Cross-linked methyl cellulose/graphene oxide rate controlling membranes for in vitro and ex vivo permeation studies of diltiazem hydrochloride

Abstract

Permeability characteristics of the anti-hypertensive drug, diltiazem hydrochloride, from uncross-linked and cross-linked methylcellulose (MC)/graphene oxide (GO) rate controlling membranes (RCMs) were investigated. The MC/GO membranes were cross-linked with different concentrations of glutaraldehyde (GLA) to examine the effect of cross-linking on the permeability characteristics. The ATR-FTIR spectra, along with solubility resistance, swelling studies, the molar mass between cross-links, and moisture absorption of cross-linked RCMs over the uncross-linked RCM confirmed the cross-linking between MC and GO. The cross sectional view of cross-linked and uncross-linked RCMs, as observed by SEM, showed that the porous and fibrillose structure of the uncross-linked RCM was disrupted after cross-linking. The cross-linked RCMs showed improved mechanical and thermal properties compared to the uncross-linked RCMs. In vitro and ex vivo drug release was found to depend on the concentration of the cross-linker, which suggests that drug delivery is controlled by the cross-link density of RCM.

---

1. Introduction

The transdermal drug delivery system (TDDS) is a potential alternative method of drug delivery, where the drug passes through the skin and into the blood stream. It maintains the plasma concentration, thus reducing the dosing frequency, and avoids gastrointestinal action. As such, it is associated with improved patient compliance. In recent years, a significant amount of research has been geared toward developing efficient transdermal drug delivery systems for non-steroidal anti-inflammatory drugs.^1–4 However, the barrier properties of skin restrict the effective systemic delivery for most of these drugs. The transdermal access of drugs into blood circulation at a desired rate can be achieved by using a suitable rate controlling membrane (RCM).⁵

Cross-linking or curing is a vital tool for the modification of existing polymers to achieve new and improved material. Cross-linking is a common way to improve the controlled release properties and mechanical strength by introducing a three-dimensional network structure.⁶ As such, the movement of drug molecules across cross-linked polymer membranes can be controlled by precisely controlling this network structure. We have therefore proposed the synthesis of cross-linked graphene oxide (GO)/methyl cellulose (MC) RCM by cross-linking it with glutaraldehyde.

Currently, graphene is becoming a “rising star” material after its successful production by a simple scotch tape approach using readily available graphite in 2004 by Geim and co-workers. Numerous efforts have been made to review the structure, preparation, properties, and applications of graphene and its composite materials.^7,8 The recent innovation of graphene has been accompanied by increasing research attention to explore this new material for drug delivery applications. Graphene, a single layer of sp²-hybridized carbon atoms arranged in a honeycomb-like two-dimensional (2-D) crystal lattice, has evoked enormous interest throughout the scientific community since its first appearance in 2004.⁹ GO, a derivative of graphene, has a large number of oxygen-containing groups, such as carboxyl, hydroxyl and epoxy, on its basal planes and edges, which impart GO with good water-solubility. These oxygen-containing groups also provide an opportunity for GO to be functionalized through covalent and noncovalent bonding, thus making it a building block for synthesizing versatile functional materials.¹⁰ Due to its unique chemical structure and geometry, GO possesses remarkable physico-chemical properties, including a high Young's modulus, high fracture strength, and large specific surface area and biocompatibility.^11,12

Cellulose and its derivatives (such as ethers and esters), together with starch, are the most important raw materials for the preparation of films. Being a biocompatible semi-synthetic polymer, MC has acquired special attention because it is a heterogeneous polymer consisting of highly substituted zones called hydrophobic zones and less substituted zones called hydrophilic zones.¹³ Among all the film-forming polymers, MC is most widely used in the pharmaceutical industry. MC, a semi-flexible linear-chain polysaccharide, has the most straight forward chemical composition among cellulose derivatives with partial replacement of hydroxyl groups by methoxy moieties.¹⁴ In effect, the inherent chemical structures of GO and MC have the potential for the cross-linking reaction through hydroxyl groups.

All multifunctional compounds capable of reacting with the hydroxyl groups may be used as cross-linkers for MC; thus, dialdehydes have been investigated for MC modification.¹⁵ In the present work, glutaraldehyde (GLA) is used as the cross-linker for MC/GO. Diltiazem is a non-dihydropyridine (DHP) member of the group of drugs known as benzothiazepines, which are a class of calcium channel blockers¹⁶ used in the treatment of angina pectoris, hypertension and some types of arrhythmia. It is a class-3 anti-angina drug and a class-4 anti-arrhythmic. Diltiazem acts as an inhibitor of the CYP3A4 enzyme obstruction.¹⁷ Diltiazem is completely absorbed on oral administration; but, due to its extensive first pass effect, its bioavailability decreases to 30–40%.¹⁶ All of these factors make diltiazem hydrochloride a suitable candidate for transdermal drug delivery. Delivery of diltiazem hydrochloride into systemic circulation via the skin has generated a lot of interest over the last ten years.¹⁸ Transdermal delivery of diltiazem hydrochloride appears to be the alternative route of administration, as a non-invasive mode of drug-delivery to maintain the drug levels in the blood for an extended period of time.¹⁹ For the said reason, diltiazem hydrochloride was selected as the model drug for studying the effectiveness of the synthesized membranes for a transdermal patch.

The inspiration behind this present research work is the development of RCMs using GLA as a cross-linker of MC/GO for its application in transdermal patches. The effects of GLA concentration on the performance of cross-linked RCMs are studied. Cross-linking between MC/GO with GLA is established with the help of attenuated total reflection-Fourier-transform infrared (ATR-FTIR) spectroscopy. Solubility resistance, swelling studies, molar mass between cross-links and moisture absorption studies are also performed to confirm the cross-linking of RCMs. The morphological analyses of uncross-linked and cross-linked RCMs are studied using scanning electron microscopy (SEM). Mechanical properties are investigated and thermal characterization (TGA) of the RCMs is carried out to evaluate the stability and to see the effect on the bioavailability of diltiazem hydrochloride. Diltiazem hydrochloride permeation through cross-linked RCMs in in vitro and ex vivo conditions has been evaluated with the objective of being used in transdermal patches.

2. Experimental

Materials

Methyl cellulose (Metolose SM-4000) of weight averaged molecular weight (M_w 2.93 × 10⁵ g mol⁻¹) was obtained from Shinetsu Chemical Co., Japan. Graphite powder was received from Sigma-Aldrich. Glutaraldehyde (GLA) 25% was received from Merck Specialties Private Ltd., India. The dialysis membrane (LA390, average flat width of 25.27 mm, average diameter of 15.9 mm and approximate capacity of 1.99 mL cm⁻¹) was purchased from Hi-Media Laboratories Pvt. Ltd., Mumbai, India. Diltiazem hydrochloride of molecular weight 450.98 daltons, was a gift sample received from Ranbaxy Int., Gurgaon, Haryana, India.

Graphene oxide synthesis by the Hummers' method

Graphene oxide (GO) was synthesized from graphite powder using the modified Hummers and Offeman method.^20–22 Initially, 23 mL of 98% concentrated H₂SO₄ were cooled down below 0 °C in an ice bath. Separately, 1 g of graphite powder and 0.5 g of NaNO₃ were mixed together. This mixture was then added to the cooled (0 °C) H₂SO₄ and was constantly stirred in an ice bath for 45 minutes. This was done so that the chemicals had a sufficient amount of time to mix with each other. Then, 3 g of KMnO₄ were slowly added to the solution. The color of the solution immediately changed from black to a greenish black. The solution was removed from the ice bath and was then left to reach room temperature. Next, the solution was kept in an oil bath with constant stirring, which caused the color to change to brown. MilliQ water (140 mL) was added, followed by 7 mL of 30% H₂O₂ solution to complete the reaction; stirring was continued for 15 minutes. The dark brown solution then changed color to yellow. The solution was left to settle overnight, then the product was filtered; the pH of the solution was very close to 2, indicating a high acidic character. The product was then centrifuged, followed by repeated washing to remove the acidity. Finally, the product was vacuum dried at 50 °C to obtain GO powder (Fig. 1).

Fig. 1 Scheme illustrating the preparation of GO.

Preparation of MC/GO cross-linked rate controlling membranes

A fixed amount of prepared GO was dispersed into an aqueous solution. Then, 1 wt% of methylcellulose (MC) was added to the suspension, and a homogeneous system was formed after 5 h of stirring. The MC/GO solution was treated with ultrasound for 4 h at a 60 °C. GLA and hydrochloric acid were added after the MC/GO solution was cooled to room temperature. A blending process was used to prepare the uncross-linked membrane (F1) of MC and GO. The GLA content was varied at 3, 6, 9 and 12 wt% for the preparation of cross-linked membranes (F2, F3, F4 and F5, respectively). In each solution, 2 drops of hydrochloric acid were added to yield a solution of pH ∼ 3. A magnetic stirrer was used at 700 rpm for 1 h to obtain the homogeneously blended solution. Approximately 30 mL of the MC/GO solution was poured into a glass Petri-plate. Dried MC/GO cross-linked membranes were obtained after the solvent was evaporated in an air-circulated chamber at ambient temperature. Finally, the resulting dried membranes were washed with distilled water in order to neutralize the membranes. The thickness of each chemically cross-linked MC/GO membrane was approximately 50 to 70 μm. The compositions of the rate controlling membrane (RCM) formulations are presented in Table 1.

Table 1 Compositions of RCMs formulations

Formulations	Composition (%)
	MC (wt%)	GO (wt% with respect to MC)	GLA (wt%)
F1	1	1	0
F2	1	1	3
F3	1	1	6
F4	1	1	9
F5	1	1	12
Pure MC	1	0	0
Cross-linked MC	1	0	12

---

Characterization

ATR-FTIR spectroscopy. Attenuated total reflection-Fourier-transform infrared (ATR-FTIR) measurements were completed for all the dried RCMs to evaluate the effects of the cross-linking agent on the absorbance spectra of the MC/GO membranes. An ATR mode Fourier-transform infrared spectrometer (Nicolet AVATAR 320, Madison, WI) with a scanning range of 4000–500 cm⁻¹ was used to obtain the FTIR absorbance spectra of each of the dried membranes. The ZnSe prism was used as an internal reflection element. The total number of scans was 64 with a resolution of 2 cm⁻¹.

Solubility and swelling measurements. The uncross-linked and cross-linked MC/GO membranes (2 cm × 2 cm) were dried under a vacuum until constant weight was observed. All of the membranes were immersed in 30 mL of distilled, deionised water in order to determine the solubility and swelling degree (SD).²³ The membranes were kept in water for 24 h to validate their solubility. They were then dried under a vacuum until constant mass was achieved. For the SD measurements, all the samples were removed from the water then dried with absorbent paper and weighed. The solubility and the SD were determined by eqn (1):where X is the solubility or the SD; W_t is the mass of the sample at time t and W_o is the mass of the dry sample. Measurements were performed in triplicate.²⁴

Molar mass between cross-links and cross-link density. Properties of RCMs depend on the network structure, which is controlled by the feed composition. As a result of the increase or decrease in the swelling ratio, we expect that the mesh size of the network will also increase or decrease considerably. The mesh size of the swollen membrane controls the molecular transport phenomena through the swollen membranes. In order to calculate the mesh size from the equilibrium swelling data (determined as explained above) the Peppas–Merrill equation²⁵ is used to determine the number average molar mass between cross-links and cross-link density.

Moisture absorption. The moisture absorption of RCMs containing MC and GO with different ratios of GLA was determined by the following method. All membrane samples were cut in the dimensions of 3 cm × 3 cm. Then, the samples were dried in an oven at 60 °C to take remove the moisture, after which they were instantaneously weighed to obtain the initial weights (W₁). The samples were kept in a 70% constant relative humidity environment generated in a hermetic glass container with aqueous saturated NaCl solutions. The RCMs were taken from the glass container and weighed immediately to obtain the final weight (W₂) after a period of 24 h. The moisture absorption of the membrane samples was then calculated using the following eqn (2):²⁶

SEM analysis. The cross sectional morphology of the RCMs was examined under a ZEISS, EVO 18, scanning electron microscope (SEM). For SEM analysis, RCMs were then coated with gold in a sputter coater and observed under SEM at an accelerating voltage of 15 kV.

Mechanical properties. A Zwick Roell machine (ZO10), with a RCM sample of 22 mm in length and 5 mm in width at a cross-head speed of 10 mm min⁻¹ at 25 °C (as per ASTM D882-95a), was used to measure the mechanical properties of the uncross-linked and cross-linked RCMs.

Thermo-gravimetric analysis. Thermo-gravimetric analysis (TGA) of uncross-linked and cross-linked RCMs was carried out in a Perkin-Elmer-PYRIS 1 TGA thermal analyzer in a dynamic atmosphere of nitrogen (flow rate = 30 cm³ min⁻¹). The samples were heated in an alumina crucible at a heating rate of 10 °C min⁻¹ over a temperature range of 30–600 °C.

In vitro drug permeation study. To prepare a standard curve for diltiazem hydrochloride release in a phosphate buffer of pH 7.4, dilute solutions were made and UV light absorption was checked at λ_max of 236 nm. Then, the standard curve was prepared by plotting absorbance data against drug concentration. Since the value of R² is 0.990, there is an acceptable linear relationship between the absorbance and the drug concentration. Also, the slope value would be valid in order to calculate the drug concentration during the cumulative release.

In vitro diltiazem release through prepared RCMs was performed in phosphate buffer of pH 7.4 (PBS) using a Franz diffusion cell. A dialysis membrane (LA390, average flat width 25.27 mm, average diameter 15.9 mm and capacity of approx. 1.99 mL cm⁻¹) made from cellulose acetate was used as a human skin replica for determining diltiazem release through a rate controlling membrane. The membrane was mounted between the donor and receptor compartments of the diffusion cell.²⁷ The RCM was placed on the cellulose acetate membrane. 1 mL of drug solution (diltiazem hydrochloride) was poured into the donor compartment which was open to air. The receptor compartment was filled with a phosphate buffer of pH 5.6 to match the pH of skin (5.4–7.4). The intact assembly was fixed on a hot plate magnetic stirrer, and the solution in the receptor compartment was continuously stirred using magnetic beads to maintain a steady temperature at 32 ± 0.5 °C, since the normal skin temperature of humans is 32 °C.²⁸ The aliquots were withdrawn at different time intervals and were replenished with an equal volume of fresh buffer. The diltiazem hydrochloride content in the aliquots was analysed spectrophotometrically at 236 nm after matching the values with a standard calibration plot.

Ex vivo drug permeation study. Ex vivo skin permeation of diltiazem hydrochloride from all samples through depilated rat abdominal skin was conducted using a modified Franz diffusion cell. Skin permeation was conducted in accordance with the Helsinki declaration and animal care and facilities in Principles and Methods of Toxicology.²⁹

The abdominal skin of the Long-Evans rat was used. Hairs on the abdominal area were shaved due to the sacrificing by prolonged chloroform inhalation. Abdominal skin was excised and the subcutaneous tissue was surgically removed. The dermis side was wiped with isopropyl alcohol (IPA) in order to remove residual adhering fat. The skin was then washed with distilled water. For 6 h, the as prepared skin was treated with 2 M sodium bromide solution in water. A cotton swab moistened with distilled water was used to separate the epidermis. Then, the epidermis sheet was cleaned by washing it with distilled water. The skin so prepared was wrapped in aluminium foil and stored in a freezer at −20 °C until further use.³⁰

The skin membranes were first hydrated for 30 minutes in the buffer solution (pH 7.4) at room temperature to remove extraneous debris and leachable enzymes. They were then placed between the donor and receptor compartments of the cells, with the dermal side in direct contact with the receptor medium. Approximately 100 mL of the phosphate buffer (pH 7.4) were placed in the receptor compartment. The temperature was maintained at 37 ± 0.5 °C using a thermostatic water bath. This whole assembly was kept on a magnetic stirrer and the solution in the receiver compartment was continuously stirred during the whole experiment using a magnetic bar. The aliquots were withdrawn at different time intervals and were replenished with an equal volume of fresh buffer. Absorbance of all of the samples was read spectrophotometrically at 236 nm, taking phosphate buffer solution (pH 7.4) as the blank. The amount of drug permeated per square centimetre at each time interval was calculated and plotted against time. A similar set was run simultaneously using the patch (without drug) at the donor compartment as a skin patch control system to avoid the influence of inherent extracts from the skin or leaching of any material from the patch without drug on the absorbance at 236 nm.

3. Results and discussion

ATR-FTIR spectroscopy

ATR-FTIR measurements were conducted on uncross-linked and cross-linked RCMs to determine whether cross-linking is taking place between GO and MC through GLA (Fig. 2). As shown in Fig. 2, the spectra are characterized by relevant molecular stretching and bending vibrations. The FTIR spectrum of GO in the uncross-linked RCM reveals a structure that is similar to pure GO, as described in detail in a previous publication.³¹ Pure GO has absorption bands related to O–H stretching at 3442 cm⁻¹, owing to the presence of a large number of hydroxyl groups in the GO backbone. The peak at 1615 cm⁻¹ is due to the C=C stretching from the skeletal vibrations of un-oxidized graphitic domains. The epoxy and alkoxy groups of pure GO give the absorbance bands at 1218 and 1056 cm⁻¹, respectively. The broad peak around 3400 cm⁻¹ corresponds to the O–H stretching of pure MC. The region from 1500 to 1250 cm⁻¹ corresponds to vibration modes of methylcellulose groups. The peaks around 1700 to 1600 cm⁻¹ correspond to the bending mode of water molecules contained in the material. In the case of uncross-linked RCM (F1), the area under the O–H stretching peak substantially decreases with respect to pure GO as the population of –OH groups decreases in the uncross-linked MC/GO RCM (F1). The spectrum of the cross-linked RCM (F5) is shown in Fig. 2. It is clear that the broadness of the O–H stretching peak further decreases compared to the uncross-linked RCM (F1). Perhaps this is due to the cross-linking through hydroxyl groups of GO and MC by GLA. We can also establish the cross-linking of RCM (F5) by analysing the peak of the alkoxy group at 1056 cm⁻¹. It is clear from Fig. 2 that the sharpness and area of the alkoxy group peak increases due to the increasing presence of alkoxy groups in the cross-linked RCM (F5), which are coming from the cross-linker GLA. It can therefore be concluded that GLA can be effectively used for the cross-linking of GO and MC. The scheme illustration of the cross-linking between MC and GO by GLA is shown in Fig. 3.

Fig. 2 ATR-FTIR spectra of RCMs.

Fig. 3 Schematic of cross-linking between MC and GO by GLA.

Solubility and swelling measurements

The cross-linking between MC/GO in RCMs (F2, F3, F4, and F5) can be confirmed by the percentage of solubility and swelling in water (Fig. 4). It is noticed that the uncross-linked RCM is around 96% soluble in water within 24 h. As observed in Fig. 4, the solubility of the cross-linked RCM (F2) with 3 wt% GLA suddenly drops to 26% and then there is no significant change in solubility with increasing concentration of GLA from 3 to 12 wt%; i.e., from F2 to F5. The solubilities of pure MC RCM and cross-linked MC RCM (12 wt% GLA) are around 98% and 25%, respectively, which shows the solubility of the cross-linked MC based RCM (without GO) in water is similar to that of the F2 formulation. However, the F5 (cross-linking MC/GO with 12 wt% GLA) formulation shows less solubility than the cross-linked MC (without GO) formulation.

Fig. 4 Solubilities of the RCMs.

The swelling of the RCMs is calculated from the weight difference relative to the final weight. The results of the swelling study are presented in Fig. 5. Uncross-linked RCM shows a swelling degree (SD) of 1.88% and the SD decreases with increasing GLA concentration from 1.4 to 0.79% in the case of cross-linked RCMs. Pure MC RCM shows a swelling degree of around 3.18%, which is more than that of the uncross-linked MC/GO RCM. We have also found that the cross-linked MC RCM shows a rate of swelling similar to the F2 formulation, which is much greater than the F5 formulation, where the cross-link density is at a maximum; this may be due to the presence of GO. Therefore, the solubility and swelling measurements established the fact that GLA is chemically bonded to GO and MC. In addition, by increasing the concentration of GLA from 3 to 12 wt% there was no significant change in solubility and SD.

Fig. 5 Swelling index of the RCMs.

Molar mass between cross-links and cross-link density

The capability of RCMs to release diltiazem hydrochloride is a function of cross-link density. For the purpose of knowing the cross-linking and cross-link density of the polymer network, two important parameters have been calculated, i.e., the molar mass between cross-links (M_c) and cross-link density (d_x), dependent on the equilibrium swelling study.^32,33 The elastic forces of the polymeric membranes are inversely proportional to the molar mass of the polymer between the points of cross-linking.³² Thus, the molar mass between two junction points in a network would be rigid and exhibit limited swelling. When M_c is large, the network is more elastic and swells rapidly. The value of M_c is calculated using the following eqn (3):

M_c = −ρ_pV_sΦ^1/3[ln(1 − Φ) + Φ + χΦ²]⁻¹ (3)

where, Φ is the volume fraction of the polymer in the swollen state, ρ_p is the density of the polymer, V_s is the molar volume of solvent and χ is the interaction parameter. The cross-linked density of the RCMs (d_x) is calculated using the following eqn (4):where, v is the specific volume of the polymer. The results of M_c and d_x are presented in Table 2. The value of the number average molar mass, M_c determined by equilibrium swelling is observed as 2494 g mol⁻¹ for uncross-linked RCM (F1). It is clear from Table 2 that in the case of the cross-linked RCMs, the value of M_c decreases from 2446 to 1132 g mol⁻¹ and d_x increases from 2.964 to 7.837 with increasing GLA concentration. It can therefore be concluded that M_c is inversely related to d_x.³⁴ Since there was no cross-linker added to the pure MC RCM, the cross-linking density is very low in this case. For cross-linked MC, by using 12 wt% GLA, the cross-linking density shows a result similar to that of F2, but much less than F5.

Table 2 Molar mass between cross-links and cross-link densities of the cross-linked RCMs (n = 3)

Formulations	Thickness (μm)	Mc^a (g mol^−1)	dx^a × 10^3	N^a
a Mc is the molar mass between cross-links, dx is the cross-link density and N is a release parameter.
F1	54.3 ± 0.02	2494	2.964	0.882
F2	63.3 ± 0.04	2446	3.215	0.917
F3	67.5 ± 0.04	1997	5.48	0.958
F4	68.8 ± 0.06	1577	6.718	0.987
F5	68.6 ± 0.05	1132	7.837	0.997
Pure MC	56.7 ± 0.03	2583	2.743	0.912
Cross-linked MC	62.8 ± 0.03	2174	3.972	0.925

---

Moisture absorption

Moisture absorption can be considered as a parameter for evaluating the extent of cross-linking in cross-linked RCMs. Uncross-linked RCM (F1) shows the maximum moisture absorption of 9.7% in 70% relative humidity at 22 °C, as shown in Fig. 6. Pure MC RCM shows the maximum moisture absorption of 10.9% in 70% relative humidity at 22 °C. Cross-linked MC RCM shows the same moisture absorption as F2, which is much greater than the F5 formulation. After cross-linking with 12 wt% GLA, the water absorption is reduced to 2.4% (F5). It is further observed that the moisture absorption of cross-linked RCMs decreases with the increasing concentration of GLA from 3 to 12 wt% (F1 to F5). This is due to more cross-linking between GO and MC through hydroxyl groups with increasing concentration of GLA, which indirectly decreases the population of polar groups. Therefore, GLA can be effectively used to improve the properties of MC/GO RCMs, as it forms inter and intra cross-linkages between MC and GO molecules, thereby reducing the hydroxyl groups in the cross-linked membranes. Thus, a decreasing trend of moisture absorption is observed with increasing GLA concentration or cross-link density.

Fig. 6 Moisture absorption study of RCMs.

SEM analysis

SEM images of uncross-linked and cross-linked RCMs are shown in Fig. 7. It can be seen that there is a significant difference in the cross-sectional morphology of uncross-linked and cross-linked RCMs. Uncross-linked RCMs display a cross-sectional morphology of porous and fibrillose structure. The porous and fibrillose structure of uncross-linked RCM is destroyed after cross-linking and the fracture surface becomes rougher with the increase in GLA concentration from 3 to 12 wt%. This kind of cross-sectional morphology indicates the brittle fracture of cross-linked RCMs. All of the uniform pores of the polymer matrix are destroyed with respect to uncross-linked RCM, which is responsible for slow drug permeation through cross-linked RCM. The rate of drug permeation is less in the case of cross-linked RCM, depending on cross-link density. Therefore, the SEM analysis can be correlated with the in vitro and ex vivo drug permeation study.

Fig. 7 Cross-sectional views of RCM formulation.

Mechanical properties

The mechanical properties of uncross-linked and cross-linked RCMs were investigated. Good mechanical properties are important for RCMs intended for use in the controlled release of drugs.²⁶ Fig. 8(a and c) illustrates the tensile strength and Young's modulus of RCMs and both are increasing with the concentration of cross-linker. Although the tensile strength and Young's modulus are increasing, we can see a continuous fall or decrease in elongation at break with the increasing concentration of cross-linker, which is shown in Fig. 8(b). Pure MC RCM shows a tensile strength and Young's modulus less than cross-linked MC RCM. Again, cross-linked MC RCM shows a lower tensile strength and Young's modulus compared to F5. From the elongation break study of pure MC RCM, we observed that it had a maximum percentage elongation at break. The decrease in elongation at break, with an increase in the cross-linker concentration, is due to the formation of a more rigid structure of the membranes, as also observed in the SEM analysis in Fig. 7. These results arise from the formation of a three-dimensional network structure by cross-linking. The variation in mechanical properties with cross-linking suggests that the mechanical properties of the cross-linked membranes can be tuned by regulating the concentration of the cross-linker.

Fig. 8 Mechanical properties of the RCM formulations.

Thermo-gravimetric analysis

In TGA a sample loses mass with increasing temperature, which directly provides information about the thermal stability and the degradation mechanism for uncross-linked and cross-linked RCMs. The thermal degradation of the uncross-linked and cross-linked RCMs as a function of GLA content is presented in Fig. 9, where it can be seen that a slight weight loss (∼3 to 6 wt%) of uncross-linked and cross-linked RCMs started below 100 °C. This indicates that the initial weight loss is probably due to the moisture and high water-retention capacity of MC and GO.²³ In Fig. 9, with the increase in the percentage of GLA from 3 to 12 wt%, we can see an increase in the degradation temperature (at 10% weight loss) from 310 °C to 400 °C, indicating that the chemically cross-linked RCMs are more thermally stable than the uncross-linked RCM. This phenomenon is similar to that reported by Park and Ruckenstein.³⁴ Above 460 °C, all curves are approaching a plateau value, as mainly char residue remains. It could be explained that the GLA acts as a cross-linker for RCMs, which enhances the overall thermal stability of the system, and occasionally assists the char formation after thermal decomposition.

Fig. 9 Thermal stability of all formulations.

In vitro drug permeation and drug release kinetics study

Fig. 10 presents the permeation profiles of diltiazem hydrochloride through RCMs (F1, F2, F3, F4, F5, pure MC and cross-linked MC). The plots of cumulative percentage of diltiazem hydrochloride release from RCMs, versus time are presented. The experiment was carried out for 8 continuous hours. The diltiazem hydrochloride release profile from pure MC RCM shows a quick permeation of 95 to 98% of the drug within 8 h, which is more than for the uncross-linked MC/GO RCM (F1). The rate of release of diltiazem hydrochloride from cross-linked MC RCM was 50% in 8 h, which is almost identical to that of cross-linked RCM (F2), and is much greater than that of F5. The diltiazem hydrochloride release profile from uncross-linked RCM (F1) shows a quick permeation of 90 to 95% of the drug within 8 h. The rate of diltiazem hydrochloride release from cross-linked RCM (F2) was 52% release in 8 h and there were signs of equilibration within 5–7 h, indicating a slower and more controlled release behavior than F1. At higher doses of GLA, the profile shows signs of sustained permeation, especially with F5, when 37% drug permeation takes place at 8 h. There is no significant change in percentage drug release from F2 to F5, which can be correlated with the fracture morphology of F2 and F5. From the SEM images (Fig. 7), it is clear that there is no change in the cross-sectional morphology of F2 and F5 with increasing concentration of GLA from 3 to 12 wt%, which is reflected in the in vitro drug release pattern. Results also indicate that the permeation of diltiazem hydrochloride from RCMs follows first order patterns; an initial burst release is observed and later, linearity is maintained in the release profiles. The drug diffusion from cross-linked RCMs depends on the GLA concentration, which is indirectly related to the cross-link density of the RCMs. An increase in GLA concentration decreases the membrane porous structure, which in turn reduces the drug diffusion rate. As the concentration of GLA is increased in the RCM, drug release is decreased appreciably. This may be due to the increased rigidity, reduced porosity and decreased swelling degree of the RCM at higher cross-link density, thereby hindering the transport of drug molecules through the membrane. It can therefore be concluded that the step involving the drug penetrating through the cross-linked RCMs is rate-limiting. Hence, the rate of drug release is considered to be membrane controlled.

Fig. 10 In vitro drug permeation through RCM formulations.

The kinetics of release profiles were investigated using the modified Korsmeyer–Peppas kinetic model.^35,36 Eqn (5) is used to account for the drug release mechanism from cross-linked membranes.

where M_t is the amount of drug released at time t, M_∞ is the total amount of drug in the donor compartment, and n indicates the type of release mechanism. The n values have been calculated and are given in Table 3. The diffusional coefficient n represents the mechanism of the drug transportation. In the equation, n = 0.50 is the Fickian release and n = 1.00 is the Case-II release. Between these two limiting cases, the anomalous release behavior is found, which is intermediate between Fickian and Case-II. It is defined as the anomalous release when n is between 0.50 and 1.0 in the semi-empirical equation. The values of the diffusional coefficient n are found to vary in the range of 0.427–0.623. Thus, the uncross-linked RCM (F1) follows the anomalous drug diffusion controlled release mechanism. The value of n for the cross-linked RCMs is much less than that of the uncross-linked RCM. Accordingly, the release mechanism of the cross-linked RCMs (F2, F3, F4 and F5) is based on the Fickian diffusion. Table 3 shows that the release kinetics are best explained by first order kinetics, since the plots show the highest linearity (r² > 0.98) for all the formulations followed by Higuchi and zero order kinetics. As the concentration of GLA is increased, the n values shift towards first order kinetics.

Table 3 Drug release kinetics of the developed RCM formulations (n = 3)

Formulations	Zero order	First order	Higuchi	Korsmeyer–Peppas
	r^2	r^2	r^2	r^2	N
F1	0.908 ± 0.008	0.983 ± 0.005	0.989 ± 0.004	0.992 ± 0.007	0.623 ± 0.04
F2	0.896 ± 0.012	0.991 ± 0.003	0.992 ± 0.004	0.975 ± 0.003	0.387 ± 0.03
F3	0.914 ± 0.006	0.993 ± 0.003	0.993 ± 0.002	0.981 ± 0.003	0.481 ± 0.01
F4	0.913 ± 0.004	0.991 ± 0.002	0.992 ± 0.005	0.995 ± 0.002	0.415 ± 0.04
F5	0.897 ± 0.002	0.993 ± 0.00	0.993 ± 0.004	0.997 ± 0.00	0.427 ± 0.02

---

Ex vivo permeation study

Ex vivo permeation profiles of diltiazem hydrochloride across rat skin with uncross-linked and cross-linked RCMs are illustrated in Fig. 11. Flux values of diltiazem hydrochloride from different RCMs decrease as follows: F5 > F4 > F3 > F2 > F1 (control). The flux (μg cm⁻² h⁻¹) of diltiazem hydrochloride is calculated from the slope of the plot of the cumulative amount of diltiazem hydrochloride permeated per cm² of skin at steady state, against time, using linear regression analysis. The permeation coefficients were obtained from the steady-state flux values using the following eqn (6).

P = J/C_o (cm h⁻¹) (6)

where P is the permeation coefficient, C_o is the initial drug concentration in the drug compartment; J represents the steady state flux obtained from eqn (7).

J = dQ/Adt (μg cm⁻² h⁻¹) (7)

where Q indicates the quantity of substances crossing the rat skin, A is the area of the exposed rat skin and t is the time of exposure. As the concentration of GLA increases in cross-linked RCMs, the amount of drug permeated decreases. This may be a result of the initial rapid dissolution of the hydrophilic polymer when the patch is in contact with the hydrated skin, which results in the accumulation of high amounts of drug on the skin surface and thus leads to the saturation of the skin with drug molecules at all times. A lower drug release rate from RCMs containing higher concentrations of GLA may be due to the relatively more hydrophobic nature of the cross-linked RCMs, which have less affinity for water. This result indicates a decrease in the thermodynamic activity of the drug and drug permeation from cross-linked RCMs. The differences found between in vitro and ex vivo drug permeation studies are related to the circumstances of the test, the capability of the artificial dialysis membrane, rat skin and the inter-species characteristics. It is seen that the ex vivo drug permeation rate is slightly lower than the in vitro drug permeation rate. This can be explained as follows. During ex vivo drug permeation, the membrane and the skin remain in contact for a very long time, resulting in the membrane becoming dehydrated, thus hindering the drug penetration and sustained drug release.

Fig. 11 Ex vivo drug permeation of RCMs through rat skin.

4. Conclusions

Novel RCMs have been successfully developed by cross-linking MC and GO with GLA. The ATR-FTIR spectra analysis, solubility studies, swelling studies, cross-link density and molar mass between cross-linking of various formulations of RCMs confirm the successful incorporation of GO platelets into the MC matrix, and cross-linking between MC and GO. SEM analysis supports the good distribution of GO in the MC matrix and the cross-sectional view of RCMs confirm that the porous and fibrillose structure of the uncross-linked RCM is destroyed after cross-linking. Mechanical properties and TGA analysis suggest that as the cross-link density increases, the RCMs become more thermally stable. In vitro and ex vivo studies established that the drug permeation rate is decreased with the increase in the cross-linking density. It can be explained in the way that the porous structure of the uncross-linked RCM got destroyed with the addition of the cross-linker, which actually hinders the drug permeation rate. Therefore, the developed MC/GO rate controlling membranes demonstrate the permeation of the cardiac drug (diltiazem hydrochloride) in a sustained way and can be used for the development of a transdermal patch.

Acknowledgements

G. Sarkar earnestly acknowledges the financial support from the University Grant commission, the Govt. of India for Rajiv Gandhi National Fellowship (RGNF) scheme. N. R. Saha thanks the University Grant commission for his fellowship. I. Roy and A. Bhattacharyya thank the TEQIP, India, for their fellowships. Also, we thank the Centre for Research in Nanoscience and Nanotechnology, the University of Calcutta for providing SEM facilities.

Notes and references

1. A. S. Chauhan, S. Sridevi, K. B. Chalasani, A. K. Jain, S. K. Jain, N. K. Jain and P. V. Diwan, J. Controlled Release, 2003, 90, 335–343.
2. J. A. Cordero, L. Alarcon, E. Escribano, R. Obach and J. Domenech, J. Pharm. Sci., 1997, 86, 503–508.
3. J. Y. Fang, C. L. Fang, C. T. Hong, H. Y. Chen, T. Y. Lin and H. M. Wei, Eur. J. Pharm. Sci., 2001, 12, 195–203.
4. J. Hadgraft, J. du Plessis and C. Goosen, Int. J. Pharm., 2000, 207, 31–37.
5. P. R. Rao and P. V. Diwan, Pharm. Acta Helv., 1997, 72, 47–51.
6. X. Qiu, S. Tao, X. Ren and S. Hu, Carbohydr. Polym., 2012, 88, 1272–1280.
7. A. H. Castro Neto, F. Guinea, N. M. R. Peres, K. S. Novoselov and A. K. Geim, Rev. Mod. Phys., 2009, 81, 109–162.
8. M. J. Allen, V. C. Tung and R. B. Kaner, Chem. Rev., 2010, 110, 132–145.
9. A. K. Geim and K. S. Novoselov, Nat. Mater., 2007, 6, 183–191.
10. H. Bao, Y. Pan, Y. Ping, N. G. Sahoo, T. Wu, L. Li, J. Li and L. H. Gan, Small, 2011, 7, 1569–1578.
11. A. K. Geim, Science, 2009, 324, 1530–1534.
12. M. Kakran and L. Li, Key Eng. Mater., 2012, 508, 76–80.
13. F. Debeaufort and A. Voilley, Cellulose, 1995, 2, 205–213.
14. M. K. Bain, D. Maity, B. Bhowmick, D. Mondal, M. R. Mollick, G. Sarkar, M. Bhowmik, D. Rana and D. Chattopadhyay, Carbohydr. Polym., 2013, 91, 529–536.
15. W. I. Cha, S. H. Hyon and Y. Ikada, Makromol. Chem., 1993, 194, 2433–2441.
16. S. E. O'Connor, A. Grosset and P. Janiak, Fundam. Clin. Pharmacol., 1999, 13, 145–153.
17. K. Kosuge, Y. Jun, H. Watanabe, M. Kimura, M. Nishimoto, T. Ishizaki and K. Ohashi, Drug Metab. Dispos., 2001, 29, 1284–1289.
18. P. R. Rao and P. V. Diwan, Drug Dev. Ind. Pharm., 1998, 24, 327–336.
19. Z. D. Roy and E. Manokian, J. Pharm. Sci., 1995, 84, 1190–1196.
20. W. S. Hummers Jr and R. E. Offerman, J. Am. Chem. Soc., 1958, 80, 1339.
21. W. Fan, Q. Lai, Q. Zhang and Y. Wang, J. Phys. Chem. C, 2011, 115, 10694–10701.
22. T. K. Ghosh, S. Gope, D. Mondal, B. Bhowmick, M. R. Mollick, D. Maity, I. Roy, G. Sarkar, S. Sadhukhan, D. Rana and D. Chattopadhyay, Int. J. Biol. Macromol., 2014, 66, 338–345.
23. M. F. Huang, J. G. Yu and X. F. Ma, Polymer, 2004, 45, 7017–7023.
24. J. B. Xu, J. P. Bartley and R. A. Johnson, J. Membr. Sci., 2003, 218, 131–146.
25. J. Berger, M. Reist, J. M. Mayer, O. Felt, N. A. Peppas and R. Gurny, Eur. J. Pharm. Biopharm., 2004, 57, 19–34.
26. K. S. Y. Hemant and H. G. Shivakumar, J. Pharma Res., 2010, 9, 197–203.
27. G. Sarkar, N. R. Saha, I. Roy, A. Bhattacharyyaa, M. Bose, R. Mishra, D. Rana, D. Bhattacharjee and D. Chattopadhyay, Int. J. Biol. Macromol., 2014, 66, 158–165.
28. H. E. Boddé, P. E. Roemelé and W. M. Star, Photochem. Photobiol., 2002, 75, 418–423.
29. J. Jaiswal, R. Poduri and R. Panchagnula, Int. J. Pharm., 1999, 179, 129–134.
30. H. Takeuchi, Y. Mano, S. Terasaka, T. Sakurai, A. Furuya, H. Urano and K. Sugibayashi, Exp. Anim., 2011, 60, 373–384.
31. E. Y. Choi, T. H. Han, J. Hong, J. E. Kim, S. H. Lee, H. W. Kim and S. O. Kim, J. Mater. Chem., 2010, 20, 1907–1912.
32. A. Dufresne, L. Reche, R. Marchessault and M. Lacroix, Int. J. Biol. Macromol., 2001, 29, 73–82.
33. T. R. R. Singh, P. A. McCarron, A. D. Woolfson and R. F. Donnelly, J. Appl. Polym. Sci., 2009, 45, 1239–1249.
34. J. S. Park and E. Ruckenstein, Carbohydr. Polym., 2001, 46, 373–381.
35. F. Hammann and M. Schmid, Materials, 2014, 7, 7975–7996.
36. S. Y. Lin, C. J. Lee and Y. Y. Lin, J. Controlled Release, 1995, 33, 375–381.

---