Antibiotics detoxification from synthetic and real effluents using a novel MTAB surfactant – montmorillonite (organoclay) sorbent

Please download to get full document.

View again

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.


  Antibiotics detoxification from synthetic and real effluents using a novel MTAB surfactant – montmorillonite (organoclay) sorbent
  Antibiotic detoxi fi cation from synthetic and reale ffl uents using a novel MTAB surfactant-montmorillonite (organoclay) sorbent † Merry Anggraini, a Al fi n Kurniawan, a Lu Ki Ong, a Mario A. Martin, a Jhy-Chern Liu, b Felycia E. Soetaredjo, a Nani Indraswati a and Suryadi Ismadji* a The growing threat of antibiotic-resistant bacteria to public health has raised interest in the propertreatment of discharged pharmaceutical wastewater before entering surface waters. In this study,adsorption was highlighted as a low cost and e ff ective pathway to remove amoxicillin and ampicillinfrom aqueous solutions. Montmorillonite (Na-MMT) and myristyltrimethylammonium (MTA)-intercalatedmontmorillonite (O-MMT) were employed as the adsorbing solids. Static adsorption experiments wereperformed at three temperatures (303.15 K, 313.15 K and 323.15 K) for single antibiotic systems. Theadsorption isotherm curves at all temperatures exhibited an L2-type isotherm. The Freundlich andLangmuir models were applied to analyze single adsorption isotherm data. The maximum sorptioncapacity of 0.124 – 0.133 mmol g  1 for amoxicillin and 0.143 – 0.157 mmol g  1 for ampicillin wasestimated for O-MMT sorbent from Langmuir  fi tting. A modi fi ed extended-Langmuir model with theinclusion of surface coverage ( q ) was proposed for analysis of binary adsorption isotherm data. The fi tness of the modi fi ed extended-Langmuir model was superior to the srcinal model. Batch adsorptiontests on real pharmaceutical wastewater demonstrated the feasibility of the O-MMT sorbent for practicalapplications. Introduction Currently, a large group of antibiotics is available in the market and they have been proven to be powerful drugs to treat variousbacterial infections, from minor to life-threatening ones. Anti-biotics are generally produced by or derived from microorgan-isms such as fungi or bacteria and they can also be chemically synthesized and particular examples are penicillins, cephalo-sporins, macrolides, rifamycins, sulfonamides, chloramphen-icol,tetracyclinesandaminoglycosides.Ofparticularinterestarepenicillin groups including amoxicillin and ampicillin. Amoxi-cillin is a moderate-spectrum of   b -lactam antibiotics and isusually the drug of choice within penicillin groups due to itsbetter absorptivity, following oral administration than other b -lactam antibiotics. Amoxicillin is widely used in the treatment of a number of bacterial infections include pneumonia, bron-chitis, laryngitis, gonorrhea, skin and urinary tract infections. 1  Ampicillin is also a  b -lactam antibiotic, part of theaminopenicillin family, which is closely related to amoxicillin interms of spectrum and activity level. 2 Both amoxicillin andampicillin work in a similar manner against Gram Positive andGram Negative bacteria by interfering cell wall synthesis so that the human antibodies can penetrate and remove them. 2 In spiteof its usefulness and valuable contributions in human therapy,antibiotic-resistant bacteria are the today's most pressing clin-ical and public health concerns that continue to grow due toabuse and overuse of antibiotics. This leads to consequent treatment complicationsandincreasedhealthcare costsbecausethe target bacteria are becoming more resistant to the exposureof therapeutic levels of an antibiotic.The dissemination of antibiotics in natural environments( e.g. , lakes and streams) may come from various sources, suchas human and animal excretion, agriculture, aquaculture andlivestock farming, hospital sewage sludge and diverse industrialroutes. 3 The relative concentrations of antibiotics in theindustrial e ffl uents are several order of magnitude higher thanthose released from veterinary and hospital sources. 3  Althoughmost of pharmaceutical products administered are particularly designed to have a short half-life, some antimicrobials liketetracycline, erythromycin, sulfamethoxazole and penicilloylgroups are persistence and poorly metabolized hence they areonly partially eliminated in the sewage treatment plants. Theenrichment of antibiotic residues and their transformed prod-ucts into receiving waters is worrying as it can impact the a  Department of Chemical Engineering, Widya Mandala Surabaya Catholic University, Kalijudan 37, Surabaya 60114, Indonesia. E-mail:; Fax:+62 31 389 1267; Tel: +62 31 389 1264 b  Department of Chemical Engineering, National Taiwan University of Science and Technology, No. 43, Sec. 4, Keelung Rd., Taipei City 106, Taiwan, Republic of China †  Electronic supplementary information (ESI) available. See DOI:10.1039/c4ra00328d Cite this:  RSC Adv. , 2014,  4 , 16298Received 13th January 2014Accepted 20th March 2014DOI: 10.1039/c4ra00328d 16298  |  RSC Adv. , 2014,  4 , 16298 – 16311 This journal is © The Royal Society of Chemistry 2014 RSC Advances PAPER  structure and activity of microbiota, spreading of antibiotic-resistant genes to pathogenic bacteria strains that can reachhumans through food chains and ultimately in the water reusescenario. 4 – 6 Several treatment methods have been developed so far forpuri  cation of antibiotic-bearing e ffl uents such as aerobic andanaerobic biological treatments, 7,8 advanced chemical oxida-tion, 9 – 11 membrane separation, 12 chlorination, 13 photocatalysis, 14 – 16 electro-oxidation, 17 and adsorption. 18 – 21  Amongst them, adsorptionis considered as the most reliable method for removal of toxicmicropollutantsfrommunicipal water andwastewater. Regardlessoftheadsorbentmaterialandsystemdesign,adsorptionprocessisgenerally simple, adaptable, economically viable and highly e ff  ec-tive across a wide range of concentrations, highlighting itsadvantages over other technologies. The treatment of water and wastewater containing antibiotics by adsorption is the key tomodulating theextent ofenvironmental occurrence,transportandfate of this micropollutant.Clays and clay minerals have found potential applications forsustaining the environment because they exhibit large adsorp-tion capacity,excellent mechanical and chemical stability, cheapand readily obtained in large quantities. So, e ff  orts toward theirapplications as a sorbent to abate antibiotics from water and wastewater have been stimulated. Over the past few years, aprogressing research on the removal of antibiotics using clays orclay minerals can be found in literatures. 18 – 21 In most, if not all,of these studies, single adsorption systems are emphasized andonly limited studies dedicate to investigate binary or multi-component systems. In the real pharmaceutical sewage treat-ment units, two or more unmetabolized antibiotics mayco-exist,thus it is necessary to study binary or multicomponent adsorp-tion equilibria and thermodynamics for e ff  ective design andoptimization ofantibiotic/claywastewater treatment systems.To  ll this gap, the goals of this study are (1) to synthesize a novelorganoclay sorbent and (2) to evaluate the performance of pris-tine and as-synthesized organoclay to remove amoxicillin andampicillin from single and binary (two components) aqueoussystems. As far as we are aware, this is the   rst study demon-strating binary adsorption of amoxicillin and ampicillin using pristine and myristyltrimethylammonium cation-intercalated(organo) montmorillonite with special attentions to adsorptionequilibria and thermodynamic aspects. We propose a modi  -cation on the extended-Langmuir model by introducing surfacecoverage for analyzing binary adsorption equilibrium data.Ultimately, batch adsorption tests on real pharmaceutical wastewater are demonstrated, along with regenerability evalua-tion of the clay sorbent. Experimental section Chemicals  Antibiotics used in this study ( i.e. , amoxicillin trihydrate andampicillin trihydrate) were kindly provided by a local pharma-ceutical industry with minimum purity of 97% and 95%,respectively. The molecular structure and some speci  c infor-mation about these compounds include their environmentalpersistence data 22,23 are presented in Table 1. Analytical-gradechemicals include myristyltrimethylammonium bromide(MTAB) cationic surfactant (99%), hydrogen peroxide solution(30%), sodium chloride (99.5%), hydrochloric acid (37%), silvernitrate (99.8%) and potassium hydroxide (85%) were purchasedfrom Sigma-Aldrich, Singapore and used as-supplied. Doubledistilled water (DDW) was used throughout the experiments. Preparation of adsorbent materials Montmorillonite lumps as the starting material were collectedfrom one of mining sites located at Pacitan town, East Java. A    er the collection, the solid was repeatedly washed with tap water to remove coarse particles and water-soluble impurities.Then, the solid was dispersed in dilute hydrogen peroxidesolution with a solid/solution ratio of 1 : 10 (w/v) and thesuspension was aged for 2 h under mechanical stirring at 500rpm.Thesolidwaswashedwithdoubledistilledwaterand0.1NNaOH solution alternately until the pH of the washing solution was near-neutral. The clay material was oven-dried at 383.15 Kand stored in an airtight plastic bag for further characteriza-tions. The mineralogical analysis of clay material was con-ducted based on the size fractionations method 24 and theresults are: 72% smectite, 4% quartz, 12% feldspar, 8% calcite,3% anatase and 1% others.To enhance the monoionic nature of clay,      y grams of clay  was treated with 250 mL 1 N NaCl solution (   ve cycles stirred at 500 rpm for an hour in each cycle) and washed until negativereaction of chloride ions with silver nitrate was obtained. Theresulting clay (denoted as Na-MMT) was oven-dried at 373.15 Kfor 24 h, pulverized and screened with 100/120 sieves to obtainsize fractions of 0.125 – 0.150 mm. The cation exchange capacity (CEC) of Na-MMT was 74.2 m eq./100 g of clay, as measured by methylene blueindex following ASTM C837-99test method. Themetal oxide compositions of Na-MMT were analyzed using aPANalytical MiniPal QC energy dispersive X-ray    uorescence(EDXRF) spectrometer and the results are shown as follows:SiO 2  of 61.28%; Al 2 O 3  of 18.33%; Na 2 O of 2.47%, K 2 O of 1.75%,MgO of 2.16%, CaO of 1.59%, MnO of 0.27%, Fe 2 O 3  of 3.35%,and TiO 2  of 0.08%.The preparation procedure of organo-montmorillonite(designated as O-MMT) was described as follows: 10 g Na-MMT was dispersed in 100 mL MTAB solution with a surfactant concentration equivalent to 1.5 times of the CEC of clay. Thesuspension was aged for 1 h under stirring at 500 rpm. Then, it  was placed in an Inextron WDS900DSL23-2 microwave oven andirradiated over 5 min at a frequency of 2.45 GHz and an output power of 500 W. The resulting solid was washed with doubledistilledwaterseveraltimesuntilitwasfreefrombromideanions(tested by titration with 0.1 M AgNO 3  solution). The product wasdried in an air-circulating oven at 383.15 K to constant weight,pulverized and sieved. The CEC of O-MMT was 14.9 m eq./100 g of clay according to methylene blue adsorption index. Characterizations of adsorbent materials Scanning electron microscopy (SEM) was performed to probethe microtopography and surface texture of the adsorbents. Thescanning was conducted on a JEOL JSM-6390   eld emission This journal is © The Royal Society of Chemistry 2014  RSC Adv. , 2014,  4 , 16298 – 16311 |  16299 Paper RSC Advances  SEM at an accelerating voltage of 20 kV. Surface characteriza-tions were conducted by physical adsorption – desorptionisotherms of N 2  at 77.15 K, on a Micromeritics ASAP 2010automated sorptometer. The samples were vacuum-outgassedunder a   ow of pure helium at 10  3 Torr and 473.15 K for 24 h.The speci  c surface area, micropore volume ( V  mic ) and external(mesoporous) surface area ( S ext  ) was determined from theadsorption branches applying the Brunauer – Emmett  – Teller(BET) and  t  -plot method, respectively. The pore size distribution was derived from desorption data by means of Barrett  –  Joyner – Halenda (BJH) method. Total pore volume ( V  T ) was estimatedfrom the volume of gas adsorbed at a relative pressure (  p /  p  )of 0.99.The pH of point of zero charge (pH pzc ) was determined by pH-dri    technique following Yang   et al. 25 method. Brie   y, asolution of 0.005 M CaCl 2  was boiled to remove dissolved CO 2 and then cooled to room temperature. A 20 mL aliquot of thesolution was poured into a series of capped vials. The pH wasadjusted by adding 0.5 M HCl or 0.5 M NaOH solution to a valuebetween 2 and 11. A known amount of Na-MMT or O-MMT(  0.05 g) was added and the suspension was equilibrated for24 h. The   nal pH was measured using a SevenEasy  ™ digitalpH-meter (Model S20, Mettler Toledo) and plotted against theinitial pH. The pH at which the curve of pH  nal  versus  pH initial crosses the line pH initial  ¼  pH  nal  is marked as pH pzc . Theresults are 5.82 for pH pzc  of Na-MMT and 7.18 for pH pzc  of O-MMT.Thermal decompositionanalysiswasperformed ona Mettler-Toledo TGA/DSC 1 thermogravimetric analyzer. Approximately 10 mg of the samples was spread uniformly at the bottom of alumina crucible. The temperature of furnace was programmedto rise from room temperature to a   nal temperature of 1123.15K at 20 K min  1 in a dynamic high-purity    owing N 2  of 100 mLmin  1 . The elemental contents of Na-MMT and O-MMTmaterials were determined using an automated CHNS/Oelemental analyzer (Model 2400-II, PerkinElmer). FT-IR analysis was carried out on a Shimadzu FTIR 8400S spectrometer using KBr disk technique. The spectra data were collected by accu-mulating200scansoverwavenumberrangeof4000 – 500cm  1 inthe transmission mode at a spectral resolution of 4 cm  1 . Dataprocessing includes baseline adjustment, normalization andspectral smoothing was performed using IRsolution so    ware(Version 1.21). The mineralogical compositions of the solids were analyzed using a Philips PANalytical X'Pert X-ray di ff  rac-tometer. The powder di ff  ractograms of the specimens wereacquired at 40 kV and 30 mA in the range of 2-theta angles of 2 – 70   with a scanning speed of 1  /min. The radiation source wasNi-  ltered Cu K a 1  ( l ¼ 0.15405 nm). Batch adsorption experiments  –  single solute systems Fresh antibiotic e ffl uents were prepared by dissolving 0.3 g amoxicillin or ampicillin into 1 L deionized water to give aninitial concentration of 0.80 mmol L  1 for amoxicillin and 0.82mmol L  1 for ampicillin. For the adsorption equilibriumexperiments, the stock e ffl uents of amoxicillin or ampicillin were pouredintoa seriesofstoppered conical  asks (each of100mL) containingNa-MMT or O-MMT withvarying doses (0.1 – 1 g).The   asks were wrapped with aluminium foil to eliminate light interference. Then, the   asks were placed in a thermostatedreciprocal shaker and shaken at 100 rpm for 24 h. Preliminary experimentsindicatedthat24hprovidedsu ffi cienttimetoreachequilibrium. The system temperature was held constant at 303.15 K, 313.15 K and 323.15 K by a built-in PID-type temper-ature controller. A    er equilibration, the clay suspension wascentrifuged at 3000 rpm for 10 min and the supernatant wastaken for analysis. The residual concentration of solute wasquanti  ed by a double beam UV-Vis spectrophotometer at a Table 1  Molecular structure and some speci fi c information about amoxicillin and ampicillin  Amoxicillin AmpicillinMolecular structurePhysical state White to o ff  -white solid White to o ff  -white solidMolar mass (g mol  1 ) 365.4 349.4Chemical formula C 16 H 19 N 3 O 5 S C 16 H 19 N 3 O 4 S Water solubility, 25   C (g L  1 ) 3.43 10.1p  K  a  2.4 (carboxylic) 2.7 (carboxylic), 7.3 (amine)7.4 (amine)9.6 (phenol)Environmental persistence dataPhotodegradation rate (s  1 )Direct  a 5.24  10  7 Not availableHydrolysis rate (s  1 ) b 4.45  10  7 2.15  10  7 a Direct photodegradation at pH 7 using a solar simulator system.  b Hydrolysis at pH 7 and room temperature (298.15 K); the hydrolysis rateconstants correspond to half-lives of 18 d for amoxicillin and 36 d for ampicillin. 16300  |  RSC Adv. , 2014,  4 , 16298 – 16311 This journal is © The Royal Society of Chemistry 2014 RSC Advances Paper  detection wavelength of 252.2 nm for amoxicillin and 245.8 nmforampicillin.Thecalibrationcurveswerepreparedfromasetof    ve standard solutions with concentration range of 50 – 300 mg L  1 . Prior to spectrophotometric measurements, all superna-tants were   ltered through a 0.45  m m syringe   lter. The amount of solute adsorbed per unit mass of adsorbent at equilibrium( q e , mmol g   1 ) was determined by the following equation: q e  ¼ C  0  C  e m V   (1) where  C  0  and  C  e  are the initial and equilibrium concentrationsof solute in the liquid phase (mmol L  1 ), respectively,  m  is themass of adsorbent (g) and  V   is the volume of solution (L). Forthe pH adsorption edge experiments, the suspension pH was variedfrom2to11andadjustment wasmadebyadding1NHClor 1 N KOH solutions. All adsorption runs were replicated twice with averages used as the results. Batch adsorption experiments  –  binary solute systems For the binary adsorption experiments, three synthetic e ffl uentscontaining amoxicillin and ampicillin were prepared (Table S1of the ESI † ). Adsorption isotherm experiments were performedin a closed batch system by equilibrating the synthetic e ffl uentscontaining a known amount of O-MMT on a reciprocal shakerfor 24 h at room temperature. The initial pH of all e ffl uentsranged between 6 and 7. The equilibrium concentration of remaining antibiotics was determined spectrophotometrically in a multi-component quantitation mode at two measurement  wavelengths of 245.8 nm and 252.2 nm. Five mixed samples with pure amoxicillin and ampicillin standards were made toconstruct the calibration curve. The following mathematicalformula was used to calculate the equilibrium amount of solutes  i   and  j   in the adsorbed phase: q e ; i  =  j   ¼  C  0 ; i  =  j   C  e ; i  =  j   m   V   (2) where  q e, i   and  q e,   j   are the equilibrium loading of solutes  i   and  j  in the solid phase (mmol g   1 ),  C  0  and  C  e  refer to the initial andequilibrium concentrations of solute in the solution (mmolL  1 ),  m  isthemass ofO-MMTused(g)and V   isthe volumeofthee ffl uents (L). Results and discussion Textural properties and surface chemistry of Na-MMT and O-MMT materials The electron micrographs of Na-MMT and O-MMT are shown inFig. 1. SEM analysis con  rmed that Na-MMT and O-MMT areboth crystalline solids of micrometer size. The lump alikemorphology of Na-MMT with smooth surface characteristic isclearly seen from Fig. 1a. In comparison, the SEM image of O-MMT displays the agglomerated sheet alike morphologicalfeature and surface roughness (Fig. 1b). N 2  adsorption – desorption isotherms (Fig. S1 of the ESI † ) ascertain that bothclay sorbents are highly mesoporous with mixed micro- andmeso-sized pores. The characteristic type H4 hysteresis loopobserved in the relative pressure range of 0.4 – 0.6 is theindication of an adsorption phenomenon of gases typical forcomplex micro-mesoporous solids, which include micropores  lling, pore condensation and cavitation-induced evaporationmechanisms. 26 The BET speci  c surface area of Na-MMT was122.2 m 2 g   1 and this value dramatically fell to 65.8 m 2 g   1 of O-MMT. Similarly, total pore volume of Na-MMT was 0.11 cm 3 g   1  while that of O-MMT was 0.06 cm 3 g   1 (Table S2 of the ESI † ).The decreased BET speci  c surface area and pore volumerevealed that some interior adsorption sites became inacces-sible by N 2  molecules due to the blocking of large surfactant cationswithinthepores.The poresizedistributioncurves(inset Fig. S1 † ) support N 2  adsorption – desorption results that highpercentage of mesopores with a diameter about 3 – 4 nm exist inNa-MMT and O-MMT. Furthermore, a notable distribution of pore sizes outside the range of 3 – 4 nm was observed in O-MMT,likely due to the surfactant cations loading into the interparticlepores within the  ‘ house-of-cards ’  structure that enlarge thecorresponding pore size. This is consistent with other studiesdealing with organoclay preparation employing long alkyl-chaincationic surfactants. 27 – 29 Con  rmation of the organi  cation of Na-MMT was also shown from elemental analysis results in theESI Table S3. †  In this table, it can be seen that O-MMT containsabout 0.83 wt% N and 12.1 wt% C (the C/N ratio is 14.58) wherethe presence of carbon and nitrogen atoms in Na-MMT is not observed. Based on the carbon and nitrogen contents, it can beestimated that each gram of clay contains 0.59 mmol of inter-calated C14-trimethylammonium cations.Thermal gravimetric analysis was used to examine the weight loss arising from organic content and the relateddegradation mechanisms. The TGA curves in the ESI Fig. S2 † show that the weight loss by about 7% below 150   C corre-sponds to the loss of surface water and water associated withNa-MMT micropore structure. Between 200   C and 400   C,about 0.0063 g H 2 O per g clay was lost, which might be attrib-uted to desorption of water from the interlayer space. Irrevers-ible dehydroxylation of the layered silicate structure takes placein the temperature range of 600 – 700   C. The weight change of the clay could be neglected above 700   C. The presence of organic moieties increases the number of decomposition stepsfor organoclay. As illustrated in Fig. S2, †  the TGA pro  le of O-MMT indicates four weight-loss steps: (I) the loss of water(dehydration) that occurs at 110   C and ends at 150   C; (II)decomposition of the bonded structure of organic modi  er inthe interlayer space at 200 – 300   C; (III) dehydroxylation of thesilicate layers around 600   C and proceeds till around 700   Cand (IV) further decomposition of the organic surfactant at  Fig. 1  SEM images of Na-MMT (a) and O-MMT (b). This journal is © The Royal Society of Chemistry 2014  RSC Adv. , 2014,  4 , 16298 – 16311 |  16301 Paper RSC Advances  720 – 800   C. The thermogram pattern of bare MTAB indicatesthe weight loss at 110   C resulted from dehydration, followed by structural degradation that occurs at 250 – 400   C. Finally, themaximum decomposition takes place around 600   C due to theincomplete oxidation of the organic moieties under N 2 atmosphere.The spectral characteristics of Na-MMT and O-MMT aredisplayed in the ESI Fig. S3. †  Several infrared absorption bandsof Na-MMT were observed at speci  c wavenumbers, which arethe typical of montmorillonitic mineral: 3614 cm  1 of O – Hstretching of structural hydroxyl groups located at the surfaceand along the broken edges, 3342 cm  1 and 1636 cm  1 of stretching and bending vibrations of OH group in water mole-cules, 1087 cm  1 of Si – O stretching, 936 cm  1 of Al –  Al – OHhydroxyl-bending vibration, 522 cm  1 of Al – O – Si bending  vibration and 475 cm  1 of Si – O – Si bending vibration. Theinsertion of organic modi  er (MTA  + cation) into the interlayerspacing gave rise to symmetric and asymmetric sp 3 C – Hstretching vibrations of methyl and methylene groups at 2900 – 2800 cm  1 and symmetric sp 3 C – H bending vibration at 1464cm  1 . It can be shown that the vibrational bands correspond toSi – O stretching, Al –  Al – OH bending, Al – O – Si bending and Si – O – Si bending between Na-MMT and O-MMT are essentially iden-tical. This suggests that the unit-cell framework of montmoril-lonite mineral (tetrahedral-octahedral-tetrahedral layeredsheets) remains intact during microwave irradiation. On theother hand, the absorption intensities of stretching andbending vibrations of hydroxyl group at   3400 cm  1 and  1600cm  1 dropped, which might be attributed to the removal of adsorbed water molecules from the clay lattice.X-ray di ff  ractograms of Na-MMT and O-MMT are given in theESI Fig. S4. †  Here, the XRD pattern of MMT lump was not shown due to its similar characteristic to that pattern of Na-MMT. A broad (001) re  ection was noted at 2-theta of 6.44  forMMT lump and 6.49  for Na-MMT, characterizing the basalspacing of montmorillonite. This information suggested that the transformation of MMT lump to Na-MMT through cationexchange did no or little alteration on the mineralogical prop-erties of clay. The occurrence of other crystalline phases such asquartz, calcite, feldspar and anatase was observed in addition tomontmorillonite crystal planes and basal re  ections. Semi-quantitativemineralanalyses ofMMTlumpandNa-MMT basedon XRD data showed that these clay impurities accounted for20 – 22% of total crystalline phases (the purity of smectite phase was 78 – 80%). The intercalation of MTA  + cations into themontmorillonite structure leads to the expansion of basalspacing from 1.36 nm to 1.95 nm and interlayer spacing from0.39 nm to 0.98 nm. The interlayer spacing was determined by subtracting the measured basal spacing with the unit-cellthickness of a single tetrahedral-octahedral-tetrahedral layeredsheet of montmorillonite, which is 0.97 nm. 30 The structuralconformation of the surfactant in the interlayer spacing can beinterpreted by considering the magnitude of increased basalspacing and the molecular structure of intercalated surfactant cations (the thickness of polar  ‘ head ’  and apolar  ‘ tail ’ ). Theloading of MTA  + cation, a single C14 alkyl chain trimethy-lammonium surfactant with a concentration of 1.5 times of theCECwouldresultinthe pseudotrilayerconformation; thatisthealkyl chain of surfactant is packed in parallel to the plane of silicate tetrahedral sheets with interlocked-chains array. 27,31 Theequivalence exchange between Na + and MTA  + cations alsorenders the clay surface to be organophilic, which is suitable tosorb organic compounds such as antibiotics. E ff  ects of solution pH on the adsorptive removal of amoxicillin and ampicillin The pH is a crucial factor both towards the surface chargedensity of adsorbent and the ionic speciation of adsorbate inthe liquid phase, which determines the e ff  ectiveness of asorption process. The presence of acid (carboxylic) and base(amino) surface functional groups within amoxicillin andampicillin structure contributes to the amphoteric nature of these drugs, in which these groups are ionisable following thepH changes. As shown in Table 1, amoxicillin has three aciddissociation constants of 2.4 (p  K  a1  of carboxylic), 7.4 (p  K  a2  of amine) and 9.6 (p  K  a3  of phenol) while for ampicillin, the p  K  a  values are 2.7 (carboxylic) and 7.3 (amine). 14  Accordingly,amoxicillin species are mainly as a cation in the acidic solution(pH < 2.4), a zwitterion between pH 2.4 and 7.4 and an anion inthe alkaline solution (pH > 7.4). Similarly, ampicillin existspredominantlyincationic, zwitterionic andanionic forms atpH< 2.7, 2.7 < pH < 7.3 and pH > 7.3, respectively (see Fig. S5 of theESI † ). The distribution diagrams showing the percentage of amoxicillin and ampicillin species at room temperature underdi ff  erent solution pHs was presented in the ESI Fig. S6. † The in  uence of pH on the adsorbed amount of amoxicillinand ampicillin in single systems is given in Fig. 2. In this  gure,the increasing amount of amoxicillin or ampicillin adsorbed by Na-MMT was seen with the increase of pH from 2 to 7 and thena progressive decrease on the removal percentage was encoun-tered at pH above 7. Similar observation was reported by Moussavi  et al.  for the removal of amoxicillin from water using NH 4 Cl-induced activated carbon. 32 The limited uptake at low pH, particularly below p  K  a1  of amoxicillin or ampicillin was dueto net repulsion between positively charged edge hydroxylsurfaces of montmorillonite crystallites (silanol and aluminolsites) and the cationic adsorbate molecules. From the specia-tion diagrams in Fig. S5, †  it can be shown that at low pH range(pH 2 – 3), the cationic amoxicillin and ampicillin accounted for84% and 86% of total species in the solution, respectively. Theremoval processes of amoxicillin (  24%) and ampicillin(  27%) could still take place in the acidic environment due tothe dipole-induced interaction between protonated aminegroup and the Si-tetrahedral basal oxygen surface in addition toion exchange between Na + interlayer cations and protonatedamoxicillin or ampicillin. The latter phenomenon (cationexchange) has been veri  ed to be the sorption controlling mechanism at low pH in the studies conducted by Jiang   et al. 33 and Wang   et al. 34 for the removal of cipro  oxacin using layeredmanganese oxide and Ca-montmorillonite, respectively. With the increase of pH approaching p  K  a2  of the adsorbate,the ratio of zwitterion to cation increases gradually for bothantibiotics and this leads to the increase of removal percentage, 16302  |  RSC Adv. , 2014,  4 , 16298 – 16311 This journal is © The Royal Society of Chemistry 2014 RSC Advances Paper
Related Search
Similar documents
View more
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!