Electrospun bi-layer carboxymethyl cellulose/polyvinyl alcohol/polycaprolactone nanofibers: fabrication, degradation, and phenytoin release kinetics

Abstract

Carboxymethyl cellulose (CMC), polyvinyl alcohol (PVA), and polycaprolactone (PCL) were used to fabricate CMC/PVA/GA nanofibers (PNCMC) and polycaprolactone and carboxymethyl cellulose nanofibers (PNPCL-CMC) nanofibers via electrospinning. PVA solutions at concentrations of 5, 6, 10, and 15% , and CMC solutions at concentration of 1 and 2% were mixed with different volume ratios of PVA: CMC (1:1, 4:6, 3:7, 6:4, and 7:3). The 10% PVA mixed with 1% CMC at a 6:4 ratio yielded the best electrospinning performance. The addition of glutaraldehyde (GA) to the solutions improved fiber uniformity by promoting cross-linking. A bi-layer PNPCL-CMC nano-composite was subsequently developed by adding the PNCMC onto electrospun polycaprolactone nanofibers (PNPCL). The active pharmaceutical ingredient of phenytoin sodium (PHT) was incorporated into the polymer solution and electrospun to prepare drug-loaded nanofibers. The average fiber diameters ranged from 100 to 500 nm. The degradation and swelling ratio of each layer were studied independently. After 30 days, 93.6% of PNCMC and 33.7% of PNPCL were degraded in phosphate buffered saline (pH = 7). The maximum swelling ratio for PNCMC (311.9%) was observed after 7 days, whereas the maximum swelling ratio for PNPCL (105%) was reported after 5 days. According to contact angel results, the bi-layer composite of superhydrophilic CMC and hydrophobic PCL nano-fibers offering high potential for wound dressing. Phenytoin release kinetics were evaluated using seven mathematical models including zero-order, first-order, Higuchi, Hixson–Crowell, Korsmeyer–Peppas, and Baker–Lonsdale and the results were analyzed by comparing R² and Akaike information criterion (AIC). The Baker–Lonsdale model provided the best fit(R² = 0.9822 and AIC=-34.8275), for phenytoin release from PNPCL-CMC. Therefore, it can be concluded the bi-layer PNPCL-CMC nanofibrous composite demonstrates controlled phenytoin release and exhibits favorable degradation, swelling and wettability properties, indicating its strong potential for wound dressing applications.

Introduction

Wound healing is a complex physiological process crucial for restoring skin integrity after injury, particularly in chronic wounds where persistent inflammation and impaired angiogenesis delay recovery and increase the risk of infection and amputation^1,2. Consequently, the development of advanced wound dressings is of critical importance³. Hybrid nanofibers and nano-composites composed of natural and synthetic polymers, such as polyvinyl alcohol (PVA)/gelatin and polycaprolactone (PCL)/carboxymethyl cellulose (CMC), have been extensively investigated for wound healing owing to their complementary physicochemical and biological properties^4,5.

Polycaprolactone nanofibers exhibit excellent mechanical strength and biocompatibility, rendering them promising candidates for tissue regeneration applications^6,7. Moreover, PCL can be readily combined with other polymers such as CMC and PVA, further expanding its applicability in drug delivery, tissue engineering scaffolds, and various other biomedical fields⁸. CMC enhances hydrophilicity and bioactivity, thereby providing additional benefits for infection control and wound healing applications⁹.

CMC is odorless, tasteless, and physiologically inert^10,11. CMC nano-composites and nanofibers exhibit a wide range of applications, including wound dressings¹², the food industry¹³, environmental remediation¹⁴, energy storage¹⁵, textiles, and tissue engineering^16,17, owing to their remarkable features such as high specific surface area, high porosity, large surface-to-volume ratio, significant elastic modulus, high chemical reactivity, enhanced hydrophilicity, and inherent biocompatibility and biodegradability¹⁸. Despite these advantageous properties, pure CMC cannot be electrospun due to its rigid molecular structure, which restricts the chain entanglement necessary for fiber formation¹⁹. Consequently, synthetic polymers such as PVA and PCL are blended with CMC to facilitate nanofiber formation²⁰.

Notably, the fabricated CMC/PVA nanofibers have been designated for applications in drug delivery and wound dressing. To increase their healing efficacy, various pharmaceutical agents, including phenytoin (PHT), have been incorporated into these nanofibers^21,22. Phenytoin was first introduced in 1937 for the effective management of convulsive disorders and subsequently applied to promote the healing of various types of wounds²³. Clinical research conducted in 1958 demonstrated that phenytoin significantly enhances gingival wound healing compared to control treatments²⁴. Between 1990 and 2000, phenytoin was successfully loaded onto nanofibers for therapeutic applications. The results showed that phenytoin led to accelerated healing, reduced pain and wound discharge, and improved stimulated granulation tissue formation than standard treatment with hydrogen peroxide and antiseptics²⁵.Phenytoin-loaded PLGA/lecithin nanoparticles were incorporated into chitosan-based nanofiber mats for wound healing applications²⁶. Subsequently, thermodynamic analysis was performed to evaluate the loading capacity of phenytoin in poly(d, l-lactic acid) (PLA) nanocarriers prepared via the solvent displacement method²⁷. Figure 1 presents a summary of studies investigating the use of phenytoin for wound healing, which is described in detail subsequently.

Fig. 1.

A summary of studies on the use of phenytoin for wound dressing applications^23–37.

Previous studies on phenytoin-loaded polymeric delivery systems have demonstrated that electrospun nanofibers and hybrid nanocomposites are effective in achieving controlled drug release while significantly improving wound healing outcomes. These systems promote essential regenerative processes such as cell adhesion and proliferation, re-epithelialization, collagen deposition, and tissue regeneration, while reducing inflammation and tissue necrosis^28–31.

Among the investigated phenytoin delivery platforms, chitosan-based hydrogels and copper nanoparticles have been shown to effectively provide controlled drug release and promote wound healing^32,33. Moreover, electrospun nanofibrous systems, including core-shell and multilayered structures based on PCL and collagen, have exhibited favorable physicochemical properties and enabled sustained phenytoin release, thereby facilitating tissue regeneration through enhanced epidermal proliferation and neovascularization^34–37.

CMC has been widely explored due to its biocompatibility, eco-friendly nature, and biomimetic properties³⁸. Nevertheless, its practical application is limited by challenges such as poor spinnability and inadequate mechanical stability. The incorporation of cellulose-based components has been shown to enhance fiber uniformity, crystallinity, surface wettability, and thermal stability, while appropriate crosslinking improves mechanical integrity and water resistance without substantially reducing absorption capacity^39,40. Furthermore, optimized CMC-based nanofibrous scaffolds have exhibited favorable degradation behavior and excellent cytocompatibility, supporting cell adhesion, proliferation, and migration. These findings highlight the potential of CMC nanofibers as multifunctional platforms for phenytoin loading, emphasizing the importance of formulation and structural optimization⁴¹.

To the best of the authors’ knowledge, loading of phenytoin onto CMC/PVA/PCL nanofibers for topical wound treatment has not been previously documented in detail. The main goal of this study was the fabrication of CMC nanofibers with an optimized structural configuration. PVA was incorporated as a co-polymer, as CMC alone is incapable of electrospinning. In addition, PCL was employed to further enhance the physicochemical properties of nanofibers. The resulting nanofibers were subsequently loaded with phenytoin.

Experimental work

Chemicals

PVA, PCL, and glutaraldehyde (GA) were prepared from Sigma-aldrich Reagent Co. Ltd. (Steinheim, Germany); whereas CMC and formic acid were sourced from Merck (Darmstadt, Germany). Glacial acetic acid was purchased from Ghatran Shimi Co (Ghazvin, Iran), and phosphate buffered saline (PBS) tablets were purchased from Tamadkala company (Tehran, Iran). The active pharmaceutical ingredient of phenytoin sodium was provided by Loghman Pharmaceutical & Hygienic Company (Tehran, Iran). Double distilled water was used throughout the experiments. All chemicals and reagents used were of analytical grade and used without further purification.

Characterization

Optical observation of the prepared samples were performed using a Nikon eclipse E100 optical microscope (Nikon Corporation, Tokyo, Japan). The morphologies of the synthesized samples were examined via scanning electron microscopy (SEM) model Seron ALS-2100 (Seron Technologies Inc, Seoul, Korea). Attenuated total reflection (ATR) spectra were recorded by a Bruker, tensor 27 spectrometer (Bruker Optik GmbH, Ettlingen, Germany) to analyze organic and inorganic compounds. Contact angle measurements were conducted using a Jikan-CAG 20 SE instrument (Jikan Surface Nano-Engineering Co., Tehran, Iran). The electrospinning machine model Full Option Lab2 ESI-II (Fanavaran Equipment Nanoazma Co., Tehran, Iran) was used in this study.

Synthesis of PVA nanofibers (PNPVA)

1, 1.2, 2, and 3 g of PVA (Mw = 85000 g/mol & Mw = 124000 g/mol) was separately dissolved in 20 mL of double-distilled water and stirred for 4.5 h at 70 to prepare solutions with concentrations of 5, 6, 10, 15% , respectively. To minimize the solvent evaporation during heating, the container was covered with a watch glass. 2 mL of the acquired solutions was subsequently electrospun under a controlled flow rate of 1 mL/h, voltage of 12 kV, tip to collector distance of 16 cm, and drum rotation speed of 100 rpm. Scheme 1. interprets the preparation steps of the PVA solutions and electrospinning process.

Scheme 1.

Schematic illustration of the preparation of PVA solution and electrospinning process.

The electrospinning results of PVA solutions were summarized in Table 1. Optical microscopy images indicated that the PVA (Mw=124000 g/mol) solution at a concentration of 10% produced the most uniform fibers, designated as the PNPVA sample.

Table 1.

List of prepared PVA fibers.

Sample	PVA concentration* %	Mw (g/mol)	Result of optical microscopy
1	5	85,000	Bad spraying and discontinuous fibers
2	6	85,000	Low fiber formation
3	10	85,000	Bad spraying and discontinuous fibers
4	15	85,000	No spraying due to high viscosity
5	5	124,000	No spraying
6	6	124,000	Fiber formation by discontinuity
7	10	124,000	Fiber formation with good apperance (PNPVA sample)
8	15	124,000	No spraying due to high viscosity

^*The volumes of all electrospinning solutions are 2 mL.

Synthesis of CMC/PVA/GA nanofibers (PNCMC)

Electrospinning of CMC is challenging due to its high viscosity and poor electrical conductivity. Therefore, PVA was employed as a crosslinker to facilitate the electrospinning CMC fibers. Briefly, 0.2 g of CMC was dissolved in 20 mL of double-distilled water at 80 for 5 h to prepare a 1% solution. To minimize the solvent evaporation during heating, the container was covered with a watch glass. Subsequently, CMC (1% ) and PVA (10% , Mw = 124000) solutions were mixed together with different volume ratios of PVA: CMC (1:1, 4:6, 3:7, 6:4, and 7:3) and subjected to electrospinning.

2 mL of the resulting mixtures was subsequently used for the electrospinning process under the following conditions: voltage of 15 kV, flow rate of 2 mL/h, distance between the needle tip and the collector of 18 cm, and drum rotation speed of 100 rpm, and this process continued for 1 h. It was observed that the addition of a specific amount of GA to the CMC/PVA solution adjusted the solution viscosity and improved fiber continuity and uniformity. Consequently, GA was added to the polymer solution, and the electrospinning process was repeated. Finally, the resulting nanofibers were collected on aluminum foil (Scheme 2).

Scheme 2.

Schematic illustration of the preparation of CMC/PVA/GA solution and electrospinning process.

The prepared CMC solutions and their electrospinning results were summarized in Table 2. Optical microscopy images indicated that solutions with PVA: CMC mixing ratios of 6:4 and 7:3, containing GA, yielded the most uniform fiber morphology (Samples 12 & 13). The optimized electrospun CMC/PVA/GA sample was designated as PNCMC.

Table 2.

List of CMC fibers.

Sample	CMC solution* %	PVA: CMC ratio**	GA*** (mL)	Result of optical microscopy
1	1	-	-	No fiber formation
2	2	-	-	No fiber formation
3	1	-	0.3	No fiber formation
4	2	-	0.3	No fiber formation
5	2	1:1	-	No fiber formation
6	1	1:1	-	No fiber formation
7	1	6:4	-	Bad spraying and formation of non-uniform fibers
8	1	7:3	-	Bad spraying and formation of non-uniform fibers
9	1	4:6	-	No fiber formation
10	1	3:7	-	No fiber formation
11	1	1:1	0.3	Formation of non-uniform fibers
12	1	6:4	0.3	Fiber formation with good apperance (PNCMC sample)
13	1	7:3	0.3	Fiber formation with good apperance
14	1	4:6	0.3	No fiber formation
15	1	3:7	0.3	No fiber formation

^*The volumes of all electrospinning solutions are 2 mL.

^**Polyvinyl alcohol (Mw = 124000 g/mol) solution with concentration of 10% was used in all entries.

^***Density of glutaraldehyde is 1.06 g/cm³. The amount of GA added are reported per 20 mL of the CMC/PVA solution.

Synthesis of polycaprolactone nanofibers (PNPCL)

PNPCL were fabricated via electrospinning, as illustrated in Scheme 3, to provide mechanical reinforcement for wound healing applications. A 20% PCL solution was prepared by dissolving 4 g of PCL (Mw = 80000 g/mol) in 20 mL of a mixture of glacial acetic acid and formic acid solvents (volume ratio of acetic acid to formic acid equal to 1:9) under continuous stirring at room temperature for 1 h. 2 mL of the obtained polymer solution was loaded into a 5 mL syringe and electrospun for 1 h under the following conditions: voltage = 11 kV, tip to collector distance = 23 cm, injection flow rate = 0.4 mL/h, and drum rotation speed = 100 rpm.

Scheme 3.

Schematic illustration of the preparation of PCL solution and electrospinning.

Synthesis of bi-layer composite of polycaprolactone and carboxymethyl cellulose nanofibers (PNPCL-CMC)

To prepare the bi-layer PNPCL-CMC nano-composite, polymer solutions of CMC and PCL were first prepared and electrospun following the methods described above. In the first layer, PCL fibers were elctrospun onto aluminum foil under the conditions outlined in the section “Synthesis of polycaprolactone nanofibers (PNPCL)”. Subsequently, CMC fibers were electrospun onto the PCL layer under the conditions specified in the section “Synthesis of CMC/PVA/GA nanofibers (PNCMC)”. The resulting bi-layer composite of PCL and CMC nanofibers was designated as PNPCL-CMC sample.

Degradability and swelling ratio tests of nanofibers

The thin film of polymeric nanofibers was cut into 5 pieces and weighed (. These pieces were then immersed in phosphate buffered saline (PBS, pH = 7) at room temperature for 3, 5, 7, 14, and 30 days (Scheme 4). At each time point, the samples were removed, blotted with filter paper, and weighed (. Subsequently, all samples were dried at room temperature for 24 h and reweighed (. The degradation percentage and swelling ratio of the samples were calculated using the following equations:

1

2

Scheme 4.

Schematic diagram of degradability & swelling ratio tests of nanofibers.

Calibration curve

The concentrations of unknown phenytoin sodium solutions were determined using a spectrophotometric calibration curve. The measured transmittance intensities were analyzed to construct the calibration curve. To construct the calibration curve, seven phenytoin sodium solutions in double distilled water were prepared at concentrations ranging from 2 to 10 µg/mL, and the absorbance of each solution was detected and recorded using UV–Vis spectrophotometer. The maximum absorption wavelength (λ_max) of phenytoin sodium was observed at 200 nm. The tests were examined triple times, and the mean values were calculated from the UV-Vis data. The Beer-Lambert law provides a quantitative method for determining the concentration of an analyte based on its absorbance. A is defined as a function of the input light intensity (I₀) and the passing light intensity (I) through the spectrophotometer cell at λ_max, as expressed in Eqs. 3 & 4:

3

4

Where, L is the width of the cell (the path length of the light), c is the phenytoin (analyte) concentration, and ε is the molar absorptivity.

Loading of phenytoin sodium on PNPCL-CMC sample

The amount of 2 g of PVA was completely dissolved in 20 mL of double-distilled water at 70 for 150 min to form a homogeneous solution. Separately, 0.2 g of CMC and 0.3 mL of GA were added to 20 mL of double distilled water at 80 for 3 h to obtain a clear CMC solution. Then, 12 mL of PVA solution and 8 mL of CMC solution were mixed and stirred for 5 h to yield a fully mixed CMC/PVA/GA solution. Subsequently, a precisely weighed amount of phenytoin sodium was added to the 2 mL of CMC/PVA/GA solution and stirred at room temperature for 30 min to obtain a homogeneous solution for electrospinning. In vitro drug loading tests were conducted using 0.1 mg of phenytoin. The resulting phenytoin-containing polymer solution was then electrospun under the following conditions: voltage of 17 kV, flow rate of 1 mL/h, tip-to-collector distance of 16 cm, and drum rotation speed of 100 rpm. The resulting nanofibers were collected on aluminum foil (Scheme 5).

Scheme 5.

Schematic description of phenytoin loading during the preparation of PNPCL-CMC nanofibers.

In vitro release study of phenytoin sodium from PNPCL-CMC nanofibers

The amount of 0.1 mg of phenytoin sodium was dissolved in 2 mL of CMC/PVA/GA solution to obtain a 50 precursor solution. This solution was subjected to electrospinning to fabricate phenytoin-loaded nanofibers. Nanofibrous samples (2) were subsequently immersed in 20 mL of phosphate buffered saline solution (pH = 7) under continuous stirring. The in vitro release profile of phenytoin sodium from PNPCL-CMC nanofibers was examined at predetermined time intervals of 15, 30, 60, and 120 min, followed by 4, 8, 16, 24, and 48 h. At each interval, the supernatant solution was centrifuged, and analyzed for absorbance intensity using UV-Vis spectrophotometry. The released concentration of phenytoin sodium was determined from the calibration curve, and the release data were further fitted to appropriate kinetic models.

Kinetic models

Several mechanisms including diffusion, dissolution, and erosion, play a key role in drug delivery and are discussed below. In diffusion-controlled release, drug molecules migrate from the polymer matrix into the surrounding environment along the concentration gradient. Whereas, in dissolution-controlled systems, drug release is governed by dissolving the polymer matrix, making polymer solubility a key determinant of the release rate. Moreover, the study of erosion mechanisms has gained considerable attention with the development of biodegradable polymers for controlled drug release. Two types of erosion can be occurred, bulk erosion and surface erosion. In bulk erosion, the entire polymeric fibers are degraded by diffusion of water molecules throughout the polymer matrix, leading to the burst release; while in surface erosion, the polymer chains on the surface are broken down, yielding a more controlled release profile. In most drug delivery systems, the release behavior is governed by a combination of diffusion, dissolution, and erosion mechanisms. The kinetic analysis is commonly applied to predict the time required for effective drug release. To this end, several mathematical models have been developed to describe drug release kinetics as a function of time. In this study, various release mechanisms were fitted to the experimental data to access the optimal pharmaceutical design of phenytoin, as summarized in Table 3. Comparative mathematical analysis of the experimental and predicted data allowed determination of the best-fitting model for phenytoin release from the nanofibers. The kinetic analysis for the predefined system of phenytoin release, was executed by taking into account 7 experimental data sets and 7 Kinetic models^42–48.

Table 3.

List of the kinetic models for phenytoin release from electrospun nano-fibers.

No.	Model name	Equation	Parameters & variables	Ref.
1	Zero-order		: Initial amount of phenytoin sodium : Remaining amount of phenytoin sodium at time t t: Release time : Zero-order model constant	42
2	First-order		: Initial amount of phenytoin sodium : Remaining amount of phenytoin sodium at time t t: Release time : First-order model constant	43
3	Higuchi		Remaining amount of phenytoin sodium at time t t: Release time : Higuchi model constant	44
4	Hixson–Crowell		: Initial amount of phenytoin sodium : Remaining amount of phenytoin sodim at time t t: Release time : Hixson–Crowell constant	45
5	Korsmeyer–Peppas		: The fraction of phenytoin sodium released at time t : Korsemeyer-Peppas release constant t: Release time n: Transport exponent	46
6	Weibull		: The fraction of phenytoin sodium released at time t : Rate constant : Shape parameter t: Release time	47
7	Baker–Lonsdale		: The fraction of phenytoin sodium released at time t : Release rate constant t: Release time	48

Results and discussion

ATR spectra

Figures 2a-c show the ATR spectra of PNCMC, PNPCL, and PNPCL-CMC nanofibers within the wavenumber range of 4500 − 400 . Figure 2a shows a broad absorption band at 3320 is attributed to the stretching vibrations of hydroxyl groups (–OH) of end chains of PVA (end chain) and CMC. The (–COOH) and (–OH) groups of CMC, together with the hydroxyl groups of PVA, can be engaged in hydrogen bonding with GA. One possible mechanism involves hydrogen bonding between the functional groups of PVA and CMC, as illustrated in Scheme 6a. GA can act as a crosslinker between CMC and PVA, thereby promoting the crosslinking process (Scheme 6b). The absorption band observed at 2950 corresponds to the asymmetric stretching vibrations of –CH and CH₂ groups of CMC and PVA, as well as the –CH group of the aldehyde in GA (Fig. 2a)⁴⁹. A distinct stretching vibration band of the carbonyl group (C = O) appeared at 1729 , which is consistent with previous findings⁵³. The carbonyl bond within the carboxyl group (–COOH) of CMC shows a stretching vibration band typically in the range of 1700–1735 . Moreover, the contributions from the carbonyl groups of PVA and GA intensified the observed stretching vibration. The carbonyl peak shifted toward lower wavenumbers, reflecting hydrogen bonding among PVA, CMC, and GA during the crosslinking process (Scheme 6b). The symmetrical bending vibration of C–O–C observed in 1080 indicates the crosslinking process via ether linkage between PVA and CMC in the presence of GA (Scheme 6b)⁴⁹.

Fig. 2.

ATR spectra of the prepared samples a PNCMC, b PNPCL, and c PNPCL-CMC.

Scheme 6.

Schematic description of molecular intraction between different polymers a PVA and CMC and b PVA, CMC, and GA⁴⁹.

The spectral representations of PNPCL nanofibers and the PNPCL-CMC nano-composite are presented in Fig. 2b and c. The absorption peaks of hydroxyl and carbonyl groups were appeared at similar wavenumbers. Additionally, the peaks observed at 2950 and 2869 in Fig. 2b and c correspond to asymmetric and symmetric stretching vibrations of CH₂ groups of polycaprolactone nanofibers. Moreover, the peaks at 1238 assigns to the stretching vibration of C–O–C in PCL (Fig. 2b and c), which is absent in Fig. 2a. The ATR spectrum of the two-layer PNPCL-CMC nano-composite (Fig. 2c) shows a combination of the charactristic peaks of both PCL and CMC. In this spectrum, the broad peak of –OH group (characteristic peak of PNCMC) disappeared, whereas the carbonyl peak intensified. The crosslinking reaction or grafting of CMC with PCL can lead to the diminution or complete disappearance of the –OH peak. Although grafting PCL onto CMC chains or crosslinking between CMC and PCL are common strategies for composite formation, this scenario is ruled out in this study. It seems the interactions between PNPCL and PNCMC during electrospinning process are primarily physical, involving van der Waals forces or polymer fiber entanglement, which may lead to the disappearance of –OH groups in PNPCL-CMC sample.

SEM images of the samples

Scanning electron microscopy (SEM) was used to examine the morphology of the nanofibers. SEM images and corresponding diameter distributions of all samples, including PNPVA, PNCMC, PNPCL, and PNPCL-CMC, are presented in Fig. 3.

Fig. 3.

a SEM images of the prepared samples including PNPVA, PNCMC, PNPCL, and PNPCL-CMC b Diameter distributions of nanofibers including PNPVA, PNCMC, PNPCL, and PNPCL-CMC.

As shown in Fig. 3a, the PNPVA and PNCMC are relatively uniform, continuous, and exhibit appropriate lengths and diameters. The fibers have smooth surfaces, with no cuts, bead formation, or other structural defects such as secondary jets or nanodroplets, owing to the precise control of structural parameters. The SEM images of PNPCL and PNPCL-CMC nanofibers reveal that the fibers were formed with smooth surfaces with no beads. The fiber diameter distributions of all samples (Fig. 3b) demonstrate a wide variation in fiber diameters. The increase in fibers’ diameters is attributed to the rise in solution viscosity caused by solvent evaporation during the electrospinning process. In this study, PNPCL was employed as a supporting layer for wound dressing applications. The integration and uniformity of fibers were well preserved in PNPCL-CMC sample. Fiber diameter distributions and mean diameters were evaluated using Image J (Fig. 3b). For each sample, 100 to 120 fibers were randomly measured from multiple independent SEM images to ensure representative sampling. All measurements were pooled, and the mean fiber diameter was calculated from the combined dataset. The average diameters of PNPVA, PNCMC, PNPCL, and PNPCL-CMC samples were estimated as 161, 238, 475, and 262 nm respectively. In fact, adding CMC to PVA increased the solution viscosity and ultimately increase the diameter of the nanofibers. Similarly, the larger diameter of PNPCL compared to PNCMC and PNPVA is attributed to the relatively high viscosity of the polymer.

EDS of the samples

Subsequently, energy dispersive X-ray (EDS) was employed for qualitative and quantitative elemental analysis of the samples including PNCMC, PNPCL, and PNPCL-CMC (Fig. 4a-c).

Fig. 4.

EDS images of the prepared samples including a PNCMC, b PNPCL, and c PNPCL-CMC.

The percentage of carbon and oxygen in the samples are reported in Table 4. The results show that carbon content in the PNPCL snd PNPCL-CMC is higher than in PNCMC sample.

Table 4.

Elemetal analysis of PNCMC, PNPCL, and PNPCL-CMC.

Sample name	C (A%)	O (A%)
PNCMC	52.13	47.87
PNPCL	61.55	38.45
PNPCL-CMC	61.33	38.67

Contact angle tests of the samples

Hydrophobicity expresses the affinity of a substance to repel water molecules and refers to the unwillingness of molecules to interact with water. In polar solvents such as water, hydrophobic molecules tend to aggregate into droplets or clusters to minimize contact with the solvent. In contrast, hydrophilicity refers to the affinity of a substance for water molecules. Hydrophilic molecules dissolve easily in water and other polar solvents due to their polarity and ability to form hydrogen bonds. These chemical properties are important features in tissue engineering and wound healing. The surface characteristics of a material strongly influence its interactions with the surrounding environment. Surface wettability is directly evaluated by measuring the contact angle between a liquid droplet and the solid surface. The contact angle plays a key role in studying the interaction between a surface and liquid, such as water. Wound dressing act as a protective layer over the wound, playing a crucial role in preventing infection, moisturizing the wound, and accelerating the healing process. The hydrophilicity of a wound dressing is a critical aspect in wound care and healing, as it facilitates the formation of a physical barrier against bacteria and other pathogens. Dry wounds are typically more painful and induce greater discomfort. The use of moist wound dressings can alleviate pain, and wounds that heal under moist conditions generally form smaller, less visible scars, which is aesthetically important for many patients.

CMC-based nanofibers demonstrate inherently low contact angle (often ), reflecting strong hydrophilicity due to hydroxyl and carboxyl functional groups⁵⁰. The hydrophilic CMC shows high potency to hydrate the wound environment, promote tissue repair, and prevent desiccation. On the other hand, hydrophobic polymers can restrict antimicrobial resistance by preventing biofilm formation. Therefore, a bifunctional composite consisting of hydrophilic and hydrophobic layers can help maintain wound moisture through the inner hydrophilic layer, while preventing bacterial penetration into the wound environment through the outer hydrophobic layer. In continuation, the hydrophilicity of the prepared samples, including PNCMC, PNPCL, and PNPCL-CMC was measured using water contact angle tests (Fig. 5a-c). In the case of the PNPCL-CMC nano-composite, the water droplet was placed on the cellulose layer surface. The results revealed that the contact angles between water droplets and the surfaces of the PNCMC and PNPCL-CMC samples were less than 10, indicating superhydrophilic surfaces that enabled the water droplets to premeate within the few seconds. In a study conducted by Mutahira et al., the contact angle between water and a CMC layer was reported in the range of 15–20 , but in the present study, due to the higher percentage of CMC, the contact angle decreased to less than 10 . In fact, CMC has significant affinity for water molecules, owing to its abundant hydroxyl groups⁵¹. The PNPCL sample exhibited hydrophobic nature, with water contact angles tipically excedding , which may limit water interaction in wound healing applications⁵². Adding PNCMC into hydrophobic polycaprolactone nanofibers enables the development of a bifunctional composite suitable for wound healing applications. This finding confirms previous reseacrches. The research show that PCL/CMC composire nanofibers exhibit substantially reduced contact angles compared to PCL nanofibers, indicating a synergistic modulation of surface properties⁵³.

Fig. 5.

Contact angles of water droplets on the prepared samples a PNCMC, b PNPCL, and c PNPCL-CMC.

Degradability and swelling ratio results of the samples

All degradability and swelling ratio experiments were examined in triplicate under identical conditions, and the results were reported as meanstandard deviation (SD). To assess the degradation behavior of the samples, weight loss in phosphate buffered saline (PBS) was monitored over predetermined time intervals. Hydrolytic degradation of polymeric nanofibers depends on the water molecule diffusivity into the nanofibers, followed by chemical reactions with the hydroxyl groups of the polymer structure. As a result of this process, the polymer chains were broken down and subsequently mechanical strength and sample weight were reduced. Figure 6a and b present the degradation profiles of CMC and polycaprolactone nanofibers in PBS. Corresponding error bars have been included in the figures to represent data variability. As shown, after 30 days, 93.6%3.16% of PNCMC was degraded. In contrast, polycaprolactone nanofibers exhibited lower degradability than carboxymethyl cellulose nanofibers, with only 33.7%3.5% of PNPCL degraded after 30 days. The degradation of PCL in phosphate buffered saline solution occurs due to hydrolysis of ester-aliphatic bonds⁵⁴, and the rate of degradation of this polymer is very slow due to its hydrophobic properties, which limits water adsorption and consequently reduces the rate of degradation.

Fig. 6.

Degradation percent of the samples a PNCMC and b PNPCL.

Figure 7a and b illustrate the swelling behavior of PNCMC and PNPCL samples respectively. The maximum swelling ratio of PNCMC was observed after 7 days, reaching 311.9%21.51%, whereas polycaprolactone nanofibers exhibited a maximum swelling ratio of 105%7.56 on the fifth day of the experiment. These findings are in agreement with the experimental results reported in the previous studies, particularly regarding the high swelling capacity associated with CMC-containing nanofibers. Previous studies have reported that electrospun PCL naofibers exhibit a swelling ratio of approximately 100%⁵⁵, that corresponds to the findings of the present study. Although few studies have quantitatively investigated the swelling behavior of nanofibers composed of both PCL and CMC in combination with PVA, the results obtained in the present study were compared with previously reported data on PCL- and CMC-based scaffolds. According to previous sudies, the addition of hydrophilic polymers such as hyaluronic acid & chitosan into PCL has been shown to enhance the swelling ratio to 130% & 270%, respectively^56,57. Furthermore, incorporation of CMC into PVA hydrogel structure significantly increased the swelling ratio form 416% for PVA pure hydrogel to 1437% for CMC/PVA hydrogel (CMC to PVA ratio of 80:20)⁵⁸.

Fig. 7.

Swelling ratio of the samples a PNCMC and b PNPCL.

Based on these findings, the higher swelling ratio of PNCMC compared to PNPCL can be attributed to the high hydrophilicity of CMC. In contrast, PCL exhibits relatively low swellingcapacity due to its hydrophobic nature.Therefore, incorporation of CMC can enhance the low swelling capacity of PCL. This property is a unique feature for wound dressing, enabling them to swell and absorb fluids secreted by injured and inflamed tissues.

Calibration curve of phenytoin

Sodium 5,5-diphenyl-2,4-imidazolidinedione with empirical formula C₁₅H₁₁N₂NaO₂ and molecular weight of 274.25 , which is known Phenytoin sodium, is an antiepileptic drug (Fig. 8). Phenytoin also, can be used for wound healing and tissue formation through different mechanisms.

Fig. 8.

Chemical structure of phenytoin sodium⁵⁹.

The concentration of phenytoin was calculated using a UV-Vis analytical method and the determined calibration curve. Specific amounts of phenytoin were dissolved in double distilled water to prepare solutions with concentrations of 2, 4, 5, 6,7, 8, and 10 . The UV-Vis absorbance intensity of each solution was measured to construct the calibration curve (Fig. 9).

Fig. 9.

The UV–Vis spectra of phenytoin sodium solutions with different concentration in the range of 2–10.

The UV absorbance values at =200 nm were plotted versus the phenytoin concentration (2–10 ) to construct the calibration curve (Fig. 10). The calibraion curve of the analyte revealed good linearity, with an R² value of 0.9632. The range of phenytoin concentrations used and the corresponding calibration results obtained in the present study are consistent with those reported in the cited literature^60,61.

Fig. 10.

Calibration curve of phenytoin solution at =200 nm.

Kinetic models and in vitro analysis results

The modeling of phenytoin release is complex due to different structural and operating parameters change over time. The release profile of phenytoin from the prepared PNPCL-CMC polymeric matrix initially exhibited a burst approach, followed by a plateau after 8 h (Table 5). The release experiments were performed in triplicate following the same experimental protocol and under identical conditions to ensure reproducibility.

Table 5.

The results of phenytoin release from electrospun nano-fibers.

No. of set	Time (h)	Mean Concentration
1	0.25	2.10.22
2	0.5	2.740.20
3	1	6.090.42
4	2	6.530.39
5	4	80.53
6	6	9.310.43
7	8	9.40.65
8	16	9.470.33
9	24	9.450.39

In burst release phase, erosion of the polymeric fibers accelerates drug release into the medium. However after reaching a plateau within 8 h, drug release was studied for 24 h, showing that the drug concentration preserved at a desirable level with no significant changes (Fig. 11). This finding is consistent with the study by Zarandi et al.⁶².

Fig. 11.

Phenytoin drug release concentration vs. time.

A better understanding of the mechanism of controlled-release of phenytoin can be gained by examining and comparing drug release kinetic models. The mathematical analysis was conducted to compare the fitting of 7 kinetic models for 7 sets of experimental data. In general, sum of the squared residuals (SSR) evalutes the deviation of the model’s predicted data from measured experimental data and can be used to find the best fitting model (Eq. 5). The sum of the squared residuals (SSR) was calculated using Microsoft Excel.

5

Where , , and N are the measured experimental value for the i^th data, predicted value for the i^th data, and number of measuring points respectively. Although the minimum SSR value indicates the best-fitted model for the measured experimental data, but in complex systems the SSR by itself is not sufficient to determine the best model. Usually, the Akaike’s method can be followed to compare the statistical mathematical models. The Akaike Information Criterion (AIC) is definedas as follows:

6

Where p is the number of model parameters, which expresses the complexity of the model. To validate the predictive potential of kinetic models, linear regression analysis were conducted to compare SSR and AIC parameters. In continuatin, several mechanistic/empirical/semi-emperical models were developed to explain the behavior of drug delivery systems by considering diffusion, dissolution, and erosion mechanisms involved in release process. The kinetic plots and determined variables were represented in Fig. 12; Table 6.

Fig. 12.

The plots of kinetic models for phenytoin release a Zero-order, b First-order, c Higuchi, d Hixson–Crowell, e Korsmeyer–Peppas, f Weibull, and g Baker–Lonsdale.

Table 6.

The results of phenytoin release from electrospun nano-fibers PNPCL-CMC according to kinetic models.

No.	Model name*	P **	n			***		SSR	AIC
1	Zero-order	2 ()	-	-	-	0.8817	0.7963	10.6816	20.5796
2	First-order	2 ()	-	-	-	−0.0162	0.6524	0.7515	2.0002
3	Higuchi	2 (, )	-	-	-	0.3328	0.9052	0.0554	−16.2522
4	Hixson–Crowell	2 ()	-	-	-	−0.0933	0.7016	0.1991	−7.2976
5	Korsmeyer–Peppas	3 (, , n)	0.4417	-	-	0.4578	0.92	0.1729	−6.2852
6	Weibull	2 )	-	0.0059	0.4664	-	0.9244	0.1815	−7.9455
7	Baker–Lonsdale	2 (, )	-	-	-	0.0641	0.9822	0.0039	−34.8275

^* is the total amount of drug release at infinite time. For all experimenatl sets, =9.47.

^**P is the number of parameters of each kinetic model.

^***k is k₀ of zero-order model, k₁ of first-order model, k_H of Higuchi model, k_HC of Hixson–Crowell model, k_KP of Korsmeyer–Peppas model, and k_BL of Baker–Lonsdale model.

In order to assess different kinetic models in predicting the release of phenytoin from PNPCL- CMC nano-composite, AIC criterion was calculated. The results show that Baker–Lonsdale is the best-fitted model with AIC value of −34.8275 (mimum value) and R² value of 0.9822.

To assess different kinetic models for predicting phenytoin release from the PNPCL-CMC nano-composite, AIC criterion was calculated. The results indicate that the Baker–Lonsdale model provided the best fit, with a minimum AIC value of −34.8275 and an R² value of 0.9822. The Baker–Lonsdale model (1974) is an extension of the pioneering Higuchi kinetic model⁶³. Although the Baker-Londsdale model was initially developed for drug release from spherical matrices such as microcapsules or microspheres, it has also been adapted for cylinderical geometries like as elecrospun nanofibers. In essence, drug release is initially controlled troughout diffusion, followed by dissolution mechanisms. It was found that the Baker–Lonsdale model accurately forecasts phenytoin release from the PNPCL-CMC ploymeric matrix, even though its non-spherical geometry. This mathematical study provides a rational, time-affordable, and cost-effective strategy for designing conventional drug delivery systems, particularly for wound dressing applications.

Conclusion

In this study, hybrid nanofibers and nano-composites were fabricated using synthetic PCL and natural CMC the electrospinning process. CMC alone cannot be electrospun due to its high viscosity and low electrical conductivity. The incorporation of PVA and GA reduced CMC solution viscosity and enhanced its electrical conductivity, enabling the formation of uniform nanofibers via electrospinning. PVA at different concentrations and mixing ratios was applied as a cross-linker to improve the mechanical and chemical properties by trapping the water molecules within the polymer chains. A bi-layer PNPCL-CMC nano-composite, composed of superhydrophilic CMC nanofibers and hydrophobic PCL fibers, was developed as a potential wound dressing material. Phenytoin sodium was incorporated at different concentrations to produce drug loaded nanofibers. Swelling and degradation studies demonstrated the potency of PNPCL-CMC for wound healing applications. Drug release was governed by a combination of diffusion and errosion mechanisms, with the Baker–Lonsdale model providing the best fit to the phenytoin release data. As the total amount of phenytoin release at infinite time () occurred within the first 8 h, the phenytoin release was primarily governed by a burst release.

Author contributions

M. H.: Writing the main manuscript, Pharmacokinetic studies, Visualization, Validation, Methodology, Investigation, Conceptualization. F. T., Synthesis and did all experiments. Z. E. G.: Writing and editing of introduction, E. S.: Drawing and editing all schemes.

Data availability

The datasets used or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.