Preparation,characterization,and in vitro evaluation of antibacterial sol–gel coated cotton textiles with prolonged release of curcumin

Abstract

The aim of this study was to develop a textile-based sol–gel release system with antibacterial properties intended for medical applications. Curcumin (a herbal compound extracted from Curcuma longa L.) was embedded in an acid-catalyzed silica xerogel coating on cotton textiles. The release of curcumin was found to be diffusion-controlled, as expressed by the Higuchi and the first-order models. Matrix dissolution release was ruled out by the absence of dissolved silica in the release medium. Results from nitrogen sorption analysis confirmed the presence of mesopores in the xerogel coating, and an increase in both the BET surface area from 1.95 to 2.23 m² g⁻¹ and in the mesopore volume from 0.003 to 0.004 cm³ g⁻¹ after the sol–gel treatment. Thicker xerogel coatings therefore provided greater porosity for curcumin entrapment, leading to retardation of the diffusion processes and consequently extended release profiles. The treated samples showed better antibacterial properties against Staphylococcus aureus (99% reduction) compared with Escherichia coli (37% reduction). Overall, the developed sol–gel coated textiles demonstrate promising potential for use as wound dressing materials.

Keywords

sol–gel coating controlled release cotton curcumin antibacterial properties

Recently, sol–gel coating techniques have been employed by several research groups to improve the properties of, or add special features to, traditional textile products. One of the most widely-used sol–gel systems utilizes an alcoholic solution of alkoxysilane as a sol–gel precursor that is transformed into a porous solid network via hydrolysis and condensation reactions. To broaden the range of textile functionalities, direct modification of the sol–gel precursor or the uses of additives are two major approaches. Sol–gel coated fabrics have been demonstrated to acquire a wide range of additional functions according to the modification route: fabrics have been produced showing fire-retardant,¹ anti-static,^2,3 water/oil repellent,^2–4 insect-repellent,⁵ anti-microbial,⁶ and catalytic characteristics.⁷ The additive initially introduced to the precursor solution during preparation may be leached out from the solidified sol–gel network through the matrix channels, or as a result of matrix dissolution when in contact with liquid media.⁸ The possibility of gaining control over the sol–gel structure by varying parameters such as the type of precursor, temperature, pH, precursor–water ratio, and consequently tuning its release characteristics, have paved the way for the development of sol–gel controlled release systems.⁹

Since one important and rapidly growing area of the textile industry is the medical sector,¹⁰ effort is currently being put into coating textiles with sol–gel technology for medical use, mainly as wound dressing materials.¹¹ Controlled-release principles based on sol–gel systems have been applied to textiles by Mahltig et al.,⁶ who developed antimicrobial sol–gel coated textiles with the use of silane precursors and four different biocides, i.e. inorganic biocide (silver metal and its nitrate salt), and organic biocides (Octenidine and CTAB) on polyester/cotton. More recent work has reported the use of silver, copper, and hexadecyltrimethylammonium p-toluenesulfonate (HTAT) for developing antimicrobial viscose textiles,¹² and capsaicin (chili pepper extract) for antimicrobial wool fabrics.¹³ Due to the potential toxicity to humans, and environmental concerns about the use of some of these biocides, naturally-occurring and biocompatible compounds are preferable for use in wound dressings and other medical textile applications. Curcumin (1,7-bis(4-hydroxy-3-methoxyphenyl)-1,6-heptadiene-3,5-dione) is a herbal compound found in turmeric (Curcuma longa L.). The WHO (World Health Organization) and FAO (Food and Agriculture Organization) have verified its safety as a natural food colorant (C.I. Natural Yellow 3).¹⁴ As shown in Figure 1, curcumin exists in at least two tautomeric keto and enol forms, with equilibrium strongly favoring the enol form. The enol structure allows inter- and intramolecular hydrogen bonding to occur. Pharmacological studies have confirmed that curcumin possesses anti-inﬂammatory, antifungal, antitumor, and antioxidant characteristics. It has been shown to promote wound healing in animals by enhancing cell proliferation and effective free-radical scavenging.¹⁵ It is believed that the methoxyl and hydroxyl groups are responsible for the microbial activity, although the actual mechanism is not fully understood.¹⁴ In light of the benefits of curcumin in wound healing, along with its ubiquitous availability and environmental compatibility, this herbal substance may be a good candidate for wound dressing applications.

Figure 1.

Chemical structure of curcumin: (a) enol form and (b) keto form.

Although much research has dealt with sol–gel release systems, data on release characteristics when textiles are used as flexible, porous substrates is limited. More detailed examinations of structural, physical and controlled release properties are therefore needed. The originalities of this work lie in firstly the use of a biocompatible herbal extract for the development of controlled release sol–gel coated textiles, and secondly, the use of mathematical drug delivery models to describe the release kinetics and mechanism since it has been realized that controlled-release behaviors are sensitive to a variety of parameters, especially those concerning the nature of the delivery device.

The goal of this investigation was to prepare and evaluate the potential of a sol–gel coating on cotton fabrics for tailoring the controlled release of curcumin. Other important factors related to potential applications, such as the appearance, weight, mechanical performance, and antibacterial properties of the material, are also reported.

Experimental details

Materials

A plain weave cotton fabric (mass per unit area = 76 g m⁻²) was used, having 42 yarns/cm in the warp direction and 35 yarns/cm in the weft direction. Curcumin (95% purity) was purchased from Changsha Kangrong Biological-Tech Co. Ltd. (Hunan, China). Tetraethyl orthosilicate (TEOS, >99%) and anhydrous sodium acetate (>99%) were supplied from Sigma–Aldrich (Germany). Analytical grade hydrochloric acid (35.4%) was obtained from Loba Chemie, Pvt. Ltd. (Mumbai, India). Absolute ethanol was purchased from RCI Labscan Ltd. (Bangkok, Thailand). Non-ionic surfactant (Sandoclean) was supplied by Clariant Ltd. (Bangkok, Thailand). Sodium hydroxide (>99%) and glacial acetic acid were obtained from Merck (Darmstadt, Germany).

Preparation of sol–gel coated fabrics containing curcumin

The cotton fabrics were scoured using a solution containing 3 g/L non-ionic surfactant and 5 g/L NaOH at a liquor ratio of 30:1 at 90℃ for 30 min. Silica sols were prepared by room-temperature mixing of TEOS, 0.01 M HCl, ethanol, and curcumin in various compositions according to Table 1. In the sample name, the notation C represents control samples (without TEOS) and T refers to treated samples (with TEOS). The numbers 10, 20, and 30 placed after T correspond to the amount of TEOS in millimoles. The numbers 30, 50, and 70 appearing after a hyphen represent the curcumin loading in milligrams. The TEOS:water:HCl molar ratios for T10, T20, and T30 were 1:42:0.0075, 1:21:0.00375, and 1:14:0.0025, respectively. The total solution volume was fixed at 25 mL. All solutions were homogenized by sonication (VGT-2013HTD Portfolio) for 30 s, giving clear, yellow solutions. Then, a 4 cm × 10 cm piece of cotton fabric was immersed in each solution for 10 minutes, taken out, and then padded at 2 kg cm⁻² pressure (Labtec, Newave Lab. Equipments, Co., Ltd.).

Table 1.

Composition of starting solutions used in preparing the fabric samples

Sample label	TEOS (mmol)	Curcumin (mg)	0.01 M HCl (mL)	EtOH (mL)
C-30	–	30	7.50	17.50
T10-30	10	30	7.50	15.25
T20-30	20	30	7.50	12.50
T30-30	30	30	7.50	10.75
C-50	–	50	7.50	17.50
T10-50	10	50	7.50	15.25
T20-50	20	50	7.50	12.50
T30-50	30	50	7.50	10.75
C-70	–	70	7.50	17.50
T10-70	10	70	7.50	15.25
T20-70	20	70	7.50	12.50
T30-70	30	70	7.50	10.75

The “% wet pickups” were calculated according to the following equation:¹⁶

% wetpickup = \frac{{(wt . ofwetfabricafterpad - wt . ofdry, untreatedfabric) \times 100}}{wt . ofdry, untreatedfabric}

The treated fabrics were dried under ambient conditions for not less than 48 h before final dry weights were recorded. The “% add-on” values are calculated as follows:¹⁷

% add-on = \frac{{wt . ofdry, treatedfabric - wt . ofdry, untreatedfabric \times 100}}{wt . ofdry, untreatedfabric}

In vitro curcumin release study

A fabric sample size of 4 cm × 5 cm was used for the dissolution study. Prior to the experiment, all samples were immersed in 50 mL of ethanol and shaken in a water bath shaker (Dye Plus DL-2003) for 1 h at 40℃, with a rotation speed of 60 rev/min, to remove unbound curcumin. The release medium was 100 mL acetate buffer containing 3% v/v ethanol and 0.5% v/v Tween 80 (pH 5.64 ± 0.01).¹⁵ The sample was immersed in the release medium at 32 ± 0.5℃ in a closed glass vessel and shaken in the same water bath at 60 rev/min. A 4 mL sample was collected, and this was replaced with fresh buffer solution of equal volume at 1, 2, or 4 h intervals (up to 24 h). The collected samples were kept refrigerated until analysis. The amount of released curcumin was determined from the absorbance values of the samples (UV-visible spectrophotometer, Perkin-Elmer Lambda 25) at 426 nm. All absorbance data points were in the linear range of the calibration curves with R² values of at least 0.98. To determine the actual amount of curcumin in each sample, curcumin extraction was done by shaking in release medium at 40℃, 60 rev/min for 48 h, followed by 30-min sonication and mechanical squeezing. All experiments were done in triplicate.

Characterization of sample specific area and porosity

Nitrogen adsorption and desorption isotherms were obtained at 77 K using a Quantachrome Autosorb iQ automated gas sorption system. The samples were outgassed at 80℃ prior to measurement to remove all adsorbed water. Sample specific surface area was calculated based on the Brunauer–Emmett–Teller (BET) method in the relative pressure (P/P₀) range of 0.05–0.30.¹⁸ Pore size distribution was analyzed from the desorption data using the conventional Barrett–Joyner–Halenda (BJH) method.¹⁹

Morphological study

Surface morphology was examined using a scanning electron microscope (JEOL, JSM-5410LV) operating at an accelerating voltage of 5 kV and 100 × magnification. All the samples were sputtered with gold prior to observation.

Antimicrobial testing

Samples were tested for their antibacterial activity according to AATCC Test Method 100-1999. All samples were sterilized by 15-min UV exposure on each side. Four circular swatches (4.8 cm in diameter) of the samples were inoculated with 1.0 mL of inoculums containing either 3.27 × 10⁵ CFU (colony forming units) of Gram-positive S. aureus (ATCC 6538) or 5.25 × 10⁵ CFU of Gram-negative E. coli (ATCC 25922). After incubation (37℃, 24 h), the percentage reduction (R) in colony numbers in the treated samples compared to that of the untreated control was calculated using the following expression:

R (%) = 100 (A - B) / A

where A and B are numbers of bacteria recovered from the untreated and curcumin-treated fabric swatches, respectively, after inoculation and incubation.

Mechanical testing

Testing of fabric mechanical properties was carried out using an Instron tensile tester (Instron 5567) according to ASTM D5035. A gauge length of 7.5 cm and a cross-head speed of 300 mm min⁻¹ were used. An Elmendorf Tear Tester (Elmatear, James H. Heal & Co. Ltd) was used to determine the tear resistance of the samples following ASTM standard D 1424. All reported data were averaged across at least five measurements.

Results and discussion

Sample weights and the calculation of related parameters: % wet pickup and % add-on

From Table 2, it can be seen that the % wet pickup values range from 73.4 to 86.9%. In general, the introduction of high-density TEOS increases the weight of a finishing solution and consequently the weight of liquid carried by the cotton fabric. The % add-on values, which signify the amount of chemical deposited onto the fabric, indicate a weight gain due to the presence of curcumin and/or silica.

Table 2.

Sample percentage wet pickups and percentage add-ons

Sample label	% wet pickup	% add-on
C-30	73.4 ± 4.0	0.2 ± 0.1
T10-30	82.4 ± 1.1	5.1 ± 0.6
T20-30	85.2 ± 2.2	9.1 ± 0.4
T30-30	84.9 ± 3.3	12.8 ± 0.2
C-50	74.8 ± 4.0	0.7 ± 0.2
T10-50	82.9 ± 4.0	5.0 ± 1.4
T20-50	83.1 ± 3.2	9.4 ± 0.2
T30-50	86.9 ± 1.3	13.4 ± 0.7
C-70	78.6 ± 2.7	1.0 ± 0.2
T10-70	83.0 ± 5.9	5.0 ± 0.4
T20-70	88.0 ± 3.1	10.1 ± 0.1
T30-70	85.6 ± 3.4	13.0 ± 0.2

Relationship between precursor concentration and percentage weight increase due to sol–gel coating

The percentage weight increase due to silica coating was calculated by excluding the adsorbed curcumin weight. As illustrated in Figure 2, the percentage weight increase was linearly dependent on the sol-gel precursor concentration. All data gave excellent correlation coefficients of 0.97–0.99, with a positive slope of 0.52. This suggests the possibility of estimating the final weight of the treated fabric if the amount of the precursor in the starting solution is known within the concentration range studied.

Figure 2.

Relationship between percentage weight increase of the treated fabrics and TEOS concentration.

Release profiles of curcumin from fabric samples

Figure 3(a) and (b) display the 24-h cumulative curcumin release profiles of the fabric samples (with three different curcumin loadings) as a function of immersion time and TEOS concentration. The cumulative release values were calculated as percentages based on the apparent surface area of the fabric (20 cm², from width × length). Although scoured cotton fabrics are generally good absorbents, their interactions with curcumin molecules involve weak intermolecular forces (van der Waals, hydrogen bonding), such that traditional curcumin-dyed cotton is prone to poor color fastness during laundering when no other chemical auxiliaries are applied. Consequently, samples without sol–gel treatment contained much lower adsorbed curcumin amounts at the beginning of the release experiment and thus released less curcumin than sol-gel treated samples. Sol–gel coating of cotton at high temperature has been reported to improve color fastness since a porous sol–gel network, or xerogel, is formed which is capable of entrapping dye molecules.²⁰ According to our results, the use of more TEOS precursor results in a larger xerogel structure and therefore significantly higher amount of curcumin available for release. The silica coating on the fabric is assumed to be thicker from the increase in fabric weight with increasing TEOS concentration, as reported in Table 2. The experiment was carried out for 24 h at which point most of the release profiles were shown to level off.

Figure 3.

Release profile of curcumin from cotton fabrics reported as the percentage of the weight of curcumin released divided by the sample surface area. Samples are treated with sol–gels containing: (a) 30 mg, and (b) 70 mg of curcumin, with different TEOS concentrations ranging from 0 to 1.2 M (0–30 mmol/25 mL).

In Figure 4, cumulative curcumin releases, based on the total curcumin content, are plotted against immersion time (for brevity only samples containing 70 mg curcumin are shown). It was found that samples treated with TEOS generally release curcumin at lower percentages than the control samples. This again confirms that the presence of sol–gel porous coating results in enhanced curcumin entrapment compared to the untreated fabric surface, with 70–80%, and 50–60% release being observed for the controls and the sol–gel treated samples, respectively, after 24 h immersion. Curcumin release from samples with low TEOS content increases only marginally after 12 h, while samples with higher levels of TEOS show a more prolonged release profile.

Figure 4.

Release profile of curcumin from cotton fabrics reported as the percentage weight based on the actual amount of curcumin. All samples are treated with sol–gels containing 70 mg curcumin.

Release kinetics

The mechanisms of drug release from polymeric delivery systems can be broadly categorized into three major processes: diffusion-controlled, degradation-controlled, and swelling-controlled.²¹ To date, it has proved impossible to derive a general mathematical model that can describe the controlled release behaviors of all types of delivery systems. In some cases, these mathematical models are derived from theoretical analysis of the occurring process, while in other cases empirical equations have proved to be useful, at least for a comparison of different drug release proﬁles.^22,23 So far, several mathematical models have been reported to be applicable to sol–gel delivery materials, mostly those associated with diffusive release mechanisms. This suggests that different physicochemical phenomena underlie each specific case. These differences include the size and shape of the delivery device, the nature of the release substance and the interaction between the matrix and the release medium. In this work, the four models having found major applications in describing drug release profiles: the zero order, the first-order, the Higuchi, and the Hixson-Crowell models, were employed for the comparison of the release kinetics. The applicability of these models to our release system was also evaluated.

Zero-order model

The model is taken as the ideal delivery process and is represented by the following equation:²²

Q_{t} = Q_{0} + K_{0} t

where Q_t is the amount of drug released in time t (expressed in %), Q₀ is the initial amount of drug in the solution (usually zero), and K₀ is the zero-order release constant. Aughenbaugh et al. reported that the sol–gel delivery profiles may follow this model when the diffusive release is suppressed, such that the main release mechanism is limited to surface degradation.²⁴ This type of release was also observed in sol–gel monoliths containing polysaccharides.²⁵

First-order model

The first-order release model predicts a first-order dependence of drug release on the concentration gradient between the static liquid layer next to the solid surface (delivery device) and the bulk liquid (release medium). According to this model, there is no change in the solid surface area, i.e. no surface disintegration of the delivery device during the release process. The first-order release equation is:^9,22

\ln [1 - (Q_{t} / Q_{\infty})] = - K_{1} t

where Q_∞ is the cumulative amount of the substance released over an infinite time and K₁ is the first-order release constant. As an example, this model renders a good fit for bupivacaine release from silica sol–gel granules and vancomycin from silica sol–gel microspheres.⁹

Higuchi model

The Higuchi square root of time model can be used to describe the release of a solid drug from a porous matrix. Based on this model, the release process involves the penetration of the release medium, the dissolution of the drug, and leaching out of the drug through the interstitial channels or pores. The process is diffusion-controlled and dependent on the porosity and tortuosity of the release device:²²

Q_{t} = [D ɛ / τ (2 A - ɛ C_{s}) C_{s} t] 0.5

in which D is the diffusion coefficient of the drug, ɛ is the porosity of the matrix, τ is the tortuosity factor of the capillary system, and A is the total amount of the drug present in the matrix. The above equation can be simplified to:

Q_{t} = K_{2} t 0.5

where K₂ is the release rate constant. This model was found to be applicable to many sol–gel delivery systems, including sol–gel silica microspheres and monoliths.^9,25,26

Hixson–Crowell model

The Hixson–Crowell cube root model describes the drug release from systems where the geometrical shape of the release matrix diminishes proportionally over time (erodible matrix). The release rate is limited by the matrix dissolution rate, not by the diffusion of drug through the polymeric matrix as shown in the following expression:²²

Q_{0}^{1 / 3} - Q_{t}^{1 / 3} = K_{3} t

where K₃ is the rate constant. This model has been tested with sol–gel nanoparticle drug delivery systems but was found to be inadequate.²⁷

How well a particular model can describe the experimental data is reflected in the correlation coefficient value (R²), which should be close to 1. However, it should be noted that, in practice, there is difficulty in discriminating between competing models.^28,29

The curcumin release data were plotted according to the aforementioned models for the first 10 hours of release. According to the literature, at levels beyond 50% release significant deviations from the models may occur.⁹ The rate constants and corresponding correlation coefficients of the fabric samples are listed in Table 3. Interestingly, all four models can describe the release profiles of both sol-gel treated, and untreated samples at R²> 0.90 but overall, the quality of fit is slightly better for the Higuchi model and the first-order model (R²> 0.95). The rate constants for every model are consistent: the samples without sol–gel treatment release curcumin faster than treated samples. Overall, sol–gel coating retards the release rate by about 30%. Radin et al. have demonstrated that silica sol–gel microspheres provide drug (bupivaciane and vancomycin) release following both the Higuchi and the first-order model.⁹ The applicability of more than one model to a certain release system is not surprising. It is worth noting that these two models share key assumptions: the release process is purely-diffusion controlled with no matrix swelling or erosion and perfect sink conditions are maintained.³⁰ These two expressions also imply decreased drug release with time, which can be explained based on the increase in diffusion path length as the degree of medium penetration, and drug extraction, proceed. Matrix swelling during immersion can be ruled out as the inorganic silica xerogel is typically rigid.

Table 3.

Release rate coefficients and their corresponding R² values derived from four different kinetic models

	Zero-order		First-order		Higuchi		Hixson–Crowell
Sample	K₀ (% h⁻¹)	R ²	K₁ (h⁻¹)	R ²	K₂ (% h^−0.5)	R ²	K₃ (% h⁻¹)	R ²
C-30	4.53	0.96	0.09	0.98	19.38	0.98	0.11	0.98
10-30	2.95	0.96	0.04	0.98	12.69	1.00	0.06	0.97
T20-30	2.78	0.97	0.04	0.98	12.23	0.95	0.06	0.9
T30-30	2.95	0.95	0.04	0.95	12.50	0.95	0.06	0.95
C-50	4.83	0.99	0.08	0.98	21.76	0.96	0.11	0.99
T10-50	3.67	0.99	0.06	1.00	16.86	1.00	0.08	1.00
T20-50	4.25	1.00	0.07	0.99	17.76	0.97	0.09	0.99
T30-50	3.36	0.97	0.05	0.98	14.27	0.98	0.07	0.98
C-70	4.82	0.93	0.08	0.96	18.92	0.97	0.10	0.94
T10-70	3.22	0.95	0.06	0.97	13.68	0.96	0.07	0.96
T20-70	3.03	0.97	0.05	0.97	12.72	0.95	0.06	0.97
T30-70	2.89	0.93	0.04	0.97	12.44	0.96	0.06	0.93

To verify that there is no dissolution of silica matrix in our experimental regime, the amount of silicon (in the form of silicic acid, Si(OH)₄) leached out of the sol-gel treated samples after 24 h immersion was determined using the heteropoly blue method.³¹ The release medium was digested with NaHCO₃ to convert unreactive silicon compounds, if present, into the molybdate-reactive form suitable for determination. Results show that no silica matrix dissolution occurred during 24 h, confirming that the curcumin release kinetics of the sol–gel coated fabrics are mainly diffusion-controlled as predicted by the first-order and Higuchi models. It is known that silica can undergo hydrolysis to form silicic acid and the hydrolysis rate would be greatly increased under alkaline conditions. For nonporous amorphous silica, the equilibrium concentration of silicic acid is only 70 ppm at 25℃, and in this case silica solubility is therefore negligible.³²

The comparable rate constants for most sol–gel coated samples can be explained by their identical chemical compositions and the homogeniety of the curcumin-embedded sol–gel coating. The coating thickness variation, due to different amounts of precursor, did not appreciably change either the release mechanism or the release kinetics. Although release is expected to be faster with greater curcumin loading, these effects were not noticeable under our experimental conditions (which was restricted due to the solubility limit of curcumin).

Specific surface area and porosity

The N₂ gas adsorption isotherms of untreated cotton and T30-30 are displayed in Figure 5(a). Based on the International Union of Pure and Applied Chemistry (IUPAC) classification,³³ the adsorption isotherm of cotton is of type II, being characterized by an intermediate plateau followed by a continuous rise at a relative pressure close to unity (unrestricted monolayer–multilayer adsorption). This type of isotherm is generally observed in non-porous, or macroporous materials. Cotton textiles contain numerous pores including intra-fiber, inter-fiber, intra-yarn and inter-yarn pores. Their pore sizes depend on fiber size, type of weave, the number of yarns per inch and the presence of yarn twists. The inter-fiber and inter-yarn pore sizes lie in a range consistent with those of macropores, which are considered too large (> 50 nm) to be reliably measured using the nitrogen gas adsorption technique. It has been reported that the inter-fiber pore sizes in woven cotton are between 2.0–4.5 µm,³⁴ while those of inter-yarns are certainly larger. In this study, the BJH method employed is suitable for the calculation of sizes (2–50 nm) spanning the mesopore to small macropore range only. Since the aim is to observe the change in porosity caused by the silica xerogel coating, the measurement of larger macropores (which is possible by mercury intrusion porosimetry) associated with the textile structure was not carried out. The T30-30 isotherm appears similar to that of cotton but with a small hysteresis loop suggesting the presence of mesopores which originate from the xerogel structure. The total pore volume was estimated from the amount of nitrogen adsorbed at a relative pressure of 0.99; the values obtained are very low, i.e. 0.003 cm³ g⁻¹ for the untreated sample and 0.004 cm³ g⁻¹ for T30-30, respectively. A slightly higher pore volume in T30-30 was attributed to the open mesopores present in the xerogel structure. Based on the BJH calculation from the desorption isotherm, the mean diameter of these mesopores is 14.3 nm. The BJH pore size distribution of T30-30 is shown as can be seen in Figure 5(b).

Figure 5.

Results of nitrogen sorption experiments: (a) adsorption–desorption isotherms of the untreated cotton and T30-30 samples; (b) BJH pore size distributions derived from the desorption isotherm of the T30-30 sample.

Regarding the determination of sample specific surface area, the BET method was applied to the adsorption isotherm at the relative pressure between 0.05–0.30. The BET specific surface area of the untreated cotton was 1.95 m² g⁻¹. For the sol–gel coated cotton, the specific area increased slightly to 2.23 m² g⁻¹. Both results were obtained with excellent correlation coefficients greater than 0.99. Similar values have been reported in another sol–gel coated cotton system.¹⁷

The higher pore volume and surface area in the treated samples are a consequence of the porous sol–gel structure. However, the difference between treated and native cotton was only small, suggesting that the extent of fabric sol–gel coating was not high. This is preferable when one wants to preserve the overall characteristics of the original textile in the fabricated material.

Because of the greater porosity creating a more tortuous pathway for any fluid to diffuse in or out of the xerogel structure, after immersing in alcoholic solution coated samples retain a greater number of curcumin molecules than the control (Figure 6). This is highlighted by the yellow color being stronger in fabrics having higher TEOS, and curcumin concentrations. This can be expected since curcumin is adsorbed only weakly onto cotton fibers. A large proportion of the curcumin molecules leach out of the uncoated cotton fabric soon after immersion, leading to poor color fastness to washing. This process can be slowed down by entrapping them within the sol–gel network. When the coated fabrics are used in wound dressing applications, the effective residence period of the biomedically active compound can be extended. The amount of the active substance as well as the release duration can be tailored easily by adjusting the sol–gel processing parameters (the concentration of the sol-gel precursor, for instance).

Figure 6.

Photographic images of samples treated with various compositions of TEOS and curcumin.

Sample morphology

SEM micrographs were taken from samples containing 0.07 g curcumin for a comparison of their morphological properties. As displayed in Figure 7, the C-70 sample shows fibers having a smooth surface, with twists and convolutions characteristic of cotton. Homogenous curcumin adsorption on the cotton fiber was obtained, with no indication of curcumin agglomeration. In our study, the sol–gel coatings do not form bridges between adjacent fibers, which is desirable as air and vapor permability is required for their potential application as wound dressings. This indicates that the coatings are thin enough such that the open structure of the fabrics is not closed up on fabrication of the material.

Figure 7.

SEM micrographs of the treated cotton fabrics: (a) C-70, (b) T10-70, (c) T20-70, (d) T30-70.

Antibacterial activity

Curcumin is an effective colorant for cotton and can also provide potential antibacterial activity to fabrics. The sol–gel coatings, however, are biologically inert and thus exhibit no antibacterial properties.³⁵ As shown in Table 4, the sol–gel coating results in higher antibacterial activity in the treated fabrics over the control. The fabrics without sol–gel treatment revealed very low antibacterial activities against E. coli and no activity against S. aureus, indicative of a very low concentration of curcumin being present. However, on sol–gel treatment, the fabrics showed improved antibacterial effects against both strains. The percentage reduction in the bacteria counts for S. aureus increased rapidly with increasing curcumin concentration, from 0 to 99% with the addition of 30–70 mg of curcumin in samples treated with 30 mmol TEOS. On the other hand, values against E. coli increased only slightly from 28 to 37% for the same sample set. Similar observations were found with the curcumin-dyed cotton: curcumin was more effective in inhibiting S. aureus than E. coli.³⁶ This is likely because the Gram-negative bacteria have an outer hydrophilic surface functioning as a preventive barrier against hydrophobic compounds. They also have enzymes capable of breaking down molecules introduced from the outside.^37,38 The Gram-positive bacteria do not possess such an outer membrane and cell wall structure.

Table 4.

Percentage reduction in bacterial levels as determined according to AATCC 100-1999

	Bacteria reduction percentage (%)
Sample	Staphylococcus aureus	Escherichia coli
C-30	0	6
C-50	0	17
C-70	0	1
T30-30	0	28
T30-50	38	31
T30-70	99	37

Mechanical properties

Light weight and high flexibility are not the only important features for functioning textile materials; having mechanical properties suitable for the targeted application is also required. In this research, the tensile strength, percentage elongation at break, and tearing force for the treated fabrics were investigated. As seen in Figure 8(a), the tensile strength of the fabrics was not altered to a significant extent by sol–gel coating. According to Figure 8(b), a decrease of percentage elongation at break was observed with increasing TEOS concentration in both warp and weft directions, with the effect more significant in the former as a result of increased stiffness imparted by the inorganic silica coating. The slight increase in tensile strength, and a smaller elongation at break were also reported for other sol–gel coated cotton fabrics.³⁸

Figure 8.

Mechanical properties of the fabric samples: (a) tensile strength; (b) tensile elongation at break; (c) tearing strength (the error bars represent the standard deviation with n = 5).

As expected, the sol–gel coating of fabrics resulted in lower tear resistance (Figure 8(c)). This can be explained partly by a reduction in yarn slippage. The silica network present on the fiber surface increases the fiber surface friction and, as a consequence, prevents yarn slippage. Since the tearing energy cannot be dissipated efficiently with poor yarn slippage, the tear propagation is then more pronounced. This results in the lowered tearing strength typically observed in sol–gel coated fabrics.^38–40 Our results reveal that the reduction in tear resistance is only slight; the sample with the lowest tearing strength (T30-30) shows about 15% reduction in the warp direction. From the SEM images, there is no interlocking between the cotton fibers arising from the silica network, therefore the fibers can still be displaced to a reasonable degree.

Conclusions

This work investigated the coating of cotton fabrics with an acid-catalyzed sol-gel and curcumin for potential application in controlled release wound dressings. Prolonged and more efficiently controlled curcumin release, as well as antibacterial properties can be realized via sol–gel coating of fabrics. A study of the release kinetics revealed that the release process is diffusion-controlled, with gas sorption analysis confirming the ability of the sol–gel coating to retain curcumin molecules. The coated fabrics exhibit reduced ductility but their strength is mostly preserved. Overall, the application of sol–gel coating with the release of nontoxic herbal substances such as curcumin may result in the eventual application of these textiles as next generation, controlled-release wound dressings.

Footnotes

Funding

This work was supported by the Thailand Research Fund (TRF); The Office of the Higher Education Commission (OHEC); and Thammasat University (grant number MRG5580207).

References

Alongi

Ciobanu

Tata

. Thermal stability and flame retardancy of polyester, cotton, and relative blend textile fabrics subjected to sol–gel treatments. J Appl Polym Sci 2011; 119: 1961–1969.

Mahltig

Fischer

. Inorganic/organic polymer coatings for textiles to realize water repellent and antimicrobial properties: a study with respect to textile comfort. J Polym Sci Part B Polym Phys 2010; 48: 1562–1568.

Textor

Mahltig

. A sol–gel based surface treatment for preparation of water repellent antistatic textiles. Appl Surf Sci 2010; 256: 1668–1674.

Bahners

Textor

Opwis

. Recent approaches to highly hydrophobic textile surfaces. J Adhes Sci Technol 2008; 22: 285–309.

Tian JC. Surface modification method for slow-release type anti-insect and anti-microbe fabric. Patent CN1807752, China, 2006.

Mahltig

Fiedler

Böttcher

. Antimicrobial sol–gel coatings. J Sol Gel Sci Technol 2004; 32: 219–222.

Xing

Ding

. Immobilization of papain on cotton fabric by sol–gel method. Enzyme Microb Technol 2007; 40: 1692–1697.

Böttcher

Slowik

Süß

. Sol–gel carrier systems for controlled drug delivery. J Sol Gel Sci Technol 1998; 13: 277–281.

Radin

Chen

Ducheyne

. The controlled release of drugs from emulsified, sol–gel processed silica microspheres. Biomaterials 2009; 30: 850–858.

10.

Rajendran

Anand

. Development in medical textiles. Text Prog 2002; 32: 1–42.

11.

Van de Schueren

De Clerck

. Coloration and application of pH-sensitive dyes on textile materials. Color Technol 2012; 128: 82–90.

12.

Mahltig

Fiedler

Fischer

. Antimicrobial coatings on textiles: modification of sol–gel layers with organic and inorganic biocides. J Sol Gel Sci Technol 2010; 55: 269–277.

13.

Liu

Lin

Peng

. Antibacterial activity of capsaicin-coated wool fabric. Text Res J 2012; 82: 584–590.

14.

Han

Yang

. Antimicrobial activity of wool fabric treated with curcumin. Dyes Pigments 2005; 64: 157–161.

15.

Suwantong

Opanasopit

Ruktanonchai

. Electrospun cellulose acetate ﬁber mats containing curcumin and release characteristic of the herbal substance. Polymer 2007; 48: 7546–7557.

16.

Schindler

Hauser

. Chemical finishing of textiles, Cambridge: Woodhead Publishing Ltd, 2004, pp. 9–9.

17.

Xing

Ding

. Silica xerogel coating on the surface of natural and synthetic fabrics. Surf Coat Technol 2008; 202: 4721–4727.

18.

Abdelmouleh

Boufi

Ben Salah

. Interaction of silane coupling agents with cellulose. Langmuir 2002; 18: 3203–3208.

19.

Brunauer

Emmett

Teller

. Adsorption of gases in multimolecular layers. J Am Chem Soc 1938; 60: 309–319.

20.

Barrett

Joyner

Halenda

. The determination of pore volume and area distributions in porous substances. I. Computations from nitrogen isotherms. J Am Chem Soc 1951; 73: 373–380.

21.

Tan

JPK

Tam

Responsive microgels for drug delivery applications. In: Hidalgo-Alvarez

(ed). Structure and functional properties of colloidal systems, Boca Raton, FL: CRC Press, 2010, pp. 387–409.

22.

Costa

Manuel

Sousa Lobo

. Modeling and comparison of dissolution profiles. Eur J Pharm Sci 2001; 13: 123–133.

23.

Mishra

Tamboli

Mitra

. Effect of hydrophobic and hydrophilic additives on sol–gel transition and release behavior of timolol maleate from polycaprolactone-based. Colloid Polym Sci 2011; 289: 1553–1562.

24.

Aughenbaugh

Radin

Ducheyne

. Silica sol–gel for the controlled release of antibiotics. II. The effect of synthesis parameters on the in vitro release kinetics of vancomycin. J Biomed Mater Res 2001; 57: 321–326.

25.

Roveri

Morpurgo

Palazzo

. Silica xerogels as a delivery system for the controlled release of different molecular weight heparins. Anal Bioanal Chem 2005; 391: 601–606.

26.

Morpurgo

Teoli

Palazzo

. Influence of synthesis and processing conditions on the release behavior and stability of sol–gel derived silica xerogels embedded with bioactive compounds. Farmaco 2005; 60: 675–683.

27.

López Goerne

López García

Rodríguez Grada

. Obtaining of sol–gel ketorolac–silica nanoparticles: characterization and drug release kinetics. J Nanomater 2013, pp. 450–483.

28.

Wong

Gilligan

. Characterisation of commercially available theophylline sustained-or controlled-release systems: in-vitro drug release profiles. Int J Pharm 1990; 66: 111–130.

29.

Kockish

Rees

Tsibouklis

. Mucoadhesive, triclosan-loaded polymer microspheres for application to the oral cavity: preparation and controlled release characteristics. Eur J Pharm Biopharm 2005; 59: 207–216.

30.

Siepmann

. Mathematical modeling of drug delivery. Int J Pharm 2008; 364: 328–343.

31.

Boltz

Mellon

. Determination of phosphorus, germanium, silicon, and arsenic. Anal Chem 1947; 19: 873–877.

32.

Finnie

Waller

Perret

. Biodegradability of sol–gel silica microparticles for drug delivery. J Sol Gel Sci Technol 2009; 49: 12–18.

33.

Sing

KSW

Everett

Haul

RAW

. Reporting physisorption data for gas/solid systems with special reference to the determination of surface area and porosity. Pure Appl Chem 1985; 57: 603–619.

34.

Burleigh

EG Jr

Wakeham

Honold

. Pore-size distribution in textiles. Text Res J 1949; 19: 547–555.

35.

Mahlting

Audenaert

Böttcher

. Hydrophobic silica sol coating on textile: the influence of solvent and sol concentration. J Sol Gel Sci Technol 2005; 34: 103–109.

36.

Reddy

Han

Zhao

. Antimicrobial activity of cotton fabrics treated with curcumin. J Appl Polym Sci 2013; 127: 2698–2702.

37.

Gao

van Belkum

Stiles

. The outer membrane of Gram-negative bacteria inhibits antibacterial activity of brochocin-C. Appl Environ Microbiol 1999; 65: 4329–4333.

38.

Duffy

Power

. Antioxidant and antimicrobial properties of some Chinese plant extracts. Int J Antimicrob Agents 2001; 17: 527–529.

39.

Cai

. Effect of acid-catalyzed sol–gel silica coating on the properties of cotton fabric. J Text Inst 2012; 103: 1099–1107.

40.

Mahltig

Textor

. Nanosols and textiles, Singapore: World Scientific Publishing Co, 2008, pp. 41–41.