Macroporous composite IPN hydrogels based on poly(acrylamide) and chitosan with tuned swelling and sorption of cationic dyes

Chemical Engineering Journal 204–206 (2012) 198–209

Contents lists available at SciVerse ScienceDirect

Chemical Engineering Journal journal homepage: www.elsevier.com/locate/cej

Macroporous composite IPN hydrogels based on poly(acrylamide) and chitosan with tuned swelling and sorption of cationic dyes Ecaterina Stela Dragan ⇑, Maria Marinela Lazar, Maria Valentina Dinu, Florica Doroftei ‘‘Petru Poni’’ Institute of Macromolecular Chemistry, Aleea Grigore Ghica Voda 41 A, 700487 Iasi, Romania

h i g h l i g h t s

g r a p h i c a l a b s t r a c t

" Macroporous hydrogels by

cryopolymerization of acrylamide in presence of chitosan. " Anionic full-IPN cryogels by crosslinking of chitosan chains in alkaline medium. " Cryogels with high sorption capacity of cationic dyes and high reusability. " High selectivity in separation of cationic dyes from the mixture with anionic dyes.

a r t i c l e

i n f o

Article history: Received 7 June 2012 Received in revised form 19 July 2012 Accepted 24 July 2012 Available online 3 August 2012 Keywords: Chitosan Cryogel Polyacrylamide Hydrolysis Methylene blue Model isotherms

a b s t r a c t The synthesis of macroporous ionic composite cryogels consisting of two independently cross-linked and oppositely charged networks is reported in the paper. Semi-interpenetrating network cryogels were prepared first by cross-linking polymerization of acrylamide (AAm) with N,N0 -methylenebisacrylamide (BAAm) in the presence of chitosan (CS), under freezing conditions, the main parameters varied being cross-linker ratio (X), pH of the CS solution, and CS molar mass. The fraction of CS trapped in the semi-IPN cryogels increased with the increase of both X and pH of CS solution. The CS chains trapped in the semi-IPN cryogels have been cross-linked with epichlorohydrin under alkaline conditions, to generate the second network, when a partial hydrolysis of the amide groups in the PAAm matrix simultaneously occurred. Examination of scanning electron microscope images showed inter-connected macropores, in both semi-IPN and full-IPN cryogels. The generation of an anionic matrix during the formation of the second network led to a high sorption capacity of a model cationic dye, methylene blue (MB), by the full-IPN cryogel, around 750 mg dye/g gel, at 25 °C. The benefits of the full-IPN cryogels consist of their high reusability and in the high selectivity in separation of MB from the mixture with methyl orange. Ó 2012 Elsevier B.V. All rights reserved.

1. Introduction Hydrogels, by their high water content, are similar to a variety of natural living tissues, and therefore have found large applicability in medicine as artificial organs, cartilage, muscles, immunoinsulation membranes, drug delivery systems [1–5], etc. For many applications, like artificial implants and controlled release systems, multicomponent hydrogels as semi- or full-interpenetrating polymer networks (IPNs) showed improved response rate and diffusion ⇑ Corresponding author. Tel.: +40 232217454; fax: +40 232211299. E-mail address: [email protected] (E.S. Dragan). 1385-8947/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.cej.2012.07.126

of solutes [6–9]. Considerable interest has been lately focused on the macroporous hydrogels, characterized by a faster response rate at small changes of the external stimuli than the conventional hydrogels. Macroporous hydrogels can be mainly achieved by: cross-linking polymerization in the presence of a pore-forming agent, when a microphase separation occurs [10], porogen leaching [11–14], cross-linking in the presence of substances releasing porogen gases [15], lyophilization of the hydrogel swollen in water [16,17], and cryogelation [18–20]. By cryogelation, the cross-linking polymerization reactions are conducted below the freezing point of the reaction solutions, when the most part of the solvent (water) forms crystals, the bound water and the soluble substances

E.S. Dragan et al. / Chemical Engineering Journal 204–206 (2012) 198–209

(monomers, initiator, polymers) being concentrated in a non-frozen liquid microphase, where the gel is formed. Cryogelation technique has been also used to produce scaffolds with controlled internal microarchitecture by the cross-linking of two polysaccharides under freezing conditions [21]. Advantages of cryogelation in the preparation of hydrogels consist of the absence of any organic porogen, the ice crystals playing the role of inert template, the microstructure of the gel being the negative replica of the ice crystals [18–27]. Thus cryogelation is a low cost and very friendly technique for the fabrication of supermacroporous gels. Not only hydrogels have been prepared by this technique, but also organogels [28]. Cryogels, by their inter-connected pore structure, allow the unhindered diffusion of solutes or even colloidal particles, making them very attractive in biomedicine and biotechnology including chromatographic materials, carriers for the immobilization of molecules and cells, matrices for cell separations, and cell culture [24– 28]. Cryogels can withstand high levels of deformations, being also characterized by superfast responsiveness at water absorption [22,29–33]. Ionic cryogels would have potential applications in controlled delivery of drugs and separation processes of small ionic species [34–36]. Poly(acrylamide) (PAAm) is a widely used polymer due to its affinity for proteins and other biomolecules, the presence of amide groups making this polymer susceptible for further reactions [6,8,37]. By their accessibility, biocompatibility, and biodegradability, PAAm hydrogels constitute one of the most investigated matrix in the preparation of semi-IPN hydrogels, achieved by cross-linking polymerization of acrylamide in the presence of synthetic or natural polymers. Such hydrogels have numerous applications, such as drug delivery systems [38,39], soil conditioners, and wastewaters remediation [40]. Chitosan (CS) is a copolymer of N-acetyl-glucosamine and N-glucosamine units distributed randomly or in blocks throughout the biopolymer chain, depending on the processing method used to obtain the biopolymer. By its outstanding properties, such as gel and film forming ability, bioadhesion, biodegradability and biocompatibility, chitosan has received a great deal of attention in the pharmaceutical field [1,3,41]. Due to the high content of amino and hydroxyl functional groups, CS has also drawn attention as a sorbent showing high potential for the adsorption of proteins, dyes, and metal ions [6,37,40]. Last decades, several works have explored the feasibility of PAAm/CS IPN composites for various applications [27,37–39,42– 45]. The most part of them refer to the conventional hydrogels [37–39,43,44]. Porous PAAm/CS IPN composite hydrogels have been prepared either in the presence of substances releasing porogen gases during the synthesis [42] or by freeze-drying of the equilibrium swollen gels [45]. Jain et al. reported on the synthesis of PAAm/CS based cryogel matrix in different formats, namely, monolith, disks, and beads used for the cell immobilization in packedbed bioreactors [27]. Recently we have reported the preparation of conventional ionic hydrogels, using poly(acrylamide) (PAAm) as a matrix and CS as trapped polycation [9]. Preliminary tests on the capacity of these composite gels to adsorb model ionic dyes evidenced some shortcomings consisting of their limited mechanical resistance, which also diminished their applicability in separation processes. Therefore, in this study, cryogelation technique has been adopted to prepare first macroporous composite semi-IPN ionic cryogels based on PAAm as a matrix and CS as trapped polycation. In the second step, full-IPN macroporous composite gels have been generated by the cross-linking with epichlorohydrin (ECH) of the CS chains trapped in the semi-IPN PAAm/CS cryogels, under alkaline conditions, when a partial hydrolysis of the amide groups in the PAAm matrix occurred. A remarkable enhancement of the equilibrium swelling ratio in basic range has been observed in the case of full-IPN cryo-

199

gel. To demonstrate that negative charges are predominant in the full-IPN, the adsorption/desorption of a model cationic dye, methylene blue (MB), has been investigated as a function of contact duration and initial concentration of the dye. It was also shown that the full-IPN composite cryogels have a selective sorption for MB from its mixture with methyl orange (MO). 2. Experimental 2.1. Materials The CS as powder, with molar mass of 235 kDa (CS1), purchased from Fluka, and CS with molar mass of 467 kDa (CS2), purchased from Sigma–Aldrich, were used as received. The viscometric average molar mass has been measured in 0.3 M CH3COOH–0.2 M CH3COONa (1:1, v/v), at 25 ± 0.1 °C according to the method previously described [46]. Degree of acetylation (DA) of CS has been evaluated by infrared spectroscopy in KBr pellets, using a Vertex 70 Bruker FTIR spectrometer as previously shown [46]. An average value of DA = 15%, resulted from three measurements, has been taken into account for both samples. Acrylamide (AAm, Fluka), N,N0 -methylenebisacrylamide (BAAm), ammonium persulfate (APS), N,N,N0 ,N0 -tetramethylethylenediamine (TEMED), all purchased from Sigma–Aldrich, were used as received. ECH purchased from Sigma–Aldrich, has been double distilled on KOH before using. Stock solutions of APS and TEMED were prepared by dissolving 0.2 g of APS and 0.625 mL of TEMED, each in 25 mL of double distilled water. Stock solutions of BAAm were prepared by dissolving BAAm, calculated for a certain cross-linker ratio defined as the mole ratio of the cross-linker BAAm to the monomer AAm, X, in 25 mL of double distilled water, at 30 °C, under magnetic stirring. Methyl orange (MO) was used after three times recrystallization from a 8 wt.% solution in a mixed solvent of water and methanol (1:1 v/v). Methylene blue (MB) from Sigma–Aldrich was used without purification. 2.2. Preparation of PAAm/CS composite cryogels For the preparation of PAAm/CS composite cryogels, the free radical cross-linking copolymerization of AAm has been performed in the presence of CS, in aqueous medium, at 18 °C. This temperature has been chosen because the previous studies concerning the formation of PAAm cryogels showed that cryogels with enhanced mechanical properties and with large pores of sizes about 70 lm could be obtained [22,29,31]. Also, it was found that the lower the gel preparation temperature, the shorter the time period until the freezing temperature of the reaction solution is reached [22]. This freezing temperature has been also adopted as appropriate in the preparation of other cryogels, both hydrogels [30] and organogels [28]. The redox initiator system used consisted of APS and TEMED. The cross-linker ratio varied in the range 1/80–1/20. The feed composition and the samples code of the composite gels are summarized in Table 1. The general code of the semi-IPN composite hydrogels consists of the term s-IPN followed by three numbers separated by dots: the first one represents CS1 or CS2, respectively, used as trapped polymer, the second one represents the mole number of AAm per one mole of BAAm, and the third one represents the pH of the CS solution. The general code of full-IPN consists of the term IPN followed by the same numbers like the semi-IPN used for their preparation. In the synthesis of semi-IPN, the initial concentration of monomers (AAm + BAAm), Co (5 w/v%), and the concentration of APS and TEMED have been kept constant in all experiments. The synthesis procedure is briefly presented below, taking the sample

200

E.S. Dragan et al. / Chemical Engineering Journal 204–206 (2012) 198–209 Table 1 Feed composition of the composite cryogels. Samplea

PAAm s-IPN1.60.5 s-IPN1.60.6 s-IPN2.60.5 s-IPN2.60.6 s-IPN1.40.5 s-IPN1.40.6 IPN1.40.6c IPN1.60.6c IPN2.60.6c a b c

BAAm:AAm molar ratio

1/80 1/60 1/60 1/60 1/60 1/40 1/40 1/40 1/60 1/60

CS Code

Molar mass, Mv, kDa

GFYb, %

CS removed in s-IPN, wt.%

– CS1 CS1 CS2 CS2 CS1 CS1 CS1 CS1 CS2

– 235 235 467 467 235 235 235 235 467

95 89.1 91.5 92.8 92.3 91.8 91.5 – – –

– 60.8 41.5 43.4 27.5 51.5 35.8 – – –

12.6 wt.% of CS added in the reaction mixture for all composite gels. Gel fraction yield. GFY and the fraction of CS removed by extraction have been not evaluated.

s-IPN1.40 (Table 1) as an example. Typically, 0.4742 g AAm, 6.3 g aqueous solution of CS1 (1 w/v%) obtained by dissolving the CS powder in 1 v/v% acetic acid aqueous solution and moderate stirring for 24 h, 0.7 mL double distilled water, 1 mL BAAm (0.643 g/ 25 mL) and 1 mL TEMED (0.625 mL/25 mL) were first mixed in a graduated flask of 10 mL. The solution was cooled at 0 °C in ice– water bath, purged with nitrogen gas for 20 min and then, 1 mL of APS stock aqueous solution has been added and the whole mixture has been further stirred about 20 s. Portions of this solution, each 1 mL, were transferred into syringes of 5 mm in diameter; the narrow tips of the syringes have been cut off before loading the reaction mixture, and thus, after polymerization, the cylindrical gels could be easily ejected. The syringe cut ends have been sealed with parafilm, and kept at 18 °C for 1 day. After polymerization, the gels were cut into pieces of about 10 mm, and immersed in water for 48 h to wash out any soluble polymers, unreacted monomers and the initiator. Each sample was washed with 80 mL of water five times, and finally the washing solutions were collected all together (about 400 mL). Thereafter, the swollen gel samples were frozen in liquid nitrogen, and freeze dried in a Martin Christ, ALPHA 1–2LD device, 24 h, at 57 °C and 0.045 mbar. For the preparation of full-IPN cryogels by the sequential strategy, the just synthesized semi-IPN gels, cut into pieces of about 10 mm, have been immersed in a flask containing 0.6 mL ECH in 60 mL aqueous solution of 2 M NaOH, and kept at 22 °C for 24 h, followed by 2 h at 37 °C. After that, the gel pieces have been separated from the basic medium and intensively washed with distilled water up to neutral pH. The swollen gels have been freeze dried in the same conditions like the semi-IPN cryogels. The conversions in the synthesis of semi-IPN cryogels have been evaluated by the gel fraction yield (GFY). For this purpose, all the semi-IPN samples freeze dried have been further dried under vacuum in the presence of P2O5, until the constant weight has been reached. GFY was calculated with following equation:

GFYð%Þ ¼ ðW d =W m Þ  100

ð1Þ

where Wd is the weight of the dried sample; Wm the weight of the monomer, CS, and cross-linker used in synthesis. The fraction of CS removed from the composite cryogel, has been determined by the polyelectrolyte titration of the washing solutions with a standard aqueous solution of poly(ethylene sulfonate) (concentration of 103 M) by the particle charge detector PCD 03, Mütek GmbH, Herrsching, Germany. The percentage of CS, which left the semi-IPN hydrogel after extraction, has been calculated with following equation:

Pex ð%Þ ¼ ðm=XÞ  100

ð2Þ

where Pex is the percentage of linear polymer removed from the gel; m the amount of polymer removed from the whole gel sample,

determined by polyelectrolyte titration, g; X the amount of CS taken in the synthesis of the gel, which was 0.063 g in each experiment. Three values have been averaged. 2.3. Apparatus The structure of the cryogels has been investigated by FT-IR spectroscopy. The freeze-dried samples have been first frozen in liquid nitrogen, and then broken in a mortar to get the samples as white powder. FT-IR spectra were recorded with a Bruker Vertex FT-IR spectrometer, resolution 2 cm1, in the range of 4000– 400 cm1 by KBr pellet technique, the amount of the sample being about 5 mg in each pellet. Surface morphology and internal structure of the dried composite gels were observed using an Environmental Scanning Electron Microscope (ESEM) type Quanta 200, operating at 20 kV with secondary electrons, in low vacuum mode. The cross-sections of the samples were performed using a sharp blade to revel the internal structures. The average diameter of the pores and the thickness of the pore walls were determined from the SEM photographs using the image analyzing program ACD Photo Editor v3.1. 2.4. Swelling kinetics Swelling properties of the composite gels were studied using the conventional gravimetric procedure [47,48]. To evaluate the swelling kinetics, the dried gels were immersed in water at pH 5.5, and 25 °C. Swollen gels were weighed by an electronic balance, after wiping the excess surface liquid by filter paper. The swelling ratio (SR) was defined by following equation:

SR ¼ ðW t  W d Þ=W d ;

g g1

ð3Þ

where Wd is the weight (g) of the dried sample, and Wt is the weight (g) of the swollen sample, at time t. Equilibrium swelling ratio as a function of pH has been evaluated according to the protocol presented for the swelling kinetics, by varying the medium pH in the range 1–12, the contact duration being 24 h. 2.5. Adsorption of MB onto composite cryogels Adsorption kinetics of the cationic dye MB, on the composite cryogels have been followed using a batch equilibrium procedure. Thus, about 0.01 g of dried cryogel was placed in a flask and contacted with 10 mL of aqueous solution of the dye with a concentration of 6.4 mg/L, the initial solution pH being 5.5. The flasks containing the dye solution and the gel were placed in a shaking water bath at 25 °C. After a certain contact time, cryogels have

E.S. Dragan et al. / Chemical Engineering Journal 204–206 (2012) 198–209

201

Fig. 1. (A) SEM images of semi-IPN cryogels synthesized with CS1, with three cross-linker ratios, X, when pH of CS solution has been adjusted at 5, at a magnitude Mag = 1000; (B) SEM images of semi-IPN with the same cross-linker ratios, X, like above, when pH of CS solution has been adjusted at 6, at a magnitude Mag = 1000.

40

SR, g/g

30

20 s-IPN1.80.5 s-IPN1.60.5 s-IPN1.40.5 s-IPN1.20.5

10

0 0.0

0.1

0.2

0.3

0.4

0.5

0.6

Time, min Fig. 2. Swelling ratio (SR) as a function of contact time for semi-IPN composite cryogels prepared with four cross-linker ratios.

been filtered off and the residual concentration of the dye remained in the filtrate was measured by the UV–Vis spectroscopy at 665 nm. The amount of the dye bound on the composite cryogels was calculated with Eq. (4), in mg dye/g cryogel.

Adsorption capacity ¼

½ðC 0  CÞV ; W

mg=g

ð4Þ

where C0 and C are the concentrations of the dye in aqueous solution (mg/L) before and after the interaction with the composite gel, respectively, V is the volume of the dye aqueous solution (L), and W is the amount of the dried composite gel (g). For each adsorption experiments, the average of three replicates was reported. The desorption of MB from the IPN composite cryogels, after the kinetic study, has been easily performed by 0.1 M NaOH, about 30 min, followed by washing at neutral pH. For the study of the influence of the initial dye concentration on the sorption efficiency, the same amount of the gel (about 0.01 g)

has been equilibrated with dye solutions with concentration ranging from 6.4 mg/L up to 1279.4 mg/L, at 25 °C, the contact duration being fixed at 6 h. For the investigation of the reusability of the gel, an aqueous solution with a concentration of 480 mg/g has been used. The desorption of MB from the gel after one cycle of adsorption was performed as follows: (i) 0.1 M HCl, 1 h, followed by washing 2–3 times with distilled water; (ii) 3 times with 0.1 M NaOH, 1 h, followed by washing at neutral pH. Selective removal of MB from a mixture with MO has been followed using a batch equilibrium procedure. Thus, 40 mL of a mixture 1:1 (v/v) of an aqueous solution of each dye with a concentration of 16 mg/L for MB and 16.4 mg/L for MO has been added to about 0.01 g of the cryogel, previously swollen in water at pH 5.5, changing the supernatant with a new portion of the dyes solution when the color changed from green to yellow. The content of each dye in supernatant has been evaluated by UV–Vis spectroscopy (at 430 nm for MO and at 664 nm for MB). The separation experiment has been finished when the presence of both dyes has been simultaneously detected in supernatant. 3. Results and discussion The results obtained in this work are discussed in three subsections. In the first two subsections, the synthesis and characterization of semi- and full-IPN composite cryogels based on PAAm and CS are presented. In the third subsection the sorption and separation capacity of the composite cryogels against MB taken as a model dye has been investigated. 3.1. Synthesis and characterization of semi-IPN composite cryogels Semi-IPN composite cryogels based on PAAm and CS were synthesized by the cross-linking polymerization of AAm in the presence of CS, under the freezing point of the solvent (18 °C). Parameters which have been varied in the synthesis of semi-IPN cryogels were: the cross-linker ratio, X, the pH of CS solution and CS molar mass. The influence of these parameters on the polymer-

202

E.S. Dragan et al. / Chemical Engineering Journal 204–206 (2012) 198–209

Scheme 1.

Fig. 3. FTIR spectra of s-IPN2.60.6 (up) and IPN2.60.6 (down).

ization conversion (GFY), the fraction of CS which left the gel in the extraction steps, the gel morphology, and the swelling kinetics will be presented in this subsection. As Table 1 shows, the GFY of the sIPN1.80.5 composite cryogel has been 86.8% compared with that of PAAm cryogel with the same cross-linker ratio, taken as a reference, which was 95%. For higher cross-linker ratios, the conversion has been in the range (89–92)%. The decrease of GFY in the case of composite gels compared with PAAm cryogel is attributed to the polymerization in a multicomponent system. Table 1 also shows that the fraction of CS which left the composite gel by extraction decreased when the cross-linker ratio increased, suggesting that the CS chains have been better entangled in the matrix of PAAm when the density of cross-links increased. The second parameter which contributed to the increase of the CS fraction kept in the composite gel has been the pH of CS solution. It is known that the solubility of CS in water is strongly depending on the medium pH, decreasing when the pH is close to its pKa (6.5), CS being insoluble in water at higher pH. This specific behavior of CS prompted us to investigate the possibility to increase the content of CS trapped in the composite cryogel by

increasing the pH of CS solution from 4 (the pH corresponding to the solution of CS obtained by dissolving 1 wt.% CS powder in 1 v/v% acetic acid aqueous solution) to either 5 or 6. As Table 1 shows, the fraction of CS which has been removed from the semi-IPN decreased with the increase of pH from 5 to 6. The percentage of CS removed from the semi-IPN composite cryogels also decreased when the CS molar mass increased (samples s-IPN1.60.5 and s-IPN2.60.5, and s-IPN1.60.6 and s-IPN2.60.6, Table 1). Morphology of semi-IPN composite cryogels has been analyzed by SEM. The SEM images taken for three samples of cryogels prepared with CS1, having the pH adjusted at 5, different only by the cross-linker ratio, at a magnitude of 1000, are presented in Fig. 1A. A statistical analysis of the pore sizes of semi-IPN has been possible using the image analyzing program ACD Photo Editor v3.1, by measuring the diameter of at least 40 pores. Thus, the images in Fig. 1A show that the cryogel formed with a cross-linker ratio of 1/80 exhibits a heterogeneous morphology consisting of polyhedral pores with an average size of 67 ± 3 lm. With the increase of the cross-linker ratio, a uniform distribution of inter-connected pores has been observed for both X = 1/60 and X = 1/40, their average size being around 54 ± 3 lm, and 51 ± 2 lm, respectively. The thickness of the pore walls has been also evaluated being of around 3, 4, and 6 lm when X has been 1/80, 1/60 and 1/40, respectively. Fig. 1B displays the SEM images taken for the gels prepared with the same cross-linker ratio and CS molar mass but with a CS solution having the pH adjusted at 6. As can be observed, the average pore size decreased at 54 ± 4 lm, 40 ± 3 lm, and 33 ± 3 lm, when X has been 1/80, 1/60 and 1/40, respectively. As Fig. 1 shows, the main difference between the morphology of semi-IPN cryogels when pH increased from 5 to 6 is consisting of the decrease of the average pore size. Moreover, the pore walls seem more compact due to the CS chains which are more rigid, when the pH of the reaction mixture was close to its pKa (6.5). The swelling ratio (SR), calculated by Eq. (3), has been plotted as a function of time in Fig. 2, for semi-IPN composite cryogels obtained with four cross-linker ratios. As Fig. 2 shows, all semi-IPN cryogels are characterized by a superfast swelling, the equilibrium swelling state being attained in 2–3 s. The main difference between the cryogel samples consists of the equilibrium swelling ratio, which increased from 28 up to about 38 g water/g gel when X decreased from 1/20 to 1/80. The

E.S. Dragan et al. / Chemical Engineering Journal 204–206 (2012) 198–209

203

Fig. 4. Cartoon for the formation of the full-IPN PAAm/CS cryogel, having two oppositely charged networks, resulted by cross-linking of CS in alkaline conditions.

Fig. 5. SEM images of IPN1.40.6, IPN1.60.6 and IPN2.60.6 (Table 1): (up) the scaling bar 200 lm and magnitude Mag = 500, and (down) the scaling bar 50 lm and Mag = 1000.

semi-IPN cryogels present also the typical behavior for the macroporous morphologies with interconnected pores, i.e., the most part of water adsorbed being driven out by squeezing and sucked up very fast when the pressure is removed [22,31]. 3.2. Synthesis of full-IPN PAAm/CS composite cryogels From the results presented in Section 3.1, it was concluded that the gels of s-IPN1.60.6, s-IPN2-60.6 and s-IPN1.40.6 would be of interest for the preparation of full-IPN hydrogels, for two reasons: (i) the fraction of CS trapped in the composite gels has been higher compared with the gels prepared with CS solution of pH 5, and (ii) a uniform distribution of inter-connected pores has been evidenced by SEM in these gels. Generation of the second network was performed by the cross-linking of CS trapped in semi-IPN with ECH in aqueous solution of NaOH with a concentration of 8 w/v%. This reaction is selective because only the functional groups of CS (primary amine groups and ACH2OH) would react with ECH (Scheme 1A). On the other hand, part of the amide bonds in PAAm

can be hydrolyzed under alkaline conditions, the simultaneous generation of an anionic network being possible according to Scheme 1B. As it was previously shown [49,50], the alkaline hydrolysis of PAAm is nonreversible and self-retarded by the effect of the already formed neighboring carboxylate groups (COO), which reject the OH groups, the rate of hydrolysis decreasing thus with the increase of the hydrolysis degree. It was also found that, because of the neighboring group effect, the distribution of COO groups is more uniform instead of random, the idealized reaction path being presented in Scheme 1B. Information about the structure of the composite cryogels has been given by FT-IR analysis. The FT-IR spectrum of IPN2.60.6 composite cryogel is compared in Fig. 3 with the spectrum of the corresponding semi-IPN cryogel. The peaks observed at 1663 cm1 and 1320 cm1 in the FTIR spectrum of s-IPN2.60.6 have been attributed to amide I band, and to CAN stretching, respectively, in PAAm matrix. The characteristic bands of CS, usually located at: 1658 cm1 (amide I), 1598 cm1 (amide II), 1323 cm1 (N-acetylglucosamine), 1078

204

E.S. Dragan et al. / Chemical Engineering Journal 204–206 (2012) 198–209

160

A

B

C

SReq, g/g

120

SR, g/g

160 140 120 100 80 60 40 20 0 0.0

IPN2.60.6 IPN1.60.6

80 40 0

0.5

1.0 1.5 Time, min

2.0

s-IPN1.60.6 IPN1.60.6

0 1 2 3 4 5 6 7 8 9 10111213 pH

Fig. 6. (A) Swelling ratio (SR) as a function of contact time for two full-IPN cryogels; (B) equilibrium swelling ratio (SReq) as a function of pH for s-IPN1.60.6 cryogel and IPN1.60.6 cryogel; (C) FTIR spectra of the IPN1.60.6 cryogel after 24 h of incubation at pH 3 and pH 4.

and 1032 cm1 (stretching of CAO bonds in anhydroglucose ring), are difficult to identify in the spectrum of s-IPN2.60.6 cryogel due to the low content within the composite, part of CS being removed in the extraction steps (see Table 1). However, the presence of CS is clearly evidenced in the FT-IR spectrum of IPN2.60.6, the peaks at 1173 cm1 (antisymmetric stretching of the CAOAC bridge), 1083 cm1 and 1032 cm1, assigned to the skeletal vibration involving the stretching of CAO bonds in anhydroglucose units, being visible (Fig. 3). The explanation is that the CS chains have been first cross-linked and thus immobilized in the gel and after that the extraction has been performed. The new peaks at 1562 cm1 and at 1403 cm1 assigned to C@O in carboxylic acids, and COO groups, which were not observed in the s-IPN cryogels, show that part of the amide bonds in PAAm have been hydrolysed under alkaline conditions. The peak at 1674 cm1, assigned to the stretching vibration of C@O in amide groups, supports the presence of non-hydrolysed acrylamide units. It is worse to be mentioned that the spectra of the cryogels s-IPN1.60.6 and s-IPN1.40.6 were similar with the spectrum of s-IPN2.60.6 selected for Fig. 3, and the spectra of the gels IPN1.60.6 and IPN1.40.6 were similar with the spectrum of IPN2.60.6 discussed above. Therefore, the structure of the PAAm matrix, after the generation of the second network (cross-linked CS), would consist of acrylamide and acrylic acid sodium salt, thus having an anionic character, as Fig. 4 shows. The degree of hydrolysis has been roughly evaluated as the ratio between the absorbance of the peak at 1562 cm1 (A1562) and the sum of the absorbance at 1562 cm1 and at 1674 cm1 (A1562 + A1674), the values for the samples IPN1.60.6 and IPN2.60.6 being 62.5% and 54.7%, respectively. The values of the hydrolysis degree are in the same range with those reported by Zhao et al. for the double-network cryogels based on PNIPA and PAAm [50]. Based on the values of GFY presented in Table 1 for the corresponding semi-IPN, and on the values of the hydrolysis degree estimated by IR spectroscopy, the ratio between ACOOH and ANH2 groups has been roughly evaluated for the IPN1.60.6 and IPN2.60.6 composite gels, as follows: around 12:1 for the IPN1.60.6, and around 11:1 for IPN2.60.6. The difference is explained by the higher degree of hydrolysis found for the IPN1.60.6 than for IPN2.60.6. The morphology of full-IPN cryogels has been also examined by SEM. Fig. 5 presents the SEM images of three cryogels, at a magnitude of 500 (up) and 1000 (down), which support the morphological changes occurring by the formation of the second network in alkaline medium. As can be observed, there is a strong difference between the morphology of IPN1.40.6 gel and IPN1.60.6 gel. The decrease of the cross-linker ratio from 1/40 to 1/60, for the same molar mass of CS (CS1), conducted to about twice larger pores (average pore

size 34 ± 5 lm compared with 75 ± 4 lm), and to less compact pore walls, these being more accessible for the diffusion of low molecular weight species than those of IPN1.40.6. The influence of CS molar mass on the morphology of IPN cryogels can be observed by comparing the images corresponding to IPN1.60.6 with those corresponding to IPN2.60.6. The average size of macropores is comparable being 75 ± 5 lm for IPN1.60.6 gel and 78 ± 2 lm for IPN2.60.6 gel, but the pore walls seem to be more compact when the gel was prepared with CS2, Mv = 467 kDa, compared with that prepared with CS1 (Mv = 235 kDa). The average thickness of the pore walls evaluated from the SEM images was around 4– 6 lm for IPN1.60.6 and 12–14 lm for IPN2.60.6, i.e., increased with the increase of the CS molar mass. By the image analyzing program ACD Photo Editor v3.1 it was also possible to evaluate the size of the cavities which appeared in the pore walls of the IPN cryogels, these being around 25 lm for IPN1.60.6 and around 12 lm for IPN2.60.6. Based on these results, the IPN-1.40.6 gel was discarded and further experiments were conducted with IPN-1.60.6 and IPN2.60.6. Fig. 6A shows the swelling ratio as a function of time, at pH 5.5, for two full-IPN gel samples. The main differences between full-IPN and semi-IPN cryogels concerning the swelling kinetics (IPN1.60.6, Fig. 6A, compared with s-IPN1.60.6, Fig. 2) consists of the much higher values of the equilibrium swelling ratio (155 g/g compared with 33 g/g) and of the time necessary to reach the equilibrium swelling, which was about 45 s compared with 3 s. The increase of the time necessary to reach the equilibrium swelling is explained by the changes in the morphology of the cryogels and by the presence of two networks, which respond independently to environmental changes. The much higher swelling ratios of full-IPN cryogels are explained by the presence of the anionic matrix, which is bearing ACOO groups, known for their high hydrophilicity. Because both semi-IPN and full-IPN cryogels contain pH responsive components, it was of interest to compare their SR as a function of pH, the contact duration being kept constant. As Fig. 6B shows, the swelling behavior of the ionic composite cryogels strongly depended on the gel structure. Thus, the SR values found for s-IPN1.60.6 decreased when pH increased from 4 to 7, because the deprotonation of the amine groups in CS occurred, and therefore the hydrophilicity of the gel decreased. On the other hand, the full-IPN hydrogel (IPN1.60.6), having two independent networks responsive at pH, behaves completely different. At low pH (pH < 3) the SR values were around 10 g/g, but abruptly increased when pH increased up to 4. The SR values monotonously increased with the increase of pH up to 7, remained almost constant when the pH varied in the range 7–10, and decreased at pH > 11. The swelling feature at pH < 3 has been dominated by the cationic network based on CS, the carboxylic groups being less hydrophilic at

205

E.S. Dragan et al. / Chemical Engineering Journal 204–206 (2012) 198–209

1000

800

6

qe, mg/g

mg MB/g cryogel

8

Exp. data IPN1.60.6 Exp. data IPN2.60.6 Exp. data s-IPN1.60.6 Pseudo-first order model Pseudo-second order model

4

2

600

experimental values Langmuir isotherm Freundlich isotherm Sips isotherm

400

200 0 0

20

40

60

80

10 0

0

Time, min

0

2

3.3. Sorption of model ionic dyes onto composite cryogels As reported in literature, the ionic dyes are used as models for drugs in preliminary tests for controlled drug release from various supports [51,52]. Furthermore, finding novel, cheaper and highly efficient sorbents for the removal and separation of small ionic species like organic dyes from aqueous solution is one of the most stringent tasks [53–55]. The possibility to adsorb anionic dyes onto semi-IPN hydrogels and cationic dyes onto full-IPN hydrogels based on PAAm and CS, prepared at room temperature, has been already shown [9]. The porosity of those gels has been generated by the sublimation of water under freeze-drying conditions. The benefits of the macroporous gels synthesized under freezing conditions, having interconnected pores, is deeply investigated in this paper, taking MB as a model cationic dye. Fig. 7 illustrates the sorption kinetics of MB onto s-IPN1.60.6, IPN1.60.6 and IPN2.60.6 composite cryogels, the time required to achieve the equilibrium sorption of the dye being about 30 min, for both full-IPN cryogels. As can be observed, the amount of the dye sorbed at equilibrium on IPN gels was much higher than that sorbed on s-IPN1.60.6 cryogel. The strong interaction of IPN with MB is explained by the presence of the anionic charges, ACOO, at pH 5.5. Two kinetic models,

6

8

10

12

14

Ce, mg/L

Fig. 7. Plots of qt vs. t MB adsorption onto three composite cryogels, at 25 °C; sorbent dose about 0.01 g.

this pH. At pH > 4, the gel dramatically swells due to the electrostatic repulsion between ACOO groups. At pH > 10, the electrostatic screening effect of the OH ions became effective, causing the decrease of the SR values. To support the abrupt change of the ionization state in the gel when pH increased from 3 to 4, the FTIR spectra of the gel IPN1.60.6 dried after the incubation at these pH values for 24 h are compared in Fig. 6C. The pick at 1734 cm1 assigned to the C@O bonds in COOH groups can be observed only in the spectrum of the gel incubated at pH 3. The pick located at 1651 cm1 shows that amide groups remained in the PAAm matrix. In the spectrum of IPN1.60.6 incubated at pH 4, the following picks are visible: at 1674 cm1, assigned to the amide bonds (amide I), at 1559 cm1, and at 1403 cm1 assigned to C@O in carboxylate groups.

4

Fig. 8. Experimental isotherm of MB sorption onto IPN2.60.6 cryogel, at pH 5.5, 25 °C, contact duration 6 h, and the model isotherms obtained by the non-linear fit of Langmuir, Freundlich, and Sips isotherms.

i.e., the pseudo-first order model by Lagergren (Eq. (5)), and the pseudo-second order model (Eq. (6)) [56], have been used to evaluate the experimental adsorption data of MB onto semi-IPN and fullIPN composite gels.

dqt ¼ k1 ðqe  qt Þ dt

ð5Þ

dqt ¼ k2 ðqe  qt Þ2 dt

ð6Þ

where qe and qt represent the amounts of the dye adsorbed at equilibrium (mg/g) and at time t, respectively, k1 is the rate constant of pseudo-first order kinetic (min1), and k2 is the rate constant of pseudo-second order kinetic (g/(mg min)). The constants corresponding to the kinetic models are included in Table 2. As Table 2 shows, the theoretical qe,calc values estimated from the pseudo-first order model were very close to the experimental values for all composite gels, the correlation coefficient R2 being also high. These results show that the composite cryogels adsorb cationic dyes (MB) mainly by electrostatic interactions. In the adsorption process of a solute onto a solid surface, the solute adsorbed is in a dynamic equilibrium with the solute remained in solution. A plot of the solute concentration adsorbed onto the solid surface (mg/g) as a function of the solute concentration in solution at equilibrium (mg/L), at a constant temperature and pH, gives an adsorption isotherm, which can be described by some model adsorption isotherms. Fig. 8 shows that the retention capacity of the IPN2.60.6 for MB increased with the increase of the equilibrium dye concentration resulting in a concave curve, i.e., a type L isotherm according to the classification of Limousin et al. [57]. Analysis of equilibrium data by isotherm models is very important to compare different sorbents under different operational conditions and to design and optimize an operating procedure. Therefore, the relationship between the amount of MB sorbed onto

Table 2 Kinetic data for the adsorption of MB onto composite cryogels. Sample

s-IPN1.60.6 IPN1.60.6 IPN2.60.6

qe,exp (mg/g)

0.533 7.83 7.202

Pseudo-first order constants

Pseudo-second order constants

qe,calc (mg/g)

k1, (min1,102)

R2

qe,calc (mg/g)

k2 (g/mg  min,102)

R2

0.568 7.59 7.002

3.1 8.2 16.04

0.994 0.993 0.994

0.769 8.56 7.32

3.51 1.41 5.83

0.996 0.997 0.997

206

E.S. Dragan et al. / Chemical Engineering Journal 204–206 (2012) 198–209

Table 3 Langmuir, Freundlich and Sips parameters for the MB sorption onto IPN2.60 composite cryogel. Langmuir isotherm parameters

Freundlich isotherm parameters

Sips isotherm parameters

qm, mg/g

1006.7

KF, mg/g

285.5

0.353 0.949 4877

N R2

0.452 0.8643 13012

qm, mg/g aS N R2

3

KL  10 , L/mg R2

v2

v2

v2

755.45 0.3 2.2 0.9964 388

Table 4 Comparison of maximum sorption capacities (qm, mg/g) for MB of different sorbents. Sorbent

Sorbent dose, g

pH T, °C qm, mg/g Refs.

Alginate/polyaspartate hydrogel gel beads Arabic gum modified with glycidyl methacrylate, polyacrylate, and PAAm PAMPS/chitosan hydrogel Alginate beads containing magnetic nanoparticles and activated carbon Sepiolite Regenerated activated carbon from spent catalyst IPN2.60.6

0.1

5.5 25

600–700 [53]

0.05

8

48

25

[54]

qe ¼ Not specified 10 25 6.6 6.7 25

74 290

[60] [61]

2.5 0.1

5 7

30 20

57 425

[62] [63]

0.01

5.5 25

750

This study

IPN2.60.6 composite cryogel and the dye concentrations at equilibrium has been described by Langmuir, Freundlich, and Sips sorption isotherm models. Langmuir isotherm is a well-known isotherm model applicable to homogeneous adsorption based on the following assumptions: (i) all the adsorption sites are identical; (ii) each site retains one molecule of the given compound; (iii) all sites are energetical and sterical independent of the adsorbed quantity [58]. The non-linear form of the Langmuir isotherm is described by following equation.

qe ¼

qm K L C e 1 þ K LCe

ð7Þ

where qm is the saturated monolayer sorption capacity (mg/g), KL is the Langmuir constant (L/mg) related to the energy of adsorption, which reflects the affinity between the sorbent and sorbate. The value of the constant KL gives indication on the affinity of MB for the composite cryogel, and the value of qm correspond to the maximum sorption capacity of the gel. The values of KL and qm obtained by the non-linear fit of the isotherm are presented in Table 3. As Table 3 shows, the maximum adsorption capacity of MB estimated from the Langmuir model (qm) is 1006.7 mg/g, being higher than the experimental value (qe,exp = 749.7 mg/g), and the correlation coefficient R2 is 0.949. These results show that the sorption process of MB onto the composite cryogel could be only approximated by the Langmuir isotherm model. Freundlich isotherm assumes heterogeneous surface with a nonuniform distribution of heat of adsorption, this isotherm being described by following equation [59]:

qe ¼ K F C Ne

If N < 1, bond energies increase with the surface density; if N > 1, bond energies decrease with the surface density, and when N = 1, all surface sites are equivalent. The values of Freundlich isotherm constants (KF and N) determined by the non-linear fit of Eq. (8) on the experimental data are also presented in Table 3. As Table 3 shows, the value of N was much lower than 1 and this indicated an unfavorable sorption of MB onto the composite cryogel. Furthermore, the value of correlation coefficient (R2) was low (0.8643) and this shows that Freundlich model is not appropriate to describe the experimental data. Sips isotherm is a combination of the Langmuir and Freundlich isotherm type models and is expected to describe much better the heterogeneous surfaces. At low sorbate concentrations it reduces to a Freundlich isotherm, while at high sorbate concentrations it predicts a monolayer adsorption capacity characteristic for the Langmuir isotherm [57]. The Sips isotherm is described by following equation:

ð8Þ

where KF, Freundlich constant, which predicts the quantity of dye per gram of composite at the equilibrium concentration; N, a measure of the nature and strength of the adsorption process and of the distribution of active sites.

qm as C Ne

ð9Þ

1 þ as C Ne

where qm is the monolayer adsorption capacity (mg/g) and aS is the Sips constant related to energy of adsorption. The values of Sips isotherm constants (qm and aS) determined by the non-linear fit of Eq. (9) are included in Table 3. According to Sips model, the monolayer sorption capacity values, qm, are very close to the experimental sorption capacity values, qe,exp (Table 3). Determining the best-fitting model is a key analysis to mathematically describe the involved sorption system and, therefore, to explore the related theoretical assumptions. Hence, several error calculation functions have been widely used to estimate the error deviations between experimentally and theoretically predicted equilibrium adsorption values [57]. In this study, two statistical functions have been used to investigate their applicability as suitable tools to evaluate isotherm models fitness, namely the correlation coefficient of determination (R2) and the non-linear Chisquare test (v2). The v2 test statistic is basically the sum of the squares of the differences between the experimental data (qe,exp, mg/g) and the data obtained by calculating from models (qe,cal, mg/g), with each squared difference divided by the corresponding data calculated using the models. This can be mathematically represented by following equation:

v2 ¼

X ðqe;exp  qe;calc Þ2 qe;calc

ð10Þ

If the data from a model are similar to the experimental data, v2 will be a small number, and if they strongly differ, v2 will be a big number. The results of the application of correlation coefficients (R2) and non-linear Chi-square test (v2) on the experimental data of the equilibrium capacity (qe,exp) are shown in Table 3. The Sips isotherm model appears to be the best fitting model for the sorption of MB onto the IPN composite cryogel, with the highest correlation coefficient (R2 = 0.9964), and the lowest Chi-square values (v2 = 388). Numerous sorbents have been used to remove MB from aqueous solutions [53,54,60–63]. A brief list of the maximum sorption capacity of various sorbents, including the results obtained in this work, is presented in Table 4. As Table 4 shows, the most part of the materials tested for their sorption capacity for MB have been anionic composite hydrogels, the interaction with the cationic dyes being electrostatic in nature [53,54,60–62]. Thus, a high value for the maximum adsorption capacity of 600–700 mg/g has been reported by Jeon et al. for the sorption of MB on the alginate/polyaspartate composite hydrogels as beads, both polymers containing carboxylic acid groups [53]. Paulino et al. have investigated the sorption of MB on the superab-

E.S. Dragan et al. / Chemical Engineering Journal 204–206 (2012) 198–209

400

qe, mg MB/g cryogel

350 300 250 200 150 100 50 0 0

1

2

3

4

5

Number of sorption/desorption cycles Fig. 9. Dye sorption capacity of IPN2.60.6 cryogel as a function of the number of sorption/desorption cycles: sorbent dose = 0.015 g, initial dye concentration = 480 mg/L, contact time 6 h at 25 °C.

Fig. 10. Selective sorption of MB from a 1:1 mixture with anionic dye MO; left image: the cryogel sample before the contact with the solution of MB and MO; middle image: the monolith sample after the selective sorption of MB; right image: the cross-section in monolith after the selective retention of MB.

sorbent and biodegradable hydrogels derived by grafting polyacrylate and polyacrylamide on the arabic gum [54]. The maximum sorption capacity has been reported for the initial dye concentration of 50 mg/L. The selective separation of MB from the mixture with orange II, a nonionic dye, has been also investigated and supported the electrostatic interaction between cationic centers of the dye and COO groups of the sorbent as the main driving of the sorption [54]. The adsorption capacity of MB onto the inorganic sorbents is also significant being 57 mg/g in the case of sepiolite [62], and 425 mg/g in the case of a regenerated activated carbon used as catalyst support [63]. The maximum sorption capacity of the IPN2.60.6 composite gel for MB, included in Table 4, shows that it could be considered a suitable sorbent for the removal of MB in aqueous solution. The preservation of the sorption capacity during the multiple sorption/desorption cycles is an important characteristic of the sorbents used in the removal of contaminants from the wastewaters. The reusability of the IPN2.60.6 composite cryogel is supported by the results presented in Fig. 9. As Fig. 9 shows, the sorption capacity of the IPN2.60.6 composite cryogel for MB has been almost constant after four cycles of dye sorption/desorption, and this support the high chemical and mechanical resistance of the macroporous and multicomponent hydrogel. Separation of the cationic dye, MB, from the mixture with the anionic dye, MO, has been investigated by the batch technique.

207

The concentration of MB and MO simultaneously present in aqueous solution is limited by their tendency to form insoluble complex at concentrations higher than 104 M. Therefore, the concentration of the dyes in the aqueous solution containing MB and MO, at a molar ratio of 1:1, has been 16 mg/L for MB and 16.4 mg/L for MO. Initially, the solution had a green color, reflecting the simultaneous presence of both dyes, and yellow after the removal of MB. The adsorption of MO was negligible, as the UV–Vis spectra of the solution containing both dyes, recorded before and after the adsorption, show (see Supplementary information). Moreover, the absorbance of MO increased after the sorption of MB, and this could be attributed to the partial complexation between the oppositely charged dyes MB and MO by their mixture, at this concentrations. This phenomenon has been not observed when the concentrations of both dyes have been two times lower (results not shown here). The capacity of the IPN2.60.6 gel to adsorb MB from its mixture with MO has been leveled off at about 150 mg MB/g gel, and the time necessary to attain the equilibrium of MB sorption for each separation step has been much longer at the concentrations selected for this experiment, compared with the dye sorption from the solution containing only MB. Fig. 10 shows the cryogel before adsorption (left) and after its charge with 320 mg MB/g gel (in the middle). The cross-section of the gel after the selective sorption of MB is also presented (on the right). As can be observed in Fig. 10 (right image), MB has been uniformly distributed in the interior of the monolith, and this also support the uniform distribution of the active sites in the gel. The capacity for selective sorption of cationic dyes and the high reusability of the IPN cyogels based on PAAm and CS recommend them as an alternative for other composite sorbents endowed with ion exchange properties, with potential applications in bioseparation or for drug delivery systems.

4. Conclusions Macroporous semi-IPN composite cryogels as rods, based on PAAm as a matrix and CS as trapped natural polycation, have been prepared in the paper. Full-IPN cationic/anionic composite cryogels have been designed by the selective cross-linking with ECH of CS chains trapped in the matrix of PAAm, under alkaline conditions, when a partial hydrolysis of amide bonds simultaneously occurred. The morphology of semi-IPN composite cryogels has been strongly influenced by the cross-linker ratio and by the pH of the CS solution. Thus, the cryogel formed with CS1 solution of pH 5, at a cross-linker ratio of 1/80, exhibited a heterogeneous morphology consisting of polyhedral pores with an average size of 67 ± 3 lm. A uniform distribution of the inter-connected pores has been observed for both X = 1/60 and X = 1/40, their average size being around 54 ± 3 lm, and 51 ± 2 lm, respectively. The average pore size decreased at 54 ± 4 lm, 40 ± 3 lm, and 33 ± 3 lm, when X has been 1/80, 1/60 and 1/40, respectively, and the pH of CS solution increased at 6. Formation of the second network strongly influenced the equilibrium swelling ratio of the full-IPN as a function of pH and the gel morphology. Also, the generation of an anionic matrix during the formation of the second network led to a high sorption capacity of MB by the full-IPN cryogel, the maximum sorption capacity being around 750 mg/g cryogel, the theoretical sorption capacity estimated by fitting Sips model being 755.5 mg MB/g cryogel. Furthermore, the full-IPN cryogels are endowed with a high reusability and showed excellent properties in separation of MB from a 1:1 mixture with MO. These sorption properties of the full-IPN composite cryogels recommend them as an alternative for other composite sorbents in the removal and separation of small ionic species like organic dyes from aqueous solution.

208

E.S. Dragan et al. / Chemical Engineering Journal 204–206 (2012) 198–209

Acknowledgement This work was supported by CNCSISUEFISCSU by the project PN-II-ID-PCE-2011-3-0300.

Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.cej.2012.07.126. References [1] J. Berger, M. Reist, J.M. Mayer, O. Felt, N.A. Peppas, R. Gurny, Structure and interactions in covalently and ionically crosslinked chitosan hydrogels for biomedical applications, Eur. J. Pharm. Biopharm. 57 (2004) 19–34. [2] N.A. Peppas, J.Z. Hilt, A. Khademhosseini, R. Langer, Hydrogels in biology and medicine: from molecular principles to bionanotechnology, Adv. Mater. 18 (2006) 1345–1360. [3] R.A.A. Muzzarelli, Genipin-crosslinked chitosan hydrogels as biomedical and pharmaceutical aids, Carbohydr. Polym. 77 (2009) 1–9. [4] F. Liu, M.W. Urban, Recent advances and challenges in designing stimuliresponsive polymers, Prog. Polym. Sci. 35 (2010) 3–23. [5] T. Thimma Reddy, A. Takahara, Simultaneous and sequential micro-porous semi-interpenetrating polymer network hydrogel films for drug delivery and wound dressing applications, Polymer 50 (2009) 3537–3546. [6] X. Zeng, W. Wei, X. Li, J. Zeng, L. Wu, Direct electrochemistry and electrocatalysis of hemoglobin entrapped in semi-interpenetrating polymer network hydrogel based on polyacrylamide and chitosan, Bioelectrochemistry 71 (2007) 135–141. [7] S. Liang, L. Liu, Q. Huang, K.L. Yam, Preparation of single or double-network chitosan/poly(vinyl alcohol) gel films through selectively cross-linking method, Carbohydr. Polym. 77 (2009) 718–724. [8] E.S. Dragan, D.F. Apopei, Synthesis and swelling behavior of pH-sensitive semiinterpenetrating polymer network composite hydrogels based on native and modified potatoes starch as potential sorbent for cationic dyes, Chem. Eng. J. 178 (2011) 252–263. [9] E.S. Dragan, M.M. Perju, M.V. Dinu, Preparation and characterization of IPN composite hydrogels based on polyacrylamide and chitosan and their interaction with ionic dyes, Carbohydr. Polym. 88 (2012) 270–281. [10] M.M. Ozmen, O. Oguz, Formation of macroporous poly(acrylamide) hydrogels in DMSO/water mixture: Transition from cryogelation to phase separation copolymerization, React. Funct. Polym. 68 (2008) 1467–1475. [11] X.S. Wu, A.S. Hoffman, P. Yager, Synthesis and characterization of thermally reversible macroporous poly(N-isopropylacrylamide) hydrogels, J. Polym. Sci. Polym. Chem. 30 (1992) 2121–2129. [12] T. Serizawa, K. Wakita, T. Kaneko, M. Akashi, Thermoresponsive properties of porous poly(N-isopropylacrylamide) hydrogels prepared in the presence of nanosized silica particles and subsequent acid treatment, J. Polym. Sci. Polym. Chem. 40 (2002) 4228–4235. [13] K. Varaprasad, N. Narayana Reddy, S. Ravindra, K. Vimala, K. Mohana Raju, Synthesis and characterizations of macroporous poly(acrylamide-2acrylamido-2-methyl-1-propanesulfonic acid) hydrogels for in vitro drug release of ranitidine hydrochloride, Int. J. Polym. Mat. 60 (2011) 490–503. [14] M. Pradny, P. Lesny, J. Fiala, J. Vacik, M. Sloui, J. Michalek, E. Sykova, Macroporous hydrogels based on 2-hydroxyethyl metacrylate. Part 1. Copolymers of 2-hydroxyethyl metacrylate with methacrylic acid, Collect. Czech. Chem. Commun. 68 (2003) 812–822. [15] E. Behravesh, S. Jo, K. Zygourakis, A.G. Mikos, Synthesis of in situ cross-linkable macroporous biodegradable poly(propylene fumarate-co-ethylene glycol) hydrogels, Biomacromolecules 3 (2002) 374–381. [16] L. Shapiro, S. Cohen, Novel alginate sponges for cell culture and transplantation, Biomaterials 18 (1997) 583–590. [17] E.S. Dragan, M. Cazacu, A. Nistor, Ionic organic/inorganic materials. III. Stimuli responsive hybrid hydrogels based on oligo (N,Ndimethylaminoethylmethacrylate) and chloroalkyl-functionalized siloxanes, J. Polym. Sci. Part A Polym. Chem. 47 (2009) 6801–6813. [18] V.L. Lozinsky, F.M. Plieva, I.Y. Galaev, B. Mattiasson, The potential of polymeric cryogels in bioseparation, Bioseparation 10 (2001) 163–188. [19] I.Y. Galaev, Smart polymers in biotechnology and medicine, Russ. Chem. Rev. 64 (1995) 471–489. [20] P. Perez, F. Plieva, A. Gallardo, J.S. Roman, M.R. Aguilar, I. Morfin, F. EhrburgerDolle, F. Bley, S. Mikhalovsky, I.Yu. Galaev, B. Mattiasson, Bioresorbable and nonresorbable macroporous thermosensitive hydrogels prepared by cryopolymerization. role of the cross-linking agent, Biomacromolecules 9 (2008) 66–74. [21] A. Tripathi, A. Kumar, Multi-featured macroporous agarose–alginate cryogel: synthesis and characterization for bioengineering applications, Macromol. Biosci. 11 (2011) 22–35. [22] M.V. Dinu, M.M. Ozmen, E.S. Dragan, O. Okay, Freezing as a path to build macroporous structures: superfast responsive polyacrylamide hydrogels, Polymer 48 (2007) 195–204.

[23] G. Baydemir, N. Bereli, M. Andac, R. Say, I.Y. Galaev, A. Denizli, Bilirubin recognition via molecularly imprinted supermacroporous cryogels, Colloid Surf. B Biointerfaces 68 (2009) 33–38. [24] T.V. Burova, N.V. Grinberg, E.V. Kalinina, R.V. Ivanov, V.I. Lozinsky, C.A. Lorenzo, V.Y. Grinberg, Thermoresponsive copolymer cryogel possessing molecular memory: synthesis, energetics of collapse and interaction with ligands, Macromol. Chem. Phys. 212 (2011) 72–80. [25] H. Kirsebom, D. Topggard, I.Yu. Galaev, B. Mattiasson, Modulating the porosity of cryogels by influencing the nonfrozen liquid phase through the addition of inert solutes, Langmuir 26 (2010) 16129–16133. [26] S. Van Vlierberghe, P. Dubruel, E. Schacht, Biopolymer-based hydrogels as scaffolds for tissue engineering applications: a review, Biomacromolecules 12 (2011) 1387–1408. [27] E. Jain, A.A. Karande, A. Kumar, Supermacroporous polymer-based cryogel bioreactor for monoclonal antibody production in continuous culture using hybridoma cells, Biotechnol. Prog. 27 (2011) 170–180. [28] D.C. Tuncaboylu, O. Okay, Hierarchically macroporous cryogels of polyisobutylene and silica nanoparticles, Langmuir 26 (2010) 7574–7581. [29] M.V. Dinu, M.M. Perju, E.S. Dragan, Porous semi-interpenetrating hydrogel networks based on dextran and polyacrylamide with superfast responsiveness, Macromol. Chem. Phys. 212 (2011) 240–251. [30] N. Orakdogen, P. Karacan, O. Okay, Macroporous, responsive DNA cryogel beads, React. Funct. Polym. 71 (2011) 782–790. [31] M.V. Dinu, M.M. Perju, E.S. Dragan, Composite IPN ionic hydrogels based on polyacrylamide and dextran sulfate, React. Funct. Polym. 71 (2011) 881–890. [32] Q. Zhao, J. Sun, X. Wu, Y. Lin, Macroporous double-network cryogels: formation mechanism, enhanced mechanical strength and temperature/pH dual sensitivity, Soft Matter 7 (2011) 4284–4293. [33] Q. Zhao, J. Sun, Q. Liang, Q. Zhou, Synthesis of macroporous thermosensitive hydrogels: a novel method of controlling pore size, Langmuir 25 (2009) 3249– 3254. [34] K. Tekin, L. Uzun, C.A. S ß ahin, S. Bektasß, A. Denizli, Preparation and characterization of composite cryogels containing imidazole group and use in heavy metal removal, React. Funct. Polym. 71 (2011) 985–993. [35] M. Liu, H. Liu, L. Bai, Y. Liu, J. Cheng, G. Yang, Temperature swing adsorption of melamine on thermosensitive poly(N-isopropylacrylamide) cryogels, J. Mater. Sci. 46 (2011) 4820–4825. [36] S. Hajizadeh, H. Kirsebom, I.Y. Galaev, B. Mattiasson, Evaluation of selective composite cryogel for bromate removal from drinking water, J. Sep. Sci. 33 (2010) 1752–1759. [37] Y.-Q. Xia, T.-Y. Guo, M.-D. Song, B.-H. Zhang, B.-L. Zhang, Hemoglobin recognition by imprinting in semi-interpenetrating polymer network hydrogel based on polyacrylamide and chitosan, Biomacromolecules 6 (2005) 2601–2606. [38] M.V. Risbud, R.R. Bhonde, Polyacrylamide-chitosan hydrogels: in vitro biocompatibility and sustained antibiotic release studies, Drug Deliv. 7 (2000) 69–75. [39] S. Ekici, D. Saraydin, Synthesis, characterization and evaluation of IPN hydrogels for antibiotic release, Drug Deliv. 11 (2004) 381–388. [40] M.J. Zohuriaan-Mehr, H. Omidian, S. Doroudiani, K. Kabiri, Advances in nonhygienic applications of superabsorbent hydrogel materials, J. Mater. Sci. 45 (2010) 5711–5735. [41] M. Rinaudo, Main properties and current applications of some polysaccharides as biomaterials, Polym. Int. 57 (2008) 397–430. [42] H. Chavda, C. Patel, Chitosan superporous hydrogel composite-based floating drug delivery system: A newer formulation approach, J. Pharm. Bioall. Sci. 2 (2010) 124–131. [43] P. Bonina, T. Petrova, N. Manolova, PH-Sensitive hydrogels composed of chitosan and polyacrylamide-preparation and properties, J. Bioact. Compat. Polym. 19 (2004) 101–116. [44] K. Varaprasad, N. Narayana Reddy, N. Mithil Kumar, K. Vimala, S. Ravindra, K. Mohana Raju, Poly(acrylamide-chitosan) hydrogels: interaction with surfactants, Int. J. Polym. Mater. 59 (2010) 981–993. [45] C. Zhou, Q. Wu, A novel polyacrylamide nanocomposite hydrogel reinforced with natural chitosan fibers, Colloids Surf. B: Biointerfaces 84 (2011) 155–162. [46] E.S. Dragan, M.V. Dinu, D. Timpu, Preparation and characterization of novel composites based on chitosan and clinoptilolite with enhanced adsorption properties for Cu2+, Bioresour. Technol. 101 (2010) 812–817. [47] B. Kim, K. La Flamme, N.A. Peppas, Dynamic swelling behavior of pH-sensitive anionic hydrogels used for protein delivery, J. Appl. Polym. Sci. 89 (2003) 1606–1613. [48] R. Zhuo, W. Li, Preparation and characterization of macroporous poly(Nisopropylacrylamide) hydrogels for the controlled release of proteins, J. Polym. Sci. Part A Polym. Chem. 41 (2003) 152–159. [49] M. Ilavsky, J. Hrouz, J. Stejskal, K. Bouchal, Phase transition in swollen gels. 6. Effect of aging on the extent of hydrolysis of aqueous polyacrylamide solutions and on the collapse of gels, Macromolecules 17 (1984) 2868–2874. [50] Q. Zhao, J. Sun, Y. Lin, Q. Zhou, Study of the properties of hydrolyzed polyacrylamide hydrogels with various pore structures and rapid pHsensitivities, React. Funct. Polym. 70 (2010) 602–609. [51] B. Kim, Y. Shin, PH-sensitive swelling and release behaviors of anionic hydrogels for intelligent drug delivery system, J. Appl. Polym. Sci. 105 (2007) 3656–3661. [52] A.J. Chung, M.F. Rubner, Methods of loading and releasing low molecular weight cationic molecules in weak polyelectrolyte multilayer films, Langmuir 18 (2002) 1176–1183.

E.S. Dragan et al. / Chemical Engineering Journal 204–206 (2012) 198–209 [53] Y.S. Jeon, J. Lei, J.-H. Kim, Dye adsorption characteristics of alginate/ polyaspartate hydrogels, J. Ind. Eng. Chem. 14 (2008) 726–731. [54] A.T. Paulino, M.R. Guilherme, A.V. Reis, G.M. Campese, E.C. Muniz, J. Nozaki, Removal of methylene blue dye from an aqueous media using superabsorbent hydrogel supported on modified polysaccharide, J. Colloid Interface Sci. 301 (2006) 55–62. [55] P.C. Thomas, B.H. Cipriano, S.R. Raghavan, Nanoparticle-crosslinked hydrogels as a class of efficient materials for separation and ion exchange, Soft Matter 7 (2011) 8192–8197. [56] Y.S. Ho, G. McKay, Sorption of dye from aqueous solution by peat, Chem. Eng. J. 70 (1998) 115–124. [57] G. Limousin, J.P. Gaudet, L. Charlet, S. Szenknect, V. Barthes, M. Krimissa, Sorption isotherms: a review on physical bases, modeling and measurement, Appl. Geochem. 22 (2007) 249–275. [58] I. Langmuir, The adsorption of gases on plane surfaces of glass, mica and platinum, J. Am. Chem. Soc. 40 (1918) 1361–1403.

209

[59] H.M.F. Freundlich, Uber die adsorption in lösungen, Z. Phys. Chem. 57 (1906) 385–470. [60] Y.H. Gad, Preparation and characterization of poly(2-acrylamido-2methylpropanesulfonic acid)/Chitosan hydrogel using gamma irradiation and its application in wastewater treatment, Rad. Phys. Chem. 77 (2008) 1101– 1107. [61] V. Rocher, J.M. Siaugue, V. Cabuil, A. Bee, Removal of organic dyes by magnetic alginate beads, Water Res. 42 (2008) 1290–1298. [62] M. Dog˘an, Y. Özdemir, M. Alkan, Adsorption kinetics and mechanism of cationic methyl violet and methylene blue dyes onto sepiolite, Dyes Pigm. 75 (2007) 701–713. [63] Z. Zhang, Z. Zhang, Y. Fernandez, J.A. Menendez, H. Niu, J. Peng, L. Zhang, S. Guo, Adsorption isotherms and kinetics of methylene blue on a low-cost adsorbent recovered from a spent catalyst of vinyl acetate synthesis, Appl. Surf. Sci. 256 (2010) 2569–2576.