Graphics & Design

Production of porous suspension polymer beads with a narrow size distribution using a cross-flow membrane and a continuous tubular reactor

Description
Production of porous suspension polymer beads with a narrow size distribution using a cross-flow membrane and a continuous tubular reactor
Published
of 9
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
  Colloids and SurfacesA: Physicochemical and Engineering Aspects 180 (2001) 301–309 Production of porous suspension polymer beads with anarrow size distribution using a cross-flow membrane and acontinuous tubular reactor Peter J. Dowding, James W. Goodwin, Brian Vincent * School of Chemistry ,  Uni   ersity of Bristol  ,  Cantock  ’  s Close ,  Bristol BS  8 1 TS  ,  UK  Received 30 August 2000; accepted 29 September 2000 Abstract The work described focuses on a two-stage process for the production of large porous suspension polymer beadsof defined particle size and narrow size distribution. Emulsification has been performed using purpose built cross-flowmembrane equipment, which allows controlled production of large emulsion droplets with a much narrower sizedistribution. The work described compares the production of large emulsion droplets of monomer prepared both byagitation and using a cross-flow membrane. The effects of variations in the pore size of the membrane and flow-rateson the size of the emulsion droplets produced are also investigated. The second stage of the process is polymerisationof preformed monomer emulsion droplets using a continuous tubular reactor. Samples polymerised using such amethod show a narrower size distribution than similar systems polymerised as a batch. © 2001 Elsevier Science B.V.All rights reserved. Keywords :   Cross-flow membrane; Emulsification; Suspension polymerisation; Continuous reactor; Porous beadswww.elsevier.nl / locate / colsurfa 1. Introduction The aim of the work described in this paper isto produce suspension polymer beads with a con-trolled particle size and a narrow size distribution.For suspension polymerisation, the size distribu-tion of the beads produced is controlled by thepolydispersity of the initial monomer emulsionand broadening effects, which occur during thepolymerisation reaction.Large emulsion droplets (  500   m in diame-ter) are generally produced by shearing the com-ponent phases with a suitable agitator. However,such methods lead to poor control over the aver-age droplet size produced, and often result in alarge polydispersity. Increasing interest is nowbeing devoted to the production of droplets of uniformed size and narrow size distribution. Theuse of more ‘non-standard’ emulsification devicesis now receiving growing attention. These tech-niques generally involve a forced passage througha constriction e.g. rotating disks [1–3] cross-flow * Corresponding author. Tel.:  + 44-117-9288160; fax:  + 44-117-9250612. E  - mail address :   brian.vincent@bris.ac.uk (B. Vincent).0927-7757 / 01 / $ - see front matter © 2001 Elsevier Science B.V. All rights reserved.PII: S0927-7757(00)00777-9  P . J  .  Dowding et al  .  /   Colloids and Surfaces A :   Physicochem .  Eng  .  Aspects  180 (2001) 301–309  302 membranes [4] or by passing the monomerthrough a porous glass membrane [5–7].The principle of expelling the oil phase througha small diameter ofifice in a cross-flow field of water containing stabiliser has been previouslyreported by Peng and Williams [8]. Peng andWilliams extended their work and produced aprototype cross-flow membrane rig for the pro-duction of emulsion droplets (with droplet sizestypically in the range  0.1–5   m) through multi-ple orifices using a cylindrical membrane geome-try [4]. Later work addressed some of the factorsinvolved in the scale-up of such an apparatus topilot plant scale. In a cross-flow membrane ap-paratus, a discontinuous (dispersed) phase ispumped through a porous substrate, into a con-tinuous phase on the other side of the membrane.Droplets are formed at the membrane / continuousphase interface. The flow of continuous phase isperpendicular to that of the discontinuous phase;this leads to detachment of droplets from themembrane. Cross-flow membrane can be used toproduce either oil-in-water or water-in-oil emul-sions, of controlled droplet size, and with a nar-row size distribution.For a system of fixed chemical composition,variations in the cross-flow velocity (of the contin-uous phase) and the flow rate of the discontinu-ous phase (through variation in trans-membranepressure) can result in a change the droplet size.Variations in the chemical composition of theemulsion can affect droplet size through changesin viscosity (of both phases) and also the oil / watersurface tension. Other factors can be used to varyemulsion droplet size produced including theporosity of the membrane and temperature. Vari-ous types of membrane have been used includingcoated ceramics, polymers and porous glass, andalso various designs of membrane shape includingtubes (in which the continuous phase flowsthrough the middle) and flat plates or discs. In thework described here, we have used a purpose-builtcross-flow membrane equipment to investigate theproduction of emulsion droplets which are muchlarger than any previously produced using such amethod (  100–800   m in diameter). The pro-duction of uniformed emulsion droplets is notstraightforward, and particularly difficult for largedroplets (having a diameter of 10   m or greater).The size distribution is dependent upon the bal-ance between droplet break-up and droplet coa-lescence. This is in turn controlled by the natureof the emulsification procedure used, the volumefraction of the monomer phase, and the type andconcentration of stabiliser used. The rate of droplet coalescence is controlled by liquiddrainage between approaching droplets, and moresignificantly by the rigidity of the two correspond-ing oil / water interfaces, since this controls thedamping of thermally or mechanically inducedoscillations in the film thickness. Adsorbed poly-mers help to increase the interfacial rigidity. Inprevious work, we have demonstrated that theaddition of relatively small amounts of polystyrene to the oil (styrene monomer) phaseresults in an increase in the viscosity of the dis-persed droplet phase [9]. This imparts increasedstability to coalescence of the droplets by againincreasing the rigidity of the droplet / waterinterface.During a suspension polymerisation reaction,as the polymerisation proceeds, there will come astage in the reaction where the partially poly-merised beads become ‘sticky’. Satellite droplets(formed by droplet break-up) may attach to thesurface of the polymer beads [10–13]. In preciouswork, we have shown that the rate of dropletbreak-up and coalescence are significantly reducedif the polymerisation is performed using a PTFEcontinuous reactor, rather than a stirred batchreactor [9]. 2. Experimental 2  . 1 .  Materials The following chemicals were obtained fromcommercial sources, styrene divinyl benzene (55%)( meta  and  para  isomers) benzoyl peroxide (BPO)(70%) 4-methyl-2-pentanol, polystyrene (averageMW = 280 000) and sodium chloride (Aldrich);glycidyl methacrylate (GMA, Fluka); poly (vinylalcohol) 88% hydrolysed (‘Mowiol 40–88’;Hoechst AG, PVA). The monomers were purifiedeither by distillation (styrene) or by filtration  P . J  .  Dowding et al  .  /   Colloids and Surfaces A :   Physicochem .  Eng  .  Aspects  180 (2001) 301–309   303 through an aluminium oxide column (glycidylmethacrylate and divinyl benzene) to remove in-hibitor. Deionised water (purite) was used formaking the aqueous phase. All purities were   97% unless otherwise stated.The component phases of the emulsion wereprepared separately. For emulsions prepared us-ing the cross-flow membrane, the continuousphase comprised 1908 g aqueous solution of PVA(0.11% w / w) and sodium chloride (5.66% w / w).The dispersed organic phase comprised 150 g of styrene (monomer) styrene, 150 g of divinyl ben-zene (crosslinking monomer), 240 g of 4-methyl-2-pentanol (porogenic diluent) and 5.0 g of polystyrene (0.93% w / w). For batch emulsifica-tions, the above amounts were divided 10-fold. 2  . 2  .  Cross -  flow membrane equipment Purpose-built cross-flow membrane equipmentwas purchased from Disperse Technologies Ltd.(Dorking, UK) and built by Memtech Ltd.(Swansea, UK). The experimental set-up adoptedis shown schematically in Fig. 1. The membraneused consisted of a stainless steel plate (0.45 mmthickness) with laser-drilled holes (100 or 150   mdiameter, ActionLaser, Sydney, Australia). Suchholes are arranged in a cubic array (shown in Fig.2), with an average distance of 323   m betweenthe holes. The active area of the membrane was20 × 20 mm.The continuous phase was circulated using aGrundfos Chi-2-20 centrifugal multistage pump(Applied Technologies Inc., Long Island, USA)fitted with an electromagnetic flow indicator. Thediscontinuous (monomer containing) oil-phasewas pumped across the membrane using aPharmicia P500 syringe pump (Pharmacia Bio-tech, Uppsala, Sweden). The pipe-work used inthe construction of the cross-flow membrane rigwas stainless steel (with an internal diameter of 19.05 mm, and a wall thickness of 1.63 mm).Silicone tubing was used for connections betweenthe membrane rig and the holding tanks. Fig. 1. Schematic representation of the cross-flow membrane emulsification rig.  P . J  .  Dowding et al  .  /   Colloids and Surfaces A :   Physicochem .  Eng  .  Aspects  180 (2001) 301–309  304Fig. 2. Optical micrograph of the stainless steel membraneused in emulsification showing the cubic arrangement of pores(laser-drilled holes) with a diameter of 150   m. 2  . 5  .  Characterisation of the emulsion droplets In situ droplet size analyses were carried outusing a ‘Lasentec’ M200L instrument. The mea-surements were made throughout emulsification.The probe tip was placed in the reaction flask (orcontinuous phase holding tank in the membranework), to the side of the agitator, approximately 1cm above the paddle.The Lasentec instrument uses the focused beamreflectance measurement technique (FBRM), inwhich particles or droplets passing the sapphirewindow of the probe reflect a laser beam. Thetime over which reflectance occurs shows a directdependence upon particle size. The manner inwhich the laser beam strikes the particles ordroplets will vary depending on which part of theparticle the beam strikes (since the wavelength of the laser used is much smaller than the particlesize) and a ‘chord-length’ of the particle size ismeasured.In the production of emulsion droplets using across-flow membrane, it should not be possible toproduce droplets with a smaller diameter than themembrane pore size. In the membrane equipmentused in this work, as already referred to, re-circu-lation of the continuous phase (containing emul-sion droplets) was required in order to increasethe dispersed phase volume fraction. This re-cir-culation can result in droplet break-up, whichmanifests itself as a reduction in the observedaverage droplet size. Volume weighted averagechord lengths were determined using the Lasentecinstrument over the entire instrument range (0.8– 1000   m). 2  . 6  .  Suspension polymerisation reactions The monomer emulsion systems prepared inthis work were used to produce solid beads bysuspension polymerisation. The polymer beadsmay be made porous by the inclusion of an inertdiluent (or ‘porogen’) in the monomer phase,which may be extracted after polymerisation [14].Suspension polymerisation reactions are generallyperformed as a stirred batch. However, in previ-ous work we have described a continuous tubularreactor (made from PTFE) for suspension poly- 2  . 3  .  Emulsification using cross -  flow membrane The effects on emulsion size of variations in themembrane pore size, cross-flow velocity and themonomer containing dispersed (oil) phase flowrate have been studied. With the design of mem-brane used, the active area is relatively small. Thisnecessitated re-circulation of the continuousphase / emulsion across the membrane severaltimes. Emulsification was typically performed for  45 min, in order to obtain an emulsion volumefraction of    10% (v / v). 2  . 4  .  Batch emulsification As a comparison, emulsions of the same com-position were prepared in a batch process, byshearing the component (organic and aqueous)phases for 1 h, at 600 rpm, using a toothed-discstirrer. These batch processes (emulsification andsuspension polymerisation reactions) were per-formed in a glass 500 ml cylindrical round bot-tom. The stirrer blade was kept at a constantdistance of 1 cm from the bottom of the flask.  P . J  .  Dowding et al  .  /   Colloids and Surfaces A :   Physicochem .  Eng  .  Aspects  180 (2001) 301–309   305 merisation reactions [9]. Polymer beads can beproduced using the continuous reactor with a sizedistribution much narrower than if polymerised asa batch reaction. This may be attributed to thefragile emulsion droplets experiencing a muchlower and more uniform shear field in the slowlaminar flow of the tubular reactor, compared witha stirred batch.Samples of emulsion produced both by shear andusing the cross-flow membrane have been poly-merised using the continuous reactor. For compari-son, a sample of beads were prepared in a batch,with emulsification performed by shear and poly-merisation as a stirred batch using the same 500 mlcylindrical vessel used for emulsification. For poly-merisation, the agitator speed was reduced to 400rpm. Polymerisation was allowed to proceed tocompletion overnight. 2  . 7  .  Polymerisation using a continuous tubularreactor A sample of emulsion (250 ml), produced eitherwith the cross-flow membrane apparatus or as abatch,wasplacedintoagentlystirredholdingtank,from which it was introduced into the continuousreactor. For emulsions produced using the cross-flow membrane, a 100   m membrane was used,with a monomer flow rate of 300 ml h − 1 and across-flow velocity of 100 l h − 1 .Previous results have shown that polymer beadscan be produced continuously using the PTFEcontinuous reactor, without blocking of the reactoroccurring, if certain modifications to the system aremade. The addition of relatively small amounts of polystyrene to the monomer phase results in anincrease in viscosity. This has a stabilising effectupon the droplets by reducing thin film fluctuationsbetweencontactingdroplets.Itcanalsohelpreducedroplet break-up during polymerisation.Secondly, by the use of bromine-containing spe-cies (in the form of a brominated porogenic dilu-ent), it is possible to match the densities of thedispersed and continuous phases. This reduces thepropensity for droplet creaming to occur.The continuous reactor consisted of a coiledPTFE tube (Teflon ® ) of internal diameter of 7.9mm and a wall thickness of 0.8 mm. Prior to theintroduction of emulsion, the tube was flushedthrough with continuous phase. The emulsion wasfedthroughthereactorataflowrateof180mlh − 1 ,using a Fluid Metering Inc. QG6 pump locatedclose to the tube exit. This type of pump operateson a valve-less positive displacement basis, whichprovides a relatively smooth flow, without subject-ing the solution in the pump head to large stresses.Polymerisation was performed at 70°C by plac-ing the continuous reactor in a thermostated waterbath. The solution was fed through the reactorusing the pump. The allowed residence time usedwasdependentuponthemonomersystemused[15],but was sufficient to allow polymerisation to pro-ceed to completion, plus an additional hour. Res-idence times were varied by the use of differinglengths of PTFE tubing, which in general wasgreater than 20 m in length.The resulting polymer beads from both thecontinuous and batch reactors were washed, se-quentially, with water, aqueous sodium hydroxidesolution (1.0 mol dm − 3 ), ethanol and acetone. Thisprocedure removes the porogenic diluent and anyunreacted monomer. The beads were then driedunder vacuum. 2  . 8  .  Characterisation of the copolymer beads The size distribution of each set of beads wasdetermined using a ‘Malvern’ Mastersizer MS1000.The beads were then fractionated using a 300   mstainless steel sieve to yield two fractions. The poresize distribution, pore volume and surface area of the larger size fraction were investigated by mer-cury porosimetry, using a ‘QuantaChrome’ Au-toscan 60 Porosimeter, operating in the range0–380 MPa. 3. Results and discussion 3  . 1 .  Effect of membrane pore size on emulsiondroplet size The effect of cross-flow velocity on the averagevolume weighted chord length (for membranes
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
SAVE OUR EARTH

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!

x