Thông tin tài liệu
membranes Article Effect of Operational Parameters on the Removal of Carbamazepine and Nutrients in a Submerged Ceramic Membrane Bioreactor Khanh-Chau Dao 1,2 , Chih-Chi Yang , Ku-Fan Chen 1 * Citation: Dao, K.-C.; Yang, C.-C.; Chen, K.-F.; Tsai, Y.-P Effect of Operational Parameters on the Removal of Carbamazepine and Nutrients in a Submerged Ceramic Membrane Bioreactor Membranes and Yung-Pin Tsai 1, * Department of Civil Engineering, National Chi Nan University, Nantou Hsien 54561, Taiwan; daokhanhchau07@gmail.com (K.-C.D.); chi813@gmail.com (C.-C.Y.); kfchen@ncnu.edu.tw (K.-F.C.) Department of Health, Dong Nai Technology University, Bien Hoa 810000, Dong Nai, Vietnam Correspondence: yptsai@ncnu.edu.tw; Tel.: +886-49-2910960 (ext 4121) Abstract: Pharmaceuticals and personal care products have raised significant concerns because of their extensive use, presence in aquatic environments, and potential impacts on wildlife and humans Carbamazepine was the most frequently detected pharmaceutical residue among pharmaceuticals and personal care products Nevertheless, the low removal efficiency of carbamazepine by conventional wastewater treatment plants was due to resistance to biodegradation at low concentrations A membrane bioreactor (MBR) has recently attracted attention as a new separation process for wastewater treatment in cities and industries because of its effectiveness in separating pollutants and its tolerance to high or shock loadings In the current research, the main and interaction effects of three operating parameters, including hydraulic retention time (12–24 h), dissolved oxygen (1.5–5.5 mg/L), and sludge retention time (5–15 days), on removing carbamazepine, chemical oxygen demand, ammonia nitrogen, and phosphorus using ceramic membranes was investigated by applying a two-level full-factorial design analysis Optimum dissolved oxygen, hydraulic retention time, and sludge retention time were 1.7 mg/L, 24 h, and days, respectively The research results showed the applicability of the MBR to wastewater treatment with a high carbamazepine loading rate and the removal of nutrients Keywords: full-factorial design; carbamazepine; membrane bioreactor; hospital wastewater; operating parameters 2022, 12, 420 https://doi.org/ 10.3390/membranes12040420 Academic Editors: Michael O Daramola and Ahmad Fauzi Ismail Received: 14 March 2022 Accepted: April 2022 Published: 14 April 2022 Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations Copyright: © 2022 by the authors Licensee MDPI, Basel, Switzerland This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/) Introduction In recent years, pharmaceuticals and personal care products (PPCPs) have caused growing concerns as emerging contaminants because of their extensive use, presence in aquatic environments, and potential impacts on wildlife and humans PPCPs comprise a large and varied group of organic compounds, including pharmaceutical drugs and components of daily personal care products (soaps, lotions, toothpaste, fragrances, sunscreens, etc.) as well as their metabolites and transformation products that are widely used in large quantities around the world [1] Previous studies showed that many PPCPs were frequently detected in surface water, groundwater, seawater, and drinking water [2–4] Pharmaceuticals and other drug components are widely used in hospitals, so hospital effluents generally have higher detection rates and concentrations of these contaminants [5,6] One of the main environmental problems caused by hospital effluents is their discharge into urban sewerage systems, without preliminary treatment, and entering into the water bodies Therefore, treatment plants should be upgraded to eliminate these PPCPs to the greatest possible extent before their effluents are released into the environment In recent investigations, different antibiotics were found in low concentrations in municipal wastewater effluents and surface water [7–9] Antibiotics in water are an environmental concern because they could disturb microbial ecology, increase the proliferation of antibiotic-resistant pathogens, and threaten human health Membranes 2022, 12, 420 https://doi.org/10.3390/membranes12040420 https://www.mdpi.com/journal/membranes Membranes 2022, 12, 420 of 15 Carbamazepine (CBZ) is a common drug that controls seizures, with about 1014 tons of it consumed worldwide annually Overdosing on CBZ and its metabolites, on the other hand, can harm the human liver and emopoietic systems As a result, it was the most common pharmaceutical residue found in water bodies [10–12] Investigations found that CBZ was persistent with removal efficiency of the WWTPs being mostly below 10% [13], due to its resistance to biodegradation at low concentrations and its less attachment to sludge [14] Researchers considered that CBZ could be a “witness molecule” confirming the presence and persistence of drugs in water bodies [15] In recent years, membrane technology has attracted attention as a new separation process for water and wastewater treatment in cities and industries Combining membranes with biological treatments is an attractive technique and has resulted in a new concept: a membrane bioreactor (MBR) An MBR was first used to treat wastewater over 50 years ago [16] Submerged MBRs (SMBR), characterized by immersing the membrane modules as separation units directly in the bioreactor, were developed for wastewater treatment in the 1990s [17] Membrane filtration processes are promising alternatives for eliminating PPCPs from wastewater [18,19] Although their effectiveness in the separation of pollutants, tolerance to high or shock loadings, stable and excellent effluent quality, ease of operation, small footprint, and effective bacterial elimination, they are currently facing some research and developmental challenges, such as membrane fouling, high membrane cost, and the need for pre-treatment [20] The current research investigated operating parameters (factors) via FFD experiments to optimize the MBR process First, the important operating conditions, such as hydraulic retention time—RT, dissolved oxygen—DO, and sludge retention time—SRT, were chosen to study their main and interaction effects on the following responses: CBZ, chemical oxygen demand—COD, ammonia nitrogen—NH4 + -N, and phosphorus—PO4 3− -P removal In addition, the ranges of the factors were selected based on the capability of the experimental setup, economic considerations, and membrane operating limits Finally, a regression model was presented for each response, and optimization of the process was carried out to maximize the removal of CBZ, COD, NH4 + -N, and PO4 3− -P Materials and Methods 2.1 Chemicals Chemicals used to prepare synthetic wastewater were analytical grade, acetonitrile (HPLC grade), acetone, ethyl acetate, and methanol (HPLC grade) purchased from SigmaAldrich, St Louis, MO, USA An Oasis HLB cc Vac Cartridge was supplied by Waters, Milford, MA, USA CBZ was of the highest purity commercially available, and it was purchased from Sigma-Aldrich Ultra-pure water was prepared with a Milli-Q water purification system (resistivity of 18.2 MΩ cm at 25 ◦ C) All solvents for HPLC application were filtered before use with the 0.45-µm membrane filter paper (Millipore, Merck, Darmstadt, Germany) and degassed by ultrasonication for 30 before use Stock solutions of CBZ were prepared in Milli-Q water from powdered substances at g/L The stock solution was made weekly from powdered substances and stored in the dark at ◦ C 2.2 Simulated Wastewater Synthetic hospital wastewater was composed of the following (g/L): peptone 0.12, meat extract 0.083, NH4Cl 0.143, NaCl 0.005, CaCl2 ·2H2O 0.003, MgSO4 ·7H2O 0.0015, CuCl2 ·2H2O 50 × 10−6, K2HPO4 ·3H2O 0.084, C6H12O6 0.19, and NaHCO3 0.83 This resulted in concentrations of COD 485.82 ± 57.11 mg/L, NH4 +-N 60.72 ± 8.67 mg/L, PO4 3− -P of 14.74 ± 1.09 mg/L, and a pH of 7.7 ± 0.2 CBZ was spiked into synthetic wastewater at 100 µg/L Membranes 2022, 12, x FOR PEER REVIEW Membranes 2022, 12, 420 of 15 of 15 in concentrations of COD 485.82 ± 57.11 mg/L, NH4+-N 60.72 ± 8.67 mg/L, PO43−-P of314.74 ± 1.09 mg/L, and a pH of 7.7 ± 0.2 CBZ was spiked into synthetic wastewater at 100 μg/L 2.3 Bioreactor BioreactorConfiguration Configurationand andOperation Operation 2.3 Thesubmerged submergedMBR MBRoperated operated a working volume 25 L flat-sheet ceramic The in ainworking volume of 25of L flat-sheet ceramic memmembrane module (GVE Environmental Co., Ltd., Taoyuan, Taiwan) with a nominal brane module (GVE Environmental Co., Ltd., Taoyuan, Taiwan) with a nominal pore size pore of 0.1anμm and anarea effective area 0.25 m1.2, The Figure Al2O 3-based flat-sheet of 0.1 size µm and effective of 0.25 m2 ,ofFigure Al21 O3The -based flat-sheet ceramic ceramic membrane had three layers: a surface layer,layer, a transition layer, layer, and aassupport membrane had three layers: a surface layer, a transition and a support shown layer, as shownelectron in the microscopy scanning electron microscopy (SEM) Materials) image The membrane was in the scanning (SEM, see Supplementary image The memoperated an on–off (8 on and off cycle) thetomembrane brane wasunder operated under mode an on–off mode (8 on and mintooffrelax cycle) relax the module The influent flow rate was adjusted equal thetoeffluent flow rate toflow maintain membrane module The influent flow rate wastoadjusted equal the effluent rate toa constant water level The systems were controlled automatically by timers and a pressure maintain a constant water level The systems were controlled automatically by timers and a gauge Air diffusers were positioned at the bottom of the reactor and the rear the pressure gauge Air diffusers were positioned at the bottom of the reactor and theend rearofend of the membrane module for aeration andscouring air scouring while air supply controlled membrane module for aeration and air while the the air supply waswas controlled by by airflow meter anan airflow meter Figure1.1.Schematic Schematicdiagram diagramof ofthe theMembrane Membranebioreactor bioreactor(MBR) (MBR)experimental experimentalsetup setup Figure The Thetransmembrane transmembranepressure pressure(TMP) (TMP)was wasmeasured measuredusing usingaapressure pressuregauge gaugeinstalled installed between and thethe permeate pump TheThe pressure gauge recorded the betweenthe themembrane membranemodule module and permeate pump pressure gauge recorded TMP daily.daily At the each membrane cleaning was performed The memthe TMP Atend theofend of experiment, each experiment, membrane cleaning was performed The brane module was was flushed withwith tap tap water to remove thethe visible membrane module flushed water to remove visiblecake cakelayer layerand andthen then immersed for a minimum of 24 h in a sodium hypochlorite solution of ‰ (v/v) Activated immersed for a minimum of 24 h in a sodium hypochlorite solution of ‰ (v/v) Actisludge was removed from the reactor during during chemical cleaning operations vated sludge was removed from the reactor chemical cleaning operations The The seed-activated seed-activated sludge sludge was was collected collected from from aaconventional conventionalwastewater wastewater system system at at National University The Theratio ratioofofthe themixed mixedliquor liquorvolatile volatilesuspended suspended solids National Chi Chi Nan Nan University solids to to mixed liquor suspended solids (MLVSS/MLSS) of the seed-activated sludge mixed liquor suspended solids (MLVSS/MLSS) of the seed-activated sludge waswas 0.8 0.8 AcActivated sludge was maintained in a batch reactor at a DO of mg/L, HRT of h, tivated sludge was maintained in a batch reactor at a DO of mg/L, HRT of 48 h, and48SRT and SRT of 20 days The components in the synthetic wastewater were adjusted to maintain of 20 days The components in the synthetic wastewater were adjusted to maintain a aBOD: BOD:N: N:PPratio ratioof of100:15:5 100:15:5 In In addition, addition, nitrogen nitrogen and and phosphorus phosphorus were were put put in in as as excess excess into the synthetic wastewater, so it was not deficient in essential nutrients for bacterial activity After each experiment, an amount of sludge was added to the MBR reactor to reach a concentration of 5000 mg/L and acclimatized for days before the experiment Membranes 2022, 12, Membranes x FOR PEER 2022, REVIEW 12, x FOR PEER REVIEW of 15 Membranes 2022, 12, Membranes x FOR PEER 2022, REVIEW 12, x FOR PEER REVIEW of 15 Membranes 2022, 12, Membranes x FOR PEER 2022,REVIEW 12, x FOR PEER REVIEW of 15 Membranes 2022, 12, 420 Factor Name A DO B HRT C SRT into the synthetic intowastewater, the syntheticsowastewater, it was not deficient so it wasinnot essential deficient nutrients in essential for bacterial nutrients for bac of 15 activity After activity each experiment, After each an experiment, amount of sludge an amount was of added sludge to was the MBR added reactor to thetoMBR into the synthetic intowastewater, the syntheticsowastewater, it was not deficient so it wasinnot essential deficient nutrients in essential for bacterial nutrients forreac bac reach a concentration reach a of concentration 5000 mg/L and of 5000 acclimatized mg/L and for acclimatized days before for the days experiment before the experime activity activity eachwastewater, experiment, After each an experiment, anit amount of added sludge to was theessential MBR added to thetoMBR into the After synthetic into the synthetic so wastewater, itamount was notofdeficient sosludge waswas in notessential deficient nutrients in forreactor bacterial nutrients forreac bac reach a concentration reach a of concentration 5000 mg/L and of 5000 acclimatized mg/L and for acclimatized days before for the days experiment before the experime activity Afterby activity each experiment, After each anexperiment, amount of an sludge amount wasofadded sludge to was the MBR addedreactor to the to MBR reac 2.4 Modeling Full-Factorial Design (FFD) Design 2.4 Modeling by2.4 Full-Factorial Modeling byDesign Full-Factorial (FFD) (FFD) reach a concentration reach a concentration of 5000 mg/L and of 5000 acclimatized mg/L andfor acclimatized days before for the days experiment before the experime Inthis this work, work, the FFDwork, was employed toDesign identify theto crucial factors, the possibility possibility of possibi 2.4 Modeling by2.4 Full-Factorial Modeling byDesign Full-Factorial (FFD)was (FFD) In the In FFD this was employed the FFD to identify employed the crucial identify factors, the crucial the factors, the of estimating interactions, and optimizing the parameters The HRT (12 and 24 h), DO (1.5 and estimating estimating and interactions, optimizing and the parameters the The parameters HRT (12 The HRT h), DO (12 and (1.5 h), DO 2.4 Modeling by 2.4 Full-Factorial Modeling by Design Full-Factorial (FFD) Design (FFD) In thisinteractions, work, the In FFD this work, was employed the FFD was tooptimizing identify employed the to crucial identify factors, theand crucial the24 possibility factors, the of 24possibi 5.5 mg/L), andand SRT5.5 (5 mg/L), and 15 and days) were set, as indays) Table Experiments were1.carried out at were ca and 5.5 mg/L), and SRT (5 and 15 days) SRT (5 were and set, 15 as in were Table set, Experiments as in Table were Experiments carried estimating interactions, estimating and interactions, optimizing and the optimizing parameters the The parameters HRT (12 and The 24 HRT h), DO (12 and (1.5 24 h), D In this work, the In this FFD work, was employed theNH FFD+ -N, was to and identify employed to crucial identify factors, the crucial the possibility factors, the of possibi 3the − -P ambient temperature CBZ, COD, concentrations were analyzed NH PO +-N,PO 3−4-P +-N, 3− out at ambient out temperature at ambient CBZ, temperature COD, CBZ, COD, NH and concentrations and PO -P concentrations were anawere and 5.5 mg/L), and and5.5 SRT mg/L), (5 and and 15 days) SRT (5 were and set, 15 days) as in were Table set, 1.HRT Experiments as in were Experiments carried estimating interactions, estimating and interactions, and the optimizing parameters the The parameters (12Table and Thewas 24 HRT h), (12 DO and (1.524were h), Dc All samples were performed inoptimizing triplicate, and the average standard deviation calculated + 3− + 3− lyzed All samples lyzed were All performed samples were in triplicate, performed and in the triplicate, average and standard the average deviation standard was deviatio out at5.5ambient out temperature at ambient CBZ, temperature COD, NH CBZ, COD, NH 415 -N, and PO 4set, -P -N,concentrations and PO4 -P concentrations were ana- were were and mg/L), and and 5.5 SRT mg/L), (5 and and 15 SRT days) (5 were and set, days) as in were Table Experiments as in Table were Experiments carried c for each sample Each experiment operated for five days calculated for each calculated sample for Each each experiment sample Each operated experiment for five operated days for five days + 3− + 3− lyzed All samples lyzed were All performed samples were in triplicate, performed and in the triplicate, average and standard the average deviation standard was deviatio out at ambientout temperature at ambient CBZ, temperature COD, NH CBZ, COD, NH4 -P -N, and PO -N,concentrations and PO4 -P concentrations were anawere calculated for each calculated sample forEach eachexperiment sample Each operated experiment foraverage five operated days foraverage fivedeviation days lyzed.1.All samples lyzed were All for samples performed were in triplicate, performed and in triplicate, the and standard the standard wasdeviatio Table Factors and levels full-factorial design (FFD) Table Factors Table and levels Factors for full-factorial and levels for design full-factorial (FFD) design (FFD) calculated for calculated each sample forEach each experiment sample Each operated experiment for five operated days for five days TableUnits Factors Table and levels Factors for full-factorial andMinimum levels for design full-factorial (FFD) design (FFD) Type Coded Low Coded High High Factor NameFactor Units Name TypeUnits Minimum Type Maximum Maximum MinimumCoded Maximum Low Coded Coded Low Coded H Table Factors Table and levels Factors for full-factorial and levels for design full-factorial (FFD) design (FFD) A mg/LName DO Factor A mg/L DO Numeric Numeric 1.5Type Maximum 5.51.5 5.5 −1 1.5 Coded +1 −1 ↔High 5.5 1.5 Coded +1 ↔H Numeric 1.5 5.5 Factor Units Name Typemg/L Units Minimum Minimum Coded Maximum Low Coded Low − 1↔ 1.5 +1↔ 5.5 B B h Name HRT Numeric h 12 Numeric 12Type Maximum 241.5 12 2412 12 24 12 A h HRT DO Factor A mg/L DO Numeric mg/L Numeric 1.5 5.5 −1 1.5 +1 −1 ↔High 5.5 1.5 +1 ↔H Numeric 245.5 Factor Name Units TypeUnits Minimum Minimum Coded Maximum Low Coded Coded Low Coded − 1↔ +1↔ 24 C SRT C days SRT Numeric days Numeric 15 15 −1 ↔ +1 −1 ↔ ↔ 15 +1 ↔ BAdays HRT B h HRT Numeric h Numeric 12 24 12 24 −1 ↔ 12 +1 −1 ↔ ↔ 24 12 +1 ↔ ↔ DO A mg/L DO Numeric mg/L5 1.5 1.5 −1 1.5 5.5 +1 Numeric 15 5.5 −1 5.551.5 +1 15 CB SRT C days SRT Numeric days Numeric 15 15 5two-level +1 −1↔ ↔ 15 to protect HRT consists B h method HRT Numeric h center 12 2412 24 12 −1 +1 ↔ 12 24 +1 ↔c The method The of adding consists of adding points to center the two-level points−1to↔ FFD the to protect FFD curvaC SRT C days SRT Numeric days Numeric 15 15 −1 ↔ +1 −1 ↔ ↔ 15 +1 be ↔c ture and ture an consists independent and anestimate independent ofofthe estimate error [21] of This the error method could also method beFFD easily could also Theallow method of center points toto the two-level FFD toThis protect curvature The consists The allow method ofadding adding consists center adding points center the two-level points to[21] FFD the two-level to protect curvato protect and and allow an respond independent of the error [21] This method could also beFFD easily upgraded to upgraded toofestimate respond designs surface for further designs optimizations for further optimizations [22] The regression [22] regre ture ture an independent and allow anestimate independent of estimate error [21] of This the error method could This also method be easily could also bec Theallow method consists The surface method adding consists center ofthe adding points to center the two-level points to[21] FFD the two-level to protect curvatoThe protect upgraded to respond surface designs for further optimizations [22] The regression equation equation based equation on the based first-order on the model first-order with three model parameters with three and parameters their interaction and their intera upgraded to respond upgraded surface to respond designs surface for further designs optimizations for further optimizations [22] The regression [22] The regre ture and allowture an independent and allow anestimate independent of theestimate error [21] of This the error method [21].could This method also be easily could also be based on the model with three parameters and their interaction terms could terms could befirst-order terms given could in the form given ofthe the in the following form ofthree expression the following [23]: expression [23]: equation based equation on the based first-order on model first-order with model parameters three and parameters their interaction and their upgraded to upgraded respond surface tobe respond designs surface for further designs optimizations forwith further optimizations [22] The regression [22].be The inter regre given in the form of the following expression [23]: terms could be terms given could in the be form given of the in the following form of expression the following [23]: expression [23]: equation based equation on the based first-order on the model first-order with three model parameters with three and parameters their interaction and their inter Yi = b0 + b1X1i +Ybi2=X2ib0++bb3X 1X 3i1i++bb 122X X1i2iX+2ib+3X b3i13X + 1ibX123iX+1iX b2i23X + 2ibX133iX+1iX b3i123+Xb1i23 XX 2iX 2iX 3i 3i + b123 (1) X1iX2iX3i terms could be terms given could in1i +the be form given of in the the following form of expression the following [23]: expression [23]: Y Y i = b + b X b i = X 2i b + + b b X X 3i 1i + + b b 12 X X 1i 2i X + 2i b + X b 3i 13 X + 1i b X 12 3i X + 1i X b 2i 23 X + 2i b X 13 3i X + 1i X b 3i 123 + X b 1i 23 X X 2i X 2i X 3i 3i + b 123 (1) Xcorrespon 1iX2iX3i Yi =Yibis b1where X +ji values b3response; X3i +(jb=121, XX1i2, X3; X3i2, +3;b8)23 X1, + the b1238) X1iindicate X2i X3i the (1) 0+ 1i + bY iXis 2iXthe 2i + 2i 2, 3i 3, where the response; ji values i = 13 1,(jX2, =1i1, 3, …, i =indicate …, corresponding Y Y b0 +coded b1YXi1iis+X bjii 2=values X bresponse; 2i0 ++b bcoded b031(j X X3i=1ithe +1, +X bforms; baverage 12 2X2i X b3i2, 13 +X1, 1iX 12 X 3i the 1i +8) Xbi2i23=indicate +X1,value b 2iX 13X 1i +bX1of bthe band 1i23 X X2i2iXbX3i33iare +bb1the 123 X 1i X2iXb3i3 a parameters ini =their parameters forms; in their is value is of average result; ,3i123 b+the 2X ,8) result; ,the b(1) 2,correspo and where response; 2, ji values 3;1iX+i 2i=b+031, (j =the 3,b 2, …, 3; 2,3i 3, …, corresponding indicate where Y Yii is is the thewhere response; Xthe ji values (j = 1, 2, 3; i = 1, 2, 3, , 8) indicate the corresponding , ji bvalues 13, bb 23 12forms; b 13,represent ,value b the 123 interaction represent the interaction coefficients linear linear and coefficients; b12X and b1, parameters parameters coded forms; in their coded 0, is average of average the of , coefficients bthe 28) , corresponding and result; b3 are b[23] 1the ,the b2,correspo and b3 a wherecoefficients; Yi is in where Yi is the response; the response; (jand =the X,b ji 2,123 values 3; ibb =2301, (jisand 2, =the 1, 3, 2, …, 3; 8) i =result; indicate 1,value 2, 3,b1…, the indicate parameters intheir their coded forms; b is the average value of the result; b1 , b2 , and b3 are Adding interaction Adding terms interaction to the main terms effects to the introduced main effects curvature introduced into curvature the response into the resp ,b btheir 13, bb 23 and ,baverage 123 b13,represent bb230 ,is and b the interaction represent linear coefficients; coefficients; b12in and b12forms; parameters in linear parameters theirand coded forms; coded 0, is the value the average of123the result; valuebof 1the , coefficients bthe 2,interaction and result; b3 are b[23] 1, the bcoefficients 2, and b3 a the linear coefficients; and 12 , b13 , b23 , and b123 represent the interaction coefficients [23] function Therefore, function if there Therefore, was slight if there curvature was slight in a curvature limited region, in a limited a first-order region, model a first-order Adding interaction Adding terms interaction main the main introduced curvature the response into the resm 12to , the bthe 13, main band 23terms , and ,to bb123 13 , represent bintroduced 23 , andeffects bcurvature the 123 curvature represent interaction the coefficients interaction [23] coefficients linear coefficients; linear and coefficients; bto b12effects Adding interaction terms effects introduced into® into the response func® 11 with interactions with was interactions appropriate was for appropriate modeling [24,25] for modeling Design-Expert [24,25] Design-Expert 11 was utilized to was function Therefore, function ifterms there Therefore, was if there curvature in a curvature limited region, in a limited a into first-order region, model awith first-order Adding interaction Adding interaction to theslight main terms effects towas the main effects curvature introduced curvature the response into theutiliz resm tion Therefore, if there was slight curvature in slight aintroduced limited region, a first-order model ® ® design the experiments, design the and experiments, analysis of and variance analysis (ANOVA) of variance was (ANOVA) used to analyze was used the to reanalyze ® with interactions with was interactions appropriate was for appropriate modeling [24,25] for modeling Design-Expert [24,25] Design-Expert 11 was utilized 11 to was utili function Therefore, function if there Therefore, was slight if there curvature was slight in acurvature limited region, in11awas limited a first-order region, model a first-order tm interactions was appropriate for modeling [24,25] Design-Expert utilized to design sults sults ® ® to design the experiments, design the and experiments, analysis of and variance analysis (ANOVA) of variance was (ANOVA) used to analyze was used the rewith interactions with was interactions appropriate for appropriate modeling [24,25] for modeling Design-Expert 11the was utilized 11 toanalyze was utilit the experiments, and analysis ofwas variance (ANOVA) wasDesign-Expert used [24,25] to analyze results sults sults the and design the experiments, design experiments, analysis of and variance analysis (ANOVA) of variance was(ANOVA) used to analyze was used thetore-analyze t 2.5 Analytical 2.5 Analytical Methods 2.5 AnalyticalMethods Methods sults sults 2.5 Analytical 2.5 Analytical Methods Standard analytical Standard methods analytical [26] were methods applied [26] in were determining inCOD, determining MLSS,MLSS, and COD, MLSS StandardMethods analytical methods [26] were applied in applied determining COD, 3−-P COD 3−measured − PO PO was -P COD was by the measured colorimetric by the method colorimetric in the method presence in of the potassium presence of MLSS potas 2.5 Analytical 2.5 Methods Analytical Methods Standard analytical Standard methods analytical [26] methods applied [26]method in were determining applied inCOD, determining and COD, and PO was measured by thewere colorimetric in the presence ofMLSS, potassium -P COD 3− 3− dichromate, and dichromate, the absorbance and the was absorbance measured was at 600 measured nm using at a 600 UV nm spectrometer using a UV (DR spectromete PO PO -P CODand was Standard measured -P.absorbance COD was by was the measured colorimetric the method colorimetric in themethod inofthe potassium presence pota dichromate, the measured at[26] 600 nmdetermining using apresence UV (DR Standard analytical methods analytical [26] methods werebyapplied were in applied in spectrometer COD, determining MLSS, COD, and ofMLSS +-N + -N 5000, Hach, CO, 5000, USA) Hach, NH CO, USA) NH 4and was measured 4+-N was by the measured indophenol by the method indophenol [27] method [27] 3− 3− 5000, Hach, CO, USA) NH was measured by the indophenol method [27] dichromate, and dichromate, the absorbance the was absorbance measured was at 600 measured nm using at a 600 UV nm spectrometer using a UV (DR spectromete PO4 -P COD PO was -P measured COD was by measured the colorimetric by themethod colorimetric in themethod presence in of thepotassium presence of pota +-N USA) +in Before Before suspended extraction, solids suspended samples solids inwere the samples removed were by filtering removed the by filterin Before extraction, suspended solids the samples were removed by filtering the 5000, Hach, extraction, CO, 5000, USA) Hach, NH CO, NH 4and was measured 4in -Nthe was by the measured indophenol byat the method indophenol [27] [27] dichromate, and dichromate, the absorbance the was absorbance measured was at 600 measured nm using 600 a UV nm spectrometer using amethod UV (DR spectromete samples through samples a 0.45-μm through glass-fiber a 0.45-μm filter glass-fiber (Millipore, filter Merck, (Millipore, Darmstadt, Merck, Germany) Darmstadt, Germ samples through a 0.45-µm glass-fiber filter (Millipore, Merck, Darmstadt, Germany) Next, + + Before suspended suspended solids inwere the removed were by filtering removed theby filterin 5000,Before Hach,extraction, CO, 5000, USA) Hach, NHextraction, CO, NH -NUSA) wassolids measured 4in -Nthe was bysamples the measured indophenol bysamples the method indophenol [27] method [27] Next, CBZ was Next, extracted CBZ was from extracted the water from samples the water using samples a selected using cartridge a selected Before cartridge B CBZ was extracted from the water samples using a selected cartridge Before loading the samples through samples aBefore 0.45-μm through glass-fiber a 0.45-μm filter (Millipore, (Millipore, Darmstadt, Merck, Germany) Darmstadt, Germ Before extraction, suspended extraction, solids suspended inglass-fiber the samples solidsfilter inMerck, were the samples removed were by filtering removed the by filterin loading the sample, loading the the solid-phase sample, the adsorbent solid-phase was adsorbent preconditioned was preconditioned with mL of methawith mL of m sample, the solid-phase adsorbent was preconditioned with 5a(Millipore, mL ofDarmstadt, methanol followed by Next, CBZ was Next, extracted CBZ was from extracted water from samples the water using samples selected using cartridge a selected Before cartridge B samples through samples a 0.45-μm through glass-fiber athe 0.45-μm filter glass-fiber (Millipore, filter Merck, Merck, Germany) Darmstadt, Germ nol followed by nol mL followed of Milli-Q by water mL of Milli-Q The sample water was The then sample passed was through then passed the cartridge through the cart 5Next, mL ofCBZ Milli-Q water The sample was then passed through the cartridge at a mL/min loading the sample, loading the the solid-phase sample, the adsorbent solid-phase was adsorbent preconditioned was preconditioned with mL of methawith mL of m was Next, extracted CBZ was from extracted the water from samples the water usingsamples a selected using cartridge a selected Before cartridge B at a followed rate mL/min flow at 5rate mL/min Subsequently, flow rate Subsequently, the cartridge was the eluted cartridge with was five eluted 1of with aliquots fiveof 15mL aliqu flow Subsequently, the cartridge was eluted with five 1passed mL aliquots ethyl acetate– nol by nol 5amL followed ofthe Milli-Q by water mL ofadsorbent Milli-Q The sample water was The then sample was through then passed the cartridge through the carm loading the sample, loading the solid-phase sample, the solid-phase was adsorbent preconditioned was preconditioned with 5mL mL of methawith mL of ethyl acetate–acetone ethyl acetate–acetone (50:50, v/v) at (50:50, a rate v/v) of at mL/min; a rate of the combined mL/min; the aliquots combined were aliquots acetone (50:50, v/v) at a rate of mL/min; the combined aliquots were evaporated under at a 5followed mL/minby flow at 5a followed 5rate mL/min flow rate Subsequently, the was thethen eluted cartridge with was five eluted mL with aliquots fiveof mL nol nol mL of Subsequently, Milli-Q by mL water of Milli-Q Thecartridge sample water was The sample passed was through then passed the cartridge through thealiqu car aatgentle flow of high purity nitrogen and redissolved in 1cartridge mL methanol The evaporated under evaporated gentle under flow of high purity flow nitrogen high purity and redissolved nitrogen inwith redissolved 1analyses mL of inaliqu m ethyl ethyl acetate–acetone (50:50, v/v)arate atgentle (50:50, aSubsequently, rate v/v) of 1of at mL/min; awas rate of the1ofcombined mL/min; the aliquots combined were aliquots a acetate–acetone mL/min at flow a 5arate mL/min Subsequently, flow the cartridge the eluted with was five eluted 1and mL aliquots five 1ofmL of CBZ were carried out on an Agilent 1200 HPLC equipped with a G1329 autosampler, methanol The methanol analyses of The CBZ analyses were carried of CBZ out were on an carried Agilent out 1200 on an HPLC Agilent equipped 1200 HPLC with equipped evaporated under evaporated a acetate–acetone gentle under flow gentle high purity flow high and redissolved nitrogen the and incombined redissolved mL of aliquots in m ethyl acetate–acetone ethyl (50:50, v/v)aofat (50:50, a rate v/v) of 1of atnitrogen mL/min; a ratepurity of the1 combined mL/min; aliquots were a G1329 G1315D diode array detector, and a carried G1316A column oven (Agilent Technologies Co Ltd.,equipped amethanol autosampler, a G1329 a autosampler, G1315D diode a G1315D array detector, diode array and a detector, G1316A and column a G1316A oven (Agcolumn The methanol analyses of The CBZ analyses were of CBZ out were on an carried Agilent out 1200 on an HPLC Agilent equipped 1200 HPLC with evaporated under evaporated a gentle under flow aofgentle high flow purityofnitrogen high purity and nitrogen redissolved andinredissolved mL of inoven m Clara, CA, USA) The detection wavelength was 210 nm, andThe the column temperailent Technologies ilent Co Technologies Ltd., Santa Co Clara, Ltd., CA, Santa USA) Clara, The CA, detection USA) wavelength detection was wavelength 210 equipped wa aSanta G1329 autosampler, a G1329 a autosampler, G1315D diode a G1315D array detector, diode array and a detector, G1316A and column a G1316A oven (Agcolumn oven methanol Themethanol analyses of The CBZ analyses were carried of CBZ out were on carried an Agilent out 1200 on an HPLC Agilent equipped 1200 HPLC with ◦ C An Eclipse XDB-C18 column (4.6 × 150 mm, particle size five µm, ture was set at 30 nm, and the column nm, and temperature the column was temperature set at 30 °C was An set Eclipse at 30 °C XDB-C18 An Eclipse column XDB-C18 (4.6 × 150 column (4.6 ilent Technologies Co Technologies Ltd., Santadiode Co Clara, CA, Santa USA) Clara, The CA, detection USA) The wavelength detection was wavelength 210 wa a G1329 autosampler, ailent G1329 aautosampler, G1315D aLtd., G1315D array detector, diode array and adetector, G1316A and column a G1316A oven (Agcolumn oven Agilent) was used for separation The mobile phase was acetonitrile–water (31:69, v/v) at a mm, particle size mm, particle μm, Agilent) size five was μm, Agilent) for separation was used for The separation mobile phase The was mobile phase wa nm, the column nm,five and temperature the column was temperature setused at 30 °C was An set Eclipse at 30 °C XDB-C18 An Eclipse column XDB-C18 (4.6 ×ace150 column (4.6 ilentand Technologies ilent Co Technologies Ltd., Santa Co Clara, Ltd., CA, Santa USA) Clara, The CA, detection USA) The wavelength detection was wavelength 210 wa mL/min flow rate The injection volume was 20 separation µL tonitrile–water tonitrile–water (31:69, v/v) a(31:69, mL/min v/v) flow atAgilent) 1rate mL/min The injection flow volume The injection was 20volume μL was 20 (4.6 μL mm, particle size mm, five particle μm, Agilent) size five was μm, for was used for The separation mobile phase The was mobile phase wa nm, and the column nm, and temperature theat column was temperature setused ata30 °C was An setEclipse at 30rate °C XDB-C18 An Eclipse column XDB-C18 (4.6 ×ace150 column tonitrile–water tonitrile–water (31:69, v/v) a(31:69, mL/min v/v) at a 1rate mL/min The used injection flow rate volume The injection was 20volume μL was 20 μL mm, particle size mm, five particle μm,atAgilent) size five was μm,flow used Agilent) for separation was for The separation mobile phase The mobile was acephase wa tonitrile–watertonitrile–water (31:69, v/v) at a(31:69, mL/min v/v) at flow a rate mL/min The flow injection rate.volume The injection was 20volume μL was 20 μL Membranes 2022, 12, 420 of 15 Results and Discussion 3.1 Biodegradation Efficiency and CBZ Removal Table gives an overview of the results achieved by the MBR treatment The system showed a relatively low efficiency in removing CBZ with an average removal of 18.41% due to its recalcitrance, but higher than previous studies by MBR using synthetic wastewater, which was commonly below 13% [28–30] Table CBZ, COD, ammonia, and phosphorus removal efficiency Removal (%) Minimum Maximum Average CBZ COD NH4 + -N PO4 3− -P 9.04 69.23 79.75 −16.87 38.36 99.37 99.71 −5.91 18.42 86.45 90.55 −10.15 However, the study results also demonstrated the effectiveness of the MBR system in removing COD and NH4 + -N, with an average removal efficiency of 86.45 and 90.55%, respectively On the other hand, the experimental results showed that the MBR system could not remove phosphorus effectively when the outlet concentration was higher than the inlet 3.2 Model Fitting and Statistical Analysis A total of 11 experiments were performed using a three-factor two-level FFD with three replicates at the center point Table shows the experimental design matrix, and the results of the response variables studied Table Experimental design table for the factors and responses Factor Factor Factor Std Run A:DO B:HRT C:SRT 10 11 3 10 11 mg/L 3.5 1.5 5.5 5.5 5.5 5.5 3.5 1.5 3.5 1.5 1.5 h 18 24 24 24 12 12 18 12 18 12 24 days 10 15 15 15 10 10 15 Response Response Response Response CBZ removal % 17.66 ± 3.33 19.65 ± 7.18 17.23 ± 5.38 14.75 ± 4.48 9.04 ± 1.12 13.67 ± 4.81 16.20 ± 4.00 24.12 ± 1.45 18.48 ± 4.35 13.39 ± 10.31 38.36 ± 4.49 COD removal % 87.18 ± 3.88 82.16 ± 4.97 99.37 ± 0.42 95.97 ± 0.99 85.87 ± 2.37 89.46 ± 1.40 87.62 ± 3.23 77.49 ± 1.46 89.54 ± 0.22 69.23 ± 2.56 87.07 ± 0.54 Ammonia removal % 91.62 ± 6.04 90.40 ± 1.11 99.58 ± 0.25 99.71 ± 0.03 89.38 ± 3.65 85.81 ± 1.38 90.35 ± 2.41 79.75 ± 6.58 92.62 ± 1.64 87.50 ± 3.51 89.30 ± 1.33 Phosphorus removal % −9.57 ± 2.22 −8.71 ± 1.05 −15.43 ± 0.33 −16.87 ± 1.51 −11.51 ± 1.75 −9.50 ± 2.95 −11.84 ± 2.47 −6.03 ± 1.73 −9.20 ± 3.27 −5.91 ± 1.23 −7.10 ± 4.70 It was suggested that for a good fit of a model, R2 should be at least 0.80 These response variables had an R2 greater than 0.80, indicating that the regression models explained the reaction well [31] The experiments were carried out in randomized runs to determine the effect of the factors on four characteristic responses: CBZ, COD, NH4 + -N, and PO4 3− -P Percent contributions of all factors are presented as a chart in Figure to determine the importance of factors It could be seen from Figure 2a,b, and d that in the case of CBZ, COD, and phosphorus removal, DO was the most influential factor However, according to Figure 2c, HRT had the most significant effect on ammonia removal 6 of 15 40 30 20 10 % contribution % contribution Membranes 2022, 12, 420 Membranes 2022, 12, x FOR PEER REVIEW 60 40 20 (b) % contribution % contribution (a) 60 40 20 (c) of 80 60 40 20 (d) Figure Percent contribution of each factor on the performance statistics of (a) CBZ removal, Figure Percent contribution each factor on the statistics of (a) CBZ removal, COD removal, (c)ofammonia removal, (d)performance phosphorus removal (b) COD removal, (c) ammonia removal, (d) phosphorus removal 3.3 CBZ Removal 3.3 CBZ Removal The overall performance of the MBR was estimated by calculating CBZ removal as The overall performance of the MBR was estimated by calculating CBZ removal as response The three-factor interaction (3FI) model described the variation of the CB a response The three-factor interaction (3FI) model describedAnalysis the variation of the(ANOVA) CBZ removal efficiency as a function of the variables of variance for t removal efficiency as a function of the variables Analysis of variance (ANOVA) for the the signi model terms is summarized in Table The F-value and p-value determined model terms is summarized Table The F-value and p-value determined thethat significance cance of eachin coefficient It was observed from ANOVA analysis the confidence lev of each coefficient was observed analysis that the confidence levelthe wasF-value an was It greater than 80%from (p < ANOVA 0.05) for the CBZ removal response, while greater than 80% (p < 0.05) formodel the CBZ removal response, the F-value p-value p-value of the were 38.44 and 0.0062, while respectively This and indicated that the es of the model were 38.44 and 0.0062, respectively This indicated that the estimated model mated model fitted the experimental data adequately fitted the experimental data adequately Furthermore, the coefficient of determination R2 of the model was reasonably clo to (0.9890), implying that the model explained about 98.90% of the variability in t Table ANOVA results for CBZ removal response data From Table 4, A (DO), C (SRT), and B (HRT) were significant model terms The i teraction between and SRT was more importantp-Value than other interactions (AB, BC, an Source Sum of Squares df DOMean Square F-Value ABC), with a probability value larger than 0.05 After elimination of insignificant p Model 582.31 83.19 38.44 0.0062 rameters, the final empirical model at 95% confidence level couldSignificant be represented as: A-DO B-HRT C-SRT AB AC BC ABC Residual Lack of fit Pure error Cor total Std dev Mean 207.62 207.62 95.94 0.0023 111.38CBZ removal 0.0056 (%) = 111.38 10.16 + 0.54 × A + 51.47 2.01 × B − 0.15 × C − 0.33 × A × B − 0.097 × A × C − ( 167.72 167.72 0.098 × B ×77.51 C + 0.02 × A0.0031 ×B×C 16.02 16.02 7.40 0.0725 62.82 62.82 29.03 0.0125 Table 4.134 ANOVA 1results for CBZ 4.13 removal response 1.91 0.2610 12.62 12.62 5.83 0.0946 Source Sum of Squares df Mean Square F-Value p-Value 6.49 2.16 Model 582.31 83.19 38.44 0.0062 Significant 3.81 3.81 2.85 0.2335 Not significant A-DO 207.62 207.62 95.94 0.0023 2.68 1.34 B-HRT 111.38 51.47 0.0056 588.80 10 111.38 C-SRT 167.72 167.72 77.51 0.0031 1.47 0.9890 R AB 16.02 16.02 7.40 0.0725 18.41 0.9632 Adjusted R21 AC 62.82 62.82 29.03 0.0125 BC 4.13 4.13 1.91 0.2610 Furthermore, the coefficient of12.62 determination the model was ABC R of12.62 5.83reasonably 0.0946 close to (0.9890), implying that the model6.49 explained about 98.90% Residual 2.16 of the variability in the data From Table 4, A (DO), (SRT), and 3.81 B (HRT) were interaction Lack ofCfit significant 3.81 model terms 2.85 The 0.2335 Not significant between DO andPure SRTerror was more important than interactions (AB, BC, and ABC), 2.68 other1.34 Cor total 588.80 10 was observed with an increase in DO and SRT, with the effect of DO being a little greater than that of SRT The figures show a slight positive effect of HRT on CBZ removal in the MBR system The maximum CBZ removal efficiency was 38.36 ± 4.49% at 1.5 mg/L DO, 24 h HRT, and days SRT, while the minimum reached 9.04 ± 1.12% at 5.5 mg/L DO, 12 h HRT, and 15 of 15 days SRT This finding was similar to previous studies, which found that anoxic conditions removed more CBZ than aerobic conditions [32] Its degradation was highly dependent on the operating conditions [33], the presence of strong electron-withdrawing with a probability value larger than 0.05 After elimination of insignificant parameters, groups, or the absence of electron-donating groups in CBZ, might explain the low rethe final empirical model at 95% confidence level could be represented as: moval effectiveness by activated sludge or MBR processes [34], even under long SRT [35] This CBZ studyremoval suggested a short SRT research (%) that = 10.16 + 0.54 × Amight + 2.01have × B a−significant 0.15 × C −impact 0.33 × on A ×the B− (2) model, and the optimization operating improve the 0.097 × A × of C −the 0.098 × B × Cfactors + 0.02 could × A ×significantly B×C overall removal of CBZ The main mechanisms forsurface removing MBR are biodegradation and sorpFigure 3a–c shows the 3D plot PPCPs of CBZ by removal versus two varying parameters tion [36] To eliminate CBZ, however, combining the MBR process with other treatments, at a fixed value of the third parameter Again, a decrease in CBZ removal efficiency was such as advanced oxidation processes (AOPs) or adsorption, required lower than the observed with an increase in DO and SRT, with the effect of DOisbeing a littletogreater concentration that of SRT of CBZ in the permeate Membranes 2022, 12, 420 30 40 20 30 10 40 30 20 10 1.5 24 20 10 24 21 2.5 21 B: HRT (h) CBZ removal (%) CBZ removal (%) CBZ removal (%) 40 18 15 12 1.5 2.5 3.5 4.5 5.5 A: DO (mg/L) (a) 10 C: SRT (days) 3.5 A: DO (mg/L) 13 4.5 15 5.5 15 13 18 10 C: SRT (days) (b) 15 B: HRT (h) 12 (c) Figure Response surface plots for CBZ removal efficiency as a function of the following: (a) HRT Figure surface as a function and DO3.atResponse SRT = 10 days; (b)plots SRT for andCBZ DO removal at HRT =efficiency 18 h; (c) HRT and SRT of at the DOfollowing: = 3.5 mg/L.(a) HRT and DO at SRT = 10 days; (b) SRT and DO at HRT = 18 h; (c) HRT and SRT at DO = 3.5 mg/L 3.4 COD Removal The figures show a slight positive effect of HRT on CBZ removal in the MBR system The 3FI model describes the variation of the COD removal in the system studied The maximum CBZ removal efficiency was 38.36 ± 4.49% at 1.5 mg/L DO, 24 h HRT, Based on ANOVA (Table 5), A (DO) and B (HRT) were significant model terms, which and days SRT, while the minimum reached 9.04 ± 1.12% at 5.5 mg/L DO, 12 h HRT, might be because of the increase in the aerobic heterotrophic bacteria, while the effect of and 15 days SRT This finding was similar to previous studies, which found that anoxic C (SRT) was not much on the overall removal efficiency Furthermore, the confidence conditions removed more CBZ than aerobic conditions [32] Its degradation was highly level of ANOVA the CODconditions removal response greater 80% (p < 0.05) for COD dependent on theofoperating [33], thewas presence of than strong electron-withdrawing response, while F-value and p-value of the model 19.38might and 0.0167, respectively groups, or the the absence of electron-donating groupswere in CBZ, explain the low reThis also indicated that the estimated model fitted the experimental data adequately It moval effectiveness by activated sludge or MBR processes [34], even under long SRT [35] was further shown that the interactions of AB, AC, BC, and ABC were not significant This study suggested that a short SRT might have a significant impact on the research model (factors) model,terms and the optimization of the operating factors could significantly improve the overall removal of CBZ The main mechanisms for removing PPCPs by MBR are biodegradation and sorption [36] To eliminate CBZ, however, combining the MBR process with other treatments, such as advanced oxidation processes (AOPs) or adsorption, is required to lower the concentration of CBZ in the permeate 3.4 COD Removal The 3FI model describes the variation of the COD removal in the system studied Based on ANOVA (Table 5), A (DO) and B (HRT) were significant model terms, which might be because of the increase in the aerobic heterotrophic bacteria, while the effect of C (SRT) was not much on the overall removal efficiency Furthermore, the confidence level of ANOVA of the COD removal response was greater than 80% (p < 0.05) for COD response, while the F-value and p-value of the model were 19.38 and 0.0167, respectively This also indicated that the estimated model fitted the experimental data adequately It was further shown that the interactions of AB, AC, BC, and ABC were not significant model terms (factors) Membranes 2022, 12, 420 of 15 Membranes 2022, 12, x FOR PEER REVIEW of Table ANOVA results for COD removal response Source Model A-DO B-HRT C-SRT AB AC BC ABC Residual Lack of fit Pure error Cor total Std dev Mean Sum of Squares df Mean Square F-Value p-Value Table ANOVA results for COD removal response 659.56 94.22 19.38 0.0167 Significant 374.40 Sum of1Squares374.40 77.02Square 0.0031 Source df Mean F-Value p-Value 226.00 226.00 46.49 0.0065 Model 659.56 94.22 19.38 0.0167 Significan 50.80 50.80 10.45 0.0481 A-DO 374.40 374.40 77.02 0.0031 0.7858 0.7858 0.1617 0.7146 B-HRT 226.00 226.00 0.3954 46.49 0.0065 4.76 4.76 0.9788 1.56 1.56 0.3218 0.6102 C-SRT 50.80 50.80 10.45 0.0481 1.25 0.2570 AB1.25 0.7858 0.7858 0.6471 0.1617 0.7146 14.58 4.86 AC11.43 4.76 4.76 0.9788Not0.3954 11.43 7.24 0.1148 significant BC3.16 1.56 1.56 0.3218 0.6102 1.58 674.15 10 ABC 1.25 1.25 0.2570 0.6471 2.20 0.9784 R Residual 14.58 4.86 86.45 0.9279 Adjusted R2 100 100 92 92 84 84 COD removal (%) COD removal (%) Not signif Lack of fit 11.43 11.43 7.24 0.1148 cant Figure 4a–c illustrates the interactive effect of the variables on COD removal The results Pure error 3.16 1.58 showed changes in DO from 1.5 to 5.5 mg/L, and HRT from 12 to 24 h increased COD Cor total 674.15 10 removal by about 13% and 11%, respectively, whereas the COD removal was lowered 5% Std dev 2.20 R2 0.9784 for the changes in SRT Overall, the system showed good performance for COD removal, Mean 86.45between 70% and 99% Adjusted 0.9279being with removal efficiencies ranging COD R removal efficiencies high throughout the experiments could be due to the filtration membrane s ability to retain Figure the interactive of the variables on COD removal T all the particulate COD4a–c [37].illustrates The maximum values foreffect the response were 99.37 ± 0.42% results showed changes in DO from 1.5 to 5.5 mg/L, and HRT from 12 to 24 at 5.5 mg/L DO, 24 h HRT, and days SRT compared to the minimum of 69.23 ± 2.56%h increas byand about 13% and respectively, whereas themodel COD removal at 1.5 mg/LCOD DO, removal 12 h HRT, 15 days SRT.11%, Therefore, the final empirical at 95% was lo ered 5% for the changes in SRT Overall, the system showed good performance for CO confidence level could be represented as: removal, with removal efficiencies ranging between 70% and 99% COD removal e ciencies being(%) high due to the COD removal = throughout 71.43 + 1.93 the × Aexperiments + 0.60 × B −could 1.45 be ×C + 0.04 × filtration membran (3) A × B + 0.20 × A × C + 0.04 × B × C − 0.007 × A × B × C ability to retain all the particulate COD [37] The maximum values for the response we 99.37 ± 0.42% at 5.5 mg/L DO, 24 h HRT, and days SRT compared to the minimum Positive coefficients indicated anDO, increasing effect of15 A days and BSRT on the response, +1.93 69.23 ± 2.56% at 1.5 mg/L 12 h HRT, and Therefore, the final empiri and +0.60, respectively, suggesting that the response was more dependent on DO (A) than model at 95% confidence level could be represented as: HRT (B) The current study confirmed a small effect of SRT on COD removal efficiency COD (%) =[38] 71.43This + 1.93 × A + 0.60 × could B − 1.45 C + to 0.04 × A × B + 0.20 × at short SRT (3, 5, and 10removal days) before phenomenon be×due a decrease A × C + 0.04 × B × C − 0.007 × A × B × C in the ratio of the active biomass to that of the total biomass (MLVSS/MLSS) following increasing SRT, indicating that the increased sludge age could decrease microbial activities 76 68 60 24 21 18 B: HRT (h) 4.5 15 12 (a) Figure Cont 1.5 2.5 5.5 3.5 A: DO (mg/L) 76 68 60 15 13 10 C: SRT (days) 4.5 (b) 1.5 2.5 5.5 3.5 A: DO (mg/L) Membranes 2022, 12, x FOR PEER REVIEW Membranes 2022, 12, 420 of 15 100 COD removal (%) 92 84 76 68 60 24 21 B: HRT (h) 18 15 12 15 10 13 C: SRT (days) (c) Figuresurface Response plots for CODefficiency removal efficiency as a function of the following: (a Figure Response plotssurface for COD removal as a function of the following: and DO at SRT = 10 days; (b) SRT and DO at HRT = 18 h; (c) HRT and SRT DO = 3.5 mg/L (a) HRT and DO at SRT = 10 days; (b) SRT and DO at HRT = 18 h; (c) HRT and SRT at DO = 3.5atmg/L 3.5 Ammonia Removal Positive coefficients indicated an increasing effect of A and B on the response, and +0.60, respectively, that the response more dependent Based on ANOVA (Table 6), A and Bsuggesting were significant model terms was In addition, the con- on D than HRT (B) The current study confirmed a small effect of SRT on fidence level for the ammonia removal response was greater than 80% (p < 0.05), whileCOD the remova ciency at short were SRT (3, 5, and days)respectively before [38] This Thissuggested phenomenon model s F-value and p-value 20.57 and 10 0.0154, thatcould the be du decrease in thematched ratio of the biomassdata to that of the totalthe biomass (MLVSS/M estimated model adequately the active experimental Furthermore, model’s following increasing indicating that the increased sludge agemodel could decreas coefficient of determination R2 was SRT, relatively close to (0.9796), meaning that the crobial activities described around 97.96% of the variability in the data Table ANOVA results for ammonia removal response 3.5 Ammonia Removal Source Model A-DO B-HRT C-SRT AB AC BC ABC Residual Lack of fit Pure error Cor total Std dev Mean Based on ANOVA (Table 6), A and B were significant Sum of Squares df Mean Square F-Value p-Value model terms In additio confidence level for the ammonia removal response was greater than 80% (p < 315.97 45.14 20.57 0.0154 Significant while the model′s1 F-value94.80 and p-value43.21 were 20.57 and 0.0154, respectively This 94.80 0.0072 gested that the estimated model adequately 167.20 167.20 76.21 matched 0.0032 the experimental data Fu was relatively close to (0 more, the model’s R 19.67 coefficient 19.67 of determination 8.97 0.0579 16.97 that the model 16.97 7.74 0.0689 meaning described around 97.96% of the variability in the data 3.33 3.33 1.52 0.3059 Positive coefficients indicated an increasing effect of A and B on the response, 5.79 0.0953 and 12.70 +0.99, respectively As12.70 could be seen, the effect of SRT on the response was 1.29 1.29 0.5872 0.4993 than 6.58 that of DO and HRT, while the interactions of AB, AC, BC, and ABC were no 2.19 nificant model terms As a result, the maximum ammonia efficiency was 4.00 4.00 3.10 0.2206 removal Not significant at DO (5.5 mg/L), HRT (24 h), and SRT (15 days) The final empirical model at 95% 2.58 1.29 322.55 10 fidence level could be represented as: 1.48 0.9796 R2 Ammonia removal (%) R = 261.29 + 1.39×A + 0.99 × B + 1.72 × C + 0.05 × A × B − 90.55 0.9320 Adjusted 0.18 × A × C − 0.07 × B × C + 0.007 × A × B × C Positive coefficients indicated increasing effect as of aAfunction and B onofthe The variation of an ammonia removal theresponse, variables+1.39 is shown in F and +0.99, respectively As could be seen, the effect of SRT on the response was lower 5a–c It was observed that an increase in ammonia removal was duethan to increased that of DO andHRT, HRT,and while the According interactionstoofseveral AB, AC, BC, and were notinsignificant SRT studies onABC nitrification MBRs, increasing model terms As a result,ammonia the maximum ammonia removal efficiency was 99.71% at DO improved removal efficiency significantly [39], while others showed th (5.5 mg/L), HRT (24 h), andefficiency SRT (15 days) The final empirical model at 95%ofconfidence high removal of ammonia was almost independent SRT [40] Membra level could be tration represented as: the system′s performance by retaining all suspended solids, pro increased and polysaccharides from the sludge supernatant Ammonia removal (%) = 61.29 + 1.39×A + 0.99 × B + 1.72 × C + 0.05 × A × B − 0.18 × A × Cresults − 0.07for×ammonia B × C + removal 0.007 ×response A×B×C Table ANOVA (4) The variationSource of ammoniaSum removal as a function the variables shown in Figure of Squares df ofMean Square isF-Value p-Value5a–c It was observed that an increase in ammonia removal was due to increased DO, HRT, Significa Model 315.97 45.14 20.57 0.0154 and SRT According to several studies in 94.80 MBRs, increasing improved A-DO 94.80on nitrification 43.21 SRT 0.0072 from the sludge supernatant 100 Ammonia removal (%) 100 96 Ammonia removal (%) 92 88 83 79 75 96 92 88 83 79 75 15 15 13 10 C: SRT (days) 21 12 15 13 C: SRT (days)10 1.5 24 18 B: HRT (h) 2.5 3.5 4.5 5.5 A: DO (mg/L) (a) (b) 100 96 Ammonia removal (%) Membranes 2022, 12, 420 B-HRT 167.20 167.20 76.21 0.0032 C-SRT 19.67 19.67 8.97 0.0579 AB 16.97 16.97 7.74 0.0689 AC 3.33 3.33 1.52 0.3059 BC 12.70 12.70 5.79 0.0953 ABC 1.29 1.29 0.5872 0.499310 of 15 Residual 6.58 2.19 Lack of fit 4.00 4.00 3.10 0.2206 Not signific Pure error 2.58 1.29 ammonia removal efficiency significantly [39], while others showed that the high removal Cor total efficiency of ammonia was almost322.55 independent10of SRT [40] Membrane filtration increased 0.9796 Std dev 1.48 R2 the system s performance by retaining all suspended solids, proteins, and polysaccharides 0.9320 Mean 90.55 Adjusted R2 92 88 83 79 75 24 21 B: HRT (h) 18 15 12 1.5 2.5 5.5 4.5 3.5 A: DO (mg/L) (c) Figuresurface Response plots for ammonia removal function of the followin Figure Response plots surface for ammonia removal efficiency as efficiency a functionasofa the following: HRT and DO at SRT = 10 days; (b) SRT and DO at HRT = 18 h; (c) HRT and SRT at DO = 3.5 m (a) HRT and DO at SRT = 10 days; (b) SRT and DO at HRT = 18 h; (c) HRT and SRT at DO = 3.5 mg/L 3.6 Phosphorus Removal 3.6 Phosphorus Removal The biological phosphorous removal process is divided anaerobic aero- and ae The biological phosphorous removal process into is divided into and anaerobic bic stages Instages the anaerobic zone, phosphate accumulating organisms (PAOs)(PAOs) releaserelease p In the anaerobic zone, phosphate accumulating organisms phosphorus and accumulate poly hydroxybutyrate (PHB), whereas, the aerobic phorus and accumulate poly hydroxybutyrate (PHB),inwhereas, in zone, the aerobic z phosphorous phosphorous is absorbed [41] is absorbed [41] According toAccording ANOVA, the confidence for phosphorus response wasresponse to ANOVA, thelevel confidence level for removal phosphorus removal s F-value and p-value were 12.77 and 0.0303, greater than 80% (p < 0.05), while the model greater than 80% (p < 0.05), while the model′s F-value and p-value were 12.77 and respectively respectively This indicated thatindicated the estimated model fitted model the experimental data well data This that the estimated fitted the experimental 2 Furthermore,Furthermore, the model s coefficient determination R was quiteRclose (0.9675), the model′sofcoefficient of determination wastoquite close to (0.9 indicating that the model described roughly 96.75% of the 96.75% data variability indicating that the model described roughly of the data variability A and B were significant model terms, as shown in Table Positive coefficients of +0.90, +0.17, and +0.35 indicated an increasing A, B, and C effect on the response, respectively As could be observed, SRT had a lower effect on the response than DO and HRT, and the interactions between AB, AC, BC, and ABC were not significant model terms Therefore, the final empirical model at 95% confidence level could be represented as follows: Phosphorus removal (%) = −7.28 + 0.90 × A + 0.17 × B + 0.35 × C − 0.12 × A × B − 0.11 × A × C − 0.02 × B × C + 0.005 × A × B × C (5) Therefore, the final empirical model at 95% confidence level could be represen follows: Phosphorus removal (%) = −7.28 + 0.90 × A + 0.17 × B + 0.35 × C − 0.12 × A × – 0.11 × A × C – 0.02 × B × C + 0.005 × A × B × C Membranes 2022, 12, 420 11 of 15 Phosphorus removal efficiency is inversely proportional to DO, HRT and SRT result, the maximum phosphorus removal achieved was −5.91% at low DO (1.5 m (12forh), and highremoval SRT (15response days), indicating an anaerobic–aerobic environmen Table ANOVAHRT results phosphorus provided at this condition Figure 6a–c shows the interactive effects of the variab Source Sum of Squares removal df Mean Square F-Value p-Value phosphorus Model A-DO B-HRT C-SRT AB AC BC ABC Residual Lack of fit Pure error Cor total Std dev Mean 121.55 17.36 12.77 0.0303 Significant Table ANOVA results for 81.66 phosphorus removal 81.66 60.06 response 0.0045 28.67 28.67 21.09 0.0194 Source Sum F-Value p-Value 3.05 of Squares 3.05 df Mean 2.24 Square 0.2312 Model 17.36 0.110112.77 0.0303 6.87 121.55 6.87 5.05 0.4812 81.66 0.4812 0.3539 A-DO 81.66 0.593860.06 0.0045 0.1661 0.1222 B-HRT 28.67 0.1661 28.67 0.749821.09 0.0194 0.6548 0.6548 0.4816 0.5376 C-SRT 3.05 2.24 0.2312 4.08 3.05 1.36 AB 6.87 0.1101 0.0131 6.87 0.0131 0.0065 0.94335.05 Not significant 4.07 0.4812 2.03 AC 0.4812 0.3539 0.5938 125.63 10 BC 0.1661 0.1661 0.1222 0.7498 1.17 0.9675 R ABC 0.6548 0.6548 0.4816 0.5376 −10.15 0.8918 Adjusted R2 Signific 0 -5 -5 Phosphorus removal (%) Phosphorus removal (%) Residual 4.08 1.36 Lack of fit 0.0131 0.0131 0.0065 0.9433 Not signi Phosphorus removal efficiency is inversely proportional to DO, HRT and SRT As a Pure error 4.07 2.03 result, the maximum phosphorus removal achieved was −5.91% at low DO (1.5 mg/L), 125.63 HRT (12 h), and Cor hightotal SRT (15 days), indicating10an anaerobic–aerobic environment was Std dev 1.17 R2 effects of the variables 0.9675 on provided at this condition Figure 6a–c shows the interactive Mean -10.15 Adjusted R 0.8918 phosphorus removal -10 -15 -20 12 1.5 2.5 3.5 A: DO (mg/L) 4.5 15 18 Membranes 2022, 12, x FOR PEER REVIEW 21 B: HRT (h) 24 5.5 -10 -15 -20 10 C: SRT (days) (a) 13 15 5.5 1.5 2.5 3.5 A: DO (mg/L) 4.5 (b) Phosphorus removal (%) -5 -10 -15 -20 10 C: SRT (days) 13 12 15 18 B: HRT (h) 21 15 24 (c) Figure Response plots for phosphorus removal efficiency as afollowing: function of the foll Figure Response surface plots forsurface phosphorus removal efficiency as a function of the (a) HRT and DO at SRT = 10 days; (b) SRT and DO at HRT = 18 h; (c) HRT and SRT at D (a) HRT and DO at SRT = 10 days; (b) SRT and DO at HRT = 18 h; (c) HRT and SRT at DO = 3.5 mg/L mg/L 3.7 Process Optimization 3.7 Process Optimization Optimization of the operating parameters based on two-level FFD was carried out Optimization the operating parameters based on two-level FFD was carried to improve the MBR process Inofaddition, a multi-response method called the desirimprove the MBR process In addition, a multi-response method called the desir function was used, which found operating conditions that yielded the “most desi responses [42,43] In this method, multiple responses could be combined into the “desirability tion” by choosing a value from to This method transformed an adequate funct (a) HRT and DO at SRT = 10 days; (b) SRT and DO at HRT = 18 h; (c) HRT and SRT at DO = 3.5 mg/L 3.7 Process Optimization Optimization of the operating parameters based on two-level FFD was carried12 out to of 15 improve the MBR process In addition, a multi-response method called the desirability function was used, which found operating conditions that yielded the “most desirable” responses [42,43].was used, which found operating conditions that yielded the “most ability function In thisresponses method, [42,43] multiple responses could be combined into the “desirability funcdesirable” tion”Inbythis choosing a value from to This method transformed an“desirability adequate function of method, multiple responses could be combined into the function” each determined response level (Yi) into a desirability score (di) within a 0–1 scale After by choosing a value from to This method transformed an adequate function of each that, all individual scores were integrated intowithin a single overall determined responsedesirability level (Yi) into a desirability score (di) a 0–1 scale.desirability After that, function optimized to determine the optimum setinto of input variables all individual desirability scores were integrated a single overall [42] desirability function For this operating parameters to be within optimized to reason, determine the optimum set of were inputset variables [42] the range, whereas CBZ, COD,For ammonia, and phosphorus removal efficiency were setthe to range, maximum Figure this reason, operating parameters were set to be within whereas CBZ, shows the graphical desirability generated from 40 optimum points At the best point COD, ammonia, and phosphorus removal efficiency were set to maximum Figure shows withgraphical a maximum overall desirability 0.72, optimum DO, HRT, andbest SRTpoint werewith found the desirability generated of from 40the optimum points At the a to be 1.7 mg/L, 24desirability h, and days, respectively Under CBZ, COD, maximum overall of 0.72, the optimum DO,optimum HRT, andconditions, SRT were found to be ammonia, and phosphorus removal were obtained at 37.08, 88.23,CBZ, 90.12, andammonia, −7.48 %, 1.7 mg/L, 24 h, and days, respectively Under optimum conditions, COD, respectively and phosphorus removal were obtained at 37.08, 88.23, 90.12, and −7.48 %, respectively 0.8 0.6 Desirability Membranes 2022, 12, 420 0.4 0.2 1.5 2.5 A: DO (mg/L)3.5 4.5 5.5 12 15 18 21 24 B: HRT (h) Figure 7 Desirability fitted fitted 3D 3D surface surface at at an an SRT SRT of of 55 days days Figure 4 Conclusions Conclusions The to to determine thethe significant parameters, The current currentwork workused usedthe theFFD FFDmethodology methodology determine significant parameinvestigate their interactions, and optimize conditions for the MBR process concerning ters, investigate their interactions, and optimize conditions for the MBR process conCBZ, COD, ammonia, and phosphorus removal As a result, the following conclusions were drawn: • • • Significant analysis of main and interaction effects revealed that the relative importance of significant parameters and interaction factors could be observed as follows: (a) for CBZ removal, when DO (A) and SRT (C) were increased, a decrease in removal efficiency was observed, with DO’s effect being a little greater than that of SRT, while a short SRT might significantly impact the research model HRT had a slight positive effect on CBZ removal; (b) COD removal: the response was more dependent on A than B while confirming a small effect of C on removal efficiency The AB, AC, BC, and ABC interactions were not significant model terms The system showed good performance for COD removal, with removal efficiencies ranging between 70% and 99% over the experiments; (c) for ammonia removal, positive coefficients indicated an increasing effect of A and B on the response, while the AB, AC, BC, and ABC interactions were not significant model terms It was observed that an increase in removal rate was due to increased DO, HRT, and SRT; (d) for phosphorus removal, A and B were significant model terms The interactions between AB, AC, BC, and ABC were not significant model terms, and removal efficiency was inversely proportional to DO, HRT, and SRT Optimization of the process was found at DO, HRT, and SRT of 1.7 mg/L, 24 h, and days for maximum CBZ, COD, ammonia, and phosphorus removal that obtained removal efficiencies 37.08, 88.23, 90.12, and −7.48 %, respectively The flat-sheet ceramic MBR demonstrated efficiency removals as high as 38.36 ± 4.49%, as CBZ is known to be a somewhat recalcitrant compound Membranes 2022, 12, 420 13 of 15 • To eliminate CBZ in the permeate, future studies require the combination of the MBR process with other treatments, such as advanced oxidation processes (AOPs) or adsorption Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/membranes12040420/s1, Figure S1: MBR experimental setup in practice, Figure S2: SEM photo of section part of flat-sheet ceramic membrane, Figure S3: Comparison of predicted and actual values of: (a) CBZ, (b) COD, (c) ammonia, and (d) phosphorus removal efficiency, Figure S4: HPLC–DAD chromatogram of CBZ standard containing 100 µg/mL at 210 nm wavelength, Table S1: Physico-chemical and pharmacological properties of CBZ, Table S2: Experimental values of trans-membrane pressure (TMP), Table S3: Average influent and effluent concentrations (±SD) of CBZ, COD, ammonia, and phosphorus through experiments Author Contributions: Conceptualization, K.-C.D and Y.-P.T.; methodology, K.-C.D and C.-C.Y.; formal analysis, K.-C.D and C.-C.Y.; writing—original draft preparation, K.-C.D.; writing—review and editing, K.-F.C and Y.-P.T.; supervision, K.-F.C and Y.-P.T.; project administration, K.-F.C and Y.-P.T.; funding acquisition, K.-F.C and Y.-P.T All authors have read and agreed to the published version of the manuscript Funding: This research received no external funding Institutional Review Board Statement: Not applicable Informed Consent Statement: Not applicable Data Availability Statement: The data presented in this study are available on request from the corresponding author Conflicts of Interest: The authors declare no conflict of interest References 10 11 12 Daughton, C.G.; Ternes, T.A Pharmaceuticals and Personal Care Products in the Environment: Agents of Subtle Change? Environ Health Perspect 1999, 107, 907–938 [CrossRef] [PubMed] Fang, T.-H.; Lin, C.-W.; Kao, C.-H Occurrence and Distribution of Pharmaceutical Compounds in the Danshuei River Estuary and the Northern Taiwan Strait Mar Pollut Bull 2019, 146, 509–520 [CrossRef] [PubMed] Lin, A.Y.-C.; Panchangam, S.C.; Chen, H.-Y Implications of Human Pharmaceutical Occurrence in the Sindian River of Taiwan: A Strategic Study of Risk Assessment J Environ Monit 2010, 12, 261–270 [PubMed] Dao, K.C.; Yang, C.-C.; Chen, K.-F.; Tsai, Y.-P Recent Trends in Removal Pharmaceuticals and Personal Care Products by Electrochemical Oxidation and Combined Systems Water 2020, 12, 1043 [CrossRef] Kosma, C.I.; Lambropoulou, D.A.; Albanis, T.A Occurrence and Removal of PPCPs in Municipal and Hospital Wastewaters in Greece J Hazard Mater 2010, 179, 804–817 [CrossRef] Oliveira, T.S.; Murphy, M.; Mendola, N.; Wong, V.; Carlson, D.; Waring, L Characterization of Pharmaceuticals and Personal Care Products in Hospital Effluent and Waste Water Influent/Effluent by Direct-Injection LC-MS-MS Sci Total Environ 2015, 518, 459–478 [CrossRef] Frick, E.A.; Henderson, A.K.; Moll, D.M.; Furlong, E.T.; Meyer, M.T Presence of Pharmaceuticals in Wastewater Effluent and Drinking Water, Metropolitan Atlanta, Georgia, July-September 1999 In Proceedings of the 2001 Georgia Water Resources Conference; Georgia Institute of Technology: Athens, GA, USA, 2001 Hartig, C.; Storm, T.; Jekel, M Detection and Identification of Sulphonamide Drugs in Municipal Waste Water by Liquid Chromatography Coupled with Electrospray Ionisation Tandem Mass Spectrometry J Chromatogr A 1999, 854, 163–173 [CrossRef] Meyer, M.T.; Bumgarner, J.E.; Varns, J.L.; Daughtridge, J.V.; Thurman, E.M.; Hostetler, K.A Use of Radioimmunoassay as a Screen for Antibiotics in Confined Animal Feeding Operations and Confirmation by Liquid Chromatography/Mass Spectrometry Sci Total Environ 2000, 248, 181–187 [CrossRef] Miao, X.-S.; Metcalfe, C.D Determination of Carbamazepine and Its Metabolites in Aqueous Samples Using Liquid Chromatography− Electrospray Tandem Mass Spectrometry Anal Chem 2003, 75, 3731–3738 [CrossRef] Molinari, R.; Pirillo, F.; Loddo, V.; Palmisano, L Heterogeneous Photocatalytic Degradation of Pharmaceuticals in Water by Using Polycrystalline TiO2 and a Nanofiltration Membrane Reactor Catal Today 2006, 118, 205–213 [CrossRef] Zhang, Y.; Geißen, S.-U.; Gal, C Carbamazepine and Diclofenac: Removal in Wastewater Treatment Plants and Occurrence in Water Bodies Chemosphere 2008, 73, 1151–1161 [CrossRef] [PubMed] Membranes 2022, 12, 420 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 14 of 15 Mohapatra, D.P.; Brar, S.K.; Tyagi, R.D.; Picard, P.; Surampalli, R.Y Analysis and Advanced Oxidation Treatment of a Persistent Pharmaceutical Compound in Wastewater and Wastewater Sludge-Carbamazepine Sci Total Environ 2014, 470, 58–75 [CrossRef] [PubMed] Ternes, T.A.; Herrmann, N.; Bonerz, M.; Knacker, T.; Siegrist, H.; Joss, A A Rapid Method to Measure the Solid–Water Distribution Coefficient (Kd) for Pharmaceuticals and Musk Fragrances in Sewage Sludge Water Res 2004, 38, 4075–4084 [CrossRef] Clara, M.; Strenn, B.; Kreuzinger, N Carbamazepine as a Possible Anthropogenic Marker in the Aquatic Environment: Investigations on the Behaviour of Carbamazepine in Wastewater Treatment and during Groundwater Infiltration Water Res 2004, 38, 947–954 [CrossRef] Smith, C.V., Jr The Use of Ultrafiltration Membrane for Activated Sludge Separation In Proceedings of the 24th Annual Purdue Industrial Waste Conference; Purdue University: West Lafayette, IN, USA, 1969; pp 1300–1310 Yamamoto, K.; Hiasa, M.; Mahmood, T.; Matsuo, T Direct Solid-Liquid Separation Using Hollow Fiber Membrane in an Activated Sludge Aeration Tank In Water Pollution Research and Control Brighton; Elsevier: Amsterdam, The Netherlands, 1988; pp 43–54 Nghiem, L.D.; Schäfer, A.I.; Elimelech, M Removal of Natural Hormones by Nanofiltration Membranes: Measurement, Modeling, and Mechanisms Environ Sci Technol 2004, 38, 1888–1896 [CrossRef] Yoon, Y.; Westerhoff, P.; Snyder, S.A.; Wert, E.C Nanofiltration and Ultrafiltration of Endocrine Disrupting Compounds, Pharmaceuticals and Personal Care Products J Membr Sci 2006, 270, 88–100 [CrossRef] Akamatsu, K.; Lu, W.; Sugawara, T.; Nakao, S Development of a Novel Fouling Suppression System in Membrane Bioreactors Using an Intermittent Electric Field Water Res 2010, 44, 825–830 [CrossRef] Montgomery, D.C Design and Analysis of Experiments; John Wiley & Sons: Hoboken, NJ, USA, 2017; ISBN 1-119-11347-4 Gottipati, R.; Mishra, S Process Optimization of Adsorption of Cr (VI) on Activated Carbons Prepared from Plant Precursors by a Two-Level Full Factorial Design Chem Eng J 2010, 160, 99–107 [CrossRef] Seki, Y.; Seyhan, S.; Yurdakoc, M Removal of Boron from Aqueous Solution by Adsorption on Al2O3 Based Materials Using Full Factorial Design J Hazard Mater 2006, 138, 60–66 [CrossRef] Hui, Y.H Handbook of Food Science, Technology, and Engineering; CRC Press: Boca Raton, FL, USA, 2006; Volume 149, ISBN 0-8493-9849-5 Onsekizoglu, P.; Bahceci, K.S.; Acar, J The Use of Factorial Design for Modeling Membrane Distillation J Membr Sci 2010, 349, 225–230 [CrossRef] American Public Health Association APHA Standard Methods for the Examination of Water and Wastewater; APHA-AWWA-WEF: Washington, DC, USA, 2005 Weatherburn, M.W Phenol-Hypochlorite Reaction for Determination of Ammonia Anal Chem 1967, 39, 971–974 [CrossRef] Alturki, A.A.; Tadkaew, N.; McDonald, J.A.; Khan, S.J.; Price, W.E.; Nghiem, L.D Combining MBR and NF/RO Membrane Filtration for the Removal of Trace Organics in Indirect Potable Water Reuse Applications J Membr Sci 2010, 365, 206–215 [CrossRef] Aubenneau, M.; Tahar, A.; Casellas, C.; Wisniewski, C Membrane Bioreactor for Pharmaceutically Active Compounds Removal: Effects of Carbamazepine on Mixed Microbial Communities Implied in the Treatment Process Biochem 2010, 45, 1826–1831 [CrossRef] García-Gómez, C.; Drogui, P.; Seyhi, B.; Gortáres-Moroyoqui, P.; Buelna, G.; Estrada-Alvgarado, M.I.; Alvarez, L.H Combined Membrane Bioreactor and Electrochemical Oxidation Using Ti/PbO2 Anode for the Removal of Carbamazepine J Taiwan Inst Chem Eng 2016, 64, 211–219 [CrossRef] Joglekar, A.M.; May, A.T Product Excellence through Design of Experiments Cereal Foods World 1987, 32, 857 Phan, H.V.; Hai, F.I.; Zhang, R.; Kang, J.; Price, W.E.; Nghiem, L.D Bacterial Community Dynamics in an Anoxic-Aerobic Membrane Bioreactor–Impact on Nutrient and Trace Organic Contaminant Removal Int Biodeterior Biodegrad 2016, 109, 61–72 [CrossRef] Hai, F.I.; Li, X.; Price, W.E.; Nghiem, L.D Removal of Carbamazepine and Sulfamethoxazole by MBR under Anoxic and Aerobic Conditions Bioresour Technol 2011, 102, 10386–10390 [CrossRef] Hai, F.I.; Nghiem, L.D.; Khan, S.J.; Asif, M.B.; Price, W.E.; Yamamoto, K Removal of Emerging Trace Organic Contaminants (TrOC) by MBR Membr Biol React Theory Model Des Manag Appl Wastewater Reuse 2019, 413–468 [CrossRef] Reif, R.; Suárez, S.; Omil, F.; Lema, J.M Fate of Pharmaceuticals and Cosmetic Ingredients during the Operation of a MBR Treating Sewage Desalination 2008, 221, 511–517 [CrossRef] Tran, N.H.; Chen, H.; Reinhard, M.; Mao, F.; Gin, K.Y.-H Occurrence and Removal of Multiple Classes of Antibiotics and Antimicrobial Agents in Biological Wastewater Treatment Processes Water Res 2016, 104, 461–472 [CrossRef] Falahti-Marvast, H.; Karimi-Jashni, A Performance of Simultaneous Organic and Nutrient Removal in a Pilot Scale Anaerobic– Anoxic–Oxic Membrane Bioreactor System Treating Municipal Wastewater with a High Nutrient Mass Ratio Int Biodeterior Biodegrad 2015, 104, 363–370 [CrossRef] Duan, L.; Moreno-Andrade, I.; Huang, C.; Xia, S.; Hermanowicz, S.W Effects of Short Solids Retention Time on Microbial Community in a Membrane Bioreactor Bioresour Technol 2009, 100, 3489–3496 [CrossRef] [PubMed] Fan, X.-J.; Urbain, V.; Qian, Y.; Manem, J Nitrification and Mass Balance with a Membrane Bioreactor for Municipal Wastewater Treatment Water Sci Technol 1996, 34, 129–136 Membranes 2022, 12, 420 40 41 42 43 15 of 15 Huang, X.; Gui, P.; Qian, Y Effect of Sludge Retention Time on Microbial Behaviour in a Submerged Membrane Bioreactor Process Biochem 2001, 36, 1001–1006 [CrossRef] Kermani, M.; Bina, B.; Movahedian, H.; Amin, M.M.; Nikaein, M Application of Moving Bed Biofilm Process for Biological Organics and Nutrients Removal from Municipal Wastewater Am J Environ Sci 2008, 4, 675 Ruby Figueroa, R.A.; Cassano, A.; Drioli, E Ultrafiltration of Orange Press Liquor: Optimization for Permeate Flux and Fouling Index by Response Surface Methodology Sep Purif Technol 2011, 80, 1–10 [CrossRef] Aksezer, S.Ç On the Sensitivity of Desirability Functions for Multiresponse Optimization J Ind Manag Optim 2008, [CrossRef] ... estimating interactions, and optimizing the parameters The HRT (12 and 24 h), DO (1.5 and estimating estimating and interactions, optimizing and the parameters the The parameters HRT (12 The HRT... indicated increasing effect as of aAfunction and B onofthe The variation of an ammonia removal theresponse, variables+1.39 is shown in F and +0.99, respectively As could be seen, the effect of. .. SRT on the response was lower 5a? ??c It was observed that an increase in ammonia removal was duethan to increased that of DO andHRT, HRT ,and while the According interactionstoofseveral AB, AC,
Ngày đăng: 22/02/2023, 22:42
Xem thêm:
