Characterization Of Selected Polymeric Membranes Used In The Separation .
membranesArticleCharacterization of Selected Polymeric MembranesUsed in the Separation and Recovery ofPalladium-Based Catalyst SystemsBongani Michael Xaba 1 , Sekomeng Johannes Modise 1 , Bamidele Joseph Okoli 1,2, * ,Mzimkhulu Ephraim Monapathi 1 and Simphiwe Nelana 112*Chemistry Department, Faculty of Applied and Computer Sciences, Vaal University of Technology, PrivateBag X021, Vanderbijlpark 1911, South Africa; BXaba@mpact.co.za (B.M.X.); joe@vut.ac.za (S.J.M.);monapathimz@gmail.com (M.E.M.); simphiwen@vut.ac.za (S.N.)Department of Chemical Sciences, Faculty of Science and Technology, Bingham University, Karu PMB005,Nasarawa State, NigeriaCorrespondence: okolibj@binghamuni.edu.ng; Tel.: 27-767619418Received: 30 May 2020; Accepted: 3 July 2020; Published: 28 July 2020 Abstract: Membrane separation processes tender a capable option for energy-demanding separationprocesses. Nanofiltration (NF) and reverse osmosis (RO) membranes are among the most explored,with a latent use in the chemical industry. In this study, four commercial membranes (NF90, NF270,BW30, and XLE) were investigated for their applicability based on the key structural performancecharacteristics in the recycling of Pd-based catalysts from Heck coupling post-reaction mixture. Purewater and organic solvent permeabilities, uncharged solute permeability, swelling, and catalystrejection studies of the membranes were conducted as well as the morphological characterizationusing Fourier transform infrared, field emission gun scanning electron microscopy, and atomic forcemicroscopy. Characterization results showed trends consistent with the manufactures’ specifications.Pure water and organic solvent fluxes generally followed the trend NF270 NF90 BW30 XLE,with the solvent choice playing a major role in the separation process. Pd(PPh3 )2 Cl2 was well rejectedby almost all membranes in 2-propanol; however, XLE rejects Pd(OAc)2 better at high pressure inacetonitrile. Our study, therefore, revealed that the separation and reuse of the two catalysts byNF90 at 10 bar resulted in 97% and 49% product yields with 52% and 10% catalyst retention forPd(OAc)2 while Pd(PPh3 )2 Cl2. gave 87% and 6% yields with 58% and 36% catalyst retention in thefirst and second cycles, respectively. Considering, the influence of membrane–solute interactionsin Pd-catalyst rejection, a careful selection of the polymeric membrane and solvent, a satisfactoryseparation, and recovery can be achieved.Keywords: palladium-based catalysts; polymeric membranes; separation; NF/RO membranes1. IntroductionSeparation technology has evolved during the 20th century, driven primarily by advances in thepetroleum industry. Numerous technologies such as distillation, extraction, and adsorption have beenuniversally used [1]. The use of membranes in nonaqueous media has drawn a lot of attention in recentyears owing to their inferior energy demands and ease of use [2]. A promising field, particularly inpressure-driven processes, is organic solvent nanofiltration (OSN) [3]. Lab-scale and commercial-scaleapplications of OSN membranes have been reported [4,5]. In the advance and application of themembrane process, characterization of modelling and optimization are essential steps [6,7]. A consistentand reliable method of measuring the separation performance of membranes is essential and allowsend-users to make an informed selection [8]. Membrane characterization parameters may be describedMembranes 2020, 10, 166; membranes
Membranes 2020, 10, 1662 of 21as either performance related or morphology related. Performance-related parameters describemembrane functionality such as flux, rejection, and molecular weight cutoff (MWCO). Morphologicalparameters which include physical and chemical parameters describe the structure of the membrane [9].Important morphological parameters include porosity and roughness of the membrane.Molecular weight cutoff is described as the molecular weight for which 90% rejection of thesolute is achieved by the membrane [10]. The MWCO concept is based on the observation thatmolecules generally get larger as their masses increase. As molecules get larger, sieving effects due tosteric hindrance increase and a larger molecule is rejected by the membrane more than the smallermolecule. MWCO may also be related to diffusion since larger molecules diffuse more slowly thansmaller molecules [11]. Hilal et al. [12] suggested that MWCO determination depends on experimentalconditions such as the nature of the feed solution and the type of membrane module. They showedthis by using a mixture of polyethyleneglycols (PEG) with different molecular weights. Their resultsalso showed that some membranes have larger and smaller pores respectively than specified bythe manufacturer.Porosity has been regarded as another useful parameter to describe separation in membranes.Porosity is usually expressed as pore size, density, pore size distribution (PSD), or the effective numberof pores in the membrane’s upper active layer [13]. Roy, Sharqawy, and Lienhard [14] have theoreticallyprobed the effects of log-normal pore size distribution on the rejection of uncharged solutes and NaCl.They showed that the theoretical log-normal function is not apposite for nanofiltration (NF) membranesdue to the large pore size tail of the distribution dictating rejection and flux. Their results also showedthat elucidation based on uncharged solute data alone cannot give functional quantitative informationabout the membrane pore size distributions. However, when used in conjunction with other surfacecharacterization techniques, they showed good agreement in pore size distribution.Košutić et al. [15] investigated the porosity of NF and reverse osmosis (RO) membranes bypermeation of uncharged compact organic molecules. To ascertain the influence of the porous structureof the membrane skin on the retention mechanism of different solutes, the PSDs were determined atalmost 690 kPa. Their results showed a distinction between the PSD of NF and RO membranes. The ROmembranes revealed a wide PSD and bimodal distribution, with maxima at 0.52 and 0.80 nm. The PSDof NF membranes exhibited maxima at larger pore sizes, the first one between 0.95 and 1.10 nm andthe second maximum around 1.55 nm. This was clear evidence of the existence of larger surface pores.Interest in the application of membrane technology in homogeneous catalyst separation wasevident in the period from 2000 to 2002. The effective application of coated polymeric ultrafiltrationmembrane for the photocatalytic degradation by organic pharmaceuticals [16] and the cationicphenothiazine dye remediation using optimized polyelectrolyte assisted ultrafiltration [17] wassome of the reported areas of interest. Other studies used solvent-resistant NF (SRNF) membranes.Of these, the polymer-based membranes suggest persuasive prospects for SRNF with unparalleledcost-efficiency [18]. Vandezande, Gevers, and Vankelecom [10], however, highlighted the dilemmaof comparing retention data from different studies, as these are application specific. The differentproperties of solvents, membranes, filtration mode, and operating conditions affect the results inmembrane separations.The early applications of OSN technology were involved in the recovery and reuse of thehigh-value palladium (Pd) catalyst in Heck, Sonogashira, and Suzuki reactions [19,20]. These processesrecorded major successes in the removal of about 95% of residual Pd and separation of the productfrom the catalyst and ionic liquid after Suzuki coupling reactions [21,22]. However, the size of thiscatalyst is the same as the product and may result in poor separation. Further, the premium utilizationof Pd is as pharmaceutical chemicals and active pharmaceutical ingredients (APIs) and the dailypermitted oral exposure to Pd in a pharmaceutical ingredient of less than 10 mg of Pd per kilo of API( 10 ppm) [23]. The premium applications of palladium in synthetic biology as an in vivo catalyst [24],the pharmaceutical industry, and electrocatalysis [25] have become a subject of interest to mostresearchers in recent years. Hence, palladium was dubbed the king of transition-metal catalysts [26].
Membranes 2020, 10, x FOR PEER REVIEW3 of 23catalyst [24], the pharmaceutical industry, and electrocatalysis [25] have become a subject of interestto most researchers in recent years. Hence, palladium was dubbed the king of transition-metalMembranes3 of 21catalysts 2020,[26].10, 166In light of this, it was of interest to investigate the separation characteristics of commonpolymeric membranes in the recovery of palladium (Pd) catalysts in aqueous and organic media withIn light of this, it was of interest to investigate the separation characteristics of common polymerican attempt to separate and reuse the catalysts from Heck coupling post-reaction mixture. Themembranes in the recovery of palladium (Pd) catalysts in aqueous and organic media with an attemptmembranes were characterized for pure water permeability, pure solvent permeability, surfaceto separate and reuse the catalysts from Heck coupling post-reaction mixture. The membranes weremorphology, chemical structure, and uncharged solute rejection measurements. The separationcharacterized for pure water permeability, pure solvent permeability, surface morphology, chemicalperformance of different membranes in different solvents was studied. This work will serve as anstructure, and uncharged solute rejection measurements. The separation performance of differentarchetype for evaluating NF/RO membranes' performance in the recovery and reuse of palladiummembranes in different solvents was studied. This work will serve as an archetype for evaluating(Pd) catalysts. The data will help identify the suitable combination of membranes and solvents forNF/RO membranes’ performance in the recovery and reuse of palladium (Pd) catalysts. The datause in order to achieve effective palladium (Pd) catalyst separation and recycling. It will also shedwill help identify the suitable combination of membranes and solvents for use in order to achievesome light on the key performance-related or morphology-related parameters responsible foreffective palladium (Pd) catalyst separation and recycling. It will also shed some light on the keyeffective catalyst separation and recycle.performance-related or morphology-related parameters responsible for effective catalyst separationand recycle.2. Experimental2. Experimental2.1. Instrumentation2.1. InstrumentationThe determination of sugar and alcohol concentrations was achieved by the use of Lambda 25UV-VISWaltham,MA, USA) and500byGasThe spectrometerdetermination(Perkin-Elmer,of sugar and alcoholconcentrationswas Clarusachievedthe Chromatographuse of respectively.Membranecharacterisationwas25 UV-VIS spectrometer (Perkin-Elmer, Waltham, MA, USA) and Clarus 500 Gas Chromatographdone on the followinginstruments:Field EmissionGun ScanningElectronMicroscopy (FEG-SEM)(Perkin-Elmer,BridgeportAvenue, Shelton,USA), respectively.Membranecharacterisationwas done(CarlSMT instruments:GmbH, Peabody,MA, USA),Atomicforcemicroscopy(AFM)(FEG-SEM)(Park ScientificontheZeissfollowingField nts,Janderstrasse,Mannheim,Germany),Spectrum400 (ParkFourier-transforminfraredZeissSMT GmbH,Peabody, MA,USA), Atomicforce andmicroscopy(AFM)Scientific Instruments,Spectrophotometer(FTIR)Germany),fitted withanda universalattenuatedtotal reflection(ATR)Spectrophotometersampler (PerkinJanderstrasse,Mannheim,Spectrum400 Fourier-transforminfraredElmer, Waltham,USA) attenuated total reflection (ATR) sampler (Perkin-Elmer, Waltham,(FTIR)fitted withMA,a universalMA, USA)2.2. Chemicals Reagents2.2. Chemicals ReagentsFive solvents were chosen for the study. These are acetonitrile, methanol, ethanol, and 2Five solventswere chosenforgrade,the study.acetonitrile,methanol,ethanol,usedand 2-propanol,propanol,all analyticalreagentandThesewater.areThesesolventsare commonlyin organicallanalyticalgrade,becauseand water.Thesesolventsare commonlyused in alsoorganicsynthesissynthesisandreagentwere chosenof theirsolvatingproperties.The solventsrepresentthe asses of solvents, those which coordinate via oxygen and those coordinating via selective donorthosecoordinateoxygenand andthosecoordinatingselective donoratomssuch as nitrogen,atomswhichsuch asnitrogen,viaknownas oxicanoxicsolvents,viarespectively.Waterand 2-propanolwereknownoxic and anoxicsolvents, respectively.Waterand 2-propanolwereinusedfor membraneused forasmembranecharacterization.Acetonitrile and2-propanolwere eandof2-propanolweresolventsused in catalystseparationphysicalstudies. The physicalpropertiesthe respectiveare givenin Table studies.S1 of theTheSupportingpropertiesof[27].the respective solvents are given in Table S1 of the Supporting Information [27].InformationUncharged solutes, glucose and sucrose, were also selected for the determinationdetermination of molecularweightthethemembranes.Thesesoluteswere weresuppliedby Merck.Two transition-metalcatalystsweight cutoffcutoffofofmembranes.Thesesolutessuppliedby Merck.Two transition-metal 1wereusedwerein thestudy,namely(II) acetateof molecularweight 224g molcatalystsusedin thestudy,palladiumnamely palladium(II)complexacetate complexof molecularweight224-1 and osphine)palladium(II) chloridecomplexcomplex(Sigma-Aldrich)of molecularweightg.mol(II) chloride(Sigma-Aldrich)of molecular 1-1701.91mol g(Figure1).weightg701.91mol (Figure1).P-OO- Pd -PClCl- Pd bis(triphenylphosphine)palladium (II) chlorideFigure 1. Structures of the catalysts used.OOpalladium (II) acetate
Membranes 2020, 10, x FOR PEER REVIEWMembranes 2020, 10, 1664 of 234 of 21Figure 1. Structures of the catalysts used.2.3.2.3. MembranesMembranesFourcommercially usedfor studythis studyFour commerciallywereusedfor isticssupplied byand XLE). These were thin-film composite membranes of various characteristics ftheSupportingInformation).Dow/FilmTec (Minneapolis, MN, USA) (Table S2 of the Supporting Information).2.4. Analytical Procedure2.4. Analytical procedureA bench-scale stainless steel dead-end module with a capacity of 1.2 litres was operated atA bench-scale stainless steel dead-end module with a capacity of 1.2 litres was operated atpressures of 25 bar with nitrogen gas was used (Figure 2).pressures of 25 bar with nitrogen gas was used (Figure 2).Figure 2. Dead-end filtration unit used for retention measurements [28].Figure 2. Dead-end filtration unit used for retention measurements [28].The unit was fitted with a Teflon-coated magnetic stirrer supported on the upper lid by a steelThe unitwaswasfitted witha Teflon-coatedmagneticsupportedon the upperlid by a steelrod. Stirringrequiredto homogenizethe sampleandstirrerto minimizeconcentrationpolarization[28].rod. samplesStirring wasrequiredto homogenizesampleand toofminimizepolarizationDiscof thedifferentmembranes thewitha diameter9 cm andconcentrationan effective areaof 0.0064[28].m2Disc cutsamplesof the differentmembraneswith Thea diameter9 cm andan effectiveof 0.0064m2wereand placedon a poroussupport erewascut 1andplacedon a The permeatewassupportcollectedfroma Tefloninto aunderneathmeasuring thecylinder.Filtrationdisc was 1 ml.wereThe permeatefromsolutionsa Teflon withtube aintoa measuringFiltrationmeasurementsperformedwasby collectedloading feedvolumerangingcylinder.from 250–600mL havolumerangingfrom250–600mlatat 24 C. The first 20 mL of permeate collected was discarded. Thereafter, 10 mL of permeate was24 eafter,10mlofpermeatewascollectedcollected at a specified time. The flux was obtained by Equation (1):at a specified time. The flux was obtained by Equation (1):VJ (1)A·t(1) . area, and t is the time.where V is the volume of permeate, A is the membranewhere V is the volume of permeate, A is the membrane area, and t is the time.2.5. Membrane Swelling ExperimentThe interactionof thesolvents with the membrane physical structure was further investigated2.5. MembraneSwellingExperimentby measuring the swelling tendency of the membranes. Membranes were cut and dried at roomThe interaction of the solvents with the membrane physical structure was further investigatedtemperature in an open dish. Each dried membrane was weighed and immersed in the selectedby measuring the swelling tendency of the membranes. Membranes were cut and dried at roomsolvents. After an equilibrium time of approximately 30 minutes, the membrane was removed fromtemperature in an open dish. Each dried membrane was weighed and immersed in the selectedthe solvent and quickly dried with a soft tissue to remove the solvent from the external surface beforesolvents. After an equilibrium time of approximately 30 minutes, the membrane was removed fromweighing. Swelling of the membrane was calculated by Equation (2) [29]:the solvent and quickly dried with a soft tissue to remove the solvent from the external surface beforeweighing. Swelling of the membrane was calculatedbyWEquation(2) [29]:dry1 Wwet Q (2)ρsWdry
Membranes 2020, 10, 1665 of 21In the equation, Q is the swelling, Wwet is the mass of wet membrane, Wdry is the mass of drymembrane, and ρs is the density of the solvent.2.6. Membrane CharacterizationFeed solutions of uncharged solutes were prepared with concentrations of 0.1 vol % for the alcoholand 0.1 wt% for the sugars. The concentrations of the feed and permeate (sugars and alcohols) wereestimatedby the Anthrone method [30] and gas chromatography, respectively. Characterizationof theMembranes 2020, 10, x FOR PEER REVIEW5 of 23surface morphology and chemical structure of the polymer gave information on the specific chemistryand orientation of the structure of the functional Wgroups Wpresent in the membrane active layer [31].Q 2.7. Catalyst Rejection1ρswetdry(2)Wdrymass of (Tablewet membrane,Wdry filteredis the massof anddry 20 bar.In the ofequation,Q isthe doneswelling,Wwet is theDissolutioncatalystswasin varioussolvents1) and thenat 10membrane,andρs is the density of the solvent.The concentration and rejection coefficient of the catalyst in the permeate and feed solutions weredeterminedby UV-VIS spectroscopy and Equation (3), respectively:2.6. Membrane CharacterizationFeed solutions of uncharged solutes wereCpreparedpermeate with concentrations of 0.1 vol % for theR concentrations 1 (3)alcohol and 0.1 wt% for the sugars. Theof the100feed and permeate (sugars and alcohols)Cfeedwere estimated by the Anthrone method [30] and gas chromatography, respectively. Characterizationthe surface morphology and chemical structure of the polymer gave information on the specificwhere Cofpermeateand Cfeed are concentrations of the catalyst in the permeate and feed, respectively.chemistry and orientation of the structure of the functional groups present in the membrane activelayer [31].!Table 1. Pure water permeability of the membranes.2.7. Catalyst RejectionMembraneAw ( ·m 2 ·h 1 ·bar 1 )Dissolution of catalysts was done in various solvents (Table 1) and then filtered at 10 and 20 bar.NF903.8The concentration and rejection coefficient of the catalyst in the permeate and feed solutions wereNF2708.9determined by UV-VIS spectroscopyand Equation (3), respectively:BW302.1XLE 1 2.4100(3)Cfeed areconcentrations of the catalyst in the permeate and feed, respectively.whereSeparationCpermeate and and2.8. CatalystReuseThe2.8.rejectionof the catalystsin a Heck coupling reaction mixture was investigated based on theCatalyst Separationand Reusehypothesis that sufficient catalyst rejection will enable separation of the catalyst from the mixture,The rejection of the catalysts in a Heck coupling reaction mixture was investigated based on thehowever,keepingthatin mindthatcatalystthe sufficientrejectionwill enableseparationinofa thecatalyst from thesolutionmixture, will beexpectedto be differentthatofthethecatalystsingle-rejectionand binary-componentsolutions. Thecouplinghowever,keeping fromin mindthatbehaviour in a multicomponentsolutionwill bereactionexpected asto bedifferent s.The couplingwas n andwas allowedto proceedas reactiondescribed wasby Nairet al. [32]with slightmodificationand wasandallowedtofor 4 to reaction6 h. Atwasthisperformedpoint, thestopped,cooledto roomtemperature,immediatelyproceed for 4 to 6 h. At this point, the reaction was stopped, cooled to room temperature, andcharged into the dead-end unit for filtration (Figure 3).immediately charged into the dead-end unit for filtration (Figure 3).Mixture is chargedinto membrane unitPermeate(product-rich)Coupling reactionRetentate-Catalyst-richFigure 3. Schematic of catalyst separation and reuse procedure.Figure 3. Schematic of catalyst separation and reuse procedure.
Membranes 2020, 10, 1666 of 21A feed sample was taken for UV-VIS analysis before filtration, after which filtration was performedat 10 bar until 70% of the volume had permeated. The retentate was also sampled for UV-VIS analysis,and catalyst rejection by the membranes was calculated according to Equation (3). This concentratedretentate solution was then transferred back to the reaction flask. Fresh reactants and solvent weretopped up for a second reaction run. The filtration protocol and reaction run were repeated for severalcycles until no further change in conversions could be observed. This procedure was repeated twotimes per membrane to get an overall concept of the efficiency of catalyst separation.2.9. Data AnalysisData analysis was performed using OriginPro 2015 Sr 1 b9.2.257, and comparisons betweendifferent membranes were carried out using a one-way analysis of variance (ANOVA). Data wereexpressed as mean SD of triplicate determinations. Significant was considered at p 0.05.3. Results3.1. Pure Water PermeabilityMeasuring the membrane’s dependence on pressure, it is possible to characterize the porosity of themembrane’s active layer [33]. Pure water permeability was investigated by using the Kedem–Katchalskymodel for irreversible thermodynamics [34]. According to the model, the relationship between purewater flux and pressure is expressed in Equation (4).Jw Aw ( P σ· π)(4)where Jw is the water flux, Aw is the membrane permeability, P is the pressure difference, σ is thereflection coefficient, and π is the osmotic pressure difference. In the case where only pure water ispresent, the osmotic pressure difference becomes zero; therefore, Equation (4) is reduced to Equation (5):Jw Aw · P(5)The results show a linear relationship between water flux and applied pressure (Figure 4).The water flux through all the membranes shows an increase with increasing pressure. Values ofAw , obtained from the slope of the model, show that NF270 has the largest pure water permeabilityfollowedNF90,XLE,PEERandREVIEWBW30. It was noted that the pure water permeability value of NF270Membranesby2020,10, x FOR7 ofis23twice that of NF90 and more than four times that of XLE and BW30 (Table 1).Pure water permeability180160Flux (l.m -2 .h -1)140NF 270NF 90BW 30XLE120100806040200510152025Pressure (Bar)Figure 4. Plot of water flux (Jw ) against pressure difference ( P) of the membranes.Figure 4. Plot of water flux (Jw) against pressure difference (ΔP) of the membranes.Consequently, it is expected that the pore size distribution would follow the same trend. FromEquation (3), the effective membrane pore radius will increase proportionally with pure waterpermeability. This is in line with the literature; NF90 and NF270 are classified as “tight” and “loose”
Membranes 2020, 10, 1667 of 21Consequently, it is expected that the pore size distribution would follow the same trend. FromEquation (3), the effective membrane pore radius will increase proportionally with pure waterpermeability. This is in line with the literature; NF90 and NF270 are classified as “tight” and “loose”membranes, respectively [35]. The results show that BW30 and XLE are similar in terms of pore sizes;however, Nghiem and Coleman [36] proposed the absence of pores in BW30. Presumptuously, XLEhas the same nonporous structure as BW30, which is also in line with the low-pressure RO membrane(LPRO) supplier’s classification [37]. NF90 has a pure water permeability which lies between that ofthe NF270 and the RO membranes. Therefore, it is expected that NF90 will behave in a similar way tothe RO membranes. Besides, the pure water permeability results agree to some extent (Table S2 of theSupporting Information).Permeate flow results at supplier’s standard test conditions show that NF90 and XLE have thelowest permeate flow followed by BW30 and NF270. However, when considering the maximum flowthrough these membranes, NF90 has the lowest maximum flow followed by BW30, XLE, and NF270.This conflicting observation points to changes in pore structure with increasing flow through eachmembrane; by implication, there is a similarity in the properties of NF90 and RO membranes (BW30and XLE). NF270 has the largest pore size with the highest flow and pure water permeability values.3.2. Organic Solvent PermeabilityThe Hagen–Poisseuille equation in Equation (6) explains the relationship between flux, pressure,and viscosity, where an increase in pressure results in a corresponding increase in flux. Hence, solventflux for viscous flow is described by Equation (6) [38].ε·r2Js 8· x·r! Pη!(6)where JS is the solvent flux, ε is the porosity, r is the average pore radius, P is the pressure difference,η is the viscosity, x is the effective membrane thickness, and τ is the tortuosity factor. Figure 5a,bMembranes 2020, 10, x FOR PEER REVIEW8 of 23showsthe plots of solvent flux vs. pressure for 2-propanol and acetonitrile, respectively.(a)(b)2-Propanol fluxNF 270NF 90BW 30XLENF 270NF 90BW 30XLE4035Flux (l.m .h )830-2-2-1-1Flux (l.m .h )10Acetonitrile flux451264252015102500051015Pressure (Bar)20250510152025Pressure (Bar)Figure 5. (a) 2-Propanol and (b) Acetonitrile fluxes through the membranes showing pressure dependence.Figure5. a(a)good2-Propanoland relationwithEquation fluxes(5); allthroughsolventstheshowedsteadyconstantfluxes whichdependence.increased with increasing pressure. Also, each solvent exposed to NF270 generally yielded higherfluxes than NF90, BW30, and XLE, with a trend almost similar to the observation in pure waterThere was a good correlation with Equation (5); all solvents showed steady constant fluxespermeability measurements. On the contrary, BW30 with the lowest pure water permeability coefficientwhich increased with increasing pressure. Also, each solvent exposed to NF270 generally yieldedhigher fluxes than NF90, BW30, and XLE, with a trend almost similar to the observation in pure waterpermeability measurements. On the contrary, BW30 with the lowest pure water permeabilitycoefficient gave higher fluxes compared to XLE. This phenomenon indicates the variation inmembrane behaviour in the presence of a different solvent. Therefore, the resultant rejectionbehaviour of the membranes will similarly be perturbed.
Membranes 2020, 10, 1668 of 21gave higher fluxes compared to XLE. This phenomenon indicates the variation in membrane behaviourin the presence of a different solvent. Therefore, the resultant rejection behaviour of the membraneswill similarly be perturbed.The same phenomenon was observed with acetonitrile; strangely with NF90, the trend of theinitial solvent fluxes appeared to be similar to what was observed in the flux of 2-propanol. However,as the pressure increases, sudden changes in pore structure become evident in NF90. The solvent fluxessuddenly increase to become the highest of all the membranes; an indication of NF90 pore structurealteration due to the interaction with acetonitrile molecules. The size of the solvent molecule has aneffect on the morphology of the polymer at a molecular level [39], causing the polymer chains to eitherrelax or contract as the solvent molecules penetrate the matrix.Membranes 2020, 10, x FOR PEER REVIEW9 of 23Equation (5) highlights the influence of viscous flow on solvent transport in nanofiltrationmembranes, which is evident from the plot of solvent flux against viscosity (Figure 6).120NF 270NF 90BW 30XLE80-2-1Flux (I.m .h )10060402000.51.01.52.02.5oViscosity @ 25 C (cP)Figure 6. Graphs showing the relationship between flux and viscosity.Figure 6. Graphs showing the relationship between flux and viscosity.The results showed an increase in solvent flux with decreasing viscosity of solvents; hence, a solventTheviscosityresults showedincreasesolvent fluxwithdecreasingsolvents;hence, awith lowwill flowanthroughtheinmembranewithmoreease thanviscositya solventofwithhigh emembranewithmoreeasethanasolventwithThe resistance to flow will, therefore, lead to lower fluxes in the membranes, which collaborates hightheviscosity. The resistanceHagen–Poisseuillemodel. to flow will, therefore, lead to lower fluxes in the membranes, whichcollaboratesmodel.The effecttheof Hagen–Poisseuilleease of flow is in linewith the resistance-in-series model developed by Machado,Theeffectof
in Pd-catalyst rejection, a careful selection of the polymeric membrane and solvent, a satisfactory separation, and recovery can be achieved. Keywords: palladium-based catalysts; polymeric membranes; separation; NF/RO membranes 1. Introduction Separation technology has evolved during the 20th century, driven primarily by advances in the
Membranes 2017, 7, 9 4 of 13 The helium permeability in a list of polymeric membranes is provided in Table2. The Robeson's plot of He against N2 is provided in Figure1and this has also been well reported in the literature [20,21]. All reported membranes have selectivity for He over N2, in part because of the smaller kinetic diameter of He compared to N2, ensuring He has a higher diffusivity .
work with major steps. af s. 1 s 2 on s 3 . 4 5 C. Scheme 1. Development of functionalized nanostructured polymeric membranes for water purification of pharmaceuticals using target specific membranes. 2. Experimental section 2.1. Reagents and chemicals Metoprolol, diclofenac, vancomycin hydrochloride, ethanol, 60 mL SPE tubes and 20
CLAY NANOPARTICLES EFFECTS ON PERFORMANCE AND MORPHOLOGY OF POLY(VINYLIDENE FLUORIDE) MEMBRANES A . (PVDF-Nanoclay and PVDF-PVP-Nanoclay) membranes is presented. All membranes were synthesized by the phase inversion process, using 18% PVDF, n-methylpyrrolidone as solvent and water as the non-solvent. Demineralized water cross-flow permeation tests were conducted to evaluate the membranes .
A membrane is a thin sheet or layer of tissue that covers a structure or lines a cavity. A.Match these terms with the Mucous membranes Serous membranes correct statement or definition: Other membranes 1. Line cavities that open to the outside of the body. 2. Line the trunk cavities and cover the organs located within the trunk cavities. 3.
for the separation and filtration of challenging industrial effluent and wastewater. Unlike outside-in filtration, as seen with submerged membrane technology, our robust . tubular UF membranes enable operators to sustain higher flux rates with less frequent cleanings and reduced maintenance. Tubular UF membranes from Berghof
Digging into the biophysical features of cell membranes with lipid-DNA conjugates. Quarterly Reviews of Biophysics 55,e5,1-11. . 2020b). In these conjugates, a lipid moiety is used to anchor into the cell membranes via hydrophobic interactions. Whereas a DNA oligonucleotide is conjugated to provide desired functions on the cell membranes .
Characterization: Characterization is the process by which the writer reveals the personality of a character. The personality is revealed through direct and indirect characterization. Direct characterization is what the protagonist says and does and what the narrator implies. Indirect characterization is what other characters say about the
Am I my Brother’s Keeper? Acts 15:19-35 Introduction: Since the beginning of time when the first man and woman rebelled against God, mankind has been separated from God. Every person since that time has been born into that rebellion and sin. Because of sin, people are separated from God and are unable to have a right relationship with Him or each other. Ill. of evil and suffering Inside of .
