GENERALNI POKROVITELJ

Transcription

1 UDK 66:54(05) 6 Vol. 66 CODEN HMIDA 8 ISSN X Časopis Saveza hemijskih inženjera Srbije BEOGRAD, NOVEMBAR-DECEMBAR HMIDA 8, 66 (6) (2012)

2 GENERALNI POKROVITELJ HEMOFARM KONCERN VRŠAC, Beogradski put bb, tel. 013/ , , BEOGRAD, Prote Mateje 70, tel. 011/ , faks: E-pošta: IZDAVANJE ČASOPISA POMOGLA JE: INŽENJERSKA KOMORA SRBIJE Bulevar vojvode Mišića Beograd SUIZDAVAČI Tehnološko-metalurški fakultet Univerziteta u Beogradu, Beograd Prirodno-matematički fakultet Univerziteta u Novom Sadu, Novi Sad Hemijski fakutet Univerziteta u Beogradu Beograd Institut za tehnologiju nuklearnih i drugih mineralnih sirovina, Beograd HIP Petrohemija a.d. Pančevo Tehnološki fakultet Univerziteta u Novom Sadu, Novi Sad Tehnološki fakultet Univerziteta u Nišu, Leskovac NU Institut za hemiju, tehnologiju i metalurgiju Univerziteta u Beogradu, Beograd Nevena Color d.o.o. Leskovac DCP Hemigal, Leskovac

3 Chemical Industry Химическая промышленность PLASTIKA I GUMA PLASTICS & RUBBER Vol. 32, No. 1 2 Časopis Saveza hemijskih inženjera Srbije Journal of the Association of Chemical Engineers of Serbia Журнал Союза химических инженеров Сербии VOL. 66 Beograd, novembar decembar 2012 Broj 6 Izdavač Savez hemijskih inženjera Srbije Beograd, Kneza Miloša 9/I Glavni urednik Branko Bugarski Zamenica glavnog i odgovornog urednika Nevenka Bošković-Vragolović Urednica časopisa «Plastika i guma» Katarina Jeremić Urednici Katarina Jeremić, Ivana Banković-Ilić, Maja Obradović, Dušan Mijin Članovi uredništva Milorad Cakić, Željko Čupić, Željko Grbavčić, Katarina Jeremić, Miodrag Lazić, Slobodan Petrović, Milovan Purenović, Aleksandar Spasić, Dragoslav Stoiljković, Radmila Šećerov-Sokolović, Slobodan Šerbanović Članovi uredništva iz inostranstva Dragomir Bukur (SAD), Jiri Hanika (Češka Republika), Valerij Meshalkin (Rusija), Ljubiša Radović (SAD), Constantinos Vayenas (Grčka) Likovno-grafičko rešenje naslovne strane Milan Jovanović Redakcija Beograd, Kneza Miloša 9/I Tel/fax: 011/ E-pošta: shi@yubc.net Izlazi dvomesečno, rukopisi se ne vraćaju Za izdavača Tatijana Duduković Sekretar redakcije Slavica Desnica Izdavanje časopisa pomaže Republika Srbija, Ministarstvo prosvete i nauke Uplata pretplate i oglasnog prostora vrši se na tekući račun Saveza hemijskih inženjera Srbije, Beograd, broj , Komercijalna banka a.d., Beograd Kompjuterska priprema Vladimir Panić Štampa Razvojno-istraživački centar grafičkog inženjerstva, Tehnološko-metalurški fakultet, Univerzitet u Beogradu, Karnegijeva 4, Beograd Indeksiranje Radovi koji se publikuju u časopisu Hemijska Industrija ideksiraju se preko Thompson Reuters Scietific servisa Science Citation Index - Expanded TM i Journal Citation Report (JCR), kao i domaćeg SCIndeks servisa Centra za evaluaciju u obrazovanju i nauci SADRŽAJ Danijela D. Maksin, Slađana O. Kljajević, Maja B. Đolić, Jelena P. Marković, Bojana M. Ekmeščić, Antonije E. Onjia, Aleksandra B. Nastasović, Kinetic modeling of heavy metal sorption by vinyl pyridine based copolymer Ivana Vukoje, Dušan Božanić, Jasna Džunuzović, Una Bogdanović, Vesna Vodnik, Surface plasmon resonance of Ag organosols: Experimental and theoretical investigations Jasna V. Džunuzović, Marija V. Pergal, Vesna V. Vodnik, Milena Špírková, Rafał Poręba, Slobodan Jovanović, Ispitivanje morfologije i površinskih svojstava umreženih poli(uretan- -estar-siloksana) Marija M. Babić, Jovana S. Jovašević, Jovanka M. Filipović, Simonida Lj. Tomić, Diffusion of drugs in hydrogels based on (meth)acrylates, poly(alkylene glycol) (meth)acrylates and itaconic acid Snežana S. Ilić-Stojanović, Ljubiša B. Nikolić, Vesna D. Nikolić, Jela R. Milić, Slobodan D. Petrović, Goran M. Nikolić, Agneš J. Kapor, Potential application of thermo-sensitive hydrogels for controlled release of phenacetin Zoran Bjelović, Ivan S. Ristić, Jaroslava Budinski-Simendić, Mirjana Jovičić, Jelena Pavličević, Branka Pilić, Suzana Cakić, Ispitivanje kinetike reakcije nastajanja poliuretana na osnovu različitih tipova diizocijanata i ricinusovog ulja Jelena Pavličević, Milena Špirková, Jaroslava Budinski-Simendić, Mirjana Jovičić, Oskar Bera, Ivan Ristić, Uticaj masenog udela tvrdih segmenata na mehanička i termička svojstva poliuretanskih materijala na osnovu alifatskog polikarbonatnog diola Mirjana C. Jovičić, Oskar J. Bera, Jelena M. Pavličević, Vesna B. Simendić, Radmila Ž. Radičević, Uticaj udela montmorilonita na kinetiku umrežavanja epoksidnih nanokompozita Dragana D. Vasiljević, Ljiljana M. Đekić, Marija M. Primorac, Ispitivanje dugoročne stabilnosti kozmetičkih u/v kremova stabilizovanih mešanim emulgatorom NOVOSTI IZ STRUKE SADRŽAJ VOLUMENA 66(1 6) INDEKS AUTORA CONTENTS Danijela D. Maksin, Slađana O. Kljajević, Maja B. Đolić, Jelena P. Marković, Bojana M. Ekmeščić, Antonije E. Onjia, Aleksandra B. Nastasović, Kinetic modeling of heavy metal sorption by vinyl pyridine based copolymer

4 CONTENTS continued Ivana Vukoje, Dušan Božanić, Jasna Džunuzović, Una Bogdanović, Vesna Vodnik, Surface plasmon resonance of Ag organosols: Experimental and theoretical investigations Jasna V. Džunuzović, Marija V. Pergal, Vesna V. Vodnik, Milena Špírková, Rafał Poręba, Slobodan Jovanović, Investigation of the morphology and surface properties of crosslinked poly(urethane-ester-siloxane)s Marija M. Babić, Jovana S. Jovašević, Jovanka M. Filipović, Simonida Lj. Tomić, Diffusion of drugs in hydrogels based on (meth)acrylates, poly(alkylene glycol) (meth)acrylates and itaconic acid Snežana S. Ilić-Stojanović, Ljubiša B. Nikolić, Vesna D. Nikolić, Jela R. Milić, Slobodan D. Petrović, Goran M. Nikolić, Agneš J. Kapor, Potential application of thermo-sensitive hydrogels for controlled release of phenacetin Zoran Bjelović, Ivan S. Ristić, Jaroslava Budinski-Simendić, Mirjana Jovičić, Jelena Pavličević, Branka Pilić, Suzana Cakić, Investigation of formation kinetics of polyurethanes based on different types of diisocyanates and castor oil Jelena Pavličević, Milena Špirková, Jaroslava Budinski-Simendić, Mirjana Jovičić, Oskar Bera, Ivan Ristić, The inluence of hard segment content on mechanical and thermal properties of polycarbonate-based polyurethane materials Mirjana C. Jovičić, Oskar J. Bera, Jelena M. Pavličević, Vesna B. Simendić, Radmila Ž. Radičević, The influence of montmorillonite content on the kinetics of curing of epoxy nanocomposites Dragana D. Vasiljević, Ljiljana M. Đekić, Marija M. Primorac, Longterm stability investigation of o/w cosmetic creams stabilized by mixed emulsifier

5 Kinetic modeling of heavy metal sorption by vinyl pyridine based copolymer Danijela D. Maksin 1, Slađana O. Kljajević 2, Maja B. Đolić 1, Jelena P. Marković 1, Bojana M. Ekmeščić 2, Antonije E. Onjia 1, Aleksandra B. Nastasović 2 1 University of Belgrade, Vinča Institute of Nuclear Sciences, Belgrade, Serbia 2 University of Belgrade, ICTM Center for Chemistry, Polymer Department, Belgrade, Serbia Abstract Commercial macroporous poly(4-vinylpyridine-co-divinylbenzene) [P4VPD], known as Reillex-425, was characterized by mercury porosimetry, nitrogen physisorption, Fourier transformed infrared (FTIR) spectroscopy and elemental analysis. Sorption rates of P4VPD for Cu(II), Co(II) and Cr(VI) ions were determined in static non-competitive experiments, at room temperature (298 K). Rapid sorption was observed, especially for Co(II), with half time, t 1/2, of 1.5 min and high experimental maximal capacity, Q max, of 3.08 mmol g 1. Four kinetic models (pseudo-first and pseudo-second order model, intraparticle diffusion and Boyd model) were used for analyzing metal sorption by P4VPD. Metal ions sorption is well represented by the pseudo-second-order model, with definite influence of pore and film diffusion on sorption rates. SCIENTIFIC PAPER UDC 678.7/.8:546.3: Hem. Ind. 66 (6) (2012) doi: /HEMIND M Keywords: poly(4-vinylpyridine-co-divinylbenzene) macroporous copolymer; Cu(II), Co(II) and Cr(VI) ions sorption; kinetic models. Available online at the Journal website: Kinetic modeling of metal sorption is extremely important since the effectiveness of the sorption process for pollutant removal from aqueous solutions strongly depends on the sorption dynamics. Predicting the rate at which sorption takes place in a given system is one of the crucial factors in sorption system design [1]. Modeling of sorption kinetics can be carried out by employing chemical reaction-based and particle diffusion-based models. Among the most commonly encountered expressions in pollutant sorption studies are Lagregen s pseudo-first-order and Ho s pseudo-second- -order [2]. However, these and other chemical reaction-based kinetic models do not consider the importance of particle diffusion processes in metal sorption by porous materials [3]. Therefore, it is necessary to include diffusion-based kinetic modeling in order to elucidate the role of diffusion processes, such as film and intraparticle diffusion. Heavy metal pollution is an environmental problem of worldwide concern. Adsorption is an efficient and cost-effective process for treatment of water and industrial effluents. Numerous inorganic and organic adsorbents can be applied for heavy metal removal from wastewaters; polymeric chelating and ion-exchange resins are widely utilized for these purposes. Correspondence: A.B. Nastasović, University of Belgrade, ICTM Center for Chemistry, Polymer Department, Njegoševa 12, Belgrade, Serbia. anastaso@chem.bg.ac.rs; anastasovic@yahoo.com Paper received: 2 October, 2012 Paper accepted: 27 November, 2012 Cross-linked functional polymers (CFPs) based on 4-vinylpyridine (4VP) have been widely used as supports for adsorbents, metal catalysts, biomedical applications, etc. For example, quaternized poly(4-vinylpyridine-co-divinylbenzene) (P4VPD, Reillex HPQ) has been used for uranium sorption from acidic sulfate solution [4], hexavalent chromium removal from aqueous solutions [5] and perrhenate sorption from the acidic solutions [6]. Poly(acrylamide) grafted onto crosslinked P4VPD (Reillex 425) and quaternized with potassium chloroacetate has been used for removal of mercury from aqueous solutions [7]. Also, iodomethylated P4VPD are considered as promising polymer-supported catalysts [8]. In general, CFPs can be obtained in two ways: by treating the pre-synthesized polymer with an appropriate reagent to introduce desired moieties or by polymerization of a monomer which already carries the required functional group. The former route predominates in practice, i.e., the pre-synthesized copolymers are functionalized prior to application. However, additional functionalization of porous copolymers is not always recommended. Some problems can arise, such as undesired side reactions, changes of pore structure as the consequence of chemical modification, etc. Also, the introduction of very selective, but bulky ligands can have a negative effect on selectivity and metal uptake capacity of a chelating polymer [9]. To the best of our knowledge, only few papers regarding application of porous 4VP based copolymers not additionally reacted to introduce functionalities other than pyridine nitrogen evidence their effecti- 795

6 D.D. MAKSIN et al.: HEAVY METAL SORPTION BY VINYL PYRIDINE BASED COPOLYMER Hem. ind. 66 (6) (2012) veness. For example, REILLEX-425 (P4VPD) supported catalyst with Co(II) and Cr(VI) ions was successfully applied in aerobic liquid/phase partial oxidation of cyclohexane [10] as well as for deperoxidation of cyclohexyl hydroperoxide under mild conditions [11]. Also, macroporous crosslinked 4VP and ethylene glycol dimethacrylate copolymer, (P4VPE) synthesized in the presence of the inert component consisting of cyclohexanol and n-alkanes was used successfully for heavy metal ions sorption from aqueous solutions [12]. In this study, four kinetic models (chemical reaction- and diffusion-based) were used for correlating the kinetic data of Cu(II), Co(II) and Cr(VI) uptake by commercial macroporous REILLEX-425, P4VPD, from aqueous solutions. EXPERIMENTAL Analysis and spectroscopy Macroporous P4VPD (REILLEX-425), produced by Reilly Tarr & Chemical Corporation was used as received in sorption experiments. All other reagents and solvents were purchased from commercial sources and used as supplied. Copper(II), cobalt(ii) and chromium(vi) solutions were prepared from reagent grade copper chloride (Kemika), cobalt chloride (Carlo Erba) and potassium dichromate salt (Sigma-Aldrich), respectively, using deionized water. The copolymer samples were analyzed for their carbon, hydrogen and nitrogen content using the Vario EL III device (GmbH Hanau Instruments, Germany). Elemental analysis was calculated from multiple determinations with ±0.2% agreement. The IR spectra were recorded in KBr pellets on a Perkin-Elmer FT-IR 1725X spectrophotometer ( cm 1 ) with DGTS detector and IRDM software. The concentration of metal ions was measured by flame atomic absorption spectrometry (FAAS, SpektrAA Varian Instruments). The pore size distribution of P4VPD was determined by mercury porosimetry using a Carlo Erba Model 2000 instrument, operating in the interval of MPa. Sample preparation was performed at room temperature and pressure of 0.5 kpa. Nitrogen adsorption desorption isotherms were determined on Sorptomatic 1990 Thermo Finnigan automatic system using nitrogen physisorption at 77 K. Before measurement, samples were outgassed at 403 K for 10 h. Batch metal-uptake experiments The sorption of Cu(II), Co(II) and Cr(VI) ions from aqueous solutions (initial metal concentrations 0.05 M) was investigated in batch experiments under noncompetitive conditions, at room temperature (298 K). P4VPD (2.0 g) was soaked in 5 cm 3 of buffer solution (NaOAc/HOAc, ph 5.5) for 1 h. After that, the copolymer was contacted with 72.5 cm 3 of metal salt solution (0.05M) and 72.5 cm 3 of buffer solution. At appropriate times in each experiment, 0.5 cm 3 aliquots were removed, diluted to 50 cm 3 and used for metal concentration measurements. The reproducibility of the sorption experiments results was verified in triplicate. Standard statistical methods were used to determine the mean values and standard deviations for each set of data. The amount of metal ions sorbed onto unit mass of macroporous copolymer beads (sorption capacity, mmol g 1 ) was calculated by using the following expression: Q ( C C) V m 0 = (1) where C 0 and C are the concentrations of the metal ions in the initial solution and in the aqueous phase after treatment for certain period of time, respectively (in mmol cm 3 ), V is the volume of the aqueous phase (cm 3 ) and m is the amount of the copolymer beads used for the experiment (g). RESULTS AND DISCUSSION Characterization of P4VPD beads The commercial macroporous poly(4-vinylpyridine- -co-divinylbenzene), REILLEX-425 (P4VPD), is declared by the manufacturer as 4-vinylpyridine crosslinked with ca. 25% DVB and particle size of ca. 60 mesh (60 mesh equates to 250 μm). The assumed structural formula representation of P4VPD is given in Scheme 1. Scheme 1. Structure of commercial macroporous P4VPD. The elemental analysis of P4VPD yields: 78.58% C, 7.65% H, and 8.77% N. From these data, pyridine group concentration of 6.27 mmol g 1 was calculated. The structure and composition of Reillex 425 (as supplied by the manufacturer) was further verified by FTIR spectroscopy. The characteristic C=C and C=N vibrations of pyridine ring give rise to the absorption bands at 1600, 1557, 1492 and 1417 cm 1 (Figure 1) [13,14]. The band at 3024 cm 1 is assigned to the C=H group vibration in the aromatic ring. The absorption 796

7 D.D. MAKSIN et al.: HEAVY METAL SORPTION BY VINYL PYRIDINE BASED COPOLYMER Hem. ind. 66 (6) (2012) Figure 1. FTIR Spectrum of P4VPD. bands at 1070 and 100 cm 1 correspond to to the inplane and out-of-plane rings C H bending, respectively [13]. The presence of the vibration at 823 cm 1 in the spectrum is attributed to disubstituted aromatic rings of the copolymer. The adsorption bands at 2924 and 1450 cm 1 are assigned to the CH 2 of the ethyl group or aliphatic chain [14]. Two complementary methods were used for a proper evaluation of porous structure of P4VPD beads, mercury porosimetry and determination of nitrogen adsorption-desorption isotherms. The first one enables the detection of porous structure from macropores down to larger mesopores, while physisorption of nitrogen provides the most reliable results for pores in the micro- and mesopore regions [15]. Cumulative pore volume, V cum, and pore diameter which corresponds to half of the pore volume, d V/2, of P4VPD were calculated from the cumulative pore volume distribution curves while specific surface area, S Hg, calculation was based on cylindrical pore model as described in literature [15,16]. Specific surface area, S BET, of P4VPD was calculated by using the BET (Brunauer, Emmett and Teller) Equation [17]. The micropores, V micro were analyzed using the Dubinin Radushkevich Method [18]. Pore diameter distribution curves in mesopore region were obtained according to the Barrett, Joyner, Halenda (BJH) Method [19]. Mean pore diameter, d BJH, was derived from the pore diameter distribution curve. The value of mesopore volume, V meso, obtained from nitrogen adsorption-desorption isotherms and calculated using the BJH method is lower than V cum obtained from mercury porosimetry data. The relevant porosity parameters of P4VPD used in sorption tests, obtained by nitrogen physisorption and mercury porosimetry measurements are collected in Table 1. Table 1. Relevant P4VPD porosity parameters S Hg / m 2 g 1 60 d V/2 / nm 80 S BET / m 2 g 1 52 [11] V micro / cm 3 g [11] V meso / cm 3 g [11] V cum / cm 3 g [11] Mean pore diameter, nm 77 [11] Sorption kinetics From the standpoint of potential application, one of the most important properties of the chelating polymers is the rate at which sorption reaches equilibrium. Rapid sorption of metal ions is advantageous, providing a short residence time required for the completion of the actual process. The ph value of 5.5 was selected for the sorption experiments as close to neutral as possible, having in mind that at ph values higher than 6, significant hydrolysis of Co(II) and Cu(II) occurs at high concentrations of the order of magnitude used in this study (0.05 M) [20 22]. Thus precipitation of poorly soluble species occurs and a decline in Co(II) and Cu(II) supernatant con- 797

8 D.D. MAKSIN et al.: HEAVY METAL SORPTION BY VINYL PYRIDINE BASED COPOLYMER Hem. ind. 66 (6) (2012) centrations cannot be ascribed to adsorptive removal by P4VPD. The amount of Co 2+ sorbed by P4VPD rapidly increases with time reaching high maximal capacity already after 30 min (Figure 2). The uptake of Cu(II) and Cr(VI) is considerably lower, suggesting higher affinity of P4VPD towards Co 2+ at given ph. The sorption of Co(II) on P4VPD was found to be very fast. After 5 min, the amount of sorbed Co(II) ions was 90%, while after 30 min 97% was sorbed. The Cu(II) and Cr(VI) maximal sorption capacities were considerably lower in comparison with Co(II). The sorption half time, t 1/2, defined as the time required to reach 50% of the total sorption capacity, for all metals was below 10 min (Table 2). ph of the aqueous solution affects the surface charge of the adsorbent as well as metal speciation [23]. In the ph range different forms of chromium ions such as dichromate (Cr 2 O 2 7 ), hydrochromate (HCrO 4 ), and polychromates (Cr 3 O 2 10, Cr 4 O 2 13 ) coexist, of which HCrO 4 predominates [24]. The dichromate ion (Cr 2 O 2 7 ) is the dimmer of HCrO 4 which is formed when the concentration of chromium exceeds 1 g dm 3, applicable here. Pyridine nitrogen is not protonated to a significant extent at the ph of the experiment (pk a = 5.25 [25]) and the only possible interaction with the resin is hydrogen bonding. Therein lies the explanation for such low sorption capacity of P4VPD for Cr(VI). It can be presumed that Cr(VI) sorption would be enhanced at low ph. Lončarević et al. [11] have shown that Cr(VI) attaches to pyridine nitrogen of P4VPD in the form of dichromate anion. On the other hand, positively charged Co(II) and Cu(II) cations form coordination complexes with the nitrogen atom of the pyridine group as was demonstrated in some previous studies of poly(4-vinylpyridine- -co-divinylbenzene) [11,13]. The evident disparity in Co(II) and Cu(II) cations behavior stems most probably from the differing nature of these cations and their affinity for pyridine nitrogen (borderline Lewis base): while Cu(II) is classified as a soft Lewis acid, Co(II) falls under the borderline category and consequently the sorption capacity of P4VPD for the latter cation is high [26,27]. The t 1/2 values obtained for P4VPD are comparable with literature data. Sugii et al. found the t 1/2 values for Cu(II) and Ni(II) sorption on P4VPD to be 3 and 5 min, respectively [28]. Also, they found higher Cu(II) sorption rates for P4VPE than for P4VPD in acetate buffer [29]. The t 1/2 values for Hg(II) sorption on P4VPD quaternized with 2-chloroacetamide were 4 and 14 min from diluted mercury acetate and mercury chloride solutions, respectively [30]. 3.5 Q t, mmol g Co(II) Cu(II) Cr(VI) t, min Figure 2. Sorption of Cu(II), Co(II) and Cr(VI) ions versus time, on P4VPD (metal ions initial concentration 0.05 M, ph 5.5). Table 2. Sorption capacities after 5 min, Q 5, and 30 min, Q 30, maximum sorption capacities, Q max, and the sorption half time, t 1/2, for heavy metal sorption on P4VPD Metal Q 5 Q 30 Q max mmol g 1 mg g 1 mmol g 1 mg g 1 mmol g 1 mg g 1 t 1/2 / min Co(II) Cu(II) Cr(VI)

9 D.D. MAKSIN et al.: HEAVY METAL SORPTION BY VINYL PYRIDINE BASED COPOLYMER Hem. ind. 66 (6) (2012) Kinetic modeling Sorption mechanisms depend on the sorbate-sorbent interactions and the system conditions, making it impossible to classify sorption mechanisms by solute type [3]. A logical classification based on kinetic models was introduced and widely accepted. In order to examine the controlling mechanism of heavy metals sorption processes by P4VPD, such as mass transfer and chemisorption, several equations (the pseudo-first, the pseudo-second order, intraparticle diffusion and Boyd model) were tested to interpret the experimental data. Pseudo-first and pseudo-second order equations Probably the earliest known and one of the most widely used kinetic equations so far for the sorption of a solute from a liquid solution is Lagergren s Equation or the pseudo-first order equation [31]: ( ) ( kt) 1 log Qe Qt = logqe (2) where k 1 is the rate constant of pseudo-first-order sorption (min 1 ), Q e and Q t denote the amounts of sorbed metal ions at equilibrium and at time t (mmol g 1 ), respectively. A plot of log (Q e Q t ) versus t should give a straight line to confirm the applicability of the kinetic model. In a true first-order process, log (Q e ) should be equal to the intercept of a plot log (Q e Q t ) against t. The pseudo-second-order rate expression is used to describe chemisorption involving valency forces through the sharing or exchange of electrons between the adsorbent and adsorbate as covalent forces, and ion exchange [32]. A pseudo-second order equation is applied in the given form [1]: t 1 1 t Q = kq + Q (3) t 2 2 e e where k 2 (g 1 mmol 1 min 1 ) is the rate constant of the pseudo-second order sorption. A plot of t/q t versus t should give a linear relationship for second-order kinetics. Additionally, the initial adsorption rate h (mmol g 1 min 1 ) can be determined using the Eq. [1]: h= k Q (4) 2 2 e The pseudo-second-order equation has the following advantages: it does not have the problem of assigning an effective adsorption capacity, i.e., the adsorption capacity, the rate constant of pseudo-secondorder, and the initial adsorption rate all can be determined from the equation without knowing any parameter beforehand [1]. The rate constants k 1, k 2 and h, equilibrium sorption capacity, Q e, and the coefficients of determination, R 2, calculated from the values of intercepts and slopes of corresponding plots for the pseudo-first and the pseudo-second order equations are given in Table 3. Plots of log (Q e Q t ) vs. t (pseudo-first order) and t/q t vs. t (pseudo-second order) for heavy metals sorption by P4VPD are shown in Figure 3. Table 3. Kinetic parameters for Co(II), Cu(II) and Cr(VI) uptake using P4VPD as sorbent Parameter Sorbate Co Cu Cr Q e / mmol g Pseudo-first order k 1 / min Q calc e / mmol g R Pseudo-second order k 2 / g mmol 1 min h, / mmol g 1 min Q calc e / mmol g R Intraparticle k id / mmol g 1 min C id / mmol g R The theoretical Q e values estimated from the firstorder kinetic model were not in accordance with the experimental ones, and the coefficients of determination were found to be rather low. This indicates that the first-order kinetic model is not applicable to the Co(II), Cu(II) and Cr(VI) sorption on P4VPD. On the other hand, the theoretical Q e values calculated from pseudosecond order model were found to be very close to the experimental values of equilibrium sorption, Q e, with coefficients of determination higher than The initial sorption rate, h, is the highest for Co(II) and the lowest for Cr(VI), which reflects the affinity of the sorbent for these sorbate species. The superior fit of the pseudo-second-order model with experimental data implies that the adsorption process may be interaction controlled, with chemisorption involving valence force through sharing or exchange of electrons between P4VPD and heavy metal ions [33]. Plazinski et al., however, showed convincing proof that the pseudo-second order equation is able to represent the kinetics of sorption in the systems for which not only the rate of surface reaction governs the overall process rate [2]. Further on, Plazinski et al. drew a general conclusion that the expressions of the pseudo-first order and the pseudo-second order model did not correspond to only one kinetic model, but they presented more or less flexible mathematical formulae, able to simulate adequately well the behavior characte- 799

10 D.D. MAKSIN et al.: HEAVY METAL SORPTION BY VINYL PYRIDINE BASED COPOLYMER Hem. ind. 66 (6) (2012) B J R log (Q eq -Q t ) t, min (a) Co Cu Cr 250 t/q t t, min (b) Figure 3. Pseudo-first (a) and pseudo-second order kinetics (b) of the heavy metals uptake by P4VPD at 25 C. ristic of physical kinetic processes of various kinds (i.e., surface reaction and intraparticle diffusion) [2]. Diffusion-based kinetic modeling Generally, if the investigated sorption process involves metal species and porous sorbent, it is a multistep process, including [34]: i) bulk diffusion, ii) external mass transfer of sorbate molecules across the liquid film around the sorbent particles, iii) binding of sorbate molecules on the active sites distributed on the outer surface of the sorbent particles, iv) intraparticle diffusion of sorbate molecules into macro-, meso- and micropores, and v) sorption of sorbate molecules onto active sites distributed within the sorbent particles. The steps iii and v are usually very fast and hence they do not have a determinant role in governing sorption rates. Any of the four previous steps may be the rate controlling factor or any combination of the steps [3]. The entire sorption process is thus affected by changes to any step, and these steps can be affected by many experimental conditions, including particle size, agitation rate and ph, among others [3]. Thus, external mass transfer and intraparticle diffusion warrant further consideration. Since the pseudo-first and the pseudo-second order kinetic models cannot identify the influence of diffusion on sorption, the Weber and Morris equation was used for calculation of the rate constants of intraparticle diffusion [35]. The intraparticle diffusion model presumes that film diffusion or boundary layer diffusion is negligible, and that intraparticle diffusion is the only ratecontrolling step. The rate of intraparticle diffusion can be calculated according to the equation [36]: Q = C + k t (5) t id 0.5 id 800

11 D.D. MAKSIN et al.: HEAVY METAL SORPTION BY VINYL PYRIDINE BASED COPOLYMER Hem. ind. 66 (6) (2012) where k id is the intraparticle diffusion rate constant (mmol g 1 min 0.5 ), and C id is the intercept which is proportional to the boundary layer thickness (mmol g 1 ) [37]. The rate constant of intraparticle transport, k id, is estimated from the slope of the linear portion of the plot of amount sorbed against square root of time. In the case of a linear plot of Q t versus t 0.5, and if the line passes through the origin, intraparticle diffusion is the only rate-controlling step [38]. If not, some other mechanisms are also involved. For investigated metal ions sorption by porous P4VPD, the plots Q t versus t 1/2 did not pass through the origin (Figure 4) suggesting that even though the adsorption process involved intraparticle diffusion, it was not the only rate-controlling step. The positive value of intercept C id is indicative of some degree of boundary layer control [38]. The multilinear shape of Q t t 1/2 relationships indicates that more than one process affects Cr(VI) adsorption. As can be seen from Figure 5, plot Q t versus t 1/2 for Co(II) has first sharper portion, which can be considered as an external surface adsorption or faster adsorption stage, followed by gradual adsorption where intraparticle diffusion is rate controlled. After that, in the final equilibrium stage the intraparticle diffusion starts to slow down due to the low adsorbate concentration in solution. Kumar et al. observed the similar for Cr(III) removal by using an amine-based polymer, aniline formaldehyde condensate (AFC) coated on silica gel [39]. The rate of uptake is limited by the size of adsorbate molecule, the adsorbate concentration and its affinity towards the adsorbent, the diffusion coefficient of the adsorbate in the bulk, the adsorbent pore size distribution and the degree of mixing step [38]. In this case, it is apparent from Table 3. that the size of the metal species and their affinity towards P4VPD are crucial for the observed trend in the k id values. Intraparticle diffusion is facilitated by the presence of macroporosity. The dependence of the amount adsorbed on the square-root of time has a concave character for the two polymers. A model investigation by Rudzinski et al. has shown that such curve shape may be due to a combined effect of the rate of surface reaction and that of the solute transport from the bulk to the surface [2], in support of deductions made in the case of this sorbate sorbent system. The contribution of boundary layer or film diffusion indicated by non-zero values of the intraparticle plot intercept is often confirmed using the model given by Boyd [36]: 6 1 i F = 1 exp π 2 2 n π Dt n= 1 n r (6) 2 2 n= 1 ( n 2 Bt) 6 1 F = 1 exp π (7) n where F is the fractional attainment of equilibrium at time t (min) obtained from the expression: Qt F = (8) Q and e 2 D i π B = (9) 2 r Q t, mmol g Co Cu Cr t 0.5, min 0.5 Figure 4. Intraparticle diffusion plots for Cu(II), Co(II) and Cr(VI) uptake using P4VPD as sorbent. 801

12 D.D. MAKSIN et al.: HEAVY METAL SORPTION BY VINYL PYRIDINE BASED COPOLYMER Hem. ind. 66 (6) (2012) 5 4 Co Cu Cr 3 Bt t, min Figure 5. The Boyd plots for sorption of Cu(II), Co(II) and Cr(VI) using P4VPD as sorbent. where B is the time constant (min 1 ), D i is the effective diffusion coefficient of the metal ions in the sorbent phase (cm 2 min 1 ), r is the radius of the sorbent particle (cm), assumed to be spherical, n is an integer that defines the infinite series solution. The approximations for Bt proposed by Reichenberg [40] are as follows: π F 2 F values < 0.85, Bt = π π 3 (10) F values > 0.85, Bt = ln(1 F) (11) Thus, the value of Bt can be computed for each value of F, and then plotted against time. The linearity of these so-called Boyd Plots (Figure 5) was employed to distinguish between sorption controlled by film diffusion and particle diffusion [20]. If the plot is in the form of straight line passing through the origin, this indicates that sorption processes are governed by particle-diffusion mechanisms; otherwise, they are controlled by film diffusion [41]. From Figure 5, it was observed that the plots were neither linear nor passed through the origin for Co(II), Cu(II) and Cr(VI) indicating the film diffusion-controlled mechanism. The strong external resistance which hinder the external mass transfer and is rate-limiting during the initial stages of sorption (0 5 min as seen in Figure 5) may be mainly due to the the absence of mixing and high affinity of adsorbate to adsorbent [42]. The intraparticle step limits the overall transfer in the sorption period of 5 60 min since the investigated systems have high concentrations of adsorbate and relatively large particle size of adsorbent. In order to make distinction between kinetic and diffusion control, a very general guideline can be used: if equilibrium is achieved within 3 h, the process is usually kinetic controlled and above 24 h, it is diffusion controlled [3]. Either or both kinetic and sorption processes may be rate controlling in the 3 to 24 h period, such as in the case of Cu(II), Co(II) and Cr(VI) sorption by commercial macroporous REILLEX-425, P4VPD. CONCLUSION Commercial macroporous crosslinked poly(4-vp-co- -DVB), P4VPD (REILLEX 425), was used as heavy metal ions sorbent from aqueous solutions without additional functionalization. In this manner, some problems, such as the undesired side reactions and the change of pore structure as a consequence of chemical modification, are avoided, simplifying the process and reducing the costs as well. Kinetics of Co(II), Cu(II) and Cr(VI) removal from aqueous solutions was tested in batch, under non-competitive conditions at room temperature and analyzed using pseudo-first order, pseudo-second order, intraparticle diffusion and Boyd s Model. The very high maximum Co(II) capacity of 181 mg g 1 of P4VPD was observed. Kinetic studies showed that the adsorption adhered to the pseudo-second-order model since theoretical and experimental sorption capacities were in excellent agreement, with R The intraparticle diffusion model revealed that pore diffusion was not the only rate-controlling step and indicated some degree of boundary layer control in the process of heavy metal sorption by the porous copolymer. Boyd plots showed that film diffusion mechanism may be involved in the process of Co(II), Cu(II) and Cr(VI) sorption by P4VPD. 802

13 D.D. MAKSIN et al.: HEAVY METAL SORPTION BY VINYL PYRIDINE BASED COPOLYMER Hem. ind. 66 (6) (2012) Acknowledgements This work was supported by the Ministry of Education, Science and Technological Development of the Republic of Serbia (Project III 43009). REFERENCES [1] Y.S. Ho, Review of second-order models for adsorption systems, J. Hazard. Mater. 136 (2006) [2] W. Plazinski, W. Rudzinski, A. Plazinska, Theoretical models of sorption kinetics including a surface reaction mechanism: A review, Adv. Colloid. Interface Sci. 152 (2009) [3] Y.S. Ho, J.C. Y. Ng, G. McKay, Kinetics of pollutant sorption by biosorbents: Review, Sep. Purif. Method. 29 (2000) [4] M. Chanda, G. Rempel, Uranium sorption behavior of a macroporous, quaternized poly(4-vinylpyridine) resin in sulfuric acid medium, React. Polym. 18 (1992) [5] V. Neagu, S. Mikhailovsky, Removal of hexavalent chromium by new quaternized crosslinked poly(4-vinylpyridines), J. Hazard. Mater. 183 (2010) [6] D. Jermakowicz-Bartkowiak, B. N. Kolarz, Poly(4-vinylpyridine) resins towards perrhenate sorption and desorption, React. Funct. Polym. 71 (2011) [7] E. Yavuz, B. F. Senkal, N. Bicak, Poly(acrylamide) grafts on spherical polyvinyl pyridine resin for removal of mercury from aqueous solutions, React. Funct. Polym. 65 (2005) [8] M. Giammatteo, L.Tauro, A.A. D Archivio, L. Galantini, A. Panatta, E. Tettamanti, K. Jerabek, B. Corain, Crosslinked poly-4-vinylpyridines as useful supports in metal catalysis: micro- and nanometer scale morphology, J. Mol. Catal., A 268 (2007) [9] P.M. Van Berkel, S.C. Van der Slot, W.L. Driessen, J. Reedijk, D.C. Sherrington, Influence of the polymer matrix on the metal-ion uptake characteristics of ligandmodified poly(glycidyl methacrylate-co-trimethylolpropane trimethacrylate) polymers, Eur. Polym. J. 33 (1997) [10] D. Lončarević, Ž. Čupić, M. Odović, Inhibition Effects in the partial oxidation of cyclohexane on polymer supported Co(II) catalysts, J. Serb. Chem. Soc. 70 (2005) [11] D. Lončarević, J. Krstić, J. Dostanić, D. Manojlović, Ž. Čupić, D.M. Jovanović, Cyclohexane oxidation and cyclohexyl hydroperoxide decomposition by poly(4-vinylpyridine-co-divinylbenzene) supported cobalt and chromium complexes, Chem. Eng. J. 157 (2010) [12] A. Nastasović, D. Đorđević, D. Jakovljević, T. Novaković, Z. Vuković, S. Jovanović, Heavy metal ions removal with macroporous poly poly(4-vinyl pyridine-co-ethylene glycol dimethacrylate), in: R.K. Bregg (Ed.), Leading Edge Polymer Research, Nova Science Publishers, New York, 2006, pp [13] K.H. Wu, Y.R. Wang, W.H. Hwu, FTIR and TGA studies of poly(4-vinylpyridine-co-divinylbenzene) Cu(II) complex, Polym. Degrad. Stab. 79 (2003) [14] J. Ortiz-Palacios, J. Cardoso, O. Manero, Production of macroporous resins for heavy-metal removal. I. Nonfunctionalized polymers, J. Appl. Polym. Sci. 107 (2008) [15] P.A. Webb, C. Orr, Analytical Methods in Fine Particle Technology, Micromeritics Instrument Corporation, Norcross, 1997, pp [16] S. Jovanović, A. Nastasović, N. Jovanović, K. Jeremić, Z. Savić, The influence of inert component composition on the porous structure of glycidyl methacrylate/ethylene glycol dimethacrylate copolymers, Angew. Makromol. Chem. 219 (1994) [17] S. Brunauer, P.H. Emmett, E. Teller, Adsorption of gases in multimolecular layers. J. Am. Chem. Soc. 60 (1938) [18] M.M. Dubinin, Physical adsorption of gases and vapors in micropores, in: D.A. Cahenhead (Ed.), Progress in Surface and Membrane Science, 9 th ed., Academic Press, New York, 1975, pp [19] E.P. Barrett, L.G. Joyner, P.P. Halenda, The determination of pore volume and area distributions in porous substances. I. Computations from nitrogen isotherms, J. Am. Chem. Soc. 73 (1951) [20] R. Djeribi, O. Hamdaoui, Sorption of copper(ii) from aqueous solutions by cedar sawdust and crushed brick, Desalination 225 (2008) [21] C.P. Huang, M.W. Tsang, Y.S. Hseigh, The removal of cobalt (II) from water by activated carbon, AIChE Symp. Ser. 243 (1985) [22] M.V. Dinu, E.S. Dragan, Heavy metals adsorption on some iminodiacetate chelating resins as a function of the adsorption parameters, React. Funct. Polym. 68 (2008) [23] S.M. Nomanbhay, K. Palanisamy, Removal of heavy metal from industrial wastewater using chitosan coated oil palm shell charcoal, Electron J. Biotechnol. 8 (2005) [24] T. Karthikeyan, S. Rajgopal, L. R. Miranda, Cr(VI) adsorption from aqueous solution by Hevea Brasilinesis sawdust activated carbon, J. Hazard. Mater. 124 (2005) [25] Y. Zubavichus, M. Zharnikov, Y. Yang, O. Fuchs, E. Umbach, C. Heske, A. Ulman, M. Grunze, XPS and NEXAFS study of water adsorption on the pyridine-terminated thiolate self-assembled monolayer, Langmuir 20 (2004) [26] R.G. Pearson, Hard and soft acids and bases, HSAB. Part I, J. Chem. Educ. 45 (1968) [27] R.G. Pearson, Hard and soft acids and bases, HSAB. Part II, J. Chem. Educ. 45 (1968) [28] A. Sugii, N. Ogawa, Y. Iinuma, H. Yamamura, Selective metal sorption on cross-linked poly(vinylpyridine) resins, Talanta 28 (1981) [29] A. Sugii, N. Ogawa, K. Harada, K. Nishimura, Metal sorption of macroreticular poly(4-vinylpyridine) resins crosslinked with oligo(ethylene glycol dimethacrylates), Anal. Sci. 4 (1988)

14 D.D. MAKSIN et al.: HEAVY METAL SORPTION BY VINYL PYRIDINE BASED COPOLYMER Hem. ind. 66 (6) (2012) [30] H.B. Sonmez, N. Bicak, Quaternization of poly(4-vinyl pyridine) beads with 2-chloroacetamide for selective mercury extraction, React. Funct. Polym. 51 (2002) [31] A.G. Ritchie, Alternative to the Elovich equation for the kinetics of adsorption of gases on solids, J. Chem. Soc. Faraday Trans. 73 (1977) [32] Y.S. Ho, G. McKay, A comparison of chemisorption kinetic models applied to pollutant removal on various sorbents, Process Saf. Environ. Prot. (Trans. Ichem. E., Part B), 76 (1998) [33] H. Zhang, Y. Tang, D. Cai, X. Liu, X. Wang, Q. Huang, Z. Yu, Hexavalent chromium removal from aqueous solution by algal bloom residue derived activated carbon: equilibrium and kinetic studies, J. Hazard. Mater. 181 (2010) [34] A.E. Ofomaja, Sorptive removal of Methylene Blue from aqueous solution using palm kernel fibre: Effect of fibre dose, Biochem. Eng. J. 40 (2008) [35] W.J. Weber, J.C. Morris, Kinetics of adsorption on carbon from solution, J. Sanit. Eng. Div., Am. Soc. Civ. Eng. 89 (1963) [36] G.E. Boyd, A.M. Adamson, L.S. Myers, The exchange adsorption of ions from aqueous solutions by organic zeolites. Part II,.J. Am. Chem. Soc. 69 (1947) [37] A.M. Donia, A.A. Atia, W.A. Al-amrani, A. M. El-Nahas, Effect of structural properties of acid dyes on their adsorption behaviour from aqueous solutions by amine modified silica, J. Hazard. Mater. 161 (2009) [38] A.S. Özcan, A. Özcan, Adsorption of acid dyes from aqueous solutions onto acid-activated bentonite, J. Colloid Interface Sci. 276 (2004) [39] P.A. Kumar, M. Ray, S. Chakraborty, Adsorption behaviour of trivalent chromium on amine-based polymer aniline formaldehyde condensate, Chem. Eng. J. 149 (2009) [40] D. Reichenberg, Properties of ion-exchange resins in rela- tion to their structure III, kinetics of exchange, J. Am. Chem. Soc. 75 (1953) [41] D. Mohan, K.P. Singh, Single- and multi-component adsorption of cadmium and zinc using activated carbon derived from bagasse-an agricultural waste, Water Res. 36 (2002) [42] V. Vadivelan, K.V. Kumar, Equilibrium, kinetics, mechanism, and process design for the sorption of methylene blue onto rice husk, J. Colloid Interface Sci. 286 (2005) IZVOD KINETIČKO MODELOVANJE SORPCIJE TEŠKIH METALA KOPOLIMEROM SA OSNOVOM VINIL PIRIDINA Danijela D. Maksin 1, Slađana O. Kljajević 2, Maja B. Đolić 1, Jelena P. Marković 1, Bojana M. Ekmeščić 2, Antonije E. Onjia 1, and Aleksandra B. Nastasović 2 1 Univerzitet u Beogradu, Institut za nuklearne nauke Vinča, Beograd, Srbija 2 Univerzitet u Beogradu, IHTM Centar za hemiju, Odeljenje za polimere, Beograd, Srbija (Naučni rad) Komercijalni makropozni poli(4-vinilpiridin-co-divinilbenzen) [P4VPD], Reillex- 425, bez dodatne funkcionalizacije korišćen je za ispitivanje mogućnosti sorpcije teških metala iz vodenih rastvora. Na taj način su izbegnuti neki problemi do kojih može doći prilikom uvođenja funkcionalnih grupa selektivnih za određene jone metala u polimerne sorbente, kao što su nepoželjne sporedne reakcije i promena porozne strukture. P4VPD je okarakterisan živinom porozimetrijom, određivanjem adsorpciono-desorpcionih niskotemperaturnih izotermi azota, infracrvenom spektroskopijom (FTIR) i elementalnom analizom. U statičkim eksperimentima su određene brzine sorpcije Cu(II), Co(II) i Cr(VI) jona pomoću P4VPD pri nekompetitivnim uslovima i pri sobnoj temperaturi (298 K). Zapažena je brza sorpcija ispitivanih jona, naročito u slučaju Co(II), pri čemu su zapažene vrednosti poluvremena sorpcije, t 1/2, od 1,5 min i visoka vrednost maksimalnog kapaciteta sorpcije, Q max, od 3,08 mmol g 1 (181 mg g 1 ). Prilikom analiziranja sorpcije ispitivanih metala pomoću P4VPD korišćena su četiri kinetička modela (model pseudo-prvog i pseudo-drugog reda, model difuzije unutar čestica i Bojdov model). Pokazalo se da sorpciju jona metala najbolje opisuje model pseudo-drugog reda, pošto je za ovaj model dobijeno odlično slaganje eksperimentalnih i teorijskih vrednosti kapaciteta sorpcije, kao i vrednosti R 2 0,99. Analiza eksperimentalnih rezultata pomoću modela difuzije unutar čestica pokazala je da difuzija kroz pore nije jedini stupanj koji određuje brzinu sorpcije i ukazala na delimičan uticaj graničnog sloja na sorpciju teških metala pomoću poroznog kopolimera. Ključne reči: Poli(4-vinilpiridin-co-divinilbenzen) Makroporozni kopolimer Sorpcija Cu(II), Co(II) i Cr(VI) jona Kinetički modeli 804

15 Surface plasmon resonance of Ag organosols: Experimental and theoretical investigations Ivana Vukoje 1, Dušan Božanić 1, Jasna Džunuzović 2, Una Bogdanović 1, Vesna Vodnik 1 1 University of Belgrade, Vinča Institute of Nuclear Sciences, Belgrade, Serbia 2 University of Belgrade, Institute of Chemistry, Technology and Metallurgy (ICTM) Center of Chemistry, Belgrade, Serbia Abstract The aim of this study was to investigate and compare the changes in surface plasmon resonance (SPR) of silver (Ag) hydrosol and organosols obtained by experimental and theoretical approaches. Silver nanoparticles (Ag NPs) of 5±1.5 nm in diameter were prepared in water by reduction of silver nitrate with sodium borohydride. Nanoparticles were subsequently transferred into different organic solvents (chloroform, hexane, toluene, 1,2-dichlorobenzene) using oleylamine as a transfer agent. These solvents were chosen because of the differences in their refractive indices. Using UV Vis absorption spectrophotometry and transmission electron microscopy (TEM), we confirmed that there were no shape and size changes of the nanoparticles upon the transfer to the organic phase. The absorption spectra of the obtained Ag organosols showed changes only in the position of the SPR band depending on the dielectric property of the used solvent. To analyze these changes, absorption spectra were modelled using Mie theory for small spherical particles. The experimental and theoretical resonance values were compared with those predicted by the Drude model and its limitations in the analysis of absorption behavior of Ag NPs in organic solvents were briefly discussed. SCIENTIFIC PAPER UDC 620.1:546.57: Hem. Ind. 66 (6) (2012) doi: /HEMIND V Keywords: silver nanoparticles; organosol; Mie theory; Drude model. Available online at the Journal website: The investigation of the specific features of metal nanoparticles have been the subject of the numerous studies for a long time, due to their unique electronic, optical, chemical and magnetic properties, that are different from those of the individual atoms as well as from their bulk counterparts [1 5]. Nanoparticles composed of noble metals such as Au and Ag exhibit strong absorption of electromagnetic radiation in the visible region, which is often referred to as a surface plasmon resonance (SPR) property. Such resonance occurs when the frequency of the incident light is resonant with the collective oscillation of the conduction band electrons of metal nanoparticles. The position and width of the SPR band of metal colloids are strongly influenced by the size, shape and the environment of the particles, and deeper understanding of these effects is necessary for understanding the optical properties of these systems [6 8]. It is well known that Mie, in 1908, was the first to develop the theory of absorption and scattering of light by spherical particles [9]. The calculations of absorption spectra of metallic nanoparticles clearly reflect the well-known dependence of nanoparticle optical properties, i.e., the resonance wavelength, the ex- Correspondence: V. Vodnik, Vinča Insitute of Nuclear Sciences, P.O. Box 522, Belgrade, Serbia. vodves@vinca.rs Paper received: 12 January, 2012 Paper accepted: 9 March, 2012 tinction cross-section, and the ratio of scattering to absorption, on the nanoparticle dimensions. On the other hand, the effects of the particle environment on the SPR band are manifested by a red-shift of the resonance wavelength with the increase of the refractive index of the surrounding medium. These effects have been used for sensor applications [10 12]. In addition, metal nanoparticles have also found application in many different fields such as biochemistry, photography or catalysis [13 18]. Thus far, a variety of synthetic methods has been reported for metallic nanoparticles synthesis [19 25]. One of the main challenges in the preparation of metal colloids is developing a way to obtain the nanoparticles in specific physicochemical environments, such as organic nonpolar liquids, specific regions within ordered surfactant phases and monolayer assemblies [26 28]. The understanding of the physicochemical properties of metal organosols is important to their potential use in future photonic applications [27, 29 35]. The silver organosols display several advantages over silver hydrosols, such as the ability to be isolated and re-dissolved in solvents without irreversible aggregation, stability in air, and so forth. Attempts have been made to obtain nearly monodisperse Ag NPs in different nonpolar solvents chloroform, hexane, toluene, dichlorobenzene, etc. In this study, we synthesized Ag NPs in water and then transferred them into organic solvents 805

16 I. VUKOJE et al.: SURFACE PLASMON RESONANCE OF Ag ORGANOSOLS Hem. ind. 66 (6) (2012) by simple surface modification using oleylamine as surfactant. A capping agent (oleylamine) was used to protect the particles from aggregation. Theoretical and experimental aspects of the optical properties of Ag NPs in different solvents have been discussed. Special attention was paid to the correlation between nanoparticles environment and their optical properties. EXPERIMENTAL Materials Silver nitrate (AgNO 3 ), sodium borohydride (NaBH 4 ), oleylamine, chloroform, hexane, toluene, and 1,2-dichlorobenzene were purchased from Merck and used as received. Milli-Q deionized water was used for synthesis. Preparation of silver sols Silver hydrosol was prepared by the reduction of silver ions using NaBH 4, as described elsewhere [24,36]. Typically, 10 mg of NaBH 4 was added into 250 ml of argon-saturated solution of M AgNO 3. Yellow silver hydrosol was formed immediately without any additional stabilizer. Due to hydrolysis of excess NaBH 4, ph increased and reached value of 9.8. The silver hydrosol was stable several hours in argon atmosphere. The transfer of Ag NPs from the hydrosol into organic phase (chloroform) was performed using oleylamine as transfer agent. To do so, 250 ml of M prepared hydrosol was mixed with 25 ml of chloroform containing M oleylamine and stirred vigorously. Two layers with a clean boundary in the bottle were observed. The bottom layer became yellowbrownish while the upper aqueous layer became colorless, evidently due to the transfer of Ag NPs from the aqueous phase to the organic phase. This can be confirmed from the photographs of Ag NPs in both aqueous and organic phases (see insets of Figures 1 and 3). Silver NPs in other organic solvents, beside chloroform, were prepared by evaporating chloroform from the 1 ml of initial organosol and subsequent re-dispersion of the sediment into equal amount of hexane, toluene, and 1,2-dichlorobenzene. All prepared Ag organosols were stable for at least few months with no evident changing. Apparatus The size distributions of Ag NPs were determined using a Philips EM-400 transmission electron microscope (TEM) operating at 100 kv. The samples were prepared by placing a drop of silver colloid onto a carbon-coated Cu grid, which was allowed to dry in air. Concentrations of the Ag NPs in water and chloroform used to prepare TEM samples were and 0.54 mg/ml, respectively. UV Vis absorption spectra of the silver hydrosol and organosols were acquired by using a Termoscientific Evolution 600 spectrophotometer at room temperature. RESULTS AND DISCUSSION The strong reducing agent (sodium borohydride) produced a sol containing silver particles, whose average diameter was smaller than the wavelength of visible light. During chemical reduction of silver ions, silver atoms are formed, aggregating into nuclei of a new phase, which can grow and yield stable Ag NPs. An important role for the final particle distribution is played by agglomeration of very small, primarily formed particles. The shape and size distribution of prepared Ag NPs were examined by TEM and a representative image of silver hydrosol and its photograph are given in Figure 1a. It can be seen that the NPs are mainly spherical and isolated. Although we did not use any stabilizing agents in the preparation procedure, the absence of aggregates in the sample can be attributed to the 4 adsorption of B(OH) on the surface of nanoparticles [36]. According to this, particle stability is achieved through the electrostatic repulsion between the negatively charged Ag NPs. Based on TEM measurements, the corresponding particle size distribution is presented in Figure 1b. The particle sizes were between 4 and 7 nm with an average diameter d av = 5.6 nm and polydispersity 32%. Polydispersity represents the ratio between the full width at half maximum (FWHM) and average diameter of Ag NPs according to Figure 1b. The UV Vis absorption spectrum of the initial Ag hydrosol (Figure 2a) shows intensive absorption of Ag NPs arising from the collective oscillations of the free conduction band electrons that are induced by the incident electromagnetic radiation. The yellow color of the hydrosol is a manifestation of narrow SPR band with maximum at around 389 nm. To model observed absorption spectra of silver colloids, we used Mie theory [9] for the calculation of the optical properties of Ag NPs. According to the theory, the extinction efficiency Q ext of a single small metallic sphere of radius r in a dielectric environment with dielectric constant ε m is given by: Q ext ( λ) ( ) ( ) ε λ εm = 4xIm ε λ + 2εm where x= εm kr and k = 2π/λ. In the preceding relation, ε(λ) stand for complex dielectric function of silver nanoparticles, which is in our study obtained using the Hao and Nordlander fit [37] of the Johnson and Christy experimental data for bulk silver [38] and after appropriate correction for small particle size according to Hovel et al. [39] (Figure 2b). Theoretical absorption (1) 806

17 I. VUKOJE et al.: SURFACE PLASMON RESONANCE OF Ag ORGANOSOLS Hem. ind. 66 (6) (2012) Figure 1. Typical TEM image of Ag NPs with photograph of Ag hydrosol (a) and corresponding particle size distribution (PSD) (b). Figure 2. Experimental (solid black line, a) and theoretical (dashed red line, b) UV Vis absorption spectra of Ag NPs. band for 5 nm Ag NPs in water (refractive index n = = 1.33) is given in Figure 2b. It can be seen that the main features of the experimental spectrum are reproduced by Mie theory. Furthermore, the theoretical band is positioned at 390 nm, which is in good agreement with the experimental measurement. To achieve the phase transfer of Ag NPs from water into nonpolar organic solvents, it was necessary to use a surfactant that forms adsorption layer on the Ag NPs. The oleylamine was used as a capping agent, which enabled the phase transfer and prevented particle aggregation, oxidation and degradation, as well as to render the particle surface hydrophobic. The oleylamine-capped metal particles are actually kinetically rather than thermodynamically stabilized and their interactions are considered to be quite weak [40]. The Ag NPs in organic solvents were also characterized by TEM measurements. As an example, a typical TEM image of the Ag NPs in the chloroform is shown in Figure 3 (TEM images of other Ag organosols are not shown). The TEM image clearly indicates that transfer of Ag NPs to chloroform resulted in their spontaneous organization in two-dimensional (2D) close-packed arrays, and has no influence on the particles morphology and size. This is a consequence of high concentration of silver particles in chloroform and low polydispersity of the system. Based on TEM investigations, the average diameter of the Ag NPs after phase transfer was 6.5 nm (histogram, Figure 3). In addition, the oleylamine-capped Ag NPs were effectively transferred in a similar fashion to other organic solvents, such as hexane, toluene and 1,2-dichloro- 807

18 I. VUKOJE et al.: SURFACE PLASMON RESONANCE OF Ag ORGANOSOLS Hem. ind. 66 (6) (2012) Figure 3. TEM Image of Ag NPs in chloroform and its photograph (a) with corresponding particle size distribution (b). benzene, and the UV Vis spectra of all organosols are presented in Figure 4 (experimental curves). In all cases, the spectra show that the organosols exhibit intensive and narrow absorption bands, which indicates that there was no aggregation of the particles upon the phase transfer. The obtained narrow half-width of the SPR band also implies that prepared Ag NPs are highly uniform. This conclusion can be further supported by calculating the extinction efficiencies of the obtained organosols (Eq. (1)) since Mie theory describes absorption behavior of non-interacting spherical nanoparticles. As can be seen in Figure 4 (red lines), the theoretical spectra of silver organosols show good agreement to those obtained experimentally. On the other hand, the only difference that can be observed between absorption spectra of the Ag nano- Figure 4. Comparison of experimental (solid black line) and theoretical (dashed red line) UV-Vis absorption spectra of Ag NPs in hexane (a), chloroform (b), toluene (c), and 1,2-dichlorobenzene (d). 808

19 I. VUKOJE et al.: SURFACE PLASMON RESONANCE OF Ag ORGANOSOLS Hem. ind. 66 (6) (2012) particles in water and organic solvents is that the peak position of the surface plasmon resonance of Ag organosols is red shifted in comparison to the absorption spectra obtained for Ag hydrosol. This is a consequence of the different nanoparticles environments (i.e., solvents). These particular solvents were chosen because of the distinctive difference in their refractive index. Namely, according to Drude theory [41], dielectric constant of the surrounding medium exclusively determines the plasmon peak position. Therefore, by changing the environment in which nanoparticles are dispersed we can alter their resonant absorption. In other words, by controlling the dielectric constant of the surrounding medium we can predict the wavelength of the SPR band, which can be used for optical applications of these nanoparticles. The process of solvent change from water to nonpolar solvents should lead to a red shift of the plasmon band, due to the higher refractive index of these solvents with respect to that of water. As an example, the SPR band of Ag NPs in chloroform (peak position at 410 nm) is red-shifted for about 21 nm compared to the position of plasmon band of the Ag hydrosol. The details of the maximum peak position and refractive index of used solvents are summarized in Table 1. Table 1. Experimentally obtained and calculated absorption maximum λ max of Ag NPs in different solvents of varying refractive index n λ max / nm Solvent n Experimental Theoretical Water Hexane Chloroform Toluene ,2-Dichlorobenzene Thus far, the theoretical calculation of Ag NPs absorption bands employed experimental values of the dielectric constants of Ag taken from [38]. To analyze the changes in SPR position further we can use Drude equation for the dielectric function of Ag NPs given by: 2 p ω εω ( ) = ε (2) 2 ω + iωγ since this equation describes absorption behavior well in the vicinity of the peak position. In the equation 2, ε represents the contribution of the vacuum and interband electronic transitions, ω p is the bulk plasmon frequency, and: F Γ = Γ 0 + A v (3) r is the size dependent damping frequency, Γ 0 is the bulk damping frequency, v F Fermi velocity, and A is the theory dependent parameter that includes details of scattering process. By introducing Eq. (2) into the Eq. (1) for extinction efficiency, one can obtain following relation for the surface plasmon peak position: 2πc λc = ε + 2εm (4) ω p It can be seen that the λ c values depend solely on ε m, since ε and ω p are assumed constant for a given metal. Note that the refractive index of the medium is directly related to its dielectric constant (ε m = n 2 ). The SPR positions as a function of dielectric constant estimated from the absorption bands of the prepared colloids, as well as those obtained using Mie theory are given in Figure 5a. It can be seen that the λ c values increase with the increase of the ε m as predicted by Eq. (4) (black line in the Figure 5a). The λ c (ε m ) dependence (Eq. (4)) was determined using the following parameters for silver: ω p = 9.5 ev, Γ 0 = 0.1 ev, ε = 5, v F = m/s and A = 1 [42]. One can notice that for given ε m, the λ c values of the experimental and theoretical curves are somewhat higher than those obtained using Eq. (4). The reason for these differences is the underestimated ε, and to demonstrate that we fitted our experimental and theoretical data with the function: F c p ( ε ) = = ( ε' + 2ε ) m 2 λω 2πc in which the proper ε values are determined for the dielectric constant of silver at high frequencies. The results of this analysis are given in Figure 5b and F(ε m ) fits yield ε = 5.7 and ε = 5.3 for the experimental and theoretical data, respectively. It is important to note that theoretical curves were obtained using experimental data for the calculation of dielectric function of Ag NPs. Therefore, for the a priori estimation of Ag NPs peak position in organic solvents by Eq. (4) we suggest that it would be more accurate to use higher value for the dielectric constant of silver at high frequencies (ε = 5.5±0.2). Bearing in mind that ε = 1 + χ, where χ represents the contribution of the interband transitions, one can conclude that inner electrons influence the absorption behavior of silver nanoparticles in organic solvents. However, the difference between the Drude model predicted and experimental peak positions cannot be attributed solely to this effect, since additional processes in silver organosols can occur, such as charge transfer between the nanoparticles surface and solvent molecules, which result in lowering of the plasmon frequency value ω p [43]. m (5) 809

20 I. VUKOJE et al.: SURFACE PLASMON RESONANCE OF Ag ORGANOSOLS Hem. ind. 66 (6) (2012) Figure 5. a) Experimental (circles) and theoretical (squares) values of surface plasmon peak positions and b) fit to F(ε m ) given by Eq. 5 (black bottom line) of Ag NPs as a function of dielectric constant. CONCLUSION A simple method of phase transfer Ag NPs with narrow size distribution from water to nonpolar organic solvents is described. The oleylamine served as both a transfer and a surface capping agent. The so prepared hydrophobic nanoparticles were stable and readily redispersable in a variety of nonpolar solvents. From the UV-Vis absorption spectra and TEM images, we confirmed that there were no shape and size changes of the nanoparticles upon transfer to the organic phase. Theoretical bands, calculated by Mie theory using experimental data for the calculation of dielectric function of Ag NPs, confirmed experimentally gained redshift of SPR band, after the phase transfer due to the increase in dielectric constant of the particles environment. The slight disagreement between experimental and theoretical curves obtained by Drude relation for the SPR peak position is attributed to underestimated value for the dielectric constant of silver at high frequencies. Therefore, after fitting experimental and theoretical data for estimation of Ag NPs peak position in organic solvent we concluded that it would be more appropriate to use higher value for the dielectric constant of silver at high frequencies ε = 5.5±0.2. The curves obtained in this way may be used to characterize the Ag NPs in different media. Furthermore, this method of phase transfer can also be applied for efficient transfer of other metal nanoparticles, such as Au and Cu, to various nonpolar organic solvents. The advantages of such organosols can be found in the applications of metal particles as catalysts and open the possibility for their interaction with polymers or organic dyes to obtain composites for fundamental optical studies and applications. Acknowledgements This work was financially supported by the Ministry of Education, Science and Technological Development of the Republic of Serbia (research project numbers: and 45020). REFERENCES [1] T. Linnert, P. Mulvaney, A. Henglein, Surface chemistry of colloidal silver: Surface plasmon damping by chemisorbed iodide, hydrosulfide (SH-) and phenilthiolate, J. Phys. Chem., B 97 (1993) [2] A. Henglein, D. Meisel, Radiolytic control of the size of colloidal gold nanoparticles, Langmuir 14 (1998) [3] S. Link, M.A. El-Sayed, Size and temperature dependence of the plasmon of colloidal gold nanoparticles, J. Phys. Chem., B 103 (1999) [4] M.M. Alvarez, J.T. Khoury, T.G. Schaaff, M.N. Shafigullin, I. Vezmar, R.L. Whetten, Optical absorption spectra of nanocrystal gold molecules, J. Phys. Chem., B 101 (1997) [5] J.R. Hea, C.M. Knobler, D.V. Leff, Pressure/temperature phase diagrams and superlattices of originally functionalized metal nanocrystal monoleyers: The influence of particle size, size distribution, and surface passivant, J. Phys. Chem., B 101 (1997) [6] S. Underwood, P. Mulvaney, Effect of the solution refractive index on the color of gold colloids, Langmuir 10 (1994)

21 I. VUKOJE et al.: SURFACE PLASMON RESONANCE OF Ag ORGANOSOLS Hem. ind. 66 (6) (2012) [7] U. Kreibig, M. Gartz, A. Hilger, H. Hovel, Mie-plasmon spectroscopy: A tool of surface science. In: E. Pelizzatti, (Ed.), Fine Particles Science and Technology, Kluwer Academic Publishers, Boston, 1996, p [8] A. Henglein, M. Giersig, Formation of colloidal silver nanoparticles: Capping action of citrate, J. Phys. Chem., B 103 (1999) [9] G. Mie, Beitrage zur optic truber medien, Ann. Physik 25 (1908) [10] F. Meriaudeau, T.R. Downey, A. Passian, A. Wig, T.L. Ferrell, Enviroment effects on surface-plasmon spectra in gold-island films potential for sensing applications, Appl. Opt. 37 (1998) [11] F. Meriaudeau, A. Wig, A. Passian, T. Downey, M. Buncick, T.L. Ferrel, Gold island fiber optic sensor for refractive index sensing, Sens. Actuators B 69 (2000) [12] T. Okamoto, I. Yamaguchi, T. Kobayashi, Local plasmon sensor with gold colloid monolayers deposited upon glass substrates, Opt. Lett. 25 (2000) [13] T.K. Sau, C.J. Murphy, Room Temperature, high-yield synthesis of multiple shapes of gold nanoparticles in aqueous solution, J. Am. Chem. Soc. 126 (2004) [14] A.S.K. Hashmi, Gold-catalyzed organic reactions, Chem. Rev. 107 (2007) [15] J.K. Cho, R. Najman, T.W. Dean, O. Ichihara, C. Muller, M.J. Bradley, Captured and cross-linked palladium nanoparticles, J. Am. Chem. Soc. 128 (2006) [16] C.J. Orendorff, T.K. Sau, C.J. Murphy, Shape-dependent plasmon-resonant gold nanoparticles, Small 2 (2006) [17] Z. Merican, T.L. Schiller, C.J. Hawker, P.M. Fredericks, I. Blakey, Self-assembly and encoding of polymer-stabilized gold nanoparticles with surface-enhanced Raman reported molecules, Langmuir 23 (2007) [18] S.P. Mulvaney, M.D. Musick, C.D. Keating, M.J. Natan, Glass-coated, analyte-tagged nanoparticles: A new tagging system based on detection with surface-enhanced Raman scattering, Langmuir 19 (2003) [19] L.M.L. Marzan, I. Lado-Tourino, Reduction and stabilization of silver nanoparticles in ethanol by nonionic surfactants, Langmuir 12 (1996) [20] D.G. Duff, G. Baiker, P.P. Edwards, A new hydrosol of gold clusters. 1. Formation and particle size variation, Langmuir 9 (1993) [21] M.-C. Daniel, D. Astruc, Gold nanoparticles: Assembly, supramolecular chemistry quantum-size-related properties, and applications toward biology, catalysis, and nanotechnology, Chem. Rev. 104 (2004) [22] C.J. Murphy, T.K. Sau, A.M. Gole, C.J. Orendorff, J. Gao, L. Gou, S. E. Hunyadi, T. Li, Anisotropic metal nanoparticles: Synthesis, assembly, and optical applications, J. Phys. Chem., B 109 (2005) [23] H.W. Lu, S.H. Liu, X.L. Wang, X.F. Qian, J. Yin, Z.K. Zhu, Silver nanocrystals by hyperbranched polyurethane-assisted photochemical reduction of Ag, Mat. Chem. Phys. 81 (2003) [24] V.V. Vukovic, J.M. Nedeljkovic, Surface modification of nanometer-scale silver particles by imidazole, Langmuir 9 (1993) [25] A. Henglein, Reduction of Ag(CN) 2 on silver and platinum colloidal nanoparticles, Langmuir 17 (2001) [26] A.C. Templeton, J.J. Pietron, R.W. Murray, P. Mulvaney, Solvent refractive index and core charge influences on the surface plasmon absorbance of alkanethiolate monolayer-protected gold clusters, J. Phys. Chem., B 104 (2000) [27] F.C. Meldrum, N.A. Kotov, J.H. Fendler, Utilization of surfactant-stabilized colloidal silver nanocrystallites in the construction of mono- and multiparticulate Langmuir Blodgett films, Langmuir 10 (1994) [28] M. Sastry, Phase transfer protocols in nanoparticle synthesis, Curr. Sci. 85 (2003) [29] W. Wang, S. Efrima, O. Regev, Directing oleate stabilized nanosized silver colloids into organic phases, Langmuir 14 (1998) [30] A. Kumar, P. Mukherjee, A. Guha, S.D. Adyantaya, A.B. Mandale, R. Kumar, M. Sastry, Amphoterization of colloidal gold particles by capping with valine molecules and their phase transfer from water to toluene by electrostatic coordination with fatty amine molecules, Langmuir 16 (2000) [31] S.H. Chen, H. Yao, K. Kimura, Reversible transference of Au nanoparticles across the water and toluene interface: A Langmuir type adsorption mechanism, Langmuir 17 (2001) [32] O. Tzhayik, P. Sawant, S. Efrima, E. Kovalev, J.T. Klug, Xanthate capping of silver, copper and gold colloids, Langmuir 18 (2002) [33] N. Gaponik, D.V. Talapin, A.L. Rogach, A. Eychmuller, H. Weller, Efficient phase transfer of luminescent thiolcapped nanocrystalas: From Water to nonpolar organic solvents, Nano Lett. 2 (2002) [34] J. Yang, J.Y. Lee, L.X. Chen, H.-P. Too, A Phase-transfer identification of core-shell structures in Ag-Pt nanoparticles, J. Phys. Chem., B 109 (2005) [35] M. Chen, G.W. Diao, C. H. Li, X. M. Zhou, Phase transition of silver nanoparticles from aqueous solution to chloroform with the help of inclusion complexes of p- sulfonated calix[4]arene and alkanethiol molecules and its application in the size sorting of nanoparticles, Nanotechology 18 (2007) [36] V.V. Vodnik, J.M. Nedeljković, Influence of negative charge on the optical properties of a silver sol, J. Serb. Chem. Soc. 65 (2000) [37] F. Hao, P. Nordlander, Efficient dielectric function for FDTD simulation of the optical properties properties of silver and gold nanoparticles, Chem. Phys. Lett. 446 (2007) [38] P.B. Jonhson, R.W. Christy, Optical constant of the noble Metals, Phys. Rev., B 6 (1972) [39] H. Hovel, S. Fritz, A. Hilger, U. Kreibig, M. Vollmer, Width of cluster plasmon resonances: Bulk dielectric functions and chemical interface damping, Phys. Rev., B 48 (1993)

22 I. VUKOJE et al.: SURFACE PLASMON RESONANCE OF Ag ORGANOSOLS Hem. ind. 66 (6) (2012) [40] D.V Leff, L. Brandt, J.R. Heath, Synthesis and characterization of hydrophobic, organically-soluble gold nanocrystals functionalized with primary amines, Langmuir 12 (1996) [41] P. Mulvaney, Surface plasmon spectroscopy of nanosized metal particles, Langmuir 12 (1996) [42] C. Oubre, P. Nordlander, Finite-difference time-domain studies of the optical properties of nanoshell dimers, J. Phys. Chem., B 109 (2005) [43] U. Kreibig, G. Bour, A. Hilger, M. Gartz, Optical properties of cluster-matter: Influences of interfaces, Phys. Stat. Sol, A 175 (1999) IZVOD REZONANCE POVRŠINSKIH PLAZMONA ORGANOSOLA SREBRA: EKSPERIMENTALNA I TEORIJSKA ISPITIVANJA Ivana Vukoje 1, Dušan Božanić 1, Jasna Džunuzović 2, Una Bogdanović 1, Vesna Vodnik 1 1 Univerzitet u Beogradu, Institut za nuklearne nauke Vinča, Beograd, Srbija 2 Univerzitet u Beogradu, Institut za hemiju, tehnologiju i metalurgiju (ICTM) Centar za hemiju, Beograd, Srbija (Naučni rad) Cilj ovog rada je eksperimentalno i teorijsko ispitivanje i upoređivanje promena površinskih plazmona nanočestica srebra u vodi i organskim rastvaračima. Nanočestice srebra prečnika 5±1,5 nm sintetisane su u vodi redukcijom srebro nitrata pomoću natrijum borhidrida. Nakon toga, nanočestice su prebačene u različite organske rastvarače (hloroform, heksan, toluen i 1,2-dihlorbenzen) pomoću oleilamina. Upotrebljeni rastvarači izabrani su na osnovu razlike u indeksima prelamanja. Na osnovu UV Vis apsorpcione spektrofotometrije i transmisione elektronske mikroskopije (TEM) potvrđeno je da nema promena u obliku i veličini čestica nakon transfera u organske rastvarače. Apsorpcioni spektri nanočestica srebra u različitim rastvaračima pokazuju promene u položaju površinskih plazmona na osnovu razlike u dielektričnim karakteristikama upotrebljenih rastvarača. Za analizu ovih promena eksperimentalni apsorpcioni spektri su modelovani pomoću Mie teorije za male sferne čestice. Eksperimentalne i teorijske vrednosti položaja površinskih plazmona nanočestica srebra upoređivane su sa Drude teorijom i diskutovana su ograničenja primene same teorije na ponašanja nanočestica srebra u organskim rastvaračima. Ključne reči: Nanočestice srebra Organosol Mie teorija Drude model 812

23 Ispitivanje morfologije i površinskih svojstava umreženih poli(uretan-estar-siloksana) Jasna V. Džunuzović 1, Marija V. Pergal 1, Vesna V. Vodnik 2, Milena Špírková 3, Rafał Poręba 3, Slobodan Jovanović 4 1 Univerzitet u Beogradu, Institut za hemiju, tehnologiju i metalurgiju (IHTM) Centar za hemiju, Beograd, Srbija 2 Univerzitet u Beogradu, Institut za nuklearne nauke Vinča, Beograd, Srbija 3 Institute of Macromolecular Chemistry AS CR, v.v.i., Nanostructured Polymers and Composites Department, Praha, Czech Republic 4 Univerzitet u Beogradu, Tehnološko metalurški fakultet, Beograd, Srbija Izvod U ovom radu sintetisane su dve serije umreženih poli(uretan-estar-siloksana) (PUS) korišćenjem α,ω-dihidroksi-(etilenoksid-poli(dimetilsiloksan)-etilenoksida) (EO-PDMS-EO), 4,4'- -diizocijanatodifenilmetana i dva Boltorn hiperrazgranata poliestra (HRP) druge i treće pseudo generacije kao umreživača. Svaka serija se sastoji od uzoraka koji imaju različiti udeo mekog segmenta (EO-PDMS-EO). Uticaj funkcionalnosti korišćenog HRP i udela mekog segmenta na morfologiju i površinska svojstva PUS ispitan je primenom FTIR spektroskopije, SAXS, AFM i SEM analize i merenjem količine apsorbovane vode. Pokazano je da sa smanjenjem funkcionalnosti umreživača i udela EO-PDMS-EO dolazi do povećanja stepena mikrofaznog razdvajanja kod sintetisanih PUS. Pored toga, uzorci PUS sa većim udelom mekog segmenta i sintetisani primenom HRP niže funkcionalnosti pokazuju bolju otpornost na vodu. NAUČNI RAD UDK : :544.7:543.4 Hem. Ind. 66 (6) (2012) doi: /HEMIND D Ključne reči: hiperrazgranati poliestar; poli(dimetilsiloksan); umreženi poliuretani; morfologija; površinska svojstva. Dostupno na Internetu sa adrese časopisa: Poliuretanske mreže su poznate po raznovrsnoj primeni, posebno u industriji premaza, zbog njihovih karakterističnih svojstava poput visokog sjaja, adhezivnosti, tvrdoće, fleksibilnosti, dobre otpornosti na rastvarače, vodu, abraziju, udar, itd. [1]. Svojstva umreženih poliuretana (PU) zavise od hemijskog sastava svih komponenata, molarne mase i polidisperznosti mekih segmenata i prirode umreživača. Zato je s ciljem dobijanja željenih svojstava poliuretanskih mreža, a radi njihove tačno određene primene, neophodno izabrati dobru kombinaciju polaznih reaktanata. Poli(dimetilsiloksani) (PDMS) pokazuju zanimljiva fizička i hemijska svojstva kao što su dobra termička stabilnost, dobra otpornost na oksidaciona sredstva i UV zračenje, nizak površinski napon i dobra otpornost na vlagu, veoma niska temperatura ostakljivanja, fleksibilnost, dobra propustljivost gasova i biokompatibilnost, ali istovremeno i vrlo loša mehanička svojstva i lošu kompatibilnost sa većinom organskih polimera [2]. Veoma efikasan način za poboljšanje mehaničkih svojstava ovog polimera je ugradnja PDMS u strukturu blok-kopolimera ili polimernih mreža. S obzirom na to da PU imaju dobra mehanička svojstva ali lošu termičku Prepiska: J.V. Džunuzović, IHTM Centar za hemiju, Univerzitet u Beogradu, Studentski trg 12 16, Beograd, Srbija. E-pošta: jasnav2002@yahoo.com Rad primljen: 4. oktobar, 2012 Rad prihvaćen: 9. oktober, 2012 stabilnost, kontrolisanom sintezom polimernih mreža polazeći od PDMS makrodiola i uretanskih komponenti mogu se poboljšati termička i površinska svojstva poliuretana uz istovremeno zadržavanje njihovih dobrih mehaničkih svojstava [3 6]. Istovremeno, prisustvo mekih PDMS segmenata dovodi do većeg stepena mikrofaznog razdvajanja poliuretana, usled termodinamičke nekompatibilnosti između tvrdih i mekih segmenata [7]. Sudeći po literaturnim podacima, hiperrazgranati poliestri (HRP) pokazali su se kao odličan izbor pri sintezi različitih polimernih mreža zbog prisustva velikog broja krajnjih funkcionalnih grupa, globularnog oblika makromolekula i jedinstvenih fizičkih i hemijskih svojstava, kao što su odsustvo prepletaja, mala viskoznost u rastvoru i rastopu, dobra rastvorljivost u velikom broju različitih rastvarača, itd. [8 17]. Pored toga, sinteza hiperrazgranatih polimera je prilično jednostavna, dok se odgovarajućom modifikacijom krajnjih funkcionalnih grupa njihova primena može lako prilagoditi željenoj nameni. U našim prethodnim radovima je pokazano da se kombinacija komercijalno dostupnih Boltorn hidroksifunkcionalnih HRP, makrodiola na bazi PDMS i odgovarajućeg diizocijanata može iskoristiti za sintezu umreženih PU dobrih termičkih i mehaničkih svojstava [18-23]. Naime, veliki broj krajnjih hidroksilnih grupa prisutnih kod HRP omogućava brzo umrežavanje i formiranje mreže koja pokazuje dobra mehanička svojstva, 813

24 J.V. DŽUNUZOVIĆ i sar.: UMREŽENI POLI(URETAN-ESTAR-SILOKSANI) Hem. ind. 66 (6) (2012) kao i dobru otpornost na dejstvo hemikalija, dok prisustvo PDMS doprinosi dobrim termičkim i površinskim svojstvima, kao i elastičnosti tako jako umreženih materijala. Cilj ovog rada je ispitivanje uticaja odnosa mekih i tvrdih segmenata i funkcionalnosti umreživača na stvaranje vodoničnih veza, morfologiju, stepen mikrofaznog razdvajanja i sposobnost apsorpcije vode dve serije umreženih poli(uretan-estar-siloksana), sintetisanih polazeći od α,ω-dihidroksi-(etilenoksid-poli(dimetilsiloksan)-etilenoksida) (EO-PDMS-EO), 4,4'-diizocijanatodifenilmetana (MDI) i Boltorn HRP druge i treće pseudo generacije. EKSPERIMENTALNI DEO Reaktanti α,ω-dihidroksi-(etilenoksid-poli(dimetilsiloksan)-etilenoksid) (ABCR, Nemačka), molarne mase M n = 1200 g/mol ( 1 H-NMR), do sinteze je čuvan iznad molekulskih sita (0,4 nm) [24]. Komercijalno dostupni (Perstorp Specialty Chemicals AB, Švedska) Boltorn hidroksi-funkcionalni hiperrazgranati alifatski poliestri druge (BH-20) i treće (BH-30) pseudo generacije su korišćeni kao agensi za umrežavanje i sušeni dva dana u vakuum sušnici pri temperaturi od 50 C pre sinteze poliuretana. Hiperrazgranati poliestri su sintetisani polazeći od 2,2- -bis(hidroksimetil)propionske kiseline kao AB 2 -monomera i etoksilovanog pentaeritritola kao tetrafunkcionalnog molekula jezgra primenom pseudo one-step procedure [25]. Korišćenjem vrednosti molarnih masa dobijenih primenom osmometrije napona pare (M n (BH- -20) = 1340 g/mol, M n (BH-30) = 3080 g/mol) i vrednosti hidroksilnih brojeva, HB, određenih metodom titracije (HB(BH-20) = 501,1 mg KOH/g, HB(BH-30) = 474,1 mg KOH/g), izračunate su vrednosti funkcionalnosti, f, korišćenih hiperrazgranatih poliestara (f(bh-20) = 12, f(bh-30) = 26) [26]. 4,4'-Diizocijanatodifenilmetan (čistoća > 98%, Aldrich, Nemačka) korišćen je bez prečišćavanja. Sadržaj NCO grupa u MDI-u je proveren određivanjem izocijanatnog broja (33,6 mas.%) [27]. Katalizator kalaj-oktoat, (Sn(Oct) 2 ; Aldrich) korišćen je kao razblaženi rastvor u anhidrovanom N-metil-2-pirolidonu (NMP). Pre korišćenja NMP (Acros) destilovan je pod sniženim pritiskom i čuvan iznad molekulskih sita (0,4 nm). Tetrahidrofuran (THF, J.T. Baker) sušen je iznad litijum-aluminijum-hidrida i destilovan pre korišćenja. Sinteza umreženih poli(uretan-estar-siloksana) Sintetisane su dve serije umreženih poli(uretan-estar-siloksana) (PUS) na bazi Boltorn HRP, EO-PDMS-EO i MDI, korišćenjem dvostepene reakcije polimerizacije u rastvoru. Kao rastvarač korišćena je smeša NMP/THF. Prva serija uzoraka (PUS2-60, PUS2-40, PUS2-30, PUS i PUS2-15) sintetisana je primenom BH-20 kao umreživača, dok je druga serija (PUS3-40, PUS3-30, PUS3-20 i PUS3-15) sintetisana korišćenjem BH-30 kao umreživača. Svaka serija se sastoji od uzoraka različitog masenog udela mekih segmenata (EO-PDMS-EO), što je označeno poslednjim brojem datim u nazivu uzoraka. Ukupan molski odnos NCO i OH grupa je pri sintezi svih uzoraka bio konstantan (NCO/OH =1,05) [27]. Detaljan opis sinteze ovih umreženih poliuretana je dat u prethodnim radovima [20-23]. Ukratko, u četvorogrli balon opremljen mehaničkom mešalicom, nastavkom za uvođenje argona, levkom za ukapavanje i kondenzatorom uneta je odgovarajuća količina EO-PDMS-EO, rastvorenog u smeši rastvarača NMP/THF, i MDI. Balon je pomoću uljanog kupatila zagrevan do temperature od 40 C u atmosferi argona. Nakon postizanja željene temperature dodata je odgovarajuća količina katalizatora rastvorenog u NMP-u (0,15 mol% u odnosu na količinu EO-PDMS-EO) [27]. Reakciona smeša je kontinualno mešana pri temperaturi od 40 C i nakon 30 min dobijen je NCO-terminirani pretpolimer [27]. Sadržaj NCO grupa je praćen primenom standardne dibutilamin titracije [28]. U drugom stupnju reakcije dodata je pomoću levka za ukapavanje odgovarajuća količina HRP rastvorenog u NMP-u i mešanje je nastavljeno pri istoj temperaturi još 10 min. Zatim je reakciona smeša prebačena u Petri šolje, podmazane silikonskim uljem, i umrežavanje je nastavljeno u sušnici 45 h pri temperaturi od 80 C 1 h pri temperaturi od 110 C i na kraju 10 h pod vakuumom pri temperaturi od 50 C. Sintetisani umreženi PUS su dobijeni u obliku filmova mrke boje, prosečne debljine oko 1,5±0,2 mm. Svi uzorci su držani u eksikatoru pri sobnoj temperaturi. Hemijska struktura sintetisanih uzoraka PUS je ispitana i potvrđena korišćenjem FTIR spektroskopije [19 23]. Karakterizacija FTIR spektri sintetisanih PUS su dobijeni korišćenjem ATR Nicolet 380 FTIR spektrometra. Dekonvolucija C=O regiona prisutnih u FTIR spektrima sintetisanih umreženih poliuretana je izvedena korišćenjem Gauss-ove funkcije (OriginPro 8). Rasipanje rendgenskih zraka na malim uglovima (SAXS) izvedeno je korišćenjem kamere Molmet/Rigaku (3 pinhole) povezane sa višeslojnom asferičnom optikom (Osmic Confocal Max-Flux), koja emitovanu monohromatsku svetlost usmerava sa mikrofokusirajuće cevi za X-zrake (Bede), pri naponu od 45 kv i jačini struje od 0,66 ma (30 W). Kamera je opremljena sa 2D detektorom ispunjenim gasom čiji je prečnik aktivne površine 0,2 m (Gabriel design). Mikroskopija atomskih sila (AFM) korišćena je za ispitivanje površinske topografije površine preloma uzoraka, dobijenih nakon preloma prethodno zamrznutih sintetisanih PUS filmova pri temperaturi tečnog azota. Korišćen je komercijalni AFM (MultiMode Digital Instruments NanoScopeTM Dimension IIIa), opremljen sa SSS-NCL sondom i Super Sharp Silicon TM SPM-sen- 814

25 J.V. DŽUNUZOVIĆ i sar.: UMREŽENI POLI(URETAN-ESTAR-SILOKSANI) Hem. ind. 66 (6) (2012) zorom (NanoSensors TM Switzerland; konstanta opruge 35 N/m, rezonantne frekvencije 170 khz). Merenja su izvedena pri sobnim uslovima, korišćenjem tzv. tapping mode AFM tehnike. Slike površina su snimljene u veličinama od (0,3 0,3) do (10 10) μm 2. Skenirajuća elektronska mikroskopija (SEM) površine sintetisanih uzoraka je izvedena primenom JEOL JSM-6610 instrumenta. Apsorpcija vode sintetisanih uzoraka PUS pri sobnoj temperaturi je merena potapanjem uzoraka dimenzija 10,0 mm 10,0 mm 1,5 mm±0,2 mm u destilovanu vodu, nakon 48 h. Maseni udeo apsorbovane vode, AV, izračunat je korišćenjem jednačine: w w = (1) 0 AV 100 w0 gde su w i w 0 mase mokrog i suvog uzorka, redom. Srednja vrednost tri zasebna merenja za svaki uzorak PUS je uzeta kao krajnja vrednost. Proračun standardne greške je izveden primenom Origin Pro-8 softvera. REZULTATI I DISKUSIJA U ovom radu sintetisane su dve serije umreženih poli(uretan-estar-siloksana) primenom dvostepene reakcije polimerizacije u rastvoru polazeći od EO-PDMS- -EO, MDI i Boltorn HRP druge i treće pseudo generacije. Sinteza je izvedena u smeši rasvarača kako bi se poboljšala kompatibilnost između nepolarnog mekog (EO-PDMS-EO) i polarnih tvrdih (MDI-Boltorn HRP) segmenata, pošto je prethodno pokazano da sintezom u rastopu dolazi do stvaranja heterogene mreže i pojave makroskopske fazne separacije kod ovih poliuretana [18]. Na slici 1 dat je šematski prikaz drugog stupnja reakcije, tj. reakcije između NCO-terminiranog pretpolimera, nastalog u prvom stupnju reakcije, i hiperrazgranatog poliestra. Korišćenjem FTIR spektroskopije ispitana je hemijska struktura sintetisanih PUS kao i prisustvo vodoničnih veza. Ilustracije radi, na slici 2 prikazan je FTIR spektar uzorka PUS3-30. Za sve ostale sintetisane uzorke položaj karakterističnih traka u FTIR spektrima bio je sličan. Traka koja je uočena na talasnom broju oko 792 cm 1 odnosi se na Si-CH 3 vezu, dok trake koje su se preklopile na talasnim brojevima oko 1014 i 1080 cm 1 pripisane su Si O Si i C O C grupama. Pored ovih, uočene su i sledeće trake u FTIR spektrima sintetisanih PUS: na talasnim brojevima oko 1257 i 1535 cm 1 (amidne II i amidne III vibracije, redom), 1598 i 1411 cm 1 (aromatske C=C), 2962, 2904 i 2877 cm 1 (simetrične i asimetrične vibracije CH 2 i CH 3 grupa) i 3306 cm 1 (vibracije vodonično vezanih NH grupa). U oblasti cm 1 prisutne su trake koje odgovaraju vibracijama karbonilne C=O grupe. Trake koje odgovaraju NCO (2270 cm 1 ) i OH (3300 cm 1 ) grupama nisu uočene u FTIR spektrima svih sintetisanih PUS, što ukazuje da su u toku reakcije sinteze ove grupe potpuno proreagovale. Radi ispitivanja uticaja udela EO-PDMS-EO na nastajanje vodoničnih veza u strukturi sintetisanih uzoraka, izvedena je dekonvolucija oblasti koja odgovara vibracijama C=O grupa, korišćenjem Gauss-ove funkcije. Ilustracije radi na slici 3 je prikazana dekonvolucija oblasti koja odgovara karbonilnim grupama za uzorak PUS3-40, dok su u tabeli 1 dati rezultati dekonvolucije za seriju PUS3. Kod svih uzoraka iz obe PUS serije je nakon dekonvolucije uočeno pet traka u C=O oblasti koje odgovaraju: vodonično vezanim estarskim C=O grupama (1645 cm 1, pik 1), vodonično vezanim uretanskim C=O grupama prisutnim u uređenim tvrdim domenima (1680 cm 1, pik 2), neuređenim vodonično vezanim uretanskim C=O grupama (1710 cm 1, pik 3), slobodnim estarskim C=O grupama (1725 cm 1, pik 4) i slobodnim uretanskim C=O grupama (1735 cm 1, pik 5). Rezultati dobijeni za uzorke PUS2 serije pokazali su da sa smanjenjem udela EO-PDMS-EO raste udeo slobodnih uretanskih C=O grupa, vodonično vezanih uretanskih grupa (uređeni tvrdi domeni) i udeo nastalih vodoničnih veza Slika 1. Šematski prikaz sinteze umreženih poli(uretan-estar-siloksana). Figure 1. Schematic representation of the synthesis of poly(urethane-ester-siloxane)s. 815

26 J.V. DŽUNUZOVIĆ i sar.: UMREŽENI POLI(URETAN-ESTAR-SILOKSANI) Hem. ind. 66 (6) (2012) Slika 2. FTIR spektar uzorka PUS3-30. Figure 2. FTIR spectrum of the PUS3-30. Slika 3. Dekonvolucija oblasti karbonilnih grupa za uzorak PUS3-40. Figure 3. Deconvolution of the FTIR absorbance region of the carbonyl groups of PUS3-40. Tabela 1. Rezultati dekonvolucije C=O regiona (%) FTIR spektara sintetisanih PUS Table 1. Deconvolution results of the C=O stretching regions (%) from the FTIR spectra of PUS Uzorak Oblast 1, estarske H-vezane CO grupe Oblast 2, uređene uretanske H-vezane CO grupe Oblast 3, neuređene uretanske H-vezane CO grupe Oblast 4, slobodne estarske CO grupe Oblast 5, slobodne uretanske CO grupe PUS ,7 14,2 31,3 22,3 15,5 PUS ,7 17,2 25,9 22,2 15,0 PUS ,5 38,4 10,1 16,8 13,2 PUS ,2 40,1 8,7 15,9 12,1 između uretanskih NH grupa i estarskih karbonilnih grupa (iz hiperrazgranatog poliestra), dok udeo slobodnih estarskih C=O grupa, kao i udeo nastalih (neuređenih) vodoničnih veza između NH uretanskih grupa i kiseonika iz etarske grupe koja pripada EO-PDMS-EO opada [23]. Isti trend je dobijen i za uzorke PUS3 serije, osim što u ovom slučaju udeo slobodnih uretanskih C=O grupa blago opada sa smanjenjem udela EO- 816

27 J.V. DŽUNUZOVIĆ i sar.: UMREŽENI POLI(URETAN-ESTAR-SILOKSANI) Hem. ind. 66 (6) (2012) -PDMS-EO (tabela 1). Pojava vodoničnog vezivanja između mekih i tvrdih segmenata dovodi do nastajanja određenog stepena faznog mešanja kod sintetisanih uzoraka. Dobijeni rezultati dekonvolucije C=O oblasti prisutne u FTIR spektrima sintetisanih PUS ukazuju da kod uzoraka sa manijm udelom EO-PDMS-EO postoji veća tendencija ka nastajanju vodoničnih veza između tvrdih segmenata, što utiče na povećanje uređenosti tvrdih domena i samim tim do pojave većeg stepena mikrofaznog razdvajanja. Uticaj broja pseudo generacije korišćenog hiperrazgranatog poliestra na morfologiju, odnosno na stepen mikrofaznog razdvajanja sintetisanih uzoraka PUS je ispitan primenom SAXS analize i dobijeni rezultati su prikazani na slici 4. Prisustvo izraženog pika na q = = 0,03 0,06 nm 1 za oba uzorka potvrđuje postojanje dvofazne mikrostrukture, tj. mikrofaznog razdvajanja kod ovih uzoraka. Rastojanje između tvrdih domena, odnosno karakteristična dužina D je izračunata korišćenjem položaja pikova i Bragg-ovog zakona, D = 2π/q, gde je q vektor rasipanja. Za uzorak PUS2-30 izračunata vrednost rastojanja između tvrdih domena iznosi 10,5 nm, dok je za uzorak PUS3-30 dobijeno D = 13,1 nm. Dobijeni rezultati pokazuju da sa povećanjem broja pseudo generacije korišćenog hiperrazgranatog umreživača rastojanje između tvrdih domena raste. Razlog za to leži u činjenici da BH-30 ima nešto duže grane od hiperrazgranatog poliestra druge pseudo generacije, što dovodi do povećanja rastojanja između domena tvrdih segmenata. Analizom SAXS krivih pri većim vrednostima vektora rasipanja (q > q max, gde q max predstavlja vrednost q za pik) uočeno je da se intenzitet rasipanja, I(q), menja na sledeći način: I(q) q n. Za uzorak PUS2-30 intenzitet rasipanja prati Porod-ov zakon, odnosno I(q) q -4, što je karakteristično za dobro mikrofazno razdvojene strukture sa jasnom granicom između faza [22,29,30]. Sa druge strane, za uzorak PUS3-30 vrednost eksponenta n je 2,5, odnosno za ovaj uzorak I(q) opada sporije nego q 4 pri većim vrednostima q, što se vidi na uvećanom prikazu rezultata datom na slici 4 [30]. Velankar i saradnici su pokazali da poliuretani kod kojih je n < 4 imaju niže vrednosti stepena mikrofaznog radzvajanja [30]. Rezultati dobijeni SAXS analizom ukazuju da sa povećanjem broja pseudo generacije hiperrazgranatog poliestra dolazi do povećanja neuređenosti tvrdih domena usled povećanja rastojanja između njih, pa samim tim i do bolje kompatibilnosti između mekih EO-PDMS-EO i tvrdih MDI-Boltorn HRP segmenata. Topografija površine preloma dva uzorka PUS kod kojih je udeo mekog segmenta 20 mas.% je ispitana primenom AFM. Dobijeni rezultati su prikazani na slikama 5 (3D slike) i 6 (2D fazne slike). Kod oba ispitana uzorka uočena je pojava agregata mikrometarske veličine (slika 5). Tamniji regioni odgovaraju mekim, dok svetliji regioni predstavljaju tvrde mikrodomene. Na faznim slikama prikazanim na slici 6 svetli i tamni regioni su međusobno razdvojeni usled mikrofaznog razdvajanja između njih. Veličina i rastojanje između tvrdih domena raste sa porastom broja pseudo generacije hiperrazranatog poliestra. Pored toga, rezultati prikazani Slika 4. SAXS profili uzoraka PUS2-30 i PUS3-30. Manja slika pokazuje uvećani prikaz rezultata pri visokim vrednostima q. Figure 4. SAXS profiles of PUS2-30 and PUS3-30. The inset shows a magnified view of the data at high q values. 817

28 J.V. DŽUNUZOVIĆ i sar.: UMREŽENI POLI(URETAN-ESTAR-SILOKSANI) Hem. ind. 66 (6) (2012) Slika 5. 3D AFM slike uzoraka PUS2-20 i PUS3-20. Figure 5. 3D AFM images of the samples PUS2-20 and PUS3-20. Slika 6. 2D AFM slike uzoraka PUS2-20 i PUS3-20. Figure 6. 2D AFM images of the samples PUS2-20 and PUS3-20. na slici 6 pokazuju da je razdvajanje mikrodomena izraženije kod uzorka PUS2-20, tj. da su u ovom slučaju tvrdi domeni bolje dispergovani u matrici nego kod uzorka sintetisanog korišćenjem BH-30 kao umreživača, što je u saglasnosti sa rezultatima dobijenim korišćenjem SAXS eksperimenata [16]. Morfologija površine sintetisanih uzoraka PUS ispitana je analizom snimljenih SEM mikrografa i na slici 7 prikazani su izabrani rezultati. Zbog velike razlike u vrednosti parametra rastvorljivosti između EO-PDMS- EO i uretanskih komponenata, kao i niske površinske energije, EO-PDMS-EO migrira na površinu sintetisanih uzoraka, čak i kada je prisutan u malim količinama, što je i glavni razlog za pojavu mikrofaznog razdvajanja kod ovih PUS [5,31]. Na površini uzoraka PUS2-60 i PUS3-40 mogu se uočiti mikrodomeni nepravilnog oblika sastavljeni od vodonično vezanih tvrdih segmenata, koji su dispergovani u meku EO-PDMS-EO matricu. Sa druge strane, EO-PDMS-EO formira sferne agregate različitog prečnika (0,5 10 µm) na površini PUS2-40. Morfologija površine uzoraka koji sadrže 15 mas.% EO-PDMS-EO je veoma slična, što ukazuje na to da formiranje mikrodomena EO-PDMS-EO na površini sintetisanih PUS zavisi od masenog udela EO-PDMS-EO segmenata. Površinska svojstva sintetisanih uzoraka PUS ispitana su takođe merenjem količine apsorbovane vode nakon 48 h. Sa slike 8, na kojoj je prikazana zavisnost količine apsorbovane vode od masenog udela EO- -PDMS-EO segmenata, vidi se da smanjenjem broja pseudo generacije hiperrazgranatog poliestra i povećanjem udela EO-PDMS-EO dolazi do smanjenja količine apsorbovane vode. Ovakvo ponašanje je očekivano zbog hidrofobnog karaktera EO-PDMS-EO i njegove površinske aktivnosti, kao i sposobnosti PDMS da migrira na površinu sintetisanih PUS [31]. Dobijeni rezultati ukazuju da uzorci PUS sa većim udelom EO- -PDMS-EO imaju bolju otpornost na vodu, odnosno hidrofobniji su. 818

29 J.V. DŽUNUZOVIĆ i sar.: UMREŽENI POLI(URETAN-ESTAR-SILOKSANI) Hem. ind. 66 (6) (2012) Slika 7. SEM mikrografi površina izabranih sintetisanih uzoraka PUS. Figure 7. SEM micrographs of surfaces of selected synthesized PUS. Slika 8. Količina apsorbovane vode u zavisnosti od udela EO-PDMS-EO sintetisanih uzoraka PUS. Figure 8. The amount of the absorbed water versus EO-PDMS-EO content for the synthesized PUS. ZAKLJUČAK Dve serije umreženih poli(uretan-estar-siloksana) sintetisane su u ovom radu polazeći od α,ω-dihidroksi- (etilenoksid-poli(dimetilsiloksan)-etilenoksida) kao mekog segmenta i 4,4'-diizocijanatodifenilmetana i Boltorn hiperrazgranatih poliestara druge i treće pseudo generacije kao komponente tvrdih segmenata. Serija PUS2 je sintetisana korišćenjem BH-20, a serija PUS3 primenom BH-30 kao umreživača. Ispitan je uticaj funkcionalnosti korišćenog HRP i udela mekog segmenta na morfologiju i površinska svojstva PUS. Rezultati dobijeni primenom FTIR spektroskopije, SAXS i AFM analize pokazali su da sa smanjenjem funkcionalnosti umreživača i udela EO-PDMS-EO dolazi do povećanja stepena mikrofaznog razdvajanja kod sintetisanih PUS. SEM analiza i merenje količine apsorbovane vode ukazali su da zbog hidrofobnog karaktera EO-PDMS-EO i njegove sposobnosti da migrira na površinu sintetisanih PUS, uzorci sa većim udelom mekog segmenta i sintetisani primenom HRP niže funkcionalnosti pokazuju bolju otpornost na vodu. Dobijeni rezultati pokazuju da se svojstva sintetisanih PUS mogu jednostavno prilagoditi željenoj primeni promenom 819

30 J.V. DŽUNUZOVIĆ i sar.: UMREŽENI POLI(URETAN-ESTAR-SILOKSANI) Hem. ind. 66 (6) (2012) funkcionalnosti umreživača ili masenog udela EO- -PDMS-EO segmenta. Zahvalnica Ovaj rad je finansiralo Ministarstvo prosvete, nauke i tehnološkog razvoja Republike Srbije (projekat br ) i Czech Science Foundation (GACR, Projekat br. P108/10/0195). Gospođa Hana Šandová (IMC) uradila je SAXS eksperimente, na čemu joj se autori zahvaljuju. LITERATURA [1] J. Dodge, in: M.E. Rogers, T.E. Long (Eds.), Synthesis Methods in Step-Growth Polymers, Wiley, New Jersey, 2003, pp [2] C.Z. Yang, C. Li, S.L.Cooper, Synthesis and characterization of polydimethylsiloxane polyurea-urethanes and related zwitterionomers, J. Polym. Sci., Part B: Polym. Phys. 29 (1991) [3] M. Shibayama, M. Suetsugu, S. Sakurai, T. Yamamoto, S. Nomura, Structure characterization of polyurethanes containing poly(dimethylsiloxane), Macromolecules 24 (1991) [4] A. Stanciu, V. Bulacovschi, V. Condratov, C. Fadei, A. Stoleriu, S. Balint, Thermal stability and the tensile properties of some segmented poly(ester-siloxane)urethanes, Polym. Degrad. Stab. 64 (1999) [5] P. Majumdar, D.C. Webster, Preparation of siloxaneurethane coatings having spontaneously formed stable biphasic microtopograpical surfaces, Macromolecules 38 (2005) [6] A. Stanciu, V. Bulacovschi, S. Oprea, S. Vlad, Thermal and mechanical behaviour of some new crosslinked poly(ester-siloxane)urethanes, Polym. Degrad. Stab. 72 (2001) [7] A. Stanciu, A. Airinei, D. Timpua, A. Ioanid, C. Ioan, V. Bulacovschi, Polyurethane/polydimethylsiloxane segmented copolymers, Eur. Polym. J. 35 (1999) [8] R.S. Mishra, A.S. Khanna, Formulation and performance evaluation of hydroxyl terminated hyperbranched polyesters based poly (ester urethane urea) coatings on mild steel, Progr. Org. Coat. 72 (2011) [9] E. Džunuzović, S. Tasić, B. Božić, K. Jeremić, B. Dunjić, Photoreactive hyperbranched urethane acrylates modified with a branched saturated fatty acid, React. Funct. Polym. 66 (2006) [10] D. Foix, X. Ramis, F. Ferrando, A. Serra, Improvement of epoxy thermosets using a thiol-ene based polyester hyperbranched polymer as modifier, Polym. Int. 61 (2012) [11] A. Tomuta, F. Ferrando, À. Serra, X. Ramis, New aromatic aliphatic hyperbranched polyesters with vinylic end groups of different length as modifiers of epoxy/ /anhydride thermosets, React. Funct. Polym. 72 (2012) [12] E.S. Džunuzović, S.V. Tasić, B.R. Božić, J.V. Džunuzović, B.M. Dunjić, K.B. Jeremić, Mechanical and thermal properties of UV cured mixtures of linear and hyperbranched urethane acrylates, Progr. Org. Coat. 74 (2012) [13] J.M. Morancho, A. Cadenato, X. Ramis, X. Fernández- Francos, M. Flores. J.M. Salla, Effect of a hyperbranched polymer over the thermal curing and the photocuring of an epoxy resin, J. Therm. Anal. Calorim. 105 (2011) [14] A. Asif, W. Shi, X. Shen, K. Nie, Physical and thermal properties of UV curable waterborne polyurethane dispersions incorporating hyperbranched aliphatic polyester of varying generation number, Polymer 46 (2005) [15] Q. Cao, P. Liu, Structure and mechanical properties of shape memory polyurethane based on hyperbranched polyesters, Polym. Bull. 57 (2006) [16] P.K. Maji, A.K. Bhowmick, Influence of number of functional groups of hyperbranched polyol on cure kinetics and physical properties of polyurethanes, J. Polym. Sci., Part A: Polym. Chem. 47 (2009) [17] P. Czech, L. Okrasa, J. Ulanski, G. Boiteux, F. Mechin, P. Cassagnau, Studies of the molecular dynamics in polyurethane networks with hyperbranched crosslinkers of different coordination numbers, J. Appl. Polym. Sci. 105 (2007) [18] J. Vuković, M. Pergal, S. Jovanović, V. Vodnik, Crosslinked polyurethanes based on hyperbranched polymers, Hem. Ind. 62 (2008) [19] M.V. Pergal, J.V. Džunuzović, M. Kićanović, V. Vodnik, M.M. Pergal, S. Jovanović, Thermal properties of poly- (urethane-ester-siloxane)s based on hyperbranched polyester, Russ. J. Phys. Chem. A 85 (2011) [20] J.V. Džunuzović, M.V. Pergal, S. Jovanović, V.V. Vodnik, Synthesis and swelling behavior of polyurethane networks based on hyperbranched polymer, Hem. Ind. 65 (2011) [21] M.V. Pergal, J.V. Džunuzović, S. Ostojić, M.M. Pergal, A. Radulović, S. Jovanović, Poly(urethane-siloxane)s based on hyperbranched polyester as crosslinking agent: synthesis and characterization, J. Serb. Chem. Soc. 77 (2012) [22] J.V. Džunuzović, M.V. Pergal, R. Poręba, S. Ostojić, N. Lazić, M. Špírková, S. Jovanović, Studies of the thermal and mechanical properties of poly(urethane siloxane)s cross-linked by hyperbranched polyesters, Ind. Eng. Chem. Res. 51 (2012) [23] J.V. Džunuzović, M.V. Pergal, R. Poręba, V.V. Vodnik, B.R. Simonović, M. Špírková, S. Jovanović, Analysis of dynamic mechanical, thermal and surface properties of poly(urethane-ester-siloxane) networks based on hyperbranched polyester, J. Non-Cryst. Solids 368 (2012) [24] M.V. Vuckovic, V.V. Antic, M.N. Govedarica, J. Djonlagic, Synthesis and characterization of copolymers based on poly(butylene terephthalate) and ethylene oxide-poly- (dimethylsiloxane)-ethylene oxide, J. Appl. Polym. Sci. 115 (2010) [25] E. Malmström, M. Johansson, A. Hult, Hyperbranched aliphatic polyesters, Macromolecules 28 (1995)

31 J.V. DŽUNUZOVIĆ i sar.: UMREŽENI POLI(URETAN-ESTAR-SILOKSANI) Hem. ind. 66 (6) (2012) [26] J. Vuković, Synthesis and characterization of aliphatic hyperbranched polyesters, Doktorska teza, Univerzitet u Osnabrück-u, Nemačka, 2006, [27] M.V. Pergal, V.V. Antic, M.N. Govedarica, D. Gođevac, S. Ostojić, J. Djonlagic, Synthesis and characterization of novel urethane-siloxane copolymers with a high content of PCL-PDMS-PCL segments, J. Appl. Polym. Sci. 122 (2011) [28] Å. Marand, J. Dahlin, D. Karlsson, G. Skarping, M. Dalene, Determination of technical grade isocyanates used in the production of polyurethane plastics, J. Environ. Monit. 6 (2004) [29] I. Krakovský, Z. Bubeníková, H. Urakawa, K. Kajiwara, Inhomogeneous structure of polyurethane networks based on poly(butadiene)diol: 1. The effect of the poly- (butadiene)diol content, Polymer 38 (1997) [30] S. Velankar, S.L. Cooper, Microphase separation and rheological properties of polyurethane melts. 2. Effect of block incompatibility on the microstructure, Macromolecules 33 (2000) [31] H.D. Hwang, H.J. Kim, Enhanced thermal and surface properties of waterborne UV-curable polycarbonatebased polyurethane (meth)acrylate dispersion by incurporation of polydimethylsiloxane, React. Funct. Polym. 71 (2011) SUMMARY INVESTIGATION OF THE MORPHOLOGY AND SURFACE PROPERTIES OF CROSSLINKED POLY(URETHANE-ESTER-SILOXANE)S Jasna V. Džunuzović 1, Marija V. Pergal 1, Vesna V. Vodnik 2, Milena Špírková 3, Rafał Poręba 3, Slobodan Jovanović 4 1 University of Belgrade, Institute of Chemistry, Technology and Metallurgy (ICTM) Center of Chemistry, Belgrade, Serbia 2 University of Belgrade, Institute of Nuclear Science Vinča, Belgrade, Serbia 3 Institute of Macromolecular Chemistry AS CR, v.v.i., Nanostructured Polymers and Composites Department, Praha, Czech Republic 4 University of Belgrade, Faculty of Technology and Metallurgy, Belgrade, Serbia (Scientific paper) Two series of crosslinked poly(urethane-ester-siloxane)s were synthesized from α,ω-dihydroxy-(ethylene oxide-poly(dimethylsiloxane)-ethylene oxide) (EO-PDMS- -EO), 4,4 -methylenediphenyl diisocyanate and Boltorn hyperbranched polyesters of the second and third pseudo generation, by two-step polymerization in solution. The effect of the EO-PDMS-EO content and functionality of the applied crosslinking agent on the morphology and surface properties of the prepared poly(urethane-ester-siloxane)s was investigated by FTIR spectroscopy, small-angle X-ray scattering (SAXS), atomic force microscopy (AFM), scanning electron microscopy (SEM) and water absorption measurement. Different techniques (FTIR peak deconvolution, SAXS and AFM) revealed that decrease of the crosslinking agent functionality and EO-PDMS-EO content promotes microphase separation in the synthesized poly(urethane-ester-siloxane)s. SEM analysis and water absorption experiments showed that, due to the hydrophobic character of EO-PDMS-EO and its ability to migrate to the surface of poly(urethane-ester-siloxane)s, samples synthesized with higher EO-PDMS-EO content and crosslinking agents of lower functionality had more hydrophobic surfaces and better waterproof performances. The obtained results indicate that the synthesis of poly(urethane-ester-siloxane)s based on EO-PDMS-EO and Boltorn hyperbranched polyesters leads to the creation of networks with interesting morphological and surface properties, which can be easily tailored by changing the content of EO-PDMS-EO segment or functionality of hyperbranched polyester. Keywords: Hyperbranched polyester Poly(dimethylsiloxane) Crosslinked polyurethane Morphology Surface properties 821

32

33 Diffusion of drugs in hydrogels based on (meth)acrylates, poly(alkylene glycol) (meth)acrylates and itaconic acid Marija M. Babić 1, Jovana S. Jovašević 2, Jovanka M. Filipović 2, Simonida Lj. Tomić 2 1 Institute Kirilo Savić, Belgrade, Serbia 2 University of Belgrade, Faculty of Technology and Metallurgy, Belgrade, Serbia Abstract The aim of this paper is to propose equations for the diffusion of drugs for investigated drug/hydrogel systems using the parameters affecting the transport of drug through poly- (2-hydroxyethylmethacrylate/itaconic acid) (P(HEMA/IA)), poly(2-hydroxyethylacrylate/itaconic acid) (P(HEA/IA)), and poly(2-hydroxyethylmethacrylate/poly(alkyleneglycol) (meth)- acrylates) (P(HEMA/BIS)) copolymeric hydrogels. Different monomer types, as well as the variable content of some components in hydrogel composition (the amount of ionizable comonomer (IA) and different type of nonionic poly(alkyleneglycol) (meth)acrylates), ultimately defined the pore size available for drug diffusion. The hydrogels synthesized ranged from nonporous to microporous, based on the classification in accordance to the pore size, and could be classified as hydrogels that contain ionic groups and hydrogels without ionic groups. The drugs selected for this study are bronchodilators-theophylline (TPH), fenethylline hydrochloride (FE), and antibiotic cephalexin (CEX). Results of in vitro drug release tests defined the release systems based on the drug type, as well as the type of hydrogel used. The diffusion coefficient of drugs and the restriction coefficient, λ, defined as the ratio of solute to pore radius (r s /r ζ ) that describes the ease of drug release from the gels, were used as factors that govern the release process. SCIENTIFIC PAPER UDC :66: Hem. Ind. 66 (6) (2012) doi: /HEMIND B Keywords: hydrogels; 2-hydroxyethyl methacrylate; 2-hydroxyethyl acrylate; poly(alkylene glycol) (meth)acrylates; itaconic acid; controlled drug release; controlled drug release parameters; diffusion equations. Available online at the Journal website: Hydrogels are hydrophilic, crosslinked polymers made of homo- or copolymers that can absorb significant amounts of water or biological fluids but do not dissolve owing to the presence of chemical or physical crosslinks [1 3]. Their hydrophilic surface is characterized by a low interfacial free energy in contact with body fluids, which results in a low tendency for proteins and cells to adhere to these surfaces. Hydrogels resemble natural living tissue more than any other class of synthetic biomaterials. This is due to their high water contents and soft consistency, which is similar to natural tissue. Furthermore, the high water content of the materials contributes to their biocompatibility. Therefore hydrogels have found widespread biomedical and pharmaceutical applications as medical devices (contact lenses, artificial hearts and skin), and as controlled drug delivery systems [4 8]. Special types of hydrogels known as stimuli-responsive have been investigated for the development of smart materials in various fields. The term stimuliresponsive implies that significant changes of key pro- Correspondence: S. Lj. Tomić, Faculty of Technology and Metallurgy, University of Belgrade, Karnegijeva 4, Belgrade, Serbia. simonida@tmf.bg.ac.rs Paper received: 6 April, 2012 Paper accepted: 28 June, 2012 perties can be induced by an external stimulus, such as ph value, temperature, ionic strength, pressure, light, or electrical and magnetic fields [5]. Due to their excellent characteristics poly(2-hydroxyethylmethacrylate) (PHEMA), poly(2-hydroxyethylacrylate) (PHEA), itaconic acid (IA) and poly(alkyleneglycol) (meth)acrylates (BIS) are favorable as components of smart materials. PHEMA has excellent biocompatibility and physicochemical properties similar to those of living tissues [9,10]. PHEA has similar biocompatibility, cytotoxicity, low thrombogenicity, and cell compatibility to the widely studied PHEMA, but it is a less frequently studied polymer. Compared to PHEMA, PHEA seems to be a better option for the case of simulating the mechanical properties of soft tissues without losing water sorption capacity [11]. Itaconic acid (IA) easily copolymerizes and provides polymer chains with carboxylic side groups, which are highly hydrophilic and able to form hydrogen bonds with corresponding groups or ionize at adequate ph values. The addition of very small amounts of IA to HEMA and HEA hydrogels renders good ph sensitivity and increases the degree of hydrogel swelling [12 16]. Furthermore, IA is expected to show high biocompatibility because of its natural source [17]. Due to good biocompatibility and safe toxicity profile, poly(ethylene glycol)s (PEGs) are applied 823

34 M.M. BABIĆ et al.: DIFFUSION OF DRUGS IN HYDROGELS BASED ON (METH)ACRYLATES Hem. ind. 66 (6) (2012) in various biomedical areas such as drug delivery, wound healing, and tissue repair systems [18 21]. Poly(2-hydroxyethylmethacrylate/itaconic acid) (P(HEMA/IA)), poly(2-hydroxyethylacrylate/itaconic acid) (P(HEA/IA)), and poly(2-hydroxyethylmethacrylate)/poly(alkyleneglycol) (meth)acrylates (P(HEMA/BIS)) copolymeric hydrogels, which were characterized in our earlier papers [12 16], showed significant potential for application in drug delivery systems. The drug release studies were carried out in vitro, using bronchodilators-theophylline (TPH) and fenethylline hydrochloride (FE), and antibiotic cephalexin (CEX). In order to give deeper insight into controlled release of used drugs from these hydrogels, as a function of gel structure and morphology, its water sorption capabilities and the drug type, model equations describing solute diffusion through these hydrogels have been proposed for each drug/hydrogel system. EXPERIMENTAL Materials 2-Hydroxyethyl methacrylate (HEMA) (Aldrich), 2- -hydroxyethyl acrylate (HEA) (Aldrich), itaconic acid (IA) (Fluka), and different poly(alkylene glycol) (meth)acrylates (BIS) (BIS1-poly(ethylene glycol) 5 monoacrylate, BIS2-poly(propylene glycol) 6 monomethacrylate, BIS3- -poly(ethylene glycol) 6 (propylene glycol) 3 monomethacrylate, and BIS4-poly(ethylene glycol) 3 (propylene glycol) 6 monomethacrylate) (Laporte Chemicals) were used as components for hydrogel preparation. All polymerizations were performed in a mixture of water/ /ethanol as solvent. Ethyleneglycol dimethacrylate (EGDMA, Aldrich) was used as crosslinking agent and potassium persulfate (KPS, Fluka) and N,N,N,N tetramethylethylene diamine (TEMED, Aldrich) as initiator and activator, respectively. Theophylline (TPH) (Sigma), fenethylline hydrochloride (FE) (Sigma) and cephalexin (CEX) (Sigma) were used as drugs (Figure 1). Buffer solution was prepared using potassium mono and dihydrogenphosphate (Fluka) and sodium hydroxide (Fluka). Demineralized water was used for all polymerizations and the preparation of the buffer solutions. Preparation of hydrogels The P(HEMA/IA) and P(HEMA/BIS) hydrogels were prepared by gamma radiation induced crosslinking/ /polymerization. The reaction solutions were irradiated in a 60 Co radiation source, under ambient conditions, at a dose rate of 0.5 kgy/h, to an absorbed dose of 25 kgy. The P(HEA/IA) hydrogels were prepared by free radical crosslinking/copolymerization. The initiator, activator and crosslinker were added to the monomer feed mixture dissolved in water/ethanol mixture. The reaction mixture was degassed prior to polymerization and placed between two glass plates sealed with a rubber spacer (2 mm thick). The polymerizations were carried out at 50 C for 24 h. After the reaction, the gels were cut into discs and immersed in water for a week, to remove unreacted components. The water was changed daily. The discs were dried to xerogels. The amount of uncrosslinked IA was determined by titration of extract against NaOH (0.05 mol/l) to phenolphthalein end point. On the other hand, the amount of uncrosslinked HEMA and BIS were determined using a UV spectroscopy. In all cases, the results indicate that the conversion during cross-linking reaction was nearly complete. Figure 1. Chemical structures of used drugs theophylline (TPH), fenethylline hydrochloride (FE), and cephalexin (CEX). 824

35 M.M. BABIĆ et al.: DIFFUSION OF DRUGS IN HYDROGELS BASED ON (METH)ACRYLATES Hem. ind. 66 (6) (2012) Drug release study The drug powder (about 5% of the xerogel weight) was dissolved in water. All xerogel discs were immersed in drug solution and swollen to equilibrium. The swollen drug-loaded samples were then dried at ambient temperature for several days to constant mass and used for the release experiments. Release studies have been carried out in vitro by placing the dried and loaded sample in a definite volume of the release medium (a buffer of ph 7.40 (simulated physiological fluid)) at 37 C. The amount of drug released was measured using a UV spectrophotometer (Shimadzu UV Vis spectrophotometer UV-1800), by taking the absorbance of the solution at regular time intervals, at a wavelength of 272, 274, and 262 nm for TPH, FE and CEX, respectively. These measurements were repeated in triplicate. The concentration of the drug in the external solution at any selected time, c t, was calculated from the corresponding calibration curve of the absorbance against drug concentration. The data were analyzed using the commercial Origin Microcal 8.0 software. RESULTS AND DISCUSSION Diffusion properties of drug/hydrogel systems The release of drug molecules from hydrogels depends on the characteristics of the network: the polymer chemical structure and the network morphology. On the other hand, the physicochemical properties of the drugs also influence the release behavior. Drug transport in hydrogels can be analyzed using the frictional characteristics of the spherical solutes (drugs) as they diffuse through cylindrical pores. Derivation of drug transport through pores is based on the equation of drug flux with additional hindrance terms for convection and restrictions due to the tortuosity of the transport path. The apparent diffusion coefficient for the drug diffusion in a gel, D d(gel), relative to the diffusion coefficient of the solute in the liquid at infinite dilution, D d(water), is related to the restriction coefficient, λ, defined as the ratio of drug radius to hydrogel pore size. The coefficient λ combines the influence of polymeric network structure and the size of the drug on release and transport properties. The diffusion coefficients for solute diffusion in a gel, D d(gel), is calculated from Eq. (1), with the mass transfer principles based on Fick's law [22]: D dgel ( ) 2 2 k π l = (1) 16 where k is kinetic constant obtained from dynamic release studies (c t /c e dependence on t) and l is the thickness of the gel. The hydrodynamic radius of the drug molecule, r d, was calculated assuming a spherical shape [23], from relation (2): ( r ) d 2 3V = 4πNA 2/3 where N A is Avogadro s number, V is the molar volume of the drug, calculated from the partial specific volume of the atomic contribution of Le Bas [24]. The calculated values of the hydrodynamic radii are presented in Table 1. Table 1. Values of molecular weights, M d, hydrodynamic radii, r d, and bulk diffusion coefficients, D d(water), of the drugs Drug M d / g mol 1 r d / nm D d(water) 10 6 / cm 2 s 1 TPH FE CEX The drug diffusion coefficient in the liquid at infinite dilution D d(water) was calculated for each drug employed in this investigation (Table 1), using the Stokes Einstein Equation: D d(water) (2) kt = (3) 6πηr d where k is the Boltzmann constant, T is the absolute temperature, η is the solvent viscosity and r d is the hydrodynamic radius of the drug. In general, a solute is considered to follow Stokes law if the solvent can be treated as a continuum (i.e., r drug >> r solvent ). The mean pore size, ξ, of hydrogels was calculated using relation from (4) [25]: 1/3 2 c ξ φ CM n = 2 l Mr 1/2 where φ 2 is volume fraction of the swollen polymer, l is the carbon carbon single bond length (0.154 nm), C n is the rigidity factor of the polymer [15], M c is the average molar mass between the crosslinks [15] and M r is the molar mass of repeat unit. The calculated mean pore size ξ of the hydrogels (from aforementioned relation) ranged from 0.85 to 3.89 nm for P(HEMA/IA), nm for P(HEMA/BIS) (nonporous), and nm for P(HEA/IA) hydrogels (microporous). The diffusion coefficients of the drugs and drug release exponents, n, for P(HEMA/IA), P(HEMA/BIS), and P(HEA/IA) hydrogels, obtained from the results of drug release studies, are presented in Table 2. The values of drug release exponents n are near 0.5 in all cases (Table 2), suggesting that the release process can be described by a Fickian transport mechanism. (4) 825

36 M.M. BABIĆ et al.: DIFFUSION OF DRUGS IN HYDROGELS BASED ON (METH)ACRYLATES Hem. ind. 66 (6) (2012) Table 2. Diffusion parameters of the investigated drug/hydrogel systems Drug Hydrogel k n D d(gel) 10 7 / cm 2 s 1 TPH PHEMA P(HEMA/2IA) P(HEMA/3.5IA) P(HEMA/5IA) FE PHEMA P(HEMA/2IA) P(HEMA/3.5IA) P(HEMA/5IA) CEX P(HEMA/BIS1) P(HEMA/BIS2) P(HEMA/BIS3) P(HEMA/BIS4) PHEA P(HEA/2IA) P(HEA/3.5IA) P(HEA/5IA) Drug transport in hydrogels Among controlled release drug delivery systems, hydrogels are interesting due to their unique tunable time-dependent swelling behavior. The purpose of controlled release systems is to deliver the drug at specified rate to a desired target, keeping the drug concentration in the body at the therapeutically effective level, with the convenient drug release profile [24]. The structure of drugs: theophylline (TPH), fenethylline hydrochloride (FE), and cephalexin (CEX), used for the investigation of the drug release behavior of hydrogels prepared in this study, are presented in Figure 1. The bronchodilatators theophylline (TPH) and fenethylline hydrochloride (FE), loaded in P(HEMA/IA) hydrogels, contain HN functional groups and the antibiotic cephalexine (CEX) (pk a = 11.9), loaded in P(HEMA/BIS) and P(HEA/IA) hydrogels, contains NH 2 and COOH functional groups which can interact effectively with hydrogel functional groups (COO, OH) through ionic or hydrogen bonds. Hydrogen bonds can be formed also between OH groups from BIS components of P(HEMA/ /BIS) hydrogels and the functional groups of CEX. Therefore, drug release process in a buffer solution of ph 7.4, at 37 C is influenced by the composition of hydrogel. The drug restriction coefficient λ describes the ease of drug release from the gels. As it can be seen from Figures 2 and 3, the solute transport, presented by normalized diffusion coefficient as a function of λ, is best correlated by exponential functions for the release of TPH, FE and CEX from P(HEMA/IA), P(HEA/IA) (Figure 2) and P(HEMA/BIS) hydrogels (Figure 3). The equations which describe the diffusion of TPH and FE through P(HEMA/IA) hydrogels, and CEX diffusion through P(HEMA/BIS) and P(HEA/IA) hydrogels are presented in Table 3. The gels investigated can be divided in two groups. In the first group are ph-sensitive networks, P(HEMA/ /IA) and P(HEA/IA), which contain ionic groups from IA, where hydrogen bonding, physical crosslinks, and electrostatic and hydrophobic interactions can take place with the corresponding functional groups of the drug TPH-P(HEMA/IA) FE-P(HEMA/IA) CEX-P(HEA/IA) D d / D o λ Figure 2. Normalized drug diffusion coefficient for TPH and FE release from P(HEMA/IA) hydrogels, and for CEX release from P(HEA/IA) hydrogels as a function of drug restriction coefficient. 826

37 M.M. BABIĆ et al.: DIFFUSION OF DRUGS IN HYDROGELS BASED ON (METH)ACRYLATES Hem. ind. 66 (6) (2012) CEX-P(HEMA/BIS) 0.4 D d / D o Figure 3. Normalized drug diffusion coefficient for CEX release from P(HEMA/BIS) hydrogels as a function of drug restriction coefficient. λ In the second group are the nonionic P(HEMA/BIS) hydrogels, which have no ionic groups so the electrostatic interactions between the gel and the drug do not take place. For the reason of easier comparison, drugs with similar dimensions were chosen. When the polymer network is ph-sensitive, the drug is entrained within a network so the drug release is a function of the expansion of the ph-sensitive network, which is regulated by the monomer ratio, the cross-linking density and the ph value of the swelling media. In our experiments, the drug release studies were performed in a buffer of ph 7.4, which is higher than both pk a values of IA, so practically all IA residues in the polymeric network are ionized. Since the monomer ratio is varied in our copolymer samples it is evident that the drug release is faster for the gels with higher IA content. For the gel containing HEMA monomer and ionic groups from IA, THP/P(HEMA/IA) and FE/P(HEMA/IA), the lowest values for normalized diffusion coefficients were obtained. It is to be expected for two reasons: pores in these systems fall into the nonporous region and interactions between the functional groups of the drugs and of the network chains are taking place. The system CEX/P(HEA/IA) has higher pores in the network (mesoporous region) due to more hydrophilic HEA residues, and the values for normalized diffusion coefficients are higher. In all cases presented in Figure 2, the normalized diffusion coefficients decline exponentially with λ, but this trend is more pronounced in the case of CEX/P(HEA/IA) system, with higher pores. In the case of P(HEMA/BIS) hydrogels, which are nonionic and have the smallest pores (nonporous region), much higher values of normalized diffusion coefficients were obtained, but the same trend of exponentially declining values with decreasing pore dimen- Table 3. Proposed diffusion equations for each drug/hydrogel system Drug Hydrogel Equation TPH P(HEMA/IA) FE P(HEMA/IA) CEX P(HEA/IA) CEX P(HEMA/BIS) D D D D d(gel) -3-3 d(water) D ( λ) = exp 6.66 d(gel) -3-3 d(water) D ( λ) = exp 5.28 d(gel) -2-2 d(water) D D ( λ) = exp d(gel) d(water) ( λ) = exp

38 M.M. BABIĆ et al.: DIFFUSION OF DRUGS IN HYDROGELS BASED ON (METH)ACRYLATES Hem. ind. 66 (6) (2012) sions is present. This behavior confirms that the diffusion of drugs depend more on the interactions (most of all electrostatic) between the drug and network chains and less on the pore size. CONCLUSION Mathematical equations describing the diffusion of model drugs through drug delivery systems based on 2- -hydroxyethyl methacrylate (HEMA), and 2-hydroxyethyl acrylate (HEA), different poly(alkylene glycol) (meth)acrylates (BIS) and itaconic acid (IA) were obtained. Hydrogels were prepared by gamma-radiation induced crosslinking/polymerization, and chemically crosslinking/polymerization. Antibiotic cephalexin (CEX), and bronchodilatators theophylline (TPH) and fenethylline hydrochloride (FE) were used as model drugs. The release of drugs from these hydrogels was investigated in vitro, in a buffer of ph 7.4, at 37 C. The results of in vitro drug release studies showed that our release systems were determined by the type of drug, as well as by the type of hydrogel. The equations which describe the diffusion of TPH and FE through P(HEMA/IA) hydrogels, and CEX diffusion through P(HEMA/BIS) and P(HEA/IA) hydrogels are given by the normalized diffusion coefficient, D d(gel) )/(D d(water), as a function of the restriction coefficient, λ. The dependence of normalized diffusion coefficient on λ, are best correlated by exponential function in all cases. From the relations obtained it can be said that the diffusion of drugs, depend more on the interactions (most of all electrostatic) between the drug and network chains and less on the pore size. Acknowledgment This work has been supported by the Ministry of Education, Science and Technological Development of the Republic of Serbia (Grants No and ). REFERENCES [1] J. Kopecek, J. Yang, Hydrogels as smart biomaterials, Polym. Int. 56 (2007) [2] S. Chaterji, I.K. Kwon, K. Park, Smart polymeric gels: Redefining the limits of biomedical devices, Prog. Polym. Sci. 32 (2007) [3] M.W. Urban, Stratification, stimuli-responsiveness, selfhealing, and signaling in polymer networks, Prog. Polym. Sci. 34 (2009) [4] Y. Dai, P. Li, A. Wang, Intelligent drug delivery system of intelligent high polymer materials, Prog. Chem. 19 (2007) [5] Y. Qiu, K. Park, Environment-sensitive hydrogels for drug delivery, Adv. Drug Deliver. Rev. 53 (2001) [6] N.A. Peppas, Y. Huang, M. Torres-Lugo, J.H. Ward, J. Zhang, Physicochemical foundations and structural design of hydrogels in medicine and biology, Annu. Rev. Biomed. Eng. 2 (2000) [7] A. Kumar, A. Srivastava, I.Y. Galaev, B. Mattiasson, Smart polymers: Physical forms and bioengineering applications, Prog. Polym. Sci. 32 (2007) [8] B. Schmaljohann, Thermo- and ph-responsive polymers in drug delivery, Adv. Drug. Delivery Rev. 58 (2006) [9] S. Brahim, D. Narinesingh, A.G. Elie, Synthesis and hydration properties of ph-sensitive p(hema)-based hydrogels containing 3-(trimethoxysilyl)propyl methacrylate, Biomacromolecules 4 (2003) [10] S. Lahooti, M.V. Sefton, Microencapsulation of normal and transfected L929 fibroblasts in a HEMA-MMA copolymer, Tissue Eng. 6 (2000) [11] G. Gallego Ferrer, M. Salmerón Sánchez, J.L. Gómez Ribelles, F.J. Romero Colomer, M. Monleón Pradas, Nanodomains in a hydroplylic hydrophobic IPN based on poly(2-hydroxyethyl acrylate) andpoly(ethyl acrylate), Eur. Polym. J. 43 (2007) [12] S.Lj. Tomić, E.H. Suljovrujić, J.M. Filipović, Biocompatible and bioadhesive hydrogels based on 2-hydroxyethyl methacrylate, monofunctional poly(alkylene glycol)s and itaconic acid, Polym. Bull. 57 (2006) [13] S.Lj. Tomić, M.M. Mićić, S.N. Dobić, J.M. Filipović, E.H. Suljovrujić, Smart poly(2-hydroxyethylmethacrylate/itaconicacid) hydrogels for biomedical application, Radiat. Phys. Chem. 79 (2010) [14] S.Lj. Tomić, M.M. Mićić, J.M. Filipović, E.H. Suljovrujić, Swelling and drug release behavior of poly(2-hydroxyethyl methacrylate/itaconic acid) copolymeric hydrogels obtained by gamma irradiation, Radiat. Phys. Chem. 76 (2007) [15] S.Lj. Tomić, M.M. Mićić, J.M. Filipović, E.H. Suljovrujić, Synthesis, characterization and controlled release of cephalexin drug from smart poly(2-hydroxyethyl methacrylate/poly(alkylene glycol)(meth)acrylates hydrogels, Chem. Eng. J. 160 (2010) [16] S.Lj. Tomić, J.S. Jovašević, J.M. Filipović, S.N. Dobić, In vitro cephalexin release study from new poly(2-hydroxyethyl acrylate/itaconic acid) copolymeric hydrogels, 2th Conference Innovation in Drug Delivery, Aix-en-Provence, France, Programme and Abstracts, 2010, p [17] M. Pertruccioli, V. Pulchi, F. Federici, Itaconic acid production by Aspergillus terreus on raw starchy materials, Lett. Appl. Microbial. 28 (1999) [18] N.A. Peppas, K.B. Keys, M. Torres-Lugo, A.M. Lowman, Poly(ethylene glycol)-containing hydrogels in drug delivery, J. Control. Release 62 (1999) [19] J.F. Lutz, Polymerization of oligo(ethylene glycol) (meth)acrylates: Toward new generations of smart biocompatible materials, J. Polym. Sci. Pol. Chem. 46 (2008) [20] Y.-Z. You, D. Oupicky, Synthesis of temperature-responsive heterobifunctional block copolymers of PEG and PNIPAM, Biomacromolecules 8 (2007) [21] M.Y. Arica, D. Tuglu, M.M. Basar, D. Kilic, G. Bayramoglu, E. Batislam, Preparation and characterization of infection-resistant antibiotics-releasing hydrogels rods of 828

39 M.M. BABIĆ et al.: DIFFUSION OF DRUGS IN HYDROGELS BASED ON (METH)ACRYLATES Hem. ind. 66 (6) (2012) poly[hydroxyethyl methacrylate-co-(poly(ethylene glycol)-methacrylate]: Biomedical application in a novel rabbit penile prosthesis model, J. Biomed. Mater. Res., B 86 (2008) [22] C.S. Brazel, N.A. Peppas, Mechanisms of solute and drug transport in relaxing swellable, hydrophilic glassy polymers, Polymer 40 (1999) [23] A. Martin, J. Swarbrick, A. Cammarata, Physical Pharmacy-Physical Chemical Principles in the Pharmaceutical Sciences, 3 rd ed., Lea and Febinger, Philadelphia, PA, [24] C.R. Wilke, P. Chang, Correlation of diffusion coefficients in dilute solutions, AIChE. J. 1 (1955) [25] S.K. Bajpai, S. Singh, Analysis of swelling behavior of poly(methacrylamide-co-methacrylic acid) hydrogels and effect of synthesis conditions on water uptake, React. Funct. Polym. 66 (2006) IZVOD DIFUZIJA LEKOVA U HIDROGELOVIMA NA BAZI (MET)AKRILATA, POLI(ALKILENGLIKOL)-(MET)AKRILATA I ITAKONSKE KISELINE Marija M. Babić 1, Jovana S. Jovašević 2, Jovanka M. Filipović 2, Simonida Lj. Tomić 2 1 Institut Kirilo Savić, Beograd, Srbija 2 Univerzitet u Beogradu, Tehnološko metalurški fakultet, Beograd, Srbija (Naučni rad) Cilj ove studije je da se predlože difuzione jednačine za ispitivane sisteme lek/hidrogel. Korišćeni su hidrogelovi poli(2-hidroksietilmetakrilat/itakonska kiselina) (P(HEMA/IK)), poli(2-hidroksietilakrilat/itakonska kiselina) (P(HEA/IK)) i poli- (2-hidroksietilmetakrilat/poli(alkilenglikol)-(met)akrilati) (P(HEMA/BIS)). Komponenta koja se menja u sastavu hidrogela HEMA, HEA, kao i udeo komponente sa promenljivim sadržajem (udeo jonizujućeg komonomera (IK) i tip BIS komponente) definiše veličinu pora koja je dostupna za difuziju leka. U ovoj studiji su korišćeni lekovi bronhodilatori teofilin (TPH) i fenetilin-hidrohlorid (FE), i antibiotik cefaleksin (CEX). Ovi gelovi su klasifikovani u režimu poroznosti kao neporozni i mikroporozni, sa veličinom pora u opsegu 0,18 24,9 nm. Kontrolisano otpuštanje lekova je izvedeno u in vitro uslovima u puferu ph 7,40 i na 37 C, da bi se odredili difuzioni koeficijenti leka u hidrogelovima. Na osnovu toga su predložene jednačine difuzije leka kroz hidrogel za svaki sistem lek/hidrogel. Rezultati dobijeni fitovanjem eksperimentalnih podataka su pokazali da difuzija leka zavisi od hemijske strukture i morfologije hidrogela i parametra λ, koji predstavlja odnos prečnika leka i veličine pora. Eksponencijalna zavisnost koeficijenta restrikcije od normalizovanog koeficijenta difuzije je dobijena za sisteme TPH/P(HEMA/IA), FE/P(HEMA/ /IA), CEX/P(HEMA/BIS) i CEX/P(HEA/IA).Utvrđeno je da veliki uticaj na difuziju leka imaju interakcije koje se odigravaju između funkcionalnih grupa leka i polimerne mreže. Ključne reči: hidrogelovi 2-hidroksietilmetakrilat 2-Hidroksietilakrilat Poli- (alkilenglikol)-(met)akrilati Itakonska kiselina Kontrolisano otpuštanje leka Parametri kontrolisanog otpuštanja leka Jednačine difuzije 829

40

41 Potential application of thermo-sensitive hydrogels for controlled release of phenacetin Snežana S. Ilić-Stojanović 1, Ljubiša B. Nikolić 1, Vesna D. Nikolić 1, Jela R. Milić 2, Slobodan D. Petrović 3, Goran M. Nikolić 4, Agneš J. Kapor 5 1 University of Niš, Faculty of Technology, Leskovac, Serbia 2 University of Belgrade, Faculty of Pharmacy, Belgrade, Serbia 3 University of Belgrade, Faculty of Technology and Metallurgy, Belgrade, Serbia 4 University of Niš, Faculty of Medicine, Niš, Serbia 5 University of Novi Sad, Faculty of Sciences, Department of Physics, Novi Sad, Serbia Abstract Over the past years, many scientific research studies have been focused on thermo-sensitive hydrogels containing N-isopropylacrylamide (NIPAM) as a monomer. The NIPAM based hydrogels with 20 mol% 2-hydroxypropyl methacrylate (HPMet) were synthesized using ethylene glycol dimethacrylate as a cross-linker. The characterization of xerogel and phenacetin using Fourier transform infrared (FTIR) spectroscopy and Scanning electron microscopy (SEM)confirm the performed synthesis with satisfactory purity as well as loading of phenacetin into hydrogel. The swelling transport mechanism at simulated physiological conditions (ph 2.20 and 7.40 at 37 C) is described by the time-independent kinetics. The potential application of synthesized hydrogels for the controlled release of phenacetin as a model drug was investigated at simulated physiological conditions by HPLC method. SCIENTIFIC PAPER UDC : : Hem. Ind. 66 (6) (2012) doi: /HEMIND I Keywords: thermo-sensitive hydrogel; N-isopropylacrylamide; 2-hydroxypropyl methacrylate, phenacetin. Available online at the Journal website: Hydrogels are cross-linked three-dimensional polymer networks, which may absorb a significant amount of water and swell several thousand times more than a dry gel mass. They are soft, wet and pliable materials, compatible with most living tissues and with a wide range of potential biomedical applications. They swell in water but do not dissolve in it [1,2]. A special class of hydrogels, which are called intelligent or stimuli-responsive, exhibits a significant volume transition in response to small changes in the environmental conditions like temperature, ph, ultrasound, electric field, magnetic field, light, pressure, glucose, etc. Their ability to swell and deswell according to environmental conditions makes them interesting for use as drug delivery systems [3 7]. During the last 30 years hydrogels have become very interesting drug carriers for a controlled delivery due to their biocompatibility and resemblance to biological tissues [4,8 12]. Thermo-sensitive hydrogels exhibit volume phase transitions at critical temperatures, i.e., lower critical solution temperatures (LCST) or upper critical solution temperatures (UCST). LCST polymers exhibit a swellingto-shrinking transition with increasing the temperature (negative thermo-sensitive hydrogels), while the UCST Correspondence: S. Ilić-Stojanović, University of Niš, Faculty of Technology, Bulevar oslobođenja 124, Leskovac, Serbia. ilic.s.snezana@gmail.com Paper received: 22 February, 2012 Paper accepted: 30 July, 2012 polymers undergo the opposite transitions (positive thermo-sensitive hydrogels) [13 15]. Poly(N-isopropylacrylamide), p(nipam), is one of the most widely used thermo-sensitive polymers. It has a hydrophilic amide group and a hydrophobic isopropyl group. p(nipam) has LCST of around 32 C [2,14]. Hydrophilic 2-hydroxypropyl methacrylate (HPMet) has excellent biocompatibility with living tissues and good tolerance by the cells. HPMet is widely used in immunocytochemical processes [16], as well as in the production of contact lenses [17]. Some investigations are based on its properties as a drug carrier or in separation processes [18 22]. Non-steroidal anti-inflammatory drugs (NSAID) are drugs with analgesic (but non-narcotics), antipyretic and anti-inflammatory effects. Phenacetin, N-(4-etoxyphenyl)-acetamide, was selected from the group of NSAID as a model drug in the study of hydrogel poly(n- -isopropylacrylamide-co-2-hydroxypropyl methacrylate), p(nipam-co-hpmet) being potential carriers for the controlled delivery [21,22]. Phenacetin was chosen as one part of the wider investigation of the selected nonsteroidal anti-inflammatory drugs: paracetamol, which is a precursor of phenacetin, then ibuprofen, naproxen, and piroxicam. Recent studies show the results of a series of formulations of phenacetin with interpolymer carboxyvinylpolymer complexes with poly(ethylene oxide) [23 25], carbopol [26] or methylcellulose [27], as well as the comparative determination of a particle size of phenacetin bulk powder [28]. 831

42 S.S. ILIĆ-STOJANOVIĆ et al.: THERMO-SENSITIVE HYDROGELS FOR CONTROLLED RELEASE OF PHENACETIN Hem. ind. 66 (6) (2012) The aims of this study are the synthesis of thermosensitive p(nipam-co-hpmet) hydrogels by free radical copolymerization, the synthesis of phenacetin, their characterization by using Fourier transform infrared spectroscopy (FTIR), scanning electron microscopy (SEM) and swelling behavior and the investigation of the potential applications of obtained hydrogels for a controlled release of phenacetin as a model drug. EXPERIMENTAL Reagents N-Isopropylacrylamide (NIPAM) 99%, 2-hydroxypropyl methacrylate (HPMet) 96.5% and 2,2'-azobis(2- -methylpropionitrile) (AZDN) 98%, from Acros, USA; ethylene glycol dimethacrylate (EGDM) 97%, Fluka, CH; acetone, Centrohem, Belgrade, RS; methanol, Unichem, Belgrade, RS; phenacetin, synthesized in the laboratory of the Faculty of Technology, Leskovac. Synthesis of p(nipam-co-hpmet) p(nipam-co-hpmet) were synthesized by radical polymerization of NIPAM with 20 mol% of HPMet in acetone (as solvent) by using the initiator (AZDN) and a varied concentration of cross-linker (1, 1.5, 2 and 3 mol% EGDM) under the temperature regime: 0.5 h at 70 C, 2 h at 80 C, 0.5 h at 85 C. The temperature range from 70 to 85 C was chosen to activate the initiator and the polymerization process as well because of the full utilization of the initiator at the temperature increase. P(NIPAM), with 2 mol% of EGDM as a crosslinker was synthesized under the same conditions. The obtained gels were extracted by methanol in order to remove all non-reacted water insoluble compounds, monomers and oligomers. Swollen gels were dried to constant mass at 40 C. Synthesis of phenacetin The procedure of phenacetin synthesis is performed in two steps (Figure 1). At the first stage of the reaction between p-aminophenol and anydrous acetic, p-acetaminopheol is obtained. In the second step, the conversion of acetaminopheol (as the type of the Williamson ether synthesis) using sodium ethanoate in the mixture of p-ethyl iodide and hydroxyacetamide is carried out to N-4-etoxyphenyl acetamide. The raw product was recrystallized from ethanol with the addition of active carbon. The obtained phenacetin was filtered and dried in air. Characterization methods Fourier transform infrared spectroscopy (FTIR) The samples were recorded with KBr technique on a BOMEM MB-100 (Hartmann & Braun, Canada) at wavelengths of cm 1. Scanning electron microscopy (SEM) The morphology of lyophilized hydrogels and samples with loaded phenacetin was investigated on a JEOL Scanning Microscope JSM The samples were first coated by gold/palladium alloy (15/85). Swelling behavior The swelling behavior of xerogels were carried out in distilled water (at 20 and 40 C) and in simulated physiological conditions (ph 2.20 and 7.40 at 37 C) and monitored gravimetrically. The swelling ratio, α, was calculated according to the equation: m m0 α = (1) m 0 where m 0 is the mass of xerogel, and m is the mass of swollen gel at time t. Loading of phenacetin Xerogel of p(nipam-co-hpmet) was swelled in the phenacetin solution, 40 mg/cm 3, in a 80/20% methanol/distilled water mixture for 48 h. The released phenacetin The swollen gel of p(nipam-co-hpmet) was soaked with 7 cm 3 of solutions at simulated physiological conditions (ph 2.20 and 7.40) and the amount of released phenacetin was monitored by high-pressure liquid chromatography (HPLC) for 24 h at a temperature of 37 C. The analysis was performed by HPLC on an Agilent Figure 1. Schematic form of phenacetin synthesis. 832

43 S.S. ILIĆ-STOJANOVIĆ et al.: THERMO-SENSITIVE HYDROGELS FOR CONTROLLED RELEASE OF PHENACETIN Hem. ind. 66 (6) (2012) 1100 Series device under the following conditions: detector DAD 1200, detection wavelength 205 nm, column ZORBAX XDB-C18, 250 mm 4.6 mm, 5 μm, eluent methanol, eluent flow 1 ml/min, sample injection volume 20 μl, column temperature 25 C. RESULTS AND DISCUSSION FTIR Analysis In the FTIR spectrum of NIPAM monomer the following bands were observed: valence vibrations of C=C double bond at 1622 cm 1, symmetric valence vibrations, ν s, of CH 3 groups at 2875 cm 1, asymmetric valence vibrations, ν as, of CH 3 groups at 2970 cm 1, ν as of CH 2 group at 2933 cm 1 and ν as of CH from vinyl group at 3072 cm 1. The valence absorption of N H bond from the amide structure was observed at 3284 cm 1. In the spectrum of HPMet monomer the following bands were observed: valence vibrations of C O at 1296 and 1173 cm 1, valence vibrations of C=C double bond at 1638 cm 1, vibrations of C=O double bond from ester at 1720 cm 1, vibrations of C H bond: ν s of CH 3 at 2894 cm 1 and ν as of CH 3 at 2980 cm 1, and valence vibration of OH group at about 3450 cm 1. In the FTIR spectrum of p(nipam-co-hpmet) xerogel with 20 mol% of HPMet, there appears to be no absorption bands that could originate from the double C=C bond, which indicates that the synthesis is performed by the initiation of radicals (Figure 2). The following bands were observed: vibrations of the double C=O bond from amide at 1652 cm 1 and the double C=O bond from ester at 1728 cm 1, vibrations of C H bond from CH 3 and CH 2 groups at 2973, 2930 and 2878 cm 1. Valence vibrations of lateral NH and OH group were observed at 3304 and 3437 cm 1, respectively. FTIR Spectrum of phenacetin indicates that phenacetin was successfully obtained with the satisfying purity. The following bands that originated from the aromatic ring were observed: valence vibration of C H bond at 3073 cm 1, out-of-plane deformation vibrations of C H bonds at 838, 826, 785 and 743 cm 1 and valence vibration of C=C bond with four characteristic bands at: 1608, 1556, 1509 and 1481 cm 1. The valence vibration of C O C bond can be observed at 1245 cm 1 (ν as ) and 1048 cm 1 (ν s ). The valence vibration of a double C=O bond of secondary amide at 1658 cm 1, the vibrations of C H bond of CH 3 groups: ν s at 2886 and 2850, ν as at 2981 and 2927 cm 1 and the valence absorption of NH bond from the secondary amide at 3432 and 3285 cm 1 were observed. Figure 2. FTIR Spectra of p(nipam-co-hpmet), phenacetin and p(nipam-co-hpmet) hydrogel with loaded phenacetin. 833

44 S.S. ILIĆ-STOJANOVIĆ et al.: THERMO-SENSITIVE HYDROGELS FOR CONTROLLED RELEASE OF PHENACETIN Hem. ind. 66 (6) (2012) The FTIR spectrum of xerogels with loaded phenacetin indicates that phenacetin was successfully inserted into the hydrogels cavity by forming intermolecular noncovalent bonds. The following bands that originated from xerogel were observed: valence vibrations of the double C=O bond from ester at 1719 cm 1 and of the double C=O bond from amide at 1660 cm 1, which were moved for 9 and 8 units to the lower frequency side, respectively. Valence vibrations of OH group were observed at 3441 cm 1, which was moved for 4 units to a higher frequency side. The following bands that originated from phenacetin were observed: valence vibrations of NH and C=O bond of secondary amide at 3286 and 1660 cm 1 which were moved for 1 and 2 units to a higher frequency side. Shifts of these bonds may indicate the interaction of the drug loaded into the hydrogel. Considering the fact that those shifts are very small, the interaction between hydrogel and phenacetin are of a noncovalent type. SEM Analysis SEM micrographs of swollen and then lyophilized samples of p(nipam-co-hpmet) hydrogel with 0.5 mol% EGDM for a) pure hydrogel and b) with loaded phenacetin are presented in Figure 3. The hydrogel morphology has a porous surface. The observed structure of hydrogel was changed after the treatment of the phenacetin solution and its loading into the hydrogel structure. Swelling behavior of p(nipam-co-hpmet) hydrogels Swelling behavior for a series of hydrogels at 20 and 40 C depending on the time and EGDM content is demonstrated in the authors previous studies [22]. It was found that the swelling ratio for p(nipam-co- -HPMet) hydrogels with 20 mol% HPMet increases extensively during the first 6 h. Along with the increase of the cross-linker content, the swelling ratio decreases. The reason is the formation of a denser network due to large amounts of the cross-linker. Polymer chains are more fixed and less able to absorb water. In contrast, when a small quantity of cross-linker is present, the length of the polymer chains between two knots is larger, the network is able to expand and absorb a greater amount of water. By adjusting the amount of cross-linkers used in the gels, the synthesis can affect the water absorption ability, and hence the swelling ratio can be controlled. In order to study the water transport mechanism from p(nipam-co-hpmet) hydrogels with different cross-linker contents, Fick s Equation was applied to fit the experimental data [5]: n F = M / M = kt (2) t e where M t /M e is the fractional sorption, M t is the amount of the water absorbed at time t, M e is the maximum amount of the absorbed water; k is a constant incorporating characteristic of the polymer network system, n is the diffusion exponent. The exponents n and k are values determined from the slope and intercept of the plots of ln M t /M e versus ln t for p(nipam- -co-hpmet) hydrogels at different EGDM contents. If n is less than 0.5, the swelling process is controlled by the Fickian diffusion mechanism. If n varies between 0.5 and 1, the diffusion and polymer relaxation control the swelling process and indicates an anomalous diffusion mechanism, which is known as non-fickian diffusion. The values of n greater than 1 are described as type III (Case III) or Super Case II. Figure 4 shows the dependence of n versus EGDM cross-linker content, according to p(nipam-co-hpmet) hydrogels swelling at 20 and 40 C. It is clear from the analysis that with the increase of (a) (b) Figure 3. SEM Micrographs of lyophilized p(nipam-co-hpmet) samples with 0.5 mol% of EGDM: a) pure and b) with loaded phenacetin. 834