Medicine, Science & Technology

A Study of Polymerization-Induced Phase Separation As a Route to Produce Porous Polymer-Metal Materials

Description
A Study of Polymerization-Induced Phase Separation As a Route to Produce Porous Polymer-Metal Materials
Published
of 6
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.
Related Documents
Share
Transcript
  A Study of Polymerization-Induced PhaseSeparation as a Route to Produce PorousPolymer–Metal Materials Stanislav Dubinsky, Alla Petukhova, Ilya Gourevich, Eugenia Kumacheva* Introduction Porouspolymermaterialscoatedwithmetalnanoparticles(NPs) show many promising applications. Metal NPspossessanumberofusefulproperties,e.g.,surfaceplasmonresonanceorcatalyticactivity,whereasaporouspolymericmatrix provides a structural stability [1] and a large activesurface. [2] Once coated with metal NPs, porous polymerscanbeutilizedinbiosensing [3] andcatalysis, [4] orcanserveas templates for the synthesis of inorganic porousmaterials. [5] The requirements for these hybrid materialsinclude a homogeneous distribution and a high density of NPs on the surface of pores, the stability of NPs againstaggregation, and a strong NP attachment to the polymersurface. [1,5] The latter requirement is important in theapplications involving flow of liquids through hybridporous materials: weakly attached NPs can be washedaway from the polymer surface.Porous polymer materials carrying metal NPs on thesurface of pores have been produced by depositing pre-formed NPs onto the polymer surface, [3c,5b,6] or by in situsynthesis of metal NPs directly on the surface of thepolymerfollowingtheuptakeofmetalionsandsubsequention reduction. [3a,7] The first approach relies on electrostaticattraction or chemical affinity between the NPs and apolymersurface,anditdoesnotprovidestrongattachmentof NPs to the surface. [5b,6] This drawback limits the use of such materials in applications exploiting continuous flowthrough the porous material. In situ synthesis leads to abroad distribution in dimensions of the NPs and theiruneven distribution on the polymer surface. [3a] Polymerization-inducedphaseseparation(PIPS)isanewefficient approach to produce porous polymer materialscarrying inorganic NPs on the surface of pores. [8] Thestrategyutilizestwoeffectsoccurringconcurrently:PIPSinthepolymer–porogensolventmixtureandthemigrationof NPs to the interface between the polymer and the porogensolvent. During PIPS, the srcinally homogeneous mixture Communication E. Kumacheva, S. Dubinsky, A. Petukhova, I. GourevichDepartment of Chemistry, University of Toronto, 80 Saint GeorgeStreet, Toronto, Canada ON M5S 3H6Fax: ( þ 1) 416 978 3576; E-mail: ekumache@chem.utoronto.ca We report the results of the experimental study of the preparation of hybrid porous polymermaterial carrying gold nanorods (NRs) on the surface of pores. The material was prepared byutilizing twoeffectsoccurringconcurrently: thephotoinitiated polymerization-induced phaseseparation in the polymer–solvent mixture and themigration of the NRs to the interface between thepolymer and the porogen solvent. We show that theenrichment of the interface with the NRs is enhancedat high polymerization rate leading to the rapid phaseseparation. By contrast, more rapid increase inviscosity achieved at high polymerization rate doesnot have a significant effect on the segregation of NRsto the surface of pores. Macromol. Rapid Commun.  2010 ,  31  ,  1635–1640   2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim  wileyonlinelibrary.com  DOI: 10.1002/marc.201000210  1635  ofamonomer,across-linkingagent,aporogensolvent,anda photoinitiator separates into two phases: a porouspolymer and a porogen. [9] The segregation of NPs to theinterface between the two phases is governed by theminimization of the free energy of the system,  D  E  , as [10] : D  E  ¼  p r  2 g  P = PM   g  P = PM   g  NP = P  g  NP = PM  h i 2 (1)where  g  NP/PM ,  g  NP/P , and  g  P/PM  are the contributions to theinterfacial energy from the NP–porogen, NP–polymer, andporogen–polymer interfaces, respectively, and  r   is the NPradius.This single-step method can be used for producingmonoliths or micrometer-size particles and it has thefollowing useful features: (i) the majority of NPs segregateto the surface of pores and hence, the NPs are not ‘‘lost’’ inthe bulk of the polymer material; (ii) the NPs are stronglyattachedtothepolymersurface,and(iii)thismethodcanbeimplemented for the combinations of different polymersand NPs.A very important feature of the method is that the twoprocesses: PIPS and the migration of NPs to the polymer–liquid interface occur concurrently. Polymerization resultsin phase separation (favoring NP segregation) and inincrease in viscosity (counteracting NP mobility). Thus itiscan be expected that polymerization rate should influencenotonlythesizeofporesofthehybridmaterialbutalsotheenrichment of the surface of pores with NPs. The objectiveof the present work was to explore the effect of the rate of photoinitiated polymerization on the structure of thehybrid porous material. The study was performed for goldnanorods (NRs) end-tethered with thiolated polystyreneand a mixture of glycidyl methacrylate (GMA), ethyleneglycol dimethacrylate (EGDMA), and a porogen solventdiisodecyl phthalate (DDP). Experimental Part Materials Monomers GMA and EGDMA, a photoinitiator 2,2-dimethoxy-2-phenylacetophenone (DMPA), porogen solvent DDP, HAuCl 4 , cetyltrimethylammonium bromide (CTAB), sodium borohydride,AgNO 3 , ascorbic acid, tetrahydrofurane, methanol, and acetonewere purchased from Aldrich Canada and used as received. Thiol-terminated polystyrene (  M  w , ¼ 21500g/mol) was purchased fromPolymer Source, Inc. (Doval, Quebec). Synthesis of Porous Polymer–NR Monoliths Gold NRs stabilized with CTAB were synthesized according toproceduredescribedelsewhere. [11] CTABattheendsoftheNRswasreplacedwiththiol-terminatedpolystyrenemolecules(laterinthetext polystyrene modified nanorods are referred to as ‘‘NRs’’). [12] TheNRs(0.4g)weredissolvedinthemixturecontaining27vol.-% of the monomer GMA, 18 vol.-% of the cross-linking agentEGDMA,1wt.-%ofDMPA(basedonthetotalcontentofmonomers),and 55 vol.-% of the porogen DDP (Later in the text this mixture isreferred to as a ‘‘monomer mixture’’). The monomer mixture wasintroduced in the 5 cm-long polytetrafluoroethylene tube withthe inner diameter of 1mm and exposed to UV irradiation(65mW  cm  2 , Hohle UV Technology) for 15min. Following thepolymerization, the porogen liquid was washed away by purgingmethanol and acetone through the hybrid monolith and subse-quently drying the monolith overnight at 60 8 C. Characterization of the Porous Monoliths and theNanorods The surface structure of the porous polymer and the hybridmonoliths was examined by scanning electron microscopy (SEM)(HitachiS-5200)attheacceleratingvoltageof1kVusingsecondary(SE) and back-scattered electron(BSE) detectors.A piece of grindedmaterial was attached to the aluminum sample holder using agraphite conductive adhesive (EMS, USA). No sputtering was usedin these experiments. In order to image the cross-sectionalstructure of the material, the monolith was filled with instantglue (Krazy, Elmer’s Products Canada Corporation), dried, andmicrotomedat room temperature into slices with the thicknessof 20nm. The slices were collected on carbon-covered copper gridsand characterized at the accelerating voltage of 100kV using thetransmitting electron microscope (TEM) (Hitachi H-7000). Themean size of the polymer globules in the porous material wasdeterminedbyanalyzingSEMimagesofca.150globulesusingtheImage Tool (UTHSCSA) Software. The specific surface area of theporousmonolithwasdeterminedbymeasuringtheadsorptionanddesorption isotherms of nitrogen on a Quantachrome AS1C-VP2apparatus with a bath temperature of 77K.AVarianCary5000UV–Vis–NEAR IR spectrometerwasusedtoacquire the transmission spectra of the NRs, the porous polymer,and the hybrid monolith material. The measurements wereconducted in dimethyl sulfoxide (DMSO, refractive index of 1.479 [13] ), in order to reduce the scattering by the porous polymermatrix.The viscosities of the polymerizing mixtures were measuredusing a Brookfield rheometer (Brookfield, USA) at 150rpm underexposure to UV irradiation (65mW  cm  2 , Hohle UV Technology). Results and Discussion Figure 1 illustrates the approach to produce a porouspolymer material carrying NPs on the surface of pores.Ligand-coatedgoldNRsaremixedwithamonomermixtureincludingamonomer,across-linkingagent,aphotoiniator,and a porogen solvent. The solubility parameters of themonomerandofthesolventarecloseandthesystemformsastablesolution.Inaddition,thesolubilityparameterofthemonomer mixture and the solubility parameter of the S. Dubinsky, A. Petukhova, I. Gourevich, E. Kumacheva 1636 Macromol. Rapid Commun.  2010  ,  31 , 1635–1640   2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim  DOI: 10.1002/marc.201000210  ligands coating the NR surface are also sufficiently close,which renders NR stability in the monomer mixture.Following photopolymerization, the system phase sepa-rates into a polymer phase and a liquid porogen phase,owing to the significant difference in the solubilityparameters between the polymer and the solvent. SincethesolubilityparameteroftheligandsontheNRsurfaceisdistinctfromthesolubilityparametersofboththepolymerandthesolvent,duringPIPStheNRsmigratetotheinterfacebetween the polymer and the porogen. The removal of theporogen from the system leaves behind a porous polymermaterial carrying NRs on the surface of pores.Figure 2a shows a typical SEM image of the structure of the hybrid polymer produced by PIPS. In the porousmaterial, the polymer globules with the average diameterof ca. 500nm are coated with gold NRs. The TEM image of the cross-section of the globule shows that the interior of the globules is deprived of the NRs (Figure 2b). The specificsurface area of the hybrid material was 3.4m 2  g  1 .FollowingPIPS,theNRsretainedtheiropticalproperties.Figure 3 shows the absorption spectra of the hybridpolymer, along with the spectra of individual gold NRs,andofthemonolithpreparedintheabsenceoftheNRs.Thespectrum of the hybrid monolith featured two absorptionpeaks at 506 and 776nm, characteristic for the transverseand longitudinal plasmon bands of gold NRs (centered at518and787nm),whereasthespectrumofthegoldNR-freemonolith showed no significant absorption in the rangefrom 400 to 1100nm. A small shift in the spectral positionof adsorption peaks of the NRs in the hybrid material wascaused by the difference in the dielectric constant of themedium surrounding the NRs. [14] Next, we examined the effect of polymerization rate onthe structure of hybrid material. We used the followingrelation. [15]  R p  /  fa ½  A   I  0 10 3 e  a ½  A   D   1 = 2 (2)where  R p  is the polymerization rate in the layer located atthe distance  D  from the surface of the polymerizationmixture,  f  the quantum yield of the initiation process,  a the absorptivity of the photoinitiator, [  A ] the concentra-tion of the photoinitiator and  I  0  is the intensity of theincident UV irradiation. In our work, we examined thestructureofthemonolithwiththethicknessofca.200 m m,whereas the concentration and the absorptivity of thephotoinitiator were 1wt.-%, and 73.6 L  mol  1  cm  1 . [16] When these values were used, Equation (2) was reduced toEquation (3) as:  R p  /  fa ½  A   I  0 10 3   1 = 2 (3)Equation (3) explicitely shows that the rate of polymeriza-tion can be tuned by changing the concentration of initiator and the intensity of incident UV irradiation as  R p   I 0 ð Þ 1 = 2 and  R p  [  A ] 1/2 , respectively.First, we examined the effect of the concentration of photoinitiatorDMPAonmaterialstructure.Figure2canddshows the SEM and the TEM images of the surface and thecross-section,respectively,ofthehybridmonoliththatwassynthesized at a tenfold reduced concentration of thephotoinitiator, in comparison with the monolith shown inFigure 2a and b, that is, at polymerization rate reduced bya factor of    3.2. With reduced polymerization rate, thenumberofNRsonthesurfaceofporeswasnotablyreduced(Figure 2c vs. 2a), and the number of NRs trapped in theinterior of polymer globules increased (Figure 2d vs. 2b).In the second series of experiments, by reducingtheintensityoftheincidentlight,  I  0 ,from65to5mW  cm  2 we decreased polymerization rate by a factor of    3.6. Thesurface coverage of pores with the NRs reduced andthe density of NRs in the polymer globules increased(Figure 2e and f), consistent with the results obtained atreduced concentration of the photoinitiator. Thus weconclude that the segregation of gold NRs to the surface of pores was suppressed at a reduced polymerization rate.To understand the role of polymerization in theenrichment of the surface of pores with the NRs, weexamined the relative rates of polymerization-drivenphase separation and increase in viscosity of the system.The time before thebeginning of PIPS wasdetermined by A Study of Polymerization-Induced Phase Separation as a . . . Figure 1. Schematicrepresentationoftheformationoftheporouspolymer monolith carrying gold NRs on the surface of pores:(a) photopolymerization of the monomer in the mixture contain-ing a monomer, a cross-linking agent, a porogen solvent, andgold NRs; (b) PIPS and segregation of the NRs to the liquid–solid interface; (c) a hybrid microporous polymer materialobtained after removal of the porogen solvent from thephase-separated system. Macromol. Rapid Commun.  2010 ,  31  ,  1635–1640   2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.mrc-journal.de  1637  measuring the cloud point of the monomer mixture as afunction of polymerization time. [17] The extinction wasmeasured at 640nm, in order to minimize the contribu-tionfromlightabsorptionbythephotoinitiatorat400nmand by gold NRs at 518 and 787nm. Figure 4a shows thevariation in extinction of the polymerizing monomermixtures with the high and low (1.0 and 0.1wt.-%)concentrations of the photoinitiator. In the course of polymerization, the extinction increased, mostly dueto increased light scattering between the solid polymerphase and the liquid porogen–monomer mixture.Taking the cloud point as the time at which extinctionlevels off, we conclude that demixing in the monomermixture containing 1wt.-% of photoinitiator (fast poly-merization) occurred after ca. 27s. In contrast, a gradualincrease in extinction coefficient in the slowly polymer-izing system ([DMPA] ¼ 0.1wt.-%) sug-gested that PIPS occurred slowly andeven within 60s it was not complete.Polymerization induced increase inviscosity of the monomer mixture con-taining 1.0 and 0.1wt.-% of the photo-initiator is shown in Figure 4b. We notethat this graph presents a trend ratherthantheexactvaluesofthevicosityofthesystem, since the measurements wereconductedintheabsenceoftheporogen,in order to avoid slip at the polymer–porogen interface. A dramatic increasein viscosity (the transition to the gelpoint) [18] was reached in approximately22 and 28s after the beginning of polymerization when the concentrationof photoinitiator was 1.0 and 0.1wt.-%,respectively.The dimensions of the polymer glo-bules in the porous polymer were alsoinfluencedbytherateofpolymerization.Themeansizeoftheglobulesobtainedathigh and low polymerization rates was0.8  0.1and1.02  0.16 m m,respectively(Figure4candd).Asmallerglobulesizeinthe rapidly polymerizing system wascaused by the greater number of free-radicals generated per unit time. [9b] BasedontheresultsshowninFigure4,we ascribe enhanced segregation of NRsto the porogen–polymer interface at thehigher polymerization rate as follows.Based on the morphology of the porouspolymer,weconcludethattheformationof the porous material in the systemstudied occurs by the  x -syneresismechanism. [19] This mechanism is char-acteristic for the systems in which a large difference existsbetween the solubility parameters of the polymer and theporogen. In the course of polymerization, before the gelpointisreached,thepolymerprecipitatesfromtheporogenand forms globules, which ultimately form a continuousporous network. [9b,19] The localization of NRs in thepolymer, in the porogen, or at the liquid–polymerinterface is determined by (i) the affinity of NRs to eachphase (determined by the difference in the correspondingsolubilityparameters)and(ii)theabilityofNRstodiffusetothe polymer–porogen interface (determined by viscositybuilt-up and the size of the polymer globules).In our work, prior to polymerization, the solubilityparametersoftheNRs(assumedtobeequaltothesolubilityparameter of the polystyrene ligand) and the monomermixture were 18.2 [20] and 16.3MPa 1/2 . The solubility of the S. Dubinsky, A. Petukhova, I. Gourevich, E. Kumacheva Figure 2. SEMimagesofthesurface(a,c,ande)andtheTEMimagesofthecross-section(b, d, and f) of the porous hybrid monoliths, synthesized at: (a and b) 1.0wt.-% of DMPAat  I 0 ¼ 65 mW  cm  2 ; (c and d) 0.1wt.-% of DMPA at  I 0 ¼ 65 mW  cm  2 ; (e and f) 1.0wt.-% of DMPA at  I 0 ¼ 5 mW  cm  2 . The scale bars are 500nm. All monomers werepolymerizedunderthesameconditions.IntheTEMimages,thebrighterareasrepresentpolymer globules and the darker regions correspond to the acrylic resin introduced inthe pores prior to microtoming. 1638 Macromol. Rapid Commun.  2010  ,  31 , 1635–1640   2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim  DOI: 10.1002/marc.201000210  monomer mixture was determined as  d mix ¼ w DDP  d DDP þ w GMA d GMA þ w EGDMA d EGDMA , [20] where d DDP ¼ 14.7MPa 1/2 , [20] d GMA ¼ 18.3MPa 1/2 , [21] and  d EGDMA ¼  18.3MPa 1/2[22] are thesolubility parameters of DDP, GMA, and EGDMA, respec-tively,and w DDP , w GMA ,and w EGDMA arethevolumefractionsof DDP,GMA,andEGDMA,respectively.Afterphaseseparation,thepolymerphaseandtheporogenphasehadthesolubilityparameters of 14.7 [20] and 24MPa 1/2[2b] , respectively. Thelocalization of the NRs in either phase was energeticallyunfavorable and they segregated to the interface betweenthe polymer and the porogen. Rapid phase separationcharacteristic for higher polymerization rate (Figure 4a)favored the segregation of the NRs to the polymer–porogeninterface.Bycontrast,atslowpolymerization,phasesepara-tionoccurredatalow rateand theemergingpolymer phasewas swollen with the porogen. Since the mean solubilityparameter of this system was close to that of the NRs, thesegregationofNRswasnotfavoredasmuchasintherapidlyphase-separatingsystem.Asmallersizeofpolymerglobulesformed at higher polymerization rate also favored thediffusionofNRs to theinterface dueto theshorter diffusionpath for the NRs moving from the polymer phase. A Study of Polymerization-Induced Phase Separation as a . . . Figure 3.  Absorption spectra acquired for (a) the porouspoly(GMA–EGDMA) monolith; (b) the solution of NRs inDMSO; (c) the porous hybrid poly(GMA–EGDMA) materialcontaining NRs. Figure 4.  Variation in (a) extinction of the polymerizing monomer mixtures containing 0.1wt.-% ( & ) and 1.0wt.-% ( ~ ) of DMPA as afunction of polymerization time. l ¼ 640nm; (b) viscosity of the monomer mixture containing 0.1wt.-% ( & ) and 1.0wt.-% ( ~ ) of thephotoinitiator DMPA, both plotted as a function of polymerization time. SEM images of poly(GMA–EGDMA) synthesized in the presence of (c) 1.0wt.-%, and (d) 0.1wt.-% of DMPA. The scale bars are 5 m m. Macromol. Rapid Commun.  2010 ,  31  ,  1635–1640   2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.mrc-journal.de  1639
Search
Similar documents
View more...
Tags
Related Search
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