Synthesis of a novel pH responsive phyllosilicate loaded polymeric hydrogel based on poly(acrylic acid-co-N-vinylpyrrolidone) and polyethylene glycol for drug delivery: modelling and kinetics study for the sustained release of an antibiotic drug

Abstract

In this study, we developed a novel pH-sensitive composite interpenetrating polymeric network (IPN) hydrogel based on polyethylene glycol (PEG) and poly(acrylic acid-co-N-vinylpyrrolidone) crosslinked with N,N-methylenebisacrylamide (MBA). This composite was used for the controlled release (CR) of cefadroxil, an antibiotic drug. A systematic method via in situ polymerization in sodium aluminosilicate dispersion media was also performed to achieve a much higher degree of swelling behaviour followed by sufficient gel strength in the simulated pH atmosphere. The resulting hydrogel imprinted was characterized by Fourier transform infrared spectroscopy (FTIR) to confirm the copolymer formation and cross linking reaction, and scanning electron microscopy (SEM) to understand the surface morphology. Differential thermal analysis thermogravimetric analysis (DTA-TGA) and X-ray diffraction (XRD) were also performed to investigate the deviations from crystallinity and swelling experiments. The in vitro release of the drug loaded hydrogel performed in the acidic and basic media affected the drug release characteristics. The release data was analysed using an empirical equation to understand the transport of a drug-containing solution through the polymeric matrices. The wt% of PEG, MBA, initiator, total monomer concentration, pH of the medium was found to strongly influence the drug release behaviour of the gels. The impression of drug loading on the encapsulation efficiency was also investigated. The release rate of the drug was much faster at pH 7.8 than at pH 1.7. The modelling and kinetics of sustained release of antibiotic is reported.

---

1. Introduction

Hydrogels are one of the forthcoming families of polymer-based sustained-release drug delivery systems. Besides displaying swelling-controlled drug release, hydrogels also show a stimuli responsive deformational attitude in their morphological network and the drug release. Because of the large variations in the physiological pH at various body sites under normal and pathological simulated conditions, pH-responsive polymeric networks have been extensively studied.^1,2 PEG hydrogels have been widely explored as water-soluble biocompatible, non-toxic, non-immunogenic and bioadhesive polymers for numerous biomedical and pharmaceutical applications.³ A wide range of material properties can be obtained by tuning the PEG and crosslinker compositions. Therefore, hydrogels based on PEG with different cross-linked structures create unique opportunities for controlling pH sensitive drug carriers and designing tissue engineering scaffolds. For free radical solution polymerization, the total monomer concentration in water should be at least 15–20 wt% but it is difficult to make an aqueous solution. In the present work, PEG was incorporated in a poly(acrylic acid-co-vinylpyrrolidone) based gel by in situ incorporation during simultaneous crosslinking between acrylic acid and MBA or N-vinylpyrrolidone and MBA in water. Acrylic acid and N-vinylpyrrolidone are both water soluble monomers and were chosen because its copolymer is a pH responsive polyelectrolyte that is reliable for drug delivery to specific sites of the gastrointestinal tract.⁴ Hydrogels based on N-vinylpyrrolidone (NVP) as a co-monomer have been applied successfully as local dressings on wound treatments such as burns, skin ulceration and postoperative dressings.⁵ In recent years, polymer/layered silicate (PLS) nanocomposites have attracted considerable interest, both in industry and in academia, because they often exhibit remarkable improvement in the material properties compared to the virgin polymer or conventional micro, nano and their composites.^6–9 The intercalation chemistry of polymer-layered silicate composites is practiced tremendously in the context of composite hydrogel development.^10–12

The aim of the study was to develop a hydrogel with reasonable biocompatibility and responsiveness to external stimuli. Therefore, N-vinylpyrrolidone (NVP) was used as an essential material of gel-forming. The synthesis of a hydrogel followed two steps. In step one, the crosslinked gel bead was synthesized through a chemical reaction. In the second step, the semi-IPN composite hydrogel was synthesized based on hydrophilic clay reinforcement stabilized through the interaction among the carboxylic acid groups, micropores and the interstices between the two networks. The copolymer formation and crosslinking reaction of the hydrogel was investigated by FTIR and swelling experiments. The morphology and crystallinity of the hydrogel was investigated by SEM, DTA-TGA and XRD. The in vitro release of the drug loaded hydrogel results performed in acidic and basic media affected the drug release characteristics. The release data was analysed using an empirical equation to understand the transport of the drug containing solution through the polymeric matrices. The wt% of PEG, MBA, initiator, total monomer concentration, pH of the medium was found to strongly influence the drug release behaviour of the gels. The modelling and kinetics for the sustained release of the antibiotic drug was studied.

2. Experimental methods

2.1. Materials

Polyethylene glycol 4000 (PEG, Merck) was used as received. The analytical grade co-monomer crosslinking agent, N,N′-methylenebisacrylamide (MBA, from Merck), and the redox pair of initiators, ammonium persulfate (APS, from Fluka) and sodium metabisulfite, were used as received. The monomer acrylic acid (AA, Merck) and N-vinylpyrrolidone (NVP, Merck Schuchardt OHG, Germany) were used after vacuum distillation. Hydrophilic sodium montmorillonite aluminosilicate filler was used after drying for 2 hours in an oven at 110 °C (rich nano size = 30–90 nm, aspect ratio = 300–500, mineral thickness = 1 nm, and cation exchange capacity = 120 m_eq./100 g). The specification of this filler is reported elsewhere;¹³ bentonite filler was supplied by Amrfeopte. Ltd., Kolkata. The drug cefadroxil was provided by Aristo Pharmaceuticals Pvt. Ltd., Solan, U.P.

2.2. Methods

2.2.1. Synthesis of composite hydrogel by in situ filler incorporation. Initially, poly(acrylic acid-co-vinyl pyrrolidone) gels were prepared at various initiators, total monomer concentrations, crosslinking agents, MBA concentrations, and PEG concentrations in a three necked reactor placed in a constant temperature bath and fitted with a stirrer, a thermometer pocket and a condenser at 60 °C. To prepare the composite gel, solutions at various concentrations, i.e., 1.0, 2.0, 3.0 and 4.0 wt% of PEG, were prepared in deionised water. The required amount of PEG solution and monomers (AA : NVP = 10 : 1) was then poured into the reactor. The temperature was maintained at 60 °C and an aqueous solution of the initiators was added to the reactor followed by the addition of MBA (crosslinker). After polymerization, the reaction mixture was cooled to ambient temperature. The hydrophilic aluminosilicate (zeolyte) filler was first dispersed in water for one an hour followed by the addition of monomers and stirring for another half an hour. Polymerization with in situ mixing of the filler with these monomers was then carried out in a similar manner as in the case of polymerization for the unfilled (without any filler) gel. In another part of the experiment, AA was partially neutralised with NaOH and then this monomer was copolymerized with NVP in the presence of PEG according to the previous method. Hydrophilic clay (nanobentonite) was also used to optimise the swelling and gel strength. The hydrogel obtained was disintegrated into small blocks and then immersed into a 1 : 3 v/v ethanol–water solution for 48 hours to remove the water soluble oligomer, unreacted monomers and the initiator fractions.

2.2.2. Yield, gel and sol content of the hydrogel. The hydrogels prepared above were first dried to a constant weight (W_c) in a vacuum oven and then it was taken in water and kept for one week with occasional shaking to remove the water soluble part from the hydrogel. The water insoluble gel sample was further dried (xerogel) in a vacuum oven to a constant weight (W_d). The yield, gel and sol% was obtained as

Sol% = 100 − gel% (2a)

where W_i is the total weight of the monomers (AA, NVP and MBA) and PEG used for synthesis of the gel.

2.3. Characterization of the hydrogels

2.3.1. Swelling experiment. The equilibrium swelling ratio (ESR) was evaluated gravimetrically. To study the dynamic swelling properties, the gel samples were immersed in double distilled water at ambient temperature, and the mass of the swollen gel sample (m_t) was taken at different time intervals until there was no change in the mass with time. The equilibrium swelling ratio (ESR) of the hydrogels was determined using eqn (3)where W_s and W_d are the weights of the swollen and dried hydrogels, respectively.

At the swelling equilibrium point, W_s = W_e. Therefore, the equilibrium swelling ratio (ESR) was obtained from eqn (3) by replacing W_s with W_e. The buffer solutions at various pH was prepared by dissolving phosphoric acid, potassium phosphate (KH₂PO₄), potassium hydrogen phosphate (K₂HPO₄), and sodium hydroxide in double distilled water. The ionic strengths of the buffer solutions were adjusted to 0.1 molar with a sodium chloride solution. The pH was determined using a pH meter. Swelling of the hydrogel samples was also carried out at different ionic strengths in the presence of NaCl, CaCl₂ and AlCl₃.

2.3.2. Factorial design for swelling study. Customarily, pharmaceutical formulations, which are developed by changing a single variable at a time, are soundly upgraded by a factorial design followed by an interaction method.¹⁴ In this factorial design, the influences of all experimental variables, either independent (total monomer concentration, PEG concentration, crosslinker concentration, initiation concentration) or dependent variables (swell ratio) with subtle interaction effects on their responses, were investigated with the 4^th order interaction effect on the responses.

2.3.3. Fourier transform infrared (FTIR) spectroscopy. The FTIR spectra of the drug free and drug loaded hydrogel samples were obtained using a FTIR spectrometer (Perkin Elmer, model-Spectrum-2, Singapore) with a KBr pellet made by mixing KBr with a fine powder of the drug free and drug loaded gel samples. (10 : 1 mass ratio of KBr to the polymer, for the drug only sample, 50 : 1 ratio of KBr to the drug).

2.3.4. Thermal analysis. Differential thermal analysis (DTA) and thermogravimetric analysis (TGA) of the hydrogel samples were carried out in a Perkin Elmer instrument in a nitrogen atmosphere at a scanning rate of 10 °C min⁻¹ over the temperature range, 25 to 600 °C.

2.3.5. X-ray diffraction (XRD). The wide angle XRD profile of the gel samples was studied at room temperature in a diffractometer (model: X'Pert PRO, made by PANalytical B.V., The Netherlands) using Ni-filtered Cu Kα radiation (λ = 1.5418 Å) and a scanning rate of 0.005° (2θ) s⁻¹. The diffraction angle was varied from 2–72° 2θ.

2.3.6. Scanning electron microscopy (SEM) studies. The morphology of the gold coated semi-IPN gels were observed by using a SEM (Scanning Electron Microscope, model no. S3400N, VPSEM, Type-II, made by Hitachi, Japan) at an accelerating voltage of 15 kV.

2.3.7. Point zero charge (PZC) analysis. For PZC analysis, a small amount of the hydrogel sample (∼50 mg) was taken in a 100 mL conical flask containing a 0.1 N potassium nitrate solution. The pH of the solution (pH_i) was adjusted between 2 and 12 by adding either 0.1 N nitric acid or 0.1 N sodium hydroxide.¹⁵ The solution containing the gel sample was then kept for 48 hours to reach equilibrium with occasional shaking. The pH of the supernatant liquid was measured (pH_f). The difference between this initial and final pH (ΔpH = pH_i − pH_f) was plotted against pH_i, and the point of intersection of the curve at ΔpH = 0 gives the PZC for the hydrogel.

2.3.8. Study of pH-reversibility. Buffer solutions with various pH values were prepared by combining KH₂PO₄, K₂HPO₄, H₃PO₄, HCl, and NaOH solutions. The pH was determined using a pH meter. The equilibrium water uptake in the various pH solutions was determined using a method similar to that in distilled water. The pH reversibility of the hydrogel samples was investigated in terms of its swelling–deswelling oscillatory behavior in a pH buffer solution of phosphate between pH 7.8 and 1.7. The sequential time interval was 30 min for each cycle.¹⁶

2.4. Evaluation of network parameter

The gel network is usually characterized in terms of the average molecular weight between crosslinks, M_c and mesh size, ς, measured by neutron scattering or quasi-elastic light scattering.¹⁷ M_c is obtained from the following equation based on network theory of Flory and Rehner.¹⁸

The molar volume of water, V_s at experimental temperature (25 °C) was calculated (18.18 cm³ mol⁻¹) from its density (0.98 g cm⁻³) and molecular weight (18 g mol⁻¹). The density of the hydrogel sample, ρ_p, was calculated from its mass and volume. The volume of the polymer sample was measured using the method reported elsewhere.¹⁹ For equilibrium swelling of m_w g water/g dry hydrogel sample, polymer volume fraction in swollen gel under equilibrium, ϕ_p will be

where ρ_p and ρ_i are the density of the polymer and solvent (water), respectively. The polymer–solvent interaction parameter, χ, between water and the polymer hydrogel was obtained using the following equation.²⁰

Crosslink density (ρ_c) of a hydrogel is obtained as

where N_A is Avogadro's number (6.023 × 10²³ mol⁻¹). The mesh size (ς in Å) of the swollen polymeric network was calculated from the following eqn.²¹

The Flory's characteristic ratio, C_n, was taken from the literature and the C–C bond length; ‘I’ was assumed to be 1.54 Å.²² M_r, the molecular weight of repeat unit, was calculated as the weight average of the repeat unit of PEG (M_r = 62) and poly(acrylic acid-co-N-vinylpyrrolidone) (M_r = 184).

2.5. Study of drug loading and entrapment efficiency of the hydrogel

The drug loading and entrapment efficiency of the hydrogel samples were carried out using similar experiments reported elsewhere. For drug loading, hydrogel samples of a specified weight (W_i) were first swollen in a 100 mL water–ethanol mixture (20% ethanol (v/v)) with a constant pH and ionic strength (pH 1.7 simulating gastric fluid and 7.8 simulating intestine fluid and ionic strength 0.1 mol L⁻¹) and containing a specified amount (W_o) of cefadroxil drug. After 72 h of swelling, the drug loaded wet hydrogel samples were carefully taken out from the solution and washed with the same solution to remove the free drug from the sample. The drug loading (DL) and entrapment efficiency of the hydrogel sample was determined to bewhere W_d is the weight of the drug loaded dry hydrogel sample.

2.6. In vitro drug release studies

The in vitro release of the drugs from the hydrogel samples was carried out at 35 ± 0.5 °C using Indian Pharmacopoeia (IP) Dissolution Test Apparatus Type 2 (paddle method) at a rotation speed of 50 rpm in 100 mL of buffer (pH 1.7 and 7.8) for 7–9 hours. The drug loaded wet samples obtained from the drug loading test were first dried overnight under ambient conditions followed by drying in a vacuum oven at 50 °C for another three days. The drug loaded dry samples were then immersed in buffer solution with the same composition. At several time intervals, 5 mL of the solution containing the released drug was withdrawn, and at the same time, 5 mL of fresh solution was added to maintain a constant solution volume. The concentration of the drug in the withdrawn solution was analyzed by UV-vis spectrophotometry (Lamda 25, Perkin Elmer, Singapore) at a λ_max of 230 nm for cefadroxil using a calibration curve constructed from a series of drug solutions with known concentrations. All release experiments were carried out in triplicate and the average values were considered. The drug release% was obtained as follows:where W_drug is the mass of the drug loaded gel sample and W_release is the mass of the drug released in the solution.

3. Results and discussion

3.1. Synthesis of composite hydrogels

In the present work, acrylic acid, N-vinylpyrrolidone (NVP) and MBA undergoes free radical crosslink copolymerization in water in the presence of PEG. During free radical polymerization, three acrylic monomers copolymerize with three vinyl (CH₂=CH–) functional groups of one MBA monomer; thus, a three dimensional network of crosslink copolymer gel is formed, as shown in the Fig. 1a followed by Scheme 1. Further, some of the (–OH) groups of PEG react with the carboxylic (–COOH) functional groups of acrylic acid to form a polyester type complex (–COO–).²³ Accordingly, a stable composite hydrogel is formed, where the dispersed phase, i.e., PEG, is chemically and covalently bonded to the continuous acrylic copolymer phase. The formation of the composite gel is shown in Scheme 1 while the structure of the two drug molecules, i.e., cefadroxil, is shown in Fig. 1b. The copolymer in the semi-IPN type gel was also confirmed by NMR spectroscopy.^4,24 A peak was reported at 81.79 ppm in the gel spectrum, which is a signal for the quaternary carbon atom (4⁰ carbon) groups in the freshly built cross bond. The change is consistent with the fact that the N–CH attached group partially altered the quaternary carbon atom (4⁰ carbon) groups in the backbone chain. The carbonyl carbon peak at 215.40 ppm vanished after crosslinking.

Fig. 1 (a) Formation of semi-IPN and its plausible interaction with the drug. (b) Chemical structure of cefadroxil. (c) Effect of the initiator (I), monomer (M), crosslinker (MBA), and polyethylene glycol (PEG) on the yield or gel% and gel time (T_gel).

Scheme 1 Formation of the composite hydrogel.

3.1.1. Effect of reaction variables on gel content and gel time of the hydrogels. The effect of the initiator (I), monomer (AA and NVP), crosslinker (MBA), and polyethylene glycol (PEG) on the gel content (%), yield (%) and gel time of the hydrogels are shown in Fig. 1c. To examine the effect of a single parameter, the other parameters were kept constant; i.e., when the initiator concentration was varied, 0.5, 0.8, 1.0 and 1.5 wt% (of the total monomer weight), the amounts of crosslinking agent and monomer were kept constant at 1 wt% and 20 wt%, respectively. Fig. 1c depicts that the increase in the initiator concentration from 0.5 to 1.5 wt% yield or gel% decrease gel time. In fact, the rate of polymerization increased at higher initiator concentrations, resulting in a polymer gel with a shorter chain length (lower molecular weight). Therefore, gelling occurs at an early stage of polymerization, resulting in a shorter gel time and lower gel%.²⁵ At very low initiator concentrations, i.e., 0.5 wt% initiator, the generation of free radicals from initiator was too low. Therefore, the yield or gel% was observed to increase from 0.5 to 1.5 wt% initiator in the polymerization mixtures. The network (gel) in the polymer was formed at a much faster rate in the presence of an increased amount of crosslinker, i.e., MBA. Similarly, the yield or gel% also increases with increasing concentration of crosslinker due to the increase in the rate of polymerization in the presence of an increased amount of reactant (crosslinker). However, at 2.5 wt% crosslinker, the yield or gel% decreased because of the formation of a network at the early stages of polymerization.¹⁵ The effects of the total monomer concentration in the reaction medium on the synthesis of the gel are also shown in Fig. 1c. With increasing monomer (acrylic acid, vinyl pyrrolidone) concentration in water, the yield or gel% was observed to increase, which may be due to the generation of a large number of active primary radicals at higher monomer concentrations in water.²⁵ In addition, the gel time decreased with increasing monomer concentration in water. This is because gelling occurs early due to the increased reaction rate at a higher monomer concentration. From Fig. 1c, it was also observed that the gel time, yield% and gel% increased with increasing wt% of PEG in the hydrogel. At a higher PEG concentration, the solution viscosity increased and the same amount of MBA and monomers took a longer time to gel in the viscous medium. However, PEG also participates in the polymerization reaction by forming macro radicals; hence, the yields or gel% are observed to increase with increasing amount of PEG in the hydrogel.²⁶ At 4 wt% PEG, the viscosities of the reaction medium increased significantly and the extent of the polymerization reaction deceased with decreasing yield or gel%.

3.1.2. Factorial design. The effects of individual variables (X₁, X₂, X₃ and X₄) and their interactions between the four variables in the factorial design of the swelling experiments were computed using the sign table (Table 2) for the counterpoint constants for the 2⁴ design, as offered by Montgomery (1997).²⁷ After performing the fourth order interaction model analysis from these contrasts, 15 factorial effects were determined. The correlation between the effective experimental variables (X₁, X₂, X₃ and X₄) and the dependant variable (swell ratio, SR) was evaluated by multiple linear regression. Fourth order interaction model analysis was used to compute the regression coefficients and the following polynomial equation was derived:²⁸

y = 16.699813 + 4.0108125X₁ + 1.992938X₂ + 1.4666875X₃ + 5.5808125X₄ + 2.046938X₁X₂ + 0.3776875X₁X₃ − 2.3306875X₁X₄ − 0.93944X₂X₃ −2.25106X₂X₄− 1.1343125X₃X₄ − 1.0854375X₁X₂X₃ − 1.3995625X₁X₂X₄ − 0.645438X₂X₃X₄ − 1.4258125X₁X₃X₄ + 3.986063.

here, the coded variables, X₁, X₂, X₃, X₄, and y represent the PEG content, crosslinker concentration, initiator concentration, total monomer concentration, and swell ratio respectively.

3.1.3. Effect of filler incorporation on swelling. The incorporation of hydrophilic phyllosilicate into the polymer gel affects the water uptake percentage as well as its equilibrium swelling ratio (ESR).²⁹ The composite meshing type network consisting of a hydrophilic filler showed higher water uptake compared to the unfilled hydrogel, i.e., the filler enhances the hydrophilicity of the gel. In contrast, it is quite noticeable that beyond a 4% filler loading, the filler particles established another network over the hydrophilicity effect in the swelling behaviour. As a result, the swelling ratio decreased.³⁰ The functional groups of the polymer as well as the hydrophilic filler participates in water absorption, and the network structure of the gel filled with the hydrophilic fillers took a much shorter time to reach saturation.

3.2. Characterisation of the hydrogels

3.2.1. FTIR spectra. In Fig. 2, for PEG, a strong but broad absorption band appearing at 3442 cm⁻¹ indicates its O–H and C–OH stretching, while the absorption band at 2885 cm⁻¹ is due to its CH₂ stretching vibration. The absorption band at 1342 cm⁻¹ stands for its –CH₂ wagging vibration. Another characteristic absorption band of PEG at 1957 cm⁻¹ was assigned to its crystalline state. Similarly, the absorption band at 1098 cm⁻¹ was attributed to C–O–C stretching and the absorption peak at 840 cm⁻¹ was assigned to the vibration of its –CH₂–CH₂–O group.³¹ The 2892 cm⁻¹ peak of PEG due to its CH₂ stretching was observed to shift to 2880 cm⁻¹ in the gel. Similarly, the 3384 cm⁻¹ peak of PEG due to its O–H and C–OH stretching was observed at 3317 cm⁻¹ in the gel sample, and 3325 cm⁻¹ in the filler loaded gel. The Si–O vibration band at 1024 cm⁻¹ of bentonite³² and the C–O–C stretching band of PEG at 1098 cm⁻¹ were shifted to 1120 cm⁻¹ in the filler loaded gel. Similarly, the 519 cm⁻¹ absorption peak corresponding to the stretching vibration of Si–O–Al of bentonite³³ also possessed a strong shoulder in that region. All these shifting and bifurcations indicate strong electrostatic interactions among various functional groups of the hydrogels and the fillers.³⁴ FTIR spectroscopy of poly(NVP-co-AA) after cross linking was carried out. The results are shown in Fig. 2. It has already been established that the carbonyl group of NVP segments exhibits a stretching vibration peak between 1650 and 1680 cm⁻¹ and the carboxylic acid group on the PAA chain exhibits a peak at approximately 1736 cm⁻¹ from the literature.³⁵ When the carbonyl group frames intermolecular hydrogen bond, there is a negative shift in the FTIR spectrum. In the present work, the carbonyl group exhibits a peak at 1654 cm⁻¹ for poly(NVP-co-AA), which shifted to a peak at 1632 cm⁻¹ for the polymeric gel. This negative shift from 1654 to 1632 cm⁻¹ indicates that a stronger inter-molecular interaction may take place. The strong shoulder appearing at about 1734 cm⁻¹ can be distinguished from the stretching vibration of the carbonyl group of carboxylic acid on the PAA segments.³⁶ In addition, a small shoulder appeared at about 1736 cm⁻¹, corresponding to the stretching vibration of the carbonyl group of the carboxylic acid group on the copolymer chain, which further showed that some intermolecular interactions due to a hydrogen bond occurred and complexation formed between the carbonyl group acid segments of the acrylic acid.^37–40

Fig. 2 FTIR of the polymer, (i) drug loaded gel, (ii) cefadroxil, (iii) filler loaded gel, (iv) pure gel, (v) filler, and (vi) PEG.

3.2.2. DTA and TGA. The DTA and TGA of the polymer samples are depicted in Fig. 3. The virgin polyethylene glycol (PEG) showed an endothermic sharp melting peak at approximately 67 °C and a weak exothermic inflection in the baseline at around 172 °C due to oxidative degradation. PEG also showed exothermic peaks at 337 °C and 372 °C due to its decomposition and charcoal evolution.⁴¹ PEG showed a weight loss of around 10 wt% up to 275 °C, which is associated with the loss of physically adsorbed water.⁴² As PEG is incorporated in the matrix of poly(acrylic acid-co-vinylpyrrolidone), the drug loaded gels show endothermic broad melting peaks in the temperature range of 200–280 °C. Similarly, the exothermic degradation peak of pure PEG at around 337 and 372 °C is also shifted to a single exothermic peak at around 450–550 °C in the copolymer gel. In the case of the pure drug, the DTA peak is combined with the pure polymer peaks and generates new broad regions of melting. The drug loaded gel shows crystallization peaks in between the pure drug and pure copolymer gel. The drug loaded sample shows multiple degradation profiles, which are also the result of the drug and copolymer TGA behaviour.

Fig. 3 (i) DTA and (ii) TGA of the polymer.

3.2.3. X-ray diffraction (XRD). The XRD patterns of pure PEG4000, montmorillonite clay and unfilled gel are shown in Fig. 4. PEG showed strong XRD peaks at two theta (2θ) of 19.8°, 23.8° and some weak peaks at 26°, 36.6°, 40.5°, and 46.2° as reported elsewhere.⁴³ Prominent intermolecular hydrogen bonding between the –OH groups of the PEG chain is responsible for its crystallinity, and therefore, the increasing intensity of the XRD peaks. Similar to PEG, the clay also showed the respective diffraction peaks of its montmorillonite “001” planes at 2θ of 7.1°, 20.3°, 29°, 35.6° and 62.3°, as reported elsewhere.⁴⁴

Fig. 4 XRD of the polymer and drug.

3.2.4. Scanning electron microscopy (SEM). The cross sectional morphology of the poly(NVP-co-AA) hydrogel with moderate cross linking levels is shown in Fig. 5. From this, we can see that the hydrogel exhibits a certain consistent and porous three-dimensional network structure. This may be due to the effective cross linking bond triggered by the free radical boosted MBA crosslinker. From this, the elastic nature and large free volume exhibited by the poly(AA-co-NVP) hydrogel would be favorable to the movement of the polymer chain segments and the migration of charged particles in the interior of the hydrogel.⁴⁵

Fig. 5 Scanning electron microscopy images of (i) unfilled gel, (ii) clay loaded gel, and (iii) drug loaded gel.

3.2.5. Point zero charge (PZC) analysis. The change in the state of ionization of the functional groups of the hydrogels with the solution pH was evaluated in terms of its point zero charge (PZC). The initial pH of a solution (pH_i) changes in the presence of a hydrogel. The difference between the final pH (pH_f) and initial pH (pH_i), i.e., the pH_i − pH_f was plotted against the initial solution pH (pH_i) in Fig. 6a. The pH of the solution (pH_PZC) at which pH_i − pH_f is zero is called the point zero charge. Therefore, the hydrogel will remain neutral at solution pH = pH_PZC, positively charged at solution pH < pH_PZC or negatively charged at solution pH > pH_PZ.⁴⁶ Fig. 6a shows that PEG has a pH_PZC of 9.98. This implies that carboxylic acid group will remain in its protonated form (–COOH) at moderate pH (almost 3 for unfilled and 6 for filled gel). It was also observed that in the presence of PEG, the negative charge of the acrylic acid segments was reduced because of the organization of the ‘polyion’ complex.⁴⁷ Thus, the pH_PZC of the IPN gel decreased. The full IPN shows a slightly lower pH_PZC than the semi-IPN, which may be due to the cross linking of PEG in the IPN hydrogels.

Fig. 6 (a) PZC of the filler loaded gel, pure gel and gel without PEG. (b) pH reversibility (switch on–off behavior) of the gels.

3.2.6. pH-sensitivity. Fig. 6b shows the dependence of water absorption for PEG/poly(acrylic acid-co-N-vinyl pyrrolidone) semi-IPN on the pH of the external buffer solution (0.10 mol L⁻¹). The hydrogel almost does not swell at pH 1.7, but it swelled sharply with increasing external pH until a plateau was reached (pH > 4). The evident change in water absorption with the pH of the external buffer solution confirmed the excellent pH-sensitive behaviour of PEG/poly(acrylic acid-co-N-vinyl pyrrolidone) semi-IPN gel. The behaviour of the semi-IPN can be attributed to the following. As an anionic polymer, the semi-IPN contains numerous hydrophilic –COO⁻ and –COOH groups that can convert with each other. In an extreme acid medium (pH 1.7), the –COO⁻ groups transform to –COOH groups. The hydrogen-bonding interaction among the –COOH groups is the additional physical cross linking. As a result, the electrostatic repulsion among the –COO⁻ groups is restricted, and so the network tends to shrink.⁴⁸ With increasing external pH, the ratio of –COO⁻ groups in the polymer network increase and the electrostatic repulsion between the carboxylate groups dominates over the hydrogen bonding interaction between the carboxylic acid groups. The pH reversibility of the swelling–deswelling oscillation behaviour was investigated in a buffer solution of phosphate between pH 1.7 and 7.8 (Fig. 6b). The hydrogel exhibited higher swelling capability at pH 7.8, but the swollen gel shrank rapidly at pH 1.7 and an intriguing on–off switching effect was observed. After three on–off cycles, the hydrogel still has better sensitivity, indicating that the semi-IPN has excellent pH reversibility.

3.3. Study of swelling of the hydrogels

Several hydrogels synthesized by varying concentrations of monomer (acrylic acid and vinyl pyrrolidone), initiator, crosslinker (MBA) and PEG were used for swelling at different time intervals in double distilled water. The results of swelling, i.e. the equilibrium swelling ratio (ESR) and equilibrium swelling time (t_eq.), in water for all these hydrogels are shown in Fig. 7a. The figure shows that both ESR and t_eq. decrease with increasing initiator concentration. At a higher initiator concentration, the hydrogels with a low molecular weight and more chain ends are organized, resulting in a low ESR and low t_eq. because of the network imperfection in the gel.⁴⁹ Similarly, the increase in crosslinker concentration results in a tighter and denser network of the gel. Consequently, the ESR decreases at a higher crosslinker concentration, while t_eq. increases because water molecules take a longer time to fill a dense network to reach swelling equilibrium. Similarly, with increasing total monomer concentration, ESR increases while t_eq. decreases. Based on the swelling results, as shown in Fig. 7a, the hydrogel synthesized with 20 wt% monomer, 1.0 wt% initiator and 1.0 wt% crosslinker was chosen and further filled with 1.0, 2.0, 3.0, and 4.0 wt% polyethylene glycol (PEG). The ESR increases further when the poly(acrylic acid-co-N-vinylpyrrolidone) is filled with PEG. However, above 2.0 wt% PEG, the ESR of the composite gel decreases. The PEG contains –OH as a pendant group, and thus the ESR increases in the presence of PEG. However, PEG also fills up the network of the gel, and above 2.0 wt% PEG, there is a marginal decrease in ESR, as observed in Fig. 3a. In the composite gel, the filled network requires a longer time for the penetration of water; hence, t_eq. increases with increasing PEG content. The hydrogel containing 2.0 wt% PEG, 1% MBA, 20% monomer concentration, and 0.5% initiator, showing the highest ESR, was also subjected to swelling at various pH, and this hydrogel showed an ESR of 4.3, 18.1 and 25.2 at pH 1.7, 6.8 and 7.8, respectively. The swelling at low pH is due to protonation of the carboxylic acid groups of the copolymer phase present in the hydrogel. This ionization causes swelling due to electrostatic repulsion. The carboxyl groups of the copolymer segments remain protonated up to its point of zero charge pH, which is around 8.0. Similarly, the carboxylic groups (COOH) of the gel ionize at a pH above its pK_a (4.26). Charge repulsion also results in an increasing ESR. In this case, this is correlated to a higher concentration COO⁻ groups, which is due not only to the dissociation of carboxylic acid group but also to the partial hydroxylation of the γ-lactam group in the alkaline solutions described in Scheme 2.^50,51

Fig. 7 (a) Effect of the concentration of initiator, total monomer, crosslinker, PEG, salt and pH on the equilibrium swelling ratio. (b) (i) Non-linear fitting of the swelling data to the 1^st order rate equation and diffusion characteristics, and (ii) deswelling of the hydrogel at crosslinker (MBA) wt% and varied PEGn wt% (inset).

Scheme 2 Partial hydrolization of N-vinylpyrrolidone.

3.3.1. Swelling kinetics, diffusion and network parameters. The swelling ratio (SR) of the hydrogels at various time intervals was observed to fit well to the following non-linear first order rate, eqn (5).⁵²

SR_t = Q_e(1 − e^−k₁t) (6)

here, k₁ is the rate constant and Q_e is the ESR of the respective hydrogel. Data fitting was carried out using the Levenberg–Marquardt (L–M) algorithm (Origin-8 software) with an adjustment of the parameter values, i.e., rate constant (k₁) and initial rate of swelling (r₀) by iteration using the chi square (χ²) and F values. The trend lines of these non-linear fittings for the hydrogels synthesized with 1, 2, 3 and 4 wt% PEG and designated as PEG1, PEG2, PEG3 and PEG4, respectively, are shown in Fig. 7b(i). Similar trend lines (not shown) were also obtained for the hydrogels synthesized with 0.5, 1, 1.5, 2 and 2.5 wt% crosslinking (designated as MBA0.5, MBA1, MBA1.5, MBA2, MBA2.5 respectively) with 15, 20, 25 and 30 wt% monomer in water (designated as AA/NVP15, AA/NVP20, AA/NVP25 and AA/NVP30, respectively). The values of k₁, the experimental and calculated ESR of all of these hydrogels are shown in Table 1. The values of the statistical parameters, i.e., r², χ² and F are also shown in Table 1. The ESR of the hydrogels calculated using the first order rate eqn (5) closely matches the experimental ESR. The regression coefficients (r²) for all these fittings were also close to unity (0.99006 to 0.99887), whereas these regressions also show a low χ² (0.003–0.14), the error value of the respective plots was in the 10⁻⁵ order and desirable high F values (3205–19 675) were obtained. These results confirm the good fitting of the swelling data to the first order rate equation.

Table 1 Swelling diffusion and network parameters of the hydrogels^a

Polymer code	K1 × 10^2	ESRexpt/ESRcal (g/g)	r^2/χ^2/F value	KD/n/D × 10^−6	Mc × 10/ρc × 10^−22/ς
a K1 (g gel/g water per minute), kD (s^−1), n (−), D (cm^2 s^−1), ς (Å).
MBA0.5	0.213	15.64/15.36	0.99628/0.11884/7482	0.0283/0.47134/3.44	2.19/3.22/1.58
MBA1	0.186	7.96/8.23	0.99586/0.03817/6296	0.02301/0.50299/4.41	1.57/4.31/1.32
MBA1.5	0.326	5.49/5.39	0.99317/0.02324/5465	0.06789/0.35822/1.43	1.73/4.64/1.38
MBA2	0.294	2.401/2.37	0.99104/0.00607/3920	0.0578/0.37974/1.79	9.93/9.07/1.03
MBA2.5	0.296	1.94/1.91	0.99263/0.00324/4769	0.05722/0.38088/1.79	5.99/1.53/0.79
I0.5	0.186	7.96/8.23	0.99586/0.03817/6296	0.02301/0.50299/4.14	1.1/7.9/1.22
I0.8	0.294	8.59/5.57	0.99527/0.04258/7296	0.05678/0.38353/1.91	2.62/3.33/1.94
I1	0.303	10.16/10.155	0.99491/0.06461/6840	0.05993/0.37631/1.78	1.32/6.19/1.34
I1.5	0.279	11.56/11.54	0.99039/0.15552/3562	0.05318/0.39148/2.02	1.42/5.19/1.41
PEG1	0.117	20.75/22.82	0.99887/0.07342/196 95	0.00958/0.61495/6.87	4.79/1.51/8.92
PEG2	0.217	12.103/12.29	0.99758/0.05029/116 84	0.03196/0.45992/3.46	1.80/4.49/1.41
PEG3	0.192	11.02/11.36	0.99745/0.04546/104 62	0.02525/0.49125/4.17	2.81/3.21/1.77
PEG4	0.180	7.96/8.23	0.99241/0.07082/3500	0.02376/0.49972/4.41	2.10/4.42/1.50
TMC15	0.308	1.458/1.46	0.99006/0.00257/3675	0.06653/0.36378/1.61	2.17/3.67/1.55
TMC20	0.180	7.96/8.23	0.99241/0.00257/3500	0.02376/0.49972/4.41	1.60/5.33/1.31
TMC25	0.238	6.89/7.006	0.99482/0.03338/5937	0.04104/0.42775/2.81	1.13/7.90/1.09
TMC30	0.134	11.172/11.53	0.99273/0.14295/3205	0.01281/0.58084/6.37	1.52/5.97/1.27

---

Table 2 A design matrix and results of the 2⁴ full factorial experiment on the swelling optimization study^a

Formulation code	Independent variable				Independent variables				Swelling ratio
	X1	X2	X3	X4	PEG%	MBA%	Initiator%	TMC%
a (−) and (+) sign indicate the low and high levels of a factor, respectively.
F1	−	−	−	−	4	1	0.5	20	7.96
F2	−	−	−	+	4	1	0.5	30	15.64
F3	−	−	+	−	4	1	1.5	20	11.172
F4	−	+	−	−	4	2	0.5	20	10.16
F5	+	−	−	−	1	1	0.5	20	12.103
F6	−	−	+	+	4	1	1.5	30	16.20
F7	−	+	−	+	4	2	0.5	30	12.64
F8	+	−	−	+	1	1	0.5	30	13.5
F9	−	+	+	−	1	1	1.5	20	11.15
F10	+	−	+	−	1	2	0.5	20	19.54
F11	+	+	−	−	4	2	1.5	30	26.112
F12	−	+	+	+	1	1	1.5	30	16.59
F13	+	−	+	+	1	2	0.5	30	21.54
F14	+	+	−	+	1	2	1.5	30	23.75
F15	+	+	+	−	1	2	1.5	20	24.19
F16	+	+	+	+	1	2	1.5	30	24.95

---

The study of diffusion through the network of hydrogel is relevant to its applications in drug release. The diffusion in the polymer is passive but it can be activated by swelling in release medium, i.e., water in the present case and also by various external physical forces such as polar, osmotic or convective forces.⁵³ To understand the diffusion mechanism, the swelling data was also fitted to eqn (6) and (7) to evaluate the diffusion characteristics, i.e., diffusion constant (k_D), diffusion exponent (n) and diffusion coefficient (D) of the hydrogels.⁵³

here F is the fractional water uptake and r is the radius of the cylindrical hydrogel sample. The data fitting and non-linear regression was similar to the swelling kinetics, as shown in the inset of Fig. 7b(i) for the PEG1, PEG2, PEG3 and PEG4 composite gels. Similar trend lines were obtained for the other hydrogels. The values of the diffusion characteristics, i.e., k_D, n and D of the hydrogels are also shown in Table 1. Table 1 shows that the hydrogels prepared with various PEG, MBA and monomer concentrations showed n values, ranging from 0.5 to around 0.7 indicating non-Fickian anomalous diffusion, i.e., in these cases, the rate of diffusion and the rate of chain relaxation of the gels are comparable. The hydrogels prepared with various initiator concentrations showed n values close to 0.5, indicating Fickian case-1 diffusion, i.e., in these cases, the rate of diffusion was slightly lower than the rate of chain relaxation.⁵⁴ The values of the statistical parameters, i.e., r², χ² and F for these non-linear fittings, as shown in Table 1, also confirmed the good fit of the swelling data to the diffusion equation.

Several network parameters, such as the average molecular weight between the cross links (M_c), crosslink density (ρ_c) and mesh size (ς) of the gels were obtained using eqn (4)–(4d) based on the experimental swelling data in double distilled water at pH 7.8. Table 1 shows that M_c and ς decrease with increasing cross linker concentration from MBA1 to MBA2.5, whereas ρ_c increases. This shows that the number of networks increases in the gel matrix with increasing crosslinker concentration. The molecular weight of the hydrogels decreases with increasing initiator dose for polymerisation.²⁵ M_c and ς also show an ascending trend with increasing PEG content, which corresponds to the formation of more branched multiple side chains in the hydrogel generated during the free radical reaction.

3.3.2. Time-dependent swelling behaviours in saline solution. As shown in Fig. 8, the water absorption of the semi-IPN in the NaCl, CaCl₂, AlCl₃, and FeCl₃ solutions increased with increasing contact time, reaching the maximum absorption, and then decreased with further increases in contact time until swelling almost disappeared; however, similar behaviour was not observed in the NaCl solution. The unnatural time dependent swelling effect can be attributed to the following reasons. In the multi-valence saline solutions, Ca²⁺ and Al³⁺ may complex with the hydrophilic –COO⁻ groups; so the degree of cross linking of the hydrogel network increases with increasing contact time.⁵⁵ As a result, the swollen network gradually collapsed and the initially absorbed water was squeezed out of the network under this action. Because Na⁺ has no complexing action with –COO⁻ groups, there is no time-dependent swelling effect in a NaCl solution. The reason for measuring the swelling behaviour of the hydrogel in the ferric ion simulated liquid is that ferric ions are also a component of human body fluid. Therefore, in the ferric ion bearing solution, the swelling is better with respect to an Al³⁺ containing solution. As Al³⁺ is smaller in size than Fe³⁺, it is justified in the data that the complexion ability of Al³⁺ is more prominent than Fe³⁺.

Fig. 8 Cumulative release% of cefadroxil drug at (i) initiator wt%, (ii) MBA wt%, (iii) monomer (acrylic acid and NVP) wt% and (iv) PEG wt%.

3.4. In vitro drug release study

Similar to the swelling ratio, the loading and entrapment efficiency of these drugs were also observed to increase with decreasing crosslinker wt%, increase in the wt% of PEG and increase in the solution pH from 1.7 to 7.8. From the FTIR results it was proven that there was no significant electrostatic interaction between the hydrogel and drug molecules. The cumulative release profile of the drug from these hydrogels is presented in Fig. 8(i) and (ii) for various crosslinker wt% and PEG wt%, respectively. The release profile of the PEG1.0 hydrogel at pH 7.8 and 1.7 is shown in Fig. 8. From these figures, it was observed that an initial burst release of the drug was followed by the sustained rate of the release for all of these hydrogels. Initially, the rapid release rate of the drug occurs from the surface of the hydrogel due to the high concentration gradient of the drug between the water and the gel surface.^56,57 As drug release continues, its concentration in the release medium increases and the concentration gradient of the drug between the gel and the release medium decreases, and entrapment of the remaining drug in the gel network slows down the further release at a low concentration gradient.⁵⁸ Similar release profiles were reported for the release of a model protein and drug from various IPN type hydrogels.^59–62

3.4.1. Fitting of drug release data to model equations. To evaluate the release kinetics, the first 60% of the drug release data of the hydrogels was fitted to the following: (i) Donbrow–Samuelov (Donbrow & Samuelov, (1980)) zero-order kinetics (eqn (9)), (ii) Higuchi model (Higuchi, 1963) (eqn (10)), and (iii) Korsmeyer–Peppas model (Korsmeyer, Gurny, Doelker, Buri & Peppas, 1983) (eqn (11)). Similar to the swelling kinetics, the nonlinear Levenberg–Marquardt (L–M) algorithm (Origin-8 software) was also used for these fittings.

(i) Zero order Donbrow–Samuelov model

m_Dt = m_De + K₀t (9)

(ii) Higuchi model

m_Dt = m_De + K_Ht^1/2 (10)

(iii) Korsmeyer–Peppas model

where m_Dt and m_De are the amount of drug released at time t and infinity (at equilibrium), respectively, K₀, K_H and K_KP are constant of the concerned model corresponding to structural and geometrical character of the dosage form, and the diffusion exponent ‘n’ indicates the mechanism of drug release. The values of ‘n’ ranged from 0.38 to 0.61, which indicates diffusion-assisted drug release of the hydrogels. The fitting of the drug release data to the models stated above are shown in Fig. 9(i)–(iii) for the Donbrow–Samuelov zero order, Higuchi and Korsmeyer–Peppas models, respectively, using the PEG1, PEG2, PEG3, and PEG4 hydrogels.^63,64

Fig. 9 Fitting of the drug release data of cefadroxil to (i) Donbrow–Samuelov zero order (ii) Higuchi and (iii) Korsmeyer–Peppas model for hydrogels.

3.4.2. Evaluation of drug action. The chemical activity of the drug was examined by detecting the UV spectra of the pure cefadroxil drug in water, and these drugs released in water from the hydrogels at a wavelength of 272 nm for cefadroxil, as exhibited in Fig. 10. The respective spectra appear to be virtually identical, suggesting that there was no significant change in the chemistry and bioactivity of the drug during its loading and sustained release. Indeed, a similar comparison to evaluate the chemical and bioactivity of the drug has also been reported.⁶⁵

Fig. 10 UV spectra of the pure drug and drug released from the hydrogel.

4. Conclusion

A sequential semi-IPN type pH-sensitive phyllosilicate filled polyethylene glycol based composite hydrogel of acrylic acid and N-vinyl pyrrolidone is constructed using N,N-methylenebisacrylamide as a gelling agent, which resulted a combination of chemical and physical crosslinking. The results of the 2⁴ factorial design experiment, which is depicted in the sign table, indicated a substantial contribution of all four main effects of the PEG content, crosslinker concentration, initiator concentration, and total monomer concentration, on the swelling response of the hydrogel. FTIR and DSC showed that the semi-IPN hydrogel is formed by a hydrogen bond between the carbonyl group in the vinyl pyrrolidone network and the carboxylic acid group on the AA segment as well as the chemical crosslinking by a reaction. The semi-IPN hydrogel is non-responsive to the ionic strength over the high concentration of inorganic salt are also investigated. This study showed that clay normally enhances the water uptake with a desirable strength. The molecular weight between the two crosslinks, crosslink density and mesh size of the hydrogels were also assessed. The pH reversibility of the hydrogels increased with increasing PEG wt%. The drug release data was also successfully fitted with the Korsmeyer–Peppas model.

References

1. P. Gupta, K. Vermani and S. Gary, Research focus, 2002, 7, 569; R. Dagani, Chem. Eng. News, 1997, 75, 26.
2. J. M. Rosiak and F. Yoshii, Nucl. Instrum. Methods Phys. Res., Sect. B, 1999, 151, 56.
3. J. M. Harris, Poly(ethylene glycol) chemistry: biotechnical and biomedical applications, Plenum Press, New York, 1992.
4. S. Jin, J. Gu, Y. Shi, K. Shao, X. You and G. Yue, Eur. Polym. J., 2013, 49, 1871.
5. M. T. Razzak, D. Darmawan and Z. Sukirno, Radiat. Phys. Chem., 2001, 62, 107.
6. S. Daniel, D. Shah and E. P. Giannelis, Curr. Opin. Solid State Mater. Sci., 2002, 6, 205.
7. K. Haraguchi and T. Takehisa, Adv. Mater., 2002, 14, 1120.
8. P. Schexnailder and G. Schmidt, Colloid Polym. Sci., 2009, 287, 1.
9. K. Haraguchi, T. Takehisa and S. Fan, Macromolecules, 2002, 35, 10162.
10. K. Haraguchi, T. Takehisa and M. Ebato, Biomacromolecules, 2006, 7, 3267.
11. X. Yuanqing, Z. Peng and D. Chen, Eur. Polym. J., 2006, 42, 2125.
12. K. Haraguchi, Curr. Opin. Solid State Mater. Sci., 2007, 11, 47.
13. S. G. Adoor, M. Sairam, L. S. Manjeshwar, K. V. S. N. Raju and T. M. Aminabhavi, J. Membr. Sci., 2006, 285, 182.
14. T. Lundstedt, E. Seifert, L. Abramo, B. Thelin, Å. Nyström, J. Pettersen and R. Bergman, Chemom. Intell. Lab. Syst., 1998, 42(1), 3.
15. I. D. Mall, V. C. Srivastava, G. V. A. Kumar and I. M. Mishra, Colloids Surf., A, 2006, 278, 175.
16. W. Wang and A. Wang, Carbohydr. Polym., 2010, 80, 1028.
17. X. Wei, S. Champ and M. B. Huglin, Polymer, 2001, 42, 3665.
18. Z. Y. Ding, J. J. Akloins and R. J. Salovey, J. Polym. Sci., Part B: Polym. Phys., 1991, 29, 1035.
19. J. Chen, H. Park and K. Park, J. Biomed. Mater. Res., 1999, 44, 53.
20. W. Xue, S. Champ and M. B. Huglin, Polymer, 2001, 42, 3665.
21. T. Canal and N. A. Peppas, J. Biomed. Mater. Res., 1989, 23(10), 1183.
22. E. R. Morris, A. N. Cutler, S. B. Ross-Murphy, D. A. Rees and J. Price, Carbohydr. Polym., 1981, 1, 5.
23. M. Oyama, H. Tamaru, W. T. Tang and C. W. Frank, Macromolecules, 1987, 20, 474.
24. X. Zhu, P. Lu, W. Chen and J. Dong, Polymer, 2010, 51(14), 3054.
25. R. S. Tomar, I. Gupta, R. Singhal and A. K. Nagpal, Polym.–Plast. Technol. Eng., 2007, 46, 481.
26. G. Odian, Radical chain polymerization, in Principle of polymerization, John Wiley & Sons Inc., Singapore, 3rd edn, 1991, pp. 198–334.
27. D. C. Montgomery, Design and Analysis of Experiments, Wiley, New York, 4th edn, 1997, pp. 290–341.
28. W. Wang and A. Wang, Carbohydr. Polym., 2010, 80, 1028–1036.
29. L. Torbjörn, E. Seifert, L. Abramo, B. Thelin, Å. Nyström, J. Pettersen and R. Bergman, Chemom. Intell. Lab. Syst., 1998, 42.1, 3.
30. C. Damm, M. R. Mallembakam, A. Voronov and W. Peukert, J. Appl. Polym. Sci., 2011, 120, 799.
31. A. Li and A. Wang, Eur. Polym. J., 2005, 41, 1630.
32. S. Nagarajan, K. Sabdham and V. Srinivasan, Macromol. Mater. Eng., 1997, 245, 9.
33. L. Zheng, S. Xu, Y. Peng, J. Wang and G. Peng, Polym. Adv. Technol., 2007, 18, 194.
34. K. Xu, J. Wang, S. Xiang, Q. Chen, W. Zhang and P. Wang, Appl. Clay Sci., 2007, 38, 139.
35. Q. Tang, J. Wu, H. Sun, S. Fan, D. Hu and J. Lin, Polym. Compos., 2009, 30, 1183.
36. A. J. Marshall, J. Blyth, C. A. B. Davidson and C. R. Lowe, Anal. Chem., 2003, 75, 4423.
37. J. Shuping, J. Gu, Y. Shi, K. Shao, X. Yu and G. Yue, Eur. Polym. J., 2013, 49, 1871.
38. J. F. Yaung and T. K. Kwei, J. Appl. Polym. Sci., 1998, 69, 921.
39. D. M. Devine and C. L. Higginbotham, Eur. Polym. J., 2005, 41, 1272.
40. D. M. Devine and C. L. Higginbotham, Polymer, 2003, 44, 7851.
41. P. Bures and N. A. Peppas, ACS Polym. Mater.: Sci. Eng., 2000, 83, 506.
42. E.-B. Mahmoud, G. Fetih and M. Fathy, Saudi Pharm. J., 2009, 17, 217.
43. Z. Huang, S. Liu, B. Zhang, L. Xu and X. Hu, Carbohydr. Polym., 2012, 88, 609.
44. D. Shanmukaraj and R. Murugan, J. Power Sources, 2005, 149, 90.
45. M. A. Hussein, B. M. Abu-Zied and A. M. Asiri, Polym. Compos., 2014, 35, 1160,  DOI:10.1002/pc.22763.
46. M. H. Cohen and D. Turnbull, J. Chem. Phys., 1959, 31, 1164.
47. S.-G. Wang, X.-F. Sun, X.-W. Liu, W.-X. Gong, B.-Y. Gao and N. Bao, Chem. Eng. J., 2008, 142, 239.
48. S.-G. Kim, G.-T. Lim, J. Jegal and K.-H. Lee, J. Membr. Sci., 2000, 174, 1.
49. W. Wang and A. Wang, Carbohydr. Polym., 2010, 80, 1028.
50. L.-liang Shen, G.-X. Liu and Y. Tang, Acta Pharmacol. Sin., 2007, 28, 2053.
51. S. Jin, M. Liu, F. Zhang, S. Chen and A. Niu, Polymer, 2006, 47(5), 1526.
52. S. Li, R. Vatanparast and H. Lemmetyinen, Polymer, 2000, 41(15), 5571.
53. Z. SedenAkdemir and A. N. Kayaman, Polym. Adv. Technol., 2007, 18, 932.
54. Y. M. P. S. Mohan, K. K. Murthy and M. Raju, React. Funct. Polym., 2005, 63, 11.
55. S. Lj Tomić, M. M. Mićić, J. M. Filipović and E. H. Suljovrujić, Radiat. Phys. Chem., 2007, 76, 801.
56. S. B. Hua and A. Q. Wang, Carbohydr. Polym., 2009, 75, 79.
57. N. S. Patil, J. S. Dordick and D. G. Rethwisch, Biomaterials, 1996, 17, 2343.
58. K. S. V Krishna Rao, B. Vijaya Kumar Naidu, M. C. S. Subha, M. Sairam and T. M. Aminabhavi, Carbohydr. Polym., 2006, 66, 333.
59. P. Matricardi, C. Di Meo, T. Coviello, W. E. Hennink and F. Alhaique, Adv. Drug Delivery Rev., 2013, 65, 1172.
60. N. S. Patil, J. S. Dordick and G. Rethwisch David, Biomaterials, 1996, 1724, 2343.
61. R. V. Kulkarni, V. Sreedhar, S. Mutalik, C. M. Setty and B. Sa, Int. J. Biol. Macromol., 2010, 47, 520.
62. R. V. Kulkarni, R. Boppana, M. G. Krishna, S. Mutalik and N. V. Kalyane, J. Colloid Interface Sci., 2012, 367, 509.
63. R. V. Kulkarni, V. Sreedhar, S. Mutalik, C. M. Setty and B. Sa, Int. J. Biol. Macromol., 2010, 47, 520.
64. R. W. Korsmeyer, R. Gurny, E. Doelkar, P. Buri and N. A. Peppas, Int. J. Pharm., 1983, 15, 23.
65. R. Chouhan and A. K. Bajpai, Release dynamics of ciprofloxacin from swellable nanocarriers of poly(2-hydroxyethyl methacrylate): an in vitro study."Nanomedicine: Nanotechnology, Biol. Med., 2010, 6(3), 453.

---