of the membrane modules and transmembrane pressure was observed. A cleaning and recovery cycle of the membrane modules was not necessary. The membranes worked at low fluxes and low permeability. Table 1 shows the working permeability values, which were maintained at 1.29-2.22 L/(m 2 h bar). For the characterization of the pilot plant, influent and effluent samples were analysed (temperature, pH, conductivity, MLSSs, MLVSSs, BOD5, and COD). The results obtained are reported in Table 2. The plant operates with combined cycles of filtration and backwashing with periods of 9.67 min and 0.33 min, respectively. Filtration is carried out from the outside to the inside by means of sucking. A recirculation pump from the membrane tank to the mixing tank allows a uniform MLSS concentration to be maintained. The membrane bioreactor is supplied with a constant air flux, and waste sludge is purged from the system.
The study was carried out during the months of December and January. Under the plant operating conditions of 12 HRT (flow rate of 7.02 L/h) and 18 HRT (flow rate of 4.67 L/h) and at low operating temperatures due to the climatic conditions of the region, no increase in the fouling of the membrane modules and transmembrane pressure was observed. A cleaning and recovery cycle of the membrane modules was not necessary. The membranes worked at low fluxes and low permeability. Table 1 shows the working permeability values, which were maintained at 1.29-2.22 L/(m 2 h bar). For the characterization of the pilot plant, influent and effluent samples were analysed (temperature, pH, conductivity, MLSSs, MLVSSs, BOD 5 , and COD). The results obtained are reported in Table 2. In the effluent, values of suspended solids surpassing 5 mg/L were detected in some cases, which may have been due to the membrane backwash from the effluent tank during the different experimental phases.

Reagents
BPA was obtained from Sigma-Aldrich Co. (St. Louis, MO, USA). A solution of BPA (97%; CAS No. 80-05-7; MW: 228.29 g/mol) was prepared by dissolving 2 mg of the compound [28] in 2 mL of HPLC-grade methanol (Merck, Darmstadt, Germany) [25,37] and added to the respirometer during the homogenization phase of the doping assay. Moreover, 2 mL of methanol was added to the respirometer during the control homogenization assay to avoid side effects generated by methanol addition in the doping test.
A total of 1 mg of carbamazepine was added during the doping phase to reach a concentration of 1000 µg/L in the respirometer, and 0.1 mg of ciprofloxacin was added in the same way to maintain a concentration of 100 µg/L [38].
Mixture assays were carried out by adding the three above compounds at concentrations identical to those in the single-doping test.

Respirometric Assays
Eight different pairs of control and doping respirometric experiments were carried out. Each test of the pair had a homogenization phase, consisting in reaching an oxygen saturation state in the sample; an exogenous phase, in which the dynamic oxygen uptake rate was measured under stirring, recirculation, and aeration conditions; and an endogenous respirometric assay, which was performed to evaluate the decay coefficient by leaving the mixed liquor without aeration so that the dissolved oxygen concentration decreased to zero. They were carried out in two different phases, varying the HRT from 12 to 18 h, as reported in Table 3. It can be seen that under the operating conditions, the F/M ratios corresponded to low organic loads with respect to the concentration of microorganisms to compensate for the low-temperature effect.
Respirometric tests were carried out using a flowing gas/static liquid-type batch respirometer called BM-Advance. The respirometer worked at 20.0 ± 0.1 • C temperature, 7.25 ± 0.50 pH, 0.906 ± 0.001 L/min air flow rate, and 2000 rpm stirring rate, and recirculation was set up to assure homogenization in the respirometer. Mixed liquor samples of one litre were transferred from the pilot plant to the respirometer for each of the assay pairs.
Firstly, each sample was homogenized and aerated for 24 h before the assay started [39]. Then, the first exogenous control respirometric assay was carried out by adding the three known sodium acetate dilutions S 1 , S 2 , and S 3 , as shown in Figure 2. decreased to zero. They were carried out in two different phases, varying the HRT from 12 to 18 h, as reported in Table 3. It can be seen that under the operating conditions, the F/M ratios corresponded to low organic loads with respect to the concentration of microorganisms to compensate for the low-temperature effect.
Respirometric tests were carried out using a flowing gas/static liquid-type batch respirometer called BM-Advance. The respirometer worked at 20.0 ± 0.1 °C temperature, 7.25 ± 0.50 pH, 0.906 ± 0.001 L/min air flow rate, and 2000 rpm stirring rate, and recirculation was set up to assure homogenization in the respirometer. Mixed liquor samples of one litre were transferred from the pilot plant to the respirometer for each of the assay pairs.
Firstly, each sample was homogenized and aerated for 24 h before the assay started [39]. Then, the first exogenous control respirometric assay was carried out by adding the three known sodium acetate dilutions S1, S2, and S3, as shown in Figure 2. Afterwards, an endogenous respiration assay without air flow was carried out, resulting in a graph similar to that in Figure 3. Afterwards, an endogenous respiration assay without air flow was carried out, resulting in a graph similar to that in Figure 3.
The R s rate is the rate of consumption in exogenous assays in the presence of a substrate and a constant supply of O 2 , while the OUR rate is the rate of consumption in the absence of a substrate and without O 2 supply.
Finally, 2 identical assays were carried out after doping the pertinent compounds into the same sludge sample.  The Rs rate is the rate of consumption in exogenous assays in the presence of a substrate and a constant supply of O2, while the OUR rate is the rate of consumption in the absence of a substrate and without O2 supply.
Finally, 2 identical assays were carried out after doping the pertinent compounds into the same sludge sample.

Kinetic Modelling
In order to model the kinetics of the analysed biomass, the yield coefficient of heterotrophic biomass referred to oxygen (YH), the maximum specific growth rate of heterotrophic biomass (µmax), the half-saturation coefficient of organic matter (KM), and the decay coefficient of heterotrophic biomass (bH) were calculated following the kinetic parameter estimation of heterotrophic biomass [28]. This estimation consisted of eight different steps.
Step 1: By integrating the Rs from Equation (1), oxygen consumption (OC) was determined for each of the three additions of sodium acetate dilutions (S1, S2, and S3).
where fcv is a conversion factor of 1.48 mgCOD/mgVSS.
Step 3: µemp can be obtained from the relation between the biomass growth rate and the substrate degradation rate, as Equation (3) shows.
Step 4: By linearizing the Monod model, KM and µm,H were assessed according to Equation (4).
Step 5: The estimation of bH was performed according to Equation (5).
OURend Figure 3. Evolution example of static oxygen uptake rate (OUR) in endogenous respirometric assay. OUR end : maximum rate of consumption in endogenous respiration assay.

Kinetic Modelling
In order to model the kinetics of the analysed biomass, the yield coefficient of heterotrophic biomass referred to oxygen (Y H ), the maximum specific growth rate of heterotrophic biomass (µ max ), the half-saturation coefficient of organic matter (K M ), and the decay coefficient of heterotrophic biomass (b H ) were calculated following the kinetic parameter estimation of heterotrophic biomass [28]. This estimation consisted of eight different steps.
Step 1: By integrating the R s from Equation (1), oxygen consumption (OC) was determined for each of the three additions of sodium acetate dilutions (S 1 , S 2 , and S 3 ).
where f cv is a conversion factor of 1.48 mgCOD/mgVSS.
Step 3: µ emp can be obtained from the relation between the biomass growth rate and the substrate degradation rate, as Equation (3) shows.
Step 4: By linearizing the Monod model, K M and µ m,H were assessed according to Equation (4).
Step 5: The estimation of b H was performed according to Equation (5).
where (1 − fp) is the fraction of volatile biomass (mgVSS/mgTSS) and OUR end is the endogenous oxygen uptake rate.
Step 6: The assessment of kinetic parameters at working temperature was performed by applying the Metcalf equation [40], as shown in Equation (6).
where r T and r 20 are the kinetic parameters at working temperature and at 20 • C, respectively; θ is a fitting standard parameter with a value of 1.04 for the MBR; and T is the working temperature.
Step 7: The evaluation of the substrate degradation rate of organic matter removal, r su,H , was performed according to Equation (7).
Step 8: The estimation of net heterotrophic biomass growth rate, r' x,H , based on the biomass growth rate and the decay rate, was performed as shown in Equation (8).

Dynamic and Static Oxygen Uptake Rates
The exogenous respiration test results of operation phases 1 (12 h HRT) and 2 (18 h HRT) are included in Figures 4 and 5, respectively.
In operation phase 1, the presence of BPA or ciprofloxacin did not modify the duration of the respirometric assays. On the other hand, the shock of carbamazepine increased the duration of the respirometric tests from 3000 to 3500 s. The mix of emerging pollutants showed the same effect as carbamazepine, extending the duration from 2600 to 3000 s approximately, which implies that there was more time for the heterotrophic biomass to degrade the organic matter substrate. This suggests a higher influence of carbamazepine on the duration of the kinetic experiments.
In relation to operation phase 2, both carbamazepine and ciprofloxacin caused an increase in the duration of the respirometric tests, around 500 s each. However, BPA did not change the duration of these experiments. The same trend was observed with the mix of emerging compounds, which suggests a higher effect of BPA over longer HRTs.
where (1 − fp) is the fraction of volatile biomass (mgVSS/mgTSS) and OURend is the endogenous oxygen uptake rate.
Step 6: The assessment of kinetic parameters at working temperature was performed by applying the Metcalf equation [40], as shown in Equation (6).
where rT and r20 are the kinetic parameters at working temperature and at 20 °C, respectively; θ is a fitting standard parameter with a value of 1.04 for the MBR; and T is the working temperature.
Step 7: The evaluation of the substrate degradation rate of organic matter removal, rsu,H, was performed according to Equation (7).
Step 8: The estimation of net heterotrophic biomass growth rate, r'x,H, based on the biomass growth rate and the decay rate, was performed as shown in Equation (8).

Dynamic and Static Oxygen Uptake Rates
The exogenous respiration test results of operation phases 1 (12 h HRT) and 2 (18 h HRT) are included in Figures 4 and 5, respectively.   In operation phase 1, the presence of BPA or ciprofloxacin did not modify the duration of the respirometric assays. On the other hand, the shock of carbamazepine increased the duration of the respirometric tests from 3000 to 3500 s. The mix of emerging pollutants showed the same effect as carbamazepine, extending the duration from 2600 to 3000 s approximately, which implies that there was more time for the heterotrophic biomass to degrade the organic matter substrate. This suggests a higher influence of carbamazepine on the duration of the kinetic experiments.
In relation to operation phase 2, both carbamazepine and ciprofloxacin caused an increase in the duration of the respirometric tests, around 500 s each. However, BPA did not change the duration of these experiments. The same trend was observed with the mix of emerging compounds, which suggests a higher effect of BPA over longer HRTs.
Regarding the values of Rs, the pharmaceuticals reduced the three maximum values of Rs of the respirometric tests, with the exception of ciprofloxacin, in phase 1. This effect was not observed in phase 2, which could have been due to the fact that operation over a longer HRT compensated the possible reduction in Rs. In light of this, the addition of ciprofloxacin even favoured the dynamic oxygen uptake rate.
In comparing the Rs values of both operation phases, it should be highlighted that phase 1 assays in the presence of pollutants had lower Rs values than phase 2 essays, which could have been due to the better operation conditions corresponding to secondphase samples, i.e., a longer HRT and slightly higher MLSS concentrations (Figures 4 and  5).
The endogenous respiration tests of the eight pairs of experiments are included in Figures 6 and 7. Regarding the values of R s , the pharmaceuticals reduced the three maximum values of R s of the respirometric tests, with the exception of ciprofloxacin, in phase 1. This effect was not observed in phase 2, which could have been due to the fact that operation over a longer HRT compensated the possible reduction in R s . In light of this, the addition of ciprofloxacin even favoured the dynamic oxygen uptake rate.
In comparing the R s values of both operation phases, it should be highlighted that phase 1 assays in the presence of pollutants had lower R s values than phase 2 essays, which could have been due to the better operation conditions corresponding to second-phase samples, i.e., a longer HRT and slightly higher MLSS concentrations (Figures 4 and 5).
The endogenous respiration tests of the eight pairs of experiments are included in Figures 6 and 7.
In general, the presence of pollutants in both phases tended to decrease the maximum value of the OUR, i.e., the endogenous oxygen uptake rate (OUR end ), compared with the values corresponding to the control tests. Additionally, as previously observed for R s , the OUR end values in the presence of the emerging pollutants were higher in phase 2 than in phase 1, which might have been caused by the most favourable operation conditions of phase 2 (a longer HRT and higher MLSS concentrations).
The different trends of R s and OUR were considered in order to evaluate the kinetic parameters following the mathematical procedure described in Equations (1)-(8).

Heterotrophic Kinetic Modelling
The kinetic parameters of heterotrophic biomass in the absence and presence of pollutants calculated following the kinetic modelling steps described in the Kinetic Modelling section are indicated in Tables 4 and 5 at respirometry temperature (20 • C) and at operation temperature, respectively.   In general, the presence of pollutants in both phases tended to decrease the maximum value of the OUR, i.e., the endogenous oxygen uptake rate (OURend), compared with the values corresponding to the control tests. Additionally, as previously observed for Rs, the OURend values in the presence of the emerging pollutants were higher in phase 2 than in phase 1, which might have been caused by the most favourable operation conditions of phase 2 (a longer HRT and higher MLSS concentrations).
The different trends of Rs and OUR were considered in order to evaluate the kinetic parameters following the mathematical procedure described in Equations (1)-(8).

Heterotrophic Kinetic Modelling
The kinetic parameters of heterotrophic biomass in the absence and presence of pollutants calculated following the kinetic modelling steps described in the Kinetic Modelling section are indicated in Tables 4 and 5 at respirometry temperature (20 °C) and at operation temperature, respectively.  In addition, Figure 8 shows the variables r su,H and r' x,H , which encompass the rest of the kinetic parameters, in the absence and presence of the emerging pollutants at 20 • C. In addition, Figure 8 shows the variables rsu,H and r'x,H, which encompass the rest the kinetic parameters, in the absence and presence of the emerging pollutants at 20 °C.  In addition, Figure 9 shows the variables r su,H and r' x,H , which encompass the rest of the parameters, in the absence and presence of the emerging pollutants at operation temperature.
In phase 1, independently of the temperature, the maximum specific growth rate of heterotrophic biomass decreased by 62.26% and 49.82% in the presence of BPA and ciprofloxacin, respectively. However, this kinetic parameter increased by 21.32% when carbamazepine was added. In the presence of the mix of emerging pollutants, µ m was higher (37.37%), which suggested a higher effect of carbamazepine. In phase 2, all contaminants reduced the µ m values by 28.86% to 62.24%, with a reduction of 38.56% in the presence of the mix of emerging pollutants. Consequently, a longer HRT did not improve the maximum specific growth rate. Moreover, operation at low values of temperature (Table 3) decreased the values of the maximum specific growth rate by 37.54% to 44.47% in phase 1 and 42.25% to 44.47% in phase 2. The overall trend of reduction in µ m implied that the heterotrophic biomass required more time to oxidize organic matter in the presence of emerging pollutants and at lower temperatures.
K M showed a similar trend in both phases. In phase 1, we observed a reduction in the half-saturation coefficient for organic matter in the presence of BPA (69.99%) and ciprofloxacin (49.25%), with increase percentages of 27.16% and 127.49% with carbamazepine and the mix of emerging compounds, respectively. In phase 2, K M was reduced in the presence of pollutants by 29.26% to 77.52% with the individual shocks and by 54.21% with the mix of pollutants. In this case, operation over a longer HRT seemed to improve the half-saturation coefficient of organic matter. The effect of the drop in temperature was identical to that obtained for the maximum specific growth rate with the same reduction percentages. The global pattern of the reduction in K M could indicate that less available substrate was required to reach µ m , which indicates that the system was not inhibited by the substrate. In phase 1, independently of the temperature, the maximum specific growth rate o heterotrophic biomass decreased by 62.26% and 49.82% in the presence of BPA and cipro loxacin, respectively. However, this kinetic parameter increased by 21.32% when carbam azepine was added. In the presence of the mix of emerging pollutants, µm was highe (37.37%), which suggested a higher effect of carbamazepine. In phase 2, all contaminant reduced the µm values by 28.86% to 62.24%, with a reduction of 38.56% in the presence o the mix of emerging pollutants. Consequently, a longer HRT did not improve the max mum specific growth rate. Moreover, operation at low values of temperature ( Table 3 decreased the values of the maximum specific growth rate by 37.54% to 44.47% in phas 1 and 42.25% to 44.47% in phase 2. The overall trend of reduction in µm implied that th heterotrophic biomass required more time to oxidize organic matter in the presence o emerging pollutants and at lower temperatures. KM showed a similar trend in both phases. In phase 1, we observed a reduction in th half-saturation coefficient for organic matter in the presence of BPA (69.99%) and cipro loxacin (49.25%), with increase percentages of 27.16% and 127.49% with carbamazepin and the mix of emerging compounds, respectively. In phase 2, KM was reduced in th Regarding the amount of heterotrophic biomass produced per substrate oxidized, represented by the yield coefficient, the results do not show any clear trend, with increases ranging from 1.88% to 1.98% with BPA and from 0.56% to 1.55% with the mix of emerging pollutants in both phases. Carbamazepine exerted a negative effect on this parameter, with reduction percentages of 1.38-2.58%, whereas ciprofloxacin increased Y H over the 12 h HRT (0.12%) and reduced it over the 18 h HRT (6.31%). Thus, the influence of the different emerging pollutants on this kinetic parameter was almost negligible. Nevertheless, a clear reduction in the amount of heterotrophic biomass produced per organic matter oxidized was observed at operation temperature, with reduction percentages similar to those obtained for µ m and K M .
Finally, the b H values decreased in the presence of pollutants by 24.76% to 66.51% with the individual shocks and by 15.19% with the mix of contaminants in phase 1. However, the trend changed in phase 2 with ciprofloxacin and the mix of emerging pollutants, with increases of 24.53% and 17.89%, respectively, which suggests a higher effect of the HRT with these shocks. Therefore, the biomass decay rate for heterotrophic biomass was reduced in the presence of pollutants with the exception of ciprofloxacin and the mix of emerging pollutants over the 18 h HRT. These latter two cases indicate a higher quantity of biomass being oxidized per day, which leads to higher loss of cell mass in the presence of ciprofloxacin, and its mix with BPA and carbamazepine over longer HRTs. As seen with µ m , the temperature also reduced the decay rate for heterotrophic bacteria, with reduction percentages similar to those observed at 20 • C.
The effect of these variations in the kinetic parameters that characterize the heterotrophic bacteria in the membrane bioreactor are included in the values of the substrate degradation rate of organic matter removal and the net heterotrophic biomass growth rate (Tables 4 and 5). Both r su,H and r' x,H showed trends similar to those observed for µ m and K M . In particular, in phase 1, r su,H decreased by 61.27% and 62.03% in the presence of BPA at 20 • C and 5 • C, respectively, and lessened by 46.10% and 47.54% in the presence of ciprofloxacin at 20 • C and 7 • C, respectively. For its part, carbamazepine caused increases in r su,H of 21.87% and 22.81% at 20 • C and 8 • C, respectively. The same pattern occurred with the mix of the three emerging pollutants, with increases in r su,H of 31.23% and 33.41% at 20 • C and 6 • C, respectively, indicating a possible higher effect of carbamazepine in the mix of contaminants. In phase 2, over a longer HRT, all contaminants reduced the degradation rate of organic matter by 21.25% to 57.43% at 20 • C and 22.44% to 59.69% at 5-6 • C, with reductions of 35.94% at 20 • C and 37.47% at 5 • C in the presence of the mix of the emerging pollutants. On the other hand, no significant differences were observed when the temperature decreased from 20 • C to the operation values, with slight rises from 0.90% to 7.58%. It must be highlighted that independently of the temperature, organic matter was degraded faster over a longer HRT with similar doping due to the longer contact time between the substrate and microorganisms within the bioreactor.
Regarding the behaviour of r' x,H , the pattern was similar to that described for r su,H , with increases in r' x,H of 58.25% and 61.08% at 20 • C and 6 • C in the presence of the mix of pollutants in phase 1. Nevertheless, in the presence of the mix of contaminants, over the 18 h HRT, the net heterotrophic biomass growth rate lessened by 43.84% and 45.19% at 20 • C and 5 • C, respectively. This could be based on lower biomass activity during operation over the 18 h HRT. Furthermore, operation at low values of temperature (Table 3) decreased the values of the net heterotrophic biomass growth rate by 35.03% to 43.66% in phase 1 and 37.18% to 42.77% in phase 2, which indicates an evident effect of temperature on r' x,H .
Concerning the synergic effect of the three compounds, it was not observed on r su,H and r' x,H , independently of the temperature, since the values of these kinetic rates were included in the ranges obtained with the individual shock of each emerging pollutant. It is important to point out that the behaviour of the heterotrophic biomass was contrary to that observed by other authors in the presence of BPA [28]. In particular, at temperatures of 12.1-31.1 • C, the above authors obtained r su,H values varying from 50 to 200 mgO 2 /(L·h) and r' x,H ranging from 20 to 170 mgVSS/(L·h) in the presence of BPA, which exceeded the values corresponding to the absence of this compound. In this study, the r su,H values varied from 18.6876 to 45.9095 mgO 2 /(L·h), and r' x,H ranged from 4.0965 to 14.5019 mgVSS/(L·h), which highlights a higher effect of very low temperatures on heterotrophic bacteria [35,36]. In light of this, [38] obtained that the presence of a mix of carbamazepine, ciprofloxacin, and ibuprofen almost doubled the r su,H of the control experiment in a membrane bioreactor, with values fluctuating from 183.97 to 192.88 mgO 2 /(L·h) at temperatures varying from 18.9 to 21.4 • C, and the same trend was obtained by [41] in a membrane bioreactor at 12.6 and 21.5 • C. The authors of [29] did not observe any inhibition of COD removal under exogenous respiration in the presence of carbamazepine at 20 • C and similar MLSS and HRT conditions. This strengthens the higher influence of very low temperatures on the kinetic behaviour of heterotrophic bacteria, which could hinder the acclimatization of bacterial populations to the presence of micropollutants.

Conclusions
Based on the kinetic results in the presence and absence of micropollutants (bisphenol A, carbamazepine, ciprofloxacin, and the mixture of the three above compounds) obtained with respirometric tests carried out on heterotrophic biomass in a membrane bioreactor working over two different hydraulic retention times (12 and 18 h), at low-temperature conditions (5-8 • C), and at low MLSS concentrations (1633-2866 mg/L), the following conclusions were drawn:  Data Availability Statement: Not applicable.

Conflicts of Interest:
The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.