Pharmaceutical Research,Vol.14.No.5,1997 Report Macromolecules as Novel phenomenon occurs in the lipid bilayers of nonliving systems, such as liposomes and red blood cell ghosts,as well as the Transdermal Transport Enhancers plasma membranes of living cells,either isolated or part of a for Skin Electroporation tissue (5).Most recently,electroporation of the multilamellar, intercellular lipid bilayers of stratum corneum has also been demonstrated(3-4).During an electric pulse which causes elec- Rita Vanbever,1 Mark R.Prausnitz,2 and troporation,membrane permeability temporarily increases by Veronique Preat13 orders of magnitude,thereby increasing transmembrane trans- port of compounds as large as macromolecules(5-7).After the pulse,membranes remain permeable for milliseconds to hours, Received January 7,1997;accepted February 19,1997 depending on conditions such as temperature,but can eventually reseal to their original structure.This transient elevation of Purpose.Macromolecules were investigated as chemical enhancers of transdermal transport by skin electroporation.Although unable to membrane permeability has found widespread use as a means enhance passive or iontophoretic transport,macromolecules are pro for introducing DNA into viable cells (8)and,more recently posed to enhance electroporation-assisted delivery by stabilizing the as a clinical tool for targeting and enhancing uptake of chemo- increased permeability caused by high-voltage pulses. therapeutic agents by tumors (9-10). Methods.To test this hypothesis,we examined the timescale of trans Electroporation of skin can increase transdermal transport port,the influence of electrical protocol and the influence of macromol- for compounds ranging in size from small ions (e.g.,Na',CI-; ecule size,structure,and charge on enhancement of transdermal ref.11)to moderate-sized molecules (e.g.calcein,metoprolol; mannitol transport in vitro by heparin,dextran-sulfate,neutral dextran, refs.3,12),to macromolecules (e.g.LHRH,heparin,oligonu- and poly-lysine. Results.Skin electroporation increased transdermal mannitol delivery cleotides;refs.13-15),to latex microspheres(16).The effects by approximately two orders of magnitude.The addition of macromole- of skin electroporation are largely or completely reversible, cules further increased transport up to five-fold,in support of the where the skin's barrier properties are restored within approxi- proposed hypothesis.Macromolecules present during pulsing enhanced mately one hour.Unlike iontophoresis,which acts primarily by mannitol transport after pulsing for hours,apparently by a macromole- electrophoretically driving compounds through existing path- cule-skin interaction.No enhancement was observed during passive ways in skin (1-2),electroporation transiently creates new path- diffusion or low-voltage iontophoresis,suggesting that macromolecules ways for transport (17).Although most work has been interact specifically with transport pathways created at high voltage. performed in vitro,animal studies in vivo show similar Although all macromolecules studied enhanced transport,those with results (3.18). greater charge and size were more effective. Conclusions.This study demonstrates that macromolecules can be Chemical enhancers which increase transdermal transport used as trandermal transport enhancers uniquely suited to skin by passive diffusion and those which increase transport by electroporation. skin electroporation are not necessarily the same.Traditional KEY WORDS:macromolecule;electroporation;iontophoresis;skin; chemical enhancers of passive transdermal delivery (e.g. permeation enhancer;transdermal transport. DMSO,Azone)act largely by disrupting stratum corneum lipid structure (1-2).In contrast,an effective chemical enhancer for INTRODUCTION electroporation-assisted delivery does not need to disrupt lipids Drug delivery across skin offers a noninvasive,user- but should stabilize the transient disruptions already created by friendly alternative to conventional oral or parenteral adminis- skin electroporation. tration (1-2).However,the skin's outer layer,the stratum cor- We hypothesized that macromolecules,although not neum,is an extremely effective barrier which prevents transport expected to enhance passive transport,could enhance electro- of most drugs at therapeutic rates.Chemical,electrical,mechan- poration-assisted delivery by stabilizing the increased perme- ical,and ultrasonic approaches to increasing transdermal trans- ability caused by high-voltage pulses.A study of cell port have been explored.Recently,the intermittent application electroporation showed that DNA could stabilize the elevated of short (e.g.msec).high-voltage (e.g.,100 V)pulses has also membrane permeability of electroporated simian Cos-1 cells, been shown to increase transport across skin by many orders possibly by being inserted into "electropores"and hinder their of magnitude,probably by a mechanism involving electropora- closure (19).In skin,an initial study concerning transdermal tion(3-4). delivery of heparin by skin electroporation indirectly suggested Electroporation (or electropermeabilization)involves the that heparin molecules could prolong increased skin permeabil- creation of transient aqueous pathways across lipid bilayer ity (14).A subsequent study supported this conclusion with membranes by applying a short,high-voltage pulse(5-7).This direct evidence of enhanced and prolonged skin permeability when heparin was present during skin electroporation (20). In this study,we investigated the possibility that macro- molecules could enhance transdermal transport by skin electro- Unite de Pharmacie galenique,Ecole de Pharmacie.Universite cathol- poration.We first confirmed the enhancing effect of heparin, ique de Louvain,Brussels,Belgium. 2 School of Chemical Engineering.Georgia Institute of Technology, and then explored the effects of dextran-sulfate,neutral dextran, Atlanta,Georgia 30332. and poly-lysine on transdermal transport of mannitol during in 3 To whom correspondence should be addressed.(e-mail:preat@ vitro skin electroporation.A broad range of electrical protocols farg.ucl.ac.be) were employed to help elucidate the mechanisms by which 0724-8741/97/0500-0638S12.500 1997 Plenum Publishing Corporation 638
Macromolecules as Enhancers for Skin Electroporation 639 macromolecules could enhance electroporation-mediated trans- Measurement of transdermal mannitol transport was car- port across skin. ried out by removing samples(0.4 ml)from the receiver com- partment at regular intervals.Samples were replaced with an MATERIALS AND METHODS equal volume of buffer B.The amount of 4C-mannitol in each sample was determined by scintillation counting (liquid The experimental approach has been described previously scintillation cocktail Ready Safe,Beckman,Belgium;liquid (12,21)and is summarized and expanded upon below. scintillation counter Wallac 1410,LKB,Pharmacia,Uppsala, Sweden). Chemicals The stability of mannitol and its 4C radiolabel during D-mannitol (crystalline),dextran (average molecular high-voltage pulsing was verified by thin layer chromatography weight,MWave=10 kDa),dextran-sulfate (MWave=5 and 10 using silica gel 60 sheets (Merck,Darmstadt,Germany).The kDa),heparin (MWave =20 kDa;sodium salt from porcine mobile phase contained 85%n-butanol and 15%water by vol- intestinal mucosa),poly-L-lysine(MW =4-15 kDa;hydro- ume.Chemical detection of mannitol was performed with iodine bromide salt)and NaCl were obtained from Sigma Chemical vapor (UCB,Drogenbos,Belgium)(22).Radioactivity was Co.(St.Louis,MO).Heparin(MWv=12 kDa;sodium salt detected by scintillation counting (Bioscan Type AO 3000 A, from porcine intestinal mucosa)was kindly donated by Phar- System 200 Auto Changer and Imaging Scanner,Canberra- macia Hepar Inc.(Franklin,OH).D-1-14C-mannitol was Packard,Washington,DC).Both mannitol and 4C-labeled man- obtained from New England Nuclear Research (Du Pont de nitol were shown to remain intact under the conditions of Nemours,Brussels).NaH2PO4 and Na2HPO4,and n-butanol this study. were supplied by Union Chimique Belge (UCB,Drogenbos Belgium).Buffer A contained 0.06 M sodium phosphate at pH 7.4.Buffer B contained 0.048 M sodium phosphate at pH Electroporation and lontophoresis 7.4 supplemented with 0.15 M glucose (Merck,Darmstadt, Germany)to provide physiological osmolarity.All solutions The electroporation device used was an Easyject Plus were prepared in ultrapure water (Sation 900,Vel,Leuven, (Equibio Ltd.,Kent,England),which delivers exponential- Belgium). decay capacitive discharge pulses.The pulse time constant (Tpulse)is defined as the length of time between the beginning Transdermal Transport Measurements of the pulse(maximum voltage)and the time when the voltage Skin electroporation was performed in vitro using upright reaches 37%of its initial value.Voltages are expressed as diffusion chambers.Freshly excised abdominal hairless rat skin voltages applied across the electrodes (Uler).However,a (mutant rat Iops hairless from Iffa Credo,Saint Germain les pulse,apparent skin resistance can drop well below 100 (23). Arbreles,France)separated the donor (or upper)and receiver As a result,significant voltage drops occur within donor and (or lower)compartments with an exposed skin area of 3 cm2 receiver solutions making the voltage across the skin (Uskin) The donor reservoir was filled with 1.5 ml of donor solution: approximately three-fold lower than the total voltage applied, mannitol (1 mg/ml)and 4C-mannitol (0.5 uCi/ml)in buffer Ueleetrodes (23).For example,pulses applied with Uelectrodes A either with or without an"enhancer"compound.The receiver 150 V and Tpulse 180 ms yielded Uskin 40 V,whether compartment(7.5 ml)was filled with buffer B,continuously 10 kDa dextran-sulfate was present or not.The total charge stirred with a magnetic stir bar and maintained at 37C.Platinum transfered across the skin by a pulsing protocol was determined electrodes were immersed in the donor and receiver solutions by the product of UlecrodeTpulse and the number of pulses and connected either to an electroporation device for pulse applied divided by the apparent chamber resistance during the application or to a current source for iontophoresis (see below). pulse (12). Unless otherwise noted,the cathode was in the donor compart- lontophoresis,alone or within 1 min subsequent to electro- ment and the anode in the receiver compartment. poration (started within I min after pulsing),was carried out "Enhancer"compounds were supplied in the donor solu- at constant current using a custom-built device.The total charge tion at the following comcentrations:10-2 M heparin(12 or transfered across the skin during an iontophoresis protocol was 20 kDa),10-2M dextran-sulfate (5 or 10 kDa),10-2 M neutral calculated by multiplying current times the duration of dextran (10 kDa),2 X 10-3 M poly-L-lysine or 0.16 M NaCI. application. The NaCl concentration was chosen to give the same electrical Abbreviated descriptions of electrical protocols are given conductivity as 10-2 M dextran-sulfate (10 kDa). in the tables and figures and can be interpreted according to After application of an electrical protocol (i.e.,electropora- the following examples:"5 X (150 V -180 ms)1/6 min" tion and/or iontophoresis),mannitol permeation was measured indicates that 5 pulses were applied with Ueteetrodes-150 V, for 6 h.Mannitol solution was present in the donor compartment Tpulse 180 ms and at a rate of I pulse every 6 min:"5 X both during and after pulsing,unless otherwise noted.In those (750 V-1.2 ms;0.1 s;74 V-350 ms)1/6 min"indicates cases when the mannitol solution was present only during puls- that 5 twin pulses were applied at a rate of I pulse every 6 ing and then removed afterwards,the donor compartment was min,where each twin pulse was composed of a first pulse with emptied,rinsed three times,and filled with buffer B within 1 Ueteetrodes=750 V and Tpulse=1.2 ms,followed after 0.I s by min after the last pulse.When mannitol solution was added a second pulse with Uelectrodes =74 V and Tpulse=350 ms;"1 only after the electrical protocol,pulsing was carried out with X (150 V-180 ms)+ionto 0.2 mA/cm2-24 min"indicates buffer A either with or without dextran-sulfate (10 kDa),and that one pulse was applied with Ueleetrode=150 V,Tpulse=180 replaced with mannitol solution within 1 min after the last pulse. ms and was followed by iontophoresis at 0.2 mA/cm-for 24 min
640 Vanbever,Prausnitz,and Preat Cumulative mannitol transported /ug/cm2 p
Macromolecules as Enhancers for Skin Electroporation 641 Table I.Influence of Electrical Parameters on Macromolecule-Enhanced Transport Cumulative mannitol transported after Total 6 h (ug/cm2)5 Total charge Without With application transfered dextran- dextran- Electrical Protocol time (min) (C) sulfate sulfate 5×(150V-180ms)1/6min 30 1.1 16±4 40±6 5×(150V-180ms)1/1tmin 5 1.1 16±2 51±10 5×(150V-180ms)1/15s 1.1 10.5±0.3 52±12 60×(500V-1.2ms)1/10-15s 12 0.4 8±1 17±3 120×(750V-1.2ms)1/15s 30 1.1 36±4 46±8 5×(750V-1.2ms:0.1s: 30 1.0 21±6 28±2 74V-350ms)1/6min ionto 0.2 mA/cm2-30 min 30 1.1 2.2±0.7 2.7±0.4 1×(150V-180ms)+ionto0.2 24 1.1 4±0.5 7±2 mA/cm2 24 min 5×(750V-1.2ms)1/15s+ionto 30 1.1 3.6±0.4 3.6±0.6 0.2 mA/cm2 -29 min passive diffusion 0.32±0.08 0.27±0.01 Explanation of electrical protocol abbreviations is given in the Methods section of the text. Expressed as mean sem. Dextran-sulfate (10 kDa). sured for 6 h.Figure 2 shows that the presence of dextran- and between pulses(Figure 2).This suggests that transport of sulfate during pulsing significantly increased post-pulse manni- mannitol occurred primarily by post-pulse diffusion through tol transport (Anova,p
642 Vanbever,Prausnitz,and Preat Enhancement Ratio greater dextran-sulfate penetration into skin can result in greater 0 1 3 enhancement of post-pulse mannitol transport. The effects of pulse length and pulse voltage were also Heparin(20 kDa) 888888888888888888* studied.In the experiments described so far,pulses of"medium" voltage and "long"time constant (e.g.,150 V,180 ms)were Heparin(12 kDa) 888888888网* used.While this type of protocol has received attention in the literature (12,21),pulses of higher voltage and shorter time 8888888388888888888889888888 constant (e.g.,500 V.1 ms)have also been studied (3.23). Dextran-sulfate (10 kDa) Figure 3 shows that the addition of dextran-sulfate during either type of protocol can enhance mannitol transport.However, Dextran-sulfate (5 kDa) 88888888* enhancement is considerably less when higher voltages and shorter pulses are used (Figure 3,Table I;t-test,p
Macromolecules as Enhancers for Skin Electroporation 643 enhancing mannitol permeation,even though the dextran-sul- pulse spacing,(b)long,medium-voltage (as opposed to short, fates tested had less charge and lower molecular weight(27). high-voltage)pulses and (c)pulse polarity which promoted When using long,medium-voltage pulses,t the positively- electrophoresis of macromolecules into skin each resulted in charged macromolecule poly-lysine increased mannitol trans- greater enhancement by macromolecules (Figures 3 and 5). port almost as well as negatively-charged dextran-sulfate.In Previous studies showed that (a)reduced inter-pulse spacing is contrast,poly-lysine significantly decreased mannitol transport associated with greater transdermal transport(12),(b)long, when short,high-voltage pulses were applied.In these experi- medium-voltage pulses may be associated with larger transder- ments with poly-lysine,the electrode polarity was reversed (i.e., mal transport pathways(25)and (c)electrophoresis contributes anode in the donor compartment)to promote electrophoresis significantly to transdermal transport of charged,hydrophilic of poly-lysine into the skin.It should be noted that under these compounds during skin electroporation (3,12).Therefore,the conditions (i.e.,anode in the donor compartment)the absolute greater enhancement of mannitol transport observed after puls- value of mannitol transport in the absence of enhancer was ing could be caused by greater introduction of macromolecules significantly lower for long,medium-voltage pulses (data not into skin during pulsing with these protocols.Direct measure- shown). ment of the amount of dextran-sulfate in skin was not performed To further evaluate the importance of molecular size and in this study,but would provide additional useful information charge,the enhancement provided by dextran-sulfate was com- Enhancement by macromolecules was observed for elec- pared to that of neutral dextran (having the same molecular troporation protocols,but not for passive diffusion or ionto- weight)and NaCl (present at a concentration which provided phoresis (Table 1,Figures I and 3).This shows that interactions the same amount of charge)using both anodal and cathodal between macromolecules and the transport pathways used dur- polarity(Figure 5).With the cathode in the donor compartment, ing non-electroporation protocols do not lead to enhancement. neutral dextran enhanced mannitol permeation while NaCl while interactions with transport pathways created at high volt- caused a significant decrease in transport.This suggests that age do enhance transport.This is consistent with the hypothesis enhancement caused by dextran-sulfate comes more from its that macromolecules stabilize electropores.These results also molecular size than its charge.When the electrode polarity was provide additional evidence that the high-voltage protocols used reversed (i.e.,anode in the donor compartment),dextran-sulfate here are not just "high-voltage iontophoresis,"but have dis- provided no enhancement,which is consistent with a mecha- tinctly different effects,probably due to the creation of new nism involving introduction of macromolecules into skin by transport pathways by skin electroporation (17). electrophoresis.However,the effects of neutral dextran and While the physicochemical properties of different macro- NaCl also disappeared when the "reverse"electrode polarity molecules affected the degree of enhancement,all macromole- was used,which cannot easily be explained by an electropho- cules examined in this study significantly enhanced mannitol retic or electroosmotic mechanism. transport(Figures 4 and 5).Although charged compounds (e.g.. dextran-sulfate)were somewhat better than uncharged ones DISCUSSION (e.g.,neutral dextran),both negatively-(e.g.,dextran-sulfate, heparin)and positively-(e.g.,poly-lysine)charged macromole- We wanted to determine if macromolecules could be effec- cules were good enhancers (Figures 4 and 5).Within each tive enhancers of transdermal transport by skin electroporation. chemical class,macromolecules of greater molecular weight. In this study we directly showed that macromolecules increase and therefore greater charge,were more effective than smaller. transdermal transport apparently by stabilizing the effects of less-charged macromolecules (e.g.,10 kDa vs.5 kDa dextran- skin electroporation.We also investigated the mechanism of sulfate;Figure 4).Small ions provided no enhancement at all macromolecular enhancement by examining the timescale of (e.g.,NaCl;Figure 5).Differences between chemical classes transport,the effects of electrical parameters,and the effects were also observed (e.g.,dextran-sulfates vs.heparins;Figure of macromolecule size,structure and charge.The results indi- 4).Together,this provides a preliminary guide for optimizing cate that a number of different macromolecules can interact with macromolecule properties for macromolecule-enhanced trans- skin during electric pulsing and thereby increase transdermal dermal transport by electroporation. transport by promoting and/or prolonging the effects of skin For applications in transdermal drug delivery,this study electroporation demonstrates ways in which macromolecules can be used and Transdermal mannitol transport was increased up to five- suggests other approaches not yet tested.The data show that fold by a number of different macromolecules present during macromolecules prolong the lifetime of skin permeabilization skin electroporation (Table 1,Figures 3-5).Levels of enhance- (Figure 2),suggesting that fewer and/or "weaker"electropora- ment were similar to those seen in previous electroporation tion pulses may be required when macromolecules are present studies involving cells (19).Significantly,if dextran-sulfate was The possibility that macromolecules may also enlarge transport present at the time of pulsing and then removed,transdermal pathways,as seen in cell membranes (19),has not been clearly transport of mannitol added after pulsing was enhanced(Figure demonstrated in skin.Pathways enlarged by macromolecule 2).This indicates that enhancement was provided by an interac- enhancers could provide a valuable approach to increasing tion between dextran-sulfate and skin rather than an interaction transdermal delivery of macromolecular drugs,such as insulin. between dextran-sulfate and mannitol.This also suggests that There is also the possibility for"self-enhancement"by macro- dextran-sulfate was introduced into skin at the time of pulsing molecular drugs,as may have occurred in studies with heparin and increased skin permeability for hours after pulsing. (14).Both charged and uncharged macromolecules enhanced The effects of electrical parameters on enhancement sug- transport of a neutral permeant (i.e.,mannitol,Figures 4 and gest that protocols which introduce more macromolecules into 5).In contrast,we previously observed that charged macromole- skin provide greater enhancement.The use of (a)reduced inter- cules may impede transdermal transport of permeants having the
644 Vanbever,Prausnitz,and Preat same charge (20),presumably by competition in electrophoretic 3.M.R.Prausnitz,V.G.Bose,R.Langer,and J.C.Weaver.Proc transport and charge repulsion.It is not yet known if neutral Natl.Acad.Sci.USA90:1050410508(1993). 4.M.R.Prausnitz.In Berner,B.and S.M.Dinh (eds),Electronically macromolecules or macromolecules having the opposite charge Controlled Drug Delivery,CRC Press,Boca Raton,FL,(in press). could enhance delivery of charged drugs.Finally,the macromol- 5. E.Neumann,A.E.Sowers,and C.A.Jordan.In Cell Biology, ecules studied here all had flexible,linear structures (28).Struc- Plenum Press,New York,1989. turally-different macromolecules (e.g.,globular)might 6. S.Orlowski and L. M.Mir.Biochim.Biophys.Acta 1154:51-63(1993). behave differently 7.J.C.Weaver.J.Cell.Biochem.51:426-435 (1993). 8. J.A.Nickoloff,eds.Animal cell electroporation and electrofusion CONCLUSIONS protocols,Humana Press,Totowa,N.J..1995. 9. R.Heller,M.J.Jaroszeski,L.F.Glass,J.L.Messina,D.P. Rapaport,R.C.DeConti,N.A.Fenske,R.A.Gilbert,L.M.Mir, Previous studies suggested that macromolecules could and D.S.Reintgen.Cancer 77:964-971 (1996). enlarge and stabilize transport pathways created by electropora- 10. C.Domenge,S.Orlowski,B.Luboinski,T.De Baere,G.Schwaab, tion.To further test this hypothesis,we examined the timescale J.Belehradek,and L.M.Mir.Cancer 77:956-963 (1996). of transport and the effects of electrical protocol and macromol- 11. U.Pliquett,T.E.Zewert,T.Chen,R.Langer,and J.C.Weaver. Biophys.Chem.,58:185-204(1996). ecule size,structure,and charge on enhancement of transdermal 12. R.Vanbever,N.Lecouturier,and V.Preat.Pharm.Res.11:1657- mannitol transport by a number of different macromolecules. 1662(1994). These studies support the hypothesis that macromolecules are 13. D.Bommannan,J.Tamada,L.Leung,and R.O.Potts.Pharm introduced into skin during electroporation and thereby stabi- Res.11:1809-1814(1994). 14.M.R.Prausnitz,E.R.Edelman,J.A.Gimm,R.Langer,and J. lize the increased skin permeability caused by high-voltage C.Weaver.Bio/Technology 13:1205-1209 (1995). pulses.This demonstrates that macromolecules can be used as 15. T.E.Zewert,U.F.Pliquett,R.Langer,and J.C.Weaver.Biochem. transdermal transport enhancers uniquely suited to skin Biophys.Res.Com.212:286-292(1995). electroporation. 16.M.R.Prausnitz,J.A.Gimm,R.H.Guy,R.Langer,J.C.Weaver, and C.Cullander.J.Pharm.Sci.85:1363-1370 (1996). 17. M.R.Prausnitz.J.Control.Rel.40:321-326 (1996). ACKNOWLEDGMENTS 18.R.Vanbever,G.Langers,S.Montmayeur,and V.Preat (submitted). We are grateful for the technical assistance of N.Lecoutur- 19. S.I.Sukharev,V.A.Klenchin,S.M.Serov,L.V.Chernomordik, ier and G.Langers.We thank Equibio for lending the electropor- and Y.A.Chizmadzhev.Biophys.J.63:1320-1327 (1992). 20.J.C.Weaver R.Vanbever,T.Vaughan,and M.R.Prausnitz ation device and Pharmacia Hepar Inc.for donating the heparin. (submitted). This work was supported in part by the Fonds National de la 21. R.Vanbever,E.Le Boulenge,and V.Preat.Pharm.Res.13:559- Recherche Scientifique (FNRS,Belgium)and a National Sci- 565(1996. 22. ence Foundation CAREER Young Investigator Award to Prof. K.Randerath.Chromatographie sur couches minces.,Gauthier- Villars (eds),Paris,1964. M.Prausnitz(BES-9624832).Prof.V.Preat is a Senior Research 23. U.Pliquett and J.C.Weaver.Bioelectrochem.Bioenerget. Associate of the FNRS (Belgium). 39:1-12(1996). 24. R.Vanbever,M.-A.Leroy,and V.Preat (submitted) 25.R.Vanbever,U.F.Pliquett,V.Preat,and J.C.Weaver (in REFERENCES oreparation ) 26. R.Vanbever and V.Preat.Bioelectrochem.Bioenerget.38:223- 1.J.Hadgraft and R.H.Guy,eds.Transdermal Drug Delivery: 228(1995) Developmental Issues and Research Initiatives,Marcel Dekker, 27.S.Budavari.The Merck Index.,Merck and Co.,Rahway,NJ,1996. New York,1989. 28.H.F.Mark,N.M.Bikales,C.G.Overberger,G.Menges,and J. 2.E.W.Smith and H.I.Maibach,eds.Percutaneous Penetration I.Kroschwitz.Encyclopedia of Polymer Science and Engineering, Enhancers,CRC Press,Boca Raton,FL,1995. Wiley-Interscience,1992